(Received for publication, November 8, 1996, and in revised form, February 24, 1997)
From the Arthritis and Rheumatism Branch, NIAMS, National Institutes of Health, Bethesda, Maryland 20892
An early event that follows aggregation of the
high affinity receptor for IgE (FcRI) is the phosphorylation of
protein tyrosines, especially those on the
- and
-subunits of the
receptor. Disaggregation of the receptors leads to their rapid
dephosphorylation, but even stably aggregated receptors undergo
continual rounds of phosphorylation and dephosphorylation. We developed
assays to study dephosphorylation of the receptors and other cellular
proteins. Whole cell extracts dephosphorylated both subunits of the
receptors rapidly and were as active against aggregated as against
disaggregated Fc
RI. Upon disaggregation, the in vivo
dephosphorylation of the Fc
RI and several other proteins followed
first-order kinetics with closely similar rate constants despite
substantial differences in the extent of phosphorylation. These results
suggest that the level of phosphorylation of Fc
RI is largely
controlled by the aggregation-induced action of kinase(s) and not from
changes in susceptibility to or activity of the phosphatases. Much of
the total phosphatase is lost when the cells are permeabilized, but the
rate of dephosphorylation of disaggregated Fc
RI was comparable in
intact and permeabilized cells. Thus, much of the activity utilized by
the cell to dephosphorylate the Fc
RI is likely to be associated with
the plasma membrane.
The high affinity IgE receptor on mast cells and basophils
(FcRI) has a central role in mediating allergic responses (1, 2).
Aggregation of the receptors results in phosphorylation of protein
tyrosines as an early event that leads to a variety of later cellular
phenomena (3, 4). The receptor itself lacks sequences typical for
intrinsic protein-tyrosine kinase activity (5), and experimentally,
receptors purified by affinity columns or well washed
immunoprecipitates show virtually no kinase activity in
vitro (6, 7). However, a variety of studies have implicated a
kinase weakly associated with the receptor (6, 8, 9). In rats, the
critical initial kinase appears to be Lyn (9). The receptor-associated
Lyn from resting cells displays tyrosine kinase activity toward
exogenous substrates, but little or no phosphorylation of the receptor
itself is observed unless the receptors are aggregated (10-13). These
observations and those made on the effect of inhibitors of phosphatases
(12, 14) imply that protein-tyrosine phosphatases (PTPs)1
are continuously modulating the resting system. The
studies with inhibitors and other experimental approaches (15) show
that the PTPs also continuously act on aggregated receptors. Finally, when individual receptors dissociate from the aggregate,
e.g. by addition of monomeric hapten after stimulation by
multivalent antigen, rapid dephosphorylation of the receptor and of
other cellular proteins is observed (7, 10-13).
The PTPs involved in these processes remain undefined, and the purpose
of the current study was to investigate some of their characteristics.
We first developed an in vitro assay to test the PTP
activity in total cell lysate toward receptors that had been
phosphorylated in response to aggregation. This assay was also used to
localize candidate PTP to subcellular fractions and to compare the
susceptibility of aggregated versus disaggregated receptors.
Finally, we examined the kinetics of dephosphorylation for FcRI and
other cellular proteins in vivo to gain insights about the
underlying regulation.
Preparation of the monoclonal anti-DNP murine IgE
from the hybridoma H1 DNP--26.82 (16) of the monoclonal antibody
against the
-subunit of Fc
RI (17) and of dinitrophenylated bovine serum albumin (DNP25-BSA, 25 mol of DNP/mol of protein)
have been described (17-19). 125I-IgE was prepared by the
chloramine-T method (20). DNP-caproate and the assay kit for lactic
dehydrogenase (used to assess the efficiency of permeabilization) were
from Sigma, an IgG fraction of goat anti-mouse IgE was from ICN
Biochemicals (Costa Mesa, CA), the permeabilizing reagent streptolysin
O was from either Life Technologies, Inc. (Gaithersburg, MD) or Murex
Diagnostics (Dartford, UK), prestained molecular weight markers used in
SDS-PAGE were from Novex (San Diego, CA), anti-SHP-1 was from Upstate
Biotechnology, Inc. (Lake Placid, NY), anti-SHP-2 was from Santa Cruz
Biotechnology Inc. (Santa Cruz, CA), anti-phosphotyrosine was from
Transduction Laboratories Inc. (Lexington, KY), and recombinant PTP,
YOP34 (21), and the PTP assay kit against a standard
tyrosine-phosphorylated peptide (22) were from Boehringer Mannheim.
The 2H3 subline of rat basophilic leukemia (RBL) cells was grown adherent in stationary culture (23) at 37 °C in a humidified atmosphere containing 5% CO2. Cells were harvested after exposure to 0.05% trypsin, 0.02% EDTA in Hanks' buffered salt solution (Biofluids). For permeabilizing sensitized adherent cells, 12 well plates (Costar Corp., Cambridge, MA) were seeded with 6 × 105 cells in the presence of 0.6 µg/ml IgE. After overnight culture, the cell monolayers were washed three times with phosphate-buffered saline (without Mg2+ or Ca2+) and then incubated for 30 min at 37 °C in the presence of 300 units/ml streptolysin O (Life Technologies, Inc.). Cells were washed twice with phosphate-buffered saline, each time allowing the wash buffer to incubate for at least 1 min before removal. The efficiency of permeabilization was assessed by following the depletion of cytosolic proteins as described under "Results." Cells were activated by adding 1 µg/ml DNP25-BSA in assay buffer (25 mM K+ PIPES, pH 7.2, 119 mM NaCl, 5 mM KCl, 5.6 mM glucose, 0.5 mM CaCl2, 1 mM MgCl2, and 2 mM ATP) at 37 °C.
Cells in suspension were sensitized with IgE, washed, and permeabilized at 1.25 × 107 cells/ml in assay buffer containing 1 unit/ml streptolysin O (Murex). The mixture was incubated for 3 min at 37 °C with gentle agitation. This protocol resulted in the permeabilization of >99% of the cells, as assessed by trypan blue uptake. Such permeabilized cells (107 cells/ml) were then stimulated for 2 min at 37 °C with 1 µg/ml DNP25-BSA in assay buffer.
Solubilization and ImmunoprecipitationCells were solubilized in lysis buffer containing 0.5% Triton X-100, 50 mM Tris, pH 7.6, 50 mM NaCl, 5 mM EDTA, 1 mM Na3VO4, 5 mM Na4P2O7, 50 mM NaF, 2 mM iodoacetate, 1 mM phenylmethylsulfonyl fluoride, and 10 µg/ml each aprotinin, leupeptin, and pepstatin A for 30 min at 4 °C. The lysates were clarified by centrifugation (16,000 × g for 10 min). The supernatants were first precleared with 50 µl of protein A-Sepharose (Pharmacia Biotech Inc.), and selected proteins were reacted with appropriate antibodies for 1 h at 4 °C. The solutions were then incubated with 60 µl of 50% protein A-Sepharose overnight at 4 °C. The beads were washed three times with 500 µl of lysis buffer and extracted with hot sample buffer (25 mM Tris, pH 6.8, 2% SDS, and 10% glycerol) by boiling for 5 min. The eluted proteins were separated by SDS-PAGE on precast gels (Novex).
Receptors aggregated with a multivalent antigen solubilize more
slowly than monomeric receptors and for this reason may be subject to
dephosphorylation during solubilization due to slow delivery of PTP
inhibitors. Therefore, except where otherwise noted, the monovalent
hapten DNP--NH2 caproate was included in the
solubilization buffer to disaggregate the receptors. When adherent
cells were solubilized by this protocol, Fc
RI but not other cellular
proteins showed enhanced levels of tyrosine phosphorylation. With cells
in suspension, where the solutions can be votexed to promote the action
of detergent (13), addition of hapten did not further enhance the
recovery of phosphotyrosine.
Analysis of protein tyrosine phosphorylation
by Western blotting with anti-phosphotyrosine antibody was performed as
described previously (15). Western blotting by various antibodies was carried out using horseradish peroxidase-conjugated secondary antibodies and enhanced chemiluminescence for detection (Amersham Corp.). Autophotographs were quantitatively analyzed by computerized densitometry (Molecular Dynamics, Inc., Sunnyvale, CA). In separate experiments, we checked that under the conditions we used here, the
chemiluminescence generated by the two critical antibodies, anti-
and anti-phosphotyrosine, was linearly proportional to the amount of
subunit and phosphotyrosine in the precipitate (data not shown).
Phosphorylated FcRI to be used as substrate for an
in vitro PTP assay was prepared from intact cells in
suspension. Stimulated cells were solubilized, and Fc
RI was
immunoprecipitated. The immunoprecipitates were first washed with lysis
buffer and then further with buffer containing 25 mM HEPES,
pH 7.2, and 5 mM EDTA.
The amount of phosphotyrosine on the receptor subunits used in the
in vitro PTP assay can be estimated as follows. On average, about 10% of the receptors that become phosphorylated in our routine stimulations can be precipitated by anti-phosphotyrosine (data not
shown). Since we routinely use 8.3 × 105 cells that
have approximately 3 × 105 receptors per cell, the
minimum number of phosphotyrosines would be 0.04 pmol. The
phosphorylation of the - and
-subunits has recently been assessed
directly, and the amount of phosphotyrosine associated with the ITAMs
of
:
2 were found to be in the ratio of
1:3.3 (24).
Then, if all phosphorylated receptors are always fully phosphorylated,
the maximal number of phosphotyrosines would be 4.3 × 0.04, or
0.17 pmol.
The total cell lysate for in vitro PTP assay was prepared by
solubilizing resting cells (4.2 × 106 cells/ml) in
0.05% Triton X-100, 25 mM HEPES, pH 7.2, 100 mM NaCl, 5 mM EDTA, 10 mM
glutathione, and 10 µg/ml each aprotinin, leupeptin, and pepstatin A
for 30 min at 4 °C. The lysates were clarified by centrifugation
(16,000 × g for 10 min). Two hundred µl of the
supernatant was added to the washed immunoprecipitates of FcRI and
incubated at 30 °C. The reaction was quenched by 1 mM
Na3VO4 at 4 °C. The immunoprecipitates were
washed once with buffer containing 25 mM HEPES, pH 7.2, 5 mM EDTA, and 1 mM
Na3VO4, extracted with hot sample buffer, and
analyzed by SDS-PAGE.
The procedures used to fractionate cells into total membranes and
cytosol have been described (12). When compared at equal cell
equivalents, >96% of the -subunit of Fc
RI was recovered in the
membrane pellet (n = 6), whereas approximately 95% of
the PTPs SHP-1 and SHP-2 were in the cytosolic fraction
(n = 3). Before the PTP assay, the fractions were
adjusted to have the same buffer composition as in the lysate.
During the in vitro PTP assay, some dissociation of the -
and
-subunits from the IgE-binding
-subunit was observed.
Therefore, to normalize the level of phosphorylation/receptor we
corrected for the amount of
. After probing with
anti-phosphotyrosine antibody, the nitrocellulose membranes were
stripped of bound antibody by incubation with a buffer containing 62.5 mM Tris, pH 6.7, 2% SDS, and 100 mM
2-mercaptoethanol for 30 min at 50 °C. The membranes were then
washed and reprobed with anti-
antibody.
The preparation of substrate (corresponding to residues 53-65 in the COOH-terminal region of hirudin (22)) for PTP assay and detection of residual phosphotyrosine after the assay were performed according to the manufacturer's instruction. Fifty µl of the cell lysate, prepared as above, was added to 4.8 pmol of peptide and incubated at 30 °C.
Our objective was to characterize the
phosphatases that with the kinase(s) regulate the degree to which the
tyrosines on the - and
-subunits of Fc
RI are phosphorylated.
We chose to examine first the phosphatase activity in total cell lysate
that could dephosphorylate receptors that had been aggregated. RBL-2H3
cells sensitized with DNP-specific IgE were reacted with antigen, and the Fc
RI was extracted with detergent. The receptors were
precipitated by reacting the receptor-bound IgE with anti-IgE and
protein A, and the precipitates were washed and subjected to PTP
assays. A detergent extract of resting RBL cells was used as the source of total PTP, and aliquots were incubated with the immunoprecipitates at 30 °C. At the end of the assay, the precipitates were extracted with SDS, and the subunits were resolved electrophoretically on polyacrylamide gels. The proteins were transferred to nitrocellulose membranes, and the latter were analyzed for phosphotyrosine by Western
blotting (Fig. 1A, anti-PY blot).
The conditions of the assay cause some of the
- and
-subunits of
Fc
RI to dissociate in unison from the IgE-bound
-subunit in the
immune complex (25). Therefore, Western blotting with anti-
was used
to quantitate the amount of
(and
) remaining in order to
normalize the phosphorylation data to a constant number of subunits
(Fig. 1B, anti-
blot). After such correction,
the data show no evidence for dephosphorylation of either the
- or
-subunits of the receptor in the presence of assay buffer only (Fig.
1C, open circles). Therefore, the
immunoprecipitates of the receptor prepared in our experiments had no
detectable associated PTP activity (cf. Ref. 26).
In the presence of total lysate, both subunits were rapidly and
completely dephosphorylated during the in vitro PTP assay (Fig. 1, A and C). By varying the dose of lysate
and the time of incubation, we developed conditions such that the
dephosphorylation of the - and
-subunits over a fixed time period
was proportional to the amount of cell lysate added (Fig.
2). These data showed that the lysate from 2.1 × 105 RBL cells hydrolyzed approximately 50% of the
phosphotyrosine on the
- and
-subunits of the receptors derived
from 8.3 × 105 cell eq/min. Thus, each cell
equivalent of lysate had adequate activity in the in vitro
assay to dephosphorylate a cell equivalent of phosphorylated Fc
RI in
approximately 30 s.
The whole cell lysate may well contain a variety of phosphatases with varying activities against the phosphorylated subunits of the receptor. Therefore, to assure that the assay was at least in principle capable of providing quantitative estimates of phosphatase activity in such a potentially complex system, we utilized the recombinant PTP, YOP34, and compared the results with it to those obtained with the RBL lysate against a phosphorylated peptide (22). The complex mixture and the recombinant phosphatase gave similarly proportional responses in a dose-response assay (data not shown).
PTPs against FcWe were interested in determining the intracellular
distribution of the PTP(s) that could dephosphorylate the receptor
subunits. Unactivated RBL cells were sonicated, and a cell-free
supernatant was prepared by centrifugation at low speed. This
supernatant was re-centrifuged at higher speed to sediment the
membranes. The latter were either washed or resuspended by sonication
in lysis buffer. The various fractions derived from equivalent numbers of cells as well as unfractionated lysate were then assayed for PTP
activity against immunoprecipitates of phosphorylated receptor or a
tyrosine-phosphorylated peptide (Table I). Relative
amounts of PTP against the receptor subunits present in each
preparation were then calculated from the extent of dephosphorylation.
These data showed that the PTP activity against the - and
-subunits was about equally distributed between cytosol and membrane
fractions. However, when sedimented and resuspended in buffer twice,
the washed membrane preparations retained only half of their initial PTP activity. When such preparations were assayed against the phosphorylated peptide, a somewhat greater fraction of the total PTP
activity was found in the cytosolic fraction.
|
We
next determined whether the cytosolic PTP dephosphorylating FcRI was
one of two PTPs that have been implicated in signal transduction
through cell surface receptors (27, 28). These are the SH2
domain-containing cytosolic enzymes, SHP-1 and SHP-2. By
immunoblotting, both SHP-1 and SHP-2 were present in RBL cell lysate
(data not shown; see Ref. 29). RBL lysate from resting cells was first
precipitated with anti-SHP-1 or anti-SHP-2, and the supernatant was
then allowed to dephosphorylate Fc
RI in vitro. The
antibodies removed about 99% of the respective PTP, as assessed by
Western blotting (data not shown). Nevertheless, the precleared supernatants dephosphorylated both
- and
-subunits as rapidly as
the control sample (data not shown), indicating that neither SHP-1 nor
SHP-2 was significantly involved in dephosphorylation of Fc
RI
in vitro.
The results of the fractionation showed that about equal amounts of PTP activity against the receptor subunits were present in the cytosol and the membranes. An alternative method was used to probe the same question, i.e. the localization of the PTP used by the cell to regulate the phosphorylation of receptor tyrosines. Sensitized adherent RBL cells permeabilized with streptolysin O were briefly stimulated with antigen, and the receptors were then disaggregated by the addition of hapten. The cells were then solubilized, and immunoprecipitates of the receptor were analyzed for residual phosphotyrosines.
Several proteins were examined to assess the loss of cytosolic components from the permeabilized cells. The average loss of lactic dehydrogenase was 88%; the corresponding losses for the two cytosolic phosphatases, SHP-1 and SHP-2, were 71 and 66%, respectively. On the other hand, the membrane-bound Lyn kinase was fully retained.
As with the intact cells, aggregating the receptors on IgE-sensitized
permeabilized cells with antigen led to a dramatic increase in
phosphotyrosine, whereas in the absence of antigen there was no signal
above background (data not shown). After addition of hapten, the rates
of dephosphorylation for the intact and permeabilized cells were
indistinguishable (Fig. 3).
PTP Dephosphorylates Aggregated or Disaggregated Fc
It was of interest to compare the dephosphorylation
of aggregated versus disaggregated receptors in the in
vitro PTP assay. Cells were activated as before, and their
receptors were solubilized (with lysis buffer that did not contain
hapten (see "Experimental Procedures")). One aliquot was
immunoprecipitated directly with anti-IgE; another was first reacted
with hapten to disaggregate the receptors before immunoprecipitation.
Fig. 4 shows that the rates of dephosphorylation of the
receptor subunits from aggregated and disaggregated receptors were
equivalent.
In this experiment it is of course important to verify that the
aggregated receptors remained aggregated during the course of the
dephosphorylation assay. Because the state of phosphorylation of only
those subunits that remained associated with the immunoprecipitates was
measured, effective disaggregation could only have occurred if there
had been substantial dissociation of the - and
-subunits during
the course of the assay. Use of anti-
blots verified that at the
earliest points (inset) >70% of the
-subunit remained with both the aggregated and hapten-disaggregated receptors.
It was of interest to
compare the in vivo dephosphorylation of the receptor with
other cellular proteins after the addition of hapten to
antigen-activated cells. We first examined the kinetics of
phosphorylation for various cellular proteins. IgE-sensitized adherent
RBL cells were stimulated with antigen, the cells were solubilized, and
the whole cell lysate (as well as immunoprecipitates of the receptor)
was assessed for phosphotyrosines. As shown in the top
panels of Fig. 5, the phosphorylation of the -
and
-subunits occurred at similar rates, reaching a maximum at 4-8
min after stimulation. These rates were about 4-fold slower than those
obtained for cells in suspension (data not shown; see Refs. 12 and 26); on the latter, using the same dose of antigen, maximum phosphorylation of the receptors was generally seen by about 1-2 min after the addition of antigen, and by 4 min it declined to about 40-75% of the
maximum. The phosphorylation of at least some of the major phosphotyrosine-containing proteins appeared to reach a plateau somewhat earlier (Fig. 5, middle panels).
We then investigated the kinetics of hapten-induced dephosphorylation
of these proteins. As shown previously, disaggregation of FcRI
induced rapid and complete dephosphorylation of the receptor subunits
(10-12). The principal other phosphotyrosine-containing proteins were
also rapidly dephosphorylated. In each case, the rate of
dephosphorylation was consistent with first-order kinetics (Fig.
6). The rate constant, k, for the
dephosphorylation of these proteins as well as their relative extent of
phosphorylation before disaggregation of Fc
RI is shown in each
panel. It is striking that the differences in the rate
constants are much smaller than the differences in the absolute levels
of phosphotyrosine.
Inhibition of Kinases Causes Dephosphorylation
We had previously shown that receptors stably aggregated by trimeric IgE are rapidly dephosphorylated when the action of kinase is inhibited (15). In the current experiments we investigated if larger aggregates of receptors formed by multivalent antigen would behave similarly. There is evidence that at least such larger aggregates become rapidly associated with specialized domains that may be critically involved in the initial signal transduction (30).
Sensitized RBL cells were permeabilized with streptolysin O and
stimulated with antigen, and EDTA was then added at various times after
stimulation to halt kinase activity. The cells were solubilized, and
immunoprecipitates of FcRI as well as the whole cell lysates were
assessed for phosphotyrosines. As previously noted for permeabilized
cells (29), the phosphorylation of the receptor was preserved, but
phosphorylation of some of the other cellular proteins was somewhat
diminished (data not shown). Addition of EDTA led to a rapid decline of
protein-bound phosphotyrosine to basal levels (Fig.
7).
Phosphorylation of tyrosines in the so-called immune recognition
tyrosine activation motifs (ITAMs) (31, 32) is the initial event
triggered by antigens when they aggregate one of the family of plasma
membrane proteins collectively referred to as the multichain immune
recognition receptors (33). With respect to FcRI, the role of the
aggregation has been intensively and quantitatively examined, but the
molecular details are still not fully elucidated. Our group has
presented direct evidence favoring a transphosphorylation mechanism
(13); others have suggested that the association of the aggregates with
specialized membrane domains (30) or immobilization of the aggregates
per se (34) also play important roles.
The phosphorylation of the ITAMs on FcRI (and likely on other
multichain immune recognition receptors) is a rapid dynamic process in
which the levels of phosphorylation are controlled by the opposing
action of kinases and phosphatases. The kinase(s) involved in these
initial events triggered by multichain immune recognition receptors are
being elucidated, but much less is known about the corresponding
phosphatase(s) (4, 35). The objectives of the present studies were to
develop a quantitative assay for the dephosphorylation of Fc
RI and
to use it to answer some fundamental questions about the responsible
tyrosine phosphatase(s).
Other than their amino acid sequences, there is virtually no structural
information on the cytoplasmic domains that contain the ITAMs.
Therefore, using arbitrary surrogates such as a peptide or a simple
chimeric construct containing one or another ITAM as substrate could
yield results that would be qualitatively or at least quantitatively
misleading. We therefore chose to utilize intact FcRI as the
substrate even though this was experimentally more demanding.
In our assays, the isolated receptors had no phosphatase associated with them (Fig. 1A). Recently Swieter et al. (26) reported such an association, but our results are not necessarily in conflict. First, Swieter et al. (26) isolated their receptors by procedures designed to minimize the dissociation of weakly interacting phosphatase, and indeed they found that the activity they measured was relatively easily dissociable. On the other hand, we deliberately used procedures that would tend to dissociate weakly interacting phosphatases, in part because it is not possible to distinguish in a straightforward way between contaminants, spurious associations, or physiological associations. Finally, we deliberately wanted to avoid assaying trace amounts of activity. Our assays were conducted at a lower temperature (30 °C versus 37 °C) and generally for much shorter times than those employed by Swieter et al. (26). Indeed, the activity we have measured appears in some instances 100-fold greater than can be estimated from their paper (see also below).
The studies employing cell fractions suggest that there are cytosolic phosphatases that in principle might participate in the dephosphorylation of receptors (Table I). However, much of the activity we observed was localized in the membrane fraction (Table I), and permeabilized cells, which had lost much of their cytosolic proteins, were as able to dephosphorylate receptors as intact cells (Fig. 3). Therefore, it appears that membrane-bound phosphatases are importantly involved. (This finding is of course consistent with a receptor-associated phosphatase such as described by Swieter et al. (26).) In vivo, a membrane-bound phosphatase could be vastly more effective than in vitro assays might reveal because of proximity and other effects (36).
That the relevant phosphatase is likely to be membrane-bound raises the
question of whether it might be CD45. Although CD45 was found to be
necessary for activating a T lymphocyte line through the FcRI with
which it had been transfected (37), the evidence that CD45 is required
for initiating IgE-mediated activation of mast cells is contradictory
(38-40). With respect to CD45, there is another consideration. As
already noted, there are interesting data related to the association of
aggregated receptors with specialized membrane domains. The latter are
resistant to solubilization by certain detergents and are enriched in
sphingolipids, glycophosphatidylinositol-anchored proteins, and
membrane-anchored kinases (30, 41). One could propose a model in which
such domains are deficient in phosphatases. Such paucity coupled with
an enrichment in kinase(s) could promote the phosphorylation of the
aggregated receptors that become associated with these specialized
regions. Indeed, CD45 is thought to be excluded from these specialized
domains (42), and the experimental data2
suggest that this exclusion may be involved in regulating the activation of the kinase Lck in T lymphocytes.
We tested the possibility that in RBL cells the specialized membrane
regions might be deficient in PTP(s) necessary to dephosphorylate FcRI but found no evidence for such a lack. We aggregated receptors under conditions shown by the Cornell investigators to promote the
association of the receptors with such domains (30, 41) and then
blocked continued kinase action with EDTA. The results showed that
there was no failure of prompt dephosphorylation of the receptors (Fig.
7). These observations parallel our previous findings on the dynamic
phosphorylation/dephosphorylation to which smaller aggregates of the
receptor are subject (15).
We also examined this latter aspect quantitatively in vitro
by comparing the susceptibility of aggregated versus
disaggregated receptors with dephosphorylation by PTP. The results
showed that aggregation did not protect the phosphorylated tyrosines
from hydrolysis (Fig. 4). Our group is attempting to analyze
quantitatively the initial response to aggregation of FcRI. A simple
scheme was proposed as a model on which to base future experiments
(Fig. 8) (43). Earlier data showed that
k
22 is substantial and effectively overwhelms
k22. The current data indicate that k
12
k
22. It is
possible that by promoting interactions with the cytoskeleton,
phosphorylation might stabilize receptors in their aggregated state,
thereby enhancing the ratio kf/kr by decreasing kr. It follows that the concentration
of the aggregated phosphorylated species, which appears to be the critical component that initiates the cascade of events, will be
independently determined by the ratio
kf/kr for both the
unphosphorylated and the phosphorylated receptors and by
k12/k
12.
The latter ratio will of course reflect not only the intrinsic properties of the enzymes involved but also their concentrations. We have previously proposed that in the cell line we are studying, the failure of phosphorylation to correlate with aggregation under some experimental conditions could be explained if the kinase responsible for k12 is limiting (43). Recent experiments have provided experimental support for this proposal (44).
Using a peptide substrate, one group has proposed that stimulation of
the RBL cells leads to an activation of a membrane-bound PTP (45). Our
studies appear to rule out any major activation of PTP that might limit
the level of phosphorylation of FcRI and some of the earliest
substrates that become tyrosine-phosphorylated. We recognize that the
data presented in Fig. 6 probably reflect the actions of multiple
phosphatases and that the individual "bands" whose phosphorylation
(Fig. 5) and dephosphorylation (Fig. 6) we assessed may represent
multiple substrates. Nevertheless, earlier data as well as all of the
results presented here suggest that the cell maintains a constant brake
on this system through the constitutive action of phosphatases.
Aggregation moves the system, principally by enhancing the
effectiveness of kinases.