| HOME | HELP | FEEDBACK | SUBSCRIPTIONS | ARCHIVE | SEARCH | TABLE OF CONTENTS |
ARTICLES |
5ß1 Integrin Are Under Developmental and Ovarian Steroid Regulation in the Normal Mouse Mammary Gland
Department of Physiology, Michigan State University, East Lansing, Michigan 48824
Address all correspondence and requests for reprints to: Dr. Sandra Z. Haslam, Department of Physiology, 108 Giltner Hall, Michigan State University, East Lansing, Michigan 48824. E-mail: shaslam{at}msu.edu
 |
Abstract |
|---|
 Top Abstract IntroductionMaterials and MethodsResultsDiscussionReferences |
|---|
5ß1, was localized in myoepithelial cells;
it increased 2.2-fold between puberty and sexual maturity and decreased
in late pregnancy and lactation. The basal localization of
5ß1 was notably increased in pubertal and
adult virgin mice. 5ß1 concentrations
decreased 40–50% after ovariectomy in pubertal and adult mice, which
was reversed by estrogen plus progesterone treatment in adult mice. The
high basal expression of 5ß1 during active
proliferation and the low expression in nonproliferating and lactating
glands indicate that fibronectin signaling may be required for
hormone-dependent proliferation in the mammary gland. |
Introduction |
|---|
|
TopAbstractIntroductionMaterials and MethodsResultsDiscussionReferences |
|---|
In vitro studies have shown a role for ECM proteins in ovarian steroid action. The progestin, R5020, significantly stimulated proliferation of primary mouse mammary epithelial cells cultured on fibronectin (FN) (5). However, this study also determined that R5020 did not stimulate proliferation of mouse mammary epithelial cells when the cells were plated on collagen I (Col I), laminin (LM), tenascin, or a nonspecific attachment factor poly-L-lysine. ECM proteins also regulate the levels of insulin-like growth factor I (IGF-I) and epidermal growth factor receptors, the levels of IGF binding proteins, and the proliferative responses to epidermal growth factor and IGF-I in primary mammary epithelial cells (7). There is also substantial in vitro evidence that the ECM protein LM, in concert with lactogenic hormones, is required for differentiation of mammary epithelial cells into lactational cells (6, 9). Therefore, cell binding to specific ECM proteins may be a prerequisite for certain growth factor and hormone responses, suggesting that the expression and regulation of ECM proteins and integrins in the mammary gland may be important in the regulation of mammary gland development and lactation.
Although FN is required for progestin-dependent mammary epithelial cell
proliferation in vitro, nothing is known about the role of
FN in the mammary gland in vivo (5). Thus, in
the present study we have quantitatively analyzed the in
vivo spatial and temporal concentrations of FN as well as those of
Col I, Col IV, and LM during postnatal mammary gland development in the
mouse. As ovarian steroids play a major role in mammary gland
proliferation, morphogenesis, and lactational function, we also
analyzed the effects of ovariectomy (OVX), and estrogen and
progesterone (E+P) treatment on ECM and integrin expression. We found
that among ECM proteins, only FN exhibited significant changes in level
as a function of developmental and hormonal states. The
5ß1 FN integrin also
exhibited changes in expression and intracellular distribution that
were regulated by E and P. The temporal and spatial patterns of FN, the
5ß1 integrin
concentration, and the pattern of intracellular
5ß1 integrin
localization were consistent with roles for FN and the
5ß1 integrin in
E-induced proliferation at puberty and in E+P-induced epithelial cell
proliferation in the adult mammary gland during pregnancy.
|
Materials and Methods |
|---|
|
TopAbstractIntroductionMaterials and MethodsResultsDiscussionReferences |
|---|
5ß1 expression and
distribution, 5- and 10-week-old mice were OVX 1 week before receiving
a single ip injection of vehicle (control), E (1 µg/mouse), or E+P (1
µg and 1 mg/mouse). Ovary-intact, OVX, and hormone-treated mice were
killed at 24, 48, and 72 h after injection, and mammary glands
were snap frozen and stored at -80 C until assayed.
Quantitative immunohistochemistry of ECM proteins and
integrins
The inguinal mammary glands from 3-, 5-, and 10-week-old,
pregnant, and lactating mice were removed and embedded in tissue
embedding media (OCT, Miles Laboratories, Elkhart, IN), snap-frozen,
and sectioned as previously described (11). For the
detection of Col I, Col IV, LM, and FN, frozen sections (5 µm) were
mounted onto poly-L-lysine-coated coverslips, fixed with
acetone/methanol, blocked with 0.5% casein/0.01 M PBS (pH
7.4), rinsed in PBS, and immunolabeled as previously described
(12). Antimouse LM (dilution, 1:150;
Engelbreth-Holm-Swarm LM-1 was used as an immunogen) and Col IV
(dilution, 1:100) antibodies were obtained from Becton Dickinson and Co. (Bedford, MA). Antimouse FN (dilution, 1:200) antibody
was obtained from Charles River Laboratories, Inc.
(Southbridge, MA), and an anti-Col I (dilution, 1:100) antibody was
purchased from Southern Biotechnology Associates
(Birmingham, AL). Alexa 488-labeled secondary antibodies (green;
dilution, 1:100; Molecular Probes, Inc., Eugene, OR) were
used to identify ECM protein localization. Alexa 488 antibody is
resistant to fading and bleaching. Additionally, the antifade agent,
Prolong (Molecular Probes, Inc.), was added to all
sections at the time of mounting. Frozen sections were also
immunolabeled for v and
5 integrin subunits as described by Vogel
et al. (13) and Hynes et al.
(14), respectively. The v
integrin subunit antibody (dilution, 1:200) was a gift from Dr. Errki
Ruoslahti (La Jolla Cancer Research Foundation, La Jolla, CA)
(13). The 5 integrin antibody
(dilution, 1:75) was supplied by Dr. Richard Hynes (Howard Hughes
Medical Institute, Massachusetts Institute of Technology, Cambridge,
MA) (14). Mounted coverslips were stored at 4 C and
analyzed within 1 week of sectioning. No fading occurred when
coverslips were stored at 4 C for at least 1 month. All tissue to be
used in comparative quantitative immunohistochemistry were processed in
the same experiment. For example, if comparing 3-, 5-, 10-week-old,
pregnant, and lactating mammary tissues for FN, all tissue were
harvested and processed at the same time. Epithelial cell type-specific
localization of integrins was distinguished by double labeling mammary
gland sections from frozen tissue with antibodies to myoepithelial
markers, - smooth muscle actin (clone 1A4; dilution, 1:2000;
Sigma, St. Louis, MO) or cytokeratin 14 (clone LL002;
dilution, 1:400; Lab Vision Corp., Freemont, CA), and the
5 integrin subunit antibody. All integrin
antibodies were raised in rabbits, and both myoepithelial cytoskeletal
antibodies were monoclonal mouse antibodies. Therefore, we used goat
antirabbit Alexa 488 antibody (green; dilution, 1:100) to detect
integrins and goat antimouse Alexa 544 antibody (red; dilution, 1:100;
Molecular Probes, Inc.) to detect myoepithelial cells. The
intensity of integrin staining of ducts and alveoli was measured on the
basal sides of the cell (the plasma membrane of cells in contact with
the basement membrane) and the rest of the cell (apical and lateral
sides). These data were used to obtain a basal ratio, a ratio of the
staining intensity on the basal side of the cell divided by the
intensity in the rest of the cell. A ratio above 1.0 indicates that
basal staining intensity is greater than staining intensity in the rest
of the cell. In all cases, fluorescent images were captured on an
Odyssey Laser Scanning Confocal Microscope (Noran Instruments, Inc.,
Madison, WI), with minimal exposure to laser before capturing the image
using identical laser intensity. No significant bleaching was found
during capture using these conditions. The fluorescence intensity of
captured images was quantified using Image-1 software (Universal
Imaging Corp., Philadelphia, PA) as previously described in detail by
Ankrapp et al. (12). A minimum of three mice
per treatment group and five sections per mouse were used in all
experiments.
Western blotting experiments for ECM proteins
Intact mammary glands or CFP were removed, homogenized, and
sonicated in RIPA buffer [10 mM
Na2HPO4 (pH 7.2), 150
mM NaCl, 2 mM EDTA, 1% Triton X-100, 1% SDS,
and 1% sodium deoxycholic acid] with protease inhibitors (2
mM phenylmethylsulfonylfluoride, 2 mM sodium
orthovanadate, 1 mM sodium fluoride, 10 mM
leupeptin, and 2 µg/ml aprotinin) as previously described
(15). The samples were normalized for protein content
using a protein assay (Bio-Rad Laboratories, Inc.,
Hercules, CA) and then were subjected to SDS-PAGE. Samples were
resolved on an 8% SDS-polyacrylamide gel (5% stacking gel) for Col I,
Col IV, and FN and on a 6% SDS-polyacrylamide gel (3% stacking gel)
for LM. Purified rat Col I (>99% pure; Becton Dickinson and Co.), mouse Col IV (>99% pure; Becton Dickinson and Co.), mouse FN (> 95% pure; Life Technologies, Inc., Gaithersburg, MD), and mouse LM (>99% pure; Becton Dickinson and Co.) were run in parallel to samples. The resolved
proteins were electrotransferred to nitrocellulose and immunostained as
described by Laird et al. (16).
125I-Conjugated secondary antibodies (SA, 15.0
µCi/µg; dilution, 1:2000; ICN Pharmaceuticals, Inc.,
Irvine, CA) were used in Western blotting. The blots were placed in a
PhosphorImager cassette for 18 h, and the band intensity was
quantified (17).
Statistics
All data were expressed as the mean ± SEM.
Differences between means were tested for statistical significance
using Student’s t test or ANOVA followed by Tukey’s test,
as appropriate.
|
Results |
|---|
|
TopAbstractIntroductionMaterials and MethodsResultsDiscussionReferences |
|---|
, A and B,
Intact lanes). The greatest change in concentration was observed for
FN, which increased nearly 3-fold between 3 and 10 weeks of age (Fig. 1, A and B, Intact lanes).
|
, A and B, CFP lanes). Therefore, the 3-fold increase in FN
concentration between 3 and 10 weeks of age in the intact mammary gland
appears to be due to the presence of epithelium.
Immunohistochemical analysis of FN localization during mammary
gland development
As the greatest changes in ECM concentration were detected for FN,
we examined FN temporal and spatial localization by
immunohistochemistry. The stages of mammary gland development examined
were 3 weeks of age (prepuberty), 5 weeks of age (puberty), 10 weeks of
age (sexual maturity), early (9–12 days) and late (18–21 days)
pregnancy, and lactation (7–10 days). In 3-week-old mammary glands, FN
staining was low around ducts, but was intense in the adipose stroma
(Fig. 2A). Between 3 and 10 weeks of age,
the staining intensity of FN surrounding small and large ducts
increased 3- to 4-fold (P < 0.05; Fig. 2, A and B). In
adult 10-week-old animals intense staining for FN was observed in a
thin layer surrounding small ducts and in thicker bands surrounding
large ducts. There were no significant differences in stromal staining
for FN at 3, 5, and 10 weeks of age (Fig. 2, A and B). During pregnancy
and lactation, the thickness of the basement membrane and the thickness
of the band of FN staining surrounding ducts decreased, but staining
intensity was not changed. No change in FN staining was observed in
mammary stroma during pregnancy. Similar to results obtained by Western
blot analysis, the staining intensities of Col I, Col IV, and LM did
not significantly change between 3 and 10 weeks of age or during
pregnancy or lactation (data not shown).
|
, A and B). Treatment
of OVX mice with E or E+P for 4 days did not increase FN concentrations
(data not shown). However, after 2 weeks of treatment with E or E+P,
the FN concentration increased 1.5- and 2-fold, respectively, in intact
mammary glands (Fig. 3, A and B). These data indicate that OVX caused a
70% decrease in FN in the intact gland within 1 week, but synthesis
and accumulation of FN in response to ovarian steroids occurred more
slowly. No changes in FN concentration were observed in mammary stroma
(CFP) after OVX or hormone replacement with E or E+P (Fig. 3). Thus,
the changes observed in FN concentration after OVX and hormone
treatment in the intact mammary gland appear to be dependent upon the
presence of the epithelium.
|
5ß1 integrin
expression during mammary gland development4ß1,
5ß1,
vß1,
vß3, and the
vß6 (18).
However, previous studies have shown that the
4-subunit is not present in normal mammary
cells (19). Furthermore, immunohistochemical analysis
revealed little or no v protein in mammary
epithelial or myoepithelial cells at any stage of development examined
herein (data not shown). Yang et al. reported that the
5 integrin subunit appears to be the key
receptor for FN, and 5 knockout mice most
closely resemble FN knockout mice (18, 20). We detected
5 integrin subunit staining in adult mouse
mammary epithelial and myoepithelial cells. As only the
ß1-subunit protein has been demonstrated to
form a heterodimer with the 5-subunit,
detection of the 5-subunit was interpreted to
mean that the 5ß1
integrin was being detected herein. Little staining for the
5ß1 integrin was
detected in either mammary epithelium or stroma of 3-week-old mice
(Fig. 4A). However, by 5 weeks of age, an
intense, thin discontinuous band of
5ß1 staining was
present at the basal side of ductal cells (Fig. 4, A–C). The intensity
of staining increased more than 2-fold between 3 and 10 weeks of age.
At 10 weeks of age,
5ß1 staining was seen
on the basal side of ductal cells and formed a more continuous band
around both mature ducts as well as smaller ducts. Overall,
5ß1 staining was much
lower in stromal cells than epithelial cells. Intense
5ß1 staining was seen
in epithelial cells in early pregnancy, but was decreased in late
pregnancy and was barely detectable during lactation (Fig. 4, A and
B).
|
5ß1 integrin revealed
modulations in staining intensity on the basal side of the epithelium.
Increased basal localization of staining indicates a higher
concentration of 5ß1
integrin adjacent to the basement membrane. At 3 weeks of age the basal
ratio was slightly higher than 1. This ratio increased to 2.0 by 5
weeks of age and remained near 2.0 at 10 weeks of age and in early
pregnancy (Fig. 4C). The basal ratio decreased in late pregnancy and
was reduced to 1.1 during lactation. The pattern of increased basal
localization of the
5ß1 integrin at 5 and
10 weeks of age and during early pregnancy coincides with stages of
mammary gland development that exhibit high proliferative activity in
the mammary epithelium. Conversely, the decrease in basal localization
during late pregnancy and lactation coincides with minimal epithelial
cell proliferation.
Cell type-specific localization of none
5ß1 integrin
To identify the specific cell type (epithelial vs.
myoepithelial) that expressed the
5ß1 integrin,
colocalization experiments were carried out with the
5 integrin subunit antibody detected by Alexa
488-green and with myoepithelial cell-specific markers, either
-smooth muscle actin or cytokeratin 14 (CK14) antibodies detected by
Alexa 544-red. Figure 5 shows colabeling
with -smooth muscle actin and the
5ß1 integrin
(top panel, A–C) or with CK14 and the
5ß1 integrin
(D). 5ß1
staining was seen around all luminal and basal epithelial cells, with
the strongest staining on the basal surface of basal cells. Nearly all
of the basal cells also stained with -smooth muscle actin and CK14,
indicating that the highest expression of the
5ß1 integrin was in
myoepithelial cells. Occasionally, a basal cell in contact with the
basement membrane would stain strongly for
5ß1, but lack both
-smooth muscle actin and CK14 staining. Thus, both basal epithelial
cells and myoepithelial cells express the
5ß1 integrin.
|
5ß1 integrin5ß1 integrin was
highest during the mammary gland developmental stages that exhibit high
proliferation rates, we hypothesized that
5ß1 integrin levels
might be regulated by E and/or P. Ductal elongation and proliferation
of mammary epithelium of immature 5-week-old mice are mediated by E,
but are unresponsive to mitogenic activity of P. In contrast, ductal
side-branching and lobuloalveolar development in the adult gland and
during early pregnancy are dependent upon E+P. In the adult gland, E
action induces epithelial cell progesterone receptors (PR), and P
acting through its receptor induces ductal side-branching and alveolar
morphogenesis (1). Therefore, OVX immature, 5-week-old
mice were treated with E, mature 10-week-old mice were treated with E
or E+P, and 5ß1
integrin protein expression was analyzed 24–72 h later. In 5-week-old
mice, OVX caused a 50% decrease in expression of the
5ß1 integrin
(P < 0.05; Fig. 6A). The
intensity of 5ß1
integrin staining was partially reversed after E treatment. Although
there was a trend toward decreased basal ratio after OVX and reversal
after E treatment in 5-week mice, it did not reach statistical
significance (Fig. 6B). In 10-week-old mice, OVX decreased
5ß1 integrin
expression by more than 40% (P < 0.05; Fig. 7A). In 10-week mice, no significant
increase in 5ß1
integrin expression was observed after treatment with E alone over the
72-h period (data not shown). In contrast, by 24 h after combined
E+P treatment 5ß1
integrin was significantly increased and reached the level in
ovary-intact mice by 48 h (P < 0.05; Fig. 7A). In
10-week-old mice OVX also caused a significant decrease in the basal
ratio of 5ß1
expression, which was reversed 48 h after E+P treatment (Fig. 7B).
|
|
|
Discussion |
|---|
|
TopAbstractIntroductionMaterials and MethodsResultsDiscussionReferences |
|---|
5ß1 integrin are
developmentally and hormonally regulated in normal mouse mammary gland.
In contrast to FN, the concentrations of Col I, Col IV, and LM did not
exhibit major changes during development. Analysis of hormonal
regulation of the 5ß1
integrin in the adult, 10-week-old mammary gland revealed that the
concentration of the
5ß1 integrin was
significantly down-regulated by OVX and rapidly up-regulated after E+P
treatment in both mammary epithelial and myoepithelial cells. In the
5-week-old pubertal mammary gland, a significant decrease in
5ß1 integrin was
observed after OVX, which was partially reversed after E treatment. The
subcellular localization of the
5ß1 integrin was also
developmentally and hormonally regulated. During periods of active
proliferation, such as puberty and early pregnancy and in OVX adult
mice after E+P treatment, the
5ß1 integrin was
preferentially localized to the basal surface of the cell.
Hormone-dependent basal localization of the
5ß1 integrin is
suggestive of binding to adjacent basement membrane proteins and is
consistent with integrin activation and signaling during E- or
P-mediated mitogenesis. High basal localization of the
5ß1 integrin was
reduced during late pregnancy and lactation, developmental stages
characterized by low proliferation. This leads us to hypothesize that
FN and the 5ß1
integrin play an important role in E- and E+P-dependent regulation of
epithelial cell proliferation. Because
5ß1 integrin
concentrations were more closely correlated with proliferation than FN
concentrations and because the
5ß1 integrin was more
rapidly regulated by ovarian steroids than FN, it is likely that
regulation of integrin expression represents a more dynamic mechanism
for controlling cellular proliferation and morphogenesis than the
synthesis and degradation of ECM proteins.
Keely et al. (4) examined the expression of Col
I, Col IV, and LM by in situ hybridization and
immunohistochemistry. Our results on the temporal spatial pattern of
Col I, Col IV, and LM deposition are in general agreement with the
findings of Keely et al. Both our study and theirs found
that Col I and LM were more concentrated in the epithelial basement
membrane than the connective tissue stroma, whereas Col IV was
uniformly present in the basement membrane and adipose stroma
(4). In contrast to the present study, Keely et
al. did not quantitate ECM levels. However, the changes that they
reported appear to be small compared with the changes in FN levels
observed herein. Keely et al. also examined localization of
the 2ß1 integrin of
the collagen/laminin receptor by immunohistochemistry. Staining of the
2ß1 integrin was high
in ducts between 2 and 7 weeks of age, decreased slightly in the adult,
increased at midpregnancy, and dramatically decreased during lactation.
Similar to our findings with the
5ß1 integrin, Keely
et al. demonstrated that the
2ß1 integrin is
regulated during postnatal development, pregnancy, and lactation and
that low levels of expression are observed during lactation. However,
in contrast to our current findings with
5ß1, high levels of
2ß1 were not
positively correlated with proliferation, as expression of the
2ß1 integrin was high
in quiescent ducts of 2-week-old mice, and basal localization was
highest in large ducts, which are also proliferatively quiescent
(4). Hormonal regulation of
2ß1 was not
investigated in their study. In addition to low proliferative activity
at 2 weeks of age, ovarian hormone levels are low and the mammary
epithelium is not responsive to E or P (1). Thus,
2ß1 integrin levels at
this age are not likely to be regulated by ovarian hormones. Instead,
2ß1 integrin levels
are more correlated with lobulo-alveolar formation, as its expression
was slightly higher during pregnancy. In contrast,
5ß1 expression is low
at the end of pregnancy when lobulo-alveolar development peaks, but
proliferation slows. Therefore,
5ß1 expression appears
to be most closely correlated with periods of proliferation.
Keely et al. (4) also observed that Col I, Col IV, and LM were synthesized exclusively by stromal cells in the mouse mammary gland. No synthesis was detected in epithelial cells. FN has also previously been reported to be a stroma-derived ECM protein (21). FN from plasma can be incorporated into the extracellular matrix (22). The FN antibody used in the current study detects both plasma and cellular FN. To determine whether the changes in mammary FN levels might be due to changes in plasma FN concentrations, we assayed plasma under the various hormonal conditions that decreased or increased mammary FN. We found that neither OVX nor treatment with E or E+P altered plasma FN levels, and we therefore conclude that the changes observed in mammary FN levels are most likely due to cellular FN derived from mammary stromal cells (Woodward, T. L., and S. Z. Haslam, unpublished observations). Analysis of FN levels in whole mammary gland vs. epithelium-devoid mammary stroma (CFP) revealed that FN levels were only decreased by OVX or increased by E or E+P treatment in the intact gland. Therefore, even though FN is synthesized by stromal cells, hormonal regulation of FN levels required the presence of mammary epithelium. These results suggest that a hormone-dependent paracrine interaction between epithelium and stroma is involved in the regulation of FN levels.
Ferguson et al. examined changes in ECM protein deposition
in the normal human breast during the menstrual cycle
(23). Their data show that deposition of all ECM proteins
in the basement membrane, except FN, was high during the first week of
the menstrual cycle, when E and P plasma levels and mammary
proliferation are known to be lowest (24). Their data also
show that the FN concentration was high in the basement membrane
between weeks 2–4, i.e. during the late follicular and
luteal phases when E and P levels are highest. These data suggest that
FN may also be involved in the regulation of proliferative changes in
the human breast during the luteal phase of the menstrual cycle that
are hormonally driven by E+P (24). In this context, it
would be interesting to analyze integrin expression in the human breast
during the menstrual cycle, because our results indicate changes in
5ß1 integrin levels in
mouse mammary gland are more closely correlated with proliferation and
are more acutely regulated by ovarian steroids than are FN levels.
Epithelial cells of the adult mouse mammary gland proliferate in
response to progestins in vivo and in serumcontaining
cultures in vitro (25). Progestin-induced
proliferation in serum-free cultures only occurs when cells were plated
on FN (5). Our current results also indicate that FN and
5ß1 integrin
expression are highly correlated with P action in vivo. It
has previously been demonstrated that P stimulates proliferation,
leading to ductal side-branching and lobuloalveolar development
(1). This proliferative effect of P occurs subsequent to E
up-regulation of PR. Although E is present and stimulates ductal
elongation in 3-week-old mice, E does not induce mammary epithelial PR
until approximately 7 weeks of age. In the present study the
concentrations of both FN and the
5ß1 integrin were
lowest at 3 weeks, when cells are not P responsive. The
5ß1 integrin level was
substantially increased and localized basally at 5 weeks of age,
preceding acquisition of progestin responsiveness at 7 weeks of age.
Plasma P increases rapidly at the onset of pregnancy, a period
characterized by high proliferation, and drops at the end of pregnancy,
and PR are absent during lactation (26). Late pregnancy
and lactation are characterized by low proliferation and maximal
differentiation of epithelium. The
5ß1 integrin
expression was highest in early pregnancy, decreased at late pregnancy,
and was barely detectable during lactation. Furthermore, the data
presented in the Ferguson et al. study (23)
indicate that the maximal FN concentration coincides with E+P-induced
proliferation in the normal breast during the menstrual cycle. Thus,
taken together, 1) the in vitro requirement for FN for
progestininduced proliferation, 2) the spatial and temporal
pattern of FN deposition and
5ß1 integrin
expression in vivo, and 3) the hormonal regulation of the
levels of FN and the
5ß1 integrin by E and
E+P lead us to conclude that FN signaling is highly correlated with and
may be required for P responsiveness.
It is well established that LM promotes mammary gland lactational
differentiation in vitro (9, 27). Because a
wide variety of cells cultured on LM express differentiated functions,
many researchers have used laminin or Matrigel, which is rich in LM, to
simulate the normal basement membrane when examining cell proliferation
and differentiation. However, our current studies have demonstrated
that FN is also an important component of the basement membrane, but
unlike LM, the major roles of FN and its integrin both in
vivo and in vitro appear to be in the regulation of
proliferation. It should be noted that serum preparations used for cell
culture usually contain high concentrations of FN. Thus, the presence
of FN may confound accurate interpretation of the effects of specific
ECM proteins in serum-containing cultures. We suggest that the presence
or absence of FN should also be considered in basement membrane
preparations in vitro when studying mammary epithelial cell
proliferation and differentiation, especially in serum-free culture
studies. Our present study also demonstrates that myoepithelial cells,
which are in direct contact with the basement membrane, are the major
cell type that expresses the
5ß1 integrin. This
observation suggests that myoepithelial cells may be the primary cell
type to convey ECM signals. However, the role of myoepithelial cells is
seldom considered in studies that investigate the effects of ECM in the
mammary gland. Our present results suggest that the role of
myoepithelial cells in ECM signaling should be examined.
In summary, our studies have demonstrated that the levels of FN and its
5ß1 integrin exhibit
substantial changes during development compared with other ECM
proteins. Furthermore, both FN and the
5ß1 integrin are
hormonally regulated and may play an important role in mediating
ovarian hormone-regulated growth. In this context, breast cancer, which
is also regulated by ovarian steroids, is usually accompanied by
striking changes in ECM and integrin expression (28, 29).
Understanding how ECM proteins and integrins affect proliferation,
differentiation, and apoptosis in normal mammary epithelial cells may
lead to an understanding of how alterations in ECM composition and
integrin expression affect breast cancer growth and provide the
conceptual basis for novel therapeutic strategies for the treatment of
this disease.
|
Acknowledgments |
|---|
5 integrin antibodies and assistance in
immunostaining procedures. Received October 23, 2000.
|
References |
|---|
|
TopAbstractIntroductionMaterials and MethodsResultsDiscussionReferences |
|---|
knockout mice. Endocrinology 141:2982–29942ß1 integrin and its
ligands, collagen I, collagen IV, and laminin, suggest important roles
in mouse mammary morphogenesis. Differentiation 59:1–13[CrossRef][Medline]
vß5 integrin to the
basic domain of the HIV Tat protein and vitronectin. J Cell Biol 121:461–468This article has been cited by other articles:
 |
|
 |
|
  C. M. Williams, A. J. Engler, R. D. Slone, L. L. Galante, and J. E. Schwarzbauer Fibronectin Expression Modulates Mammary Epithelial Cell Proliferation during Acinar Differentiation Cancer Res., May 1, 2008; 68(9): 3185 - 3192. [Abstract] [Full Text] [PDF]
|
|
|
|
 |
|
 O. Shynlova, S. J. Williams, H. Draper, B. G White, D. J MacPhee, and S. J Lye Uterine Stretch Regulates Temporal and Spatial Expression of Fibronectin Protein and Its Alpha 5 Integrin Receptor in Myometrium of Unilaterally Pregnant Rats Biol Reprod, November 1, 2007; 77(5): 880 - 888. [Abstract] [Full Text] [PDF]
|
|
|
|
 |
|
 C. Burrows, J. M. P. Holly, N. J. Laurence, E. G. Vernon, J. V. Carter, M. A. Clark, J. McIntosh, C. McCaig, Z. E. Winters, and C. M. Perks Insulin-Like Growth Factor Binding Protein 3 Has Opposing Actions on Malignant and Nonmalignant Breast Epithelial Cells that Are Each Reversible and Dependent upon Cholesterol-Stabilized Integrin Receptor Complexes Endocrinology, July 1, 2006; 147(7): 3484 - 3500. [Abstract] [Full Text] [PDF]
|
|
|
|
 |
|
 A. J M O'Donnell, K. G Macleod, D. J Burns, J. F Smyth, and S. P Langdon Estrogen receptor-{alpha} mediates gene expression changes and growth response in ovarian cancer cells exposed to estrogen Endocr. Relat. Cancer, December 1, 2005; 12(4): 851 - 866. [Abstract] [Full Text] [PDF]
|
|
|
|
 |
|
 D. R. Boverhof, K. C. Fertuck, L. D. Burgoon, J. E. Eckel, C. Gennings, and T. R. Zacharewski Temporal- and dose-dependent hepatic gene expression changes in immature ovariectomized mice following exposure to ethynyl estradiol Carcinogenesis, July 1, 2004; 25(7): 1277 - 1291. [Abstract] [Full Text] [PDF]
|
|
|
|
 |
|
 J.-L. Lee, C.-T. Lin, L.-L. Chueh, and C.-J. Chang Autocrine/Paracrine Secreted Frizzled-related Protein 2 Induces Cellular Resistance to Apoptosis: A POSSIBLE MECHANISM OF MAMMARY TUMORIGENESIS J. Biol. Chem., April 9, 2004; 279(15): 14602 - 14609. [Abstract] [Full Text] [PDF]
|
|
|
|
 |
|
 S. D. K. Berry, R. D. Howard, and R. M. Akers Mammary Localization and Abundance of Laminin, Fibronectin, and Collagen IV Proteins in Prepubertal Heifers J Dairy Sci, September 1, 2003; 86(9): 2864 - 2874. [Abstract] [Full Text] [PDF]
|
|
|
|
 |
|
 K. J. Turner, B. S. McIntyre, S. L. Phillips, N. J. Barlow, C. J. Bowman, and P. M. D. Foster Altered Gene Expression during Rat Wolffian Duct Development in Response to in Utero Exposure to the Antiandrogen Linuron Toxicol. Sci., July 1, 2003; 74(1): 114 - 128. [Abstract] [Full Text] [PDF]
|
|
|
|
 |
|
 U. M. Liegibel, U. Sommer, P. Tomakidi, U. Hilscher, L. van den Heuvel, R. Pirzer, J. Hillmeier, P. Nawroth, and C. Kasperk Concerted Action of Androgens and Mechanical Strain Shifts Bone Metabolism from High Turnover into an Osteoanabolic Mode J. Exp. Med., November 18, 2002; 196(10): 1387 - 1392. [Abstract] [Full Text] [PDF]
|
|
|
|
|
|
K. H. Burns, G. E. Owens, J. M. Fernandez, J. H. Nilson, and M. M. Matzuk Characterization of Integrin Expression in the Mouse Ovary Biol Reprod, September 1, 2002; 67(3): 743 - 751. [Abstract] [Full Text] [PDF]
|
|
| ||||||||||||||||||||||||||||||||||||||||||||||
