of May 25, 2017.This information is current as Lymphocytes in Rheumatoid Arthritis

Hyporesponsiveness of Synovial Fluid T Redox Balance Alterations Results inCells from the Plasma Membrane Due to Displacement of Linker for Activation of T

C. Breedveld and Cornelis L. VerweijPapendrecht-van der Voort, Philip H. J. Remans, Ferdinand Sonja I. Gringhuis, Angela Leow, Ellen A. M.

http://www.jimmunol.org/content/164/4/2170doi: 10.4049/jimmunol.164.4.2170

2000; 164:2170-2179; ;J Immunol 

Referenceshttp://www.jimmunol.org/content/164/4/2170.full#ref-list-1

, 23 of which you can access for free at: cites 60 articlesThis article

Subscriptionhttp://jimmunol.org/subscription

is online at: The Journal of ImmunologyInformation about subscribing to

Permissionshttp://www.aai.org/About/Publications/JI/copyright.htmlSubmit copyright permission requests at:

Email Alertshttp://jimmunol.org/alertsReceive free email-alerts when new articles cite this article. Sign up at:

Print ISSN: 0022-1767 Online ISSN: 1550-6606. Immunologists All rights reserved.Copyright © 2000 by The American Association of1451 Rockville Pike, Suite 650, Rockville, MD 20852The American Association of Immunologists, Inc.,

is published twice each month byThe Journal of Immunology

by guest on May 25, 2017

http://ww

w.jim

munol.org/

Dow

nloaded from

by guest on May 25, 2017

http://ww

w.jim

munol.org/

Dow

nloaded from

Displacement of Linker for Activation of T Cells from thePlasma Membrane Due to Redox Balance Alterations Resultsin Hyporesponsiveness of Synovial Fluid T Lymphocytes inRheumatoid Arthritis

Sonja I. Gringhuis,1 Angela Leow, Ellen A. M. Papendrecht-van der Voort, Philip H. J. Remans,Ferdinand C. Breedveld, and Cornelis L. Verweij

The T lymphocytes that reside in the synovium of the inflamed joints in patients with rheumatoid arthritis display severe hypo-responsiveness upon antigenic stimulation, which is probably due to their constant subjection to high levels of oxidative stress.Here we report that the synovial fluid T lymphocytes exert severely impaired phosphorylation of the adaptor protein linker foractivation of T cells (LAT), a crucial component of the TCR-mediated signaling pathways. In healthy T lymphocytes, LAT is amembrane-bound protein and becomes phosphorylated byz-associated protein of 70 kDa (ZAP-70) upon TCR engagement. Themolecular basis underlying the deficient phosphorylation of LAT and consequently the hyporesponsiveness of the synovial fluidT lymphocytes lies in the membrane displacement of LAT. We demonstrate that the subcellular localization of LAT is sensitiveto changes in the intracellular levels of the antioxidant glutathione. The membrane anchorage of LAT, and consequently thephosphorylation of LAT and the cellular activation of the synovial fluid T lymphocytes upon TCR engagement, is restored insynovial fluid T lymphocytes after supplementation of the intracellular glutathione levels withN-acetyl-L-cysteine. These datasuggest a role for the membrane displacement of LAT in the hyporesponsiveness of the synovial fluid T lymphocytes as aconsequence of oxidative stress.The Journal of Immunology,2000, 164: 2170–2179.

Rheumatoid arthritis (RA)2 is a chronic inflammatory jointdisease that eventually leads to the destruction of thejoint architecture. The synovium of the inflamed joints is

invaded by T lymphocytes, neutrophils, monocytes, B lympho-cytes, and dendritic cells. The role of the T lymphocytes in theorigin and propagation of the inflammation of the joints remainscontroversial. The synovial fluid (SF) T lymphocytes display anactivated phenotype (1–3), but react hyporesponsive to triggeringof the TCR/CD3 complex as determined by low proliferative re-sponses and the minimal production of IL-2 and IFN-g in responseto TCR stimulation (4–7). Although unable to respond to the Ag-dependent triggering of the TCR/CD3 complex, evidence existsthat the T lymphocytes in the inflamed joint induce damage of thejoint cartilage through Ag-independent cell-cell contact with mac-rophages (8–11). Furthermore, Salojin and coworkers recentlysuggested that the hyporesponsiveness of regulatory T lympho-cytes that normally confer protection from autoimmune diseasedue to altered TCR signaling might contribute to the breakdown of

tolerance and increase susceptibility to autoimmune disease (12).The prolonged residence of the T lymphocytes in the synoviumthus seems to contribute to and perpetuate the disease.

The hyporesponsive state of the SF T lymphocytes in RA cor-relates with markers of oxidative stress: the intracellular levels ofthe antioxidant glutathione (GSH) are significantly decreased in SFT lymphocytes. Moreover, the extracellular levels of another im-portant redox regulator, thioredoxin are significantly increased inthe SF (13, 14). GSH is an important regulator of the cellular redoxbalance and plays a major role in the protection against oxidativestress both by reacting directly with reactive oxygen radicals andacting as a substrate in the detoxification of H2O2 and organicperoxides. In addition, GSH is involved in maintaining the sulf-hydryl groups of intracellular proteins in a reduced state, which isrequired for their functional conformation (15, 16). It has previ-ously been shown that the depletion of the intracellular GSH levelsin T lymphocytes through treatment withDL-buthionine (S,R)-sul-foximine (BSO) results in hyporesponsiveness due to the abroga-tion of the proximal TCR-mediated signaling events (17).

The proximal signaling pathways induced by the engagement ofthe TCR/CD3 complex by ligand include the activation of twoclasses of protein tyrosine kinases, theSrc andSykfamilies. Therecruitment of theSykkinasez-associated protein of 70 kDa (ZAP-70) to the TCRz-chain plays a central role as it serves to bring thekinase in close proximity of its substrates (18–21). Two substratesof ZAP-70 have been identified: the adaptor proteins Src homol-ogy 2 (SH2) domain-containing leukocyte protein of 76 kDa (SLP-76) and linker for activation of T cells (LAT) (22–24). LAT hasrecently been cloned and identified as the 36- to 38-kDa proteinthat had previously been shown to play an important role in theTCR-mediated signaling pathways (22, 25–27). LAT is selectivelyexpressed by T lymphocytes and NK cells, where it is localized inthe plasma membrane through its N-terminala-helical structure

Department of Rheumatology, Leiden University Medical Center, Leiden, The Neth-erlands

Received for publication April 28, 1999. Accepted for publication December 3, 1999.

The costs of publication of this article were defrayed in part by the payment of pagecharges. This article must therefore be hereby markedadvertisementin accordancewith 18 U.S.C. Section 1734 solely to indicate this fact.1 Address correspondence and reprint requests to Dr. Sonja I. Gringhuis, Departmentof Rheumatology, C4-R, Leiden University Medical Center, P.O. Box 9600, 2300 RCLeiden, The Netherlands. E-mail address: gringhuis@mail.medfac.leidenuniv.nl2 Abbreviations used in this paper: RA, rheumatoid arthritis; SF, synovial fluid; GSH,glutathione; BSO,DL-buthionine (S,R)-sulfoximine; ZAP-70,z-associated protein of70 kDa; SH2, Src homology 2; SLP-76, SH2 domain-containing leukocyte protein of76 kDa; LAT, linker for activation of T cells; PLCg1, phospholipase Cg1; PB, pe-ripheral blood; NAC,N-acetyl-L-cysteine; pAb, polyclonal Ab; RT, room tempera-ture; ECL, enhanced chemiluminescence; MFI, mean fluorescence intensity; FasL,Fas ligand; PerCP, peridinin chlorophyl protein.

Copyright © 2000 by The American Association of Immunologists 0022-1767/00/$02.00

by guest on May 25, 2017

http://ww

w.jim

munol.org/

Dow

nloaded from

(22, 28). Several signaling proteins, including Grb2, the p85 sub-unit of phosphatidylinositol 3-kinase, Grb2-like accessory protein(Grap), and phospholipase Cg1 (PLCg1) associate with phosphor-ylated LAT through their SH2 domains to generate multiproteincomplexes that amplify the TCR-induced signals (22, 26–29).

The molecular basis underlying the hyporesponsiveness of theSF T lymphocytes in RA remains elusive, although some leadshave evolved. First, it has been observed that the expression of thesignaling chain of the TCR/CD3 complex, the TCRz-chain, isdown-regulated in SF T lymphocytes (30, 31). Second, the tyrosinephosphorylation of substrates upon TCR stimulation shows an ab-errant pattern in SF T lymphocytes. Specifically, the phosphory-lation of a protein of 36–38 kDa is absent in SF but not peripheralblood (PB) T lymphocytes from RA patients (31). Third, a moredistal event in the TCR-mediated signaling pathways, the influx ofCa21 from the endoplasmic reticulum, is reduced in PB T lym-phocytes from RA patients, which also display some degree ofhyporesponsiveness (32–34). These observations seem to be con-sistent with a defect in the proximal TCR-mediated signalingevents in SF T lymphocytes in RA.

In the present study, we set out to elucidate the molecular mech-anisms underlying the hyporesponsiveness of the T lymphocytes inthe synovium of RA patients and establish the role oxidative stressplays in this process. We present evidence that the 36- to 38-kDaphosphoprotein that does not become phosphorylated in TCR-stimulated SF T lymphocytes is the adaptor protein LAT. We alsoshow that LAT is displaced from the membrane in SF T lympho-cytes due to the severely reduced intracellular GSH levels. Thedisplacement of LAT is responsible for the abrogation of its phos-phorylation upon TCR stimulation, as well as the subsequent as-sociation of PLCg1 with LAT, hence blocking the TCR-inducedsignaling pathways leading to the influx of Ca21 and eventually tothe expression of IL-2. The treatment of SF T lymphocytes withN-acetyl-L-cysteine (NAC) elevates the intracellular GSH levelsand consequently restores the membrane localization and phos-phorylation of LAT and ultimately the cellular activation of the SFT lymphocytes.

Materials and MethodsT lymphocyte isolation

T lymphocytes from heparine-collected PB or SF of RA patients wereisolated through a negative selection procedure. Mononuclear cell suspen-sions were prepared by Ficoll-Hypaque density-gradient centrifugation.Monocytes, B lymphocytes, and NK cells were depleted by incubating withmAbs against CD14, CD16, and CD19 (10mg of each mAb per 203 106

of mononuclear cells; Central Laboratory of the Netherlands Red CrossBlood Transfusion Service (CLB), Amsterdam, The Netherlands) andsheep anti-mouse IgG coated-dynabeads (Dynal, Oslo, Norway) for 1.5 h,after which cells rosetted with immunomagnetic beads were removed witha Dynal magnetic particle concentrator. The remaining cell preparationscontained.95% T lymphocytes as assessed by flow cytometric analysisafter staining with a peridinin chlorophyl protein (PerCP)-conjugated anti-CD3 mAb (Becton Dickinson, San Jose, CA). Control human PB T lym-phocytes were obtained from healthy volunteer platelet donors, and mono-nuclear cell suspensions were prepared by Ficoll-Hypaque density-gradientcentrifugation. T lymphocytes were isolated by 2-aminoethylisothiouro-nium bromide-treated SRBC rosetting. The SRBC were lysed with 155mM NH4Cl, 10 mM KHCO3, and 0.1 mM EDTA according to standardprocedures. The remaining cell preparations contained.92% T lympho-cytes as assessed by FACS analysis after staining with a PerCP-conjugatedanti-CD3 mAb (Becton Dickinson). After isolation, T lymphocytes werekept at 37°C, 5% CO2 in IMDM (Life Technologies, Gaithersburg, MD)containing 10% FCS (Life Technologies) supplemented with 100 U/mlpenicillin and 100mg/ml streptomycin (Boehringer Mannheim, Mannheim,Germany).

Stimulation

T lymphocytes (53 106/ml) were incubated for various time periods with1 mg/ml anti-CD3 mAb (1XE; CLB), in combination with 1mg/ml anti-CD28 mAb (15E8; CLB) as indicated. NAC (Sigma, St. Louis, MO) wasadded at a final concentration of 5 mM, while BSO (Sigma) was added ata final concentration of 200mM.

Immunoprecipitation, Western blotting, and immunodetection

Whole-cell lysates were prepared from 53 106 T lymphocytes, eitherunstimulated or stimulated with anti-CD3 for 3 min, when indicated afterincubation with either NAC or BSO. Cells were harvested, washed twicewith PBS, and lysed in 300ml lysis buffer (10 mM triethanolamine, pH 7.8,150 mM NaCl, 5 mM EDTA, 10 mM Na3VO4, 1% Nonidet P-40) sup-plemented with protease inhibitors (10mg/ml leupeptin (Sigma), 0.4 mMPMSF (Sigma)) during 45 min on ice. Insoluble debris was spun down ina microcentrifuge for 15 min at 4°C.

To detect tyrosine phosphorylation of LAT or association of PLCg1with LAT, LAT was immunoprecipitated from 300ml of cell lysate (equiv-alent to 53 106 cells) by incubating with 3mg rabbit anti-LAT pAb(06-807; Upstate Biotechnology, Lake Placid, NY) during 16 h at 4°Cwhile rotating, and an additional 2 h after the addition of 25ml of proteinA-Agarose beads (50% slurry; Santa Cruz Biotechnology, Santa Cruz,CA). The immunoprecipitates were spun down in a microcentrifuge for30 s at 4°C, washed twice with lysis buffer, and then boiled in 13SDS-PAGE sample buffer. The proteins were separated by SDS-PAGE on a 5%(PLCg1 detection) or 10% (LAT detection) gel, using Rainbow-coloredprotein m.w. markers (Amersham, Little Chalfont, U.K.) as a reference,and transferred onto a polyvinylidene difluoride membrane (Millipore,Bedford, MA). The membrane was blocked in PBS containing 5% skinmilk and 0.01% Tween 20 during 1 h. Detection of tyrosine phosphorylatedproteins was performed by incubating the membranes with a mouse mAbagainst phospho-Tyr (PY99) (1:1000;sc-7020; Santa Cruz Biotechnology),while LAT and PLCg1 were detected by incubating the membranes withthe rabbit polyclonal Abs (pAbs) against LAT (1:1000) and PLCg1 (1:250)for 16 h. The membranes were subsequently incubated with the appropriatesecondary Abs (HRP-conjugated rabbit anti-mouse Ig (1:5000; Dako,Glostrup, Denmark) or swine anti-rabbit Ig-HRP (1:5000; Dako)) for 3 h,and then assayed using the enhanced chemiluminescence (ECL) detectionsystem (Amersham). Membranes were stripped of bound Abs by incubat-ing the membranes for 30 min at 50°C in stripping buffer (100 mM 2-ME,2% SDS, 62.5 mM Tris-HCl, pH 6.7) to allow for a second round ofdetection.

Flow cytometric analysis

A total of 2.5 3 105 T lymphocytes were fixed for 10 min in 1%para-formaldehyde in PBS at room temperature (RT), washed twice with PBS,and stored overnight at 4°C. The expression of LAT, ZAP-70, or PLCg1was measured via a two-step intracellular staining procedure. Cells wereincubated for 10 min at RT in PBS containing 1 mg/ml BSA (Sigma), 10%FCS, and 100mg/ml saponin (Sigma) as a permeabilizing agent. Cells werewashed three times with PBS/BSA/saponin and incubated for 1 h at 4°Cwith appropriate dilutions of the primary Abs (LAT, 1:250; ZAP-70 pAb(sc-574; Santa Cruz Biotechnology), 1:35; PLCg1 pAb (sc-81; Santa CruzBiotechnology), 1:35). After three washes (PBS/BSA/saponin), cells wereincubated for 30 min at 4°C with PE (R-PE)-conjugated goat anti-rabbitIgG (Harlan Sera-Lab, Crawley Down, U.K.). A negative control was in-cubated with the secondary Ab only. After three washes (PBS/BSA/sapo-nin), the free F(ab9)2 sites of cell-bound goat anti-rabbit-PE were blockedusing 10% normal rabbit serum, whereafter cells were stained with aPerCP-conjugated anti-CD3 mAb (Becton Dickinson) (1:25) for 1 h at 4°C.CD3-positive cells were analyzed on a FACScan using the CellQuest soft-ware (Becton Dickinson). The levels of LAT, ZAP-70, and PLCg1 areexpressed as the ratio of the mean fluorescence intensity (MFI), i.e., theMFI of LAT, ZAP-70, or PLCg1 divided by the MFI of the negativecontrol.

ZAP-70 kinase activity assay

Total cell lysates were prepared from 53 106 T lymphocytes, either un-stimulated or stimulated with anti-CD3 for 3 min. Cells were harvested andwashed once with PBS, resuspended in 500ml lysis buffer (20 mM HEPES,pH 7.4, 2 mM EGTA, 1 mM DTT, 1% Triton X-100, 10% glycerol) sup-plemented with protease inhibitors (10mg/ml leupeptin, 0.4 mM PMSF)and phosphatase inhibitors (50 mMb-glycerophosphate, 1 mM Na3VO4),and incubated on ice for 30 min. Insoluble debris was spun down in amicrocentrifuge for 15 min at 4°C.

2171The Journal of Immunology

by guest on May 25, 2017

http://ww

w.jim

munol.org/

Dow

nloaded from

ZAP-70 was immunoprecipitated from 500ml of cell lysate (equivalentto 53 106 cells) by incubating with 800 ng rabbit anti-ZAP-70 pAb (SantaCruz Biotechnology) during 1 h at 4°C while rotating, and an additional16 h after the addition of 25ml of protein A-Agarose beads (50% slurry;Santa Cruz Biotechnology). The immunoprecipitates were spun down in amicrocentrifuge for 30 s at 4°C, washed twice with lysis buffer, twice withLiCl buffer (500 mM LiCl, 100 mM Tris-HCl, pH 7.6, 1 mM DTT, 0.1%Triton X-100), and finally twice with assay buffer (25 mM Tris-HCl, pH7.0, 150 mM NaCl, 10 mM MnCl2, 1 mM DTT, 0.1 mM Na3VO4, 0.1%Triton X-100). The immunoprecipitates were assayed for kinase activity in30 ml assay mix containing 25 mM Tris-HCl, pH 7.0, 150 mM NaCl, 10mM MnCl2, 1 mM DTT, 0.1 mM Na3VO4, 0.1% Triton X-100, 250mMATP, and 10mCi [g-32P]ATP (3000 Ci/mmol; Amersham) during 20 minat 30°C, using 10mg tubulin (Sigma) as substrate. The reactions wereterminated by the addition of 53SDS-PAGE sample buffer and boiling.Proteins were separated by SDS-PAGE on a 10% gel, using Rainbow-colored protein m.w. markers (Amersham) as a reference. Quantification ofphosphorylated substrates was performed using a PhosphorImaging systemand the ImageQuant software (Molecular Dynamics, Sunnyvale, CA).

Immunofluorescence staining and microscopy

A total of 13 105 T lymphocytes were mounted onto adhesive microscopeslides, air dried, and kept frozen until staining. Before staining, the cellswere fixed in 4%para-formaldehyde in PBS for 15 min at RT. After threewashes in PBS containing 5% BSA, cells were permeabilized using 0.1%Triton X-100 in PBS for 4 min at RT, washed three times (PBS/BSA), andpreblocked for 45 min at RT in PBS containing 10% FCS. Cells were thenincubated with a rabbit pAb against LAT (1:250) in PBS/FCS for 45 minat RT, washed three times (PBS/BSA), and incubated with FITC-conju-gated swine anti-rabbit Ig (1:100; Dako) in PBS/FCS for 45 min at RT. Anegative control was incubated with the secondary Ab only. After threefinal washes (PBS), cells were imbedded in 1 mg/mlp-phenyleendiamine(Sigma) in 90% glycerol, 10% PBS and covered with a coverslip. To de-termine whether the cell membrane was intact after BSO treatment, cellswere stained immediately after fixation with a mouse mAb against CD3(1:25; Becton Dickinson) and detected using tetramethylrhodamineisothiocyanate (TRITC)-conjugated rabbit anti-mouse Ig (1:200; Dako).Cells were viewed using a Leitz Aristoplan microscope (Leica, Wetzlar,Germany) equipped with a 1003objective and optics for FITC andTRITC. Images were taken with a Sony 3 CCD Color Video Camera,model DXC-9508 (Sony, Tokyo, Japan).

Preparation of membrane and cytoplasmic cell fractions

A total of 10 3 106 T lymphocytes were disrupted in 500ml extractionbuffer (50 mM Tris-HCl, pH 7.0, 10 mM KCl, 1 mM CaCl2, 1 mM MgCl2,1 mM Na3VO4) supplemented with protease inhibitors (10mg/ml leupep-tin, 0.4 mM PMSF) by shearing through a 25-gauge needle. After centrif-ugation at 16003 g for 10 min at 4°C, this extraction step was repeated on

the pellet. The membrane and cytoplasmic fractions were separated fromthe pooled supernatants at 100,0003 g for 60 min at 4°C. The supernatantafter ultracentrifugation contains the cytoplasmic proteins (Triton X-100was added to 1%). The pellet containing the membrane proteins was re-suspended in 500ml buffer (10 mM triethanolamine, pH 7.8, 150 mMNaCl, 5 mM EDTA, 1 mM Na3VO4, 1% Nonidet P-40) supplemented withprotease inhibitors (10mg/ml leupeptin, 0.4 mM PMSF). LAT was immu-noprecipitated from the extracts and detected with ECL Western blotting asdescribed above.

Measurement of secreted IL-2 protein

Human T lymphocytes (33 106/ml) were left unstimulated or stimulatedwith anti-CD3 and anti-CD28 for 24 h when indicated after incubation witheither NAC or BSO. Secreted IL-2 protein was quantified in cell-free su-pernatants using a human IL-2 ELISA kit (Genzyme, Cambridge, MA) asrecommended by the manufacturer.

Statistical analysis

Statistical analyses were performed on the data using the Fisher’s exact testfor paired observations. Statistical significance of the data was set atp , 0.05.

Patient identification

Patients enrolled in this study fulfilled the 1987 revised criteria of theAmerican College of Rheumatology for rheumatoid arthritis, differed be-tween 27 and 72 years of age (mean age 57 years), with a female:male ratioof 9:8 (see Table I).

ResultsLAT remains unphosphorylated in SF T lymphocytes, but not PBT lymphocytes from RA patients upon TCR stimulation

The tyrosine phosphorylation pattern upon TCR stimulation of SFT lymphocytes from RA patients is aberrant, with, most particular,the absence of the phosphorylation of a 36- to 38-kDa protein (31).To determine whether this phosphoprotein of 36–38 kDa is iden-tical with the adaptor protein LAT, we immunoprecipitated LATfrom whole-cell extracts of SF and PB T lymphocytes from RApatients that were either left unstimulated or stimulated for 3 minwith anti-CD3 and assessed the phosphorylation status of LATthrough Western blotting analysis. We observed no phosphory-lated LAT in unstimulated T lymphocytes. In PB T lymphocytesfrom RA patients, LAT becomes extensively phosphorylated ontyrosine residues upon TCR stimulation, as occurs in T lympho-cytes from healthy controls (Fig. 1). However, after stimulation of

Table I. Clinical and demographic data of RA patients

No. GenderAge

(years)

DiseaseDuration(years)

RheumatoidFactor Medicationa Experimentsb

1 M 40 6 1 d, s 4E2 F 51 14 1 CsA, h 1; 23 F 70 17 1 p, paracetamol 24 F 74 11 – N 2; 4,A–D5 M 48 5 1 M, p, s 1; 4,A–D6 M 55 24 1 M 27 M 55 21 1 2; 4, A–D8 M 64 14 1 M, N 19 M 46 2 – N, s 3

10 F 71 4 – d, l 2; 4,A–D11 F 53 17 – d, M 612 F 56 15 1 M, N, paracetamol 613 F 65 22 1 h, M, s, ibuprofen 614 F 66 43 1 N 315 M 39 5 1 d, s 6; 716 F 27 8 1 d, M, N 317 M 72 5 1 d 4E

a d, Diclofenac; s, salazopyrine; CsA, cyclosporin A; h, hydroxychloroquine; p, prednisone; N, nonsteroidal antiinflammatorydrug; M, methotrexate; l, leflunomide.

b Numbers refer to figure numbers.

2172 MEMBRANE DISPLACEMENT OF LAT IN RA

by guest on May 25, 2017

http://ww

w.jim

munol.org/

Dow

nloaded from

the TCR/CD3 complex in SF T lymphocytes from RA patients, wedid not detect any phosphorylated LAT, indicating that the missing36- to 38-kDa phosphoprotein in SF T lymphocytes is indeed LAT(Fig. 1). Western blotting analysis revealed that the impaired phos-phorylation of LAT is not due to a lack of expression of LAT,because we could detect LAT in extracts from SF as well as PB Tlymphocytes from RA patients, although the level of LAT expres-sion seemed slightly lower in both SF and PB T lymphocytes com-pared with the expression of LAT in T lymphocytes from healthycontrols (Fig. 1). To quantitatively determine the expression of

LAT at the individual cell level, we measured the MFI after theintracellular immunofluorescence staining of LAT by flow cyto-metric analysis. The MFI ratio for LAT in both PB and SF Tlymphocytes was;2-fold lower compared with the LAT MFI ra-tio in T lymphocytes from healthy controls: 104.46 5.1 (mean6SEM; n 5 5) for healthy controls vs 52.26 2.2 (mean6 SEM;n 5 4; p 5 0.003) for PB T lymphocytes and 48.96 1.7 (mean6SEM;n 5 4; p , 0.001) for SF T lymphocytes (Fig. 2). These datasupport the conclusion that the significantly decreased expressionof LAT in SF T lymphocytes does not account for the lack ofphosphorylation of LAT, because the phosphorylation of LAT re-mains intact in PB T lymphocytes from RA patients that containsimilar levels of LAT.

The expression and kinase activity of ZAP-70 is intact in SF Tlymphocytes

The phosphorylation of LAT upon TCR ligation is regulated by thesyk family tyrosine kinase ZAP-70 (22). One explanation for thelack of phosphorylation of LAT in SF T lymphocytes could be thateither the expression of ZAP-70 or the activation of the kinaseactivity of ZAP-70 is affected by the severely reduced intracellularGSH levels. The expression of ZAP-70 was similar in T lympho-cytes from healthy controls and RA patients (MFI ratios: 24.360.9 (mean6 SEM; n 5 6) for healthy controls, 26.56 1.6(mean6 SEM;n 5 4) for PB T lymphocytes, 25.16 1.5 (mean6SEM; n 5 5) for SF T lymphocytes; Fig. 2). To determine thekinase activity of ZAP-70, we performed a kinase assay specificfor ZAP-70. ZAP-70 was immunoprecipitated from whole-cell ex-tracts of SF and PB T lymphocytes from RA patients that wereeither left unstimulated or stimulated for 3 min with anti-CD3, andthe specific ZAP-70 kinase activity was determined by measuringthe phosphorylation of its substrate, tubulin. The basal kinase ac-tivity of ZAP-70 present in unstimulated T lymphocytes from

FIGURE 1. LAT is phosphorylated in PB but not SF T lymphocytesfrom RA patients upon TCR stimulation. T lymphocytes from healthy con-trols (HC) and PB (PB T/RA) and SF T lymphocytes (SF T/RA) from RApatients were left unstimulated or stimulated with anti-CD3 for 3 min. LATwas immunoprecipitated from whole-cell lysates of an equivalent numberof cells, and the phosphorylation status of LAT was determined by ECLWestern blotting with anti-phosphotyrosine (PY) Abs and HRP-conjugatedrabbit anti-mouse Abs (top panel). The expression of LAT was determinedby ECL Western blotting with anti-LAT Abs and HRP-conjugated swineanti-rabbit Abs. The detected H chain comes from the LAT Abs used forthe immunoprecipitation (bottom panel). The data are representative of sixindependent experiments.

FIGURE 2. The expression of LAT is reducedin both PB and SF T lymphocytes from RA pa-tients, while the expression of ZAP-70 andPLCg1 is comparable to the expression in T lym-phocytes from healthy controls.A, PermeabilizedT lymphocytes from healthy controls (HC) as wellas PB (PB T/RA) and SF T lymphocytes (SFT/RA) from RA patients were labeled with spe-cific Abs for ZAP-70, LAT, and PLCg1 and thenstained with PE-conjugated goat anti-rabbit Abs.The MFI was measured by flow cytometric anal-ysis. The MFI of cells stained with the PE-conju-gated Abs only was also determined as a negativecontrol and (B) used to calculate the MFI ratio forZAP-70, LAT, and PLCg1 as described inMate-rials and Methods. The mean values6 SEM forthe MFI ratios found in four to six independentexperiments are shown.

2173The Journal of Immunology

by guest on May 25, 2017

http://ww

w.jim

munol.org/

Dow

nloaded from

healthy controls was enhanced;2.2-fold 6 0.23 (mean6 SEM;n 5 3) in anti-CD3-stimulated T lymphocytes (Fig. 3). The basalZAP-70 kinase activity in both unstimulated PB and unstimulatedSF T lymphocytes from RA patients is up-regulated almost 2.1-fold (2.1-fold 6 0.16 (mean6 SEM; n 5 3; p , 0.001) for PB Tlymphocytes, 2.1-fold6 0.07 (mean6 SEM; n 5 3; p 5 0.003)for SF T lymphocytes) upon TCR/CD3 stimulation to reach a levelof kinase activity that is only slightly lower than the kinase activityobserved in stimulated T lymphocytes from healthy controls (p 50.37 for PB T lymphocytes,p 5 0.17 for SF T lymphocytes; Fig.3). These results demonstrate that the defective phosphorylation ofLAT in SF T lymphocytes is not due to a defect in the kinaseactivity of ZAP-70.

LAT is displaced from the membrane and present in thecytoplasm of SF T lymphocytes

The recruitment of signaling proteins to the membrane is a com-mon theme in the proximal events of many signaling pathways (20,22, 35). To explore whether LAT is absent from the membrane inSF T lymphocytes and therefore the recruitment of ZAP-70 to themembrane upon TCR engagement does not result in the phosphor-ylation of LAT, we analyzed the subcellular localization of LATby immunofluorescence staining of T lymphocytes on cytospins. Inhealthy controls, LAT is clearly expressed in the plasma mem-brane of the T lymphocytes (Fig. 4B). In PB T lymphocytes fromRA patients, LAT is also localized in the cellular membrane, al-though the level of expression is clearly diminished compared withT lymphocytes from healthy controls (Fig. 4C). In contrast, in SFT lymphocytes from RA patients, we detected LAT in the cyto-plasm of the cell instead of the membrane (Fig. 4D). Western blotdetection of LAT in separated membrane and cytoplasm fractionsconfirmed the membrane and cytoplasm localization of LAT in,respectively, PB and SF T lymphocytes from RA patients (Fig.4E). These results signify that the displacement of LAT from the

membrane in SF T lymphocytes is responsible for its deficientphosphorylation.

The localization of LAT is sensitive to alterations in theintracellular GSH levels

The intracellular GSH levels are significantly reduced in SF butnot PB T lymphocytes from RA patients (13). To examine whetherthe decreased intracellular GSH levels play a role in the membranedisplacement of LAT, we treated T lymphocytes from healthy con-trols with BSO to decrease the intracellular GSH reserves (13, 17)and determined the subcellular localization of LAT at successivetime points. Immunofluorescence staining of LAT revealed thatLAT still resided in the cellular membrane of the T lymphocytesafter 16 h of treatment with BSO (Fig. 5B). However, after 48 h ofBSO treatment, LAT started to accumulate in the cytoplasm, al-though we also could detect membrane-associated LAT (Fig. 5C).After 72 h of BSO treatment, we did not observe any expression ofLAT in the plasma membrane, while the cytoplasm showed mas-sive expression of LAT (Fig. 5D). Immunofluorescence stainingfor the membrane receptor CD3 ascertained that the cellular mem-

FIGURE 3. The ZAP-70 kinase activity is similar in T lymphocytesfrom healthy controls and RA patients. T lymphocytes from healthy con-trols (HC) and PB (PB T/RA) and SF T lymphocytes (SF T/RA) from RApatients were left unstimulated or stimulated with anti-CD3 for 3 min.ZAP-70 was immunoprecipitated from whole-cell lysates of an equivalentnumber of cells and was assayed for kinase activity using tubulin as asubstrate. The kinase assay shown is representative of three independentexperiments. The specific ZAP-70 kinase activity was determined by quan-tification of phosphorylated tubulin using a PhosphorImaging system. Thelower graph shows the relative phosphorylation of tubulin. The phosphor-ylation of tubulin detected in unstimulated T lymphocytes from healthycontrols was set at 1. The values shown correspond to the kinase assaydepicted above.

FIGURE 4. LAT is present in the cytoplasm of SF T lymphocytes fromRA patients instead of the cellular membrane as in T lymphocytes fromhealthy controls and PB T lymphocytes from RA patients. Fixed, perme-abilized T lymphocytes from healthy controls (B) and PB (C) and SF Tlymphocytes (D) from RA patients were labeled with specific Abs for LATand then stained with FITC-conjugated swine anti-rabbit Abs. A negativecontrol (A) was stained with secondary Abs only. The data are represen-tative of four independent experiments.E, Membrane and cytoplasmicfractions of PB and SF T lymphocytes were prepared. LAT was immuno-precipated and then detected by ECL Western blotting with anti-LAT Absand HRP-conjugated swine anti-rabbit Abs. The detected H chain comesfrom the LAT Abs used for the immunoprecipitation. The data are repre-sentative of two independent experiments.

2174 MEMBRANE DISPLACEMENT OF LAT IN RA

by guest on May 25, 2017

http://ww

w.jim

munol.org/

Dow

nloaded from

brane of the T lymphocytes was still intact after 72 h of BSOtreatment (Fig. 5E). Moreover, the cellular activation of the T lym-phocytes treated for 72 h with BSO was completely impaired be-cause the cells were unable to produce any IL-2 while untreated Tlymphocytes produced high levels of IL-2 after 24 h of stimulationwith anti-CD3 and anti-CD28 (data not shown). T lymphocytescultured for 72 h in medium showed normal LAT membrane lo-calization (Fig. 5F). These results suggest that the membrane dis-placement of LAT in SF T lymphocytes from RA patients is dueto the severely decreased intracellular levels of GSH.

Next, we wanted to determine whether increasing the intracel-lular GSH concentration would restore the membrane localizationof LAT in SF T lymphocytes. The supplementation of SF T lym-phocytes from RA patients with the antioxidant NAC, which is aprecursor in the synthesis of GSH, has previously been shown torestore the intracellular GSH levels (13). Immunofluorescencestaining of LAT in SF T lymphocytes that had been treated withNAC for 48 h showed that LAT was exclusively expressed in theplasma membrane (Fig. 6B). In SF T lymphocytes that were leftuntreated for 48 h, LAT was still present in the cytoplasm of thecells, excluding the possibility that LAT returns spontaneously tothe membrane after 48 h of culture (Fig. 6A). These data denotethat the subcellular localization of LAT is highly sensitive to theintracellular concentration of GSH.

Increasing the intracellular GSH levels in SF T lymphocytesrestores the phosphorylation of LAT, the association of PLCg1with LAT, and the secretion of IL-2 upon TCR stimulation

Because the elevation of the intracellular GSH levels restored themembrane localization of LAT in SF T lymphocytes from RApatients, we also examined whether the phosphorylation of LATand downstream signaling events in the TCR-mediated signalingpathways were restored. We immunoprecipitated LAT fromwhole-cell extracts of either resting or activated SF T lymphocytesboth before and after a 48-h treatment with NAC and determinedthe tyrosine phosphorylation of LAT through Western blottinganalysis. As shown before, LAT remained unphosphorylated inuntreated SF T lymphocytes from RA patients after TCR stimu-lation (Fig. 7A). However, in SF T lymphocytes that had beentreated with NAC for 48 h, LAT becomes extensively phosphor-ylated on tyrosine residues upon TCR ligation (Fig. 7A).

The association of PLCg1 with LAT upon TCR engagement,necessary to transmit the TCR-induced signal from LAT to theinositol 1,4,5-triphosphate-sensitive Ca21 channels, is dependenton the phosphorylation of LAT because the N-terminal SH2 do-main of PLCg1 interacts with the multiple phosphorylated tyrosineresidues of LAT (22, 25, 28, 36). FACS analysis showed that theexpression of PLCg1 was slightly but not significantly reduced inboth PB and SF T lymphocytes from RA patients compared withT lymphocytes from healthy controls: 31.26 1.8 (mean6 SEM;n 5 5) for healthy controls vs 27.26 1.6 (mean6 SEM; n 5 4)for PB T lymphocytes and 26.56 2.9 (mean6 SEM; n 5 4) forSF T lymphocytes (Fig. 2). To determine whether PLCg1 associ-ates with LAT, we subjected whole-cell extracts of either resting orstimulated T lymphocytes to immunoprecipitation with anti-LATAbs followed by Western blotting detection of PLCg1. PLCg1coimmunoprecipitated with LAT only in stimulated T lympho-cytes from healthy controls and also from PB T lymphocytes ofRA patients, but not in stimulated untreated SF T lymphocytesfrom RA patients due to the lack of phosphorylation of LAT (Fig.7B). To determine whether the modulation of the intracellular GSHlevels would restore the association of PLCg1 with LAT, we eval-uated the ability of PLCg1 to coimmunoprecipitate with LAT fromwhole-cell extracts of NAC-treated SF T lymphocytes. Indeed, af-ter treatment of the SF T lymphocytes with NAC for 48 h, not onlythe phosphorylation of LAT but also the association of PLCg1with LAT upon TCR stimulation was restored (Fig. 7B).

To establish whether the restoration of the membrane localiza-tion of LAT, and hence the phosphorylation of LAT and the as-sociation of PLCg1 with LAT upon TCR ligation, through increas-ing the intracellular GSH levels, also results in a reversal of thehyporesponsiveness of the SF T lymphocytes from RA patients,

FIGURE 5. LAT is gradually displaced from the plasma membrane in Tlymphocytes from healthy controls due to treatment with BSO. T lympho-cytes from healthy controls were cultured in medium for 72 h (F) or treatedwith BSO and after 0 h (A), 16 h (B), 48 h (C), or 72 h (D), fixed, per-meabilized, labeled with specific Abs for LAT, and then stained withFITC-conjugated swine anti-rabbit Abs.E, Fixed T lymphocytes fromhealthy controls after 72 h of treatment with BSO were also labeled withmonoclonal anti-CD3 Abs followed by staining with tetramethylrhodamineisothiocyanate-conjugated rabbit anti-mouse Abs to check the integrity ofthe plasma membrane. The data are representative of five independentexperiments.

FIGURE 6. The membrane localization of LAT is restored in SF T lym-phocytes from RA patients after treatment with NAC. SF T lymphocytesfrom RA patients were cultured in medium (A) or treated with NAC (B) for48 h and afterward fixed, permeabilized, labeled with specific Abs forLAT, and then stained with FITC-conjugated swine anti-rabbit Abs. Thedata are representative of four independent experiments.

2175The Journal of Immunology

by guest on May 25, 2017

http://ww

w.jim

munol.org/

Dow

nloaded from

we determined the secretion of IL-2 by the T lymphocytes in re-sponse to stimulation with anti-CD3 and anti-CD28 for 24 h. Un-stimulated T lymphocytes secreted only very low levels of IL-2.Stimulation of T lymphocytes from healthy controls with anti-CD3/anti-CD28 for 24 h resulted in a secretion of 23816 319pg/ml IL-2 (mean6 SEM; n 5 7; Fig. 7C). The IL-2 secretion ofanti-CD3 plus anti-CD28-stimulated PB T lymphocytes from RApatients was;2-fold lower: 13806 58 pg/ml IL-2 (mean6 SEM;n 5 4; Fig. 7C). The stimulation of untreated SF T lymphocytesinduced an IL-2 production of only 2936 58 pg/ml IL-2 (mean6SEM;n 5 4; Fig. 7C). The treatment of the SF T lymphocytes withNAC for 48 h elevated the IL-2 secretion upon anti-CD3/anti-CD28 stimulation significantly to a level of expression comparableto the expression by PB T lymphocytes: 13166 137 pg/ml IL-2(mean6 SEM; n 5 3; p 5 0.023; Fig. 7C). These results indicatethat the reduced GSH levels play a critical role in the abrogation ofthe cellular activation of the SF T lymphocytes in RA through themembrane displacement of LAT.

DiscussionIn RA, a common chronic inflammatory joint disease characterizedby the progressive, erosive destruction of joint cartilage and bone,the functioning of the T lymphocytes present at the site of inflam-mation, is severely impaired. In the present study, we have exam-ined the molecular basis for this hyporesponsive state, and wedemonstrate that the phosphorylation of the adaptor protein LAT,which plays a crucial role in the TCR-mediated signaling path-

ways, is dramatically impaired in SF T lymphocytes from RA pa-tients. Immunohistochemical analysis reveals that the reducedphosphorylation of LAT upon engagement of the TCR/CD3 com-plex correlates with the displacement of LAT from the membrane,a phenomenon that could be induced by modulating the intracel-lular redox balance, suggesting that the membrane displacement ofLAT is induced by oxidative stress. The supplementation of theGSH levels in SF T lymphocytes could reverse the membranedisplacement of LAT and the hyporesponsiveness of the cells.

Several observations have suggested that the T lymphocytes inthe inflamed joints of RA patients exhibit defects in the TCR-induced signal transduction routes rendering them unresponsiveupon mitogenic or antigenic stimulation. These observations in-clude a decrease in the expression levels of the TCRz-chain, anaberrant pattern of protein tyrosine phosphorylation, and a block inthe Ca21 influx upon TCR activation, any of which could contrib-ute to the hyporesponsiveness of the T lymphocytes (30–34).

The reduced expression of the TCRz-chain has been proposedas underlying mechanism responsible for the negative regulationof the TCR signaling pathways. The decreased expression of theTCR z-chain is not only observed in the hyporesponsive SF Tlymphocytes from RA patients but also in tumor-infiltrating T lym-phocytes from human cancer patients and in the hyporesponsive Tlymphocytes from HIV-infected individuals. In all cases, only theexpression of the TCRz-chain is reduced, while the expression ofother components of the TCR/CD3 complex remains constant (30,31, 37–39). Although the correlation between the hyporesponsive

FIGURE 7. The phosphorylation ofLAT, the association of PLCg1 withphosphorylated LAT, and the secretion ofIL-2 upon T cell activation is restored inSF T lymphocytes from RA patients aftertreatment with NAC. T lymphocytes fromhealthy controls (HC), PB T lymphocytesfrom RA patients (PB T/RA), and SF Tlymphocytes from RA patients (SFT/RA), either untreated or treated for 48 hwith NAC, were left unstimulated orstimulated with anti-CD3 for 3 min. LATwas then immunoprecipitated fromwhole-cell lysates of an equivalent num-ber of cells.A, The phosphorylation ofLAT was determined by ECL Westernblotting with anti-phosphotyrosine (PY)Abs and HRP-conjugated rabbit anti-mouse Abs.B, The association of PLCg1with LAT was determined by ECL West-ern blotting with anti-PLCg1 Abs andHRP-conjugated swine anti-rabbit Abs.The data are representative of three inde-pendent experiments.C, T lymphocytesfrom healthy controls (HC), PB T lym-phocytes from RA patients (PB T/RA),and SF T lymphocytes from RA patients(SF T/RA), either untreated or treated for48 h with NAC, were left unstimulated orstimulated with anti-CD3 plus anti-CD28.Cell-free supernatants were harvested af-ter 24 h and analyzed for secreted IL-2protein. The mean values6 SEM for theIL-2 secretion measured in three to sevenindependent experiments are shown.

2176 MEMBRANE DISPLACEMENT OF LAT IN RA

by guest on May 25, 2017

http://ww

w.jim

munol.org/

Dow

nloaded from

state of the T lymphocytes and the reduced TCRz expressionsuggests a causal relationship, formal evidence for this hypothesisis not available. Our data show that the kinase activity of ZAP-70is not affected in SF T lymphocytes from RA patients when com-pared with the ZAP-70 kinase activity in T lymphocytes fromhealthy controls and PB T lymphocytes from RA patients, sug-gesting that the signaling pathway from the TCRz-chain leading tothe activation of ZAP-70 is intact and functional.

Previously, we have reported the absence of a 36-to 38-kDaphosphoprotein in SF T lymphocytes upon TCR stimulation (31).A similar observation has been described for autoimmune nono-bese diabetic mice, where it was shown that a decrease in thephosphorylation of a 36-kDa protein upon TCR engagement inhyporesponsive T lymphocytes correlates with a decreased mem-brane targeting of PLCg1 and the Grb2/Sos complex, and an im-paired activation of Ras (40, 41). The current data indicate that inSF T lymphocytes from RA patients this phosphoprotein repre-sents the membrane-anchored 36-kDa adaptor protein LAT. Theexpression of LAT is reduced by 50% in SF T lymphocytes fromRA patients compared with T lymphocytes from healthy controls.However, the reduced expression is not likely to account for thedeficient phosphorylation of LAT, because the level of LAT ex-pression is SF T lymphocytes is approximately similar to the levelof expression in PB T lymphocytes from RA patients, which dem-onstrate efficient phosphorylation of LAT upon TCR engagement.Furthermore, the deficient phosphorylation of LAT is not due to adefect in the kinase activity of ZAP-70, because this activity isintact in SF T lymphocytes. Intracellular immunofluorescencestaining of LAT reveals that the impaired phosphorylation of LATin SF T lymphocytes correlates with a cytoplasmic localization ofLAT, whereas in T lymphocytes from healthy controls and PB Tlymphocytes from RA patients, which exert intact LAT phosphor-ylation, LAT is exclusively expressed in the membrane. These dataindicate that, due to the displacement of LAT from the cellularmembrane, the membrane recruitment of ZAP-70 in SF T lym-phocytes upon TCR engagement is ineffective in evoking the phos-phorylation of LAT. Furthermore, the membrane displacement andthe impaired phosphorylation of LAT in SF T lymphocytes alsocorrelate with an impaired recruitment of PLCg1 to the membranethrough association with phosphorylated LAT. These data indicatethat the disturbed subcellular localization of LAT accounts for thehyporesponsiveness displayed by the SF T lymphocytes throughthe abrogation of the TCR-induced signaling pathways.

The functioning of T lymphocytes is markedly influenced byalterations in the intracellular redox balance. Exposure to reactiveoxygen radicals has been demonstrated to down-regulate the ac-tivity of T lymphocytes (42–44). In SF T lymphocytes from RApatients, a significant decrease of the intracellular GSH levels hasbeen shown to correlate with the hyporesponsive state of thesecells (13). A similar decrease in the intracellular GSH levels hasbeen reported for T lymphocytes from HIV-infected individuals(45). Moreover, the hyporesponsiveness displayed by the SF Tlymphocytes from RA patients clearly resembles the hyporespon-siveness observed in aging T lymphocytes. Aging T lymphocytescontain severely reduced intracellular levels of antioxidants in-cluding GSH, which inhibits the phosphorylation of a 35- to 36-kDa protein, which is most likely LAT, the phosphorylation ofPLCg1, the generation of inositol 1,4,5-triphosphate, and the in-flux of Ca21 upon TCR engagement. Increasing the intracellularGSH levels in aged T lymphocytes by either GSH or NAC treat-ment restores the generation of inositol 1,4,5-triphosphate and theCa21 influx in response to TCR ligation (46, 47). The data pre-sented here indicate that redox balance alterations play a criticalrole in the abrogation of the cellular activation of the SF T lym-

phocytes in RA through the membrane displacement of LAT. Thetreatment of SF T lymphocytes with NAC not only restores themembrane localization of LAT, but also the phosphorylation ofLAT upon TCR engagement and the subsequent association ofPLCg1 with phosphorylated LAT. By mimicking the situationof the SF T lymphocytes through the treatment of T lympho-cytesfrom healthy controls with BSO, which depletes the intra-cellular GSH levels, we could induce the membrane displacementof LAT. Similar results were reported upon the treatment of Tlymphocytes withN-ethylmaleimide, which induces the sulfhydryloxidation of intracellular proteins. InN-ethylmaleimide-treated Tlymphocytes, the phosphorylation of a 35-kDa protein, which mostlikely represents LAT, upon TCR stimulation is inhibited. Also,the tyrosine phosphorylation of PLCg1 is abrogated, probably as aresult of the impaired membrane recruitment of PLCg1 becausethe phosphorylation of LAT is also abrogated (48).

Preliminary data indicate that the membrane displacement ofLAT due to decreased GSH levels is relatively specific for RA ascompared with other arthritic diseases. Only in patients with pso-riatic arthritis have we so far observed a partial displacement ofLAT from the membrane, but not in patients with spondylarthropa-thy or juvenile RA (our unpublished observations). Further studiesare in progress to establish the disease-specificity of the membranedisplacement of LAT.

Several hypotheses can be proposed to explain the displacementof LAT from the membrane. First, a posttranslational modificationof LAT resulting in a conformational change could account for themembrane displacement of LAT. The primary functions of GSHinclude conserving the reduced state of the sulfhydryl groups ofintracellular proteins to maintain the correct conformation of theproteins (16, 49). LAT contains four cysteine residues, of whichtwo are within the N-terminala-helix, which mediates the mem-brane localization of LAT, one is just proximal of thea-helix, andthe fourth is present in the cytoplasmic part of the protein (22). Itseems probable that in SF T lymphocytes from RA patients thatcontain decreased levels of the main antioxidant GSH, the sulfhy-dryl groups of LAT become oxidized and form disulfide bondseither intramolecular or intermolecular with small cytoplasmicsulfhydryl-containing proteins and hence interfere with the place-ment of the N-terminala-helix within the cellular membrane. Sec-ond, a change in the cellular architecture could result in the mem-brane displacement of LAT. Oxidative stress has been reported toinduce severe changes in the cytoskeletal structure (50–52). Sev-eral proteins involved in the TCR-mediated signaling pathwaysincluding the TCRz-chain, the guanine nucleotide exchange factorVav, and SLP-76 are or upon TCR ligation become physicallylinked with components of the cytoskeleton (15, 53–55). It seemspossible that LAT is also connected either directly or indirectlywith the cytoskeleton, and any changes in the cytoskeletal structuredue to oxidative stress therefore might somehow result in the dis-placement of LAT from the cellular membrane.

An important point of consideration is the role of functionallydefective T lymphocytes in chronic inflammation. It is generallyaccepted that T lymphocytes contribute to the perpetuation of thedisease in RA. Because SF T lymphocytes exert impaired respon-siveness to TCR-mediated activation, it is clearly important toknow what mechanism might be involved. In this respect it isworth note that despite suppressed cytokine expression and pro-liferation, SF T lymphocytes display an activated phenotype (1–3),which may be important for the activation of macrophages andfibroblasts via direct Ag-independent cell-cell contact. Chronicallyactivated macrophages and fibroblasts produce matrix metallopro-teinases and proinflammatory cytokines like TNF-a and IL-1,which may aggravate the disease (8, 10, 11). The induction of

2177The Journal of Immunology

by guest on May 25, 2017

http://ww

w.jim

munol.org/

Dow

nloaded from

oxidative stress through the stimulation of neutrophils by TNF-athus creates a vicious circle, responsible for the chronicity of thedisease. Evidence exists that the persistence of T lymphocytes inthe chronically inflamed joints is a consequence of the active in-hibition of T cell apoptosis by environmental factors (56). Re-cently, it has been shown that the expression of Fas ligand (FasL)is deficient in SF T lymphocytes from RA patients that expresshigh levels of Fas (57). Because the intact TCR-signaling path-ways, and especially the ZAP-70-mediated Ca21 influx, are re-quired to control the expression of FasL and the homeostasis of Tlymphocytes through Fas/FasL-mediated apoptosis (58–60), themembrane displacement of LAT is likely to underlie this defect.Therefore, we propose that the treatment of RA patients with an an-tioxidant like NAC might contribute to the increased apoptosis of theSF T lymphocytes and eventually ameliorate the disease in RA.

In summary, our present findings indicate that the hyporespon-sive state of the T lymphocytes in the SF of the inflamed joints ofpatients with RA results from the displacement of LAT from theplasma membrane due to the reduced intracellular GSH levels. Thelocalization of LAT in the cytoplasm of the SF T lymphocytesinterferes with its phosphorylation and subsequent association withother signaling proteins, and hence abrogates the TCR-mediatedsignaling pathways.

AcknowledgmentsWe are grateful to the Department of Cell Biology and Genetics at theErasmus University (Rotterdam, The Netherlands) for the use of their im-munofluorescence microscope/video equipment.

References1. Afeltra, A., M. Galeazzi, G. D. Sebastiani, G. M. Ferri, D. Caccavo,

M. A. Addessi, R. Marcolongo, and L. Bonomo. 1997. Coexpression of CD69and HLADR activation markers on synovial fluid T lymphocytes of patientsaffected by rheumatoid arthritis: a three-colour cytometric analysis.Int. J. Exp.Pathol. 78:331.

2. Takahashi, H., K. Soderstrom, E. Nilsson, R. Kiessling, and M. Patarroyo. 1992.Integrins and other adhesion molecules on lymphocytes from synovial fluid andperipheral blood of rheumatoid arthritis patients.Eur. J. Immunol. 22:2879.

3. Hemler, M. E., D. Glass, J. S. Coblyn, and J. G. Jacobson. 1986. Very lateactivation antigens on rheumatoid synovial T lymphocytes: association withstages of T cell activation.J. Clin. Invest. 78:696.

4. al-Janadi, N., A. al-Dalaan, S. al-Balla, and S. Raziuddin. 1996. CD41 T cellinducible immunoregulatory cytokine response in rheumatoid arthritis.J. Rheu-matol. 23:809.

5. Firestein, G. S., W. D. Xu, D. Townsend, J. Broide, J. Alvaro-Gracia,A. Glasebrook, and N. J. Zvaifler. 1988. Cytokines in chronic inflammatory ar-thritis. I. Failure to detect T cell lymphokines (IL-2 and IL-3) and presence ofmacrophage colony-stimulating factor (CSF-1) and a novel mast cell growth fac-tor in rheumatoid synovitis.J. Exp. Med. 168:1573.

6. Keystone, E. C., L. Poplonski, R. G. Miller, R. Gorczynski, D. Gladman, andK. Snow. 1988. Reactivity of T cells from patients with rheumatoid arthritis toanti-CD3 antibody.Clin. Immunol. Immunopath. 48:325.

7. Firestein, G. S., and N. J. Zvaifler. 1987. Peripheral blood and synovial fluidmonocyte activation in inflammatory arthritis. II. Low levels of synovial fluid andsynovial tissue interferon suggest thatg-interferon is not the primary macrophageactivating factor.Arthritis Rheum. 30:864.

8. McInnes, I. B., B. P. Leung, R. D. Sturrock, M. Field, and F. Y. Liew. 1997.Interleukin-15 mediates T cell-dependent regulation of tumor necrosis factor-aproduction in rheumatoid arthritis.Nat. Med. 3:189.

9. Panayi, G. S. 1997. T-cell-dependent pathways in rheumatoid arthritis.Curr.Opin. Rheumatol. 9:236.

10. Sebbag, M., S. L. Parry, F. M. Brennan, and M. Feldmann. 1997. Cytokinestimulation of T lymphocytes regulates their capacity to induce monocyte pro-duction of tumor necrosis factor-a, but not interleukin-10: possible relevance topahtophysiology of rheumatoid arthritis.Eur. J. Immunol. 27:624.

11. Migita, K., K. Eguchi, Y. Kawabe, Y. Ichinose, T. Tsukada, T. Aoyagi,H. Nakamura, and S. Nagataki. 1996. TNF-a-mediated expression of membrane-type matrix metalloproteinase in rheumatoid synovial fibroblasts.Immunology89:553.

12. Salojin, K. V., J. Zhang, J. Madrenas, and T. L. Delovitch. 1998. T-cell anergyand altered T-cell receptor signaling: effects on autoimmune disease.Immunol.Today 19:468.

13. Maurice, M. M., H. Nakamura, E. A. M. Van der Voort, A. I. Van Vliet,F. J. T. Staal, P.-P. Tak, F. C. Breedveld, and C. L. Verweij. 1997. Evidence forthe role of an altered redox state in hyporesponsiveness of synovial T cells inrheumatoid arthritis.J. Immunol. 158:1458.

14. Yoshida, S., T. Katoh, T. Tetsuka, K. Uno, N. Matsui, T. Okamoto. 1999. In-volvement of thioredoxin in rheumatoid arthritis: its costimulatory roles in theTNF-a-induced production of IL-6 and IL-8 from cultured synovial fibroblasts.J. Immunol. 163:351.

15. Fischer, K.-D., Y.-Y. Kong, N. Nishina, K. Tedford, L. E. M. Marengere,I. Kozieradzki, T. Sasaki, M. Starr, G. Chan, S. Gardener, et al. 1998. Vav is aregulator of cytoskeletal reorganization mediated by the T-cell receptor.Curr.Biol. 8:554.

16. Ziegler, D. M. 1985. Role of reversible oxidation-reduction of enzyme thiols-disulfides in metabolic regulation.Annu. Rev. Biochem. 54:305.

17. Staal, F. J. T., M. T. Anderson, G. E. J. Staal, L. A. Herzenberg, C. Gitler, andL. A. Herzenberg. 1994. Redox regulation of signal transduction: tyrosine phos-phorylation and calcium influx.Proc. Natl. Acad. Sci. USA 91:3619.

18. Wange, R. L., and L. E. Samelson. 1996. Complex complexes: signaling at theTCR. Immunity 5:197.

19. Van Leeuwen, J. E. M., and L. E. Samelson. 1999. T cell antigen-receptor signaltransduction.Curr. Opin. Immunol. 11:242.

20. Graef, I. A., L. J. Holsinger, S. Diver, S. L. Schreiber, and G. R. Crabtree. 1997.Proximity and orientation underlie signaling by the non-receptor tyrosine kinaseZAP70.EMBO J. 16:5618.

21. Neumeister, E. N., Y. Zhu, S. Richard, C. Terhorst, A. C. Chan, and A. S. Shaw.1995. Binding of ZAP-70 to phosphorylated T-cell receptorz andh enhances itsautophosphorylation and generates specific binding sites for SH2 domain-con-taining proteins.Mol. Cell. Biol. 15:3171.

22. Zhang, W., J. Sloan-Lancaster, J. Kitchen, R. P. Trible, and L. E. Samelson. 1998.LAT: the ZAP-70 tyrosine kinase substrate that links T cell receptor to cellularactivation.Cell 92:83.

23. Raab, M., A. J. Da Silva, P. R. Findell, and C. E. Rudd. 1997. Regulation ofVav-SLP-76 binding by ZAP-70 and its relevance to TCRz/CD3 induction ofinterleukin-2.Immunity 6:155.

24. Wardenburg, J. B., C. Fu, J. K. Jackman, H. Flotow, S. E. Wilkinson,D. H. Williams, R. Johnson, G. Kong, A. C. Chan, and P. R. Findell. 1996.Phosphorylation of SLP-76 by the ZAP-70 protein-tyrosine kinase is required forT-cell receptor function.J. Biol. Chem. 271:19641.

25. Motto, D. G., M. A. Musci, S. E. Ross, and G. A. Koretzky. 1996. Tyrosinephosphorylation of Grb2-associated proteins correlates with phospholipase Cg1activation in T cells.Mol. Cell. Biol. 16:2823.

26. Fukazawa, T., K. A. Reedquist, G. Panchamoorthy, S. Soltoff, T. Trub, B. Druker,L. Cantley, S. E. Shoelson, and H. Band. 1995. T cell activation-dependent as-sociation between the p85 subunit of the phosphatidylinositol 3-kinase and Grb2/phospholipase C-g1-binding phosphotyrosyl protein pp36/38.J. Biol. Chem. 270:20177.

27. Buday, L., S. E. Egan, P. Rodriguez Viciana, D. A. Cantrell, and J. Downward.1994. A complex of Grb2 adaptor protein, Sos exchange factor and a 36-kDamembrane-bound tyrosine phosphoprotein is implicated inras activation in Tcells.J. Biol. Chem. 269:9019.

28. Weber, J. R., S. Ørstavik, K. M. Torgersen, N. C. Danbolt, S. F. Berg, J. C. Ryan,K. Tasken, J. B. Imboden, and J. T. Vaage. 1998. Molecular cloning of the cDNAencoding pp36, a tyrosine-phosphorylated adaptor protein selectively expressedby T cells and natural killer cells.J. Exp. Med. 187:1154.

29. Trub, T., J. D. Frantz, M. Miyazaki, H. Band, and S. E. Shoelson. 1997. The roleof a lymphoid-restricted, Grb2-like SH3-SH2-SH3 protein in T cell receptor sig-naling.J. Biol. Chem. 272:894.

30. Matsuda, M., A.-K. Ulfgren, R. Lenkei, M. Petersson, A. C. Ochoa, S. Lindblad,P. Andersson, L. Klareskog, and R. Kiessling. 1998. Decreased expression ofsignal-transducing CD3z chains in T cells from the joints and peripheral bloodof rheumatoid arthritis patients.Scand. J. Immunol. 47:254.

31. Maurice, M. M., A. C. Lankester, A. C. Bezemer, M. F. Geertsma, P.-P. Tak,F. C. Breedveld, R. A. W. Van Lier, and C. L. Verweij. 1997. Defective TCR-mediated signaling in synovial T cells in rheumatoid arthritis.J. Immunol. 159:2973.

32. Carruthers, D. M., W. G. Naylor, M. E. Allen, G. D. Kitas, P. A. Bacon, andS. P. Young. 1996. Characterization of altered signalling in T lymphocytes frompatients with rheumatoid arthritis (RA).Clin. Exp. Immunol. 105:291.

33. Allen, M. E., S. P. Young, R. H. Michell, and P. A. Bacon. 1995. Altered Tlymphocyte signaling in rheumatoid arthritis.Eur. J. Immunol. 25:1547.

34. Mirza, N. M., V. Relias, E. J. Yunis, W. N. Pachas, and J. D. Dasgupta. 1993.Defective signal transduction via T-cell receptor-CD3 structure in T cells fromrheumatoid arthritis patients.Hum. Immunol. 36:91.

35. Peterson, E. J., J. L. Clements, N. Fang, and G. A. Koretzky. 1998. Adaptorproteins in lymphocyte antigen-receptor signaling.Curr. Opin. Immunol. 10:337.

36. Stoica, B., K. E. DeBell, L. Graham, B. L. Rellahan, M. A. Alava, J. Laborda, andE. Bonvini. 1998. The amino-terminal Src homology 2 domain of phospholipaseCg1 is essential for TCR-induced tyrosine phosphorylation of phospholipaseCg1. J. Immunol. 160:1059.

37. Trimble, L. A., and J. Lieberman. 1998. Circulating CD8 T lymphocytes in hu-man immunodeficiency virus-infected individuals have impaired function anddownmodulate CD3z, the signaling chain of the T-cell receptor complex.Blood91:585.

38. Stefanova, I., M. W. Saville, C. Peters, F. R. Cleghorn, D. Schwartz,D. J. Venzon, K. J. Weinhold, N. Jack, C. Bartholomew, W. A. Blattner, et al.1996. HIV infection-induced posttranslation modification of T cell signaling mol-ecules associated with disease progression.J. Clin. Invest. 98:1290.

39. Matsuda, M., M. Petersson, R. Lenkei, J. L. Taupin, I. Magnusson, H. Mellstedt,P. Anderson, and R. Kiessling. 1995. Alterations in the signal-transducing mol-ecules of T cells and NK cells in colorectal tumor-infiltrating, gut mucosal and

2178 MEMBRANE DISPLACEMENT OF LAT IN RA

by guest on May 25, 2017

http://ww

w.jim

munol.org/

Dow

nloaded from

peripheral lymphocytes: correlations with the stage of the disease.Int. J. Cancer61:765.

40. Salojin, K., J. Zhang, M. Cameron, B. Gill, G. Arreaza, A. Ochi, andT. L. Delovitch. 1997. Impaired plasma membrane targeting of Grb2-murine Sonof sevenless (mSos) complex and differential activation of the Fyn-T cell receptor(TCR)-z-Cbl pathway mediate T cell hyporesponsiveness in autoimmune nono-bese diabetic mice.J. Exp. Med. 186:887.

41. Rapoport, M. J., A. H. Lazarus, A. Jaramillo, E. Speck, and T. L. Delovitch.1993. Thymic T cell anergy in autoimmune nonobese diabetic mice is mediatedby deficient T cell receptor regulation of the pathway of p21Rasactivation.J. Exp.Med. 177:1221.

42. Lahdenpohja, N., K. Savinainen, and M. Hurme. 1998. Pre-exposure to oxidativestress decreases the nuclear factor-kB-dependent transcription in T lymphocytes.J. Immunol. 160:1354.

43. Otsuji, M., Y. Kimura, T. Aoe, Y. Okamoto, and T. Saito. 1996. Oxidative stressby tumor-derived macrophages suppresses the expression of CD3z chain ofT-cell receptor complex and antigen-specific T-cell responses.Proc. Natl. Acad.Sci. USA 93:13119.

44. Kono, K., F. Salazar-Onfray, M. Petersson, J. Hansson, G. Masucci,K. Wasserman, T. Nakazawa, P. Anderson, and R. Kiessling. 1996. Hydrogenperoxide secreted by tumor-derived macrophages down-modulates signal-trans-ducing z molecules and inhibits tumor-specific T cell- and natural killer cell-mediated cytotoxicity.Eur. J. Immunol. 26:1308.

45. Staal, F. J. T., M. Roederer, D. M. Israelski, J. Bubp, L. A. Mole, D. McShane,S. C. Deresinski, W. Ross, H. Sussman, P. A. Raju, et al. 1992. Intracellularglutathione levels in T cell subsets decrease in HIV-infected individuals.AIDSRes. Hum. Retroviruses 8:305.

46. Weber, G. F., N. M. Mirza, E. J. Yunis, D. Dubey, and H. Cantor. 1997. Local-ization and treatment of an oxidative-sensitive defect within the TCR-coupledsignalling pathway that is associated with normal and premature immunologicaging.Growth Dev. Aging 61:191.

47. Grossmann, A., P. S Rabinovitch, T. J. Kavanagh, J. C. Jinneman, L. K. Gilliland,J. A. Ledbetter, and S. B. Kanner. 1995. Activation of murine T-cells via phos-pholipase-Cg1-associated protein tyrosine phosphorylation is reduced with aging.J. Gerontol. A. Biol. Sci. Med. Sci. 50:B205.

48. Kanner, S. B., T. J. Kavanagh, A. Grossmann, S.-L. Hu, J. B. Bolen,P. S. Rabinovitch, and J. A. Ledbetter. 1992. Sulfhydryl oxidation down-regu-lates T-cell signaling and inhibits tyrosine phosphorylation of phospholipaseCg1. Proc. Natl. Acad. Sci. USA 89:300.

49. Fischer, E. H., H. Charbonneau, and N. K. Tonks. 1991. Protein tyrosine phos-phatases: a diverse family of intracellular and transmembrane enzymes.Science253:401.

50. Milzani, A., I. DalleDonne, and R. Colombo. 1997. Prolonged oxidative stress onactin.Arch. Biochem. Biophys. 339:267.

51. Malorni, W., G. Rainaldi, R. Rivabene, M. T. Santini, S. W. Peterson, U. Testa,and G. Donelli. 1994. Cytoskeletal oxidative changes lead to alterations of spe-cific cell surface receptors.Eur. J. Histochem. 38:91.

52. Bellomo, G., F. Mirabelli, M. Vairetti, F. Iosi, and W. Malorni. 1990. Cytoskel-eton as a target in menadione-induced oxidative stress in cultured mammaliancells. I. Biochemical and immunocytochemical features. J. Cell Physiol. 143:118.

53. Holsinger, L. J., I. A. Graef, W. Swat, T. Chi, D. M. Bautista, L. Davidson,R. S. Lewis, F. W. Alt, and G. R. Crabtree. 1998. Defects in actin-cap formationin Vav-deficient mice implicate an actin requirement for lymphocyte signal trans-duction.Curr. Biol. 8:563.

54. Huby, R. D., A. Weiss, and S. C. Ley. 1998. Nocodazole inhibits signal trans-duction by the T cell antigen receptor.J. Biol. Chem. 273:12024.

55. Rozdzial, M. M., B. Malissen, and T. H. Finkel. 1997. Tyrosine-phosphorylatedT cell receptorz chain associates with the actin cytoskeleton upon activation ofmature T lymphocytes.Immunity 3:623.

56. Salmon, M., D. Scheel-Toellner, A. P. Huissoon, D. Piling, N. Shamsadeen,H. Hyde, A. D. D’Angeac, P. A. Bacon, P. Emery, and A. N. Akbar. 1997.Inhibition of T cell apoptosis in the rheumatoid arthritis.J. Clin. Invest. 99:439.

57. Cantwell, M. J., T. Hua, N. J. Zvaifler, and T. J. Kipps. 1997. Deficient Fas ligandexpression by synovial lymphocytes from patients with rheumatoid arthritis.Ar-thritis Rheum. 40:1644.

58. Esser, M. T., D. M. Haverstick, C. L. Fuller, C. A. Gullo, and V. Lam Braciale.1998. Ca21 signaling modulates cytolytic T lymphocyte effector functions.J. Exp. Med. 187:1057.

59. Eischen, C. M., B. L. Williams, W. G. Zhang, L. E. Samelson, D. H. Lynch,R. T. Abraham, and P. J. Leibson. 1997. Zap-70 tyrosine kinase is required for theupregulation Fas ligand in activation-induced T cell apoptosis.J. Immunol. 159:1135.

60. Ju, S. T., D. J. Panka, H. Cui, R. Ettinger, M. el-Khatib, D. H. Sherr,B. Z. Stanger, and A. Marshak-Rothstein. 1995. Fas(CD95)/FasL interactionsrequired for programmed cell death after T-cell activation.Nature 373:444.

2179The Journal of Immunology

by guest on May 25, 2017

http://ww

w.jim

munol.org/

Dow

nloaded from