of September 13, 2016.This information is current as

from Aged MicePeptide-Specific Activation of Naive T Cells Single-Cell Analyses Reveal Two Defects in

Gonzalo G. Garcia and Richard A. Miller

http://www.jimmunol.org/content/166/5/3151doi: 10.4049/jimmunol.166.5.3151

2001; 166:3151-3157; ;J Immunol 

Referenceshttp://www.jimmunol.org/content/166/5/3151.full#ref-list-1

, 12 of which you can access for free at: cites 39 articlesThis article

Subscriptionshttp://jimmunol.org/subscriptions

is online at: The Journal of ImmunologyInformation about subscribing to

Permissionshttp://www.aai.org/ji/copyright.htmlSubmit copyright permission requests at:

Email Alertshttp://jimmunol.org/cgi/alerts/etocReceive 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 © 2001 by The American Association of9650 Rockville Pike, Bethesda, MD 20814-3994.The American Association of Immunologists, Inc.,

is published twice each month byThe Journal of Immunology

by guest on September 13, 2016

http://ww

w.jim

munol.org/

Dow

nloaded from

by guest on September 13, 2016

http://ww

w.jim

munol.org/

Dow

nloaded from

Single-Cell Analyses Reveal Two Defects in Peptide-SpecificActivation of Naive T Cells from Aged Mice1

Gonzalo G. Garcia* and Richard A. Miller 2*†‡

Confocal fluorescent microscopy was used to study redistribution of membrane-associated proteins in naive T cells from young andold mice from a transgenic stock whose T cells express a TCR specific for a peptide derived from pigeon cytochrome C. About 50%of the T cells from young mice that formed conjugates with peptide-pulsed APC were found to form complexes, at the site ofbinding to the APC, containing CD3e, linker for activation of T cells (LAT), and Zap-70 in a central area and c-Cbl, p95vav, Grb-2,PLCg, Fyn, and Lck distributed more uniformly across the interface area. Two-color staining showed that those cells that wereable to relocalize c-Cbl, LAT, CD3e, or PLCg typically relocalized all four of these components of the activation complex. About75% of conjugates that rearranged LAT, c-Cbl, or PLCg also exhibited cytoplasmic NF-AT migration to the T cell nucleus. Aginghad two effects. First, it led to a diminution of ;2-fold in the proportion of T cell/APC conjugates that could relocalize any of thenine tested proteins to the immune synapse. Second, aging diminished by;2-fold the frequency of cytoplasmic NF-AT migrationamong cells that could generate immune synapses containing LAT, c-Cbl, or PLCg. Thus naive CD4 T cells from old mice exhibitat least two separable defects in the earliest stages of activation induced by peptide/MHC complexes.The Journal of Immunology,2001, 166: 3151–3157.

A ctivation of T lymphocytes by APC involves a complexseries of interactions at the area of cell-to-cell contact.The three-dimensional organization of this TCR-APC

interaction was described by Monk et al. using deconvolution im-aging to study fixed T cell/APC conjugates (1, 2), and later byothers using digital time-lapse microscopy to study live T cells(3–5). These studies have led to models in which the area of Tcell/APC contact, the supramolecular activation cluster (SMAC)3

or “immune synapse,” is divided into a central area (c-SMAC)containing the TCR and associated proteins bound to it, and animmediately concentric peripheral area (p-SMAC) containingother key components including LFA-1 and cytoskeletal proteins(2). Engagement of the TCR by agonist peptides or anti-receptorAbs induces a reorganization of the T cell’s cytoskeleton (6) andaccumulation of specific adaptor protein and enzymes (for reviewsee Ref. 7). Localization of the tyrosine kinase Zap-70 within thesecomplexes leads to the phosphorylation of adaptor proteins linkerfor activation of T cells (LAT), Slp-76, and p95vav (8, 9). Theseproteins in turn bring to the TCR complex other adaptor molecules,such as Grb-2 (10), and enzymes such as PLCg and c-Cbl (11, 12)that together trigger later stages of the signal transduction cascade.It has recently been proposed that glycolipid-enriched membrane

microdomains, known as rafts or glycolipid-enriched membranes(GEMS), may play an important role in the translocation of en-zymes and adaptor proteins to the area of TCR-APC contact (13,14). In particular, it has been proposed that the raft domains helpto concentrate the constitutively palmitoylated LAT to thec-SMACs, while excluding other molecules with negative regula-tory roles, such as CD45 (15).

Aging leads to a decline in T cell response to new and previ-ously encountered Ags (16). Several laboratories have shown thatT cells from healthy old humans and mice exhibit a multitude ofdefects at early stages of the T cell signaling pathways, includingchanges in serine/threonine and tyrosine phosphorylation (17–23),development of calcium signals (24), activation of the Raf-1/mi-togen-activated protein/extracellular signal-related kinase kinase/extracellular signal-related kinase (19) and c-Jun N-terminal ki-nase (JNK) pathways (25–26) and translocation of cytoplasmicNF-AT (NF-ATc) to the nucleus (27). Most of these alterations aredemonstrable within the first 15 min of TCR engagement, and it isnot yet clear which of the defects are primary, and which aremerely secondary consequences of earlier abnormalities. One ofthe earliest detectable age-related defects is the decline with age inthe phosphorylation of LAT by Zap-70 (28). In vitro kinase assaysshowed no effects of aging on Zap-70 protein kinase activity inCD4 T cells from young and old mice stimulated by anti-CD3/anti-CD4 cross-linking (29), suggesting that defective LAT phos-phorylation might result from altered accessibility, i.e., differentialcompartmentalization of Zap-70 and its substrates. Indeed, confo-cal microscopic studies using anti-CD3 hybridoma cells as poly-clonal APC analogs showed that the majority of CD4 T cells fromold mice could not efficiently relocate protein kinase C (PKC)u,LAT, and p95vav (Vav) to the immune synapse (28, 30–31), sug-gesting that at least some of the age-associated defects in the earlystages of signal transduction could be the result of defects in theformation of immune synapses at the site of T cell/APC interac-tion. However, one weakness of this initial study was its relianceon anti-CD3 stimulation. Interactions between the 2C11 anti-CD3eand the TCR are of much higher affinity, with a much sloweroff-rate, than the typical interaction between peptide-bearing MHC

*Department of Pathology, University of Michigan School of Medicine,†Universityof Michigan Institute of Gerontology, and‡Ann Arbor Department of Veteran AffairsMedical Center, Ann Arbor, MI 48109

Received for publication September 28, 2000. Accepted for publication December28, 2000.

The costs of publication of this article were defrayed in part by the payment of pagecharges. This article must therefore be hereby markedadvertisementin accordancewith 18 U.S.C. Section 1734 solely to indicate this fact.1 This work was supported by National Institutes of Health Grants AG08808 andAG15942.2 Address correspondence and reprint requests to Dr. Richard A. Miller, 5316 CCGC,1500 East Medical Center Drive, Ann Arbor, MI 48109-0940. E-mail address:millerr@umich.edu3 Abbreviations used in this paper: SMAC, supramolecular activation cluster; LAT,linker for activation of T cells; NF-ATc, cytoplasmic NF-AT; PCC, pigeon cyto-chrome-C; PKC, protein kinase C; Vav, p95vav; c-SMAC, central area; p-SMAC,peripheral area.

Copyright © 2001 by The American Association of Immunologists 0022-1767/01/$02.00

by guest on September 13, 2016

http://ww

w.jim

munol.org/

Dow

nloaded from

molecules and the TCR (32, 33). Furthermore, the original studiesused responder cells that contained both naive and memory T cells;because aging leads to major increases in memory T cell numbersat the expense of naive T cells (for review see Ref. 16), it seemedpossible that differences in synapse formation between naive andmemory CD4 might have contributed to the decline with age in theproportion of cells able to relocalize LAT, Vav, and PKCu to theT cell/APC interface.

To help resolve these issues, we have now used fluorescenceconfocal microcopy to study the responses of T cells from youngand old transgenic mice bearing TCR specific for a peptide frag-ment of pigeon cytochromec (PCC), as presented by the CH12 Bcell line. The results suggest two age-dependent changes in theearly activation process, one that interferes with recruiting of avariety of proteins to the immune synapse, and another, postsyn-aptic defect that prevents NF-AT translocation even in T cells withapparently normal synapse composition.

Materials and MethodsAnimals and cell culture

Breeding pairs of the AND line of TCR-transgenic mice, whose T cellsrespond to PCC, were a generous gift from Dr. Susan Swain (TrudeauInstitute, Saranac Lake, NY). Transgene-positive mice were aged in a spe-cific pathogen-free colony at the University of Michigan (Ann Arbor, MI)and given free access to food and water. Sentinel animals from this colonywere examined quarterly for serological evidence of viral infection; allsuch tests were negative during the course of these studies. Transgenicmice that were found to have splenomegaly or macroscopically visibletumors at the time of sacrifice were not used for experiments. Mice usedwere at 6–8 (young) and 18–20 (old) mo of age. The CH12 B cell line wasobtained from American Type Culture Collection (Manassas, VA) andmaintained in RPMI 1640 with 10% FCS and 2 mML-glutamine at 37°Cand 10% CO2.

Abs and reagents

Rabbit polyclonals anti-PLCg, Vav, c-Cbl, and Grb-2 were purchased fromSanta Cruz Biotechnology (Santa Cruz, CA), the rabbit anti-Zap-70, Lck,Fyn, and LAT from Upstate Biotechnology (Lake Placid, NY) and theanti-CD3e from Dako (Carpinteria, CA). The mAb anti-c-Cbl was pur-chased from Transduction Laboratories (Lexington, KY), and the mAb toNF-ATc1 was purchased from Santa Cruz Biotechnology. Single-color de-tection of proteins for confocal microscopy analysis was performed usingrabbit polyclonal Abs and goat anti-rabbit Ig coupled to FITC (JacksonImmunoResearch, West Grove, PA). Two-color analysis was performedusing first mAb and goat anti-mouse Fcg coupled to FITC (Jackson Im-munoResearch), then the rabbit polyclonal and a goat anti-rabbit coupled toAlexa-594 (Molecular Probes, Eugene, OR).

Peptides were synthesized in the Protein Core Facilities of the Univer-sity of Michigan. The agonist peptide sequence represents aa 88–103 ofPCC (ANERADLIAYLKQATK). The nonagonist peptide had the samesequence, with the exception of a single amino acid substitution (K to N)at position 99 (marked in bold), (34).

Cell preparation

CD41 T cells were obtained from transgenic mice using the negative se-lection methods described in (28). Flow-cytometry analysis of a typicalpreparation showed it to be 90–95% positive for both CD3 and CD4.

For each experiment, CH12 cells in log-phase were pulsed with 20mMagonist or nonagonist peptide for 2 h in fresh medium at 37°C.

Slide preparation and confocal microscopy

A total of 6 3 105 TCR-transgenic CD41 T cells (resuspended at 43 106

cells/ml in RPMI 1640 plus 1% FCS) was combined with 33 105 CH12cells (resuspended at 23 106 cells/ml) to achieve a 2:1 ratio, respectively.Cell mixtures were incubated at 37°C for 20 min and then gently resus-pended and spread onto prewarmed poly(L-lysine)-coated slides (Sigma,St. Louis, MO). Slides were incubated for another 20 min at 37°C, fixedwith 3.7% formaldehyde, 1 mM MgCl2 in PBS (pH 8.5) for 10 min andwashed three times for 5 min each with PBS. Slides were permeabilizedwith 0.2% Triton X-100 in PBS for 10 min at 4°C, washed with PBS andblocked overnight with 1% BSA/PBS. For single-color experiments, theslides were stained with the appropriate primary rabbit Abs diluted in

blocking solution at 1mg/ml for 1 h at 4°C, washed three times with PBSand then stained with goat anti-rabbit-FITC at 10mg/ml in blocking solu-tion for 1 h at 4°C. The slides were washed four times with PBS, mountedusing SlowFade Light antifade reagents (Molecular Probes) and sealedwith nail polish. All slides were coded for blind analysis, and then storedat 4°C protected from light. For two-color protocols, the slides were ini-tially stained with the appropriate mAb diluted in blocking solution at 1mg/ml for 1 h at 4°C, washed, and incubated with goat anti-mouse Fcg-FITC at 10mg/ml in blocking solution for 1 h at 4°C. The second stain wasthen applied using the method given for single stains, but using goat anti-rabbit reagent coupled to Alexa-594 as the second Ab.

Single- and two-color analyses were performed at3100 magnificationon a Nikon Diaphot microscope (Nikon, Melville, NY) equipped with aBio-Rad MRC 600 confocal laser imaging system (Bio-Rad, Hercules,CA). Randomly selected T cell-APC conjugates were analyzed only if thefollowing criteria were first met: 1) tightly formed cell-to-cell contact, 2) Tcells in contact with only one APC, and 3) both cells in the samez-axisplane. At least 100 accepted conjugates on each slide were analyzed andscored as either positive or negative for translocation of the protein to theAPC interface or for accumulation of NF-ATc in the nucleus. All slideswere coded and scored blind, i.e., without knowledge of the age of the Tcell donor. Two slides were examined for each sample (n. 200 totalconjugates), and the mean of the two values was used for further statisticalanalysis.

Statistical analyses

Data in the text and tables represent the means6 SD of three mice fromeach age group, tested as one young and old pair in each set of experiments.Statistical significance was assessed using the paired Student’st test atp 5 0.05.

ResultsPeptide-specific formation of immune synapses by naive CD41

transgenic T cells

Studies by Monks et al. (1, 2) have shown that T cells interactingwith APCs form highly organized protein clusters at the site of cellcontact. The TCR becomes concentrated in the c-SMAC of theinterface, with other signaling proteins associated to this cluster orsurrounding it. In our own investigations we made use of the CH12B cell line, which expresses the MHC-II in the context of I-Ek andI-Ak, and high levels of ICAM-1, B7.1, and B7.2 (data not shown).As our source of T cells we used the AND line of PCC-specifictransgenic mice, in which the majority of CD41 T cells remainnaive throughout life, and express the Va11 and Vb3 specific forPCC (35). In preliminary studies using proliferation as endpoint (I.Dozmorov and R.A. Miller, unpublished observations), we con-firmed previously published data showing that the agonist peptidesequence ANERADLIAYLKQATK, at 2–20mM, triggers strongproliferation of the transgenic T cells, and that a peptide in whichN is substituted for K (shown in boldface) provides a nonstimu-latory control.

Fig. 1A shows a representative set of digital images of CD3elocalization in transgenic T cells conjugated to peptide-pulsedCH12 cells. The T cells in each image are smaller than the CH12cells, and show strong CD3e fluorescence signals. The image atleft shows a conjugate between a T cell and CH12 bearing thenonagonist control peptide; in this as in other negative controls,tight conjugation was confirmed by Nomarski optics (not shown;but see Ref. 28 for examples). In the presence of the control pep-tide, CD3e remains distributed evenly around the outside of the Tcell, as expected, consistent with the lack of immune synapse for-mation. CH12 cells pulsed with agonist peptide generate two typesof tight conjugates. Some conjugates, like the one shown in themiddle of Fig. 1A, show no redistribution of the CD3e-chainwhereas others, such as the one shown at theright of Fig. 1A,exhibit highly concentratede-chain staining in the center of thearea of cell-to-cell contact. These latter, positive responses closelyresemble the images generated previously by other authors using

3152 DEFECTIVE T CELL SIGNALS IN NAIVE CD4 CELLS OF OLD MICE

by guest on September 13, 2016

http://ww

w.jim

munol.org/

Dow

nloaded from

parallel methods (2, 5) and represent the c-SMAC area of the im-mune synapse.

Age-dependent decline in the proportion of naive CD41 T cellsthat can form effective immune synapses after conjugation withpeptide-pulsed APC

Fig. 1Bshows the proportion of conjugates, from three young andthree old mice, that showed strong CD3e relocalization in re-sponses to agonist and nonagonist peptide stimuli. As expected,the nonagonist peptide triggers CD3e relocalization in only a smallproportion (,5%) of conjugates from young or old donors. Theagonist peptide triggerede-chain redistribution in 48% of T cellsfrom young mice, and in 20% of the cells from old mice; thisdifference is significant atp , 0.03. These data suggest that agingdiminishes the proportion of naive T cells that can form effectiveSMACs at the site of interaction with peptide-bearing APC.

The redistribution of multiple proteins to the immune synapsedeclines with age in naive CD41 T cells

In some systems (1, 2, 28), the percentage of CD3e moleculesrelocating to the c-SMAC is much smaller than the fraction ofother TCR-dependent signaling proteins, such as PKCu and LAT,that move to the site of the synapse. Thus the defect in CD3e-chainredistribution in T cells from aged transgenic mice does not ex-clude the possibility that the small numbers of TCR molecules atthe site of APC interaction might still be able to complex withother signaling proteins and form functionally efficient immune

synapses. To examine this problem we performed a series of ex-periments with nonagonist and agonist peptides, staining the con-jugates for proteins known in other models to be associated withthe TCR complex. We selected three groups of proteins for theseanalyses. The first group included PLCg and c-Cbl, enzymesthought to play positive and negative roles, respectively, in theprogress of the TCR signal transduction. The second group in-cluded the tyrosine kinases Lck, Fyn, and Zap-70, which are re-sponsible for phosphorylation of both CD3 molecules and otherproteins in the TCR signal transduction complex. The third groupincluded LAT, Grb-2, and Vav, adapter molecules that help torecruit other elements of the transduction chain. Fig. 2 shows rep-resentative digital images from this series of experiments. Theup-per left panelof Fig. 2 shows a negative control, using nonagonistpeptide stained for distribution of c-Cbl; similar negative controlsusing nonagonist peptide were obtained for all the Abs used (notshown). The other eight panels show examples of conjugates inwhich the indicated target molecules did indeed relocalize to thesynapse in response to agonist peptide-pulsed CH12 cells.

We noted consistent differences in the degree to which the testedmolecules were or were not concentrated in the central area of theT cell/APC interface. Specifically, a high percentage of the con-jugates positive for LAT and Zap-70 rearrangement showed fluo-rescence localized in a small area of cell-to-cell contact, a patternresembling that noted for CD3e (see Fig. 1); these proteins seemlikely to be bound tightly to the TCR/CD3 complex within thec-SMAC. In contrast, distribution of c-Cbl, PLCg, Lck, Fyn, Grb2,and Vav was typically much more uniform along the area of mem-brane contact. These differences in distribution were not alteredappreciably by age (not shown).

A series of three experiments, each using an old and a youngmouse, was conducted to determine the effects of age on the pro-portion of CD41 naive T cells that could rearrange these proteinsto the immune synapse. Table I summarizes these results. Non-agonist peptide controls were included for c-Cbl, Lck, and LAT,and showed that fewer than 10% of conjugates from young or oldmice were able to induce protein redistribution in response to the

FIGURE 1. Age-associated decline in peptide-specific induced translo-cation of CD3e to the immune synapse in naive CD4 T cells. A, CD4 Tcells from TCR-transgenic, PCC-responsive mice were conjugated toCH12 APC cells pulsed with either nonagonist (left) or agonist peptide(middleand right). In each image the T cells are smaller and more fluo-rescent than the APC. Theleft andmiddleimages are scored as negative forCD3e relocalization. The image at the right shows a positive synapse, inwhich thee-chain is relocalized to a small, central area of the cell-to-cellinterface (arrow). B shows means6 SD of three experiments, each per-formed with two mice, one young (Y) and one old (O). Each value wascalculated as the proportion of positive conjugates among at least 200counted, using two slides for each age group and peptide. There were noappreciable age effects on the proportions of CD4 cells able to form con-jugates with APC (not shown). There is a significant difference (p, 0.03)in the proportion of young and old T cells able to respond to the agonistpeptide.

FIGURE 2. Differential localization of signaling proteins in T cell/APCimmune synapses. CD4 T cells from TCR transgenic mice were coincu-bated with CH12 cells pulsed with agonist peptide, fixed and then stainedusing Abs specific for the indicated target proteins; the figure shows rep-resentative confocal digital images. In each case the T cell is the smaller ofthe two cells illustrated. Nomarski optics were used to demonstrate tightconjugation of T cell to APC in each case (not shown). The top image atthe top left (labeled as “negative”) is representative of conjugates in whichprotein migration does not occur; this image is from a c-Cbl-stained slide.All other images show positive synapse responses for the proteins indi-cated, with the arrows showing the area of greatest concentration of fluo-rescent signal.

3153The Journal of Immunology

by guest on September 13, 2016

http://ww

w.jim

munol.org/

Dow

nloaded from

control peptide, in good agreement with the CD3e andproliferation data.

CH12 cells pulsed with agonist peptide induced relocalization ofeach of the eight tested molecules in a higher proportion of youngT cells than of T cells from aged mice. For Lck, Zap-70, Fyn, LAT,Grb2, and Vav, the proportion of responding T cells in young miceranged from 45 to 51%, and in old mice from 17 to 22%; each ofthese is significant by pairedt test atp , 0.05, except for Fyn,which is marginal atp 5 0.06. PLCg rearrangement was noted in62% of T cells from young mice (p , 0.05 compared with the24% of T cells from old donors); the slightly higher results forPLCg may represent a technical artifact, in that the fluorescentsignal was very strong for this reaction. The effect of aging onc-Cbl relocalization is more difficult to assess, in part because thec-Cbl localization to the APC interface was noted in;10% ofconjugates generated using the nonagonist control peptide. Theeffect of age is somewhat smaller than for the other moleculesexamined, and is not statistically significant (p 5 0.15). Takentogether, the data suggest that aging can decrease the translocationof many TCR-associated proteins to the immune synapse, and areconsistent with models in which altered protein relocation contrib-utes to defects in LAT phosphorylation (28) and in later stages ofactivation in T cells from aged donors.

All-or-none redistribution of multiple proteins in the immunesynapses of naive CD4 T cells

The close agreement between the proportions of T cells showingrearrangement of each protein (see Table I) suggests that most Tcells either rearrange all the tested proteins, or none of them. How-ever, single-color experiments cannot formally exclude the hy-pothesis that defects in the ability to relocalize specific couplingmolecules might be differentially distributed among the cells ofyoung or old mice. We tested this possibility by using a double-staining system, in which conjugates are examined using both anti-c-Cbl Ab and Abs to CD3e, LAT, or PLCg. The method uses amouse anti-c-Cbl monoclonal and goat anti-mouse Fcg-FITC toavoid cross-reaction with the IgM expressed on the surface ofCH12 cells. No significant background interference was seen usingthis secondary Ab alone (not shown). The second molecule (LAT,CD3e, or PLCg) was then detected using specific rabbit antiserafollowed by goat anti-rabbit coupled to Alexa-594. Fig. 3 showstypical pairs of digital images showing both fluorescent channelsfor individual conjugates. The examples chosen are ones in whichboth c-Cbl and the other tested molecule are rearranged in the

same cell; these constitute the majority of cells examined (seebelow). The distribution of the proteins within the synapses is sim-ilar to that noted in single-color experiments (see Fig. 2), withCD3e and LAT more centrally located than c-Cbl or PLCg. TableII presents results of these two-color tests, from three mice of eachage, showing the proportion of cells with each staining pattern asa percentage of all cells in which at least one protein was rear-ranged. Double-positive cells made up at least 83% of all stainedcells for each case. Fewer than 15% of the positive cells stained forc-Cbl alone, and fewer than 5% stained positive for CD3e, LAT,or PLCg but not for c-Cbl. There were no statistically significantdifferences for any of these values between the young and oldsamples. These findings are consistent with our previous data look-ing at colocalization of LAT and PKCu in individual CD4 cellsfrom nontransgenic young mice stimulated by an anti-CD3 hybrid-oma cell line (28). The observations show that defects in the abilityto relocalize effector proteins to the immune synapse are not ran-domly distributed among cells, but instead that the redistributionreaction may be “all-or-nothing” at least with respect to the spe-cific proteins we have examined.

Age-dependent defect in nuclear translocation of NF-ATc aftersuccessful synapse formation in CD4 naive T cells

Formation of the immune synapse is followed within minutes byactivation of several downstream pathways whose linkage to theTCR signal is not yet fully elucidated. Many of these early stepsare diminished by aging, including activation of mitogen-activatedprotein kinases of the extracellular signal-related kinase (36) andJNK families (25, 26), and activation of RAF-1 (19), together with

Table I. Single-color analysis of c-Cbl, PLCg, Lck, Zap-70, Fyn, LAT,Grb-2, and Vav localization in naive CD41 cells conjugated with CH12cells bearing either nonagonist or agonist peptide

Protein Stain

Percent of Positive Conjugatesa

Nonagonist Agonist

Young Old Young Old

c-Cbl 106 1 96 1 466 20 306 9PLCg –b – 626 9 246 8*Lck 7 6 3 76 3 506 11 226 2*Zap-70 – – 506 11 196 2*Fyn – – 496 13 196 1LAT 4 6 1 46 1 516 9 186 1*Grb-2 – – 476 12 176 1*Vav – – 456 8 176 3*

a Mean6 SD.b Not determined.p, Value was statistically different (p # 0.05) from that for the same stain in the

young mice.

FIGURE 3. Two-color experiments; representative images. Each pair ofimages shows a single T cell/APC conjugate stained with Abs to c-Cbl(left) as well as with Abs to CD3e, LAT, or PLCg (right). The exampleschosen are of conjugates in which both c-Cbl and the other molecule bothtranslocated to the synapse site.

Table II. Two-color analyses of protein localization to immune synapsein CD4 naive, TCR-transgenic T cells from young and old mice inresponse to agonist peptide

Protein Stain

Percent of Positive Conjugatesa

Young Old

1/1 1/– –/1 1/1 1/– –/1

c-Cbl/LAT 866 5 126 4 26 3 836 6 146 4 36 4c-Cbl/CD3e 916 3 66 2 36 3 886 8 76 5 56 5c-Cbl/PLCg 916 6 96 6 0 916 5 86 6 16 1

a Mean6 SD.1/1, Relocalization of both c-Cbl and the indicated protein (LAT,CD3e–chain, or PLCg). 1/–, Relocalization of c-Cbl only. –/1, Relocalization ofLAT, CD3e, or PLCg, but not c-Cbl. There were no significant differences betweenage groups.

3154 DEFECTIVE T CELL SIGNALS IN NAIVE CD4 CELLS OF OLD MICE

by guest on September 13, 2016

http://ww

w.jim

munol.org/

Dow

nloaded from

impaired activation of transcriptional factors such as NF-ATc (27,28). To see whether defects in the assembly of protein complexesat the site of TCR/APC interaction might contribute to age-depen-dent alterations in these downstream events, we performed a seriesof experiments examining both NF-ATc translocation and relocal-ization of LAT, PLCg, or c-Cbl in individual T cells. Fig. 4 in-cludes a set of digital images derived from these experiments,showing examples of conjugates in which relocalization of LAT,PLCg, or c-Cbl to the synapse either is, or is not, accompanied byNF-ATc migration to the nucleus in the conjugated T cell. Con-jugates positive for NF-ATc translocation but negative for themembrane rearrangement were extremely rare (not shown).

Using this approach we then conducted a series of tests involv-ing three young and three old mice. For each mouse and eachstaining combination we examined at least 100 conjugates thatexhibited a positive response either for NF-ATc, for the synapse-localized protein, or for both. Table III summarizes these results.For naive CD4 cells from young mice,;75% of the cells thatscored positive for the synaptic protein also exhibited NF-ATctranslocation (scored as1/1). Approximately 24% of the respond-ing conjugates showed a relocalization of the synapse componentbut not translocation of NF-ATc (2/1), and fewer than 3%showed translocation of NF-ATc alone (1/2). The data show thatyoung T cells that are able to form synapses involving LAT, c-Cbl,and PLCg typically proceed to NF-ATc migration. In contrast,fewer than 32% of those CD4 naive T cells from old mice thatrelocalize LAT, c-Cbl, or PLCg also exhibit NF-ATc migration tothe nucleus. The majority (;70%) of old T cells that show syn-aptic responses fail to proceed to NF-ATc translocation, with re-arrangement of NF-ATc alone as rare among aged T cells as in theones from younger donors. The proportion of LAT1 cells that alsoscore as NF-ATc-positive is significantly different (p 5 0.03) be-tween young and old mice, and the effects of age in the experi-ments using c-Cbl and PLCg markers are also significant atp , 0.04.

DiscussionReorganization of kinases, substrates, and coupling molecules atthe site of T cell/APC interaction plays an important role in T cellactivation (37). Using the B cell line CH12 as APC and TCR-transgenic mice as the source of naive CD4 T cells, we have stud-ied the effect of aging on formation of the immune synapse. Wenoted that in this system CD3e, Zap-70, and LAT relocalized to thecentral domain of the synapse, similar to the c-SMAC pattern pre-viously documented in other systems (2). The distribution ofPLCg, Grb-2, Vav, c-Cbl, and Fyn resembles that seen for cy-toskeletal proteins, such as talin, which localize to the p-SMAC ofthe synapse (2, 4).

However, not all naive CD4 T cells, freshly isolated from mousespleens, are able to form synapses of this composition after con-jugation to peptide-loaded APC. When the T cells are derived fromyoung donors, only;50% of those forming tight conjugates withAPC translocate any of the listed proteins to the synapse (Table I).These values are consistent with those noted in previous studies ofLAT and PKCu translocation in T cells activated by exposure to

FIGURE 4. Two-color analysis of NF-ATc localization inT cells showing translocation of LAT, PLCg, or c-Cbl; rep-resentative images. Each pair of images shows a single con-jugate, stained for NF-ATc and either LAT, PLCg, or c-Cbl,as indicated. The first pair in each row shows a conjugate inwhich NF-ATc remained in the cytoplasm, and the secondpair in each row shows a conjugate in which NF-ATc movedto the T cell nucleus.

Table III. Two-color analysis of NF-ATc nuclear translocation andLAT, PLCg, and c-Cbl localization in naive CD41 T cells conjugated toCH12 in the presence of agonist peptide

Protein Stain

Percent of Positive Conjugatesa

Young Old

1/1 1/– –/1 1/1 1/– –/1

NF-ATc/LAT 76 6 12 16 1 236 14 316 5 16 1 686 4NF-ATc/c-Cbl 746 17 26 1 246 17 276 3 26 2 716 4NF-ATc/PLCg 746 12 26 1 246 13 316 7 26 2 676 7

a Mean 6 SD. 1/1, Relocalization of both NF-ATc and the indicated protein(LAT, e-chain, or PLCg). 1/–, Relocalization of NF-ATc only. –/1, Relocalizationof the synaptic protein, but not of NF-ATc.

3155The Journal of Immunology

by guest on September 13, 2016

http://ww

w.jim

munol.org/

Dow

nloaded from

hybridoma cells bearing cell surface Ab to mouse CD3e (28, 30,31). In the current study, the proportion of T cells showing trans-location of CD3e was higher than in our previous work (28) and inreports from other groups (2), perhaps because our current systemuses higher concentrations of agonist peptide (20mM, instead of 2mM). The distribution of Lck also deserves comment. In our stud-ies we found Lck to be distributed throughout the area of APC-Tcell contact (presumably within the p-SMAC). Consistent with ourfindings, Krummel et al. (38) found that CD4/Lck complexes, ini-tially present within the central area of the synapse, migratedwithin minutes of synapse formation to more peripheral areas ofthe contact zone.

The one-color data summarized in Table I show that relativelyfew naive CD4 T cells from old mice are able to generate synapsesthat include any of the tested molecules: c-Cbl, PLCg, Lck, Zap-70, Fyn, LAT, Grb-2, and Vav. In each case the proportion ofresponsive naive CD4 cells from aged donors is about half that ofcells from young donors. These results are similar to those seen inour previous studies of unseparated CD4 cells from young and oldnontransgenic donors responding to anti-CD3 stimulation, as mon-itored by translocation of LAT and Vav (28). The new data showthat aging affects T cell responses to peptide Ags in addition toresponses triggered by high affinity anti-receptor Abs, show thatthe changes affect all eight of the tested components of the syn-apse, and show that the aging effect is seen in naive T cells, andthus not due simply to the accumulation of memory CD4 T cells inold age. The two-color experiments of Table II show that cells,from young or old donors, which exhibit defects in any one of thefour proteins tested (c-Cbl, LAT, PLCg, and CD3e) usually showdefects in all of them; in this sense the changes are “all or nothing”at the single-cell level. We have not examined the effects of agingon the time course of translocation of these proteins to the synapsein transgenic T cells, and, therefore, it is possible that differencesin response to peptide-APC conjugates may be slowed, rather thanabsent, in T cells from old donors, even though our previous stud-ies using an anti-CD3 hybridoma cell line as a polyclonal stimu-lator found no evidence for an effect of aging on the time course ofsynapse assembly in CD4 or CD8 cells (31). Although not all ofthe eight molecules summarized in Table I have been examined intwo-color experiments, the excellent agreement among them in theproportions of responsive cells in mice of either age are consistentwith the idea that they, too, show parallel responses, all on, or alloff, within individual cells. Although many of these proteins areeither constitutively associated with high viscosity raft domains inthe T cell membrane, or can become raft-associated by binding toLAT (13, 14), our previous work (28) found no evidence for an ageeffect on distribution of LAT, RAFT-constitutive GM-1, or raft-excluded CD45 in resting or anti-CD3-activated T cells. Thus de-fects in immune synapse formation probably cannot be attributedwith changes in the initial composition of the raft microdomains(28). However, it is possible that further studies of raft-associatedproteins may provide clues to the mechanism of the alterations incomplex formation we have documented. It is also plausible thatage effects on formation of functional immune synapses might bedue to changes in cytoskeleton reorganization during responses toTCR stimulation. In this context it is noteworthy that at least oneof the cytoskeletal proteins, talin, can apparently become relocal-ized to the immune synapse in T cells exposed to nonagonist pep-tides (1, 2), whereas other proteins, such as PKCu, migrate only inresponse to agonist peptides. Biochemical and microscopic tech-niques will help to sort out the ways in which aging might alterassociation of signaling proteins with cytoskeletal proteins beforeand during responses to agonist peptides.

Two-color experiments can also test the linkages betweenevents immediately tied to TCR stimulation, and those down-stream events, such as migration of transcription factors and in-duction of new gene expression, that are triggered by kinase-de-pendent cascades. The data presented in Table III show that T cellsfrom aged mice have, over and above the problems in synapseformation summarized in Table II, a diminished ability to trans-locate NF-ATc to the nucleus. Among young CD4 T cells thatshow LAT migration, for example, 77% proceed to NF-ATc mi-gration, but this figure falls to 31% for cells from aged donors,with very similar changes seen when c-Cbl or PLCg is used as theindex of functional synapse formation. Thus there seem to be threeclasses of CD4 T cells that can be discriminated in responses topeptide-bearing APCs: 1) those that form synapses and induce NF-ATc migration; 2) those that form synapses but do not undergoNF-ATc migration; and 3) those that do neither, with the latter twoclasses increasing as a function of age. T cells from aged donorsalso show defects in activation of the Raf-1 and JNK-dependentprotein kinase pathways (19, 25), the latter of which depends uponCD28-mediated signaling. It is possible that alterations in CD28/JNK signals or in PLCg-independent generation of calcium signals(39) might contribute to derangements in the postsynaptic pro-cesses required for NF-ATc translocation and induction of IL-2gene expression. Further work should now be able to define addi-tional biochemical and functional differences among these threeclasses of cells, and may help to develop a more detailed picture ofthe ways in which T cells discriminate between agonist and non-agonist peptides, and of the changes that impair activation of Tcells in old mice.

References1. Monks, C. R., H. Kupfer, I. Tamir, A. Barlow, and A. Kupfer. 1997. Selective

modulation of protein kinase C-u during T-cell activation.Nature 385:83.2. 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.

3. Shaw, A. S., and M. L. Dustin. 1997. Making the T cell receptor go the distance:a topological view of T cell activation.Immunity 6:361.

4. Grakoui, A., S. K. Bromley, C. Sumen, M. M. Davis, A. S. Shaw, P. M. Allen,and M. L. Dustin. 1999. The immunological synapse: a molecular machine con-trolling T cell activation.Science 285:221.

5. Wulfing, C., Y. H. Chien, and M. M. Davis. 1999. Visualizing lymphocyte rec-ognition. Immunol. Cell Biol. 77:186.

6. Penninger, J. M., and G. R. Crabtree. 1999. The actin cytoskeleton and lympho-cyte activation.Cell 96:9.

7. Myung, P. S., N. J. Boerthe, and G. A. Koretzky. 2000. Adapter proteins inlymphocyte antigen-receptor signaling.Curr. Opin. Immunol. 12:256.

8. Salojin, K. V., J. Zhang, and T. L. Delovitch. 1999. TCR and CD28 are coupledvia ZAP-70 to the activation of the Vav/Rac-1-/PAK-1/p38 MAPK signalingpathway.J. Immunol. 163:844.

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

10. Schneider, H., Y. C. Cai, K. V. Prasad, S. E. Shoelson, and C. E. Rudd. 1995. Tcell antigen CD28 binds to the GRB-2/SOS complex, regulators of p21ras. Eur.J. Immunol. 25:1044.

11. Finco, T. S., T. Kadlecek, W. Zhang, L. E. Samelson, and A. Weiss. 1998. LATis required for TCR-mediated activation of PLCg1 and the Ras pathway.Im-munity 9:617.

12. Lupher, M. L. J., N. Rao, M. J. Eck, and H. Band. 1999. The Cbl protoonco-protein: a negative regulator of immune receptor signal transduction.Immunol.Today 20:375.

13. Montixi, C., C. Langlet, A. M. Bernard, J. Thimonier, C. Dubois, M. A. Wurbel,J. P. Chauvin, M. Pierres, and H. T. He. 1998. Engagement of T cell receptortriggers its recruitment to low-density detergent-insoluble membrane domains.EMBO J. 17:5334.

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

15. Rodgers, W., and J. K. Rose. 1996. Exclusion of CD45 inhibits activity of p56lckassociated with glycolipid-enriched membrane domains.J. Cell Biol. 135:1515.

16. Miller, R. A. 1996. The aging immune system: primer and prospectus.Science273:70.

17. Garcia, G. G., and R. A. Miller. 1997. Differential tyrosine phosphorylation ofzchain dimers in mouse CD4 T lymphocytes: effect of age.Cell Immunol. 175:51.

3156 DEFECTIVE T CELL SIGNALS IN NAIVE CD4 CELLS OF OLD MICE

by guest on September 13, 2016

http://ww

w.jim

munol.org/

Dow

nloaded from

18. Ghosh, J., and R. A. Miller. 1995. Rapid tyrosine phosphorylation of Grb2 andShc in T cells exposed to anti-CD3, anti-CD4, and anti-CD45 stimuli: differentialeffects of aging.Mech. Ageing Dev. 80:171.

19. Kirk, C. J., and R. A. Miller. 1998. Analysis of Raf-1 activation in response toTCR activation and costimulation in murine T-lymphocytes: effect of age.Cell.Immunol. 190:33.

20. Miller, R. A., G. Garcia, C. J. Kirk, and J. M. Witkowski. 1997. Early activationdefects in T lymphocytes from aged mice.Immunol. Rev. 160:79.

21. Patel, H. R., and R. A. Miller. 1992. Age-associated changes in mitogen-inducedprotein phosphorylation in murine T lymphocytes.Eur. J. Immunol. 22:253.

22. Shi, J., and R. A. Miller. 1993. Differential tyrosine-specific protein phosphory-lation in mouse T lymphocyte subsets: effect of age.J. Immunol. 151:730.

23. Grossmann, A., P. S. Rabinovitch, T. J. Kavanagh, J. C. Jinneman, L. K. Gilli-land, J. A. Ledbetter, and S. B. Kanner. 1995. Activation of murine T-cells viaphospholipase-Cg1-associated protein tyrosine phosphorylation is reduced withaging.J. Gerontol. A. Biol. Sci. Med. Sci. 50:B205.

24. Miller, R. A. 1996. Calcium signals in T lymphocytes from old mice.Life Sci.59:469.

25. Kirk, C. J., A. M. Freilich, and R. A. Miller. 1999. Age-related decline in acti-vation of JNK by TCR- and CD28-mediated signals in murine T-lymphocytes.Cell Immunol. 197:75.

26. Kirk, C. J., and R. A. Miller. 1999. Age-sensitive and -insensitive pathwaysleading to JNK activation in mouse CD41 T-cells.Cell Immunol. 197:83.

27. Whisler, R. L., L. Beiqing, and M. Chen. 1996. Age-related decreases in IL-2production by human T cells are associated with impaired activation of nucleartranscriptional factors AP-1 and NF-AT.Cell Immunol. 169:185.

28. Tamir, A., M. D. Eisenbraun, G. G. Garcia, and R. A. Miller. 2000. Age-depen-dent alterations in the assembly of signal transduction complexes at the site of Tcell/APC interaction.J. Immunol. 165:1243.

29. Garcia, G. G., and R. A. Miller. 1998. Increased zap-70 association with CD3zin CD4 T cells from old mice.Cell Immunol. 190:91.

30. Eisenbraun, M. D., A. Tamir, and R. A. Miller. 2000. Altered composition of theimmunological synapse in an anergic, age-dependent memory T cell subset.J. Immunol. 164::6105.

31. Yang, D., and R. A. Miller. 1999. Cluster formation by protein kinase Cu duringmurine T cell activation: effect of age.Cell Immunol. 195:28.

32. 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.

33. Valitutti, S., and A. Lanzavecchia. 1997. Serial triggering of TCRs: a basis for thesensitivity and specificity of antigen recognition.Immunol. Today 18:299.

34. Reay, P. A., R. M. Kantor, and M. M. Davis. 1994. Use of global amino acidreplacements to define the requirements for MHC binding and T cell recognitionof moth cytochrome c (93–103).J. Immunol. 152:3946.

35. Linton, P. J., L. Haynes, N. R. Klinman, and S. L. Swain. 1996. Antigen-inde-pendent changes in naive CD4 T cells with aging.J. Exp. Med. 184:1891.

36. Gorgas, G., E. R. Butch, K. L. Guan, and R. A. Miller. 1997. Diminished acti-vation of the MAP kinase pathway in CD3-stimulated T lymphocytes from oldmice.Mech. Ageing Dev. 94:71.

37. Viola, A., and A. Lanzavecchia. 1999. T-cell activation and the dynamic worldof rafts.APMIS 107:615.

38. Krummel, M. F., M. D. Sjaastad, C. Wulfing, and M. M. Davis. 2000. Differentialclustering of CD4 and CD3z during T cell recognition.Science 289:1349.

39. Miller, R. A. 1991. Accumulation of hyporesponsive, calcium extruding memoryT cells as a key feature of age-dependent immune dysfunction.Clin. Immunol.Immunopathol. 58:305.

3157The Journal of Immunology

by guest on September 13, 2016

http://ww

w.jim

munol.org/

Dow

nloaded from