(Received for publication, September 20, 1996, and in revised form, December 13, 1996)
From the Division of Clinical Immunology, Department of Medicine, The Johns Hopkins University School of Medicine, Asthma and Allergy Center, Baltimore, Maryland 21224
Interleukin 4 (IL-4) is a potent cytokine produced by T cells and to a lesser extent by tumor-associated natural killer cells, basophils, and mast cells. IL-4 treatment of T cells and macrophages leads to augmentation of their cytotoxic activity. In human B cells, IL-4 is a potent stimulator of Ig class switching from IgM to IgE. The diverse biological responses induced by IL-4 are mediated through a high affinity receptor complex (IL-4R). Although a wealth of information has accumulated regarding IL-4R, the exact mechanisms of IL-4R-mediated signaling pathways in human B cells are not well defined. In an attempt to characterize the IL-4-induced signals in human B cells, we have found that IL-4 treatment induced rapid dephosphorylation of the 85-kDa regulatory subunit of phosphatidylinositol 3-kinase. To identify the protein-tyrosine phosphatase involved in the IL-4-mediated dephosphorylation, we performed Western blot analysis using monoclonal antibodies specific to protein-tyrosine phosphatases. Upon IL-4 treatment, SHP-1 was specifically translocated to the cellular membrane fraction. Furthermore, immunoprecipitation studies revealed that SHP-1 could be specifically coimmunoprecipitated with the IL-4R as well as with phosphatidylinositol 3-kinase (p85). Collectively, our observations suggest that in addition to protein phosphorylation, protein tyrosine dephosphorylation may play a role in the IL-4-induced signaling pathways.
IL-41 is a potent cytokine with pleiotropic effects on many cell types. The biological effects of IL-4 include induction of IgE class switching, induction of proliferation in T and B cells, and up-regulation of CD23 and major histocompatibility complex class II molecules on human B cells (1-4).
The IL-4 receptor complex is present on many hematopoietic and non-hematopoietic cell lines (5). Although many members of the cytokine receptor family, including the IL-4 receptor (IL-4R), lack protein kinase consensus domains, ligand binding to these receptors results in tyrosine phosphorylation as well as dephosphorylation and subsequent biological responses (6, 7).
Early studies of Morla et al. (8) reveal IL-4-induced tyrosine phosphorylation of proteins of 110 and 170 kDa in a murine mast cell line, IC 2.9. Subsequently, Wang et al. (9) reported IL-4-induced tyrosine phosphorylation of a 170-kDa polypeptide, termed 4PS in murine myeloid cell lines. Keegan et al. (10) reported that 4PS may be antigenically and functionally similar to the insulin receptor substrate-1. Additional evidence of IL-4-induced signal transduction events was demonstrated by the IL-4-induced association of the 85-kDa subunit of phosphatidylinositol 3-kinase (p85) with the phosphorylated form of 4PS; however, p85 itself was not phosphorylated (11).
Further studies have shown that IL-4 treatment induces the association
of IL-4R with the chain of the IL-2 receptor complex (12) that
participates in IL-4-mediated signaling by
chain-associated JAK
kinases (13, 14). However, experiments by He and Malek (15) provide
evidence for the presence of two distinct IL-4R-mediated signaling
events,
chain-dependent and -independent, providing an
explanation for the pleiotropic effects exerted by IL-4. Also, Hou
et al. (16) reported that IL-4 could activate a
tyrosine-phosphorylated DNA-binding protein termed IL-4-Stat (Stat-6)
that is involved in IL-4-mediated transcription. Recently, Stat-6
knockout mice have been developed, and these mice are deficient in
biological responses to IL-4 treatment (17).
In this report we provide evidence for IL-4-induced protein tyrosine dephosphorylation of the 85-kDa subunit of the PI 3-kinase (p85). Also, we report the association of SHP-1, previously known as PTP-1C, with the IL-4R and PI 3-kinase (p85). It therefore appears that the IL-4-induced protein tyrosine dephosphorylation may play a role in the IL-4-induced signals, leading to diverse biological responses.
The highly IL-4-sensitive subtype of the human Burkitt's lymphoma B cell line, Ramos 2G6.4C10, was purchased from ATCC (Rockville, MD). Cells (1 × 105-1 × 106/ml) were grown in RPMI 1640 supplemented with 10% fetal calf serum, 0.1 mM nonessential amino acids, 1 mM sodium pyruvate, and gentamicin sulfate at 5 µg/ml at 37 °C in a 5% CO2-humidified chamber.
Cytokine Treatment and Cell FractionationRecombinant human
IL-4 was purchased from Sigma. For membrane
preparations, cells were washed once and incubated in serum-free medium
for 1.5 h before the addition of 10 ng/ml recombinant human IL-4.
After incubation for 10 min, cells were washed twice with isotonic
buffer containing 20 mM Hepes, 120 mM KCl, 5 mM magnesium acetate, and 1 mM dithiothreitol.
After washing, the cells were swollen for 5 min in hypotonic buffer
containing 10 mM Hepes (pH 7.5), 10 mM KCl, 1.5 mM magnesium acetate, and 1 mM dithiothreitol. Cells were then disrupted by Dounce homogenization with 30 strokes using a tight pestle. The buffer was made isotonic with a 0.1 volume of
10 × hypertonic buffer containing 100 mM Hepes (pH
7.5), 1100 mM KCl, 35 mM magnesium acetate, and
10 mM dithiothreitol. After the nuclei were separated by
centrifugation at 800 × g, the membrane and
cytoplasmic fractions were separated by centrifugation at 16,000 × g. The pelleted membrane fractions were washed once with
isotonic buffer and solubilized in 1 × SDS-sample buffer, or the
aliquots were frozen at 80 °C.
For the kinetics of IL-4-induced dephosphorylation, Ramos cells (5 × 105/ml) were serum-starved for 1.5 h before the
addition of 10 ng/ml recombinant human IL-4. At indicated time points,
cells were harvested and lysed in buffer containing 1% Nonidet P-40,
25 mM Tris-HCl (pH 7.5), 150 mM NaCl, 10 µg/ml leupeptin, 1 µg/ml pepstatin A, 10 mM
benzamidine, 100 µg/ml TPCK, 1 mM phenylmethylsulfonyl
fluoride, 1 mM Na3VO4, and 2 mM -glycerol phosphate. The nuclei and cell debris were
removed by centrifugation at 10,000 × g for 5 min. Equal amounts of cell extracts, as determined by BCA protein
determination assay (Pierce), were prepared by the addition of an equal
volume of 2 × SDS-PAGE buffer containing 5%
-mercaptoethanol
and boiling for 2 min. The proteins were subjected to electrophoresis
through 10% SDS-PAGE and electrotransferred to nitrocellulose membrane for Western blot analysis.
Ramos cells at 5 × 105/ml were washed
twice and lysed for 5 min in buffer containing 1% digitonin or 1%
Nonidet P-40, 25 mM Tris-HCl (pH 7.8), 150 mM
NaCl, 10 µg/ml leupeptin, 1 µg/ml pepstatin A, 10 mM
benzamidine, 100 µg/ml TPCK, 1 mM phenylmethylsulfonyl fluoride, 1 mM Na3VO4, and 2 mM -glycerol phosphate. The nuclei and cell debris were
removed by centrifugation at 10,000 × g for 5 min. The
cell extracts were precleared once with protein A-Sepharose. After the
addition of appropriate antibodies, the reactions were then allowed to
continue for 4 h at 4 °C. To remove the antibody-antigen complexes, prewashed protein A-Sepharose (Pharmacia Biotech Inc.) or
goat anti-mouse IgG-agarose (Sigma) was added to the
reaction. After incubation for 1 h, the complexes were removed by
centrifugation at 2,000 × g and washed extensively
with buffer containing 0.5% digitonin or 0.5% Nonidet P-40, 25 mM Tris-HCl (pH 7.8), and 150 mM NaCl. Samples
were prepared by the addition of an equal volume of 1 × SDS-PAGE
buffer containing 2.5%
-mercaptoethanol to the washed resin and
boiling for 2 min before electrophoresis through 10% SDS-PAGE.
For Western blot analysis, equal amounts of protein from detergent extracts, as determined by BCA protein assay (Pierce), or immunoprecipitated proteins were subjected to SDS-PAGE and electrotransferred onto nitrocellulose membranes. The nitrocellulose membranes were blocked using buffer containing 10 mM Tris-HCl (pH 7.8), 150 mM NaCl, and 3% bovine serum albumin for 24 h at 4 °C. Monoclonal antibody to phosphotyrosine (4G10) (UBI, Lake Placid, NY), SHP-1, or SHP-2 (Transduction Laboratories, Lexington, KY) were added at optimal concentrations, and the blots were incubated for 1 h at room temperature. The immunoblotted proteins were visualized using horseradish peroxidase-conjugated goat anti-mouse IgG (Sigma) and the chemiluminescence (ECL) Western blot detection system (Amersham Corp.). Stripping of nitrocellulose membranes was performed as described by the manufacturer (Amersham).
The molecular weights of proteins were determined using the Ferguson plot (18). Quantitation of the protein phosphorylation was performed by densitometric scanning of the autoradiograms using peak integration analysis on a Molecular Dynamics densitometer. The relative change in the level of membrane-bound SHP-1 was determined by normalizing the area under the peak for SHP-1 against the peaks for SHP-1 in the cytoplasmic fractions.
To determine the effects of IL-4 on the
phosphorylation pattern of cellular proteins in human B cells, Ramos
cells at 8 × 105/ml were serum-starved and treated
with 10 ng/ml IL-4. After different times posttreatment, cells were
harvested, washed, and lysed in buffer containing Nonidet P-40. Equal
amounts of detergent lysates (approximately 1 × 106
cells) were subjected to SDS-PAGE and Western blot analysis using anti-phosphotyrosine (anti-Tyr(P)) mAb 4G10. The results showed a
reproducible (n = 3) and significant decrease in the
phosphorylation state of a major polypeptide at approximately 85 kDa
(p85) after 1 min of IL-4 treatment (Fig. 1). However,
the phosphorylation state of several other polypeptides did not change
at this time point.
The 85-kDa Subunit of the PI 3-Kinase Is Rapidly Dephosphorylated upon IL-4 Treatment
Since studies by Paul and co-workers (11)
showed that the 85-kDa subunit of PI 3-kinase (p85) may participate in
the IL-4-induced signaling, we speculated that the IL-4-induced
dephosphorylated p85 polypeptide may in fact be the 85-kDa subunit of
the PI 3-kinase (p85). To determine this, we performed Western blot
analysis using mAb to PI 3-kinase (p85). Ramos cells were left
untreated or were treated with 10 ng/ml IL-4. After a 1-min incubation,
detergent extracts were prepared, and an equal amount of protein was
subjected to SDS-PAGE and Western blot analysis using
anti-phosphotyrosine mAb 4G10 (Fig. 2A,
left panel). The blot was then stripped and subjected to
immunodetection using anti-PI 3-kinase (p85) mAb. The results revealed
that p85 migrated with identical mobility to PI 3-kinase (p85) (Fig.
2A, right panel), suggesting that the two polypeptides may
be the same.
Subsequently, to confirm that IL-4 treatment could induce the dephosphorylation of PI 3-kinase (p85), we performed immunoprecipitation studies. Ramos cells were left untreated or were treated with 10 ng/ml IL-4 for 1 min. After preparation of detergent extracts and immunoprecipitation using anti-PI 3-kinase (p85), the proteins were resolved by SDS-PAGE. After electrotransfer, the proteins were visualized by Western blot analysis using anti-phosphotyrosine mAb 4G10 (Fig. 2B, left and middle panels). An identical blot was also probed with mAb to PI 3-kinase (p85) (Fig. 2B, right panel). The results revealed that the PI 3-kinase (p85) was dephosphorylated upon IL-4 treatment of Ramos cells. Interestingly, an IL-4-induced dephosphorylated polypeptide at approximately 170 kDa was also coimmunoprecipitated with PI 3-kinase (p85) (Fig. 2B, middle panel).
IL-4 Treatment Can Induce Specific Translocation of SHP-1 to the Cellular Membrane FractionOur data showed that PI 3-kinase (p85) was rapidly dephosphorylated by IL-4 treatment; therefore, we attempted to determine the phosphatase that may be involved in this event.
In concert with the signaling molecules, phosphatases are thought to play a key role in cellular proliferation and differentiation (19, 20). For example, both SHP-1 and SHP-2 have been reported to participate in insulin receptor-mediated signaling (21, 22).
Interestingly, IL-4R shares common signal transduction molecules such
as 4PS and PI 3-kinase (p85) with the insulin receptor (10, 11, 23).
Based on the reported similarities between IL-4R and the insulin
receptor, we performed Western blot experiments using monoclonal
antibodies to two candidate protein-tyrosine phosphatases, SHP-1 and
SHP-2. Ramos cells were serum-starved for 1 h and treated with 5 ng/ml recombinant human IL-4 for 15 min before cell disruption and
fractionation. The membrane and cytoplasmic fractions were subjected to
Western blot analysis using anti-SHP-1 and anti-SHP-2. The results
revealed that the level of SHP-1 increased in the membrane fractions
prepared from IL-4-treated cells (Fig. 3, left
panel). Densitometric scan of the Western blot showed that the
level of SHP-1 increased approximately 9.5-fold in membranes prepared
from IL-4-treated Ramos cells. In contrast, IL-4 treatment did not
increase the association of SHP-2 with cellular membrane fractions in
preparations (Fig. 3, right panel) (the data are
representative of three different experiments).
SHP-1 Associates with the IL-4R as Well as with PI 3-Kinase (p85)
Since our data showed that upon IL-4 treatment SHP-1 was
specifically translocated to the cellular membrane, we performed immunoprecipitation experiments to determine whether SHP-1 could associate with the IL-4R. Ramos cells were harvested and lysed using
digitonin lysis buffer. The lysate was subjected to immunoprecipitation in the presence or absence of IL-4 using anti-IL-4R serum. The immunoprecipitated proteins were subjected to SDS-PAGE and Western blot
analysis first using anti-SHP-1 mAb, and the immunoblotted proteins
were visualized by the ECL detection method. The data revealed that
SHP-1 could be coimmunoprecipitated with IL-4R (Fig. 4A, upper panel). Furthermore, the presence
of IL-4 during the immunoprecipitation resulted in a significant
increase in the association of SHP-1 with the IL-4R, suggesting an
IL-4-induced association of SHP-1 with the IL-4R. Normal rabbit serum
failed to immunoprecipitate SHP-1 (Fig. 4A, upper panel). To
determine whether SHP-2 could also be coimmunoprecipitated with the
IL-4R, the blot was then stripped and subjected to probing with mAb to SHP-2. The result revealed that SHP-2 was not associated with the IL-4R
in Ramos cells (Fig. 4A, lower panel).
It has been reported that SHP-1 and SHP-2 play a role in the insulin receptor signaling (21, 22). Since we observed the IL-4-induced PI 3-kinase (p85) dephosphorylation and SHP-1 association with the IL-4R, we tested the possibility that SHP-1 or SHP-2 may also interact with PI 3-kinase (p85) in Ramos cells. IL-4-treated Ramos cell extracts were prepared as above, and immunoprecipitation was performed with anti-PI 3-kinase (p85) mAb. Western blot analysis of the immunoprecipitated proteins was performed using, first, anti-SHP-1 (Fig. 4B, upper panel) and then, after stripping with SHP-2 (Fig. 4B, lower panel), mAb. The results revealed that SHP-1 could specifically associate with PI 3-kinase (p85) in Ramos cells. Control mAb (IgG1) failed to immunoprecipitate SHP-1 (Fig. 4B).
We have used the highly IL-4-sensitive subtype (2G6.4C10) of the human Burkitt's lymphoma B cell line Ramos. These cells are IgM+ and respond to IL-4 treatment by up-regulation of CD23 (24). In this paper we report the IL-4-induced tyrosine dephosphorylation of an 85-kDa polypeptide in Ramos cells.
This polypeptide appears to be identical to the 85-kDa regulatory subunit of PI 3-kinase (p85). Immunoprecipitation studies revealed that IL-4 treatment induced tyrosine dephosphorylation of PI 3-kinase (p85). Interestingly, an IL-4-induced tyrosine-dephosphorylated polypeptide at approximately 170 kDa was also coimmunoprecipitated with anti-PI 3-kinase (p85) (Fig. 2B, middle panel). At present the identity of the 170-kDa polypeptide (p170) is unknown, but it is possible that this molecule may represent insulin receptor substrate-2.
Also, IL-4 treatment of Ramos cells resulted in specific translocation of SHP-1 to a cellular membrane fraction. Zhao et al. (25) reported SHP-1 phosphorylation and translocation to the cellular membrane by phorbol ester treatment of a human promyelocytic leukemia cell line. Furthermore, by performing immunoprecipitation studies we showed that SHP-1 could specifically associate with IL-4R as well as with PI 3-kinase (p85), suggesting a role for this phosphatase in IL-4-induced signaling pathways.
The biological responses induced by IL-4 are transmitted by the high
affinity IL-4R present in low abundance
(102-103 copies/cell) on many hematopoietic
and non-hematopoietic cells (26). Cross-linking and immunoprecipitation
studies of cell surface radioiodinated cells showed that IL-4 treatment
induces the association of IL-4R and the chain shared by the IL-2,
IL-5, IL-7, and IL-9 receptors (12). Since the IL-4R chain does not possess any protein kinase activity, IL-4-induced signals are thought
to be mediated through IL-4R-associated protein kinase(s).
Further evidence of IL-4-induced signal transduction was demonstrated by the observation that the 85-kDa subunit (p85) of PI 3-kinase was associated with the phosphorylated form of insulin receptor substrate-2 (4PS), suggesting that this interaction may be critical to IL-4-induced activation (11). Hou et al. (16) revealed the presence of an IL-4-activated transcription factor termed Stat-6. Stat-6 is phosphorylated upon IL-4 treatment, and the phosphorylated form of Stat-6 is translocated to the nucleus where it can bind and activate the transcription of IL-4-responsive genes.
In contrast to the reported IL-4-induced protein phosphorylation
events, in vivo phosphorylation studies by Mire-Sluis and Thorpe (7) in human TF-1 cells showed the IL-4-induced
dephosphorylation of an 80-kDa phosphotyrosine polypeptide.
Pretreatment of cells with TGF-, a known down-regulator of
IL-4-induced IgE class switching (28, 29), blocked this
dephosphorylation (30). Although it is not yet determined, it is
possible that p80 (7) is the same polypeptide as p85, and the
difference in the molecular weight is due to different electrophoresis
conditions and molecular weight markers.
The discrepancy in the reports of the IL-4-induced signals may represent the differences between human and mouse systems. Alternatively, these differences in the IL-4-induced signals may be due to the different cell types used in the experiments.
In our experiments, we have detected rapid protein tyrosine dephosphorylation upon IL-4 treatment. At this point we do not know the exact physiologic relevance of the IL-4-induced PI 3-kinase (p85) dephosphorylation. But based on previous reports showing tyrosine phosphorylation of PI 3-kinase (p85) by growth factors such as platelet-derived growth factor and insulin (31, 32), it is possible that the IL-4-induced inhibition of leukemic and normal B cell proliferation (27) may be due to the IL-4-induced PI 3-kinase (p85) dephosphorylation. Alternatively, based on our preliminary results showing that Na3VO4, a potent protein-tyrosine phosphatase inhibitor, could block the IL-4-induced IgE germline transcript expression,2 it is tempting to speculate that protein tyrosine dephosphorylation may be necessary for the IL-4-induced signaling that leads to IgE class switching in human B cells. Further experiments are necessary to determine the exact role of PI 3-kinase (p85) dephosphorylation in the IL-4-induced signaling events.
We gratefully acknowledge Drs. David Proud and Stephen Desiderio (The Johns Hopkins University) for helpful suggestions and critical review of this manuscript.