UBC Theses and Dissertations

UBC Theses Logo

UBC Theses and Dissertations

Enteropathognic Escherichia coli (EPEC) interactions with the host epithelial cell actin cytoskeleton Goosney, Danika Louise 2001

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

Item Metadata

Download

Media
831-ubc_2001-610950.pdf [ 12.38MB ]
Metadata
JSON: 831-1.0090489.json
JSON-LD: 831-1.0090489-ld.json
RDF/XML (Pretty): 831-1.0090489-rdf.xml
RDF/JSON: 831-1.0090489-rdf.json
Turtle: 831-1.0090489-turtle.txt
N-Triples: 831-1.0090489-rdf-ntriples.txt
Original Record: 831-1.0090489-source.json
Full Text
831-1.0090489-fulltext.txt
Citation
831-1.0090489.ris

Full Text

ENTEROPATHOGENIC ESCHERICHIA COLI (EPEC) INTERACTIONS WITH THE HOST EPITHELIAL CELL ACTIN CYTOSKELETON  by  DANIKA LOUISE GOOSNEY B.Sc. (Hons.), St. Francis Xavier University (1996)  A THESIS SUBMITTED IN PARTIAL FULFILLMENT OF THE REQUIREMENTS FOR THE DEGREE OF  DOCTOR OF PHILOSOPHY in THE FACULTY OF GRADUATE STUDIES Department of Microbiology and Immunology And the Biotechnology Laboratory  We accept this thesis as conformed to the required standard  THE UNIVERSITY OF BRITISH COLUMBIA March 2001 © Danika Louise Goosney, 2001  In presenting  this  degree at the  thesis  in  partial fulfilment  of  University of  British Columbia,  I agree  freely available for reference copying  of  department publication  this or  and study.  thesis for scholarly by  of this  his  or  her  Department The University of British Columbia Vancouver, Canada  requirements that the  I further agree  purposes  representatives.  may be It  thesis for financial gain shall not  permission.  DE-6 (2788)  the  that  advanced  Library shall make it  by the  understood be  an  permission for extensive  granted  is  for  allowed  that without  head  of  my  copying  or  my written  Abstract Enteropathogenic Escherichia coli (EPEC) is a gram-negative bacterial pathogen that adheres to human intestinal epithelial cells, resulting in watery, persistent diarrhea. EPEC subverts the host cell actin cytoskeleton, causing a rearrangement of cytoskeletal components into a characteristic structure, the attaching and effacing lesion, or pedestal. EPEC transmits signals through the host cell plasma membrane via direct injection of virulence factors by the type III secretion system. One injected factor is Tir, which functions as the receptor for the EPEC outer membrane protein, intimin. Upon binding intimin, Tir initiates pedestal formation. Both the amino and carboxyl termini of Tir are oriented inside the host cell cytosol, where each mediates effects on the actin cytoskeleton. Tyrosine phosphorylation at the C-terminus of Tir is required for actin polymerization to occur. A major goal of this thesis was to define the composition of the EPEC pedestal and determine which cytoskeletal components contributed to its formation. To this end, over 30 cytoskeletal proteins were identified in the pedestal tip and length. Recruitment of these proteins depended on either initial EPEC attachment, Tir insertion in the membrane, or Tir tyrosine phosphorylation. Among the proteins recruited to EPEC pedestals were N-WASP and the Arp2/3 complex, key regulators of actin polymerization. N-WASP was recruited to the pedestal tip through its GTPase binding domain and initiated pedestals through its acidic domain by recruiting the Arp2/3 complex. These data suggest that N-WASP and the Arp2/3 complex are critical effectors of actin polymerization leading to pedestal formation. To identify a binding partner of Tir, the yeast two-hybrid system and a Tir-affinity column were employed, resulting in the identification of cc-actinin as a binding partner of Tir. Direct binding was confirmed by ELISA, far western, and co-immunoprecipitation.  ii  a-actinin overexpression in HeLa cells resulted in a two-fold increase in pedestal length, suggesting it plays a key role in pedestal formation. Collectively, these results demonstrate that the EPEC pedestal is a complex structure, with Tir playing the central role in mediating its formation. This work has significantly increased our understanding of the cytoskeletal components involved in pedestal formation, and provides insight into the mechanisms of disease for this important pathogen.  Table of Contents  page ii  Abstract Table of Contents  iv  List of Figures  viii  List of Tables  x  List of Abbreviations  xi  Acknowledgements  xiv  Chapter 1 - Introduction  1  L I E . coli history and epidemiology 1.1.1. Current epidemiology 1.2 Mechanisms of EPEC infection  1 1 2  1.2.1 Attachment and effacement  2  1.2.2 Localized adherence  2  1.2.3 Pathogenicity islands, type III secretion,  3  virulence factors 1.3 Other A/E pathogens  9  1.4 The intestinal epithelial cell  13  1.4.1 Brush border  13  1.4.2 Tight junctions  16  1.4.3. Attachment to the basement membrane:  17  Integrins and focal adhesions 1.4.4. Actin polymerization at the plasma membrane 1.5 EPEC effects on the host cell  21 23  1.5.1 Disruption of tight j unctions  23  1.5.2 The Ca debate  24  1.5.3 Effects on ion secretion  24  1.5.4 EPEC interactions with phagocytic cells  27  1.5.5 Pedestal formation  27  1.5.6. Diarrhea  30  2+  iv  1.6 Objectives Chapter 2- Materials and Methods  30 33  2.1 Cell lines and bacterial strains used in studies  33  2.2 Antibodies used  33  2.3 SDS-PAGE and western immunoblot analysis  34  2.4 Protein purification  36  2.5 Affinity chromatography  38  2.6 Immunofluorescence  38  2.7 Far western analysis  39  2.8 Enzyme-linked immunosorbent assays  40  2.9 Immunoprecipitations  40  2.10 Construction of GFP-cc-actinin  41  2.11 Transfection of HeLa cells  42  2.12 Yeast two-hybrid system  42  Chapter 3 - Recruitment of cytoskeletal and signaling proteins to enteropathogenic  46  and enterohemorrhagic E. coli pedestals 3.1 Summary  46  3.2 Introduction  47  3.3 EPEC recruits several cytoskeletal and signaling proteins to  47  the pedestal 3.4 CD44 and calpactin are recruited independently of Tir  48  delivery 3.5 Gelsolin, tropomyosin, ezrin, talin and oc-actinin are  48  recruited to EPEC independently of Tir tyrosine phosphorylation 3.6 EHEC does not recruit the adaptor molecules Nek, Grb2,  48  and Crkll to its pedestal 3.7 Discussion Chapter 4 - The Wiskott Aldrich Syndrome family of proteins and the Arp2/3  58 66  complex participate in EPEC pedestal formation 4.1 Summary  66  4.2 Introduction  67  4.3 Proteins of the WASP family are recruited to the EPEC  67  pedestal  v  4.4 The WASP acidic and WH2 domains are required for pedestal  71  formation 4.5 The Arp2/3 complex is recruited to pedestals  76  4.6 The WASP GBD is necessary and sufficient for WASP localization to  76  the pedestal tip 4.7 Localization of WASP via the GBD is required for pedestal formation  82  4.8 Discussion  82  Chapter 5 - Yeast two hybrid system to identify binding partners of Tir  88  5.1 Summary  88  5.2 Introduction  89  5.3 Positive Tir-interacting proteins by yeast two hybrid screen  90  5.4 Domain mapping of Tir interactions with CD44 and profilin  92  5.5 CD44 is not required for pedestal formation  92  5.6 Tir does not co-immunoprecipitate with CD44  96  5.7 Profilin does not localize to the pedestal  96  5.8 Profilin does not bind Tir by ELISA  96  5.9 Discussion  102  Chapter 6 - Enteropathogenic E. coli translocated intimin receptor, Tir, interacts  104  directly with ct-actinin 6.1 Summary  104  6.2 Introduction  105  6.3 cc-Actinin is recruited to the pedestal in a Tir-dependent manner  105  6.4 a-Actinin elutes with Tir on a Tir affinity column  111  6.5 Tir binds purified a-actinin directly  111  6.6 a-Actinin binds the N-terminus of Tir  112  6.7 a-Actinin co-immunoprecipitates with Tir  113  6.8 a-Actinin is functional in the EPEC pedestal  122  6.9 The actin binding domain and spectrin like repeats are recruited to the  122  pedestal 6.10 Discussion  127  Chapter 7 - Discussion  128  7.1 The pedestal: A cell biologist's friend (unless taken internally)  128  7.1.1 A model for focal adhesions?  128  7.1.2 A model for receptor tyrosine kinase signaling?  131  vi  7.1.3 Two different cytosolic Tir domains, two distinct  132  populations of actin? 7.2 Mimicry of host cell function by other pathogens  136  7.2.1 EHEC: Pedestals without Tir tyrosine phosphorylation  136  7.2.2 Model for receptor tyrosine kinase signaling: Vaccinia  137  virus 7.2.3 Models for lamellipodia: Listeria monocytogenes and  137  Shigella flexneri 7.3 The pedestal: Role in disease?  141  7.4 Future directions  143  Bibliography  146  Appendix 1: EPEC inhibits its own phagocytosis  174  Appendix 2: Cdc42-homolgous protein (Chp) involvement in EPEC pedestal  199  formation Appendix 3: Actin incorporation into EPEC pedestals  202  Appendix 4: Contribution of others  204  Appendix 5: Publications arising from graduate work  205  vii  List of Figures F  i  g  U  r  e  Title  #  Page#  Fig. 1.1  The type III secretion system of EPEC  4  Fig. 1.2  The EPEC locus for enterocyte effacement (LEE)  7  Fig. 1.3  Diagram of the intestinal epithelial cell  15  Fig. 1.4  EPEC induced cytoskeletal rearrangements in the intestinal  26  epithelial cell Fig. 1.5  EPEC induced signaling events within the host intestinal  29  epithelial cell Fig.3.1  Cytoskeletal proteins recruited to the EPEC pedestal  50  Fig.3.2  CD44 and calpactin are recruited to the site of EPEC  53  adherence independently of Tir translocation Fig.3.3  Talin, tropomyosin, gelsolin, ezrin and a-actinin are  55  recruited to the site of EPEC adherence independently of Tir tyrosine phosphorylation Fig.3.4  EHEC does not recruit the adaptor proteins Crkll, Grb2  57  and Nek to the EHEC pedestal Fig.4.1  WASP is recruited beneath EPEC during actin pedestal  70  formation Fig.4.2  The C terminus of WASP is required for pedestal  73  formation Fig.4.3  Arp2/3 complex localization to the EPEC pedestal requires  75  the WASP acidic domain Fig.4.4  The GBD targets WASP to the pedestal tip  79  Fig.4.5  Model of WASP and Arp2/3 action in EPEC pedestal  87  formation Fig.5.1  Domain mapping of Tir interactions with CD44 and  93  profilin using the yeast two hybrid system Fig.5.2  CD44 is not required for pedestal formation  95  Fig.5.3  Tir does not co-immunoprecipitate with CD44  98  Fig.5.4  Profilin is not recruited to the pedestal  100  Fig.5.5  Profilin does not bind Tir directly  101  Fig.6.1  a-Actinin is recruited to the pedestal in a Tir-dependent  108  viii  manner Fig.6.2  a-Actinin recruitment coincides with Tir insertion in the  110  Hela plasma membrane Fig.6.3  a-Actinin elutes with Tir off a His-Tir Ni-NTA column  115  Fig.6.4  Tir binds a-actinin directly  117  Fig.6.5  Tir binds a-actinin through its N-terminal 1-200 amino  119  acids Fig.6.6  a-Actinin co-immunoprecipiates with Tir from EPEC  121  infected Hela cells Fig.6.7  Overexpression of a-actinin results in a two-fold increase  124  in pedestal length Fig.6.8  Domain mapping of a-actinin recruitment to the pedestal  Fig.7.1  Proposed model of EPEC-induced cytoskeletal  126 135  rearrangements Fig.7.2  Comparison of pathogens using similar components of the  140  actin cytoskeleton for their actin-based motility in the host  ix  List of Tables #  Title  Page #  1.1  LEE-encoded effector proteins  6  1.2  Pathogens that cause A/E lesions  10  1.3  Focal contact proteins  20  2.1  Antibodies used in studies  35  2.2  Recombinant constructs used in studies  37  3.1  Host proteins characterized in E. coli pedestals  60  4.1  Summary of effects of wild type and mutant WASP  81  family members on EPEC pedestal formation 5.1  Sequencing results from positive Tir-interacting  91  proteins as determined by the yeast two hybrid screen  x  List of Abbreviations A domain  Acidic domain  ABD  Actin binding domain  ABM  Actin based motility  AE  Attaching and effacing  ARP  Actin related proteins  BFP  Bundle forming pilus  BSA  Bovine serum albumin  C  Carboxyl  CBD  Calcium binding domain  CesT  Chaperone of EPEC-secreted Tir  cfm  Class-four mutant  CM  Complete media  CRIB  Cdc42/Rac interactive binding motif  DAPI  diamidino-2-phenylindole  DEPEC  Dog enteropathogenic E. coli  DMEM  Dulbecco's miminal eagle's medium  DMSO  Dimethyl sulfoxide  DSS  Di-succinyl suberate  ECL  Enhanced chemiluminescence  ECM  Extracellular matrix  EDTA  Ethyenediaminetetraacetic acid  EHEC  Enterohemorrhagic E. coli  ELISA  Enzyme-linked immunosorbent assay  EPEC  enteropathogenic E. coli  Esp  EPEC secreted protein  F-actin  Filamentous actin  FAK  Focal adhesion kinase  FCS  Fetal calf serum  GBD  GTPase binding domain  GFP  Green fluorescence protein  GST  Glutathione S-trasferase  HA  haemmagluttin  HC0 "  Bicarbonate  HeLa  Human cervix epithelial cell line  His  Histidine  HRP  Horse-radish peroxidase  3  HUS  Hemolytic uremic syndrome  IL-8  Interleukin 8  Int282  Tir binding domain of intimin  IP  inositol trisphosphate  IPTG  isopropylthio-B-D-galactosidase  Isc  Short circuit current  LB  Lurani Bertani  LEE  Locus of enterocyte effacement  Leu  Leucine  LPP  Lipoma preferred partner  MALT  Mucosal associated lymhoid tissue  MLC  Myosin light chain  MLCK  Myosin light chain kinase  N  Amino  NaPhos  Sodium Phosphate  NFkB  Nuclear factor kB  NGS  Normal goat serum  N-WASP  Neural Wiscott Aldrich Syndrome protein  OD^  Optical density at 600nm  OPD  o-phenylenediamine  PAGE  Polyacrylamide gel electrophoresis  PBS  Phosphate buffered saline  PEG  polyetheleneglycol  PEPEC  Porcine enteropathogenic E. coli  PFA  Paraformaldehyde  PI3K  Phophatidylinositol 3 kinase  PIP  Phosphatifylinositol-4,5,-bisphosphate  PKC  Protein kinase C  PLC  Phospholipase C  PMN  Polymorphonuclear monocytes  PMSF  Phenymethylsulfonyl fluoride  PP  Poly proline  REPEC  Rabbit enteropathogenic E. coli  SDS  Sodium dodecyl sulfate  SH2  Src homology 2  SH3  Src homology 3  SLR  Spectrin like repeats  TBS-T  Tris-buffered saline with Tween-20  3  2  xii  TE  Tris-EDTA  TEM  Transmission electron microscopy  Tir  Translocated intimin receptor  TMD  Transmembrane domain  Tip  Tryptophan  Ura  Uracil  VASP  Vasodilator stimulated phosphoprotein  WASP  Wiscott-Aldrich Syndrome protein  WH1/2  WASP homology 1 or 2  WIP  WASP interacting protein  WT  Wild type  ZO-1  zona occludens 1  Acknowledgements Without trying to sound like an overemotional actress receiving her very first Academy Award, I'd like to thank pretty much everybody who's been around me these past few years. First of all, a heartfelt thanks to Brett Finlay for letting me do the project I wanted even though I was a young, inexperienced graduate student who didn't know the first thing about the actin cytoskeleton or bacterial pathogenesis. What were you thinking? Thanks for giving me direction and encouragement when I was "off and letting me do my own thing when I was "on". Thanks to my committee (Rachel, Cal, and Wayne) for being as enthusiastic about my project as I was! Thanks also to Lori Graham, Bill Marshall, and Tony Miller who got me excited about lab work during my undergraduate and encouraged me to do my Ph.D. Thanks to the Natural Sciences and Engineering Research Council, the University of British Columbia, and the Biotechnology Laboratory for supporting me throughout my degree. A big kiss to the entire Finlay lab, past and present, for putting up with my hyper and excitable approach to life and science. I'm sure it was as exhausting for you as it was for me (actually, it was probably worse for you). Special thanks to Brendan who took me under his wing when I first started off in the lab (once again I ask, what were you  xiv  thinking?), Gwen for dealing with my sudden and urgent requests for couriers, references letters, addresses, phone numbers, etc., the grad students past and present with whom I could commiserate about the pains of grad school, Sam for working side-by-side with me on this project without ever losing her cool and for showing me how awesome science is done, Rebekah for the "signaling and tequila" nights, Bruce for the helpful advice on thesis writing and science in general, and Michelle and Annick for being faithful gym partners (and listeners) over the course of my degree. Big hugs to Ian for putting up with my nuttiness for the better part of a decade and to Suz, Shauna, Antonia, Brandee and Maria for the useful bitch sessions (and girls' nights out). Hugs, kisses and a whole lot more to Joe for sticking by me through the uglier parts of my Ph.D. and making me believe I am superwoman (even though he has his doubts). My utmost love and gratitude to my wonderful family (la belle Louise, Dale and Julian) who encouraged me to do whatever I was passionate about and gave me every opportunity to pursue that passion. Their support, strength, and love are why I'm still (and will continue to be) so excited about my science and my life in general.  xv  Introduction 1.1 E. coli history and epidemiology E. coli was first characterized as a cause of diarrhea in the United Kingdom in the 1940s based on epidemiological investigations of infantile gastroenteritis outbreaks (Bray, 1945). Dr. John Bray is accredited with the initial discovery based on his findings that antigenically homogenous 'Bacterium coli' strains were associated with 95% of infants with severe diarrhea (Bray, 1945). Following Bray's publication, two serogroups of Escherichia coli (as it was later renamed) were identified in association with infantile diarrhea outbreaks, 055 and 0111 (Kauffmann, 1950). The term enteropathogenic E. coli (EPEC) was first used in 1955 by Neter et al. to describe strains associated with infantile gastroenteritis (Neter, 1955). It was soon realized, however, that not all E. colis were equally pathogenic. E. coli serotypes (based on the flagellar (H) antigen as well as the lipopolysaccharide (O) antigen) varied within the serogroups, and this variability reflected in its pathogenicity. The difference in pathogenicity necessitated the term EPEC to be restricted to strains that caused diarrhea using a mechanism shared by the initial reported isolates. The term diarrheagenic was then employed as the generic term for E. coli strains capable of causing gastroenteritis. 1.1.1 Current epidemiology Today, EPEC is a predominant cause of diarrhea worldwide, posing a major health threat to children in developing countries, particularly infants less than 6 months old. Although EPEC is not a major concern in developed nations, isolated outbreaks in daycares and nurseries in developed countries occur. Additionally, the recent AIDS epidemic may provide a new source of susceptible hosts (Kotler and Orenstein, 1993). EPEC is estimated to kill several hundred thousand children a year worldwide, and is the leading cause of bacterially mediated diarrhea in children (Levine and Edelman, 1984). EPEC primarily causes chronic, watery diarrhea, often accompanied by a low-grade fever, dehydration and vomiting. Like most diarrheagenic pathogens, EPEC is  1  transmitted person-to-person via the fecal-oral route. EPEC has been isolated from asymptomatic children (and occasionally asymptomatic adults), and these carriers may be the main reservoir of infection (Chatkaeomorakot, et al., 1987, Cobeljic, et al., 1989, Gunzburg, et al., 1992). Animal reservoirs of EPEC do not exist, and contamination of food or water cause only rare and sporadic outbreaks (Schtoeder, 1968). 1.2 Mechanism of EPEC infection 1.2.1 Attachment and effacement Early ultrastructure studies of piglet intestinal tissue infected with EPEC demonstrated changes in the gut epithelium termed the attaching and effacing (A/E) lesion (Moon, 1983, Peeters, et al., 1984, Peeters, et al., 1984b, Staley, et al., 1969). This phenotype is marked by a loss, or effacement, of microvilli on the intestinal epithelial surface at sites of bacterial attachment. The bacteria intimately associate with the host cell membrane and mediate their pathogenic effects on the host cell while remaining on the intestinal cell surface. In addition to the dissolution of brush border in the vicinity of the bacterium, there is a marked change in the underlying epithelial cell cytoskeleton. A/E pathogens induce an accumulation of actin and associated proteins beneath the site of adherence. This recruitment is so characteristic that it forms the basis of a diagnostic test for A/E lesions (Knutton, et al., 1989). In some cases, A/E pathogens initiate more extensive actin polymerization resulting in the formation of an extended pseudopod, or pedestal, upon which they reside. More detail on this structure is given in section 1.5.5. 1.2.2 Localized adherence In addition to A/E lesion formation, EPEC infection is characterized by the ability to adhere to intestinal epithelial cells in a distinct pattern called localized adherence(Scaletsky, et al., 1984). Localized adherence is typified by EPEC aggregrating together in microcolonies rather than a simple diffuse adherence on the cell surface. Microcolony formation is highly correlated with clinical isolates of EPEC. It was  2  discovered that localized adherence was dependent on the presence of a large, 60MDa plasmid, pMAR2 (Baldini, et al., 1983). Plasmid-cured strains lost their ability to form microcolonies and were significantly less pathogenic in human volunteers than the parent strain from which it was derived (Levine, et al., 1985). The pMAR2 plasmid encodes fimbriae termed bundle-forming pili (BFP) (Giron, et al., 1991). BFP are surface organelles that mediate adherence of EPEC to host cells and to neighboring EPEC. They are designated as bundle-forming because pilus filaments emanating from the EPEC surface align in a parallel fashion to form rope-like bundles that intertwine and link individual bacteria together into microcolonies (Giron, et al., 1991). Bfp biogenesis is dependent on a 14 gene operon that encodes the major subunit of Bfp, bfpA, the transcriptional factor per A that is encoded elsewhere on the pMAR2 plasmid, and the chromosomal dsbA gene that is required for disulfide bond formation of BfpA (Puente, et al., 1996, Sohel, et al., 1996).  1.2.3 Pathogenicity islands, type III secretion, virulence factors Attaching and effacing pathogens such as EPEC use a specialized apparatus, the type III secretion system, to translocate a small number of virulence factors directly into host cells (Jarvis, et al., 1995) (Fig.1.1). Type III secretion systems are multimeric (-20 protein components) molecular machines that are exclusively present in pathogenic strains of gram negative bacteria (reviewed in (Hueck, 1998)). Analysis of the genes encoding type III secretion systems in various distantly related pathogens reveals that the subunits that comprise the secretion apparatus are highly conserved between bacterial species, although the secreted factors themselves are quite divergent (Hueck, 1998). Type III secretion system genes are usually clustered on plasmids or on chromosomal "pathogenicity islands" - regions of the bacterial chromosome that differ from the rest of the genome with respect to G + C content and codon usage (Hueck, 1998). It is thought that type III secretion systems are recent genetic acquisitions to pathogenic genomes, although their origin is unknown.  3  Fig.1.1: The type III secretion system of E P E C used to deliver virulence factors, including Tir, into the host cell cytosol or membrane. Several gram-negative pathogens use this conserved secretion system to deliver diverse effectors into host cells to mediate several different effects within mammalian and even plant cells (from Goosney et al.  2000).  4  In EPEC, a 35 kb chromosomal pathogenicity island known as the locus for enterocyte effacement (LEE) is necessary and sufficient for ATE lesion formation (McDaniel and Kaper, 1997) . The LEE contains at least 41 open reading frames that are organized into 5 polycistronic operons (Elliott, et al., 1998, Mellies, et al., 1999)(Fig.l.2). Three of these operons encode the structural subunits of the secretion apparatus that form the channel across the inner and outer membrane of EPEC(sep and esc genes), as well as the transcriptional regulator, ler. A fourth operon contains the genes mediating intimate attachment of the bacteria to host cells: eae, tir, and Tir's chaperone cesr(Tablel.l). Intimate attachment of bacteria to cells is mediated by intimin, a bacterial outer membrane adhesin, binding its receptor, Tir, which is inserted into host plasma membranes. It was originally proposed that the receptor for intimin was a host protein that was phosphorylated upon bacterial infection (Rosenshine, et al., 1992). However, this receptor has recently been shown to be a bacterially-encoded protein that is translocated into the host cell via the type III secretion apparatus (Kenny, et al., 1997). Accordingly, what used to be referred to as Hp90 (host protein, 90 kDa), was renamed Tir (translocated intimin receptor). Tir is injected by bacteria into host cells, where it is serine/threonine and tyrosine phosphorylated and inserted into the host cell plasma membrane (Kenny, et al., 1997). Tir spans the plasma membrane in a "hairpin" structure, with two transmembrane domains, an extracytoplasmic loop that contains the intimin binding region, and intracytoplasmic amino (N)- and carboxyl (C)-termini (deGrado, 1999, Kenny, 1999). The C-terminal of Tir contains several tyrosine residues, one of which (Y474) is phosphorylated upon translocation into host cells. Phosphorylation of this tyrosine is required for A/E lesion formation, although the kinase involved and the downstream effects of this phosphorylation has not yet been characterized (Kenny, 1999). Intimin, anchored in the outer membrane of bacterial cells, binds and clusters Tir in the host cell membrane. This clustering is thought to trigger changes in associated host cell  5  cytoskeletal proteins leading to the formation of characteristic pedestals underneath adherent bacteria.  Table 1.1: LEE-encoded effector proteins Protein  Localization  Intimin  Secreted  Translocated  Function  N  N  adhesin  EPEC outer  binds Tir  membrane Tir  host cell plasma  Y  Y  Receptor for intimin Binds host cell  membrane  cytoskeletal proteins initiates actin pedestal formation cesT  bacterial cytoplasm  N  N  Tir's chaperone within bacteria  EspA  bridges bacteria  Y  N  delivery tube?  Y  Y  translocation pore?  and host EspB  host membrane  cytosolic effector?  /cytosol EspD  host membrane  Y  Y  translocation pore?  EspF  host  Y  Y  unknown  6  LEE operons Premolars  LEE1  LEE 2  hEEl  Tir  hEE2 HEE3 __di  rori1,2 ' ori2345e$cRSTU  rortf  ler/orf1  LEE 3  rH tir  I Orf/O.fi  r__ eae  escD  espADB  aspf  orffj  sep7 orftf  r> m a p p e d p r o m o t e r - * •O p e r o n  LEE 4  t e n c o d i n g t y p e III s e c r e t i o n s y s t e  me n c o d i n g s e c r e t e d p r o t e i n s encoding other or cryptic functions  Fig. 1.2. The EPEC locus for enterocyte effacement (LEE; from Elliot et al. 1998).  7  The fifth operon of the LEE region contains the esp (E. coli secreted protein) genes, which encode proteins (other than Tir) that are secreted by the type III system. There are four such identified proteins in EPEC: EspA, B, D and F. Deletion of EspA, B or D abolishes A/E lesion formation (Foubister, et al., 1994, Kenny, et al., 1996, Lai, et al., 1997), although this may be due to a role of these proteins in Tir delivery to the host cells, rather than through a direct role as effectors. EspA is an integral part of a filamentous organelle that bridges the EPEC bacteria to host cells preceding A/E lesion formation (Knutton, et al., 1998). It has been proposed that this filament forms a hollow tube that serves as a conduit for translocating bacterial proteins including Tir. EspB and EspD are thought to form a pore complex in the host cell plasma membrane. EspD and EspB are both predicted to possess transmembrane domains, and have been found associated with infected host cell membranes (Kresse, et al., 1999, Wachter, et al., 1999, Wolff, et al., 1998). Once in the host cell membrane, EspD is resistant to extraction with high salt but sensitive to externally added trypsin, indicating that part of the protein remains extracellular (Wachter, et al., 1999). It has been proposed that the extracellular portions of EspD may be involved in interaction with EspAcontaining filaments (Fivaz and van der Goot, 1999). This may explain the observation that EspD is required for proper formation of the EspA-containing filaments (Knutton, et al., 1998). Further evidence for a role of EspB and D in pore formation comes from an assay of hemoglobin release by red blood cells (Warawa, et al., 1999). In this assay, EPEC co-incubation with erythrocytes results in release of hemoglobin, presumably through a pore created in the erythrocyte membrane. Hemoglobin release requires an intact type III secretion apparatus, bacterial contact with the erythrocytes, as well as EspA, B, and D, suggesting that these proteins are needed for pore formation. EspB may play an additional role in host cells as a cytosolic effector protein. Translocated EspB has been localized to the host cell cytosol (Taylor, et al., 1998) and  8  transfection of EspB in HeLa cells results in a cytoplasmic localization of the protein (Taylor, et al., 1999). Cells expressing transfected EspB display abnormal cellular morphology and a decrease in the number of actin stress fibers. This has prompted the hypothesis that EspB depolymerizes filamentous actin within host cells, increasing the number of actin monomers available for pedestal assembly. However, transfected EspB may not act analogously to the bacterially translocated protein. The heterologous expression of EspB in HeLa cells could not correct the inability of an espB mutant to form ATE lesions on host cells. It is tempting to speculate that this non-complementation was due to a requirement for membrane-localized EspB to form a translocation pore for other bacterial proteins. Indeed, Tir is not translocated into host cells by an espB mutant bacterium. Thus, it appears that EspB may be both part of the translocation apparatus, and a translocated effector. EspF is the most recently described type III secreted effector protein in EPEC (McNamara and Donnenberg, 1998). EspF is a proline-rich protein that is delivered to the host cell cytosol. Deletion of espF in EPEC has no effect on A/E lesion formation, and its function remains undefined. 1.3 Other A/E pathogens There are other members of the A/E family of pathogens in addition to EPEC. These include enterohemorrhagic E. coli (EHEC; 0157:H7), a human pathogen, and numerous animal pathogens that cause disease in rabbits (REPEC, RDEC-1), pigs (PEPEC), dogs (DEPEC) and mice (Citrobacter rodentium) (Table 1.2). These all induce actin polymerization and pedestal formation in a manner analogous to EPEC. EPEC serves as the prototype for this small family of gram negative pathogens that form A/E lesions on host cells. Enterohemorrhagic E. coli (EHEC; 0157:H7) is a closely related pathogen that is responsible for numerous high-profile food poisoning outbreaks in North America. The causative agent of "hamburger disease", EHEC produces acute gastroenteritis, hemorrhagic colitis, and in severe cases, a systemic kidney disease known as  9  hemolytic uremic syndrome (HUS). The increased severity of EHEC-mediated disease in comparison to EPEC-mediated disease is thought to be due to the presence of other additional virulence factors in EHEC, including the Shiga-toxin. Although these additional virulence factors are responsible for some of the more severe symptoms associated with EHEC disease, A/E lesion formation is thought to contribute to the development of non-bloody diarrhea early in the infection (Tzipori, et al., 1987).  Table 1.2: Pathogens that cause A/E lesions. Strain  Host  Symptoms  EPEC  Human; children <2 yr and  profuse watery diarrhea,  adults at high inocula  low grade fever, vomiting  Human; both adult and  profuse bloody diarrhea;  children  hemolytic uremic  EHEC  syndrome; hemorrhagic colitis; REPEC/RDEC-1  Rabbit  profuse, watery diarrhea  Citrobacter rodentium  Mouse  colonic hyperplasia; rectal prolapse in young mice  10  The LEE is quite conserved between EPEC and EHEC, although there are some notable differences. Whereas transfer of the LEE of EPEC is sufficient to confer A/E lesion ability to non-pathogenic E. coli strains (McDaniel and Kaper, 1997), transfer of the EHEC LEE is not sufficient (Elliott, et al., 1999). This suggests that EHEC requires other non-LEE encoded factors to produce cytoskeletal changes in host cells. These factors have not yet been identified, but probably include regulators and other accessory factors. While the genes encoding the components of the type III apparatus of EPEC and EHEC are highly conserved, there is some divergence in the sequence of the secreted proteins, probably due to immune response pressure. Tir is quite divergent between EPEC and EHEC, particularly in the C-terminal of the protein. Notably, the C-terminal tyrosine that is required to be phosphorylated for pedestal formation in EPEC is not present in EHEC, and A/E lesion formation occurs independently of tyrosine phosphorylation (Deibel, et al., 1998). The mechanistic differences that allow EHEC to form pedestals in the absence of Tir tyrosine phophorylation have not yet been characterized. EPEC and EHEC have a narrow host specificity, and are human pathogens. However, there exist some related bacteria that share virulence factors and mechanisms with EPEC that cause analogous disease in animals. These provide convenient animal models to study the role of A/E lesion formation in pathogenesis. Rabbit enteropathogenic E. coli, or REPEC, encompasses several E.coli serotypes that cause A/E lesion formation and diarrheal disease in rabbits. The two most studied serotypes of REPEC are 015 (also known as RDEC or rabbit diarrheagenic E. coli) and 0103. Although the initial attachment of REPEC to intestinal surfaces differs from that described for human EPEC, the mechanism for intimate attachment and host cell cytoskeletal perturbation is very similar. In all cases this involves aggregation of actin and phosphotyrosine beneath adherent bacteria leading to gross cytoskeletal perturbations including microvilli effacement and pedestal formation (De Rycke, et al., 1997, Philpott,  11  et al., 1995). The LEE region has been characterized in RDEC and REPEC O103, and is highly homologous to that of EPEC (Jerse, et al., 1990). REPEC and EPEC have similar profiles of LEE-encoded secreted proteins (EspA, B, D, and Tir) and produce the outer membrane protein intimin (Abe, et al., 1998). The rabbit model has facilitated definitive demonstration of the requirement of functional type III secretion, EspA and EspB for A/E lesion formation and for disease (Abe, et al., 1998), defining these as true virulence factors. Citrobacter is a gram negative bacterial species that is phylogenetically classified between E. coli and its close relative Salmonella. Citrobacter rodentium (formerly known as Citrobacter freundii, biotype 4280) is a pathogen of mice that contains a LEE and forms A/E lesions in mouse intestinal cells (Schauer and Falkow, 1993). The A/E lesions produced by Citrobacter are morphologically identical to those produced by EPEC, and intimin from EPEC can complement a Citrobacter intimin mutant (Frankel, et al., 1996). However, in contrast to the diarrheal diseases mediated by EPEC, EHEC, or REPEC, Citrobacter rodentium infection results in proliferation of the epithelial cells in the mucosa of the colon, a disease called transmissible murine colonic hyperplasia (Barthold, et al., 1976). The mechanism leading to this hyperplasia is not well understood. Intimin and EspB are both required for development of hyperplasia, suggesting a possible link between A/E lesion formation and the proliferative disease (Newman, et al., 1999, Schauer and Falkow, 1993, Schauer and Falkow, 1993). However, in one report, mice innoculated with formalin-killed bacteria still developed an intermediate degree of hyperplasia (Higgins, et al., 1999). Since dead bacteria would not be predicted to form A/E lesions, the link between A/E lesion formation and hyperplasia requires further study.  12  1.4 The intestinal epithelial cell EPEC infection occurs in the small intestine of its host and alters the host cell actin cytoskeleton to produce the A/E lesion (or pedestal). It is important, therefore, to gain an understanding of the epithelial cell structure and function with respect to the actin cytoskeleton. The intestinal epithelial cell is polarized, comprised of an apical brush border, basolateral attachments to the underlying matrix, and tight junctions that form a barrier between apical and basolateral surfaces (Fig. 1.3). Cell polarity allows epithelial cells to perform specialized functions, including absorption and secretion. Establishing and maintaining cell polarity requires: (1) proper sorting through the trans Golgi network of basolateral and apical proteins; (2) targeting of carrier vesicles to the correct surfaces; (3) retention of the polarized components within the proper area once they have been delivered (Rodriguez-Boulan and Nelson, 1989, Simons and Wandinger-Ness, 1990) (Rodriguez-Boulan and Salas, 1989). The cytoskeleton appears to play a key role in these mechanisms. For example, the microtubule network is oriented along an apicalbasal axis that is critical in positioning transport carrier vesicles (Bacallao, et al., 1989). The intermediate filament network, particularly a 53 kDa cytokeratin, may also play a role in apical polarization (Rodriguez, et al., 1994). Actin, although found in the apical, basolateral, and lateral submembrane cytoskeleton in a non-polarized manner, is highly polarized in the apical surface of intestinal cells in the brush border.  1.4.1 Brush border The apical surface of the intestinal epithelium is made up of a crypt-villus functional unit, in which the villi are lined by absorptive, goblet, and endocrine cells while the crypt cells consist primarily of stem cells (Beaulieu, 1999). Proliferating cells migrate, differentiate, and mature from the stem cells in the crypt to the tip of the villus. During the migration of the intestinal epithelial cell from the crypt to the tip of the villus, microvilli are created.  13  Microvilli are extensions from the surface of most cells, but are best characterized in the brush border of intestinal epithelial cells. They function to increase the surface area of the gut to allow for increased absorption. Each brush border cell of a vertebrate intestinal epithelium has over 15,000 microvilli extending from its apical surface, each containing approximately 20 filaments (DeRosier and Tilney, 2000). The barbed, or fast growing, ends of these filaments are facing the tip of the microvillus, embedded in a currently poorly characterized dense patch of material called the terminal web. It is believed that the cyoskeletal protein ezrin may mediate linkages from the actin cytoskeleton to the tip of the microvillus (Hanzel, et al., 1991). Actin filaments are bundled by three actin-binding proteins: villin, fimbrin/plastin and small espin (DeRosier and Tilney, 2000). Villin is a Ca and phosphoinositide2+  regulated protein that caps, nucleates, severs, and bundles actin (Bretscher and Weber, 1980, Craig and Powell, 1980, Glenney, et al., 1981, Mooseker, et al., 1980). Bundling activity is dispensible in vivo, presumably because either espin, fimbrin, or both may compensate for the loss of villin (Ferrary, et al., 1999). Villin severing activity, however, is critical in vivo (Ferrary, et al., 1999). Fimbrin/plastin is also calcium sensitive, with its actin-bundling activity being inhibited in high levels of calcium (Hanein, et al., 1998). Small espin appears to accumulate later during brush border development and its presence is minor in comparision with villin and fimbrin/plastin (Bartles, et al., 1998). It may function in an already assembled microvillus to lock the filaments in place in a Ca  2+  insensitive fashion.  14  Fig. 1.3. Schematic diagram of the intestinal epithelial cell, including microvilli, tight junctions, and attachment to the ECM. See text for details.  t  1.4.2 Tight junctions Tight junctions, or zona occludens, play two main roles in the intestinal epithelium: they restrict movement across the epithelium of small particles and ions by forming a tightly regulated barrier, and they separate the apical and basolateral surface of the polarized cell (reviewed in (Stevenson and Keon, 1998)). Tight junctions are comprised of eleven known proteins - nine peripheral proteins, one integral membrane protein and actin. ZO-1, ZO-2, and ZO-3 are all specifically localized to tight junction and may function in signaling events and possibly tumour suppression (Haskins, et al., 1998, Jesaitis and Goodenough, 1994, Stevenson and Keon, 1998, Stevenson, et al., 1986). Cingulin, 7H6, symplekin, and 19B1 have all been localized to the tight junction but have currently undefined functions (Citi, et al., 1988, Keon, et al., 1996, Merzdorf and Goodenough, 1997, Zhong, et al., 1993). Rab3b is a GTPase that is involved in membrane trafficking events and its presence in the tight junction suggests a role for trafficking in junctional assembly and maintenance (Weber, et al., 1994). AF-6, a ZO-1 binding partner, is a target of another GTPase, Ras, further suggesting a role for the GTPase in junctional function (Prasad, et al., 1993). Occludin is the only integral membrane protein and is critical in maintaining proper paracellular permeability (Wong and Gumbiner, 1997). Disruption of occludin in the epithelium results in an increase in transepithelial flux of nontransported solutes (Balda, et al., 1996) (McCarthy, et al., 1996). In addition to the proteins specifically localized to tight junctions, actin is also required for tight junction formation. Actin plays an active role in maintaining junctions as actin disruption results in loss of paracellular integrity (reviewed in Stevenson and Keon, 1999). Furthermore, myosin, although excluded from tight junctions, can control the underlying contractility of the cytoskeletal network underlying the tight junction, thereby regulating its permeability (Broschat, et al., 1983). 16  1.4.3 Attachment to the basement membrane: Integrins and focal adhesions The intestinal epithelium is built upon a continuous sheet of specialized extracellular matrix (ECM), the basement membrane. The basement membrane is comprised of glycoproteins such as collagens, laminins, and proteoglycans, which contribute to both cell adhesion and structural stability to the enterocyte during its development (Beaulieu, 1999). Binding of the epithelium to the ECM occurs primarily through a family of proteins called integrins. ECM-integrin interactions are mainly studied in tissue culture where the cultured cell attaches to its substratum through adhesion sites termed focal contacts or focal adhesions. Integrins are aP heterodimeric transmembrane proteins. There are at least 14 different a subunits and 8 different P subunits, that when combined make over 20 different integrin receptors (Hynes, 1992). Upon binding to the ECM, integrins are clustered and undergo conformational changes that permit them to signal to the cytoskeleton (referred to as 'outside-in' signaling). In order for ligand-integrin interactions to occur, the integrin itself needs to be activated on its cytoplasmic side, a process referred to as 'inside-out' signaling. Further complexities of integrin signaling manifest themselves in the specific hierarchy of cytoskeletal activation. Integrin aggregation, ligand occupancy, or both trigger a series of signaling events within the cell (Miyamoto, et al., 1995). Tyrosine phosphorylation of the pi integrin cytoplasmic subunit adds yet another level of signaling once clustering and ligand occupancy occurs (Miyamoto, et al., 1995). The ECM is linked to the cytoskeleton through 'anchor proteins' that bind the cytoplasmic tail of pi integrins. The most common and well-defined linkages occur through two structural proteins, a-actinin and talin (Otey, et al., 1990, Rees, et al., 1990). a-Actinin is a major constituent of adhesion plaques, and is also found in stress fibres, lamellae, and the cortical actin network. It is an actin cross-linking protein that forms a rod-like anti-parallel homodimer (Puius, et al., 1998). It interacts with several  17  cytoskeletal and signaling proteins and membrane lipids, including pi integrins, filamentous (F)-actin, phosphatidylinositol(3,4,5)trisphosphate (PIP ), 3  phosphatidylinositol 4,5-bisphosphate (PIP ), zyxin and vinculin (Blanchard, et al., 1989, 2  Fukami, et al., 1992, Greenwood, et al., 2000, Kroemker, et al., 1994, Reinhard, et al., 1999). Talin is a 270kDa protein that also forms an actin cross-linking rod-like dimer (reviewed in(Critchley, 2000)). It interacts with pi integrins, F- actin, vinculin, and lipid membranes (Bass, et al., 1999, Heise, et al., 1991, Hemmings, et al., 1996). The interaction of a-actinin and talin with integrin occurs following ligand occupancy and clustering, but upstream of tyrosine phosphorylation events and F-actin recruitment (Miyamoto, et al., 1995). In addition to a-actinin and talin, numerous other structural and signaling molecules are recruited to the focal contact in response to ECM binding. A list of these proteins and their functions within the focal contact is given in Table 1.3. Structural analyses of the components listed in Table 1.3 show motifs that are shared among different groups. Actin-binding domains are found on several of the proteins, including talin, a-actinin, fimbrin, vinculin, tensin and VASP (Bendori, et al., 1989, Kuhlman, et al., 1992, Lo, et al., 1994, Niggli, et al., 1994, Reinhard, et al., 1992). LIM domains (named after proteins encoded by the Lin-11, Isl-1, and Mec-3 genes) are Zn binding 2+  regions that mediated protein-protein interactions in proteins like zyxin, LPP, paxillin, and cysteine-rich protein (CRP) (Petit, et al., 2000, Sadler, et al., 1992, Turner and Miller, 1994). Src-homology 2 (SH2) domains mediate binding of proteins to specific phosphotyrosine motifs and are found in numerous proteins, including paxillin, tensin, Grb2, pp60 src, Crk, pl30cas, and the p85 subunit of P-I-3kinase (reviewed in Buday, 1999). Src-homology 3 (SH3) domains mediate protein-protein interactions by binding to poly-proline rich regions (reviewed in(Buday, 1999)). Proteins in focal contacts that contain SH3 domains include Grb2, Crk, pl30cas, and pp60src. Given the large number of proteins found in focal contacts, it is clear that not all the interactions that have been  18  reported in the literature are actually happening in vivo. One also has to assume that a certain amount of redundancy of function in these proteins occurs in the cell.  19  Table 1.2: Focal adhesion proteins Proteins found in Focal Adhesions a-actinin  Function  Reference  Cross-links actin; binds integrin and vinculin  actin  Major subunit of the microfilament  CRP (Cysteine rich protein)  Binds zyxin  tensin  Caps and cross-links F-actin; binds phosphotyrosine residues of several proteins Protein tyrosine kinase; binds integrin, paxillin, and SH2 domains of several proteins Binds vinculin, actin and integrins  reviewed in (Critchley, 1999) reviewed in (Chen, et al., 2000) (Sadler, et al., 1992) (Lo, et al., 1994)  FAK (focal adhesion kinase) talin pl30cas PI3 kinase Vinculin  Src family and FAK substrate; Binds SH2 domain of several proteins Generates PIP ; targetting subunit p85 binds PY residues on several proteins Binds talin, a-actinin , actin, paxillin, and itself 3  Paxillin  Binds vinculin, FAK, SH2 and SH3 domains of several proteins  Src family kinases  Protein tyrosine kinases;  VASP (vasodilatorstimulated phosphoprotein)  Binds actin, zyxin, LPP and profilin  zyxin  Binds a-actinin and VASP; can target to the nucleus  Lipoma-preferred partner  Homologue of zyxin  profilin  Monomeric actin sequestering protein  PLCy (Phopholipase C)  Gerenates diacylglycerol (DAG) and IP from PIP Binds vinculin, talin;Serine/threonine kinase 3  PKC (protein kinase C)  2  (Schaller and Parsons, 1994) reviewed in (Critchley, 2000) reviewed in (O'Neill, et al., 2000) reviewed in (Leevers, et al., 1999) reviewed in (Critchley, 2000) reviewed in(Turner, 1998) reviewed in (Thomas and Brugge, 1997) reviewed in (Holt, et al., 1998) (Crawford and Beckerle, 1991) (Petit, et al., 2000) (Sohn and GoldschmidtClermont, 1994) (McBride, et al., 1991) (Hyatt, et al., 1994)  20  1.4.4 Actin polymerization at the plasma membrane A common thread among the structures discussed above (microvilli, tight junctions and focal adhesions) is the actin cytoskeleton. The actin cytoskeleton is a dynamic network that is essential for cellular processes such as movement, polarization, morphogenesis, endocytosis, phagocytosis, and cell division. Actin is a 42kDa protein that exists in monomeric and polymeric pools. For actin polymerization to occur, a trimer (or nucleus) of actin monomers must first form. This nucleation is the ratelimiting step of actin polymerization. Following nucleation, actin monomers add onto the fast-growing end of the actin filament (called the barbed end). Actin monomers are removed from the opposite end of the actin filament (the slow-growning or pointed end). This process is referred to as "treadmilling" of actin. Production of free fast-growning ends of actin occurs in three main ways: (1) de novo synthesis of actin filaments (Zigmond, 1998); (2) severing preexisting filaments (Arber, et al., 1998); and (3) uncapping preexisting filaments (Hartwig, et al., 1995). De novo synthesis requires an initial nucleation step, bringing actin monomers together to form a nucleus of actin. From this trimer grows a newly formed actin filament. Nucleation occurs through a complex of 7 proteins, termed the Arp2/3 complex (Machesky, et al., 1994, Welch, et al., 1997). Actin-related proteins (Arp) 2 and 3, together with one monomer of actin, form the nucleus (Mullins, et al., 1998). In addition to its nucleating activity, the Arp2/3 complex also caps actin filaments at its pointed (or slow-growing) ends to initiate growth in the barbed direction (Kelleher, et al., 1995). The Arp2/3 complex is targeted to the plasma membrane and activated by proteins in the Wiskott-Aldrich Syndrome protein family (WASP), including the ubiquitously expressed neural-WASP (N-WASP) (Miki, et al, 1996, Symons, et al., 1996). N-WASP is capable of interacting with numerous signaling molecules, including: Arp2/3 at its acidic Ctermini; the GTPases Cdc42, Rac and Chp through its N-terminal Cdc42/Rac-interactive binding motif (CRIB; or GTPase binding domain (GBD)); PIP and WASP-interacting 2  21  protein (WIP) through its N-terminal WH1 domain; actin through its C-terminal cofilin homology domain; the SH3 domains of several proteins, including the adaptor molecules Nek and Grb2; and the actin-sequestering protein profilin, through its large stretch of poly-prolines (reviewed in Mullins, 2000). The large number of binding motifs found in N-WASP suggests numerous ways to regulate actin polymerization. Currently, N-WASP is known to be activated by Cdc42 and PIP (Prehoda, et al., 2000, Rohatgi, et al., 2000). 2  Binding of Cdc42 to the CRIB domain of N-WASP dramatically increases its ability to nucleate actin through the Arp2/3 complex. PIP acts synergistically with Cdc42 to 2  enhance the nucleation capacity of Arp2/3 through N-WASP (Prehoda, et al., 2000, Rohatgi, et al., 2000). It is thought that a conformational change in N-WASP is caused by Cdc42 and PIP binding, unfolding the protein to expose its Arp2/3-binding C2  terminus. This theory is supported by evidence that a C-terminal truncated form of NWASP appears constitutively active in binding the Arp2/3 complex (Prehoda, et al., 2000). In addition to actin nucleation by N-WASP/Arp2/3 activation, free barbed ends may be generated by severing preexisting filaments. Severing occurs by weakening covalent bonds between actin monomers in an actin filament. The most potent actinsevering protein to date is gelsolin. Gelsolin binds to the side of an actin filament and induces a conformational change (a kink) in the actin filament, providing a mechanical basis for severing (McGough, et al., 1998). Following severing, gelsolin remains attached to the barbed end of the filament in a process called capping, thereby preventing newly severed filaments from interacting with one another. Gelsolin activity is regulated by Ca and PIP (Hartwig, 1992, Janmey and Stossel, 1987). Ca binding to the C-terminus 2+  2+  2  of gelsolin changes its conformation to expose its actin-binding domain (Selden, et al., 1998). PIP inactivates gelsolin activity by binding at its N-terminus and dissociating it 2  from actin (Janmey, et al., 1992).  22  Barbed end nucleating sites can also be created by uncapping them without severing. CapG and capping protein are two proteins that cap existing filaments and terminate actin polymerization. PIP regulates capping in the cell, eliciting a dissociation 2  of the capping proteins from actin, making them available for actin polymerization (Bamburg, 1999). Cells probably use a combination of de novo synthesis, severing, and uncapping to elicit the wide variety of responses of cytoskeletal rearrangement in response to numerous external stimuli.  1.5 E P E C effects on the host cell 1.5.1 Disruption of tight junctions In addition to the dramatic actin restructuring leading to A/E lesion formation, EPEC also alters the phosphorlyation states of several host proteins. Among the proteins prominently phosphorylated upon EPEC infection is myosin light chain (MLC), a 20 kDa protein involved in the contractility of the cytoskeleton underlying the tight junction (Fig. 1.4). MLC can be phosphorylated on serine or threonine to produce two different isoforms of the protein (Manjarrez-Hernandez, et al., 1996). During EPEC infection, MLC is initially found in the cytosol and is phosphorylated only on the threonine residue. During EPEC infection, MLC is phosphorylated on the serine residue and becomes associated with the cytoskeletal fraction of the host cell. The initial phosphorylation event probably occurs via protein kinase C (PKC) (Baldwin, et al., 1990), which has been shown to be activated upon EPEC infection (Crane and Oh, 1997). The second phosphorylation event of MLC occurs by a different kinase, most likely myosin light chain kinase (MLCK; Manjarrez-Hernandez, et al., 1996). Indeed, inhibition of MLCK causes a decrease in MLC phosphorylation induced by EPEC infection (Yuhan, et al., 1997). These effects on MLC ultimately lead to a disruption of tight junction integrity and an increase in paracellular permeability.  23  1.5.2 The C a  2+  debate  Several studies have indicated that EPEC infection of epithelial cells stimulates a rise in the levels of intracellular Ca  2+  concentration, due to the release of Ca from 2+  intracellular stores, possibly 1,4,5-inositol trisphosphate (IP ) sensitive stores (Baldwin, 3  et al, 1993, Baldwin, et al., 1991, Dytoc, et al., 1994). Earlier studies showed that EPEC could indeed affect inositol phosphate signaling within the host cell (Foubister, et al., 1994). Furthermore, there are data that demonstrate stimulation of PLC-yl upon EPEC infection, which could be responsible for the increase in IP (Kenny and Finlay, 1997). 3  Despite the evidence supporting a role for intracellular C a  2+  in EPEC infection,  Bain et al. provided evidence suggesting there was no increase in Ca  2+  during the course  of infection (Bain, et al., 1998). In an attempt to analyze the temporal and spatial distribution of C a  2+  during EPEC infection, the authors saw no increase at all during the  course of their studies. The differences between the conflicting results could lie with the different techniques used to measure Ca  2+  fluxes, but still need to be resolved.  1.5.3 Effects on ion secretion With a disruption of tight junction permeability in polarized epithelial cells, it follows that there is a change in transepithelial resistance induced by EPEC (Philpott, et al., 1996, Spitz, et al., 1995). Increasing tight junction permeability could alter normal ion transport mechanisms, allowing electrochemical gradients to reach an equilibrium. In addition to this disruption, EPEC alters net ion secretion in polarized epithelial cells. In fact, it inhibits net ion secretion stimulated by classic secretagogues, resulting in a decreased short circuit current (Isc) (Hecht and Koutsouris, 1999, Philpott, et al., 1996). This attenuation is not due to Cf secretion, but may be partly dependent on HC0 " 3  secretion (Hecht and Koutsouris, 1999). Conflicting data from Collington et al. (Collington, et al., 1998) suggests that EPEC can induce a transient, rapid increase in Isc  24  that is at least partially Cl" dependent. The differences between the method of infection and the cell lines used could account for these discrepancies.  25  EPEC  Fig. 1.4: EPEC induced signaling events within the host intestinal epithelial cell. See text for details.  26  1.5.4 EPEC interactions w i t h phagocytic cells In addition to the effects observed in intestinal epithelial cells, EPEC also interacts with macrophage-like cells in culture and inhibits its own phagocytosis (Goosney, et al., 1999). This inhibition is partially dependent on pedestal formation, and requires type III secretion of the Esps. Upon delivery of secreted proteins to the host cell, several currently unidentified host proteins are tyrosine dephosphorylated. This dephosphorylation directly correlates with the anti-phagocytic phenotype, indicating that either EPEC injects a tyrosine phosphatase directly into the host or, more likely, activates a host tyrosine phosphatase that dephosphorylates host proteins essential for phagocytosis. 1.5.5 Pedestal formation  The secretion of EPEC virulence proteins leads to the loss of microvillar structure and the reorganization of the underlying actin cytoskeleton into the characteristic A/E lesion, or pedestal. These structures can extend up to 10 um in length beneath the pathogen (Rosenshine, et al., 1996). Tir is located at the very tip of the pedestal, where it functions as the receptor for EPEC intimate attachment to the cell via intimin (Fig. 1.5). Binding of Tir to intimin focuses Tir beneath EPEC, initiating pedestal formation. Tir that has been delivered to the host cell but does not bind its ligand (using an intimin mutant) still recruits cytoskeletal proteins, but the recruitment is diffuse and unfocused, appearing as a cloud surrounding adherent bacteria. There is evidence that correlate pedestal formation with EPEC-mediated disease, although to date there is no causal relationship. Mutants of REPEC (the rabbit form of EPEC) lacking either Tir or intimin (but still retaining type III secretion and other virulence proteins) do not form pedestals nor did they cause diarrhea in a rabbit model system (Marches, et al., 2000). These data suggest that Tir and intimin are virulence factors in EPEC disease, although their exact roles remain unknown.  27  Fig. 1.5. EPEC-induced cytoskeletal rearrangements in the intestinal epithelial cell, a) EPEC pedestals on a HeLa cell: Tir (yellow; arrow) is at the tip of the pedestal; actin (green) is at the tip and along the length of the pedestal. The bacterial D N A is stained with DAPI (blue), b) A model of EPEC cytoskeletal rearrangements that leads to pedestal formation. See text for details.  28  29  1.5.6 Diarrhea Despite the advances made in EPEC-host cell interactions, little is known about how it actually triggers diarrhea in the host. There are probably several mechanisms involved in initiating and maintaining chronic diarrhea. The effacement of microvillar structure may lead to loss of absorptive surfaces and enzymes crucial to proper absorption. However, this is not the only factor involved since the brush border disruptions are very specifically localized to the site of bacterial attachment and do not affect large areas of the intestine. Additionally, diarrhea appears too quickly in volunteers to account for this mechanism (Donnenberg, et al., 1993). Pedestal formation following the degradation of microvilli may also add to the continued production of diarrhea, since recruitment of various cytoskeletal proteins to the site of bacterial attachment and effects on MLC may contribute to the disruption in tight junction integrity, which then leads to paracellular permeability and a decrease in transepithelial resistance. Tight junction disruption could thus contribute to diarrhea in several ways. The onset of diarrhea occurs as quickly as 4 hours post-ingestion of EPEC in human volunteers (Donnenberg, et al., 1993). It is likely that a rapid effect on ion secretion, perhaps through Cf or HC0 ", triggers the initial onset of diarrhea, and the following 3  cytoskeletal changes contribute to the prolonged disease. 1.6 Objectives The formation of pedestals is strongly correlated with EPEC-induced diarrhea, and it is therefore important to understand their structure, function, and role in disease. My research has focused on defining the cytoskeletal components and rearrangements that lead to pedestal formation. Because Tir is essential to this process, much of my research has studied how Tir mediates these effects on the host cell actin cytoskeleton. The initial observation that EPEC induced actin polymerization during pedestal formation led to the hypothesis that pedestals are comprised of several actin-associated proteins that could function as structural or regulatory components. Furthermore, the  30  observation that Tir was located at the tip of the pedestal and was required for pedestal formation led to the hypothesis that Tir functions as the direct linkage between extracellular bacteria and the host cell actin cytoskeleton through one or more actinassociated proteins. Two major experimental approaches were used to test these hypotheses: (1) identification and characterization of many host cytoskeletal proteins that comprise the pedestal and defining the hierarchy of recruitment; and (2) identification of cytoskeletal components that interact directly with Tir. To test the first hypothesis, over 30 structural and signaling proteins were examined, including their location in the pedestal (tip or length), their dependence on Tir insertion into the plasma membrane, and their dependence on Tir tyrosine phosphorylation for recruitment (Chapter 3; Goosney et al. 2001). A comparative screening was done with EHEC to define the similarities and differences in the cytoskeletal and signaling events leading to pedestal formation by EHEC and EPEC. The identification of two key components of actin polymerization in the pedestal, N-WASP and the Arp2/3 complex, led to the hypothesis that they were functionally important in mediating pedestal formation, especially given their central role in normal cell actin based polymerization processes. To test this hypothesis, mutational analysis of N-WASP was performed. It was observed that N-WASP was required to initiate pedestal formation through recruitment of the Arp2/3 complex, and together these proteins acted as distal effectors of pedestal formation (Chapter 4; Kalman et al. 1999). To test the hypothesis that Tir was directly binding a cytoskeletal protein, a HeLa cell cDNA library was screened for Tir-interacting proteins using a yeast two hybrid system. Several proteins were identified using this assay, but after secondary biochemical and cell biological analysis, these turned out to be false positives (Chapter 5). A more biochemical approach was then employed which used a Tir affinity column to identify Tir-interacting proteins that were present in HeLa cell extracts. This experimental approach resulted in the identification of a-actinin as a direct binding  31  partner for Tir (Chapter 6; Goosney et al. 2000). Collectively, these findings have significantly advanced our knowledge about the mechanisms and components involved in pedestal formation. They suggest that the EPEC pedestal is a complex structure with Tir playing a central role in its formation, mediating several effects on the host cell cytoskeleton through both its N and C termini. They also suggest that these pathogens subvert a normal cellular process (actin polymerization) and use normal cellular components for pedestal formation, which are coordinated through Tir. A discussion on the role of Tir and the pedestal in EPEC pathogenicity and as a cell biological tool is presented in Chapter 7.  32  Chapter 2: Materials and Methods  2.1 Cell lines and bacterial strains used in studies HeLa cells, a human cervical epithelial cell line (CCL-2; American Type Culture Collection), were used for all studies except in addressing the role of CD44 in Chapter 5. They were maintained between passages 5 and 25 in Dulbecco's minimal Eagle medium (DMEM, Gibco, Grand Island, NY) supplemented with 10% fetal calf serum (FCS, Gibco) at 37°C in a humidified atmosphere with 5 % C02. In Chapter 5, CD44-/- Swiss 373 fibroblasts were used (courtesy of Dr. C.J. Eaves, Terry Fox Laboratory, University of British Columbia) to address the role of CD44 in EPEC pedestal formation. These were grown in D M E M with 10% calf serum at 37°C in a humidified atmosphere with 5 % C02. The EPEC strains used in all studies were E2348/69 (Donnenberg, et al., 1990) CVD206 (E2348/69 with a eaeA deletion) (Donnenberg and Kaper, 1991), Atir (E2348/69 with a tir deletion) (Kenny, et al., 1997), Atir complemented with pACYCftVY474F (Goosney et al., 2000), and cfm 14-1-1(1) (cfm::TnPhoA) (Donnenberg, et al., 1990). The E H E C strain used was 86-24 (serotype 0157:H7) (Griffin, et al., 1988). A l l EPEC and E H E C strains were grown in Lurani-Bertani (LB) broth (Difco Laboratories, Detroit, MI) at 37°C in overnight cultures without shaking. 2.2 Antibodies used A l l primary antibodies used in immunofluorescence, western immunoblotting and enzyme-linked immunosorbent assays are listed in Table 2.1. To prevent non-specific binding of rabbit polyclonal antisera to EPEC and EHEC, all rabbit polyclonal antisera were pre-incubated with paraformaldehyde-fixed EPEC and E H E C (PFA; 2.5% in phosphate-buffered saline pH 7.4). To prepare PFA-fixed bacteria, 1 ml of a standing overnight culture of EPEC or E H E C was centrifuged at 10,000xg to pellet bacteria. Pellets were resuspended in 500ul PFA and incubated for 15 min, room temperature.  33  The antibodies were incubated with the PFA-fixed bacteria for 20 min at room temperature then centrifuged at 10,000xg for 5 min to remove bacteria before they were added to coverslips. 2.3 SDS-PAGE and western immunoblot analysis Samples were prepared for electrophoresis by resuspending in Laemmli sample buffer to a final concentration of lOOmM Tris-HCl, pH6.8, 0.1M P-mercaptoethanol, 2% SDS and 10% glycerol and boiling for 5 min (Laemmli, 1970). Samples were loaded onto polyacrylamide gels and separated using a Bio-Rad mini-gel apparatus. Proteins were transferred to nitrocelluose (AB-S 83, Schleicher and Schuell, Keene, NH) and blocked with 5% skim milk (Carnation) in Tris-buffered saline with Tween-20 (TBS-T; 150 mM NaCl and 20 mM Tris-HCl, 0.1% Tween-20 pH 7.5) for 2 hours room temperature or 4°C overnight. The primary antibody (see Table 2.1 for dilutions) was diluted in TBS-T with 1% skim milk and incubated for one hour, room temperature. Blots were washed three times for 5 min with TBS-T prior to incubation with the horseradish peroxidase (HRP)-conjugated secondary antibody (anti-rat, anti-mouse, or anti-rabbit depending on the species of the primary antibody; Jackson Laboratories) for one hour. Blots were washed three times for 5 min and developed using enhanced chemiluminescence (ECL; Amersham). Controls of primary or secondary antibody alone were performed to determine the level of background staining.  34  Table 2.1: Antibodies used during studies Dilution in IF  Source  Mouse Mouse Mouse Rabbit  Dilution in Western/ELISA 1:200 1:100 1:100 N/A  1:100 N/A N/A 1:500  Finlay Finlay Finlay Finlay  Mouse Mouse Rat Rabbit mouse mouse Mouse Rabbit Rabbit Rabbit  1:2000 N/A 1:1000 N/A 1:1000 N/A N/A N/A N/A N/A  1:200 1:100 1:100 1:100 1:100 1:100 1:50 1:100 1:50 1:200  Anti-She Anti-paxillin Anti-gelsolin Antitropomyosin Anti-Arp3  Rabbit mouse mouse mouse  N/A N/A 1:2000 N/A  1:100 1:200 1:200 1:100  Sigma Transduction Labs Transduction Labs (TL) Cytoskeleton Inc. Sigma TL Sigma Santa Cruz Santa Cruz Upstate BiotechnologyUBI UBI TL Sigma Sigma  Rabbit  N/A  1:100  Anti-p34  Rabbit  N/A  1:50  Anti-N-WASP  Rabbit  N/A  1:300  Anti-HA 3F10 Anti-Myc 9E10 Anti- Flag Anti-T7 Anti-p85 Anti-pp60src Anti-cortactin Anti-Shp Anti-talin Anti-profilin  Mouse Mouse  N/A N/A  1:500 1:200  Mouse Mouse Rabbit Mouse Mouse Mouse Mouse Rabbit  N/A 1:2000 1:1000 N/A N/A N/A 1:1000 1:500  1:200 1:100 1:100 1:100 1:100 1:100 1:100 1:100  Anti-VASP Anti-vinculin Anti-LPP  Mouse Mouse Rabbit  1:2000 N/A 1:1000  1:200 1:100 1:100  Anti-zyxin  Mouse  N/A  1:100  Antibody  Species  Anti-Tir 2A8 Anti-Tir 2A5 Anti-Tir 2A2 Anti-Tir polyclonal Anti-a-actinin Anti-calpactin Anti-CD44 Anti-cofilin Anti-ezrin Anti-FAK Anti-pl30cas Anti-Crkll Anti-Grb2 Anti-Nck  Lab Lab Lab Lab  Edith Gouin, Institut Pasteur Matt Welch, U. Berkeley Jeffery Peterson, Harvard Boehringer Sigma Sigma Novagen TL UBI UBI TL Sigma Roy Golsteyn, Institut Curie TL Sigma Roy Golsteyn Institut Curie Jurghen Wehland, Braunschweig,Germany  35  2.4 Protein purification BL21 bacteria containing the pET28a vector (a His-tagging vector; Novagen) with TirCesT, Tir truncations (residues 1-200, 1-391, 202-391, 202-550, 391-550), CesT, and int282 (see Table 2.2) were grown overnight at 37°C in LB supplemented with kanamycin (30 ug/ml), shaking. The following day, the culture was subcultured to 1:50 in LB kanamycin and grown to an OD  600  of 0.6-0.7. 1 mM isopropylthio-B-D-  galactosidase (IPTG; Gibco BRL; final concentration) was added to the culture to induce protein expression, and the bacteria were grown for another 3h. The bacteria were pelleted by centrifugation for 20 min at 3,000xg and frozen at -70°C in sonication buffer (lOOmM NaPhos, 100 mM NaCl, pH 8.0). Following thawing of the pellet, the bacteria were sonicated, the insoluble pellet spun down at 12,000xg and the supernatant added to preequilibrated Ni-NTA agarose (Qiagen) for 60 min, 4°C stirring. The slurry was then washed with wash buffer (100 mM NaPhos, 100 mM NaCl, pH 7.4) to get rid of nonspecific binding proteins. His-tagged proteins were eluted with an increasing imidazole (Sigma, St. Louis, MO) gradient (0-250mM imidazole in wash buffer).  36  Table 2.2: Recombinant constructs used (residues) Construct Flag-WASP (1-502) HA-N-WASP (1-505) Flag-WASP-AC (A443-502) Flag-C (443-502)  Reference (Kalman, et al., 1999) (Kalman, et al., 1999) (Kalman, et al., 1999) (Kalman, et al, 1999)  HA-N-WASP-AAcidic (A485-505) HA-NWASP-ACofilin (A473-476) HA-N-WASP-AAcidic/ACofilin(A473476, A485-505) Flag- WASP-AGBD (A235-268) Flag-W ASP-GBD(207-313) Myc-PAK-GBD(61-150) Flag-WASP-AC/AGBD (A235-268, A443502) Flag-WASP-AN (A10-150) Flag-WASP-AWHl (A47-149) Flag-WH 1(47-149) Flag-WASP-APPP (A319-430) Flag-PPP (319-430) Flag-WASP-AWH2 (A428-446) Flag-WH2 (428-446) GFP GFP-a-actinin (1-890) GFP-actin binding domain (1-253) GFP-spectrin-like repeats (235-713) GFP-calcium-binding domain (704-890) GFP-profilin His-Tir/CesT (1-550) His-Tir (1-200) His-Tir (1-392) His-Tir (201-392) His-Tir (201-550) His-Tir (392-550) His-CesT His-T7-intimin282  (Kalman, et al., 1999) (Kalman, et al., 1999) (Kalman, et al., 1999) (Kalman, et al., 1999) (Kalman, et al., 1999) (Kalman, et al., 1999) (Kalman, et al., 1999) (Kalman, et al., 1999) (Kalman, et al., 1999) (Kalman, et al., 1999) (Kalman, et al., 1999) (Kalman, et al., 1999) (Kalman, et al., 1999) (Kalman, et al., 1999) Clonetech S. Gruenheid and D. Goosney, unpublished S. Gruenheid and D. Goosney, unpublished S. Gruenheid and D. Goosney, unpublished S. Gruenheid and D. Goosney, unpublished Heinzen. R. unpublished (U. Wyoming) (Goosney, et al., 2000) (Goosney, et al., 2000) (Goosney, et al., 2000) (Goosney, et al., 2000) (Goosney, et al., 2000) (Goosney, et al., 2000) (Goosney, et al., 2000) (Luo, et al., 2000)  37  2.5 Affinity chromatography Full length Tir-CesT was purified as described above and was maintained on the Ni-NTA agarose as described above instead of being eluted with imidazole. HeLa cells were lysed in 1% Triton X-100 in wash buffer at 4°C for 20 min, the insoluble pellet spun out at 12,000xg, and the Triton soluble fraction added to the Ni-NTA agarose with His-Tir bound for 60 min, 4°C, shaking. The slurry was again poured into a 1.5 ml column and left to settle prior to washing with wash buffer at a flow rate of 2.0 ml/min. Tir and its interacting proteins were eluted with an increasing imidazole gradient (0250mM imidazole in wash buffer) at a flow rate of 1.0 ml/min, collecting 1 ml fractions. Each fraction was analysed by SDS-PAGE followed by Coomassie blue staining (Sigma), and further studied by western blot analysis probing for Tir and a-actinin to identify the fractions containing each. The same procedure was employed with the Ni-NTA agarose alone and with His-T7Tir (392-550) with the HeLa lysates and analysed for a-actinin elution. 2.6 Immunofluorescence HeLa cells were seeded onto 11 mm coverslips at a density of 2xl0 cells/ml to 4  achieve a final confluency of 60-70%. The following day, they were infected with a standing overnight EPEC or EHEC culture (multiplicity of infection 20:1) for 3 hours (for EPEC) or 4 hours (for EHEC) in DMEM at 37°C and 5% C 0 . The DMEM was 2  changed at 3 (or 4) hours to remove non-adherent bacteria and the remaining adherent EPEC and EHEC allowed to infect for 2 more hours. Following infection, the coverslips were washed three times with PBS and fixed with 2.5% paraformaldehyde for 15 min, 37°C. The coverslips were washed extensively with PBS after fixing and the cells permeabilized with 0.5% Triton X-100 in PBS. Following permeabilization, the cells were washed with 0.1% Triton X-100 in PBS, blocked with 10% normal goat serum (NGS; Gibco BRL) in PBS, then probed with one or a combination of primary antibodies listed above. Following the primary antibody, the cells were washed extensively with  38  0.1% Triton X-100 in PBS and probed with Alexa dye-conjugated secondary antibodies (either anti-mouse, anti-rat, or anti-rabbit depending on the species the primary antibody was raised in) and Alexa-conjugated phalloidin to detect actin (Molecular Probes). HeLa and bacterial DNA were visualized using diamidino-2-phenylindole (DAPI; lu,g/ml; Sigma). For experiments using antibodies that only recognize denatured epitopes (antip34, N-WASP, cofilin), cells were fixed with PFA as described above, treated with 100% methanol (stored at -20°C) for 90 seconds, and blocked in 10% NGS. For all chapters except Chapter 4, the coverslips were mounted in Mowiol (Aldrich) and viewed at 350nm, 488nm and 594 nm on a Zeiss Axiophot epifluorescence microscope. For Chapter 4, all images were acquired with a scientific-grade cooled charge-coupled device on a multiwave wide-field three-dimensional microscopy system. Immunofluorescent samples were imaged in successive 0.25 um focal planes through the samples and out-offocus light was removed with a constrained iterative deconvolution algorithm. All images represent single optical sections of immunofluorescence data. All experiments were repeated a minimum of three times. 2.7 Far Western analysis Purified a-actinin (Sigma) and int282 were prepared in Laemelli sample buffer, boiled for 5 min and loaded onto an 8% SDS-PAGE at a concentration of 5 u,g/well. The gel was transferred to nitrocellulose then washed with 50 mM Tris, pH 8.0 for 15 min. Following washing, the blot was treated with 20% isopropanol in 50 mM Tris, pH 8.0 for 60 min then washed again in 50mM Tris, pH 8.0 for 60 min. The proteins were then denatured with 6M guanidine-HCl in 50 mM Tris, pH 8.0 for 60 min then extensively washed with Tris pH 8.0 for greater than 18 hours with numerous changes of the wash buffer. Once the proteins slowly renatured, the nitrocellulose was blocked in 5% skim milk powder in 0.1% Tween-20 in PBS for 60 min, then overlaid with full-length TirCesT or the Tir truncations (1-391 and 392-550) in PBS for 2h. The nitrocellulose was then washed for 30 min in 0.1%Tween-20 in PBS and probed for Tir using the Tir  39  monoclonal antibody 2A8 (or 2A5 which was raised against the N-terminus and used for the 1-391 truncation). The nitrocellulose was washed again in PBS and probed with a HRP-conjugated anti-mouse secondary. Tir was detected using chemiluminescence. 2.8 Enzyme-linked immunosorbent assays Purified proteins (int282, BSA, a-actinin, Tir, CesT, profilin) in PBS were bound to Immulon-2 96 well plates (Dynatec) for lh, 37°C. The wells were washed with 0.1%Tween-20 in PBS and blocked in 5% skim milk powder (Carnation) in 0.1% Tween20 in PBS for lh. Purified Tir, Tir truncations (as described above), profilin (Cytoskeleton Inc.) or a-actinin (Sigma) were incubated in the wells for 1 h in 0.1% Tween-20 and PBS at 37°C. The wells were washed again and the overlaying proteins detected with the appropriate antibody (described above) for 1 h, 37°C. After further washing, HRP-conjugated secondary antibodies were added to the wells for 30 min at 37°C, the wells washed, and the reaction detected colorimetrically using developing buffer (0.1M citrate, 0.2M NaPhos, o-phenylenediamine (OPD; Sigma), 60 ul H 0 (Fischer)). The absorbance at 490 nm was then read using a SpectraMax or TECAN 2  2  plate reader. 2.9 Immunoprecipitations Two 100 cm plates were seeded with HeLa cells at 2 x 10 cells/ml the day prior 5  to infection. One plate was infected with EPEC for 3 hours, the other plate remained uninfected. The cells were lysed at 4°C with 1% Triton-X 100 in PBS in the presence of inhibitors (phenylmethylsufonyl fluoride (PMSF), sodium fluoride (NaF), sodium orthovanadate (Na V0 ) all from Sigma). The cells were centrifuged at 10,000xg for 3  4  5min to remove the Triton insoluble fraction. A pre-immunoprecipitation sample was set aside and resuspended in SDS-sample buffer. For CD44 immunoprecipitation, 5ug rat anti-CD44 (Transduction Labs) was added to both uninfected and infected lysates and incubated at 4°C for 1 hour, rotating. After 1 hour, 30 ul of a 50% protein G sepharose bead slurry (Pharmacia Biotech AB) was added to both uninfected and infected lysates  40  and incubated for an hour at 4°C, rotating. Beads (with the immunoprecipitate) were then pelleted by gentle centrifugation (l,000xg) for 20 seconds. The supernatant (post-IP) was removed and resuspended in SDS-sample buffer. The beads were washed three times in 0.1% Triton X-100 in PBS. The immunoprecipitates were resuspended in SDS-sample buffer and all samples boiled for 5 min prior to loading onto SDS-PAGE. Samples were run on an acrylamide gel, transferred to nitrocellulose, blocked in 5% skim milk in PBS 0.1% Tween-20, and probed for CD44 and Tir. HRP-conjugated secondary antibodies followed by chemiluminescence were used to develop the blot. For Tir immunoprecipitations, a cross-linker, disuccinyl suberate (DSS), was used prior to lysing the cells. DSS is an amine-reactive, thiol-cleavable cross-linker (Pierce). To extract proteins that are normally Triton-insoluble (as are most structural cytoskeletal proteins) RIPA buffer (0.2% SDS, 1% Triton X 100, 0.1% deoxycholate, 50mM Tris pH 7.4, 2mM EDTA, 10 mM Na V0 , 10 mM PMSF) was used. Given the "harsh" 3  4  detergent nature of this buffer, many protein-protein interactions were disrupted. Therefore, a cross-linker was used in these conditions to maintain Tir interactions with its binding partner(s). HeLa cells (uninfected or EPEC infected) were incubated with O.lmM DSS in PBS for 30 mins at room temperature. Saturated Tris was added to quench any unreacted amine groups for 10 min after cross-linking. Cells were then lysed in RIPA buffer at 4°C and the detergent-insoluble fraction removed by centrifugation at 10,000xg for 5 min. Lysates were diluted 1:4 with PBS (to dilute out the strong detergent nature of RIPA) and 30ul of mouse anti-Tir (2A8) were incubated with the lysates for 1 hour, 4°C, rotating. Protein G beads were added and the remaining steps performed as described above. 2.10 Construction of GFP-cc-actinin Human a-actinin 1 (non-muscle isoform) cDNA (from Alan Beggs, Harvard, Boston, MA) was amplified by PCR (Youssoufian, et al., 1990). Full-length a-actinin was cloned from nucleotide 129-2800, actin binding domain from 129-889, spectrin-like  41  repeats 834-2267, and the calcium binding domain from 2240-2800. Full-length aactinin and the three truncations were cloned into a pEGFP vector (Clonetech) at Nhel and Hindlll restriction sites. Clones were sequenced at the Nucleic Acid and Protein Sequencing (NAPS) facility at UBC. 2.11 Transfection of HeLa cells HeLa cells were seeded onto coverslips at a density of 2xl0 approximately 8 4  hours prior to transfection. Plasmid expression vectors containing the constructs listed in Table 2.2 were transfected into HeLa cells using Fugene-6 (Boehringer). Briefly, 2ul of Fugene-6 was added to lOOul serum-free DMEM for 5 min, room temperature. This mixture was added to 0.5ug of plasmid DNA and incubated at room temperature for 30 min. The DNA/Fugene/DMEM mixture was then added drop-wise to the HeLa cells and incubated at 37°C and 5%C0 . For the N-WASP studies, cells were transfected for 3 2  days prior to infection with EPEC. For the a-actinin and profilin studies, cells were transfected for 18-24 hours prior to infection with EPEC. The transfection efficiency for all experiments was 30-40%. 2.12 Yeast two hybrid system The basic protocol used was adapted from Current Protocols in Molecular Biology, Unit 20 (1996). A HeLa cell cDNA library was obtained from the Brent laboratory in the vector pJG4-5 (containing the activating domain for LexA transcription; see Chapter 5 for background on the interaction) (Gyuris, et al., 1993). This library was the "prey" in the interaction trap. The library was transformed into DH10B E. coli, grown on LB artipicillin (lOOug/ml) to select for successful transformants, and amplified to obtain 6 x 10 colonies with which to conduct the interaction trap. All colonies were 6  scraped into LB broth using a sterile, clean glass slide and centrifuged at 4,000 g for 15 min. The pellet was washed three times in PBS and the HeLa cDNA was purified using a Qiagen maxi-prep kit.  42  Saccharomyces cerivisiae strain EGY48 containing the reporter plasmid pSH1834 (S.Hanes, Wadsworth University, Albany, NY) was transformed with a pEG202 vector (E. Golemis, Fox Chase Cance r Center, Philadelphia, PA) containing Tir (the "bait"). S. cerivisiae was grown in non-inducing conditions (glucose, complete minimal media (CM)/-His/ -Ura). Yeast cells were maintained at 30°C for 2-3 days to select for colonies containing transformants. These transformants were tested to ensure they didn't appreciably activate transcription on their own. pEG202Tir (test), pSH17-4 (positive control for activation; it encodes Lex A fused to the activation domain of Gal4 and strongly activates transcription; S. Hanes, Wadsworth University, Albany, NY), and pPvFHMl+pSH 18-34 (negative control for transcription; plasmid encodes lexA fused to the N-terminus of the Drosophila protein bicoid and has no ability to activate transcription; S. Hanes Wadsworth University, Albany, NY and R. Finley, Wayne State University, Detroit, MI) were transformed into yeast and plated on CM/ glucose/ -Ura/ His dropout plates. These plates were incubated for 2 days at 30°C to select for yeast that contain both plasmids in each test condition. Five colonies from each transformation (test, positive control, negative control) were picked and streaked on a glucose/CMAUra/His master dropout plate and incubated overnight at 30°C. From this master plate, the colonies were streaked onto plates containing X-gal, incubated at 30°C, and examined for colour development over 2 days (strongly activating fusions turn blue over 12 -24 hours). Upon confirming that the pEG202-Tir fusion was not appreciably activating transcription, the yeast were transformed with the pJG4-5 HeLa cDNA library. A 20 ml culture of EGY48 yeast containing the pEG202Tir vector was grown in Glucose/CM Ura-His liquid media overnight at 30°C. The following morning, the culture was diluted into 300 ml Glucose/CM-Ura-His liquid medium and incubated at 30°C until it reached an O D  600  of 0.5. The yeast were centrifuged at l,500xg for 5 min, harvested, resuspended  in 30 ml sterile water, and transferred to a 50 ml conical tube. The cells were centrifuged again at l,500xg for 5 min, the supernatant removed, and the cells  43  resuspended in 1.5ml Tris-EDTA (TE) buffer/0.1M lithium acetate. One pig of library DNA and 50ug sheared salmon sperm carrier DNA (Sigma) was added to 30 microcentrifuge tubes. Fifty \il of the resuspended yeast solution was then added, followed by 300 ul of sterile 40% polyethylene glycol (PEG) 4000/0.1M lithium acetate/TE buffer, pH 7.5. The mixture was mixed and incubated at 30°C for 30 min. Forty u,l of DMSO was added to each tube and inverted to mix. The cells were heat shocked for 10 min at 42°C. The contents of each tube were plated on a large (24x24 cm) Glucose/CM/-Ura/-His/-Trp plate and incubated at 30°C. For two of the tubes, dilutions were plated to determine transformation efficiency. To collect all primary transformed cells, the large plates were cooled for several hours, then scraped gently with a sterile glass microscope slide. The cells from the 30 plates were pooled into two 50 ml conical tubes, washed with TE buffer, and centrifuged for 5 min at l,500xg. These yeast cells were then replated on Galactose/Raff/CM/-Ura/-His/-Trp media and incubated for 23 days at 30°C. Colonies were picked and plated on a new master plate containing Glu/CM/-Ura/-His/-Trp on day 2 and day 3, then incubated at 30°C overnight. Each colony was restreaked onto the following media: (1) Glu/Xgal/CM-Ura-His-Trp; (2) Gal/Raff/Xgal/CM-Ura-His-Trp; (3) Glu/CM/-Ura-His-Trp-Leu; (4) Gal/Raff/CM-UraHis-Trp-Leu. Colonies that grew on Gal/Raff plates but not Glu/CM-Leu plates and colonies that turned blue on Gal/Raff plates but not Glu/X-gal plates were considered to be posititve interactions. For positive colonies, plasmid DNA was isolated by yeast miniprep, resuspended in TE buffer, and KC8 bacteria were transformed by electroporation using 1 u4 of isolated DNA. The KC8 bacteria were incubated overnight on LB/ampicillin plates at 37°C. The following day, a bacterial mini-prep was performed to isolate the library-containing plasmids. These plasmids were retested for specificity by retransforming EGY48 yeast containing: (1) pSH 18-34 +pEG202Tir and (2) pSH 18-34 and pRFHM-1 (negative control). Each transformant was replated on Glu/CM-Ura-HisTrp masterplates, incubated for 2-3 days until colonies appeared, and streaked onto the  44  same series of test plates as above (1-4). True positive cDNA were blue on the Gal plates and not on the Glu plates and they grew on the GalALeu plates but not the Glu/Leu plates. The cDNAs that were positive under these conditions were then isolated and sent for sequencing (NAPS Unit, University of British Columbia). The resultant DNA sequences were analysed using the National Center for Biotechnology Information (NCBI) BLAST program to identify potential interacting partners for Tir.  45  Chapter 3 Recruitment of cytoskeletal and signaling proteins to enteropathogenic and enterohemorrhagic E. coli pedestals  3.1 Summary Pedestal formation is associated with EPEC and EHEC induced diarrhea, yet very little is known about the composition and organization. The composition of the EPEC and EHEC pedestals was analysed by examining numerous cytoskeletal, signaling, and adaptor proteins. In EPEC, pedestal formation requires Tir tyrosine 474 phosphorylation. In EHEC Tir is not tyrosine phosphorylated, yet the pedestals appear similar. Of the 30 proteins examined, only two, calpactin and CD44, were recruited to the site of bacterial attachment independently of Tir. Several others, including ezrin, talin, gelsolin, a-actinin and tropomyosin required Tir in the host membrane for recruitment but were independent of Tir tyrosine 474 phosphorylation. The remaining proteins were recruited to the pedestal in a manner dependent on Tir tyrosine phosphorylation or were not recruited in at all. Differences were also found between the EPEC and EHEC pedestals: the adaptor proteins Nek, Grb2, and Crkll were recruited to the EPEC pedestal but were absent in the EHEC pedestal. These results demonstrate that EPEC pedestals are comprised of numerous regulatory and strucutral components that are recruited in a Tir-independent, Tir phosphotyrosine-independent, or Tir phosphotyrosine-dependent manner. Furthermore, EPEC and EHEC recruit similar cytoskeletal proteins, but there are also significant differences in pedestal composition.  46  3.2 Introduction The initial observation that EPEC and EHEC induced actin polymerization during pedestal formation led to the hypothesis that pedestals are comprised of several actin-associated proteins that could function as structural or regulatory components. This study was performed to identify components of both structures, serving as a basis for future functional analysis of the pedestal. Tir is critical in mediating the cytoskeletal rearrangements seen during infection of both pathogens. Upon Tir insertion in the host cell plasma membrane, several cytoskeletal proteins are recruited to the site of EPEC attachment. These include a-actinin, ezrin, talin, fimbrin, tropomyosin, and villin. Little, however, is known about the cytoskeletal composition of the EHEC pedestal. a-Actinin and actin are the only proteins to date shown to be specifically recruited to the EHEC pedestal (Ismaili, et al., 1995). In this study, 30 cytoskeletal, signaling, and adaptor proteins were screened for recruitment to EPEC and EHEC pedestals and the specific role of Tir and its tyrosine phosphorylation was also examined. A comparision of the EPEC and EHEC pedestal composition illustrates that although the pedestals are similar, there are also significant differences. Results 3.3 E P E C recruits several cytoskeletal and signaling proteins to the pedestal HeLa cells infected with EPEC formed elongated pedestals ranging between 1 and 4 um in length. They were prepared for immunofluorescence, and probed for cytoskeletal and signaling proteins. Of the proteins tested, the following were found to be in the pedestal: Crkll, Grb2, ADF/cofilin, LPP, pl30cas, She, gelsolin, CD44, calpactin, zyxin, Nek, VASP, N-WASP, cortactin, vinculin, and Arp3 (Fig. 3.1a-r). Other host proteins that were tested but not recruited to the EPEC pedestal include (31 and a5 integrin, src family kinases, FAK, paxillin, and Shpl (Fig. 3.1s and data not shown). Of the proteins recruited only Nek, N-WASP, Arp2/3 and VASP colocalized with Tir at the  47  tip of the pedestal. Although previously identified as components of the pedestal, aactinin and talin were also found to go to the tip of the pedestal as well as the length in this study (Fig.3.1 p, q). 3.4 CD44 and calpactin are recruited independently of Tir delivery HeLa cells were infected with EPEC Atir, which delivers Esps to the host cell but is incapable of forming pedestals due to the absence of Tir. This mutant can attach loosely to the host cell through its bundle-forming pilus (BFP). Of all the proteins tested, only CD44 and calpactin were localized to the site of bacterial adherence, indicating that these proteins are recruited independently of Tir (Fig.2a-f). All other proteins had a staining pattern similar to uninfected cells (data not shown). 3.5 Gelsolin, tropomyosin, ezrin, talin and a-actinin are recruited to EPEC independently of Tir tyrosine phosphorylation HeLa cells were infected with EPEC Atir/tirY474F, an EPEC strain capable of delivering Tir to the host, but lacking the tyrosine residue that is phosphorylated in the host cell. Phosphorylation of EPEC Tir tyrosine 474 is critical for pedestal formation. Of the proteins tested, gelsolin, tropomyosin, ezrin, talin, and a-actinin were recruited to the site of EPEC adherence without Tir tyrosine phosphorylation (Fig. 3a-e). The other proteins had a staining similar to uninfected cells, as seen with Crkll staining (Fig. 3f).  3.6 E H E C does not recruit the adaptor proteins Nek, Crkll or Grb2 to its pedestal HeLa cells were infected with EHEC and pedestals examined for recruitment of cytoskeletal and signaling proteins. Of the proteins tested, Nek, Crkll and Grb2 differed, being recruited specifically to EPEC, but not EHEC, pedestals (Fig. 4). Nek was of particular interest as it was recruited specifically to the tip of the EPEC pedestal, colocalizing with Tir whereas Grb2 and Crkll localized to the length of the pedestal. All other proteins were recruited similarly beneath both EPEC and EHEC, indicating the similar, but not identical, cytoskeletal composition of these pedestals.  48  Fig. 3.1: Cytoskeletal proteins recruited to the EPEC pedestal. HeLa cells were infected with EPEC for 5 hours. Upon Tir translocation to the cells (A-K; top panel), numerous cytoskeletal proteins were recruited to the pedestal (A-I; middle panel), including Crkll (A), Grb2 (B), cofilin (C), LPP (D), pl30cas (E), She (F), gelsolin (G), CD44 (H), calpactin (I), zyxin (J), vinculin (K), cortactin (L), VASP (M), Nek (N), N-WASP (O), talin (P), and a-actinin (Q), and Arp3 (of the Arp2/3 complex; R). In contrast, paxillin (S) was not recruited to the pedestal. The bottom panels represent a merge of Tir (green) the cytoskeletal protein (red) and DAPI-stained EPEC (blue). Arrows indicate EPEC pedestals. Scale bars represent 5|im.  49  Fig.3.2: CD44 and calpactin are recruited to the site of EPEC adherence independently of Tir translocation. HeLa cells were infected with EPECAn'r for 5 hours. EPEC Atir microcolonies were labelled by DAPI (A, D). CD44 (B) and calpactin (E) were recruited beneath EPECAft'r in a honeycomb pattern. The bottom panels represent a merge of EPECAft'r (blue), CD44 (C; red) and calpactin (F; red). Arrows indicate sites of EPEC attachment to the HeLa cells. Scale bars represent 5um.  52  «  A  LJ F  c  #  A  53  Fig.3.3: Talin, tropomyosin, gelsolin, ezrin and a-actinin are recruited to the site of EPEC adherence independently of Tir tyrosine phosphorylation. HeLa cells were infected with EPECArir/hVY474F for 5 hours. Upon TirY474F translocation to the cells (A-E, top panel), talin (A, middle panel) tropomyosin (B, middle panel), gelsolin (C, middle panel), ezrin (D, middle panel), and a-actinin (E, middle panel) were recruited to the site of bacterial attachment. Other cytoskeletal proteins, like Crkll (F, middle panel) were not recruited. The bottom panels represent a merge of Tir Y474F (red) and the cytoskeletal protein (green). Arrows indicate sites of EPEC attachment to the HeLa cells. Scale bars represent 5|im.  54  Fig.3.4: EHEC does not recruit the adaptor proteins Crkll, Grb2 and Nek to the EHEC pedestal. HeLa cells were infected with EHEC for 6 hours. Upon translocation of Tir to the cell (top panel), Crkll (A), Grb2 (B), and Nek (C) were not rearranged as seen during EPEC infection (see Fig. 1). The bottom panels represent a merge of Tir (green), Crkll (A, red), Grb2 (B, red) or Nek (C, red). Scale bars represent 5um.  56  A  *  »  c  B  • *  *  t  '  -•  V  •A  "t  57  3.7 Discussion Since the initial survey of cytoskeletal proteins recruited to the site of EPEC attachment was performed (Finlay, et al., 1992), our knowledge of EPEC and the cytoskeleton has expanded significantly. The identification of Tir, a bacterial protein, as the receptor for EPEC intimate adherence has allowed us to dissect pedestal formation even further through genetic manipulation of Tir (Kenny, et al., 1997). Elongated pedestal formation is not only dependent on EPEC Tir, but dependent on its tyrosine phosphorylation by a currently unidentified kinase. Additionally, the discovery that EHEC Tir is not tyrosine phosphorylated warrants comparison of these two pedestals (Deibel, et al., 1998, DeVinney, et al., 1999). Pathogenic E. coli may be used as a model system to study signaling to the actin cytoskeleton across the plasma membrane in response to external stimuli. Indeed, there are many parallels between EPEC pedestal formation and the formation of focal adhesions. Focal adhesions are found at sites of eukaryotic cell attachment to the extracellular matrix (ECM). This attachment is mediated through a family of integral membrane proteins called integrins, which link the ECM to the cytoskeleton. Many of the cytoskeletal and signaling proteins that were examined in this study are also involved in focal adhesion formation (Table 3.1). Both oc5 and [31 integrins were screened in this study, but they were not found in the EPEC or EHEC pedestal. This was not unexpected as a previous report suggested that P1 integrins play no role in EPEC infection (Liu, et al., 1999). However, it is interesting that so many focal adhesion proteins were localized to the pedestal in the absence of pi integrins. This suggests that Tir might function much like an integrin (see Chapter 7 for a detailed comparison). Several other proteins found in focal adhesions are recruited to the EPEC pedestal, including zyxin, LPP, a-actinin, talin, vinculin, VASP and pl30cas. a-Actinin and talin interact directly with pi integrins and link the receptor directly to the actin  58  cytoskeleton. Both also function to bind and cross-link actin (Jockusch and Isenberg, 1981). Vinculin does not directly bind to integrins, but links actin filaments to integrins through other proteins, such as talin (Burridge and Mangeat, 1984). VASP targets profilin and F-actin to the site of focal adhesions (Reinhard, et al., 1995). To date, profilin has not been seen in the pedestal either by indirect immunofluorescence or by GFPprofilin transfections. Zyxin and its homologue, LPP, function in focal adhesions and cell-cell contacts (Petit, et al., 2000) (Reinhard, et al., 1995). Both shuttle between the cytoplasm and the nucleus and have transcriptional activation capacity (Nix and Beckerle, 1997) (Petit, et al., 2000). The primary difference between them in cultured epithelial cells is that zyxin colocalizes with stress fibers as well as contact sites, whereas LPP is found only in the sites of contact (Petit, et al., 2000). Both proteins bind and presumably target VASP to focal adhesions.  59  Table 3.1: Host proteins characterized in E. coli pedestals Protein  Found in EPEC pedestals  Actin associated proteins Arp2/3  •  a-actinin  •  Found in EHEC pedestals  Tir dependent  Tir PY dependent  Location in pedestal  Reference  •  •  Tip and length  X  X  Tip and length Tip and length Tip and length Tip  This study and Kalman et al 1999 Goosney et al. 2000 This study  X  Along length  Tip and length Along length  • X  calpactin  X  cofilin  •  cortactin  ezrin  X  X  V  gelsolin  •  X  lipoma preferred partner (LPP) N-WASP  •  •  •  •  •  paxillin pl30cas  X  X  X  X  •  •  talin  •  •  zyxin Membrane roteins  V  This study and Kalman et al 1999 This study This study  X  Along length  V  Tip  •  Along length  s  •  Along length  This study and Finlay et al. 1992 This study and Sanger et al. 1996 This study and Goosney et al. 2000 This study and Finlay et al. 1992 This study  •  vinculin  This study  X  •  VASP  This study and Cantarelli et al. 1999 Finlay et al. 1992; this study This study  X Tip and length Along length  V  tropomyosin  Tip  This study  0c5 integrin  X  X  X  X  X  This study  (}1 integrin  X  X  X  X  X  •  X  X  Tip  This study and Liu et al. 1999 This study  •  Tip and length  CD44 Adaptor proteins Crkll  X  This study  60  Grb2 Nek She Kinases/ phosphatases pp60src Shp FAK  • •  X X  V  •/  X X X  X X X  X X X  •  Along length Tip Along length  This study This study This study  X X X  X X X  This study This study This study  61  pl30cas is also a substrate for FAK and src, and can interact with the adaptor molecule Crk (Harte, et al., 1996) (Burnham, et al., 1996). It is surprising that both src family kinases and FAK were not localized to the pedestal during EPEC or EHEC infection. This could be simply due to problems detecting the proteins using the antibodies available, or could suggest that another signaling pathway is being used by the pathogens to initiate pedestal formation. Alternatively, the tyrosine phosphorylation event may be occurring only transiently, and therefore was missed using this immunofluorescence technique. Small GTP-binding proteins are essential for focal adhesion and stress fiber formation. Rho activation during adhesion results in activation of PI 3-kinase (Zhang, et al., 1993) (Murakami, et al., 1999) which plays a role in restructuring the focal adhesion. Several reports indicate that EPEC does not use GTPases during pedestal formation (BenAmi, et al., 1998; Ebel, et al., 1998). These studies were performed using inhibitors (such as compactin and Clostridium difficile ToxB) or dominant negative alleles to specifically inhibit the Rho family of GTPases. Surprisingly, pedestal formation was greatly enhanced in the presence of these inhibitors (Ben-Ami et al., 1998). This is an exciting discovery as most actin rearrangements in the cell are mediated in part by the Rho subfamily of GTPases. Several other proteins were recruited to the pedestal in addition to those normally found in focal adhesions - CD44, calpactin, ezrin, ADF/cofilin, gelsolin, She, Crkll, Grb2, N-WASP, the Arp2/3 complex, and tropomyosin - illustrating differences between the two structures. CD44 is a membrane receptor for hyaluronic acid, whereas calpactin (pi 1) acts together with annexin II at the plasma membrane to function in membrane fusion and host cell exocytosis. Both CD44 and the annexin II-pl 1 complex colocalize in lipid rafts (Oliferenko, et al., 1999) which act to concentrate signaling proteins in various cellular functions, including signal transduction and protein sorting. Although the role of  62  CD44 and calpactin in EPEC infection is not known, it is tempting to speculate that lipid rafts may be involved in the initial signaling processes leading to pedestal formation. The recruitment of N-WASP and the Arp2/3 complex was of particular interest as both are important regulators of actin nucleation at the plasma membrane and have been implicated in actin-based motility of other pathogens, including vaccinia and Shigella (see Chapter 7 for a discussion on these pathogens). The role of these proteins in EPEC pedestal formation will be the focus of Chapter 4. Once Tir is delivered to the host cell, it is tyrosine phosphorylated in EPEC, but not in EHEC. We addressed the role of EPEC Tir tyrosine phosphorylation by characterizing cytoskeletal proteins recruited to the site of bacterial attachment independently of this phosphorylation event. In this study, we show that ezrin, talin, gelsolin, a-actinin and tropomyosin are also recruited in the absence of tyrosine phosphorylation. a-Actinin, talin and ezrin are all proteins that link the actin cytoskeleton to the plasma membrane. As mentioned previously, talin and a-actinin link to integrins and function in focal adhesions. Ezrin links the cytoskeleton to the plasma membrane in structures such as microvilli and microspikes. These three proteins may be involved in mediating a stable link from EPEC to the host cell cytoskeleton (see Chapter 6 for the role of a-actinin). Gelsolin is a Ca -sensitive F-actin severing protein that also caps barbed ends of 2+  actin filaments and functions to increase free pointed ends (Southwick, 2000). Gelsolin may be recruited early to the site of EPEC attachment to provide EPEC with a source of free end filaments from which to build the pedestal. Tropomyosin is an actin-binding protein that can be targeted to sites of active actin rearrangements by gelsolin (Koepf and Burtnick, 1992). The wide range of proteins recruited either dependently or independently of Tir tyrosine phosphorylation suggests that Tir may have more than one function in the host cell - recruiting proteins that serve to stabilize Tir interactions with  63  the cytoskeleton and proteins that can provide the actin polymerization machinery for pedestal formation. Given the differences in recruitment between phosphorylated and unphosphorylated Tir, we also investigated the composition of the EHEC pedestal. EHEC triggers an elongated pedestal similar to EPEC, but without Tir tyrosine phosphorylation. All proteins were recruited and localized to the tip or the length of the pedestal in the same manner as EPEC, with the exception of the adaptor proteins Nek, Grb2 and Crkll. However, the adaptor protein She was recruited to both EPEC and EHEC pedestals in the same pattern. Grb2 and Crkll were recruited along the tip and length of the EPEC pedestal but were not in the EHEC pedestal. Both are comprised of two SH3 and one SH2 domain which allows them to form multimeric complexes involved in signaling and cytoskeletal rearrangements. Grb2 was recently identified as an activator of the N-WASP /Arp2/3 cascade (Carlier, et al., 2000). Nek was of particular interest as it was very specifically recruited to the EPEC pedestal at the tip, the dynamic point of actin polymerization. Nek, like Grb2 and Crkll, is comprised of two SH3 domains and an SH2 domain. It interacts with several proteins involved in actin polymeriation, including N-WASP and WIP. It is tempting to speculate that EPEC may use Nek to recruit N-WASP and Arp2/3 to the pedestal. This would not be unprecedented, as another pathogen, vaccinia, uses Nek to recruit N-WASP and the Arp2/3 complex to its actin tail (Frischknecht, et al., 1999; see Chapter 7 for details on vaccinia-induced actin rearrangements). This is the first reported difference between EPEC and EHEC pedestals, and indicates that the role of tyrosine phosphorylation may be related to recruitment of adaptor proteins to the site of bacterial adherence. EHEC may build a slightly different pedestal independent of known host adaptors perhaps by delivering its own bacterial adaptor to the host cell. Alternatively, one or more adaptors may interact with the  64  tyrosine phosphorylated residue of EPEC to initiate pedestal formation, whereas this interaction is bypassed in EHEC pedestals. The results presented here provide significant insight into how the EPEC and EHEC pedestals are formed during infection. CD44 and calpactin were recruited independently of Tir in the host cell, suggesting that they are recruited before pedestal formation occurs. Other proteins, including a-actinin, ezrin, talin, gelsolin, and tropomyosin, were recruited to the site of EPEC attachment independently of Tir tyrosine phosphorylation. Tir tyrosine phosphorylation may then allow recruitment of additional factors, such as N-WASP and the Arp2/3 complex that would target actin-polymerizing machinery to the plasma membrane, initiating full pedestal formation. Although EHEC recruited most of the same proteins as EPEC, there were key differences, namely in the adaptor proteins Nek, Grb2 and Crkll. EPEC may require these adaptor proteins to bind tyrosine phosphorylated Tir and recruit additional factors to the pedestal, whereas EHEC may by-pass this step as it does not require tyrosine phosphorylation of Tir. Collectively, these results provide a basis from which to perform functional analysis of cytoskeletal rearrangements leading to pedestal formation.  65  Chapter 4: The Wiskott Aldrich Syndrome family of Proteins and the Arp2/3 complex participate in EPEC pedestal formation  4.1 Summary: EPEC pedestals are comprised of numerous signaling and structural cytoskeletal proteins. Among these proteins are N-WASP and the Arp2/3 complex, key regulators of actin nucleation. In this report, the role of the Wiskott Aldrich Syndrome family of proteins (WASP) in pedestal formation is addressed. EPEC recruits WASP to the tip of the pedestal via the WASP Cdc42/Rac interaction domain (CRIB; or GTPase binding domain (GBD)). The C-terminal acidic domain of N-WASP is required for recruitment and activation of members of the Arp2/3 complex, proteins that nucleate actin polymerization. N-WASP and the Arp2/3 complex, therefore, are among the key mediators of the EPEC signaling cascade. This study is the first demonstration of cellular mediators of EPEC pedestal formation.  66  4.2 Introduction Extracellular stimuli can induce localized actin rearrangements at the site of stimulation. EPEC pedestal formation may be used as a model to study actin dynamics at the plasma membrane in response to external stimuli. Of particular interest in recent years are the proteins in the WASP family of actin regulators, which are involved in initiating actin nucleation at the plasma membrane. WASP is predominantly expressed in hematopoeitic cells, and N-WASP, a WASP homolog, is broadly expressed. Mutations in WASP are the primary cause of the human immunodeficiency disorder WiskottAldrich syndrome, an X-linked primary immunodeficiency disease that results in severe defects in T cell signaling, chemotaxis, and actin polymerization (Abo, 1998, Gallego et al., 1997). WASP family proteins participate in coordinating responses to extracellular stimuli to the actin cytoskeleton. They contain several protein-binding domains through which they signal from the plasma membrane to the actin cytoskeleton (see Chapter 1.4.4 for more detail on WASP and Arp2/3). The acidic C terminus of WASP provides a binding region for the Arp2/3 complex, key actin nucleating components (Machesky and Insall, 1998). Given the presence of N-WASP and the Arp2/3 complex in the pedestal, and their importance in actin polymerization, their roles in EPEC pedestal formation were addressed. The working hypothesis for this study was that N-WASP was functionally important for pedestal formation, recruiting the Arp2/3 complex and initiating actin polymerization. Results 4.3 Proteins of the WASP family are recruited to the EPEC pedestal HeLa cells were transfected with a plasmid expressing a wild type-WASP tagged with a FLAG epitope, infected with EPEC for six hours,fixed,and stained for actin, WASP and EPEC (Fig 4. IB). WASP constructs (instead of N-WASP) were used in these studies as these were previously described and characterized, and functionally  67  mimicked N-WASP (A.Abo, unpub. obs). The EPEC are shown in Figure 4.1C and the WASP-WT, detected by FLAG staining, is shown in Fig. 4. ID. WASP-WT had no effect on actin pedestal formation as measured by rhodamine phalloidin staining (Fig. 4. IE). As demonstrated previously in Fig. 3.1, endogenous N-WASP is also recruited to the pedestal.  68  Fig.4.1 WASP is recruited beneath EPEC during actin pedestal formation. A. Domain structure of WASP. WASP and N-WASP are similar except that N-WASP contains an additional WH2 domain. B. HeLa cell infected with EPEC for 6 hours. Cells were stained with rhodamine phalloidin which recognizes actin, DAPI which recognizes bacterial and nuclear DNA, and FITC-labelled anti-EPEC antisera. C-F. HeLa cell transfected with Flag-WT-WASP and infected with EPEC. Cells were stained with DAPI (C), anti-FLAG (D) and rhodamine palloidin (E). In the merge (F), EPEC were coloured red, WASP blue, and actin green. Arrowheads represent sites of WASPWT recruitment to EPEC pedestals in transfected HeLa cells. Scale bar represents 5 um.  69  C  EPEC  ^  Actin  "4'  c  -  t  e  r  m  m  u  s  WASP-WT D ^  n  F  ^  *  • »  1  $#|VEPEC  Actin  Merge  70  4.4 The WASP acidic and WH2 domains are required for pedestal formation To determine if WASP was required for pedestal formation, HeLa cells were transfected with WASP and N-WASP proteins containing mutations or deletions in the various domains, as well as the various domains by themselves. Deletion of the C terminus of WASP (AC) results in pronounced defects in the ability of WASP to polymerize actin. Expression of WASP AC blocked pedestal formation in a dominant negative fashion (Fig. 4.2A-D). Pedestals, as detected by phalloidin staining (Fig. 4.2C) were not evident beneath attached bacteria (Fig. 4.2A) in cells expressing WASP-AC (Fig. 4.2B). This inhibition occurred even in cells expressing very low levels of WASP AC (Fig 4.2E; cell on right). In such cells, WASPAC (green, Fig. 4.2E) colocalized with EPEC (blue, Fig. 4.2E) but no pedestals (red) were evident. The C terminus itself neither localized nor blocked pedestal formation (Table 4.1). The WASP C terminus contains a verprolin-homology domain (WH2), a cofilin domain and an acidic (A) domain (Fig. 4.1 A). N-WASP alleles with mutations in the cofilin domain have been reported to act in a dominant negative fashion. However, we could detect no effect of N-WASP Acofilin on pedestal formation or localization, even when the protein was expressed at high levels (Table 4.1). In contrast, the WASP protein lacking the WH2 domain continued to localize beneath the bacteria and blocked pedestal formation when expressed at high levels. N-WASP AA blocked when expressed at low levels and was indistinguishable from WASPAC (Fig. 4.3D-G). These results suggest that both the WH2 and acidic domains are required for pedestal formation, and that deletions in either confer a dominant negative phenotype.  71  Fig.4.2. The C terminus of WASP is required for pedestal formation. A-D. HeLa cells were transfected with WASPAC and infected with EPEC. Cells were stained with DAPI (A), anti-FLAG to recognize WASPAC (B), and rhodamine phalloidin (C). In the merged image (D), EPEC appear as red, WASPAC blue, and actin green. Note that the bacteria on cells expressing WASPAC have no actin pedestals (arrowheads). Scale bars represent 10 um. E . HeLa cell expressing low levels of WASPAC and infected with EPEC. Cells were stained as in A-D. In this image, EPEC appear as blue, actin red, and WASPAC green. The cell on the right expressed low levels of WASPAC that is recruited to the site of bacterial attachment and blocks pedestal formation. The cell on the left was untransfected and developed pedestals. Scale bar represents lOum.  72  Fig. 4.3. Arp2/3 complex localization to the EPEC pedestal requires the N-WASP acidic domain. A - C . HeLa cells were infected with EPEC for 6 hours and stained with DAPI (A), anti-actin (B) and anti- p21, an Arp2/3 complex member (C). Scale bar represents lOpim. D-G. HeLa cells were transfected with N-WASPAA, infected with EPEC, and stained with DAPI (D), anti-HA to recognize N-WASPAA (E), rhodamine phalloidin (F) and anti-p41 (an Arp2/3 complex member) antibody (G). Note that p41 fails to accumulate beneath EPEC in cells expressing N-WASPAA and pedestals are not formed. Arrowhead denotes the lack of p41 recruitment to the site of EPEC adherence. Scale bar represents lOum.  74  75  4.5 The Arp2/3 complex is recruited to pedestals The observation that N-WASPAA blocked pedestal formation suggested that the Arp2/3 complex may be involved. To test this, HeLa cells were infected with EPEC, fixed, and stained for members of the Arp2/3 complex: p21 (Fig. 4.3A-C), p40 and p41 (Fig. 4.3D-G). All three were recruited to the pedestals uniformly throughout the length and at the tip (see also Fig. 3.1R). Expression of WASPAC or N-WASPAA prevented pedestal formation but were still recruited beneath adherent EPEC. Expression of both dominant negatives also prevented localization of p41 and p40 beneath the bacteria. The Arp2/3 complex, therefore, depends on the WASP acidic domain for recruitment to the pedestal. This is in accordance with what is observed for actin nucleation events at the plasma membrane in leading edge formation (Rohatagi, 1999; Machesky, 1999). 4.6 The WASP GBD is necessary and sufficient for localization to the pedestal tip To determine which domains of WASP were required for localization to the pedestal tip, WASP mutations or deletions of various functional domains outside the C terminus were transfected into HeLa cells. HeLa cells were then infected with EPEC and screened for effects on pedestal formation (Table 4.1). Deletion of the GBD domain prevented WASP from localizing in pedestal tips (Fig.4.4A-D). In cells expressing high levels of the protein, WASPAGBD was present in the pedestals but was not enriched relative to the cell body, nor did it localize to the tip. WASP proteins with mutations in the cofilin domain, the polyproline region, and the N-terminus encompassing the WH1 and PH domains, localized in a manner comparable to the wild type protein (Table 4.1). These results suggest that WASP is recruited to the pedestal tip by a GTPase. When expressed at low levels, the WASP GBD alone localized to the pedestal tips (Fig. 4.4 E-H). When expressed at high levels, the WASP GBD blocked pedestal formation but continued to localize beneath the bacterium (not shown), suggesting that  76  the GBD can competitively inhibit binding of endogenous WASP-like proteins. The effect of this domain alone was specific.  77  Fig. 4.4. The GBD targets WASP to the pedestal tip. A-D. HeLa cells were transfected with WASPAGBD, infected with EPEC and stained with DAPI (A), anti-Flag to recognize WASPAGBD (B) and rhodamine phalloidin (C). In the merged image (D), the bacteria are coloured blue, WASPAGBD green, and actin red. Note that the absence of the GBD prevents WASP localization exclusively at the tip. The protein was in the pedestal but it was not enriched relative to the cell cytoplasm. Scale bar represents lum. E-H. HeLa cells were transfected with low levels of WASP-GBD and infected with EPEC. Cells were stained as in A-D. In the merged image (H), actin is red, WASP-GBD is green, and EPEC blue. Note that the GBD localizes to the pedestal tip. Scale bar represents ljim. I-L. HeLa cells were transfected with WASPAC/AGBD and infected with EPEC. Cells were stained with DAPI (I), anti-Flag to recognize WASPACAGBD (J) and rhodamine phalloidin (K). In the merged image (L), EPEC are blue, actin red, and WASPACAGBD green. WASPACAGBD was present in the pedestals though it was not enriched relative to the cytoplasm. Despite the presence of high levels of WASPACAGBD, pedestals were formed. Scale bar represents lOum.  78  79  Table 4.1. Summary of effects of wild type and mutant WASP family members on EPEC pedestal formation. Localization refers to the distribution of transfected epitopetagged protein in the pedestal; "+" refers to localization in the tip directly beneath the bacterium whereas "-"refers to nonspecific localization throughout the pedestal without enrichment relative to the rest of the cell cytoplasm. In the case of WASP mutants that blocked pedestal formation "+" refers to protein accumulation directly beneath the bacterium. Effects on pedestal formation were determined by examination of actin in transfected cells compared to untransfected cells, and were considered all or none. In cases where the construct was determined to "block" no pedestals were evident whether the construct was expressed at low or high levels. In some cases ("no effect/block") pedestal formation was only blocked when the protein was expressed at high levels.  80  Construct Flag-WASP-WT HA-N-WASP GFP Fl-WASP-AC Fl-C HA-N-WASPAAcidic HA-N-WASP-ACofilin HA-NWASPAAcidic/Acofilin Fl-WASPAGBD Fl-WASP-GBD Myc-PAK-GBD Fl-WASP-ACAGBD Fl-WASPAN Fl-WASP-N F1-WASPAWH1 F1-WH1 Fl-WASP-APP Fl-PP F1-WASPAWH2 F1-WASP-WH2  Localization + +  -  +  -  + + +  +  -  +  -  +  -  +  -  +  Pedestals No effect No effect No effect Block No effect Block No effect Block No effect No effect/block No effect No effect No effect No effect No effect No effect No effect No effect No effect/block No effect  The PH, PP, and WH1 domains by themselves were without effect of pedestal formation and did not localize (Table 4.1). Moreover, a GBD from PAK3, which shares a 70% amino acid homology with WASP GBD, neither blocked pedestal formation nor localized to the pedestal (Table 4.1). Therefore, the WASP GBD was necessary and sufficient for localization to the pedestals. However, the possibility that additional domains contribute to the localization cannot be ruled out.  4.7 Localization of WASP via the GBD is required for pedestal formation To determine if WASP activity was dependent on localization of WASP to the tip, we asked whether the inhibitory effects of WASPAC required an intact GBD. Expression of a mutant WASP with deletions in both the C terminus and the GBD (AC/AGBD) neither blocked pedestal formation nor localized beneath adherent bacteria (Fig.4.4 I-L). These results suggest that WASP is localized to the tip of the EPEC pedestal via its GBD and recruits the Arp2/3 complex through its acidic C-terminus for pedestal formation.  4.8 Discussion How do extracellular factors induce actin polymerization at the site of stimulation? In this study, a role for WASP and the Arp2/3 complex was described for EPEC pedestal formation. These are the first host cell factors identified as having a role in the EPEC signaling pathway (Kalman, et al., 1999). Recent biochemical experiments have linked WASP directly to the Arp2/3 complex, which nucleates actin polymerization. In particular, the C-terminus of WASP associates with and potentiates the nucleating activity of the Arp2/3 complex. The acidic domain of WASP directly binds the Arp2/3 complex member p21, but both the acidic and WH2 domains are required for activation of the complex. WH2 binds actin directly, perhaps as a means of restricting new polymerization to existing filaments. The results presented here show a requirement for both WASP acidic and WH2 domains in EPEC pedestal formation and the Arp2/3 complex recruitment in vivo.  82  The ability of WASP family members to activate actin polymerization is regulated in cells by signaling molecules such as Cdc42 (Prehoda, et al., 2000, Rohatgi, et al., 2000, Symons, et al., 1996). This signaling cascade has been reconstituted in cell extracts where Cdc42 and PIP2 act through N-WASP to activate the Arp2/3 complex (Rohatgi, et al., 1999). There is evidence that the WASP GBD in cis may inhibit the catalytic function of the C terminus and binding of a GTPase may alleviate this inhibition. The presence of GBD prevents the ability of WASP or N-WASP to stimulate actin polymerization in cells or to activate Arp2/3 in extracts (Rohatgi, et al., 1999). The addition of Cdc42 can overcome that inhibition. That the GBD appeared to be necessary for recruitment of WASP to the pedestal tip suggests that members of the small GTPases family are involved in pedestal formation. Two studies, however, showed that not only were they not involved, but that inhibition of the Rho subfamily actually resulted in more pronounced pedestal formation (Ben-Ami, et al., 1998, Ebel, et al., 1998). Inhibition of all GTPases by compactin (which prevents isoprenylation of the GTPases required for activity) or the Rho subfamily in particular by the Clostridium difficile toxin B (ToxB) (which glucosylates and inactivates Cdc42, Rac, and Rho) resulted in elongated pedestal formation (Ben-Ami, et al., 1998). Dominant negative alleles of Cdc42, Racl, and RhoA had no effect on pedestals, nor did their wild type or activated alleles localize to the pedestal tip (BenAmi, et al., 1998) (Ebel, et al., 1998), (Kalman, Weiner, and Goosney, unpub. obs). These data suggested the possibility that a ToxB and compactin insensitive GTPase was involved in pedestal formation. A good candidate for this GTPase was Chp (Cdc42 homologous protein). Chp has been shown to bind WASP in vitro, and as its name implied, has a strong homology to Cdc42 (Aronheim, et al., 1998). It also contains additional N and C terminal sequences not found in the rest of the GTPase family. As shown in Appendix 2, Chp is indeed insensitive to isoprenylation and to ToxB glucosylation. Furthermore, it was recruited specifically to the tip of the pedestal when  83  transfected into HeLa cells. Dominant negative alleles of Chp failed to block pedestal formation when expressed in HeLa cells, however since Chp has no function ascribed to it, it could be that these putative dominant negative proteins are in fact not inhibitory at all. These data suggests the possibility that a Chp-like protein likely participates in pedestal formation by recruiting N-WASP to the tip. Also, the possibility cannot be ruled out that the Chp-like molecule is of bacterial origin. This would not be unprecedented, as other pathogenic bacteria, like Salmonella typhimurium, possess regulators of GTPase function (Hardt, et al., 1998). Intriguingly, EPEC itself has a putative GTPase called BipA, which has effects on pedestal formation but as of yet has not been demonstrated to be secreted by EPEC (Farris, et al., 1998). Recent evidence implicates the Arp2/3 complex, WASP and Rho family GTPases in the spatial control of actin polymerization (Higgs and Pollard, 1999). Artificial targeting of WASP or Cdc42 to a localized site on the plasma membrane is sufficient to generate actin accumulation and filopodia-like structures at that site. In accordance with these data, EPEC triggers localized recruitment and activation of WASP and the Arp2/3 complex to localized site of actin rearrangements in vivo. A possible model for the role of WASP and the Arp2/3 complex in EPEC pedestal formation is presented in Fig. 4.5, based on the data presented here and in vitro studies. Upon loose attachment to the host cell, EPEC secretes Tir into the host cell via the type III secretion system. Once delivered and inserted into the plasma membrane, Tir binds intimin on the outer membrane of EPEC. Tir then recruits, through an unknown mechanism, a Chp-like GTPase, which then recruits WASP through its GBD. GTPase binding of WASP causes exposure of the WH2 and acidic domains in the WASP C terminus, which then recruits and activates the Arp2/3 complex, stimulating actin nucleation and polymerization. In conclusion, WASP family members and the Arp2/3 complex are the first identified mediators of a signaling cascade initiated at the cell surface by EPEC and  84  culminating in actin polymerization and pedestal formation. Future studies in understanding how EPEC interfaces with the cellular signaling machinery will provide insight into the molecules that link membrane receptors with the actin cytoskeleton.  85  Fig. 4.5 Model of W A S P and Arp2/3 action in E P E C pedestal formation. EPEC attach to the host cell loosely (stage 1) and secrete virulence factors into the host including Tir. Tir is inserted into the host cell membrane and binds intimin on the EPEC outer membrane (stage 2). Tir-intimin binding leads to recruitment of a Chp-like GTPase (stage 3) which in turn recruits WASP through its GBD (stage 4). Once WASP is bound to the GTPase, the WASP acidic domain is free to bind and activate the Arp2/3 complex (stage 5). The activated Arp2/3 complex then nucleates actin polymerization leading to pedestal formation (stage 6).  86  6  Chp-hke GTPase  WASP  ArpZO  Chapter 5: Yeast two hybrid system to identify binding partners of Tir 5.1 Summary: Although N-WASP and the Arp2/3 complex had been identified as effectors in pedestal formation, the direct binding partner of Tir remained unknown. The yeast twohybrid system was used to screen for interactions between the protein Tir (the bait) and proteins encoded by a HeLa cell cDNA library (the prey). HeLa c D N A library was amplified to 6 xlO clones, from which 13 positively interacting clones were isolated and 6  sequenced. Two of these clones were picked as potential candidates for Tir interactions CD44 and profilin. CD44 is a membrane protein that binds hyaluronic acid in the extracellular matrix, whereas profilin is a key regulator of the actin cytoskeleton. Domain mapping by yeast two hybrid showed that CD44 interactions with Tir required both Tir transmembrane domains (amino acids 156-475), whereas profilin required Tir N terminus from amino acids 156-253. To confirm the binding by cell biology or biochemical means, the localization of CD44 in EPEC pedestals was determined by immunofluorescence. CD44 was localized to the EPEC pedestal tip and recruited independently of Tir (Chapter 3). CD44-deficient 3T3 fibroblasts were capable of making elongated pedestals as determined by Tir and actin staining. Immunoprecipitation of CD44 from HeLa cells did not pull down Tir. Therefore CD44 binding to Tir could not be confirmed by cell biological or biochemical means, nor did it seem to be functional in pedestal formation. The same approach was taken with profilin. Profilin could not be detected in the EPEC pedestal in HeLa cells overexpressing GFP-profilin. ELISAs were performed using purified proteins which showed no direct binding of profilin to Tir. Taken together, these data indicate that the binding of profilin to Tir in yeast could not be confirmed by cell or biochemical means.  88  5.2 Introduction An important part of determining the role of a protein in a given system is to identify and characterize the interactions it has with other proteins. In the case of Tir, it had been shown to bind intimin, but there were no known interacting proteins in the host cell. There are numerous approaches to identify interacting partners for specific proteins, such as co-purification and sequencing of components, but these techniques are laborious and difficult to perform. In recent years, a genetic approach was developed called the yeast two-hybrid system that was easier and faster than traditional biochemical methods. It relies on two classes of chimeric proteins used in the screen: (1) the bait, a fusion of the protein of interest (in this case, Tir) to a DNA binding domain (LexA) (Brent and Ptashne, 1980), and (2) the prey, a fusion of a cDNA library (a HeLa cell library for these studies (Gyuris, et al., 1993)) to a transcriptional activation domain (in this system, the "acid blob" B42; Ruden, et al., 1991). The bait is transformed into a suitable strain of yeast with a dual reporter system that contains binding sites for LexA. The dual reporter system reduces the chance of false positives in the system. In this case, the lexA operator sites are upstream of lacZ, encoding (3-galactosidase, and LEU2, a gene required in yeast leucine biosynthesis. Yeast containing bait only do not turn on the reporters and thus are unable to break down galactose (remaining white when grown on X-Gal) and are unable to synthesis leucine (failing to grow in leucine-deficient medium). These yeast are transformed with the HeLa cDNA library and plated on the appropriate selective media. If there is an interaction with Tir, the activation domain is brought to the lexA sites, and colonies turn blue when plated on X-Gal media and grow on medium lacking leucine. This technique was used to initially identify potential binding partners of Tir. Several potential Tir binding partners were identified (Table 5.1), but two proteins were further studied: CD44 and profilin. The other proteins were not cytoskeletal-associated, not of HeLa origin, or had too low an identity with the query sequence to warrant an  89  initial biochemical and cell biology analysis. CD44 is an integral membrane that mediates binding to hyaluronic acid in the ECM (Goodison, et al., 1999). It links to the actin cytoskeleton through the linker proteins ezrin and ankyrin (Bourguignon, et al., 1992, Tsukita, et al., 1994). It also associates with annexin II in lipid rafts, microdomains that cluster signaling molecules together (Oliferenko, et al., 1999). Profilin is also strongly associated with the actin cytoskeleton. It is a small, monomeric-actin sequestering protein that is involved in actin polymerization (Sohn and GoldschmidtClermont, 1994). Both proteins were associated with the actin cytoskeleton and therefore seemed to be likely candidates for further characterization for Tir binding, including colocalization experiments by immunofluorescence and direct biochemical binding assays.  Results 5.3 Positive Tir-interacting proteins by yeast two hybrid screen Of the 6xl0 HeLa cDNA clones used in the screen, 26 positive clones were 6  identified. Of those cDNAs, 13 were successfully sequenced (the other 13 had repeated slippage problems after the poly-T region and could not be sequenced after repeated attempts) and the results are tabulated in Table 5.1. Of the clones sequenced, two proteins- CD44 and profilin- seemed to be the most likely candidates interacting with Tir in the pedestal. The other proteins were not cytoskeletal-associated, not of HeLa origin, or had too low an identity with the query sequence to warrant an initial biochemical and cell biology analysis.  90  Table 5.1 Sequencing results from positive Tir-interacting proteins as determined by the yeast two hybrid screen Clone # 1,6 2 3 4 5 7 8 9  10 11  12 13  Encoded protein Translation initiation factor 3 47kDa subunit Ubiquitin Conjugating enzyme PZE40 gene product Thyroid Hormone receptor coactivation protein Human Spliceosome Associated Protein Calpactin light chain Mettallothionein-II Myelin basic protein  CD44  Yeast hypothetical protein Profilin Homeobox protein  % Identity with Query 86%  Species  98%  Homo sapien  36% 98%  Hordeum vulgare Homo sapien  75%  Homo sapien  83%  Homo sapien  98% 62%  Homo sapien Mus musculus Homo sapien Saccharomyces cerevisiae Homo sapien Gallus gallus  100%  29%  99%  44%  Homo sapien  91  The region of CD44 that was pulled out in the yeast two-hybrid screen as interacting with Tir was residues 197-326, which corresponds to the extracellular and transmembrane domain of CD44. The region of profilin that interacted with Tir in yeast was residues 1-121, which corresponds to all but the very C-terminus of profilin. 5.4 Domain mapping of Tir interactions with CD44 and profilin  Tir truncations were constructed, put into the bait vector pEG202, and tested for interactions with CD44 and profilin in the yeast two-hybrid system. CD44 binding to Tir required the Tir transmembrane domains (amino acids 156-475) whereas binding to profilin required the N-terminus from residues 156-253 (Fig. 5.1). As a positive control for Tir binding, intimin282 was used and was found to bind in the region between the two transmembrane domains (now referred to as the intimin binding domain). 5.5 CD44 is not required for pedestal formation  The identification of CD44 as a Tir binding partner was an intriguing result as early work in identifying Tir (or as it was known then, Hp90) suggested it was CD44 from co-purification experiments with intimin (D.J. Reinscheid, unpublished observations). CD44 was studied initially using cell biology and biochemical techniques. As shown in Chapter 3 (see Fig. 3.1 and Fig.3.2), CD44 was indeed recruited to the EPEC pedestal, but its recruitment to the site of attachment was independent of Tir translocation and insertion in the plasma membrane. To determine if CD44 was needed in the pedestal, CD44-deficient Swiss 3T3 cells were infected with EPEC for 5 hours. EPEC made elongated pedestals as observed by Tir and actin staining in the absence of CD44 (Fig. 5.2). Therefore, CD44 appeared to have no functional role in the EPEC pedestal. Using this CD44-deficient cell line, it was observed that Tir was still efficiently transferred and tyrosine phosphorylated in the cell (D.J. Reinscheid, unpublished observations).  92  Int282  II  Tir 1-550  '  Tir1-362  I  II  Tir 253-550 i  I  +  +  +  .  I  I  Tir 156-475  II  n  II n  I  Tir 156-550  I I  D  •  U  U  profilin  +  +  +  I  +  Tir 1-475  CD44  ~l  +  +  +  ^  +  +  +  Fig. 5.1 Domain mapping of Tir interactions with CD44 and profilin using the yeast two hybrid system. Tir truncations were constructed and inserted into the "bait" vector pEG202. The CD44 ,profilin, and intimin282 cDNAs in the "prey" vector pJG34-18 were transformed into yeast containing the different Tir baits and screened for interactions as described in the Materials and Methods. "+" designates a positive interaction, whereas "-" indicates no interaction.  93  F i g . 5.2. C D 4 4 is not required for pedestal formation. Swiss  3T3 fibroblasts  (CD44-/-) were infected with EPEC for 5 hours. Pedestals were formed in the absence of CD44 as determined by Tir (A) and actin (B) staining. A merge of both (C; Tir, green; actin, red) is shown in the bottom panel. Scale bars represent 5 |im. (D) Western immunoblot of CD44 expression in HeLa, Swiss 3T3 and Swiss 3T3 CD44-deficient cells  94  A  Tir  Actin  C  Merge  i). HeLa 3T3  3T3 ___44  85kDa  probe: anti-CD44  95  5.6 Tir does not co-immunoprecipitate with CD44 HeLa cells were infected with EPEC and lysed with 1% Triton X-100. CD44 was immunoprecipitated from the HeLa cell lysate and the immunoprecipitate was probed for Tir. As seen in Fig. 5.3, CD44 was immunoprecipitated but Tir remained in the supernatant. These results indicate that Tir and CD44 do not associate within the host cell. 5.7 Profilin does not localize to the pedestal Given the disappointing results with CD44, the second likely candidate, profilin, was examined. Profilin antibodies proved to be difficult to obtain (at the time of the experiments, few labs made their own profilin antibody, and I was fortunate to receive aliquots from two sources. However, problems arose with both antisera: they were from rabbits and cross-reacted strongly with EPEC (despite absorbing them against fixed EPEC), and they did not work in immunofluorescence). I therefore used a GFP-profilin transfection vector to overexpress profilin in HeLa cells. As seen in Fig. 5.4a-d, GFPprofilin was not recruited specifically to EPEC pedestals. It had the same diffuse staining pattern as GFP vector alone (Fig. 5e-h), being present diffusely in the cytosol, but not enriched in the pedestals specifically. 5.8 Profilin does not bind Tir by ELISA To test if profilin could bind Tir in vitro, purified profilin, int282 and BSA were incubated in a 96-well plate and overlaid with purified Tir (Fig. 5.5). Tir bound int282 strongly, but did not bind profilin or the BSA negative control. Together with the immunofluorescence data, it was concluded that profilin was not Tir's binding partner in the pedestal.  96  Fig. 5.3. Tir does not co-immunoprecipitate with C D 4 4 . HeLa cells were infected with EPEC for three hours, lysed in PBS- 1% Triton-XlOO, and CD44 immunoprecipitated using a rat anti-CD44 polyclonal antibody. The immunoprecipitate (IP), and remaining supernatant (post-IP) were probed for CD44 and Tir. Tir is present in the post-IP, but does not IP with CD44.  97  Uninfected  EPEC Infected  IP  IP  post-IP  post-IP 80kDa  —80kDa Anti-Tir  Fig.5.4 Profilin is not recruited to the pedestal. H e L a cells overexpressing G F P profilin or G F P only were infected with E P E C for 5 hours. E P E C were stained with D A P I (C and F) and pedestals were observed by phalloidin staining ( A and D ) . G F P profilin was not enriched i n the pedestal compared to G F P only (B and E ) . The bottom panels (D and G) represent a merge of A - C and D - F respectively, with actin pseudocoloured red, G F P green, and D A P I blue. Scale bars represent 5 \im.  99  GFP-profilin  GFPonly  100  0.7  T  0.6 0.5 M <  0.4  Q profilin • int282 BSA  0.3 0.2 0.1  T  0 -0.1  Fig. 5.5. Profilin does not bind Tir directly. Purified profilin (Cytoskeleton Inc.), intimin282, and BSA were incubated on a 96 well Immunlon-2 plate and overlaid with l[ig of Tir. Binding was detected using Tir monoclonal antibodies followed by horesradish-peroxidase conjugated anti-mouse secondary antibody. The reaction was developed colorimertrically and absorbance read at 492nm. Data represent averages and standard deviations of three independent experiments performed in triplicate. Tir binding to profilin was not significantly different from the BSA negative control as determined by a Student's t-test.  101  5.9 D i s c u s s i o n  The yeast two-hybrid system can be a very powerful tool to detect protein-protein interactions, but as observed here, it has its limitations. Any positively identified interacting protein requires additional data by biochemical or cell biology approaches to confirm the positive interactions. In this particular screening, there were several problems that may have contributed to incorrect identification of Tir's binding partner. The HeLa cell cDNA library that was used in this study may or may not have been complete at the onset. This, together with an amplification process that resulted in only the minimal number of clones needed to do the screening, could account for some of the problems encountered with this system. 9.6 x 10 clones were required to ensure that all 6  HeLa genes were presented (Unit 20, Current Protocols in Molecular Biology, 1996), and we used 6 x 10 clones to do this screening. Another major problem was that Tir is a 6  membrane protein, and as such, has hydrophobic domains that can participate in hydrophobic interactions with other proteins non-specifically. This may well be the case with CD44, as domain mapping demonstrated that both Tir and CD44 transmembrane domains were involved in mediating their interactions. At the time of this screening, the yeast two-hybrid system had not been well characterized for membrane proteins. In hindsight, it would have been more beneficial to screen with only the N or C terminus of Tir lacking the transmembrane domains. However, Tir had only been recently identified and very little was known about its orientation in the host cell membrane. Identifying the part of Tir with which to perform the screen would have been difficult to determine. In fact, this same system was initially used to successfully map out the Tir-intimin binding domain concurrently with this study. CD44 was a very interesting candidate for Tir binding. As mentioned earlier, CD44 had initially been a candidate for Tir itself, before the discovery that Tir was bacterial in origin. This hypothesis was based on an observation that CD44 co-purified with intimin, was tyrosine phosphorylated, and was of the right molecular weight to be  102  Hp90 (the original name of Tir). This line of study was no longer pursued following the observation that Hp90 was still present as a tyrosine phosphorylated protein in CD44deficient cell lines. Upon identifying CD44 as a potential binding partner for Tir, I was encouraged by these previous findings, believing that the co-purification experiments could be explained by Tir and CD44 interacting, not that CD44 was Tir. Unfortunately, further analysis by immunoprecipitation and CD44-deficient fibroblasts did not support this theory. The binding seen in the yeast, therefore, is probably due to hydrophobic interactions between two membrane proteins. Profilin was also an interesting candidate due to its importance in regulating actin polymerization events. It interacts with Arp2/3, WASP-family members and VASP, all of which were found in the pedestal (Mullins, et al., 1998, Reinhard, et al., 1995, Suetsugu, et al., 1998). Since WASP and the Arp2/3 complex were functionally important in pedestal formation, I was excited by the possibility that profilin would therefore be a key regulator of pedestal formation as well. However, the ELISA to detect direct binding and the colocalization experiments were negative, indicating that the binding in the yeast was a false positive. At the time of the CD44 and profilin experiments, I performed another screening for Tir-interacting proteins using a Tir affinity column. This approach was a more direct method for identifying a binding partner, and worked very well in obtaining other likely candidates for binding (see Chapter 6). Other less likely candidates from the yeast twohybrid screen were therefore not examined.  103  Chapter 6: Enteropathogenic E. coli translocated intimin receptor, Tir, interacts directly with a-actinin 6.1 Summary: Tir is believed to anchor EPEC firmly to the host cell, although its direct linkage to the cytoskeleton was unknown. In this chapter, Tir is shown to directly bind the cytoskeletal protein a-actinin, a key component of focal adhesions. a-Actinin co-eluted off an affinity column with Tir, suggesting an indirect or direct association. Far Western and enzyme-linked immunosorbent assays demonstrated that Tir and a-actinin bind directly in a concentration-dependent and saturatable manner. Tir and a-actinin interacted in HeLa cells as well, as shown by co-immunoprecipitation. a-Actinin was recruited to the pedestal in a Tir-dependent manner and colocalized with Tir in infected host cells. Binding was mediated through the amino (N)-terminal 200 amino acids of Tir. Recruitment of a-actinin occurs independently of Tir tyrosine 474 phosphorylation (Chap.3). a-Actinin appears to be functional in the pedestal, as its overexpression resulted in a two-fold increase in pedestal length. Overexpression of a-actinin actin binding domain (ABD), spectrin-like repeats (SLR) and the calcium-binding domain (CBD) had no marked effect on pedestal formation or length, though both the ABD and SLR were recruited specifically to pedestal. These data suggest that full length a-actinin may be required for proper folding and activity. The results presented here indicate that Tir plays at least three roles in the host cell during infection: binding intimin on the EPEC outer membrane to attach to the bacterium; mediating a stable anchor with a-actinin through its N-terminus in a phosphotyrosine-independent manner; and recruiting additional cytoskeletal proteins at the C-terminus in a phosphotyrosine-dependent manner. These findings demonstrate the first known direct linkage between extracellular EPEC, through the transmembrane protein Tir, to the host cell actin cytoskeleton via aactinin. 104  6.2 Introduction The mechanism by which Tir anchors itself to the mammalian cell cytoskeleton remains unknown. In this study, we show that Tir binds a-actinin directly (Goosney, et al., 2000). a-Actinin is a 100 kDa protein that functions as an anti-parallel homodimer, cross-linking actin filaments. It also links actin filaments to membrane receptors, including {3,integrins, and soluble cytoskeletal regulators, such as vinculin, and zyxin (Heiska, et al., 1996, McGregor, et al., 1994, Otey, et al., 1990, Pavalko, et al., 1995, Papa, et al., 1999, Eilertsen, et al., 1997, Crawford, et al., 1992). a-Actinin is a key component of focal adhesions, stress fibers, and microvilli. By binding a-actinin, Tir forms a continuous transmembrane link between intimin on the bacterial surface and the host cell cytoskeleton. Results 6.3 a-Actinin is recruited to the pedestal in a Tir-dependent manner Tir delivery to the host cell by the type III secretion system of EPEC is the first step in restructuring the actin cytoskeleton to form A/E lesions. To determine if a-actinin recruitment to the pedestal was Tir -dependent, HeLa cells were infected for 5 hours with wild type EPEC, the intimin deletion mutant CVD206, the Tir mutant Atir, and a type III secretion mutant cfm 14-1-1 (Fig.6.1a-d). Following infection, cells were fixed and prepared for immunofluoresence. a-Actinin was recruited beneath the wild type bacterium along the length and concentrated at the tip of the EPEC pedestal (Fig.6.1a). The intimin mutant, which translocates Tir but doesn't cluster it, had an unfocused recruitment pattern of a-actinin beneath the pathogen, which matched the unfocused Tir in the membrane (Fig. 6.1b). The Atir mutant did not recruit a-actinin nor made pedestals (Fig.6.1c). The type III secretion mutant cfm 14-1-1, unable to secrete Tir, behaved like the Atir mutant, as no a-actinin was recruited (Fig. 6.1d).  105  The kinetics of a-actinin recruitment was determined to see if they matched Tir delivery to the host membrane (Fig.6.2). HeLa cells were infected with wild type EPEC for 45, 60, 75, 90, 105, and 120 min and prepared for immunofluoresence. It has been well established that once Tir is delivered to the host cell, it is phosphorylated and clustered by intimin, leading to pedestal formation. a-Actinin recruitment to the site of EPEC attachment occurred at 90 min after infection and remained throughout the length of the infection. This recruitment corresponded well with the time of insertion and phosphorylation of Tir in the host cell membrane. Note that initial EPEC attachment (not mediated through Tir) causes no accumulation of a-actinin beneath EPEC (Fig 6.2 60 min).  106  Fig.6.1. a-Actinin is recruited to the pedestal in a Tir-dependent manner. HeLa cells were infected with A) wild type EPEC, B) the intimin deletion mutant CVD206, C) EPECArir, orD) the type III secretion mutant cfm-14 for 5 hr at 37°C and prepared for immunofluoresence as described in the Materials and Methods. Cells were probed for aactinin (green) or Tir (red) and the nuclear and bacterial DNA stained with DAPI (blue). Scale bars represent 2 \im. Arrows indicate attached bacteria.  107  108  Fig. 6.2. a-Actinin recruitment coincides with Tir insertion in the HeLa plasma membrane. HeLa cells were infected with EPEC for 45, 60, 75, 90, 105, and 120 min with EPEC. Cells were prepared for immunofluorescence and stained for a-actinin (red) and Tir (green). Phase contrast shows EPEC attached to the HeLa cell. a-Actinin recruitment occurs at 90 min, following Tir insertion in the membrane.  109  60min  90min  105min  75min  1  2  0  m  i  n  a-actinin (r) Tir(g)  110  6.4 a-Actinin elutes with Tir on a Tir affinity column To determine if a-actinin associated with Tir, a Tir affinity column was constructed. His-tagged Tir was attached to a Ni-NTA column. HeLa cell lysates were incubated with Tir-Ni NT A for 1 hour, then washed extensively. Tir and the HeLa proteins associated with it were eluted with increasing amounts of imidazole (Fig 6.3). Tir elution began at fraction 33 (180 mM imidazole), peaked at fraction 37 (200mM imidazole), and was slowly eluted until fraction 43 (225 mM imidazole) (Fig.6.3a). a Actinin eluted at fraction 35 (185 mM imidazole), peaked at fraction 37 (200 mM imidazole), and slowly eluted until fraction 41 (215 mM imidazole), matching the Tir elution profile (Fig.6.3b). a-Actinin did not elute with the same pattern in the absence of Tir or when His-Tir (392-550) was present (Fig.6.3c). These results confirmed the above immunqfluoresence data, indicating that a-actinin interacts directly or indirectly with Tir. 6.5 Tir binds purified a-actinin directly To determine if Tir binds a-actinin directly, two experimental approaches were used: Far Westerns and ELISAs. Purified a-actinin (Sigma), int282, and BSA (Sigma) were run on an 8% acrylamide gel, transferred to nitrocellulose, and renatured. The nitrocellulose was overlaid with purified Tir for 2 hr, then probed with a Tir monoclonal antibody to detect Tir binding to intimin or a-actinin. Tir bound directly to intimin as expected, but also to a-actinin on the nitrocellulose (Fig.6.4a and b). There was no nonspecific binding of Tir to BSA or the molecular weight markers. To determine if this binding was concentration dependent, an ELISA was developed. Int282, a-actinin, and BSA were bound to an Immulon-2 96 well plate (0.5 ug /well each, or 165 nM of int282, 50 nM of a-actinin and 75 nM of BSA) and overlaid with increasing amounts of Tir+CesT (0,18 nM, 35 nM, 90 nM, 180 nM, and 360 nM /well). CesT is a bacterial Tir chaperone that prevents Tir degradation and aggregation in solution (Abe, et al., 1999). Tir interacted with a-actinin, saturating at 180 nM Tir  111  (Fig.6.4c). Int282 also interacted with Tir as expected, saturating at 18 nM Tir. Tir did not bind to the BSA control nor did it bind non-specifically to the Immulon-2 plate. Purified CesT (350 nM) did not bind int282, a-actinin, or BSA (Fig.6.4d). To further establish direct binding, the reverse assay was also performed. Tir+CesT (90 nM/well) was bound to the Immulon plate and overlaid with increasing amounts of purified a-actinin (0, 100 nM, 200 nM, and 500 nM). As shown in Fig.6.4e, a-actinin bound to Tir directly in a saturatable manner. a-Actinin did not bind nonspecifically to the controls. Collectively, the above results show that Tir binds purified a-actinin directly. 6.6 a-Actinin binds the N-terminus of Tir Tir is comprised of two transmembrane domains, an extracellular intimin binding domain (IBD) and two intracellular domains corresponding to the N and C termini (deGrado et al., 1999, Kenny, 1999). The C terminus contains several tyrosine residues, one of which (tyrosine 474) is phosphorylated in the host cell, a process essential for pedestal formation (Kenny, 1999). A point mutation of this tyrosine residue abolishes phosphorylation and actin accumulation, but does not affect its delivery to host cells. To determine which domain of Tir binds a-actinin directly, truncations of Tir were constructed, purified, and used in an ELISA as described above. Int282, a-actinin, and were bound to an Immulon-2 plate and overlaid with the various purified Tir truncations. As expected, full length Tir bound to int282 and a actinin, but not to the BSA control as demonstrated above. Tir (1-391), which includes the cytoplasmic N-terminus plus the extracellular intimin binding domain (IBD), bound to both int282 and a-actinin (Fig.6.5). Tir( 1-200), containing only the N-terminus but lacking the IBD, bound a-actinin but not intimin. The cytoplasmic C-terminus of Tir (392-550) did not bind to either protein. Tir(202-550), containing the IBD and the Cterminus, bound to intimin but not to a-actinin. None of the Tir truncations bound to the  112  controls (data not shown). These results demonstrate that the first 200 amino acids of Tir are required for direct binding to a-actinin. 6.7 a-Actinin co-immunoprecipitates with Tir HeLa cells were infected with EPEC for 3 hours and cross-linked with the aminereactive cross-linker disuccinyl-suberate then lysed with RIPA buffer. Following crosslinking, cell lysates were immunoprecipitated with Tir monoclonal antibody, run on SDSPAGE, transferred to nitrocellulose and probed for Tir and a-actinin (Fig.6.6). a-Actinin co-immunopreciptated with Tir in infected HeLa cells, but did not come down in the immunoprecipitate in uninfected HeLa cells. These results suggest that the interaction seen in vitro is occurring in the cell as well.  113  Fig.6.3. a-Actinin elutes with Tir off a His-Tir Ni-NTA column. Western blot of column fractions of increasing imidazole concentrations, probed for (A) Tir and (B) a-actinin. a-Actinin did not elute off a His-Tir(392-550) Ni-NTA column with the same pattern (C).  114  97 kDa 71 kDa  Fraction  4  31  33 35 37 39 41 43 45  B.  a-actmin  97 kDa 71 kDa Fraction  4 29 31 33 35 37 39 41 43  c.  T i r (392-550) a-actinin  Hela lysate  1  2 3 4 5 6 7 8 9  Hela lysate+ NiNTA-Tir(392-550)  1  Fig.6.4. Tir binds a-actinin directly. A. Far Western of immobilized a-actinin (on nitrocellulose) binding soluble Tir. The nitrocellulose was probed with anti-Tir monoclonal antibodies. Tir bound int282 and aactinin but not to BSA or other high-molecular weight markers. B. Coommassie stained acrylamide gel showing the relative loading of a-actinin, int282 and BSA for the Far Western in A. C. Concentration-dependent binding of Tir to a-actinin in an ELISA. a-Actinin (black diamonds), int282 (open squares), and BSA (black circles) were incubated on an Immunlon-2 96 well plate and overlaid with increasing amounts of Tir. Tir was detected with an anti-Tir monoclonal and a peroxidase-conjugated secondary antibody. The ELISA shown is a representative of 10 separate experiments done in triplicate. D. Concentration-dependent binding of a-actinin to Tir in an ELISA. Tir (open squares) and BSA (black diamonds) were incubated on a 96-well plate and overlaid with increasing amounts of a-actinin. a-Actinin was detected with an anti-a-actinin monoclonal antibody and a peroxidase-conjugated secondary antibody. The ELISA shown is a representative of 4 separate experiments done in triplicate. E. CesT dos not bind a-actinin. To maintain Tir solubility in the purification process, its chaperone, CesT, is purified with Tir. To determine that binding to a-actinin of the TirCesT complex is not mediated through CesT, an ELISA was performed using 1 jig/well purified CesT overlaid on 1 [ig/well a-actinin, Tir and intimin.  116  a-actinin. nM  a-actinin  117  Fig. 6.5 Tir binds a-actinin through its N-terminal 1-200 amino acids. ELISA detection of truncated Tir interactions with a-actinin. a-Actinin was incubated on an Immunlon-2 96 well plate and overlaid with Tir (residues 1-391), Tir (residues 1-200), Tir (residues 202-391), Tir (residues 202-550) and Tir (residues 392-550). Tir was detected with monoclonal anti-Tir antibodies and a peroxidase-conjugated secondary antibody. The ELISA shown is representative of 7 independent experiments done in triplicate +/- standard errors of means (SEM).  118  • N term • N term+IBD • IBD IBD+C term •Cterm N o Tir D  D  Tir truncations  119  Fig. 6.6: a-Actinin co-immunoprecipiates with Tir from EPEC infected HeLa cells. Uninfected and EPEC-infected HeLa cells were cross-linked with DSS, lysed, and immunoprecipitated for Tir. HeLa immunopreciptates (IP) and post-immunoprecipitates (supernatants; post-IP) were probed for Tir and a-actinin. a-Actinin is found in the IP of EPEC-infected cells only.  120  Uninfected  EPEC infected  IP post-IP IP  post-IP 90kDa 78kDa  anti-Tir  lOOkDa  anti-a-actinin  6.8 a-Actinin is functional in the EPEC pedestal HeLa cells were transfected with GFP-a-actinin and infected with EPEC for 5 hours. a-Actinin was specifically recruited to the tip and along the length of the pedestal in the identical pattern as endogenous a-actinin staining (Fig. 6.7A). However, these pedestals differed from untransfected cells or GFP-transfected cells in that they were twofold longer (Fig. 6.7B). These data suggest that a-actinin plays a functional role in EPEC pedestal formation.  6.9 The actin binding domain and spectrin like repeats are recruited to the pedestal To determine which part of a-actinin was important for its recruitment and function in the pedestal, GFP constructs of different a-actinin domains were made (Fig. 6.8E). HeLa cells were transfected with the GFP-ABD, G F P-SLR, GFP-CBD, and GFP alone and infected with EPEC for 5 hours. GFP-ABD and GFP-SLR were both recruited to the pedestal, but there was no effect on pedestal formation or length (Fig. 6.8a and b). GFP-CBD and GFP alone did not specifically recruit to the pedestal (Fig 6.8c and d). These results suggest that the A B D and SLR may be involved in recruiting a-actinin to the pedestal, but that full-length a-actinin may be needed for proper folding and function in the cell.  122  Fig.6.7.  Overexpression of a-actinin results in a two-fold increase in pedestal  length. (A) HeLa cells were transfected with GFP-a-actinin, infected with EPEC for 5 hours, and stained for phalloidin (red) or DAPI (blue). (B) Pedestal lengths from GFP-aactinin transfected, GFP only transfected, or untransfected HeLa cells were measured using Northern Eclipse software. The lengths, in p,m, are an average from 50 pedestals/ experimental condition. Error bars represent standard deviations of the averages from 3 independent experiments. Pedestals from GFP-a-actinin transfected cells are significantly different (p < 0.05) from GFP-only or untransfected cells as calculated by Student t-tests.  123  Fig6.8. Domain mapping of a-actinin recruitment to the pedestal. a-Actinin was truncated into three domains, the actin binding domain, the spectrin-like repeats, and the calcium binding domain ( E ) . The ABD and SLR are recruited to the pedestal (A and B) whereas the CBD and GFP alone are not recruited ( C and D). Scale bars represent 5 um Arrows indicate sites of EPEC pedestals.  125  126  6.10 Discussion Delivery of Tir via the type III secretion system is a critical step in pedestal formation. How Tir directs this major cytoskeletal rearrangement, however, was not well characterized. It was believed that Tir linked to the cytoskeleton and functioned as an anchor for bacterial attachment. This study is the first to document a direct binding of Tir with a cytoskeletal component, a-actinin. This interaction was shown by several methods: colocalization in immunofluoresence of Tir and a-actinin; affinity chromatography with Tir and HeLa cell lysates; Far Western and ELISA analysis with purifed Tir and a-actinin; and co-immunoprecipitation of a-actinin and Tir. a-Actinin bound to purified Tir that was not tyrosine phosphorylated. Domain mapping of Tir-aactinin interactions demonstrated that the binding occurred within the first 200 amino acids of Tir, in a region that is not tyrosine phosphorylated. More evidence for Tir-aactinin binding was recently published, using gel overlay and GST-Tir pulldowns from Caco-2 cell lysates (Freeman, et al., 2000). Furthermore, a-actinin was recruited beneath the EPEC Atir/tirYAlAF mutant, independent of Tir tyrosine phosphorylation (Fig. 3.3). Other cytoskeletal proteins involved in pedestal formation were no longer recruited in the absence of Tir tyrosine phosphorylation (Fig3.3). These results suggest that Tir is a multifunctional protein. It binds intimin in the EPEC outer membrane, which focuses Tir beneath the pathogen. Secondly, it mediates a stable cytoskeletal anchor through its cytosolic N terminus by directly binding to a-actinin. Thirdly, Tir probably mediates active actin polymerization processes through its C terminus, dependent on Tir tyrosine phosphorylation and N-WASP/Arp2/3 (Kalman, et al., 1999). The implications and potential uses of this multi-functional binding will be discussed in detail in Chapter 7. The results presented here give us new insight into how the EPEC pedestal is formed during infection, and provides a new role for Tir as a multifunctional protein. Tir not only binds the EPEC ligand intimin, but anchors the bacterium directly to the host cytoskeleton. 127  Chapter 7: Discussion At the onset of this thesis work, we knew that pedestals were implicated in EPEC-mediated disease. All clinical EPEC strains isolated from infected patients formed the characteristic attaching and effacing lesion. Few details, however, were available on its structure, composition and function. The work presented here has focused on the identification, characterization and functional role of cytoskeletal proteins in the pedestal. This research has expanded our knowledge about the structure of EPEC pedestals and allowed us to consider its role both in EPEC infection and as a cell biology tool for studying the actin cytoskeleton. 7.1 The pedestal: A cell biologist's friend (unless taken internally) Pathogenic E. coli may be used as a model system to study signaling to the actin cytoskeleton across the plasma membrane in response to external stimuli. EPEC provides a simple, reproducible system to induce and manipulate actin polymerization. At the center of this model is Tir. Tir delivery, insertion and tyrosine phosphorylation in the epithelial plasma membrane provides the starting point for membrane-based actin polymerization events. Binding of intimin to Tir acts as the external stimuli required to "snap" the cytoskeletal components together into a functional structure. Tir's unique hairpin structure allows both the N and C termini to act as two distinct domains through which to signal to the cytoskeleton - one domain acting much like a f3l integrin and the other functioning in receptor tyrosine kinase signaling. 7.1.1 A model for focal adhesions? It is interesting that so many focal adhesion proteins were localized to the pedestal in the absence of (31 integrins (Chap 3). This suggests that Tir functions in a manner analogous to pi integrins. There are several lines of evidence to support this hypothesis. First, both Tir and pi integrin span the plasma membrane and upon binding of their extracellular ligand, signal to the actin cytoskeleton. Secondly, Tir links to the  128  cytoskeleton through a-actinin as do p\ integrins in focal adhesion complexes (Otey et al., 1993) (Goosney, et al., 2000). Third, Tir has recently been shown to bind talin, a cytoskeletal protein that directly binds pi integrins and functions in focal adhesions (Freeman, et al., 2000). Both a-actinin and talin co-localize at the tip of the pedestal and bind Tir in vitro and in the cell (Freeman, et al., 2000, Goosney, et al., 2000). That Tir binds both at its N-terminus suggests that it contains a structural homology to the pi cytoplasmic tail, although to date no sequence homology has been identified (Freeman, et al., 2000, Goosney, et al., 2000). Fourth, EPEC intimin binds Tir to initiate signaling inside the host cell. Pi integrins are also bound by the extracellular matrix (ECM) to initiate signaling cascades (Geiger, et al., 1995). This outside-in signaling in both cases leads to the recruitment of signaling molecules and other cytoskeletal proteins to the plasma membrane to build a focal contact or EPEC pedestal. Fifth, a-actinin recruitment to both Tir and Pi integrins occurs independently of their phosphorylation (Miyamoto, 1995) (Goosney, et al., 2000). It is interesting to speculate that different signaling molecules may be recruited to the pedestal in different sequences depending on Tir ligand occupancy, clustering, tyrosine phosphorylation, and actin polymerization. This appears to be the case, since a subset of cytoskeletal proteins, including N-WASP and the Arp2/3 complex, require Tir tyrosine phosphorylation, while others, like a-actinin and talin, recruit independently of Tir tyrosine phosphorylation and actin polymerization. A sixth similarity between pedestals and focal adhesions is that even a modest overexpression of a-actinin results in elongation in both structures. In EPEC pedestals, transfection with wild type a-actinin resulted in a doubling of pedestal length. In focal adhesions, wild type a-actinin transfection resulted in an increase in both the number and size of focal adhesions (Gluck and Ben-Ze'ev, 1994). A seventh similarity lies with the recruitment of the motor proteins myosin light chain and tropomyosin. In both structures, MLC and tropomyosin are recruited to the  129  base, but are excluded from the tips (Geiger, et al., 1995, Sanger, et al., 1996). This is also the case for microvilli (Mooseker, 1985). An eighth similarity lies in the phosphoinositol signaling pathways used in both systems. PIP is recruited to the tips of both focal adhesions and pedestals (Fukami, et 2  al., 1994) (DeVinney, unpub. obs). Many proteins recruited to the tip of the pedestal bind PIP , including talin, vinculin and a-actinin (Fukami, et al., 1994). It may be that 2  PIP -binding of the structural components of focal contacts and pedestals affects their 2  binding to each other (Gilmore and Burridge, 1995). This could provide a means of cross-talk between these structures and the plasma membrane. Interestingly, PLCy, the enzyme that cleaves PIP into IP and DAG, and PKC, the enzyme that is activated by 2  3  IP , both colocalize in focal adhesions, and both are activated in EPEC infection (Kenny 3  and Finlay, 1997, Crane and Oh, 1997, Hyatt, et al., 1994, McBride, et al., 1991). A ninth similarity between EPEC pedestals and focal adhesions is found in the intimin structure. Intimin shares a high degree of homology with the Yersinia outer membrane protein, invasin (Frankel, et al., 1995). Invasin mediates Yersinia binding to Pi integrins on host cells, resulting in the uptake of Yersinia into the intestinal cell during infection. Intimin has been shown to bind Pi integrins (Frankel, et al., 1996), although this binding is not physiologically relevant (Liu, et al., 1999). The Tir intimin-binding area (TIBA, or IBD) is homologous to the extracellular matrix-binding domain of integrins (Kenny, 1999), which may explain why intimin can bind pi integrins. Finally, it should be noted that several groups have found that EPEC infection of cultured epithelial cells results in the loss of focal adhesions over time (Goosney, unpub. obs., Freeman, et al., 2000, Rosenshine et al, unpub. obs.). This suggests that EPEC may be competing with existing focal adhesions for pools of focal adhesion proteins. Why should EPEC mediate integrin-type signaling? One explanation is that it provides a stable attachment from the bacterium to the host cytoskeleton through the plasma membrane, much like integrins mediate stable attachments for mammalian cells  130  to the ECM. Another reason, as suggested by the observations presented above, may be to recruit cytoskeletal proteins away from tight junctions and focal adhesions to increase the paracellular permeability and induce diarrhea (see section 7.3 for the role of the pedestal in diarrhea). 7.1.2 A model for receptor tyrosine kinase signaling? Although the EPEC Tir N-terminus has similarities with (31 integrins, Tir's Cterminus is strikingly similar to sequences seen in receptor tyrosine kinase signaling. Indeed, the sequence flanking EPEC Tir's Tyrosine474 (EHIYDEVA) is almost identical with the c-src optimal substrate sequence. Although pp.60src was not found in the EPEC pedestal, it is tempting to speculate that another src-family member maybe involved in phosphorylating Tir given this high degree of similarity with src substrate sequences. To date, the identity of Tir's kinase remains unknown. Interestingly, this same sequence harbours a binding sequence (YDEV) for the adaptor protein Nek, which is recruited specifically to EPEC pedestal tips in a Tir phosphotyrosine-dependent manner (Chap 3). Recently, Nek has been shown to directly bind EPEC Tir at the C terminus through its SH2 domain (Gruenheid et al., unpub. obs). This binding requires Tyr474 to be phosphorylated. Overexpression of Nek dominant negative alleles results in inhibition of pedestal formation. Furthermore, a fibroblast cell line deficient in Nek and Grb4 (an adaptor highly homologous to Nek) cannot make pedestals when infected with EPEC. In the absence of Nek, both N-WASP and the ARP2/3 complex are no longer recruited to the site of EPEC attachment, suggesting that Nek is upstream of these effectors (Gruenheid unpub. obs). As shown in Chap 4, WASP/N-WASP plays a critical role in pedestal formation. The GTPase binding domain (GBD) of WASP is responsible for the recruitment of WASP to the site of pedestal formation. Deletion of this region leads to loss of recruitment to the tip of the pedestal, and the WASP GBD alone recruits specifically to the pedestal tip. A newly identified GTPase, Chp, was recruited  131  specifically to the pedestal tip (Appendix 2), but to date, has not be shown to be functional in pedestal formation. In this respect, although the GBD is necessary and sufficient for WASP localization, other domains may also involved because the GBD did not efficiently block pedestal formation on its own, as seen with the WASPAC or NWASPAA. Given that Nek is upstream of N-WASP, it is likely that other domains are involved in bringing N-WASP to the pedestal tip. One such candidate is WASPinteracting protein (WIP; Ramesh, et al., 1997). WIP is an adaptor protein that binds both Nek and N-WASP, as well as other key actin regulators like profilin (Ramesh, et al., 1999). Nek signaling through WIP to N-WASP has already been demonstrated in another system, vaccinia virus actin tails, which will be discussed in detail below (Frischknecht, et al., 1999, Moreau, et al., 2000). It is interesting that EPEC mimics both integrin-type signaling and receptor tyrosine kinase signaling, as the two types of receptors often work synergistically in normal mammalian cell functioning. Both types of receptors are associated with the cytoskeleton and integrated within focal adhesion complexes (Plopper, et al., 1995). Focal adhesions, therefore, provide more than adherence to a substrate - they provide spatial organization to other membrane receptors and a local concentration of cytosolic effector proteins like adaptors (reviewed in (Aplin, et al., 1999)). EPEC may be taking advantage of this normally occurring collaboration between receptors to initiate a full signaling cascade from one protein. 7.1.3 Two different cytosolic Tir domains, two distinct populations of actin? It is tempting to speculate that the EPEC pedestal is comprised of two distinct populations of actin filaments. There are several lines of evidence that support this hypothesis. First, there are dramatic differences between the signaling events occurring through the N-terminus of Tir and those occurring through the C-terminus. Several cytoskeletal proteins, including talin and a-actinin, are recruited to the site of bacterial attachment independently of Tir tyrosine phosphorylation at the C-terminus. Both these  132  proteins were demonstrated to bind directly to the N-terminus. Phosphorylation of the C-terminus resulted in recruitment of the actin polymerization machinery: Nek, NWASP, and the Arp2/3 complex. Only when Tir is tyrosine phosphorylated do we see extended pedestals made. Furthermore, N-WASP and Arp2/3 are not recruited in the Nck-Grb4- fibroblasts, but a-actinin is (Gruenheid, unpub. obs.). This strongly suggests that the events occurring at the C terminus dependent on Tir tyrosine phosphorylation are separate from the events occurring at the N-terminus. A second piece of evidence for the bipartite functioning of Tir stems from experiments using the actin polymerization inhibitor cytochalasin D. Cytochalasin D functions by preventing the addition of actin monomers to an existing actin filament. In the case of actin treadmilling, the addition of cytochalasin D would disassemble the actin structure. In the case of a stable, non-treadmilling actin population, cytochalasin D would not affect the existing structure. In the case of EPEC pedestals, cytochalasin D does not disassemble pedestals as actin is maintained beneath the bacteria throughout treatment (Sanger, et al., 1996). However, treatment does inhibit pedestal twisting and movement (see below for details on pedestal movement and function). More evidence for the existence of two distinct actin populations comes from actin incorporation assays (Appendix 3). EPEC pedestals are treated with rhodamine-labeled monomeric actin, which is allowed to incorporate into the pedestal (providing that active polymerization events are occurring in the pedestal). The monomeric actin is incorporated at the tip of the pedestal. EPEC pedestal length is fairly constant in a given experiment, suggesting that disassembly at the base of the pedestal must be occurring if addition to the filament is occurring at the tip (Sanger, et al., 1996). Collectively, the results presented here suggest that the EPEC pedestal is comprised of stable and treadmilling actin filaments. The N-terminus is likely anchoring Tir to the stable filaments through a actinin and talin, :  whereas the C-terminus is initiating actin polymerization following tyrosine  133  phosphorylation and binding of Nek. This leads to downstream recruitment of N-WASP and the Arp2/3 complex (Fig. 7.1).  134  EPEC  Fig7.1. Proposed model of N-terminal and C-terminal signaling in EPEC infection. See text for details.  135  7.2 Mimicry of host cell function by other pathogens EPEC is one of several pathogens, bacterial and viral, that hijack the host cell actin cytoskeleton during their infection process. Their use in understanding various aspects of cellular regulation of actin dynamics has been critical. Cell biologists have traditionally used drugs or genetic manipulation to induce changes in cell behaviour, but this requires specific knowledge of the protein of interest and some understanding of function. Pathogens provide the ideal tools for analysing specific aspects of a cytoskeletal system. They are (often) easily genetically manipulated and provide an inducible system to invoke the desired response in the host. The pathogens most commonly used to manipulate the actin cytoskeleton are pathogenic E. coli (EPEC and EHEC) vaccinia virus, Listeria and Shigella (Fig.7.2). It is interesting to see how each pathogen manipulates similar components of the actin cytoskeleton in a different manner to survive and persist in the host. 7.2.1 E H E C : Pedestals without Tir tyrosine phosphorylation Until very recently, it was believed that EPEC and EHEC were signaling the same way to the cytoskeleton to build a pedestal. The discovery that EHEC Tir was not tyrosine phosphorlyated in host cells was the first clue that they could be hijacking the actin cytoskeleton in different ways (Deibel, et al., 1998, DeVinney, et al., 1999). This was followed by the information that EHEC did not recruit the adaptor proteins Grb2, Crkll or Nek to its pedestal (Chap 3; Goosney et al. 2001). Of particular interest was Nek, the adaptor that has now been shown to bind EPEC Tir through its SH2 domain directly at Tir's phosphotyrosine in the C-terminus. Nek, although critical in mediating EPEC pedestal formation, is not necessary for EHEC pedestals (Gruenheid, unpub.obs). How, then, is EHEC recruiting the actin polymerization machinery N-WASP and the Arp2/3 complex to the pedestal? It could be directly binding N-WASP to its C-terminus, or it could be using its own adaptor, a bacterial effector, to mediate recruitment. It is  136  interesting to speculate that the glycine-rich region in the C-terminus of EHEC Tir could mediate direct binding of N-WASP, as is the case for Shigella (see section 7.2.3). 7.2.2 Model for receptor tyrosine kinase signaling: Vaccinia virus Vaccinia virus is the prototype of the orthopox genus of large enveloped double stranded DNA viruses (Moss, 1990). It has a very complex morphogenesis that results in two distinct infectious forms, the intracellular mature virus (IMV) and the intracellular enveloped virus (IEV)(Moss, 1990). The IEV stems from an extra wrapping step of the IMV by a membrane derived from the trans-Golgi network (Schmelz, et al., 1994). The IEV then hijacks the actin cytoskeleton to form actin tails with which to propel itself through the cytoplasm to the plasma membrane, where its outer membrane fuses with the host plasma membrane, releasing the virus (Cudmore, et al., 1995). Actin tail formation relies on the vaccinia integral membrane protein A36R (Frischknecht, et al., 1999, Rottger, et al., 1999). A36R must be phosphorylated on residue Tyrl 12 to induce actintail formation (Frischknecht, et al., 1999). As with EPEC Tir, the sequence surrounding Tyrl 12 is highly homologous to the substrate seqeunce for the src family kinases and for Nek SH2 binding. Indeed, src appears to be involved, as tail formation is inhibited in srefamily kinase inhibitor-treated cells and in sre-family kinase-deficient cells (Frischknecht, et al., 1999). Following Tyrl 12 phosphorylation, Nek binds to the region surrounding the phosphorylated residue. This recruits a WIP (WASP-interacting protein)/N-WASP complex which in turn activates the Arp2/3 complex, leading to actin nucleation and tail formation (Moreau, et al., 2000). Given the high degree of similarities between vaccinia and EPEC signaling, it is tempting to believe that the sre-family kinases are involved in Tir's Tyr474 phosphorylation. 7.2.3 Models for Iamellipodia: Listeria monocytogenes and Shigella flexneri Listeria monocytogenes is a food-borne pathogen that causes listeriosis, a disease that results in menigo-encephalitis, abortion, and death in immuno-compromised patients (Cossart and Lecuit, 1998). It enters the host cell, escapes its vacuole and resides in the  137  host cytoplasm. It uses actin-based motility (ABM) to move about the host cytoplasm and propel itself into neighboring host cells(Cossart, 1995). It has an outer membrane protein, ActA, that is necessary and sufficient to drive the ABM (Sheehan, et al., 1995) Smith, et al., 1995). ActA is polarized to one end of the bacterium, allowing unidirectional ABM to occur (Smith, et al., 1995). It binds VASP, Mena, and Arp2/3 directly (Chakraborty, et al., 1995, Gertler, et al., 1996, Welch, et al., 1998). ActA harbours a sequence similarity (residues 146-150) to the C-terminus of WASP-family proteins, which mediates binding to the Arp2/3 complex (Skoble, et al., 2000, Zalevsky, et al., 2000). Deletion of this basic region results in loss of Arp2/3 recruitment and inhibition of actin tail formation.  ActA also has a proline-rich region that binds VASP,  which, although not essential for motility, enhances the rate of motility in v/fro(Chakraborty, et al., 1995, Smith, et al., 1996). This enhancement could be due to the recruitment of profilin-actin complexes by VASP to the site of actin polymerization behind ListeriaCTheriot, et al., 1994). Shigella species are gram-negative enteric pathogens that cause bacillary dysentery (Nhieu and Sansonetti, 1999). They invade intestinal epithelial cells and escape the vacuole in a manner analogous to Listeria. Upon escape from the vacuole, Shigella initiates actin-based motility through its outer protein IcsA. Although IcsA shares no sequence homology with ActA from Listeria, it is required and sufficient for Shigella motility (Goldberg and Theriot, 1995, Kocks, et al., 1995). IcsA binds vinculin and N-WASP in vitro, but only N-WASP is essential in vivo (Egile, et al., 1999, Moreau, et al., 2000, Suzuki, et al., 1996). N-WASP is recruited to the glycine-rich repeats of IcsA through its CRIB domain (Egile, et al., 1999, Suzuki, et al., 1998). Given that the small GTPases are not localized to existing actin tail, IcsA may be mimicking GTPases like Cdc42 in its recruitment of N-WASP. However, it is likely that Cdc42 may play some role in Shigella ABM. Cdc42 enhances motility when injected into Shigellainfected cells and its inhibition blocks bacterial movement (Suzuki, et al., 2000).  138  Furthermore, it does localize to Shigella at the onset of actin polymerization, but not in existing tails (Suzuki, et al., 2000). Therefore, Cdc42 may only be involved in the initial recruitment of N-WASP to IcsA, not in the continuance of Shigella ABM.  139  Extracellular  actin  Intracellular  Arp2/3  Fig.7.2. Comparison of pathogens using similar components of the actin cytoskeleton for their actin-based motility in the host. See text for details.  140  7.3 The pedestal: Role in disease? EPEC provides an excellent model for studying actin dynamics at the plasma membrane. It is easily genetically manipulated to provide a simple, inducible system to study the mechanics of actin polymerization at the plasma membrane in response to external stimuli. However, it did not evolve to be a useful model for cell biologists. What, then, is the purpose of the pedestal? At every conference, every seminar, and even most Finlay lab meetings, this same question is asked and met with the same response much speculation, but nothing definitive. We know it is important in disease, as only strains capable of making pedestals are virulent. The actual mechanism of this disease production, however, is still unclear. Discussed below are possible answers to the dreaded "what are pedestals good for" question, assimilating the research I've done throughout my Ph.D. with what I've read and discussed with others in the field. One view is that the pedestal is an aborted phagocytic cup. EPEC has been shown to block its own phagocytosis by macrophages in tissue culture (Appendix 1; Goosney et al. 1999). It has been suggested that during EPEC infection, bacteria attaching to the phagocytic M cells prevent their uptake by forming this actin-rich structure. Although extracellular EPEC attached to phagocytes do indeed have actin recruited beneath them, recent evidence suggests that this anti-phagocytic phenotype is distinct from pedestal formation (Celli, unpub. obs). Another view is that the pedestal is formed to reduce the absorptive surface area of the intestinal epithelium. By effacing the microvilli (through a currently unidentified mechanism) in the area surrounding the pathogen and replacing them with a much larger pedestal structure, it is possible that there is a significant reduction in the surface area. This could explain some of the chronic diarrhea associated with EPEC infections. The loss in absorption could possibly provide EPEC with more nutrients from the lumen.  141  Yet another explanation for pedestals is that they may affect the rest of the cellular cytoskeleton, such as tight junctions. Recruitment of actin and various components of adherens junctions to the pedestal (and away from tight junctions and E C M attachments) may disrupt the integrity of these structures and promote paracellular permeability leading to diarrhea (Spitz, et al., 1995). A similar idea of disrupting sites of cell-cell or cell-ECM attachment is based on the theory of tensegrity (Ingber, 1993). Briefly, tensegrity suggests that the entire cytoskeleton is an interconnected network of compression-resistant struts (the microtubule system) and tensile elements (actin filaments). Tension applied to one part of the network is relayed throughout the system and the system readjusts its shape to compensate. With pedestals, EPEC may "pull" the actin cytoskeleton in the direction of the lumen creating a tension that is relayed throughout the intestinal cell. This may disrupt tight junction integrity and lead to in increase paracellular permeability and diarrhea. A fifth theory as to the role of the pedestal is that it allows the bacterium to be tethered to the intestinal cell and move about with the flow of the intestinal lumen. EPEC pedestals are not stationary structures in intestinal epithelial cells, but instead move about on the cell surface. EPEC surf along the cell surface at speeds ranging from 0.03-0.07 u,m/second, significantly slower than rates seen for intracellular Listeria A B M (0.2 am/second) (Sanger, et al., 1996) (Theriot, 1992). The pathogen also remains tethered at the base of the pedestal and moves, twisting and turning, sometimes elongating 0.01 0.02 urn (Sanger, et al., 1996). This movement is inhibited by the addition of cytochalasin D, suggesting that it needs active actin polymerization to move. It has been suggested that this movement allows EPEC to "go with the flow" yet remain firmly attached to the cell (Freeman, et al., 2000). It is my belief that the pedestal may be functioning in a number of ways in the host. It most definitely provides a strong attachment to the cell, and the surfing movement observed is probably not a lab artifact. Given the disruption of tight junctions,  142  the loss of attachment to the ECM in tissue culture, and the reduction in absorptive surfaces, I suspect the pedestal is affecting these cellular processes as well. 7.4 Future directions It is clear that much characterization of the structure and function of the EPEC pedestal remains to be done. There are several avenues for future research that can be pursued, ranging from the molecular aspects of protein-protein interactions in the pedestal to the in vivo implications of EPEC infection. Listed below are several questions that should be addressed in the future. 1. What is the identity of the kinase phosphorylating Tir? Although some of the downstream effectors of pedestal formation have been identified and characterized (Nek, N-WASP, Arp2/3), the initial step at the C terminus, tyrosine phosphorylation of Tir, is still unknown. Homology exists between the region surrounding the Tyr474 and the substrate-sequence of sre-family kinases, suggesting members of this group may be the best candidates to investigate. Given the functional redundancy of the sre kinases, a sre family-specific inhibitor, such as PP1 or PP2 (Calbiochem) may be useful in initially determining the role of any sre kinases in pedestal formation. 2. How is Nek recruiting N-WASP to the pedestal? Following phosphorylation of the Tyr474 region and binding by Nek, N-WASP and the Arp2/3 complex are recruited to the pedestal through Nek in a currently unidentified manner. It will be interesting to see how these proteins are recruited to the pedestal through Nek. Is N-WASP being recruited through another domain in addition to the GBD? Is Chp or a Chp-like GTPase involved in pedestal formation, and if so, is it downstream of Nek? Is WIP in the cascade of signals? Several tools are now available to tease apart this signaling cascade, including mutants in N-WASP that no longer bind Nek or WIP, GFP-WIP constructs, GFPNckSH2 and SH3 domains (deletions or the independent domains), and numerous constructs containing other adaptor proteins (to address the involvement of Grb2, Crkll, and She).  143  3. How is E H E C recruiting N-WASP and the Arp2/3 complex to its pedestal? The adaptor Nek is not involved in EHEC pedestal formation, but N-WASP and the Arp2/3 complex are still recruited. As seen with Shigella, EHEC Tir could be binding N-WASP directly. EHEC Tir contains a glycine-rich region at its C-terminus that may be responsible for N-WASP binding and recruitment. Shigella IcsA glycine-rich region is responsible for mimicking Cdc42 and recruiting N-WASP through its GBD. EHEC could be hijacking the signaling cascade in the same manner. Direct binding assays (ELISAs, gel overlays, co-immunoprecipitations) could be performed with the EHEC Cterminus and purified N-WASP or HeLa cell lysates to determine if N-WASP is binding directly. 4. What are the roles of other key actin regulators in the cell? For instance, VASP has been implicated in Listeria motility, and is found at the tip of the EPEC and EHEC pedestals. Zyxin and LPP both function in focal adhesions, and provide a link from the cytoskeleton to the nucleus(Petit, et al., 2000). Reagents, including dominant negative constructs and knockout cell lines, have recently been made available to study the roles of each of these proteins in the EPEC pedestals. 5. What is the arrangement of actin in the pedestal? There are currently no data on the ultrastructure of the pedestal. Are the actin filaments bundled in a parallel fashion like microvilli? Are the fast-growing ends of actin oriented towards the plasma membrane beneath EPEC? Is significant branching occurring in the structure due to the presence of the Arp2/3 complex? Is the pedestal a mixed population of treadmilling, branched filaments and stable, parallel ones? To address these questions, ultrastructure studies have to be done, using primarily transmission electron microscopy (TEM). These studies have been done to determine the structures of Listeria, Shigella, vaccinia, and another pathogen that alters actin in the host, Rickettsia. Other studies should include actin incorporation assays to determine not only the sites of incorporation but the rate of treadmilling as well. Does the rate of treadmilling correspond with the rate of movement  144  seen in EPEC surfing? Determining the nature and role of the actin population in the EPEC pedestal will aid in determining the roles of actin-associated proteins in the pedestal.  6. What is happening in a more physiologically relevant, polarized cell model? Currently, most of the work done in EPEC pedestal formation occurs in HeLa cells, an unpolarized epithelial cell line. Although this provides an uncomplicated means of identifying and characterizing host proteins involved, it may not present the full story. For instance, what is happening to the microvilli during infection? How does effacement occur? What are the roles of microvillar proteins in pedestal formation? How does pedestal formation alter adherens junctions? These questions can be initially addressed using a polarized tissue culture cell line, such as the intestinally-derived Caco-2. 7. What is the role of Tir tyrosine phosphorylation in vivo? Work done in a polarized cell can then be taken further to address what is happening in vivo. With the Tir Y474F mutant now available and characterized in tissue culture, it will be exciting to determine what effects it has during infection. There are obviously numerous lines of research that can be pursued with this project. It my hope that much of the groundwork has been laid for addressing these questions, and that in the not-too-distant future, a greater understanding of the EPEC and EHEC pedestals will be achieved.  145  Bibliography 1.  Abe, A., U. Heczko, R. G. Hegele, and B. Brett Finlay. 1998. Two  enteropathogenic Escherichia coli type III secreted proteins, EspA and EspB, are virulence factors. J Exp Med. 188:1907-16. 2.  Abo, A. 1998. Understanding the molecular basis of Wiskott-Aldrich syndrome.  Cell Mol Life Sci. 54:1145-53. 3.  Aplin, A. E., A. K. Howe, and R. L. Juliano. 1999. Cell adhesion molecules,  signal transduction and cell growth [see comments]. Curr Opin Cell Biol. 11:737-44. 4.  Arber, S., F. A. Barbayannis, H. Hanser, C. Schneider, C. A. Stanyon, O.  Bernard, and P. Caroni. 1998. Regulation of actin dynamics through phosphorylation of cofilin by L I M - kinase [see comments]. Nature. 393:805-9. 5.  Aronheim, A., Y. C. Broder, A. Cohen, A. Fritsch, B. Belisle, and A. Abo.  1998. Chp, a homologue of the GTPase Cdc42Hs, activates the JNK pathway and is implicated in reorganizing the actin cytoskeleton. Curr Biol. 8:1125-8. 6.  Bacallao, R., C. Antony, C. Dotti, E . Karsenti, E . H. Stelzer, and K. Simons.  1989. The subcellular organization of Madin-Darby canine kidney cells during the formation of a polarized epithelium. J Cell Biol. 109:2817-32. 7.  Bain, C , R. Keller, G. K. Collington, L. R. Trabulsi, and S. Knutton. 1998.  Increased levels of intracellular calcium are not required for the formation of attaching and effacing lesions by enteropathogenic and enterohemorrhagic Escherichia coli. Infect Immun. 66:3900-8. 8.  Balda, M. S., J. A. Whitney, C. Flores, S. Gonzalez, M. Cereijido, and K.  Matter. 1996. Functional dissociation of paracellular permeability and transepithelial electrical resistance and disruption of the apical- basolateral intramembrane diffusion barrier by expression of a mutant tight junction membrane protein. J Cell Biol. 134:103149.  146  9.  Baldini, M . M . , J . B . Kaper, M . M . Levine, D . C . Candy, and H . W . Moon.  1983. Plasmid-mediated adhesion in enteropathogenic Escherichia coli. J Pediatr Gastroenterol Nutr. 2:534-8. 10.  Baldwin, T . J . , S. F. Brooks, S. Knutton, H . A . Manjarrez Hernandez, A .  Aitken, and P. H . Williams. 1990. Protein phosphorylation by protein kinase C in HEp2 cells infected with enteropathogenic Escherichia coli [published erratum appears in Infect Immun 1990 Jun;58(6):2024]. Infect Immun. 58:761-5. 11.  Baldwin, T . J . , M . B. Lee-Delaunay, S. Knutton, and P. H . Williams. 1993.  Calcium-calmodulin dependence of actin accretion and lethality in cultured HEp-2 cells infected with enteropathogenic Escherichia coli. Infect Immun. 61:760-3. 12.  Baldwin, T . J . , W . Ward, A . Aitken, S. Knutton, and P. H. Williams. 1991.  Elevation of intracellular free calcium levels in HEp-2 cells infected with enteropathogenic Escherichia coli. Infect Immun. 59:1599-604. 13.  Bamburg, J . R. 1999. Proteins of the ADF/cofilin family: essential regulators of  actin dynamics. Annu Rev Cell Dev Biol. 15:185-230. 14.  Barthold, S. W., G. L . Coleman, P. N . Bhatt, G. W . Osbaldiston, and A. M .  Jonas. 1976. The etiology of transmissible murine colonic hyperplasia. Lab Anim Sci. 26:889-94. 15.  Bartles, J . R., L . Zheng, A. L i , A. Wierda, and B. Chen. 1998. Small espin: a  third actin-bundling protein and potential forked protein ortholog in brush border microvilli. J Cell Biol. 143:107-19. 16.  Bass, M . D., B. J . Smith, S. A . Prigent, and D. R. Critchley. 1999. Talin  contains three similar vinculin-binding sites predicted to form an amphipathic helix. Biochem J. 341:257-63. 17.  Beaulieu, J . F. 1999. Integrins and human intestinal cell functions. Front Biosci.  4:D310-21.  147  18.  Ben-Ami, G., V. Ozeri, E. Hanski, F. Hofmann, K. Aktories, K. M . Hahn, G.  M . Bokoch, and I. Rosenshine. 1998. Agents that inhibit Rho, Rac, and Cdc42 do not block formation of actin pedestals in HeLa cells infected with enteropathogenic Escherichia coli. Infect Immun. 66:1755-8. 19.  Bendori, R., D. Salomon, and B. Geiger. 1989. Identification of two distinct  functional domains on vinculin involved in its association with focal contacts. J Cell Biol. 108:2383-93. 20.  Blanchard, A., V. Ohanian, and D. Critchley. 1989. The structure and function  of alpha-actinin. J Muscle Res Cell Motil. 10:280-9. 21.  Bourguignon, L . Y., V. B. Lokeshwar, J. He, X. Chen, and G. J .  Bourguignon. 1992. A CD44-like endothelial cell transmembrane glycoprotein (GP116) interacts with extracellular matrix and ankyrin. Mol Cell Biol. 12:4464-71. 22.  Bray, J. 1945. isolation of antigenically homogeneous strains of Bact. coli  nepaolitanum from summer diarrhoea of infants. J. Pathol. Bacteriol. 57:239-247. 23.  Brent, R., and M . Ptashne. 1980. The lexA gene product represses its own  promoter. Proc Natl Acad Sci U S A . 77:1932-6. 24.  Bretscher, A., and K. Weber. 1980. Villin is a major protein of the microvillus  cytoskeleton which binds both G and F actin in a calcium-dependent manner. Cell. 20:839-47. 25.  Broschat, K. O., R. P. Stidwill, and D. R. Burgess. 1983. Phosphorylation  controls brush border motility by regulating myosin structure and association with the cytoskeleton. Cell. 35:561-71. 26.  Buday, L . 1999. Membrane-targeting of signalling molecules by SH2/SH3  domain-containing adaptor proteins. Biochim Biophys Acta. 1422:187-204. 27.  Burnham, M . R., M . T. Harte, A. Richardson, J. T. Parsons, and A. H.  Bouton. 1996. The identification of pl30cas-binding proteins and their role in cellular transformation. Oncogene. 12:2467-72.  148  28.  Burridge, K., and P. Mangeat. 1984. An interaction between vinculin and talin.  Nature. 308:744-6. 29.  Carlier, M . F., P. Nioche, I. Broutin-L'Hermite, R. Boujemaa, C. Le  Clainche, C. Egile, C. Garbay, A. Ducruix, P. Sansonetti, and D. Pantaloni. 2000. GRB2 links signaling to actin assembly by enhancing interaction of neural WiskottAldrich syndrome protein (N-WASp) with actin-related protein (ARP2/3) complex. J Biol Chem. 275:21946-52. 30.  Chakraborty, T., F. Ebel, E. Domann, K. Niebuhr, B. Gerstel, S. Pistor, C. J.  Temm-Grove, B. M . Jockusch, M . Reinhard, U. Walter, and et al. 1995. A focal adhesion factor directly linking intracellularly motile Listeria monocytogenes and Listeria ivanovii to the actin-based cytoskeleton of mammalian cells. Embo J. 14:131421. 31.  Chatkaeomorakot, A., P. Echeverria, D. N. Taylor, K. A. Bettelheim, N. R.  Blacklow, O. Sethabutr, J. Seriwatana, and J. Kaper. 1987. HeLa cell-adherent Escherichia coli in children with diarrhea in Thailand. J Infect Dis. 156:669-72. 32.  Chen, H., B. W. Bernstein, and J. R. Bamburg. 2000. Regulating actin-filament  dynamics in vivo. Trends Biochem Sci. 25:19-23. 33.  Citi, S., H. Sabanay, R. Jakes, B. Geiger, and J. Kendrick-Jones. 1988.  Cingulin, a new peripheral component of tight junctions. Nature. 333:272-6. 34.  Cobeljic, M., D. Mel, B. Arsic, L . Krstic, B. Sokolovski, B. Nikolovski, E.  Sopovski, M . Kulauzov, and S. Kalenic. 1989. The association of enterotoxigenic and enteropathogenic Escherichia coli and other enteric pathogens with childhood diarrhoea in Yugoslavia. Epidemiol Infect. 103:53-62. 35.  Collington, G. K., I. W. Booth, and S. Knutton. 1998. Rapid modulation of  electrolyte transport in Caco-2 cell monolayers by enteropathogenic Escherichia coli (EPEC) infection. Gut. 42:200-7. 36.  Cossart, P. 1995. Actin-based bacterial motility. Curr Opin Cell Biol. 7:94-101.  149  37.  Cossart, P., and M . Lecuit. 1998. Interactions of Listeria monocytogenes with  mammalian cells during entry and actin-based movement: bacterial factors, cellular ligands and signaling. Embo J. 17:3797-806. 38.  Craig, S. W., and L. D. Powell. 1980. Regulation of actin polymerization by  villin, a 95,000 dalton cytoskeletal component of intestinal brush borders. Cell. 22:73946. 39.  Crane, J. K., and J. S. Oh. 1997. Activation of host cell protein kinase C by  enteropathogenic Escherichia coli. Infect Immun. 65:3277-85. 40.  Crawford, A. W., and M . C. Beckerle. 1991. Purification and characterization  of zyxin, an 82,000-dalton component of adherens junctions. J Biol Chem. 266:5847-53. 41.  Crawford, A. W., J. W. Michelsen, and M . C. Beckerle. 1992. An interaction  between zyxin and alpha-actinin. J Cell Biol. 116:1381-93. 42.  Critchley, D., Flood G. 1999. alpha-actinins, p. 24-27. In T. Kreis, Vale, R (ed.),  Guidebook to the cytoskeletal and motor proteins. Oxford University Press, Oxford, U.K. 43.  Critchley, D. R. 2000. Focal adhesions - the cytoskeletal connection. Curr Opin  Cell Biol. 12:133-9. 44.  Cudmore, S., P. Cossart, G. Griffiths, and M . Way. 1995. Actin-based motility  of vaccinia virus. Nature. 378:636-8. 45.  De Rycke, J., E. Comtet, C. Chalareng, M . Boury, C. Tasca, and A. Milon.  1997. Enteropathogenic Escherichia coli 0103 from rabbit elicits actin stress fibers and focal adhesions in HeLa epithelial cells, cytopathic effects that are linked to an analog of the locus of enterocyte effacement. Infect Immun. 65:2555-63. 46.  deGrado, M., Abe, A., Gauthier, A., Steele-Mortimer, O., DeVinney, R., and  B.B. Finlay. 1999. Identification of the intimin binding domain of Tir of enteropathogenic Escherichia coli. Cellular Microbiol. 1:7-19. 47.  Deibel, C., S. Kramer, T. Chakraborty, and F. Ebel. 1998. EspE, a novel  secreted protein of attaching and effacing bacteria, is directly translocated into infected  150  host cells, where it appears as a tyrosine-phosphorylated 90 kDa protein. Mol Microbiol. 28:463-74. 48.  DeRosier, D. J., and L . G. Tilney. 2000. F-actin bundles are derivatives of  microvilli: What does this tell us about how bundles might form? J Cell Biol. 148:1-6. 49.  DeVinney, R., M . Stein, D. Reinscheid, A. Abe, S. Ruschkowski, and B. B.  Finlay. 1999. Enterohemorrhagic Escherichia coli 0157:H7 produces Tir, which is translocated to the host cell membrane but is not tyrosine phosphorylated. Infect Immun. 67:2389-98. 50.  Donnenberg, M . S., S. B. Calderwood, A. Donohue-Rolfe, G. T. Keusch, and  J. B. Kaper. 1990. Construction and analysis of TnphoA mutants of enteropathogenic Escherichia coli unable to invade HEp-2 cells. Infect Immun. 58:1565-71. 51.  Donnenberg, M . S., and J. B. Kaper. 1991. Construction of an eae deletion  mutant of enteropathogenic Escherichia coli by using a positive-selection suicide vector. Infect Immun. 59:4310-7. 52.  Donnenberg, M . S., C. O. Tacket, S. P. James, G. Losonsky, J. P. Nataro, S.  S. Wasserman, J. B. Kaper, and M . M . Levine. 1993. Role of the eaeA gene in experimental enteropathogenic Escherichia coli infection [see comments]. J Clin Invest. 92:1412-7. 53.  Dytoc, M., L. Fedorko, and P. M . Sherman. 1994. Signal transduction in human  epithelial cells infected with attaching and effacing Escherichia coli in vitro. Gastroenterology. 106:1150-61. 54.  Ebel, F., C. von Eichel-Streiber, M . Rohde, and T. Chakraborty. 1998. Small  GTP-binding proteins of the Rho- and Ras-subfamilies are not involved in the actin rearrangements induced by attaching and effacing Escherichia coli. FEMS Microbiol Lett. 163:107-12. 55.  Egile, C., T. P. Loisel, V. Laurent, R. Li, D. Pantaloni, P. J . Sansonetti, and  M . F. Carlier. 1999. Activation of the CDC42 effector N-WASP by the Shigella flexneri  151  IcsA protein promotes actin nucleation by Arp2/3 complex and bacterial actin-based motility. J Cell Biol. 146:1319-32. 56.  Eilertsen, K. J., S. T. Kazmierski, and T. C. Keller, 3rd. 1997. Interaction of  alpha-actinin with cellular titin. Eur J Cell Biol. 74:361-4. 57.  Elliott, S. J., L. A. Wainwright, T. K. McDaniel, K. G. Jarvis, Y. K. Deng, L .  C. Lai, B. P. McNamara, M . S. Donnenberg, and J. B. Kaper. 1998. The complete sequence of the locus of enterocyte effacement (LEE) from enteropathogenic Escherichia coli E2348/69. Mol Microbiol. 28:1-4. 58.  Elliott, S. J., J. Yu, and J. B. Kaper. 1999. The cloned locus of enterocyte  effacement from enterohemorrhagic Escherichia coli 0157:H7 is unable to confer the attaching and effacing phenotype upon E. coli K-12. Infect Immun. 67:4260-3. 59.  Farris, M., A. Grant, T. B. Richardson, and C. D. O'Connor. 1998. BipA: a  tyrosine-phosphorylated GTPase that mediates interactions between enteropathogenic Escherichia coli (EPEC) and epithelial cells. Mol Microbiol. 28:265-79. 60.  Ferrary, E., M . Cohen-Tannoudji, G. Pehau-Arnaudet, A. Lapillonne, R.  Athman, T. Ruiz, L. Boulouha, F. El Marjou, A. Doye, J. J. Fontaine, C. Antony, C. Babinet, D. Louvard, F. Jaisser, and S. Robine. 1999. In vivo, villin is required for Ca(2+)-dependent F-actin disruption in intestinal brush borders. J Cell Biol. 146:819-30. 61.  Finlay, B. B., I. Rosenshine, M . S. Donnenberg, and J. B. Kaper. 1992.  Cytoskeletal composition of attaching and effacing lesions associated with enteropathogenic Escherichia coli adherence to HeLa cells. Infect Immun. 60:2541-3. 62.  Foubister, V., I. Rosenshine, M . S. Donnenberg, and B. B. Finlay. 1994. The  eaeB gene of enteropathogenic Escherichia coli is necessary for signal transduction in epithelial cells. Infect Immun. 62:3038-40. 63.  Foubister, V., I. Rosenshine, and B. B. Finlay. 1994. A diarrheal pathogen,  enteropathogenic Escherichia coli (EPEC), triggers a flux of inositol phosphates in infected epithelial cells. J Exp Med. 179:993-8.  152  64.  Frankel, G., D. C. Candy, E. Fabiani, J. Adu-Bobie, S. Gil, M . Novakova, A.  D. Phillips, and G. Dougan. 1995. Molecular characterization of a carboxy-terminal eukaryotic-cell- binding domain of intimin from enteropathogenic Escherichia coli. Infect Immun. 63:4323-8. 65.  Frankel, G., O. Lider, R. Hershkoviz, A. P. Mould, S. G. Kachalsky, D. C. A.  Candy, L. Cahalon, M . J. Humphries, and G. Dougan. 1996. The cell-binding domain of intimin from enteropathogenic Escherichia coli binds to betal integrins. J Biol Chem. 271:20359-64. 66.  Frankel, G., A. D. Phillips, M . Novakova, H. Field, D. C. Candy, D. B.  Schauer, G. Douce, and G. Dougan. 1996. Intimin from enteropathogenic Escherichia coli restores murine virulence to a Citrobacter rodentium eaeA mutant: induction of an immunoglobulin A response to intimin and EspB. Infect Immun. 64:5315-25. 67.  Freeman, N. L., D. V. Zurawski, P. Chowrashi, J. C. Ayoob, L. Huang, B.  Mittal, J. M . Sanger, and J. W. Sanger. 2000. Interaction of the enteropathogenic Escherichia coli protein, translocated intimin receptor (Tir), with focal adhesion proteins. Cell Motil Cytoskeleton. 47:307-318. 68.  Frischknecht, F., V. Moreau, S. Rottger, S. Gonfloni, I. Reckmann, G.  Superti-Furga, and M . Way. 1999. Actin-based motility of vaccinia virus mimics receptor tyrosine kinase signalling. Nature. 401:926-9. 69.  Fukami, K., T. Endo, M . Imamura, and T. Takenawa. 1994. alpha-Actinin and  vinculin are PIP2-binding proteins involved in signaling by tyrosine kinase. J Biol Chem. 269:1518-22. 70.  Fukami, K., K. Furuhashi, M . Inagaki, T. Endo, S. Hatano, and T.  Takenawa. 1992. Requirement of phosphatidylinositol 4,5-bisphosphate for alphaactinin function [see comments]. Nature. 359:150-2.  453  71.  Gallego, M . D., M . Santamaria, J. Pena, and I. J. Molina. 1997. Defective  actin reorganization and polymerization of Wiskott-Aldrich T cells in response to CD3mediated stimulation. Blood. 90:3089-97. 72.  Geiger, B., S. Yehuda-Levenberg, and A. D. Bershadsky. 1995. Molecular  interactions in the submembrane plaque of cell-cell and cell- matrix adhesions. Acta Anat. 154:46-62. 73.  Gertler, F. B., K. Niebuhr, M . Reinhard, J. Wehland, and P. Soriano. 1996.  Mena, a relative of VASP and Drosophila Enabled, is implicated in the control of microfilament dynamics. Cell. 87:227-39. 74.  Gilmore, A. P., and K. Burridge. 1995. Cell adhesion. Cryptic sites in vinculin  [news; comment]. Nature. 373:197. 75.  Giron, J. A., A. S. Ho, and G. K. Schoolnik. 1991. An inducible bundle-forming  pilus of enteropathogenic Escherichia coli. Science. 254:710-3. 76.  Glenney, J. R., Jr., P. Kaulfus, and K. Weber. 1981. F actin assembly  modulated by villin: Ca++-dependent nucleation and capping of the barbed end. Cell. 24:471-80. 77.  Gluck, U., and A. Ben-Ze'ev. 1994. Modulation of alpha-actinin levels affects  cell motility and confers tumorigenicity on 3T3 cells. J Cell Sci. 107:1773-82. 78.  Goldberg, M . B., and J. A. Theriot. 1995. Shigella flexneri surface protein IcsA  is sufficient to direct actin- based motility. Proc Natl Acad Sci U S A . 92:6572-6. 79.  Goodison, S., V. Urquidi, and D. Tarin. 1999. CD44 cell adhesion molecules.  Mol Pathol. 52:189-96. 80.  Goosney, D., R. DeVinney, R. Pfuetzner, E. Frey, Strynadka NC, and F. BB.  2000. Enteropathogenic E. coli translocated intimin receptor, Tir, interacts directly with alpha-actinin. Curr. Biol. 10:735-738. 81.  Goosney, D. L., J. Celli, B. Kenny, and B. B. Finlay. 1999. Enteropathogenic  Escherichia coli inhibits phagocytosis. Infect Immun. 67:490-5.  154  82.  Goosney, D. L., R. DeVinney, and B.B. Finlay. 2001. Recruitment of  cytoskeletal and signaling proteins to enteropathogenic and enterohemorrhagic Escherichia coli pedestals. In Press, Infect. Immun. 83.  Greenwood, J. A., A. B. Theibert, G. D. Prestwich, and J. E. Murphy-Ullrich.  2000. Restructuring of focal adhesion plaques by PI 3-kinase. Regulation by Ptdlns (3,4,5)-p(3) binding to alpha-actinin. J Cell Biol. 150:627-42. 84.  Griffin, P. M., S. M . Ostroff, R. V. Tauxe, K. D. Greene, J. G. Wells, J. H .  Lewis, and P. A. Blake. 1988. Illnesses associated with Escherichia coli 0157:H7 infections. A broad clinical spectrum. Ann Intern Med. 109:705-12. 85.  Gunzburg, S. T., B. J. Chang, V. Burke, and M . Gracey. 1992. Virulence  factors of enteric Escherichia coli in young Aboriginal children in north-west Australia. Epidemiol Infect. 109:283-9. 86.  Gyuris, J., E. Golemis, H. Chertkov, and R. Brent. 1993. Cdil, a human G l  and S phase protein phosphatase that associates with Cdk2. Cell. 75:791-803. 87.  Hanein, D., N. Volkmann, S. Goldsmith, A. M . Michon, W. Lehman, R.  Craig, D. DeRosier, S. Almo, and P. Matsudaira. 1998. An atomic model of fimbrin binding to F-actin and its implications for filament crosslinking and regulation [published erratum appears in Nat Struct Biol 1998 Oct;5(10):924]. Nat Struct Biol. 5:787-92. 88.  Hanzel, D., H. Reggio, A. Bretscher, J. G. Forte, and P. Mangeat. 1991. The  secretion-stimulated 80K phosphoprotein of parietal cells is ezrin, and has properties of a membrane cytoskeletal linker in the induced apical microvilli [published erratum appears in EMBO J 1991 Dec;10(12):3978-81]. Embo J. 10:2363-73. 89.  Hardt, W. D., L. M . Chen, K. E. Schuebel, X. R. Bustelo, and J. E. Galan.  1998. S. typhimurium encodes an activator of Rho GTPases that induces membrane ruffling and nuclear responses in host cells. Cell. 93:815-26.  155  90.  Harte, M . T., J. D. Hildebrand, M . R. Burnham, A. H. Bouton, and J. T.  Parsons. 1996. pl30Cas, a substrate associated with v-Src and v-Crk, localizes to focal adhesions and binds to focal adhesion kinase. J Biol Chem. 271:13649-55. 91.  Hartwig, J. H . 1992. Mechanisms of actin rearrangements mediating platelet  activation. J Cell Biol. 118:1421-42. 92.  Hartwig, J. H., G. M . Bokoch, C. L. Carpenter, P. A. Janmey, L . A. Taylor,  A. Toker, and T. P. Stossel. 1995. Thrombin receptor ligation and activated Rac uncap actin filament barbed ends through phosphoinositide synthesis in permeabilized human platelets. Cell. 82:643-53. 93.  Haskins, J., L. Gu, E. S. Wittchen, J. Hibbard, and B. R. Stevenson. 1998.  ZO-3, a novel member of the MAGUK protein family found at the tight junction, interacts with ZO-1 and occludin. J Cell Biol. 141:199-208. 94.  Hecht, G., and A. Koutsouris. 1999. Enteropathogenic E. coli attenuates  secretagogue-induced net intestinal ion transport but not CI- secretion. Am J Physiol. 276:G781-8. 95.  Heise, H., T. Bayerl, G. Isenberg, and E. Sackmann. 1991. Human platelet P-  235, a talin-like actin binding protein, binds selectively to mixed lipid bilayers. Biochim Biophys Acta. 1061:121-31. 96.  Heiska, L., C. Kantor, T. Parr, D. R. Critchley, P. Vilja, C. G. Gahmberg,  and O. Carpen. 1996. Binding of the cytoplasmic domain of intercellular adhesion molecule-2 (ICAM-2) to alpha-actinin. J Biol Chem. 271:26214-9. 97.  Hemmings, L., D. J. Rees, V. Ohanian, S. J. Bolton, A. P. Gilmore, B. Patel,  H. Priddle, J. E. Trevithick, R. O. Hynes, and D. R. Critchley. 1996. Talin contains three actin-binding sites each of which is adjacent to a vinculin-binding site. J Cell Sci. 109:2715-26.  156  98.  Higgins, L . M., G. Frankel, I. Connerton, N. S. Goncalves, G. Dougan, and T.  T. MacDonald. 1999. Role of bacterial intimin in colonic hyperplasia and inflammation. Science. 285:588-91. 99.  Holt, M . R., D. R. Critchley, and N. P. Brindle. 1998. The focal adhesion  phosphoprotein, VASP. Int J Biochem Cell Biol. 30:307-11. 100.  Hueck, C. J. 1998. Type III protein secretion systems in bacterial pathogens of  animals and plants. Microbiol Mol Biol Rev. 62:379-433. 101.  Hyatt, S. L., L . Liao, C. Chapline, and S. Jaken. 1994. Identification and  characterization of alpha-protein kinase C binding proteins in normal and transformed REF52 cells. Biochemistry. 33:1223-8. 102.  Hynes, R. O. 1992. Integrins: versatility, modulation, and signaling in cell  adhesion. Cell. 69:11-25. 103.  Ingber, D. E. 1993. Cellular tensegrity: defining new rules of biological design  that govern the cytoskeleton. J Cell Sci. 104:613-27. 104.  Ismaili, A., D. J. Philpott, M . T. Dytoc, R. Soni, S. Ratnam, and P. M .  Sherman. 1995. Alpha-actinin accumulation in epithelial cells infected with attaching and effacing gastrointestinal pathogens. J Infect Dis. 172:1393-6. 105.  Janmey, P. A., J. Lamb, P. G. Allen, and P. T. Matsudaira. 1992.  Phosphoinositide-binding peptides derived from the sequences of gelsolin and villin. J Biol Chem. 267:11818-23. 106.  Janmey, P. A., and T. P. Stossel. 1987. Modulation of gelsolin function by  phosphatidylinositol 4,5- bisphosphate. Nature. 325:362-4. 107.  Jarvis, K. G., J. A. Giron, A. E. Jerse, T. K. McDaniel, M . S. Donnenberg,  and J. B. Kaper. 1995. Enteropathogenic Escherichia coli contains a putative type III secretion system necessary for the export of proteins involved in attaching and effacing lesion formation. Proc Natl Acad Sci U S A . 92:7996-8000.  157  108.  Jerse, A. E., J. Yu, B. D. Tall, and J. B. Kaper. 1990. A genetic locus of  enteropathogenic Escherichia coli necessary for the production of attaching and effacing lesions on tissue culture cells. Proc Natl Acad Sci U S A . 87:7839-43. 109.  Jesaitis, L . A., and D. A. Goodenough. 1994. Molecular characterization and  tissue distribution of ZO-2, a tight junction protein homologous to ZO-1 and the Drosophila discs-large tumor suppressor protein. J Cell Biol. 124:949-61. 110.  Jockusch, B. M., and G. Isenberg. 1981. Interaction of alpha-actinin and  vinculin with actin: opposite effects on filament network formation. Proc Natl Acad Sci U S A . 78:3005-9. 111.  Kalman, D., O. D. Weiner, D. L. Goosney, J. W. Sedat, B. B. Finlay, A. Abo,  and J. M . Bishop. 1999. Enteropathogenic E. coli acts through WASP and Arp2/3 complex to form actin pedestals. Nat Cell Biolog. 1:389-391. 112.  Kauffmann, F. D., A. 1950. Escherichia strains from infantile epidemic  gastroenteritis. Acta. Path. 27:552-564. 113.  Kelleher, J. F., S. J. Atkinson, and T. D. Pollard. 1995. Sequences, structural  models, and cellular localization of the actin- related proteins Arp2 and Arp3 from Acanthamoeba. J Cell Biol. 131:385-97. 114.  Kenny, B. 1999. Phosphorylation of tyrosine 474 of the enteropathogenic  Escherichia coli (EPEC) Tir receptor molecule is essential for actin nucleating activity and is preceded by additional host modifications. Mol Microbiol. 31:1229-41. 115.  Kenny, B., R. DeVinney, M . Stein, D. J. Reinscheid, E. A. Frey, and B. B.  Finlay. 1997. Enteropathogenic E. coli (EPEC) transfers its receptor for intimate adherence into mammalian cells. Cell. 91:511-20. 116.  Kenny, B., and B. B. Finlay. 1997. Intimin-dependent binding of  enteropathogenic Escherichia coli to host cells triggers novel signaling events, including tyrosine phosphorylation of phospholipase C-gammal. Infect Immun. 65:2528-36.  158  117.  Kenny, B., L. C. Lai, B. B. Finlay, and M . S. Donnenberg. 1996. EspA, a  protein secreted by enteropathogenic Escherichia coli, is required to induce signals in epithelial cells. Mol Microbiol. 20:313-23. 118.  Keon, B. H., S. Schafer, C. Kuhn, C. Grund, and W. W. Franke. 1996.  Symplekin, a novel type of tight junction plaque protein. J Cell Biol. 134:1003-18. 119.  Knutton, S., T. Baldwin, P. H. Williams, and A. S. McNeish. 1989. Actin  accumulation at sites of bacterial adhesion to tissue culture cells: basis of a new diagnostic test for enteropathogenic and enterohemorrhagic Escherichia coli. Infect Immun. 57:1290-8. 120.  Knutton, S., I. Rosenshine, M . J. Pallen, I. Nisan, B. C. Neves, C. Bain, C.  Wolff, G. Dougan, and G. Frankel. 1998. A novel EspA-associated surface organelle of enteropathogenic Escherichia coli involved in protein translocation into epithelial cells. Embo J. 17:2166-76. 121.  Kocks, C , J. B. Marchand, E. Gouin, H. d'Hauteville, P. J. Sansonetti, M . F.  Carlier, and P. Cossart. 1995. The unrelated surface proteins ActA of Listeria monocytogenes and IcsA of Shigella flexneri are sufficient to confer actin-based motility on Listeria innocua and Escherichia coli respectively. Mol Microbiol. 18:413-23. 122.  Koepf, E . K., and L . D. Burtnick. 1992. Interaction of plasma gelsolin with  tropomyosin. FEBS Lett. 309:56-8. 123.  Kotler, D. P., and J. M . Orenstein. 1993. Chronic diarrhea and malabsorption  associated with enteropathogenic bacterial infection in a patient with AIDS. Ann Intern Med. 119:127-8. 124.  Kresse, A. U., M . Rohde, and C. A. Guzman. 1999. The EspD protein of  enterohemorrhagic Escherichia coli is required for the formation of bacterial surface appendages and is incorporated in the cytoplasmic membranes of target cells. Infect Immun. 67:4834-42.  159  125.  Kroemker, M . , A. H. Rudiger, B. M . Jockusch, and M . Rudiger. 1994.  Intramolecular interactions in vinculin control alpha-actinin binding to the vinculin head. FEBS Lett. 355:259-62. 126.  Kuhlman, P. A., L . Hemmings, and D. R. Critchley. 1992. The identification  and characterisation of an actin-binding site in alpha-actinin by mutagenesis. FEBS Lett. 304:201-6. 127.  Laemmli. 1970. Cleavage of structural proteins during the assembly of the head  of bacteriophage T4. Nature. 227:680-5. 128.  Lai, L . C , L. A. Wainwright, K. D. Stone, and M . S. Donnenberg. 1997. A  third secreted protein that is encoded by the enteropathogenic Escherichia coli pathogenicity island is required for transduction of signals and for attaching and effacing activities in host cells. Infect Immun. 65:2211-7. 129.  Leevers, S. J., B. Vanhaesebroeck, and M . D. Waterfield. 1999. Signalling  through phosphoinositide 3-kinases: the lipids take centre stage. Curr Opin Cell Biol. 11:219-25. 130.  Levine, M . M., and R. Edelman. 1984. Enteropathogenic Escherichia coli of  classic serotypes associated with infant diarrhea: epidemiology and pathogenesis. Epidemiol Rev. 6:31-51. 131.  Levine, M . M., J. P. Nataro, H. Karch, M . M . Baldini, J. B. Kaper, R. E .  Black, M . L. Clements, and A. D. O'Brien. 1985. The diarrheal response of humans to some classic serotypes of enteropathogenic Escherichia coli is dependent on a plasmid encoding an enteroadhesiveness factor. J Infect Dis. 152:550-9. 132.  Liu, H., L. Magoun, and J. M . Leong. 1999. betal-chain integrins are not  essential for intimin-mediated host cell attachment and enteropathogenic Escherichia coli-induced actin condensation. Infect Immun. 67:2045-9.  160  133.  Lo, S. H., Q. An, S. Bao, W. K. Wong, Y. Liu, P. A. Janmey, J . H. Hartwig,  and L . B. Chen. 1994. Molecular cloning of chick cardiac muscle tensin. Full-length cDNA sequence, expression, and characterization. J Biol Chem. 269:22310-9. 134.  Lo, S. H., P. A. Janmey, J. H. Hartwig, and L . B. Chen. 1994. Interactions of  tensin with actin and identification of its three distinct actin-binding domains. J Cell Biol. 125:1067-75. 135.  Luo, Y., E . A. Frey, R. A. Pfuetzner, A. L . Creagh, D. G. Knoechel, C. A.  Haynes, B. B. Finlay, and N. C. Strynadka. 2000. Crystal structure of enteropathogenic Escherichia coli intimin-receptor complex. Nature. 405:1073-7. 136.  Machesky, L . M . , S. J. Atkinson, C. Ampe, J. Vandekerckhove, and T. D.  Pollard. 1994. Purification of a cortical complex containing two unconventional actins from Acanthamoeba by affinity chromatography on profilin-agarose. J Cell Biol. 127:107-15. 137.  Machesky, L . M . , and R. H. Insall. 1998. Scarl and the related Wiskott-Aldrich  syndrome protein, WASP, regulate the actin cytoskeleton through the Arp2/3 complex. Curr Biol. 8:1347-56. 138.  Manjarrez-Hernandez, H. A., T. J . Baldwin, P. H. Williams, R. Haigh, S.  Knutton, and A. Aitken. 1996. Phosphorylation of myosin light chain at distinct sites and its association with the cytoskeleton during enteropathogenic Escherichia coli infection. Infect Immun. 64:2368-70. 139.  Marches, O., J. P. Nougayrede, S. Boullier, J. Mainil, G. Charlier, I.  Raymond, P. Pohl, M . Boury, J. De Rycke, A. Milon, and E . Oswald. 2000. Role of tir and intimin in the virulence of rabbit enteropathogenic Escherichia coli serotype O103:H2. Infect Immun. 68:2171-82. 140.  McBride, K., S. G. Rhee, and S. Jaken. 1991. Immunocytochemical localization  of phospholipase C-gamma in rat embryo fibroblasts. Proc Natl Acad Sci U S A . 88:7111-5. 161  141.  McCarthy, K. M., I. B. Skare, M . C. Stankewich, M . Furuse, S. Tsukita, R.  A. Rogers, R. D. Lynch, and E. E. Schneeberger. 1996. Occludin is a functional component of the tight junction. J Cell Sci. 109:2287-98. 142.  McDaniel, T. K., and J. B. Kaper. 1997. A cloned pathogenicity island from  enteropathogenic Escherichia coli confers the attaching and effacing phenotype on E. coli K-12. Mol Microbiol. 23:399-407. 143.  McGough, A., W. Chiu, and M . Way. 1998. Determination of the gelsolin  binding site on F-actin: implications for severing and capping. Biophys J. 74:764-72. 144.  McGregor, A., A. D. Blanchard, A. J. Rowe, and D. R. Critchley. 1994.  Identification of the vinculin-binding site in the cytoskeletal protein alpha-actinin. Biochem J. 301:225-33. 145.  McNamara, B. P., and M . S. Donnenberg. 1998. A novel proline-rich protein,  EspF, is secreted from enteropathogenic Escherichia coli via the type III export pathway. FEMS Microbiol Lett. 166:71-8. 146.  Mellies, J. L., S. J. Elliott, V. Sperandio, M . S. Donnenberg, and J. B. Kaper.  1999. The Per regulon of enteropathogenic Escherichia coli: identification of a regulatory cascade and a novel transcriptional activator, the locus of enterocyte effacement (LEE)-encoded regulator (Ler). Mol Microbiol. 33:296-306. 147.  Merzdorf, C. S., and D. A. Goodenough. 1997. Localization of a novel 210 kDa  protein in Xenopus tight junctions. J Cell Sci. 110:1005-12. 148.  Miki, H., K. Miura, and T. Takenawa. 1996. N-WASP, a novel actin-  depolymerizing protein, regulates the cortical cytoskeletal rearrangement in a PIP2dependent manner downstream of tyrosine kinases. Embo J. 15:5326-35. 149.  Miyamoto, S., H. Teramoto, O. A. Coso, J. S. Gutkind, P. D. Burbelo, S. K.  Akiyama, and K. M . Yamada. 1995. Integrin function: molecular hierarchies of cytoskeletal and signaling molecules. J Cell Biol. 131:791-805.  162  150.  Moon, H. W., Whipp, S.C., Argenzio, R.A., Levine, M . M . Giannella, R.A.  1983. Attaching and effacing activities of rabbit and human enteropathogenic Escherichia coli in pig and rabbit intestines. Infect. Immun. 41:1340-51. 151.  Mooseker, M . S. 1985. Organization, chemistry, and assembly of the cytoskeletal  apparatus of the intestinal brush border. Annu Rev Cell Biol. 1:209-41. 152.  Mooseker, M . S., T. A. Graves, K. A. Wharton, N. Falco, and C. L. Howe.  1980. Regulation of microvillus structure: calcium-dependent solation and cross-linking of actin filaments in the microvilli of intestinal epithelial cells. J Cell Biol. 87:809-22. 153.  Moreau, V., F. Frischknecht, I. Reckmann, R. Vincentelli, G. Rabut, D.  Stewart, and M . Way. 2000. A complex of N-WASP and WIP integrates signalling cascades that lead to actin polymerization. Nat Cell Biol. 2:441-8. 154.  Moss, B. 1990. Poxviridae and their replication, p. 2079-2111. In D. M. K. B.N.  Fields, R.M. Chanock, M.S. Hirsch, J.L. Melnick, T.P. Monath and B. Roizman (ed.), Virology. Raven Press, New York. 155.  Mullins, R. D. 2000. How WASP-family proteins and the Arp2/3 complex  convert intracellular signals into cytoskeletal structures. Curr Opin Cell Biol. 12:91-6. 156.  Mullins, R. D., J. A. Heuser, and T. D. Pollard. 1998. The interaction of Arp2/3  complex with actin: nucleation, high affinity pointed end capping, and formation of branching networks of filaments. Proc Natl Acad Sci U S A . 95:6181-6. 157.  Mullins, R. D., J. F. Kelleher, J. Xu, and T. D. Pollard. 1998. Arp2/3 complex  from Acanthamoeba binds profilin and cross-links actin filaments. Mol Biol Cell. 9:84152. 158.  Murakami, H., T. Iwashita, N. Asai, Y. Iwata, S. Narumiya, and M .  Takahashi. 1999. Rho-dependent and -independent tyrosine phosphorylation of focal adhesion kinase, paxillin and pl30Cas mediated by Ret kinase. Oncogene. 18:1975-82.  163  159.  Neter, E., Westphal, O, Luderitz, O, Gino, R M , Gorzynski, EA. 1955.  demonstration of antibodies against enteropathogenic Escherichia coli in sera of children of various ages. Pediatrics. 16:801-808. 160.  Newman, J. V., B. A. Zabel, S. S. Jha, and D. B. Schauer. 1999. Citrobacter  rodentium espB is necessary for signal transduction and for infection of laboratory mice. Infect Immun. 67:6019-25. 161.  Nhieu, G. T., and P. J. Sansonetti. 1999. Mechanism of Shigella entry into  epithelial cells. Curr Opin Microbiol. 2:51-5. 162.  Niggli, V., S. Kaufmann, W. H. Goldmann, T. Weber, and G. Isenberg. 1994.  Identification of functional domains in the cytoskeletal protein talin. Eur J Biochem. 224:951-7. 163.  Nix, D. A., and M . C. Beckerle. 1997. Nuclear-cytoplasmic shuttling of the focal  contact protein, zyxin: a potential mechanism for communication between sites of cell adhesion and the nucleus. J Cell Biol. 138:1139-47. 164.  O'Neill, G. M., S. J. Fashena, and E. A. Golemis. 2000. Integrin signalling: a  new Cas(t) of characters enters the stage. Trends Cell Biol. 10:111-9. 165.  Oliferenko, S., K. Paiha, T. Harder, V. Gerke, C. Schwarzler, H. Schwarz, H.  Beug, U. Gunthert, and L. A. Huber. 1999. Analysis of CD44-containing lipid rafts: Recruitment of annexin II and stabilization by the actin cytoskeleton. J Cell Biol. 146:843-54. 166.  Otey, C. A., F. M . Pavalko, and K. Burridge. 1990. An interaction between  alpha-actinin and the beta 1 integrin subunit in vitro. J Cell Biol. 111:721-9. 167.  Papa, I., C. Astier, O. Kwiatek, F. Raynaud, C. Bonnal, M . C. Lebart, C.  Roustan, and Y. Benyamin. 1999. Alpha actinin-CapZ, an anchoring complex for thin filaments in Z-line. J Muscle Res Cell Motil. 20:187-97. 168.  Pavalko, F. M., D. M . Walker, L . Graham, M . Goheen, C. M . Doerschuk, and i  G. S. Kansas. 1995. The cytoplasmic domain of L-selectin interacts with cytoskeletal  164  proteins via alpha-actinin: receptor positioning in microvilli does not require interaction with alpha-actinin. J Cell Biol. 129:1155-64. 169.  Peeters, J. E., G. J. Charlier, and P. H. Halen. 1984. Pathogenicity of attaching  effacing enteropathogenic Escherichia coli isolated from diarrheic suckling and weanling rabbits for newborn rabbits. Infect Immun. 46:690-6. 170.  Peeters, J. E., P. Pohl, L . Okerman, and L. A. Devriese. 1984. Pathogenic  properties of Escherichia coli strains isolated from diarrheic commercial rabbits. J Clin Microbiol. 20:34-9. 171.  Petit, M . M., J. Fradelizi, R. M . Golsteyn, T. A. Ayoubi, B. Menichi, D.  Louvard, W. J. Van de Ven, and E. Friederich. 2000. LPP, an actin cytoskeleton protein related to zyxin, harbors a nuclear export signal and transcriptional activation capacity. Mol Biol Cell. 11:117-29. 172.  Philpott, D. J., A. Ismaili, M . T. Dytoc, J. R. Cantey, and P. M . Sherman.  1995. Increased adherence of Escherichia coli RDEC-1 to human tissue culture cells results in the activation of host signaling pathways. J Infect Dis. 172:136-43. 173.  Philpott, D. J., D. M . McKay, P. M . Sherman, and M . H. Perdue. 1996.  Infection of T84 cells with enteropathogenic Escherichia coli alters barrier and transport functions. Am J Physiol. 270:G634-45. 174.  Plopper, G. E., H. P. McNamee, L . E. Dike, K. Bojanowski, and D. E. Ingber.  1995. Convergence of integrin and growth factor receptor signaling pathways within the focal adhesion complex. Mol Biol Cell. 6:1349-65. 175.  Prasad, R., Y. Gu, H. Alder, T. Nakamura, O. Canaani, H. Saito, K.  Huebner, R. P. Gale, P. C. Nowell, K. Kuriyama, and et al. 1993. Cloning of the ALL-1 fusion partner, the AF-6 gene, involved in acute myeloid leukemias with the t(6; 11) chromosome translocation. Cancer Res. 53:5624-8.  165  176.  Prehoda, K. E., J. A. Scott, R. Dyche Mullins, and W. A. Lim. 2000.  Integration of multiple signals through cooperative regulation of the N- WASP-Arp2/3 complex [In Process Citation]. Science. 290:801-6. 177.  Puente, J. L., D. Bieber, S. W. Ramer, W. Murray, and G. K. Schoolnik.  1996. The bundle-forming pili of enteropathogenic Escherichia coli: transcriptional regulation by environmental signals. Mol Microbiol. 20:87-100. 178.  Puius, Y. A., N. M . Mahoney, and S. C. Almo. 1998. The modular structure of  actin-regulatory proteins. Curr Opin Cell Biol. 10:23-34. 179.  Ramesh, N., I. M . Anton, J. H. Hartwig, and R. S. Geha. 1997. WIP, a protein  associated with wiskott-aldrich syndrome protein, induces actin polymerization and redistribution in lymphoid cells. Proc Natl Acad Sci U S A . 94:14671-6. 180.  Ramesh, N., I. M . Anton, N. Martinez-Quiles, and R. S. Geha. 1999. Waltzing  with WASP [see comments]. Trends Cell Biol. 9:15-9. 181.  Rees, D. J., S. E. Ades, S. J. Singer, and R. O. Hynes. 1990. Sequence and  domain structure of talin. Nature. 347:685-9. 182.  Reinhard, M., K. Giehl, K. Abel, C. Haffner, T. Jarchau, V. Hoppe, B. M .  Jockusch, and U. Walter. 1995. The proline-rich focal adhesion and microfilament protein VASP is a ligand for profilins. Embo J. 14:1583-9. 183.  Reinhard, M., M . Halbrugge, U. Scheer, C. Wiegand, B. M . Jockusch, and U.  Walter. 1992. The 46/50 kDa phosphoprotein VASP purified from human platelets is a novel protein associated with actin filaments and focal contacts. Embo J. 11:2063-70. 184.  Reinhard, M . , K. Jouvenal, D. Tripier, and U. Walter. 1995. Identification,  purification, and characterization of a zyxin-related protein that binds the focal adhesion and microfilament protein VASP (vasodilator-stimulated phosphoprotein). Proc Natl Acad Sci U S A . 92:7956-60.  166  185.  Reinhard, M . , J. Zumbrunn, D. Jaquemar, M . Kuhn, U. Walter, and B.  Trueb. 1999. An alpha-actinin binding site of zyxin is essential for subcellular zyxin localization and alpha-actinin recruitment. J Biol Chem. 274:13410-8. 186.  Rodriguez, M . L., M . Brignoni, and P. J. Salas. 1994. A specifically apical sub-  membrane intermediate filament cytoskeleton in non-brush-border epithelial cells. J Cell Sci. 107:3145-51. 187.  Rodriguez-Boulan, E., and W. J. Nelson. 1989. Morphogenesis of the polarized  epithelial cell phenotype. Science. 245:718-25. 188.  Rodriguez-Boulan, E., and P. J. Salas. 1989. External and internal signals for  epithelial cell surface polarization. Annu Rev Physiol. 51:741-54. 189.  Rohatgi, R., H. Y. Ho, and M . W. Kirschner. 2000. Mechanism of N-WASP  activation by CDC42 and phosphatidylinositol 4, 5- bisphosphate [see comments]. J Cell Biol. 150:1299-310. 190.  Rohatgi, R., L . Ma, H. Miki, M . Lopez, T. Kirchhausen, T. Takenawa, and  M . W. Kirschner. 1999. The interaction between N-WASP and the Arp2/3 complex links Cdc42- dependent signals to actin assembly. Cell. 97:221-31. 191.  Rosenshine, I., M . S. Donnenberg, J. B. Kaper, and B. B. Finlay. 1992. Signal  transduction between enteropathogenic Escherichia coli (EPEC) and epithelial cells: EPEC induces tyrosine phosphorylation of host cell proteins to initiate cytoskeletal rearrangement and bacterial uptake. Embo J. 11:3551-60. 192.  Rosenshine, I., S. Ruschkowski, M . Stein, D. J. Reinscheid, S. D. Mills, and B.  B. Finlay. 1996. A pathogenic bacterium triggers epithelial signals to form a functional bacterial receptor that mediates actin pseudopod formation. Embo J. 15:2613-24. 193.  Rottger, S., F. Frischknecht, I. Reckmann, G. L . Smith, and M . Way. 1999.  Interactions between vaccinia virus LEV membrane proteins and their roles in LEV assembly and actin tail formation. J Virol. 73:2863-75.  167  194.  Ruden, D. M., J. Ma, Y. Li, K. Wood, and M . Ptashne. 1991. Generating yeast  transcriptional activators containing no yeast protein sequences. Nature. 350:250-2. 195.  Sadler, L, A. W. Crawford, J. W. Michelsen, and M . C. Beckerle. 1992. Zyxin  and cCRP: two interactive LIM domain proteins associated with the cytoskeleton. J Cell Biol. 119:1573-87. 196.  Sanger, J. M., R. Chang, F. Ashton, J. B. Kaper, and J. W. Sanger. 1996.  Novel form of actin-based motility transports bacteria on the surfaces of infected cells. Cell Motil Cytoskeleton. 34:279-87. 197.  Scaletsky, I. C , M . L. Silva, and L . R. Trabulsi. 1984. Distinctive patterns of  adherence of enteropathogenic Escherichia coli to HeLa cells. Infect Immun. 45:534-6. 198.  Schaller, M . D., and J. T. Parsons. 1994. Focal adhesion kinase and associated  proteins. Curr Opin Cell Biol. 6:705-10. 199.  Schauer, D. B., and S. Falkow. 1993. Attaching and effacing locus of a  Citrobacter freundii biotype that causes transmissible murine colonic hyperplasia. Infect Immun. 61:2486-92. 200.  Schauer, D. B., and S. Falkow. 1993. The eae gene of Citrobacter freundii  biotype 4280 is necessary for colonization in transmissible murine colonic hyperplasia. Infect Immun. 61:4654-61. 201.  Scheffzek, K., M . R. Ahmadian, W. Kabsch, L . Wiesmuller, A. Lautwein, F.  Schmitz, and A. Wittinghofer. 1997. The Ras-RasGAP complex: structural basis for GTPase activation and its loss in oncogenic Ras mutants [see comments]. Science. 277:333-8. 202.  Schmelz, M., B. Sodeik, M . Ericsson, E. J. Wolffe, H. Shida, G. Hiller, and G.  Griffiths. 1994. Assembly of vaccinia virus: the second wrapping cisterna is derived from the trans Golgi network. J Virol. 68:130-47.  168  203.  Schtoeder, S., Caldwell, JR, Vernon, T M , White, PS, Granger, SI, Bennett  JV. 1968. A waterborne outbreak of gatroenteritis in adults associated with enteropathogenic Escherichia coli. Lancet. 1:737-740. 204.  Selden, L. A., H. J. Kinosian, J. Newman, B. Lincoln, C. Hurwitz, L . C.  Gershman, and J. E. Estes. 1998. Severing of F-actin by the amino-terminal half of gelsolin suggests internal cooperativity in gelsolin. Biophys J. 75:3092-100. 205.  Simons, K., and A. Wandinger-Ness. 1990. Polarized sorting in epithelia. Cell.  62:207-10. 206.  Skoble, J., D. A. Portnoy, and M . D. Welch. 2000. Three regions within ActA  promote Arp2/3 complex-mediated actin nucleation and Listeria monocytogenes motility. J Cell Biol. 150:527-38. 207.  Smith, G. A., D. A. Portnoy, and J. A. Theriot. 1995. Asymmetric distribution  of the Listeria monocytogenes ActA protein is required and sufficient to direct actinbased motility. Mol Microbiol. 17:945-51. 208.  Smith, G. A., J. A. Theriot, and D. A. Portnoy. 1996. The tandem repeat  domain in the Listeria monocytogenes ActA protein controls the rate of actin-based motility, the percentage of moving bacteria, and the localization of vasodilator-stimulated phosphoprotein and profilin. J Cell Biol. 135:647-60. 209.  Sohel, I., J. L. Puente, S. W. Ramer, D. Bieber, C. Y. Wu, and G. K.  Schoolnik. 1996. Enteropathogenic Escherichia coli: identification of a gene cluster coding for bundle-forming pilus morphogenesis. J Bacteriol. 178:2613-28. 210.  Sohn, R. H., and P. J. Goldschmidt-Clermont. 1994. Profilin: at the crossroads  of signal transduction and the actin cytoskeleton. Bioessays. 16:465-72. 211.  South wick, F. S. 2000. Gelsolin and ADF/cofilin enhance the actin dynamics of  motile cells. Proc Natl Acad Sci U S A . 97:6936-8.  169  212.  Spitz, J., R. Yuhan, A. Koutsouris, C. Blatt, J. Alverdy, and G. Hecht. 1995.  Enteropathogenic Escherichia coli adherence to intestinal epithelial monolayers diminishes barrier function. Am J Physiol. 268:G374-9. 213.  Staley, T. E., E. W. Jones, and L . D. Corley. 1969. Attachment and penetration  of Escherichia coli into intestinal epithelium of the ileum in newborn pigs. Am J Pathol. 56:371-92. 214.  Stevenson, B. R., and B. H. Keon. 1998. The tight junction: morphology to  molecules. Annu Rev Cell Dev Biol. 14:89-109. 215.  Stevenson, B. R., J. D. Siliciano, M . S. Mooseker, and D. A. Goodenough.  1986. Identification of ZO-1: a high molecular weight polypeptide associated with the tight junction (zonula occludens) in a variety of epithelia. J Cell Biol. 103:755-66. 216.  Suetsugu, S., H. Miki, and T. Takenawa. 1998. The essential role of profilin in  the assembly of actin for microspike formation. Embo J. 17:6516-26. 217.  Suzuki, T., H . Miki, T. Takenawa, and C. Sasakawa. 1998. Neural Wiskott-  Aldrich syndrome protein is implicated in the actin- based motility of Shigella flexneri. Embo J. 17:2767-76. 218.  Suzuki, T., H. Mimuro, H. Miki, T. Takenawa, T. Sasaki, H. Nakanishi, Y.  Takai, and C. Sasakawa. 2000. Rho family GTPase Cdc42 is essential for the actinbased motility of Shigella in mammalian cells. J Exp Med. 191:1905-20. 219.  Suzuki, T., S. Saga, and C. Sasakawa. 1996. Functional analysis of Shigella  VirG domains essential for interaction with vinculin and actin-based motility. J Biol Chem. 271:21878-85. 220.  Symons, M . , J. M . Derry, B. Karlak, S. Jiang, V. Lemahieu, F. McCormick,  U. Francke, and A. Abo. 1996. Wiskott-Aldrich syndrome protein, a novel effector for the GTPase CDC42Hs, is implicated in actin polymerization. Cell. 84:723-34.  170  221.  Taylor, K. A., P. W. Luther, and M . S. Donnenberg. 1999. Expression of the  EspB protein of enteropathogenic Escherichia coli within HeLa cells affects stress fibers and cellular morphology. Infect Immun. 67:120-5. 222.  Taylor, K. A., C. B. O'Connell, P. W. Luther, and M . S. Donnenberg. 1998.  The EspB protein of enteropathogenic Escherichia coli is targeted to the cytoplasm of infected HeLa cells. Infect Immun. 66:5501-7. 223.  Theriot, J. A., J. Rosenblatt, D. A. Portnoy, P. J. Goldschmidt-Clermont, and  T. J. Mitchison. 1994. Involvement of profilin in the actin-based motility of L. monocytogenes in cells and in cell-free extracts. Cell. 76:505-17. 224.  Thomas, S. M., and J. S. Brugge. 1997. Cellular functions regulated by Sre  family kinases. Annu Rev Cell Dev Biol. 13:513-609. 225.  Tsukita, S., K. Oishi, N. Sato, J. Sagara, and A. Kawai. 1994. ERM family  members as molecular linkers between the cell surface glycoprotein CD44 and actinbased cytoskeletons. J Cell Biol. 126:391-401. 226.  Turner, C. E. 1998. Paxillin. Int J Biochem Cell Biol. 30:955-9.  227.  Turner, C. E., and J. T. Miller. 1994. Primary sequence of paxillin contains  putative SH2 and SH3 domain binding motifs and multiple LIM domains: identification of a vinculin and ppl25Fak-binding region. J Cell Sci. 107:1583-91. 228.  Wachter, C , C. Beinke, M . Mattes, and M . A. Schmidt. 1999. Insertion of  EspD into epithelial target cell membranes by infecting enteropathogenic Escherichia coli. Mol Microbiol. 31:1695-707. 229.  Warawa, J., B. B. Finlay, and B. Kenny. 1999. Type III secretion-dependent  hemolytic activity of enteropathogenic Escherichia coli. Infect Immun. 67:5538-40. 230.  Weber, E., G. Berta, A. Tousson, P. St John, M . W. Green, U.  Gopalokrishnan, T. Jilling, E. J. Sorscher, T. S. Elton, D. R. Abrahamson, and et al. 1994. Expression and polarized targeting of a rab3 isoform in epithelial cells. J Cell Biol. 125:583-94.  171  231.  Welch, M . D., A. H. DePace, S. Verma, A. Iwamatsu, and T. J. Mitchison.  1997. The human Arp2/3 complex is composed of evolutionarily conserved subunits and is localized to cellular regions of dynamic actin filament assembly. J Cell Biol. 138:37584. 232.  Welch, M . D., J. Rosenblatt, J. Skoble, D. A. Portnoy, and T. J. Mitchison.  1998. Interaction of human Arp2/3 complex and the Listeria monocytogenes ActA protein in actin filament nucleation. Science. 281:105-8. 233.  Wolff, C , I. Nisan, E . Hanski, G. Frankel, and I. Rosenshine. 1998. Protein  translocation into host epithelial cells by infecting enteropathogenic Escherichia coli. Mol Microbiol. 28:143-55. 234.  Wong, V., and B. M . Gumbiner. 1997. A synthetic peptide corresponding to the  extracellular domain of occludin perturbs the tight junction permeability barrier. J Cell Biol. 136:399-409. 235.  Youssoufian, H., M . McAfee, and D. J . Kwiatkowski. 1990. Cloning and  chromosomal localization of the human cytoskeletal alpha- actinin gene reveals linkage to the beta-spectrin gene. Am J Hum Genet. 47:62-71. 236.  Yuhan, R., A. Koutsouris, S. D. Savkovic, and G. Hecht. 1997.  Enteropathogenic Escherichia coli-induced myosin light chain phosphorylation alters intestinal epithelial permeability. Gastroenterology. 113:1873-82. 237.  Zalevsky, J., I. Grigorova, and R. D. Mullins. 2000. Activation of the Arp2/3  complex by the Listeria ActA protein: ActA binds two actin monomers and three subunits of the Arp2/3 complex. J Biol Chem. 238.  Zhang, J., W. G. King, S. Dillon, A. Hall, L. Feig, and S. E. Rittenhouse.  1993. Activation of platelet phosphatidylinositide 3-kinase requires the small GTPbinding protein Rho. J Biol Chem. 268:22251-4.  172  239.  Zhong, Y., T. Saitoh, T. Minase, N. Sawada, K. Enomoto, and M . Mori. 1993.  Monoclonal antibody 7H6 reacts with a novel tight junction-associated protein distinct from ZO-1, cingulin and ZO-2. J Cell Biol. 120:477-83. 240.  Zigmond, S. H . 1998. Actin cytoskeleton: the Arp2/3 complex gets to the point.  Curr Biol. 8:R654-7.  173  Appendix IrEnteropathogenic Escherichia coli (EPEC) inhibits phagocytosis Danika L. Goosney , Jean Celli , Brendan Kenny ' , and B.Brett Finlay '* 1  1  1  2  Running Title: EPEC inhibits phagocytosis  Correspondent Footnote: ^Biotechnology Laboratory, and the Departments of Microbiology & Immunology, and Biochemistry & Molecular Biology, University of British Columbia, Vancouver, B.C., Canada, V6T 1Z3 604-822-2210 (ph); 604-822-9830 (fax). Email: bfinlay@unixg.ubc.ca  ^Current address: Department of Pathology and Microbiology, School of Medical Sciences, University of Bristol, Bristol, GB, BS81TD  * To whom correspondence should be addressed.  174  ABSTRACT  Enteropathogenic Escherichia coli (EPEC) interacts with intestinal epithelial cells, activating host signaling pathways leading to cytoskeletal rearrangements and ultimately diarrhea. In this study, we demonstrate that EPEC interacts with the macrophage-like cell line J774A.1 to inhibit phagocytosis by these cells. Antiphagocytic activity was also observed in cultured RAW macrophage-like cells upon EPEC infection. The EPEC anti-phagocytic phenotype was dependent on the type III secretion pathway of EPEC and its secreted proteins, including EspA, EspB, and EspD. Intimin and Tir mutants displayed intermediate anti-phagocytic activity, suggesting that intimate attachment mediated by intimin-Tir binding may also play a role in antiphagocytosis. Tyrosine dephosphorylation of several host proteins was observed following infection with secretion competent EPEC, but not with secretion deficient mutants. Dephosphorylation was detectable 120 min after infection with EPEC, directly correlating with the onset of the anti-phagocytic phenotype. Inhibition of protein tyrosine phosphatases by pervanadate treatment increased the number of intracellular wild type EPEC to levels seen with secretion deficient mutants, suggesting that dephosphorylation events are linked to the anti-phagocytic phenotype. No tyrosine phosphatase activity was detected with the EPEC secreted proteins, suggesting that EPEC induces antiphagocytosis via a different mechanism than Yersinia species. Taken together, the present findings demonstrate a novel function for EPEC-secreted proteins in triggering macrophage protein tyrosine dephosphorylation and inhibition of phagocytosis.  175  INTRODUCTION Enteropathogenic Escherichia  coli (EPEC), a human pathogen, is a leading cause  of infantile diarrhea in developing countries, killing up to one million children per year worldwide (21). Despite the significance of this disease, the mechanisms by which EPEC causes diarrhea remain poorly characterized. It has been demonstrated both in vitro and in vivo that EPEC initially adhere non-intimately in microcolonies to host intestinal epithelial cells via bundle forming pili (bfp) (13). Upon initial adherence, several bacterial proteins are secreted by a Type III secretion pathway, of which at least three, EspA, EspB, and EspD, are involved in activating host signals (10, 14, 15, 16, 20). Induction of host signaling events leads to rearrangement of the host cytoskeleton to form the characteristic attaching and effacing (A/E) lesion, resembling a pedestal structure upon which the organism resides (25). The signaling events include induction of host inositol triphosphate (IP3) and Ca^+ fluxes, as well as host protein phosphorylation (3, 8, 9, 16, 24). The bacterial receptor on the host cell, Tir, is tyrosine phosphorylated prior to intimate binding of EPEC through the outer membrane protein, intimin. Tir is a bacterially secreted protein which is inserted into host cell membranes and tyrosine phosphorylated (18). Binding of Tir to the EPEC outer membrane protein, intimin, leads to pedestal formation (25, 18). Tyrosine dephosphorylation of host proteins also occurs within the host epithelial cell following EPEC infection, but its role in the attachment process remains undefined (16). This dephosphorylation occurs independently of intimin. Furthermore, inhibition of tyrosine dephosphorylation does not prevent pedestal formation, suggesting that it may play another role in the infection process (16). Pathogenic Yersinia species induce tyrosine dephosphorylation of macrophage proteins to paralyze host phagocytic activity  176  (26, 27). This event requires a bacterial tyrosine phosphatase, YopH, which is translocated into the host cell (4, 2). The rabbit form of EPEC, RDEC-1, inhibits its uptake by membranous (M) cells on intestinal surfaces, but the mechanism mediating anti-phagocytosis remains undefined (12). This pathogen is very similar to EPEC, secreting EspA, EspB, and Tir, and producing A/E lesions (1). M cells are specialized epithelial phagocytic cells which transport many bacteria (except RDEC-1), bacterial components, and other antigens from the lumen into contact with the underlying antigen presenting cells of the gastrointestinalassociated lymphoid tissue (GALT) (30). The anti-phagocytic phenotype may be advantageous in that it allows the bacteria to colonize the intestinal lining without being transported to the GALT, and thereby delaying an elicited immune response. Given the anti-invasive phenotype of RDEC-1 and the occurrence of tyrosine dephosphorylation events upon EPEC infection of epithelial cells, we investigated the possible relationship between EPEC- induced dephosphorylation and anti-phagocytic activity.  MATERIALS AND METHODS Cell culture and bacterial growth The mouse phagocytic cell lines J774A. 1 (22) and RAW (23) were grown in Dulbecco's minimal Eagle medium (DMEM) supplemented with 10% fetal calf serum at 37°C in a humidified atmosphere with 5% C02- EPEC and Y. pseudotuberculosis strains used in this study are listed in Table 1. EPEC strains were cultured in Luria-Bertani (LB) broth overnight at 37°C without shaking. Y. pseudotuberculosis strains were grown in Brain-Heart Infusion (BHI) broth at 26°C overnight on a rotary shaker. These cultures were diluted to 10^ bacteria/ml (O.D.550 = 0.1) and further incubated at 26°C for lh, and then at 37°C for 2h prior to infection of J774 cells. Determination of bacterial uptake by J774 cells by immunofluorescence 177  Two days prior to infection, J774 cells were seeded onto coverslips (11 mm diameter; 5x10^ cells/well) in a 24-well microtitre plate. Monolayers were routinely infected with EPEC (multiplicity of infection (m.o.i.) = 50) for 180 min (except time course experiments). Bacterial uptake was stopped by placing the microtitre plates on ice. Intra- and-extracellular bacteria were determined as described previously for F. pseudotuberculosis (26, 2). Briefly, to stain extracellular bacteria, coverslips were washed and incubated with rabbit anti-EPEC antibodies (1:100 dilution) (11) and incubated for 30 min at 4°C. Monolayers were washed four times with PBS and fixed with ice-cold methanol for 90 sec. After methanol removal, extracellular bacteria were labeled with Texas-Red conjugated goat anti-rabbit antiserum (Jackson Laboratories;!:200 dilution) for 20 min at 37°C, followed by four washes in PBS. Both intra- and extracellular bacteria were then stained by incubating coverslips for 1 h at 37°C with anti-EPEC antisera, and washed four times in PBS. Coverslips were then incubated for 20 min at 37°C with FITC-conjugated goat-anti-rabbit antiserum (Jackson Laboratories; 1:200 dilution). After four washes in PBS, coverslips were mounted in Mowiol mounting medium on a glass slide. Extracellular bacteria were detected by excitation at 596 nm and total cell-associated bacteria at 490 nm. For each experiment, 50 cells per coverslip with 10-20 cell-associated bacteria were randomly selected and the number of extracellular and total cell-associated bacteria were determined. To determine the effects of bacterial number on anti-phagocytosis, 50 cells per coverslip were counted with either 0-5, 5-15, or 15-50 bacteria per cell. Preparation of cell lysates for Western blotting J774 monolayers were cultured in 60 mm plates to confluency and infected with an overnight EPEC culture for 3 h (m.o.i. = 50) in serum-free DMEM. The monolayers were washed three times in PBS and resuspended directly in boiling Laemmli sample buffer (19). Protein samples were resolved by 8% SDS-PAGE (19) and the proteins transferred to nitrocellulose as described elsewhere (29). Blots were blocked in 4% BSA  178  in PBS prior to incubation with anti-phosphotyrosine (4G10: Upstate Biotechnology Inc; 1:2000 dilution) and phosphotyrosine proteins were detected by alkaline phosphatase conjugated secondary antibodies as described previously (24). Cell viability assays Trypan blue (0.08% final volume; Gibco) was added to wells containing infected monolayers of J774 cells (200 ul/ml DMEM) for 10 mins, followed by four washes in PBS. Cells were visualized by phase contrast microscopy and the percentage of blue cells determined. Inhibition of protein tyrosine phosphatases by pervanadate Hydrogen peroxide (2 mM) was added to 0.1 mM vanadate (VO4; Sigma) to make pervanadate in DMEM. This solution was added to infected J774 cells following 90 min of EPEC infection, and incubated for an additional 60 min. Controls of untreated, H2O2 only, or VO4 only were performed in parallel. After treatment, cells were prepared for fluorescence microscopy or for Western analysis as described above. Detection of protein tyrosine phosphatase activity To detect phosphatase activity of EPEC secreted proteins and of Y. pseudotuberculosis Yops in vitro, a tyrosine phosphatase assay kit (Boehringer Mannheim) was used according to manufacturer's instructions. EPEC was grown under optimal conditions for secretion of Esps as described elsewhere (17). Briefly, EPEC transformed with the plasmid CVD450 were grown standing overnight at 37°C in LB broth supplemented with tetracycline (25 ug/ml final concentration). The culture was diluted then 1:100 in M9 media supplemented with 0.45% glucose, 0.2% Casamino acids, and 0.4% NaHC03 (w/v). EPEC cultures were grown to an OD600 of 0.6 and centrifuged (16 K RCF) for 2 min to pellet out the bacteria. M9 supernatant containing the secreted proteins was concentrated using a Centricon-10 (Amicon). To collect Y. pseudotuberculosis Yops, the bacteria were grown as described above and concentrated in the Centricon-10. 179  RESULTS Phagocytosis of EPEC by J774 cells To examine whether EPEC exhibited any anti-phagocytic activity, cultured J774 cells were infected with wild type EPEC and a type III secretion deficient mutant, cfm, followed by immunofluorescence to determine the number of extracellular and total cellassociated bacteria. Wild type EPEC inhibited its own uptake into J774 cells (only 36% of cell-associated bacteria were intracellular), whereas cfm was internalized in significantly greater numbers (87% intracellular; p < 0.01) (Fig. 1). Similar results were obtained with RAW cells, with 19% of wild type EPEC being intracellular compared to 66% of the cfm strain being internalized. To determine the potential roles of different EPEC gene products in mediating anti-phagocytosis, several EPEC mutants were examined by immunofluorescence. The Bfp mutant inhibited its own uptake into J774 cells at levels comparable to that seen by wild type EPEC (30% intracellular). By contrast, type III secretion and signaling defective mutants, AespA, AespB, and AespD, were internalized in significantly greater numbers than the parental strain (p< 0.01), with 87%, 79%, and 96% of cell-associated bacteria found inside the cell, respectively. The Tir and intimin (eae) mutants showed intermediate phenotypes, with 65% and 68% intracellular bacteria, respectively. It is possible that the lack of intracellular EPEC could be simply due to EPEC inducing J774 cell death and not preventing uptake at all. To examine this possibility, EPEC infected macrophages were assayed for membrane integrity by trypan blue exclusion. No significant J774 membrane permeability was observed during the three hour infection period with any of the EPEC strains. Kinetics of EPEC-induced anti-phagocytosis To further characterize the anti-phagocytic effect of EPEC on J774 cells, the number of cell-associated EPEC (wild type and cfm) and the percentage of those that  180  were intracellular was determined at 60, 120, and 180 mins post-infection (Fig. 2a, b). EPEC exhibited increasing adherence and anti-phagocytosis (i.e. decreasing number of intracellular bacteria) with increasing time. At 60 mins after bacterial addition, J774 cells had approximately 5 EPEC associated per cell, 78% of which were intracellular. At 120 and 180 min after bacterial addition, EPEC adherence to J774 cells increased significantly relative to 60 min infection (p < 0.05). Correlated with the increased adherence was a significant increase in anti-phagocytic activity (26% and 16% of EPEC were intracellular at 120 and 180 min, respectively). This was in contrast to cfm, as increased adherence did not result in increased anti-phagocytic activity. The percentage of intracellular cfm remained similar at 60, 120, and 180 min after bacterial addition. The above results correlate anti-phagocytic activity with increased numbers of adherent secretion competent EPEC. To further examine this relationship, J774 cells were infected for 180 min with wild type EPEC and cfm, and the percentage of intracellular bacteria was determined for cells with 0-5, 5-15, or 15-50 cell-associated bacteria. The anti-phagocytic effect of the wild type EPEC was increased significantly (at p < 0.05) with increasing number of bacteria per J774 cell (Fig. 2c). This differed from cfm, as the percentage of intracellular bacteria remained independent of the number of cell-associated bacteria. Phosphorylation profiles of EPEC infected J774 cells To determine whether EPEC induced tyrosine dephosphorylation in J774 cells, as well as in HeLa cells (16), J774 cells were infected with wild type and EPEC mutants and protein extracts resolved by SDS-PAGE. Phosphotyrosine profiles were visualized by Western blot analysis using anti-phosphotyrosine antibodies (see Materials and Methods). Wild type EPEC and the Bfp mutant induced tyrosine dephosphorylation of a triplet of proteins migrating around 50 kDa, a 60 kDa protein, a 100 kDa protein, and a 110 kDa protein (Fig. 3). The eae and tir mutants induced dephosphorylation of the 60 kDa protein as well as the 50 kDa, 100 kDa, or 110 kDa proteins, although to a lesser  181  extent. The secretion deficient mutants (espA, espB, espD, and cfm-14) did not cause tyrosine dephosphorylation of host proteins relative to the uninfected control. The strains that caused dephosphorylation were the same as those that blocked bacterial uptake, correlating dephosphorylation with the observed anti-phagocytic phenotype. Interestingly, the eae and tir mutants which demonstrated intermediate anti-phagocytic effects induced intermediate levels of dephosphorylation. Tyrosine dephosphorylation kinetics of EPEC-infected J774 cells To correlate the onset of anti-phagocytosis with tyrosine dephosphorylation, tyrosine phosphorylation profiles were determined for J774 cells infected with wild type EPEC and cfm for 60, 120, and 180 min (Fig. 4). At 60 min infection, both EPEC and cfm infected cells showed a similar tyrosine phosphorylation profile as the uninfected control. However, at 120 and 180 min infection, significant tyrosine dephosphorylation of the 110 kDa, 100 kDa, 60 kDa and 50 kDa proteins had occurred in the EPEC-infected samples relative to the uninfected control. No such dephosphorylation was observed with the cfm infected cells at 180 min. Tir phosphorylation was observed at 120 min and remained phosphorylated at 180 min. These data further suggest a role for tyrosine dephosphorylation in EPEC-mediated anti-phagocytosis. Inhibition of protein tyrosine phosphatases by pervanadate To determine if the anti-phagocytic phenotype was dependent upon tyrosine dephosphorylation, pervanadate was used to inhibit protein tyrosine phosphatases in the J774 cells. Monolayers were infected with EPEC for 90 min prior to the addition of pervanadate for the remaining 60 min infection as described in the Materials and Methods. No significant loss of membrane integrity was observed during these experiments as determined by trypan blue staining. EPEC-infected J774 cells treated with pervanadate showed a 3.5-fold increase in the percentage of total cell-associated EPEC that were intracellular (83%), relative to no treatment (24%) (Fig. 5). However, the cfm EPEC-infected cells maintained similar numbers of intracellular bacteria, with or  182  without pervanadate treatment (88% and 86%, respectively). The Y. pseudotuberculosis wild type strain was used as a control to demonstrate the ability of pervanadate to inhibit Y. pseudotuberculosis anti-phagocytotic activity (15% of Y. pseudotuberculosis were intracellular without the pervanadate treatment; 60% of the bacteria were intracellular with the pervanadate treatment; data not shown). Additionally, the tyrosine phosphorylation profile of EPEC-infected cells with pervanadate showed strong inhibition of dephosphorylation compared to the untreated control (data not shown). T y r o s i n e phosphatase activity o f E P E C secreted proteins  To determine if the observed tyrosine dephosphorylation was caused by one of the EPEC secreted proteins, phosphatase activity was monitored (Table 2). No significant decrease in absorbance was detected when the EPEC secreted proteins were added to a tyrosine phosphorylated substrate in vitro compared to the substrate alone, suggesting that tyrosine dephosphorylation did not occur under these conditions. However, a very significant phosphatase activity (p< 0.01) was detected with the Y. pseudotuberculosis secreted proteins, as would be expected due to the presence of YopH. The addition of pervanadate caused a significant increase in absorbance with Y. pseudotuberculosis (p<0.01), suggesting that pervanadate inhibited phosphatase activity of the Yops. No effect was seen with the EPEC secreted proteins upon pervanadate addition. DISCUSSION  This study demonstrates that EPEC inhibits its own phagocytosis into cultured macrophage-like cells. This ability is dependent upon the EPEC type III secretion system and its secreted proteins (EspA, EspB, and EspD) which mediate effects on tyrosine phosphorylated proteins within the host cell. Two lines of evidence suggest the involvement of the secreted proteins. Mutations in the secretion apparatus, as seen in the cfm strain, resulted in the inability of EPEC to inhibit phagocytosis by J774 cells. EPEC strains lacking EspA, EspB, or EspD also lost any anti-phagocytic capacity, indicating a  183  role in mediating this process, either directly as effector molecules or indirectly as part of a transport delivery apparatus. The time course data indicate that EPEC-mediated antiphagocytosis does not occur within the first hour of infection, but is evident after two hours. Protein secretion and significant bacterial attachment do not occur until 90 min of infection (14). Indeed, only low level adherence to J774 cells was observed until 120 min after bacterial addition, and as bacterial adherence increased, anti-phagocytosis also increased, correlating EPEC attachment and protein secretion/transfer with antiphagocytic activity. The intermediate anti-phagocytic phenotype observed for the Tir and intimin mutants correlated with an intermediate dephosphorylation of the same host proteins that are dephosphorylated following infection with both wild type and Bfp mutant strains. Tir and intimin mutants thus seem to be less efficient to induce anti-phagocytosis, although still signaling-competent compared to EspA, EspB, EspD, or cfm mutants. This argues for a role of intimate attachment in anti-phagocytosis. Prolonged intimate attachment of EPEC to the host cell leads to cytoskeletal rearrangements and pedestal formation. The role of tyrosine dephosphorylation in pedestal formation remains unclear. Prolonged treatment of EPEC-infected J774 cells with pervanadate leads to cell death (data not shown), and as a consequence it cannot be determined if pedestals are formed in the presence of this phosphatase inhibitor. Therefore, it cannot be ruled out that cytoskeletal rearrangements are not involved in the EPEC-mediated anti-phagocytic process. Another explanation would be that intimate attachment favors translocation of bacterial effectors required for inducing anti-phagocytosis, since it has been shown that EspB is less efficiently translocated into HeLa cells by the intimin mutant than by the wild type strain (31). The number of cell-associated bacteria also affected the levels of anti-phagocytic activity, probably through a dose effect on the signaling events triggered by the secreted proteins. It might be expected, therefore, that microcolony formation, mediated by Bfp,  184  would be important in the process by increasing the number of adherent bacteria and thereby delivering a stronger anti-phagocytic signal. However, there were no differences in levels of uptake between the wild type and bfp suggesting that this was not the case. Microcolony formation could possibly function in anti-phagocytosis by interconnecting large numbers of bacteria to the macrophage surface. However, the secretion mutants are still capable of microcolony formation, and very few bacteria are found extracellular on the J774 cells, suggesting that this is probably not the case. Anti-phagocytic activity was correlated with tyrosine dephosphorylation events, and the presence of pervanadate specifically inhibited the anti-phagocytic activity of wild type EPEC without hindering the uptake of the signaling mutant, cfm. Pervanadate also inhibited the anti-phagocytic activity of Y. pseudotuberculosis, as previously reported (2). Since Y. pseudotuberculosis transfers a phosphatase into host cells that leads to the activity, we hypothesized that EPEC may also inject a phosphatase. However, no tyrosine phosphatase activity was detected with the secreted proteins using an in vitro assay system. This suggest that EPEC, unlike Y. pseudotuberculosis, does not secrete a functional phosphatase under these conditions. It is possible that E P E C may secrete a phosphatase that requires a eukaryotic cofactor for activity. Alternatively, the secreted proteins may trigger the activation of a mammalian tyrosine phosphatase. Taken together, these present findings demonstrate that E P E C blocks its own uptake into phagocytic cells by a mechanism associated with tyrosine dephosphorylation of host proteins. The EPEC-induced dephosphorylation events are dependent upon the secretion of several bacterial proteins, whose specific functions remain undefined. Research is on-going in the laboratory to further elucidate the mechanism by which EPEC mediates its anti-phagocytic phenotype. ACKNOWLEDGEMENTS  We would like to thank Rebekah DeVinney for helpful discussions and technical advice. D.L.G. was supported by a Canadian Natural Sciences and Engineering Research  185  Council Post-Graduate Scholarship, J. C. by a post-doctoral fellowship from the Fondation pour la Recherche Medicale, and B.K. by a post-doctoral fellowship from the Human Frontiers Program. This work was supported by a Howard Hughes International Research Scholar Award and operating grants from the Medical Research Council of Canada to B.B.F.  REFERENCES  1. Abe, A., B. Kenny, M . Stein, and B.B. Finlay. 1997. Characterization of two virulence proteins secreted by rabbit enteropathogenic Escherichia coli, EspA and EspB, whose maximal expression is sensitive to host body temperature. Infect. Immun. 65: 3547-3555.  2. Andersson, K., N.Carballeira, K.E. Magnusson, C. Persson, O.Stendahl, H. WolfWatz, and M.Fallman. 1996. YopH of Yersinia pseudotuberculosis interrupts early phosphotyrosine signaling associated with phagocytosis. Mol. Micro. 20:1057-1069.  3. Baldwin, T.J., W.Ward, A. Aitken, S. Knutton, and P.H. Williams. 1991. Elevation of intracellular free calcium levels in HEp-2 cells infected with enteropathogenic Escherichia coli. Infect. Immun. 59:1599-1604.  4. Bliska, J.B., K.L. Guan, J.E. Dixon, and S. Falkow. 1991. Tyrosine phosphate hydrolysis of host proteins by an essential Yersinia virulence determinant. Proc. Natl. Acad. Sci. USA. 88: 1187-91.  186  5. Bolin, I., L.Norlander, and H. Wolf-Watz. 1982. Temperature-inducible outer membrane protein of the Yersinia pseudotuberculosis and Yersinia enterocolitica is associated with the virulence plasmid. Infect. Immun. 37:506-512.  6. Donnenberg, M.S., S.B. Calderwood, A. Donohue-Rolfe, G.T. Keusch, and J.B. Kaper. 1990. Construction of and analysis of TnphoA mutants of enteropathogenic Escherichia coli unable to invade HEp-2 cells. Infect. Immun. 58:1565-1571.  7. Donnenberg, M.S. and J.B. Kaper. 1991. Construction of an eaeA deletion mutant of enteropathogenic Escherichia coli by using a positive-selection suicide vector. Infect. Immun. 59:4310-4317.  8. Dytoc, M., L.Fedorko, and P.M. Sherman. 1994. Signal transduction in human epithelial cells infected with attaching and effacing Escherichia coli in vitro. Gastroenterology. 106:1150-1161.  9. Foubister, V., I. Rosenshine, and B.B. Finlay. 1994. A diarrheal pathogen, enteropathogenic Escherichia coli (EPEC), triggers a flux of inositol phosphates in infected epithelial cells. J. Exp. Med. 179:993-998.  10. Foubister,V., I. Rosenshine, M.S. Donnenberg, and B.B. Finlay. 1994. The eaeB gene of enteropathogenic Escherichia coli is necessary for signal transduction in epithelial cells. Infect. Immun. 62:3038-3040.  11. Gomez-Duarte, O.G. and J.B. Kaper. 1995. A plasmid-encoded regulatory region activates chromosomal eaeA expression in enteropathogenic Escherichia coli. Infect. Immun. 63:1767-1776.  187  12. Inman, L.R. and J.R. Cantey. 1983. Specific adherence of Escherichia coli (strain RDEC-1) to membranous (M) cells of the Peyer's patch in Escherichia coli diarrhea in the rabbit. J. Clin. Invest. 71:1-8.  13. Jerse, A.E., J . Y u , B.D.Tall, and J.B. Kaper. 1990. A genetic locus of enteropathogenic E. coli necessary for the production of attaching and effacing lesions on tissue culture cells. Proc. Natl. Acad. Sci. USA. 87: 7839-7843.  14. Kenny, B. and B.B. Finlay. 1995. Protein secretion by enteropathogenic Escherichia coli is essential for transduction signals to epithelial cells. Proc. Natl. Acad. Sci. USA 92:7991-7995.  15. Kenny, B., L.C.Lai, B.B.Finlay, and M.S. Donnenberg. 1996. EspA, a protein secreted by enteropathogenic Escherichia coli, is required to induce signals in epithelial cells. Mol. Micro. 20:313-323.  16. Kenny, B. and B.B.Finlay. 1997. Intimin-dependent binding of enteropathogenic E. coli (EPEC) to host cell triggers novel signaling events including tyrosine phosphorylation of PLCy-1. Infect. Immun. 65:2528-2536.  17. Kenny, B., A. Abe, M . Stein, and B. B. Finlay. 1997. Enteropathogenic Escherichia coli protein secretion is induced in response to conditions similar to those found in the gastrointestinal tract. Infect. Immun. 65: 2606-2612.  188  18. Kenny, B., R. DeVinney, M . Stein, D J . Reinscheid, E.A. Frey, and B.B. Finlay. 1997. Enteropathogenic Escherichia coli (EPEC) transfers its receptor for intimate adherence into mammalian cells. Cell 91:511-520.  19. Laemmli, U.K. 1970. Cleavage of structural proteins during the assembly of the head of bacteriophage T4. Nature 227:680-685.  20. Lai, L.C., L.A.Wainwright, K.D.Stone, and M.S. Donnenberg. 1997. A third secreted protein that is encoded by the enteropathogenic Escherichia coli pathogenicity island is required for transduction of signals and for attaching and effacing activities in host cells. Infect. Immun. 65:2211-2217.  21. Levine, M . M , and R. Edelman. 1984. Enteropathogenic Escherichia coli of classic serotypes associated with infant diarrhea: epidemiology and pathogenesis. Epidemiol. Rev. 162:1285-1292.  22. Ralph, P. and I. Nakoinz. 1975. Phagocytosis and cytolysis by a macrophage tumour and its cloned cell line. Nature 257: 393-394.  23. Raschke, W.C., S. Baird, P. Ralph, and I. Nakoinz. 1978. Functional macrophage cell lines transformed by Abelson leukemia virus. Cell 15:261-267.  24. Rosenshine, I., Donnenberg, M.S., Kaper, J.B., and Finlay, B.B. 1992. Signal exchange between enteropathogenic Escherichia coli (EPEC) and epithelial cells: EPEC induces tyrosine phosphorylation of host cell protein to initiate cytoskeletal rearrangment and bacterial uptake. EMBO J. 11:3551-3560.  189  25. Rosenshine, I., S. Ruschkowski, M . Stein, D.J. Reinscheid, S.D. Mills, and B.B. Finlay. 1996. A pathogenic bacterium triggers epithelial signals to form a functional bacterial receptor that mediates actin pseudopod formation. EMBO. J. 15:2613-2624.  26. Rosqvist, R., I. Bolin, and H. Wolf-Watz. 1988. Inhibition of phagocytosis in Yersinia pseudotuberculosis: a virulence plasmid-encoded ability involving the Yop2b protein. Infect. Immun. 56:2139-2143.  27. Rosqvist, R., A. Forsberg, M . Rimpilainen, T. Bergman, and H. Wolf-Watz. 1990. The cytotoxic protein YopE of Yersinia obstructs the primary defence. Mol. Micro. 4:657-667.  28. Rosqvist, R., A. Forsberg, and H. Wolf-Watz. 1991. Intracellular targeting of the Yersinia YopE cytotoxin in mammalian cells induces actin microfilament disruption. Infect. Immun. 59:4562-4569.  29. Sambrook, J., E.F. Fritsch, and T. Maniatis. 1989. Molecular cloning, a laboratory manual. Cold Spring Harbor Laboratory Press, Cold Spring Harbor, N.Y., USA.  30. Siebers, A. and B.B. Finlay. 1996. M cells and the pathogenesis of mucosal and systemic infections. Trends in Micro. 4:22-29.  31. Wolff, C , I. Nisan, E. Hanski, G . Frankel, and I. Rosenshine. 1998. Protein translocation into host epithelial cells by infecting enteropathogenic Escherichia coli. Mol. Micro. 28:143-155.  190  TABLE 1. Bacterial strains and plasmids used Strain or plasmid  Genotype or description  Reference or source  Strains EPEC Wild type  Donnenberg et al.  E2348/69 with espB deletion  Gomez-Duarte  E2348/69 with espA deletion  Kenny et al.  UMD870  E2348/69 with espD deletion  Lai etal. 1997.  CVD206  E2348/69 with eaeA deletion  Donnenberg and  cjhr.-.TnPhoA  Donnenberg et al.  bfpAwTnPhoA  Jerse etal. 1990.  YPIII(pIB102)  Wild type  Bolin et al. 1982.  YPIII(p-)  plasmid cured  Bolin etal. 1982.  Encodes the per regulon  Gomez-Duarte  E2348/69 1990. UMD864 and Kaper 1995. UMD872 1996.  Kaper 1991. cfm 14-2-1(1) 1990. 31-6-1 Yersinia pseudotuberculosis  Plasmid pCVD450 and  Kaper  1995.  191  Table 2. EPEC secreted proteins have no detectable protein tyrosine phosphatase activity in vitro  Sample  Absorbance (405 nm)  No sample  1.190 ±0.063  Yops from Y. pseudotuberculosis  0.334 ± 0.023^  Yops from Y. pseudotuberculosis +  0.715 ±0.164^  fl  pervandate  a  EPEC-secreted proteins  1.223 ±0.032  EPEC-secreted proteins + pervanadate  1.258 ±0.010  Values are means ± standard errors of the means (3 samples per group). Significantly b  different (p < 0.01) from value for no sample control group as calculated by Student ttests.  192  FIGURE LEGENDS  Figure 1. EPEC inhibits its own uptake by J774 cells. J774 monolayers were infected with various EPEC strains for 3 h and prepared for fluorescence microscopy studies as described in Materials and Methods. Data represent averages and standard errors of the means for three independent experiments performed in duplicate. * indicates a significant difference from wild type at p < 0.05.  Figure 2. EPEC-mediated anti-phagocytosis is evident after 120 min infection of J774 cells and correlates with increasing cell associated bacteria. J774 cells were infected with EPEC or cfm for 60, 120, or 180 min. Monolayers were prepared for fluorescence microscopy studies as described in the Materials and Methods. (A) The number of total cell-associated bacteria was determined as a function of time. (B) The percentage of cellassociated bacteria that were intracellular was determined for 60, 120, and 180, min infections. (C) The percentage of intracellular bacteria was determined for cells with 0-5, 5-15, or 15-50 bacteria associated per cell 180 min after bacterial addition to the macrophages. Data represent averages and standard errors of the means (S.E.M's) of three independent experiments performed in duplicate.  Figure 3. Signaling competent EPEC induce tyrosine dephosphorylation in J774 cells. J774 cells were infected with various EPEC strains for 3 h prior to isolation of J774 cell proteins as described in Materials and Methods. Samples were resolved by SDS-PAGE (8%), transferred onto nitrocellulose, and probed with antiphosphotyrosine antibodies. Molecular weights of proteins are in kilodaltons (kD). * indicates proteins that are tyrosine dephosphorylated in response to signaling-competent EPEC infection. The arrow indicates the phosphorylated form of Tir.  193  Figure 4. EPEC-induced tyrosine dephosphorylation is evident after 120 min infection. Monolayers were left uninfected or infected with wild type EPEC or cfm for 60, 120, and 180 min prior to isolation of J774 cell proteins as described in Materials and Methods. Samples were resolved by SDS-PAGE (8%), transferred to nitrocellulose, and probed with antiphosphotyrosine antibodies. Molecular weights of proteins are in kilodaltons (kD). * indicates proteins that are tyrosine dephosphorylated in response to signalingcompetent EPEC infection. The arrow indicates the phosphorylated form of Tir.  Figure 5. EPEC-mediated anti-phagocytosis is dependent upon tyrosine dephosphorylation. J774 cells were infected with EPEC or cfm for 90 min prior to a 60 min incubation with 0.1 mM VO4 and 2 mM H2O2. Data represent averages and standard errors of the means for three independent experiments performed in duplicate. ** indicates a significant difference from wild type at p< 0.01 as obtained by Student ttests.  194  KXH T  2  * T  50A  0  < a,  u w  OH  GO  w  T  CQ ex  I  Q ex CO  EPEC strains  195  196  197  - pervanadate  + pervanadate  m  EPEC  •  cfm  Appendix 2: Cdc42-homolgous protein (Chp) involvement in EPEC pedestal formation  Based on the work presented in Chapter 4, Dan Kalman and Orion Weiner at UCSF continued to characterize the recruitment of N-WASP further. Since WASP was recruited to the pedestal through its GBD, they attempted to find the GTPase involved. Previous reports demonstrated that known GTPases were not involved in pedestal formation (Kalman, Weiner and Goosney, unpub. obs; Ben-Ami, et al., 1998, Ebel, et al., 1998). Dominant negative Cdc42, Rac, and Rho had no effect on pedestal formation, nor did traditional inhibitor treatments (Clostridium  difficile  ToxB was used to specifically  glucosylate and inhibit the Rho family of GTPases and compactin was used to inhibit isoprenylation of all GTPases ). A newly identified GTPase, Chp (Cdc42-homologous protein) was considered as a candidate for WASP recruitment to the pedestal, since it bound WASP in vitro (Aronheim, et al., 1998). Chp is highly homologous to Cdc42, but has additional N- and C-terminal domains. The C-terminus precludes any lipid modification (making it resistant to inhibition by compactin ). The effector domain of Chp contains a threonine at position 63 that is conserved in other Rho family members and required for GTPase activity (Scheffzek, et al., 1997). The threonine residue is also the site of glucosylation in Cdc42 by C. difficile ToxB. Despite this conserved residue, the residues surrounding Thr61 were not conserved in Chp, suggesting that it may not be sensitive to ToxB glucosylation. Purified active Chp and Cdc42 were used in an UDP-glucosylation assay to determine Chp's sensitivity to the toxin (Fig.A2.1). The labeled glucose was readily incorporated into Cdc42 but not Chp. These data show that Chp is the first known Rho GTPase that is insensitive to ToxB, potentially explaining the resistance of EPEC pedestals to drug treatment.  199  Both endogenous and myc-tagged Chp were localized to the pedestals, although localization improved somewhat when co-expressed with WASP or N-WASP (FigA2C). Several mutants of Chp were tested for function in the pedestal, including a point mutation in the active site of the GTPase that binds no nucleotide (G43A), a point mutation that allows binding to GDP only (S45N) and an allele that does not bind effectors but can bind xanthine instead of guanine nucleotides (Q89L/D146N). None of these alleles blocked pedestal formation. Therefore, a role for Chp could not be determined in mediating pedestal formation. However, the subcellular localization of Chp and the ToxB-insensitivity suggests that an undefined member of the Chp sub-family likely participates in pedestal formation by recruiting WASP to the pedestal tip.  Fig.A2. The Cdc42-related protein Chp is insensitive to ToxB and localizes to EPEC pedestals. A. Effector domains of Rho family GTPases. The threonine (T; arrow) is conserved in all Rho family GTPases, including Chp, and serves as a site of glucosylation by C. difficile ToxB, in Cdc42, Ras, Racl and RhoA, Althoguh several amino acids are conserved among the family members (square boxes) the arginine residues (R, circles) in Chp are not found in effector domains from any other GTPase inside or outside the Rho family. B. ToxB catalyses glucosylation of purfied Cdc42 but not Chp. ToxB and 14CUDP-glucose were incubated together with purified Chp (0.1 ug, lane 1; 2ug, lane 2; and 5u.g, lane 3) or purified Cdc42 (O.lug lane 4; 0.5ug, lane 5). Only Cdc42 incorporated the labeled glucose. Chp protein was detected by western blotting at the arrow above the 32 kDa marker. C. Chp colocalizes with transfected N-WASP in the pedestal tip. Cells were transfected with N-WASP-WT and Chp-WT, exposed to EPEC and stained with DAPI, rhodamine phalloidin, a-myc/FITC-conjugated mouse secondary antibody to recognize Chp, and cc-HA/Cy5-conjugated rat secondary antibody to recognize N-WASP. In the upper merged image, EPEC are pseudocoloured blue, actin red, and Chp green. In  200  the lower merged image, EPEC are pseudocoloured blue, actin red, and N-WASP green. Scale bar represents 1p.m.  cdc42 : K > \ p i Rail : K \ 1 i kho\ : V \ V p i ,  (hp B  K  ToxB  Chp cdc42 Chpcdc42-  \  i  p  + +  + +  V F 1) \ V r 1) \ V 1 1 N  i  1. 1) 1  + +  \  CHP-WT  i  + + - + +  NWASP-WT -47 -32 -25 -17  201  Appendix 3: Actin incorporation into EPEC pedestals The presence of N-WASP, the Arp2/3 complex, and other components of the actin polymerixation/depolymerization machinery (VASP, ADF/cofilin, gelsolin) suggests that the actin in the EPEC pedestal is dynamic. To determine whether EPEC pedestals were treadmilling (and if so, where monomeric actin was incorporating into the EPEC pedestal), HeLa cells were infected with EPEC for 5 hours and permeabilized in 0.2% saponin in the presence of 0.45um rhodamine-labeled actin (Molecular Probes), 20mM HEPES pH 7.5, 138mM KC1, 4mM MgCl , 3mM EGTA, ImM ATP and 1% 2  BSA. Rhodamine actin was incorporated for 30 seconds. Cells were then washed with PBS, fixed in 2.5% PFA, and stained for filamentous actin using Alexa 488 phalloidin (Molecular Probes) and EPEC using DAPI (lu.g/ml; Sigma). The results are shown in Fig.A3. These data suggest that the EPEC pedestal is incorporating actin at the tip of the pedestal, just beneath adherent bacteria. As the actin treadmills, it becomes incorporated into the filamentous actin (Fig. A3B- yellow staining represents colocalization of rhodamine monomeric actin and filamentous actin (pseudocoloured green)). Actin incorporation usually occurs at the barbed, or fast-growing, ends, suggesting the actin is oriented with its barbed ends towards EPEC. However, TEM needs to be done to visualize actin arrangements in the pedestal to confirm this hypothesis.  Fig.A3 Monomeric actin incorporates into the EPEC pedestal directly beneath the bacterium. Rhodamine-labeled monomeric actin was added to existing EPEC pedestals as described above. It added to the tip of the pedestal (A) and was incorporated into the filmentous actin (B). EPEC are stained blue, monomeric actin, red, and filamentous actin, green.  202  Appendix 4: Contribution of others Scientific research is a collaborative pursuit of knowledge. The work presented in this thesis would not have been as successful without the help of others. This appendix outlines the contribution of myself and others to the data collection.  Chapter 3: All experiments and writing were done by myself at UBC.  Chapter 4: This research was done at University of California, San Francisco from March 1998-June 1998 by myself, Daniel Kalman (a post-doctoral fellow in J. Michael Bishop's labaortory), and Orion Weiner (a graduate student in John Sedat's laboratory). Dan and I did all the transfections, EPEC infections, and immunofluorescence preparations. Orion acquired images on the multiwave wide-field three-dimensional microscopy system using a constrained iterative deconvolution algorithm to remove out-of-focus light. The three of us contributed equally to the intellectual input of this project. Dan has given me permission to use the data we collected together for Chapter 4 and to use his Chp data for Appendix 2.  Chapter 5: All experiments and writing were done by myself at UBC. Special thanks to Myriam deGrado (post-doctoral fellow, Finlay lab) for help setting up the yeast twohybrid system.  Chapter 6: All experiments and writing were done by myself at UBC. His-Tir and its trunctations were initially constructed by Elizabeth Frey (post-doctoral fellow; Finlay and Strynadka laboratories) for use in crystallography studies. Samantha Gruenheid (postdoctoral fellow, Finlay lab) was the primary person involved in the construction the GFPoc-actinin for the transfection studies.  204  Appendix 5: Publications arising from this graduate work Peer-reviewed journals 2001  Gruenheid, S., DeVinney, R., F. Bladt, D.L. Goosney, A . Pawson, and B.B.Finlay. The adaptor molecule Nek binds Tir and recruits N - W A S P and the Arp2/3 complex in EPEC but not in E H E C . Submitted, Nature Cell Biology.  2001  DeVinney, R., Puente, J.L., Goosney. D.L.. Gauthier, A., and B.B. Finlay. 2001. Enterohemorrhagic E. coli uses a different Tirbased mechanism of pedestal formation than enteropathogenic E. coli. Submitted, Molecular Microbiology.  2001  Goosnev. D.L.. DeVinney, R., and B.B. Finlay. 2001. Recruitment of cytoskeletal and signaling proteins to enteropathogenic and enterohemorrhagic E. coli pedestals. In Press, Infection and Immunity.  2000  Goosney. D.L.. DeVinney, R., Pfuetzner, R., Frey, E.A., Stryndaka, N.C., and B.B. Finlay. 2000. Enteropathogenic E. coli translocated intimin receptor, Tir, interacts with a-actinin directly. Current Biology 10:735-8.  2000  Goosney. D.L.. Gruenheid. S.. and B.B. Finlay. 2000. Gut feelings: EPEC interactions with the host. Annual Reviews in Cell and Developmental Biology 16:173-189.  205  1999  Kalman, D., Weiner, O.D., Goosney. D.L.. Sedat, J., Finlay, B.B., Abo, A., and J.M. Bishop. 1999. The Wiscott Aldrich Syndrome family of proteins and the ARP2/3 complex participate in actin pedestal formation initiated by enteropathogenic E. coli.. Nature Cell Biology. 1:389-91.  1999  Goosney. D.L.. deGrado. M.. and B.B. Finlay. 1999. Putting E. coli on a pedestal: a unique system for studying signal transduction and the actin cytoskeleton. Trends in Cell Biol. 9:11-14.  1999  Goosnev. D.L.. Knoechel. D.G.. and B.B. Finlav. 1999. Eneteropathogenic E. coli, Salmonella, and Shigella: masters of  host cell cytoskeletal exploitation. Emerg. Infect. Dis. J. 5:216223.  1999  Goosnev. D.L.. J. Celli, B. Kenny, and B.B. Finlay. 1999. Enteropathogenic Escherichia coli (EPEC) inhibits phagocytosis. Infect. Immun. 67:490-95.  Abstracts  2000  Gruenheid, S., Goosney. D.L.. DeVinney, R., and B.B. Finlay. Molecular characterization of EPEC pedestals. American Society for Cell Biology, Dec. 9-13. San Francisco, CA.  206  2000  Goosney. D.L.. DeVinney, R., Gruenheid, S., B.B. Finlay. Enteropathogenic E. coli translocated intimin receptor, Tir, mediates actin cytoskeleton rearrangements in host epithelial cells. Frontiers in Cellular Microbiology and Cell Biology, Oct. 7-12 Toulon, France.  2000  DeVinney, R., Goosney. D.L.. Gruenheid, S. and B.B. Finlay. Subversion of host signalling pathways by enteropathogenic Escherichia coli and enterohemorrhaghic Escherichia coli. UA-  UC Conference on Infectious Diseases, May 14-17, Panorama, B.C.  2000  Goosney. D.L.. DeVinney, R., Pfuetzner, R., Frey, E.A., Gruenheid, S., Strynadka, N., and B.B. Finlay. Enteropathogenic E. coli Tir links to the mammalian cell cytoskeleton via a direct interaction with a-actinin. The Dynamics of the Cytoskeleton, Feb 3-9, 2000, Keystone, CO.  1999  Goosney. D.L.. and B.B. Finlay. The effects of E. coli on the actin cytoskeleton. Workshop on Mathematical Cellular Biology, Pacific Institute for the Mathematical Sciences, Aug 16-27, 1999, Vancouver, B.C.  1999  Goosney. D.L.. Celli. J.. and B.B. Finlay. Inhibition of phagocytosis by enteropathogenic Escherichia coli (EPEC). INSERM 105 meeting - methods and analysis of phagocytosis:  207  application to bacterial infections, March 11-12, 1999, Le Vesinet, France.  Goosney. D.L., Celli, J., Kenny, B., and B.B. Finlay. Enteropathogenic E. coli inhibits phagocytosis. 2nd Louis Pasteur Conference on Infectious Diseases. Paris, France.  Goosney. D.L.. Kenny, B., DeVinney, R., Stein, M., Reinscheid, D.L, Frey, E.A., Kalman, D., Weiner, O.D., Bishop, J.M., Sedat, L, Bourne, H., and B.B. Finlay. Bacterial piracy: Enteropathogenic Escherichia coli (EPEC) hijacks the host actin cytoskeleton to form pseudopods at the plasma membrane. Motile and Contractile Systems, Gordon Research Conference, ColbySawyer College, New London, New Hampshire, U.S.A.  Goosney. D.L.. Kenny, B., and B.B. Finlay. Enteropathogenic Escherichia coli (EPEC) inhibits phagocytosis by cultured macrophages. Microbial Pathogenesis and Host Defense Meeting, Cold Spring Harbor Laboratory, New York, U.S.A.  208  

Cite

Citation Scheme:

        

Citations by CSL (citeproc-js)

Usage Statistics

Share

Embed

Customize your widget with the following options, then copy and paste the code below into the HTML of your page to embed this item in your website.
                        
                            <div id="ubcOpenCollectionsWidgetDisplay">
                            <script id="ubcOpenCollectionsWidget"
                            src="{[{embed.src}]}"
                            data-item="{[{embed.item}]}"
                            data-collection="{[{embed.collection}]}"
                            data-metadata="{[{embed.showMetadata}]}"
                            data-width="{[{embed.width}]}"
                            async >
                            </script>
                            </div>
                        
                    
IIIF logo Our image viewer uses the IIIF 2.0 standard. To load this item in other compatible viewers, use this url:
http://iiif.library.ubc.ca/presentation/dsp.831.1-0090489/manifest

Comment

Related Items