help button home button Endocrine Society Endocrinology
HOME HELP FEEDBACK SUBSCRIPTIONS ARCHIVE SEARCH TABLE OF CONTENTS
Endocrinology, doi:10.1210/en.2006-1590
This Article
Right arrow Abstract Freely available
Right arrow Full Text (PDF)
Right arrow Alert me when this article is cited
Right arrow Alert me if a correction is posted
Right arrow Citation Map
Services
Right arrow Email this article to a friend
Right arrow Similar articles in this journal
Right arrow Similar articles in PubMed
Right arrow Alert me to new issues of the journal
Right arrow Download to citation manager
Right arrow Request Copyright Permission
Citing Articles
Right arrow Citing Articles via Google Scholar
Google Scholar
Right arrow Articles by Andreu-Vieyra, C.
Right arrow Articles by Matzuk, M. M.
Right arrow Search for Related Content
PubMed
Right arrow PubMed Citation
Right arrow Articles by Andreu-Vieyra, C.
Right arrow Articles by Matzuk, M. M.
Endocrinology Vol. 148, No. 8 3837-3849
Copyright © 2007 by The Endocrine Society

Effects of Granulosa Cell-Specific Deletion of Rb in Inha-{alpha} Null Female Mice

Claudia Andreu-Vieyra, Ruihong Chen and Martin M. Matzuk

Departments of Pathology (C.A.-V., R.C., M.M.M.), Molecular and Cellular Biology (M.M.M.), and Molecular and Human Genetics (M.M.M.) and Program in Developmental Biology (M.M.M.), Baylor College of Medicine, Houston, Texas 77030

Address all correspondence and requests for reprints to: Martin M. Matzuk, M.D., Ph.D., Stuart Wallace Chair and Professor, Baylor College of Medicine, Department of Pathology, One Baylor Plaza, Smith Building S217, Houston, Texas 77030. E-mail: mmatzuk{at}bcm.tmc.edu.


   Abstract
Top
 Abstract
Introduction
 Materials and Methods
 Results
 Discussion
 References
 
Our laboratory is interested in the gonadal growth regulatory properties of inhibins, members of the TGFß superfamily. We have previously shown that female mice lacking inhibins (Inha–/–) develop granulosa cell tumors and that concurrent loss of p27 accelerates tumor development. It has also been shown that the retinoblastoma protein RB regulates the G1 to S phase transition of the cell cycle by controlling the activity of transcription factors and stabilizing the levels of the cell cycle inhibitor P27. Based on these data, we hypothesized that concurrent loss of Rb and inhibins in the ovary will exacerbate tumor formation. To test this hypothesis, we generated an ovarian granulosa cell conditional knockout (cKO) of Rb using the Cre/lox recombination system in the background of Inha–/– mice. Inha–/–/Rb cKO females show a modest increase in mortality rates compared with Inha–/– females. Although histologically similar to Inha–/– ovarian tumors, tumors from Inha–/–/Rb cKO females show increased number of mitotic figures and apoptotic rates. Interestingly, P27 levels are decreased in Inha–/–/Rb cKO ovarian tumors, likely due to the combined effect of Rb loss and increased Skp2 expression, which targets P27 to the proteosome. We propose that Rb loss may cause cell cycle delay or arrest, followed by apoptosis and that increases in p107 and p130 levels may compensate for Rb loss. These findings confirm the importance of P27 as a cell cycle regulator in granulosa cells and suggest functional compensation between RB-like proteins in ovarian tumorigenesis.


   Introduction
Top
 Abstract
 Introduction
Materials and Methods
 Results
 Discussion
 References
 
MAMMALIAN CELL CYCLE progression involves the activity of cyclin-dependent kinases (CDKs), which are complexes of cyclin and cdc/cdk proteins (1, 2). The G1-S transition of the cell cycle is dependent on the activity of cyclin D-CDK4/6 and cyclin E2/CDK2 (1). Once activated, these CDK complexes phosphorylate retinoblastoma (RB) and related family members (RB-like 1/p107 and RB-like 2/p130) and release sequestered E2F transcription factors, thereby promoting the transcription of genes required for progression through the cell cycle (3, 4, 5, 6, 7). Such progression is under the control of CDK inhibitors (CDKIs), including p16ink4, p27kip, and p21cip, which cause transient or permanent cell cycle arrest in cells carrying DNA damage (2). Alternatively, apoptotic pathways can be activated to suppress the propagation of abnormal cells (8). Because cell cycle transitions are tightly regulated, changes in the expression of CDKs or CDKIs may lead to exacerbated cell proliferation and malignant transformation (2).

In the adult ovary, granulosa cells undergo proliferation during follicular development and exit the cell cycle during luteinization or terminal differentiation (9, 10). The relevance of cell cycle control in folliculogenesis and fertility is highlighted by knockout mouse studies in which Cyclin D2 (Ccnd2), CDK4, or CDKIs (p27ki, p21ci/p27ki) have been deleted (11, 12, 13, 14). Ccnd2 knockout (KO) females show defects in granulosa cell proliferation and a block in follicle development at the preantral stage, whereas p27kip KO females show infertility associated, at least in part, with increased granulosa cell proliferation and defects in luteinization. Although p21cip null females are fertile (15, 16), granulosa cells from p27kip/p21cip double-KO females exhibit prolonged proliferation span, suggesting that P21 and P27 cooperate to restrict the proliferative span of granulosa cells (14).

Progression through the cell cycle is regulated by external stimuli, including hormones and growth factors. In granulosa cells, FSH and estradiol initiate the G1-S transition by activation of cyclin D2 (10, 17). Inhibins and activins, which are heterodimeric ({alpha}:ßA/ßB) and homodimeric (ßA: ßA, ßB: ßB, and ßA: ßB) members of the TGFß superfamily, respectively, also play an important role in cell cycle control (18, 19). Female mice lacking the inhibin {alpha}-subunit (Inha–/–) undergo uncontrolled granulosa cell proliferation, formation of gonadal tumors, and premature death (19, 20, 21). Ovarian tumors from Inha–/– females retain the expression of several granulosa cell markers including cyclin D2 and depend on FSH and LH to grow (22, 23, 24). Inha–/– mice also show high serum levels of FSH and activins, the latter being responsible for a cachexia-like wasting syndrome (19, 20, 21, 25). Importantly, studies in double-KO mice show that two cell cycle regulators, cyclin D2 and P27, modify the development of ovarian tumors in Inha–/– females. Mice lacking both Inha and cyclin D2 (Ccnd2) show delayed tumor formation and prolonged survival rates (26), whereas absence of both p27kip and Inha results in earlier tumor formation and death, probably due to the lack of cell cycle arrest and incomplete cell differentiation (27). In rat granulosa cells, treatment with activins has a synergistic effect with FSH on promoting the G1-S transition and the inhibitory phosphorylation of RB (17).

Loss of heterozygosity at the Rb and p130 loci occurs in malignancies of various origins, including the ovary (28). However, it is unclear whether Rb loss of heterozygosity contributes to the tumor phenotype in all tissues (28) because in vitro and in vivo mouse models show that Rb loss may trigger or inhibit apoptosis via an increase in E2F transcription factors and downstream targets (7, 29). Mice heterozygous for Rb develop pituitary and thyroid gland tumors between 8 and 11 months of age, whereas Rb null mice are embryonic lethal (30). The use of conditional KO models to overcome embryonic lethality has helped to unmask the contribution of RB to several malignancies including medulloblastoma and lung and ovarian epithelia tumors (31, 32, 33, 34, 35, 36). However, the role of RB in tumors arising from granulosa cells remains uncharacterized.

To further investigate the role of RB in ovarian tumors, we generated a granulosa cell-specific knockout model for Rb in the Inha null background [Inha–/–/Rb conditional KO (cKO)]. We hypothesized that deletion of Rb in Inha–/–mice would accelerate tumor development and/or progression. Our results suggest that Rb loss accelerates tumorigenesis in Inha null mice through a decrease in P27 protein levels, likely by targeting P27 to the proteosome and an increase in cell proliferation. A compensatory pathway might be activated by an increase in p107 and p130 expression levels. In addition, increases in E2f1, Trp53 (p53), and Bcl-related apoptotic protein (Bax) levels as well as an increase in caspase activity may contribute to the higher levels of apoptosis observed in ovarian tumors from double-KO females. Overall, our results confirm the importance of P27 as a regulator of the cell cycle in granulosa cells and suggest the existence of functional redundancy between RB family members in ovarian tumors.


   Materials and Methods
Top
 Abstract
 Introduction
 Materials and Methods
Results
 Discussion
 References
 
Generation and genotyping of Inha–/–; Rb conditional KO mice
The retinoblastoma protein (Rb) null allele (Rb1tm1Tyj) (designated throughout as Rb+/–), the Rb conditional allele (Rb1tm2Brn) (designated throughout as Rbflox), and the inhibin-{alpha} null allele (Inha) have been previously characterized and were maintained in a C57BL/6J;129S5/SvEvBrd mixed hybrid background. All experiments described in this manuscript were conducted in agreement with accepted standards of humane animal care, as outlined by the Institutional Animal Care and Use Committee of Baylor College of Medicine. The Rb null allele was generated by insertion of a PGK-neo cassette and three base changes in exon 3, which introduces two stop codons and produces a truncated protein (30). The Rb conditional allele creates a truncated protein by deletion of exon 19, which is functionally equivalent to the null allele (33, 37). Rb and Inha mutant mice were genotyped from tail genomic DNA using PCR primers as previously described (20, 37). To generate Inha–/–/Rb cKO granulosa cells, Rb+/– animals were first bred to Amhr2cre+ mice to generate Rb+/–/Amhr2cre+ mice. Rb+/–/Amhr2cre+ and Rbflox/flox mice were then bred to Inha+/– mice. Inha+/–/Rb+/–/Amhr2cre+ mice were subsequently crossed to Inha+/–/Rbflox/flox mice to generate Rbflox/–/Inha–/–/Amhr2cre positive (experimental, designated throughout Inha–/–/Rb cKO) and negative (control, designated throughout Inha–/–/Rbflox/–) females. In some experiments, Inha–/–/Rbflox/+/Amhr2cre– littermates (designated throughout as Inha–/–) were used as controls. Recombination of Rb conditional allele was confirmed in granulosa cells and ovarian tumors by PCR and quantitative PCR (QPCR) analyses using genomic DNA and cDNA, respectively.

Weight and survival studies of Inha and Rb conditional KO female mice
Mice were weighed once per week for a period of 4–14 wk. Mice were monitored for kyphoscoliosis as described (19, 21) and killed when they developed signs of cachexia. Livers from 6-wk-old and adult females were dissected and weighed. Liver weights are presented as a ratio to total body weight to correct for weight differences between controls and Inha–/– females, which are particularly marked in adult mice. Ovarian tumors were dissected and fixed in formalin or snap frozen for RNA or protein.

Serum and hematocrit analysis
Mice were anesthetized by isofluorane inhalation (Abbott Laboratories, North Chicago, IL) and blood was collected by cardiac puncture. Serum was prepared using microtainer tubes (Becton and Dickinson, Franklin Lakes, NJ) and stored at –20 C until assayed. FSH, LH, progesterone, and estradiol measurements were made by the University of Virginia Ligand Core Facility (Specialized Cooperative Center Program in Reproductive Research National Institute of Child Health and Human Development/National Institutes of Health U54 HD28934) as described. Assay information is available (http://www.healthsystem.virginia.edu/internet/crr/ligand.cfm). Sensitivity and coefficient of variation for these assays have been previously reported (38). The percentage hematocrit was measured in blood samples obtained by retroorbital bleed, using heparinized capillary tubes.

Histological and immunohistochemical analyses
For morphological studies, tissues were fixed overnight in 10% neutral buffered formalin, embedded, sectioned, and stained with hematoxylin and eosin (39). Immunohistochemistry was performed on 5-µm-thin tissue sections using the Vectastain ABC kit (Vector Laboratories, Burlingame, CA) according to the manufacturer’s instructions. The monoclonal antibody TROMA I [anti-cytokeratin 8/EndoA/keratin 8 (KRT8)], developed by Philippe Brulet and Rolf Kemler, was obtained from the Developmental Studies Hybridoma Bank developed under the auspices of the National Institute of Child Health and Human Development and maintained by The University of Iowa (Department of Biological Sciences, Iowa City, IA). To evaluate cell proliferation in ovarian tumors, we used a rabbit polyclonal antibody raised against phosphorylated histone 3 (Ser10, catalog no. 06-570; Upstate Laboratories, Upstate, NY), which is a marker for mitosis and condensed chromatin (40). Immunoreactivity was visualized by diaminobenzidine (Vector Laboratories) and counterstained with hematoxylin. Results were confirmed by immunostaining with a monoclonal antibody raised against Ki-67 (TEC-3, code M7249; Dako, Glostrup, Denmark), which is present in cycling but not resting cells (41). For direct comparison, control and experimental ovarian tumor sections from at least five individual females were processed together. Negative (no primary antibody) controls and positive (mouse intestine for proliferation markers and ovarian epithelium for KRT8) controls were run in parallel (data not shown). Cells unstained with phosphorylated histone 3, KRT8, or Ki-67 in sections treated with primary antibodies served as internal negative controls. Total and phosphorylated histone 3- or Ki-67-positive cell numbers were quantified in 10 high-power fields (HPF; x40) per section and five individual sections were quantified per specimen. Results are presented as the average of the percent of positive cells per total number of cells per HPF (mitotic index) ± SEM.

Terminal deoxynucleotidyl transferase-mediated deoxyuridine triphosphate nick end labeling (TUNEL) staining
Analysis of apoptosis in ovarian tumors was carried out by TUNEL assay using the ApopTag Plus fluorescein in situ apoptosis detection kit (catalog no. S7111; Chemicon International, Temecula, CA). Ovarian tumors were fixed in 10% formalin and embedded, and three consecutive 5-µm-thin sections were placed onto one slide. At least five different specimens from Inha–/– and Inha–/–/Rbflox/– cKO were analyzed in parallel. TUNEL assay was performed according to the manufacturer’s instructions, and slides were mounted in Vectashield (Vector Laboratories) containing propidium iodine to visualize chromatin. Slides were analyzed by confocal microscopy. Total and TUNEL-positive cell numbers were quantified in 10 HPFs (x40) per section, and three individual sections were quantified per specimen. Results are presented as the average of the percent of positive cells per total cell number per HPF (apoptotic index) ± SEM.

Caspase activity assay
Caspase-3/7 and caspase-9 activities were measured in ovarian tumors from Inha–/– and Inha–/–/Rbflox/– cKO females using the caspase-Glo 3/7 (catalog no. G8090) and caspase Glo 9 kits (catalog no.10) as per the manufacturer’s instructions (Promega, Madison, WI). Protein extraction and assays were carried out as previously described (42). Briefly, ovarian tumors were dissected out, homogenized in a hypotonic extraction buffer [25 mM HEPES (pH 7.5), 5 mM MgCl2, 1 mM EGTA, and protease inhibitors (Sigma, St. Louis, MO)] and centrifuged for 12 min at 13,000 rpm at 4 C. Protein concentration was measured by the bicinchoninic assay protein assay kit (Pierce, Rockford, IL) and standardized to 5 µg of total protein per assay. Assays were run per duplicate. To demonstrate assay specificity, tumor lysates were coincubated with caspase substrates and 20 µM of the pan-specific caspase inhibitor Z-Val-Ala-Asp(OMe)-fluoromethylketone (catalog no. G7231; Promega). Plates were incubated at room temperature for 1 h after the addition of the caspase substrate and read using a luminometer (Berthold Technologies, Oak Ridge, TN). Values are presented as average luminescence signal per microgram of total protein.

Granulosa cell collection
Twenty-one-day-old control and experimental mice were injected with pregnant mare serum gonadotropin (PMSG), and granulosa cells were collected 24 h later as previously described (43, 44). Briefly, granulosa cells were harvested by puncturing the larger follicles in DMEF/F12 medium (Invitrogen, Carlsbad, CA), supplemented with 0.3% BSA, 10 mM HEPES, and 10 U/ml penicillin and streptomycin. Cells were filtered through a 40-µm nylon mesh (Nalgene, Rochester, NY) to remove tissue debris and oocytes. Granulosa cells were spun down and snap frozen for RNA isolation. RNA was extracted using the RNAeasy kit (QIAGEN, Valencia, CA).

Real-time QPCR analysis
One microgram total RNA from granulosa cells or ovarian tumors was reverse transcribed in 50 µl reaction using 250 Superscript II reverse transcriptase (Invitrogen) and random primers (Invitrogen). Samples were diluted 10-fold and 10 µl were used for each QPCR. Real-time QPCR was performed on the ABI Prism 7500 sequence detection system (Applied Biosystems, Foster City, CA) using predesigned Taqman Assays-On-Demand (Applied Biosystems) PCR primer and probe sets and mouse glyceraldehyde-3-phosphate dehydrogenase (Gapd) as an endogenous control. The following Taqman assays were used: Cdkn1b/p27, Mm00438167; Ccne2, Mm00438077; keratin 8 (Krt8/cytokeratin 8), Mm00835759; and Gapd (4352339E, primer limited). Taqman PCR was performed using the TaqMan universal PCR master mix (Applied Biosystems) in 20 µl of reaction. S-phase kinase-associated protein 2 (p45/Skp2), E2f transcription factor 1 (E2f1), transformation-related protein 53 (Trp53), cyclin-dependent kinase inhibitor 1A (p21Cip1/Cdkn1), Igf1, and Bax protein cDNAs were amplified using Sybr Green master mix (Applied Biosystems) with primers designed using the Primer Express software (Applied Biosystems) (Table 1Go). Rb1 exon 19 was amplified using Sybr Green master mix (Applied Biosystems). Ten-fold serial dilutions were used to determine amplification efficiency for each primer set. Reaction conditions were the same reported by Pangas et al. (38). The relative amount of transcript was calculated by the cycle threshold method as described by Applied Biosystems using the 7500 System software (version 1.2.3; Applied Biosystems) and normalized to the endogenous reference (Gapd). The calibrator sample was randomly chosen from the wild-type samples. The relative amount of target gene expression for each sample was calculated and plotted as the average ± SE.


View this table:
[in this window]
[in a new window]

  		TABLE 1. Primer sequences for real-time PCR analysis

 
Western blot analysis
Lysates from ovarian tumors were subjected to electrophoresis through a 15% sodium dodecyl sulfate-polyacrylamide gel at 100 V. Immunoblotting was performed using polyclonal antibodies raised against P27 (1:2500 dilution; sc528), Bax (1:1000 dilution), cyclin B1 (1:1000 dilution), cdc2, proliferating nuclear antigen (PCNA) (1:1000 dilution), cyclin D2 (1:500 dilution), and P53 (1:1000 dilution). Antibodies were purchased from Santa Cruz Biotechnology (Santa Cruz, CA), with the exception of anticyclin D2 (ab-2, clone DCS-3.1; Lab Vision Corp., Fremont, CA). Blots were reprobed with a 1:1000 dilution of mouse anti-ß-tubulin (Sigma; T4026) to verify equivalent loading of the samples. Bands were visualized using the Supersignal West pico chemiluminescence substrate (Pierce) and quantified by densitometry using Image J (National Institutes of Health). After quantification, the concentration of each target protein was normalized against ß-tubulin and plotted as the average ratio ± SE. A minimum of three samples per genotype was analyzed in three independent experiments.

Statistical analysis
Statistical analysis was carried out using the JMP version 5.1 statistical package (SAS Software, Cary, NC). Statistical differences were tested using the nonparametric Mann-Whitney U for single comparisons or Kruskal-Wallis analysis of ranks test for multiple comparisons. P < 0.05 or P < 0.01 was considered statistically significant.


   Results
Top
 Abstract
 Introduction
 Materials and Methods
 Results
Discussion
 References
 
Conditional disruption of Rb in the ovary
To determine the role of Rb in granulosa cell tumorigenesis, we generated a cKO of Rb in the background of Inha null mice (Inha–/–/Rb cKO) using the Cre-loxP system. We crossed mice carrying the Rb null allele to mice carrying the Amhr2cre knock-in allele (Amhr2cre+) (45). We then crossed Rb+/–/Amhr2cre+ mice and the previously characterized Rb floxed (Rbflox) allele to Inha heterozygous mutant mice. Rb conditional allele contains loxP sites flanking exon 19 of Rb, and its deletion by Cre recombination generates a functional knockout that is equivalent to the null allele (Fig. 1Go) (33, 37). Expression of the Amhr2cre allele in the ovary has been previously described (45), and the allele has been successfully used to delete follistatin, SMAD4 [MAD (mothers against decapentaplegic) homolog 4], and steroidogenic factor 1 floxed alleles in the ovary (38, 46, 47).


Figure 1
View larger version (24K):
[in this window]
[in a new window]

  		FIG. 1. Generation of Inha–/–/Rb cKO mice. A, Schematic representation of the conditional (Flox) allele of Rb. LoxP sites (triangles) flank exon 19. Recombination by Cre recombinase results in an allele functionally equivalent to the null allele. PCR analysis using primers (Rb212, Rb19E, Rb18) for genotyping (D). B, Rb transcript levels measured in granulosa cells (GCs) by quantitative PCR in wild-type, Inha–/–, Inha–/–/Rbflox/– (Inha–/–/Rbflox/–/Amhr2cre–), and Inha–/–/Rb cKO (Rbflox/–/Inha–/–/Amhr2cre+) experimental mice. Levels of Rb (RQ) are shown relative to the wild-type. C, Rb transcript levels measured in ovarian tumors (OTs) by quantitative PCR in Inha–/– and Inha–/–/Rb cKO experimental mice. Levels of Rb (RQ) are shown relative to Inha–/–. D, PCR analysis of genomic DNA extracted from Inha–/–/Rbflox/– and Inha–/–/Rb cKO OTs. Genomic DNA from Rb flox/flox mouse tail was included as control. Note the amplification of the null and floxed alleles in Inha–/–/Rbflox/– (Rbflox/–/Inha–/–/Amhr2cre–) OTs and the amplification of the recombined and null alleles in Inha–/–/Rb cKO (Rbflox/–/Inha–/–/Amhr2cre+) OTs. **, P < 0.01.

 
We confirmed Rb loss of expression in granulosa cells and ovarian tumors from experimental Inha–/–/Rb cKO (Inha–/–/Rbflox/–/Amhr2cre+) females, using primers located in exons 18 (forward) and exon 19 (reverse) (Fig. 1Go, A–C). We used wild-type and Inha–/–/Rbflox/– (Inha–/–/Rbflox/–/Amhr2cre–) granulosa cells as well as ovarian tumors from Inha–/– females as controls. In these experiments, granulosa cells were obtained by puncture of large follicles 24 h after PMSG. As expected, the presence of one null allele caused a 53% decrease in the levels of Rb transcripts in granulosa cells from Inha–/–/Rbflox/– females, whereas a 71% decrease was observed with additional recombination of the floxed allele in Inha–/–/Rb cKO granulosa cells, compared with wild-type granulosa cells. This decrease represented a 49% reduction with respect to granulosa cells from Inha–/–/Rbflox/– females. Ovarian tumors from Inha–/–/Rb cKO showed a 95.5% decrease in the levels of Rb, compared with Inha–/– ovarian tumors. The increased rate of recombination in ovarian tumors vs. granulosa cells is expected because ovarian tumors represent a population enriched in cells lacking Rb, whereas growing follicles represent a mixed population of recombined and nonrecombined cells (38, 48). Recombination of the Rb conditional allele in ovarian tumors was further confirmed by PCR genotyping of genomic DNA using primers described by Marino et al. (33) (Fig. 1DGo).

Inha–/–/Rb cKO mice have decreased survival rates and develop a wasting syndrome
Inha–/– mice are infertile and develop ovarian and testicular granulosa/Sertoli cell tumors and a cancer cachexia-like wasting syndrome, with significant loss in liver and body weights (19, 20, 21, 25). Thus, the wasting syndrome in this model can be used as an indicator for tumor development. To evaluate the effect of Rb loss on tumor development and/or progression, we weekly weighed control (Inha+/–/Rb+/–, Inha–/–, and Inha–/–/Rbflox/–) and experimental (Inha–/–/Rb cKO) females for a period of 14 wk. Whereas Inha+/–/Rb+/– showed increased weight with age, all animals that were null for Inha (Inha–/–, Inha–/–/Rbflox/–, and Inha–/–/Rb cKO) developed kyphoscoliosis and signs of cachexia, irrespective of their Rb status.

As expected, the survival rate of Inha+/–/Rb+/– females was 100% over the tested 4- to 14-wk period (Fig. 2AGo). Survival rates were also 100% for Inha–/–, Inha–/–/Rbflox/–, and Inha–/–/Rb cKO females up to 7 wk of age; however, all Inha–/– and Inha–/–/Rbflox/– females died by 13 wk of age (0% survival) (Fig. 2AGo). Fifty percent of Inha–/–/Rbflox/– and Inha–/– females died at 10 and 11.5 wk of age, respectively, whereas 50% of Inha–/–/Rb cKO females died at 8 wk of age (Fig. 2AGo). A further 10% per week decrease in the survival rates was observed for those females that survived past 8 wk of age; interestingly, a few females survived past wk 13, when 0% survival was observed for Inha–/– females (Fig. 2AGo). The minimal change in survival rates observed in Inha–/–/Rb cKO females was unexpected, given the relevance of Rb in cell cycle control.


Figure 2
View larger version (76K):
[in this window]
[in a new window]

  		FIG. 2. Survival curves and histological analysis of ovarian tumors from Inha–/–/Rb cKO females. A, Mice were counted once per week between 4 and 14 wk of age. Results are presented as a percentage of survival. Note that 50% of Inha–/–/Rb cKO females die by 8 wk of age, whereas 50% of Inha–/– females die by 11.5 wk of age. B, Hematoxilin-eosin staining of experimental (Inha–/–/Rb cKO) and control (Inha–/–) ovarian tumors. No morphological differences were observed in the seven specimens per genotype that were analyzed.

 
To better understand the small differences observed in the double-KO females, we analyzed liver weights and hematocrit levels in adult and 6-wk-old mice. At 6 wk of age, no significant differences in liver weight were observed between experimental (Inha–/–/Rb cKO) and control mice (Inha+/–/Rb+/–, Inha–/–, Inha–/–/Rbflox/–) (Table 2Go). In contrast, at the same age, a small but significant reduction in hematocrit was observed in Inha–/–/Rbflox/– and Inha–/– females. In end-of-life measurements, a significant decrease in liver weight was observed in Inha–/–/Rb cKO females, compared with Inha+/–/Rb+/– controls, but this decrease was similar to that observed in Inha–/– and Inha–/–/Rbflox/– females (Table 2Go). Hematocrit levels in adult animals were significantly different among all groups, with the lowest levels observed in Inha–/– females, followed by Inha–/–/Rbflox/–and Inha–/–/Rb cKO females (Table 2Go). The reasons for this phenomenon are still unclear, although higher hematocrit levels in Inha–/–/Rb cKO females correlated with slightly higher body weights (data not shown). Taken together, these results suggest that Rb deletion in granulosa cells does not have a major impact on the development of the cancer cachexia syndrome induced by Inha loss. Furthermore, hematocrit levels suggest that the absence of Rb prevents or delays the onset of anemia observed in Inha–/– females. Although Rb loss decreases the survival of double-KO females, this decrease is less dramatic than expected, and it suggests that other factors or processes may play a compensatory role in the development of Inha–/– ovarian tumors.


View this table:
[in this window]
[in a new window]

  		TABLE 2. Liver weights and hematocrit for 6-wk-old and adult Inha–/–/Rb cKO females

 
Inha–/–/Rb cKO females develop hemorrhagic ovarian tumors similar to those of Inha–/– females
To begin to understand the effect of the loss of Rb on tumor formation and progression, we performed macroscopic and histological examination of ovarian tumors. Macroscopic examination of ovarian tumors was uninformative because high variability in ovarian tumor size was observed in specimens from individual Inha–/–/Rb cKO females but also from Inha–/– females. Histological analysis of ovarian tumors from Inha–/–/Rb cKO adult animals revealed high heterogeneity among specimens and failed to unmask morphological differences with Inha–/– tumors (Fig. 2Go, B and C). Large hemorrhagic cysts were found throughout the ovaries of 6-wk-old Inha–/–/Rb cKO females, which persisted in a number of older mice (Fig. 2CGo); hemorrhagic cysts were also observed in Inha–/– adult ovarian tumors (Fig. 3BGo). A major difference between Inha–/–/Rb cKO and Inha–/– ovarian tumors was the presence of KRT8-positive cells in the former (Fig. 3Go, A and D). Seventy-one percent of Inha–/–/Rb and 50% of Inha–/–/Rbflox/– ovarian tumors showed significantly higher expression of KRT8, compared with Inha–/– ovarian tumors (Fig. 4Go, E and F). KRT8-positive cells were distributed in small patches throughout the tumors (Fig. 3Go, A and B) and lining ovarian cysts (Fig. 3DGo) in Inha–/–/Rb cKO ovaries, whereas Inha–/– ovarian tumors were negative (Fig. 3CGo).


Figure 3
View larger version (81K):
[in this window]
[in a new window]

  		FIG. 3. Expression and localization of KRT8 in ovarian tumors from Inha–/– and Inha–/–/Rb cKO females. A, B, and D, Immunohistochemistry of Krt8 in experimental (Inha–/–/Rb cKO) and control (Inha–/–) ovarian tumors (C). Brown staining indicates immunoreactivity; nuclei are counterstained blue (hematoxylin). Note that immunostaning was present in Inha–/–/Rb cKO ovarian tumors but not Inha–/– ovarian tumors. Images were captured at the indicated original magnification. E, Percentage of KRT8-positive specimens; a total of seven specimens per genotype were analyzed. F, QPCR analysis showed that Krt8 levels were significantly increased by 400-fold in Inha–/–/Rb cKO ovarian tumors (n = 5) vs. Inha–/– ovarian tumors (n = 5). Mean and SEM are shown. RQ, relative quantity. *, P < 0.05.

 

Figure 4
View larger version (68K):
[in this window]
[in a new window]

  		FIG. 4. Inha–/–/Rb cKO showed increased number of mitotic and cycling cells, compared with Inha–/– ovarian tumors. Immunohistochemistry of the mitotic marker phosphorylated histone 3 (A) and the proliferation marker Ki-67 (B) in experimental (Inha–/–/Rb cKO) and control (Inha–/–) ovarian tumors. Brown staining indicates immunoreactivity; nuclei are counterstained blue (hematoxylin). Note the increased number of immunopositive cells per field in Inha–/–/Rb cKO ovarian tumors. Images were captured at the indicated original magnification. C, The mitotic index (percent number of positive cells per x40 field per total number of cells) was increased by 1.6-fold in Inha–/–/Rb cKO ovarian tumors, compared with Inha–/– ovarian tumors (n = 5). D, The number of Ki-67-positive cells (cycling cells) per x40 field per total number of cells was also significantly increased in Inha–/–/Rb cKO ovarian tumors (n = 5). Mean and SEM are shown. *, P < 0.05; **, P < 0.01.

 
Inha–/–/Rb cKO female mice show significantly higher LH and progesterone but not FSH or estradiol serum levels, compared with Inha–/– mice
Inha–/– females have increased serum FSH levels due to the loss of its negative regulator, inhibin, and consequent increase in its positive regulator, activin (20). We were unable to detect significant differences in FSH or estradiol levels in end-of-life serum samples taken from Inha–/–/Rb cKO and Inha–/– females (Table 3Go); however, samples from Inha–/–/Rb cKO females showed significantly higher levels of LH and progesterone (Table 3Go). Because both gonadotropin hormones (FSH and LH) have been shown to function as positive modifiers of ovarian tumors in Inha–/– mice (23, 24), it is possible that higher LH levels may contribute to the decreased survival rate observed in Inha–/–/Rb cKO females.


View this table:
[in this window]
[in a new window]

  		TABLE 3. Serum hormone levels for adult Inha–/–, Inha–/–/Rbflox/–, and Inha–/–/Rb cKO females

 
Ovarian tumors from Inha–/–/Rb cKO females show increased number of mitotic figures and apoptotic cells, compared with Inha–/– ovarian tumors
Loss of Rb has been shown to lead to increased cell proliferation or cell death under a variety of experimental conditions and in a tissue-specific manner (7, 33, 34, 36).

To further investigate the effect of Rb loss in ovarian tumors, we first analyzed the rate of proliferation of these tumors by phosphorylated histone 3 and Ki-67 immunostaining. We found a 1.6-fold increase in the number of phosphorylated histone 3-positive cells in Inha–/–/Rb cKO ovarian tumors as compared with Inha–/– tumors, suggesting an increase in the number of mitotic cells in the ovarian tumors from double-KO females (P < 0.01; Fig. 4Go, A, and C). Inha–/–/Rb cKO ovarian tumors also showed a significant increase in the percentage of Ki-67-positive cells (P < 0.05; Fig. 4Go, B and D), which reflects an increase in the number of total cycling cells. These findings correlated with an increase in the expression E2f1 and Ccne2, two genes known to be involved in the G1-S transition of the cell cycle and deregulated in the absence of Rb (7) (Fig. 5Go, A and B). Granulosa cells from both Inha–/– and Inha–/–/Rb cKO females also showed a significant increase in the expression of these genes, compared with granulosa cells from wild-type mice, but no differences in expression levels were observed in granulosa cells from Inha–/– and Inha–/–/Rb cKO females (Fig. 5Go, A and B). In addition, mitotic markers cyclin D2 (CcnD2), PCNA, cyclin B1 (CcnB1), and CDC2 (1) were significantly increased in Inha–/–/Rb cKO ovarian tumors, compared with Inha–/–, tumors by Western blot and densitometry analyses (n = 3, P < 0.05; Fig. 6Go, A–E). Although Cdc2 was also significantly increased in Inha–/–/Rbflox/– tumors (Fig. 6Go, A and E), we observed large variability in the levels of PCNA and cyclin B1 in these tumors (Fig. 6Go, A, C, and D).


Figure 5
View larger version (20K):
[in this window]
[in a new window]

  		FIG. 5. Changes in cyclin E2 (Ccne2) and E2f1 expression levels in ovarian tumors and granulosa cells of Inha–/–/Rb cKO females. QPCR analysis of mRNA from 3-wk-old control [wild-type (WT) and Inha–/–] and experimental (Inha–/–/Rb cKO) granulosa cells (GCs) as well as Inha–/– and Inha–/–/Rb cKO ovarian tumors (OTs). Granulosa cells were collected from three independent control and four experimental ovaries stimulated with PMSG for 24 h. Ovarian tumors were collected from five control and five experimental females. Mean and SEM are shown. Significant changes were seen in the relative quantity (RQ) of Ccne2 (A) and E2f1 (B) in Inha–/– and Inha–/–/Rb cKO granulosa cells. Significant changes were also seen in the relative quantity of Ccne2 (A) and E2f1 (B) in Inha–/–/Rb cKO ovarian tumors, compared with Inha–/– ovarian tumors. *, P < 0.05; **, P < 0.01.

 

Figure 6
View larger version (25K):
[in this window]
[in a new window]

  		FIG. 6. Inha–/–/Rb cKO ovarian tumors showed increased levels of mitotic markers PCNA cyclin D2, cyclin B1, and CDC2, compared with Inha–/– tumors. A, Representative Western blot analysis of PCNA cyclin D2, cyclin B1, and CDC2 in ovarian tumors from experimental (Inha–/–/Rb cKO) and control (Inha–/– or Inha–/–/Rbflox/–) females. Three independent ovarian tumor samples were evaluated in three separate experiments. Note the increase in the levels of these markers in Inha–/–/Rb cKO samples. B–E, Western blot band intensity was analyzed by densitometry; values were corrected by ß-tubulin intensity. B, Cyclin D2. C, PCNA. D, Cyclin B1. E, CDC (p34). Mean and SEM are shown. *, P < 0.05. RQ, Relative quantity.

 
Next, we investigated whether Rb loss increased the rate of apoptosis in Inha–/– ovarian tumors. TUNEL analysis revealed a 2-fold increase in the number of TUNEL-positive cells in Inha–/–/Rb cKO tumors, compared with Inha–/– ovarian tumors (P < 0.05; Fig. 7Go, A and B). Because loss of Rb has been shown to induce apoptosis via p53-dependent or independent mechanisms (7), we evaluated P53 levels as well as the levels of the P53 target genes Bax and p21, which is also an E2F1 target (49, 50). QPCR analysis showed significant increases in Trp53, p21cip, and Bax expression in Inha–/–/Rb cKO ovarian tumors compared with Inha–/– tumors (P < 0.05; Table 4Go and Fig. 8AGo); however, no significant differences among groups were observed in granulosa cell samples (Table 4Go). In addition, Western blot and densitometry analyses showed a significant increase in P53 and Bax protein levels in Inha–/–/Rb cKO ovarian tumors (n = 3, P < 0.05; Fig. 8Go, B and C). Increased DNA fragmentation and expression of proapoptotic markers in Inha–/–/Rb cKO tumors correlated with a significant increase in the activities of caspases-3/7, an effector caspase, and caspase-9, a caspase involved in the mitochondrial apoptotic pathway (P < 0.05; Fig. 8DGo) (51, 52). Caspase activity was significantly decreased by coincubation with a pan-specific caspase inhibitor (P < 0.05; Fig. 8DGo). Taken together, these results suggest that although positive signals for cell proliferation are in place in Inha–/–/Rb cKO ovarian tumors, induction of apoptosis could limit ovarian tumor growth.


Figure 7
View larger version (37K):
[in this window]
[in a new window]

  		FIG. 7. Evaluation of apoptosis in Inha–/–/Rb cKO and Inha–/– ovarian tumors. A, TUNEL assay and confocal microscopy analysis of experimental (Inha–/–/Rb cKO) and control (Inha–/–) ovarian tumors. Green staining indicates TUNEL-positive, apoptotic cells; nuclei are counterstained red (propidium iodine); yellow indicates the merge between TUNEL and propidium iodine staining. Note the increased number of TUNEL-positive cells per field in Inha–/–/Rb cKO ovarian tumors. Images were captured at x20 (original magnification). B, The apoptotic index (percentage of TUNEL-positive cells per x40 field per number of total cells) was increased by 2-fold in Inha–/–/Rb cKO ovarian tumors, compared with Inha–/– ovarian tumors (n = 5). Mean and SEM are shown. **, P < 0.01.

 

View this table:
[in this window]
[in a new window]

  		TABLE 4. QPCR results for changes in gene expression in granulosa cells and ovarian tumors derived from 3-wk-old and adult Inha –/–/Rb cKO female mice

 

Figure 8
View larger version (16K):
[in this window]
[in a new window]

  		FIG. 8. Inha–/–/Rb cKO ovarian tumors showed increased levels of P53 and Bax as well as increased caspase-3/7 and caspase-9 activities. A, QPCR analysis of Bax levels in ovarian tumors from experimental (Inha–/–/Rb cKO) and control (Inha–/– or Inha–/–/Rbflox/–) females. Three independent ovarian tumor samples were evaluated in three separate experiments. B, Representative Western blot analysis of Bax. Western blot band intensity was analyzed by densitometry and values were corrected by ß-tubulin intensity. Mean and SEM are shown. Note the significant increase in Bax protein levels in Inha–/–/Rb cKO ovarian tumors (P < 0.01), which correlates with the significant increase in mRNA levels (P < 0.05). C, Representative Western blot analysis of P53. Western blot band intensity from three independent experiments was analyzed by densitometry; values were corrected by ß-tubulin intensity. Mean and SEM are shown. Note the significant increase in P53 protein levels in Inha–/–/Rb cKO ovarian tumors (P < 0.05). D, Caspase-3/7 and caspase-9 activities were significantly higher in Inha–/–/Rb cKO ovarian tumors; activities decreased by cotreatment with a pan-specific caspase inhibitor (n = 3; *, P < 0.01; **, P < 0.05). RQ, Relative quantity; casp, caspase; subst, substrate.

 
Loss of Rb causes a decrease in P27 protein levels in Inha–/–/Rb cKO ovarian tumors
The stabilization of P27 is one of RB’s many functions (53); therefore, we predicted that deletion of Rb in Inha–/– animals would cause a reduction in p27 levels and accelerate ovarian tumorigenesis. However, the results presented above show that Rb deletion has only a small impact on ovarian tumorigenesis. To corroborate the predicted decrease in P27 levels, we performed QPCR and Western blot and densitometry analyses in Inha–/–/Rb cKO tumors (Fig. 9Go, A and B). As expected, P27 protein levels were significantly lower in ovarian tumors from Inha–/–/Rb cKO females, compared with Inha–/– tumors (Fig. 9Go). We found that the observed decrease in P27 protein levels did not correlate with changes in mRNA levels because p27 expression was similar in granulosa cells and ovarian tumors from Inha–/–/Rb cKO and Inha–/– females (Fig. 9CGo). Proteosomal degradation of P27 occurs via its phosphorylation by cyclin E2 and its binding to the F-box protein SKP2 (54). We have already shown that Rb loss significantly increases Ccne2 expression in Inha–/–/Rb cKO ovarian tumors (Fig. 5AGo), and this will likely have a negative impact on P27 levels. We then analyzed whether changes in Skp2 expression levels may also contribute to the decrease in P27. As shown in Table 4Go, no significant differences were observed in Skp2 mRNA levels in granulosa cells from wild-type, Inha–/– and Inha–/–/Rb cKO females; however, Skp2 mRNA levels were significantly increased in Inha–/–/Rb cKO vs. Inha–/– ovarian tumors (Table 4Go). In this context, it has been shown that Skp2 is a target for the mitogen IGF-I (55, 56). QPCR analysis showed a significant increased in Igf1 mRNA levels in Inha–/–/Rb cKO vs. Inha–/– ovarian tumors (Table 4Go). Thus, these results confirm the prediction that Rb loss destabilizes P27 in Inha–/– ovarian tumors, and therefore, other factors may contribute to the attenuated phenotype observed in Inha–/–/Rb cKO females.


Figure 9
View larger version (20K):
[in this window]
[in a new window]

  		FIG. 9. Inha–/–/Rb cKO ovarian tumors showed decreased P27 protein levels but not p27 mRNA levels, compared with Inha–/– tumors. A, Representative Western blot analysis of P27 in ovarian tumors (OTs) from experimental (Inha–/–/Rb cKO) and control (Inha–/– or Inha–/–/Rbflox/–) females. Three independent ovarian tumor samples were evaluated in three separate experiments. Note the decrease in P27 levels in Inha–/–/Rb cKO samples. B, Western blot band intensity was analyzed by densitometry and values were corrected by ß-tubulin intensity. Mean and SEM are shown. Note a significant decrease in P27 levels in Inha–/–/Rb cKO, compared with Inha–/– ovarian tumors. *, P < 0.05. C, Real-time PCR analysis of p27 mRNA from 3-wk-old control [wild-type (WT) or Inha–/–] and experimental (Inha–/–/Rb cKO) granulosa cells (GCs) as well as Inha–/– and Inha–/–/Rb cKO ovarian tumors. Granulosa cells were collected from four independent control and five experimental ovaries stimulated with PMSG for 24 h. Ovarian tumors were collected from six control and four experimental females. Mean and SEM are shown. No significant changes were seen in the relative quantity (RQ) of p27 mRNA in granulosa cells or ovarian tumors.

 
Inha–/–/ Rb cKO ovarian tumors showed increased expression of RB family members p107 and p130
In certain cell types, loss of Rb causes a compensatory increase in the levels of the RB-related protein P107. In addition, P130, the other RB-related protein, has been recently shown to be important to prevent ovarian carcinoma cell growth (49, 57, 58, 59).

We next asked whether the concurrent loss of Rb and Inha in the ovary could lead to a compensatory increase in the RB-related family members p107 and p130. To answer this question, we analyzed p107 and p130 mRNA levels by QPCR in granulosa cells and ovarian tumors (Fig. 10Go). Granulosa cells from Inha–/–/Rb cKO females showed a significant increase in p107 mRNA levels, compared with those from wild-type and Inha–/– females (P < 0.05; Fig. 10AGo). p130 mRNA levels were also increased in granulosa cells from Inha–/–/Rb cKO females, compared with those from wild-type females, although these differences were not statistically significant (Fig. 10BGo). In contrast, both p107 and p130 mRNA levels were significantly increased in Inha–/–/Rb cKO ovarian tumors, compared with Inha–/– tumors (Fig. 10Go, A and B), suggesting that compensation by RB family members may occur in ovarian tumors lacking Rb.


Figure 10
View larger version (19K):
[in this window]
[in a new window]

  		FIG. 10. Inha–/–/Rb cKO ovarian tumors and granulosa cells (GCs) showed increased expression of the RB family members p107 and p130. Real-time PCR analysis of mRNA from 3-wk-old control [wild-type (WT) or Inha–/–] and experimental (Inha–/–/Rb cKO) granulosa cells as well as Inha–/– and Inha–/–/Rb cKO ovarian tumors (OTs). Granulosa cells were collected from three independent control and seven experimental ovaries stimulated with PMSG for 24 h. Ovarian tumors were collected from five control and four experimental females. Mean and SEM are shown. Significant changes were seen in the relative quantity (RQ) of p107 (A) in Inha–/–/Rb cKO granulosa cells; although not significant, an increase in p130 (B) was also observed in Inha–/–/Rb cKO granulosa cells. Significant changes were seen in the relative quantity of p107 (A) and p130 (B) in Inha–/–/Rb cKO ovarian tumors, compared with Inha–/– ovarian tumors. *, P < 0.05; **, P < 0.01.

 

   Discussion
Top
 Abstract
 Introduction
 Materials and Methods
 Results
 Discussion
References
 
During the cell cycle, cyclin D2 and cyclin E CDK complexes phosphorylate and inactivate RB, causing the release of bound E2F transcription factors, which are required for cell cycle progression (1, 3, 5, 6, 7). Rb mutations can be found in human malignancies of various origins, including ovarian tumors (28). However, it is unclear the extent to which these mutations contribute to ovarian tumor formation and progression.

To analyze the effect of Rb loss in ovarian tumors, we generated a granulosa-cell specific Rb KO model in the background of Inha–/– mice, which are known to develop ovarian tumors with 100% penetrance. In this study, 50% of Inha–/–/Rb cKO females died 3.5 wk earlier than Inha –/– females, showing increased levels of progesterone and LH, which may exacerbate tumor growth (19, 20, 21, 24). The mechanism by which LH production is increased in double KO females is unclear. Estradiol has been shown to sensitize gonadotrophs to GnRH action, increasing LH production (60), whereas progesterone has been shown to augment LH secretion in pituitary cells (61). Thus, the increase in serum LH levels observed in Inha–/–/Rb cKO females could be attributed to the combination of higher levels of progesterone and estradiol.

Inha–/–/Rb cKO ovarian tumors showed KRT8 immunoreactivity, which was absent in Inha–/– tumors. The origin of KRT2–8-positive cells is currently unknown; however, expression of cytokeratins has been previously reported in human nonepithelial ovarian tumors and fetal as well as transformed rat granulosa cells (62, 63). Therefore, it is possible that the expression of cytokeratin in Inha–/–/Rb cKO ovarian tumors may reflect the presence of immature or fetal-like granulosa cells. Alternatively, cytokeratin expression may be related to epithelization of granulosa cells during cystogenesis, as has been demonstrated in mural granulosa cells of a rat model for polycystic ovary (64). Ovaries from Inha–/–/Rb cKO females develop large cysts before tumor growth, and high LH levels observed in these females may contribute to cystogenesis as previously reported (65).

Our immunostaining analysis also showed significant increases in the number of mitotic and total cycling cells in Inha–/–/Rb cKO ovarian tumors. In addition, we observed increased levels of G1-S and mitotic markers, including cyclin E2, E2F1, PCNA, CDC2, cyclin D2, and cyclin B1, further supporting our immunostaining results. Increases in cyclin E2 and E2F1 mRNA levels were expected because they are both targets of E2F1 (3, 66). Despite the fact that the total number of cells undergoing cell cycle was increased in Inha–/–/Rb cKO tumors, their size was similar to that of Inha–/– ovarian tumors. Increases in DNA fragmentation and caspase activity, as well as increases in the levels of Trp53 and its targets, p21cip and the proapoptotic protein Bax, in Inha–/–/Rb cKO ovarian tumors, suggest that apoptosis limits tumor growth (51). The mechanism by which p53 mRNA levels are increased is unclear. However, recent studies have suggested that the activation of Che-1/Aatf, a RNA polymerase II binding protein that is involved in the transcription of E2F1-responsive genes, can also stimulate the transcription of both p21cip and p53 after DNA damage, leading to cell cycle arrest (67). Deregulated cell proliferation and Rb inactivation have been shown to lead to DNA damage and increased aneuploidy in human cells (66, 68, 69). A plausible scenario that will explain our results is that Inha–/–/Rb cKO cells may become aneuploid as a result of Rb loss, which will trigger a mitotic checkpoint. Cell cycle delay or arrest could be established in an attempt to repair the damage, followed by apoptotic cell death involving mitochondrial factors, including Bax. Increase in Bax levels will offset the balance between pro- and antiapoptotic proteins, causing the release of mitochondrial factors that will activate caspase-9 and downstream caspases (e.g. caspase-3/7) and DNA fragmentation (52). In this context, early studies showed that E2F1 overexpression overcomes p53-mediated cell cycle arrest and that conflicting signals trigger apoptosis (70). More recently, deregulated E2F1 expression has been shown to stimulate the expression of several apoptotic proteins and trigger a p53-dependent apoptotic pathway (7, 29, 71). At present, we can neither confirm nor rule out the possibility that cell cycle delay or arrest takes place in double-KO tumor cells, although our Ki-67 results will argue against this possibility. Further studies aimed at analyzing expression/activation of the RNA polymerase II binding protein Che-1 and aneuploidy in Inha–/–/Rb cKO tumor cells will help clarify these points. Although several genes were up-regulated in Inha–/–/Rb cKO tumors, compared with Inha–/– tumors, no significant differences were observed in the expression of such genes in double-KO granulosa cells, with the exception of RB-related proteins. This could be attributed to the degree of chimerism for Amhr2cre expression and, consequently, recombination of the Rb floxed allele in ovarian follicles (Fig. 1BGo). Alternatively, it is possible that further up-regulation of these genes in granulosa cells from Inha–/–/Rb cKO females may require more than a PMSG pulse to occur.

Several studies have shown that RB stabilizes P27, thereby inhibiting cell cycle progression (53, 72); based on such studies, we predicted that Rb loss in Inha–/– ovarian tumors would lead to decreased P27 levels and increased tumorigenesis. We found that, as expected, there was a significant decrease in P27 protein levels in Inha–/–/Rb cKO tumors, compared with Inha–/– ovarian tumors. Because expression of Skp2, which encodes a component of the SCF (SKP1, cullin 1, and F-box protein)-ubiquitin ligase complex that targets P27 to degradation (73, 74), was also significantly increased in Inha–/–/Rb cKO ovarian tumors, we propose that the decrease in P27 protein levels observed in double-KO tumors is due to its increase degradation by the proteosome. Although further studies are required to confirm this hypothesis, our results show that degradation of P27 appears to be specific because other proteins, including cyclins, P53, Bax, and CDC2 were up-regulated in Inha–/–/Rb cKO tumors. Furthermore, our results show that the expression of Igf1, which has been shown to decrease P27 protein levels via an increase in SKP2 and to be expressed in human granulosa cell tumors (55, 75), was significantly increased in Inha–/–/Rb cKO tumors. Because FSH, estradiol, and IGF-I have been shown to cooperate to promote cell cycle progression in granulosa cells (76, 77), our results suggest that changes in the expression of cell cycle regulators (E2F1 and cyclin E2) and mitogenic factors (IGF-I) contribute to the increased proliferative capability of Inha–/–/Rb cKO ovarian cells.

There is now evidence from studies in mouse models that P107 and P130 can partially compensate for Rb loss, i.e. in epidermis and cardiac muscle development (31, 78). In addition, overexpression of p107 and p130 has been shown to cause cell cycle arrest in a number of cell lines (79), and P107 has also been shown to suppress tumorigenesis at least in RB-deficient cells (80). Therefore, changes in the levels of all three RB-related proteins could have an impact on cell cycle progression. In this study, we found that p107 and p130 were up-regulated in ovarian tumors from Inha–/–/Rb cKO mice, suggesting that functional compensation by RB proteins may occur in this model. The finding that p107 and p130 expression is also increased in granulosa cells from Inha–/–/Rb cKO mice before ovarian tumor formation further supports this hypothesis. At present, we cannot rule out the possibility that other events associated to the many functions of RB may also take place and contribute to the double-KO phenotype. Undoubtedly, further studies are required to evaluate these possibilities.

Studies focused on the analysis of the molecular pathways involved in granulosa cell tumorigenesis are scarce; using in vivo models, we have previously shown that several cell cycle proteins are key to the development/progression of granulosa cell tumors lacking inhibin, including P27 and cyclin D2 (26, 27). The results presented here are the first to show that tumor granulosa cells in mice undergo up-regulation of RB family members and apoptosis in response to Rb loss, as previously reported in other cell types. Whether these pathways are conserved in human ovarian cells awaits investigation.


   Acknowledgments
 
The authors thank Dr. R. Behringer for the gift of the Amhr2-Cre transgenic mice, Mr. S. Ogbonna for his help in setting up the Rb and Inha mouse colonies, and Dr. D. Vieyra for critical reading of the manuscript.


   Footnotes
 
This work was supported by National Institutes of Health Grant CA60651 (to M.M.M.) and a grant from the Lalor Foundation (to C.A.-V.).

First Published Online May 17, 2007

Abbreviations: Bax, Bcl-related apoptotic protein; CDK, cyclin-dependent kinase; CDKI, CDK inhibitor; cKO, conditional KO; Gapd, glyceraldehyde-3-phosphate dehydrogenase; HPF, high-power field; Inha–/–, lacking the inhibin {alpha}-subunit; KO, knockout; KRT8, keratin 8; PCNA, proliferating nuclear antigen; PMSG, pregnant mare serum gonadotropin; QPCR, quantitative PCR; RB, retinoblastoma; Skp2, S-phase kinase-associated protein 2; Trp53, transformation-related protein 53; TUNEL, terminal deoxynucleotidyl transferase-mediated deoxyuridine triphosphate nick end labeling.

Received November 29, 2006.

Accepted for publication May 8, 2007.


   References
Top
 Abstract
 Introduction
 Materials and Methods
 Results
 Discussion
 References
 

  1. Pestell RG, Albanese C, Reutens AT, Segall JE, Lee RJ, Arnold A 1999 The cyclins and cyclin-dependent kinase inhibitors in hormonal regulation of proliferation and differentiation. Endocr Rev 20:501–534[Abstract/Free Full Text]
  2. Elledge SJ, Winston J, Harper JW 1996 A question of balance: the role of cyclin-kinase inhibitors in development and tumorigenesis. Trends Cell Biol 6:388–392[CrossRef][Medline]
  3. Grana X, Garriga J, Mayol X 1998 Role of the retinoblastoma protein family, pRB, p107 and p130 in the negative control of cell growth. Oncogene 17:3365–3383[CrossRef][Medline]
  4. Classon M, Harlow E 2002 The retinoblastoma tumour suppressor in development and cancer. Nat Rev Cancer 2:910–917[CrossRef][Medline]
  5. Nevins JR 2001 The Rb/E2F pathway and cancer. Hum Mol Genet 10:699–703[Abstract/Free Full Text]
  6. Harbour JW, Dean DC 2000 The Rb/E2F pathway: expanding roles and emerging paradigms. Genes Dev 14:2393–2409[Free Full Text]
  7. Harbour JW, Dean DC 2000 Rb function in cell-cycle regulation and apoptosis. Nat Cell Biol 2:E65–E67
  8. Delston RB, Harbour JW 2006 Rb at the interface between cell cycle and apoptotic decisions. Curr Mol Med 6:713–718[Medline]
  9. Rolaki A, Drakakis P, Millingos S, Loutradis D, Makrigiannakis A 2005 Novel trends in follicular development, atresia and corpus luteum regression: a role for apoptosis. Reprod Biomed Online 11:93–103[Medline]
  10. Robker RL, Richards JS 1998 Hormone-induced proliferation and differentiation of granulosa cells: a coordinated balance of the cell cycle regulators cyclin D2 and p27Kip1. Mol Endocrinol 12:924–940[Abstract/Free Full Text]
  11. Sicinski P, Donaher JL, Geng Y, Parker SB, Gardner H, Park MY, Robker RL, Richards JS, McGinnis LK, Biggers JD, Eppig JJ, Bronson RT, Elledge SJ, Weinberg RA 1996 Cyclin D2 is an FSH-responsive gene involved in gonadal cell proliferation and oncogenesis. Nature 384:470–474[CrossRef][Medline]
  12. Nakayama K, Ishida N, Shirane M, Inomata A, Inoue T, Shishido N, Horii I, Loh DY 1996 Mice lacking p27(Kip1) display increased body size, multiple organ hyperplasia, retinal dysplasia, and pituitary tumors. Cell 85:707–720[CrossRef][Medline]
  13. Kiyokawa H, Kineman RD, Manova-Todorova KO, Soares VC, Hoffman ES, Ono M, Khanam D, Hayday AC, Frohman LA, Koff A 1996 Enhanced growth of mice lacking the cyclin-dependent kinase inhibitor function of p27(Kip1). Cell 85:721–732[CrossRef][Medline]
  14. Jirawatnotai S, Moons DS, Stocco CO, Franks R, Hales DB, Gibori G, Kiyokawa H 2003 The cyclin-dependent kinase inhibitors p27Kip1 and p21Cip1 cooperate to restrict proliferative life span in differentiating ovarian cells. J Biol Chem 278:17021–17027[Abstract/Free Full Text]
  15. Brugarolas J, Chandrasekaran C, Gordon JI, Beach D, Jacks T, Hannon GJ 1995 Radiation-induced cell cycle arrest compromised by p21 deficiency. Nature 377:552–557[CrossRef][Medline]
  16. Deng C, Zhang P, Harper JW, Elledge SJ, Leder P 1995 Mice lacking p21CIP1/WAF1 undergo normal development, but are defective in G1 checkpoint control. Cell 82:675–684[CrossRef][Medline]
  17. Ogawa T, Yogo K, Ishida N, Takeya T 2003 Synergistic effects of activin and FSH on hyperphosphorylation of Rb and G1/S transition in rat primary granulosa cells. Mol Cell Endocrinol 210:31–38[CrossRef][Medline]
  18. Bachman KE, Park BH 2005 Duel nature of TGF-ß signaling: tumor suppressor vs. tumor promoter. Curr Opin Oncol 17:49–54[CrossRef][Medline]
  19. Matzuk MM, Kumar TR, Shou W, Coerver KA, Lau AL, Behringer RR, Finegold MJ 1996 Transgenic models to study the roles of inhibins and activins in reproduction, oncogenesis, and development. Recent Prog Horm Res 51:123–154; discussion 155–157[Medline]
  20. Matzuk MM, Finegold MJ, Su JG, Hsueh AJ, Bradley A 1992 {alpha}-Inhibin is a tumour-suppressor gene with gonadal specificity in mice. Nature 360:313–319[CrossRef][Medline]
  21. Matzuk MM, Finegold MJ, Mather JP, Krummen L, Lu H, Bradley A 1994 Development of cancer cachexia-like syndrome and adrenal tumors in inhibin-deficient mice. Proc Natl Acad Sci USA 91:8817–8821[Abstract/Free Full Text]
  22. Burns KH, Owens GE, Ogbonna SC, Nilson JH, Matzuk MM 2003 Expression profiling analyses of gonadotropin responses and tumor development in the absence of inhibins. Endocrinology 144:4492–4507[Abstract/Free Full Text]
  23. Kumar TR, Palapattu G, Wang P, Woodruff TK, Boime I, Byrne MC, Matzuk MM 1999 Transgenic models to study gonadotropin function: the role of follicle-stimulating hormone in gonadal growth and tumorigenesis. Mol Endocrinol 13:851–865[Abstract/Free Full Text]
  24. Kumar TR, Wang Y, Matzuk MM 1996 Gonadotropins are essential modifier factors for gonadal tumor development in inhibin-deficient mice. Endocrinology 137:4210–4216[Abstract]
  25. Coerver KA, Woodruff TK, Finegold MJ, Mather J, Bradley A, Matzuk MM 1996 Activin signaling through activin receptor type II causes the cachexia-like symptoms in inhibin-deficient mice. Mol Endocrinol 10:534–543[Abstract/Free Full Text]
  26. Burns KH, Agno JE, Sicinski P, Matzuk MM 2003 Cyclin D2 and p27 are tissue-specific regulators of tumorigenesis in inhibin {alpha} knockout mice. Mol Endocrinol 17:2053–2069[Abstract/Free Full Text]
  27. Cipriano SC, Chen L, Burns KH, Koff A, Matzuk MM 2001 Inhibin and p27 interact to regulate gonadal tumorigenesis. Mol Endocrinol 15:985–996[Abstract/Free Full Text]
  28. Scambia G, Lovergine S, Masciullo V 2006 RB family members as predictive and prognostic factors in human cancer. Oncogene 25:5302–5308[CrossRef][Medline]
  29. Lazzerini Denchi E, Helin K 2005 E2F1 is crucial for E2F-dependent apoptosis. EMBO Rep 6:661–668[CrossRef][Medline]
  30. Jacks T, Fazeli A, Schmitt EM, Bronson RT, Goodell MA, Weinberg RA 1992 Effects of an Rb mutation in the mouse. Nature 359:295–300[CrossRef][Medline]
  31. Wikenheiser-Brokamp KA 2006 Retinoblastoma family proteins: insights gained through genetic manipulation of mice. Cell Mol Life Sci 63:767–780[CrossRef][Medline]
  32. MacPherson D, Sage J, Crowley D, Trumpp A, Bronson RT, Jacks T 2003 Conditional mutation of Rb causes cell cycle defects without apoptosis in the central nervous system. Mol Cell Biol 23:1044–1053[Abstract/Free Full Text]
  33. Marino S, Vooijs M, van Der Gulden H, Jonkers J, Berns A 2000 Induction of medulloblastomas in p53-null mutant mice by somatic inactivation of Rb in the external granular layer cells of the cerebellum. Genes Dev 14:994–1004[Abstract/Free Full Text]
  34. Meuwissen R, Linn SC, Linnoila RI, Zevenhoven J, Mooi WJ, Berns A 2003 Induction of small cell lung cancer by somatic inactivation of both Trp53 and Rb1 in a conditional mouse model. Cancer Cell 4:181–189[CrossRef][Medline]
  35. Minna JD, Kurie JM, Jacks T 2003 A big step in the study of small cell lung cancer. Cancer Cell 4:163–166[CrossRef][Medline]
  36. Flesken-Nikitin A, Choi KC, Eng JP, Shmidt EN, Nikitin AY 2003 Induction of carcinogenesis by concurrent inactivation of p53 and Rb1 in the mouse ovarian surface epithelium. Cancer Res 63:3459–3463[Abstract/Free Full Text]
  37. Vooijs M, van der Valk M, te Riele H, Berns A 1998 Flp-mediated tissue-specific inactivation of the retinoblastoma tumor suppressor gene in the mouse. Oncogene 17:1–12[CrossRef][Medline]
  38. Pangas SA, Li X, Robertson EJ, Matzuk MM 2006 Premature luteinization and cumulus cell defects in ovarian-specific Smad4 knockout mice. Mol Endocrinol 20:1406–1422[Abstract/Free Full Text]
  39. Cipriano SC, Chen L, Kumar TR, Matzuk MM 2000 Follistatin is a modulator of gonadal tumor progression and the activin-induced wasting syndrome in inhibin-deficient mice. Endocrinology 141:2319–2327[Abstract/Free Full Text]
  40. Grimm SL, Contreras A, Barcellos-Hoff MH, Rosen JM 2005 Cell cycle defects contribute to a block in hormone-induced mammary gland proliferation in CCAAT/enhancer-binding protein (C/EBPß)-null mice. J Biol Chem 280:36301–36309[Abstract/Free Full Text]
  41. Gerdes J, Lemke H, Baisch H, Wacker HH, Schwab U, Stein H 1984 Cell cycle analysis of a cell proliferation-associated human nuclear antigen defined by the monoclonal antibody Ki-67. J Immunol 133:1710–1715[Abstract]
  42. Samuel MA, Morrey JD, Diamond MS 2007 Caspase-3 dependent cell death of neurons contributes to the pathogenesis of West Nile virus encephalitis. J Virol 81:2614–2623[Abstract/Free Full Text]
  43. Elvin JA, Yan C, Matzuk MM 2000 Growth differentiation factor-9 stimulates progesterone synthesis in granulosa cells via a prostaglandin E2/EP2 receptor pathway. Proc Natl Acad Sci USA 97:10288–10293[Abstract/Free Full Text]
  44. Pangas SA, Jorgez CJ, Matzuk MM 2004 Growth differentiation factor 9 regulates expression of the bone morphogenetic protein antagonist gremlin. J Biol Chem 279:32281–32286[Abstract/Free Full Text]
  45. Jamin SP, Arango NA, Mishina Y, Hanks MC, Behringer RR 2003 Genetic studies of the AMH/MIS signaling pathway for Mullerian duct regression. Mol Cell Endocrinol 211:15–19[CrossRef][Medline]
  46. Jorgez CJ, Klysik M, Jamin SP, Behringer RR, Matzuk MM 2004 Granulosa cell-specific inactivation of follistatin causes female fertility defects. Mol Endocrinol 18:953–967[Abstract/Free Full Text]
  47. Jeyasuria P, Ikeda Y, Jamin SP, Zhao L, De Rooij DG, Themmen AP, Behringer RR, Parker KL 2004 Cell-specific knockout of steroidogenic factor 1 reveals its essential roles in gonadal function. Mol Endocrinol 18:1610–1619[Abstract/Free Full Text]
  48. Boerboom D, Paquet M, Hsieh M, Liu J, Jamin SP, Behringer RR, Sirois J, Taketo MM, Richards JS 2005 Misregulated Wnt/ß-catenin signaling leads to ovarian granulosa cell tumor development. Cancer Res 65:9206–9215[Abstract/Free Full Text]
  49. Buchmann AM, Swaminathan S, Thimmapaya B 1998 Regulation of cellular genes in a chromosomal context by the retinoblastoma tumor suppressor protein. Mol Cell Biol 18:4565–4576[Abstract/Free Full Text]
  50. Hiyama H, Iavarone A, LaBaer J, Reeves SA 1997 Regulated ectopic expression of cyclin D1 induces transcriptional activation of the cdk inhibitor p21 gene without altering cell cycle progression. Oncogene 14:2533–2542[CrossRef][Medline]
  51. Andreu-Vieyra CV, Habibi HR 2000 Factors controlling ovarian apoptosis. Can J Physiol Pharmacol 78:1003–1012[CrossRef][Medline]
  52. Earnshaw WC, Martins LM, Kaufmann SH 1999 Mammalian caspases: structure, activation, substrates, and functions during apoptosis. Annu Rev Biochem 68:383–424[CrossRef][Medline]
  53. Ji P, Jiang H, Rekhtman K, Bloom J, Ichetovkin M, Pagano M, Zhu L 2004 An Rb-Skp2–p27 pathway mediates acute cell cycle inhibition by Rb and is retained in a partial-penetrance Rb mutant. Mol Cell 16:47–58[CrossRef][Medline]
  54. Slingerland J, Pagano M 2000 Regulation of the cdk inhibitor p27 and its deregulation in cancer. J Cell Physiol 183:10–17[CrossRef][Medline]
  55. Lu Y, Zi X, Pollak M 2004 Molecular mechanisms underlying IGF-I-induced attenuation of the growth-inhibitory activity of trastuzumab (Herceptin) on SKBR3 breast cancer cells. Int J Cancer 108:334–341[CrossRef][Medline]
  56. Dupont J, Le Roith D 2001 Insulin-like growth factor 1 and oestradiol promote cell proliferation of MCF-7 breast cancer cells: new insights into their synergistic effects. Mol Pathol 54:149–154[Abstract/Free Full Text]
  57. Mulligan GJ, Wong J, Jacks T 1998 p130 is dispensable in peripheral T lymphocytes: evidence for functional compensation by p107 and pRB. Mol Cell Biol 18:206–220[Abstract/Free Full Text]
  58. Donovan SL, Schweers B, Martins R, Johnson D, Dyer MA 2006 Compensation by tumor suppressor genes during retinal development in mice and humans. BMC Biol 4:14[CrossRef][Medline]
  59. Soprano KJ, Purev E, Vuocolo S, Soprano DR 2006 Rb2/p130 and protein phosphatase 2A: key mediators of ovarian carcinoma cell growth suppression by all-trans retinoic acid. Oncogene 25:5315–5325[CrossRef][Medline]
  60. Messinis IE 2006 Ovarian feedback, mechanism of action and possible clinical implications. Hum Reprod Update 12:557–571[Abstract/Free Full Text]
  61. Turgeon JL, Waring DW 2006 Differential expression and regulation of progesterone receptor isoforms in rat and mouse pituitary cells and LßT2 gonadotropes. J Endocrinol 190:837–846[Abstract/Free Full Text]
  62. Van Niekerk CC, Ramaekers FC, Hanselaar AG, Aldeweireldt J, Poels LG 1993 Changes in expression of differentiation markers between normal ovarian cells and derived tumors. Am J Pathol 142:157–177[Abstract]
  63. Pan J, Kwan RW, Auersperg N 1995 Keratin expression and steroidogenesis in rat granulosa cells, transformed with the Kirsten-ras and SV40 oncogenes singly and in combination. Mol Cell Endocrinol 112:231–239[CrossRef][Medline]
  64. Anderson E, Lee GY 1997 The polycystic ovarian (PCO) condition: apoptosis and epithelialization of the ovarian antral follicles are aspects of cystogenesis in the dehydroepiandrosterone (DHEA)-treated rat model. Tissue Cell 29:171–189[CrossRef][Medline]
  65. Abbott DH, Dumesic DA, Franks S 2002 Developmental origin of polycystic ovary syndrome—a hypothesis. J Endocrinol 174:1–5[Abstract]
  66. Bracken AP, Ciro M, Cocito A, Helin K 2004 E2F target genes: unraveling the biology. Trends Biochem Sci 29:409–417[CrossRef][Medline]
  67. Halazonetis TD, Bartek J 2006 DNA damage signaling recruits the RNA polymerase II binding protein Che-1 to the p53 promoter. Mol Cell 24:809–810[CrossRef][Medline]
  68. van Deursen JM 2007 Rb loss causes cancer by driving mitosis mad. Cancer Cell 11:1–3[CrossRef][Medline]
  69. Pickering MT, Kowalik TF 2006 Rb inactivation leads to E2F1-mediated DNA double-strand break accumulation. Oncogene 25:746–755[CrossRef][Medline]
  70. Wu X, Levine AJ 1994 p53 and E2F-1 cooperate to mediate apoptosis. Proc Natl Acad Sci USA 91:3602–3606[Abstract/Free Full Text]
  71. Roos WP, Kaina B 2006 DNA damage-induced cell death by apoptosis. Trends Mol Med 12:440–450[CrossRef][Medline]
  72. Ji P, Zhu L 2005 Using kinetic studies to uncover new Rb functions in inhibiting cell cycle progression. Cell Cycle 4:373–375[Medline]
  73. Nakayama K, Nagahama H, Minamishima YA, Matsumoto M, Nakamichi I, Kitagawa K, Shirane M, Tsunematsu R, Tsukiyama T, Ishida N, Kitagawa M, Hatakeyama S 2000 Targeted disruption of Skp2 results in accumulation of cyclin E and p27(Kip1), polyploidy and centrosome overduplication. EMBO J 19:2069–2081[CrossRef][Medline]
  74. Nakayama K, Nagahama H, Minamishima YA, Miyake S, Ishida N, Hatakeyama S, Kitagawa M, Iemura S, Natsume T, Nakayama KI 2004 Skp2-mediated degradation of p27 regulates progression into mitosis. Dev Cell 6:661–672[CrossRef][Medline]
  75. Alexiadis M, Mamers P, Chu S, Fuller PJ 2006 Insulin-like growth factor, insulin-like growth factor-binding protein-4, and pregnancy-associated plasma protein-A gene expression in human granulosa cell tumors. Int J Gynecol Cancer 16:1973–1979[CrossRef][Medline]
  76. Kolodziejczyk J, Gertler A, Leibovich H, Rzasa J, Gregoraszczuk EL 2003 Synergistic action of growth hormone and insulin-like growth factor I (IGF-I) on proliferation and estradiol secretion in porcine granulosa and theca cells cultured alone or in coculture. Theriogenology 60:559–570[CrossRef][Medline]
  77. Richards JS, Russell DL, Ochsner S, Hsieh M, Doyle KH, Falender AE, Lo YK, Sharma SC 2002 Novel signaling pathways that control ovarian follicular development, ovulation, and luteinization. Recent Prog Horm Res 57:195–220[Abstract/Free Full Text]
  78. Mulligan G, Jacks T 1998 The retinoblastoma gene family: cousins with overlapping interests. Trends Genet 14:223–229[CrossRef][Medline]
  79. Classon M, Dyson N 2001 p107 and p130: versatile proteins with interesting pockets. Exp Cell Res 264:135–147[CrossRef][Medline]
  80. Rodier G, Makris C, Coulombe P, Scime A, Nakayama K, Nakayama KI, Meloche S 2005 p107 inhibits G1 to S phase progression by down-regulating expression of the F-box protein Skp2. J Cell Biol 168:55–66[CrossRef][Medline]




This Article
Right arrow Abstract Freely available
Right arrow Full Text (PDF)
Right arrow Alert me when this article is cited
Right arrow Alert me if a correction is posted
Right arrow Citation Map
Services
Right arrow Email this article to a friend
Right arrow Similar articles in this journal
Right arrow Similar articles in PubMed
Right arrow Alert me to new issues of the journal
Right arrow Download to citation manager
Right arrow Request Copyright Permission
Citing Articles
Right arrow Citing Articles via Google Scholar
Google Scholar
Right arrow Articles by Andreu-Vieyra, C.
Right arrow Articles by Matzuk, M. M.
Right arrow Search for Related Content
PubMed
Right arrow PubMed Citation
Right arrow Articles by Andreu-Vieyra, C.
Right arrow Articles by Matzuk, M. M.

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