| HOME | HELP | FEEDBACK | SUBSCRIPTIONS | ARCHIVE | SEARCH | TABLE OF CONTENTS |
|
|
|
|||||||
|
|||||||||
Departments of
*
Medicinal Chemistry and Molecular Pharmacology and
Biological Sciences, Purdue University, West Lafayette, IN 47907
 |
Abstract |
|---|
 Top Abstract IntroductionMaterials and MethodsResultsDiscussionReferences |
|---|
|
Introduction |
|---|
|
TopAbstractIntroductionMaterials and MethodsResultsDiscussionReferences |
|---|
(CD79a) and Ig-
(CD79b) subunits (4, 5). The Syk-receptor interaction is initiated by the
phosphorylation of conserved tyrosines present within immunoreceptor
tyrosine-based activation motifs (ITAMs) located within the cytoplasmic
domains of Ig- and Ig- (6, 7). This ITAM
phosphorylation creates docking sites for the tandem pair of Syk SH2
domains. Binding to the phosphorylated ITAM and subsequent tyrosine
phosphorylation leads to the activation of Syk. Syk then couples the
BCR to multiple downstream signaling pathways that are stimulated
following B cell activation, including the mobilization of
intracellular stores of calcium, activation of the mitogen-activated
protein kinase cascade, and generation of phosphatidylinositide
3-phosphates (7, 8, 9, 10). A Src family kinase such as Lyn, which is expressed predominately in hemopoietic cells, is required for the efficient activation of Syk following BCR aggregation. It probably plays two important roles in this regard: the phosphorylation of ITAM tyrosines and the phosphorylation of Syk itself on tyrosines located within the activation loop. This general model for the early events in BCR-mediated signaling is analogous to that proposed for signaling through the TCR and its associated kinases, Lck and ZAP-70 (11, 12). In T cells, Lck is needed for both receptor-ITAM phosphorylation and ZAP-70 phosphorylation and activation. However, several studies have found fundamental differences between ZAP-70 and Syk and their reliance on Src family kinases for activation. For example, Syk, but not ZAP-70, can restore TCR-stimulated signaling to Jurkat T cells deficient in Lck (13, 14). Also, recent genetic studies have shown that, in many contexts, the expression of Lyn has a negative effect on signaling in B cells. For example, certain BCR-stimulated events such as inositol 1,4,5-trisphosphate production and activation of the serine/threonine kinase, Akt/protein kinase B, are actually more robust in DT40 B cells that lack Lyn, the only Src family kinase expressed in these cells at a detectable level (1, 15). Likewise, IgM cross-linking in primary B cells from Lyn knockout mice leads to hyperactivation of the mitogen-activated protein kinase and c-Jun kinase pathways and a hyperproliferative response (16, 17, 18).
The recruitment of Syk to the aggregated BCR has not been visualized in intact cells, and the biochemical demonstration of this is difficult due to the association of the aggregated receptor with cytoskeletal components, which hampers the isolation of intact BCR-Syk complexes in high yield. To approach this problem, we generated cDNAs coding for a chimeric molecule of Syk fused with green fluorescent protein (GFP). We then used this construct as a tool for visualizing the effects of BCR cross-linking on the redistribution of Syk within cells that either express or do not express endogenous Lyn. In this study we demonstrate that Syk is present in nuclear and cytoplasmic compartments and is recruited from both to the site of the aggregated BCR. Syk is associated with receptors aggregated to form both membrane patches and polar caps. The recruitment of Syk to the BCR is most efficient when Syk retains catalytic activity. Lyn is not required for the interaction of Syk with theaggregated Ag receptor. We demonstrate further that Lyn and Syk differentially regulate the dynamics of receptor internalization. Syk stabilizes BCR signaling complexes at the plasma membrane, while Lyn is required for receptor internalization.
|
Materials and Methods |
|---|
|
TopAbstractIntroductionMaterials and MethodsResultsDiscussionReferences |
|---|
An XhoI/HpaI DNA fragment encoding Syk was cut from EPB:SykMyc (19). Insertion of this fragment into XhoI/SmaI sites of the pEGFP-N2 vector (Clontech, Palo Alto, CA) resulted in a fusion of Syk and enhanced GFP (SykEGFP). The same approach was used to add the open reading frame of enhanced GFP (EGFP) to the C-terminus of a catalytically inactive form of Syk (Syk(K396R)) (20) to generate KDEGFP. Two oligonucleotides (5'-GCTCTGCAGAAAAAGTTGGAAGAGCTTGAGCTGGATGAGC-3' and 5'-GCTCATCCAGCTCAAGCTCTTCCAACTTTTTCTGCAGAGCCT-3'), which encode the nuclear export signal from Xenopus mitogen-activated protein kinase kinase (21), were synthesized, annealed, and inserted in-frame into BsrGI/NotI sites at the 3' end of the coding sequence for SykEGFP to make a construct for the expression of pSykEGFPNES. Syk-deficient (1) or Syk- and Lyn-deficient (Syk/Lyn-deficient) (22) chicken DT40 B cells (1 x 106) were transfected with 20 µg of each DNA construct as previously described (20). Human DG75 B cells were electroporated using 20 µg of each DNA construct at 300 V and 800 µF. Cells were analyzed 24–36 h post-transfection. DT40 B cell lines were obtained from Drs. Tomohiro Kurosaki (Kansai Medical University, Osaka, Japan) and Ellen Puré (The Wistar Institute, Philadelphia, PA).
Fluorescence and confocal microscopy
Unstimulated cells were adhered to coverslips precoated with
poly-L-lysine (100 µg/ml; Sigma, St. Louis, MO) for 10
min at room temperature. Cells stimulated by either goat
anti-chicken IgM Abs (Bethyl Laboratories, Montgomery, TX) or Texas
Red-conjugated anti-IgM Abs were adhered to coverslips 5–10 min
before the end of stimulation at 37°C. Texas Red was conjugated to
goat anti-chicken IgM using the FluoReporter Texas Red-X protein
labeling kit from Molecular Probes (Eugene, OR). Cells were fixed in
3.7% paraformaldehyde in PBS for 15 min at room temperature, washed
three times with PBS, stained with 4,6-diamidinao-2-phenylindole
(Sigma) and viewed by fluorescence microscopy. Cells undergoing
subsequent staining with anti-human Ig- Ab were preincubated
with blocking buffer (3% BSA, 1% goat serum in PBS) for 30 min
followed by incubation with R-PE-conjugated monoclonal anti-human
Ig- Ab (Ancell, Bayport, MN) diluted in the blocking buffer for
1 h. Conventional fluorescence was examined using an Olympus
BH2-RFCA fluorescence microscope with x60 objective equipped with a
Sony DXC-950 3CCD color video camera and Northern Eclipse 5.0 software
from Empix Imaging (Mississauga, Canada). Laser scanning microscopy was
performed with a Nikon Optiphot microscope (Melville, NY) and a Bio-Rad
MRC 1024 system (Hercules, CA) using an argon/krypton laser and an
E2/UBHS filter set. The reported percentages of cells responding to a
given stimulus represent data from a typical experiment and include an
examination in each case of >100 cells. Similar data were obtained in
two additional trials. In one experiment cells expressing SykEGFPNES
were pretreated with leptomycin B (23) at a concentration
of 10 ng/ml for 60 min before fixation and examination by fluorescence
microscopy. Leptomycin B was a gift from Dr. Minoru Yoshida (University
of Tokyo, Tokyo, Japan).
For real-time observation of live cells, 1 x 106 transfected cells were adhered to glass coverslips. Immediately after exposure to RPMI medium containing 20 µg/ml of goat anti-IgM Ab, cells were observed under the microscope, and a picture was taken as time zero of stimulation. The cells remained in the presence of anti-IgM Abs and were photographed at timed intervals.
Promoter-linked luciferase assays
Syk-deficient DT40 cells (1 x 107) were transfected by electroporation with vectors containing cDNAs for the various EGFP- or epitope-tagged forms of Syk or Syk mutants (20 µg) and an NF-AT-luciferase reporter plasmid (10 µg) (20). Cells were harvested 36 h following transfection, plated at a density of 1 x 106 cells/ml, and activated with anti-IgM Abs (10 µg/ml) or a mixture of PMA (50 ng/ml) and ionomycin (1.0 µM) for 6 h at 37°C. Luciferase activity was determined using the luciferase assay system kit (Promega, Madison, WI). Luciferase activity is expressed as a fraction of that activity observed with activation by PMA plus ionomycin.
Western blot analysis
For the detection of fusion proteins by Western blot analysis, proteins in extracts from resting or activated cells were separated by SDS-PAGE, transferred to polyvinylidene difluoride membranes, and probed with an anti-GFP Ab (Clontech) and an HRP-conjugated secondary Ab. The Ag-Ab complex was visualized by an enhanced chemiluminescence detection system. For the detection of epitope-tagged Syk, proteins were first immunoprecipitated from cell lysates with an anti-Myc mAb and then probed by Western blotting with anti-Syk Abs (24). The 9E10 anti-Myc hybridoma cell line was obtained from American Type Culture Collection (Manassas, VA).
|
Results |
|---|
|
TopAbstractIntroductionMaterials and MethodsResultsDiscussionReferences |
|---|
To prepare a form of Syk that could be visualized in intact cells,
we constructed a mammalian expression vector with the coding sequence
for a form of GFP with enhanced fluorescent properties (EGFP) inserted
downstream from the cDNA coding for wild-type murine Syk (SykEGFP). A
schematic diagram showing the orientation of the two proteins within
the chimera is shown in Fig. 1
A. The resulting cDNA was
transfected into Syk-deficient DT40 B cells. Immunoblotting analysis
using anti-GFP Abs confirmed the expression of the fusion protein
only in cells transfected with the plasmid containing the cDNA coding
for SykEGFP (Fig. 1B). A single protein was detected that
migrated at Mr 100,000, the expected
size of the intact fusion protein. Smaller, proteolytic fragments of
SykEGFP containing the GFP domain were not detected.
|
C). The extent of NF-AT
activation observed in cells expressing SykEGFP and Syk(WT) was
comparable over a wide range of concentrations of transfected plasmid
and a wide range of concentrations of activating Ab (data not shown).
The extent of receptor-mediated activation of NF-AT in DT40 cells
expressing Syk(WT) or SykEGFP is comparable to that observed in
wild-type DT40 cells (data not shown). As a control, a catalytically
inactive form of SykEGFP was prepared with a point mutation at a
critical lysine residue in the active site (KDEGFP). The expression of
KDEGFP was unable to rescue signaling in the Syk-deficient cells (Fig. 1C). Likewise, the expression of EGFP alone was unable to
rescue signaling (Fig. 1C). These studies indicate that
SykEGFP can replace endogenous Syk for coupling the BCR to the
activation of NF-AT. Localization of SykEGFP in DT40 B cells
To examine the intracellular localization of the fusion protein,
we transfected Syk-deficient DT40 B cells with the SykEGFP expression
plasmid. Cells were then fixed and examined by both fluorescence and
confocal microscopy. A composite image is shown in Fig. 2. The Syk fusion protein was localized
throughout the interior of the cell in both the cytoplasm and nucleus,
but was excluded from nucleoli (Fig. 2, A and B).
This same distribution was observed for the catalytically inactive
KDEGFP (Fig. 2, C and D), indicating that the
presence or the absence of catalytic activity did not affect
localization of the kinase in resting cells. SykEGFP and KDEGFP
appeared in both the nucleus and cytoplasm of cells, whether expressed
at abundant levels or in trace amounts.
|
, E and F; the location of the nucleus was verified
in all cells by staining with 4,6-diamidinao-2-phenylindole; data not
shown). The staining pattern observed by fluorescence or confocal
microscopy of cells expressing SykEGFPNES was clearly distinct from
that observed for cells expressing SykEGFP or KDEGFP. The exclusion of
SykEGFPNES from the nucleus could be reversed by treatment of the cells
with leptomycin B, a specific inhibitor of NES-mediated nuclear export
(23) (data not shown). This result indicated that the
peptide that was added to the carboxyl terminus of SykEGFP to form
SykEGFPNES was indeed functioning as a nuclear export signal (data not
shown). This nuclear-excluded form of Syk was able to restore the
BCR-stimulated activation of NF-AT to Syk-deficient cells as
efficiently as SykEGFP or Syk(WT) (Fig. 1C). Redistribution of SykEGFP to the cross-linked BCR
BCR cross-linking is thought to promote the physical association
of SykEGFP with the receptor. To confirm this, we treated Syk-deficient
DT40 cells expressing SykEGFP with affinity-purified anti-IgM Abs
conjugated with Texas Red dye to allow visualization of the surface IgM
component of the BCR by confocal microscopy. In untreated cells, IgM
was uniformly distributed around the cell surface (Fig. 3C), and there was no obvious
colocalization of SykEGFP with the receptor (Fig. 3, A and
B). When cells were stimulated by the addition of the
fluorescently tagged anti-IgM Abs, the surface IgM first formed
membrane patches of variable sizes (Fig. 3F). The patching
of cell surface receptors was more easily observed at 25 than at 37°C
because the redistribution of the patches into caps occurred more
slowly. When the patches of cell surface IgM formed, SykEGFP also began
to form aggregates (Fig. 3D). These colocalized with the
patched IgM (Fig. 3E), consistent with recruitment of
SykEGFP to the clustered receptors. By 20 min at 37°C, the patched
receptors had aggregated to form a cap at one pole of the cell (Fig. 3I). SykEGFP was also redistributed to a single cap that
colocalized with the capped surface Ig (Fig. 3, G and
H). Further incubation of the stimulated cells resulted in
the appearance of IgM in small aggregates that were present in the
interior of the cell due to internalization of the IgM in endocytic
vesicles (Fig. 3L). At this stage, there was no obvious
accumulation of SykEGFP at the sites of internalized IgM (Fig. 3J).
|
mAb and a
tetramethylrhodamine isothiocyanate-conjugated secondary Ab. Human B
cells were used for this experiment due to the availability of
anti-human Ig- Abs. After a 30-min stimulation, Ig- was
clustered near one pole of the cell (Fig. 3O). A portion of
the SykEGFP was also clustered and colocalized with Ig- (Fig. 3, M and N). In these anti-IgM-stimulated DG75
cells, both SykEGFP and Ig- also colocalized with surface IgM (data
not shown). These data are consistent with the recruitment of SykEGFP
to the site of the clustered BCR complex.
Several control experiments were conducted to further confirm the
receptor-mediated relocalization of SykEGFP into membrane-associated
patches and caps. To confirm that the redistribution of SykEGFP
occurred in this fashion within individual cells, we examined changes
in the localization of SykEGFP following BCR cross-linking in live
cells as a function of time. Syk-deficient DT40 cells expressing
SykEGFP were treated with anti-IgM Abs at room temperature and then
examined by fluorescence microscopy at various time points without
fixation to allow events in single cells to be monitored (Fig. 4, A–D). In most cells,
SykEGFP initially redistributed to the plasma membrane, moved to one
hemisphere of the cell, and eventually formed a single cap at one pole.
This was especially prominent in cells expressing lower levels of
SykEGFP. The redistribution of the fusion protein was more difficult to
visualize in cells with high levels of expression of SykEGFP,
suggesting that the membrane-associated binding sites to which SykEGFP
was recruited were saturable. SykEGFP translocation to the plasma
membrane could also be observed in live, SykEGFP-expressing, DG75 B
cells that were then treated with anti-human IgM Abs (data not
shown).
|
, E and F). Its
distribution, however, remained unchanged following BCR aggregation
when examined at multiple time points following receptor engagement
(shown here for a 30-min treatment at room temperature). Secondly,
Syk-deficient DT40 B cells expressing SykEGFP were treated with Con A
to examine whether the redistribution of SykEGFP was a general
phenomenon that accompanied the clustering and capping of a cell
surface receptor other than IgM. Following a 1-h treatment with Con A,
Con A receptors were aggregated into a single cap without a concomitant
redistribution of SykEGFP (Fig. 4, G and H).
Thus, the redistribution of SykEGFP was specific for both the Syk
portion of the fusion protein and the cross-linking of surface
Ig.
Finally, to determine the fraction of transfected cells that responded
to treatment with anti-IgM Abs through the redistribution of
SykEGFP, we monitored the changes in SykEGFP localization in response
to receptor cross-linking in populations of SykEGFP-expressing cells as
a function of time after B cell activation. Cells were classified into
one of five categories depending on the distribution of SykEGFP in the
cell (Fig. 5A). These
categories were: resting, a distribution of SykEGFP throughout the
cytoplasm and nucleus; patched (I), patches of SykEGFP appearing at the
plasma membrane before capping; capped, SykEGFP appearing in a single
polar cap occupying less than one-quarter of the cell surface; patched
(II), only a small number of brightly staining clusters of SykEGFP
still present at the plasma membrane after dissociation of the tight
cap structure; and nuclear excluded, diffuse distribution of SykEGFP
throughout the cytoplasm, but excluded from the nucleus. In >98% of
unstimulated cells, SykEGFP exhibited a diffuse distribution throughout
the cell corresponding to the resting phenotype (Fig. 5B).
After 5 min of treatment with anti-IgM Abs at 37°C, SykEGFP was
present in membrane patches (58%) or caps (33%) in the majority of
the cells. After BCR ligation for 10 min, SykEGFP appeared
predominately in cell surface caps (85%). By 1 h poststimulation,
most of the SykEGFP was absent from the plasma membrane, except for a
few brightly staining clusters (patched, II). These were still
associated with the few small clusters of IgM (as determined by
staining with Texas Red-conjugated anti-IgM Abs, data not shown) that
had not undergone internalization. By 3 h after stimulation,
SykEGFP was no longer observed in clusters in the majority of cells and
was once again present in a diffuse staining pattern. In approximately
50% of these cells, however, SykEGFP was confined to the cytoplasm and
excluded from the nucleus. By 6 h, 90% of the cells had returned
to the diffuse SykEGFP distribution pattern that characterized the
unstimulated cells.
|
Lyn is thought to be an important kinase acting upstream of Syk to
promote its recruitment to the BCR. To determine whether Lyn expression
was required for the redistribution of SykEGFP to the cross-linked BCR,
we expressed the fusion protein in DT40 B cells in which either the
gene for Syk (Syk-deficient) or the genes for both Syk and Lyn
(Syk/Lyn-deficient) had been disrupted. In unstimulated cells, the
intracellular distribution of SykEGFP in Syk/Lyn-deficient cells was
indistinguishable from that described above for SykEGFP expressed in
Syk-deficient cells (data not shown). Following a 10-min treatment with
activating Ab at 37°C, SykEGFP was recruited to the plasma membrane
in both cell types and appeared in most cells as a single cap at the
cell surface (Fig. 6, A and
C) that colocalized with the capped surface IgM (data not
shown). An analysis of multiple transfected cells indicated that the
initial cocapping of SykEGFP with the BCR was similar for the
Syk-deficient and the Syk/Lyn-deficient cells after 10 min of
stimulation (Table I). Interestingly,
however, SykEGFP expressed in the Syk/Lyn-deficient cells remained in
membrane caps following 60 min of activation in the majority of the
cells (Fig. 6D and Table I). These caps persisted for 6
h (the longest time point examined) following addition of the
anti-IgM Ab (data not shown). This is in sharp contrast to SykEGFP
expressed in Syk-deficient cells, which was present in only a few small
clusters at the plasma membrane after 60 min of treatment with
anti-IgM Abs (Fig. 6B and Table I) and was absent from
the membrane after 6 h of treatment (Fig. 5B). Thus,
the presence of SykEGFP in membrane caps is prolonged in the absence
of Lyn.
|
|
The "kinase-dead" KDEGFP mutant was used to monitor the
receptor-mediated redistribution of the enzyme in the absence of
intrinsic catalytic activity. KDEGFP was expressed in both the
Syk-deficient and the Syk/Lyn-deficient DT40 B cells, and the effect of
receptor cross-linking on its intracellular localization was monitored.
In resting cells of both types, KDEGFP was distributed normally
throughout the nucleus and cytoplasm (Fig. 2 and data not shown). In
the majority of Syk-deficient cells, BCR cross-linking led to the
appearance of clusters of KDEGFP at the plasma membrane (Fig. 6E and Table I). However, this redistribution was less
robust than that observed in cells expressing the catalytically active
SykEGFP. After 10 min of treatment with anti-IgM Abs, 63% of the
transfected cells exhibited some degree of KDEGFP redistribution to the
plasma membrane. At this time, KDEGFP appeared primarily in clusters of
small patches that resembled those observed in Syk-deficient cells
expressing SykEGFP that had been treated with anti-IgM Abs for 60
min (Fig. 6B). Most of these small patches of KDEGFP had
disappeared by 60 min of treatment with anti-IgM (Fig. 6F). The appearance of clusters of KDEGFP at the plasma
membrane was further blunted in cells lacking both Syk and Lyn (Fig. 6G and Table I). In approximately 50% of the cells, no
obvious change in the subcellular distribution of KDEGFP could be
observed following aggregation of the BCR. In the other 50%, only a
relatively limited amount of KDEFGP appeared in membrane patches (Fig. 6G). These aggregates of KDEGFP at the plasma membrane
persisted for at least 60 min following receptor aggregation (Fig. 6H). It is interesting to note that some clustering of
KDEGFP can occur in response to BCR cross-linking even in cells lacking
Lyn and an active form of Syk.
Syk and Lyn differentially regulate receptor clustering and internalization
Differences in the appearance of the clustered, fluorescently
tagged Syk molecules in cells lacking catalytically active Syk or Lyn
suggested that these kinases might play important roles in modulating
BCR dynamics following receptor aggregation. As a consequence, the fate
of aggregated BCR complexes in DT40 cells treated with anti-IgM Abs
for various periods of time was monitored using the fluorescently
tagged anti-IgM Ab. In wild-type DT40 cells, which express normal
levels of endogenous Lyn and Syk, treatment with anti-IgM Abs led
to the formation of surface IgM aggregates, which had redistributed
into caps in the majority of cells by 10 min (Figs. 7B and 8). Only 12% of the
cells exhibited internalized receptors at this time. By 60 min in the
majority of cells the receptors had been internalized (Fig. 8). By fluorescence microscopy,
internalized IgM appeared as a mass of punctate spots seen when
focusing on the interior of the cell (Fig. 7C).
|
|
, D and E). These capped
receptors persisted at the cell surface and failed to become
internalized even after 60 min of treatment with anti-IgM Abs (Fig. 8). The caps appeared in immunofluorescence images as bright clusters
of stained receptors seen by focussing on the surfaces of the cells
(Fig. 7F). These results are consistent with a requirement
for Lyn for receptor internalization.
In cells lacking Syk, treatment with anti-IgM Abs also led to
receptor clustering and the movement of clustered receptors to a single
pole of the cell (Fig. 7G). However, few cells formed
tightly capped receptors, and receptor internalization occurred
rapidly, with 48% of the cells exhibiting internalized receptors
within 10 min of treatment with anti-IgM Abs (Fig. 8). By 60 min,
nearly all cells had internalized receptors (Fig. 8), which appeared
within the cell as small, punctate clusters (Fig. 7I). Thus,
Syk appears to be necessary for the stable appearance of tightly capped
receptor complexes at the cell surface.
DT40 cells lacking both Syk and Lyn demonstrated a composite of the
defects observed in the single kinase-deficient lines. In these cells,
receptors were also aggregated by treatment with anti-IgM Abs (Fig. 7, J and K). The clustered BCR complexes
redistributed to one hemisphere of the cell, but failed to form tightly
clustered caps. These clustered Igs remained on the cell surface for an
extended period of time, with no receptor internalization observed by
60 min following receptor aggregation (Figs. 7I and
8).
Similar results were observed using an FITC-conjugated goat anti-chicken IgM Ab. The fluorescence of the extracellular fluorescein can be quenched selectively by a reduction in the pH of the medium. In a typical experiment, FACS analysis of Lyn-expressing cells or Lyn-deficient cells treated on ice (conditions under which internalization does not occur) with the fluorescein-conjugated Ab indicated that 95 and 96%, respectively, of the fluorescence could be quenched after acidification of the medium to pH 4.0. The fluorescence of Lyn-expressing cells treated with labeled Ab at 37°C for 30 min was quenched by only 32% due to internalization of a majority of the receptors. In contrast, the fluorescence of Lyn-deficient cells treated with Ab at 37°C for 30 min was quenched by 92%, indicating that the receptors remained at the cell surface.
A role for Syk in the formation of tightly capped receptors could be
confirmed by examining a mixed population of Syk/Lyn-deficient cells
and Syk/Lyn-deficient cells expressing SykEGFP. Following treatment
with Texas Red-conjugated anti-IgM Abs, tightly capped receptor
complexes appeared only in those cells expressing SykEGFP. Tightly
capped receptors were not observed in cells that did not express the
tagged fusion protein (Fig. 7, M and N). The
appearance of these tightly capped receptors required Syk catalytic
activity, since they also did not form in anti-IgM-treated cells
that expressed the catalytically inactive KDEGFP fusion protein (Fig. 7, O and P). The small amount of KDEGFP that did
relocate to the plasma membrane, however, colocalized with aggregated
surface IgM. However, many clusters of receptors lacked associated
KDEGFP.
Syk-dependent activation of NF-AT is elevated in Lyn-negative DT40 B cells
The redistribution studies indicated that in DT40 cells lacking
Lyn, receptor cross-linking still led to the recruitment of SykEGFP to
the clustered BCR, resulting in the formation of a long-lived
receptor-kinase complex. Therefore, the ability of Syk to restore
BCR-dependent signaling to cells deficient in Syk or deficient in both
Syk and Lyn was examined. In Syk-deficient cells, receptor-stimulated
tyrosine phosphorylation of intracellular proteins was reduced compared
with that in wild-type cells and was undetectable in cells lacking both
Syk and Lyn (Fig. 9A).
Ligation of the BCR by treatment with anti-IgM Abs was uncoupled
from the activation of NF-AT, as measured using an NF-AT-driven
luciferase reporter plasmid, in both cell types. The restoration of Syk
by transfection with a plasmid coding for Syk(WT) restored the
BCR-stimulated activation of NF-AT to both cell types. However, the
BCR-stimulated activation of NF-AT in the cells lacking Lyn was 4-fold
greater than that observed in the Lyn-expressing cells (Fig. 9B) despite the expression of similar amounts of Syk(WT)
(Fig. 9C). The enhanced signaling in Lyn-deficient cells was
also observed in cells transfected with the expression plasmid coding
for SykEGFP in place of Syk(WT) (data not shown).
|
|
Discussion |
|---|
|
TopAbstractIntroductionMaterials and MethodsResultsDiscussionReferences |
|---|
In unstimulated cells, SykEGFP is localized to both cytoplasmic and
nuclear compartments, but is excluded from the nucleoli (Fig. 2). In
preliminary studies not illustrated here, we have been able also to
demonstrate a similar distribution of endogenous Syk in human DG75 B
cells by indirect immunofluorescence. This subcellular distribution is
consistent with the previously reported localization of a ZAP-70 GFP
chimera (25). The nuclear localization of SykEGFP and
ZAP-70 GFP raise the intriguing possibility that Syk family kinases
have unrecognized nuclear functions that are independent of their roles
as mediators of proximal Ag receptor signals. At least one Syk
function, coupling the BCR to the activation of NF-AT, is independent
of its nuclear localization, since the attachment of a nuclear
exclusion signal to the carboxyl terminus of SykEGFP does not reduce
its activity in this assay (Fig. 1C). The large size of the
SykEGFP chimera (
100 kDa) precludes the possibility that it diffuses
freely into and out of the nucleus through nuclear pores, and there is
no evidence that the observed nuclear fluorescence of
SykEGFP-expressing cells is a consequence of proteolysis of the fusion
protein, as we could not detect the formation of smaller m.w. species
of protein containing the EGFP tag (Fig. 1B). Thus, the
fusion protein must interact directly or indirectly through associated
proteins with components of nuclear transport systems. The transport of
SykEGFP into and out of the nucleus is likely to be a dynamic process.
Upon BCR cross-linking, it appears that SykEGFP in both nuclear and
cytoplasmic compartments is eventually recruited to the plasma
membrane. It is possible that downstream signal transduction pathways
stimulated by the initial cross-linking of the BCR function actively to
enhance the nuclear export of Syk. Then, once Ag receptor complexes
have been internalized, the reappearance of SykEGFP in the nucleus
requires several hours. This prolonged exclusion of the kinase from the
nucleus occurs in cells expressing SykEGFP, but not in those expressing
the catalytically inactive KDEGFP, which cannot reconstitute
receptor-mediated signaling.
Phosphorylation of the ITAMs of the BCR complex that leads to the
recruitment of Syk is an activity normally attributed to members of the
Src family of protein tyrosine kinases, of which Lyn is the only member
expressed at any significant level in DT40 cells. In fact, a
catalytically inactive form of SykEGFP (KDEFGP) can be recruited to the
cross-linked BCR in cells expressing Lyn and lacking Syk (Fig. 6E). This process appears relatively inefficient compared
with the recruitment of a catalytically active form of SykEGFP, due in
part to the rapid rate at which the receptors become internalized in
the absence of an active Syk kinase (Figs. 7, G—I, and 8).
However, as long as Syk retains intrinsic catalytic activity, Lyn is
not required for its recruitment to the receptor, since this process is
not compromised in Lyn-deficient cells (Fig. 6, C and
D). The most straightforward explanation for this
observation is that Syk itself can catalyze the phosphorylation of
receptor ITAMs. The ability of Syk to catalyze the phosphorylation of
Ig- in immune complexes (26) and to phosphorylate ITAM
tyrosines in Lck-deficient Jurkat T cells (27) and
CD45-deficient B cells (28) has been suggested by previous
studies. This apparent lack of a requirement for a Src family kinase
for the recruitment of Syk to the BCR probably explains why receptor
engagement in Lyn-deficient DT40 cells is still coupled to inositol
1,4,5-trisphosphate production, calcium mobilization, and the
activation of Akt/protein kinase B (1, 15).
The Lyn-independent recruitment of Syk to the BCR requires that
alternative mechanisms exist to promote the initial interaction of Syk
with the receptor. A possible explanation is suggested by the
observation that even in the absence of both Lyn and a catalytically
active form of Syk, a detectable fraction of kinase-inactive Syk
(KDEGFP) still appears at sites of aggregated BCR complexes (Fig. 6G and Fig. 7, O and P). This
interaction probably occurs independently of receptor-stimulated
protein tyrosine phosphorylation, since this cannot be detected in the
Syk/Lyn-deficient DT40 cells following receptor aggregation (Fig. 9A). We think it is likely that the appearance of KDEGFP at
the site of cross-linked BCR complexes reflects the aggregation of
small amounts of KDEGFP preassociated with the unligated BCR. A low
level of Syk associated with the BCR was reported previously in
unstimulated B cells (29). This preassociated Syk could
conceivably initiate receptor phosphorylation and the recruitment of
additional Syk molecules to the cross-linked receptor, bypassing the
requirement for a member of the Src family in the early events of B
cell signaling.
The presence or the absence of Syk and Lyn in DT40 B cells has major
effects on the dynamics of receptor aggregation and internalization.
The treatment of B cells with anti-IgM Abs leads first to the
appearance of the BCR and its signaling component Ig- (and
associated Ig-) in surface patches. The formation of these
receptor-containing patches and their initial redistribution to one
hemisphere of the cell are independent of the expression of either Syk
or Lyn. If Syk is present, it is being actively recruited to the sites
of the aggregated receptors during formation of the BCR-containing
membrane patches. These patches, with time, redistribute into a single
cap. The appearance of these caps is strongly influenced by the
presence or the absence of Syk within the clustered BCR complex. If Syk
is absent, the receptor caps are large, relatively diffuse, and
internalized rapidly if Lyn is present in the cells. If Lyn is also
absent, the receptor clusters that form are broad and cover one-quarter
to one-half of the cell surface, where they remain for an extended
period of time, indicating that Lyn is important for receptor
internalization. In the presence of Syk, the BCR redistributes into
tightly clustered caps regardless of whether Lyn is expressed. This
effect requires the expression of a catalytically active form of Syk.
If Lyn is also expressed, the receptors in these caps are internalized,
although at a slower rate than in Syk-deficient cells. In the presence
of Syk, but in the absence of Lyn, the tightly clustered receptors
remain on the cell surface for an extended period of time (at least
6 h). Thus, Syk is required for the formation of tightly clustered
caps of aggregated Ag receptors, and Lyn is required for receptor
internalization.
A role for Lyn in receptor internalization is consistent with several
studies that have used pharmacologic tyrosine kinase inhibitors to
block BCR endocytosis (30, 31, 32). This suggests that one
role for Lyn may be the down-regulation of activated receptors, since,
in its absence, the Syk-BCR complexes that form following receptor
cross-linking remain at the cell surface for a prolonged period of time
(Figs. 6D and 7F). Thus, the activation of
Lyn-deficient cells leads to a slow, but prolonged, increase in
cytosolic free calcium (1) and an enhanced activation of
NF-AT (Fig. 9B). It is possible that reduced receptor
internalization may contribute to the hyper-responsiveness to IgM
cross-linking that is observed in B cells from Lyn-deficient mice,
which also demonstrate a delayed, but prolonged, calcium response
(17), but this has not been determined.
The capping of surface IgM is an energy-dependent process that is
mediated by the contractile activity of the actomyosin peripheral
cytoskeleton and is inhibited by the treatment of cells with
cytochalasin D, which disrupts actin microfilaments. A requirement for
Syk has been demonstrated previously for the increased assembly of
cortical actin in DT40 cells following the cross-linking of expressed
Fc 
subunits (33). Thus, Syk might influence the
clustering of receptors by regulating aspects of the actin assembly
process important for the formation of the tightly clustered and capped
receptor complexes. How this affects the overall process of receptor
internalization and subsequent trafficking is not known. However, it is
known that dominant-negative mutants of Syk inhibit Ag presentation
(34), which requires the proper uptake and delivery of
BCR-Ag complexes to MHC II-containing endosomes, the assembly of which
is inhibited by tyrosine kinase inhibitors (35). Thus,
both Lyn and Syk play important roles in regulating Ag receptor
dynamics in B cells.
|
Footnotes |
|---|
2 Address correspondence and reprint requests to Dr. Robert L. Geahlen, Department of Medicinal Chemistry and Molecular Pharmacology, Hansen Life Sciences Research Building, Purdue University, West Lafayette, IN 47907.
3 Abbreviations used in this paper: BCR, B cell Ag receptor; ITAM, immunoreceptor tyrosine-based activation motif; GFP, green fluorescent protein; EGFP, enhanced GFP; WT, wild type; NES, nuclear export signal; KD, kinase dead.
Received for publication May 4, 2000. Accepted for publication November 15, 2000.
|
References |
|---|
|
TopAbstractIntroductionMaterials and MethodsResultsDiscussionReferences |
|---|
2. J. Exp. Med. 184:31.-tubulin on tyrosine. J. Biol. Chem. 271:4755.This article has been cited by other articles:
|
|
 |
|
  O. O. Onabajo, M. K. Seeley, A. Kale, B. Qualmann, M. Kessels, J. Han, T.-H. Tan, and W. Song Actin-Binding Protein 1 Regulates B Cell Receptor-Mediated Antigen Processing and Presentation in Response to B Cell Receptor Activation J. Immunol., May 15, 2008; 180(10): 6685 - 6695. [Abstract] [Full Text] [PDF]
|
|
|
|
 |
|
 M. Li, S. S. Ong, B. Rajwa, V. T. Thieu, R. L. Geahlen, and M. L. Harrison The SH3 Domain of Lck Modulates T-Cell Receptor-Dependent Activation of Extracellular Signal-Regulated Kinase through Activation of Raf-1 Mol. Cell. Biol., January 15, 2008; 28(2): 630 - 641. [Abstract] [Full Text] [PDF]
|
|
|
|
|
|
C. Lau, X. Wang, L. Song, M. North, S. Wiehler, D. Proud, and C.-W. Chow Syk Associates with Clathrin and Mediates Phosphatidylinositol 3-Kinase Activation during Human Rhinovirus Internalization J. Immunol., January 15, 2008; 180(2): 870 - 880. [Abstract] [Full Text] [PDF]
|
|
|
|
 |
|
 H. Oh, E. Ozkirimli, K. Shah, M. L. Harrison, and R. L. Geahlen Generation of an Analog-sensitive Syk Tyrosine Kinase for the Study of Signaling Dynamics from the B Cell Antigen Receptor J. Biol. Chem., November 16, 2007; 282(46): 33760 - 33768. [Abstract] [Full Text] [PDF]
|
|
|
|
|
|
M. M. Al-Alwan, K. Okkenhaug, B. Vanhaesebroeck, J. S. Hayflick, and A. J. Marshall Requirement for Phosphoinositide 3-Kinase p110{delta} Signaling in B Cell Antigen Receptor-Mediated Antigen Presentation J. Immunol., February 15, 2007; 178(4): 2328 - 2335. [Abstract] [Full Text] [PDF]
|
|
|
|
 |
|
 C. M. Mutch, R. Sanyal, T. L. Unruh, L. Grigoriou, M. Zhu, W. Zhang, and J. P. Deans Activation-induced endocytosis of the raft-associated transmembrane adaptor protein LAB/NTAL in B lymphocytes: evidence for a role in internalization of the B cell receptor Int. Immunol., January 1, 2007; 19(1): 19 - 30. [Abstract]< |
) or were activated by treatment
with anti-IgM Abs (
). Relative luciferase activity is reported
as the activity observed under the experimental conditions divided by
the activity produced in response to stimulation with a mixture of PMA
and ionomycin. The values represent the mean and SD of three separate
analyses.
), or internalized (