| HOME | HELP | FEEDBACK | SUBSCRIPTIONS | ARCHIVE | SEARCH | TABLE OF CONTENTS |
ARTICLES |
Department of Veterinary Biosciences (D.L.B., P.S.C.), University of Illinois, Urbana, Illinois 61802; Department of Anatomy (T.K., G.R.C.), University of California, San Francisco, California 94143; and Departments of Biochemistry and Child Health (J.A.T., D.B.L.), University of Missouri, Columbia, Missouri 65211
Address all correspondence and requests for reprints to: Paul Cooke, Department of Veterinary Biosciences, University of Illinois, 2001 South Lincoln Avenue, Urbana, Illinois 61802.
 |
Abstract |
|---|
 Top Abstract IntroductionMaterials and MethodsResultsDiscussionReferences |
|---|
(ER) in these E2-induced events. Vaginal
epithelium (E) and stroma (S) from adult ER knockout (ko) and
wild-type (wt) neonatal Balb/c mice were enzymatically separated and
used to produce four types of tissue recombinants in which epithelium,
stroma, or both lack functional ER. Tissue recombinants were grafted
into female nude mice, which were subsequently ovariectomized and
treated with oil or E2. In response to E2
treatment, grafts prepared with wt-S (wt-S + wt-E and wt-S + ko-E)
showed similar large increases in epithelial labeling index, indicating
that E2 stimulated epithelial proliferation despite a lack
of epithelial ER in wt-S + ko-E tissue recombinants. Conversely, in
tissue recombinants prepared with ko-S (ko-S + wt-E and ko-S + ko-E),
epithelial labeling index remained at baseline levels after
E2 or oil treatment, even though epithelial ER were
detected in ko-S + wt-E grafts. Epithelial cornification was present in
wt-S + wt-E grafts from E2-treated hosts, whereas
epithelium in all other tissue recombinants failed to cornify. Grafts
composed of wt-S + wt-E from E2-treated hosts had highly
stratified epithelium, whereas epithelial thickness was reduced almost
60% in wt-S + ko-E tissue recombinants grown in E2-treated
hosts and was atrophic in all other tissue recombinants. In addition,
cytokeratin 10, a marker of epithelial differentiation, was strongly
expressed in wt-S + wt-E tissue recombinants grown in
E2-treated hosts but was markedly reduced or absent in all
other tissue recombinants. These results indicate that
E2-induced vaginal epithelial proliferation is mediated
indirectly through stromal ER, consistent with our recent findings
in uterus. Conversely, both epithelial and stromal ER are required
for E2-induced cornification and normal epithelial
stratification. These are the first known functions attributed to
epithelial ER in vivo and the first time any
epithelial response to E2 has been shown to involve both
stromal and epithelial ER. |
Introduction |
|---|
|
TopAbstractIntroductionMaterials and MethodsResultsDiscussionReferences |
|---|
(ER), as
shown by the lack of vaginal epithelial proliferation, stratification,
and cornification in E2-treated ER knockout (ERKO)
mice (2). ER is normally present in both vaginal epithelial and stromal
cells (3). It has been widely assumed that E2-induced
epithelial proliferation and differentiation in female reproductive
organs are mediated directly through the epithelial ER (4, 5). In
contrast, other evidence suggests that vaginal and uterine stromal ER
may play an important role in the E2-induced epithelial
response. Indeed, we have shown previously that uterine epithelium
isolated from the ERKO mouse proliferates in response to E2
when associated with an ER-positive uterine stroma (6). It has not
been established, however, whether epithelial proliferation is mediated
indirectly by stromal ER in other E2-responsive tissues
of the female reproductive tract, such as the vagina.
In addition to stimulating vaginal epithelial proliferation,
E2 treatment also elicits a complex pattern of
differentiative events in vaginal epithelium. The atrophic vaginal
epithelium of the ovariectomized mouse is only 2–3 cell layers thick.
In response to E2, basal epithelial cells proliferate
rapidly, leading to the formation of a highly stratified epithelium.
The suprabasal cells, which are no longer mitotic, undergo a
well-characterized differentiative sequence as they move up through the
epithelium; they become enlarged and undergo structural and
morphological changes indicative of cornification, so that the apical
layer becomes heavily keratinized. These morphological changes, in
response to E2, are accompanied by the production of
cytokeratins 1 and 10, markers common to epidermal and vaginal
epithelial differentiation (7, 8, 9). Although ER is essential for
normal vaginal stratification and cornification (2), the respective
roles of epithelial vs. stromal ER are unknown in these
proliferative and differentiative responses to E2.
The aim of the present study was to determine the respective roles of
epithelial vs. stromal ER in the response of vaginal
epithelium to E2. To address this question, ERKO and
wild-type (wt) mice were used to prepare vaginal tissue recombinants
that lacked functional ER in epithelium, stroma, or both. This study
examined the roles of epithelial vs. stromal ER in
E2-induced vaginal epithelial proliferation to determine
whether the indirect mediation of epithelial mitogenesis by stromal
ER, previously seen in uterus, may be common to other female
reproductive organs. We also determined the role of stromal and
epithelial ER in E2-induced differentiative responses,
such as vaginal epithelial stratification and cornification. Our
results indicate that E2-induced vaginal epithelial
proliferation is mediated indirectly through stromal ER. Conversely,
E2-induced vaginal epithelial cornification and normal
stratification require both epithelial and stromal ER.
|
Materials and Methods |
|---|
|
TopAbstractIntroductionMaterials and MethodsResultsDiscussionReferences |
|---|
Tissue recombination
Vaginae were removed from adult (90- to 120-day-old) ERKO and
neonatal (0- to 3-day-old) Balb/c mice that had been killed by
CO2 asphyxiation or decapitation, respectively. The present
experiments were performed in parallel with a previously published
study using uteri from ERKO and Balb/c mice (6). To obtain pure uterine
luminal epithelium from Balb/c mice for that study, uteri were removed
before the start of uterine gland formation, at approximately 5 days of
age. Because uterine gland formation is rudimentary in ERKO mice, adult
ERKO mice were used, to obtain greater amounts of tissue. Therefore,
due to availability, neonatal Balb/c vaginae and adult ERKO vaginae
were used in the present experiments. Previous work has shown that
vaginal tissue recombinants prepared from stromal and epithelial
components derived from different age mice show normal growth,
development, and hormonal responsiveness, after grafting under the
renal capsule (11, 12).
The homozygous ERKO genotype, initially established by multiplex PCR, was verified at the time of tissue removal by confirming the presence of hypoplastic uteri and hyperemic ovaries (10). The tissue separation and recombination procedure for vaginal epithelium and stroma has been described (11, 13). Briefly, vaginae from Balb/c and ERKO mice were trimmed, opened lengthwise, and incubated with 1% trypsin (Life Technologies, Grand Island, NY) in a calcium- and magnesium-free HBSS for 90 min at 4 C. Vaginal stroma and epithelium were then separated by removing the epithelium from the underlying stroma using a von Graefe knife and fine forceps. Either tissue fraction obtained by this method is devoid of contamination with the other (14). Vaginal tissue recombinations were prepared and cultured overnight on 1% agar medium (Difco, Detroit, MI), as described previously (6). The following four tissue recombinants were prepared with ERKO (ko) and wt Balb/c vaginal stroma (S) and epithelium (E): wt-S + wt-E, wt-S + ko-E, ko-S + wt-E, and ko-S + ko-E. Tissue recombinants were transplanted under the renal capsules of intact, adult female athymic (nude) mice (Harlan), as described previously (11). The grafts were allowed to grow 4 weeks, then all hosts were ovariectomized.
Hormone treatments, autoradiography, and tissue staining
One week after ovariectomy, hosts were given oil or different
regimens of hormone treatments. Some host animals used to determine the
mechanism of E2-induced epithelial mitogenesis were
injected ip with 100 ng E2 (Sigma Chemical Co., St. Louis,
MO) in 0.05 ml corn oil, whereas other host animals received oil
vehicle as a control. To determine whether stimulation of epithelial
proliferation seen in wt-S + wt-E and wt-S + ko-E tissue recombinants
in response to E2 was mediated through ER, some hosts were
given daily sc injections of 1 mg/kg of the antiestrogen ICI 182,780
(Zeneca Pharmaceuticals, Cheshire, UK) in oil on days 5–7 post
ovariectomy, and then they were given E2 on day 7 post
ovariectomy, in addition to the antiestrogen. To assess vaginal
epithelial proliferation, some hosts were injected with
[3H]-thymidine (specific activity = 80 Ci/mmol; 1
Ci = 37 GBq; Amersham, Arlington Heights, IL) at a dose of 2
µCi/g BW, 16 h after E2 or oil treatments. Two hours
later (18 h after E2 or oil injection), hosts were killed,
and grafts were harvested. Tissue recombinants were fixed in 10%
neutral buffered formalin (Sigma) for 12 h at 4 C,
paraffin-embedded, and then sectioned at 6 µm. For
[3H]-thymidine autoradiography, tissue sections were
deparaffinized, dried, dipped in NTB-2 nuclear emulsion (Kodak,
Rochester, NY), and stored at 4 C for 3–4 weeks. Slides were then
developed, using Kodak processing chemicals, and stained with
hematoxylin and eosin for labeling index (L.I.) determination.
To determine the mechanism by which E2 elicits epithelial stratification and cornification in vaginal tissue recombinants, host animals received one injection per day of E2 (100 ng) or oil vehicle alone (control) over days 7–9 post ovariectomy (72 h total treatment); tissue recombinants were harvested 24 h after the final injection. The tissue recombinants were fixed, embedded, and sectioned as above, and then stained with hematoxylin and eosin.
ER and cytokeratin 10 immunohistochemistry
ER was detected by immunohistochemistry in mouse vaginal
tissue recombinants using a protocol similar to that described
previously for uterus (6), with some modification. Briefly, tissue
sections were deparaffinized, immersed in antigen retrieval solution
(Vector Laboratories, Burlingame, CA) diluted 1/100, and microwaved on
high power for 16 min. Avidin- and biotin-blocking solutions (DAKO
Corp., Carpinteria, CA) were used, and nonspecific binding was blocked
using Super Block (Pierce, Rockford, IL). Sections were incubated at
room temperature for 2 h with either the primary monoclonal
antibody (NCL-ER-LH2, Novocastra, Burlingame, CA) diluted 1/50 in
SuperBlock (Pierce) or a control nonspecific IgG (DAKO) and rinsed in
PBS (Dulbecco’s PBS, Sigma). After inactivation of endogenous
peroxidases with Peroxidase Suppressor (Pierce), slides were incubated
with a secondary biotinylated antimouse antibody (DAKO, LSAB2 kit).
Finally, streptavidin, conjugated to horseradish peroxidase (DAKO,
LSAB2 kit), was applied; and tissue sections were rinsed in PBS.
Vaginal ER was visualized by a 5-min incubation in 0.5%
diaminobenzidine (DAKO) and 0.01% H2O2 in PBS.
Slides were counterstained with hematoxylin.
Cytokeratin 10 is an intermediate filament protein expressed in suprabasal layers of cornifying stratified epithelia (7, 15). It is also found in a variable number of suprabasal cells in normally stratified epithelia that are noncornifying (7). Because the expression of this cytokeratin is limited to these types of epithelia, it serves as a marker of E2-induced differentiation in vaginal epithelia. For cytokeratin 10 immunohistochemistry in vaginal tissue recombinants, tissue sections were deparaffinized and subjected to antigen-retrieval treatment using the microwave method as above. Endogenous peroxidases were inactivated by 0.03% hydrogen peroxide in methanol, and nonspecific binding was blocked using 1% BSA in Tris-HCl buffer saline (pH 7.6). Sections were incubated with the monoclonal horseradish peroxidase-labeled cytokeratin 10 antibody (DAKO) or a control nonspecific IgG (DAKO) and were rinsed in PBS. Immunoreactivity was developed using diaminobenzidine (Sigma) as the chromogen. Slides were counterstained using hematoxylin. The immunostaining procedures were repeated three times on different tissue recombinants from each group, derived from three to four separate experiments.
Image and data analysis
All images were captured using an Olympus Vanox Photomicroscope
with planapochromatic lenses and a Sony Digital Photo Camera
DKC-5000 (Sony Corp., Tokyo, Japan) interfaced to a Macintosh
computer using Adobe Photoshop software. Epithelial L.I. in various
tissue recombinants was measured as
[3H]-thymidine-labeled cells per total basal cells, as
described previously (16). For each group, 1000–3000 cells were
counted. Morphometric analysis of epithelial stratification was
performed on a Macintosh computer using the public domain National
Institutes of Health Image program (version 1.6, NIH Shareware)
to measure epithelial thickness at 6 random points in each replicate of
each tissue recombinant type. Epithelial height was defined as the
distance from the basal lamina to the apical surface. Data on
epithelial proliferation and height were analyzed by one-way ANOVA, and
the significance between two groups was determined using Orthogonal
contrasts; differences between means were considered significant at
P < 0.05. Each data point for epithelial proliferation
and height is based on at least 12 tissue recombinants for each group,
derived from three to four separate experiments.
|
Results |
|---|
|
TopAbstractIntroductionMaterials and MethodsResultsDiscussionReferences |
|---|
). Epithelial proliferative activity was
similar in wt-S + wt-E and wt-S + ko-E tissue recombinants from
E2-treated hosts, despite a lack of epithelial ER in the
latter. In contrast, epithelium in wt-S + wt-E and wt-S + ko-E tissue
recombinants from oil-treated hosts showed only minimal proliferative
activity. Epithelial L.I. in tissue recombinants prepared with ko-S
(ko-S+ wt-E and ko-S + ko-E) was uniformly low and was not stimulated
by E2 treatment (Fig. 1). After administration of the
antiestrogen ICI 182,780 plus E2, vaginal epithelial L.I.
was low and similar to oil-treated controls in wt-S + wt-E and wt-S +
ko-E grafts (Fig. 1).
|
expression in
epithelium and stroma of the various tissue recombinants (Fig. 2) and demonstrated the efficacy of the
tissue separation technique. Tissue recombinants composed of wt-S +
wt-E showed nuclear staining in both epithelial and stromal cells, thus
indicating the presence of ER in both tissues. In wt-S + ko-E tissue
recombinants, nuclear ER was detected in stroma but not in
epithelium. Conversely, ER was detected by immunohistochemistry in
epithelium but not stroma in ko-S + wt-E tissue recombinants and was
undetectable in both epithelium and stroma in ko-S + ko-E tissue
recombinants. Substitution of nonimmune IgG for the primary antibody
resulted in an absence of staining (data not shown) in all cases.
|
). Morphometric analysis
demonstrated that mean epithelial thickness in wt-S + wt-E tissue
recombinants from E2-treated hosts was approximately 60
µm, compared with about 11 µm in wt-S + wt-E tissue recombinants
derived from oil-treated hosts, which corresponded to an 81.2%
reduction in epithelial thickness. Mean epithelial thickness in wt-S +
ko-E tissue recombinants from E2-treated hosts was
significantly decreased, by 57.4%, compared with wt-S + wt-E tissue
recombinants recovered from E2-treated hosts, though this
epithelium was approximately 3-fold thicker than that of wt-S + ko-E
tissue recombinants from oil-treated hosts. Mean epithelial thicknesses
was similar in ko-S + wt-E and ko-S + ko-E grafts from both
E2- and oil-treated hosts. These epithelia ranged from
4.8–9.5 µm (Fig. 4), corresponding to
approximately 85–90% less than that of wt-S + wt-E tissue
recombinants from E2-treated hosts. In all cases,
epithelial height in these grafts was not significantly different from
that in wt-S + wt-E tissue recombinants from oil-treated hosts.
Consistent with the morphometric results, immunohistochemical staining
for cytokeratin 10 was very intense in the suprabasal epithelium of
wt-S + wt-E tissue recombinants from E2-treated hosts (Fig. 5). In contrast, the epithelium in wt-S +
ko-E tissue recombinants from E2-treated hosts showed only
weak, patchy staining for cytokeratin 10. Immunohistochemical staining
for cytokeratin 10 was minimal or absent in tissue recombinants
prepared with ko-S (ko-S + wt-E and ko-S + ko-E) or in tissue
recombinants from oil-treated hosts.
|
|
|
|
Discussion |
|---|
|
TopAbstractIntroductionMaterials and MethodsResultsDiscussionReferences |
|---|
in E2-induced vaginal
epithelial responses in vivo using vaginal tissue
recombinants that contain ER in either stroma, epithelium, or both.
The present results indicate that E2 produces similar
increases in epithelial L.I. in wt-S + wt-E and wt-S + ko-E tissue
recombinants, despite the lack of epithelial ER in the latter. This
suggests that E2 induces epithelial proliferation
indirectly through stromal ER. Corroborating this interpretation is
the finding that E2 does not stimulate epithelial
proliferation in koS + wtE tissue recombinants, even though ER is
expressed in the epithelium. Therefore, epithelial ER are neither
necessary nor sufficient for the vaginal epithelial mitogenic response
to E2. The apparent indirect mediation of
E2-induced epithelial mitogenesis is consistent with
earlier reports that E2 does not stimulate proliferation or
differentiation in isolated vaginal epithelium in culture (18, 19).
However, when cultured vaginal epithelium is reassociated with vaginal
stroma and grafted in vivo, the epithelial mitogenic
response to E2 is reestablished (13). The present results,
in conjunction with our recent findings that E2-induced
epithelial mitogenesis in uterine (6) and mammary (20) tissue
recombinants is mediated indirectly via stromal ER, suggest that
indirect ER mediation of E2-induced epithelial
proliferation may be universal in female reproductive tissues.
Further verification that the proliferative response to E2
was mediated through stromal ER was obtained by use of the potent
antiestrogen, ICI 182,780, which competes with E2 for ER
binding (21). The ability of ICI 182,780 to block the proliferative
effects of E2 on vaginal epithelial proliferation in wt-S +
ko-E tissue recombinants indicates that the stimulation of epithelial
proliferation in tissue recombinants in response to E2 was
mediated through stromal ER and not through some other receptor or some
nonreceptor-mediated pathway.
Both uterine and vaginal epithelial cells differentiate in response to
E2 stimulation. In the absence of E2, vaginal
epithelium is atrophic, and it consists of 2–3 layers of squamous
epithelial cells. E2 treatment in vivo induces
vaginal epithelium to proliferate, to thicken, and to produce
flattened, cornified cells that accumulate in the superficial layers.
In contrast, uterine epithelium normally remains simple columnar, with
or without E2 stimulation, but it is induced by
E2 to produce secretory proteins such as lactoferrin. Thus,
in contrast to the simple columnar uterine epithelium, the
differentiation response of vaginal epithelium is more complex and
involves the generation and differentiation of multiple suprabasal cell
layers. Isolated vaginal epithelium in vitro fails to
stratify or cornify in response to E2 (19); but when this
epithelium is reassociated with vaginal stroma and grown in
vivo, stratification and cornification, in response to
E2 treatment, is reestablished (13). Furthermore, isolated
vaginal epithelium will not grow and develop normally in
vivo when grafted by itself under the renal capsule (Cooke and
Cunha, unpublished observations). Such results suggest the importance
of stromal-epithelial interactions in vaginal epithelial
differentiation and imply that vaginal epithelium is dependent on
stromal regulatory signals to express E2- induced
cornification and stratification. However, it has not been clear
whether stroma simply plays a permissive role in E2-induced
cornification and stratification or whether direct E2
action on stroma was necessary to allow the stroma to support the
typical epithelial changes seen in these responses. Similarly, it has
not been clear whether epithelial ER plays a role in the normal
progression of epithelial differentiative events in response to
E2. Therefore, an additional aim of the present study was
to determine the respective roles of stromal and epithelial ER in
ER-dependent vaginal epithelial stratification and
cornification.
The complete vaginal response to E2 that includes
epithelial proliferation, stratification, cornification, and
cytokeratin 10 expression only occurred in wt-S + wt-E tissue
recombinants, in which ER is expressed in both epithelium and
stroma. Conversely, epithelium in wt-S + ko-E and ko-S + wt-E tissue
recombinants from E2-treated hosts did not stratify
normally or express normal amounts of cytokeratin 10, indicating that
neither stromal nor epithelial ER is individually capable of
mediating E2-induced vaginal epithelial cornification or
normal stratification. Thus, the full epithelial differentiative
response requires both stromal and epithelial actions of
E2; this is the first time any epithelial response to
E2 in vivo has been shown to require both
stromal and epithelial ER. The sequence of events that culminate in
vaginal epithelial differentiation seem to first require epithelial
proliferation, which most likely is elicited by paracrine mechanisms
via stromal ER. The production of multiple suprabasal epithelial
layers may then provide the tissue organization required for direct
E2 action on the epithelium mediated by epithelial ER.
The direct effect of E2 may secondarily elicit terminal
differentiation, which involves the expression of differentiation-type
cytokeratins 1 and 10, as well as other proteins, such as involucrin
and loricrin (7, 8, 9). The result is the formation of a stratified
epithelium similar to fully mature epidermis. If proliferation does not
occur, as in ko-S + wt-E tissue recombinants, E2 cannot
elicit terminal differentiation by direct action through the epithelial
ER. Likewise, in the absence of epithelial ER in wt-S + ko-E
tissue recombinants, proliferation generates a moderately thickened
epithelium through the stromal ER, but terminal differentiation does
not occur. Only the coexpression of epithelial and stromal ER
insures the full differentiative response to E2 in vaginal
epithelium.
The demonstration that E2 induction of epithelial
mitogenesis involves only stromal ER raised obvious questions
concerning the role of epithelial ER in vagina. The present results,
showing that epithelial ER are necessary for epithelial
cornification and normal stratification in vaginal epithelium, provide,
at least, a partial answer. Direct E2 effects on
reproductive epithelial function have been described in
vitro. Specifically, E2 acts directly on epithelium in
tissue culture to inhibit proliferation (22, 23) and regulate secretion
of certain proteins, though uterine epithelial secretory proteins, such
as complement component C3, which are regulated by E2in vivo, are not E2-responsive in
vitro (24, 25). In addition, estrogen treatment has been shown to
increase epithelial progesterone receptor expression in isolated
vaginal epithelial cells in vitro (23). However, the present
results, showing the essential role of epithelial ER in normal
E2-induced vaginal epithelial differentiation, represent
the first function described for epithelial ER in vivo.
Additional work from our laboratory (26) indicates that both stromal
and epithelial ER are necessary for E2-induced
production of uterine secretory proteins, such as lactoferrin and C3
in vivo. Thus, in addition to its role in stratification and
cornification in the vaginal epithelium, epithelial ER seems to be
required for production of estrogen-induced secretory protein products
in uterine epithelium, a finding consistent with earlier data that
epithelial androgen receptor is necessary for androgen-induced
secretory protein production in seminal vesicle and prostate (27, 28).
The recent discovery of a second ER, termed ERß (29), raises the
possibility that this receptor could play a role in mediating
E2-induced epithelial proliferation and differentiation in
the tissue recombinants. ERß messenger RNA is expressed in low (but
detectable) quantities in ERKO uterus and in wt mouse uterus and vagina
(30). Although the in vivo role of ERß has not yet been
established, characteristic estrogen-induced responses (such as
increased uterine wet weight and lactoferrin production, increased
uterine and vaginal epithelial proliferation, and vaginal epithelial
cornification) are not observed when ERKO mice are treated with
E2 (2). It is uncertain whether ERß plays any role in
these characteristic responses to E2. However, it is clear
that ERß alone is not sufficient to mediate the
E2-stimulated induction of vaginal epithelial
proliferation, cornification, or stratification in the absence of
ER.
Immunohistochemical staining for ER in the tissue recombinants provided
an important control for verifying the completeness of tissue
separation and allowed verification of tissue origin in heterotypic
tissue recombinants. According to the manufacturer, the LH2 antibody
used in this study is specific to several epitopes along the entire
length of the ER molecule, some of which have high homology to
ERß. Based on the lack of ER immunostaining in ERKO vaginal tissue
with this antibody and despite the known expression of ERß messenger
RNA in ERKO reproductive tract (30), it seems that the LH2 antibody
either recognizes only ER or that ERß protein is not present in
ERKO vagina. Regardless, immunostaining for ER showed that epithelial
and stromal cells in tissue recombinants maintain expression of ER,
independent of the presence or absence of ER in the adjoining
tissue.
The nature of the stromal-epithelial communication and how it is modified by E2 are unknown. Stromal-epithelial communication is clearly reciprocal; in addition to stromal influences on epithelium, unknown factors produced in the epithelium may influence the differentiation and function of the stroma (31, 32, 33). Stromal-epithelial signaling may involve the production of growth factors or the decreased production of a tonic inhibitor of estrogen action (reviewed in Ref. 34). Paracrine communication could also occur by stromal effects on processes such as enzymatic modification (digestion, phosphorylation, or glycosylation) or changes in bioavailability (secretion, binding, and release from extracellular matrix) of growth factors themselves, their receptors, binding proteins, or enzymes that modify activity of the molecules involved (34). Stromal cells may also communicate with epithelial cells and modify their activity by altering the composition of basement membrane components. For example, progestin-induced proliferation of isolated nulliparous mammary epithelial cells occurs only when these cells are cultured on extracellular matrix components, such as Type IV collagen and fibronectin, that are normally secreted by stromal cells (35). Thus, E2 interactions with stromal ER may induce changes in the stromal cells that affect basement membrane composition and, in turn, permit epithelial proliferation and/or differentiation.
The mechanisms of E2-induced stromal-epithelial
interactions have clinical implications. Estrogen is typically
involved, at least as a permissive agent, in the initiation and
progression of neoplasias of estrogen target organs (such as mammary
gland, endometrium, vagina, and cervix). Because normal
E2-induced epithelial proliferation in uterus and vagina is
mediated through stromal ER, stromal-epithelial communication is
likely to be altered in endometrial, vaginal, and cervical cancers. In
this regard, epithelial cells of these tumors may show direct mitogenic
responses to estrogen, despite the loss of their normal stromal
association. This altered stromal-epithelial communication may not
involve only epithelial cells; stromal cells may also be a target for
epigenetic alterations that can lead to carcinogenesis (36).
Understanding the alterations in stromal-epithelial interactions during
the onset and progression of carcinogenesis may allow mechanistic
insight into the etiology of cancer and potentially provide new targets
for chemopreventative agents.
In conclusion, E2-induced vaginal epithelial proliferation
is mediated indirectly by stromal ER, which seems to be a universal
mechanism for E2-induced epithelial mitogenesis in female
reproductive organs. More important, vaginal epithelial differentiative
responses, such as cornification and normal stratification induced by
E2, require both stromal and epithelial ER and are
mechanistically different from E2-induced epithelial
proliferation. These are the first known functions attributed to
epithelial ER in vivo, and this is the first time any
epithelial response to E2 has been shown to require both
stromal and epithelial ER.
|
Acknowledgments |
|---|
|
Footnotes |
|---|
Received March 18, 1998.
|
References |
|---|
|
TopAbstractIntroductionMaterials and MethodsResultsDiscussionReferences |
|---|
(ER) and estrogen receptor-ß (ERß) messenger
ribonucleic acid in the wild-type and ER-knockout mouse.
Endocrinology 138:4613–4621This article has been cited by other articles:
 |
|
 |
|
  L. Yao, A. I. Agoulnik, P. S. Cooke, D. D. Meling, and O. D. Sherwood Relaxin Acts on Stromal Cells to Promote Epithelial and Stromal Proliferation and Inhibit Apoptosis in the Mouse Cervix and Vagina Endocrinology, May 1, 2008; 149(5): 2072 - 2079. [Abstract] [Full Text] [PDF]
|
|
|
|
|
|
M. Welsh, P. T. K. Saunders, N. I. Marchetti, and R. M. Sharpe Androgen-Dependent Mechanisms of Wolffian Duct Development and Their Perturbation by Flutamide Endocrinology, October 1, 2006; 147(10): 4820 - 4830. [Abstract] [Full Text] [PDF]
|
|
|
|
 |
|
 K. Inada, S. Hayashi, T. Iguchi, and T. Sato Establishment of a primary culture model of mouse uterine and vaginal stroma for studying in vitro estrogen effects. Experimental Biology and Medicine, March 1, 2006; 231(3): 303 - 310. [Abstract] [Full Text] [PDF]
|
|
|
|
|
|
H.-Y. Lee, S. Zhao, P. A. Fields, and O. D. Sherwood The Extent to which Relaxin Promotes Proliferation and Inhibits Apoptosis of Cervical Epithelial and Stromal Cells Is Greatest during Late Pregnancy in Rats Endocrinology, January 1, 2005; 146(1): 511 - 518. [Abstract] [Full Text] [PDF]
|
|
|
|
 |
|
 A. Y. Ting, A. D. Blacklock, and P. G. Smith Estrogen Regulates Vaginal Sensory and Autonomic Nerve Density in the Rat Biol Reprod, October 1, 2004; 71(4): 1397 - 1404. [Abstract] [Full Text] [PDF]
|
|
|
|
|
|
V. Selvaraj, M. A. Zakroczymski, A. Naaz, M. Mukai, Y. H. Ju, D. R. Doerge, J. A. Katzenellenbogen, W. G. Helferich, and P. S. Cooke Estrogenicity of the Isoflavone Metabolite Equol on Reproductive and Non-Reproductive Organs in Mice Biol Reprod, September 1, 2004; 71(3): 966 - 972. [Abstract] [Full Text] [PDF]
|
|
|
|
 |
|
 T. Kurita, A. A. Mills, and G. R. Cunha Roles of p63 in the diethylstilbestrol-induced cervicovaginal adenosis Development, April 1, 2004; 131(7): 1639 - 1649. [Abstract] [Full Text] [PDF]
|
|
|
|
 |
|
 S. C. Hewitt, B. J. Deroo, K. Hansen, J. Collins, S. Grissom, C. A. Afshari, and K. S. Korach Estrogen Receptor-Dependent Genomic Responses in the Uterus Mirror the Biphasic Physiological Response to Estrogen Mol. Endocrinol., October 1, 2003; 17(10): 2070 - 2083. [Abstract] [Full Text] [PDF]
|
|
|
|
|
|
Y. Katsu, D. B. Lubahn, and T. Iguchi Expression of a Novel C-Type Lectin in the Mouse Vagina Endocrinology, June 1, 2003; 144(6): 2597 - 2605. [Abstract] [Full Text] [PDF]
|
|
|
|
|
|
G. P. Risbridger, H. Wang, M. Frydenberg, and G. Cunha The Metaplastic Effects of Estrogen on Mouse Prostate Epithelium: Proliferation of Cells with Basal Cell Phenotype Endocrinology, June 1, 2001; 142(6): 2443 - 2450. [Abstract] [Full Text] [PDF]
|
|
|
|
 |
|
 D. L. Buchanan, T. Sato, R. E. Peterson, and P. S. Cooke Antiestrogenic Effects of 2,3,7,8-Tetrachlorodibenzo-p-dioxin in Mouse Uterus: Critical Role of the Aryl Hydrocarbon Receptor in Stromal Tissue Toxicol. Sci., October 1, 2000; 57(2): 302 - 311. [Abstract] [Full Text] [PDF]
|
|
|
|
|
|
T. Kurita, K.-j. Lee, P. S. Cooke, J. A. Taylor, D. B. Lubahn, and G. R. Cunha Paracrine Regulation of Epithelial Progesterone Receptor by Estradiol in the Mouse Female Reproductive Tract Biol Reprod, April 1, 2000; 62(4): 821 - 830. [Abstract] [Full Text]
|
|
|
|
|
|
T. Kurita, K.-j. Lee, P. S. Cooke, J. P. Lydon, and G. R. Cunha Paracrine Regulation of Epithelial Progesterone Receptor and Lactoferrin by Progesterone in the Mouse Uterus Biol Reprod, April 1, 2000; 62(4): 831 - 838. [Abstract] [Full Text]
|
|
|
|
|
|
L. Zhao, C. S. Samuel, G. W. Tregear, F. Beck, and E. M. Wintour Collagen Studies in Late Pregnant Relaxin Null Mice Biol Reprod, March 1, 2000; 63(3): 697 - 703. [Abstract] [Full Text]
|
|
|
|
 |
|
 J. F. Couse and K. S. Korach Estrogen Receptor Null Mice: What Have We Learned and Where Will They Lead Us? Endocr. Rev., June 1, 1999; 20(3): 358 - 417. [Abstract] [Full Text]
|
|
|
|
|
|
D. L. Buchanan, T. Setiawan, D. B. Lubahn, J. A. Taylor, T. Kurita, G. R. Cunha, and P. S. Cooke Tissue Compartment-Specific Estrogen Receptor-{alpha} Participation in the Mouse Uterine Epithelial Secretory Response Endocrinology, January 1, 1999; 140(1): 484 - 491. [Abstract] [Full Text]
|
|
|
|
 |
|
 J. Teng, Z.-Y. Wang, and D. E. Bjorling Estrogen-induced proliferation of urothelial cells is modulated by nerve growth factor Am J Physiol Renal Physiol, June 1, 2002; 282(6): F1075 - F1083. [Abstract] [Full Text] [PDF]
|
|
| ||||||||||||||||
