Interleukin-10 and Immunity against Prokaryotic and Eukaryotic Intracellular Pathogens

Abstract

The generation of an effective immune response against an infection while also limiting tissue damage requires a delicate balance between pro- and anti-inflammatory responses. Interleukin-10 (IL-10) has potent immunosuppressive effects and is essential for regulation of immune responses. However, the immunosuppressive properties of IL-10 can also be exploited by pathogens to facilitate their own survival. In this minireview, we discuss the role of IL-10 in modulating intracellular bacterial, fungal, and parasitic infections. Using information from several different infection models, we bring together and highlight some common pathways for IL-10 regulation and function that cannot be fully appreciated by studies of a single pathogen.

INTRODUCTION

Interleukin 10 (IL-10) is an anti-inflammatory cytokine most readily associated with macrophages, both as a source of IL-10 and as the population most impacted by its action (82). Numerous other cells have, however, been shown to secrete IL-10, including dendritic cells (DCs), T cells, B cells, neutrophils, eosinophils, and mast cells (82). The secretion of IL-10 is mediated by several cytokines, including IL-12 (116), IL-6 (126), transforming growth factor β (TGFβ) (78), and IL-27 (38, 84), although the exact pathways that lead to IL-10 secretion are currently unclear. The action of IL-10 on target cells is more clearly described and is mediated by the IL-10 receptor (IL-10R), a dimer consisting of an α and β subunit (69, 125). Engagement of the IL-10R results in the activation of the Jak1 and Tyk2 protein tyrosine kinases and the activation, and DNA binding, of signal transducer and activator of transcription 3 (STAT3), leading to a downstream alteration in the biological function of the target cell (90, 106). The action of IL-10 results in the downregulation of major histocompatibility complex class II (MHC II) proteins and costimulatory molecules, such as CD80 and CD86, on the surfaces of target macrophages (8, 31). IL-10 also suppresses the production of reactive oxygen and nitrogen intermediates in activated macrophages (41). Therefore, IL-10 diminishes the capacity of innate immune cells to kill pathogens, as well as reduces their capacity to generate and maintain responsive antigen-specific T cells.

The suppression of effector function after IL-10R engagement is mediated by several key molecules, including heme oxygenase-1 (HO-1) and suppressor of cytokine signaling-3 (SOCS3) (71, 104). Both HO-1 and SOCS3 induce potent changes in JAK/STAT signaling through the mitogen-activated protein kinase (MAPK) system, leading to disruption of effector cell function (113, 117). Although the specifics of IL-10R signaling are still being elucidated, it is evident that SOCS3 and, potentially, SOCS1 are mediators of cellular responses, including tumor necrosis factor (TNF), gamma interferon (IFN-γ), and nitric oxide production (32, 67, 104). Furthermore, it appears that SOCS1 and SOCS3 can negatively impact the production of IL-10, indicating an internal SOCS-dependent negative-feedback loop for the regulation of IL-10 (32, 67, 104). Since IL-10 has such a potent effect on protective immunity, controlling its activity is a vital component of immune competence. The intricacies of IL-10 regulation have previously been reviewed (116), but briefly, several studies report that epigenetic remodeling of the iL10 locus is the primary regulatory mechanism of IL-10 production (56, 61, 115, 132). Though chromatin remodeling may be an important first step in IL-10 expression, other molecular factors also regulate IL-10 production. Various transcription factors, such as CREB, ATF-1 (1), GATA3 (18, 122), and musculoaponeurotic fibrosarcoma (MAF) (15), bind to the iL10 promoter, but no single factor is sufficient to induce IL-10 expression. Posttranscriptional control of iL10 mRNA is regulated by elements in the 3′ untranslated region (UTR) (103) and microRNAs (61, 121) that lead to iL10 mRNA degradation. Most of these control mechanisms are induced by IL-10 itself in a negative-feedback loop (13, 68, 87). The subtle balance of pro- and anti-inflammatory signals determines the extent of IL-10 expression and production.

The action of IL-10 leads to inhibition of secretion of inflammatory cytokines, including IFN-γ, TNF, IL-1, IL-2, and granulocyte-monocyte colony-stimulating factor (GM-CSF), as well as several chemokines (29, 49, 62, 76). Therefore, after the generation of a proinflammatory immune response, IL-10 serves to dampen inflammation that could be deleterious to the host, limiting potential tissue damage. However, the immunosuppressive properties of IL-10 can also be exploited by pathogens, leading to a reduction in proinflammatory and antigen-specific responses that are normally required to control or clear infection. In this minireview, we focus primarily on the role of IL-10 in modulating intracellular bacterial, fungal, and parasitic infections. By evaluating experimental data across several different intracellular infectious disease model systems, we identify some of the common characteristics of IL-10 regulation, expression, and function that enable us to better understand the function of this cytokine in vivo during infection.

THE ABSENCE OF IL-10 LEADS TO IMPROVED CONTROL OF INFECTION

Studies in the absence of IL-10, either by disruption of the IL-10 gene in mice or by an antibody blockade of the IL-10 receptor (IL-10R), indicate that the majority of intracellular infections are controlled better or cleared faster in the absence of IL-10 (summarized in Table 1). In some instances, however, the removal of IL-10 during an infection results in inflammation-associated death, highlighting the critical immunosuppressive role of IL-10 in vivo. Polymorphisms in the human IL-10 promoter have also been linked to altered susceptibility (35, 40, 66, 85, 98, 107), supporting the findings in animal models. In general, the absence of IL-10 leads to extended survival after infection that is associated with an enhanced adaptive immune response, including CD4 T cell IFN-γ production and improved control of infection as the normal proinflammatory responses are extended or enhanced. IL-10 is also capable of modulating the innate immune response. For example, intravenous infection with Candida albicans is quickly cleared in IL-10 knockout (Ko) mice compared to wild-type mice, which is attributed to more-efficient pathogen killing by neutrophils (129). It is unclear why IL-10 has such a significant influence on cells of the innate immune system yet impacts predominantly adaptive immune responses for most infections that have been studied, but IL-10 likely serves as an important molecule in the cross talk between innate and adaptive immune cells.

Table 1.

Effects of IL-10 manipulation on pathogenicity

Intracellular pathogen	Alteration(s) of IL-10: impact(s) of IL-10 alterationa
Bacteria
Chlamydia trachomatis	KO: ↑ chlamydial-antigen-presenting ability
Chlamydophila pneumoniae	KO: ↑ pathogen clearance, ↑ immunopathology
Coxiella burnetii	KO: ↓ bacterial replication; OE: ↓ TH1, ↑ TH2
Klebsiella pneumoniae	Ab: ↑ TH1 cytokines, bacterial clearance
Legionella pneumophila	Ab: ↑ TH1 cytokines
Listeria monocytogenes	Ab: ↑ infection resistance (R); ↓ fatality in chronic infection (S)
Mycobacterium avium	Ab: ↓ bacterial replication, ↓ serum IgM
Mycobacterium bovis BCG	KO: ↑ macrophage activation, ↑ NO, ↑ TNF
Mycobacterium tuberculosis	OE: ↑ bacterial load, ↓ TH1 (R); KO: ↓ bacterial load, chronic immunopathology (S)
Pseudomonas aeruginosa	KO: ↑ wt loss, ↑ lung inflammation; OE: ↑ mortality, delayed PMN recruitment
Yersinia enterocolitica	KO: resolve infection, ↑ TNF
Yersinia pestis	KO: restore TLR function, resolve infection
Parasites
Leishmania donovani	KO: resistant to infection, ↑ TH1
Leishmania major	Ab: drives TH2 response
Plasmodium chabaudi	KO: ↑ disease severity, mortality
Plasmodium yoelii	Ab: ↑ survival
Toxoplasma gondii	KO: ↑ disease severity, mortality
Trypanosoma cruzi	Ab: ↓ parasitemia, ↑ survival (S)
Fungi
Candida albicans	KO: ↑ pathogen clearance, ↑ infection resistance
Histoplasma capsulatum	KO: ↑ pathogen clearance

^aAb, anti-IL-10(R) monoclonal antibody; KO, knockout; OE, overexpression; (R), resistant mice; (S), susceptible mice; PMN, polymorphonuclear leukocytes; ↑, increased; ↓, decreased.

Although IL-10 clearly plays an important role in many infectious diseases, understanding the exact mechanism of its action has been challenging because each infection model studied has addressed only a small piece of the puzzle. Reviewing the literature from a single intracellular pathogen is not sufficient to determine how this cytokine exerts its biological effect. Only in combination can we fully appreciate some of the common pathways and mechanisms that the host and the pathogen have implemented to modulate IL-10 production. In this minireview, we highlight some of the common themes that become visible as we view the role of IL-10 across a broad spectrum of pathogens. Furthermore, as data from different infection models and studies in humans are evaluated, the immunomodulatory action of IL-10 draws attention to the critical interplay between pathogen virulence and the variability of expression of host factors that control inflammation.

PATHOGEN-INDUCED IL-10 PRODUCTION VIA PRRs

Secretion of IL-10 is induced by several cytokines, but the trigger for IL-10 production during infection is becoming apparent through the study of in vitro systems and mice that lack specific pattern recognition receptors (PRRs). Using clearly defined infection models or purified antigens from pathogens, it is now evident that the interaction of specific PRRs with their ligands not only drives proinflammatory cytokine production but also stimulates the production of IL-10, which then plays a critical regulatory role in moderating the inflammatory response.

Using purified antigens from intracellular bacterial pathogens of the genus Yersinia, it has been shown that pathogen-Toll-like receptor 2 (TLR2) interactions play an important role in the stimulation of IL-10 production by macrophages. Yersinia pestis and Yersinia enterocolitica LcrV antigens interact with TLR2, leading to IL-10 secretion by stimulated cells (124). This process was dependent on the presence of TLR2 as well as the lipopolysaccharide (LPS) binding protein CD14. IL-10 subsequently induced a hypo-responsive state upon TLR2-mediated cell activation, leading to a dampening of inflammatory cytokine production and the regulation of inflammatory responses. Perhaps most intriguing was the finding that IL-10 production, mediated by LcrV, led to a general hypo-responsiveness of several other TLRs. This LcrV-induced TLR hypo-responsiveness was absent in cells isolated from IL-10 KO mice, confirming a role for IL-10 in this process and suggesting inhibition of a common downstream signaling event via the action of IL-10 (105). The capacity of antigens from Y. pestis to stimulate IL-10 production is not limited to LcrV, as the culture of cells with YopM also stimulated the production of IL-10 (79), suggesting a selection for this specific immunomodulatory activity by Y. pestis. For Yersinia spp., the induction of IL-10 clearly favors pathogen survival because, in contrast to wild-type mice, IL-10-deficient mice are fully capable of resolving infection (102, 123).

IL-10 is a critical biomarker for poor disease outcome during infection with the intracellular parasite Leishmania donovani (46). Like Y. pestis, L. donovani stimulates IL-10 production via TLR2- and TLR4-mediated pathways, leading to suppressed IL-12 secretion and increased susceptibility (17). Furthermore, the parasite Toxoplasma gondii is capable of shutting down TLR4-mediated LPS signaling in a manner that specifically blocks TNF but allows for the production of IL-10 (73). This indicates an advantageous role for generating abundant IL-10 for some parasitic infections. For the fungal pathogens Histoplasma capsulatum (25) and C. albicans (111), the absence of IL-10 results in accelerated pathogen clearance. TLR2-mediated pathogen recognition has also been implicated in this infection model. TLR2 KO mice are more resistant to infection with C. albicans than wild-type mice, which correlates with increasing levels of IL-10 in vivo (88).

For mycobacterial infections, the outcome of depleting IL-10 is more variable. Mice deficient in IL-10 or those receiving anti-IL-10R antibodies have varied phenotypes, with reports of no effect, a transient protective effect, or a long-term beneficial effect during infection with either virulent Mycobacterium tuberculosis, the environmental mycobacterium Mycobacterium avium, or the vaccine strain Mycobacterium bovis BCG (28, 51, 52, 58, 89). The outcome of infection in the absence of IL-10 can be linked to the species/substrain of mycobacteria or the particular strain of mouse being used, and this impact of host variation will be discussed below. Like Y. pestis, purified M. tuberculosis antigens, or the whole mycobacterium, can be recognized via TLR2, which results in TNF and IL-10 production in vitro (43, 108). Interestingly, the culture of human-monocyte-derived macrophages with lipid fractions from the hypervirulent Beijing strain of M. tuberculosis, compared to lipid from a less virulent strain, led to increased IL-10 production and an associated loss of expression of TLR2, TLR4, and MHC II molecules on the cell surface (108). These studies indicate that the virulence of the M. tuberculosis strain may play a role in dictating the production and biological influence of IL-10. It is unclear whether less virulent strains of M. tuberculosis drive the production of IL-10 via lipid-TLR2-associated pathways; however, additional mycobacterial antigens have also been shown to interact with other PRRs and to influence the production of IL-10. Lipoarabinomannan is a major component of the virulent M. tuberculosis cell wall and is a potent inducer of IL-10 when recognized by the mannose receptor on macrophages (24, 127), and binding of DC-SIGN by bacterial products also stimulates IL-10 production (45, 133). Additionally, the PPE18 antigen of M. tuberculosis H37Rv induces IL-10 production in a mechanism that utilizes the MAPK signaling pathway (86). In vivo studies with CARD9 KO mice also implicate nod2 and/or dectin-1 in the recognition of M. tuberculosis (33). Interestingly, CARD9 KO mice developed overwhelming inflammation and had decreased survival following infection with M. tuberculosis, which was associated with a specific absence of IL-10 production (33). It is unclear whether IL-10 directly interacts with any type of inflammasome. Certainly, receptors like dectin-1 can signal through CARD9 and induce IL-10 production (109), and IL-10 can downregulate dectin-1 surface expression (134), but interactions between components of the NLRP₃ inflammasome and IL-10 signaling pathways during infection have not yet been described.

Overall, evidence is growing for a TLR-mediated induction of IL-10 production in vitro and in vivo that is common across several intracellular pathogens. There is evidence that the recognition of antigens from infectious agents via TLR2, and potentially TLR4 and CD14, lead to IL-10 secretion and downregulation of proinflammatory responses, including the decreased expression or reactivity of other TLRs. We can speculate that TLR2-mediated activation by some pathogens drives an anti-inflammatory response that sustains an infection by dampening inflammation and the generation of adaptive immunity, thereby providing a niche for pathogen survival and persistence (Fig. 1). In this regard, the ability of a pathogen to decorate its surface with ligands for TLRs may provide an important evolutionary advantage for some infectious agents. Interestingly, studies with zymosan, a glucan from yeast cell wall, show that cells receiving zymosan (representing infection) are actually refractory to the immunosuppressive activities of IL-10, as they internalize the IL-10R in response to TLR2 ligation (34). Therefore, in this scenario, infected cells will continue to generate antimicrobial mediators and eradicate the infectious agent, yet the surrounding cells will be suppressed by IL-10 in a process that limits associated tissue damage. Therefore, the capacity for microbes to secrete in vivo antigens that bind PRRs and modulate neighboring cells would be a way to modify the local environment. This would provide a survival advantage for some pathogens for which inflammation is an important survival mechanism, whereas for most infections, the host response to soluble immunogenic mediators would be advantageous for clearance of infection.

Fig. 1.

The interplay between pathogen and host dictates the requirements for IL-10. The initial interaction between the pathogen and host likely dictates the necessity for the immunoregulatory role of IL-10. (A) If pathogen recognition leads to proinflammatory responses that can rapidly and successfully eradicate the pathogen, the anti-inflammatory properties of IL-10 are minimal and an IL-10 blockade or removal has little effect on disease outcome. (B) If pathogen recognition triggers a strong inflammatory response with abundant TNF, IL-10 is essential to counter this reaction and prevent tissue damage. The removal of IL-10 leads to a severe immunopathology that is frequently lethal. (C) For pathogens that are able to avoid killing by host cells, a balance is established between the generation of proinflammatory responses to eradicate the pathogen and anti-inflammatory responses to limit inflammation. In this instance, pathogen-driven IL-10 production can establish persistence.

IL-10 DAMPENS TNF-MEDIATED INFLAMMATION DURING INFECTION

Regardless of the stimulation pathway, IL-10 secretion appears to directly contribute to the downregulation of TNF secretion, reducing inflammation and associated cell death. This mechanism is evident for several different pathogens, including T. gondii (42) and Trypanosoma cruzi (54), for which infection normally leads to an upregulation of TNF and chemokines, recruitment of neutrophils, and cell death via apoptosis or necrosis, which can be detrimental if not moderated. For many infections, TNF is secreted in abundance and is further amplified in mice lacking IL-10. While this additional burst of TNF can be favorable in enhancing the clearance of some infections, an absence of IL-10 can also be detrimental. For example, mice lacking IL-10 are highly susceptible to infection with the parasites T. cruzi (55), T. gondii (42), and Plasmodium chabaudi chabaudi (74), which is linked to the overwhelming inflammation and TNF-mediated shock that occurs in the absence of the immunosuppressive properties of IL-10. In this regard, IL-10 production is essential to dampen inflammation-associated mortality.

The mechanism by which IL-10 dampens the production of TNF occurs via the inhibition of NF-κB (30) and, as indicated previously, the establishment of a negative-feedback loop that inhibits TLR-mediated cellular activation. In several infections, the pathogen is able to modify host responses that enhance IL-10 production in a manner that promotes infection. For example, uptake of Pseudomonas aeruginosa by macrophages stimulates IL-10 production and the regeneration of IκBα, leading to inhibition of NF-κB activation (22). Furthermore, M. tuberculosis infection of alveolar macrophages leads to IL-10-mediated BCL-3 expression and inhibition of NF-κB, again leading to lower levels of TNF (99). Infection of macrophages with T. gondii blocks the binding of transcription factors NF-κB, CREB, and c-Jun to the TNF promoter, despite the fact that these transcription factors translocate to the nucleus (72). Therefore, T. gondii appears to disrupt chromatin remodeling of the TNF promoter and interferes with histone modification to promote IL-10 while reducing TNF production (73). Furthermore, T. gondii activates STAT3, leading to reduced TNF and IL-12 production (14). The various mechanisms that T. gondii employs to generate IL-10 production by host cells indicates that IL-10 provides a significant survival (or persistence) advantage to the parasite. It is therefore no surprise that the removal of IL-10 can be fatal for T. gondii-infected mice, most notably due to overwhelming inflammation (42).

Because TNF is also a potent activator of cell death via apoptosis, increasing levels of IL-10 can moderate the extent of apoptosis that is induced in response to infection. In a Chlamydia pneumoniae model, where bacterial clearance was enhanced in the absence of IL-10, mice also developed severe inflammation and had elevated levels of apoptosis (100). To confirm a role for IL-10 in the control of cell death, C. pneumoniae-infected IL-10-deficient mice were supplemented with IL-10, which reduced the level of apoptosis (44). Apoptosis itself also serves as an important trigger for IL-10 production. In Coxiella burnetii infection, the presence of apoptotic lymphocytes drives macrophage polarization to an M2 phenotype, which is associated with IL-10 secretion in addition to other alternatively activated macrophage properties (mannose receptor, arginase-1 expression) (6). Increased IL-10 production was linked to enhanced bacterial replication, as blocking the action of IL-10 during infection reduced C. burnetii replication (47). Therefore, interaction with apoptotic cells elevates the production of IL-10 by macrophages, which in turn decreases TNF production and reduces apoptosis further. This mechanism is clearly linked to pathogen survival, as naturally increased production, or the overexpression, of IL-10 in vivo is associated with enhanced macrophage polarization and persistent replicative infection of C. burnetii in mice (81). In this regard, increased IL-10 serves to create an environment for the persistence of C. burnetii.

When apoptosis occurs as a result of a potent proinflammatory response and cell clearance is critical for resolution of inflammation, IL-10 secretion is a relevant and essential mediator in limiting tissue damage. In contrast, if apoptosis favors the clearance of a pathogen, then it benefits the host to limit IL-10 production and promote this cell death pathway. Correspondingly, if apoptosis is used as a natural host defense mechanism to eradicate a pathogen, it can also benefit the pathogen to promote the production of IL-10, either by driving IL-10 production via PRR interactions or by interfering with the signaling pathways that lead to IL-10 regulation. For example, some virulent strains of M. tuberculosis induce less apoptosis of cells after infection than less virulent strains (65). This correlates well with the levels of IL-10 produced in culture and indicates that elevating the levels of IL-10, perhaps by potent TLR2 ligation, promotes pathogen survival. A careful balance between TNF (-associated apoptosis) and IL-10 production may be an effective mechanism for M. tuberculosis to establish a latent and persistent infection in the host. Indeed, M. avium-infected mice that overexpressed IL-10 had evidence of less apoptosis and increased bacterial loads (37), indicating that apoptosis is associated with the clearance of this mycobacterial species.

Since TNF is also critical for establishing competent mycobacterial granulomas (16), perturbation of the IL-10-TNF balance should lead to disruption of the granulomatous control of M. tuberculosis infection. The influence of IL-10 on granuloma formation has not been fully elucidated; however, granuloma formation is moderately altered in IL-10 KO/anti-IL-10R-treated mycobacterial models (21, 39, 57, 120) as well as in models with other pathogens (135). C57BL/6 transgenic mice that overexpress IL-10 exhibit altered granuloma formation during chronic M. tuberculosis infection, with a transition from the compact, lymphocyte-rich granulomas that are normally observed in this strain to macrophage-dominated and disorganized lesions that resemble those usually seen in mice of the CBA/J background, a strain that naturally expresses abundant IL-10 (128). Whether IL-10 disrupts TNF function during chronic M. tuberculosis infection is unclear, however, because TNF mRNA levels appear to be intact during late stages of infection in CBA/J and C57BL/6 IL-10 transgenic mice (128).

Overall, the study of intracellular pathogens has shown that IL-10 plays a significant role in modulating TNF production to reduce inflammation and apoptotic cell death (Fig. 2). The varied mechanisms of modulation implemented by both host and pathogen that have been described are all centered on the inhibition of NF-κB and TLR-mediated signaling. For the host, there is a fine balance between controlling inflammation and the clearance of an infection, especially when the downregulation of inflammatory signatures via the immunosuppressive action of IL-10 favors pathogen persistence. From the perspective of the pathogen, there is a benefit to interfering with and increasing the production of IL-10. For some infections, such as T. cruzi (54, 55) and P. chabaudi chabaudi (74) infections, the anti-inflammatory role of IL-10 is critical for reducing damage, as is evident by the inflammation-induced lethality of removing IL-10 in mouse models of these infections. For other infections, such as H. capsulatum infections, where the inflammatory response appears to be more subtle, the role of IL-10 seems less significant (25). Furthermore, some pathogens, such as M. tuberculosis, appear to manipulate the levels of IL-10 that are produced in response to infection to facilitate their own survival (96).

Fig. 2.

IL-10 and the balance between apoptosis and pathogen persistence. The presence or absence of IL-10 in infected tissue dictates not only the strength of the immune response but subsequently the amount of apoptosis incurred due to inflammation. A heavily inflamed apoptotic environment is not conducive to pathogen survival; thus, some pathogens induce IL-10 via pattern recognition receptors (PRRs) to subvert the development of acute inflammation. An abundance of IL-10 in the infected tissue decreases activation-induced inflammation and apoptosis, allowing the pathogen to persist intracellularly. If IL-10 is not induced, or it is removed, the inflammation that develops is robust enough to clear the infection, but high levels of apoptosis and parenchymal damage consequently occur.

CELLULAR SOURCES OF IL-10

Macrophages have classically been considered the dominant source of IL-10, and IL-10 production has certainly been detected from CD14⁺ cells in response to intracellular pathogens such as M. tuberculosis (26), Y. pestis (124), and Chlamydia trachomatis (92), as well as from DCs in response to Plasmodium yoelii infection (136). However, studies of intracellular pathogens have revealed alternate cellular sources of IL-10 that include natural killer (NK) cells, B cells, and T cells. In systemic infections, such as those caused by Y. pestis, T. gondii, and Listeria monocytogenes, NK cells respond to IL-12 from DCs and become a dominant source of IL-10 (101). While IL-12 also stimulates NK cells to produce IFN-γ, IL-10 production seems to predominate in disseminated infections, generating a negative-feedback loop where antigen-presenting cell (APC)-derived IL-12 is subsequently downregulated. During infection with C. albicans (110) and T. gondii (7), neutrophils were also found to be a source of IL-10. Other studies implicate the importance of B-1 cells as an early innate source of IL-10. Peritoneal B-1 cells can produce abundant IL-10 after stimulation with the TLR agonists LPS (TLR4) and CpG (TLR9) (91), and B-1 cells (CD5⁺) have also been shown to secrete IL-10 in vivo in response to Leishmania major infection in BALB/c mice (112). Furthermore, BALB/c xid mice, which have a mutation in Bruton's tyrosine kinase gene and lack B-1 cells, had a reduced burden of L. major infection relative to that of wild-type mice which was associated with a loss of IL-10 production (53). In contrast, CBA/J xid mice were still capable of producing IL-10 during M. tuberculosis infection (63), indicating that the role for B-1 cells may be infection specific and likely linked to the route and primary location of infection. Previously, B-1 cells were shown to secrete IL-10 (91, 112) in response to infection or stimulation, but current work indicates that other B cell subsets, including plasma cells, also contribute to IL-10 production and are critical for controlling virus-specific CD8⁺ T cell responses and plasma cell expansion (77). Clearly, when early IL-10 production is implicated in infection models, it is important to also consider other immune cells, such as NK cells, neutrophils, and B cells, as alternate sources of IL-10.

From the adaptive immune system, IL-10-producing T cells have been identified. It is clear from studies with the intracellular parasites L. major and Leishmania guyanensis that CD8 T cells are a source of IL-10 during infection (4, 9); however, the biological relevance of CD8 T cell-derived IL-10 has not been fully realized. The discovery of regulatory T cells (114) has shifted the focus of this research area toward CD4 T cells, and CD4 T cells are a dominant source of IL-10 in several different experimental infections with intracellular pathogens. IL-10-producing FoxP3⁺ regulatory CD4 T cells have been detected in parasite infections, including those caused by T. cruzi (2), L. major (5), L. guyanensis (10), T. gondii (95), and P. yoelii (19), although the exact mechanism of their action has not been fully elucidated. For L. major, IL-10-producing CD4⁺ CD25⁺ FoxP3⁺ T cells have been shown to inhibit the generation of memory recall immunity upon challenge with heat-killed parasites (93, 94), indicating that IL-10-producing CD4 T cells can have a detrimental impact on immune responses, at least for the functional capacity of memory recall responses. IL-10 is also found within CD4 FoxP3⁻ populations in many infection models (60). For T. gondii infection, CD4 FoxP3⁻ T cells are considered to play a protective role during infection, and the discovery of T cells with dual expression of IL-10 and IFN-γ has led to the hypothesis that these cells self-regulate (59). Indeed, IL-10-producing IFN-γ CD4 T cells have been shown to inhibit IL-12 production while simultaneously promoting antiparasitic activity by antigen-presenting cells (60). Experimental data suggest that this dual-expression phenotype is a natural progression of cell differentiation in response to elevated IL-12 and antigen load (59), as a way to mediate local and potentially damaging immunity (60).

For the intracellular bacterial pathogen M. tuberculosis, FoxP3⁺ regulatory CD4 T cells have been found at the site of infection in humans (20), and in vitro, these cells are capable of suppressing antigen-specific IFN-γ production via the action of IL-10 (12). Furthermore, T cells from the peripheral blood of PPD-negative tuberculosis patients were capable of secreting IL-10 and modulating in vitro IFN-γ production (27), and in an independent study, CD25⁺ cells from purified protein derivative (PPD)-positive individuals, indicating exposure to mycobacteria, could inhibit T cell proliferation and IFN-γ production in a manner that was partially dependent on IL-10 (75). There is good supportive evidence that IL-10-producing regulatory T cells play a role in modulating immune responses during M. tuberculosis infection in humans; however, studies in mice have been less conclusive. The depletion of CD25⁺ cells in mice leads to very modest changes in IFN-γ production in response to M. bovis BCG vaccination, with very little impact on the ability to control a subsequent infection with M. tuberculosis (97). In contrast, recent studies have determined that antigen-specific regulatory T cells are capable of delaying the generation of IFN-γ-producing protective T cells, which was associated with increased lung bacterial burden during M. tuberculosis infection (119). Furthermore, adoptive transfer of FoxP3⁺ CD25⁺ CD4⁺ T cells into Rag^−/− recipient mice also led to a decrease in protective immunity against infection with M. tuberculosis (70). For murine models, however, the role of IL-10 in regulatory T cell function is unclear.

It is becoming increasingly evident that cells of both the innate and adaptive arms of immunity contribute IL-10 and influence the control of infection-associated inflammation. Transgenic mice that produce IL-10 from either T cells (induced) or macrophages (constitutive) have provided a model system to address the relative contributions of IL-10 derived from these two cell populations. Mice that constitutively overexpressed IL-10 within the macrophage compartment showed increasing numbers of M. tuberculosis CFU in the lung and earlier mortality than wild-type controls (where IL-10 was detected at modest levels) (118). Increased susceptibility to M. tuberculosis infection was highly associated with the presence of alternatively activated macrophages within the lung. Interestingly, in this model system, as well as in experiments using M. bovis BCG (83) or M. avium (37), TH1 cell function was neither impaired nor enhanced, indicating that the increased levels of IL-10 did not overtly impact the function of T cells during infection but perhaps served as an autocrine regulatory mechanism that impaired macrophage function, facilitating intracellular M. tuberculosis growth in vivo.

Mice that expressed IL-10 under the control of the IL-2 promoter, leading to induced production from the T cell compartment, also showed increasing numbers of M. tuberculosis CFU, indicating again that elevated levels of IL-10 increase the susceptibility of mice to infection with M. tuberculosis (128). The IL-12 message was reduced, demonstrating some influence on macrophage function. In contrast to mice that expressed IL-10 in the macrophage compartment, T cell IL-10 transgenic mice had fewer activated T cells within the lung and reduced antigen-specific IFN-γ production from lung cells (with associated increased levels of IL-10 in culture). In this specific model, T cell-derived IL-10 led to changes in both the macrophage and the T cell phenotype or function in response to M. tuberculosis infection, indicating that T cell-derived IL-10 can directly modulate T cell function. In contrast, macrophage-derived IL-10 appears to have little direct impact on the adaptive arm of immunity (118). In other infection models, IL-10 production from macrophages similarly led to enhanced susceptibility to infection with L. monocytogenes and L. major, again with no measurable impact on T cell function (48). IL-10 transgenic mice under the control of the IL-2 promoter were fully capable of controlling L. major infection, but not M. tuberculosis infection (50), perhaps indicating a more significant requirement for macrophage-derived IL-10 in the control of this intracellular infection.

The role of IL-10-secreting T cells and regulatory T cells has not yet been fully realized for intracellular infections, but studies to date demonstrate that T cells contribute to the pool of IL-10 that is generated during several different infections. The level of IL-10-secreting T cells is a good correlate with disease progression, at least for several intracellular parasitic infections (2, 5, 11, 19), and it is likely that they play an as-yet-unidentified role in other intracellular infections. IL-10-secreting T cells seemingly arise to regulate the local proinflammatory response, and we anticipate that this is mediated in an antigen-specific manner. Several intracellular parasite infection models have determined that IL-10-secreting T cells express a skewed Vβ T cell receptor repertoire, Vβ9 in T. cruzi infection (131) and Vβ6 within the CD8 T cell pool of L. major infection (23), indicating an antigen-dependent generation. It is currently unclear how antigen-specific IL-10 secretion, versus innate IL-10 secretion, differentially influences the control of infection. Dissecting the relative contributions of T cell-derived IL-10 and reevaluating the role of CD8 T cells from other cellular sources may uncover new mechanisms for IL-10-mediated regulation of the immune response.

GENETIC VARIABILITY CAN INFLUENCE THE ROLE OF IL-10

The study of IL-10 using a variety of different intracellular infections highlights the complex role that this cytokine plays in the development of disease. For some pathogens, such as M. bovis BCG, IL-10 plays a modest role in regulating inflammation, and the removal of IL-10 in this system barely impacts disease outcome (36). In this scenario, it is likely that the pathogen does not generate abundant inflammation during infection or that the clearance of the pathogen is rapid, negating a potent anti-inflammatory response when IL-10 production is critical. For other infections, such as M. tuberculosis (52) or T. gondii (42), that are persistent and/or stimulate abundant TNF, IL-10 is absolutely critical for the survival of the host, and in its absence, infection can be fatal. What is clear from the large quantity of studies of IL-10 KO, or anti-IL-10R-treated, mice is that our knowledge of the role of IL-10 is tightly linked to the variability of the model system that is being studied. In the mouse model, the critical need for IL-10 often changes depending on the background of the mouse strain being tested, and it is likely that a similar variability is evident in humans. For example, DBA/2 mice are inherently more susceptible to C. albicans infection than C57BL/6 mice, and although the removal of IL-10 increases resistance in both mouse strains, the impact of an IL-10 blockade on DBA/2 mice is much more dramatic (111). Additionally, BALB/c mice are more susceptible to L. major infection than C57BL/6 mice, which is linked in part to the naturally elevated levels of IL-10 measured in BALB/c mice (64). Administration of IL-10 to C57BL/6 mice exacerbates infection in this normally resistant mouse strain, indicating that an abundance of IL-10 increases susceptibility (130). Furthermore, IL-10-secreting B-1 cells are more prominent in L. major-infected BALB/c mice than in C57BL/6 mice (113). For CBA/J mice, elevated amounts of IL-10 are responsible for the increased susceptibility of this mouse strain to infection with M. tuberculosis in comparison to the susceptibility of the relatively resistant C57BL/6 (3) or BALB/c (80) strain. The blockade of IL-10 activity leads to a moderate reduction of bacterial load in the lungs of C57BL/6 and BALB/c mouse strains (104a), whereas the blockade of IL-10 in CBA/J mice significantly reduces the bacterial loads and extends survival (3). The positive impact of IL-10 on M. tuberculosis persistence that is apparent in the C57BL/6 mouse strain is not observed in CBA/J mice. It is clear that the natural predisposition of the host to respond to infection with a potent proinflammatory signature followed by subsequent IL-10-mediated suppression may ultimately dictate the course of infection.

SUMMARY

Following an infection, it is essential for the host to limit inflammation. If an infectious agent is quickly cleared, the anti-inflammatory role of IL-10 is transient; its absence is rarely problematic and can in fact enhance the clearance of infection (Fig. 1 A). For infections that are highly immunogenic and inflammatory, the role of IL-10 is essential, and its absence can lead to unimpeded inflammation that can be fatal (Fig. 1B). For many pathogens that fall within these two extremes, particularly those that establish persistence, there is a critical balance where IL-10 is beneficial either for the host or for the pathogen (Fig. 1C). TLR2 signaling seems to be a general pathway for interaction between host and pathogen that leads not only to proinflammatory cytokine production but also to the production of IL-10. While there are significant advantages for the host to generate IL-10 in response to infection, it is likely that some pathogens have evolved mechanisms to promote interactions with TLR, induce IL-10, and interfere with the generation of proinflammatory responses for their own advantage. Therefore, highly virulent pathogens that are rich in ligands for TLR (and survive within an intracellular environment) may stimulate IL-10 production to facilitate their own persistence (Fig. 2). The role of IL-10-producing, antigen-specific T cells in this scenario is unclear but may provide a critical pathway for IL-10-mediated regulation at the single-cell level by a host that is focused specifically on infected cells via the specific presentation of antigen. In this regard, the cellular source of IL-10 during infection with intracellular pathogens and its interaction with infected versus noninfected cells may be critical factors in the regulation of inflammation that have not yet been fully realized.