| HOME | HELP | FEEDBACK | SUBSCRIPTIONS | ARCHIVE | SEARCH | TABLE OF CONTENTS |
|
|
|
|||||||
|
|||||||||
Johns Hopkins Asthma and Allergy Center, Baltimore, MD 21224
 |
Abstract |
|---|
 Top Abstract IntroductionMaterials and MethodsResultsDiscussionReferences |
|---|
and that there are no changes
in association between p21ras and p110 in
stimulated basophils. We used the generation of phospho-Akt as a marker
of the presence of phosphatidylinositol-3,4,5-trisphosphate and found
that phospho-Akt is transient on a time scale consistent with
p21ras activity. On the basis of information
obtained in these and other studies, we localize down-regulation of
IgE-mediated LTC4 secretion to a region of the signaling
cascade antecedent to p21ras activation,
downstream of phosphatidylinositol 3 kinase activity and probably
involving regulation of phosphatidylinositol-3,4,5-trisphosphate
levels. |
Introduction |
|---|
|
TopAbstractIntroductionMaterials and MethodsResultsDiscussionReferences |
|---|
This is not likely to be a subpopulation issue. From single cell studies, we know that the cytosolic calcium response that accompanies activation with any secretagogue does not occur in subpopulations of basophils (8, 9). If cells are stimulated sequentially with anti-IgE Ab and FMLP, FMLP induces further elevations in cytosolic calcium in the same cells in which an IgE-mediated elevation is observed. In other words, taken together, it appears that for the IgE-mediated signal, the basophil has down-regulated its response to this stimulus, but only with respect to histamine/LTC4 release. Since IL-4 secretion occurs much later and requires the persistence of cell surface aggregates (or newly forming aggregates) (10), the down-regulation is focused on secretory pathways leading to either histamine or LTC4 secretion. In this context, it is also notable that there is no loss of receptor and its IgE during the first hours of the reaction (11).
We do not have a very complete picture of the signaling pathways leading to degranulation (histamine release), but have recently identified some of the components probably relevant to LTC4 secretion from human basophils. We have recently demonstrated that extracellular signal-related kinase 1/2 (ERK1/2) and cytosolic phospholipase A2 (cPLA2) are phosphorylated during stimulation of basophils with anti-IgE Ab (12), FMLP, C5a, and IL-3 (13, 14). For all stimuli, inhibition of mitogen-activated protein/ERK kinase (MEK) activity with PD98059 inhibits ERK1/2 and cPLA2 phosphorylation, without effects on some other putative upstream signaling elements. In addition, for anti-IgE, FMLP, and C5a, PD98059 selectively inhibits LTC4 secretion, but not histamine and IL-4 secretion. This and other related evidence strongly support the role of ERK1/2 and cPLA2 in the secretory pathway leading to LTC4 generation. Studies in a variety of cell types have linked the MEK1/2 and ERK1/2 enzymes into a pathway that includes Raf-1 and p21ras (15), although there are alternative pathways to ERK1/2 activation (16).
Studies in rat basophilic leukemia (RBL) cells and various murine mast cells have identified some of the early signaling components that follow IgE-mediated stimulation (15, 17, 18, 19, 20, 21, 22, 23). With respect to the activation of the pathway that includes ERK1/2, MEK1/2, Raf-1, and p21ras, it has been demonstrated that activation of syk kinase, phosphorylation of shc (by syk), and the subsequent association of shc to Grb2 and son of sevenless (SOS) may be involved in the activation of p21ras. A few studies in basophils have also identified the presence and probably activation of one or two of these components. Notably, the presence and activation of syk kinase have been shown to follow stimulation of peripheral blood basophils with anti-IgE Ab (24, 25, 26, 27, 28, 29, 30). We have also recently demonstrated activation of p21ras following stimulation of human basophils with anti-IgE Ab (26).
The goal of the following studies is to determine which IgE-mediated
signaling steps persist for times that reflect the transient nature of
LTC4 release, i.e., show their maximum state of
activation between 5–8 min and quickly return to near resting levels
by 15–30 min, and which are more persistent and therefore consistent
with the long secretory process needed for IL-4 secretion. For example,
from our previous studies of ERK activation, we found that ERK
phosphorylation was transient on a time scale consistent with the
activation and cessation of LTC4 release
(12). These studies ask the question: where is control
exerted that it may differentiate between the cessation of
histamine/LTC4 release vs IL-4 secretion? A
directly testable hypothesis is that some early signaling steps will be
sustained for periods far longer than previously observed for ERK
phosphorylation. In the past, it has been useful to distinguish
basophil behavior when stimulated through Fc
RI vs stimulation with
FMLP (operating through a heterotrimeric G protein-linked receptor)
because, while sharing some functional outcomes, signaling also appears
independently regulated. We, therefore, examined some early signals in
the context of both anti-IgE Ab and FMLP to help localize the
down-regulatory alterations induced by anti-IgE.
|
Materials and Methods |
|---|
|
TopAbstractIntroductionMaterials and MethodsResultsDiscussionReferences |
|---|
The following were purchased: PIPES, BSA, EGTA, EDTA,
D-glucose, NaF,
Na4P2O7,
Na3VO4, 2-ME, Nonidet P-40,
and FMLP (Sigma, St. Louis, MO); crystallized human serum albumin (HSA;
Miles Laboratories, Elkhart, IN); FCS and RPMI 1640 containing 25
mM HEPES and L-glutamine (BioWhittaker, Walkersville, MD);
Percoll (Pharmacia, Piscataway, NJ); Tris and Tween 20 (Bio-Rad,
Hercules, CA); leupeptin, DTT, and PMSF (Boehringer Mannheim,
Indianapolis, IN); anti-phosphotyrosine mAb (4G10) and rabbit
anti-SOS1 Ab (Upstate Biotechnology, Lake Placid, NY); rabbit
anti-phospho-ERK Ab, rabbit anti-phospho-MEK Ab, rERK-2
(p42MAPK) protein, anti-phospho Akt
(Thr308-specific), anti-Akt, and biotinylated
molecular mass markers (New England Biolabs, Beverly, MA, now
Cell Signaling); anti-shc pAb (for immunoprecipitation),
anti-shc mAb (for Western blotting), anti-Grb2 mAb
(for Western blotting), anti-ras mAb, and anti-SOS1
mAb (Transduction Laboratories, San Diego, CA); rabbit anti-Grb2
Ab, rabbit anti-SOS2 Ab, anti-syk mAb, and
anti-phosphatidylinositol 3 (PI3) kinase p110 
, 
, 
,
(Santa Cruz Biotechnology, Santa Cruz, CA); anti-MEK1 mAb (MBL,
Watertown, MA); HRP-conjugated donkey anti-rabbit Ig Ab,
HRP-conjugated sheep anti-mouse Ig Ab, and protein G-Sepharose
beads (Amersham Life Science, Arlington Heights, IL); PP1 (Biomol,
Plymouth Meeting, PA); LY294002 (Calbiochem, San Diego, CA); goat
anti-human IgE Ab was prepared as previously described
(31). Stock solutions of PP1 and LY294002 were prepared in
DMSO; controls were incubated with an equal concentration of
DMSO.
Buffers
PIPES-albumin-glucose (PAG) buffer consisted of 25 mM PIPES, 110 mM NaCl, 5 mM KCl, 0.1% glucose, and 0.003% HSA. PAGCM was PAG supplemented with 1 mM CaCl2 and 1 mM MgCl2. PAG-EDTA consisted of PAG supplemented with 4 mM EDTA. Countercurrent elutriation was conducted in PAG containing 0.25% BSA in place of 0.003% HSA. ESB is NOVEX (San Diego, CA) electrophoresis sample buffer containing 5% 2-ME. Complete lysis buffer (CLB) is 20 mM Tris-HCl, pH 7.5, 2 mM EDTA, 2 mM EGTA, 100 µg/ml aprotinin, 10 mM benzamidine, 5 mM DTT, 1 mM PMSF, 100 µg/ml leupeptin, 50 mM NaF, 5 mM Na4P2O7, 1 mM Na3VO4, 1% Nonidet P-40, and 10% glycerol. Incomplete lysis buffer is CLB without the protease inhibitors, Nonidet P-40, glycerol, or vanadate. Stripping buffers were either 7 M guanidine hydrochloride or 65 mM Tris-HCl (pH 6.7), 100 mM 2-ME, and 2% SDS. The sensitivity of the subsequent blotting to the choice of stripping agent determined which of these two were used.
Basophil purification
For most of these experiments, residual cells of normal donors undergoing leukapheresis were enriched in basophils using a combination of Percoll density gradients and countercurrent flow elutriation, as previously described (32, 33). The cells were further purified by negative selection using MACS basophil isolation kit (Miltenyi Biotec, Aubum, CA). More recently, we have used a mixture of Abs for negative selection from Stem Cell Technologies (Vancouver, British Columbia, Canada; basophil purification kit) and columns from Miltenyi. The purity of basophils was determined by Alcian blue staining (34) and, from these leukapheresis packs, generally exceeded 99%.
Phosphorylation of ERKs, MEK, and Akt
The phosphorylation of ERKs, MEK, and Akt was assessed using phospho-ERK Ab (12, 13, 35) phospho-MEK Ab (36), or phospho-Akt Ab, respectively. After stimulating basophils (0.5–1 x 106 cells per sample) in PAGCM buffer, reactions were stopped by adding ice-cold PAG and microfuged for 5–10 s. After collecting the supernatant (for the measurement of histamine and LTC4), cell pellets were immediately lysed in ESB and separated on 10% or 4–20% Tris glycine gels (Novex). Electrophoresis and transfer were performed as previously described (12, 37). The membranes were immersed in TBST containing 5% nonfat dried skim milk (Carnation, Los Angeles, CA) overnight to block nonspecific binding. Immunoreactive proteins were detected using the Abs that were diluted in TBST containing 1% skim milk for 90 min. After washing, the membranes were incubated with HRP-conjugated anti-rabbit Ab for 1 h. After washing, ECL detection was performed as described previously (12, 37). In some experiments, the same membranes were sequentially blotted with both anti-phospho-ERK Ab and anti-MEK1 mAb. Between each blotting, membranes were stripped with stripping buffer (usually 7 M guanidine) (12).
Immunoprecipitation
After stimulating basophils (1.5–5 x 106 cells per sample) in PAGCM buffer at 37°C, the reactions were stopped by adding ice-cold PAG, and the tubes were microfuged for 5–10 s. The cell pellets were immediately lysed in CLB buffer and after a 10-min incubation on ice, centrifuged for 3 min at 16,000 x g to remove nuclei or undissolved components. Lysates were precleared with protein G-Sepharose beads for 1 h at 4°C to remove any nonspecific binding to the beads. The clarified lysates were then incubated with specific Ab prebound to protein G-Sepharose beads (usually 1 µg per 20 µl beads) at 4°C. After a 1-h incubation, the beads were washed three times with CLB buffer. The immunoprecipitated proteins were eluted by boiling in ESB. Electrophoresis, transfer, and immunoblotting with anti-phosphotyrosine Ab (4G10) were performed as described previously (12). The Ab was stripped from the membranes, and then membranes were reprobed with indicated Abs.
Activated ras affinity precipitation assay
Activated ras affinity precipitation assay was
performed, as described previously, with slight modifications
(38, 39). A GST fusion protein containing the RBD of
raf-1 (aa 1–149 of raf-1), which binds only
GTP-bound (activated) ras, was immobilized on
glutathione-agarose beads (Upstate Biotechnology, Lake Placid, NY).
After stimulating basophils (
5 x 106
cells per condition), reactions were stopped by adding ice-cold PAG and
microfuged for 5–10 s. The cell pellets were immediately lysed in
ras affinity precipitation buffer (25 mM HEPES, pH 7.5, 2 mM
EGTA, 150 mM NaCl, 10 mM MgCl2, 10% glycerol, 50
µg/ml aprotinin, 5 mM benzamidine, 50 µg/ml leupeptin, 25 mM NaF, 1
mM Na3VO4, 1% Nonidet
P-40, and 1 mM PMSF). Clarified lysates were incubated with the GST-RBD
beads (5 µl per sample) for 1 h at 4 °C with rocking. The
GST-RBD beads were washed three times with ras affinity
precipitation buffer. Bound proteins were eluted by boiling in ESB for
5 min. Affinity-precipitated ras was detected by
immunoblotting with anti-ras mAb.
LTC4 and histamine measurements
An RIA was performed using 100 µl supernatant to determine LTC4 levels, as previously described (31, 40). Histamine was measured by automated fluorometry (41). The percentage of total histamine release was calculated after subtraction of spontaneous histamine release (42). Each condition tested was performed in duplicate.
|
Results |
|---|
|
TopAbstractIntroductionMaterials and MethodsResultsDiscussionReferences |
|---|
, A and B, shows
that for both histamine and LTC4 release, adding
FMLP to the cells after they have reached their plateau of IgE-mediated
release, but at a time that still largely preceded IL-4 secretion (as
shown in other studies), resulted in significantly greater secretion of
both mediators (Fig. 1C will be discussed in detail below).
In addition, the secretion of both mediators was essentially
additive.
|
Before an assessment of activation kinetics, we first established
the expression and characteristics/conditions for activating signaling
elements within the canonical syk to ERK pathway. These
results are presented in Figs. 2 through 4. Previous studies have established that syk is rapidly
phosphorylated during stimulation with anti-IgE Ab, so these
data are not shown in Fig. 2. We have examined lyn kinase,
at least with respect to its phosphorylation state during
activation4 as
detected by blotting with anti-phosphotyrosine (4G10). Given the
mechanism by which this kinase is activated, it is not surprising that
there is little change in the ability of 4G10 to detect a change in
phosphorylation state. For 13 experiments, examining activation across
a range of times, there was a nonstatistically significant increase in
phosphorylation; the ratio of stimulated to control was 1.26 ±
0.17. Fig. 2A shows one example of this observation, and
also shows that for the same cells, syk kinase was
significantly phosphorylated. Not shown in the figure is that
immunoprecipitation with anti-lyn also resulted in
coimmunoprecipitation of syk kinase.
|
|
|
B shows that basophils express this protein and that
anti-IgE Ab also induces phosphorylation of shc, as we
previously demonstrated (26). In contrast, stimulation
with FMLP does not induce shc phosphorylation. Within these
experiments, in parallel tubes, histamine and
LTC4 release were measured. For FMLP stimulation,
histamine release was 94 ± 2%, and LTC4
release was 96 ± 13 pmol/106 basophils,
while for stimulation with anti-IgE Ab, histamine release was
45 ± 19% and LTC4 release was 23 ±
10 pmol/106 basophils. Tyrosine phosphorylation
of shc (p46) was also detected with fairly faint bands in a
similar manner to phosphorylation of shc (p52). Provided
that the anti-shc Ab detects the various forms
equivalently, blotting showed greater expression of shc
(p52) than shc (p46), and shc (p66) was not
detected. We further confirmed the same expression pattern of
shc (p52) and shc (p46) without shc
(p66) in whole basophil lysates using different Abs (data not shown).
Likewise, basophils express Grb2, and anti-IgE Ab induces an
association between Grb2 and shc (Fig. 2D). In
these experiments, cells ± stimulation were lysed, and Grb2 was
captured by immunoprecipitation on anti-Grb2/protein G beads.
Proteins eluted from the beads were subjected to Western blot analysis.
Shc was coimmunoprecipitated with Grb2 only in cells stimulated with
anti-IgE Ab. Once again, stimulation with FMLP did not induce the
association.
Basophils were tested for the presence of SOS protein. There are two
forms of SOS (SOS1 and SOS2), and basophils were not found to express
measurable quantities of SOS1 with two different anti-SOS1 Abs, as
compared with RBL-2H3 cells and Jurkat cells (Fig. 2C).
However, SOS2 was easily detected in these lysates. Using anti-Grb2
to immunoprecipitate proteins, Grb2 and SOS2 were found to be
associated in resting cells, and stimulation with anti-IgE Ab for 5
min did not have a significant influence on the association of these
two proteins (Fig. 2D). FMLP stimulation also did not affect
the association of Grb2 and SOS2.
We have previously shown that stimulation with anti-IgE Ab results
in activation of p21ras (26). This
assay requires a large number of basophils even when there is a strong
secretory response, so we were limited to basic observations concerning
this signaling element. Fig. 3A demonstrates that
p21ras was activated in basophils stimulated
with either FMLP or anti-IgE Ab. Notably, an FMLP-induced transient
activation of ras was observed. For these studies, cell
lysates were adsorbed to agarose beads to which the RBD in
raf-1 was linked. Activated ras protein
(ras-GTP) bound to RBD on beads, and the bead-associated
protein was eluted and analyzed by Western blotting with
anti-ras Ab (see details in Materials and
Methods). Fig. 3, B and C, demonstrates that
both anti-IgE Ab and FMLP stimulated the appearance of the
phosphorylated form of MEK-1. The bands detected by phospho-MEK Ab were
also recognized by anti-MEK1 Ab, indicating a role for MEK-1 as a
relevant MEK in human basophils (Fig. 3B). Previous studies
have already established that stimulation with anti-IgE Ab or FMLP
induces ERK1/2 phosphorylation (12). As shown in Fig. 3C, the kinetics of MEK-1 and ERK1/2 induced by stimulation
with FMLP was transient with a peak at 1 min. Five minutes after FMLP,
both MEK and ERK phosphorylation had decreased to <20% of the value
at 1 min, and by 15 min, phosphorylation had returned to levels
indistinguishable from resting cells.
To confirm that some of these steps in the activation cascade
were sensitive to the activity of the earliest kinase, lyn
kinase, the cells were treated with the src family
kinase inhibitor PP1,
4-amino-5-(4-methylphenyl)-7-(t-butyl)pyrazolo[3,4-d]pyrimidine
(43), which we have recently shown to effectively inhibit
syk phosphorylation (presumably by inhibiting
lyn) in human basophils stimulated with anti-IgE Ab
(27). Fig. 4 shows that
shc, MEK-1, and ERK1/2 phosphorylation were completely
inhibited by PP1. Notably, when basophils were stimulated with FMLP,
PP1 did not inhibit the phosphorylation of either MEK-1 or ERK1/2.
Kinetics of signaling element phosphorylation
Since 1) LTC4 secretion (and histamine
release) is complete within 15 min when the cells are stimulated with
an optimal concentration of anti-IgE Ab, 2) the peak activation of
syk phosphorylation occurs between 5 and 15 min for most
preparations of basophils (as shown below), and 3) the kinetics of ERK
phosphorylation was previously demonstrated (with a peak at 5 min)
(12), we simplified, and made more practical, the analysis
of all of the signaling elements by examining the relationship between
phosphorylation or association states at two time points, 5 or 30 min
poststimulation with anti-IgE Ab. However, to obtain a more
complete sense of the kinetic time course for one of these signaling
elements, we did focus more attention on syk
phosphorylation. Fig. 5 demonstrates the
kinetics of syk phosphorylation; it shows that
syk phosphorylation is a sustained event, persisting not
only for the primary period of interest in these studies (30 min), but
showing marked elevation after 2 h. Fig. 6 summarizes the additional findings of
these studies. Beginning at the top of Fig. 6, it can be seen that the
ratio of syk and shc phosphorylation,
Grb2/shc association, or Grb2/SOS2 association (30 min/5
min) was 0.96 ± 0.12, 0.99 ± 0.05, 1.12 ± 0.09, and
0.99 ± 0.04, respectively. In data not shown, we examined
phosphorylation of shc at 1, 5, 15, and 30 min, and found
that phosphorylation at 1 min was approximately 50% of the
phosphorylation at 5 min, with a plateau beyond 5 min. These results
indicated that there was sustained activity of the pathway extending
from syk to the association of shc/Grb2 and SOS2.
Parenthetically, we could not detect a gel shift of SOS2 following
stimulation (under conditions in which electrophoresis was performed
for a longer time period to tease out any differences in mobility). In
contrast to the sustained activation of the nonshared signaling
elements, the ratios (30 min/5 min) of activation/phosphorylation for
ras, MEK1, ERK1, and ERK2 were 0.27 ± 0.09, 0.25
± 0.04, 0.22 ± 0.14, and 0.14 ± 0.04, respectively. In
addition to these experiments in which the various elements were
measured simultaneously, we have accumulated many other studies that
demonstrate the same results for each of the components shown in Fig. 6. For example, we have additional experience examining syk
kinase and find that in addition to the results in Figs. 5 and 6, the
average behavior of syk kinase phosphorylation, for the 30
min/5 min ratio is 1.20 ± 0.11 (n = 21).
Likewise, for shc phosphorylation, the average 30 min/5 min
is 0.91 ± 0.07 (n = 6), and for
Grb2/shc association, 1.08 ± 0.06 (n =
3). For ERK phosphorylation, our experience also includes basophils
isolated from selected donors by standard venipuncture; the average 30
min/5 min ratio is 0.42 ± 0.06 (n = 12). The
clearest discrimination between ERK and syk phosphorylation
kinetics is at 60 min, in which syk phosphorylation is still
elevated (0.82 ± 0.09, n = 11, relative to the
5-min point) and ERK phosphorylation is quite low (0.09 ± 0.02,
n = 6, relative to the 5-min point).
|
|
, we demonstrated
that the sustained phosphorylation of shc required the
maintenance of antecedent signaling. These studies were done by adding
PP1 to cells that were first stimulated with anti-IgE Ab for 60
min. Shc phosphorylation was significant even after 60 min of
stimulation, but within 5 min of the addition of PP1 (10 µM)
shc phosphorylation was reduced to resting levels.
|
, sequential stimulation with
anti-IgE Ab and then FMLP indicated that FMLP-induced release was
essentially unaffected by the prior stimulation with anti-IgE Ab.
From the data shown in Fig. 6, it appears that the transition from
sustained to transient signaling occurs around or preceding the
p21ras signaling element. In Fig. 1C,
we show that the ERK1/2 phosphorylation response that follows
stimulation with FMLP is unaltered by prior stimulation with
anti-IgE Ab. In this experiment, rather than stimulate with 1 µM
FMLP to obtain the best additional mediator release, the FMLP
concentration was reduced in anticipation that any effect of prior
anti-IgE treatment might be easier to detect with a weaker second
stimulus. At 50 nM FMLP, there was no significant effect of
anti-IgE treatment on the FMLP response. At a concentration of 1
µM FMLP, there also was no difference ± prior anti-IgE Ab
(data not shown).
We have recently demonstrated that PI3 kinase activity appears to be
required for the activation of p21ras
(26). It has been shown in other cells that some isoforms
of PI3 kinase have RBD. We examined human basophils for the expression
of the three known forms of p85-regulated PI3 kinase (, , )
and, as shown in Fig. 8A,
basophils appear to express only PI3 kinase . In two experiments, we
have also detected in basophils the presence of the non-p85-regulated
PI3 kinase, p110 . We next examined whether activation with
anti-IgE Ab induced association between PI3 kinase (p110 )
and p21ras, and whether this association
persisted for at least 30 min. Fig. 8B shows that
stimulation with anti-IgE Ab did not induce any measurable increase
in association between PI3 kinase and p21ras for
either the 5-min or 30-min time points.
|
shows that
phosphorylation of Akt was transient on a time scale consistent with
the p21ras/ERK pathway activation profile. The
30 min/5 min ratio was 0.35 ± 0.07. Also shown in this figure is
the observation that the addition of LY294002 at the 30-min time point
reduced Akt phosphorylation to resting levels after only an additional
5 min of incubation. In data not shown, we found that the addition of
LY294002 at any time up to 30 min (tested at 5–25 min at 5-min
intervals) also reduced Akt phosphorylation to resting levels within 5
min. In two of these experiments, the kinetics of syk
phosphorylation was studied in parallel and the results were as found
in Fig. 5 (while Akt phosphorylation was transient). In addition, we
found that simultaneous immunoblotting with both the
anti-phospho-Akt and anti-phospho-ERK showed well-matched
kinetic profiles for ERK and Akt phosphorylation at several
concentrations of anti-IgE Ab (data not shown).
|
|
Discussion |
|---|
|
TopAbstractIntroductionMaterials and MethodsResultsDiscussionReferences |
|---|
summarizes our
experience with various signaling elements in human basophils and,
based on extensive studies in other cells types (in particular the RBL
cell) and a few pharmacological studies in basophils, the cartoon makes
tentative assignments for sequencing these events. By and large, those
elements most likely involved in LTC4 secretion
and shared among the various secretagogues, notably FMLP in these
studies, were transiently activated following an optimal stimulation
with anti-IgE Ab. These elements are also shared with IL-3
(unpublished studies) and probably C5a (although we have not yet
examined p21ras activation following C5a). The
transient activity of the ERK pathway probably accounts for the
cessation of LTC4 following either stimulus
(12), but as we have previously shown, the activity of ERK
itself does not influence histamine or IL-4 secretion (12, 14, 37). It remains possible that signaling upstream of MEK-1, e.g.,
p21ras, has an influence on histamine release
with p21ras or Raf-1 mediating another signaling
pathway. The interesting observation from these and our recent studies
of PI3 kinase signaling is that events normally found to be upstream of
p21ras, with the possible exception of PIP3, are
sustained following stimulation with anti-IgE Ab.
|
, we have now characterized three interrelated
pathways with sustained function. First, our older studies of
IgE-mediated elevations in cytosolic calcium have shown that it is
sustained for a period that remains somewhat undefined
(10). As noted in our previous studies, there are issues
related to how the basophil processes the fura 2 probe used to detect
cytosolic calcium levels that make it difficult to fully characterize
the magnitude of the late elevation relative to the early elevation.
However, it is clear that by 60 min, the cytosolic calcium response
remains elevated. We have little information from studies of human
basophils concerning the elements regulating the intracellular
Ca2+ concentration
([Ca2+]i) response. In
studies yet to be published, we have observed the expected
phosphorylation of phospholipase C, and our more recent studies have
found that there is some relationship between PI3 kinase activity and
[Ca2+]i. However, unlike
the characteristics of B cell receptor signaling in B cells, the
influence of PI3 kinase on the
[Ca2+]i response appears
only moderate for human basophils. A concentration of LY294002 that
completely inhibits histamine, LTC4, and IL-4
release inhibits the
[Ca2+]i response by only
50% (26). In the current study, we find that syk and shc phosphorylation and the IgE-mediated association between shc, Grb2, and SOS2 is maintained for at least 30 min. Some of these experiments indicate that these elements remain active for far longer. The lengthy phosphorylation of syk appears to be an indication of sustained activity because shutting it off with PP1 immediately ablates the sustained phosphorylation of shc, a putative substrate of syk. Therefore, although syk has known inhibitory phosphorylation sites (44) and not all phosphorylation of syk can be equated with its activity (45), our results suggest that it is indeed active. As near as we can discern, this arm of early signaling is sustained and therefore should have had a sustained influence on the activity of p21ras.
Sequencing of this arm of the early response is limited to our studies
using PP1 (43). This drug has proven to be a specific
inhibitor of IgE-mediated release in our hands and inhibits the
phosphorylation of syk, as expected (27). The
IC50 of 2 µM is high relative to its potency
for isolated src family kinases, but consistent with all
previous studies in intact cells (suggesting a permeability problem)
(27, 43, 46). Our results indicate the linkage from
lyn through syk and shc and
shc/Grb2 association, and are also consistent with
lyn being upstream of the ras-MEK-ERK pathway.
However, this kinase probably acts as the first step in all
FcRI-associated cascades, so its inhibition would be expected to
affect every pathway. We have examined the effects of piceatannol, a
putative selective inhibitor of syk kinase. However, in
studies to be reported in detail elsewhere (47), we found
that while piceatannol is a reasonably potent and efficacious inhibitor
of histamine release from human basophils (presumably indicating its
ability to enter the cell), it has no effect on the IgE-mediated
cytosolic calcium response or shc phosphorylation at
relevant concentrations and also inhibits PMA-induced histamine release
and FMLP-induced LTC4 release. Therefore, its
selectivity is suspect (an observation now reported by others using
other cell types (see Ref. 47 for details)). The data
either suggest that it does not inhibit syk kinase at the
concentrations relevant for its inhibition of mediator release (and
therefore must be acting on other signaling elements) or that
syk kinase is not involved in phosphorylation of
shc or the elevation of free cytosolic calcium in basophils.
This latter possibility seems unlikely given recent indications
that syk expression correlates well with the ability of the
basophil to secrete in response to stimulation through FcRI
(24, 48).
The third partially characterized arm of the early response in
basophils comes from our recent studies of PI3 kinase. The relative
positioning of PI3 kinase in the sequence of events following
stimulation varies considerably among cell types. As noted above, in
basophils, we find that inhibition of PI3 kinase with wortmannin and
LY294002 completely inhibits p21ras activation
(26). This situation appears unique to basophils, and the
reasons for this cause and effect relationship between PI3 kinase and
ras activation are, as of yet, unclear. The RBD of the p110
subunit of PI3 kinase offers one possible connection between these
two signaling elements (49, 50, 51, 52, 53). There have been in vitro
studies that suggest that the binding of ras to p110 can
influence the activity of p110, and there are indications that this
occurs in vivo (49, 50, 51, 52). However, the direction we need to
understand is how PI3 kinase activity influences ras.
There is no evidence that PI3 kinase p110 binding to ras
influences ras behavior in other cell types, but there are
sequenced reactions that begin with PI3K (54), which by
means unknown induce the activity of a c-Src kinase, which
phosphorylates shc, thereby recruiting Grb2/SOS and
activating p21ras. However, the data in
basophils are not consistent with such a scheme because PI3 kinase
inhibitors do not block shc phosphorylation or its binding
to Grb2/SOS2. Furthermore, we are trying to explain the transient
activation of ras, and the data from the current studies
show that shc/Grb2/SOS2 remain associated while
p21ras is down-regulated. We find no
stimulus-induced association between PI3 kinase and
p21ras, although there appears to be a weak
constitutive association whose significance is not clear. Therefore, in
Fig. 10, we have placed PI3K upstream of p21ras
with some indeterminate connectors. Since our data indicate that
p21ras activity is dependent on the activity of
PI3 kinase, we ask whether there is any evidence that PI3 kinase
activity is transient. Currently, this question can only be answered
indirectly. Our previous data demonstrate that p85 phosphorylation is
maintained for at least 30 min (26). The precise
significance of p85 phosphorylation as it relates to the activity of
the p110 catalytic subunit is not yet clear. The published literature
provides only conflicting conclusions (55, 56). Further
studies are necessary to understand the significance of the observation
in human basophils. Recent studies have also shown an association
between shc and PI3 kinase activation, through an
interaction with Grb2, Gab2, and p85 (57, 58). Since we
also find sustained shc phosphorylation, even this potential
pathway to PI3 kinase activation appears sustained.
Therefore, the current evidence indicates that the transient behavior
of p21ras cannot be explained by either a
transient activation of PI3 kinase (upon which it appears to depend) or
a transient association between shc, Grb2, and SOS2 (for
which we have no direct evidence that this is needed for
p21ras activity). It seems likely that the
regulation is exerted before ras because the addition of
FMLP, after the IgE-mediated activation of this pathway has reached its
near basal state, results in ERK stimulation that is similar to that
observed without prior anti-IgE Ab stimulation. If there were
active GTPase-activating protein (GAP)-like factors reducing its
activity, one might expect some effect on FMLP stimulation, unless, of
course, the pools of ras used by these two stimuli are
distinct or FMLP utilizes alternative entry points into the ERK
pathway. Therefore, the box in Fig. 10 indicating the point of control
during activation is around a currently unknown element (or chain of
elements) that leads from PI3 kinase to ras.
Just as it is unclear how PI3 kinase itself may interact with ras, so too is it unclear how the generation of PIP3 could influence ras activation. However, our data suggest that the presence of PIP3 is transient on a time scale that is consistent with the transient activity of ras. Of course, the observations are dependent on the assumption that Akt phosphorylation is an accurate indicator of the presence of PIP3. In support of this assumption, we found that the addition of LY294002 to the ongoing reaction rapidly reduced the phosphorylation of Akt. This suggests that Akt phosphorylation remains sensitive to the presence of PIP3 throughout the reaction and therefore, that its relatively rapid decline during stimulation with anti-IgE Ab could indicate the transient presence of PIP3. Taken together with the data discussed above concerning the possibly sustained activity of PI3 kinase, these results suggest that something is acting to eliminate newly generated PIP3. One obvious candidate that has currently received a great deal of attention as a gatekeeper in rodent mast cells is Src homology 2-containing inositol 5-phosphatase (59). This possibility is currently under study. However, even if PIP3 is only transiently present, we continue to need an explanation for the linkage between its presence and the activity of p21ras. Alternatively, while ras activation may be a part of some PI3 kinase-dependent pathway, it remains possible (since we haven’t shown the cause and effect relationship between ras activation and MEK activation in human basophils) that the linkage between PIP3 levels and activation of MEK-1/ERK/cPLA2 occurs through some other non-ras-dependent pathway, and the ras data are a red herring in the context of LTC4 release.
A surprising interpretation of the available data is that PI3 kinase activity is necessary for IL-4 secretion only in the earliest stages of the reaction. There has been no previous demonstration of an early reaction being required for a secretory endpoint only at the onset of the reaction. These results require further study to break down what is actually required. The increase in expression of mRNA for IL-4 significantly precedes the generation of the protein, so future studies may reveal that mRNA expression remains tightly associated with the activity of PI3 kinase, while protein generation does not. The only partial maintenance of IL-4 secretion after the addition of LY294002 at 30 min indicates that there continues to be some requirement for PI3 kinase activity even later in the reaction, but we know from previous studies that even increased expression of mRNA does not subside toward resting levels for 1–2 h. Since our previous studies have shown that cross-links are required throughout the reaction, we would tentatively conclude that some signaling is necessary throughout IL-4 secretion, just not the activity of PI3 kinase or its products.
The upstream events leading to ras activation following
stimulation by G protein-coupled receptor are also complex
(60, 61, 62, 63). However, it was not the intent of the current
studies to determine the upstream FMLP-driven events in basophils, but
to delineate which IgE-mediated pathways did or did not have
commonality to stimulation through a GTP-binding protein and to use
this information to support interpretation of results for IgE-mediated
signaling. In this context, the data in Fig. 1 demonstrated that there
was no difference in the FMLP response, with respect to signaling (ERK
phophorylation) or LTC4 secretion, between cells
having been treated with or without anti-IgE Ab. These data
strongly suggest that the down-regulation imposed on the IgE-dependent
signals occurred upstream of the otherwise shared pathway of
p21ras through ERK1/2. Several pathways for
activation of ras by a G protein-coupled receptor (e.g.,
shc/Grb2/SOS formation-dependent or independent
ras activation) are reported using a variety of cell types
and agonists. These findings are not consistent, suggesting that the
pathway leading to activation may be dependent on certain cell types or
stimuli. Activation of human basophils with FMLP did not result in
phosphorylation of shc or the association of shc
and Grb2. A similar inability for FMLP to induce shc
phosphorylation was reported for FMLP receptor-transfected cells, and
any tyrosine phosphorylation induced by FMLP is not required for
p21ras activation in human neutrophils (as
assessed using tyrosine kinase inhibitors) (63, 64),
although this point remains controversial (65, 66, 67) and may
be related to the presence and activity of PI3 kinase . Basophils
clearly express this form of PI3 kinase, but the absence of inhibition
of FMLP-induced secretion by wortmannin or LY294002 suggests that it
does not play a role in either histamine or LTC4
release. Despite the absence of shc phosphorylation in human
basophils stimulated with FMLP, p21ras was
activated to a degree equivalent to anti-IgE Ab when directly
comparing stimulation with these two stimuli. Recently, Zheng et al.
(63) have demonstrated that FMLP stimulation results in
transient inhibition of ras p120-GAP with kinetics that
correlated well with the kinetics of ras activation.
Collectively, tyrosine kinase-dependent ras guanine
nucleotide exchange factors (SOS) do not contribute to the FMLP-induced
ras activation, while inhibition of p120-GAP is associated
with ras activation. The same mechanism may exist for
activation of ras following stimulation with FMLP in human
basophils. However, further study of the events in human basophils is
needed.
In summary, these studies identify a region of down-regulation during
stimulation of human basophils with anti-IgE Ab. Stimulation of
basophils through Fc
RI is associated with transient activation of
ras and its downstream signaling elements, while many events
preceding ras activation are sustained. Notably, while PI3
kinase may remain active for long periods, one of its products, PIP3,
appears present only transiently. When combined with results from other
recent studies of IgE-mediated stimulation in human basophils, these
results suggest that down-regulation of IgE-mediated
LTC4 secretion, initiated by strong stimuli like
anti-IgE Ab, involves regulation of events between the activation
of PI3 kinase, most probably regulation of PIP3 levels, and the
activation of p21ras.
|
Acknowledgments |
|---|
|
Footnotes |
|---|
2 Address correspondence and reprint requests to Dr. Donald W. MacGlashan, Jr., Johns Hopkins Asthma and Allergy Center, 5501 Hopkins Bayview Circle, Baltimore, MD 21224. E-mail address: dmacglas{at}jhmi.edu
3 Abbreviations used in this paper: LTC4, leukotriene C4; [Ca2+]i, intracellular Ca2+ concentration; CLB, complete lysis buffer; cPLA2, cytosolic phospholipase A2; ERK, extracellular signal-regulated kinase; ESB, electrophoresis sample buffer; GAP, GTPase-activating protein; HSA, human serum albumin; MEK, mitogen-activated protein/ERK kinase; PI3, phosphatidylinositol 3; PAG, PIPES-albumin-glucose; PIP3, phosphatidylinositol-3,4,5-trisphosphate; RBD, Ras-binding domain; RBL, rat basophilic leukemia; SOS, son of sevenless.
4 As will be noted, immunoprecipitation of lyn kinase during activation results in the coimmunoprecipitation of at least one other kinase, syk. For this reason, in vitro kinase assays of immunoprecipitated lyn, unless performed with well-defined substrates, would not discriminate between the activity of the various kinases that assemble in the developing aggregate that forms during stimulation.
5 We have recently attempted to measure the presence of PIP3 using the anti-PIP3 Ab developed by Echelon (Salt Lake City, UT), but have been unsuccessful detecting PIP3 this way for the IgE-mediated reaction.
6 We have noted previously, as have other laboratories, that there is a low level of IL-4 that appears to be prestored. This is especially true in the overnight cultured basophils used in these studies. Prestored IL-4 accounts for 3–15% of the final secreted IL-4 when compared with release from basophils challenged with a concentration of anti-IgE Ab optimal for IL-4 secretion (0.02–0.06 µg/ml, depending on the basophil preparation). The concentration for optimal histamine and LTC4 release (i.e., that used in the signaling experiments, 0.2 µg/ml) induces less IL-4 secretion, but better release of the prestored IL-4, resulting in a ratio of newly synthesized to prestored of 2:1. Thus, there is essentially equivalent release of IL-4 at 10 min and 30 min (reflecting the release of prestored IL-4 with granule release), but its release is not sensitive to cycloheximide, while the newly synthesized material is sensitive to cycloheximide. The IL-4 released from 30 to 180 min represents only newly synthesized IL-4, and it was this IL-4 secretion that was used in the calculations of inhibition by LY294002 during the later part of the reaction.
Received for publication July 10, 2001. Accepted for publication October 10, 2001.
|
References |
|---|
|
TopAbstractIntroductionMaterials and MethodsResultsDiscussionReferences |
|---|
R1 is not shared by a G protein-coupled receptor. J. Biol. Chem. 270:10960.RI chain in the RBL-2H3 mast cell line: evidence for a constitutive in vivo association of Vav with Grb2, Raf-1, and ERK2 in an active complex. J. Biol. Chem. 271:26962.R1 in the mast cell line RBL2H3. J. Biol. Chem. 270:9500.RI and the Ras/mitogen-activated protein kinase signaling pathway through SOS and Grb2. J. Biol. Chem. 271:16268.RI-mediated signaling. J. Immunol. 161:6733.RI-activated tyrosine kinases Lyn, Syk, and Zap-70 in human basophils. J. Allergy Clin. Immunol. 102:304.[Medline]
RI signaling by FcRII costimulation in human blood basophils. J. Allergy Clin. Immunol. 106:337.[Medline]
