helper cell differentiation - Wiley Online Library

DOI: 10.1111/exd.12493

Viewpoint

www.wileyonlinelibrary.com/journal/EXD

Microbe driven T-helper cell differentiation: lessons from Candida albicans and Staphylococcus aureus Christina E. Zielinski1,2 1

Cellular Immunoregulation Group, Department of Dermatology and Allergology, Charit
e-Universit€atsmedizin Berlin, Berlin, Germany; Berlin-Brandenburg-Center for Regenerative Therapies, Berlin, Germany Correspondence: Christina E. Zielinski, MD, Cellular Immunoregulation Group, Department of Dermatology and Allergology, Charit
e–Universit€atsmedizin Berlin, Charit
eplatz 1, 10117 Berlin, Germany, Tel.: +49 30 450 618213, Fax: +49 (0)30 450 518947, email: [email protected] 2

Abstract: T-helper cells integrate signals from their T-cell receptor, co-stimulatory molecules and cytokine receptors to polarize into effector T-helper subsets with specialized functions in antigen clearance or tolerance. To this end, antigen presenting cells and the local microenvironment tailor effector T-helper cells to respond appropriately to microbial challenges. These challenges comprise protection from pathogens on the one hand and tolerance for the commensal microbiota on the other hand. To accomplish these complex tasks, the host immune system needs to be highly specialized and stringently regulated. In this viewpoint article, we will concentrate on how microbes shape human

T-helper cell responses and how this could relate to the emergence of chronic inflammatory diseases. Understanding the intricate communication between adaptive immunity and microbes will be important for the rational design of novel immunomodulatory therapies and also for anticipating infectious complications upon therapeutic intervention with cytokine depleting therapies, such as biologicals in dermatology.

Introduction

billion years of co-evolution with the host (8). The commensal microflora outnumbers host cells at least 10-fold and encodes about 100-fold more genes than the host genome (9). This warrants well-established tolerance mechanisms. Treg cells represent an abundant T-helper cell population in the skin and gut and help to maintain barrier as well as systemic immune homoeostasis (10,11). As germ-free mice display reduced numbers of Treg cells in the colon as well as lower IL-10 expression by Treg cells, the microbiota has been proposed to be involved in the induction of Treg cells (12). For example, clostridia strains from murine as well as human microbiota have been demonstrated to induce Treg cells (13). Recently, it has been proposed as a potential mechanism that short-chain fatty acids, such as butyrate and propionate, produced by commensal micro-organisms during starch fermentation, facilitate extrathymic generation of Treg cells (14). The differentiation of Th17 cells, which are also highly enriched in the gut mucosa, is also selectively driven by a commensal bacterial stain, the segmented filamentous bacteria (SFB) (15). Previously, the ability of SFB to induce serum amyloid A has been suggested as the mechanism to induce Th17 cells. Recently, antigen specificity and MHC class II restricted mechanisms have further been established for SFB driven Th17 cell differentiation in mice (16,17). This has an important impact on mice, as colonization with SFB, through its induction of Th17 cells, prevents Citrobacter rodentium-induced gut inflammation but could also contribute to Th17 cell-mediated autoimmune disease at peripheral sites such as the joints (18). In humans, which do not host SFBs, the colonic commensal microbe Prevotella copri has recently been strongly correlated with new-onset untreated rheumatoid arthritis, although a causal relationship still remains to be established (19). Together, this demonstrates that distinct microbial

The discovery of Th1 and Th2 cells about three decades ago provided a useful conceptual framework for categorizing specialized T-helper cells responses (1). Th1 cells produce IFN-c under the control of their master transcription factor T-bet and efficiently clear intra-cellular bacteria and viruses. Th2 cells, on the other hand, produce IL-4 under the control of Gata-3 and are specialized in the control of parasitic infections (2). However, it was soon realized that this dualism of T-helper cell responses was too simplistic to account for the broad repertoire of T-helper cell functions. But it is only seven years ago that the 20-year old Th1-Th2 paradigm was shaken by the identification of a novel T-helper cell subset, the Th17 cells, and subsequently several more new T-helper subsets (3,4) (Fig. 1). Th17 cells exert their functions by multiple effector mechanisms, most importantly by IL-17-induced recruitment of neutrophils, which further contribute to the elimination of a variety of extracellular bacteria and fungi (5) (Fig. 1). Most of the Th17 cells’ prominence can, however, be attributed to their crucial involvement in the pathogenesis of autoimmune diseases (6). Interestingly, host–microbial interactions are strongly implicated in initiating autoimmunity: Host defense responses have to be followed by counter-inflammatory mechanisms to reduce collateral tissue damage and release of selfantigens. Otherwise, autoimmune diseases might develop. Not surprisingly, exploration of the reciprocal interactions between the host immune system and the microbiota has taken centre stage in the field of immunology over the last few years.

The role of microbes in shaping T-helper cell functions The microbiota in humans is well-tolerated and resides at epithelial surfaces (7). This mutualistic relationship follows almost half a

Key words: immunoregulation – microbiota – T helper cells

Accepted for publication 4 July 2014

ª 2014 The Authors. Experimental Dermatology Published by John Wiley & Sons A/S. Published by John Wiley & Sons Ltd Experimental Dermatology, 2014, 23, 795–798 This is an open access article under the terms of the Creative Commons Attribution-NonCommercial-NoDerivs License, which permits use and distribution in any medium, provided the original work is properly cited, the use is non-commercial and no modifications or adaptations are made.

795

Zielinski

Priming cytokines/ T helper subsets Master transcription factors

Th1

Effector cytokines

Target microbes

IFN-γ

Virus

IL-4, IL-5, IL-13

Parasite

IL-17 IL-10–

C. albicans

Microbial products adjuvants Th2

Naive Th17

IL-17 IL-10+

Dendritic Cell

ThGM-CSF

GM-CSF

S. aureus

?

Figure 1. Human T-helper cell subsets: priming requirements, cytokine profile and microbial antigen specificities. Human T-helper cell subsets can be distinguished based on priming requirements, cytokine profiles and master transcriptional regulators. These factors translate into distinct specializations in the clearance of microbial antigens.

species can be closely linked to pathologies of the immune system. This also has an impact on pathologies at sites distant from the microbe’s residence as evidenced by gut dysbiosis driven rheumatoid arthritis in mice and man. Interestingly, however, the skin microbiota has an autonomous role in controlling local immune homoeostasis. It is not affected by distinct gut microbiota or gut dysbiosis, at least in mice (20). Genome-wide mapping studies in mice have recently demonstrated that certain genetic loci have an impact on the variability of the skin microbiota and could therefore influence disease risk, such as in the case of epidermolysis bullosa acquisita, an autoimmune skin blistering disease (21). In atopic dermatitis, a chronic, relapsing, pruritic inflammatory skin condition, it has long been recognized that disease flares are associated with a decrease in the overall diversity of microbial communities of the skin due to an expansion of Staphylococcus aureus (22,23). Resolution of disease flares is preceded by a restoration of microbial diversity. Mechanistic evidence is, however, still scarce. In psoriasis, a chronic inflammatory skin condition, which is characterized by erythematous, scaly plaques, no consensus micro-organism has been directly linked to disease pathogenesis despite extensive investigations (24). Streptococcal throat infections have, however, for a long time been associated with psoriasis flares, most likely through mechanisms involving molecular mimicry or superantigens (25). This recent appreciation of the role of the microbiota in shaping local as well as systemic immune responses and its impact on health and disease warrants a detailed analysis of the underlying mechanisms. Recently, differential priming requirements for the generation of human Th17 cells have been demonstrated, which depended on the microbial antigen specificity of the respective na€ıve T-helper cell precursors: IL-1b was required for the generation of C. albicans-specific Th17 cells but not for the generation of S. aureusspecific Th17 cells (26). This has important consequences for the functionality of the respective Th17 cells, as IL-1b-dependent Th17 cells display an IL-10—IFN-c+ cytokine profile, while IL-1bindependent Th17 cells co-express IL-10, but not IFN-c (26). This highlights that IL-1b, which is differentially induced by distinct microbes upon engagement of innate receptors on antigen

796

presenting cells, is an important switch factor for determining pro- versus anti-inflammatory Th17 cell functions (27,28) (Fig. 2). Inhibition of IL-1R signalling with the drug anakinra in patients with the cryopyrin-associated periodic syndrome (CAPS), who have increased IL-1b levels due to inflammasome mutations, correlated with increases of anti-inflammatory IL-10 production by T-helper cells (26). Whether IL-1b depletion and thus induction of anti-inflammatory Th17 cells correlates with changes in the microbiota or infectious risks remains to be determined. IL-1b can, however, also have anti-inflammatory functions, if produced by tissue-resident mononuclear phagocytes (MNPs) upon microbiota driven induction. This has recently been demonstrated by its ability to induce GM-CSF production in Type 3 innate lymphoid cells (ILC3), which in turn induce colonic Treg cells through regulatory factors such as retinoic acid and IL-10 (29). ILC3 therefore translate microbial cues into immunoregulatory signals in the intestine. This was also shown to be important for maintaining oral tolerance to dietary antigens (29). In addition, it has recently been shown that IL-1b also has the ability to reduce pro-inflammatory GM-CSF production in human T-helper cells, but not in mouse T-helper cells (3). The relevance of this finding for health and disease remains to be further explored, especially with respect to the potential involvement of GM-CSF in the pathogenesis of multiple sclerosis (30–32). Together, recent research efforts regarding host–microbiota relationships in mouse and man have opened up avenues for the treatment of inflammatory and allergic diseases by manipulation of the microbiota. Conversely, it should also be noted that certain diseases and

C. albicans

S. aureus

IL-1β +

––

Th17

Th17

v

v v

Inflammation

v

v v

Self-regulation

Figure 2. Two types of Th17 cells with different microbial antigen specificities. C. albicans induces pro-inflammatory Th17 cells, which are incapable of selfregulatory IL-10 production, whereas S. aureus induces anti-inflammatory Th17 cells, which can co-produce IL-10. The molecular switch factor for pro- versus anti-inflammatory Th17 cell properties is IL-1b, which is differentially induced by C. albicans but not S. aureus stimulated antigen presenting cells. Red circles indicate IL-17; blue circles indicate IL-10.

ª 2014 The Authors. Experimental Dermatology Published by John Wiley & Sons A/S. Published by John Wiley & Sons Ltd Experimental Dermatology, 2014, 23, 795–798

T cell responses against microbes

treatments modulate the microbiota causing an increased risk of infections by pathogen overgrowth or cause de novo development of autoimmunity.

Modulation of T-helper cell responses by sensors for microbial recognition Toll-like receptors (TLRs), expressed primarily on antigen presenting cells, are known to have critical roles in the initiation of innate immunity (33). As evolutionary conserved pattern recognition receptors, they integrate microbial signals and induce the secretion of cytokines, chemokines and antimicrobial peptides, as well as the maturation of antigen-specific cells and the upregulation of co-stimulatory molecules (34,35). Microbial stimuli include viable, dormant as well as non-viable microbial products, such as free bacterial DNA (36). This translation of microbial stimuli into a TLR-conditioned micromilieu indirectly affects adaptive T-helper cell polarization. However, it has also been reported that T-helper cells themselves express TLRs and are therefore directly receptive to microbial signals (37–39). This particular property of T-helper cells has largely been underappreciated. For example, TLR2 stimulation of murine T-helper cells resulted in enhanced proliferation and Th17 cell polarization in both the early and late stages of their lineage specification (37). This had important consequences for autoimmune brain inflammation in the experimental autoimmune encephalomyelitis mouse model (EAE), because TLR2 deficient mice displayed reduced disease activity (37). TLR2 is receptive to a number of bacterial products, such a lipopeptides. Other TLRs are also increased in Th17 cells, such as TLR4, which makes them more sensitive to LPS signalling (37). Treg cells are also receptive to TLR stimulation. While conflicting data exist with respect to LPS effects through TLR4 signalling in Treg cells (40,41), TLR2 ligands such as Pam3Cys induce proliferation in otherwise anergic Treg cells in vitro and in vivo. TLR stimulation has therefore been considered to assume ‘co-stimulatory’ functions. It can also transiently revert FoxP3 expression and suppressive properties (41,42). Expression of these innate receptors by adaptive immune cells does, however, not per se have to translate into innate properties. This has elegantly been demonstrated by the role of NLR caspase recruitment domain (CARD) containing 5 (NLRC5), which is an intra-cellular protein classically involved in pathogen recognition. In adaptive lymphocytes, it acts, instead, as a transcriptional regulator of major-histocompatibility complex (MHC) class I (43). Therefore, adaptive immune cells, such as T-helper cells, have the ability to integrate microbial signals by their expression of pattern recognition receptors (PRR). This provides them with innate functions, but they may also exploit the expression of those receptors for other purposes, such as those classically associated with adaptive properties.

Technical tools to analyse microbial antigen specificities of human T-helper cell subsets During immune responses, T-helper cells recognize microbial antigens. This initiates clonal expansion and polarization into distinct subsets, which exert tailored responses against the respective microbes. Each subset is classically defined by its cytokine profile, priming requirements and expression of a specific master transcriptional regulator. The recent explosion in the identification of novel T-helper cell subsets, such as Th22 and Th9 cells, which show some limitations in matching the classical subset criteria, has

warranted a readjustment of subset definition and classification criteria. Therefore, profiling active enhancers on a genome-wide scale (enhancer landscapes) is now promoted to better define functional subsets (44–46). Also, the differential expression of chemokine receptor surface markers helps to distinguish T-helper cell subsets based on their preferential migration properties (47,48) (Fig. 3). Together, this categorization of T-helper cells into distinct subsets provides a convenient basis to match T-helper cell function with antigen/microbial TCR specificity. The Sallusto laboratory has recently established a powerful tool to dissect antigen specificities in human T-helper cell subsets using an ‘amplified library method’ (49,50). Here, well-defined T-helper cell subsets are polyclonally expanded and restimulated with antigens of interest, which are presented by autologous monocytes. The original frequency of antigen-specific T cells can be calculated based on the number of non-proliferating cultures and the number of input cells according to the Poisson distribution. This method allows for the quantification of antigen-specific cells reactive against natural unprocessed antigens such as microbial lysates and is independent of MHC class II restriction in contrast to the tetramer technology. It also enables further functional analyses of the responding cultures, such as TCR affinity measures and crossreactivity tests (51). As with most cell-culture-based methods, differential expansion and survival capabilities in mixed cultures cannot completely be excluded and could bias the frequency calculation towards the ‘fittest’ cell type. Overall, this methodology has the potential to enumerate microbe-specific T-helper cells in humans and to reveal a potential enrichment within a distinct T-cell subset. Enrichment of C. albicans-specific T-helper cell subsets within the Th17 cell compartment has already been corroborated with this method (50). Recently, a method has been established to dissect in vitro the molecular differentiation requirements of human microbe-specific T-helper cells (26). Na€ıve T-helper cells are co-cultured with microbe pulsed antigen presenting cells (APC), that is monocytes, in the absence of exogenous polarizing factors. The cytokines that are induced by the respective microbes upon innate receptor

CCR6

CCR4

CXCR3

CCR10

Th1

–

–

+

–

Th2

–

+

–

–

Th17

+

+

–

–

Th22

+

+

–

+

ThGMCSF

–

+

–

+

Figure 3. Detection of T-helper cell subsets by chemokine receptors. The differential expression of the chemokine receptors CCR6, CCR4, CXCR3 and CCR10 on memory T-helper cells allows the detection and ex vivo isolation of multiple distinct T-helper cell subsets.

ª 2014 The Authors. Experimental Dermatology Published by John Wiley & Sons A/S. Published by John Wiley & Sons Ltd Experimental Dermatology, 2014, 23, 795–798

797

Zielinski

engagement on APCs provide the polarization milieu for antigenspecific T-helper cells. Perturbation of the in vitro culture system, that is by adding neutralizing antibodies, allows to systematically dissect the cytokine requirements for T-helper cell polarization. With this method, the need for IL-1b for the generation of C. albicans-specific Th17 cells was established. It also revealed differential priming requirements for Th17 cells, depending on their respective antigen specificity as S. aureus-specific Th17 cells, in contrast to C. albicans-specific Th17 cells, were shown not to be dependent on IL-1b (26). The validity of this in vitro method is demonstrated by the fact that the phenotype of ex vivo isolated C. albicans as well as S. aureus-specific T-helper cells exactly matches that of in vitro-generated microbe-specific T cells (26). Other well-established techniques are available for the analysis of antigen-specific T cells. The MHC-multimers, for example, directly bind the T-cell receptor to its fluorescently labelled MHCpeptide ligand. Here, however, the antigenic epitope needs to be well characterized, which rarely applies to microbial antigens. In addition, a defined peptide has to be restricted to a specific MHC haplotype (51,52). Also, upregulation of activation markers can identify microbe responsive T-helper cells. CD40L (CD154) upregulation, in particular, has proven to be a very reliable marker due to its low ex vivo background expression, its fast upregulation in response to antigen recognition and its abundant expression on virtually all activated T-helper cells (53). Taken together, several technical tools exist, which by themselves or in combination, already have the potential to interrogate the human TCR repertoire for microbial antigen specificities, reveal the association of certain microbes with distinct T-helper

References 1 Mosmann T R, Coffman R L. Annu Rev Immunol 1989: 7: 145–173. 2 Zhu J, Yamane H, Paul W E. Annu Rev Immunol 2010: 28: 445–489. 3 Noster R, Riedel R, Mashreghi M F et al. Sci Transl Med 2014: 6: 241ra280. 4 Park H, Li Z, Yang X O et al. Nat Immunol 2005: 6: 1133–1141. 5 Korn T, Bettelli E, Oukka M et al. Annu Rev Immunol 2009: 27: 485–517. 6 Tesmer L A, Lundy S K, Sarkar S et al. Immunol Rev 2008: 223: 87–113. 7 Harder J, Schroder J M, Glaser R. Exp Dermatol 2013: 22: 1–5. 8 Maynard C L, Elson C O, Hatton R D et al. Nature 2012: 489: 231–241. 9 Ley R E, Peterson D A, Gordon J I. Cell 2006: 124: 837–848. 10 Zielinski C E, Zuberbier T, Maurer M. Curr Opin Allergy Clin Immunol 2012: 12: 498–503. 11 Izcue A, Coombes J L, Powrie F. Annu Rev Immunol 2009: 27: 313–338. 12 Nagano Y, Itoh K, Honda K. Curr Opin Immunol 2012: 24: 392–397. 13 Atarashi K, Tanoue T, Oshima K et al. Nature 2013: 500: 232–236. 14 Arpaia N, Campbell C, Fan X et al. Nature 2013: 504: 451–455. 15 Ivanov I I, Atarashi K, Manel N et al. Cell 2009: 139: 485–498. 16 Yang Y, Torchinsky M B, Gobert M et al. Nature 2014: 510: 152–156. 17 Goto Y, Panea C, Nakato G et al. Immunity 2014: 40: 594–607. 18 Wu H J, Ivanov I I, Darce J et al. Immunity 2010: 32: 815–827.

798

cell functions and to dissect the T-cell polarizing micromilieu induced by specific microbes.

Outlook Our understanding of host–pathogen or host–commensal interactions is still at the beginning despite the recent explosion of interest within an interdisciplinary scientific and clinical community. Insights into the mechanisms of how microbes instruct adaptive T-cell effector functions will provide valuable therapeutic tools for immunomodulation strategies. In particular, chronic inflammatory diseases at barrier sites such as the skin and gut are expected to profit from manipulation of the resident microbiota or their respective effector molecules. Caution is, however, warranted, as therapeutic cytokine depletion can also cause infectious complications. This is occasionally observed upon treatment with biologicals and stresses that therapeutic immunomodulatory therapies can cause shifts in the microbial community, improving autoimmune diseases, on the one hand, but causing infectious complications, on the other hand. Promoting close collaborations between clinicians as well as immunologists will help to translate in a ‘from bedside to bench approach’ clinical responses seen upon such treatments into a better understanding of the intricate regulation of host–microbial interactions in health and disease.

Acknowledgements This work was supported by the Deutsche Forschungsgemeinschaft (SFB650, ZI 1262/2-1) and the Celgene Award of the European Society for Research in Dermatology and the award of the Wolfgang Schulze Stiftung.

Conflict of interest The author has no conflicts of interest.

19 Scher J U, Sczesnak A, Longman R S et al. Elife 2013: 2: e01202. 20 Naik S, Bouladoux N, Wilhelm C et al. Science 2012: 337: 1115–1119. 21 Srinivas G, Moller S, Wang J et al. Nat Commun 2013: 4: 2462. 22 Kong H H, Oh J, Deming C et al. Genome Res 2012: 22: 850–859. 23 Abeck D, Mempel M. Br J Dermatol 1998: 139 (Suppl 53): 13–16. 24 Sanford J A, Gallo R L. Semin Immunol 2013: 25: 370–377. 25 Valdimarsson H, Thorleifsdottir R H, Sigurdardottir S L et al. Trends Immunol 2009: 30: 494– 501. 26 Zielinski C E, Mele F, Aschenbrenner D et al. Nature 2012: 484: 514–518. 27 Zielinski C E. Z Rheumatol 2013: 72: 457–461. 28 Sallusto F, Zielinski C E, Lanzavecchia A. Eur J Immunol 2012: 42: 2215–2220. 29 Mortha A, Chudnovskiy A, Hashimoto D et al. Science 2014: 343: 1249288. 30 Noster R, Riedel R, Mashreghi M-F et al. Sci Transl Med 2014: 6: 241ra80. 31 Codarri L, Gyulveszi G, Tosevski V et al. Nat Immunol 2011: 12: 560–567. 32 El-Behi M, Ciric B, Dai H et al. Nat Immunol 2011: 12: 568–575. 33 Kawai T, Akira S. Nat Immunol 2010: 11: 373– 384. 34 Komai-Koma M, Jones L, Ogg G S et al. Proc Natl Acad Sci U S A 2004: 101: 3029–3034. 35 Percoco G, Merle C, Jaouen T et al. Exp Dermatol 2013: 22: 800–806. 36 Alexeyev O A. Exp Dermatol 2013: 22: 443– 446.

37 Reynolds J M, Pappu B P, Peng J et al. Immunity 2010: 32: 692–702. 38 Reynolds J M, Dong C. Trends Immunol 2013: 34: 511–519. 39 Kabelitz D. Curr Opin Immunol 2007: 19: 39– 45. 40 Caramalho I, Lopes-Carvalho T, Ostler D et al. J Exp Med 2003: 197: 403–411. 41 Sutmuller R P, den Brok M H, Kramer M et al. J Clin Invest 2006: 116: 485–494. 42 Liu H, Komai-Koma M, Xu D et al. Proc Natl Acad Sci U S A 2006: 103: 7048–7053. 43 Neerincx A, Castro W, Guarda G et al. Front Immunol 2013: 4: 397. 44 Samstein R M, Arvey A, Josefowicz S Z et al. Cell 2012: 151: 153–166. 45 Heintzman N D, Hon G C, Hawkins R D et al. Nature 2009: 459: 108–112. 46 Vahedi G, Takahashi H, Nakayamada S et al. Cell 2012: 151: 981–993. 47 Sallusto F, Lanzavecchia A. Eur J Immunol 2009: 39: 2076–2082. 48 Loetscher P, Uguccioni M, Bordoli L et al. Nature 1998: 391: 344–345. 49 Geiger R, Duhen T, Lanzavecchia A et al. J Exp Med 2009: 206: 1525–1534. 50 Zielinski C E, Corti D, Mele F et al. Immunol Rev 2011: 240: 40–51. 51 Bacher P, Scheffold A. Cytometry A 2013: 83: 692–701. 52 Altman J D, Moss P A, Goulder P J et al. Science 1996: 274: 94–96. 53 Frentsch M, Arbach O, Kirchhoff D et al. Nat Med 2005: 11: 1118–1124.

ª 2014 The Authors. Experimental Dermatology Published by John Wiley & Sons A/S. Published by John Wiley & Sons Ltd Experimental Dermatology, 2014, 23, 795–798