Regulatory T cells (Tregs) are a subpopulation of lymphocytes that play a role in suppressing and regulating immune responses. Recently, it was suggested that controlling the functions and activities of Tregs might be applicable to the treatment of human diseases such as autoimmune diseases, organ transplant rejection, and graft-versus-host disease. TNF receptor type 2 (TNFR2) is a target molecule that modulates Treg functions. In this study, we investigated the role of TNFR2 signaling in the differentiation and activation of mouse Tregs. We previously reported the generation of a TNFR2-selective agonist TNF mutant, termed R2agoTNF, by using our unique cytokine modification method based on phage display. R2agoTNF activates cell signaling via mouse TNFR2. In this study, we evaluated the efficacy of R2agoTNF for the proliferation and activation of Tregs in mice. R2agoTNF expanded and activated mouse CD4+CD25+ Tregs ex vivo. The structural optimization of R2agoTNF by internal cross-linking or IgG-Fc fusion selectively and effectively enhanced Treg expansion in vivo. Furthermore, the IgG-Fc fusion protein suppressed skin-contact hypersensitivity reactions in mice. TNFR2 agonists are expected to be new Treg expanders.

This article is featured in Top Reads, p.1679

Regulatory T cells (Tregs) are essential for the maintenance of self-tolerance and immune cell homeostasis. They also have an important role in regulating immune responses responsible for autoimmune diseases, transplantation rejection, and graft-versus-host disease (GvHD) (13). Tregs were originally identified as a specialized CD4+ T cell subset that had immunosuppressive functions, and which expressed CD25, the IL-2R α-chain. Subsequently, the Foxp3 gene encoding a master transcription factor was identified as responsible for autoimmune symptoms in scurfy mice and immunodysregulation polyendocrinopathy enteropathy X-linked syndrome patients. Foxp3 has an essential role in Treg differentiation and function and, therefore, is an important Treg-specific molecular marker. Tregs can be classified into three populations: thymus-derived Tregs, peripherally derived Tregs, and inducible Tregs (iTreg). The expression of Foxp3 is essential for their suppressive function. Although iTregs have lower stability, the immunosuppressive activity of these Tregs are thought to induce immune tolerance. Tregs exhibit immunosuppressive functions mediated by the following mechanisms: 1) CD25 neutralizes IL-2, which is required for the activation and proliferation of effector T cells (4); 2) CTLA-4 (CD152) binds to CD80/86 on APCs to suppress CD28-mediated costimulatory signals (5, 6); and 3) IL-10 secreted by Tregs reduces the expression level of CD80/86 on APCs to suppress the production of inflammatory cytokines such as IL-6 and TNF-α (7, 8). Therefore, cell surface molecules expressed on Tregs such as CD25 and CTLA-4 are considered to play an important role in the immunosuppressive function of Tregs. OX-40, GITR, and FR-4 were also identified as target molecules that regulate Tregs (911). Recent studies reported that TNF receptor type 2 (TNFR2) is preferentially expressed on Tregs (1215) and is a target molecule that controls Treg functions (1618). The expression of Treg surface molecules such as CD25 or CTLA-4 is controlled by Foxp3. In contrast, the mechanism related to TNFR2 expression in Tregs is poorly understood.

Although Tregs are potent inhibitors of immune responses, they are not an abundant population of cells. IL-2 expands populations of Tregs with promising effects related to immune tolerance (19, 20). However, IL-2 can also activate effector T cells or induce Treg expansion with low purity or heterogeneity (21, 22). Therefore, new Treg expanders that selectively and markedly induce the proliferation and activation of Tregs are being explored for the treatment of immune diseases. One approach is to generate a TNFR2 agonist that selectively activates TNFR2 signaling. TNF-α transmits intracellular signaling via two receptor subtypes, TNFR1 and TNFR2. TNFR1 is ubiquitously expressed, but TNFR2 is restricted to specific cell types such as endothelial cells, nerve cells, and immune cells (2325). Recent studies reported that TNFR2 was highly expressed on mouse and human Tregs (12, 13). TNFR2-expressing Tregs had a higher suppressive activity on the proliferation of effector T cells compared with TNFR2-nonexpressing Tregs (14). Furthermore, TNFR2-deficient mice developed exacerbated symptoms of experimental autoimmune encephalomyelitis, a model of multiple sclerosis (26). Several other studies reported that oligomeric TNFR2-selective TNF muteins (27), S95C/G148C (TNF07) double-mutated TNF (28) or anti-TNFR2 Ab (22) increased Tregs via TNFR2.

We previously generated a mouse TNFR2-selective agonist TNF-α mutant (R2agoTNF) derived from human TNF-α by multiple amino acid replacement using phage display (29). R2agoTNF has a unique property because it binds to mouse TNFR2, but not TNFR1, and transmits TNFR2 signaling selectively. In this study, we confirmed that our new TNFR2 agonist, R2agoTNF, expanded Tregs and enhanced immunosuppressive activity ex vivo as a potent Treg expander. In addition, we demonstrated that R2agoTNF derivatives, which have undergone internal cross-linking or Fc fusion for structural optimization, expanded murine Tregs in vivo.

C57BL/6 wild-type (WT) mice were purchased from Oriental Yeast (Tokyo, Japan). B6.129S2-Tnfrsf1btm1Mwm/J (TNFR2–/–) mice were purchased from The Jackson Laboratory (Bar Harbor, ME) and bred under specific pathogen–free conditions. All animal studies were approved by the Kobe Gakuin University Experimental Animal Care and Use Committee.

Single-cell suspensions from mesenteric lymph nodes (LNs), inguinal LNs, axillary LNs, or spleen were prepared by filtration through a 70-μm cell strainer (BD Biosciences, San Jose, CA). Erythrocytes were lysed using BD Pharm Lyse (BD Biosciences). Freshly prepared cells were used in each experiment.

LN cells or splenocytes (SPLs) were cultured with RPMI-1640 (FUJIFILM Wako Pure Chemical Corporation, Osaka, Japan) supplemented with 10% FBS (Biowest, Nuaillé, France), 1% antibiotic mixture (10,000 U/ml penicillin, 10 mg/ml streptomycin, and 25 μg/ml amphotericin B; FUJIFILM Wako Pure Chemical Corporation), 50 μM 2-ME (FUJIFILM Wako Pure Chemical Corporation), and 1% nonessential amino acids solution (FUJIFILM Wako Pure Chemical Corporation). mTNFR2/mFas preadipocytes (TNFR1–/–/TNFR2–/– mouse preadipocytes expressing the chimeric extracellular and transmembrane domains of mouse TNFR2 and the intracellular domain of mouse Fas), which were previously established (30), were cultured in DMEM (FUJIFILM Wako Pure Chemical Corporation) supplemented with 10% FBS, 1% antibiotic mixture, and 5 μg/ml blasticidin S HCl (FUJIFILM Wako Pure Chemical Corporation).

Cells were suspended in PBS containing 2% FBS and stained. The following fluorescent-labeled anti-mouse surface molecule Abs were used for surface marker staining: CD4 (RM4-5)/PerCP, CD25 (PC61)/BV421, CD8 (53-6.7)/FITC, TNFR2 (TR75-89)/PE, TNFR1 (55R-286)/PE, GITR (DTA-1)/PE, CTLA4 (UC10-4F10-11)/PE, BAFFR (7H22-E16)/PE, OX-40 (OX-86)/PE, NK1.1 (PK136)/BV421, CD11c (N418)/BV421, CD19 (6D5)/allophycocyanin, CD45R/B220 (RA3-6B2)/Alexa488, and CD90.2 (30-H12)/PE (all from BioLegend, San Diego, CA). Live cells were stained by eBioscience Fixable Viability Dye eFluor 506 (Thermo Fisher Scientific, Waltham, MA). Anti-mouse CD16/CD32 (2.4G2) (BD Biosciences) was used for Fc blocking. For intracellular staining, anti-Foxp3 (FJK-16s)/allophycocyanin (Thermo Fisher Scientific), anti–TGF-β1 (TW7-16B4)/FITC, and anti–IL-10 (JES5-16E3)/PE were used after fixation and permeabilization with eBioscience Foxp3/Transcription Factor Staining Buffer Set (Thermo Fisher Scientific) according to the manufacturer’s protocol. Isotype controls matched for each Ab were used. Flow cytometry (FCM) was performed using a CytoFLEX Cell Analyzer (Beckman Coulter, Brea, CA). Data were analyzed using FlowJo software (BD Biosciences).

CD4+CD25 conventional T cells (Tconvs) were purified from the LNs of WT mice. CD4+CD25+ Tregs of WT or TNFR2–/– mice were separately purified from each LN. CD90.2 cells were also isolated from the spleens of WT mice as APCs. Cells were separated by an FACSAria Cell Sorter (BD Biosciences). CD4+CD25 Tconvs were labeled with CFSE (3 μM) using a CellTrace CFSE Cell Proliferation Kit (Thermo Fisher Scientific) for 5 min at room temperature. CFSE-labeled Tconvs (5 × 104 cells per well) were cocultured with WT Tregs or TNFR2–/– Tregs at the desired ratio plus APCs (2 × 105 cells per well) and purified anti-mouse CD3ε mAb (0.5 μg/ml) (145-2C11; BD Biosciences). After 72 h, the proliferation of Tconvs was estimated from CFSE dilution measured by FCM.

Purified anti-mouse CD3ε mAb (BD Biosciences) was bound to a 96-well plate. CD4+CD25 CD44low CD62Lhigh naive T cells isolated from lymphocytes were seeded and cultured with purified anti-mouse CD28 mAb (37.51; BD Biosciences) and recombinant human TGF-β (R&D Systems, Minneapolis, MN). After 72 h, cells were collected and stained for CD4, Foxp3, and TNFR2. Populations of Tregs and Tconvs were analyzed by FCM.

Total RNA was prepared using an RNeasy Mini Kit (QIAGEN) from purified CD4+CD25+ Tregs, CD4+CD25 Tconvs, or TGF-β induced naive T cells. Total RNA was reverse transcribed using an ReverTra Ace qPCR RT Kit (TOYOBO, Tokyo, Japan). Real-time PCR was performed using the TaqMan probe (mouse Foxp3; Mm00475162_m1, mouse Tnfrsf1a; Mm00441883_g1, mouse Tnfrsf1b; MN00441889_m1, mouse hypoxanthine phosphoribosyltransferase 1 [HPRT1]; Mm01545399_m1 [Applied Biosystems, Foster City, CA] with THUNDERBIRD Probe qPCR Mix [TOYOBO] and a StepOnePlus Real-Time PCR System [Applied Biosystems]). Expression levels of Foxp3, Tnfrsf1a, and Tnfrsf1b were determined relative to that of HPRT1.

The Foxp3 gene is encoded on the X chromosome. SPLs were prepared from female depletion of Tregs (DEREG)–Scurfy mice generated by crossing female heterozygous scurfy mice with male DEREG mice. Because of the heterogeneous inactivation of Foxp3, scurfy mice harbor normal Foxp3-expressing Tregs and inactivated Foxp3-expressing Tregs. DEREG mice, a bacterial artificial chromosome–transgenic mouse, express a fusion protein of the human diphtheria toxin receptor and enhanced GFP under control of the Foxp3 promoter. Tregs are tracked by their GFP expression (31, 32). GFP-positive and Foxp3-positive normal Tregs and GFP-positive and Foxp3-negative inactivated mutant Tregs were fractionated from SPLs of female DEREG-Scurfy mice using anti-mouse Foxp3 mAb, and the TNFR2 expression level of each cell type was analyzed.

CD4+CD25+ Tregs or CD4+CD25 Tconvs isolated from the LNs of WT mice were labeled by CFSE (3 μM) using a CellTrace CFSE Cell Proliferation Kit (Thermo Fisher Scientific) for 5 min at room temperature. CFSE-labeled cells (5 × 104 cells per well) were cultured with anti-mouse CD3ε mAb (0.5 μg/ml), human IL-2 (10 U/ml) (Imunase35; Shionogi, Osaka, Japan), or R2agoTNF (100 ng/ml) in a U-bottom 96-well plate for 72 h. Anti-human IL-2 Ab (MQ1-17H12) was used to neutralize IL-2. After the collection of cells, the proliferation of each cell type was measured by FCM.

The R2agoTNF structure was modeled on that of R1antTNF (Protein Data Bank code 2E7A) using UCSF Chimera and MODELER. The structure was superimposed on that of lymphotoxin-α derived from the lymphotoxin-α·TNFR1 complex (Protein Data Bank code 1TNR).

Mouse TNFR2/mouse Fas preadipocytes (1.5 × 104 cells per well) were cultured with serially diluted recombinant murine TNF-α (PeproTech, Rocky Hill, NJ), R2agoTNF, or scR2agoTNF-Fc for 48 h. Mouse CD4+CD25+ Tregs (3 × 104 cells per well) were cultured with serially diluted scR2agoTNF-Fc for 24 h. After incubation at 37°C, cell viability was measured using a WST-8 colorimetric assay (Cell Counting Kit-8; Dojindo Laboratories, Kumamoto, Japan) according to the manufacturer’s protocol.

Cells were cultured with Cell Activation Mixture (including PMA and ionomycin) containing brefeldin A (BioLegend) for 4 h. After collecting cells by centrifugation, cell surface markers were stained with a fluorescent-labeled Ab as described in the Flow Cytometry section. Then, cells were fixed and permeabilized with a Foxp3/Transcription Factor Staining Buffer Set (Thermo Fisher Scientific) according to the manufacturer’s instructions and stained with Abs.

Leukocytes were prepared from the LNs of WT mice. Then, the expressions of cytokine receptors including TNFR2, TNFR1, GITR, CTLA-4, BAFFR, and OX-40 on CD4+Foxp3+ Tregs, CD4+Foxp3 Tconvs, CD8+ CTLs, CD11c+ dendritic cells (DCs), CD19+ CD45R+ B cells, and NK1.1+ NK cells were analyzed by FCM using fluorescent-labeled Abs as described in the Flow Cytometry section.

For the S95C/G148C mutations and 6×His-tag insertion to the C-terminal, inverse PCR of the pYas-R2agoTNF plasmid DNA (29) was repeated using a mutation primer. Primer sets used are shown in Supplemental Fig. 1. The template plasmid DNA was digested by DpnI (TOYOBO). Self-ligation of PCR products was performed by a reaction with T4 polynucleotide kinase (TOYOBO) and Ligation high version 2 (TOYOBO). Escherichia coli was transformed by self-ligated PCR products. scR2agoTNF cDNA was artificially synthesized (GenScript, Piscataway, NJ) and inserted into the pCAG-human IgG-Fc fusion vector (FUJIFILM Wako Pure Chemical Corporation).

The protocol for the expression and purification of recombinant proteins was described previously (29, 33). Briefly, R2agoTNF-DS protein was overexpressed in E. coli BL21(DE3)-RIL (Agilent Technologies, Santa Clara, CA). Expressed protein was recovered from the soluble fraction using BugBuster Protein Extraction Reagent (Merck MilliporeSigma, Burlington, MA). Protein was purified by affinity chromatography using Ni-Sepharose Excel Resin (GE Healthcare Biosciences, Piscataway, NJ) followed by size-exclusion chromatography (SEC) using an HiLoad 16/600 Superdex 200 Prep Grade Column (GE Healthcare Biosciences). Protein expression of scR2agoTNF-Fc was analyzed using the Expi293 expression system (Thermo Fisher Scientific) in accordance with the manufacturer’s instructions. Briefly, the pCAG-scR2agoTNF-Fc plasmid vector was transfected into Expi293F cells by using an ExpiFectamine 293 Transfection Kit (Thermo Fisher Scientific). After culture for 7 d, cultured medium was collected by centrifugation. The scR2agoTNF-Fc protein was recovered from the supernatant using KANEKA KanCapA (FUJIFILM Wako Pure Chemical Corporation) and eluted with 0.1 M glycine-HCl (pH 2.8). Recovered protein was further purified by SEC using a Superdex column.

Proteins were mixed in Laemmli sample buffer without 2-ME. SDS-PAGE, under unreduced conditions, was performed using Tris-glycine buffer (pH 8.3) and proteins were detected by Coomassie Brilliant Blue G-250 staining. For Western blotting, proteins were transferred onto a polyvinylidene fluoride membrane. After the blocking of unspecific binding proteins, proteins were probed with human TNF-α Biotinylated Ab (R&D Systems) and Streptavidin HRP Conjugates (Jackson ImmunoResearch, West Grove, PA). Visualization was performed by chemiluminescence with ECL Prime (GE Healthcare Biosciences).

R2agoTNF (250 μg/ml) and R2agoTNF-DS (250 μg/ml) were serially diluted 2-fold with PBS (pH 7.4). The diluted proteins were mixed with SYPRO Orange (Protein Thermal Shift Dye Kit; Applied Biosystems) and heat-denatured by raising the temperature from 25°C to 90°C at a rate of 0.16°C/10 s. Fluorescence intensity was measured using a StepOnePlus Real-Time PCR System (Applied Biosystems). To determine the melting temperature (Tm) values, data were analyzed using Protein Thermal Shift Software v1.0 (Applied Biosystems).

R2agoTNF-DS was administered i.p. to WT mice twice a day for 5 d. scR2agoTNF-Fc was administered i.p. to WT mice twice a week for 1, 2, and 4 wk. Saline was administered i.p. as the negative control. After each period of administration, leukocytes were prepared from LNs and stained with fluorescent-labeled Abs as described in the Flow Cytometry section. CD4+Foxp3+ Tregs, CD4+Foxp3 Tconvs, CD8+ T cells, and granulocytes from each group were measured by FCM.

Mice were sensitized epicutaneously at day 0 by applying 100 μl of 0.5% 2,4-dinitrofluorobenzene (DNFB) diluted in acetone to the abdominal skin and challenged at day 5 by applying 20 μl of 0.5% DNFB onto the ear as previously described (34). scR2agoTNF-Fc (5 and 50 μg per mouse) were administered by i.p. at day −2 and day 1 before challenge. Ear swelling and populations of IFN-γ+ cells and CD4+Foxp3+ T cells were assessed at 24 h after DNFB challenge.

The significance of differences was determined by one-way ANOVA followed by a secondary test (Tukey) or unpaired Student t test, and *p < 0.05, **p < 0.01, and ***p < 0.001 were considered statistically significant. All statistical analyses were carried out using GraphPad Prism for Windows version 6.0 (GraphPad Software, San Diego, CA).

TNFR2-deficiency was reported to decrease the population of Tregs in lymphoid tissues and the suppressive activity of Tregs (14, 15). We confirmed the presence of a Treg population in TNFR2–/– mice (Fig. 1A). There were significantly lower numbers of CD4+Foxp3+ Tregs in the LNs and spleen of TNFR2–/– mice compared with WT mice. However, the numbers of CD4+Foxp3 Tconvs (Fig. 1A), CD8+ T cells, and granulocytes (data not shown) were similar between WT mice and TNFR2–/– mice. We also confirmed the immunosuppressive function of Tregs in TNFR2–/– mice was partial (Fig. 1B). After CD4+CD25+ Tregs and CD4+CD25 Tconvs were isolated from the LNs of WT and TNFR2–/– mice, CFSE-labeled WT Tconvs were cocultured with WT Tregs or TNFR2–/– Tregs at several ratios for 72 h. WT Tregs suppressed the proliferation of WT Tconvs by a ratio-dependent mechanism (Fig. 1B, upper). In contrast, TNFR2–/– Tregs had a lower suppressive activity than WT Tregs. A plot of Tconv/Treg ratio versus Tconv inhibition by Tregs indicated TNFR2 deficiency decreased the suppressive activity of Tregs (Fig. 1B, lower). CFSE-labeled mouse CD4+CD25+ WT Tregs and TNFR2–/– Tregs were stimulated by mouse TNF-α with or without anti-mouse CD3 mAb. Mouse TNF-α induced the proliferation of WT Tregs but not TNFR2–/– Tregs under anti-CD3 stimulation (Fig. 1C). These results indicated that TNFR2 signaling is involved in the proliferation of Tregs.

FIGURE 1.

TNFR2 deficiency decreases the immunosuppressive activity of Tregs. (A) Populations of CD4+Foxp3+ Tregs and CD4+Foxp3 Tconvs in lymphocytes and SPLs were compared between WT and TNFR2–/– mice. Representative dot plots are shown. Data are the mean ± SD (n = 3). (B) Suppressive activities of WT Tregs and TNFR2–/– Tregs were examined by culturing with CFSE-labeled WT Tconvs at several cell ratios (Treg suppression assay). Representative cell division histograms for each ratio are shown. Tconv inhibition rates (%) by WT Tregs or TNFR2–/– Tregs were calculated from the Treg suppression assay. Data are the mean ± SD (n = 3). (C) Differences in proliferation between WT Tregs or TNFR2–/– Tregs by mouse TNF-α simulation are shown. *p < 0.05, **p < 0.01 (unpaired Student t-test).

FIGURE 1.

TNFR2 deficiency decreases the immunosuppressive activity of Tregs. (A) Populations of CD4+Foxp3+ Tregs and CD4+Foxp3 Tconvs in lymphocytes and SPLs were compared between WT and TNFR2–/– mice. Representative dot plots are shown. Data are the mean ± SD (n = 3). (B) Suppressive activities of WT Tregs and TNFR2–/– Tregs were examined by culturing with CFSE-labeled WT Tconvs at several cell ratios (Treg suppression assay). Representative cell division histograms for each ratio are shown. Tconv inhibition rates (%) by WT Tregs or TNFR2–/– Tregs were calculated from the Treg suppression assay. Data are the mean ± SD (n = 3). (C) Differences in proliferation between WT Tregs or TNFR2–/– Tregs by mouse TNF-α simulation are shown. *p < 0.05, **p < 0.01 (unpaired Student t-test).

Close modal

Induced Tregs are differentiated from naive T cells in peripheral lymph tissues and TGF-β has a critical role in this process (35, 36). We confirmed the relationship between TGF-β–induced Treg differentiation and TNFR2 expression in mice. CD4+CD25CD44low CD62Lhigh naive T cells were isolated from the LNs of WT mice by cell sorting, then stimulated with TGF-β for 72 h. Naive T cells were differentiated to CD4+Foxp3+ Tregs and an increase in the number of Tregs was dependent on the increase in TGF-β concentration (Fig. 2A, upper). TNFR2 expression on CD4+Foxp3+ Tregs was increased dependent on Treg differentiation (Fig. 2A, lower), whereas TNFR2 expression on Tconvs was not increased. We also measured the transcript levels of Foxp3, Tnfrsf1b, and Tnfrsf1a under the same differentiation conditions using reverse transcription quantitative PCR (RT-qPCR) (Fig. 2B). Foxp3 mRNA was increased by a TGF-β concentration–dependent mechanism. Moreover, increases in TNFR2 mRNA and Foxp3 mRNA were positively correlated. There was no correlation between the increase of Foxp3 mRNA and TNFR1 mRNA. In the absence of mouse anti-CD3 stimulation, Foxp3 mRNA and TNFR2 mRNA were not increased by TGF-β stimulation (data not shown). Therefore, TNFR2 was upregulated by costimulation of the TGF-βR and TCR. DEREG-Scurfy mice harbor two types of Tregs, Foxp3GFP+ inactivated mutant Tregs and Foxp3+ GFP+ normal Tregs. We measured the level of TNFR2 in each Treg population isolated from the spleens of DEREG-Scurfy mice and found TNFR2 was expressed by normal Tregs but not inactivated mutant Tregs (Fig. 2C). To investigate the relationship between Foxp3 and TNFR2 in iTreg, naive T cells from DEREG-Scurfy mice were differentiated by TGF-β (Fig. 2D). TNFR2 expression levels were higher in GFP+Foxp3+ T cells than in GFP+ Foxp3 or GFPFoxp3 T cells. Thus, these data showed that the transcriptional activity of Foxp3 upregulated TNFR2 expression on Tregs.

FIGURE 2.

Relationship between Foxp3 and TNFR2 in Treg differentiation in vitro. (A) TGF-β–dependent differentiation of naive T cells into CD4+Foxp3+ Tregs was analyzed by FCM. TNFR2 expression levels on CD4+Foxp3+ Tregs or CD4+Foxp3 Tconvs is shown in the lower panel. (B) mRNA levels of Foxp3, Tnfrsf1b, and Tnfrsf1a in differentiated CD4+Foxp3+ Tregs were measured by RT-qPCR. Data are the mean ± SD (n = 3). (C) SPLs from DEREG-Scurfy mice and DEREG mice were fractionated to Foxp3+GFP+ normal Tregs (WT Tregs) and Foxp3GFP+ inactivated Tregs (Mut Tregs) by anti-mouse Foxp3 mAb. Representative TNFR2 expression levels in WT Tregs and Mut Tregs are shown as a histogram. (D) Naive T cells were prepared from DEREG-scurfy mice. TNFR2 expression levels of TGF-β–induced Tregs are shown. Bar graph data are the mean ± SD (n = 3).

FIGURE 2.

Relationship between Foxp3 and TNFR2 in Treg differentiation in vitro. (A) TGF-β–dependent differentiation of naive T cells into CD4+Foxp3+ Tregs was analyzed by FCM. TNFR2 expression levels on CD4+Foxp3+ Tregs or CD4+Foxp3 Tconvs is shown in the lower panel. (B) mRNA levels of Foxp3, Tnfrsf1b, and Tnfrsf1a in differentiated CD4+Foxp3+ Tregs were measured by RT-qPCR. Data are the mean ± SD (n = 3). (C) SPLs from DEREG-Scurfy mice and DEREG mice were fractionated to Foxp3+GFP+ normal Tregs (WT Tregs) and Foxp3GFP+ inactivated Tregs (Mut Tregs) by anti-mouse Foxp3 mAb. Representative TNFR2 expression levels in WT Tregs and Mut Tregs are shown as a histogram. (D) Naive T cells were prepared from DEREG-scurfy mice. TNFR2 expression levels of TGF-β–induced Tregs are shown. Bar graph data are the mean ± SD (n = 3).

Close modal

R2agoTNF is a mutant in which nine amino acids of the TNFR2-binding region of human TNF-α have been replaced resulting in selective TNFR2 binding and agonistic activity (Fig. 3A). The concentration-dependent cytotoxicity of mTNFR2/Fas preadipocytes (30) showed that R2agoTNF activated signaling via TNFR2 similar to mouse TNF-α (Fig. 3B). We examined the Treg proliferation activity of R2agoTNF ex vivo. CD4+CD25+ Tregs and CD4+CD25 Tconvs prepared from mouse lymphocytes were labeled with CFSE and stimulated by R2agoTNF or human IL-2 with mouse anti-CD3 mAb. Proliferation rates of CD4+CD25+ Tregs and CD4+CD25 Tconvs were measured by CFSE division (Fig. 3C). The mean proliferation rate indicated that human IL-2 and R2agoTNF significantly expanded CD4+CD25+ Tregs (Fig. 3D, left). Tregs were further expanded by a combination of R2agoTNF and IL-2. Although CD4+CD25 Tconvs were significantly increased by IL-2, the expansion of Tconvs was not induced by R2agoTNF stimulation (Fig. 3D, right). In addition, Tconvs were not expanded by a combination of R2agoTNF and IL-2 compared with IL-2 stimulation. Therefore, R2agoTNF was confirmed to selectively expand Tregs. To verify the accuracy of Treg expansion by TNFR2 agonist, Tregs were stimulated with IL-2 or R2agoTNF in the presence of anti-human IL-2 mAb (Fig. 3E). Treg expansion by IL-2, but not R2agoTNF, was decreased when they were cultured with anti–IL-2 mAb. This data showed that Tregs were induced to proliferate by R2agoTNF, but not by mouse IL-2 derived from Tconvs, which were contaminating cells during Treg isolation. IL-2, but not R2agoTNF, expanded TNFR2–/– Tregs (Fig. 3F). These results indicate that R2agoTNF expands Tregs and enhances the proliferation of CD4+CD25+ Tregs by a concentration-dependent mechanism (Fig. 4A). Anti-inflammatory cytokines such as IL-10 and TGF-β secreted from CD4+CD25+ Tregs after stimulation with IL-2, R2agoTNF, or a combination of IL-2 and R2agoTNF were measured by intracellular staining (Fig. 4B). The mean positive-cell rate demonstrated that R2agoTNF enhanced the production of IL-10 and TGF-β from Tregs similar to that of IL-2 (Fig. 4C, 4D). The combination of IL-2 and R2agoTNF effectively enhanced the production of anti-inflammatory cytokines from Tregs. Increased expressions of CD25 and CTLA-4 were also assessed when Tregs were stimulated by IL-2 or R2agoTNF. Both IL-2 and R2agoTNF significantly enhanced CD25 expression on Tregs ex vivo and acted in an additive manner (Fig. 4E). Increased CTLA-4 expression on Tregs was also observed by stimulation with R2agoTNF (Fig. 4F).

FIGURE 3.

Ex vivo Treg expansion activity of R2agoTNF. (A) Ligand-receptor binding structure between R2agoTNF and TNFR2 was modeled by MODELER. Spheres indicate nine mutated amino acids (aa 29, 31–33, 143, 145–147, and 149) of R2agoTNF, which are in the TNFR2-binding region. (B) Signal transduction activity of R2agoTNF was measured by cytotoxic assay using mouse TNFR2/Fas overexpressing preadipocytes. Mouse TNF-α was used as a positive control. Data are the mean ± SD (n = 3). (C) Proliferation of CFSE-labeled CD4+CD25+ Tregs and CD4+CD25 Tconvs isolated from lymphocytes of WT mice was measured by cell proliferation assay. Representative proliferations are shown as a histogram. (D) Proliferation rates of CD4+CD25+ Tregs and CD4+CD25 Tconvs ex vivo are shown. Cells were stimulated by IL-2, R2agoTNF, and a combination of IL-2 and R2agoTNF, with or without anti-mouse CD3 mAb. (E) Proliferation rates of CD4+CD25+ Tregs cultured with anti-human IL-2 mAb are shown. (F) Proliferation rates of CD4+CD25+ TNFR2–/– Tregs by stimulation with IL-2 or R2agoTNF are shown. Data are the mean ± SD (n = 3). *p < 0.05, **p < 0.01, ***p < 0.001 (one-way ANOVA with Tukey multiple comparisons test).

FIGURE 3.

Ex vivo Treg expansion activity of R2agoTNF. (A) Ligand-receptor binding structure between R2agoTNF and TNFR2 was modeled by MODELER. Spheres indicate nine mutated amino acids (aa 29, 31–33, 143, 145–147, and 149) of R2agoTNF, which are in the TNFR2-binding region. (B) Signal transduction activity of R2agoTNF was measured by cytotoxic assay using mouse TNFR2/Fas overexpressing preadipocytes. Mouse TNF-α was used as a positive control. Data are the mean ± SD (n = 3). (C) Proliferation of CFSE-labeled CD4+CD25+ Tregs and CD4+CD25 Tconvs isolated from lymphocytes of WT mice was measured by cell proliferation assay. Representative proliferations are shown as a histogram. (D) Proliferation rates of CD4+CD25+ Tregs and CD4+CD25 Tconvs ex vivo are shown. Cells were stimulated by IL-2, R2agoTNF, and a combination of IL-2 and R2agoTNF, with or without anti-mouse CD3 mAb. (E) Proliferation rates of CD4+CD25+ Tregs cultured with anti-human IL-2 mAb are shown. (F) Proliferation rates of CD4+CD25+ TNFR2–/– Tregs by stimulation with IL-2 or R2agoTNF are shown. Data are the mean ± SD (n = 3). *p < 0.05, **p < 0.01, ***p < 0.001 (one-way ANOVA with Tukey multiple comparisons test).

Close modal
FIGURE 4.

Ex vivo anti-inflammatory cytokine production and cell surface marker expression in Tregs by R2agoTNF stimulation. (A) R2agoTNF expanded mouse CD4+CD25+ Tregs by a concentration-dependent mechanism under CD3 stimulation. (B) Production of IL-10 and TGF-β in R2agoTNF-stimulated Tregs was measured by intracellular staining. Representative histograms of the expressions of IL-10 and TGF-β stimulated by IL-2, R2agoTNF, and a combination of IL-2 and R2agoTNF, with or without anti-mouse CD3 mAb are shown. Mean ex vivo expression rates of IL-10 (C) and TGF-β (D) in CD4+CD25+ Tregs were calculated. Expression levels of CD25 (E) and CTLA-4 (F) in R2agoTNF-stimulated Tregs are shown. Data are the mean ± SD (n = 3). *p < 0.05, **p < 0.01, ***p < 0.001 (one-way ANOVA with Tukey multiple comparisons test).

FIGURE 4.

Ex vivo anti-inflammatory cytokine production and cell surface marker expression in Tregs by R2agoTNF stimulation. (A) R2agoTNF expanded mouse CD4+CD25+ Tregs by a concentration-dependent mechanism under CD3 stimulation. (B) Production of IL-10 and TGF-β in R2agoTNF-stimulated Tregs was measured by intracellular staining. Representative histograms of the expressions of IL-10 and TGF-β stimulated by IL-2, R2agoTNF, and a combination of IL-2 and R2agoTNF, with or without anti-mouse CD3 mAb are shown. Mean ex vivo expression rates of IL-10 (C) and TGF-β (D) in CD4+CD25+ Tregs were calculated. Expression levels of CD25 (E) and CTLA-4 (F) in R2agoTNF-stimulated Tregs are shown. Data are the mean ± SD (n = 3). *p < 0.05, **p < 0.01, ***p < 0.001 (one-way ANOVA with Tukey multiple comparisons test).

Close modal

TNFR2 was reported to be expressed on mouse and human Tregs (1215, 37). To investigate the potential of TNFR2 as a target molecule for Treg expansion, we confirmed the cytokine receptor profiles including TNFR2, TNFR1, GITR, CTLA-4, BAFFR, and OX-40 on CD4+Foxp3+ Tregs, CD4+Foxp3 Tconvs, CD8+ cytotoxic T cells (CTLs), CD11c+ DCs, CD19+CD45R+ B cells, and NK1.1+ NK cells (Fig. 5A). TNFR2 has a similar expression profile to GITR and OX40, which are Treg-specific markers. TNFR2 was confirmed to be preferentially expressed on CD4+Foxp3+ Tregs. We measured the TNFR2 expression on CD4+Foxp3+ Tregs from several LNs (Fig. 5B). TNFR2 was similarly expressed on Tregs derived from SPLs, inguinal lymphocytes, axillary lymphocytes, mesenteric lymphocytes, and PBLs. TNFR1 expression was low on Tregs from each lymph tissue. RT-qPCR showed that Foxp3 mRNA and Tnfrsf1b mRNA were highly expressed in CD4+Foxp3+ Tregs but were present at low levels in CD4+Foxp3 Tconvs (Fig. 5C). These results indicated that TNFR2 is preferentially expressed on Tregs.

FIGURE 5.

TNFR2 expression specificity in peripheral leukocyte subsets. (A) Expression levels of TNFR2, TNFR1, GITR, CTLA-4, BAFFR, and OX-40 in CD8+ CTLs, CD4+Foxp3+ Tregs, CD4+Foxp3 Tconvs, CD11c+ DCs, CD19+ CD45R+ B cells, and NK1.1+ cells fractionated from WT mouse lymphocytes were measured by FCM. Each cell population was gated as shown in the dot plots. (B) The protein levels of TNFR2 and TNFR1 in several lymphoid tissues from WT mice were compared by FCM. Data are the mean ± SD (n = 3). (C) mRNA levels of Foxp3, Tnfrsf1b, and Tnfrsf1a in mouse CD4+CD25+ Tregs and mouse CD4+CD25 Tconvs were measured by RT-qPCR. Data are the mean ± SD (n = 3). AXL, axillary lymphocyte; INL, inguinal lymphocyte; MEL, mesenteric lymphocyte.

FIGURE 5.

TNFR2 expression specificity in peripheral leukocyte subsets. (A) Expression levels of TNFR2, TNFR1, GITR, CTLA-4, BAFFR, and OX-40 in CD8+ CTLs, CD4+Foxp3+ Tregs, CD4+Foxp3 Tconvs, CD11c+ DCs, CD19+ CD45R+ B cells, and NK1.1+ cells fractionated from WT mouse lymphocytes were measured by FCM. Each cell population was gated as shown in the dot plots. (B) The protein levels of TNFR2 and TNFR1 in several lymphoid tissues from WT mice were compared by FCM. Data are the mean ± SD (n = 3). (C) mRNA levels of Foxp3, Tnfrsf1b, and Tnfrsf1a in mouse CD4+CD25+ Tregs and mouse CD4+CD25 Tconvs were measured by RT-qPCR. Data are the mean ± SD (n = 3). AXL, axillary lymphocyte; INL, inguinal lymphocyte; MEL, mesenteric lymphocyte.

Close modal

To enhance the TNFR2 agonistic activity in vivo, we optimized the R2agoTNF structure by internal covalent cross-linking with a disulfide bond. Ban et al. (28) created a stable human TNF-α trimer by the covalent internal cross-linking of three TNF-α monomers by a double mutation S95C/G148C to enhance its signal transduction activity. We applied the S95C/G148C replacement to R2agoTNF (Fig. 6A). These mutations were inserted into the pYas-R2agoTNF plasmid vector by a repeating point mutation. This R2agoTNF derivative, termed R2agoTNF-DS, was expressed in E. coli BL21(DE3) and recovered from the soluble fraction. R2agoTNF-DS was purified by 6×His-tag affinity chromatography and SEC. SEC results showed that the retention time of R2agoTNF-DS was similar to that of R2agoTNF (Fig. 6B). R2agoTNF and R2agoTNF-DS were detected by Western blotting as an ?17-kDa protein under reduced conditions with 2-ME, indicating a monomeric structure (Fig. 6C). However, only R2agoTNF-DS was detected at a high molecular mass under unreduced conditions without 2-ME, indicating a trimeric structure. The presence of disulfide bonds that cross-linked between R2agoTNF monomers was confirmed. These results demonstrated that R2agoTNF-DS was generated in the E. coli expression system. To evaluate the enhanced molecular stability by internal cross-linking, a thermal shift assay was performed (Fig. 6D). The thermal peak of R2agoTNF-DS was shifted to a higher temperature compared with R2agoTNF. The Tm values (°C) of R2agoTNFa and R2agoTNF-DS were 73.9 ± 1.5 and 87.7 ± 1.5, respectively. These results indicated that internal cross-linking enhanced molecular stability. The CFSE proliferation assay was performed to examine whether internal cross-linking promoted Treg expansion ex vivo (Fig. 6E). R2agoTNF-DS enhanced the proliferation of CD4+CD25+ Tregs compared with R2agoTNF. The internal cross-linking of R2agoTNF improved its TNFR2 agonistic activity.

FIGURE 6.

Structure modification of R2agoTNF by internal disulfide cross-linking. (A) Primary amino acid sequence of R2agoTNF-DS. S95C/G148C double mutations and 6×His-tag inserted by PCR are shown in bold font. (B) Molecular masses of R2agoTNF and R2agoTNF-DS were compared using SEC. (C) Structural multimerization of R2agoTNF and R2agoTNF-DS were estimated by Western blot analysis under reduced (with 2-ME) or unreduced (without 2-ME) conditions. (D) Thermal stabilities of R2agoTNF and R2agoTNF-DS were compared from thermal peaks and Tm values (°C) by thermal shift assay. (E) Ex vivo proliferation of CFSE-labeled CD4+CD25+ Tregs stimulated by R2agoTNF or R2agoTNF-DS was measured. Representative proliferation histograms (right) and mean proliferation rates (left) of each group are shown. Data are the mean ± SD (n = 3). (F) R2agoTNF-DS (50 μg per mouse) or saline was administered to WT mice twice a day for 5 d. Saline was administered as a negative control. (G) CD4+CD25+ Tregs, CD4+CD25 Tconvs, CD8+ T cells, and granulocytes were gated from lymphocytes in each group using FCM. Representative gating strategies and cell populations are shown. (H) The mean population rates of CD4+Foxp3+ Tregs, CD4+Foxp3 Tconvs, CD8+ T cells, and granulocytes were calculated from FCM data. Data are the mean ± SD (n = 3). *p < 0.05 (unpaired Student t test).

FIGURE 6.

Structure modification of R2agoTNF by internal disulfide cross-linking. (A) Primary amino acid sequence of R2agoTNF-DS. S95C/G148C double mutations and 6×His-tag inserted by PCR are shown in bold font. (B) Molecular masses of R2agoTNF and R2agoTNF-DS were compared using SEC. (C) Structural multimerization of R2agoTNF and R2agoTNF-DS were estimated by Western blot analysis under reduced (with 2-ME) or unreduced (without 2-ME) conditions. (D) Thermal stabilities of R2agoTNF and R2agoTNF-DS were compared from thermal peaks and Tm values (°C) by thermal shift assay. (E) Ex vivo proliferation of CFSE-labeled CD4+CD25+ Tregs stimulated by R2agoTNF or R2agoTNF-DS was measured. Representative proliferation histograms (right) and mean proliferation rates (left) of each group are shown. Data are the mean ± SD (n = 3). (F) R2agoTNF-DS (50 μg per mouse) or saline was administered to WT mice twice a day for 5 d. Saline was administered as a negative control. (G) CD4+CD25+ Tregs, CD4+CD25 Tconvs, CD8+ T cells, and granulocytes were gated from lymphocytes in each group using FCM. Representative gating strategies and cell populations are shown. (H) The mean population rates of CD4+Foxp3+ Tregs, CD4+Foxp3 Tconvs, CD8+ T cells, and granulocytes were calculated from FCM data. Data are the mean ± SD (n = 3). *p < 0.05 (unpaired Student t test).

Close modal

Inducing Treg proliferation in vivo is considered an important strategy for the treatment of diseases. We examined whether R2agoTNF-DS promoted Treg expansion in vivo. R2agoTNF-DS or saline (control) were administered to WT mice twice a day for 5 d (Fig. 6F). After this period, the populations of CD4+Foxp3+ Treg, CD4+Foxp3 Tconvs, CD8+ T cells, and granulocytes in mouse lymphocytes were measured (Fig. 6G). Numbers of CD4+Foxp3+ Tregs in R2agoTNF-DS–administered mice were significantly increased compared with those in saline administered mice (Fig. 6H). However, no increase in CD4+Foxp3 Tconvs, CD8+ T cells, and granulocytes was observed in R2agoTNF-DS–administered mice compared with saline administered mice. Therefore, R2agoTNF-DS preferentially expanded Tregs in vivo.

We structurally optimized R2agoTNF by IgG-Fc fusion to improve its activity in vivo. We previously reported that IgG-Fc fusion of a single-chain TNF-α mutant improved its molecular stability and prolonged the half-life (38, 39). According to this strategy, we created a single-chain R2agoTNF followed by IgG-Fc fusion and named it scR2agoTNF-Fc. scR2agoTNF-Fc cDNA was transfected into Expi293F cells, and then cells were cultured for protein expression for 1 wk. Protein was purified from supernatants by protein A affinity chromatography followed by SEC. The peak of scR2agoTNF-Fc was detected at an optimal molecular mass by SEC (Fig. 7A). Cytotoxic assay using mTNFR2/Fas preadipocytes showed that scR2agoTNF-Fc induced higher cell death activity compared with mouse TNF-α (Fig. 7B). This result indicated that scR2agoTNF-Fc, which has a bivalent structure, had enhanced agonistic activity via TNFR2. Furthermore, the TNFR2 agonist had low cytotoxicity because scR2agoTNF-Fc did not induce cell death in CD4+CD25+ Tregs (Fig. 7C) Then, we examined the Treg expansion activity of scR2agoTNF-Fc in vivo (Fig. 7D). scR2agoTNF-Fc was administered to WT mice twice a week for 1, 2, and 4 wk. After the administration period, populations of CD4+Foxp3+ Tregs, CD4+Foxp3 Tconvs, and CD8+ T cells in LNs were measured (Fig. 7E). Numbers of CD4+Foxp3+ Tregs in scR2agoTNF-Fc administered mice were significantly increased compared with saline administration mice (Fig. 7E, upper). No increase in the numbers of CD4+Foxp3 Tconvs and CD8+ T cells in scR2agoTNF-Fc administered mice was observed. To investigate the effect of scR2agoTNF-Fc on Treg suppressive activity, IL-10 production was measured from CD4+Foxp3+ Tregs, CD4+Foxp3 Tconvs, and CD8+ T cells by intracellular staining. Although a population of IL-10 positive cells was increased in CD4+Foxp3+ Tregs, it was not increased in CD4+Foxp3 Tconvs or CD8+ T cells (Fig. 7F). scR2agoTNF-Fc expanded Tregs dose dependently in vivo (Fig. 7G). Then, to evaluate the effects of long-term administration, Treg populations in LNs were measured after treatment with scR2agoTNF-Fc at 1, 2, and 4 wk. Numbers of CD4+Foxp3+ Tregs in scR2agoTNF-Fc–treated mice were maintained at higher levels compared with those saline-treated mice (Fig. 7H). Furthermore, CTLA-4 was significantly upregulated on CD4+Foxp3+ Tregs for 4 wk (Fig. 7I). No body-weight loss was observed by scR2agoTNF-Fc treatment for 4 wk. scR2agoTNF-Fc significantly and selectively enhanced the numbers and suppressive activity of Tregs in vivo.

FIGURE 7.

Molecular characterization and in vivo Treg expansion activity of scR2agoTNF-Fc. (A) The SEC chart indicates the molecular mass of scR2agoTNF-Fc. (B) Signal transduction activity of scR2agoTNF-Fc via TNFR2 was measured by cytotoxic assay using mouse TNFR2/Fas overexpressing preadipocytes. Mouse TNF-α was used as a positive control. Data are the mean ± SD (n = 3). (C) Cytotoxicity of scR2agoTNF-Fc on CD4+CD25+ Tregs was evaluated. (D) To evaluate their expansion and immunosuppressive activities, scR2agoTNF-Fc (50 μg per mouse) or saline was administered i.p. to WT mice twice a week for 1, 2, and 4 wk. Saline was administered as a negative control. (E) The population rates of CD4+Foxp3+ Tregs, CD4+Foxp3Tconvs, and CD8+ T cells in LNs of mice after a week were measured by FCM and plotted in a scatter graph. Data are the mean ± SD (n = 6). (F) In vivo IL-10 expression rates of CD4+Foxp3+ Tregs, CD4+Foxp3 Tconvs, and CD8+ T cells in LNs were measured after a week (n = 6). (G) scR2agoTNF-Fc (0.5, 5, 50 μg per mouse, respectively) was administered i.p. twice a week to WT mice. One week later, the numbers of CD4+Foxp3+ Tregs in LNs were measured. (H) Numbers and (I) CTLA-4 expression of CD4+Foxp3+ Tregs induced by the administration of scR2agoTNF-Fc (50 μg per mouse i.p., twice a week) for 1, 2, and 4 wk were assessed. **p < 0.01, ***p < 0.001 (unpaired Student t test).

FIGURE 7.

Molecular characterization and in vivo Treg expansion activity of scR2agoTNF-Fc. (A) The SEC chart indicates the molecular mass of scR2agoTNF-Fc. (B) Signal transduction activity of scR2agoTNF-Fc via TNFR2 was measured by cytotoxic assay using mouse TNFR2/Fas overexpressing preadipocytes. Mouse TNF-α was used as a positive control. Data are the mean ± SD (n = 3). (C) Cytotoxicity of scR2agoTNF-Fc on CD4+CD25+ Tregs was evaluated. (D) To evaluate their expansion and immunosuppressive activities, scR2agoTNF-Fc (50 μg per mouse) or saline was administered i.p. to WT mice twice a week for 1, 2, and 4 wk. Saline was administered as a negative control. (E) The population rates of CD4+Foxp3+ Tregs, CD4+Foxp3Tconvs, and CD8+ T cells in LNs of mice after a week were measured by FCM and plotted in a scatter graph. Data are the mean ± SD (n = 6). (F) In vivo IL-10 expression rates of CD4+Foxp3+ Tregs, CD4+Foxp3 Tconvs, and CD8+ T cells in LNs were measured after a week (n = 6). (G) scR2agoTNF-Fc (0.5, 5, 50 μg per mouse, respectively) was administered i.p. twice a week to WT mice. One week later, the numbers of CD4+Foxp3+ Tregs in LNs were measured. (H) Numbers and (I) CTLA-4 expression of CD4+Foxp3+ Tregs induced by the administration of scR2agoTNF-Fc (50 μg per mouse i.p., twice a week) for 1, 2, and 4 wk were assessed. **p < 0.01, ***p < 0.001 (unpaired Student t test).

Close modal

To assess the potent suppressive activity of the TNFR2 agonist via the expansion of Tregs, the anti-inflammatory effects of scR2agoTNF-Fc were confirmed using a skin-contact hypersensitivity model (Fig. 8A). The prophylactic administration of scR2agoTNF-Fc before DNFB challenge suppressed inflammation (Fig. 8B) and swelling (Fig. 8C) in the ears. IFN-γ production by effector T cells in the regional LNs was decreased in scR2agoTNF-Fc–treated mice in a dose-dependent manner compared with saline-treated mice (Fig. 8D). Furthermore, the percentage of CD4+Foxp3+ T cells in the regional LNs was increased by the administration of scR2agoTNF-Fc (Fig. 8E).

FIGURE 8.

Effects of scR2agoTNF-Fc in DNFB-sensitized contact hypersensitivity mice. (A) BALB/c mice were sensitized with 0.5% DNFB painting of the skin (n = 5). scR2agoTNF-Fc (5 and 50 μg per mouse i.p., respectively) was administered twice before DNFB challenge as described in the schedule chart. (B) Histology of the ear with H&E staining (scale bar, 100 μm) and (C) the degree of ear swelling were assessed. (D) IFN-γ production from CD4+ T cells and (E) numbers of CD4+Foxp3+ T cells in the regional LNs were measured 24 h after DNFB challenge. Data are the mean ± SD. *p < 0.05, **p < 0.01, ***p < 0.001 (one-way ANOVA with Tukey multiple comparisons test).

FIGURE 8.

Effects of scR2agoTNF-Fc in DNFB-sensitized contact hypersensitivity mice. (A) BALB/c mice were sensitized with 0.5% DNFB painting of the skin (n = 5). scR2agoTNF-Fc (5 and 50 μg per mouse i.p., respectively) was administered twice before DNFB challenge as described in the schedule chart. (B) Histology of the ear with H&E staining (scale bar, 100 μm) and (C) the degree of ear swelling were assessed. (D) IFN-γ production from CD4+ T cells and (E) numbers of CD4+Foxp3+ T cells in the regional LNs were measured 24 h after DNFB challenge. Data are the mean ± SD. *p < 0.05, **p < 0.01, ***p < 0.001 (one-way ANOVA with Tukey multiple comparisons test).

Close modal

In this study, we investigated the benefits of TNFR2 targeting for the proliferation and activation of Tregs in mice and assessed Treg expansion methods using a TNFR2 agonist. We demonstrated that TNFR2–/– mice had reduced numbers of Tregs in the LNs and spleen compared with WT mice. TNFR2–/– Tregs also had a lower inhibitory effect on the proliferation of effector T cells compared with WT Tregs. In addition, we demonstrated that mouse TNF-α expanded CD4+CD25+ TNFR2+ Tregs, but not CD4+CD25+ TNFR2 Tregs, ex vivo. These findings indicated that Tregs required TNFR2 signaling to proliferate and exert their immunosuppressive activity. CD25 and CTLA-4 are regulated by Foxp3 in Tregs (6, 40). Although TNFR2 is considered an important molecule for the immunosuppressive function of Tregs, the mechanism of TNFR2 expression on Tregs is poorly understood. Genome-wide analysis using chromatin immunoprecipitation–PCR or chromatin immunoprecipitation sequencing reported that Foxp3 was associated with the expression of the TNFR2 gene, Tnfrsf1b (41, 42). We demonstrated that TGF-β increased the population of CD4+Foxp3+ Tregs in vitro by enhancing the cell differentiation of CD4+ naive T cells to CD4+Foxp3+ Tregs. Furthermore, TNFR2 expression was increased by a TGF-β concentration–dependent mechanism. This positive correlation was observed between transcripts of Foxp3 and TNFR2. Foxp3 might upregulate TNFR2 expression in iTregs. In addition, phenotype analysis using DEREG-Scurfy mice (32, 43), which harbor normal Foxp3-expressing Tregs and inactivated Foxp3–expressing Tregs, showed that normal Foxp3 is needed for the expression of TNFR2 in thymus-derived Tregs and peripherally derived Tregs. These results indicated that TNFR2 is a target receptor that controls Tregs.

We also reported a Treg expansion strategy using a novel TNFR2-selective agonist TNF-α mutant. Low-dose IL-2 administration to expand Tregs is used clinically to treat GvHD (4446), but the high-dose administration of IL-2 activates effector T cells and promotes the risk of dose-dependent adverse effects. Recently, Treg expanders to replace IL-2 have been investigated. TNF-α, a candidate molecule, expands Tregs in vitro. However, the administration of TNF-α in vivo for Treg expansion has disadvantages because TNFR1 agonism induces liver failure, hepatic failure, and systemic shock (47). Several groups have demonstrated TNFR2 agonism suppressed diseases such as rheumatoid arthritis, GvHD, or T1D in mouse models or human subjects via Treg expansion or activation. STAR2, a mouse TNF-based agonist of TNFR2, prolonged the survival and decreased the severity of disease in a GvHD mouse model (48). It also ameliorated arthritis in an rheumatoid arthritis mouse model (49). TNF07, a S95C/G148C double-mutated human TNF-α, induced the selective death of autoreactive CD8+ T cells in type-1 diabetic patients and expanded CD4+ Tregs (28). Our mouse TNFR2 agonist, R2agoTNF, specifically transmits signaling via mouse TNFR2. We investigated the Treg expansion effect of R2agoTNF ex vivo and in vivo. In vitro cell proliferation assays demonstrated that IL-2 expanded CD4+CD25+ Tregs and CD4+CD25 Tconvs. In contrast, R2agoTNF selectively expanded CD4+CD25+ Tregs with the upregulated expression of CD25 and CTLA-4 and increased the production of IL-10 and TGF-β. R2agoTNF is an in vitro Treg expander that enhances the production of anti-inflammatory cytokines.

Because R2agoTNF is a soluble protein derived from human TNF-α, it has poor in vivo stability and a half-life in blood of <24 h. In our preliminary in vivo experiment, CD4+Foxp3+ Tregs were barely expanded in WT mice after the administration of R2agoTNF (data not shown). We attempted the structural optimization of R2agoTNF to improve its agonistic activity and molecular stability. Covalent internal cross-linking of human soluble TNF-α by double mutations of TNF monomers, S95C/G148C, improved TNFR2 signaling (28). Based on this previous report, we generated R2agoTNF-DS, which is a S95C/G148C mutation of R2agoTNF with an internal disulfide bond, to enhance its in vivo stability and activity. Internal cross-linking complicated the structural folding of R2agoTNF-DS, but characterization by SEC and Western blotting indicated that R2agoTNF-DS was produced in the E. coli expression system. The results of thermal shift assay and Treg proliferation assay showed that R2agoTNF-DS had enhanced thermal stability and signal transduction activity via mouse TNFR2 signaling. Internal cross-linking dramatically improved the molecular characteristics of R2agoTNF. The in vivo administration of R2agoTNF-DS significantly expanded Tregs in the LNs of WT mice. Importantly, other leukocyte subsets such as granulocytes, CD8+ T cells, and CD4+ helper T cells were not expanded. It was considered that internal cross-linkage by disulfide bonds might be beneficial for the structural optimization of R2agoTNF.

We previously reported that a single-chain TNFR1-selective antagonistic TNF-α mutant with IgG-Fc fusion prolonged its half-life in blood (38, 39). Following this report, we attempted to create scR2agoTNF-Fc, a single-chain R2agoTNF fused to human IgG-Fc, as another structural optimization. TNF-α has two molecular forms, soluble TNF and membrane-bound TNF. It was reported that the oligomerization of TNFR2 by membrane-bound TNF alone fully activates its intracellular signaling pathways (50, 51). scR2agoTNF-Fc is thought to promote TNFR2 signaling by mimicking the membrane-bound form in contrast to R2agoTNF, because scR2agoTNF-Fc has a bivalent R2agoTNF domain formed by the IgG-Fc fusion. In vivo stabilization of scR2agoTNF-Fc reduced the frequency of administration required to expand CD4+Foxp3+ Tregs and increase IL-10 production in the LNs of WT mice. Moreover, scR2agoTNF-Fc efficiently maintained a population of Tregs and increased the expression of CTLA-4 in vivo over a long period. Potential toxicity such as body-weight loss was not observed. Skin-contact hypersensitivity reactions are classified as a type IV allergy response with delayed hypersensitivity reaction. DNFB-induced hypersensitivity is evoked by CD4+ Th1 lymphocytes that express IFN-γ+, a Th1 cytokine (52). Administration of scR2agoTNF-Fc prior to DNFB challenge suppressed the ear inflammation. The suppression of IFN-γ+ T cells by expanded Tregs might explain the anti-inflammatory effects of scR2agoTNF-Fc. These R2agoTNF derivatives, R2agoTNF-DS and scR2agoTNF-Fc, might exert potential therapeutic effects via Treg expansion in vivo.

In this study, we showed that Foxp3 upregulated TNFR2 expression and that TNFR2 is a favorable target molecule to expand Tregs. We also demonstrated that R2agoTNF and its derivatives efficiently expanded Tregs ex vivo and in vivo. These Treg expanders might be powerful tools to facilitate Treg function analysis by targeting mouse TNFR2. We previously reported the generation of a mutant TNF-α, R2-7, a selective agonist for human TNFR2 (33). Human Tregs have multiple fractions with different immunosuppressive activities including naive Tregs, effector Tregs, and non-Tregs (53). The selective expansion of CD4+CD45RAFoxp3high effector Tregs is thought to exert the maximal immunosuppressive effect for the treatment of immune diseases. Future studies should demonstrate whether R2-7 or its derivatives expand or activate human Tregs.

This work was supported by Japan Society for the Promotion of Science KAKENHI Grants JP18K14877, JP18H02699, and JP18K19567 and by the Takeda Science Foundation, the Mochida Memorial Foundation for Medical and Pharmaceutical Research, and Kobe Gakuin University Joint Research (B).

The online version of this article contains supplemental material.

Abbreviations used in this article:

DC

dendritic cell

DEREG

depletion of Tregs

DNFB

2,4-dinitrofluorobenzene

FCM

flow cytometry

GvHD

graft-versus-host disease

iTreg

inducible Treg

LN

lymph node

RT-qPCR

reverse transcription quantitative PCR

SEC

size-exclusion chromatography

SPL

splenocyte

Tconv

conventional T cell

Tm

melting temperature

TNFR2

TNF receptor type 2

Treg

regulatory T cell

WT

wild-type.

1
Di Ianni
,
M.
,
F.
Falzetti
,
A.
Carotti
,
A.
Terenzi
,
F.
Castellino
,
E.
Bonifacio
,
B.
Del Papa
,
T.
Zei
,
R. I.
Ostini
,
D.
Cecchini
, et al
.
2011
.
Tregs prevent GVHD and promote immune reconstitution in HLA-haploidentical transplantation.
Blood
117
:
3921
3928
.
2
Leclerc
,
M.
,
S.
Naserian
,
C.
Pilon
,
A.
Thiolat
,
G. H.
Martin
,
C.
Pouchy
,
C.
Dominique
,
Y.
Belkacemi
,
F.
Charlotte
,
S.
Maury
, et al
.
2016
.
Control of GVHD by regulatory T cells depends on TNF produced by T cells and TNFR2 expressed by regulatory T cells.
Blood
128
:
1651
1659
.
3
Todo
,
S.
,
K.
Yamashita
,
R.
Goto
,
M.
Zaitsu
,
A.
Nagatsu
,
T.
Oura
,
M.
Watanabe
,
T.
Aoyagi
,
T.
Suzuki
,
T.
Shimamura
, et al
.
2016
.
A pilot study of operational tolerance with a regulatory T-cell-based cell therapy in living donor liver transplantation.
Hepatology
64
:
632
643
.
4
Yamaguchi
,
T.
,
A.
Kishi
,
M.
Osaki
,
H.
Morikawa
,
P.
Prieto-Martin
,
K.
Wing
,
T.
Saito
,
S.
Sakaguchi
.
2013
.
Construction of self-recognizing regulatory T cells from conventional T cells by controlling CTLA-4 and IL-2 expression.
Proc. Natl. Acad. Sci. USA
110
:
E2116
E2125
.
5
Takahashi
,
T.
,
T.
Tagami
,
S.
Yamazaki
,
T.
Uede
,
J.
Shimizu
,
N.
Sakaguchi
,
T. W.
Mak
,
S.
Sakaguchi
.
2000
.
Immunologic self-tolerance maintained by CD25(+)CD4(+) regulatory T cells constitutively expressing cytotoxic T lymphocyte-associated antigen 4.
J. Exp. Med.
192
:
303
310
.
6
Wing
,
K.
,
Y.
Onishi
,
P.
Prieto-Martin
,
T.
Yamaguchi
,
M.
Miyara
,
Z.
Fehervari
,
T.
Nomura
,
S.
Sakaguchi
.
2008
.
CTLA-4 control over Foxp3+ regulatory T cell function.
Science
322
:
271
275
.
7
Chinen
,
T.
,
K.
Komai
,
G.
Muto
,
R.
Morita
,
N.
Inoue
,
H.
Yoshida
,
T.
Sekiya
,
R.
Yoshida
,
K.
Nakamura
,
R.
Takayanagi
,
A.
Yoshimura
.
2011
.
Prostaglandin E2 and SOCS1 have a role in intestinal immune tolerance.
Nat. Commun.
2
:
190
.
8
Izcue
,
A.
,
J. L.
Coombes
,
F.
Powrie
.
2009
.
Regulatory lymphocytes and intestinal inflammation.
Annu. Rev. Immunol.
27
:
313
338
.
9
Piconese
,
S.
,
B.
Valzasina
,
M. P.
Colombo
.
2008
.
OX40 triggering blocks suppression by regulatory T cells and facilitates tumor rejection.
J. Exp. Med.
205
:
825
839
.
10
Shimizu
,
J.
,
S.
Yamazaki
,
T.
Takahashi
,
Y.
Ishida
,
S.
Sakaguchi
.
2002
.
Stimulation of CD25(+)CD4(+) regulatory T cells through GITR breaks immunological self-tolerance.
Nat. Immunol.
3
:
135
142
.
11
Yamaguchi
,
T.
,
K.
Hirota
,
K.
Nagahama
,
K.
Ohkawa
,
T.
Takahashi
,
T.
Nomura
,
S.
Sakaguchi
.
2007
.
Control of immune responses by antigen-specific regulatory T cells expressing the folate receptor.
Immunity
27
:
145
159
.
12
Chen
,
X.
,
M.
Bäumel
,
D. N.
Männel
,
O. M.
Howard
,
J. J.
Oppenheim
.
2007
.
Interaction of TNF with TNF receptor type 2 promotes expansion and function of mouse CD4+CD25+ T regulatory cells.
J. Immunol.
179
:
154
161
.
13
Chen
,
X.
,
J. J.
Subleski
,
R.
Hamano
,
O. M.
Howard
,
R. H.
Wiltrout
,
J. J.
Oppenheim
.
2010
.
Co-expression of TNFR2 and CD25 identifies more of the functional CD4+FOXP3+ regulatory T cells in human peripheral blood.
Eur. J. Immunol.
40
:
1099
1106
.
14
Chen
,
X.
,
J. J.
Subleski
,
H.
Kopf
,
O. M.
Howard
,
D. N.
Männel
,
J. J.
Oppenheim
.
2008
.
Cutting edge: expression of TNFR2 defines a maximally suppressive subset of mouse CD4+CD25+FoxP3+ T regulatory cells: applicability to tumor-infiltrating T regulatory cells.
J. Immunol.
180
:
6467
6471
.
15
Chen
,
X.
,
X.
Wu
,
Q.
Zhou
,
O. M.
Howard
,
M. G.
Netea
,
J. J.
Oppenheim
.
2013
.
TNFR2 is critical for the stabilization of the CD4+Foxp3+ regulatory T. cell phenotype in the inflammatory environment.
J. Immunol.
190
:
1076
1084
.
16
Faustman
,
D.
,
M.
Davis
.
2010
.
TNF receptor 2 pathway: drug target for autoimmune diseases.
Nat. Rev. Drug Discov.
9
:
482
493
.
17
Medler
,
J.
,
H.
Wajant
.
2019
.
Tumor necrosis factor receptor-2 (TNFR2): an overview of an emerging drug target.
Expert Opin. Ther. Targets
23
:
295
307
.
18
Yang
,
S.
,
J.
Wang
,
D. D.
Brand
,
S. G.
Zheng
.
2018
.
Role of TNF-TNF receptor 2 signal in regulatory T cells and its therapeutic implications.
Front. Immunol.
9
:
784
.
19
Malek
,
T. R.
,
A.
Yu
,
L.
Zhu
,
T.
Matsutani
,
D.
Adeegbe
,
A. L.
Bayer
.
2008
.
IL-2 family of cytokines in T regulatory cell development and homeostasis.
J. Clin. Immunol.
28
:
635
639
.
20
Setoguchi
,
R.
,
S.
Hori
,
T.
Takahashi
,
S.
Sakaguchi
.
2005
.
Homeostatic maintenance of natural Foxp3(+) CD25(+) CD4(+) regulatory T cells by interleukin (IL)-2 and induction of autoimmune disease by IL-2 neutralization.
J. Exp. Med.
201
:
723
735
.
21
He
,
X.
,
S.
Landman
,
S. C.
Bauland
,
J.
van den Dolder
,
H. J.
Koenen
,
I.
Joosten
.
2016
.
A TNFR2-agonist facilitates high purity expansion of human low purity Treg cells.
PLoS One
11
:
e0156311
.
22
Okubo
,
Y.
,
T.
Mera
,
L.
Wang
,
D. L.
Faustman
.
2013
.
Homogeneous expansion of human T-regulatory cells via tumor necrosis factor receptor 2.
Sci. Rep.
3
:
3153
.
23
Ban
,
L.
,
J.
Zhang
,
L.
Wang
,
W.
Kuhtreiber
,
D.
Burger
,
D. L.
Faustman
.
2008
.
Selective death of autoreactive T cells in human diabetes by TNF or TNF receptor 2 agonism.
Proc. Natl. Acad. Sci. USA
105
:
13644
13649
.
24
Dopp
,
J. M.
,
T. A.
Sarafian
,
F. M.
Spinella
,
M. A.
Kahn
,
H.
Shau
,
J.
de Vellis
.
2002
.
Expression of the p75 TNF receptor is linked to TNF-induced NFkappaB translocation and oxyradical neutralization in glial cells.
Neurochem. Res.
27
:
1535
1542
.
25
Irwin
,
M. W.
,
S.
Mak
,
D. L.
Mann
,
R.
Qu
,
J. M.
Penninger
,
A.
Yan
,
F.
Dawood
,
W. H.
Wen
,
Z.
Shou
,
P.
Liu
.
1999
.
Tissue expression and immunolocalization of tumor necrosis factor-alpha in postinfarction dysfunctional myocardium.
Circulation
99
:
1492
1498
.
26
Tsakiri
,
N.
,
D.
Papadopoulos
,
M. C.
Denis
,
D. D.
Mitsikostas
,
G.
Kollias
.
2012
.
TNFR2 on non-haematopoietic cells is required for Foxp3+ Treg-cell function and disease suppression in EAE.
Eur. J. Immunol.
42
:
403
412
.
27
Fischer
,
R.
,
J.
Marsal
,
C.
Guttà
,
S. A.
Eisler
,
N.
Peters
,
J. R.
Bethea
,
K.
Pfizenmaier
,
R. E.
Kontermann
.
2017
.
Novel strategies to mimic transmembrane tumor necrosis factor-dependent activation of tumor necrosis factor receptor 2.
Sci. Rep.
7
:
6607
.
28
Ban
,
L.
,
W.
Kuhtreiber
,
J.
Butterworth
,
Y.
Okubo
,
E. S.
Vanamee
,
D. L.
Faustman
.
2015
.
Strategic internal covalent cross-linking of TNF produces a stable TNF trimer with improved TNFR2 signaling.
Mol. Cell. Ther.
3
:
7
.
29
Ando
,
D.
,
M.
Inoue
,
H.
Kamada
,
S.
Taki
,
T.
Furuya
,
Y.
Abe
,
K.
Nagano
,
Y.
Tsutsumi
,
S. I.
Tsunoda
.
2016
.
Creation of mouse TNFR2-selective agonistic TNF mutants using a phage display technique.
Biochem. Biophys. Rep.
7
:
309
315
.
30
Ando
,
D.
,
H.
Kamada
,
M.
Inoue
,
S.
Taki
,
T.
Furuya
,
Y.
Abe
,
K.
Nagano
,
Y.
Tsutsumi
,
S.
Tsunoda
.
2016
.
Generation of a sensitive TNFR2-specific murine assays system.
Pharmazie
71
:
235
237
.
31
Komatsu
,
N.
,
S.
Hori
.
2007
.
Full restoration of peripheral Foxp3+ regulatory T cell pool by radioresistant host cells in scurfy bone marrow chimeras.
Proc. Natl. Acad. Sci. USA
104
:
8959
8964
.
32
Ohkura
,
N.
,
M.
Hamaguchi
,
H.
Morikawa
,
K.
Sugimura
,
A.
Tanaka
,
Y.
Ito
,
M.
Osaki
,
Y.
Tanaka
,
R.
Yamashita
,
N.
Nakano
, et al
.
2012
.
T cell receptor stimulation-induced epigenetic changes and Foxp3 expression are independent and complementary events required for Treg cell development.
Immunity
37
:
785
799
.
33
Abe
,
Y.
,
T.
Yoshikawa
,
M.
Inoue
,
T.
Nomura
,
T.
Furuya
,
T.
Yamashita
,
K.
Nagano
,
H.
Nabeshi
,
Y.
Yoshioka
,
Y.
Mukai
, et al
.
2011
.
Fine tuning of receptor-selectivity for tumor necrosis factor-α using a phage display system with one-step competitive panning.
Biomaterials
32
:
5498
5504
.
34
Mikami
,
N.
,
R.
Kawakami
,
K. Y.
Chen
,
A.
Sugimoto
,
N.
Ohkura
,
S.
Sakaguchi
.
2020
.
Epigenetic conversion of conventional T cells into regulatory T cells by CD28 signal deprivation.
Proc. Natl. Acad. Sci. USA
117
:
12258
12268
.
35
Chen
,
W.
,
W.
Jin
,
N.
Hardegen
,
K. J.
Lei
,
L.
Li
,
N.
Marinos
,
G.
McGrady
,
S. M.
Wahl
.
2003
.
Conversion of peripheral CD4+CD25- naive T cells to CD4+CD25+ regulatory T cells by TGF-beta induction of transcription factor Foxp3.
J. Exp. Med.
198
:
1875
1886
.
36
Wan
,
Y. Y.
,
R. A.
Flavell
.
2007
.
‘Yin-Yang’ functions of transforming growth factor-beta and T regulatory cells in immune regulation.
Immunol. Rev.
220
:
199
213
.
37
Chen
,
X.
,
J. J.
Oppenheim
.
2011
.
The phenotypic and functional consequences of tumour necrosis factor receptor type 2 expression on CD4(+) FoxP3(+) regulatory T cells.
Immunology
133
:
426
433
.
38
Inoue
,
M.
,
D.
Ando
,
H.
Kamada
,
S.
Taki
,
M.
Niiyama
,
Y.
Mukai
,
T.
Tadokoro
,
K.
Maenaka
,
T.
Nakayama
,
Y.
Kado
, et al
.
2017
.
A trimeric structural fusion of an antagonistic tumor necrosis factor-α mutant enhances molecular stability and enables facile modification.
J. Biol. Chem.
292
:
6438
6451
.
39
Inoue
,
M.
,
Y.
Tsuji
,
C.
Yoshimine
,
S.
Enomoto
,
Y.
Morita
,
N.
Osaki
,
M.
Kunishige
,
M.
Miki
,
S.
Amano
,
K.
Yamashita
, et al
.
2020
.
Structural optimization of a TNFR1-selective antagonistic TNFα mutant to create new-modality TNF-regulating biologics.
J. Biol. Chem.
295
:
9379
9391
.
40
Hori
,
S.
,
T.
Nomura
,
S.
Sakaguchi
.
2003
.
Control of regulatory T cell development by the transcription factor Foxp3.
Science
299
:
1057
1061
.
41
Arvey
,
A.
,
J.
van der Veeken
,
G.
Plitas
,
S. S.
Rich
,
P.
Concannon
,
A. Y.
Rudensky
.
2015
.
Genetic and epigenetic variation in the lineage specification of regulatory T cells.
Elife
4
:
e07571
.
42
Sadlon
,
T. J.
,
B. G.
Wilkinson
,
S.
Pederson
,
C. Y.
Brown
,
S.
Bresatz
,
T.
Gargett
,
E. L.
Melville
,
K.
Peng
,
R. J.
D’Andrea
,
G. G.
Glonek
, et al
.
2010
.
Genome-wide identification of human FOXP3 target genes in natural regulatory T cells.
J. Immunol.
185
:
1071
1081
.
43
Lahl
,
K.
,
C.
Loddenkemper
,
C.
Drouin
,
J.
Freyer
,
J.
Arnason
,
G.
Eberl
,
A.
Hamann
,
H.
Wagner
,
J.
Huehn
,
T.
Sparwasser
.
2007
.
Selective depletion of Foxp3+ regulatory T cells induces a scurfy-like disease.
J. Exp. Med.
204
:
57
63
.
44
Koreth
,
J.
,
K.
Matsuoka
,
H. T.
Kim
,
S. M.
McDonough
,
B.
Bindra
,
E. P.
Alyea
III
,
P.
Armand
,
C.
Cutler
,
V. T.
Ho
,
N. S.
Treister
, et al
.
2011
.
Interleukin-2 and regulatory T cells in graft-versus-host disease.
N. Engl. J. Med.
365
:
2055
2066
.
45
Matsuoka
,
K.
,
J.
Koreth
,
H. T.
Kim
,
G.
Bascug
,
S.
McDonough
,
Y.
Kawano
,
K.
Murase
,
C.
Cutler
,
V. T.
Ho
,
E. P.
Alyea
, et al
.
2013
.
Low-dose interleukin-2 therapy restores regulatory T cell homeostasis in patients with chronic graft-versus-host disease.
Sci. Transl. Med.
5
:
179ra43
.
46
Tahvildari
,
M.
,
R.
Dana
.
2019
.
Low-dose IL-2 therapy in transplantation, autoimmunity, and inflammatory diseases.
J. Immunol.
203
:
2749
2755
.
47
Leist
,
M.
,
F.
Gantner
,
S.
Jilg
,
A.
Wendel
.
1995
.
Activation of the 55 kDa TNF receptor is necessary and sufficient for TNF-induced liver failure, hepatocyte apoptosis, and nitrite release.
J. Immunol.
154
:
1307
1316
.
48
Chopra
,
M.
,
M.
Biehl
,
T.
Steinfatt
,
A.
Brandl
,
J.
Kums
,
J.
Amich
,
M.
Vaeth
,
J.
Kuen
,
R.
Holtappels
,
J.
Podlech
, et al
.
2016
.
Exogenous TNFR2 activation protects from acute GvHD via host T reg cell expansion.
J. Exp. Med.
213
:
1881
1900
.
49
Lamontain
,
V.
,
T.
Schmid
,
D.
Weber-Steffens
,
D.
Zeller
,
Z.
Jenei-Lanzl
,
H.
Wajant
,
R. H.
Straub
,
D. N.
Männel
.
2019
.
Stimulation of TNF receptor type 2 expands regulatory T cells and ameliorates established collagen-induced arthritis in mice.
Cell. Mol. Immunol.
16
:
65
74
.
50
Grell
,
M.
,
E.
Douni
,
H.
Wajant
,
M.
Löhden
,
M.
Clauss
,
B.
Maxeiner
,
S.
Georgopoulos
,
W.
Lesslauer
,
G.
Kollias
,
K.
Pfizenmaier
,
P.
Scheurich
.
1995
.
The transmembrane form of tumor necrosis factor is the prime activating ligand of the 80 kDa tumor necrosis factor receptor.
Cell
83
:
793
802
.
51
Rauert
,
H.
,
A.
Wicovsky
,
N.
Müller
,
D.
Siegmund
,
V.
Spindler
,
J.
Waschke
,
C.
Kneitz
,
H.
Wajant
.
2010
.
Membrane tumor necrosis factor (TNF) induces p100 processing via TNF receptor-2 (TNFR2).
J. Biol. Chem.
285
:
7394
7404
.
52
Cher
,
D. J.
,
T. R.
Mosmann
.
1987
.
Two types of murine helper T cell clone. II. Delayed-type hypersensitivity is mediated by TH1 clones.
J. Immunol.
138
:
3688
3694
.
53
Miyara
,
M.
,
Y.
Yoshioka
,
A.
Kitoh
,
T.
Shima
,
K.
Wing
,
A.
Niwa
,
C.
Parizot
,
C.
Taflin
,
T.
Heike
,
D.
Valeyre
, et al
.
2009
.
Functional delineation and differentiation dynamics of human CD4+ T cells expressing the FoxP3 transcription factor.
Immunity
30
:
899
911
.

The authors have no financial conflicts of interest.

Supplementary data