Null Mutation of Kru¨ppel-Like Factor9/Basic Transcription Element ...

BIOLOGY OF REPRODUCTION 73, 472–481 (2005) Published online before print 25 May 2005. DOI 10.1095/biolreprod.105.041855

Null Mutation of Kru¨ppel-Like Factor9/Basic Transcription Element Binding Protein-1 Alters Peri-Implantation Uterine Development in Mice1 Michael C. Velarde,3 Yan Geng,3 Renea R. Eason,4 Frank A. Simmen,3,4 and Rosalia C.M. Simmen2,3,4 Department of Physiology and Biophysics,3 University of Arkansas for Medical Sciences and Arkansas Children’s Nutrition Center,4 Little Rock, Arkansas 72202 ABSTRACT

(E) and progesterone (P) [4, 5, 7, 8]. Detailed studies of the function of E and P in the mouse uterus have elucidated the distinct temporal and cell type-specific effects of these hormones [9, 10]. In particular, while E increases luminal epithelial (LE) cell proliferation during the first 2 days postcoitum (dpc; 0.5 dpc 5 day of vaginal plug), the synthesis of P by the corpus luteum beginning at 2.5 dpc initiates stromal (ST) cell proliferation, which is further enhanced by E at 3.5 dpc. Also at this time, P bound to its nuclear receptor PGR in ST inhibits, in a paracrine manner, the proliferation of LE, allowing it to differentiate [11, 12]. The balance of proliferation and differentiation in these endometrial cell types results from modifications in their respective gene expression programs, which underlie the transition from a refractory endometrium to one that is receptive to the implantation-ready embryo [5, 13]. Gene ablation in the mouse has revealed the importance of specific genes during the period surrounding implantation [14–17]. Of these, perhaps the most important is that encoding PGR since mice null for Pgr are totally infertile [18]. Interestingly, while ablation of the gene for PGR-A, the shorter (94 kDa) of the two PGR isoforms [19, 20], resulted in severe uterine hyperplasia and ovarian abnormalities, leading to infertility [21], ablation of the gene encoding PGR-B (114 kDa) did not result in an obvious uterine phenotype but rather in pregnancy-associated mammary gland abnormalities [22]; these suggest an obligatory role for PGR-A in reproduction. Nevertheless, distinct subsets of P-target genes for each individual PGR isoform were demonstrated in the uterine endometrium, suggesting a function for both PGR-A and PGR-B in this tissue [21]. Since P has an essential role in establishment and maintenance of pregnancy, factors that affect PGR function and/ or expression markedly alter uterine sensitivity to P and hence pregnancy success. Our laboratory has identified basic transcription element binding protein-1 (BTEB1), a member of the Sp1 family of transcription factors [23], as a novel PGR-interacting protein [24, 25] whose expression in the early pregnancy endometrium was coincident with PGR expression [25, 26]. In subsequent studies, we reported that null mutation of Bteb1 [27] resulted in subfertility in female mice due to multiple effects in the uterus [28]. In particular, we found uterine growth retardation in prepubertal and young adult females, fewer numbers of embryo implantation sites at 6.5 dpc and hence decreased litter size, and partial uterine insensitivity to P-stimulation. These results suggested that BTEB1 may function as a PGR coactivator during peri-implantation, when the P-dominant environment of the uterus necessitates a functional PGR to subserve the timely progression of endometrial cell proliferation and differentiation requisite for endometrial receptivity. The present study was designed to evaluate the mecha-

Female mice null for the basic transcription element binding protein-1 (Bteb1) gene have reduced numbers of implanting embryos. We hypothesized that the implantation defect, resulting in subfertility, is a consequence of developmental asynchrony between the embryo and uterine endometrium at peri-implantation. To address this, endometrium from wild-type (WT) and Bteb1(2/2) females at 0.5 to 5.5 days postcoitum (dpc) were evaluated for proliferation (BrdU labeling), apoptosis (TUNEL), and steroid hormone receptor expression (immunohistochemistry). Loss of BTEB1 did not affect serum estrogen (E) and progesterone (P) levels. In stroma (ST), the numbers of progesterone receptor (PGR) and HomeoboxA10 (HOXA10)-expressing cells were lower (3.5 and 4.5 dpc), while those of estrogen receptoralpha (ESR1) were higher (3.5 dpc), with Bteb1 ablation. The peak of proliferation in luminal epithelium (LE), glandular epithelium (GE), and ST was delayed, while the apoptotic index in all cell types was increased (2.5 dpc) in Bteb1(2/2) relative to WT mice. The numbers of PGR-positive ST cells was negatively correlated with LE proliferation in WT mice; this correlation was lost in Bteb1(2/2) mice and was not observed before 2.5 dpc for both genotypes. Proliferation and apoptosis in all endometrial compartments, as well as the numbers of PGR-, HOXA10-, and ESR1-expressing ST cells, were lower in Bteb1(2/2) relative to WT mice after ovariectomy and E 1 P treatment. Results suggest that BTEB1, by regulating ST PGR expression and transactivation, participates in the paracrine control of LE proliferation by PGR and thus is important for establishment of a receptive uterus critical for successful implantation.

female reproductive tract, implantation, progesterone, steroid hormone receptors, uterus

INTRODUCTION

An out-of-phase uterus is considered to be a major cause of infertility in mammals [1–4]. Hence, for a successful pregnancy to occur, the endometrium must develop in synchrony with the embryo, support its development to an implantation-competent blastocyst, and allow it to adhere and successfully implant [4–6]. To attain this receptive state, the endometrium undergoes dynamic changes in cell proliferation, differentiation, and apoptosis, processes that are controlled largely by the ovarian steroid hormones estrogen Supported by National Institutes of Health grant HD21961. Correspondence: Rosalia C.M. Simmen, University of Arkansas for Medical Sciences and Arkansas Children’s Nutrition Center, 1120 Marshall St., Little Rock, AR 72202. FAX: 501 364 3161; e-mail: [email protected]

1 2

Received: 10 March 2005. First decision: 6 April 2005. Accepted: 4 May 2005. Q 2005 by the Society for the Study of Reproduction, Inc. ISSN: 0006-3363. http://www.biolreprod.org

472

BTEB1 AND PERI-IMPLANTATION UTERINE DEVELOPMENT

473

FIG. 1. Lack of effect of Bteb1 ablation on serum levels of progesterone (P) and estrogen (E) during peri-implantation. The patterns of serum P (A) and E (C) levels as a function of day postcoitum (dpc) are shown for WT mice. Values are presented as mean 6 SEM (n 5 5–6 mice per dpc). Significant differences across 0.5 to 5.5 dpc were identified by one-way ANOVA, followed by Tukey test. Means with different superscripts differ at P , 0.05. Comparison of serum P (B) and E (D) levels in WT and Bteb1 null females. Values are presented as mean 6 SEM (n 5 5–6 mice per genotype per dpc). No significant differences between genotypes were observed (Student t-test).

nism by which BTEB1, acting in concert with PGR in the endometrial stroma, mediates endometrial receptivity. Toward this end, we examined the consequence of Bteb1 ablation on endometrial cell proliferation and apoptosis as well as on the expression of PGR and ESR1 in ST, the site of predominant BTEB1 expression [28], during peri-implantation. Further, we assessed the sensitivity of the endometrium to E 1 P, using mice of both Bteb1 genotypes on ovariectomy. MATERIALS AND METHODS

Animals Animal protocols were approved by the Institutional Animal Care and Use Committee of the University for Arkansas for Medical Sciences. WT and Bteb1(2/2) mouse lines were propagated and genotyped as previously described [28]. In timed pregnancy studies, WT and Bteb1(2/2) females of comparable ages (6–8 wk) were monitored for stage of the estrous cycle by the appearance of vaginal smears using a Protocol Hema stain kit (Fischer Scientific, Pittsburgh, PA) and were mated at proestrus of the second estrous cycle. The presence of a vaginal plug was considered as 0.5 day postcoitum (dpc). WT and Bteb1(2/2) females were mated with Bteb1(2/2) and WT males, respectively, to generate embryos of the same phenotype (heterozygote); this strategy precluded the possibility that any observed changes in the endometrium of Bteb1(2/2) mice might be due to lack of embryonic BTEB1 expression. For steroid hormone treatment studies, 6to 8-wk-old adult WT and Bteb1(2/2) females were subjected to bilateral ovariectomy and were then rested for 2 wk before receiving a single subcutaneous injection (0.1 ml) of sesame oil (vehicle) or estradiol-17b (E; 100 ng/mouse; Sigma Biochemicals, St. Louis, MO) 1 progesterone (P; 1 mg/mouse; Sigma) in sesame oil. Mice were killed 24 h after injection.

Steroid Hormone Measurements Serum E and P concentrations for WT and Bteb1(2/2) female mice (n 5 5–7 mice per genotype per dpc) were measured using Coat-A-Count Estradiol or Coat-A-Count Progesterone Diagnostic kits following the manufacturer’s protocols (Diagnostic Products Corp., Los Angeles, CA) as previously described [28]. All samples were assayed in one run.

Immunocytochemistry and TUNEL Assay Whole uteri from WT and Bteb1(2/2) females at peri-implantation (dpc 5 0.5, 1.5, 2.5, 3.5, 4.5, 5.5; n 5 5–6 mice per dpc per genotype) were fixed in 10% neutral-buffered formaldehyde overnight and embedded in paraffin. Procedures for the preparation of tissue sections and staining were previously described [28]. Prior to incubation with primary antibody, sections were treated with 3% hydrogen peroxide to quench endogenous peroxidase activity, microwaved sequentially (105 sec at power 10 and then 10 min at power 1) in Citra Plus (Biogenex, San Ramon, CA) to unmask antigen, and incubated in a blocking solution (VectaStain ABC kits; Vector Laboratories, Burlingame, CA) for 30 min to decrease nonspecific staining. Incubation with rabbit anti-human PGR antibody (1:650 dilution; Dako Corp., Carpinteria, CA) and goat anti-human HOXA10 antibody (1: 1000 dilution; Santa Cruz Biotech, Santa Cruz, CA) were carried out for 1 h and overnight, respectively, at 48C in a humidity chamber. Incubation with rabbit anti-human ESR1 antibody (1:500 dilution; Santa Cruz) for 1 h at room temperature followed previously described procedures [29], except for the antigen retrieval step and the use of VectaStain kits, as described previously. Following incubation with appropriate anti-goat or anti-rabbit secondary antibodies (Vectastain ABC kits), immunoreactive proteins were detected with the chromagen 3,39-diaminobenzidine tetrahydrochloride (Dako), and sections were then counterstained with hematoxylin, dehydrated, cleared, and coverslipped for examination under a microscope. A total of 1000 ST, 300 LE, and 200 GE cells, respectively, were counted on average from four randomly selected fields (2003 magnification) per slide and two slides for each mouse, with a total of five to six mice per dpc or treatment group used for the analyses. TUNEL assay to detect apoptotic cells was performed as previously described [30]. The same numbers of ST, GE, and LE cells, as indicated previously, were examined for positive staining per slide and two slides for each mouse, with a total of five to six mice per dpc or treatment group evaluated.

BrdU Labeling WT and Bteb1(2/2) females at the indicated pregnancy day (0.5–5.5 dpc; n 5 5–6 mice per genotype per dpc) or 2 wk after ovariectomy and steroid treatments (n 5 5–6 per genotype for sesame oil or E 1 P) were administered intraperitoneal injections of BrdU reagent (1 ml per 100 g of body weight; Zymed Laboratories, San Francisco, CA) 2 h before sacrifice. Uteri were fixed in 10% neutral-buffered formalin and processed

474

VELARDE ET AL.

FIG. 2. PGR (A and B) and HOXA10 (C and D) expression in uterine stroma (ST) of WT and Bteb1(2/2) mice at day postcoitum (dpc) 0.5 to 5.5. Values (A and C) are mean 6 SEM for n 5 5–6 mice per dpc per genotype. Significant differences were identified by two-way ANOVA, followed by Tukey test. Means with different superscripts differ at P , 0.05. Representative immunostaining for PGR- (B) and HOXA10- (D) positive cells at 3.5 and 4.5 dpc. GE, Glandular epithelium; LE, luminal epithelium; ST, stroma. Original magnification 3200.

for immunohistochemistry as described previously. BrdU uptake was detected using the BrdU-detection kit following the manufacturer’s protocol (Zymed). Immunopositive ST, LE, and GE cells were counted from a total of 1000, 300, and 200 cells, respectively, from four randomly selected fields (2003 magnification) per slide and two slides for each mouse, with a total of five to six mice per group used for analyses.

Data Analysis All data are presented as mean 6 SEM. The values were subjected to analysis by Student t-test, one-way ANOVA, or two-way ANOVA, as indicated under each figure legend. Differences between means in one-

way and two-way ANOVA were further analyzed by Tukey test. Correlation coefficients were calculated by Spearman rank order correlation. P values , 0.05 were considered statistically significant.

RESULTS

Serum E and P Levels in WT and Bteb1(2/2) Females During Early Pregnancy

Serum levels of E and P in WT mice from 0.5 to 5.5 dpc were in agreement with previously published data for mouse and rat strains [2, 31, 32]. An increase in P levels occurred from 2.5 to 3.5 dpc and then decreased slightly through to 5.5 dpc (Fig. 1A). By contrast, E levels did not significantly change from 0.5 dpc until 4.5 dpc, rising only at 5.5 dpc (Fig. 1C). WT and Bteb1(2/2) females showed comparable serum E (P . 0.5) and P (P . 0.5) levels at each of the time points examined (Fig. 1, B and D). Expression of PGR, HOXA10, and ESR1 in Uterine Stroma of WT and Bteb1(2/2) Mice

FIG. 3. Expression of ESR1 in uterine stroma (ST) of WT and Bteb1(2/2) mice at day postcoitum (dpc) 0.5 to 5.5. Values are mean 6 SEM for n 5 5–6 mice per dpc per genotype. Significant differences were identified by two-way ANOVA, followed by Tukey test. Means with different superscripts differ at P , 0.05.

The expression of PGR (A/B isoforms) and its downstream target HOXA10 were evaluated in uterine ST of pregnant (0.5–5.5 dpc) WT and Bteb1(2/2) mice by immunohistochemistry. In WT mice, the percentage of PGR-expressing ST cells increased temporally from 0.5 dpc to maximal levels at 3.5 dpc, and this level of expression was maintained through 5.5 dpc (Fig. 2, A and B). The percentage of PGR-expressing cells differed between WT and Bteb1(2/2) mice only at 3.5 and 4.5 dpc and was comparable for both genotypes at the other pregnancy days (Fig. 2, A and B). The temporal pattern of HOXA10 immunoreactivity followed that of PGR; the percentage of HOXA10 immunopositive cells increased from 0.5 dpc through 5.5 dpc and was higher in WT than Bteb1(2/2) mice at 3.5 and 4.5 dpc (Fig. 2, C and D). The contribution of BTEB1 to ST expression of ESR1

FIG. 4. BrdU labeling of luminal epithelium (LE), glandular epithelium (GE), and stroma (ST) of WT and Bteb12/2) mice at day postcoitum (dpc) 0.5 to 5.5. A) Values are mean 6 SEM for n 5 5–6 mice per dpc per genotype. Significant differences were identified by two-way ANOVA, followed by Tukey test. Means with different superscripts differ at P , 0.05. B) Representative immunostaining for BrdU. GE, Glandular epithelium; LE, luminal epithelium; ST, stroma. Original magnification 3200.

BTEB1 AND PERI-IMPLANTATION UTERINE DEVELOPMENT

475

FIG. 5. TUNEL-positive cells in the luminal epithelium (LE), glandular epithelium (GE), and stroma (ST) of WT and Bteb1(2/2) mice at day postcoitum (dpc) 0.5 to 5.5. A) Values are mean 6 SEM for n 5 5–6 mice per dpc per genotype. Significant differences were identified by two-way ANOVA, followed by Tukey test. Means with different superscripts differ at P , 0.05. B) Representative immunostaining for TUNEL-positive cells. GE, Glandular epithelium; LE, luminal epithelium; ST, stroma. Original magnification 3200.

476 VELARDE ET AL.

FIG. 7. Effect of estrogen (E) plus progesterone (P) on endometrial cell apoptosis in ovariectomized WT and Bteb1(2/2) mice. The percentage of TUNEL-positive cells in luminal epithelium (LE) (A), glandular epithelium (GE) (B), and stroma (ST) (C) of ovariectomized WT and Bteb1(2/2) mice treated with sesame oil (control) or E 1 P is presented as mean 6 SEM (n 5 5–6 mice per genotype per treatment group). Significant differences were identified by two-way ANOVA, followed by Tukey test. Means with different superscripts differ at P , 0.05.

FIG. 6. Effect of estrogen (E) plus progesterone (P) on endometrial cell proliferation in ovariectomized WT and Bteb1(2/2) mice. The percentage of BrdU-labeled cells in luminal epithelium (LE) (A), glandular epithelium (GE) (B), and stroma (ST) (C) of ovariectomized WT and Bteb1(2/2) mice treated with sesame oil (control) or E 1 P is presented as mean 6 SEM (n 5 5–6 mice per genotype per treatment group). Significant differences were identified by two-way ANOVA, followed by Tukey test. Means with different superscripts differ at P , 0.05.

BTEB1 AND PERI-IMPLANTATION UTERINE DEVELOPMENT

477

FIG. 8. Effect of estrogen (E) plus progesterone (P) on PGR (A), HOXA10 (B), and ESR1 (C) expression in uterine stroma (ST) of ovariectomized WT and Bteb1(2/2) mice. Values are presented as mean 6 SEM for n 5 5–6 mice per genotype per treatment group. Significant differences were identified by two-way ANOVA, followed by Tukey test. Means with different superscripts differ at P , 0.05.

478

VELARDE ET AL.

was examined by quantifying the percentage of ESR1-expressing cells in WT and Bteb1(2/2) mice. In contrast to that for PGR, the percentage of ESR1-expressing ST cells was relatively constant throughout early pregnancy (Fig. 3). However, similar to that for PGR, a difference in the percentage of ESR1-expressing ST cells between the two genotypes was observed at 3.5 dpc, although WT exhibited lower numbers of immunopositive cells than Bteb1(2/2) mice (Fig. 3). Effect of Bteb1 Null Mutation on Uterine Endometrial Cell Proliferation and Apoptosis

Endometrial cell proliferation in WT and Bteb1(2/2) females from 0.5 to 5.5 dpc was quantified from BrdU-labeled tissue sections by counting the percentage of BrdUpositive cells in LE, GE, and ST. In WT mice, the percentage of BrdU-labeled LE and GE cells was low at 0.5 and 1.5 dpc, increased to maximal levels at 2.5 dpc, dramatically decreased at 3.5 dpc to the levels found for 0.5 dpc, and remained low through 5.5 dpc (Fig. 4). By comparison, BrdU labeling of cells in Bteb1(2/2) mice displayed a different pattern; LE exhibited a delayed peak in cell proliferation that occurred at 3.5 dpc (vs. 2.5 dpc for WT mice), while proliferation in GE was significantly blunted at 2.5 dpc and persisted for 1 day longer (3.5 dpc) than in WT mice. The pattern of ST proliferation also differed between WT and Bteb1(2/2) mice (Fig. 4). While ST cells from WT mouse uteri showed a dramatic increase in percentage of BrdU labeling at 2.5 dpc, which was essentially maintained through 4.5 dpc and then diminished to 0.5 dpc levels by 5.5 dpc, those of Bteb1(2/2) mutants had a delayed increase in BrdU labeling, which peaked at 3.5 dpc. Given the effects of Bteb1 ablation on cell proliferation, the apoptotic status of endometrial cells in WT and Bteb1(2/2) mouse uteri was also evaluated (Fig. 5, A and B). For all cell types, the temporal pattern of apoptosis (measured as percentage of TUNEL-positive cells) was comparable between genotypes up to 3.5 dpc, except at 2.5 dpc, when Bteb1(2/2) mutants showed a higher percentage of apoptotic cells relative to WT. After 3.5 dpc, the pattern of apoptosis differed slightly among cell types: GE at 4.5 dpc showed higher apoptosis in WT than in Bteb1(2/2), while ST at 5.5 dpc had higher apoptosis in Bteb1(2/2) than in WT. On the other hand, LE from WT and Bteb1(2/2) mice between 3.5 and 5.5 dpc had undetectable apoptotic cells (Fig. 5A). The observed differences in cell proliferation and apoptosis occurring between 2.5 and 3.5 dpc for WT and Bteb1(2/2) mice were temporally correlated with rising serum P levels (which initiates at 2.5 dpc) and with the decrease in percentage of ST PGR-immunopositive cells in Bteb1(2/2) relative to WT mice at 3.5 dpc. To evaluate whether loss of BTEB1 might be linked to alterations in an existent paracrine relationship among these parameters, the correlation between PGR-positive ST cells and proliferation or apoptosis in endometrial LE and GE cells was examined as a function of Bteb1 genotype before (0.5–2.5 dpc) and during and subsequent to (2.5–5.5 dpc) the rise in serum P levels. Between 0.5 and 2.5 dpc, the percentage of PGRimmunopositive ST cells was not correlated with proliferation status in LE (P 5 0.442 and 0.304 for WT and Bteb1(2/2), respectively) or GE (P 5 0.852 and 0.304 for WT and Bteb1(2/2), respectively) regardless of Bteb1 genotype. By contrast, between 2.5 and 5.5 dpc, there was a strong negative correlation between ST expression of PGR

BTEB1 AND PERI-IMPLANTATION UTERINE DEVELOPMENT

and proliferation of LE (correlation coefficient r 5 20.552; P 5 0.031) and GE (r 5 20.556; P 5 0.030), respectively, in WT mice. Interestingly, this correlation was lost with Bteb1 null mutation in LE (r 5 20.074; P 5 0.783) and in GE (r 5 20.360; P 5 0.180). A negative correlation between PGR expression in ST and apoptosis in LE was noted in WT and Bteb1(2/2) mice before (WT: r 5 20.643; P 5 0.022; Bteb1(2/2): r 5 20.676; P 5 0.010) and after (WT: r 5 20.778; P , 0.001; Bteb1(2/2): r 5 20.766; P , 0.001) the rise in P levels. The negative correlation in WT mice (r 5 20.587; P 5 0.042) between PGR expression in ST and GE apoptosis was also observed in Bteb1(2/2) (r 5 20.648; P 5 0.016) mice at low P but was lost in WT (r 5 20.081; P 5 0.708) when compared to Bteb1(2/2) (r 5 20.638; P 5 0.010) mice, under conditions of high P. Effect of BTEB1 Null Mutation on Endometrial Cell Responsiveness to E and P

Given the altered proliferation in uterine LE, GE, and ST of Bteb1(2/2) relative to WT mice, in the background of comparable serum E and P profiles, we tested whether loss of responsiveness to exogenous E 1 P may underlie these differences between genotypes. Ovariectomy of WT and Bteb1(2/2) mice resulted in low basal levels of proliferation in all three uterine cell types for both Bteb1 genotypes. After 24 h of exposure to E 1 P, dramatic increases in BrdU labeling were observed for LE (by 30-fold) and GE (by 6fold) in WT uteri (Fig. 6, A and B). By comparison, GE cells of Bteb1(2/2) mouse uteri did not show any increase in BrdU labeling, while LE cells displayed an increase that was 50% lower than observed in WT. Stromal cells from OVX WT mouse uteri, although responsive to E 1 P, had a lower degree of induced proliferation when compared to LE and GE cells, respectively; by comparison, ST cells from corresponding Bteb1(2/2) uteri were unresponsive to this treatment (Fig. 6C). Uterine LE and GE cells of OVX WT mice had comparable levels of apoptosis that were significantly diminished with E 1 P treatments (Fig. 7). Ablation of Bteb1 did not influence their apoptotic responses to E 1 P. By contrast, uterine ST cells in OVX WT mice displayed higher levels of apoptosis than in those of OVX Bteb1(2/2) mice; E 1 P treatment reduced the level of apoptosis in WT uteri but had no effect in uteri of Bteb1(2/2) mice. To evaluate whether blunted (LE) or loss of (GE, ST) proliferative responses to E 1 P could be a function of BTEB1-mediated regulation of PGR or ESR1 expression in ST, the percentage of ST cells expressing PGR, HOXA10, and ESR1 was evaluated in control and E 1 P-treated cells from uteri of WT and Bteb1(2/2) mice. OVX WT and Bteb1(2/2) mice had comparable ST expression of PGR, HOXA10, and ESR1 (Fig. 8). Treatment of WT mice with E 1 P increased the numbers of ST cells expressing PGR (Fig. 8A), HOXA10 (Fig. 8B), and ESR1 (Fig. 8C). By comparison, in uteri of OVX Bteb1(2/2) mice, the percentage of ST cells expressing HOXA10 and ESR1 was not altered with E 1 P treatment, while that of PGR-expressing cells in ST was increased, albeit 50% lower than in uteri from OVX, E 1 P-treated WT mice (Fig. 8). DISCUSSION

The present report describes alterations in the patterns of proliferation and apoptosis and in the numbers of ESR1and PGR-expressing cells in uterine endometrium of early pregnant female mice as a consequence of null mutation of

479

the Bteb1 gene. These differences in peri-implantation uterine parameters between Bteb1(2/2) and WT mice were manifested in the absence of variations in their steroid hormone profiles and appeared to be partly a function of loss and/or blunted response to E 1 P due to decreased numbers of PGR- and ESR1-expressing ST cells on Bteb1 ablation. Results support a model in which the development of the periimplantation uterus involves ST BTEB1, with paracrine effects on LE cell proliferation but not on apoptosis, particularly influenced by the presence of ligand-bound ST PGR. Because the appropriate levels of ST proliferation as well as the timely progression of LE from proliferation to differentiation are essential for receptivity of the uterus to embryo implantation [2, 4, 6], our findings suggest that disruption in the temporal occurrence and alterations in the proliferative status of uterine LE, GE, and ST cells with Bteb1 ablation could result in developmental asynchrony between the endometrium and the implantation-ready embryo, likely leading to the demonstrated subfertility phenotype of Bteb1(2/2) mice [28]. Our hypothesis of the involvement of ST BTEB1, acting in concert with ligand-bound PGR in the control of ST and LE cell proliferation at peri-implantation, is supported by a number of observations from this and previous studies. First, we have shown that BTEB1 expression in early pregnant mouse uteri was predominantly in ST, with undetectable expression in LE [28]. Moreover, ST expression of BTEB1 was found to be dependent on pregnancy day, with 3.5 dpc showing higher level of expression relative to 0.5 dpc, respectively [28], a pattern paralleled by ST PGR (this study). Second, LE proliferation during early pregnancy in the mouse has been reported to be negatively regulated by PGR in the ST [33, 34]. Consistent with this, we found that at 2.5 and 3.5 dpc, when the most dramatic differences in LE proliferation between WT and Bteb1(2/2) were observed, there was no difference in the percent of PR-expressing LE cells in uteri of mice of either Bteb1 genotype (data not shown). Third, we report here that the number of PGR-expressing ST cells is negatively correlated with LE proliferation at or after 2.5 dpc (coincident with the initiation in rising P levels), and this correlation was lost with Bteb1 ablation. Last, we observed that the number of ST cells expressing PGR was greater in both pregnant and OVX, E 1 P-treated WT mice, relative to Bteb1(2/2) mice, suggesting that establishment and/or maintenance of threshold levels of functional PGR by BTEB1 in ST is important for control of LE proliferation. Interestingly, the control of apoptosis by BTEB1 at peri-implantation appeared to be independent of P-bound PGR and varied with cell type, suggesting that although cell proliferation and cell death may be coordinately regulated by BTEB1 during early pregnancy, distinct mechanisms are involved. The negative correlation between ST PGR and LE proliferation that was demonstrated in WT but not Bteb1(2/2) mice was observed only at or after the rise in serum P levels. Prior to this time (earlier than 2.5 dpc), this correlation was nonexistent, regardless of BTEB1 genotype. In particular, maximal proliferation of LE cells in WT occurred at 2.5 dpc, when ST PGR and serum P levels are coincidentally low, suggesting that under conditions of low (or no) PGR activation by its ligand P, BTEB1 in ST positively affected LE cell proliferation. Consistent with this, we have shown in an earlier study [35] that stable transfection of BTEB1 expression construct in a human endocarcinoma cell line Hec-1A lacking PGR resulted in increased cell proliferation coincident with up-regulated ex-

480

VELARDE ET AL.

pression of positively acting growth regulatory genes [35, 36]. Moreover, we also demonstrated that cotransfection of PGR expression constructs in the Hec-1A clonal line expressing high levels of BTEB1 resulted in increased promoter activity of a P-dependent gene, uteroferrin, whose expression is associated with uterine differentiation [24]. Hence, we suggest that BTEB1 plays a dual role in the control of uterine development; under a PGR/P-independent cellular context, BTEB1 mainly exhibits proliferative activity, while under conditions where levels of activated (functional) PGR are high, BTEB1 interacts with the PGR/ P complex to induce cell differentiation. The latter is supported by our present finding that ST expression of the PGR/P-downstream target gene Hoxa10, which encodes a homeodomain transcription factor important for development of endometrial receptivity [14], is altered in Bteb1(2/2) mice only at 3.5 and 4.5 dpc. Thus, the regulation of uterine development at peri-implantation, specifically of the LE, may involve ST-derived factors whose temporal expression is regulated by BTEB1 but to differing extents by PGR/P. The ability of BTEB1 to affect the proliferation as well as apoptosis of the three major endometrial cell types at peri-implantation and in OVX mice treated with E 1 P demonstrates that this transcription factor affects endometrial sensitivity to steroid hormones. The latter is in part due to the ability of BTEB1 to influence the numbers of ESR1- and PGR-expressing cells; this was demonstrated here for ST cells, which exhibited differing numbers of PGR (lower)- and ESR1 (higher or lower)-expressing cells in Bteb1(2/2) relative to corresponding WT mice. Interestingly, this activity of BTEB1 appears to be P dependent since in OVX mice injected only with sesame oil or in mice before the P surge, which initiates at 2.5 dpc, the percentage of ESR1- and PGR-expressing cells did not differ with presence or absence of Bteb1. These results suggest the existence of a regulatory loop involving PGR, ESR1, and BTEB1 and highlight the importance of additional studies to carefully investigate this linkage, given its implication in the control of physiological and pathophysiological states regulated by P/PGR in numerous P-target tissues [7, 37]. Several interesting observations reported here provide a framework for mechanistically addressing BTEB1 involvement in PGR/P-dependent processes. The first relates to the ST PGR isoform functionally interacting with BTEB1 to elicit PGR/P inhibition of LE cell proliferation. It has been reported that although both PGR isoforms are expressed in uterine ST, the levels of PGR-A are higher than PGR-B [38]. Since ablation of Pgr-A but not Pgr-B resulted in uterine hyperplasia and severely compromised implantation [21], our findings of aberrant uterine LE proliferation at 3.5 dpc in Bteb1(2/2) mice coincident with decreased PGR expression levels imply a role for BTEB1 in PGR-A-mediated control of LE proliferation. The second relates to the narrow window of peri-implantation (2.5 and 3.5 dpc) at which BTEB1 effects on proliferation, apoptosis, and PGR and ESR1 expression were manifested. After 3.5 dpc, with a few exceptions, we found no major differences in these biological parameters for the two genotypes, suggesting either the loss of the relative importance of BTEB1 postimplantation and/or the expression of other nuclear proteins that can compensate for BTEB1 function at this time [24, 28]. Third, given that PGR has many other coactivators implicated in its transactivity, which include the SRC family of proteins, p300/CBP, and p300/CBP-associated factor [39, 40], it is of interest to determine the cellular context

under which BTEB1 is preferentially used by PGR among all other comparably expressed nuclear proteins. Finally, the observation that in OVX mice treated with E 1 P, absence of BTEB1 resulted in complete refractoriness of GE and ST cells to these steroid hormones (as measured by proliferation index), in contrast to LE cells, which showed sensitivity, albeit blunted, raises questions on the molecular mechanisms underlying cell type-specific regulation of proliferation involving BTEB1. In this regard, it is noteworthy that GE and ST express BTEB1, in contrast to LE, which lacks BTEB1 expression [28]. We suggest that insights from these observations may provide a strategy for ameliorating subinfertility in a cohort of women whose etiology is not dependent on abnormal ovarian function [41, 42]. The normal differentiation of endometrial glands is essential for pregnancy since the timely synthesis and secretion of growth factors and other regulatory proteins by the differentiated GE are requisite for uterine receptivity and other functions of the LE [4, 43]. Although our results also showed that loss of BTEB1 function affected the proliferation, apoptosis, and steroid receptor gene expression in GE to almost the same extent as observed for LE, the mechanisms underlying these changes are likely distinct for each cell type since GE, unlike LE, is a site of BTEB1 expression [28]. Given that GE coexpress BTEB1 and PGR during early pregnancy [44] and that PGR-B is considered to be the predominant PGR isoform involved in endometrial glandular gene and secretory activity [38], further studies to evaluate the consequence of loss of functional PGR-P/ BTEB1 interactions in GE on Bteb1 null mutation could provide insights into the relative paracrine contributions of GE and ST on LE function. In summary, we have demonstrated that the subfertility phenotype of Bteb1(2/2) mice characterized by lower numbers of implanting embryos [28] may be due in part to alterations in the temporal progression and level of proliferation of endometrial LE cells during peri-implantation, likely leading to developmental asynchrony of the uterine endometrium and the implantation-ready embryo. These alterations, which were manifested within a narrow window of uterine development (2.5–3.5 dpc), are correlated with decreased numbers of PGR-expressing ST cells, suggesting a possible signaling pathway between ST and LE involving BTEB1. Thus, BTEB1 constitutes a novel molecular mediator of endometrial receptivity via its PGR coregulator function. Future studies to mechanistically evaluate the potential regulatory loop linking BTEB1, PGR, and ESR1 as well as to identify the paracrine factors elicited by ST cells in response to BTEB1, acting alone and in concert with PGR/P, will provide interesting insights into the intricate and complex processes essential for uterine receptivity and pregnancy success. ACKNOWLEDGMENT We thank Anne Wiley and Frank Bartol (Auburn University) for advice on HOXA10 immunolabeling.

REFERENCES 1. Dickmann Z, Noyes RW. The fate of ova transferred into the uterus of the rat. J Reprod Fertil 1960; 1:197–212. 2. Psychoyos A. Hormonal control of implantation. Vitam Horm 1973; 31:201–256. 3. Li TC, Dockery P, Cooke ID. Endometrial development in the luteal phase of women with various types of infertility: comparison with women of normal fertility. Hum Reprod 1991; 6:325–330.

BTEB1 AND PERI-IMPLANTATION UTERINE DEVELOPMENT 4. Carson DD, Bagchi I, Dey SK, Enders AC, Fazleabas AT, Lessey BA, Yoshinaga K. Embryo implantation. Dev Biol 2000; 223:217–237. 5. Dey SK, Lim H, Das SK, Reese J, Paria BC, Daikoku T, Wang H. Molecular cues to implantation. Endocr Rev 2004; 25:341–373. 6. Abrahamson PA, Zorn TMT. Implantation and decidualization in rodents. J Exp Zool 1993; 266:603–628. 7. Clarke CL, Sutherland RL. Progestin regulation of cellular proliferation. Endocr Rev 1990; 11:266–301. 8. Pampfer S, Donnay I. Apoptosis at the time of embryo implantation in mouse and rat. Cell Death Differ 1999; 6:533–545. 9. DeMayo FJ, Zhao B, Takamoto N, Tsai SY. Mechanisms of action of estrogen and progesterone. Ann N Y Acad Sci 2002; 955:48–59. 10. Aitken RJ. The hormonal control of implantation. Ciba Found Symp 1978; 64:53–83. 11. Kurita T, Young P, Brody JR, Lydon JP, O’Malley BW, Cunha GR. Stromal progesterone receptors mediate the inhibitory effects of progesterone on estrogen-induced uterine epithelial cell deoxyribonucleic acid synthesis. Endocrinology 1998; 139:4708–4713. 12. Cooke PS, Buchanan DL, Young P, Setiawan T, Brody J, Korach KS, Taylor J, Lubahn DB, Cunha GR. Stromal estrogen receptors mediate mitogenic effects of estradiol on uterine epithelium. Proc Natl Acad Sci U S A 1997; 94:6535–6540. 13. Reese J, Das SK, Paria BC, Lim H, Song H, Matsumoto H, Knudtson KL, DuBois RN, Dey SK. Global gene expression analysis to identify molecular markers of uterine receptivity and embryo implantation. J Biol Chem 2001; 276:44137–44145. 14. Benson GV, Lim H, Paria BC, Satokata I, Dey SK, Maas RL. Mechanisms of reduced fertility in Hoxa-10 mutant mice; uterine homeosis and loss of maternal Hoxa-10 expression. Development 1996; 122: 2687–2696. 15. Gendron RL, Paradis H, Hseh-Li HM, Lee DW, Potter SS, Markoff E. Abnormal uterine stromal and glandular function associated with maternal reproductive defects in Hoxa-11 null mice. Biol Reprod 1997; 56:1097–1105. 16. Lim H, Paria BC, Das SK, Dinchuk JE, Langenbach R, Trzaskos JM, Dey SK. Multiple female reproductive failures in cyclooxygenase 2deficient mice. Cell 1997; 91:197–208. 17. Stewart CL, Kaspar P, Brunet LJ, Bhatt H, Gadi I, Kontgen F, Abbondanzo SJ. Blastocyst implantation depends on maternal expression of leukaemia inhibitory factor. Nature 1992; 359:76–79. 18. Lydon JP, DeMayo FJ, Funk CR, Mani SK, Hughes AR, Montgomery CA Jr, Shyamala G, Conneely OM, O’Malley BW. Mice lacking progesterone receptor exhibit pleiotropic reproductive abnormalities. Genes Dev 1995; 9:2266–2278. 19. Kastner P, Krust A, Turcotte B, Stropp U, Tora L, Gronemeyer H, Chambon P. Two distinct estrogen-regulated promoters generate transcripts encoding the two functionally different human progesterone receptor forms A and B. EMBO J 1990; 9:1603–1614. 20. Schott DR, Shymala G, Schneider W, Parry G. Molecular cloning, sequence analyses, and expression of complementary DNA encoding murine progesterone receptor. Biochemistry 1991; 30:7014–7020. 21. Mulac-Jericevic B, Mullinax RA, DeMayo FJ, Lydon JP, Conneely OM. Subgroup of reproductive functions of progesterone mediated by progesterone receptor-B isoform. Science 2000; 289:1751–1754. 22. Mulac-Jericevic B, Lydon JP, DeMayo FJ, Conneely OM. Defective mammary gland morphogenesis in mice lacking the progesterone receptor B isoform. Proc Natl Acad Sci U S A 2003; 100:9744–9749. 23. Imataka H, Sogawa K, Yasumoto K, Kikuchi Y, Sasano K, Kobayashi A, Hayami M, Fujii-Kuriyama Y. Two regulatory proteins that bind to the basic transcription element (BTE), a GC box sequence in the promoter region of the rat P-4501A1 gene. EMBO J 1992; 11:3663– 3671. 24. Zhang D, Zhang XL, Michel FJ, Blum JL, Simmen FA, Simmen RCM. Direct interaction of the Kru¨ppel-like family (KLF) member, BTEB1, and PR mediates progesterone-responsive gene expression in endometrial epithelial cells. Endocrinology 2002; 143:62–73. 25. Zhang XL, Zhang D, Michel FJ, Blum JL, Simmen FA, Simmen RCM. Selective interactions of Kru¨ppel-like factor 9/basic transcription element-binding protein with progesterone receptor isoforms A and B determine transcriptional activity of progesterone-responsive

26.

27.

28.

29.

30.

31.

32.

33. 34.

35.

36.

37. 38. 39. 40. 41. 42. 43. 44.

481

genes in endometrial epithelial cells. J Biol Chem 2003; 278:21474– 21482. Wang Y, Michel FJ, Wing A, Simmen FA, Simmen RCM. Cell-type expression, immunolocalization, and deoxyribonucleic acid-binding activity of basic transcription element binding transcription factor, an Sp-related family member in porcine endometrium of pregnancy. Biol Reprod 1997; 57:707–714. Morita M, Kobayashi A, Yamashita T, Shimanuki T, Nakajima O, Takahashi S, Ikegami S, Inokuchi K, Yamashita K, Yamamoto M, Fujii-Kuriyama Y. Functional analysis of basic transcription element binding protein by gene targeting technology. Mol Cell Biol 2003; 23:2489–2500. Simmen RCM, Eason RR, McQuown JR, Linz AL, Kang TJ, Chatman L Jr, Till SR, Fujii-Kuriyama Y, Simmen FA, Oh SP. 2004. Subfertility, uterine hypoplasia, and partial progesterone resistance in mice lacking the Kruppel-like factor 9/basic transcription element-binding protein-1 (Bteb1) gene. J Biol Chem 2004; 279:29286–29294. Pendaries C, Darblade B, Rochaix P, Krust A, Chambon P, Korach KS, Bayard F, Arnal JF. The AF-1 activation-function of ER-a may be dispensable to mediate the effect of estradiol on endothelial NO production in mice. Proc Natl Acad Sci U S A 2002; 99:2205–2210. Eason RR, Velarde MC, Chatman L Jr, Till SR, Geng Y, Ferguson M, Badger TM, Simmen RCM. Dietary exposure to whey proteins alters rat mammary gland proliferation, apoptosis, and gene expression during postnatal development. J Nutr 2004; 134:3370–3377. Parkening TA, Lau IF, Saksena SK, Chang MC. Circulating plasma levels of pregnenolone, progesterone, estrogen, luteinizing hormone, and follicle stimulating hormone in young and aged C57BL/6 mice during various stages of pregnancy. J Gerontol 1978; 33:191–196. Watson J, Anderson FB, Alam M, O’Grady JE, Heald PJ. Plasma hormones and pituitary luteinizing hormone in the rat during the early stages of pregnancy and after post-coital treatment with tamoxifen (ICI 46,474). J Endocrinol 1975; 65:7–17. Martin L, Das RM, Finn CA. The inhibition by progesterone of uterine epithelial proliferation in the mouse. J Endocrinol 1973; 57:549– 554. Tong W, Pollard JW. Progesterone inhibits estrogen-induced cyclin D1 and cdk4 nuclear translocation, cyclin E- and cyclinA-cdk2 kinase activation, and cell proliferation in uterine epithelial cells in mice. Mol Cell Biol 1999; 19:2251–2264. Zhang XL, Simmen FA, Michel FJ, Simmen RCM. Increased expression of the Zn-finger transcription factor BTEB1 in human endometrial cells is correlated with distinct cell phenotype, gene expression patterns, and proliferative responsiveness to serum and TGF-b1. Mol Cell Endocrinol 2001; 181:81–96. Simmen RCM, Zhang XL, Michel FJ, Min SH, Zhao G, Simmen FA. Molecular markers of endometrial epithelial cell mitogenesis mediated by the Sp/Kru¨ppel-like factor BTEB1. DNA Cell Biol 2002; 21:115– 118. Inoue M. Current molecular aspects of the carcinogenesis of the uterine endometrium. Int J Gynecol Cancer 2002; 11:339–348. Mote PA, Balleine RL, McGowan EM, Clarke CL. Heterogeneity of progesterone receptors A and B expression in human endometrial glands and stroma. Hum Reprod 2000; 15(suppl 3):48–56. McKenna NJ, Xu J, Nawaz Z, Tsai SY, Tsai MJ, O’Malley BW. Nuclear receptor coactivators: multiple enzymes, multiple complexes, multiple functions. J Steroid Biochem Mol Biol 1999; 69:3–12. Freedman LP. Multimeric coactivator complexes for steroid/nuclear receptors. Trends Endocrinol Metab 1999; 10:403–407. Sharkey AM, Smith SK. The endometrium as a cause of implantation failure. Best Pract Res Clin Obstet Gynaecol 2003; 17:289–307. Li TC, Iqbal T, Anstie B, Gillham J, Amer S, Wood K, Laird S. An analysis of the pattern of pregnancy loss in women with recurrent miscarriage. Fertil Steril 2002; 78:1100–1106. Simmen FA, Simmen RCM. Peptide growth factors and protooncogenes in mammalian conceptus development. Biol Reprod 1991; 44: 1–5. Tan J, Paria BC, Dey SK, Das SK. Differential uterine expression of estrogen and progesterone receptors correlates with uterine preparation for implantation and decidualization in the mouse. Endocrinology 1999; 140:5310–5321.