| HOME | HELP | FEEDBACK | SUBSCRIPTIONS | ARCHIVE | SEARCH | TABLE OF CONTENTS |
|
|
|
|||||||
|
|||||||||
,
Departments of
*
Medicine and
Microbiology, University of Alabama, Birmingham, AL 35294; and
Birmingham Veterans Affairs Medical Center, Birmingham, AL 35233
 |
Abstract |
|---|
 Top Abstract IntroductionMaterials and MethodsResultsDiscussionReferences |
|---|
2. These molecules also were found associated with native CD19. In
mapping studies with altered constructs, CD19 Y330 and/or Y360 were
necessary for binding Grb2 and Sos. Vav associated with CD19
constitutively in unstimulated cells by a tyrosine-independent
mechanism requiring the portion of CD19 encoded by exons 9–12. After B
cell Ag receptor stimulation, Vav association was tyrosine-dependent,
but binding was influenced by multiple residues. However, when
maximally phosphorylated by pervanadate, Y391 and, to a lesser extent,
Y421 were sufficient. CD19 Y391 was also both necessary and sufficient
for binding phospholipase C-2. Thus, different tyrosines along the
CD19 cytoplasmic domain provide scaffolding for the formation of
complexes of different signaling molecules. |
Introduction |
|---|
|
TopAbstractIntroductionMaterials and MethodsResultsDiscussionReferences |
|---|
Phosphorylation of CD19 cytoplasmic tyrosines provides a mechanism for
enhancement of BCR signaling by CD19 (8, 10). The
phosphorylated CD19 tyrosines provide docking sites for Src homology 2
(SH2) domain-containing proteins, serving as a scaffold for the
assembly of complexes of signaling molecules. Vav and
phosphatidylinositol 3-kinase (PI3K) are easily demonstrable in
immunoprecipitates of CD19 (12, 13, 14). Their binding is
reduced by mutation of CD19 tyrosine (Y)391 (numbering as per Ref.
15) and Y482 and Y513, respectively (10, 12, 16). Phosphopeptides corresponding to CD19 Y403 and Y443 bound
to Fyn SH2 domains, although coprecipitation of the intact molecules
was not shown (16). An association of Lyn and CD19 has
been reported but not mapped (17, 18). A phospholipase
(PLC)-1 SH2-Ig fusion protein precipitated a 95-kDa
phosphoprotein thought to be CD19 (16).
Coligation of CD19 and membrane IgM (mIgM) on B cell lines has served as a model of the enhancement of B cell downstream signaling pathways by CD19 (19, 20). Although CD19 may have independent functions, cross-linking is required for CD19 to have a positive effect on c-Jun N-terminal kinase activation or Bcl-2 expression in murine B cells and on ERK2 activation, DNA synthesis, or Ab secretion in mature human B cells (6, 8, 9, 10, 21, 22). In addition, C3d-opsonized Ag has been postulated to coligate the BCR and the CD21/CD19/CD81 complex, and covalently linked C3d is a potent adjuvant to hen egg lysozyme (23, 24).
We and others have used the coligation of CD19 and mIgM in Daudi B cells to study the function of the CD19 cytoplasmic domain (8, 20, 25, 26, 27). Chimeric molecules containing the extracellular region of CD4 linked to the transmembrane and cytoplasmic portions of CD19 mimic the signaling functions of native CD19 (25). Using such chimeric molecules expressed in Daudi cells, we previously demonstrated that CD19 Y391 was required for normal function but that Y482 and Y513 were not in this system (8, 10). In CD45-transfected plasmacytoma cells, mutation of Y482 and Y513 in cotransfected CD19 reduced the magnitude of the increase in [Ca2+]i induced by ligation of the BCR alone (28, 29). However, CD19 has six additional cytoplasmic tyrosines of unknown function.
To systematically dissect the role of all CD19 cytoplasmic tyrosines,
we compared the enhancement of the increase in
[Ca2+]i and ERK2 by
CD4/19 chimeric molecules containing mutations, singly or in pairs, of
each CD19 cytoplasmic tyrosine (Fig. 1
).
We used this information to construct chimeras containing the minimally
sufficient tyrosines. These results indicate an important role for Y330
and Y391 and enhancing roles for Y360 and Y421. Synthetic
phosphopeptides provided initial evidence for binding of signaling
molecules to these peptides, and association with native CD19 was
confirmed. The associations were then mapped with mutant chimeric
constructs. The results demonstrate that different tyrosines in the
CD19 cytoplasmic domain provide a membrane anchor for the formation of
different groups of signaling molecules.
|
|
Materials and Methods |
|---|
|
TopAbstractIntroductionMaterials and MethodsResultsDiscussionReferences |
|---|
CD4/19 chimeric constructs, in which the extracellular region of CD19 is replaced with that of CD4, were as previously described (8, 10, 25, 26). Mutation of codons encoding tyrosines (TAT, TAC) was achieved using mutagenesis oligos consisting of the phenylalanine codon (TTT, TTC) and 10–15 flanking base pairs using Altered Sites II Mutagenesis System (Promega, Madison, WI) as described (8, 10, 12). Truncation mutants exon 8F (ex8F) and ex12-6F were derived from the cDNA for CD4/19 ex8 and ex12, as previously described (26). The presence of the desired mutation and the absence of additional mutations were confirmed by sequencing. CD4/19 chimeras were stably transfected and expressed in Daudi cells as described (25). Clones were selected for analysis based on equivalent expression by flow cytometric analysis.
Cell culture, Abs, and CD19-derived peptides
The human Daudi B cell line was cultured in IMDM supplemented
with 10% FCS, penicillin (100 U/ml), and streptomycin (100 µg/ml).
mAb against Vav and Grb2 were purchased (Upstate Biotechnology, Lake
Placid, NY). Polyclonal antisera against ERK2 (Santa Cruz
Biotechnology, Santa Cruz, CA), Sos (Upstate Biotechnology), and
PLC-2 (Santa Cruz Biotechnology) were purchased. Polyclonal goat
F(ab')2 anti-human IgM was purchased
(BioSource International, Camarillo, TX). Biotinylated Fab' DA4.4 mouse
anti-human IgM and biotinylated F(ab')2
ADF4.2 anti-CD19 were prepared as described previously (5, 30). Biotinylated CD19 peptides PQNQY(330)GNVLSL,
TAPSY(360)GNPSSD, EGEGY(391)EEPDSE, and DGSGY(421)ENPEDE in tyrosine
phosphorylated and unphosphorylated form were purchased (Quality
Control Biochemicals, Hopkington, MA).
ERK2 in vitro kinase assays and Ca2+ assays
ERK2 kinase assay and flow cytometric analysis of [Ca2+]i were performed as previously described (5, 8). Replicate samples of cells were incubated with buffer only, 5 µg/ml anti-CD19 alone, 5 µg/ml anti-CD4 alone, anti-IgM alone, both anti-IgM and 5 µg/ml anti-CD19, and both anti-IgM and 5 µg/ml anti-CD4 for 10 min and washed. These Ab fragments had no effect on [Ca2+]i or ERK without addition of avidin. The concentration of anti-IgM was 20 ng/ml in the Ca2+ assays and 1 µg/ml in the ERK2 assays. In the Ca2+ assays, cells were analyzed by flow cytometry for a 20-s baseline period before the addition of 2 µg/ml avidin, and analysis was continued for a total of 3 min. In the ERK2 assays the cells were stimulated with 5 µg/ml avidin for 1 min (the results of the ERK assays are similar at 30 sec and 5 min, X.L., unpublished observations). In the Ca2+ assays, the fold enhancement induced by coligation of CD4 or CD19 with mIgM was calculated using the peak [Ca2+]i after coligation. At this time point the response to CD4 alone or CD19 alone was negligible. Therefore, the increase in [Ca2+]i after coligation of either CD4 or CD19 with mIgM was compared with the response with anti-IgM alone at the time of the peak of the relevant synergistic response. In the ERK2 assays the response to CD4 or CD19 ligation alone was also minimal. For both Ca2+ and ERK2 assays, the fold increase induced by CD4/mIgM coligation over mIgM alone and the fold increase induced by CD19/mIgM coligation over mIgM alone are compared.
Stimulations, immunoprecipitations, peptide precipitations, and immunoblots
For anti-IgM stimulation, cells were incubated with 20 µg/ml polyclonal anti-IgM for 1 min. For pervanadate stimulation, cells were incubated with 10 mM VO4 and 10 µM H2O2 for 20 min. For immunoprecipitations, 2 x 107 unstimulated or stimulated Daudi cells were lysed with Nonidet P-40 lysis buffer (1% Nonidet P-40, 50 mM Tris (pH 7.4), 80 mM KCl, 10 mM EDTA, 5 mM iodoacetamide, and a mixture of protease inhibitors). Lysates were incubated with 4 µg/ml anti-CD19 or anti-CD4 mAb for 45 min at 4°C. Forty microliters of 50% Protein A-trisacryl or UltraLink Immobilized Protein A beads (Pierce, Rockford, IL) were added to lysates for 45 min at 4°C. Coprecipitations of Grb2 and Sos used Nonidet P-40 lysates of 8 x 107 unstimulated or pervanadate-stimulated cells. For two-step precipitations, beads were recovered after incubation with lysate of pervanadate-stimulated cells, washed, then added to Nonidet P-40 lysates of 8 x 107 unstimulated cells for 45 min at 4°C. Beads were washed, eluted, subjected to SDS-PAGE, and transferred to nitrocellulose. Blots were blocked with 3% milk or 2% BSA, probed with primary Ab followed by HRP conjugated secondary Ab (Jackson) and visualized with ECL (Amersham). The amount of associated protein was estimated by densitometry (NIH Image Version 1.58). Peptide precipitation was performed as described with modifications (10). Briefly, biotinylated peptides (2 pmol) were incubated with 200 µl of 25% UltraLink Immobilized NeutrAvidin Plus (Pierce) in PBS for 1 h at 4°C. Washed beads were incubated with Nonidet P-40 lysates of 2.5 x 107 Daudi cells for 30 min. at 20°C and washed. Bound proteins were eluted, resolved on 10% SDS-PAGE, and stained with silver (Bio-Rad, Richmond, CA).
Replication
All experiments have been replicated at least once and the results shown are representative.
|
Results |
|---|
|
TopAbstractIntroductionMaterials and MethodsResultsDiscussionReferences |
|---|
Previous studies have identified an important role for CD19
cytoplasmic Y391 for the synergistic increase in
[Ca2+]i and ERK2
activation after CD19-mIgM coligation (8, 10) and for Y482
and Y513 in the increase in
[Ca2+]i after BCR
ligation (28, 29). In this study, we compared the role of
all cytoplasmic tyrosines of CD19 in Ca2+ and
ERK2 signaling (Fig. 2A).
Different CD4/19 chimeric molecules containing point mutations, singly
or in pairs, of each of the CD19 cytoplasmic tyrosines were expressed
on Daudi cells. These cells also expressed native CD19, which served as
a positive control. To minimize the impact of any non-CD19 factors that
might alter signaling between different populations of cells, the
function of the chimeric and native molecules on the same cells were
compared.
|
Mutation of Y391 indeed had the most profound effect on both the
increase in [Ca2+]i and
ERK2 activation (Fig. 2
, A and B). The additional
experiments performed here also confirm the lack of effect of mutation
of Y482 and Y513 in this system. However, other CD19 tyrosines also
were required for optimal signaling in the Ca2+
assays. Mutation of Y421 and of Y330 and Y360 also resulted in
decreases in responses that were significantly different from the
unmutated chimera. Mutation of Y490 or of Y403 and Y443 had no
statistically significant effect. In the ERK2 assays, mutation of Y391
was the only change that resulted in a significant loss of activity.
Mutation of Y330 and Y360 resulted in a small but statistically
significant increase in response. The discrepancy in the requirement of
Y330 and Y360 for the increase in
[Ca2+]i but not for ERK2
suggests that there are differences in the signaling pathways linked to
CD19 that enhance the increase in
[Ca2+]i and ERK2
activation.
These experiments suggested that Y330, Y360, Y391, and Y421 play a role
in CD19 signaling in this system. We found that chimeric molecules
containing only these four tyrosines, with point mutations of the other
five, were sufficient for synergistic enhancement of the increase in
[Ca2+]i and ERK2
activation (Fig. 2, C and D). Chimeras containing
substitutions in the remaining four tyrosines were constructed and
tested. In these experiments, the effects of the mutations were similar
in both Ca2+ and ERK2 assays. In both types of
analysis, any further mutation reduced the enhancement of signaling,
suggesting that, in the absence of other tyrosines, Y330, Y360, Y391,
and Y421 all can contribute to CD19 signaling. Single further mutation
of either Y330 or Y391 resulted in a >70% reduction in enhancement,
suggesting a crucial role for these tyrosines. Mutation of either Y360
(Y330, Y391, and Y421 remaining) or of Y421 (Y330, Y360, and Y391
remaining) resulted in 40% and 32% reduction in enhancement in the
Ca2+ assays and 33% and 51% reduction in
enhancement in the ERK2 assays, respectively. Further mutation of Y360
from Y330, Y360, and Y391 (Y330 and Y391 remaining) or of Y421 from
Y330, Y391, and Y421 (Y330 and Y391 remaining) resulted in further loss
of function. These data suggest a cooperative role for Y360 and Y421
with Y330 and Y391 in the synergistic increase in
[Ca2+]i and ERK2
activation after CD19-mIgM coligation. The differential requirement for
Y330 and Y360 in ERK2 activation in the presence or absence of
additional tyrosines suggests some redundancy when other tyrosines are
present.
CD19-derived phosphopeptides associate with proteins of apparent Mr of 24, 113, 139, and 149
Phosphorylation of CD19 tyrosines provides docking sites for SH2
domain-containing proteins, and these associations are likely to
mediate CD19 function. We searched for proteins associating with the
tyrosines shown above to play a role in synergistic increase in
[Ca2+]i and ERK2
activation after CD19-mIgM coligation. Biotinylated 11-mer peptides
corresponding to unphosphorylated and phosphorylated Y330, Y360, Y391,
and Y421 and flanking sequences were bound to avidin-coated beads and
incubated with Daudi lysates. Bound proteins were eluted, resolved by
SDS-PAGE, and silver stained. Phospho-Y330 bound two proteins with
apparent Mr of 24 and 149, whereas
nonphosphorylated Y330 did not (Fig. 3A, filled arrows).
Phospho-Y391 and, to a lesser extent, phospho-Y421 bound two proteins
with apparent Mr of 113 and 139 (Fig. 3B), whereas nonphosphorylated Y391 and Y421 peptides did
not. No specific associations with phospho-Y360 were identified. The
homology with Y330 raises the question of a secondary role in binding
of similar molecules, below the limits of detection here.
|
To identify the proteins of apparent
Mr of 24 and 149 that were associated
with phospho-Y330, precipitates were formed from Daudi lysates as
above, transferred to nitrocellulose, and probed with Abs specific for
Grb2 and Sos. Phospho-Y330 was able to coprecipitate Grb2 and Sos,
whereas nonphosphorylated Y330 could not (data not shown). To determine
whether native CD19 could coprecipitate Grb2, CD19 was precipitated
from lysates of 8 x 107 unstimulated or
pervanadate-stimulated cells. Replicate CD19 precipitates from
stimulated cells were also subjected to an additional incubation with
lysates of 8 x 107 unstimulated cells. CD19
could coprecipitate Grb2 in stimulated but not unstimulated lysates
(Fig. 4A). More Grb2 could be
coprecipitated when a second incubation with unstimulated lysates was
performed. This suggests that other proteins are competing with
phospho-CD19 for binding Grb2 in stimulated lysates. An
anti-phosphotyrosine probe of the same blot confirms that tyrosine
phosphorylation was induced by pervanadate. A CD19 probe shows a
slightly lower recovery of CD19 from pervanadate-stimulated cells. CD19
could coprecipitate Sos from 8 x 107 cells
of pervanadate-stimulated Daudi lysates, whereas UPC10, a control Ab,
could not (Fig. 4B). Again, a two-step precipitation first
using pervanadate-stimulated and then unstimulated lysates was able to
precipitate more Sos than a single-step precipitation from
pervanadate-stimulated cells. The same blot was probed with
anti-phosphotyrosine to demonstrate that pervanadate stimulation
induced tyrosine phosphorylation. The CD19 probe shows equivalent
recovery. To map the tyrosines required for association of CD19 and
Grb2, chimeric CD4/19 was precipitated from stably transfected
Daudi cell lines (Fig. 4C). A chimera with an intact CD19
cytoplasmic domain (wild type (WT)) was able to coprecipitate Grb2 from
lysates of 8 x 107 pervanadate-stimulated
cells but not from those of unstimulated cells. Mutation of all nine
tyrosines to phenylalanine (Y9F) abolishes this association,
demonstrating the requirement for cytoplasmic phosphotyrosines of CD19
in mediating this association. Mutation of 330 and 360 (330,360F)
ablates the ability of the chimera to coprecipitate Grb2, establishing
a requirement of 330 and/or 360 for Grb2 association. An
anti-phosphotyrosine probe of the same blot demonstrates
phosphorylation of the WT and 330,360F chimeras and an absence of
tyrosine phosphorylation in Y9F. The CD4 probe demonstrates equivalent
recovery of chimera from each sample. In preliminary experiments, a
construct containing only Y330 and Y360 failed to precipiate Grb2, but
the significance of this is as yet unclear, given the ability of the
phospho-Y330 peptide to do so.
|
Previous studies demonstrated activation-induced association of
CD19 and Vav and that CD19 cytoplasmic Y391 is important in this
interaction (8, 10, 14). The protein with apparent
Mr of 113 that coprecipitated with
peptides phospho-Y391 and phospho-Y421 (Fig. 3B) comigrated
with Vav, as demonstrated by immunoblotting (data not shown). The
requirement for other CD19 cytoplasmic tyrosines in the
activation-induced association of CD19 and Vav was investigated by
precipitating the CD4/19 chimera from stably transfected Daudi cells
before and after stimulation with polyclonal anti-IgM. The
precipitates were probed with anti-Vav (shown on the left in Fig. 5). The amount of coprecipitating Vav was
estimated by densitometry. For each chimera, the fold increase in Vav
association after stimulation over the basal association in
unstimulated cells is reported (shown on the right in Fig. 5). The WT
cytoplasmic domain exhibits a low basal association and an increase in
Vav association on stimulation. Y9F, a construct in
which all nine tyrosines are mutated to phenylalanine, shows an
equivalent basal association that decreases on stimulation. When
tyrosines 330 and 360 are mutated to phenylalanine (330,360F), the
increase in Vav association on stimulation is not significantly
different from WT. There is a similar increase when the cytoplasmic
domain is truncated at the end of ex12. However, if any of tyrosines
391, 403/443, 421, or 490 are mutated to phenylalanine, after
stimulation there is more Vav coprecipitated than with Y9F but less
than with WT. This suggests a role for each of these tyrosines in
maximal association of Vav with CD19 after stimulation with
anti-IgM.
|
The ability of Y9F to coprecipitate Vav in unstimulated but not
stimulated lysates suggested that there is a constitutive association
of CD19 and Vav that is independent of tyrosine phosphorylation. To
determine the site of this interaction, CD4/19 chimeras that are
truncations of Y9F were tested for their ability to associate with Vav
in a constitutive manner (Fig. 6). Y9F is
able to coprecipitate Vav before but not after pervanadate stimulation,
similar to what is seen with anti-IgM stimulation. A chimera that
is truncated at the end of exon 8 with all tyrosines mutated to
phenylalanine (ex8F) is unable to coprecipitate Vav in either
unstimulated or stimulated cells. However, a chimera that is truncated
at the end of exon 12 with all tyrosines mutated to phenylalanine
(ex12-6F) is able to coprecipitate Vav, suggesting that the
constitutive association of CD19 and Vav is mediated by the CD19
cytoplasmic region encoded by exons 9–12. The blot was reprobed with
anti-CD4 Ab to demonstrate equivalent recovery of the chimera.
|
2 and tyrosines 330, 360, 391, and 421
are sufficient for the BCR-induced association
To identify the protein of apparent
Mr of 139 that bound to peptides
phospho-Y391 and phospho-Y421 (Fig. 3B), precipitates formed
with these peptides were probed with Abs specific for PLC-2.
Phospho-Y391 and phospho-Y421 were able to coprecipitate PLC-2,
whereas nonphosphorylated Y391 and Y421 could not (data not shown). To
determine whether native CD19 could coprecipitate PLC-2,
precipitates were formed from lysates of 2 x
107 unstimulated and pervanadate-stimulated cells
with anti-CD19 Ab. Immune complexes were eluted, resolved by
SDS-PAGE, transferred to nitrocellulose, and probed with
anti-PLC-2 Ab (Fig. 7A). CD19 is able to
coprecipitate PLC-2 in stimulated but not unstimulated cells. The
blot was stripped and reprobed with anti-phosphotyrosine to
demonstrate tyrosine phosphorylation of CD19 after pervanadate
stimulation. The blot was further reprobed with anti-CD19 to show
equivalent recovery. To determine the requirement for CD19 cytoplasmic
tyrosines in the anti-IgM-induced association, anti-CD4
immunoprecipitates were formed from unstimulated and
anti-IgM-stimulated Daudi cells that were either untransfected or
that expressed CD4/19 chimeras containing either the WT cytoplasmic
domain or a cytoplasmic domain containing mutations in all nine
tyrosines (Fig. 7B). Probing for PLC-2 reveals a
low-level, non-CD4-dependent, nonspecific precipitation that was
observed consistently in untransfected cells and in cells expressing
the Y9F chimera. The slight increase observed in the latter relative to
untransfected cells was variably observed but is unlikely to be
meaningful given the absence of coprecipitation with native CD19 from
untransfected cells. However, the non-CD4-dependent precipitation
precluded analysis of constitutive association of PLC-2. An increase
(2.4-fold) in association after anti-IgM stimulation is observed
only with the intact cytoplasmic domain (WT), suggesting that the
activation-induced association between CD19 and PLC-2 is dependent
on tyrosine phosphorylation. In other experiments, activation with
anti-IgM induced an increase in PLC-2 association with a
construct in which all but four tyrosines (Y330, Y360, Y391, and Y421)
are mutated to phenylalanine (Y5F), suggesting that these tyrosines
bind PLC-2 (data not shown). Additional experiments were conducted
to map this association.
|
2 in
pervanadate-stimulated cells
The data in Fig. 5 for Vav and similar experiments with PLC-2
suggest that multiple CD19 tyrosines had an effect on binding when
cells were stimulated with anti-IgM. This suggested that certain
CD19 cytoplasmic tyrosines are required to induce maximal
phosphorylation of other tyrosines, possibly by recruiting and/or
activating the tyrosine kinases Lyn (17, 18) or Fyn
(16) or others. To eliminate this variable, pervanadate
stimulation was used to maximally phosphorylate tyrosines in CD4/19
chimeras to map the requirements for CD19 cytoplasmic tyrosines in Vav
and PLC-2 association (Fig. 8). A WT
cytoplasmic domain is able to associate with Vav and PLC-2 in a
stimulation-dependent manner. Mutation of Y391, Y421, and Y490 (Y3F)
eliminates stimulation-induced association of PLC-2, as does
mutation of Y391 and Y421 (Y2F). A chimera in which all tyrosines
except Y391 are mutated (391Y) associates with PLC-2 as well as WT,
whereas Y421 alone (421Y) does not show any PLC-2 association above
background although the chimera is well phosphorylated. A light band
observed in the PLC-2 probe in stimulated lanes Y3F, Y2F, Y9F, and
421Y represents an increase in non-CD4-dependent precipitation (as
above) in pervanadate-treated cells and is insignificant compared with
the increase observed in WT or 391Y. The same blot was stripped and
reprobed for Vav. Vav association exhibits a similar pattern of
association except that there is a constitutive association in
unstimulated cells. 391Y exhibits a stimulation-induced increase
in Vav association comparable to that observed with WT. Y9F again
exhibits a decrease in Vav association on stimulation. Y3F and Y2F have
lost the stimulation-induced increase in association seen in WT but do
not show the loss of association observed with Y9F. This suggests that
other CD19 tyrosines can mediate weaker binding of Vav in the absence
of Y391, perhaps through indirect association. The blot was stripped
and reprobed with anti-phosphotyrosine. Chimeras with fewer
tyrosines show a proportionally lower but still substantial level of
tyrosine phosphorylation after pervanadate stimulation. Blots were
reprobed with anti-CD4 to demonstrate equivalent recovery of
chimera.
|
|
Discussion |
|---|
|
TopAbstractIntroductionMaterials and MethodsResultsDiscussionReferences |
|---|
Our functional studies suggest that Y330, Y360, Y391, and Y421 can each
interact with downstream signaling pathways. In the presence of the
other tyrosines, Y330 and Y360 are dispensable for ERK2 but not
Ca2+ signaling. However, Y330 and Y360 are
required for both ERK2 and Ca2+ signaling in the
absence of Y403, Y443, Y482, Y490, and Y513. We suspect that these
differences may be due to alternative binding sites for effectors or
activation of different effectors that may converge into these
pathways. Introduction of structural changes by the tyrosine mutations
are difficult to exclude, but current evidence suggests that this is
not the case. Phosphorylation of each of these constructs (including
those with mutations of Y403 and Y443, the putative sites of Fyn
association) is proportional to the number of remaining tyrosines (data
not shown). Association with PI3K is intact if Y482 and Y513 are
present (data not shown). Mutation of Y330 and Y360 by themselves did
not reduce ERK2 activation (Fig. 2C). These observations
suggest that the tyrosine to phenylalanine substitutions do not result
in gross structural changes that disrupt the binding at the intact
residues.
Previous studies have identified a role for Y391 in ERK2 and
Ca2+ signaling and in Vav association. Vav can
enhance Ca2+ signaling via activation of
phosphatidylinositol 4-phosphate 5-kinase and increased production of
phosphatidylinostol 4,5-bisphosphate (8, 10), the
substrate for PLC-2. We now show that PLC-2 also associates with
CD19 via Y391. Therefore, CD19 may play a role in membrane localization
of PLC-2. We have demonstrated by coprecipitation that Vav and
PLC-2 inducibly associate in pervanadate stimulated Daudi cells
(data not shown). Together, this suggests a model in which, upon
activation, a Vav/PLC-2 complex binds phospho-CD19 and coordinately
regulates the activation of PLC-2 and the pathway which produces its
substrate, phosphatidylinostol 4,5-bisphosphate.
We previously had interpreted our results with mutation of Y391 (n
which the amount of associated Vav was equivalent in precipitates of
the CD4/19 chimera containing mutation of Y391 from unstimulated and
stimulated cells, as opposed to the increase observed with WT CD19
cytoplasmic domain) as reflecting an absence of activation-induced
association when Y391 was mutated (8). However, our
current results show that when all tyrosines are mutated, there is a
dramatic reduction in the amount of associated Vav after stimulation
(Fig. 5, 6, and 8). Thus, the residual binding, after activation, to
CD4/19 with mutation of Y391 represents partial, tyrosine-based
binding. Other tyrosines also can mediate Vav binding after mutation of
Y391, although at lower levels than Y391 (Fig. 8). The complete loss of
postactivation binding after mutation of all tyrosines suggests that
the basal association of Vav with CD19 is of relatively low affinity
and that, after activation, Vav shifts to SH2-mediated,
tyrosine-dependent binding to CD19 or an associated molecule (e.g.,
PLC-2). CD19 may provide a reservoir of Vav near the site of
postactivation binding.
BLNK/SLP-65, an adaptor protein that binds effectors/adaptors including
PLC-2, Vav, and Grb2 (32, 33, 34), has been shown to be
crucial for activation of PLC-2 in DT40 cells (35, 36).
We have been unable to detect BLNK associated with CD19 in stimulated
Daudi cells, although such precipitates contain Vav, PLC-2, Grb2,
and Sos. BLNK/SLP-65 is detectable in Daudi whole cell lysates (data
not shown). CD19 may provide an alternative pathway for membrane
localization and activation of PLC-2.
The enhanced association of Grb2 and Sos when phosphorylated CD19 is recovered from activated lysates and then incubated with unactivated lysates suggests that, under conditions in which phosphorylation of tyrosine residues on intracellular proteins is maximal, the affinity of binding to phospho-CD19 is less than that of other phosphoproteins. This suggests the hypothesis that, under conditions in which CD19 is preferentially phosphorylated such as after ligation of CD19 alone (12), CD19 might serve as a sink for Grb2 and Sos and play a negative role (6, 37, 38). Such an effect might explain the small increase in ERK2 activity observed with the chimera with mutations in Y330 and Y360. The mutational analysis does not permit conclusions as to whether the proteins identified as associating with particular residues mediate the functional role demonstrated for those residues. The tyrosine residues may play other roles in addition to SH2-mediated protein binding, such as forming part of the motif for phosphorylation of nearby serine/threonine residues. In addition, we suspect that mutation of some tyrosines may affect the phosphorylation of other CD19 tyrosines or of CD19-associated molecules. In the studies of association after stimulation with anti-IgM, alterations of different tyrosines reduced the level of Vav association, even when Y391, shown in the pervanadate studies to be sufficient by itself for maximal Vav association, was intact.
Native CD19 is present on these cells, and thus these experiments only address the question of which tyrosines are required on the CD19 molecule that is cross-linked with the BCR. This might explain the lack of effect of mutation of Y482 and Y513. Buhl et al. (28) have shown that these residues are important for optimal signaling after ligation of the BCR alone. The native CD19 in the transfected Daudi cells could serve this function in our system. In other studies we have found that wortmannin suppresses the Ca2+ and ERK signals after CD19-mIgM coligation in Daudi cells (X. Li and R. H. Carter, unpublished observations).
This is the first report to demonstrate a function for the upstream
CD19 tyrosines and the association between CD19 and the effectors
PLC-2, Grb2, and Sos and to map associations with CD19 Y330 and
Y421. Additional studies will be required to further elucidate the
relationship of Vav and PLC-2 in binding CD19, the role of the
association of Grb2/Sos with CD19, and the conditions under which
different cytoplasmic tyrosines of CD19 are phosphorylated. However,
identification of the rich array of effectors that associate with CD19
suggests a model in which CD19 acts as a point of convergence for
multiple signaling pathways. Combining this and previous studies, we
can now assign association for most CD19 tyrosines: Y330 with Grb2/Sos,
Y391 and Y421 with Vav/PLC-2, Y403/Y443 with Fyn, and Y492/Y513 with
PI3K. Tyrosine phosphorylation of these residues could occur on a
concerted basis or individual tyrosines may be phosphorylated
independently, depending on stage of development, strength of signaling
through the BCR, complement activation, stimulation by coreceptors,
anatomical location, or other factors. By such a mechanism, CD19 could
play a modulated role in multiple signaling pathways that are central
to BCR-mediated B cell activation.
|
Acknowledgments |
|---|
|
Footnotes |
|---|
2 Address correspondence and reprint requests to Dr. Robert H. Carter, 409 LHRB, 701 South 19th Street, Birmingham, AL 35294. E-mail address:
3 Abbreviations used in this paper: BCR, B cell Ag receptor; [Ca2+]i, intracellular free calcium concentration; ERK, extracellular signal-regulated kinase; SH2, Src homology 2; mIg, membrane Ig; PI3K, phosphatidylinositol 3-kinase; mIgM, membrane IgM; WT, wild type; ex, exon; PLC, phospholipase C; WCL, whole-cell lysate.
Received for publication September 16, 1999. Accepted for publication December 27, 1999.
|
References |
|---|
|
TopAbstractIntroductionMaterials and MethodsResultsDiscussionReferences |
|---|
, Grb2, and Vav after B cell antigen receptor activation. J. Biol. Chem. 272:27362. 2 Src homology 2 domains with BLNK is critical for B cell antigen receptor signaling. J. Immunol. 163:1746.2 and Rac1-JNK in B cells. Immunity 10:117.[Medline]
This article has been cited by other articles:
|
|
 |
|
  Y. Aiba, M. Kameyama, T. Yamazaki, T. F. Tedder, and T. Kurosaki Regulation of B-cell development by BCAP and CD19 through their binding to phosphoinositide 3-kinase Blood, February 1, 2008; 111(3): 1497 - 1503. [Abstract] [Full Text] [PDF]
|
|
|
|
 |
|
 J. Yan, M. J. Wolff, J. Unternaehrer, I. Mellman, and M. J. Mamula Targeting antigen to CD19 on B cells efficiently activates T cells Int. Immunol., July 1, 2005; 17(7): 869 - 877. [Abstract] [Full Text] [PDF]
|
|
|
|
 |
|
 E. Vigorito, G. Bardi, J. Glassford, E. W.-F. Lam, E. Clayton, and M. Turner Vav-Dependent and Vav-Independent Phosphatidylinositol 3-Kinase Activation in Murine B Cells Determined by the Nature of the Stimulus J. Immunol., September 1, 2004; 173(5): 3209 - 3214. [Abstract] [Full Text] [PDF]
|
|
|
|
 |
|
 A. Cherukuri, R. H. Carter, S. Brooks, W. Bornmann, R. Finn, C. S. Dowd, and S. K. Pierce B Cell Signaling Is Regulated by Induced Palmitoylation of CD81 J. Biol. Chem., July 23, 2004; 279(30): 31973 - 31982. [Abstract] [Full Text] [PDF]
|
|
|
|
|
|
S. R. Brooks, P. M. Kirkham, L. Freeberg, and R. H. Carter Binding of Cytoplasmic Proteins to the CD19 Intracellular Domain Is High Affinity, Competitive, and Multimeric J. Immunol., June 15, 2004; 172(12): 7556 - 7564. [Abstract] [Full Text] [PDF]
|
|
|
|
|
|
A. Cherukuri, T. Shoham, H. W. Sohn, S. Levy, S. Brooks, R. Carter, and S. K. Pierce The Tetraspanin CD81 Is Necessary for Partitioning of Coligated CD19/CD21-B Cell Antigen Receptor Complexes into Signaling-Active Lipid Rafts J. Immunol., January 1, 2004; 172(1): 370 - 380. [Abstract] [Full Text] [PDF]
|
|
|
|
|
|
T. Yokozeki, K. Adler, D. Lankar, and C. Bonnerot B Cell Receptor-Mediated Syk-Independent Activation of Phosphatidylinositol 3-Kinase, Ras, and Mitogen-Activated Protein Kinase Pathways J. Immunol., August 1, 2003; 171(3): 1328 - 1335. [Abstract] [Full Text] [PDF]
|
|
|
|
 |
|
 C. H. Nielsen and R. G. Q. Leslie Complement's participation in acquired immunity J. Leukoc. Biol., August 1, 2002; 72(2): 249 - 261. [Abstract] [Full Text] [PDF]
|
|
|
|
 |
|
 J P Mitchell, E J Enyedy, M P Nambiar, A Lees, and G C Tsokos Engagement of complement receptor 2 on the surface of B cells from patients with systemic lupus erythematosus contributes to the increased responsiveness to antigen stimulation Lupus, May 1, 2002; 11(5): 299 - 303. [Abstract] [PDF]
|
|
|
|
|
|
L. Chakravarty, M. D. Zabel, J. J. Weis, and J. H. Weis Depletion of Lyn kinase from the BCR complex and inhibition of B cell activation by excess CD21 ligation Int. Immunol., February 1, 2002; 14(2): 139 - 146. [Abstract] [Full Text] [PDF]
|
|
|
|
|
|
M. Hasegawa, M. Fujimoto, J. C. Poe, D. A. Steeber, C. A. Lowell, and T. F. Tedder A CD19-Dependent Signaling Pathway Regulates Autoimmunity in Lyn-Deficient Mice J. Immunol., September 1, 2001; 167(5): 2469 - 2478. [Abstract] [Full Text] [PDF]
|
|
|
|
|
|
P. A. Zipfel, M. Grove, K. Blackburn, M. Fujimoto, T. F. Tedder, and A. M. Pendergast The c-Abl Tyrosine Kinase Is Regulated Downstream of the B Cell Antigen Receptor and Interacts with CD19 J. Immunol., December 15, 2000; 165(12): 6872 - 6879. [Abstract] [Full Text] [PDF]
|
|
| ||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||
| HOME | HELP | FEEDBACK | SUBSCRIPTIONS | ARCHIVE | SEARCH | TABLE OF CONTENTS |
