| HOME | HELP | FEEDBACK | SUBSCRIPTIONS | ARCHIVE | SEARCH | TABLE OF CONTENTS |
|
|
|
|||||||
|
|||||||||
Regular Article |
a Laboratory of Genetics and Physiology, National Institute of Diabetes, Digestive and Kidney Diseases, National Institutes of Health, Bethesda, Maryland 20892
b Eppley Institute for Research in Cancer and Allied Diseases, University of Nebraska Medical Center, Omaha, Nebraska 68198
c Animal Sciences Unit, University of Missouri, Columbia, Missouri 65211
d Department of Epidemiology and Preventive Medicine, University of Maryland Medical School, Baltimore, Maryland 21201
 |
ABSTRACT |
|---|
 TOP ABSTRACT INTRODUCTIONMATERIALS AND METHODSRESULTSDISCUSSIONNOTE ADDED IN PROOFREFERENCES |
|---|
apoptosis, corpus luteum, follicle, granulosa cells
|
INTRODUCTION |
|---|
|
TOPABSTRACTINTRODUCTIONMATERIALS AND METHODSRESULTSDISCUSSIONNOTE ADDED IN PROOFREFERENCES |
|---|
Deletion of the bcl-x gene from the mouse results in lethality at approximately Embryonic Day 12.5 (E12.5) [7]. Thus, it is not possible to establish the role of Bcl-x in the ontogeny of ovaries in the postnatal mouse. To test whether Bcl-x is required for the maintenance of follicles, granulosa and luteal cells, and oocytes, we have inactivated the bcl-x gene in these cells using Cre-loxP-based recombination. We have generated mice that carry loxP sites in the promoter and second intron of the bcl-x gene [6] as well as mice that express the Cre recombinase in several cell types in the ovary [8].
In addition, we have used the conditional Bcl-x-null mice to test the hypothesis that bcl-x gene transcription in the ovary is activated by prolactin (Prl) through Stat5. Prolactin is released during estrus and is thought to be obligatory for the maintenance of luteal function. Previous studies have shown that Prl receptor (PrlR) levels increase when granulosa cells differentiate, and that PrlR levels are highest in luteal cells [9]. Previous studies have also shown that PrlR-mediated activation of Stat5 coincides with luteinization [9], and that transcription of the bcl-x gene can be activated by Stat5 in some cell lines [10, 11]. To our knowledge, however, no in vivo studies have directly tested whether bcl-x gene transcription in the ovary is activated by Prl through Stat5. Thus, we used mice deficient in Bcl-x, PrlR, and Stat5 to test this hypothesis.
|
MATERIALS AND METHODS |
|---|
|
TOPABSTRACTINTRODUCTIONMATERIALS AND METHODSRESULTSDISCUSSIONNOTE ADDED IN PROOFREFERENCES |
|---|
Mice were housed in a 12L:12D photoperiod at 22°C and 80% relative humidity with food and water ad libitum. All animal procedures were conducted in accordance with the National Institutes of Health (NIH) Guide for the Care and Use of Laboratory Animals and were approved by the Animal Care and Use Committees of the University of Maryland and the NIH. These animals were covered under the NIH protocol K990631.
Transgenic and Gene Knock-Out Mice
Generation of mice carrying an MMTV-Cre transgene and of mice carrying a bcl-x gene with loxP sequences (floxed gene) has been reported previously [6, 8]. Polymerase chain reaction (PCR) was used to distinguish the floxed bcl-x gene from the wild-type allele using primers 5'-CGGTTG-CCTAGCAACGGGGC-3' and 5'-CTCCCACAGTGGAGACCTCG-3'. Reaction conditions were 4 min at 96°C (1 cycle); 45 sec at 96°C, 45 sec at 58°C, and 1 min at 72°C (30 cycles); and 5 min at 72°C (1 cycle). The PCR products were 300 base pairs (bp) for the floxed allele and 200 bp for the wild-type allele. The presence of the Cre-mediated recombined bcl-x allele (resulting in a bcl-x-null allele) was detected using primers 5'-AATGGCCAGTACTAGTGAACC-3' and 5'-TCAGAAGCCGCAAT-ATCCCC-3' with the same cycling conditions as described above. The bcl-x-null allele was presented by a PCR product of 150 bp. A PCR method to genotype the LacZ reporter in the targeted Rosa26 locus has been published previously [12].
The PrlR-null [13] and Stat5-null [14] mice were obtained from Drs. Paul Kelly (Paris, France) and Jim Ihle (Memphis, TN), respectively. In the Stat5-null mice, both the Stat5a and 5b genes had been inactivated. The Stat5- and PrlR-null mice were used for superovulation studies, Western blot analysis of ovaries, histological analysis of ovaries, and TUNEL assay of ovaries from pseudopregnant mice. These mice were infertile and not bred.
PCR assays were used to determine the time during mouse development at which Cre-mediated bcl-x gene inactivation and deletion in granulosa cells occurred. Four-week-old mice that carried two floxed bcl-x alleles and the MMTV-Cre (A) transgene (referred to as bcl-x fl/fl; Cre) as well as bcl-x fl/+ mice were injected with 5 IU of eCG i.p. Ovaries from three bcl-x fl/fl; Cre and three bcl-x fl/+ mice were harvested 48 h later. Large antral follicles were punctured with a needle to collect granulosa cells. Between 50 and 100 granulosa cells were lysed in a solution consisting of 1% (v/v) NP-40, 6 mM dithiothreitol, and 10 U/µl of RNase inhibitor (Promega, Madison, WI). The PCR assays were performed using 1 µl of lysed granulosa cells for the floxed and recombined Bcl-x allele as described above. Whole ovaries were collected from mice on E15.5, as determined by plug formation, and PCR was performed on isolated DNA for the presence of the recombined bcl-x allele. Ovaries from four bcl-x fl/fl; Cre and three bcl-x fl/fl fetuses were analyzed. Embryonic Day 15.5 was chosen because the MMTV-LTR is activated during that time [15].
Superovulation
To determine any defect in ovulation, mice were superovulated by i.p. injection of 5 IU of eCG followed 48 h later by i.p. injection of 5 IU of hCG. Twenty-four hours after hCG injection, ovaries and oviducts were collected, and oviducts were punctured with a needle to release oocyte and cumulus cells. Released oocytes were then counted. Three bcl-x fl/fl, three bcl-x fl/fl; Cre, four Stat5 -/-, four Stat5 +/-, four PrlR -/-, and three wild-type mice were used in these experiments. Data are presented as mean ± SEM.
Histological Evaluation of Ovaries
Ovaries were fixed in Bouin solution overnight and then washed in 70% ethanol. Ovaries were embedded in paraplast (VWR Scientific, Buffalo Grove, IL) and then serial sectioned (thickness, 8 µm) through the entire tissue, mounted on glass slides, and stained with Weigert hematoxylin-picric acid methylene blue. Every 10th section was analyzed for primordial, primary, and preantral/antral follicle numbers. Three virgin mice each were used for the 5-wk bcl-x fl/fl; Cre and bcl-x fl/fl as well as for the 3-mo bcl-x fl/fl time points. Six virgin mice were used for the 3-mo bcl-x fl/fl; Cre time point. All data are presented as mean ± SEM. The X-Gal staining technique has been reported previously [8].
Tissues were fixed in 10% buffered formalin overnight for Bcl-x immunohistochemistry. The Bcl-x antibody used for staining (SC-634; Santa Cruz Biotechnology, Santa Cruz, CA) was used at a 1:100 (v/v) dilution. Biotinylated secondary antibodies were detected using Vectastain Elite ABC kit (Vector, Burlingame, CA). The slides were counterstained with hematoxylin.
To identify nuclei with DNA strand breaks, TUNEL was performed using the Apoptag protocol (Intergen, Purchase, NY). Slides were counterstained with 0.1% methyl green for 1 min. Pseudopregnancy was induced in mature PrlR-null, Stat5-null, bcl-x fl/fl; Cre, and bcl-x fl/fl virgin mice (i.p. injection of 5 IU of eCG followed 48 h later by i.p. injection of 5 IU of hCG every 24 h for 5–7 days). At least three mice of each genotype were analyzed.
Western Blot Analysis
Tissues were homogenized in ice-cold lysis buffer consisting of 40 mM Tris-HCl (pH 8.0), 280 mM NaCl, 20% glycerol, 1% NP-40, 4 mM EDTA, 20 mM NaF, 100 µg/ml of PMSF, 40 µg/ml of aprotinin, 40 µg/ml of leupeptin, and 2 mM Na-vanadate. Protein concentration was determined using the BCA Protein Assay Kit (Pierce, Rockford, IL) with BSA as a standard. Forty micrograms of protein were loaded per lane, separated by SDS-PAGE on a 16% gel, and transferred to a polyvinylidene difluoride membrane (Invitrogen, Carlsbad, CA). The membranes were blocked overnight at 4°C with TBST (50 mM Tris-HCl [pH 7.5], 0.15 M NaCl, and 0.05% Tween-20) containing 3% nonfat dry milk. They were incubated for 1 h with rabbit anti-Bcl-x (SC-634), anti-Bax (SC-493; Santa Cruz Biotechnology), anti-Bcl-2 (no. 06-474; Upstate Biotechnology, Lake Placid, NY), anti-Stat5a, or anti-Stat5b serum [16], diluted at 1:500, 1:500, 1:500, 1:20 000, or 1:10 000, respectively, with 3% nonfat dry milk in TBST. Horseradish peroxidase-conjugated goat anti-rabbit immunoglobulin G was used as the secondary antibody and reacted at 1:5000 with 3% nonfat dry milk in TBST for 30 min. Bands were visualized using SuperSignal West Pico Chemiluminescent Substrate (Pierce). Protein was collected from pseudopregnant ovaries (i.p. injection of 5 IU of eCG followed 48 h later by i.p. injection of 5 IU of hCG every 24 h for 5–7 days) from three PrlR-null, PrlR +/-, wild-type, Stat5-null, and Stat5 +/- mature mice each for Western blot analysis. Protein was also collected from four 4-wk-old virgin bcl-x fl/fl; Cre and four bcl-x fl/fl mice.
Progesterone Levels
Progesterone levels were measured by RIA using Coat-A-Count (Diagnostic Products Corporation, Los Angeles, CA). Mice were anesthetized, and blood was collected by phlebotomy from the retro-orbital plexus. Serum was separated from cells by centrifugation and stored at -80°C. Serum from three bcl-x fl/fl; Cre and three bcl-x fl/fl mice was analyzed. Data are presented as mean ± SEM.
Statistical Analysis
The mean number of ovarian follicles was calculated using ovaries from at least three different animals. The mean number of oocytes recovered following superovulation was determined using at least three different mice for each genotype. The mean progesterone levels were calculated using three different animals. Differences in follicle numbers, recovered oocytes, and progesterone levels were evaluated by a t-test for equality of means, with statistical significance being assigned at P < 0.05.
|
RESULTS |
|---|
|
TOPABSTRACTINTRODUCTIONMATERIALS AND METHODSRESULTSDISCUSSIONNOTE ADDED IN PROOFREFERENCES |
|---|
Because Bcl-x-null mice die at approximately E12.5, we deleted the bcl-x gene in distinct cell types using a Cre-loxP recombination approach. One line of transgenic mice that expresses Cre recombinase under control of the MMTV-LTR exhibits Cre activity in the ovary [8]. We have now used in situ and PCR assays to establish the cell types within the ovary that express Cre. The Rosa26 reporter allele was introduced into MMTV-Cre mice, and the ovaries were analyzed. Blue-stained cells indicated that Cre-mediated recombination had been accomplished. Recombination was observed in oocytes, granulosa cells, and corpora lutea (CL) (Fig. 1A).
|
To delete the bcl-x gene in these cells, we generated females that carry two floxed bcl-x alleles (fl/fl) and the MMTV-Cre transgene (referred to as bcl-x fl/fl; Cre mice, as previously defined). These mice were overtly normal but exhibited reduced fertility. Most females that carried two floxed alleles but no Cre transgene conceived within one estrous cycle and gave birth to normal-sized litters, but only approximately 30% of the bcl-x fl/fl; Cre females (n > 10) gave birth to small litters (six pups or less). Invariably, the offspring carried bcl-x-null alleles (data not shown), establishing that Bcl-x was not required for oocyte survival. Western blot analyses from ovaries and spleen tissue of bcl-x fl/fl; Cre females showed reduced levels of Bcl-x compared to control mice (Fig. 2A). To evaluate whether the reduction of Bcl-x resulted in altered levels of other Bcl-2 family members, we determined the levels of cell survival factor Bcl-2 and of cell death factor Bax. Levels of Bcl-2 and Bax were maintained in bcl-x fl/fl; Cre ovaries compared to control littermates (Fig. 2A). The absence of Bcl-2 in spleen from Bcl-x-null mice reflects the loss of T and B cells as well as of mature erythrocytes, in which Bcl-2 is highly expressed [12]. Furthermore, we performed PCR analyses on isolated granulosa cells and detected the bcl-x-null allele (Fig. 2B), demonstrating that the bcl-x gene was deleted in this cell type. Deletion of the bcl-x gene in ovaries was detected as early as E15.5 (Fig. 2C), a time when the MMTV-Cre transgene is active.
|
Because ovaries from bcl-x fl/fl; Cre mice still contained residual Bcl-x levels (Fig. 2A), we determined by immunohistochemistry the cell types that contain abundant Bcl-x (Fig. 3). Bcl-x was detected in oocytes and many granulosa and luteal cells from bcl-x fl/fl mice. On deletion of the bcl-x gene in bcl-x fl/fl; Cre mice, Bcl-x levels were sharply reduced in these cell types (Fig. 3). Thus, the loss of Bcl-x expression is seen in those cells that expressed the Cre gene based on the Rosa reporter gene (Fig. 1).
|
Maintenance of Follicle Numbers in Postnatal Life in the Absence of Bcl-x
Bcl-x is required for the survival of primordial germ cells [6]. To evaluate whether Bcl-x is required for maintaining the postnatal pool of follicles, we performed morphometric analyses on ovaries from bcl-x fl/fl and bcl-x fl/fl; Cre females. At 5 wk and 3 mo of age, no differences were observed for primordial, primary, and antral follicle numbers in bcl-x fl/fl; Cre ovaries compared with bcl-x fl/fl control littermates (Table 1). Bcl-x fl/fl; Cre ovaries were also healthy in appearance and morphologically similar to ovaries of control littermates. For 5 wk, n = 3 for bcl-x fl/fl; Cre and bcl-x fl/fl mice, P = 0.184 for primordial follicles, P = 0.305 for primary follicles, and P = 0.989 for antral follicles. For 3 mo, n = 6 for bcl-x fl/fl; Cre mice and n = 3 for bcl-x fl/fl mice, P = 0.156 for primordial follicles, P = 0.357 for primary follicles, and P = 0.206 for antral follicles.
|
Ovulation and Progesterone Levels in the Absence of Bcl-x
To evaluate whether the reduced fertility of the bcl-x fl/fl; Cre females resulted from failure to ovulate, we superovulated mice and determined the number of oocytes. Both bcl-x fl/fl (14 ± 3.5) and bcl-x fl/fl; Cre (19 ± 5, P = 0.459) females had similar numbers of oocytes recovered in their oviducts. We also superovulated Stat5- and PrlR-null mice and determined the number of oocytes. Stat5 -/- mice had 37 ± 9 and Stat5 +/- mice 41 ± 19 (P = 0.858) oocytes recovered from their oviducts. The PrlR -/- mice had 10 ± 1 and wild-type mice 22 ± 6.7 (P = 0.086) oocytes recovered from their oviducts. Bcl-x fl/fl; Cre females had a normal estrous cycle, and progesterone levels in pseudopregnant bcl-x fl/fl; Cre mice (30 ± 12 ng/ml) were comparable to those in bcl-x fl/fl control mice (19 ± 7 ng/ml, P = 0.450).
Expression of the bcl-x Gene in Ovaries Is Not Controlled by Stat5
To determine whether expression of the bcl-x gene in ovaries is dependent on the PrlR-Stat5 pathway, we analyzed Bcl-x levels in ovaries from mature PrlR- and Stat5-null mice. Similar levels of Bcl-x were found in the presence and absence of the PrlR and Stat5 (Fig. 1A). The specificity of the Bcl-x signal was established with spleens from mice in which the bcl-x gene had been deleted in the erythroid compartment [12]. The strong signal observed in spleen from bcl-x fl/fl mice was lost in bcl-x fl/fl; Cre mice (Fig. 4A). The presence of Stat5a and Stat5b in ovarian tissue was confirmed by Western blot analysis (Fig. 4, B and C, respectively). As expected, no Stat5 was detected in Stat5-null mice. The absence of Stat5 in spleen from Bcl-x-null mice reflects the loss of T and B cells as well as of mature erythrocytes from this organ [12].
|
Maintenance of CL Depends on PrlR But Not Stat5 and Bcl-x
The PrlR- [13] and Stat5-null [14] mice are infertile, possibly as the result of nonfunctional CL. Whereas CL of normal size and appearance were observed in PrlR-null females [13], only a few small CL were reported in Stat5-null mice [14]. We have now evaluated whether PrlR, Stat5, and Bcl-x are required for establishment and maintenance of the CL using the respective gene-deletion mice. Histological analyses demonstrated the presence of CL in mature PrlR- and Stat5-null virgins and in mice from which the bcl-x gene had been deleted in granulosa and luteal cells (Fig. 5, A, D, and G, respectively). Because the loss of PrlR, Stat5, and Bcl-x results in severely reduced fertility, it was not possible to investigate pregnancy-mediated maintenance of the CL. Pseudopregnancies were induced with eCG-hCG, and the ovaries were examined after 1 wk. The PrlR-null mice developed hemorrhagic CL (Fig. 5B) and exhibited extensive apoptosis as detected by TUNEL assay (Fig. 5C). In contrast, no overt changes in the CL were observed in the absence of Stat5 (Fig. 5, E and F) and Bcl-x (Fig. 5, H and I). Similarly, the CL were maintained in wild-type mice (Fig. 5, K and L). Whereas we observed CL of regular size in the Stat5-null mice, others have described Stat5-null mice containing no, or only small, CL [14]. The reason for this difference is not clear.
|
|
DISCUSSION |
|---|
|
TOPABSTRACTINTRODUCTIONMATERIALS AND METHODSRESULTSDISCUSSIONNOTE ADDED IN PROOFREFERENCES |
|---|
Although Bcl-x is the most abundant member of the Bcl-2 family, other antiapoptotic proteins might compensate for the loss of Bcl-x. It could be hypothesized that Bcl-2 compensates for the loss of Bcl-x, but such compensation has not been observed in the fetal gonad [6] or in erythroid cells [12]. In addition, we did not detect increased levels of Bcl-2 in the Bcl-x-null ovaries, suggesting the absence of compensation through Bcl-2.
Evidence from the hematopoietic system suggests that cytokines can activate bcl-x gene transcription through the transcription factor Stat5. In erythroid cells, bcl-x gene expression is activated by erythropoietin through the Jak2/Stat5 pathway [17]. We have now shown that Bcl-x levels in ovaries are unaltered in the absence of PrlR or Stat5, suggesting that this pathway does not control transcription of the bcl-x gene. However, this does not exclude that bcl-x gene expression in the ovary can be induced under certain circumstances through the PrlR-Stat5 axis.
Corpora lutea cannot be maintained in the absence of PrlR, and after 1 wk of pseudopregnancy, most luteal cells have undergone apoptosis. In contrast, no significant loss of luteal cells is observed in the absence of Stat5 and Bcl-x. This demonstrates that the Prl-induced cell survival signal is not mediated by the Jak2/Stat5/Bcl-x pathway. Because cytokine-induced cell survival of other cell types is mediated by Stat5, we suggest that luteal cells are unique and exploit other, yet-to-be-defined cell survival pathways.
|
NOTE ADDED IN PROOF |
|---|
|
TOPABSTRACTINTRODUCTIONMATERIALS AND METHODSRESULTSDISCUSSIONNOTE ADDED IN PROOFREFERENCES |
|---|
|
ACKNOWLEDGMENTS |
|---|
|
FOOTNOTES |
|---|
1 Supported in part by NIH R01 HD 38955 (J.F.). 
2 Correspondence: Lothar Hennighausen, Laboratory of Genetics and Physiology, National Institutes of Health, Bldg. 8, Room 101, 8 Center Drive, Bethesda, MD 20892. FAX: 301 480 7312; hennighausen{at}nih.gov
Accepted: September 19, 2001.
Received: August 15, 2001.
|
REFERENCES |
|---|
|
TOPABSTRACTINTRODUCTIONMATERIALS AND METHODSRESULTSDISCUSSIONNOTE ADDED IN PROOFREFERENCES |
|---|
This article has been cited by other articles:
|
|
 |
|
  Y. Oshima, T. Akiyama, A. Hikita, M. Iwasawa, Y. Nagase, M. Nakamura, H. Wakeyama, N. Kawamura, T. Ikeda, U.-i. Chung, et al. Pivotal Role of Bcl-2 Family Proteins in the Regulation of Chondrocyte Apoptosis J. Biol. Chem., September 26, 2008; 283(39): 26499 - 26508. [Abstract] [Full Text] [PDF]
|
|
|
|
 |
|
 J. M. Savitt, S. S. Jang, W. Mu, V. L. Dawson, and T. M. Dawson Bcl-x Is Required for Proper Development of the Mouse Substantia Nigra J. Neurosci., July 20, 2005; 25(29): 6721 - 6728. [Abstract] [Full Text] [PDF]
|
|
|
|
 |
|
 Z.-J. Lan, X. Xu, and A. J. Cooney Differential Oocyte-Specific Expression of Cre Recombinase Activity in GDF-9-iCre, Zp3cre, and Msx2Cre Transgenic Mice Biol Reprod, November 1, 2004; 71(5): 1469 - 1474. [Abstract] [Full Text] [PDF]
|
|
|
|
 |
|
 I. Ben-Shlomo, U. A. Vitt, and A. J. W. Hsueh Perspective: The Ovarian Kaleidoscope Database--II. Functional Genomic Analysis of an Organ-Specific Database Endocrinology, June 1, 2002; 143(6): 2041 - 2044. [Abstract] [Full Text] [PDF]
|
|
| ||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||
| HOME | HELP | FEEDBACK | SUBSCRIPTIONS | ARCHIVE | SEARCH | TABLE OF CONTENTS |
