(Received for publication, June 5, 1996, and in revised form, November 1, 1996)
From the The Biomedical Research Centre, 2222 Health
Sciences Mall, University of British Columbia, Vancouver, British
Columbia, V6T 1Z3 Canada, § The Pharmacology Group, School
of Pharmacy and Pharmacology, University of Bath, Bath, BA2 7AY United
Kingdom, and the
Laboratory of Cell and Molecular Biology,
National Institutes of Health, Bethesda, Maryland 20892
Insulin receptor substrate 1 (IRS-1), and its
structural relative IRS-2, are both phosphorylated on tyrosine
following treatment of cells with interleukin-4 (IL-4) and insulin. We
have investigated whether both IRS-1 and IRS-2 are expressed in murine
lymphohemopoietic cells. T and B lymphocytes and macrophages from
primary cultures expressed only IRS-2, which became phosphorylated on
tyrosine following stimulation with both IL-4 and insulin. Likewise,
the murine myeloid cell line FD-5 expressed only IRS-2, which was tyrosine phosphorylated in response to IL-4 and insulin, as well as
interleukin-3 and granulocyte-macrophage colony stimulating factor.
Neither IRS-1 nor IRS-2 were expressed at detectable levels in primary
bone marrow mast cells although these cells do respond to IL-4.
Moreover, a factor-dependent lymphocyte cell line, CT.4S, which grows continuously in IL-4, did not express detectable levels of
IRS-1 or IRS-2. IRS-2 from FD-5 cells stimulated with either IL-4 or
insulin bound to glutathione S-transferase fusion proteins of the p85 subunit of phosphoinositol 3-kinase, Grb2, and Syp, paralleling reported associations of IRS-1 with these molecules and
indicating phosphorylation of the corresponding residues on IRS-2.
IL-41 is a pleiotropic cytokine which
has distinct actions on many different cells of the lymphohemopoietic
system (1). Its most important functions include induction of
immunoglobulin class switching in B cells, the differentiation of Th2
cells, and down-regulation of many of the pro-inflammatory actions of macrophages (1). Unlike other cytokines, IL-4 fails to activate p21ras (2, 3, 4), Raf-1 (5), or the mitogen-activated protein kinases erk-1 and erk-2 (4, 6) in most cell
types. IL-4 also fails to induce tyrosine phosphorylation of Shc (7),
p120 GAP (2, 8), phospholipase C1 (8), and Syp (9). However, IL-4
activates PI 3
-kinase (8, 10, 11) potentially through inducing
association of the p85 regulatory subunit of PI 3
-kinase with a
170-kDa phosphotyrosine protein termed 4PS or p170 (8, 10, 12). IL-13
induces tyrosine phosphorylation of a similar 170-kDa protein (13,
14).
4PS shares key structural features with the insulin receptor
substrate-1 (IRS-1) molecule (15), and has been termed IRS-2. The
pleckstrin homology and phosphotyrosine-binding domains of IRS-1 and
IRS-2 are well conserved and of the 20 tyrosine residues IRS-1
contains, 14 are conserved in IRS-2 (15). Several of these form
consensus SH2 binding motifs for known signaling proteins, including
the p85 subunit of PI 3-kinase, Grb-2, Nck, and Syp (16). This has led
to the speculation that IRS-1 and IRS-2 may interact with similar
downstream elements (15) and experiments in hemopoietic cells suggest
that IRS-1/IRS-2 are critical for the mitogenic action of IL-4 (17,
18).
Not only insulin, insulin-like growth factor-1 and IL-4, but also
interferon-, interferon-
, leukemia inhibitory factor, and IL-9
can induce tyrosine phosphorylation of IRS-1 (19, 20, 21, 22, 23). In human
peripheral blood T cells, both IRS-1 and IRS-2 are expressed and
inducibly tyrosine phosphorylated by IL-2, IL-4, IL-7, and IL-15 (24).
Interestingly, IRS-1 knockout mice exhibit growth retardation and
insulin and insulin-like growth factor-1-resistance (25, 26),
indicating that IRS-2 cannot wholly substitute for IRS-1, either
because of differences in tissue distribution, or function.
We provide evidence that IRS-2 is the major substrate of tyrosine kinases activated by either IL-4 or insulin in primary murine T and B lymphocytes and macrophages, and in the murine myeloid cell line, FD-5. IRS-1 was not detectable in any of these cells. In primary bone marrow-derived mast cells and a T lymphocyte cell line, CT.4S, both of which respond to IL-4, neither IRS-1 nor IRS-2 were expressed at detectable levels. Finally, we show that like IRS-1, tyrosine-phosphorylated IRS-2 binds to p85, Grb2, and Syp.
Cells were cultured in humidified incubators at 37 °C, 5% CO2 (v/v) in RPMI 1640 medium supplemented with 10% fetal bovine serum, 20 µM 2-mercaptoethanol, 100 units of penicillin/streptomycin, and 2 mM glutamine (Life Technologies Inc., Paisley, Scotland). FDMAC11/4.5 (FD-5) and CT.4S were cultured as described previously (4, 27). Primary cultures of murine cells were derived as described using concanavalin A to activate splenic T lymphocytes (28), lipopolysaccharide to activate splenic B lymphocytes (29), and IL-3 and IL-4 to generate mast cells from bone marrow (30). Macrophages were generated by culturing bone marrow cells in media supplemented with 20% L-cell conditioned medium (as a source of CSF-1) for 3-4 weeks. Tissues were derived from BDF1 mice, with the exception of splenic B lymphocytes, which were from BALB/c (nu/nu) mice.
Cell Stimulation and ImmunoprecipitationsStimulation of cells with different growth factors was carried out as described previously (7) using chemically synthesized murine IL-4, IL-13, IL-3, and GM-CSF, kindly provided by Dr. Ian Clark-Lewis (Biomedical Research Centre, Vancouver). mIL-2 was purchased from Genzyme and porcine insulin from Sigma. L-cell conditioned media was used as a source of CSF-1. The concentration of growth factors and duration of stimulation (2-10 min) had been previously optimized to obtain maximal levels of tyrosine phosphorylation (4, 7, 14, 28). Cell pellets were lysed in immunoprecipitation buffer between 1 and 2 × 107 cells/ml, as described previously (14). Immunoprecipitations were carried out as described previously (9, 14), using either 5 µl of a polyclonal rabbit antiserum raised against the C terminus of rat IRS-1 (UBI, Lake Placid, NY, number 06-248) or 2 µl of a polyclonal rabbit antiserum raised against a GST fusion protein containing residues 619-746 of mouse IRS-2 (13).
GST Fusion ProteinsThe construction, expression,
purification, and precipitation of full-length Grb2, p85 PI 3kinase
N-SH2 domain, and Syp-SH2 domains as GST fusion proteins have been
previously described (7, 9, 31, 32).
SDS-PAGE and immunoblotting were carried out as described previously (4, 33). The polyclonal anti-IRS-1 antibody was used at 0.5 µg/ml and anti-IRS-2 antiserum at 1:1000. Blots were developed using the ECL detection system (Amersham) and subsequently stripped and reprobed as described previously (14).
We first investigated whether the 170-kDa protein (p170),
which is distinct from IRS-1 and tyrosine phosphorylated in response to
IL-4 in an IL-4-dependent myeloid cell line FD-5 (14), was IRS-2 (15). Immunoprecipitates were prepared from FD-5 cells treated
with insulin or IL-4, or from 3T3 cells treated with insulin, using
either an antiserum raised against the C terminus of IRS-1 (anti-IRS-1), that did not cross-react with IRS-2, or an antiserum that
was specific for IRS-2 (anti-IRS-2; 13). Immunoprecipitates were
separated by SDS-PAGE and immunoblotted with the antiphosphotyrosine antibody 4G10. The anti-IRS-1 antibody precipitated a
tyrosine-phosphorylated 165-kDa protein from 3T3 cells treated with
insulin (see Fig. 1, upper panel, lane 9).
This was confirmed to be IRS-1 by reprobing the same blot with the
anti-IRS-1 antibody (Fig. 1, lower panel). The anti-IRS-2
serum precipitated a protein of slightly slower mobility from 3T3 cells
following insulin stimulation (Fig. 1, upper panel, lane
10). However, this protein was not detected by immunoblotting with
anti-IRS-1 antibodies (Fig. 1, lower panel). In contrast, in
experiments using FD-5 cells treated with insulin or IL-4, the
anti-IRS-2 serum precipitated a tyrosine-phosphorylated 170-kDa protein
(Fig. 1, upper panel, lanes 4 and 6), but nothing was immunoprecipitated with the anti-IRS-1 antibody. Reprobing this
same blot with the anti-IRS-1 antibodies confirmed the absence of
detectable levels of IRS-1 protein in FD-5 cells (Fig. 1, lower panel). Reprobing IRS-2 precipitates from FD-5 with the anti-IRS-2 antibody confirmed that the 170-kDa protein was IRS-2 (see Fig. 3A, lower panel, and data not shown). Thus, FD-5 cells
expressed only IRS-2, in contrast to 3T3 cells which expressed both
IRS-1 and IRS-2. These experiments demonstrate that the antisera were operationally specific for either IRS-1 or IRS-2 (see Figs. 1 and
3A).
To determine whether the absence of IRS-1 in the FD-5 myeloid cell line
was peculiar to this cell line, we investigated the expression of IRS-1
and IRS-2 in primary cultures of murine myeloid and lymphoid cells. T
and B lymphocytes were generated by short term activation of splenic
lymphocytes and were stimulated with either insulin or IL-4 or left
untreated as a control. Immunoprecipitates were prepared using either
the IRS-1- or IRS-2-specific antisera and immunoblotted as before. The
anti-IRS-1 antibody failed to precipitate detectable
tyrosine-phosphorylated 165-kDa protein from either T or B lymphocytes
(see Fig. 2A, upper panel) stimulated with
either insulin or IL-4. Reprobing this blot with anti-IRS-1 antibodies
confirmed that no IRS-1 protein was detectable in either T or B
lymphocytes (see Fig. 2A, lower panel). However, the
anti-IRS-2 serum immunoprecipitated a 170-kDa tyrosine-phosphorylated
protein from both T cells and B cells stimulated with insulin and IL-4 (Fig. 2A, lanes 4, 6, 10, and 12), suggesting
that primary T and B lymphocytes express predominantly IRS-2, which is
a substrate for both IL-4 and insulin-induced tyrosine kinases.
Similar experiments were performed on macrophages generated from bone marrow. Once again the anti-IRS-1 antiserum failed to precipitate a tyrosine-phosphorylated 165-kDa species from insulin or IL-4 treated macrophages (Fig. 2B, lanes 3 and 5), whereas the anti-IRS-2 antiserum precipitated a 170-kDa species from the insulin- or IL-4-stimulated macrophages (see Fig. 2B, lanes 4 and 6). The same blot was reprobed with anti-IRS-2 confirming that the primary macrophages expressed IRS-2 (Fig. 2B, lower panel).
IL-4 Responsive Primary Mast Cells and an IL-4-dependent T Lymphocyte Line Failed to Express Either IRS-1 or IRS-2We performed similar experiments on primary cultures of bone marrow-derived mast cells. Neither anti-IRS-1 nor anti-IRS-2 antisera precipitated a 165-170-kDa phosphotyrosine protein from cells stimulated with IL-4 or insulin (Fig. 2C). Immunoblotting with both these antisera also failed to detect IRS-1 and IRS-2 in the immunoprecipitates (see Fig. 2C, middle and lower panels). Thus, primary cultures of mast cells do not express detectable levels of either IRS-1 or IRS-2.
In the IL-4 dependent T lymphocyte line CT.4S (4, 27) we had previously shown that IL-4 fails to induce the tyrosine phosphorylation of a 170-kDa protein, judged by immunoblotting whole cell lysates (14). Immunoprecipitation of lysates of IL-4, IL-2, or insulin-treated CT.4S, with IRS-1 or IRS-2 antisera failed to detect any tyrosine-phosphorylated species of 165-170 kDa (Fig. 2D). Immunoblotting with anti-IRS-1 or IRS-2 antibodies failed to detect either protein in the immunoprecipitates (data not shown). Thus, CT.4S, which grows continuously in IL-4, expresses neither IRS-1 nor IRS-2.
Effects of Other Cytokines on the Tyrosine Phosphorylation of IRS-1 and IRS-2We next investigated whether cytokines, other than IL-4 and insulin could induce tyrosine phosphorylation of IRS-2. FD-5 cells were either left untreated as a control or stimulated with IL-3, GM-CSF, or CSF-1, cell lysates prepared and immunoprecipitated with either anti-IRS-1 antibodies or the anti-IRS-2 serum. The anti-IRS-1 antibodies did not precipitate any 165-kDa proteins from FD-5 cells treated with IL-3, GM-CSF, or CSF-1, as would be expected (data not shown). However, following stimulation of FD-5 cells with either IL-3 or GM-CSF, IRS-2 was inducibly tyrosine-phosphorylated (Fig. 3A, upper panel). CSF-1 did not detectably affect IRS-2 phosphorylation when compared to control. Reprobing the same immunoblot with anti-IRS-2 antibodies showed the loading to be equivalent (Fig. 3A).
IL-13 induces the tyrosine phosphorylation of IRS-2 in FDC-P1 and TF-1 cells (13) and a 170-kDa protein in the murine plasmacytoma line B9 (14). Immunoprecipitation analyses demonstrated this 170-kDa protein was IRS-2, and that IRS-1 was not detectable in B9 (data not shown). Treatment of bone marrow macrophages with IL-13 resulted in the very faint, but consistent, tyrosine phosphorylation of a 170-kDa protein (Fig. 2B, lane 8; difficult to discern in reproductions) precipitated by the anti-IRS-2 serum, suggesting IRS-2 is a substrate for IL-13-induced kinases in normal cells, and cell lines. GM-CSF and CSF-1 failed to induce detectable tyrosine phosphorylation of IRS-2 in these cells.
Association of IRS-2 with SH2-containing ProteinsThe
conservation of tyrosine residues, their relative positions, and
context in IRS-1 and IRS-2 (15) raises the possibility that both IRS-1
and IRS-2 are able to interact with a similar set of SH2-containing
mediators of signaling pathways and thus, perhaps control similar
pathways. Therefore, we investigated the ability of fusion proteins of
GST and either the SH2 domains of the p85 subunit of PI 3-kinase or
Syp or full-length Grb2, to bind to tyrosine-phosphorylated IRS-2,
using cell lysates from untreated or factor-stimulated FD-5 cells.
After insulin, IL-4 and IL-3 treatment of FD-5 cells, IRS-2 co-purified with the N-terminal SH2 domain of p85 (Fig. 3B) and full-length Grb2 (Fig. 3C). The amount of IRS-2 co-precipitated following IL-3 treatment was lower in each case. As seen in Fig. 3D, a 170-kDa phosphotyrosine protein co-precipitated with the Syp-SH2 fusion protein from FD-5 cells stimulated with insulin and to a lesser extent with IL-4, but not from IL-3-treated cells, where the IL-3 receptor, AIC2A, co-precipitated with Syp-SH2.2
Here we have demonstrated that murine splenic T and B lymphocytes and bone marrow macrophages express IRS-2, which is inducibly tyrosine-phosphorylated following either IL-4 or insulin stimulation of these cells. In none of these primary murine lymphohemopoietic cultures could we detect any IRS-1. Our results with murine T lymphocytes contrast to those with human T lymphoblasts (24), where both IRS-1 and IRS-2 were expressed and tyrosine-phosphorylated in response to IL-4, IL-2, or insulin. Comparable results on primary human B lymphocytes and macrophages are not available, but it is possible that there is a species difference, the functional significance of which is not clear.
Intriguingly, when we examined primary cultures of bone marrow-derived mast cells, treated with a number of cytokines, including IL-4 and insulin, we failed to detect any tyrosine-phosphorylated proteins precipitated by either anti-IRS-1 or anti-IRS-2 antibodies. Indeed, using a combination of immunoprecipitation and immunoblotting, we were unable to detect either IRS-1 or IRS-2 in these cells. These mast cells respond to IL-4, which synergizes with other cytokines, e.g. IL-3 and SLF, in stimulating proliferation and determining their differentiation status. Therefore, neither IRS-1 nor IRS-2 appear to be necessary for the signaling pathways activated by IL-4 in the physiologically important mast cell. Such pathways could include Jak/Stat, reported to be IRS-1/2 independent in cell lines (34), and transcriptional up-regulation of the c-myc, c-fos, and c-jun genes. Likewise, the IL-4-dependent T cell line CT.4S, in which IL-4 promotes both growth and survival signals, did not express detectable levels of IRS-1 or IRS-2, again indicating that IRS-1 and IRS-2 are not essential for IL-4-stimulated mitogenesis and survival. IL-4 also induces transcription of c-fos and c-jun in CT.4S cells3 indicating that these processes do not require participation of IRS-1 or IRS-2. However, in the myeloid cell line 32D, expression of IRS-1 or IRS-2 strongly enhances the growth of these cells in response to IL-4 (18), suggesting that the requirement for signals mediated by IRS-1 or IRS-2 for cell proliferation may vary depending on the cell type.
The reasons for the apparent variation in the requirement of IRS-1 and IRS-2 for signaling in different cell types are complex and warrant further analyses. A number of potential factors exist, for example, the expression of IRS-1 and IRS-2 may be differentially regulated within particular cell lineages. In primary T and B lymphocytes and macrophages, we show that the specificity is determined by IRS protein expression, only IRS-2 is detectable. In mast cells and CT.4S, there was no detectable expression of either IRS-1 or IRS-2 and it is possible that alternative pathways are involved in IL-4 signaling in these cells. In addition, the tyrosine kinases responsible for IRS protein phosphorylation may be expressed in a cell-type specific manner. Hence, different kinases, which have distinct substrate specificities, may be recruited to the IL-4 receptor complex resulting in phosphorylation of IRS proteins at different residues, enabling coupling to different downstream pathways (see below). Further detailed molecular and biochemical analyses are required in the future to define the molecular basis and importance of the cell type variations in the requirement for IRS proteins.
The downstream pathways which are coupled to IRS-2 following IL-4 or
insulin treatment are largely uncharacterized. Our data indicate that
IRS-2, which is tyrosine-phosphorylated in response to either IL-4 or
insulin, binds the p85 subunit of PI 3-kinase and Grb2, agreeing with
other reports (13, 35). While these data suggest that IRS-2 has the
ability to couple to similar downstream effectors as IRS-1, we have
previously shown that in the same FD-5 cells, IL-4 fails to induce
tyrosine phosphorylation of Shc, activation of p21ras, or
activation of the mitogen-activated protein kinases erk1 and
erk2 (4, 7). Therefore, despite the fact that IRS-2 is
capable of binding Grb2, this is not sufficient to activate the
ras/mitogen-activated protein kinase pathway.
We also observed that the SH2 domains of the tyrosine phosphatase Syp could associate with IRS-2 from FD-5 cells that had been treated with IL-4 or insulin. It has been previously shown that binding of Syp to IRS-1 activates the catalytic activity of the phosphatase and this appears to be important for insulin-stimulated proliferation (36). Previously we have shown that IL-4 fails to induce the tyrosine phosphorylation of Syp in FD-5 cells (9). However, it is still possible that binding of Syp to IRS-2 could play a role in IL-4 signaling, distinct from activation of the ras pathway and this possibility requires further investigation.
The tyrosine phosphorylation of IRS-1 and IRS-2 is induced by a variety
of cytokines, including IL-2, IL-7, IL-9, IL-15, and interferons and
(19, 20, 21, 23). We have shown that IL-3 and GM-CSF can also
stimulate tyrosine phosphorylation of IRS-2 in FD-5 cells, although to
a lesser extent than observed with IL-4 and insulin. At least some of
the tyrosines involved include those that can bind to p85 PI 3
-kinase
and Grb2. This is in contrast to FDC-P1 and FDC-P2 cells where it has
been shown that Grb2 can associate with tyrosine-phosphorylated IRS-2
after IL-4 and insulin-like growth factor-1, but not IL-3 treatment (13). Interestingly, in FD-5 cells Syp was not able to bind to IRS-2
following IL-3 stimulation, although it did associate with the IL-3
receptor
chain, AIC2A.2 These data suggest that
cytokines differ in the efficiency with which they induce tyrosine
phosphorylation of different sites on IRS-2, and thus in coupling to
combinations of signaling molecules.
The data we have presented here demonstrates that in primary cultures of murine lymphohemopoietic cells only IRS-2 is detectably expressed. The precise role of IRS-2 in cytokine signaling in these cells remains to be determined, but it appears from our results that its phosphorylation may not necessarily be a pre-requisite for IL-4-mediated biological effects.