Interleukin-7 Compartmentalizes Its Receptor SignalingComplex to Initiate CD4 T Lymphocyte Response*□S

Received for publication, January 15, 2010 Published, JBC Papers in Press, February 18, 2010, DOI 10.1074/jbc.M110.104232

Thierry Rose‡1, Anne-Helene Pillet‡2, Vincent Lavergne‡, Blanche Tamarit‡, Pascal Lenormand§,Jean-Claude Rousselle§, Abdelkader Namane§, and Jacques Theze‡

From the ‡Institut Pasteur, Unite d’Immunogenetique Cellulaire, Departement Infection et Epidemiologie, Departementd’Immunologie, and §Plate-Forme Proteomique, Genopole, 25 rue du Dr Roux, 75724 Paris Cedex 15, France

Interleukin (IL)-7 is a central cytokine that controls homeo-stasis of the CD4 T lymphocyte pool. Here we show on humanprimary cells that IL-7 binds to preassembled receptorsmadeupof proprietary chain IL-7R� and the common chain �c sharedwith IL-2, -4, -9, -15, and -21 receptors. Upon IL-7 binding, bothchains are driven in cholesterol- and sphingomyelin-rich raftswhere associated signaling proteins Jak1, Jak3, STAT1, -3, and-5 are found to be phosphorylated. Meanwhile the IL-7�IL-7Rcomplex interacts with the cytoskeleton that halts its diffusionas measured by single molecule fluorescence autocorrelatedspectroscopy monitored by microimaging. Comparative immu-noprecipitations of IL-7R� signaling complex from non-stimu-lated and IL-7-stimulated cells confirmed recruitment of pro-teins such as STATs, but many others were also identified bymass spectrometry from two-dimensional gels. Among re-cruited proteins, two-thirds are involved in cytoskeleton andraft formation. Thus, early events leading to IL-7 signal trans-duction involve its receptor compartmentalization into mem-brane nanodomains and cytoskeleton recruitment.

Interleukin (IL)3-7 is a central cytokine that regulates CD4 Tcell homeostasis (1–5). It is secreted by a number of cell sourc-es: stromal cells in the red marrow and thymus, keratinocytes,dendritic cells, and endothelial cells (6). CD4 T cell lymphope-nia increases the expression of circulating IL-7, resulting in theexpansion of CD4 T cell subsets by increasing the metabolismand inactivating cell cycle inhibitors (7, 8). It also lowers thethreshold of TCR activation by foreign and self-antigens (9, 10),and it prolongs these effects by inducing anti-apoptotic factorsBcl2 and Bcl-xL, and inhibiting pro-apoptotic factors such asBad and Bax (11–13).

IL-7 binds to its receptor, IL-7R, which is made up of twoglycosylated chains anchored to themembrane by a single heli-cal transmembrane domain (14): IL-7R� and the common �chain. IL-7R�, also known as CD127 (65 kDa, 459 amino acids),is also shared by thymic stromal lymphopoietin (15, 16). Thecommon� chain (�c), also known asCD132 (56 kDa, 369 aminoacids), is shared by IL-2, -4, -9, -15, and -21 (5). IL-7R� is highlyand �c weakly expressed at the surface of resting CD4 T cells.Their stimulation by IL-7 down-regulates the expression ofIL-7R�, which disappears from the cell surface after 12 h, andup-regulates quickly the expression of �c, optimal after 12 h(17–20). IL-7 has high affinity for the IL-7R heterodimer (Kd �35 � 10�12 M), low affinity for its single proprietary chainIL-7R� (Kd � 3 � 10�9 M), and very low affinity for �c (Kd �250� 10�9 M) (21). IL-7 has been cocrystallized with the extra-cellular fragment of IL-7R� in the absence of any �c fragment(22). IL-7R� has a long cytoplasmic domain (195 amino acids),which is responsible for binding a large array of proteinsinvolved in the signaling pathways that support cell survival andproliferation pathways (4, 23). These include Jak1, which isinvolved in the Jak/STAT pathway. The �c chain has a shortercytoplasmic domain (86 amino acids), which binds Jak3. BothJak1 and Jak3, carried by their respective receptor chains, arerequired to phosphorylate themselves, then the IL-7R� carbox-yl-terminal Tyr (Tyr(P)-456) upon IL-7 binding. This Tyr(P)-456 provides a single binding site for STAT1, STAT3, andmainly STAT5a and STAT5b. Bound STATs are then phos-phorylated by the activated pJak1�pJak3 complex (4, 23). Afterphosphorylation, the STATs dissociate, dimerize, and aretranslocated into the nucleus where they induce transcrip-tion of gene clusters involved in cell programs (23). Mitogen-activated protein kinase and phosphatidylinositol 3-kinasepathways are also triggered by IL-7�IL-7R binding and giverise to mitogenic and anti-apoptotic signals. Although awealth of information is available on the various kinasesinvolved in IL-7 signal transduction, less is known of the IL-7signaling complexes.This work aimed to describe the IL-7R chain assembly pro-

cess and the protein content of the signaling complexes beforeand after IL-7 binding. Our observations in primary CD4 Tlymphocytes were based on kinetic investigations of fluores-cent-tagged cell components followed by confocal micros-copy on living cells, and on biochemical investigations ofpurified complexes by immunoblotting and mass spectrom-etry. For the first time, a �c-cytokine receptor is shown to beassociated with lipid rafts and cytoskeleton is shown to be

* This work was supported in part by the Institut Pasteur.□S The on-line version of this article (available at http://www.jbc.org) contains

supplemental Fig. S1 and Table S1.1 To whom correspondence should be addressed. Fax: 33-0-1-45-68-86-39;

E-mail: rose@pasteur.fr.2 Supported by a fellowship from the Ministere de l’Education Nationale et de

la Recherche.3 The abbreviations used are: IL, interleukin; ACF, autocorrelated function;

CCF, cross-correlated function; DRM, detergent-resistant membranedomain; FCS, fluorescence autocorrelated spectroscopy; FCCS, fluores-cence cross-correlated spectroscopy; IP, immunoprecipitation; Jak, Januskinase; mAbb, biotinylated monoclonal antibody; pAb, polyclonal anti-body; MS, mass spectrometry; SA488/633, streptavidin-Alexa Fluor 488/633; STAT, signal transducer and activator of transcription; TCR, T cellreceptor; ERK, extracellular signal-regulated kinase; MALDI-TOF, matrix-assisted laser desorption ionization time-of-flight.

THE JOURNAL OF BIOLOGICAL CHEMISTRY VOL. 285, NO. 20, pp. 14898 –14908, May 14, 2010© 2010 by The American Society for Biochemistry and Molecular Biology, Inc. Printed in the U.S.A.

14898 JOURNAL OF BIOLOGICAL CHEMISTRY VOLUME 285 • NUMBER 20 • MAY 14, 2010

This article has been withdrawn by Thierry Rose, Blanche Tamarit, Pascal Lenormand, Abdelkader Namane, and Jacques Thèze. Anne-Hélène Pillet, Vincent Lavergne, and Jean-Claude Rousselle could not be reached. Fig. 4 was inappropriately presented. The withdrawing authors assert that the

results of this article are valid.

by guest on September 22, 2020

http://ww

w.jbc.org/

Dow

nloaded from

by guest on September 22, 2020

http://ww

w.jbc.org/

Dow

nloaded from

by guest on September 22, 2020

http://ww

w.jbc.org/

Dow

nloaded from

involved in IL-7R compartmentalization at the level of a sin-gle complex in a single living primary human cell. Thisapproach, at the molecular level and in real time, describesthe very first steps in IL-7 response initiation that is crucialto CD4 T cell activation.

MATERIALS AND METHODS

Human CD4 T Lymphocyte Purification—Venous blood wasobtained from healthy volunteers through the EFS (Etablisse-ment Francais du Sang, Centre Cabanel, Paris). Peripheralbloodmononuclear cells were purified by density gradient cen-trifugation on Lymphoprep solution (Axis-Schield). CD3�/CD4� NT cells were prepared from human peripheral bloodmononuclear cells by separation on magnetic beads (CD4�

negative purification kit, Miltenyi Biotec). The enriched CD4 Tcell population contained�95%CD3�/CD4� cells. The recov-eredCD4Tcellswere not activated as controlled by the absenceof CD69 and CD25 expression.Cell purity of preparations and IL-7R chain expression at the

cell surface were analyzed by flow cytometry with labeled anti-bodies. Cells were harvested and resuspended in 50 �l ofcytometer buffer (phosphate-buffered saline with 0.02%sodium azide and 5% fetal bovine serum) and labeled for 1 h at4 °C with antibodies to CD4 (eBioscience). CD4 receptorexpression wasmeasured by flow cytometry in CD4T cells on aCyan LXTM cytometer (DakoCytomation). Data were acquiredwith Summit version 4.1 software (Dako) and analyzed usingFlowJo version 8.3.3 software (Tree Star).For cytokine activation, CD4 T cells were resuspended at 106

cells/ml in RPMI 1640medium (Lonza, Verviers, Belgium) sup-plemented with 5% fetal bovine serum, 50 mM HEPES, pH 7.4,glutamine, penicillin, streptomycin, and fungizone (completemedium) in 24-well plates, were treated with 1 nM recombinantglycosylated human IL-7 (Cytheris) at 37 °C in a 5% CO2humidified atmosphere.FCS and FCCS Analysis of IL-7R Chain Assembly and Diffu-

sion at the Surface of Living Cells—Fluorescence auto- andcross-correlated spectroscopy (FCS/FCCS) measurementswere made on living cells using an inverted laser scanning con-focal microscope (LSM510), combined with a ConfoCor2 FCSsystem (Zeiss). Depth of field was spatially filtered through a30–300-�m pinhole and fluorescence light was split into twodetection channels with the following excitation/emissionwavelengths: Alexa Fluor 488 (Invitrogen) (Ar, 488 nm/550–600 nm) and Alexa Fluor 633 (Invitrogen) (He/Ne, 633nm/BP680 nm). FCS and image data were acquired and thenanalyzed by LSM software (Zeiss). The observation volumes ofthe LSM and the ConfoCor systems were aligned by bleachingspots with the ConfoCor in a dried layer of rhodamine-6G(Sigma) on a 0.16-mm glass slide from cross-hair positions inLSM images. Pinhole setting, volume overlap, and ConfoCor/LSM alignment were processed before each measurement ses-sion when laser power and temperature were at equilibrium.Pinhole x-, y-, and z-positions were carefully aligned for bothchannels 488 and 633 nm, and their confocal emission volumeoverlapwas tuned by adjusting the collimator. Confocal volumewas calibrated with rhodamine-6G and tested with beads asdescribed under supplemental Fig. S1. Autocorrelation and

cross-correlation functions were extracted from fluorescenceintensity fluctuations during 30–60-s acquisition times and fit-ted to three-, two-, or three-/two-dimensional mixed mathe-matical models according to Ref. 24 with Origin software(OriginLab).Three-dimensional diffusions of rhodamine-6G, strepta-

vidin-Alexa 488 (SA488, Invitrogen), streptavidin-Alexa633 (SA633, Invitrogen), IgG-biotin�streptavidin-Alexa 488(mAbb�SA488), and IL-7-biotin�streptavidin-Alexa 633 (IL-7b�SA633) were analyzed in RPMI medium at 37 °C. Autocor-relation functions (ACF) of fluorescent particle three-dimen-sional diffusion were fitted with the following model,

G��� � 1/N�1 � ftriplet � f1e�� /�,triplet/�1 � ftriplet�i� fi�1

� ��/�i��1�1 � ��/�i�/s2�1/ 2 (Eq. 1)

where N is the number of fluorescent molecules in the actualdetection volume defined by Veff �3/20

3, and s is the struc-tural parameter representing its shape, as derived from the ratiobetween its axial and lateral radius s z0/0, i is the componentnumber, ftriplet and �triplet are the fraction and diffusion time ofparticles with their fluorophore in the triplet state, fi and �i arethe fraction and diffusion time of fluorescent component i (24).The structure factor swas approximated to 5.0 frombest fits for0

2 in the range 50� 103—150� 103 nm2. Below 50� 103 andbeyond 150 � 103 nm2, s deviation indicated that the volumeshape was clearly affected (waved zones are shown insupplemental Fig. S1) and restricted the exploration field withour commercial system. The following values were calculatedfor �d at0

2 70� 103 nm2 in RPMI at 37 °C: �D[SA488] 243�s, �d[SA633] 245 �s, �D[mAbb�SA488] 312 �s. Up to 10ACF were recorded for each of 6 to 10 0

2 values. �D wereaveraged from these 10 ACF per 0

2 values, fitted with Equa-tion 1 for one component. Effective lateral diffusion rates Deffwere calculated from the linear regression of the diffusion time�d versus 0

2 plots.

�D � �0 � 02/4Deff (Eq. 2)

Accordingly, the following values were calculated for controlsin RPMI at 37 °C: Deff(SA488sol) 72 �m2/s, Deff(SA633sol) 72 �m2/s, and Deff(mAbb/SA488sol) 56 �m2/s.

CD4 T cells were observed before and 10 min after additionof 1, 10, and 100 nM IL-7. When indicated, cells were washedtwice in RPMI, 10 mM HEPES, then treated at 37 °C in 10 mM

RPMI/HEPES either for 25 min with 2 and 10 �M cytochalasinD (CytD, Sigma) or for 25 min with cholesterase oxidase (1unit/ml, Sigma) and/or for 5 min with sphingomyelinase (0.1unit/ml, Sigma) as described previously (25) prior to addingIL-7. IL-7R� and �c chains expressed at the cell surface were,respectively, labeled with anti-IL-7R� (mAb 40131, R&D Sys-tems Inc.) tagged with anti-mouse IgG-Alexa 488 (sAb/A488,Invitrogen) or biotinylated anti-IL-7R� (mAbb) tagged withstreptavidin-Alexa 488 and biotinylated anti-�c (mAbbTugH4,Pharmingen) tagged with streptavidin-Alexa 633 in a 1:4 molarratio to avoid mAb aggregation by tetrameric streptavidins.Antibodies were used in large excess to avoid Ab-inducedaggregation of receptors. Biotinylated IL-7 was labeled with

IL-7-induced Human CD4 T Cell Response Initiation

MAY 14, 2010 • VOLUME 285 • NUMBER 20 JOURNAL OF BIOLOGICAL CHEMISTRY 14899

by guest on September 22, 2020

http://ww

w.jbc.org/

Dow

nloaded from

streptavidin-Alexa 633 (IL-7b�SA633) in a 1:4 molar ratio aswell to avoid cytokine aggregation. Receptor diffusionmeasure-ments were acquired within a 10–30-min time frame in culturemedium to minimize internalization effects and IgG-inducedreceptor aggregation. Confocal volumes in the FCS optical sys-tem and imaging laser scanning microscope were adjusted andcentered on the cytoplasmic membrane. All images and FCSdata were acquired at 37 °C using a thermostated dish holderand objective ring. Ten 30-s acquisitions were recorded at 6 pin-hole values over 10 points spread over a 3 � 3-�m square of asingle cell immobilized on poly-L-lysine-coated glass slides at thebottom of the dish. Briefly, autocorrelation curves, G(�), were fit-ted with one up to three components: one fast three-dimensiondiffusion based component (free mAbb�SA or IL-7b�SA) charac-terized by its diffusion time in the confocal volume (�D,fast) anddescribed by Equation 1, then one intermediate (�D,inter) and oneslow (�D,slow) two-dimensional diffusion-based component asdescribedbyEquation3, corresponding to themembrane-embed-ded receptor chains in different aggregation states.

G��� � 1/N�1 � ftriplet � ftriplete�� /�,triplet/

�1 � ftriplet�i�fi�1 � ��/�Di��1� (Eq. 3)

Overall the model used with 3 components is given by Equa-tion 4.

G��� � 1/N�1 � ftriplet

� ftriplete�� /�,triplet/�1 � ftriplet�ffast�1 � ��/�D,fast�

�1�1

� ��/�D,fast/s2)�1/ 2 � finter�1

� ��/�D,inter��1 � fslow�1 � ��/�D,slow��1� (Eq. 4)

where ffast, finter, and fslow are relative fractions of componentpopulations. Effective diffusion rates Deff were also extractedfrom the linear regression of �D versus the observed membranesurface area 0

2 and confinement time, �0 at the y intercept atthe waist origin, 0

2 0 in Equation 2. We counted the diffus-ing particles inVeff from the extrapolation at the origin of auto-correlation functions.

G�� � 0� � 1/N (Eq. 5)

Autocorrelation functions were normalized for comparisonwith Equation 6.

Normalized G��� � N�G��� � 1� (Eq. 6)

Cross-correlation functions (CCF) between IL-7R��mAbb�SA488 and IL-7b�SA633 were fitted using the following three-dimensional (Equation 7) and two-dimensional (Equation 8)diffusion models, respectively

Gij��� � 1/Nij�1 � ftriplet,I � fie�� /�,triplet,i/�1 � ftriplet,i�1 � ftriplet,j

� fje�� /�,triplet,j/�1 � ftriplet,j� fij�1 � ��/�ij�

�1�1 � ��/�ij�/s2]�1/ 2

(Eq. 7)

Gij��� � 1/Nij�1 � ftriplet,i � fie�� /�,triplet,i/�1 � ftriplet,i�1

� ftriplet,j � fje�� /�,triplet,j/�1 � ftriplet,j�fij�1 � ��/�ij�

�1 (Eq. 8)

where �ij is the diffusion time of the complex IL-7R��mAbb�SA488�IL-7b�SA633 considered as a unique species car-rying both fluorophores mostly anchored to the membrane.SA488/SA633 fluorescence cross-talkwas corrected froma sin-gle labeling experiment of the same system recorded in bothchannels (24).Among our major concerns was receptor aggregation

induced by antibodies or streptavidin, and receptor internal-ization. Diffusion plots (Equation 2) give control of receptorchain aggregation slowing down their diffusion with increas-ing time: the plot is curving up when acquisition starts fromsmall to large 0

2 as observed for diffusion of IL-7R� labeledwith the primary�secondary antibody complex mAb�sAb�A488 beyond 10 min of acquisition time and beyond 20–30min with mAbb�SA488. Controls were done with either freeunlabeled streptavidin, biotin (Sigma), or mouse IgG addedin culture medium. Usual conditions applied to lower recep-tor internalization were unfavorable: decrease of tempera-ture affected membrane fluidity and cytoskeleton adhesion,and deoxyglucose and sodium azide affected receptor com-partmentalization. The linearity of the diffusion plot (Equa-tion 2) was used to reject acquisitions affected by internal-ization events and/or antibody aggregation.Cell Lysate Ultracentrifugation through a Sucrose Gradient—

Purified primary CD4 T lymphocytes were stimulated or notby IL-7 (2 nM, 5 min), then harvested, cooled by adding ice-cold phosphate-buffered saline, centrifuged, and lysed in0.5% Triton X-100 buffer (50 mM Tris/HCl, pH 7.4, 5 mM

EGTA, 5 mM EDTA, 30 mM NaF, 20 mM Na-pyrophosphate,1 mM Na3VO4, 1 mM phenylmethylsulfonyl fluoride, 10 �M

leupeptin). Briefly, 300 �l of lysate supernatant from 5 � 106

cells was loaded on the top of 5 ml of a preformed 5–40%sucrose/Triton buffer gradient and centrifuged for 16 h at 50krpm using a SW50 rotor in a Beckman ultracentrifuge.Tube contents were divided into 18 280-�l samples and 60 �lwere loaded on SDS-PAGE (7% acrylamide-bisacrylamide)and transferred onto a Hybond-ECL nitrocellulose mem-brane (GE Healthcare) overnight at 4 °C. Membranes weresaturated with bovine serum albumin (Sigma), incubatedwith primary antibodies, and washed in TBS, 0.5% Tween 20buffer before being incubated with horseradish peroxidase-coupled anti-mouse (Jackson), anti-rat (Amersham Bio-sciences), or anti-goat (Jackson) secondary antibodies. Pro-teins were then revealed by ECL-plus Western blottingdetection reagents (GEHealthcare).Western blots were ana-lyzed with anti-IL-7R� (mAb 40131, R&D Systems), anti-�c(pAb AF284, R&D Systems, or mAb TugH4, Pharmingen),and anti-flottilin-1 as a detergent-resistant membrane do-main (DRM) marker (mAb 610821, BD Biosciences).When indicated, pooled fractions of the sucrose gradient

were immunoprecipitated using anti-IL-7R�. Immunoprecipi-tated proteins were loaded on SDS-PAGE (7% acrylamide-bi-sacrylamide) and analyzed by Western blotting using anti-Tyr(P) (Ab 4G10, Upstate), anti-pJak (Tyr(p)1022/1023, Ab44–422G, BIOSOURCE), anti-pSTAT3 (Tyr(p)705, Ab 3E2,Cell Signaling), and anti-pSTAT5 (Tyr(p)694, Ab 14H2, CellSignaling).

IL-7-induced Human CD4 T Cell Response Initiation

14900 JOURNAL OF BIOLOGICAL CHEMISTRY VOLUME 285 • NUMBER 20 • MAY 14, 2010

by guest on September 22, 2020

http://ww

w.jbc.org/

Dow

nloaded from

Immunoprecipitation of IL-7R�-bound Proteins for Two-di-mensional PAGEandMSAnalysis—After 8–24 h of incubationat 37 °C, purified primary CD4 T lymphocytes were activatedwith or without 1 nM IL-7 for 5 min. Cells were harvested,washed twice in phosphate-buffered saline, and lysed for 20min on ice with 0.5% Triton X-100 lysis buffer as describedabove. Immunoprecipitationswere processed from lysates con-taining 350 �g (two-dimensional PAGE) of proteins pulleddown with anti-IL-7R� mAb (mAb 40131, R&D Systems) for1 h at 4 °C on a spinning wheel with 250 pM IL-7 when neces-sary.G-protein coupled-Sepharose beads (GEHealthcare)werethen added for 1 h at 4 °C. After four washes in 0.5% TritonX-100 buffer containing 250 pM IL-7 when necessary, sampleswere resuspended in Triton/urea buffer for IEF (7 M urea, 2 M

thiourea, 1 M dithiothreitol, 0.5% ampholyte, 0.5% TritonX-100, 50 mM Tris/HCl, pH 7.4, 5 mM EGTA, 5 mM EDTA, 30mM NaF, 20 mM sodium pyrophosphate, 1 mM Na3VO4, 1 mM

phenylmethylsulfonyl fluoride, 10 �M leupeptin).Western Blot Analysis—Immunoprecipitated protein sam-

ples used for two-dimensional PAGE were also analyzed byWestern blotting. Immunoprecipitated fractions from lysednon-stimulated and IL-7-stimulated CD4 T cells were loadedon one-dimensional SDS-PAGE (7% acrylamide-bisacrylam-ide) and transferred onto a nitrocellulosemembrane treated forWestern blotting as described above. The following primaryantibodies were used: anti-IL-7R� (mAb 40131, R&D Systems),anti-�c (pAb AF284, R&D Systems), anti-Jak1 (pAb sc-295,Santa Cruz Biotechnology), anti-Jak3 (pAb sc-513, SCB), anti-STAT1(pAb sc-592, SCB), anti-STAT3 (pAb sc-482, SCB),anti-STAT5 (pAb sc-835, SCB), anti-ERK1/2 (pAb 9102, CellSignaling), anti-actin (pAb sc-1616, SCB), anti-�-tubulin (mAbsc-5286, SCB), anti-ezrin (rabbit IgG pAb, kind gift of Dr. M.Arpin, Institut Curie), and anti-moesin (mAb 610401, BDBiosciences).Two-dimensional PAGEAnalysis of IL-7R Signaling Complex

Components, Identification by Mass Spectrometry—IEF gelstrips (16 cm, pH 3–10, Bio-Rad) were loaded with IP proteinswith anti-IL-7R� from 350 �g of proteins from non-stimulatedor IL-7-stimulated CD4 T cell lysates according to the Bio-Radprocedure. Strips equilibrated with dithiothreitol buffer, theniodoacetamide buffer were loaded over 16� 16-cm SDS-PAGE(12% acrylamide-bisacrylamide). The gels were stained withSypro-Ruby (Invitrogen). Acrylamide gel plugs were cut andremoved from Sypro-stained spots using an automat and pro-cessed as described previously (21). Trypsin-proteolyzed sam-ples were loaded by the robot onto the MALDI plate target,then dried and mass spectra were acquired as described previ-ously (21) on a 4800MALDI/TOF/TOFAnalyzer (Applied Bio-systems) performingMS andMS/MS in automatic mode (3000laser shots). Subspectra of 50 laser shots were accumulated forthe 15 most intense m/z peaks from MS followed by air colli-sion-induced peptide fragmentation (2 kV) and MS/MS analy-sis to determine the peptide sequence. Trypsin autolysis pep-tides were used for calibration. Mascot GPS Explorer version3.6 software (Matrix Science) was used to scout Swissprot database release 94 (SIB, EBI) supplemented with variant and sig-nal-free sequences. Only monoisotopic masses between 800and 4000m/zwere used with amaximumpeptidemass error of

50 ppm for MS and 0.3 Da for MS/MS. One incomplete cleav-age was allowed per peptide and possible methionine oxidationand cysteine derivatization were considered.

RESULTS

Preassembled Free IL-7R Is Compartmentalized after IL-7Binding—The mechanism of IL-7R� and �c assembly to formthe active receptor, and involvement of IL-7 in this formationwere analyzed by FCS measured by confocal microscopy at thesurface of living primary human CD4 T lymphocytes. Fluores-cent particle species were separated and quantified from theirdiffusion times in the microscope confocal volume. Diffusiontimes �D of individual receptor chains IL-7R� (Fig. 1E, e, �D 25.8 ms for 0

2 70 � 103 nm2) and �c (Fig. 1E, d, �D 23.7ms) were calculated from autocorrelation functions of fluores-cence intensity fluctuations of their labels, fitted with adapteddiffusion models (Fig. 1, A–C) for increasing sizes of theobserved membrane surface areas (Fig. 1F). ACF were normal-ized with Equation 6 for comparison. IL-7R� and �c effectivediffusion rates Deff were extracted from the plots shown in Fig.2A.Deff is inversely correlated to slopes of diffusion times versusobserved membrane surface areas 0

2: accordingly with Equa-tion 2, Deff 0.215 � 0.025 and 0.252 � 0.030 �m2/s, forIL-7R� and �c, respectively. Diffusion time of the IL-7R���ccomplex was calculated from cross-correlation functions ofboth channel fluorescence intensity fluctuations: one forIL-7R��mAbb�A488 and the other for �c�mAbb�SA633 asdetailed in Fig. 1E, f. The diffusion rate for the IL-7R���c com-plex was determined to be Deff 0.173 � 0.026 �m2/s, about20% slower than for IL-7R� (Fig. 2A). Oligomerization does notsignificantly affect the IL-7R� diffusion rate. Eight hours aftertheir purification start, we counted the chains present at thesurface of the CD4 T lymphocytes by extrapolation of the auto-correlation function at the origin (Equation 5) in the observedsurface 0

2. Total concentrations of both chains at the surfacecorrespond to [IL-7R�] 108 pmol/m2 and [�c] 12.3 pmol/m2. The concentration of the complex was extrapolated fromthe cross-correlated function: [IL-7R���c] 10.0 pmol/m2. Aswe had now determined the concentration of different molec-ular species, we were able to evaluate the dissociation constantat apparent equilibrium:Kd 24.3 pmol/m2. About 80.2% of �cwere assembled in IL-7R���c complexes, whereas only 10.1% ofIL-7R� were engaged in heterodimers. We checked formationof the IL-7R� homodimer by labeling IL-7R� with anti-IL-7R�mAbb coupled to SA488 (50%) and SA633 (50%). From cross-correlations that accounted for half the IL-7R��IL-7R�homodimers (50% SA488/SA633 versus 25% SA488/SA488 and25% SA633/SA633) wemeasured theKd 336 pmol/m2. In all,17.4% of IL-7R� were engaged in homodimers and were thenunavailable for �c-binding. The correctedKd for IL-7R���c dis-sociation was therefore 19.6 pmol/m2. IL-7R� affinity for itselfis thus 17-fold weaker than for �c.Interestingly, IL-7 binding affects its receptor diffusion as

shown with the longer diffusion times at the inflection point inFig. 1E, curve f, by comparison with curve e for the sameobserved membrane surface area. IL-7 slows down 2.6-fold theIL-7R� diffusion rate, as shown in Fig. 2B,Deff 0.065 � 0.014�m2/s. The effect was even more obvious when we observed

IL-7-induced Human CD4 T Cell Response Initiation

MAY 14, 2010 • VOLUME 285 • NUMBER 20 JOURNAL OF BIOLOGICAL CHEMISTRY 14901

by guest on September 22, 2020

http://ww

w.jbc.org/

Dow

nloaded from

the complex by analyzing the cross-correlation betweenIL-7R��mAbb�SA488 diffusion and the diffusion of biotinylatedIL-7 labeledwith SA633 (IL-7b�SA633) (Fig. 1E, curve h, andG):Deff 0.024 � 0.0038 �m2/s, 7.2-fold slower than ���c and9-fold than � (Fig. 2B). As can be seen in Fig. 2,A and B, extrap-olation of the IL-7R���c linear regression crosses the y axisbelow the origin before (�0 �62 � 9.8 ms) and after IL-7binding (�0 �242� 38ms). �0 was described byMarguet andcolleagues (25) as the confinement time assessing the retentionduration of particles by the cytoskeleton meshwork fencingreduced area below the plasma membrane. �0 increases4-fold after IL-7-binding, suggesting not only collision of thereceptor complex but also tighter interactions with cytoskel-eton components.

Mechanism Involved in IL-7RCompartmentalization—IL-7 bind-ing slowed down the diffusion rateof the receptor and increased itsconfinement time. In attempts todemonstrate that the cytoskeletonis involved in confinement time andthe diffusion brake, we used a drugthat depolymerizes actin fibers:CytD. Fig. 3A shows that CytD didnot significantly affect the diffusionrates of IL-7-free receptors: Deff 0.227 � 0.030 �m2/s, �0 �2.4 �0.4 ms for 10 mM CytD comparedwith Deff 0.173 �m2/s, �0 �62ms without CytD. By contrast, aCytD dose-dependent increase ofIL-7R diffusion was noted in thepresence of IL-7: Deff 0.210 �0.033�m2/s, �0 �8.1� 1.3ms for10 mM CytD compared with 0.024�m2/s, �0 �242mswithoutCytD.This 9-fold increase suggests thatthe IL-7�IL-7R complex is no longerinteracting with the cytoskeletonmeshwork and is released from itsconfinement state. Sphingomyeli-nase and cholesterol oxidase, whichinhibit lipid raft formation, had nosignificant effect on receptor slow-ing in the presence of its ligand (Fig.3B): Deff 0.026 � 0.0036 �m2/s,�0 �182 � 26 ms. However,sphingomyelinase and cholesteraseoxidase did have a noticeable effectafter CytD treatment as theIL-7�IL-7R complex was once morefreely diffused, as observed for theligand-free receptor:Deff 0.349 �0.047 �m2/s, �0 �11.4 � 1.7 ms(Fig. 3C).IL-7�IL-7R Is Extracted into De-

tergent-resistant Microdomains whenActivated—As sphingomyelinase and

cholesterase oxidase affect the IL-7�IL-7R diffusion rate in thepresence of CytD, this suggests that as well as interacting withthe cytoskeleton, the receptor is also interacting with lipid raftsor proteins embedded in these rafts. To investigate receptorchain partition inside and outside lipid rafts upon ligand bind-ing, we extracted DRM by ultracentrifugation in sucrose gradi-ents. We lysed the CD4 T cells with 0.5% Triton X-100,removed by centrifugation unlysed cells and organelles, andseparated the insoluble and soluble membrane fractions byultracentrifugation on a sucrose viscosity gradient according totheir sedimentation velocities. By using flottilin-1 as a fractionmarker of DRMmicrodomains, we demonstrated, in theWest-ern blot shown in Fig. 4, that IL-7R� and the �c chain arelocated in Triton-solubilized fractions 13–17 before IL-7 bind-

FIGURE 1. FCS analysis of IL-7R chain diffusion. Normalized ACF G(�) were plotted from 10 averaged acqui-sitions versus diffusion times �D (log scale in seconds) in panel E for rhodamine-6G (a), SA488 (b), and anti-IL-7R�mAbb�SA488 (c) in colorless completed RPMI medium fitted as the three-dimensional diffusion particle accord-ing to Equation 1. ACF of �c (d) and IL-7R� (e) were labeled with their respective mAbb�SA488, IL-7R��mAbb�SA488 in the presence of unlabeled IL-7 (g) was plotted as mixed two-/three-dimensional diffusionparticles at the surface of CD4 T cells analyzed at 0

2 70 � 103 nm2 and fitted with Equation 4. Normal-ized CCF were also plotted from 10 averaged acquisitions for IL-7R��mAbb�SA488��c�mAbb�SA633 (f) andIL-7R��mAbb�SA488��c�IL-7b�SA633 (h) considered as two-dimensional diffusion particles at the surface ofCD4 T cells. Residuals from fitting with different mathematical diffusion models are given, for example, inpanels A-C for IL-7R��mAbb�SA488 after ACF fitting with models accounting for one two-dimensionalcomponent (A, Equation 3), one two-dimensional and one three-dimensional (B, Equation 4), and twotwo-dimensional and one thre-dimensional (C, Equation 4). Residuals are also given in panel D forIL-7R��mAbb�SA488��c�IL-7b�SA633 after CCF fitting with models accounting for one two-dimensionalcomponent (A, Equation 8). Dashed lines indicates half-height cross-correlation function inflection pointsof single component diffusion. Arrows indicate diffusion times for slow and fast diffusing particles. Six ACFare plotted at increasing 0

2 values for IL-7R��mAbb�SA488 (panel F) and six CCF for IL-7R��mAbb�SA488��c�IL-7b�SA633 (panel G). Corresponding curves in e and h plotted in panel E acquired at 0

2 70 �103 nm2 are shown with arrows. Corresponding �D values extracted from their fitting were used to builddiffusion plots (Equation 2) as shown in Figs. 2 and 3.

IL-7-induced Human CD4 T Cell Response Initiation

14902 JOURNAL OF BIOLOGICAL CHEMISTRY VOLUME 285 • NUMBER 20 • MAY 14, 2010

by guest on September 22, 2020

http://ww

w.jbc.org/

Dow

nloaded from

ing, and in insoluble fractions 6–10 after IL-7 binding. Thisappearance of IL-7R chains in DRM domains suggests that thereceptors are driven into lipid rafts upon IL-7 binding, or thatlipid rafts are formed around the cytokine-bound receptor.To characterize the role played by receptor partition in sig-

naling, we checked whether Jak/STAT phosphorylation wasassociatedwith receptors inside or outsideDRM.To do this, weimmunoprecipitated proteins with anti-IL-7R�: 1) frompooledfractions 6–10 (DRM), or 2) pooled fractions 13–17 (solublefraction of the membrane) of the IL-7-stimulated cell lysatesamples. Protein phosphorylation on the Western blots wasdetected using anti-Tyr(P). Fig. 5 shows that proteinswith IL-7-induced phosphorylation of Tyr were mainly located insideDRM. More specifically, phosphorylated Jak1, Jak3, STAT3,and STAT5 after IL-7 activation were found in DRM, withbarely traces outside.Proteins Immunoprecipitated with IL-7R� before and after

IL-7 Binding—It was clear that some proteins are carried byIL-7R before and after IL-7 binding, and their interactions arepreserved in Triton cell-lysis buffer. We therefore investigated

FIGURE 2. Receptor chain diffusion and assembly at the surface of livingCD4 T lymphocytes. A, IL-7-free IL-7R�, �c, and IL-7R���c diffusion analysisby FCS/FCCS. The following diffusion times �D (in 10�3 s) acquired in theabsence of IL-7 are plotted versus the surface area 2 intercepted by theconfocal volume (in 103 nm2) accordingly to Equation 2: IL-7R��anti-IL-7R��mAbb�SA488 ACF (E), �c�anti-�c�mAbb�SA633 ACF (‚), IL-7R��anti-IL-7R��mAbb�SA488 with �c�anti-�c�mAbb�SA633 CCF (�). Slopes of the linearregression give effective diffusion rates Deff and intercepts at the y axis extrap-olate confinement times �0. B, IL-7-bound IL-7R�, �c, and IL-7R���c diffusionanalysis by FCS�FCCS. The following diffusion times �D (in 10�3 s) in the pres-ence of IL-7b�SA633 are plotted versus the surface area 2 intercepted by theconfocal volume (in 103 nm2): ACF of IL-7R��anti-IL-7R��mAbb�SA488 in theabsence of IL-7 (E), ACF of IL-7R��anti-IL-7R��mAbb�SA488 in the presence ofIL-7 (�), CCF of IL-7R��anti-IL-7R��mAbb�SA488 with IL-7b�SA633 (f). Slopesof the linear regression give Deff and intercepts at the y axis extrapolate �0.Error bars give S.E.

FIGURE 3. Effect of the cytoskeleton and lipid rafts on IL-7R diffusion.A, IL-7R compartmentalization is released after addition of CytD. The follow-ing diffusion times, �D (in 10�3 s), in the presence of IL-7-biotin�streptavidin-A633 are plotted versus the surface area 2 intercepted by the confocal vol-ume (in 103 nm2): ACF of IL-7R��mAbb�SA488 in the absence of IL-7 (E), CCF ofIL-7R��mAbb�SA488 with IL-7b�SA633 without CytD (f), in the presence of 2(gray diamond) and 10 �M CytD (�). Slopes of the linear regression give effec-tive diffusion rates, Deff, and intercepts at the y axis extrapolate confinementtime, �0. B, IL-7R is not affected by the addition of cholesterase oxidase(COase) (1 unit/ml) and sphingomyelinase (Smase) (0.1 unit/ml). The follow-ing diffusion times, �D (in 10�3 s), are plotted versus the surface area 0

2

intercepted by the confocal volume (in 103 nm2): CCF of IL-7R��mAbb�SA488with IL-7b�SA633 before (f) and after treatment with cholesterase oxidase (1unit/ml) (�) and sphingomyelinase (0.1 units/ml) (E) or both (‚). C, IL-7R isfreed by cholesterase oxidase (1 unit/ml) and sphingomyelinase (0.1 unit/ml)only after CytD treatment. The following diffusion times, �D (in 10�3 s), areplotted versus the surface area 0

2 intercepted by the confocal volume (in 103

nm2): ACF of IL-7R��mAbb�SA488 in the absence of IL-7 (E), CCF ofIL-7R��mAbb�SA488 with IL-7b�SA633 in the presence of 10 �M CytD (�), 10�M CytD with cholesterase oxidase (1 unit/ml) and sphingomyelinase (0.1unit/ml) (Œ). Slopes of the linear regression give Deff and intercepts at the yaxis extrapolate �0. Error bars give S.E.

IL-7-induced Human CD4 T Cell Response Initiation

MAY 14, 2010 • VOLUME 285 • NUMBER 20 JOURNAL OF BIOLOGICAL CHEMISTRY 14903

by guest on September 22, 2020

http://ww

w.jbc.org/

Dow

nloaded from

the protein cortege pulled down with the receptor by immuno-precipitation using anti-IL-7R� from lysed CD4 T lympho-cytes. The first step consisted in checking the presence of pro-teins involved in IL-7R signaling that were separated from theIP complex by heat denaturation in SDS andmigration on one-dimensional SDS-PAGE. Protein-specific Western blots areshown in Fig. 6. IL-7R� was detected as a control. �c was foundin cells activated by IL-7 and to a lesser extent in resting cells.This demonstrated that �c interacts spontaneously with

IL-7R� at the surface of resting CD4 T lymphocytes prior tobecoming embedded in lipid rafts. IL-7 stabilizes the IL-7R���cinteraction as suggested by the darker band of pulled down �c.The same amount of Jak1 was carried out before and after IL-7binding, and thiswas consistentwith the constitutive binding ofJak1 onto the IL-7R� cytoplasmic domain. The Jak3 band wasdarker in the activated signaling complexes and its quantitycorrelated to the amount of �c. Jak1 and Jak3 are resident pro-teins on IL-7R chains and some complexes were fully preasso-ciated prior to IL-7 binding. As expected, STAT1, STAT3, andSTAT5 were pulled down only after IL-7 binding as STATrecruitment after IL-7-induced phosphorylation of the IL-7R�cytoplasmic domain on Tyr456 provides a STAT binding site.Proteins ERK1 and ERK2 were also clearly recruited on thesignaling complex after IL-7 binding to IL-7R. As IL-7R inter-action with cytoskeleton was expected, recruitment of actin(microfilament), �-tubulin (microtubule), ezrin and moesin(FERM linking rafted proteins with cytoskeleton) were testedand the presence of all four proteins was confirmed after IL-7activation (Fig. 6).Signaling Complex Analyzed by Two-dimensional PAGE and

MALDI-TOF/TOF—The composition of the signaling complexevidently formed after IL-7 binding was investigated by sepa-rating the IP components on two-dimensional SDS-PAGE andSypro-Ruby staining. Fig. 7 shows a representative pair of two-dimensional PAGEs from non-stimulated (NS) and activated(�IL-7) cells. Eight gel pairs were analyzed from non-stimu-lated and IL-7-stimulated purified CD4 T cells from 8 healthyhuman blood donors. The intensity of 314 spots was measuredand calibrated per gel, then aligned between gels to comparetheir Sypro-staining intensity. In all, 249 reproducible spotsfound in at least 6 of the 8 gels were plugged from two-dimen-

FIGURE 4. IL-7R� and �c are found in DRM after IL-7 stimulation of CD4 Tcells. CD4 T lymphocyte lysates were loaded on a 5– 40% sucrose gradientand divided into 18 fractions after 16 h of centrifugation at 50 krpm at 4 °C.Fractions: 1 left, tube top 5%; 18 right, tube bottom 40%) were loaded onSDS-PAGE (7% acrylamide-bisacrylamide). Flottilin, IL-7R�, and �c werelocated in the membrane fractions by immunoblotting. Fractions corre-sponding to DRM are indicated above the membrane strip according to flot-tilin distribution.

FIGURE 5. Phosphorylated Jaks and STAT5 are found mainly in DRM afterIL-7 stimulation of CD4 T cells. Materials were prepared as described in thelegend to Fig. 4. a, after centrifugation, fractions 6 to 10 were pooled to pro-vide a “DRM” sample and fractions 13 to 17 were pooled to form a “solubi-lized” sample. Both samples were loaded on SDS-PAGE (7% acrylamide-bis-acrylamide). Tyr-phosphorylated proteins Tyr(P) (b and c), pJak (d and e),pSTAT3 (f and g), and pSTAT5 (h and i) were revealed by immunoblotting.

FIGURE 6. Proteins immunoprecipited with IL-7R� before and after IL-7stimulation of CD4 T cells. Proteins were immunoprecipitated with anti-IL-7R� from CD4 T lymphocyte lysate and separated on SDS-PAGE (7% acrylam-ide-bisacrylamide). Corresponding bands were cut out of images of specificimmunoblots from non-stimulated (NS) and IL-7-stimulated (�IL-7) cells.A, receptor chains; B, “resident” protein selection; C, “IL-7-recruited” proteinselection.

IL-7-induced Human CD4 T Cell Response Initiation

14904 JOURNAL OF BIOLOGICAL CHEMISTRY VOLUME 285 • NUMBER 20 • MAY 14, 2010

by guest on September 22, 2020

http://ww

w.jbc.org/

Dow

nloaded from

sional PAGE. Proteinswere digested in trypsin, eluted, and ana-lyzed by combined MS and MS/MS from MALDI-TOF/TOFprocedures. The 109 proteins identified from at least 5sequenced peptides specific to the protein sequence in the Uni-prot data base are detailed in supplemental Table S1. Theywerethen sorted according to the increase in spot staining intensityafter IL-7 binding. 78 proteins were increased by a factor of 4 ormore, whereas 26 proteins were increased by a factor from 2 to3.9. We also noted that four proteins were decreased after IL-7binding.Among the 109 proteins identified, Table 1 lists the 78 pro-

teins recruited by the IL-7R signaling complex after IL-7binding, i.e. that were at least 4-fold more abundant in theimmunoprecipitated complex after IL-7 binding than before.Two-thirds of the pulled down proteins (48/78) are involved inthe cytoskeleton (42/78) or have been described as associatedwith lipid rafts (6/78). None of the proteins in the IL-7R signa-losome and none of those found byWestern blotting (Fig. 6) inthe same IP complex preparation were among the proteinsidentified until the cut off was lowered to select peptides withm/z peaks bellow the top 15most intense. The concentration of

abundant proteins overwhelmed those present at low levels.Interestingly, the cytoskeleton group included actin and itsbinding proteins that regulate filament assembly (gelsolin, cofi-lin, profilin, actin capping proteins, and coactosin) and mem-brane filament anchors or intermediate linkers (vinculin, zyxin,ezrin, and moesin), proteins involved in microfilament poly-merization (HSP70), proteins involved in calcium-inducedrecruitment by the cytoskeleton (calmodulin, calreticulin, andsorcin), and actin-binding linkers and carriers (myosin, tropo-myosin, plastin-2, and derbrin-like). Microtubule subunits (�-and -tubulin) were abundantly recruited upon IL-7 binding.We also foundproteins known for their embedding in lipid rafts(integrin) and their interactions with the cytoskeleton throughvinculin, ezrin, and moesin. Plekstrin, which interacts withphosphatidylinositol 4,5-bisphosphate in lipid rafts was alsofound. Certain recruited proteins are mainly involved in fold-ing/unfolding and degradation processes (HSP, proteasomeactivator, thioredoxine, glutathione transferase, and disulfideisomerase) and oxidation-reduction reactions (peroxiredoxinand superoxide dismutase) and might be carried with thecytoskeleton.Fig. 8 summarizes the IL-7-driven compartmentalization

steps according to our IP analysis before (Fig. 7A) and after (Fig.7B) IL-7 binding. Cytokine-free IL-7R diffuses freely with areduced carriage (Jaks) at the surface of CD4 T lymphocytes,then upon cytokine binding IL-7R is recruited by the cytoskel-eton and interacts with other raft-embedded proteins.

DISCUSSION

IL-7 induces a variety of responses ex vivo in CD4 T lympho-cytes, e.g. cell activation, survival, and proliferation, mainlythrough �c-signaling Jak/STAT, AKT/phosphatidylinositol3-kinase, and mitogen-activated protein kinase pathways. Sig-

FIGURE 7. Two-dimensional PAGE analysis of proteins immunoprecipi-tated with IL-7R� before and after IL-7 stimulation of CD4 T cells. Proteinswere immunoprecipitated with anti-IL-7R� from CD4 T lymphocyte lysateand separated on two-dimensional PAGE (pH 3–10 and 12% acrylamide-bi-sacrylamide). Gels were stained with Sypro-Ruby: A, top, non-stimulated cells(NS); B, bottom, IL-7-stimulated cells (IL-7). pH and molecular weight scales aredisplayed.

TABLE 1Proteins identified by MALDI-TOF/MS/MS at least 4-fold moreabundant among those immunoprecipitated with IL-7R� from CD4 Tcell lysates when cells were stimulated for 5 min with 2 nM IL-7

a CD4 lymphocytes were stimulated or not by IL-7 and their lysates were immuno-precipitated by anti-IL-7R� mAb. Immunoprecipitated proteins were separatedby two-dimensional-PAGE and SYPRO-stained spots were analyzed for theirintensity with SameSpot software and compared between stimulated and non-stimulated conditions over the CD4T cell lysates from eight blood donors. Repro-ducible spots were analyzed by MALDI-TOF/MS/MS. The 78 proteins wereselected according to the ratio � 4 from the full table (109 proteins) provided assupplemental Table S1 and divided into four classes. Numbers of classified pro-teins are bracketed.

IL-7-induced Human CD4 T Cell Response Initiation

MAY 14, 2010 • VOLUME 285 • NUMBER 20 JOURNAL OF BIOLOGICAL CHEMISTRY 14905

by guest on September 22, 2020

http://ww

w.jbc.org/

Dow

nloaded from

naling proteins involved have been mainly identified throughtheir cytokine-induced phosphorylation and have been func-tionally grouped together in the IL-7R “signalosome” (4, 5, 23).However, the early IL-7 response has yet to be described at themolecular level of its receptor and interacting proteins thatform a physical entity and are considered the “signaling com-plex.” The mechanism involved in assembly of the receptoritself is unknown and the protein involved in signaling trans-duction is known for phosphorylated ones but very little isknown concerning the number and function of the “dark” pro-teins involved. New time-resolved microimaging tools havenow enabled us to analyze receptor assembly on living primarycells revealing the story board of the early receptor response fora few entities at a time, and proteomics approaches the sum ofthe composition of these complexes.In this work, we demonstrate for the first time that IL-7R�

and �c spontaneously form heterodimers (1:1 stoichiometry) atthe surface of living human CD4 T cells even in the absence ofIL-7. IL-7R� is highly expressed in the resting CD4 T cells ofhealthy humans (108 pmol/m2), whereas �c (12 pmol/m2) ispresent only at low concentrations and thus restricts het-erodimer formation (10 pmol/m2). The heterodimer was seento have a dissociation constantKd of 19 pmol/m2. The affinity isgreater than the �c chain concentration and explains why 80%of �c is found in the bound state in cultures of living cells. Asalready noted for soluble fragments (21), this very high affinitysuggests that the �c chain is titrated by abundant IL-7R� at thecell surface. We also observed that IL-7R� forms homodimers,as previously suggested (21, 23) and demonstrated for homologIL-2R (26–28).We confirmed that these homodimers assem-ble spontaneously in the absence of IL-7with 1:1 stoichiometry,and noted that the affinity of IL-7R� is 17-fold greater for �cthan for itself, meaning that formation of the heterodimer islargely favored and does not limit �c binding. Both the het-erodimer and the homodimer coexist in resting cells.IL-7 binding to its receptor is followed by receptor complex

compartmentalization. In our studies, the binding of IL-7 to its

receptor almost halted its lateral diffusion and, in accordancewith the criteria established byMarguet and colleagues (25, 29),receptor retention time in the cytoskeleton meshwork wasincreased at least 4-fold after IL-7 binding, suggesting not onlycollision of the receptor complex but also tighter interactionswith cytoskeleton components in living cell cultures at 37 °C.This was first demonstrated by showing that the interactionwas lost when IL-7-activated cells were treated with a cytoskel-eton polymerization inhibitor (cytochalasin D). Our subse-quent analysis by two-dimensional PAGE further supportedthese results as discussed below. IL-7 binding to its receptoralso altered its distribution in detergent-resistant membranenanodomains. This observation suggests that migration wastaking place into lipid rafts or that lipid rafts were being formedaround the receptor.We have therefore for the first time shownthat phosphorylated Jak and STAT5 are associated with raft-embedded receptors. Interestingly, although lipid raft forma-tion was not dependent on the cytoskeleton meshwork, recep-tor dissociation from the lipid rafts was dependent ondepolymerization of actin microfilaments.Our studies of IP complex composition showed that three

categories of proteinswere pulled downwith anti-IL-7R�. First,we found both receptor chains themselves, IL-7R� and �c,moderately associated in the absence of cytokine and then sta-bilized in the presence of cytokine. Second, some proteins werebound to receptor chains independently of the presence ofcytokine, e.g. Jaks. These resident proteins were pulled downbefore and after IL-7 binding. Third, some proteins wererecruited transiently upon cytokine binding and were thenphosphorylated and released, for instance, STAT and ERK.These recruited proteinswere pulled downonly after IL-7 bind-ing. Many proteins involved in IL-7 signaling pathways werenot found in the MS protein list according to the stringentselection criteria used for protein identification: at least 5 of the15 most intense m/z peaks yielded a positive sequence matchwith theMS/MSanalysis.However, the presence of several pro-teins was validated by Western blotting and many were foundfromMS lists using less stringent criteria. To date, among cyto-kine receptors only, the signaling complex associated withmouse IL-1R has been studied by proteomics (30). Interferon,prolactin, growth hormone, and erythropoietin are amongthose best resolved for their assembly at the molecular level,and although their signaling complexes have not yet been elu-cidated, pioneer works on epidermal growth factor receptorand prolactin receptor have suggested that machinery carriedby the cytoskeleton might be involved (31, 32).In our work, we noted that two-thirds of the proteins

recruited upon IL-7 binding have already been described as partof the cytoskeleton or are associated with lipid rafts. Cytoskel-eton-associated proteins identified includedmicrofilament andmicrotubule compounds, proteins regulating their polymeriza-tion and depolymerization, and intermediates such as myosinand tropomyosin.We also found proteins involved in cytoskel-eton connection to the membrane (ezrin, moesin, and vincu-lin). These FERM proteins (for 4.1 protein, ezrin, radixin, andmoesin) regulate the anchorage of plasma membrane proteinsto actin cytoskeleton. These proteins have already beendescribed in the regulation of signal transduction pathways (33,

FIGURE 8. Sketch views of the IL-7R-signaling complex assembly uponIL-7 binding. Left (NS conditions), the IL-7-free heterodimer IL-7R���c isembedded in the lipid bilayer out of rafts; Jak1 and Jak3 are constitutivelybound to their cognate cytoplasmic receptor chains. Right (�IL-7 conditions),the complex IL-7�IL-7R���c is embedded in the lipid raft. STAT is bound toIL-7R���c�Jak1�Jak3. FERM proteins (E) connect IL-7R� to F-actin. Integrinchains are linked to F-actin through praxillin (P) or talin (T) complexed tovinculin (V) and Arp2-3 (42). ABP represents actin-binding proteins, S symbol-izes proteins inhibiting F-actin elongation, and R represents proteins involvedin filament ramification. Tubulin assemblies and proteins unrelated tocytoskeleton are not represented.

IL-7-induced Human CD4 T Cell Response Initiation

14906 JOURNAL OF BIOLOGICAL CHEMISTRY VOLUME 285 • NUMBER 20 • MAY 14, 2010

by guest on September 22, 2020

http://ww

w.jbc.org/

Dow

nloaded from

34). Ezrin and moesin are well expressed in T cells, whereasradixin is low. FERM proteins interact with positively chargedamino acid clusters in the juxtamembrane region of proteinsembedded in lipid rafts. There is a putative FERM-bindingdomain at the juxtamembrane sequence of the cytoplasmicdomain of IL-7R�, characterized by several basic residues(265KKRIKPIVWPSLPDHKKTLEHLCKKPRK292) in the ex-tension of the helical transmembrane domain (240–264) (35)as also found in IL-2R (266NCRNTGPWLKKVLKCNTP-DPSK287) but not in �c. This FERM-binding domain could beinvolved in receptor recruitment of FERM proteins, anchoringthe complex to the cytoskeleton. Full-length FERM proteinsshow a low level of binding activity to both membrane andactin. These inactive states are believed to be expressed by amasking mechanism in which the FERM domain binds theC-terminal half to suppress the actin filament and membranebinding activities (36–39). Biochemical studies have shownthat phosphatidylinositol 4,5-bisphosphate also binds FERMdomains and stimulates the binding of FERM proteins to theirtargets (39, 40). Interestingly, the FERMdomains bind the Rho-specific GDP-dissociation inhibitor (RhoGDI) found amongIL-7R-recruited proteins and accelerate the release of Rho toactivate Rho-dependent processes (39, 41), suggesting involve-ment by the Rho signaling pathway.Integrin �IIb receptor chain is among the most IL-7R-re-

cruited proteins upon IL-7 binding. This receptor chain, whichinteracts with the extracellular matrix (fibronectin and fibrino-gen �), has been described as embedded in lipid rafts (42). Vin-culin, which is also among the highly recruited proteins, hasbeen described as interacting with integrin through the talinintermediate. Talin is very heavy (270 kDa) andwas not selectedfrom our 12% acrylamide SDS-PAGE. Vinculin links the inte-grin receptor-talin complex with contribution of actin-relatedprotein 2–3 (ARP2–3) to actin filaments. The functional roleplayed by talin/vinculin/ARP2–3 in T cells has not beenassessed but might be involved in linking de novo actin poly-merization to integrin activation.Although lipid rafts have been implicated in a number of

cellular functions, their role in lymphocytes has so far beenstudiedmainly in the context of immunoreceptor signaling andis now accepted that lipid rafts are dynamic platforms in T cellswhere proteins implicated in the TCR signaling cascade aretransiently recruited following receptor engagement. Recentdata have also shown that the TCR signaling-initiationmachin-ery is actually preassembled in lipid rafts (42). Protein associa-tion with lipid rafts is likely to be facilitated by raft coalescencefollowing TCR triggering, a process promoted by cortical actinreorganization and enhanced by the engagement of co-stimu-latory receptors such as CD28 (42) and clustered at the immu-nological synapse. No such coalescence is observed during for-mation of the IL-7R signaling complex: its distribution lookedisotrope.This work describes the early steps in IL-7R responses to its

cognate cytokine at the surface of human resting primary CD4T lymphocytes. �c chains are weakly expressed at the cell sur-face, whereas IL-7R� are abundant. Most of the �c chains areassociated with IL-7R� and form high affinity active receptorsthat diffuse laterally out of lipid rafts at the cell surface. When

IL-7 binds to its preformed receptor, this drives it into lipid raftsor the lipid rafts are formed around the receptor. TheIL-7�IL-7R complex then recruits proteins to build the signalingcomplex, e.g. FERM proteins that anchor the rafted receptorand its carriage to the cytoskeleton meshwork and halt its dif-fusion as summarized in Fig. 8. Lipid rafts are rich in cholesteroland sphingomyelin, which increase membrane thickness andhold the long single transmembrane helical domain straight,thus favoring the juxtaposition of the cytoplasmic domains.The H-bond network between cholesterol and sphingomyelinreduces the lateral diffusion of lipids and embedded proteins.We hypothesize that viscous rafts slow down the dissociationprocess between receptor subunits and thereby prolongIL-7�IL-7R association. Cytokine residency time is crucial toamplify the response translated by STAT phosphorylationturnover (43). The cytoskeleton might provide a scaffold foruploading and downloading of signaling proteins, their storage,andmight involve largemachineries to dispatch signaling prod-ucts to their targets.In more general terms, our functional proteomics approach

has led to the discovery of mechanisms linking lipid rafts andthe cytoskeleton with a number of additional functions involv-ing cell membranes and cytoplasmic proteins in T cells. Ourfindings have thrown light on an unexpected dynamic patternof interactions between lipid rafts and not only proteins partic-ipating in the IL-7R signaling cascade, but also a wide array ofproteins associated directly or indirectly with the plasmamem-brane and intracellular membranes, and implicated in a varietyof cell functions. Although the biological significance of theassociation between lipid rafts, the cytoskeleton, and the vari-ous proteins identified requires clarification, this study pro-vides insight into the profound and far reaching changesinduced in a cell by the triggering of surface receptors.

Acknowledgments—We thank Dr. Andres Alcover and Dr. RemiLasserre (Institut Pasteur) for helpful discussion. We thank PascalRoux (Plate-Forme d’Imagerie Dynamique, Institut Pasteur) forexpertise and technical help in confocal microscopy, Dr. Klaus Weis-shart and Dr. Bernhard Gotze (Zeiss) for FCS expert advice, ChristineLaurent (IP) for expert advice in two-dimensional PAGE, and Phil-ippe Bogart (Non-Linear) for assistance in gel analysis. We thank Dr.Mark Jones for text editing (Transcriptum). We extend our gratitudeto the volunteer blood donors and staff at the Etablissement Francaisdu Sang (Centre Cabanel-Paris).

REFERENCES1. Ma, A., Koka, R., and Burkett, P. (2006) Annu. Rev. Immunol. 24, 657– 6792. Fry, T. J., and Mackall, C. L. (2005) J. Immunol. 174, 6571– 65763. Lee, S. K., and Surh, C. D. (2005) Immunol. Rev. 208, 169 –1804. Palmer, M. J., Mahajan, V. S., Trajman, L. C., Irvine, D. J., Lauffenburger,

D. A., and Chen, J. (2008) Cell Mol. Immunol. 5, 79 – 895. Rochman, Y., Spolski, R., and Leonard, W. J. (2009) Nat. Rev. Immunol. 9,

480 – 4906. Goodwin, R. G., Lupton, S., Schmierer, A., Hjerrild, K. J., Jerzy, R., Clev-

enger, W., Gillis, S., Cosman, D., and Namen, A. E. (1989)Proc. Natl. Acad.Sci. U.S.A. 86, 302–306

7. Bolotin, E., Annett, G., Parkman, R., and Weinberg, K. (1999) Bone Mar-row Transplant. 23, 783–788

8. Napolitano, L. A., Grant, R. M., Deeks, S. G., Schmidt, D., De Rosa, S. C.,Herzenberg, L. A., Herndier, B. G., Andersson, J., and McCune, J. M.

IL-7-induced Human CD4 T Cell Response Initiation

MAY 14, 2010 • VOLUME 285 • NUMBER 20 JOURNAL OF BIOLOGICAL CHEMISTRY 14907

by guest on September 22, 2020

http://ww

w.jbc.org/

Dow

nloaded from

(2001) Nat. Med. 7, 73–799. Fry, T. J., and Mackall, C. L. (2001) Trends Immunol. 22, 564 –571

10. Sportes, C., Hakim, F. T., Memon, S. A., Zhang, H., Chua, K. S., Brown,M. R., Fleisher, T. A., Krumlauf, M. C., Babb, R. R., Chow, C. K., Fry, T. J.,Engels, J., Buffet, R., Morre, M., Amato, R. J., Venzon, D. J., Korngold, R.,Pecora, A., Gress, R. E., and Mackall, C. L. (2008) J. Exp. Med. 205,1701–1714

11. Jiang, Q., Li, W. Q., Hofmeister, R. R., Young, H. A., Hodge, D. R., Keller,J. R., Khaled, A. R., and Durum, S. K. (2004)Mol. Cell. Biol. 24, 6501– 6513

12. Khaled, A. R., Li, W. Q., Huang, J., Fry, T. J., Khaled, A. S., Mackall, C. L.,Muegge, K., Young, H. A., and Durum, S. K. (2002) Immunity 17, 561–573

13. Maraskovsky, E., O’Reilly, L. A., Teepe, M., Corcoran, L. M., Peschon, J. J.,and Strasser, A. (1997) Cell 89, 1011–1019

14. Goodwin, R. G., Friend, D., Ziegler, S. F., Jerzy, R., Falk, B. A., Gimpel, S.,Cosman, D., Dower, S. K., March, C. J., Namen, A. E., Park, L. S. (1990)Cell60, 941–951

15. Pandey, A., Ozaki, K., Baumann, H., Levin, S. D., Puel, A., Farr, A. G.,Ziegler, S. F., Leonard, W. J., and Lodish, H. F. (2000) Nat. Immunol. 1,59 – 64

16. Park, L. S., Martin, U., Garka, K., Gliniak, B., Di Santo, J. P., Muller, W.,Largaespada, D. A., Copeland, N. G., Jenkins, N. A., Farr, A. G., Ziegler,S. F., Morrissey, P. J., Paxton, R., and Sims, J. E. (2000) J. Exp. Med. 192,659 – 670

17. Alves, N. L., van Leeuwen, E. M., Derks, I. A., and van Lier, R. A. (2008)J. Immunol. 180, 5201–5210

18. Colle, J. H., Moreau, J. L., Fontanet, A., Lambotte, O., Delfraissy, J. F., andTheze, J. (2007) AIDS 21, 101–103

19. Colle, J. H., Moreau, J. L., Fontanet, A., Lambotte, O., Joussemet, M.,Delfraissy, J. F., and Theze, J. (2006) Clin. Exp. Immunol. 143, 398 – 403

20. Park, J. H., Yu, Q., Erman, B., Appelbaum, J. S., Montoya-Durango, D.,Grimes, H. L., and Singer, A. (2004) Immunity 21, 289 –302

21. Rose, T., Lambotte, O., Pallier, C., Delfraissy, J. F., and Colle, J. H. (2009)J. Immunol. 182, 7389 –7397

22. McElroy, C. A., Dohm, J. A., and Walsh, S. T. (2009) Structure 17, 54 – 6523. Leonard, W. J., Imada, K., Nakajima, H., Puel, A., Soldaini, E., and John, S.

(1999) Cold Spring Harbor Symp. Quant. Biol. 64, 417– 42424. Bacia, K., and Schwille, P. (2007) Nat. Protoc. 2, 2842–285625. Lenne, P. F., Wawrezinieck, L., Conchonaud, F., Wurtz, O., Boned, A.,

Guo, X. J., Rigneault, H., He, H. T., and Marguet, D. (2006) EMBO J. 25,

3245–325626. Eckenberg, R., Rose, T., Moreau, J. L., Weil, R., Gesbert, F., Dubois, S.,

Tello, D., Bossus, M., Gras, H., Tartar, A., Bertoglio, J., Chouaïb, S., Gold-berg, M., Jacques, Y., Alzari, P. M., and Theze, J. (2000) J. Exp. Med. 191,529 –540

27. Rose, T., Moreau, J. L., Eckenberg, R., and Theze, J. (2003) J. Biol. Chem.278, 22868 –22876

28. Pillet, A. H., Juffroy, O., Mazard-Pasquier, V., Moreau, J. L., Gesbert, F.,Chastagner, P., Colle, J. H., Theze, J., and Rose, T. (2008) Eur. CytokineNetw. 19, 49 –59

29. Lasserre, R., Guo, X. J., Conchonaud, F., Hamon, Y., Hawchar, O., Bernard,A. M., Soudja, S. M., Lenne, P. F., Rigneault, H., Olive, D., Bismuth, G.,Nunes, J. A., Payrastre, B., Marguet, D., and He, H. T. (2008) Nat. Chem.Biol. 4, 538 –547

30. Brikos, C., Wait, R., Begum, S., O’Neill, L. A., and Saklatvala, J. (2007) Mol.Cell. Proteomics 6, 1551–1559

31. Lopez-Perez, M., and Salazar, E. P. (2006) Int. J. Biochem. Cell Biol. 38,1716 –1728

32. Zoubiane, G. S., Valentijn, A., Lowe, E. T., Akhtar, N., Bagley, S., Gilmore,A. P., and Streuli, C. H. (2004) J. Cell Sci. 117, 271–280

33. Bretscher, A., Edwards, K., and Fehon, R. G. (2002)Nat. Rev.Mol. Cell Biol.3, 586 –599

34. Fievet, B., Louvard, D., and Arpin, M. (2007) Biochim. Biophys. Acta 1773,653– 660

35. Girault, J. A., Labesse, G., Mornon, J. P., and Callebaut, I. (1999) TrendsBiochem. Sci. 24, 54 –57

36. Andreoli, C., Martin, M., Le Borgne, R., Reggio, H., and Mangeat, P. (1994)J. Cell Sci. 107, 2509 –2521

37. Thuillier, L., Hivroz, C., Fagard, R., Andreoli, C., and Mangeat, P. (1994)Cell. Immunol. 156, 322–331

38. Gary, R., and Bretscher, A. (1995) Mol. Biol. Cell 6, 1061–107539. Hirao, M., Sato, N., Kondo, T., Yonemura, S., Monden, M., Sasaki, T.,

Takai, Y., Tsukita, S., and Tsukita, S. (1996) J. Cell Biol. 135, 37–5140. Niggli, V., Andreoli, C., Roy, C., and Mangeat, P. (1995) FEBS Lett. 376,

172–17641. Takahashi, K., Sasaki, T., Mammoto, A., Takaishi, K., Kameyama, T., Tsu-

kita, S., and Takai, Y. (1997) J. Biol. Chem. 272, 23371–2337542. Chichili, G. R., and Rodgers, W. (2009) Cell Mol. Life Sci. 66, 2319 –232843. Stroud, R. M., and Wells, J. A. (2004) Sci. STKE 2004, re7

IL-7-induced Human CD4 T Cell Response Initiation

14908 JOURNAL OF BIOLOGICAL CHEMISTRY VOLUME 285 • NUMBER 20 • MAY 14, 2010

by guest on September 22, 2020

http://ww

w.jbc.org/

Dow

nloaded from

Lenormand, Jean-Claude Rousselle, Abdelkader Namane and Jacques ThèzeThierry Rose, Anne-Hélène Pillet, Vincent Lavergne, Blanche Tamarit, Pascal

T Lymphocyte ResponseInterleukin-7 Compartmentalizes Its Receptor Signaling Complex to Initiate CD4

doi: 10.1074/jbc.M110.104232 originally published online February 18, 20102010, 285:14898-14908.J. Biol. Chem. 

  10.1074/jbc.M110.104232Access the most updated version of this article at doi:

 Alerts:

  When a correction for this article is posted• 

When this article is cited• 

to choose from all of JBC's e-mail alertsClick here

Supplemental material:

  http://www.jbc.org/content/suppl/2010/02/18/M110.104232.DC1

  http://www.jbc.org/content/285/20/14898.full.html#ref-list-1

This article cites 43 references, 17 of which can be accessed free at

by guest on September 22, 2020

http://ww

w.jbc.org/

Dow

nloaded from

VOLUME 285 (2010) PAGES 14898 –14908DOI 10.1074/jbc.W118.007350

Withdrawal: Interleukin-7 compartmentalizes itsreceptor signaling complex to initiate CD4 T lymphocyteresponse.Thierry Rose, Anne-Helene Pillet, Vincent Lavergne, Blanche Tamarit,Pascal Lenormand, Jean-Claude Rousselle, Abdelkader Namane,and Jacques Theze

This article has been withdrawn by Thierry Rose, Blanche Tamarit,Pascal Lenormand, Abdelkader Namane, and Jacques Theze. Anne-Helene Pillet, Vincent Lavergne, and Jean-Claude Rousselle could notbe reached. Fig. 4 was inappropriately presented. The withdrawingauthors assert that the results of this article are valid.

WITHDRAWALS/RETRACTIONS

1436 J. Biol. Chem. (2019) 294(4) 1436 –1436

© 2019 by The American Society for Biochemistry and Molecular Biology, Inc. Published in the U.S.A.