Th17 and regulatory T cells: implications for AIDS pathogenesis

Abstract

Purpose of review

The present review discusses recent reports showing that reciprocal changes in T helper interleukin-17-secreting CD4⁺ Th17 cells and CD4⁺CD25^highFoxP3⁺ regulatory T cells (Tregs) may play a role in the progressive disease caused by the HIV and by simian immunodeficiency virus.

Recent findings

Studies in nonhuman primate models of lentiviral infection and in HIV-infected human individuals have shown that pathogenic infection is associated with loss of Th17 cells and an increase in the frequency of Tregs. Because interleukin-17 serves to maintain the integrity of the mucosal barrier, loss of Th17 cells may permit the increase in microbial translocation across the gastrointestinal mucosa that is observed in pathogenic lentiviral disease. It remains unclear, however, whether Th17 cells are preferentially infected or if, instead, their loss is induced by bystander effects of lentiviral infection, for example, the induction of indoleamine 2,3-dioxygenase.

Summary

Progressive lentiviral disease is associated with preferential depletion of Th17 cells and loss of Th17/Treg balance. Further analysis of such changes in the composition of subset CD4⁺ T helper and Tregs may shed new light on the immunopathology of HIV disease and suggest new strategies for therapeutic and preventive interventions.

Introduction

The Th1/Th2 paradigm has shaped our understanding of adaptive immune responses by delineating the interactions that can occur between T helper subsets [1]. More recently, attention has focused on the reciprocal relationship betweenCD4⁺T cells secreting interleukin-17 (Th17 cells) and CD4⁺CD25^highFoxP3⁺ regulatory T cells (Tregs). Cells of these two lineages are derived from a common progenitor [2–5,6••,7•,8] and their differentiation pathways are reciprocally modulated in a number of aberrant immune disorders related to host–pathogen interactions, inflammatory syndromes, autoimmune diseases, and primary immune deficiencies [9,10]. In the present article, we discuss the hypothesis that it is the relative balance of these two subsets, rather than the function of either alone, that may drive the immunopathology of progressive disease caused by simian immunodeficiency virus (SIV) and HIV.

The acute phase of SIV and HIV infection is a period of aberrant immune activation and rapid loss of CD4⁺T cells in lymphoid tissues such as the gastrointestinal tract. Early studies suggested that, during this time, resting memory CD4⁺ T cells are preferentially infected and activated effector memory T cells undergo apoptosis, and that early treatment with highly active antiretroviral therapy can reverse these effects [11,12]. In subsequent studies by Brenchley et al. [13], circulating levels of microbial products such as lipopolysaccharide (LPS) were found to be significantly increased in chronically infected HIV patients and in SIV-infected rhesus macaques, suggesting that breaches in the gut epithelial barrier facilitate microbial translocation and prompt the persistent immune activation that characterizes disease progression to the AIDS. By contrast, even though massive depletion of CD4⁺ T cells also occurs in the gastrointestinal tract of SIV-infected sooty mangabeys and African green monkeys (nonhuman primate species that do not progress to AIDS despite chronic infection with SIV) [14], microbial translocation is not observed [15,16]. Put together, these findings suggest that a critical determinant of disorder in lentiviral disease lies not so much in the quantity of CD4⁺ T cells lost but in qualitative changes that may instead relate to the relative composition of CD4⁺ T cell subsets.

The reciprocal development of Th17 and regulatory T cells in mucosal tissue

The microbial flora of the intestinal tract is composed primarily of bacteria that are commensal to and in some situations symbiotic with the host. In part, it is the role of the intestinal immune system to confer tolerance toward these organisms whereas also prompting immune responses against the occasional pathogen that enters [17,18•–20•]. At least two populations of CD3⁺CD4⁺ T cells have been found to be important in mediating this balance. Tregs expressing the transcription factor, forkhead box P3 (FoxP3), and Th17 cells secreting the cytokine, interleukin-17. Interestingly, Tregs and Th17 cells share common chemokine receptors (CCR6, CCR4) and homing properties (CCL20) [21] and are derived from a common progenitor cell [22–24], the differentiation of which is dependent upon the stimulation of mucosal dendritic cells and macrophages by microbial, parasitic, or fungal products [25–27] and cytokines including interleukin-23, interleukin-1β, and interleukin-6 [26,28].

Th17 cells are critical in the defense against bacteria and fungi, and also contribute to the homeostasis of enterocytes [2,29–31]. This T cell subset, however, has also recently been implicated in the pathology of human inflammatory bowel diseases (IBD) [32]. Th17 cells express the interleukin-23 receptor (IL23R) and, in genome-wide association studies, the IL23R as well as other genes involved in the differentiation of Th17 cells have been recognized as IBD susceptibility genes [33]. In small animal models of IBD and of autoimmune inflammation of the brain and nervous system, models that were typically thought to be Th1-dependent, mice deficient in the interleukin-23p19 subunit did not develop inflammation whereas those deficient in the interleukin-12p35 subunit did [34,35]. This seminal observation identified interleukin-23 as a critical driver of autoimmune inflammation and, in so doing, underscored the importance of the Th17 lineage. In subsequent studies, Annunziato et al. [36] described a population of human gut-associated lymphoreticular tissue (GALT) T cells that express both interleukin-17A and interferon-γ (IFNγ), raising the possibility that these cells may play a role in the pathophysiology of human IBD [36–38].

In contrast to the pro-inflammatory effects associated with Th17 cells, naturally arising CD4⁺CD25^highFoxP3⁺ + Tregs induce tolerance against self antigens and prevent autoimmunity. As the initial description of such cells nearly 15 years ago by Sakaguchi et al. [39], their importance in various rodent models of experimental colitis has been highlighted by their ability to reverse both acute and chronic inflammation [40–46]. In studies of human IBD, however, such immunosuppressive effects have been less clearly demonstrated, prompting Maul et al. [47] to suggest that it may be more relevant to focus instead on the balance between Tregs and other proinflammatory T cell subsets (e.g., Th1 and Th17 cells). A number of studies have shown that FoxP3⁺ Tregs exert anti-inflammatory functions and control self-reactive T cells, including Th1, Th2, and Th17 cells [48,49]. In the context of acute and chronic infectious diseases, the net outcome of these effects remains unclear [50]. In the context of HIV and SIV disease, for instance, Tregs might decrease chronic immune activation, thereby slowing disease progression [51]; conversely, they might inhibit antiviral immune responses, thereby hastening disease progression [52,53]. Discriminating between these two extremes has been hampered by the difficulty of precisely identifying distinct subsets of FoxP3⁺ Tregs in vivo [54] and quantifying their mechanism(s) of suppression in vitro [55].

Studies in mice and in humans have defined a developmental link between Th17 cells and Tregs. Thus, transforming growth factor-β (TGF-β) is essential for the development of Th17 cells [2,3,56,57], primarily because it upregulates the retinoic acid receptor-related orphan receptor-γt (ROR-γt, encoded by RORc gene) [22,58], the master transcription factor of Th17 differentiation. Interestingly, TGF-β is also known to induce the Treg specific transcription factor, FoxP3, a key regulatory gene for the development of Tregs [59,60]. In a study first demonstrating the reciprocal nature of Treg versus Th17 differentiation, Bettelli et al. [4] found that the addition of interleukin-6 to TGF-β inhibits the generation of Tregs and induces the development of Th17 cells. Further support for the reciprocal development of Th17 and Treg cells was obtained in studies of retinoic acid [5,61] (which induces FoxP3, inhibits ROR-γt in Th17-inducing conditions and promotes the development of Tregs), of ligands of the aryl hydrocarbon receptor (AhR) [7•], and of Stat3 deficiency (a critical transcriptional factor related to Th17 development in humans [62•,63•] and directly involved in Treg contol over Th17 responses in mice [64••]). The above observations, coupled with the known effects of lentiviral infection of the gastrointestinal tract, beg the question: is it possible that the relative proportion of Th17 cells and Tregs might also influence the progression of lentiviral disease in vivo?

Sustained immune activation and the shift in Th17/regulatory T cells balance

Although CD4⁺ T cell depletion is the hallmark of HIV disease, the mechanisms leading to such depletion in vivo remain unclear. Possible mechanisms include viral lysis, apoptosis, and/or immune clearance of HIV-infected cells [65], bystander activation-induced cell death (AICD) and apoptosis of neighboring noninfected T cells [66,67], and impaired regenerative capacity, coupled with the destruction of essential hematopoietic progenitor cells and hastened by the chronic immune activation that attends HIV infection [13,68,69]. As early as 1984, Kotler et al. [70] observed histological abnormalities and lymphocyte depletion in GALT mucosa of infected individuals and first suggested a role for the mucosal immune system in disease progression. More recently, in HIV-infected patients, chronic immune activation has been found to be associated with a higher frequency of circulating T cells with an activated phenotype as well as with increased levels of pro-inflammatory cytokines and chemokines [71–73]. Such chronic CD4⁺ and CD8⁺ T cell activation ultimately leads to clonal exhaustion of memory T cell pools and also provides an increased frequency of target cells for viral infection, thus leading to increased viral loads and systemic dissemination [74,75].

To better define the role of Th17 and Tregs in lentiviral disease progression, we studied SIV infection in a pathogenic model (pigtailed macaques) and in a nonpathogenic model (African green monkeys), and found that disease progression was associated with the loss of Th17 cells and an increase in the frequency of CD4⁺FoxP3⁺ Tregs [76••]. The loss of Th17/Treg balance was associated with sustained systemic immune activation and a shift toward cellular stress pathways, nuclear factor-κB signaling, and Th1 profiles [76••,77••]. The loss of Th17 cells was also evident in blood and in rectosigmoid mucosal biopsies from chronically HIV-infected patients [78] and associated with sustained immune activation, microbial translocation, and disease progression (Favre et al., unpublished observation). Because Th17 cells can enhance host defenses against microbial agents [79–81], thus maintaining the integrity of the mucosal barrier [2,29–31,82,83], loss of Th17 cells in HIV disease might account for an increase in microbial translocation across the gastrointestinal mucosa [76••,78,84•]. A current working model proposes that maintaining robust Th17 function in mucosal tissues during HIV infection may prevent immune activation that would otherwise occur after microbial translocation and spread from the gut [75,85].

Possible mechanisms and consequences of loss of the Th17/regulatory T cells balance

The mechanisms responsible for selective depletion of Th17 cells during pathogenic SIV and HIV infection remain undefined. As IFNγ can directly impair subsequent Th17 development [86], the Th1/interleukin-12 prone inflammatory environment generated during pathogenic HIV and SIV infection may tilt the balance in the favor of Treg differentiation. However, as preferential Th17 depletion occurs quite quickly (e.g., in a matter of days after acute infection in nonhuman primates) [76••,87,88•], it seems more likely that destruction of existing Th17 cells must also occur. At this point, it is not clear whether these cells are preferentially infected by virus [88•] or, instead, indirectly destroyed as bystanders [78]. Cecchinato et al. [87] first showed that Th17 cells were infected by SIV and that depletion of Th17 cells in mucosal tissues strongly correlated with viral loads. Kader et al. [88•] extended this analysis to show that most Th17 cells in the peripheral blood were resting CD4⁺ T cells expressing high levels of integrin α4β7 heterodimer, and were preferentially infected and depleted during primary SIV infection. Ancuta et al. also found that peripheral blood memory CD4⁺ T cells expressing the chemokine receptor, CCR6 (a phenotype associated with Th17 cells and also Tregs [21,89]), displayed increased susceptibility to viral infection by CCR5 and CXCR4-tropic viruses in vitro and increased viral DNA content in HIV-infected individuals (Gosselin et al. Keystone 2009 and P. Ancuta, personal communication). Together, these results suggest that preferential HIV infection may account, at least in part, for the selective depletion of Th17 cells.

A number of considerations suggest, however, that viral and/or nonviral bystander mechanisms of selective Th17 depletion and Treg expansion are also likely at play. First, preferential infection of Th17 cells is not always found in viremic patients [78] or in SIV-infected macaques [87]. Second, the same virus inoculated in different hosts can result in massive Th17 depletion (in pathogenic infection in macaque) or not (in nonpathogenic infection in African green monkeys), despite similar viral loads [76••]. Finally, and perhaps most importantly, Tregs (and, to some degree, Th1 cells) are instead increased in fraction and/or numbers in lymphoid tissues during progressive HIV disease [53,76••, 90], even though they share common features with Th17 cells, including memory phenotype, chemokine receptors (CCR6, CCR4), homing properties (CCL20) [21], levels of CCR5 expression [89], and the propensity for lentiviral infection in vitro [91].

Apoptosis and AICD of noninfected CD4⁺ T cells have been correlated with chronic immune activation and inflammation during HIV disease progression [66,67]. One might speculate that selective loss of Th17 cells is due to the fact that they are more susceptible to cell death caused by such activation than are Tregs. For example, an intriguing developmental link exists between indoleamine 2,3-dioxygenase (IDO) metabolism and the differentiation of Th17 and Tregs from naive T cells. IDO is the rate-limiting enzyme involved in the catabolism of the amino acid tryptophan through the kynurenine pathway [92]. Tryptophan catabolites are able to induce the expression of FoxP3 and the generation of Tregs, and to suppress the expression of RORc and the generation of Th17 cells [93,94••]. Similarly, IDO-mediated tryptophan deprivation and the amino acid starvation response can induce Treg development and blunt Th17 conversion [95•,96]. Predominantly found in macrophages and dendritic cells, the expression of IDO is upregulated by interferons and by agonists of toll-like receptors [92]. When catalytically active, the enzyme suppresses T cell responses in a variety of settings, including autoimmune disorders [97], allograft rejection [98], viral infections [99], cancer [100], and pregnancy [101]. Such suppression is thought to occur either because IDO depletes the essential amino acid tryptophan and/or because it produces tryptophan catabolites that are toxic to T cells [102,103]. In either case, the ability of IDO to suppress immune responses has raised the possibility that it may contribute to the immunodeficiency found in individuals with progressive HIV disease [104]. As IDO metabolism is related both to the Treg to Th17 developmental switch and to HIV pathogenesis [104], we have explored the relationships between HIV disease, Th17 and Treg subsets, and IDO metabolism. In ongoing studies, we have now shown that enhanced IDO activity is associated with HIV disease progression, and that such activity results in an imbalance of Th17 cells and Tregs in the peripheral blood and in rectosigmoid tissue that is mediated by the tryptophan catabolite, 3-hydroxyanthranilic acid (Favre and Mold et al., submitted).

Although IDO-mediated depletion of Th17 cells could occur in concert with other mechanisms (e.g., bystander CD4⁺ T cell death through interactions with gp120/gp41 [105–107], Tat [108], Nef [109], or Vpr [110]), the induction of IDO may represent a critical initiating event that results in inversion of the Th17/Treg balance and in the maintenance of the chronic inflammatory state of progressive disease (see Fig. 1). This chronic activation of the IDO pathway diminishes the host’s capacity to generate Th17 cells and favors the generation of Tregs. The net outcome is a progressive loss of the mucosal immune barrier that is critically dependent upon Th17 cells coupled with a rise in Tregs, which may dampen effective T cell immune responses to HIV and other pathogenic organisms.

Figure 1. Loss of Th17/regulatory T cells balance and AIDS pathogenesis.

Direct simian immunodeficiency virus (SIV) or HIV infection and/or upregulation of indoleamine 2,3-dioxygenase (IDO) by plasmacytoid (pDCs) and myeloid (mDCs) dendritic cells leads to the preferential depletion of Th17 cells with a reciprocal increase of CD4⁺CD25^highFoxP3⁺ regulatory T cells (Tregs) in peripheral blood and in the mucosa of the gastrointestinal tract (1). Loss of Th17/Treg balance results in progressive disruption of the mucosal barrier, permitting increased translocation of the gut microbial flora (2). Setting up a positive feedback loop, microbial products then activate antigen-presenting cells (APCs) in the mucosa and in lymphoid organs, enhancing local and systemic inflammation and immune activation, and resulting in sustained loss of the Th17/Treg balance (3). The picture of ‘gut mucosa’ represents the colonic mucosa of an SIV-infected monkey, with CD3⁺ T cells stained in black. Bacterial translocation from the lumen to the lamina propria is represented schematically. RORγt, receptor-related orphan receptor-γt.

Conclusion

Although additional work is required to clearly define the mechanisms of selective Th17 depletion as well as to definitively link changes in the Th17/Treg balance with the immunopathology of HIV infection, we believe that further attention to the balance of Th17 and Treg subsets will reveal much about the relationship between disease progression and inflammation. For example, several tryptophan derivatives are known to be natural ligands of the AhR, a receptor also involved in the balance of Th17 and Treg cells in vivo [7•,8]. An intriguing possibility for immune intervention may be to delineate the AhR signaling pathways associated with IDO metabolism so that strategies to blunt and to reverse Th17/Treg loss could be devised. For prophylactic vaccine strategies, on the contrary, it may be favorable to generate combined HIV-specific Th17 and Th1 responses in mucosal tissue that will sustain and/or generate a balanced representation of Th17/Treg and Th1 subsets therein. In this manner, the host may be able to establish an environment that is more likely to minimize inflammation and to limit the extent of viral replication and spread [111••].