Glucocorticoid-Induced Tumor Necrosis Factor Receptor Family ...

1521-0103/347/1/164–172$25.00 THE JOURNAL OF PHARMACOLOGY AND EXPERIMENTAL THERAPEUTICS Copyright ª 2013 by The American Society for Pharmacology and Experimental Therapeutics

http://dx.doi.org/10.1124/jpet.113.207605 J Pharmacol Exp Ther 347:164–172, October 2013

Glucocorticoid-Induced Tumor Necrosis Factor Receptor FamilyRelated Ligand Triggering Upregulates Vascular Cell Adhesion Molecule-1 and Intercellular Adhesion Molecule-1 and Promotes Leukocyte Adhesion

Laboratory of Molecular Oncology, “Istituto Dermopatico dell’Immacolata-IRCCS,” Rome, Italy (P.M.L., F.R.); Department of Medicine, University of Perugia, Perugia, Italy (M.G.P., R.B., S.R., G.M., C.R., G.N.); and Department of System Medicine, University of Rome “Tor Vergata,” Rome, Italy (A.M., G.G.) Received June 29, 2013; accepted July 24, 2013

ABSTRACT The interaction of glucocorticoid-induced tumor necrosis factor receptor-family related (GITR) protein with its ligand (GITRL) modulates different functions, including immune/inflammatory response. These effects are consequent to intracellular signals activated by both GITR and GITRL. Previous results have suggested that lack of GITR expression in GITR2/2 mice decreases the number of leukocytes within inflamed tissues. We performed experiments to analyze whether the GITRL/GITR system modulates leukocyte adhesion and extravasation. For that purpose, we first evaluated the capability of murine splenocytes to adhere to endothelial cells (EC). Our results indicated that adhesion of GITR2/2 splenocytes to EC was reduced as compared with wild-type cells, suggesting that GITR plays a role in adhesion and that this effect may be due to GITRL-GITR interaction. Moreover, adhesion was increased when EC were pretreated with an agonist GITR-Fc fusion protein,

Introduction Glucocorticoid-induced tumor necrosis factor (TNF) receptor family-related (GITR, also named TNFRSF18) protein, is a member of the TNF receptor superfamily that interacts with its specific ligand (GITRL, also named TNFSF18). After interaction, both GITRL and GITR are reciprocally stimulated and This work was supported by a research grant from the Italian Association for Cancer Research (AIRC) in Milan, Italy, and by the Italian Ministry of Health. G.G. and G.N. contributed equally to this work (joint senior investigators). 1 Current affiliation: Clinical Research Center Salzburg GmbH (CRCS), Landeskrankenhaus Salzburg (SALK) and Universitätsklinikum der Paracelsus Medizinischen Privatuniversität (PMU), Salzburg, Austria. dx.doi.org/10.1124/jpet.113.207605.

thus indicating that triggering of GITRL plays a role in adhesion by EC regulation. In a human in vitro model, the adhesion to human EC of HL-60 cells differentiated toward the monocytic lineage was increased by EC pretreatment with agonist GITRFc. Conversely, antagonistic anti-GITR and anti-GITRL Ab decreased adhesion, thus further indicating that GITRL triggering increases the EC capability to support leukocyte adhesion. EC treatment with GITR-Fc favored extravasation, as demonstrated by a transmigration assay. Notably, GITRL triggering increased intercellular adhesion molecule-1 (ICAM-1) and vascular cell adhesion molecule-1 (VCAM-1) expression and anti–ICAM-1 and anti-VCAM-1 Abs reversed GITR-Fc effects. Our study demonstrates that GITRL triggering in EC increases leukocyte adhesion and transmigration, suggesting new antiinflammatory therapeutic approaches based on inhibition of GITRL-GITR interaction.

activate intracellular signals regulating immune functions (Nocentini and Riccardi, 2009; Azuma, 2010; Schaer et al., 2012). In particular, GITR-driven T-cell costimulation was found to be the main mechanism by which the GITRL-GITR system contributes to tumor rejection and the development of autoimmune/inflammatory diseases (Kohm et al., 2004; Cuzzocrea et al., 2005; Lee et al., 2006; Morris and Kong, 2006; Santucci et al., 2007; Ronchetti et al., 2012; Schaer et al., 2012). This was also demonstrated by the mild inflammatory response of GITR knockout (GITR2/2) mice (Nocentini and Riccardi, 2009). Despite the likelihood that GITR-triggered T cells are involved in exacerbation of chronic inflammatory responses (Cuzzocrea et al., 2007; Santucci et al., 2007; Liao et al., 2012),

ABBREVIATIONS: BSA, bovine serum albumin; calcein-AM, calcein acetoxymethylester; DC, dendritic cells; EBM-2, endothelial basal medium-2; EC, endothelial cells; EGM-2, endothelial growth factor medium; FBS, fetal bovine serum; GITR, glucocorticoid-induced TNF receptor familyrelated; GITR2/2, GITR knockout; GITR-Fc, GITR-Fc fusion protein; GITRL, GITR ligand; ICAM-1, intercellular adhesion molecule-1; LPS, lipopolysaccharide; NF-kB, nuclear factor k-light-chain-enhancer of activated B cells; PCR, polymerase chain reaction (PCR); PMA, phorbol 12-myristate 13-acetate; STAT, signal transducers and activators of transcription; TNF, tumor necrosis factor; VCAM-1, vascular cell adhesion molecule-1; WT, wild-type. 164

Downloaded from jpet.aspetjournals.org at Harvard University, Museum of Comparative Zoology on December 10, 2013

Pedro Miguel Lacal, Maria Grazia Petrillo, Federica Ruffini, Alessia Muzi, Rodolfo Bianchini,1 Simona Ronchetti, Graziella Migliorati, Carlo Riccardi, Grazia Graziani, and Giuseppe Nocentini

GITRL/GITR System and Leukocyte Adhesion

the development of acute inflammation (Cuzzocrea et al., 2004, 2006; Galuppo et al., 2011a,b) is very likely favored by GITR activation in cells other than T cells. Indeed, GITR is expressed in B cells and in cells of the innate immune system (Krausz et al., 2007; Azuma, 2010; Ronchetti et al., 2011). Interestingly, in the pleurisy model, we detected differences in the number of proinflammatory cells in the pleura of GITR2/2 compared with wild-type (WT) mice soon after the proinflammatory stimulus. The number of all leukocyte subsets was reduced, suggesting that the extravasation process is affected in GITR2/2 mice (Cuzzocrea et al., 2006). Of note, during the inflammatory response, the expression level of adhesion molecules was reduced in GITR2/2 as compared with WT mice (Cuzzocrea et al., 2004, 2006, 2007; Galuppo et al., 2011b). Other studies demonstrated that GITR triggering in T cells upregulates adhesion molecules (Mahesh et al., 2006) and favors migration of cutaneous dendritic cells (DC) from the skin to the draining lymph nodes (Kamimura et al., 2009). GITRL is expressed in DC, monocytes/macrophages, and other cells of the immune system where it activates nuclear factor k-light-chain-enhancer of activated B cells (NF-kB) and extracellular signal-regulated kinase (ERK) and increases expression of chemoattractants such as interleukin-8 (IL-8) and monocyte chemotactic protein-1 (Grohmann et al., 2007; Bae et al., 2008; Baessler et al., 2009; Byrne et al., 2009; Nocentini and Riccardi, 2009; Azuma, 2010). Therefore, it stands to reason that the phenotypic differences of GITR2/2 mice may also be due to the lack of GITR interaction with GITRL and the consequent lack of GITRL signaling. It is noteworthy that GITRL is expressed at high levels in endothelial cells (EC), particularly when activated (Gurney et al., 1999; Nardelli et al., 2006). For this reason, it has been hypothesized to be an adhesion molecule (Gurney et al., 1999; Kwon et al., 1999; Krausz et al., 2007), and a direct role of the GITRL-GITR system in the regulation of extravasation cannot be ruled out. However, to date the issue has yet to be investigated. We examined the role of the GITRL-GITR system in leukocyte adhesion to EC and extravasation. We demonstrate that GITRL triggering by an agonist GITR-Fc fusion protein increases signal transducers and activators of transcription 1 (STAT1) phosphorylation, vascular cell adhesion molecule-1 (VCAM-1) and intercellular adhesion molecule-1 (ICAM-1) expression as well as leukocyte adhesion. In contrast, GITRL inhibition by antagonist anti-GITRL and anti-GITR Abs decreases adhesion and extravasation. Our results suggest that GITRL in EC is a regulator of the inflammatory process and that therapeutic approaches based on inhibition of GITR interaction with GITRL may represent a new line of investigation for the treatment of inflammatory diseases.

Materials and Methods Animals. Male mice with a targeted disruption of the GITR gene (GITR2/2) and corresponding WT mice were obtained on a Sv129 background as previously described elsewhere (Ronchetti et al., 2004). Animal care was in compliance with regulations in Italy (D.M. 116192), Europe (O.J. of E.U. L276/33 20/10/2010), and the United States (Animal Welfare Assurance No. A5594-01; U.S. Department of Health and Human Services). Cell Lines and Cell Culture Reagents. The murine EC line HYKO6P (a gift from Dr. J. Yélamos, IMIM-Hospital del Mar,

165

Barcelona, Spain) was maintained in 10% fetal bovine serum (FBS)/endothelial basal medium-2 (EBM-2). Splenocytes were isolated from the spleen of WT and GITR2/2 mice and cultured in RPMI 1640 medium supplemented with 10% FBS, 100 mg/ml streptomycin, 10 mM HEPES, 0.1% nonessential amino acids, 1 mM sodium pyruvate, and 50 mM 2-mercaptoethanol. The immortalized human EC line HUV-ST was maintained in culture in EGM-2 supplemented with 0.4 mg/ml Geneticin (gentamycin) and 5 mg/ml puromycin (Tentori et al., 2005). HL-60 cells (American Type Culture Collection [ATCC], Manassas, VA) were cultured in 10% FBS/RPMI 1640 medium. EBM-2 and EGM-2 were obtained from Clonetics (BioWhittaker, Walkersville, MD), FBS was from Euroclone (Milan, Italy), and RPMI 1640 medium and other cell culture reagents were from Lonza (Verviers, Belgium). The fatty acid-free bovine serum albumin (BSA) and gentamycin were from Invitrogen (Groningen, The Netherlands); puromycin, heparin, phorbol 12-myristate 13-acetate (PMA), lipopolysaccharide (LPS), and other chemicals were from Sigma-Aldrich (St. Louis, MO). The recombinant human TNF-a was from ImmunoTools (Friesoythe, Germany). Cell Adhesion. The EC monolayers were prepared by seeding in 96-well plates 5 104 HYKO6P or HUV-ST cells/well in 100 ml of 0.1% FBS/EBM-2. After overnight incubation at 37°C, cells were activated with TNF-a (100 ng/ml, 18 hours, 37°C), washed with prewarmed medium, and tested. The HL-60 cells, differentiated toward the monocytic lineage (10 ng/ml PMA, 24 hours, 37°C), and murine splenocytes (activated with 50 mg/ml LPS for 72 hours at 37°C) were resuspended in complete medium at 2 106 cells/ml, and were labeled and protected from light with calcein acetoxymethylester (calcein-AM; Molecular Probes/Invitrogen, Monza, Italy) (1 mM, 15 minutes at 37°C). The cells were then washed three times and resuspended at 2 106 cells/ml in 0.5% BSA/RPMI 1640 medium. Before adhesion, EC and/or leukocytes were pretreated as indicated in the text with recombinant murine GITR-human Fc, human GITRhuman Fc, or Thy-1-human Fc proteins (all from Alexis Biochemicals, Postfach, Switzerland), and/or with anti-human GITR Ab (clone 110416; R&D Systems, Minneapolis, MN), anti-human GITRL (clone 109114; R&D Systems), anti-human ICAM-1 (clone 1H4; ImmunoTools), anti-human VCAM-1 (clone STA; ImmunoTools), or control Abs (mouse IgG1 control, clone 11711; R&D Systems; or mouse antihuman CD19 Ab, clone LT19; Serotec, Kidlington, UK). When we tested the effect of anti-human GITR Ab in comparison with control Ab, we added Fc protein (2 mg/ml) to differentiated HL-60 cells soon before Ab to saturate the Fc receptors and inhibit binding of antiGITR and control Ab to the Fc receptor. Adhesion of leukocytes to EC was tested by placing 50 ml of calceinAM-labeled cells into the appropriate wells of the 96-well plate and allowing them to adsorb to the EC monolayer for 30 minutes at 37°C. Afterward wells were washed three times with prewarmed medium, and remaining cells were lysed with 100 ml of 1% Triton X-100/PBS. Cell extracts were transferred to 96-well Opti-Plates (PerkinElmer, Zaventem, Belgium) and fluorescence measured in a LS 55 Luminescence Spectrometer (PerkinElmer) set at absorption/emission of 494/517 nm. Transmigration Assay. HUV-ST cells (2 105 in 300 ml of complete medium) were plated in gelatin-coated (20 mg/ml) 24-well transwell inserts with 8 mm diameter pore (Becton Dickinson, San Jose, CA), and allowed to form a cell monolayer at 37°C. Cells were starved in 0.1% FBS/EBM-2 for 6 hours and then activated with TNF-a (100 ng/ml, 18 hours, 37°C). Cell monolayer was then washed with prewarmed medium and treated with GITR-Fc or control-Fc (4 mg/ml, 30 minutes, RT) in 200 ml of transmigration medium (0.5% BSA, 2 mM CaCl2, 2 mM MgCl2 RPMI 1640 medium). HL-60 cells (5 105 in 200 ml of transmigration medium), differentiated with PMA and labeled with calcein-AM, were loaded on EC monolayer, and the plates containing Transwell inserts were incubated at 37°C for 6 hours. We transferred 400 ml of the bottom medium comprising migrated

166

Lacal et al.

cells to a clean well, filled with 100 ml of 5% Triton X-100/PBS, incubated for 10 minutes at room temperature with shaking and 200 ml of the mixture transferred to 96-well Opti-Plates (PerkinElmer, Zaventem, Belgium) to measure fluorescence as indicated above. Cell number in the samples was determined using a reference curve fluorescence versus number of HL-60–labeled cells. Real-Time Polymerase Chain Reaction. RNA was purified and cDNA was prepared as previously described elsewhere (Rinaldi et al., 2011). The analysis of GITR (Hs00188345_m1) and GITRL (Hs00183225_m1) expression was performed as previously described elsewhere (Bianchini et al., 2006) using a FAM-labeled TaqMan probe (investigated gene); a VIC-labeled TaqMan probe (HPRT-1) (Applied Biosystems, Foster City, CA) was used as the housekeeping gene for data normalization. For each experimental group, amplification was performed in quadruplicate. Flow Cytometry. Analysis of cell surface molecule expression was performed using standard procedures. For murine cell staining, the following Abs were used: anti-mouse GITR-PE (clone DTA-1;

Biolegend, San Diego, CA), anti-mouse GITRL-PE (clone YGL386; eBioscience, Hatfield, UK), anti-mouse B220-FITC (clone RA3-6B2; Miltenyi Biotech, Auburn, CA), anti-mouse GR-1-FITC (clone RB68C5; BD Pharmingen, Erembodegem, Belgium), anti-mouse CD3FITC (clone 17A2; BD Pharmingen). For human cell staining, HUV-ST cells were detached from the culture flasks by gentle trypsin digestion and allowed to recover in complete medium for 2 hours at room temperature in a rotating wheel in the presence of 100 ng/ml TNF-a. Differentiated HL-60 cells were gently scraped from the culture flask and directly stained with Abs. For each sample (0.5 106 cells) 2.5 mg of an unlabeled anti-human GITRL mouse Ab (clone 109114; R&D Systems), a secondary goat anti-mouse IgG (Fc specific)-FITC Ab (1:100; Sigma-Aldrich), or 5 ml of a PE-conjugated anti-human CD357 (GITR) (clone 621; Biolegend, Aachen, Germany) were used. In other experiments, anti-human GITRL-FITC (clone 10910; R&D Systems) was used. GITRL detection in PMA-treated HL-60 cells required the presence of protease inhibitors in the culture medium during differentiation. For staining of

Fig. 1. GITRL and GITR expression in LPStreated splenocytes and murine EC. (A–C) LPStreated splenocytes from WT mice were stained with anti-B220/anti-GITRL, anti-B220/anti-GITR, anti-CD3/anti-GITRL, anti-CD3/anti-GITR, antiGR-1/anti-GITRL, anti-GR-1/anti-GITR Abs, and the relative control Abs and were evaluated by flow cytometry. The percentage of cells positive for each antigen is reported above the marker (A–C), and the mean fluorescence intensity of control stained and specific Ab stained cells is also shown (B and C). (A) B220+ (B cells), CD3+ (T cells), and GR-1+ (granulocytes) cells in LPS-treated cells. (B) GITRL+ cells within the B220+, CD3+, and GR-1+ cell subsets. (C) GITR+ cells within the B220+, CD3+, and GR-1+ cell subsets. One representative experiment out of three is shown. (D–E) GITRL and GITR mRNA expression in the endothelial cell line HYKO6P, untreated or treated with TNF-a, evaluated by real-time PCR. Values are the mean (6 S.D.) of the relative GITRL expression respect to that of macrophages (evaluated in the same experiments) and the relative GITR expression respect to that of CD4+ regulatory T cells (evaluated in the same experiments), arbitrarily set equal to 100. ***P , 0.001 according to Student’s t test comparing GITRL mRNA of TNF-a treated cells to untreated control.


167

adhesion molecules, anti-human ICAM-1-FITC (clone 1H4; ImmunoTools) and anti-human VCAM-1-FITC (clone STA; ImmunoTools) were used. All isotype controls were purchased from Dako (Glostrup, Denmark). Analysis of Protein Phosphorylation. EC monolayers were prepared by seeding in 96-well plates 5 104 HUV-ST cells/well in 100 ml of 0.1% FBS/EBM-2. After overnight incubation at 37°C, cells were activated with TNF-a (100 ng/ml, 18 hours, 37°C), washed with prewarmed medium, and treated with control-Fc or GITR-Fc (4 mg/ml in 0.5% BSA/PhosStop/RPMI 1640 medium). Samples were run in an 8% SDS-polyacrylamide gel, Western blot analysis was performed using anti-phospho-STAT1, and the goat anti–phospho-STAT1 (sc-7988) (Tyr701, 1:200) from Santa Cruz Biotechnology (Santa Cruz, CA) and anti–b-tubulin (H-286, 1:500) were used as as loading control. The phosphatase inhibitors PhosStop and the protease inhibitors cocktails from Roche (Mannheim, Germany) were used during the procedure (Lacal et al., 2009). Statistics. For each experimental group, adhesion was performed six times. Data are reported as the mean 6 S.D. of at least three independent experiments. Statistical significance was evaluated according to Student’s t test: *P , 0.05; **P , 0.01; ***P , 0.001.

Results GITRL/GITR System Favors Adhesion of Murine Splenocytes to EC. To evaluate the effect of the GITRLGITR system on cell adhesion, we first set up an in vitro assay in which labeled LPS-activated splenocytes were seeded on a monolayer of murine activated EC. LPS-activated WT splenocytes expressed both GITRL and GITR whereas activated EC displayed high GITRL but very low GITR expression levels (Fig. 1, D and E). When we evaluated the number of adherent cells by a fluorometric method, we found (Fig. 2A) that activated splenocytes of WT mice adhered to murine EC better than the activated splenocytes of GITR2/2 mice, suggesting that GITRL-GITR interaction favors the adhesion process. We then used an agonist GITR-Fc fusion protein to trigger GITRL in EC. The results, as shown in Fig. 2B, were that GITR-Fc pretreatment of murine EC statistically significantly enhanced the adhesion of activated GITR2/2 splenocytes as compared with control-Fc in a dose-dependent manner (P , 0.01), demonstrating that GITRL triggering in EC increases leukocyte adhesion to EC. GITRL and GITR Are Expressed in Human Cells That Participate in the Cell Adhesion Process. Previous studies had shown that mouse and human GITRL and GITR have different tridimensional structures and different functions (Chattopadhyay et al., 2009; Nocentini et al., 2012). We investigated whether GITRL plays a similar role in humans and in mice. For that purpose, we used a human promyelocytic leukemia cell line (HL-60), differentiated in vitro toward the monocytic lineage by treatment with PMA and labeled with calcein-AM. HL-60 cells were cocultured with immortalized human EC (HUV-ST cells), and the number of adherent cells was evaluated after washing out nonadherent cells by a fluorometric method. As shown in Fig. 3A, differentiated HL-60 cells adhered only to activated EC. Therefore, studies were performed using PMA differentiated HL-60 cells and TNF-a activated HUV-ST cells. Interestingly, TNF-a treatment markedly upregulated GITRL in HUV-ST cells, as shown by real-time polymerase chain reaction (PCR) and flow cytometry analysis (Fig. 3, B

Fig. 2. The GITRL-GITR system favors adhesion of murine splenocytes to EC. (A) Adhesion of LPS-activated calcein-AM-labeled WT and GITR2/2 splenocytes to murine EC line HYKO6P after 30 minutes of contact, as evaluated by measuring fluorescence of cell lysates after washing out nonadherent cells. Values are the mean of six samples 6 S.D. of four independent experiments. (B) Increased adhesion of activated/labeled GITR2/2 splenocytes to EC pretreated with GITR-Fc fusion protein (2 mg/ml, 15 minutes) shown on a logarithmic scale (2 basis). The fusion protein was maintained in cell coculture (30 minutes). Adhesion was evaluated as previously specified, and adhesion of Fc-treated cells was used as the control. Results are the mean 6 S.D. of four independent experiments in which each experimental group was evaluated six times. **P , 0.01, according to Student’s t test comparing adhesion of GITR2/2 splenocytes with WT cells (A), GITR-Fc-treated to Fc-treated cells (B).

and D). In addition, GITR was upregulated in TNF-a– activated HUV-ST cells (Fig. 3C), although at very low level (~30 times lower than in Treg cells) so that it was undetectable by flow cytometry analysis (Fig. 3D). On the other hand, both GITR and GITRL were upregulated in PMAdifferentiated HL-60 cells (Fig. 3, B–D). GITRL Triggering Favors Cell Adhesion and Increases Transmigration. The expression of GITRL in activated EC and that of GITR and GITRL in differentiated HL-60 cells suggest that the GITRL-GITR system is involved in up-modulation of human cell adhesion. Pretreatment of HUV-ST cells with agonist GITR-Fc favored the adhesion of HL-60 cells to human EC, whereas pretreatment of HL-60 cells with GITR-Fc had no effect on cell adhesion (Fig. 4A). The effect of HUV-ST pretreatment was dose-dependent, being irrelevant at 0.5 mg/ml and significant at 1 and 2 mg/ml (Fig. 4A). At the highest concentration, adhesion of HL-60 cells to GITR-Fc-treated EC increased, and difference was statistically significant (60% increase, P , 0.01), compared with adhesion of control-Fc–treated cells. Results suggest that similar to the mouse system the GITR-Fc effect on cell adhesion is attributed to GITRL activation in EC cells. To confirm this conclusion, EC or HL-60 cells were pretreated with antagonistic anti-GITRL Abs (Satoguina et al., 2008; Kamimura et al., 2009) (referred to as pretreatment 1 in Fig. 4B) before treatment with GITR-Fc or control Fc (referred to as pretreatment 2 in Fig. 4B).

168

Lacal et al.

Fig. 3. Adhesion of human myelomonocytic cells to EC is associated with GITRL and GITR expression. (A) Adhesion of calcein-AM–labeled HL-60 cells (differentiated or not with PMA) to human EC (HUV-ST) (activated or not with TNF-a) after 30 minutes of coculture, as evaluated by measuring the fluorescence of cell lysates after washing out nonadherent cells. Adhesion was evaluated six times. Mean fluorescence 6 S.D. of four independent experiments is shown. (B) GITRL mRNA expression in HUV-ST and HL-60 cells (untreated or treated with TNF-a or PMA, respectively) evaluated by real-time PCR. Values are the mean (6 S.D.) of the relative GITRL expression versus TNF-a–treated HUV-ST arbitrarily set equal to 100. (C) GITR mRNA expression in HUV-ST and HL-60 cells (untreated or treated with TNF-a or PMA, respectively) was evaluated by real-time PCR. Values are the mean (6 SD) of the relative GITR expression versus CD4+CD25lowGITR+ peripheral lymphocytes (Treg) (evaluated in the same experiments), arbitrarily set equal to 100. (D) GITRL and GITR expression in HUV-ST and HL-60 cells (untreated or treated with TNF-a or PMA, respectively) evaluated by flow cytometry using isotype control (gray pattern) or specific anti-GITR or anti-GITRL Abs (black line). **P , 0.01; ***P , 0.001, according to Student’s t test comparing the adhesion of TNF-a- and PMA-treated cells with untreated cells (A), comparing GITRL or GITR mRNA of TNF-a–treated HUV-ST or PMA-treated HL-60 cells with untreated controls (B and C).

Results in Fig. 4B indicate that the increased adhesion induced by agonist GITR-Fc was counteracted when HUVST cells were pretreated with antagonist anti-GITRL Abs (column 6 versus column 5). On the other hand, no inhibitory effect was observed when HL-60 cells were pretreated with the Abs (column 9 versus column 8). These data provide evidence for a predominant role of GITRL triggering in HUV-ST cells. Figure 4B also shows that the anti-GITRL Abs statistically significantly (P , 0.05) inhibited basal adhesion (i.e., adhesion to untreated or control-Abs treated EC; column 3 versus columns 1 and 2) by inhibiting the physiologic interaction between GITR in HL-60 and GITRL in EC. These data suggest that up-modulation of leukocyte adhesion to EC occurs, at least in part, upon interaction of endothelial GITRL with GITR physiologically expressed in leukocytes and that it can be further increased by pharmacological treatment with GITR-Fc fusion protein. To test this hypothesis, we verified whether inhibition of GITRL triggering by GITR expressed in differentiated HL-60 cells influences adhesion. To this end, we pretreated HL-60 cells with a GITR-blocking Ab. Indeed, the anti-GITR Ab significantly inhibited adhesion of HL-60 cells to EC, confirming that GITRL triggering in EC by GITR expressed in leukocytes favors cell adhesion during the inflammatory response. Results of the effects of GITR-Fc and anti-GITR on cell adhesion are shown in Fig. 4C. Improved adhesion after GITR-Fc treatment may cause an increased transmigration. Therefore, we tested the influence of GITR-Fc on the ability of leukocytes to migrate through an EC monolayer in a transmigration assay. PMA-treated HL-60 cells were loaded on a TNF-a–activated EC monolayer grown into a Transwell insert and were allowed to migrate through the monolayer. The transmigration of differentiated HL-60 cells was 3-fold higher than the control-Fc pretreated EC when the EC monolayer was pretreated with GITR-Fc (Fig. 4D), suggesting that GITRL triggering in EC favors extravasation.

This increased extravasation is likely favored by the increased adhesion promoted by GITRL triggering. GITRL Triggering Upregulates Adhesion Molecules in HUV-ST Cells and Increases Phosphorylation of STAT1 Proteins. To investigate how GITRL triggering in EC favors cell adhesion, we treated activated HUV-ST cells with GITR-Fc and evaluated the expression of adhesion molecules known to be upregulated in EC upon TNF-a activation and to be involved in cell-to-cell interactions. Flow cytometry studies showed that ICAM-1 and VCAM-1 expression on the cell surface was upregulated by GITR-Fc treatment (Fig. 5A): the mean fluorescence intensity of ICAM-1 and VCAM-1 staining increased by 34% (P , 0.01) and 296% (P , 0.001), respectively. Conversely, GITR-Fc did not increase VCAM-1 and ICAM-1 expression when HUV-ST cells were not activated (data not shown). The GITR-Fc–dependent upregulation of adhesion molecules in activated HUV-ST cells has a functional meaning, because blocking Abs against the adhesion molecules inhibited the effect of GITR-Fc treatment (Fig. 5B). This inhibition was not due to the downregulation of GITRL expression as a consequence of blocking Abs treatment (data not shown). In an attempt to investigate how GITRL triggering upregulates ICAM-1 and VCAM-1, we evaluated the phosphorylation level of STAT family members known to be involved in the modulation of adhesion molecules (Duff et al., 1997; Naik et al., 1997). Western blotting analysis demonstrated that GITRL triggering increased phosphorylation of STAT1 (Fig. 6A). Actually, the densitometric analysis of the immunoreactive bands showed that the phosphorylation level of both STAT1a and STAT1b increased within 5 minutes after GITR-Fc treatment (Fig. 6B).

Discussion It is well known that the GITRL-GITR system plays a proinflammatory role in murine disease models and that


169

Fig. 4. GITR-Fc and anti-GITR treatment modulates adhesion of differentiated HL-60 cells to activated EC. (A) GITR-Fc-dependent increase of the adhesion of differentiated/labeled HL-60 cells to activated HUV-ST EC after 30 minutes of contact, as evaluated by measuring the fluorescence of cell lysates after washing out nonadherent cells. EC were pretreated with GITR-Fc or control-Fc (2 mg/ml, 15 minutes), which were maintained in cell coculture (30 minutes). Adhesion was evaluated six times, and the mean fluorescence 6 S.D. of one representative experiment out of five is shown. (B) Adhesion of differentiated/labeled HL-60 cells to activated EC, evaluated as previously specified, after pretreatment with anti-GITRL Ab and GITR-Fc or control-Fc proteins. HUV-ST or HL-60 cells were pretreated with anti-GITRL Ab or control Ab (mouse IgG1 control, 2.5 mg/ml, 15 minutes) (pretreat 1). HUV-ST cells were then treated with GITR-Fc or control-Fc (2 mg/ml, 15 minutes) (pretreat 2) which were maintained in cell coculture (30 minutes). Adhesion was evaluated six times, and the mean fluorescence 6 S.D. of one representative experiment out of three is shown. (C) Modulation of the adhesion of differentiated/labeled HL-60 cells to activated EC, evaluated as specified in panel A legend, after anti-GITR treatment of HL-60 cells compared with GITR-Fc treatment of EC, shown on a logarithmic scale (2 basis). Adhesion of cells treated with anti-CD19 Ab or control-Fc protein was used as control. Values are the mean 6 S.D. of four/five independent experiments in which each experimental group was evaluated six times. (D) Increased transmigration of differentiated/labeled HL-60 cell through an activated EC monolayer in response to GITR-Fc. Activated EC monolayer was treated with GITR-Fc or control-Fc (4 mg/ml, 30 minutes at RT) and then differentiated/labeled HL-60 cells were loaded and allowed to migrate for 6 hours. Fluorescence of migrated cells was measured after lysis, as described in Materials and Methods. Values are the mean (6 S.D.) of three independent determinations. *P , 0.05; **P , 0.01; ***P , 0.001, according to Student’s t test analysis comparing GITR-Fc with Fc-treated cells (A, B, and C) and anti-GITR with control Ab-treated cells (C). #P , 0.05; ##P , 0.01, according to Student’s t test comparing anti-GITRL to control Ab (B).

GITR triggering in cells of the adaptive and innate immune system has an activating function (Krausz et al., 2007; Nocentini and Riccardi, 2009; Azuma, 2010; Ronchetti et al., 2011). Our study demonstrates for the first time that GITRL triggering in EC promotes adhesion in both mice and humans, suggesting that the proinflammatory activity of GITRL/GITR system is due, at least in part, to increased expression of adhesion molecules in EC that may then favor extravasation during the inflammatory response. We also demonstrate that increased adhesion is accompanied by increased STAT-1 phosphorylation and the consequent augmented expression of adhesion molecules such as VCAM-1 and ICAM-1. Previous studies indicate that extravasation is promoted by activation of immune cells, and some data suggest that GITR-stimulated cells have the same behavior. A recent

study performed on mice demonstrated that GITR triggering favors migration of cutaneous DC by increasing inflammatory cytokine production (Kamimura et al., 2009), and another study showed that GITR triggering in T cells upregulates expression of adhesion molecules (Mahesh et al., 2006). Our data, while demonstrating a role for GITRL signaling in EC-activation, do not exclude the possibility that GITR triggering in peripheral blood leukocytes plays a role in increasing their adhesion to EC. Indeed, the inhibition of adhesion caused by antagonist anti-GITR Abs and shown in Fig. 4C may be due not only to inhibition of GITRL triggering on EC but also to inhibition of GITR triggering on leukocytes. In this respect, data from our and other laboratories suggest that the antihuman GITR mAbs currently available are antagonistic (Bianchini et al., 2011; Buechele et al., 2012; Nocentini et al., 2012). If the antagonistic properties of anti-human GITR Abs

170

Lacal et al.

Fig. 5. ICAM-1 and VCAM-1 are involved in the increased adhesion of cells after GITR-Fc treatment. (A) Increased ICAM-1 and VCAM-1 expression in TNF-a-activated HUV-ST cells treated with GITR-Fc. Activated HUV-ST cells were detached from the culture flasks and allowed to recover in complete medium for 2 hours at room temperature in a rotating wheel, treated with 4 mg/ml GITR-Fc or control-Fc protein (30 minutes at room temperature) and stained. The expression levels of adhesion molecules were evaluated by flow cytometry. One representative experiment out of three is shown. Mean fluorescence intensities were as follows: ICAM-1, 527 and 683 for TNF-a and TNF-a/GITR-Fc–treated cells, respectively; VCAM-1, 144 and 494 for TNF-a and TNF-a/GITR-Fctreated cells, respectively. The mean values of three independent experiments were: ICAM-1, 541 6 14 and 663 6 19 (P , 0.01); VCAM, 132 6 11 and 434 6 60 (P , 0.001). (B) Modulation of the adhesion of differentiated/ labeled HL-60 cells to activated HUV-ST EC after 30 minutes of contact when pretreated with GITR-Fc (2 mg/ml, 15 minutes) alone or in association with 5 mg/ml anti–ICAM-1, anti–VCAM-1 or a control Ab. Adhesion was evaluated by measuring the fluorescence of cell lysates after washing out nonadherent cells, and the adhesion of cells treated with control-Fc was used as control, shown on a logarithmic scale (2 basis). Data

are confirmed by in vivo studies, such Abs will deserve to be tested as anti-inflammatory drugs. The evaluation of anti-GITR Abs as potential antiinflammatory drugs is hampered by the structural complexity of the ligands belonging to the TNF superfamily and of their cognate receptors. In fact, both ligands and receptors activate intracellular signaling. Moreover, the differences between human and mouse GITRL-GITR systems clearly suggest that the effects of anti-GITR Abs in mouse experiments cannot be easily extrapolated to predict the activity in human models. For example, GITR-Fc fusion protein has been demonstrated to have various effects subjected to the experimental settings. In some murine in vivo models, GITR-Fc shows an antiinflammatory effect, and its main activity appears to be the inhibition of GITR triggering (Cuzzocrea et al., 2004, 2006; Galuppo et al., 2011a,b). In in vitro models, particularly in human cells, GITR-Fc and soluble GITR exert proinflammatory activity, and their main activity appears to be GITRL triggering (Kim et al., 2006). Moreover, while anti-human GITR mAbs are antagonistic, all anti-murine GITR Abs available possess agonistic properties (Nishioka et al., 2008; Bianchini et al., 2011; Buechele et al., 2012; Nocentini et al., 2012). Such differences may depend on 1) the different primary sequence of the extracellular domain of GITR giving rise also to different alternatively spliced products (Nocentini et al., 2000; Krausz et al., 2007); 2) the tridimensional structures of GITR and GITRL in mice and humans that may cause different triggering/inhibition potentials of GITRFc and of Abs or even different functions in mouse and humans (Chattopadhyay et al., 2009; Nocentini et al., 2012); 3) their partially different tissue distribution (Nocentini et al., 2012). Therefore, future studies on the potential use of antiGITR Abs should consider these differences. Our study demonstrates that GITRL triggering enhances STAT1 phosphorylation and suggests that this increment could be responsible, at least in part, for VCAM-1 and ICAM-1 upregulation. Indeed, induction of adhesion molecule expression and cell adhesion through a STAT1-dependent pathway has been previously described in response to interferon-g (Naik et al., 1997; Lee et al., 2011). Moreover, the STAT1 pathway can upregulate VCAM-1 and ICAM-1 independently of interferon-g (Duff et al., 1997; Stojanovic et al., 2009). There is evidence that STAT1 synergizes with NF-kΒ in inducing VCAM-1 and ICAM-1 (Jahnke and Johnson, 1994; Ohmori et al., 1997; Nizamutdinova et al., 2009; Lee et al., 2011) and that GITRL triggering activates the noncanonical pathway leading to NF-kΒ activation (Grohmann et al., 2007). In our experimental model, it is likely that NF-kΒ is already activated due to treatment with TNF-a (Mackay et al., 1993) and GITRL-dependent-NF-kΒ–activation might synergize with TNF-a in potentiating the upregulation of adhesion molecules. In conclusion, we have demonstrated that the GITRL-GITR system promotes leukocyte adhesion to EC. This effect is consequent to intracellular signaling activated by GITRL

are the mean 6 S.D. of four independent experiments in which each experimental group was evaluated four times. ***P , 0.001, according to Student’s t test comparing anti–ICAM-1 or anti–VCAM-1 treated to control Ab-treated cells.


Fig. 6. STAT1 phosphorylation is increased by GITR-Fc treatment. EC were activated with TNF-a for 24 hours (lanes 3–6) and then washed and treated for different time points with control-Fc or GITR-Fc. Lanes 1–2 represent not activated and buffer-treated cells. Western blot analysis was performed using anti–phospho-STAT1 and anti–b-tubulin as a loading control. A representative immunoblot is shown (A). The levels of protein phosphorylation were evaluated by densitometric analysis, and the results were expressed as the ratio of the densitometric values obtained for GITR-Fc and control-Fc-treated cell extracts, shown on a logarithmic scale (2 basis) (B). Data are the mean 6 S.D. of three independent experiments. *P , 0.05; **P , 0.01; ***P , 0.001, according to Student’s t test comparing GITR-Fc-treated with control-Fc–treated cells.

triggering in EC. Future studies will evaluate the in vivo relevance of such mechanisms and the possible antiinflammatory efficacy of therapeutic use of GITRL and GITR signaling inhibitors. Acknowledgments

The authors thank Dr. J. Yélamos who generously provided us with the murine EC line HYKO6P. Authorship Contributions

Participated in research design: Lacal, Riccardi, Graziani, Nocentini. Conducted experiments: Lacal, Petrillo, Ruffini, Muzi, Bianchini, Ronchetti. Performed data analysis: Lacal, Petrillo, Bianchini, Migliorati, Nocentini. Wrote or contributed to the writing of the manuscript: Lacal, Migliorati, Riccardi, Graziani, Nocentini. References Azuma M (2010) Role of the glucocorticoid-induced TNFR-related protein (GITR)GITR ligand pathway in innate and adaptive immunity. Crit Rev Immunol 30: 547–557. Bae EM, Kim WJ, Suk K, Kang YM, Park JE, Kim WY, Choi EM, Choi BK, Kwon BS, and Lee WH (2008) Reverse signaling initiated from GITRL induces NF-kB activation through ERK in the inflammatory activation of macrophages. Mol Immunol 45:523–533. Baessler T, Krusch M, Schmiedel BJ, Kloss M, Baltz KM, Wacker A, Schmetzer HM, and Salih HR (2009) Glucocorticoid-induced tumor necrosis factor receptor-related protein ligand subverts immunosurveillance of acute myeloid leukemia in humans. Cancer Res 69:1037–1045.

171

Bianchini R, Bistoni O, Alunno A, Petrillo MG, Ronchetti S, Sportoletti P, Bocci EB, Nocentini G, Gerli R, and Riccardi C (2011) CD41 CD25(low) GITR1 cells: a novel human CD4(1) T-cell population with regulatory activity. Eur J Immunol 41: 2269–2278. Bianchini R, Nocentini G, Krausz LT, Fettucciari K, Coaccioli S, Ronchetti S, and Riccardi C (2006) Modulation of pro- and antiapoptotic molecules in doublepositive (CD41CD81) thymocytes following dexamethasone treatment. J Pharmacol Exp Ther 319:887–897. Buechele C, Baessler T, Wirths S, Schmohl JU, Schmiedel BJ, and Salih HR (2012) Glucocorticoid-induced TNFR-related protein (GITR) ligand modulates cytokine release and NK cell reactivity in chronic lymphocytic leukemia (CLL). Leukemia 26:991–1000. Byrne AM, Goleva E, and Leung DY (2009) Identification of glucocorticoid-induced TNF receptor-related protein ligand on keratinocytes: ligation by GITR induces keratinocyte chemokine production and augments T-cell proliferation. J Invest Dermatol 129:2784–2794. Chattopadhyay K, Lazar-Molnar E, Yan Q, Rubinstein R, Zhan C, Vigdorovich V, Ramagopal UA, Bonanno J, Nathenson SG, and Almo SC (2009) Sequence, structure, function, immunity: structural genomics of costimulation. Immunol Rev 229: 356–386. Cuzzocrea S, Ayroldi E, Di Paola R, Agostini M, Mazzon E, Bruscoli S, Genovese T, Ronchetti S, Caputi AP, and Riccardi C (2005) Role of glucocorticoid-induced TNF receptor family gene (GITR) in collagen-induced arthritis. FASEB J 19:1253–1265. Cuzzocrea S, Nocentini G, Di Paola R, Agostini M, Mazzon E, Ronchetti S, Crisafulli C, Esposito E, Caputi AP, and Riccardi C (2006) Proinflammatory role of glucocorticoid-induced TNF receptor-related gene in acute lung inflammation. J Immunol 177:631–641. Cuzzocrea S, Nocentini G, Di Paola R, Mazzon E, Ronchetti S, Genovese T, Muià C, Caputi AP, and Riccardi C (2004) Glucocorticoid-induced TNF receptor family gene (GITR) knockout mice exhibit a resistance to splanchnic artery occlusion (SAO) shock. J Leukoc Biol 76:933–940. Cuzzocrea S, Ronchetti S, Genovese T, Mazzon E, Agostini M, Di Paola R, Esposito E, Muià C, Nocentini G, and Riccardi C (2007) Genetic and pharmacological inhibition of GITR-GITRL interaction reduces chronic lung injury induced by bleomycin instillation. FASEB J 21:117–129. Duff JL, Quinlan KL, Paxton LL, Naik SM, and Caughman SW (1997) Pervanadate mimics IFNgamma-mediated induction of ICAM-1 expression via activation of STAT proteins. J Invest Dermatol 108:295–301. Galuppo M, Nocentini G, Mazzon E, Ronchetti S, Esposito E, Riccardi L, Di Paola R, Bruscoli S, Riccardi C, and Cuzzocrea S (2011a) GITR gene deletion and GITR-FC soluble protein administration inhibit multiple organ failure induced by zymosan. Shock 36:263–271. Galuppo M, Nocentini G, Mazzon E, Ronchetti S, Esposito E, Riccardi L, Sportoletti P, Di Paola R, Bruscoli S, and Riccardi C, et al. (2011b) The glucocorticoid-induced TNF receptor family-related protein (GITR) is critical to the development of acute pancreatitis in mice. Br J Pharmacol 162:1186–1201. Grohmann U, Volpi C, Fallarino F, Bozza S, Bianchi R, Vacca C, Orabona C, Belladonna ML, Ayroldi E, and Nocentini G, et al. (2007) Reverse signaling through GITR ligand enables dexamethasone to activate IDO in allergy. Nat Med 13: 579–586. Gurney AL, Marsters SA, Huang RM, Pitti RM, Mark DT, Baldwin DT, Gray AM, Dowd AD, Brush AD, and Heldens AD, et al. (1999) Identification of a new member of the tumor necrosis factor family and its receptor, a human ortholog of mouse GITR. Curr Biol 9:215–218. Jahnke A and Johnson JP (1994) Synergistic activation of intercellular adhesion molecule 1 (ICAM-1) by TNF-alpha and IFN-gamma is mediated by p65/p50 and p65/c-Rel and interferon-responsive factor Stat1 alpha (p91) that can be activated by both IFN-gamma and IFN-alpha. FEBS Lett 354:220–226. Kamimura Y, Iwai H, Piao J, Hashiguchi M, and Azuma M (2009) The glucocorticoidinduced TNF receptor-related protein (GITR)-GITR ligand pathway acts as a mediator of cutaneous dendritic cell migration and promotes T cell-mediated acquired immunity. J Immunol 182:2708–2716. Kim SJ, Shin HH, Park SY, Lee DS, Lee EA, Cho SD, Cho HR, Miyazawa K, and Choi HS (2006) Induction of MMP-13 expression by soluble human glucocorticoidinduced tumor necrosis factor receptor in fibroblast-like synovial cells. Osteoarthritis Cartilage 14:146–153. Kohm AP, Williams JS, and Miller SD (2004) Cutting edge: ligation of the glucocorticoid-induced TNF receptor enhances autoreactive CD41 T cell activation and experimental autoimmune encephalomyelitis. J Immunol 172:4686–4690. Krausz LT, Bianchini R, Ronchetti S, Fettucciari K, Nocentini G, and Riccardi C (2007) GITR-GITRL system, a novel player in shock and inflammation. ScientificWorldJournal 7:533–566. Kwon B, Yu KY, Ni J, Yu GL, Jang IK, Kim YJ, Xing L, Liu D, Wang SX, and Kwon BS (1999) Identification of a novel activation-inducible protein of the tumor necrosis factor receptor superfamily and its ligand. J Biol Chem 274:6056–6061. Lacal PM, Tentori L, Muzi A, Ruffini F, Dorio AS, Xu W, Arcelli D, Zhang J, and Graziani G (2009) Pharmacological inhibition of poly(ADP-ribose) polymerase activity down-regulates the expression of syndecan-4 and Id-1 in endothelial cells. Int J Oncol 34:861–872. Lee JK, Kim JK, Park SH, Sim YB, Jung JS, and Suh HW (2011) Lactosylceramide mediates the expression of adhesion molecules in TNF-a and IFNg-stimulated primary cultured astrocytes. Korean J Physiol Pharmacol 15:251–258. Lee SK, Choi BK, Kim YH, Kang WJ, Kim KH, Sakaguchi S, Suh JH, Kim TY, and Kwon BS (2006) Glucocorticoid-induced tumour necrosis factor receptor family-related receptor signalling exacerbates hapten-induced colitis by CD41 T cells. Immunology 119:479–487. Liao G, Detre C, Berger SB, Engel P, de Waal Malefyt R, Herzog RW, Bhan AK, and Terhorst C (2012) Glucocorticoid-induced tumor necrosis factor receptor family-related protein regulates CD41T cell-mediated colitis in mice. Gastroenterology 142: 582–591.e588 DOI: 10.1053/j.gastro.2011.11.031.

172

Lacal et al.

Mackay F, Loetscher H, Stueber D, Gehr G, and Lesslauer W (1993) Tumor necrosis factor alpha (TNF-alpha)-induced cell adhesion to human endothelial cells is under dominant control of one TNF receptor type, TNF-R55. J Exp Med 177:1277–1286. Mahesh SP, Li Z, Liu B, Fariss RN, and Nussenblatt RB (2006) Expression of GITR ligand abrogates immunosuppressive function of ocular tissue and differentially modulates inflammatory cytokines and chemokines. Eur J Immunol 36:2128– 2138. Morris GP and Kong YC (2006) Interference with CD41CD251 T-cell-mediated tolerance to experimental autoimmune thyroiditis by glucocorticoid-induced tumor necrosis factor receptor monoclonal antibody. J Autoimmun 26:24–31. Naik SM, Shibagaki N, Li LJ, Quinlan KL, Paxton LL, and Caughman SW (1997) Interferon gamma-dependent induction of human intercellular adhesion molecule1 gene expression involves activation of a distinct STAT protein complex. J Biol Chem 272:1283–1290. Nardelli B, Zaritskaya L, McAuliffe W, Ni Y, Lincoln C, Cho YH, Birse CE, Halpern W, Ullrich S, and Moore PA (2006) Osteostat/tumor necrosis factor superfamily 18 inhibits osteoclastogenesis and is selectively expressed by vascular endothelial cells. Endocrinology 147:70–78. Nishioka T, Nishida E, Iida R, Morita A, and Shimizu J (2008) In vivo expansion of CD41 Foxp31 regulatory T cells mediated by GITR molecules. Immunol Lett 121:97–104. Nizamutdinova IT, Kim YM, Chung JI, Shin SC, Jeong YK, Seo HG, Lee JH, Chang KC, and Kim HJ (2009) Anthocyanins from black soybean seed coats preferentially inhibit TNF-alpha-mediated induction of VCAM-1 over ICAM-1 through the regulation of GATAs and IRF-1. J Agric Food Chem 57:7324–7330. Nocentini G and Riccardi C (2009) GITR: a modulator of immune response and inflammation. Adv Exp Med Biol 647:156–173. Nocentini G, Ronchetti S, Bartoli A, Spinicelli S, Delfino D, Brunetti L, Migliorati G, and Riccardi C (2000) Identification of three novel mRNA splice variants of GITR. Cell Death Differ 7:408–410. Nocentini G, Ronchetti S, Petrillo MG, and Riccardi C (2012) Pharmacological modulation of GITRL/GITR system: therapeutic perspectives. Br J Pharmacol 165: 2089–2099. Ohmori Y, Schreiber RD, and Hamilton TA (1997) Synergy between interferongamma and tumor necrosis factor-alpha in transcriptional activation is mediated by cooperation between signal transducer and activator of transcription 1 and nuclear factor kB. J Biol Chem 272:14899–14907.

Rinaldi B, Donniacuo M, Esposito E, Capuano A, Sodano L, Mazzon E, Di Palma D, Paterniti I, Cuzzocrea S, and Rossi F (2011) PPARa mediates the antiinflammatory effect of simvastatin in an experimental model of zymosan-induced multiple organ failure. Br J Pharmacol 163:609–623. Ronchetti S, Nocentini G, Petrillo MG, Bianchini R, Sportoletti P, Bastianelli A, Ayroldi EM, and Riccardi C (2011) Glucocorticoid-induced TNFR family related gene (GITR) enhances dendritic cell activity. Immunol Lett 135:24–33. Ronchetti S, Nocentini G, Petrillo MG, and Riccardi C (2012) CD81 T cells: GITR matters. ScientificWorldJournal 2012:308265. Ronchetti S, Zollo O, Bruscoli S, Agostini M, Bianchini R, Nocentini G, Ayroldi E, and Riccardi C (2004) GITR, a member of the TNF receptor superfamily, is costimulatory to mouse T lymphocyte subpopulations. Eur J Immunol 34: 613–622. Santucci L, Agostini M, Bruscoli S, Mencarelli A, Ronchetti S, Ayroldi E, Morelli A, Baldoni M, and Riccardi C (2007) GITR modulates innate and adaptive mucosal immunity during the development of experimental colitis in mice. Gut 56:52–60. Satoguina JS, Adjobimey T, Arndts K, Hoch J, Oldenburg J, Layland LE, and Hoerauf A (2008) Tr1 and naturally occurring regulatory T cells induce IgG4 in B cells through GITR/GITR-L interaction, IL-10 and TGF-beta. Eur J Immunol 38: 3101–3113. Schaer DA, Murphy JT, and Wolchok JD (2012) Modulation of GITR for cancer immunotherapy. Curr Opin Immunol 24:217–224. Stojanovic T, Wagner AH, Wang S, Kiss E, Rockstroh N, Bedke J, Gröne HJ, and Hecker M (2009) STAT-1 decoy oligodeoxynucleotide inhibition of acute rejection in mouse heart transplants. Basic Res Cardiol 104:719–729. Tentori L, Vergati M, Muzi A, Levati L, Ruffini F, Forini O, Vernole P, Lacal PM, and Graziani G (2005) Generation of an immortalized human endothelial cell line as a model of neovascular proliferating endothelial cells to assess chemosensitivity to anticancer drugs. Int J Oncol 27:525–535.

Address correspondence to: Dr. Carlo Riccardi, Department of Medicine, University of Perugia, Piazza Gambuli, Bldg D, 2° floor, I-06132 San Sisto, Perugia, Italy. E-mail: [email protected]