| HOME | HELP | FEEDBACK | SUBSCRIPTIONS | ARCHIVE | SEARCH | TABLE OF CONTENTS |
|
|
|
|||||||
|
|||||||||
/CD3 Complex, But Not CD28, Interact with Clathrin Adaptor Complexes AP-1 and AP-2
,||,||
*
Department of Cancer Immunology and AIDS, Dana-Farber Cancer Institute, Boston, MA 02115;
Sackler School of Biomedical Sciences, Tufts University, and
Center for Blood Research, and Departments of
§
Medicine,
¶
Pathology, and
||
Cell Biology, Harvard Medical School, Boston, MA 02115
 |
Abstract |
|---|
 Top Abstract IntroductionMaterials and MethodsResultsDiscussionReferences |
|---|
/CD3 complex (but not CD28) also binds to AP-1 and AP-2
complexes, thus providing a possible link between these two receptors
in the regulation of T cell function. |
Introduction |
|---|
|
TopAbstractIntroductionMaterials and MethodsResultsDiscussionReferences |
|---|
chain and the adaptor protein LAT
(13, 14). CTLA-4 may therefore disrupt TCR/CD3
signaling (15, 16), as well as down-regulating
CD28-mediated potentiation of activation (17). The
blockage of CD28-CD80 binding has also been reported to reverse the
hyperresponsive phenotype of CTLA-4-negative mice
(18). In addition to negative signaling, CTLA-4 can be distinguished from CD28 by the fact that it is primarily an intracellular Ag (19). Surface expression is low and tightly regulated, occurring as early as 6 h postactivation and reaching its highest levels by some 36 h (5). Even at its highest levels, however, CTLA-4 is expressed at only some 3% relative to CD28 (5). This may be due to a high level of compartmentalization in the trans-Golgi network (TGN) and secretory vesicles (19, 20), and rapid internalization from the surface of T cells (20, 21). The molecular basis for this intracellular accumulation is unclear, although it has been reported to be linked to a tyrosine-containing motif in the CTLA-4 cytoplasmic tail (19). This intracellular localization is unusual for a surface receptor, and has led to speculation regarding its function, possibly allowing intracellular signaling, or serving as a mechanism to control the polarized release of CTLA-4 at sites of contact between T cells and APC (21).
Little is known regarding the processes that control the intracellular
accumulation of CTLA-4 in T cells. Clathrin-coated vesicles are
involved in selective intracellular trafficking of membrane proteins in
eukaryotes (22). At least three distinct tetrameric
adaptor complexes, AP-1, AP-2, and AP-3, associate with clathrin
(23). AP-1 complexes are localized in the TGN and are
involved in lysosomal and cell surface targeting, while AP-2 complexes
are found at the plasma membrane and regulate endocytosis
(24, 25, 26). AP-3 complexes play a role in cargo-selective
transport from the Golgi to intracellular vacuoles
(27, 28, 29). AP-1 and AP-2 complexes contain two large
subunits (
and ß-1 vs 
and ß-2), one medium chain (µ-1
(AP-47) vs µ-2 (AP-50), and a small chain (
1 (AP-19) vs -2
(AP-17) (30, 31, 32, 33). The ß-1 and ß-2 chains are most
closely related to each other, and bind to clathrin (34).
AP-1 localization with the Golgi membrane requires the core complex and
a small GTP-binding protein ADP-ribosylation factor (35, 36). Subsequent recruitment of cytosolic clathrin is dependent
on the presence of intact AP-1 (33, 37, 38).
Although somewhat poorly defined, the µ-1 and µ-2 chains bind to nonphosphorylated tyrosine-based sorting motifs with the sequence YXXØ, where X can be any amino acid and Ø is a large hydrophobic amino acid (39, 40, 41, 42). Using the two-hybrid screening assay, several laboratories found that the CTLA-4 cytoplasmic tail can bind to the µ-2 chain of the AP-2 complex (43, 44, 45, 46). A similar tyrosine-based motif may function in the case of TGN 38, Lamp-1, CD69, and H2-Mb, proteins that are sorted at the TGN to the endosomal/lysosomal pathway (39). By contrast, a similar motif YMNM in the cytoplasmic tail of CD28 did not interact with the µ-2 chain. AP-2 binding to CTLA-4 provides a possible mechanism by which CTLA-4 becomes endocytosed; however, it leaves unresolved the mechanism responsible for the accumulation of CTLA-4 in the TGN (19, 20).
In this study, we report the identification of an interaction between
the clathrin adaptor complex AP-1 and intracellular forms of CTLA-4,
found primarily in the Golgi compartment. AP-1 appears to play a role
in regulating the shuttling of excess receptor from the Golgi to the
lysosomal compartments for degradation thereby maintaining a constant
level of intracellular receptor. CTLA-4/AP-1 binding serves as one
mechanism to regulate intracellular levels of CTLA-4 in T cells.
Finally, we demonstrate that the TCR/CD3 complex (but not CD28) also
binds to AP-1 and AP-2 complexes, thus providing a potential link
between these two receptors in the regulation of T cell function.
|
Materials and Methods |
|---|
|
TopAbstractIntroductionMaterials and MethodsResultsDiscussionReferences |
|---|
The murine T cell hybridoma DC27.10 (gift of Dr. R. Zamoyska, Medical Research Council, London, U.K.) was cultured in RPMI 1640 medium supplemented with 5% (v/v) FBS (Intergen, Purchase, NY), 100 U/ml penicillin,100 mg/ml streptomycin (Life Technologies, Grand Island, NY), 2 mM L-glutamine (Life Technologies), and 50 µM 2-ME. PBL were isolated from the buffy coat by lymphocyte separation medium (Ficoll-Paque) density-gradient centrifugation. Adherent cells were depleted from the PBL by plastic absorption. The nonadherent cells were cultured for 2 days in RPMI 1640 supplemented with 10% (v/v) FBS, 100 U/ml penicillin, 100 mg/ml streptomycin, 2 mM L-glutamine, PHA (1 µg/ml), and human rIL-2 (50 U/ml).
Anti-CTLA-4 (11D4) has been previously described (47),
while anti-CTLA-4 (BNI3.1) was purchased from Immunotech
(Marseille, France). Anti-CD28 (4B10) was a kind gift from Coulter
(Hialeah, FL). Anti-murine CD3 (145-2C11) was purchased from American
Type Culture Collection (Manassas, VA). Anti-p85 of PI 3-kinase was
provided by Dr. M. White (Joslin Diabetes Center, Boston, MA).
Anti--subunit Ab specific for AP-1 was kindly provided by Dr. L.
Traub (Washington University, St. Louis, MO); the specific
anti-ß-chain antiserum by Dr. T. Kirchhausen (Center for Blood
Research, Boston, MA). Unlabeled and PE-labeled mouse IgG2a (specific
for 2,4,6-trinitrophenol; TNP) and PE-labeled anti-CTLA-4 were
purchased from PharMingen (San Diego, CA). Monoclonal anti-Golgi
58K protein was obtained from Sigma (St. Louis, MO). mAb lamp-2 was
purchased from Developmental Studies Hybridoma Bank (Iowa). Ionomycin
and PMA were obtained from Calbiochem (San Diego, CA).
Transfections of human CTLA-4 and chimera
Human CTLA-4 was obtained by PCR using pCDM8-CTLA-4 (kindly
provided by G. Freeman, Dana-Farber Cancer Institute, Boston, MA) as a
template. Primers for PCR were: GAA TTC ATG GCT TGC CTT GGA TTT. The
PCR product was digested with EcoRI and subsequently cloned
into the EcoRI site of the SR expression vector
(48). Amino acid residue at position Y-201 as well as
CD28GVYVKM and CTLA-4SDYMNM were subjected to site-directed mutagenesis
based on the protocol provided by Promega (Madison, WI). Mutation of
the designated tyrosine motif and the generation of the chimera were
verified by dideoxynucleotide sequence analysis. Stable transfections
were generated using the T cell hybridoma DC27.10 transfected with
human CTLA-4WT, mutant Y201F, CTLA-4CD28(SDYMNM), or CD28CTLA-4(GVYVKM)
plasmid together with pSVneo containing a neomycin resistance gene.
Electroporation was conducted at 260 V and 1600 µF. Cells were
selected with 1.5 mg/ml of G418 for 2 wk, and cells from different
populations were assayed for Ag expression by FACS, as described
(49).
Subcellular fractionation
DC27.10 cells were harvested, resuspended in 0.25 M sucrose, pH 7.4, homogenized, and centrifuged at 3000 rpm for 15 min to pellet the nuclei and unbroken cells. Cytoplasmic organelles from the postnuclear supernatant were separated by a Percoll/metrizamide discontinuous density gradient, as described (50). Lysosomal and lysosomal/mitochondrial membranes were obtained as described (51). The supernatants contained matrix proteins. The membranes were washed twice with 0.5 M NaCl and 0.1 M DTT, and then resuspended in PBS. A mixture of protease inhibitors was added to the lysosomal fractions before lysis (1 mM EDTA, 0.1 mM pepstatin A, 0.1 mM leupeptin, and 0.1 mM AEBSF). The purity of the isolated fractions was tested by immunoblotting using Abs for Golgi and lysosomes. In addition, the purity of lysosomes was verified by the lysosomal marker enzyme ß-hexosaminidase, as described (50). Protein determinations were performed using the Lowry assay (52). For precipitations, the amount of protein was about 900 µg, and for whole cell lysate generally 30 µg. In some experiments, cell fractionations were prepared following treatment of cells for 48 h with 20 mM ammonium chloride to prevent lysosomal degradation.
Peptide competition assay
Peptides were synthesized and HPLC purified by the Molecular Biology Core Facility (Dana-Farber Cancer Institute). The sequence of the peptides used was as follows: TTGVYVKMPPTE (unphosphorylated peptide), TTGVpYVKMPPTE (phosphorylated peptide). Purification of AP-1 and peptide photocross-linking analysis were conducted as described (42, 53).
Immunoprecipitation and immunoblotting
For immunoprecipitations, T cells were lysed in ice-cold lysis buffer containing 1% Triton X-100 (v/v) in 20 mM Tris-HCl, pH 8.3, 150 mM NaCl. The lysis buffer contained 1 mM PMSF, 1 mM Na4VO3, 10 mM NaF, and 1 mM Na4P2O7. For precipitations, 100 x 106cells/ml were generally used, while whole lysates are comprised of 1 x 106 cells/lane. Lysates were incubated for 20 min on ice before centrifugation at 15,000 x g for 15 min at 4°C. Postnuclear lysates were incubated for 1 h with agitation at 4°C with the indicated mAb. Protein A-Sepharose beads (30 µl; Pharmacia), swollen and washed in lysis buffer, were added and incubated for 1 h at 4°C. The beads were washed three times in cold lysis buffer, and proteins were eluted by boiling for 5 min in SDS sample buffer, separated by SDS-PAGE, and transferred to nitrocellulose for immunoblotting. The membranes were blocked with 5% milk in TBS (10 mM Tris-HCl, pH 7.6, 150 mM NaCl) and incubated with the indicated Ab. Bound Ab was revealed with HRP-conjugated rabbit anti-mouse or donkey anti-rabbit Abs or HRP-conjugated protein A/G using enhanced chemoluminescence (ECL; Amersham, Arlington Heights, IL).
Metabolic labeling
To monitor the synthesis of newly synthesized CTLA-4, CTLA-4 wild-type cells were treated with ionomycin (1 µM) or PMA (50 ng/ml). After metabolic labeling for 30 min with 1 mCi/ml [35S]methionine, cells were lysed, immunoprecipitated with CTLA-4 mAb or isotype-specific mAb, and analyzed by 10% SDS-PAGE.
|
Results |
|---|
|
TopAbstractIntroductionMaterials and MethodsResultsDiscussionReferences |
|---|
Stable CTLA-4 transfectants were generated from the mouse T cell
hybridoma DC27.10 using previously described approaches
(54). As seen in Fig. 1
A, human CTLA-4WT
(b) and CTLA-4 mutant Y201F (tyrosine at residue 201
was mutated to phenylalanine) (c) were expressed at moderate
levels on cell surface of the T cells. Generally, the CTLA-4 Y210F
mutant was expressed at slightly higher levels on the surface of cells,
consistent with the importance of the YVKM site in regulating surface
expression of the receptor (19).
|
-chain of the AP-1 complex. As seen in Fig. 1
B (upper left panel), anti-CTLA-4
precipitated a polypeptide at 95–100 kDa that was recognized by the
anti--chain antiserum (lane 3). A band of the
same size was also recognized in the cell lysate (lane
4). By contrast, anti-CD28 failed to precipitate the protein
(lane 2). As a negative control, anti-CTLA-4
failed to precipitate the -chain from nontransfected CTLA-4-negative
DC27.10 control cells (lane 1). As a positive
control, both CD28 and CTLA-4 precipitated PI 3-kinase as detected by
anti-p85 immunoblotting (Fig. 1B, middle
panel, lanes 2 and 3, respectively). PI
3-kinase binding to CD28 and CTLA-4 has been described (45, 57, 58). Recent studies using the two-hybrid system found that
CTLA-4 binds the µ-chain of the AP-2 complex (37, 43, 44, 45). Using
our transfectants and a specific anti-ß-chain antiserum,
anti-CTLA-4 was found to precipitate ß-subunits of both the AP-1
and AP-2 (Fig. 1B, lower panel, lane
3). The migration of the protein bands corresponding to the AP-1
and AP-2 ß-chains is identical to what has been previously documented
(37, 66). The interaction was specific in that no
association between CD28 and AP-1/AP-2 was observed (lane
2) and anti-CTLA-4 did not precipitate Ag from vector
transfectants (lane 1).
To verify that the CTLA-4/AP-1 interaction could occur in normal T
cells, peripheral T cells that had been activated for 48 h with
PHA and rhIL-2 were subjected to anti-CTLA-4 precipitation and
anti--chain immunoblotting (Fig. 1B, upper
right panel, lanes 5–7). Again, an AP-1 -chain was
readily detected in anti-CTLA-4 precipitates (lane
6), and in cell lysates (lane 7). These data
demonstrate that CTLA-4 associates in a specific manner with the AP-1
complex in T cells, a feature that distinguishes CTLA-4 from CD28. In
addition, coprecipitation of CTLA-4 with AP-1 is unusual, being the
first example of an AP-1 interaction with a protein of sufficient
avidity to withstand detergent and immunoprecipitation conditions.
Binding of AP-1 to the GVYVKM motif of CTLA-4 in T cells
Tyrosine-based endocytic motifs (tyrosine-polar-polar-large
hydrophobic) have been shown to interact with the µ-1 chain of the
AP-1 complex (39). CTLA-4 possesses a YVKM motif that
could meet this requirement (4). To assess this further, a
peptide competition assay was conducted using phosphorylated and
nonphosphorylated peptides encoding TTGVYVKMPPTE and a peptide
containing the UV-induced activation of cross-linker
benzoylphenylalanine (42, 53). Peptide YQTI corresponds to
the sequence of the complete 11-amino-acid cytoplasmic tail of lamp-1
containing the Y-motif YQTI. Under this regime, nonphosphorylated
peptide could compete in a range of 1–1000 µM for the association
between AP-1 and the photoreactive *YQTI peptide (Fig. 2A, lanes 7–10).
Densitometric readings of the exposed film showed a clear
dose-dependent competition between 1–100 µM peptide (right
panel). The absence of CTLA-4 peptide served as a positive control
(lanes 1 and 6). By contrast,
phosphorylated peptide did not dissociate the complex between 1–100
µM (lanes 3–5), while a competition was observed
at 1000 µM (lane 2). Whether the effect of this
high peptide concentration is related to a small contamination of
nonphosphorylated peptide is unclear. The concentration range of
peptide needed for effective displacement of *YQTI/AP-1 binding
corresponds to previously documented examples (42).
|
B, upper panel). DC27.10 cells were then stably
transfected with either human CD28, the CD28-CTLA-4 (GVYVKM) chimera,
hCTLA-4, or the CTLA-4-CD28(SDYMNM) chimera. Immunoprecipitation was
conducted using anti-CD28 or anti-CTLA-4 Abs, followed by
immunoblotting with an antiserum against AP-1 (-chain specific). As
shown in Fig. 2B (lower panel),
anti-CD28 and anti-CTLA-4 Abs precipitated the -chain of
AP-1 from the CD28-CTLA-4(GVYVKM)- and CTLA-4WT-transfected cells,
respectively (lanes 2 and 3). By contrast,
CD28WT and the CTLA-4-CD28(SDYMNM) chimera failed to precipitate the
protein (lanes 1 and 4). Therefore, the
GVYVKM motif alone provided minimal conditions for binding, albeit less
efficiently than WTCTLA-4. The ability of the GVYVKM motif alone to
confer AP-1 binding to the CD28 receptor provided further in vivo
evidence implicating this motif in the binding to AP-1.
Finally, to define the role of the tyrosine residue itself within
CTLA-4, Y-201 was mutated to phenylalanine within WTCTLA-4 and
investigated for AP-1 binding in T cells (Fig. 2C). In this
case, the loss of the tyrosine abrogated AP-1 binding
(lane 3 vs lane 4). Together, these data
demonstrate that the tyrosine-based YVKM motif serves as a binding site
for the AP-1 complex.
Intracellular localization of CTLA-4/AP-1 complexes
The majority of the AP-1 complexes have been shown in
well-documented studies to reside in the TGN (25, 30, 33).
This suggested that unlike CTLA-4 binding to intracellular Ags such as
the SHP-2 phosphatase and PI 3-kinase, CTLA-4 binding to AP-1 may occur
primarily in the Golgi compartment, the site of accumulated
intracellular CTLA-4. To address this further, intracellular CTLA-4 was
separated from surface CTLA-4 and compared for the presence of AP-1
binding (Fig. 3). Surface CTLA-4 was
distinguished from intracellular CTLA-4 by preincubating intact cells
with anti-CTLA-4 Ab at 4°C, followed by washing to remove excess
Ab, detergent lysis, and immunoprecipitation. Following this, lysates
were subjected to a round of preclearing with protein A-Sepharose,
followed by a second precipitation to purify intracellular CTLA-4. As
seen in Fig. 3 (upper panel), more than 90% of CTLA-4/AP-1
complexes were found in the intracellular lysate (lane
4), while anti-CTLA-4 precipitates from the cell surface
contained only a minor fraction (lane 3). Given the
limitations of the Ab-prebinding assay with some possible
redistribution of Ab in cell lysates, the significance of some 5–10%
AP-1 binding to cell surface-expressed CTLA-4 is unclear. As a control,
surface and intracellular CTLA-4/AP-1 was also precipitated from
untreated cells (lane 2). As expected, the
combination of surface and intracellular CTLA-4 precipitated AP-1,
which equaled the total precipitated by anti-CTLA-4 against
untreated cells (Fig. 3, lower panel). Anti-CD28
precipitates served as a negative control (lane 1).
These data show that the majority of CTLA-4/AP-1 complexes are located
in intracellular compartments of T cells.
|
CTLA-4 accumulates as an intracellular Ag in T cells
(19). AP-1 binding to CTLA-4 could play a role in this
event by regulating the shuttling of CTLA-4 to the lysosomal
compartments for degradation. AP-1 has been shown in several other
systems to regulate the trafficking of proteins such as the mannose
6-phosphate receptor to lysosomes (55, 59, 60, 61, 62). Initially,
wild-type and a form of CTLA-4 with a mutation at tyrosine 201 (which
disrupts the AP-1 binding site (Y201F)) were compared for localization
in various intracellular compartments. Intracellular membranes were
subfractionated on a Percoll/metrizamide discontinuous density gradient
to purify Golgi-enriched, mitochondrial, and lysosomal vesicles, as
described in Materials and Methods (63). In
addition to the Golgi-enriched fraction, two populations of lysosomes
were defined according to their density in the Percoll/metrizamide
gradients, one light and another heavier lysosomal fraction that
cosediments with the mitochondria. Immunoblotting with antisera against
a Golgi-specific 58-kDa protein confirmed the identity of the Golgi
fraction (Fig. 4A, upper
left panel, lane 1), and a marker for lysosomes lamp-2
was found in the two lysosomal fractions (lower
panel, lanes 2 and 3). In addition, the
purity of lysosomes was verified by the lysosomal marker enzyme
ß-hexosaminidase. As shown in Fig. 4A (upper right
panel), significant ß-hexosaminidase activity could be
demonstrated in both the light and heavy lysosomal fractions, with no
activity found in the cytosolic and Golgi-enriched fractions. We were
therefore successful in separating Golgi and lysosomes from each
other.
|
B, middle left panel,
lanes 2 and 3). Within the lysosomal fraction,
wild-type CTLA-4 was found primarily in the light lysosomal fraction
(Fig. 4B, middle panel, lane 3). By
contrast, AP-1 was found in the Golgi-enriched fraction and the cytosol
(Fig. 4B, lower left panel, lanes 1
and 2). Anti-CTLA-4 precipitation followed by
anti--chain immunoblotting showed that CTLA-4/AP-1 complexes
were detected exclusively in the Golgi-enriched fraction (Fig. 4B, upper left panel, lane 2).
To address whether the intracellular binding of CTLA-4 and AP-1
influenced the intracellular accumulation and processing of CTLA-4, the
steady state expression of wild-type CTLA-4 was compared to the mutant
Y201F in the Golgi membranes. Significantly, Y201F transfectant showed
a marked loss of CTLA-4 in the Golgi and lysosomal compartments (Fig. 4B, middle right panel, lanes 2 and
3). With the overexposure of the gel, it was possible to
detect only small amounts of intracellular material (lane
6 vs 5). Furthermore, as expected with the loss of
CTLA-4, anti-CTLA-4 coprecipitated no detectable AP-1 (Fig. 4B, upper right panel, lane 2). This
occurred despite the presence of normal levels of AP-1 in the
Golgi-enriched and cytosolic fractions of the Y201F-expressing cells
(Fig. 4B, lower panel, lanes 1 and
2). These findings indicate that the disruption of the AP-1
binding site in CTLA-4 resulted in the loss of CTLA-4 from the Golgi
and the lysosomes.
A difference in shuttling between the WTCTLA-4 vs the Y201F mutant to
the lysosomes was also apparent. In this case, cells were treated with
ammonium chloride to neutralize lysosomal pH and block degradation
(64). Only under these conditions could one begin to
visualize the Y201F mutant in the lysosomal compartment (Fig. 4C, lanes 5 and 6). Although untreated
cells showed no protein (Fig. 4C, lanes 1–3),
ammonium chloride treatment allowed for the detection of small amounts
of the mutant Y201F in both light and heavy lysosomal fractions
(lanes 5 and 6). Further enrichment and
separation of the matrix and membranes of the lysosomes showed receptor
binding primarily to the lysosomal membranes (lanes 7
and 9 vs 8 and 10). These data
demonstrate that the GVYVKM motif influences the accumulation of CTLA-4
in the Golgi and shuttling CTLA-4 from the Golgi to the lysosomes.
Pulse-chase experiments showed that the difference in the levels of
WTCTLA-4 vs Y201F mutant in the Golgi fraction was not due to
differences in rates of protein biosynthesis (data not shown). AP-1 is
therefore necessary for the efficient transport and degradation of
CTLA-4 to the lysosomal compartment.
Ca2+-dependent increase in cell surface expression
Accumulated intracellular CTLA-4 can be released from
intracellular stores to the cell surface by exposure to ionomycin, and
to a lesser extent, by phorbol ester and TCR ligation
(19, 20, 21). Our data combined with the data of others
(21) on the effect of the Y201F mutation on the
accumulation of CTLA-4 in the Golgi compartment support the idea that
the YVKM motif is needed for this retention. One possibility is
that AP-1 binding could also influence the release of CTLA-4 from the
Golgi to the cell surface. Unfortunately, dominant-negative forms of
the AP-1 complex are not available to test this question directly.
Instead, it was possible to assess whether the release of intracellular
CTLA-4 was accompanied by a loss of CTLA-4/AP-1 binding in cells. As
seen in Fig. 5A, the
Ca2+ ionophore ionomycin caused a time-dependent
increase in the surface expression of CTLA-4 (left
panel). Exposure to extracellular EGTA completely inhibited the
increase in surface expression, indicating a requirement for a
Ca2+ influx (Fig. 5A). The effect of
ionomycin in increasing surface expression of CTLA-4 was not the result
of a disruption of receptor endocytosis, as shown by internalization
and uptake studies (data not shown). Significantly, we found that this
increase in surface expression occurred in the absence of de novo
transcription or protein synthesis. Actinomycin D, a potent inhibitor
of transcription (65), had no effect on the increased
expression (left panel). Furthermore, pulse
labeling for 30 min with [35S]methionine at
various times following ionomycin treatment showed no change in the
synthesis of the protein (right panel, lanes
2–5). The ability of ionomycin to induce the release of receptor
without inducing an increase in protein synthesis allowed an
examination of changes in AP-1 binding upon the release of receptor to
the cell surface.
|
-chain of AP-1 (Fig. 5B). Interestingly,
during the time course of ionomycin treatment, no alteration in the
levels of AP-1 binding was observed (upper panel,
lanes 2–4; also see histogram). The same result was
observed over four separate experiments. As a control, equal amounts of
CTLA-4 receptor were found to be present in the anti-CTLA-4
precipitates (middle panel). Similarly, the level of AP-1 in
cells remained the same during the treatment (lower
panel). These data demonstrate that the release of intracellular
CTLA-4 to the cell surface is not accompanied by a detectable change in
AP-1/CTLA-4 binding. Increased CTLA-4 synthesis is accompanied by increased AP-1 binding and accumulation in the lysosomes
Because AP-1/CTLA-4 binding is not influenced by the release of
CTLA-4 to the cell surface, another issue was whether the interaction
could regulate steady state levels of intracellular receptor. Under
conditions in which there is an increase in CTLA-4 synthesis, but no
release of receptor to the cell surface, the cell would need a
mechanism to maintain steady state levels of the receptor in the Golgi.
Otherwise, the Golgi would become overburdened with newly synthesized
accumulated CTLA-4. From this analysis, we found that treatment of
cells with phorbol ester (PMA) caused a 2–3-fold increase in protein
synthesis, as detected by a pulse labeling with
[35S]methionine (Fig. 6A, right panel).
This occurred with only a moderate effect on surface expression (Fig. 6A, left panel). Significantly, the increased
synthesis was accompanied by a 2–3-fold increase in AP- 1 binding
(Fig. 6B, middle panel; see histogram). The
2–3-fold increase in binding matched the increase in protein
synthesis. At the same time, despite the increase in biosynthesis, as
shown in two experiments (Fig. 6B), the overall level of
intracellular CTLA-4 remained more or less the same, as detected by
anti-CTLA-4 immunoblotting (upper panel). This
maintenance of intracellular levels of receptor may be accounted for by
an increase in AP-1 binding and its ability to facilitate trafficking
to lysosomes for degradation. PMA also increased the expression of the
Y201F mutant, although its ultimate fate remains to be established.
|
C). In the
absence of ammonium chloride, both untreated and phorbol ester-treated
cells showed degradation of protein by 2–4 h (lanes
1–3 and 7–9). As expected from increased synthesis,
labeling of cells during the initial pulse was higher in phorbol
ester-treated cells than untreated cells (lane 7 vs
1). Exposure to ammonium chloride caused a moderate rescue
from degradation in untreated cells (lanes 4–6). By
contrast, the combination of phorbol ester and ammonium chloride caused
an increased accumulation of CTLA-4 in the lysosomes, which was
significantly higher than in nonstimulated cells (lanes
10–12 vs 4–6). These data are therefore consistent
with the hypothesis that the increase in CTLA-4 synthesis is
accompanied by a preferential increase in shuttling of newly
synthesized material to the lysosomal compartment. These observations
support a model in which increased synthesis of receptor is accompanied
by increased binding to AP-1 and an increase in its appearance in the
lysosomal compartment without an increase in the accumulated receptor
in the intracellular compartment. Overall, these observations are
consistent with a role for CTLA-4/AP-1 binding as a homeostatic
regulator of steady state levels of intracellular CTLA-4.
TCR/CD3 complex shares an ability to interact with AP-1 and AP-2
complexes
AP-1 and AP-2 failed to bind to the CD28 Ag, as assessed by
immunoblotting of anti-CD28 precipitates (Fig. 1), and two-hybrid
analysis (43, 44, 45, 46). However, one possibility is that the
adaptor complexes might bind to the TCR/CD3 complex in addition to
CTLA-4. Such an interaction could hypothetically act to bridge the
CTLA-4 with TCR in a manner that is not available to CD28. To assess a
possible interaction, anti-CD3 was used to precipitate protein
followed by blotting with the anti--chain antiserum (Fig. 7A, upper panel,
lane 2). From this, it is evident that a well-represented
-chain was present that comigrated with the same band precipitated
by anti-CTLA-4 (lane 4). A band of the same size
was also recognized in the cell lysate (lane 5). By
contrast, anti-CD28 failed to precipitate the protein
(lane 3). As an additional negative control,
anti-CTLA-4 also failed to precipitate the -chain from
nontransfected CTLA-4-negative DC27.10 control cells (lane
1). Similarly, using a specific anti-ß-chain antiserum,
anti-CD3 was found to precipitate ß-subunits of both the AP-1 and
AP-2 (Fig. 7B, lower panel, lane 2).
As discussed in Fig. 1, the position of the AP-1 and AP-2 ß-chains
has been previously well documented with the upper band corresponding
to AP-1 and the lower band to AP-2 ß-chains (37, 66).
For specificity, we were also interested in whether AP-1 specifically
bound to the TCR/CD3 complex in the Golgi compartment. As shown in Fig. 4 and described in Materials and Methods, intracellular
membranes were subfractionated on a Percoll/metrizamide discontinuous
density gradient to purify Golgi-enriched, mitochondrial, and lysosomal
vesicles (63). As in the case of CTLA-4,
anti-CD3-precipitated -chain was found restricted to the
Golgi-enriched fraction (Fig. 7B, lane 2). As
positive controls, AP-1 was found in the Golgi-enriched fraction,
cytosol, and to a smaller extent in the lysosomal fraction (Fig. 7B, lanes 5–7). These data indicate that CTLA-4
and the TCR/CD3 complex share the property of being associated with
the AP-1 and AP-2 complexes in T cells.
|
|
Discussion |
|---|
|
TopAbstractIntroductionMaterials and MethodsResultsDiscussionReferences |
|---|
). Peptide competition and in vivo
expression of chimeras further showed that AP-1 binds to the
nonphosphorylated GVYVKM motif within the CTLA-4 cytoplasmic tail. AP-1
binding appears linked to its classical role in regulating the
shuttling of proteins from the Golgi compartment to lysosomes for
degradation (24, 25, 26). Consistent with this function,
phorbol ester-induced increases in CTLA-4 synthesis were accompanied by
an increase in AP-1 binding, and a concomitant increase in the
appearance of [35S]methionine pulse-labeled
CTLA-4 in neutralized lysosomes. At the same time, the overall level of
intracellular CTLA-4 remained the same despite the increased rate of
receptor synthesis. Together, these observations support a role for
AP-1/CTLA-4 binding in the maintenance of steady state levels of CTLA-4
in the intracellular compartments of T cells. Finally, we demonstrate
that the TCR/CD3 complex also binds to AP-1 and AP-2 complexes,
providing a possible link between these two receptors in the regulation
of T cell function.
The interpretation of the data was complicated by the fact that the
AP-1 binding site (i.e., the GVYVKM site) also plays a role in
regulating the release of the receptor from the Golgi to the cell
surface (19, 20, 67). This latter observation was clearly
supported by the observation of an almost complete loss of the Y201F
mutant in purified Golgi-enriched membranes of T cells (Fig. 4). One
possibility was that the AP-1 complex binding is responsible for this
retention, a possibility that is difficult to establish without a
dominant-negative form of AP-1 for testing in transfection studies.
However, with the limited tools presently available, we could find no
evidence to implicate AP-1 in retaining CTLA-4 in the Golgi. Instead,
an ionomycin-induced Ca2+ influx was observed to
induce surface expression of the receptor without having an effect on
the binding of AP-1 to the receptor. If AP-1 controlled the release of
the receptor to the cell surface, one would have expected a decreased
association between AP-1 and the receptor. Future studies will be
needed to determine the nature of the YVKM-recognizing protein that
regulates the surface expression of the CTLA-4 molecule. Candidate
proteins include PI 3-kinase, SHP-2, and another adaptor complex,
AP-3. Instead, our data are most compatible with the well-documented
role of the AP-1 complex in sorting proteins in the TGN to the
endosomal/lysosomal pathway for degradation. Consistent with this,
AP-1/CTLA-4 complexes were found primarily in the Golgi-enriched
membrane fraction, the site of shuttling to the lysosomes. Moreover,
increases in CTLA-4 synthesis induced by phorbol ester and monitored by
[35S]-methionine labeling were reproducibly
accompanied by a 2–3-fold increase in CTLA-4 binding to AP-1 (Fig. 6).
This in turn was followed by a marked increase in the accumulation of
CTLA-4 in the purified lysosomal compartment, as monitored by
[35S]methionine labeling under conditions in
which the degradation was blocked. At the same time, although there was
an increase in synthesis, there was no increase in the overall level of
intracellular receptor, thus supporting the view that AP-1 serves
to maintain intracellular CTLA-4 in defined level. Pulse-chase
comparisons between WTCTLA-4 and Y201F mutant also showed a more rapid
degradation of the WT receptor (data not shown). Without a mechanism to
maintain steady state levels of receptor, the Golgi would become
overburdened with newly synthesized CTLA-4. This might prevent
leakiness in the expression of CTLA-4 at the cell surface. AP-1 binding
to CTLA-4 would therefore act to maintain intracellular levels of
CTLA-4 in the Golgi for its eventual release to the cell surface upon
TCR/CD3 ligation. Small alterations in the levels of CTLA-4
expression could have major consequences on T cell function. For
example, anti-CTLA-4 Abs can block proliferation as early as 6
h postactivation, where the levels of CTLA-4 expression are too low to
be detected by FACS analysis (17).
The exact nature of the AP-1 subunit responsible for binding the CTLA-4GVYVKM motif remains to be clarified. Our in vitro and in vivo analysis showed binding to intact AP-1 tetrameric complexes. Previous studies using the two-hybrid screen failed to detect µ-1 binding to the CTLA-4 cytoplasmic domain, suggesting that other subunits are involved (43, 44, 45, 46). Other chains in the complex are capable of binding proteins, such as in the case of the ß-chain that was recently reported to bind CD4 (53). Future studies will be needed to identify the binding subunit.
Another feature of interest is the ability of AP-1 to form a stable biochemical complex in detergent lysates. This complex is the first example of binding between the AP-1 complex and a protein or receptor that can withstand exposure to detergent. In this context, a comparison can be made with AP-2 complex binding to the epidermal growth factor receptor (40). In both cases, some 1.5–3% of AP complexes remain stably associated with the receptor in detergent. In the case of the epidermal growth factor receptor, despite the low stoichiometry of binding as estimated in the formation of detergent-resistant complexes, AP-2 is widely accepted as the major mechanism by which the receptor is removed from the cell surface (40). A similar stoichiometry of binding was noted in our transfectants in the case of AP-2/CTLA-4 binding (data not shown). Detergent sensitive AP complexes are known to regulate the trafficking of other proteins. For example, recognition motifs have been identified in the two-hybrid system, but the integrity of these complexes is lost during the biochemical purification (39, 40, 41, 42).
Finally, we demonstrate that CTLA-4 and the TCR/CD3 complex share an
ability to bind to the AP-1 and AP-2 complexes, thus providing a
potential link between these two receptors (Fig. 7). AP-1 binding to
the TCR/CD3 complex was also found localized to the Golgi apparatus.
The binding mechanism is unknown, but may involve dileucine-based
motifs that are present in CD3 subunits and bind to the ß-chain of
AP-1 (53). The shared use of AP-1 and AP-2 with binding to
potentially distinct binding sites could facilitate an interaction
between these receptors, both in the Golgi and the cell surface. AP
complexes in turn have been noted to bind to each other
(68). This could provide a mechanism to favor a CTLA-4
association with the TCR relative to CD28. This in turn could help
facilitate the association between CTLA-4 and the TCR/CD3 complexes
described by others (14). In addition, the shared AP-2
binding suggests a mechanism by which CTLA-4 might alter the
redistribution of the TCR/CD3 complex needed for appropriate signaling.
We are presently exploring the possible function of AP-1/AP-2
complex binding to the two receptors in T cell signaling.
|
Acknowledgments |
|---|
|
Footnotes |
|---|
2 Abbreviations used in this paper: CTLA-4, cytolytic T lymphocyte-associated Ag; PI, phosphatidylinositol; TGN, trans-Golgi network; WT, wild-type.
Received for publication February 26, 1999. Accepted for publication June 7, 1999.
|
References |
|---|
|
TopAbstractIntroductionMaterials and MethodsResultsDiscussionReferences |
|---|
promoter: an efficient and versatile mammalian cDNA expression system composed of the simian virus 40 early promoter and the R-U5 segment of human T cell leukemia virus type 1 long terminal repeat. Mol. Cell. Biol. 8:466.This article has been cited by other articles:
|
|
 |
|
  M. Berg and N. Zavazava Regulation of CD28 expression on CD8+ T cells by CTLA-4 J. Leukoc. Biol., April 1, 2008; 83(4): 853 - 863. [Abstract] [Full Text] [PDF]
|
|
|
|
 |
|
 B. Wei, S. da Rocha Dias, H. Wang, and C. E. Rudd CTL-Associated Antigen-4 Ligation Induces Rapid T Cell Polarization That Depends on Phosphatidylinositol 3-Kinase, Vav-1, Cdc42, and Myosin Light Chain Kinase J. Immunol., July 1, 2007; 179(1): 400 - 408. [Abstract] [Full Text] [PDF]
|
|
|
|
|
|
D. Bernard, B. Riteau, J. D. Hansen, R. B. Phillips, F. Michel, P. Boudinot, and A. Benmansour Costimulatory Receptors in a Teleost Fish: Typical CD28, Elusive CTLA4 J. Immunol., April 1, 2006; 176(7): 4191 - 4200. [Abstract] [Full Text] [PDF]
|
|
|
|
 |
|
 M. Lundholm, V. Motta, A. Lofgren-Burstrom, N. Duarte, M.-L. Bergman, S. Mayans, and D. Holmberg Defective Induction of CTLA-4 in the NOD Mouse Is Controlled by the NOD Allele of Idd3/IL-2 and a Novel Locus (Ctex) Telomeric on Chromosome 1 Diabetes, February 1, 2006; 55(2): 538 - 544. [Abstract] [Full Text] [PDF]
|
|
|
|
 |
|
 H. Schneider, E. Valk, S. da Rocha Dias, B. Wei, and C. E. Rudd CTLA-4 up-regulation of lymphocyte function-associated antigen 1 adhesion and clustering as an alternate basis for coreceptor function PNAS, September 6, 2005; 102(36): 12861 - 12866. [Abstract] [Full Text] [PDF]
|
|
|
|
|
|
K. I. Mead, Y. Zheng, C. N. Manzotti, L. C. A. Perry, M. K. P. Liu, F. Burke, D. J. Powner, M. J. O. Wakelam, and D. M. Sansom Exocytosis of CTLA-4 Is Dependent on Phospholipase D and ADP Ribosylation Factor-1 and Stimulated during Activation of Regulatory T Cells J. Immunol., April 15, 2005; 174(8): 4803 - 4811. [Abstract] [Full Text] [PDF]
|
|
|
|
|
|
A. L. Szymczak and D. A. A. Vignali Plasticity and Rigidity in Adaptor Protein-2-Mediated Internalization of the TCR:CD3 Complex J. Immunol., April 1, 2005; 174(7): 4153 - 4160. [Abstract] [Full Text] [PDF]
|
|
|
|
 |
|
 L. A. Yi, S. Hajialiasgar, and E. Chuang Tyrosine-mediated inhibitory signals contribute to CTLA-4 function in vivo Int. Immunol., April 1, 2004; 16(4): 539 - 547. [Abstract] [Full Text] [PDF]
|
|
|
|
|
|
M. A. Putnam, A. E. Moquin, M. Merrihew, C. Outcalt, E. Sorge, A. Caballero, T. A. Gondre-Lewis, and J. R. Drake Lipid Raft-Independent B Cell Receptor-Mediated Antigen Internalization and Intracellular Trafficking J. Immunol., January 15, 2003; 170(2): 905 - 912. [Abstract] [Full Text] [PDF]
|
|
|
|
|
|
L. Yin, H. Schneider, and C. E. Rudd Short cytoplasmic SDYMNM segment of CD28 is sufficient to convert CTLA-4 to a positive signaling receptor J. Leukoc. Biol., January 1, 2003; 73(1): 178 - 182. [Abstract] [Full Text] [PDF]
|
|
|
|
 |
|
 M. P. Pistillo, P. L. Tazzari, G. L. Palmisano, I. Pierri, A. Bolognesi, F. Ferlito, P. Capanni, L. Polito, M. Ratta, S. Pileri, et al. CTLA-4 is not restricted to the lymphoid cell lineage and can function as a target molecule for apoptosis induction of leukemic cells Blood, January 1, 2003; 101(1): 202 - 209. [Abstract] [Full Text] [PDF]
|
|
|
|
|
|
H. Schneider, D. A. Mandelbrot, R. J. Greenwald, F. Ng, R. Lechler, A. H. Sharpe, and C. E. Rudd Cutting Edge: CTLA-4 (CD152) Differentially Regulates Mitogen-Activated Protein Kinases (Extracellular Signal-Regulated Kinase and c-Jun N-Terminal Kinase) in CD4+ T Cells from Receptor/Ligand-Deficient Mice J. Immunol., October 1, 2002; 169(7): 3475 - 3479. [Abstract] [Full Text] [PDF]
|
|
|
|
 |
|
 S. Chikuma and J. A. Bluestone CTLA-4: Acting at the Synapse Mol. Interv., July 1, 2002; 2(4): 205 - 208. [Abstract] [Full Text] [PDF]
|
|
|
|
|
|
A. Nakajima, H. Iijima, M. F. Neurath, T. Nagaishi, E. E. S. Nieuwenhuis, R. Raychowdhury, J. Glickman, D. M. Blau, S. Russell, K. V. Holmes, et al. Activation-Induced Expression of Carcinoembryonic Antigen-Cell Adhesion Molecule 1 Regulates Mouse T Lymphocyte Function J. Immunol., February 1, 2002; 168(3): 1028 - 1035. [Abstract] [Full Text] [PDF]
|
|
|
|
|
|
S. da Rocha Dias and C. E. Rudd CTLA-4 blockade of antigen-induced cell death Blood, February 15, 2001; 97(4): 1134 - 1137. [Abstract] [Full Text] [PDF]
|
|
|
|
|
|
T. Iida, H. Ohno, C. Nakaseko, M. Sakuma, M. Takeda-Ezaki, H. Arase, E. Kominami, T. Fujisawa, and T. Saito Regulation of Cell Surface Expression of CTLA-4 by Secretion of CTLA-4-Containing Lysosomes Upon Activation of CD4+ T Cells J. Immunol., November 1, 2000; 165(9): 5062 - 5068. [Abstract] [Full Text] [PDF]
|
|
|
|
|
|
A. Qadri, C. G. Radu, J. Thatte, P. Cianga, B. T. Ober, R. J. Ober, and E. S. Ward A Role for the Region Encompassing the c'' Strand of a TCR V{alpha} Domain in T Cell Activation Events J. Immunol., July 15, 2000; 165(2): 820 - 829. [Abstract] [Full Text] [PDF]
|
|
|
|
|
|
E. L. Masteller, E. Chuang, A. C. Mullen, S. L. Reiner, and C. B. Thompson Structural Analysis of CTLA-4 Function In Vivo J. Immunol., May 15, 2000; 164(10): 5319 - 5327. [Abstract] [Full Text] [PDF]
|
|
|
|
|
|
T. Cinek, A. Sadra, and J. B. Imboden Cutting Edge: Tyrosine-Independent Transmission of Inhibitory Signals by CTLA-4 J. Immunol., January 1, 2000; 164(1): 5 - 8. [Abstract] [Full Text] [PDF]
|
|
| ||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||
| HOME | HELP | FEEDBACK | SUBSCRIPTIONS | ARCHIVE | SEARCH | TABLE OF CONTENTS |
), ionomycin plus actinomycin D (1 µM,
), or ionomycin plus EGTA (5 mM, 
) for increasing periods of time
at 37°C. Following incubation with the indicated reagents, cells were
washed and incubated with PE-labeled CTLA-4 mAb at 4°C for 30 min.
Cells stained with PE-labeled isotype-specific mAb served as a negative
control. Expression levels of CTLA-4 were analyzed by FACS.
Right panel, CTLA-4 wild-type cells were treated with
ionomycin (1 µM) for the indicated periods of time at 37°C.
Following this treatment, cells were incubated for 1 h in
methionine-free medium and then metabolically labeled with
[35S]methionine for 30 min. Cells were lysed,
immunoprecipitated with CTLA-4 mAb (lanes 2–5) or
isotype-specific mAb (lane 1), and analyzed by 10%
SDS-PAGE. B, CTLA-4/AP-1 binding is not altered by
ionomycin. CTLA-4 wild-type cells treated with ionomycin (1 µM) for
the indicated periods of time were lysed, immunoprecipitated with
CTLA-4 mAb (lanes 2–4) or isotype-specific mAb
(lane 1), and subjected to immunoblotting with
anti-
