| HOME | HELP | FEEDBACK | SUBSCRIPTIONS | ARCHIVE | SEARCH | TABLE OF CONTENTS |
ARTICLES |
Department of Physiology, Michigan State University, East Lansing, Michigan 48824
Address all correspondence and requests for reprints to: Sandra Z. Haslam, Department of Physiology, Michigan State University, East Lansing, Michigan 48824.
 |
Abstract |
|---|
 Top Abstract IntroductionMaterial and MethodsResultsDiscussionReferences |
|---|
We report the novel finding that the proliferative responses of mammary epithelial cells to progestin is influenced by specific ECM molecules. However, the primary determinant of hormonal responsiveness is the developmental state of the gland from which the epithelial cells were derived. Nulliparous-derived epithelial cells, proliferated in response to R5020 only on fibronectin (FN) and collagen IV (Col IV). The more highly differentiated, pregnancy-derived epithelial cells were not responsive to E2 or R5020 on any ECM. To determine if steroid hormone receptors were targets of ECM-mediated effects, ER and PR levels were analyzed. In both nulliparous and pregnancy-derived cultures, PR binding levels were maintained at similar levels on all ECMs. However, ER levels were not maintained in nulliparous-derived cultures, and this may have contributed to the lack of a significant response to E2. Alternatively or in addition, E2-induced responses may require additional signals or growth factors that are provided by stromal cells in vivo or by serum supplementation in vitro.
These results demonstrate the ECM molecules, fibronectin and collagen IV, can modulate responsiveness of mammary epithelial cells to R5020 in vitro, and may be the mediators of stromal influences on hormone responsiveness in vivo. However, the specific effects of ECM and hormones are also determined by the developmental state of the mammary gland from which the cells are derived. Thus, mammary gland differentiation, ovarian hormones, and ECM composition may act in concert to determine the outcome of hormone treatment on cell proliferation.
|
Introduction |
|---|
|
TopAbstractIntroductionMaterial and MethodsResultsDiscussionReferences |
|---|
Although epithelial-stromal cell interactions are known to play an important role in ovarian hormone-regulated mammary cell growth both in vivo and in vitro, virtually nothing is known about the role of specific ECM molecules in this process. Thus, in the present study we have investigated the effects of several different ECM molecules (collagen I and IV, tenascin, laminin, and fibronectin) on epithelial cell attachment, growth, modulation of E2 and/or progestin (R5020) stimulation of cell proliferation and regulation of estrogen receptor (ER) and progesterone receptor (PR) binding levels. To assess the role of differentiation in interactions with ECM molecules, the responses of mammary epithelial cells from mammary glands of nulliparous (predominance of less differentiated ductal epithelium with some alveolar buds) and pregnant (highly differentiated and extensive lobuloalveolar, secretory epithelium) mice were compared.
We report the novel finding that ECM molecules can modulate responsiveness of mammary epithelial cells to the proliferative effects of R5020 in vitro and thus may be the mediators of stromal influences on hormone responsiveness in vivo. However, the primary determinant of hormonal responsiveness is the developmental state of the gland from which the epithelial cells were derived. In the case of nulliparous-derived epithelial cells, proliferative responses to R5020 could be obtained with fibronectin (FN) and collagen IV (Col IV), but not on other ECM molecules. The more highly differentiated, pregnancy-derived epithelial cells were not responsive to ovarian hormones. Thus, as lactational differentiation proceeds the attenuation of hormonal responsiveness is also accompanied by altered influence of ECM molecules.
|
Material and Methods |
|---|
|
TopAbstractIntroductionMaterial and MethodsResultsDiscussionReferences |
|---|
Cell culture
Nulliparous (10–12 weeks old) or pregnant (12–15 days) Balb/c
mice from our own colony were the source of mammary tissues. Epithelial
cells were obtained by enzymatic dissociation as previously described
(11). Briefly, mammary tissues were minced and digested with 0.1%
collagenase plus 0.01% pronase at 35 C for 2 h, and the
fibroblasts were removed by centrifuging at 80 x g for
30 sec. The epithelial cells were washed and collected after Percoll
gradient centrifugation. Cell viability was determined by exclusion of
trypan blue and viability was about 95%. On day 0, freshly dissociated
mammary epithelial cells were plated at 1.4 x 105
cells/well and cultured in 96 well flat-bottomed plates in serum- and
phenol red-free DME/F12 nutrient mixture supplemented with 0.1
mM nonessential amino acids, 0.3 mg/ml
L-glutamine, 0.1 µg/ml insulin, 100 µg/ml penicillin
and 50 µg/ml streptomycin and kept in 5% CO2 at 37 C. On
day 1 (20 h after cell plating), the cultures received media with or
without hormones, and then the media were changed every other day.
E2 was used at 23 nM and the synthetic
progestin, R5020 (promegestone, New England Nuclear Corp., Boston, MA)
at 10 nM.
Treatment of culture dishes with ECM
The coating concentrations and conditions used for ECM molecules
were based on the supplier’s instructions and our own titration for
effects on mammary epithelial cell attachment and growth in basal
media. TN, FN, LN, and Cols I and IV were tested at 0.625, 1.3, 3.25,
6.5 and 13 µg/cm2 and Poly-L-lysine (PL) was
tested at 0.13, 1.3, 13 and 130 µg/cm2. The
concentrations that provided good attachment (see Fig. 2
) and minimal
cell proliferation (see Fig. 3) in the absence of hormones or growth
factors were chosen. One day before cell plating, ECM molecules were
coated onto each culture well as follows: Col IV, Col I, FN and LN at
6.5 µg/cm2; human TN at 1.3 µg/cm2; PL at
130 µg/cm2. After 1 h incubation at room
temperature, each well was rinsed with HBSS.
|
|
[3H]-thymidine incorporation and DNA content
[3H]-thymidine incorporation was used to measure
DNA synthesis of cell cultures with or without hormone treatments.
Cells were incubated with 0.1 µCi/well (96-well plates)
[3H]-thymidine (specific activity 50 Ci/mmol, ICN
Pharmaceutical, Inc., Costa Mesa, CA). at 37 C for 6 h. This was
followed by washing with HBSS, ice-cold 90% ethanol, 10%
trichloroacetic acid, and 90% ethanol. After air-drying, the cells
were dissolved with 1 M NH4OH/NaOH (pH 12.3)
for 4–5 h and neutralized with 1 M
KH2PO4. The radioactivity was determined by
liquid scintillation counting. DNA content for each well was determined
by a fluorometric assay using Hoescht 33258 as described by West
et al. (12). Calf thymus DNA was used as a standard.
Immunocytochemical assay of ß-casein
For ease of handling, cells were cultured on ECM-coated glass
coverslips. Frozen tissue sections (6 µm) from 10-day lactating and
10-week-old nulliparous mice were used as positive and negative
controls for the presence of ß-casein, respectively. Cultured cells
or tissue sections were fixed with methanol:acetone 1:1 at -20 C for 6
min. All other steps were carried out at room temp. The coverslips were
rinsed with PBS (0.01 M pH 7.4), permeabilized with 0.5%
Triton X-100 for 90 sec and incubated in PBS 1%BSA for 1 h.
Samples were then incubated with sheep antimouse ß-casein antiserum
(1:160 dilution) (a generous gift of Dr. B. K. Vonderhaar, NCI) or
normal sheep serum for 1 h. After three washes in PBS 0.1% BSA,
the samples were incubated with secondary antibody, horseradish
peroxidase-conjugated rabbit antisheep IgG (Organon Teknika Co.,
Durham, NC) (1:1000 dilution) for 1 h. After three additional
washes in PBS, 0.1% BSA the samples were reacted with
3,3'-diaminobenzidine (0.05% in 0.003% H2O2)
for 3 min followed by a brief rinse in distilled H2O. This
step produced a dark brown reaction product in cells that contained
ß-casein.
Steroid hormone binding assay
Ligand binding assay was used to determine estrogen receptor
(ER) or progesterone receptor (PR) content as described previously (1).
Briefly, cells were cultured in 96-well plates on different ECM in
basal media for days indicated and then incubated with 10
nM [3H] estradiol (specific activity 142.0
Ci/mmol; New England Nuclear, Boston, MA) with or without 100-fold
excess unlabeled estradiol for 1 h at 37 C. To determine PR
binding levels, cells were cultured in 96-well plates on different ECM
in media with or without 23 nM E2 for the days
indicated, and then incubated with 10 nM
[3H]-R5020 (specific activity 84.8 Ci/mmol; New England
Nuclear, Boston, MA) plus 100-fold excess dexamethasone (to suppress
R5020 binding to glucocorticoid binding sites) or 100-fold excess
unlabeled R5020. The specific binding was calculated by subtracting
nonspecific binding obtained in the presence of unlabeled R5020 from
total binding obtained in the presence of unlabeled dexamethasone. DNA
content was determined as described above.
Statistical analysis
All data were expressed as the mean ± SEM and
statistical significance was determined using the Student t
test or ANOVA as appropriate.
|
Results |
|---|
|
TopAbstractIntroductionMaterial and MethodsResultsDiscussionReferences |
|---|
. To identify ECM
specific effects, cells were also cultured on PL, a nonspecific
attachment factor. In the case of PL, fewer of the organoids formed
monolayers, even by day 3. The overall morphological appearance of
pregnancy-derived cultures, except for the presence of intracellular
lipid accumulation, was very similar to nulliparous cultures.
|
). In nulliparous-derived cultures, the highest
attachment was obtained with Col I and FN (50–55%) followed by Col
IV, LN, and TN (40–45%). The lowest attachment obtained was on PL
(34%). In pregnancy-derived cultures, a similar pattern of attachment
efficiency was observed.
Cell proliferation on different ECM in basal media
To assess the contribution of ECM molecules themselves to cell
proliferation, DNA synthesis was determined in basal media, without
added hormones. In nulliparous-derived cultures (Fig. 3), DNA synthesis was similar on all ECMs and PL,
indicating that the various ECMs did not exhibit significant growth
stimulating activity. Identical results were obtained with
pregnancy-derived cultures (data not shown).
Effects of steroid hormones on cell proliferation in relation to
ECM composition
To determine the peak time of proliferation, DNA synthesis was
measured for nulliparous and pregnancy-derived cells on all ECM
molecules at 24, 48, and 72 h after addition of hormones. For time
course studies, epidermal growth factor (EGF) was included as a
positive control to demonstrate the capacity of the cells to
proliferate under the serum-free culture conditions that were used for
these studies. Figure 4A shows that nulliparous-derived
cells cultured on FN exhibit maximal proliferation in response to both
R5020 and EGF at 48 h after treatment. Increased proliferation was
also observed with E2 treatment. The small stimulatory
effect of E2 seen at 24 h was not significantly
different from the control, R5020 or E2 + R5020 treatment
groups. Maximal proliferation in response to EGF also occurred at
48 h after treatment for pregnancy-derived cultures on FN (Fig. 4B) and on all other ECMs for both nulliparous and pregnancy-derived
cultures (data not shown). Figure 5 shows the effect of
hormone treatment on DNA synthesis when nulliparous-derived cells were
cultured on different ECMs. R5020 increased DNA synthesis by 50–60%
on FN and Col IV. The increased DNA synthesis observed with R5020 on FN
was significantly greater that that obtained with E2 or
E2 + R5020. E2 also increased DNA synthesis on
FN, although the increase was not statistically significant across
experiments. In several individual experiments, the stimulation was
significant as shown above (Fig. 4A), but a high degree of variability
between experiments obscured the statistical significance of
stimulatory effects when the values were averaged.
|
|
). Neither E2 nor
R5020 was able to stimulate cell proliferation on any ECM
molecules.
|
and 6, PL did
not promote cell proliferation in nulliparous- or pregnancy-derived
cultures with added hormones.
ECM molecules can interact with cells via binding to specific receptor
molecules called integrins. Many integrins recognize the peptide
sequence Arg-Gly-Asp (RGD) in their ligands (FN, TN, and LN) and the
RGD tripeptide by itself can in some instances initiate
integrin-mediated cellular effects (13). Thus it was of interest to
determine if integrin binding via RGD was involved in steroid hormone
responsiveness that was observed with certain ECMs. Nulliparous-derived
cells were plated on PL, a nonspecific attachment factor that did not
promote steroid hormone responsiveness but promoted attachment under
serum-free conditions. The cultures were then treated with various
concentrations of RGD with or without R5020. The RGE (Arg-Gly-Glu)
peptide that does not bind to integrins served as a control. Figure 7 shows that RGD did not mediate R5020 effects on cells.
Higher concentrations RGD and RGE had negative effect on cell growth
(data not shown). The effect of RGD with and without R5020 was also
tested on Col I, and no increase in DNA synthesis was observed (data
not shown). To assure adequate access of the RGD peptide to the
integrin binding sites on the cell surface, RGD was also added to
freshly dissociated cells, before plating. Under this condition RGD did
not inhibit cell attachment or increase DNA synthesis on either PL or
Col I (data not shown).
|
). In contrast, ER levels were maintained on all ECMs
in pregnancy-derived cultures.
|
. In freshly dissociated cells, PR
levels were 13.4 ± 2.5 and 5.3 ± 1.0 in nulliparous- and
pregnancy-derived cells, respectively. PR levels in both pregnancy- and
nulliparous-derived cells increased with time after plating, regardless
of ECM composition. However, no increases in PR levels were observed
after treatment with E2 on any ECM in either nulliparous-
or pregnancy-derived cultures (Fig. 9).
|
, a very low number, only about 5–10%, of pregnancy-derived cells
cultured on FN, TN, LN, or Col IV were positive for casein and there
were no significant differences among the various ECMs. Thus, loss of
steroid hormone responsiveness did not appear to be correlated with the
extensive induction of a lactational phenotype.
|
|
Discussion |
|---|
|
TopAbstractIntroductionMaterial and MethodsResultsDiscussionReferences |
|---|
In the case of the less differentiated, nulliparous-derived epithelium, we have found that FN and Col IV promoted a proliferative response to R5020, with the greatest stimulation observed on them at 48 h after hormone treatment. The fact that stromal cell-derived ECM molecules (FN and Col IV) promoted hormone responsiveness supports our previous findings both in vivo (15) and in vitro (1, 16) that mammary stromal cells can confer steroid hormone responsiveness on mammary epithelium. These results indicate that one underlying mechanism of the modulation of steroid hormone responsiveness in vivo and in vitro by mammary stroma may be mediated by ECM molecules.
It should be noted that in previous reports using ECMs such as Col I
(17) or matrigel (18) for investigation of rodent mammary cells and
steroid hormone regulation, the cells were kept in culture from 10 days
to 4 weeks. According to Streuli and Bissell (19), mammary epithelial
cells can begin to synthesize basement membrane or produce ECM
components after 2 days in culture, depending on the culture
substratum. Thus, from previous experiments, it is difficult to
determine the actual source (exogenous vs. endogenous) of
the ECM molecules that influenced the behavior of the cultured cells.
In the present experiments, cells were cultured for only 3 or 4 days
and R5020 stimulation was observed as soon as 24 h after treatment
(Fig. 4). Thus, the stimulation of cell proliferation with R5020
observed herein was most likely mediated by the exogenously supplied FN
or Col IV.
Integrin binding to ECM can initiate cellular responses and the
tripeptide RGD sequence by itself can in some instances initiate
integrin-mediated effects for FN, LN, or TN (13). While R5020
stimulated proliferation of cells cultured on FN, RGD plus R5020 failed
to reproduce this proliferative effect. One explanation for this is
that the integrins that bind to RGD may not be involved in mediating
FN’s effect on the progestin response. While the RGD sequence has been
identified as a key adhesive motif, larger fragments of FN possess
synergistic regions that interact with RGD to influence cell functions
(20). Thus other regions of the FN molecule may be involved in
ECM-mediated progestin responses. It should also be noted that because
the maximum purity of currently available ECM molecules is 95–99% we
cannot completely rule out the possibility that some unidentified
contaminating molecule(s) contributed to the progestin responsiveness
that we observed. For example, previous studies using the ER positive,
E2 responsive MCF-7 human breast carcinoma cell line have
demonstrated that estrogenic effects such as regulation of PR levels in
serum-free medium required the presence of a growth factor such as
IGF-I (21). However, the low amount of proliferation that we observed
in basal media on all ECMs tested (Fig. 3) indicates that there was not
significant contamination of the ECMs with growth factors. Furthermore,
initial studies on the combined effects of growth factors and steroid
hormones on various ECMs in serum-free culture have not shown any
positive interactions between either EGF or IGF-I and E2 or
R5020 (Xie, J., and S. Z. Haslam, unpublished observations)
During puberty, between 3 and 6 weeks of age, ductal elongation is stimulated by estrogen. However, at this stage of development PR levels cannot be increased by estrogen. But by 7–10 weeks of age, after the ducts have grown to fill the limits of the stromal fat pad, both cell proliferation and PR levels are increased by estrogen. However, if the immature, pubertal mammary epithelium is transplanted to a mature mammary stroma, the full adult response to estrogen, including PR up-regulation, is precociously acquired (15). In the present studies, FN and Col IV mediated R5020 stimulation of epithelial cells, supporting our hypothesis that certain ECM produced by stromal cells may modulate steroid hormone responsiveness. Interestingly, culturing cells on all ECMs and PL increased PR levels during culture. However, only nulliparous cells cultured on FN and Col IV proliferated in response to progestin. These results suggest that the presence of PR is not sufficient to obtain a progestin response.
In general, estrogen alone had little or no stimulatory effect on cell proliferation or PR regulation. A stimulatory effect on cell proliferation could be obtained with E2 on FN in nulliparous-derived cultures. However, this response was not consistently observed. Furthermore, the synergistic effect of E2 plus progestin on cell proliferation that is observed in vivo (3) was not observed in culture. We considered the possibility that carryover of endogenous in vivo hormones might reduce the response to exogenous hormones in culture. However, ovariectomizing the mice to reduce endogenous levels of hormones before culture, or preculturing in the absence of hormones did not increase cell proliferation in response to E2 or R5020 (unpublished observations, Xie, J., and S. Z. Haslam). It is likely that a major contributing factor to the absence of E2-induced effects was the low levels of ER that were present after plating. Another explanation is that additional factors are needed to obtain a proliferative response to E2. Astrahantseff and Morris (22) have demonstrated that E2 stimulates proliferation of uterine cells only when they are cocultured with stromal cells. Furthermore, stromal cells need to be in direct contact with or reside very short distances from the epithelial cells. We have previously reported similar results with mammary cells cocultured with mammary fibroblasts on plastic in the presence of serum (16). Thus, E2-induced responses may require additional signals or growth factors that may be produced by stromal cells in vivo or that are provided by serum supplementation in vitro. In contrast, the proliferative effect of progestin may be mediated by ECM molecules (e.g. FN and Col IV) produced by stromal cells and does not require the presence of the stromal cells. Studies are in progress to investigate the interaction of steroid hormones, growth factors and ECM molecules in control of mammary epithelial cell proliferation under serum-free culture conditions.
We have previously shown that in vivo, as the mammary gland becomes progressively differentiated toward lactation, ovarian steroid hormone responsiveness is lost (14). This reduced responsiveness appears to be paralleled in vitro. We analyzed the pregnancy-derived cultures on various ECMs, for the production of the milk specific protein, ß-casein, but only small amounts were detectable. This was not surprising because the culture media were lacking in the lactogenic hormones, PRL and hydrocortisone. However, it appears that the extent of differentiation induced in vivo and exhibited by the pregnancy-derived cells in vitro, although it fell short of complete lactational differentiation, may have been sufficient to produce refractoriness to hormone responsiveness. Our studies also indicate that as lobulalveolar differentiation progresses, the influence of other ECM molecules on proliferative response to ovarian hormone responsiveness becomes attenuated. Hormonal regulation of lactational function in cell culture is strongly influenced by ECM composition, and laminin has been shown to play an important role in increasing casein gene expression (10). In this context, it is interesting to note that in the present studies, laminin had no significant effect on the proliferative response of nulli-parous- or pregnancy-derived cultures.
In summary, we have found that two stromal cell-derived ECM molecules, FN and Col IV, can specifically influence mammary epithelial cell proliferative response to progestin. However, the extent of the influence of ECM and hormones is predetermined by the developmental state of the mammary gland. These results support our previous observations that mammary stroma can modulate mammary epithelial cell hormone responsiveness in vivo and indicate that ECM molecules may mediate some stromal cell influences.
The loss of hormone responsiveness is a critical factor in the treatment of breast cancer. Advancing our understanding of the various mechanisms by which modulations in steroid hormone responsiveness is achieved in normal mammary epithelium may broaden the conceptual framework for understanding loss of hormone responsiveness in breast cancer. Identifying the role of mammary stroma in this process may lead to novel therapeutic strategies that focus on mammary stroma as a target.
|
Footnotes |
|---|
Received December 26, 1996.
|
References |
|---|
|
TopAbstractIntroductionMaterial and MethodsResultsDiscussionReferences |
|---|
50 ß 49 integrin and its ligands,
collagen I, collagen IV, and laminin, suggest important roles in mouse
mammary morphogenesis. Differentiation 59:1–13[CrossRef][Medline]
This article has been cited by other articles:
 |
|
 |
|
  K. Roarty and R. Serra Wnt5a is required for proper mammary gland development and TGF-{beta}-mediated inhibition of ductal growth Development, November 1, 2007; 134(21): 3929 - 3939. [Abstract] [Full Text] [PDF]
|
|
|
|
 |
|
 V. Novaro, C. D. Roskelley, and M. J. Bissell Collagen-IV and laminin-1 regulate estrogen receptor {alpha} expression and function in mouse mammary epithelial cells J. Cell Sci., July 15, 2003; 116(14): 2975 - 2986. [Abstract] [Full Text] [PDF]
|
|
|
|
 |
|
 H.-Z. Zhang, J. M. Bennett, K. T. Smith, N. Sunil, and S. Z. Haslam Estrogen Mediates Mammary Epithelial Cell Proliferation in Serum-Free Culture Indirectly via Mammary Stroma-Derived Hepatocyte Growth Factor Endocrinology, September 1, 2002; 143(9): 3427 - 3434. [Abstract] [Full Text] [PDF]
|
|
|
|
|
|
N. Sunil, J. M. Bennett, and S. Z. Haslam Hepatocyte Growth Factor Is Required for Progestin-Induced Epithelial Cell Proliferation and Alveolar-Like Morphogenesis in Serum-Free Culture of Normal Mammary Epithelial Cells Endocrinology, August 1, 2002; 143(8): 2953 - 2960. [Abstract] [Full Text] [PDF]
|
|
|
|
|
|
T. L. Woodward, A. S. Mienaltowski, R. R. Modi, J. M. Bennett, and S. Z. Haslam Fibronectin and the {{alpha}}5{beta}1 Integrin Are Under Developmental and Ovarian Steroid Regulation in the Normal Mouse Mammary Gland Endocrinology, July 1, 2001; 142(7): 3214 - 3222. [Abstract] [Full Text] [PDF]
|
|
|
|
|
|
T. L. Woodward, J. Xie, J. L. Fendrick, and S. Z. Haslam Proliferation of Mouse Mammary Epithelial Cells in Vitro: Interactions among Epidermal Growth Factor, Insulin-Like Growth Factor I, Ovarian Hormones, and Extracellular Matrix Proteins Endocrinology, October 1, 2000; 141(10): 3578 - 3586. [Abstract] [Full Text] [PDF]
|
|
|
|
|
|
T. L. Woodward, H. Lu, and S. Z. Haslam Laminin Inhibits Estrogen Action in Human Breast Cancer Cells Endocrinology, August 1, 2000; 141(8): 2814 - 2821. [Abstract] [Full Text] [PDF]
|
|
|
|
|
|
D. L. Buchanan, T. Kurita, J. A. Taylor, D. B. Lubahn, G. R. Cunha, and P. S. Cooke Role of Stromal and Epithelial Estrogen Receptors in Vaginal Epithelial Proliferation, Stratification, and Cornification Endocrinology, October 1, 1998; 139(10): 4345 - 4352. [Abstract] [Full Text] [PDF]
|
|
| |||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||
| HOME | HELP | FEEDBACK | SUBSCRIPTIONS | ARCHIVE | SEARCH | TABLE OF CONTENTS |
| Endocrinology | Endocrine Reviews | J. Clin. End. & Metab. |
| Molecular Endocrinology | Recent Prog. Horm. Res. | All Endocrine Journals |
) and pregnancy-derived (
) cells were plated
at 1.4 x 105cells/well in 96-well plates on different
ECM molecules as indicated. The percent cell attachment on each ECM for
both pregnant and nulliparous cells was calculated by dividing the
numbers of cells attached after 17 h by the numbers of cells
plated x 100. Cell numbers were determined by MTT assay. Each
value represents the M ± SEM. of triplicate values
from three to four separate experiments. *, P <
0.01 that the numbers of nulliparous-derived cells attached on Col I
and FN were greater than on other ECMs and that numbers of
pregnancy-derived cells attached on Col I and FN were greater than on
other ECMs except TN.
), R5020 (10
nM, 
) or EGF (5 ng/ml 
). [3H]-thymidine
incorporation into DNA was determined at indicated times as described
in Materials and Methods. Each value represents the
M ± SEM of five replicates. *, P
< 0.05–0.005 that hormone- and growth factor-treated groups are
greater than the hormone-free control group.
