Immunity, Vol. 17, 809–822, December, 2002, Copyright 2002 by Cell Press
Dynamics of p56lck Translocation
to the T Cell Immunological Synapse
following Agonist and Antagonist Stimulation
al., 1990). Engagement of CD4 along with CD3 has been
shown to enhance T cell responses in an lck-dependent
manner (Abraham et al., 1991; Anderson et al., 1987;
Glaichenhaus et al., 1991). This augmented activation is
394. However, it has not been clearly demonstrated that
(Luo and Sefton, 1990). Nonetheless, lck, CD4, and CD3
can form a trimolecular complex (Rudd et al., 1991),
bringing lck into proximity with CD3 ITAMs to initiate
Lck SH3 and SH2 domains are also important regula-
tors of kinase activity. Lck can be activated via SH3
domain-mediated interactions with the costimulatory
can interact with the tyrosine kinase ZAP-70 and with
CD3? ITAMs. Deletion of or mutation within this domain
reduces CD3? and ZAP-70 association and phosphory-
lation and greatly impairs T cell activation (Lewis et al.,
1997; Straus et al., 1996). Furthermore, lck can be acti-
vated by interactions with the tyrosine phosphatase
CD45 (Trowbridge and Thomas, 1994).
These protein-protein interactions regulate lck’s ac-
tivity and proximity to effectors, making lck a likely can-
didate for spatial regulation during T cell activation. A
number of studies have recently described the spatial
and temporal recruitment of lck-interacting proteins at
we mean a stable contact region between T cells and
APCs that displays a broad, flat interface. Seminal work
by the laboratories of Kupfer and Dustin on T cell:APC
as LFA-1 were recruited to the synapse periphery,
termed the p-SMAC (peripheral supramolecular activa-
tion cluster), while signaling molecules such as TCR/
CD3 and PKC-? were concentrated in the central region
of thesynapse, the c(central)-SMAC (Dustin etal., 1998;
real time studies of labeled APC molecules in artificial
membranes (Grakoui et al., 1999) and GFP-labeled sur-
face molecules in T and B cells (Krummel et al., 2000).
Previously, in retrospective mature T cell conjugates,
we made the preliminary observation that lck was local-
ized to the periphery of the mature T cell synapse, while
CD3? was in the center (Richie et al., 2002). Given the
complex distribution of lck-interacting proteins at the
immunological synapse, we wished to extend these ob-
servations to precisely define the temporal and spatial
formation. To this end, we expressed an lck-GFP fusion
gate formation in real time. We find that lck is rapidly
recruited to the immunological synapse following agonist
stimulation. Within 15 s of calcium flux, lck accumulates
at the synapse periphery, but in the next few minutes
lck moves dynamically within the synapse, eventually
settling into a peripheral accumulation pattern, colocal-
izing with CD4. These data suggest a limited time for
Lauren I. Richie Ehrlich,1Peter J.R. Ebert,1
Matthew F. Krummel,2, 6Arthur Weiss,3
and Mark M. Davis1,2,4,5
1Program in Immunology
2Department of Microbiology and Immunology
Stanford University School of Medicine
Stanford, California 94305
3Department of Medicine and the
Howard Hughes Medical Institute
University of California, San Francisco
San Francisco, California 94143
4Howard Hughes Medical Institute
Stanford University School of Medicine
Stanford, California 94305
To study the spatio/temporal recruitment of lck during
immunological synapse formation, we utilize high-
speed time-lapse microscopy to visualize green fluo-
however, lck becomes excluded to the periphery of
mature synapses, while most CD3? is centrally local-
ized, suggesting a limited time frame within which lck
can efficiently phosphorylate CD3 molecules during
synapse maturation. Exposure of T cells to specific
APLsaffects theefficiency ofconjugate formationand
lck accumulation. Most surprisingly, we find an intra-
cellular pool of lck associated with recycling endo-
somes that translocates to mature synapses within 10
min of calcium flux. This bolus of lck may contribute
to intermediate-late signal transduction.
T cell activation is initiated when T cell receptors (TCRs)
bind stimulatory peptides in the context of self-major
histocompatibility complex (MHC) molecules. This cell-
tion (Kane et al., 2000). One of the earliest intracellular
modifications downstream of TCR recognition is p56lck
(lck)-mediated tyrosine phosphorylation of CD3 ITAMs
(Iwashima et al., 1994). While much is known about bio-
chemical regulation of lck, little is known about its prox-
imity to the TCR/CD3 during and after immunological
Lck has all of the prototypic domains of other src
family kinases, rendering it capable of multiple protein-
protein interactions (Wange and Samelson, 1996). Lck
binds to the cytoplasmic tails of CD4 and CD8 via cys-
teine motifs in its unique N-terminal region (Turner et
6Present address: Department of Pathology, University of California
San Francisco, San Francisco, California 94143.
formation. We also examined the recruitment of lck to
the synapse following altered peptide ligand (APL) stim-
ulation. We find that the temporal and spatial patterns
agonists, but the efficiency of conjugate formation is
reduced in the later case. Furthermore, antagonists do
not induce efficient conjugate formation. We also de-
scribe an intracellular pool of lck in both T cell systems,
confirming the previous observations of Ley et al. (1994)
in Jurkat T cells. We find that this pool surrounds the
microtubule organizing center (MTOC), colocalizes with
recycling endosomes, and translocates to the interface
following agonist but not antagonist stimulation.
a flat interface, rapidly fluxes calcium, and accumulates
membrane localized lck-GFP at the interface (see Sup-
plemental Movie S1 at http://www.immunity.com/cgi/
content/full/17/6/809/DC1 for the full timecourse). Inter-
estingly, a large intracellular pool of lck-GFP is present
in D10 cells, and it translocates toward the interface,
frequently resulting in a trail of lck-GFP between the
intracellular pool and the membrane accumulation (Fig-
ure 1A, last two frames). Within the synapse, lck-GFP
moves rapidly from the periphery to a more central re-
gion within the first couple of minutes post calcium flux
(compare the 1 min to the 2 min time point). Then lck-
GFP moves to the periphery of the synapse until the
the precise membrane lck localizations (compare 4 and
5 min time points to the 7 min time point).
We also examined lck-GFP accumulations at the im-
munological synapse following stimulation with the an-
tagonist E8T (Dittel et al., 1997). When a D10 cell con-
tacts an E8T-pulsed CH27 B cell, a tight interface can
form between the cells. However, calcium flux is tran-
sient, and lck-GFP does not accumulate at the con-
tacting membranes. Furthermore, the intracellular lck
pool does not translocate to the interface (Figure 1B
and Supplemental Movie S2 at http://www.immunity.
com/cgi/content/full/17/6/809/DC1 for the complete
30 min after T cell:B cell admixture, the intracellular pool
is no longer a distinct entity, as it accumulates at the
contacting interface (Supplemental Figure 2A and Sup-
plemental Movie S3 at http://www.immunity.com/cgi/
content/full/17/6/809/DC1). In contrast, retrospective
E8T conjugates continue to display a distinct intracellu-
lar pool (Supplemental Figure 2B and Supplemental
Movie S4 at http://www.immunity.com/cgi/content/full/
17/6/809/DC1). Specifically, 86% of CA retrospective
conjugates display translocation of the intracellular lck
pool, in contrast to 35% of E8T conjugates.
To quantitate the relative amount of lck-GFP at the
immunological synapse following agonist and antago-
nist stimulation, the average intensity at the T cell:B cell
during conjugate formation. Following CA stimulation,
lck-GFP accumulates to a 1.5-fold level at the interface
within 3 min of calcium flux, and this level is maintained
for at least 13 min. In contrast, there is no significant
accumulation of lck-GFP at the interface following stim-
ulation with the antagonist E8T (Figure 3A).
and antagonist stimulation, the efficiency of conjugate
formation was compared. On a given day, random fields
of D10 cells interacting with either CA-pulsed or E8T-
cell front contacts and the number of conjugates. Cells
were scored positive for conjugation if they stopped
formation is 80% with the agonist peptide CA, but only
with 10.3% with the antagonist E8T. In summary, CA
cumulation, and translocation of the intracellular lck
Lck-GFP Is Functional and Does Not Interfere
with T Cell Receptor-Mediated Activation
GFP fusion with a four amino acid linker, PVAT (Richie et
al., 2002). To ensure that lck-GFP was functional, it was
expressed in JCAM1.6 cells, a Jurkat mutant lacking
lck activity and unable to flux calcium following TCR
stimulation (Straus and Weiss, 1992). Both wild-type lck
and the lck-GFP fusion are able to restore calcium flux
in JCAM1.6 cells following TCR ligation (Supplemental
Figure S1A at http://www.immunity.com/cgi/content/
full/17/6/809/DC1). The identity of the GFP-fusion was
confirmed by Western blot (data not shown). Therefore,
our lck-GFP fusion is functional and can be activated
by TCR ligation.
We also ascertained whether expression of lck-GFP
altered T cell proliferative capacity. Lck-GFP was sub-
cloned into a retroviral vector, and the D10 T cell line
was stably transduced. The D10 TCR is activated by a
peptide of conalbumin (CA134-146) in the context of
I-Ak(Kaye et al., 1983). A comparison of wild-type and
transduced D10 T cells shows that lck-GFP does not
alter D10 proliferation over a range of conalbumin con-
centrations (Supplemental Figure S1B at http://www.
fore, we proceeded to examine the spatial and temporal
localization of lck-GFP at the immunological synapse.
Lck-GFP Translocates to the Immunological Synapse
following Agonist but Not Antagonist Stimulation
in D10 T Cells
To observe lck translocation to the immunological syn-
apse, lck-GFP D10 cells were imaged as they interacted
with CA-pulsed CH27 B cells. The T cells were loaded
with the calcium indicator dye fura-2 before imaging.
Agonist-pulsed CH27 B cells were added to the T cells,
and images were acquired for 15 min at 15 s time inter-
vals. At each interval, four images were acquired: first,
a differential interference contrast (DIC) image, second
and third 340 nm and 380 nm images for calcium analy-
sis, and fourth a 20 ?m z stack of GFP images acquired
at 1 ?m intervals. From this z stack, we generated
3-dimensional (3-D) reconstructions of lck-GFP that
In addition, we followed a central horizontal GFP slice
of the cell over time.
Lck Recruitment to the Immunological Synapse
Figure 1. lck-GFP Is Rapidly Recruited to the D10 T Cell Immunological Synapse following Agonist but Not Antagonist Stimulation
Fura-2 loaded lck-GFP D10 T cells were imaged as they interacted with agonist CA (A)- or antagonist E8T (B)-pulsed CH27 cells. Time-lapse
images were taken every 15 s for 15 min. The top row consists of DIC images; the second row contains fura-2 excitation ratios; the third row
contains GFP images of single horizontal slices through the center of the cell; the fourth row (A) contains a 3-D reconstruction of lck-GFP at
the immunological synapse viewed en face. Time 0 corresponds to the initial rise in intracellular calcium. Scale bars, 10 ?m. See Supplemental
Movies S1 and S2 at http://www.immunity.com/cgi/content/full/17/6/809/DC1 for complete timecourses of (A) and (B), respectively.
Lck-GFP Accumulates Efficiently at the Interface
to Agonist and Weak Agonist but Not Antagonist
Stimulation in 5C.C7 T Cell Blasts
To confirm our results in the D10 T cell line, we utilized
a second T cell system, 5C.C7 TCR transgenics. We
transduced freshly activated 5C.C7 T cell blasts with
lck-GFP and imaged conjugate formation with peptide-
pulsed CH27 cells several days later. The 5C.C7 TCR is
activated by a peptide of moth cytochrome C (MCC 88-
103) in the context of I-Ek(Fazekas de St. Groth et al.,
1993). 102S is a well-characterized weak agonist pep-
As seen in Figure 2A, when a 5C.C7 blast contacts a
MCC-pulsed CH27 B cell, it rapidly fluxes calcium and
forms a tight interface. Membrane localized lck-GFP
quickly translocates to the immunological synapse (see
Supplemental Movie S5 at http://www.immunity.com/
cgi/content/full/17/6/809/DC1 for the complete time-
course). The 3-D reconstructions demonstrate that lck-
(see the 0 min time point), though the very first strong
flux in Supplemental Movie S5 at http://www.immunity.
com/cgi/content/full/17/6/809/DC1). Within a few min-
utes, lck accumulates preferentially in the p-SMAC
(compare 1:30 min time point to 2:15 and subsequent
time points). Unlike D10 cells, 5C.C7 blasts do not dis-
play a bright intracellular pool of lck-GFP. However,
upon closer examination, such a pool can be seen, as
at the 2:15 and the 4:30 time points just behind the
tal Movie S5 (http://www.immunity.com/cgi/content/
full/17/6/809/DC1), indicating that it is not a result of
activation-induced endocytosis. Interestingly, this pool
often corresponds to a region of less intensity in the
(compare the fura-2 images to the GFP images at 2:15
and 4:30 in Figure 2A).
CD4 capping was sufficient to induce lck recruitment.
We observed that antibodies to CD4 but not to CD3
Figure 2. lck-GFP Translocates to the Immunological Synapse in 5C.C7 T Cell Blasts following Agonist and Weak Agonist but Not Antagonist
Fura-2 loaded Lck-GFP 5C.C7 T cell blasts were imaged as they interacted with agonist MCC (A)-, weak agonist 102S (B)-, or antagonist 99R
(C)-pulsed CH27 B cells. The rows are arranged as for Figure 1A. Time is displayed above each panel as min:sec elapsed after the onset of
calcium flux, which is set as time 0. Color scale as for Figure 1. Scalebar, 10 ?m. See Supplemental Movies S5, S6, and S7 at http://
www.immunity.com/cgi/content/full/17/6/809/DC1 for complete timecourses of (A), (B), and (C), respectively.
Lck Recruitment to the Immunological Synapse
Figure 3. Quantitation of lck Accumulations, Calcium Flux, and Efficiency of Conjugate Formation following Agonist and APL Stimulation
(A) During conjugate formation, lck-GFP D10 cells were scored for the average GFP intensity at the interface/average intensity on the remaining
cell membrane. Regions at the interface or the noncontacting portion of the cell were traced in Metamorph (Universal Imaging), and after
subtraction of background fluorescence from an adjacent empty region, the ratio of the average intensity of the two regions was calculated.
The mean of ratios from different conjugates was calculated and displayed versus time. n ? 20 agonist and 9 antagonist conjugates.
(B) 5C.C7 conjugates were scored for lck-GFP accumulations as in (A). n ? 19 agonist, 14 weak agonist, and 3 antagonist conjugates.
(C) lck-GFP 5C.C7 blasts were scored for fold increase of intracellular calcium following activation. n ? 10 agonist, 10 weak agonist, and 4
antagonist conjugates. Time 0 is defined as the onset of calcium flux in all graphs.
(D) Random fields of D10 cells or 5C.C7 cells on the same day were observed for conjugate formation with peptide pulsed CH27 cells. Front
contacts were scored for D10 cells when the leading edge of the cell contacted an APC. 5C.C7 cells were scored for the total number of B
cell contacts since the front edge could not be easily distinguished. The efficiency of conjugate formation was determined as the # conjugates/#
(front) contacts ? 100.
were able to co-cap lck-GFP, indicating that synapse
localized lck is CD4-associated (see Supplemental Fig-
ure S3 at http://www.immunity.com/cgi/content/full/17/
6/809/DC1). Furthermore, lck-GFP was not co-capped
with CD28 (see Supplemental Figure S3 at http://
To determine whether lck recruitment is affected by
APL stimulation, we examined translocation of lck-GFP
in response to the weak and strong agonists. In Figure
2B, the weak agonist 102S induces calcium flux and
rapid lck-GFP accumulation at the interface. Lck-GFP
first accumulates in a central region (see 45 min time
point), then at the synapse periphery over the next sev-
eral minutes (compare:45 to 1:30 and subsequent time
points. See Supplemental Movie S6 at http://www.
complete timecourse). Thus, strong and weak agonists
induce similar temporal and spatial recruitment of lck-
GFP. In contrast, the antagonist 99R does not induce
lck-GFP translocation even when calcium elevation and
a flat interface are observed (Figure 2C and Supplemen-
tal Movie S7 at http://www.immunity.com/cgi/content/
full/17/6/809/DC1). Furthermore, translocation of the in-
tracellular lck-GFP pool can be observed in blasts fol-
lowing 102S stimulation (compare fura-2 and GFP im-
ages between :45 and 3 min time points in Figure 2B),
but not 99R stimulation (see 3:45-5:15 time points in
To quantitate lck-GFP accumulations during agonist
and APL stimulation, the average intensity of GFP at
contacting interfaces was compared to the rest of the
cell membrane. Figure 3B shows that within 1 min of
calcium flux, lck-GFP accumulates to a 1.5-fold level at
the immunological synapse following MCC and 102S
stimulation. In contrast, there is no accumulation seen
for the antagonist 99R.
Two additional criteria were considered for distin-
First, the magnitude and duration of calcium flux were
measured. As seen in Figure 3C, both MCC and 102S
stimulation induce a rapid rise in calcium that plateaus
at 2-fold over background. In contrast, 99R induces a
smaller calcium flux that is transient and periodic. As
the graph represents an average, it does not reflect that
calcium levels return to baseline and spike periodically
following antagonist stimulation in individual cells (data
not shown). The second criterion considered was the
efficiency of conjugate formation. In random fields,
5C.C7 blasts were scored for total contacts made and
the number of conjugates formed with CH27 B cells. As
seen in Figure 3D, the efficiency of 5C.C7 blast conju-
gate formation is 65.3% to MCC, 40.7% to 102S, and
6.8% to 99R stimulation. In summary, while neither the
spatial or temporal accumulations of lck-GFP nor the
magnitude or duration of calcium flux distinguish be-
tween strong and weak agonist stimulation, conjugate
formation is more efficient with the strong agonist MCC.
The antagonist 99R fails to induce lck-GFP accumula-
tion, generates a smaller and less sustained calcium
flux, and reduces the efficiency of conjugate formation.
lations (Figure 5C). These data suggest that lck is re-
cruited to the periphery of mature synapses providing
a limited time in which it can interact efficiently with
To address the temporal window of lck and CD3?
interactions more directly, the synapse accumulation
patterns of both molecules were compared throughout
the duration of conjugate formation. To this end, four
synapse patterns were scored as previously described
(Richie et al., 2002). In brief, interfaces were defined by
that extended to the periphery of the interface but were
accumulations were those in the middle of the interface
that did not extend to the periphery; uniform accumula-
tions were evenly distributed across the interface; and
excluded ? central patterns displayed discreet central
percentage of conjugates displaying the four accumula-
tion patterns over time. Immediately following calcium
flux, both lck and CD3? are predominantly at the periph-
ery of the synapse. Within the next few minutes both
mulations. During these first few minutes, lck has an
opportunity to interact with and phosphorylate CD3?
ITAMs. Over the next 10–15 min, an increasing percent-
age of conjugates display peripheral lck accumulations
ecules become spatially separated. However, it should
be noted that both lck and CD3? move dynamically
throughout the interface at all time points. Thus, while
CD3? and lck are predominantly localized centrally and
peripherally, respectively, at later time points, conju-
gates that deviate from this pattern are found at lower
frequencies (Figure 5D). Such a central accumulation of
lck during a period in which lck is predominantly ex-
cluded can be seen at the 4:30 time point in Figure 2A.
lar accumulation patterns, we performed immunofluo-
rescence on retrospective 5C.C7-CH27 conjugates. As
seen in Figure 6, lck and CD4 largely colocalize at the
immunological synapse periphery, while V?3 is found in
fusions. However, it should be noted that this spatial
ecules at the cellular membrane and those within vesi-
cles just below the surface. Therefore, it is possible that
V?3 has been internalized. The peripheral CD4 accumu-
tions in D10 cells (Krummel et al., 2000). To establish a
direct comparison between CD4 and the prototypical
peripheral synapse marker LFA-1, we performed immu-
nofluorescent analysis on these twoproteins. As seen in
Figure 6, both CD4 and LFA-1 colocalize to the synapse
periphery, confirming that CD4 is a peripheral marker.
CD3?-GFP Accumulates with Similar Dynamics
but Different Spatial Localization within
the Immunological Synapse
We examined the spatial and temporal recruitment of
CD3? in the 5C.C7 system for direct comparison to lck.
5C.C7 T cell blasts were transduced with CD3?-GFP
(Richie et al., 2002) and were imaged several days later
during conjugate formation. As seen in Figure 4A, CD3?
is rapidly recruited to the immunological synapse fol-
lowing calcium flux (compare 1:30 to 1:30, and see Sup-
plemental Movie S8 at http://www.immunity.com/cgi/
content/full/17/6/809/DC1 for the complete timecourse).
At time point 0, the first accumulations of CD3? are at
the periphery of the synapse, but over time, CD3? is
ter of the synapse (compare time point 0 to time points
of CD3? accumulation as reported previously in D10
cells (Krummel et al., 2000).
To compare CD3? and lck, we quantified the fold in-
crease of CD3?-GFP intensity at the interface during
conjugate formation. As seen in Figure 4B, both mole-
cules are recruited to the synapse with remarkably simi-
lar kinetics. One minute aftercalcium flux, they accumu-
late at slightly more than a 1.5-fold level, and this
accumulation is maintained for at least 13 min post cal-
nated at the synapse. We next compared the spatial
accumulations of lck and CD3? in more detail.
Lck-GFP and CD3?-GFP 5C.C7 blasts were allowed
which were rotated to view the immunological synapse
enface. Asseen inFigure 5Aand B,lck-GFPis predomi-
nantly found in peripheral accumulations, while CD3?-
An Intracellular Pool of lck Surrounds the MTOC
and Is Colocalized with Recycling Endosomes
Though an intracellular pool of lck has been previously
described in Jurkat T cells (Ley et al., 1994), we wanted
to ensure that this pool was not a result of mislocaliza-
tion of the GFP fusion protein. Therefore, we stained
endogenous lck in both D10 and 5C.C7 T cells. The
Lck Recruitment to the Immunological Synapse
Figure 4. Accumulation of CD3?-GFP at the Immunological Synapse Is Temporally Similar but Spatially Distinct from lck-GFP
(A) Fura-2 loaded CD3?-GFP 5C.C7 blasts were imaged as they interacted with MCC-pulsed CH27 B cells. The images, color scale, and
time are arranged as for Figure 1. See Supplemental Movie S8 at http://www.immunity.com/cgi/content/full/17/6/809/DC1 for the complete
timecourse. Size bar, 10 ?m.
(B) To quantitate the accumulation of CD3?-GFP at the synapse over time, the relative GFP intensity of 18 conjugates at contact membranes
versus noncontacting membranes was scored and averaged as in Figure 3. The graph was then overlayed with the comparable lck-GFP graph
to allow for direct comparison.
Figure 5. lck-GFP Is Progressively Recruited to the Periphery, while CD3?-GFP Is Recruited to the Center of the Synapse
Retrospective conjugates of lck-GFP (A) or CD3?-GFP (B) 5C.C7 blasts interacting with MCC-pulsed CH27 cells were generated and 3-D
reconstructions of the GFP images were made. Z stacks were deconvolved using 60 iterations of a blind 3D algorithm (AutoQuant). Each pair
of panels represents a side view of the conjugated cell on the left and an en face view of the immunological synapse on the right. (C) 111
CD3?-GFP and 83 lck-GFP 5C.C7 retrospective conjugates were scored for synapse accumulation patterns. (D) Seventeen lck-GFP and fifteen
CD3?-GFP conjugates were scored for accumulation patterns over time. Around calcium flux, a large percentage of conjugates have both lck
and CD3? at the synapse periphery. A diversity of patterns is observed as the synapse begins to mature. By 4 min after flux, lck-GFP is
predominantly recruited to the synapse periphery, while CD3?-GFP is recruited to the center.
deconvolved images in Figure 7A show an intracellular
pool of endogenous lck in both. We then stained lck-
GFP-expressing D10 cells with an anti-? tubulin anti-
body. This shows that the intracellular lck pool tightly
surrounds the microtubule organizing center (MTOC),
as identified by the high-intensity region in the ? tubulin
lular lck might be associated with microtubules. Indeed,
the pool is dispersed into numerous small accumula-
tions following treatment with nocodazole (Figure 7C).
These data suggest that the intracellular pool translo-
catesto theinterface concomitantlywith MTOCreorien-
To narrow the identity of the intracellular pool with
respect to cellular organelles, we stained lck-GFP D10
cells with anti-?-adaptin to identify the trans-golgi net-
work or with anti-transferrin receptor to stain recycling
endosomes. In addition, TRITC-dextran uptake labels
late endosomes. As seen in Figure 7D, neither ?-adaptin
nor TRITC-dextran colocalize with the intracellular pool;
however, anti-transferrin receptor colocalizes exten-
sively. To confirm this result, a pulse-chase experiment
was performed with labeled transferrin. Initially, trans-
ferrin is seen only on cell membranes (Figure 7E top
row), but over time it is internalized so that by 20 min,
it is largely colocalized with the intracellular pool, sug-
gesting colocalization with recycling endosomes. The
labeled T cells were then imaged during conjugate for-
mation with CA-pulsed CH27 cells, demonstrating that
transferrin and intracellular lck translocate to the inter-
face simultaneously (Figure 7F and Supplemental Movie
S9 at http://www.immunity.com/cgi/content/full/17/6/
Translocation of intracellular lck to the immunological
synapse suggests that lck may contribute to T cell sig-
naling in the late synapse. To address this possibility,
we examined antigen-induced 5C.C7 T cell blast prolif-
eration upon addition of PP2, a potent lck and fyn inhibi-
tor (Hanke et al., 1996), at various time points after T
cell:APC admixture. Proliferation is inhibited completely
Lck Recruitment to the Immunological Synapse
Figure 6. Endogenous lck and CD4 Are Predominantly at the Periphery, while TCR Is in the Center of the Mature Immunological Synapse
5C.C7 blasts interacted with MCC-pulsed CH27 cells for 30 min. The conjugates were then fixed, permeabilized, and stained for endogenous
(A) lck, CD4, and V?3 or (B) CD4 and LFA-1. Z stacks of each fluorophore were acquired at 1 ?m intervals and were used to generate 3-D
reconstructions. The top panel of each conjugate represents single z slices through the center of the cell, and the asterisk indicates the
location of the APC. The remaining panels represent an immunological synapse viewed en face as labeled. Note that lck and CD4 largely
colocalize at the periphery, leaving a region of lower intensity in the center of the synapse. TCR is predominantly found in the center of the
synapse (c-SMAC), while LFA-1 and CD4 colocalize to the peripheral region (p-SMAC).
Figure 7. Endogenous lck Can Be Found in an Intracellular Pool in 5C.C7 Blasts and in D10 Cells, and This Pool Surrounds the MTOC and
Colocalizes with Recycling Endosomes
(A) D10 cells and 5C.C7 blasts were fixed, permeabilized, and stained with an anti-lck antibody as in Figure 6. Images were captured at 1 ?m
intervals and were deconvolved using 2-D deconvolution with a nearest-neighbors algorithm. Each panel is a single slice through the middle
of different cells, showing a clear intracellular pool of lck in both 5C.C7 and D10 cells.
(B) lck-GFP D10 cells were fixed, permeabilized, and stained for ?-tubulin. The cells were then imaged as in (A) and a central region of three
different cells is displayed, showing that the lck intracellular pool tightly surrounds the MTOC, which is seen as a bright red accumulation.
(C) lck-GFP D10 cells were treated with 100 ?M nocodazole to disrupt the microtubule cytoskeleton and 3-D reconstructions of the GFP
images were then made. Note that the intracellular pool is dispersed.
(D) lck-GFP D10 cells were fixed, permeabilized, and stained with anti-?-adaptin or anti-transferrin receptor, or they were allowed to uptake
Lck Recruitment to the Immunological Synapse
when PP2 is added within 1 hr of admixture, and is
significantly reduced when PP2 is added after 24 hr
(Supplemental Figure S4 at http://www.immunity.com/
cgi/content/full/17/6/809/DC1). These data strongly in-
dicate that lck and/or fyn tyrosine kinase activity is es-
sential for T cell activation for well over an hour after
to result in complete ITAM phosphorylation, generating
different or reduced signals. In support of this are data
showing that even 2-fold changes in the number of
genes encoding signaling molecules can have dramatic
effects on lymphocyte activation (Cornall et al., 1998).
Therefore, even a 2-fold decrease in association be-
tween lck and CD3? would alter TCR-mediated signals
as the synapse matures, allowing the immunological
synapse to generate varying signals throughout the du-
ration of T cell activation.
It has been suggested that lck activity at the synapse
is short lived and limited to the periphery (Holdorf et al.,
2002; Lee et al., 2002). Consistent with these data, we
find that lck is in close proximity to CD3? at the nascent
synapse periphery, but over the next couple of minutes,
lck and CD3? are found throughout the synapse, where
they may have opportunities to interact. However, in
contrast to the conclusion of Holdorf et al. that lck is
released from CD4 and maintained in the c-SMAC, we
find that lck accumulates peripherally with CD4 in ma-
ture T cell synapses (Figures 5 and 6). Furthermore,
it has been demonstrated that autophosphorylation of
but not following anti-CD3 treatment or other stimuli
that are known to induce T cell activation via lck (Luo
and Sefton, 1990). Therefore, while anti-phopho-Y394
may recognize a particular form of activated lck, it does
not detect all forms, so lck activity at the immunological
synapse may be longerlived than previously suggested.
Indeed, we find that blocking lck and/or fyn activity as
late as 1 hr after T cell:APC admixture blocks T cell
proliferation, consistent with the importance of lck
activityat thesynapsewell beyondtheonset ofsynapse
formation (Supplemental Figure S4 at http://www.
PP2 blocks both lck and fyn, the effect that we see is
more likely due to an lckblockade as analyses of knock-
out mice show that lck is more critical in peripheral T
cell activation than fyn (Appleby et al., 1992; Molina et
al., 1992; Stein et al., 1992). Thus, translocation of lck
after synapse formation is likely to make an important
contribution to T cell activation.
As lck is the first tyrosine kinase to initiate signaling
downstream of an engaged TCR, and CD3? ITAM phos-
phorylation patterns have been shown to vary following
APL stimulation (Kersh et al., 1998), we reasoned that
reported an alteration in the relative localizations of lck
and CD3? within thymocyte immunological synapses
apse architecture contributes to distinct TCR-mediated
signals (Richie et al., 2002). However, in mature T cells,
we did not find a difference in the extent, timing, or
pattern of lck recruitment to the synapse following weak
agonist signaling (Figure 3). Indeed, the only reproduc-
ible difference was a reduced efficiency of conjugate
Using GFP fusions, we characterized the relative spatial
and temporal recruitment of lck and CD3? to the immu-
nological synapse. We find that both are rapidly re-
cruited to the synapse where they are stably maintained
for at least 15 min (Figures 3 and 4). Both lck and CD3?
are initially recruited to the synapse periphery and then
move dynamically throughout the synapse during the
first few minutes after calcium flux. CD3? then becomes
enriched in the center while lck is enriched at the syn-
apse periphery (Figure 5). Endogenous lck and TCR ex-
hibit the same distinct accumulation patterns (Figure 6).
These data suggest a limited window of opportunity
for lck and CD3? to interact efficiently at the nascent
ated throughout synapse formation. The disparate spa-
tial accumulations of CD3? and lck mirror our previous
studies of CD3? and CD4 segregation in a T cell line
(Krummel et al., 2000) and are in contrast to studies of
Zal and colleagues (2002) who found CD3? and CD4
dispersed throughoutthe synapse ina Tcell hybridoma.
Indeed, we find that lck colocalizes with CD4 at the
mature immunological synapse periphery (Figure 6).
Most importantly, anti-CD4 capping recruits most of the
detectable membrane proximal lck (Supplemental Fig-
ure S1 at http://www.immunity.com/cgi/content/full/17/
6/809/DC1), suggesting CD4 association of lck at the
patterns do not reflect an absolute molecular segrega-
tion. Instead, we observe a change in the relative distri-
bution of molecules, which could have a substantial
impact on signals transduced via CD3 ITAM phosphory-
lation. If lck and CD3 cannot interact efficiently, many
fewer second messengers would be generated, signifi-
cantly impacting signals transduced through the TCR.
This concept has been mathematically verified in the
kinetic proofreading (McKeithan, 1995) and kinetic dis-
crimination (Rabinowitz et al., 1996a) models in which
changes in the kinetics of TCR-peptide:MHC binding
alter the relative concentrations of signaling intermedi-
ates, such that early intermediates result in different
signals than later intermediates. Thus, enrichment of
lck and CD3? in different regions of the synapse could
reduce the number of interactions of sufficient duration
TRITC-dextran before imaging. A single slice through the center of two different cells for each stain is shown. Note that transferrin receptor
colocalizes with the intracellular pool of lck-GFP.
(E) lck-GFP D10 cells were pulse/chased for the indicated times with Texas Red-transferrin. A central slice for each cell is shown.
(F) Time points from conjugate formation of an lck-GFP D10 cell with internalized TR-transferrin. Note the simultaneous translocation of internal
lck and transferrin to the interface. See Supplemental Movie S9 at http://www.immunity.com/cgi/content/full/17/6/809/DC1 for the complete
formation (Figure 3D). Previous work has shown that
weak agonists induce less T cell proliferation but similar
acid release and calcium flux as strong agonists (Rabi-
nowitz et al., 1996b). Our data are consistent with these
observations: if a smaller percentage of T cells are in-
duced to form conjugates by weak agonists, then the
maximum proliferation of the population would be re-
to induce significant calcium flux or lck accumulation
at the synapse (Figure 3). Furthermore, they did not
induce conjugate formation any more efficiently than
unpulsed CH27 cells (data not shown and Figure 3D).
These dataare consistent with previousstudies demon-
strating that antagonists fail to induce acid release, cal-
cium flux, or T cell proliferation above the level induced
by endogenous APC peptides (Rabinowitz et al., 1996b;
Wulfing et al., 1997). However, our data differ from the
results of Huang et al. (2000) who find that antagonists
induce efficient conjugate formation. This is likely to
reflect differences in experimental design, as Huang et
al. used a much higher concentration of antagonist pep-
tides (55 ?M versus 1 ?M).
lck to the synapse of both D10 and 5C.C7 cells (Supple-
mental Movies S1 and S5 at http://www.immunity.
com/cgi/content/full/17/6/809/DC1). Staining for en-
dogenous lck revealed that this pool was not a product
of lck-GFP mislocalization (Figure 7A). A similar intracel-
lular pool has been described in Jurkat cells (Ley et
al., 1994), and lck has been biochemically identified in
endosomes of nonactivated T cells (Luton et al., 1997).
As in Jurkat cells, intracellular lck surrounds the MTOC
in D10 cells and can be dispersed by the microtubule
dissociating agent nocodazole (Figures 7B and 7C).
Consistent with localization to recycling endosomes,
intracellular lck is colocalized with transferrin receptor
and internalized transferrin (Figures 7D, 7E, and 7F).
Thus, intracellular lck might recycle back to the mem-
brane, providing an additional source of the kinase at
mature synapses. Indeed, both labeled transferrin and
the intracellular lck pool translocate to the T cell:APC
interface following agonist stimulation (Figure 7F and
Supplemental Movie S9 at http://www.immunity.com/
cgi/content/full/17/6/809/DC1). Kuhn and colleagues
(2002) have recently demonstrated that the MTOC reori-
ents such that it is located at the periphery of the syn-
apse, consistent with localization of lck at the synapse
periphery in retrospective conjugates (Figure 5A). There
is precedence for the fusion of an intracellular pool of
signaling molecules with the plasma membrane, as
CTLA-4 has recently been shown to act in a comparable
manner (Egen and Allison, 2002). It is known that a T
cell must interact with an APC for at least 2 hr to commit
to activation (Karttunen and Shastri, 1991; Lee et al.,
2002). Possibly, intracellular lck provides a bolus of ki-
nase activity needed for full activation of the T cell.
Interestingly, this translocation of intracellular lck may
place it in proximity to CD45 molecules that are on intra-
cellular membranes proximal to the synapse (Johnson
et al., 2000), potentially leading to lck activation at this
location. Alternatively, lck could play a regulatory role
in T cell signaling along with CTLA-4 (Egen and Allison,
2002). It is also possible that intracellular lck could
counter the activity of CTLA-4-associated phospha-
Our data on lck recruitment to the immunological syn-
apse, along with a wealth of previous experimentation,
suggests a multistage role for lck in T cell activation.
First, lck initiates signaling by phosphorylating CD3
number of TCRs (Iwashima et al., 1994). Rapid reorgani-
per, 2000), recruiting molecules such as CD3, CD4,
CD28, CD45, and lck to the immunological synapse
(Bromley et al., 2001; Johnson et al., 2000; Krummel et
al., 2000). At this point, lck may play a second signaling
role by contributing to CD3 ITAM phosphorylations in
nascent synapses, where we find that lck and CD3?
are colocalized first at the synapse periphery and then
throughout the synapse (Figure 3). This second wave of
lck activity may strengthen TCR-mediated signals and/
or establish cellular polarity (Davis and van der Merwe,
2001). In addition, lck may play a role in integrin affinity
maturation, allowing for conjugate stabilization. It has
been shown that src-family kinases can contribute to
integrin signaling (Cary et al., 2002), and T cell:B cell
conjugation is dependent on lck, but not ZAP-70, im-
plying a distinct role for lck in conjugate stabilization
(Morgan et al., 2001). Once molecules have spatially
segregated within the mature synapse, the MTOC is
recruited to the T cell:B cell interface (Kuhn and Poenie,
2002; Kupferet al.,1987; Lowin-Kropfet al.,1998). Inter-
estingly, we find that a bolus of intracellular lck trans-
locates to the synapse, concomitant with MTOC re-
orientation (Supplemental Figure S2, Figure 7, and
Supplemental Movie S9 at http://www.immunity.com/
cgi/content/full/17/6/809/DC1). Thus, lck may play a
third role in T cell signaling by contributing to signals
emanating from late mature synapses. Indeed, Supple-
mental Figure S4 (http://www.immunity.com/cgi/content/
full/17/6/809/DC1) demonstrates that lck and/or fyn ac-
tivity isessential forT cellactivation forat least1 hrafter
T cell:APC admixture, suggesting that lck contributes to
T cell activation following synapse formation.
Constructs and Generation of Cell Lines and Blasts
cDNAs for CD3? and lck were cloned into EGFP-N1 (Clontech) to
generate in frame carboxy terminal EGFP fusion constructs (Krum-
mel et al., 2000). JCAM1.6 cells were transfected by electroporation
using standard protocols. Transfectants were selected in G418
(Sigma) and screened for GFP expression. cDNAs for lck-GFP and
CD3?-GFP were subcloned into the retroviral vector pIB (gift from
cells (gift of Dr. Garry Nolan), which were stably selected and main-
tained in 10 ?g/ml blasticidin (Invitrogen). The producer lines were
monitored for GFP expression. Virus was collected and concen-
trated as previously described (Richie et al., 2002).
5C.C7 blasts were transduced with retrovirus and utilized for con-
jugate formation as previously described (Richie et al., 2002). D10
cells were maintained by stimulation of 106cells with 107fresh,
irradiated B10.BR splenocytes pulsed with 100 ?g/ml whole conal-
bumin (Sigma) every 7 days in 10 ml complete RPMI containing 10%
FBS, 100 U/ml penicillin/streptomycin, 2 mM L-glutamine, and 50
?M ?-mercaptoethanol. 24 hr poststimulation one flask of D10 cells
was spin infected in a 24-well plate in 1 ml concentrated retrovirus
(Richie et al., 2002). 24 hr later, the cells were transferred to 5 ml
Lck Recruitment to the Immunological Synapse
of complete RPMI ? 2% IL-2 ? 20 ?g/ml blasticidin. GFP? cells
were sorted by FACS and maintained as for nontransduced cells
with the addition of 10 ?g/ml blasticidin. Both GFP? 5C.C7 blasts
and D10 cells were monitored by FACS for normal expression of
TCR, CD4, and CD28.
incubated at 37?C for the indicated periods of time prior to imaging.
After 20 min, CA-pulsed CH27 cells were added and timelapse im-
ages were obtained every 15 s for 15 min.
Calcium Flux Analysis
Jurkat, JCAM1.6,and reconstitutedJCAM1.6 lineswere loadedwith
1 ?M indo-1, AM (Molecular Probes) according to manufacturer’s
instructions. Propidium iodide was used to exclude dead cells. Live
cells were analyzed at 37?C on a modified FACStar (FACS facility,
Stanford University) with a clock that recorded elapsed time. The
ratio of 405/485 emission, an indicator of intracellular calcium con-
centration, was monitored following addition of a saturating amount
of monoclonal antibody C305 and then 1 ?M ionomycin (Sigma).
We thank Cenk Sumen for the pIB construct and Bjo ¨rn Lillemeier
for assistance. L.I.R.E. and P.J.R.E are Howard Hughes Medical
tor. Supported by grants from the National Institutes of Health and
the Howard Hughes Medical Institute (to M.M.D).
Received: May 31, 2002
Revised: September 30, 2002
104D10 cells or 5C.C7 blasts that had been rested for 5 days were
incubated with 5 ? 104irradiated B10.BR splenocytes/well in 100
?l in a 96-well plate. 0-100 ?g/ml whole conalbumin were added to
triplicate wells of D10 cells, and 5 ?M MCC was added to the
indicated wells of 5C.C7 blasts. PP2 was added to the blasts at the
indicated times after T cell:splenocyte admixture. 72 hr later for D10
and 48 hr later for 5C.C7 blasts 1 ?Ci
well. After 16 hr, the plates were harvested, and the filter paper was
analyzed by liquid scintillation. Counts from triplicate wells were
Abraham, N., Miceli, M.C., Parnes, J.R., and Veillette, A. (1991).
Enhancement of T-cell responsiveness by the lymphocyte-specific
tyrosine protein kinase p56lck. Nature 350, 62–66.
Cross-linking of T3 (CD3) with T4 (CD4) enhances the proliferation
of resting T lymphocytes. J. Immunol. 139, 678–682.
Appleby, M.W., Gross, J.A., Cooke, M.P., Levin, S.D., Qian, X., and
Perlmutter, R.M. (1992). Defective T cell receptor signaling in mice
lacking the thymic isoform of p59fyn. Cell 70, 751–763.
Bromley, S.K., Iaboni, A., Davis, S.J., Whitty, A., Green, J.M., Shaw,
A.S., Weiss, A., and Dustin, M.L. (2001). The immunological synapse
and CD28-CD80 interactions. Nat. Immunol. 2, 1159–1166.
Cary, L.A., Klinghoffer, R.A., Sachsenmaier, C., and Cooper, J.A.
(2002). SRC catalytic but not scaffolding function is needed for
integrin-regulated tyrosine phosphorylation, cell migration, and cell
spreading. Mol. Cell. Biol. 22, 2427–2440.
Cornall, R.J., Cyster, J.G., Hibbs, M.L., Dunn, A.R., Otipoby, K.L.,
Lyn, CD22, and SHP-1 are limiting elements of a biochemical path-
way regulating BCR signaling and selection. Immunity 8, 497–508.
Davis, S.J., and van der Merwe, P.A. (2001). The immunological
synapse: required for T cell receptor signalling or directing T cell
effector function? Curr. Biol. 11, R289–291.
antagonists inhibit proliferation and the production of IL-4 and/or
IFN-gamma in T helper 1, T helper 2, and T helper 0 clones bearing
the same TCR. J. Immunol. 158, 4065–4073.
Dustin, M.L., and Cooper, J.A. (2000). The immunological synapse
and the actin cytoskeleton: molecular hardware for T cell signaling.
Nat. Immunol. 1, 23–29.
Dustin, M.L., Olszowy, M.W., Holdorf, A.D., Li, J., Bromley, S., Desai,
N., Widder, P., Rosenberger, F., van der Merwe, P.A., Allen, P.M.,
et al. (1998). A novel adaptor protein orchestrates receptor pat-
terning and cytoskeletal polarity in T-cell contacts. Cell 94, 667–677.
Egen, J.G., and Allison, J.P. (2002). Cytotoxic T lymphocyte antigen-4
accumulation in the immunological synapse is regulated by TCR
signal strength. Immunity 16, 23–35.
Enns, C.A., Shindelman, J.E., Tonik, S.E., and Sussman, H.H. (1981).
Radioimmunochemical measurement of the transferrin receptor in
receptor antibody. Proc. Natl. Acad. Sci. USA 78, 4222–4225.
Fazekas de St. Groth, B., Patten, P.A., Ho, W.Y., Rock, E.P., and
Davis, M.M. (1993). An analysis of T cell receptor-ligand interaction
using a transgenic antigen model for T cell tolerance and T cell
receptor mutagenesis. In Molecular Mechanisms of Immunological
Self-Recognition, F.W. Alt and H.J. Vogel, eds. (San Diego: Aca-
demic Press), pp. 123–127.
Glaichenhaus, N., Shastri, N., Littman, D.R., and Turner, J.M. (1991).
Requirement for association of p56lck with CD4 in antigen-specific
signal transduction in T cells. Cell 64, 511–520.
Grakoui, A., Bromley, S.K., Sumen, C., Davis, M.M., Shaw, A.S.,
Allen, P.M., and Dustin, M.L. (1999). The immunological synapse: a
3H-Thymidine was added/
Transduced D10 or 5C.C7 conjugates were loaded with 1 ?M fura-2,
AM (Molecular Probes) for 20 min prior to imaging. The cells were
washed and 2 ? 105cells were put into an 8-well glass-bottom
chamber slide (Labtek/Nunc). 105peptide-pulsed CH27 B cells were
allowed to settle onto the slide at the onset of imaging.
Imaging was carried out on a previously described microscope
taken every 15 s for 15 min. Microscope control, data acquisition,
and image analysis were performed in Metamorph (Universal Im-
aging). In time-lapse experiments, bleaching was corrected using
a previously described algorithm (Krummel et al., 2000).
Co-Capping, Blocking, and Cytoskeletal Inhibitor Experiments
For co-capping, lck-GFP D10 cells were incubated on ice for 20
min with biotinylated primary antibodies against CD4 (GK1.5), CD28
(37.51), or CD3 (145-2C11) (Pharmingen) followed by Av-PE (Phar-
mingen) at37?C for20 min.The cellswere fixedin 2%paraformalde-
hyde, mounted on slides, and imaged. To disrupt the actin or micro-
tubule cytoskeleton, T cells were incubated with 100 ?M colchicine
(Sigma) or 100 ?M nocodazole (Sigma) for 30 min prior to imaging.
Immunofluorescence and Transferrin or Dextran Uptake
5C.C7 blastswere allowed toconjugate for 30 minwith MCC-pulsed
CH27 B cells on poly-L-lysine coated slides (Wescor). The cells
were fixed in 4% paraformaldehyde for 10 min and blocked in PBS
containing 1% FBS ? .3% saponin for 1 hr at room temperature.
The following primary antibodies were added to the slides for 1 hr in
thedark:monoclonal anti-lckclone3A5(Santa CruzBiotechnology),
gated anti-mouse Ig, Fc? fragment-specific antibody (Jackson Im-
munoResearch) was added for 1 hr in the dark. The slides were
mounted in ProLong Antifade (Molecular Probes).
For staining of intracellular organelles, lck-GFP D10 cells were
fixed, permeabilized in 1% Triton x-100, and stained with primary
antibodies against ?-adaptin (AP1m, Pharmingen), human trans-
ferrin receptor (Enns et al., 1981), or ?-tubulin (Sigma) followed
by PE-conjugated anti-mouse, rhodamine-conjugated anti-goat, or
rhodamine-conjugated anti-mouse antibodies, respectively. For dex-
tranuptake, lck-GFP D10 cellswere resuspended in 5 mg/ml TRITC-
dextran (Sigma) and incubated at 37?C for 45 min.
For the transferrin pulse-chase, lck-GFP D10 cells were chilled on
ice and incubated with 10 ?g/ml Texas Red-conjugated transferrin
(Molecular probes) for 30 min. The cells were washed twice and
Immunity Download full-text
molecular machine controlling T cell activation. Science 285,
Hanke, J.H., Gardner, J.P., Dow, R.L., Changelian, P.S., Brissette,
W.H., Weringer, E.J., Pollok, B.A., and Connelly, P.A. (1996). Discov-
tor. J. Biol. Chem. 271, 695–701.
Holdorf, A.D., Green, J.M., Levin, S.D., Denny, M.F., Straus, D.B.,
Link, V., Changelian, P.S., Allen, P.M., and Shaw, A.S. (1999). Proline
residues in CD28 and the Src homology (SH)3 domain of Lck are
required for T cell costimulation. J. Exp. Med. 190, 375–384.
Holdorf, A.D., Lee, K.H., Burack, W.R., Allen, P.M., and Shaw, A.S.
(2002). Regulation of Lck activity by CD4 and CD28 in the immuno-
logical synapse. Nat. Immunol. 3, 259–264.
Huang, J., Sugie, K., La Face, D.M., Altman, A., and Grey, H.M.
(2000). TCR antagonist peptides induce formation of APC-T cell
conjugates and activate a Rac signaling pathway. Eur. J. Immunol.
Iwashima, M., Irving, B.A., van Oers, N.S., Chan, A.C., and Weiss,
A. (1994). Sequential interactions of the TCR with two distinct cyto-
plasmic tyrosine kinases. Science 263, 1136–1139.
A supramolecular basis for CD45 tyrosine phosphatase regulation
in sustained T cell activation. Proc. Natl. Acad. Sci. USA 97, 10138–
Kane, L.P., Lin, J., and Weiss, A. (2000). Signal transduction by the
TCR for antigen. Curr. Opin. Immunol. 12, 242–249.
activation in single viable T cells using the lacZ reporter gene. Proc.
Natl. Acad. Sci. USA 88, 3972–3976.
Both a monoclonal antibody and antisera specific for determinants
unique to individual cloned helper T cell lines can substitute for
antigen and antigen-presenting cells in the activation of T cells. J.
Exp. Med. 158, 836–856.
Kersh, E.N., Shaw, A.S., and Allen, P.M. (1998). Fidelity of T cell
activation through multistep T cell receptor zeta phosphorylation.
Science 281, 572–575.
Krummel, M.F., Sjaastad, M.D., Wulfing, C., and Davis, M.M. (2000).
Differential clustering of CD4 and CD3zeta during T cell recognition.
Science 289, 1349–1352.
tubule cytoskeleton during CTL-mediated killing. Immunity 16,
Kupfer, A., Swain, S.L., and Singer, S.J. (1987). The specific direct
interaction of helper T cells and antigen-presenting B cells. II. Reori-
entation of the microtubule organizing center and reorganization of
the membrane-associated cytoskeleton inside the bound helper T
cells. J. Exp. Med. 165, 1565–1580.
Lee, K.H., Holdorf, A.D., Dustin, M.L., Chan, A.C., Allen, P.M., and
Shaw, A.S. (2002). T cell receptor signaling precedes immunological
synapse formation. Science 295, 1539–1542.
Lewis, L.A., Chung, C.D., Chen, J., Parnes, J.R., Moran, M., Patel,
V.P., and Miceli, M.C. (1997). The Lck SH2 phosphotyrosine binding
site is critical for efficient TCR-induced processive tyrosine phos-
phorylation of the zeta-chain and IL-2 production. J. Immunol. 159,
Ley, S.C., Marsh, M., Bebbington, C.R., Proudfoot, K., and Jordan,
P. (1994). Distinct intracellular localization of Lck and Fyn protein
tyrosin kinases in human T lymphocytes. J. Cell Biol. 125, 639–649.
Lowin-Kropf, B., Shapiro, V.S., and Weiss, A. (1998). Cytoskeletal
polarization of T cells is regulated by an immunoreceptor tyrosine-
based activation motif-dependent mechanism. J. Cell Biol. 140,
Luo, K.X., and Sefton, B.M. (1990). Cross-linking of T-cell surface
molecules CD4 and CD8 stimulates phosphorylation of the lck tyro-
sine protein kinase at the autophosphorylation site. Mol. Cell. Biol.
Luton, F., Legendre, V., Gorvel, J.P., Schmitt-Verhulst, A.M., and
ated with ligand-induced internalized TCR/CD3 complexes. J. Im-
munol. 158, 3140–3147.
McKeithan, T.W. (1995). Kinetic proofreading in T-cell receptor sig-
nal transduction. Proc. Natl. Acad. Sci. USA 92, 5042–5046.
Molina, T.J., Kishihara, K., Siderovski, D.P., van Ewijk, W., Naren-
dran, A., Timms, E., Wakeham, A., Paige, C.J., Harmann, K.-U., Veil-
lette, A., et al. (1992). Profound block in thymocyte development in
mice lacking p56lck. Nature 357, 161–164.
Monks, C.R., Freiberg, B.A., Kupfer, H., Sciaky, N., and Kupfer, A.
(1998). Three-dimensional segregation of supramolecular activation
clusters in T cells. Nature 395, 82–86.
D.B., and Burkhardt, J.K. (2001). Superantigen-induced T cell:B cell
conjugation is mediated by LFA-1 and requires signaling through
Lck, but not ZAP-70. J. Immunol. 167, 5708–5718.
Rabinowitz, J.D., Beeson, C., Lyons, D.S., Davis, M.M., and McCon-
nell, H.M. (1996a). Kinetic discrimination in T-cell activation. Proc.
Natl. Acad. Sci. USA 93, 1401–1405.
M.M., and McConnell, H.M. (1996b). Altered T cell receptor ligands
trigger a subset of early T cell signals. Immunity 5, 125–135.
Richie, L.I., Ebert,P.J.R., Wu, L.C., Krummel, M.F.,Owen, J.J.T., and
Davis, M.M. (2002). Imaging synapse formation during thymocyte
selection: inability of CD3zeta to form a stable central accumulation
during negative selection. Immunity 16, 595–606.
Rudd, C.E., Barber, E.K., Burgess, K.E., Hahn, J.Y., Odysseos, A.D.,
Sy, M.S., and Schlossman, S.F. (1991). Molecular analysis of the
interaction of p56lck with the CD4 and CD8 antigens. Adv. Exp.
Med. Biol. 292, 85–96.
Stein, P.L., Lee, H.M., Rich, S., and Soriano, P. (1992). pp59fyn
mutant mice display differential signaling in thymocytes and periph-
eral T cells. Cell 70, 741–750.
Straus, D.B., and Weiss, A. (1992). Genetic evidence for the involve-
ment of the lck tyrosine kinase in signal transduction through the
T cell antigen receptor. Cell 70, 585–593.
Straus, D.B., Chan, A.C., Patai, B.,and Weiss, A. (1996). SH2 domain
function is essential for the role of the lck tyrosine kinase in T cell
receptor signal transduction. J. Biol. Chem. 271, 9976–9981.
Trowbridge, I.S., and Thomas, M.L. (1994). CD45: an emerging role
as a protein tyrosine phosphatase required for lymphocyte activa-
tion and development. Annu. Rev. Immunol. 12, 85–116.
Turner, J.M., Brodsky, M.H., Irving, B.A., Levin, S.D., Perlmutter,
R.M., and Littman, D.R. (1990). Interaction of the unique N-terminal
region of tyrosine kinase p56lck with cytoplasmic domains of CD4
and CD8 is mediated by cysteine motifs. Cell 60, 755–765.
Wange, R.L., and Samelson, L.E. (1996). Complex complexes: sig-
naling at the TCR. Immunity 5, 197–205.
Wulfing, C., Rabinowitz, J.D., Beeson, C., Sjaastad, M.D., McCon-
nell, H.M., and Davis, M.M. (1997). Kinetics and extent of T cell
activation as measured with the calcium signal. J. Exp. Med. 185,
Zal, T., Zal, M.A., and Gascoigne, N.R. (2002). Inhibition of T cell
receptor-coreceptor interactions by antagonist ligands visualized
by live FRET imaging of T-hybridoma immunological synapse. Im-
munity 16, 521–534.