Outer Membrane Protease/Adhesin PgtE ofSalmonella enterica:

Role in Salmonella-Host Interaction

Päivi Ramu

General MicrobiologyDepartment of Biological and Environmental Sciences

Faculty of Biosciencesand

Viikki Graduate School in BiosciencesUniversity of Helsinki

Academic dissertation in General Microbiology

To be presented, with the permission of the Faculty of Biosciences,University of Helsinki, for public criticism in the Auditorium 2 at Viikki

Infocenter Korona (Viikinkaari 11, Helsinki) on May 4th, 2007 at 12o’clock noon.

Helsinki 2007

2

Supervisors Professor Timo Korhonen, PhDDepartment of Biological and Environmental SciencesFaculty of BiosciencesUniversity of Helsinki, Finland

Docent Kaarina Lähteenmäki, PhDDepartment of Biological and Environmental SciencesFaculty of BiosciencesUniversity of Helsinki, Finland

Reviewers Docent Kaisa Granfors, PhDDepartment of Bacterial and Inflammatory DiseasesNational Public Health InstituteDepartment in Turku, Finland

Sakari Jokiranta, MD, PhDDepartment of Bacteriology and ImmunologyHaartman InstituteUniversity of Helsinki, Finland

Opponent Professor Levente Emödy, MD, PhD, DScDepartment of Medical Microbiology and ImmunologyFaculty of MedicineUniversity of Pécs, Hungary

Cover Figure A model for PgtE-mediated migration of Salmonella throughtissue barriers. Macrophage (M ); Alpha2-antiplasmin ( 2AP);Plasminogen (Plg); Matrix metalloproteinase-9 (MMP-9);Extracellular matrix (ECM).

ISSN 1795-7079ISBN 978-952-10-3901-0 (paperback)ISBN 978-952-10-3902-7 (pdf)http://ethesis.helsinki.fiYliopistopaino, Helsinki 2007

3

To Petri and my family

4

5

Preface

This work was carried out at the General Microbiology, Department ofBiological and Environmental Sciences, Faculty of Biosciences in the Universityof Helsinki. I warmly thank the Head of the Division, Professor Timo Korhonenand the Head of the Department, Professor Kielo Haahtela for the opportunity toperform this work and for creating great working and educational facilities.

I have been happy to have two excellent supervisors, Professor Timo Korhonenand Docent Kaarina Lähteenmäki, and I owe my deepest gratitude for theirguidance and support during my PhD studies. Timo and Kaarina have beenamazing supervisors and advisors and I have loved our conversations aboutscience and life in general.

I sincerely thank the reviewers of this thesis, Docent Kaisa Granfors and DrSakari Jokiranta for the smooth reviewing process and for their good andvaluable comments.

It has been a pleasure to work in the Bacterial adhesion lab and I want to thankall the present and former members of our research group and technical staff forall the help and for creating a friendly atmosphere. Maini is thanked for teachingme during my early days in lab. Leandro, Marjo and Riikka are acknowledgedfor being good and cheerful room mates. Veera is thanked for being such a goodfriend.

My special thanks go to my parents Ritva and Pekka, my brother Mika and allmy friends for supporting me in my never-ending studies and for keeping meattached to the real world. To the “new” members of my family: Ole, Salme,Niina and Janne, I am grateful for taking me into your homes and hearts. Andfinally I want to thank my dear husband Petri, whose continuous support helpedme to finish this work. Thank you for believing me so much!

Helsinki, April 2007

6

CONTENTS

Preface 5List of original publications 7Summary 81. Introduction: Pathogenesis of Salmonella infections 102. Interaction of Salmonella with macrophages and dendritic cells 123. The outer membrane protein PgtE of Salmonella enterica 143.1 The omptin family of bacterial surface proteins 143.2 The PhoP/Q two-component regulatory system in regulation

of PgtE and other outer membrane components of Salmonella 174. Lipopolysaccharide variability and modification in Salmonella 195. Interaction of Salmonella with the complement system 205.1 Complement as an antimicrobial defense system 205.2 Resistance mechanisms of Salmonella

against the complement system 226. Bacterial interference with the host extracellular matrix

and matrix-degrading proteolytic systems 236.1 Interaction of Salmonella with the host extracellular matrix 236.2 Matrix metalloproteinases in bacterial infections 247. Aims of the study 278. Materials and methods 289. Results and discussion 319.1 Role of lipopolysaccharide for the activity of PgtE of

Salmonella enterica and Pla of Yersinia pestis 319.2 Activity of PgtE in Salmonella released from macrophages 359.3 Role of PgtE in survival of Salmonella inside macrophages in vitro 369.4 Effect of PgtE on survival of Salmonella in mice 379.5 Identification of novel substrates for PgtE 38

9.5.1 Alpha2-antiplasmin 389.5.2 Complement components C3b, C4b and C5 409.5.3 Matrix metalloproteinase-9 and gelatin 42

9.6 Structural basis for the substrate specificity of PgtE 4410. Conclusions 4811. References 51

7

List of original publications

This thesis of Päivi Ramu (née Kyllönen) is based on the following articles,which in the text are referred to by their Roman numerals.

I. Kukkonen, M., Suomalainen, M., Kyllönen, P., Lähteenmäki, K.,Lång, H., Virkola, R., Helander, I.M., Holst, O., Korhonen, T.K.(2004) Lack of O-antigen is essential for plasminogen activation byYersinia pestis and Salmonella enterica. Mol. Microbiol. 51: 215-225.(Includes supplementary material: http://www.blackwellpublishing.com/products/ journals/suppmat/mmi/mmi3817/mmi3817sm.htm)

II. Lähteenmäki, K., Kyllönen, P., Partanen, L., Korhonen, T.K. (2005)Antiprotease inactivation by Salmonella enterica released from infectedmacrophages. Cell. Microbiol. 7: 529-538.

III. Ramu, P., Tanskanen, R., Holmberg, M., Lähteenmäki, K.,Korhonen, T.K., Meri, S. (2007) The surface protease PgtE ofSalmonella enterica affects complement activity by proteolyticallycleaving C3b, C4b and C5. FEBS Lett., in press.(doi:10.1016/j.febslet.2007.03.049)

IV. Ramu, P., Lobo, L.A., Kukkonen, M., Bjur, E., Suomalainen, M.,Raukola, H., Miettinen, M., Julkunen, I., Holst, O., Rhen, M.,Korhonen, T.K., Lähteenmäki, K. Activation of pro-matrixmetalloproteinase-9 and degradation of gelatin by the Salmonellaenterica surface protease PgtE. Submitted to Int. J. Med. Microbiol.

The articles are reprinted by kind permission from the publishers.

8

Summary

Salmonella enterica serovar Typhimurium is a common cause of gastroenteritisin humans and, occasionally, also causes systemic infection. During systemicinfection an important characteristic of Salmonella is its ability to survive andreplicate within macrophages (M ). The outer membrane protease PgtE of S.enterica is a member of the omptin family of outer membrane aspartateproteases, which are -barrel proteins with five surface-exposed loops. The maingoals of this study were to characterize biological substrates and pathogenesis-associated functions of PgtE and to determine the conditions where PgtE is fullyactive.

In this study we found that PgtE requires rough lipopolysaccharide (LPS) to befunctional but is sterically inhibited by the long O-antigen side chain in smoothLPS. Salmonella isolates normally are smooth with a long oligosaccharide O-antigen, and PgtE remains functionally cryptic in wild-type Salmonellacultivated in vitro. Interestingly, our results showed that due to increasedexpression of PgtE and to reduced length of the LPS O-antigen chains, the wild-type Salmonella expresses highly functional PgtE when isolated from mouseM -like J774A.1 cells. Salmonella is thought to be continuously released fromM s to infect new ones, and our results suggest that PgtE is functional duringthese transient extracellular growth phases.

Six novel host protein substrates were identified for PgtE in this work. PgtE waspreviously known to activate human plasminogen (Plg) to plasmin, a broad-spectrum serine protease, and in this study PgtE was shown to interfere with thePlg system by inactivating the main inhibitor of plasmin, alpha2-antiplasmin.PgtE also interferes with another important proteolytic system of mammals byactivating pro-matrix metalloproteinase-9 to an active gelatinase. PgtE alsodirectly degrades gelatin, a component of extracellular matrices. PgtE alsoincreases bacterial resistance against complement-mediated killing in humanserum and enhances survival of Salmonella within murine M s as well as in theliver and spleen of intraperitoneally infected mice.

Taken together, the results in this study suggest that PgtE is a virulence factor ofSalmonella that has adapted to interfere with host proteolytic systems and to

9

modify extracellular matrix; these features likely assist the migration ofSalmonella during systemic salmonellosis.

10

1. Introduction: Pathogenesis of Salmonella infections

Salmonellosis causes over 3 million deaths every year and is economically themost important foodborne disease world-wide (Pang et al, 1995). The genusSalmonella contains two species: Salmonella enterica that is pathogenic towarm-blooded animals and Salmonella bongori that is rarely associated withinfectious diseases in humans. Salmonella is genetically related to Escherichiacoli and their genomic sequences are approximately 80 % identical (McClellandet al, 2001). It has been estimated that E. coli and Salmonella diverged from acommon ancestor some 120 million years ago after which they have gainednumerous phenotypic characteristics that distinguish these enteric species fromeach other (Ochman and Wilson, 1987). Salmonella has acquired fourteen largegenomic insertions, Salmonella pathogenicity islands 1 to 14 (SPI-1 to SPI-14)that encode many genes important for virulence of Salmonella (Morgan, 2007).SPI-1 was acquired very early in the evolution of Salmonella and both S.enterica and S. bongori possess it. SPI-2, on the other hand, is only present in S.enterica and was acquired by this species after its divergence from S. bongori(Groisman and Ochman, 1997). Based on comparative genomics and biotyping(Crosa et al, 1973; LeMinor and Popoff, 1987)), S. enterica has been dividedinto 6 subspecies and 2519 serovars (sv) (Popoff et al, 2004). Subspecies Iincludes 99% of the serovars that are pathogenic to mammals and cause diseasesvarying from self-limiting gastroenteritis to systemic bacteremia (Chan et al,2003). In humans, S. enterica sv Typhimurium and sv Enteritidis are the mostcommon causes of gastroenteritis, and also cause systemic infection inapproximately 5% of infected individuals (Hohmann, 2001). A more severeSalmonella-infection, with a mortality of 10-15% without treatment, is typhoidfever caused by sv Typhi (Ohl and Miller, 2001). In a murine host,Typhimurium infection resembles human typhoid fever, and therefore mousemodels are widely used to study pathogenesis of systemic Salmonella infections.

Salmonella infections begin with ingestion of bacteria in contaminated food orwater. A fraction of Salmonella cells survive the low pH of the stomach andreach the small intestine, where they replicate and colonize. Salmonellapenetrates the intestinal mucus layer of the small bowel and adheres to theintestinal epithelium. Salmonella traverses to the basolateral side of theepithelium. Salmonella-mediated invasion of the epithelium in the smallintestine occurs preferably via antigen-sampling M-cells that overlie the Peyer´s

11

patches, but also via the nonphagocytic enterocytes. The invasion processrequires the type three secretion system (TTSS) and effector proteins encoded bySPI-1 (Ohl and Miller, 2001). After invasion into the epithelial cells,gastroenteritis-causing Salmonella induces a fluid secretion in the intestine andinitiates transmigration of neutrophils into the intestinal lumen (Galyov et al,1997; McGovern and Slavutin, 1979). These events cause diarrhea which is oneof the symptoms in salmonellosis (Giannella, 1979).

Once across the epithelium, Salmonella initiates systemic infection by enteringsubmucosal macrophages (M ) and dendritic cells (DC) by host-cell mediatedphagocytosis as well as by Salmonella-induced macropinocytosis. Survivalinside phagocytic cells is crucial for Salmonella during systemic infection, and ithas been shown that Salmonella mutants unable to survive inside macrophagesare avirulent (Fields et al, 1986). Intracellular Salmonella disseminates from thegut to the organs of the reticuloendothelial system, such as the liver and thespleen, where the bacteria are mainly associated with M s and DCs (Richter-Dahlfors et al, 1997).

At some point during the infection, Salmonella induces cell death in M s (Chenet al, 1996; Hueffer and Galán, 2004; Knodler and Finlay, 2001; Lindgren et al,1996; Richter-Dahlfors et al, 1997) and this way escapes from the host cell andmay spread to new ones. The infected, apoptotic cells are also targeted forengulfment by other macrophages and this way Salmonella spreads directly fromone cell to another (Guiney, 2005). Sheppard et al (2003) have shown thatduring acute infection, the number of Salmonella cells in an individualphagocyte is low, with the majority of the phagocytes containing only onebacterium. They hypothesized that in tissues Salmonella replicates intracellularlyin pathological lesions and that bacterial growth mainly results from an increasein the number of infected phagocytes. It follows that at some point during theinfection Salmonella escapes from infected phagocytes to infect new ones and todisseminate to secondary infection sites. This way, the amount of lesions alsoincreases as the infection proceeds. The repeated escape from the host cells andthe infection of new ones are crucial for full spread within the host and thevirulence of Salmonella (Mastroeni and Sheppard 2004).

12

2. Interaction of Salmonella with macrophages anddendritic cells

M s and DCs are Trojan horses in systemic Salmonella infections asSalmonella survives and replicates within these phagocytic cells that normallyare critical for innate immunity and the acquired antibacterial responses.Professional phagocytes include the circulating monocytes and the resident M sin tissues, such as Kupffer cells in the liver. M s are also found in mucosalstructures such as the submucosal M s in the Peyer’s patches of the smallintestine. They are important in many biological functions such as development,tissue remodeling, immune responses and inflammation (Aderem and Underhill,1999) and effectively recognize, internalize and degrade microbes, apoptotichost cells and inert particles (Méresse et al, 1999). Another cell group that isimportant in systemic Salmonella infection is the antigen-presenting DCs. Theyare found in almost all peripheral tissues as well as in primary (thymus and bonemarrow) and secondary (lymph nodes, Peyer’s patches, spleen) lymphoidorgans. Their function is to collect and present antigenic material to thelymphocytes (Bonasio and von Andrian, 2006).

Phagocytosis of microbes is a complex mechanism, which in the phagocytes ismediated by a combination of different surface receptors, such as complementreceptors that mediate internalization of complement-opsonized microbes andFc-receptors that mediate internalization of antibody-opsonized microbes(Aderem and Underhill, 1999; Ravetch, 1997; Underhill and Ozinsky, 2002).Salmonella also actively invades phagocytes utilizing effector proteinstranslocated by SPI-1 TTSS (Mèresse et al, 1999; Drecktah et al, 2006). Onceinternalized, Salmonella persists and multiplies in a Salmonella containingvacuole (SCV; Alpuche-Aranda et al, 1994). SCV avoids fusion with lateendosomes and lysosomes (Hashim, et al, 2000; Holden, 2002; Brumell andGrinstein, 2004; Buchmeier and Heffron, 1991) and Salmonella directs thematuration of SCV by a SPI-2 dependent fashion. The SPI-2 encoded proteinSpiC inhibits the fusion of SCV with lysosomes and endosomes (Uchiya et al,1999) and another SPI-2 gene product, SifA, is required for the SCV membraneintegrity (Beuzón et al, 2000; Holden, 2002; Waterman and Holden, 2003).

Conventional phagolysosome is very antimicrobial and contains, for example,reactive oxygen and nitrogen species and cationic antimicrobial peptides

13

(CAMP) that effectively kill bacteria (Linehan and Holden, 2003). Salmonellasenses the conditions within SCV and activates transcription of the virulencegenes, which protect it against the antimicrobial effectors in M s. SPI-2 TTSSencodes several enzymes that directly inactivate reactive oxygen and nitrogenmolecules. The PhoP/Q two-component regulatory system, on the other hand,activates expression of components that protect Salmonella against antimicrobialpeptides (Ohl and Miller, 2001).

Besides actively invading through the epithelium of the small intestine,Salmonella can also traverse the epithelial barrier within CD18-positivephagocytes (Vazquez-Torres et al, 1999). The migration of Salmonella from thegastrointestinal tract into the liver and spleen has been shown to mainly takeplace inside M s and DCs (Vazquez-Torres et al, 1999) and is most likelydependent on host cell-mediated proteolytic activity. Interestingly, gentamicin,which poorly penetrates eukaryotic cells, reduces bacterial loads in mice infectedwith Salmonella, which indicates that a proportion of Salmonella cells are, atleast transiently, outside the phagocytes (Bonina et al, 1998). Further, in typhoidfever, one third of the bacteria in patient blood are found outside phagocyticcells (Wain et al, 1998), suggesting that Salmonella disseminates alsoextracellularly and, therefore, is likely to require bacterial proteolytic activity.

14

3. The outer membrane protein PgtE of Salmonellaenterica

During the growth of S. enterica sv Typhimurium in SCV in a mouse M -likecell line J774A.1, the gene expression of Salmonella goes through significantchanges and, for example, the transcription of the pgtE gene of Typhimurium isupregulated and is five to ten times higher than in bacteria cultivated in the cellculture medium (Eriksson et al, 2003; Eriksson-Ygberg et al, 2006). The pgtEgene is present and is 98 % identical in all fully or partially sequencedchromosomes of S. enterica but, so far, the functions of the PgtE surfaceprotease that it encodes have only been studied with Typhimurium.

3.1 The omptin family of bacterial surface proteins

PgtE of S. enterica is a member of the omptin family of outer membraneaspartate proteases in Gram-negative bacteria. The biological functions andvirulence potential of the omptins have been characterized for some members ofthe family, in particular for Pla of Yersinia pestis (Kukkonen and Korhonen,2004). Pla has a central role in the virulence of Y. pestis, whereas the virulenceroles of the other omptins, including PgtE, have remained less clear. Expressionof Pla activity is associated with the invasive character of plague; deletion of plaincreased the LD50 of Y. pestis from 50 to approximately 107 bacteria in asubcutaneous infection model in mice. No effect of the pla deletion was detectedin an intravenous infection model (Sodeinde et al, 1992). These results indicatethat Pla enables the spread of Y. pestis from the subdermal infection site into thelymph nodes and the circulation. More recently, Sebbane et al (2006a; 2006b)showed that the pla gene is needed for the establishment of bubonic plague by Y.pestis and is one of the most highly expressed genes in bacteria in the bubo.

Y. pestis Pla activates the mammalian plasma proenzyme plasminogen (Plg) intothe broad-spectrum serine protease plasmin. Plg is a 90-kDa glycoprotein thatcirculates in plasma in high concentration (2 M) and also is present in otherbody fluids and within the extracellular matrix (ECM; Stephens and Vaheri,1993). Plg is activated to plasmin by a proteolytic cleavage of the site Arg561-Val562 by two mammalian Plg activators, tissue-type Plg activator (tPA) andurokinase-type Plg activator (uPA) (Myöhänen and Vaheri, 2004). Pla resemblesthe mammalian Plg activators in activating Plg by a cleavage of the same peptide

15

bond. Also other invasive human pathogens, streptococci and staphylococci,express Plg activators. Streptokinase and staphylokinase are secreted non-enzymatic Plg activators that form molecular complexes with Plg and plasminleading to the changes in conformation and specificity of Plg and plasmin(Rabijns et al, 1997; Wang et al, 1998). Plasmin has a central role in fibrinolysis(Collen and Lijnen, 1991), and it also degrades non-collagenous components ofthe ECM and activates pro-matrix metalloproteinases (proMMPs) (Liotta et al,1981; Saksela, 1985; Carmeliet et al, 1997). Pla proteolytically inactivates themain physiological inhibitor of plasmin, alpha2-antiplasmin ( 2AP) (Kukkonenet al, 2001) which is a 70 kDa serine protease inhibitor and circulates in plasmaat a concentration of 60-70 g/ml. 2AP is a fast inhibitor of free plasmin andforms an equimolar complex with plasmin in a two-stage process; first in areversible fashion and then with a covalent bonding between the AP reactivesite and the active site of plasmin (Coughlin, 2005). Pla-mediated Plg activationand 2AP inactivation result in an uncontrolled plasmin-mediated proteolysis.Further, Y. pestis Pla mediates bacterial binding to laminin and heparan sulfateproteoglycans that are major components of basement membranes (BM), and tomouse BM and human epithelial cell-derived ECM (Lähteenmäki et al, 1998;2001b). Pla-mediated Plg activation and 2AP inactivation promote damage ofBM and ECM in vitro (Lähteenmäki et al, 2000; 2005). Migrating eukaryoticcells and many bacterial pathogens that cause systemic infections utilize hostproteolytic systems to break tissue barriers and to potentiate cell metastasis(Lähteenmäki et al, 2001a; 2005; Mignatti and Rifkin, 1993; Travis et al, 1995;Vassalli et al, 1992); this may apply to Pla-positive bacteria as well.

Pla also enhances bacterial invasiveness into human epithelial and endothelialcells (Cowan et al, 2000; Lähteenmäki et al, 2001b) and Pla-mediated invasioninto HeLa cells has been shown to involve intracellular signalling andcytoskeletal rearrangement of the host cell (Benedek et al, 2004). Pla alsodegrades the complement component C3 (Sodeinde et al, 1992). Taken together,Pla is an astonishingly multifunctional virulence factor that functions inproteolysis, adhesion and invasion.

The members of the omptin family are orthologs with approximately 50-70 %sequence identity. Their mature forms have 290-301 amino acids and lack orhave a low content of cysteine (Kukkonen and Korhonen, 2004; Fig 1). Theycleave polypeptides after basic residues (Dekker et al, 2001). The structure of

16

OmpT of E. coli has been solved by crystallography (Vandeputte-Rutten et al,2001). It is a -barrel protein with 10 anti-parallel -strands connected by fourperiplasmic turns and five surface-exposed loops. The catalytic residues (Asp83,Asp85, Asp210 and His212) reside in a negatively charged cleft surrounded bythe five surface-exposed loops (L1-L5; Vandeputte-Rutten et al, 2001).Activation of OmpT into an enzymatically active form was seen with E. coli K-12 lipopolysaccharide (LPS) and probably leads to a subtle conformationalchange in the omptin structure (Kramer et al, 2000; 2002). The crystal structureof a ferrichrome-iron receptor FhuA, which is a -barrel outer membrane proteinof E. coli, has been solved in complex with LPS (Coulton et al, 1986; Fergusonet al, 1998). The three-dimensional structure shows that FhuA non-covalentlybinds a single LPS molecule. An LPS-binding motif of four basic amino acidswas identified in FhuA and subsequently detected in several prokaryotic andeukaryotic LPS-binding proteins (Ferguson et al, 2000). OmpT has three basicamino acids (Arg138, Arg175 and Lys226) of the motif in the correct spatialorganization (Vandeputte-Rutten et al, 2001). PgtE has two arginines (Arg138and Arg171) of the motif (Fig 1).

Figure 1. The surface protease

PgtE of S. enterica. The structureof PgtE was modelled based on the

coordinates of the OmpT crystalstructure (ExPDB1I78; Vandeputte-

Rutten et al, 2001) and the figurewas drawn using Swiss-Pdb viewer

(Guex and Peittsch, 1997;http://www.expasy.org/spdbv/).

The catalytic residues Asp84,Asp86, Asp206 and His208, the

residues Arg138 and Arg171 of theLPS binding motif, and two girdles

of aromatic residues that border theouter membrane are indicated.

Sodeinde and Goguen (1989) found that PgtE activates mammalian Plg toplasmin when expressed from a multi-copy plasmid in recombinant E. coli.However, they did not detect PgtE-mediated Plg activation in wild-type isolates

17

of Typhimurium (Sodeinde and Goguen, 1989). Guina et al (2000) have shownthat PgtE degrades -helical CAMPs, which suggests that PgtE may have a rolein bacterial resistance to innate immunity. In a vaccine study, PgtE was found todegrade viral epitopes that were fused into fimbriae of an attenuatedTyphimurium strain, and deletion of pgtE resulted in a higher antibody responseagainst the viral epitopes (Chen and Schifferli, 2003). These studies suggest thatPgtE is able to cleave various types of substrates. However, the biologicalfunctions and significance of PgtE in virulence of S. enterica have not beensystemically addressed.

PgtE resembles Pla in that it activates Plg, whereas OmpT is dramatically lessactive towards this substrate (Sodeinde and Goguen 1989; Kukkonen et al,2001). Substitution of critical surface-exposed amino acids converted OmpT intoa Pla-like enzyme (Kukkonen et al, 2001), which indicates that the differingsubstrate specifities in omptin proteolysis are dependent on their surface loopstructures.

3.2 The PhoP/Q two-component regulatory system in regulation of PgtEand other outer membrane components of Salmonella

Transcription of the pgtE gene of Salmonella is under the control of the PhoP/Qtwo-component regulatory system (Guina et al, 2000; Bader et al, 2003). ThePhoP/Q system regulates expression of c. 40 genes, including at least 15 genesencoding outer membrane components. The PhoP/Q system is important inintramacrophage survival of Salmonella (Fields et al, 1986), in bacterialresistance to antimicrobial peptides and acidic pH (Fields et al, 1989; Foster andHall, 1990; Groisman et al, 1992), in bacterial invasion into epithelial cells(Behlau and Miller, 1993), in formation of SCV (Alpuche-Aranda et al, 1994)and in reduction of the presentation of antigens by phagocytic cells (Wick et al,1995). It has also been shown that the PhoP/Q system has a role in Salmonella-induced human macrophage cell death (Detweiler et al, 2001) and that aTyphimurium mutant constitutively expressing PhoP has a reduced amount ofO-antigen which is the outermost part of the LPS (Guo et al, 1997; Baker et al,1999).

The PhoP/Q system consists of the inner membrane sensor/kinase PhoQ and thecytoplasmic response regulator PhoP. The PhoP/Q system responds to the levels

18

of Mg2+ and Ca2+ ions and to CAMPs in the periplasmic space of Salmonellacells (Bader et al, 2003; 2005; García Véscovi et al, 1996; Groisman, 2001).Growth of bacteria in millimolar concentrations of Mg2+ represses expression ofthe PhoP-activated genes, whereas the growth in micromolar concentrations ofMg2+ activates their expression. Ca2+ also is a physiologically relevant cation inregulation of the PhoP/Q system and high Ca2+ concentration repressestranscription of PhoP-activated genes (García Véscovi et al, 1996). Morerecently, Bader et al (2003) reported that the PhoP/Q system senses sublethalconcentrations of CAMPs. Bader et al (2005) have also shown that CAMPs actas the activating signals for the PhoP/Q system whereas the divalent cationsrepress the system by competitive inhibition of CAMP binding to PhoQ. Theysuggest that CAMPs displace the divalent cations from cation-binding sites inPhoQ and this way activate the PhoP/Q system inside SCVs.

Expression of the PhoP-activated genes is high in the SCV, where pH is acidic,the amounts of Mg2+ and Ca2+ ions are low and the concentration of CAMPs ishigh (Alpuche-Aranda et al, 1992; Bader et al, 2005). Some of the genes that areactivated by the PhoP/Q system are also dependent on another two-componentsystem, PmrA-PmrB (Roland et al, 1993). The PmrA/B system can either beactivated by the PhoP/Q system or by extracellular acidification (Soncini andGroisman, 1996). Thus, the expression of the genes that are under the control bythe PhoP/Q or the PmrA/B systems can be activated by a spectrum ofenvironmental signals. Recently, it was shown that the upregulation of pgtE bythe PhoP/Q system is indirect and mediated by a PhoP-regulated transcriptionalregulator SlyA (Navarre et al, 2005). Taken together, the regulation of geneexpression in Salmonella is complex and dependent on various regulatorysystems.

19

4. Lipopolysaccharide variability and modification inSalmonella

LPS is a major component of the outer leaflet of the outer membrane in Gram-negative bacteria. The LPS molecule consists of the endotoxic lipid A, theoligosaccharide core (further divided into the inner and the outer cores) and theO-specific side chain (O-antigen) that is composed of repeating oligosaccharideunits (Raetz and Whitfield, 2002). The lipid A and the inner core are highlyconserved within Enterobacteriaceae (Heinrichs et al, 1998, Bruneteau andMinka, 2003), whereas the O-antigens are remarkably diverse (Raetz andWhitfield, 2002). The O-antigen is a major factor in antigenic variability of theSalmonella cell surface and S. enterica has been divided into 46 O-serogroupsbased on differences in the O-antigen (Wang et al, 2002). Smooth LPS with longO-antigen is important virulence factor of pathogenic Salmonella isolates andprotects Salmonella against complement-mediated killing in serum (Joiner et al,1982; Murray et al, 2003; Yethon et al, 2000).

Many PhoP/Q-regulated genes function in the remodelling of the LPS during thegrowth of Salmonella within M s (Ernst et al, 2001). Further, Wzz-proteinscontrol the length and density of the O-antigen and are important for the serumresistance as well as for the interaction of Salmonella with M s. WzzST isrequired for the polymerization of the O-antigen with 16-35 serotype-specificoligosaccharide units of 3-4 carbohydrate moieties and WzzfepE forpolymerization of the O-antigen with over 100 oligosaccharide units (Murray etal, 2003; 2005; 2006). Murray et al (2006) suggested that an O-antigen withover 100 oligosaccharide units protects Salmonella against complement-mediated killing more efficiently than does a shorter O-antigen. On the otherhand, Murray et al (2006) showed that Salmonella with a shorter O-antigen arephagocytosed more effectively by M s. This is also supported by the result thatrough Salmonella isolates lacking the O-antigen are taken up by phagocytesmore efficiently than smooth Salmonella (Wick et al, 1994). Taken together, itmight be that the bimodal O-antigen length is advantageous for Salmonelladuring different stages of infection. Interestingly, inside the murine M s, thethree main gene clusters acting in the LPS biosynthesis (Reeves et al, 1996),wecDE (genes required for the oligosaccharide synthesis and processing), waa(genes required for the biosynthesis of the core region) and rfb (many of thegenes required for the synthesis of O-antigen) are downregulated (Eriksson et al,

20

2003, Eriksson-Ygberg et al, 2006). However; it is not known whetherexpression of wzzST and wzzfepE genes is modified inside M s.

Changes in the LPS structure may prevent the interaction of the positivelycharged CAMPs with the negatively charged lipid A and this way promote theresistance of Typhimurium against CAMPs (Ernst et al, 2001). PagP, forexample, is a PhoP-activated gene product and required for the addition ofpalmitate to lipid A (Guo et al, 1998), which decreases the permeability of theouter membrane and promotes the bacterial resistance to CAMPs. ThepmrE/prmF operon, on the other hand, is regulated by the PmrA/B and PhoP/Qsystems. PmrE and PmrF mediate the substitution of lipid A with cationicaminoarabinose and ethanolamine and this way mask negative charges of thecore and the lipid A and reduce the affinity of the positively chargedantimicrobial peptides to the bacterial surface (Gunn et al, 1998; Gunn, 2001).

5. Interaction of Salmonella with the complementsystem

5.1 Complement as an antimicrobial defence system

The complement system of serum is a key component of the innate immunitydefence against bacterial infections and bridges the innate and adaptiveimmunity (Rautemaa and Meri, 1999). It consists of a group of proteincomponents that circulate in blood and other body fluids mainly as inactivezymogens and whose activation cascade is initiated by the formation of theantibody-antigen complexes at the infection sites (classical pathway), by directrecognition of the foreign material such as bacterial cells (alternative pathway)or by binding of the mannose-binding lectin to mannose structures on thebacterial surfaces (lectin pathway) (Walport et al 2001a; 2001b). Activation ofthe complement cascade has many biological consequences. After activation,complement components induce the migration of phagocytes into the infectionsite, enhance the receptor-mediated phagocytosis by opsonizing microbes andlyse the bacterial cells by inserting the membrane attack complex (MAC) on thebacterial surfaces (Table 1).

The complement cascade is tightly regulated to prevent unnecessaryconsumption of the complement components and to confine the functions of the

21

complement within the infection site. The activation of any of the threecomplement pathways leads to the cleavage of the complement component C3into C3b that efficiently activates the complement cascade. The amount of C3 inserum is high (1.25 mg/ml) and therefore, the amount of C3b is easily increasedduring bacterial infections. To prevent undesired activation of the complementcascade, C3b is regulated in plasma by factor H (fH) and factor I (fI) that blockthe formation of C3b from C3 and enzymatically degrade C3b.

Table 1. Functions of the selected complement components. Many of the complement

components are synthesized as inactive proenzymes and are proteolytically converted to thefragments that either bind to the bacterial surface or are released into the microenvironment.

Complement protein Function

C3 Precursor of C3a and C3bC3a Anaphylatoxin: increases vascular permeability

C3b Opsonin; component of the C3-convertase of thealternative pathway; component of C5-convertase of all

pathwaysC4 Precursor of C4a and C4b

C4a AnaphylatoxinC4b Opsonin; component of the C3-convertases of the classical

and lectin pathwaysC5 Precursor of C5a and C5b

C5a Anaphylatoxin; chemotaxin: induces the migration of thephagocytes into the infection site

C5b Initiator of the MAC formationC5bC6C7C8C9 MAC complex

22

5.2 Resistance mechanisms of Salmonella against the complementsystem

In general, complement is cytolytic against Gram-negative bacteria. Manypathogens have developed mechanisms to evade complement attack and killing.S. enterica sv Typhi protects itself against complement attack with the Vipolysaccharide capsule (Looney and Steiqbiqel, 1986). The uncapsulatedTyphimurium, on the other hand, is protected by long heterogenic O-antigen sidechains of the smooth LPS. Smooth LPS is the major determinant of the serumresistance in Salmonella and mutants with an incomplete LPS are more easilylysed by the MAC (Grossman et al, 1987; Murray et al, 2003; Rautemaa andMeri, 1999).

Three outer membrane proteins of Salmonella, TraT, Rck and PagC, have alsobeen shown to protect Salmonella against complement. TraT is a surface proteinof E. coli and Typhimurium. In E. coli, traT resides in the F-like plasmid and thefunction of TraT is to prevent the conjugational transfer of DNA and formationof mating pairs between F+-cells (Rhen and Sukupolvi, 1988). TraT also protectsE. coli against complement-mediated killing and phagocytosis by M s (Moll etal, 1980; Agüero et al, 1984). In Typhimurium, TraT is encoded by the 94-kbvirulence plasmid (Rychlik et al, 2006) and it increases the serum resistance byinhibiting the insertion of MAC on the bacterial surface (Rhen and Sukupolvi,1988; Pramoonjago et al, 1992). Rhen and Sukupolvi (1988) suggested that TraTmay also have a role in survival of Salmonella inside M s.

Rck is a 17-kDa outer membrane protein of Typhimurium. It is also encoded bythe 94-kb virulence plasmid (Rychlik et al, 2006), and overexpressed Rckprotects rough Salmonella and E. coli against complement (Heffernan et al,1992a). The serum-resistance mechanism of Rck is that it inhibits the properMAC insertion on the bacterial surface and releases the complex from thebacterial cell (Heffernan et al, 1992a). Based on the predicted amino acidsequence, Rck belongs to the homologous family of outer membrane proteins ofEnterobacteriaceae. Other members of this family are Ail of Y. enterocoliticathat also mediates the bacterial serum resistance, OmpX of Enterobacter cloacaeand Lom of E. coli that have not been found to be involved in the virulence, andPagC of S. enterica (Heffernan et al, 1992b; Cirillo et al, 1996). InTyphimurium, pagC is a PhoP/Q regulated gene and required for the intra-M

23

survival and the full virulence of Typhimurium in mice (Miller et al, 1989).PagC has also been shown to protect S. enterica sv Choleraesuis against thecomplement-mediated killing by a so-far uncharacterized mechanism (Nishio etal, 2005).

6. Bacterial interference with the host extracellularmatrix and matrix-degrading proteolytic systems

6.1 Interaction of Salmonella with the host extracellular matrix

Adhesiveness to tissues is an important virulence factor of bacterial pathogens(Pizarro-Cerda and Cossart, 2006; Westerlund and Korhonen, 1993). Salmonellaexpresses nonfimbrial and fimbrial adhesins that bind to ECM and BMcomponents, mainly to fibronectin and laminin. The outer membrane proteinsRck and PagC, which also have a role in resistance of Salmonella againstcomplement, bind to laminin and to reconstituted BM, and Rck also binds tofibronectin (Crago and Koronakis, 1999). ShdA is a large surface protein thatbinds fibronectin (Kingsley et al, 2002) and seems to be important for thecolonization of the murine cecum by Typhimurium (Kingsley et al, 2003). MisLbinds to type IV collagen and fibronectin (Dorsey et al, 2005). Expression ofMisL was shown to mediate bacterial invasion into human colonic carcinomacell line T84 and colonization of mouse intestine (Dorsey et al, 2005).

From fimbrial adhesins, type 1 fimbriae mediate binding of Salmonella to high-mannose chains in laminin (Kukkonen et al, 1993) and the thin aggregativefimbriae (curli) to fibronectin (Collinson et al, 1993). Type 1 fimbriae and curliof Salmonella also bind Plg on the bacterial surface (Sjöbring et al, 1994;Kukkonen et al, 1998), and curli also binds tPA (Sjöbring et al, 1994). Cell-bound Plg is then effectively activated to plasmin by host tPA. tPA-activatedplasmin on Salmonella surface degrades laminin and ECM and enhancespenetration of Salmonella through the reconstituted BM (Lähteenmäki et al,1995). Bound plasmin is protected from 2AP because the binding of plasmin tobacterial cell and to 2AP involve the same lysine-binding structures in theplasmin molecule (Coughlin, 2005). As described above, PgtE activates Plg invitro (Sodeinde and Goguen, 1989), and Salmonella thus has both direct andindirect mechanisms to generate active plasmin. Taken together, the binding toECM and BM components and plasmin-mediated degradation of ECM and BM

24

may assist Salmonella to invade through tissue barriers and to colonize theinfection sites.

6.2 Matrix metalloproteinases in bacterial infections

The MMPs are a family of 25 zinc-dependent endopeptidases targeting the ECMcomponents. They have a role in cell migration, tissue remodelling andmodulation of inflammation and innate immunity (Parks et al, 2004; Sternlichtand Werb, 2001). Different MMPs have overlapping proteolytic affinities and,collectively, cause degradation of the ECM components (Sternlicht and Werb,2001). Based on the substrate specifities, MMPs are divided into collagenases,gelatinases, stromelysins and elastases. Additionally, MMPs include membrane-type MMPs that are surface-anchored. In addition to ECM components,substrates of MMPs include other proteinases, proteinase inhibitors, cytokinesand chemokines (McCawley and Matrisian, 2001).

The MMPs are secreted as inactive proenzymes by various cell types, e.g., bymonocytes, M s and DCs (Harper et al, 1971; Sternlicht and Werb, 2001). TheMMPs are homologous in their overall structure (Fig 2) and have a zinc-bindingcatalytic domain, an N-terminal pre-domain with a signal sequence and a pro-domain with a conserved cysteine (Cys) that chelates the catalytic Zn2+-ion andkeeps the enzyme in a pro-enzyme form (Stöcker et al, 1995). Most MMPs alsocontain a C-terminal hemopexin/vitronectin-like domain that binds substratesand tissue inhibitors of MMPs and is involved in activation of some MMPs(Sternlicht and Werb, 2001). Various molecules, such as other MMPs, plasminand trypsin, are able to disrupt the Cys-Zn2+ binding that maintains the latentform of proMMP. This leads to an intermediate activation of MMPs. Anothercleavage, which can be autocatalytic, is needed to fully activate the enzyme(Johansson et al, 2000; Woessner and Nagase, 2002). Some MMPs also haveadditional binding sites and, for example, MMP-9 has collagen/gelatin bindingdomains (Allan et al, 1995; Woessner and Nagase, 2002).

The gelatinases, MMP-9 (92 kDa gelatinase; gelatinase B) (Fig 2) and MMP-2(72 kDa-gelatinase; gelatinase A), have a role in many physiological andpathological conditions and their role in cancer biology has been intensivelystudied. MMP-9 and MMP-2 are expressed by various cell types, such asmonocytes and M s (Johansson et al, 2000; Mohan et al, 1998; Munaut et al,

25

1999). Gelatinases degrade a variety of ECM proteins; i.e. type IV collagen thatis a central component in BMs, type V, VII, X, XI and XIV collagens, gelatin(denaturated collagen), elastin, proteoglycan core proteins and fibronectin(Johansson et al, 2000). MMP-2 also degrades native type I collagen (Aimes andQuigley, 1995) suggesting that it has a role in the remodelling of collagenousECM. However, more important in the initial degradation of ECM collagens arethe collagenases (MMP-1, -8 and -13) that cleave fibrillar collagens to fragmentsthat are rapidly denaturated to gelatin at the body temperature and therebybecome susceptible to the degradation by gelatinases (Billinghurst et al, 1997;Johansson et al, 1997).

Figure 2. Domain structure of proMMP-9. Pre; N-terminal signal sequence, Pro; propeptidewith a conserved Cys residue, Catalytic; catalytic domain includes the collagen binding domains

(II) and the Zn2+-binding site (Zn). Hemopexin; four C-terminal hemopexin/vitronectin-likedomains, the first and last of which are linked by a disulfide bond. The catalytic domain is linked

to the hemopexin domain by a hinge region.

Most MMPs are not constitutively expressed by cells in vivo, but theirexpression is induced by exogenous signals such as contact with ECM ormicrobial surface components (Johansson et al, 2000; Opdenakker et al, 2001).Both bacterial and viral infections have been shown to induce production ofMMPs from host cells such as monocytes and M s (Elkington et al, 2005). Alimited number of bacteria, Pseudomonas aeruginosa, Vibrio cholerae,Porphyromonas gingivalis, Treponema denticola, pneumococci and group Astreptococci have been shown to produce proteolytic enzymes that activate hostproMMPs (Oggioni et al, 2003; Okamoto et al, 1997; Pattamapun et al, 2003;Sorsa et al, 1992; Tamura et al, 2004). Some bacterial species have also beenshown to express collagenases and gelatinases (Harrington, 1996; Potempa et al,2000; Watanabe, 2004). LPS from Salmonella and other enteric bacteria inducessecretion of proMMP-9 in M s (Xie et al, 1994). No reports of Salmonella-mediated activation of proMMPs have been published; however, mice deficient

26

in MMP-9 are more resistant to Typhimurium-induced enterocolitis than wild-type mice (Castaneda et al, 2005) and MMP-3 deficient mice are more resistantto systemic Typhimurium infection than wild-type mice (Handley and Miller,2007). These results suggest a role for MMPs in Salmonella infections.

27

7. Aims of the study

When this study was initiated, it was already known that PgtE of S. enterica is aPlg activator. However, the activation was only seen with recombinant bacteriaand remained cryptic in isolates of Salmonella grown in laboratory culture media(Sodeinde and Goguen, 1989; Lähteenmäki et al, 1995). On the other hand, Plaof Y. pestis was known to be a multifunctional Plg activator and OmpT of E. coliwas known to require LPS to be proteolytically active. It was also known thatduring growth of Salmonella inside SCV of M , pgtE is upregulated and severalgenes involved in LPS synthesis are downregulated (Eriksson et al, 2003). ThisPhD work aimed to address the role of PgtE activity in S. enterica pathogenesis.In particular, important goals in this study were

-to assess the conditions where PgtE is active-to study how LPS effects the activity of PgtE-to compare the proteolytic and nonproteolytic functions of Pla and PgtE-to identify biologically important substrates for PgtE

28

8. Materials and methods

Table 2. Bacterial strains and cultured eukaryotic cell line used in this study.

Bacterial strains Relevant characteristics Article Reference

E. coli

XL1-Blue MRF´ Rough LPS I, II, III, IV StratageneIHE3070 Rough LPS I Selander et al, 1986IHE3074 Rough LPS I Selander et al, 1986IHE3034 O18 LPS I Selander et al, 1986IHE1049 O1 LPS I Pere et al, 1988M15 I, IV QiagenBL21 DE F- hsd ompT lon I Novagen

S. enterica sv

TyphimuriumSL1102 Re LPS I Wilkinson et al, 1972SL1102-1 pgtE derivative of SL1102 I This study14028 Smooth LPS I, II, IV ATCC14028-1 pgtE derivative of 14028 I, II, IV This study14028R Rough LPS derivative of 14028 II, III, IV Wick et al, 199414028R-1 pgtE derivative of 14028R II, III, IV This studySH401 Rough LPS I Kukkonen et al, 1993SL696 Smooth LPS I Wilkinson et al, 1972SL901 Semirough LPS I Wilkinson et al, 1972

MinnesotaSF1112 Ra LPS I Schmidt and Lüderitz, 1969SF1127 Rb1 LPS I Schmidt and Lüderitz, 1969SF1119 Rc LPS I Schmidt and Lüderitz, 1969SF1121 Rd1 LPS I Schmidt and Lüderitz, 1969SF1118 Rd2 LPS I Schmidt and Lüderitz, 1969SF1167 Re LPS I Schmidt and Lüderitz, 1969

Y. pseudotuberculosis

Pa3606 01b LPS, pYV- I Tsubokura and Aleksic, 1995PB1 01b LPS, pYV- I Perry and Brubaker, 1983PB1 wb Lacks O-antigen

due to wb, pYV- I Obtained from J.A. Bengoechea

Eukaryotic cell lines

J774A.1 Murine M -like cell line II, IV ATCC (TIB-67)ECV-304 Endothelial-like cell line I ATCC (CRL-1998)

29

Table 3. Plasmids used and constructed in this study.

Plasmid Relevant characteristics Article Reference

pSE380 trc promoter, lacO operator, lacI I, II, III, IV InvitrogenpMRK3 pgtE in pSE380 I, II, III, IV This studypMRK31 PgtE D206A I, II, III, IV This studypMRK32 PgtE R138E I This studypMRK33 PgtE R171E I This studypMRK34 PgtE R138E R171E I This studypMRK1 pla in pSE380 I, IV Kukkonen et al, 2001pMRK2 ompT in pSE380 I Kukkonen et al, 2001pQE30 T5 promoter, lac operator, His6-tag I, IV QiagenpREP4 lacI I, IV Qiagen

pla/ pgtE-hybrids in pSE380

pMRK11 Pla with PgtE 1-40 IV This studypMRK12 Pla with PgtE 52-130 IV This studypMRK13 Pla with PgtE 143-189 IV This studypMRK14 Pla with PgtE 190-230 IV This studypMRK15 Pla with PgtE 235-292 IV This studypMRK1.32 Pla 161KGVRVIGYN 161HGVRGIGYS IV This studypMRK1.41 Pla 212D 212K IV This studypMRK1.51 Pla 259TIDKN 259IIDKT IV This studypMRK1.52 Pla 267SVSI 267TAYF IV This studypMRK1.53 Pla 278SNK 278ANN IV This study

pla/ pgtE-hybrids in pMRK11 (all contain PgtE 1-40)

pMRK1122 Pla 89NENQSE 89SSEQPG IV This studypMRK1123 Pla 98HSS 98RSI IV This studypMRK1131 Pla 149SYNNGAYT 149IYDNGRYI IV This studypMRK1132 Pla 161KGVRVIGYN 161HGVRGIGYS IV This studypMRK113 Pla 149SYNNGAYT 149IYDNGRYI and

Pla 161KGVRVIGYN 161HGVRGIGYS IV This studypMRK1141 Pla 212D 212K IV This studypMRK1151 Pla 259 TIDKN 259IIDKT IV This studypMRK1152 Pla 267SVSI 267TAYF IV This studypMRK1153 Pla 278SNK 278ANN IV This studypMRK115 Pla 259TIDKN 259IIDKT and

Pla 267SVSI 267TAYF andPla 278SNK 278ANN IV This study

30

Table 4. Methods used in this study.

Method Described and / or used in

Cultivation of S. enterica I, II, III, IVInduction of PgtE expression under the PhoP/Q inducing conditions I, II, III, IVInduction of omptin production in recombinant bacteria I, II, III, IVCultivation of eukaryotic cells I, II, IV

Allelic replacements I, IIGeneration of point mutations IGeneration of pla/pgtE hybrid constructs IVDNA sequencing I, IVWestern blotting to characterize the omptin expression I, II, IV

Expression of the recombinant His6-tagged omptin in E. coli I, IVPurification and reactivation of the His6-tagged omptin I, IV

Plg cleavage and activation I, II, IVAdhesion to Matrigel IInvasion into eukaryotic cells I

2AP cleavage and inactivation IILaminin degradation IIInfection of J774A.1 M -like cells with the phagocytosis assay II, IVPreparing of J774A.1 cell lysates II, IVIsolation of SCVs from J774A.1 cells II, IVSilver staining IIComplement protein cleavage IIISerum sensitivity assays IIIDegradation of gelatin IVActivation of proMMP-9 IVGelatin zymography IVMouse infection experiments IV

31

9. Results and discussion

9.1 Role of lipopolysaccharide for the activity of PgtE of Salmonellaenterica and Pla of Yersinia pestis (I, II, IV)

Virulent S. enterica isolates normally have smooth LPS, and although thesebacteria have an intact pgtE gene, they do not express PgtE activity (Sodeindeand Goguen, 1989; Lähteenmäki et al, 1995). However, PgtE was active whenoverexpressed in recombinant E. coli K-12 (Sodeinde and Goguen, 1989) whichhas a rough, Rc type of LPS (Holst, 2000). To assess the effect of the O-antigenon PgtE functions we first studied Plg activation by the smooth, widely usedvirulent strain S. enterica sv Typhimurium 14028 and two rough Salmonellastrains, SL1102 and 14028R. Bacteria were cultivated in the presence ofPhoP/Q-inducing N-minimal medium, which mimics the intracellular conditionsof SCV and has a low Mg2+ concentration. The growth in the N-minimalmedium was found to induce the expression of PgtE in all strains (Fig 1B in I;Fig 4 in II). The rough SL1102 and 14028R efficiently activated Plg, whereasthe smooth 14028 only poorly promoted formation of plasmin (Fig 1A in I; Fig3). S. enterica 14028R, which is an isogenic derivative of 14028, was morepowerful in Plg activation than SL1102 (Fig 3). As rough LPS seemed to becritical for PgtE activity we studied the effect of O-antigen length on PgtEfunctions. PgtE-encoding pMRK3 was expressed in a series of S. entericaisolates, which express LPS molecules of different length and were tested for Plgactivation. The strains with a rough LPS (chemotypes Ra, Rb1, Rc, Rd1, Rd2, Re)were efficient, whereas the strain with smooth LPS was poor in Plg activation(Fig 1C in I). Plg activation by the strain with a semirough LPS (LPS containsonly one O-antigen unit) was slightly decreased as compared to the rough strains(Fig 1C in I). Further, Plg activation was seen in rough but not in smoothrecombinant E. coli transformed with pMRK3 (Fig 4 in I).

To confirm that PgtE has a role in Plg activation by S. enterica, we constructed asite-specific deletion of pgtE in 14028, SL1102 and 14028R. The resulting pgtE-deficient derivatives 14028-1, SL1102-1 and 14028R-1 were negative in Plgactivation (Fig 1A in I; Fig 3). Complementation of the pgtE-negative SL1102-1and 14028R-1 with the PgtE-encoding pMRK3 restored Plg activator activity(Fig 1A in I; Fig 3). A low Plg activator activity was also seen with 14028-1(pMRK3) (Fig 1A in I), which overexpressed pgtE. The expression levels of

32

PgtE were similar in the complemented recombinant derivatives used in theassays (Fig 1B and 1D in I).

Figure 3. Plg activation by smooth and rough S. enterica. Plg activation was measured by

incubating bacteria and human Plg with the chromogenic plasmin substrate. The strains used inthis assay were (A) the smooth pgtE-proficient S. enterica 14028 and the rough pgtE-proficient S.

enterica 14028R and SL1102 and (B) S. enterica 14028R, 14028, pgtE-negative 14028R-1 and14028-1, the PgtE-expressing 14028R-1(pMRK3) and 14028R-1(pSE380) with the vector

plasmid. The means of duplicate samples in a representative assay are shown.

We found that Pla of Y. pestis also is active only in bacterial strains with roughLPS. The Pla-encoding pMRK1 was transformed into the temperature-sensitiveYersinia pseudotuberculosis strains Pa3606 and PB1 and into the rough wbderivative of PB1. Plg activation was tested at 25°C and 37°C, which are thetemperatures where Y. pseudotuberculosis Pa3606 and PB1 express smooth andrough LPS, respectively (Bengoechea et al, 1998). Pa3606(pMRK1) andPB1(pMRK1) were efficient in Plg activation after cultivation at 37°C. No Plgactivation was detected with cells from 25°C (Fig 2A, 2B and 2C in I). Therough strain PB1 wb(pMRK1) activated Plg after growth at both temperatures(Fig 2C in I). The LPS types in these strains were confirmed by silver stainingand electrophoresis (data not shown). Similarly, Pla activity was detected inrough but not in smooth recombinant E. coli (Fig 4 in I). The requirement of theLPS on Pla and PgtE activity was also shown with the purified proteins (Fig 2D

33

in I; Fig 4). As previously noted for the purified OmpT of E. coli (Kramer et al,2000; 2002), His6-Pla and His6-PgtE that were purified under denaturizingconditions were enzymatically active after reconstitution with rough LPS andinactive when reconstituted in the absence of LPS or in the presence of smoothLPS (Fig 2D in I; Fig 4).

Figure 4. Plg activation by purified PgtE. Plg

activation by purified His6-PgtE in the presenceof rough, Rc type of LPS or in the absence of

LPS. Plg activation by LPS alone is alsoshown. Plg activation was measured by

incubating the purified PgtE and human Plgwith the chromogenic plasmin substrate. The

means of triplicate samples are shown.

A consensus LPS-binding motif of four basic amino acids has been identified inprokaryotic and eukaryotic LPS-binding proteins. This LPS-binding motif ispartially present in the OmpT sequence, and two amino acids, Arg138 andArg171, of the LPS-binding motif are present in PgtE and Pla in the correctspatial organization to form electrostatic and hydrogen bonds with 4’ phosphatein lipid A (Ferguson et al, 2000; Vandeputte-Rutten et al, 2001; Kramer et al,2002; Fig 1). To study the role of Arg138 and Arg171 in PgtE functions, wesubstituted the arginines for glutamates separately (in plasmids pMRK32 andpMRK33) and together (pMRK34). Salmonella SL1102-1 was complementedwith each of the three plasmids and tested for Plg activation. The derivativeswith the substitutions PgtE R138E and PgtE R171E were as efficient in Plgactivation as SL1102-1(pMRK3) complemented with pgtE, whereas noactivation was seen with the derivative harbouring the substitutions PgtE R138ER171E (Fig 3A in I).

34

Pla is an invasin and an adhesin (Lähteenmäki et al, 1998; 2001b) and westudied whether PgtE also is active in these functions. Both Pla-encodingpMRK1 and PgtE-encoding pMRK3 enhanced adhesion of rough recombinantE. coli isolates onto the ECM secreted by the ECV-304 cells (Fig 4 in I). Pla alsomediated the invasion of rough E. coli strains into the ECV-304 cells whereasPgtE was negative in this function (Fig 4 in I). These results indicate that PgtE isan adhesin but not an invasin. The smooth Pla- and PgtE-expressing E. coliisolates were negative in both invasion and adhesion (Fig 4 of I), which is inagreement with the O-antigen masking functions of Pla and PgtE. Adhesion ofPla does not depend on the proteolytic activity as the protease-negative Pla isalmost as adhesive to the ECM secreted by the ECV-304 cells as the wild-typePla (Lähteenmäki et al, 2001b). It remains open whether also the PgtE-mediatedadhesion is independent of the proteolytic activity. Plg activation furtherconfirmed that only E. coli isolates with rough LPS expressed functions of eitherPgtE or Pla (Fig 4 in I). Our results also confirm that OmpT-expressing bacteriaare poor in invasion, adhesion and Plg activation (Fig 4 in I; Lähteenmäki et al,1998; 2001b).

These results indicate that PgtE and Pla have a dual interaction with LPS. Theyrequire rough LPS to be functional but are sterically inhibited by the long O-antigen side chains of LPS. Similarly, the O-antigen has been shown to interferewith the function of Y. pseudotuberculosis invasin-protein when expressed inShigella flexneri (Voorhis et al, 1991). Y. pestis has evolved from Y.pseudotuberculosis O1b and has a rough LPS due to inactivation of five genes inthe gene clusters encoding the LPS biosynthetic pathway (Achtman et al, 1999;Skurnik et al, 2000). This far, the advantage of the rough LPS for Y. pestis hasremained unknown (Skurnik et al, 2000; Parkhill et al, 2001). Our resultsindicate that the lack of O-antigen allows efficient Pla-mediated proteolysis,invasion and adhesion of Y. pestis, and our hypothesis is that this is the mainadvantage for the rough LPS in Y. pestis. In virulent S. enterica isolates, on theother hand, activity of PgtE is sterically inhibited by the long O-antigen and isonly seen in rough derivatives of Salmonella when cultivated in vitro (I, II, IV).

35

9.2 Activity of PgtE in Salmonella released from macrophages (II, IV)

Transcription of pgtE is under the control by the PhoP/Q system and upregulatedin mouse M s. Therefore, we studied the effect of intracellular conditions onPgtE functions by infecting murine J774A.1 M -like cells with Salmonella14028 and the pgtE-negative mutant 14028-1. The M s were lysed after 20hours of infection and tested for 2AP degradation. 2AP was chosen here as amodel substrate for PgtE-mediated proteolysis after our finding that 14028R, butnot 14028R-1, cleaved and inactivated this biologically important antiproteaseafter growth in the PhoP/Q inducing conditions (Fig 1A and 1B in II; see 9.5.1).For comparison, Salmonella 14028 was cultivated in vitro in the PhoP/Qinducing N-minimal medium as well as in Luria broth. When bacteria weregrown in the laboratory media, the PgtE activity in 14028R was significantlyhigher than in 14028 (Fig 1 and 4 in II), which is compatible with theinterference of PgtE functions by the O-antigen discussed above. Notably, thelysate from M s infected with Salmonella 14208 was far more efficient in 2APdegradation than 14028 from Luria broth or from N-minimal medium (Fig 3 inII). The lysates from J774A.1 cells infected with the pgtE-negative mutant14028-1 and from uninfected J774A.1 cells were inactive in 2AP degradation(Fig 3 in II), indicating that PgtE was needed for the antiprotease cleavage. Wealso isolated SCVs from 14028- and 14028R-infected J774A.1 cells to decreasethe amount of J774A.1 cell material contaminating the test samples. Bacteriafrom SCVs were active in 2AP degradation, with 14028 and 14028R showingsimilar activity (Fig 4 in II). Likewise, Salmonella 14028 from M s efficientlydegraded also another polypeptide substrate, gelatin (Fig 3A and 3B in IV; seechapter 9.5.3), which indicates that the enhancement of the proteolytic activityof PgtE in M s is not associated with the antiprotease substrate but rather is ageneral phenomenon.

To study whether the increased proteolytic activity of Salmonella 14028 fromthe infected M s results from the increased production of PgtE, or from themodification of the O-antigen of LPS, or both, we analyzed the bacteria fromSCVs by Western blotting with anti-PgtE antibodies and by silver staining fordetection of LPS. PgtE expression was high in both 14028 and 14028R fromSCVs compared to bacteria grown in Luria broth (Fig 4 in II). PgtE expressionwas also increased in N-minimal medium but was not as high as in bacteria fromSCVs (Fig 4 in II). Further, silver staining revealed that O-antigen length was

36

reduced during growth of Salmonella 14028 inside M s as compared to 14028grown in Luria broth (Fig 4 in II). Growth of 14028 in N-minimal mediumpromoted only a slight modification in the LPS carbohydrate chain pattern (Fig 4in II).

Our results indicate that the PgtE expression is upregulated and the O-antigensynthesis downregulated during the growth of Salmonella within M . Theseresults are in accordance with the data showing increased transcription of 14028pgtE in J774A.1 M -like cells (Eriksson et al, 2003) and the role of the PhoP/Qsystem in pgtE expression (Guina et al, 2000; Bader et al, 2003). Our resultsalso show that both the increased PgtE-expression level and the reduced O-antigen length are important for proteolytic activity of PgtE and that PgtE isfully active after the growth of Salmonella within M . As it is thought thatSalmonella is continuously released from M s to spread into new ones(Sheppard et al, 2003; Sheppard and Mastroeni, 2004), we suggest that PgtE isfunctional during these transient extracellular growth phases and thus canfacilitate bacterial dissemination from one M to another.

9.3 Role of PgtE in survival of Salmonella inside macrophages in vitro(II)

In vitro, PgtE degrades CAMPs and promotes resistance of Salmonella againstCAMP-mediated killing (Guina et al, 2000), which led us to evaluate thepossible role of PgtE in survival of Salmonella within M s. During the 20-hourinfection of murine J774A.1 M s, the pgtE-proficient Salmonella 14028 cellsmultiplied to approximately 170 % of internalized bacteria. On the contrary, thenumber of pgtE-negative 14028-1 cells reduced to approximately 50 % (II; Fig5), suggesting that PgtE promotes the intracellular survival of Typhimuriuminside murine M s. It remains open whether the increased survival observed inthis study results from the previously reported degradation of CAMPs by PgtE(Guina et al, 2000) or whether additional factors are involved. Typhimurium isknown to survive better in murine M s than in human M s (Schwan et al,2000), and PgtE was found not to have a role in survival of Typhimurium 14028within human primary M s and DCs (Pietilä et al, 2005). This may indicate thatPgtE degrades CAMPs of murine M s, but not those of human M s. It mayalso be that the major targets of PgtE are not intracellular but extracellular (see

37

chapter 9.5) and that PgtE is more important after release of Salmonella from thehost cell.

Figure 5. Effect of PgtE on survival of

Salmonella inside murine M s. J774A.1 cellswere infected with Salmonella 14028 and the pgtE-

negative derivative 14028-1. After a 20-hourincubation, J774A.1 cells were lysed and the

amount of bacteria were calculated by viablecounting relative to the bacterial amount at time

point 0 hour. The means and ranges from triplicatewells of a representative assay are shown.

9.4 Effect of PgtE on survival of Salmonella in mice (IV)

To study whether PgtE has a role in virulence of Salmonella in vivo we infectedBALB/c mice intraperitoneally with Salmonella 14028 or 14028-1. After one orthree days of infection, bacterial loads in the livers and spleens were determinedby viable counting of homogenized organs. One day post infection, the amountof Salmonella 14028 was nearly ten times higher than the number of 14028-1both in the liver and in the spleen (Fig 4 in IV). Three days post infection thedifference between 14028 and 14028-1 was not as high and the amount ofbacteria had increased at the same rate, even if Salmonella 14028-1 stillremained at a lower level than 14028 (Fig 4 in IV). These results indicate thatPgtE has a role in survival of Salmonella within the murine host in vivo, andsuggest that it can be important during early stages of infection when Salmonellais migrating into the host organs.

38

9.5 Identification of novel substrates for PgtE

9.5.1 Alpha2-antiplasmin (II)

As shown above (see chapter 9.2), we found that the rough S. enterica 14028Rcleaved 2AP when grown to produce PgtE. Cleavage of the 70 kDa 2AP to aapproximately 50 kDa cleavage product was detected with the rough 14028R butnot with the pgtE-negative derivative 14028R-1 (Fig 1A in II). Complementationof 14028R-1 with pgtE in trans with the plasmid (pMRK3) restored the abilityof 14028R-1(pMRK3) to degrade 2AP (Fig 1A in II). Similarly to Salmonella14028R and 14028R-1(pMRK3), E. coli XL1(pMRK3) cleaved 2AP into a 50kDa cleavage product (Fig 2A in II). These results confirm the role of PgtE in

2AP degradation. The catalytic residue Asp206 (Vandeputte-Rutten et al, 2001)is conserved in the omptin proteins and needed for Plg activation by Pla(Kukkonen et al, 2001). We substituted Asp206 with alanine in plasmidpMRK31 and observed that the derivative 14028R-1(pMRK31) failed to degrade

2AP (Fig 1A in II), which indicates that the proteolytic activity of PgtE isimportant for 2AP degradation.

To test whether PgtE-mediated degradation of 2AP also leads to inactivation ofthe protease inhibitor, we incubated bacteria first with 2AP and then withplasmin. The residual inhibitory 2AP activity was estimated by measuringplasmin activity. 2AP incubated with PgtE-expressing Salmonella 14028R,14018R-1(pMRK3), and E. coli XL1(pMRK3) failed to inhibit plasmin activity,whereas after incubation with 14028R-1 and 14028R-1(pMRK31), 2AP stillefficiently inhibited plasmin activity (Fig 1B and 2B in II). This indicates thatPgtE inactivates the protease inhibitor 2AP.

We next studied the effect of PgtE-mediated Plg activation and 2APinactivation on degradation of laminin. Laminin is an important component ofECM and a well-characterized substrate of plasmin (Salonen et al, 1984) andincubation of XL1(pMRK3) with Plg on immobilized, 125I-labelled lamininpromoted release of radioactivity from laminin (Fig 2C in II). Only a smallrelease of radioactivity was detected with XL1(pSE380) harbouring the vectorplasmid. Addition of 2AP only slightly reduced the laminin degradationmediated by suspensions of XL1(pMRK3) and Plg, whereas laminin degradationwith plasmin alone was efficiently inhibited when 2AP was added (Fig 2C in

39

II). The result shows that PgtE potentiates plasmin-mediated degradation of abiologically important substrate, laminin, in the presence of an importantantiprotease.

The importance of the 2AP degradation in the interference with the host Plgsystem was further indicated with the lysates of Salmonella-infected J774A.1M -like cells. M s activate Plg to plasmin by an uPA-dependent mechanism(Vassalli et al, 1992) and lysates from both Salmonella-infected and uninfectedM s converted Plg to plasmin (Fig 5 in II). Addition of 2AP completelyinhibited the slow activation of Plg by the lysates from uninfected or 14028-1infected cells (Fig 5 in II), but had only a limited effect on the rapid Plgactivation by the lysate from 14028-infected M s (Fig 5 in II). It is noteworthythat 14028R cleaved 2AP nearly completely after being cultivated in N-minimal medium, whereas degradation by 14028R from SCVs was only partial(Fig 4 in II). A possible explanation for this is that during isolation of SCVs,inhibitory factors from the M s are enriched into SCV preparations andtherefore lower the ability of PgtE to cleave 2AP. However, the lysates ofinfected M s were highly proteolytic (Fig 3 and 5 in II, Fig 3 in IV), suggestingthat the putative inhibitors of PgtE are absent or low in the lysates.

The Plg system and other proteolytic systems in the host are strictly regulated byprotease inhibitors that represent almost 10 % of the total protein mass of plasma(Travis and Salvesen, 1983). A few invasive bacterial species including oralpathogens, Y. pestis (Pla-mediated inactivation of 2AP), Staphylococcus aureusand Pseudomonas aeruginosa have been found to inactivate host antiproteases(Carlsson et al, 1984; Potempa et al, 1986; Grenier, 1996; Kukkonen et al,2001). The inactivation is thought to enhance uncontrolled proteolysis and topromote the migration of these bacteria within the host. Our results indicate thatalso PgtE of Salmonella proteolytically degrades and inactivates the antiprotease

2AP. Both 2AP inactivation and Plg activation interfere with the balance ofthe host Plg system and may enhance the dissemination of Salmonella throughtissue barriers by plasmin-mediated degradation of ECM components. Comparedto Pla of Y. pestis PgtE is less effective in Plg activation (Fig 4 in I). As plasminconverts prouPA into an active enzyme (Vassalli et al, 1992), Plg activation byPgtE may be adequate to initiate the uPA-mediated plasmin formation on thesurface of M s which secrete uPa and express uPA receptors (Vassalli et al,1992). Therefore, the ability of PgtE to efficiently inactivate the plasmin

40

inhibitor 2AP may be more important than Plg activation itself for creatinguncontrolled plasmin proteolysis of tissue barriers during the Salmonella-infection.

9.5.2 Complement components C3b, C4b and C5 (III)

Smooth Salmonella isolates are normally resistant against the complementmediated killing due to presence of long O-antigen chains in the LPS (Joiner etal, 1982; Grossman et al, 1987; Murray et al, 2003; 2005; 2006). However, asshown in article II, the O-antigen length of Typhimurium is dramaticallyreduced during the growth of Typhimurium within M s. Therefore, wereasoned that Salmonella requires other mechanisms to protect itself against thecomplement. Pla of Y. pestis has been reported to cleave C3 (Sodeinde et al,1992) and we studied whether also PgtE degrades the complement componentsC3b, C4b and C5. These proteins belong to the group of 2-macroglobulins(Budd et al, 2004) and are crucial components of the complement cascade. ThePgtE-expressing S. enterica 14028R, 14028R-1(pMRK3) and E. coliXL1(pMRK3) cleaved C3b, C4b and C5 into smaller degradation fragments (Fig1 and 2 in III). XL1(pMRK3) also degraded intact C3 and C4 (data not shown).The PgtE-mediated fragmentation of C3b was partly similar to the C3b cleavagemediated by fI and fH, suggesting that the PgtE-mediated degradation of C3bleads to its inactivation (Fig 1A in III). The pgtE-deficient 14028R-1, 14028R-1(pSE380) and XL1(pSE380) as well as 14028R-1(pMRK31) andXL1(pMRK31) expressing PgtE with a catalytic-residue substitution D206Awere inefficient in complement protein cleavage (Fig 1 and 2 in III). Theseresults indicate that complement protein cleavage is due to PgtE-mediatedproteolysis. Plasmin is known to degrade C3b (Seya et al, 1985), and Plgactivation by staphylokinase of S. aureus has been shown to lead to C3bcleavage (Rooijakkers et al, 2005). We found that cleavage of C3b by 14028Rwas increased in the presence of Plg (Fig 3 in III).

To study whether the PgtE-mediated cleavage of the complement componentshas a role in the serum resistance of Salmonella, Salmonella 14028R and14028R-1 were first cultivated in PhoP/Q-inducing conditions to activate PgtE-expression and then incubated in the presence of 2% normal human serum(NHS). The pgtE-proficient 14028R grew to higher numbers than did 14028R-1(Fig 4A in III). On the other hand, in heat-inactivated serum both strains grew

41

equally well and the amount of 14028R was 330% and amount of 14028R-1 was370% higher than in the beginning of the incubation (data not shown). Takentogether, the results suggest a positive effect of PgtE on the bacterial serumresistance. In accordance with the role of PgtE in the serum resistance, weobserved that after a 1-hour incubation, approximately 30% of Salmonella14028R-1(pMRK3) cells survived in serum whereas 14028R-1(pSE380) wasalmost totally killed (Fig 4B in III). Also E. coli XL1(pMRK3) was moreresistant to NHS than XL1(pSE380) (Fig 4C in III).

These results suggest that the PgtE-mediated proteolytic cleavage of thecomplement components C3b, C4b and C5 promotes the serum resistance ofSalmonella. C3b cleavage was enhanced by the PgtE-mediated Plg activation,further suggesting the importance of PgtE in the interference with the Plgsystem. PgtE is not the only surface protein of Salmonella that has an impact onbacterial survival in serum (Rhen and Sukupolvi, 1988; Heffernan et al, 1992a;Nishio et al, 2005), but proteolytic degradation of complement components bySalmonella has not been published earlier.

As expected, Salmonella 14028 and 14028-1 were highly serum resistant aftergrowth in Luria broth or in the PhoP/Q inducing conditions (data not shown). Incontrast, Luria-grown 14028R and 14028R-1 were killed in serum (data notshown), which was expected as they lack complement resistance conferred bythe O-antigen chains and because cultivation of Salmonella in Luria broth doesnot upregulate the PgtE expression. However, after growth in the PhoP/Qinducing conditions both 14028R and 14028R-1 showed increased serumresistance, suggesting that the simultaneous upregulation of PgtE and othercomponents protect rough Salmonella against complement-mediated killing (Fig4 in III). Interestingly, the PhoP/Q system is a positive regulator in thetranscription of pagC of Salmonella (Eriksson-Ygberg et al, 2006), and it ispossible that PagC acts together with PgtE against complement-mediated killingafter the bacteria are released from the M s.

42

9.5.3 Matrix metalloproteinase-9 and gelatin (IV)

The Plg system acts together with the MMPs to degrade the ECM componentsand thus promotes eukaryotic cell migration (Sternlicht and Werb, 2001). M suse the Plg system and secrete proMMPs such as proMMP-9 to enhance theirmigration through ECM (Myöhänen and Vaheri, 2004; Parks et al, 2004). Usinggelatin zymography (Snoek-van Beurden and Von den Hoff, 2005), we foundcleavage of proMMP-9 in assays where serum-free medium from humanprimary M s or purified human proMMP-9 were incubated with the PgtE-expressing derivatives of Salmonella 14028R. Salmonella 14028R and 14028R-1(pMRK3), but not 14028R-1, processed the 92 kDa proMMP-9 into smallerfragments (Fig 1A in IV). Also E. coli XL1 transformed with pMRK3 cleavedproMMP-9 (Fig 1C in IV). The cleavage products of proMMP-9 differedslightly between the PgtE-expressing bacteria and trypsin (Fig 1A and 1C in IV).The approximated sizes of the PgtE-generated cleavage products of proMMP-9were 88 kDa, 82 kDa and 74kDa (Fig 1A and 1C in IV). Trypsin, as mostmammalian proteases, is known to process proMMP-9 first into a proteolyticallyinactive intermediate form that is approximately 88 kDa in size. Activation ofthe intermediate form results from further cleavage into fragments with sizes of80 kDa, 74 kDa and 66 kDa (Woessner and Nagase, 2002; Snoek-van Beurdenand Von den Hoff, 2005).

The possible activation of proMMP-9 by PgtE was measured as cleavage offluorescein conjugated gelatin (DQ-gelatin). When DQ-gelatin was incubated inthe presence of proMMP-9 and S. enterica 14028R, 14028R-1(pMRK3) and E.coli XL1(pMRK3), DQ-gelatin degradation was increased (Fig 1B and 1C in IV)whereas no degradation of DQ-gelatin was detected with 14028R-1 (Fig 1B inIV). This indicates that PgtE converts proMMP-9 into an active enzyme.

An interesting observation was that 14028R and 14028R-1(pMRK3) alsodegraded DQ-gelatin in the absence of proMMP-9 (Fig 1B and 2A in IV). Theresults suggest that PgtE both activates proMMP-9 to MMP-9 as well as directlydegrades gelatin. To assess this in more detail, Salmonella 14028R derivativeswere incubated with DQ-gelatin to study the kinetics of gelatin degradation andwith porcine skin gelatin to study the fragmentation pattern of gelatin by SDSPAGE. Salmonella 14028R and 14028R-1(pMRK3) rapidly degraded both

43

gelatin preparations whereas the pgtE-negative 14028R-1 was inactive in gelatindegradation (Fig 2A and 2B in IV).

As shown above, Salmonella 14028 from mouse M s expresses functionalPgtE. As a first step to assess whether the observed gelatin degradation isbiologically relevant, we probed the ability of Salmonella 14028 to degradegelatin once the cells were released from M s. The Salmonella 14028-infectedM lysate efficiently degraded both DQ-gelatin (Fig 3A in IV) and porcine skingelatin (Fig 3B in IV). The role of PgtE in gelatin degradation was confirmedwith Salmonella 14028 from the SCVs of infected J774A.1 M s. The pgtE-negative Typhimurium strain 14028-1 from the infected lysates and from theSCVs was almost as inefficient in gelatin degradation as was the uninfected Mlysate (Fig 3A and 3B in IV). Taken together, these results show that theSalmonella infection promotes the gelatinase activity in mouse M s by a PgtE-mediated fashion and that Salmonella also expresses a gelatinase on its surfaceafter release from mouse M s. The M s utilize the MMPs and the Plg systemin their migration and we suggest that during systemic salmonellosis PgtE is oneof the important components that induce the M -mediated degradation of ECMcomponents and induces the migration of infected host cells.

PgtE differs both structurally and in catalytic mechanisms from MMPs as wellas from clostridial collagenases, which are the best-characterized groups ofgelatinases/collagenases. The MMPs and the clostridial collagenases are zinc-dependent and secreted metalloproteinases with separate catalytic and substratebinding domains (Woessner and Nagase, 2002; Wilson et al, 2003), whereasPgtE and other omptins are transmembrane aspartate proteases. Despite thestructural differences, PgtE shares some functions with MMP-3 which alsodegrades gelatin, activates proMMP-9 and inactivates 2AP (Ogata et al, 1992;Lijnen et al, 2001; Sternlicht and Werb, 2001). Thus, we suggest that even ifPgtE is not structurally similar to collagenases it functionally mimics them.

44

9.6 Structural basis for the substrate specificity of PgtE (IV)

The omptin family has spread through horizontal gene transfer, and Pla, PgtEand the Erwinia omptin form a closely related subfamily (Kukkonen andKorhonen, 2004). PgtE shares 72% amino acid sequence identity with Pla and48% identity with OmpT of E. coli (Grodberg and Dunn, 1989; Sodeinde andGoguen, 1989). Based on the sequence similarity and crystallized three-dimensional structure of OmpT, the structures of PgtE and Pla were modelledand found to share the overall structure with OmpT (Kramer et al, 2000;Kukkonen et al, 2001; Fig 1). Pla and PgtE are similar in structure and also, to adegree, share putative virulence-associated functions. They activate Plg toplasmin (Sodeinde and Goguen, 1989; I), inactivate 2AP (Kukkonen et al,2001; II), degrade complement components (Sodeinde et al, 1992; III) andadhere onto BM (Lähteenmäki et al, 1998; 2001b; Fig 4 in I).

Using recombinant bacteria, we compared the ability of PgtE and Pla to activateproMMP-9 and to degrade gelatin and found that Pla is inactive in thesefunctions (Fig 5A and 5B in IV). Interestingly, Pla, as well as PgtE carrying thecatalytic site substitution D206A cleaved proMMP-9 into an 88 kDa cleavageproduct, but this cleavage did not lead to MMP-9 activity (Fig 1C, 5A and 5B inIV). This suggests that both PgtE D206A and Pla modify proMMP-9 into anintermediate form, but do not produce enzymatically active MMP-9 (Woessnerand Nagase, 2002).

Transmembrane -barrel proteins such as omptins are very stable and toleratelarge substitutions in their surface-exposed areas (Wimley et al, 2003). OmpTwas turned into a Pla-like protease by changing critical amino acids in thesurface-exposed loops (L1-L5, see Fig 1) between OmpT and Pla (Kukkonen etal, 2001). To characterize the protein regions important for the gelatinolyticactivity in PgtE, we constructed Pla/PgtE hybrid proteins where each surfaceloop of Pla was separately replaced by the corresponding loop of PgtE (Fig 6;Fig 6A and 6B in IV). Analysis of gelatin degradation revealed that thesubstitution of the N-terminal residues 1-40, which contains L1, significantlyincreased gelatin degradation, whereas the constructs harbouring L2, L3 and L4from PgtE were all negative and L5 marginally increased gelatin degradation(Fig 6C in IV). The substitution analysis was continued by using the L1-hybridXL1(pMRK11) as a template. Derivatives XL1(pMRK1132), including L1 and

45

the sequence 161HGVRGIGYS close to L3, XL1(pMRK1141) including L1 andthe non-identical amino acid 212K from L4 and XL1(pMRK1152) including L1and 267TAYF close to L5 (Fig 6; Fig 6B in IV) significantly increased thegelatinase activity when compared to L1 substitute on Pla (Fig 7A in III).Simultaneous substitutions of the amino acids in L3 and L5 with L1 resulted inefficient gelatin degradation, derivative XL1(pMRK115) with L1 and L5 waseven more effective than PgtE (Fig 7A in IV). The importance of L5 in gelatindegradation was further demonstrated with small substitutions independently ofthe amino terminal region. XL1(pMRK1.51) with substitution 259IIDKT andXL1(pMRK1.52) with substitution 267TAYF were more efficient in gelatindegradation than Pla (Fig 7B in IV).

The Pla/PgtE hybrids harbouring the amino acids 161HGVRGIGYS from the L3area of PgtE processed proMMP-9 significantly more efficiently than Pla did.This was seen with XL1(pMRK1.32) (Fig 7B and 7C in IV), XL1(pMRK1132)(Fig 7A and 7C in IV) and XL1(pMRK113) including L1 and amino acids149IYDNGRYI and 161HGVRGIGYS from PgtE (Fig 7A and 7C in IV). Also L5together with L1 was found to be important in proMMP-9 activation andXL1(pMRK115) generated active MMP-9 even more efficiently than PgtE (Fig7A and 7C in IV). These results suggest that the sequence 161HGVRGIGYS fromthe L3 area, together with L1, is important in proMMP-9 activation. The resultsalso show the significance of L5 in turning Pla into a PgtE-like proMMP-9activator.

Figure 6. Schematic model and substitution strategy of Pla. The gray boxes denote -strand

regions and the white boxes L1 through L5 denote the outermost regions of the five surface loops.The individual residues, as well as the larger regions substituted in Pla with the corresponding

PgtE sequence are indicated. The numbers refer to residues in the mature proteins, and thesequence of Pla is given above the sequence of PgtE.

46

In our substitution analysis, all of the non-identical amino acids between Pla andPgtE were covered by large substitutions (Fig 6). The expression levels of thehybrid proteins in the E. coli XL1 host were analyzed by anti-Pla and anti-PgtEantisera and found to be essentially similar (Fig 6D in IV and data not shown).The proteolytic activity of the hybrids was confirmed in Plg activation assay andall hybrids, excluding pMRK1151, were at least as efficient in plasminformation as PgtE (Fig 6E in IV). XL1(pMRK1151) was inefficient in gelatindegradation, proMMP-9 activation and Plg activation, and we think that in thismutant the orientation of Ile259 and Thr263 residues somehow disturbs thecatalytically important amino acids Asp206 and His208 and therefore theproteolytic activity of this mutant is prevented. Interestingly, the L3-associatedsubstitutions that increased proMMP-9 activation in Pla, decreased the Plgactivation to the level seen with PgtE (Fig 6E in IV and data not shown).

In summary, the results show that PgtE of S. enterica and Pla of Y. pestis differin their substrate specifities and that the sequence differences at the surfaceloops are responsible for these differences. Pla has adapted to efficiently activatePlg and PgtE to be a gelatin degrading and proMMP-9 activating enzyme. Placould be converted into a gelatinase by relatively small substitutions, such astwo amino acid change in L5 (pMRK1.51), and substitutions of 27 amino acidsin L1 and L5 (pMRK115) converted Pla into a more efficient gelatinase thanwild-type PgtE. Our results suggest the important role of L1 and L5 forgelatinolytic activity of PgtE and L1, L5 and L3 in proMMP-9 activation.

Different substrate specifities of Pla and PgtE probably reflect the differentpathogenic mechanisms that Salmonella and Y. pestis utilize to cause disease. Y.pestis disseminates mainly extracellularly (Perry and Fetherston, 1997) andactivates circulating Plg efficiently, thus promoting the plasmin-mediateddegradation of ECM components (Lähteenmäki et al, 2005). Plasmin activatesseveral proMMPs, and therefore Pla-generated plasmin is able to promote asufficient amount of collagen and gelatin degradation for infection to proceed.Recently, Lathem et al (2007) showed that Plg activation activity of Pla isessential for Y. pestis in pneumonic plague, which indicates that the Pla-mediated interference with the Plg system promotes the virulence of Y. pestis.Salmonella is a facultatively intracellular pathogen and M s are important hostcells for the migration of Salmonella. The M s secrete proMMP-9 as response

47

to Salmonella LPS (Xie et al, 1994) and our results suggest that PgtE hasadapted to activate proMMP-9 secreted by M s and also directly degradegelatin of ECM. This way Salmonella can compensate its weaker ability tofunction as Plg activator as compared to Y. pestis.

48

10. Conclusions

PgtE of S. enterica is a broad-spectrum outer membrane protease that activatesPlg, inactivates 2AP, mediates serum resistance by degrading the complementcomponents C3b, C4b and C5, degrades gelatin, activates proMMP-9, degradesCAMPs and mediates bacterial adhesion onto BM. Salmonella is a facultativelyintracellular pathogen that has a complex pathogenesis involving extracellularand intracellular phases in the epithelial cells and the phagocytes (Ohl andMiller, 2001; Mastroeni and Sheppard, 2004). Salmonella has many virulencefactors important in bacterial entry and survival in the M during the systemicsalmonellosis (Ohl and Miller, 2001), and we found that also PgtE promotessurvival of Salmonella within murine M s and in mice in vivo. Taken together,PgtE is multifunctional in nature and can have a role at various steps insalmonellosis (summarized in Fig 7).

Figure 7. A model of the functions of PgtE of S. enterica during systemic infection. In thisstudy, PgtE was found to have various functions. It has a role in survival of Salmonella within

M s, mediates bacterial adhesion onto ECM, potentiates migration through tissue barriers andalso is important in bacterial resistance to serum.

49

This work addresses the importance of the coordinated regulation of the surfacestructures of Salmonella in different environments. Lipid A and core regions ofLPS are remodelled in response to CAMPs (Ernst et al, 2001) and regulation ofthe O-antigen length is thought to be important in interaction of Salmonella withcomplement and M s (Murray et al, 2006). One of the major findings in thisPhD work is the dual interaction of omptins with LPS. We found that rough LPSis crucial for PgtE and Pla functions but that they are sterically inhibited by afull-length O-antigen. We propose that the full activity of Pla is a major benefitfor Y. pestis, which is genetically rough. On the other hand, virulent Salmonellaisolates have smooth LPS, and therefore the functions of PgtE had remainedcryptic before this work. A second major finding in this thesis work is that theO-antigen length is reduced during the growth of Typhimurium within M s atthe same time when the expression of PgtE is upregulated, which in a concertlead to high PgtE-mediated proteolysis. Typhimurium multiplying inside M salter the LPS structure in the other molecular aspects as well, and it remains aninteresting task to analyze how, e.g., the PhoP/Q-controlled substitutions ofphosphates in lipid A affect PgtE activity. Taken together, the results in this PhDstudy stress that the induction of the efficient surface proteolysis in Y. pestis andS. enterica is not a matter of the protease gene expression only but is dependenton the over-all alteration of the surface architecture.

Another main finding in this thesis work is the description of how S. enterica svTyphimurium interferes with three important proteolytic systems of human, thePlg and the MMP systems and the complement system. PgtE was also found tomimic mammalian MMPs in many functions. PgtE was known to be a Plgactivator and was here shown to degrade and inactivate the main inhibitor ofplasmin, antiprotease 2AP. This will lead to uncontrolled plasmin activityduring infection, and this study gave evidence that 2AP inactivation isimportant in the M -mediated plasmin formation. Our results show that PgtEhas a positive effect on growth of Salmonella in human serum and at early stagesof systemic infection, supporting the view that PgtE is specifically focused tomediate the migration of Salmonella. The mechanisms of how S. enterica and Y.pestis interact with the Plg and MMP systems are mechanistically different andprobably reflect their very different pathogenic mechanisms. Y. pestis is mainlyan extracellular pathogen, and Pla is needed for its spread from the subdermalskin sites into the lymph nodes (Sodeinde et al, 1992), whereas systemicSalmonella infection is thought to include infective cycles of phagocytes, where

50

Salmonella is transiently released from M s to infect new ones (Sheppard et al,2003; Mastroeni and Sheppard, 2004). The M s utilize the Plg system andMMPs in their migration (Vassalli et al, 1992), and Salmonella LPS induces thesecretion of proMMP-9 from M s (Xie et al, 1994). Our finding that PgtE ofSalmonella activates proMMP-9 and promotes bacterial survival in mice in vivoled to the hypothesis that PgtE could have an effect on M motility and also hasa role in migration of Salmonella within M s. Recently, a mechanism of Mmobility involving the SrfH protein encoded by SPI-2 was proposed (Worley etal, 2006), and it seems likely that multiple mechanisms exist to enhance themigration of M during salmonellosis.

The last part of this thesis work addressed the possible evolution of gelatinaseactivity in the omptin -barrel and identification of active domains and regionsin the PgtE molecule. We found that Pla and PgtE, which are similar in structureand 72% identical in sequences, differ in their substrate specificity in MMP-associated functions. It has been proposed that Pla has evolved from PgtE(Kukkonen and Korhonen, 2004), and earlier studies in this laboratory indicatedthat sequences near or at the surface-exposed loops are important for targetprotein specificity in omptin proteolysis (Kukkonen et al, 2001). In this study,conversion of Pla into a PgtE-like gelatinase and proMMP-9 activator wasachieved by substitution of amino acids in L1, L3 and L5. This exemplifies howdiffering functions can be created in a horizontally spread protein fold, i.e. theomptin -barrel.

51

11. ReferencesAchtman, M., Zurth, K., Morelli, G., Torrea, G., Guiyole, A. and Carniel, E. (1999) Yersinia pestis, thecause of plague, is a recently emerged clone of Yersinia pseudotuberculosis. Proc. Natl. Acad. Sci. USA 96,14043-14048.

Aderem, A. and Underhill, D.M (1999) Mechanisms of phagocytosis in macrophages. Annu. Rev. Immunol.17, 593-623.

Agüero, M.E., Aron, L., DeLuca, A.G., Timmis, K.N. and Cabello, F.C. (1984) A plasmid-encoded outermembrane protein, TraT, enhances resistance of Escherichia coli to phagocytosis. Infect. Immun. 46, 740-746.

Aimes, R.T and Quigley, J.P. (1995) Matrix metalloproteinase-2 is an interstitial collagenase. J. Biol. Chem.270, 5872-5876.

Allan, J.A., Docherty, J.P., Barker, P.J., Huskisson, N.S., Reynolds, J.J. and Murphy, G. (1995) Bindingof gelatinases A and B to type-I collagen and other matrix components. Biochem. J. 309, 299-306.

Alpuche-Aranda, C.M., Racoosin, E.L., Swanson, J.A. and Miller, S.I. (1994) Salmonella stimulatemacrophage macropinocytosis and persist within spacious phagosomes. J. Exp. Med. 179, 601-608.

Alpuche-Aranda, C.M., Swanson, J.A., Loomis, W.P. and Miller, S.I. (1992) Salmonella typhimuriumactivates virulence gene transcription within acidified macrophage phagosomes. Proc. Natl. Acad. Sci. USA 89,10079-10083.

Bader, M.W., Sanowar, S., Daley, M.E., Shcneider, A.R., Cho, U., Xu, W., Klevit, R.E., Moual, H.L. andMiller, S.I. (2005) Recognition of antimicrobial peptides by a bacterial sensor kinase. Cell 122, 461-472.

Bader, M.W., Navarre, W.W., Shiau, W., Nikaldo, H., Frye, J.G., McClelland, M., Fang, F.C. andMiller, S.I. (2003) Regulation of Salmonella typhimurium virulence gene expression by cationic antimicrobialpeptides. Mol. Microbiol. 50, 219-230.

Baker, S.J., Gunn, J.S. and Morona, R. (1999) The Salmonella typhi melittin resistance gene pqaB affectsintracellular growth in PMA-differentiated U937 cells, polymyxin B resistance and lipopolysaccharide.Microbiology 145, 367-378.

Behlau, I. and Miller, S.I. (1993) A PhoP-repressed gene promotes Salmonella typhimurium invasion ofepithelial cells. J. Bacteriol. 175, 4475-4484.

Benedek, O., Nagy, G. and Emödy, L. (2004) Intracellular signalling and cytoskeletal rearrangementinvolved in Yersinia pestis plasminogen activator (Pla) mediated HeLa cell invasion. Microb. Pathog. 37, 47-54.

Bengoechea, J.A., Brandenburg, K., Seydel, U., Díaz, R. and Moriyón I. (1998) Yersiniapseudotuberculosis and Yersinia pestis show increased outer membrane permeability to hydrophobic agentswhich correlates with lipopolysaccharide acyl-chain fluidity. Microbiology 144, 1517-1526.

Beuzón, C.R., Méresse, S., Unsworth, K.E., Ruíz-Albert, J., Garvis, S., Waterman, S.R., Ryder, T.A.,Boucrot, E. and Holden, D.W. (2000) Salmonella maintains the integrity of its intracellular vacuole throughthe action of SifA. EMBO J. 19, 3235-3249.

Billinghurst, R.C., Dahlberg, L., Ionescu, M., Reiner, A., Bourne, R., Rorabeck, C., Mitchell, P.,Hambor, J., Diekmann, O., Tschesche, H., Chen, J., Van Wart, H. and Poole, A.R. (1997) Enhancedcleavage of type II collagen by collagenases in osteoarthritic articular cartilage. J. Clin. Invest. 99, 1534-1545.

Bonasio, R. and von Andrian, U.H. (2006) Generation, migration and function of circulating dendritic cells.Curr. Opin. Immunol. 18, 503-511.

Bonina, L., Costa, G.B. and Mastroeni, P. (1998) Comparative effect of gentamicin and pefloxacin treatmenton the late stages of mouse typhoid. New Microbiol. 21, 9-14

52

Brumell, J.H. and Grinstein, S. (2004) Salmonella redirects phagosomal maturation. Curr. Opin. Microbiol.7, 78-84.

Bruneteau, M. and Minka, S. (2003) Lipopolysaccharides of bacterial pathogens from the genus Yersinia: amini-review. Biochimie 85, 145-152.

Buchmeier, N.A. and Heffron, F. (1991) Inhibition of macrophage phagolysosome-lysosome fusion bySalmonella typhimurium. Infect. Immun. 59, 2232-2238.

Budd, A., Blandin, S., Levashina, E.A. and Gibson, T.J. (2004) Bacterial 2-macroglobulins: colonizationfactors acquired by horizontal gene transfer from the metazoan genome? Genome Biol. 5, R38.

Carlsson, J., Herrmann, B.F., Höfling, J.F. and Sundqvist, G.K. (1984) Degradation of the humanproteinase inhibitors alpha-1-antitrypsin and alpha-2-macroglobulin by Bacteroides gingivalis. Infect. Immun.43, 644-648.

Carmeliet, P., Moons, L., Lijnen, R., Baes, M., Lemaître, V., Tipping, P., Drew, A., Eeckhout, Y.,Shapiro, S., Lupu, F. and Collen, D. (1997) Urokinase-generated plasmin activates matrix metalloproteinasesduring aneurysm formation. Nat. Genet. 17, 439-444.

Castaneda, F.E., Walia, B., Vijay-Kumar, M., Patel, N.R., Roser, S., Kolachala, V.L., Rojas, M., Wang,L., Oprea, G., Garg, P., Gewirtz, A.T., Roman, J., Merlin, D. and Sitaraman, S.V. (2005) Targeteddeletion of metalloproteinase 9 attenuates experimental colitis in mice: central role of epithelial-derived MMP.Gastroenterology 129, 1991-2008.

Chan, K., Baker, S., Kim, C.C., Detweiler, C.S., Dougan, G. and Falkow, S. (2003) Genomic comparisonof Salmonella enterica serovars and Salmonella bongori by use of an S. enterica serovar Typhimurium DNAmicroarray. J. Bacteriol. 185, 553-563.

Chen, L.M., Kaniga, K. and Galan, J.E. (1996) Salmonella spp. are cytotoxic for cultured macrophages.Mol. Microbiol. 21, 1101-1115.

Chen, H. and Schifferli, M. (2003) Construction, characterization, and immunogenicity of an attenuatedSalmonella enterica serovar Typhimurium pgtE vaccine expressing fimbriae with integrated viral epitopesfrom the spiC promoter. Infect. Immun. 71, 4664-4673.

Cirillo, D.M., Heffernan, E.J., Wu, L., Harwood, J., Fierer, J., Guiney, D.G. (1996) Identification of adomain in Rck, a product of the Salmonella typhimurium virulence plasmid, required for both serum resistanceand cell invasion. Infect. Immun. 64, 2019-2023.

Collen, D. and Lijnen, H.R. (1991) Basic and clinical aspects of fibrinolysis and thrombolysis. Blood 78,3114-3124.

Collinson, S.K., Doig, P.C., Doran, J.L., Clouthier, S., Trust, T.J. and Kay, W.W. (1993) Thin,aggregative fimbriae mediate binding of Salmonella enteritidis to fibronectin. J. Bacteriol. 175, 12-18.

Coughlin, P.B. (2005) Antiplasmin: the forgotten serpin? FEBS J. 272, 4852-4857.

Coulton, J.W., Mason, P., Cameron, D.R., Carmel, G., Jean, R. and Rode, H.N. (1986) Protein fusions ofbeta-galactosidase to the ferrichrome-iron receptor of Escherichia coli K-12. J. Bacteriol. 165, 181-92.

Cowan, C., Jones, H.A., Kaya, Y.H., Perry, R.D. and Straley, S.C. (2000) Invasion of epithelial cells byYersinia pestis: evidence for a Y. pestis-specific invasin. Infect. Immun. 68, 4523-4530.

Crago, A.M. and Koronakis, V. (1999) Binding of extracellular matrix laminin to Escherichia coli expressingthe Salmonella outer membrane proteins Rck and PagC. FEMS Microbiol. Lett. 176, 495-501.

Crosa, J.H., Brenner, D.J., Ewing, W.H. and Falkow, S. (1973) Molecular relationships among theSalmonellae. J. Bacteriol. 115, 307-315.

53

Dekker, N., Cox, R.C., Kramer, A. and Egmond, M.R. (2001) Substrate specificity of the integralmembrane protease OmpT determined by spatially addressed peptide libraries. Biochemistry 40, 1694-1701.

Detweiler, C.S., Cunahan, D.B. and Falkow, S. (2001) Host microarray analysis reveals a role for theSalmonella response regulator PhoP in human macrophage cell death. Proc. Natl. Acad. Sci. USA 98, 5850-5855.

Dorsey, C.W., Laarakker, M.C., Humphries, A.D., Weening, E.H. and Bäumler, A.J. (2005) Salmonellaenterica serotype Typhimurium MisL is an intestinal colonization factor that binds fibronectin. Mol. Microbiol.57, 196-211.

Drecktah, D., Knodler, L.A., Ireland, R. and Steele-Mortimer, O. (2006) The mechanism of Salmonellaentry determines the vacuolar environment and intracellular gene expression. Traffic 7, 39-51.

Elkington, P.T.G., O’Kane C.M. and Friedland J.S. (2005) The paradox of matrix metalloproteinases ininfectious disease. Clin. Exp. Immunol. 142, 12-20.

Eriksson Ygberg, S., Clements, M.O., Rytkönen, A., Thompson, A., Holden, D.W., Hinton, J.C.D. andRhen, M. (2006) Polynucleotide phosphorylase negatively controls spv virulence gene expression inSalmonella enterica. Infect. Immun. 74, 1243-1254.

Eriksson, S., Lucchini, S., Thompson, A., Rhen, M. and Hinton, J.C.D. (2003) Unravelling the biology ofmacrophage infection bygene expression profiling of intracellular Salmonella enterica. Mol. Microbiol. 47, 103-112.

Ernst, R.K., Guina, T. and Miller, S.I. (2001) Salmonella typhimurium outer membrane remodeling: role inresistance to host innate immunity. Microbes Infect. 3, 1327-1334

Ferguson, A.D., Welte, W., Hofmann, E., Lindner, B., Holst, O., Coulton, J.W. and Diederichs, K. (2000)A conserved structural motif for lipopolysaccharide recognition by procaryotic and eucaryotic proteins.Structure 8, 585-592.

Ferguson, A.D., Hofmann, E., Coulton, J.W., Diederichs, K. and Welte, W. (1998) Siderophore-mediatediron transport: crystal structure of FhuA with bound lipopolysaccharide. Science 282, 2215-2220.

Fields, P.I., Groisman, E.A. and Heffron, F. (1989) A Salmonella locus that controls resistance tomicrobicidal proteins from phagocytic cells. Science 243, 1059-62.

Fields, P.I., Swanson, R.V., Haidaris, C.G. and Heffron, F. (1986) Mutants of Salmonella typhimurium thatcannot survive within the macrophage are avirulent. Proc. Natl. Acad. Sci. USA 83, 5189-5193.

Foster, J.W. and Hall, H.K. (1990) Adaptive acidification tolerance response of Salmonella typhimurium.J.Bacteriol. 172, 771-778.

Galyov, E.E., Wood, M.W., Rosqvist, R., Mullan, P.B., Watson, P.R., Hedges, S. and Wallis, T.S. (1997)A secreted effector protein of Salmonella dublin is translocated into eukaryotic cells and mediatesinflammation and fluid secretion in infected ileal mucosa. Mol. Microbiol. 25, 903-912.

Garcia Véscovi, E., Soncini, F.C. and Groisman, E.A. (1996) Mg2+ as an extracellular signal: environmentalregulation of Salmonella virulence. Cell 84, 165-174.

Giannella, R.A. (1979) Importance of the intestinal inflammatory reaction in salmonella-mediated intestinalsecretion. Infect. Immun. 23, 140-145.

Grenier, D. (1996) Degradation of host protease inhibitors and activation of plasminogen by proteolyticenzymes from Porphyromonas gingivalis and Treponema denticola. Microbiology 142, 955-961.

Grodberg, J. and Dunn, J.J. (1989) Comparison of Escherichia coli K-12 outer membrane protease OmpTand Salmonella typhimurium E protein. J. Bacteriol. 171, 2903-2905.

54

Groisman, E.A. (2001) The pleiotropic two-component regulatory system PhoP-PhoQ. J. Bacteriol. 183,1835-1842.

Groisman, E.A. and Ochman, H. (1997) How Salmonella became a pathogen. Trends Microbiol. 5, 343-349.

Groisman, E.A., Parra-Lopez, C., Salcedo, M., Lipps, C.J. and Heffron, F. (1992) Resistance to hostantimicrobial peptides is necessary for Salmonella virulence. Proc. Natl. Acad. Sci. USA 89, 11939-11943.

Grossman, N., Schmetz, M.A., Foulds, J., Klima, E.N., Jiminez, V., Leive, L.L., Joiner, K.A. (1987)Lipopolysaccharide size and distribution determine serum resistance in Salmonella montevideo. J. Bacteriol.169, 856-863.

Guex, N. and Peittsch, M.C. (1997) SWISS-MODEL and the Swiss-PdbViewer: an environment forcomparative protein modeling. Electrophoresis 18, 2714-2723.

Guina, T., Yi, E.C., Wang, H., Hackett, M. and Miller, S.I. (2000) A PhoP- regulated outer membraneprotease of Salmonella enterica serovar Typhimurium promotes resistance to alpha- helical antimicrobialpeptides. J. Bacteriol. 182, 4077- 4086.

Guiney, D.G. (2005) The role of host cell death in Salmonella infections. Curr. Top. Microbiol. Immunol. 289,131-150.

Gunn, J.S. (2001) Bacterial modification of LPS and resistance to antimicrobial peptides. J. Endotoxin Res. 7,57-62.

Gunn, J.S., Lim, K.B., Krueger, J., Kim, K., Hackett, M. and Miller, S.I. (1998) PmrA-PmrB-regulatedgenes necessary for 4-aminoarabinose lipid A modification and polymyxin resistance. Mol. Microbiol. 27,1171-1182.

Guo, L., Lim, K.B., Poduje, C.M., Daniel, M., Gunn, J.S., Hackett, M. and Miller, S.I. (1998) Lipid Aacylation and bacterial resistance against vertebrate antimicrobial peptides. Cell 95, 189-198.

Guo, L., Lim, K.B., Gunn, J.S., Bainbridge, B., Darveau, R.P., Hackett, M. and Miller, S.I. (1997)Regulation of lipid A modifications by Salmonella typhimurium virulence genes phoP-phoQ. Science 276,250-253.

Handley, S.A. and Miller, V.L. (2007) General and specific host responses to bacterial infection in Peyer’spatches: a role for stromelysin-1 (matrix metalloproteinase-3) during Salmonella enterica infection. Mol.Microbiol. 64, 94-110.

Harper, E., Bloch, K.J. and Gross, J. (1971) The zymogen of tadpole collagenase. Biochemistry 10, 3035-3041.

Harrington, D.J. (1996) Bacterial collagenases and collagen-degrading enzymes and their potential role inhuman disease. Infect. Immun. 64, 1885-1891.

Hashim, S., Mukherjee, K., Raje, M., Basu, S.K. and Mukhopadhyay, A. Live Salmonella modulateexpression of Rab proteins to persist in a specialized compartment and escape transport to lysosomes. J. Biol.Chem. 275, 16281-16288.

Heffernan, E.J., Reed, S., Hackett, J., Fierer, J., Roudier, C. and Guiney, D. (1992a) Mechanism ofresistance to complement-mediated killing of bacteria encoded by the Salmonella typhimurium virulenceplasmid gene rck. J. Clin. Invest. 90, 953-964.

Heffernan, E.J., Harwood, J., Fierer, J. and Guiney, D. (1992b) The Salmonella typhimurium virulenceplasmid complement resistance gene rck is homologous to a family of virulence-related outer membraneprotein genes, including pagC and ail. J. Bacteriol. 174, 84-91.

Heinrichs, D.E., Yethon, J.A. and Whitfiels, C. (1998) Molecular basis for structural diversity in the coreregions of the lipopolysaccharides of Escherichia coli and Salmonella enterica. Mol. Microbiol. 30, 221-232.

55

Hohmann, E.L. (2001) Montyphoidal salmonellosis. Food Safety 32, 263-269.

Holden, D.W. (2002) Trafficking of the Salmonella vacuole in macrophages. Traffic 3, 161-169.

Holst, O. (2000) Deacylation of lipopolysaccharides and isolation of oligosaccharide phosphates. MethodsMol. Biol. 145, 345-353.

Hueffer, K. and Galán, J.E. (2004) Salmonella-induced macrophage death: multiple mechanisms, differentoutcomes. Cell. Microbiol. 6, 1019-1025.

Johansson, N., Ahonen, M. and Kähäri, V.M. (2000) Matrix metalloproteinases in tumor invasion. Cell.Mol. Life. Sci. 57, 5-15.

Joiner, K.A., Hammer, C.H., Brown, E.J., Cole, R.J. and Frank, M.M. (1982) Studies on the mechanismof bacterial resistance to complement-mediated killing. I. Terminal complement component components aredeposited and released from Salmonella minnesota S218 without causing bacterial death. J. Exp. Med. 155,797-808.

Kingsley, R.A., Humphries, A.D., Weening, E.H., de Zoete, M.R., Winter, S., Papaconstantinopoulou, A.,Dougan, G. andBäumler, A.J. (2003) Molecular and phenotypic analysis of the CS54 island of Salmonella enterica serotypeTyphimurium: identification of intestinal colonization and persistence determinants. Infect. Immun. 71, 629-40.

Kingsley, R.A., Santos, R.L., Keestra, A.M., Adams, L.G. and Bäumler, A.J. (2002) Salmonella entericaserotype Typhimurium ShdA is an outer membrane fibronectin-binding protein that is expressed in theintestine. Mol. Microbiol. 43, 895-905.

Knodler, L.A. and Finlay, B.B. (2001) Salmonella and apoptosis: to live or let die? Microbes infect. 3, 1321-1326.

Kramer, R.A., Brandenburg, K., Vandeputte-Rutten, L., Werkhoven, M., Gros, P., Dekker, N. andEgmond, M.R. (2002) Lipopolysaccharide regions involved in the activation of Escherichia coli outermembrane protease OmpT. Eur.J.Biochem. 269, 1746-1752.

Kramer, R.A., Zandwijken, D., Egmond, M.E. and Dekker, N. (2000) In vitro folding, purification andcharacterization of Escherichia coli outer membrane protease OmpT. FEBS Lett. 267, 885-893.

Kukkonen, M. and Korhonen, T.K. (2004) The omptin family of enterobacterial surface proteases/adhesins:from housekeeping in Escherichia coli to systemic spread of Yersinia pestis. Int. J. Med. Microbiol. 294, 7-14.

Kukkonen, M., Lähteenmäki, K., Suomalainen, M., Kalkkinen, N., Emödy, L., Lång, H. and Korhonen,T.K. (2001) Protein regions important for plasmingen activation and inactivation 2-antiplasmin in the surfaceprotease Pla of Yersinia pestis. Mol. Microbiol. 40, 1097-1111.

Kukkonen, M., Saarela, S., Lähteenmäki, K., Hynönen, U., Westerlund-Wikström, B., Rhen, M. andKorhonen, T.K. (1998) Identification of two laminin-biding fimbriae, the type 1 fimbria of Salmonellaenterica serovar Typhimurium and the G fimbria of Escherichia coli, as plasminogen receptors. Infect. Immun.66, 4965-4970.

Kukkonen, M., Raunio, T., Virkola, R., Lähteenmäki, K., Mäkelä, P.H., Klemm, P., Clegg, S. andKorhonen, T.K. (1993) Basement membrane carbohydrate as a target for bacterial adhesion: binding of type Ifimbriae of Salmonella enterica and Escherichia coli to laminin. Mol. Microbiol. 7, 229-237.

Lathem, W.W., Price, P.A., Miller, V.L. and Goldman, W.E. (2007) A plasminogen-activating proteasespecifically controls the development of primary pneumonic plague. Science 315, 509-513.

Lindgren, S.W., Stojiljkovic, I. and Heffron, F. (1996) Macrophage killing is an essential virulencemechanism of Salmonella typhimurium. Proc. Natl. Acad. Sci. USA 93, 4197-201

Linehan, S.A. and Holden, D.W. (2003) The interplay between Salmonella typhimurium and its macrophagehost-what can it teach us about innate immunity. Immunol. Lett. 85, 183-192.

56

Lijnen, H.R., Van Hoef, B. and Collen, D. (2001) Inactivation of the serpin 2-antiplasmin by stromelysin-1.Biochim. Biophys. Acta. 1547, 206-213.

Liotta, L.A., Goldfarb, R.H., Brundage, R., Siegal, G.P., Terranova, V. and Garbisa, S. (1981) Effect ofplasminogen activator (urokinase), plasmin, and thrombin on glycoprotein and collagenous components ofbasement membrane. Cancer Res. 41, 4629-4636.

Looney, R.J. and Steiqbiqel, R.T. (1986) Role of the Vi antigen of Salmonella typhi in resistance to hostdefense in vitro. J. Lab. Clin. Med. 108, 506-516.

Lähteenmäki, K., Edelman, S. and Korhonen, T.K. (2005) Bacterial metastasis: the host plasminogensystem in bacterial invasion. Trends Microbiol. 13, 79-85.

Lähteenmäki, K., Kuusela, P. and Korhonen, T.K. (2001a) Bacterial plasminogen activators and receptors.FEMS Microbiol. Rev. 25, 531-552.

Lähteenmäki, K., Kukkonen, M. and Korhonen, T.K. (2001b) The Pla surface protease/adhesin of Yersiniapestis mediates bacterial invasion into human endothelial cells. FEBS Lett. 504, 69-72.

Lähteenmäki, K., Kuusela, P. and Korhonen, T.K. (2000) Plasminogen activation in degradation andpenetration of extracellular matrices and basement membranes by invasive bacteria. Methods 21, 125-132.

Lähteenmäki, K., Virkola, R., Sarén, A., Emödy, L. and Korhonen, T.K. (1998) Expression of plasmingenactivator Pla of Yersinia pestis enhances bacterial attachment to the mammalian extracellular matrix. Infect.Immun. 66, 5755-5762.

Lähteenmäki, K., Virkola, R., Pouttu, R., Kuusela, P., Kukkonen, M. and Korhonen, T.K. (1995)Bacterial plasminogen receptors: in-vitro evidence for a role in degradation of the mammalian extracellularmatrix. Infect. Immun. 63, 3659-3664.

Mastroeni, P. and Sheppard, M. (2004) Salmonella infections in the mouse model: host resistance factorsand in vivo dynamics of bacterial spread and distribution in the tissues. Microbes Infect. 6, 398-405.

McCawley, L.J. and Matrisian, L.M. (2001) Matrix metalloproteinases: they’re not just for matrix anymore.Curr. Opin. Cell Biol. 13, 534-540.

McClelland, M., Sanderson, K.E., Spleth, J., Clifton, S.W., Latreille, P., Courtney, L., Porwollik, S., Ali,J., Dante, M., Du, F., Hou, S., Layman, D., Leonard, S., Nguyen, C., Scott, K., Holmes, A., Grewal, N.,Mulvaney, E., Ryan, E., Sun, H., Florea, L., Miller, W., Stoneking, T., Nhan, M., Waterston, R. andWilson, R.K. (2001) Complete genome sequence of Salmonella enterica serovar Typhimurium LT2. Nature413, 852-856.

McGovern, V.J. and Slavutin, L.J. (1979) Pathology of Salmonella colitis. Am. J. Surg. Pathol. 3, 483-490.

Méresse, S., Steele-mortimer, O., Moreno, E., Desjardins, M., Finlay, B. and Gorvel, J. (1999) Controllingthe maturation of pathogen-containing vacuoles: a matter of life and death. Nat. Cell Biol. 1, E183-E188.

Mignatti, P. and Rifkin, D.B. (1993) Biology and biochemistry of proteinases in tumor invasion. Physiol.Rev. 73, 161-95.

Miller, S.I., Kukral, A.M. and Mekalanos, J.J. (1989) A two component regulatory system (phoP phoQ)controls Salmonella typhimurium virulence. Proc. Natl. Acad. Sci. USA 86, 5054-5058.

Mohan, R., Rinehart, W.B., Bargagna-Mohan, P. and Fini, M.E. (1998) Gelatinase B/lacZ transgenic mice,a model for mapping gelatinase B expression during developmental and injury-related tissue remodeling. J.Biol. Chem. 273, 25903-25914.

Moll, A., Manning, P.A. and Timmis, K.N. (1980) Plasmid-determined resistance to serum bactericidalactivity: a major outer membrane protein, the traT gene product, is responsible for plasmid-specified serumresistance in Escherichia coli. Infect. Immun. 2, 359-67.

57

Morgan, E. (2007) Salmonella pathogenicity islands. In: Rhen, M., Maskell, D., Mastroeni, P. and Threlfall, J.(eds.): Salmonella molecular biology and pathogenesis. Horizon Bioscience, UK. 67-88.

Munaut, C., Salonurmi, T., Kontusaari, S., Reponen, P. and Morita, T. (1999) Murine matrixmetalloproteinase 9 gene. 5’-upstream region contains cis-acting elements for expression in osteoclasts andmigrating keratinocytes in transgenic mice. J. Biol. Chem. 274, 5588-5596.

Murray, G.L., Attridge, S.R. and Morona, R. (2006) Altering the length of the lipopolysaccharide O antigenhas an impact on the interaction of Salmonella enterica serovar Typhimurium with macrophages andcomplement. J. Bacteriol. 188, 2735-2739.

Murray, G.L., Attridge, S.R. and Morona, R. (2005) Inducible serum resistance in Salmonella typhimuriumis dependent on wzzfepE-regulated very long O antigen chains. Microbes Infect. 7, 1296-1304.

Murray, G.L., Attridge, S.R. and Morona, R. (2003) Regulation of Salmonella typhimuriumlipopolysaccharide O antigen chain length is required for virulence; identification of FepE as a second Wzz.Mol. Microbiol. 47, 1395-1406.

Myöhänen, H. and Vaheri, A. (2004) Regulation and interactions in the activation of cell-associatedplasminogen. Cell. Mol. Life. Sci. 61, 2840-2858.

Navarre, W.W., Halsey, T.A., Walthers, D., Frye, J., McClelland, M., Potter, J.L., Kenney, L.J., Gunn,J.S., Fang, F.C. and Libby, S.J. (2005) Co-regulation of Salmonella enterica genes required for virulence andresistance to antimicrobial peptides by SlyA and PhoP/PhoQ. Mol. Microbiol. 56, 492-508.

Nishio, M., Okada, N., Miki, T., Haneda, T. and Danbara, H. (2005) Identification of the outer-membraneprotein PagC required for the serum resistance phenotype in Salmonella enterica serovar Choleraesuis.Microbiology 151, 863-873.

Ochman, H. and Wilson, A.C. (1987) Evolution in bacteria: evidence for a universal substitution rate incellular genomes. J. Mol. Evol. 26, 74-86.

Ogata, Y., Enghild, J.J. and Nagase, H. (1992) Matrix metalloproteinase 3 (stromelysin) activates theprecursor for the human matrix metalloproteinase 9. J. Biol. Chem. 267, 3581-3584.

Oggioni, M.R., Memmi, G., Maggi, T., Chiavolini, D., Iannelli, F. and Pozzi, G. (2003) Pneumococcal zincmetalloproteinase ZmpC cleaves human matrix metalloproteinase 9 and is a virulence factor in experimentalpneumonia. Mol Microbiol. 49, 795-805.

Ohl, M.E. and Miller, S.I. (2001) Salmonella: A model for bacterial pathogenesis. Annu. Rev. Med. 52, 259-274.

Okamoto, T., Akaike, T., Suga, M., Tanase, S., Horie, H., Miyajima, S., Ando, M., Ichinose, Y. andMaeda, H. (1997) Activation of human matrix metalloproteinases by various bacterial proteinases. J. Biol.Chem. 272, 6059-6066.

Opdenakker, G., Van den Steen, P.E., Dubois, B., Nelissen, I., Van Coillie, E., Masure, S., Proost, P. andVan Damme, J. (2001) Gelatinase B functions as regulator and effector in leukocyte biology. J. Leukoc. Biol.69, 851-859.

Pang, T., Bhutta, Z.A., Finlay, B.B. and Altwegg, M. (1995) Typhoid fever and other salmonellosis: acontinuing challenge. Trends Microbiol. 3, 253-255.

Parkhill, J., Wren, B.W., Thomson, N.R., Titball, R.W., Holden, M.T., Prentice, M.B., Sebaihia, M.,James, K.D., Churcher, C., Mungall, K.L., Baker, S., Basham, D., Bentley, S.D., Brooks, K., Cerdeno-Tarraga, A.M., Chillingworth, T., Cronin, A., Davies, R.M., Davis, P., Dougan, G., Feltwell, T., Hamlin,N., Holroyd, S., Jagels, K., Karlyshev, A.V., Leather, S., Moule, S., Oyston, P.C., Quail, M., Rutherford,K., Simmonds, M., Skelton, J., Stevens, K., Whitehead, S. and Barrell, B.G. (2001) Genome sequence ofYersinia pestis, the causative agent of plague. Nature 413, 523-527.

58

Parks, W.C., Wilson, C.L. and López-Boado, Y.S. (2004) Matrix metalloproteinases as modulators ofinflammation and innate immunity. Nat. Rev. Immunol. 4, 617-629.

Pattamapun, K., Tiranathanagul, S., Yongchaitrakul, T., Kuwatanasuchat, J. and Pavasant, P. (2003)Activation of MMP-2 by Porphyromonas gingivalis in human periodontal ligament cells. J. Periodontal Res.38, 115-121.

Pere, A., Selander, R.K. and Korhonen, T.K. (1988) Characterization of P fimbriae on O1, O7, O75, rough,and nontypable strains of Escherichia coli. Infect. Immun. 56, 1288-1294.

Perry, R.D. and Fetherston, J.D. (1997) Yersinia pestis-etiologic agent of plague. Clin. Microbiol. Rev. 10,35-66.

Perry, R.D. and Brubaker, R.R. (1983) Vwa+ phenotype of Yersinia enterocolitica. Infect. Immun. 40, 166-171.

Pietilä, T.E., Veckman, V., Kyllönen, P., Lähteenmäki, K., Korhonen, T.K. and Julkunen, I. (2005)Activation, cytokine production, and intracellular survival of bacteria in Salmonella-infected human monocyte-derived macrophages and dendritic cells. J. Leukoc. Biol. 78, 909-920.

Pizarro-Cerda, J. and Cossart, P. (2006) Bacterial adhesion and entry into host cells. Cell 124, 715-727.

Popoff, M.Y., Bockemühl, J. and Gheesling, L.L. (2004) Supplement 2002 (no. 46) to the Kauffmann-Whitescheme. Res. Microbiol. 155, 568-570.

Potempa, J., Watorek, W. and Travis, J. (1986) The inactivation of human plasma 1-proteinase inhibitor byproteinases from Staphylococcus aureus. J. Biol. Chem. 261, 14330-14334.

Potempa, J., Banbula, A. and Travis, J. (2000) Role of bacterial proteinases in matrix destruction andmodulation of host responses. Periodontol. 2000 24, 153-192.

Pramoonjago, P., Kaneko, M., Kinoshita, T., Ohtsubo, E., Takeda, J., Hong, K., Inagi, R. and Inoue, K.(1992) Role of Trat protein, an anticomplementary protein produced in Escherichia coli by R100 factor, inserum resistance. J. Immunol. 148, 827-836.

Rabijns, A., De Bondt, H.L. and De Ranter, C (1997) Three-dimensional structure of staphylokinase, aplasminogen activator with therapeutic potential. Nature Struct. Biol. 4, 357-360.

Raetz, C.R. and Whitfield, C. (2002) Lipopolysaccharide endotoxins. Annu. Rev. Biochem. 71, 635-700.

Rautemaa, R. and Meri, S. (1999) Complement-resistance mechanisms of bacteria. Microbes Infect. 1, 785-794.

Ravetch, J.V. (1997) Fc receptors. Curr. Opin. Immunol. 9, 121-125.

Reeves, P.R., Hobbs, M., Valvano, M.A., Skurnik, M., Whitfield, C., Coplin, D., Kido, N., Klena, J.,Maskell, D., Raetz, C.R.H. and Rick, P.D. (1996) Bacterial polysaccharide synthesis and gene nomenclature.Trends Microbiol. 4, 495-503.

Rhen, M. and Sukupolvi, S. (1988) The role of the traT gene of the Salmonella typhimurium virulenceplasmid for serum resistance and growth within liver macrophages. Microb. Pathog. 5, 275-285.

Richter-Dahlfors, A., Buchan, A.M.J. and Finlay, B.B. (1997) Murine salmonellosis studied by confocalmicroscopy: Salmonella typhimurium resides intracellularly inside macrophages and exerts a cytotoxic effecton phagocytes in vivo. J. Exp. Med. 186, 569-580.

Roland, K.L., Martin, L.E., Esther, C.R. and Spitznagel, J.K. (1993) Spontaneous pmrA mutants ofSalmonella typhimurium LT2 define a new two-component regulatory system with a possible role in virulence.J. Bacteriol. 175, 4154-4164.

59

Rooijakkers, S.H.M., van Wamel, W.J.B., Ruyken, M., van Kessel, K.P.M. and van Stripj, J.A.G. (2005)Anti-opsonic properties of staphylokinase. Microbes Infect. 7, 476-484.

Rychlik, I., Gregorova, D. and Hradecka, H. (2006) Distribution and function of plasmids in Salmonellaenterica. Vet. Microbiol. 112, 1-10.

Saksela, O. (1985) Plasminogen activation and regulation of pericellular proteolysis. Biochim. Biophys. Acta.823, 35-65.

Salonen, E., Zitting, A. and Vaheri, A. (1984) Laminin interacts with plasminogen and its tissue-typeactivator. FEBS Lett. 172, 29-32.

Schmidt, G. and Lüderitz, O. (1969) Investigations on the classification of Salmonella R forms. II.Classification of S. minnesota mutants by phage typing. Zentralbl. Bakteriol. 210, 381-387.

Schwan, W.R., Huang, X.Z., Hu, L. and Kopecko, D.J. (2000) Differential bacterial survival, replication,and apoptosis-inducing ability of Salmonella serovars within human and murine macrophages. Infect. Immun.68, 1005-1013.

Sebbane, F., Jarrett, C.O., Gardner, D., Long, D. and Hinnebusch, B. J. (2006a) Role of the Yersinia pestisplasminogen activator in the incidence of distinct septicemic and bubonic forms of flea-borne plague. Proc.Natl. Acad. Sci. USA 103, 5526-5530.

Sebbane, F., Lemaître, N., Sturdevant, D.E., Rebeil, R., Virtaneva, K., Porcella, S.F. and Hinnebusch, B.J.(2006b) Adaptive response of Yersinia pestis to extracellular effectors of innate immunity during bubonicplague. Proc. Natl. Acad. Sci. USA 103, 11766–11771

Selander, R.K., Korhonen, T.K., Väisänen-Rhen, V., Williams, P.H., Pattison, P.E. and Caugant D.A.(1986) Genetic relationships and clonal structure of strains of Escherichia coli causing neonatal septicemia andmeningitis. Infect. Immun. 52, 213-222.

Seya, T., Nagasawa, S., Matsukura, M., Hasegawa, H. and Atkinson, J.P. (1985) Generation of C3d,g andC3d by urokinase-treated plasma in association with fibrinolysis. Complement 2, 165-174.

Sjöbring, U., Pohl, G. and Olsen A. (1994) Plasminogen, absorbed by Escherichia coli expressing curli or bySalmonella enteritidis expressing thin aggregative fimbriae, can be activated by simultaneously capturedtissue-type plasminogen activator (t-PA). Mol. Microbiol. 14, 443-452.

Skurnik, M., Peippo, A. and Ervelä, E. (2000) Characterization of the O-antigen gene clusters of Yersiniapseudotuberculosis andthe cryptic O-antigen gene cluster of Yersinia pestis shows that the plague bacillus is most closely related toand has evolved from Y. pseudotuberculosis serotype O:1b. Mol. Microbiol. 37, 316-330.

Sheppard, M., Webb, C., Heath, F., Mallows, V., Emilianus, R., Maskell, D. and Mastroeni, P. (2003)Dynamics of bacterial growth and distribution within the liver during Salmonella infection. Cell. Microbiol. 5,593-600.

Snoek-van Beurden, P.A.M. and Von den Hoff, J.W. (2005) Zymographic techniques for the analysis ofmatrix metalloproteinases and their inhibitors. Biotechniques 38, 73-83.

Sodeinde, O.A., Subrahmanyam, Y.V.B.K., Stark, K., Quan, T., Bao, Y. and Goguen, J.D. (1992) Asurface protease and the invasive character of plague. Science 258, 1004-1007

Sodeinde, O.A. and Goguen, J.D. (1989) Nucleotide sequence of the plasminogen activator gene of Yersiniapestis: relationship to ompT of Escherichia coli and gene E of Salmonella typhimurium. Infect. Immun. 57,1517-1523.

Soncini, F.C. and Groisman, E.A. (1996) Two-component regulatory systems can interact to process multipleenvironmental signals. J. Bacteriol. 178, 6796-6801.

60

Sorsa, T., Ingman, T., Suomalainen, K., Haapasalo, M., Konttinen, Y.T., Lindy, O., Saari, H. and Uitto,V. (1992) Identification of proteases from periodontopathogenic bacteria as activators of latent humanneutrophil and fibroblast-type interstitial collagenases. Infect. Immun. 60, 4491-4495.

Stephens, R.W. and Vaheri, A. (1993) Plasminogen. Plasminogen activator inhibitor 1 (PAI-1). Plasminogenactivator inhibitor 2 (PAI-2). In: T. Kreis and R. Vale (eds.) Guidebook to the extracellular matrix andadhesion proteins. Oxford University Press, New York. 81-85.

Sternlicht, M.D. and Werb, Z. (2001) How matrix metalloproteinases regulate cell behavior. Annu. Rev. Cell.Dev. Biol. 17, 463-516.

Stöcker, W., Grams, F., Baumann, U., Reinemer, P. and Gomis-Rüth, F.X. (1995) The metzincins-topological and sequential relations between the astacins, adamalysins, serralysins, and matrixins(collagenases) define a superfamily of zinc-peptidases. Protein Sci. 4, 823-840.

Tamura, F., Nakagawa, R., Akuta, T., Okamoto, S., Hamada, S., Maeda, H., Kawabata, S. and Akaike,T. (2004) Proapoptotic effect of proteolytic activation of matrix metalloproteinases by Streptococcus pyogenesthiol proteinase (Streptococcus pyrogenic exotoxin B). Infect. Immun. 72, 4836-4847.

Travis, J., Potempa, J. and Maeda, H. (1995) Are bacterial proteinases pathogenic factors. Trends Microbiol.3, 405-407.

Travis, J. and Salvesen, G.S. (1983) Human plasma proteinase inhibitors. Annu. Rev. Biochem. 52, 655-709.

Tsubokura, M. and Aleksic, S. (1995) A simplified antigenic scheme for serotyping of Yersiniapseudotuberculosis: phenotypic characterization of reference strains and preparation of O and H factor sera.Contrib. Microbiol. Immunol. 13, 99-105.

Uchiya, K., Barbieri, M.A., Funato, K., Shah, A.H., Stahl, P.D. and Groisman, E.A. (1999) A Salmonellavirulence protein that inhibits cellular trafficking. EMBO J. 18, 3924-3933.

Underhill, D.U. and Ozinsky, A. (2002) Phagocytosis of microbes: complexity in action. Annu. Rev.Immunol. 20, 825-852.

Vandeputte-Rutten, L., Kramer, R.A., Kroon, J., Dekker, N., Egmond, M.R. and Gros, P. (2001) Crystalstructure of the outer membrane protease OmpT from Escherichia coli suggests a novel catalytic site. EMBO J.20, 5033-5039.

Vassalli, J.-D., Wohlwend, A. and Belin, D. (1992) Urokinase catalyzed plasminogen activation at themonocyte/macrophage cell surface: a localized and regulated proteolytic system. Curr. Top. Microbiol.Immunol. 181, 65-86.

Vazquez-Torres, A., Jones-Carson, J., Bäumler, A.J., Falkow, S., Valdivia, R., Brown, W., Le, M.,Berggren, R., Parks, W.T. and Fang, F.C. (1999) Extraintestinal dissemination of Salmonella by CD18-expressing phagocytes. Nature 401, 804-808.

Voorhis, D.L., Dillon, S., Formal, S.B. and Isberg, S.S. (1991) An O antigen can interfere with the functionof the Yersinia pseudotuberculosis invasin protein. Mol. Microbiol. 5, 317-325.

Wain, J., Diep, T.S., Ho, V.A., Walsh, A.M., Ngyen, T.T.H., Parry, C.M. and White, N.J. (1998)Quantitation of bacteria in blood of typhoid fever patients and relationship between counts and clinicalfeatures, transmissibility, and antibiotic resistance. J. Clin. Microbiol. 36, 1683-1687.

Walport, M.J. (2001a) Complement- First of two parts. N. Engl. J. Med. 14, 1058-1066.

Walport, M.J. (2001b) Complement- Second of two parts. N. Engl. J. Med. 14, 1140-1144.

Wang, L., Andrianopoulos, K., Liu, D., Popoff, M.Y. and Reeves, P.R. (2002) Extensive variation in the O-antigen gene cluster within one Salmonella enterica serogroup reveals an unexpected complex history. J.Bacteriol. 184, 1669-1677.

61

Wang, X., Lin, X., Loy, J.A., Tang, J. and Zhang, X.C. (1998) Crystal structure of the catalytic domain ofhuman plasmin complexed with streptokinase. Science 281, 1662-1665.

Watanabe, K. (2004) Collagenolytic proteases from bacteria. Appl. Microbiol. Biotechnol. 63, 520-526.

Waterman, S.R. and Holden, D.W. (2003) Functions and effectors of the Salmonella pathogenicity island 2type III secretion system. Cell. Microbiol. 5, 501-511.

Westerlund, B. and Korhonen, T.K. (1993) Bacterial proteins binding to the mammalian extracellularmatrix. Mol. Microbiol. 9, 687-694.

Wick, M.J., Harding, C.V., Twesten, N.J., Normark, S.J. and Pfeifer, J.D. (1995) The phoP locusinfluences processing and presentation of Salmonella typhimurium antigens by activated macrophages. Mol.Microbiol. 16, 465-476.

Wick, M.J., Harding, C.V., Normark, S.J. and Pfeifer, J.D. (1994) Parameters that influence the efficiencyof processing antigenic epitopes expressed in Salmonella typhimurium. Infect. Immun. 62, 4542-4548.

Wilkinson, R.G., Gemski, P. and Stocker, B.A. (1972) Non-smooth mutants of Salmonella typhimurium:differentiation by phage sensitivity and genetic mapping. J. Gen. Microbiol. 70, 527-554.

Wilson, J.J., Matsushita, O., Okabe, A. and Sakon, J. (2003) A bacterial collagen-binding domain withnovel calcium-binding motif controls domain orientation. EMBO J. 22, 1743-1752.

Wimley, W.C. (2003) The versatile -barrel membrane protein. Curr. Opin. Struct. Biol. 13, 404-411.

Woessner, J.F. and Nagase, H. (2002) Matrix Metalloproteinases and TIMPs. Oxford University Press,Oxford.

Worley, M.J., Nieman, G.S., Geddes, K. and Heffron, F. (2006) Salmonella typhimurium disseminateswithin its host by manipulating the motility of infected cells. Proc. Natl. Acad. Sci. USA 47, 17915-17920.

Xie, B., Dong, Z. and Fidler, I.J. (1994) Regulatory mechanisms for the expression of type IVcollagenases/gelatinases in murine macrophages. J. Immunol. 152, 3637-3644.

Yethon, J.A., Gunn, J.S., Ernst, R.K., Miller, S.I., Laroche, L., Malo, D. and Whitfield, C. (2000)Salmonella enterica serovar typhimurium waaP mutants show increased susceptibility to polymyxin and lossof virulence In vivo. Infect. Immun. 68, 4485-4491.