How T cells take developmental decisions by using thearyl hydrocarbon receptor to sense the environmentThomas Korn1

Klinikum rechts der Isar, Department of Neurology, Technical University Munich, 81675 Munich, Germany

For the last 25 y or so, T helper(Th)-cell responses have beenclassified according to either Th1or Th2 depending on whether

these T cells produce IFN-γ or IL-4. WhileTh1 cells produce IFN-γ and induce mac-rophage-driven inflammation, Th2 cellssecrete IL-4 and drive eosinophilic in-flammation. Following the discovery ofTh17 cells whose signature cytokine isIL-17A, a vigorous debate on the classifi-cation of Th-cell lineages has been un-leashed. In addition to IL-17, Th17 cellssecrete IL-17F, IL-21, and IL-22. Themain argument for Th17 cells to be dis-tinct from Th1 and Th2 cells was that theycould be generated in the absence of Th1-associated (T-bet, STAT1, and STAT4)and Th2-associated (STAT6) transcrip-tion factors, respectively (reviewed in 1).However, the fact that Th17 cells appearunstable in vivo and the unresolved issueson the role of IL-23 for either their dif-ferentiation or stabilization, as well as thediscovery of even more “T-cell lineages”like Th9 and Th22, have again spurred thedebate on the plasticity of Th cells. Theclassification into Th1, Th2, and Th17 isstill and foremost based on the distinctprofile of cytokines that are produced bythese T-cell subsets. In an effort to char-acterize other genes that would be differ-entially expressed in Th-cell subsets, thearyl hydrocarbon receptor (Ahr) gene wasrecognized to be silent in Th1 and Th2cells but highly transcribed in Th17 cells(2, 3). In a study published in PNAS,Quintana et al. (4) investigate the func-tional relevance of AHR expression byspecific Th-cell subsets in a disease modelof CNS autoimmunity.The AHR system is a ligand-driven

transcription factor system. AHR is a basichelix–loop–helix protein that is seques-tered within a protein complex in thecytosol unless it is bound by one of itsmembrane-permeating ligands. On ligandbinding, the ligand–AHR complex trans-locates to the nucleus, where it dimerizeswith the AHR nuclear translocator andbinds to specific response elements closeto the promoters of its target genes(reviewed in 5). The AHR system has beenwell characterized in terms of its majortarget genes, Cyp1a1 and Cyp1b1, in hep-atocytes. On activation of AHR, theseoxygenases are up-regulated and increasethe intracellular availability of oxidizingagents. However, at the same time, cyto-

chrome oxidases are believed to be themain metabolizing enzymes of natural orendogenous AHR ligands, and thus pro-vide an intrinsic negative feedback loopresulting in only transient induction ofCyp1a1 and Cyp1b1. Notably, AHR is alsothe receptor for a series of xenobioticagents like 2,3,7,8-tetrachlorodibenzo-p-dioxin (TCDD) or polychlorinated bi-phenyls that cannot be metabolized andlead to permanent activation of AHRtarget genes. These ligands only exist inan industrialized world. Therefore, it ishard to imagine that they could haveprovided the selection pressure for thehigh degree of conservation of the AHRsystem during evolution (6). Yet, theAHR system has a role in embryonic de-velopment and was even hijacked by theimmune system in the vertebrate cosmos.AHR is expressed in cells of the innate

immune system, such as dendritic cells,macrophages, natural killer cells, andlymphoid tissue inducer-like cells (7, 8). Incontrast to other danger-sensing systems,such as the TLR pathways, AHR signalsare thought to convey intrinsic metabolicor oxidative stress in a cell type-specificmanner (9). Engagement of AHR withdifferent ligands modulates the expressionof surface molecules of dendritic cells

and the secretion of cytokines witheither proinflammatory or tolerogenicnet effects (4, 9).As an alternative sensing system of the

cellular milieu, AHR is particularly in-teresting because it is differentiallyexpressed in various T-cell subsets andcould have a direct effect on the shapingof specific Th-cell responses (Fig. 1). Incommitted Th17 cells, activation of AHRwith 6-formylindolo-(3,2-b)-carbazole in-duces the production of IL-22 (2). Ac-cordingly, Th17 cells differentiated fromAHR-deficient T cells (disrupted exon 2 ofAhr) lack production of IL-22 and are alsounable to up-regulate IL-22 in response toIL-23. Thus, it is possible that AHRmodulates the responsiveness of Th17cells to IL-23, although a direct impact ofAHR deficiency on the expression of theIL-23 receptor (IL-23R) has not beenformally tested. IL-22 is a member of theIL-10 family of cytokines. Its receptor isa heterodimer, which is composed ofIL-10R2 and IL-22R. The IL-22R is notexpressed on lymphocytes but on epithelialcells and specific parenchymal cells, andIL-22 has an important role in maintainingthe integrity of epithelial barriers (10).Thus, at epithelial barriers and in certainparenchymal organs, Th17 cells may havea protective and beneficial role becauseof their production of IL-22. Interestingly,the T cell-restricted AHR pathway mayalso be involved in inducing IL-10. IL-27,a cytokine produced by innate immunecells under inflammatory conditions, isspecifically able to switch on IL-10 pro-duction in T cells. It has been proposedthat IL-10–producing T regulatory-1(Tr-1) cells develop on stimulation withTGF-β plus IL-27 (11, 12). Mechanisti-cally, IL-27 induces AHR, and AHRcooperates with the transcription factorc-Maf to transactivate the Il10 gene(13, 14) (Fig. 1). This mechanism may bean important way to limit exaggeratedimmunopathology during ongoingtissue inflammation.Thus, the best evidence that the AHR

pathway is directly involved in tuning thephenotype of activated T cells exists forTh17 cells and Tr-1 cells (Fig. 1). Th17

TCnaïve

Th1

Th2

Treg(AHR)Foxp3

Th17AHRROR- t

Tr-1AHRc-Maf

FICZEndogenousligands?

TCDDEndogenousligands?

TCDD, ITEEndogenousligands?

TGF-b+ IL

-6

TGF-b

TGF-b + IL-27

IFN-g, IL-12

IL-4

IL-17IL-22

CD39

IL-10

Fig. 1. Although Th1 (TH1) and Th2 (TH2) cells failto up-regulate AHR, this ligand-driven transcriptionfactor is expressed in Th17 (TH17) cells, Foxp3+

Tregs, and Tr-1 cells. Ligation of AHR with partic-ular ligands, such as TCDD, 6-formylindolo-(3,2-b)-carbazole (FICZ), ITE, or endogenous ligands, likelypromotes and stabilizes the transcriptional pro-gram of the AHR-expressing T-cell subsets (in thetext and refs. 2, 3, 13, 14). It is interesting to hy-pothesize (and with the available tools, it can nowbe tested) whether AHR ligation might also convertprecommitted precursors of one T-cell lineage intoanother T-cell subset, and thus exhibit truly im-munomodulatory properties. TC, T cell.

Author contributions: T.K. wrote the paper.

The author declares no conflict of interest.

See companion article on page 20768.1E-mail: korn@lrz.tu-muenchen.de.

www.pnas.org/cgi/doi/10.1073/pnas.1015420107 PNAS | November 30, 2010 | vol. 107 | no. 48 | 20597–20598

COMMENTARY

cells and Tr-1 cells can be differentiatedfrom naive T cells with TGF-β plus IL-6 orIL-27, respectively. It is possible that thepresence of TGF-β is required for the up-regulation of AHR (15). Consistent withthis idea, AHR is expressed in Th17 cellsthat have been differentiated with TGF-βplus IL-6 but not in Th17 cells that havebeen generated in an alternative mannerin the absence of TGF-β (16). Neither isAHR-expressed in Th1 or Th2 cells, forthe differentiation of which TGF-β hasa profound inhibitory effect. Th17 cellshave a close ontogenetic relationship withFoxp3+ regulatory T cells (Tregs) (17).Tregs and Th17 cells share common fac-tors necessary for their differentiation.Whereas TGF-β induces Foxp3+ in naiveT cells, IL-6 abrogates the TGF-β–driveninduction of Foxp3 and TGF-β and IL-6cooperate to initiate the expression ofROR-γt, which is an essential member ofthe transcription factor complex thattransactivates IL-17. TGF-β plus IL-6 alsolead to the expression of AHR, and it isvery likely that AHR will then sense en-vironmental cues to modulate further de-velopment of the functional phenotype ofa given T cell. Specific AHR ligands (e.g.,high affinity vs. low affinity) might skewthe fate of a T cell in specific niches to-ward either the induced Foxp3+ Tregphenotype or the Th17 phenotype (3).In their recent study, Quintana et al. (4)

provide evidence that activation of AHRis able to influence the generation of in-duced Foxp3+ Tregs in vivo. Mice thatexpress a low-affinity variant of AHR(AHRd) harbor fewer Foxp3+ Tregs intheir mesenteric lymph nodes, most likely

because of reduced conversion of con-ventional T cells into Tregs. Low-affinityligation of AHRd by endogenous ligandsmight fail to limit STAT1 activation,which, in turn, will impair both Treg andTh17 generation, a phenomenon that has

Quintana et al.

investigate the functional

relevance of AHR

expression by specific

Th-cell subsets.

also been described in AHR-deficientmice. In pharmacological interventionswith 2-(1′H-indole-3′-carbonyl)-thiazole-4-carboxylic acid methyl ester (ITE),a natural agonist of AHR, Foxp3+ Tregscould be induced in vivo. Thus, pharma-cological manipulation of the AHR systemwith nontoxic ligands is a potent means tomodulate fate decisions of developingT cells in vivo.What are the real endogenous ligands

for AHR? A number of natural sub-stances have been identified as ligandsfor AHR, including indigoids, equilenin,arachidonic acid metabolites, hememetabolites, tryptophan metabolites, andITE (5). Some natural ligands with ex-tremely high affinity to AHR appear tobe generated in the gastrointestinal tractfrom dietary compounds. For example,the secondary metabolite of gluco-brassicin, indole-3 carbinol, and its higher

order derivatives, indolo[3,2-b]carbazole(ICZ), 3,3′-diindolylmethane, and 2-(indol-3-ylmethyl)-3,3′-diindolylmethane,are generated naturally from cruciferousplants like broccoli in the stomach, andICZ binds to AHR with an affinity of1.9 × 10−10 M, which is only two ordersof magnitude lower than the affinity ofTCDD. It is likely that these metabolitescontribute to shape the gut immunemilieu in vivo. These are testable hy-potheses that have to be addressed infuture studies.Taken together, the discovery of the

importance of the AHR system for thedevelopment of Th-cell subsets providesa molecular platform to explain how di-etary components, the composition of thecommensal gut flora, or the exposure toindustrial chemicals (by the generationof specific AHR ligands) might have animpact on fate decisions of T cells, andthus constitute a framework for the overallimmunological susceptibility to autoim-munity or chronic inflammation of a givenindividual. Further investigation of thisligand-driven transcription factor systemmight not only allow us to understandhow endogenous ligands drive fate deci-sions of T cells but enable us to exploitthis potentially “druggable” system for in-telligent immunomodulatory interventionsin autoimmunity, chronic inflammation,and tumor immunology.

ACKNOWLEDGMENTS. T.K. is supported by theDeutsche Forschungsgemeinschaft (Grants KO2964/3-1, KO 2964/4-1, and KO 2964/5-1) and theGemeinnützige Hertie-Stiftung.

1. Korn T, Bettelli E, Oukka M, Kuchroo VK (2009) IL-17and Th17 Cells. Annu Rev Immunol 27:485–517.

2. Veldhoen M, et al. (2008) The aryl hydrocarbon recep-tor links TH17-cell-mediated autoimmunity to environ-mental toxins. Nature 453:106–109.

3. Quintana FJ, et al. (2008) Control of T(reg) and T(H)17cell differentiation by the aryl hydrocarbon receptor.Nature 453:65–71.

4. Quintana FJ, et al. (2010) An endogenous aryl hy-drocarbon receptor ligand acts on dendritic cellsand T cells to suppress experimental autoimmuneencephalomyelitis. Proc Natl Acad Sci USA 107:20768–20773.

5. Nguyen LP, Bradfield CA (2008) The search for endog-enous activators of the aryl hydrocarbon receptor.Chem Res Toxicol 21:102–116.

6. Powell-Coffman JA, Bradfield CA, WoodWB (1998) Cae-norhabditis elegans orthologs of the aryl hydrocarbonreceptor and its heterodimerization partner the aryl hy-

drocarbon receptor nuclear translocator. Proc Natl AcadSci USA 95:2844–2849.

7. Frericks M, Meissner M, Esser C (2007) Microarray anal-ysis of the AHR system: Tissue-specific flexibility in sig-nal and target genes. Toxicol Appl Pharmacol 220:320–332.

8. Cua DJ, Tato CM (2010) Innate IL-17-producing cells:The sentinels of the immune system. Nat Rev Immunol10:479–489.

9. Matsumura F, Vogel CF (2006) Evidence supporting thehypothesis that one of the main functions of the arylhydrocarbon receptor is mediation of cell stress re-sponses. Biol Chem 387:1189–1194.

10. Ouyang W, Kolls JK, Zheng Y (2008) The biologicalfunctions of T helper 17 cell effector cytokines in in-flammation. Immunity 28:454–467.

11. Stumhofer JS, et al. (2007) Interleukins 27 and 6 induceSTAT3-mediated T cell production of interleukin 10.Nat Immunol 8:1363–1371.

12. Awasthi A, et al. (2007) A dominant function for in-terleukin 27 in generating interleukin 10-producinganti-inflammatory T cells. Nat Immunol 8:1380–1389.

13. Apetoh L, et al. (2010) The aryl hydrocarbon receptor inter-acts with c-Maf to promote the differentiation of type 1regulatoryTcells inducedby IL-27.Nat Immunol11:854–861.

14. GandhiR, et al. (2010)Activationof the aryl hydrocarbonreceptor induces human type 1 regulatory T cell-like andFoxp3(+) regulatory T cells. Nat Immunol 11:846–853.

15. Kimura A, Naka T, Nohara K, Fujii-Kuriyama Y,Kishimoto T (2008) Aryl hydrocarbon receptor regulatesStat1 activation and participates in the development ofTh17 cells. Proc Natl Acad Sci USA 105:9721–9726.

16. Ghoreschi K, et al. (2010) Generation of pathogenic T(H)17 cells in the absence of TGF-beta signalling. Na-ture 467:967–971.

17. Bettelli E, et al. (2006) Reciprocal developmental path-ways for the generation of pathogenic effector TH17and regulatory T cells. Nature 441:235–238.

20598 | www.pnas.org/cgi/doi/10.1073/pnas.1015420107 Korn