INTRODUCTION

Horizontal gene transfer and bacterial virulence

Bacteria show a remarkable ability to adapt rapidly to new habitats. This observation also applies to pathogenic bacteria that have evolved strategies to colonize various anatomical niches of their multi-cellular hosts. The acquisition of genetic material by a process termed horizontal gene transfer is considered to be a driving force for the rapid evolution of bacteria as pathogens. Extrachromosomal DNA such as plasmids conferring resistance to antibiotics were the first horizontally transferred DNA elements to be identified, but later it became obvious that there are also mechanisms that allow the horizontal transfer of chromosomal DNA elements. The observation that certain virulence functions are clustered in distinct regions of the chromosome and that these regions are genetically unstable and were deleted with high frequencies gave the first clue about the existence of the new form of genetic elements. The term ‘Pathogenicity Island’ (PAI) was first introduced in 1983 by Hacker and colleagues who observed genetic instability of genes associated with the haemolytic activity of uropathogenic strains of Escherichia coli (Hacker et al., 1983). PAI were initially defined as large, unstable regions of the chromosome. With the identification of a large number of additional PAIs in various groups of bacterial pathogens, further common features were found.

Definition of pathogenicity islands

The following common characteristics were defined for PAIs:

• They are large, distinct genetic entities on the bacterial chromosome.

• They harbor one or more virulence genes.

• They are often genetically unstable. This feature correlates with the presence of genetic elements involved in DNA-mobility, such as direct repeats, integrases, transposases and bacteriophage genes.

• […]

INTRODUCTION

Dendritic cells (DC) are efficient antigen-presenting cells and are likely to be involved in the initiation of T-cell responses to Salmonella. However, it is not known what type of DC initiate immune responses to Salmonella or where this initiation takes place. Studies on interactions between Salmonella and DC are emerging and are shedding light on this topic. This chapter will review how Salmonella interacts with DC, following the course the bacteria take after oral infection. One of the earliest sites of Salmonella replication is within the Peyer's patches of the gut. Thereafter, Salmonella can be found in the gut-draining mesenteric lymph nodes. After systemic release of bacteria or bacteria-containing cells, Salmonella spread to the spleen and liver and replicate further. The relevance of the interactions between Salmonella and DC in these organs for initiating antibacterial T-cell responses is discussed. This is preceded by a brief overview of the biology of DC.

DENDRITIC CELLS

DC originate from precursors in the bone marrow and were named because of their morphology having long, branched dendrites (Steinman and Cohn, 1973; Steinman et al., 1974). DC are widely distributed in lymphoid as well as non-lymphoid tissues (Steptoe et al., 2000; Vremec and Shortman, 1997; Steiniger et al., 1984).

INTRODUCTION

Typhoid and paratyphoid fever result from infection with Salmonella enterica serovars Typhi and Paratyphi respectively. Humans are the only reservoir of these infections that are spread by the fecal-oral route. The control and near elimination of this disease in Western countries has been achieved largely because of improved sanitation, surveillance, contact tracing and successful therapy. In locations where this infrastructure does not exist, vaccines can be used as one measure to control the incidence of typhoid fever (Tarr et al., 1999) and possibly even to contribute to the eventual eradication of the disease.

Non-typhoidal S. enterica infections are a major public health problem world-wide and reduction of the incidence of these diseases presents quite different challenges to reducing the incidence of typhoid fever. In fact, these diseases have several animal reservoirs and, in humans, a large number of different S. enterica serovars cause gastroenteritis probably making vaccines very difficult to realize and/or use commercially.

This chapter will outline the current development of vaccines that address disease caused by S. enterica, with an emphasis on typhoid fever.

TYPHOID VACCINES

A short history of typhoid vaccines

Typhoid fever vaccine development began shortly after the so-called “Golden Age” of microbiology, at the end of the 19th Century. The initial discovery of the typhoid bacillus was made by Eberth in 1880 and it was first isolated from stools by Pfeiffer in 1896 (Warren and Hornick, 1979).

INTRODUCTION

Salmonella infections in humans can range from a self-limiting gastroenteritis, usually associated with non-typhoidal Salmonella (NTS), to typhoid fever with complications such as a fatal intestinal perforation. The World Health Organization (WHO) estimates that the annual global incidence of typhoid fever is about 21 million cases with a mortality of 1% (Crump et al., 2004). This may be an underestimate because typhoid is predominantly a disease of developing countries, where not all cases present to the healthcare services and data collection may be difficult. In addition, financial constraints limit outbreak investigation and antibiotics are often widely available without prescription. Not only does this compound problems with data gathering, but it is likely to add to the burden of resistant disease circulating in the community. The situation is even less clear for NTS because most patients do not need to consult the health services. Despite this, as reported in 1999 in the USA alone there were an estimated 1.4 million cases of NTS infection annually, resulting in approximately 600 deaths (Mead et al., 1999).

There is no doubt that antibiotic resistance in Salmonella infections poses a major threat to human health, especially in cases of invasive NTS in immunocompromised patients and in typhoid fever. The cost of resistance in human terms is shown in Table 2.1. There is also a potential increase in the cost of food production.

Salmonella enterica encompasses a diverse range of bacteria that cause a spectrum of diseases in many hosts. Typhoid fever is still a major killer of people in the developing world and rears its ugly head whenever war or natural disaster strikes. The increase in antibiotic resistance that has been observed in S. enterica serovar Typhi makes the understanding of this pathogen ever more important. But typhoid fever is not the only Salmonella-related disease that causes concern, with human gastrointestinal disease a major problem in developed and developing countries, not forgetting salmonelloses in livestock that bring with them economic losses as well as the problems of zoonoses and food-borne disease.

The different salmonellae make up a versatile and fascinating group of organisms that have inspired both of the Editors of this book since we were scientific juveniles studying the pathogenesis and immunity of these bacteria for our Ph. D. degrees. As we have moved through the stages of our scientific careers, other bacteria and immunological questions may have caught our attention for a while, but always the salmonellae persisted, providing the bedrock of our interests and the centrepiece of our scientific enquiries.

So why edit a book on salmonellae now? The easy answer to this question is that the study of the salmonellae is entering a brave new world with the completion of the genome sequences of serovars Typhi, Paratyphi A and Typhimurium, with other sequences hot on their tail.

INTRODUCTION

Salmonella enterica affects humans and animals worldwide. It can be found in sewage-, sea-, and river- water and can contaminate food. Asymptomatic carriage in domestic animals can result in the introduction of the bacteria into the food chain.

Interest in understanding the mechanisms of pathogenesis and immunity that operate in S. enterica infections is twofold. Firstly, development of vaccines against salmonellosis has been too empirical due to insufficient understanding of how the host controls these infections, and how the bacteria evade immune surveillance. The fact that S. enterica-based vaccines are also being evaluated as systems to deliver recombinant antigens or DNA vaccines to the immune system and as new tools for the therapy of cancer has further increased the need to study how these vaccines work (Chabalgoity et al., 2002; Mastroeni et al., 2001; Reisfeld et al., 2004).

Secondly, S. enterica provides a model to understand how bacterial pathogens interact with the immune system. S. enterica is an intriguing bacterium in the way it interacts with the immune system and the immunological requirements for host resistance to this bacterium are affected by a very large number of variables.

MODELS FOR THE STUDY OF IMMUNITY TO S. ENTERICA

The study of the immunobiology of S. enterica infections has been facilitated by the availability of reliable models and by improved genetic tools that allow identification of polymorphic differences or mutations in genes involved in immune functions.

INTRODUCTION

Any determinant that enables a Salmonella serotype to enter a host, to find a unique niche to multiply, to avoid or subvert the host defenses, to cause disease and to be transmitted to the next susceptible host may be considered a virulence determinant. Essential genes required for growth in standard laboratory medium are usually not included under this broad definition of virulence genes. The total number of virulence genes present in the Salmonella genome can be estimated by screening a bank of mutants generated by random transposon mutagenesis using an animal model of infection. The genome of Salmonella enterica serovar Typhimurium strain LT2 contains 4552 intact open reading frames (McClelland et al., 2001). Of these, approximately 490 genes are essential during growth in rich medium (Knuth et al., 2004). Thus, only the function of about 4062 genes is assessed when S. enterica serovar Typhimurium transposon mutants are generated and analyzed. Analysis of 197 randomly generated transposon mutants of S. enterica serovar Typhimurium for virulence in mice upon intra gastric infection identified 8 mutants that were more than 1000 fold attenuated (Bowe et al., 1998). Extrapolating to the actual number of intact genes present in the genome (4062 genes) this study suggests that mutations in approximately 165 S. enterica serovar Typhimurium genes result in attenuation of more than 1,000-fold in mice.

INTRODUCTION

Typhoid fever is an acute systemic infection caused by the bacterium Salmonella enterica serovar Typhi. Salmonella enterica serovars Paratyphi A, B, and C cause the clinically similar condition, paratyphoid fever. Typhoid and paratyphoid fevers are collectively referred to as enteric fevers. In most endemic areas, approximately 90% of enteric fever is typhoid. Typhoid is transmitted by the fecal-oral route via contaminated food and water and is therefore common where sanitary conditions are inadequate and access to clean water is limited. Although typhoid fever was common in the United States and Europe in the 19th century, it is now encountered mostly throughout the developing world. In the last fifteen years, the emergence of resistance to the antibiotics used for treatment has led to large epidemics, and complicated the management of this serious disease.

Before the 19th century, typhoid fever was commonly confused with other prolonged febrile syndromes, particularly typhus fever. Following the observations of Huxham, Louis, Bretonneau, Gerhard and William Jenner, by the middle of the 19th century the two conditions were clearly differentiated (Richens, 1996). In 1873, William Budd described the contagious nature of the disease and incriminated fecally contaminated water sources in transmission. The causative organism was visualized in tissue sections from Peyer's patches and spleens of infected patients by Eberth in 1880 and was grown in pure culture by Gaffky in 1884. The organism has been variously known as Bacillus typhosus, Erbethella typhosa, Salmonella typhosa and Salmonella typhi.

INTRODUCTION AND HISTORICAL PERSPECTIVE

Salmonella enterica is a food-borne pathogen of global significance and no population is spared. It has been reported that there are over 1.3 billion cases of human salmonellosis annually worldwide, with three million deaths (Pang et al., 1995). More recent data suggest that each year in the USA there are 1.3 million cases of human S. enterica infections with 600 deaths. Other recent work from Denmark claims that mortality rates of people who have had S. enterica are three times those of controls, in the year after infection (Helms et al., 2003). Infection with S. enterica can lead to long-term sequelae such as irritable bowel syndrome and reactive arthritis (Rees et al., 2004). There are clear public health benefits to be had from better control of these bacteria in the food chain. In addition, S. enterica serovars have an enormous economic impact (Roberts et al., 2003; Voetsch et al., 2004), and are the most important foodborne pathogens in terms of deaths caused (Adak et al., 2002; Kennedy et al., 2004; Mead et al., 1999).

Gaffky may have been the first person to isolate S. enterica micro-organisms. These were S. enterica serovar Typhi from an infected patient, in 1884. Around this time it was also known that similar bacteria could cause non-typhoid disease in both animals and humans. In 1885, two American veterinarians, Salmon and Smith, isolated the bacterium causing ‘hog cholera’ from infected pigs.

INTRODUCTION

Salmonella enterica has been proposed as a highly efficient vector for the delivery of heterologous molecules to the immune system of the host. For more than two decades recombinant live attenuated salmonellae expressing antigens from other pathogens have been extensively assessed as oral multivalent vaccines and tested in a great diversity of experimental models. More recently, it has been demonstrated that S. enterica can also be used as a vector for DNA vaccines. New emerging applications for recombinant S. enterica include its use in the treatment of cancer and possible applications in gene therapy.

Some distinctive features of S. enterica have strongly contributed to make it an attractive delivery system. Among them are features of the immunobiology of S. enterica infections and the genetics of the bacteria. S. enterica naturally enter the host by the oral route, elicit strong mucosal and systemic immune responses and are eventually cleared from the tissues leaving long lasting immunological memory. Once inside the host, S. enterica can be found within macrophages and dendritic cells (DC), which are professional antigen presenting cells (APC). Thus, oral administration of recombinant S. enterica can be an effective way of directing the expression of relevant molecules (antigens or immunomodulatory molecules) to APC.

The genetics of S. enterica are very similar to those of E. coli and the full genome sequences of several Salmonella species and serovars are available. Therefore, the molecular tools and techniques currently available enable the rational construction of vaccine vectors based on S. enterica.

INTRODUCTION

Genome sequences of different salmonellae are available or are close to completion providing a rich data set to support studies on these micoorganisms. Comparative sequence analysis has been used to redefine the relationships between different Salmonella species and serovars and the first functional genomic analyses have been completed. In the near future genomic studies will facilitate a redefinition of the Salmonella genus from an evolutionary perspective and we can expect novel typing systems, diagnostic approaches and possibly therapies to emerge.

FULL GENOME SEQUENCES FACILITATE THE STUDY OF SALMONELLA

The availability of full genome sequences for several Salmonella serovars has radically advanced the fields of functional and comparative Salmonella genomics. The genomic era brings an opportunity to analyze more comprehensively the phylogenetic relationships between Salmonella, the evolution of pathogenicity and the genetic variability within natural populations – comparative genomics. The precise genetic makeup of the bacterium combined with host factors are thought to account for the observed differences in the disease spectra and host specificities for different salmonellae. The recent rapid expansion of bacterial genome sequence information has enhanced our ability to investigate the activities of the genes involved on the bacterial side of this equation – functional genomics. There is hope that these genetic insights may contribute not only to a clearer understanding of Salmonella pathogenicity and epidemiology, but also to the design of better vaccines, diagnostic kits and surveillance tools.

INTRODUCTION

The bacterial species Salmonella enterica subspecies enterica can be divided into over 2400 antigenically distinct serovars and the pathogenicity of most of these serovars is undefined. The majority of incidents of salmonellosis in humans and domestic animals are caused by relatively few serovars and these can be subdivided into three groups on the basis of host prevalence. The first group consists of host-specific serovars. These typically cause systemic disease in a limited number of phylogenetically related species. For example, S. enterica serovar Typhi, serovar Gallinarum and serovar Abortusovis are almost exclusively associated with systemic disease in humans, fowl and sheep respectively. The second group consists of host-restricted strains. These are primarily associated with one or two closely related host species but may also infrequently cause disease in other hosts. For example, S. enterica serovar Dublin and serovar Choleraesuis are generally associated with severe systemic disease in ruminants and pigs respectively (Sojka et al., 1977). However, these serovars are potentially capable of infecting other animal species and humans. The third group consists of the ubiquitous S. enterica serovars, such as Typhimurium and Enteritidis that usually induce gastroenteritis in a broad range of unrelated host species.

Clearly the nature and severity of Salmonella infections in different animal species varies enormously and is influenced by many factors including the infecting Salmonella serovar, strain virulence, infecting dose, host animal species, age and immune status of the host, and the geographical region. All these factors are likely to inter-relate.

INTRODUCTION

Salmonellosis in domestic animal species is important in terms of animal welfare and productivity. Infection may lead to decreased yields of milk, eggs or meat, and in certain cases loss of livestock. Salmonellosis in domestic species is also important for public health as the major reservoir and source of food-borne human infections.

A number of Salmonella enterica serovars can induce a systemic typhoid-like disease in healthy adults of a restricted range of host animal species. Other serovars colonize the intestine of the host and in some cases may induce severe enteritis. The severity of the disease will be dependent on the virulence and dose of the challenge and immune status of the host. Thus, some S. enterica strains that would normally induce enteritis in adult hosts are able to induce systemic disease in immuno-compromised hosts. Immunity to S. enterica is dependent on the nature of the disease that different serovars induce in different hosts. Thus, mucosal immunity is more likely to be important in protecting against serovars that induce enteritis, whereas systemic immunity would be more important in protecting against serovars that induce systemic disease.

Our understanding of the interaction of the host's immune system with different S. enterica serovars is still rudimentary. Effective control of salmonellosis affecting domestic host species requires a greater understanding of immunological mechanisms during such infections. This will provide the basis from which rational control measures, such as more effective vaccines, vaccination strategies, diagnostic tools or other non-immunological tools may be developed.

INTRODUCTION

Mononuclear phagocytes associate with S. enterica early in the disease process before acute inflammatory abscesses are formed, as well as during later stages of the acquired immune response in which macrophages form part of well-organized granulomas (Mastroeni et al., 1995; Richter- Dahlfors et al., 1997). The ability to survive within macrophages is a key event in the pathogenesis of Salmonella enterica (Fields et al., 1986). A growing body of information indicates that macrophages can serve as sites for S. enterica replication, even though they can be activated to exert potent anti- S. enterica activity. The great majority of the intimate interactions between S. enterica and macrophages take place inside a specialized endocytic vacuole named the phagosome. This chapter discusses the dynamic S. enterica phagosome as it pertains to the pathogenesis of this intracellular Gram-negative bacterium.

Immunological and genetic manipulations in animal models of infection, as well as the observation of naturally occurring genetic traits, have revealed that genetic loci encoding Nramp1, TLR4, NADPH oxidase and IFNγ play key roles in resistance to S. enterica infection. These host defenses are expressed directly by macrophages or, as in the case of IFNγ, up-regulate the anti-S. enterica activity of mononuclear phagocytes. In the following sections, we will discuss both the mechanisms by which these host defenses contribute to the anti-S. enterica activity of macrophages, and the virulence factors used by S. enterica to avoid these components of the antimicrobial arsenal of professional phagocytes.

INTRODUCTION

With the benefit of hindsight, the last 20 years in microbial ecology will probably be referred to as the census period that dramatically changed our perception of biodiversity within the three domains of life. Bacteria and archaea are no longer viewed as groups of peculiar and morphologically simple organisms that show relatively little diversification despite their long evolutionary history, but have now been recognized to harbour a perplexing number of novel phylogenetic lineages (Rappé & Giovannoni, 2003). Current estimates assume that the number of prokaryotic species ranges in the millions and thus vastly exceeds the fewer than 10 000 described prokaryotic species that have been isolated to date in pure culture (Curtis et al., 2002). This dramatic paradigm shift was only made possible by the development of cultivation-independent molecular approaches for surveying microbial diversity in nature. Whilst it is now evident that most prokaryotes cannot be cultured easily, due to their living in complex communities and their intimate metabolic links with both their abiotic and biotic environments, the powerful arsenal of techniques at our disposal enables us to see beyond the ‘cultured few’ and gain valuable insights into the realm of uncultured microorganisms (Wagner, 2004). It is now relatively straightforward to determine the species richness of natural microbial communities by comparative sequence analysis of environmentally retrieved 16S rRNA gene sequences (Olsen et al., 1986; Schloss & Handelsman, 2004).