|Home | About | Journals | Submit | Contact Us | Français|
The lower gastrointestinal tract is densely populated with resident microbial communities (microbiota), which do not elicit overt host responses but rather provide benefit to the host, including niche protection from pathogens. However, introduction of bacteria into the underlying tissue evokes acute inflammation. Non-typhoidal Salmonella serotypes (NTS) elicit this stereotypic host response by actively penetrating the intestinal epithelium and surviving in tissue macrophages. Initial responses generated by bacterial host cell interaction are amplified in tissue through the interleukin (IL)-18/interferon (IFN)-γ and the IL-23/IL-17 axes, resulting in the activation of mucosal barrier functions against NTS dissemination. However, the pathogen is adapted to survive antimicrobial defenses encountered in the lumen of the inflamed intestine. This strategy enables NTS to exploit inflammation to outcompete the intestinal microbiota, and promotes their transmission by the fecal/oral route.
Non-typhoidal Salmonella serotypes (NTS) are a leading cause of acute food borne disease worldwide (Box 1). The most common human clinical isolates are Salmonella enterica serotypes Typhimurium (S. Typhimurium) and Enteritidis (S. Enteritidis) (reviewed in 1). In immunocompetent individuals, NTS are associated with gastroenteritis, a localized infection of the terminal ileum and colon that manifests in fever, diarrhea and intestinal cramping. However, a breach of mucosal barrier functions in immunocompromised individuals can result in the development of a life threatening bacteremia.
NTS are the single most common cause of death from diarrheal disease associated with viruses, parasites or bacteria 83 and the leading cause of foodborne disease outbreaks in the United States 84, producing between $0.5 billion to $2.3 billion in annual costs for medical care and lost productivity 85. The press coverage of high profile outbreaks contributes to a good visibility of this public health problem in the US and Europe. What is less known publicly is the enormous impact that NTS infections have in developing countries, particularly in Sub-Saharan Africa. Diarrheal diseases result in an estimated 2-3 million annual deaths among children in developing countries, a significant portion of which is caused by NTS. In addition, NTS are a leading cause of bacteremia in immunocompromised individuals in Sub-Saharan Africa and symptoms of gastroenteritis are frequently absent in these cases. The main risk factors in African children for developing NTS bacteremia are malnutrition, acquired immunodeficiency syndrome (AIDS) and severe malarial anemia (reviewed in 86). The magnitude of this problem is little publicized but it contributes considerably to morbidity and mortality in Sub-Saharan Africa. For example, NTS are currently the most common blood isolates from children 87 and the second most common cause of neonatal meningitis in Malawi 88, resulting in mortality rates exceeding 20%. In contrast to children, NTS bacteremia in African adults is associated with AIDS as the sole major risk factor. Due to the human immunodeficiency virus (HIV) epidemic, NTS have become one of the most common blood isolates from adults admitted to hospital, which represents an important but under-recognized emerging infectious disease problem in Sub-Saharan Africa 86. Annually, about 10% of HIV positive African adults develop invasive NTS infections, resulting in mortality rates above 20% despite antibiotic therapy.
Current research on S. Typhimurium pathogenesis is beginning to paint a novel picture of the unique challenges and opportunities encountered during life in the inflamed intestine. Recent studies identify host factors that are critical for activating mucosal barrier functions in the inflamed intestine of immunocompetent hosts. In turn, these findings provide clues about the identity of mucosal barrier defects that put immunocompromised hosts at risk of developing bacteremia. New insights into the consequences that inflammation has on the growth conditions encountered by microbes residing in the intestinal lumen reveal how the pathogen might benefit from inducing antimicrobial host responses. Here we review these novel hypotheses that help to understand both the patient presentation and the selective forces shaping the inflammation-adapted pathogenic lifestyle of S. Typhimurium.
The key virulence traits that enable S. Typhimurium to elicit inflammation are its ability to penetrate the intestinal epithelium and to survive within macrophages. These properties are conferred by two type III secretion systems (T3SS) that function in injecting bacterial proteins, termed effectors, into the eukaryotic cytoplasm where they alter host cell physiology. The invasion-associated type III secretion system (T3SS-1) allows the pathogen to induce actin rearrangements in epithelial cells, resulting in membrane ruffling and bacterial internalization (reviewed in 2). The second type III secretion system (T3SS-2) is employed to alter trafficking of the Salmonella containing vacuole, resulting in macrophage survival (reviewed in 3). Both type III secretion systems contribute to intestinal inflammation after oral infection of calves 4 or streptomycin-pretreated mice 5, 6 (Box 2).
Mice are naturally susceptible to infection with S. Typhimurium and this animal model offers some clear advantages, such as the availability of knockout animals and immunological reagents. A limitation of using the mouse model to study gastroenteritis is the fact that these animals do not develop exudative inflammation in the intestinal mucosa during S. Typhimurium infection. This limitation can be overcome by pretreatment of mice with streptomycin, which results in the consistent development of neutrophil influx in the cecum during S. Typhimurium infection and can be used to model acute intestinal inflammation 89. A remaining caveat of the mouse model is that bacteremia and systemic dissemination are observed in mouse lineages that are either genetically resistant or genetically susceptible to S. Typhimurium, thus making it difficult to study mucosal barrier functions in this model. Furthermore, development of exudative inflammation requires a disruption of the intestinal microbiota through antibiotic treatment. In contrast to mice, the natural or experimental infection of calves with S. Typhimurium results in a localized enteric disease with clinical and pathological features that parallel the disease in humans 43. However, this model is more challenging experimentally, it requires specialized animal facilities and the availability of immunological reagents and genetic tools is limiting. A notable difference between the mouse and the calf appears to be related to mechanisms involved in an early initiation of intestinal inflammation. Signaling through bacterial specific TLRs does not contribute to cecal inflammation during the first one or two days after infection of streptomycin pretreated mice 90, while this mechanism contributes to cytokine production and neutrophil recruitment observed within hours after S. Typhimurium infection of the bovine ileal mucosa 9, 20. Neither mice nor calves are well suited to model how an underlying HIV infection predisposes individuals to develop NTS bacteremia. A model using rhesus macaques, which are naturally susceptible to both S. Typhimurium and SIV, has recently been developed to study this question 24.
The action of T3SS-1 and T3SS-2 enables S. Typhimurium to reside and survive in intestinal tissues where the pathogen is located intracellularly within epithelial cells and professional phagocytes (macrophages, dendritic cells and neutrophils) 7. Several mechanisms by which these direct interactions between bacteria and host cells result in the production of proinflammatory cytokines have been described using tissue culture models (reviewed in 8). These include the detection of bacterial products, termed pathogen associated molecular patterns (PAMPs), by pattern recognition receptors (PRRs) of the innate immune system. Examples of relevant PRRs are Toll-like receptors (TLRs) and nucleotide-binding and oligomerization domain (NOD)-like receptors (NLRs). For instance, the biofilm matrix of S. Typhimurium contains amyloid fibrils, termed curli, whose protein subunits (CsgA) are detected by the innate immune system through TLR2 9, a PRR that also responds to host amyloids 10. Flagella are proteinaceous surface appendages composed of a major subunit (flagellin), which is recognized by TLR5, a PRR expressed on the basolateral surface of intestinal epithelial cells 11. Flagellin can also serve as an agonist for a cytosolic NLR termed IPAF, an acronym for IL-1β converting enzyme (ICE)-protease activating factor. T3SS-1 translocates flagellin into the cytosol of macrophages 12, a process that directly or indirectly results in the activation of IPAF 13. In turn, IPAF activates ICE, also known as caspase-1, thereby rendering this protein proteolytically active. The resulting protein complex, which is known as the inflammasome, proteolytically activates IL-1β and IL-18 14. A second pathway activating the inflammasome during S. Typhimurium infection involves injection of SopE, a T3SS-1 effector protein 15. Lipopolysaccharide (LPS), a component of the Gram-negative outer membrane, is a potent TLR4 agonist triggering production of interleukin (IL)-23 by dendritic cells in response to S. Enteritidis infection 16. The relative contribution to intestinal inflammation of these and other mechanisms remains to be worked out in animal models (Box 2).
Direct contact between bacteria and host cells can result in the production of neutrophil chemoattractants known as CXC chemokines. However, in vivo these mediators are substantially produced by epithelial cells located in the crypts and at the base of villi, areas of the epithelium that are not invaded by S. Typhimurium 17. These observations suggest that CXC chemokine production in these areas of the epithelium is induced in vivo primarily by paracrine signaling mechanisms (Figure 1).
Since only a very small fraction of cells in infected tissue contain bacteria, the total capacity for cytokine production by these cells is probably limited in scope. However, host cells infected with S. Typhimurium are a source of cytokines, such as IL-18 and IL-23, which help to amplify responses in tissue through paracrine signaling mechanisms (Figure 1). T cells play an important role in amplifying inflammatory responses induced by S. Typhimurium in the intestinal mucosa by contributing to the production of key cytokines, including interferon (IFN)-γ, IL-17 and IL-22 18. IL-18 can stimulate antigen experienced T cells to rapidly secrete IFN-γ during bacterial infection by an antigen-independent mechanism, thereby significantly amplifying early effector responses in vivo 19. Caspase-1, which proteolyticallly activates IL-18, is essential for inducing IFN-γ expression in the cecal mucosa 12 hours after infecting mice with S. Typhimurium 20, and contributes to pathological changes observed two days after infection in the cecum 15. However, the inflammasome is not required for inducing expression of IL-17 20.
The cytokine IL-6 in combination with transforming growth factor (TGF)-β can initiate the differentiation of naïve T cells into antigen-specific IL-17 producing CD4+ memory T cells (Th17 cells) 21-23, an adaptive response that develops over a period of several days. However, a dramatic increase in IL-17 and IL-22 expression is observed in bovine or simian ligated ileal loops within two to five hours after infection of naïve animals with S. Typhimurium, pointing to an innate mechanism for inducing the production of these cytokines in the intestinal mucosa early after infection 24, 25. In a naïve host, this early, innate amplification of cytokine expression likely involves antigen experienced T cells that secrete IL-17 and IL-22 in response to an antigen-independent stimulation with IL-23. Studies in the mouse model show that IL-23 is essential for the expression of IL-17, and contributes to the production of IL-22 by T cells in the intestinal mucosa early during S. Typhimurium infection 26. In addition to T cells, natural killer (NK) cells 27, 28 and dendritic cells 29 are sources for IL-22 production in the intestine. The receptor for IL-23 is expressed on the surface of several distinct intestinal T cell populations, including Th17 cells, γδ T cells and natural killer T (NKT) cells 26, 30-32. The number of γδ T cells expressing the receptor for IL-23 increases within two days after S. Typhimurium infection, while expression by other T cell subsets remains constant 26. IL-23 does not affect production of IFN-γ, whose expression in the intestine is induced normally during S. Typhimurium infection of IL-23 deficient mice 26.
In conclusion, an innate, T cell-dependent amplification of inflammatory responses proceeds through two independent pathways, which are initiated by IL18 and IL-23, respectively. Consistent with the idea that their production results from an amplification of responses, IFN-γ, IL-17 and IL-22 are among the cytokines whose expression is most prominently induced during S. Typhimurium infection of the intestinal mucosa 18, 24.
The cytokine storm ensuing from the amplification of inflammatory responses in the intestinal mucosa results in the activation of two broad arms of an innate immune response. One important component of this antimicrobial response is the recruitment of neutrophils. IL-17 is a cytokine involved in orchestrating this arm of the host response, in part by stimulating epithelial cells to secrete CXC chemokines and by stimulating granulopoiesis in the bone marrow through inducing production of granulocyte colony-stimulating factor (G-CSF) 33. During S. Typhimurium infection, IL-17 deficient mice exhibit a defect in recruiting neutrophils into the intestinal mucosa 24. An additional mechanism contributing to CXC chemokine production in epithelial cells might be mediated through IL-1β 34, which becomes proteolytically activated by the caspase-1 inflammasome. Consistent with this idea, caspase-1 deficient mice exhibit reduced CXC chemokine production in the cecal mucosa early (12 hours) after S. Typhimurium infection 20 (Figure 1).
Neutrophils help to prevent S. Typhimurium dissemination from the gut, a notion supported by clinical data, showing that neutropenia is a risk factor for bacteremia with NTS 35, 36. The neutrophil barrier orchestrated by the IL-23/IL-17 axis is thought to constitute a defense against extracellular bacteria 37, which at first seems to be at odds with the fact that this host defense is effective against the facultative intracellular S. Typhimurium. A possible explanation is the brief extracellular phase that must exist when S. Typhimurium transits from epithelial cells to phagocytes or when it moves to a new host cell after killing a macrophage through pyroptosis 38 and this might render the pathogen susceptible to neutrophil attack. S. Typhimurium typically is not observed extracellularly in infected intestinal tissue 7. However, after depletion of neutrophils, bacteria grow extracellularly, which demonstrates the importance of this arm of the host defense in controlling the extracellular phase of S. Typhimurium in vivo 39, 40. In summary, clinical and experimental data support a role for neutrophil recruitment in checking systemic dissemination of S. Typhimurium from the gut.
AIDS patients have an increased risk of developing NTS bacteremia (Box 1) and experimental evidence from a rhesus macaque model points to IL-17 deficiency as the underlying immune defect responsible for this clinical observation. Simian immunodeficiency virus (SIV) induces a depletion of Th17 cells in the ileal mucosa of rhesus macaques, and this results in increased translocation of S. Typhimurium to the mesenteric lymph nodes 24. Interestingly, SIV infected rhesus macaques exhibit a defect in producing IL-17 and IL-22 in response to S. Typhimurium challenge while production of IFN-γ is normal. IL-17 deficiency 24 or depletion of CD4+ T cells 41 in mice results in increased bacterial translocation from the gut. These data suggest that Th17 depletion impairs the mucosal neutrophil barrier, which is responsible for the increased risk of individuals infected with human immunodeficiency virus (HIV) to develop NTS bacteremia. HIV positive individuals also exhibit an increased translocation of normal gut microbiota, which does not manifest as bacteremia. Instead, translocation of microbiota leads to chronic immune activation by introducing LPS into the circulation, which presumably originates from killed gut commensals 42. In contrast to the resident microbiota, S. Typhimurium possesses potent virulence factors, such as T3SS-2, and the resulting ability to resist host defenses upon crossing the mucosal barrier might help explain why NTS are leading blood isolates from AIDS patients in Sub-Saharan Africa (Box 1).
In addition to the benefits conferred upon the host, neutrophil influx can also cause collateral tissue damage and is associated with necrosis of the surface mucosa in large areas of the terminal ileum and colon 4. The neutrophil-associated injury to the intestinal epithelium, and the resulting loss of epithelial barrier function lead to leakage of extravascular fluids, thereby contributing to diarrhea 43. Neutrophils might also contribute to diarrhea by stimulating chloride secretion in intestinal epithelial cells 44. Thus neutrophil recruitment might be a double-edged sword that is potentially harmful to the host since the resulting diarrhea can lead to severe dehydration. IL-17 deficiency is predicted to impair neutrophil recruitment and neutrophil functionality during AIDS, which might explain why NTS bacteremia frequently manifests in the absence of gastroenteritis in these individuals 45, 46 (Box 1).
A second arm of the antimicrobial innate immune response encountered during acute inflammation is epithelial-derived antimicrobial proteins and peptides (Box 3). An important cytokine for orchestrating this arm of the response is IL-22, which induces expression in epithelial cells of host defense effector molecules that seem to be directed against luminal bacteria (Figure 1). During Citrobacter rodentium infection of mice, the IL-23/IL-22 axis is required for expression in the colonic mucosa of calprotectin, an antimicrobial protein mediating zinc deprivation, and RegIIIγ (regenerating islet-derived 3 gamma), a bactericidal C-type lectin 29. RegIIIγ expression in the cecal mucosa is markedly increased by an IL-23 dependent mechanism during S. Typhimurium infection of streptomycin pretreated mice 26. Secretion of RegIIIγ into the intestinal lumen of wild type mice contributes to clearance of luminal bacteria, including Listeria monocytogenes and vancomycin resistant Enterococcus, but this response is absent in mice deficient for myeloid differentiation primary response protein 88 (MyD88) 47, 48, an adaptor protein for all TLRs except TLR3.
A variety of molecules in the mammalian intestine might impair growth and survival of microbes in this anaerobic environment. One source of inhibitory molecules stems from many of the colonizing microbes themselves. Bacteriocins, including colicins, microcins, lantibiotics and others, are proteinaceous toxins that inhibit the growth of related bacteria 91. These molecules likely contribute to niche protection and might inhibit colonization by potential pathogens. The other source of inhibitory molecules are products of the host, some of which are host defense molecules per se, whereas others have primary function in nutrient absorption. For example, bile salts and hydrolytic enzymes required for digestion are toxic for microbes and might contribute to controlling microbial proliferation and survival. Within the group of primary host defense molecules, epithelial cells make various peptides along the intestinal tract with some constitutively expressed and others transcriptionally induced. In the small intestine, Paneth cells are the source of abundant quantities of antimicrobials in most mammals 92. These granule-rich secretory cells reside at the base of small intestinal out pouches called crypts. Among the most abundant and extensively studied antimicrobials of Paneth cells are the defensins 93. Constitutively expressed at high levels, the Paneth cell defensins typically have membrane-targeted antimicrobial activity against a wide range of bacteria, as well as some fungi and parasites. Transgenic and knock-out mouse models have provided evidence that Paneth cell defensins contribute to the host defense against bacterial pathogens 94, 95. Other secretory-granule-associated constitutive antimicrobials of these cells are lysozyme and secretory phospholipase A2. Paneth cells also make some inducible antimicrobials, including RegIIIγ, a C-type lectin with selective activity against Gram-positive bacteria 48, 96. In the colon, epithelial cells inducibly express a collection of antimicrobial peptides and proteins, including Resistin-like beta (RELM-b), RegIIIγ, calprotectin, and β-defensins.
In vitro, IL-22 induces the expression in human colonic epithelial cells of inducible nitric oxide synthase (iNOS), mucin (MUC4) and lipocalin-2, an antimicrobial protein that prevents bacterial iron acquisition 49. While lipocalin-2 secretion is induced upon IL-22 stimulation, S. Typhimurium infection of colonic epithelial cell lines does not induce expression, suggesting that activation of this epithelial antimicrobial response requires paracrine IL-22 signaling rather than direct interaction of bacteria with enterocytes. In vivo, epithelial cells in the ileal mucosa of rhesus macaques produce large quantities of lipocalin-2 in response to S. Typhimurium infection, resulting in accumulation of this antimicrobial in the intestinal lumen 49. Other antimicrobial proteins and peptides whose transcripts are prominently induced in the intestinal mucosa during S. Typhimurium infection include iNOS, calprotectin, MUC4, dual oxidase 2 (Duox2) and bovine enteric β defensin 18, 24, 25.
From these discussions, it can be concluded that acute intestinal inflammation causes dramatic changes in the environment encountered by luminal bacteria through at least two distinct mechanisms 50. First, the release of antimicrobial products orchestrated through the IL-23/IL-22 axis converts the intestinal lumen in an increasingly hostile environment. Second, coordination of neutrophil recruitment through the IL-23/IL-17 axis is associated with diarrhea, during the course of which the intestine becomes devoid of contents, thereby severely limiting the presence of nutrients that normally support bacterial growth in the intestinal lumen. Together, these dramatic changes in the luminal environment trigger changes in the composition of the intestinal microbiota 51. The numbers of commensal microbes, mostly belonging to the Firmicutes and Bacteroides phyla, are reduced 51, 52. Although S. Typhimurium cannot resist the onslaught of neutrophils in intestinal tissue, the pathogen appears to be custom-built to bloom in the lumen of the inflamed intestine, because its numbers in this niche increase dramatically during inflammation 52, 53.
The strategy by which S. Typhimurium appears to be able to adjust to the limited nutrient availability in the inflamed intestine is to specialize on mucus carbohydrates as a major source of high-energy nutrients. This pathway might take advantage of the increased mucus (MUC4) production elicited by IL-22 during S. Typhimurium infection 49. Flagella-mediated motility and chemotaxis are required for enhanced growth in the inflamed intestine, because D-galactose gradients emanating from the mucus layer drive S. Typhimurium chemotaxis towards this niche 54. Once it reaches the mucus, S. Typhimurium can attach using Std fimbriae, an adhesin that binds terminal α(1,2)fucose moieties in mucus carbohydrates of the murine cecum 55. Mucus colonization (Figure 2) enables S. Typhimurium to efficiently utilize mucus carbohydrates, which is thought to be a good energy source for the pathogen 54.
To exploit inflammation, S. Typhimurium must also possess virulence genes that confer resistance to antimicrobial defense mechanisms encountered during growth in the mucus layer; however, the underlying mechanisms are largely unknown. Support for this idea comes from studies on lipocalin-2 resistance of S. Typhimurium. Lipocalin-2 is an antimicrobial secreted into the lumen of the inflamed intestine 49, which prevents bacterial iron acquisition by binding enterobactin (also known as enterochelin), a low molecular weight iron chelator produced by most members of the Enterobacteriaceae, including S. Typhimurium 56-58. The iroBCDEN gene cluster of S. Typhimurium encodes proteins involved in the biosynthesis and uptake of salmochelin, a glycosylated derivative of enterobactin 59. Salmochelin is not bound by lipocalin-2 and its production therefore confers resistance against this antimicrobial protein 60, 61. Mutational inactivation of the iroBCDEN gene cluster renders S. Typhimurium sensitive to lipocalin-2, because it now depends on enterobactin for iron acquisition. The iroBCDEN gene cluster confers a growth advantage in the inflamed intestine, while no benefit is observed in the absence of intestinal inflammation or in lipocalin-2 deficient mice 49. These data suggest that the iroBCDEN gene cluster enables S. Typhimurium to cope with the increased lipocalin-2 concentration encountered in the lumen of the inflamed intestine. In addition, S. Typhimurium has virulence genes that confer resistance to antimicrobial peptides, including defensins 62-64, which might serve a similar purpose. Collectively, these data are beginning to support the concept that resistance to epithelial-derived antimicrobial proteins and peptides represents a specific adaptation of S. Typhimurium to life in the inflamed intestine.
The picture emerging from these studies is that efficient utilization of mucus carbohydrates and an intrinsic resistance to inducible antimicrobials produced by epithelial cells might allow S. Typhimurium to exploit inflammation by gaining an advantage during its competition with other microbes for intestinal colonization. This virulence strategy enables S. Typhimurium to thrive in the lumen of the acutely inflamed intestine. The benefit of this tactic is that suppression of the competing microbiota by intestinal inflammatory responses increases the numbers of S. Typhimurium in the intestinal contents 52, 53, thereby promoting its transmission by the fecal oral route 65 (Figure 1).
Production of IFN-γ in the early phase of intestinal inflammation contributes to antimicrobial responses in the intestinal mucosa 66 and is caspase-1 dependent 20. With the onset of adaptive immunity, IL-12 dependent mechanisms further increase IFN-γ levels, thereby restricting bacterial growth in macrophages and the pathogen is cleared after the development of antigen-specific CD4+ T cells responses 67. Thus, a location in tissue is a dead end and the bacterial cells triggering intestinal inflammation by invading the intestinal mucosa cannot benefit from the resulting outgrowth of their comrades in the intestinal lumen. It has been proposed that such a self-destructive cooperation can evolve through population heterogeneity, where one fraction is capable of invading intestinal tissue while the other fraction can benefit from the resulting environmental changes in the intestinal lumen (Figure 2) 68. For example, Std fimbriae are only expressed by a fraction of bacteria in a population, which is probably due to phase variation 69. As a result, bacterial cells expressing this adhesin might be able to colonize the mucus layer 55, while the remaining bacterial cells might be able to pass through it and invade the intestinal epithelium. These or other mechanisms might enable S. Typhimurium to successfully use self-destructive cooperation to thrive in the lumen of the inflamed intestine.
Recent advances in understanding the pathogenesis of S. Typhimurium-induced gastroenteritis draw a fascinating picture of a pathogen that triggers intestinal inflammation to create specific growth conditions in the intestinal lumen, which enable it to outcompete the resident microbiota. This inflammation-adapted lifestyle requires S. Typhimurium to be resistant against many of the host defense mechanisms encountered in the lumen of the inflamed intestine. One defense to which S. Typhimurium is not resistant is neutrophils, which can check systemic spread of the pathogen in immunocompetent individuals. However, in immunocompromised individuals, defects in neutrophil recruitment and neutrophil antimicrobial activity can give rise to a fulminant bacteremia, which often develops in the absence of gastroenteritis. Once S. Typhimurium enters the bloodstream of immunocompromised individuals, its considerable resistance against antimicrobial proteins and peptides leaves the host with precious few innate defenses, which might explain the high mortality rates associated with NTS bacteremia.
The evolution of this inflammation-adapted pathogenic lifestyle was driven by acquisition of key virulence determinants through horizontal gene transfer 70. After it split from the closely related Escherichia coli lineage some 100-140 million years ago 71, 72, the genus Salmonella acquired Salmonella pathogenicity island (SPI) 1, a genetic region carrying genes encoding the T3SS-1 secretion apparatus 73. SPI1 encodes only two effector proteins, SipA and SipC, which are involved in inducing actin rearrangements required for bacterial entry into epithelial cells. Other T3SS-1 effector proteins required for invasion 74, including SopB, SopE2 and SopD, were acquired through independent horizontal gene transfer events that occurred at a similar time during evolution, as indicated by the presence of the corresponding genes in all members of the genus Salmonella 75. Genes on SPI2 encoding the T3SS-2 secretion apparatus were acquired some 71-100 million years ago by the S. enterica lineage after its divergence from S. bongori 76, 77,78, the second species within the genus Salmonella.
While commensal E. coli serotypes are adapted to colonize the normal intestine, acquisition of SPI1 and SPI2 enabled Salmonella serotypes to trigger acute intestinal inflammation in their vertebrate hosts. To survive the consequences of acute intestinal inflammation, acquisition of SPI1 and SPI2 by the Salmonella lineage was likely accompanied by horizontal transfer of genes that enable the pathogen to survive the dramatic environmental changes encountered in the lumen of the inflamed gut. Consistent with this idea, the iroBCDEN gene cluster is present in all members of S. enterica, suggesting that these genes were acquired at a similar time during evolution as SPI2 79, 80. The iroBCDEN gene cluster is absent from the genomes of commensal E. coli isolates but can be found in uropathogenic E. coli (UPEC) isolates 81, 82, which are able to grow and survive at the mucosal surface of the inflamed bladder. These observations support the concept that lipocalin-2 resistance evolved as an adaptation to colonize inflamed mucosal surfaces.
It appears likely that acquisition of SPI1 and SPI2 was accompanied by horizontal transfer of additional genes mediating efficient utilization of mucus carbohydrates and an intrinsic resistance against inducible antimicrobials produced by epithelial cells. Identification of these Salmonella-specific genes and their respective roles during the inflammation-adapted pathogenic lifestyle of S. Typhimurium represents an exciting area for future study.
We would like to thank Sebastian Winter for critical reading of the manuscript. This work was supported by Public Health Service grants AI050553 (R.M.T.), AI076246 (A.J.B., C.L.B. and L.G.A.), AI040124, AI044170 and AI079173 (A.J.B.). CT is supported by Scientist Development Grant 0835248N from the American Heart Association.
Publisher's Disclaimer: This is a PDF file of an unedited manuscript that has been accepted for publication. As a service to our customers we are providing this early version of the manuscript. The manuscript will undergo copyediting, typesetting, and review of the resulting proof before it is published in its final citable form. Please note that during the production process errors may be discovered which could affect the content, and all legal disclaimers that apply to the journal pertain.