of May 25, 2018.This information is current as

Responses in Systemic Lupus ErythematosusDynamics Contribute to Abnormal T Cell Alterations in Lipid Raft Composition and

and George C. TsokosWarke, Carolyn U. Fisher, Jeanne Mitchell, Nancy Delaney Sandeep Krishnan, Madhusoodana P. Nambiar, Vishal G.

http://www.jimmunol.org/content/172/12/7821doi: 10.4049/jimmunol.172.12.7821

2004; 172:7821-7831; ;J Immunol 

Referenceshttp://www.jimmunol.org/content/172/12/7821.full#ref-list-1

, 31 of which you can access for free at: cites 68 articlesThis article

        average*  

4 weeks from acceptance to publicationFast Publication! •    

Every submission reviewed by practicing scientistsNo Triage! •    

from submission to initial decisionRapid Reviews! 30 days* •    

Submit online. ?The JIWhy

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 © 2004 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, 2018

http://ww

w.jim

munol.org/

Dow

nloaded from

by guest on May 25, 2018

http://ww

w.jim

munol.org/

Dow

nloaded from

Alterations in Lipid Raft Composition and DynamicsContribute to Abnormal T Cell Responses in Systemic LupusErythematosus1,2

Sandeep Krishnan,* Madhusoodana P. Nambiar,*† Vishal G. Warke,* Carolyn U. Fisher,*Jeanne Mitchell,‡ Nancy Delaney,‡ and George C. Tsokos3*†‡

In response to appropriate stimulation, T lymphocytes from systemic lupus erythematosus (SLE) patients exhibit increased andfaster intracellular tyrosine phosphorylation and free calcium responses. We have explored whether the composition and dynamicsof lipid rafts are responsible for the abnormal T cell responses in SLE. SLE T cells generate and possess higher amounts ofganglioside-containing lipid rafts and, unlike normal T cells, SLE T cell lipid rafts include FcR� and activated Syk kinase. IgManti-CD3 Ab-mediated capping of TCR complexes occurs more rapidly in SLE T cells and concomitant with dramatic accelerationof actin polymerization kinetics. The significance of these findings is evident from the observation that cross-linking of lipid raftsevokes earlier and higher calcium responses in SLE T cells. Thus, we propose that alterations in the lipid raft signaling machineryrepresent an important mechanism that is responsible for the heightened and accelerated T cell responses in SLE.The Journalof Immunology, 2004, 172: 7821–7831.

Systemic lupus erythematosus (SLE)4 is a complex autoim-mune disease of unknown etiology. Recent studies havereported several functional abnormalities of T cells in me-

diating pathogenesis of SLE. However, the biochemical processesunderlying these abnormal T cell responses are only beginning tobe understood (1, 2). Previous studies have demonstrated abnor-malities at multiple levels in the TCR signaling cascade in SLE Tcells. SLE T cells display marked reduction in the expression ofthe CD3� chain and the kinase LCK (3–6), increased activity ofmitogen-activated protein kinase (7), and a concomitant up-regu-lation of CD3� homologue FcR�. FcR� associates with TCR andutilizes Syk kinase instead of ZAP-70 to mediate downstream sig-naling (8). This “rewiring” of T cells in SLE assumes significanceconsidering that in its context, biochemical signaling events suchas intracellular protein tyrosine phosphorylation and calcium re-sponses (intracellular Ca2 concentration ([Ca2�]i)) become aug-mented (3). While attempting to identify mechanisms that mediateheightened T cell responses in SLE, we found that transfection ofthe FcR� chain into normal T cells amplifies TCR signaling events

such as protein tyrosine phosphorylation and calcium signaling ina Syk kinase-dependent manner and can thus mimic SLE T cellsignaling abnormalities (9). Conversely, we detected that height-ened TCR signaling responses observed in SLE T cells can bedampened by replenishing CD3� into SLE T cells (10). These ob-servations suggest the presence of mechanisms responsible for theaberrant amplification of signals in SLE T cells by the “rewiredTCR” that involve close interactions between disparate sets of sig-naling proteins (11).

Lipid rafts are highly ordered cholesterol and ganglioside-richplatforms that can facilitate and coordinate close interactions be-tween critical signaling molecules to amplify downstream signal-ing (12, 13). In normal T cells, ligation of the TCR induces rapidlipid raft clustering that leads to concentration of signaling proteinsat the area of contact between APCs and T cells known as theimmunological synapse (14–16). Efficient formation of the immu-nological synapse is critical to amplification of signals downstreamof the TCR and subsequent T cell activation (15). Conversely, lossof integrity of lipid raft processes have been shown to play a majorrole in the pathogenesis of several diseases such as infections,allergies, and neoplasms (17).

We hypothesized a prominent role for lipid rafts in augmentingTCR-mediated responses in SLE T cells for several reasons. First,both CD3� and FcR� can signal through lipid rafts in differentcellular contexts (18, 19). Second, increased tyrosine phosphory-lation of cytoplasmic proteins, such as that observed in SLE Tcells, has been shown to induce their recruitment to lipid rafts (20).Third, we observed that in SLE T cells, CD3� is localized withindiscrete surface membrane clusters that are also enriched for linkerfor activation of T cells (LAT), an important resident of lipid rafts,suggesting that the bulk of CD3� resides within lipid rafts in SLE(21). Fourth, we observed that in SLE T cells, the bulk of theresidual CD3� is associated with the detergent-insoluble fractionand could be solubilized by methyl-�-cyclodextrin (MbCD), alipid raft disrupting agent, thus suggesting an increased localiza-tion of CD3� within lipid rafts in these cells (21).

Several factors can influence the strength of signals controlledby lipid rafts such as, lipid raft pool size, membrane distribution

*Department of Cellular Injury, Walter Reed Army Institute of Research, SilverSpring, MD 20910;†Department of Medicine, Uniformed Services University of theHealth Sciences, Bethesda, MD 20814; and‡Walter Reed Army Medical Center,Washington, DC 20307

Received for publication February 5, 2004. Accepted for publication April 13, 2004.

The costs of publication of this article were defrayed in part by the payment of pagecharges. This article must therefore be hereby markedadvertisement in accordancewith 18 U.S.C. Section 1734 solely to indicate this fact.1 This work was supported by research grants from the National Institutes of Health(RO1-AI42269 and RO1-AI49954) awarded to G.C.T.2 The opinions expressed herein are the private ones of the authors and they do notrepresent those of the Department of Defense or the Department of the Army.3 Address correspondence and reprint requests to Dr. George C. Tsokos, Departmentof Cellular Injury, Walter Reed Army Institute of Research, Silver Spring, MD 20910-7500. E-mail address: gtsokos@usuhs.mil4 Abbreviations used in this paper: SLE, systemic lupus erythematosus; ZAP-70,�-as-sociated protein 70; [Ca2�]i, intracellular Ca2� concentration; LAT, linker for acti-vation of T cells; MbCD, methyl-�-cyclodextrin; SLEDAI, SLE disease activity in-dex; F-actin, filamentous actin; PVDF, polyvinylidene difluoride; CT-B, cholera toxinB; PLC�1, phospholipase C�1; MFI, mean fluorescence intensity; TRITC, tetram-ethylrhodamine isothiocyanate; p-Syk, phosphorylated Syk.

The Journal of Immunology

Copyright © 2004 by The American Association of Immunologists, Inc. 0022-1767/04/$02.00

by guest on May 25, 2018

http://ww

w.jim

munol.org/

Dow

nloaded from

pattern, protein content, and the kinetics of cytoskeletal rearrange-ments following T cell stimulation. Thus, we assessed the contri-bution of lipid rafts to abnormal T cell responses in SLE by com-paring these parameters between normal and SLE T cells. Wedemonstrate that the lipid raft composition and dynamics are al-tered in SLE T cells and that these changes may contribute toamplification of TCR-mediated signals in SLE.

Materials and MethodsPatient population

All patients studied fulfilled the American College of Rheumatology clas-sification criteria for SLE (22). Our study included 20 inactive SLE pa-tients, 32 active patients with SLE disease activity index (SLEDAI) rang-ing from 1 to 16, 5 disease controls with Sjogren’s syndrome andrheumatoid arthritis, and 48 healthy volunteers (Table I). All SLE patients,disease controls, and healthy volunteers were from the age group of 19–71years. Patients who were being treated with prednisone were asked not totake this medication for at least 24 h before blood was drawn. Because ofpaucity of cell numbers obtained from the disease group, many of whomwere leukopenic, different sets of experiments were performed with dif-ferent patient groups but care was taken to include patients of SLEDAIscores ranging from low (0–5 range) to high (6–16 range) in each group.Within each group, patient samples were matched with normal samples ofsimilar ages and gender. Monocyte-depleted lymphocytes from healthycontrols were kindly provided by Dr. D. Hoover (Walter Reed Army In-stitute of Research (WRAIR), Silver Spring, MD). The study protocol wasapproved by the Health Use Committees of WRAIR and Walter ReedArmy Medical Center. Written informed consent was obtained from allparticipating patients and volunteers.

T lymphocyte isolation

PBMC were obtained by density gradient centrifugation over Ficoll and Tcells were isolated by positive depletion of non-T cells by magnetic sep-aration (Miltenyi Biotec, Auburn CA) as described previously (23) or usingRosette Sep (StemCell Technology, Vancouver, British Columbia, Can-ada) following the manufacturer’s instructions. In all cases, the percentageof T cells in the isolated subpopulation was �98% as determined by anti-CD3� staining and flow cytometry (FACS) using FACSCalibur (BD Bio-sciences, Mountain View, CA).

Cell stimulation, CD3� capping, and confocal microscopy

Cells (0.5 � 106) were adhered on polylysine-coated glass slides for 1 h atroom temperature. For capping experiments, cells were stimulated with1/100 IgM anti-CD3 Ab for various time points at 37oC and immediatelyfixed with 4% paraformaldehyde solution. For intracellular staining, cells

were permeabilized with a buffer containing 0.05% (w/v) saponin in RPMI1640 medium. Cells were stained with matching pairs of primary and flu-orochrome-labeled secondary Abs (Jackson Immunoresearch Laboratories,West Grove, PA) for 1 h and 30 min, respectively, at room temperature.Cells were washed, air dried, and mounted using Gel/Mount (Biomeda,Foster City, CA) and coverslips applied. Samples were analyzed with alaser scanning confocal fluorescence microscope (1X70; Olympus, LakeSuccess, NY) with Lasersharp-2000 software (Bio-Rad, Richmond, CA).For some images where comparison was made between cytoplasmic andmembrane distribution of proteins, the Z-stack feature of the software wasused to obtain a library of images of various sections of cells, and sectionsshowing both cytoplasmic and membrane portions clearly were selected.All images were processed to obtain the final images using Adobe Photo-shop 7.0 Adobe Systems, Mountain View, CA).

GM1 staining and flow cytometric analysis

T cells, resting or activated with anti-CD3 (2 �g/ml, clone OKT3; OrthoBiotech, Raritan, NJ) and anti-CD28 (1 �g/ml; BD PharMingen, San Di-ego, CA) immobilized on plastic for 72 h as described before (24), werefixed with 0.75% paraformaldehyde and permeabilized with a buffer con-taining 0.05% (w/v) saponin in PBS and stained with cholera toxin B(CT-B)-FITC for 1 h. Cells were washed twice and fluorescence was de-tected by FACS.

Actin polymerization assay

The procedure was done as described before (25), with minor changes asfollows: cells were washed twice with PBS and stimulated with IgM anti-CD3 (1/100) as required and treated with 3.7% paraformaldehyde for 8 minto stop the reaction. Cells were washed twice with PBS and permeabilizedand stained with staining buffer containing 0.1% Triton X-100 and 0.2 �Mphalloidin-FITC (Sigma-Aldrich, St. Louis, MO) for 40 min. Cells werewashed, resuspended in PBS and F-actin content was analyzed by FACS.In each experiment, normal and SLE T cells were activated and analyzedtogether.

Preparation of detergent-resistant membrane fractions

We modified the protocols described before (20, 26) as follows: T cells(25 � 106) from healthy volunteers and lupus patients were used, eitherresting or following activation with IgM anti-CD3 Ab for 2 min. Cells werewashed twice with PBS and resuspended in 1 ml of ice-cold lysis buffer(1% Brij 58, 25 mM MES, 150 mM NaCl, supplemented with 1 mMsodium orthovanadate, 2 mM EDTA, 1 mM PMSF, and 1 �g/ml aprotinin).Cells were homogenized with 12 strokes of Dounce homogenizer and in-cubated on ice for 30 min, mixed with an equal volume of 85% w/v sucrose(prepared in 25 mM MES, 150 mM NaCl, pH 6.5) and placed at the bottomof a 12-ml ultracentrifuge tube. A step gradient consisting of 6 ml of 35%sucrose and 4 ml of 5% sucrose was layered on top of it and the tubes werecentrifuged for 18–20 h at 200,000 � g (39,000 rpm) at 4°C in a SW41rotor (Beckman Instruments, Fullerton, CA). Twelve 1-ml fractions werecollected starting from the top. Fractions 4 and 5 were indicative of raftsand were pooled for each experiment while fraction 12 was used as thecytosolic fraction.

Immunoblotting

Proteins from raft and cytosolic fractions were concentrated using a stan-dard TCA precipitation protocol and then suspended in denaturing samplebuffer (27). Following resolution on a 4–12% bis-Tris NuPage (Invitrogen,San Diego, CA), proteins were transferred onto polyvinylidene difluoride(PVDF) membranes and then incubated with specific Abs. The followingAbs were used: anti-FcR� (Upstate Biotechnology, Charlottesville, VA),and anti-Lck (2102), anti-CD3� (clone 6B10.2), anti-phosphorylated Syk(p-Syk; clone 4D10), anti-LAT (FL-233), and anti-actin (I-19) from SantaCruz Biotechnology (Santa Cruz, CA). HRP-coupled Abs (Santa Cruz Bio-technology) were used as secondary Abs and detection was performed withECL (Amersham Pharmacia Biotech, Piscataway, NJ). For Vav1 associa-tion experiments, collection of actin cytoskeletal fractions has been de-scribed before (28). Briefly, the detergent-insoluble pellets obtained bylysis of 5 � 106 cells as described above were treated with cytochalasin B(0.1 mg/ml; Sigma-Aldrich) for 1 h at 37°C, samples were spun, and su-pernatants were collected. Samples were boiled for 2 min and reduced with1 M DTT and immunoblotting experiments were performed with anti-Vav(Upstate Biotechnology) or anti-phospho-Vav (Santa Cruz Biotechnology),as discussed above.

Table I. Details of the patients involved in this study

Patients/Features Parameters n

SLESLEDAI 0 20

1–5 136–10 1111–16 8

Sex Females 51Males 1

Race Caucasian 9African American 27

Hispanic 4Asian 2

Unknown 10Medications None 19

Plaq. 21Pred. 4

Cytoxan 1Plaq. � Pred. 4

Plaq. � Pred. � Metho. 3Disease controls

Sjogren’s syndrome Females 3Pheumatoid arthritis Females 2

a Plaq., plaquenil; Pred., prednisone; Metho., methotrexate.

7822 LIPID RAFT-MEDIATED SIGNALING IN SLE

by guest on May 25, 2018

http://ww

w.jim

munol.org/

Dow

nloaded from

Cross-linking of lipid rafts and measurement of Ca2� influx

Five million cells were washed with RPMI 1640 and incubated with 1�g/ml Indo-AM (Molecular Probes, Eugene, OR) for 30 min at 37°C,washed with RPMI 1640, and kept on ice. Cells were analyzed using anEpics Altra (Coulter, Hialeah, FL) flow cytometer equipped with a high-power dual wavelength (365- and 488-nm argon laser). CT-B subunitpatching of rafts was adopted as described before (29), with changes asfollows: cells were stained with CT-B (10 �g/ml; Calbiochem, San Diego,CA) for 10 min followed by anti-CT-B (1/250; Calbiochem) for 4 min onice and then warmed up to 37°C in 1 min. For some experiments, cells weretreated with varying doses of MbCD for 30 min at 37°C as reported before(20, 30). Samples were run and at 40 s, either OKT3 (10 �g/ml) or theisotype control mIgG2a was added followed by goat anti-mouse cross-linker at 1 min and the ratio of the fluorescence, which is directly propor-tional to free cytosolic Ca2�, was recorded for a period of 10 min asdescribed before (3, 10). In some cases, MbCD-treated cells were treatedwith 1 �M thapsigargin (Sigma-Aldrich) and calcium response was ana-lyzed immediately.

Densitometry and statistical analysis

Densitometric analysis of the autoradiograms was performed with the soft-ware program GelPro (Media Cybernetics, Silver Spring, MD). Statisticalanalysis of the data was done by paired t test using the software MINITAB,version 13 (Minitab, State College, PA). A value of p � 0.05 was consid-ered to be significant.

ResultsSLE T cells display robust capacity to generate lipid rafts

To understand the role of lipid rafts in altered signal transductionin SLE T cells, we first asked whether SLE T cells contained largerpools of lipid rafts compared with normal T cells by analyzing thesurface and total (surface and intracellular) GM1 expression inthese cells. GM1 ganglioside has been traditionally used as a lipidraft marker (27, 29, 31–33) and can be detected by flow cytometryby virtue of its ability to bind to CT-B (32, 34, 35). Normal andSLE T cells, either freshly isolated or stimulated with anti-CD3and anti-CD28 for 72 h, were stained with CT-B-FITC and ana-lyzed by flow cytometry. Although similar surface levels of GM1were found in freshly isolated normal and SLE T cells (Fig. 1A),SLE T cells have a significantly larger intracellular pool of GM1than normal T cells (Fig. 1B; normals � 1-fold, SLE � 1.43-fold,p � 0.009). Anti-CD3- and anti-CD28-mediated activation re-sulted in dramatic increases in both surface and total GM1 expres-sion in both cell types. However, unlike freshly isolated T cells,activated SLE T cells generated higher amounts of GM1 gangli-oside than activated normal T cells detected both in the total andmembrane pool (surface (Fig. 1A), normals � 15.96-fold, SLE �22.06-fold, p � 0.036; total (Fig. 1B), normals � 2.84-fold,SLE � 4.16-fold, p � 0.008). These observations were unaffectedby the SLEDAI scores of the patients or the treatment regimen,suggesting that higher levels of GM1 expression occurred inde-pendent of SLE disease activity. To determine whether the higherlevels of GM1 expression following activation were due to in-creased production of GM1 by SLE T cells, we repeated the ac-tivation studies in T cells isolated from normal and SLE patients inthe presence or absence of sphingosine N acetyltransferase inhib-itor, fumonisin B1 (32, 36). The dose of fumonisin B1 used (25�M) was monitored carefully to ensure that it did not affect thefunction of T cells, as suggested by optimum production of IL-2 bynormal T cells (Ref. 32 and data not shown). Activation of cells inthe presence of the vehicle DMSO alone did not limit the produc-tion of GM1 in SLE or normal T cells (Fig. 1C; normals � 3.17-fold, SLE � 4.57-fold, p � 0.04). However, in the presence offumonisin B1, both subsets displayed markedly lower levels oftotal GM1 (normals � 1.32-fold, SLE � 2.91-fold, p � 0.015; Fig.1C), whereas unstimulated normal and SLE T cells cultured in thepresence of fumonisin B1 did not demonstrate appreciable differ-

ences in the levels of GM1 expression (data not shown). Thesefindings were observed at both 48 h (data not shown) and 72 h(Fig. 1C). This experiment confirms that the observed increase inGM1 levels in these cells was due to increased production inducedby T cell activation. We conclude that SLE T cells possess moreextensive basal levels of lipid rafts and greater capacity to generatelipid rafts than normal T cells in response to stimulation.

Qualitative alterations in the composition of lipid rafts in SLE

Next, we asked whether the protein composition of lipid rafts dif-fered between SLE and normal T cells. We had previously dem-onstrated that in SLE T cells, residual CD3� was limited to deter-gent-insoluble fractions, thus suggesting that CD3� might belocalized to lipid rafts of SLE T cells (21). However recent evi-dence has accumulated suggesting that detergent-insoluble frac-tions may not necessarily depict lipid rafts (37, 38). Therefore, toclarify this point and to identify possible alterations in the protein

FIGURE 1. SLE T cells have larger lipid raft compartments and pro-duce lipid rafts more robustly than normal T cells. T cells from normalsubjects and SLE patients were cultured in 24-well plates coated with anti-CD3 and anti-CD28 for 72 h as described in Materials and Methods. Cellswere harvested, fixed with 0.75% paraformaldehyde, and stained directlyor following permeabilization with CT-B-FITC and GM1 staining detectedby FACS. Values on the y-axis represent the ratio of fold increases in meanfluorescence intensity (MFI) of each time point over MFI of normal T cellsat 0 h. Demonstration of (A) surface GM1 expression (n � 5, �, p � 0.05,(B) total GM1 expression (n � 5, �, p � 0.05, and (C) inhibition of GM1production by fumonisin B1 (n � 3, �, p � 0.05.

7823The Journal of Immunology

by guest on May 25, 2018

http://ww

w.jim

munol.org/

Dow

nloaded from

composition of lipid rafts in SLE, we performed immunofluores-cence confocal microscopy to compare the location of critical pro-teins such as LAT, CD3�, the CD3� homologue FcR�, and phos-pholipase C�1 (PLC�1) in normal and SLE T cells, relative to lipidraft marker GM1. Several interesting findings emerged from theseexperiments. We noted that consistent with our previous results(21), both LAT and CD3� were limited to discrete clusters on themembrane of SLE T cells that also colocalized with GM1, whereasthey were uniformly distributed on the normal T cell membrane(Fig. 2 and data not shown). Additionally, we noted that FcR�chain was also present within GM1-enriched clusters on the mem-brane of SLE T cells. These observations were in contrast to thosein normal T cells in which FcR� chain expression was not detected(Fig. 2 and Refs. 8 and 39). In addition, while a majority of normalT cells displayed intracellular stores of PLC�1, in a majority ofSLE T cells, PLC�1 was restricted to the cell membrane (percent-age of cells displaying intracellular stores of PLC�1: normals �67 � 15%, SLE � 28 � 14.9%, p � 0.046). Activation of normaland SLE T cells with anti-CD3 induced a shift in the localizationof PLC�1 to the membrane (percentage of cells displaying intra-cellular stores of PLC�1: normal cells � 9 � 5.03%, SLE � 17 �7.06%, p � 0.16), consistent with the findings reported by othersin Jurkat cells (Ref. 40 and Fig. 2). The above findings suggest thatthe majority of T cells in SLE resemble activated T cells. Ourobservations that the pattern of distribution of these crucial pro-teins did not vary between T cells from normal subjects and indi-viduals with Sjogren’s disease suggest that the observed alterationsin the lipid raft composition are unique to SLE T cells (Fig. 2).

To confirm our observations from confocal microscopy and todetect possible alterations in the distribution of other crucial sig-naling molecules in the lipid raft fractions of SLE T cells, weisolated lipid rafts from normal and SLE T cells by performingsucrose density gradient ultracentrifugation. Lipid raft fractionswere identified as fractions 4 and 5 (Fig. 3A) by the presence ofGM1 expression and enrichment for LAT protein, that is consti-tutively associated with lipid rafts (41), and fraction 12 was con-sidered as the cytosolic fraction (31). Because such experimentsrequired large numbers of T cells and we were limited by the lowyield of cells from SLE patients, the proteins were concentrated bythe standard TCA precipitation method for further analysis (27).Comparison between raft and cytosolic fractions of normal andSLE T cells revealed that CD3� was present in low amounts in

lipid rafts of SLE T cells and in even lower levels in the cytosolicfractions (Fig. 3B). Additionally, we detected FcR� chain exclu-sively in SLE T cells in both raft and cytosolic fractions, while theexpression and distribution of LAT was similar between normaland SLE T cells (Fig. 3C). Our results showing that LAT levelsremain unaltered despite increases in GM1-containing lipid rafts inSLE are consistent with observations made by others in effectorCD8 T cells that bore higher levels of lipid rafts and yet displayedsimilar levels of LAT compared with resting naive and memorycells (33). Next, we compared the levels of LCK within raft andnon-raft fractions of normal and SLE T cells. In normal T cells,substantial amounts of LCK are found within lipid rafts (42, 43).Consistent with previous reports (6), levels of LCK were markedlylower in both lipid raft and cytosolic fractions of SLE T cellscompared with respective fractions of normal T cells (Fig. 3D). Inaddition, we noted that detectable levels of active Syk kinase(phosphorylated Syk) were consistently observed only in the lipidraft fractions of SLE cells (Fig. 3D), while it was undetected in raftand non-raft fractions of normal cells. Activation of SLE T cells byanti-CD3 induced association of activated Syk with lipid rafts ofSLE but not normal T cells. We could not detect ZAP-70 in thelipid rafts of SLE T cells (data not shown). These results comple-ment our previous observations that SLE T cells preferentially uti-lize Syk kinase whereas normal T cells utilize ZAP-70 (8, 44). Thepreferential usage of Syk kinase by SLE T cells also suggests thatSLE T cells may be in a state of activation. Previously, we haveshown that activated effector T cells generated from normal T cellsutilize Syk instead of ZAP-70 (39). We could not detect PLC�1 ineither raft or cytosolic fractions of both cell types by Westernblotting, possibly due to low levels of expression of this protein orbecause of the limitations imposed by a low number of T cells inthe density gradient experiments (data not shown). There was alsono detectable difference in the phosphorylation pattern of LAT inSLE and normal T cells (data not shown). We also observed thatthe above qualitative changes in the lipid raft composition wereunaffected by disease activity (SLEDAI scores).

Increased lateral membrane mobility in SLE T cells

While exploring alterations in the lipid raft composition in SLE Tcells by confocal microscopy, we unexpectedly noted that activa-tion of SLE T cells with the strong CD3 cross-linker, IgM anti-

FIGURE 2. Altered composition of lipid rafts inSLE T cells demonstrated by confocal microscopy.Normal and SLE T cells were adhered to polylysine-coated slides (Sigma-Aldrich) and stained directly orfollowing activation with IgM anti-CD3 for 2 min.Cells were fixed and stained with CT-B-FITC (green),followed by permeabilization and staining with anti-LAT (1/50), anti-FcR� (1/50), or anti-PLC�1 (1/50)followed by corresponding tetramethylrhodamine iso-thiocyanate (TRITC)-conjugated secondary Abs (red),and analyzed by a laser scanning confocal microscopeas described in Materials and Methods. The mergedimages (yellow) indicate colocalization of respectiveproteins with GM1. Twenty-five cells were countedper field and the percentage of positive cells was cal-culated. A representative cell is shown for each sam-ple. Magnification used was �100 oil objective.These results are representative of six experiments.

7824 LIPID RAFT-MEDIATED SIGNALING IN SLE

by guest on May 25, 2018

http://ww

w.jim

munol.org/

Dow

nloaded from

CD3� for 2 min, resulted in the formation of CD3 caps in a ma-jority of cells, whereas T cells from normal and Sjogren’ssyndrome demonstrated only receptor clustering (Fig. 2). We sus-pected an increase in lateral membrane mobility of proteins andlipids in SLE T cells as a possible mechanism behind this phe-nomenon. To address this issue, we compared the kinetics of anti-CD3-induced capping of TCR-CD3 complexes in normal and SLET cells by fluorescent microscopy. Cells were activated with IgManti-CD3� for various time points as depicted in Fig. 4, and thenumber of cells demonstrating capping of CD3� were counted.Similar to our earlier observations in CD4 cells (39) and as ob-served in Fig. 2, in normal T cells, TCR-CD3 complexes wereuniformly distributed on the surface membrane. Cross-linking ofCD3� induced TCR-CD3 clustering within 2 min (8 � 3.27%capped, n � 6, p � 0.014) and, within 10 min, a majority of cellscapped (75 � 11.9% capped, n � 6, p � 0.001; Fig. 4, top panel,boxes 1a, 2a, and 3a, and bottom right panel). By contrast, freshlyisolated SLE T cells consistently exhibited clustering of TCR-CD3complexes on the surface (12 � 3% capped) and demonstratedreceptor capping in a majority of T cells within 2 min of CD3�cross-linking (79 � 5.26% capped, n � 6, p � 0.002; Fig. 4, toppanel, boxes 1d, 2d, and 3d, and bottom right panel), demonstrat-ing that the kinetics of receptor capping or lateral membrane mo-bility is accelerated in SLE T cells. The caps were maintained upto 30 min of CD3 cross-linking without internalization of the re-ceptor complexes in all three cell types (Fig. 4, top panel, column

4). All of these observations occurred independent of SLE diseaseactivity and treatment regimen. Our observations that disease con-trol cells from patients with Sjogren’s syndrome displayed a pat-tern similar to that of normal T cells (cf Fig. 4, top panel, boxes 1a,2a, and 3a and 1g, 2g, and 3g, and bottom right panel) suggest thatfaster lateral membrane mobility of proteins is a feature specific toSLE T cells. Because the CD3 clusters and caps also colocalizedGM1 (Fig. 4, lower left panel), these results suggest that SLE Tcells are characterized by faster kinetics of lipid raft clustering andpolarization.

Alterations in actin dynamics in SLE T cells

Remodeling of cytoskeleton is an essential component of receptorcap formation and integrity of the immunological synapse. Be-cause actin polymerization is a critical event in this process (45–47), we asked whether faster kinetics of receptor capping observedin SLE T cells was associated with accelerated kinetics of actinpolymerization in these cells. Comparison of F-actin polymeriza-tion levels (detected by binding of polymerized F-actin to phalloi-din) in normal and SLE T cells activated for various time pointswith IgM anti-CD3 (Fig. 5A, top panel) demonstrated that while innormal T cells, peak levels of actin polymerization occurred at 1min of activation and in SLE T cells it occurred between 30 s and1 min (Fig. 5A, bottom panels). Another notable difference wasthat, while in normal T cells, high levels of polymerized F-actinwere maintained up to 15 min following activation; in SLE T cells,

FIGURE 3. Altered composition of lipid rafts in SLE T cells demonstrated by sucrose density gradient analysis. Normal and SLE T cells (25 � 106)were lysed with a lysis buffer containing 1% Brij 58 as such or following activation with IgM-anti-CD3 (1/100) for 2 min and raft and cytosolic fractionswere extracted by discontinuous sucrose density gradient ultracentrifugation. Proteins of each fraction were concentrated by standard TCA precipitation andresolved on a 4–12% bis-Tris-NuPage and transferred to PVDF membranes. The blot was probed with anti-LAT (1/1000), anti-CD3� (1/1000), anti-actin(1/1000), anti-FcR� (1/1000), anti-p-Syk (1/1000), or anti-LCK (1/1000). A, Identification of raft fractions as fractions 4 and 5 in SLE T cells by probingblot with CT-B-HRP (Sigma-Aldrich). Various fractions depict portions of the cell with varying density, fraction 12 being the heaviest and depictingcytosolic portion (B) nonreduced gel showing CD3� association with raft and cytosolic fractions of normal (n � 2) and SLE (n � 3), (C) nonreduced gelshowing FcR� and LAT association with raft and cytosolic fractions of normal (n � 6) and SLE (n � 6), (D) reduced gel (1 M DTT) comparing localizationof p-Syk and LCK in normal (n � 6) and SLE (n � 6). Densitometric ratios of various proteins with LAT or �-actin are shown below the gels. R, Raftfraction; C, cytosolic fraction; Stim., stimulus.

7825The Journal of Immunology

by guest on May 25, 2018

http://ww

w.jim

munol.org/

Dow

nloaded from

F-actin levels dropped rapidly to values less than those observed innormal T cells by 15 min following activation (Fig. 5A, bottomright panel). These observations demonstrate that compared withnormal T cells, SLE T cells display faster kinetics of actin poly-merization and depolymerization. We used IgM anti-CD3 Ab forthese experiments because of its ability to cross-link CD3 withoutaddition of a cross-linker Ab as would have been required withOKT3 (39), thus useful in short time point analysis of actinpolymerization.

To understand the mechanisms that alter the rate of actin poly-merization in SLE, we explored the expression and activity of keyproteins involved in actin polymerization such as WASP, Vav1,and Rac-1 (48–50). Immunoblot analysis of lysates obtained fromnormal and SLE T cells demonstrated that expression of WASP

and Rac-1 remained unchanged in SLE T cells (data not shown).However, we observed that in up to 55% of SLE T cells, there wasa significant increase in the expression of Vav1 (normal subjects,n � 13; SLE, n � 21, p � 0.014; Fig. 5B). This increase in Vav1expression was unrelated to SLE disease activity and was observedat random in patients with SLEDAI scores ranging from 0 to 16.Of the 45% patients that did not demonstrate an increase in aVav1:actin ratio, only one patient demonstrated a decrease,whereas all other patients displayed similar levels of Vav1:actinratio as normal subjects (data not shown). Vav1 has been shown tobe important for reorganization of actin cytoskeleton, and in Vav1-deficient mice, Ab-mediated TCR capping is abrogated (50, 51).Therefore, we explored the possibility that Vav1 may be involvedin mediating rapid kinetics of actin polymerization in SLE. Phos-phorylation of tyrosine residues is essential for the activation ofVav1 (52, 53), and Vav1 induces actin polymerization primarilyby its association with actin-binding proteins such as Rac-1, Rac-2,and Rho G proteins (49, 50, 54). Thus, we compared the amountsof phosphorylated Vav1 that associates with actin-bound proteinsin normal and SLE T cells following treatment of 1% TritonX-100-insoluble fractions with cytochalasin B, which is known torelease proteins from the cytoskeletal network (28). Analysis ofactin cytoskeletal fractions of normal and SLE T cells revealedsimilar amounts of Vav1 in normal and SLE T cells (data notshown). However, the levels of phosphorylated Vav1 were higherwithin the cytoskeletal fractions of freshly isolated SLE T cellsthan within normal T cells (Fig. 5C). Following anti-CD3 Ab stim-ulation, the levels of phosphorylated Vav1 remained high in SLET cells with 2.7 � 0.23 ( p � 0.034)- and 3.1 � 0.5, ( p � 0.04)-foldincreases observed at 15 and 30 s, respectively (Fig. 5C). Taken to-gether, the increase in Vav1 expression in SLE T cells and the pres-ence of higher amounts of phosphorylated Vav1 within actin cytoskel-etal fractions strongly suggest a role for Vav1 in mediatingaccelerated kinetics of actin polymerization in SLE T cells.

Enhanced calcium signaling in SLE T cells is mediated bylipid rafts

Calcium flux has been shown to be lipid raft dependent (20, 29).Thus, we asked whether previous cross-linking of lipid raftsevoked different levels of TCR-CD3-mediated calcium responsesin normal and SLE T cells. Cross-linking of lipid rafts by treatingcells with CT-B followed by anti-CT-B has been described before(29, 55) and has been shown to recruit additional phosphoproteinsto lipid rafts (29). Consistent with this view, we observed in nor-mal T cells that recruitment of PLC�1 to membrane patches, anevent critical for TCR-induced calcium flux, could be induced bycross-linking either TCR-CD3 receptor complex by anti-CD3� orlipid rafts by CT-B/anti-CT-B (Fig. 6A). We observed that restingT cells demonstrated PLC�1 in both membrane and cytoplasm,with a majority of cells demonstrating cytoplasmic stores ofPLC�1 (Fig. 6A), a result consistent with that reported previouslyfor Jurkat cells (20). Following anti-CD3 and CT-B/anti-CT-Btreatment, a majority of cells demonstrated a shift in PLC�1 lo-cation to the membrane fractions (Fig. 6A, right panel). Analysisof calcium responses of normal and SLE T cells to anti-CD3 re-vealed that although previous cross-linking of lipid rafts inducedcalcium fluxes in both normal and SLE T cells, calcium responsesobserved in SLE were more rapid and higher than those observedin normal cells (Fig. 6B). Our results are consistent with our pre-vious findings that in SLE T cells, there is higher and sustainedcalcium responses than in normal T cells following stimulationwith anti-CD3 (56). We also reasoned that if the augmented cal-cium responses observed in SLE T cells were mediated throughlipid rafts, disruption of lipid rafts by MbCD should abrogate this

FIGURE 4. IgM anti-CD3-induced receptor capping is accelerated inSLE T cells. Freshly isolated T cells from normal, SLE, and Sjogren’spatients were adhered to polylysine-coated slides, directly fixed or acti-vated with IgM-anti-CD3� Ab at 37°C at 2, 10, and 30 min to induce CD3�cap formation before fixation with 4% paraformaldehyde. Cells were thensurface stained with CT-B FITC (red) and anti-CD3� Ab (clone UCHT1)and counterstained with anti-mouse-TRITC (red) and analyzed by a laserscanning confocal microscope. The merged images (yellow) represent co-localization of respective proteins with GM1. Twenty-five cells werecounted per field and the percentage of positive cells was calculated. Arepresentative cell is shown for each sample. Magnification used was �100oil objective. These results are representative of six experiments. Bottomleft panel, Single-field image of clusters of cells representative of 2-minactivation. Bottom right panel, Statistical representation of the capping datashowing percentages of cells that demonstrate CD3 capping (�, p � 0.05).

7826 LIPID RAFT-MEDIATED SIGNALING IN SLE

by guest on May 25, 2018

http://ww

w.jim

munol.org/

Dow

nloaded from

heightened response. MbCD disrupts lipid rafts by depleting cho-lesterol and can alter signaling by altering associations of signalingproteins with the lipid raft compartment (30). We noted that al-though we could achieve near total abrogation of anti-CD3-induced calcium responses by treating both normal and SLET cells with MbCD, SLE T cells required higher doses of MbCD(12 mM) to achieve the same decrease observed in normal T cells(10 mM; Fig. 6C), thus arguing for a direct role for lipid rafts inmediating the enhanced calcium response in SLE T cells. Theseresults were consistently observed in SLE patients of varying lev-els of disease activity (SLEDAI scores 0, 4, 8, 12, and 12) asshown in the right panel of Fig. 6C. We observed that decreases incalcium responses following treatment with 12 mM MbCD couldbe reversed following resting both normal and SLE T cells for10 h, thus showing that at the doses used for the above experi-ments, MbCD is not irreversibly cytotoxic (Fig. 6C). We also ob-served that prior treatment of these cells with MbCD did not abol-ish calcium response to thapsigargin (Fig. 6E), thus ruling out thepossibility that the observed reductions in calcium response weredue to depletion of calcium reserves caused by MbCD (57). Takentogether, these results demonstrate that calcium signaling is aug-

mented in SLE T cells, at least in part, via a lipid raft-dependentmechanism.

DiscussionIn this study, we provide evidence of significant differences in thelipid raft composition and dynamics between normal and SLE Tcells. First, SLE T cells have a more extensive lipid raft pool andare more robust in their capacity to generate lipid rafts. Second,SLE T cells qualitatively differ in their lipid raft composition.Third, actin cytoskeletal dynamics are accelerated in SLE. Fourth,calcium responses are augmented in SLE T cells, at least in part,via a lipid raft-dependent mechanism. These findings were ob-served independent of the disease activity or treatment. Our resultsshed new light on molecular mechanisms that contribute to in-creased and accelerated TCR-mediated signaling in SLE T cells.

Among mechanisms that control lipid raft-mediated signaling,the size and distribution of lipid rafts are considered to be signif-icant. The small size of individual rafts, for example, is believed tobe important for maintaining the signaling proteins present in themin an inactive state, whereas clustering of multiple rafts inducesactivation of these proteins (17, 20). Thus, preclustering of lipid

FIGURE 5. Accelerated actin cytoskeleton rear-rangement events in SLE T cells. A, FACS analysis ofphalloidin-FITC staining of normal and SLE T cellsfollowing activation with IgM-anti-CD3 (1/100) forindicated time periods. The reaction was stopped aftereach time point by treatment with 3.7% paraformal-dehyde. The top panel represents overlay of histo-grams representing each time point (open) with that of0 s (shaded). The bottom left panel is a representativehistogram overlay comparing phalloidin staining atselected time points (open) with that at 0 s (shaded).The bottom right panel is a statistical comparison ofphalloidin staining at various time points. Values onthe y-axis represent ratios of fold increases in MFI ofeach sample over MFI at 0 h (�, p � 0.05). Data arerepresentative of five separate experiments. B, Whole-cell lysates from normal and SLE T cells, nonacti-vated or activated with IgM-anti-CD3 (1/100) for 2min, were loaded (20 �g/lane) and resolved on4–12% bis-Tris NuPage gels, blotted onto PVDFmembranes, and probed with anti-Vav1 (1/1000), anti-phospho-Vav1 (p-Vav1; 1/1000), and anti-actin(1/1000). A representative gel of six experiments isshown. The right panel shows a statistical analysis ofVav1 expression in 13 normal subjects and 21 SLEpatients. The values on the y-axis are represented asthe mean of ratios of densitometric values of Vav1:actin of each sample. C, Comparison of phospho-Vav1 expression in cytochalasin-B treated TritonX-100-insoluble fractions from 5 � 106 normal orSLE T cells. Data are representative of six separateexperiments.

7827The Journal of Immunology

by guest on May 25, 2018

http://ww

w.jim

munol.org/

Dow

nloaded from

FIGURE 6. Role of lipid rafts in mediating increased calcium responses in SLE T cells. A, PLC�1 distribution in freshly isolated normal T cells or aftercross-linking CD3� with anti-CD3 (OKT3) and goat anti-mouse IgG Abs or after cross-linking surface GM1 with CT-B followed by anti-CT-B as detailedin Materials and Methods. Cells were fixed with paraformaldehyde, permeabilized, and stained with anti-PLC�1 (1/1000) and secondary anti-rabbit-TRITC(red) and analyzed by confocal microscopy. Twenty-five cells were counted per field. Data are representative of cells positively staining for PLC�1. Dataare representative of three similar experiments. Magnification used was �100 oil objective. Right panel shows statistical representation of PLC�1localization. B, [Ca2�]i measurement was performed in normal and SLE T cells loaded with Indo-AM either without or after cross-linking lipid rafts withCT-B/anti-CT-B. Cells were stimulated with OKT3 and anti-mouse IgG and [Ca2�]i was measured with EPICS Altra flow cytometer. A representative offive separate experiments is shown. [Ca2�]i was also measured in normal and SLE T cells upon stimulation with OKT3 and goat anti-mouse IgG (C)following incubation for 30 min with varying concentrations of MbCD in the left panel, and statistical analysis in the right panel (D) following rest for10 h after treatment with 12 mM MbCD (only SLE T cells) or (E) following incubation with 12 mM MbCD for 30 min followed by treatment with 1 �Mthapsigargin. Mean ratio:mean fluorescence ratio corresponding to [Ca2�]i, �, not determined.

7828 LIPID RAFT-MEDIATED SIGNALING IN SLE

by guest on May 25, 2018

http://ww

w.jim

munol.org/

Dow

nloaded from

rafts observed herein in SLE T cells might tilt the balance in favorof preactivation of signaling proteins contained within lipid raftsand may mediate faster downstream signaling. This view is alsosupported by our observations that SLE T cells bear phosphopro-teins within lipid rafts such as Syk and Vav1 and by previousobservations that effector CD4 T cells also display clustering ofsurface receptors (39). Similarly, a larger pool of lipid rafts canalso contribute to accelerated signaling. Thus, memory CD4 andCD8 T cells that have lower thresholds of activation possess largerlipid raft pools than naive T cells (32, 33).

Although it is not known presently what drives preclustering oflipid rafts in SLE T cells, our observations that the lateral mem-brane mobility of proteins (e.g., CD3 complex) and lipids (e.g.,GM1) is significantly enhanced in SLE T cells suggest its role infacilitating preclustering of lipid rafts. It must be mentioned thatour results differ from those reported 20 years ago and suggesteda defect in receptor capping in SLE T cells (58). We believe thatthis difference is due to the fact that IgM anti-CD3 Ab used in ourstudies is a stronger cross-linker of CD3� than OKT3, which re-quires an additional cross-linker such as goat anti-mouse IgG toexert its cross-linking effect, as observed by us previously (G. C.Tsokos, D. L. Farber, and S. Krishnan, unpublished observationand Ref. 24). Therefore, IgM anti-CD3 Ab may overcome the lim-itations of weaker responses to cross-linkers such as OKT3 used inthe previous study. In support of this view is our observation thatpeak receptor capping occurred in normal T cells within 10 minusing IgM anti-CD3, whereas OKT3-induced maximum cappingoccurred within 30 min (our unpublished observation). Addition-ally, the previous study used lower concentrations of OKT3 (625ng/ml) than us. When we used higher concentrations of OKT3 (10�l/ml), we could demonstrate receptor capping in SLE T cells,albeit at a slower rate (between 2 and 5 min, data not shown).Other processes such as dysregulation of glycosylation of proteinsin SLE may also be responsible in mediating lipid raft clustering.It has been shown that dysregulation of �1,6-N-acetylglucosami-nyltransferase V (Mgat5), an enzyme in the N-glycosylation path-way, can induce T cell activation by directly enhancing TCR clus-tering and thus lowering the threshold for T cell activation (59).We had previously observed that defects in glycosylation of thetranscription factor Elf-1 might contribute to defective CD3� chainexpression in SLE T cells (60, 61). Thus, it is likely that otherdefects in protein glycosylation may also coexist in SLE T cellsthat mediate lipid raft clustering.

The role of actin cytoskeletal rearrangements in mediating lipidraft clustering is unclear. However, it has been shown that lipid raftpolarization to the immunological synapse is dependent upon actincytoskeleton reorganization in T cells (49). Thus, during interac-tions between SLE T cells and APC, faster kinetics of actin poly-merization may lead to rapid formation of the immunological syn-apse and thus accelerate TCR signaling. Because polarization oflipid rafts to immunological synapse is Vav1 dependent and Ab-mediated TCR capping is affected in the absence of Vav1 (49–51),our observation of higher levels of Vav1 expression, combinedwith higher levels of activated Vav1 association with actin-bindingproteins in SLE T cells, strongly suggests that Vav1 plays an im-portant role in mediating faster kinetics of receptor capping and actinpolymerization and immunological synapse formation in SLE.

At the biochemical level, interactions between disparate signal-ing proteins present within lipid rafts impose another level of reg-ulation upon TCR signaling. Thus, lipid rafts of naive, effector, andmemory CD8 T cells display distinct protein profiles with distinctfunctional outcomes (33), and Th1 and Th2 cells differentially re-cruit TCR complexes to lipid rafts and display different functions(62). Previously, we had reported “rewiring of TCR” in SLE T

cells in which CD3� is replaced by the FcR� chain in the TCR-CD3 complex and that this modified receptor complex preferen-tially utilizes Syk instead of ZAP-70 kinase (11, 44). Furthermore,we demonstrated that the excitation threshold of normal T cellscould be lowered akin to SLE T cells by overexpressing FcR� thatalso increased the expression and activation of Syk kinase (9).Thus, our observations that FcR� and p-Syk are recruited to lipidrafts in SLE T cells suggest a prominent role for lipid rafts inmediating heightened T cell responses in SLE. Although the pre-cise mechanisms mediating increased calcium responses in SLEare not known, there are important leads in this direction. There isgrowing support for a connection between Vav1 and PLC�1 ininducing calcium flux (54). Previous studies in Vav1 knockoutmice have revealed that Vav1 is important for maintaining sus-tained calcium levels following T cell activation (63). Our obser-vations of higher Vav1 activity in SLE, greater association ofPLC�1 with lipid rafts, and higher calcium flux in SLE T cellsfollowing cross-linking lipid rafts strongly support this theory andsuggest that lipid rafts act as a platform for mediating these events,although we cannot rule out contribution from other important fac-tors. It must be mentioned here that although we could not deter-mine the phosphorylation status of PLC�1, previous studies havedemonstrated that localization of PLC�1 within lipid rafts wouldbe sufficient to induce its activation (40). It is presently unclearwhy FcR� is up-regulated in SLE. However, the dependence ofCD3� on Src kinases for phosphorylation and association with mi-crofilament cytoskeleton (64, 65), the reduced expression of Srckinase Lck in SLE (6), and the ability of FcR� to associate withlipid rafts, combined with the increased association of Syk withlipid rafts (Fig. 3D) that has been shown to phosphorylate FcR�(66), suggest FcR� to be an ideal candidate to functionally asso-ciate with the TCR signaling complex in the context of reducedCD3� in SLE (8).

Similar to our findings in human SLE, previous studies havenoted lowering of activation threshold for T cells from lupus-pronemice that are also hyperresponsive to TCR stimulation (67). Ourpresent findings are significant in this regard and alterations inlipid raft dynamics could be an important contributor of enhancedT cell functions in murine SLE models as well. Currently it is notknown what drives T cells in SLE to display activation pheno-types. We have drawn significant parallels between TCR signalingin SLE and effector T cells and raised the possibility that theseactivation phenotypes displayed by SLE T cells may be due to thepresence of effector cells perpetuated via chronic stimulation byautoantigens (11, 44). This view is supported by 1) our previousobservations that, similar to SLE T cells, effector CD4 T cells alsodisplay membrane clustering of TCR complexes (11, 39); 2) ob-servations reported herein that higher levels of PLC�1 localize tocell membrane of SLE T cells similar to activated Jurkat T cells(40); and 3) association of active Syk, a kinase used preferentiallyin effector T cells (39), is present within lipid rafts of SLE T cells.In addition, these findings lend support to the view that biochem-ical markers of T cell markers are more reliable than phenotypicmarkers such as CD25 and CD69 in determining the long-termactivation status of T cells (39, 68). The T cell population analyzedby us included various T cell subsets such as CD4 and CD8, andwithin CD4 T cells, probably Th1 and Th2 cells. Presently, it isunclear whether there exist differences in lipid raft-mediated sig-naling in these different subsets.

In conclusion, we demonstrate that alterations in the lipid raftdistribution, composition, and biochemical associations contributeto the increased and accelerated TCR-mediated responses in SLET cells. In addition, our studies add SLE to the list of diseases in

7829The Journal of Immunology

by guest on May 25, 2018

http://ww

w.jim

munol.org/

Dow

nloaded from

which abnormal lipid raft processes are used to generate disease-specific immune abnormalities (17). Elucidating the precisesignaling pathways regulated by lipid rafts in SLE may serve toeffectively target these pathways and dampen increased T cellresponses.

AcknowledgmentsWe gratefully acknowledge Dr. Ellis Reinherz (Dana-Farber Cancer Insti-tute, Boston, MA) for the kind gift of IgM anti-CD3 Ab, Dr. MichaelKoenig (WRAIR, Silver Spring, MD) for help with confocal microscopy,Drs. Ashima Saxena and Ramchandra Naik (WRAIR) for help with sucrosegradient experiments, Dr. Juliann Kiang (WRAIR) for help with statisticalanalysis, and Leah Perkins (WRAIR) for her excellent technical assistance.We also thank Drs. Donna Farber (University of Maryland, Baltimore,MD) and Mojgan Ahmadzadeh (National Institutes of Health, Bethesda,MD) for critical reading of this manuscript and Dr. Gary Kammer (WakeForest University School of Medicine, Winston Salem, NC) for extensivecritical discussions.

References1. Kammer, G. M., A. Perl, B. C. Richardson, and G. C. Tsokos. 2002. Abnormal

T cell signal transduction in systemic lupus erythematosus. Arthritis Rheum. 46:1139.

2. Kong, P. L., J. M. Odegard, F. Bouzahzah, J. Y. Choi, L. D. Eardley,C. E. Zielinski, and J. E. Craft. 2003. Intrinsic T cell defects in systemic auto-immunity. Ann. NY Acad. Sci. 987:60.

3. Liossis, S. N., X. Z. Ding, G. J. Dennis, and G. C. Tsokos. 1998. Altered patternof TCR/CD3-mediated protein-tyrosyl phosphorylation in T cells from patientswith systemic lupus erythematosus: deficient expression of the T cell receptor �chain. J. Clin. Invest. 101:1448.

4. Brundula, V., L. J. Rivas, A. M. Blasini, M. Paris, S. Salazar, I. L. Stekman, andM. A. Rodriguez. 1999. Diminished levels of T cell receptor � chains in periph-eral blood T lymphocytes from patients with systemic lupus erythematosus. Ar-thritis Rheum. 42:1908.

5. Pang, M., Y. Setoyama, K. Tsuzaka, K. Yoshimoto, K. Amano, T. Abe, andT. Takeuchi. 2002. Defective expression and tyrosine phosphorylation of the Tcell receptor � chain in peripheral blood T cells from systemic lupus erythema-tosus patients. Clin. Exp. Immunol. 129:160.

6. Jury, E. C., P. S. Kabouridis, A. Abba, R. A. Mageed, and D. A. Isenberg. 2003.Increased ubiquitination and reduced expression of LCK in T lymphocytes frompatients with systemic lupus erythematosus. Arthritis Rheum. 48:1343.

7. Yi, Y., M. McNerney, and S. K. Datta. 2000. Regulatory defects in Cbl andmitogen-activated protein kinase (extracellular signal-related kinase) pathwayscause persistent hyperexpression of CD40 ligand in human lupus T cells. J. Im-munol. 165:6627.

8. Enyedy, E. J., M. P. Nambiar, S. N. Liossis, G. Dennis, G. M. Kammer, andG. C. Tsokos. 2001. Fc� receptor type I � chain replaces the deficient T cellreceptor � chain in T cells of patients with systemic lupus erythematosus. Ar-thritis Rheum. 44:1114.

9. Nambiar, M. P., C. U. Fisher, A. Kumar, C. G. Tsokos, V. G. Warke, andG. C. Tsokos. 2003. Forced expression of the Fc receptor �-chain renders humanT cells hyperresponsive to TCR/CD3 stimulation. J. Immunol. 170:2871.

10. Nambiar, M. P., C. U. Fisher, V. G. Warke, S. Krishnan, J. P. Mitchell,N. Delaney, and G. C. Tsokos. 2003. Reconstitution of deficient T cell receptor� chain restores T cell signaling and augments T cell receptor/CD3-induced in-terleukin-2 production in patients with systemic lupus erythematosus. ArthritisRheum. 48:1948.

11. Krishnan, S., D. L. Farber, and G. C. Tsokos. 2003. T cell rewiring in differen-tiation and disease. J. Immunol. 171:3325.

12. Simons, K., and E. Ikonen. 1997. Functional rafts in cell membranes. Nature387:569.

13. Brown, D. A., and E. London. 2000. Structure and function of sphingolipid- andcholesterol-rich membrane rafts. J. Biol. Chem. 275:17221.

14. Monks, C. R., B. A. Freiberg, H. Kupfer, N. Sciaky, and A. Kupfer. 1998. Three-dimensional segregation of supramolecular activation clusters in T cells. Nature395:82.

15. Dustin, M. L., and A. C. Chan. 2000. Signaling takes shape in the immunesystem. Cell 103:283.

16. Davis, D. M. 2002. Assembly of the immunological synapse for T cells and NKcells. Trends Immunol. 23:356.

17. Simons, K., and R. Ehehalt. 2002. Cholesterol, lipid rafts, and disease. J. Clin.Invest. 110:597.

18. Krummel, M. F., M. D. Sjaastad, C. Wulfing, and M. M. Davis. 2000. Differentialclustering of CD4 and CD3zeta during T cell recognition. Science 289:1349.

19. Ezumi, Y., K. Kodama, T. Uchiyama, and H. Takayama. 2002. Constitutive andfunctional association of the platelet collagen receptor glycoprotein VI-Fc recep-tor �-chain complex with membrane rafts. Blood 99:3250.

20. Xavier, R., T. Brennan, Q. Li, C. McCormack, and B. Seed. 1998. Membranecompartmentation is required for efficient T cell activation. Immunity 8:723.

21. Nambiar, M. P., E. J. Enyedy, C. U. Fisher, S. Krishnan, V. G. Warke,W. R. Gilliland, R. J. Oglesby, and G. C. Tsokos. 2002. Abnormal expression of

various molecular forms and distribution of T cell receptor � chain in patientswith systemic lupus erythematosus. Arthritis Rheum. 46:163.

22. Tan, E. M., A. S. Cohen, J. F. Fries, A. T. Masi, D. J. McShane, N. F. Rothfield,J. G. Schaller, N. Talal, and R. J. Winchester. 1982. The 1982 revised criteria forthe classification of systemic lupus erythematosus. Arthritis Rheum. 25:1271.

23. Nambiar, M. P., E. J. Enyedy, V. G. Warke, S. Krishnan, G. Dennis, H. K. Wong,G. M. Kammer, and G. C. Tsokos. 2001. T cell signaling abnormalities in sys-temic lupus erythematosus are associated with increased mutations/polymor-phisms and splice variants of T cell receptor � chain messenger RNA. ArthritisRheum. 44:1336.

24. Krishnan, S., V. G. Warke, M. P. Nambiar, H. K. Wong, G. C. Tsokos, andD. L. Farber. 2001. Generation and biochemical analysis of human effector CD4T cells: alterations in tyrosine phosphorylation and loss of CD3� expression.Blood 97:3851.

25. Anton, I. M., M. A. de la Fuente, T. N. Sims, S. Freeman, N. Ramesh,J. H. Hartwig, M. L. Dustin, and R. S. Geha. 2002. WIP deficiency reveals adifferential role for WIP and the actin cytoskeleton in T and B cell activation.Immunity 16:193.

26. Darlington, P. J., M. L. Baroja, T. A. Chau, E. Siu, V. Ling, B. M. Carreno, andJ. Madrenas. 2002. Surface cytotoxic T lymphocyte-associated antigen 4 parti-tions within lipid rafts and relocates to the immunological synapse under condi-tions of inhibition of T cell activation. J. Exp. Med. 195:1337.

27. Dupre, L., A. Aiuti, S. Trifari, S. Martino, P. Saracco, C. Bordignon, andM. G. Roncarolo. 2002. Wiskott-Aldrich syndrome protein regulates lipid raftdynamics during immunological synapse formation. Immunity 17:157.

28. Lee, K. H., S. C. Meuer, and Y. Samstag. 2000. Cofilin: a missing link betweenT cell co-stimulation and rearrangement of the actin cytoskeleton. Eur. J. Immu-nol. 30:892.

29. Janes, P. W., S. C. Ley, and A. I. Magee. 1999. Aggregation of lipid rafts ac-companies signaling via the T cell antigen receptor. J. Cell Biol. 147:447.

30. Guo, B., R. M. Kato, M. Garcia-Lloret, M. I. Wahl, and D. J. Rawlings. 2000.Engagement of the human pre-B cell receptor generates a lipid raft-dependentcalcium signaling complex. Immunity 13:243.

31. Cheng, P. C., M. L. Dykstra, R. N. Mitchell, and S. K. Pierce. 1999. A role forlipid rafts in B cell antigen receptor signaling and antigen targeting. J. Exp. Med.190:1549.

32. Tuosto, L., I. Parolini, S. Schroder, M. Sargiacomo, A. Lanzavecchia, andA. Viola. 2001. Organization of plasma membrane functional rafts upon T cellactivation. Eur. J. Immunol. 31:345.

33. Kersh, E. N., S. M. Kaech, T. M. Onami, M. Moran, E. J. Wherry, M. C. Miceli,and R. Ahmed. 2003. TCR signal transduction in antigen-specific memory CD8T cells. J. Immunol. 170:5455.

34. Fra, A. M., E. Williamson, K. Simons, and R. G. Parton. 1994. Detergent-insol-uble glycolipid microdomains in lymphocytes in the absence of caveolae. J. Biol.Chem. 269:30745.

35. Harder, T., P. Scheiffele, P. Verkade, and K. Simons. 1998. Lipid domain struc-ture of the plasma membrane revealed by patching of membrane components.J. Cell Biol. 141:929.

36. Merrill, A. H., Jr., G. van Echten, E. Wang, and K. Sandhoff. 1993. Fumonisin B1inhibits sphingosine (sphinganine) N-acyltransferase and de novo sphingolipidbiosynthesis in cultured neurons in situ. J. Biol. Chem. 268:27299.

37. Heerklotz, H. 2002. Triton promotes domain formation in lipid raft mixtures.Biophys. J. 83:2693.

38. Pizzo, P., and A. Viola. 2003. Lymphocyte lipid rafts: structure and function.Curr. Opin. Immunol. 15:255.

39. Krishnan, S., V. G. Warke, M. P. Nambiar, G. C. Tsokos, and D. L. Farber. 2003.The FcR� subunit and Syk kinase replace the CD3�-chain and ZAP-70 kinase inthe TCR signaling complex of human effector CD4 T cells. J. Immunol. 170:4189.

40. Veri, M. C., K. E. DeBell, M. C. Seminario, A. DiBaldassarre, I. Reischl,R. Rawat, L. Graham, C. Noviello, B. L. Rellahan, S. Miscia, et al. 2001. Mem-brane raft-dependent regulation of phospholipase C�-1 activation in T lympho-cytes. Mol. Cell. Biol. 21:6939.

41. Zhang, W., R. P. Trible, and L. E. Samelson. 1998. LAT palmitoylation: itsessential role in membrane microdomain targeting and tyrosine phosphorylationduring T cell activation. Immunity 9:239.

42. Rodgers, W., B. Crise, and J. K. Rose. 1994. Signals determining protein tyrosinekinase and glycosyl-phosphatidylinositol-anchored protein targeting to a glyco-lipid-enriched membrane fraction. Mol. Cell. Biol. 14:5384.

43. Kabouridis, P. S., A. I. Magee, and S. C. Ley. 1997. S-acylation of LCK proteintyrosine kinase is essential for its signalling function in T lymphocytes. EMBO J.16:4983.

44. Tsokos, G. C., M. P. Nambiar, K. Tenbrock, and Y. T. Juang. 2003. Rewiring theT-cell: signaling defects and novel prospects for the treatment of SLE. TrendsImmunol. 24:259.

45. Valitutti, S., M. Dessing, K. Aktories, H. Gallati, and A. Lanzavecchia. 1995.Sustained signaling leading to T cell activation results from prolonged T cellreceptor occupancy: role of T cell actin cytoskeleton. J. Exp. Med. 181:577.

46. Rozdzial, M. M., B. Malissen, and T. H. Finkel. 1995. Tyrosine-phosphorylatedT cell receptor � chain associates with the actin cytoskeleton upon activation ofmature T lymphocytes. Immunity 3:623.

47. Fuller, C. L., V. L. Braciale, and L. E. Samelson. 2003. All roads lead to actin:the intimate relationship between TCR signaling and the cytoskeleton. Immunol.Rev. 191:220.

7830 LIPID RAFT-MEDIATED SIGNALING IN SLE

by guest on May 25, 2018

http://ww

w.jim

munol.org/

Dow

nloaded from

48. Symons, M., J. M. Derry, B. Karlak, S. Jiang, V. Lemahieu, F. Mccormick,U. Francke, and A. Abo. 1996. Wiskott-Aldrich syndrome protein, a novel ef-fector for the GTPase CDC42Hs, is implicated in actin polymerization. Cell84:723.

49. Villalba, M., K. Bi, F. Rodriguez, Y. Tanaka, S. Schoenberger, and A. Altman.2001. Vav1/Rac-dependent actin cytoskeleton reorganization is required for lipidraft clustering in T cells. J. Cell Biol. 155:331.

50. Fischer, K. D., Y. Y. Kong, H. Nishina, K. Tedford, L. E. 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.

51. 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.

52. Crespo, P., K. E. Schuebel, A. A. Ostrom, J. S. Gutkind, and X. R. Bustelo. 1997.Phosphotyrosine-dependent activation of Rac-1 GDP/GTP exchange by the vavproto-oncogene product. Nature 385:169.

53. Han, J., B. Das, W. Wei, L. Van Aelst, R. D. Mosteller, R. Khosravi-Far,J. K. Westwick, C. J. Der, and D. Broek. 1997. Lck regulates Vav activation ofmembers of the Rho family of GTPases. Mol. Cell. Biol. 17:1346.

54. Tybulewicz, V. L., L. Ardouin, A. Prisco, and L. F. Reynolds. 2003. Vav1: a keysignal transducer downstream of the TCR. Immunol. Rev. 192:42.

55. Gouy, H., P. Deterre, P. Debre, and G. Bismuth. 1994. Cell calcium signaling viaGM1 cell surface gangliosides in the human Jurkat T cell line. J. Immunol. 152:3271.

56. Vassilopoulos, D., B. Kovacs, and G. C. Tsokos. 1995. TCR/CD3 complex-mediated signal transduction pathway in T cells and T cell lines from patientswith systemic lupus erythematosus. J. Immunol. 155:2269.

57. Holleran, B. J., E. Barbar, M. D. Payet, and G. Dupuis. 2003. Differential re-cruitment of alpha2�1 and alpha4�1 integrins to lipid rafts in Jurkat T lympho-cytes exposed to collagen type IV and fibronectin. J. Leukocyte Biol. 73:243.

58. Kammer, G. M. 1983. Impaired T cell capping and receptor regeneration in activesystemic lupus erythematosus. Evidence for a disorder intrinsic to the T lympho-cyte. J. Clin. Invest. 72:1686.

59. Demetriou, M., M. Granovsky, S. Quaggin, and J. W. Dennis. 2001. Negativeregulation of T-cell activation and autoimmunity by Mgat5 N-glycosylation. Na-ture 409:733.

60. Juang, Y. T., E. E. Solomou, B. Rellahan, and G. C. Tsokos. 2002. Phosphory-lation and O-linked glycosylation of Elf-1 leads to its translocation to the nucleusand binding to the promoter of the TCR �-chain. J. Immunol. 168:2865.

61. Tsokos, G. C., M. P. Nambiar, and Y. T. Juang. 2003. Activation of the Etstranscription factor Elf-1 requires phosphorylation and glycosylation: defectiveexpression of activated Elf-1 is involved in the decreased TCR � chain geneexpression in patients with systemic lupus erythematosus. Ann. NY Acad. Sci.987:240.

62. Balamuth, F., D. Leitenberg, J. Unternaehrer, I. Mellman, and K. Bottomly. 2001.Distinct patterns of membrane microdomain partitioning in Th1 and Th2 cells.Immunity 15:729.

63. Costello, P. S., A. E. Walters, P. J. Mee, M. Turner, L. F. Reynolds, A. Prisco,N. Sarner, R. Zamoyska, and V. L. Tybulewicz. 1999. The Rho-family GTPexchange factor Vav is a critical transducer of T cell receptor signals to thecalcium, ERK, and NF-�B pathways. Proc. Natl. Acad. Sci. USA 96:3035.

64. van Oers, N. S., N. Killeen, and A. Weiss. 1996. Lck regulates the tyrosinephosphorylation of the T cell receptor subunits and ZAP-70 in murine thymo-cytes. J. Exp. Med. 183:1053.

65. Rozdzial, M. M., C. M. Pleiman, J. C. Cambier, and T. H. Finkel. 1998. pp56Lck

mediates TCR �-chain binding to the microfilament cytoskeleton. J. Immunol.161:5491.

66. Shiue, L., J. Green, O. M. Green, J. L. Karas, J. P. Morgenstern, M. K. Ram,M. K. Taylor, M. J. Zoller, L. D. Zydowsky, J. B. Bolen, et al. 1995. Interactionof p72syk with the � and � subunits of the high-affinity receptor for immuno-globulin E, Fc�RI. Mol. Cell. Biol. 15:272.

67. Vratsanos, G. S., S. Jung, Y. M. Park, and J. Craft. 2001. CD4� T cells fromlupus-prone mice are hyperresponsive to T cell receptor engagement with lowand high affinity peptide antigens: a model to explain spontaneous T cell activa-tion in lupus. J. Exp. Med. 193:329.

68. Harrington, L. E., M. Galvan, L. G. Baum, J. D. Altman, and R. Ahmed. 2000.Differentiating between memory and effector CD8 T cells by altered expressionof cell surface O-glycans. J. Exp. Med. 191:1241.

7831The Journal of Immunology

by guest on May 25, 2018

http://ww

w.jim

munol.org/

Dow

nloaded from