UBC Theses and Dissertations

UBC Theses Logo

UBC Theses and Dissertations

Novel overlapping roles of Salmonella pathogenicity islands 1 and 2 in intestinal salmonellosis Coburn, Bryan 2006

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

Item Metadata

Download

Media
831-ubc_2007-266977.pdf [ 23.16MB ]
Metadata
JSON: 831-1.0100343.json
JSON-LD: 831-1.0100343-ld.json
RDF/XML (Pretty): 831-1.0100343-rdf.xml
RDF/JSON: 831-1.0100343-rdf.json
Turtle: 831-1.0100343-turtle.txt
N-Triples: 831-1.0100343-rdf-ntriples.txt
Original Record: 831-1.0100343-source.json
Full Text
831-1.0100343-fulltext.txt
Citation
831-1.0100343.ris

Full Text

N O V E L OVERLAPPING ROLES 0¥_SALMONELLA PATHOGENICITY ISLANDS 1 AND 2 IN INTESTINAL SALMONELLOSIS By B R Y A N C O B U R N BSc. (Hon), University of Toronto, 2000 A THESIS SUBMITTED IN PARTIAL F U L F I L L M E N T OF THE REQUIREMENTS OF THE DEGREE OF COMBINED DOCTOR OF MEDICINE A N D DOCTOR OF PHILOSOPHY in THE F A C U L T Y OF G R A D U A T E STUDIES (Microbiology and immunology) THE UNIVERSITY OF BRITISH C O L U M B I A April, 2006 © Bryan Coburn, 2006 Abstract Non-typhoidal Salmonella species are a significant cause of human diarrheal disease, incurring worldwide morbidity and mortality. The prevailing dogma arising from animal models of Salmonella enteropathogenesis is that the virulence associated genomic regions, Salmonella pathogenicity island (SPI) -1 and SPI-2, are essential for intracellular invasion/intestinal disease and intracellular survival/systemic disease, respectively. This paradigm partly reflects limitations of animal models currently used to study in vivo pathogenesis! In this thesis, a new model of murine Salmonella enteropathogenesis is presented which allows a novel examination of this theoretical dichotomy. Using this model, SPI-2 was shown to be required for complete enteropathogenesis in Salmonella enterica serovar Typhimurium infection. In addition, murine and bovine intestinal inflammation was identified in the absence of SPI-1, previously thought to be essential for intestinal disease. These findings are corroborated in human disease by the identification of a SPI-1 deficient human clinical diarrheal Salmonella enterica isolate. These strains were isolated from patients affected with severe diarrheal disease in Shenzhen, China. These are the first findings that demonstrate that SPI-2 is required for intestinal pathogenesis early in murine infection, and that SPI-1 is dispensible for enteropathogenesis in animal and human infections with S. enterica. These observations indicate that disease models, diagnostic and therapeutic approaches predicated on the requirement for SPI-1 in intestinal disease do not accurately describe intestinal salmonellosis. n Table of Contents Abstract i i Table of Contents i i i List of Tables vi List of Figures vii Non-standard Abbreviations ix Acknowledgements x Chapter 1 - Introduction 1 1.1 Epidemiology and natural history of and therapeutic approach to Salmonella enterica infections 2 1.1.1 Typhoid 3 1.1.2 Enterocolitis and Diarrhea 4 1.2 Model infections in Salmonella pathogenesis 5 1.2.1 In vitro infection models 6 1.2.2 Murine typhoid 9 1.2.3 Bovine and Rabbit Typhimurium infection (enterocolitis) 11 1.3 Fundamentals of S. enterica serovar Typhimurium pathogenesis 13 1.3.1 Salmonella pathogenicity islands : 14 1.3.2 SPI-1: the invasion associated locus 15 1.3.3 SPI-2: the intracellular survival locus 17 1.3.4 Determinants of S. enterica intestinal virulence.'. 18 1.3.5 The SPI-l/SPI-2 dogma ; 20 1.3.6 Hints from the literature: Data discordant with the prevailing model of pathogenesis 21 1.4 A new model for assessing the role of SPI-1 and SPI-2 in intestinal pathogenesis 24 1.5 Research objectives 24 Chapter 2 - Murine enterocolitis caused by Salmonella infection 26 2.1 Introduction 27 2.2 Results 29 2.2.1 Oral Streptomycin treatment causes transient diarrhea, kills normal flora and decreases host resistance to Salmonella enterica serovar Typhimurium infection 29 2.2.2 Inflammatory features of murine serovar Typhimurium enterocolitis include changes in the lumen, surface epithelium, mucosa and submucosa of infected ceca 30 2.2.3 Pathology scores correlate with external quantitative criteria of inflammation severity 35 2.2.4 Signs of intestinal inflammation occur by 8 hours post-infection and plateau between 24-48 hours 36 2.2.5 Inflammatory changes occur with low bacterial doses 36 2.2.6 Inflammatory severity is comparable in N R A M P f 6 ™ and N R A M P l s u s c e p t i b l e mouse strains 39 2.3 Discussion 41 Chapter 3 - Salmonella typhimurium pathogenicity island 2 is necessary for complete virulence in a mouse model of infectious enterocolitis 44 .3.1 Introduction 45 3.2 Results 47 i n 3.2.1 S. enterica serovar Typhimurium elicits intestinal inflammation that is most pronounced in the cecum of streptomycin pretreated mice. 47 3.2.2 SPI-2 contributes to cecal inflammation in serovar Typhimurium infection 47 3.2.3 SPT2 mutants induce ICAM-1 expression and neutrophil recruitment less strongly than wild-type serovar Typhimurium 49 3.2.4 S. enterica serovar Typhimurium in the ceca of infected mice is predominantly lumenal and sensitive to gentamicin 51 3.2.5 SPI-2 is expressed prior to penetration of the intestinal epithelium 52 3.3 Discussion 54 Chapter 4 - Analysis of the contribution of Salmonella Pathogenicity Island 1 and 2 to enteric disease progression 56 4.1 Introduction 57 4.2 Results 59 4.2.1 Ceca of streptomycin-treated mice infected with SPI-1 and SPI-2 mutants show attenuated pathology at 48 h post-infection 59 4.2.2 Murine ceca infected with SPI-l/SPI-2 double mutants display little pathology at 5 days after infection, while ceca infected with SPI-1 mutants show severe pathological changes. 61 4.2.3 S. enterica serovar Typhimurium invades the intestinal mucosa by 5 days after infection following streptomycin treatment 62 4.2.4 Calf intestinal loops infected with SPI-1 and SPI-2 mutants show reduced secretory response and similar pathological changes at 24 h post infection 63 4.2.5 Calf ileal loops infected with SPI-2 mutants display decreased pathology at 5 days after infection 64 66 4.2.6 SPI-1 mutants produce severe intestinal pathology in calf ileal loops at five days post infection 66 4.2.7 Both SPI-1 and SPI-2 independent intestinal inflammation require aromatic amino acid biosynthesis 67 4.3 Discussion 70 Chapter 5: Discussion 73 5.1 The non-dichotomous role of Salmonella Pathogenicity Islands 1 and 2 in Salmonella enterica serovar Typhimurium infection 74 5.1.1 The murine entercolitis model of intestinal salmonellosis 74 5.1.2 SPI-2 in enterocolitic salmonellosis 75 5.1.3 SPI-1 independent colitis 79 5.1.4 Caveats of experimental murine enterocolitis 82 5.1.5 Human Diarrheal disease can be induced by SPI-1 deficient S. enterica 83 5.2 A synthesized, non-dichotomous model of Salmonella enteropathogenesis 87 5.3 Implications for research: Salmonella, inflammation and the host 89 5.4 Directions for intestinal salmonellosis research 89 Chapter 6: Methods 93 6.1 Methods used throughout thesis 93 6.1.1 Bacterial culture 93 6.1.2 Murine enterocolitis (Streptomycin pre-treatment) infections 93 •6.1.3 Bacterial enumeration 93 6.1.4 Histopathology 93 6.1.5 Immunohistochemistry 94 iv 6.1.6 Quantitative measures of inflammation 94 6.1.7 Statistical analysis 95 6.2 Methods from Chapter 2 95 6.2.1 Assessment of Normal Flora 95 6.3 Methods from Chapter 3 95 6.3.1 Mast cell staining 95 6.3.2 Gentamicin protection assay 96 6.3.3 Murine ligated ileal loop infections 96 6.3.4 Gene expression reporter (RIVET) assays..... 96 6.4 Methods from Chapter 4 97 6.4.1 Calf intestinal loop surgeries 97 6.4.2 Calf intestinal tissue collection, specimen handling and histopathology 98 6.5 Methods from Chapter 5 98 6.5.1 Bacterial strains 98 6.5.2 Bacterial Identification 98 6.5.3 Pulse-Field Gel Electrophoresis(PFGE) 98 6.5.4 PCR Assay 99 6.5.5 Preparation of PCR probes 100 6.5.6 Restriction endonuclease digestion and Southern transfer: 100 6.5.7 Southern Blot hybridization: 100 6.5.8 Invasion Assay 101 Chapter 7: References 102 Appendix 1: List of Peer-reviewed published primary research 119 Appendix 2: Animal Ethical Approvals 120 v L i s t of Tables Table 1.1 Animal models of Salmonella enterica infection 6 Table 1.2 Genes in the SPI-1 and SPI-2 pathogenicity islands and their classes of function 14 Table 1.3 Salmonella effectors secreted by the SPI-1 type III secretion system 16 Table 1.4 Salmonella effectors secreted by the SPI-2 type III secretion system 20 Table 1.5 Landmark papers of Salmonella in vivo pathogenesis 21 Table 1.6 Non-SPI-1 secreted virulence factors tested in animal models of intestinal salmonellosis 22 Table 2.1 - Salmonella murine enterocolitis histopathology scoring criteria 34 Table 2.2 The timecourse of intestinal and systemic illness in Salmonella enterocolitis 38 Table 3.1 Pathological features of Salmonella enterica enterocolitis after 48 hours infection induced by strains lacking SPI-1 and SPI-1 type III secretion...: 49 Table 5.1 Features of early and late murine enterocolitis 76 Table 5.2 Characteristics of human diarrheal strains and controls. 87 vi Lis t of Figures Figure 1.1 Interactions of Salmonella and host cells in vitro 8 Figure 0 Type III secretion system 15 Figure 1.3 The prevailing view of pathogenicity island involvement in Salmonella enterica infection in vivo 23 Figure 2.1 The histopathological features of murine Salmonella enterica induced cecal inflammation 29 Figure 2.2 The histopathological features of murine Salmonella enterica induced cecal inflammation 31 Figure 2.3 Pathological changes in murine Salmonella enterica induced intestinal inflammation. 32 Figure 2.4 Histopathology scores correlate with objective quantitative criteria of inflammation in Salmonella infected murine ceca 35 Figure 2.5 The timecourse of inflammatory pathogenesis in the ceca of mice infected with Salmonella enterica serovar Typhimurium 37 Figure 2.6 Murine intestinal inflammation can be induced with low doses of Salmonella enterica serovar Typhimurium 38 Figure 2.7 Intestinal inflammation is induced in mouse ceca by S. enterica serovar Typhimurium in the presence and absence of the host resistance factor Nrampl 40 Figure 3.1 Salmonella enterica serovar Typhimurium elicited enterocolitis is most severe in the ceca of wild-type infected mice and, is partially attenuated in the absence of SPI-2 type III secretion 48 Figure 3.2 SPI-2 but not SPI-1 T3SS mutant Salmonella enterica serovar Typhimurium induces ICAM-1 in the intestines of streptomycin treated mice 48 hours after infection 50 Figure 3.3 Neutrophil infiltration is markedly reduced in Salmonella enterica serovar Typhimurium induced typhlitis in the absence of SPI-2 type III secretion (AssaR) 51 Figure 3.4 Salmonella enterica serovar Typhimurium in the ceca of infected mice is primarily extracellular 52 Figure 3.5 The majority of Salmonella enterica serovar Typhimurium are susceptible to the extracellular killing of gentamicin in the cecum and colon of infected mice 2 days after infection 53 Figure 3.6 Salmonella pathogenicity island 2 (SPI-2) expression is induced prior to invasion of vii the intestinal epithelium by Salmonella enterica serovar Typhimurium 53 Figure 4.1 Salmonella enterica serovar Typhimurium SPI-1 mutants cause delayed typhlitis in streptomycin treated mice in the presence of functional SPI-2 type III secretion 60 Figure 4.2 Histopathological changes in Salmonella enterica infected mice 60 Figure 4.3 Bacterial loads in murine Salmonella enterocolitis 61 Figure 4.4 Murine Salmonella typhlitis is characterized by bacterial invasion of the mucosa at five days but not two days post-infection 63 Figure 4.5 Salmonella enterica serovar Typhimurium SPI-1 mutants cause delayed typhlitis in inoculated bovine ileal loops, but SPI-2 mutants are attenuated 65 Figure 4.6 Histopathological changes in calf ileal loops at 24h and 5 days post-infection 66 Figure 4.7 Both SPI-1 and SPI-2 independent delayed inflammation in Salmonella enterica infected murine ceca requires the aromatic amino acid biosynthesis gene aroA 68 Figure 4.8 Histopathological changes at 5 days post-infection in streptomycin-pretreated mice infected with wild-type Salmonella enterica serovar Typhimurium 69 Figure 5.1 Possible roles for Salmonella pathogenicity island-2 (SPI-2) type III secretion in early enterocolitis induced by Salmonella enterica serovar Typhimurium 77 Figure 5.2 Distinct host proinflammatory pathways are potentially influenced by Salmonella pathogenicity islands 1 and 2 (SPIs 1 and 2) to induce inflammation 81 Figure 5.3 Pulsed field gel electrophoresis of stool isolates from an outbreak of diarrhea in Shenzhen, China 84 • Figure 5.4 Genetic analysis of Salmonella pathogenicity island (SPI) -1 and -2 of strains isolated from diarrheal patients in Shenzhen, China compared to control strains 85 Figure 5.5 Southern hybridization analysis for SPI-1 genes of diarrheal stool isolates from Shenzhen, China and controls : 86 Figure 5.6 A revised view of SPI-l/SPI-2 contributions to intestinal Salmonella infection, in which both SPI-1 and SPI-2 play critical and overlapping roles in the induction of disease in the intestine and where SPI2 activation and involvement in disease has both early, extracellular and late, intracellular phases 88 vm Non-standard Abbreviat ions A M P antimicrobial peptide CD (e.g. CD 18) cluster of differentiation CDC Centres for Disease Control C F U colony forming units DC dendritic cell et al et alii G A L T gut-associated lymphoid tissue HIV human immunodeficiency virus ICAM-1 intercellular adhesion molecule 1 IL (e.g. IL-1) interleukin i.p. intraperitoneal i.v. intravenous LPS lipopolysaccharide M A P K M A P kinase N F K B nuclear factor kappa B N L R Nod-like receptor P A M P pathogen associated molecular pattern PEEC pathogen elicited epithelial chemoattractant PCR polymerase chain reaction PFGE pulsed-field gel electrophoresis P M N , neutrophil polymorphonuclear leukocyte; neutrophil p.o. per os PRR pattern recognition receptor RES reticuloendothelial system S. enterica Salmonella enterica SARS severe acute respiratory syndrome SCV Salmonella containing vacuole Serovar Typhi Salmonella enterica serovar Typhi Serovar Typhimurium Salmonella enterica serovar Typhimurium Serovar Senftenberg Salmonella enterica serovar Senftenberg Serovar Dublin Salmonella enterica serovar Dublin Serovar Enteritidis Salmonella enterica serovar Enteritidis Sif Salmonella induced filaments SPI Salmonella pathogenicity island TLR Toll-like receptor TNFa tumour necrosis factor alpha T3SS type three (III) secretion system WHO World Health Organization WT wild-type Acknowledgements I would like to acknowledge the contributions of many people to the genesis of this thesis. M y supervisor, Brett Finlay, provided me with the resources, guidance, intellectual support, freedom and enthusiasm to make my time in his lab enjoyable, productive, exciting and an opportunity to learn. Other current and former lab members, in particular Carrie Rosenberger, Bruce Vallance, Nat Brown, Mark Wickham, Wanyin Deng and Brian Coombes were excellent collaborators and effective, critical sounding boards for my ideas. M y thesis supervisory committee; Bob Hancock, Rob McMaster and David Speert kept me focused on my academic goals while keeping an eye on the clock, yet allowed me to come through the process of maturing as a grad student in a natural way. M y family has been encouraging, supportive, and lofty in their goals for me, which helped me feel successful even when things weren't going well. M y friend and perpetual classmate Liam Brunham has been a great educational partner and friend. The conversations I've had with him have ranged from absurd, to career-changing insight. Elizabeth, my wife, is an inspiration. The quality of science that she produced as a PhD student, the hard work she's applied as a medical student, the love and support for me and my work (and everything she's done while I've been stuck in the lab, at the computer, or in some kind of scientifically induced stupor) always set the bar higher. If I ever catch up, it will be because she let me. Finally, I'd like to specially acknowledge Yuling L i . She's been a tireless help to me, and has made coming to work fun every day. She always brightens my mood and made everything I did turn out. Financial support was provided by studentships from the Canadian Institutes for Health Research and the Michael Smith Foundation for Health Research. x Chapter 1 - Introduction Communicable diseases account for 25-40% of deaths worldwide (World Health Organization, 1999), and the morbidity and mortality associated with viral, bacterial, fungal and parasitic infections is incalculable. Due to the variable nature and consequences of human infections and their disproportionate prevalence in developing countries where the diagnostic, clinical and infrastructural means to identify and track infectious diseases are frequently nonexistent, only estimates of disease occurrence are available. Such estimates indicate both a growing problem and diminishing solutions. While global threats such as the H5N1 strain of the avian influenza virus and its potential spread to humans or global outbreaks of corona-virus induced severe acute respiratory syndrome (SARS) preoccupy the developed world, old standards and recent pandemics such as tuberculosis, infectious diarrhea and human immunodeficiency virus (HIV) kil l millions of people per year in the developing world. Diarrheal disease is the greatest cause of death in children, and infections with diarrheal and other pathogens grow increasingly resistant to both historical cures and recently developed treatments (World Health Organization, 1999; World Health Organization, 2005). This thesis focuses on a pathogen with particular importance in the developing world, but also of significant health and economic impact in wealthy nations; the Gram-negative facultative intracellular Salmonella enterica (S. enterica) bacterium. 1.1 Epidemiology and natural history of and therapeutic approach to Salmonella enterica infections. Over 2500 serovars of S. enterica have been identified belonging to six subspecies (Fierer and Guiney, 2001; Ochman and Groisman, 1994). Serovars are differentiated by their flagellar, carbohydrate and lipopolysaccharide structures. Bacteria of this species cause a variety of important diseases in humans and commercial livestock (Kingsley and Baumler, 2000). Although frequently a cause of septic arthritis and focal infections of extraintestinal organs, S. enterica species are typically orally-acquired pathogens that cause one of 4 major syndromes: enteric fever (Typhoid), enterocolitis/diarrhea, bacteremia and asymptomatic carriage. Chronic carriage was made infamous by 'Typhoid' Mary Mallon, a 19 t h century cook and a carrier of serovar Typhi. Host adaptation of Salmonella serovars has resulted in specific associations between serovar, host and syndrome. The disease manifestation depends on both host susceptibility and the infectious S. enterica serovar (Fierer and Guiney, 2001). While younger (<20 years) and older (>70 years) individuals have the highest attack rates, susceptibility is also 2 bacteria and host species specific. In humans, serovars Typhi, Paratyphi and Sendai cause enteric fever, while most serovars cause enterocolitis/diarrhea. Several serovars including Cholerasuis and Dublin are more commonly associated with bacteremia in humans (Fierer and Guiney, 2001). While serovar Typhi is largely restricted to humans, other serovars are more broadly host adapted and cause natural animal infection. Serovars Dublin, Typhimurium and Cholerasuis cause disease in both humans and animals, but cause distinct syndromes in different hosts. Serovar Dublin causes intestinal inflammatory disease, bacteremia and abortion in cows; serovar Typhimurium causes a Typhoid-like systemic illness in mice; and serovar Cholerasuis causes septicemia in pigs (Baumler et al, 1998). Human Typhoid fever and intestinal/diarrheal disease are of particular interest, however, as they not only represent the most common syndromes associated with S. enterica infection, but also involve the pathogenic processes of both bacteria and host most thoroughly investigated in infectious models of Salmonella pathogenesis. 1.1.1 Typhoid An estimated 17-21 million cases of Typhoid with 600,000 deaths per year are caused by systemic infection with the S. enterica serovar Typhi. Typhoid is an historically important disease in both the developing and developed world, particularly in historical periods and regions of high population density and poor sanitation. It is now predominantly a disease of the developing world, where incidence is between 100-1000 times as high as in developed nations (World Health Organization, 2005). Disease occurs following the ingestion of between 103-106 bacteria, usually from contaminated water or animal products, or from close contact with an infected individual or carrier (Hornick, 1970). Disease manifests one to two weeks following inoculation with generalized fever and malaise, abdominal pain with or without other symptoms including headache, myalgias, nausea, anorexia and constipation. Diarrhea occurs occasionally but is typical only of infection in the immunocompromised. Hepatosplenomegaly is common but not present in all cases and diffuse abdominal tenderness is usual. Fever is typically mild at first and worsening as disease progresses (reviewed in (Parry et al, 2002)). In the absence of complications, disease resolves following varied periods of infection although carriage of the bacteria can continue in post-symptomatic patients for months or years and relapse occurs in a minority of patients. 3 Diagnosis cannot be made based on clinical appearance alone but requires microbiologic investigation. The primary treatment for serovar Typhi infection is fluoroquinolones, although nalidixic acid and other antimicrobial agents are also used. Treatment is effective in the vast majority of cases and decreases time to bacterial clearance, carriage rates and infection-associated morbidity and mortality (Parry et al, 2002). 1.1.2 Enterocolitis and Diarrhea Although estimates vary greatly due to a lack of consistent diagnosis and reporting, between 200 million and 1.3 billion cases of intestinal disease including 3 million deaths due to non-typhoidal Salmonella are estimated to occur each year worldwide (World Health Organization, 2005). Like Typhoid, the incidence of intestinal disease caused by non-Typhoidal Salmonella species is highest in the developing world. The disease is of considerable importance in developed countries as well, where infections due to the consumption of contaminated animal products (usually chicken or eggs) cause an estimated 1.4 million cases annually in the United States, 12,000 cases annually in Canada and over 600 deaths per year in North America (Health Canada, 2006; National Center for Infectious Disease, 2006; World Health Organization, 2005). The total economic burden of the disease in the United States alone is estimated to be 3 Billion USD (World Health Organization, 2005). Worldwide, over 75% of reported Salmonella cases are caused by two S. enterica serovars: Enteritidis and Typhimurium (World Health Organization, 2005). Disease in humans typically follows the ingestion of greater than 50,000 bacteria in contaminated food or water with symptoms occurring between 6 and 72 hours after consumption. Onset of symptoms is marked by acute onset, crampy, abdominal pain and diarrhea with or without blood. Nausea and vomiting are also common. Capable of infecting the entire bowel and commonly thought to be a disease of the ileum, inflammation in non-Typhoidal disease preferentially occurs in the large bowel, with common infections in the ileum and rare infections in the jejunem, duodenum and stomach (Boyd, 1985; Mcgovern and Slavutin, 1979). Although the primary disease associated with serovars Enteritidis and Typhimurium infection is a self-limited enterocolitis, infection is commonly associated with septic arthritis (2% of cases), bacteremia (5% of cases), sepsis and other complications such as toxic megacolon, psoas abscess and granulomata of bone or soft tissues. Enterocolitic infection in children is marked by increased inflammatory severity, bloody diarrhea, increased duration of infection and risk of complication. 4 Diagnosis of intestinal salmonellosis is made based on a positive identification of the presence of Salmonella bacteria in stool samples using a number of criteria including microbiological investigations, serotyping and bacterial biochemistry as well as newly developed molecular techniques such as polymerase chain reaction (PCR) and pulsed-field gel electrophoresis (PFGE) based genotyping. Positive blood cultures or culturable bacteria from joint aspirates are also an important diagnostic tool in cases of invasive disease, sepsis or septic arthritis (Parry et al, 2002; Rahn et al, 1992; Winokur, 2003). In the absence of treatment for gut-limited infections, symptoms usually last between 5-7 days and resolve spontaneously. Treatment of fluid and electrolyte imbalances by oral or intravenous rehydration is necessary in cases where fluid loss is substantial. In adults, specific antimicrobial therapy with amoxicillin, sulfamethoxazole/trimethoprim, or ciprofloxacin is indicated only in the presence of positive signs of invasive disease, and does not decrease the duration of illness or the severity of symptoms. Neonatal gut infection also requires treatment to prevent invasion. 1.2 Model infections in Salmonella pathogenesis Despite the numerous serovars of S. enterica capable of infecting humans and the broad spectrum of illness induced by these serovars, the study of human Salmonella pathogenesis has been simplified to the study of a limited number of serovars and two syndromes: Typhoid (enteric) fever and diarrhea/enterocolitis. Although the latter manifestation of S. enterica infection may be further subdivided based on the clinical features of intestinal illness in human infection (e.g. diarrheal versus inflammatory infection), it has largely been investigated as a single entity in animal models of disease. However, the study of human-specific serovars is confounded by their host-adaptation and the consequent limitation of serovars to specific host species (reviewed in (Kingsley and Baumler, 2000)). For example, serovar Typhi is restricted to human and higher primate hosts and does not cause disease in experimental rodents, cows or pigs. Serovars Dublin and Cholerasuis are essentially pathogens of cattle and pigs respectively, causing intestinal disease and septicemia, occasionally causing bacteremia in humans, but rarely affecting mice. In contrast, the broad-host-range serovar Typhimurium infects a wide variety of hosts but manifests in diverse clinical outcomes ranging from gut-limited intestinal inflammation and diarrhea in humans and cattle to the systemic infection 'murine typhoid' without intestinal disease during natural infections of genetically susceptible mice (Santos et al, 2001b). 5 Summarized in Table 1.1 are the models of human salmonelloses that have emerged, in which limited serovars,of S. enterica are used to model specific human diseases based on the disease induced in their respective host organisms. The most widely investigated models of human Salmonella infection are murine infection with serovar Typhimurium that models human Typhoid and bovine infection with serovars Dublin and Typhimurium that models human intestinal disease. Although important as models of colonization and diarrheal disease, discussion of chicken and rabbit infections will be limited to the contextualized discussion of subsequent sections. Table 1.1 A n i m a l models of Salmonella enterica infection Salmonella enterica Route of Intestinal Model System serovar(s) inoculation Manifestations Systemic Manifestations Reference Murine Typhimurium Oral, Mild ileal Septicemia, Loeffler, 1892 Typhoid intreperitoneal, inflammation Hepatosplenomegaly, Intravenous Death Bovine enterocolitis Ligated Loops Dublin Direct injection Secretory response, N/A due to short duration (Santos et al, into loop Inflammation 2001b) Typhimurium Direct injection Secretory response, N/A due to short duration into loop Inflammation Oral Infection Dublin Oral Diarrhea, Bacteremia, Wasting, enterocolitis Abortion, Fever Typhimurium Oral Diarrhea, +/- bacteremia enterocolitis Rabbit Ileal Typhimurium, Direct injection Secretory response, N/A due to short duration (Giannella et Loops Enteritidis into loop Inflammation al, 1973) 1.2.1 In vitro infection models Infection of cultured and primary cell lines in vitro has been critical for the exploration of Salmonella virulence, cell biology, biochemistry and pathogenesis. Although Salmonella is capable of infecting a wide variety of cells in culture conditions including dendritic cells, macrophages, hepatocytes, neutrophils, colonocytes and other epithelial cells, cells typically used are either macrophages or macrophage-like cell lines and epithelial cells and cell lines. In the context of S. enterica infection, these cell types represent professional phagocytes (macrophages) and non-phagocytic cells (epithelial cells). Alternatively, they may be considered representative of the two major tissues/systems encountered during Salmonella infection in vivo - intestinal epithelium and cells of the reticuloendothelial system (RES). Although the RES is composed of many more cell types than simply macrophages, bacterial residence in macrophages is perhaps 6 the defining feature of systemic infection in vivo (Richter-Dahlfors et al., 1997). Multiple cultured lines of each type are used including the macrophage-like R A W and ill A.l cells, and the epithelial cell lines Caco-2, T84 and HeLa. The epithelial cell-lines Caco-2 and T84 can be polarized into tight-junction containing monolayers useful for studying the bacterial disruption of epithelial integrity and the cell-biology of Sa/wo«e//a-epithelial interactions which depend on epithelial polarity. Because bacteria are believed to encounter these cell types during different stages of infection or during different diseases (Typhoid or enterocolitis), they have often been used to investigate specific pathogenic behaviours. The interaction of Salmonella with cultured cells can be divided into stages: initial contact between bacteria and host cells (adhesion); bacterial or cell-mediated engulfment of the bacteria, and; intracellular bacterial residence (Figure 1.1). For clarity, I will describe these stages of infection separately for bacteria-epithelial cell and bacteria-macrophage interactions. The bacterial virulence machinery required for these events and their impact on disease pathogenesis will be more thoroughly discussed in later sections. Multiple fimbrial adhesins are involved in the attachment of serovar Typhimurium to epithelial cells in vitro and in vivo (Baumler et al, 1996b; Baumler et al, 1996c; Baumler et al, 1996a; Baumler et al, 1997; van der Velden et al, 1998). In vitro adhesion is associated with the invasion of cultured cells, and bacteria with mutant adhesins invade epithelial cells less efficiently than wild-type (Baumler et al, 1996b; Duguid et al, 1976). The function of the adhesins in vivo is likely to promote tight attachment to tissues thereby preventing clearance of bacteria by the peristaltic activity of the bowel, and to bring bacteria into close contact with target cells so that they can mediate other contact-dependent behaviours promoting virulence, such as invasion/bacterial uptake. Indeed, mutations of Jim, Ipf, pef and agt encoded fimbriae increase the 50% lethal dose (LD 5o) of Salmonella 26-fold following oral infection (van der Velden et al, 1998), and single fimbrial mutants also attenuate the enteropathogenicity of serovar Typhimurium (Baumler et al, 1996a; Baumler et al, 1997). Attachment to phagocytes is likely mediated by host cells and serum factors, rather than fimbrial adhesion. In vitro, Salmonella is opsonized by a range of opsonins including complement and immunoglobulins resulting in binding of ospsonic receptors on phagocytes and subsequent phagocytosis of bacteria (Ishibashi and Arai, 1990; Makela et al, 1988)(Figure 1.1 A). 7 A. Attachment B. Induced Uptake C. SCV formation D. Replication E. Cytopathy/Escape M m O 0) LU Fimbrial Adhesins Cell receptors Actin Cytoskeleton Rearrangements Evasion of Phagosomal Maturation/ Sif Formation Nutrient Acquisition Apoptosis/ Cytotoxicity Opsonins/ Receptors Phagocytosis 6 2 _ 4 Evasion of Phagosomal Maturation Nutrient Acquisition - A Induced Apoptosis Figure 1.1 Interactions of Salmonella and host cells in vitro. Facultative intracellular parasitism is a central feature of Salmonella infection of cultured cells. The dynamics of bacterial-host cell interactions have been described predominantly in epithelial cells (represented here in yellow) and macrophage-like cells or primary macrophages (represented here in blue). (A) Salmonella mediate contact with target cells via specific fimbrial adhesins (epithelium) or opsonic receptors (macrophages). (B) Bacteria induce epithelia to undergo transformation of the actin cytoskeleton or undergo phagocytosis to gain access to the inside of the cell. (C) Following uptake, Salmonella evade phagolysosomal killing by forming a specialized intracellular niche, the Salmonella containing vacuole (SCV). (D) Within the SCV, Salmonella acquire and exploit host resources to replicate and (E) finally escape the cell, usually by inducing host cell death. Once Salmonella come within sufficiently close contact with target cells, they can enact other virulence behaviours. Within minutes of contact with cells, Salmonella are internalized (Figure L I B ) . While professional phagocytes are capable of engulfing bacteria by receptor-mediated phagocytosis, Salmonella can also induce bacterially-controlled invasion in these cells (Drecktrah et al, 2006). Contact of Salmonella with non-professional phagocytes results in the induction of microscopically distinct host-cell membrane rearrangements (Finlay and Falkow, 1990). In bacterially-mediated invasion, bacterial signals stimulate host CDC42, Rac-1 and Rho-GTPase signaling which are required for actin cytoskeletal rearrangements and bacterial uptake (Ffardt et al, 1998). In phagocytes, opsonin-receptor binding of opsonized Salmonella results in 8 internalization of bacteria into 'spacious phagosomes', an altered macropincoytic compartment. Spacious phagosome formation can also occur in the absence of opsonization (Alpuche-Aranda etal, 1994). Upon internalization, Salmonella reside in a unique membrane-bound compartment distinct from a phagosome or lysosome, termed the Salmonella containing vacuole (SCV) (Figure 1.1C). In infected cells, the SCV undergoes a maturation process distinct from the normal endocytic pathway. Unlike normal endosomes, the SCV does not acquire fluid-phase endosomal markers, or the mannose-6-phosphate receptor. Although it does not fuse with the lysosome, the SCV does acquire some lysosomal glycoproteins (e.g. Lampl) (Gorvel and Meresse, 2001; Meresse et a/., 1999; Steele-Mortimer et al, 1999). In epithelial cells, but not macrophages, the SCV forms long tubular structures continuous with the SCV membrane called Salmonella-induced filaments or Sifs (Garcia-del Portillo et al, 1993a; Garcia-del Portillo et al, 1993b)(Figure 1.1C). Within macrophages, Salmonella SCV formation has the important function of evading endosomal fusion with the phagocyte oxidase complex (Vazquez-Torres et al, 2000). Following a lag of several hours, Salmonella within the SCV begin to replicate (Figure 1.1D). Although the extent and duration of this replication is variable and depends on host and bacterial factors, ultimately it leads to destruction of the cell by cytotoxic effects or cell death with features of both necrosis and apoptosis (Guiney, 2005) (Figure LIE) . The coordinated progression of these cellular events has largely been established in vitro. The interactions of Salmonella with host cells have a variety of roles in intracellular pathogenesis, and in vitro models have served as an important tool in identifying virulence behaviours important for bacterial pathogenesis in vivo. 1.2.2 Murine typhoid Perhaps the best understood disease induced by Salmonella in any host is the fatal systemic disease induced by serovar Typhimurium in genetically susceptible inbred mice. Naturally occurring murine infection by serovar Typhimurium was initially described in 1892 by Loeffler, who noted that the disease course, bacterial dynamics and tissue pathology of this murine infection recapitulated those seen in sufferers of human Typhoid (Loeffler, 1892). This comparison has been the hallmark of the study of Typhoid pathogenesis in the century following Loeffler's observations. Both targeted and serendipitous disruption of bacterial and host 9 processes has enabled the further elucidation of critical mechanisms of bacterial virulence and host defense required for the induction and control of disease. Based on these findings, a model of disease pathogenesis has been elaborated and divided into discrete but overlapping phases and pathogenic processes. Following oral inoculation, virulent serovar Typhimurium survive gastric acidity and colonize the ileum and cecum by outcompeting the resident microflora (Bohnhoff et al, 1954; Stecher et al, 2005). Via invasion of the phagocytic epithelial M-cells covering Peyer's patches, as well as through uptake by dendritic cells (DCs), bacteria are translocated across the intestinal epithelium and gain access to the host circulation or are carried from the gut within CD 18 expressing phagocytes (Jones et al, 1994; Rescigno et al, 2001; Vazquez-Torres et al, 1999). Upon extraintestinal infection, bacteria disseminate via the RES and take up residence in granulomatous foci within various splenocytes, predominantly macrophages, DCs and polymorphonuclear leukocytes (PMNs), as well as hepatocytes and other non-professional phagocytes in the liver (Nakoneczna and Hsu, 1980; Richter-Dahlfors et al, 1997; Yrlid et al, 2001). Bacterial proliferation and dissemination in these organs is accompanied by cytokine induction, RES activation, hepatosplenomegaly, sepsis, bacteremia and ultimately death, which occurs in 3-5 days following systemic inoculation of bacteria (intraperitoneal or intravenous), or 6-10 days following oral infection (Mastroeni and Sheppard, 2004). Although infection of mice is not limited to those with specific genetic susceptibilities, the induction of murine typhoid requires that the host lack a functional Nrampl protein encoded by the bcg/ity/lsh or Nrampl locus on chromosome 1 (Vidal et al, 1995). Thought to encode a divalent cation transporter and alternately named 'solute carrier H a l ' (Slcl l a l ) , the presence of an NramplTeSKtant allele prevents the lethal infection of mice with serovar Typhimurium as well as Leishmania and Mycobacterium species (Govoni et al, 1996; Vidal et al, 1995). The mechanisms of Nrampl -mediated resistance are not yet clear. Infection of resistant mice is non-lethal and although carriage of serovar Typhimurium continues for as long as 1 year following infection, the mice do not display significant pathology at any timepoint following oral or intraperitoneal (i.p.) infection with doses of serovar Typhimurium that would be lethal in other mouse lines (Monack et al, 2004). The study of murine typhoid is therefore confined to those strains of mice (e.g. C57B16 and BALB/c) that have homozygous mutations in Nrampl that render the protein non-functional. Unless otherwise stated, further discussion of murine typhoid wil l refer to infections of susceptible mice that do not produce functional Nrampl. 10 Murine typhoid can be induced following oral, i.p. or intravenous (i.v.) inoculation of mice with serovar Typhimurium. Although oral infection most faithfully recreates naturally occurring infections in humans and mice, systemic administration of the bacteria to the host has become a common experimental technique, as these routes of administration decrease the variance in clinical outcome, asynchronicity of infection, infectious dose and time to death compared to oral infection. The LD50 for intraperitoneal infection of B A L B / c mice with the commonly used serovar Typhimurium laboratory strain SL1344 is 20 colony forming units (CFU) compared to an LD50 of 4><104 C F U for oral inoculation of the same strain (Galan and Curtiss, III, 1989). Disease induced by these routes of infection involves many (but not all) of the same bacterial and host factors. In fact, by bypassing the intestinal phase of infection with i.v. or i.p. inoculation, investigation of those host and bacterial processes required for the systemic phase of illness in the absence of the confounding influence of intestinal virulence factors and host defenses is possible. In the absence of intestinal infection, intracellular replication and survival may be considered the central feature of murine typhoid. Upon translocation to systemic sites, or upon inoculation of the bacteria into the peritoneal cavity, survival of phagocytic killing is an essential component of bacterial virulence. Fields and colleagues, in perhaps the defining paper identifying bacterial virulence behaviours in murine typhoid, demonstrated that bacterial survival within phagocytes was essential for virulence (Fields et al, 1986). Because of its independence from intestinal processes, experimental murine typhoid can be considered a reflection of the host processes involved in the systemic phase of disease in mice, e.g. the adaptation of bacteria to the compartment in which they reside in vivo: the cell. 1.2.3 Bovine and Rabbit Typhimurium infection (enterocolitis) In contrast to the murine infections with serovar Typhimurium, experimental infection of cattle and rabbits results in the induction of significant intestinal disease. While this serovar is not exclusively used to model intestinal disease in bovine models of enteritis, it recapitulates the human disease induced by serovar Typhimurium better than bovine infections with serovar Dublin, which causes significant systemic disease in cows (Clarke and Gyles, 1987; Rings, 1985; Santos et al, 2001b). Induction of intestinal disease in bovine or rabbit infections is achieved primarily by either orally inoculating animals with bacteria, or constructing multiple ligated intestinal loops within a single, animal and inoculating them with experimental and control 11 bacterial strains. The latter technique has the advantages of controlling for inter-animal variation in disease responses thereby decreasing the number of animals required, as well as allowing for the direct assessment of pathogenic process such as fluid secretion in a single loop (Clarke and Gyles, 1987; Giannella et al, 1973). Disadvantages of this approach include the technical difficulty performing surgical procedures, the relatively short duration of experiments due to the obstruction of the bowel by ligatures and the high cost of animals. Differences in host response also must be considered when interpreting data from these models of enterocolitis. While oral bovine infections mimic human enterocolitis in the time to onset of symptoms (8-48h), duration of illness (6-10 days) pathological changes and self-limited nature (with suitably low doses), oral infections of rabbits results in a fatal systemic disease within 7-10 days of infection (Hanes et al, 2001). Bovine experiments are constrained by the considerable cost of animals and difficulty in performing experiments with large experimental groups. Furthermore, since genetic manipulations of cows and rabbits are not practical, assessments of host response during diarrheal pathogenesis is limited to descriptions of host responses upon challenge with various virulent and avirulent bacterial strains. Despite these limitations, important insights regarding the induction of Salmonella enterocolitis and diarrhea have been made using these models, primarily relating to the nature of inflammatory cell recruitment and fluid secretion at the site of infection and the virulence machinery required by S. enterica to elicit these changes. Upon colonization of the intestine by virulent S. enterica, bacteria localize to the apical epithelium, induce invasion-associated virulence machinery and elicit neutrophil recruitment and fluid secretion (Clarke and Gyles, 1987; Finlay et al, 1989; Giannella et al, 1973). While the mild diffuse intestinal pathology associated with serovar Typhi infection in humans and serovar Typhimurium infection in mice is concentrated in the ileum and primarily composed of diffuse infiltrate of mononuclear leukocytes (Shirai et al, 1979; Sprinz et al, 1966), human, bovine and rabbit serovar Typhimurium enterocolitis is most severe in the caudal ileum and proximal colon and is dominated by both focal and diffuse P M N infiltrate, crypt abscesses, epithelial necrosis and edema (Clarke and Gyles, 1987; Giannella et al, 1973; Mcgovern and Slavutin, 1979). Although in vivo evidence is correlative, in vitro P M N recruitment to cultured epithelial monolayers occurs via the induction of interleukin-8 (IL-8) by Salmonella proximate to the apical epithelium (McCormick et al, 1993). Furthermore, the ability of various S. enterica strains to cause human intestinal disease correlated to their ability to attract PMNs across T84 12 cell monolayers, notably without requiring epithelial invasion (McCormick et al, 1995b). Both Wallis and colleagues (Wallis et al, 1990) and Giannella and colleagues (Giannella, 1979) demonstrated that depletion of the neutrophil pool in rabbits by nitrogen mustard administration prevented the inflammatory pathology caused by serovar Typhimurium infection, indicating that P M N recruitment is essential for enterocolitis. While neutrophil recruitment by serovar Typhimurium occurs within the first 1-3 hours of infection, massive neutrophil migration and the secretion of protein-rich exudates into the intestinal lumen does not occur until 8-10 hours following infection and diarrhea begins approximately 12-48 hours after bacterial colonization (Tsolis et al, 1999a; Wray and Sojka, 1978). Both the temporal separation of inflammation and secretory diarrhea and other evidence indicating that Salmonella in different growth stages show differential induction of inflammation versus secretory responses (Wallis et al, 1989) suggest that, although perhaps related, diarrhea and inflammation occur independently in enteropathogenesis. There is some evidence that diarrhea may depend on chloride secretion in a manner similar to that induced by cholera toxin involving the Salmonella stn gene (Chopra et al, 1999) and host-produced prostaglandins (Eckmann et al, 1997), however these processes may be limited to specific hosts. Resolution of intestinal disease relies on a combination of the virulence of the infecting serovar as well as the resistance of the host to systemic illness. Cows infected orally with serovar Typhimurium show symptomatic resolution by 6-10 days, with little residual bacterial colonization, while rabbits infected with serovar Typhimurium and cows infected with the invasive serovar Dublin succumb to systemic infection (Baumler et al, 1998). While the oral inoculation of cows and rabbits with serovar Typhimurium resemble human infection in many ways, because of their practical and theoretical limitations, they have only been scarcely applied to the study of serovar Typhimurium enteropathogenesis. The more experimentally versatile ligated loop models of infection, although limited to the study of the early events in intestinal salmonellosis, have therefore provided the bulk of insights into the intestinal pathogenesis of serovar Typhimurium infection, although oral infection with serovar Typhimurium likely most closely resembles human disease. 1.3 Fundamentals of S. enterica serovar Typhimurium pathogenesis Although clearly successful as tools for investigating Salmonella infection in vivo, the practical and theoretical limitations of the murine typhoid and ligated ileal loop models of 13 pathogenesis have resulted in the partitioning of in vivo pathogenesis into fragments. The murine typhoid model of systemic disease has biased discovery towards those bacterial and host processes critical during the intracellular phase of infection, particularly during residence in phagocytes in the liver and spleen. Conversely, due to the limited duration of ligated loop infections, discoveries made from the exploitation of these models reflect the early events associated with colonization of the gut,, induction of inflammation and invasion of epithelial cells; the hallmarks of early pathogenesis. Due to these limitations, existing animal models have favoured the discovery of discrete pathogenic mechanisms that are both temporally and mechanistically dichotomous. Table 1.2 Genes in the SPI-1 and SPI-2 pathogenicity islands and their classes of function. Function SPI-1 genes SPI-2 genes Regulators HilA.D InvF, SirC, SprB SsrAB SpaOPQRS, InvABCEGIJ PrgHIJ, T3SS Apparatus OrgB SsaBCDEGHI JKLMVNOPQRSTU T3SS Effectors SipABCD, AvrA, SptP SseBFG, SpiC T3SS Translocon/Chaperones SspBCD SicAP, InvB, lacP, SipB SseABCD, SscAB Other/Unassigned Functions lacA.P, OrgA.C InvE.I TlrABCSR, SseE 1.3.1 Salmonella pathogenicity islands Within its arsenal of genes required for bacterial physiology and disease pathogenesis, Salmonella harbours several bacterial regions required for disease pathogenesis. These genetic 'islands' have nucleotide ( 'GC') content that differs significantly from the average GC content of the rest of the Salmonella genome, indicating that they likely evolved in another organism and were acquired horizontally. Several such regions have been identified in the S. enterica genome, including two Salmonella pathogenicity islands (SPIs) of particular interest for their involvement in disease pathogenesis in vivo: SPI-1 and SPI-2. SPI-1 is believed to have been acquired prior to speciation of S. enterica from Salmonella bongori, and SPI-2 acquired sometime after this event but prior to divergence of S. enterica (Groisman and Ochman, 1994; Groisman and Ochman, 1996; Ochman and Groisman, 1996). Both SPI-1 and SPI-2 encode a molecular apparatus called a type III secretion system (T3SS, Figure 1.2) capable of injecting bacterial proteins known as 'effectors' through bacterial and host membranes into host cells (translocation) or the extracellular milieu (secretion)(Gophna et al, 2003). A multi-component molecular machine, T3SSs are powerful virulence tools, as they can directly influence host biochemistry and cell physiology by translocating effectors from within the bacterial cytoplasm 14 to that of the host. Encoded within each SPI are several T3SS structural genes, effectors, regulators and molecular chaperones and 'translocases' required for the efficient translocation or secretion of effectors via the T3SS (Table 1.2). In addition to effectors encoded within the pathogenicity islands themselves, genes located elsewhere on the bacterial chromosome, or within phages, encode effectors that are translocated or secreted by the SPI-1 and SPI-2 type III secretion systems. Several functions of SPI-1 and SPI-2 type III effectors have been identified and their impact on virulence in vivo assessed (Tables 1.3 and 1.4). Figure 1.2 Type III secretion system. A variety of pathogens including Salmonella have acquired a specialized molecular needle capable of introducing bacterial proteins into host cells called a type III secretion system (T3SS). This multi-component structure has modules that span the bacterial inner membrane (IM), periplasm containing peptidoglycan (PGN) and outer membrane (OM), and host cell or vacuolar membrane (HM). Using ATP, this apparatus is capable of delivering bacterial proteins called effectors from within the bacterial cytoplasm to the host cytoplasm (translocation) where they can effect virulence behaviours necessary for bacterial pathogenesis. The T3SS is composed of a large number of bacterial proteins necessary to perform this function, and has similarities to the flagellar apparatus from which it likely evolved. ATP ADP 1.3.2 SPI-1: the invasion associated locus In 1989, Galan and Curtiss (Galan and Curtiss, III, 1989) complemented the invasion defect of noninvasive serovar Typhimurium strains with cloned genes from strains capable of invasion. Upon identifying four invasion associated genes; invA, invB, invC and invD; they demonstrated that inv ~ strains were attenuated for oral, but not i.p. infections of BALB/c mice. While only 20 C F U of wild-type or inv " bacteria were required to produce lethal infection in 50% of infected mice, the LD50 of invasion mutants was one to two logs greater than wild-type bacteria following oral infection. In 1995, Mills et al. postulated that 40 contiguous kilobases from the S. enterica chromosome containing the invasion associated genes were a horizontally acquired pathogenicity island, as it is absent from the closely related non-invasive pathogen E. coli K-12 (Mills et al, 1995). Genes encoded by SPI-1 include regulators of SPI-1 genes, components of the T3SS and effectors that influence a variety of host cell machinery including the actin cytoskeleton, tight junctions and caspases (Table 1.2, Table 1.3). The result of SPI-1 activity includes the induced uptake of Salmonella into non-phagocytic cells, caspase-1 mediated 15 cell death and tight junction disruption (Table 1.3, for reviews of SPI-1 function see (Altier, 2005; Finlay and Brumell, 2000; Hansen-Wester and Hensel, 2001; Lostroh and Lee, 2001)). In vivo, although not exclusively so, SPI-1 is also critical for the induction of intestinal disease in both cows and rabbit ileal loops (Santos et al, 2003; Zhang et al, 2003). Table 1.3 Salmonella effectors secreted by the SPI-1 type III secretion system. Effector Location of Gene SipA SPI-1 SipB SPI-1 SipC SPI-1 SopA SopB/ SPl-5 SigD SopC Chromosome SopD Chromosome SopE Phage SopE2 Chromosome SptP SPI-1 AvrA SlrP SspH1 Function in vitro Actin binding and rearrangement; Early bacterial internalization; neutrophil recruitment to epithelium Pore formation and effector translocation; Caspase-1 activation; induction of apoptosis Nucleation and bundling of actin; Necessary for effector translocation Invasion (with sopB,D,E2,sipA) Invasion (with sopA,D,E2,sipA) Invasion (with sopB,D,E2,sipA) Invasion Invasion (with sopA,D,E2,sipA) Tyrosine phosphatase; Disrupts actin cytoskeleton Mutant phenotype in vivo Mildly reduced enteropathogenicity in calves; Complete attenuation (early) in the absence of sopA,B,D,E2 Attenuated enteropathogenicity in calves; po. attenuation in murine typhoid (mild) virulent; Total intestinal attenuation (early) in the absence of sipA,sopB,D,E2 significantly reduced enteritis; Total attenuation (early) in the absence of sipA,sopA,D,E2 Significantly reduced enteritis; Total attenuation (early) in the absence of sipA,sopA,B,E2 virulent virulent; Total intestinal attenuation (early) in the absence of sipA,sopA,B,D virulent Reference (Kahiga era/., 1995a; Lee eta/., 2000; Raffatellu et al., 2005; Zhang etal., 2002b; Zhou etal., 1999) (Hayward et al., 2000; .Kaniga etal., 1995b) (Hersh etal., 1999; Tsolis etal., 1999a) (Hayward and Koronakis, 1999) (Raffatellu et al., 2005; Wood etal., 1998; Wood et al., 2000; Zhang et al., 2002b; Zhou era/., 2001) (Raffatellu et al., 2005; Wood era/., 1998; Wood et al., 2000; Zhang et al., 2002b; Zhou et al., 2001) (Raffatellu et al., 2005; Wood etal., 1998; Wood et al., 2000; Zhang et al., 2002b; Zhou et al., 2001) (Raffatellu et al., 2005; Wood etal., 1998; Wood et al., 2000; Zhang etal., 2002b; Zhou et al., 2001) (Fu and Galan, 1998) SPI-1 Chromosome Gifsy Prophage Inhibits N F K B activation Attenuated for per os infection of mice Enteropathogenic like WT, but attenuated for lethality in cows in the absence of SspH2 (Collier-Hyams et al., 2002) (Tsolis etal., 1999c) (Miao etal., 1999) 16 The contribution of SPI-1 to intestinal Salmonellosis has been tested in many model systems. This stems from two important observations already described; that SPI-1 mutants have a higher oral LD50 than wild-type serovar Typhimurium in murine typhoid, and that SPI-1 is necessary for invasion of epithelial cell lines in vitro (Galan and Curtiss, III, 1989). Subsequent investigations further established the nature of the SPI-1 associated invasion process and indicated that SPI-1 was required for the bacterially-induced internalization event which occurs following the contact between bacteria and host cells (Collazo and Galan, 1997). By comparing the invasiveness of serovar Typhimurium strains bearing mutations in SPI-1 and other genes with their enteropathogenicity in rabbit intestine ex vivo and bovine or rabbit ileal loops, several groups correlated the ability to invade intestinal epithelial cells in vitro with the ability to induce intestinal disease in vivo (Lodge et al, 1995; Watson et al., 1995Galyov et al., 1997; Jones et al, 1998; Wallis et al, 1999a; Watson et al, 1998; Wood et al, 1998; Wood et al, 2000). Further investigations in bovine and tissue culture infections identified specific SPI-1 T3SS effectors required for this process (Raffatellu et al, 2005; Zhang et al, 2002a). The accumulation of data in bovine and rabbit ileal loop models of intestinal disease, and the repeated observations that serovar Typhimurium strains with mutant versions of various SPI-1 T3SS components and effectors were virulent in models of systemic disease, led to the emergence of a dominant theoretical paradigm in which SPI-1 virulence machinery is considered essential for intestinal disease but not systemic disease (Fierer and Guiney, 2001; Finlay and Brumell, 2000; Hansen-Wester and Hensel, 2001; Lostroh and Lee, 2001; Santos et al, 2001b; Santos et al, 2003; Tsolis et al, 1999a; Tsolis et al, 1999b; Zhang et al, 2003). 1.3.3 SPI-2: the intracellular survival locus In addition to the SPI-1 virulence island, Salmonella contain several other horizontally acquired pathogenicity islands. More recently identified by the Holden and Groisman laboratories (Hensel et al, 1995; Ochman et al, 1996; Shea et al, 1996), the SPI-2 pathogenicity island has been considered the counterpart to the invasion associated SPI-1. By identifying a series of serovar Typhimurium mutants from a transposon library that were attenuated in murine typhoid, Hensel et al, (Hensel et al, 1995), as well as Ochman et al. (Ochman et al, 1996), identified virulence genes required for both murine typhoid and intracellular survival in vitro. Although the relationship between intracellular survival and systemic virulence of S. enterica had been described a decade earlier by Fields and colleagues (Fields et al, 1986), this was the 17 first identification of a specific virulence system unique to Salmonella required for both behaviours. Unlike Salmonella with mutations of SPI-1, mutants lacking critical components for SPI-2 required a median dose 10,000 times that of wild-type bacteria to induce the same lethality in susceptible mice by any route of infection. Clearly critical for survival within wild-type mice, further investigations revealed that SPI-2 encodes an additional T3SS necessary for maintaining the SCV (SPI-2 function is reviewed in (Fierer and Guiney, 2001; Finlay and Brumell, 2000; Waterman and Holden, 2003)). Although complex interactions with the endocytic machinery of host cells are involved, of central importance to intracellular survival of serovar Typhimurium is the evasion of the phagocyte oxidase machinery. SPI-2 proficient Salmonella prevent this important bactericidal process from accessing the SCV, and virulence of SPI-2 mutant Salmonella is restored in mice lacking a functional phagocyte oxidase complex (Vazquez-Torres et al, 2000). The net consequence of SPI-2 dysfunction is clear in intracellular survival and systemic disease in murine typhoid. Evidence for a role of SPI-2 in intestinal disease is conflicting. Salmonella SPI-2 mutants are equally attenuated in murine typhoid regardless of the route of infection (oral or systemic). Comparisons of SPI-2 mutant strains in ligated loop models of enteropathogenesis have been made with various outcomes. Everest (Everest et al, 1999) and colleagues and Tsolis and colleagues (Tsolis et al, 1999a) demonstrated that SPI-2 T3SS apparatus mutants were as virulent as wild-type serovar Typhimurium in rabbit ileal loop infections and bovine oral infections with serovar Typhimurium, while Bispham et al. (Bispham et al, 2001) demonstrated that serovar Dublin SPI-2 mutants incapable of SPI-2 T3SS-mediated secretion and translocation caused only mildly reduced intestinal secretion and inflammatory pathology in bovine ileal loop infections. The difference in these results may be related to the host adaptation of serovar Dublin for cows, in which serovar Dublin, but not Typhimurium, is invasive. Because of weak and conflicting evidence regarding a role for SPI-2 in enteric disease and significant evidence that SPI-2 is required absolutely only for systemic infection of mice, it has become widely accepted (by investigators including the authors of the latter bovine study) that SPI-2 plays little or no role in intestinal disease (Zhang et al, 2003). 1.3.4 Determinants ofS. enterica intestinal virulence There is abundant evidence that S. enterica strains lacking SPI-1 effectors are attenuated for intestinal pathogenesis in both ligated ileal loops and oral infection of calves. Of particular 18 importance for intestinal pathogenesis in these models are the serovar Dublin and Typhimurium SPI-1 secreted effectors SipA, SopA, SopB, SopD, SopE, and SopE2. The SPI-1 effector and translocase SipB is also critical for inflammatory disease in vivo (Zhang et al, 2002b), and in vitro is required for the induction of specific inflammatory cascades (Hersh et al, 1999), however, it is also required for translocating other SPI-1 effectors, and the relative contribution of its translocation and effector functions to in vivo pathogenesis have not been evaluated. In contrast, the study of other SPI-1 secreted effectors has yielded specific insight into their cooperative function in vitro and in vivo and their role in eliciting enterocolitis. Invasion of cultured epithelial cells by serovar Typhimurium is drastically reduced in the absence of the SPI-1 effectors SipA, SopA, SopB, SopD and SopE2 (Raffatellu et al, 2005). While complementation with SopE2 alone provides the most substantial rescue of this defect, it is not sufficient to restore wild-type levels of invasion. In vivo, the absence of these effectors results in reduced inflammatory severity in bovine ligated ileal loops at 8 hours to levels equivalent to strains lacking any capacity for SPI-1 effector translocation, indicating that these effectors are critical for induction of intestinal disease early in infection (Zhang et al, 2002b). Furthermore, these strains cause less diarrhea and mortality in orally infected cows between 3-8 days post infection (Zhang et al, 2002b). However, the precise mechanism of this induction is not clear, and although significant overlap exists between effectors required for cell invasion in vitro and intestinal pathogenesis in vivo, no causal connection has been demonstrated. In addition to effectors secreted by SPI-1, a number of other virulence factors have been tested for a role in the induction of intestinal disease in various animal models (Table 1.5). The Salmonella flagellar apparatus and chemotaxis are required for inflammatory disease in mice (Stecher et al, 2004). Aromatic amino acid synthesis is required for enteropathogenicity in bovine oral infections and rabbit ligated ileal loops (Everest et al, 1999; Tsolis et al, 1999a). Salmonella enterotoxin is required for fluid secretion into murine ligated intestinal loops (Chopra et al, 1999), and an additional pathogenicity island - SPI-5 - is required for the enteropathogenicity of serovar Dublin in bovine ileal loops (Wood et al, 1998). In contrast, several bacterial factors involved in other model infections such as murine typhoid, have been shown to have limited or no role in intestinal disease at various timepoints (Table 1.5). The predominant role of SPI-1 effectors has therefore emerged as the dominant virulence mechanism controlling enterocolitic infection in animal models of intestinal salmonellosis. 19 Table 1.4 Salmonella effectors secreted by the SPI-2 type III secretion system. Effector Location •_ Function in vitro Mutant Phenotype in vivo References SifA Chromosome Maintenance of the SCV; Attenuated in murine (Beuzon et al, 2000; Brumell et al, Sif formation; Intracellular typhoid 2001; Stein et al, 1996) replication Siffi Chromosome Targeted to SCV/Sifs - (Freeman et al, 2003) SseE SPI-2 - Virulent (Hensel a/., 1998) SseF SPI-2 Aggregation of endosomal Slightly attenuated in (Hensel et al, 1998; Kuhle and compartments; Intracellular murine typhoid Hensel, 2002) Replication in macrophages SseG SPI-2 Aggregation of endosomal Slightly attenuated in (Hensel et al, 1998; Kuhle and compartments; Intracellular murine typhoid Hensel, 2002) Replication in macrophages SseJ Phage Maintenance of the SCV; Decreased replication in (Ohlson et al, 2005; Ruiz-Albert et Deacylation of host protein? mouse tissues; Partial al, 2003) attenuation Ssel/SrfH Gifsy Actin Remodelling? - (Miao et al, 2003) prophage SseKl/2/3 Chromosome Virulent (Kujat Choy et al, 2004) SopD2 Chromosome Sif formation Decreased replication in (Brumell et al, 2003; Jiang et al, mouse tissues when 2004) competed against WT SpiC SPI-2 Prevention of Attenuated in murine (Uchiya et al, 1999; Yu et al, phagolysosomal fusion; typhoid 2002) Effector translocation, PipB SPI-5 Maintenance of the SCV or - (Knodlere? al, 2002) Sif formation? PipB2 Chromosome Late endosomal/lysosomal - (Knodler and Steele-Mortimer, reorganization 2005) SspHl Gifsy Enteropathogenic like (Miao et al, 1999) Prophage WT, but attenuated for lethality in cows in the absence of SspH2 SspH2 Phage - Enteropathogenic like (Miao et al, 1999) WT, but attenuated for lethality in cows in the absence o/SspH2 SlrP Chromosome Attenuated for per os (Tsolis etal, 1999c) infection of mice 1.3.5 The SPI-1/SPI-2 dogma Since the critical observations that SPI-1 mutants of serovar Typhimurium are attenuated following oral, but not systemic infection in vivo and that SPI-2 is essential for both intracellular survival and systemic pathogenesis (Table 1.6), a dichotomous view of the contributions of SPI-1 and SPI-2 to Salmonella pathogenesis has arisen (Figure 1.3). Based on in vitro and in vivo 20 evidence, including evidence that SPI-1 repression is mediated by the same regulatory system that induces SPI-2 expression (Behlau and Miller, 1993; Bijlsma and Groisman, 2005) it has become the prevailing dogma of Salmonella pathogenesis that SPI-1 and SPI-2 have temporospatially distinct roles in pathogenesis (Baumler et al, 1998; Fierer and Guiney, 2001; Finlay and Brumell, 2000; Hansen-Wester and Hensel, 2001; Kingsley and Baumler, 2000; Lostroh and Lee, 2001; Mastroeni and Sheppard, 2004; Patel et al, 2005; Santos et al, 2001b; Waterman and Holden, 2003; Zhang et al, 2003). Table 1.5 L a n d m a r k papers of Salmonella in vivo pathogenesis. Virulence Factor/ Behaviour Model System Finding Investigators Intracellular Survival Murine Typhoid Intracellular survival is required for systemic disease Fields et al (1986) SPI-1 Murine Typhoid invasion (SPI-1) mutants are attenuated for oral but not intraperitoneal infection Galan and Curtiss (1989) SPI-1 SPI-2 Bovine and Rabbit enterocolitis Murine Typhoid SPI-1 is required for intestinal Various (see text for details) virulence SPI-2 is required for intracellular Hensel et al (1995), Shea et al survival and systemic virulence (1996), Ochman et al (1996) SPI-2 Murine Typhoid SPI-2 is required for evasion of phagocyte oxidase (Vazquez-Torres et al., 2000) 1.3.6 Hints from the literature: Data discordant with the prevailing model ofpathogenesis In addition to the perhaps murky and conflicting evidence that SPI-2 plays a minor role in early S. enterica enteropathogenesis, in vitro and in vivo studies indicate that the relationship between SPI-1, SPI-2, invasion, inflammation and intracellular survival is more overlapping than discrete. Three important lines of evidence indicate that invasion and enteropathogenicity are not causally related. Firstly, in vitro invasiveness is not predictive of enteropathogenicity. Despite the correlation between invasiveness of SPI-1 mutants in cultured cells and enteropathogenicity in cows (Raffatellu et al, 2005; Zhang et al, 2002b), McCormick et al. demonstrated that invasion of polarized epithelial cell cultures by S. enterica serovars Pullorum, Arizonae, Typhi and Paratyphi was not predictive of P M N recruitment across epithelial monolayers or enteropathogenecity in humans (McCormick et al, 1995b). Secondly, invasion-independent inflammation can be induced by a Salmonella secreted protein. Further evidence 21 from McCormick while in the Madara laboratory and subsequently from her own research group demonstrated that recruitment of neutrophils to and across cultured epithelial monolayers required production of IL-8 and pathogen elicited epithelial chemoattractant (PEEC) and translocation of the SPI-1 effector SipA, but not bacterial internalization (Gewirtz et al, 1999; Lee et al, 2000; McCormick et al, 1998). Thirdly, inflammatory responses can be induced in the absence of SPI-1. Interestingly, although incapable of inducing PEEC mediated transmigration of neutrophils across model epithelia, comparison of the inflammatory gene expression profiles of cultured intestinal epithelial monolayers infected with wild-type serovar Typhimurium or those with a complete deletion of the SPI-1 pathogenicity island demonstrated little difference in the proinflammatory potential of these strains (Zeng et al, 2003). Thus, at least in vitro inflammation can occur in the absence of either bacterial invasion or SPI-1. Table 1.6 Non-SPI-1 secreted virulence factors tested in animal models of intestinal salmonellosis Virulence Phenotype in Factor Description vivo Disease Model Reference Attenuated FliGHI Flagellar apparatus genes attenuated Murine enterocolitis Stecher 2004 CheY Chemotaxin attenuated Murine enterocolitis Stecher 2004 AroA aromatic amino acid attenuated Bovine oral infection, Tsolis 1999, biosynthesis Rabbit ligated ileal loops Everest 1999 Stn enterotoxin attenuated Murine ligated ileal loops Chopra 1999 PipA* SPI-5 encoded Gene attenuated Bovine ligated ileal loops Wood 1998 PipB* SPI-5 encoded Gene attenuated Bovine ligated ileal loops Wood 1998 PipD* SPI-5 encoded Gene attenuated Bovine ligated ileal loops Wood 1998 Intermediate virulence RfaJ LPS outer core biosynthesis Intermediate Bovine Oral infection Tsolis 1999 SsaT* SPI-2 T3SS apparatus protein Mild Bovine ligated ileal loops Bispham 2001 attenuation SseD* SPI-2 T3SS translocon Mild Bovine ligated ileal loops Bispham 2001 component attenuation Virulent MsbB lipid A biosynthesis Virulent Rabbit ligated ileal loop Everest 1999 SsrA SPI-2 Two component regulator Virulent Rabbit ligated ileal loop Everest 1999 SpiB SPI-2 Virulent Bovine Oral infection Tsolis 1999 SpvR virulence plasmid Virulent Bovine Oral infection Tsolis 1999 * experiments used S.enterica serovar Dublin 22 Figure 1.3 T h e prevailing view of pathogenicity island involvement in Salmonella enterica infection in vivo. Rather than overlapping roles, Salmonella pathogenicity islands (SPIs) 1 and 2 are thought to play temporospatially discrete roles in pathogenesis. During intestinal colonization and invasion, critical SPI-1 mediated behaviours such as attachment to intestinal cells, barrier disruption and cell invasion are thought to be involved in disease. Behaviours such as intracellular survival within the reticuloendothelial system (RES) and target organs are considered to be independent of SPI-1 and rely solely on SPI-2. In this model, there is no role for SPI-2 in intestinal disease and no possible intestinal disease in the absence of SPI-1. I have provided little evidence that the role of SPI-2 is confined to anything but intracellular survival and replication of Salmonella and systemic virulence. The weak and conflicting evidence from bovine and rabbit models is generally not considered an indication that SPI-2 is involved in Salmonella enteropathogenesis by most reviewers of this field. There is evidence, however, that SPI-2 is important for more than the maintenance of the SCV. Both the regulation of SPI-1 expression and resistance to antimicrobial substances has been shown to partially depend on SPI-2 (Deiwick et al, 1998). Also, the proinflammatory potential of 23 flagellin at least partly requires its SPI-2 dependent translocation to the basolateral membrane of the intestinal epithelium (Lyons et al, 2004). In a series of papers, Uchiya and colleagues also demonstrate that SPI-2 is involved in the induction of cyclooxygenase as well as the modulation of host cytokine expression and signaling (Uchiya et al, 2004; Uchiya and Nikai, 2004; Uchiya and Nikai, 2005). Due to the conflicting nature of the literature on the roles of SPI-1 and SPI-2 in intestinal disease, it is clear that further investigation of the roles of these important virulence determinants in infection in vivo is warranted. Because of the previously described practical and theoretical limitations of the in vivo models available, an alternative model of pathogenesis is necessary to fully explore these questions. 1.4 A new model for assessing the role of SPI-1 and SPI-2 in intestinal pathogenesis In August, 2003, the Hardt laboratory published a report detailing a murine enterocolitis model of serovar Typhimurium infection that had not previously been described in detail (Barthel et al, 2003). This model relies on the pretreatment of mice with streptomycin which decreases the colonization resistance of the intestine to Salmonella. Disease in this model can be compartmentalized into three categories; bacterial colonization of organs, inflammatory pathology and mortality. Importantly, this model can be used to examine the course of illness in a large number of genetically identical mice over a range of timepoints, using a variety of bacterial mutants and hosts with various genetic susceptibilities. This in vivo model is a new and valuable tool to study serovar Typhimurium enteropathogenesis. 1.5 Research objectives Intestinal salmonellosis is a worldwide problem. A clear understanding of the bacterial and host factors influencing disease pathogenesis is vital to the development of efficacious therapies and strategies for the approach to this disease. Multiple animal models have been exploited to gain understanding of the contributions of specific virulence mechanisms to S. enterica in vivo pathogenesis. These models have led to a specific theoretical paradigm of the involvement of two such virulence mechanisms, SPIs 1 and 2, to intestinal and systemic disease, suggesting that they are involved in separate phases of pathogenesis. However, theoretical and practical restraints encumber the interpretation of these experiments. Furthermore, in vitro evidence indicates possible novel roles for SPI-2 in intestinal disease, and a potential for intestinal inflammatory disease in the absence of SPI-1. 24 The objectives of this thesis are: 1) To establish and characterize the murine enterocolitis model of Salmonella infection and to develop criteria to assess and compare murine intestinal inflammatory pathology of ranging severity induced by Salmonella; 2) To exploit this new tool to examine the contributions of SPIs 1 and 2 to intestinal inflammatory disease over an extended course of infection. Specifically, I aim to compare the severity of intestinal inflammation induced by SPI-1 and SPI-2 mutant Salmonella enterica serovar Typhimurium in early intestinal inflammatory pathogenesis to test validity of the dichotomous model of SPI-l/SPI-2 in vivo pathogenesis in a newly described experimental system. Furthermore I intend to explore possible SPI-1-independent inflammation in murine enterocolitis. Finally, I aim to validate the findings from these experiments to other intestinal salmonellosis models and, i f possible, human infections. I expect that the use of this model will indicate novel overlapping roles for SPIs 1 and 2 in intestinal disease in vivo and indicate opportunities for further exploration of bacterial and host mechanisms of disease pathogenesis. 25 Chapter 2 - Murine enterocolitis caused by Salmonella infection Preface: The contribution of the authors to this chapter are as follows: I designed and performed all experiments described in this chapter except: David Owen externally verified the histopathology scoring method and results, but did not perform the initial assessments or develop the scoring method. Yuling L i provided technical assistance for mouse infections and tissue harvesting. Claudia Lupp performed Sybr Green staining for normal host flora (Figure 2.1). 26 2.1 Introduction mflarnrnatory diarrheal disease is the most common manifestation of Salmonella enterica infection in humans, and is a global problem (World Health Organization, 2005). The understanding of this infection is still incomplete. Intestinal salmonellosis has been modeled in a number of experimental systems in vitro, and numerous in vivo infections including bovine oral infections, and infections of ligated intestinal loops in cows and rabbits (Clarke and Gyles, 1987; Giannella et al, 1973; Hanes et al, 2001; Rings, 1985; Santos et al, 2001b). Information on human infection has come from necropsy and biopsy studies and rare infections of healthy volunteers (Boyd, 1985; Mcgovern and Slavutin, 1979), as well as a recently described human biopsy explant system (Haque et al, 2004). While these systems have been exploited to gain significant insights into intestinal pathogenesis of human diarrheal infection, they have significant limitations. Bovine infections are expensive and cumbersome, and require significant expertise and infrastructure unavailable to most investigators. Ligated intestinal loops are limited to short durations of infection as the ligatures typically obstruct the bowel. Human tissue samples are rare and inconsistently available, and in vitro organ culture is available only when biopsy is indicated, usually in the presence of preexisting intestinal pathology. Furthermore, none of these systems provide the opportunity to manipulate the host genome in order to assess the contributions of specific host factors to intestinal infection. Simple murine infections with human enteropathogenic S. enterica species do not result in significant intestinal disease, and can cause fatal systemic disease in the absence of intestinal pathology. In 2003, Barthel et al published a report in which they described the induction of intestinal inflammation by S. enterica in mice. This new model of Salmonella enterocolitis requires the treatment of mice with streptomycin one day prior to infection. The decrease in colonization resistance to Salmonella following streptomycin administration has been extensively described (Bohnhoff et al, 1954; Bohnhoff et al, 1964; Bohnhoff and Miller, 1962; Que and Hentges, 1985), and has been used to experimentally increase colonization by S. enterica in mice (McCormick et al, 1988; Murray and Lee, 2000). Streptomycin is thought to increase colonization by Salmonella due to perturbations of the normal intestinal flora and physiology (Hentges et al, 1985; Que et al, 1986). 27 Human intestinal inflammation in S. enterica infection has specific features. Neutrophil infiltration into the intestinal mucosa and lumen, epithelial architectural disruption, the formation of crypt abscesses, crypt foreshortening and mucosal and submucosal edema are all common features of intestinal pathology in human infection (Boyd, 1969; Boyd, 1985; Mcgovern and Slavutin, 1979). Although ulceration of the mucosa is common, intestinal perforation is rare and infection is limited to the gut in the absence of other disease pathology or an immunocompromised state. Murine infection following streptomycin treatment was reported to share many of these intestinal pathological features, however, infection of susceptible mice (BALB/c, C57B16 and other experimental strains) results in systemic disease with features of murine typhoid in addition to intestinal disease (Barthel et al, 2003). In murine typhoid, the intestinal phase of disease is dominated by bacterial interactions with the specialized lymphoid organs of the ileum, specifically the interaction of bacteria with M cells and macrophages associated with Peyer's patches (Penheiter et al, 1997). In contrast, intestinal pathology in streptomycin treated mice occurs regardless of. the presence of gut associated lymphoid tissue, as mice lacking these specialized organs are as affected by serovar Typhimurium as wild-type mice (Barthel et al, 2003). The goals of this study were to characterize the infection model described by Barthel et al, (2003) in our hands, and develop methods to compare the histopathology induced by a wide variety of experimental S. enterica serovar Typhimurium strains in infected mice and to verify the validity of these methods. 28 2.2 Results 2.2.1 Oral Streptomycin treatment causes transient diarrhea, kills normal flora and decreases host resistance to Salmonella enterica serovar Typhimurium infection In their initial characterization of murine enterocolitis caused by serovar Typhimurium, Barthel et al. (Barthel et al, 2003) postulated that oral treatment with streptomycin decreased intestinal flora and consequently decreased colonization resistance to serovar Typhimurium. We sought to test this hypothesis by directly assessing the number of resident microbes in the intestine prior to and after oral administration of 20mg streptomycin. One day following oral administration of antibiotic or vessel (distilled water) to 10 week old male C57B16 mice, the contents of the cecum were removed, homogenized and stained to visualize bacterial nucleic acid (Figure 2.1). The constituent flora in antibiotic treated mice was drastically reduced in both number and variety. Notably, the intestines of treated but not control mice were grossly distended with fluid feces one day, but not two days, following antibiotic administration. This observation is consistent with the hypothesis that intraluminal streptomycin exerts both direct antimicrobial activity and induces secretory diarrhea due to increased osmolality of intestinal contents. Streptomycin administration and elimination of intestinal microflora caused significantly decreased host resistance to oral inoculation with serovar Typhimurium. The oral LD50 o f streptomycin treated mice decreased from 4><104 C F U to <150 C F U (all reproducible doses induced lethal infection in susceptible mice following oral administration of streptomycin). Furthermore, mean time to death decreased significantly for similar doses of bacteria from ~8 days to ~5 days for inocula of 10 6 C F U wild-type serovar Typhimurium. Figure 2.1 The effect of streptomycin treatment on the normal flora. Intestinal homogenate from mice 24 hours after oral gavage with vessel (Left) or 20mg streptomycin (Right) stained with anti-nucleic acid stain (Sybr Green) to visualize bacteria. The absence of green staining indicates the loss of intestinal flora following streptomycin treatment. L n t r e a t e d Post 2(>mg Streptomycin po 29 2.2.2 Inflammatory features of murine serovar Typhimurium enterocolitis include changes in the lumen, surface epithelium, mucosa and submucosa of infected ceca Infection of mice with serovar Typhimurium following administration of streptomycin results in significant inflammatory pathology of the cecum, with less pathology in the colon and little or no pathology in the ileum (Chapter 3, (Barthel et al, 2003; Coburn et al, 2005)). Inflammatory features of murine typhlitis (inflammation of the cecum) can be approximately divided into the following categories: gross pathology; histopathology of the intestinal lumen, surface epithelium, mucosa and submucosa, and; alterations of tissue physiology or function. In order to assess and compare the severity of inflammatory pathology in murine serovar Typhimurium enterocolitis, it is necessary to characterize these changes. Histopathological changes were assessed using fixed intestinal tissue cross-sections stained with a morphological stain (hematoxylin and eosin). The intestinal inflammatory changes in murine ceca were evaluated at two days following oral infection of 10 week old male C57B16 mice with wild-type serovar Typhimurium strain SL1344. Mice were infected with 108 C F U in L B broth or vessel alone one day following oral gavage with 20mg streptomycin sulfate p.o as described in Materials and Methods. Lumen: In vitro, infection of polarized epithelial cells with virulent serovar Typhimurium results in both epithelial cytopathy and recruitment and transcytosis of neutrophils to and across polarized epithelial monolayers (Finlay and Falkow, 1990; McCormick et al, 1993; McCormick et al, 1995a; McCormick et al, 1995b). Furthermore, infiltration of neutrophils into the intestine is a hallmark of human intestinal salmonellosis (Boyd, 1985; Mcgovern and Slavutin, 1979). Epithelial necrosis and neutrophil transmigration into the lumen of infected ceca was substantial in murine serovar Typhimurium enterocolitis and resulted in significant purulent exudate that was both grossly and histologically apparent (Figure 2.2 B,C). By several days of infection, the intestinal lumen was completely occupied by necrotic epithelial cells and infiltrating neutrophils (Figure 2.2C, inset). Although the presence of occasional necrotic epithelial cells within the lumen is common in uninfected intestines and likely represents the normal physiological cell turnover of this tissue, a high density of epithelial cells and the presence of neutrophils are features that accompanied only severe inflammation. The presence of other cell types in the lumen was infrequent, although lumenal lymphocytes were common adjacent to gut associated lymphoid tissue (GALT) such as Peyer's and cecal patches. This is not normally a feature of inflammation in murine inflammatory intestinal salmonellosis. 30 Figure 2.2 The histopathological features of murine Salmonella enterica induced cecal inflammation. Mice were treated with 20mg streptomycin orally 24hours prior to inoculation with 108 Salmonella enterica serovar Typhimurium SL1344 (B-F) or vessel (A). Two days after infection ceca were isolated, and fixed and stained with hematoxylin and eosin. The histological features of inflammation induced by infection are apparent at low power (50 x magnification, B) when compared to uninfected tissue (A) and can be compartmentalized by region. The lumen (L) is infiltrated by significant inflammatory cells and necrotic epithelial debris (li=lumenal infiltrate, panel C) and is completely obscured by purulent necrotic and inflammatory exudates by day 5 post-infection (C, inset). The surface epithelium (SE) shows abnormal epithelial morphology (rc=regenerative change), nuclei within the epithelium (en), desquamation (ds) and ulceration by 5 days (panel D, inset). The archecture of the mucosa is disrupted and loss of goblet cells is evident, as is the presence of crypt abscesses (ca, panel E). The submucosa (F) is marked by edema (e) and the aggregation of extravascular neutrophils (na, panel F, inset). Magnification for A= lOOx, C,F = 50x, D,E=400x. 3 1 Surface Epithelium: Mucosal changes in murine serovar Typhimurium typhlitis can be divided into those changes affecting the epithelium of the intestinal crypt apices (surface epithelium) and those affecting crypt bases and mucosal tissue (mucosa). As intestinal pathology increased in severity, surface epithelial morphology was substantially altered (Figure 2.2D). Epithelial nuclei became variable in size and position within the cell, cell size and shape became irregular and the presence of proliferative cells in the crypt apices was more common (Figure 2.2D, 2.3A, collectively these represent regenerative changes in the epithelium). Epithelial cells were commonly shed into the intestinal lumen (desquamation). Furthermore, epithelial cells responded to infectious insult by expressing the NFKB-regulated intercellular adhesion molecule-1 (ICAM-1, Figure 2.3B), an important marker of the epithelial inflammatory response involved in neutrophil recruitment to epithelia (Dippold et al, 1993; Parkos et al, 1996; Patarroyo and Makgoba, 1989). Neutrophils were seen within the surface epithelial layer (Figure 2.3C). In mild or focal inflammation, these features were confined to discrete regions, but as inflammation became more severe, these changes were evident in the entire circumference of the cecal cross-section (Figure 2.2B). Ulceration of the epithelium exposing the lamina propria, although an uncommon event early in intestinal inflammation, was frequently observed as infection progressed (Figure 2.2D, inset). uninfected 5tfec\ed*«„ L \ B B B B S S S S ^ H Q I I S S ^ B B uninfected infected" Figure 2.3 Pathological changes in murine Salmonella enterica induced intestinal inflammation. (A) Mucosal epithelial regeneration is visualized by staining for proliferating cell nuclear antigen (PCNA, brown staining cells). (B) Epithelial and endothelial (inset) intercellular adhesion molecule-1 (ICAM-1) expression (green) is induced in the cecal mucosa of Salmonella infected mice during inflammation. C - Neutrophil recruitment to the mucosa (red stained cells) is evident following infection. Panels show ceca from uninfected mice or mice infected orally with 108 Salmonella enterica serovar Typhimurium for two days following the administration of 20mg streptomycin. Tissue was fixed in formalin and paraffin embedded (A,C) or fixed in paraformaldehyde and cryopreserved (B) prior to sectioning and staining as described in Materials and Methods. Magnification = 400 x in all panels. 32 Mucosa: Inflammatory changes of the cecal mucosa largely reflected two processes: recruitment of neutrophils to crypt lumens and bases and the destruction of mucosal architecture (Figure 2.2E). The presence of neutrophils within or surrounding crypt bases (crypt abscesses) increased in frequency as inflammation worsened (Figure 2.3C). The discharge of mucous and inflammatory infiltrate into the crypt lumen resulted in mucinous plugs which lead to dilation of the crypt lumen. The formation of granulation tissue (vascular connective tissue investing the crypt walls) also marked inflammatory pathogenesis. Perturbations of normal mucosal physiology and cell turnover resulted in a loss of mucous producing goblet cells and increased mucosal thickness (Figure 2.3A). Submucosa: The most dramatic tissue pathology of the submucosa was the fluid accumulation in the space between the external and mucosal muscle layers (submucosal edema). Normally adjacent, these two layers were often separated by a fluid-filled space of a depth as great as or greater than the intestinal mucosa; itself already significantly thickened (Figure 2.2F). This edema likely reflects the increased vascular permeability of submucosal vessels which accompanies endothelial ICAM-1 expression and the extravasation of neutrophils (Figure 2.3B,C). Neutrophil aggregation was particularly common on the submucosal surface of the muscularis propria (Figure 2.2F, inset, Figure 2.3C). Aggregation of monocytic inflammatory cells (including lymphocytes and monocyte/macrophages) was normal in the uninflamed intestine, but lymphoid aggregates increased in both number and size as pathology worsened. The inflammatory pathology did not extend to the external muscular layer of the intestine and cannot therefore be considered transmural. There was no evidence of intestinal perforation in any mice. The inflammatory features described above clearly indicate a severe inflammatory pathology that is induced in the murine model of Salmonella colitis which shares common features with human intestinal salmonellosis. Using these observations, I have developed a novel semi-quantitative method of scoring intestinal inflammation in murine serovar Typhimurium induced enterocolitis (Table 2.1). This method has several advantages. Tissue pathology of various intestinal regions (lumen, surface epithelium, mucosa and submucosa) can be independently assessed. With sufficient training, pathology scoring can be repeated in an observer-independent manner. Furthermore, the use of a pathology score of sufficient magnitude (total possible score of 25) allows the differentiation of tissues with a wide range of pathological severity. Finally, the exclusion of some quantitative 33 measures of inflammation (mucosal edema, goblet cell number, cecal weight) allows the calibration and external verification of this scoring method with quantitative parameters. Table 2.1 - Salmonella murine enterocolitis histopathology scoring criteria. Region Description Severity Score Lumen Empty 0 Necrotic Epithelial Cells Scarce 1 Moderate 2 Dense 3 Neutrophils Scarce 2 Moderate 3 Dense 4 TOTAL 11 Surface Epithelium No Pathological Changes 0 Regenerative Change Mild/Patchy (<20%) 1 Moderate/Diffuse 2 Severe/Diffuse 3 Desquamation Patchy (<30%) 1 Diffuse (>30%) 2 Neutrophils in Epithelium 1 Ulceration 1 TOTAL 11 Mucosa No Pathological Changes 0 Crypt Abscesses Rare (<15%) 1 Moderate (15-50%) 2 Abundant (>50%) 3 Mucinous Plugs 1 Granulation Tissue 1 TOTAL 15 Submucosa No Pathological Changes 0 Monocytes/lymphocytes 1 small lymphoid aggregate 0 >1 small lymphoid aggregates 1 Large lymphoid aggregate and/or greatly increased cells 2 Neutrophils Absent 0 Single extravascular neutrophils 1 Neutrophil aggregate(s) 2 Edema Mild (<10% mucosal thickness) 0 Moderate (10-80% mucosal thickness) 1 Severe (>80% mucosal thickness) 2 TOTAL 16 TOTAL 125 34 2.2.3 Pathology scores correlate with external quantitative criteria of inflammation severity A s noted above, the semi-quantitative subjective pa thology score described here can be compared to external measures o f inf lammat ion . In order to test the v a l i d i t y o f the scor ing method developed, m ice were infected w i t h strains o f serovar T y p h i m u r i u m S L 1 3 4 4 harbour ing mutations i n var ious genes result ing i n the part ial or total attenuation o f inf lammatory severity (these strains w i l l be discussed i n subsequent chapters). Th i s resulted i n intestinal tissues f rom m ic e w i t h a range o f disease severity. Tissues from mice infected w i t h var ious strains and uninfected controls were isolated and scored, and the pathology scores were compared to three independent criteria: mucosa l thickness, goblet c e l l number and cecal weight . A s noted previous ly , mucosa l edema and hypertrophy results i n increases i n cecal mucosa l thickness w i t h increases i n inf lammatory pathology. Goble t c e l l number decreases as in f lammat ion worsens, as does cecal weight (due to expuls ion o f lumenal contents and tissue contraction). F o r a l l o f these parameters, there is robust (P<0.0001) correlat ion o f the non-parametric pathology score and quantitative cri teria (Figure 2 . 4 A - C ) . The use o f this his topathology scor ing method therefore accurately reflects the severity o f in f lammat ion across a broad range o f intestinal inf lammatory pathology induced b y serovar T y p h i m u r i u m . A 0.7T „ 2 N M ^ A Spearman r= -0.8351 P-Value <0.0001 B 5<H Q. £ 4 0 4 o 5 10 15 Pathology Score Spearman r*= -0.7227 P-Value <0.0001 i i 5 10 15 Pathology Score T 20 25 C 5 2 5 -Pathology Score Figure 2.4 Histopathology scores correlate with objective quantitative criteria of inflammation in Salmonella infected murine ceca. Streptomycin pre-treated mice were infected with a variety of Salmonella enterica serovar Typhimurium strain SL1344 mutants or wild-type bacteria or left uninfected to elicit a range of inflammatory pathological severity. The severity of intestinal pathology as assessed by pathology score (x-axes) was compared to quantitative measures of inflammation (y-axes). Cecal weight and goblet cell number both decrease as inflammatory pathology worsens, and were negatively correlated with histopathology * score (PO.0001, A,B). Mucosal thickness increases with inflammatory severity and correlated positively with histopathology score (P<0.0001, C). 35 2.2.4 Signs of intestinal inflammation occur by 8 hours post-infection and plateau between 24-48 hours In humans, onset of diarrhea and abdominal pain following ingestion of non-typhoidal Salmonella species occurs relatively rapidly, usually within 8 hours to 3 days. I sought to assess the timing of intestinal inflammatory pathology in susceptible C57B16 mice following serovar Typhimurium infection. To accomplish this, mice were infected with 106-108 C F U of wild-type serovar Typhimurium following streptomycin treatment as described above. Following inoculation of mice, tissues were harvested and assessed for inflammatory histopathology. Mi ld inflammatory changes in the cecum were evident by 8 hours post-infection (Figure 2.5A). Inflammation continued to develop and was characterized by significant pathological change in all affected regions of the cecum by 48 hours post-infection. Notably, intestinal pathology showed the most rapid increase in the lumen and surface epithelium of mice (Figure 2.5 B,C) with slower progression in the mucosa (Figure 2.5 D) and the slowest progression in the submucosa, where maximal pathology was not evident until 48 hours after infection (Figure 2.5 E). Subsequent increases in intestinal pathology scores (at five days) were noticeable and included several features not present at 48 hours such as consolidation of the intestinal lumen with purulent exudates and necrotic cells, and epithelial/mucosal ulceration (Figure 2.2 C,D, Figure 2.5). Peak bacterial loads in the cecum were achieved by 48 hours, but extraintestinal bacterial loads continued to increase until the death of the animals at approximately day 5. Late in infection, mice were affected with both localized pathology (e.g. intestinal inflammation) and systemic symptoms including ruffling of fur, hunching and weight loss consistent with the features of murine typhoid (Table 2.2). .The superimposition of systemic disease in susceptible mice limits the length of infection in this model to a maximum of five days. 2.2.5 Inflammatory changes occur with low bacterial doses In order to assess the inflammatory response of susceptible mice to various doses of serovar Typhimurium, mice were treated with streptomycin as above and 24 hours later infected orally with either vessel alone (sterile saline) or 102, 104, 106 or 108 C F U wild-type serovar Typhimurium. Following an infection of 48 hours, ceca were isolated, weighed, fixed, sectioned and stained for histopathology assessment, and spleens and colons were isolated for enumeration of Salmonella. Inoculation with sterile saline did not produce any evident gross or histological inflammatory changes in the cecum (Figure 2.6 A,B,E) and colonization of the colon and spleen 36 by Salmonella was not detected (Figure 2.6 C,D). Significant inflammatory pathology and intestinal and systemic bacterial colonization was achieved with the lowest administered dose of serovar Typhimurium, although intestinal inflammation only achieved maximal severity with administrations of at least 104 C F U Salmonella (Figure 2.6 A-D, F-I). These data indicate the sensitivity of susceptible streptomycin treated mice to intestinal inflammation induced by serovar Typhimurium. 25-i g 20-1 I'd a> f i O H ra Q. 7-i o 6-O c. o 3-o £ 2-ra Q. 5H o-l & T - r -48 16 24 32 Time (h) 72 Total 120 96 B .-0 o-l • 16 ^ 32 Lumen -T 48 Time (h) 72 96 - I 120 Figure 2.5 The timecourse of inflammatory pathogenesis in the ceca of mice infected with Salmonella enterica serovar T y p h i m u r i u m . Mice were pretreated with streptomycin and infected with 108 Salmonella enterica serovar Typhimurium as described in Materials and Methods. At various timepoints pos-infection, tissues were harvested and assessed for pathology. Total scores (A), and the scores for individual histological regions (B-E) were analyzed separately. Points indicate the mean value for 6-20 mice/timepoint. Error bars indicate standard deviations. 7-i %* >; 4-1 3" B 2-n fjJ • i i i i i 0 8 16 24 32 Surface Epithelium —i " - r -48 I I I 72 96 120 Time (h) 5-i o 4. co o O 2-D 5— 1 1 1 1 16 24 32 .-—•5 48 Time (h) 72 Mucosa 1 — 96 - I 120 6' <u O 5-o V) 4. O 3-2 2. *-< ra Q. 1 . 1 E ~r Submucosa 16 24 32 48 Time (h) 72 —r-96 -1 120 37 Table 2.2 The timecourse of intestinal and systemic illness in Salmonella enterocolitis Timepoint . Intestinal Inflammatory Features |8h Patchy P M N recruitment and epithelial irritation |16h diffuse P M N recruitment, epithelial regeneration |26h Diffuse P M N recruitment with foci, crypt abscesses , epithelial erosions |2 days Massive P M N recruitment including into lumen, epithelial necrotic damage and regeneration, crypt destruction, submucosa l edema |5 days Obliteration of lumen with P M N s and necrotic cells, epithyelial ulceration, crypt obliteration, submucosa l edema with infiltrate Systemic Features limited colonization of mesenteric lymph nodes, liver, spleen Established colonization of mesenteric lymph nodes, liver, spleen Established colonization of mesenteric lymph nodes, liver, spleen Mesenteric lymph node, liver, spleen colonization,,granulomae, necrotic death of hepatocytes and splenocytes Cytokine upregulation, septicemia, hepatosplenomegaly, organ failure, death e o o o lmm l lllim llilim llliim llliuq l mm l lllim 1 0 2 1 0 3 1 0 " 1 0 5 1 0 6 1 0 7 10 Bacterial Dose 106 Bacterial Dose a 7-6-O 3-5-1 D w 3- J c o Q) °- 3-1 u. o a> o | I I l l l i q 0 10 2 10 3 10* 10 s 10 6 10 7 10 8 10 s QJ-rH t Bacterial Dose • num i num i num iimm iimm iimm I U U U M 0 10 2 10 3 10 4 10 5 10 6 10 7 10 8 10 9 Bacterial Dose Figure 2.6 Murine intestinal inflammation can be induced with low doses of Salmonella enterica serovar Typhimurium. Mice were pretreated with streptomycin and infected with various doses of Salmonella enterica serovar Typhimurium as described in Materials and Methods. Total histopathology scores (A), cecal weight (B) and bacterial loads in the were assessed two days after infection with wild-type bacteria. Points indicate arithmetic means (A-B) or geometric means (C-D) for 8-20 mice/dose +/- standard deviations. Dashed lines indicate the threshold of detection of bacteria for each organ. A dose of 0 indicates uninfected mice. 38 2.2.(5 Inflammatory severity is comparable in NRAMP1 resistant and NRAMPlsusceptible mouse strains. As noted, Salmonella enterocolitis in the Nrampl susceptible mouse strain C57B16 is confounded by the presence of a systemic disease with features of murine typhoid. The presence of functional Nrampl increases the resistance of mice to murine typhoid (Govoni et al, 1996; Vidal et al, 1995). In order to determine i f Nrampl also determined susceptibility to intestinal inflammation induced by serovar Typhimurium, congenic mice with wild-type or mutant alleles of the Nrampl gene were infected with serovar Typhimurium following oral treatment with streptomycin, and the resulting intestinal inflammatory responses were compared. Mice were infected with 106 wild-type serovar Typhimurium as above, and infection was allowed to proceed for 2 or 5 days. These timepoints were chosen because at both, significant intestinal pathology is present in susceptible mice, but only at the later timepoint are features of systemic disease overt. Tissue pathology and systemic bacterial loads were assessed following removal of ceca and spleens. Two days following infection, there were no significant differences in either systemic tissue pathology or bacterial burden in the spleen (Figure 2.7 A-F). Mice from neither group at this time displayed any signs of morbidity. In contrast, by five days, Nrampl-/- but not Nrampl +/+ mice displayed significant morbid behaviour (ruffling of fur, hunching and weight loss) and Nrampl -/- mice had higher bacterial loads in the spleen (P=0.0012, T-Test, Figure 2.7D). Despite significant differences in systemic pathology, there was no difference in the severity of intestinal inflammation at day 5 (P>0.05, Figure 2.7 B,G-H), suggesting that disease pathology in the gut occurs independently of Nrampl. 39 25 22 o o to > Ol o o 20 15 10 4 £ 5 c at a> a. CO 3 LL o ro o Day 2 0 Submucosa • Mucosa O Epithelium • Lumen 25 -| O 20 -o (A 15 -O) o 10 -o • • B Day 5 • Submucosa O Mucosa • Epithelium • Lumen 1 Nramp -/- Nramp +/+ Nramp -I- Nramp +/+ i C 6H 5-4-3-2-1-0-7-, D c 6-«D 3 ro 3 3 H Nramp -/- Nramp +/+ Nramp -/- Nramp +/+ rrjp +/+ Figure 2.7 Intestinal inflammation is induced in mouse ceca by S. enterica serovar Typhimurium in the presence and absence of the host resistance factor Nrampl. Congenic Nrampl +/+ or -/- mice (Nramp -/-, Nramp +/+) were streptomycin treated and infected for 2 or 5 days with 106 wild-type Salmonella enterica serovar Typhimurium. At the indicated timepoints, tissues were isolated and cecal pathology (A, B, E-H)) and splenic bacterial load (C, D) were assessed. Pathology scores are the stacked means of pathology of the histological regions indicated + the standard deviation of the total pathology score for each group. Lines indicate the geometric mean of splenic load and each point represents a single mouse. Representative hematoxylin and eosin stained ceca from day 2 and day 5 are shown (100* magnification). 40 2.3 Discussion Following treatment with streptomycin, colonization resistance to Salmonella enterica serovar Typhimurium is reduced. Intestinal colonization of pre-treated, but not untreated mice occurs following infection with 102 C F U bacteria. Infection at higher levels is also established in the mesenteric lymph nodes, spleen and liver in this model. Streptomycin kills the majority of the normal host flora and similar colonization dynamics occur in germfree mice (Stecher et al, 2005) consistent with the hypothesis that efficient colonization of the intestine by Salmonella is inhibited by the normal flora. Following colonization, mice are affected by significant intestinal inflammation that is most severe in the cecum. Within 8 hours of oral infection with serovar Typhimurium one day following streptomycin treatment there is evidence of increased intestinal inflammation compared to uninfected control mice. Inflammatory changes progressively worsen until the mice succumb to superimposed systemic infection approximately 5 days following inoculation. Inflammatory pathology includes the infiltration of neutrophils into the intestinal lumen, epithelial necrosis and desquamation, mucosal architectural disruption and hyperplasia, submucosal edema and diffuse, transmural inflammatory infiltrate composed of predominantly of neutrophils. Like human infection, but unlike bovine and rabbit ileal loop infections, murine inflammatory pathology is,most severe in the cecum, with significant pathology in the colon and little pathology in the small intestine. Interestingly, although large doses of bacteria provide consistent inocula and a greater degree of reproducibility, as few as 300 bacteria induce significant inflammation and colonization by two days post-infection in this model. By assessing the presence of the histopathological features of Salmonella inflammation, a novel inflammatory pathology scoring method was developed. This method was sufficient to discriminate a large range of disease severity induced by a variety of fully virulent and partially or totally attenuated bacterial strains. Pathology scores correlated closely with objective, quantitative measures of inflammation including cecal weight, goblet cell number and mucosal thickness. This measure represents an important tool for the assessment and discrimination of histopathology induced in either infections with various mutant bacterial strains, or infections of mice with genetic defects that may impair or exacerbate their inflammatory response. This method was applied to both a dose-response and timecourse of intestinal inflammation and demonstrated significant discriminative resolution, not only for the total severity of 41 inflarnrnation, but also the tissue compartmentalization of tissue pathology. Specifically, application of this measure indicated that onset of inflammatory pathology progressed from the lumen outward, with early onset in the lumen and surface mucosa, slower onset in the mucosa and lastly inflammatory pathology of the submucosa. Notably, onset in the lumen and surface epithelium correlated closely, which is consistent with the hypothesis that lumenal infiltration of necrotic cells and neutrophils occur concurrently with pathogen-elicited changes in the surface epithelium, the likely point of first contact for Salmonella. The ability to spatially separate inflammatory pathology represents a significant advantage over other pathology scoring methods used to score intestinal inflammation, which group pathological features in a non-tissue specific manner (Barthel et al, 2003). While many of the inflammatory features of murine enterocolitis are common to the rabbit and bovine ileal loop infections, inflammation in murine infection can be practically examined in a large number of experimental animals of identical genetic composition over longer durations of infection. Extensive research tools (e.g. antibodies raised against murine epitopes, biochemical assay kits, etc.) have been generated for experimental mice which are unavailable for other experimental animals. Furthermore, mice with specific genetic defects can be compared to assess the contribution of various host-defense processes to disease pathogenesis. One major limitation of the murine enterocolitis model is the confounding influence of systemic infection on the course of intestinal disease. Due to the susceptibility of the commonly used inbred mouse strains C57B16 and B A L B / c to murine typhoid, mice infected, even following streptomycin treatment, succumb to systemic infection before the course of intestinal disease is completed. Using this model, we demonstrated that intestinal pathology occurs in both Nramplresislant and Nramplsusceptlble mice. While the use of resistant mouse strains and bacterial mutants attenuated for systemic infection provide opportunities to circumvent this critical limitation, the examination of enterocolitis in a susceptible mouse has an important theoretical basis. The molecular and cellular characterization of the roles of SPI-1 and SPI-2 in systemic and intestinal pathogenesis has largely been performed in susceptible model systems and the presence of an Nramplresistant allele would confound the comparison of murine enterocolitis and previously published work. Furthermore, in our case, there were practical limitations relating to the acquisition and production of sufficient mice of the Nramplresistant genotype to thoroughly investigate the model in this context. In addition, since the model was initially characterized in susceptible mice, I felt it appropriate and necessary to validate my results with the original data 42 as reference, which required the use of the same mouse strain. Largely, the rationale for not using mutant strains of bacteria to prevent the occurrence of murine typhoid is the same. Commonly used mutant bacterial strains that cause attenuated murine typhoid include metabolic mutants such as those deficient for the production of aromatic amino acids. These have been tested in rabbit ileal loop infection and as vaccine candidates in bovine and human infections and are attenuated and as such are unsuitable for assessing SPI-1 and SPI-2 function (Dougan et al, 1988; Everest et al, 1999; Hindle et al, 2002; Jones et al, 1991; Villarreal-Ramos et al, 1998). Despite these limitations, the advantages of the murine Salmonella enterocolitis model make it an appealing model system to study. Because infections can be performed in a large number of inbred animals over longer durations of infection than are possible with traditional ileal loop infections, we can compare a significant number of bacterial strains and the contribution of a greater number of bacterial virulence factors to an expanded course of infectipn in vivo. Thus, this model is an important new tool for the study of Salmonella enteropathogenesis, and provides an opportunity to reevaluate previously postulated models of intestinal infection and inflammation. 43 Chapter 3 - Salmonella typhimurium pathogenicity island 2 is necessary for complete virulence in a mouse model of infectious enterocolitis. Preface: This chapter is a modified version of a manuscript published in Infection and Immunity as: Coburn B , L i Y , Owen D , Vallance B , Finlay B B . Salmonella typhimurium pathogenicity island 2 is necessary for complete virulence in a mouse model of infectious colitis. InfectImmun. 2005; 73.-3219-27. Additional material (Figure 3.6 - SPI-2 expression analysis) was included from the manuscript published as: Brown N F , Vallance B A , Coombes B K , Valdez Y , Coburn B A , Finlay B B Salmonella Pathogenicity Island 2 Is Expressed Prior to Penetrating the Intestine. PLoS Pathogens 2005; 1(e32):0001-000?'. The contribution of the authors to this chapter are as follows: I designed and performed all experiments described in this chapter except: David Owen externally verified the histopathology scoring method and results, but did not perform the initial assessments. Bruce Vallance provided strategic guidance and intellectual input regarding the types of outcome measures used. SPI-2 expression analysis ( ' R I V E T ' experiment, Figure 3.6) was performed in collaboration with Nat Brown, who performed bacterial plating and enumeration of R I V E T strains. Yul ing L i provided technical assistance for mouse infections and tissue harvesting. 44 3.1 Introduction Salmonella species are facultative intracellular Gram-negative bacteria that cause a wide array of disease including systemic disease and enterocolitis in a multitude of hosts (reviewed in (Baumler et al, 1998)). Murine infection with Salmonella enterica serovar Typhimurium has been used predominantly to model human typhoid (caused by serovar Typhi), while bovine infection with serovar Dublin or serovar Typhimurium has been a prevailing model of intestinal disease. These models have been exploited to gain critical insight into the pathogenesis of disease caused by Salmonellae including for example: that invasion associated genes are required for intestinal secretory and inflammatory disease; that intracellular survival in both the intestinal epithelium and macrophages is essential for systemic pathogenesis; and that M cells of the ileal Peyer's patches are the site of invasion for systemic infection in murine typhoid prior to dissemination to liver and spleen via the reticuloendothelial system (Fields et al, 1986; Jones et al, 1994; Leung and Finlay, 1991; Watson et al, 1998). A central hypothetical theme that has emerged as a result of these discoveries is the distinct role of different virulence systems -Salmonella pathogenicity island 1 (SPI-1) and SPI-2 - in the pathogenesis of intestinal and systemic disease. Both of these horizontally acquired genomic islands encode a type III secretion system (T3SS), capable of secreting bacterial proteins into the host cell or extracellular milieu. The prevailing view is that SPI-1 is necessary for cell invasion and essential for intestinal disease, while SPI-2 is required for intracellular survival and persistence in target organs such as the spleen and liver (Santos et al, 2001b). Soon after the identification of these virulence systems, it was demonstrated that in mice, SPI-1 deficient serovar Typhimurium are attenuated during systemic disease following oral infection, but are not attenuated when introduced intraperitoneally (Galan and Curtiss, III, 1989; Watson et al, 1998), while SPI-2 mutants inoculated by the latter route are attenuated for systemic infection (Shea et al, 1996) but maintain intestinal virulence in cows and rabbits after oral infection (Everest et al, 1999; Tsolis etal, 1999a). Whether SPI-2 plays a role in intestinal inflammatory disease caused by serovar Typhimurium is unclear. It has been shown in cattle that diarrhea caused by serovar Typhimurium is SPI-1 but not SPI-2 dependent (Tsolis et al, 1999a) and that Salmonella strains lacking SPI-2 regulatory genes and wild-type Salmonella are equally pathogenic in a rabbit 45 model of gastroenteritis (Everest et al, 1999). However, it has also been reported that SPI-2 plays at least some role in bovine inflammatory disease (Bispham et al, 2001). Bispham et al, (Bispham et al, 2001) demonstrated that serovar Dublin lacking the SPI-2 translocon component SseD or the SPI-2 T3SS apparatus protein ssaT induced less fluid secretion into bovine ileal loops inoculated with mutant bacteria than did wild-type and that there was diminished neutrophil influx into infected tissue. There are significant practical and theoretical limitations of these models, however. Due to the technically demanding nature of bovine experiments, the number of experimental animals used in these experiments is limited. Furthermore, cattle used are typically outbred and show significant variation in disease severity depending on age. In addition, although investigations have demonstrated disease in the ilea of these animals, the role of SPI-2 in tissues more commonly affected in human serovar Typhimurium infection e.g. the cecum and colon, have not been investigated. Consequently, it is difficult to extrapolate the findings of these studies to human disease. A recently characterized model of infectious cecal inflammation (typhlitis) in mice provides an alternate model for the study of human serovar Typhimurium induced intestinal disease with significant advantages over the bovine model (Barthel et al, 2003). Following oral administration of streptomycin, mice challenged with serovar Typhimurium display signs of intestinal inflammatory pathology with many histopathological similarities to human disease including severe inflammation in the large bowel with little or no inflammatory pathology in the ileum (Barthel et al, 2003; Boyd, 1985; Mcgovern and Slavutin, 1979). This model has been utilized to demonstrate that Salmonella induced intestinal inflammation requires bacterial activity dependent on the SPI-1 effector SipA, as well as functional flagella and chemotaxis (Barthel et al, 2003; Hapfelmeier et al, 2004; Stecher et al, 2004). Antibiotic treatment is a risk factor for acquiring serovar Typhimurium colitis (Dore et al, 2004) and the dependence of inflammation on streptomycin treatment is likely due to alterations in the host microflora resulting in environmental changes and decreased colonization resistance (Que et al, 1986; Que and Hentges, 1985). In this study, we have exploited the streptomycin pre-treatment model to show that SPI-2 is required for complete intestinal virulence in the cecum and colon in vivo in a disease model that shares clinical and histopathological features with human disease. 46 3.2 Results 3.2.1 S. enterica serovar Typhimurium elicits intestinal inflammation that is most pronounced in the cecum of streptomycin pretreated mice. In order to assess induction of intestinal inflammation in response to serovar Typhimurium, we infected NramplSusceptible C57B1/6 mice orally with 3><108 wild-type serovar Typhimurium SL1344 24 hours after oral gavage with 20mg streptomycin. We observed extensive inflammatory changes in the large bowel of wild-type Salmonella infected mice similar to those reported previously (Barthel et al, 2003), as well as bacterial translocation to liver and spleen and ultimately fatal systemic disease (Figure 3.1). Common features in the intestine following Salmonella infection included neutrophil infiltrate into the intestinal lumen, surface epithelial erosion/desquamation, inflammatory infiltrate into the lamina propria and submucosa, crypt abscesses and submucosal edema. These features were absent from control mice given vessel alone. Epithelial proliferation in the mucosa was increased resulting in increased mucosal thickness (Figure 3.IF, Table 3.1). Submucosal edema was also evident (Figure 3.1 F, Table 3.1). Inflammatory features including mast cell recruitment and loss of goblet cells were common at 48 hours in infected mice (Table 3.1), but absent from controls. Pathological changes were also uncommon in mice infected with wild-type serovar Typhimurium but not pretreated with streptomycin. Bacterial translocation to the liver and spleen occurred as early as 6 hours post infection and bacterial colonization of these sites occurred consistently by 24 hours. At 48 hours post-infection intestinal inflammation was near maximal. Consistent with previous studies (Barthel et al, 2003), we observed that wild-type serovar Typhimurium-elicited intestinal inflammation at 48 hours is most severe in the cecum, with less severe inflammation in the colon and little or no inflammatory change in the ileum (Figure 3.1 A-C). We focused subsequent comparisons of intestinal pathology on the ceca at this timepoint due to the consistent severity of histopathology. 3.2.2 SPI-2 contributes to cecal inflammation in serovar Typhimurium infection In order to assess the role of SPI-2 in intestinal disease, we compared intestinal pathology at 48 hours in mice infected with wild-type serovar Typhimurium or bacterial strains lacking a functional SPI-1 or SPI-2 T3SS due to the mutation of an essential apparatus component (invA and ssaR respectively). In agreement with previous studies, we observed that a functional SPT1 47 T3SS is essential to elicit intestinal inflammation at 48 hours (Barthel et al., 2003; Hapfelmeier et al, 2004). Significant histopathological changes were absent from all SPI-1 infected mice in all three tissues examined (Figure 3.1 A-C) and histopathology scores were statistically indistinguishable from uninfected mice. Mucosal hypertrophy and submucosal edema were absent from SPI-1 infected mice and mast cell and goblet cell numbers were not different than uninfected controls (Table 3.1). Colon Cecum Ileum Lumen Surface Epithelium Mucosa Submucosa Figure 3.1 Salmonella enterica serovar Typhimurium elicited enterocolitis is most severe in the ceca of wild-type infected mice and, is partially attenuated in the absence of SPI-2 type III secretion. Streptomycin treated C57B1/6 mice were infected with 3><108 wild-type (WT), SPI-1 (AinvA) or SPI-2 (AssaR) T3SS mutant serovar Typhimurium. Histopathology scores from wild-type, AinvA mutant and ssaR mutant infected ceca (A), colons (B), and ilea (C) plus representative hematoxylin and eosin stained ceca (D-F) are shown. Histopathology is most severe in the cecum of all infected mice and is completely attenuated in AinvA infected mice. Inflammation in the ceca and colons of mice infected with SPI-2 mutants was statistically intermediate between wild-type and SPI-1 T3SS mutant infected mice by Kruskall-Wallis test (P<0.0001). P-values indicated represent Mann-Whitney U test comparison of totals between groups. (A-C) Each bar represents a single mouse. Images are shown at a magnification of 200*. Intestinal inflammation in the absence of SPI-2 type III secretion was also significantly attenuated compared to wild-type (P<0.0001, Figure 3.1, Table 3.1). In all three tissues assessed, intestinal pathology was intermediate between SPI-1 and wild-type infected mice (Figure 3.1, Table 3.1). The 'intermediate' phenotype associated with SPI-2 mutant infection was not due to the absence of one specific pathological feature, but rather reflected the diffuse attenuation of the inflammatory phenotype in all measured parameters (Figure 3.1 A , Table 3.1). This intermediate pathology was not due to differences in bacterial content of the tissues examined as bacterial 48 loads of all strains were comparable at the timepoint assessed (Table 3.1), suggesting that SPI-2 is actively involved in the induction of colitis and typhlitis. Table 3.1 Pathological features of Salmonella enterica enterocolitis after 48 hours infection induced by strains lacking SPI-1 and SPI-1 type III secretion. Strain P-value Pathological Change ASPI1 ASPI2 Wild-type (ANOVA) Logio CFU/g Colon 7.4 +/- 0.4 7.7 +/- 0.6 7.8 +/- .7 NS Cecal Weight (g) 0.47+/-.102 3 0.34+/- .091**3*** 0.16+/-.051 2 PO.0001 Submucosal edema 37+/- 151* (% Wall Thickness) 22 +/- 63* 26 +/- 11 P=0.0298 Mucosa (pm) 90 +/- 133*** 115+/- 153 192 +/- 521 2 P<0.0001 Goblet Cells/HPF 38 +/- 192*3*** 12+/-11*3*** 6 +/-1 p < 0.0001 Mast Cells/100 crypts 1.3 +/- 1.43** 1.2+/- 1.23** 15 +/- 121**2** P=0.0006 Values shown represent means +/- standard deviation for each group * PO.05, ** PO.01, *** P< 0.001 by Tukey's Multiple Comparison test 1 vs ASPI1, 2 vs ASPI2, 3 vs WT 3.2.3 SPI-2 mutants induce ICAM-1 expression and neutrophil recruitment less strongly than wild-type serovar Typhimurium. Since SPI-2 mutant bacteria caused attenuated intestinal inflammation, we sought to determine whether this was due a decreased ability to induce leukocyte recruitment. To do this, mice were treated with streptomycin and infected as before and after 48 hours ceca were cryosectioned and stained for the CD18-p 2 integrin receptor ICAM-1. In response to infection with wild-type serovar Typhimurium, we noted significant ICAM-1 expression in the mucosa and submucosal vasculature as has been previously reported (Barthel et al, 2003). Expression of ICAM-1 was absent from SPI-1 infected tissues, confirming that SPI-1 activity is essential to induce an inflammatory response in infected intestines at 48 hours. Epithelial cells maintained some ICAM-1 expression in response to SPI-2 deficient Salmonella, although it was less intense and extensive than that induced by wild-type bacteria (Figure 3.2, D-I). In order to determine i f differences in ICAM-1 expression correlated with differences in neutrophil recruitment, we stained sections of infected ceca with a neutrophil detection kit that detects activity of the neutrophil specific esterase. 49 ICAM-1 DAPI Merge G H 1 e Figure 3.2 SPI-2 but not SPI-1 T3SS mutant Salmonella enterica serovar Typhimurium induces ICAM-1 in the intestines of streptomycin treated mice 48 hours after infection. Streptomycin treated mice were infected with SPI-1 (AinvA, A-C), SPI-2 (AssaR, D-F) or wild-type (WT, G-I) serovar Typhimurium for 48 hours. Ceca were retrieved, paraformaldehyde-fixed, OCT-embedded and cryosectioned prior to staining for ICAM-1 (A, D, G, green in merge) and nuclei (DAPI - B, E, H, blue in merge) by immunohistochemistry. Isotype control antibody staining of wild-type infected mice was not significant (insets). ICAM-1 expression was evident in the epithelium and submucosal vessels of both wild-type and AssaR infected mice, but decreased in the latter. Ceca of AinvA infected animals showed minimal ICAM-1 expression. Epithelium = e, submucosal vessel = v. Pictures are representative of 8 mice/group and 4 tissue sections per mouse. Rare neutrophils were present in all sections, but were abundant only in wild-type infected ceca (Figure 3.3). Occasional small aggregates of neutrophils were present in SPI-2 infected ceca, but large aggregates were not observed (Figure 3.3B). Infection with wild-type Salmonella however, resulted in numerous large aggregates and greatly increased single cells in the mucosa, lamina propria and submucosa (Figure 3.3C). 5 0 Figure 3.3 Neutrophil infiltration is markedly reduced in Salmonella enterica serovar Typhimurium induced typhlitis in the absence of SPI-2 type III secretion (AssaR). Forty-eight hours after infection with SPI-1 T3SS mutant (AinvA,A), SPI-2 T3SS mutant (AssaR, B), or wild-type (WT, C) serovar Typhimurium ceca were retrieved, fixed and stained for the neutrophil specific chloroacetate esterase (red staining). Neutrophils were rare in ceca infected with SPI-1 or SPI-2 T3SS deficient Salmonella (A, B), but abundant in wild-type infected tissues. Pictures are representative of 8 mice/group and 4 tissue sections per mouse. 3.2.4 S. enterica serovar Typhimurium in the ceca of infected mice is predominantly lumenal and sensitive to gentamicin. In order to assess whether intestinal inflammation was attenuated in SPI-2 mutants due to changes in the distribution of the bacteria within the gut, we assessed bacterial localization by immunohistochemistry and a gentamicin protection assay. Twenty-four hours after streptomycin administration, mice were infected with wild-type, SPI-1 or SPI-2 mutant serovar Typhimurium as described above. Tissues were harvested 48 hours after infection and stained for Salmonella LPS, actin and nuclei. As observed previously, mice infected with SPI-1 or SPI-2 mutants or wild-type Salmonella displayed no, moderate and severe inflammatory pathology respectively (Figure 3.4 D, H, L). Bacterial staining was predominantly confined to the lumen of infected tissues in mice infected with all bacterial strains (Figure 3.4 insets). Bacteria that were in close association with the epithelium and mucosa were present in all infected mice. Severe inflammation was common in the absence of infiltrating bacteria. Intracellular bacteria were not observed in any infected tissues at this timepoint. To approximate the number of bacteria residing within cells, bacterial number from intestinal tissues were compared in the presence and absence of treatment with gentamicin, which does not penetrate cells and consequently spares intracellular bacteria. We found few detectable bacterial colonies from gentamicin treated intestines compared to those untreated (>10,000 difference, pO.0001, Figure 3.5). 51 • J • K L Figure 3.4 Salmonella enterica serovar Typhimurium in the ceca of infected mice is primarily extracellular. Bacterial localization of SPI-1 (AinvA, A-D), SPI-2 (AssaR, E-H) and wild-type (WT, I-L) serovar Typhimurium 48 hours after infection in the ceca of streptomycin treated C57B1/6 mice. Paraformaldehyde-fixed OCT-embedded cryosections were stained for actin (A, E , I, red in merge), nuclei (DAPI - B, F, J, blue in merge) or against Salmonella LPS (C, G, K, green in merge). Mucosal infiltration was uncommon and confined to areas of tissue destruction inflammation worsened (merges D, H, K). Images are a pseudocolour and are shown at 200* magnification. Insets are 100 x. 3.2.5 SPI-2 is expressed prior to penetration of the intestinal epithelium. Previous work has shown that SPI-2 is critical during intracellular infection. Our evidence that SPI-2 is important for intestinal disease prior to significant invasion of the intestine suggests an extracellular role for SPI-2 during enteropathogenesis. Presumably, this would require activation of SPI-2 expression. Using a sensitive recombinase-based in vivo reporter assay (RIVET) developed by Camilli and colleagues (Camilli et al, 1994) we assessed the expression 52 of SPI-2 during extracellular infection. In this technique, bacteria that have induced a SPI-2 gene activate a site-specific recombinase that excises a selectable marker (in this case a chloramphenicol resistance cassette) from the bacterial chromosome (resolution). We assessed the chloramphenicol sensitivity of wild-type and non-invasive SPI-1 mutant bacteria isolated from the lumen of murine ligated ileal loops after 15 minutes of infection. For these experiments, ligated intestinal loops were surgically constructed and inoculated with reporter strains of serovar Typhimurium. Bacterial localization was assessed in parallel by fixing tissue and staining for Salmonella LPS by immunohistochemistry. In this assay, the majority of bacteria had activated SPI-2 but had not yet invaded the intestinal epithelium (Figure 3.6 A,B). These data indicate that Salmonella activate SPI-2 prior to invading the intestine (a detailed investigation of this phenomenon was published in 2005 (Brown et al, 2005)). 9-i p<0.0001 p<0.0001 F i g u r e 3 5 X h e majority of Salmonella enterica g-j " serovar Typhimurium are susceptible to the Q • • extracellular killing of gentamicin in the cecum u 0 and colon of infected mice 2 days after infection. O • 0 C57B16 mice were streptomycin treated and infected O with serovar Typhimurium as described in Materials and Methods. Two days following infection, ceca 9 (open circles) and colons (closed circles) were isolated 0 and incubated for lhr at 37°C with 200ug/mL _ gentamicin or media alone prior to homogenization and enumeration of bacteria. P-values are the results of student's T-tests. Gm + w t invA~ Relevant genotype Figure 3.6 Salmonella pathogenicity island 2 (SPI-2) expression is induced prior to invasion of the intestinal epithelium by Salmonella enterica serovar Typhimurium. Murine ileal loops were inoculated with 100u.l Salmonella enterica serovar Typhimurium containing a SPI-2 reporter gene in strains with a functional or non-functional SPI-1 T3SS (WT, invA-respectively). Activation of SPI-2 after 15 minutes was assessed (A). Percent resolution indicates the percentage of Salmonella that had activated SPI-2 expression by 15 minutes. (B) Intestinal tissue was fixed and sectioned in parallel and stained for Salmonella LPS (red), nuclei (blue) and actin (green). 53 3.3 Discussion The data presented here is the first demonstration that SPI-2 is necessary for complete virulence in a model that recapitulates serovar Typhimurium elicited human colitis. Furthermore, our data suggest that SPI-2 is necessary for complete intestinal virulence during the bacterial-epithelium interaction rather than during intracellular parasitism. Although there are differences in the overall presentation of murine and human serovar Typhimurium elicited intestinal disease, including the absence of diarrhea and the development of systemic 'typhoid-like' infection in mice, this model has numerous advantages over existing models of Salmonella enterocolitis including low cost, availability o f mice and reproducibility. Like human intestinal disease induced by serovar Typhimurium murine inflammation is predominantly colitic with little or no ileal inflammation (Boyd, 1985; Mcgovern and Slavutin, 1979). In contrast, although colitis is observed in cows upon serovar Typhimurium infection, previously published models of serovar Typhimurium pathogenesis in cows have focused primarily on inflammatory disease o f the ileum. This focus may have arisen from substantial evidence that bacterial invasion in serovar Typhimurium infection occurs in the M cells of ileal Peyer's Patches, a behaviour required for systemic virulence (Jones et al, 1994; Penheiter et al, 1997). Although this may represent an important pathogenic mechanism in murine typhoid, it has not been demonstrated that intestinal invasion per se is essential for intestinal disease. Rather, it has been demonstrated that ileal inflammation requires invasion associated virulence genes (Wallis and Galyov, 2000; Watson et al, 1998) and that intestinal invasion in inoculated ilea occurs in a SPI-1 dependent manner (Bispham et al, 2001; Frost et al, 1997). It is in this paradigm that conflicting evidence for the involvement of SPI-2 in intestinal disease has arisen (Bispham et al, 2001; Everest et al, 1999). B y exploiting the murine serovar Typhimurium colitis model we have demonstrated for the first time that SPI-2 is necessary for complete virulence in a model representative of human disease. The SPI-2 virulence system is important in a range of bacterial adaptations to the host, including vacuolar remodeling, intracellular survival and resistance to the host immune response. Although survival of many host defenses is SPI-2 dependent, it is not clear whether this role of SPI-2 is important in inflammation of the large bowel. Our evidence that SPI-2 is expressed during extracellular infection is consistent with the hypothesis that SPI-2 is performing a different role at this stage of disease in the intestine than it does during murine typhoid. We 54 observe inflammation in this model at a timepoint at which the vast majority of bacteria are lumenal, suggesting that extracellular bacteria may be responsible for intestinal inflammation. Others have shown that the attenuation of intestinal virulence in SPI-2 mutant infected bovine ilea was not accompanied by a decrease in the number of intracellular bacteria (Bispham et al., 2001) and only a small proportion of total intestinal bacterial burden in infected human intestines is intracellular (Haque et al, 2004). SPI-2 is expressed by bacteria prior to invasion of the intestine. Intestinal epithelium-derived peptides such as a-/p-defensins and cathelicidin-related antimicrobial peptide (CRAMP) have anti-Salmonella activity in vitro and in vivo (Ayabe et al, 2000; Rosenberger et al, 2004; Salzman et al, 2003), and SPI-2 apparatus genes have been shown to play a role in resistance to antimicrobial peptides (Deiwick et al, 1998). It is possible that SPI-2 influences survival of bacteria close to the intestinal epithelium by enhancing resistance to antimicrobial peptides secreted into the intestine thereby promoting pathogenicity. While it is not clear what role SPI-2 plays in intestinal inflammation, our data imply that SPI-2 is important for induction of inflammation at the epithelial interface, rather than in deeper tissues, as ICAM-1 expression, a marker of epithelial inflammation and an inducer of neutrophil recruitment, was diminished in SPI-2 mutant infections. The uniform 'intermediacy' of the inflammatory phenotype associated with SPI-2 mutant infection suggests a decrease in the dose of inflammatory stimulus in the gut rather than the complete loss of a single virulence strategy. Although bacterial numbers are similar in the entire gut, proinflammatory stimuli such as flagellin may be compartmentalized differently within mice infected with various mutants. Flagellin is an important serovar Typhimurium inflammatory stimulus (Zeng et al, 2003), Salmonella lacking intact flagella are attenuated in this murine colitis model (Stecher et al, 2004) and it has recently been demonstrated that SPI-2 mediated vesicular transport is necessary for the transcytosis of flagellin across polarized epithelia (Lyons et al, 2004). The attenuation of cecal inflammation in the absence of SPI-2 may be a result of the decreased SPI-2-dependent delivery of proinflammatory flagellin to the basolateral epithelium. We have clearly demonstrated that SPI-2 is essential for complete virulence in the large intestines of infected mice. It is important now to carefully reconsider the dichotomous roles of SPI-1 and SPI-2 in the intestinal/systemic paradigm of serovar Typhimurium infection. 55 Chapter 4 - Analysis of the contribution of Salmonella Pathogenicity Island 1 and 2 to enteric disease progression Preface: This chapter is a modified version of a manuscript published in Infection and Immunity as: Coburn B*, Coombes B K * , Potter A A , Gomis S, Mirakhur K , L i Y , Finlay B B . Analysis of the contribution of salmonella pathogenicity islands 1 and 2 to enteric disease progression using a novel bovine ileal loop model and a murine model of infectious enterocolitis. InfectImmun. 2005; 73:7161-9. * Authors contributed equally to this work I designed and performed all experiments described in this chapter except: Brian Coombes and Andrew Potter performed bovine ileal loop surgeries with the assistance of K Mirakhur. Susantha Gomis took pictures of bovine ileal loop sections. Yuling L i provided technical assistance during the infection of mice and harvesting of murine organs for bacterial enumeration and pathology. 56 4.1 Introduction Salmonella enterica are facultative intracellular enteric bacteria that cause a range of clinically important infections in humans and commercial livestock. Within the S. enterica species there are numerous serovars, discriminated by their lipopolysaccharide, flagellar and capsular antigens. The natural progression of Salmonella disease is strongly influenced by both the S. enterica serovar and the host. For example, serovar Typhimurium is a pathogen with a wide host range, infecting humans, cattle, mice and chickens. In humans and cattle, infection with serovar Typhimurium manifests as enterocolitis that rarely spreads to systemic organs. In susceptible mice however, serovar Typhimurium efficiently crosses the gut epithelium and colonizes the spleen and the liver (Santos et al, 2001b; Tsolis et al, 1999b). Infection of chickens is mainly a colonization model that rarely manifests in disease. Two major virulence determinants involved in Salmonella pathogenesis are encoded in large chromosomal pathogenicity islands called Salmonella pathogenicity island (SPI)-l (Galan, 1996) and SPI-2 (Ochman et al, 1996; Shea et al, 1996). Both SPI-1 and SPI-2 encode separate type III secretion systems (T3SS) that introduce virulence proteins (called effectors) into the host-environment either by translocation directly into host cells or possibly by secretion into the vicinity of host cells(Galan, 2001; Ghosh, 2004; Hueck, 1998). Effector proteins translocated by the SPT1 T3SS influence early host cell cytoskeletal and membrane rearrangements involved in bacterial uptake into target cells (Galan, 1999) whereas SPI-2 is generally thought to play a role during intracellular infection by allowing the formation of Salmonella replicative vacuoles and evading host cell defenses (Waterman and Holden, 2003). Animal models are available to study both intestinal and systemic phases of salmonellosis caused by serovar Typhimurium. Cattle have been used to study the enteric disease caused by this organism, whereas infected mice develop a systemic infection that shares features with human typhoid. Salmonella enterica serovar Dublin is also commonly used in bovine infection models because this particular serotype is able to initiate both intestinal and systemic phases of infection in cows. An important theme to emerge from these models is the seemingly dichotomous role played by the SPI-1 and SPI-2 T3SS during the intestinal and systemic phases of salmonellosis. Whereas the SPI-1-encoded T3SS plays an essential role in colonization of the bovine intestine and in bovine enteropathogenesis (Galyov et al, 1997; Jones et al, 1998; Wallis et al, 1999b; Wood et al, 1998), this virulence trait has been reported to play little to no role in 57 systemic infection (Galan and Curtiss, III, 1989). Conversely, the SPI-2-encoded T3SS is more strongly associated with systemic virulence and associated pathology (Cirillo et al, 1998; Hensel et al, 1998; Ochman et al, 1996; Shea et al, 1999) than intestinal disease. The role of the SPI-2-encoded T3SS in enteric disease has been less studied and appears more contentious. Earlier work demonstrated that SPI-2 mutants maintain intestinal virulence during oral infection of cows (Tsolis et al, 1999a) and in ileal loops in rabbits (Everest et al, 1999). However, one study using bovine ileal loops infected with serovar Dublin (Bispham et al, 2001) indicated that mutations in SPI-2 reduced the intestinal secretory response compared to wild-type serovar Dublin. Another study using oral infections of calves with a bovine isolate of serovar Typhimurium (Tsolis et al, 1999a) indicated that SPI-2 mutants produced intestinal lesions of reduced severity in calves but nevertheless caused mortality and acute diarrhea. One of the confounding factors for the study of enteric disease in cattle caused by serovar Dublin is the highly invasive nature of the infection that results in a disseminating bacteremia (Zhang et al, 2003). In contrast, natural and experimental infection of calves with serovar Typhimurium results in an enteric disease localized to the gut without systemic involvement, which shares similar pathologic and clinical features to human enterocolitis caused by this same organism (Zhang et al, 2003). Recently a mouse model of Salmonella-induced colitis has been developed that relies on the pretreatment of animals with streptomycin (Barthel et al, 2003). This model has been used to demonstrate that Salmonella-induced colitis at day two after infection requires the SPI-1 effectors SipA, SopE and SopE2 (Hapfelmeier et al, 2004), plus flagella and chemotaxis (Stecher et al, 2004). Using this model, we have extended these findings by showing that the SPI-2 type III secretion system contributes to the intestinal inflammatory phenotype in the cecum and proximal colon two days following oral infection of mice with S. enterica serovar Typhimurium (Coburn et al, 2005). Here, we elaborate on these findings and demonstrate that, not only is SPI-2 involved in intestinal disease, but significant intestinal inflammation can occur in the absence of SPI-1. Our results identify novel phenotypes for serovar Typhimurium mutants with deficiencies in SPI-1 and SPI-2 type III secretion systems, which we have demonstrated in a murine model of infectious colitis and confirmed in the extended bovine ileal loop model. The working model presented here supports the view that during the progression of enteric disease, the presence of SPI-2 overcomes the previously reported requirement of SPI-1 for intestinal inflammatory disease. 58 4.2 Results 4.2.1 Ceca of streptomycin-treated mice infected with SPI-1 and SPI-2 mutants show attenuated pathology at 48 h post-infection A model of Salmonella-induced colitis using mice pretreated with streptomycin has been developed (Barthel et al, 2003). Using this model we (Coburn et al, 2005) and others (Hapfelmeier et al, 2005) reported that SPI-2 contributes to the induction of intestinal inflammation in mice. In the latter study, a SPI-1 apparatus mutant was compared to a SPI-2 translocon mutant, however a direct comparison in mice of SPI-1 and SPI-2 apparatus mutants devoid of all type III secretion and translocation activity has not been previously described later than two days post-infection. In order to directly compare the relative contribution of SPI-1 and SPI-2 to the full course of murine intestinal inflammation, streptomycin-treated mice were infected with wild-type S. enterica serovar Typhimurium or mutants lacking functional SPI-1, SPI-2 or both SPI-1 and SPI-2 type III secretion systems (invA and ssaR respectively). Intestinal inflammation was evaluated and bacterial loads were enumerated in the intestine and spleen. By day two, significant differences in the severity of histopathology elicited by wild-type, SPI-1 and SPI-2 mutant bacteria were evident (Figure 4.1 A). Wild-type Salmonella elicited severe and diffuse inflammatory changes in the ceca of infected mice. In contrast, significant attenuation of inflammatory pathology was apparent in SPI-2 infected ceca, including decreased lumenal inflammatory cell infiltrate and epithelial debris, mild or patchy surface epithelial necrosis and regenerative change, little inflammatory cell recruitment into the mucosa and diminished submucosal edema (Figure 4.2). Inflammatory pathology was virtually absent at two days in tissues infected with Salmonella lacking SPI-1 or both SPI-l/SPI-2 type III secretion systems (Figure 4.1 A). Intestinal colonization by wild type Salmonella and each of the mutant strains was similar at two days and translocation to the spleen was observed at similar levels in all strains under investigation (Figure 4.3 A,B). 59 25 0 ; invA ssaR P-value vs WT • Submucosa • Mucosa I Surface Epithelium • Lumen < 001 + <.05 < 05 25 2C B P-value vsWT • Submucosa O Mucosa Q Surface Epithelium B Lumen <.001 Figure 4.1 Salmonella enterica serovar Typhimurium SPI-1 mutants cause delayed typhlitis in streptomycin treated mice in the presence of functional SPI-2 type III secretion. (A) Pathological scores in streptomycin-pretreated mice 48 h post-infection with wild-type serovar Typhimurium or strains lacking functional SPI-1 (invA), SPI-2 (ssaR)or both type III secretion systems. (B) Pathological scores from mice infected for 120 h (5 days). Pathological scores are stacked averages of pathological changes in the lumens, surface epithelia, mucosa and submucosa from 8 mice in each group, error bars indicate the standard deviation of the total score. P-values represent the results of Dunn's post-Kruskall-Wallis tests. All p-values are derived from comparisons to wild type Salmonella (invA+, ssaR+). WT invA- ssaR- invA- ssaR-Figure 4.2 Histopathological changes in Salmonella enterica infected mice. Tissue sections from murine ceca infected for 48 h (A, C, E, G) and 120 h (B, D, F, H) post-infection with wild type Salmonella enterica serovar Typhimurium (A-B) or strains lacking functional SPI-1 (invA, C-D), or SPI-2 (ssaR-, E-F) type III secretion systems or both (invA-, ssaR-, G-H). Paraffin embedded 5 micron hematoxylin and eosin stained tissue sections are shown at 400 x magnification. Photographs are of representative tissues for each group. 60 A Colon Load 48 10-9 -O) 8 -17-1 M 3 2 invA ssaR • P-value vs WT c ra O o 8 74 6 5 4 3 2 1 invA ssaR + + ns ns ns Spleen Load 120 Q Spleen Load 48 c 54. °> on2H H O) 3 U. O O) o 10-, 9 8 4 7 6-5-4-3 0 -invA ssaR + + ns ns ns • Colon Load 120 invA ssaR + + <001 ns <001 < 001 ns ns Figure 4.3 Bacterial loads in murine Salmonella enterocolitis. Bacterial loads in the colon (A,C) and spleen (B,D) of streptomycin-treated mice infected with indicated strains of bacteria 48 h and 120 h after infection. Each black circle represents data from one mouse. Bars represent the geometric mean for the group. The type III secretion status of each strain is given below each panel next to the respective locus. Bacterial strains lacking components of the type III secretion systems encoded by SPI-1 (invA) and SPI-2 (ssaR) or both were compared to wild-type Salmonella enterica serovar Typhimurium SL1344. P-values represent Tukey's post-ANOVA tests compared to wild-type (invA+, ssaR+). 4.2.2 Murine ceca infected with SPI-1/SPI-2 double mutants display little pathology at 5 days after infection, while ceca infected with SPI-1 mutants show severe pathological changes. Our previous work using streptomycin-pretreated mice demonstrated a role for SPI-2 in an inflammatory phenotype two days after infection, however we did not assess the combined contributions of SPI-1 and SPI-2 to this disease process later than two days after infection. In order to compare the relative contributions of SPI-1 and SPI-2 virulence mechanisms during the 61 course of intestinal inflammatory pathogenesis in mice, we infected streptomycin-treated mice orally with wild-type, SPI-1, SPI-2 or SPI-1/SPI-2 mutant serovar Typhimurium and assessed intestinal disease at day 5. In contrast to the lack of intestinal inflammation elicited by SPI-1 mutants at two days after infection, significant histopathological changes were apparent in the ceca of mice infected with either wild type or SPI-1 mutant Salmonella at day 5 after infection (Figure 4.IB, Figure 4.2). Serovar Typhimurium SPI-1 mutants that retained a functional SPI-2 T3SS secretion system elicited severe and diffuse inflammatory changes in the ceca of infected mice. In contrast, in mouse tissues infected with mutants lacking both a functional SPI-1 and SPI-2 T3SS, we observed a near total attenuation of inflammatory pathology including little or no luminal inflammatory cell infiltrate and epithelial debris, normal surface epithelial appearance and architecture, little inflammatory cell recruitment into the mucosa and no submucosal edema (Figure 4.1, Figure 4.2). The presence of SPI-1 was not sufficient to induce an inflammatory response at this time in the absence of SPI-2. These data indicate that although SPI-1 is important for the induction of inflammatory responses in the gut of Salmonella infected mice early in infection, it is not required for inflammatory induction later in enteropathogenesis. Furthermore, SPI-2 is crucial for intestinal inflammatory pathogenesis in mice in the absence of SPI-1 encoded type III secretion. 4.2.3 S. enterica serovar Typhimurium invades the intestinal mucosa by 5 days after infection following streptomycin treatment. As we have previously described, early in enteropathogenic infection with serovar Typhimurium (2 days) bacteria are predominantly confined to the intestinal lumen. By immunohistochemistry, we sought to assess whether wild-type infection was accompanied by significant tissue infiltration by 5 days after infection. Unlike early infection, late infection is marked by significant and extensive infiltration of Salmonella into the intestinal wall (Figure 4.4). Bacteria are visible within surface epithelium, lamina propria and submucosa by 5 days after infection. The evident intestinal invasion associated with Salmonella infection at this timepoint is consistent with the hypothesis that distinct pathological processes may be involved in early and late enteropathogenesis in murine infection. 62 L Figure 4.4 Murine Salmonella typhlitis is characterized by bacterial invasion of the mucosa at five days but not two days post-infection. Invasion of the intestinal epithelium (e), lamina propria (lp) and submucosa (sm) by wild-type Salmonella enterica (green) is significant only at the later timepoint in streptomycin treated mice. Paraformaldehyde-fixed O C T -embedded cryosections were stained for actin (red), nuclei (DAPI, blue) or against Salmonella LPS. Images are a pseudocolour and are shown at 200x magnification. Insets are 4 0 0 . 4.2.4 Calf intestinal loops infected with SPI-1 and SPI-2 mutants show reduced secretory response and similar pathological changes at 24 h post infection. A l t h o u g h these studies have identif ied a role for SPI-2 i n the ceca o f susceptible mice , as yet these data have not been extended to the bovine i l ea l loop infect ion mode l . The l igated i lea l loop m o d e l i n cattle is a p roven tool for s tudying enteropathogenesis o f Salmonella serotypes ( W a l l i s and G a l y o v , 2000; Z h a n g et al, 2003). Howeve r , due to obstruction o f the smal l b o w e l w i t h mul t ip le ligatures, one recognized l imi ta t ion o f this mode l i n its present usage is the inab i l i ty to moni tor pa thologica l changes i n response to Salmonella at t imes longer than 12 h post- infection (Zhang et al, 2002b). Th i s has precluded the invest igat ion o f the relative contr ibut ion o f SPI-1 and SPI-2 to the progression o f intestinal disease i n bovine loops and the compar i son to other intestinal inf lammatory models o f serovar T y p h i m u r i u m infect ion. T o address this l imi ta t ion , w e have developed a nove l c a l f i l ea l loop mode l that restores the patency o f the smal l intestine fo l l owing surgery. Th i s feature overcomes the t ime constraints o f tradit ional loop experiments, permit t ing the examinat ion o f vi rulence determinants at later stages o f enteric disease i n a h i g h l y control led environmental context. W e examined intestinal pathology at 24 h and five days f o l l o w i n g infect ion w i t h w i l d -type S. enterica serovar T y p h i m u r i u m , or the same SPI-1 and SPI-2 apparatus mutants used i n our investigations o f the mur ine enterocolitis mode l ( i n v A and ssaR mutants). Infected i lea l loops were examined for pa thologica l changes after infect ion. A t 24 h , there was litt le difference i n the degree o f co lon iza t ion between w i l d type Salmonella and the SPI-1 or SPI-2 mutants. 63 H o w e v e r , b y 5 days after infect ion, a role for SPI -2 , but not S P I - 1 , i n intestinal co lon iza t ion was apparent, (Figure 4.5 A ) . M u c o s a l necrosis and edema w i t h acute in f lammat ion o f the intestinal loops due to inocula t ion w i t h w i l d type Salmonella or the SPI-1 or SPI-2 mutants was s imi la r at 24 h after infect ion (Figure 4 . 5 B , F igure 4.6). A s another measure o f pa thologica l change i n the epi thel ium, w e measured the v i l l u s height i n several tissue sections o f intestinal epi thel ium from loops infected w i t h wi ld - type Salmonella and each o f the SPI-1 and SPI-2 mutants. V i l l u s height correlated w i t h the pathologica l score and demonstrated more severe v i l l u s atrophy i n tissues w i t h greater pa thologica l changes (Figure 4.5 D , E ) . A t 24 h after infect ion there was li t t le correlat ion between the degree o f pa thologica l change and the in i t i a l infectious dose o f Salmonella. 4.2.5 Calf ileal loops infected with SPI-2 mutants display decreased pathology at 5 days after infection T o date, no studies have examined pa thologica l changes i n Salmonella-infected ileal loops later than 12 hours post-infection. In order to examine whether the natural progress ion o f intestinal disease i n calves f o l l o w i n g infect ion w i t h various Salmonella mutants fo l lows the same course as that observed i n mice , w e infected bov ine i l ea l loops for f ive days and examined intestinal pathology. Af te r 5 days o f infect ion, tissue damage and in f l ammat ion i n the intestinal loops infected w i t h w i l d type Salmonella was severe and demonstrated tissue necrosis, submucosal edema and f lu id accumulat ion into the intestinal l umen (Figure 4 . 5 C , F igure 4.6). In contrast, intestinal tissue infected w i t h SPI-2 mutant Salmonella had m i l d to moderate in f lammat ion w i t h m i l d v i l l u s atrophy, and b y day 5 f o l l o w i n g infect ion, began to show signs o f tissue regeneration and restoration (Figure 4.6). N o pa thologica l features o f disease were observed i n control loops inoculated w i t h saline (Figure 4 . 5 D , F igure 4.6) ind ica t ing that the surgical procedure d i d not account for the observed changes. 64 WT ssaR- invA- WT ssaR- invA-24 hours 5 days Figure 4.5 Salmonella enterica serovar Typhimurium SPI-1 mutants cause delayed typhlitis in inoculated bovine ileal loops, but SPI-2 mutants are attenuated. (A) Bacterial loads from calf ileal loops at 24 h after infection with saline alone, wild type Salmonella enterica serovar Typhimurium or strains lacking SPI-2 or SPI-1 type III secretion (ssaR-, invA-, respectively). Pathology scores from calf ileal loops at 24 hours (B) and 5 days (C) after inoculation with villus height from matched tissue sections was also determined from loops infected with 1><106 CFU at 24 h (D) and 5 days (E) after infection. B 25 • e o o CO >» O ) o o J C « 5 • 0_ 20 • 15 • 10 • • Submucosa • Mucosa • Surface Epithelium • Lumen Saline WT 24 h ssaR- invA-„ ! ..... cn • J 1.0 JC J 0.9 H > 0.8 H 0.7 saline WT ssaR invA c 25 cu s_ 20 o o CO > 15 o — 10 o JC •*-> TO 5 CL • Submucosa • Mucosa • Surface Epithelium • Lumen 120 h R Saline WT ssaR- invA-I.OO-i (mm' 0.75 JO gl 0.50 £ > 0.25 0.00-saline WT ssaR invA 65 24 h 5 days Figure 4.6 Histopathological changes in calf ileal loops at 24h and 5 days post-infection. Histopathological changes in calf ileal loops at 24 h and 5 days after inoculation with either saline alone (A, B, B'), wild type Salmonella enterica serovar Typhimurium (C, D, D'), or mutants lacking SPI-1 (invA-, E, F, F') or SPI-2 (ssaR-, G, H, H') type III secretion systems. Shown are hematoxylin and eosin-stained sections of the intestinal epithelium. White bars correspond to 0.250 mm. 4.2.6 SPI-1 mutants produce severe intestinal pathology in calf ileal loops at five days post infection Previous studies employing the 8-12 h infection model of calf ileal loops have established a role for SPI-1 in the early stages of intestinal pathology (Tsolis et al., 1999a; Zhang et al, 2002b), however, our observations and those of Hapfelmeier et al. (Hapfelmeier et al, 2005) demonstrate that SPI-1 independent colitis occurs in murine enteropathogenesis. In order to 6 6 examine whether the pathology at later t imes after infect ion was also SPI-1 independent i n bovine infections, w e infected c a l f i l ea l loops for f i ve days p r io r to examinat ion o f intestinal pathology. A s observed i n mice , intestinal tissues infected w i t h wi ld - type or SPI-1 mutant Salmonella d i sp layed severe pa thologica l lesions characterized b y severe inf lammatory c e l l inf i l t ra t ion, submucosal edema, necrosis, and vascular thrombosis w i t h v i l l u s atrophy at f ive days after infect ion (Figure 4 .5E , F igure 4.6). L o o p s inoculated w i t h saline or w i t h SPI-2 mutants (described above) had d imin i shed or absent pathologica l changes indica t ing that neither the surgical procedure nor the presence of Salmonella per se i n the intestinal loops contributed to this associated pathology. L o o p s infected w i t h wi ld - type Salmonella or SPI-1 mutants also showed a greater degree o f v i l l u s atrophy compared to loops infected w i t h SPI-2 mutants (Figure 4 .5F) . These data indicate that the later progress ion o f bov ine enteric disease induced b y S. enterica serovar T y p h i m u r i u m can occur independently o f the SPI-1 T 3 S S and is par t ly dependent o n SPI -2 , as was observed i n mice . 4.2.7 Both SPI-1 and SPI-2 independent intestinal inflammation require aromatic amino acid biosynthesis. The ava i lab i l i ty o f aromatic amino acids is restricted w i t h i n the S C V and Salmonella incapable o f synthesizing these amino acids are avirulent i n m ice (Hoise th and Stocker, 1981; Stocker et al, 1983). Studies o f the v i ru lence o f serovar T y p h i m u r i u m and D u b l i n strains l ack ing aromatic synthesis genes {pro genes) i n murine, bovine , rabbit and human infections indicate that they are effective vaccine candidates and cause attenuated systemic and intestinal disease (Dougan et al, 1988; Everest et al, 1999; H i n d l e et al, 2002; Jones et al, 1991; V i l l a r r e a l - R a m o s et al, 1998). Salmonella l a ck ing aro genes fa i l to achieve h i g h bacterial loads i n the spleens and l ivers o f infected mice , and are thought to be incapable o f intracel lular repl icat ion. In their recent characterization o f SPI-1 and SPI-2 independent col i t i s i n mice , Hapfe lmeier et al. (Hapfelmeier et al, 2005) show that Salmonella l ack ing either a functional SPI-1 T 3 S S or aro A cause inf lammatory disease, but those w i t h mutations i n both are attenuated. Since SPI-1 mutants require both aromatic amino ac id synthesis and SPI-2 they conclude that SPI-2 dependent in f lammat ion requires intracel lular repl icat ion. Furthermore, they c l a i m that this dependence o f SPI-1 independent in f lammat ion o n aroA is unique. Howeve r , they d i d not test mutations o f the SPI-2 T 3 S S and aromatic amino ac id biosynthesis i n combinat ion . In order to test the assertion that o n l y SPI-1 T 3 S S deficient serovar T y p h i m u r i u m requires aroA, w e 67 constructed bacterial strains w i t h mutations i n aroA, invA (SPI-1 T 3 S S ) and ssaR (SPI-2 T 3 S S ) alone and i n combinat ion , and assessed the vi rulence o f these strains i n mice . Contrary to what was postulated b y Hapfe lmeier et al, Salmonella strains l ack ing SPI-1 or SPI-2 both required aroA for v i rulence (Figure 4.6). In fact, the inf lammatory pathology i n SFl-2/aroA double mutants showed an even greater degree o f attenuation than that observed i n the SPl-l/aroA double mutant. In addit ion, bacterial co lon iza t ion o f the c o l o n was greatly reduced i n the S P I -llaroA mutant to a greater degree than that seen for the SFl-l/aroA double mutant (Figure 4 . 6D) . Predictably, co lon iza t ion o f the spleen was not detected i n SFl-2/aroA double mutants. A » • Submucosa • Mucosa • Surface Epithelium • Lumen w 0.9-, 0.8-3 °-7-£ 0.6-ro 5 0.4 0.H 0.0-aroA + invA + ssaR + + + + + B « , 7-c _a> 6-Q. co 5-u. 4. O ro o 3 2-aroA invA ssaR D 11, I 9" " 8-f * & 6-? 5-3 4-+ + + + + ns + + + + *** aroA + invA + ssaR + + + + + Figure 4.7 Both SPI-1 and SPI-2 independent delayed inflammation in Salmonella enterica infected murine ceca requires the aromatic amino acid biosynthesis gene aroA. Mice were infected with bacterial strains lacking SPI-1, SPI-2 type III secretion (invA-, ssaR- respectively) or both (invA-, ssaR-) in the presence or absence of the aromatic amino acid biosynthesis pathway gene aroA. Pathology scores (A), splenic bacterial loads (B), cecal weights (C) and intestinal bacterial loads (D), were assessed 5 days after infection. Pathology scores indicate stacked averages of pathological changes as described in Materials and Methods plus standard deviations. Each closed circle for other measures indicates a single mouse. P-values represent Dunn's post-Kruskall-Wallis tests (A) or Bonferroni's multiple comparison post-ANOVA tests (B-D). 68 Because Hapfe lmeier and colleagues used Salmonella strains l ack ing the SPI-2 translocon component sseD rather than a SPI-2 T 3 S S apparatus mutant as a SPI-2 deficient strain for the majori ty o f their studies, w e also constructed and tested bacterial strains w i t h non-polar mutations i n sseD alone or both sseD and aroA and tested their virulence i n mice . U n l i k e the T 3 S S apparatus mutants, t ranslocon mutants are able to secrete bacterial effectors into the extracellular m i l i e u , although l ike apparatus mutants are unable to inject them into host cel ls . L i k e the SPI-2 T 3 S S apparatus (ssaR) mutant, SPI-2 translocon mutants required aro A to induce inf lammat ion. These data conf i rm that a requirement for aromatic amino ac id biosynthesis is not unique to SPI-1 independent col i t i s , but is also a requirement o f SPI-2 independent col i t i s . AinvA AssaR Figure 4.8 Histopathological changes at 5 days post-infection in streptomycin-pretreated mice infected with wild-type Salmonella enterica serovar Typhimurium (A) or strains lacking functional SPI-1 (invA, B, E), or SPI-2 (ssaR-, C,F) type III secretion systems in the presence (A-C) or absence (D-F) of the aromatic amino acid biosynthesis pathway gene aroA. Paraffin embedded 5 micron hematoxylin and eosin stained tissue sections are shown at 400* magnification. Photographs are of representative tissues for each group. 6 9 4 .3 Discussion O u r data c lear ly demonstrate that, not o n l y is SPI-2 important i n mul t ip le models o f Salmonella enterocolitis, but significant intestinal disease can be induced b y serovar T y p h i m u r i u m i n the absence o f the SPI-1 T 3 S S . Prev ious studies o f Salmonella enteropathogenesis have indicated that SPI-1 is essential for intestinal disease i n mul t ip le mode l systems. Indeed, it has been reported us ing the mouse streptomycin-pretreatment mode l that the SPI-1 effectors, S ip A , S o p E and S o p E 2 , are required for Sa lmonel la - induced intestinal pa thology at day two after ora l infect ion (Hapfelmeier et al, 2004). O u r data and a report f rom another group (Hapfelmeier et al, 2005) indicate that serovar T y p h i m u r i u m can overcome this p rev ious ly postulated requirement for SPI-1 i n mur ine infectious enterocolit is. Furthermore, w e demonstrate that i n bov ine i l ea l loop infections o f sufficient duration, there is a s imi la r SPI-1 independent in f lammat ion induced b y serovar T y p h i m u r i u m . One o f the important f indings o f this w o r k was the concordance between the mouse m o d e l o f infectious col i t i s and the 5-day i l ea l loop m o d e l i n calves for d i sc r imina t ing the phenotypes associated w i t h major Sa lmone l l a vi rulence factors. A requirement for SPI-2 i n the progression o f intestinal in f lammat ion was found i n both an imal models , as was a s imi la r induct ion o f in f lammat ion b y Salmonella i n the absence o f SPI -1 . These observations were consistent despite noteworthy differences between mur ine and bov ine infections. S u c h differences include (i) the presence o f co-occurr ing systemic infect ion i n m ice and the absence o f this systemic component i n cows , (i i) the induct ion o f in f lammat ion i n the smal l b o w e l i n bov ine infect ion and i n the large b o w e l i n mur ine infect ion and ( i i i ) the necessity to treat m ice w i t h antibiotics i n order to induce intestinal disease, wi thout any such requirement i n cows . The dependence o n SPI-2 for the progression o f intestinal disease i n an infect ion m o d e l largely confined to the intestine (bovine) is noteworthy, as the superimposed systemic disease i n streptomycin treated mice can confound the interpretation o f intestinal phenotypes. A s imi la r progression o f S. enterica serovar D u b l i n infect ion o f cattle, i n w h i c h the h i g h l y invas ive bacteria disseminate to systemic sites o f infect ion, can also encumber the interpretation o f enteric phenotypes o f certain bacterial mutants under study. A l t h o u g h the microenvi ronment and phys io logy o f a l igated intestinal loop and a patent gut l i k e l y differ, the use o f l igated loops affords desirable experimental condit ions and endpoints for the study o f serovar T y p h i m u r i u m pathogenesis. L i k e w i s e , the early co lon iza t ion 70 b y Salmonella o f the mur ine intestine f o l l o w i n g s t reptomycin treatment requires careful interpretation due to the absence o f normal microbio ta . It has been postulated that SPI-1 independent col i t i s un ique ly requires both SPI-2 and aromatic amino ac id synthesis. Bac te r ia l a ck ing aro genes do not replicate as eff icient ly as w i l d -type i n the target organs o f infect ion i n mur ine typhoid (Hoise th and Stocker, 1981; Stocker et al, 1983). U n l i k e systemic infect ion, i n murine enterocolitis and other models o f intestinal Salmonellosis, aro mutants induce significant inf lammatory disease, al though w i t h observable differences f rom wi ld - type Salmonella induced in f lammat ion (Everest et al, 1999; Hapfe lmeier et al, 2005; Tso l i s et al, 1999a). O u r data are consistent w i t h these findings, as aro A mutants induce significant in f lammat ion that is not as severe as in f lammat ion induced b y wi ld - type bacteria. H o w e v e r , w e demonstrate here that both SPI-1 and SPI -2 mutants, not just SPI-1 mutants, require aromatic ac id biosynthesis to induce inf lammatory pa thology at day 5. The dependence o f in f lammat ion on aro A i n the absence o f SPI-1 is predictable, and consistent w i t h the need for SPI-2 for this fo rm o f col i t i s . That is , i f aroA and SPI-2 are both required for intracel lular surv iva l , and intracellular surv iva l is necessary for v i ru lence i n the absence o f S P I -1, then SPI-1 /SPI -2 and SVl-llaroA mutants should both be avirulent. H o w e v e r , i f aro A and SPI-2 perform o n l y over lapping functions, the absence o f aroA i n a SPI-2 mutant should not further attenuate the vi ru lence o f this strain. O u r observation that the SVI-2/aroA double mutant is to ta l ly attenuated i n intestinal inf lammatory disease suggests that SPl-llaroA m a y perform non-over lapping functions perhaps not exc lus ive ly l imi t ed to intracel lular stages o f infect ion. W h i l e it is clear that Salmonella must synthesize aromatic amino acids de novo w h i l e occupy ing the S C V in vivo, the phenotypic consequence o f impairments o f this process have o n l y been assessed indi rec t ly b y measur ing bacterial burden i n tissues. Consequent ly , several poss ibi l i t ies regarding the role o f aro A i n SPI-1 and SPI-2 independent typhl i t is are possible , i nc lud ing intracel lular repl icat ive defects, or the need for synthesized aromatic amino acids for other vi rulence behaviours. Despi te caveats associated w i t h these in vivo enteropathogenesis models , w e have used them to identify a nove l phenomenon i n intestinal Salmonella pathogenesis: SPI-1 independent col i t i s . The concordant inf lammatory responses o f mice and cows to infect ion w i t h SPI-1 deficient serovar T y p h i m u r i u m refute the prev ious ly postulated necessity for SPI-1 i n Salmonella enterocolit is. O u r data are consistent w i t h a new m o d e l o f intestinal pathogenesis i n w h i c h enterocoli t ic Salmonella infect ion does not depend exc lus ive ly o n S P T 1 , but depends at 71 various t imes o n both SPI-1 and SPI-2 and can occur even i n the absence o f the former vi rulence mechan i sm prev ious ly thought essential for intestinal Salmonella induced inf lammatory disease. 72 Chapter 5: Discussion Preface: The unpubl ished human diarrhea data i n this chapter (Figures 5.3-5.5 and Table 5.2) is a mod i f i ed vers ion o f the f o l l o w i n g manuscript i n preparation: C o b u r n B * . H u Q * , D e n g W , L i Y , S h i X , L a n Q , W a n g B , F i n l a y B B . Food Poisoning and diarrhea in humans caused by SPI-1 deficient Salmonella enterica. * Authors contributed equal ly to this w o r k Exper iments i n this chapter were performed b y the f o l l o w i n g people: I p rov ided the intel lectual guidance and experimental protocols and reagents for P C R o f SPI-1 and SPI-2 genes (Figure 5.4). C l i n i c a l characterization and treatment o f patients, P F G E and southern blots (Figures 5.3 and 5.5) were performed i n Shenzhen, G u a n g D o n g province , C h i n a b y the laboratory o f Q inghua H u , under her supervision. Invasion assays were performed b y m e i n col laborat ion w i t h W a n y i n D e n g and Q i n g h u a L i . Cons t ruc t ion o f SPI-2 mutant S. enterica serovar Senftenberg strains and mur ine infections o f these strains (data not shown) was performed b y me w i t h the technical assistance o f Y u l i n g L i . I wrote the manuscript and prepared it for submiss ion. 73 5.1 The non-dichotomous role of Salmonella Pathogenicity Islands 1 and 2 in Salmonella enterica serovar Typhimurium infection The theoretical f ramework o f in vivo Salmonella enterica infect ion evo lved under the selective pressure o f m o d e l systems w i t h specific theoretical and pract ical l imitat ions. U n t i l recently, mur ine infections reflected processes i n v o l v e d i n a systemic disease that recapitulates features o f human typhoid , i.e. intracel lular parasi t ism. Infections i n other systems i nc lud ing cultured c e l l l ines and l igated i l ea l loops were constrained b y the pract ical l imi ta t ion o f infections to those o f short duration, w h i l e oral infections o f cows and rabbits are expensive, cumbersome and lack the molecula r and genetic tools available to those w o r k i n g on mur ine models o f disease. The result ing experimental constraints resulted i n the product ion o f a synthesized mode l o f disease i n w h i c h two virulence factors c r i t i ca l for S. enterica pathogenesis, Salmonella pathogenici ty islands 1 and 2, were d i v i d e d into discrete, non-over lapping roles i n intestinal and systemic infect ion, respectively. W i t h the advent o f new approaches, i n part icular a m o d e l o f Salmonella enterocolit is i n mice , the opportunity was created to re-evaluate this theoretical m o d e l i n a system w i t h different constraints. The result ing discoveries contradicted the dichotomous v i e w o f Salmonella pathogenici ty and indicated instead that: SPI-2 is i n v o l v e d i n intestinal inf lammatory salmonel losis and; intestinal disease induced b y S. enterica can occur i n the absence o f SPI -1 . In the subsequent sections I w i l l b r ie f ly describe: (i) the n e w l y ident i f ied roles o f SPIs 1 and 2 i n Salmonella enteropathogenesis; (i i) the extension o f these f indings to human intestinal S. enterica infect ion and; ( i i i ) a new m o d e l o f in vivo pathogenesis ref lect ing these discoveries. I w i l l also speculate o n what roles SPI-1 and SPI-2 are p l a y i n g i n infect ion, describe the l imita t ions and caveats o f these data and their interpretation and postulate what interactions w i t h the host m a y be important for intestinal Salmonella infect ion. F i n a l l y , the impl ica t ions o f this w o r k for future investigations i n this area w i l l be discussed. 5.1.1 The murine enter colitis model of intestinal salmonellosis The descr ipt ion o f a new m o d e l o f inf lammatory intestinal S. enterica infect ion p rov ided an important new opportunity to reevaluate exis t ing theoretical models o f the mechanisms o f inf lammatory pathogenesis in vivo. U s i n g the in i t i a l characterization o f this m o d e l (Barthel et al, 2003) as a foundation, important features o f murine enteropathogenesis were ident i f ied and a 74 n o v e l scor ing method capable o f evaluating a broad range o f intestinal inf lammatory severity induced b y Salmonella in vivo was developed. The generation o f this scor ing method enabled a detailed descr ipt ion o f the parameters o f this infect ion inc lud ing the t imecourse o f infect ion and the dose-response o f m ice to serovar -Typhimur ium. The appl icat ion o f this m o d e l to the study o f the roles o f SPI-1 and SPI-2 i n serovar T y p h i m u r i u m enteropathogenesis has produced nove l insights into disease mechan i sm that had been p rev ious ly undescribed i n other models o f Salmonella intestinal infect ion. 5.1.2 SPI-2 in enterocolitic salmonellosis O u r descr ipt ion o f the role o f SPI-2 i n Salmonella enteropathogenesis can be summar ized as fo l lows : SPI-2 is c r i t i ca l for complete mur ine enterocolitis dur ing extracellular intestinal infect ion w i t h S. enterica serovar T y p h i m u r i u m and is essential dur ing invas ive intestinal disease i n the absence o f SPI -1 . The precise functions o f SPI-2 at these stages o f infect ion are not clear, but have specific features from w h i c h w e can infer mul t ip le roles. Ev idence presented i n this thesis (Chapter 3) and more thoroughly described elsewhere ( B r o w n et al, 2005) indicates that w i t h i n minutes o f co lon iza t ion o f the intestine, p r ior to epi thel ial invas ion , SPI-2 genes are expressed. Furthermore, after two days o f infect ion, SPI-2 is required for complete vi rulence dur ing a stage at w h i c h extracellular bacteria predominate. Importantly, at this stage o f infect ion, in f lammat ion is induced b y Salmonella on ly i n the presence o f SPI-1 (Table 5.1). Infection o f this type, dur ing w h i c h SPI-1 is essential for in f lammat ion , has been consistently reproduced i n a large number o f experimental systems in vitro and in vivo us ing Salmonella strains l a ck ing a variety o f SPI-1 v i ru lence factors (Barthel et al, 2003; C o b u r n et al, 2005; Coombes et al, 2005; Everest et al, 1999; S i l v a et al, 2004; Z h a n g et al, 2002a; Z h a n g et al, 2002b). Interestingly, w e have shown that SPI-2 is activated i n the l umen o f the gut, pr ior to invas ion dur ing mur ine infect ion. W h e n contemplat ing the potential roles o f SPI-2 at this stage o f infect ion, it is therefore important to consider the dominant influence o f SPI-1 and also the largely extracellular nature o f bacteria. Several SPI-2 mediated behaviours consistent w i t h these constraints m a y p l a y a role i n this stage o f disease (Figure 5.1). T h e y m a y be d i v i d e d into two general classes o f behaviour: 1) increasing the eff ic iency o f de l ive ry o f SPI-1 effectors to the intestinal epi thel ium, and 2) SPI-1 independent behaviours w h i c h increase the inf lammatory responsiveness o f the intestine or the s t imulat ing potential o f the bacteria. 75 Table 5.1 Features of early and late murine enterocolitis. E a r l y (1-2 days) Late (5 days) Murine Bovine Murine Bovine A S P I - 1 F u l l y Attenuated F u l l y Attenuated V i r u l e n t V i r u l e n t A S P I - 2 Par t i a l ly Par t ia l ly Attenuated Attenuated V i r u l e n t Par t i a l ly Attenuated A S P I - l / S P I - 2 F u l l y Attenuated F u l l y Attenuated F u l l y Attenuated F u l l y Attenuated Bac te r ia l L o c a l i z a t i o n Ext race l lu la r N / A Invasive N / A Systemic Disease Absen t Absen t Present Absen t Bac te r ia l resistance to an t imicrobia l peptides secreted into the intestinal l umen w o u l d increase contact between bacteria and the epi the l ium (and therefore de l ivery o f pro inf lammatory SPI-1 effectors). A role for the SPI-2 T 3 S S i n surv iva l o f p o l y m y x i n B and gentamic in in vitro has been reported ( D e i w i c k et al, 1998) and intest inal ly secreted peptides have been shown to have antibacterial act ivi ty against extracellular Salmonella ex vivo and the attaching and effacing non- invas ive mur ine pathogen Citrobacter rodentium in vivo (Ayabe et al, 2000; I imura et al, 2005). Intestinally-secreted peptides such as C R A M P and (3-defensins have both been shown to p l ay a role i n Salmonella infections i n other models (Rosenberger et al, 2004; Sa l zman et al, 2003). It w o u l d be interesting to assess the effect o f these peptides o n enteropathogenesis and their interaction w i t h SPI-2 i n this context (Figure 5.1 A ) . In addi t ion to a possible role i n su rv iva l o f an t imicrobia l defenses, D e i w i c k and colleagues ident i f ied a regulatory interaction between SPI-1 and SPI-2 ( D e i w i c k et al, 1998). T h e exquisite dependence o f early inf lammatory pathogenesis o n the SPI-1 T 3 S S m a y require some regulatory cross-talk between SPI-1 and SPI-2 . The eff ic iency o f SPI-1 ac t iv i ty i n the intestinal l umen m a y therefore par t ly depend o n SPI-2 ac t iv i ty (Figure 5 . I B ) . W e have shown that SPI-2 expression is induced pr ior to i nvas ion o f intestinal mucosa , and this induc t ion m a y be sufficient to enhance SPI-1 vi rulence. Since perturbation o f SPI-2 type III secretion has been shown to influence the expression o f SPI-2 genes (Coombes et al, 2004) this m a y affect downstream regulat ion o f SPI -1 and therefore pathogenesis. W h i l e it is possible that SPI-2 influences early intestinal pathogenesis b y potentiating S P I -1 act ivi ty, SPI-1 independent SPI-2 mediated vi rulence mechanisms m a y also be important for p rov id ing inf lammatory st imulus to the gut. Salmonella f l age l l in is a potent inducer o f host 76 in f lammat ion i n po la r i zed epi thel ia l monolayers (Zeng et ah, 2003), but on ly when del ivered to the basolateral surface o f the epi thel ium (Gewir tz et ah, 2001). F lage l la r Salmonella mutants are avirulent i n mur ine enterocolit is (Stecher et ah, 2004). Once del ivered there, Salmonella f lagel l in induces I L - 8 secretion v i a calcium-dependent N F K B act ivat ion b y st imulat ing basolateral T L R - 5 (Gewi r t z et ah, 2000; G e w i r t z et ah, 2001 ; Y u et ah, 2003; Z e n g et ah, 2006). The transcytosis o f f lage l l in occurs w i t h i n 15 minutes o f contact w i t h intestinal epi thel ium, and does not require bacterial internal izat ion (Gewi r t z et ah, 2001). Th i s potent proinf lammatory behaviour requires SPI-2 (Lyons et ah, 2004). Poss ib ly , i n mur ine enterocolitis, SPI-1 and SPI-2 del iver the proinf lammatory s t imul i S i p A / S i p B and f lagel l in independently but i n paral le l pr ior to epithelial invas ion (Figure 5 . 1 B , C ) . I f this is the case, SPI-2 deficient Salmonella w o u l d induce in f lammat ion o f equal severity i n T L R - 5 deficient mice . A. Secreted AMP Resistance i i B. Effecient SPI-1 activity s | - i SPI-1 Effectors (SipA, sopE etc) Host Cell Signals (IL-6, Caspase-1) C. Transcytosis of Flagellin D. Bacterial Translocation SM-2 Legend • i t Salmonella Flagellin Type 111 Secretion System — v Salmonella (40f*) Containing Vacuole Cytokine Release/ Cell Recruitment E. Intracellular Survival in Subepithelial Phagocytes Toll-like Receptor 5 Secreted Antimicrobial Peptides (AMP) Figure 5.1 Possible roles for Salmonella pathogenicity island-2 (SPI-2) type III secretion in early enterocolitis induced by Salmonella enterica serovar Typhimurium. (A) In vitro SPI-2 enhances bacterial resistance to soluble antimicrobial peptides. This may enhance contact of bacteria with the intestinal epithelium (yellow cells), where bacterial virulence mechanisms can induce pro-inflammatory host cell signaling. (B) SPI-2 regulatory cross-talk with SPI-1 may enhance the activity of SPI-1 effectors (SipA, SopB, SopE/E2) important for induction of host-cell inflammatory signaling. (C) Independently of bacterial internalization, Salmonella transcytoses flagellin to the basolateral surface of the intestinal epithelium, where it activates the proinflammatory pattern-recognition receptor TLR5. This process has been shown to require SPI-2. (D) SPI-2 is required for intracellular survival of Salmonella, and may be required for the efficient translocation of bacteria from the intestinal lumen to the subepithelium. Once translocated, subepithelial bacteria may interact with inflammatory cells/phagocytes (blue cell) in the lamina propria and stimulate innate immune responses leading to inflammation (E). 77 A l t h o u g h at two days post-infection, the vast majori ty o f Salmonella are extracellular, and SPI-2 is b road ly expressed b y these bacteria, the few potent ia l ly invad ing bacteria m a y exert a significant pro inf lammatory effect on the intestine. In the c o l o n o f s t reptomycin treated mice , a smal l but detectable number o f bacteria are gentamicin-protected at 2 days post-infection. In infect ion o f human intestinal explants smal l but detectable numbers o f internal ized bacteria are also detectable (Haque et al, 2004). U s i n g a sensitive reporter o f SPI-2 act ivat ion w e showed that the majori ty o f lumena l bacteria activate SPI-2 early i n infect ion ( B r o w n et al, 2005). U s i n g a robust assay that identif ied o n l y genes w i t h very h i g h levels o f expression, Ro l l enhagen and B u m a n n (Rol lenhagen and B u m a n n , 2006) demonstrated that a sma l l propor t ion o f bacteria induce h i g h levels o f SPI-2 expression dur ing mur ine enterocolitis b y day 2. T h e y hypothesize that this m a y represent a smal l propor t ion o f bacteria that have invaded into the intestinal mucosa . It is possible that this smal l popula t ion o f internal ized bacteria is responsible for the contr ibut ion o f SPI-2 to col i t i s . Such bacteria m a y induce specific inf lammatory responses from either the intracel lular compartment o f intestinal epi thel ial cel ls (Figure 5 . I D ) or f rom non-epi thel ia l cel ls such as macrophages, D C s or neutrophils from the subepi thel ium (Figure 5 . I E ) . In this case, S P T 2 expression i n the lumen m a y represent ' p r i m i n g ' p r ior to intestinal invas ion rather than functional expression i n the intestinal l umen itself. W h i l e early i n infect ion the function o f SPI-2 seems to be extracellular, this differs from later stages o f intestinal pathogenesis, dur ing w h i c h w e (Chapter 4) and others (Hapfelmeier et al, 2005) have demonstrated significant mucosa l invas ion w h i c h is at least par t ly SPI-2 dependent. Hapfe lmeier and colleagues demonstrated that Salmonella mutants l a c k i n g the SPI-2 t ranslocon component SseD d i d not eff iciently co lon ize C D 11-expressing cel ls i n the l amina propr ia late i n mur ine col i t i s . Furthermore, they demonstrated that SPI-2 dependent col i t is required intact T o l l - l i k e receptor s ignal ing, as infect ion o f m ice l ack ing the T L R adapter prote in M y D 8 8 was attenuated for SPI-2 dependent col i t is (Hapfelmeier et al, 2005). T h i s evidence suggests an intracel lular role for SPI-2 dur ing late enterocolit is, perhaps one more consistent w i t h its role i n systemic disease and intracel lular surv iva l . H o w e v e r , it is important to note that the use o f a SPI-2 mutant incapable o f translocation - the inject ion o f effectors into host cel ls -rather than an apparatus mutant incapable o f any secretion whatsoever, m a y bias the resul t ing mutant phenotype towards processes that require effector translocation such as intracel lular 78 surv iva l . In total, however , these data suggest discrete roles for SPI-2 dur ing early and late intestinal infect ion i n v i v o (Table 5.1). A s the role for SPI-2 i n Salmonella enteropathogenesis emerges, it is o f considerable importance to explore h o w this v i rulence system imparts its function i n mul t ip le models o f Salmonella enterocolitis us ing mul t ip le SPI-2 mutant strains. The exploi ta t ion o f mouse strains w i t h specific genetic defects i n an t imicrobia l peptide secretion, T L R 5 and T L R signal ing, or l ack ing specific ce l l types (e.g. C D 1 8 / C D 1 1 knockouts) provide an opportunity to explore the potential v i rulence contributions o f SPI-2 to enterocolitis proposed here. 5.1.3 SPI-1 independent colitis In the in t roduct ion to this thesis, in vitro evidence that mul t ip le pro inf lammatory pathways are i n v o l v e d i n the infect ion o f intestinal epithelia, some independent o f SPI-1 was described (Gewi r t z et al, 1999; L e e et al, 2000; M c C o r m i c k et al, 1998; Z e n g et al, 2003). O u r data f rom murine and bovine infections substantiate this observation in vivo. W e have shown unequ ivoca l ly that i n the absence o f a functional SPI-1 encoded T 3 S S , Salmonella are able to induce significant intestinal inf lammatory disease. W e and others (Hapfelmeier et al, 2005) have also shown that SPI-1 independent disease requires functional SPI-2 type III secretion and translocation and l i k e l y occurs dur ing intracel lular infect ion. W e have also shown that both S P I -1 and SPI-2 independent typhl i t is requires the presence o f the aromatic amino ac id biosynthesis gene aroA. That mul t ip le pathways o f intestinal in f lammat ion i n intestinal epi the l ium independent o f SPI-1 and invas ion exist is inconsistent w i t h the p rev ious ly postulated necessity for invas ion and SPI-1 i n S. enterica enteropathogenesis. W h i l e SPI-2 dependent in f lammat ion seems to require intact T L R s ignal ing, this is not true for SPI-1 dependent inf lammat ion . Salmonella mutants w i t h an intact SPI-1 T 3 S S but a non-functional SPI-2 T 3 S S induce in f lammat ion irrespective o f the presence o f the T L R adapter M y D 8 8 and these bacteria appear to reside w i t h i n the intestinal surface epi thel ium late i n infect ion (Hapfelmeier et al, 2005). These data suggest that independent processes are i n v o l v e d i n SPI-1 and SPI-2 induced inf lammat ion late i n infect ion. A clear pattern to emerge f rom these in vivo experiments is the presence o f two virulence pathways that result i n the induc t ion o f in f lammat ion i n the gut. The host pathways i n v o l v e d i n this induc t ion are not clear, however , in vitro and in vivo evidence provides two interesting candidates: the act ivat ion o f caspase-1 and 79 I L - i p / i L - 1 8 , and; the act ivat ion o f M y D 8 8 dependent T L R - s i g n a l i n g resul t ing i n N F K B act ivat ion. T h e interaction o f the SPI-1 effector S i p B and caspase-1 is w e l l described in vitro and i n mur ine typhoid . Translocated S i p B activates caspase-1 and I L - i p / I L - 1 8 release and causes rap id ce l l death i n macrophages that has features o f both apoptosis and necrosis ( C h e n et al, 1996; H e r s h et al, 1999). Th is act ivat ion is cr i t ica l for mur ine typhoid pathogenesis, as caspase-1 knockout m ice have a substantially increased L D 5 0 (>1,000 fold) compared to wi ld - type mice ( M o n a c k et al, 2000). Salmonella induced caspase-1 act ivat ion requires the intracellular adapters A S C and Ipaf, but is independent o f T L R s ignal ing (Figure 5.2, (Mariathasan et al, 2004)). A l t h o u g h a potent inducer o f systemic in f lammat ion and sepsis, the role o f caspase-1 activation i n Salmonella induced intestinal inf lammatory disease has not been investigated. T h i s i n part reflects the d i f f icul ty i n separating two important v i rulence functions o f S i p B ; act ivat ion o f caspase-1 and the translocation o f other SPI-1 effectors. S ince S i p B mutants cannot activate caspase-1 nor translocate other pro inf lammatory SPI-1 effectors (e.g. S i p A ) , the unique role o f S i p B i n inf lammatory pathogenesis has p roven diff icul t to study. In one study o f early enteropathogenesis i n cows , a S o p B mutant was shown to cause m i l d l y attenuated f lu id secretion and in f lammat ion but the same amount o f ce l l death as measured b y terminal deoxynucleotidyltransferase-mediated d U T P - b i o t i n n i ck end labe l ing ( T U N E L ) . Howeve r , the S o p B mutant was o n l y m i l d l y attenuated and intact SipB/caspase-1 ac t iv i ty m a y have resulted i n the significant in f lammat ion s t i l l present dur ing this infect ion (Santos et al, 2001a). Furthermore, o n l y early t imepoints were assessed i n these experiments, and our data indicate that early and late enteropathogenesis m a y i n v o l v e d mul t ip le and distinct disease mechanisms. A l t h o u g h caspase-1 act ivat ion has not been di rect ly assessed i n models o f inf lammatory intestinal sa lmonel losis , it has been shown that chemica l ly - induced col i t i s i n caspase-1 knockout m ice is reduced i n in f lammat ion o f short duration, and near ly absent i n extended inf lammatory disease (S iegmund et al, 2001). Furthermore, the increased L D 5 n o f Salmonella i n caspase-1 knockout m i c e is evident f o l l o w i n g ora l but not i .p. infect ion w i t h wi ld - type bacteria, indica t ing that caspase-1 mediated processes are i n v o l v e d i n the intestinal phase o f mur ine salmonel losis ( M o n a c k et al, 2000). The act ivat ion o f IL -1 p b y caspase-1 m a y therefore be an important pro inf lammatory process i n Salmonella-induced enterocolit is. O f part icular interest is the rel iance o f caspase-1 act ivat ion on the SPI-1 effector S i p B , but its independence f rom T L R 80 s ignal ing. Caspase-1 act ivat ion is an interesting candidate pathway for SPI-1 dependent, SPI-2 independent intestinal inf lammat ion . SipB E f f e c t o r s Adapte r (MyDSS) Ipaf R I C K A c t i v a t o r ( I R A K ) 1 A S C I Caspase-1 I K K s , M A P K s IL1P, IL18 N F K B — * TNFa Figure 5.2 Distinct host proinflammatory pathways are potentially influenced by Salmonella pathogenicity islands 1 and 2 (SPIs 1 and 2) to induce inflammation. The activation of the caspase-1 "inflammasome" (several caspase-1 containing protein complexes capable of activating ILip by cleavage) by the Salmonella enterica SPI-1 effector SipB results in the release of proinflammatory ctokines (ILlp7IL18) independently of toll-like receptor (TLR), Nod-like receptor (NLR) or N F K B activation. Whereas, SPI-2 has been shown to require MyD88 to induce intestinal inflammation in the absence of SPI-1, SPI-1 does not require MyD88. In addition to inflammasome and T L R activation, SPI-1 effectors can directly induce N F K B via M A P / E R K kinase cascade activation, and intracellular pathogen-associated molecular patterns (PAMPs) can activate both inflammasome and N F K B by N L R activation of various pathway intermediates. Potentially, this provides host-signaling cascades that could be independently activated by SPI-1 and SPI-2 virulence mechanisms. U n l i k e act ivat ion o f T L R s , caspase-1 act ivat ion does not direct ly result i n the translocation o f N F K B to the nucleus and act ivat ion o f NFtcB-regula ted genes (Figure 5-2). In contrast to act ivat ion o f caspase-1, SPI-1 does not appear to be required for M y D 8 8 mediated inf lammat ion. Rather, SPI-2 dependent col i t is requires this T L R adapter molecu le (Hapfelmeier et al, 2005). A l t h o u g h it is not clear whether T L R act ivat ion requires intracel lular parasi t ism b y Salmonella, the presence o f SPI-2 increases the intestinal pathogenici ty o f serovar T y p h i m u r i u m i n the presence o f M y D 8 8 and this vi rulence mechan i sm is associated w i t h intracellular surv iva l . S igna l ing downstream o f T L R - M y D 8 8 act ivat ion results i n N F K B translocation to the nucleus 81 and induct ion o f potent inf lammatory regulators such as T N F a (Figure 5-2). A l t h o u g h convergent o n the same f inal outcome ( inflammation) act ivat ion o f T L R s and caspase-1 are largely independent and thus represent potential targets for SPI-2 and SPI-1 mediated in f lammat ion respectively. No tab ly , w e found that both SPI-1 and SPI-2 dependent intestinal in f lammat ion required biosynthesis o f aromatic amino acids. Whether requirement for aroA indicates a dependence o f pathogenesis on intracel lular repl icat ion is debatable, however , i f this is the case, inf lammatory behaviours that are induced b y bacteria must be sensitive to intracel lular pathogens. In addi t ion to the caspase-1 and T L R proinf lammatory pathways, a n e w l y described intracel lular innate immune pathway has been described. A f a m i l y o f intracel lular pattern-recognition receptors ( P R R s ) that respond to pathogen-associated molecula r patterns ( P A M P s ) has been identif ied. The N O D - l i k e receptor f ami ly ( N L R s ) behaves analogously to T L R receptors, but for intracellular agonists (Inohara et al, 2004; Inohara and N u n e z , 2003). Interestingly, N L R s that activate either the caspase-1 pathway or N F K B v i a M A P kinases have been identif ied (Figure 5.2). Sa lmone l l a res iding w i t h i n intracel lular compartments cou ld potent ia l ly activate either o f these pathways us ing different v i ru lence mechanisms. The relative ease o f creating knockout mouse l ines, the existence o f mouse strains w i t h mutations i n m a n y o f the pathways described and the ava i lab i l i ty o f Salmonella strains l a ck ing various SPI-1 and SPI-2 components a l lows the invest igat ion o f the interaction o f the inf lammatory pathways described here w i t h bacterial v i rulence mechanisms. T h i s pathway is a potential target o f SPI-1 dependent, SPI-2 independent intestinal inf lammatory pathways. The mur ine Salmonella enterocolitis infect ion m o d e l can be easi ly exploi ted to investigate the interactions o f SPI-1 and SPI-2 w i t h these important inf lammatory pathways in vivo. 5.1.4 Caveats of experimental murine enterocolitis The an imal models o f disease used i n our and previous experiments have significant l imita t ions and interpretation o f data ar is ing f rom their use must be careful ly considered when app ly ing findings o f the models to human disease. F o r example, mur ine infections are encumbered b y concurrent systemic ' t ypho id - l i ke ' disease ( in susceptible mice) w h i l e bov ine infections are induced b y inocula t ing l igated intestinal segments w i t h large numbers o f bacteria, or o ra l ly inocula t ing animals w i t h large doses o f laboratory g r o w n strains. Furthermore, al though disease i n these animals has s imi la r i ty to human disease, differences i n host phys io logy 82 and infect ion suscept ibi l i ty result i n host-specific disease processes that cannot a lways be compared to human disease pathogenesis. F i n a l l y , w h i l e laboratory-created S. enterica strains l ack ing a functional SPI-1 T 3 S S m a y induce disease ar t i f ic ia l ly and SPI-1 deficient S. enterica has been isolated from environmental reservoirs (Rahn et al, 1992), it does not necessari ly f o l l o w that SPI-1 deficient strains cause disease in* a natural ly occur r ing human infect ion. The SPI-1 and SPI-2 T 3 S S mutants used i n our studies lacked o n l y a cr i t ica l component o f the T 3 S S and consequently lack secretion. H o w e v e r , the pathogenici ty islands themselves, i nc lud ing regulatory components and a fu l l complement o f effectors, were otherwise intact. It is possible that remain ing elements o f the pathogenici ty islands inf luenced bacter ial p h y s i o l o g y i n the absence o f type III secretion. A recent col laborat ive project has p rov ided an opportunity to test the va l id i t y o f our an imal models i n human infect ion w i t h natural ly occurr ing S. enterica strains. T h i s project resulted i n the ident if icat ion o f a human diarrheagenic serovar o f S. enterica w h i c h lacks SPI -1 . In this study, w e have isolated S. enterica serovar Senftenberg f rom human diarrhea patients infected i n September, 2002, i n Shenzhen, Guangdong province , C h i n a . W e show that these strains lack SPI-1 yet cause c l i n i c a l l y significant disease, demonstrating both that SPI-1 independent enteropathogenesis is an important human phenomenon and that the currently p reva i l ing m o d e l o f intestinal sa lmonel losis , founded upon a requirement for S P T 1 for intestinal pathogenesis, is incorrect. 5.1.5 Human Diarrheal disease can be induced by SPI-1 deficient S. enter ica Bacte r ia l cultures o f S. enterica serovar Senftenberg were isolated from an outbreak o f diarrhea i n Shenzhen, C h i n a . W i t h i n 12 hours o f sharing a mea l o f beef, ch icken , mushrooms and cucumber on Sep 10, 2002, 7 immunocompetent ind iv idua ls (age range 26-62) were str icken w i t h symptoms ranging from abdominal c ramping and diarrhea to severe diarrhea, nausea, v o m i t i n g , fever and dehydrat ion requi r ing hospi ta l izat ion. A l l patients were treated w i t h a m p i c i l l i n , after w h i c h they reported the resolut ion o f their symptoms. O f the four stool samples provided , serovar Senftenberg was ident i f ied i n samples from two patients affected w i t h systemic symptoms (nausea, v o m i t i n g , fever and dehydration). Isolates o f S. enterica serovar Senftenberg were i n i t i a l l y identif ied us ing standard serological and b iochemica l testing methods. T h e y were then compared to a series o f reposi tory serovar Senftenberg strains b y pulsed-f ie ld gel electrophoresis (Figure 5.3). The strains isolated 83 from the two patients had ident ical genotypes that were d i ss imi la r to other Senftenberg strains tested. o '.N LL II I III I ll | i l l l II II 11 m a I I I I I ! I I i IHII M I i i n i i i i n in i HI I I I III H I I CM CI2 SC1 SC2 SC3 SC4 SC5 S T M STy Figure 5.3 Pulsed field gel electrophoresis of stool isolates from an outbreak of diarrhea in Shenzhen, China. Bacterial D N A from stool samples from diarrhea patients from a single outbreak (CI1, CI2), or control strains of Salmonella enterica serovar Senftenberg (SC1-4) and Salmonella enterica serovar Typhimurium (STM) were digested with restriction the enzyme Xbal and subject to pulsed-field gel electrophoresis as described in Materials and Methods and compared to the reference Salmonella strain X9844 (Sty). The stool isolates have an identical genotype, distinct from repository strains of S. enterica serovar Senftenberg. A s part o f an ongo ing project intended to develop a molecula r method for typ ing Salmonella species, our collaborators had been assessing outbreak strains for the h i g h l y conserved SPI-1 locus by P C R for the apparatus gene encoding I n v A , a component cr i t ica l for SPI-1 T 3 S S function and encoded w i t h i n the SPI-1 is land. No tab ly , the c l i n i c a l isolates o f serovar Senftenberg from the Shenzhen outbreak, but not control strains, were negative for P C R o f this gene (Figure 5.4). Furthermore, the Shenzhen strains, but not controls, apparently lacked the SPI-1 gene encoding the secreted effector S i p A . A l l strains i nc lud ing the Shenzhen isolates and controls were P C R - p o s i t i v e for the phage-encoded SPI-1 effector S o p E , thought to have been acquired independently o f the SPI-1 genomic is land (Ehrbar et al, 2003). W e also assessed b y P C R the presence o f the SPI-2 pathogenici ty i s land gene ssaR and the non-SPI-2 encoded, SPI-2 secreted effector sifA gene, located i n the potABCD operon (Figure 5.4). It is thought that the SPI-2 i s land was acquired b y S. enterica species subsequent to its acquis i t ion o f SPI-1 ( O c h m a n and G r o i s m a n , 1996). T o ensure that the strains d i d not mere ly conta in sequence var ia t ion i n the P C R pr imer b ind ing sites o f the invA and sipA genes, w e assessed the genome o f the strains for these genes 84 b y Southern blot hybr id iza t ion . In the Shenzhen isolates, but not the control strains, sipA and invA, the latter assessed b y two different probes, were negative (Figure 5.5). These data conc lus ive ly demonstrate that SPI-1 is absent from these strains. Y e t , despite the loss o f SPI-1 at some point dur ing the evolu t ion o f the Shenzhen strains, they were able to cause c l i n i c a l l y significant intestinal disease i n humans. These f inding refute the w i d e l y he ld v i e w that SPI-1 is essential for the development o f gastrointestinal disease dur ing S. enterica infect ion. The poss ib i l i ty remained, however , that despite the absence o f the SPI-1 encoded virulence machinery, the Shenzhen strains retain or have complemented their ab i l i ty to effect SPI-1 induced invas ion . Figure 5.4 Genetic analysis of Salmonella pathogenicity island (SPI) -1 and -2 of strains isolated from diarrheal patients in Shenzhen, China compared to control strains. Stool isolates (CI1, CI2) were compared to S. enterica serovar Senftenberg (Senftenberg-1, Senftenberg-2), serovar Enteritidis (Enteritidis) and serovar Typhimurium (STM1344, STMinvA-) strains by polymerase chain reaction (PCR) for the presence of the SPI-1 genes invA and sipA, the phage encoded SPI-1 effector gene sopE, the SPI-2 encoded gene ssaR and the non-SPI-2 encoded SPI-2 effector gene sif A. SPI-1 encoded genes, but not those encoded outside SPI-1 were PCR-negative in both stool isolates and the negative control strain STMinvA- which contains an insertional mutation. One strain (Senftenberg-2) of S. enterica serovar Senftenberg lacked the non-SPI-2 encoded SPI-2 effector sif A. In order to ascertain whether these strains had complemented the invas ion associated function o f SPI-1 and were effecting ' S P I - 1 - l i k e ' invas ion b y some other mechanism, w e assessed the capaci ty for these strains to invade cultured human epithelial c e l l l ines ( H e L a cel ls) . Compared to control strains, SPI-1 deficient strains were s ignif icant ly attenuated for c e l l invas ion (Table 5.2). A l t h o u g h control SPI-1 posi t ive Senftenberg strains d i d not invade cel ls as efficiently as laboratory passaged S. enterica serovar T y p h i m u r i u m controls, they were 85 s ignif icant ly more efficient than both the SPI-1 deficient Shenzhen strains and the ar t i f ic ia l ly created invA mutant serovar T y p h i m u r i u m control strain. Th i s observation suggests that, i n the absence o f S P I - 1 , the Shenzhen isolates are attenuated for the active invas ion process w h i c h is the ha l lmark o f SPI-1 mediated v i ru lence in vitro. The loss o f SPI-1 b y these strains w o u l d therefore seem to represent a genuine loss o f SPI-1 associated vi rulence function. The abi l i ty o f these strains to cause disease therefore relies on vi rulence mechanisms and pathogenic behaviors independent o f SPI -1 , further conf i rming that models o f enteropathogenesis i n S. enterica infect ion r e ly ing on SPI-1 induced pathology incomple te ly describe intestinal salmonellosis . invA 1 i n v A 2 sipA Figure 5.5 Southern hybridization analysis for SPI-1 genes of diarrheal stool isolates from Shenzhen, China and controls. By Southern hybridization, the presence of the SPI-1 genes invA and sip A were assessed in stool isolated (CI1, CI2) and control strains of S. enterica serovars Senftenberg (Senftenberg-1, Senftenberg-2) and Typhimurium (STM, STMinvA-). The presence of invA was assessed using two independent probes. SPI-1 genes were absent only in the stool isolates. N.B. STMinvA- contains an insertion with inv A and is consequently larger than wild-type inv A. W h i l e the role o f SPI-1 is c lear ly cr i t ica l i n early intestinal pathogenesis i n murine and bov ine enterocolitis, it does not seem to be so for human S. enterica induced diarrhea. The k e y difference i n these infections m a y be the absence o f a functional T 3 S S i n murine and bovine infections versus the complete absence o f the SPI-1 i s land i n human infections. In murine infections w i t h complete SPI-1 deletions, M u r r a y and L e e demonstrated that the absence o f regulatory (hilA) or T 3 S S components (invG) resulted i n decreased co lon iza t ion o f the mouse gut dur ing oral infect ion, whereas the absence o f the entire pathogenici ty i s land d i d not ( M u r r a y and L e e , 2000). Based o n this evidence, M u r r a y and L e e proposed that SPI-1 encoded proteins el ici t a host response and that other SPI-1 encoded factors are required for the defense against them. One candidate interaction is the resistance to the ant imicrobia l ac t iv i ty o f neutrophils that have transmigrated across the intestinal epi thel ium into the gut lumen. Neu t roph i l recruitment is a ha l lmark o f human Salmonella enterocolitis ( B o y d , 1969; B o y d , 1985; M c g o v e r n and Slavut in , 86 1979). It has been shown that f o l l o w i n g SPI-1 dependent bacter ia l ly induced transmigrat ion and act ivat ion across intestinal epithelia, neutrophils exhibi t increased k i l l i n g o f serovar T y p h i m u r i u m (Nadeau et al, 2002). T h e evas ion o f extracellular neutrophi l k i l l i n g is greatly enhanced b y the SPI-1 dependent invasiveness o f serovar T y p h i m u r i u m and these constitute a significant antibacterial mechan i sm o f these activated neutrophils. Table 5.2 Characteristics of human diarrheal strains and controls. Strain Salmonella enterica serovar Source SPI-1 SPI-2 Invasiveness Index* Notes CI 1 Senftenberg Clinical Isolate Negative Positive 0.00045 Stool sample isolate, Shenzhen outbreak CI 2 Senftenberg Clinical Isolate Negative Positive 0.006 Stool sample isolate, Shenzhen outbreak Control Senftenberg Repository Positive Positive 0.056 Control strain STM Typhimurium Lab Strain Positive Positive 1.00 Invasion Positive Control SL1344 STMinvA- Typhimurium Lab Strain Insertional Positive 0.0026 Invasion Negative Control mutant (invA) "Invasion/Invasion of STM SL1344 The difference i n t ime o f onset to SPI-1 independent col i t i s i n humans and experimental animals m a y represent differences i n either the nature o f the SPI-1 muta t ion or strain-specific pathogenici ty (a testable hypothesis) , or host phys io logy . T h i s remains an area o f cr i t ica l importance for future invest igat ion. W h a t is not disputable is that the occurrence o f SPI-1 independent intestinal disease induced b y Salmonella is not conf ined to art i f ic ial experimental systems, but represents a c l i n i c a l l y important human disease phenomenon. 5.2 A synthesized, non-dichotomous model of Salmonella enteropathogenesis B a s e d o n the data presented i n this thesis, as w e l l as concurrent ly publ i shed evidence from Hapfe lmeier and colleagues (Hapfelmeier et al, 2005), I propose a new m o d e l o f the roles o f SPI-1 and SPI-2 i n intestinal sa lmonel losis (Figure 5.6). Rather than discrete, non-over lapping roles in vivo, SPIs 1 and 2 are i n v o l v e d i n a complex array o f interactions w i t h the host result ing i n intestinal in f lammat ion and systemic disease i n mice . I contend that these roles, w h i l e perhaps i n v o l v i n g over lapping processes, differ i n part from those p rev ious ly described i n murine typho id pathogenesis. The impl ica t ions o f these findings for both the study o f S. enterica 87 pathogenesis and the strategic approach to detection and treatment o f Salmonella enterocolitis are significant. E/C SPI-2 function? MyD88 activation SPI1 independent Inflammation Early Late Figure 5.6 A revised view of SPI-l/SPI-2 contributions to intestinal Salmonella infection, in which both SPI-1 and SPI-2 play critical and overlapping roles in the induction of disease in the intestine and where SPI2 activation and involvement in disease has both early, extracellular and late, intracellular phases. A role for SPI-2 i n early and late pathogenesis has n o w been established, the latter independent o f SPI-1 and dependent on host s ignal ing v i a the T o l l - l i k e receptor adapter M y D 8 8 . A l t h o u g h a specific extracellular function o f SPI-2 has not been established, SPI-2 is activated pr ior to invas ion and is important dur ing a phase o f infect ion at w h i c h bacteria are predominant ly lumenal , and w h i l e bacterial v i rulence mechanisms important for early intestinal in f lammat ion have been described, the ro le o f SPI-1 late i n infect ion is not clear. 88 5.3 Implications for research: Salmonella, inflammation and the host O u r data demonstrate that, w i t h i n its arsenal, Salmonella contains mul t ip le v i ru lence mechanisms that, w h e n activated, result i n the induc t ion o f an inf lammatory response w i t h i n the host. Inf lammation i n any host is a heterogeneous process and is the cu lmina t ion o f the act ivat ion o f numerous complex and interacting pro inf lammatory cascades that are co l l ec t ive ly inf luenced b y host and pathogenic behaviours . A double-edged sword , in f lammat ion is the strategy b y w h i c h the host controls infect ion, the Trojan horse b y w h i c h some pathogens gain influence over host phys io logy and ul t imate ly the cause o f death for either the pathogen or host i n a l l acute infections. Whether in f lammat ion favours the host or the bacteria m a y depend not o n l y o n the severity but also the nature o f the inf lammatory response. O f specific interest w i t h regards to Salmonella intestinal inf lammatory pathogenesis is h o w SPI-1 and SPI-2 influence this important phenomenon to the detriment o f either the host or bacteria. In mur ine typhoid , SPI-1 and SPI-2 influence specific host responses to exploi t them to the advantage o f bacteria. Via the SPI-1 effector S i p B , serovar T y p h i m u r i u m induces rapid caspase-1 mediated c e l l death i n macrophages and the induc t ion o f IL1 (3 and I L 1 8 , potent pro inf lammatory cytokines . The induc t ion o f caspase-1 b y Salmonella is essential to the pathogen and deleterious to the host. E m p l o y i n g SPI -2 , Salmonella influence a variety o f i m m u n e responses i n a manner d i ss imi la r to those inf luenced b y SPI -1 . Rather than a rapid induc t ion o f apoptosis, Salmonella induce a delayed apoptosis i n macrophages i n a SPI-2 dependent manner (van der V e l d e n et al., 2000). In addi t ion, SPI -2 dependent evasion o f phagolysosomal maturat ion inhibi ts bacterial antigen presentation and T L R mediated nuclear factor kappa B ( N F K B ) act ivat ion (Cheminay et al, 2005; R o y l e et al, 2003). U s i n g these or other mechanisms, it m a y be that SPI -1 , SPI-2 or both are required to tip the host-pathogen balance i n favour o f bacteria. Further e lucidat ion o f the bacterial v i ru lence mechanisms and host response pathways i n v o l v e d i n the induct ion o f intestinal inf lammatory disease should i nvo lve the considerat ion o f both SPI-1 and SPI-2 dependent process. 5.4 Directions for intestinal salmonellosis research The ident if icat ion o f nove l intestinal phenotypes associated w i t h SPI-1 and SPI-2 T 3 S S mutants requires significant explorat ion o f w h i c h two broad themes are par t icular ly important: (i) Bac te r ia l factors required for enterocolitis ( i i ) Hos t pathways i n v o l v e d i n disease promot ion . 89 The first theme can be addressed b y genet ical ly manipula t ing bacteria, the latter b y assessing and perturbing specific host responses. In our studies, w e spec i f ica l ly targeted the disrupt ion o f the T 3 S S . W h i l e this approach has the advantage o f interfering w i t h numerous convergent v i rulence processes, it does not identify specific v i ru lence factors and is unsuitable for the study o f host-pathogen b iochemica l interactions. D i s rup t ion o f bacterial type III effectors alone or i n combina t ion has been a product ive study i n other models o f Salmonella pathogenesis both in vivo and in vitro, however , this is often accompl ished b y mutat ing effectors o f one pathogenici ty i s land i n the presence o f other v i ru lence elements, for example, assessing the phenotypes o f SPI-1 effector mutants i n the presence o f a complete and functional SPI-2 T 3 S S . O u r f indings indicate that, for future studies o f intestinal pathogenesis, it is important to assess the phenotypes o f SPI-1 and SPI-2 effectors i n the absence o f the complementary v i ru lence determinants (i.e. on SPI -2 and SPI-1 deficient backgrounds respect ively) . Several SPI-1 and SPI-2 effectors have been characterized i n other m o d e l systems and the contr ibut ion o f their functions to SPI-1 and SPI-2 independent intestinal disease can be assessed us ing the murine enterocolitis m o d e l as an in vivo screening tool . The interaction o f bacter ial v i rulence mechanisms and specific host processes can also be assessed b y the examinat ion and characterization o f host responses to bacteria l a c k i n g functional SPI-1 and SPI-2 T 3 S S or both. G e n o m i c and proteomic approaches such as tissue-specific microarrays or cytokine profiles can be exploi ted to broad ly characterize differences i n the host response to SPI-1 and SPI-2 deficient bacteria and to identify potential n o v e l pathways i n v o l v e d i n disease pathogenesis. H o w e v e r , exis t ing data f rom other models o f disease pathogenesis provide ample clues and in i t i a l targets. O f part icular interest are those host pathways i n v o l v e d i n SPI-1 and SPI-2 dependent inf lammatory induct ion. Bac te r i a l l y induced inf lammatory responses l i k e l y i nvo lve mul t ip le convergent pathways. B y e l imina t ing SPI-1 and SPI -2 mediated vi ru lence mechanisms, it m a y be possible to investigate the contr ibut ion o f single inf lammatory pathways to disease. F o r example, an assessment o f SPI-2 mediated act ivat ion o f T L R or N L R s ignal ing can be performed us ing SPI-1 deficient bacteria, theoret ical ly prevent ing the confounding act ivat ion o f caspase- 1 / I L i p . S i m i l a r l y , caspase-1 dependent inf lammatory processes can be investigated i n the absence o f immune act ivat ion b y SPI -2 dependent T L R or N L R induct ion. W h i l e these investigations w i l l elucidate mechanisms and pathways i n v o l v e d i n disease pathogenesis, it is also interesting to consider the net effect o f these processes; the outcome o f infect ion. 90 In order to investigate h o w perturbations o f the host inf lammatory response b y bacteria influence outcomes o f intestinal disease, infect ion must be a l l owed to continue i n the absence o f systemic disease. Th i s can be accompl i shed b y either al tering the bacteria suff icient ly to prevent systemic i l lness , or us ing genet ical ly resistant hosts. Because the use o f bacterial mutants m a y affect bo th systemic and intestinal vi rulence, as has been described i n our and other studies (Coburn et al, 2005; Everest et al, 1999), it is preferable to emp loy the latter strategy. The use o f Nramplresistant m ice provides an opportunity to do so. Infection o f Nramplresistant 129Sv mi c e results i n intestinal in f lammat ion as significant as that w h i c h occurs i n susceptible mice , but mice survive infect ion. The assessment o f bacterial intestinal v i ru lence i n infections o f resistant m ice us ing SPI-1 and SPI-2 mutants m a y prov ide specific evidence indica t ing the relative importance o f SPI-1 and SPI-2 dependent processes i n disease progression. Outcome measures such as the duration, severity and resolut ion o f intestinal in f lammat ion as w e l l as bacterial co lon iza t ion and persistence should be considered. The nove l insights into SPI-1 and SPI-2 dependent intestinal pathogenesis described i n this thesis are important for our understanding and invest igat ion o f Salmonella pathogenesis i n experimental systems and also consequential for future approaches to the detection, diagnosis and treatment o f this important infect ion. The exposure o f ind iv idua l s to SPI-1 deficient S. enterica strains is l i k e l y to be a significant p rob lem, par t icular ly i n areas o f h i g h popula t ion density, poor sanitation, and close p r o x i m i t y o f l ives tock and human habitation. The presence o f SPI-1 deficient strains o f S. enterica serovars i n var ious environmental reservoirs i n c l u d i n g l ives tock feed and marine environments have been reported i n the literature (Rahn et al, 1992). The use o f SPI-1 l o c i as targets for the molecu la r detection o f Salmonella enterica species has been the subject o f intense and cont inuing invest igat ion and is b road ly appl ied ( A r n o l d et al, 2004; B u l t e and Jakob, 1995; C h i u and O u , 1996; C o c o l i n et al, 1998a; C o c o l i n et al, 1998b; M a l o r n y et al, 2003; Pathmanathan et al, 2003; Perel le et al, 2004; Scho lz et al, 2001). D e v e l o p i n g therapies directed towards the molecular targets encoded w i t h i n SPI-1 has also been proposed (Patel et al, 2005). T h e use o f SPI-1 as a detectable marker and therapeutics target has arisen due to the b e l i e f that SPI-1 is both h i g h l y conserved across Salmonella species due to its early phylogenet ic acquis i t ion (Ochman and Gro i sman , 1996) and the perceived requirement o f Salmonella for S P I -1 dur ing pathogenesis. The ident i f icat ion o f SPI-1 independent intestinal disease i n mice , cows and humans requires us to reevaluate these approaches. N 91 The research presented in this thesis identifies previously undescribed pathogenic processes critical for experimental and clinical intestinal salmonellosis. The discovery of new roles for SPI-1 and SPI-2 i n intestinal pathogenesis has resulted i n a new, non-dichotomous model of in vivo salmonellosis and represents a significant new approach to the study and understanding of S. enterica infection. 92 Chapter 6: Methods 6.1 Methods used throughout thesis 6.1.1 Bacterial culture Salmonella typhimurium S L 1 3 4 4 (wild- type, W T ) , i n v A mutant ( i n v A : : k a n S B 1 0 3 , ' S P I - 1 ' ) and ssaR mutant (AssaR, ' S P I 2 ' , ( B r u m e l l et al, 2001)) were g rown overnight shaking (200rpm) i n 3 m L Lur i a -Ber t an i broth ( L B ) w i t h 5 0 u g / m L streptomycin +/- 5 0 u g / m L k a n a m y c i n at 3 7 ° C for 18 hours. 6.1.2 Murine enterocolitis (Streptomycin pre-treatment) infections Inbred C57B16 (Jackson Laboratories, B a r Harbor , M a i n e , U S A ) or congenic 129SvJ Nrampl knockout mice (Bred at W e s b r o o k A n i m a l U n i t , U B C ) were depr ived o f food and water for 4 hours pr ior to administrat ion o f 20mg/mouse o f s t reptomycin b y oral gavage. T w o hours subsequently food and water were p rov ided ad libitum. T w e n t y hours after oral s t reptomycin treatment food and water were once again w i thd rawn for 4 hours after w h i c h bacteria i n 3x 10 2 -3><108 l O O u L L B broth or sterile saline were administered b y oral gavage. C o n t r o l m i c e were g iven l O O u L sterile L B broth or saline. Water and food were p rov ided ad libitum. M i c e were euthanized w i t h C O 2 at designated t imepoints and tissues harvested aseptical ly for bacterial enumeration and histopathology. A l l an imal experiments were conducted i n a manner consistent w i t h the ethical requirements o f the A n i m a l Care Commi t tee at the U n i v e r s i t y o f B r i t i s h C o l u m b i a and the Canad ian C o u n c i l o n the U s e o f Labora tory A n i m a l s . 6.1.3 Bacterial enumeration Tissues were col lected at var ious t imepoints into 1.5mL sterile P B S and homogenized us ing a tissue homogenizer (Poly t ron M R 21—, K i n e m a t i c a , Lucerne , Swi tzer land) . Ser ia l di lut ions o f the result ing mixture were plated on L B - a g a r plates containing 100 u,g/mL streptomycin. The threshold o f detection was 50 co lony fo rming units ( C F U s ) per organ. 6.1.4 Histopathology Colons , ceca and i l ea o f experimental animals were f ixed i n 3 % formal in for 18 hours fo l l owed b y 18 hours i n 7 0 % ethanol pr ior to be ing embedded i n paraffin, sectioned and stained w i t h 93 hematoxy l in and eosin, or f ixed for 3 hours i n 3 % paraformaldehyde pr ior to embedding i n op t imal cutt ing temperature compound and cryosect ioning. r Patho log ica l scores were determined b y averaging 6 fields/sample as fo l lows , based on a r ev i s ion o f p rev ious ly publ i shed methods (quantitave and qualitative cri ter ia were separated) and descriptions o f human disease: L u m e n - S u m of: empty (score=0); necrotic epi thel ia l cel ls (scant=l , moderate=2, dense=3); P M N s (scant=2, moderate=3, dense=4). Surface E p i t h e l i u m - S u m of: no pa thologica l changes (0); Mild/moderate /severe regenerative change (score=l/2/3); patchy/diffuse desquamation (score=l/2) ; P M N s i n epi the l ium (score=l) ; U lce ra t ion (score 1). M u c o s a - S u m of: no pa thologica l changes (0); rare(<15%)/moderate(15-50%)/Abundant(>50%) crypt abscesses (score=l/2/3); presence o f mucinous plugs (score=l) ; presence o f granulat ion tissue (score=l) . Submucosa - S u m of: N o pathologica l changes (score=0); M o n o n u c l e a r c e l l infiltrate (1 sma l l aggregated-1 aggregate/large aggregates plus increased single cel ls , score=0/l /2) ; P M N infiltrate (no extravascular P M N s / s i n g l e extravascular P M N s / P M N aggregates, score=0/ l /2) ; mild/moderate/severe edema (score=0/l /2) . 6.1.5 Immunohistochemistry Paraff in embedded tissues were deparaffmized i n xy lene (2x 5 min ) and rehydrated i n 100%, 9 5 % , 7 0 % ethanol (5 m i n each) then washed P B S containing 0 . 1 % bov ine serum a lbumin ( P B S -B S A , B S A from Sigma) . Sections were outl ine w i t h a w a x pen and b l o c k e d 3 0 m i n i n 10% goat serum i n P B S - B S A at r o o m temperature. Sections were then washed 3 t imes i n P B S - B S A pr ior to incubat ion over night at 4 ° C or 1 hour r o o m temperature w i t h p r imary antibody (anti-Salmonella L P S 6 . 8 p g / m L , a n t i - I C A M - 1 , 5 u g / m L - B D Biosc iences ) . Sections were then washed 3x i n P B S - B S A pr ior to incubat ion w i t h appropriate f luorochrome or streptavidin conjugated secondary antibodies ( O . l m g / m L , 30 m i n , R T ) . Imaging was performed on a Zeiss A x i o s k o p epifluorescence microscope us ing M e t a M o r p h software for fluorescent images and o n a Zeiss A x i o s t a r microscope w i t h a N i k o n Powershot G 5 camera. L i g h t and fluorescent images were cropped and scaled i n A d o b e Photoshop vers ion 7.0.1. 6.1.6 Quantitative measures of inflammation M a s t cells from 100 sequential crypts were counted at a magni f ica t ion o f 400x . Goble t cel ls were enumerated from 10 random high-powered fields spanning muscular is mucosa to surface 94 epi the l ium us ing hematoxy l in and eosin stained sections. M u c o s a l and submucosal thickness was measured at 6 evenly spaced points per section for each experimental an imal . M u c o s a l thickness was defined as distance f rom the surface epi thel ium to the inner edge o f the muscular is mucosa . Submucosa thickness was defined as the distance from muscular is mucosa to muscular is externa. Averages for each mouse were compared. 6.1.7 Statistical analysis Tota l pa thologica l scores were compared us ing M a n n - W h i t n e y U and K r u s k a l l - W a l l i s nonparametric tests w i t h D u n n ' s post-tests. Cor re la t ion o f pathology scores and goblet c e l l number, cecal weight and mucosa l thickness were compared us ing Spearman's non-parametric correlat ion coefficients. M u c o s a l values were logar i thmica l ly transformed to satisfy homoscedast ic i ty requirements o f statistical tests. Bac te r ia l load was compared us ing A N O V A and T u k e y ' s M u l t i p l e comparisons post-tests. A l l analyses were performed us ing Graphpad P r i s m vers ion 4.0. 6.2 Methods from Chapter 2 6.2.1 Assessment of Normal Flora 10 week o l d male C57B16 mic e were depr ived o f food and water as described above and treated w i t h either 2 0 m g streptomycin sulfate ( S i g m a - A l d r i c h ) i n lOOul sterile water or water alone. One day f o l l o w i n g inocula t ion , m i ce were CO2 euthanized and ceca were harvested aseptically. Tissues were homogenized i n 1 m l sterile phosphate buffered saline us ing a tissue homogenizer (Poly t ron M R 21—, K i n e m a t i c a Lucerne , Switzer land) homogenates were ser ia l ly di luted i n sterile saline pr ior to f ixat ion w i t h 3 % formal in . Homogenates were incubated w i t h Sybr Green anti-nucleic ac id stain and plated onto coversl ips pr ior to moun t ing o n glass mic roscopy slides. Imaging was performed on a Zeiss A x i o s k o p epifluorescence microscope us ing M e t a M o r p h software for fluorescent images. 6.3 Methods from Chapter 3 6.3.1 Mast cell staining Paraff in embedded sections were deparaffmized as above pr ior to treatment w i t h the N a p h t h o l A S - D Chloroacetate Esterase staining k i t (Sigma). Sections were treated for 15 minutes to 95 visua l ize mast cel ls , or 1 hour to v i sua l ize neutrophils as p rev ious ly described ( W y l l i e et al, 2002). 6.3.2 Gentamicin protection assay Q 10 week o l d male C57B16 mic e were treated w i t h s t reptomycin and infected w i t h 3 x 1 0 w i l d -type serovar T y p h i m u r i u m S L 1 3 4 4 as described above. T w o days after infect ion, m ice were euthanized and tissues were aseptical ly recovered, and cut longi tud ina l ly to expose the intestinal lumen. Tissues were p laced i n 2 m L D M E M : N T C - 1 3 5 m e d i a (1:1) + 10% fetal c a l f serum w i t h or wi thout 100ug /ml gentamicin and incubated for 1 hour at 3 7 ° C pr ior to homogeniza t ion and bacterial enumeration as described above. 6.3.3 Murine ligated ileal loop infections Female C 5 7 B L / 6 mic e (6 -10 w k o f age) were purchased f rom Jackson Labora tory (Bar Harbor , M a i n e , U n i t e d States), and housed i n the an imal fac i l i ty at the U n i v e r s i t y o f B r i t i s h C o l u m b i a i n direct accordance w i t h guidelines drafted b y the U n i v e r s i t y o f B r i t i s h C o l u m b i a ' s A n i m a l Care Commi t t ee and the Canadian C o u n c i l o n the U s e o f Labora tory A n i m a l s . F o r i l ea l loop experiments, bacterial i nocu la o f approximately 10 7 C F U were prepared i n l O O u l , and the resolut ion status o f the strain was conf i rmed di rec t ly before inocula t ion . I leal loop experiments were mod i f i ed from those p rev ious ly described (Jones et al, 1994). In brief, m i c e were anaesthetized b y intraperitoneal inject ion o f ketamine and xy laz ine . F o l l o w i n g a m i d l i n e abdomina l inc i s ion , the sma l l b o w e l was exposed and the b o w e l was l igated twice , close to the cecum, to create a loop approximate ly 4 c m i n length into w h i c h the i n o c u l u m was injected. The b o w e l was then returned to the abdomina l cavi ty and the i n c i s i o n c losed w i t h discontinuous sutures. A t g iven t ime points, the mi ce were euthanized and tissues col lected for bacterial enumeration and R I V E T . The intestinal l umen was r insed w i t h P B S to remove non-adherent bacteria. Tissues were homogenized i n P B S us ing a P o l y t r o n homogenizer (Kinemat ica , Lucerne , Switzer land) . 6.3.4 Gene expression reporter (RIVET) assays Reporter Salmonella strains containing sseA reporter fusions ( B r o w n , 2005) were employed for SPI-2 reporter assays. Reso lu t ion assays were performed as described ( B r o w n , 2005) b y p la t ing serial di lut ions o f sample mater ial onto L B a m p i c i l l i n plates and incubat ing these overnight at 96 3 7 ° C . The f o l l o w i n g day, plates w i t h between 50 and 200 colonies were replica-plated onto L B a m p i c i l l i n plates and L B a m p i c i l l i n ch loramphenico l plates, w h i c h were incubated overnight at 3 7 ° C . Co lon ie s that grew o n the a m p i c i l l i n on ly plates, but not on the a m p i c i l l i n ch loramphenico l plates were considered to have undergone the resolut ion event. 6.4 Methods from Chapter 4 6.4.1 Calf intestinal loop surgeries A l l an imal experiments were conducted i n accordance w i t h the G u i d e to the Care and U s e o f Exper imenta l A n i m a l s , p rov ided b y the Canad ian C o u n c i l o n A n i m a l Care. One month o ld , male H o l s t e i n calves were housed i n single i so la t ion cubicles at the V a c c i n e and Infectious Disease Organiza t ion ( V I D O ) an imal faci l i ty . A l l animals were c l i n i c a l l y healthy pr ior to surgery and rectal swabs f rom a l l calves were tested for Salmonella p r ior to experimentat ion and were found to be negative i n a l l cases. The calves were fasted for 24 h pr ior to surgery and then pre-medicated w i t h butorphanol (0.2 mg/kg) and d iazepam (0.1 mg/kg) . Anes thes ia was induced w i t h 6-8 m l o f 5% thiopental sod ium pr ior to placement o f an endotracheal tube and maintenance w i t h isoflurane. The intestinal i l ea l ' l o o p ' m o d e l developed i n sheep (13) was adapted for use i n one-month o l d male calves. A laparotomy was performed and the sma l l intestine was exter ior ized unt i l s ix consecutive Peyer ' s patches were identif ied. The exposed sma l l intestine was frequently mois tened w i t h sterile P B S pre-warmed to 3 7 ° C . A n intestinal segment containing two Peyer ' s patches separated b y interspaces wi thout Peyer ' s patches was demarcated b y intestinal c lamps approximate ly 30 c m p r o x i m a l and 30 c m distal to the first and last Peyer ' s patch and then transected. Th i s intestinal segment was flushed twice w i t h 100 m l o f w a r m sterile saline to clean the gut o f its contents. The patency o f the intestine was restored w i t h end-to-end anastomoses b y a l ign ing the mesenteric and antirhesenteric borders o f the transected intestine and c los ing w i t h interrupted and continuous sutures. S i l k ligatures were t ied approximately 8 c m p r o x i m a l and distal to each Peyer ' s patch to create - 1 6 - 1 8 c m isolated segments containing a Peyer ' s patch, separated b y interspaces o f var ious length that l acked Peyer ' s patches. Three anastamoses were created i n each animal , generating three intestinal segments each containing two loops that p rov ided two independent sites for duplicate samples. F o r each animal , the intestinal segments received (i) - sterile saline, ( i i ) w i ld - type Salmonella and ( i i i ) a Salmonella mutant under invest igation. The loops were infected w i t h 3 m l o f sterile P B S or 3 m l o f a . 97 bacterial suspension di luted i n P B S to contain either l x 10 3 or l x l O 6 C F U . Intestinal segments were marked b y s i lk sutures and the succession and size o f each internal loop and interspace was recorded. The loops were then replaced into the abdominal cavi ty and the surgical i nc i s ion i n the abdominal w a l l was sutured. T h i s surgery was completed i n a total o f s ix animals w i t h two calves euthanized at 24 h post-surgery and four calves euthanized at 5-days post-surgery. 6.4.2 Calf intestinal tissue collection, specimen handling and histopathology Immedia te ly after euthanasia, intestinal loops were exposed f rom the abdomina l cav i ty and the f lu id v o l u m e was collected. F o r his topathology o f intestines, sections o f i l e u m were f ixed i n 10% neutral-buffered formal in , embedded i n paraffin, sectioned at 5 Lim thickness and stained w i t h hematoxy l in and eosin. His topa tho logy was scored as above for mur ine h i s to log ica l sections. 6.5 Methods from Chapter 5 6.5.1 Bacterial strains. C o n t r o l Sa lmone l l a strains are f rom Na t iona l Institute for the C o n t r o l o f Pharmaceut ica l and B i o l o g i c a l P roduc t s fNTCPB) . Exper imenta l strains were isolated f rom stool samples o f patients affected b y food po i son ing i n Shenzhen, Guangdong P rov ince , C h i n a . Salmonella enterica serovar T y p h i m u r i u m S L 1 3 4 4 ( w i l d type) and i n v A mutant ( i n v A : : k a n S B 1 0 3 ; SPI-1) were p rov ided b y Bret t F i n l a y at the M i c h a e l S m i t h Laboratories, U n i v e r s i t y o f B r i t i s h C o l u m b i a . 6.5.2 Bacterial Identification Stoo l samples were inoculated o n S S agar for 18-24hr at 3 5 ° C . Co lon ie s were selected and identif ied b y us ing b iochemica l and serological method based on the B e r g e y ' s Bac te r ia l M a n u a l . 6.5.3 Pulse-Field Gel Electrophoresis (PFGE) C o n t r o l serovar Senftenberg strains, experimental strains and serovar T y p h i H 9 8 1 2 standard strains were g r o w n shaking at 37 °C , overnight. 1ml o f the overnight culture was subcultured into 5 m l L B at 37 °C , shaking 1-1.5hr unt i l the Op t i ca l density equalled 0.6. Cul ture (1ml) was pel leted i n a 1 .5mL eppendorf tube and washed pr io r to resuspension i n 1ml T E buffer. A n equal v o l u m e o f mol ten 2 % agarose was added to the c e l l suspension and m i x e d b y gently pipet t ing up 98 and down. The result ing suspension was dispensed into 2 p l u g molds w h i c h were then a l l owed to c o o l for 5 m i n at r o o m temperature. The plugs were transferred to 15ml tubes w i t h 3 m l lysis buffer ( 5 0 m M Tr i s p H 8.0, 5 0 m M E D T A , p H 8.0, 1% sod ium deoxycholate, 0 .15mg/ml Proteinase K , l m g freshly added l y s o z y m e per m l and 2ug o f R N a s e per m l ) , inoculated for 2hrs at 5 0 - 5 4 ° C i n an orbital water bath shaker. Af t e r proteolysis , the lys is buffer solut ion was removed and the plugs were washed twice w i t h 15ml sterile water for 10 m i n each fo l l owed b y four washes w i t h 15ml o f T E buffer i n the orbital water bath shaker ( room temperature) at 200 rpm. Af te r the f inal T E wash , one p l u g was s l iced and the p l u g s l ice was digested w i t h X b a l at 37 °C overnight. The D N A restr ict ion fragments were separated b y electrophoresis through 1% S e a K e m G o l d agarose ge l i n 0.5 X so lu t ion o f T r i s - b o r a t e - E D T A buffer at 14 °C i n a GenePath P F G E apparatus ( B i o - R a d ) . The electrophoretic parameters used were as fo l lows : in i t i a l swi tch t ime, 4.0s; f inal swi tch t ime, 40.0s; run t ime, 22h ; angle, 120 °; r amping factor, linear. Af t e r electrophoresis, the gel was stained for 15-20 m i n i n 2 5 0 m l o f de ion ized water containing 25u l o f e th id ium bromide ( l O m g / m l ) and destained 20-30 m i n b y us ing 500 m l o f de ionized water. The D N A restr ict ion fragments were compared to X 9 8 4 4 . 6.5.4 PCR Assay The f o l l o w i n g pr imers o f SPI-1 and SPI -2 effectors and vi ru lence factors were used for P C R : invA forward: C A G C G A T A T C C A A A T G T T G C , reverse: A A A T G G C A G A A C A G C G T C G T A ; sipA forward: A T G G G T A C C A G G C G G C T A C T A A A A T C C , reverse: A T G G A G C T C C A A G C G A G A G A A A A A T C T A C A C ; sopE forward: C A A C A A C A T C A A C A C C C G C G C C A C C , reverse: A T A G G C A A T A G C T T C C T C C A C C ; ssaR forward: G T T C G G A T T C A T T G C T T C G G , reverse: T C T C C A G T G A C T A A C C C T A A C C A A ; sifA forward: A T G G T C G A C A T G C C G A T T A C T A T A G G G A A T G G , reverse: A T G G G A T C C T T A T A A A A A A C A A C A T A A A C A G C C G . Bac te r ia l strains were g r o w n overnight at 3 7 ° C i n 3 m l Lur i a -Ber t an i broth. 1ml bacterial culture was used for D N A extraction. Af te r be ing heated at 100 °C for 10 m i n , the suspension was coo led o n ice for 10 m i n then pelleted b y centrifugation i n microcentr i fuge tubes (13000rpm, l O m i n ) . P C R was carried out i n a 20u l vo lume w i t h l O m M Mgc i2 , IX buffer, l U T a q 99 polymerase, 2 0 0 n M primers . W e ampl i f ied i n v A ( 95 °C for l O m i n , 35cycles o f 94 °C for 30s, 55 °C for I m i n , 72 °C for 2 . 5 m i n , 72 °C for l O m i n ) , s i p A ( 9 4 °C for 2 m i n , 35cycles o f 94 °C for 30s, 55 °C for I m i n , 68 °C for 4 m i n ,68 °C for lOmin) , SopE(94 °C for 2 m i n , 35cycles o f 94 °C for 30s, 6 0 ° C for I m i n , 68 °C for 3 .5min ,68 °C for l O m i n ) , ssaR(95 °C for l O m i n , 35cycles o f 94 °C for 30s, 50 °C for I m i n , 72 °C for 2 m i n , 72 °C for l O m i n ) , s i fA(95 °C for l O m i n , 35cycles o f 94 °C for 30s, 52 °C for I m i n , 72 °C for I m i n , 72 °C for l O m i n ) . P C R products were separated o n 1% agarose w i t h e th id ium bromide . 6.5.5 Preparation of PCR probes The D N A probes used i n Southern hybr id iza t ion were prepared b y i n v A and s i p A ampl i f ica t ion o f S L 1 3 4 4 strain b y P C R us ing the f o l l o w i n g primers: invA-l forward: T C C C T T T G C G A A T A A C A T C C , reverse: T A C G G T T C C T T T G A C G G T G C ; invA-2 forward: G G G T C A A G G C T G A G G A A G , reverse: C T G A T C G C A C T G A A T A T C G T A C ; sipA forward: C G G C T T C A C A T T C A C A A , reverse: C G G G C T C T T T C G T T C A . Probe lengths were as fo l lows : invA-l 267bp; invA-2 1736bp; sipA 1126bp i n length. The probes were labeled b y bio t in . 6.5.6 Restriction endonuclease digestion and Southern transfer: 6 strains i nc lud ing two c l i n i c a l isolates (S. Senftenberg) and control strains were used for southern blot . G e n o m i c D N A from each o f the strains was extracted us ing phenol -ch loroform extraction o f the chromosomal D N A . A p p r o x i m a t e l y l u g o f D N A was digested w i t h 2 0 U o f restr ict ion endonuclease E c o R I at 3 7 ° C overnight . The D N A fragments were separated b y electrophoresis i n 0 .7% agarose gels and capi l la ry transferred to H y b o n d N * membranes . 6.5.7 Southern Blot hybridization: The southern blots were hybr id i zed w i t h prepared probes at 55 °C i n hybr id iza t ion buffer. The membrane was washed once for 20 m i n i n 0.1 X S S C ( I X S S C is 0 . 1 5 M N a c l plus 0 . 0 1 5 M sod ium citrate), f o l l o w e d b y three washes for 2 0 m i n each i n 2 X S S C - 0 . 5 % S D S at 55 °C and then exposed. 100 6.5.8 Invasion Assay 11 strains i nc lud ing two c l i n i c a l isolates and other Sa lmone l l a strains were cultured i n 1.5ml L B at 37 °C overnight. 24 w e l l plates were seeded w i t h H e L a cells i n med ia (10% F B S ) at 37 °C overnight and cel ls were washed once. Bac te r i a were added ( l O u l o f O.D .6oonm=0.5) to each w e l l p r ior to incubat ion for 30 m i n . 30 minutes after inocula t ion , cel ls were washed w i t h P B S twice pr ior to the addi t ion o f fresh med ia containing 100u.g/ml gentamicin. Infected cel ls were incubated for 2 hours (37 °C) to k i l l the extracellular bacteria. Af te r incubat ion, cel ls were washed 3 t imes w i t h P B S , fo l lowed b y adding 500u l lys is buffer ( 1 % Tr i ton , 0 . 1 % S D S and P B S ) for 10 minutes at r o o m temperature. L y s e d cel ls were resuspended b y pipet t ing up and d o w n 4-5 t imes p r io r to serial d i lu t ion i n P B S . D i lu t i ons were then plated onto L B - a g a r and incubated overrnight at 37 °C . The f o l l o w i n g day, bacterial colonies were enumerated and d i v i d e d b y the bacterial i n o c u l u m , then norma l i zed b y the percent invasiveness o f wi ld - type serovar T y p h i m u r i u m S L 1 3 4 4 . 101 Chapter 7: References Alpuche-Aranda, C M ; Racoosin, E L ; Swanson, J A ; Mi l le r , SI. Salmonella Stimulate Macrophage Macropinocytosis and Persist W i t h i n Spacious Phagosomes. (1994). J.Exp Med. 179:601. Altier, C . Genetic and Environmental Con t ro l of Salmonella Invasion. (2005). J.Microbiol. 43 SpecNo:85. Arnold, T ; Scholz, H C ; Marg, H ; Rosier, U ; Hensel, A . Impact of I n v A - P C R and Cul ture Detection Methods on Occurrence and Survival of Salmonella in the Flesh, Internal Organs and L y m p h o i d Tissues of Experimental ly Infected Pigs. (2004). J. Vet.Med.B Infect.Dis. Vet.Public Health 51:459. Ayabe, T ; Satchell, D P ; Wilson, C L ; Parks, W C ; Selsted, M E ; Ouellette, A J . Secretion of M i c r o b i c i d a l Alpha-Defensins by Intestinal Paneth Cells in Response to Bacteria. (2000). Nat.Immunol. 1:113. Barthel, M ; Hapfelmeier, S; Quintanilla-Martinez, L ; Kremer, M ; Rohde, M ; Hogardt, M ; Pfeffer, K ; Russmann, H ; Hardt, W D . Pretreatment of M i c e W i t h Streptomycin Provides a Salmonella Enter ica Serovar T y p h i m u r i u m Coli t is M o d e l That Al lows Analysis of Both Pathogen and Host. (2003). Infect.Immun. 71:2839. Baumler, A J ; Tsolis, R M ; Ficht, T A ; Adams, L G . Evolut ion of Host Adaptat ion in Salmonella Enter ica. (1998). Infect.Immun. 66:4579. Baumler, A J ; Tsolis, R M ; Bowe, F A ; Kusters, J G ; Hoffmann, S; Heffron, F . The Pef F i m b r i a l Operon of Salmonella T y p h i m u r i u m Mediates Adhesion to M u r i n e Smal l Intestine and Is Necessary for F l u i d Accumulat ion in the Infant Mouse. (1996a). Infect.Immun. 64:61. Baumler, A J ; Tsolis, R M ; Heffron, F . Contr ibut ion of F i m b r i a l Operons to Attachment to and Invasion of Epi thel ia l C e l l Lines by Salmonella T y p h i m u r i u m . (1996b). Infect.Immun. 64:1862. Baumler, A J ; Tsolis, R M ; Heffron, F . The L p f F i m b r i a l Operon Mediates Adhesion of Salmonella T y p h i m u r i u m to M u r i n e Peyer's Patches. (1996c). Proc.Natl.Acad.Sci. U.S.A 93:279. Baumler, A J ; Tsolis, R M ; Heffron, F . F i m b r i a l Adhesins of Salmonella Typh imur ium. Role in Bacterial Interactions W i t h Epi thel ia l Cells. (1997). Adv.Exp.Med.Biol. 412:149. Behlau, I and Mil le r , SI . A PhoP-Repressed Gene Promotes Salmonella T y p h i m u r i u m Invasion of Epi thel ia l Cells. (1993). J.Bacteriol. 175:4475. 102 B e u z o n , C R ; Meresse , S; U n s w o r t h , K E ; R u i z - A l b e r t , J ; Garv i s , S; Waterman, S R ; Ryder , T A ; Boucro t , E ; H o l d e n , D W . Salmonella Maintains the Integrity of Its Intracellular Vacuole Through the Action of SifA. (2000). Embo Journal 19:3235. B i j l s m a , J J and G r o i s m a n , E A . The PhoP/PhoQ System Controls the Intramacrophage Type Three Secretion Systemof Salmonella Enterica. (2005). Mol.Microbiol. 57:85. B i s p h a m , J ; Tr ipa th i , B N ; Watson , P R ; W a l l i s , T S . Salmonella Pathogenicity Island 2 Influences Both Systemic Salmonellosis and Salmonella-Induced Enteritis in Calves. (2001). Infect.Immun. 69:367. Bohnhoff , M ; Drake , B L ; M i l l e r , C P . Effect of Streptomycin on Susceptibility of Intestinal Tract to Experimental Salmonella Infection. (1954). Proc.Soc.Exp Biol.Med. 86:132. Bohnhoff , M and M i l l e r , C P . Enhanced Susceptibility to Salmonella Infection in Streptomycin-Treated Mice. (1962). J.Infect.Dis. 111:117. Bohnhoff , M ; M i l l e r , C P ; M R t i n , W R . Resistance of the Mouse's Intestinal Tract to Experimental Salmonella Infection, II. Factors Responsible for its Loss Following Streptomycin Treatment. (1964). J.Exp Med. 120:817. B o y d , I F . Salmonella Typhimurium, Colitis, and Pancreatitis. (1969). Lancet 2 :901. B o y d , J F . Pathology of the Alimentary Tract in Salmonella Typhimurium Food Poisoning. (1985). Gut 26:935. B r o w n , N F ; V a l l a n c e , B A ; Coombes , B K ; V a l d e z , Y ; Coburn , B A ; F i n l a y , B B . Salmonella Pathogenicity Island 2 Is Expressed Prior to Penetrating the Intestine. (2005). PLoS.Pathog. I :e32. B r u m e l l , J H ; K u j a t - C h o y , S; B r o w n , N F ; V a l l a n c e , B A ; K n o d l e r , L A ; F i n l a y , B B . SopD2 Is a Novel Type III Secreted Effector of Salmonella Typhimurium That Targets Late Endocytic Compartments Upon Delivery into Host Cells. (2003). Traffic 4:36. B r u m e l l , J H ; Rosenberger, C M ; Got to , G T ; M a r c u s , S L ; F i n l a y , B B . SifA Permits Survival and Replication of Salmonella Typhimurium in Murine Macrophages. (2001). Cell.Micro. 3:75. Bu l t e , M and Jakob, P . The Use of a PCR-Generated InvA Probe for the Detection of Salmonella Spp. in Artificially and Naturally Contaminated Foods. (1995). Int.JFood Microbiol. 26:335. C a m i l l i , A ; Beatt ie, D T ; Meka l anos , J J . Use of Genetic Recombination As a Reporter of Gene Expression. (1994). Proc.Natl.Acad.Sci. U.S.A 91:2634. Cheminay , C ; M o h l e n b r i n k , A ; Hense l , M . Intracellular Salmonella Inhibit Antigen Presentation by Dendritic Cells. (2005). J.Immunol 174:2892. 103 Chen , L M ; K a n i g a , K ; G a l a n , J E . Salmonella Spp. Are Cytotoxic for Cultured Macrophages. (1996). Mol.Microbiol 21 :1101. C h i u , C H and O u , J T . Rapid Identification of Salmonella Serovars in Feces by Specific Detection of Virulence Genes, InvA and SpvC, by an Enrichment Broth Culture-Multiplex PCR Combination Assay. (1996). J.Clin Microbiol. 34:2619. Chopra , A K ; Huang , J H ; X u , X ; Burden , K ; N i e s e l , D W ; Rosenbaum, M W ; Popov , V L ; Peterson, J W . Role of Salmonella Enterotoxin in Overall Virulence of the Organism. (1999). Microb.Pathog. 27:155. C i r i l l o , D M ; V a l d i v i a , R H ; M o n a c k , D M ; F a l k o w , S. Macrophage-Dependent Induction of the Salmonella Pathogenicity Island 2 Type III Secretion System and Its Role in Intracellular Survival. (1998). Mol.Micro. 30:175. C la rke , R C and G y l e s , C L . Virulence of Wild and Mutant Strains of Salmonella Typhimurium in Ligated Intestinal Segments of Calves, Pigs, and Rabbits. (1987). Am.J.Vet.Res. 48:504. Coburn , B ; L i , Y ; O w e n , D ; V a l l a n c e , B A ; F i n l a y , B B . Salmonella Enterica Serovar Typhimurium Pathogenicity Island 2 Is Necessary for Complete Virulence in a Mouse Model of Infectious Enterocolitis. (2005). Infect.Immun. 73:3219. C o c o l i n , L ; M a n z a n o , M ; A s t o r i , G ; Bot ta , G A ; Can ton i , C ; C o m i , G . A Highly Sensitive and Fast Non-Radioactive Method for the Detection of Polymerase Chain Reaction Products From Salmonella Serovars, Such As Salmonella Typhi, in Blood Specimens. (1998a). FEMSImmunolMed.Microbiol. 22:233. C o c o l i n , L ; M a n z a n o , M ; Can ton i , C ; C o m i , G . Use of Polymerase Chain Reaction and Restriction Enzyme Analysis to Directly Detect and Identify Salmonella Typhimurium in Food. (1998b). J.Appl.Microbiol. 85:673. C o l l a z o , C M and G a l a n , J E . The Invasion-Associated Type-Ill Protein Secretion System in Salmonella-a Review. (1997). Gene 192:51. C o l l i e r - H y a m s , L S ; Z e n g , H ; Sun, J ; T o m l i n s o n , A D ; B a o , Z Q ; C h e n , H ; M a d a r a , J L ; Or th , K ; N e i s h , A S . Cutting Edge: Salmonella AvrA Effector Inhibits the Key i Proinflammatory, Anti-Apoptotic NF-Kappa B Pathway. (2002). J.Immunol 169:2846. Coombes , B K ; B r o w n , N F ; V a l d e z , Y ; B r u m e l l , J H ; F i n l a y , B B . Expression and Secretion of Salmonella Pathogenicity Island-2 Virulence Genes in Response to Acidification Exhibit Differential Requirements of a Functional Type III Secretion Apparatus and SsaL. (2004). J.Biol.Chem. 279:49804. Coombes , B K ; C o b u r n , B A ; Potter, A A ; G o m i s , S; M i r a k h u r , K ; L i , Y ; F i n l a y , B B . Analysis of the Contribution of Salmonella Pathogenicity Islands 1 and 2 to Enteric Disease Progression Using a Novel Bovine Ileal Loop Model and a Murine Model of Infectious Enterocolitis. (2005). Infect.Immun. 73:7161. 104 Deiwick, J; Nikolaus, T ; Shea, JE; Gleeson, C ; Holden, DW; Hensel, M. Mutations in Salmonella Pathogenicity Island 2 (SPI2) Genes Affecting Transcription of SPI1 Genes and Resistance to Antimicrobial Agents. (1998). J.Bacteriol. 180:4775. Dippold, W; Wittig, B; Schwaeble, W; Mayet, W; Meyer zum Buschenfelde, KH. Expression of Intercellular Adhesion Molecule 1 (ICAM-1, CD54) in Colonic Epithelial Cells. (1993). Gut 34:1593. Dore, K; Buxton, J; Henry, B; Pollari, F; Middleton, D; Fyfe, M; Ahmed, R; Michel, P; King, A; Tinga, C ; Wilson, JB. Risk Factors for Salmonella Typhimurium DT104 and Non-DT104 Infection: a Canadian Multi-Provincial Case-Control Study. (2004). Epidemiol.Infect. 132:485. Dougan, G; Chatfield, S; Pickard, D; Bester, J; O'callaghan, D; Maskell, D. Construction and Characterization of Vaccine Strains of Salmonella Harboring Mutations in Two Different Aro Genes. (1988). J.Infect.Dis. 158:1329. Drecktrah, D; Knodler, LA; Ireland, R; Steele-Mortimer, O . The Mechanism of Salmonella Entry Determines the Vacuolar Environment and Intracellular Gene Expression. (2006). Traffic 7:39. Duguid, JP; Darekar, MR; Wheater, DW. Fimbriae and Infectivity in Salmonella Typhimurium. (1976). J.Med.Microbiol. 9:459. Eckmann, L; Stenson, WF; Savidge, TC; Lowe, DC; Barrett, KE; Fierer, J; Smith, JR; Kagnoff, MF. Role of Intestinal Epithelial Cells in the Host Secretory Response to Infection by Invasive Bacteria. Bacterial Entry Induces Epithelial Prostaglandin h Synthase-2 Expression and Prostaglandin E2 and F2alpha Production. (1997). J.Clin.Invest 100:296. Ehrbar, K; Friebel, A; Miller, SI; Hardt, WD. Role of the Salmonella Pathogenicity Island 1 (SPI-1) Protein InvB in Type III Secretion of SopE and SopE2, Two Salmonella Effector Proteins Encoded Outside of SPI-1. (2003). J. Bad. 185:6950. Everest, P; Ketley, J; Hardy, S; Douce, G; Khan, S; Shea, J; Holden, D; Maskell, D; Dougan, G. Evaluation of Salmonella Typhimurium Mutants in a Model of Experimental Gastroenteritis. (1999). Infect.Immun. 67:2815. Fields, PI; Swanson, RV; Haidaris, CG; Heffron, F. Mutants of Salmonella Typhimurium That Cannot Survive Within the Macrophage Are Avirulent. (1986). Proc.Natl.Acad.Sci.U.S.A 83:5189. Fierer, J and Guiney, DG. Diverse Virulence Traits Underlying Different Clinical Outcomes of Salmonella Infection. (2001). J.Clin.Invest 107:775. Finlay, BB and Brumell, JH. Salmonella Interactions With Host Cells: in Vitro to in Vivo. (2000). Philos.Trans.R.Soc.LondB Biol.Sci. 355:623. 105 Finlay, B B and Falkow, S. Salmonella Interactions With Polarized Human Intestinal Caco-2 Epithelial Cells. (1990). J.Infect.Dis. 162:1096. Finlay, B B ; Heffron, F; Falkow, S. Epithelial Cell Surfaces Induce Salmonella Proteins Required for Bacterial Adherence and Invasion. (1989). Science 243:940. Freeman, J A ; O h l , M E ; Miller, SI . The Salmonella Enterica Serovar Typhimurium Translocated Effectors SseJ and SifB Are Targeted to the Salmonella-Containing Vacuole. (2003). Infect.Immun. 71:418. Frost, A J ; Bland, A P ; Wallis, T S . The Early Dynamic Response of the Calf Ileal Epithelium to Salmonella Typhimurium. (1997). Vet.Pathol. 34:369. Fu, Y and G a l a n , J E . The Salmonella Typhimurium Tyrosine Phosphatase SptP Is Translocated into Host Cells and Disrupts the Actin Cytoskeleton. (1998). Mol.Microbiol. 27:359. Galan, J E . Molecular Genetic Bases of Salmonella Entry into Host Cells. (1996). Mol.Microbiol. 20:263. G a l a n , J E . Interaction of Salmonella With Host Cells Through the Centisome 63 Type III Secretion System. (1999). Curr.Opin.Microbiol. 2:46. G a l a n , J E . Salmonella Interactions With Host Cells: Type III Secretion at Work. (2001). Annu.Rev. Cell Dev.Biol. 17:53. Galan, J E and Curtiss, R , III. Cloning and Molecular Characterization of Genes Whose Products Allow Salmonella Typhimurium to Penetrate Tissue Culture Cells. (1989). Proc.Natl.Acad.Sci.U.S.A 86:6383. Galyov, E E ; W o o d , M W ; Rosqvist, R ; Mullan, P B ; Watson, P R ; Hedges, S; Wallis, T S . A Secreted Effector Protein of Salmonella Dublin Is Translocated into Eukaryotic Cells and Mediates Inflammation and Fluid Secretion in Infected Ileal Mucosa. (1997). Mol.Microbiol. 25:903. Garcia-del Portillo, F; Zwick, M B ; Leung, K Y ; Finlay, B B . Intracellular Replication of Salmonella Within Epithelial Cells Is Associated With Filamentous Structures Containing Lysosomal Membrane Glycoproteins. (1993a). Infect.Agents Dis. 2:227. Garcia-del Portillo, F; Zwick, M B ; Leung, K Y ; Finlay, B B . Salmonella Induces the Formation of Filamentous Structures Containing Lysosomal Membrane Glycoproteins in Epithelial Cells. (1993b). Proc.Natl.Acad.Sci.U.S.A 90:10544. Gewirtz, A T ; R a o , A S ; Simon, P O , Jr.; Merlin, D; Carnes, D; Madara, J L ; Neish, A S . Salmonella Typhimurium Induces Epithelial IL-8 Expression Via Ca(2+)-Mediated Activation of the NF-KappaB Pathway. (2000). J.Clin Invest 105:79. 106 Gewirtz, AT; Siber, AM; Madara, JL; McCormick, BA. Orchestration of Neutrophil Movement by Intestinal Epithelial Cells in Response to Salmonella Typhimurium Can Be Uncoupled From Bacterial Internalization. (1999). Infect.Immun. 67:608. Gewirtz, AT; Simon, PO, Jr.; Schmitt, CK; Taylor, LJ; Hagedorn, CH; O'Brien, AD; Neish, AS; Madara, JL. Salmonella Typhimurium Translocates Flagellin Across Intestinal Epithelia, Inducing a Proinflammatory Response. (2001). J.ClinJnvest 107:99. Ghosh, P. Process of Protein Transport by the Type III Secretion System. (2004). Microbiol.Mol.Biol.Rev. 68:771. Giannella, RA. Importance of the Intestinal Inflammatory Reaction in Salmonella-Mediated Intestinal Secretion. (1979). Infect.Immun. 23:140. Giannella, RA; Formal, SB; Dammin, GJ; Collins, H. Pathogenesis of Salmonellosis. Studies of Fluid Secretion, Mucosal Invasion, and Morphologic Reaction in the Rabbit Ileum. (1973). J.ClinJnvest 52:441. Gophna, U ; Ron, EZ; Graur, D. Bacterial Type III Secretion Systems Are Ancient and Evolved by Multiple Horizontal-Transfer Events. (2003). Gene 312:151. Gorvel, JP and Meresse, S. Maturation Steps of the Salmonella-Containing Vacuole. (2001). Microbes Infect. 3:1299. Govoni, G; Vidal, S; Gauthier, S; Skamene, E; Malo, D; Gros, P. The Bcg/Ity/Lsh Locus: Genetic Transfer of Resistance to Infections in C57BL/6J Mice Transgenic for the Nrampl Glyl69 Allele. (1996). Infect.Immun. 64:2923. Groisman, EA and Ochman, H. How to Become a Pathogen. (1994). Trends Microbiol. 2:289. Groisman, EA and Ochman, H. Pathogenicity Islands: Bacterial Evolution in Quantum Leaps. (1996). Cell 87:791. Guiney, DG. The Role of Host Cell Death in Salmonella Infections. (2005). Curr. Top.Microbiol.Immunol 289:131. Hanes, DE; Robl, MG; Schneider, CM; Burr, DH. New Zealand White Rabbit As a Nonsurgical Experimental Model for Salmonella Enterica Gastroenteritis. (2001). Infect.Immun. 69:6523. Hansen-Wester, I and Hensel, M. Salmonella Pathogenicity Islands Encoding Type III Secretion Systems. (2001). Microbes.Infect. 3:549. Hapfelmeier, S; Ehrbar, K; Stecher, B; Barthel, M; Kremer, M; Hardt, WD. Role of the Salmonella Pathogenicity Island 1 Effector Proteins SipA, SopB, SopE, and SopE2 in Salmonella Enterica Subspecies 1 Serovar Typhimurium Colitis in Streptomycin-Pretreated Mice. (2004). Infect.Immun. 72:795. 107 Hapfelmeier , S; Stecher, B ; Bar the l , M ; K r e m e r , M ; M u l l e r , A J ; He ikenwalder , M ; Sta l lmach, T ; Hense l , M ; Pfeffer, K ; A k i r a , S; Hardt , W D . The Salmonella Pathogenicity Island (SPI)-2 and SPI-1 Type III Secretion Systems Allow Salmonella Serovar Typhimurium to Trigger Colitis Via MyD88-Dependent and MyD88-Independent Mechanisms. (2005). J.Immunol. 174:1675. Haque , A ; B o w e , F ; F i tzhenry , R J ; F ranke l , G ; T h o m s o n , M ; Heuschke l , R ; M u r c h , S; Stevens, M P ; W a l l i s , T S ; P h i l l i p s , A D ; Dougan , G . Early Interactions of Salmonella Enterica Serovar Typhimurium With Human Small Intestinal Epithelial Explants. (2004). Gut 53:1424. Hardt , W D ; C h e n , L M ; Schuebel , K E ; Bus te lo , X R ; Ga lan , J E . S. Typhimurium Encodes an Activator of Rho GTPases That Induces Membrane Ruffling and Nuclear Responses in Host Cells. (1998). Cell 93:815. H a y w a r d , R D and K o r o n a k i s , V . Direct Nucleation and Bundling of Actin by the SipC Protein of Invasive Salmonella. (1999). EMBO J. 18:4926. H a y w a r d , R D ; M c G h i e , E J ; K o r o n a k i s , V . Membrane Fusion Activity of Purified SipB, a Salmonella Surface Protein Essential for Mammalian Cell Invasion. (2000). Mol.Microbiol. 37:727'. Hea l th Canada. Salmonella Prevention. (2006). www,hc-scsc. ca/iyh-vsv/food-aliment/salmonella e.html Hense l , M ; Shea, J E ; Gleeson , C ; Jones, M D ; Da l ton , E ; H o l d e n , D W . Simultaneous Identification of Bacterial Virulence Genes by Negative Selection. (1995). Science 269:400. Hense l , M ; Shea, J E ; Waterman, S R ; M u n d y , R ; N i k o l a u s , T ; B a n k s , G ; Vazquez-Tor res , A ; Gleeson , C ; Fang , F C ; H o l d e n , D W . Genes Encoding Putative Effector Proteins of the Type III Secretion System of Salmonella Pathogenicity Island 2 Are Required for Bacterial Virulence and Proliferation in Macrophages. (1998). Mol.Microbiol. 30:163. Hentges, D J ; Stein, A J ; Casey, S W ; Que, J U . Protective Role of Intestinal Flora Against Infection With Pseudomonas Aeruginosa in Mice: Influence of Antibiotics on Colonization Resistance. (1985). Infect.Immun. 47:118. Hersh , D ; M o n a c k , D M ; Smi th , M R ; G h o r i , N ; F a l k o w , S; Z y c h l i n s k y , A . The Salmonella Invasin SipB Induces Macrophage Apoptosis by Binding to Caspase-1. (1999). Proc.Natl.Acad.Sci. U.S.A 96:2396. H i n d l e , Z ; Chat f ie ld , S N ; P h i l l i m o r e , J ; Bent ley , M ; Johnson, J ; Cosgrove , C A ; G h a e m -M a g h a m i , M ; Sexton, A ; K h a n , M ; Brennan , F R ; Everest , P ; W u , T ; P i cka rd , D ; H o l d e n , D W ; Dougan , G ; G r i f f i n , G E ; House , D ; Santangelo, J D ; K h a n , S A ; Shea, J E ; Fe ldman , R G ; L e w i s , D J . Characterization of Salmonella Enterica Derivatives Harboring Defined AroC and Salmonella Pathogenicity Island 2 Type III Secretion System 108 (SsaV) Mutations by Immunization of Healthy Volunteers. (2002). Infect.Immun. 70:3457. Hoiseth, SK and Stacker, BA. Aromatic-Dependent Salmonella Typhimurium Are Non-Virulent and Effective As Live Vaccines. (1981). Nature 291:238. Hornick, RB. Pathogenesis of Typhoid Fever. (1970). J.Egypt.Public Health Assoc. 45:247. Hueck, CJ. Type III Protein Secretion Systems in Bacterial Pathogens of Animals and Plants. (1998). Microbiol.Mol.Biol.Rev. 62:379. Iimura, M; Gallo, RL; Hase, K; Miyamoto, Y ; Eckmann, L; Kagnoff, MF. Cathelicidin Mediates Innate Intestinal Defense Against Colonization With Epithelial Adherent Bacterial Pathogens. (2005). J.Immunol 174:4901. Inohara, N; Chamaillard, M; McDonald, C; Nunez, G. NOD-LRR Proteins: Role in Host-Microbial Interactions and Inflammatory Disease. (2004). Annu.Rev.Biochem. Inohara, N and Nunez, G. NODs: Intracellular Proteins Involved in Inflammation and Apoptosis. (2003). Nat.Rev.Immunol 3:371. Ishibashi, Y and Arai, T . Roles of the Complement Receptor Type 1 (CRI) and Type 3 (CR3) on Phagocytosis and Subsequent Phagosome-Lysosome Fusion in Salmonella-Infected Murine Macrophages. (1990). FEMSMicrobiol.Immunol 2:89. Jiang, X; Rossanese, OW; Brown, NF; Kujat-Choy, S; Galan, JE; Finlay, BB; Brumell, JH. The Related Effector Proteins SopD and SopD2 From Salmonella Enterica Serovar Typhimurium Contribute to Virulence During Systemic Infection of Mice. (2004). Mol.Microbiol. 54:1186. Jones, BD; Ghori, N; Falkow, S. Salmonella Typhimurium Initiates Murine Infection by Penetrating and Destroying the Specialized Epithelial M Cells of the Peyer's Patches. (1994). J.Exp.Med. 180:15. Jones, MA; Wood, MW; Mullan, PB; Watson, PR; Wallis, TS; Galyov, EE. Secreted Effector Proteins of Salmonella Dublin Act in Concert to Induce Enteritis. (1998). Infect.Immun. 66:5799. Jones, PW; Dougan, G; Hayward, C; Mackensie, N; Collins, P; Chatfield, SN. Oral Vaccination of Calves Against Experimental Salmonellosis Using a Double Aro Mutant of Salmonella Typhimurium. (1991). Vaccine 9:29. Kaniga, K; Trollinger, D; Galan, JE. Identification of Two Targets of the Type III Protein Secretion System Encoded by the Inv and Spa Loci of Salmonella Typhimurium That Have Homology to the Shigella IpaD and IpaA Proteins. (1995a). J.Bacteriol. 177:7078. 109 Kaniga, K ; Tucker, S; Trollinger, D ; Galan, JE. Homologs of the Shigella IpaB and IpaC Invasins Are Required for Salmonella Typhimurium Entry into Cultured Epithelial Cells. (1995b). J.Bacteriol. 177:3965. Kingsley, R A and Baumler, AJ . Host Adaptation and the Emergence of Infectious Disease: the Salmonella Paradigm. (2000). Mol.Microbiol. 36:1006. Knodler, L A ; Celli, J; Hardt, W D ; Vallance, B A ; Yip, C; Finlay, B B . Salmonella Effectors Within a Single Pathogenicity Island Are Differentially Expressed and Translocated by Separate Type III Secretion Systems. (2002). Mol.Microbiol. 43:1089. Knodler, L A and Steele-Mortimer, O. The Salmonella Effector PipB2 Affects Late Endosome/Lysosome Distribution to Mediate Sif Extension. (2005). Mol.Biol.Cell 16:4108. Kuhle, V and Hensel, M . SseF and SseG Are Translocated Effectors of the Type III Secretion System of Salmonella Pathogenicity Island 2 That Modulate Aggregation of Endosomal Compartments. (2002). Cell. Microbiol. 4 :813. Kujat Choy, SL; Boyle, EC; Gal-Mor, O; Goode, DL; Valdez, Y ; Vallance, B A ; Finlay, B B . SseKl and SseK2 Are Novel Translocated Proteins of Salmonella Enterica Serovar Typhimurium. (2004). Infect.Immun. 72:5115. Lee, C A ; Silva, M ; Siber, A M ; Kelly, A J ; Galyov, E; McCormick, B A . A Secreted Salmonella Protein Induces a Proinflammatory Response in Epithelial Cells, Which Promotes Neutrophil Migration. (2000). Proc.Natl.Acad.Sci.U.S.A 97:12283. Leung, K Y and Finlay, B B . Intracellular Replication Is Essential for the Virulence of Salmonella Typhimurium. (1991). Proc.Natl.Acad.Sci.U.S.A 88:11470. Lodge, J; Douce, GR; Amin, II; Bolton, A J ; Martin, GD; Chatfield, S; Dougan, G; Brown, N L ; Stephen, J. Biological and Genetic Characterization of TnphoA Mutants of Salmonella Typhimurium T M L in the Context of Gastroenteritis. (1995). Infect.Immun. 63:762. Loeffler, F. Epidemieen Unter Den Im Hygienischen Institute Zu Griefswald Gehaltenen Mausen Und Uber Die BekaUeber Mpfung Der Feldmausplage. (1892). Zentralbl.Bakteriol.fOrig.A] 11:129. Lostroh, CP and Lee, CA. The Salmonella Pathogenicity Island-1 Type III Secretion System. (2001). Microbes Infect. 3:1281. Lyons, S; Wang, L; Casanova, JE; Sitaraman, SV; Merlin, D ; Gewirtz, AT. Salmonella Typhimurium Transcytoses Flagellin Via an SPI2-Mediated Vesicular Transport Pathway. (2004). J.Cell Sci. Makela, PH; Hovi, M ; Saxen, H ; Valtonen, M ; Valtonen, V . Salmonella, Complement and Mouse Macrophages. (1988). Immunol Lett. 19:217. 110 M a l o r n y , B ; Hoorfar , J ; Bunge , C ; H e l m u t h , R . Multicenter Validation of the Analytical Accuracy of Salmonella PCR: Towards an International Standard. (2003). Appl.Environ.Microbiol. 69:290. Maria thasan, S; N e w t o n , K ; M o n a c k , D M ; V u c i c , D ; French , D M ; L e e , W P ; R o o s e - G i r m a , M ; E r i c k s o n , S; D i x i t , V M . Differential Activation of the Inflammasome by Caspase-1 Adaptors ASC and Ipaf. (2004). Nature 430:213. Mast roen i , P and Sheppard, M . Salmonella Infections in the Mouse Model: Host Resistance Factors and in Vivo Dynamics of Bacterial Spread and Distribution in the Tissues. (2004). Microbes Infect. 6:398. M c C o r m i c k , B A ; C o l g a n , S P ; D e l p - A r c h e r , C ; M i l l e r , SI; Mada ra , J L . Salmonella Typhimurium Attachment to Human Intestinal Epithelial Monolayers: Transcellular Signalling to Subepithelial Neutrophils. (1993). J.Cell Biol. 123:895. M c C o r m i c k , B A ; H o f m a n , P M ; K i m , J ; Carnes, D K ; M i l l e r , SI; M a d a r a , J L . Surface Attachment of Salmonella Typhimurium to Intestinal Epithelia Imprints the Subepithelial Matrix With Gradients Chemotactic for Neutrophils. (1995a). J.Cell Biol. 131:1599. M c C o r m i c k , B A ; M i l l e r , SI ; Carnes, D ; Mada ra , J L . Transepithelial Signaling to Neutrophils by Salmonellae - A Novel Virulence Mechanism for Gastroenteritis. (1995b). Infect.Immun. 63:2302. M c C o r m i c k , B A ; Parkos , C A ; C o l g a n , S P ; Carnes, D K ; Mada ra , J L . Apical Secretion of a Pathogen-Elicited Epithelial Chemoattractant Activity in Response to Surface Colonization of Intestinal Epithelia by Salmonella Typhimurium. (1998). J.Immunol. 160:455. M c C o r m i c k , B A ; Stocker, B A ; L a u x , D C ; Cohen , P S . Roles of Motility, Chemotaxis, and Penetration Through and Growth in Intestinal Mucus in the Ability of an Avirulent Strain of Salmonella Typhimurium to Colonize the Large Intestine of Streptomycin-Treated Mice. (1988). Infect.Immun. 56:2209. M c g o v e r n , V J and Slavut in , L J . Pathology of Salmonella Colitis. (1979). Am. J. Surg. Path. 3:483. Meresse , S; Stee le-Mort imer , O ; F i n l a y , B B ; G o r v e l , J P . The R a b 7 GTPase Controls the Maturation of Salmonella Typhimurium-Containing Vacuoles in HeLa Cells. (1999). EMBO J. 18:4394. M i a o , E A ; Brit tnacher, M ; Haraga, A ; Jeng, R L ; W e l c h , M D ; M i l l e r , SI . Salmonella Effectors Translocated Across the Vacuolar Membrane Interact With the Actin Cytoskeleton. (2003). Mol.Microbiol. 48 :401. M i a o , E A ; Scherer, C A ; Tso l i s , R M ; K i n g s l e y , R A ; A d a m s , G ; Baumle r , A J ; M i l l e r , SI . Salmonella Typhimurium Leucine-Rich Repeat Proteins Are Targeted to the SPI1 and SPI2 Type III Secretion Systems. (1999). Mol.Microbiol. 34:850. I l l c Mills, DM; Bajaj, V ; Lee, CA. A 40 Kb Chromosomal Fragment Encoding Salmonella Typhimurium Invasion Genes Is Absent From the Corresponding Region of the Escherichia Coli K-12 Chromosome. (1995). Mol.Microbiol 15:749. Monack, DM; Bouley, DM; Falkow, S. Salmonella Typhimurium Persists Within Macrophages in the Mesenteric Lymph Nodes of Chronically Infected Nrampl+/+ Mice and Can Be Reactivated by IFNgamma Neutralization. (2004). J.Exp.Med. 199:231. Monack, DM; Hersh, D; Ghori, N ; Bouley, D; Zychlinsky, A; Falkow, S. Salmonella Exploits Caspase-1 to Colonize Peyer's Patches in a Murine Typhoid Model. (2000). J.Exp Med. 192:249. Murray, RA and Lee, CA. Invasion Genes Are Not Required for Salmonella Enterica Serovar Typhimurium to Breach the Intestinal Epithelium: Evidence That Salmonella Pathogenicity Island 1 Has Alternative Functions During Infection. (2000). Infect.Immun. 68:5050. Nadeau, WJ; Pistole, TG; McCormick, BA. Polymorphonuclear Leukocyte Migration Across Model Intestinal Epithelia Enhances Salmonella Typhimurium Killing Via the Epithelial Derived Cytokine, IL-6. (2002). Microbes Infect. 4:1379. Nakoneczna, I and Hsu, HS. The Comparative Histopathology of Primary and Secondary Lesions in Murine Salmonellosis. (1980). Br.J.Exp Pathol. 61:76. National Center for Infectious Disease, CfDC. Salmonella Infection (Salmonellosis). (2006). www, cdc. sov/ncidod/diseases/submenus/sub salmonella, htm Ochman, H and Groisman, EA. The Origin and Evolution of Species Differences in Escherichia Coli and Salmonella Typhimurium. (1994). EXS 69:479. Ochman, H and Groisman, EA. Distribution of Pathogenicity Islands in Salmonella Spp. (1996). Infect.Immun. 64:5410. Ochman, H; Soncini, FC; Solomon, F; Groisman, EA. Identification of a Pathogenicity Island Required for Salmonella Survival in Host Cells. (1996). Proc.Natl.Acad.Sci.U.S.A 93:7800. Ohlson, MB; Fluhr, K ; Birmingham, CL; Brum ell, JH; Miller, SI. SseJ Deacylase Activity by Salmonella Enterica Serovar Typhimurium Promotes Virulence in Mice. (2005). Infect.Immun. 73:6249. Parkos, CA; Colgan, SP; Diamond, MS; Nusrat, A; Liang, TW; Springer, TA; Madara, JL. Expression and Polarization of Intercellular Adhesion Molecule-1 on Human Intestinal Epithelia: Consequences for CDllb/CD18-Mediated Interactions With Neutrophils. (1996). Mol.Med. 2:489. Parry, CM; Hien, TT; Dougan, G; White, NJ; Farrar, JJ. Typhoid Fever. (2002). N.Engl.J.Med. 347:1770. 112 Patarroyo, M and M a k g o b a , M W . Leucocyte Adhesion to Cells in Immune and Inflammatory Responses. (1989). Lancet 2:1139. Patel , J C ; Rossanese, O W ; Ga lan , J E . The Functional Interface Between Salmonella and Its Host Cell: Opportunities for Therapeutic Intervention. (2005). Trends Pharmacol.Sci. 26:564. Pathmanathan, S G ; Cardona-Castro , N ; Sanchez-J imenez, M M ; Cor rea-Ochoa , M M ; Puthucheary, S D ; T h o n g , K L . Simple and Rapid Detection of Salmonella Strains by Direct PCR Amplification of the HilA Gene. (2003). J.Med.Microbiol. 52:773. Penheiter, K L ; Mathur , N ; G i l e s , D ; Fah len , T ; Jones, B D . Non-Invasive Salmonella Typhimurium Mutants Are Avirulent Because of an Inability to Enter and Destroy M Cells of Ileal Peyer's Patches. (1997). Mol.Microbiol. 24:697. Perel le , S; Di lasser , F ; M a l o r n y , B ; Grout , J ; Hoorfar , J ; Fach , P . Comparison of PCR-ELISA and LightCycler Real-Time PCR Assays for Detecting Salmonella Spp. in Milk and Meat Samples. (2004). Mol. Cell Probes 18:409. Que, J U ; Casey, S W ; Hentges, D J . Factors Responsible for Increased Susceptibility of Mice to Intestinal Colonization After Treatment With Streptomycin. (1986). Infect.Immun. 53:116. Que , J U and Hentges, D J . Effect of Streptomycin Administration on Colonization Resistance to Salmonella Typhimurium in Mice. (1985). Infect.Immun. 48:169. Raffate l lu , M ; W i l s o n , R P ; Chessa, D ; A n d r e w s - P o l y m e n i s , H ; Tran , Q T ; L a w h o n , S; K h a r e , S; A d a m s , L G ; Baumle r , A J . SipA, SopA, SopB, SopD, and SopE2 Contribute to Salmonella Enterica Serotype Typhimurium Invasion of Epithelial Cells. (2005). Infect.Immun. 73:146. R a h n , K ; D e Grandis , S A ; C la rke , R C ; M c e w e n , S A ; G a l a n , J E ; G i n o c c h i o , C ; Curt iss , R , III; G y l e s , C L . Amplification of an InvA Gene Sequence of Salmonella Typhimurium by Polymerase Chain Reaction As a Specific Method of Detection of Salmonella. (1992). Mol.CellProbes 6:271. Resc igno , M ; Urbano , M ; V a l z a s i n a , B ; F r a n c o l i n i , M ; Rot ta , G ; Bonas io , R ; Granucc i , F ; Kraehenbuhl , J P ; R icc ia rd i -Cas tagno l i , P . Dendritic Cells Express Tight Junction Proteins and Penetrate Gut Epithelial Monolayers to Sample Bacteria. (2001). Nat.Immunol 2 :361. Richter -Dahl fors , A ; B u c h a n , A M ; F i n l a y , B B . Murine Salmonellosis Studied by Confocal Microscopy: Salmonella Typhimurium Resides Intracellular^ Inside Macrophages and Exerts a Cytotoxic Effect on Phagocytes in Vivo. (1997). J.Exp Med. 186:569. R i n g s , D M . Salmonellosis in Calves. (1985). Vet.Clin North Am.FoodAnim Pract. 1:529. Rol lenhagen , C and B u m a n n , D . Salmonella Enterica Highly Expressed Genes Are Disease Specific. (2006). Infect.Immun. 74:1649. ' 113 Rosenberger, C M ; Gallo, R L ; Finlay, B B . Interplay Between Ant ibacter ia l Effectors: a Macrophage An t imic rob ia l Peptide Impairs Intracellular Salmonella Replicat ion. (2004). Proc.Natl.Acad.Sci.U.S.A 101:2422. Royle, M C ; Totemeyer, S; Alldridge, L C ; Maskel l , D J ; Bryant, C E . Stimulation of T o l l - L i k e Receptor 4 by Lipopolysaccharide Dur ing Cel lu la r Invasion by L i v e Salmonella T y p h i m u r i u m Is a C r i t i c a l but Not Exclusive Event Lead ing to Macrophage Responses. (2003). J.Immunol 170:5445. Ruiz-Albert, J; Mundy, R; Y u , X J ; Beuzon, C R ; Holden, D W . SseA Is a Chaperone for the SseB and SseD Translocon Components of the Salmonella Pathogenicity-Island-2-Encoded Type III Secretion System. (2003). Microbiology-Sgm 149:1103. Salzman, N H ; Ghosh, D ; Huttner, K M ; Paterson, Y ; Bevins, C L . Protection Against Enter ic Salmonellosis in Transgenic M i c e Expressing a H u m a n Intestinal Defensin. (2003). Nature 422:522. Santos, R L ; Tsolis, R M ; Baumler, A J ; Adams, L G . Pathogenesis of Salmonella-Induced Enteri t is . (2003). Braz.J.Med.Biol.Res. 36:3. Santos, R L ; Tsolis, R M ; Zhang, S; Ficht, T A ; Baumler, A J ; Adams, L G . Salmonella-Induced C e l l Death Is Not Required for Enterit is in Calves. (2001a). Infect.Immun. 69:4610. Santos, R L ; Zhang, S; Tsolis, R M ; Kingsley, R A ; Adams, L G ; Baumler, A J . A n i m a l Models of Salmonella Infections: Enterit is Versus Typhoid Fever. (2001b). Microbes Infect. 3:1335. Scholz, H C ; Arnold, T; Marg, H ; Rosier, U ; Hensel, A . Improvement of an InvA-Based P C R for the Specific Detection of Salmonella T y p h i m u r i u m in Organs of Pigs. (2001). Berl Munch. Tier arztl.Wochenschr. 114:401. Shea, JE; Beuzon, C R ; Gleeson, C ; Mundy, R; Holden, D W . Influence of the Salmonella T y p h i m u r i u m Pathogenicity Island 2 Type III Secretion System on Bacter ial Growth in the Mouse. (1999). Infect.Immun. 67:213. Shea, JE; Hensel, M ; Gleeson, C ; Holden, D W . Identification of a Virulence Locus Encoding a Second Type III Secretion System in Salmonella T y p h i m u r i u m . (1996). Proc.Natl.Acad.Sci.U.S.A 92>:2592>. Shirai, Y ; Sunakawa, K ; Ichihashi, Y ; Yamaguchi, H . A Morpholog ica l Study in Germfree M i c e (Salmonella Infection). (1979). Exp Pathol.(Jena) 17:158. Siegmund, B ; Lehr, H A ; Fantuzzi, G ; Dinarello, C A . I L - l Beta -Convert ing Enzyme (Caspase-1) in Intestinal Inflammation. (2001). Proc.Natl.Acad.Sci.U.S.A'98:13249. Silva, M ; Song, C; Nadeau, W J ; Matthews, J B ; McCormick, B A . Salmonella T y p h i m u r i u m SipA-Induced Neutrophi l Transepithelial Mig ra t i on : Involvement of a P K C - A l p h a -Dependent Signal Transduction Pathway. (2004). Am.J.Physiol.-Gastrointest. Liver Physiol. 286:G1024. 114 Spr inz , H ; Gangarosa, E J ; W i l l i a m s , M ; H o r n i c k , RJ3; W o o d w a r d , T E . Histopathology of the Upper Small Intestines in Typhoid Fever. Biopsy Study of Experimental Disease in Man. (1966). AmJ.Dig.Dis. 11:615. Stecher, B ; Hapfelmeier , S; M u l l e r , C ; Kremer , M ; Sta l lmach, T ; Hardt , W D . Flagella and Chemotaxis Are Required for Efficient Induction of Salmonella Enterica Serovar Typhimurium Colitis in Streptomycin-Pretreated Mice. (2004). Infect.Immun. 72:4138. Stecher, B ; Macphe r son , A J ; Hapfelmeier , S; Kremer , M ; Sta l lmach, T ; Hardt , W D . Comparison of Salmonella Enterica Serovar Typhimurium Colitis in Germfree Mice and Mice Pretreated With Streptomycin. (2005). Infect.Immun. 73:3228. Stee le-Mort imer , O ; Meresse , S; G o r v e l , J P ; T o h , B H ; F i n l a y , B B . Biogenesis of Salmonella Typhimurium-Containing Vacuoles in Epithelial Cells Involves Interactions With the Early Endocytic Pathway. (1999). Cell Microbiol. 1:33. Stein, M A ; L e u n g , K Y ; Z w i c k , M ; Garc ia -de l Po r t i l l o , F ; F i n l a y , B B . Identification of a Salmonella Virulence Gene Required for Formation of Filamentous Structures Containing Lysosomal Membrane Glycoproteins Within Epithelial Cells. (1996). Mol.Microbiol. 20:151. Stocker, B A ; Hoise th , S K ; Smi th , B P . Aromatic-Dependent "Salmonella Sp." As Live Vaccine in Mice and Calves. (1983). Dev.Biol.Stand. 53:47. Tso l i s , R M ; A d a m s , L G ; F ich t , T A ; Baumle r , A J . Contribution of Salmonella Typhimurium Virulence Factors to Diarrheal Disease in Calves. (1999a). Infect.Immun. 67:4879. Tso l i s , R M ; K i n g s l e y , R A ; Townsend , S M ; F ich t , T A ; A d a m s , L G ; Baumle r , A J . Of Mice, Calves, and Men. Comparison of the Mouse Typhoid Model With Other Salmonella Infections. (1999b). Adv.Exp.Med.Biol. 473:261. Tso l i s , R M ; Townsend , S M ; M i a o , E A ; M i l l e r , SI; F ich t , T A ; A d a m s , L G ; Baumle r , A J . Identification of a Putative Salmonella Enterica Serotype Typhimurium Host Range Factor With Homology to IpaH and YopM by Signature-Tagged Mutagenesis. (1999c). Infect.Immun. 67:6385. U c h i y a , K ; Barb ie r i , M A ; Funato, K ; Shah, A H ; Stahl , P D ; Gro i sman , E A . A Salmonella Virulence Protein That Inhibits Cellular Trafficking. (1999). Embo Journal 18:3924. U c h i y a , K ; Gro i sman , E A ; N i k a i , T . Involvement of Salmonella Pathogenicity Island 2 in the Up-Regulation of Interleukin-10 Expression in Macrophages: Role of Protein Kinase a Signal Pathway. (2004). Infect.Immun. 72:1964. U c h i y a , K and N i k a i , T . Salmonella Enterica Serovar Typhimurium Infection Induces Cyclooxygenase 2 Expression in Macrophages: Involvement of Salmonella Pathogenicity Island 2. (2004). Infect.Immun. 72:6860. 115 Uchiya, K and Nika i , T . Salmonella Pathogenicity Island 2-Dependent Expression of Suppressor of Cytokine Signaling 3 in Macrophages. (2005). Infect.Immun. 73:5587. v a n der Velden, A W ; Baumler, A J ; Tsolis, R M ; Heffron, F . Mul t ip l e F i m b r i a l Adhesins A r e Required for F u l l Virulence of Salmonella T y p h i m u r i u m in M i c e . (1998). Infect.Immun. 66:2803. van der Velden, A W M ; Lindgren, S W ; Worley, M J ; Heffron, F . Salmonella Pathogenicity Island 1-Independent Induction of Apoptosis in Infected Macrophages by Salmonella Enter ica Serotype Typh imur ium. (2000). Infect.Immun. 68:5702. Vazquez-Torres, A ; Jones-Carson, J; Baumler, A J ; Falkow, S; Valdivia , R ; Brown, W ; L e , M ; Berggren, R ; Parks, W T ; Fang, F C . Extraintestinal Dissemination of Salmonella by CD18-Express ing Phagocytes. (1999). Nature 401:804. Vazquez-Torres, A ; X u , Y S ; Jones-Carson, J; Holden, D W ; Lucia, S M ; Dinauer, M C ; Mastroeni, P ; Fang, F C . Salmonella Pathogenicity Island 2-Dependent Evasion of the Phagocyte N A D P H Oxidase. (2000). Science 287:1655. Vida l , S; Tremblay, M L ; Govoni, G ; Gauthier, S; Sebastiani, G ; Malo , D ; Skamene, E ; Olivier, M ; Jothy, S; Gros, P . The I ty /Lsh/Bcg Locus : Natura l Resistance to Infection W i t h Intracellular Parasites Is Abrogated by Disrupt ion of the N r a m p l Gene. (1995). J.Exp.Med. 182:655. Villarreal-Ramos, B ; Manser, J; Collins, R A ; Dougan, G ; Chatfield, S N ; Howard, CJ . Immune Responses in Calves Immunised Ora l l y or Subcutaneously W i t h a L i v e Salmonella T y p h i m u r i u m A r o Vaccine. (1998). Vaccine 16:45. Wallis , T S and Galyov, E E . Molecu la r Basis of Salmonella-Induced Enteri t is . (2000). Mol.Microbiol. 36:997. Wall is , T S ; Hawker, R J ; Candy, D C ; Q i , G M ; Clarke, G J ; Worton, K J ; Osborne, M P ; Stephen, J. Quantification of the Leucocyte Influx into Rabbi t Ileal Loops Induced by Strains of Salmonella T y p h i m u r i u m of Different Virulence. (1989). J.Med.Microbiol. 30:149. Wall is , T S ; Vaughan, A T ; Clarke, G J ; Q i , G M ; Worton, K J ; Candy, D C ; Osborne, M P ; Stephen, J. The Role of Leucocytes in the Induction of F l u i d Secretion by Salmonella T y p h i m u r i u m . (1990). J.Med.Microbiol. 31:27. Wall is , T S ; W o o d , M ; Watson, P ; Paulin, S; Jones, M ; Galyov, E . Sips, Sops, and SPIs but Not S T N Influence Salmonella Enteropathogenesis. (1999a). Mech. Pathogen. Enteric Dis. 2 473:275. Wallis , T S ; W o o d , M ; Watson, P ; Paulin, S; Jones, M ; Galyov, E . Sips, Sops, and SPIs but Not Stn Influence Salmonella Enteropathogenesis. (1999b). Adv.Exp Med.Biol. 473:275. Waterman, S R and Holden, D W . Functions and Effectors of the Salmonella Pathogenicity Island 2 Type III Secretion System. (2003). Cell. Microbiol. 5:501. 116 Watson , P R ; G a l y o v , EE; P a u l i n , S M ; Jones, P W ; W a l l i s , T S . Mutation of InvH, but Not Stn, Reduces Salmonella-Induced Enteritis in Cattle. (1998). Infect.Immun. 66:1432. Watson , P R ; P a u l i n , S M ; B l a n d , A P ; Jones, P W ; W a l l i s , T S . Characterization of Intestinal Invasion by Salmonella Typhimurium and Salmonella Dublin and Effect of a Mutation in the InvH Gene. (1995). Infect.Immun. 63:2743. W i n o k u r , P L . Molecular Epidemiological Techniques for Salmonella Strain Discrimination. (2003). Front Biosci. 8 : c l 4 . W o o d , M W ; Jones, M A ; Wat son , P R ; Hedges, S; W a l l i s , T S ; G a l y o v , EE. Identification of a Pathogenicity Island Required for Salmonella Enteropathogenicity. (1998). Mol.Microbiol. 29:883. W o o d , M W ; Jones, M A ; Watson , P R ; Siber, A M ; M c C o r m i c k , B A ; Hedges , S; Rosqvis t , R ; W a l l i s , T S ; G a l y o v , EE. The Secreted Effector Protein of Salmonella Dublin, SopA, Is Translocated into Eukaryotic Cells and Influences the Induction of Enteritis. (2000). Cell Microbiol. 2:293. W o r l d Hea l th Organiza t ion . The World Health Report. Making a Difference. (1999). Zentralbl.BakteriolfB] W o r l d Hea l th Organiza t ion . Drug-Resistant Salmonella. (2005). ht tp: / /www.who.int/mediacentre/factsheets/fsl39/en/print .html. W r a y , C and Sojka , W J . Experimental Salmonella Typhimurium Infection in Calves. (1978). Res.Vet.Sci. 25:139. W y l l i e , S; Seu, P ; G a o , F Q ; Gros , P ; Goss , J A . Disruption of the Nrampl (Also Known As Slcllal) Gene in Kupffer Cells Attenuates Early-Phase, Warm Ischemia-Reperfusion Injury in the Mouse Liver. (2002). J.Leukoc.Biol. 72:885. Y r l i d , U ; Svensson, M ; Hakansson , A ; Chambers , B J ; Ljunggren, H G ; W i c k , M J . In Vivo Activation of Dendritic Cells and T Cells During Salmonella Enterica Serovar Typhimurium Infection. (2001). Infect.Immun. 69:5726. Y u , X J ; R u i z - A l b e r t , J ; Unswor th , KE; Ga rv i s , S; L i u , M ; H o l d e n , D W . SpiC Is Required for Secretion of Salmonella Pathogenicity Island 2 Type III Secretion System Proteins. (2002). Cell Microbiol. 4 :531. Y u , Y ; Z e n g , H ; L y o n s , S;. Car l son , A ; M e r l i n , D ; N e i s h , A S ; G e w i r t z , A T . TLR5-Mediated Activation of P38 MAPK Regulates Epithelial IL-8 Expression Via Posttranscriptional Mechanism. (2003). Am.J.Physiol Gastrointest.Liver Physiol 285 :G282 . Z e n g , H ; Car l son , A Q ; G u o , Y W ; Y u , Y M ; C o l l i e r - H y a m s , L S ; M a d a r a , J L ; G e w i r t z , A T ; N e i s h , A S . Flagellin Is the Major Proinflammatory Determinant of Enteropathogenic Salmonella. (2003). J. Immunol. 171:3668. 117 Zeng, H ; W u , H ; Sloane, V ; Jones, R; Y u , Y ; L i n , P; Gewirtz, A T ; Neish, A S . F l a g e l l i n / T L R 5 Responses in Epi thel ia Reveal Intertwined Act ivat ion of Inflammatory and Apoptot ic Pathways. (2006). Am.J.Physiol Gastrointest.Liver Physiol 290:G96. Zhang, S; Kingsley, R A ; Santos, R L ; Andrews-Polymenis, H ; Raffatellu, M ; Figueiredo, J; Nunes, J; Tsolis, R M ; Adams, L G ; Baumler, A J . Molecu la r Pathogenesis of Salmonella Enter ica Serotype Typhimurium-Induced Diarrhea. (2003). Infect.Immun. 71:1. Zhang, S; Santos, R L ; Tsolis, R M ; Miro ld , S; Hardt, W D ; Adams, L G ; Baumler, A J . Phage Mediated Hor izonta l Transfer of the S o p E l Gene Increases Enteropathogenicity of Salmonella Enter ica Serotype T y p h i m u r i u m for Calves. (2002a). FEMS Microbiol.Lett. 217:243. Zhang, SP; Santos, R L ; Tsolis, R M ; Stender, S; Hardt, W D ; Baumler, A J ; Adams, L G . The Salmonella Enter ica Serotype T y p h i m u r i u m Effector Proteins S i p A , SopA, SopB, SopD, and SopE2 A c t in Concert to Induce Diar rhea in Calves. (2002b). Infect.Immun. 70:3843. Zhou, D ; Chen, L M ; Hernandez, L ; Shears, S B ; Galan, JE. A Salmonella Inositol Polyphosphatase Acts in Conjunction W i t h Other Bacter ial Effectors to Promote Host C e l l A c t i n Cytoskeleton Rearrangements and Bacter ial Internalization. (2001). Mol.Microbiol. 39:248. Zhou, D ; Mooseker, M S ; Galan, JE. Role of the S. T y p h i m u r i u m Ac t in -B ind ing Protein S i p A in Bacter ial Internalization. (1999). Science 283:2092. 118 Appendix 1: List of Peer-reviewed published primary research 1. Zahar ik M L , C u l l e n V L , F u n g A M , L i b b y S J , Kuja t C h o y S L , C o b u r n B . Kehres D G , M a g u i r e M E , F a n g F C , F i n l a y B B . The Salmonella enterica serovar typhimurium divalent cation transport systems MntH and SitABCD are essential for virulence in an NramplG169 murine typhoid model. Infect Immun. 2004;72:5522-5. 2. C o b u r n B , L i Y , O w e n D , V a l l a n c e B , F i n l a y B B . Salmonella typhimurium pathogenicity island 2 is necessary for complete virulence in a mouse model of infectious colitis. Infect Immun. 2005; 73:3219-27. 3. C o b u r n B*. Coombes B K * , Potter A A , G o m i s S, M i r a k h u r K , L i Y , F i n l a y B B . Analysis of the contribution of salmonella pathogenicity islands 1 and 2 to enteric disease progression using a novel bovine ileal loop model and a murine model of infectious enterocolitis. Infect Immun. 2005; 73:7161-9. 4. B r o w n N F , V a l l a n c e B A , Coombes B K , V a l d e z Y , C o b u r n B A , F i n l a y B B Salmonella Pathogenicity Island 2 Is Expressed Prior to Penetrating the Intestine. PLoS Pathogens 2005; 1 (e32): 0001-0007. 5. Dimaras H , C o b u r n B , Pa jovic S, C h e n K , G a l l i e B L . Loss of p75 Neurotrophin Receptor Expression Accompanies Malignant Progression to Human and Murine Retinoblastoma. Mol. Carcinogenesis. Epub ahead of print. 6. B r u n h a m L R , K r u i t J K , Iqbal J , F ievet C , T i m m i n s J M , Pape T M , C o b u r n B A , B i s sada N , Staels B , G r o e n A K , H u s s a i n M M , Parks J S , K u i p e r s F , H a y d e n M R . Tissue-Specific Deletion of Intestinal ABCA1 Reveals that the Intestine Directly Contributes to Plasma HDL Cholesterol Levels. J Clin Invest. 2006; 116(4): 1052-1062 * Author s Contr ibuted equal ly to this w o r k In press 7. W i c k h a m M E , L u p p C , Mascarenhas M , V a z q u e z A , Coombes B K , B r o w n N F , C o b u r n B A , D e n g W , Puente J L , K a r m a l i M A , F i n l a y B B . Genetic determinants of outbreaks and Haemolytic-Uraemic Syndrome following non-0157 STEC infection J Infect Dis. N . B . Three addi t ional papers have been submitted for considerat ion. 119 Appendix 2: Animal Ethical Approvals Canadian C o u n c i l o n A n i m a l Care /Nat iona l Institutional A n i m a l T ra in ing Certif icate N u m b e r N u m b e r : 0173 Trainee: Bryan Coburn A n i m a l Care E t h i c a l A p p r o v a l Certificates: 1) Number : A04-0058 Pr inc ip l e Investigator: Brett Finlay Ti t l e : Bacterial Infections and Innate Immunity 2) Number : A04-0280 Pr inc ip l e Investigator: Brett Finlay Ti t le : Genomics of Infection: Salmonella and Pathogenic E. Coli 120 

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.831.1-0100343/manifest

Comment

Related Items