2741
Journal of Cell Science 111, 2741-2751 (1998)
Printed in Great Britain © The Company of Biologists Limited 1998
JCS9813
Cell-type specific and estrogen dependent expression of the receptor
tyrosine kinase EphB4 and its ligand ephrin-B2 during mammary gland
morphogenesis
Zariana Nikolova, Valentin Djonov, Gisela Zuercher, Anne-Catherine Andres and Andrew Ziemiecki*
Department of Clinical Research, University of Berne, Tiefenaustrasse 120, CH-3004 Berne, Switzerland
*Author for correspondence (e-mail: [email protected])
Accepted 8 July; published on WWW 27 August 1998
SUMMARY
Morphogenesis of the mammary gland occurs mainly
during adult life and is dependent on a complex interplay
of hormonal, cell-cell and cell-matrix interactions. The
molecular mechanisms involved in pattern formation of the
mammary epithelium in adult life are poorly understood.
Recently, several members of the Eph family of receptor
tyrosine kinases and their ligands have been shown to
participate in pattern formation during embryogenesis and
conceivably may fulfill similar functions during adult
morphogenesis. We have investigated the expression of a
member of this family, EphB4, and its cognate ligand,
ephrin-B2, during normal and malignant mouse mammary
morphogenesis. A spatially, temporarily and hormonally
coordinated expression of both the receptor and ligand was
observed. The receptor was predominantly localized in the
myoepithelial cells surrounding the ducts and alveoli
whereas ligand expression was limited to the luminal
epithelial cells. Expression of both was induced at the onset
of gland morphogenesis at puberty and was differentially
regulated during the estrus cycle. Ovariectomy of prepubertal or adult females abolished the expression of both
receptor and ligand and administration of estrogen alone
was sufficient to restore their normal expression.
Disruption of the balanced expression was observed during
experimental mouse mammary carcinogenesis. Ligand
expression was lost at the onset of tumorigenesis and
receptor expression shifted from myoepithelial to epithelial
cells with progressive malignancy. These results implicate
both the EphB4 receptor and its ligand ephrin-B2 in the
hormone dependent morphogenesis of the mammary
gland. Furthermore, their deregulated expression may
contribute to mammary carcinogenesis.
INTRODUCTION
TGF-β family, hepatocyte growth factor and epimorphin,
which modulate epithelial morphogenesis (Imagawa et al.,
1994; Hirai et al., 1998).
At birth, a modestly ramifying ductal system extends under
the skin in a small mass of adipose tissue and is connected to
the outside via the nipple (Kratochwil, 1987). With the onset
of ovarian function during puberty, the epithelial anlagen
elongate as a result of massive, directional growth of the endbuds. These structures invade the surrounding stroma and, by
twisting and branching, give rise to a ductal network (Williams
and Daniel, 1983; Howlett and Bissell, 1993). This
developmental process is induced by the increased secretion of
estrogens and depends not only on the presence of growth
hormone, prolactin and epidermal growth factor but also on
intact epithelial:stromal interactions (Howlett and Bissell,
1993; Kleinberg, 1997). Progesterone responsiveness is
acquired later at the end of puberty and stimulates lobuloalveolar development (Haslam, 1989).
In the mature female, the mammary fat pad is filled with a
The mammary gland represents a unique model for the study
of inductive processes occurring in the adult organism.
Functional development is strictly dependent on cell-cell and
cell-matrix interactions as well as local, paracrine and
endocrine interactive stimuli (Lin and Bissell, 1993; Rosekelly
et al., 1994). Furthermore, the tissue composition and state of
differentiation change drastically according to the functional
requirement.
The mammary gland is composed of two main components,
the ectodermal parenchyma consisting of secretory epithelium
and contractile myoepithelium and the mesodermal adipose
stroma within which the parenchyma develops and functions.
Several studies have shown that the stroma plays an active,
inductive role in mammary gland development and function
(Kratochwil, 1987; Sakakura, 1991; Lin and Bissell, 1993).
This inductive role is evidenced by the presence in the stroma
of paracrine factors, such as members of the wnt family, the
Key words: Estrus cycle, Mammary carcinogenesis, Myoepithelium,
Luminal epithelial cell, Immunohistochemistry
2742 Z. Nikolova and others
highly branched ductal network consisting of myoepithelial
and luminal epithelial cells limited by a basement membrane
and enveloped by a thin sheet of fibrous connective tissue. In
the absence of pregnancy, the mammary gland parenchyma
undergoes hormonally dictated changes during the menstrual
cycle. We have recently demonstrated that during the estrus
cycle the mouse mammary epithelium undergoes moderate
proliferation, differentiation and cell death in a tightly
controlled spatial and temporal manner (Andres et al., 1995).
This represents a limited version of the parenchymal life cycle
accomplished by pregnancy, lactation and involution, the so
called reproductive cycle. Any escape from this tightly
regulated, balanced equilibrium can result in neoplastic
transformation, a development to which the mammary gland
epithelium is particularly prone (Marshall, 1993).
Although the mediators of the complex interplay involved in
mammary gland morphogenesis are not fully characterized,
protein tyrosine kinases (PTKs) either as receptors or
intracellular signal transducers have been implicated (Fox and
Harris, 1997; Hynes et al., 1997). The Eph family of receptor
PTKs, with 14 characterized members to date (Pasquale, 1997;
Orioli and Klein, 1997), represents the largest family of PTKs.
The extracellular moiety of these receptors possess an
immunoglobulin-like loop, a cysteine-rich domain and two
fibronectin type III repeats (Brambilla and Klein, 1995). The
kinase domain is located intracellularly together with a
conserved sterile alpha motif (SAM) thought to represent a
protein binding module involved in developmental regulation
(Schultz et al., 1997). The expression of most members of the
Eph family is ubiquitous during embryonic development and
in adults is limited to cells of neural origin. A minority of
family members are expressed during adult life also in other
organs (Hirai et al., 1987; Lindberg and Hunter, 1990; Andres
et al., 1994). The putative protein ligands of the Eph family
receptor PTKs also constitute a large family, being themselves
transmembrane proteins or membrane-anchored by a glycosylphosphatidylinositol tail (Pandey et al., 1995). Physiologically
relevant interaction between the receptor and the ligand
probably requires cell-cell contact since ligand has been
reported to be most effective in receptor activation when cellsurface bound (Davis et al., 1994). Recently, the cytoplasmic
domain of two transmembrane ligands has been shown to
undergo phosphorylation on conserved intracellular tyrosine
residues following receptor interaction (Brückner et al., 1997;
Holland et al., 1996). This observation suggests that this
receptor/ligand family can provoke bi-directional signaling and
enables mutual cell:cell communication. To date, the Eph
family of receptors and their ligands have been implicated in
the control of neuronal guidance and cell movement during
embryonic development leading to the determination of the
body plan (Xu et al., 1996; Tessier-Lavigne and Goodman,
1996; Gale et al., 1996; Monschau et al., 1997). The role(s) of
Eph family receptors and their ligands in adult organs has not
been investigated, however, it is possible that they may also be
involved in inductive processes such as the morphogenic
changes observed in the mammary gland.
In a previous study we have described the isolation of a
novel member of the Eph family, EphB4, and have observed
differential expression of EphB4 at the RNA level during
normal mammary gland development and deregulated
expression in invasive mouse mammary tumours (Andres et al.,
1994). Recently, the cognate ligand of EphB4, ephrin-B2, has
been identified (Bennett et al., 1995). Northern blot analyses
have shown that ephrin-B2 is also expressed in the mammary
gland (unpublished observation). In this report we describe the
immunohistochemical localization of the EphB4 receptor and
its cognate ligand ephrin-B2 at the different stages of normal
postnatal mammary gland development and carcinogenesis.
MATERIALS AND METHODS
Animals and tissues
Whole mammary glands were taken from female Swiss Moro mice at
various stages of development: 3 weeks (pre-puberty); 3.5, 4, 4.5, 5,
5.5, 6 weeks (puberty); from mature adults in the proestrus, estrus,
metestrus and anestrus phases of the estrus cycle. The phases of the
estrus cycle were determined by histological examination of vaginal
smears. The transgenic Wap-myc and Wap-ras mice have been
described in detail previously (Andres et al., 1991; Schoenenberger et
al., 1988).
Female Swiss Moro mice were anaesthetized with NembutalR
sodium solution, 1 µg/g body weight, (Abbott Laboratories, North
Chicago, USA), ovariectomized or sham operated at 3 weeks of age
(pre-puberty) and sacrificed at 6.5 weeks of age. Alternatively, mature
females at 12 weeks of age were ovariectomized or sham operated
and sacrificed 7, 21 or 30 days thereafter. Estrogen stimulation was
accomplished by daily intraperitoneal injection for 10 days of 50 ng
per g body weight of water soluble Estradiol (Sigma, St Louis, MO)
dissolved in PBS. Estrogen treatment of pre-pubertally
ovariectomized females began at 5 week of age and of ovariectomized
adults 30 days after surgery. Mammary glands were removed, snap
frozen in liquid N2 and stored at −70°C.
Northern analysis
Total RNA was prepared using the guanidinium thiocyanate extraction
method according to Strange et al. (1992) and was enriched for
poly(A)+ RNA as described (Maniatis et al., 1982). Aliquots of either
5 µg poly(A) enriched or total RNA were denatured with glyoxal,
separated by electrophoresis on a 1% agarose gel and transferred to
cellulose nitrate membranes (Schleicher and Schuell, Dassel,
Germany). Filters were pre-hybridized for 3-4 hours in 50%
formamide, 4% SSC, 5× Denhardt’s, 0.2% SDS, 0.1% tetra-Nadiphosphate-10-hydrate and 100 µg/ml denatured salmon sperm DNA
and hybridized in the same solution containing randomly primed 32P
labeled EphB4, ephrin-B2 or ribosomal protein L7a probe.
Hybridization was performed at 42°C for 18 hours and the filters were
washed to a final stringency of 0.1% SSC, 0.5% SDS at 65°C 3× 2030 minutes and exposed to X-ray film.
Antibody production
For EphB4-specific antiserum production, an AluI fragment spanning
amino acids (aa) 825 to 991 was blunt ended and cloned into the SmaI
site of pGEX3 bacterial expression plasmid. This fragment encodes
the 50 C-terminal aa of the kinase domain and the C-terminal tail of
the EphB4 receptor. For ephrin-B2 specific antibody production, a
ScaI/SmaI fragment was blunt ended and cloned into the SmaI site of
pGEX1. This fragment encodes the entire ligand protein without the
signal peptide. Polyclonal rabbit antisera were raised against gelpurified GST-EphB4 or GST-ephrin-B2 bacterial fusion proteins.
Rabbits were immunized by repeated intradermal injections of the
isolated proteins emulsified with Freund’s adjuvant. Antibodies were
affinity purified by absorption to antigen immobilized on Reacti-Gel
(HW-65F) Support (Pierce, Rockford, IL). Specific antibody was
eluted with 100 mM glycine, pH 2.5, and dialyzed against PBS
(Harlow and Lane, 1988). Affinity purified antibody was tested for
reactivity by western blotting. The specificity was verified (a) by
EphB4 and ephrin-B2 in mammary morphogenesis 2743
antibody absorption with either the corresponding or unrelated fusion
proteins, and (b) by immunoprecipitation followed by western blot
analysis. Affinity purified antibody was FITC coupled using the
FluoroTagTM FITC Conjugation Kit (Sigma, St Louis, MO) following
the manufacturer’s instructions.
Immunocytochemistry
Mammary gland cryosections (5 µm) were air dried for 20-30
minutes at room temperature (RT), fixed in acetone for 20 minutes
at RT and air dried for 20 minutes at RT. Sections were re-hydrated
in TBS for 10 minutes at RT and the endogenous peroxidase activity
was blocked by incubation in 0.3% H2O2 in TBS (25 mM Tris-HCl,
pH 7.5, 140 mM NaCl) for 15 minutes at RT. After 3× washing for
5 minutes at RT in TBS, sections were incubated in 5% BSA
solution in water for 60 minutes at RT. Incubation with affinity
purified antibody against EphB4 (1:50 in TBS) or ephrin-B2 (1:5 in
TBS) containing 10% Physiogel was performed in a moist chamber
at 4°C for 4 days. After 3× washing in TBS for 5 minutes at RT, the
sections were incubated with horseradish peroxidase labeled swine
anti-rabbit immunoglobulins, 1:100 diluted in TBS (DAKO A/S,
Glostrup, Denmark), for 1 hour at RT. Peroxidase activity was
visualized by incubation in detection solution (3-amino-9-ethylcarbazole tablets (Sigma) dissolved in 10% N,N-dimethylformamide, 100 mM imidazole, 20 mM citric acid, pH 7, 0.01%
H2O2) for 15-20 minutes at RT. Sections were counterstained with
hematoxylin (Merck, Darmstadt, Germany) for 30 seconds and
embedded in Aquatex (Merck, Darmstadt, Germany). Controls
included omission of the primary antibody as well as specific
antibody absorption with the immunizing fusion proteins. Indirect
immunocytochemistry was also performed using specific antibodies
against well-known marker proteins for myoepithelial cells;
monoclonal mouse anti-α-smooth muscle actin (Sigma) or
polyclonal rabbit anti-53 kD desmin (Cappel, ICN, Costa Mesa,
CA), epithelial cells; polyclonal rabbit anti-46 kDa cytokeratin
antibody (a gift from Dr Ernstli Reichmann, ISREC, Lausanne,
Switzerland) and basement membrane; polyclonal rabbit anti-αlaminin (Gibco-BRL Life Technologies, Merelbeke, Belgium). The
antibody was visualized either with horseradish peroxidase labeled
second antibody or with rhodamine and FITC labeled second
antibodies (Cappel, ICN, Costa Mesa, CA).
Morphometric analysis
Morphometric analyses were performed using the Olympus CUE-2
computer image analyzing system. The area and intensity of EphB4
Fig. 1. A schematic representation of alveolar crosssections showing the morphological changes seen in
mouse mammary glands during the major stages of the
estrus cycle.
and ephrin-B2 specific staining per stromal unit of 1.6 mm2 were
measured. 20 ducts and alveoli were measured for all developmental
stages examined. The values for all parameters are expressed as mean
± mean standard error.
RESULTS
We have recently demonstrated by northern blot analysis that
the expression of the EphB4 receptor PTK and its ligand
ephrin-B2 is tightly controlled during mammary gland
development, in particular, during the estrus cycle (Andres et
al., 1994; unpublished observation). Furthermore, deregulated
expression was seen in invasive, metastasizing mammary
tumours. These observations suggested that the EphB4
receptor and its cognate ligand, ephrin-B2, may be involved in
the homeostasis of the mammary gland parenchyma. We have
applied indirect immunocytochemistry, using affinity purified
specific antisera, to identify the cell type(s) responsible for the
expression of both the receptor and the ligand and to investigate
their developmental regulation.
Cell-type specificity of EphB4 receptor and ephrinB2 ligand expression
Upon reaching sexual maturity and in the absence of
pregnancy, the mouse mammary gland is subjected to repeated
cycles of limited proliferation, differentiation and cell death in
response to the fluctuating hormonal stimulation of the
menstrual cycle. This is reflected by the morphological
appearance of the epithelial alveoli schematically presented in
Fig. 1. During the proestrus phase of the mouse ovarian cycle
a multilayered alveolar epithelium is indicative of epithelial
expansion. At ovulation (estrus) lumen formation is increased
in the multilayered epithelial alveoli. During the luteal phase,
metestrus, the epithelium differentiates into single layered
alveoli with limited secretory activity. The anestrus phase of
the cycle is characterized by alveoli with a flattened epithelium
and regional cell death accounting for the constancy of the
parenchymal cell numbers in the mature mammary gland
(Andres et al., 1995).
2744 Z. Nikolova and others
Fig. 2. Localization of EphB4 receptor and ephrin-B2 ligand
expression in the mammary gland. Sections of mammary
glands in the estrus phase were reacted with anti-EphB4 and
visualized with FITC labeled anti-rabbit IgG (A), anti-smooth
muscle actin (SMA) visualized with rhodamine labeled antimouse IgG (B), anti-EphB4 and anti-SMA visualized with
FITC labeled anti-rabbit and rhodamine labeled anti-mouse
IgG respectively (C), FITC labeled anti-ephrin-B2 (D), FITC
labeled anti-ephrin-B2 and anti-SMA visualized with
rhodamine labeled anti-mouse IgG (E), FITC labeled antiephrin-B2 and anti-EphB4 visualized with rhodamine labeled
anti-rabbit IgG (F). (C,E,F) Photographed using a
green/blue/red filter (Leica, 513836). Bars, 50 µm.
In the mature mammary gland, the EphB4 receptor was
localized to the myoepithelial cells of alveolar and ductal
structures (Fig. 2A-C). This is evidenced by the complete
overlap of the receptor staining (green, Fig. 2A) with that of
smooth muscle actin (SMA), a mammary myoepithelial cell
marker (red, Fig. 2B), yielding the yellow appearance of
staining when superimposed (Fig. 2C). In contrast, the cognate
ligand ephrin-B2 was localized to the luminal epithelial cells
(green, Fig. 2D) surrounded by the SMA-positive
myoepithelial cells (red, Fig. 2E). Interestingly, not all luminal
epithelial cells were positive for the ligand protein (Fig. 2D and
F). The distinct localization of the receptor and ligand was
confirmed by double immunofluorescence using FITC
conjugated ligand antibody and rhodamine based indirect
detection of the receptor antibody (Fig. 2F). The yellow
staining in Fig. 2F probably reflects overlapping light emission
in the contact areas between luminal epithelial and
myoepithelial cells.
Receptor and ligand expression during normal
development
In the immature female, a rudimentary mammary ductal
system is embedded in the adipose stroma and restricted to the
nipple region of the mammary gland. Immunostaining of 2 and
3 week old mammary glands with either EphB4 or ephrin-B2
specific antibodies failed to detect the receptor and ligand
proteins (Fig. 3A,B). This lack of staining was not due to the
absence of either myoepithelial or epithelial cells since these
cells were readily detectable with SMA and cytokeratin 46kD
specific antisera respectively (Fig. 3C, and data not shown).
The pubertal mammary epithelium exhibits characteristics
of invasive growth, penetration of the basement membrane and
invasion of the fatty tissue (Williams and Daniel, 1983).
Expression of both the EphB4 receptor and the ephrin-B2
ligand was induced at the onset of puberty. Intensive expression
of the EphB4 receptor protein was observed in the
myoepithelial cells. In addition, weak staining of some
epithelial cells was evident at this developmental stage (Fig.
3D,F). Ephrin-B2 ligand expression was confined to the
epithelial cells of ducts and end-buds (Fig. 3E).
In the mammary glands of adult females during the estrus
cycle, both the ephrin-B2 ligand as well as the EphB4 receptor
persisted in the luminal epithelial and myoepithelial cells,
respectively, during the estrogen dominated proliferative
phases of the cycle (Fig. 3G-L). Significantly less ephrin-B2
ligand and EphB4 receptor protein was detected at metestrus,
the progesterone dominated differentiation phase of the cycle
(Fig. 3M-O). During the anestrus phase of the cycle, after the
collapse of the corpus luteum, less ephrin-B2 ligand protein
could be detected in the epithelial cells (Fig. 3Q). These results
have been confirmed by a semi-quantitative analysis of EphB4
and ephrin-B2 expression during development (Fig. 4A,B).
Strikingly, at the anestrus stage of the cycle, no EphB4 receptor
protein could be detected although the myoepithelium was
clearly present (Fig. 3P,R). The decrease of EphB4 receptor
expression during the cycle, culminating in total absence at the
anestrus
phase,
was
substantiated
by
double
immunofluorescence with anti-EphB4 receptor and anti-SMA
antibodies (Fig. 3I,L,O,R).
Receptor and ligand expression are dependent on
estrogen
The parallels between maximal receptor and ligand expression
and elevated estrogen levels prompted us to investigate the
influence of estrogen on the expression of the EphB4 receptor
and the ephrin-B2 ligand in ovariectomized mice. Animals
were ovariectomized either pre-pubertally at 3 weeks of age or
as adults at 12 weeks of age and analyzed three weeks after
surgery. Sham operated animals served as controls.
In ovariectomized animals, regardless of the age at
ovariectomy, no receptor or ligand protein was detected (Fig.
5A,B,G,H), even though the myoepithelial-epithelial
EphB4 and ephrin-B2 in mammary morphogenesis 2745
organization of the parenchyma was maintained (Fig. 5C,I).
Strikingly, the expression of both the EphB4 receptor and the
ephrin-B2 ligand could be restored by injection of estradiol for
10 days, the staining pattern for both being indistinguishable
from non-ovariectomized animals (Fig. 5D,E,J,K). A semiquantitative analysis of EphB4 receptor expression is presented
in Fig. 6A. Expression was not affected in either sham operated
animals or in sham operated, estrogen treated adult animals. In
Fig. 3. Developmental regulation
of EphB4 receptor and ephrin-B2
ligand expression. Sections of
mammary glands of 3 week old
females (A-C), 6 week old females
at puberty (D-F), in the proestrus
(G-I), in the estrus (J-L), in the
metestrus (M-O) and in the
anestrus (P-R) phases of the estrus
cycle were reacted with either antiEphB4 (A,D,G,J,M,P), antiephrin-B2 (B,E,H,K,N,Q) antisera
and visualized with peroxidase
coupled anti-rabbit IgG. Sections
for double immunofluorescence
were reacted with anti-EphB4 and
anti-SMA, visualized with FITC
labeled anti-rabbit and rhodamine
labeled anti-mouse IgG,
respectively, and photographed
using a green/blue/red filter
(C,F,I,L,O,R). Sections were
counterstained with hematoxylin.
Bars, 50 µm.
2746 Z. Nikolova and others
A
B
Fig. 4. Morphometric quantification of the EphB4 receptor and
ephrin-B2 ligand proteins in mammary glands at puberty and during
the estrus cycle. The density of EphB4 antibody staining in
myoepithelial cells (A) and of ephrin-B2 antibody staining in
epithelial cells (B) is represented schematically. Bars indicate the
mean standard error, n=20.
contrast, ovariectomy led to the complete ablation of receptor
protein. Administration of estrogen was sufficient to induce
receptor expression to almost normal levels in mature or
pubertal ovariectomized females.
In order to investigate at which level estrogen influences the
expression of the EphB4 receptor, RNA isolated from control
and experimental animals was probed for EphB4 receptor
expression and ribosomal protein L7a expression as an internal
control (Fig. 6B). Densitometric scanning and calculation of
the ratio between EphB4 receptor and L7a ribosomal protein
RNA levels revealed that the EphB4 receptor RNA was not
significantly affected by ovariectomy eventhough the EphB4
receptor protein was completely absent (Fig. 6A). Estrogen
treatment of either ovariectomized adult or pubertal females
led to a two-fold increase in EphB4 receptor mRNA levels.
These results indicate that the marked effects of estrogen seen
at the protein level are not reflected by the amounts of RNA
detected. A similar discrepancy between the protein and the
RNA levels was also seen for the ephrin-B2 ligand (data not
shown).
Receptor and ligand expression in malignant
mammary development
The preferential expression of the EphB4 receptor and ephrinB2 ligand during the proliferative, estrogen controlled stages
of mammary gland development prompted us to investigate
their expression during carcinogenesis. We have analyzed
immunohistochemically the expression of the EphB4 receptor
and the ephrin-B2 ligand proteins in Wap-ras (undifferentiated,
invasive) and in Wap-myc (well-differentiated, non-invasive)
induced mammary neoplasms from transgenic mice. These
mice develop tumours of different grades of malignancy
thereby allowing the analysis of progressive stages of
mammary carcinogenesis (Andres et al., 1991). Both types of
tumours express estrogen receptor as determined by northern
blot analyses (data not shown).
In the highly differentiated, non-invasive Wap-myc tumours,
EphB4 receptor expression was restricted to the myoepithelial
cells surrounding the dilated ducts and cysts (Fig. 7A), whereas
the ephrin-B2 ligand protein was absent (Fig. 7B). The Wapras transgene induces undifferentiated tumours which rapidly
acquire an invasive phenotype (Andres et al., 1991). In the in
situ stage of Wap-ras induced carcinogenesis, intense EphB4
receptor staining was confined to the myoepithelium
surrounding the hyperplastic foci (Fig. 7C). In the more
advanced stages of tumorigenesis, characterized by stromal
invasion of the rapidly proliferating tumor cells and the
disappearance of the myoepithelial cells, EphB4 receptor
protein could no longer be detected in the periphery of the
ductal structures. Instead, EphB4 receptor expression was seen
in single anaplastic cells interspersed in the mainly EphB4
receptor negative tumor mass (Fig. 7E). In the highly
anaplastic, aggressively growing tumors variable EphB4
receptor expression was observed in all anaplastic cells (Fig.
7G). Counter-staining with either SMA or cytokeratin
antibodies confirmed the epithelial nature of the tumor mass
(data not shown). Strikingly, the ephrin-B2 ligand was
undetectable at all stages of the invasive Wap-ras
carcinogenesis (Fig. 7D,F,H). Analysis of multiple tumor
samples from each stage revealed a general pattern of EphB4
and ephrin-B2 expression during mouse mammary
carcinogenesis: the loss of ligand expression at the earliest
stages of carcinogenesis and the transition of EphB4 receptor
expression from myoepithelial to epithelial cells with
progression of the malignant process.
DISCUSSION
Immunohistochemical analyses have revealed a spatially and
temporarily regulated expression of the EphB4 receptor and its
cognate ephrin-B2 ligand during mammary gland
development. Both receptor and ligand exhibited their highest
expression during the proliferative stages of the mammary
epithelium, associating them with the processes of growth
and/or tissue remodelling. Accordingly, expression of both
receptor and ligand was drastically down-regulated during
pregnancy and absent during the end-differentiated state at
lactation (data not shown). The major localization of the
EphB4 receptor was in the myoepithelial cells of ducts and
alveoli, while the ephrin-B2 ligand was limited to the luminal
epithelial cells. The myoepithelial cells are located, depending
on the functional state of the gland, either as a discontinuous
or continuous layer between the basement membrane and the
luminal epithelial cells and are connected to the epithelial cells
by gap junctions and desmosomes. Although a mainly
contractile function has been attributed to the myoepithelium,
EphB4 and ephrin-B2 in mammary morphogenesis 2747
Fig. 5. Estrogen dependent
EphB4 receptor and ephrin-B2
ligand expression. Sections of
mammary glands of 6.5 week
old pubertal females
ovariectomized pre-pubertally
at 3 weeks (A-C), females
ovariectomized at 3 weeks and
treated at 5 weeks with
estrogen (D-F), mature
females 3 weeks after
ovariectomy (G-I) and
ovariectomized mature
females treated with estrogen
30 days after surgery (J-L).
Sections were reacted either
with anti-EphB4 (A,D,G,J),
anti-ephrin-B2 (B,E,H,K) or
anti-SMA (C,F,I,L) antisera
and visualized with
peroxidase coupled anti-rabbit
or anti-mouse IgG,
respectively. Sections were
counterstained with
hematoxylin. Bars, 50 µm.
the intense intercellular communication suggests that the
myoepithelium contributes significantly to the morphological
and functional organization of the gland. This is supported by
the observation that the myoepithelium participates in the
constitution and degradation of the basement membrane and in
the hormone dependent cell proliferation and death
(Monteaguo et al., 1990; Djonov et al., 1995; Andres et al.,
1995). Interestingly, modest EphB4 receptor expression was
found also in the epithelial cells during puberty, a
developmental stage characterized by estrogen dependent
proliferation. In the mammary gland, both epithelial and
myoepithelial cells are of ectodermal origin and are thought to
originate from the same stem cell population interspersed in
the parenchyma (Sonnenberg et al., 1986). These stem cells
have a high capacity to proliferate and are the main source of
epithelial cell growth during puberty and possibly also during
the estrus cycle (Russo and Russo, 1994). It remains to be
elucidated if such undifferentiated, proliferating stem cells are
the source of the epithelial EphB4 expression prior to their
differentiation into myoepithelial cells.
Our results demonstrate the crucial importance of estrogens
for the presence of both EphB4 receptor and ephrin-B2 ligand
proteins. Although the effect of progesterone on the expression
of EphB4 receptor and ephrin-B2 ligand expression remains to
be investigated, the onset of their expression at early puberty
when progesterone is not yet active (Haslam, 1989) and the
efficient induction of the proteins by estrogen alone supports
the notion that estrogen represents the main inducer of the
EphB4 receptor and ephrin-B2 ligand proteins. This suggests
that receptor-ligand interactions may link changes in estrogen
levels to a signaling pathway that imposes cellular growth and
organization. This contention is supported by the observation
that estrogen depletion severely impairs mammary epithelial
outgrowth (Korach et al., 1996; Topper and Freeman, 1980;
2748 Z. Nikolova and others
Fig. 6. Morphometric protein quantification and RNA analysis of
EphB4 receptor expression in mammary glands of ovariectomized
and estrogen treated females. (A) The density of EphB4 antibody
staining in myoepithelial cells is represented schematically. Bars
indicate the mean standard error, n=20. (B) Analysis of EphB4
receptor and L7a ribosomal protein RNA in mammary glands of
ovariectomized and estrogen treated females. 5 µg of poly(A)
enriched RNA prepared from mammary glands at the stages
indicated were analyzed on northern blots. The probes used for
hybridization (EphB4 and L7a) and the positions of the 28S and 18S
rRNA on the ethidium bromide stained gel are indicated. The ratio
between EphB4 and L7a RNA levels were 0.5 (Ad C), 0.6 (Ad E),
1.0 (Ad Ov, E), 0.4 (Ad Ov), 0.9 (Pu Ov, E) and 0.5 (Pu Ov),
respectively. Ad C: mature adult control animal; Ad E: mature adult
female treated with estrogen; Ad Ov, E: mature adult female treated
30 days after ovariectomy with estrogen; Ad Ov: mature adult female
3 weeks after ovariectomy; Pu Ov, E: estrogen treated pubertal 6.5
week old females ovariectomized pre-pubertally at 3 weeks; Pu Ov:
pubertal 6.5 week old females ovariectomized pre-pubertally at 3
weeks.
unpublished observation). The drastic effect of estrogen was
not seen at the RNA levels. Since the estrogen receptor is a
transcription factor (Evans, 1988), this observation suggests an
indirect effect of the activated estrogen receptor on the
translation and/or stabilization of the EphB4 receptor and
ephrin-B2 ligand proteins. An indirect mitogenic effect of
estrogen via induction of epidermal growth factor (EGF) has
been observed in the estrogen dependent outgrowth of the
epithelial cells of terminal end-buds at puberty, which
themselves are devoid of estrogen receptors (Snedeker et al.,
1991; Daniel et al., 1987). Several hormone-initiated signal
transduction cascades modulate the expression of growth
factors and/or their receptors (De Bortoli and Dati, 1997) and
it has been proposed that polypeptide growth factors may act
as autocrine and paracrine mediators of estrogen-induced
mitogenesis (Dickson and Lippman, 1987). Estrogens induce
mRNA and protein for both EGF (Huet-Hudson et al., 1990;
DiAugustine et al., 1988) and its receptor PTK (Gardner et al.,
1989; Mukku and Stancel, 1985) in the rodent reproductive
tract. Moreover, it has been observed that EGF may partially
mediate estrogen-induced growth and differentiation by an
interaction between the EGF signaling pathway and the
classical estrogen receptor (Curtis et al., 1996). Thus the crosstalk between receptor PTKs and steroid hormone receptors
may be of importance for modulation and/or targeting of
hormone induced signals.
Induction, involving mutual interaction of cells of different
origin, is the major principle in early embryogenesis. Since
proper functional and spatial development of the mammary
gland is strictly dependent on mesenchymal-epithelial
interactions, this organ offers an excellent example of the
inductive principle maintained in adult life. Indeed,
developmental stage specific expression of Wnt and Hox genes
in the adult mammary gland has been observed (Gavin and
McMahon, 1992; Friedman et al., 1994), two gene families
specifying developmental pathways during embryonic
development. The main mediators of inductive signals at the
earliest cellular determination stages in amphibians and
Drosophila include ligands of the PTK receptor family
(Yamaguchi and Rossant, 1995; Schüpbach and Roth, 1994).
The Eph family and its ligands have been shown to participate
in the pattern formation and establishment of the body plan
during embryogenesis (Gale et al., 1996). Recently it has been
shown that a transmembrane Eph family ligand, ephrin-B1,
participates in the early blastula formation in Xenopus laevis
by modulating cell adhesion (Jones et al., 1998). The gradientlike expression of Eph family receptors and their ligands
observed during both neural and embryonic development
underscore their participation in cellular induction and/or
movement (Monschau et al., 1997; Gale et al, 1996). It has long
been established that the differentiative pathway of the
mammary epithelial cells depends on their position within the
ductal system (Williams and Daniel, 1983), however, the
determinants of positioning and ductal pattern formation are
largely unknown. The strict compartmentalization and the tight
temporal control of receptor and ligand expression suggests
that EphB4 and its ligand ephrin-B2 may participate in
positioning and pattern formation in the adult mammary gland.
Receptor ligand interaction may enable pathfinding and correct
organization of the growing epithelium similar to the role of
EphA5 and ephrin-A5 during axonal growth (Drescher et al.,
1995).
Deregulated expression of PTKs and autocrine ligand
stimulation has been implicated in a variety of neoplastic
diseases including those of the mammary gland (Rodriguez and
Park, 1994). We have analyzed the expression of EphB4
receptor and ephrin-B2 ligand during malignant mammary
epithelial development using a transgenic mouse model
system. Carcinogenesis was characterized by a transition of
EphB4 receptor expression from the myoepithelial cells to the
anaplastic tumor cells. Considering the modest epithelial
localization of the EphB4 receptor during the proliferative
stages at puberty and in the estrus cycle, the localization of the
receptor in tumor cells could reflect their origin from this
(pluripotent?) epithelial cell population. Alternatively, the
EphB4 and ephrin-B2 in mammary morphogenesis 2749
Fig. 7. EphB4 receptor and
ephrin-B2 ligand expression
in experimental mouse
mammary tumors. Sections of
a differentiated Wap-myc
induced tumor (A,B), a Wapras induced in situ carcinoma
(C,D), an invasive Wap-ras
induced carcinoma (E,F) and a
highly anaplastic, invasive
Wap-ras induced tumor (G,H)
were reacted either with antiEphB4 (A,C,E,G) or antiephrin-B2 antisera (B,D,F,H)
and visualized with
peroxidase coupled anti-rabbit
IgG. Sections were
counterstained with
hematoxylin. Bars, 50 µm.
EphB4 receptor expression only in the anaplastic tumor cells
could reflect their increasing degree of de-differentiation. In
contrast to the EphB4 receptor, no ephrin-B2 ligand expression
could be detected in mammary tumors of both Wap-myc and
Wap-ras transgenic mice irrespective of the grade of
malignancy. These observations suggest that the presence of
the ligand is of vital importance for the normal growth in the
mammary parenchyma, uncontrolled growth being the
consequence of ligand loss. It is conceivable that signal(s)
normally emanating from the cytoplasmic face of the
transmembrane ligand molecule may be obligatory for the
maintenance of organ structure and function. In this sense, the
ligand may be exerting a ‘tumor suppresser’ function.
Alternatively, other member(s) of the ligand family could be
expressed in the tumor cells compensating for the loss of
ephrin-B2 ligand and/or supporting their malignant
progression.
Epidemiological and experimental studies have shown that
2750 Z. Nikolova and others
breast cancer occurs almost exclusively in sexually mature
women. The prepubertal status, pregnancy at an early age
followed by full-term lactation and ovariectomy at early age
are among the best known protective factors (Marshall, 1993;
Russo and Russo, 1994). This indicates that the estrogen
induced changes which occur at puberty are not only essential
for the normal mammary gland development and homeostasis,
but also render the mammary parenchyma refractive to
neoplastic development. The hormonal fluctuations occurring
throughout the estrus cycle activate a complex growth program
in the mammary parenchymal cells and may influence their
neoplastic alteration. Indeed, experimental studies have shown
that susceptibility of the mammary epithelium to malignant
transformation varies during the estrus cycle being especially
high during the proliferative, estrogen-dominated phase of the
cycle (Anderson and Beattie, 1992). It is this stage that exhibits
the strongest expression of both the EphB4 receptor and the
ephrin-B2 ligand. Deregulation of the balance between ligand
and receptor, such as the down-regulation of ligand expression
observed in tumors, may be instrumental in the development
of malignant growth characteristics. Interestingly, several Eph
family receptor and ligand genes are clustered in a region of
human chromosome 1 which is lost in 42% of all human breast
tumors examined (Cerretti et al., 1996; Nagai et al., 1995).
We thank Prof. Robert R. Friis (Department of Clinical Research,
University of Berne) for continuing support, encouragement and
critical reading of the manuscript. The help of Dorothea Brazitzova
(Freie Universität, Berlin) with the morphometric analyses is
gratefully acknowledged. This work was supported by the Bernese
Cancer League, Boehringer Ingelheim International, Stiftung Prof.
Max Cloëtta, the Swiss Foundation for Clinical and Experimental
Cancer Research and the Swiss National Science Foundation (3142278.94 and 31-45599.95).
REFERENCES
Anderson, C. H. and Beattie, C. W. (1992). Cellular kinetics of rat mammary
gland terminal end bud epithelium exposed to N-methyl-N-nitrosourea in
vivo. Cancer Res. 52, 5076-5081.
Andres, A.-C., Bchini, O., Schubaur, B., Dolder, B., LeMeur, M. and
Gerlinger, P. (1991). H-ras induced transformation of mammary epithelium
is favored by increased oncogene expression or by inhibition of mammary
regression. Oncogene 6, 771-779.
Andres, A.-C., Reid, H. H., Zürcher, G., Blaschke, R. J., Albrecht, D. and
Ziemiecki, A. (1994). Expression of two novel eph-related receptor protein
tyrosine kinases in mammary gland development and carcinogenesis.
Oncogene 9, 1461-1467.
Andres, A.-C., Zürcher, G., Djonov, V., Flück, M. and Ziemiecki, A. (1995).
Protein tyrosine kinase expression during the estrous cycle and
carcinogenesis of the mammary gland. Int. J. Cancer 63, 288-296.
Bennett, B. D., Zeigler, F. C., Gu, Q., Fendly, B., Goddard, A. D., Gillett,
N. and Matthews, W. (1995). Molecular cloning of a ligand for the ephrelated receptor protein-tyrosine kinase Htk. Proc. Nat. Acad. Sci. USA 92,
1866-1870.
Brambilla, R. and Klein, R. (1995). Telling axons where to grow: a role for
Eph receptor tyrosine kinases in guidance. Mol. Cell. Neurosci. 6, 487-495.
Brückner, K., Pasquale, E. B. and Klein, R. (1997). Tyrosine
phosphorylation of transmembrane ligands for Eph receptors. Science 275,
1640-1643.
Cerretti, D. P., Lyman, S. D., Kozlosky, C. J., Copeland, N. G., Gilbert, D.
J., Jenkins, N. A., Valentine, V., Kirstein, M. N., Shapiro, D. N. and
Morris, S. W. (1996). The genes encoding the eph-related receptor tyrosine
kinase ligands LERK-1 (EPLG1, Epl1), LERK-3 (EPLG3, Epl3), and
LERK-4 (EPLG4, Epl4) are clustered on human chromosome 1 and mouse
chromosome 3. Genomics 33, 277-282.
Curtis, S. W., Washburn, T., Sewall, C., DiAugustine, R., Lindzey, J.,
Couse, J. F. and Korach, K. S. (1996). Physiological coupling of growth
factor and steroid receptor signaling pathways: estrogen receptor knockout
mice lack estrogen-like response to epidermal growth factor. Proc. Nat.
Acad. Sci. USA 93, 12626-12630.
Daniel, C. W., Silberstein, G. B. and Strickland, P. (1987). Direct action of
17β-estradiol on mouse mammary ducts analyzed by sustained release
pellets and steroid autoradiography. Cancer Res. 47, 6052-6057.
Davis, S., Gale, N. W., Aldrich, T. H., Maisonpierre, P. C., Lothak, V.,
Pawson, T., Goldfarb, M. and Yancopoulos, G. D. (1994). Ligands for
Eph-related receptor tyrosine kinases that require membrane attachment or
clustering for activity. Science 266, 816-819.
De Bortoli, M. and Dati, C. (1997). Hormonal regulation of type I receptor
tyrosine kinase expression in the mammary gland. J. Mammary Gland Biol.
Neoplasia 2, 175-187.
DiAugustine, R. P., Petrusz, P., Bell, G. I., Brown, C. F., Korach, K. S.,
McLachlan, J. A. and Teng, C. T. (1988). Influence of estrogens on mouse
uterine epidermal growth factor precursor protein and messenger ribonucleic
acid. Endocrinology 122, 2355-2363.
Dickson, R. B. and Lippman, M. E. (1987). Estrogenic regulation of growth
and polypeptide growth factor secretion in human breast carcinoma. Endocr.
Rev. 8, 29-43.
Djonov, V., Altermatt, H. J., Arnold, A.-M., Gerber, H. A. and Andres, A.C. (1995). Recapitulation of a normal cellular growth program in early
invasive breast cancer. Int. J. Oncol. 7, 311-318.
Drescher, U., Kremoser, C., Handwerker, C., Löschinger, J., Noda, M. and
Bonhoeffer, F. (1995). In vitro guidance of retinal ganglion cell axons by
RAGS, a 25 kDa tectal protein related to ligands for Eph-receptor tyrosine
kinases. Cell 82, 359-370.
Evans, R. M. (1988). The steroid and thyroid hormone receptor superfamily.
Science 240, 889-895.
Fox, S. B. and Harris, A. L. (1997). The epidermal growth factor in breast
cancer. J. Mammary Gland Biol. Neoplasia 2, 131-142.
Friedmann, Y., Daniel, C. A., Strickland, P. and Daniel, C. W. (1994). Hox
genes in normal and neoplastic mouse mammary gland. Cancer Res. 54,
5981-5985.
Gale, N. W., Holland, S. J., Valenzuela, E. M., Flenniken, A., Pan, L., Ryan,
T. E., Henkemeyer, M., Strebhardt, K., Hirai, H., Wilkinson, D. G. et
al. (1996). Eph receptors and ligands comprise two major specificity
subclasses and are reciprocally compartmentalized during embryogenesis.
Neuron 17, 9-19.
Gardner, R. M., Verner, G., Kirkland, J. L. and Stancel, G. M. (1989).
Regulation of uterine epidermal growth factor (EGF) receptors by estrogen
in the mature rat and during the estrous cycle. J. Steroid Biochem. 32, 339343.
Gavin, B. J. and McMahon, A. P. (1992). Differential regulation of the wnt
gene family during pregnancy and lactation suggests a role in postnatal
development of the mammary gland. Mol. Cell. Biol. 12, 2418-2423.
Harlow, E. and Lane, D. (1988). Antibodies, A Laboratory Manual Cold
Spring Harbor Laboratory Press. Cold Spring Harbor. NY. 726 pp.
Haslam, S. Z. (1989). The ontogeny of mouse mammary gland responsiveness
to ovarian steroid hormones. Endocrinology 125, 2766-2772.
Hirai, H., Maru, Y., Hagiwara, K., Nishida, J. and Takaku, F. (1987). A
novel putative tyrosine kinase receptor encoded by the eph gene. Science
238, 1717-1720.
Hirai, Y., Lochter, A., Galosy, S., Koshida, S., Niwa, S. and Bissell, M. J.
(1998). Epimorphin functions as a key morphoregulator for the mammary
gland. J. Cell Biol. 140, 159-169.
Holland, S. J., Gale, N. W., Mbamalu, G., Yancopoulos, G. D.,
Henkemeyer, M. and Pawson, T. (1996). Bidirectional signaling through
the EPH-family receptor Nuk and its transmembrane ligands. Nature 383,
722-725.
Howlett, A. R. and Bissell, M. J. (1993). The influence of tissue
microenvironment (stroma and extracellular matrix) on the development and
function of mammary epithelium. Epith. Cell Biol. 2, 79-89.
Hynes, N., Cella, N. and Wartmann, M. (1997). Prolactin mediated
intracellular signalling in mammary epithelial cells. J. Mammary Gland
Biol. Neoplasia 2, 19-28.
Huet-Hudson, Y. M., Chakraborty, C., De, S. K., Suzuki, Y., Andrews, G.
K. and Dey, S. K. (1990). Estrogen regulates the synthesis of epidermal
growth factor in mouse uterine epithelial cells. Mol. Endocrinol. 4, 510-523.
Imagawa, W., Yang, J., Guzman, R. and Nandi, S. (1994). Control of
mammary gland development. In The Physiology of Reproduction, vol. 2,
EphB4 and ephrin-B2 in mammary morphogenesis 2751
2nd edn (ed. E. Knobil and J. D. Neill), pp. 1033-1065. Raven Press, New
York.
Jones, T. L., Chong, L. D., Kim, J., Xu, R.-H., Kung, H.-F. and Daar, I. O.
(1998). Loss of cell adhesion in Xenopus laevis embryos mediated by the
cytoplasmic domain of Xlerk, an erythropoietin-producing hepatocellular
ligand. Proc. Nat. Acad. Sci. USA 95, 576-581.
Kleinberg, D, L. (1997). Early mammary development: growth hormone and
IGF-1. J. Mammary Gland Biol. Neoplasia 2, 49-57.
Korach, K. S., Couse, J. F., Curtis, S. W., Washburn, T. F., Lindzey, J.,
Kimbro, K. S., Eddy, E. M., Migliaccio, S., Snedeker, S. M, Lubahn, D.
B. et al. (1996). Estrogen receptor gene disruption: molecular
characterization and experimental and clinical phenotypes. Recent Prog.
Horm. Res. 51, 159-186.
Kratochwil, K. (1987). Epithelium-mesenchyme interactions in the fetal
mammary gland. In Cellular and Molecular Biology of Mammary Cancer
(ed. D. Medina, W. Kidwell, G. Heppner and E. Anderson), pp. 67-80.
Plenum Press, New York.
Lin, C. Q. and Bissell, M. J. (1993). Multi-faceted regulation of cell
differentiation by extracellular matrix. FASEB J. 7, 37-743.
Lindberg, R. A. and Hunter, T. (1990). cDNA cloning and characterization
of eck, an epithelial cell receptor protein-tyrosine kinase in the eph/elk
family of protein kinases. Mol. Cell Biol. 10, 6316-6324.
Maniatis, T., Fritsch, E. F. and Sambrook, J. (1982). Molecular Cloning. A
Laboratory Manual. Cold Spring Harbor Laboratory Press, Cold Spring
Harbor, New York. 545 pp.
Marshall, E. (1993). Special report: The politics of breast cancer. Science 259,
616-628.
Monteaguo C., Merino, M. J., San-Juan, J., Liotta, L. A. and StetlerStevenson, W. G. (1990). Immunohistochemical distribution of type IV
collagenase in normal, benign and malignant breast tissue. Am. J. Pathol.
136, 585-592.
Monschau, B., Kremoser, C., Ohta, K., Tanaka, H., Kaneko, T., Yamada,
T., Handwerker, C., Hornberger, M. R., Loeschinger, J., Pasquale, E. B.
et al. (1997). Shared and distinct functions of RAGS and ELF-1 in guiding
retina axons. EMBO J. 16, 1258-1267.
Mukku, V. R. and Stancel, G. M. (1985). Regulation of epidermal growth
factor receptor by estrogen. J. Biol. Chem. 260, 9820-9824.
Nagai, H., Negrini, M., Carter, S. L., Gillum, D. R., Rosenberg, A. L.,
Schwartz, G. F. and Croce, C. M. (1995). Detection and cloning of a
common region of loss of heterozygosity at chromosome 1p in breast cancer.
Cancer Res. 55, 1752-1757.
Orioli, D. and Klein, R. (1997). The Eph receptor family: axonal guidance
by contact repulsion. Trends Genet. 13, 354-359.
Pandey, A., Lindberg, R. A. and Dixit, V. M. (1995). Receptor orphans find
a family. Curr. Biol. 5, 986-989.
Pasquale, E. B. (1997). The Eph family of receptors. Curr. Biol. 9, 608-615.
Rodriguez, G. A. and Park, M. (1994). Oncogenic activation of tyrosine
kinases. Curr. Opin. Genet. Dev. 4, 15-24.
Roskelley, C. D., Desperez, P. Y. and Bissell, M. J. (1994). Extracellular
matrix-dependent tissue-specific gene expression in mammary epithelial
cells requires both physical and biochemical signal transduction. Proc. Nat.
Acad. Sci. USA 91, 12378-12382.
Russo, J. and Russo, I. H. (1994). Toward a physiological approach to breast
cancer prevention. Cancer Epidemiol. Biomarkers Prevent. 3, 353-364.
Sakakura, T. (1991). New aspects of stroma-parenchyma relations in
mammary gland differentiation. Int. Rev. Cytol. 125, 165-202.
Schoenenberger, C.-A., Andres, A.-C., Groner, B., van der Valk, M. A.,
LeMeur, M. and Gerlinger, P. (1988). Targeted c-myc gene expression in
mammary glands of transgenic mice induces mammary tumors with
constitutive milk protein gene transcription. EMBO J. 7, 169-175.
Schüpbach, T. and Roth, S. (1994). Dorsoventral patterning in drosophila
oogenesis. Curr. Opin. Genet. Dev. 4, 502-507.
Schultz, J., Ponting, C. P., Hofmann, K. and Bork, P. (1997). SAM as a
protein interaction domain involved in developmental regulation. Protein
Sci. 6, 249-253.
Snedeker, S. M., Brown, C. F. and DiAugustine, R. P. (1991). Expression
and functional properties of transforming growth factor (and epidermal
growth factor during mouse mammary gland ductal morphogenesis. Dev.
Biol. 88, 276-280.
Sonnenberg, A., Daams, H., van der Valk, M. A., Hilkens, J. and Hilgers,
J. (1986). Development of mouse mammary gland: Identification of stages
in differentiation of luminal and myoepithelial cells using monoclonal
antibodies and polyvalent antiserum against keratin. J. Histochem.
Cytochem. 34, 1037-1046.
Strange, R., Li, F., Burkhardt, A. and Friis, R. R. (1992). Apoptotic cell
death and tissue remodelling during mouse mammary gland involution.
Development 115, 49-58.
Tessier-Lavigne, M. and Goodman, C. S. (1996). The molecular biology of
axon guidance. Science 274, 1123-1132.
Topper, Y. J. and Freeman, C. S. (1980). Multiple hormone interactions in
the developmental biology of the mammary gland. Physiol. Rev. 60, 10491106.
Williams, J. M. and Daniel, C. W. (1983). Mammary ductal elongation:
Differentiation of myoepithelium and basal lamina during branching
morphogenesis. Dev. Biol. 97, 274-290.
Xu, Q., Alldus, G., Macdonald, R., Wilkinson, D. G. and Holder, N. (1996).
Function of the Eph-related kinase rtk1 in patterning of the zebrafish
forebrain. Nature 381, 319-322.
Yamaguchi, T. P. and Rossant, J. (1995). Fibroblast growth factors in
mammalian development. Curr. Opin. Genet. Dev. 5, 485-491.