Plasticity of Th17 Cells in Autoimmune Kidney Diseases Christian F

Plasticity of Th17 Cells in Autoimmune Kidney Diseases Christian F. Krebs, Jan-Eric Turner, Hans-Joachim Paust, Sonja Kapffer, Tobias Koyro, Sonja Krohn, Friederike Ufer, This information is current as Manuel A. Friese, Richard A. Flavell, Brigitta Stockinger, of October 2, 2021. Oliver M. Steinmetz, Rolf A. K. Stahl, Samuel Huber and Ulf Panzer J Immunol 2016; 197:449-457; Prepublished online 6 June 2016; doi: 10.4049/jimmunol.1501831 Downloaded from http://www.jimmunol.org/content/197/2/449

Supplementary http://www.jimmunol.org/content/suppl/2016/06/04/jimmunol.150183 http://www.jimmunol.org/ Material 1.DCSupplemental References This article cites 35 articles, 14 of which you can access for free at: http://www.jimmunol.org/content/197/2/449.full#ref-list-1

Why The JI? Submit online.

• Rapid Reviews! 30 days* from submission to initial decision by guest on October 2, 2021

• No Triage! Every submission reviewed by practicing scientists

• Fast Publication! 4 weeks from acceptance to publication

*average

Subscription Information about subscribing to The Journal of Immunology is online at: http://jimmunol.org/subscription Permissions Submit copyright permission requests at: http://www.aai.org/About/Publications/JI/copyright.html Email Alerts Receive free email-alerts when new articles cite this article. Sign up at: http://jimmunol.org/alerts

The Journal of Immunology is published twice each month by The American Association of Immunologists, Inc., 1451 Rockville Pike, Suite 650, Rockville, MD 20852 Copyright © 2016 by The American Association of Immunologists, Inc. All rights reserved. Print ISSN: 0022-1767 Online ISSN: 1550-6606. The Journal of Immunology

Plasticity of Th17 Cells in Autoimmune Kidney Diseases

Christian F. Krebs,* Jan-Eric Turner,* Hans-Joachim Paust,* Sonja Kapffer,* Tobias Koyro,* Sonja Krohn,* Friederike Ufer,† Manuel A. Friese,† Richard A. Flavell,‡ Brigitta Stockinger,x Oliver M. Steinmetz,* Rolf A. K. Stahl,* Samuel Huber,{ and Ulf Panzer*

The ability of CD4+ T cells to differentiate into pathogenic Th1 and Th17 or protective T regulatory cells plays a pivotal role in the pathogenesis of autoimmune diseases. Recent data suggest that CD4+ T cell subsets display a considerable plasticity. This plasticity seems to be a critical factor for their pathogenicity, but also for the potential transition of pathogenic effector T cells toward a more tolerogenic phenotype. The aim of the current study was to analyze the plasticity of Th17 cells in a mouse model of acute crescentic glomerulonephritis and in a mouse chronic model of lupus nephritis. By transferring in vitro generated, highly purified Th17 cells and by using IL-17A fate reporter mice, we demonstrate that Th17 cells fail to acquire substantial expression of the Th1 and Th2 signature cytokines IFN-g and IL-13, respectively, or the T regulatory transcription factor Foxp3 throughout the course Downloaded from of renal inflammation. In an attempt to therapeutically break the stability of the Th17 phenotype in acute glomerulonephritis, we subjected nephritic mice to CD3-specific Ab treatment. Indeed, this treatment induced an immunoregulatory phenotype in Th17 cells, which was marked by high expression of IL-10 and attenuated renal tissue damage in acute glomerulonephritis. In summary, we show that Th17 cells display a minimum of plasticity in acute and chronic experimental glomerulonephritis and introduce anti- CD3 treatment as a tool to induce a regulatory phenotype in Th17 cells in the kidney that may be therapeutically exploited. The Journal of Immunology, 2016, 197: 449–457. http://www.jimmunol.org/

he CD4+ T cells play a central role in the pathogenesis glomerulonephritis (i.e., nephrotoxic nephritis [NTN]) (8–10) and of autoimmune diseases, including human and experi- a model of pristane-induced lupus nephritis (11, 12). Time course T mental crescentic glomerulonephritis. Classically, it was analyses of the renal CD4+ T cell response in the NTN model assumed that the cell-mediated immune reactions responsible for revealed that Th17 cell numbers peak at ∼days 6–10 after ne- tissue injury are predominantly mediated by IFN-g–producing phritis induction and then decline, whereas the numbers of Th1 Th1 cells (1, 2). However, the identification and functional char- and T regulatory (Treg) cells continuously increase from days 6 to acterization of Th17 cells have challenged this paradigm (3, 4). 10 onward (13–16). This sequential infiltration of the different by guest on October 2, 2021 Indeed, numerous studies have clearly shown that Th17 cells drive CD4+ T cell subsets into inflamed target organs with an early autoimmune processes that were previously thought to be exclu- increase of Th17 cells that is followed by Th1 and Treg cell in- sively Th1 mediated, such as multiple sclerosis (5), rheumatoid filtration is a phenomenon that has also been reported in other arthritis (6), and psoriasis (7). inflammatory diseases (17, 18). Whether the gradual decline of We and others recently demonstrated that the IL-17A/Th17 axis Th17 cells observed during progression of renal autoimmunity is contributes to renal tissue injury in both experimental crescentic due to transdifferentiation into cells of another Th cell subtype, a result of apoptosis, or a consequence of emigration of Th17 cells is currently unclear. + *III. Medizinische Klinik, Universita¨tsklinikum Hamburg-Eppendorf, 20246 Ham- In this context, recent studies indicating that CD4 Tcell † burg, Germany; Institute of Neuroimmunology and Multiple Sclerosis, Universita¨ts- subsets in other organs demonstrate a higher grade of plastic- klinikum Hamburg-Eppendorf, 20246 Hamburg, Germany; ‡Department of Immunobiology, School of Medicine, Yale University, New Haven, CT 06520; ity than previously appreciated are of great interest (18–21). In xDivision of Molecular Immunology, Medical Research Council National Institute { particular, Th17 and Treg cells might be highly flexible, which for Medical Research, London NW7 1AA, United Kingdom; and I. Medizinische would allow these cells to switch to a different Th cell program Klinik, Universita¨tsklinikum Hamburg-Eppendorf, 20246 Hamburg, Germany and thereby adapt to various inflammatory settings during an ORCIDs: 0000-0002-8412-1840 (T.K.); 0000-0001-8781-336X (B.S.); 0000-0002- 6225-8600 (R.A.K.S.). immune response. However, whether the potential plasticity of Received for publication August 24, 2015. Accepted for publication May 4, 2016. Th17 cells in autoimmunity is a general finding, and whether this process is beneficial or detrimental remains to be fully This work was supported by Deutsche Forschungsgemeinschaft Grants SFB1192/A5 (to C.F.K. and S.H.), SFB1192/A1 (to U.P.), and KFO 228: PA 754/7-2 (to U.P. and elucidated. C.F.K.); Emmy-Noether-Programme Grants TU316/2-1 (to J.-E.T.) and HU1714/3-1 The aim of the current study was to investigate the potential (to S.H.); and grants from the Deutsche Gesellschaft fur€ Nephrologie (to C.F.K. and J.-E.T.). T.K. received a stipend from the Werner Otto Foundation. plasticity of Th17 cells in murine models of glomerulonephritis. We therefore did the following: 1) assessed the fate of adoptively Address correspondence and reprint requests to Dr. Christian F. Krebs, Universita¨tsklinikum Hamburg-Eppendorf, III. Medizinische Klinik, Martinistraße 52, 20246 transferred Th17 cells in nephritic Rag1-knockout mice; 2) studied Hamburg, Germany. E-mail address: [email protected] the plasticity of in vivo differentiated Th17 cells in two models of The online version of this article contains supplemental material. glomerulonephritis (NTN and pristane-induced lupus nephritis) by Cre YFP Abbreviations used in this article: EAE, experimental autoimmune encephalomyeli- using IL-17A fate reporter mice (Il17a 3 R26r ); and 3) tis; GN, glomerulonephritis; NTN, nephrotoxic nephritis; PAS, periodic acid–Schiff; modulated Th17 cell plasticity by CD3-specific Ab treatment to Tr1, type 1 regulatory; Treg, T regulatory. induce a regulatory phenotype of Th17 cells in crescentic glo- Copyright Ó 2016 by The American Association of Immunologists, Inc. 0022-1767/16/$30.00 merulonephritis (GN). www.jimmunol.org/cgi/doi/10.4049/jimmunol.1501831 450 Th17 CELL PLASTICITY IN GN

Materials and Methods Ag retrieval was performed by incubation with proteinase type XXIV (5 mg/ml; Animals Sigma-Aldrich) for 15 min at 37˚C. Tissue sections were developed with the Vectastain ABC-AP kit (Vector Laboratories, Burlingame, CA). To All mice in this study were on C57BL/6J background. Il-17aGFP/Ifn-gFP635/ quantify tubulointerstitial GR1+ cells, at least 20 low power fields (original Foxp3mRFP and Il-10GFP reporter mice as well as IL-17ACre 3 R26rYFP magnification 3200) were counted. mice were used, as described before (18, 22–24). Rag12/2 mice were All slides were evaluated using an Axioskop light microscope (Zeiss) and purchased from The Jackson Laboratory (Bar Harbor, ME). All mice were photographed with an Axiocam HRc (Zeiss) using the ZEN lite 2011 raised under specific pathogen-free conditions in our animal facility at the software (Zeiss). University Medical Center Hamburg-Eppendorf. Animal experiments were performed according to national animal care and ethical guidelines, and Isolation of leukocytes from kidney, spleen, and CNS were approved by local ethics committees. Previously described methods for leukocyte isolation from murine kidneys Animal models and functional studies were used (9). In brief, kidneys were digested for 45 min at 37˚C with 0.4 mg/ml collagenase D (Roche, Mannheim, Germany) and 0.01 mg/ml NTN was induced in 8- to 10-wk-old male mice by i.p. injection of 0.5 ml DNase I in RPMI 1640 medium (Life Technologies, Darmstadt, Ger- nephrotoxic sheep serum per mouse, as previously described (10). In many) supplemented with 10% heat-inactivated FCS (Invitrogen, Carlsbad, adoptive cell transfer experiments, cells were administered i.v. 1 d prior to CA), and finely minced using GentleMACS digester (Miltenyi Biotec, induction of glomerulonephritis i.v. Lupus nephritis was induced by i.p. Bergisch-Gladbach, Germany). Subsequently, 37% Percoll gradient cen- injection of 0.5 ml pristane (Sigma-Aldrich, St. Louis, MO) (11). trifugation was performed (GE Healthcare, Uppsala, Sweden). Urinary albumin excretion was determined by standard ELISA anal- Spleens were minced, sequentially filtered through 70-mm and 40-mm ysis (Mice-Albumin Kit; Bethyl). Urinary creatinine was measured using nylon meshes, and then washed with HBSS without Ca2+ and Mg2+ standard laboratory methods. Anti-CD3 experiments were performed, as (Invitrogen). Cell viability was assessed by trypan blue staining before previously described (25). In brief, 10 mg of either hamster anti-mouse flow cytometry or cell transfer experiments.

CD3 (clone 145-2C11) or isotype control Ab was injected i.p. (both To isolate leukocytes from CNS, mice were perfused via the vascular Downloaded from BioLegend, San Diego, CA) at the time points indicated. system with PBS by intracardial injection, and thereafter, brains and spinal Experimental autoimmune encephalomyelitis (EAE) was induced by cords were extracted. Brains and spinal cords were combined and digested immunization of female mice (12–14 wk) s.c. with 200 mg myelin oli- for 45 min at 37˚C with 0.4 mg/ml collagenase D (Roche) and 0.01 mg/ml godendrocyte glycoprotein peptide fragment 35–55 (NeoMPS, San Diego, DNase I in RPMI 1640 medium (Life Technologies) supplemented with CA) in CFA (BD Biosciences, Franklin Lakes, NJ) containing 4 mg/ml 10% heat-inactivated FCS (Invitrogen), minced using GentleMACS di- Mycobacterium tuberculosis (H37Ra; BD Biosciences), as described pre- gester (Miltenyi Biotec), and finely filtered through a 70-mm cell strainer viously (26). We injected 200 ng pertussis toxin (Merck Millipore, (BD Biosciences). Subsequently, 37% Percoll gradient centrifugation was Darmstadt, Germany) i.v. on the day of immunization and 48 h later. Mice performed (GE Healthcare). http://www.jimmunol.org/ were scored for clinical signs by the following system: 0, no clinical deficits; 1, tail weakness; 2, hind limb paresis; 3, partial hind limb paral- Flow cytometry ysis; 3.5, full hind limb paralysis; 4, full hind limb paralysis and forelimb For FACS analysis, the following Abs were used: CD45 (30-F11), CD3 paresis; and 5, premorbid or dead (26). Mice were analyzed at day 14 after (17A2), CD4 (GK1.5), CD8 (53-6.7), CD11b (M1/70), IL-17A (TC11- EAE induction. 18H10), IL-13 (eBio13A), and IFN-g (XMG1.2) (BD Biosciences, eBio- Splenocyte cultures and T cell polarization science, or BioLegend) (14). Dead-cell staining was performed using the LIVE/DEAD Fixable Read Dead Stain Kit (Invitrogen). Analyses were For splenocyte cultures, a total of 4 3 106 cells/ml was incubated for 60 h performed on a BD Biosciences LSRII system with Diva software, and in the presence of sheep IgG. Supernatants were collected, and IL-17A and data were analyzed using the Flowjo software (Tree Star). For transfer IFN-g were measured by ELISA (BioLegend) or by cytometric bead array experiments, cells were sorted with an Aria II system (BD Biosciences). by guest on October 2, 2021 (BD Biosciences) (27), according to the manufacturer’s instructions. For intracellular cytokine staining, cells were activated by incubation at + For polarization of naive CD4 cells, splenocytes were isolated from 37˚C, 5% CO2, for 4 h with PMA (50 ng/ml; Sigma-Aldrich) and ion- naive IL-17aGFP/IFN-g-FP635 reporter mice using the MACS CD4+ T Cell omycin (1 mg/ml; Calbiochem-Merck, Darmstadt, Germany) in X-VIVO Isolation Kit II (Miltenyi Biotec, Bergisch-Gladbach, Germany), according medium (Lonza AG, Walkersville, MD). After 30 min of incubation, to the manufacturer’s protocol. A total of 5 3 105 cells/well was plated in brefeldin A (10 mg/ml; Sigma-Aldrich) was added. anti-CD3 Ab-coated 96-well plates (2 mg/ml; eBioscience, San Diego, CA) and incubated for 65 h with soluble anti-CD28 at 5 mg/ml. Th17 polari- Statistical analysis zation was achieved in the presence of anti–IL-4 (10 mg/ml), anti–IFN-g Data were expressed as mean 6 SEM. All statistical analyses were per- (12 mg/ml), IL-6 (50 ng/ml), TGF-b (1 ng/ml), IL-1b (20 ng/ml), and IL- formed with the SPSS package and Graph Pad Prism 5 software. A p value 23 (5 ng/ml) (all from BioLegend). ,0.05 was considered to be statistically significant. The Student t test was Real-time RT-PCR analysis used for comparison between two groups. In case of three or more groups, one-way ANOVA was used, followed by a post hoc analysis with Bon- Total RNA of the renal cortex was prepared according to standard labo- ferroni’s test for multiple comparisons. ratory methods. Real-time RT-PCR was performed on a StepOnePlus Real- Time PCR system (Applied Biosystems, Foster City, CA), as previously described (10). Results Stability of Th17 cells after transfer into nephritic Morphological analysis Rag12/2 mice Glomerular crescent formation was assessed in a blinded fashion in 30 NTN is a well-characterized, T cell–mediated mouse model of glomeruli per mouse in periodic acid–Schiff (PAS)–stained paraffin sec- crescentic glomerulonephritis, which is induced by the injection of tions (28). For the assessment of tubulointerstitial injury, low original magnification field photographs (3200) of four nonoverlapping cortical sheep antiserum raised against cortical kidney components (10, areas from PAS-stained kidney sections were taken per mouse. The in- 29). We first determined the pathogenic potential of Th17 cells in terstitial area was then determined by superimposing the photographs with transfer experiments. To this end, we compared the effect of a a grid containing 40 points with subsequent counting of points located in transfer of a bulk population of in vitro generated Th17 cells the interstitial area (excluding glomeruli, blood vessels, and tubules) in a (purity of 30%, Supplemental Fig. 1) with the transfer of Th0 cells blinded fashion. The percentage of points in the interstitial area corre- sponds to the tubulointerstitial injury score (13, 28). The glomerular tuft on the course of experimental glomerulonephritis in Rag1-defi- area was measured in glomeruli from five nonoverlapping cortical areas cient mice. The transfer of Th17 cells resulted in aggravated renal per mouse from PAS-stained kidney sections using ZEN lite 2011 software damage, including glomerular crescent formation and tubu- (Zeiss, Jena, Germany). The average glomerular tuft area per individual lointerstitial injury (Supplemental Fig. 1A–C). In addition, renal mouse was analyzed. For immunohistochemical staining, paraffin-embedded sections were neutrophil infiltration was increased after transfer of Th17 cells deparaffinized, rehydrated, and stained with the following Abs: GR-1 (1:50, (Supplemental Fig. 1D). Therefore, our results demonstrated the Ly6 G/C, NIMP-R14; Hycult Biotech) and Foxp3 (FJK-16s; eBioscience). increased pathogenicity of in vitro polarized Th17 cells in this The Journal of Immunology 451 model. Of note, immunohistochemistry showed no Foxp3+ cells IL-13–expressing YFP2 cells at later time points (Fig. 3). Taken in the kidney of mice after transfer of cells, indicating that Th0 together, this indicates that Th17 cells have a rather stable phe- and Th17 cells did not transdifferentiate into Foxp3+ Treg cells notype in acute experimental glomerulonephritis and that ex- (Supplemental Fig. 1D). Interestingly, the reanalysis by flow pression of IFN-g and IL-13 by renal CD4+ T cells is restricted to cytometry revealed that only a minority of the transferred CD4+ non-Th17 cells. T cells remained IL-17A positive (∼7%) in the inflamed kidney, Importantly, we also induced EAE in Il17aCre fate reporter mice whereas the majority produced IFN-g (∼64%) (Supplemental Fig. (Il17aCre 3 R26RYFP), a model in which Hirota et al. (18) ob- 1E). However, technical limitations might restrict the significance served a high degree of Th17 cell plasticity. EAE was induced by of these results, because the transferred cell population included injection of myelin oligodendrocyte glycoprotein peptide frag- significant amounts of IL-17A–negative cells, which might be ment 35–55. Mice showed signs of disease with increasing EAE responsible for the emergence of IFN-g–producing CD4+ T cells score starting at day 8 (Supplemental Fig. 2A). CD45+CD3+CD4+ in nephritic Rag12/2 mice. CD11bneg T cells from CNS (Supplemental Fig. 2B) and from the To determine the potential plasticity of Th17 cells in more detail, draining inguinal lymph node (Supplemental Fig. 2C) were ana- we isolated CD4+CD62L+ naive T cells from IL-17aGFP mice that lyzed at day 14 for IL-17A–positive cells. In agreement with the express GFP under the control of the IL-17A promotor and po- study of Hirota et al. (18) and in contrast to the situation in au- larized them into Th17 cells in vitro (23). Th17-differentiated cells toimmune kidney disease, intracerebral Th17 cells and Th17 cells were FACS sorted based on GFP expression to obtain a purity of from the draining lymph node showed indeed a high degree of .95% and were then transferred i.v. into Rag1-deficient recipi- plasticity to Th1 cells (Supplemental Fig. 2). ents. Mice that had received Th17 cells showed signs of renal damage, as determined by glomerular crescent formation, tubu- Stability of Th17 cells in pristane-induced lupus nephritis Downloaded from lointerstitial injury (Fig. 1A, 1B), and proteinuria (Fig. 1C) at day We next examined Th17 cell stability in a chronic model of lupus 10 after induction of glomerulonephritis. In line with a predomi- nephritis (pristane-induced lupus nephritis) using Il17aCre fate nant Th17 response, staining of renal tissue sections for the neu- reporter mice. This model depends on the Th17 immune response, trophil marker GR-1 revealed massive neutrophil recruitment into as shown before (11, 30). Six months after induction of the dis- the kidney in nephritic mice (Fig. 1D). ease, PAS staining of renal tissue sections showed glomerular To evaluate whether the transferred Th17 cells retain their cy- hypercellularity and glomerular hypertrophy, as a sign of immune- http://www.jimmunol.org/ tokine pattern, we used a fluorescence reporter system (Il-17aGFP/ mediated kidney damage resembling certain aspects of human Ifn-gFP635/Foxp3RFP) (24) to analyze cytokine expression of donor lupus nephritis (Fig. 4A, 4B). In line with a role of Th17 cells in CD4+ T cells after restimulation with PMA and ionomycin. The this model of renal autoimmunity, flow cytometry revealed in- analysis of CD4+ T cells before the transfer and after recovery creased infiltration of Th17 cells into the kidney at the 6-mo time from the inflamed kidney at day 10 showed that a major propor- point (Fig. 4C). Most importantly, staining for intracellular cyto- tion of Th17 cells maintained IL-17A expression (∼40%), whereas kines showed IL-17A expression in ∼80% of renal CD4+YFP+ only a minority initiated IFN-g production (∼20%) (Fig. 1E). (IL-17Afm+) cells, but only in ,2% of YFP2 (IL-17Afm2) cells Another major proportion of renal CD4+ T cells became negative (Fig. 4D). In contrast, a high percentage of IL-17Afm2 cells by guest on October 2, 2021 for IL-17A but also did not upregulate IFN-g (∼40%) (Fig. 1E). expressed IFN-g, but not IL-17A. IL-13 expression was found Immunohistochemical staining for Foxp3 (Fig. 1D) and fluores- only in a small percentage of CD4+ T cells (Fig. 4E). Taken to- cence analysis of the Foxp3-reporter (mRFP) (data not shown) gether, these analyses underscore the stability of Th17 cells in revealed the absence of Treg cells in the kidney of Rag-1–deficient renal autoimmunity even in a chronic model that develops grad- mice after transfer of Th17 cells. ually over months. Stability of Th17 cells in acute crescentic GN CD3-specific Ab treatment induces an immunoregulatory Despite the transfer of a highly purified population of Th17 cells, phenotype of Th17 cells marked by IL-10 expression and we could not exclude that proliferation of contaminating non-Th17 ameliorates GN cells in a lymphopenic environment might account for the con- After having shown that Th17 cells are pathogenic in experimental siderable proportion of IL-17A–negative cells observed in the glomerulonephritis and have a stable phenotype in renal autoim- experiments delineated above. Therefore, we aimed at studying mune disease, we aimed at interfering with Th17 cell stability to the fate of Th17 cells under more physiological conditions in induce a tolerogenic or regulatory phenotype in Th17 cells. We immune-competent nephritic mice. To this end, we induced NTN have previously shown that anti-CD3–specific Ab treatment in- in Il17aCre fate reporter mice (Il17aCre 3 R26RYFP) (18) and induces a regulatory phenotype in Th17 cells in the intestine in vivo, vestigated parameters of tissue damage and cytokine expression of which is characterized by high IL-10 production (25). We there- CD4+ T cells in the kidney at different time points. Glomerular fore aimed to test whether the phenotype of Th17 cells could be crescents increased over time to .35% at day 20 (Fig. 2A, 2B). therapeutically manipulated in the kidney. To that end, we induced The percentage of Th17 fate-mapped (YFP+) cells of all CD4+ NTN and treated animals with anti-CD3 Abs (2 3 10 mg i.p.) in T cells peaked at day 10. In the course of NTN, YFP+ cells were the course of the disease. To analyze the impact of anti-CD3 the major population of CD4+ T cells expressing IL-17A treatment on IL-10 expression in IL-17A+ cells, we induced (Fig. 2C). IL-17A expression by some YFP2 cells (non-Th17) NTN in IL-10 reporter mice (IL-10GFP) (15, 22). Mice were (Fig. 2C) might be a result of new Th17 cells coming up that treated with anti-CD3 at day 8, and 4 h before analysis at day 10. have not had time to switch on enough Cre. Renal CD4+ T cells were analyzed for IL-17A, IFN-g, and IL-13 To analyze the cytokine production of Th17 cells, we compared expression after stimulation by intracellular Ab staining, and for YFP+ (Fig. 3A, upper panel) with YFP2 cells (lower panel). YFP+ IL-10 using the GFP reporter system. Flow cytometry showed IL- cells maintained IL-17A expression, but only a small subset of 10 expression in 5% of renal IL-17A–expressing T cells from these cells expressed IFN-g, and there was virtually no expression isotype-treated mice. After anti-CD3 treatment, .30% of renal of the Th2 cytokine IL-13 in this population. YFP2 cells IL-17A+ cells were positive for IL-10 (Fig. 5A, 5B). Analysis of expressed high amounts of IFN-g, and there was also a subset of IL-17A–negative cells also revealed significant induction of IL-10 452 Th17 CELL PLASTICITY IN GN Downloaded from

FIGURE 1. Stability of Th17 cells in nephritic Rag12/2 mice. (A) PAS staining of renal cortex from Rag12/2 mice at day 10 after transfer of 2 3 105 (n =4)or43 105 (n = 4) in vitro polarized Th17 cells after NTN induction. Original magnification 3400. (B) Crescent formation and http://www.jimmunol.org/ tubulointerstitial injury in the different groups. (C) Urinary albumin-to-creatinine ratio at day 8 after NTN induction. (D) Immunohistochemical staining and quantification of neutrophils (top) and Treg cells (bottom) in renal cortex by staining of the granulocyte differentiation Ag (GR1)-positive cells and Foxp3-positive cells, respectively. Original magnification 3400. (E) Cytokine expression of FACS-sorted Th17 cells 2/2 directly before transfer into Rag1 mice, and by guest on October 2, 2021 analysis of renal T cells at day 10 after NTN induction. The diagram shows the quantification of cytokine expression after transfer as quantified by flow cytometry. Bars represent mean 6 SEM (*p , 0.05).

in this population. However, the expression of IFN-g and IL-13 treatment. Therefore, fate reporter mice were treated with was not inﬂuenced by anti-CD3 treatment (Fig. 5A, 5B). anti-CD3 at days 8 and 10 after induction of experimental glo- Next, we used IL-17ACre fate reporter mice to investigate merulonephritis. Still, Th17 cells (YFP+) maintained IL-17A ex- whether Th17 cells maintain IL-17A expression after anti-CD3 pression after anti-CD3 treatment (Fig. 5C, 5D). IFN-g and IL-13 The Journal of Immunology 453 Downloaded from

FIGURE 2. Time kinetics of YFP and IL-17A expression in acute crescentic GN. (A) PAS staining of renal cortex from Il-17aCre 3 R26RYFP fate reporter mice at indicated time points after NTN induction. Original magnification 3400. (B) Crescent formation at indicated time points (n = 3–6 per group) after http://www.jimmunol.org/ NTN induction. (C) Flow cytometry of renal CD3+CD4+ T cells for IL-17A fate (IL-17Afm/YFP) and IL-17A expression (anti–IL-17A mAb [mab]). (D) Quantification of renal YFP+ cells in relation to all renal CD4+ T cells. Bars represent mean 6 SEM. were not induced in these cells by anti-CD3. In addition, anti-CD3 the mechanisms driving plasticity or stability, as well as their did not reduce the number of renal CD4+ T cells in the kidney relevance in autoimmunity, remain to be fully elucidated. significantly (Fig. 5H). The pathogenic impact of Th1 and Th17 cells in human and In addition, we analyzed the systemic effect of anti-CD3 treat- experimental GN has recently been demonstrated (8, 9, 16, 32). ment in mice with crescentic glomerulonephritis. We therefore Interestingly, during the course of murine experimental GN, we by guest on October 2, 2021 measured in vitro cytokine production of sheep IgG-stimulated and others observed a shift from a Th17-dominated inflammation splenocytes from mice with anti-CD3 or isotype using cytometric to a Th1-dominated immune response in the kidney (13, 33). The bead array analyses. Interestingly, IL-10 was significantly released rapid decline in Th17 cell numbers observed between days 10 and into the supernatants of splenocytes from anti-CD3–treated mice 20 in the NTN model of GN raises the question what the fate of (Supplemental Fig. 3A), whereas Th1, Th2, and Th17 cytokines renal Th17 cells might be. We therefore transferred in vitro po- were not altered (Supplemental Fig. 3B, 3C). larized Th17 cells into nephritic Rag12/2 mice.Inlinewithre- To test whether anti-CD3 treatment induces Foxp3+ Treg cells cent studies, the transferred Th17 cells induced severe renal in vivo, we also analyzed Foxp3 expression in nephritic IL-17ACre injury in terms of glomerular crescent formation and tubu- fate reporter mice (Fig. 5E, 5F). Our data revealed that anti-CD3 lointerstitial damage (10, 12, 34). Reanalysis of renal CD4+ treatment did not induce Foxp3 expression in Th17 cells (YFP+). T cells after transfer revealed a very poor stability of the Th17 To evaluate the impact of anti-CD3 treatment on renal tissue phenotype and a shift to IFN-g production. However, technical injury, PAS-stained kidney sections were analyzed at day 10 in limitations restrict the significance of these results because a bulk mice after treatment with anti-CD3 (days 6 and 8). Interestingly, in Th17 population that included significant amounts of IL-17A– mice treated with anti-CD3, renal tissue damage was reduced as negative cells was transferred. To analyze whether the emergence compared with isotype controls (Fig. 5I, 5J), indicating that anti- of IFN-g–producing T cells is caused by cellular or population CD3 treatment has beneficial effects in experimental glomerulo- plasticity, we used acute reporter mice that express GFP under the nephritis. control of the IL-17A promotor (23) to obtain highly purified Th17 cells (.95%). After transfer of these cells into Rag12/2 Discussion mice, the majority of Th17 cells continued to produce IL-17A, CD4+ T cell responses can be subdivided into proinflammatory whereas only a minor population produced IFN-g, indicating and anti-inflammatory reactions. Although Th1, Th2, and Th17 minimal single cell plasticity of in vitro polarized Th17 cells in cells are well-characterized subsets that provide protection against GN. It is noteworthy that immunohistochemistry revealed no exogenous harmful agents, but also drive autoimmune diseases, Foxp3-positive cells in the kidney after transfer of polarized Treg cells avoid autoimmune reactions and overwhelming effector Th17 cells, excluding significant transdifferentiation of Th17 cells responses (31). These CD4+ T cell subsets have classically been into Treg cells. This observation nicely extends our earlier data viewed as terminally differentiated lineages with limited flexibil- showing that renal and systemic Treg cells develop from neither ity. Recent data suggest that Th17 cells show a high degree of transferred Foxp3-negative T effector cells (35) nor RORgt+ Th17 plasticity in the brain and intestine, in particular upon inflamma- cells (36). tory stimuli, where Th17 cells can produce IFN-g and trans- Cells transferred into lymphopenic Rag12/2 mice encounter a differentiate into highly pathogenic Th1 cells (18, 21). However, rather nonphysiological environment. We therefore aimed at 454 Th17 CELL PLASTICITY IN GN Downloaded from http://www.jimmunol.org/

FIGURE 3. Stability of Th17 cells in acute crescentic GN. (A) Gating for renal IL-17A fate-mapped (IL-17Afm+/YFP+) and IL-17Afm2 (YFP2) cells by flow cytometry. FACS analyses for IL-17A, IFN-g, and IL-13 expression at days 0 and 10 after NTN expression. (B) Quantification of cytokine expression of IL-17Afm+ (red) and IL-17Afm2 cells (blue) in the course of NTN (days 0–20, n = 3–6 per group). investigating the plasticity of Th17 cells in immune-competent pression of IL-17A and IFN-g, the induction of T-bet in Th17 mice. To this end, we induced experimental glomerulonephritis cells, and the subsequent shift from Th17 to Th1 cells (18), might (NTN)infatereportermice(Il-17aCre 3 R26rYFP)inwhichthe be of great interest in this context. by guest on October 2, 2021 activation of the IL-17A promotor results in the expression of After having shown that Th17 cells have a rather stable phe- Cre recombinase. Subsequently, Cre-mediated excision of a notype in two models of renal autoimmune diseases, we sought to floxed stop codon upstream of the fluorescent protein YFP under therapeutically interfere with the stability of these cells. To induce control of a ubiquitously active promoter leads to permanent transdifferentiation of renal Th17 cells into anti-inflammatory IL- labeling of IL-17A–producing cells, regardless of their actual 10–producing T cells, we induced NTN in IL-10 reporter mice production of this cytokine at the time of analysis (18). Similar (IL-10GFP) (22) and treated them with an anti-CD3 Ab, a treat- to the data obtained with intracellular cytokine staining after ment strategy that we previously used to induce tolerogenic Th17 restimulation (13), time course analysis revealed that the per- cells mainly in the intestine (25). The mechanisms of CD3- centage of Th17 fate-mapped cells (YFP+)ofallCD4+ Tcells specific Ab-induced tolerance have been extensively studied be- peaked at day 10. To analyze the cytokine production of Th17 fore, but the exact mechanisms are still incompletely understood cells, we compared YFP+ Th17 with non-Th17 cells. The main and a matter of ongoing debate. Potential mechanisms include finding of this study is that Th17 cells maintained IL-17A ex- expansion of Foxp3+ Treg (37) and type 1 regulatory (Tr1) cells pression in experimental glomerulonephritis, and only a small (25), induction of a regulatory phenotype in Th17 cells (23), and subset of these cells expressed IFN-g. In contrast, non-Th17 transdifferentiation of Th17 cells into Tr1 cells (24). Our results cells expressed high amounts of IFN-g, and a minor IFN-g– demonstrate that IL-10 can be induced in renal IL-17A+ Tcells negative subset also expressed IL-13. Results obtained with the by anti-CD3 treatment in a model of crescentic glomerulone- IL-17A fate reporter in the chronic model of pristane-induced phritis. Importantly, Th17 cells maintained IL-17A expression lupus nephritis revealed almost identical results, demonstrating and there was no induction of IFN-g in Th17 fate-mapped cells. that Th17 cells have a stable phenotype in autoimmune kidney These data show that Th17 cells in mice treated with anti-CD3 diseases. Abs may acquire a regulatory phenotype. The potential thera- In two recent landmark studies, a shift of Th17 to Th1 cells in the peutic relevance of this intervention was further substantiated by course of experimental autoimmune encephalomyelitis and ex- the observation of reduced renal injury in nephritic mice after perimental colitis was shown (18, 21), whereas acute cutaneous anti-CD3 treatment. However, we do not yet know whether these infection with Candida albicans did not result in the deviation of IL-10–producing Th17 cells represent a stable Treg cell subset. In Th17 cells to the production of alternative cytokines (18). This addition, it is unclear whether a further transdifferentiation into suggests that the specific inflammatory environment in the target Tr1 T cells can be induced. organ can directly influence the plasticity of Th17 cells, but more In conclusion, using adoptive transfer experiments and studies are needed for a better understanding of the underlying studies in fate reporter mice, we demonstrated that renal Th17 mechanisms. The role of IL-23 stimulation of Th17 cells in dif- cells have very limited spontaneous plasticity in acute and ferent organs, which seems to be required for simultaneous ex- chronic models of glomerulonephritis. However, using CD3- The Journal of Immunology 455 Downloaded from

FIGURE 4. Stability of Th17 cells in pristane-induced lupus nephritis. (A) PAS staining of renal cortex from Il-17aCre 3 R26RYFP fate reporter mice 6 mo after induction of lupus nephritis by exposure to pristane (n = 11), and from age- http://www.jimmunol.org/ matched controls (n = 10). Original magnification 3400. (B) Quantification of glomerular hypertrophy, as measured by glomerular tuft area. (C) Intracellular cytokine staining for IL- 17A and analysis by flow cytometry of renal CD4+ T cells of mice with lupus nephritis and healthy control mice. The diagram shows the relative quantification of renal IL-17A fate mapped (IL-17Afm+/YFP+)(*p , 0.05). (D) Analysis of renal CD4+ IL-17Afm+ (YFP+) and IL-17Afm2 (YFP2) cells for expression of IL-17A, IFN-g, and IL-13 by flow cytometry. (E) + Quantification of cytokine expression of IL-17Afm (red) and by guest on October 2, 2021 IL-17Afm2 cells (blue) (*p , 0.05, **p , 0.01).

specific Ab treatment, we were able to promote an IL-10– vide the basis for interventions that modulate effector Th17 producing regulatory phenotype of Th17 cells in the kidney. A cells and that push immune responses from proinflammatory to better understanding of renal Th17 cell flexibility might pro- beneficial anti-inflammatory responses. 456 Th17 CELL PLASTICITY IN GN Downloaded from http://www.jimmunol.org/ by guest on October 2, 2021

FIGURE 5. CD3-specific Ab treatment induced an immunoregulatory phenotype of Th17 cells marked by IL-10 expression and ameliorated GN. (A and B) Cytokine expression of renal T cells in IL-10 acute reporter mice (Tiger) at day 10 after induction of NTN. Anti-CD3 (10 mg/mouse, n =4)or isotype control Ab (n = 5) was applied i.p. 2 d and 4 h prior to the analysis. CD4+IL-17+ and CD4+IL-172 T cells were analyzed for IL-10, IFN-g, and IL- 13 expression, as indicated. (C and D) Nephritic Il-17aCre 3 R26RYFP fate reporter mice were treated with anti-CD3 (n = 4) or isotype control Ab (n =4). Cytokine expression of renal CD4+ IL-17Afm+ (YFP+) and IL-17Afm2 (YFP2) cells, as measured by flow cytometry and intracellular staining for IL- 17A, IFN-g, and IL-13 by mAb at day 10 after induction of glomerulonephritis. (D) The percentage of IL-17A+, IL-13+,andIFN-g+ of IL-17Afm+ (upper panel) and IL-17Afm2 (lower panel) was quantified. (E and F)CD4+ T cells from fate reporter mice treated with anti-CD3 or isotype control Abs were analyzed for ROR-gt, T-bet, and Foxp3 expression. (G) The percentage of renal Th17 cells (IL-17Afm+) and (H) the number of CD4+ cells/kidney were quantified from the kidney of mice treated with anti-CD3 or isotype control Abs. (I) PAS staining of renal cortex from wild-type mice 10 d after NTN induction. Original magnification 3400. NTN mice were treated on days 6 and 8 with i.p. injection of anti-CD3 mAb (n = 6) or isotype control (n =8).(J) Renal damage as measured by quantification of glomerular crescents and tubulointerstitial injury. Bars represent mean 6 SEM (*p , 0.05, **p , 0.01, ***p , 0.001). The Journal of Immunology 457

Disclosures 19. Murphy, K. M., and B. Stockinger. 2010. Effector T cell plasticity: flexibility in the face of changing circumstances. Nat. Immunol. 11: 674–680. The authors have no financial conflicts of interest. 20. Coomes, S. M., V. S. Pelly, and M. S. Wilson. 2013. Plasticity within the ab⁺CD4⁺ T-cell lineage: when, how and what for? Open Biol. 3: 120157. Available at: http://rsob.royalsocietypublishing.org/content/3/1/120157, References 21. Morrison, P. J., D. Bending, L. A. Fouser, J. F. Wright, B. Stockinger, A. Cooke, Helicobacter hepaticus 1. Traugott, U., and P. Lebon. 1988. Multiple sclerosis: involvement of interferons and M. C. Kullberg. 2013. Th17-cell plasticity in -in- in lesion pathogenesis. Ann. Neurol. 24: 243–251. duced intestinal inflammation. Mucosal Immunol. 6: 1143–1156. 2. Kitching, A. R., S. R. Holdsworth, and P. G. Tipping. 1999. IFN-gamma me- 22. Kamanaka, M., S. T. Kim, Y. Y. Wan, F. S. Sutterwala, M. Lara-Tejero, ´ diates crescent formation and cell-mediated immune injury in murine glomer- J. E. Galan, E. Harhaj, and R. A. Flavell. 2006. Expression of interleukin-10 in ulonephritis. J. Am. Soc. Nephrol. 10: 752–759. intestinal lymphocytes detected by an interleukin-10 reporter knockin tiger 3. Cua, D. J., J. Sherlock, Y. Chen, C. A. Murphy, B. Joyce, B. Seymour, L. Lucian, mouse. Immunity 25: 941–952. W. To, S. Kwan, T. Churakova, et al. 2003. Interleukin-23 rather than 23. Esplugues, E., S. Huber, N. Gagliani, A. E. Hauser, T. Town, Y. Y. Wan, interleukin-12 is the critical cytokine for autoimmune inflammation of the brain. W. O’Connor, Jr., A. Rongvaux, N. Van Rooijen, A. M. Haberman, et al. 2011. Nature 421: 744–748. Control of TH17 cells occurs in the small intestine. Nature 475: 514–518. 4. Harrington, L. E., R. D. Hatton, P. R. Mangan, H. Turner, T. L. Murphy, 24. Gagliani, N., M. C. Amezcua Vesely, A. Iseppon, L. Brockmann, H. Xu, K. M. Murphy, and C. T. Weaver. 2005. Interleukin 17-producing CD4+ effector N. W. Palm, M. R. de Zoete, P. Licona-Limoń, R. S. Paiva, T. Ching, et al. 2015. T cells develop via a lineage distinct from the T helper type 1 and 2 lineages. Th17 cells transdifferentiate into regulatory T cells during resolution of in- Nat. Immunol. 6: 1123–1132. flammation. Nature 523: 221–225. 5. Langrish, C. L., Y. Chen, W. M. Blumenschein, J. Mattson, B. Basham, 25. Huber, S., N. Gagliani, E. Esplugues, W. O’Connor, Jr., F. J. Huber, A. Chaudhry, J. D. Sedgwick, T. McClanahan, R. A. Kastelein, and D. J. Cua. 2005. IL-23 M. Kamanaka, Y. Kobayashi, C. J. Booth, A. Y. Rudensky, et al. 2011. Th17 cells ⁻ drives a pathogenic T cell population that induces autoimmune inflammation. J. express interleukin-10 receptor and are controlled by Foxp3 and Foxp3+ reg- Exp. Med. 201: 233–240. ulatory CD4+ T cells in an interleukin-10-dependent manner. Immunity 34: 554– 6. Nakae, S., A. Nambu, K. Sudo, and Y. Iwakura. 2003. Suppression of immune 565. induction of collagen-induced arthritis in IL-17-deficient mice. J. Immunol. 171: 26. Schattling, B., K. Steinbach, E. Thies, M. Kruse, A. Menigoz, F. Ufer, V. Flockerzi, W. Bruck,€ O. Pongs, R. Vennekens, et al. 2012. TRPM4 cation 6173–6177. Downloaded from 7. van der Fits, L., S. Mourits, J. S. Voerman, M. Kant, L. Boon, J. D. Laman, channel mediates axonal and neuronal degeneration in experimental autoimmune F. Cornelissen, A. M. Mus, E. Florencia, E. P. Prens, and E. Lubberts. 2009. encephalomyelitis and multiple sclerosis. Nat. Med. 18: 1805–1811. Imiquimod-induced psoriasis-like skin inflammation in mice is mediated via the 27. Riedel, J. H., H. J. Paust, S. Krohn, J. E. Turner, M. A. Kluger, O. M. Steinmetz, IL-23/IL-17 axis. J. Immunol. 182: 5836–5845. C. F. Krebs, R. A. Stahl, and U. Panzer. 2016. IL-17F promotes tissue injury 8. Ooi, J. D., R. K. Phoon, S. R. Holdsworth, and A. R. Kitching. 2009. IL-23, not in autoimmune kidney diseases. J. Am. Soc. Nephrol. DOI: 10.1681/ IL-12, directs autoimmunity to the Goodpasture antigen. J. Am. Soc. Nephrol. ASN.2015101077. 20: 980–989. 28. Riedel, J. H., H. J. Paust, J. E. Turner, A. P. Tittel, C. Krebs, E. Disteldorf, € 9. Paust, H. J., J. E. Turner, O. M. Steinmetz, A. Peters, F. Heymann, C. Ho¨lscher, C. Wegscheid, G. Tiegs, J. Velden, H. W. Mittrucker, et al. 2012. Immature renal G. Wolf, C. Kurts, H. W. Mittrucker,€ R. A. Stahl, and U. Panzer. 2009. The IL- dendritic cells recruit regulatory CXCR6(+) invariant natural killer T cells to http://www.jimmunol.org/ 23/Th17 axis contributes to renal injury in experimental glomerulonephritis. J. attenuate crescentic GN. J. Am. Soc. Nephrol. 23: 1987–2000. Am. Soc. Nephrol. 20: 969–979. 29. Panzer, U., O. M. Steinmetz, H. J. Paust, C. Meyer-Schwesinger, A. Peters, 10. Krebs, C. F., S. Kapffer, H. J. Paust, T. Schmidt, S. B. Bennstein, A. Peters, J. E. Turner, G. Zahner, F. Heymann, C. Kurts, H. Hopfer, et al. 2007. Che- G. Stege, S. R. Brix, C. Meyer-Schwesinger, R. U. Muller,€ et al. 2013. mokine receptor CXCR3 mediates T cell recruitment and tissue injury in MicroRNA-155 drives TH17 immune response and tissue injury in experimental nephrotoxic nephritis in mice. J. Am. Soc. Nephrol. 18: 2071–2084. crescentic GN. J. Am. Soc. Nephrol. 24: 1955–1965. 30. Summers, S. A., A. Hoi, O. M. Steinmetz, K. M. O’Sullivan, J. D. Ooi, 11. Summers, S. A., D. Odobasic, M. B. Khouri, O. M. Steinmetz, Y. Yang, D. Odobasic, S. Akira, A. R. Kitching, and S. R. Holdsworth. 2010. TLR9 and S. R. Holdsworth, and A. R. Kitching. 2014. Endogenous interleukin (IL)-17A TLR4 are required for the development of autoimmunity and lupus nephritis in promotes pristane-induced systemic autoimmunity and lupus nephritis induced pristane nephropathy. J. Autoimmun. 35: 291–298. by pristane. Clin. Exp. Immunol. 176: 341–350. 31. Sakaguchi, S., T. Yamaguchi, T. Nomura, and M. Ono. 2008. Regulatory T cells 12. Summers, S. A., O. M. Steinmetz, M. Li, J. Y. Kausman, T. Semple, and immune tolerance. Cell 133: 775–787. K. L. Edgtton, D. B. Borza, H. Braley, S. R. Holdsworth, and A. R. Kitching. 32.Kitching,A.R.,A.L.Turner,G.R.Wilson,T.Semple,D.Odobasic, by guest on October 2, 2021 2009. Th1 and Th17 cells induce proliferative glomerulonephritis. J. Am. Soc. J. R. Timoshanko, K. M. O’Sullivan, P. G. Tipping, K. Takeda, S. Akira, and Nephrol. 20: 2518–2524. S. R. Holdsworth. 2005. IL-12p40 and IL-18 in crescentic glomerulonephritis: 13. Paust, H. J., J. E. Turner, J. H. Riedel, E. Disteldorf, A. Peters, T. Schmidt, IL-12p40 is the key Th1-defining cytokine chain, whereas IL-18 promotes C. Krebs, J. Velden, H. W. Mittrucker,€ O. M. Steinmetz, et al. 2012. Chemokines local inflammation and leukocyte recruitment. J.Am.Soc.Nephrol.16: 2023– play a critical role in the cross-regulation of Th1 and Th17 immune responses in 2033. murine crescentic glomerulonephritis. Kidney Int. 82: 72–83. 33. Odobasic, D., P. Y. Gan, S. A. Summers, T. J. Semple, R. C. Muljadi, Y. Iwakura, 14. Turner, J. E., C. Krebs, A. P. Tittel, H. J. Paust, C. Meyer-Schwesinger, A. R. Kitching, and S. R. Holdsworth. 2011. Interleukin-17A promotes early but S. B. Bennstein, O. M. Steinmetz, I. Prinz, T. Magnus, T. Korn, et al. 2012. IL- attenuates established disease in crescentic glomerulonephritis in mice. Am. J. 17A production by renal gd T cells promotes kidney injury in crescentic GN. J. Pathol. 179: 1188–1198. Am. Soc. Nephrol. 23: 1486–1495. 34. Tulone, C., A. Giorgini, S. Freeley, A. Coughlan, and M. G. Robson. 2011. 15. Ostmann, A., H. J. Paust, U. Panzer, C. Wegscheid, S. Kapffer, S. Huber, Transferred antigen-specific T(H)17 but not T(H)1 cells induce crescentic glo- R. A. Flavell, A. Erhardt, and G. Tiegs. 2013. Regulatory T cell-derived IL-10 merulonephritis in mice. Am. J. Pathol. 179: 2683–2690. ameliorates crescentic GN. J. Am. Soc. Nephrol. 24: 930–942. 35. Steinmetz, O. M., S. A. Summers, P. Y. Gan, T. Semple, S. R. Holdsworth, and 16. Paust, H. J., A. Ostmann, A. Erhardt, J. E. Turner, J. Velden, H. W. Mittrucker,€ A. R. Kitching. 2011. The Th17-defining transcription factor RORg t promotes T. Sparwasser, U. Panzer, and G. Tiegs. 2011. Regulatory T cells control the Th1 glomerulonephritis. J. Am. Soc. Nephrol. 22: 472–483. immune response in murine crescentic glomerulonephritis. Kidney Int. 80: 154–164. 36. Kluger, M. A., M. C. Meyer, A. Nosko, B. Goerke, M. Luig, C. Wegscheid, G. 17. Reboldi, A., C. Coisne, D. Baumjohann, F. Benvenuto, D. Bottinelli, S. Lira, Tiegs, R. A. Stahl, U. Panzer, and O. M. Steinmetz. 2016. RORgammat+Foxp3+ A. Uccelli, A. Lanzavecchia, B. Engelhardt, and F. Sallusto. 2009. C-C che- cells are an independent bifunctional regulatory T cell lineage and mediate mokine receptor 6-regulated entry of TH-17 cells into the CNS through the crescentic GN. J. Am. Soc. Nephrol. 27: 454-465. choroid plexus is required for the initiation of EAE. Nat. Immunol. 10: 514–523. 37.Perruche,S.,P.Zhang,Y.Liu,P.Saas,J.A.Bluestone,andW.Chen.2008. 18. Hirota, K., J. H. Duarte, M. Veldhoen, E. Hornsby, Y. Li, D. J. Cua, H. Ahlfors, CD3-specific antibody-induced immune tolerance involves transforming C. Wilhelm, M. Tolaini, U. Menzel, et al. 2011. Fate mapping of IL-17- growth factor-beta from phagocytes digesting apoptotic T cells. Nat. Med. 14: producing T cells in inflammatory responses. Nat. Immunol. 12: 255–263. 528–535.