| HOME | HELP | FEEDBACK | SUBSCRIPTIONS | ARCHIVE | SEARCH | TABLE OF CONTENTS |
|
|
|
|||||||
|
|||||||||
Departments
*
Immunology and
of Organic Chemistry, Weizmann Institute of Science, Rehovot, Israel
 |
Abstract |
|---|
 Top Abstract IntroductionMaterials and MethodsResultsDiscussionReferences |
|---|
RI stimulus. Here we show that the tyrosine
of the ITIM undergoes phosphorylation, on MAFA clustering, that is
markedly enhanced on pervanadate treatment of the cells. Furthermore,
the Src homology 3 domain of the protein tyrosine kinase Lyn binds
directly to a peptide containing nonphosphorylated MAFA ITIM and PAAP
motif. Results of both in vitro and in vivo experiments suggest that
Lyn is probably responsible for this ITIM phosphorylation, which
increases the Src homology domain 2 (SH2) affinity of Lyn for the
peptide. In vitro measurements established that
tyrosine-phosphorylated MAFA ITIM peptides also bind the SH2 domains of
inositol 5'-phosphatase (SHIP) as well as protein tyrosine
phosphatase-2. However, the former single domain is bound 8-fold
stronger than both of the latter. Further support for the role of SHIP
in the action of MAFA stems from in vivo experiments in which
tyrosine-phosphorylated MAFA was found to bind primarily SHIP. In
RBL-2H3 cells overexpressing wild-type SHIP, MAFA clustering causes
markedly stronger inhibition of the secretory response than in control
cells expressing normal SHIP levels or cells overexpressing either
wild-type protein tyrosine phosphatase-2 or its dominant negative form.
In contrast, on overexpression of the SH2 domain of SHIP, the
inhibitory action of MAFA is essentially abolished. Taken together,
these results suggest that SHIP is the primary enzyme responsible for
mediating the inhibition by MAFA of RBL-2H3 cell response to the
FcRI stimulus. |
Introduction |
|---|
|
TopAbstractIntroductionMaterials and MethodsResultsDiscussionReferences |
|---|
More recently, the genes encoding the human and mouse MAFA homologues were cloned and sequenced. The human homologue was expressed, at least at the RNA level, by NK cells as well as by basophils. It also possesses an intracellular ITIM sequence, VIYSML, and an extracellular C-lectin-like carbohydrate recognition domain (6). The mouse MAFA homologue was recently renamed as the killer cell lectin-like receptor G1, because it is expressed by NK cells and by activated CD8 T cells but not by mast cells and is up-regulated after expression of class I MHC-encoded molecules (7, 8). An ITIM is also present in the cytoplasmic tail of mouse MAFA with a sequence (SIYSTL) identical with that of rat MAFA. The high degree of sequence conservation among the three species suggests that MAFA serves an important and evolutionary conserved immunological function that still needs to be resolved.
MAFA clustering on RBL-2H3 cells by its specific mAb (G63) suppresses
the secretory response of the cell to the type 1 FcRI stimulus
(1). Functional studies have shown that MAFA clustering
inhibits both the FcRI-mediated hydrolysis of
phosphatidylphosphoinositides and the transient rise in the
intracellular concentration of free calcium ions
([Ca2+]i) but does not
affect degranulation induced by the Ca2+
ionophore A23187. Therefore, it was concluded that MAFA interferes with
the FcRI coupling cascade upstream to phospholipase C
(PLC-)
activation (1). Further studies have shown that in cell
lysates, tyrosine-phosphorylated MAFA ITIM peptides bind two Src
homology domain 2 (SH2)-containing phosphatases, SH2 domain-containing
protein tyrosine phosphatase-2 (SHP-2) and SH2 domain-containing
inositol 5'-phosphatase (SHIP) (9). Thus far, only the
role of SHP-2 in the inhibitory function of MAFA has been investigated.
Results of these studies suggest that SHP-2 is involved in the action
of MAFA because it causes a reduction in FcRI-induced Syk
phosphorylation and activity (10). However, because
SHP-2-mediated inhibition accounts only partially for the inhibitory
effect of MAFA, it appears that SHP-2 involvement represents only one
of the pathways that participate in the inhibitory activity of MAFA.
Therefore, we investigated the role of SHIP. Evidence was now obtained
that MAFA is phosphorylated on its ITIM tyrosine by the protein
tyrosine-kinase (PTK) Lyn. Tyrosine phosphorylated MAFA was then found
to bind SHIP, thereby recruiting it to the plasma membrane of the
cells, proximal to the SHIP substrate phosphatidylinositol
3,4,5-trisphosphate (PIP3). At the plasma
membrane, SHIP presumably hydrolyzes PIP3 and
reduces its levels. This interferes with the FcRI
stimulus-response-coupling cascade and leads to suppression of the
secretory response of the mast cells. Measurements of the inhibitory
action of MAFA on the secretory response of RBL-2H3 cells
overexpressing either wild-type or mutated SHIP and SHP-2 further
support the role of SHIP as the primary mediator of the inhibitory
function of MAFA.
|
Materials and Methods |
|---|
|
TopAbstractIntroductionMaterials and MethodsResultsDiscussionReferences |
|---|
BSA conjugated with an average of 11 molecules of 2,4-dinitrobenzene (DNP11-BSA) was prepared in our laboratory from fraction V (Sigma, St. Louis, MO) by reaction with 1-fluoro-2,4-dinitrobenzene (11). Sodium vanadate was purchased from Sigma. Protein G-Sepharose 4 Fast Flow beads and ECL detection reagents were from Amersham Pharmacia Biotech U.K. (Little Chalfont, U.K.). Protease inhibitor cocktail was from Calbiochem (La Jolla, CA). Protein assay and reagents for electrophoresis were from Bio-Rad Laboratories (Rishon Le Zion, Israel). The 2,4-dinitrobenzene-specific, IgE class mAb A2IgE preparation was described previously (12). F(ab')2 of mAb G63 was prepared in our laboratory according to a previously described method (13) and conjugated to CNBr-activated Sepharose 4B (Amersham Pharmacia Biotech) according to supplier’s instruction. Polyclonal Abs specific for phosphotyrosine (PY-40) were prepared in our laboratory by rabbit immunization with a mixture of keyhole limpet hemocyanin-phosphotyrosine and tetanus toxoid-phosphotyrosine conjugates and purified as described earlier (14). Rabbit antiserum for a peptide having the 35-aa sequence of the intracellular domain of MAFA (MP35) was prepared in our laboratory by rabbit immunization with a keyhole limpet hemocyanin conjugate of the peptide (see below) prepared using m-maleimidobenzoyl-N-hydroxysuccinimide ester as cross-linker (peptide-protein (w/w) ratio, 1:2). Lyn- and Syk-specific rabbit polyclonal Abs that react with the 1–131 domain of Lyn and the 257–352 linker domain of Syk, respectively, were gifts of Dr. J. Cambier (National Jewish Center, Denver, CO). Polyclonal Abs specific for SHIP (SH2 domain) were purchased from Upstate Biotechnology (Lake Placid, NY). The recombinant vaccinia viruses encoding 1) wild-type Lyn or Syk, 2) dominant negative forms of Lyn or Syk (15), 3) wild-type SHIP, 4) a deletion mutant of SHIP containing only its SH2 domain (16), 5) the wild-type form of SHP-2, and 6) SHP-2 with a point mutation in its catalytic site were kindly provided by Dr. J.-P. Kinet (Harvard University, Boston, MA).
Peptides
MP35, a peptide with a sequence corresponding to the 35-aa of the IC domain of MAFA (MADNSIYSTLELPAAPRVQDDSRWKVKAVLHRPCV), was synthesized by Prof. C. Gilon (The Hebrew University, Jerusalem, Israel). MAFA ITIM peptides nonphosphorylated (MADNSIYSTLELC), tyrosine phosphorylated (MADNSIpYSTLELC), phosphorylated on both tyrosine and serine (MADNSIpYpSTLELC), as well as phosphorylated only on its Y-2 serine residue (MADNpSIYSTLELC) were synthesized and conjugated to beads as previously described (9). For surface plasmon resonance (SPR) studies, the MAFA peptides LDNSIYSTLEL, LDNSIpYSTLEL, ELPAAPRVQDDSR, and DNSIYSTLELPAAPR were synthesized and biotinylated at their N termini as described (9).
Cell culture and stimulation
The rat mucosal-type mast cells of the RBL-2H3 line were originally obtained from Dr. R. Siraganian (National Institutes of Health, Bethesda, MD) and grown in DMEM (Life Technologies, Gaithersburg, MD) supplemented with 10% FCS (Life Technologies), 2 mM glutamine, and combined antibiotics (Bio-Lab, Jerusalem, Israel) in a humidified atmosphere with 7% CO2 at 37°C. Cells were detached by incubation with 10 mM EDTA (in DMEM) for 10 min at 37°C. Then, they were incubated (2 x 106 cells/ml) with 2 µl/ml ascitic fluid containing the monoclonal 2,4-dinitrobenzene-specific IgE (A2IgE) for 1–2 h at 37°C in DMEM and washed (three times) with Tyrode’s buffer (130 mM NaCl, 5 mM KCl, 1.4 mM CaCl2, 1 mM MgCl2, 5.6 mM glucose, 10 mM HEPES, 0.1% BSA, pH 7.4). MAFA was clustered by incubating the cells with G63 F(ab')2 conjugated to Sepharose beads (1.5 mg Ab/ml beads); 50 x 106 cells were incubated with 40 µl beads in 1 ml Tyrode’s buffer for 5 min. In controls, the cells were incubated with the same amount of Sepharose beads conjugated with BSA. Then, the cell samples were either stimulated by Ag (100 ng/ml DNP11-BSA) for the indicated periods, or left untreated as control, all at 37°C. Pervanadate used for cell treatment was always freshly prepared by mixing 50 µl 200 mM sodium vanadate with 20 µl 30% hydrogen peroxide in 930 µl deionized water, followed by 5 min incubation at room temperature. Cells (50 x 106/sample) were incubated with 20 µl of this mixture in 1 ml Tyrode’s buffer for the indicated time at 37°C.
Vaccinia virus infection
RBL-2H3 cells were seeded in a 175-cm2 flask (15 x 106 cells/flask) or 96-well plates (7 x 104 cells/well) and cultured overnight. Cells were washed once with DMEM, 5% FCS, kept in 5 ml (for flasks) or 50 µl (for plates) DMEM, 5% FCS, and infected with 5 PFU/cell recombinant vaccinia viruses or control virus for 1.5 h at 37°C, with hand rocking of the flask at 15-min intervals. The medium was removed, 25 ml/flask or 100 µl/well DMEM/10% FCS were added, and culturing was continued for another 6–8 h at 37°C (15, 16).
Immunoprecipitation and immunoblotting
After the indicated treatments, 30 x 106 cells/ml were sedimented by centrifugation and solubilized in lysis buffer (1% Triton X-100, 50 mM HEPES, 100 mM NaF, 10 mM EDTA, 2 mM sodium orthovanadate, 10% glycerol, 10 mM sodium pyrophosphate, 1/200 dilution of protease inhibitor cocktail set III, pH 7.4) on ice for 30 min. Lysates were centrifuged for 15 min at 15,000 x g and 4°C, and after precleaning by incubation with BSA-Sepharose for 1 h at 4°C, the postnuclear supernatants were treated with G63 F(ab')2-conjugated Sepharose beads or by other specific Abs carried on protein G-Sepharose beads. After 2 h equilibration with rotation at 4°C, the beads were washed (three times) with ice-cold lysis buffer and once with PBS, and the bound proteins were eluted by boiling for 5 min in sample buffer. The eluted proteins were separated by SDS-PAGE, electrotransferred to nitrocellulose membranes, and blocked overnight at 4°C with a 0.1% gelatin solution (only for phosphotyrosine blotting) or 20 min at room temperature with 3% fat milk. This was followed by Western blotting with the specific Abs and detection by ECL.
In the experiments in which MAFA clustering was done by G63 F(ab')2 conjugated to Sepharose beads, following the indicated treatments, the cells (50 x 106/sample) and Sepharose beads were sedimented together by centrifugation and solubilized in 1% Triton X-100 lysis buffer (3 ml/sample) on ice for 10 min. The Sepharose beads were allowed to sediment by gravity and separated from the lysates. The beads were again suspended in 1 ml lysis buffer, sedimented by gravity and separated from the lysis buffer. The BSA-Sepharose beads were discarded, whereas the G63 F(ab')2-Sepharose beads were kept for later immunoisolation. This lysis buffer was combined with original cell lysates and kept on ice for another 20 min; then it was centrifuged for 15 min at 15,000 x g and 4°C to separate the nuclear debris. MAFA was immunoisolated by mAb G63 F(ab')2-conjugated Sepharose beads from the postnuclear supernatants by 4 h equilibration with rotation at 4°C. To avoid loss of MAFA, in the MAFA aggregated samples, the same Sepharose beads used for aggregation were further used for the immunoisolation.
In vitro peptide phosphorylation
Lyn or Syk was immunoprecipitated as described above from
Ag-stimulated cell lysates (20 x 106/0.7
ml). The beads carrying the proteins were washed (three times) by lysis
buffer, once with 0.5 M LiCl, followed by washing once with the kinase
assay buffer (20 mM HEPES, 100 mM NaCl, 5 mM
MnCl2, 10 mM MgCl2, pH
7.5). The in vitro kinase reaction was started by adding 50 µl of the
kinase assay buffer containing 10 µCi
[-32P]ATP, 4 µM ATP, and 2.5 µg
intracellular domain of MAFA (MAFA-IC) peptide (MP35) to the above
beads or 0.6 µg recombinant GST murine Syk (rSyk), produced by
Spodoptera frugiperda (Sf9) insect cells infected by
GST-murine Syk recombinant baculovirus (17). The reaction
was allowed to proceed for 5 min at 30°C with agitation and stopped
by adding sample buffer. Solution was boiled for 5 min, and products
were resolved by Tricine SDS-PAGE (16.5% T, 3%
C, where T denotes the total percentage
concentration of both acrylamide and bisacrylamide and C
denotes the percentage concentration of the cross-linker relative to
the total concentration T) (18), and the gel
was dried and visualized by autoradiography.
Affinity isolation by ITIM peptides
RBL-2H3 cell lysates (30 x 106cells/ml sample) were precleaned with Sepharose beads and incubated with the peptide-conjugated Sepharose beads (40 µl/ml lysate) for 2 h at 4°C with shaking. Beads were washed (four times) with ice-cold lysis buffer and once with PBS, and the bound proteins were eluted by boiling for 5 min with reducing SDS-sample buffer.
SPR analysis
Biotin or synthetic biotinylated MAFA peptides were immobilized
on streptavidin-coated sensor chips (Sensor Chip SA; BIAcore, Uppsala,
Sweden) at 25°C, at a flow rate of 5 µl/min in HEPES-buffered
saline (HBS) (10 mM HEPES, 150 mM NaCl, 3.4 mM EDTA, 0.005% surfactant
P20, pH 7.4). The concentration (density) of immobilized peptide
was determined by the change in refractive index of the hydrogel
(expressed in resonance units (RU)). All peptides used were bound at
the level of 
200–300 RU and the pY-ITIM + PAAP peptide also at the
level of 40RU (because of the high affinity of SHIP). Ligand binding
was monitored after injection (250 µl/sample) of GST fusion proteins
(9) (previously dialyzed against HBS) at different
concentrations with a flow rate of 20 µl/min in HBS running buffer.
Dissociation was monitored during subsequent washing of the chip with
HBS running buffer for either 1800 s (for SH2 and SH3 domain of
Lyn) or 3600 s (SH2 domains of SHP-2 and SHIP) with a flow rate of
20 µl/min. Chip regeneration was performed by injection of 10 µl
HBS running buffer supplemented with 0.01% SDS. SPR measurements used
the BIAcore 2000 apparatus (Pharmacia Biosensor, Uppsala, Sweden). All
real time curves were best fitted by a two-state reaction model, and
all kinetic and affinity parameters were calculated using BIAevaluation
3.0.2 software.
Secretory response assay
RBL-2H3 cells were plated in 96-well plates (7 x
104/100 µl DMEM/well) and cultured overnight at
37°C. After 6 h of infection by control vaccinia virus or
viruses containing either wild-type SHIP or a deletion mutant
containing only its SH2 domain and saturated with
A2IgE (0.2 µl/well), the cells were washed
(three times) with Tyrode’s buffer and incubated in 50 µl Tyrode’s
buffer with G63 F(ab')2 (1 x
10-7M) for 5 min at 37°C. After that,
50 µl Tyrode’s buffer containing different Ag concentrations were
added, and incubation was continued for 50 min at 37°C. Supernatant
aliquots (15 µl) were then transferred to a separate plate, and 40
µl 
-hexosaminidase substrate solution (1.3 mg/ml
p-nitrophenyl-N-acetyl--D-glucosamine
in 0.1 M citrate, pH 4.5) were added to the samples. The plate was then
incubated for 60 min at 37°C, and the reaction was terminated by
addition of 150 µl 0.2 M glycine solution, pH 10.7. The OD change
caused by substrate hydrolysis was measured at 405 nm in an ELISA plate
reader. Net secretion was calculated as percent of the total enzyme
activity measured in 1% Triton X-100-lysed cells, and all these assays
were done at least four times.
|
Results |
|---|
|
TopAbstractIntroductionMaterials and MethodsResultsDiscussionReferences |
|---|
It has previously been shown that MAFA isolated from RBL-2H3 cells
is phosphorylated at very low levels on both its tyrosine and serine
residues as detected by [32P]phosphate
(5). The tyrosine phosphorylation following MAFA
clustering by soluble mAb G63 or its F(ab')2 was
only weakly detected by Western blotting with specific Abs. Therefore,
we used mAb G63 F(ab')2 conjugated to Sepharose
beads to achieve more effective aggregation. MAFA was than further
immunoisolated by the mAb G63 F(ab')2-conjugated
Sepharose beads from the lysates. The bound proteins were eluted and
separated by a reducing SDS-PAGE, electrotransferred to nitrocellulose
membranes, and Western blotted with phosphotyrosine-specific polyclonal
Abs (PY-40). Fig. 1
A shows
that tyrosine phosphorylation of MAFA took place in the cells where
MAFA was clustered by beads carrying G63 F(ab')2,
whereas Ag stimulation did not cause that change. Similar results were
obtained from three independent experiments. The same membranes were
further blotted sequentially by Abs specific for Lyn, SHIP, or SHP-2.
All three proteins were found to coisolate with
tyrosine-phosphorylated MAFA (Fig. 1C).
Tyrosine-phosphorylated MAFA was also coisolated when Lyn
or SHIP immunoprecipitates were examined by Western blotting with
MAFA-specific Abs, although the signals were relatively weak. In
addition, Lyn coisolated with the non-tyrosine-phosphorylated MAFA from
both untreated and Ag-stimulated cells, suggesting that Lyn also binds
to the PAAP motif of MAFA via its SH3 domain, although with
considerably lower affinity (see below). On tyrosine phosphorylation of
MAFA, this coisolation was markedly enhanced (Fig. 1C).
|
RBL-2H3 cells (50 x 106/sample) were
either left untreated as control or treated with 200 µM freshly
prepared sodium pervanadate for 5 or 15 min at 37°C. The cells were
then lysed, and MAFA was isolated from the lysates by immunoaffinity
using mAb 63 F(ab')2 conjugated to Sepharose
beads. The isolated proteins were eluted and analyzed by Western
blotting with PY-40. Results (Fig. 2A) show that pervanadate
treatment markedly increased tyrosine phosphorylation of both the
dimeric and monomeric MAFA at 5 min. Significantly, the level of
tyrosine-phosphorylated MAFA declined after 15 min. Reblotting the
membrane with mAb G63 showed that all three samples contained the same
amount of MAFA (Fig. 2B).
|
To identify the PTK responsible for the tyrosine phosphorylation
of MAFA, we have first examined the in vitro tyrosine phosphorylation
of a 35-aa-long synthetic peptide with a sequence corresponding to
MAFA-IC (MP35). IgE-saturated RBL-2H3 cells (20 x
106/sample) were stimulated by Ag (100 ng/ml) for
2 min, and lysed. Lyn or Syk was isolated using polyclonal-specific Abs
carried on protein G-Sepharose beads. The in vitro peptide
phosphorylation reaction was initiated by adding 2.5 µg MP35 together
with 10 µCi [-32P]ATP to the reaction
mixture containing beads with isolated Lyn or Syk in 50 µl kinase
assay buffer (cf. Materials and Methods). Fig. 3A shows that MAFA-IC peptide
(molecular mass, 3.5 kDa) underwent phosphorylation by both
Lyn and Syk isolated from Ag-stimulated cell lysates. However,
phosphorylation by Lyn was markedly higher than that caused by Syk. In
addition, we tested the phosphorylation capacity of rSyk under the same
conditions. Although an increase in rSyk autophosphorylation was
observed, only a relatively modest increase in that of the MAFA-IC
could be detected. These results indicate that Lyn phosphorylates
MAFA-IC much more efficiently than Syk.
|
B).
Clustering MAFA by F(ab')2 of G63 also caused its
tyrosine phosphorylation yet at a very low level in cells transfected
with control viruses, and it became prominent only in those cells
overexpressing wild-type Lyn (Fig. 3B).
As stated above, Lyn was found to coisolate with MAFA even from the
untreated cells, and coisolation was markedly increased by MAFA
clustering (Fig. 1C). Coisolation of Lyn with MAFA was also
enhanced by pervanadate treatment. MAFA was isolated by immunoaffinity
using mAb G63 F(ab')2 conjugated to Sepharose
beads from the lysates of cells (50 x
106/sample in 1.7 ml) that were treated with 200
µM pervanadate for the indicated time. Lyn was monitored by Western
blotting with its specific polyclonal Ab. Fig. 3C shows that
after pervanadate treatment the amount of Lyn bound to MAFA was almost
twice that detected in controls.
Tyrosine-phosphorylated MAFA binds SHIP
RBL-2H3 cells (50 x 106/sample) were
treated by pervanadate for 5 min as before or left untreated as control
and then lysed. MAFA was immunoisolated, and the isolated proteins were
eluted and analyzed as above by Western blotting with SHIP-specific
Abs. Fig. 4A shows that the
tyrosine-phosphorylated MAFA bound SHIP and rather low amounts of
SHP-2, but no SH2 domain-containing protein tyrosine phosphatase-1
(SHP-1) could be detected (data not shown). Reblotting the membrane by
mAb G63 showed that similar amounts of MAFA were isolated from the
above samples.
|
RI
subunit associates in vitro with SHIP (19). Moreover,
evidence that some form of association exists between MAFA and the
FcRI has emerged from several different studies (Ref.
20 and unpublished data). Therefore, to examine the
possibility that MAFA may indirectly bind SHIP through the FcRI
subunit, we investigated the association of MAFA and SHIP in an RBL-2H3
subline defective in expression of the FcRI (RBL
-/- cells) (21). FACS analysis
using fluorescently labeled IgE or mAb G63 (Fig. 4B)
established that there is no FcRI
expression by the RBL
-/- cells and therefore none of the other
subunits of this receptor. Also shown in the above analysis is a lower
MAFA expression by these cells than by the regularly used parental
RBL-2H3 cells. Equal numbers (50 x
106/sample) of parental RBL and
-/- cells were either left untreated as
controls or treated by pervanadate for 5 min and lysed. MAFA was
immunoisolated by mAb G63 F(ab')2-conjugated
Sepharose beads from the lysates and analyzed as above. Fig. 4C shows that after pervanadate treatment, MAFA bound SHIP
in both cell types. Because the RBL -/-
cells express only about one-third of MAFA compared with the parental
cells (see Fig. 4B), about one-third of the amount of SHIP
was monitored by densitometry. MAFA can bind SHIP and Lyn directly through their SH2 or SH3 domains
To analyze the interactions between MAFA and SHIP or Lyn, peptides
with a sequence corresponding to residues 4–12 of the MAFA,
phosphorylated on tyrosine (MADNSIpYSTL), serine (MADNpSIYSTL), or both
(MADNSIpYpSTL), were synthesized and conjugated to Sepharose beads.
These were then used to isolate proteins from RBL-2H3 cell lysates. The
proteins bound to the different peptides were eluted, separated
as before, and Western blotted using polyclonal Abs specific for SHIP
or Lyn. The results (Fig. 5A)
show that whereas the nonphosphorylated ITIM peptide did not bind any
of these proteins, the tyrosine-phosphorylated ITIM bound both
SHIP and Lyn. The serine phosphorylated ITIM, and in particular the
tyrosine- and serine-phosphorylated ITIM peptide, also bound SHIP and
Lyn but in considerably lower amounts (Fig. 5A).
|
B and summarized in Table I. These results show that 1) none of the
examined proteins binds the nonphosphorylated MAFA ITIM and 2) neither
N- nor C- terminal SH2 domains of SHP-1 bound the MAFA pY-ITIM. 3)
Although SH2 domain of SHIP and the two SH2 domains of SHP-2 bind the
MAFA pY-ITIM peptide, the former (single SH2
domain) binds with considerably higher affinity than the two latter
domains (Kd 1.02 and 8.7 nM,
respectively) (Fig. 5B1 and Table I). 4) The SH2 domain of
Lyn also binds MAFA pY-ITIM peptide
(Kd 28 nM), whereas its SH3 domain
binds the nonphosphorylated peptide containing both ITIM and the PAAP
motif (Kd 55 nM) (Fig. 5B2). 5) The peptides containing only the ITIM or the PAAP
motif did not bind the SH3 domain of Lyn (Fig. 5B2). This
suggests that the ITIM may serve as a spacer to facilitate binding of
the SH3 domain of Lyn to the PAAP motif or help in maintaining the
peptide conformation. This finding is in line with results presented in
Fig. 1C which indicates that Lyn also binds, probably via
its SH3 domain, to the nonphosphorylated intracellular tail of intact
MAFA. On MAFA clustering, this binding may be enhanced and lead to the
initial tyrosine phosphorylation of the ITIM, which enables subsequent
binding of Lyn and SHIP to the tyrosine-phosphorylated ITIM via their
SH2 domains with the noted higher affinity.
|
RI secretory response depends on
SHIP
To obtain further evidence for the involvement of SHIP in the
MAFA-mediated inhibition of the secretory response to the FcRI
stimulus, we used the recombinant vaccinia virus system to overexpress
in RBL-2H3 cells either wild-type SHIP or a mutant containing only its
SH2 domain (15, 16). Because SHP-2 is also found to bind
to the tyrosine-phosphorylated MAFA, we have performed, in parallel,
transfections of the RBL-2H3 cells with vaccinia viruses that led to a
similar overexpression of either wt SHP-2 or of its inactive, dominant
negative mutant. The secretory response of all these cells was then
compared with those of cells infected by the control virus. After
6 h transfection, the densitometric analysis of Western blots
revealed that protein expression levels of wild-type SHIP or SHP-2 were
almost 7-fold that of control, nontransfected cells. Cells transfected
by the mutant SHIP expressed both the regular amount of wild-type SHIP
and the mutant containing only its SH2 domain (with an apparent
molecular mass of 25 kDa; data not shown).
The inhibitory action of MAFA on the FcRI-induced secretion was
subsequently assayed over a range of Ag concentrations in cells
transfected by these recombinant or control vaccinia viruses. The
-hexosaminidase secretion in response to Ag
(DNP11-BSA) stimulation after MAFA clustering by
G63 F(ab')2 was compared with that of cells that
were stimulated only by Ag (Fig. 6, A and B). Net Ag-induced secretion by the
differently transfected cells was in the range of 40–60% of their
-hexosaminidase activity content. In cells overexpressing wild-type
SHIP, MAFA induced a considerably higher inhibition (Fig. 6C) than in control transfected cells (up to 14%).
Furthermore, in the cells overexpressing only the SH2 domain of SHIP,
the inhibitory action of MAFA was essentially abolished (Fig. 6, A and C (left)). The effect of
MAFA clustering on the secretory response of cells overexpressing wt
SHP-2 or only its dn mutant was only marginally different
from that observed in the cells transfected with the control virus. All
the transfected cells exhibited a slightly lower secretion when MAFA
was clustered, and those transfected by either two recombinant SHP-2
viruses exhibited essentially the same inhibition as those transfected
with the control virus (Fig. 6, B and C
(right)) further supporting the limited role of SHP-2 in the
action of MAFA. More significantly, these results show that the degree
of secretion was markedly lower in cells overexpressing wt SHIP,
whereas the overexpression of only the SH2 domain of SHIP led to higher
degranulation. There results are consistent with recent findings
showing that SHIP has a key role in regulating the secretory response
of mast cells (38).
|
20%
(Fig. 6C).
Taken together, these results provide functional evidence that SHIP has
a major role in the inhibitory effect of MAFA on the response of
RBL-2H3 cells to the FcRI stimulus.
|
Discussion |
|---|
|
TopAbstractIntroductionMaterials and MethodsResultsDiscussionReferences |
|---|
- and Ig-chains of the B cell Ag
receptor, thereby initiating the positive downstream activating signal.
However, it also phosphorylates the ITIMs tyrosines of the FcRIIb,
CD22, and PIR-B, initiating inhibitory signaling (22, 23). Similarly, in mast cells, Lyn was shown to phosphorylate
the ITAM tyrosines of both FcRI and subunits as well as the
ITIM tyrosine of the FcRIIb (24, 25). Data emerging
from the present biochemical experiments suggest that the
tyrosine of the ITIM of MAFA is phosphorylated by Lyn in the rat
mucosal type line RBL-2H3. However, when these cells overexpressed a
dominant negative form of Lyn, the tyrosine phosphorylation of MAFA was
not clearly suppressed (Fig. 3B). This suggests that either
the remaining active Lyn in the cells is sufficient for phosphorylating
MAFA to the level observed in control cells or that other PTKs such as
Syk, especially when overexpressed, may also phosphorylate
MAFA.
We further investigated the interactions between Lyn and MAFA by
measuring the binding of either SH2 or SH3 domains of Lyn to synthetic
MAFA peptides using the SPR method. The results show a considerable
affinity (Kd 55 nM) between the SH3
domain of Lyn and a peptide containing both the MAFA-ITIM and PAAP
motif. Further, the SH2 domain of Lyn binds the tyrosine-phosphorylated
ITIM of MAFA with higher affinity (Kd
28 nM). Results of experiments in the intact cells (Figs. 1C
and 3, B and C) resolved in untreated RBL-2H3
cells a basal association between Lyn and non-tyrosine-phosphorylated
MAFA. Therefore, the association of MAFA with the SH3 domain of Lyn
probably already occurs in untreated cells and MAFA clustering further
enhances it. This association may be required for phosphorylation of
the ITIM tyrosine of MAFA by Lyn, which then takes place upon MAFA
clustering. This tyrosine phosphorylation will then enhance Lyn binding
to MAFA via its SH2 domain (Figs. 1C and 3C), and
potentially promote phosphorylation of other MAFA molecules in the
clusters.
The tyrosine-phosphorylated MAFA bound both SHIP and SHP-2 in the cells
(Fig. 1C). It has previously been shown that SHP-2 plays a
limited role in the inhibitory action of MAFA by suppressing the
activity of Syk (10). The partial inhibition therefore
suggested that SHP-2 represents only one pathway of the inhibitory
action of MAFA. SHIP has been established as a key negative regulator
("gate keeper") (26, 27, 28, 29, 30, 31, 32) of the secretory response of
mast cells. SHIP activity was proposed to set the threshold for and
limit the response to the FcRI-mediated secretory response. More
recently, SHIP was also shown to serve a similar role for the Steel
factor-induced signaling in mast cells (26). As reported
here, the tyrosine-phosphorylated MAFA ITIM binds SHIP single SH2
domain almost 9-fold stronger than it binds to combined two SH2 domains
of SHP-2. Moreover, overexpression of the SH2 domain of SHIP
essentially abolished the inhibitory function of MAFA. These results
suggest that SHIP represents another, most probably the major, element
in the inhibitory action of MAFA.
In addition to its catalytic domain, which hydrolyzes the 5'-phosphate
in inositol 1,3,4,5-tetrakisphosphate and PIP3
(27, 28), SHIP contains domains involved in mediating
protein-protein interactions. These include the N-terminal SH2 domain,
two NPXY motifs, and several proline-rich sequences in its C terminus
that resemble SH3 domain binding sites (29). These
features suggest that SHIP is able to interact with several
signal-coupling molecules and thus also serve as an adaptor. Indeed,
SHIP has recently been implicated in pathways that cause both
inhibition and feedback regulation of different cell types
(30). In particular, it has been shown to inhibit the
activation via ITAM-containing receptors of mast and B cells; SHIP was
shown to be recruited, via its SH2 domain, to the
tyrosine-phosphorylated ITIM of the FcRIIB and subsequently to
inhibit the FcRI and B cell receptor activation signals (31, 32). The generation of a SHIP knockout mouse opened the way to
further investigate its modulatory role.
SHIP-/- mast cells were found to have 4-fold
higher secretory response to the FcRI stimulus than those of
SHIP+/- or SHIP+/+ cells.
Thus, as already stated, the critical role of SHIP is to establish a
threshold for mast cell secretory response (32, 33, 34, 35, 36, 37, 38) .
Present results of overexpression of either wild-type SHIP or only its
SH2 domain in RBL-2H3 cells now provide evidence that SHIP is directly
involved in MAFA-mediated inhibition of mast cell degranulation: In the
former case, MAFA clustering induced a higher inhibition than that
observed in control cells; whereas in the latter, the inhibitory effect
of MAFA was essentially abolished. These results are rationalized by
the observation that MAFA associates with SHIP via its SH2 domain and
that overexpressing only this SH2 domain leads to a competition for the
interaction of endogenous SHIP (and probably also of SHP-2) with the
tyrosine-phosphorylated MAFA, thus functioning as a dominant negative
form of SHIP (34). SHIP recruitment by
tyrosine-phosphorylated MAFA most probably leads to translocation of
this enzyme from the cytoplasm to the cell membrane. This brings SHIP
in proximity to and action on its substrate PIP3
(27), reducing its membranal levels. This in turn,
suppresses the PH domain-mediated association of the Bruton’s tyrosine
kinase with the membrane, thereby reducing its kinase activity which is
responsible for activating PLC- (35, 36, 37). Therefore,
MAFA clustering brings about an increase in the SHIP-catalyzed
hydrolysis of PIP3 and thereby also to the
observed decrease in PLC- activity (1). This, in turn
leads to inhibition of the transient rise in
[Ca2+]i (1, 26), culminating in suppression of the secretory response of the
cells.
|
Acknowledgments |
|---|
|
Footnotes |
|---|
2 Address correspondence and reprint requests to Dr. Israel Pecht, Department of Immunology, Weizmann Institute of Science, Rehovot 76100, Israel. E-mail address: Israel.Pecht{at}weizmann.ac.il
3 Abbreviations used in this paper: MAFA, mast cell function-associated Ag; RBL-2H3, rat basophilic leukemia 2H3 cells; PTK, protein tyrosine kinase(s); PTP, protein tyrosine phosphatase(s); SHP-2, SH2 domain-containing protein tyrosine phosphatase-2; SH2 domain-containing protein tyrosine phosphatase-1 (SHP-1); SHIP, SH2 domain-containing inositol polyphosphate 5'-phosphatase; PLC-, phospholipase C; SH2, Src homology 2 domain; SH3, Src homology 3 domain; ITAM, immune receptor-based tyrosine activation motif; ITIM, immune receptor-based tyrosine inhibition motif; SPR, surface plasmon resonance; HBS, HEPES-buffered saline; PIP3, phosphatidylinositol 3,4,5-trisphosphate; DNP11-BSA, BSA conjugated with an average of 11 molecules of 2,4-dinitrobenzene; rSyk, recombinant GST murine Syk; MAFA-IC, intracellular domain of MAFA.
Received for publication March 14, 2001. Accepted for publication September 26, 2001.
|
References |
|---|
|
TopAbstractIntroductionMaterials and MethodsResultsDiscussionReferences |
|---|
or Ig heavy chain haplotype. Eur. J. Immunol. 11:527.[Medline]
RIIb1 require distinct phosphatases. J. Exp. Med. 186:473. receptor signaling. J. Immunol. 163:1110. subunit of the high affinity IgE receptor. J. Biol. Chem. 272:13991. (FcRI) and the mast cell function-associated antigen (MAFA) studied by donor photobleaching fluorescence resonance energy transfer microscopy. Eur. J. Immunol. 26:84.[Medline]
receptors on rat mucosal mast cells using a mutant FcRI-deficient rat basophilic leukemia line. Eur. J. Immunol. 25:2948.[Medline]
RI. Nature 402:B24.[Medline]
receptor I-associated lyn-dependent phosphorylation of Fc receptor IIB during negative regulation of mast cell activation. J. Immunol. 160:1647.RIIB. Nature 383:263.[Medline]
RIIB1-effector interactions and inhibitory functions. J. Immunol. 164:631.This article has been cited by other articles:
|
|
 |
|
  M. S. Tessmer, C. Fugere, F. Stevenaert, O. V. Naidenko, H. J. Chong, G. Leclercq, and L. Brossay KLRG1 binds cadherins and preferentially associates with SHIP-1 Int. Immunol., April 1, 2007; 19(4): 391 - 400. [Abstract] [Full Text] [PDF]
|
|
|
|
 |
|
 K. Yogo, M. Mizutamari, K. Mishima, H. Takenouchi, N. Ishida-Kitagawa, T. Sasaki, and T. Takeya Src Homology 2 (SH2)-Containing 5'-Inositol Phosphatase Localizes to Podosomes, and the SH2 Domain Is Implicated in the Attenuation of Bone Resorption in Osteoclasts Endocrinology, July 1, 2006; 147(7): 3307 - 3317. [Abstract] [Full Text] [PDF]
|
|
|
|
 |
|
 S. Zhang and J. H. Phillips Identification of tyrosine residues crucial for CD200R-mediated inhibition of mast cell activation J. Leukoc. Biol., February 1, 2006; 79(2): 363 - 368. [Abstract] [Full Text] [PDF]
|
|
|
|
 |
|
 H. M. Cherwinski, C. A. Murphy, B. L. Joyce, M. E. Bigler, Y. S. Song, S. M. Zurawski, M. M. Moshrefi, D. M. Gorman, K. L. Miller, S. Zhang, et al. The CD200 Receptor Is a Novel and Potent Regulator of Murine and Human Mast Cell Function J. Immunol., February 1, 2005; 174(3): 1348 - 1356. [Abstract] [Full Text] [PDF]
|
|
|
|
|
|
Y. Okoshi, S. Tahara-Hanaoka, C. Nakahashi, S.-i. Honda, A. Miyamoto, H. Kojima, T. Nagasawa, K. Shibuya, and A. Shibuya Requirement of the tyrosines at residues 258 and 270 of MAIR-I in inhibitory effect on degranulation from basophilic leukemia RBL-2H3 Int. Immunol., January 1, 2005; 17(1): 65 - 72. [Abstract] [Full Text] [PDF]
|
|
|
|
|
|
S. Zhang, H. Cherwinski, J. D. Sedgwick, and J. H. Phillips Molecular Mechanisms of CD200 Inhibition of Mast Cell Activation J. Immunol., December 1, 2004; 173(11): 6786 - 6793. [Abstract] [Full Text] [PDF]
|
|
RI stimulation was assayed as above in
RBL-2H3 cells transfected with vaccinia viruses leading either to
overexpression of wild-type (wt) or forms lacking catalytic activity
(SHIP-SH2; left) or (SHP2-SH2; right) or with
control transfected virus (both panels). Secretion
(D) was calculated as percent of the total enzyme
activity measured in 1% Triton X-100-lysed cells.
D1, Secretion of Ag-stimulated cells;
D0, secretion of cells where MAFA was
preclustered by G63 F(ab')2 and then Ag stimulated. The
percent inhibition ([I]) was calculated as
[I] = 100 x (D1
- D0)/D1. Data
are the average ± SD of the average results of at least three
experiments. Values are those in which the inhibition reached its
maximum.