UBC Theses and Dissertations

UBC Theses Logo

UBC Theses and Dissertations

S-layer biogenesis studies in Caulobacter crescentus : RsaA anchoring and the localization of the S-layer-associated.. Ford, Matthew James 2006

You don't seem to have a PDF reader installed, try download the pdf

Item Metadata

Download

Media
[if-you-see-this-DO-NOT-CLICK]
ubc_2006-0041.pdf [ 9.19MB ]
[if-you-see-this-DO-NOT-CLICK]
Metadata
JSON: 1.0092468.json
JSON-LD: 1.0092468+ld.json
RDF/XML (Pretty): 1.0092468.xml
RDF/JSON: 1.0092468+rdf.json
Turtle: 1.0092468+rdf-turtle.txt
N-Triples: 1.0092468+rdf-ntriples.txt
Original Record: 1.0092468 +original-record.json
Full Text
1.0092468.txt
Citation
1.0092468.ris

Full Text

S-layer biogenesis studies mCaulobacter crescentus: J t s a A anchoring a n d the localization of the S-layer-associated protease by Matthew James F o r d B. Sc., Bishop's University, 2002  A THESIS SUBMITTED IN PARTIAL FULFILLMENT OF THE REQUIREMENTS FOR THE DEGREE OF M A S T E R OF S C I E N C E  in THE FACULTY OF GRADUATE STUDIES ( Microbiology and Immunology)  THE UNIVERSITY OF BRITISH COLUMBIA October 2005 ©Matthew James Ford, 2005  A  B  S  T  R  A  C  T  Despite the widespread occurrence of bacterial S-layers, little is known about the mechanisms of attachment to the cell surface, especially in the case of Gram-negative organisms. The S-layer of  Caulobacter crescentus is composed of a single protein,  R s a A . After export, R s a A assembles into a hexagonal crystalline array that covers the bacterium. In this array, some R s a A monomers are directly attached to the cell surface, while others are surface anchored only by interacting to other R s a A monomers. Since truncations (9) and mutations (8) in the R s a A N-terminus result in S-layer shedding into the culture medium, we hypothesized that the N-terminus of R s a A anchors the monomer to the cell surface. However, since disruption of the R s a A N-terminus and disruption of the putative R s a A subunit-subunit interaction domain both result in the same phenotype (S-layer shedding), when a particular mutation results in a shedding phenotype, it is difficult to know whether R s a A anchoring or R s a A subunit-subunit interaction has been perturbed. T o tease apart these issues, we have developed an assay where small R s a A fragments are incubated with S-layer-negative cells to assess the ability of the fragments to re-attach. In doing so we found that the R s a A anchoring region lies in the first -225 amino acids, that this R s a A anchoring region requires a smooth lipopolysaccharide molecule found on the outer membrane, and that even minor perturbations within the first -225 amino acids of R s a A cause loss of anchoring. Mutations that lie outside of the R s a A anchoring region but still result in the shedding phenotype are likely disrupting R s a A - R s a A subunit-subunit interactions rather than directly disrupting R s a A anchoring.  ii  A s a by-product of these anchoring studies, we have recent preliminary data that Sap, an S-layer editing protease, is likely to be an extracellular membrane-bound protease, rather than an intracellular protease as previously proposed. Additionally, we have found that Sap is likely to be secreted to the cell surface via the same Type I secretion transporter that the S-layer protein utilizes.  iii  TABLE OF CONTENTS ABSTRACT  ii  T A B L E OF CONTENTS  . . . .iv  LIST OF T A B L E S  vi  LIST OF FIGURES  vii  LIST O F A B B R E V I A T I O N S  viii  ACKNOWLEDGEMENTS  ix  1. I N T R O D U C T I O N  1  1.1 1.2 1.3 1.4 1.5  -  General features of S-layer composition Frequency of S-layer occurrence in nature Functions of S-layers: Gram-positive bacteria Functions of S-layers: Gram-negative bacteria Summary of S-layer functions and applied utility of S-layers  1.6 - Caulobacter crescentus 1.7 - C. crescentus S-layer function 1.8 - Composition and assembly of C. crescentus S-layer 1.9 - S-layer secretion apparatus in C. crescentus 1.10 - Components of the R s a A secretion system. 1.11 - R s a A editing: Sap metalloprotease 1.12 - The "weak spot" in R s a A 1.13 - Region of R s a A that mediates secretion 1.14 - Region of R s a A that mediates cell surface anchoring 1 . 1 5 - Mechanism of S-layer anchoring to the cell surface: Gram-positive bacteria 1.16 - Mechanism of S-layer anchoring to the cell surface: Gram-negative bacteria 1.17 - Tolerated insertions in RsaA 1.18 - The N-terminus of R s a A 1.19 - Summary of the study  2 2 3 5 6  7 8 8 .'  10 11 11 12 13 14 15 16 17 17 18  2. M A T E R I A L S A N D M E T H O D S 20 2.1 - Bacterial strains, plasmids, and growth conditions 20 2.2 - Plasmid and D N A manipulations 20 2.3 - C. crescentus expression vectors 20 2.4 - Construction of plasmids carrying rsaA with collagenase cleavage sites.. ..24 2.5 - Construction of plasmids carrying rsaA with collagenase cleavage at R s a A position 277 with additional mutations in the R s a A N-terminus 26 2.6 - Construction of plasmids used for gene disruptions 28 2.7 - Construction of gene disruptions 31  2.8 - S-layer reattachment assays 2.9 - Protein techniques  36 37  3. R E S U L T S - R s a A anchoring studies 40 3.1 - Assay development elucidates the importance of R s a A secondary structure for wild type RsaA anchoring 40 3.2 - S L P S is required for RsaA anchoring 47 3.3 - T w o anchoring regions in RsaA? 49 3.4 - R s a A 1-277 is sufficient for R s a A anchoring 54 3.5 - The R s a A C-terminal fragment reattachment requires full-length R s a A . . . . 5 5 3.6 - Mutations in R s a A 1-277 cause the loss of R s a A anchoring 59 4. R E S U L T S - Sap localization studies 4.1 - Sap cleaves R s a A 1-277...., '. 4.2 - Sap is an extracellular enzyme 4.3 - S-layer Type I secretion O M P s F a and/or Fb are involved in Sap secretion '. 4.4 - A third F outer membrane protein may be present in strain JS4000 4.5 - Sap is unlikely to be exported into the culture media  64 64 66 67 70 72  5. D I S C U S S I O N A N D C O N C L U S I O N  76  REFERENCES  82  v  LIST OF TABLES Table 2-1. Bacterial strains and plasmids  :  21a/b  LIST OF FIGURES  Figure 1-1. B o x model representation of the C.  crescentus S-layer monomer, 1026  amino acid R s a A  9  Figure 2-1. S-layer reattachment in calcium-supplemented water  42  Figure 2-2. S-layer reattachment in P Y E  43  Figure 2-3. S-layer reattachment by co-culturing S-layer donors and recipients  45 '  Figure 2-4. S-layer reattachment using soluble R s a A from the supernatant of shedder cultures  46  Figure 2-5. Proposed N-acetylperosamine biosynthetic pathway (2)  48  Figure 2-6. S L P S levels in SLPS-positive, SLPS-deficient, and SLPS-negative C.  crescentus strains  ...50  Figure 2-7. S-layer attachment requires S L P S  51  Figure 2-8. T w o anchoring regions in R s a A ?  53  Figure 2-9. R s a A 1-277 is sufficient for R s a A anchoring  56  Figure 2-10. R s a A C-terminal fragment anchoring requires full-length R s a A .  58  Figure 2-11. B o x model of mutant R s a A used for reattachment studies  60  Figure 2-12. Mutations, at R s a A amino acids 7, 29, and 69 disrupt R s a A anchoring  61  Figure 2-13. Mutations at R s a A amino acids 154, 169, and 222 disrupt R s a A anchoring  62  Figure 3-1. Sap cleaves R s a A 1-277  65  Figure 3-2. Sap is probably an extracellular enzyme  68  Figure 3-3. RsaFa and RsaFb levels in w i l d type and RsaF knockout strains  69  Figure 3-4. S-layer Type I secretion O M P s F a and/or Fb are involved in the secretion of Sap  71  Figure 3-5. A third F outer membrane protein may be present in strain JS4000  73  Figure 3-6. Sap is unlikely to be exported into the culture media.  75  vii  LIST OF ABBREVIATIONS ABC Amp Ap Cm Cm C-terminus DNA DNase EDTA . EGTA GSP HEPES Kan kb Km' kDa r  r  Hg pi mg ml min MCS MFP N-terminus OD OM OMP MCC PAGE PCR PGN PMN PYE RNase RTX SDS S-layer SLH SLPS Sm Sm Sue Tris 6 0 0  r  5  A T P - B i n d i n g Cassette Ampicillin A m p i c i l l i n resistance Chloramphenicol Chloramphenicol resistance carboxy terminus deoxyribonucleic acid deoxyribonuclease ethylenediaminetetraacetic acid p-aminoethyl ether)-N, N , N \ N'-tetraacetic acid General Secretory Pathway 4-(2-hydroxyethyl)-1 -piperazineethanesulfonic acid Kanamycin kilobases Kanamycin resistance kilodaltons microgram microlitre milligram millilitre minute Multiple Cloning Site Membrane Fusion Protein amino terminus Optical Density at absorbance of 600 nm Outer membrane Outer Membrane Protein Multiple Cleavage Cassette Polyacrylamide gel electrophoresis Polymerase Chain Reaction Peptidoglycan polymorphonuclear leukocytes Peptone Yeast Extract ribonuclease Repeat in Toxin Sodium dodecyl sulfate Surface layer Surface Layer Homology Smooth Lipopolysaccharide Streptomycin Streptomycin resistance Sucrose sensitivity Tris (hydroxymethyl) methylamine  ACKNOWLEDGEMENTS  I would like to thank my supervisor Dr. John Smit for his guidance, patience, support and understanding throughout my project.  D r . John N o m e l l i n i was instrumental i n  clarifying cloning strategies and other parts of my experimentation, and was also there to support me i n general and I thank h i m for that. Lab members also helped at various stages, including Dr. Peter A w r a m , M i k e Toporowski, Andrea Pusic, Janny L a u , & Corin Forrester. I thank Jodi Y u e for listening and providing feedback to me regarding my project. M a n y members of the Fernandez lab and the Murphy lab were very generous and helpful to me through this project. I also thank my family and friends, without whom I would have never made it here.  This work is dedicated to my grandfather, whom I miss very much.  ix  1 - INTRODUCTION  The  Caulobacter crescentus S-layer has been the subject of extensive investigation  and manipulation i n our lab.  Our lab has essentially two major areas o f focus:  manipulating the S-layer for biotechnological applications, and understanding S-layer biogenesis from a basic science perspective. O n the biotechnology front, we have been successful in modifying the,S-layer to cause it to shed from the bacterium into the culture medium, by itself or together with fused passenger proteins (9).  This shed protein  aggregates and is easily retrieved by simple filtration. W e have also displayed many heterologous proteins within the S-layer on the bacterial cell surface (7). In terms of wild-type S-layer biogenesis, we have learned about aspects of R s a A production, secretion, and regulation (1), (71). W e have had some indication that the N terminus of the S-layer monomer, R s a A , mediates R s a A anchoring, because R s a A N terminal truncations (9) and mutations (8) lead to the shedding phenotype.  This thesis  aimed to define the R s a A anchoring region more precisely, and to discover which regions or residues within this anchoring domain are involved in cell surface attachment. Results from this study suggest that indeed the R s a A N-terminus mediates R s a A anchoring, but surprisingly we learned that all minor perturbations constructed thus far within the first -225 amino acids of R s a A disrupt surface attachment. Through this work, we also clarified some other aspects of S-layer biogenesis, such as the requirement for the cell surface molecule smooth lipopolysaccharide (SLPS) for S-layer attachment. A s a by-product of these S-layer anchoring studies, we have learned more about a  C. •  crescentus metalloenzyme, the S-layer-associated protease, Sap. W e have recent data  1  that Sap, an S-layer editing protease, is likely to be an extracellular membrane-bound protease, rather than an intracellular protease as previously proposed. Additionally, we found that Sap is likely secreted to the cell surface v i a the same Type I secretion transporter that the S-layer protein utilizes. Current investigation is directed to confirm and extend these hypotheses.  1.1 - General features of S-layer composition Bacterial surface layers (S-layers) are composed of protein or glycoprotein subunits. Although the S-layers of the Gram-positive bacteria  Clostridium difficile and Bacillus  anthracis contain two different S-layer subunits (43, 69), most S-layers are composed of a single protein or glycoprotein species, with molecular masses ranging from 40 to 200 k D a (59). These subunits assemble on the outermost surface of the cell, forming twodimensional lattices that cover the organism (5). The S-layer is the first contact point between the S-layer-possessing cell and the external environment. For this reason, it has been suggested that differing S-layer compositions among organisms reflects specific \  adaptations to the environmental and ecological conditions in which these organisms live (4).  1.2 - Frequency of S-layer occurrence in nature M a n y bacteria possess S-layers. However, the high frequency of occurrence of Slayer-possessing organisms has only been recognized over the last 25 years. It is now known that S-layers represent an almost universal feature of archaebacteria (65).  S-  layers have also been found on hundreds of different species of almost every taxonomic  2  group of eubacteria (57). Even some eukaryotic algae have been reported to have Slayers (52).  G i v e n the frequency of occurrence of S-layers-possessing organisms, it  seems important to understand various S-layer functions, and how S-layer structures and compositions enable S-layers to perform these functions. Despite the large number of Slayer-possessing organisms and the considerable knowledge that has accumulated regarding S-layer structure, assembly, biochemistry, and genetics, relatively little is known about specific S-layer functions (63). The next several paragraphs describe what is currently known about the functions of S-layers in different bacterial species.  1.3 - Functions of S-layers: Gram-positive bacteria Efforts have been mounted to understand the functions of S-layers from several Grampositive bacteria  including  Bacillus cereus, Staphylococcus aureus, Bacillus  stearothermophilus, and Clostridium thermosulfurogenes (reviewed i n (5)). The S-layers of  B. cereus and S. aureus offer these organisms protection against phagocytosis (5). B.  cereus is often found i n oral infections i n humans.  T o clear these types of infections,  phagocytosis is the most important defense mechanism (5). The phagocytic process can be enhanced by the opsonization of the bacteria by host opsonins such as antibodies and complement proteins. However, a shortage of host opsonins might occur as a result of bacterial defense proteases secreted during infection which destroy the opsonins (34, 68). Accordingly, phagocytosis of  B. cereus i n the absence of host opsonins has been a  research focus. Studies suggest that cells from young cultures of  B. cereus are readily  ingested by polymorphonucleur leukocytes ( P M N s ) , while cells from aged cultures are not (35). This phenomenon has been attributed to changes i n the properties of the B.  3  cereus S-layer during the aging of the culture which result in increased resistance to nonopsonin-mediated phagocytosis (35). This suggests that for B. cereus, a mature S-layer offers protection against this form of phagocytosis. prevent opsonin-mediated phagocytosis.  S-layers can also play a role to  S. aureus is an opportunistic pathogen that  usually infects immuno-comprimised individuals, and is often found to be the causative agent of hospital-acquired infection. S-layer encapsulation of this bacterium interferes with opsonin-mediated phagocytosis by P M N s by preventing opsonin binding to S.  aureus (49). Evidently, these S-layers help these pathogens evade host immune responses during infection. Some Gram-positive S-layers are anchoring points for exo-proteins, as is the case for the S-layers of B. stearothermophilus and C. thermosulfurogen.es. The S-layer of B.  stearothermophilus acts as an adhesion site for an exo-enzyme (amylase) secreted by the bacterium  into .the  culture  m e d i u m (18).  Similarly,  the pullulanase o f C .  thermosulfurogenes remains anchored to the cell surface by binding to the S-layer of C. thermosulfurogenes (41). Both the amylase and the pullulanase are surface-anchored by the same mechanism, as w i l l be discussed later. Thus S-layers are acting as platforms for attaching other exo-proteins in these bacteria. Since the S-layer is often a first contact point between the S-layer-possessing cell and the external environment, it is intuitive that S-layers serve as barriers that protect organisms from harmful substances in their external environments, such as lytic enzymes. Since S-layer lattices possess pores identical in size and diameter in the 2- to 8-nm range, they function as precise molecular sieves, excluding most macromolecules and proteins  (59). The S-layer of Bacillus stearothermophilus has been shown to act in this way,  4  excluding molecules with molecular weights greater than 30, 000 D a (58).  Since wild-  type strains are often outgrown by S-layer-negative strains under optimal growth conditions i n the laboratory (64), there must be strong selective pressure to maintain Slayers i n competitive environments, further demonstrating that S-layers play a vital role for organisms in their natural environments.  1.4 - Functions of S-layers: Gram-negative bacteria Although most information regarding S-layer function pertains to Gram-positive organisms, the functionality o f some S-layers from Gram-negative organisms is beginning to become understood.  The S-layers from the Gram-negative bacteria  Aeromonus salmonicida and Campylobacter fetus are important virulence factors. A. salmonicida was originally characterized as a pathogen that infects salmon, but it is now known that many fish species are susceptible to A. salmonicida infection (77).  A.  salmonicida possesses an S-layer that assembles into tetragonal arrays on the cell surface (42). It has been shown that S-layer-expressing but not S-layer-deficient A. salmonicida, can adhere to, enter, and survive within macrophages (24). This S-layer appears to be critical for the adhesion step o f infection, since S-layer-expressing A. salmonocida adheres to trout and murine macrophages successfully, while S-layer-deficient A. salmonicida cannot adhere to the same macrophages (22, 23). Another pathogen, C. fetus, is a causative agent of extraintestinal infections in humans (11), and o f infertility and abortion i n sheep and cattle (16).  C. fetus possesses an S-  layer that assembles into a hexagonal array (20). S-layer-deficient C. fetus has never been seen in human nor ovine C. fetus clinical isolates (5), suggesting the C. fetus S-layer  5  plays a key role in successful pathogenesis. The presence of the C. fetus S-layer results in impaired human complement binding, which may explain why S-layer-expressing C. fetus is resistant to serum and opsonization (11).  C. fetus S-layer-containing subunit  vaccines offer sheep protection against C. /erws-induced ovine abortion (42). For these reasons, the C. fetus S-layer is thought to be the predominant virulence factor in ovine  and human C. fetus infections (48). S-layers offer protection from predators found in a competitive environment (reviewed in (64). For example, S-layers from the Gram-negative bacteria  Aeromanas  salmonocida, Campylobacter fetus, Aquaspirillum serpens, and Caulobacter crescentus protect the cells from attack by bacterial parasites such as  Bdellovibrio bacteriovorus (5).  These S-layer-possessing prey cells are resistant to  Bdellovibrio predation, whereas  isogenic S-layer-deficient prey cells are not (36).  Interestingly, some predatory  organisms may have adapted to overcome the barrier role that S-layers serve. example, the C.  For  crescentus S-layer acts as a phage receptor for transducing O C R 3 0 , and  S-layer-deficient C.  crescentus is actually OCR30-resistant (17). This type of use of S-  layers as phage receptors has been reported in a few other instances (30, 38).  1.5 - Summary of S-layer functions and applied utility of S-layers Taken together, the evidence to date suggests that S-layers have evolved to perform a variety of functions. S-layers can act as molecular sieves, excluding harmful molecules found i n a competitive environment.  S-layers also protect organisms from predation  from other organisms found in nature. Some S-layers have been shown to aid pathogens to evade host immune responses, while others provide platforms for exo-protein  6  anchoring.  However, since there has been little study on S-layer function i n many  organisms, this list of S-layer functions is unlikely to be exhaustive. Additionally, the S-layers of some bacteria (including  C. crescentus ) have been  engineered to express foreign peptides or proteins, giving rise to many potential biotechnology applications (reviewed i n (44)). Because S-layers occur frequently, are poorly understood, and have promising potential in biotechnological applications, efforts are being made to better understand various aspects in this developing field.  1.6 - Caulobacter crescentus There are many reasons why the S-layer of the Gram-negative bacterium Caulobacter  crescentus is a good choice for further study. First, C . crescentus is a harmless bacterium. Indeed, C.  crescentus can often be isolated from "clean" drinking water (40).  Therefore risk involved with working with this bacterium i n the laboratory is minimal. Second, the Caulobacter genus o f bacteria is very common, and perhaps even abundant in many aquatic and terrestrial environments (40, 50, 51). Caulobacter is therefore o f general scientific interest i f we wish to understand the organisms that are prevalent in our environment.  T h i r d , C.  crescentus has been widely used as a model system for  understanding fundamental aspects of cell development and differentiation (14, 28). Caulobacter differs from other bacteria i n that it exhibits an unusual biphasic lifestyle, alternating between a stalked cell and a non-stalked dispersal (swarmer) cell (61). The switch between these two phases o f the Caulobacter life cycle has been studied as a differentiation process. Fourth, large amounts of genetic studies have been performed on this bacterium; the sequenced genome is available ( w w w . t i g r . o r g ) . Fifth, the C.  7  crescentus S-layer has been engineered to display foreign peptides of biotechnological interest, and has also been used to effect the secretion of fusion proteins, ideal for use in antibody production, into the culture medium (46). Therefore many data are available regarding C. crescentus, but not in the area of wild-type S-layer biogenesis.  1.7 - C. crescentus S-layer function The only known function of the S-layer of C. crescentus is the protective role it plays in the predation by a Bdellovibrio-like organism; S-layer-positive C. crescentus is resistant to this predation, whereas isogenic S-layer-negative C. crescentus is not (36). Since C. crescentus forms biofilms in nature, an important role for the C. crescentus Slayer may be protection by selective porosity from various predatory assaults that are likely to occur in complex bacterial biofilm communities (5). Difficulty in transforming C. crescentus by electroporation may be a reflection of this protective role (27).  1.8 - Composition and assembly of C. crescentus S-layer The S-layer of C. crescentus is composed of a single protein, RsaA  (26), and a  representation of RsaA is illustrated in Fig. 1-1. RsaA is a 1,026 amino acid protein, with a predicted molecular weight of 98 kDa (26). RsaA is not modified post-translation, aside from the cleavage of the initial methionine residue, nor is it glycosylated (26). RsaA assembles on the cell surface into a two-dimensional hexagonal lattice that completely covers the bacterium (67). Proper S-layer assembly is dependent on smooth lipopolysaccharide (SLPS) found on the outer membrane of C. crescentus, and mutants  8  Putative anchoring domain (First -225 residues)  Insertion-tolerant region  Type I secretion signal (last -80 residues)  Fig. 1-1. Box model representation of the C. crescentus S-layer monomer, 1026 amino acid R Mutation of the N-terminus causes S-layer shedding, as does mutation of the Repeat in Toxin ( R T X ) region, a calcium-binding motif (8). Mutation beyond amino acid 945 causes loss of S-layer secretion (9). Many sites within the middle of R s a A can tolerate insertion of foreign peptides without causing any adverse effects on S-layer anchoring or crystallization (8).  that are S-layer-proficient but SLPS-deficient shed RsaA into the culture medium (2). The prominent morphological unit in the hexagonal array has six RsaA monomers, and is spaced at 22 nm, as determined by electron microscopy and three-dimensional image reconstruction (67).  Current research is focused on building correlations between  primary RsaA structure and the spatial organization of the S-layer as evidenced by electron microscopy, in order to understand which RsaA regions are mediating the anchoring of RsaA to the cell surface, and which regions are involved in RsaA subunitsubunit interactions. Clearly, a crystal structure at the atomic level for RsaA would aid tremendously in this task, but so far such a structure is not available. In fact, there is no 3-dimensional crystal structure available for any S-layer protein to date. This is probably because the intrinsic nature of S-layers to form 2-dimensional structures likely impedes the formation of 3-dimensional structures that are needed to obtain a crystal structure.  1.9 - S-layer secretion apparatus in C. crescentus C. crescentus employs an ABC transporter (Type I) secretion apparatus to export RsaA monomers to the cell surface (1). In most bacteria, S-layer secretion does not occur via the Type I system, but rather by the general secretory pathway, GSP (12). In contrast to the Type II GSP, the Type I secretion system for Gram-negative bacteria exports substrates to the cell surface without a periplasmic intermediate, and utilizes a noncleaved C-terminal secretion signal present on secretion substrates to effect export. Two well-described Type I secretion systems are those required for the secretion of E. coli hemolysin, HlyA (6), and a Pseudomonas aeruginosa alkaline protease, AprA (55). Aside from the S-layer of C. crescentus, only two other S-layer proteins are known to be  10  exported by a Type I secretion apparatus: the and the  Campylobacter fetus S-layer protein (70),  Serratia marcescens S-layer protein (33). W i t h the exception of the S-layer  proteins from these three organisms, all S-layer proteins sequenced to date are produced with an N-terminal secretion signal peptide that is cleaved during translocation through the plasma membrane (59).  1.10 - Components of the RsaA secretion system The C. crescentus S-layer Type I secretion system has three components: an A B C transporter, its accessory molecule (membrane fusion protein, M F P ) , and an outermembrane protein ( O M P ) (6). The S-layer transporter genes in C. cloned and sequenced.  The  crescentus have been  rsaD, rsaE, and rsaF genes encode the putative A B C  transporter, M F P and O M P proteins, respectively. This secretion apparatus is responsible for secreting a large amount of protein (RsaA constitutes 10-12% of total cell protein (1)). It is perhaps not surprising that recent gene deletion studies have found that there are two homologous genes that encode for the O M P (RsaF) component of this transport system, perhaps enabling the cell to continuously export the large amounts of R s a A produced, even in the event that one of these O M P genes becomes non-functional (71). It may also be that  C. crescentus has developed other mechanisms to ensure that these large  amounts of R s a A do not build up intracellularly, should R s a A transport break down (71).  1.11 - RsaA editing: Sap metalloprotease A n editing process occurs during biogenesis of the  C.  crescentus S-layer, which  appears to eliminate bad (mutant) copies of R s a A that might otherwise impair S-layer  11  crystallization and tight packing of the R s a A monomers. A metalloprotease has been identified, which cleaves some mutated versions of RsaA, and has been named Sap (72). The C-terminal of Sap is homologous with the N-terminal of R s a A . Because the Sap sequence does not appear to contain (a) a signal leader peptide at the extreme N-terminal, (b) a Type I C-terminal secretion signal, and (c) obvious transmembrane domains, it was previously suggested that Sap is an intracellular enzyme. Although the native function of Sap is unknown, our previous model of Sap-mediated editing of recombinant S-layer suggested that Sap associates with nascent R s a A monomers inside the cell to "scan" for bad copies of R s a A (72).  A s an extension of this model, when Sap detects a bad R s a A  copy, it cleaves the mutant R s a A , which separates the R s a A C-terminal secretion signal from the portion of R s a A that is N-terminal to the cleavage site, and may also result in the intracellular degradation of the cleavage products (72).  Confusingly, however,  sometimes both Sap cleavage products ended up on the cell surface, though only one of these products contains the C-terminal secretion signal to effect secretion to the cell surface. Because Sap sequence suggests it is an intracellular enzyme, we hypothesized that R s a A intramolecular forces (such as hydrogen bonding) keep the two Sap cleavage products linked together, which could account for the secretion of both products. However, the current study challenges our previous model of Sap, suggesting that it is in fact an outer membrane-bound enzyme.  1.12 - T h e "weak spot" i n R s a A I found that Sap was responsible for undesired proteolysis of some R s a A mutants that I constructed. In a particular R s a A mutant that I constructed that contained a collagenase  12  cleavage site, after treatment with collagenase, an additional Sap-mediated cleavage of R s a A occurred. This Sap cleavage at this particular site has been observed in other R s a A mutants (8). This suggests that there is a weak Sap recognition site only exposed when certain mutations in R s a A alter the folding of the protein. This may be a Sap-mediated strategy employed by  C. crescentus to release from the cell surface various R s a A mutants  that might perturb S-layer assembly. In terms of R s a A anchoring studies, i f a particular mutation caused loss of R s a A anchoring, it was difficult to assess whether this loss of anchoring was a result of the mutation perturbing the R s a A anchoring domain itself, or whether the mutation converted the protein into an Sap substrate, and the proteolysis of the R s a A mutant caused the loss of anchoring.  Therefore, when investigating the  importance of R s a A regions or residues for R s a A attachment or crystallization, after mutating R s a A regions or residues, Sap-negative strains were employed to eliminate this variable. In that way I could test the effects of the particular R s a A mutations and be confident that the phenotype observed (loss of anchoring) was a consequence of the mutation perturbing the R s a A anchoring domain, rather than a consequence of simply exposing or creating a Sap cleavage site on RsaA.  1.13 - Region of RsaA that mediates secretion The secretion of R s a A to the cell surface is directed by a C-terminal secretion signal, localized to the last 82 amino acids of R s a A (9). This secretion signal can be genetically fused to peptides of choice, creating hybrid proteins that get secreted to the culture medium and aggregate into macroscopic particles that can be recovered by simple filtration (9). The primary R s a A sequence reveals the presence of a calcium binding  13  motif, the so-called R T X (repeat in toxin) motif, near the C-terminal of R s a A (26). The R T X motif is a tandemly-repeated nine amino acid ( L - X - G - G - X - G - ( N / D ) - D - X ) sequence found in Type I secreted proteins, of which the prototype (reviewed in (76)).  E. coli hemolysin protein H l y A is the  Proper cell-surface assembly of R s a A into the native  hexagonal array requires calcium (67, 75).  The R T X region of R s a A is thought to  mediate R s a A crystallization (subunit-subunit interaction) on the cell surface via calcium bridging between R s a A monomers (67).  C a l c i u m does not directly bind to the cell  surface: attempts to bind radioactively-labeled calcium to the surface of S-layer-deficient cells have been unsuccessful.  Therefore we believe that c a l c i u m does not directly  mediate R s a A attachment to the cell surface. Perturbations i n R s a A near or at the R T X motif results in the shedding of R s a A into the culture medium, possibly by damaging R s a A subunit-subunit interaction which may be mediated by calcium bridging at the R T X region. The sequences of C. crescentus R s a A and the  C. fetus S-layer protein Sap A share  significant homology in their C-termini. This, and the fact that SapA is also exported to the cell surface by a Type I secretion apparatus (70) suggests a common secretion mechanism for these two S-layers proteins.  R s a A and S a p A are not, however,  homologous i n their N-termini, which are believed to mediate anchoring o f these proteins, suggesting a different surface anchoring mechanism exists for these proteins.  1.14 - Region of RsaA that mediates cell surface anchoring The regions of R s a A that mediate S-layer anchoring to the cell surface remain unknown, but since mutations (8) and truncations (9) i n the R s a A N-terminus lead to the  14  shedding phenotype, it is likely that the N-terminus mediates R s a A anchoring. Insight into R s a A anchoring would further the utility df  C. crescentus as a tool for surface  display of peptides or proteins. If the R s a A anchoring regions were determined, then their maintenance when designing heterologous proteins for surface display would facilitate the effective anchoring of these proteins on the  C. crescentus cell surface. A  better understanding of the R s a A anchoring region may aid in manipulating R s a A such that it could anchor to other inert surfaces, opening a door to many applications in biotechnology.  1.15 - Mechanism of S-layer anchoring to the cell surface: Gram-positive bacteria S-layers from most Gram-positive bacteria are considered to remain surface associated by binding to peptidoglycan ( P G N ) and/or lipoglycans, two molecules that are abundant in the cell wall of this type of organism (59). Specifically, Gram-positive S-layers are thought to bind to the carbohydrates moieties of these molecules, and therefore S-layers can be thought of as surface-located carbohydrate binding proteins (59). The ability of Slayers to bind to these carbohydrates on the cell surface is often a consequence of the presence of a particular amino acid motif on the S-layer protein, the so-called Surface Layer Homology ( S L H ) domain (39). The S L H domain, found in many Gram-positive S-layer proteins, is a conserved sequence of approximately 55 amino acids, usually repeated i n tandem 3 times (39).  The S L H domain is proposed to function as a  peptidoglycan ( P G N ) binding structure (39), or more generally, a carbohydrate binding structure (19). A s an example, the S-layer protein of  Thermus thermophilus has an S L H  domain that anchors this S-layer protein to P G N found i n the cell wall of this Gram-  15  positive organism (47).  One or more copies of the S L H domain has also been identified  in at least 40 exo-enzymes and exo-proteins (19, 21, 37, 41, 53, 54). S L H domains present on these S-layer-associated proteins are thought to mediate the attachment of the S-layer-associated protein to the S-layer itself (19).  B o t h the amylase of  B.  stearothermophilus and the pullulanase of C. thermosulfuro genes anchor to their respective S-layers via their S L H domains (5).  1.16 - Mechanism of S-layer anchoring to the cell surface: Gram-negative bacteria Little information is available regarding the cell surface anchoring of S-layers of Gram-negative bacteria. Most evidence suggests the involvement of the N-terminus of the S-layer protein monomers. The S-layer of A. salmonicida is thought to be anchored to the cell surface via the N-terminus of the S-layer subunits because the N-termini of these S-layer subunits are inaccessible to trypsin digest when S-layer-proficient A. salmonicida cells are treated with this protease (15). However, mutagenesis studies have yet to be performed to confirm this hypothesis. Similarly, the C. fetus S-layer is probably anchored to lipopolysaccharide on the cell surface via the N-terminus of the C. fetus Slayer monomer, SapA. The evidence for this is that deletions in the SapA N-terminal region result in the loss of SapA anchoring to the cell surface (16), and C-terminal SapA truncations retain the ability to reattach to SapA-deficient C. fetus cells (78). anchoring of the C.  The  crescentus S-layer protein R s a A is perturbed by N-terminal mutations  and truncations (8), and is discussed in greater detail below.  16  1.17 - Tolerated insertions in RsaA The  C. crescentus S-layer has been engineered to display foreign peptides inserted  within the R s a A sequence (7, 8, 46). Sites in  rsaA have been prepared for insertion of  heterologous material at gene positions corresponding to R s a A amino acids 266, 622, 690, 723, 784, 860, and 944, and heterologous sequences have been displayed at these positions (reviewed in (46)). Since heterologous insertion at the above-mentioned sites does not impair R s a A anchoring or crystallization, it is likely that those insertion-tolerant R s a A regions are not involved in R s a A anchoring or crystallization.  Recent R s a A  mutagenesis studies suggest that amino acids 229, 316, 450 and 510 can be mutated without causing adverse affects on R s a A attachment or crystallization, and so it is likely that these R s a A regions are also not involved in R s a A attachment or crystallization (7, 8).  1.18 - The N-terminal region of RsaA The function of the N-terminal region of R s a A may be to anchor R s a A to the  C.  crescentus cell surface, since R s a A N-terminal truncations (9) and mutations (8) result in the shedding of R s a A into the culture medium. Indeed, it is on this premise that proteins of interest can be collected and easily purified from  C. crescentus cultures: the secreted  hybrid proteins (that lack the R s a A N-terminal sequence) aggregate i n the culture supernatant, and are not found to be cell surface-associated.  Thus it appears that  perturbations in the R s a A N-terminal region disrupt R s a A anchoring. However, the exact R s a A N-terminal regions or residues that interact with the cell surface to effect R s a A  17  anchoring remain unknown. Additionally, it is unknown i f there are additional R s a A regions other than the N-terminal region that might mediate R s a A anchoring.  1.19 - Summary of the study Shedding of the  C. crescentus S-layer occurs when the putative anchoring domain is  removed (RsaA N-terminal truncations shed into the culture medium (9). However, since disruption of the R s a A anchoring domain and disruption of R s a A crystallization (subunitsubunit interaction) domain both result in the same phenotype (S-layer shedding), when a particular mutation results in a shedding phenotype, it is difficult to know whether R s a A anchoring or R s a A subunit-subunit interaction has been perturbed.  T o focus on R s a A  anchoring rather than R s a A crystallization (subunit-subunit interaction), we developed an assay where small R s a A fragments are incubated with S-layer-negative cells to assess the ability of the fragments to re-attach. In this way we investigated the possibility of other regions in R s a A that might contain anchoring information. W e found that only the N terminus of R s a A contained anchoring information, and that the smallest S-layer fragment sufficient for R s a A attachment was 277 amino acids, which is ~ l / 4 of the size of full-length R s a A . This small R s a A fragment would not be expected to retain its cell surface crystallization (subunit-subunit) capabilities, since it is missing -3/4 of the native R s a A . Therefore this small R s a A fragment likely directly anchors to the cell surface, rather than being surface-associated  by interacting with other R s a A  monomers.  Consequently, loss of anchoring mutations created in this small R s a A fragment can be attributed to perturbations in the R s a A anchoring domain, rather than perturbations in the R s a A crystallization domain.  18  M y results suggest that the N-terminus of R s a A mediates attachment of the  C.  crescentus S-layer, and even minor perturbations within the first - 2 2 0 R s a A residues cause loss of R s a A anchoring. The project also shows that R s a A anchoring requires smooth lipopolysaccharide (SLPS). A s a by-product of the S-layer anchoring studies, we refute the previous notion that the S-layer Associated Protease, Sap, is intracellular. Results from this work suggest that Sap is in fact an outer membrane-bound protease, and furthermore, it is probably secreted by the S-layer Type I secretion system.  19  2. MATERIALS AND METHODS  2.1 - Bacterial strains, plasmids. and growth conditions A l l of the strains and plasmids used in this study are listed in Table 1. E. coli D H 5 a was used for all E. coli cloning manipulations. E. coli was grown at 37°C in Luria broth (1% tryptone, 0.5% NaCl, 0.5% yeast extract) with 1.3% agar for plates. C. crescentus strains were grown at 30°C in P Y E medium (0.2% peptone, 0.1% yeast extract, 0.01% CaCl , 2  0.02% M g S 0 ) with 1.2% agar for plates. Ampicillin, kanamycin, and streptomycin 4  were used at 50 pg/ml, and chloramphenicol was used at 20 p-g/ml in E. coli cultures. Kanamycin and streptomycin were used at 25 pg/ml and chloramphenicol was used at 2p,g/ml in C. crescentus cultures, when needed.  2.2 - Plasmid and DNA manipulations Standard methods of D N A manipulation and isolation were used (56). Electroporation of C. crescentus was performed as previously described (27).  A l l P C R products were  generated using Platinum Pfx D N A polymerase (Invitrogen, Burlington, ON) following the manufacturers suggested protocols.  2.3 - C. crescentus expression vectors pUC8CVXASDA19 The peptide display vector p U C 8 C V X A S D is a pUC8-based vector that contains a fulllength copy of the rsaA gene under the control of a modified rsaA promoter. It contains an origin of replication, oriV, which allows for replication in both E. coli strains and C.  20  Strain or p l a s m i d  Relevant characteristics  Reference  ATCC 19089  JS1001  A p syn -1000; variant of wild-type strain CB15 that synchronizes well S-LPS mutant of NA1000, sheds S-layer into medium  JS1003  NA1000 with rsaA  interrupted by KSAC K m cassette  [17]  JS1004  JS 1001 with rsaA  interrupted by KSAC Km' cassette  [17]  JS1008'  Cmr, NA1000, rsaFb -negative  [71]  JS1010  NA1000, rsaFa -negative strain  This study  JS1011 JS1012  ,Cmr, NA1000, rsaFa -negative, rsaFb -negative, rsaA -negative This study strain JS1001, Sap-negative, S-layer-negative This study  JS1013  NA1000, S-layer-negative (amber codon in rsaA )  This study  JS1014  This study  JS4015  NA1000, S-layer-negative (amber codon in rsaA ) with manB interuption rendering this strain SLPS-neaative NA1000, rsaA knocked out deleting rsaA promoter and portion of rsaA NA1000 with Tn5 insertion in manB rendering this strain SLPSnegative NA1000 with Tn5 insertion in manB rendering this strain SLPSnegative Spontaneous RsaA-negative mutant of strain CB2 maintained in the laboratory of J. Smit Sap-negative UV-NTG mutant of strain JS4000  JS4022  JS4015, recA -negative. repBAC -positive  This study  JS4023  JS4000, rsaFa -negative strain  This study  JS4024  Kanr, JS 4015 with manB interuption rendering this strain SLPS This study negative JS 4000, rsaFa -negative and rsaFb -negative This study  Bacterial strains C. crescentus NA1000  r  CB155rsaA CB15ATn5F9 CB15ATn5F23 JS 4000  JS4025  r  [17]  This study [2] [2] [66] [72]  E. coli Invitrogen  DH5ct  recA endA  Rb404  F-dam-3, dam-6, metB1, galK2, galT22lacY1, 78. mtl-1. suoE44  thi-1, tonA31, tsx-  [13]  Plasmids [8]  pWB9:rsa/\ AP  Cmr, Smr, pKT215-derived expression vector incorporating the rsaA promoter Cm', Sm , rsaA gene and rsaA promoter  pWB9:rsa4 AP:B162MCCA  Cmr, Smr, pWB9.rsaAAP with collagenase site at a.a. 162  This study  pWB9:rsaA AP:Hps1MCCA  Cmr, Smr, pWB9:rsaAAP with collagenase site at a.a. 277  This study  pWB9:rsaA AP:Hps4MCCA  Cmr, Smr, pWB9:rsaAAP with collagenase site at a.a. 690  This study  Cmr, Smr, pWB9:rsaAAP with collagenase site at a.a. 723  This study  site at a.a. 7 and  This study  linker at a.a. 29 and  This study  linker at a.a. 69 and  This study  site at a.a. 154 and  This study  linker at a.a. 169 and  This study  linker at a.a. 222 and  This study  pBSKII  Cmr, Smr, pWB9:rsaAAP with BamHI collagenase site at a.a. 277 Cmr, Smr, pWB9:rsaAAP with BamHI collagenase site at a.a. 277 Cmr, Smr, pWB9:rsaAAP with BamHI collagenase site at a.a. 277 Cmr, Smr, pWB9:rsaAAP with BamHI collagenase site at a.a. 277 Cmr, Smr, pWB9:rsaAAP with BamHI collagenase site at a.a. 277 Cmr, Smr, pWB9:rsaAAP with BamHI collagenase site at a.a. 277 Apr, ColE1 cloning vector; lacZ Ap'  pTZ18UCHE:rsafib ANAC  Cmr, cloning vector, rsaFb missing N and C termini  [71]  pAL1  Smr, Sues, E. coli-based pNPTSI 38 suicide vector with A-rsaA fragment  This study  pWB9  p\NB9:rsaA  AP:Hps12MCCA  pWB9:rea4 AP:B7Hps1MCCA pWB9:rsaA AP:Taq29Hps1 MCCA p\NB9:rsaA  AP:Mps4Hps1MCCA  pWB9:rsa/A AP:B154Hps1MCCA pBBR3ASD:Taq169Hps1 MCCA pWB9:rsaA AP:B222Hps1MCCA  r  [6]  Stratagene  pK18mobsacB  Kmr, Sues, E. coli-based suicide vector  [60]  pK18mobsacB:rsaFa AKP  Kmr, Sues, E. coli-based suicide vector with RsaFa internallydeleted between Kpnl and Pst I sites Kmr, Sues, E. coli-based suicide vector with manB fragment missing N and C termini Cmr, Smr, pWB9:rsaAAP with 112 amino acid segment of VP2 qlvcoDrotein of IPNV at RsaA a.a 723 Apr,, Cmr pUC9 vector carrying a modified multiple cloning site; a promotorless Cm-resistance gene from Tn9 is inserted in the Bgl II site, and a multiple cleavage cassette is inserted into the Xhol-Stul MCS Smr, repBAC -positive, modified rsaA promoter  This study  pK18mobsacB:manB ANAC pWB9:rsaAAP(723/VP2CA) PUC9CXSMCC  pBBR3ASD puc8CVXASDA19 pBSKI+ v  pK18mobsacBrea/A 3 5 3 0 B pBSkllESH pK18mobsacBpw£> A R pBBR3 pBSkllEEH  This study [72] This study  This study  Cmr, high copy number puc8-based E.coli/C.crescentu s shuttle This study vector with modified rsaA promoter and wild tvoe rsaA aene Apr, LacZ, cloning vector Stratagene Kmr, Sues, E. coli-based suicide vector with rsaA with amber codon at a position corresponding to RsaA a.a. 353 Apr, modified pBSkll cloning vector with EcoRI- Stul-, Hindlllmodified MCS Kmr, Sues, E. coli-based suicide vector with 200 bp internal deletion of S-laver associated protease aene (sap ) Smr, broad host range vector  This study [71] This study [31]  Apr, modified pBSkll cloning vector with EcoRI- EcoRV-, Hindlll- [71] modified MCS  crescentus strains that express the repBAC genes (73). It was constructed by D r . John N o m e l l i n i as follows: the oligonucleotides I 1060 5' G A G G C C T A C T C T T C C T T T T T C A A T A T T A T T G A A 3' (StuI underlined) and 1920 5' G A G G C C T A G T A C T C T G T C A G A C C A A G T T T A C T C A T A 3' (Seal site underlined) were used to do inverse P C R on the plasmid p U C 8. This P C R product was digested with Stul and Seal and ligated to Hpal digested chloramphenicol gene called C H E . The resulting plasmid was called p U C 8 C X . The C H E gene was made by using the oligonucleotides J N C H E - 1 5' G G A A G A T C T G T T A A C T T T T C A G G A G C T A A G G A A G C T 3' and J N C H E - 2 5' G G A A G A T C T G T T A A C A C A A T A A C T G C C T T A A A A A A A T T A 3' (Hpal sites are underlined) to P C R the chloramphenicol gene from the plasmid p M M B 2 0 6 (44), which does not have an E c o R I site in the middle of the gene. The o r i V was inserted by first cutting the p U C 8 C X plasmid with EcoO109, filling in the recessed ends resulting from this digestion with the Klenow enzyme, recircularizing the plasmid by ligation, then digesting the resulting plasmid with Hindlll.  This Hindlll/Hindlll  fragment was  removed and replaced with the 521bp Hindlll/Xmnl  o r i V fragment from plasmid pCR2.1  o r i V (73), and the resulting plasmid was called p U C 8 C V X . The next step was to remove the lac promoter and replace it with a modified rsaA promoter. This was done in the follow manner. The EcoRUHindlll  fragment from plasmid pSSa49ASD (10) containing  the modified rsaA promoter was cloned into p U C 8 cut with E c o R I and Hindlll,  called  p U C 8 A S D . Then both this new p U C 8 plasmid and p U C 8 C V X were cut with E c o R I and ligated together and selected on A m p / C m plates. The correct orientation of the plasmids was determined with an Ndel digest.  The correct plasmid fusion was then  digested with Sapl and Pstl, filled in with the Klenow enzyme and ligated back together  22  and selected on C m plates. This last manipulation removed almost all of the p U C 8 A S D plasmid except the modified  rsaA promoter with is now upstream of the EcoRUHindlll  multiple cloning site. This plasmid was called p U C 8 C V X A S D . A l l previous full length R s a A gene constructs can be easily cloned into this vector as and w i l l be transcribed off the modified  EcoRUHindlll fragments  rsaA promoter. The w i l d type rsaA gene was  excised from p T Z 1 8 U B : r , s a A A P (8) as an  EcoRUHindlll fragment and cloned into  EcoRUHindlll cut p U C 8 C V X A S D , resulting in the plasmid p U C 8 C V X A S D A 1 9 .  pBBR3ASD The p B B R 3 A S D vector (constructed by D r . John Nomellini) was used as a shuttle E.  coli/C. crescentus shuttle vector, and does not require the repBAC genes to be integrated into the chromosome of the C.  crescentus host cell in order to replicate, in contrast to the  p U C 8 C V X A S D A 1 9 vector described above. The modified  rsaA promoter was included  in this vector, which results in increased R s a A production (10). T o create this vector, the plasmid p U C 8 C V X A S D A 1 9 (which contains the modified with  rsaA promoter) was digested  Nspl and the recessed ends were filled in using the Klenow enzyme. The resulting  linearized, blunt-ended plasmid was ligated into p B B R 3 (31) cut with Smal, creating a - 1 2 kb plasmid fusion. Correct orientation of the plasmid fusion was confirmed by E c o R I digestion, where the desired clone gave a - 6 kb and a 6.7 kb band upon E c o R I digestion.  The plasmid fusion with the correct orientation was then digested with  Hindlll, releasing most of the p U C 8 C V X A S D A 1 9 plasmid, but leaving the modified  rsaA promoter with the p B B R 3 vector. The remaining fragment containing the modified  23  rsaA promoter with the p B B R 3 vector was circularized by ligation, and this resulted in the creation of p B B R 3 A S D .  2.4 - Construction of plasmids carrying rsaA with collagenase cleavage sites pUC9CXSMCC p U C 9 C X S M C C was created by ligating the annealed oligonucleotides 5 ' - T C G A G G C A T G A T C G A G G G T C G C G G C C C G C A C G G T C C C G C C G G C C C G G - 3 ' and 5'C C G G G C C G G C G G G A C C G T G C G G G C C G C G A C C C T C G A T C A T G C C-3' into the p U C 9 C X S plasmid (7) cut with Xhol-Stul, inserted in the same orientation as the promotorless Cm-resistance gene.  These oligonucleotides encode, from 5'-3', a Factor  X protease cleavage site, followed by two collagenase cleavage sites i n tandem (Multiple Cleavage. Cassette, M C C ) .  pWB9:rsaAAP:HpslMCCA, pWB9:rsaAAP:Hps4MCCA, and pWB9:ra*AAP:Hpsl2MCCA p U C 9 C X S M C C was cut with BamHI, releasing a BamHl-BamHl the M C C + promotorless Cm-resistance gene. ligated into p T Z 1 8 U B \ r s a A {HinVllll  fragment that included  The entire BamHI  cassette was then  BamHI) (7) to place the M C C at the position in  rsaA corresponding to amino acid 277, p T Z 1 8 U B : r s a A (HinPI690BamHI)  (7) to place  the M C C at the position in rsaA corresponding to amino acid 690, or p T Z 1 8 U B : r s a A (//mPI723BaraHI) (7) to place the M C C at the position in rsaA corresponding to amino acid 723. Proper orientation of the M C C achieved by selecting for clones with C m resistance (the C m gene is driven by the lac promoter in the p T Z plasmid i f the C m -  24  resistance cassette was oriented properly). The Cm-resistance genes were then excised by cutting with ligation. The  Bglll, and subsequent circularization o f the plasmids was achieved by  EcoRl-Sstl fragments of the resulting plasmids were excised and ligated  into the p W B 9 plasmid that had been cut with EcoRl-Sstl. This resulted i n the creation of  plasmids  pWB9:rsaAAP:HpslMCCA,  p W B 9 : r s a A A P:Hps4MCCA,  and  pWB9:rcaAAP:Hpsl2MCCA.  pWB9:rsaAAP:B162MCCA The  rsaA EcoRllClal fragment was excised from p T Z 1 8 U B : rsaA (HinPl723BamHT) (7)  that was harbored by and subsequently derived from R B 4 0 4 (13), a n o n - D N A methylating strain of  E. coli that allowed the Clal digestion to proceed. This rsaA  EcoRl/Clal (wild type) fragment was then ligated into p B S K I I cut with E c o R I and Clal. T o destroy the BamHI site i n the M C S o f p B S K I I , the p B S K I I plasmid was cut with  BamHI, and the resulting overhangs were filled i n using the K l e n o w enzyme, then the blunt ends were ligated to circularize the plasmid. Mutation at R s a A amino acids 162/163 was achieved using the Quickchange method (Stratagene, L a Jolla, C A ) , following the manufacturer's suggested protocols, using the primers 5 ' - G T T G G C C T G G C G G C T G G A T C C A G C C A C G G C G G C C G C - 3 ' and 5 ' - G C G G C C G C C G T G G C T G G A T C C A G C C G C C A G G C C A A C - 3 ' where the bases that were non-complementary to the template are shown i n bold.  This changed R s a A amino acids F 1 6 2 / L 1 6 3 to  G162/S163, and created a BamHI site at a position corresponding to R s a A amino acids 162/163. The M C C cassette from p U C 9 C X S M C C was inserted into this BamHI site and the  Cm-resistance  gene  was  removed  as  described  above,  yielding  25  pBSKll^Bamlll:rsaAEcoRl-ClaIBamHI162. The plasmid p u c 8 C V X A S D A 1 9 (described above) was cut with  PstI, releasing a Pstl-Pstl fragment containing two undesired Clal  sites from a non-coding region of the plasmid. The remaining backbone plasmid was recircularized by ligation, creating puc8CVXASDA19<|)P,stI. This plasmid, derived from  E. coli strain R B 4 0 4 to allow for Clal digestion, was cut with E c o R I and Clal, and then the  rsaA EcoRl/Clal  fragment  w i t h the desired mutation,  pBSlQl^Bamm:rsaAEcoRl-ClalBamlill62  with E c o R I and  ligation, creating puc8CVXASDA19<])PstI£amHI162. fragment (that contained the  from digestion o f  Clal, was inserted by  Next, to replace the Pstl-Pstl  Clal sites that were now inconsequential, but that also  contained an Sstl site required for the next step of the cloning strategy), the AvrlAIHindlll fragment  (containing within it the  Pstl-Pstl fragment) was obtained from cutting  p u c 8 C V X A S D A 1 9 w i t h A v r l l and Hindlll,  and this fragment  was ligated into  p u c 8 C V X A S D A 1 9 0 P 5 r I 5 a m H I 1 6 2 cut with A v r l l and Hindlll,  g i v i n g the plasmid  p u c 8 C V X A S D A 1 9 £ a m H I 1 6 2 . Finally, the mutant roaA-encoding  EcoRl/Sstl fragment  was excised from puc8CVXASDA19<|)PsflitamHI162 and ligated into the p W B 9 plasmid that  had been  cut  with  EcoRl-Sstl.  This  resulted  i n the creation o f  pWB9:raaAAP:B 162MCCA.  2.5 - Construction of plasmids carrying rsaA with collagenase cleavage at RsaA position 277 with additional mutations in the RsaA N-terminus pWB9:rsaAAP:B7HpslMCCA, pWB9:rsaAAP:B154HpslMCCA, and pWB9:rsaAAP:B222HpslMCCA  26  The p T Z 1 8 U B : r s a A (HinPlTll  BamHI) plasmid with the M C C cassette at R s a A position  277 (described above) was used as the template for P C R for site-directed mutagenesis of rsaA to achieve the mutation at R s a A position 7, 154, and 222. Mutation at R s a A amino acid 7 was achieved using the Quickchange method (Stratagene, L a Jolla, C A ) , following the manufacturer's suggested protocols. The mutation at amino acid 7 was generated using the primers 5 ' - G C C T A T A C G A C G G C C G G A T C C G T G A C T G C G T A C A C C - 3 ' and 5 ' - G G T G T A C G C A G T C A C G G A T C C G G C C G T C G T A T A G G C - 3 ' , where the bases that were non-complementary to the template are shown in bold. This changed R s a A amino acids Q 7 / L 8 to G7/S8.  The mutation at amino acid 154 was  generated using the primers 5 ' - G C G A C C G C C G C T G G C G G A T C C G T C G C G G C C G C C G T G - 3 ' and 5 ' - C A C G G C G G C C G C G A C G G A T C C G C C A G C G G C G G T C G C - 3 ' , where the bases that were non-complementary to the template are shown in bold.  This changed R s a A amino acids V154/D155 to G154/S155. The mutation at  amino acid 222 was generated using the primers 5 ' - G C C G C G A T G A T C A A C G G A T C C T C G G A C G G C G C C C T G - 3 ' and 5 ' - C A G G G C G C C G T C C G A G G A T C C G T T G A T - C A T C G C G G C - 3 ' , where the bases that were non-complementary to the template are shown in bold. G222/S223.The EcoRl-Sstl  This changed R s a A amino acids D 2 2 2 / L 2 2 3 to  fragments from the resulting Quikchange products were  excised and ligated into the p W B 9 plasmid that had been cut with EcoRl-Sstl, pWB9:rsoAAP:B7HpslMCCA,  p W B 9 : r s a A A P:B154HpslMCCA,  creating and  pWB9:ra*AAP:B222HpslMCCA.  27  pWB9:rsaAAP:Taq29HpslMCCA, pWB9:rsaAAP:Mps4HpslMCCA The EcoRl-Notl  fragments  from  pTZ18UB\rsaA  (Taql29BamHl) (8) or the  p T Z 1 8 U B : r a z A (Mspl69BamHl) (7) were excised and ligated into p T Z 1 8 U B : r s a A (HinPimBamHI)  plasmid with the M C C cassette at R s a A position 277 (described  above) that had been cut with EcoRl-Notl.  The rsaA EcoRUSstl fragments were then  excised from the resulting plasmids and subsequently ligated into the p W B 9 plasmid that had  been  cut w i t h EcoRl-Sstl, creating p W B 9 : r s a A A P : T a q 2 9 H p s l M C C A and  pWB9:rsoAAP:Mps4HpslMCCA.  pBBR3ASD:Taql69HpslMCCA The p T Z 1 8 U B : r s a A {HinVYlllBamHI) plasmid with the M C C cassette at R s a A position 277 (described above) was cut with Pstl and then recircularized by ligation, eliminating two undesired Clal sites in the released non-coding Pstl-Pstl fragment, creating plasmid p T Z 1 8 U B : rsaA (//mPI2775amHI)MCCOPsfI.  The rsaA EcoRUClal fragment from  RB404-derived p T Z 1 8 U B : r s a A ( T a g i l 69flamffl) (8) was excised and then ligated into RB404-derived, EcoRUClal cut pTZ18UB:rsaA (#mPI277flamHI)MCCOPs/I. From the resulting plasmid, the rsaA EcoRllHindlll  fragment was excised and ligated into  EcoRVHindlll cut p B B R 3 A S D , creating p B B R 3 A S D : T a q l 6 9 H p s l M C C A .  2.6 - Construction of plasmids used for gene disruptions pALl p A L l plasmid construction was carried out by A s s a f L e v i .  P l a s m i d p A L l was  constructed in order to create an in-frame deletion of the complete rsaA coding region. A  28  P C R product encoding a 1.0 k B region upstream of the  rsaA gene was created using  N A 1 0 0 0 chromosomal D N A as the template and using the primers 5 ' - G G A T C C G G C G T T C G A G C T G C T G C T G A - 3 ' and 5 ' - G A A T T C T C A C C T G G C G G G T G A G T G A G - 3 ' , introducing BamHI and E c o R I sites. Another P C R product was created using the primers 5 ' - G A A A T T C C G C T C G C C T A A G C G A A C G T C - 3 ' and 5 ' - A C T A G T G G C C G A G A T C T T G C C G T C G A - 3 ' , amplifying a 1.0 k B region containing the end of  rsaA and incorporating E c o R I and Spel sites. Fragments were ligated into the p G E M 5ZT(+) vector at the E c o R V site using the p G E M - T ® easy kit (Promega). The resulting fragments were digested with E c o R I ,  BamHI, and Spel, and ligated into BamHI and Spel  cut p N P T S (32) plasmid. This resulted in the creation of p A L l , which was transformed by electroporation into the E .  coli D H 1 0 B strain (Invitrogen) and selected by blue-white  screening.  pK18mobsacBmanl?ANAC This plasmid was created by D r . Peter A w r a m , and was used to create a strain that was null for Sap. A P C R product encoding  manB that was deleted in the regions encoding the  N - and C-termini of M a n B was generated using N A 1 0 0 0 chromosomal D N A and the primers 5 ' - C C T G G G T C T G G G A A C C T A T A T C C - 3 ' (ManB 169) and 5 ' - C A G T G C G G G C T C A T G G T C A G - 3 ' (IManB 1202), and then blunt-end ligated into E c o R V - c u t p B S K I I E E H (71). The  EcoRUHindlll fragment from the resulting plasmid (containing  the desired deleted form of  manB was then ligated into EcoRI/Hindlll-cut pK18mobsacB  (60), giving pK18mobsacBman5ANAC.  29  pK18mobsacB :pwb A R This plasmid, created by Theo Blake, was used to create an internal deletion i n the Slayer-Associated Protease gene, sap (formerly known as pwb). A P C R product encoding sap was created using JS4000 chromosomal D N A as the template and the primers 5'C C G C C C G A G C G A G C G C T G T G C G A A C - 3 ' and 5'- A C C T T T T C G G G G G A G G G C C G C C C G C - 3 ' and then blunt-end ligated into E c o R V - c u t p B S K I I (Stratagene). The  sap gene was then excised from the resulting plasmid as an EcoRl/Hindlll fragment,  and ligated into  EcoRI/Hindlll-cut p B S K T I E S H (71). The resulting plasmid was cut with  Pstl, which removed 1023 bp of the sap sequence (approximately 1/2 of the sap gene), and the remaining plasmid was recircularized by ligation. The internally-deleted form of sap was then excised as an EcoRl/Hindlll fragment, and ligated into  EcoRI/Hindlll-cut  pK18mobsacB, creating pK18mobsacB:pw&AR.  pK18mobsacB :rsaFaAKP The plasmid, created by M i k e Toporowski, was used for the internal deletion of and contained  some  o f the rsaFa flanking regions to encourage  rsaFa,  homologous  recombination. A P C R product containing the rsaFa gene and flanking regions of 1008 bp 5' and 139 bp 3' was generated using N A 1 0 0 0 chromosomal D N A as the template, and the primers 5 ' - G C C A C G C C C G G C G T C C A G T C C G A - 3 ' and 5 ' - G A G C T C C C T A G A G C G T T C T C C G A T C C G T G C G - 3 ' . This fragment was blunt end ligated into the p B S K I + plasmid (Stratagene) at the EcoRV site and the resulting construct was called pBSKLrsaFaEX.  The pBSKV.rsaFaEX plasmid was digested with  Kpnl and Pstl and  30  blunt-ended using T 4 D N A polymerase and ligated. The resulting plasmid has an 852 bp deletion. fragment,  The internally-deleted version of rsaFa was excised as an EcoRl-Hindill and then  ligated  into  EcoRl-Hindill  cut p K 1 8 m o b s a c B , creating  p K 18mobsacB: rsaFa&KP.  pK18mobsacBrxaA3530B This plasmid was created by D r . John Nomellini. The BamHI site i n p T Z 1 8 U B : r s a A (AdI353BamHI) (7) was destroyed by cutting the plasmid with BamHI, filling i n the recessed ends with the Klenow enzyme, then ligating the Klenow-blunted ends together to recircularize the plasmid. This resulted i n putting the rsaA sequence out of frame, introducing an early stop codon at a position corresponding to R s a A amino acid 358. The  EcoRUHindlll  fragment  containing the mutated rsaA gene was then ligated into  EcoRUHindlll cut pK18mobsacB, creating pK18mobsacBr5aA3530B.  2.7 - Construction of gene disruptions JS1010 Knockout  o f rsaFa  in NA1000  was done by M i k e T o p o r o w s k i using the  pK18mobsacB: rsaFaAKP plasmid. Primary recombination of the plasmid was selected for using Km-resistance. Three consecutive sub-culturing events were done to encourage a second recombination event.  Secondary selection on 5% sucrose P Y E plates and  subsequent replica plating on P Y E and P Y E K m plates was used to confirm a second recombination event. Colonies were then screened using the primers 5 ' - C G C C G G C T T C G C A G C G A T G A C C C - 3 ' and 5 ' - C C C G G A G G C C T C C C A G G C G G C G T A - 3 ' to  31  confirm that the appropriate gene replacement occurred. A strain confirmed to possess only the internally-deleted form of rsaFa was designated JS1010.  JS1011 Knockout o f rsaFa i n JS1008 (70) was done as described above for the knockout of rsaFa in N A 1 0 0 0 , except that colonies determined to have had a second recombination event were first screened by colony western (7) using R s a A antiserum. Colonies that were S-layer negative according to the colony western were further screened by P C R using the same primers as described above to confirm that the appropriate gene replacement occurred. A strain confirmed to possess only the internally-deleted form of rsaFa was then subjected to the knocking out o f rsaA, using the p A l l plasmid. Primary recombination of the plasmid was selected for using Km-resistance, as w e l l as C m resistance resulting from the Cm-cassette already integrated into the chromosome as a result of the rsaFb knockout in this strain. Three consecutive sub-culturing events i n the presence of C m (to maintain the rsaFb deletion) but not K m were done to encourage a second recombination event. Secondary selection on 5% sucrose P Y E C m plates and subsequent replica plating on P Y E C m and P Y E K m C m plates was used to confirm a second recombination event. Colonies were then screened by P C R using the primers 5 ' G C G G C G G A G G T C T T G C A C C T - 3 ' and 5 ' - C A T C T G G A T C G G G T T C T T G G T G-3'.  A strain confirmed to possess only the internally-deleted form o f  rsaA was  designated JS 1011.  32  JS1012 The plasmid pK18mobsacB:pwoAR was used by Andrea Prusic to knockout sap in JS1001, replacing the w i l d type gene with and internally-deleted version of sap that was missing -1/2 of the sap sequence. Primary recombination of the plasmid was selected for using Km-resistance. Three consecutive sub-culturing events were done to encourage a second recombination event.  Secondary selection on 5% sucrose P Y E plates and  subsequent replica plating on P Y E and P Y E K m plates was used to confirm a second recombination event. Gene replacement was confirmed by P C R using the primers 5'C C G C C C G A G C G A G C G C T G T G C G A A C - 3 ' and 5'- A C C T T T T C G G G G G A G G G C C G C C C G C - 3 ' . Next, the  rsaA gene was knocked out using the p A L l plasmid.  Primary recombination of the plasmid was selected for using Km-resistance.  Three  consecutive sub-culturing events were done to encourage a second recombination event. Secondary selection on 5% sucrose P Y E plates and subsequent replica plating on P Y E and P Y E K m plates was used to confirm a second recombination event. Colonies were then screened by P C R using the primers 5 ' - G C G G C G G A G G T C T T G C A C C T - 3 ' and 5 ' - C A T C T G G A T C G G G T T C T T GG.T G - 3 ' . A strain confirmed to possess only the internally-deleted form of  rsaA was designated JS1012.  JS1013 Dr. John N o m e l l i n i used pK18mobsacBrsaA353d>B to knockout replacing the w i l d type  rsaA in N A 1 0 0 0 ,  rsaA gene with rsaA containing an early stop codon.  Primary  recombination of the plasmid was selected for using Km-resistance. Three consecutive sub-culturing events were done to encourage a second recombination event. Secondary  33  selection on 5% sucrose P Y E plates and subsequent replica plating on P Y E and P Y E K m plates was used to confirm a second recombination event. Colonies determined to have had a second recombination event were screened by colony western (7) using R s a A antiserum.  Colonies that were S-layer negative were further determined to have no  surface-presented R s a A by low p H extraction (see protein techniques below).  JS1014 p K 1 8 m o b s a c B m a n 5 A N A C was used to knockout  manB i n JS 1013, leaving two non-  functional copies of the gene in the chromosome, one missing the part of the gene encoding the M a n B N-terminus and one missing the part of the gene encoding the M a n B C-terminus. Recombination of the plasmid p K 1 8 m o b s a c B m a « 5 A N A C was selected for using Km-resistance.  A secondary recombination event was not sought after, thus  stability of the primary recombination event was assured by always using K m when growing JS 1014.  CB15ArsaA. Knockout of  rsaA in N A 1 0 0 0 to create CB15ArsaA was done using the plasmid p A L l by  Assaf L e v i . Delivery of the p A L l plasmid into N A 1 0 0 0 was done by conjugation with the  E. coli L S 9 8 0 (match maker) and M T 6 0 7 helper strain (D. A l l e y ) . The helper strain  utilizes vector p R K 6 0 0 , a derivative of p R K 2 0 1 3 , Cm-resistance, containing a Tn9 insertion, C o l E l ori, and tra functions from p R K 2 0 1 3 (62).  Gene replacement was  confirmed by P C R analysis (not shown) and the resulting strain was named  CB15ArsaA.  34  CB15ATn5F9/CB15ATn5F23  The C B 1 5 A Tn5 mutants F9 and F23 were made/confirmed as previously described (2).  JS4022 JS4022 was constructed by D r . John Nomellini in the Sap-negative strain JS4015. The  repBAC genes were introduced into the recA gene i n the same way that resulted i n the construction of JS4019, described previously (73).  JS4023 Knockout  o f rsaFa  i n JS4000  was done b y M i k e  T o p o r o w s k i using the  p K 1 8 m o b s a c B : r m F a A K P plasmid as described above for the creation of JS1010.  JS4024 Knockout o f manB i n JS4015 (72) was achieved using the p K 1 8 m o b s a c B m a n 5 A N A C plasmid in the same manner utilized to create strain JS1014 (described above).  JS4025 Knockout o f rsaFb i n JS4023 was done via insertional inactivation using an N - and C terminally  deleted  rsaFb  fragment.  The  non-replicating  plasmid  pTZ18UCHE:rsaFZ?ANAC (71) was electroporated into N A 1 0 0 0 competent cells and the Cm-resistance cassette was used for plasmid selection for insertional inactivation. Recombination o f the p T Z 1 8 U C H E : r 5 f l F & A N A C plasmid resulted i n loss o f the full rsaFb gene, leaving independent N-and C-terminally-deleted rsaFb gene fragments.  35  Recombination of the plasmid was confirmed by P C R using the primers 5 ' - G A G G C C T A C T C T T C C T T T T T C A A T A T T A T T G A A - 3 ' and 5 ' - G G A C G A C G C T G A C C A GCACCCCCTGCT-3'.  2.8 - S-layer reattachment assays Co-culturing assay Co-culturing of S-layer donors (JS1001) with S-layer recipients (JS1003 & JS1004) was done as follows. JS 1001, JS1003 and JS1004 were grown to m i d log phase. Aliquots of JS1001 and either JS1003 or JS1004 i n a cellular ratio of 8 donors: 1 recipient were used to inoculate 10 mis of P Y E . These co-cultures were then grown to mid-log phase, then filtered through Whatman 52 hardened filter paper (Whatman, Kent England) to remove any aggregated protein. Equivalent numbers o f cells were then subjected to l o w p H extraction (see Protein techniques below) to assess the extent of S-layer reattachment.  RsaA reattachment assay To  produce  soluble protein  for reattachment purposes,  cultures  o f smooth  lipopolysaccharide (SLPS)-negative cultures (JS1001, for w i l d type R s a A ) , or S L P S negative, S-layer-negative  C. crescentus strains that are Sap-positive (JS1004) or Sap-  negative (CB15ACa5AP6Ar,saA) harboring the plasmids encoding R s a A or R s a A mutants were grown to O D - 0 . 9 , pelleted by centrifugation three times at 13, 000 rpm for 10 min at 4°C; the pellet was discarded after each centrifugation i n an effort to clear the supernatant of any cells. Supernatants were stored at 4°C until needed.  S-layer-negative  target cells possessing various levels of S L P S (JS1003, JS1004, and JS4024) were grown  36  to O D  600  ~ 0 . 9 , then pelleted by centrifugation at 13, 000 rpm for 5 m i n at 4°C. C e l l pellets  were washed with 1ml of P Y E then resuspended with the supernatants containing the soluble R s a A or R s a A mutant. The volume of supernatant used for resuspension was 1.21.5 times the volume of the target cells initially pelleted. The resulting mixtures were incubated at room temperature for 10-30 minutes with slow inversion. These cultures were then pelleted, washed twice with 10 m M H E P E S p H 7 for subsequent l o w p H extraction, or washed twice with 10 m M T r i s - H C l p H 8 for subsequent boiling of the entire sample or for subsequent whole cell protein preparations.  2.9 - Protein techniques Low pH extraction Surface protein from C. crescentus cells was extracted by l o w p H extraction or by EGTA-treatment as previously described (75). Briefly, cell pellets were washed twice using 10 m M H E P E S p H 7.2 and then release of the S-layer was facilitated using 100 m M H E P E S , p H 2.0 (low p H extraction) or 10 m M ethylene glycol bis (P-aminoethyl ether)-N, N , N \ N'-tetraacetic acid ( E G T A ) in 10 m M H E P E S p H 7.2. T o compare the amounts of S-layer protein from C crescentus cells expressing R s a A to S-layer that had been reattached to S-layer-negative cells, equivalent amounts of cells (determined by OD  6 0 0  ) growing at log phase were used and equal amounts of extracted protein samples  were loaded onto protein gels.  37  Whole cell protein preparation Whole cell protein preparations were done with equivalent amounts of cells (determined by O D  6 0 0  ) growing at log phase. The cultures were centrifuged and the cell pellets were  washed twice with 10 m M T r i s - H C l p H 8. The cells were resuspended in 10 m M TrisH C l pH8 and lysozyme (100 pg/ml) was added and the mixture incubated at 25 °C for 15 minutes.  R N a s e A (50 pg/ml) and D N a s e l (1 pg/ml) were added and the incubation  continued at 37°C for 30 minutes. Equal amounts of whole cell protein preparations were loaded onto protein gels. Some samples were subjected to l o w p H extraction before whole cell protein preparation was done to disrupt R s a A monomer-monomer interactions, in order to assess R s a A that was directly anchored to the cell surface.  Whole cell protein preparation by boiling A s an alternative to whole cell protein preparations, equivalent numbers of cells were washed twice with 10 m M T r i s - H C l p H 8 then subsequently boiled for 5 minutes.  Collagenase digests Collagenase digests performed on soluble R s a A carrying a collagenase recognition sequence were done i n P Y E supplemented with collagenase buffer as follows. Supernatant from S-layer shedding C.  crescentus strains containing the desired soluble  R s a A protein were harvested as described above (see R s a A reattachment assay). The resulting culture supernatant was supplemented such that final concentrations of the following substances were achieved: 10 m M N a C l , 4 m M C a C l , 2 m M T r i s - H C l p H 7, 2 2  m M 6-mercaptoethanol). Collagenase was added to 7 U / m l , and digestion was performed  38  for various lengths of time at 37°C, and subsequently stored at 4°C (if required) prior to use.  S D S - P A G E and Western blotting Sodium dodecyl sulfate (SDS)-polyacrylamide gel electrophoresis ( P A G E ) was done using 4% stacking, and 7.5% or 12% (as indicated) separating gels. Coomassie stained S D S - P A G E gels and Western immunoblotting were done as previously described (56). After transfer of proteins to 0.2 p m B i o T r a c e N T nitrocellulose membrane  (Life  Sciences, Pensacola, F L ) , membranes were blocked using 3% skim milk, 0.9% N a C l , and 20 m M T r i s - H C l p H 8 . Western blots were probed with primary rabbit polyclonal antibodies, and antibody binding was visualized by colorimetric developing methods. Colorimetric blotting was done using goat anti-rabbit Ig coupled to horseradish peroxidase and color forming reagents as previously described (66).  Antibodies R s a A A188-784 antiserum (71) was used at a 1/10, 000 dilution and S L P S antiserum (74) was used at a 1/6, 000 dilution for colorimetric detection.  39  3. RESULTS - RsaA anchoring studies 3.1 - Assay development elucidates the importance of RsaA secondary structure for wild type RsaA anchoring Full-length R s a A can be obtained by several different methods. For example, treating cells to low p H conditions extracts R s a A from the cell surface. EGTA-treatment of Slayer possessing  cells also removes  the S-layer.  C e l l s deficient i n smooth  lipopolysaccharide (SLPS) shed their S-layer into the culture medium, and a fraction of this protein aggregates and precipitates out of solution. This protein can be collected by simple filtration, and solubilized by urea treatment. The remaining fraction of this shed S-layer remains soluble in the supernatant of these shedder cultures (this study). This soluble R s a A fraction can be collected by centrifugation of the culture and collection of the supernatant. Our lab had previously found that to achieve in vitro  recrystallization of R s a A onto  lipid vesicles, the best method for R s a A preparation was the low p H extraction (45), but at that time it was not known that soluble R s a A is found in the supernatant of shedder cultures. I wanted to find the best method to achieve in vivo recrystallization of R s a A onto S-layer-negative cells rather than lipid vesicles, since I was trying to investigate R s a A attachment to the cell surface. T o test the ability of R s a A to reattach to S-layer negative cells, I first obtained R s a A by low p H or E G T A extraction (see Materials and Methods) of w i l d type C.  crescentus cells (NA1000). These R s a A preparations were then  incubated with S-layer negative cells that either possessed w i l d type (JS1003) or deficient (JS1004) levels of S L P S (since S L P S is known to be involved in S-layer assembly on the cell surface), in water supplemented with C a C l (since C a 2  2 +  is required for normal levels  40  of S-layer assembly). The results indicated that R s a A obtained by E G T A extraction did not reattach to cells, while R s a A obtained by low p H extraction did reattach to cells in an SLPS-dependent manner (Fig. 2-1).  Unfortunately, S-layer recrystallization did not  occur to w i l d type levels; it appears that S-layer recrystallization occurred to about 10% of wild type S-layer levels (Fig. 2-1). M y next hypothesis was that perhaps S-layer  recrystallization could be optimized by  incubating S-layer protein with cells i n the normal growth media P Y E , rather than i n calcium-supplemented water, and that dialyzing the prepared R s a A might render the protein more amenable to reattachment. T o test this hypothesis, R s a A was first obtained by l o w p H or E G T A extraction of w i l d type C. subsequently dialyzed (or not) against water.  crescentus cells ( N A 1 0 0 0 ) and  These R s a A preparations were then  incubated with S-layer-negative cells that either possessed w i l d type (JS1003) or deficient (JS1004) levels of S L P S , in the normal C.  crescentus growth media, P Y E . The  results show that again S-layer reattachment was achieved in an SLPS-dependent manner (except for the E G T A - e x t r a c t e d R s a A that was not dialyzed against water prior to addition to S-layer-negative cells), but unfortunately, not back to w i l d type levels (Fig. 22). Co-culturing of S-layer donors (strains that lack S L P S and shed their S-layer) and Slayer recipients (strains that possess S L P S but lack S-layer) achieves S-layer recrystallization i n  Aeromonas salmonicida (25).  I thought that perhaps the same  experiment could be done with C. crescentus, and that this method might achieve better S-layer recrystallization than that achieved by the addition of low pH-extracted R s a A to S-layer-negative cells, since the acidic conditions during the low p H extraction might  41  100% A m o u n t of S L P S on cells:  wt  Input R s a A extracted b y :  N/A  10% wt  d  EGTA  wt  cl  low p H  wt N/A  100 k D a  Fig. 2-1. S-layer reattachment in calcium-supplemented water. W i l d type R s a A was extracted from the C. crescentus cell surface by l o w p H or E G T A treatment. Resulting protein was subsequently incubated with equivalent amounts of Slayer negative cells that possess wild type (wt) or deficient (d) levels of S L P S , in calcium-supplemented water. Reattached RsaA was subsequently extracted by low p H treatment. Extractions were separated on a 7.5% S D S - P A G E gel, and R s a A was detected by western blotting using R s a A antiserum. For comparison, the lane labeled 100% is a low p H extraction of an equivalent number of cells expressing w i l d type levels of S-layer, and the lane labeled 10% is a low p H extraction of 10% of the number of cells used in the reattachment assay.  42  A m o u n t of S L P S on cells: Input R s a A extracted b y :  wt  wt  d  N/A EGTA/dial  wt EGTA  d  wt  d  low pH/dial  wt  d  low p H  *  RsaA  100 kDa  Fig. 2-2. S-layer reattachment in PYE. W i l d type R s a A was extracted from the C. crescentus cell surface by low p H or E G T A treatment, and the resulting protein was dialyzed (dial) or not against dH20. These R s a A preparations were subsequently incubated with equivalent amounts of S-layer negative cells that possess w i l d type (wt) or deficient (d) levels of S L P S . Reattached R s a A was subsequently extracted by low p H treatment. Extractions were separated on a 7.5% S D S - P A G E gel, and RsaA was detected by western blotting using R s a A antiserum. For comparison, the lane labeled with * is a low p H extraction of an equivalent number of cells expressing w i l d type levels of S-layer.  43  cause loss of R s a A secondary structure that might be important for R s a A anchoring. T o test this hypothesis, S-layer shedding cells (JS1001) were co-cultured with S-layernegative cells that possessed w i l d type (JS1003) or deficient (JS1004) levels of S L P S . A s predicted, more S-layer reassembled selectively to SLPS-positive cells (JS1003) in this method as compared to the methods aforementioned: SLPS-dependent S-layerrecrystallization (Fig. 2-3, lane 3) occurred to levels approaching those of w i l d type cells (Fig. 2-3, lane 1).  These results suggested that perhaps that soluble, reattachment-  competent R s a A is present in the supernatant of S-layer shedding strains, in addition to the R s a A that aggregates into macroscopic particles i n these cultures. supernatant from an S-layer-shedding  Accordingly,  C. crescentus culture (JS1001) was analyzed, and  soluble R s a A was indeed found in appreciable quantity ( F i g . 2-4, lane 1). Protein prepared in this way was then tested for its ability to reattach to S-layer negative cells. R s a A obtained in this manner reassembled on S-layer-negative (JS1003) cells to w i l d type (NA1000) levels (Fig. 2-4, compare lane 6 to lane 2). These results suggest that R s a A obtained in this manner retains the proper secondary structure for optimum levels of S-layer recrystallization. The varied ability of R s a A prepared in different ways to reattach to S-layer-negative cells suggests that obtaining R s a A by some methods causes a loss of secondary structure that is important for R s a A anchoring. These results not only address the importance of the maintenance of proper S-layer folding for normal levels of S-layer presentation on the cell surface, but also provide the basis of an assay to test S-layer-mutants for their ability to reattach, in an effort to investigate the importance of R s a A regions or residues for R s a A anchoring.  44  RsaA  100 k D a  Fig. 2-3. S-layer reattachment by co-culturing S-layer donors and recipients. S-layer shedding cells (donors) were co-cultured with S-layer-negative cells (recipients) that possess wild type or deficient levels of S L P S , in a cell ratio of 8 donors: 1 recipient. Equivalent numbers wild type cells alone (lane 1), donor cells alone (lane 2), donors + wild type-SLPS recipients (lane 3), or donors + deficient-SLPS recipients (lane4) were subjected to low p H extraction. Extractions were separated on a 7.5% S D S - P A G E gel, and RsaA was detected by western blotting using RsaA antiserum. SLPS-dependent S-layerrecrystallization (lane 3) occurred to levels close to wild type levels (lane 1).  45  Fig.  2-4.  S-layer reattachment  using soluble RsaA from  the supernatant  of shedder  cultures.  Soluble R s a A harvested from the supernatant of an S-layer shedding strain was incubated with an equivalent number of S-layer negative cells that possess w i l d type or deficient levels of S L P S . Reattached R s a A was subsequently extracted by l o w p H treatment. Extractions were separated on a 7.5% S D S - P A G E gel, and RsaA was detected by western blotting using RsaA antiserum. Lanes: 1, 12 u L supernatant from shedder strain; 2 & 3, to assess the extent of S-layer reattachment, low p H extractions of 100% (lane 2) and 10% (lane 3) of the number of cells used in the reattachment assay were performed; 4, l o w p H extraction of S-layernegative cells possessing w i l d type levels of S L P S (control); 5, low p H extraction o f S-layernegative cells possessing deficient levels of S L P S (control); 6, low p H extraction o f S-layernegative cells possessing w i l d type levels of S L P S after incubation with soluble R s a A ; 7, low p H extraction of S-layer-negative cells possessing deficient levels o f S L P S after incubation with soluble R s a A .  46  3.2 - S L P S is r e q u i r e d for R s a A anchoring A fraction of the lipopolysaccharide (LPS) on the surface of  C. crescentus cells has an  O antigen polymer attached to the core to form a "smooth" L P S ( S L P S ) . This O antigen polymer consists at least in part of N-acetylperosamine, a 4-amino-4,6-dideoxymannose (2).  C. crescentus strains that are deficient in S L P S shed their S-layer into the culture  medium, suggesting that S L P S is involved in S-layer anchoring. However, our lab did not have a strain that was completely devoid of S L P S ; rather, JS1004 was isolated as a strain grown in the absence of calcium which likely generated a "leaky" point mutation that drastically reduced S L P S production, but did not ablate it entirely (2).  In  reattachment assays of R s a A to JS1004, a very low level of R s a A still reattached to the cells (Fig. 2-4, lane 7). This left the possibility that either S L P S or some other molecule on the cell surface was mediating the low level of S-layer reattachment observed with JS1004. T o determine i f other molecules could mediate R s a A anchoring i n the absence of S L P S , a strain completely devoid of S L P S and R s a A needed to be constructed. The putative N - acetylperosamine biosynthetic pathway was previously proposed  (2)  and is shown in F i g . 2-5. M a n B is a putative phosphomannomutase involved in this pathway, catalyzing the conversion of mannose 6-phosphate to mannose 1-phosphate. Previous transposon mutagenesis studies have shown that the disruption of  manB results  in the loss of S L P S production in vivo (2). T o construct a strain completely devoid of S L P S and also R s a A , I disrupted  manB i n a strain that was already lacking R s a A  (JS1013); the resulting strain was named JS1014. JS1014 was to be used as an S-layernegative, SLPS-negative target for S-layer reattachment, to determine i f any S-layer at all  47  Gluco kinase  T-'" ' 1  PhosphomannoSsometase  Glucose 6-phosphate Pgl  ^  PbosphoglucQisomerase  Fructose 6-phosphate ManC ^  Pnosphonwnnolsomerase  Mannose 6-phosphate ManB ^  Phosphomannomutase  ^~"yt Gmtl  I  O  Per  GDP-mannose GDP-mannose  4,6-dehydratase  GD P-4-keto-6-deoxyma n n ose  Perosamfne  synthetase  GDP-perosamlne (G D P-4-am I no-4,6-d ideoxynna n nose)  Mannose 1-phosphate  F i g . 2-5. Proposed N-acetylperosamine biosynthetic pathway (2).  48  could reattach to cells completely devoid of S L P S . The disruption of  manB resulted in a  strain completely devoid of S L P S , as expected (Fig. 2-6, lane 6). To determine i f R s a A reattachment could be mediated by another molecule in the absence of S L P S , soluble, reattachment-competent R s a A (isolated from the supernatant of JS1001) was incubated with S-layer-negative cells that had no S L P S (JS1014), low levels of S L P S (JS1004), or w i l d type levels of S L P S (JS1003). N o S-layer reattachment was observed to cells that completely lack S L P S , a low amount of S-layer reattached to cells with l o w levels of S L P S , and a large amount of S-layer reattached to cells possessing w i l d type levels of S L P S (Fig. 2-7). These results confirm that S L P S is required for S-layer anchoring, and that no other molecule on the cell surface can mediate S-layer attachment in the absence of S L P S .  3.3 - Two anchoring regions in RsaA? Previous work has suggested that the N-terminus of R s a A mediates cell surface anchoring, since N-terminal truncations of R s a A lead to shedding of the S-layer into the culture medium. To begin to investigate the importance of regions or residues within the R s a A anchoring region, my first goal was to find a region or regions in R s a A sufficient for R s a A anchoring. In order to generate R s a A fragments that retained proper secondary structure necessary for R s a A anchoring, I first engineered collagenase cleavage sites at various points throughout R s a A . Full-length R s a A bearing the collagenase cleavage sites at various positions were retrieved from the supernatant of  C. crescentus shedder  cultures, treated with collagenase, and the resulting R s a A fragments were tested for their ability to selectively reattach to SLPS-positive cells. For example, one collagenase  49  1  2  3  4  5  6  SLPS  Fig. 2-6. SLPS levels in SLPS-positive, SLPS-deficient, and SLPS-negative C. crescentus strains. Cells possessing varying amounts of S L P S were grown to m i d log phase. Whole cell protein preparations from equivalent numbers of cells were performed. Samples were run on a 7.5% S D S P A G E gel. S L P S was detected by western blotting using S L P S antiserum. Lanes: 1 , Kan-cassette interruption of rsaA in spontaneous calcium-independent mutant (deficient levels of S L P S ) ; 2, K a n cassette interruption of rsaA (wt levels of S L P S ) ; 3, Transposon insertion mutant Tn5F9 (interruption of manB (2)); 4, Transposon insertion mutant Tn5F23 (interruption of manB (2)); 5, wt C. crescentus; 6, manB knock-out in S-layer-negative strain (no S L P S ) .  50  S L P S levels on target cells  •  N/A  wt  d  * RsaA  Fig. 2-7. S-layer attachment requires SLPS. Wild-type RsaA was incubated with cells possessing varying S L P S levels (wild type, wt, deficient, d, or none, -). Whole cell protein preparations from equivalent numbers of cells were then performed. Extracted protein was separated on a 10% S D S - P A G E gel. R s a A was detected by western blotting using R s a A antiserum. Lane marked with * is input R s a A alone.  51  cleavage site was engineered at a position in R s a A corresponding to amino acid 690. This protein was harvested from the supernatant of an S-layer-negative shedder culture (JS1004) harboring a plasmid-borne copy of  C. crescentus  rsaA with a collagenase site  at the position corresponding to R s a A amino acid 690, and subsequently treated with collagenase. The resulting R s a A fragments were incubated with cells possessing varying levels of S L P S , (JS1014, JS1004, and JS1003) and then the cells were analyzed for the presence of reattached R s a A fragments.  Equivalent amounts of the R s a A N - and C -  terminal fragment resulted from collagenase cleavage of the R s a A mutant, as expected and as evidenced by Coomassie-staining of the digest products separated on an S D S P A G E gel (data not shown). However, R s a A antiserum reacted poorly with the R s a A N terminal fragment  (Fig. 2-8, lane marked with *). Unexpectedly, both the N - and C -  terminal R s a A fragments were detected on cells possessing w i l d type levels of S L P S (JS1003) (Fig. 2-8). A n appreciable amount of the R s a A C-terminal fragment reattached to the cells, and given the limited reactivity of the antibody towards the R s a A N-terminal fragment, an appreciable amount of the R s a A N-terminal fragment probably reattached to those cells as well (Fig. 2-8). The phenomenon of R s a A N - and C-terminal fragment reattachment was observed for several other R s a A mutants as w e l l (data not shown). This opened the possibility that two R s a A anchoring regions may exist in R s a A , or that some type of co-operative binding may be occurring. However, since R s a A N-terminal truncations shed into the culture medium (9), the R s a A C-terminus alone likely cannot mediate cell surface anchoring. W e favored the idea that the observed R s a A C-terminal fragment reattachment was a result of its co-operative binding to reattached R s a A N terminal fragment or to reattached full-length RsaA. The observed R s a A C-terminal  52  Amount of SLPS on target cells  *  N/A  wt  d 160 kDa  Full-length RsaA  •  RsaA N-terminal , . fragment  •  f-  4"4t0  110 kDa  90 kDa „„ , _ 70 kDa 55 kDa 45 kDa  RsaA C-terminal fragment  •  S=  F i g 2.8. Two anchoring regions in RsaA ? R s a A fragments (resulting from collagenase treatment of mutant R s a A bearing a collagenase cleavage site at residue 690) were incubated with cells possessing varying S L P S levels (wild type, wt, deficient, d, or none, -). Whole cell protein preparations from equivalent numbers o f cells were then performed. Extracted protein was separated on a 10% S D S - P A G E gel. R s a A and R s a A fragments were detected by western blotting using R s a A antiserum. Lane marked with * is input (cleaved) R s a A alone.  53  fragment reattachment turned out to require full-length R s a A , as I discuss in a later part of this Results section.  3.4 - RsaA 1-277 is sufficient for RsaA anchoring Since disruption of R s a A anchoring domain(s) and disruption of R s a A subunit-subunit interaction domains both result in the same phenotype (the S-layer is shed from the bacterium), when a particular mutation results in a shedding phenotype, it is difficult to know whether that mutation disrupted R s a A anchoring or crystallization (subunit-subunit interaction). M y approach to determining the regions of the R s a A N-terminus that are important for R s a A anchoring was to first find a small R s a A N-terminal fragment that is sufficient for R s a A anchoring. Since such a protein would be lacking much of the native sequence putatively involved i n R s a A subunit-subunit interactions (crystallization), I would have confidence that subsequent mutations within this R s a A N-terminal fragment that lead to loss of reattachment of this small fragment could be ascribed to disruptions in the R s a A - S L P S interaction, rather than disruptions in R s a A subunit-subunit interactions (recall that some R s a A monomers are thought to be directly anchored to S L P S on the cell surface, while others are thought to be tethered to the cell surface only by interacting with other RsaA monomers rather than by a direct interaction with S L P S ) . To find a small R s a A N-terminal fragment sufficient for reattachment, I first engineered collagenase cleavage sites at various positions within full-length R s a A . These proteins were collected from C.  crescentus shedder strains, treated with collagenase, and  the resulting R s a A fragments were incubated with S-layer-negative C.  crescentus cells to  assess their ability to reattach to these cells. R s a A 1-277 was identified as a small R s a A  54  N-terminal fragment that was sufficient for R s a A reattachment to these cells (Fig. 2-9, lane 3), whereas R s a A 1-162 could not be detected by available R s a A antiserum (Fig 2-9, lane 2). Note that consistent with the data aforementioned, both the R s a A N-and C terminal fragments reattached to cells in the case of R s a A fragment pair 1-723 & 7241026, as well as full length R s a A (Fig. 2-9, lane 5), leaving the question of whether two anchoring regions exist in R s a A or whether co-operative binding is occurring unresolved. This issue is addressed in the next part of this Results section.  3.5 - The RsaA C-terminal fragment reattachment requires full-length RsaA Results from several experiments indicated that both the R s a A N - and C-terminal fragments (resulting from collagenase treatment of R s a A bearing a collagenase site) can reattach to S-layer-negative cells. One way to resolve the issue of whether or not there are two anchoring regions within R s a A or whether co-operative binding occurs is to separate R s a A fragments resulting from collagenase digestion, then test the individual fragments for reattachment capability. I attempted to separate these R s a A fragments by several standard methods, including using A m i c o n Ultra Centrifugal Filter D e v i c e s ™ (Bedford, Massachusetts) with the appropriate molecular weight cut-offs, size exclusion chromatography, and affinity chromatography using a Histidine-tagged version of R s a A . A l l attempts to separate the fragments were unsuccessful, perhaps due to the intrinsic nature of R s a A to multimerize. W e have observed in the lab that S-layer fragments appear to associate with one another, rendering separation of these fragments quite difficult (Smit, unpublished). For example, a Histidine-tagged R s a A fragment w i l l bind to a nickel column, but unfortunately the other R s a A fragment (from which separation is  55  1 110 kDa  90 kDa mmm 70 kDa  2  3  4  asfc,  5 —  6  7 4  Full-length  RsaA  55 kDa 45 kDa 35 kDa 25 kDa 15 kDa  Fig. 2-9. RsaA 1-277 is sufficient for RsaA anchoring. RsaA bearing a collagenase cleavage site at residue 162, 277 or 723 was cleaved by collagenase and the resulting RsaA fragments were incubated with Sap-negative, S-layer-negative cells. After washing, protein preparations from equivalent numbers of cells were then separated on a 12% S D S - P A G E gel. R s a A and R s a A fragments were detected by western blotting using R s a A antiserum. Lanes: 1, cleaved RsaA (cleavage site at residue 162) + cells; 2, cleaved R s a A (cleavage site at residue 162) alone; 3, cleaved RsaA (cleavage site at residue 277) + cells; 4, cleaved R s a A (cleavage site at residue 277) alone; 5, cleaved R s a A (cleavage site at residue 723) + cells; 6, cleaved R s a A (cleavage site at residue 723) alone; 7, cells alone.  56  desired) apparently associates with the column-bound R s a A fragment, and thus both fragments come off of the column upon elution (data not shown).  Similarly, R s a A  fragments tend to elute from a size exclusion column together as one peak in the void volume, even in the presence of detergents or E D T A (data not shown). This suggests that these R s a A fragments may form soluble microaggregates that cannot enter into the column matrix beads. Upon closer inspection of the data from several experiments, it appeared that the RsaA N-terminal fragment always reattached to cells. However, the R s a A C-terminal fragment only reattached when residual full-length (uncleaved) R s a A , left over despite collagenase treatment, was available and also reattached. This suggested that the reattachment of the R s a A C-terminal fragment may be due to an association of C-terminal fragments with residual full-length S-layer, rather than direct attachment of the R s a A C-terminal fragment to the cell surface. T o test this hypothesis, R s a A bearing a collagenase cleavage site at residue 277 was treated with collagenase for various lengths of time, in order to generate complete and partial digests of the protein. The resulting R s a A and R s a A fragments were then incubated with S-layer-negative cells (JS4025) to test their ability to reattach. A s hypothesized, the R s a A C-terminal fragment (278-1026) only reattached when some full-length R s a A was available and also reattached (Fig. 2-10, compare lane 2 with lane 5). Since the R s a A C-terminal fragment did not reattach in the absence of fulllength R s a A , these results indicate that the R s a A C-terminal fragment contained no anchoring information.  In contrast, the R s a A N - t e r m i n a l fragment  reattaches  independently of full-length R s a A (or the R s a A C-terminal fragment), indicating that the anchoring region of R s a A is located in the R s a A N-terminus.  57  110 kDa 90 kDa 70 kDa  ..•'-'III*!,,  <«  Full-length RsaA  ^  RsaA 278-1026  55 kDa 45 kDa  <«MWK  35 kDa  ^ ^ ^ ^ ^  .  ...**«*  RsaA 1-277  25 kDa 1*1: Fig. 2-20. /?saA C-terminalfragmentanchoring requires full-length RsaA. RsaA bearing a collagenase cleavage site at residue 277 was cleaved by collagenase for increasing lengths of time. The resulting R s a A fragments were incubated with SAP-negative, Slayer negative cells. After washing, protein preparations from equivalent numbers of cells were then separated on a 12% S D S - P A G E gel. RsaA and R s a A fragments were detected by western blotting using R s a A antiserum. Lanes: 1, cells (alone); 2 through 5, cells + R s a A cut by collagenase for a specific length of time: 2, 60 min; 3, 120 m i n ; 4, 180 m i n ; 5, overnight; 6, RsaA cut by collagenase overnight (alone).  58  3.6 - Mutations in RsaA 1-277 cause the loss of RsaA anchoring M y next goal was to determine which regions within the R s a A N-terminus are important for anchoring. Accordingly, I constructed independent mutations i n the N terminus at positions in R s a A corresponding to amino acids 7, 29, 69, 154, 169, & 222 (Fig. 2-11). These mutations were chosen as an attempt to cover the entire putative R s a A anchoring region, since little information is gained from secondary structure predictions for R s a A using current algorithms. The mutations at residues 7, 154 and 222 resulted in a two amino acid exchange for Gly/Ser, while the mutations at residues 29, 69, and 169 resulted in the insertion of four amino acids (N-Asp-Gly-Ser-Val) at these positions (see Materials and Methods section).  A l l of these mutants also possessed the collagenase  cleavage site at residue 277. These full-length proteins were collected from C.  crescentus  shedder strains and subsequently treated with collagenase. The resulting fragments were then incubated with S-layer negative cells, and after washing, the cells were analyzed for the presence of R s a A and R s a A fragments. Surprisingly, all mutations in the R s a A N terminus caused loss of anchoring (Figs. 2-12 and 2-13). These results suggest that all of the regions of R s a A that were mutated contribute in some way to the anchoring of the R s a A N-terminus to S L P S . Additionally, the digest of R s a A bearing the collagenase site at residue 277, but without further mutation in the N-terminus (the control input protein), appeared to go to completion (Fig. 2-12, lane marked with *). Although the resulting R s a A 1-277 fragment clearly reattached to SLPS-positive cells, there is no evidence of any C-terminal fragment (RsaA 278-1026) binding (Fig. 2-12, lanes marked wt). This is consistent with the aforementioned suggestion that some full-length protein is required for any C-terminal fragment to bind, and that only the N-terminus of R s a A carries the  59  Collagenase site at residue 277  I  N  C  Fig. 2-11. Box model of mutant RsaA used for reattachment studies. X represents independent mutations at positions in the R s a A N-terminus corresponding to amino acids 7, 29, 69, 154, 169, and 222. These full-length proteins were isolated from C. crescentus shedder strains and subsequently treated with collagenase. The resulting R s a A fragments were assessed for their ability to reattach to S-layer-negative cells.  60  29 110 kDa 90 kDa 70 kDa 55 kDa 45 kDa 35 kDa  69  wt  Mutation in N? SLPS on target? RsaA 278-1026  ^ - -M M Hi  RsaA 1-277  25 kDa • a  Loading control  F i g . 2-12. Mutations at RsaA amino acids 7,29, and 69 disrupt RsaA anchoring. RsaA bearing a collagenase cleavage site at residue 277 that also possessed or did not possess (wt) an additional mutation in the N-terminus were cleaved by collagenase. The resulting R s a A fragments were incubated with Sap-negative cells that either possessed (+) or did not possess (-) S L P S . After washing, whole cell protein preparations from equivalent numbers of cells were then performed. Extracted protein was separated on a 12% S D S - P A G E gel. R s a A and R s a A fragments were detected by western blotting using R s a A antiserum. The lane marked with * is a sample o f the collagenase-treated input protein that did not possess an additional mutation in its N-terminus.  61  154  169 +  222 +  -  wt +  < + ^  110 kDa 90 kDa ~— 70 kDa ;  Mutation in N? SLPS on target?  =  RsaA 278-1026  55 kDa 45 kDa RsaA 1-277  35 kDa 25 kDa 15 kDa  ^  ••••••••  Loading control  Fig. 2-13. Mutations at RsaA amino acids 154,169, and 222 disrupt RsaA anchoring. RsaA bearing a collagenase cleavage site at residue 277 that also possessed or did not possess an additional mutation i n the N-terminus were cleaved by collagenase. The resulting R s a A fragments were incubated with SAP-negative cells that either possessed or did not possess S L P S . After washing, whole cell protein preparations from equivalent numbers of cells were then performed. Extracted protein was separated on a 12% S D S - P A G E gel. R s a A and R s a A fragments were detected by western blotting using RsaA antiserum. The lane marked with * is a sample of the collagenase-treated input protein that did not possess an additional mutation in its N-terminus.  62  anchoring information for the protein.  In the case of the R s a A mutants, there is no  evidence of any C-terminal fragment (RsaA 278-1026) binding. This is consistent with the hypothesis that only the R s a A N-terminus can mediate anchoring: i f there were any residual full-length mutant protein remaining after the collagenase treatment, its mutated N-terminus w o u l d be incapable of mediating anchoring, thus preventing its - and consequently the C-terminal fragment's - reattachment to the cell surface.  Taken  together, these data confirm that the R s a A N-terminus mediates R s a A anchoring, and that small perturbations within the first -225 amino acids disrupt R s a A anchoring.  63  4. RESULTS - Sap localization studies 4.1 - Sap cleaves RsaA 1-277 Throughout the course of my initial R s a A reattachment studies, I observed that reattachment o f R s a A 1-277 to the cell surface o f S-layer-deficient cells resulted i n the cleavage o f R s a A 1-277. Interestingly, this cleavage resulted i n a product that was about 28 k D a , the same sized cleavage product previously observed for some other R s a A mutants (8). One possible candidate for this proteolytic activity was the S-layerassociated protease, Sap (72). T o determine i f Sap was responsible for the cleavage of R s a A 1-277, soluble R s a A bearing the collagenase site at residue 277 was isolated from cells that harbored a plasmid-borne copy of rsaA with the collagenase site engineered at residue 277; these cells were either Sap-positive (JS1004) or Sap-negative (internal deletion i n sap, see Materials and Methods) cells (JS1012). The isolated protein was subsequently treated with collagenase, and the resulting R s a A fragments were incubated with S-layer-negative target cells that were either Sap-positive (JS1003) or Sap-deficient (JS4025). The results indicate that the source of the input protein did not matter i n terms of the cleavage o f R s a A 1-277, however, the use of Sap-positive target cells (JS1003) leads to complete cleavage of this protein, whereas the use of Sap-deficient target cells largely prevented this cleavage (Fig. 3-1, compare lanes 1 & 4 or lanes 2 & 5). The reason that the cleavage was not completely ablated when Sap-deficient target cells (JS4025) were used may be that i n this strain, there is only a point mutation i n the active site o f sap, (72) which significantly reduces its activity, but may not abolish it completely, allowing for some residual cleavage of R s a A 1-277. These data indicate that target cell-harbored Sap is responsible for cleavage o f R s a A 1-277.  64  Sap' target  Sap target +  No No Sap" Sap protein SapSap protein ^ I ] J 1 1 J 11 i L'VmAAm* „i I J . . . ^ sheddershedder added sheddershedder added +  Input protein  +  f t *  derived from  110 kDa 90 kDa 70 kDa 55 kDa 45 kDa 35 kDa 25 kDa  <  RsaA 1-277  ^~  Cut RsaA j 277  15 kDa  Fig. 3-1. Sap cleaves RsaA 1-277. RsaA bearing a collagenase site at 277 was isolated from a Sap" or Sap shedder strain, then treated with collagenase. The resulting RsaA fragments were then incubated with equivalent numbers of S-layer-negative cells that are Sap or Sap". After washing, whole cell preparations were performed. Samples were separated on a 12% SDS-PAGE gel. RsaA fragments were detected by western blotting using RsaA antiserum. +  +  65  This was a fortuitous result, since our S-layer reattachment assay involved mutating specific regions within R s a A 1-277, then testing these fragments for their ability to reattach to S-layer-negative cells. Undesired cleavage of R s a A 1-277 (that contained a particular mutation within these residues) being tested for its ability to reattach to S-layer deficient cells is prevented by using Sap-deficient cells as targets i n the reattachment assay.  4.2 - Sap is an extracellular enzyme In the S-layer reattachment assay, protein is added exogenously to intact cells. Upon analysis of reattached protein, we found that R s a A 1-277 was cleaved by Sap. Initially, to examine whether or not a particular R s a A fragment had reattached to a target cell, whole cell protein preparations were performed (using lysozyme, see Materials and Methods). This resulted in the lysis of the target cells. For Sap-positive (target) cells, it was possible that a) Sap is localized on the outer membrane or in the supernatant of the target cells and can therefore easily access and cleave R s a A 1-277 during the reattachment assay, or b) Sap is cytoplasmic, but upon cell lysis, Sap encounters and cleaves R s a A 1-277. T o investigate the localization of Sap, the reattachment of R s a A 1277 to Sap-positive target cells was performed, and then samples were either analyzed after the usual whole cell protein preparation, or by boiling samples after reattachment (see Materials and Methods). L i k e the whole cell protein preparations, boiling the samples would cause cell lysis.  However, unlike the whole cell protein preparation,  boiling would also be expected to denature the proteins in the sample, and thus Sap, i f intracellular, may be released upon this boiling and thus could come into contact with R s a A 1-277, but Sap would be expected to be denatured. The denatured Sap would be  66  expected to be inactive and therefore unable to cleave R s a A 1-277. ,RsaA 1-277 was incubated with S-layer-negative, Sap-positive cells (JS1003), and then after washing, one aliquot of the resulting cells was boiled immediately, while another aliquot containing an equivalent amount of cells was subjected to whole cell protein preparation. The results indicate that i n comparison to preparing the protein sample by whole cell protein preparation, boiling the samples immediately after the reattachment assay still resulted in a significant amount of cleaved R s a A 1-277 (Fig. 3-2, lane 1). This suggests that R s a A 1-277 encounters and is cleaved by Sap during the reattachment assay rather than after reattachment/cell lysis, which necessitates that Sap be localized extracellularly.  4.3 - S-layer Type I secretion OMPs Fa and/or Fb are involved in Sap secretion Since data were accumulating that suggested that Sap is an extracellular enzyme, our focus shifted to how the protease might get secreted from the bacterium. One hypothesis is that Sap is secreted by the same Type I secretion system that R s a A utilizes. Although there is no high degree of homology between the C-terminus of R s a A ,and the C-terminus of Sap (i.e., no indication that Sap may contain a Type I secretion signal similar to that of R s a A ) , since it has been shown that another protease called A p r A from Pseudomonas aeruginosa, whose C-terminus shares even less homology to the C-terminus of R s a A , can be secreted by the  Caulobacter crescentus T y p e T secretion system (1), we thought it  possible that Sap could also be secreted by this system. T o test this hypothesis, a strain that I constructed (JS1011) that was devoid of the outer membrane proteins RsaFa and RsaFb involved in this system (Fig. 3-3, lane 3) as well as S-layer (data not shown) was used as a target for the reattachment of R s a A 1-277. The cleavage of R s a A 1-277 was  67  Fig. 3-2. SAP is probably an extracellular enzyme RsaA bearing a collagenase site at residue 277 was isolated from an SAP-negative C. crescentus shedder strain (lane 3), then treated with collagenase (lane 4). The resulting R s a A fragments were then incubated with equivalent numbers of S A P positive cells. After washing, cells were immediate boiled i n the presence of loading buffer (lane 1), or whole cell preparations were performed (lane 2). Samples were separated on a 12% S D S - P A G E gel. R s a A fragments were detected by western blotting using R s a A antiserum.  68  JS4000  NA1000 wt  F b  F -/F b  wt  F - F -/F a  a  b  unknown protein  Fig. 3.3. RsaFa and RsaFb levels in wild type and RsaF knockout  <  RsaF  <  RsaF  a  b  strains.  Whole cell protein preparations from equivalent numbers of N A 1 0 0 0 (wt) or JS4000 (wt) cells, or strains derived from them that are null for RsaFa (F -), RsaFb (F -), or both RsaFa & RsaFb (F -/F -) were separated on a 7.5% S D S - P A G E gel. RsaFa and RsaFb were detected using RsaFa antiserum (which has cross-reactivity with RsaFb). a  a  b  b  69  significantly reduced when JS 1011 was used compared to using a target (JS1015) that possesses the w i l d type O M P s of the S-layer Type I transporter proteins (Fig. 3-4, compare lanes 3 & 4).  This suggests that RsaFa and/or RsaFb are involved in the  secretion of Sap, and thus when this/these O M P s are not present, Sap does not get secreted and therefore cannot access and cleave R s a A 1-277 that has been reattached to the cell surface.  4.4 - A third F outer membrane protein may be present in strain JS4000 C. crescentus strain JS4000 is a spontaneous mutant that is deficient in S-layer production due to a frameshift mutation in the S-layer gene that results in an early stop codon a third o f the way into  rsaA (66). There are a few other differences between  JS4000 and N A 1 0 0 0 , including the fact that JS4000 appears to shed more heterologous protein than NAlOOO-derived (Smit, unpublished).  rsaA knockouts when harboring the appropriate plasmid  Since one focus of our laboratory has been to optimize C .  crescentus protein expression/yield for various biotechnology applications, we have used JS4000 routinely for this purpose. The genome of N A 1 0 0 0 has been sequenced and is available (www.tigr.org). A search of the N A 1 0 0 0 genome suggests there are only two chromosomal O M P genes  (rsaFa and rsaFb) involved i n S-layer Type I secretion, and  this has been confirmed experimentally (71) for this  C. crescentus strain. In an unrelated  study, I constructed an NAlOOO-derived strain that was null for rsaFa, rsaFb, and rsaA, called JS1011 (Fig. 3-3, lane 3), as well as a JS4000-derived strain that was null for rsaFa and rsaFb , called JS4023 (Fig. 3-3, lane 6). Unexpectedly, in contrast to JS1011,  70  1  2  3  4 110 kDa 90 kDa 70 kDa 55 kDa 45 kDa  RsaA 1-277  35 kDa  Cut RsaA 1-277 25 kDa  Fig. 3-4. S-layer Type I secretion OMPs Fa and/or Fb are involved in the secretion of Sap. RsaA bearing a collagenase site at residue 277 was isolated from an Sap C. crescentus shedder -  strain (lane 2), then treated with collagenase (lane 1). The resulting R s a A fragments were then incubated with equivalent numbers of S-layer-negative C. crescentus cells that either possess (lane 4) or do not possess (lane 3) RsaFa & RsaFb. After washing, cells were immediate boiled in the presence of loading buffer. Equal loadings of sample were separated on a 12% S D S P A G E , and R s a A and R s a A fragments were detected by western blotting using R s a A antiserum.  71  upon introduction of a plasmid-borne copy of rsaA into JS4023, R s a A still gets secreted to the cell surface, evidenced by a) low p H extraction of these cells yields S-layer, and b) S-layer is detected on the surface of these cells i n immunofluorescence experiments (data not shown).  This suggests that a third  rsaF gene may be present on the JS4000  chromosome, allowing for R s a A secretion in this strain even when rsaFa and rsaFb have been deleted. T o gather more evidence suggesting there is a third RsaF O M P i n JS4000, and that Sap can utilize this putative O M P for its secretion, the cleavage o f R s a A 1-277 was assessed using JS1011 or JS4023 as targets for reattachment o f R s a A 1-277. A s predicted, full cleavage of R s a A 1-277 was observed when JS4023 was used as a target (Fig. 3-5, lane 2) but not when JS 1011 was used as a target (Fig. 3-5, lane 1). Taken together, these data suggest that a third RsaF protein may well form part of the S-layer Type I secretion system i n JS4000, and that this third R s a F transporter protein is also involved i n the secretion of Sap. A third O M P i n JS4000-based strains may explain the improved protein expression/yield observed in our lab (Smit, unpublished) when utilizing these strains for heterologous protein production.  ,  4.5 - Sap is unlikely to be exported into the culture media Having implicated RsaF O M P s i n the secretion of Sap, we wanted to determine whether Sap is localized to the cell surface, or secreted into the culture medium. T o address this issue, an R s a A mutant that had been previously shown to get cleaved by Sap called R s a A : V P 2 C A (RsaA with the insertion of a 112 amino acid segment of a salmonid virus glycoprotein at a position corresponding to R s a A amino acid 723) (72) was first isolated from an Sap-negative shedder strain (JS1012) harboring a plasmid encoding  72  1 2  3  4  110 kDa 90 kDa 70 kDa 55 kDa 45 kDa 35 kDa  4  RsaA 1-277  <  Cut RsaA 1-277  25 kDa  Fig. 3-5. A third F outer membrane protein may be present in strain JS4000. RsaA bearing a collagenase site at residue 277 was isolated from an Sap" C. crescentus shedder strain, then treated with collagenase (lane 4). The resulting R s a A fragments were then incubated with equivalent numbers of Sap-positive, S-layer-negative, RsaFa- & RsaFbnegative cells derived from N A 1 0 0 0 (lane 1) or JS4000 (lane 2). A s a control, the R s a A fragments were also incubated with Sap-negative (JS4025) cells (lane 3). After washing, cells were immediate boiled in the presence of loading buffer. Equal loadings of sample were separated on a 12% S D S - P A G E , and RsaA and R s a A fragments were detected by western blotting using R s a A antiserum.  73  R s a A : V P 2 C A . This isolated protein was subsequently reattached to Sap-deficient target cells (JS4025), and then excess mutant R s a A was washed away. Next, S-layer-negative, Sap-positive cells (JS4000) or supernatant from an Sap-positive cell culture was incubated with the Sap-negative cells that had (uncleaved) R s a A : V P 2 C A reattached to them, in an effort to cleave the reattached R s a A : V P 2 C A .  Cleavage of R s a A : V P 2 C A  would indicate the presence of Sap on either the cell surface or in the supernatant, suggesting where Sap is localized after its secretion.  Incubation o f cell-bound  R s a A : V P 2 C A with Sap-positive cells did not result in the cleavage of the Sap substrate (Fig. 3-6, lane 2), but this might be because Sap that is bound to the cell surface of one cell cannot effectively access or cleave its substrate when it is bound to another cell. Importantly, however, incubation of cell-bound R s a A : V P 2 C A with supernatant from the Sap-positive culture also failed to result in the cleavage of the Sap substrate, since the amount of full-length R s a A : V P 2 C A found anchored to these cells was roughly equal to that found anchored to an equivalent number of cells that were not treated with Sappositive cells or supernatant (Fig. 3-6, compare lane 1 with lane 3). This preliminary evidence suggests that Sap is not found in the supernatant of Sap-positive cells: i f Sap were secreted into the supernatant, Sap would not be hampered by being bound to a cell surface, and thus Sap would be expected to have access to its cell-bound substrate and should be able to cleave it with ease. It is therefore unlikely that Sap is exported into the culture media. Taken together with the evidence of the involvement of the O M P s F a and/or F b i n the secretion of Sap, this suggests that Sap is an outer membrane-bound protease.  74  1  2  3  RsaA:VP2CA  4  110 kDa  Fig. 3-6. SAP is unlikely to be exported into the culture media.  RsaA with the insertion of a 112 amino acid segment of the V P 2 surface glycoprotein of infectious pancreatic necrosis virus ( I P N V ) at residue 723 (70), hereafter called R s a A : V P 2 C A , was isolated from an Sap" C. crescentus shedder strain. R s a A : V P 2 C A was then incubated with S-layernegative, Sap-negative C. crescentus cells (lanes 1-3), or S-layer-negative, SAP-positive C. crescentus cells (lane 4), then excess R s a A : V P 2 C A was washed away. Aliquots o f SAP-negative cells with reattached R s a A : V P 2 C A were then subsequently incubated with S-layer-negative, S A P positive cells (lane 2) or culture supernatant (lane 1). After washing, cells were immediate boiled in the presence of loading buffer. Equal loadings of sample were separated on a 12% S D S - P A G E , and R s a A : V P 2 C A was detected by western blotting using R s a A antiserum.  75  5. DISCUSSION AND CONCLUSION W e have developed an assay that can be utilized systematically to investigate regions or residues of the C.  crescentus S-layer protein, R s a A , that are important for S-layer  anchoring in vivo. Through this work, we have learned that R s a A secondary structure is important for R s a A anchoring, and therefore care must be taken when preparing R s a A and R s a A mutants destined for reattachment. W e demonstrated conclusively for the first time that S L P S is required for R s a A anchoring, and that i n the complete absence of S L P S , R s a A anchoring does not occur. W e have established that it is the N-terminus of R s a A that mediates anchoring of S-layer to the cells surface, and that the first 277 amino acids are sufficient for R s a A anchoring. Surprisingly, all of the six mutations created thus far within the first 222 amino acids have resulted in loss of R s a A anchoring, suggesting that this entire region may contribute in some way to S-layer anchoring. W e have accumulated evidence suggesting that the S-layer-associated protease, Sap, is likely an O M - b o u n d protein, rather than a cytoplasmic enzyme as previously suggested. Finally, we have implicated the S-layer Type I secretion O M P s RsaFa and/or RsaFb in the secretion of Sap to the O M . There has been a significant amount of investigation into the S-layer anchoring regions of Gram-positive bacteria, but little work has been done to define S-layer anchoring regions of Gram-negative bacteria.  In fact, to date, there has only been one  other study that evaluated the ability of truncated S-layer protein to reattach to S-layerdeficient Gram-negative cells. In that study, deletion mutagenesis revealed that C. fetus S-layer proteins bound serospecifically to the C. fetus lipopolysaccharide v i a their conserved N-terminal regions, which include approximately 189 amino acids (16).  76  Those findings are comparable to the results of this study for the  C. crescentus S-layer  protein: we have found that the first -225 amino acids of R s a A are involved in S-layer anchoring, suggesting that a large amount of residues mediate S-layer anchoring in Gram-negative bacteria.  This is in accordance with anchoring domains for S-layer  proteins of Gram-positive bacteria, where anchoring domains are usually 3 repeats of the - 5 5 amino acid S-layer homology ( S L H ) domain (4). Although an S-layer anchoring motif has been defined and is well documented for Gram-positive bacteria (19), there is no such motif defined for S-layer protein for Gramnegative bacteria. In the C. fetus S-layer study (16), an investigation of the importance of regions or residues found within the conserved N-terminal lipopolysaccharide binding domain was not performed, thus it is not known which regions or residues within that domain are important for the anchoring of the C. fetus S-layer proteins. The present study is the first to elucidate the involvement of regions or residues within a defined Slayer anchoring domain in Gram-negative bacteria. Because all mutations created thus far within the putative R s a A anchoring region abolish R s a A anchoring, a complementary approach to the current mutagenesis strategy may be to complete comparative studies of S-layer sequences from other Caulobacter species, in an effort to search for conserved residues or motifs that may mediate S-layer anchoring. Further understanding of S-layer anchoring i n Caulobacter may aid in the understanding of S-layer anchoring in other Gram-negative bacteria. There have been several instances where we wondered whether or not Sap was located on the cell surface rather than in the cytoplasm. For example, when the 112-amino acid segment of the V P 2 surface glycoprotein of infectious pancreatic necrosis virus ( I P N V )  77  strain S P is inserted at a position corresponding to amino acid 723 of R s a A , Sap-positive cells subjected to low p H conditions to extract S-layer protein from the cell surface yields both the N - and C-terminal fragments resulting from Sap cleavage (72). Since the C terminus of R s a A contains the information required for R s a A secretion (9), i f Sap cleaves the heterologous S-layer protein in the cytoplasm, then only the C-terminal fragment should get secreted and thus be retrievable by low p H extraction of protein localized on the cell surface. A t the time of our last publication regarding Sap we had no reason to predict that Sap could be on the cell surface (an analysis of the Sap sequence reveals no predicted N-terminal signal leader peptide, no predicted Type I secretion signal, and no predicted transmembrane domains). Additionally, no experimental data suggested that this might be the case.  Therefore we previously suggested that R s a A intramolecular  forces (such as hydrogen bonding) might account for both the N - and C-terminal products being secreted. Sap may cleave the particular mutant protein in the cytoplasm, but these R s a A intramolecular forces may hold the two cleavage products together and allow for their simultaneous secretion. However we have now accumulated experimental evidence suggesting that Sap is located on the cell surface, and can access and cleave reattached R s a A 1-277. If Sap is localized on the cell surface, some Sap substrates (such as R s a A : V P 2 C A and R s a A 1277) would likely be cleaved by Sap on the cell surface rather than i n the cytoplasm, which would explain how both cleavage products are located on the cell surface. Clearly, final confirmation of the localization of Sap using Sap-specific antibody needs to be done, and efforts are underway to generate antibodies against Sap to perform these experiments.  78  We have observed that Sap apparently cleaves various R s a A mutants at the same position in R s a A (about a third of the way C-terminal from the N-terminus), yielding a ~28 k D a R s a A N-terminal cleavage product. This "weak Sap cleavage site" is present in wild-type RsaA, but Sap does not cleave wild-type RsaA. W e hypothesize that this weak Sap recognition site is buried or inaccessible to Sap in the wild-type R s a A case due to native folding of the S-layer. In contrast, some R s a A mutants may cause altered folding of S-layer, thus exposing this weak cleavage site to Sap. It is possible that this is a Sapmediated strategy that C.  crescentus utilizes to rid the cell surface or R s a A mutants that  disrupt normal S-layer assembly on the cell surface. Efforts are underway to determine the exact sequence and position of this weak Sap cleavage site. Several questions arise from the Sap studies. W e have implicated the S-layer Type I secretion O M P s RsaFa and/or RsaFb in the secretion of Sap. A n investigation of the effect on Sap secretion i n strains that are null for only RsaFa or RsaFb w i l l suggest whether one or both of these proteins are important for Sap secretion.  Furthermore,  results are consistent with the presence of a third O M P i n JS4000-derived JS4025 that may secrete Sap. The construction and subsequent screening of members of a JS4025 transposon library for loss of Sap secretion may aid in the identification of this putative third O M P . Identification and characterization of this third O M P might explain why protein expression/yield is improved in JS4000-based strains. The involvement of RsaFa and/or RsaFb in Sap secretion suggests that Sap may be secreted by the S-layer Type I secretion system. Further studies such as investigations of the effects on Sap secretion in strains lacking other components of the S-layer Type I secretion system (RsaD and RsaE) w i l l need to be done to resolve these possibilities. If  79  it is found that indeed the S-layer Type I A B C transporter secretes Sap as well as R s a A , this would not be the first time a Type I secretion system has been shown to secrete more than one protein. The L i p B C D Type I secretion system in  Serratia marcescens has in  fact been shown to mediate the secretion of three proteins: a lipase, a metalloprotease, and the  S. marcescens S-layer protein (33).  From primary protein sequence, RsaFa and RsaFb are predicted to be O M P beta barrel proteins. Since experimental evidence suggests that Sap is O M - b o u n d (as is RsaA), after RsaFa and/or RsaFb-mediated translocation past the O M , Sap (and R s a A ) must get transferred to molecules on the cell surface that tether these proteins to the cell surface. The question arises: i f Sap is OM-bound, what O M molecule is Sap bound to? The C terminus of Sap is homologous to the N-terminus of RsaA, and it may mediate anchoring of Sap to the cell surface. Our previous model of Sap activity in the case of heterologous S-layer was that the Sap C-terminus associated with the N-terminus of R s a A , in the cytoplasm, as it scanned for and cleaved some mutated versions of R s a A . However, in light of the data suggesting that Sap is OM-bound, it could be that the Sap C-terminal homology to the N-terminus of R s a A reflects the possibility that the C-terminus of Sap anchors to S L P S on the cell surface, in the same way that the N-terminus of R s a A anchors to S L P S . Interestingly, another  C. crescentus OM-bound protein, H f a D , remains  anchored to the cell surface in the absence of S-layer (Brun, personal communications), as does Sap.  H f a D is 28% identical to the first 277 amino acids of R s a A (data not  shown), suggesting that perhaps H f a D also anchors to S L P S on the cell surface. Therefore, it is possible that at least three proteins are bound to the  C. crescentus cell  surface via a conserved protein domain that interacts with S L P S . Reattachment of the  80  purified proteins (Sap or HfaD) to SLPS-positive but not SLPS-negative cells would confirm that these proteins anchor to the cell surface by interacting with S L P S . The Type V  Autotransporter secretion (29) and lipid-linked proteins  (3) are two  mechanisms by which multiple proteins are secreted and subsequently anchored to the cell envelope of Gram-negative bacteria. Both of these transport systems utilize the Secdependent pathway also known as the General Secretory Pathway, G S P to traverse the inner membrane, and thus generate periplasmic intermediates.  These periplasmic  intermediates can be subjected to proteolysis by periplasmic enzymes, and may also require chaperones for successful translocation (29), (3). In contrast to the G S P , the Type I secretion system can secrete a variety of proteins across both the inner and outer membranes, without generating periplasmic intermediates (55). If it is true that the C.  crescentus S-layer Type I secretion system also secretes other proteins such as Sap and H f a D which subsequently anchor to the cell surface, then we may have discovered another, less cumbersome secretion pathway that C.  crescentus utilizes to efficiently  secrete and subsequently anchor proteins to the cell surface by interaction with S L P S . This concept of Type I secretion systems utilized for secretion and subsequent surface presentation of multiple proteins via the same Type I system may extend to other Gramnegative bacteria in general.  81  REFERENCES  1.  2.  3. 4.  5.  Awram, P., and J. Smit. 1998. The Caulobacter crescentus paracrystalline Slayer protein is secreted by an A B C transporter (type I) secretion apparatus. J Bacteriol 180:3062-9. Awram, P., and J. Smit. 2001. Identification of lipopolysaccharide O antigen synthesis genes required for attachment of the S-layer of Caulobacter crescentus. Microbiology 147:1451-60. Bernstein, H. D. 2000. The biogenesis and assembly of bacterial membrane proteins. Curr Opin M i c r o b i o l 3:203-9. Beveridge, T. J., and L. L. Graham. 1991. Surface layers of bacteria. M i c r o b i o l Rev 55:684-705.  Beveridge, T. J., P. H. Pouwels, M. Sara, A. Kotiranta, K. Lounatmaa, K. Kari, E. Kerosuo, M. Haapasalo, E. M. Egelseer, I. Schocher, U . B. Sleytr, L. Morelli, M. L. Callegari, J. F. Nomellini, W. H. Bingle, J. Smit, E. Leibovitz, M. Lemaire, I. Miras, S. Salamitou, P. Beguin, H. Ohayon, P. Gounon, M. Matuschek, and S. F. Koval. 1997. Functions of S-layers. F E M S M i c r o b i o l Rev 20:99-149.  6.  Binet, R., S. Letoffe, J. M. Ghigo, P. Delepelaire, and C. Wandersman. 1997.  7.  Protein secretion by Gram-negative bacterial A B C exporters—a review. Gene 192:7-11. Bingle, W. H., J. F. Nomellini, and J. Smit. 1997. Cell-surface display of a Pseudomonas aeruginosa strain K pilin peptide within the paracrystalline S-layer of Caulobacter crescentus. M o l M i c r o b i o l 26:277-88.  8.  Bingle, W. H., J. F. Nomellini, and J. Smit. 1997. Linker mutagenesis of the  9.  10.  11.  12. 13.  Caulobacter crescentus S-layer protein: toward a definition of an N-terminal anchoring region and a C-terminal secretion signal and the potential for heterologous protein secretion. J Bacteriol 179:601-11. Bingle, W. H., J. F. Nomellini, and J. Smit. 2000. Secretion of the Caulobacter crescentus S-layer protein: further localization of the C-terminal secretion signal and its use for secretion of recombinant proteins. J Bacteriol 182:3298-301. Bingle, W. H., and J. Smit. 1990. High-level expression vectors for Caulobacter crescentus incorporating the transcription/translation initiation regions of the paracrystalline surface-layer-protein gene. Plasmid 24:143-8.  Blaser, M. J., P. F. Smith, J. E. Repine, and K. A. Joiner. 1988. Pathogenesis of Campylobacter fetus infections. Failure of encapsulated Campylobacter fetus to bind C3b explains serum and phagocytosis resistance. J C l i n Invest 81:1434-44. Boot, H. J., and P. H. Pouwels. 1996. Expression, secretion and antigenic, variation of bacterial S-layer proteins. M o l M i c r o b i o l 21:1117-23. Brent, R., and M. Ptashne. 1980. The l e x A gene product represses its own promoter. Proc Natl A c a d S c i U S A 77:1932-6.  82  14.  15.  16.  17.  18.  19.  20.  21.  22.  23.  24.  25. 26.  27. 28.  Brun, Y. V., G. Marczynski, and L. Shapiro. 1994. The expression of asymmetry during Caulobacter cell differentiation. A n n u Rev Biochem 63:41950. Doig, P., W. D. McCubbin, C. M. Kay, and T. J. Trust. 1993. Distribution of surface-exposed and non-accessible amino acid sequences among the two major structural domains of the S-layer protein of Aeromonas salmonicida. J M o l B i o l 233:753-65. Dworkin, J., M. K. Tummuru, and M. J. Blaser. 1995. A lipopolysaccharidebinding domain of the Campylobacter fetus S-layer protein resides within the conserved N terminus of a family of silent and divergent homologs. J Bacteriol 177:1734-41. Edwards, P., and J. Smit. 1991. A transducing bacteriophage for Caulobacter crescentus uses the paracrystalline surface layer protein as a receptor. J Bacteriol 173:5568-72. Egelseer, E., I. Schocher, M. Sara, and U . B. Sleytr. 1995. The S-layer from Bacillus stearothermophilus D S M 2358 functions as an adhesion site for a highmolecular-weight amylase. J Bacteriol 177:1444-51. Engelhardt, H., and J. Peters. 1998. Structural research on surface layers: a focus on stability, surface layer homology domains, and surface layer-cell wall interactions. J Struct B i o l 124:276-302. Fujimoto, S., A. Takade, K. Amako, and M. J. Blaser. 1991. Correlation between molecular size of the surface array protein and morphology and antigenicity of the Campylobacter fetus S layer. Infect Immun 59:2017-22.  Fujino, T., P. Beguin, and J. P. Aubert. 1993. Organization of a Clostridium thermocellum gene cluster encoding the cellulosomal scaffolding protein C i p A and a protein possibly involved in attachment of the cellulosome to the cell surface. J Bacteriol 175:1891-9. Garduno, R. A., and W. W. Kay. 1992. Interaction of the fish pathogen Aeromonas salmonicida with rainbow trout macrophages. Infect Immun 60:461220. Garduno, R. A., E. J. Lee, and W. W. Kay. 1992. S-layer-mediated association of Aeromonas salmonicida with murine macrophages. Infect Immun 60:4373-82.  Garduno, R. A., A. R. Moore, G. Olivier, A. L. Lizama, E. Garduno, and W. W. Kay. 2000. Host cell invasion and intracellular residence by Aeromonas salmonicida: role of the S-layer. Can J Microbiol 46:660-8. Garduno, R. A., B. M. Phipps, and W. W. Kay. 1995. Physical and functional S-layer recdnstitution in Aeromonas salmonicida. J Bacteriol 177:2684-94. Gilchrist, A., J. A. Fisher, and J. Smit. 1992. Nucleotide sequence analysis of the gene encoding the Caulobacter crescentus paracrystalline surface layer protein. Can J Microbiol 38:193-202. Gilchrist, A., and J. Smit. 1991. Transformation of freshwater and marine caulobacters by electroporation. J Bacteriol 173:921-5. Gober, J. W., and M. V. Marques. 1995. Regulation of cellular differentiation in Caulobacter crescentus. Microbiol Rev 59:31-47.  83  29.  Henderson, I. R., F. Navarro-Garcia, M. Desvaux, R. C. Fernandez, and D. Ala'AIdeen. 2004. Type V protein secretion pathway: the autotransporter story. Microbiol M o l B i o l Rev 68:692-744.  30.  31. 32. 33.  34.  35.  36. 37.  38.  39.  40.  41.  Howard, L., and D. J. Tipper. 1973. A polypeptide bacteriophage receptor: modified cell wall protein subunits in bacteriophage-resistant mutants of Bacillus sphaericus strain P - l . J Bacteriol 113:1491-504. Iuga, M., P. Awram, J. F. Nomellini, and J. Smit. 2004. Comparison of S-layer secretion genes in freshwater caulobacters. Can J Microbiol 50:751-66. Janakiraman, R. S., and Y. V. Brun. 1997. Transcriptional and mutational analyses of the rpoN operon in Caulobacter crescentus. J Bacteriol 179:5138-47. Kawai, E., H. Akatsuka, A. Idei, T. Shibatani, and K. Omori. 1998. Serratia marcescens S-layer protein is secreted extracellularly via an ATP-binding cassette exporter, the L i p system. M o l Microbiol 27:941-52. Kilian, M. 1981. Degradation of immunoglobulins A 2 , A 2 , and G by suspected principal periodontal pathogens. Infect Immun 34:757-65.  Kotiranta, A., M. Haapasalo, K. Kari, E. Kerosuo, I. Olsen, T. Sorsa, J. H. Meurman, and K. Lounatmaa. 1998. Surface structure, hydrophobicity, phagocytosis, and adherence to matrix proteins of Bacillus cereus cells with and without the crystalline surface protein layer. Infect Immun 66:4895-902. Koval, S. F., and S. H. Hynes. 1991. Effect of paracrystalline protein surface layers on predation by Bdellovibrio bacteriovorus. J Bacteriol 173:2244-9. Lemaire, M., H. Ohayon, P. Gounon, T. Fujino, and P. Beguin. 1995. O l p B , a new outer layer protein of Clostridium thermocellum, and binding of its S-layerlike domains to components of the cell envelope. J Bacteriol 177:2451-9. Lewis, L. O., A. A. Yousten, and R. G. Murray. 1987. Characterization of the surface protein layers of the mosquito-pathogenic strains of Bacillus sphaericus. J Bacteriol 169:72-9.  Lupas, A., H. Engelhardt, J. Peters, U. Santarius, S. Volker, and W. Baumeister. 1994. Domain structure of the Acetogenium kivui surface layer revealed by electron crystallography and sequence analysis. J Bacteriol 176:122433. MacRae, J. D., and J. Smit. 1991. Characterization of caulobacters isolated from wastewater treatment systems. A p p l Environ Microbiol 57:751-8.  Matuschek, M., G. Burchhardt, K. Sahm, and H. Bahl. 1994. Pullulanase of Thermoanaerobacterium thermosulfurigenes E M I (Clostridium thermosulfurogenes): molecular analysis of the gene, composite structure of the enzyme, and a common model for its attachment to the cell surface. J Bacteriol 176:3295-302.  42.  McCoy, E. C , D. Doyle, K. Burda, L. B. Corbeil, and A. J. Winter. 1975. Superficial antigens of Campylobacter (Vibrio) fetus: characterization of antiphagocytic component. Infect Immun 11:517-25.  43.  Mesnage, S., E. Tosi-Couture, P. Gounon, M. Mock, and A. Fouet. 1998. The capsule and S-layer: two independent and yet compatible macromolecular structures in Bacillus anthracis. J Bacteriol 180:52-8.  84  44.  45.  46.  Morales, V. M., A. Backman, and M. Bagdasarian. 1991. A series of widehost-range low-copy-number vectors that allow direct screening for recombinants. Gene 97:39-47. Nomellini, J. F., S. Kupcu, U . B. Sleytr, and J. Smit. 1997. Factors controlling in vitro recrystallization of the Caulobacter crescentus paracrystalline S-layer. J Bacteriol 179:6349-54. Nomellini, J. F., M. Toporowski, and J. Smit. 2003. Secretion or presentation of recombinant proteins and peptides mediated by the S-layer o f Caulobacter crescentus, in Expression Technologies: Current Status and Future Trends. Horizon Scientific Press, Norfolk, in press.  47.  Olabarria, G., J. L. Carrascosa, M. A. de Pedro, and J. Berenguer. 1996. A  48.  conserved motif in S-layer proteins is involved in peptidoglycan binding i n Thermus thermophilus. J Bacteriol 178:4765-72. Pei, Z., and M. J. Blaser. 1990. Pathogenesis of Campylobacter fetus infections. Role of surface array proteins i n virulence in a mouse model. J C l i n Invest 85:1036-43.  49. 50. 51. 52.  53.  Peterson, P. K., B. J. Wilkinson, Y . Kim, D. Schmeling, and P. G. Quie. 1978. Influence of encapsulation on staphylococcal opsonization and phagocytosis by human polymorphonuclear leukocytes. Infect Immun 19:943-9. Poindexter, J. S. 1964. Biological Properties and Classification of the Caulobacter Group. Bacteriol Rev 28:231-95. Poindexter, J. S. 1981. The caulobacters: ubiquitous unusual bacteria. Microbiol Rev 45:123-79. Roberts, K., P. J. Shaw, and G. J. Hills. 1981. High-resolution electron microscopy of glycoproteins: the crystalline cell wall of Lobomonas. J C e l l Sci 51:295-313.  Salamitou, S., M. Lemaire, T. Fujino, H. Ohayon, P. Gounon, P. Beguin, and J. P. Aubert. 1994. Subcellular localization of Clostridium thermocellum ORF3p, a protein carrying a receptor for the docking sequence borne by the catalytic components of the cellulosome. J Bacteriol 176:2828-34.  54.  Salamitou, S., O. Raynaud, M. Lemaire, M. Coughlan, P. Beguin, and J. P.  59.  Aubert. 1994. Recognition specificity of the duplicated segments present in Clostridium thermocellum endoglucanase C e l D and in the cellulosomeintegrating protein C i p A . J Bacteriol 176:2822-7. Salmond, G. P., and P. J. Reeves. 1993. Membrane traffic wardens and protein secretion in gram-negative bacteria. Trends Biochem S c i 18:7-12. Sambrook, J., E. F. Fritsch, and T. Maniatis. 1989. Molecular cloning: a laboratory manual. C o l d Spring Harbor Laboratory Press, C o l d Spring Harbor, N.Y. Sara, M., and U . B. Sleytr. 1996. Crystalline bacterial cell surface layers (Slayers): from cell structure to biomimetics. Prog Biophys M o l B i o l 65:83-111. Sara, M., and U . B. Sleytr. 1987. Molecular sieving through S layers of Bacillus stearothermophilus strains. J Bacteriol 169:4092-8. Sara, M., and U . B. Sleytr. 2000. S-Layer proteins. J Bacteriol 182:859-68.  60.  Schafer, A., A. Tauch, W. Jager, J. Kalinowski, and G. Theirbach. 1994.  55. 56.  57. 58.  Small mobilizable multi-purpose cloning vectors derived from the Escherichia  85  61.  62.  63. 64. 65. 66.  67. 68.  69.  70.  coli plasmids p K 1 8 and pK19: selection of defined deletions i n the chromosome of Cornyebacterium glutamicum. Gene 145:69-73. Shapiro, L. 1985. Generation of polarity during Caulobacter cell differentiation. Annu Rev Cell B i o l 1:173-207.  Sharma, S. B., and E. R. Signer. 1990. Temporal and spatial regulation of the symbiotic genes of Rhizobium meliloti i n planta revealed by transposon Tn5gusA. Genes Dev 4:344-56. Sleytr, U. B., and T. J. Beveridge. 1999. Bacterial S-layers. Trends M i c r o b i o l 7:253-60. Sleytr, U. B., and P. Messner. 1983. Crystalline surface layers on bacteria. A n n u Rev Microbiol 37:311-39. Sleytr, U. B., P. Messner, D . Pum, and M. Sara. 1996. Crystalline Bacterial C e l l Surface Proteins. Academic Press, Landes Company, Austin. Smit, J., and N. Agabian. 1984. Cloning of the major protein of the Caulobacter crescentus periodic surface layer: detection and characterization of the cloned peptide by protein expression assays. J Bacteriol 160:1137-45.  Smit, J., H. Engelhardt, S. Volker, S. H. Smith, and W. Baumeister. 1992. The S-layer of Caulobacter crescentus: three-dimensional image reconstruction and structure analysis by electron microscopy. J Bacteriol 174:6527-38. Sundqvist, G., J. Carlsson, B. Herrmann, and A. Tarnvik. 1985. Degradation of human immunoglobulins G and M and complement factors C 3 and C 5 by black-pigmented Bacteroides. J M e d Microbiol 19:85-94. Takeoka, A., K. Takumi, T. Koga, and T. Kawata. 1991. Purification and characterization of S layer proteins from Clostridium difficile G A I 0 7 1 4 . J Gen Microbiol 137 (Pt 2):261-7.  Thompson, S. A., O. L. Shedd, K. C. Ray, M. H. Beins, J. P. Jorgensen, and M. J. Blaser. 1998. Campylobacter fetus surface layer proteins are transported by a type I secretion system. J Bacteriol 180:6450-8.  71.  Toporowski, M. C , J. F. Nomellini, P. Awram, and J. Smit. 2004. Two outer membrane proteins are required for maximal type I secretion of the Caulobacter crescentus S-layer protein. J Bacteriol 186:8000-9.  72.  Umelo-Njaka, E., W. H. Bingle, F. Borchani, K. D . Le, P. Awram, T. Blake, J. F. Nomellini, and J. Smit. 2002. Caulobacter crescentus synthesizes an S-layerediting metalloprotease possessing a domain sharing sequence similarity with its paracrystalline S-layer protein. J Bacteriol 184:2709-18.  73:  Umelo-Njaka, E., J. F. Nomellini, H. Yim, and J. Smit. 2001. Development of small high-copy-number plasmid vectors for gene expression i n Caulobacter crescentus. Plasmid 46:37-46.  74.  Walker, S. G., D . N. Karunaratne, N. Ravenscroft, and J. Smit. 1994.  75.  Characterization of mutants of Caulobacter crescentus defective i n surface attachment of the paracrystalline surface layer. J Bacteriol 176:6312-23. Walker, S. G., S. H. Smith, and J. Smit. 1992. Isolation and comparison of the paracrystalline surface layer proteins of freshwater caulobacters. J Bacteriol 174:1783-92.  86  76. 77.  78.  Welch, R. A., C. Forestier, A. Lobo, S. Pellett, W. Thomas, Jr., and G. Rowe. 1992. The synthesis and function of the Escherichia coli hemolysin and related R T X exotoxins. F E M S Microbiol Immunol 5:29-36. Wiklund, T., and I. Dalsgaard. 1998. Occurrence and significance of atypical Aeromonas salmonicida in non-salmonid and salmonid fish species: a review. Dis Aquat Organ 32:49-69. Yang, L. Y., Z. H. Pei, S. Fujimoto, and M . J. Blaser. 1992. Reattachment of surface array proteins to Campylobacter fetus cells. J Bacteriol 174:1258-67.  87  

Cite

Citation Scheme:

    

Usage Statistics

Country Views Downloads
United States 5 2
China 2 2
Canada 2 0
Norway 1 0
City Views Downloads
Ashburn 3 0
Vancouver 2 0
Sunnyvale 2 0
Shenzhen 1 2
Trondheim 1 0
Beijing 1 0

{[{ mDataHeader[type] }]} {[{ month[type] }]} {[{ tData[type] }]}
Download Stats

Share

Embed

Customize your widget with the following options, then copy and paste the code below into the HTML of your page to embed this item in your website.
                        
                            <div id="ubcOpenCollectionsWidgetDisplay">
                            <script id="ubcOpenCollectionsWidget"
                            src="{[{embed.src}]}"
                            data-item="{[{embed.item}]}"
                            data-collection="{[{embed.collection}]}"
                            data-metadata="{[{embed.showMetadata}]}"
                            data-width="{[{embed.width}]}"
                            async >
                            </script>
                            </div>
                        
                    
IIIF logo Our image viewer uses the IIIF 2.0 standard. To load this item in other compatible viewers, use this url:
http://iiif.library.ubc.ca/presentation/dsp.831.1-0092468/manifest

Comment

Related Items