Immune Suppression via Glucocorticoid-Stimulated Monocytes: A Novel Mechanism To Cope with Inflammation This information is current as of September 29, 2021. Georg Varga, Jan Ehrchen, Anne Brockhausen, Toni Weinhage, Nadine Nippe, Michael Belz, Athanasios Tsianakas, Matthias Ross, Dominik Bettenworth, Tilmann Spieker, Marc Wolf, Ralph Lippe, Klaus Tenbrock, Pieter J. M. Leenen, Johannes Roth and Cord Sunderkötter Downloaded from J Immunol 2014; 193:1090-1099; Prepublished online 2 July 2014; doi: 10.4049/jimmunol.1300891 http://www.jimmunol.org/content/193/3/1090 http://www.jimmunol.org/

Supplementary http://www.jimmunol.org/content/suppl/2014/07/02/jimmunol.130089 Material 1.DCSupplemental References This article cites 51 articles, 17 of which you can access for free at: http://www.jimmunol.org/content/193/3/1090.full#ref-list-1 by guest on September 29, 2021

Why The JI? Submit online.

• Rapid Reviews! 30 days* from submission to initial decision

• 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 © 2014 by The American Association of Immunologists, Inc. All rights reserved. Print ISSN: 0022-1767 Online ISSN: 1550-6606. The Journal of Immunology

Immune Suppression via Glucocorticoid-Stimulated Monocytes: A Novel Mechanism To Cope with Inflammation

Georg Varga,*,†,‡,1 Jan Ehrchen,*,†,1 Anne Brockhausen,*,† Toni Weinhage,‡ Nadine Nippe,*,† Michael Belz,*,† Athanasios Tsianakas,† Matthias Ross,x Dominik Bettenworth,x Tilmann Spieker,{ Marc Wolf,* Ralph Lippe,* Klaus Tenbrock,‖ Pieter J. M. Leenen,# Johannes Roth,* and Cord Sunderko¨tter*,†

Glucocorticoids (GCs) are used as first-line therapies for generalized suppression of inflammation (e.g., allergies or autoimmune diseases), but their long-term use is limited by severe side effects. Our previous work revealed that GCs induced a stable anti- inflammatory phenotype in monocytes, the GC-stimulated monocytes (GCsMs) that we exploited for targeted GC-mediated

therapeutic effects. We demonstrate that GCsMs interact with T cells in suppressing proliferation, as well as cytokine release Downloaded from of CD8+ and, especially, CD4+ T cells in vitro, and that they support generation of Foxp3+ cells. Therefore, we tested their immunosuppressive potential in CD4+ T cell–induced colitis in vivo. We found that injection of GCsMs into mice with severe colitis abolished the inflammation and resulted in significant clinical improvement within a few days. T cells recovered from GCsM-treated mice exhibited reduced secretion of proinflammatory cytokines IFN-g and IL-17. Furthermore, clusters of Foxp3+ CD4+ T cells were detectable at local sites of inflammation in the colon. Thus, GCsMs are able to modify T cell responses in vitro and

in vivo, as well as to downregulate and clinically cure severe T cell–mediated colitis. The Journal of Immunology, 2014, 193: 1090–1099. http://www.jimmunol.org/

lucocorticoids (GCs) are the most widely used agents for flammatory cytokines as the result of an interaction between GC suppression of both acute and chronic inflammation, but receptor monomers and proinflammatory transcription factors, such G their wide spectrum of adverse effects limits long-term as NF-kB or AP-1, in a process called “transrepression” (4, 5). GCs treatment. Their clinical efficacy in treating inflammation has also induce anti-inflammatory mediators in monocytes/macrophages been ascribed mainly to their direct inhibitory effects on activated (e.g., annexin A1) (6, 7). immune cells, such as T cells (1) or monocytes/macrophages (2, 3). Moreover, we demonstrated previously that GCs not only down- In monocytes and macrophages they inhibit production of proin- regulate proinflammatory cytokines, they also generate a functionally

active, anti-inflammatory phenotype in human monocytes that sup- by guest on September 29, 2021 presses inflammatory processes and, thus, induces resolution of in- *Institute of Immunology, University of Muenster, 48149 Muenster, Germany; flammation (8–10). †Department of Dermatology, University of Muenster, 48149 Muenster, Germany; ‡Pediatric Rheumatology and Immunology, University of Muenster, 48149 Muenster, Monocytes represent a central part of innate immunity. After Germany; xDepartment of Medicine B, University of Muenster, 48149 Muenster, differentiation from stem cells in the bone marrow, monocytes enter Germany; {Institute of Pathology, University of Muenster, 48149 Muenster, Germany; ‖ the circulation and are present in the blood until they migrate into Department of Pediatrics, University of Aachen, Rheinisch Westfa¨lische Technische Hochschule, 52062 Aachen, Germany; and #Department of Immunology, Erasmus tissues where they differentiate into macrophages or dendritic cells Medical Center, 3000 Rotterdam, the Netherlands (DCs). These cells give rise to subtypes that are crucial for nearly 1G.V. and J.E. contributed equally to this work. every step of an immune reaction, including the initiation of an Received for publication April 3, 2013. Accepted for publication May 16, 2014. adaptive immune response, clearance of infectious agents, and res- This work was supported by grants from the Deutsche Forschungsgemeinschaft olution of inflammation (11–16). Subpopulations have been defined (DFG SU 195/3-1 to C.S. and G.V.; DFG SU 195/3-2 to C.S., G.V., and A.B; at the level of circulating monocytes in the human (14) and in the DFG DA 1161/4-1 to T.W.; and DFG EH 397/1-1 to J.E. and C.S.), as well as by murine system (17–19). Reflected by their expression level of IMF VA 22 07 07, IZKF RO 20120, and IZKF Eh2/019/11 (to C.S. and J.E.) (Uni- versity of Muenster). N.N. was supported by SkinStaph Grant FKZ: 01kI07100 from CX3CR1 and CCR2 (as well as Ly6C in mice), monocytes either Bundesministerium fur€ Forschung und Bildung. represent a subtype that selectively migrates into inflamed tissue G.V. and J.E. designed research, performed research, and wrote the manuscript; A.B., M.W., (17, 18, 20) or that forms tissue macrophages and DCs under N.N.,M.B.,A.T.,M.R.,T.S.,R.L.,D.B.,andT.W.performedresearch;K.T.andP.J.M.L. steady-state conditions (reviewed in Refs. 12, 14, 19). designed research; and J.R. and C.S. designed research and wrote the manuscript. In analogy to our studies on human monocytes, we previously The sequences presented in this article have been submitted to the Gene Expression Omnibus (http://www.ncbi.nlm.nih.gov/geo/query/acc.cgi?acc=GSE54778) under ac- defined a GC-stimulated murine monocyte subset generated cession number GSE54778. from bone marrow–derived monocytes (21). These murine GC- Address correspondence and reprint requests to Dr. Cord Sunderko¨tter, Department stimulated monocytes (GCsMs) exhibit a stable and distinct phe- of Dermatology, University of Muenster, Von-Esmarch-Straße 58, D-48149 Muen- notype and share with their human counterparts properties such as ster, Germany. E-mail address: [email protected] production of anti-inflammatory mediators and increased migra- The online version of this article contains supplemental material. tory potential (8, 21). In the murine system, GCsMs are charac- Abbreviations used in this article: Ctr-Mo, control monocyte; DC, dendritic cell; GC, low 2 med/high + glucocorticoid; GCsM, glucocorticoid-stimulated monocyte; GITR, glucocorticoid- terized phenotypically as CX3CR1 CCR2 Ly6C CD80 + + induced TNFR-related protein; GO, Gene Ontology; LPMC, lamina propria mono- CD124 CD163 cells (21). This cell surface signature and in- nuclear cell; MDSC, myeloid-derived suppressor cell; MLN, mesenteric lymph node; creased expression of the IL-4Ra-chain (CD124) suggest simi- Treg, regulatory T cell. larity of GCsM to so-called “myeloid-derived suppressor cells” Copyright Ó 2014 by The American Association of Immunologists, Inc. 0022-1767/14/$16.00 (MDSCs) that were described to suppress T cells and adaptive www.jimmunol.org/cgi/doi/10.4049/jimmunol.1300891 The Journal of Immunology 1091 immune responses (22, 23). Therefore, we hypothesized that analyzed by GCOS Software (Affymetrix), using data from corresponding GCsMs also contribute to regulation of adaptive immune responses, control samples as baseline, and further evaluated statistically, as described especially those mediated by effector T cells. previously (8). They were submitted to the Gene Expression Omnibus under accession number GSE54778. In this study, we approached this hypothesis by genome-wide We retained only genes that were significantly regulated in every ex- expression screening of murine GCsMs and ensuing functional periment (change p value , 0.05; fold change .2, expression over clustering. The data indicated that GC treatment significantly background), as well as in the complete set of experiments (fold-change . , affected the capacity of monocytes to interact with T cells. In agree- 2.0, p value 0.05, paired t test). ment with this, we identified GCsMs as potent suppressors of both Functional clustering CD8+ and CD4+ effector T cell activation in vitro and in vivo. Most To analyze microarray data in the context of biological functions, we used importantly, treatment of mice suffering from T cell–mediated information available from the Gene Ontology (GO) consortium (http:// transfer colitis with GCsMs (i.v.) resulted in a complete and fast www.geneontology.org) (8). The GO terms represent a defined vocabulary cure of disease. describing the biological process, cellular components, and molecular functions of genes in a hierarchical directed acyclic graph structure. Sta- tistical analysis was performed using GO-Elite software (8). Materials and Methods Mice Isolation of T cells from spleen and CFSE labeling for T cell proliferation C57BL/6 and Rag2/2 mice were kept under specific pathogen–free con- ditions, and experiments were performed according to approved protocols T cells were isolated from spleens, as described by Ahlmann et al. (26), and + (#87-51.042010A113) of the animal welfare committee Landesamt fuer CD4 T cells were purified using MACS technology, according to the Natur Umwelt und Verbraucherschutz (Recklinghausen, Germany). Mice manufacturer’s instructions. Cells were labeled with CFSE (0.5 mM) and Downloaded from were used for experiments at the age of 10–12 wk. Specific pathogen–free used in coculture assays with monocytes (T cells/monocytes = 10:1) or DC conditions were according to the recommendation of the Federation of (T cells/DCs = 50:1). CFSE staining was routinely evaluated 2 h post- European Laboratory Science Association. staining by FACS to ensure efficiency of labeling. T cell proliferation was assessed as CFSE dilution in FACS. Abs and reagents Repetitive stimulation of naive T cells Anti-CD3e (145-2C11), anti-CD4 allophycocyanin (RM4-5), anti-CD8 allo- phycocyanin (53-6.7), anti-CD11b (M1/70), anti-CD28 (37.51), anti-CD80 Naive T cells were isolated as described above and cocultured with either http://www.jimmunol.org/ 6 + 5 (1G10/B7), anti-CD86 (GL1), anti-CD121b (4E2), anti-CD124 (mouse IL- Ctr-Mos or GCsM. A total of 1 3 10 CD4 T cells/ml or 1 3 10 monocytes/ 4R–M1), and IFN-g, IL-4, IL-13, and IL-17 Flex Sets were from BD Bio- ml was used. After 5 d, cells were harvested and reapplied at the same ratio to sciences (Heidelberg, Germany). Anti-Foxp3–FITC (FJK 16s) was from fresh monocytes for three cycles. Subsequently, cells were stained intracellu- eBioscience (Frankfurt, Germany). Anti-Ly6C (HK1.4), anti–CTLA-4 PE larly for Foxp3 expression and on the surface for CD4 (27). Cells were (UC10-4B9), anti-CD274 (MIH5), and anti-GITR (CD357, YGITR765) were measured by FACS and analyzed using Flow Jo 8.7 software. from BioLegend. Cell culture medium generally was RPMI 1640 contain- Suppression assay ing10%FCS,2mML-glutamine, 1 mM sodium pyruvate, 13 nonessential amino acids, 100 mg/ml Pen Strep, and 10 mM HEPES. To induce their proliferation, T cells were stimulated with anti-CD3 and anti- T cell isolation kits, anti-CD25-, anti-CD19-, anti-CD11c-, and anti- CD28 Abs (5 mg/ml each) coated to 96-well round-bottom plates. A total of CD90 magnetic beads were from Miltenyi Biotec (Bergisch-Gladbach, 5 1 3 10 splenic T cells was applied to each well. To test the suppressive by guest on September 29, 2021 Germany). CFSE and DAPI were from Invitrogen (Karlsruhe, Ger- activity of monocytes on stimulated T cells, 1 3 104 monocytes were added many). Medium and supplemental substances were from Biochrom (triplicates for each condition) and incubated for 5 d. For controls, T cells (Berlin, Germany). Dexamethasone was from Sigma (Taufkirchen, were cultured either alone (positive control) or without stimulation by anti- Germany). CD3/anti-CD28 Abs (negative control). After 4–6 d in culture, 100 mlcell Generation of GCsMs and DCs from bone marrow supernatants was stored until cytokine analysis was performed. Cells were harvested and stained for CD4 and CD8 and analyzed for CFSE dilution by DCs from bone marrow cells were generated essentially as described earlier FACS. (24, 25). For monocyte enrichment, fresh bone marrow cells were applied to a density gradient (Ficoll) and subsequently deprived of CD11c+,CD19+, Transfer colitis + and CD90 cells using MACS technology. Monocytes were then cultured for + 27 Syngeneic CD4 T cells were prepared from spleen of C57BL/6 mice, and 48 h with 10 M (40 ng/ml) dexamethasone in medium supplemented with CD25+ cells were removed using MACS technology. A total of 1 3 106 CD4+ M-CSF (50 ng/ml) using Teflon bags; they are referred to as GCsMs. Control CD252 T cells was adoptively transferred i.v. into Rag2/2 mice (on a C57BL/6 monocytes (Ctr-Mos) were cultured identically with M-CSF (50 ng/ml) but background), which do not develop functional T cells (28, 29). The weight of with only the solvent (PBS; Ctr-Mo) instead of GC. Finally, cells were animals was monitored frequently (at least every 2 d) until they lost body washed three times and used as GCsMs (dexamethasone-treated monocytes) weight on consecutive days and colitis was established (30). All animals that or Ctr-Mos, respectively. In coculture with T cells, the ratio of T cells/ fulfilled these criteria of fully developed colitis by days 19–22 were injected monocytes was always 10:1. i.v. with 2 3 106 monocytes (CGsMs or Ctr-Mos), and weight was monitored Flow cytometry (FACS) analysis and cytokine measurement for an additional 9–12 d. Animals were sacrificed between days 29 and 31 and divided into two subgroups. Colons were removed from animals of one Cells were surface stained with 1 mg/ml the indicated Ab for 30 min at 4˚C subgroup for histological analysis. Mice of the other group were used to and washed twice with PBS/1% BSA. For intracellular staining, cells were obtain mesenteric lymph nodes (MLNs) from which single-cell suspensions fixed and permeabilized according to the manufacturer’s protocol (Cytofix/ were prepared, as well as to isolate lamina propria mononuclear cells Cytoperm; BD Biosciences) and subsequently incubated with 1 mg/ml of the (LPMCs). For T cell experiments, single-cell suspensions of spleens were respective Ab for 30 min at 4˚C. For intracellular cytokine measurement, prepared, and cells were labeled with CFSE for coculture with allogeneic cells were additionally stimulated with PMA/ionomycin and treated with DCs. After 5 d of restimulation, T cell proliferation and cytokine pro- GolgiStop for 6 h before fixation. After washing twice, cells were measured duction were determined (Supplemental Fig. 1). using a FACSCalibur or FACSCanto, and data were analyzed with FlowJo 8.7 software. Cytokines were determined from supernatants of cocultures Preparation of LPMC suspensions using cytometric bead array technology (BD Biosciences), which was LPMCs were isolated from colon of mice suffering from colitis by a standard performed according to the manufacturer’s instructions. Data were analyzed method (31). Briefly, the colon was removed, opened longitudinally, cut into using FCAP Array (v1.0.1) software. 5-mm pieces, and washed with cold Ca2+/Mg2+-free HBSS. The intestinal tissue DNA microarray and statistical data analysis specimens were transferred into HBSS with EDTA to remove intraepithelial lymphocytes. After 30 min of gentle shaking at 37˚C, the samples were vor- In three independent experiments, total RNA from GCsMs and Ctr-Mos was texed, and intraepithelial lymphocyte–containing supernatant was removed. isolated and processed for microarray hybridization using Affymetrix This step was repeated twice. LPMC suspensions were prepared from the Murine Genome (430-2.0), as reported earlier (8). Microarray data were EDTA-treated de-epithelialized intestinal tissue by incubation with 100 U/ml 1092 REGULATORY MONOCYTES Downloaded from

FIGURE 1. Phenotype of GCsMs. Bone marrow monocytes were cultured for 48 h with dexamethasone (GCsM; 40 ng/ml) or with solvent (Ctr-Mo). Cells were surface stained for CD11b and CD124 (p , 0.05) or CD121b (p , 0.001) (A) or for CD11b and CD80, CD86, CD274, or GITR (CD357) (B). Cells were measured using FACS and analyzed by FlowJo 8.7 software. Cells were gated for CD11b+ cells and analyzed for the indicated surface molecules on CD11b+ cells. Graphs include three to five independent experiments. *p , 0.05, **p , 0.01, ***p , 0.001.

collagenase and 5 U/ml DNase for 30 min at 37˚C. LPMCs were washed, re- NJ), underlain with 66% Percoll solution, and centrifuged for 30 min at http://www.jimmunol.org/ suspendedin44%Percollsolution(Amersham Pharmacia Biotech, Piscataway, 600 3 g. The LPMC fraction was harvested from the interface. by guest on September 29, 2021

FIGURE 2. GCsMs inhibit proliferation of CD8+ and CD4+ T cells. Splenic T cells from naive C57BL/6 mice were isolated and labeled with CFSE. A total of 1 3 105 bulk T cells was plated in a well of a 96-well round bottom plate that was coated with 5 mg/ml anti-CD3/anti-CD28 Abs. Where indicated, 4 Ctr-Mos or GCsMs (1 3 10 ) were added to the T cells (at a T cell/monocyte ratio of 10:1). After 5 d of incubation at 37˚C, 5% CO2, cells were harvested and stained for either surface CD8 or CD4 and subsequently analyzed for proliferation by FACS. (A) One representative example of CD8+ proliferation by CFSE dilution. (B) Quantification of CD8+ T cell proliferation of eight independent experiments (p , 0.0001). (C) One representative example of CD4+ proliferation by CFSE dilution. (D) Quantification of CD4+ T cell proliferation of 11 independent experiments (p = 0.0001, p =0.022).*p , 0.05, ***p , 0.001. The Journal of Immunology 1093

Histology nomenclature, we identified functional clusters among regulated The colon was opened longitudinally, embedded as “Swiss rolls” (31) in genes by detailed research of published literature, analyzing the O.C.T. compound (Tissue-Tek; Sakura Finetek, Zoeterwoude, the Neth- occurrence of subject headings related to the functional cluster in the erlands), and kept frozen at 280˚C until further use. Sections of 5 mm National Center for Biotechnology Information’s gene-centric were stained with H&E. Histological analysis of the proximal, medial, and “Gene” database using an algorithm described previously (8). In distal colon focused on epithelial hyperplasia, leukocyte infiltration and the agreement with described functional similarities between murine and amounts of goblet cells, ulcerations, and crypt abscesses. Mean histolog- ical damage was determined by having two experienced examiners score human GC-treated monocytes (8, 21) we found an overrepresenta- each colonic section in a blinded fashion using the colitis score described tion of identical or similar gene clusters among upregulated genes by Maloy (32). like “inflammatory response,” “signal transduction,” “antioxidative/ Statistical analysis oxidative,” “scavenger receptor activity,” “antiapoptosis,” “anti-in- flammatory,” and “tolerance induction” (Supplemental Table III). Results are mean values 6 SEM. The p values are given in the figures and/ Likewise, we found an overrepresentation of functional clusters or figure legends; p . 0.05 was considered not significant. Statistical analysis was by the Student’s t test (two tailed and unpaired). like “immune response” and “IFN-gamma induced” and, strik- ingly, “T cell activation” and “Ag binding” among downregu- lated genes (Supplemental Table III). Downregulated genes that Results are important for T cell stimulation included genes like MHC-II Genome-wide expression screening and phenotypic and CD83. In contrast, and in agreement with our previous results characterization (21), we confirmed upregulation of the cell surface receptor

We performed genome-wide expression screening of murine GC- IL-4Ra (CD124) at the protein level (Fig. 1A). Expression of Downloaded from stimulated monocytes to obtain more detailed information about CD124 presents a phenotypic quality shared with a group of MDSCs. their functional potential. Using microarray technology on murine MDSCs originally were discovered in tumor-bearing mice or monocytes stimulated for 16 h with GCs, we detected significant humans, and they are known to inhibit T cell responses (22, 34). upregulation of 149 genes and downregulation of 110 genes Therefore, we analyzed expression of molecules with functional compared with Ctr-Mos (Supplemental Tables I and II); raw data relevance in the regulation of T cells. The decoy receptor for IL-1,

have been submitted to the Gene Expression Omnibus under ac- CD121b (IL-1RII), was significantly upregulated on GCsMs http://www.jimmunol.org/ cession number GSE54778. Many GC-regulated genes, such as (Fig. 1A). CD121b is expressed on myeloid cells but recently was IL-10, CD163, or IL-1RII (CD121b), showed similar regulation as demonstrated to be a unique cell surface marker of activated that for human GC-stimulated monocytes (8). Upregulation of regulatory T cells (Tregs) (35). Other molecules with known CD121b also was confirmed at the protein level (Fig. 1A). functions for monocyte–T cell interactions, such as costimulatory To gain functional insight into the pattern of regulated genes, we molecules CD80, CD86, and CD274 (PD-L1), were not regulated applied an automated unbiased functional clustering using GO- by GC (Fig. 1B). GC-induced TNFR-related protein (GITR) is Elite software (33) and gene ontology annotations to determine another molecule relevant for the function of immune cells which functional clusters among regulated genes were statistically (reviewed in Ref. 36). GITR is expressed on cells of the innate and overrepresented (Supplemental Table III). In addition to the GO adaptive immune system, but its functional role was mainly ex- by guest on September 29, 2021

FIGURE 3. GCsMs regulate cytokine production of T cells. Supernatants of anti-CD3/anti-CD28–induced T cell proliferation (black bars) and cocultures of T cells and Ctr-Mos (white bars) or GCsMs (gray bars) were harvested after 5 d of culture at 37˚C, 5% CO2 and analyzed for cytokine content. (A) IFN-g production of five independent experiments. (B) IL-17 production of five independent experiments. (C) IL-4 production of five independent experiments (p , 0.05, p , 0.0001). (D) IL-13 production of five independent experiments (p , 0.005, p , 0.0005). Intracellular staining for IFN-g (E), IL-17 (F), IL-4 (G), and IL-13 (H) in CD4+ and CD8+ T cells from cocultures in (A). Cells were stimulated with PMA/ionomycin (6 h) prior to intracellular staining. Percentage of either CD4+ or CD8+ T cells producing the respective cytokine is shown. Data are mean and SEM of four independent experiments. *p , 0.05, **p , 0.01, ***p , 0.001. 1094 REGULATORY MONOCYTES amined on T cells (reviewed in Refs. 36, 37). In this study, we IL-17, IL-4, IL-13) (Fig. 3, gray bars). Addition of Ctr-Mos also show that GITR was downregulated on GCsMs after 48 h of led to a reduction in IL-4 (Fig. 3C, white bar) and IL-13 (Fig. 3D, culture (Fig. 1B). white bar), but the effect of GCsMs was significantly stronger. In Because of the phenotypic similarity between GCsMs and contrast, release of IFN-g (Fig. 3A, white bar) and IL-17 (Fig. 3B, MDSCs, and because our functional clustering indicated that GCs white bar) was significantly reduced exclusively by GCsMs and markedly affect the capacity of monocytes to interact with T cells, not by Ctr-Mos. We additionally stained for intracellular cyto- we tested the potential of murine GCsMs to regulate T cells. kines, which revealed that the percentage of CD4+ or CD8+ T cells that produced IFN-g, IL-17, IL-4, or IL-13 was not changed by T cell proliferation is suppressed by GCsMs in vitro GCsMs (Fig. 3E–H). This indicated that GCsMs are capable of We stimulated CFSE-labeled splenic T cells from naive C57BL/6 suppressing T cell activation in terms of proliferation, which mice with anti-CD3/anti-CD28 Abs and analyzed the influence of subsequently leads to a reduction in total cytokine release. GCsMs on T cell proliferation. GCsMs significantly reduced the To test whether the upregulated surface molecules CD124 and + + proliferation of CD8 T cells (Fig. 2A, 2B) and CD4 Tcells CD121b (Fig. 1) are responsible for regulation of T cell functions + (Fig. 2C, 2D). Strikingly, the proliferation of CD4 Tcellswas by GCsMs, we used blocking Abs to CD121b and CD124 on diminished more by GCsMs (Fig. 2C, right panel)thanwasthat monocytes and subsequently tested whether this interference abol- + of CD8 Tcells. ished the suppression of T cell proliferation by culture with GCsMs compared with Ctr-Mos. We did not observe differences in either Cytokine production is downregulated by GCsMs in vitro T cell proliferation or cytokine production after blocking CD121b

Next, we analyzed whether and how GCsMs influence cytokine or CD124 in cocultures with GCsMs or with Ctr-Mos (data not Downloaded from production by activated T cells (Fig. 3). We measured the release shown). of proinflammatory cytokines IFN-g and IL-17, as well as that of Thus, CD121b and CD124, as well as their increased expression, Th2 cytokines IL-4 and IL-13, from supernatants of bulk T cells do not appear to be functionally involved in the suppression of that were stimulated with anti-CD3/anti-CD28 and cocultured T cells by GCsMs in vitro. with the respective monocytes. Because IL-13 also can be released GCsMs induce Foxp3+ Tregs through repetitive stimulation of

by monocytes/macrophages, we first measured LPS-stimulated http://www.jimmunol.org/ Ctr-Mos and GCsMs. Neither unstimulated nor LPS-stimulated naive T cells Ctr-Mos or GCsMs produced measurable IL-13 in our setting Other myeloid cells, such as CD40L- and IFN-g2treated macro- (data not shown). Addition of GCsMs to activated T cells resulted phages (38) and some MDSCs (39), were shown to regulate CD4+ in a marked and significant reduction in all four cytokines (IFN-g, T cell activation because of the induction of Tregs. Therefore, we by guest on September 29, 2021

FIGURE 4. Repetitive stimulation of CD4+ T cells by GCsMs generates Foxp3+ T cells in vitro. (A) Experimental setting for repetitive stimulation of T cells. (B) Purified splenic CD4+ T cells (1 3 106/ml) were cocultured with the indicated monocytes at a 10:1 ratio for 5 d. Subsequently, bulk coculture cells were harvested and incubated in a second cycle with fresh monocytes at the same ratio for an additional 5 d. A third cycle was conducted in the same way. First- and third-cycle cells were collected, stained for CD4 on the surface, and stained intracellularly for Foxp3 expression. One representative example is shown for coculture of T cells and Ctr-Mos or GCsMs. (C) Percentage of Foxp3+ cells among CD4+ cells after the third cycle of repetitive stimulation (six independent experiments). *p , 0.05. The Journal of Immunology 1095 analyzed whether CD4+ T cells expressed Foxp3 (Fig. 4B, upper GCsMs cure CD4+ T cell–induced colitis in vivo + panels) and CTLA-4 (CD152), two signature molecules of CD4 Considering the marked effects, especially on CD4+ T cells Tregs, after 5 d of coculture with GCsMs. The percentages of + + + + in vitro, we wondered whether GCsMs similarly influence Foxp3 CD4 T cells (Fig. 4B) and CTLA-4 CD4 T cells (data not CD4+ T cells in vivo and, thus, mediate GC effects in situ. shown) did not differ after coculture of T cells with GCsMs or Ctr- Therefore, we investigated GCsMs’ effects on CD4 cell–de- Mos. However, it was shown previously that repetitive stimulation pendent inflammatory reactions in vivo. We chose CD4-dependent of naive T cells with immature myeloid cells [originally reported for inflammatory bowel disease (32, 40–42) because, in this model, DCs (27)] is required to induce Tregs. We transferred this approach 2 syngeneic CD4+CD25 T cells induce severe and fatal colitis to our murine system and showed that repetitive stimulation of 2 2 when adoptively transferred into Rag / mice (41). Colitis was naive splenic T cells with GCsMs (Fig. 4A) indeed increased the monitored clinically by weight loss (Fig. 5). To test the influence proportion of Foxp3+ Tregs more strongly (from 7 to 41.5%) than did repetitive coculture with Ctr-Mos (from 5.9 to 20.8%) (Fig. 4B). of GCsMs on T cells already activated in vivo, we injected vehicle The overall proportion of Foxp3+ CD4+ Tregs increased signifi- (PBS), Ctr-Mos, or GCsMs i.v. in a therapeutic setting [i.e., when cantly when naive T cells were repetitively cocultured with GCsMs mice had already developed severe colitis (Fig. 5A, 5B, days 19– (Fig. 4B, 4C). 22)]. This stage of disease is reflected by weight loss on consec- However, the generation of Foxp3+CD4+ Tregs was not man- utive days. For the ensuing experiments, we only used animals datory for suppression of proliferation and cytokine release that fulfilled this criterion between days 19 and 22 (Fig. 5B). Mice mentioned above, because these processes also took place without that received GCsMs showed dramatic and significant clinical repetitive stimulation of naive splenic T cells when no Tregs were improvement over a period of 9–11 d postcell transfer (Fig. 5A, Downloaded from generated (Figs. 2, 4B). 5C). In contrast, mice that received Ctr-Mos continued to lose http://www.jimmunol.org/ by guest on September 29, 2021

FIGURE 5. Impact of GCsMs on T cell–mediated colitis in vivo. (A) Weight monitoring (% of initial weight) in model of T cell transfer colitis. On day 0, T cell transfer was performed i.v.; between days 19 and 22, monocyte transfer was done i.v. Mice were sacrificed between days 29 and 31. Colon, MLNs, and spleen were harvested for further analysis. Graph shows mean and SEM of n = 6 (PBS) and n = 10 (Ctr-Mo and GCsM) mice from three independent experiments. p , 0.05 (days 25–27), p , 0.001 (days 29–31), Ctr-Mo versus GCsM, Student t test. (B) Weight of individual Rag2/2 mice in different groups from day 0 to day 19/22 before monocyte transfer. (C) Weight of individual Rag2/2 mice of different groups from monocyte transfer (days 19/22) to termination of the experiment (days 29/31). Weight difference between days 19/22 and days 29/31 were calculated for each group and tested for sig- nificance. p , 0.001 (GCsM versus PBS), p , 0.0001 (GCsM versus Ctr-Mo). (D). Colon histology (H&E staining) after termination of the experiment. Images are representative of at least five mice for each group: healthy control (naive mouse, upper left image), colitis induced and injected with PBS (upper right image), colitis induced and treatment with Ctr-Mo (lower left image), colitis induced and treatment with GCsMs (lower right image). (E) Mean histology score for animals in (A) was determined as described in Materials and Methods (p , 0.0001). *p , 0.05, **p , 0.01, ***p , 0.001. 1096 REGULATORY MONOCYTES weight (Fig. 5A, 5C), as did mice that received vehicle (PBS) only treated animals. Also, IL-17 production was strongly downregu- (Fig. 5A, 5C). In those experiments in which mice with colitis lated in MLN cells (as well as in T cells from spleen; Sup- were monitored for longer periods after treatment with GCsMs, plemental Fig. 1) but not in cells from lamina propria (LPMCs). three of four animals survived, whereas three of four untreated Other than in the in vitro system, production of IL-4 and IL-13 mice died by day 60 (data not shown). was not influenced in single-cell suspensions from LPMCs, Histological analysis of colons at days 29–31 revealed dense and MLN cells and splenic T cells after treatment of mice with in- large inflammatory infiltrates in untreated mice with colitis that jection of GCsM (Fig. 6, Supplemental Fig. 1). In addition, IL- received no GCsMs (Fig. 5D, upper right image) or Ctr-Mos 10 was determined from restimulated LPMCs and MLN cells, (Fig. 5D, lower left image) compared with naive mice (Fig. 5D, but it was below the detection limit of the assay system in all upper left image). After injection of GCsMs, these infiltrates re- cases (data not shown). solved almost completely (Fig. 5D, lower right image), reflecting Thus, therapeutic transfer of GCsMs resulted in alteration of the the marked clinical improvement. To quantify this observation, we cytokine pattern released by T cells from mucosal lamina propria, generated histological scores from colon sections of these mice. MLN cells, and spleen, exhibiting suppression of IFN-g by all The scores depicted in Fig. 5E clearly emphasize the highly sig- populations and suppression of IL-17 by T cells from MLNs and nificant effect of GCsMs on the improvement of inflammation spleen. compared with treatment with Ctr-Mos or vehicle (PBS). Mechanisms of GCsMs on CD4+ T cells in vivo: inducing Taken together, treatment of established, severe CD4+ T cell– differentiation of Foxp3+CD4+ Tregs induced colitis by injection of GCsMs resulted in complete res- olution, whereas Ctr-Mos and vehicle had no effect. Thus, GCsMs To explore whether downregulation of colitis by GCsMs in vivo Downloaded from exert regulatory effects on activated CD4+ T cells, with marked would also encompass differentiation of T cells into Tregs (as effects on inflammation in vivo. seen in vitro after repetitive interactions), we analyzed Foxp3 + expression ex vivo and in situ. First, we analyzed Foxp3 expression Mechanisms of GCsMs on CD4 T cells in vivo: suppression of in CD4+ T cells from lamina propria (LPMCs) and MLNs, all IL-17 and IFN-g obtained from mice suffering from colitis. Although the T cells

To investigate whether inhibition of colitis by GCsMs was due to transferred into mice on day 0 were negative for Foxp3 (data not http://www.jimmunol.org/ alteration or downregulation of activated T cells, we examined shown), a certain percentage of Foxp3+ cells emerged physio- LPMCs, MLN cells, and splenocytes from mice with colitis. Be- logically in all treatment groups (Fig. 7A), and a small amount cause naive Rag2/2 mice do not harbor intrinsic T cells, all T cells already was detectable on the day of treatment (days 19–22, data derive from initially transferred cells. The yield of CD4+ T cells in not shown) when colitis had fully developed. However, as shown lamina propria (LPMCs), MLNs, and spleen was very low, and we in Fig. 7A, both the numbers and percentages of Foxp3+CD4+ used cell suspensions of these for restimulation with anti-CD3/ T cells in LPMCs and MLNs were not different between the anti-CD28 Abs. Prior to stimulation, we ascertained the presence groups tested. Histological analysis of cells in situ in the colon, the of CD4+ T cells in suspensions of LPMCs, MLNs, and spleens by organ primarily affected in inflammatory bowel disease, revealed + FACS (data not shown). After 3 d of stimulation, supernatants that, although few Foxp3 cells (Fig. 7B) were detected in LPMCs by guest on September 29, 2021 of cultured LPMCs, MLN cells, and splenic T cells were ana- by FACS, they were not found in cryosections of naive mice (left lyzed for cytokine production. Unstimulated cultures revealed no panel). In colon sections of mice that developed colitis and were detectable cytokine production (data not shown). IFN-g pro- treated with Ctr-Mos, few single Foxp3+ cells were detectable duction from LPMCs, MLN cells (Fig. 6), and splenic T cells (middle panel). However, colon sections of mice that received (Supplemental Fig. 1C) was significantly reduced in GCsM- GCsMs (right panel) contained clusters of Foxp3+ cells, which

FIGURE 6. CD4+ T cells are regulated by GCsMs in vivo. MLN and LPMC single-cell suspensions from mice used in colitis experiments (Fig. 5) were prepared as described in Materials and Methods. LPMC and MLN single cells (1 3 105/well of a 96-well plate) were restimulated with precoated anti-CD3/ anti-CD28 Abs (5 mg/ml each) for 3 d, and supernatants were harvested. Cytokines of supernatants were measured by FACS using cytometric bead array technology. Data are mean 6 SEM of 4–10 individual mice from three independent experiments. *p , 0.05, ***p , 0.001. The Journal of Immunology 1097 Downloaded from http://www.jimmunol.org/

FIGURE 7. Presence of GCsMs in vivo leads to accumulation of Foxp3+CD4+ T cells. (A) LPMC and MLN single-cell suspensions from mice used in transfer colitis experiments (Fig. 5) were stained with Abs against CD4 on the surface and Foxp3 intracellularly. Graph shows individual and mean values 6 SEM for seven to nine mice from three independent experiments. (B) Foxp3 Ab staining of colon sections. Colon section of naive Rag2/2 mouse without colitis (left panel). No Foxp3+ cells are detectable (see arrow and magnified inset). Colon section of mouse with colitis treated with Ctr-Mos (middle panel). Some Foxp3+ cells are detectable (see arrow and magnified inset). Colon section of mouse with colitis treated with GCsMs (right panel). Large clusters of by guest on September 29, 2021 Foxp3+ cells can be detected (see arrow and magnified inset). Scale bar, 50 mm. (C) Clusters of Foxp3+ cells/section (more than five positive cells = one cluster) were counted in eight sections in each group by four independent observers. Data are mean 6 SEM. p = 0.0366. (D) Fluorescence microscopy of colon sections of mice with colitis treated with GCsMs. Sections were stained with DAPI for cell nuclei (blue). Foxp3 was stained in parallel in all sections (green). In addition, cells were costained for CD4 (upper left panel), F4/80 (upper right panel), Ly6C (lower left panel), or Ly6G (lower right panel) (all red) and analyzed by fluorescence microscopy (Zeiss Axio Observer Z1; Zeiss, Jena, Germany). apparently developed locally at the site of inflammation. To IL-17 in vivo. Thus, GCsMs are capable of downregulating an quantify Foxp3 expression, clusters (defined as more than five already established severe immune response in vivo. Foxp3+ cells) were analyzed from eight sections for each group Because GCsMs share the expression of CD11b, Gr-1, and CD124 (Fig. 7C). Colons of GCsM-treated animals displayed significantly (21) with monocytic MDSCs, which are known suppressors of more Foxp3+ clusters compared with colons of mice treated with T cell activation (30, 37), and because genes involved in T cell Ctr-Mos. To identify the type of cell that presented with Foxp3 interaction were overrepresented among GC-regulated genes, we expression, we performed double staining of tissue sections for wondered whether GCsMs would suppress T cell activation. We fluorescence microscopy (Fig. 7D). Counterstaining Foxp3+ cells demonstrated that GCsMs suppress activation of CD8+ Tcells (green) with CD4, F4/80, Ly6C, or Ly6G (red) and with DAPI in vitro, as was described for MDSCs (39, 43–45). Interestingly, (blue) showed that only CD4+ cells also were Foxp3+; F4/80+, GCsMs were even more effective suppressors of CD4+ Tcell Ly6C+, and Ly6G+ cells were negative for Foxp3. responses. GCsMs inhibited proliferation of polyclonally activated Takentogether,thesedataindicatethatCD4+Foxp3+ Tregs CD4+ T cells and substantially reduced their cytokine production accumulated locally but did not expand in draining lymph nodes in vitro. (MLNs) and LPMCs in animals treated with GCsMs. It was proposed that GCs downregulate Th1 responses and in- duce a Th2 shift (by a combination of direct effects on T cells and Discussion indirect effects on APCs) (42), but it only has been shown that GCs In this study, we demonstrate that GCsMs regulate activated CD8+ reduce the release of IL-12 after classical activation of monocytes and, especially, CD4+ T cells by reducing their proliferation and, or macrophages (46, 47) and favor Th2 responses in this setting consequently, their total cytokine release in vitro. Also, repetitive (42, 47). In our study, regulation of cytokine production of T cells encounter between CD4+ T cells and GCsMs induces Foxp3+ by GCsMs did not result in a Th2 shift in vitro or in vivo. In vitro, Tregs in vitro. GCsMs inhibited activation of naive T cells and suppressed all In vivo injection of GCsMs into mice suffering from CD4+ cytokines examined. However, in vivo, the inflammatory cytokines T cell–induced colitis resulted in clinical healing. This remarkable IFN-g and IL-17 released by activated T cells appeared to be downregulation was associated with altered release of IFN-g and regulated selectively by GCsMs. 1098 REGULATORY MONOCYTES

Hence, we now describe a direct effect of GC on naive monocytes that Tregs generate intrinsically (see PBS group) as a natural that results in a subtype that actively modifies T cell responses. mechanism to counter-regulate inflammation. None of the mono- To explore the mechanisms of T cell regulation by GCsMs, we cytic subtypes transferred into colitis mice changed the amount of used neutralizing Abs against the immunosuppressive cytokine IL-10 Tregs in the LPMCs and MLNs. However, the presence of CD4+ and the surface molecules CD80, CD121b, and CD124 in cocultures Foxp3+ T cell clusters in colonic epithelium after treatment with of GCsMs and T cells. There are conflicting data about the mech- GCsMs provides evidence for local accumulation of Tregs and, anisms of MDSC-mediated regulation. The IL-4Ra-chain (CD124) possibly, for local activation at the site of inflammation. Hence, we was proposed to be involved in the suppressive activity of MDSCs speculate that GCsMs may be able to induce local accumulation of toward CD8+ T cells (44), but those results were called into ques- Tregs in vivo at sites of inflammation while directly suppressing tion by a study that demonstrated that CD124 is not involved in the activated T cells. Whether this generation of Tregs at sites of regulatory function of MDSCs (48). Furthermore, CD80 was de- inflammation might also contribute to the initial control of estab- scribed to be used by MDSCs to inhibit IFN-g production and lished colitis or whether it would induce a longer-lasting state of proliferation of T cells (49). In addition, CD121b was reported tolerance has to be explored. Because the in vitro effect of GCsMs to be a possible marker for Tregs that could be involved in their to induce Tregs takes some time, they might not be involved in the suppressive capacity on effector T cells by sequestering IL-1 from initial control of inflammation, but they may be important for the environment (35). We did not observe any influence of IL-10, maintaining control. CD80, CD121b, or CD124 on the GCsM-mediated regulation of Taken together, we suggest that GCsMs use at least two distinct T cell activation in vitro (data not shown). mechanisms to control and regulate CD4+ T cells. One is a fast and

For human DCs, it was shown that GC treatment, together with very efficient (due to a T cell/GCsM ratio of 10:1) direct inhibition Downloaded from either vitamin D (50) or CD40 ligation (51), generates tolerogenic of activated T cells without the participation of Tregs (as shown DCs that require high IL-10 or CD274 (PDL-1) expression to directly in vitro). The other mechanism is their capacity to induce induce Tregs. We were unable to detect increased IL-10 protein Tregs in vitro and locally at sites of inflammation in vivo. Thus, it production (data not shown) or CD274 on GCsMs. Nevertheless, is tempting to speculate that GCsMs have the unique dual ability we explored the possibility that inhibition of activated CD4+ to actively inhibit acute inflammation mediated by CD4+ T cells + T cells was also indirect and mediated by the induction of Foxp3 (proliferation and cytokine pattern) and simultaneously induce a http://www.jimmunol.org/ Tregs. When we analyzed Foxp3 expression in T cells after 5 d of longer-lasting state of tolerance via Tregs. coculture with GCsMs, we did not observe an increase in the Moreover, GCsMs could be a promising novel target of thera- Foxp3+ Treg population, although suppressive activity was fully peutic strategies to exploit the efficacy of GCs without their adverse developed (Figs. 2, 3). effects and, in this context, to support distinctly active and long- However, GCsMs are able to differentiate Foxp3+ Tregs from lasting suppression of T cell–mediated inflammation through the naive T cells in vitro when stimulated repetitively (Fig. 4), as was innate immune system. described for immature APCs (27). Expression of GITR protein is implicated in costimulation of Acknowledgments cells of the innate and adaptive immunity (52). In the setting of by guest on September 29, 2021 We thank Andrea Stadtba¨umer, Eva Nattkemper, and Tarek Chehab for transfer colitis, GITR expression on APCs was critical for the excellent technical assistance. development of Treg function (52). Therefore, we analyzed the expression of GITR on GCsMs; however, GITR was significantly downregulated on GCsMs, indicating that it is not involved in Disclosures GCsM-mediated mechanisms on T cells. The authors have no financial conflicts of interest. Taken together, these facts point to an as yet unidentified mecha- nism that is used by GCsMs to suppress T cell activation in vitro. It References + exerts its effects faster than Foxp3 Tregs and prior to their ap- 1. Almawi, W. Y., H. N. Beyhum, A. A. Rahme, and M. J. Rieder. 1996. Regulation pearance. It is noteworthy that GCsMs confer their regulatory activity of cytokine and cytokine receptor expression by glucocorticoids. J. Leukoc. Biol. at a 10:1 ratio (T cell/monocyte) and, thus, are far more effective than 60: 563–572. 2. Stahn, C., M. Lo¨wenberg, D. W. Hommes, and F. Buttgereit. 2007. Molecular Tregs, which need ratios of 2:1 (T cell/Treg) to be inhibitory in vitro. mechanisms of glucocorticoid action and selective glucocorticoid receptor The marked capability of these GCsMs to suppress CD4+ agonists. Mol. Cell. Endocrinol. 275: 71–78. 3. Ma, W., K. Gee, W. Lim, K. Chambers, J. B. Angel, M. Kozlowski, and A. Kumar. T cells responses in vitro prompted us to investigate whether 2004. Dexamethasone inhibits IL-12p40 production in lipopolysaccharide- they are able to suppress established T cell responses in vivo. stimulated human monocytic cells by down-regulating the activity of c-Jun Transfer of GCsMs led to rapid amelioration, and even healing, N-terminal kinase, the activation protein-1, and NF-kappa B transcription factors. J. Immunol. 172: 318–330. of T cell–induced colitis. The clinical effect correlated with an 4. Kleiman, A., and J. P. Tuckermann. 2007. Glucocorticoid receptor action in inhibited secretion of the inflammatory cytokines IFN-g and IL- beneficial and side effects of steroid therapy: lessons from conditional knockout 17 by CD4+ T cells from corresponding LPMCs (for IFN-g), mice. Mol. Cell. Endocrinol. 275: 98–108. 5. Ogawa, S., J. Lozach, C. Benner, G. Pascual, R. K. Tangirala, S. Westin, MLNs, and spleens ex vivo. A. Hoffmann, S. Subramaniam, M. David, M. G. Rosenfeld, and C. K. Glass. IL-10 does not seem to play a role systemically because it was not 2005. Molecular determinants of crosstalk between nuclear receptors and toll- like receptors. Cell 122: 707–721. detectable upon restimulation of LPMCs and MLN cells (data not 6. Patel, H. B., K. N. Kornerup, A. L. F. Sampaio, F. D’Acquisto, M. P. Seed, 2/2 shown). Because Rag mice do not have intrinsic T cells (28), A. P. Girol, M. Gray, C. Pitzalis, S. M. Oliani, and M. Perretti. 2012. The impact T cells that were regulated by GCsMs were the colitogenic CD4+ of endogenous annexin A1 on glucocorticoid control of inflammatory arthritis. 2 Ann. Rheum. Dis. 71: 1872–1880. CD25 T cells that were transferred i.v. on day 0. 7. Perretti, M., and F. D’Acquisto. 2009. Annexin A1 and glucocorticoids as Because we observed that GCsMs have the potential to generate effectors of the resolution of inflammation. Nat. Rev. Immunol. 9: 62–70. € Tregs in vitro (although they were not required for the short-term 8. Ehrchen, J., L. Steinmuller, K. Barczyk, K. Tenbrock, W. Nacken, M. Eisenacher, U. Nordhues, C. Sorg, C. Sunderko¨tter, and J. Roth. 2007. suppressive effects) and because Tregs can arise locally at the site Glucocorticoids induce differentiation of a specifically activated, anti- of the immune response (53), we investigated whether Tregs would inflammatory subtype of human monocytes. Blood 109: 1265–1274. + + 9. Barczyk, K., J. Ehrchen, K. Tenbrock, M. Ahlmann, J. Kneidl, D. Viemann, and be generated in colonic mucosa. Examination of CD4 Foxp3 J. Roth. 2010. Glucocorticoids promote survival of anti-inflammatory macro- T cells from lamina propria (LPMCs) and MLNs clearly showed phages via stimulation of adenosine receptor A3. Cancer 116: 1–3. The Journal of Immunology 1099

10. Tsianakas, A., G. Varga, K. Barczyk, G. Bode, N. Nippe, N. Kran, J. Roth, 33. Zambon, A. C., S. Gaj, I. Ho, K. Hanspers, K. Vranizan, C. T. Evelo, T. A. Luger, J. Ehrchen, and C. Sunderkoetter. 2012. Induction of an anti- B. R. Conklin, A. R. Pico, and N. Salomonis. 2012. GO-Elite: a flexible solution inflammatory human monocyte subtype is a unique property of glucocorti- for pathway and ontology over-representation. Bioinformatics 28: 2209–2210. coids, but can be modified by IL-6 and IL-10. Immunobiology 217: 329–335. 34. Sica, A., and V. Bronte. 2007. Altered macrophage differentiation and immune 11. Mosser, D. M. 2003. The many faces of macrophage activation. J. Leukoc. Biol. dysfunction in tumor development. J. Clin. Invest. 117: 1155–1166. 73: 209–212. 35. Tran, D. Q., J. Andersson, D. Hardwick, L. Bebris, G. G. Illei, and 12. Gordon, S., and P. R. Taylor. 2005. Monocyte and macrophage heterogeneity. E. M. Shevach. 2009. Selective expression of latency-associated peptide (LAP) Nat. Rev. Immunol. 5: 953–964. and IL-1 receptor type I/II (CD121a/CD121b) on activated human FOXP3+ 13. Wells, C. A., T. Ravasi, and D. A. Hume. 2005. Inflammation suppressor genes: regulatory T cells allows for their purification from expansion cultures. Blood please switch out all the lights. J. Leukoc. Biol. 78: 9–13. 113: 5125–5133. 14. Auffray, C., M. H. Sieweke, and F. Geissmann. 2009. Blood monocytes: de- 36. Nocentini, G., S. Ronchetti, M. G. Petrillo, and C. Riccardi. 2012. Pharmaco- velopment, heterogeneity, and relationship with dendritic cells. Annu. Rev. logical modulation of GITRL/GITR system: therapeutic perspectives. Br. J. Immunol. 27: 669–692. Pharmacol. 165: 2089–2099. 15. Gordon, S. 2003. Alternative activation of macrophages. Nat. Rev. Immunol. 3: 23–35. 37. Schaer, D. A., J. T. Murphy, and J. D. Wolchok. 2012. Modulation of GITR for 16. Mosser, D. M., and J. P. Edwards. 2008. Exploring the full spectrum of mac- cancer immunotherapy. Curr. Opin. Immunol. 24: 217–224. rophage activation. Nat. Rev. Immunol. 8: 958–969. 38. Brem-Exner, B. G., C. Sattler, J. A. Hutchinson, G. E. Koehl, K. Kronenberg, 17. Geissmann, F., S. Jung, and D. R. Littman. 2003. Blood monocytes consist of S. Farkas, S. Inoue, C. Blank, S. J. Knechtle, H. J. Schlitt, et al. 2008. Macro- two principal subsets with distinct migratory properties. Immunity 19: 71–82. phages driven to a novel state of activation have anti-inflammatory properties in 18. Sunderko¨tter, C., T. Nikolic, M. J. Dillon, N. Van Rooijen, M. Stehling, mice. J. Immunol. 180: 335–349. D. A. Drevets, and P. J. M. Leenen. 2004. Subpopulations of mouse blood 39. Solito, S., V. Bronte, and S. Mandruzzato. 2011. Antigen specificity of immune monocytes differ in maturation stage and inflammatory response. J. Immunol. suppression by myeloid-derived suppressor cells. J. Leukoc. Biol. 90: 31–36. 172: 4410–4417. 40. Powrie, F., J. Carlino, M. W. Leach, S. Mauze, and R. L. Coffman. 1996. A 19. Strauss-Ayali, D., S. M. Conrad, and D. M. Mosser. 2007. Monocyte sub- critical role for transforming growth factor-beta but not interleukin 4 in the populations and their differentiation patterns during infection. J. Leukoc. Biol. suppression of T helper type 1-mediated colitis by CD45RB(low) CD4+ T cells. 82: 244–252. J. Exp. Med. 183: 2669–2674. 20. Biermann, H., B. Pietz, R. Dreier, K. W. Schmid, C. Sorg, and C. Sunderko¨tter. 41. Mudter, J., S. Wirtz, P. R. Galle, and M. F. Neurath. 2002-2003. A new model of Downloaded from 1999. Murine leukocytes with ring-shaped nuclei include granulocytes, mono- chronic colitis in SCID mice induced by adoptive transfer of CD62L+ CD4+ cytes, and their precursors. J. Leukoc. Biol. 65: 217–231. T cells: insights into the regulatory role of interleukin-6 on apoptosis. Patho- 21. Varga, G., J. Ehrchen, A. Tsianakas, K. Tenbrock, A. Rattenholl, S. Seeliger, biology 70: 170–176. M. Mack, J. Roth, and C. Sunderkoetter. 2008. Glucocorticoids induce an acti- 42. Elenkov, I. J. 2004. Glucocorticoids and the Th1/Th2 balance. Ann. N. Y. Acad. vated, anti-inflammatory monocyte subset in mice that resembles myeloid- Sci. 1024: 138–146. derived suppressor cells. J. Leukoc. Biol. 84: 644–650. 43. Nagaraj, S., M. Collazo, C. A. Corzo, J.-I. Youn, M. Ortiz, D. Quiceno, and 22. Gabrilovich, D. I., and S. Nagaraj. 2009. Myeloid-derived suppressor cells as D. I. Gabrilovich. 2009. Regulatory myeloid suppressor cells in health and

regulators of the immune system. Nat. Rev. Immunol. 9: 162–174. disease. Cancer Res. 69: 7503–7506. http://www.jimmunol.org/ 23. Gabrilovich, D. I., S. Ostrand-Rosenberg, and V. Bronte. 2012. Coordinated 44. Gallina, G., L. Dolcetti, P. Serafini, C. De Santo, I. Marigo, M. P. Colombo, regulation of myeloid cells by tumours. Nat. Rev. Immunol. 12: 253–268. G. Basso, F. Brombacher, I. Borrello, P. Zanovello, et al. 2006. Tumors induce 24. Krummen, M. B. W., G. Varga, M. Steinert, A. Stuetz, T. A. Luger, and a subset of inflammatory monocytes with immunosuppressive activity on CD8+ S. Grabbe. 2006. Effect of pimecrolimus vs. corticosteroids on murine bone T cells. J. Clin. Invest. 116: 2777–2790. marrow-derived dendritic cell differentiation, maturation and function. Exp. 45. Huang, B., P.-Y. Pan, Q. Li, A. I. Sato, D. E. Levy, J. Bromberg, C. M. Divino, Dermatol. 15: 43–50. and S.-H. Chen. 2006. Gr-1+CD115+ immature myeloid suppressor cells me- 25. Varga, G., S. Balkow, M. K. Wild, A. Stadtbaeumer, M. Krummen, T. Rothoeft, diate the development of tumor-induced T regulatory cells and T-cell anergy in T. Higuchi, S. Beissert, K. Wethmar, K. Scharffetter-Kochanek, et al. 2007. tumor-bearing host. Cancer Res. 66: 1123–1131. Active MAC-1 (CD11b/CD18) on DCs inhibits full T-cell activation. Blood 109: 46. Blotta, M. H., R. H. DeKruyff, and D. T. Umetsu. 1997. Corticosteroids inhibit 661–669. IL-12 production in human monocytes and enhance their capacity to induce IL-4 26. Ahlmann, M., G. Varga, K. Sturm, R. Lippe, K. Benedyk, D. Viemann, synthesis in CD4+ lymphocytes. J. Immunol. 158: 5589–5595. € T. Scholzen, J. Ehrchen, F. U. Muller, M. Seidl, et al. 2009. The cyclic AMP 47. DeKruyff, R. H., Y. Fang, and D. T. Umetsu. 1998. Corticosteroids enhance the by guest on September 29, 2021 response element modulator alpha suppresses CD86 expression and APC func- capacity of macrophages to induce Th2 cytokine synthesis in CD4+ lymphocytes tion. J. Immunol. 182: 4167–4174. by inhibiting IL-12 production. J. Immunol. 160: 2231–2237. 27. Jonuleit, H., E. Schmitt, G. Schuler, J. Knop, and A. H. Enk. 2000. Induction of 48. Sinha, P., K. H. Parker, L. Horn, and S. Ostrand-Rosenberg. 2012. Tumor- interleukin 10-producing, nonproliferating CD4(+) T cells with regulatory induced myeloid-derived suppressor cell function is independent of IFN-g and properties by repetitive stimulation with allogeneic immature human dendritic IL-4Ra. Eur. J. Immunol. 42: 2052–2059. cells. J. Exp. Med. 192: 1213–1222. 49. Yang, R., Z. Cai, Y. Zhang, W. H. Yutzy, IV, K. F. Roby, and R. B. S. Roden. 28. Mombaerts, P., J. Iacomini, R. S. Johnson, K. Herrup, S. Tonegawa, and 2006. CD80 in immune suppression by mouse ovarian carcinoma-associated V. E. Papaioannou. 1992. RAG-1-deficient mice have no mature B and Gr-1+CD11b+ myeloid cells. Cancer Res. 66: 6807–6815. T lymphocytes. Cell 68: 869–877. 50. Unger, W. W., S. Laban, F. S. Kleijwegt, A. R. van der Slik, and B. O. Roep. 29. Kannengiesser, K., C. Maaser, J. Heidemann, A. Luegering, M. Ross, 2009. Induction of Treg by monocyte-derived DC modulated by vitamin D3 or T. Brzoska, M. Bohm, T. A. Luger, W. Domschke, and T. Kucharzik. 2008. dexamethasone: differential role for PD-L1. Eur. J. Immunol. 39: 3147–3159. Melanocortin-derived tripeptide KPV has anti-inflammatory potential in murine 51. Rea, D., C. van Kooten, K. E. van Meijgaarden, T. H. Ottenhoff, C. J. Melief, and models of inflammatory bowel disease. Inflamm. Bowel Dis. 14: 324–331. R. Offringa. 2000. Glucocorticoids transform CD40-triggering of dendritic cells 30. Strober, W., I. J. Fuss, and R. S. Blumberg. 2002. The immunology of mucosal into an alternative activation pathway resulting in antigen-presenting cells that models of inflammation. Annu. Rev. Immunol. 20: 495–549. secrete IL-10. Blood 95: 3162–3167. 31. Lugering,€ A., T. Kucharzik, D. Soler, D. Picarella, J. T. Hudson, III, and 52. Liao, G., C. Detre, S. B. Berger, P. Engel, R. de Waal Malefyt, R. W. Herzog, I. R. Williams. 2003. Lymphoid precursors in intestinal cryptopatches express A. K. Bhan, and C. Terhorst. 2012. Glucocorticoid-induced tumor necrosis factor CCR6 and undergo dysregulated development in the absence of CCR6. J. receptor family-related protein regulates CD4(+)T cell-mediated colitis in mice. Immunol. 171: 2208–2215. Gastroenterology 142: 582–591.e8. 32. Maloy, K. J. 2007. Induction and regulation of inflammatory bowel disease in 53. Barnes, M. J., and T. Griseri, A. M. Johnson, W. Young, F. Powrie, and A. Izcue. immunodeficient mice by distinct CD4+ T-cell subsets. Methods Mol. Biol. 380: 2012. CTLA-4 promotes Foxp3 induction and regulatory T cell accumulation in 327–335. the intestinal lamina propria. Mucosal Immunol. 6: 324–334. Supplemental Figure 1: CD4+ T-cells are regulated by GCsM in vivo

A. Spleen cells restimulated with C. E. allogeneic DC (50:1) – 5 days IL-4

Ctr-Mo GCsM

*

CD4 DC (50:1)

CFSE

B. D. IL-17 F. IL-13

* *

DC (50:1)

A. Spleen cells of mice from colitis experiments (day 29) were CFSE-labeled and restimulated with allogeneic dendritic cells (DC) for 5 days to assess activation state of in vivo differentiated T-cells (ratio T:DC = 50:1). Cells from co-culture were stained with anti-CD4 antibody and analyzed for proliferation (CFSE-dilution) using FACS. Plots show representative T-cell proliferation of CD4+ T -cells from mice of colitis experiment (see Fig. 5). B. Summary of proliferation of splenic CD4+ T-cells from 8 mice of each group from 3 independent experiments. Mean and SEM of unstimulated and restimulated (T:DC = 50:1) spleen cells of colitis mice from Figure 5 is shown. Ctr-Mo: 12.6% +/- 1.7 vs. Mregs: 8.4% +/- 0.9 *p=0.0449. C-F. Supernatants of co-cultures from B) were analyzed for cytokines using CBA technology. Graphs show mean values +/- SEM of 4-8 individual mice from 3 independent experiments. C. IFNg: Ctr-Mo group: 976.9 pg/ml +/- 77.3 vs. GCsM group: 348.5 pg/ml +/- 54.4 D. IL-17: Ctr-Mo group: 105.3 pg/ml +/- 28.6 vs. GCsM group: 36.6 pg/ml +/- 11.0 E. IL-4: Ctr-Mo group: 145.3 pg/ml +/- 52.25 vs. GCsM group: 164.5 pg/ml +/- 37.7 F. IL-13: Ctr-Mo group: 275.6 pg/ml +/- 54.64 vs. GCsM group: 183.8 pg/ml +/- 55.03 Supplemental Table 1: Up-regulated genes Affymetrix identifier Gene symbol N-fold P-Value 1452614_at Gm566 266.5 0.0048 1427345_a_at Sult1a1 45.3 0.0473 1424968_at 2210023G05Rik 24.6 0.0058 1418858_at Aox3 21.7 0.0090 1451932_a_at Adamtsl4 21.5 0.0002 1418797_at Ms4a8a 20.5 0.0012 1457438_at Gm11827 18.6 0.0048 1448233_at Prnp 17.7 0.0024 1419874_x_at Zbtb16 15.0 0.0030 1419532_at Il1r2 10.0 0.0018 1448231_at Fkbp5 9.8 0.0043 1451054_at Orm1 8.9 0.0008 1428306_at Ddit4 8.2 0.0184 1441799_at 6030422H21Rik 7.9 0.0006 1448681_at Il15ra 7.8 0.0197 1437056_x_at Crispld2 7.6 0.0158 1425295_at Ear11 7.6 0.0006 1418509_at Cbr2 7.4 0.0416 1448613_at Ecm1 7.4 0.0049 1428332_at Pik3ip1 7.2 0.0021 1428288_at 2310051E17Rik 7.0 0.0024 1425281_a_at Tsc22d3 6.5 0.0001 1452864_at Med12l 6.5 0.0003 1440879_at Abca9 6.3 0.0376 1417829_a_at Rab15 6.3 0.0342 1423854_a_at Rasl11b 6.3 0.0070 1419144_at Cd163 6.2 0.0134 1450639_at RP23-357I14.1 /// Slc28a2 6.1 0.0052 1449106_at Gpx3 5.4 0.0262 1426526_s_at Ovgp1 5.3 0.0078 1422869_at Mertk 5.3 0.0203 1428662_a_at Hod 5.1 0.0022 1438377_x_at Slc13a3 4.9 0.0087 1448830_at Dusp1 4.8 0.0305 1422851_at Hmga2 4.7 0.0134 1423467_at Ms4a4b 4.7 0.0008 1426642_at Fn1 4.7 0.0168 1417995_at Ptpn22 4.7 0.0200 1418216_at Ggtla1 4.6 0.0447 1434366_x_at C1qb 4.5 0.0147 1434207_at Rnf169 4.4 0.0026 1455345_at Phf15 4.3 0.0038 1417529_at Rab33a 4.3 0.0066 1426235_a_at Glul 4.3 0.0028 1435280_at Pde4d 4.2 0.0102 1416926_at Trp53inp1 4.2 0.0172 1421977_at Mmp19 4.2 0.0005 1424256_at Rdh12 4.2 0.0326 1430700_a_at Pla2g7 4.1 0.0275 1420150_at Spsb1 4.0 0.0006 1447903_x_at Ap1s2 4.0 0.0002 1422824_s_at Eps8 4.0 0.0091 1434414_at Foxred2 4.0 0.0044 1441761_at Sla 3.9 0.0017 1449851_at Per1 3.9 0.0213 1431609_a_at Acp5 3.9 0.0012 1437932_a_at Cldn1 3.9 0.0159 1455826_a_at Bace1 3.9 0.0256 1437176_at Nlrc5 3.8 0.0037 1448489_at Pafah2 3.8 0.0194 1434372_at AW112010 3.8 0.0004 1442358_at AA409587 3.7 0.0088 1419538_at Flt3 3.7 0.0284 1437886_at Klhl6 3.7 0.0021 1416666_at Serpine2 3.7 0.0413 1422754_at Tmod1 3.6 0.0141 1437136_at 5830436I19Rik 3.6 0.0029 1448881_at Hp 3.6 0.0208 1425107_a_at Lifr 3.5 0.0348 1448355_at Prss16 3.5 0.0002 1456393_at 2310002J21Rik 3.4 0.0004 1449901_a_at Map3k6 3.4 0.0151 1425451_s_at Chi3l4 3.4 0.0004 1448470_at Fbp1 3.4 0.0068 1420089_at Nfkbia 3.3 0.0141 1420764_at Scrg1 3.2 0.0092 1418840_at Pdcd4 3.2 0.0132 1448842_at Cdo1 3.2 0.0205 1449405_at Tns1 3.1 0.0459 1441823_at Zmiz1 3.1 0.0175 1438805_at Ccnd3 3.1 0.0013 1429055_at 4930506M07Rik 3.0 0.0085 1426223_at 2810439F02Rik 3.0 0.0016 1428622_at LOC100505173 /// Depdc6 3.0 0.0087 1425888_at Klra17 3.0 0.0072 1418069_at Apoc2 2.9 0.0219 1460616_at Slco4c1 2.8 0.0307 1438169_a_at Frmd4b 2.8 0.0330 1438511_a_at 1190002H23Rik 2.8 0.0304 1418992_at F10 2.8 0.0263 1419481_at Sell 2.8 0.0376 1427974_s_at Cacna1d 2.8 0.0083 1421392_a_at Birc3 2.7 0.0282 1440451_at D730048J04Rik 2.7 0.0483 1449366_at Mmp8 2.7 0.0161 1422471_at Pex13 2.6 0.0001 1436499_at Sgms1 2.6 0.0333 1433743_at Dach1 2.6 0.0140 1451263_a_at Fabp4 2.6 0.0411 1434070_at Jag1 2.5 0.0454 1454727_at Afap1l1 2.5 0.0001 1427994_at Cd300lf 2.5 0.0435 1439956_at Ms4a6b 2.5 0.0384 1419599_s_at Ms4a6d 2.5 0.0278 1429194_at Tigd2 2.4 0.0105 1424414_at Ogfrl1 2.4 0.0071 1432543_a_at Klf13 2.4 0.0026 1436617_at Cetn4 2.4 0.0350 1437304_at Cblb 2.4 0.0064 1419764_at Chi3l3 2.4 0.0012 1422010_at Tlr7 2.4 0.0040 1434548_at Serinc3 2.4 0.0079 1438971_x_at Ube2h 2.3 0.0133 1434378_a_at Mxd4 2.3 0.0137 1448647_at Man2a1 2.3 0.0163 1451584_at Havcr2 2.3 0.0251 1434046_at AA467197 2.3 0.0023 1455002_at LOC100044742 /// Ptp4a1 2.3 0.0082 1420973_at Arid5b 2.3 0.0075 1455222_a_at Ubp1 2.3 0.0322 1433488_x_at Gns 2.2 0.0059 1450234_at Ms4a6c 2.2 0.0204 1460337_at Sh3kbp1 2.2 0.0069 1450350_a_at Jundm2 2.2 0.0303 1426979_at Mlxip 2.2 0.0145 1429642_at Anubl1 2.2 0.0355 1424265_at Npl 2.2 0.0199 1450199_a_at Stab1 2.2 0.0387 1448929_at F13a1 2.1 0.0019 1418435_at Mkrn1 2.1 0.0178 1456812_at Abcd2 2.1 0.0098 1428143_a_at Pnpla2 2.1 0.0123 1438928_x_at Ninj1 2.1 0.0302 1422526_at Acsl1 2.1 0.0017 1450291_s_at Ms4a4c 2.1 0.0004 1451196_at Ypel5 2.1 0.0218 1418932_at Nfil3 2.1 0.0135 1449954_at Hyal1 /// Nat6 2.1 0.0476 1435536_at Ddi2 2.0 0.0395 1418553_at Arhgef18 2.0 0.0370 1448810_at Gne 2.0 0.0268 1452309_at Cgnl1 2.0 0.0030 1417290_at Lrg1 2.0 0.0007 1435171_at 2810416G20Rik 2.0 0.0086 1424917_a_at Wipi1 2.0 0.0282 Supplemental Table 2: Down-regulated genes Affymetrix identifier Gene symbol N-fold P-Value 1436039_at Cmah -14.38 0.0400 1427381_at Irg1 -11.38 0.0100 1427038_at Penk1 -11.19 0.0500 1439036_a_at Atp1b1 -10.95 0.0135 1427419_x_at Ccr9 -10.76 0.0221 1451721_a_at H2-Ab1 -10.43 0.0243 1416111_at Cd83 -9.95 0.0176 1426112_a_at Cd72 -8.15 0.0331 1452815_at P2ry10 -6.39 0.0001 1449025_at Ifit3 -6.21 0.0011 1449195_s_at Cxcl16 -6.10 0.0375 1422891_at H2-Ea -6.02 0.0057 1452431_s_at H2-Aa -6.01 0.0265 1455090_at Angptl2 -5.73 0.0472 1426622_a_at Qpct -5.70 0.0274 1451780_at Blnk -5.38 0.0496 1419709_at Stfa3 -5.38 0.0079 1422139_at Plau -5.08 0.0116 1417025_at H2-Eb1 -5.07 0.0291 1419307_at Tnfrsf13c -4.84 0.0178 1427329_a_at Igh-6 -4.60 0.0258 1425519_a_at Cd74 -4.14 0.0454 1436058_at Rsad2 -4.06 0.0035 1436329_at Egr3 -4.01 0.0396 1417420_at Ccnd1 -3.93 0.0300 1450276_a_at Scin -3.85 0.0061 1454966_at Itga8 -3.83 0.0017 1424305_at Igj -3.76 0.0082 1421596_s_at H28 -3.72 0.0447 1437401_at Igf1 -3.68 0.0049 1451798_at Il1rn -3.67 0.0019 1427455_x_at Igkj1 /// Igk-C /// Igk-V28 /// Igkv8-30 /// Igkv4-53 /// Igkv6-23 -3.67 0.0112 1420699_at Clec7a -3.60 0.0060 1418293_at Ifit2 -3.58 0.0016 1423555_a_at Ifi44 -3.57 0.0068 1455796_x_at Olfm1 -3.55 0.0112 1418191_at Usp18 -3.53 0.0171 1418998_at Kmo -3.44 0.0070 1449249_at Pcdh7 -3.43 0.0142 1453196_a_at Oasl2 -3.37 0.0402 1418497_at Fgf13 -3.34 0.0228 1450783_at Ifit1 -3.33 0.0011 1434920_a_at Evl -3.32 0.0490 1417930_at Nab2 -3.32 0.0256 1416410_at Pafah1b3 -3.22 0.0130 1459923_at Bex6 -3.19 0.0002 1423756_s_at Igfbp4 -3.13 0.0019 1419128_at Itgax -3.07 0.0037 1448162_at Vcam1 -3.04 0.0377 1450871_a_at Bcat1 -2.92 0.0307 1455452_x_at AI449310 -2.89 0.0188 1437025_at Cd28 -2.85 0.0177 1455106_a_at Ckb -2.83 0.0135 1424011_at Aqp9 -2.64 0.0172 1454632_at 6330442E10Rik -2.64 0.0331 1423586_at Axl -2.59 0.0343 1449580_s_at H2-DMb1 /// H2-DMb2 -2.59 0.0230 1431591_s_at Isg15 /// Gm9706 -2.57 0.0028 1421027_a_at Mef2c -2.56 0.0031 1450753_at Nkg7 -2.52 0.0291 1428547_at Nt5e -2.52 0.0027 1437939_s_at Ctsc -2.52 0.0073 1435761_at BC100530 /// Stfa1 -2.47 0.0077 1417074_at Ceacam10 -2.41 0.0056 1434944_at Dmpk -2.41 0.0288 1421653_a_at Igh-2 /// Igh-VJ558 -2.39 0.0177 1427102_at Slfn4 -2.36 0.0459 1418580_at Rtp4 -2.36 0.0034 1422873_at Prg2 -2.35 0.0200 1417856_at Relb -2.33 0.0180 1454788_at Arl4c -2.29 0.0289 1418326_at Slc7a5 -2.26 0.0337 1427327_at Pilra -2.25 0.0132 1426725_s_at Ets1 -2.25 0.0156 1458802_at Hivep3 -2.24 0.0152 1418401_a_at Dusp16 -2.23 0.0143 1423141_at Lipa -2.23 0.0216 1416258_at Tk1 -2.22 0.0490 1415904_at Lpl -2.21 0.0101 1416239_at Ass1 -2.20 0.0049 1422411_s_at Ear1 /// Ear12 /// Ear2 /// Ear3 -2.17 0.0202 1424254_at Ifitm1 -2.12 0.0149 1454086_a_at Lmo2 -2.11 0.0250 1452912_at 2600005O03Rik -2.10 0.0336 1430080_at Fcnb -2.10 0.0144 1424208_at Ptger4 -2.10 0.0117 1422474_at Pde4b -2.08 0.0318 1430127_a_at Ccnd2 -2.07 0.0186 1429189_at Arsb -2.06 0.0414 1425374_at Oas3 -2.06 0.0267 1455161_at AI504432 -2.06 0.0150 1417244_a_at Irf7 -2.03 0.0010 1424339_at Oasl1 -2.01 0.0383 1421186_at Ccr2 -2.00 0.0222 1418219_at Il15 -1.99 0.0280 1430466_at 2310005G13Rik -1.97 0.0228 1417928_at Pdlim4 -1.96 0.0021 1442339_at Stfa2l1 -1.96 0.0030 Supplemental Table 3: Functional gene clusters Overrepresented among genes up-regulated by 16h GC treatment Z score P-Value % of selection % of all Molecular function chitinase activity 11.89 <0.001 2.03 0.08 protease binding 5.93 <0.001 2.70 0.48 pattern binding 5.31 <0.001 4.73 1.52 scavenger receptor activity 4.60 0.001 2.03 0.43 drug binding 3.03 0.015 2.03 0.79 carboxylic ester hydrolase activity 2.65 0.018 2.03 0.93 Biological process aminoglycan catabolic process 13.96 <0.001 3.38 0.16 inflammatory response 6.41 <0.001 6.76 2.15 negative regulation of cell activation 5.53 <0.001 3.38 0.80 sulfur compound metabolic process 3.98 <0.001 3.38 1.28 response to glucocorticoid stimulus 3.89 <0.001 3.38 1.32 response to cAMP 3.55 0.007 2.03 0.63 negative regulation of protein kinase activity 3.36 0.011 2.70 1.11 coagulation 3.34 0.007 2.03 0.69 signal transduction 3.15 0.003 25.00 29.31 response to hydrogen peroxide 3.15 0.011 2.03 0.75 protein complex subunit organization 2.96 0.003 6.08 4.45 regulation of cell proliferation 2.89 0.003 9.46 8.47 regulation of myeloid cell differentiation 2.65 0.033 2.03 0.93 angiogenesis 2.62 0.023 2.70 1.48 cellular response to cytokine stimulus 2.59 0.020 2.70 1.50 anti-apoptosis 2.49 0.035 2.70 1.57 activation of immune response 2.37 0.031 2.70 1.65 negative regulation of response to stimulus 2.31 0.017 5.41 4.62 leukocyte activation 2.25 0.028 3.38 2.41 antioxidative/oxidative* 22.10 <0.001 14.76 0.62 tolerance induction* 3.72 <0.001 4.03 1.00 IFN-γ induced* 3.60 <0.001 4.03 1.04 anti-inflammatory* 3.03 0.004 4.03 1,26 Overrepresented among genes down-regulated by 16h GC treatment Z score P-Value % of selection % of all Molecular function antigen binding 13.64 <0.001 5.10 0.26 cysteine-type endopeptidase inhibitor activity 6.65 <0.001 3.06 0.37 carboxylic ester hydrolase activity 3.81 0.005 3.06 0.92 receptor activity 3.48 <0.001 14.29 11.96 transferase activity, transferring phosphorus-containing groups 2.35 0.042 9.18 8.69 Biological process antigen processing and presentation of exogenous peptide antigen via MHC class II 19.17 <0.001 5.10 0.14 immune response 14.07 <0.001 18.37 2.89 positive regulation of T cell activation 12.78 <0.001 9.18 0.92 positive regulation of lymphocyte proliferation 10.49 <0.001 7.14 0.81 T cell selection 9.20 <0.001 2.73 0.09 regulation of T cell differentiation 8.62 <0.001 4.55 0.28 response to virus 8.61 <0.001 6.12 0.86 regulation of interleukin-2-produktion 7.45 <0.001 2.73 0.14 regulation of adaptive immune response 7.42 <0.001 4.55 0.36 lymphocyte activation 6.21 <0.001 7.14 1.96 regulation of MAPKKK cascade 6.09 <0.001 6.36 0.93 response to chemical stimulus 5.54 <0.001 21.43 14.21 regulation of myeloid cell differentiation 5.52 0.001 4.08 0.86 homeostasis of number of cells 4.60 0.004 2.73 0.32 positive regulation of cytokine production 4.47 0.003 4.08 1.20 anti-apoptosis 3.70 0.010 4.08 1.57 regulation of inflammatory response 3.29 0.015 3.06 1.13 positive regulation of organelle organization 3.29 0.030 3.06 1.13 locomotion 2.81 0.015 6.36 2.41 cell surface receptor linked signaling pathway 2.65 0.008 11.82 6.08 regulation of vesicle-mediated transport 2.54 0.043 2.73 0.73 response to external stimulus 2.48 0.018 5.45 2.15 IFN-γ induced* 20.58 <0.001 20.90 1.04 antioxidative/oxidative* 15.02 <0.001 11.82 0.62 anti-inflammatory* 5.66 <0.001 7.27 1.26 tolerance induction* 2.78 0.007 3.64 1.00 * Since only a small percentage of information available in the scientific literature is annotated in the GO classification we additionally performed a detailed literature search for regulated genes and performed statistical testing of identified functional gene clusters using information available in the NCBI “Gene” database as described in detail previously (8).