| HOME | HELP | FEEDBACK | SUBSCRIPTIONS | ARCHIVE | SEARCH | TABLE OF CONTENTS |
ARTICLES |
Department of Anatomy and Human Biology, The University of Western Australia (P.J.B., B.J.W.), Perth, Western Australia; and Baker Medical Research Institute (Z.S.K.), Melbourne, Victoria, Australia
Address all correspondence and requests for reprints to: Dr. Brendan J. Waddell, Department of Anatomy and Human Biology, The University of Western Australia, Nedlands, Western Australia 6907, Australia. E-mail: bwaddell{at}anhb.uwa.edu.au
 |
Abstract |
|---|
 Top Abstract IntroductionMaterials and MethodsResultsDiscussionReferences |
|---|
|
Introduction |
|---|
|
TopAbstractIntroductionMaterials and MethodsResultsDiscussionReferences |
|---|
Given the range of glucocorticoid effects in the uterus, local expression of the 11ßHSD enzymes may have an important role in normal uterine physiology by regulating access of endogenous glucocorticoid to the GR. Expression of 11ßHSD-1 messenger RNA (mRNA) has been observed in whole uterus of the rat (9) and in endometrium of the sheep (10) and human (11), and we recently reported a dramatic up-regulation of 11ßHSD-1 mRNA and protein expression specifically within the uterine myometrium late in rat pregnancy (12). Expression of 11ßHSD-2 mRNA has also been observed in pregnant rat uterus (13), and bioactivity indicative of 11ßHSD-2 expression is present in human endometrial stromal cells (11). Importantly, expression of both enzymes in human endometrial cells in vitro is stimulated by the synthetic progestin, medroxyprogesterone acetate, particularly in the presence of estrogen (11). This suggests that uterine expression of both 11ßHSD enzymes may vary across the normal cycle in relation to changes in ovarian steroid secretion, with potentially important implications for glucocorticoid action. This possibility was investigated in the present work by immunocytochemical localization of 11ßHSD-1 and -2 at each stage of the 4-day estrous cycle (proestrus, estrus, postestrus, and diestrus). These initial studies demonstrated distinct immunolocalization and clear cyclic variation in uterine 11ßHSD-1 and -2 immunoreactivities, with both enzymes being more abundant at times of maximal estrogen secretion. Therefore, in a second experiment we assessed the effects of ovariectomy with and without estrogen and/or progesterone replacement on uterine expression of 11ßHSD-1 and -2.
|
Materials and Methods |
|---|
|
TopAbstractIntroductionMaterials and MethodsResultsDiscussionReferences |
|---|
To assess the effects of estrogen and progesterone, bilateral ovariectomies were performed under halothane/nitrous oxide anesthesia at diestrus via a midline abdominal incision. Rats were then treated immediately with estradiol (in propylene glycol at a rate of 150 ng/h via miniosmotic pumps; Alzet 2001, Alza Corp., Palo Alto, CA) and/or progesterone (2 mg/0.2 ml peanut oil, sc); progesterone injections were repeated 24 and 48 h later. The use of miniosmotic pumps for estrogen administration ensured that plasma levels of this relatively potent steroid could be maintained within the physiological range at all times. Control ovariectomized (OVX) rats received vehicle alone. Approximately 72 h after ovariectomy, uteri were removed and weighed, and portions were collected for immunocytochemistry as described above. The remainder was frozen on liquid nitrogen and stored at -80 C for subsequent Western blot analysis. All procedures involving animals were conducted after approval by the animal ethics committee of The University of Western Australia.
Immunocytochemistry
Immunocytochemistry was performed on sections of whole uterus as
previously described for myometrium (12). Briefly, after sectioning,
paraffin was removed, and nonspecific staining was blocked by
incubation of tissue sections (4 µm) with 3% hydrogen peroxide in
methanol for 10 min, followed by incubation with 2% BSA-PBS-0.02%
Triton X-100 for 20 min. Tissue sections were then exposed to
11ßHSD-1 and -2 antibodies. The 11ßHSD-1 antiserum, 56-127 was
raised against purified rat liver 11ßHSD (15), and the 11ßHSD-2
antibody was raised against the last 16 residues of the rat 11ßHSD-2
protein and immunopurified (16). The 11ßHSD-1 antiserum was diluted
1:500 in 0.1% BSA-PBS-0.02% Triton X-100, and the immunopurified
11ßHSD-2 antibody was diluted to a final concentration of 0.5
µg/ml. Positive immunostaining was identified by the addition of an
antirabbit IgG biotinylated secondary antibody followed by
avidin-biotin-peroxidase complex (Vectastain Elite ABC kit, Vector
Laboratories, Burlingame, CA) and diaminobenzidine. Sections were
counterstained with Gill’s hematoxylin. The absence of nonspecific
staining for both 11ßHSD-1 and -2 was demonstrated by negative
control sections (i.e. incubations carried out as described
above but without 11ßHSD-1 or -2 antibodies) and further supported by
the lack of immunostaining for either enzyme in uterine sections from
OVX rats (see Results).
Western blot analysis
Whole uteri (n = 3/group) were homogenized in 4 vol 10
mM sodium phosphate buffer (pH 7.0) containing 0.25
M sucrose, 1 µM EDTA, 1 µM
phenylmethylsulfonylfluoride, and 100 µg/ml trypsin inhibitor.
Microsomes were recovered by sequential centrifugation and subjected to
Western blot analysis using a Vectastain Elite ABC kit as previously
described (17). For quantification of the 11ßHSD-1 and -2 signals,
blots were incubated with a 125I-labeled antirabbit IgG
secondary antibody (Amersham Australia, Sydney, Australia) and exposed
on a Fuji Imaging Plate (Fuji, Tokyo, Japan). The resultant images were
quantified using a Fuji Bioimager and a MacBas 1000 program, and in
each case, a background reading was obtained from the area immediately
adjacent to the immunopositive signal.
Statistical analysis
Changes in uterine wet weight during the estrous cycle and after
ovariectomy with or without estrogen and/or progesterone replacement
were assessed by one-way ANOVA and least significant difference tests
(18). Similar analyses were used to assess changes in 11ßHSD-1 and -2
immunoreactive signals detected by Western blot analysis in OVX rats
with or without estrogen and/or progesterone replacement.
|
Results |
|---|
|
TopAbstractIntroductionMaterials and MethodsResultsDiscussionReferences |
|---|
).
|
) and maximal at
proestrus (Fig. 1b). Positive immunostaining was also observed in the
endometrial stroma and myometrium, but in these regions was limited to
cells that appeared morphologically to be eosinophils rather than
stromal or myometrial cells per se (Fig. 1, a and n). The
number of these positively stained cells was maximal at proestrus (Fig. 1n) and estrus. No immunostaining was evident in endometrial stromal
cells (Fig. 1, a, b, and n) or myometrial cells at any stage of the
cycle or in sections incubated in the absence of primary antibody (Fig. 1c). Western blot analysis of 11ßHSD-1 protein in whole uteri from
rats at proestrus showed the presence of a single 34-kDa immunoreactive
species (see Fig. 2a).
|
|
, e, f, and o) and
estrus in myometrium (both layers of smooth muscle) and in endometrial
stroma. The relative paucity of immunostaining at diestrus was
particularly evident in the myometrium (Fig. 1d). No immunostaining was
observed in the luminal and glandular epithelium or in eosinophils at
any stage of the cycle (Fig. 1, f and o), and sections incubated in the
absence of the 11ßHSD-2 antibody were also immunonegative. Western
blot analysis of 11ßHSD-2 protein in whole uteri from rats at
proestrus showed the presence of a single 40-kDa immunoreactive species
(Fig. 2b).
Effect of estrogen and progesterone manipulation
Uterine weights. To assess the effectiveness of the
ovariectomy and steroid replacement treatments, uterine wet weight was
measured in each group. As shown in Table 1, uterine weight in OVX rats
varied considerably among the different treatment groups (F =
56.1; P < 0.001, by one-way ANOVA), being minimal
after ovariectomy alone but increasing after treatment with estrogen
(P < 0.001) and to a lesser extent with estrogen plus
progesterone (P < 0.01). Progesterone replacement
alone had no effect on uterine wet weight.
11ßHSD-1. Uterine immunoreactivity for 11ßHSD-1 by Western
blot analysis was low in OVX rats, but was partly restored to proestrus
levels by replacement of estrogen with or without progesterone (Fig. 3a). Progesterone replacement alone had
no effect on the uterine 11ßHSD-1 signal. The reduction in 11ßHSD-1
immunoreactivity in OVX rats appeared to be due to reduced
immunostaining in luminal and glandular epithelium (Fig. 1g). Moreover,
there was an apparent decrease in the number of positively stained
cells (eosinophils) in the stroma after ovariectomy, and this effect
was reversed after estrogen replacement (with or without progesterone)
as was immunostaining for 11ßHSD-1 in the luminal and glandular
epithelium (Fig. 1, h and i). Progesterone replacement alone had no
detectable effect on 11ßHSD-1 immunostaining.
|
4-fold; P < 0.001) than the
maximal levels observed during the cycle (i.e. at proestrus;
Fig. 3b). Immunocytochemical analysis showed that although this effect
of estrogen on 11ßHSD-2 occurred in both layers of myometrium and in
endometrial stromal cells (Fig. 1, j, k, and m), the intensity of
staining after steroid replacement was particularly enhanced in the
myometrium (Fig. 1m). This suggests that the much higher 11ßHSD-2
signal detected by Western analysis after steroid replacement compared
with that at proestrus (Fig. 3b) primarily reflects myometrial
11ßHSD-2. Coadministration of progesterone with estrogen had no
apparent additive effect on total 11ßHSD-2 immunoreactivity (by
Western analysis), although it clearly enhanced the intensity of
immunostaining in stromal cells (compare Fig. 1, k and l). Progesterone
replacement alone had a clear, but slight, stimulatory effect on
11ßHSD-2 immunostaining in stromal cells (data not shown), but this
effect was also not detectable by Western analysis of whole uterus
(Fig. 3b). |
Discussion |
|---|
|
TopAbstractIntroductionMaterials and MethodsResultsDiscussionReferences |
|---|
The presence of 11ßHSD-1 in uterine luminal epithelium is similar to our recent observation of its immunolocalization to the reformed uterine epithelium during late pregnancy, although in the latter, 11ßHSD-1 immunostaining was far more intense (12). Similarly, 11ßHSD-1 immunostaining has recently been localized to the uterine epithelium of sheep during the estrous cycle (10). The effective absence of 11ßHSD-1 immunoreactivity in myometrium during the cycle is also consistent with the observation that this enzyme is highly expressed in rat myometrium only in the last days of pregnancy (12). It is noteworthy that even after 3 days of estrogen exposure the myometrium did not show immunolocalization of 11ßHSD-1, suggesting that its induction late in pregnancy is due to factors other than the rising estrogen titer characteristic of this time (19). The present study also demonstrates immunolocalization of 11ßHSD-1 to isolated cells in the endometrial stroma, the number of which appeared to increase at proestrus and estrus. The morphological appearance of these cells and their variable presence (in parallel with estrogen changes) indicate that they are eosinophils. These cells comprise 95% of blood-derived cells in the rat uterus (20), and vast numbers are known to migrate into the uterus at proestrus (21) under the influence of estrogen (1, 22, 23). To our knowledge this is the first evidence of 11ßHSD-1 immunostaining in eosinophils or any other white blood cells, but its physiological role in these cells remains to be established.
The presence of 11ßHSD-2 immunostaining in the stroma and myometrium, but not in epithelium, is the first demonstration of 11ßHSD-2 protein expression in the uterus during the estrous cycle and is similar to the reported distribution of 11ßHSD-2 mRNA in pregnant uterus (13). It is also consistent with the recent observation of 11ßHSD-2 bioactivity in human endometrial stromal cells in vitro (11).
The observed cyclic variation in 11ßHSD-1 and -2 protein expression during the estrous cycle is probably due to variations in ovarian estrogen and progesterone secretion, as both 11ßHSD-1 and -2 were depleted by ovariectomy and restored by steroid replacement. Although there was no difference in total 11ßHSD-2 immunoreactivity measured by Western blot analysis in uteri from OVX plus estrogen-treated and OVX plus estrogen/progesterone-treated rats, immunostaining was clearly more intense and widespread in the endometrial stroma of rats in which both steroids were replaced. This apparent inconsistency presumably reflects the very high expression of 11ßHSD-2 in the myometrium, which far exceeded that in endometrial stroma, and the associated dominance of myometrial 11ßHSD-2 in the Western analyses of whole uteri. This contention is further supported by the observation that a low level of 11ßHSD-2 immunostaining was apparent in endometrial stromal cells of rats receiving only progesterone replacement after ovariectomy, even though no effect of progesterone alone was detected by Western blot analysis. These observations highlight the greater sensitivity of immunocytochemistry and the need for cautious interpretation when analyses are based on whole uteri, especially when a particular protein is differentially expressed among subgroups of cells. The apparent additive effects of progesterone and estrogen on 11ßHSD-2 protein expression are consistent with the recent observation that 11ßHSD-2 bioactivity in human endometrial stromal cells in vitro was stimulated 2.5-fold by either estrogen or the synthetic progestin medroxyprogesterone acetate, but was increased 8-fold when cells were exposed to both steroids together (11). In contrast to the present study, however, 11ßHSD-1 expression in human endometrial stromal cells in vitro (measured by Northern blot analysis) was stimulated by estrogen only when given together with medroxyprogesterone acetate (11). The relatively potent effect of estrogen alone (in the absence of exogenous progesterone) with respect to that of both 11ßHSD-1 and -2 in our model may be due to the continued presence of some endogenous progesterone in OVX rats, as the adrenal cortex provides an additional source of progesterone in the rat (24, 25).
The physiological importance of 11ßHSD-1 in uterine luminal epithelial cells will depend to a large extent on the relative proportions of 11ß-dehydrogenase and 11-oxoreductase activities in vivo. Although both reactions are theoretically possible for this enzyme, recent evidence suggests that 11-oxoreductase generally prevails in intact cells (26) and tissue (12), in which case 11ßHSD-1 would be expected to enhance local glucocorticoid actions (27, 28). Glucocorticoids could potentially affect the cell cycle in uterine epithelium, as dexamethasone has been shown to inhibit both uterine epithelial cell death (2, 3) and proliferation (29). Alternatively, glucocorticoids may influence electrolyte transport across the uterine epithelium, which is highly dynamic during the rat estrous cycle (30). Indeed, exogenous glucocorticoids have been shown to block the estrogen-induced increase in uterine fluid volume and electrolyte concentration (31), and Whorwood et al. (32) have proposed that regulation of glucocorticoid access to GR by 11ßHSD-1 modulates sodium/potassium transport in colon and kidney epithelia. Whether 11ßHSD-1 plays a similar role in the uterus and the specific mechanisms involved remain to be determined.
Expression of 11ßHSD-2 in the uterine stroma may play a role in the normal dynamic physiology of the uterus by local inactivation of endogenous glucocorticoids, as this form of the enzyme almost exclusively exhibits 11ß-dehydrogenase activity (7, 33). Exogenous glucocorticoids have been shown to inhibit several uterine functions, including PG synthesis (4), estrogen-induced eosinophil infiltration (1), and uterine growth (34, 35). Perhaps the most important of these is PG synthesis, which is crucial for the induction of increased stromal vascular permeability (36) that precedes and accompanies implantation (37, 38, 39). Indeed, administration of dexamethasone has been shown to dramatically reduce the number of implantation sites in pregnant rats, an effect reversed by treatment with PGs (40). In the present study administration of both progesterone and estrogen to OVX rats closely mimicked the pattern of ovarian output of these steroids during the preimplantation period; therefore, the consequent increase in 11ßHSD-2 immunostaining in the endometrial stroma is consistent with a protective role (by its inactivation of glucocorticoids) for this enzyme around the time of implantation. In addition, Roland and Funder (13) proposed that 11ßHSD-2 expression in female reproductive tissues may be important to ensure that progesterone-responsive genes are directed only by progesterone and not by glucocorticoids. In this case, 11ßHSD-2 metabolism of active glucocorticoids would confer progesterone selectivity on otherwise nonselective response elements present in cells that coexpress the progesterone receptor (PR) and GR (13). Indeed, both the PR (41) and GR (12, 42) are expressed in rat uterus, and the pattern of PR expression in endometrial stromal cells and myometrial cells (41) is similar to that of 11ßHSD-2, with most expression evident at proestrus.
The specific and distinct localization of 11ßHSD-1 and -2 to the uterine epithelium and stroma, respectively, may have important implications in relation to a physiological interplay between these enzymes. A major concern with regard to the functional significance of 11ßHSD-1 is its very high Km and the associated requirement for substrate concentrations that are likely to exceed those in plasma. We suggest that high 11ßHSD-2 expression in stromal cells and their close proximity to epithelial cells and eosinophils (both with 11ßHSD-1 expression) provides a mechanism through which a supply of 11-dehydrocorticosterone substrate to epithelial cells and eosinophils is maintained at the very high concentrations demanded by 11ßHSD-1. In addition to this potential variation in substrate supply, the activities of both 11ßHSD-1 and -2 in different cell types may be affected by variations in a range of other determinants, including inhibitory steroids and available cofactors. For example, 11ßHSD-1 activity appears particularly sensitive to variations in cofactor supply and pH (5), and so its ultimate biological effect in eosinophils and/or uterine epithelial cells may be governed by local variations in these factors.
In conclusion, these results establish the presence of both 11ßHSD-1 and -2 in the nonpregnant rat uterus and show distinct distributions and cyclic variation for the two enzymes. Moreover, uterine expression of both enzymes is highly responsive to estrogen, an effect enhanced by coadministration of progesterone with respect to 11ßHSD-2. Estrogen-induced 11ßHSD-2 expression in the stroma is likely to limit potentially disruptive effects of glucocorticoids on uterine function during the cycle by reducing the local concentration of active glucocorticoid. On the other hand, expression of 11ßHSD-1 in the epithelium may provide the uterus with the capacity to selectively reactivate glucocorticoids in these cells.
|
Acknowledgments |
|---|
|
Footnotes |
|---|
2 Supported by a Research Studentship from The University of Western
Australia
Received May 19, 1997.
|
References |
|---|
|
TopAbstractIntroductionMaterials and MethodsResultsDiscussionReferences |
|---|
, and histamine in
delayed implantation in the mouse. Acta Endocrinol (Copenh) 91:801–807
This article has been cited by other articles:
 |
|
 |
|
  P. J. Mark, S. Augustus, J. L. Lewis, D. P. Hewitt, and B. J. Waddell Changes in the Placental Glucocorticoid Barrier During Rat Pregnancy: Impact on Placental Corticosterone Levels and Regulation by Progesterone Biol Reprod, June 1, 2009; 80(6): 1209 - 1215. [Abstract] [Full Text] [PDF]
|
|
|
|
 |
|
 A. Naray-Fejes-Toth and G. Fejes-Toth Reply to Geerling and Loewy Am J Physiol Renal Physiol, July 1, 2007; 293(1): F442 - F443. [Full Text] [PDF]
|
|
|
|
|
|
A. Naray-Fejes-Toth and G. Fejes-Toth Novel mouse strain with Cre recombinase in 11beta-hydroxysteroid dehydrogenase-2-expressing cells Am J Physiol Renal Physiol, January 1, 2007; 292(1): F486 - F494. [Abstract] [Full Text] [PDF]
|
|
|
|
 |
|
 J. T Smith, P. J Mark, and B. J Waddell Developmental increases in plasma leptin binding activity and tissue Ob-Re mRNA expression in the rat J. Endocrinol., March 1, 2005; 184(3): 535 - 541. [Abstract] [Full Text] [PDF]
|
|
|
|
 |
|
 J. W. Tomlinson, E. A. Walker, I. J. Bujalska, N. Draper, G. G. Lavery, M. S. Cooper, M. Hewison, and P. M. Stewart 11{beta}-Hydroxysteroid Dehydrogenase Type 1: A Tissue-Specific Regulator of Glucocorticoid Response Endocr. Rev., October 1, 2004; 25(5): 831 - 866. [Abstract] [Full Text] [PDF]
|
|
|
|
 |
|
 B. J. Waddell, S. Hisheh, Z. S. Krozowski, and P. J. Burton Localization of 11{beta}-Hydroxysteroid Dehydrogenase Types 1 and 2 in the Male Reproductive Tract Endocrinology, July 1, 2003; 144(7): 3101 - 3106. [Abstract] [Full Text] [PDF]
|
|
|
|
|
|
J. T. Smith and B. J. Waddell Leptin Receptor Expression in the Rat Placenta: Changes in Ob-Ra, Ob-Rb, and Ob-Re with Gestational Age and Suppression by Glucocorticoids Biol Reprod, October 1, 2002; 67(4): 1204 - 1210. [Abstract] [Full Text] [PDF]
|
|
|
|
|
|
R. M. Seeber, J. T. Smith, and B. J. Waddell Plasma Leptin-Binding Activity and Hypothalamic Leptin Receptor Expression During Pregnancy and Lactation in the Rat Biol Reprod, June 1, 2002; 66(6): 1762 - 1767. [Abstract] [Full Text] [PDF]
|
|
|
|
|
|
B. J. Waddell and P. J. Burton Full Induction of Rat Myometrial 11{beta}-Hydroxysteroid Dehydrogenase Type 1 in Late Pregnancy Is Dependent on Intrauterine Occupancy Biol Reprod, April 1, 2000; 62(4): 1005 - 1009. [Abstract] [Full Text]
|
|
|
|
|
|
R. M. Sapolsky, L. M. Romero, and A. U. Munck How Do Glucocorticoids Influence Stress Responses? Integrating Permissive, Suppressive, Stimulatory, and Preparative Actions Endocr. Rev., February 1, 2000; 21(1): 55 - 89. [Abstract] [Full Text]
|
|
|
|
|
|
P. J. Burton and B. J. Waddell Dual Function of 11ß-Hydroxysteroid Dehydrogenase in Placenta: Modulating Placental Glucocorticoid Passage and Local Steroid Action Biol Reprod, February 1, 1999; 60(2): 234 - 240. [Abstract] [Full Text] [PDF]
|
|
|
|
|
|
B. J. Waddell, R. Benediktsson, R. W. Brown, and J. R. Seckl Tissue-Specific Messenger Ribonucleic Acid Expression of 11{beta}-Hydroxysteroid Dehydrogenase Types 1 and 2 and the Glucocorticoid Receptor within Rat Placenta Suggests Exquisite Local Control of Glucocorticoid Action Endocrinology, April 1, 1998; 139(4): 1517 - 1523. [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 |
