Oncogene (1997) 15, 2841 ± 2848
 1997 Stockton Press All rights reserved 0950 ± 9232/97 $12.00
Epiregulin binds to epidermal growth factor receptor and ErbB-4 and
induces tryosine phosphorylation of epidermal growth factor receptor,
ErbB-2, ErbB-3 and ErbB-4
Toshi Komurasaki, Hitoshi Toyoda, Daisuke Uchida and Shigeo Morimoto
Molecular Biology Laboratory, Medicinal Research Laboratories, Taisho Pharmaceutical Co., Ltd., 1-403, Yoshino-cho,
Ohmiyashi, Saitama 330, Japan
Epiregulin is a member of the epidermal growth factor
(EGF) family, and has certain characteristics that are
di€erent from that of EGF, including mitogenic
responses and binding to EGF receptor (EGFR).
Epiregulin may also have another cell surface receptor
and/or induces di€erent receptor heterodimerizations for
intracellular signaling. We investigated the binding
ability of epiregulin to four ErbB family receptors using
four human breast carcinoma cell lines that expressed
di€erent subsets of receptors. Chemical cross-linking
experiments showed that [125I]epiregulin directly bound to
each of EGFR and ErbB-4 but not to ErbB-2 and ErbB3. Furthermore, although epiregulin stimulated tyrosine
phosphorylation of all four ErbB receptors, the main
intracellular signal was mediated by ErbB-4 and/or
EGFR. The pattern of activation of ErbB family
receptors was di€erent from that of other EGF-related
ligands. Our ®ndings indicate that ErbB-4 and EGFR
are receptors for epiregulin, and suggest that EGFrelated ligands transduce signals for di€erent biological
responses by the hierarchical mechanism.
Keywords: epiregulin; EGF family; ErbB receptors;
tyrosine phosphorylation
Introduction
The ErbB family receptors include epidermal growth
factor receptor (EGFR), also called ErbB-1/HER1),
ErbB-2 (Neu/HER2), ErbB-3 (HER3), and ErbB-4
(HER4). These four receptors are widely expressed in
epithelial, mesenchymal, and neuronal tissues and have
been implicated in the progression of certain types of
human cancer as well as in development (Ullrich et al.,
1984; Slamon et al., 1987; Plowman et al., 1990; Lai
and Lemke, 1991; Lemoine et al., 1992; Prigent et al.,
1992; reviewed in Hynes and Stern, 1994). The
receptors possess a high degree of sequence homology
(40 ± 50%) and have a similar molecular structure
consisting of an extracellular ligand binding domain,
transmembrane domain, protein tyrosine kinase
domain, and a C-terminal autophosphorylation domain (Ullrich et al., 1984; Yamamoto et al., 1986;
Kraus et al., 1989; Plowman et al., 1990, 1993a).
Several ligands that bind to and stimulate the kinase
activity of ErbB receptors have also been reported.
These can be subdivided into two main groups based
Correspondence: T Komurasaki
Received 11 March 1997; revised 25 July 1997; accepted 28 July 1997
on their direct binding to speci®c receptors and
structure organization (Groenen et al., 1994; Lee et
al., 1995). One such group of ligands binds speci®cally
to and activates EGFR but does not interact directly
with ErbB-2, ErbB-3 or ErbB-4. These include
epidermal growth factor (EGF) (Savage et al., 1972),
transforming growth factor-a (TGF-a) (Marquardt et
al., 1984), amphiregulin (AR) (Shoyab et al., 1989),
also known as a schwanoma derived growth factor
(Kimura et al., 1990) or keratinocyte autocrine growth
factor (Cook et al., 1991), heparin binding-EGF likegrowth factor (HB-EGF) (Higashiyama et al., 1991),
and betacellulin (Shing et al., 1993).
TGF-a, AR, or HB-EGF stimulates phosphorylation
of ErbB-2 and ErbB-3 and ErbB-4 in an EGFRdependent manner (Beerli and Hynes, 1996; Riese et
al., 1996b). Notably, a recent report shows that BTC
activates EGFR and ErbB-4 in the absence of
additional ErbB family receptor expression (Riese et
al., 1996a). The other group, neuregulins, is comprised
of alternative spliced isoforms from a single gene. The
group includes neu di€erentiation factor (NDF) (Wen
et al., 1992)/heregulin (HRG) (Homes et al., 1992),
acetylcholine receptor-inducing activity (Falls et al.,
1993) and glial growth factor (Marchionni et al., 1993).
These ligands do not interact directly with EGFR and
ErbB-2 but bind with low anity to ErbB-3 and high
anity to ErbB-4 (Plowman et al., 1993b; Carraway et
al., 1994). Although none of these ligands interact
directly with ErbB-2 itself, when ErbB-2 is expressed
with other members of ErbB receptor family,
transphosphorylation of the receptor is observed
through ligand-induced heterodimerization of ErbB-2/
EGFR, ErbB-2/ErbB-3, ErbB-2/ErbB-4 (King et al.,
1988; Kokai et al., 1989; Plowman et al., 1993b;
Carraway and Cantley, 1994; Sliwkowski et al., 1994;
Beerli et al., 1995; Graus-Porta et al., 1995).
Heterodimers of ErbB-2/ErbB-3 are reported to
constitute a high anity binding site for HRG
(Sliwkowski et al., 1994). Regulation of signaling of
ErbB receptors is very complex since several EGF
family ligands have been reported and transactivation
occurs through the formation of ligand-induced
receptor dimelization and receptor cross-phosphorylation. This may be re¯ective of the diversity of
biological responses of EGF-related ligands.
Epiregulin is a recently isolated member of the EGF
family and originally found in the conditioned medium
of mouse tumorigenic ®broblast NIH3T3/clneT7 cells
established from NIH3T3 cells (Toyoda et al., 1995a).
Epiregulin inhibits the growth of several epithelial
tumor cells but stimulates the proliferation of
®broblasts, primary rat hepatocytes and human
Epiregulin receptors
T Komurasaki et al
αErbB4
IP:
unlabeled ER:
αErbB3
a
αErbB2
smooth muscle cells. Puri®ed epiregulin is approximately 5.4 kDa single chain polypeptide composed of
46-amino acid residues. Approximately 24 ± 50% of the
amino acid sequence of epiregulin is identical to that of
EGF-related growth factors. In the intact animal, the
expression of epiregulin transcript is detected during
the early stages of embryonic life (Toyoda et al.,
1995b). Furthermore, results from our laboratory have
shown that the human epiregulin peptide deduced from
cDNA sequence is highly homologous to mouse
epiregulin and is biologically active (Toyoda et al.,
1997). Although epiregulin inhibits the binding of
[125I]EGF to EGFR on A431 cells more weakly than
EGF (Toyoda et al., 1995a), the direct binding of
epiregulin to ErbB receptors has not been analysed. In
the present study, we investigated whether the signal
transduction by epiregulin is mediated by EGFR alone
or other ErbB receptors. Our results showed that
recombinant human epiregulin binds to EGFR and
ErbB-4, stimulates receptor phosphorylation of distinct
subset of EGF receptors.
αEGFR
2842
– + – + – + – +
1 2 3 4 5 6 7 8
MW(kDa)
MDA-MB-468
— 200
1 2 3 4 5 6 7 8
b
MW(kDa)
SK-BR-3
— 200
1 2 3 4 5 6 7 8
c
MW(kDa)
Results
MDA-MB-453
Binding of [125I]epiregulin to ErbB receptors
To investigate the membrane binding receptors for
epiregulin, four human breast carcinoma cell lines,
which express di€erent subsets of ErbB family
receptors,
were
chemically
cross-linked
with
[125I]recombinant human epiregulin and immunoprecipitated with speci®c antibodies against each receptor.
The MDA-MB-468 cell line expresses a high level of
EGFR (Filmus et al., 1985) and a moderate level of
ErbB-3 protein (Kraus et al., 1987; Alimandi et al.,
1995), but does not express of ErbB-2 (Lupu et al.,
1990; Hancock et al., 1991) or ErbB-4 mRNA
(Plowman et al., 1993a). The SK-BR-3 cell line
expresses EGFR, ErbB-2 (Kraus et al., 1987) and
ErbB-3 (Alimandi et al., 1995) but not ErbB-4 mRNA
(Plowman et al., 1993a). The MDA-MB-453 cell line
expresses ErbB-2 (Kraus et al., 1987), ErbB-3
(Alimandi et al., 1995) and ErbB-4 mRNA (Plowman
et al., 1990), but not EGFR mRNA or protein (Yarden
and Weiberg, 1989). T47D cell line expresses moderate
levels of the four ErbB family receptors and NDFinduced phosphorylation of ErbB-4 is detected in these
cells (Beerli et al., 1995; Graus-Porta et al., 1995; Beerli
and Hynes, 1996).
As shown in lane 1 of Figure 1, when immunoprecipitation was performed with anti-EGFR mAb, a
protein of approximately 170 ± 185 kDa and large
molecular weight protein complexes (4200 kDa)
were observed in all cell lines expressing EGFR
including MDA-MB-468, SK-BR-3, and T47D cells.
In addition, the immunoprecipitated protein was
detected by only anti-EGFR mAb in MDA-MB-468
cells. These results indicate that epiregulin is a ligand
for EGFR. On the other hand, bands of about
180 kDa were detected in proteins immunoprecipitated
with each of anti-ErbB-2 and anti-ErbB-4 antibodies
in MDA-MB-453 and T47D cells (lane 3 and 7,
Figure 1c or d), suggesting that ErbB-2 and ErbB-4
are candidate for receptors of epiregulin. However,
since the band was not detected in SK-BR-3 cell that
— 200
1 2 3 4 5 6 7 8
d
MW(kDa)
T47D
— 200
Figure 1 Chemical cross-linking of [125I]epiregulin to the cell
surface receptor of human breast carcinoma cell lines.
[125I]epiregulin (50 ng/ml) was incubated with the indicated cell
lines for 4 h at 48C in the absence (7) or presence (+) of
unlabeled epiregulin (10 mg/ml). Cross-linking was performed as
described in Materials and methods. Equal amounts of protein
(2 mg) were immunoprecipitated with 2 mg/ml of the indicated
four anti-ErbB family receptor antibodies. Immune complexes
were resolved by 7.5% SDS ± PAGE gel, and radioactive, crosslinked complexes were detected by Bioimaging analyzer
express ErbB-2 but not ErbB-4, the bands detected is
probably co-precipitated ErbB-4 monomers, due to
failure of chemical cross-linking to stably associate
with ErbB-2. To con®rm this issue, we performed
cross-linking analysis in the presence of increasing
concentration of unlabeled epiregulin or HRG-a, that
binds speci®cally to ErbB-3 and ErbB-4 (Marchionni
et al., 1993; Plowman et al., 1993b), using MDA-MB453 cells. Cross-linking of [125I]epiregulin to MDAMB-453 cells were inhibited with epiregulin and
HRG-a in a dose-dependent manner when immunoprecipitation was carried out with each of anti-ErbB2-and anti-ErbB-4 antibodies (Figure 2). HRG-a
completely blocked the binding of [125I]epiregulin to
ErbB-2 or ErbB-4 at a concentration of 100 ng/ml,
respectively. In contrast, epiregulin was required at a
concentration above 1 mg/ml to completely block the
Epiregulin receptors
T Komurasaki et al
MDA-MB-453
1 2 3 4 5
IP:
unlabeled
HRG-α
:
(ng/ml)
αErbB2
0
10
100
1000
10000
unlabeled
ER
:
(ng/ml)
αErbB2
0
10
100
1000
10000
IP:
1 2 3 4 5
MW(kDa)
MW(kDa)
200 —
unlabeled
ER
:
(ng/ml)
MW(kDa)
200 —
αErbB4
0
10
100
1000
10000
IP:
1 2 3 4 5
IP:
unlabeled
HRG-α
:
(ng/ml)
αErbB4
0
10
100
1000
10000
200 —
1 2 3 4 5
MW(kDa)
200 —
Figure 2 Competition of [125I]epiregulin binding by HRG-a in
MDA-MD-453 cells. MDA-MB-453 cells were incubated with
50 ng/ml of [125I]epiregulin for 4 h at 48C in the presence of 10 ±
10 000 ng/ml of unlabeled epiregulin or HRG-a. Cross-linking,
immunoprecipitations and analysis by electrophoresis on SDS ±
PAGE gel were performed as described in Materials and methods
formation of cross-linked complexes. [125I]EGF did not
generate any radioactive protein by immunoprecipitation with antibodies for four ErbB receptors in MDAMB-453 cells (data not shown). These results indicate
that the labeled protein observed in anti-ErbB-2
immunoprecipitates is ErbB-4, and strongly suggest
that epiregulin is a ligand for ErbB-4. The results also
demonstrate that epiregulin does not directly bind to
ErbB-2 but induces heterodimerization of ErbB-4/
ErbB-2, as well as HRG (Carraway et al., 1994;
Sliwkowski et al., 1994; Beerli et al., 1995; GrausPorta et al., 1995), and EGFR/ErbB-2/ErbB-4. No
radioactive protein could be immunoprecipitated with
anti-ErbB-3 antibodies in four breast carcinoma cell
lines (Figure 1a ± d, lane 5).
In order to con®rm a direct interaction between
epiregulin and EGFR, ErbB-2, ErbB-3 and ErbB-4, the
cross-linking analysis was performed using an antiEGFR mAb that binds to the extracellular domain of
EGFR but does not activate tyrosine kinase of EGFR
(Kawamoto et al., 1983). No protein chemically crosslinked with [125I]epiregulin could be immunoprecipitated by pretreatment with anti-EGFR mAb in SKBR-3 cells (Figure 3a). Activation of tyrosine
phosphorylation of proteins by epiregulin was also
inhibited completely in the presence of anti-EGFR
mAb in SK-BR-3 cells (Figure 3c), indicating that the
intracellular signal for epiregulin is directly mediated
by EGFR rather than ErbB-2 or ErbB-3 and forms a
heterodimerization of EGFR/ErbB-2. Anti-EGFR
mAb had no e€ect on the binding of [125I]epiregulin
to ErbB-4 in MDA-MB-453 cells, showing a speci®city
of the inhibiting activity of this mAb (Figure 3b). On
the other hand, anti-EGFR mAb readily blocked the
binding of [125I]EGF to EGFR in T47D cells. However,
a protein of approximately 170 ± 180 kDa and bands of
molecular
weight
4200 kDa
labeled
with
[125I]epiregulin were still detected in proteins immunoprecipitated by anti-EGFR-, anti-ErbB-2, and antiErbB-4 antibodies (Figure 4), suggesting that ErbB-4 is
another receptor for epiregulin. The labeled bands
observed in anti-ErbB-2 immunoprecipitates also
competed with HRG-a in similar experiments shown
in Figure 2 but in the presence of 10 mg/ml of antiEGFR mAb (data not shown). Based on these ®ndings,
we concluded that epiregulin binds directly to each of
EGFR and ErbB-4 but not to ErbB-2 and ErbB-3.
Although no protein cross-linked with [125I]epiregulin
was immunoprecipitated with anti-ErbB-3 antibody in
four breast carcinoma cell lines in this study, we were
able to detect epiregulin-induced phosphorylation of
ErbB-3 in a manner that was as e€ective as that of
EGF in SK-BR-3 cells as described below. This is
probably due to the transphosphorylation of EGFR/
ErbB-3. Alternatively, this may be due to the high
sensitivity of the functional assay (i.e. tyrosine
phosphorylation) compared with the structural assay
(anity labeling).
Epiregulin induces tyrosine phosphorylation of ErbB
receptors
In the next step, we compared the induction of tyrosine
phosphorylation of ErbB receptors by human epiregulin with that by EGF and HRG-a in four human
breast carcinoma cell lines (Figure 5). The results are
summarized in Table 1. Epiregulin at a concentration
of 20 ng/ml was slightly less e€ective than EGF at
phosphorylating EGFR in MDA-MB-468, SK-BR-3
and T47D cells (Figure 5a), in agreement with previous
observations that epiregulin has a weak anity to
EGFR. On the other hand, the ability of epiregulin to
activate ErbB-2 tyrosine phosphorylation was very
weak compared with EGF in SK-BR-3 and T47D cells
(Figure 5b). This is probably due to the ability of
epiregulin to heterodimerize and transactivate ErbB-2
through EGFR. Epiregulin and EGF did not clearly
stimulate tyrosine phosphorylation of ErbB-2 in MDAMB-453 cells, in contrast to HRG-a. In addition,
epiregulin and EGF were signi®cantly less e€ective
than HRG-a in inducing tyrosine phosphorylation of
ErbB-3 in SK-BR-3, MDA-MB-453, and T47D cells
(Figure 5c). Epiregulin-induced tyrosine phosphorylation of ErbB-3 was also con®rmed by immunoprecipitation with anti-phosphotyrosyl mAb and blotted with
anti-ErbB-3, but not vice versa. The di€erence may be
due to the low expression level of ErbB-3 in MDAMB-468 cells (Sliwkowski et al., 1994) or the
immunoprecipitation ability of anti-ErbB-3 antibodies
used in the present study. These results indicate that
epiregulin induces heterodimerization and transphosphorylation of EGFR/ErbB-3. As expected, we were
able to observe a clear activation of tyrosine
phosphorylation of ErbB-4 by epiregulin in T47D
cells (Figure 5d). Epiregulin exhibited a marked
activity and was as ecient as HRG-a, while EGF
had a weak e€ect. Although the level of induction was
2843
Epiregulin receptors
T Komurasaki et al
2844
SK-BR-3 cells
MDA-MB-453 cells
αEGFR MoAb
no treated
unlabeled EGF
normal IgG
αEGFR MoAb
4
5
6
7
8
MW(kDa)
200 —
200 —
97 —
97 —
66 —
66 —
αEGFR MoAb
normal IgG
3
normal IgG
unlabeled ER
2
unlabeled ER
no treated
1
MW(kDa)
SK-BR-3 cells
αErbB4
IP:
1
2
3
4
no treated
ER
αErbB2
no treated
αEGFR
IP:
c
1 2 3 4 5
αEGFR MoAb
b
normal IgG + ER
αEGFR MoAb + ER
normal IgG
a
6
MW(kDa)
200 —
97 —
66 —
Figure 3 Inhibition by anti-EGFR monoclonal antibody of [125I]epiregulin binding to cell surface receptor of SK-BR-3 and MDAMB-453 cells, and epiregulin-induced tyrosine phosphorylation of proteins of SK-BR-3 cells. (a, b) SK-BR-3 cells or MDA-MB-453
cells was incubated for 30 min at 378C with 10 mg/ml of anti-EGFR mAb or anti-ErbB-4 antibodies, respectively. [125I]epiregulin
(50 ng/ml) was then added in the absence or presence of 10 mg/ml of unlabeled epiregulin and incubated for 4 h at 48C. Chemical
cross-linking and immunoprecipitation with indicated antibodies (2 mg/ml) were performed as described in Materials and methods.
Radiolabeled signals were detected by Bioimaging analyzer after resolving the samples by 7.5% SDS ± PAGE gel. (c) SK-BR-3 cells
were labeled with [32P]orthophosphate, and medium containing 20 ng/ml of epiregulin was added in the absence or presence of
10 mg/ml of anti-EGFR mAb or isotype control mAb. After 10 min at 378C, tyrosine phosphorylated proteins were isolated as
described in Materials and methods and resolved in 7.5% SDS ± PAGE gel. Radiolabeled proteins were detected with Bioimaging
analyzer
weaker than in T47D cell, similar results were found in
MDA-MB-453 cells. Therefore, epiregulin probably
induces transactivation of phosphotyrosine of ErbB-2
and ErbB-3 through ErbB-4 in MDA-MB-453, and of
the four receptors through EGFR and ErbB-4 in T47D
cells.
Discussion
We have previously demonstrated that epiregulin
inhibits the binding of [125I]EGF to EGFR on an
epidormoid carcinoma A431 cell, suggesting that
epiregulin is a ligand for EGFR (Toyoda et al.,
1995a). In this study, we analysed the ability of
epiregulin to bind to and stimulate tyrosine phosphorylation of ErbB family members by using four human
breast carcinoma cell lines. Our results showed that
[125I]epiregulin bound directly to ErbB-4 as well as
EGFR, but not to ErbB-2 and ErbB-3. Furthermore,
epiregulin activated tyrosine phosphorylation of all
four ErbB receptors with an eciency equivalent to
that of HRG-a in stimulating ErbB-4 in T47D cells.
Thus, our data indicate that EGFR and ErbB-4 are
receptors for epiregulin. TGF-a, AR and HB-EGF
stimulate tyrosine phosphorylation of ErbB-2, ErbB-3,
and ErbB-4 through EGFR (Riese et al., 1996b). On
the other hand, HRG/NDF activates EGFR and
ErbB-2 through ErbB-3 and ErbB-4 (reviewed in
Carraway and Burden, 1995; Pinkas-Kramarski et al.,
1996). Epiregulin exhibited a pattern of receptor
binding and activation that was di€erent from these
peptides. Recent reports described the ability of BTC
to induce tyrosine phosphorylation of all four ErbB
receptors and a possible direct binding to ErbB-4 as
well as EGFR (Beerli and Hynes, 1996; Riese et al.,
1996a). Although similar receptor binding pro®les were
observed, epiregulin is probably di€erent from BTC
with regard to the magnitude of activation of tyrosine
phosphorylation of the four ErbB family receptors.
The e€ects of BTC on tyrosine phosphorylation of
EGFR and ErbB-2 in T47D cells are similar to those
of EGF and NDF, respectively (Beerli and Hynes,
1996). On the other hand, epiregulin was signi®cantly
less active in phosphorylation of EGFR and ErbB-2
than EGF and HRG-a in T47D cells, respectively.
Furthermore, although BTC had a much stronger
e€ect on ErbB-3 phosphorylation than EGF in T47D
cells (Beerli and Hynes, 1996), epiregulin-induced
ErbB-3 tyrosine phosphorylation was very weak.
These results point to the existence of di€erences
between epiregulin and BTC in their ability of ErbB
receptor heterodimer formation following transactivation of ErbB family receptors.
Based on the present results, we can summarize the
potency of epiregulin-induced tyrosine phosphorylation
Epiregulin receptors
T Komurasaki et al
the present study. Epiregulin is probably unable to
induce the complete formation of ErbB-2/ErbB-3
complex. These di€erences between epiregulin and
HRG-a are partially responsible for the activation of
tyrosine phosphorylation of ErbB-2. In summary, the
signi®cantly low potency of epiregulin in transactivating ErbB-2, relative to EGF, HRG-a, and probably
BTC, is probably due to the very low anity to EGFR
and the absence of a direct binding to ErbB-3.
of ErbB receptors as follows: ErbB-45EGFR4ErbB24ErbB-3. Epiregulin-induced stimulation of tyrosine
phosphorylation of ErbB-2 was weaker than that of
EGF in SK-BR-3 and T47D cells, and HRG-a in
T47D and MDA-MD-453 cells. NDF-induced heterodimerization of the receptors in cells that ectopically
expressed ErbB receptors showed that the interaction
between ErbB-3 and ErbB-2 is more extensive than the
cross-talk between ErbB-4 and ErbB-2 (Tzahar et al.,
1996). On the other hand, epiregulin bound directly to
ErbB-4 but not ErbB-3 in four cell lines examined in
MDAMDAMB-468 SK-BR-3 MB-453
[125I]-ER [125I]-EGF
[125I]-ER [125I]-EGF
no treated
unlabeled ER
normal IgG
αEGFR MoAb
no treated
unlabeled EGF
normal IgG
αEGFR MoAb
no treated
unlabeled ER
normal Ig
αEGFR MoAb
no treated
unlabeled EGF
normal IgG
αEGFR MoAb
1 2 3 45 6 78
1 23 4567 8
MW(kDa)
MW(kDa)
200 —
200 —
1 2 3 4 5 6 7 8 9 101112 13141516
MW(kDa)
200 —
IP : αEGFR
blot Ab : αPY
1 2 3 4 5 6 7 8 9 10 1112
MW(kDa)
97 —
97 —
66 —
66 —
c
200 —
IP : αErbB2
IP : αEGFR
1 2 3 45 67 8
MW(kDa)
200 —
200 —
97 —
97 —
66 —
66 —
IP : αErbB3
MW(kDa)
200 —
Figure 4 Chemical cross-linking of [125I]epiregulin or [125I]EGF
to ErbB family receptors in T47D cells in the presence of antiEGFR antibody. After pretreatment with 10 mg/ml of anti-EGFR
mAb or isotype control mAb for 30 min at 378C, T47D cells were
incubated with [125I]epiregulin or [125I]EGF (each at 50 ng/ml) in
the absence or presence of unlabeled 10 mg/ml of epiregulin or
EGF, respectively. Immunoprecipitation with indicated antibodies
(2 mg/ml) was performed as described in Materials and methods.
(a) anti-EGFR mAb; (b) anti-ErbB-2 antibody; (c) anti-ErbB-3
antibody; (d) anti-ErbB-4 antibody. Radiolabeled signals were
detected by Bioimaging analyzer after resolving the samples on
7.5% SDS ± PAGE gel
+++
7
7
7
++
7
7
7
IP : αErbB3
blot Ab : αPY
1 2 3 4 5 6 7 8 9 10 1112
IP : αErbB4
blot Ab : αPY
Di€erential activation of ErbB receptors by epiregulin, EGF and HRG-a
MDA-MB-468
EGF
ER
HRG
EGFR
ErbB-2
ErbB-3
ErbB-4
1 2 3 4 5 6 7 8 9 10 1112
Figure 5 Epiregulin-induced tyrosine phosphorylation of ErbB
family receptors. Four cell lines were incubated for 10 min at
378C in the absence or presence of ligands (epiregulin, EGF, or
HRG-a, each at 20 ng/ml). The cells were then washed, lysed and
equal amount of cell lysates (2 mg) were subjected to
immunoprecipitation with the indicated antibodies (2 mg/ml).
The immunoprecipitates were resolved by SDS ± PAGE and
phosphorylated tyrosine was detected with anti-phosphotyrosine
mAb (aPY). A very faint band in anti-ErbB-2 immunoprecipitates
of MDA-MB-468 cells was occasionally observed when immunoprecipitation was carried out with anti-ErbB-2 antibody and
blotted with anti-phosphotyrosyl mAb, however, the band was
not observed vice versa, suggesting that it was probably a nonspeci®c band
IP : αErbB4
Table 1
IP : αErbB2
blot Ab : αPY
MW(kDa)
200 —
1 23 45 6 7 8
d
MW(kDa)
T47D
no treated
ER
EGF
HRGα
no treated
ER
EGF
HRGα
no treated
ER
EGF
HRGα
no treated
ER
EGF
HRGα
b
a
7
7
7
7
EGF
SK-BR-3
ER
HRG
EGF
+++
++
+
7
+
+
+
7
7
7
++
7
7
7
7
7
MDA-MB-453
ER
HRG
7
7
7
+
7
++
++
+
EGF
T47D
ER
HRG
++
++
7
+
+
+
7
++
7
++
++
++
Stimulation of receptor tyrosine phosphorylation is summarized in Figure 5. ER, epiregulin. `+' indicates increased receptor tyrosine
phosphorylation over basal levels, `7' indicates no increase in receptor tyrosine phosphorylation
2845
Epiregulin receptors
T Komurasaki et al
2846
Tzahar et al. (1996) reported that directional and
hierarchical networks of interreceptor interactions of
ErbB family receptors determine signal transduction
by NDF/HRG and EGF. Although EGF-related
ligands can be divided into subgroups based on the
receptor binding, their ability for receptor phosphorylation and biological e€ects in breast carcinoma
and BaF3 cells that ectopically expressed one or two
ErbB receptors are ligand-speci®c. Epiregulin binds
to EGFR and ErbB-4 and seems to transduce the
main signals by ErbB-4 and EGFR but not by
ErbB-2 and ErbB-3. Recent reports and our results
suggest that EGF-related ligands transduce signals
for di€erent biological responses by the hierarchical
mechanism.
Materials and methods
Materials
Recombinant human EGF was obtained from Upstate
Biotechnology, Inc. (Lake Placid, NY). Recombinant
heregulin-a (HRG-a) was purchased from R&D (Minneapolis, MN). Recombinant human epiregulin of 46 amino
acids was produced by Bacillus brevis (Nakazawa et al., in
preparation). The biological activity of puri®ed epiregulin
was equal to that of chemically-synthesized mouse
epiregulin based on stimulation of DNA synthesis in
Balb3T3 A31 cells. Monoclonal antibody directed against
EGFR (Ab-3), and anti-phosphotyrosine coupled to
agarose were obtained from Oncogene Science (Manhasset, NY). Anti-ErbB-2 (C-18), anti-ErbB-3 (C-17) and antiErbB-4 (C-18) antibodies, and anti-phosphotyrosine
(PY20) mAb were purchased from Santa Cruz Biotechnology (Santa Cruz, CA). The IgG 1 isotype control mAb
MOPC.21 was purchased from Sigma (St Louis, MO).
Cell cultures and cell growth assays
Human breast carcinoma cell lines, MDA-MB-468, MDAMB-453, SK-BR-3, and T47D cells, were obtained from the
American Type Culture Collection (Rockville, MD) and
maintained in 1 : 1 mixture of DMEM and Ham F12 (DF
medium, GIBCO BRL) containing 10% fetal bovine serum
(FBS, GIBCO BRL).
Labeling of recombinant human epiregulin and EGF
Human epiregulin and EGF were radioiodinated using
the lactoperoxidase (LPO) method (Marchalonis, 1969)
with a slight modi®cation. Brie¯y, epiregulin or EGF
(2 nmol in 5 ml of water) and 25 ml of 0.5 M sodium
phosphate, pH 7.5, were added to 1 mg/ml of LPO
(Boehringer Mannheim, Germany). In the next step, we
added 4 ml of 0.03% hydrogen peroxide and 2 mCi of
Na 125I (Carrier-free, 3.7 Gbq/ml, Amersham). The reaction mixture was incubated for 10 min at room
temperature and applied onto a Sephadex G-15 column
(5650 mm, Pharmacia, LKB) pre-equilibrated with
50 m M sodium phosphate bu€er (pH 7.5), containing
0.05% bovine serum albumin (BSA). After 5 min,
radiolabeled epiregulin or EGF was eluted with the
same bu€er described above and stored with the addition
of BSA at a ®nal concentration of 1% at 7708C. The
speci®c activities of [125I]epiregulin and [125I]EGF were
3.36105 c.p.m./ng and 3.06105 c.p.m./ng, respectively.
The respective recovery of [125I]epiregulin and [125I]EGF
was about 50%. This was determined by iodination in the
absence of Na125I and LPO. The amount of epiregulin
and EGF recovered at the end of the experiment were
determined by HPLC using a cold epiregulin or EGF as a
standard. The biological activity of [ 125I]epiregulin or
[125I]EGF was examined by morphological changes in
HeLa cells and the growth stimulatory activity on
Balb3T3 A31 cells as previously described (Toyoda et
al., 1995a).
Chemical cross-linking and immunoprecipitation
Cells were seeded at 56105/dish in 90 mm tissue culture
dish and cultured for 48 h at 378C. The medium was
replaced with serum-free DF medium containing 0.1%
BSA and cultured for another 24 h. In the next step, the
medium was discarded and the cells were washed twice
with an ice cold binding bu€er (DMEM containing 20 mM
HEPES, pH 7.5 and 0.1% BSA), and incubated with 2 ml
of a binding bu€er containing 50 ng/ml of [125I]epiregulin
or 50 ng/ml of [125I]EGF for 4 h at 48C in the absence or
presence of indicated unlabeled epiregulin, EGF, or HRGa, respectively. Disuccinimidyl suberate (DSS, Pierce) was
then added to a ®nal concentration of 0.15 mM and the
cells were incubated for another 15 min at room
temperature. The reaction was terminated by adding
200 ml of a quenching bu€er containing 10 mM Tris-HCl,
pH 7.5, 200 mM glycine and 2 mM EDTA. The cells were
then washed twice with ice cold PBS(7) and solubilized in
2 ml of cold lysis bu€er (10 mM Tris-HCl, pH 7.6, 1%
Triton X-100, 5 mM EDTA, 150 mM NaCl, 30 mM sodium
pyrophosphate, 50 mM sodium ¯uoride, 100 mM sodium
orthovanadate, 1 mM PMSF, 1 mM pepstatin, 1 mM
leupeptin, 1 U/ml apro®nin) for 30 min at 48C. The
lysates were clari®ed by centrifugation for 30 min at
15 000 r.p.m. For immunoprecipitation experiments, equal
amounts of cell lysates were incubated overnight with
speci®c antibodies at 48C. Protein G-Sepharose (50% v.v,
Pharmacia) was then added with continuous stirring for
another 2 h at 48C. Complexes were collected by
centrifugation for 5 min at 3000 r.p.m., washed ®ve times
with lysis bu€er, boiled for 5 min in an equal volume of
twofold SDS ± PAGE sampler bu€er, clari®ed by centrifugation and analysed by SDS ± PAGE in 7.5% SDS ± PAGE
gel (Daiichi Pure Chemical Co, Tokyo) under reduced
conditions. The gels were ®xed for 30 min in 10 ± 30%
acetic acid-methanol solution, exposed to Bioimaging plate
(Fuji®lm, Inc, Tokyo) for 4 h and the resultant cells were
analyzed with a Bioimaging analyzer (Model BAS2000,
Fuji®lm, Inc, Tokyo).
Metabolic labeling and immunoprecipitation of tyrosinephosphorylated protein
Experiments were performed as described by Johnson et al.
(1993). Cells prepared as described above were washed
twice with phosphate-free DMEM (GIBCO BRL) and
labeled for 4 h with 2 ml of phosphate-free DMEM
containing 2 mCi of [ 32P]orthophosphate (Carrier-free,
370 MBq/ml, Amersham) at 378C. At the end of
incubation, 5 ml of epiregulin was added to the cells to
yield a ®nal concentration of 20 ng/ml and further
incubated at 378C. After exactly 10 min, the culture plate
was placed on ice, the culture medium was removed and
washed twice with ice cold phosphate-free DMEM. The
cells were solubilized in lysis bu€er and immunoprecipitated with 50 ml of anti-phosphotyrosyl mAb-agarose (50%
v/v) that was pre-equilibrated with a lysis bu€er. The
resultant pellet was washed ®ve times with a lysis bu€er,
and tyrosine-phosphorylated proteins were eluted with
500 ml of lysis bu€er containing 10 m M phenylphosphate
(Sigma). Finally, 20 ml of the eluate was resolved by SDS ±
PAGE and analysed as described.
Epiregulin receptors
T Komurasaki et al
Western blot analysis
Following stimulation, the culture medium was removed
and cells were washed twice with ice-cold PBS(7),
solubilized in lysis bu€er and immunoprecipitated with
speci®c antibodies as described above. Immunoprecipitates
were subjected to SDS ± PAGE in 7.5% SDS ± PAGE gel,
electrotransferred onto ProBlott membrane (Applied
Biosystems, Inc, CA). After blocking with 5% BSA in
PBS(7) containing 0.05% Tween 20, the ®lters were
incubated with speci®c antibodies and proteins were
detected by peroxidase-coupled secondary antibodies
using the ECL detection system (Amersham).
Abbreviations
EGFR, epidermal growth factor receptor; ER, epiregulin;
EGF, epidermal growth factor; TGF-a, transforming
growth factor-a; BTC, betacellulin; HRG, heregulin;
NDF, neu di€erentiation factor; mAb, monoclonal antibody; PAGE, polyacrylamide gel electrophoresis.
Acknowledgements
We thank Dr K Hanada for his support of our research.
We also thank our colleagues in the Molecular Biology
Laboratory of Taisho Pharmaceutical Co, Ltd. for helpful
discussions.
References
Alimandi M, Romano A, Curia MC, Muraro R, Fedi P,
Aaronson SA, Fiore PPD and Kraus MH. (1995).
Oncogene, 10, 1813 ± 1821.
Beerli RR, Graus-Porta D, Woods-Cook K, Chen X, Yarden
Y and Hynes NE. (1995). Mol. Cell. Biol., 15, 6496 ± 6505.
Beerli RR and Hynes NE. (1996). J. Biol. Chem., 271, 6071 ±
6076.
Carraway III KL, Sliwkowski MX, Akita R, Platko JV, Guy
PM, Nuijens A, Diamonti AJ, Van den RL, Cantley LC
and Cerione RA. (1994). J. Biol. Chem., 269, 14303 ±
14306.
Carraway III KL and Cantley LC. (1994). Cell, 78, 5 ± 8.
Carraway III KL and Burden SJ. (1995). Curr. Opin.
Neurobiol., 5, 606 ± 612.
Cook PW, Mattox PA, Keeble WW, Pittelkow MR,
Plowman GD, Shoyab M, Adelman JP and Shipley GD.
(1991). Mol. Cell. Biol., 11, 2547 ± 2557.
Falls DL, Rosen KM, Corfas G, Lane WS and Fischbacb
GD. (1993). Cell, 72, 801 ± 805.
Filmus J, Pollak MN, Cailleau R and Buick RN. (1985).
Biochem. Biophys. Res. Commun., 128, 898 ± 905.
Graus-Porta D, Beerli RR and Hynes NE. (1995). Mol. Cell.
Biol., 15, 1182 ± 1191.
Groenen LC, Nice EC and Burgess AW. (1994). Growth
Factors, 11, 235 ± 257.
Hancock MC, Langton BC, Chan T, Toy P, Monahan JJ,
Mischak RP and Shawver LK. (1991). Cancer Res., 51,
4575 ± 4580.
Higashiyama S, Abraham JA, Miller J, Fiddes JC and
Klagsbrun M. (1991). Science, 251, 936 ± 939.
Homes WE, Sliwkowski MX, Akita RW, Henzel WJ, Lee J,
Park JW, Yansura D, Abadi N, Raab H, Lewis GD,
Shepard HM, Kuang W-JL, Wood WI, Goeddel DV and
Vandlen RL. (1992). Science, 256, 1205 ± 1210.
Hynes NE and Stern DF. (1994). Biochem. Biophis. Acta,
1198, 165 ± 184.
Johnson GR, Kannan B, Shoyab M and Stromberg K.
(1993). J. Biol. Chem., 268, 2924 ± 2931.
Kawamoto T, Sato JD, Le A, Poliko€ J, Sato GH and
Mendelsohn J. (1983). Proc. Natl. Acad. Sci. USA, 80,
1337 ± 1341.
Kimura H, Fischer WH and Schubert D. (1990). Nature, 348,
257 ± 260.
King CR, Borrello I, Bellot F, Comoglio P and Schlessinger
J. (1988). EMBO J., 7, 1647 ± 1651.
Kokai Y, Meyers JN, Wada T, Brown IV, Levea CM, Davis
JG, Dobashi K and Green MI. (1989). Cell, 58, 287 ± 292.
Kraus MH, Popescu NC, Amsbaugh SC and King CR.
(1987). EMBO J., 6, 605 ± 610.
Kraus MH, Issing W, Miki T, Popescu NC and Aaronson
SA. (1989). Proc. Natl. Acad. Sci. USA, 86, 9193 ± 9197.
Lai C and Lemke G. (1991). Neuron, 6, 691 ± 704.
Lee DC, Fenton SE, Berkowitz EA and Hissong MA. (1995).
Pharm. Rev., 47, 51 ± 85.
Lemoine NR, Barnes DM, Hollywood DP, Hughes CM,
Smith P, Dublin E, Prigent SA, Gullick WJ and Hurst HC.
(1992). Br. J. Cancer, 66, 1116 ± 1121.
Lupu R, Colomer R, Zugmaier G, Sarup J, Shepard M,
Slamon D and Lippman M. (1990). Science, 249, 1552 ±
1555.
Marchalonis JJ. (1969). Biochem. J., 113, 299 ± 305.
Marchionni MA, Goodearl ADJ, Chen MS, BerminghamMcDonogh O, Kirk C, Hendricks M, Danehy F, Misumi
D, Sudhalter J, Kobayashi K, Wroblewski D, Lynch C,
Baldassare M, Ian H, Davis JB, Hsuan JJ, Totty NF, Otsu
M, McBurney RN, Water®eld MD, Stroobant P and
Gwynne D. (1993). Nature, 362, 312 ± 318.
Marquardt H, Hunkapiller MW, Hood LE and Todaro GJ.
(1984). Science, 223, 1079 ± 1082.
Pinkas-Kramarski R, Shelly M, Glathe S, Ratzkin BJ and
Yarden Y. (1996). J. Biol. Chem., 271, 19029 ± 19032.
Plowman GD, Whitney GS, Neubauer MG, Green JM,
McDonald VL, Todaro GJ and Shoyab M. (1990). Proc.
Natl. Acad. Sci. USA, 87, 4905 ± 4909.
Plowman GD, Culouscou J-M, Whitney GS, Green JM,
Carlton GW, Foy L, Neubauer MG and Shoyab M.
(1993a). Proc. Natl. Acad. Sci. USA, 90, 1746 ± 1750.
Plowman GD, Green JM, Culouscou J-M, Carlton GW,
Rothwell VM and Buckley S. (1993b). Nature, 366, 473 ±
475.
Prigent SA, Lemoine NR, Hughes CM, Plowman GD,
Selden C and Gullick WJ. (1992). Oncogene, 7, 1273 ±
1278.
Riese II DJ, Bermingham Y, van Raaij TM, Buckley S,
Plowman GD and Stern DF. (1996a). Oncogene, 12, 345 ±
353.
Riese II DJ, Kim ED, Elenius K, Buckley S, Klagsbru M,
Plowman GD and Stern DF. (1996b). J. Biol. Chem., 271,
20047 ± 20052.
Savage CR Jr, Inagami T and Cohen S. (1972). J. Biol.
Chem., 247, 7612 ± 7621.
Shing Y, Christofori G, Hanahan D, Ono Y, Sasada R,
Igarashi K and Folkman J. (1993). Science, 259, 1604 ±
1607.
Shoyab M, Plowman GD, McDonald VL, Bradley JG and
Todaro GJ. (1989). Science, 243, 1074 ± 1076.
Slamon DJ, Clark GM, Wong SG, Levin WJ, Ullrich A and
McGuire WL. (1987). Science, 235, 177 ± 182.
Sliwkowski MX, Shaefer G, Akita RW, Lofgren JA,
Fitzpatrick VD, Nuijeus A, Fendly BM, Cerione RA,
Vandlen RL and Carraway III KL. (1994). J. Biol. Chem.,
269, 14661 ± 14665.
Toyoda H, Komurasaki T, Uchida D and Hanada K.
(1995a). J. Biol. Chem., 270, 7495 ± 7500.
Toyoda H, Komurasaki T, Ikeda Y, Yoshimoto M and
Morimoto S. (1995b). FEBS Lett., 377, 403 ± 407.
Toyoda H, Komurasaki T, Uchida D and Morimoto S.
(1997). Biochem. J., 326, 69 ± 75.
2847
Epiregulin receptors
T Komurasaki et al
2848
Tzahar E, Waterman H, Chen X, Levkowitz G, Karunagaran D, Lavi S, Ratzkin BJ and Yarden Y. (1996).
Moll. Cell. Biol., 16, 5276 ± 5287.
Ullrich A, Coussens L, Hay¯ick JS, Dull TJ, Gray A, Tam
AW, Lee J, Yarden Y, Libermann TA, Schlessinger J,
Downward J, Mayes ELV, Whittle N, Water®eld MD and
Seeburg PH. (1984). Nature, 309, 418 ± 425.
Wen D, Peles E, Capples R, Suggs SV, Bacus SS, Luo Y,
Trail G, Hu S, Silbiger SM, Levy RB, Koski RA, Lu HS
and Yarden Y. (1992). Cell, 69, 559 ± 572.
Yamamoto T, Ikawa S, Akiyama T, Semba K, Nomura N,
Miyajima N, Saito T and Toyoshima K. (1986). Nature,
319, 230 ± 234.
Yarden Y and Weiberg RA. (1989). Proc. Natl. Acad. Sci.
USA, 86, 3179 ± 3188.