| HOME | HELP | FEEDBACK | SUBSCRIPTIONS | ARCHIVE | SEARCH | TABLE OF CONTENTS |
Prince Henry’s Institute of Medical Research (T.J.K.-L., N.B.M., L.A.S.) and Department of Obstetrics and Gynecology (T.J.K.-L.), Monash University, Clayton, Victoria 3168, Australia
Address all correspondence and requests for reprints to: T. J. Kaitu’u-Lino, Prince Henry’s Institute of Medical Research, Level 4, Block E, Monash Medical Centre, 246 Clayton Road, Clayton, Victoria 3168, Australia. E-mail: tuuhevaha.kaituu{at}princehenrys.org.
 |
Abstract |
|---|
 Top Abstract IntroductionMaterials and MethodsResultsDiscussionReferences |
|---|
|
Introduction |
|---|
|
TopAbstractIntroductionMaterials and MethodsResultsDiscussionReferences |
|---|
Due to the limited availability of animal models for study of these processes, we established a mouse model of endometrial breakdown and repair (9) first described in the 1980s (10). In this model, decidualization is artificially induced and progesterone (P) support withdrawn. Tissue breakdown is initiated within 12 h of P withdrawal and is complete in most animals by 24 h. Tissue restoration or repair is equally rapid and, as in women, begins even while breakdown is still underway in adjacent areas. In most mice, the endometrium is restored to its original state by 48 h after P withdrawal. Previous studies (9, 11, 12) have focused upon teasing out the factors that are essential for these dynamic processes.
The current dogma surrounding endometrial restoration after menses suggests a need for estrogen-primed proliferation. Although estrogen may not be required for the initial reepithelialization of the uterine surface, it is widely believed that it is essential for successful stromal renewal, with the advent of stromal mitoses coinciding with the rising estrogen levels in women (13) and rhesus monkeys (14). However, in the mouse model, restoration of both the stromal and epithelial components proceeds rapidly after breakdown and results in what appears to be a normal endometrium (9) despite the mice being ovariectomized. Likewise, after menstruation in human endometrial explants in ovariectomized immunodeficient mice, complete endometrial repair is seen in the absence of exogenous hormonal support (15). However, in all studies to date, potential estrogenic influences from extraovarian sources (particularly the diet and fat) remain.
The main aim of the present study was to ascertain whether full endometrial restoration is possible in the complete absence of estrogen, achieved by the use of ovariectomized animals maintained on a soy-free diet and injected with the aromatase inhibitor letrozole. The data clearly demonstrate that estrogen is not essential to endometrial restoration after a menses-like event. The endometrium from estrogen-deplete animals repairs to a normal phenotype containing all cellular components.
|
Materials and Methods |
|---|
|
TopAbstractIntroductionMaterials and MethodsResultsDiscussionReferences |
|---|
Mouse model of endometrial breakdown and repair
Endometrial breakdown and repair was induced in mice according to a standard protocol as previously described (9). Briefly, under xylazine/ketamine-induced anesthesia, mice were ovariectomized 7 d before the first of three daily sc injections of 100 ng 17ß-estradiol (Sigma Chemical Co., St. Louis, MO) in arachis oil at approximately 0900 h. After resting the mice for 3 d, P implants were inserted sc into the back of each mouse. Simultaneously, the mice received a single injection sc of P in arachis oil, and an injection of 5 ng 17ß-estradiol in arachis oil was given at approximately 0900 h on that and the subsequent 2 d. At approximately 1100 h on the day of the final 17ß-estradiol injection, 20 µl sesame oil was injected into the lumen of the right uterine horn of each mouse to induce decidualization. The left horn remained untreated as a control. The P implants were removed 48 h after oil injection, and mice were killed 24 and 48 h after P removal. Uteri were cleaned of fat and weighed. Tissue was cut into pieces and fixed in Carnoy’s fixative for 4 h or phosphate-buffered formalin overnight and processed to wax in an orientation that would provide cross-sections. For RNA studies, intact uteri were snap-frozen and stored at –80 C. Any mouse in which the oil-treated horn had not decidualized (as evidenced by weight 
400% of the untreated contralateral horn and histological assessment) was excluded from the study.
Study groups
Two main groups (control and estrogen-deplete, both ovariectomized) were included for analysis in this study, with a third estrogen-replacement group used as a positive control for serum and mRNA expression analysis. All groups were subjected to the standard protocol with minor modifications. The control and estrogen-replacement groups were maintained before and throughout the period of study on regular mouse chow (containing 0.146 mg/g isoflavones), whereas the estrogen-deplete group was placed on a soy-free chow (Glen Forrest Stokfeeders, Glen Forres, Western Australia, Australia) (contains undetectable levels of isoflavones) 7 d before ovariectomy and maintained on this diet throughout the periods of the study. Animals in the estrogen-deplete group were also administered sc letrozole (a third-generation aromatase inhibitor; Pharma AG, Basel, Switzerland) injections from the time of P withdrawal (see below), whereas the control group was administered vehicle, PBS without Ca2+ and Mg2+ (PBS–) alone. The estrogen-replacement group received 250 ng sc injections of 17ß-estradiol dissolved in arachis oil in place of the letrozole injections.
Preparation and administration of letrozole
Letrozole was initially dissolved in ethanol to a final concentration of 6.25 mg/ml by shaking at 37 C for 1 h. The stock ethanol solution was then dissolved in PBS– to a final concentration of 200 µg/ml, of which 100 µl was injected sc into the animals to give a dose of 20 µg/injection. Working solutions were prepared immediately before use with injections administered at the time of P withdrawal and 24 h later (two injections per animal). A dose of 20 µg per animal of letrozole was chosen, with evidence from published data showing 10–20 µg to be an effective dose in mice (16). Control animals were administered 100 µl PBS– sc at the time of letrozole injections.
Analysis of epithelial, smooth muscle and endothelial cells, proliferating cells, and estrogen receptor 
(ER)-positive cells
Immunohistochemistry was conducted on cross-sections of uteri from control animals at various stages of endometrial breakdown and restoration to provide further characterization of the model appropriate to this study. Rabbit antimouse polyclonal pan-cytokeratin (Santa Cruz Biotechnology, Inc., Santa Cruz, CA) was used to detect epithelial cells, mouse antihuman monoclonal smooth muscle actin (Dako, Glostrup, Denmark) to detect myometrial smooth muscle cells, rabbit antimouse polyclonal Ki67 (Novocastra Laboratories, Newcastle, UK) to detect proliferating cells, and mouse monoclonal antihuman ER (Novocastra Laboratories) to detect ER-positive cells.
Paraffin sections (5 µm) of formalin-fixed (for smooth muscle actin and ER) or Carnoy’s-fixed (for cytokeratin and Ki67) tissues were dewaxed in Histosol (Sigma) and rehydrated through descending grades of ethanol. Sections for detection of cytokeratin, Ki57, and ER were then immersed in 0.1 M citrate buffer and heated for 5 min on high (for cytokeratin) or 5 min on high and 5 min on medium low (for Ki67 and ER) in a 700-W microwave. No antibody retrieval was carried out for smooth muscle actin. Once the slides had returned to room temperature, they were washed for 10 min in Tris-buffered saline (TBS; pH 7.6) (smooth muscle actin, Ki67, and ER) or for 10 min in TBS containing 0.6% Tween 20 (pan-cytokeratin), rinsed in distilled H2O, and immersed in 3% H2O2 in methanol for 10 min at room temperature.
Sections were then incubated with blocking solution containing 10% normal goat serum in TBS for 30 min at room temperature and incubated for 1 h at room temperature with primary antibodies diluted in 1% fetal calf serum/TBS to 4 µg/ml (pan-cytokeratin), 0.18 µg/ml (smooth muscle actin), 0.15 µg/ml (Ki67), or 3.5 µg/ml (ER) before washing in TBS (smooth muscle actin, Ki67, and ER) for 10 min or sequentially in 0.6% (vol/vol) Tween 20 in TBS and thrice in TBS (pan-cytokeratin). For pan-cytokeratin and Ki67, biotinylated goat antirabbit IgG diluted at 1:200 in 1% fetal calf serum/TBS was applied for 30 min at room temperature and the slides washed as described above. The StrepABC horseradish peroxidase kit (Dako) was then applied according to manufacturer’s instructions. For smooth muscle actin and ER, the mouse Envision (horseradish peroxidase) system (Dako) was applied for 30 min at room temperature and the slides washed as described above. Diaminobenzidine solution (Dako) revealed the smooth muscle actin, Ki67 pan-cytokeratin, and ER staining. Sections were then lightly counterstained with Harris hematoxylin (Accustain; Sigma Diagnostics, Castle Hill, New South Wales, Australia), dehydrated, and mounted using DPX mounting medium (BDH Laboratory Supplies, Poole, UK).
Negative controls were included for each tissue section by substitution of the primary antibody with a matching concentration of normal rabbit IgG (for pan-cytokeratin and Ki67) or normal mouse IgG (smooth muscle actin and ER).
Analysis of serum estradiol levels
The ultrasensitive estradiol RIA kit (Diagnostic Systems Laboratories, Webster, TX) was used according to manufacturer’s specifications to analyze serum estradiol levels. Due to the amount of serum required for assay, equal amounts of serum from three to four animals under each treatment regime were pooled, with four to five pools measured for each treatment. Serum from an estrogen-replacement pool was included in analyses as a positive control. Sensitivity of the assay (supplied by manufacturer) was 2.2 pg/ml, with the lowest standard at 5 pg/ml.
Quantification of mRNA for estrogen-responsive genes lactoferrin and P receptor (PR)
RNA was isolated from frozen uteri from control, estrogen-deplete, and estrogen-replacement animals using TRIZOL (Invitrogen Australia, Pty. Ltd., Mount Waverley, Victoria, Australia) according to the manufacturer’s instructions with the exception of the inclusion of a second chloroform-only extraction to prevent any phenol carryover. RNA samples were then resuspended in 50 µl diethylpyrocarbonate H2O. Contaminating DNA was removed using the DNase-free kit (Ambion, Austin, TX) according to the manufacturer’s instructions. Total RNA concentrations were determined using the Ribogreen fluorescence RNA assay (Molecular Probes, Eugene, OR) (17).
Total RNA (1 mg) was reverse transcribed in triplicate reactions using random hexameric primers and AMV-RTase (Roche) at 46 C for 90 min. Omission of reverse transcriptase served as a negative control. Triplicate RTs were analyzed for efficiency and reproducibility of the RT reaction by real-time PCR for ribosomal 18S subunit using a LightCycler (Roche) as previously described (17). Primers used were sense 5'-CGG CTA CCA CAT CCA AGG AA-3' and antisense 5'-GCT GGA ATT ACC GCG GCT-3'. Amplified DNA quantitation was performed using SYBR Green I, by comparison with serially diluted cDNA standards generated by PCR. Cycle conditions were 95 C for 10 min and 35 cycles of 95 C for 15 sec, 60 C for 5 sec, and 72 C for 10 sec. The 18S concentrations (picograms per milliliter) were compared between triplicates, and the intraassay variability of the RT and PCR steps was evaluated. Within triplicates, samples outside 15% variability of the average 18S concentration were excluded as outliers. Otherwise, triplicates were pooled, creating cDNA reaction products that are close representations of the initial mRNA population. Negative controls were performed by omission of RT.
cDNA used for 18S real time RT-PCR was then used to assess lactoferrin and PR genes. Primers used were lactoferrin sense 5'-AGC CCC GGA CTC ACT ACT ATG-3' and antisense 5'-AGG TAT GGA AGT GTC CCT-3' and PR sense 5'-CCA ACT CAC AAA ACT TCT CGA CA-3' and antisense 5'-GGC AGC AAT AAC TTC AGA CAT CA-3'. Amplified DNA quantitation was performed using SYBR Green I, by comparison with serially diluted cDNA standards generated by PCR. Cycle conditions for both primer sets were 95 C for 10 min and 40 cycles of 95 C for 15 sec, 67 C for 5 sec, and 72 C for 10 sec. Lactoferrin and PR concentrations were obtained by averaging duplicate samples from triplicate runs.
Assessment of extent of tissue breakdown/restoration
Cross-sections of formalin- and Carnoy’s-fixed uterine tissue from control and estrogen-deplete animals were deparaffinized and hydrated by processing sections through Histosol (Sigma) and a graded series of ethanol to distilled H2O. The hydrated sections were stained with hematoxylin and eosin using standard procedures. Stained sections were dehydrated and mounted under coverslips using DPX mounting medium. Sections were then examined for extent of tissue breakdown and/or repair and analyzed using a scoring system from 1–5 as previously described (11), where 1 = intact decidualized tissue, 3 = complete breakdown, and 5 = complete repair. For each animal, two to four sections from different areas of the uterus were assessed and scored to determine the intra-animal variation. Inter-animal variation was determined by calculation and comparison of the mean score for each animal. All tissue sections were scored blind by two independent observers.
Statistical analysis
After testing for normal distribution, statistical analysis was performed using the Kruskal-Wallis test followed by Dunn’s multiple comparison post hoc test (P 
0.05 was taken as significant) using PRISM version 3.00 for Windows (GraphPad, San Diego, CA).
|
Results |
|---|
|
TopAbstractIntroductionMaterials and MethodsResultsDiscussionReferences |
|---|
). These cells generally appeared as tall and columnar, although in some sections, cells on the antimesometrial side of the lumen (adjacent to the decidual cells) appeared cuboidal and less organized than those on the mesometrial side of the lumen (Fig. 1A). Glands were also identified in decidualized sections, within the outer basal stroma, adjacent to the myometrium (data not shown). In all sections examined, glandular epithelium had lower immunoreactivity for the pan-cytokeratin antibody than the luminal epithelium.
|
), whereas luminal epithelium on the mesometrial side, away from the breaking-down decidualized stroma, remained intact and displayed no alterations in structural integrity. In some sections, undergoing breakdown, there was complete loss of luminal epithelium (data not shown). Glands in the outer basal region were not disturbed by endometrial breakdown.
Both luminal and glandular epithelium were readily detectable in restored/restoring endometrium (Fig. 1C). Luminal epithelium was often tortuous in appearance during early restoration with individual epithelial cells being columnar. In sections where restoration had progressed more fully, epithelium was indistinguishable from that in nonstimulated tissue (data not shown). Glandular epithelium was identified throughout the restored stromal compartment, no longer confined to the basal region of the endometrium. Sections in which the primary antibody was replaced with a nonspecific IgG were negative (Fig. 1D).
ER.
ER is the dominant ER in the endometrium. Strong nuclear staining for ER was abundant in the predecidual stromal cells in tissue in which decidualization was progressing (0 h, Fig. 1E), but the highly decidualized stromal cells (identified morphologically with very large nuclei) were consistently negative. The staining profile of nondecidualized stromal cells located in the basal region of the endometrium was variable between sections. In decidualized sections, some luminal epithelial cells were also identified as ER positive. ER was also detected during endometrial breakdown (Fig. 1F). In sections where breakdown of the central zone was still underway and reepithelialization of the uterine lumen had not proceeded, stromal cells in the basal region of the endometrium lying between the breakdown zone and inner myometrium were stained intensely for ER, with some luminal epithelial cell positive staining also apparent (data not shown). Once reepithelialization had proceeded and the endometrium was beginning to slough from the myometrium, staining was abundant both within the luminal epithelium itself and within the subepithelial stromal population (data not shown).
As endometrial restoration progressed, some luminal epithelial staining for ER was apparent (Fig. 1G). This staining extended to the subepithelial stromal cells, with glandular epithelial cells also identified as ER positive as the glands formed. Sections in which the primary antibody was replaced with a nonspecific IgG were negative (Fig. 1H).
Cellular proliferation
Ki67 was used to identify proliferating cells during endometrial breakdown and repair (Fig. 1, I–K). In decidualized sections (Fig. 1I), a high proportion of cells were Ki67 positive, reflecting the considerable expansion of the tissue. In particular, significant numbers of decidualizing/decidualized stromal cells stained positive (Fig. 1I), whereas proliferation was also apparent within luminal epithelial cells (Fig. 1I). Few if any glandular epithelial cells were Ki67 positive (data not shown).
In sections in which tissue breakdown was apparent, Ki67-positive cells were again abundant (Fig. 1J) but only in areas where breakdown was incomplete. The most marked proliferation in this tissue was identified in the layer of stromal cells within the basal region of the endometrium between the breakdown zone and myometrium. At this stage, a high proportion of luminal epithelial cells were also positive for Ki67. However, the distribution of these cells was not limited to the boundaries of the lumen (the area from which it is presumed the reepithelialization was extending), with cells throughout the lumen positive for Ki67 (data not shown). Few if any glandular epithelial cells were positive.
By the time restoration of the endometrium was nearly complete, only limited numbers of stromal cells remained Ki67 positive (Fig. IK). Ki67 was particularly marked within luminal epithelial cells, even once reepithelialization of the uterine surface appeared to have reached completion. Sections in which the primary antibody was replaced with a nonspecific IgG were negative (Fig. 1L).
-Smooth muscle actin
-Smooth muscle actin was used to delineate the myometrium from the adjacent stroma, although the antibody also detected pericytes/smooth muscle cells associated with some vessels (18). At all experimental times, two clear layers of myometrium were identifiable after smooth muscle actin staining (Fig. 1, M–O). Delineation of these was most important in tissue undergoing breakdown because it defined clearly a layer of stroma that was neither decidualized nor breaking down immediately adjacent to the inner myometrial layer (Fig. 1N). A clear network of blood vessels was identifiable in fully restored tissue (Fig. 1O) using this stain. Sections in which the primary antibody was replaced with a nonspecific IgG were negative (Fig. 1P).
Uterine weight
Uterine weight is influenced by estrogen exposure (19, 20). To monitor effects of estrogen depletion and replacement, the wet weights of the nonstimulated uterine horns were measured and expressed as a percentage of total body weight at 48 h. The mean uterine weight as a percentage of total body weight for the estrogen-added group (0.19 ± 0.01%) was significantly (P 0.001) higher than that of both the control and estrogen-deplete groups (0.06 ± 0.02 vs. 0.08 ± 0.01%). There was no significant difference in uterine weight as a percentage of total body weight between the latter two groups.
Assessment of estrogen-responsive genes: lactoferrin and PR mRNA expression
To assess the effects of estrogen depletion on estrogen-responsive genes, uterine lactoferrin and PR mRNA expression was quantitated by real time RT-PCR in selected control (n = 5), estrogen-deplete (n = 4), and estrogen-replacement (n = 3) animals (Fig. 2), with expression levels normalized for the housekeeping gene 18S. For both genes, some variation was noted between animals in all treatment groups but was most apparent for lactoferrin mRNA expression. For lactoferrin mRNA, control levels were not significantly different from the estrogen-deplete group (1.0 ± 0.75 compared with 3.1 ± 3.0 relative units) but significantly lower (P 0.05) than the estrogen-replacement group (176.8 ± 190.6 relative units). Similarly, when PR mRNA levels were quantitated, levels in the control group were not significantly different from those in the estrogen-deplete group (6.8 ± 5.9 vs. 6.0 ± 3.6 relative units) but were significantly lower (P 0.05) than in the estrogen-replacement group (87.2 ± 60.7 relative units).
|
Full endometrial restoration is not dependent on estrogen
In accord with previous studies (11), intra-animal variability at 48 h was low in this study (data not shown). Generally, this variation encompassed only two ordinal morphological stages and had little confounding effect on data. Similarly, as previously documented (11), little inter-animal variation is observed in this model at 48 h after P withdrawal, possibly reflecting tight regulation of restorative processes.
No morphological differences in the rate of endometrial breakdown were observed among control, estrogen-deplete, and estrogen-added animals (data not shown). Likewise, at 48 h after P withdrawal, when the endometrium is fully restored to a predecidualized state in control animals, no obvious differences between treatment groups were apparent (Fig. 3, A and B). Under all three treatment regimens, the regenerated endometrium morphologically resembled the nonstimulated uterine horn that had not undergone breakdown and repair (Fig. 3C). To quantify the morphological changes in the control and estrogen-deplete groups, tissue was scored by two independent observers according to a previously described scale (11) (Fig. 4). This revealed that for control animals, 85% of tissue had undergone complete restoration (stage 5), 7.5% were undergoing early restoration (stage 4), and 7.5% were still in the early stages of endometrial breakdown (stage 2). In the estrogen-deplete group, 88% of tissues showed complete restoration (stage 5), whereas 12% of tissues were undergoing early restoration (stage 4), demonstrating no significant difference in the rate or extent of endometrial restoration.
|
|
|
Discussion |
|---|
|
TopAbstractIntroductionMaterials and MethodsResultsDiscussionReferences |
|---|
At the peak of endometrial breakdown, undisturbed glandular structures in the basal region, portions of intact luminal epithelium, and a distinct layer of nondecidualized stromal basalis was observed. Both the luminal epithelium and basal layer of stromal cells were proliferating and ER positive. Although a clear demarcation between a basal and functional layer is apparent in the endometrium of women, the existence of a true basal layer in mice has not been clarified but is strongly suggested in this study. In primate models of the menstrual cycle (24, 25), strong epithelial proliferative activity and ER reactivity have been demonstrated in the deepest basalis compartment (basalis IV), where it has been suggested that stem-progenitor cells lie (24). Similarly, application of a label-retaining cell technique in mice has demonstrated the presence of endometrial stromal label-retaining cells near the endometrial-myometrial junction (26, 27). Although the presence of stem cells within our model remains to be established, the location of proliferative and ER activity within the stroma and epithelium of this basal zone, agree with human data that it is from such cells that endometrium regenerates (28, 29).
The limited number of proliferating stromal cells once restoration has proceeded may be indicative of limited hormonal support (note this staining was carried out in control ovariectomized animals); however, the finding that both stromal and epithelial components expressed ER at this time demonstrates the capability of these cells to react to estrogen (and potentially proliferate). The intense Ki67 staining in luminal epithelium at the time of restoration is presumed to reflect the rapid proliferation involved in reepithelialization of the uterine surface.
Interestingly, we found no difference in the uterine weight or expression of estrogen-responsive genes in our estrogen-deplete group compared with controls. Because provision of a soy-free diet and administration of letrozole would completely deplete estrogenic action in these animals, we had anticipated that differences would be detected between the two groups of ovariectomized mice. Indeed, Wang et al. (19) were able to detect significantly higher lactoferrin and PR expression levels in mice maintained on a high compared with low phytoestrogen-containing diet. Our inability to detect such differences in this study may be due to the lower levels of phytoestrogens in our control diet compared with Wang’s high phytoestrogen diet (compare >400 vs. 146 ppm). These results may also have been impacted by the high level of inter-animal variability in our measurement of expression of these genes and our relatively small n value.
The findings presented in this study can in part be extrapolated to what is known to occur in women and nonhuman primates. In women, the total generation time for postmenstrual surface reepithelialization is approximately 48 h after the first two menstrual days (29), a very rapid time course as in our mouse model. Also in agreement with our findings, in women, estrogen-dependent changes such as mitosis are not apparent before the time that the surface epithelium is restored, suggesting that the regenerative process must be influenced by normal wound-healing factors other than estrogen (29), with estrogenic effects on stromal expansion detected only on d 5 or 6 of the menstrual cycle coincident with rising estrogen levels (13). Importantly, a recent study (15) using a model of human endometrial tissue transplants in mice demonstrated that when hormone support is withdrawn, the transplanted tissue undergoes destruction accompanied by bleeding and that 7 d after the cessation of hormonal support, dense stromal cells and many small narrow glands were observed, both containing mitotic structures. Similar events of endometrial restoration have also been recorded in the nonhuman primate endometrium of the rhesus monkey, where epithelial restoration precedes stromal expansion with only limited hormonal support (21) and that glandular proliferation in the basal zone is independent of ovarian estrogen (14).
In summary, it is clear that in this mouse model, even very low levels of estrogen are not required for complete restoration of all cellular components of the endometrium after breakdown. The rapid and effective scar-free restoration of the human endometrium each menstrual cycle offers a unique system of tissue repair seen only elsewhere in the fetus. The demonstration that this repair is estrogen independent clearly suggests that other highly regulated wound-healing mechanisms must be at play. Successful identification of such mechanisms will aid in providing better treatment strategies for women with abnormal bleeding disorders and also provide novel information into mechanisms of scar-free repair with implications for reducing scarring after wound healing.
|
Footnotes |
|---|
Disclosure Statement: The authors of this manuscript have nothing to declare.
First Published Online July 19, 2007
Abbreviations: ER, Estrogen receptor ; P, progesterone; PBS–, PBS without Ca2+ and Mg2+; PR, progesterone receptor; TBS, Tris-buffered saline.
Received May 30, 2007.
Accepted for publication July 10, 2007.
|
References |
|---|
|
TopAbstractIntroductionMaterials and MethodsResultsDiscussionReferences |
|---|
-actin is reduced in the endometrium of women with progestin-only contraceptive breakthrough bleeding. Reproduction 15:78–84This article has been cited by other articles:
 |
|
 |
|
  E. Menkhorst, L. Salamonsen, L. Robb, and E. Dimitriadis IL11 Antagonist Inhibits Uterine Stromal Differentiation, Causing Pregnancy Failure in Mice Biol Reprod, May 1, 2009; 80(5): 920 - 927. [Abstract] [Full Text] [PDF]
|
|
|
|
 |
|
 T. J. Kaitu'u-Lino, D. J. Phillips, N. B. Morison, and L. A. Salamonsen A New Role for Activin in Endometrial Repair after Menses Endocrinology, April 1, 2009; 150(4): 1904 - 1911. [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 |
; ovariectomized plus normal diet plus vehicle injections), estrogen-deplete (
; ovariectomized plus soy-free diet plus letrozole injections), and estrogen-added (
; ovariectomized plus normal diet plus estradiol injections) groups. No significant difference in expression levels were identified for either gene in control vs. estrogen-deplete (E-free) group. Significantly higher (P < 0.05) expression levels were observed in the estrogen-added (E-added) group. Bars represent mean values.
, Stage 2; 
, stage 5. No significant difference between groups was identified.