Human cumulus granulosa cell gene expression: a predictor of fertilization and embryo selection in women undergoing IVF

L.J. McKenzie1, S.A. Pangas2,3, S.A. Carson1,5, E. Kovanci1, P. Cisneros1, J.E. Buster1, P. Amato1 and M.M. Matzuk2,3,4

Departments of 1 Obstetrics and Gynecology, 2 Pathology, 3 Molecular and Cellular Biology, and 4 Molecular and Human Genetics, Baylor College of Medicine, Houston, TX 77030, USA

5 To whom correspondence should be addressed at: Baylor College of Medicine, Department of Obstetrics and Gynecology, 6550 Fannin Street, Suite 801, Houston, TX 77030, USA. Email: scarson{at}bcm.tmc.edu


    Abstract
 Top
 Abstract
 Introduction
 Methods and materials
 Results
 Discussion
 Acknowledgements
 References
 
BACKGROUND: A biochemical marker for embryo development would increase the chance of a successful pregnancy with IVF by optimizing oocyte and embryo selection, and allow fewer embryos to be transferred. In this study, we correlated cumulus granulosa cell gene expression of hyaluronic acid synthase 2 (HAS2), cyclooxygenase 2 (COX2; PTGS2) and gremlin (GREM1) with subsequent embryo development in search of a parameter for embryo selection. METHODS: Cumulus cell gene expression was determined prospectively on eight consecutive patients undergoing IVF with ICSI. Immediately following oocyte retrieval, the cumulus was stripped from the oocyte, and cumulus gene expression for PTGS2, HAS2 and GREM1 was assessed using a one-step real-time quantitative RT–PCR assay. Oocyte quality, fertilization and embryo morphology were correlated to relative gene expression. RESULTS: Gene expression data were available on cumulus cells from 108 oocytes that developed into 70 embryos (64.8% fertilization rate). Cumulus PTGS2, HAS2 and GREM1 expression was higher from oocytes that developed into higher quality embryos (grades 3, 4 and 5) compared with lower quality embryos (grades 1 and 2) (P<0.05, P<0.001 and P<0.001, respectively). HAS2 and GREM1 expression was also higher from the cumulus surrounding oocytes that gave rise to higher grade embryos (P<0.001). The expression of PTGS2 and HAS2 was 6-fold higher, and that of GREM1 was 15-fold higher in cumulus yielding higher grade embryos versus lower grade embryos. CONCLUSION: PTGS2, HAS2 and GREM1 gene expression correlates to morphological and physiological characteristics and provides a novel approach to predict human embryo development. Ultimately, with better predictors of follicular and embryonic health, higher quality embryos can be selected and transferred, reducing higher order pregnancy rates.

Key words: embryo selection/growth differentiation factor 9/oocyte-regulated genes/reproductive success


    Introduction
 Top
 Abstract
 Introduction
 Methods and materials
 Results
 Discussion
 Acknowledgements
 References
 
After IVF, the primary criterion for embryo selection is the morphological appearance of the embryos based on a combination of cleavage rates and fragmentation of cells (Van Royen et al., 2001Go). The morphological appearance of the oocyte and embryo, however, does not accurately predict the health of the embryo. Seventy to eighty percent of morphologically abnormal embryos have abnormal chromosomes, but up to 40% of embryos with normal morphology have abnormal chromosomes (Munné et al., 1993Go, 1994Go). Thus, the only option to maintain high pregnancy rates is to transfer 3–4 early stage embryos or 2–3 blastocysts. However, this results in frequent high order multiple gestations associated with increased fetal morbidity and mortality (Keith and Oleszczuk, 2002Go; Blickstein and Keith, 2003Go). Therefore, any biochemical markers for fertilization potential and subsequent embryo development would increase the chance of a successful pregnancy by optimizing oocyte and embryo selection, allowing fewer embryos to be transferred.

Pre-ovulatory follicles contain subpopulations of granulosa cells, including mural and cumulus cells. Mural granulosa cells line the follicle wall, are in close proximity to the theca cells and express genes important for follicular rupture. Cumulus cells, which remain apposed to the oocyte during follicular development and ovulation, are likely mediators of oocyte development and fertilization, and are in turn regulated by oocyte factors (Matzuk et al., 2002Go). In the peri-ovulatory period, the cumulus cells expand to form a radiating layer in response to the mid-cycle surge of luteinizing hormone or exogenous HCG. This cumulus expansion denotes a grade III oocyte, the optimal stage for fertilization (Veek, 1999Go). During expansion, cumulus cells secrete a hyaluronic acid-rich matrix that binds the oocyte and cumulus together, facilitates follicular extrusion and oviductal fimbrial capture, and allows sperm penetration and fertilization (Salustri et al., 1996Go). Cumulus expansion is imperative for normal oocyte development in vivo, and oocytes not associated with cumulus expansion have limited potential for implantation (Veek, 1999Go). Expansion of cumulus is regulated by growth differentiation factor 9 (GDF9), an oocyte-secreted member of the transforming growth factor-{beta} (TGF-{beta}) superfamily (Elvin et al., 1999aGo; Chang et al., 2002Go; Vanderhyden et al., 2003Go).

GDF9 has been shown to induce several genes in cumulus granulosa cells, including those for hyaluronic acid synthase 2 (HAS2), cyclooxygenase 2 (COX2; PTGS2), GREM1 and steroidogenic acute regulator protein (StAR), and to repress those for luteinizing hormone receptor (LHR) and urokinase plasminogen activator (uPA), which play critical roles in follicular development and cumulus expansion (Elvin et al., 1999aGo; Pangas et al., 2004Go). Thus, regulated expression of downstream GDF9 targets in the cumulus cells may reflect GDF9 activity, and could ultimately predict oocyte health.

Alterations in cumulus expansion may occur as women age and may be responsible for their reproductive disadvantage, either as a direct cause or as a reflection of a decline in the functional and structural qualities of the oocyte (Navot et al., 1991Go). The biochemical assessment of follicles, including the cumulus and oocytes, may generate additional information necessary for an understanding of successful and unsuccessful fertilization, both in vivo and in vitro. The expression of genes (e.g. PTGS2, HAS2 and GREM1) expressed in the cumulus during expansion, and discarded at the time of IVF, may give a direct assessment of the fertility potential of an individual oocyte without compromising the oocyte integrity. Thus, by determining the ‘optimal’ embryos that are produced and transferred, we may be able to replace fewer embryos and still maintain an acceptable pregnancy rate.


    Methods and materials
 Top
 Abstract
 Introduction
 Methods and materials
 Results
 Discussion
 Acknowledgements
 References
 
Study population
Baylor College of Medicine Institutional Review Board approval was obtained for this study (IRB # H-10863). The study population included eight consecutive patients undergoing IVF with ICSI at the Baylor Assisted Reproductive Technology programme. Patient age, diagnosis, size and number of follicles generated, estradiol level at time of HCG trigger, number of oocytes retrieved, medication used for stimulation and number of prior IVF cycles were recorded for all patients. Exclusion criteria were recent cigarette use and body mass index above 25 kg/m2.

All patients underwent a standard institutional stimulation protocol of ovarian downregulation with GnRH agonist (Lupron®, TAP Pharmaceuticals, Lake Forest, IL) followed by controlled ovarian stimulation exclusively with HMG (Pergonal®, Serono, Rockland, MA). Follicular monitoring was performed via ovarian ultrasonography and serum estradiol measurements.

Average patient age was 30 years (range 22–36) and diagnoses included male factor only (n=3), unexplained (n=1), male factor and endometriosis (n=1), male factor and tubal disease (n=1) and known genetic disorder (n=2). The two couples with known genetic disorders underwent preimplantation genetic diagnosis (PGD); a known female carrier of Hunter's disease and a known male balanced translocation carrier involving chromosomes 5 and 9. Average estradiol prior to HCG administration was 2319 pg/ml (range 600–4942). Oocyte retrieval was performed 36 h following HCG administration. A total of 108 oocytes (average = 13.6 oocytes per patient, range 6–25) were obtained from the 134 scanned follicles >10 mm in maximal diameter.

Semen analyses were obtained on all male partners. Mean semen concentration was 20 x 106/ml (range 1–51 x 106/ml), mean motility was 33.5% (range 3–75) and mean normal morphology (by Kruger's strict criteria) was 2.3% (range 0–7).

Pregnancy outcomes were recorded on all eight patients. One patient had no unaffected embryos available for transfer following PGD. Three patients have ongoing pregnancies, for a 37.5% ongoing pregnancy rate per initiated cycle.

Specimens
Immediately following ooctye retrieval, cumulus cells were stripped from the oocyte with a micropipette and frozen at –80°C within 10 min of oocyte aspiration. Oocytes were maintained in individual microdroplets of culture medium during cumulus recovery, insemination and culture.

ICSI
ICSI was routinely performed 5 h after aspiration using straight needles (Humagen, Charlottesville, VA) in sperm wash medium with HEPES buffer (Irvine Scientific, Santa Ana, CA) on a heated stage (Narishige micromanipulator McHenry, IL).

RNA isolation
RNA from the cumulus cells was extracted using a commercial RNA isolation kit (Picopure RNA Isolation Kit, Arcturus, Mountain View, CA, catalogue no. KIT0204). Total RNA was eluted in a volume of 25 µl of elution buffer and stored at –80°C.

Real-time PCR
RNA expression was assessed using a one-step real-time quantitative RT–PCR assay to measure PTGS2, HAS2 and GREM1 mRNA and 18S rRNA transcripts. The procedure was performed for all mRNA expression using the ABI Prism 7700 Sequence Detector (PE Applied Biosystems) and the proprietary Taq Man® assay. Primers for PTGS2, GREM1 and HAS2 were intron-spanning and obtained through Applied Biosystems [Assays on Demand®: Hs00153133_m1 (PTGS2), Hs00171951_m1 (GREM1) and Hs00193435_m1 (HAS2)]. 18S rRNA was used as a normalizer to correct for the amount of RNA in each sample (Applied Biosystems PDAR with MGB probe; catalogue no. 4319413E). RT–PCR was performed in a reaction volume of 20 µl containing 10 µl of One Step PCR Master Mix (Applied Biosystems catalogue no. 4309169), 0.5 µl of reverse transcriptase, 1 µl of primer, 6.5 µl of RNase-free water (Ambion®, Austin, TX) and 2 µl of extracted RNA. Each sample was analysed in duplicate, and multiple water blanks were included with the analysis.

The thermal profile used for the analysis was as follows: reverse transcription was performed at 48°C for 30 min. Following a 10 min denaturation at 95°C, 40 cycles of PCR were performed at 95°C for 15 s, and 60°C for 1 min.

Amplification data were collected and analysed with the ABI Prism 7700 Sequence Detector software SDS1.2. (PE Applied Biosystems). Relative gene expression was calculated via the Delta-delta C (T) method as described by Livak and Schmittgen (2001)Go. Replicate RT–PCR analysis indicated that the coefficients of variation of PTGS2, GREM1 and HAS2 mRNA relative to 18S rRNA expression were 3.1, 2.8, 3.3 and 1.4%, respectively.

Outcomes
Morphological characteristics of the cumulus and embryos were recorded. The cumulus mass for each cumulus–oocyte complex was assessed as compacted, expanded or hyperexpanded, and the percentage of dark versus light cells was recorded. Based on these characteristics, the oocytes were classified as either mature or immature. Fertilization was assessed the morning after fertilization and was defined by the presence of two pronuclei. Embryo morphology (blastomere number, shape, size, fragmentation and ooplasm characteristics) was recorded at 72 h. Embryos were graded according to a standard institutional grading scale as described (Wolf, 1984Go).

Statistical analysis
Statistical analysis was performed using Spearman's rank correlation test and Mann–Whitney rank sum test (SigmaStat 2.0 software). Logistic regression analysis and receiver operating characteristic (ROC) curves were constructed to determine the predictive value of clinical outcomes using SPSS® (SPSS Inc., Chicago, IL). Data points determined to be outliers (<5% of data points) were excluded from graphic representation, but they were, however, included in statistical analysis.


    Results
 Top
 Abstract
 Introduction
 Methods and materials
 Results
 Discussion
 Acknowledgements
 References
 
Gene expression data were available on cumulus cells from all 108 oocytes. A total of 103 oocytes had expression data for all four genes of interest (PTGS2, HAS2, GREM1 and 18S rRNA), yielding 412 data points. Five out of five total oocytes from one patient did not reveal any expression data for PTGS2 or HAS2; however, GREM1 and 18S rRNA assays yielded amplification results (10 data points). All five cumulus–oocyte complexes were expanded, and one of the five oocytes failed to fertilize. The subsequent embryo development of those four samples was variable; two were high grade and two were low grade.

Cumulus cell PTGS2 expression is 6-fold higher (0.5 versus 3.2, respectively; P<0.05), GREM1 expression is 15-fold higher and HAS2 expression is 6-fold higher (1.0 versus 15 for GREM1 and 1.1 versus 6.5 for HAS2; P<0.001 for both) in cumulus cells yielding healthier embryos (grades 3, 4 and 5) versus poor quality embryos (grades 1 and 2) (Figures 13).



View larger version (11K):
[in this window]
[in a new window]
 
Figure 1. Relative expression of cyclooxygenase (PTGS2) in low grade versus high grade embryos. The horizontal line represents the median expression value, and the box encompasses 50% of data points (1st quartile to 3rd quartile). The upper adjacent value and lower adjacent value (vertical lines) represent maximum and minimum data points.

 


View larger version (11K):
[in this window]
[in a new window]
 
Figure 3. Relative expression of GREM1 in low grade versus high grade embryos.

 
After statistically significant correlations of gene expression to embryo quality were found, ROC curves were constructed to determine the predictive value of PTGS2, HAS and GREM1 for oocyte maturity, fertilization and embryo development. Although all three genes were found to be statistically significant predictors of clinical outcomes (P<0.05) (Figures 46), the genes were not found to be statistically significantly different from one another.



View larger version (21K):
[in this window]
[in a new window]
 
Figure 4. Receiver operating characteristics (ROC) curves for PTGS2, HAS2 and GREM1 expression as predictors of oocyte maturity. The ROC is a graphic representation of the false positives and false negatives for a particular diagnostic test (false positive on the x-axis; 1–false negative on the y-axis). The larger the area under the curve, the better the diagnostic test.

 


View larger version (21K):
[in this window]
[in a new window]
 
Figure 6. Receiver operating characteristics (ROC) curves for PTGS2, HAS2 and GREM1 expression as predictors of embryo quality.

 
Logistic regression was then performed to characterize further the candidate genes as predictors of ooctye maturity, fertilization and embryo grade. By logistic regression, HAS2 expression is predictive of oocyte maturity (P<0.05), and both PTGS2 and GREM1 are predictors of fertilization (P<0.05). PTGS2 and GREM1 are predictive of embryo development (P<0.05).

The best single predictor for oocyte maturity, fertilization and embryo quality is GREM1 expression, with areas under the curve of 0.71, 0.76 and 0.81, respectively (Figures 46). A 5.2-fold increased relative expression of GREM1 yields a sensitivity and specificity for oocyte maturity of 63 and 93%, for fertilization 72 and 78%, and for embryo quality 83 and 81% (Table 1). Combining PTGS2 and GREM1 expressions to improve predictive efficiency increases the predictive value for embryo development only slightly (area under the curve of 0.82 versus 0.81), a difference which is not statistically significant.


View this table:
[in this window]
[in a new window]
 
Table I. Specificity and sensitivity of GREM1, PTGS2 and HAS2 expression for embryo development

 
Additional genes were analysed including pentraxin 3 (PTX3) (Applied Biosystems #Hs00357041_m1) and tumour necrosis factor-{alpha}-induced protein 6 (TNFAIP6) (Applied Biosystems #Hs00200178_m1). PTX3 gene expression was not detected in the cumulus granulosa cells, and TNFAIP6 expression did not differ as a function of embryo quality (data not shown).


    Discussion
 Top
 Abstract
 Introduction
 Methods and materials
 Results
 Discussion
 Acknowledgements
 References
 
The developmental potential of human embryos is reflected in cumulus cell gene expression. Specifically, we demonstrate that expression of three genes downstream of GDF9 activity (HAS2, PTGS2 and GREM1) correlates with embryo development. Cumulus cells from oocytes that develop into higher quality embryos (grades 3, 4 and 5) have 6- to 15-fold increased median expression of these GDF9 targets, and may be of clinical use in human assisted reproduction.

GDF9, a TGF-{beta} superfamily member, is required for normal folliculogenesis in female mice and regulates human folliculogenesis (Dong et al., 1996Go; Aaltonen et al., 1999Go; Elvin et al., 1999aGo, bGo). Elvin et al. demonstrated that GDF9 protein is expressed in all mouse oocytes beginning at the 3a follicle stage including antral follicles. GDF9 gene mutation leads to infertility due to a block at the type 3b (primary) follicle stage, absence of thecal layer formation and defects in oocyte meiotic competence (Dong et al., 1996Go; Carabatsos et al., 1998Go; Aaltonen et al., 1999Go; Elvin et al., 1999bGo; Yan et al., 2001Go). Additional data have shown that GDF9 is also required for later stage development (Elvin et al.,1999aGo; Yan et al., 2001Go).

The oocyte itself plays an active role in cumulus expansion (for review, see Eppig, 2001Go; Matzuk et al., 2002Go), inducing expansion by secretion of GDF9. GDF9 induces HAS2 in cumulus cells (Elvin et al., 1999aGo) that stimulates the production of hyaluronic acid and results in cumulus expansion (Buccione et al., 1990Go). GDF9 also facilitates cumulus expansion via the induction of cyclooxygenase 2 (PTGS2) and the resultant prostaglandin E2 (PGE2) production (Elvin et al., 2000bGo), both of which are involved in cumulus expansion in cattle (Calder et al., 2001Go). Mice lacking functional PTGS2 have defects in ovulation, fertilization, decidualization and implantation (Lim et al., 1997Go). Systemic treatment with PGE2 restores the ability of PTGS2 knockout mice to ovulate (Davis et al., 1999Go). Calder et al. (2001)Go examined the expression of PTGS2 and prostaglandin receptor mRNA in bovine cumulus–oocyte complexes, and found that expression of PTGS2 and EP2 mRNA is dependent on time of maturation and oocyte quality. Complexes in which the oocyte is microsurgically removed and cultured with recombinant GDF9 can be induced to undergo cumulus expansion (Elvin et al., 1999aGo).

GREM1, recently described as another downstream gene upregulated by GDF9, selectively inhibits bone morphogenic protein (BMP) signalling and not GDF9 signalling (Pangas et al., 2004Go). It is hypothesized that the differential signalling inhibition may facilitate the luteinization of mural granulosa cells, while allowing cumulus cell expansion (Pangas et al., 2004Go). If this hypothesis is correct, continued expression of GREM1 in cumulus granulosa cells should represent a mature oocyte and may predict improved embryo development. The role of GREM1 in human reproduction clearly merits further investigation.

Interestingly, despite the known role of PTGS2 in cumulus expansion,PTGS2 had a lower ROC curve for oocyte maturity than GREM1 or HAS2. The ROC curves for PTGS2 increased for fertilization and embryo development, although GREM1 was the best predictor for all three clinical parameters. HAS2 has an established role with cumulus expansion (Buccione et al., 1990Go), and GREM1 may function in cumulus expansion via the luteinization of mural granulosa cells (Pangas et al., 2004Go). Further elucidation of the role of GREM1 in cumulus expansion is required, as it may prove to be a better predictor of cumulus expansion than PTGS2 expression.

In this report, we describe the gene expression of PTGS2, HAS2 and GREM1, which function under the influence of GDF9. Expression profiling of other GDF9-induced genes in periovulatory granulosa cells, such as TNFAIP6, EP2 PGE2 receptor and StAR (Elvin et al., 2000aGo; Varani et al., 2002Go; Fulop et al., 2003Go) may also prove to be predictive of embryo quality. Oocyte-secreted factors other than GDF9 may also mediate cumulus physiology. Bone morphogenic protein 15 (BMP15), which is largely uncharacterized in humans, has been shown to be essential for normal follicular development in sheep (Galloway et al., 2000Go; Juengel et al., 2002Go; Hanrahan et al., 2004Go). In contrast to results in mice, sheep with inactivating mutations of the BMP15 gene are infertile, with primary stage arrest of follicular development. Expression of downstream BMP15 targets may further define gene expression profiles in high grade embryos and be of additional clinical use.

A mutation in BMP15 has been described recently in two sisters with premature ovarian failure (Di Pasquale et al., 2004Go), indicating that the action of BMP15 is required for normal human folliculogenesis. Dysregulation of downstream targets of BMP15 or GDF9 may result in abnormal embryo development. Patients with markedly abnormal PTGS2, HAS2 or GREM1 expression may have an underlying defect in GDF9 or other oocyte-secreted mediators of cumulus physiology.

Ultimately, a finite set of cumulus genes may function as an assay for clinical utilization, creating an expression profile useful for embryo selection. In this study, the use of a single gene, GREM1, could accurately predict all three outcomes: maturity, fertilization potential and embryo quality. No additional predictive value was gained by including PTGS2 or HAS2 in the analysis. Cumulus cells are an ideal surrogate for the developmental potential of the oocyte as they are typically discarded tissues during IVF and represent a non-invasive means to identify embryo quality. Thus, cumulus expression profiling may be a parameter, other than morphology, by which to facilitate embryo selection.

The expression of these candidate genes (PTGS2, HAS2 and GREM1) can be correlated to morphological and physiological characteristics and may provide a novel approach to predict human embryo development. Ultimately, with better predictors of follicular and embryonic health, we can better select the embryos for transfer and reduce higher order pregnancy rates.



View larger version (10K):
[in this window]
[in a new window]
 
Figure 2. Relative expression of hyaluronic acid synthase 2 (HAS2) in low grade versus high grade embryos.

 


View larger version (21K):
[in this window]
[in a new window]
 
Figure 5. Receiver operating characteristics (ROC) curves for PTGS2, HAS2 and GREM1 expression as predictors of oocyte fertilization.

 

    Acknowledgements
 Top
 Abstract
 Introduction
 Methods and materials
 Results
 Discussion
 Acknowledgements
 References
 
These studies were supported by the following grants: American College of Obstetrics and Gynecology Ortho-McNeil Academic Fellowship (L.J.M), NIH grants HD32067 and HD33438 (M.M.M.) and a postdoctoral fellowship from the Center for Reproductive Training Grant HD007165 at Baylor College of Medicine (S.A.P.).


    References
 Top
 Abstract
 Introduction
 Methods and materials
 Results
 Discussion
 Acknowledgements
 References
 
Aaltonen J, Laitinen M, Vuojolainen K, Jaatinen R, Horelli-Kuitunen N, Seppa L, Louhio H, Tuuri T, Sjoberg J, Butzow R, Hovatta O, Dale L and Ritvos O (1999) Human growth differentiation factor 9 and its novel homolog growth differentiation factor-9B are expressed in oocytes during early folliculogenesis. J Clin Endocrinol Metab 84, 2744–2750.[Abstract/Free Full Text]

Blickstein I and Keith LG (2003) Outcome of triplets and high-order multiple pregnancies. Curr Opin Obstet Gynecol 15, 113–117.[CrossRef][ISI][Medline]

Buccione R, Vanderhyden BC, Caron PJ and Eppig JJ (1990) FSH-induced expansion of the mouse cumulus oophorus in vitro is dependent upon a specific factor(s) secreted by the oocyte. Dev Biol 167, 371–378.[CrossRef]

Calder MD, Caveney AN, Westhusin ME and Watson AJ (2001) Cyclooxygenase 2 and prostaglandin E2 receptor messenger RNAs are affected by bovine oocyte maturation time and cumulus oocyte complex quality, and PGE2 induces moderate expansion of the bovine cumulus in vitro. Biol Reprod 65, 135–140.[Abstract/Free Full Text]

Carabatsos M, Elvin J, Matzuk M and Albertini D (1998) Characterization of oocyte and follicle development in growth differentiation factor-9-deficient mice. Dev Biol 203, 373–384.[CrossRef]

Chang H, Brown CW and Matzuk MM (2002) Genetic analysis of the mammalian transforming growth factor-beta superfamily. Endocr Rev 6, 787–823.[CrossRef]

Davis BJ, Lennard DE, Lee CA, Tiano HF, Morham SG, Wetsel WC and Langenbach R (1999) Anovulation in cyclooxygenase deficient mice is restored by PGE2 and interleukin-1{beta}. Endocrinology 140, 2685–2695.[Abstract/Free Full Text]

Di Pasquale E, Beck-Peccoz P and Persani L (2004) Hypergonadotropic ovarian failure associated with an inherited mutation of BMP15 gene. Am J Hum Genet 75, 106–111.[CrossRef][ISI][Medline]

Dong J, Albertini DF, Nishimori K, Kumar TR, Lu N and Matzuk MM (1996) Growth differentiation factor-9 is required during early ovarian folliculogenesis. Nature 383, 531–535.[CrossRef][ISI][Medline]

Elvin J, Clark A, Wang P, Wolfman N and Matzuk M (1999a) Paracrine actions of growth differentiation factor-9 in the mammalian ovary. Mol Endocrinol 13, 1035–1048.[Abstract/Free Full Text]

Elvin JA, Yan C, Wang P, Nishimori K and Matzuk MM (1999b) Molecular characterization of the follicle defects in the growth differentiation factor 9-deficient ovary. Mol Endocrinol 13, 1018–1034.[Abstract/Free Full Text]

Elvin J, Yan C and Matzuk M (2000a) Oocyte expressed TGF-{beta} superfamily members in female fertility. Mol Cell Endocrinol 159, 1–5.[CrossRef][ISI][Medline]

Elvin JA, Yan C and Matzuk MM (2000b) Growth differentiation factor-9 stimulates progesterone synthesis in granulosa cells via a prostaglandin E2/EP2 receptor pathway. Proc Natl Acad Sci USA 18, 10288–10293.[CrossRef]

Eppig JJ (2001) Oocyte control of ovarian follicular development and function in mammals. Reproduction 122, 829–838.[Abstract/Free Full Text]

Fulop C, Szanto S, Mukhopadhyay D, Bardos T, Kamath RV, Rugg MS, Day AJ, Salustri A, Hascall VC, Glant TT and Mikecz K (2003) Impaired cumulus mucification and female sterility in tumor necrosis factor-induced protein-6 deficient mice. Development 130, 2253–2261.[Abstract/Free Full Text]

Galloway SM, McNatty KP, Cambridge LM, Laitinen MP, Juengel JL, Jokiranta TS, McLaren RJ, Luiro K, Dodds KG, Montgomery GW, Beattie AE, Davis GH and Ritvos O (2000) Mutations in an oocyte-derived growth factor gene (BMP15) cause increased ovulation rate and infertility in a dosage-sensitive manner. Nature Genet 25, 279–283.[CrossRef][ISI][Medline]

Hanrahan JP, Gregan SM, Mulsant P, Mullen M, Davis GH, Powell R and Galloway SM (2004) Mutations in the genes for oocyte-derived growth factors GDF9 and BMP15 are associated with both increased ovulation rate and sterility in Cambridge and Belclare sheep (Ovis aries). Biol Reprod 74, 900–909.

Juengel JL, Hudson NL, Heath DA, Smith P, Reader KL, Lawrence SB, O'Connell AR, Laitinen MP, Cranfield M, Groome NP, Ritvos O and McNatty KP (2002) Growth differentiation factor 9 and bone morphogenetic protein 15 are essential for ovarian follicular development in sheep. Biol Reprod 67, 1777–1789.[Abstract/Free Full Text]

Keith LG and Oleszczuk JJ (2002) Triplet births in the United States An epidemic of high-risk pregnancies. J Reprod Med 47, 259–265.[ISI][Medline]

Lim H, Paria BC, Das SK, Dinchuk JE, Lagenbach R, Trzaskos JM and Dey SK (1997) Multiple female reproductive failures in cyclooxygenase deficient mice. Cell 91, 197–208.[ISI][Medline]

Livak KJ and Schmittgen TD (2001) Analysis of relative gene expression using real-time quantitiative PCR and the 2–{Delta}{Delta}ct method. Methods 25, 402–408.[CrossRef][ISI][Medline]

Matzuk MM, Burns KH, Viveiros MM and Eppig JJ (2002) Intercellular communication in the mammalian ovary: oocytes carry the conversation. Science 21, 2178–2180.[CrossRef]

Munné S, Lee A, Rosenwaks Z, Grifo J and Cohen J (1993) Diagnosis of major chromosome aneuploidies in human preimplantation embryos. Hum Reprod 8, 2185–2191.[Abstract]

Munné S, Grifo J, Cohen J and Weier H (1994) Chromosome abnormalities in human arrested preimplantation embryos: a multiple probe FISH study. Am J Hum Genet 55, 150–159.[ISI][Medline]

Navot D, Bergh PA, Williams MA, Garrisi GJ, Guzman I, Sandler B and Grunfeld L (1991) Poor oocyte quality rather than implantation failure as a cause of age-related decline in female fertility. Lancet 337, 1375–1377.[CrossRef][ISI][Medline]

Pangas SA, Jorgez CJ and Matzuk MM (2004) Growth differentiation factor nine regulates expression of the bone morphogenic protein antagonist, gremlin. J Biol Chem 279, 32281–32286.[Abstract/Free Full Text]

Salustri A, Camaioni A and D'Alessandris C (1996) Endocrine and paracrine regulation of cumulus expansion. Zygote 4, 313–315.[ISI][Medline]

Vanderhyden BC, Macdonald EA, Nagyova E and Dhawan A (2003) Evaluation of members of the TGFbeta superfamily as candidates for the oocyte factors that control mouse cumulus expansion and steroidogenesis. Reproduction Suppl 61, 55–70.

Van Royen E, Mangelschots K, De Neubourg D, Laureys I, Ryckaert G and Gerris J (2001) Calculating the implantation potential of day 3 embryos in women younger then 38 years of age: a new model. Hum Reprod 16, 326–332.[Abstract/Free Full Text]

Varani S, Elvin JA, Yan C, DeMayo J, DeMayo FJ, Horton HF, Byrne MC and Matzuk MM (2002) Knockout of pentraxin 3, a downstream target of growth differentiation factor-9, causes female subfertility. Mol Endocrinol 16, 154–167.

Veek L (1999) An Atlas of Human Gametes and Conceptuses. Parthenon, New York.

Wolf D (1984) Human In Vitro Fertilization and Embryo Transfer. Plenum Publishing Corp,.

Yan C, Wang P, DeMayo J, DeMayo FJ, Elvin JA, Carino C, Prasad SV, Skinner SS, Dunbar BS, Dube JL, Celeste AJ and Matzuk MM (2001) Synergistic roles of bone morphogenetic protein 15 and growth differentiation factor 9 in ovarian function. Mol Endocrinol 15, 854–866.[Abstract/Free Full Text]

Submitted on June 29, 2004; accepted on August 27, 2004.