CD25 and Protein Phosphatase 2A Cooperate to Enhance IL-2R Signaling in Human Regulatory T Cells

This information is current as Ying Ding, Aixin Yu, George C. Tsokos and Thomas R. of September 30, 2021. Malek J Immunol published online 13 May 2019 http://www.jimmunol.org/content/early/2019/05/10/jimmun ol.1801570 Downloaded from

Supplementary http://www.jimmunol.org/content/suppl/2019/05/10/jimmunol.180157 Material 0.DCSupplemental http://www.jimmunol.org/ Why The JI? Submit online.

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

• No Triage! Every submission reviewed by practicing scientists

• Fast Publication! 4 weeks from acceptance to publication

by guest on September 30, 2021 *average

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

The Journal of Immunology is published twice each month by The American Association of Immunologists, Inc., 1451 Rockville Pike, Suite 650, Rockville, MD 20852 Copyright © 2019 by The American Association of Immunologists, Inc. All rights reserved. Print ISSN: 0022-1767 Online ISSN: 1550-6606. Published May 13, 2019, doi:10.4049/jimmunol.1801570 The Journal of Immunology

CD25 and Protein Phosphatase 2A Cooperate to Enhance IL-2R Signaling in Human Regulatory T Cells

Ying Ding,* Aixin Yu,* George C. Tsokos,† and Thomas R. Malek*,‡

Low-dose IL-2 therapy is a direct approach to boost regulatory T cells (Tregs) and promote immune tolerance in autoimmune patients. However, the mechanisms responsible for selective response of Tregs to low-dose IL-2 is not fully understood. In this study we directly assessed the contribution of CD25 and protein phosphatase 2A (PP2A) in promoting IL-2R signaling in Tregs. IL-2– induced tyrosine phosphorylation of STAT5 (pSTAT5) was proportional to CD25 levels on human CD4+ T cells and YT human NK cell line, directly demonstrating that CD25 promotes IL-2R signaling. Overexpression of the PP2A catalytic subunit (PP2Ac) by lentiviral transduction in human Tregs increased the level of IL-2R subunits and promoted tyrosine phosphorylation of Jak3 and STAT5. Interestingly, increased expression of CD25 only partially accounted for this enhanced activation of pSTAT5, indicating

that PP2A promotes IL-2R signaling through multiple mechanisms. Consistent with these findings, knockdown of PP2Ac in Downloaded from human Tregs and impaired PP2Ac activity in mouse Tregs significantly reduced IL-2–dependent STAT5 activation. In contrast, overexpression or knockdown of PP2Ac in human T effector cells did not affect IL-2–dependent pSTAT5 activation. Overexpres- sion of PP2Ac in human Tregs also increased the expressions of proteins related to survival, activation, and immunosuppressive function, and upregulated several IL-2–regulated genes. Collectively, these findings suggest that CD25 and PP2A cooperatively enhance the responsiveness of Tregs to IL-2, which provide potential therapeutic targets for low-dose IL-2 therapy. The Journal

of Immunology, 2019, 203: 000–000. http://www.jimmunol.org/

nterleukin-2 is a key cytokine that promotes immune re- immune system attacks self-tissues (5). Completed clinical trials sponses and is also essential for immune tolerance through its indicate that low-dose IL-2 therapy is safe, increases Tregs in most I action on Foxp3+ regulatory T cells (Tregs) (1). The reali- patients, and is accompanied by clinical benefit in patients with zation that low IL-2R signaling in mice effectively promotes Treg chronic graft-versus-host disease, hepatitis C virus–induced vas- development and homeostasis, but not T effector (Teff) responses culitis, alopecia areata, and systemic lupus erythematosus (6–9). (2), favors the concept that low amounts of IL-2 may selectively Low-dose IL-2 is in a range of 0.5–3 3 106 IU/m2, administered at boost Treg activity in the context of autoimmune diseases. Pre- various frequencies (from daily to biweekly). These levels of IL-2 clinical studies showed that low doses of IL-2 or agonist IL-2/ are ∼30- to 100-fold lower than used in cancer immunotherapy, by guest on September 30, 2021 anti–IL-2 complexes supported immune tolerance in the context of where the goal has been to boost Teff and NK cells. A critical diabetes-prone NOD mice, experimental autoimmune encephalo- aspect of low-dose IL-2 therapy in autoimmunity is that so far myelitis, and allogenic islet transplantation (3, 4). Low-dose IL-2 there has been no indication of activation of autoreactive Teff is now being advanced as a promising therapeutic approach in cells, although sometimes regulatory CD56hi NK cells and eo- patients with autoimmune diseases or other situations in which the sinophils increase (7, 10). IL-2 signaling is initiated by binding of IL-2 to the IL-2R, which is expressed on the cell surface as either the intermediate-affinity *Department of Microbiology and Immunology, Miller School of Medicine, Univer- IL-2R, a dimer of IL-2Rb (CD122) and gc (CD132), or the high- † sity of Miami, Miami, FL 33136; Department of Medicine, Beth Israel Deaconess affinity IL-2R, a trimer of IL-2Ra (CD25), IL- 2Rb and gc (11). Medical Center, Harvard Medical School, Boston, MA 02215; and ‡Diabetes Re- search Institute, Miller School of Medicine, University of Miami, Miami, FL 33136 Because IL-2 can stimulate both Tregs and autoreactive T cells, ORCIDs: 0000-0002-0557-8077 (A.Y.); 0000-0001-9589-2360 (G.C.T.). important considerations to advance this therapy are related to Received for publication November 30, 2018. Accepted for publication April 20, the window of selectivity of low-dose IL-2 toward Tregs and 2019. the mechanisms that impose this selectivity. In this regard, we This work was supported by National Institutes of Health (NIH) Grant R01 AI131648 previously showed that IL-2–dependent STAT5 activation and and the Diabetes Research Institute Foundation (to T.R.M.), and NIH Grants R01 downstream gene activation in human Tregs occurred at ∼10–15- AI068787 and R01 AI136924 (to G.C.T.). and 100-fold lower concentrations of IL-2, respectively, than in Y.D. designed, performed, and analyzed most of the experiments and wrote the CD4+ CD45RO+ T memory (Tm) cells (12), where the latter manuscript; A.Y. performed and analyzed experiments on CD4+ T cells and YT cell line; G.C.T. provided the Ppp2r1aflox/flox mice and edited the manuscript; represents a viable pharmacologic range to target Tregs. These T.R.M. conceived the project, designed the overall study, and wrote the paper. selective responses by human Tregs correlated with their higher Address correspondence and reprint requests to Dr. Thomas R. Malek, Department of expression of CD25 than CD4+ Tm cells (13). Indeed, in vitro Microbiology and Immunology, Miller School of Medicine, University of Miami, fully activated T cells exhibited over a 1000-fold range of re- 1600 NW 10th Avenue, Miami, FL 33136. E-mail address: [email protected] sponse to IL-2 as measured by pSTAT5 activation (13), supporting The online version of this article contains supplemental material. the notion that CD25 levels dictate the sensitivity of their re- Abbreviations used in this article: MFI, mean fluorescence intensity; PP2A, protein phosphatase 2A; PP2Aa, PP2A scaffold subunit; PP2Ac, PP2A catalytic subunit; sponses to IL-2. Nevertheless, in vitro activated human T cells SFM, OpTmizer CTS T cell expansion medium; shRNA, short hairpin RNA; Teff, remain less responsive to IL-2 than human Tregs, even though T effector; TEM, Teff memory; Tm, T memory; Treg, regulatory T cell; YFP, yellow the former expressed higher levels of all IL-2R subunits (12). fluorescent protein. These latter data suggest that other cell intrinsic factors, separate Copyright Ó 2019 by The American Association of Immunologists, Inc. 0022-1767/19/$37.50 from CD25 levels, contribute to the high IL-2 sensitivity of Tregs

www.jimmunol.org/cgi/doi/10.4049/jimmunol.1801570 2 CD25 AND PP2A IN IL-2R SIGNALING SENSITIVITY and that assessment of IL-2 responsiveness by a heterogeneous Human CD4+ T cells were enriched by negative selection with the MACS + population of activated T cells may not directly relate to differ- CD4 T Cell Isolation Kit II (Miltenyi Biotec, Auburn, CA). The purified CD4+ T cells were stained and sorted using a BD FACSAria into Tregs ential responses by Tregs and Teff cells. + hi lo + lo hi (CD4 CD25 CD127 ) and Teff memory (TEM) (CD4 CD25 CD127 Protein phosphatase 2A (PP2A) is a ubiquitously expressed, CD45RA2) cells. Sorted cells were typically .90% pure. Tregs were + 2 highly conserved serine/threonine phosphatase that contributes to $85% Foxp3 and TEM cells were $93% Foxp3 . Purified Tregs and Treg function as assessed by Treg-specific knockout of PP2A TEM cells were initially cultured with anti-CD3/CD28 beads (Dynabeads activity (14). PP2A consists of three subunits: a scaffold subunit Human Treg Expander; ThermoFisher Scientific, Vilnius, LT) and human IL-2 (500 U/ml) and then were subcultured with IL-2 (500 U/ml) in (PP2Aa), a catalytic subunit (PP2Ac), and a regulatory subunit OpTmizer CTS T cell expansion medium (designated as SFM) (Life (PP2Ab). The scaffold (a, PPP2R1A and b, PPP2R1B) and cat- Technologies, Grand Island, NY). The anti-CD3/anti-C28/IL-2–activated alytic (a, PPP2CA, and b, PPP2CB) subunits are each encoded by and expanded TEM cells are referred to as Teff cells. two homologous genes, with the a isoform being ∼10-fold more Plasmid construction abundant than the b isoform (15). In contrast, the regulatory subunits are coded by a large variety of genes that are grouped into The lentiviral vector pRRL-sin-cPPT-MCS-IRES-emdGFP (a gift from four families (B, B’, B’’, B’’’) (16). PP2Aa and PP2Ac form a Dr. N. Kasahara, University of Mimi, FL) containing a CMV promoter was used for overexpression of PP2Ac. The full coding sequences of human heterodimeric core enzyme, which further interacts with a variable PP2Ac was PCR amplified from PP2Ac-a (PPP2CA) (NM_002715) human PP2Ab to assemble into a functional heterotrimeric holoenzyme. untagged clone (no. SC321401, Origene, Rockville, MD) using the following PP2A regulates cell cycle progression, cell division and death, primers: PPP2CA-cloning-for 59- TATGGATCCATGGACGAGAAGGT- cytoskeleton dynamics, and multiple signaling pathways (17). GTTCACCAAGG-39; PPP2CA-cloning-rev 59- GTGTGGAATTCTTAC- AGGAAGTAGTCTGGGGTACGACG -39. They were then incorporated Several lines of evidence support the notion that regulation of Downloaded from using BamHI and EcoRI restriction enzymes into the pRRL-sin-cPPT- serine/threonine phosphorylation of the IL-2R or associated sig- MCS-IRES-emdGFP vector. For PP2Ac knockdown experiments, the naling machinery might also contribute to the IL-2–dependent pLKO.3G vector containing hU6 promoter (no. 14748, Addgene, Cam- responses of Tregs. Use of small molecule inhibitors of PP2A bridge, MA) was used. The oligonucleotides (PPP2CA–short hairpin RNA in the human YT NK-like cell line enhanced serine/threonine (shRNA)–sense 59-AATTGGATATTATTCAGTTGAAACACTCGAGTG- TTTCAACTGAATAATATCCTTTTTTTAT-39 and PPP2CA–shRNA–anti- phosphorylation of IL-2Rb, JAK3 and STAT5 to negatively reg- sense-59-AAAAAAAGGATATTATTCAGTTGAAACACTCGAGTGTTT- ulate IL-2 signaling (18). Blockade of PP2A with calyculin A, an CAACTGAATAATATCC-39) were annealed and inserted into pLKO.3G http://www.jimmunol.org/ inhibitor of protein phosphatase 1 (PP1) and PP2A, had a greater vector using EcoRI and PacI restriction enzymes, encoding the PP2Ac- effect on the sensitivity to IL-2 in human Tregs when compared specific shRNA transcript targeting 666–686 nt of the mRNA coding with CD4+ CD45RO+ Tm cells (12). However, the direct contri- region of PPP2CA. bution of PP2A in Treg sensitivity remains uncertain, as these Lentivirus generation and transduction inhibitors have off-target effects, and the effect of Treg-specific Lentivirus was generated in HEK293T cells. The lentiviral plasmids deletion of Ppp2r1a on Treg function in the mouse did not assess constructed for PP2Ac overexpression or knockdown (described above) effects on IL-2R signaling. were cotransfected with the pMDLGg/p packaging plasmid, the pRSV-REV In this study, we have directly examined the contribution of plasmid, and the pMD.G envelop plasmid (all kindly provided by CD25 levels and PP2A in responses by human Tregs to IL-2. By Dr. N. Kasahara, University of Miami, FL) (23–25) into H293T cells by guest on September 30, 2021 directly assessing Tregs and a cloned cell line, we find that their using calcium phosphate transfection. Viral supernatants were collected, filtered (0.45 mm), and titrated. Lentiviral transduction of expanded Tregs selective response to IL-2 is in part due to high CD25 levels. By and Teff cells was performed 24 h after activation with anti-CD3/CD28 evaluating human Tregs with genetically increased or reduced beads and IL-2. Lentiviruses (multiplicity of infection is ∼10) were added PP2Ac and mouse Tregs that lacked PP2Ac activity, we defined a in SFM at 500 ml/well and centrifuged at 2200 rpm for 90 min at 30˚C in the direct role for this phosphatase as another cell intrinsic mechanism 24-well plate precoated with 40 mg/ml RetroNectin (TaKaRa, Mountain View, CA). After centrifugation, Tregs and Teff cells (1 3 105 per well) in for the high IL-2R signaling sensitivity of Tregs. 500 ml SFM were added. Cells were cultured with anti-CD3/CD28 beads and 500 U/ml human IL-2 and analyzed 3 d later. Materials and Methods Abs and flow cytometry Mice Anti-PP2Ac-PE (1D6) was purchased from EMD Millipore (Billerica, MA). Ppp2r1aflox/floxFoxp3YFP-cre (PP2Aflox) mice were generated as previously The following mAbs (with the clone names in parentheses) were obtained Ppp2r1aflox/flox described (14). mice carrying the loxP sites across exons from BD Biosciences (San Jose, CA), BioLegend (San Diego, CA), or 5–6 of Ppp2r1a (strain: 017441) (19) and Foxp3YFP-cre mice expressing eBioscience (San Diego, CA): CD4 (SK3), CD8a (RPA-T8), CD127 (HIL- yellow fluorescent protein (YFP) and Cre recombinase via the Treg- 7R-M21), CD25 (PC61), CD122 (Mik-b3), CD132 (TUGh4), CD45RA specific gene Foxp3 (strain: 016959) (20) were obtained from The Jackson (HI100), CD45RO (UCHL1), ICOS/CD278 (DX29), CD69 (FN50), Laboratory. These mice were bred and maintained within the specific CD62L (DREG-56), CD39 (TU66), CD73 (AD2), Tigit (MBSA43), Foxp3 pathogen-free animal facility at the University of Miami. Both age- and (259D), Ki67 (B56), Bcl-2 (100), pSTAT5 (pY694), CD4 (GK1.5), CD8 sex- matched male and female mice at the ages of 6–8 wk were used for (53-6.7), CD25 (PC-61), and Foxp3 (FJK-16S). Cells were surface stained experiments. All animal studies were reviewed and approved by the In- with Abs in FACS buffer (HBSS, 0.2% BSA, 0.1% sodium azide) for stitutional Animal Care and Use Committee at the University of Miami. 15 min at 4˚C. Foxp3, Bcl-2, and Ki67 staining was performed after fixing Cell culture and permeabilizing using Foxp3/Transcription Factor Staining Buffer Set (eBioscience) according to the manufacturer’s instructions. For PP2Ac and Human NK cell line, CD25-expressing YT (21, 22), was maintained in pSTAT5 staining, cells were permeabilized using ice-cold 100% methanol (see RPMI 1640 medium (Invitrogen, Carlsbad, CA) with 5% FBS (Atlanta below). FACS analysis was performed using BD LSRFortessa or BD LSR II Biologicals, Flowery Branch, GA) and penicillin (50 U/ml)/streptomycin flow cytometers, where typically 100,000 events (for PBMCs and mouse splenic (50 mg/ml) (Life Technologies by ThermoFisher Scientific). Human em- cells) or 10,000 events (for YT and cultured Tregs and Teff cells) were collected. bryonic kidney 293T (HEK293T) adherent cells were grown in DMEM Data were analyzed using BD FACSDiva 8.0.1 software, where viable cells (Invitrogen) supplemented with 10% FBS and penicillin (50 U/ml)/ were gated based on forward versus side light scatter profiles, and doublets were streptomycin (50 U/ml). excluded based on forward light scatter area versus scatter width. PBMC isolation and Treg and Teff cell expansion pSTAT5 analysis Peripheral blood and buffy coat samples from healthy adult donors (20–45 y Human PBMCs, expanded human Tregs and Teff cells, and mouse old) were purchased from the Continental Blood Bank, Miami, FL. splenocytes were cultured in 1 ml RPMI complete medium (Tregs and Teff PBMCs were isolated as previously described on the day of collection (12). cells in SFM) for 1, 4, and 0.5 h, respectively. After this “rest culture,” IL-2 The Journal of Immunology 3

(human IL-2: aldesleukin [Proleukin], Novartis, East Hanover, NJ; murine proliferation of CD8+ cells (used as responder cells) was assessed by flow IL-2: eBioscience) was added for 15 min at 37˚C. pSTAT5 staining was cytometry as the dilution of CFSE or CellTrace violet. Percentage of inhi- performed as previously described (12). bition was calculated by [12(proliferation [responder and suppressor])/ proliferation (responder only)] 3 100. Western blot analysis Cell extracts were prepared using cell extraction buffer (ThermoFisher PP2A enzymatic activity assay Scientific, Rockford, IL) with protease inhibitors, phosSTOP, and 1mM PP2A enzymatic activity was assessed using PP2A Immunoprecipitation PMSF (All from Sigma-Aldrich). Immunoblotting was performed on 10% Phosphatase Assay Kit (EMD Millipore, Temecula, CA) following the SDS-PAGE gels after 200 mg total protein was loaded per lane as previ- manufacturer’s instructions. In brief, Tregs and TEM cells were lysed in ously described (26). The following Abs from Cell Signaling Technology cell extraction buffer (2 3 106 cells in 30 ml buffer). These protein (Danvers, CA) were used: anti-JAK3 (B32-32), anti-phosphorylated JAK3 extracts were incubated with immobilized anti-PP2Ac Ab (clone 1D6; (pJAK3) (Y980/981) (D44E3), anti-pSTAT5 (Y694) (C11C5), b-Tubulin EMD Millipore) and with threonine phosphopeptide (K-R-pT-I-R-R, (9F3), HRP-linked anti-rabbit IgG, HRP-linked anti-mouse IgG, and where pT indicates phosphorylated threonine) for 10 min at 30˚C in a HRP-linked anti-rat IgG. Purified anti-human PP2Aca was purchased shaking incubator. Supernatants (25 ml) from this reaction were then from BD Biosciences. Purified anti-mouse PPP2R1A (6F9) was ob- incubated with malachite green phosphate detection solution for col- tained from BioLegend. Proteins were visualized after incubation of the orimetric reaction in a 96-well plate. Released phosphates were calcu- blots with ECL chemiluminescence agent (ThermoFisher Scientific) and ana- lated from a standard curve generated using a phosphate standard lyzed using Odyssey Fc Dual Mode Imaging System (Li-COR Biosciences, solution. Lincoln, NE). Statistics Quantitative real-time PCR Graphical representations of the data and statistical analyses were per- cDNA was prepared with the High-Capacity cDNA Reverse Transcription formed using GraphPad Prism 7.0 software. Data are shown as mean 6 Downloaded from Kit using oligo(dT) primers (Life Technologies). The primer pairs for SEM. To determine the EC50 for the pSTAT5 dose-response studies, indicated targets are listed in Supplemental Table I. Real-time PCR was nonlinear regression analyses were used. A one-sample two-tailed t test done in triplicate using Power SYBR Green PCR Master Mix (Life and paired or unpaired two-tailed t test were used to calculate statistical Technologies), primers (0.15 mM), and cDNA (2.5 ng/ml). PCR conditions significance among groups, as listed in each figure legend. For a one- were 95˚C for 10 min (1 cycle), followed by 95˚C for 15 s and 60˚C for sample t test, each data point for the experiment under consideration 1 min (40 cycles). Results were normalized based on the GAPDH as an was generated from a process that involved a comparison with a control

endogenous control. condition, where each control group value was normalized to 1. A p value http://www.jimmunol.org/ ,0.05 was considered significant. In vitro suppression assay

6 Human PBMCs (1 3 10 ) were suspended in 1 ml SFM and labeled with Results 0.5 mM CFSE (ThermoFisher Scientific) or 5 mM CellTrace Violet (ThermoFisher Scientific) according to the manufacturer’s instructions. IL-2–induced pSTAT5 activation is dependent on CD25 levels + For in vitro suppression assays, labeled PBMCs (100,000 cells per well) in human CD4 T cells and YT cell line were incubated with titrated amounts of Tregs (in serially decreasing ra- + tios) in SFM in 96-well round-bottom plates (Corning, NY). Cells were CD4 T cells are heterogeneous with respect to CD25 expression, stimulated with anti-CD3 (2 mg/ml) (OKT3; BioLegend) and anti-CD28 but express similar amount of IL-2R b and gc (12). To assess the

(2 mg/ml) (BioLegend), and incubated at 37˚C under 5% CO2.After3d, extent that the level of CD25 contributes to the activation of IL-2R by guest on September 30, 2021

FIGURE 1. IL-2–induced pSTAT5 activation is dependent on CD25 levels in human CD4+ T cells. Unfractionated PBMCs from different healthy adult donors (n = 5) were stimulated with the indicated amount of human IL-2 for 15 min at 37˚C. (A) pSTAT5 activation of CD4+ T cells gated on CD25 expression. (B) Foxp3 expression of CD4+ T cells based on the gates shown in (A). The color code distinguishes the gating scheme. The three gates with the highest CD25 level are Foxp3+ and labeled as Treg. The remaining three gates are Foxp32 and labeled as CD4+ T conventional cells (Tcon). The numbers represent the percentage of Foxp3+ cells. (C) pSTAT5 dose-response curves of the indicated cell populations. (D) Nonlinear regression analysis of the binding data in (C). (E)EC50 for pSTAT5 activation determined by (D) versus MFI of CD25. The EC50s from the representative cell populations are shown in the graph. Data in (C)–(E) are shown as mean 6 SEM and were analyzed by an unpaired two-tailed t test. ***p , 0.001, ****p , 0.0001. 4 CD25 AND PP2A IN IL-2R SIGNALING SENSITIVITY signaling, we measured tyrosine phosphorylation of STAT5 (pSTAT5) these data showed that signaling sensitivity was related to CD25 in human CD4+ T cells 15 min after stimulation with IL-2. Higher levels (Fig. 2D, 2E), with the most sensitive responses to a low IL-2 activation of pSTAT5 was associated with CD4+ T cells that IL-2 concentration at high levels of CD25. Collectively, these expressed greater amounts of CD25 (Fig. 1A). After gating of findings indicate that IL-2R levels represent one mechanism that these cells based on CD25 levels (Fig. 1B), the three gates with the contributes to the high signaling sensitivity of Tregs to low-dose highest levels of CD25 were largely Foxp3hi and were designated IL-2. as Tregs that expressed CD25 at high (CD25hi), intermediate (CD25med), and low (CD25lo) levels. In contrast, the three gates PP2A promotes IL-2R signaling in human Tregs 2 with lowest levels of CD25 were largely Foxp3 and were des- Past results implicated PP2A as another potential cell intrinsic ignated as CD25hi, CD25med, and CD25lo conventional CD4+ factor that regulates IL-2R signaling sensitivity (12, 18). To di- T cells (Fig. 1B). Dose-response studies indicated that IL-2 acti- rectly assess the role of PP2A in human Tregs, we sought to vation of pSTAT5 was related to the levels of CD25 on these CD4+ manipulate PP2A mRNA levels in Tregs. Short-term in vitro ex- T cells (Fig. 1C). Nonlinear regression analysis of the pSTAT5 panded Tregs were used for this purpose (Supplemental Fig. 1A). data based on this scheme (Fig. 1D) revealed that the EC50 of Initially, the properties of these cultured Tregs were evaluated in the CD25med (0.5 pM) and CD25lo (4.3 pM) Tregs was ∼70- and relationship to the input primary Tregs isolated from the PBMCs. 8-fold more responsive to IL-2, respectively, than CD25hi CD4+ FACS-sorted CD4+ CD25hi CD127lo Tregs (.90% Foxp3+ T conventional cells (EC50, 35.5 pM). Furthermore, these data Supplemental Fig. 1B) were initially stimulated with anti-CD3/ directly showed that the EC50 of all these cells were proportional CD28 and IL-2 and further expanded after subculture with IL-2. to CD25 levels (Fig. 1E). After 5 d in culture, the expanded Tregs remained 90–95% Foxp3+ Downloaded from The above experiment and past studies (12) explored the re- (Supplemental Fig. 1B), and IL-2 activated pSTAT5 in a manner sponse of heterogeneous populations of T cells, where other cell similar to the input primary Tregs (Supplemental Fig. 1C). The intrinsic factors might also contribute to the distinctive pSTAT5 expanded Tregs also retained suppressive activity in vitro activation. Therefore, we examined pSTAT5 activation of a more (Supplemental Fig. 1D). homogenous cell by using an IL-2Ra–expressing YT human NK PP2Ac was overexpressed in these expanded human Tregs

cell line (22). These cells expressed CD25 at heterogeneous levels by lentiviral transduction. The transduction efficiency was http://www.jimmunol.org/ but CD122 and CD132 at relatively high homogenous amounts 85–90% and 35–45% for vector and PP2Ac-encoding plasmids, (Fig. 2A). pSTAT5 activation was assessed for IL-2 stimulation respectively (Fig. 3A). Compared with control vector-transduced of YT cells that expressed high, intermediate, low, and negative Tregs, PP2Ac mRNA levels increased ∼10- to 15-fold (Fig. 3B) levels of CD25 (Fig. 2B, 2C). Nonliner regression analysis of and protein levels ∼2- to 3-fold (Fig. 3C). FACS analyses revealed by guest on September 30, 2021

FIGURE 2. IL-2–induced pSTAT5 activation is de- pendent on CD25 levels in YT cells. (A) Representative histograms of IL-2R subunit expression on YT cells. Unfilled histograms represents unstained cells. (B) YT cells were stimulated with the indicated amount of human IL-2 for 15 min at 37˚C. Cells were gated into four groups (high, medium, low, negative) based on CD25 expression. (C) IL-2–dependent pSTAT5 dose- response curves of the indicated cell populations in (B) (n =3).(D) Nonlinear regression analysis of the binding data in (C). (E) A representative plot of EC50 determined by the analysis in (D) versus MFI of CD25. Data in (C) and (D) are shown as mean 6 SEM of three independent experiments. The Journal of Immunology 5

FIGURE 3. Overexpression of PP2Ac activates JAK3/STAT5 signal- ing in human Tregs. Expanded human Tregs were lentiviral transduced with control vector or a plasmid encoding PP2Ac and cultured in SFM with anti- CD3/CD28 beads and IL-2 for 3 d before analysis. (A) Transduction effi- Downloaded from ciency was assessed by the percentage of GFP+ cells. (B–D) Expression of PP2Ac in GFP+ cells was assessed at the mRNA and protein levels using quantitative PCR (B), Western blot- ting (C), and flow cytometry (D) E (n =3).() A representative im- http://www.jimmunol.org/ munoblot (left) and densitometry analysis of the resulting data (right) of protein extracts from transduced Tregs were probed for tyrosine- phosphorylated JAK3 (Tyr980/981), total JAK3, tyrosine-phosphorylated STAT5 (Tyr694), PP2Ac, and b-Tubulin (n = 3). (F) IL-2–induced pSTAT5 dose-response curves of transduced Tregs (GFP+Foxp3+) (left) and non- by guest on September 30, 2021 linear regression analysis of the bind- ing data (middle) to determine EC50 (right) for IL-2–induced pSTAT5 activation (n = 6). Data in plots are shown as mean 6 SEM and were analyzed by a one-sample two-tailed t test (B–F). *p , 0.05, **p , 0.01, ***p , 0.001, ****p , 0.0001. ns, not significant.

that mean fluorescence intensity (MFI) of PP2Ac increased ∼2- contribution of PP2A on these aspects of IL-2R signaling was to 3-fold, with a peak shift in PP2Ac-overexpressed Tregs assessed after overexpression of PP2Ac. pJAK3 at Tyr980/981 compared with control (Fig. 3D). The staining of PP2Ac by and pSTAT5 at Tyr694 was enhanced in the PP2Ac transduced this mAb was specific as unlabeled anti-PP2Ac inhibited the cells (Fig. 3E). Dose-response studies were performed to assess binding of PE-labeled mAb to the expanded Tregs (Supplemental the effect of overexpression of PP2Ac on IL-2 signaling sen- Fig. 2). The lower increase of PP2Ac at the protein level is likely sitivity by measuring pSTAT5 activation by flow cytometry. due to posttranscriptional regulatory mechanisms (27). The resulting dose-response curves and nonlinear regression Tyrosine phosphorylation of JAK3 and STAT5 is an imme- analysis of these data showed that Tregs with increased amounts diate consequence of IL-2–induced signaling (28, 29). The of PP2Ac were ∼3-fold more responsive to IL-2 than the control 6 CD25 AND PP2A IN IL-2R SIGNALING SENSITIVITY

FIGURE 4. Overexpression of PP2Ac upregulates IL-2–dependent genes in human Tregs. Tregs from different healthy adult donors (n = 3) were expanded and lenti- viral transduced in vitro. Total RNA was isolated from vector-transduced or PP2Ac- overexpressed Tregs 3 d after lentiviral trans- duction and analyzed by real-time quantitative PCR. Data were normalized to the mRNA level in the control vector-transduced cells. The sample in each graph with the highest fold-change is from the same donor. Downloaded from

Tregs (Fig. 3F). These data indicate the PP2A can directly pro- the consequence of overexpression of PP2A on downstream mote IL-2R signaling in Tregs. IL-2–dependent gene activation, we evaluated six of these genes http://www.jimmunol.org/ in control- and PP2Ac-transduced Tregs from three normal sub- PP2A promotes expression of IL-2–dependent targets jects. These genes encode molecules involved in IL-2 binding in human Tregs (CD25), Treg function (Foxp3), lymphocyte survival (BCL2), and Past genome-wide mRNA profiling of human Tregs identified 388 immune system processes (AHR, SOCS1, SOCS3). All the mRNAs that were differentially regulated by IL-2 (12). To assess mRNAs were upregulated in PP2A-overexpressed Tregs, although by guest on September 30, 2021

FIGURE 5. Increased expression of CD25 induced by PP2Ac overexpression does not fully account for enhanced pSTAT5 activation. Expanded human Tregs from different healthy adult donors (n = 4) were lentiviral transduced with control vector or a plasmid encoding PP2Ac and analyzed 3 d later. (A) CD25 level of Foxp3+ GFP+ transduced Tregs. Numbers represent MFI of CD25 for the indicated cell population. (B) Representative gating strategy of FACS plots to identify transduced Tregs with a similar MFI of CD25 (represented in the P5 gate). (C) IL-2–induced pSTAT5 dose-response curves (left) of control or PP2Ac-overexpressed Tregs with similar CD25 levels, as shown in (B). Nonlinear regression analysis of the binding data (middle) was conducted to determine the EC50 (right) for IL-2–induced pSTAT5. (D) MFI of pSTAT5 versus CD25 levels in control and PP2Ac-overexpressed Tregs after stim- ulation with IL-2 (1 U/ml) for 15 min (n = 4). Representative gating strategy (left) and quantified data (right) where the MFI of CD25 was normalized to 1 based for the gated cell with the lowest amount of CD25. Data in (C) and (D) are shown as mean 6 SEM and were analyzed by one-sample two-tailed t test (C) or a paired two-tailed t test (D). *p , 0.05, **p , 0.01, ****p , 0.0001. The Journal of Immunology 7 the level of increase varied between individuals, with one being compared the consequence of increased PP2Ac expression on other much more responsive to IL-2 (Fig. 4). This trend, however, was markers that are important for Treg function. Representative FACS not statistically significant, owing to the variance of the response. analysis (Fig. 6A) and composite data (Fig. 6B) revealed that The distinctive responses are consistent with potential high/low overexpression of PP2Ac not only increased cell surface levels of responders to IL-2, which we have previously noted (30). CD25 but also CD132, which could also contribute to enhanced The increase in CD25 mRNA after overexpression of PP2Ac IL-2R signaling. Foxp3, ICOS, TIGIT, CD69, CD73, and BCL2 raised the possibility that the increased IL-2–dependent pSTAT5 significantly increased whereas expression of CD45RA decreased. sensitivity (Fig. 3F) might simply reflect higher amount of this These data suggest that PP2A might act to enhance Treg stability, IL-2R subunit on the PP2Ac-tranduced cells (Fig. 5A). We, function, and persistence in part by promoting activated or effector therefore, directly assess the impact of PP2Ac on IL-2R signaling Tregs. Consistent with this view, cultured Tregs that overexpressed independent of varied CD25 levels. IL-2 activation of pSTAT5 PP2Ac were also more suppressive in the in vitro suppressive was assessed after gating CD25 of control- and PP2Ac-tranduced assay (Fig. 6C). These data are consistent with the notion that GFP+ Foxp3+ cells such that the MFI of CD25 was equivalent PP2A is an important regulator of IL-2R signaling and function in (Fig. 5B). Under these conditions, the PP2Ac transduced cells human Tregs. remained more sensitive to IL-2, but at a reduced level; that is, ∼2- Decreased PP2Ac in human Tregs lowers responses to IL-2 fold with normalized CD25 levels (Fig. 5C) versus 3-fold when gating on the total population GFP+-transduced Tregs. (Fig. 3F). To further explore the role of PP2A in the regulation of IL-2R IL-2–induced pSTAT5 activation was also compared over a range signaling, the levels of PP2Ac were reduced in cultured human of similar amounts of CD25 for control- and PP2Ac-transduced Tregs after lentiviral transduction of a shRNA designed to target Downloaded from Tregs, using a fixed and limiting amount of IL-2. Under these PP2Aca. The transduction efficiency was ∼85% in both control conditions, pSTAT5 levels were always greater for the PP2Ac- and shRNA-transduced Tregs (Fig. 7A). This PP2Ac-directed overexpressed Tregs, except when CD25 levels were the highest shRNA reduced mRNA by ∼75% and protein by 50–60% as (Fig. 5D). These data indicate that some of the enhanced pSTAT5 assessed by Western blots and FACS analyses (Fig. 7B). These response mediated by increased PP2Ac is due to effects on Tregs PP2Aca transduced cells were gated to identify Tregs with ap-

independent of its ability to enhance CD25 expression. proximately a 40 and 80% reduction in PP2Ac (Fig. 7C). IL-2– http://www.jimmunol.org/ induced pSTAT5 dose-response curves showed that percentage of PP2A broadly regulates the function of human Tregs pSTAT5+ Tregs was markedly lower in Tregs with an 80%, but not The capacity of PP2A to enhance IL-2R signaling and downstream a 40%, decrease of PP2Ac (Fig. 7D). Nonlinear regression anal- gene expression suggests that multiple activities of human Tregs ysis of these dose-responses curves revealed a statistically non- are regulated by PP2A. To address this possibility, we further significant trend for a somewhat higher EC50 after an 80% by guest on September 30, 2021

FIGURE 6. PP2Ac increases survival, activation, and immunosuppressive function of human Tregs. Representative histograms (A) and quantitative evaluation (B) of expression of the indicated markers for control and PP2Ac-overexpressed Tregs (n = 4). (C) In vitro suppression assay of CD8+ T cells (responders) by control or PP2Ac-overexpressed Tregs (n = 3). Data in (B) and (C) are shown as mean 6 SEM and were analyzed by a one-sample two- tailed t test (B) or an unpaired two-tailed t test (C). *p , 0.05, **p , 0.01, ***p , 0.001. ns, not significant. 8 CD25 AND PP2A IN IL-2R SIGNALING SENSITIVITY Downloaded from http://www.jimmunol.org/ by guest on September 30, 2021

FIGURE 7. Knockdown of PP2Ac reduces pSTAT5 response to IL-2 by human Tregs. Expanded human Tregs were lentiviral transduced with control or PP2Ac shRNA and analyzed after 3 d. (A) Transduction efficiency was assessed by the percentage of GFP+ cells. (B) Expression of PP2Ac was assessed by quantitative PCR (n = 3), Western blotting (n = 3), and flow cytometry (n = 3) in control and PP2Ac knockdown Tregs. (C) FACS gating strategy for pSTAT5 activation analysis. Tregs transduced with PP2Ac shRNA were gated into two groups according to the knockdown efficiency. Numbers represent MFI of PP2Ac for the indicated cell population. (D) IL-2–induced pSTAT5 dose-response curves (left) for the indicated group of Tregs shown in (C)(n = 6). Nonlinear regression analysis of the binding data (middle) was used to determine EC50 (right) (E) Representative histograms (left) and quantitative evaluation (right) of pSTAT5 MFI for indicated cell population after treatment with 40 and 400 pM of IL-2 (n = 6). The numbers represent MFI of the gated pSTAT5+ cells. (F) Representative histograms of IL-2R subunit expression on control and PP2Ac shRNA-transduced Tregs. Data in (B)–(E) are shown as mean 6 SEM and were analyzed by a one- sample two-tailed t test (B, D,andE)oranunpairedtwo-tailedt test (D). *p , 0.05, **p , 0.01, ***p , 0.001, ****p , 0.0001. ns, not significant. reduction in PP2Ac. Furthermore, when the amount of pSTAT5 similar amount of PP2Ac (Fig. 8A, 8B), but Tregs displayed was measured for Tregs at high concentrations of IL-2 by enu- greater PP2A activity than TEM cells (Fig. 8C). To test the effect of merating the MFI, the amount of pSTAT5 was reduced only in PP2A on IL-2R signaling for Teff cells, IL-2–dependent pSTAT5 Tregs with an 80% decrease of PP2Ac (Fig. 7E). Importantly, levels were measured in control and after overexpression or levels of IL-2R subunits were comparable for the control and knockdown of PP2A. With respect to overexpression of PP2A, transduced PP2Ac shRNA-transduced Tregs (Fig. 7F). Thus, upon transduction efficiency was 75–85% and 35–45% for vector substantial reduction in PP2A, human Tregs showed an impair- and PP2Ac-encoding plasmids, respectively (Fig. 8D). For the ment in response to IL-2 that is independent of CD25 levels. knockdown experiments, transduction efficiency was .90% in both control and PP2Ac shRNA-transduced Teff cells (Fig. 8E). PP2A does not regulate IL-2R signaling in Teff cells Compared with control vector-transduced Teff cells, FACS anal- Given the capacity of PP2A to affect IL-2R signaling in Tregs, we ysis revealed that PP2Ac increased ∼2-fold in PP2Ac-transduced assessed the extent by which PP2A affected CD4+ T conventional Teff cells (Fig. 8F). For PP2Ac-shRNA–transduced Teff cells, the cells. Expression and catalytic activity of PP2Ac were compared cells were gated to discriminate a decrease of ∼40 and 80% of in freshly isolated Tregs and TEM cells. Both cells expressed a PP2Ac (Fig. 8G). Overexpression or knockdown of PP2Ac did not The Journal of Immunology 9 Downloaded from http://www.jimmunol.org/ by guest on September 30, 2021

FIGURE 8. Overexpression and knockdown of PP2Ac does not alter pSTAT5 response to IL-2 by human Teff cells. (A and B) Expression of PP2Ac in freshly isolated Treg and TEM cells were assessed using Western blotting (n =3)(A) and flow cytometry (n =3)(B). (C) Quantification of the enzymatic activity of PP2A in freshly isolated Treg and TEM cells (n = 3). (D and E) Transduction efficiency in vector-transduced and PP2Ac-overexpressed Teff cells (D) or in control and PP2Ac shRNA-transduced cells (E) was assessed by the percentage of GFP+ cells. (F and G) PP2Ac expression in control and PP2Ac- transduced overexpressed cells (F) or in control and PP2Ac shRNA knockdown cells (G) was assessed by flow cytometry (n = 3). Teff cells transduced with PP2Ac shRNA were gated into two groups according to the knockdown efficiency and this gating strategy was used for pSTAT5 analysis in (K). (H and I) Representative histograms for CD25 expression for control and PP2Ac-overexpressed (H) or for control and PP2Ac shRNA-transduced Teff cells (I). The numbers in the FACS plots are the MFI of CD25 for the indicated cell populations. (J and K) IL-2–induced pSTAT5 dose-response curves (left) of control and PP2Ac-overexpressed (J) or control and PP2Ac shRNA-transduced Teff cells (K). Nonlinear regression analysis of the binding data (middle) was performed to determine EC50 (right) (n = 3). Data in (A)–(C), (F), (G), (J), and (K) are shown as the mean 6 SEM and were analyzed by a one-sample two- tailed t test (A–C, F, J, and K). **p , 0.01, ***p , 0.001. ns, not significant. alter the amounts of CD25 in Teff cells (Fig. 8H, 8I). Furthermore, Foxp3YFP-Cre mice resulted in mice (designated as PP2Aflox) that IL-2–induced pSTAT5 dose-response curves, and nonlinear re- do not express PP2Aaa in Tregs. Knockout of PP2Aaa disrupts gression analyses showed that increasing (Fig. 8J) or lowering the binding between PP2Aa and PP2Ac and impairs the phos- (Fig. 8K) the levels of PP2Ac in Teff cells did not affect IL-2– phatase activity of PP2Ac. Tregs in these mice also expressed the dependent activation of STAT5. Thus, in contrast to Tregs, PP2A YFP reporter because of the cross to Foxp3YFP-Cre. Wild-type does not readily modify proximal IL-2R signaling in Teff cells. littermate mice were used as controls for these experiments. Western blot analysis confirmed the absence of PP2Aaa in Tregs Impaired PP2A decreased pSTAT5 activation in mouse Tregs (Fig. 9A). FACS analysis of Tregs from PP2Aflox mice indicated When PP2A activity is conditionally absent in Tregs through de- that they expressed ∼10% lower levels of CD25 than Tregs from letion of the PP2Aa a isoform (PP2Aaa encoded by Ppp2r1a), the control mice (Fig. 9B, 9C). dominant isoform in lymphoid cells, mice develop lethal auto- To study the consequence of the lack of PP2Aaa, which leads to immunity, illustrating the importance of this phosphatase in Treg nonfunctional PP2A, on IL-2R signaling, we evaluated IL-2– function (14). Appropriate crossing of Ppp2r1aflox/flox mice to dependent pSTAT5 activation in Tregs from PP2Aflox mice. Dose- 10 CD25 AND PP2A IN IL-2R SIGNALING SENSITIVITY Downloaded from http://www.jimmunol.org/ by guest on September 30, 2021

FIGURE 9. Expression of CD25 and IL-2–induced pSTAT5 activation in Tregs are reduced in PP2Aflox mice. Tregs from spleen of control (Foxp3 YFP-CrePpp2r1aWT)andPP2Aflox(Foxp3YFP-CrePpp2r1aflox/flox) mice at the age of 6–8 wk were used for analysis. (A) Expression of PP2Aaa was assessed using immunoblot of protein extracts from the splenic FACS sorted Tregs from control and PP2Aflox mice. (B) FACS gating strategy for total Tregs (P5 gate) and Tregs with the same MFI of CD25 (P6 gate). The numbers with the histograms (right) represent the MFI of CD25 on the gated cell population. (C) MFI of CD25 on total splenic Tregs in control and PP2Aflox mice (n = 5). (D and E) IL-2–induced pSTAT5 dose-response curves (left) of gated total Tregs (P5 gate) (D) or of gated Tregs with the same MFI of CD25 (P6 gate) (E) shown in (B) and nonlinear regression analysis of binding data (middle) to calculate the

EC50 (right) (n = 5). Data in (C)–(E)areshownasmean6 SEM and were analyzed by an unpaired two-tailed t test. *p , 0.05, **p , 0.01. response experiments demonstrated that Tregs without PP2Aaa provide direct data consistent with two distinct mechanisms re- were less responsive to low levels of IL-2 than the Tregs in control sponsible for the high sensitivity of Tregs to IL-2. One mechanism mice (Fig. 9D). The EC50 increased ∼2.9-fold, from 2.13 versus is related to the high cell surface levels of CD25 on Tregs. Past work 6.18 pM, for Tregs from the PP2Aflox mice (Fig. 9D). The 3-fold demonstrated that IL-2–dependent signaling by in vitro mouse increase of EC50 indicated that loss of PP2Ac activity decreased Teff cells was proportional to IL-2R levels (13). In this study, we pSTAT5 activation in Tregs. When the Tregs were regated to take provide evidence that this mechanism is operative for human into account their slightly lower levels of CD25 (Fig. 9C), Tregs, as IL-2–induced tyrosine pSTAT5 in human Tregs and the the EC50 value for IL-2–induced activation of pSTAT5 was still YT NK cell line is directly associated to heightened sensitivity 2.5-fold higher than control Tregs (Fig. 9E). This value was only with increasing CD25 levels. slightly lower than the fold-change in total Tregs. Thus, lower The other mechanism is related to the activity of the Ser/Thr IL-2 sensitivity in Tregs without PP2Aaa is largely independent of phosphatase PP2A. When PP2A expression was increased or re- CD25 levels. duced in human Tregs or absent in mouse Tregs, IL-2–dependent pSTAT5 was modulated in a way where increasing levels of PP2A Discussion was associated with more effective responses by Tregs to lower Understanding Treg-intrinsic properties that account for their se- levels of IL-2. Enhancing PP2Ac amounts in Tregs also upregu- lective responsiveness to low amounts of IL-2 is important to lated IL-2–dependent downstream genes, including CD25, but provide a solid mechanistic underpinning for advancing low-dose also other genes that are expected to enhance their survival and IL-2 as a therapy to treat autoimmune diseases. In this study, we activation. Increased PP2Ac was also associated with enhanced The Journal of Immunology 11 human Treg suppressive function in vitro. Although PP2A activity that PP2A promotes the activation of STAT5, providing another is linked to regulation of CD25, this effect did not fully account level by which PP2A regulates Treg function. Analysis of STAT5 for the ability of PP2A to affect IL-2R signaling. Thus, these two binding sites using chromatin immunoprecipitation identified a mechanisms are not synonymous but are partially overlapping. number of STAT5 target genes involved in survival, development, Our data also show a differentially capacity of PP2A to regulate and function of T cells, such as Foxp3, Il2ra (CD25), Bcl2, Socs1, IL-2R signaling in Treg and Teff cells. Although the amount of and Socs3 (50, 51). All these genes were upregulated by over- PP2Ac was similar in Treg and TEM cells, its catalytic activity was expression of PP2A in our study. Therefore, PP2A likely promotes higher in Tregs, suggesting that PP2A phosphatase activity is survival, activation, and immunosuppressive function of Tregs distinctively regulated in these cell populations. Correspondingly, at least in part by promoting STAT5 activation while limiting PP2A is associated with enhancing IL-2R signaling in Treg but not TORC1. As such, PP2A has the potential to influence both Teff cells. The mechanisms behind this distinctive activity are not proximal and distal IL-2R signaling. clear but may be related to cell type specific factors that control Overall, our study helps to better understand the basis by which substrate specificity. For example, PP2A is a negative regulator of Tregs selectively respond to low-dose IL-2. Pharmacological ap- IL-2 production in Teff cells through dephosphorylation of the proaches that enhance PP2A activity in Tregs have the potential to transcription factor Elf-1, which interrupts TCR signaling by ul- further promote this selectivity by directly enhancing IL-2R sig- timately decreasing CD3z-chain within the CD3 complex (31). naling and by indirectly increasing the levels of CD25 (52). PP2A also dephosphorylates another transcription factor, speci- Moreover, this effect influences other downstream activities that ficity protein-1 (SP-1), which inhibits transcription of IL-2 (32). enhance Treg survival and suppressive function. Thus, enhancing

This specific regulation is irrelevant in Tregs as they do not pro- PP2A activity in Tregs may improve low-dose IL-2 therapy Downloaded from duce IL-2. for autoimmune diseases by minimizing off-target effects JAK1/3-STAT5 is the dominant IL-2–dependent signaling while enhancing efficacy. pathway in Tregs because of the inhibitory effects associated with high levels of PTEN on the PI3K and MAPK pathways (33, 34). Although we demonstrated that PP2A activity is associated with Acknowledgments We thank the Flow Cytometry Cores of the Diabetes Research Institute enhanced tyrosine phosphorylation of JAK3 and STAT5, the basis http://www.jimmunol.org/ for this effect remains unclear. Treatment of YT cells with PP1/ and the Sylvester Comprehensive Cancer Center at the University of Miami. We also thank Dr. Warren J. Leonard for providing the YT cell line PP2A inhibitors increased Ser/Thr phosphorylation and decreased and Dr. Noriyuki Kasahara for the help with plasmid construction and IL-2–dependent Tyr phosphorylation of JAK3 and STAT5 (18), lentivirus generation. suggesting that Ser/Thr phosphatase activity of PP2A may nor- mally promote IL-2 signaling, which is consistent with our data. One simple idea is that PP2A directly interacts with one or more Disclosures components of an IL-2R signaling complex. However, such an The authors have no financial conflicts of interest. interaction has not yet been demonstrated. Indeed a physical in- teraction between PP2Ac and JAK3 or STAT5 was not detected by guest on September 30, 2021 using protein interaction assays after immunoprecipitation of ac- References 1. Malek, T. R. 2008. The biology of interleukin-2. Annu. Rev. Immunol. 26: 453– tivated PP2Ac in Jurkat T cells (14), suggesting JAK3 and STAT5 479. are not direct substrates of PP2Ac. This result, however, does 2. Yu, A., L. Zhu, N. H. Altman, and T. R. Malek. 2009. A low interleukin-2 re- not preclude a direct interaction with IL-2R subunits to promote ceptor signaling threshold supports the development and homeostasis of T reg- ulatory cells. Immunity 30: 204–217. IL-2R signaling. 3. Grinberg-Bleyer, Y., A. Baeyens, S. You, R. Elhage, G. Fourcade, S. Gregoire, Alternatively, PP2A may dephosphorylate Ser/Thr on a kinase or N. Cagnard, W. Carpentier, Q. Tang, J. Bluestone, et al. 2010. IL-2 reverses another target(s) to enhance its activity, ultimately leading to more established type 1 diabetes in NOD mice by a local effect on pancreatic regu- latory T cells. J. Exp. Med. 207: 1871–1878. robust IL-2R signaling. In this regard, the serine/threonine kinases 4. Webster, K. E., S. Walters, R. E. Kohler, T. Mrkvan, O. Boyman, C. D. Surh, Mst1 and Mst2 have been implicated, through Rac1-dependent S. T. Grey, and J. Sprent. 2009. In vivo expansion of T reg cells with IL-2-mAb complexes: induction of resistance to EAE and long-term acceptance of cytoskeletal reorganization, to amplify IL-2R signaling in Tregs islet allografts without immunosuppression. J. Exp. Med. 206: 751–760. (35). PP2A antithetically regulates Mst2 and Rac, where PP2A 5. Klatzmann, D., and A. K. Abbas. 2015. The promise of low-dose interleukin-2 acts to inactivate MST2 (36), but potentiates Rac1 (37). These therapy for autoimmune and inflammatory diseases. Nat. Rev. Immunol. 15: 283–294. and/or other substrates of PP2A may engage multiple pathways 6. Matsuoka, K., J. Koreth, H. T. Kim, G. Bascug, S. McDonough, Y. Kawano, where the net effect results in PP2A is to promote IL-2R–dependent K. Murase, C. Cutler, V. T. Ho, E. P. Alyea, et al. 2013. Low-dose interleukin-2 signaling in Tregs. therapy restores regulatory T cell homeostasis in patients with chronic graft- versus-host disease. Sci. Transl. Med. 5: 179ra43. PP2A is also involved in multiple key intercellular signaling 7. Saadoun, D., M. Rosenzwajg, F. Joly, A. Six, F. Carrat, V. Thibault, D. Sene, pathways, including but not limited to MAPK, AKT, NF-kB, P. Cacoub, and D. Klatzmann. 2011. Regulatory T-cell responses to low-dose interleukin-2 in HCV-induced vasculitis. N. Engl. J. Med. 365: 2067–2077. and mTOR pathways (38–41), depending on the cell type and 8. Castela, E., F. Le Duff, C. Butori, M. Ticchioni, P. Hofman, P. Bahadoran, biological response. IL-2 contributes to T cell function mainly J. P. Lacour, and T. Passeron. 2014. Effects of low-dose recombinant interleukin through two signaling pathways associated with the IL-2R/ 2 to promote T-regulatory cells in alopecia areata. JAMA Dermatol. 150: 748–751. JAK1/3-STAT5 and mTORC1 pathways (42). Increased activ- 9. von Spee-Mayer, C., E. Siegert, D. Abdirama, A. Rose, A. Klaus, T. Alexander, ity of mTOR pathway negatively affects the generation and the P. Enghard, B. Sawitzki, F. Hiepe, A. Radbruch, et al. 2016. Low-dose function of Tregs (43, 44). PP2A has been reported to inhibit the interleukin-2 selectively corrects regulatory T cell defects in patients with sys- temic lupus erythematosus. Ann. Rheum. Dis. 75: 1407–1415. activity of mTORC1 complex (14), suggesting a link between 10. Koreth, J., K. Matsuoka, H. T. Kim, S. M. McDonough, B. Bindra, E. P. Alyea, PP2A activity and Treg suppressive function. However, the role of 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. PP2A in JAK-STAT pathway in Tregs is poorly understood. Ac- 11. Malek, T. R., and I. Castro. 2010. Interleukin-2 receptor signaling: at the in- tivation of STAT5-mediated transcriptional programs by IL-2 terface between tolerance and immunity. Immunity 33: 153–165. is essential for Tregs (11, 45, 46). Correspondingly, STAT5 defi- 12. Yu, A., I. Snowhite, F. Vendrame, M. Rosenzwajg, D. Klatzmann, A. Pugliese, and T. R. Malek. 2015. Selective IL-2 responsiveness of regulatory T cells ciency in human and mouse Tregs is associated with their im- through multiple intrinsic mechanisms supports the use of low-dose IL-2 ther- paired development and homeostasis (47–49). Our data indicate apy in type 1 diabetes. Diabetes 64: 2172–2183. 12 CD25 AND PP2A IN IL-2R SIGNALING SENSITIVITY

13. Feinerman, O., G. Jentsch, K. E. Tkach, J. W. Coward, M. M. Hathorn, 34. Li, L., W. R. Godfrey, S. B. Porter, Y. Ge, C. H. June, B. R. Blazar, and M. W. Sneddon, T. Emonet, K. A. Smith, and G. Altan-Bonnet. 2010. Single-cell V. A. Boussiotis. 2005. CD4+CD25+ regulatory T-cell lines from human cord quantification of IL-2 response by effector and regulatory T cells reveals critical blood have functional and molecular properties of T-cell anergy. Blood 106: plasticity in immune response. Mol. Syst. Biol. 6: 437. 3068–3073. 14. Apostolidis, S. A., N. Rodrı´guez-Rodrı´guez, A. Sua´rez-Fueyo, N. Dioufa, 35. Shi, H., C. Liu, H. Tan, Y. Li, T. M. Nguyen, Y. Dhungana, C. Guy, P. Vogel, E. Ozcan, J. C. Crispı´n, M. G. Tsokos, and G. C. Tsokos. 2016. Phosphatase PP2A G. Neale, S. Rankin, et al. 2018. Hippo kinases Mst1 and Mst2 sense and am- is requisite for the function of regulatory T cells. Nat. Immunol. 17: 556–564. plify IL-2R-STAT5 signaling in regulatory T cells to establish stable regulatory 15. Shi, Y. 2009. Serine/threonine phosphatases: mechanism through structure. Cell activity. Immunity 49: 899–914.e6. 139: 468–484. 36. Zheng, Y., B. Liu, L. Wang, H. Lei, K. D. Pulgar Prieto, and D. Pan. 2017. 16. Janssens, V., S. Longin, and J. Goris. 2008. PP2A holoenzyme assembly: in Homeostatic control of Hpo/MST kinase activity through autophosphorylation- cauda venenum (the sting is in the tail). Trends Biochem. Sci. 33: 113–121. dependent recruitment of the STRIPAK PP2A phosphatase complex. Cell Rep. 17. Sontag, E. 2001. Protein phosphatase 2A: the Trojan Horse of cellular signaling. 21: 3612–3623. Cell. Signal. 13: 7–16. 37. Kong, M., T. V. Bui, D. Ditsworth, J. J. Gruber, D. Goncharov, V. P. Krymskaya, 18. Ross, J. A., H. Cheng, Z. S. Nagy, J. A. Frost, and R. A. Kirken. 2010. Protein T. Lindsten, and C. B. Thompson. 2007. The PP2A-associated protein alpha4 phosphatase 2A regulates interleukin-2 receptor complex formation and JAK3/ plays a critical role in the regulation of cell spreading and migration. J. Biol. STAT5 activation. J. Biol. Chem. 285: 3582–3591. Chem. 282: 29712–29720. 19. Ruediger, R., J. Ruiz, and G. Walter. 2011. Human cancer-associated mutations 38. Silverstein, A. M., C. A. Barrow, A. J. Davis, and M. C. Mumby. 2002. Actions in the Aa subunit of protein phosphatase 2A increase lung cancer incidence in of PP2A on the MAP kinase pathway and apoptosis are mediated by distinct Aa knock-in and knockout mice. Mol. Cell. Biol. 31: 3832–3844. regulatory subunits. Proc. Natl. Acad. Sci. USA 99: 4221–4226. 20. Rubtsov, Y. P., J. P. Rasmussen, E. Y. Chi, J. Fontenot, L. Castelli, X. Ye, 39. Lin, C. F., C. L. Chen, C. W. Chiang, M. S. Jan, W. C. Huang, and Y. S. Lin. P. Treuting, L. Siewe, A. Roers, W. R. Henderson, Jr., et al. 2008. Regulatory 2007. GSK-3beta acts downstream of PP2A and the PI 3-kinase-Akt pathway, T cell-derived interleukin-10 limits inflammation at environmental interfaces. and upstream of caspase-2 in ceramide-induced mitochondrial apoptosis. J. Cell Immunity 28: 546–558. Sci. 120: 2935–2943. 21. Kim, H. P., J. Kelly, and W. J. Leonard. 2001. The basis for IL-2-induced IL-2 40. Tsuchiya, Y., K. Osaki, M. Kanamoto, Y. Nakao, E. Takahashi, T. Higuchi, and receptor alpha chain gene regulation: importance of two widely separated IL-2 H. Kamata. 2017. Distinct B subunits of PP2A regulate the NF-kB signalling response elements. Immunity 15: 159–172. pathway through dephosphorylation of IKKb,IkBa and RelA. FEBS Lett. 591: Downloaded from 22. Yodoi, J., K. Teshigawara, T. Nikaido, K. Fukui, T. Noma, T. Honjo, 4083–4094. M. Takigawa, M. Sasaki, N. Minato, M. Tsudo, et al. 1985. TCGF (IL 2)-re- 41. Di Conza, G., S. Trusso Cafarello, S. Loroch, D. Mennerich, S. Deschoemaeker, ceptor inducing factor(s). I. Regulation of IL 2 receptor on a natural killer-like M. Di Matteo, M. Ehling, K. Gevaert, H. Prenen, R. P. Zahedi, et al. 2017. The cell line (YT cells). J. Immunol. 134: 1623–1630. mTOR and PP2A pathways regulate PHD2 phosphorylation to fine-tune HIF1a 23. Naldini, L., U. Blo¨mer, P. Gallay, D. Ory, R. Mulligan, F. H. Gage, I. M. Verma, levels and colorectal cancer cell survival under Hypoxia. Cell Rep. 18: 1699– and D. Trono. 1996. In vivo gene delivery and stable transduction of nondividing 1712. cells by a lentiviral vector. Science 272: 263–267. 42. Ross, S. H., and D. A. Cantrell. 2018. Signaling and function of interleukin-2 in

24. Zufferey, R., T. Dull, R. J. Mandel, A. Bukovsky, D. Quiroz, L. Naldini, and T lymphocytes. Annu. Rev. Immunol. 36: 411–433. http://www.jimmunol.org/ D. Trono. 1998. Self-inactivating lentivirus vector for safe and efficient in vivo 43. Liu, G., K. Yang, S. Burns, S. Shrestha, and H. Chi. 2010. The S1P(1)-mTOR gene delivery. J. Virol. 72: 9873–9880. axis directs the reciprocal differentiation of T(H)1 and T(reg) cells. Nat. 25. Dull, T., R. Zufferey, M. Kelly, R. J. Mandel, M. Nguyen, D. Trono, and Immunol. 11: 1047–1056. L. Naldini. 1998. A third-generation lentivirus vector with a conditional pack- 44. Battaglia, M., A. Stabilini, and M. G. Roncarolo. 2005. Rapamycin selectively aging system. J. Virol. 72: 8463–8471. expands CD4+CD25+FoxP3+ regulatory T cells. Blood 105: 4743–4748. 26. Yu, A., and T. R. Malek. 2001. The proteasome regulates receptor-mediated 45. Cheng, G., A. Yu, and T. R. Malek. 2011. T-cell tolerance and the multi- endocytosis of interleukin-2. J. Biol. Chem. 276: 381–385. functional role of IL-2R signaling in T-regulatory cells. Immunol. Rev. 241: 27. Baharians, Z., and A. H. Scho¨nthal. 1998. Autoregulation of protein phosphatase 63–76. type 2A expression. J. Biol. Chem. 273: 19019–19024. 46. O’Shea, J. J., D. M. Schwartz, A. V. Villarino, M. Gadina, I. B. McInnes, and 28. Malek, T. R., and A. L. Bayer. 2004. Tolerance, not immunity, crucially depends A. Laurence. 2015. The JAK-STAT pathway: impact on human disease and on IL-2. Nat. Rev. Immunol. 4: 665–674. therapeutic intervention. Annu. Rev. Med. 66: 311–328.

29. Kirken, R. A., H. Rui, M. G. Malabarba, O. M. Howard, M. Kawamura, 47. Snow, J. W., N. Abraham, M. C. Ma, B. G. Herndier, A. W. Pastuszak, and by guest on September 30, 2021 J. J. O’Shea, and W. L. Farrar. 1995. Activation of JAK3, but not JAK1, is critical M. A. Goldsmith. 2003. Loss of tolerance and autoimmunity affecting multiple for IL-2-induced proliferation and STAT5 recruitment by a COOH-terminal organs in STAT5A/5B-deficient mice. J. Immunol. 171: 5042–5050. region of the IL-2 receptor beta-chain. Cytokine 7: 689–700. 48. Yao, Z., Y. Kanno, M. Kerenyi, G. Stephens, L. Durant, W. T. Watford, 30. He, J., X. Zhang, Y. Wei, X. Sun, Y. Chen, J. Deng, Y. Jin, Y. Gan, X. Hu, R. Jia, A. Laurence, G. W. Robinson, E. M. Shevach, R. Moriggl, et al. 2007. Nonre- et al. 2016. Low-dose interleukin-2 treatment selectively modulates CD4(+) T cell dundant roles for Stat5a/b in directly regulating Foxp3. Blood 109: 4368–4375. subsets in patients with systemic lupus erythematosus. Nat. Med. 22: 991–993. 49. Cohen, A. C., K. C. Nadeau, W. Tu, V. Hwa, K. Dionis, L. Bezrodnik, A. Teper, 31. Juang, Y. T., Y. Wang, G. Jiang, H. B. Peng, S. Ergin, M. Finnell, A. Magilavy, M. Gaillard, J. Heinrich, A. M. Krensky, et al. 2006. Cutting edge: decreased V. C. Kyttaris, and G. C. Tsokos. 2008. PP2A dephosphorylates Elf-1 and de- accumulation and regulatory function of CD4+ CD25(high) T cells in human termines the expression of CD3zeta and FcRgamma in human systemic lupus STAT5b deficiency. J. Immunol. 177: 2770–2774. erythematosus T cells. J. Immunol. 181: 3658–3664. 50. Kanai, T., S. Seki, J. A. Jenks, A. Kohli, T. Kawli, D. P. Martin, M. Snyder, 32. Juang, Y. T., T. Rauen, Y. Wang, K. Ichinose, K. Benedyk, K. Tenbrock, and R. Bacchetta, and K. C. Nadeau. 2014. Identification of STAT5A and STAT5B G. C. Tsokos. 2011. Transcriptional activation of the cAMP-responsive modu- target genes in human T cells. PLoS One 9: e86790. lator promoter in human T cells is regulated by protein phosphatase 2A-mediated 51. Basham, B., M. Sathe, J. Grein, T. McClanahan, A. D’Andrea, E. Lees, and dephosphorylation of SP-1 and reflects disease activity in patients with systemic A. Rascle. 2008. In vivo identification of novel STAT5 target genes. Nucleic lupus erythematosus. J. Biol. Chem. 286: 1795–1801. Acids Res. 36: 3802–3818. 33.Walsh,P.T.,J.L.Buckler,J.Zhang,A.E.Gelman,N.M.Dalton,D.K.Taylor, 52. Sharabi, A., M. G. Tsokos, Y. Ding, T. R. Malek, D. Klatzmann, and S. J. Bensinger, W. W. Hancock, and L. A. Turka. 2006. PTEN inhibits IL-2 receptor- G. C. Tsokos. 2018. Regulatory T cells in the treatment of disease. Nat. Rev. mediated expansion of CD4+ CD25+ Tregs. J. Clin. Invest. 116: 2521–2531. Drug Discov. 17: 823–844. Supplementary Figure 1

A PBMCs (Day 0) B Pre-culture D5 Post-culture

+ CD4 Cells Enrichment (Day 0) 92.2% 95.7% (Microbeads Selection)

Sorting

Sorted cells (Day 0) Foxp3 (CD4+CD25highCD127low) D In vitro expansion (Day 0) (SFM+IL2+Dynabeads CD3/CD28) (Transduction on Day 1)

Subculture (Day 3) % Inhibition (SFM+IL2)

Analysis (Day 4/5) C Responders/Tregs Ratio

Supplementary Figure 1: In vitro expansion of human Tregs for gene modification experiments. (A) Flow chart of Tregs expansion and transduction in vitro. (B) Representative histogram of Foxp3 expression by purified input Tregs and 5-day expanded Tregs without transduction. (C) pSTAT5 dose-response curves (left) and nonlinear regression analysis of the binding data (middle) of purified Tregs on day 0 and cultured Tregs on day 5. The EC50 was normalized based on the values on day 0 and analyzed using a one-sample one-sided t test (right) (n=3). (D) In vitro suppression assay of non-transduced Tregs after 5-day expansion (n=3). Data in plots indicate means± SEM of three independent experiments (C, D); ns, not significant (P > 0.05). Supplementary Figure 2

Supplementary Figure 2: PP2Ac-PE staining is blocked by pretreatment of unconjugated anti-PP2Ac antibody in expanded human Tregs. Expanded Tregs were pretreated with unlabeled anti-PP2Ac with the indicated amounts for 1 h, followed by staining with PP2Ac-PE for 1 h. Representative histograms show the peak shifts as the amount of the unlabeled antibody increases. Numbers represent the MFI of PP2Ac. Supplementary Table 1: Primers used for RT-qPCRs.

Gene Gene ID Forward Reverse Size Location Identifier (bp)

NM_002715 PPP2CA GCC TCT GCG AGA AGG CTA AA TAG ATG GCG AGA GAC CAC CA 432 487-918 (Exon 1-4)

NM_000417 CD25 CGT CTG CAA AAT GAC CCA CG CTG AGG CTT CTC TTC ACC TGG 103 717-819 (Exon 2-3)

NM_014009 FOXP3 TTT CAT GCA CCA GCT CTC AAC CTG ATC ATG GCT GGG CTC TC 87 491-577 (Exon 3-4)

NM_000633 BCL2 CTT TGA GTT CGG TGG GGT CA GGG CCG TAC AGT TCC ACA AA 162 943-1104 (Exon 2-3)

NM_001621 AHR GCC GGT GCA GAA AAC AGT AAA GGT CTC TAT GCC GCT TGG AA 74 664-737 (Exon 1-2)

NM_003745 SOCS1 GTG CAC GCA GCA TTA ACT GG GGA GGG TAC CCA CAT GGT TC 71 806-876 (Exon 2)

NM_003955 SOCS3 ATT CGG GAC CAG CCC CC AAA CTT GCT GTG GGT GAC CA 121 317-437 (Exon 1-2)

NM_002046 GAPDH ACC ACA GTC CAT GCC ATC AC TCC ACC ACC CTG TTG CTG T 451 602-1053 (Exon 8-9)