th17 cells induce a distinct graft rejection response that ... · of effector mechanisms able to...

American Journal of Transplantation 2012; 12: 835–845Wiley Periodicals Inc.

C© Copyright 2012 The American Society of Transplantationand the American Society of Transplant Surgeons

doi: 10.1111/j.1600-6143.2011.03971.x

Th17 Cells Induce a Distinct Graft Rejection ResponseThat Does Not Require IL-17A

E. I. Agorogiannis, F. S. Regateiro, D. Howie,

H. Waldmann and S. P. Cobbold∗

University of Oxford, Sir William Dunn School ofPathology, South Parks Road, Oxford OX1 3RE, UK*Corresponding author: Stephen P. Cobbold,[email protected]

IL-17A-producing helper T (Th17) cells have been impli-cated in the pathogenesis of autoimmune disease, in-flammatory bowel disease and graft rejection, howeverthe mechanisms by which they cause tissue damageremain ill-defined. We examined what damage Th17cell lines could inflict on allogeneic skin grafts in theabsence of other adaptive lymphocytes. CD4+ Th17 celllines were generated from two TCR transgenic mousestrains, A1(M).RAG1−/− and Marilyn, each monospe-cific for the male antigen Dby. After prolonged in vitroculture in polarizing conditions, Th17 lines producedhigh levels of IL-17A with inherently variable levels ofinterferon gamma (IFNc ) and these cells were able tomaintain IL-17A expression following adoptive trans-fer into lymphopenic mice. When transferred into lym-phopenic recipients of male skin grafts, Th17 lineselicited a damaging reaction within the graft associ-ated with pathological findings of epidermal hyper-plasia and neutrophil infiltration. Th17 cells could befound in the grafted skins and spleens of recipients andmaintained their polarized phenotype both in vivo andafter ex vivo restimulation. Antibody-mediated neu-tralization of IL-17A or IFNc did not interfere with Th17-induced pathology, nor did it prevent neutrophil in-filtration. In conclusion, tissue damage by Th17 cellsdoes not require IL-17A.

Key words: graft rejection, mouse skin allograft, T-cellcytokine, T helper cells,

Abbreviations: APC, antigen-presenting cell; BMDCs,bone marrow-derived dendritic cells; ChIP, chromatinimmunoprecipitation; cDNA, complementary DNA;FBS, fetal bovine serum; FACS, fluorescence-activatedcell sorting; CSF3, granulocyte colony stimulating fac-tor; CSF2, granulocyte-macrophage colony stimulatingfactor; GVHD, graft-versus-host disease; rm, recombi-nant murine; H&E, hematoxylin and eosin; IHC, im-munohistochemical; IFNc , interferon gamma; IMDM,Iscove’s modified Dulbecco’s medium; MPO, myeloper-oxidase; nM, nanomolar; PMA phorbol myristate ac-etate; PBS, phosphate-buffered saline; (RT-)PCR, (re-verse transcriptase) polymerase chain reaction; RAG,

recombination activating gene; SD, standard devia-tion.

Received 5 September 2011, revised 17 November 2011and accepted for publication 12 December 2011

Introduction

Th17 cells can orchestrate a wide range of proinflammatoryactivities in both normal host defense and autoimmunedisease pathogenesis (1,2). A contribution to graft rejectionhas been proposed (3), but despite early claims for a roleof IL-17A (4–10), convincing experimental evidence for theinvolvement of Th17 cells in this process has only recentlyemerged (11–13). Further work has implicated Th17 cellsin dysfunction and acute rejection of lung transplants (14–16), in chronic kidney rejection (17) and also graft-versus-host disease (GVHD) (18). Nonetheless, the redundancyof effector mechanisms able to induce transplant damage(19) has not allowed for clear evidence of a specific role ofTh17 cells in this setting. Even though antagonism of IL-17Aactivity could interfere with forms of transplant rejection(5–7), Th17 cells are not essential to the process, as Th1or Th2 cell lines were able to elicit graft rejection in theabsence of other lymphocytes (20).

Th17 cells can mediate graft rejection of MHC-II-mismatched cardiac allografts in T-bet-deficient mice (13).IL-17A produced by CD4+ T cells and neutrophils wereshown to be important in this case. For rejection of fullyMHC-mismatched heart transplants by T-bet-deficient micethe main source of IL-17A was CD8+ T cells (21,22). Pre-vious findings had revealed increased neutrophil infiltra-tion within grafts in IFNc (23) or IL-4 deficient recipients(24), and depletion of neutrophils was associated with pro-longed transplant survival. The involvement of IL-17A wasnot assessed. There may, however, be a necessary rolefor Th17 cells in rejection of male skin grafts by femaleC57BL/6 hosts that absolutely depended on IL-17A andneutrophil infiltration (12). Th17 cells were detected earlyin both skin grafts and draining lymph nodes, their abun-dance decreasing over time paralleled by the emergenceof IFNc -producing CD8+ T cells (12).

As previous studies did not exclude the presence of otherlymphocyte populations, the exact contribution of Th17cells to transplant rejection remained unclear. We needed

835

Erick

Destacar

Agorogiannis et al.

to know whether Th17 cells could induce graft rejectionin their own right or whether they merely form part of aninflammatory cascade involving other types of T cells. Herewe describe the effector properties of long-term polarized,stable Th17 cells in single minor histocompatibility antigen(H-Y)-mismatched transplantation, in the absence of otherT- or B-cell populations.

Materials and Methods

Mice

C57BL/6, CBA/Ca, Marilyn (25) bred to a C57BL/6 recombination ac-tivating gene 1 deficient (RAG1−/−) background, A1(M).RAG1−/− (20),C57BL/6.RAG1−/− and CBA/Ca.RAG1−/− mice were bred and maintained inthe specific pathogen-free animal facility of the Sir William Dunn School ofPathology (Oxford, UK), and used at the age of 5–25 weeks. All experimentswere performed in accordance with the Animals (Scientific Procedures) Act1986.

Generation of long-term CD4+ T-cell lines

For the generation of long-term Th1 and Th17 lines, A1(M).RAG1−/− orMarilyn naive CD4+ T cells, purified using the CD4+ T cell isolation kit(Miltenyi Biotec, CD4+ cell purity >80%), were stimulated weekly in thepresence of female syngeneic mitomycin C-treated T-cell-depleted spleno-cytes and cognate peptide (100 nM) under the following conditions: forTh1 cells 5 ng/mL recombinant murine (rm)IL-12, 10 ng/mL rmIL-2 and10 lg/mL anti-IL-4 (clone 11B11) (26) and for Th17 cells 2 ng/mL recom-binant human TGFb1, 50 ng/mL rmIL-6, 50 ng/mL rmIL-21, 10 lg/mL anti-IFNc (clone XMG1.2) (27) and 10 lg/mL anti-IL-4. For the preparation of T-cell-depleted splenocytes, mice were injected with 2 mg of each YTA3.1.2and YTS191.1.2 (anti-CD4), and YTS156.7.7 and YTS169.4.2.1 (anti-CD8)antibodies (28,29) on days −5 and −3, and spleens were recovered on day0. Recombinant cytokines were purchased from R&D Systems and Pepro-Tech EC. Peptide sequences recognized by T cells from A1(M).RAG1−/−

and Marilyn mice were REEALHQFRSGRKPI (Dby479–493) in the context ofH2Ek and NAGFNSNRANSSRSS (Dby608–622) in the context of H2Ab respec-tively (30). At the end of each activation round, live cells were separatedon a histopaque (Sigma-Aldrich, St. Louis, MO, USA) gradient and culturedagain with fresh splenocytes, antigen, cytokines and neutralizing antibod-ies. Th1 cells were cultured in RPMI 1640 with 10% fetal bovine serum(FBS), whereas for Th17 cells IMDM with 10% FBS was used.

Measurement of cell proliferation using tritiated (3H) thymidine

To assess T-cell proliferation, CD4+ T cells were cultured in 96-well roundbottom plates at 5×104 per well with 1×104 bone marrow-derived dendriticcells (BMDCs) or 25×104 T-cell-depleted splenocytes in the presence ofthe indicated conditions. Generation of BMDCs has been described before(31). Cells were incubated for a total of 72 h, for the last 18–24 h in thepresence of 0.5 lCi 3H-thymidine (Amersham, Buckinghamshire, UK) perwell of culture. Incorporation of 3H-thymidine was assessed using a flat-bedscintillation counter (Perkin-Elmer, Waltham, MA, USA). Data are expressedas means and standard deviations (SD) of triplicate cultures.

In vivo T-cell transfers and skin transplantation

AutoMACS-isolated naive CD4+ T cells or live-cultured CD4+ T cells sepa-rated on a histopaque gradient were resuspended at appropriate concentra-tions in phosphate-buffered saline (PBS) and injected in the lateral tail vein.Control mice were injected with PBS. When necessary, skin transplantationwas performed 1 day after injection, as previously described (28,29). Skingrafts were considered rejected when no viable tissue was visible.

In vivo antibody treatment

Neutralizing antibodies against IFNc (clones XMG1.2 and R46-A2, both ratIgG1) (27,32) and IL-17A (ab13, UCB Celltech) were injected intraperitoneally

at the indicated time points and concentrations. Isotype controls were an-ticanine antibody YCATE55 (rat IgG1) (33) and murinized IgG104.1 (isotypecontrol for anti-IL-17A antibody, UCB Celltech, Brussels, Belgium).

Imaging of skin grafts

Mice were anesthetized using isoflurane (IsoFlo, Abbott Animal Health,Abbott Park, IL, USA) and black and white photos of skin grafts were takenusing a Maestro in vivo imaging system (CRi) weekly starting on day 7 aftertransplantation. The skin graft surface area was calculated using AdobePhotoshop CS2 software (version 9.0.2). Data are expressed as relativegraft surface area normalized to the baseline area measured on day 7 aftergrafting. Color photos of skin grafts were taken with an Olympus SP-500UZ digital camera.

Histological analysis of skin grafts

Skin grafts were recovered and fixed in a solution of 4% wt/vol formalde-hyde in PBS. After 24 h, skin grafts were embedded in paraffin and mi-croscopic sections were stained with hematoxylin and eosin (H&E) at thehistology facility of the Sir William Dunn School of Pathology. Immunohis-tochemical (IHC) analysis was performed with polyclonal rabbit antihumanmyeloperoxidase (Dako, Glostrup, Denmark, A0398) and antihuman CD3(clone SP7, rabbit IgG, Lab Vision). Photos of histological slides were takenwith a Nikon CoolScope digital microscope (Nikon, Tokyo, Japan).

Fluorescence-activated cell sorting (FACS) analysis

Live cells separated on a histopaque gradient or total splenocytes recov-ered from mice injected with CD4+ T cells, were restimulated for 4–5 h with50 ng/mL phorbol myristate acetate (PMA) and 500 ng/mL ionomycin, forthe last 2–3 h in the presence of 10 lg/mL brefeldin A (all from Sigma-Aldrich). Restimulated cells were stained for membrane markers and sub-sequently fixed with formaldehyde and permeabilized with the saponin so-lution (Sigma-Aldrich) in PBS before staining with anticytokine antibodies.Antibodies used were anti-CD3e (145–2C11), anti-CD4 (RM4–5), anti-Vb6(RR4–7), anti-Vb8 (F23.1), anti-IL-17A (TC11–18H10.1), anti-IFNG (XMG1.2)(all from BD Biosciences, Franklin Lakes, NJ, USA) and anti-FoxP3 (FJK-16s) (eBioscience, San Diego, CA, USA). Cells were analyzed on a 4-colourFACSCalibur flow cytometer (BD Biosciences). Data were analyzed usingFlowJo software version 7.6.1 (Tree Star, Ashland, OR, USA). In all FACSplots, numbers in gates represent percentages of cells.

Analysis of gene expression

DNase-treated total RNA was purified using the SV total RNA isolation sys-tem (Promega) from cell lysates or from skin graft samples homogenizedusing a FastPrep-24 system (MP Biomedicals, Irvine, CA, USA). The Super-Script III first-strand synthesis system for reverse transcriptase polymerasechain reaction (RT-PCR) with random hexamers (Invitrogen, Carlsbad, CA,USA) was used to synthesize complementary DNA (cDNA) from 1 lgof total RNA. TaqMan gene expression assays (Applied Biosys-tems, Foster City, CA, USA) for Hprt1 (Mm01545399_m1), Cd3g(Mm00438095_m1), Il17c (Mm00439619_m1), Ifng (Mm00801778_m1),Mpo (Mm00447886_m1), Rorc (Mm03682796_m1) and Tbx21(Mm00450960_m1) were used for real-time RT-PCR analysis on theApplied Biosystems 7500 Fast Real-Time PCR System with SDS softwarev1.3.1. Using the 2−��Ct method, expression of a target gene wasnormalized to an endogenous control gene, and data were presentedrelative to the same sequence in a calibrator sample (34).

Chromatin immunoprecipitation (ChIP) analysis

The “fast ChIP” method (35) was used for the immunoprecipitation ofhistone 3 trimethylated at lysine 4 (H3K4me3) or 27 (H3K27me3) withanti-H3K4me3 (17–614, Millipore, Billerica, MA, USA) and anti-H3K27me3(17–622, Millipore) antibodies, respectively. Real-time PCR was performed

836 American Journal of Transplantation 2012; 12: 835–845

Th17-Mediated Graft Rejection

using SYBR green master mix (Applied Biosystems) on the Applied Biosys-tems 7500 fast real-time PCR system with SDS software v1.3.1. The den-sity of an immunoprecipitated factor (target) at a locus was expressedrelative to 10% input DNA according to the equation “relative density= 2Ct(input)−Ct(target)”, where Ct(input) and Ct(target) indicate the thresh-old PCR cycles for amplification of input DNA and immunoprecipitatedfactor-associated DNA respectively. The following primers were used forreal-time PCR analysis; RORc t promoter: 5′-CATTCCTAGGCCCGCTGAA-3′, 5′-GGGACACACCCTGCCAAAG-3′ (36); T-bet promoter: 5′-TGGGC-ATACAGGAGGCAGCA-3′, 5′-TCGCTTTTGGTGAGGACTGAAG-3′ (37).

Statistical analysis

Where indicated, statistical analysis was performed using GraphPad Prism(version 5.01) for Windows (GraphPad Software, San Diego, CA, USA) withunpaired two-tailed Student’s t-test. p-Values were classified as not signifi-cant (ns; p > 0.05), significant (∗0.01<p < 0.05), very significant (∗∗0.001< p < 0.01) or extremely significant (∗∗∗p < 0.001).

Results

Prolonged culture under polarizing conditions confers

greater stability of IL-17A expression

Previous studies have highlighted the propensity of Th17cells, generated in short-term cultures, to lose IL-17A ex-pression and up-regulate IFNc following transfer in vivo(38–43). Although such phenotypic flexibility might be aphysiological property, it does not allow a clear interpre-tation of adoptive transfer experiments. We hypothesizedthat repetitive stimulation in the presence of polarizing con-ditions would enable the generation of Th17 cells with amore stable phenotype. As shown in Figure 1A and B, long-term culture of either A1(M).RAG1−/− or Marilyn CD4+ Tcells in the presence of TGFb, IL-6, IL-21 plus neutraliz-ing anti-IFNc and anti-IL-4 antibodies, led to a persistentlyhigh production of IL-17A with variable levels of IFNc co-expression. Importantly, even after 160 days of culture,cells still maintained specific antigen-dependent prolifera-tion. When transferred into female RAG1−/− hosts, lackingany endogenous B or T cells, these long-term Th17 lineswere able to maintain expression of IL-17A in the majorityof cells and did not generate any detectable IL-17A−IFNc +

cells (Figure 1C), suggesting they could be used for in vivotransfer studies.

Th17 cells induce a slow skin graft rejection

In order to test the direct influence of Th17 cells ongraft survival, female C57BL/6.RAG1−/− hosts were in-fused with Marilyn “long-term” Th17 cells, and subse-quently grafted with male C57BL/6.RAG1−/− skin. As pos-itive controls we monitored graft rejection caused by naiveMarilyn CD4+ T cells or “long-term” Th1 cells. Flow cy-tometry analysis validated the Th1 and Th17 cells asproperly polarized (Figure 2A). As shown in Figure 2B,both Th1 and Th17 cells were able to enter skin grafts,as transplants recovered on day 7 from Th1- or Th17-injected hosts had increased levels of CD3c mRNA, com-pared to PBS-injected mice. Nevertheless, grafts on Th1-injected hosts had higher levels of CD3c transcripts,

Figure 1: Prolonged culture of Th17 cells under polarizing con-

ditions confers greater stability of IL-17A expression. (A–B)Summary FACS analysis for the expression of IL-17A and IFNcfrom n = 3 A1(M).RAG1−/− Th17 lines growing for 77 days inthe presence of TGFb, anti-IFNc and anti-IL-4, with either IL-6or IL-21, or both and n = 4 Marilyn Th17 lines growing for 135days in the presence of TGFb, IL-6, anti-IFNc and anti-IL-4, withor without IL-21 and/or IL-23. IL-23 was used at 20 ng/mL. Num-bers next to gated populations indicate percentages presentedas means ± SD. Also shown is antigen-dependent proliferationby A1(M).RAG1−/− (A) and Marilyn (B) Th17 cells growing in thepresence of polarizing conditions for 160 and 57 days, respec-tively. For 3H-thymidine incorporation study, cells were restimu-lated for 3 days with female syngeneic splenocytes or BMDCsunder different conditions. Peptides were used at a concentrationof 100 nM, unless otherwise indicated. (C) A1(M).RAG1−/− Th17cells growing in the presence of polarizing conditions for 162 dayswere transferred into female CBA/Ca.RAG1−/− mice. Mice weresacrificed on days 10 or 21, and FACS analysis was performed onsplenocytes restimulated with PMA and ionomycin. Numbers nextto gated populations indicate percentages presented as means ±SD of n = 3 mice/group. FACS plots are gated on live CD4+ cells (Aand B) or live CD3+CD4+ splenocytes (C). Data are representativeof at least two experiments.

suggesting a more efficient migration and/or increasedproliferation of Th1 cells (Figure 2B). The total numbersof Vb6+ cells detected in spleens of graft recipients onday 22 after transplantation were also higher when Th1cells were injected (Figure S1), suggesting that Th17

American Journal of Transplantation 2012; 12: 835–845 837

Agorogiannis et al.

Figure 2: Th17 cells enter skin grafts

and induce distinct transplant dam-

age compared to Th1 or naive CD4+T cells. A total of 0.1×106 Marilyn naiveCD4+ T cells or 10×106 long-term cul-tured Th1 or Th17 cells were transferredinto female C57BL/6.RAG1−/− hosts onday −1. Other mice were injected onlywith PBS. Mice were grafted with maleor female C57BL/6.RAG1−/− tail skin onday 0. (A) Phenotype of Marilyn long-term Th1 and Th17 cells before trans-fer. (B) Real-time RT-PCR analysis for theexpression of CD3c , IL-17A, IFNe andMPO in male skin grafts recovered on day7. Data are normalized to expression ofHPRT1 and presented as means and SD ofn = 3 mice/group, except for Th17-injected mice (n = 5). Grafts on PBS-injected mice were used as calibrator. Sta-tistical analysis was performed using un-paired two-tailed Student’s t-test. (C) Rep-resentative photographs (not to scale) ofmale skin grafts on mice injected withnaive CD4+ T cells or PBS only, and fe-male or male skin grafts on Th17-injectedmice (day 14). White-dashed rectanglesare indicative of the position of the graftonly, and do not represent the measure-ments described in (D). (D) Survival curvesare shown for male skin grafts on PBS-,CD4+- or Th1-injected mice, whereas therelative surface area is presented at theindicated time points as mean ± SD formale grafts on PBS-injected hosts, andmale or female grafts on Th17-injectedmice. Statistical analysis was performedusing unpaired two-tailed Student’s t-testin order to compare male grafts on PBS-and Th17-injected mice.

cells might also have a reduced ability to homeostaticallyexpand in lymphopenic recipients compared to Th1cells.

As expected, intragraft levels for IL-17A and IFNc mRNAwere higher after transfer of Th17 and Th1 cells respec-tively (Figure 2B). Myeloperoxidase (MPO) expression atthis time point, a measure of neutrophil infiltration, wasnot increased by the transfer of either Th1 or Th17 cellsabove the background levels induced by nonspecific inflam-matory changes associated with skin grafting (Figure 2B).Similar findings were observed following the transfer of

A1(M).RAG1−/− Th17 cells into female CBA/Ca.RAG1−/−

hosts and transplantation with male CBA/Ca.RAG1−/− skin(Figure S2A).

Both Th1 and naive CD4+ T cells induced acute rejectionof male skin grafts (Figure 2C and D), manifesting as mas-sive epithelial necrosis and scab formation within 2 weeksafter grafting, followed by contraction and replacement byscar tissue. Male grafts on PBS-injected hosts remainedhealthy, similar to female grafts on Th17-injected mice(Figure 2C and D). Nevertheless, Marilyn Th17 cells did notinduce a classic acute rejection, but caused a number of



rapid pathological changes in male skin grafts (Figure 2C),characterized by hair loss and focal erythema that soonresolved, and succeeded by a gradual loss of graft volumethat could be followed over time (Figure 2D). A similarpattern of rejection was induced by A1(M).RAG1−/− Th17cells (Figures S2B and C). Forty-eight days after transplan-tation, residual tissue could still be observed in skin graftson hosts that received Th17 cells (not shown), so we couldnot define a specific time of rejection, but the absence ofany normal skin texture (e.g. hair or dermal ridges) wasproof for the potential of transferred Th17 cells to induceantigen-specific tissue damage in the absence of otherlymphocyte populations.

Th17 cells promote unique pathological changes and

neutrophil infiltration

The differential rejection phenotypes induced by Th1 andTh17 cells were further investigated by histological analy-sis. As shown in Figure 3A, PBS-injected hosts retainedhealthy grafts, whereas both Th1 and Th17 cells wereable to induce distinct pathological patterns explaining themacroscopic changes. As early as day 7 after transplanta-tion, Th1 cells had elicited dermal hemorrhage, signalingimminent scab formation, and epithelial damage had led toepidermal thinning (Figure 3A). On the contrary, skin graftson Th17-injected hosts showed epidermal hyperplasia, lossof skin appendages (hair follicles and sebaceous glands)and extensive dermal inflammation (Figure 3A), withoutany obvious signs of inflammatory involvement of the vas-culature. Such histological findings were reproducible inthe A1(M).RAG1−/− transplantation system (Figure S2D).In order to further characterize the Th17-mediated inflam-matory reaction, IHC staining for CD3 and MPO was per-formed in grafts from mice injected with Th17 cells. Figure3B shows that on day 48 after transplantation dense MPO+

cell infiltrates were present in these grafts, suggesting neu-trophils as mediators of Th17 effector function, whereasthe presence of CD3+ cells verified efficient homing ofTh17 cells into grafted tissue. As no difference was ob-served in MPO mRNA levels between grafts from PBS-and Th17-injected mice on day 7 (Figures 2B and S2A), itis likely that neutrophil infiltration remains active at laterstages during Th17-mediated rejection.

Transferred Th17 cells maintain memory of IL-17A

expression in vivo

We subsequently examined the phenotype of transferredcells in spleens of grafted mice. Th1 cells maintained theirphenotype, and naive CD4+ T cells differentiated into Th1cells expressing high levels of IFNc , but no IL-17A (Figure4A), IL-22 or IL-4 (data not shown). Most Marilyn-derivedTh17 cells maintained IL-17A expression. There was, how-ever, an increase in the percentage of IL-17A−IFNc + cellsthat represented only a minor fraction prior to transfer (Fig-ure 4A). This contrasted with A1(M).RAG1−/− Th17 cellsthat maintained IL-17A expression without up-regulation ofIFNc (Figure S3A). The phenotype of transferred Marilyn-

Figure 3: Th17 cells induce pathological changes distin-

guishable from Th1-mediated rejection and drive intragraft

neutrophil infiltration. (A) Female C57BL/6.RAG1−/− mice, in-jected with Marilyn Th1 or Th17 cells, were grafted with maleC57BL/6.RAG1−/− skin on day 0, as described in Figure 2. Othermice were injected with PBS only. H&E staining was performed ongrafts from mice killed on day 7. Different examples of skin graftsare shown. In grafts from PBS-injected hosts, there is a normalappearance of epidermis and dermis, with retention of hair folli-cles and sebaceous glands. Th1 cells-induced epithelial damagewith thinning of the epidermis and red blood cell extravasation,indicative of dermal hemorrhage. Th17 cells-promoted epidermalhyperplasia and dermal inflammatory cell infiltration, associatedwith loss of skin appendages. (B) Female C57BL/6.RAG1−/− micewere injected with PBS or Marilyn Th17 cells on day −1, andgrafted with male C57BL/6.RAG1−/− skin on day 0, as describedin Figure 2. IHC staining for CD3 and MPO (in brown) was per-formed on grafts from mice killed on day 48. Note the abundanceof CD3+ and MPO+ cells in grafts from Th17-injected hosts rela-tive to PBS-injected mice.

derived T cells was also examined after in vitro restimula-tion in the presence of Dby608–622 peptide. After 4 days ofculture, the vast majority of CD4+ and Th1 cells expressedIFNc (Figure 4B). The expression of both IL-17A and IFNcby Th17 cells increased, and IL-17A+IFNc + cells now repre-sented the most abundant population (Figure 4B). A smallpercentage of Th17 cells were seen to up-regulate FoxP3in vivo, although they lacked any FoxP3 expression beforetransfer (Figure S4). It is unclear whether these FoxP3+

cells can interfere with Th17-mediated rejection.

As Th17 lines maintained memory of IL-17A expres-sion in vivo, unlike short-term cultured cells in previous


Agorogiannis et al.

Figure 4: Maintenance of permissive hi-

stone modifications in the Rorc locus

allows transferred Th17 cells to retain

IL-17A expression, while up-regulating

IFNc , in recipients of male skin grafts. (A)Female mice injected with Marilyn CD4+(n = 4), Th1 (n = 3), or Th17 (n = 7) cells,and grafted with male skin as describedin Figure 2, were killed on day 48 aftertransplantation, and FACS analysis was per-formed on splenocytes restimulated withPMA and ionomycin. Expression of IL-17Aand IFNc by Vb6+ cells is shown. (B) To-tal splenocytes from (A) were stimulatedat a density of 2.5×105 cells per well in96-well plates with 100 nM Dby608–622 inIMDM. After 4 days, cells were restimu-lated with PMA and inomycin and analyzedby flow cytometry for expression of IL-17Aand IFNc . All FACS plots are gated on liveVb6+ cells. Numbers next to gated popu-lations indicate percentages presented asmeans ± SD. (C) ChIP analysis of long-term cultured Marilyn Th17 cells stimulatedfor 7 days in the presence (Th17) or ab-sence (Th17medium) of polarizing conditions.As control, long-term cultured Marilyn Th1cells were analyzed. Real-time PCR wasperformed using DNA from samples im-munoprecipitated with antibodies againstH3K4me3 and H3K27me3. Data are nor-malized to 10% input DNA and presentedas ratios of H3K4me3 to H3K27me3 den-sity at the promoters of RORc t (Rorc+5.0)and T-bet (Tbx21) genes. Means and SD ofn = 3 technical replicates from one experi-ment (representative of two) are shown.

studies (38,40), we tested transferred cells for histone 3modifications in the promoters of RORc t (Rorc+5.0) andT-bet (Tbx21), two gene loci regulating fidelity to the Th17phenotype. RORc t drives the differentiation program ofTh17 cells and induces IL-17A expression (44), whereasT-bet has been implicated in the phenotypic instability ofTh17 cells (38,40,45). ChIP analysis was performed on Mar-ilyn long-term Th17 lines stimulated for 7 days in vitro inthe presence or absence of polarizing cytokines. As shownin Figure 4C, withdrawal of polarizing conditions was asso-ciated with an increase in the H3K4me3/H3K27me3 ratiofor Tbx21, but despite this the permissive modificationsin the RORc t promoter were maintained. In contrast, Th1cells cultured for the same duration showed predominanceof repressive marks in the RORc t promoter and a predom-inance of H3K4me3 in Tbx21 as previously described (46).Similar findings were observed for A1(M).RAG1−/− Th17cells activated under various conditions (Figure S3B), andimportantly histone 3 modifications in the promoters ofRORc t and Tbx21 mirrored gene expression (Figure S3C).These results indicate that activation of polarized Th17 cellsin the absence of Th17-inducing cytokines, as may have oc-

curred within skin grafts, should facilitate stable expressionof IL-17A while allowing up-regulation of IFNc .

Th17-induced rejection proceeds independently of

IL-17A and IFNcWe sought to determine whether neutralization of IL-17Aor IFNc could have an effect on Th17-induced rejection. IL-17A is a potent inducer of neutrophil mobilization (1, 2) andFigure 3B clearly indicated participation of neutrophils inthe response to the graft. Additionally, because long-termTh17 lines produced IFNc as an inherent feature (Figures1A and B), the significance of this cytokine for Th17 ef-fector function needed clarification. Exposure to IFNc canup-regulate MHC molecules by nonimmune cells (47,48),and could thus enhance antigen presentation and the prop-agation of Th17-associated tissue damage. A combinationof two antibodies, XMG1.2 (27) and R46-A2 (49), known toneutralize in vivo IFNc activity, and confirmed by ourselvesto block an IFNc -specific ELISA in vitro (data not shown),was used.



Figure 5: Th17-mediated pathology

proceeds independently of IL-17A

or IFNc neutralization. A total of10×106 long-term cultured MarilynTh17 cells were transferred into fe-male C57BL/6.RAG1−/− hosts on day−1. Mice were grafted with maleC57BL/6.RAG1−/− tail skin on day 0and treated either with IgG (n = 8) oranti-IL-17A antibody (n = 10) (3 doses× 0.75 mg, weekly, starting on the dayof grafting), or with IgG1 (n = 4) oranti-IFNc antibodies (n = 8) (8 doses× 1 mg of each R46-A2 and XMG1.2,every other day, starting on the dayof grafting). (A) Relative surface areaon day 21 is presented as mean andSD for male grafts on antibody-treatedrecipients as described above or fe-male grafts (n = 4). Statistical analy-sis was performed using unpaired two-tailed Student’s t-test. (B) H&E and IHCstainings for MPO (in brown) were per-formed on grafts from IgG- or anti-IL-17A-treated mice killed on day 22. Notethat the abundance of MPO+ cells re-mains unchanged despite IL-17A neu-tralization. The arrows indicate the bor-der between host and grafted skin. (C)Grafted mice were killed on day 22 af-ter transplantation, and FACS analysiswas performed on splenocytes restim-ulated with PMA and ionomycin. Ex-pression of IL-17A and IFNc by Vb6+cells is shown. Statistical analysis wasperformed using unpaired two-tailedStudent’s t-test in order to compare rel-evant groups for the same gated pop-ulation (∗∗0.001<p < 0.01, highlightedin red). (D) Expression of IL-22 by Vb6+cells is shown like in (C). Statisticalanalysis was performed using unpairedtwo-tailed Student’s t-test.

Neither IL-17A nor IFNc neutralization interfered withTh17-mediated graft contraction on day 21, which wasa consistent finding in both the Marilyn (Figure 5A) andA1(M).RAG1−/− (Figures S2B and C) systems. Despite anapparent increase in graft size in the control IgG- comparedto anti-IL-17A-treated Marilyn groups (Figure 5A), histolog-ical analysis confirmed the absence of any biological dif-ference between them (Figure 5B and Figure S2D). Neu-trophil infiltration, as judged by IHC staining for MPO, wasnot influenced by IL-17A neutralization (Figure 5B), suggest-ing compensatory activity by other Th17-derived moleculessuch as IL-17F (50).

Blockade of IL-17A activity did not alter the expression ofIL-17A or IFNc by Th17 cells (Figure 5C and Figure S3A),

whereas IFNc neutralization reduced the frequency of IL-17A−IFNc + cells amongst transferred Marilyn Th17 cells(Figure 5C). Antigen exposure was responsible for driv-ing IFNc up-regulation by Th17 cells, as Vb6+ splenocytesrecovered from female graft recipients did not show in-creased expression of IFNc (Figure 5C). Finally, IL-22 ex-pression by transferred cells remained unchanged despiteantibody treatments (Figure 5D).

Discussion

Despite previous reports implicating Th17 cells in trans-plant rejection (51,52), the exact dimensions of such in-volvement remain unclear. To our knowledge, our study is


Agorogiannis et al.

the first to investigate the individual contribution of Th17cells to graft rejection in the absence of other adaptivelymphocytes.

It is still uncertain whether Th17 cells can arise endoge-nously in response to grafted tissues in wild-type recip-ients. The implication of IL-17A and Th17 cells in trans-plant rejection by T-bet-deficient mice (13,21,22), unable tomount efficient Th1 responses, may suggest a redundantrole for Th17 cells in wild-type hosts. Moreover, transferrednaive Marilyn CD4+ T cells did not seem to differentiate intoTh17 cells (Figure 4A), unless regulatory T cells were alsopresent (12). Even if Th17 cells have little or no involvementin graft rejection in unmanipulated animals, it is possiblethey might participate in patients receiving immunosup-pression that inhibits other effector populations, or wheregraft antigens are recognized by preexisting memory Th17cells acquired through heterologous immunity (53).

Th17 lines with sustainable IL-17A expression were gen-erated after long-term culture under polarizing conditions.Cells polarized either with IL-23 or without IL-21 displayedcomparable phenotypes, and variable levels of IFNc ex-pression remained an inherent property of these Th17 lineseven after neutralization of both IL-12 and IL-23 (with anti-IL-12p40 antibody) (data not shown). Marilyn Th17 cellswere found to generally produce higher levels of IFNccompared to A1(M).RAG1−/− both in vitro and after in vivotransfer. The dose of antigen has been shown to influ-ence the level of IFNc expression of Th17 cells (43) andthe higher surface expression of the TCR by Marilyn com-pared to A1(M).RAG1−/− T cells (54) may have a similarconsequence.

Importantly, epigenetic stability in the Rorc locus followingwithdrawal of polarizing conditions indicated that long-termlines can maintain their core Th17 differentiation status andexpression of IL-17A, despite enrichment for permissivemodifications in the T-bet promoter and subsequent IFNcproduction. Overall, while these data might be explainedby plasticity in the Th17 lineage (38–43), an alternative viewwould be that production of IFNc is an inherent effectorproperty of Th17 cells elicited following recognition of anti-gen. In any case, as tissue damage was not amelioratedfollowing IFNc neutralization and as the graft pathologyinduced by Marilyn and A1(M).RAG1−/− Th17 cells was in-distinguishable, it is unlikely that IFNc expression by Th17cells plays an essential part in this model of graft rejection.

Injected Th17 cells were able to enter skin grafts and in-duce rapid pathological changes evident as early as day 7after grafting. Epidermal hyperplasia (acanthosis), loss ofskin appendages and dermal inflammation with neutrophilinfiltration were consistent, suggesting parallels with pso-riasis models (55,56), although Munro’s microabscesseswere absent. A murine model of Th17-induced GVHD wasalso associated with epidermal hyperplasia (57), and long-term Th17 lines grown with additional IL-23 +/− IL-1b, in-

duced identical pathology (unpublished). These changeswere distinct from Th1-induced acute rejection that wascharacterized by epidermal thinning and dermal hemor-rhage.

Th17-induced pathology was not prevented following IL-17A or IFNc neutralization, suggesting redundancy of ef-fector mechanisms. For example, IL-17F is also known topromote neutrophil recruitment (50), and IL-17A-deficientmice show increased neutrophil counts associated withincreased IL-17F and granulocyte colony stimulating fac-tor (CSF3) plasma concentrations (58). It is also possiblethat neither IL-17A nor IL-17F are essential to the rejec-tion process mediated by Th17 cells (59), with granulocyte-macrophage colony stimulating factor (CSF2) (60,61) andIL-22 (62) as other candidates to be excluded. Indeed, IL-22 has been recognized as an important player in tissue-specific inflammation (63, 64), and IL-22 (65), but not IL-17A(66), is involved in IL-23-induced acanthosis, and in a mousemodel of psoriasis IL-22 neutralization prevented diseasemanifestations (63).

In summary, our data indicate that Th17 cells can inducegraft damage in their own right, as is the case for Th1and Th2 cells (20), but with a distinct pathology due to dif-ferent underlying effector mechanisms (67–69). Our studyconfirms IL-17A-independent pathways to Th17-induced in-flammation (59–61) and suggests new experimental op-portunities to resolve mechanisms by which these cellsmediate tissue damage.

Acknowledgments

We are indebted to Mr. Richard Stillion for preparation of histological slides,Dr. Teresa Marafioti for IHC staining of histological sections, Dr. Diane Mar-shall (UCB Celltech) for the kind gift of anti-IL-17A and control IgG antibodiesand members of the Pathology Support Building Staff for excellent care ofexperimental mice. This work was supported by a MRC grant and an EdwardPenley Abraham studentship (to E.I.A.).

Disclosure

The authors of this publication have no conflicts of inter-est to disclose as described by the American Journal ofTransplantation.

References

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

2. Miossec P, Korn T, Kuchroo VK. Interleukin-17 and type 17 helperT cells. N Engl J Med 2009; 361: 888–898.

3. Mitchell P, Afzali B, Lombardi G, Lechler RI. The T helper 17-regulatory T cell axis in transplant rejection and tolerance. CurrOpin Organ Transplant 2009; 14: 326–331.

4. Van Kooten C, Boonstra JG, Paape ME, et al. Interleukin-17 ac-tivates human renal epithelial cells in vitro and is expressed



during renal allograft rejection. J Am Soc Nephrol 1998; 9: 1526–1534.

5. Antonysamy MA, Fanslow WC, Fu F, et al. Evidence for a role ofIL-17 in alloimmunity: A novel IL-17 antagonist promotes heart graftsurvival. Transplant Proc 1999; 31: 93.

6. Antonysamy MA, Fanslow WC, Fu F, et al. Evidence for a roleof IL-17 in organ allograft rejection: IL-17 promotes the functionaldifferentiation of dendritic cell progenitors. J Immunol 1999; 162:577–584.

7. Tang JL, Subbotin VM, Antonysamy MA, Troutt AB, Rao AS, Thom-son AW. Interleukin-17 antagonism inhibits acute but not chronicvascular rejection. Transplantation 2001; 72: 348–350.

8. Hsieh HG, Loong CC, Lui WY, Chen A, Lin CY. IL-17 expressionas a possible predictive parameter for subclinical renal allograftrejection. Transpl Int 2001; 14: 287–298.

9. Loong CC, Hsieh HG, Lui WY, Chen A, Lin CY. Evidence for theearly involvement of interleukin 17 in human and experimentalrenal allograft rejection. J Pathol 2002; 197: 322–332.

10. Strehlau J, Pavlakis M, Lipman M, et al. Quantitative detectionof immune activation transcripts as a diagnostic tool in kidneytransplantation. Proc Natl Acad Sci USA 1997; 94: 695–700.

11. Chen L, Ahmed E, Wang T, et al. TLR signals promote IL-6/IL-17-dependent transplant rejection. J Immunol 2009; 182: 6217–6225.

12. Vokaer B, Van Rompaey N, Lemaitre PH, et al. Critical role of regu-latory T cells in Th17-mediated minor antigen-disparate rejection.J Immunol 2010; 185: 3417–3425.

13. Yuan X, Paez-Cortez J, Schmitt-Knosalla I, et al. A novel role of CD4Th17 cells in mediating cardiac allograft rejection and vasculopathy.J Exp Med 2008; 205: 3133–3144.

14. Burlingham WJ, Love RB, Jankowska-Gan E, et al. IL-17-dependentcellular immunity to collagen type V predisposes to obliterativebronchiolitis in human lung transplants. J Clin Invest 2007; 117:3498–3506.

15. Yoshida S, Haque A, Mizobuchi T, et al. Anti-type V collagen lym-phocytes that express IL-17 and IL-23 induce rejection pathologyin fresh and well-healed lung transplants. Am J Transplant 2006;6: 724–735.

16. Fan L, Benson HL, Vittal R, et al. Neutralizing IL-17 prevents oblit-erative bronchiolitis in murine orthotopic lung transplantation. AmJ Transplant 2011; 11: 911–922.

17. Deteix C, Attuil-Audenis V, Duthey A, et al. Intragraft Th17 infil-trate promotes lymphoid neogenesis and hastens clinical chronicrejection. J Immunol 2010; 184: 5344–5351.

18. Dander E, Balduzzi A, Zappa G, et al. Interleukin-17-producing T-helper cells as new potential player mediating graft-versus-hostdisease in patients undergoing allogeneic stem-cell transplanta-tion. Transplantation 2009; 88: 1261–1272.

19. Le Moine A, Goldman M, Abramowicz D. Multiple pathways toallograft rejection. Transplantation 2002; 73: 1373–1381.

20. Zelenika D, Adams E, Mellor A, et al. Rejection of H-Y disparateskin grafts by monospecific CD4+ Th1 and Th2 cells: No require-ment for CD8+ T cells or B cells. J Immunol 1998; 161: 1868–1874.

21. Burrell BE, Csencsits K, Lu G, Grabauskiene S, Bishop DK. CD8+Th17 mediate costimulation blockade-resistant allograft rejectionin T-bet-deficient mice. J Immunol 2008; 181: 3906–3914.

22. Yuan X, Ansari MJ, D’Addio F, et al. Targeting Tim-1 to overcomeresistance to transplantation tolerance mediated by CD8 T17 cells.Proc Natl Acad Sci USA 2009; 106: 10734–10739.

23. Miura M, El-Sawy T, Fairchild RL. Neutrophils mediate parenchy-mal tissue necrosis and accelerate the rejection of complete ma-jor histocompatibility complex-disparate cardiac allografts in theabsence of interferon-gamma. Am J Pathol 2003; 162: 509–519.

24. Surquin M, Le Moine A, Flamand V, et al. IL-4 deficiency preventseosinophilic rejection and uncovers a role for neutrophils in therejection of MHC class II disparate skin grafts. Transplantation2005; 80: 1485–1492.

25. Lantz O, Grandjean I, Matzinger P, Di Santo JP. Gamma chainrequired for naive CD4+ T cell survival but not for antigen prolifer-ation. Nat Immunol 2000; 1: 54–58.

26. Ohara J, Paul WE. Production of a monoclonal antibody to andmolecular characterization of B-cell stimulatory factor-1. Nature1985; 315: 333–336.

27. Cherwinski HM, Schumacher JH, Brown KD, Mosmann TR. Twotypes of mouse helper T cell clone: III. Further differences inlymphokine synthesis between Th1 and Th2 clones revealed byRNA hybridization, functionally monospecific bioassays, and mon-oclonal antibodies. J Exp Med 1987; 166: 1229–1244.

28. Cobbold SP, Jayasuriya A, Nash A, Prospero TD, Waldmann H.Therapy with monoclonal antibodies by elimination of T-cell sub-sets in vivo. Nature 1984; 312: 548–551.

29. Qin S, Cobbold S, Tighe H, Benjamin R, Waldmann H. CD4 mono-clonal antibody pairs for immunosuppression and tolerance induc-tion. Eur J Immunol 1987; 17: 1159–1165.

30. Scott D, Addey C, Ellis P, et al. Dendritic cells permit identificationof genes encoding MHC class II-restricted epitopes of transplan-tation antigens. Immunity 2000; 12: 711–720.

31. Inaba K, Inaba M, Romani N, et al. Generation of large num-bers of dendritic cells from mouse bone marrow cultures supple-mented with granulocyte/macrophage colony-stimulating factor. JExp Med 1992; 176: 1693–1702.

32. Havell EA, Spitalny GL. Production and characterization of anti-murine interferon-gamma sera. J Interferon Res 1983; 3: 191–198.

33. Cobbold S, Metcalfe S. Monoclonal antibodies that define caninehomologues of human CD antigens: Summary of the First In-ternational Canine Leukocyte Antigen Workshop (CLAW). TissueAntigens 1994; 43: 137–154.

34. Livak KJ, Schmittgen TD. Analysis of relative gene expressiondata using real-time quantitative PCR and the 2(-Delta Delta C(T))Method. Methods 2001; 25: 402–408.

35. Nelson JD, Denisenko O, Bomsztyk K. Protocol for the fast chro-matin immunoprecipitation (ChIP) method. Nat Protoc 2006; 1:179–185.

36. Mukasa R, Balasubramani A, Lee YK, et al. Epigenetic instabilityof cytokine and transcription factor gene loci underlies plasticityof the T helper 17 cell lineage. Immunity 2010; 32: 616–627.

37. Thomas RM, Chen C, Chunder N, et al. Ikaros silences T-bet ex-pression and interferon-gamma production during T helper 2 dif-ferentiation. J Biol Chem 2010; 285: 2545–2553.

38. Bending D, De La Pena H, Veldhoen M, et al. Highly purifiedTh17 cells from BDC2.5NOD mice convert into Th1-like cells inNOD/SCID recipient mice. J Clin Invest 2009; 119: 565–572.

39. Croxford AL, Kurschus FC, Waisman A. Cutting edge: An IL-17F-CreEYFP reporter mouse allows fate mapping of Th17 cells. JImmunol 2009; 182: 1237–1241.

40. Lee YK, Turner H, Maynard CL, et al. Late developmental plasticityin the T helper 17 lineage. Immunity 2009; 30: 92–107.

41. Martin-Orozco N, Chung Y, Chang SH, Wang YH, Dong C. Th17cells promote pancreatic inflammation but only induce diabetesefficiently in lymphopenic hosts after conversion into Th1 cells.Eur J Immunol 2009; 39: 216–224.

42. Nurieva R, Yang XO, Chung Y, Dong C. Cutting edge: In vitro gen-erated Th17 cells maintain their cytokine expression program innormal but not lymphopenic hosts. J Immunol 2009; 182: 2565–2568.


Agorogiannis et al.

43. O’Connor RA, Prendergast CT, Sabatos CA, et al. Cutting edge:Th1 cells facilitate the entry of Th17 cells to the central nervoussystem during experimental autoimmune encephalomyelitis. J Im-munol 2008; 181: 3750–3754.

44. Ivanov, II, McKenzie BS, Zhou L, et al. The orphan nuclear receptorRORgammat directs the differentiation program of proinflamma-tory IL-17+ T helper cells. Cell 2006; 126: 1121–1133.

45. Mathur AN, Chang HC, Zisoulis DG, et al. T-bet is a critical deter-minant in the instability of the IL-17-secreting T-helper phenotype.Blood 2006; 108: 1595–1601.

46. Wei G, Wei L, Zhu J, et al. Global mapping of H3K4me3 andH3K27me3 reveals specificity and plasticity in lineage fate de-termination of differentiating CD4+ T cells. Immunity 2009; 30:155–167.

47. Goes N, Sims T, Urmson J, Vincent D, Ramassar V, Halloran PF.Disturbed MHC regulation in the IFN-gamma knockout mouse.Evidence for three states of MHC expression with distinct rolesfor IFN-gamma. J Immunol 1995; 155: 4559–4566.

48. Skoskiewicz MJ, Colvin RB, Schneeberger EE, Russell PS.Widespread and selective induction of major histocompatibilitycomplex-determined antigens in vivo by gamma interferon. J ExpMed 1985; 162: 1645–1664.

49. Havell EA. Purification and further characterization of an anti-murine interferon-gamma monoclonal neutralizing antibody. J In-terferon Res 1986; 6: 489–497.

50. Oda N, Canelos PB, Essayan DM, Plunkett BA, Myers AC, HuangSK. Interleukin-17F induces pulmonary neutrophilia and amplifiesantigen-induced allergic response. Am J Respir Crit Care Med2005; 171: 12–18.

51. Burrell BE, Bishop DK. Th17 cells and transplant acceptance. Trans-plantation 2010; 90: 945–948.

52. Chadha R, Heidt S, Jones ND, Wood KJ. Th17: Contributors toallograft rejection and a barrier to the induction of transplantationtolerance? Transplantation 2011; 91: 939–945.

53. Adams AB, Pearson TC, Larsen CP. Heterologous immunity: Anoverlooked barrier to tolerance. Immunol Rev 2003; 196: 147–160.

54. Farquhar CA, Paterson AM, Cobbold SP, et al. Tolerogenicity is notan absolute property of a dendritic cell. Eur J Immunol 2010; 40:1728–1737.

55. Gudjonsson JE, Johnston A, Dyson M, Valdimarsson H, Elder JT.Mouse models of psoriasis. J Invest Dermatol 2007; 127: 1292–1308.

56. Nestle FO, Kaplan DH, Barker J. Psoriasis. N Engl J Med 2009;361: 496–509.

57. Carlson MJ, West ML, Coghill JM, Panoskaltsis-Mortari A, BlazarBR, Serody JS. In vitro-differentiated TH17 cells mediate lethalacute graft-versus-host disease with severe cutaneous and pul-monary pathologic manifestations. Blood 2009; 113: 1365–1374.

58. von Vietinghoff S, Ley K. IL-17A controls IL-17F production andmaintains blood neutrophil counts in mice. J Immunol 2009; 183:865–873.

59. Haak S, Croxford AL, Kreymborg K, et al. IL-17A and IL-17F do notcontribute vitally to autoimmune neuro-inflammation in mice. JClin Invest 2009; 119: 61–69.

60. Codarri L, Gyulveszi G, Tosevski V, et al. RORgammat drives pro-duction of the cytokine GM-CSF in helper T cells, which is essen-tial for the effector phase of autoimmune neuroinflammation. NatImmunol 2011; 12: 560–567.

61. El-Behi M, Ciric B, Dai H, et al. The encephalitogenicity of T(H)17cells is dependent on IL-1- and IL-23-induced production of thecytokine GM-CSF. Nat Immunol 2011; 12: 568–575.

62. Liang SC, Tan XY, Luxenberg DP, et al. Interleukin (IL)-22 andIL-17 are coexpressed by Th17 cells and cooperatively enhance

expression of antimicrobial peptides. J Exp Med 2006; 203: 2271–2279.

63. Ma HL, Liang S, Li J, et al. IL-22 is required for Th17 cell-mediatedpathology in a mouse model of psoriasis-like skin inflammation. JClin Invest 2008; 118: 597–607.

64. Sonnenberg GF, Nair MG, Kirn TJ, Zaph C, Fouser LA, Artis D.Pathological versus protective functions of IL-22 in airway inflam-mation are regulated by IL-17A. J Exp Med 2010; 207: 1293–1305.

65. Zheng Y, Danilenko DM, Valdez P, et al. Interleukin-22, a T(H)17cytokine, mediates IL-23-induced dermal inflammation and acan-thosis. Nature 2007; 445: 648–651.

66. Chan JR, Blumenschein W, Murphy E, et al. IL-23 stimulates epi-dermal hyperplasia via TNF and IL-20R2-dependent mechanismswith implications for psoriasis pathogenesis. J Exp Med 2006;203: 2577–2587.

67. Jager A, Dardalhon V, Sobel RA, Bettelli E, Kuchroo VK. Th1,Th17, and Th9 effector cells induce experimental autoimmune en-cephalomyelitis with different pathological phenotypes. J Immunol2009; 183: 7169–7177.

68. Kroenke MA, Carlson TJ, Andjelkovic AV, Segal BM. IL-12- and IL-23-modulated T cells induce distinct types of EAE based on histol-ogy, CNS chemokine profile, and response to cytokine inhibition.J Exp Med 2008; 205: 1535–1541.

69. Stummvoll GH, DiPaolo RJ, Huter EN, et al. Th1, Th2, and Th17effector T cell-induced autoimmune gastritis differs in pathologicalpattern and in susceptibility to suppression by regulatory T cells.J Immunol 2008; 181: 1908–1916.

Supporting Information

Additional supporting information may be found in the on-line version of this article.

This section contains four supplementary figures (S1–S4).

Figure S1: Reduced expansion of Th17 cells in female

lymphopenic recipients of male skin grafts compared

to Th1 and naive CD4+ T cells. Female mice injected withMarilyn CD4+ (n = 4), Th1 (n = 3) or Th17 (n = 7) cells,and grafted with male skin as described in Figure 2, werekilled on day 48 after transplantation, and FACS analysiswas performed on splenocytes. Total numbers of Vb6+

cells within recovered spleens are shown.

Figure S2: A1(M).RAG1−/− Th17 cells enter skin grafts

and induce transplant pathology independently of IL-

17A, similar to Marilyn Th17 cells. A total of 10×106long-term cultured A1(M).RAG1−/− Th17 cells were trans-ferred into female CBA/Ca.RAG1−/− hosts on day −1.Other mice were injected only with PBS. Mice weregrafted with male CBA/Ca.RAG1−/− tail skin on day 0. Th17-injected mice were left untreated (n = 3), or injected withIgG (n = 5) or anti-IL-17A antibody (n = 5) (3 doses × 0.75mg, weekly, starting on the day of grafting). (A) Real-timeRT-PCR analysis for the expression of CD3Y, IL-17A, IFNcand MPO in male skin grafts recovered on day 7. Dataare normalized to expression of HPRT1 and presented asmeans and SD of n = 4 (PBS) or 3 (Th17) mice/group.Grafts on PBS-injected mice were used as calibrator.



Statistical analysis was performed using unpaired two-tailed Student’s t-test. (B) Representative photographs (notto scale) of male skin grafts on day 21. White-dashed rect-angles are indicative of the position of the graft only, anddo not represent the measurements described in (C). (C)Relative surface area is presented at the indicated timepoints as mean ± SD for male grafts on PBS- or Th17-injected hosts. Statistical analysis was performed usingunpaired two-tailed Student’s t-test in order to comparemale grafts on IgG- and anti-IL-17A-injected mice. (D) H&Estaining was performed on grafts from mice killed on day22. Normal appearance of epidermis and dermis in graftson PBS-injected hosts contrasts with epidermal hyperpla-sia, dermal inflammatory cell infiltration and loss of skinappendages observed in grafts from recipients of Th17cells, regardless of IL-17A neutralization. The arrow indi-cates the border between host and grafted skin, whereasthe arrowheads show fibrin aggregates interspersed withneutrophils.

Figure S3: A1(M).RAG1−/−Th17 cells maintain IL-17A

expression without upregulation of IFNc , in recipients

of male skin grafts. (A) Female CBA/Ca.RAG1−/− miceinjected with A1(M).RAG1−/−Th17 cells were grafted withmale skin and either left untreated (n = 3) or treated withIgG (n = 5) or anti-IL-17A antibody (n = 5), as described inFigure S2. Mice were killed on day 22 after transplantation,and FACS analysis was performed on splenocytes restim-ulated with PMA and ionomycin. Expression of IL-17A andIFNc by Vb8+ cells is shown. (B) ChIP analysis of long-term cultured A1(M).RAG1−/− Th17 cells stimulated for 7

days in the presence of different conditions. Real-time PCRwas performed using DNA from samples immunoprecip-itated with antibodies against H3K4me3 and H3K27me3.Data are normalized to 10% input DNA and presented asratios of H3K4me3 to H3K27me3 density at the promotersof the RORc t (Rorc+5.0) and T-bet (Tbx21) genes. Meansand SD of n = 3 technical replicates from one experiment(representative of two) are shown. (C) Cells from (B) wereanalyzed for expression of Rorc and Tbx21 mRNA. Dataare normalized to expression of HPRT1 and presented asmeans and SD of n = 3 biological replicates for each dif-ferent condition, except for cells activated in the presenceof Th17-polarizing conditions (n = 4). CBA/Ca splenocyteswere used as calibrator.

Figure S4: Th17 cells upregulate FoxP3 expression in

female lymphopenic recipients of male skin grafts. (A)Expression of FoxP3 by Marilyn and A1(M).RAG1−/− Th17cells before in vivo transfer. (B) Female mice injected withMarilyn CD4+ (n = 4), Th1 (n = 3) or Th17 (n = 7) cells,and grafted with male skin as described in Figure 2, werekilled on day 48 after transplantation, and FACS analysiswas performed on splenocytes. Expression of FoxP3 byCD4+ cells is shown.

Please note: Wiley-Blackwell is not responsible for the con-tent or functionality of any supporting materials suppliedby the authors. Any queries (other than missing material)should be directed to the corresponding author for thearticle.


th17 cells induce a distinct graft rejection response that ... · of effector mechanisms able to...

Documents