Age-Dependent Defects in TCR-Triggered
Cytoskeletal Rearrangement in CD4 + T Cells
Gonzalo G. Garcia and Richard A. Miller
This information is current as
of June 17, 2017.
Subscription
Permissions
Email Alerts
This article cites 33 articles, 17 of which you can access for free at:
http://www.jimmunol.org/content/169/9/5021.full#ref-list-1
Information about subscribing to The Journal of Immunology is online at:
http://jimmunol.org/subscription
Submit copyright permission requests at:
http://www.aai.org/About/Publications/JI/copyright.html
Receive free email-alerts when new articles cite this article. Sign up at:
http://jimmunol.org/alerts
The Journal of Immunology is published twice each month by
The American Association of Immunologists, Inc.,
1451 Rockville Pike, Suite 650, Rockville, MD 20852
Copyright © 2002 by The American Association of
Immunologists All rights reserved.
Print ISSN: 0022-1767 Online ISSN: 1550-6606.
Downloaded from http://www.jimmunol.org/ by guest on June 17, 2017
References
J Immunol 2002; 169:5021-5027; ;
doi: 10.4049/jimmunol.169.9.5021
http://www.jimmunol.org/content/169/9/5021
The Journal of Immunology
Age-Dependent Defects in TCR-Triggered Cytoskeletal
Rearrangement in CD4ⴙ T Cells1
Gonzalo G. Garcia* and Richard A. Miller2*†‡
he kinetic segregation model of T cell activation postulates that the initial contact of a T cell with an APC involves recognition of LFA-1 and CD28 on the T cell surface by ICAM-1 and B-7.1 on the APC (for a review, see Ref. 1).
This initial interaction creates the necessary proximity for the TCR
to interact with peptides presented by the MHC. These first steps
lead to a series of cytoskeletal reorganizations, including development of lamellopodia and molecular segregation with displacement of large glycoproteins, including CD43 and CD45, from the
central area of TCR/MHC contact (2, 3). The process culminates in
the formation of immunosynapses, also referred to as supramolecular activation clusters (SMAC).3 Detailed models of these processes are just starting to emerge, models in which TCR interaction
with actin (4) and LFA-1 interaction with talin (5) have important
roles. When the TCR encounters agonist or even antagonist peptides, LFA-1 and talin are translocated to the T cell/APC contact
area, perhaps contributing to the stability necessary for the TCR to
recognize the different peptides presented by the MHC. Encounters
with antagonist peptides do not yield an efficient signal, and the T
cell and APC become dissociated. In contrast, encounter with agonist peptides leads to signals sufficiently strong or long lasting to
trigger SMAC formation (6). During the initial T cell-APC interaction leading to the formation of lamellopodia, there is a localized
elevation of intracellular Ca2⫹ and active reorganization of the
cytoskeleton, with polymerization and depolymerization of actin
T
(7). The details of this process are not yet well characterized, and
little is known as to whether other focal adhesion proteins, such as
paxillin and vinculin, play key roles in immunosynapse formation.
CD4⫹ T cells from aged mice show defects in TCR signal transduction that include deficits in serine/threonine and tyrosine phosphorylation, diminished activation of Raf-1, c-Jun N-terminal kinase, and extracellular signal-regulated kinase, and diminished
elevation of intracellular Ca⫹, among other problems (for comprehensive review, see Ref. 8). In contrast, our experiments (9)
have shown no effect of aging on the level of activation of the
TCR-associated tyrosine kinase Zap-70 after CD3-CD4 stimulation in freshly isolated mouse CD4⫹ T cells. The contrast between
the age-insensitive activation of Zap-70 and the age sensitivity of
multiple downstream events led us to postulate that age-related
defects in TCR-mediated activation signals are related to a lack of
accessibility of the TCR-associated kinases to their substrates. Experiments using two-color confocal microscopy (10 –13) supported
this hypothesis by showing age-dependent defects in migration of
multiple substrates, including the linker for activation of T cells
(LAT), Vav, Grb2, and phospholipase C (PLC)-␥, to the site of T
cell/APC contact in peptide-specific and CD3-dependent conjugates. In this study, we describe a series of experiments to explore
the hypothesis that age-dependent alterations in T cell cytokeletal
organization and lamellopodia formation might contribute to the
decline with age in the proportion of T cells that can form functional immune synapses when conjugated to stimulatory APC.
*Department of Pathology, School of Medicine, and †Institute of Gerontology, University of Michigan, Ann Arbor, MI 48109; and ‡Ann Arbor Department of Veterans’
Affairs Medical Center, Ann Arbor, MI 48109
Materials and Methods
Received for publication June 11, 2002. Accepted for publication September 5, 2002.
The costs of publication of this article were defrayed in part by the payment of page
charges. This article must therefore be hereby marked advertisement in accordance
with 18 U.S.C. Section 1734 solely to indicate this fact.
1
This work was supported by National Institutes of Health Grants AG19619 and Core
Grants AG13283 and AG08808.
2
Address correspondence and reprint requests to Dr. Richard A. Miller, 5316 CCGC,
1500 E. Medical Center Drive, Ann Arbor, MI 48109-0940. E-mail address:
[email protected]
3
Abbreviations used in this paper: SMAC, supramolecular activation cluster; LAT,
linker for activation of T cells; PCC, pigeon cytochrome c; PLC, phospholipase C.
Copyright © 2002 by The American Association of Immunologists, Inc.
Animals and cell culture
Breeding pairs of the AND line of TCR transgenic mice, whose T cells
respond to pigeon cytochrome c (PCC), were a generous gift from S. Swain
(Trudeau Institute, Saranac Lake, NY). Transgene-positive mice were aged
in a specific pathogen-free colony at the University of Michigan. Specific
pathogen-free male (BALB/c ⫻ C57BL/6)F1 (CB6F1) mice were purchased from the National Aging Institute contract colonies at the Charles
River Laboratories (Kingston, NJ) and at Harlan (Indianapolis, IN). Mice
were given free access to food and water. Sentinel animals were examined
quarterly for serological evidence of viral infection; all such tests were
negative during the course of these studies. Mice that were found to have
splenomegaly or macroscopically visible tumors at the time of sacrifice
were not used for experiments. Mice used were at 6 – 8 (young) or 18 –20
0022-1767/02/$02.00
Downloaded from http://www.jimmunol.org/ by guest on June 17, 2017
Previous research has shown that many of the CD4 T cells from older mice do not form functional immune synapses after
conjugation with peptide-pulsed APC. We now show that the defect lies at a very early stage in the cytoskeletal reorganization that
precedes movement of protein kinases and their substrates to the TCR/APC interface. Antagonist peptides presented to T cells
from young mice induce migration of talin (but not paxillin, vinculin, or F-actin) to the APC contact zone, but CD4 T cells from
older donors typically fail to show the talin polarization response. A spreading assay in which contact with anti-CD3-coated slides
induces CD4 T cells to assume a conical shape and develop lammelopodia also shows a decline with age in the proportion of T cells
that can initiate cytoskeletal changes in response to this simplified stimulus. Finally, the transition from detergent-soluble to
cytoskeletal forms of the p16, p21, and p23 isoforms of CD3␨ in response to CD3/CD4/CD28 cross-linking is much stronger in
young than in old T cells. Thus, defects in cytoskeletal reorganization triggered by initial contact between TCR and peptidebearing APC precede, and presumably contribute to, defective activation of protein kinase-mediated signals in the first few
minutes of the activation cascade in T cells from aged mice. The Journal of Immunology, 2002, 169: 5021–5027.
5022
EFFECT OF AGE ON CD4 T CELL CYTOSKELETON REARRANGEMENT
(old) mo of age. The CH-12 B cell line (expressing the I-Ak MHC-II and
high levels of ICAM and B-7.1) was obtained from American Type Culture
Collection (Manassas, VA) and maintained in RPMI with 10% FCS and 2
mM L-glutamine at 37°C and 10% CO2.
Abs and reagents
Cell preparation
CD4⫹ T cells were obtained and stimulated from transgenic or CB6F1 mice
using the negative selection methods previously described (12, 17). Flow
cytometry analysis of a typical preparation showed it to be 90 –95% positive for both CD3 and CD4.
TCR-APC conjugation, spreading assay, and confocal
microscopy
CH-12 cells pulsed with 20 ␮g/ml of agonist or antagonist peptide were
conjugated with CD4⫹ T cells, as described (11). For the two-color analyses, we first used specific rabbit or mouse Abs to cytokeletal proteins (at
1 ␮g/ml in PBS containing 0.2% BSA), or in some cases Alexa-488 phalloidin (at 2 U/ml in PBS/BSA), followed by biotin-GST-Grb2 (0.5 ␮g/ml
in PBS/BSA). The slides were developed by incubation for 1 h with goat
anti-rabbit or anti-mouse Fc␥ Abs coupled to Alexa-488 (except for phalloidin) and then streptavidin Alexa-594 (Molecular Probes), all at 1 ␮g/ml
in PBS/BSA.
For the spreading assays, polylysine slides were incubated with antiDNP (clone UC-8), anti-CD28 (clone 37.51), or anti-CD3 (clone 1452C11) in PBS for 2 h at 37°C using the concentrations indicated in the text.
The slides were then washed several times with PBS and prewarmed at
37°C in PBS before use. CD4⫹ T cells at 2 ⫻ 106 cells/ml in RPMI 1640
were then gently deposited on the Ab-coated slide, and incubated in a
humidified chamber at 37°C for the times indicated in the text. The slides
were fixed, permeabilized with 1% saponin in PBS for 10 min at 4°C, and
stained with Alexa-488 phalloidin. After extensive washing with PBS, the
slides are mounted, as described (11).
The two-color analyses were performed at ⫻100 magnification on a
Nikon Diaphot microscope equipped with a Bio-Rad MRC 600 confocal
laser imaging system (Bio-Rad, Hercules, CA), as described (10). For the
spreading assay, digital images for at least 100 cells/slide were recorded.
The digital images were converted to TIFF files, and the area of individual
cells was measured using Scion Image software (www.scioncorp.com). All
slides were coded and scored blind, i.e., without knowledge of the age of
the T cell donor.
Cytoskeleton purification, immunoprecipitation, and Western
blots
Approximately 10 ⫻ 106 CD4⫹ T cells were stimulated for 5 min by
cross-linking CD3⑀ with CD4 and CD28, as described previously (17). The
cells were lysed in 1 ml of 1% Brij-58, PBS, pH 7.4, 10 ␮g/ml aprotinin,
100 ␮g/ml leupeptin, 10 mM PMSF, and 0.1 mM sodium orthovanadate for
30 min on ice. The lysates were ultracentrifuged at 100,000 ⫻ g for 1 h at
4°C. The pellet (cytoskeleton fraction) was washed once with lysis buffer,
centrifuged at 100,000 ⫻ g for 10 min, resuspended in 50 ␮l of SDS
sample buffer, and boiled for 15 min. The CD3␨ contained in the supernatant (soluble fraction) was immunoprecipitated overnight with rabbit antiCD3␨ chain precoupled to protein G-Sepharose. The beads were washed
Statistical analyses
Unless otherwise indicated, results are presented as means ⫾ SD of at least
six young and three old subjects tested in at least three separate experiments. Statistical significance was assessed using a Mann-Whitney test at
p ⫽ 0.05.
Results
Peptide-specific formation of immune synapses by naive CD4⫹
transgenic T cells, as detected by biotin-GST-Grb2 fusion
protein
We have previously used transgenic mice, whose naive CD4⫹ T
cells are specific for a peptide derived from PCC, to show ageassociated decline in T cell immunosynapse formation (11). We
have now extended these studies to document the usefulness of
biotin-GST-Grb2 as a single cell marker for immunosynapse formation. We chose biotin-Grb2 for this purpose based on previous
demonstrations that it interacts with specific domains of Slp-76
(18), Cbl, and LAT (19), proteins that are themselves components
of the SMAC. The use of biotin-GST-Grb2 also provides a technical advantage as a second color stain for confocal microscopy, in
that it avoids the problems of cross-reactivity between secondary
Abs used to localize other intracellular molecules of interest. Fig.
1A shows digital images of T cells, conjugated with peptide-bearing CH-12 cells and then stained for GST-Grb2 as well as for
PLC-␥, Lck, or the cytoskeletal proteins paxillin, vinculin, and
FIGURE 1. Single cell analysis of the differential localization of signaling proteins in T cell/APC immunosynapses using two-color confocal
microscopy. A, CD4⫹ T cells from TCR transgenic mice were incubated
with CH-12 cells pulsed with agonist peptide, fixed, and stained for the
indicated target proteins as well as for biotin-GST-Grb2; each pair of images shows a single T cell/APC conjugate. In each case, the T cell is the
smaller of the two cells illustrated. Nomarski imaging (not shown) was
used to demonstrate tight conjugation of T cell to APC. The top image (a
double-negative cell) is representative of conjugates in which protein migration does not occur. All other images show positive synapse responses
for the proteins indicated, with the arrows showing the area of greatest
concentration of fluorescent signal. B, Talin and GST-Grb2 colocalization
in conjugates of transgenic T cells to APCs pulsed with antagonist peptide.
The top pair of images shows a double-negative conjugate in which the T
cell has relocated neither talin nor the proteins that interact with biotinGST-Grb2. The middle pair (⫹/⫺) shows an example of translocation of
talin without the translocation of the GST-Grb2-binding proteins. The
lower pair of images shows a T cell in which both talin and GST-Grb2
have moved to the immunosynapse area (⫹/⫹).
Downloaded from http://www.jimmunol.org/ by guest on June 17, 2017
Rabbit polyclonals anti-PLC-␥, c-Cbl, and CD3␨ (clone 6B10) were purchased from Santa Cruz Biotechnology (Santa Cruz, CA). Rabbit anti-Lck
was purchased from Upstate Biotechnology (Lake Placid, NY). Anti-talin
(clone 8, day 4) and anti-vinculin (clone hVIN-1) were purchased from
Sigma-Aldrich (St. Louis, MO). Alexa-488 phalloidin was purchased from
Molecular Probes (Eugene, OR). Anti-paxillin was purchased from Transduction Labs (Lexington, KY). The GST-Grb2 construct was a donation of
G. Koretzky (University of Pennsylvania, Philadelphia, PA). The GSTband III construct was a gift from A. Weiss (Howard Hughes Medical
Institute, San Francisco, CA). A construct encoding the cytoplasmic tail of
CD3␨ was a gift of M. Exley (Beth Israel Deaconess Medical Center,
Boston, MA). The GST-Grb2, GST-band III, and GST-CD3␨ fusion proteins were purified (14) and biotinylated in our lab, as described (15).
Rabbit anti-CD3␨ (recognizing all phosphorylated and nonphosphorylated
forms of CD3␨) was developed in our lab using the full cytoplasmic domain of ␨-chain.
Peptides were synthesized in the Protein Core Facilities of the University of Michigan. The agonist peptide sequence represents aa 88 –103 of
PCC (ANERADLIAYLKQATK). The antagonist peptide had the same
sequence, with the exception of a single amino acid substitution (K to N)
at position 99, marked in bold (16).
and then boiled for 5 min with 50 ␮l of SDS sample buffer. Aliquots from
cytoskeleton and the CD3␨ soluble fraction were resolved in 12% polyacrylamide gel, transferred to polyvinylidene difluoride, and incubated
with anti-CD3␨ mouse monoclonal (clone 6B10); the bands were visualized by chemifluorescence and quantified, as previously described (9).
The Journal of Immunology
5023
nearly all experiments and are therefore not shown in Table I.
Antagonist peptide, as expected, induced formation of synapses
containing both talin and GST-Grb2 in only a small percentage
(3–11%) of the conjugates, consistent with previous reports that
antagonist peptides do not induce the formation of functional synapses containing Grb2-binding proteins. When the antagonist peptide was used, 51% of the CD4⫹ T cells from young mice showed
talin redistribution without GST-Grb2 translocation, consistent
with models in which antagonist peptides induce early, but not late
stages in immunosynapse formation. In CD4⫹ T cells from old
mice, however, only 12% of the conjugates showed talin redistribution by itself. This age effect is statistically significant at p ⬍
0.001, and shows that T cells from old mice have a defect at a very
early stage of the interaction with APCs, i.e., before the discrimination by the TCR between agonist and antagonist peptides,
which leads to formation of the SMAC. Table I also shows that
neither paxillin, vinculin, nor F-actin moved to the area of T cellAPC contact in responses to antagonist peptides, even in CD4⫹ T
cells from young mice. The results suggest that the translocation of
these focal adhesion proteins and detectable polymerization of actin are, unlike talin translocation, dependent on SMAC formation,
or on processes triggered by agonist peptides that lead to SMAC
formation.
Studies using agonist peptide, also shown in Table I, demonstrated that during immunosynapse formation, 61–75% of the conjugates from young donors show translocation of paxillin and vinculin and polymerization of actin in the T cell-APC contact area.
In the same experiments, only 15% of the T cells from young mice
are single positive for talin, and fewer than 2% are single positive
for F-actin or for the other three focal adhesion proteins; thus,
nearly all of the young CD4 cells that are able to relocalize the
focal adhesion proteins to the synapse proceed to the stage of Grb2
relocalization as well. Consistent with our previous studies of immunosynapse formation (11), there is a far smaller proportion of
CD4⫹ T cells from aged transgenic mice that can form immunosynapses in response to agonist-pulsed APC, ranging from 19 to
25%, depending on the specific cytoskeletal protein examined
along with GST-Grb2. These age-related declines in cytoskeleton
protein relocalization are statistically significant at p ⬍ 0.05. These
results show that, with the exception of talin, translocation of cytoskeletal proteins to the T cell-APC contact area requires agonist
peptide. Thus, the age-related defects in T cell activation include
changes at early stages of TCR-APC interaction, changes that precede recruitment of other cytoskeletal elements, as well as proteins
that bind to Grb2, to the APC interaction site.
Age-related decline in lamellopodia formation by CD4⫹ T cells
T cells form lamellopodia, a flattening of the contact area with the
APC with a partial reorganization of the cytoskeleton, before the
formation of a complete functional immune synapse (21–23). To
Table I. Two-color analysis of talin, F-actin, paxillin, and vinculin localization to the immunosynapse in naive CD4⫹ cells conjugated with CH-12
cells bearing either antagonist or agonist peptide
Antagonista
Young
Agonista
Old
Young
Old
Protein Stain
⫺/⫺
⫹/⫺
⫹/⫹
⫺/⫺
⫹/⫺
⫹/⫹
⫺/⫺
⫹/⫺
⫹/⫹
⫺/⫺
⫹/⫺
⫹/⫹
Talin/Grb-2
F-actin/Grb-2
Paxillin/Grb-2
Vinculin/Grb-2
38 ⫾ 7
82 ⫾ 6
87 ⫾ 9
96 ⫾ 1
51 ⫾ 5
0
0
0
11 ⫾ 4
18 ⫾ 5
10 ⫾ 7
3⫾2
75 ⫾ 10
87 ⫾ 6
91 ⫾ 8
94 ⫾ 1
12 ⫾ 4
0
0
0
3⫾2
12 ⫾ 4
7⫾5
4⫾1
24 ⫾ 3
25 ⫾ 6
36 ⫾ 4
36 ⫾ 5
15 ⫾ 3
0
0
0
61 ⫾ 5
75 ⫾ 6
64 ⫾ 4
63 ⫾ 5
67 ⫾ 9
81 ⫾ 10
80 ⫾ 4
74 ⫾ 5
9⫾4
0
0
0
23 ⫾ 5
19 ⫾ 10
19 ⫾ 2
25 ⫾ 5
a
Mean ⫾ SD. ⫹/⫹, Relocalization of both Grb-2 and the indicated cytoskeleton protein. ⫹/⫺, Relocalization of cytoskeleton proteins, but not Grb-2.
Downloaded from http://www.jimmunol.org/ by guest on June 17, 2017
F-actin. The top panel shows a T cell/CH-12 conjugate in which
neither Grb2 nor PLC-␥ has moved to the immunosynapse; double-negative cells of this kind are seen, at varying proportions as
discussed below, for all pairs of stains used. The next two rows of
Fig. 1A show double-positive immunosynapses in which PLC-␥
(or Lck) and biotin-GST-Grb2 become localized to the area of T
cell-APC contact. We found very few single-positive cells, i.e.,
cells in which the synapse area contained either Grb2 alone or
PLC-␥ (or Lck) alone. Results using Abs to LAT or c-Cbl together
with GST-Grb2 were similar to those for PLC-␥ and Lck (not
shown). For controls of the GST-Grb2 specificity, we use a biotinGST-band III fusion protein (for a description of this construct, see
Ref. 9). Under our experimental conditions, we find that the GSTband III has no detectable binding of any protein from the CD4⫹
T cell. These results indicate that the biotin-GST-Grb2 construct
can be used to identify CD4⫹ T cells that have formed immunosynapses containing PLC-␥, Lck, LAT, and c-Cbl. During immunosynapse formation, there is a reorganization of the cytoskeleton
characterized by polymerization of actin into F-actin filaments and
accumulation of actin-related protein 2–3 (Arp2–3) at the immunosynapse (20). We therefore also examined translocation of paxillin, vinculin, and F-actin, in relation to GST-Grb2 localization.
We found that all three of these molecules can move to the GSTGrb2-containing immunosynapses; results with F-actin and paxillin are shown in the bottom rows of Fig. 1A.
Talin, another cytoskeleton protein, differs from PLC-␥, Lck,
F-actin, vinculin, and paxillin in that it has been reported to move
to the area of APC-T cell interaction even when antagonist peptides are used in place of agonists for stimulation (6). This translocation requires TCR/MHC interactions, but does not result in the
formation of a SMAC (6). To confirm these results in our own
transgenic system, we studied redistribution of talin using CH-12
cells pulsed with an antagonist peptide. Fig. 1B shows typical results using Abs to talin along with GST-Grb2. We noted doublenegative conjugates in which neither protein had moved to the
synapse area (Fig. 1B, top), others in which talin, but not GSTGrb2 had been redistributed (Fig. 1B, middle), and still others (Fig.
1B, bottom) in which both molecules had translocated to the immunosynapse. Conjugates in which only GST-Grb2 had moved to
the synapse without talin redistribution were very rare (not shown).
Using this methodology, we then quantified the percentage of
transgenic CD4⫹ T cells from young and old mice that can produce F-actin fibrils and can redistribute focal adhesion proteins to
the immunosynapses using GST-Grb2 as our indicator for synapse
formation. Table I shows mean values for double-negative (⫺/⫺),
single-positive (i.e., positive for the cytoskeletal proteins, but not
for Grb2) (⫹/⫺), and double-positive (⫹/⫹) cells from three experiments involving six young and three old mice. Conjugates in
which only GST-Grb2 had relocated, without relocation of the
cytoskeletal element, represented fewer than 1% of the total in
5024
EFFECT OF AGE ON CD4 T CELL CYTOSKELETON REARRANGEMENT
FIGURE 2. Spreading assay using CD4⫹ T cells. CD4⫹ T cells from
young and old CB6F1 mice were incubated for 10 min on slides covered
with anti-DNP, anti-CD28, or anti-CD3. The cells were fixed, stained with
phalloidin, and analyzed by confocal microscopy. A, Shows a montage of
images through successive Z planes from single T cells adherent to slides
coated either with anti-CD28 (top) or anti-CD3 (bottom). The anti-CD3binding cell shows induction of spreading, an overall conical shape, and a
ring of F-actin at the periphery of the T cell in the sections in which the cell
is in contact with the slide. B, Shows a wider field at the Z plane level at
which the T cells, from young or old donors, are in contact with slides
coated with anti-DNP, anti-CD28, or anti-CD3. The arrows in the lower
right panel of B indicate two types of cells from old mice, one (upper
arrow) resembling young cells stimulated by anti-CD3, and the other (lower arrow) resembling old cells exposed to the control Abs anti-CD28 and
anti-DNP. C, Shows typical frequency distributions for T cells of different
size classes (from 1 to 6 U of area), in which the measurement reflects the
area of the cell in its contact zone (Z plane) with the coated slide, normalized by setting the mean area of one of the two young CD28-exposed CD4
cells per experiment to 1 arbitrary area unit. 䡺, Young cells, CD28; f,
young cells, CD3; E, old cells, CD28; F, old cells, CD3.
CD4⫹ T cells from young mice generate contact zones that are
round and relatively small when the slides are coated with antiDNP or anti-CD28. Contact zones of CD4⫹ T cells from old mice
are similarly round and homogeneous, but for unknown reasons
are typically somewhat larger than those produced by T cells from
young donors. Incubation on anti-CD3-coated slides induces in
young CD4 cells a larger contact zone characterized by a bright
ring and an inner zone of diminished phalloidin-staining intensity.
CD4⫹ T cells from old mice, incubated on anti-CD3 slides, fall
into two classes. Some cells show with no detectable spreading
(similar to images seen on anti-CD28), but others closely resemble
T cells from young mice on anti-CD3 (see arrows in Fig. 2B for
contrasting examples).
We used an image analysis program to estimate the proportion
of CD4⫹ T cells from young and old mice that respond in this
anti-CD3-spreading assay. The procedure measured the area of the
contact zone for at least 100 cells of each type in each experiment;
the raw data were normalized so that the mean area of responses to
anti-CD28 was set to 1.0 U for the first young mouse tested within
each experiment. Fig. 2C shows a representative result, comparing
cells from one young and one old CB6F1 mouse. T cells from the
old mice incubated on anti-CD28 slides (E) have slightly larger
contact zones compared with T cells from young donors (䡺), i.e.,
1.8 arbitrary units (SD ⫽ 0.8). When T cells from young mice are
incubated with anti-CD3 (f), the mean area increases from 1 to
3.2 ⫾ 0.7. The mean value for old T cells on anti-CD3 slides
increases only slightly, i.e., from 1.8 to 2.2 ⫾ 0.9 U (F). The
distribution of cell areas for the anti-CD3-treated old CD4⫹ cells
is much broader than that seen for young CD4⫹ cells, and there is
substantial overlap with the distribution of areas seen in the antiCD28 control condition. We interpreted this distribution of old
CD4 cells as a composite of two overlapping populations, one that
responds slightly, if at all, to anti-CD3 contact, and a second that
can spread nearly as well as T cells from young mice.
We also performed this spreading assay using CD4⫹ T cells
from the transgenic mice. The anti-CD3 induced a 3.5 ⫾ 0.2
(mean ⫾ SD)-fold induction of spreading in T cells from the
young mice, but only a 2.1 ⫾ 0.2-fold induction in T cells from the
old mice. We also noted a bimodal distribution, including some
responsive and some nonresponsive cells, in T cells from the aged
transgenic mice, similar to that seen in experiments with CB6F1
mice (data not shown).
It has been reported that PMA-induced increases in the mobility
of membrane LFA-1 reflect partial cytoskeleton reorganization and
release of LFA-1 from the cytoskeleton matrix as a result of protein kinase C activation or increases in membrane fluidity (26). To
test the hypothesis that age-related defects in cytoskeleton reorganization can be overridden by pharmacological agents, we treated
T cells from young and old with 5 nM of PMA for 10 min and then
performed the spreading assay, as described above. The results of
three separate experiments showed no significant differences in the
frequency distribution between young and old after PMA treatment (data not shown). The result suggests that cytoskeleton rearrangement in the old CD4⫹ T can be triggered by CD3 in the
presence of low doses of PMA.
Table II shows the results of a more comprehensive analysis,
using two concentrations of anti-CD28 or anti-CD3 (20 and 2 ␮g/
ml) and two different incubation times (10 and 20 min) for CB6F1
mice, and using a single dose and time for cells from transgenic
mice. Because before stimulation the cells from the old CB6F1
mice are larger than those of young donors, for these mice we used
two criteria for a positive response, one set at a level exceeded by
only 5% in the first of the two young mice used on CD28 slides in
Downloaded from http://www.jimmunol.org/ by guest on June 17, 2017
test whether formation of lamellopodia is affected by aging, we
developed an assay to study the spreading of CD4⫹ T cells, purified from CB6F1 mice, on slides covered with anti-CD3 Ab; slides
covered with anti-CD28 or anti-DNP are used as negative controls.
The upper panel of Fig. 2A shows an example of a young CD4⫹
T cell incubated for 10 min on an anti-CD28-coated slide. The
figure compiles images of the different Z planes of a single cell,
stained with phalloidin to show F-actin. In this cell, F-actin is
distributed uniformly around the cell periphery from the top of the
cell (left part of the series) to the area in which the cell is in contact
with the anti-CD28 on the surface of the glass slide. Identical
results are found using anti-DNP-coated slides (not shown). In
contrast, cells incubated on slides covered with anti-CD3 typically
spread and acquire a conical shape, as illustrated in the bottom
panel of Fig. 2A. The Z plane of contact with the anti-CD3 (the
rightmost Z plane section in the series) shows high levels of polymerized actin localized as a ring at the edge of the cell, with a
central area containing lower levels of F-actin. The T cell shape
and F-actin redistribution and cytoskeleton reorganization are similar to alterations previously seen in T cells conjugated to APC (24,
25), although the SMAC is not formed in such slide-adherence
cultures (24). Fig. 2B shows typical digital images of CD4⫹ cells
from young or old mice obtained from the contact zone using
anti-DNP-, anti-CD28-, or anti-CD3-coated slides. As expected,
The Journal of Immunology
5025
Table II. Induction of spreading of CD4⫹ T cells on Ab-coated glass slides
a
b
c
Incubation
Time (min)
Animal
Model
10
10
20
20
10
10
20
20
10
10
F1
F1
F1
F1
F1
F1
F1
F1
Tg
Tg
Anti-CD28a
Anti-CD3a
Age
Ab
Concentration
(␮g/ml)
Young cutoff (%)b
Old cutoff (%)c
Young cutoff (%)b
Old cutoff (%)c
Young
Old
Young
Old
Young
Old
Young
Old
Young
Old
20
20
20
20
2
2
2
2
20
20
7⫾2
42 ⫾ 14
6⫾3
38 ⫾ 24
7⫾6
44 ⫾ 23
5⫾4
24 ⫾ 10
5
6⫾5
0
5
0
5
0
5
0
5
5
5
98 ⫾ 2
76 ⫾ 6
94 ⫾ 5
70 ⫾ 7
90 ⫾ 4
70 ⫾ 18
78 ⫾ 14
52 ⫾ 7
97 ⫾ 1
52 ⫾ 8
51 ⫾ 10
29 ⫾ 6
47 ⫾ 14
25 ⫾ 7
31 ⫾ 5
17 ⫾ 7
35 ⫾ 10
19 ⫾ 4
96 ⫾ 4
49 ⫾ 1
Each value shows the mean ⫾ SD obtained using either CB6F1 (F1; six young and three old) or transgenic (Tg, three young and three old) mice tested over three experiments.
Indicates the percentage of cells with an area ⬎95% of the cells from the first-tested young mouse on CD28-coated control slides in the same experiment.
Indicates the percentage of cells with an area ⬎95% of the cells from the old mouse using CD28-coated control slides in the same experiment.
decrease the amount of TCR translocated to the immunosynapse.
To test this hypothesis, we stimulated CD4⫹ T cells purified from
spleen of CB6F1 mice, by cross-linking the TCR with CD4 and
CD28, and then separated the proteins associated with the cytoskeleton matrix (cytoskeletal fraction) from the proteins in the
detergent-soluble fraction, which includes proteins from both
membrane and cytosolic compartments. The purity of the cytoskeletal fraction was evaluated by Western blots showing that Raf-1
and ribosomal S6 protein kinase were found in the soluble fraction,
but not in the cytoskeleton preparation (data not shown). Fig. 3
shows CD3␨ Western blots for the cytoskeleton and soluble fractions purified from young and old CD4⫹ T cells. Three predominant ␨ forms can be detected; p16␨ and p21␨, both in the cytoskeleton and soluble fractions, and p23␨, detected only in the
cytoskeleton fractions. These results are similar to previous observations using T cell lines (4). Fig. 3 illustrates effects of donor age
and of stimulation, which we quantitated and show in Table III as
mean values from five individual mice of each age.
Even before stimulation, there is a difference between young
and old CD4⫹ cells in the distribution of CD3␨ molecules: a shift
from the p16 form to the p21 form in the soluble fraction. CD4⫹
T cells from old mice have 2-fold lower levels of p16␨ ( p ⫽ 0.02)
and almost 2-fold higher levels of p21␨ in the soluble fraction
( p ⫽ 0.04), compared with T cells from young donors.
Activation induces changes in the distribution of all three forms
of CD3␨ in T cells of young mice. After 5 min of activation, T cells
from young mice show increases in the amount of p16␨, p21␨, and
Age-related decline in translocation of CD3␨ from the plasma
membrane to cytoskeleton
FIGURE 3. Anti-CD3␨ Western blots showing CD3␨ forms associated
with the cytoskeleton or the detergent-soluble fraction of CD4⫹ T cells
from young and old mice. CD4⫹ T cells from young and old mice were
stimulated (⫹) or not stimulated (⫺) by anti-CD3 cross-linked with CD4
and CD28. The cytoskeleton and soluble fractions were purified, and the
CD3␨ forms of the soluble fraction were immunoprecipitated and compared with cytoskeleton-associated proteins using anti-CD3 Western blots.
The arrows indicate the three major CD3␨ forms present.
The lack of talin redistribution and lamellopodia formation in
CD4⫹ T cells from the old mice could be the result of age-related
changes in the interactions between cytoskeleton and the TCR
complex. Age-related changes in CD3␨ association with the cytoskeleton could limit the initial cytoskeleton reorganization and
Downloaded from http://www.jimmunol.org/ by guest on June 17, 2017
each experiment, and the other, higher, criterion set at a level exceeded by only 5% of the old cells on CD28 slides in the same
experiment. As suggested by Fig. 2, CD4⫹ T cells from the old
CB6F1 mice were indeed slightly larger than young cells on control slides with anti-CD28; for the 10-min incubation at high Ab
dose, for example, 42% of the old cells exceeded the cutoff area
value, compared with only 7% of the cells from young mice. A
10-min incubation on anti-CD3-coated slides induced 98% of the
young cells to spread out to an area above the criterion (7% background), but only 76% of the old cells (despite their higher, 42%
background value). On the control slides coated with anti-CD28,
there was no difference in spreading between young and old transgenic mice (with 5% of cells about threshold for both genotypes),
consistent with the idea that the difference in the size of unstimulated CB6F1 cells might reflect the age-dependent increase in
memory T cells, which is much less dramatic in transgenic mice.
A total of 96 –97% of the cells from young transgenic mice spread
onto anti-CD3-coated slides, compared with only 49 –52% of cells
from old transgenic mice, showing that the age-dependent decline
in spreading responses is not limited to CB6F1 animals.
Qualitatively similar results were seen at each Ab concentration
and each time point using the CB6F1 animal model, and regardless
of whether the arbitrary threshold was set based on young or old
CD28-binding cells. When the criterion for response was based
instead on the 95% cutoff for unstimulated old T cells, no more
than 30% of the old CD4⫹ cells can be scored as responsive. The
age-related declines are in all cases statistically significant (at p ⬍
0.05) when compared with young for the same time of stimulation.
Taken together, the data in Table II show that 70 –98% of the
CD4⫹ T cells from young mice can produce lamellopodia and
spread out onto anti-CD3-coated slides, but that only 17–30% of
the T cells from old mice can trigger a detectable membrane flattening when compared with the responses of old cells to control
slides coated with anti-CD28 in the same experiments. The lack of
T cell response cannot be attributed to the shift between the naive
to memory phenotype, because the T cells from old transgenic
mice remain in naive state (11) and show similar decline in the
spreading assay.
5026
EFFECT OF AGE ON CD4 T CELL CYTOSKELETON REARRANGEMENT
Table III. Distribution of CD3␨ forms between cytoskeletal and detergent-soluble fractions in CD4 T cells from young and old mice
Cytoskeleton Fractiona
Young
CD3␨
p16
p21
p23
Soluble Fraction
Old
Young
Old
⫺
⫹
⫺
⫹
⫺
⫹
⫺
⫹
1 ⫾ 0.15
1 ⫾ 0.16
0.04 ⫾ 0.01
2.5 ⫾ 0.45
3.3 ⫾ 0.77
0.2 ⫾ 0.04
0.8 ⫾ 0.2
1.2 ⫾ 0.35
0.1 ⫾ 0.04
0.8 ⫾ 0.16
1.7 ⫾ 0.19
0.13 ⫾ 0.03
2.3 ⫾ 0.3
2.8 ⫾ 0.12
ND
1 ⫾ 0.14
2.1 ⫾ 0.11
ND
0.9 ⫾ 0.13
5.1 ⫾ 0.69
ND
0.7 ⫾ 0.12
5 ⫾ 0.34
ND
a
The data are means ⫾ SEM using five young and five old mice tested in two experiments. Raw densitometry values were normalized so that the mean level for p21␨ in
the cytoskeletal fraction of resting young T cells was set at 1.0 in each experiment.
Discussion
The segregation model of T cell activation proposes that the initial
T cell-APC contact takes place through the interaction of accessory molecules such as LFA-1 and CD28, followed by displacement from the contact area of large glycosylated proteins, such as
CD45 and CD43, that could interfere with TCR-MHC interaction
(for review, see Ref. 1). In the next phase, TCR recognition of
agonist peptide stabilizes the T cell-APC interaction and initiates a
process that leads to the formation of the SMAC. According to this
model, MHC molecules bearing antagonist peptides can during
this phase induce the translocation of focal adhesion molecules
such as talin to the T cell-APC contact area, but not induce formation of functional immunosynapses containing protein kinase
C-␪ and other signal-coupling elements. Our current results are
consistent with this model, in that both antagonist and agonist peptides induced the translocation of talin to the T cell-APC contact
area (see Fig. 1B), while only agonist peptide can trigger the translocation of several TCR-associated proteins and formation of the
SMAC (11). To study the relation between the immunosynapse
and proteins belonging to the focal adhesion family, we have analyzed the localization of paxillin and vinculin with respect to
SMAC formation. We selected paxillin and vinculin because these
have been reported to associate, respectively, with talin and LFA-1
(for a review, see Ref. 27). In addition, paxillin can associate to Src
homology 2 domains of the Src family of tyrosine kinases (27). As
shown in Table I, agonist peptides can trigger the translocation of
these proteins to the immunosynapse in 60 –75% of the young
naive CD4⫹ T cells conjugated to agonist-bearing CH-12 APC.
The results suggest that these focal adhesion molecules play a role
in immunosynapse formation, although further studies will be
needed to determine whether blocking of their translocation abrogates the formation of SMACs. When we analyzed naive CD4⫹ T
cells from old mice, only ⬃20% of the cells show this pattern of
translocation of cytoskeletal proteins to the APC contact area, consistent with our previous work (11) showing age-related defects in
movement of many kinases and substrates to the SMAC in aged T
cells.
Our observations of talin translocation suggest that age-related
changes in reorganization of the cytoskeleton may be a cause,
rather than merely a consequence, of defects in SMAC formation.
When redistribution was analyzed using an antagonist peptide,
62% of young CD4 cells, but only 15% of old CD4 cells, were able
to relocate talin to the area of APC contact. Thus, the lack of talin
translocation in the naive CD4⫹ T cells from old mice suggests
age-related defects in the initial TCR-APC interactions before recruitment of other elements of the SMAC.
The kinetic segregation model of TCR signal transduction suggests several possibilities that could explain defects early in the
interaction between TCR and APC, including not only alterations
in cytoskeleton reorganization, but also potentially changes in
TCR-MHC-II interaction caused by negative regulators such as
CD43 (28) or other large glycoproteins (29), or a decline in the
calcium signal in the initial T cell-APC contact area. To evaluate
these ideas, we used a simplified system in which the APC surface
was replaced by glass slides coated with stimulatory (anti-CD3) or
nonstimulatory (anti-CD28 or anti-DNP) Abs, and in which T cell
cytoskeletal changes were monitored by evaluating lamellopodia
formation. This system has been shown previously to mimic, in
part, the initial TCR-APC interaction by inducing a TCR-dependent cytoskeleton reorganization, localized calcium flux, and flattening of the T cell membrane without formation of immunosynapses (24). Using this system (Fig. 2 and Table II), we noted that
anti-CD3 induces spreading of CD4⫹ T cells, which acquire a
conical shape that resembles T cells adhering to authentic APC
(see Fig. 1 for examples), and triggers a limited reorganization of
the cytoskeleton including an F-actin ring localized in the area of
the T cell periphery that is in contact with the slide (Fig. 2). AntiCD3 induces this spreading response in 70 –90% of the CD4⫹ T
cells from young mice, but only 17–31% of the CD4⫹ T cells from
old CB6F1 mice. The response of transgenic T cells in the spreading system also declines by ⬃50% in old age.
We hypothesized that this age-related change in cytoskeletal
responsiveness might reflect underlying changes in interactions between proteins of the TCR/CD3 complex and proteins responsible
for cytoskeletal structure. Previous work has shown that different
CD3␨ forms are associated with different fractions of the T cell
membrane (4). Some, including p23␨, are anchored to the cytoskeletal matrix, while other forms of CD3␨ remain relatively free of
cytoskeleton interaction in resting cells. After TCR engagement,
however, all isoforms of CD3␨ are translocated from the soluble
Downloaded from http://www.jimmunol.org/ by guest on June 17, 2017
p23␨ associated with the cytoskeleton, and corresponding declines
in the levels of p16␨ and p21␨ in the soluble fraction; each of these
changes is statistically significant at p ⬍ 0.05. These translocations
from the soluble fraction to the cytoskeleton were, however, not
detected after activation of CD4⫹ T cells from old mice. These
results show that aging inhibits the translocation of CD3␨ from the
detergent-soluble fraction to the cytoskeletal matrix. Such a
change might constrain the movement of TCR complexes to the
immunosynapse by interfering with the ability of TCR complexes
to flow within plasma membrane domains to the site of APC
contact.
Two experiments of this kind were conducted using purified
CD4⫹ T cells from four young and two old transgenic mice (data
not shown). The CD4⫹ T cells from the young transgenic mice
showed a 2.5-fold induction in p16 translocation to the cytoskeleton (mean 1.0 ⫾ 0.2 SD in resting cells, compared with 2.5 ⫾ 0.5
after CD3 stimulation), with a corresponding decline in p16 content in the soluble fraction (2.1 ⫾ 0.3 in resting cells compared
with 1.1 ⫾ 0.3 after CD3 stimulation). In contrast, there were no
significant changes in distribution of p16 in CD4⫹ T cells from the
old transgenic donors.
The Journal of Immunology
References
1. Van der Merwe, A., S. J. Davis, A. S. Shaw, and M. L. Dustin. 2000. Cytoskeletal
polarization and redistribution of cell-surface molecules during T cell antigen
recognition. Semin. Immunol. 12:5.
2. Sperling, A. I., J. R. Sedy, N. Manjunath, A. Kupfer, B. Ardman, and
J. K. Burkhardt. 1998. TCR signaling induces selective exclusion of CD43 from
the T cell-antigen-presenting cell contact site. J. Immunol. 161:6459.
3. Janes, P. W., S. C. Ley, A. I. Magee, and P. S. Kabouridis. 2000. The role of lipid
rafts in T cell antigen receptor (TCR) signalling. Semin. Immunol. 12:23.
4. Caplan, S., and M. Baniyash. 1996. Normal T cells express two T cell antigen
receptor populations, one of which is linked to the cytoskeleton via ␨ chain and
displays a unique activation-dependent phosphorylation pattern. J. Biol. Chem.
271:20705.
5. Sampath, R., P. J. Gallagher, and F. M. Pavalko. 1998. Cytoskeletal interactions
with the leukocyte integrin ␤2 cytoplasmic tail: activation-dependent regulation
of associations with talin and ␣-actinin. J. Biol. Chem. 273:33588.
6. Monks, C. R., B. A. Freiberg, H. Kupfer, N. Sciaky, and A. Kupfer. 1998. Threedimensional segregation of supramolecular activation clusters in T cells. Nature
395:82.
7. Serrador, J. M., M. Nieto, and F. Sanchez-Madrid. 1999. Cytoskeletal rearrangement during migration and activation of T lymphocytes. Trends Cell Biol. 9:228.
8. Miller, R. A. 2000. Effect of aging on T lymphocyte activation. Vaccine 18:1654.
9. Garcia, G. G., and R. A. Miller. 1998. Increased Zap-70 association with CD3␨
in CD4 T cells from old mice. Cell. Immunol. 190:91.
10. Eisenbraun, M. D., A. Tamir, and R. A. Miller. 2000. Altered composition of the
immunological synapse in an anergic, age-dependent memory T cell subset.
J. Immunol. 164:6105.
11. Garcia, G. G., and R. A. Miller. 2001. Single-cell analyses reveal two defects in
peptide-specific activation of naive T cells from aged mice. J. Immunol. 166:
3151.
12. Tamir, A., M. D. Eisenbraun, G. G. Garcia, and R. A. Miller. 2000. Age-dependent alterations in the assembly of signal transduction complexes at the site of T
cell/APC interaction. J. Immunol. 165:1243.
13. Yang, D., and R. A. Miller. 1999. Cluster formation by protein kinase C␪ during
murine T cell activation: effect of age. Cell. Immunol. 195:28.
14. Smith, D. B., and K. S. Johnson. 1988. Single-step purification of polypeptides
expressed in Escherichia coli as fusions with glutathione S-transferase. Gene
67:31.
15. Hnatowich, D. J., F. Virzi, and M. Rusckowski. 1987. Investigations of avidin
and biotin for imaging applications. J. Nucl. Med. 28:1294.
16. Reay, P. A., R. M. Kantor, and M. M. Davis. 1994. Use of global amino acid
replacements to define the requirements for MHC binding and T cell recognition
of moth cytochrome c (93–103). J. Immunol. 152:3946.
17. Garcia, G. G., and R. A. Miller. 1997. Differential tyrosine phosphorylation of ␨
chain dimers in mouse CD4 T lymphocytes: effect of age. Cell. Immunol. 175:51.
18. Motto, D. G., S. E. Ross, J. Wu, L. R. Hendricks-Taylor, and G. A. Koretzky.
1996. Implication of the GRB2-associated phosphoprotein SLP-76 in T cell receptor-mediated interleukin 2 production. J. Exp. Med. 183:1937.
19. Motto, D. G., S. E. Ross, J. K. Jackman, Q. Sun, A. L. Olson, P. R. Findell, and
G. A. Koretzky. 1994. In vivo association of Grb2 with pp116, a substrate of the
T cell antigen receptor-activated protein tyrosine kinase. J. Biol. Chem. 269:
21608.
20. Dustin, M. L., and J. A. Cooper. 2000. The immunological synapse and the actin
cytoskeleton: molecular hardware for T cell signaling. Nat. Immun. 1:23.
21. Stowers, L., D. Yelon, L. J. Berg, and J. Chant. 1995. Regulation of the polarization of T cells toward antigen-presenting cells by Ras-related GTPase CDC42.
Proc. Natl. Acad. Sci. USA 92:5027.
22. Rabinowitz, J. D., C. Beeson, C. Wulfing, K. Tate, P. M. Allen, M. M. Davis, and
H. M. McConnell. 1996. Altered T cell receptor ligands trigger a subset of early
T cell signals. Immunity 5:125.
23. Wulfing, C., M. D. Sjaastad, and M. M. Davis. 1998. Visualizing the dynamics
of T cell activation: intracellular adhesion molecule 1 migrates rapidly to the T
cell/B cell interface and acts to sustain calcium levels. Proc. Natl. Acad. Sci. USA
95:6302.
24. Parsey, M. V., and G. K. Lewis. 1993. Actin polymerization and pseudopod
reorganization accompany anti-CD3-induced growth arrest in Jurkat T cells.
J. Immunol. 151:1881.
25. Bunnell, S. C., V. Kapoor, R. P. Trible, W. Zhang, and L. E. Samelson. 2001.
Dynamic actin polymerization drives T cell receptor-induced spreading: a role for
the signal transduction adaptor LAT. Immunity 14:315.
26. Kucik, D. F., M. L. Dustin, J. M. Miller, and E. J. Brown. 1996. Adhesionactivating phorbol ester increases the mobility of leukocyte integrin LFA-1 in
cultured lymphocytes. J. Clin. Invest. 97:2139.
27. Clark, E. A., and J. S. Brugge. 1995. Integrins and signal transduction pathways:
the road taken. Science 268:233.
28. Manjunath, N., M. Correa, M. Ardman, and B. Ardman. 1995. Negative regulation of T-cell adhesion and activation by CD43. Nature 377:535.
29. Wild, M. K., A. Cambiaggi, M. H. Brown, E. A. Davies, H. Ohno, T. Saito, and
P. A. van der Merwe. 1999. Dependence of T cell antigen recognition on the
dimensions of an accessory receptor-ligand complex. J. Exp. Med. 190:31.
30. Osman, N., S. C. Lucas, H. Turner, and D. Cantrell. 1995. A comparison of the
interaction of Shc and the tyrosine kinase ZAP-70 with the T cell antigen receptor
␨ chain tyrosine-based activation motif. J. Biol. Chem. 270:13981.
31. Osman, N., S. Lucas, and D. Cantrell. 1995. The role of tyrosine phosphorylation
in the interaction of cellular tyrosine kinases with the T cell receptor ␨ chain
tyrosine-based activation motif. Eur. J. Immunol. 25:2863.
32. Van Oers, N. S., N. Killeen, and A. Weiss. 1994. ZAP-70 is constitutively associated with tyrosine-phosphorylated TCR ␨ in murine thymocytes and lymph
node T cells. Immunity 1:675.
33. Isakov, N., R. L. Wange, W. H. Burgess, J. D. Watts, R. Aebersold, and
L. E. Samelson. 1995. ZAP-70 binding specificity to T cell receptor tyrosinebased activation motifs: the tandem SH2 domains of ZAP-70 bind distinct tyrosine-based activation motifs with varying affinity. J. Exp. Med. 181:375.
Downloaded from http://www.jimmunol.org/ by guest on June 17, 2017
fraction to the cytoskeletal matrix, as illustrated for young CD4
cells in Fig. 3. Our data show that aging alters the pattern of CD3␨
distribution in both resting and activated cells. In the resting cells,
there seems with age to be a replacement of p16␨ by p21␨ in the
soluble fraction, and after activation there is an age-related decline
in the extent to which all three forms of CD3␨ translocate to the
cytoskeletal fraction. These shifts could affect the migration of the
T cell protein kinase Zap-70 to the SMAC, because there is evidence that Zap-70 does not bind to p16␨ (4), but does associate
with p21␨ and p23␨ (30 –33). This differential Zap-70 association,
along with the age-related increase in p21␨ levels, could potentially explain our previous observation of higher levels of Zap-70
associated to TCR complex in CD4⫹ T cells from old mice (9).
The age-related decline in TCR-stimulated translocation to the cytoskeleton of all forms of CD3␨ is consistent with, and may prove
to account for, our previous report of defects in translocation of
Lck and c-Cbl to the SMAC in CD4 T cells of age TCR transgenic
mice (11). In addition, we have unpublished data to show agerelated defects in the translocation of Lck and c-Cbl from the soluble to the cytoskeletal fraction in activated CD4 cells (not
shown). All three approaches, synapse analysis, spreading, and
immunoprecipitation, reveal age-related changes in T cells from
TCR transgenic mice, making it unlikely that our results can be
explained by the age-related shift from naive to memory T cells,
which is greatly inhibited in the transgenic model. Thus, each of
these results supports the notion of an age-related change in interaction of the cytoskeleton with components of the TCR complex,
at least in naive CD4⫹ T cells.
We do not know whether the age-related changes in the association of CD3␨ isoforms with the cytoskeleton are the sole explanation for the signal transduction defects in aged naive CD4
cells. The TCR segregation model suggests other possibilities, including potentially alterations in interaction of TCR or other T cell
surface molecules with APC proteins caused by interferences mediated by large proteins such as CD43 and CD45. Further studies
will need to be performed to differentiate among these and other
mechanistic hypotheses.
5027