|Home | About | Journals | Submit | Contact Us | Français|
The mucosal surfaces are often the first site of interaction between pathogenic microorganisms and the host. Activation of the mucosal immune response has the important function of containing an infection and preventing dissemination of pathogens to systemic sites (barrier function). Numerous lines of evidence suggest that the barrier function is orchestrated by a subset of cytokines (interleukin (IL-)17 and IL-22), which belong to the Th17 family. IL-17 and IL-22 induce expression of antimicrobial peptides and neutrophil chemoattractants at mucosal sites, and thus play an important role in controlling mucosal infections. However, there is increasing evidence that mucosal pathogens achieve greater colonization during inflammation because they are resistant to a subset of these antimicrobial responses. In this review we compare the antimicrobial responses elicited by Th17 cytokines during mucosal infections with four different pathogens: Klebsiella pneumoniae, Citrobacter rodentium, Candida albicans and Salmonella typhimurium. We will then discuss which responses may constitute the mucosal barrier, thus providing a benefit to the host, and which ones may promote the colonization of pathogens, thereby providing a benefit to the microbes.
Mucosal surfaces are often the first interface between the host, the resident microbiota, and pathogenic microorganisms. In normal individuals, the mucosa constitutes a barrier against the systemic spread of microorganisms. Both the microbiota and pathogens are normally confined to mucosal sites by the local innate immune responses. However, in certain individuals the mucosal defenses are altered and thus commensals or pathogens may spread systemically. For instance, patients infected with the human immunodeficiency virus (HIV) clinically present with high concentrations of lipopolysaccharide (LPS) in the blood, indicating that Gram negative commensals may enter the blood stream from the gut . Also, HIV-infected patients are more susceptible to bacteremia . Elderly patients are also at high risk for bacteremia, mostly originating from the urinary tract, intra-abdominal sites and the lungs . While it is accepted that these groups of patients have altered mucosal immunity, the precise defects have yet to be elucidated.
When pathogens approach the mucosal surface, the mucosal epithelium constitutes both a physical barrier and a first line of defense via mechanisms of innate immunity. To counter these impediments, pathogens can employ a repertoire of virulence factors that include enabling penetration through thick mucous layers (flagella), surface adherence (fimbriae and other adhesins) and induction of a host inflammatory response.
Direct interaction of microbes with epithelial cells and resident macrophages induces a stereotypic host response. Conserved structures (pathogen associated molecular patterns, PAMPs) are recognized by pattern recognition receptors (Toll-like receptors and Nod-like receptors) of epithelial cells and antigen presenting cells [4, 5]. These cells are then activated and secrete chemokines and cytokines. For a long time, major research efforts have been directed towards investigating the direct interaction of pathogens with epithelial cells and antigen-presenting cells, which was considered sufficient for mounting an effective mucosal response. However, there is increasing evidence that several subsets of T cells are activated in the mucosa in response to pathogens .
Clinical evidence for the role of T cells as “gate keepers” in the mucosa comes from patients infected with HIV, wherein CD4+ T cells are severely depleted by the virus over time [7, 8]. Importantly, while the number of CD4+ T cells in the blood is reduced quite slowly, the kinetics of CD4+ T cell depletion at mucosal sites (particularly in the gut) are much faster, with severe depletion occurring within the first few weeks following HIV infection [7, 8]. This drop in mucosal CD4+ T cells correlates with increased susceptibility to mucosal infections. Notably, HIV-infected patients are more susceptible to bacteremia caused by intestinal pathogens like non-typhoidal Salmonella, and less commonly Shigella and Campylobacter [9-11]. From these epidemiological observations, it follows that T cells in the mucosa may be essential components of mucosal responses to pathogens.
A pioneering study on the role of T cells in the mucosal barrier was published in 1994 by Gautreaux and his colleagues, who showed that depletion of either the CD4+ or CD8+ T cell subsets resulted in increased translocation of non-pathogenic E. coli to the mesenteric lymph nodes . Also, depletion of CD4+ and CD8+ T cells resulted in increased systemic dissemination of S. typhimurium to the spleen and liver . Adoptive transfer of CD4+ and/or CD8+ T cells in mice previously depleted of T cells inhibited the translocation of E. coli from the GI tract . Several other studies point towards a role of T cells in orchestrating mucosal responses to infections. In particular, a new subset of T cells, termed Th17 cells, has been recognized as a key component of mucosal immunity (reviewed in ).
Th17 cells are a distinct lineage from Th1 and Th2 cells , characterized by the release of a subset of cytokines: interleukin (IL-) 17A and IL-17F, IL-22 and IL-26 [16, 17]. Receptors for IL-17A and IL-17F (Il-17Ra and IL-17Rc) are present in several cell types, including antigen presenting cells and epithelial cells [18, 19]. In contrast, receptors for IL-22 and IL-26 appear to be localized to the epithelium [19-21]. Very little is known about the role of IL-26 during mucosal infection because rodents do not express this cytokine. For IL-17A/F and IL-22, expression increases at mucosal sites after infection with a number of pathogens including lung infection with Klebsiella pneumoniae [22, 23], intestinal infections with Citrobacter rodentium [24-26] or Salmonella typhimurium [27, 28], and colonization of the oral cavity with Candida albicans . In each of these examples IL-17 and/or IL-22 are induced and contribute to localizing the infection to the mucosa, thus impeding dissemination of these pathogens beyond the initial site of infection.
K. pneumoniae infection of IL-17Ra-/- mice results in increased death and systemic dissemination from the lung . Moreover, K. pneumoniae infection is worsened if IL-22 is depleted from IL-17Ra-/- mice . Th17 cells are also activated during colonic infection to C. rodentium [24-26]. All the known mouse Th17 cytokines (IL-17A and IL-17F, IL-22) appear to play a role in controlling C. rodentium infection and the severity of gut pathology [24-26]. IL-17 but not IL-22 seems to be important for controlling C. albicans infection of the oral mucosa . S. typhimurium infection induces expression of IL-17 and IL-22 in the intestinal mucosa of mice, calves and rhesus macaques [27, 28]. In rhesus macaques, infection with the Simian Immunodeficiency Virus (SIV) causes depletion of Th17 cells, resulting in increased dissemination of S. typhimurium to the mesenteric lymph nodes . This can be recapitulated in IL-17Ra-/- mice, which have an increased bacterial load of S. typhimurium in the mesenteric lymph nodes and spleen .
Considering that IL-17 and IL-22 appear to play a role in orchestrating the mucosal barrier against different pathogens, it is not surprising that there is substantial overlap between expression of some known Th17-induced genes in response to various infectious agents, as shown in Table 1. It thus appears that early Th17 activation constitutes a stereotypic host response to mucosal infections.
The cytokines IL-17 and IL-22 orchestrate the mucosal response by inducing expression of several chemokines and antimicrobial peptides (Table 1). IL-17 and IL-22 stimulation of lung and gut epithelial cells induces expression of several antimicrobial responses and chemokines, including some CXC chemokines that are neutrophil chemoattractants [23, 30-36]. Thus, neutrophil influx in the mucosa is at least partly dependent on IL-17 and IL-22. Moreover, both IL-17 and IL-22 stimulate granulopoiesis by inducing expression of G-CSF [22, 23, 35, 37, 38]. Neutrophils are an important first line of mucosal defense against bacterial infections. A low neutrophil count (neutropenia) is an important risk factor for bacteremia. One of the most common treatments for neutropenia is the administration of G-CSF, which is also employed in HIV patients [39, 40]. In animal models, Th17 deficiency results in a decrease in neutrophil recruitment during lung infection with K. pneumoniae, gut infection with S. typhimurium, and oral infection with C. albicans [22, 23, 28, 29]. Given that these pathogens are susceptible to neutrophil killing, it is likely that a defect in neutrophil recruitment may explain why these infections are less well controlled in the absence of Th17 responses.
Several antimicrobial responses are induced by IL-17 and IL-22 in the mucosa. Expression of secreted C-type lectins of the RegIII family is dependent on IL-22 . IL-22 deficient mice infected with C. rodentium show increased mortality when compared to wild-type mice . Nevertheless, administration of just Reg3γ to IL-22-/- mice is sufficient to control infection with C. rodentium to a similar level as wild-type mice . Induction of Reg3γ is also important to limit growth of vancomycin-resistant enterococci in the gut . During S. typhimurium infection the expression of Reg3γ is dependent on IL-23, a cytokine upstream of IL-17 and IL-22 . Reg3γ is not an effective antimicrobial against S. typhimurium in vitro, making it highly unlikely that it plays a role in controlling S. typhimurium bacterial numbers in vivo .
β-defensins are a major family of antimicrobial peptides present in mammals . Expression of β-defensins is induced by Th17 cytokines in some infection models. For instance, induction of the antimicrobial peptide β-defensin 3 in response to infection with C. albicans is dependent on IL-17 . β-defensin 3 has candidacidal activity in vitro [29, 45] and saliva from wild-type mice, but not IL-17Ra-/- mice, has candidacidal activity, indicating that IL-17 also controls C. albicans proliferation by promoting secretion of antimicrobial peptides . Several β-defensins are induced in the gut in response to C. rodentium infections, with induction of β-defensin 1, 3, 4 dependent on IL-17 and and IL-17F .
As different subsets of Th17-mediated responses appear to control infection with different mucosal pathogens, further work is needed to investigate which aspects of the Th17 response are effective in controlling specific pathogens.
There are many studies indicating that Th17-induced responses are beneficial to the host as they provide containment of infections to the mucosa. In spite of this, pathogens colonize mucosal surfaces during inflammation quite successfully. There is increasing evidence that inflammation can be beneficial to a pathogen and promote its colonization and host-to-host transmission. The case of the gut is emblematic as this mucosal surface is colonized by a barrier of millions of protective bacteria (the resident microbiota) that is severely compromised upon initiation of inflammation .
Colonic infection with C. rodentium results in changes of microbial composition, with preferential elimination of a subset of microbiota and overgrowth of C. rodentium itself and other gamma-Proteobacteria . It is thus highly likely that a subset of antimicrobial responses induced by C. rodentium in the gut alters the composition of the microbiota and promotes C. rodentium gut colonization during inflammation. In a similar fashion, infection with S. typhimurium results in alteration of the gut microbiota composition, with growth suppression of Bacteriodes spp. and Firmicutes spp. [48, 49]. This results in a higher degree of S. typhimurium colonization of the inflamed gut. The observation that avirulent S. typhimurium mutants which fail to trigger gut inflammation are outcompeted by the resident microbiota indicates that both microbiota suppression and the ability to thrive in the inflamed intestine benefit S. typhimurium. In line with the observation that inflammation is beneficial for pathogens is the observation that only S. typhimurium strains which trigger an inflammatory response are transmitted to the next susceptible host . From these studies, it appears that intestinal pathogens exploit the mucosal host response to suppress growth of the resident microbiota and colonize the inflamed gut.
One of the most important strategies for successful colonization is a pathogen’s ability to access nutrients in the context of inflammation. Motility and chemotaxis are important to direct pathogens towards sources of nutrients like those provided by mucins  which are upregulated in vivo and in vitro by IL-17 and IL-22 stimulation [23, 31, 33]. Enteroaggregative Escherichia coli (EAEC), uropathogenic E. coli (UPEC) and Shigella flexneri secrete a serine protease autotransporter (Pic) which degrades mucin [52, 53]. The Pic mucinase enhances EAEC colonization of the mouse gastrointestinal tract and growth in the presence of mucin, thereby constituting a virulence factor which enhances the fitness of a pathogen in the hostile gut environment .
Other essential nutrients for bacterial growth are metal ions including iron, zinc and manganese. Consequently, one host strategy to fight bacterial infection is to deprive bacteria of said ions. Iron is largely bound to serum transferrin and, in inflamed mucosal sites, to lactoferrin secreted by neutrophils . To overcome these host defense mechanisms, bacteria secrete iron chelators with higher affinity than transferrin and lactoferrin, termed siderophores [55, 56]. Once siderophores are bound to iron, they are internalized by specialized transport systems. One of these siderophores, termed enterochelin, is a target of the host antimicrobial protein lipocalin-2 . Enterochelin is produced by most members of the Enterobacteriaceae including Citrobacter spp., E. coli spp., Klebsiella spp., and Salmonella spp. [58-61]. Lipocalin-2 is an antimicrobial peptide secreted at high levels by epithelial cells at mucosal surfaces in response to IL-17 and IL-22 stimulation, thus mostly in response to infection [23, 29, 31]. Lipocalin-2 binds to iron-laden enterochelin, effectively suppressing the growth of strains lacking other means of iron acquisition [57, 62]. Some enteric pathogens (Salmonella spp, uropathogenic E. coli, and some strains of Klebsiella) produce additional siderophores and their receptors . One of these siderophores, termed salmochelin, is a glycosylated form of enterochelin which is not bound by lipocalin 2 [64-67] and its expression renders pathogens resistant to lipocalin-2-mediated iron withholding [31, 68-70]. Salmochelin-mediated resistance to lipocalin-2 promotes S. typhimurium colonization of the inflamed gut , but provides no benefit in the absence of gut inflammation or lipocalin-2 . This indicates that iron acquisition during mucosal inflammation is an important virulence trait of pathogens trying to survive and proliferate in the inflamed gut.
In addition to salmochelin, other siderophores and their receptors may facilitate pathogen colonization. Some K. pneumoniae clinical isolates are susceptible to lipocalin-2 expressed in the lung . However, K. pneumoniae expressing the siderophore yersiniabactins colonize the lung better than isogenic mutants impaired in yersiniabactin synthesis . Some pathogenic E. coli produce the siderophores yersiniabactin and aerobactin in addition to salmochelin . Acquisition of siderophores and additional iron uptake mechanisms may circumvent the inhibition of enterobactin mediated by lipocalin-2.
Lipocalin-2 expression is induced by several other mucosal pathogens. Both Streptococcus pneumoniae and Haemophilus influenzae induce expression of lipocalin-2 in the nasal mucosa . In rhesus macaques, Helicobacter pylori infection induces expression of lipocalin-2 and other antimicrobials in the stomach . Also C. albicans infection of the oral cavity in mice induces expression of lipocalin-2 . All of these pathogens are not susceptible to lipocalin-2 antimicrobial activity, and the role of lipocalin-2 in these infection models has yet to be established. It is feasible that the expression of lipocalin-2 and other antimicrobial peptides promotes colonization of mucosal pathogens other than S. typhimurium by suppressing the growth of competitors.
Several other antimicrobial responses are induced in the mucosa in a Th17-dependent fashion. Expression of the S100A8 and S100A9 subunits of calprotectin - an antimicrobial peptide that chelates the essential nutrients zinc and manganese  - is also induced in response to IL-17 and IL-22 [17, 26]. Inducible nitric oxide synthase (iNOS) expression also appears to be dependent on IL-22 and IFN-γ . It will be important determine whether these antimicrobial responses play a role in mucosal immunity.
Recent studies reveal a role for Th17-mediated mucosal immunity in the early control of infection with a variety of organisms. The responses elicited by the cytokines IL-17 and IL-22 in the mucosa are important for controlling dissemination of infectious agents beyond the mucosa and constitute the mucosal barrier. These responses are thus beneficial to the host during an infection.
While some aspects of the Th17-mediated responses are beneficial to the host, others are exploited by mucosal pathogens to achieve greater colonization when competing for a niche with other microbes. This is a prime example of pathogens evolving to take advantage of key components of the immune response. Thus, Th17-mediated responses constitute a double-edged sword: on one end, these responses are necessary to contain an infection to the mucosa, and on the other end they favor pathogen colonization (Figure 1). While a defect in Th17-mediated responses would likely impair mucosal colonization of pathogens, the reduced expression of antimicrobial peptides would also promote the dissemination of both pathogens and commensals from the mucosal surfaces. Sustained activation of Th17 cells is a hallmark of several autoimmune disorders, like multiple sclerosis, inflammatory bowel diseases, and psoriasis [78-80]. For this reason, blocking Th17 cell function is currently considered as a therapeutic option for many of these conditions. However, blockade of Th17 responses may also result in increased severity of mucosal infections and bacteremia, a side effect that requires careful evaluation.
The authors would like to thank Andreas Bäumler and Sean-Paul Nuccio for helpful discussions. Work in MR laboratory is supported by Public Health Service Grant AI083619 and the IDSA ERF/NIFID Astellas Young Investigator Award.
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.