| HOME | HELP | FEEDBACK | SUBSCRIPTIONS | ARCHIVE | SEARCH | TABLE OF CONTENTS |
|
|
|
|||||||
|
|||||||||
| ||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||
Department of Animal Science, Cornell University, Ithaca, New York 14853
 |
ABSTRACT |
|---|
 TOP ABSTRACT INTRODUCTIONMATERIALS AND METHODSRESULTSDISCUSSIONREFERENCES |
|---|
follicle, granulosa cells, growth factors, ovary
|
INTRODUCTION |
|---|
|
TOPABSTRACTINTRODUCTIONMATERIALS AND METHODSRESULTSDISCUSSIONREFERENCES |
|---|
GDF9 has been observed in ovarian follicles of mice at all stages of differentiation except at the primordial stage [4]. In addition to results from gene inactivation studies, in vitro experiments have revealed many effects of GDF9 relating to differentiation and function of the ovarian follicle. Recombinant rat GDF9 increased the diameter of cultured rat preantral follicles to a level similar to that stimulated by FSH and in an additive manner when combined with FSH [6]. Follicular steroid synthesis [11, 12] is affected and inhibin B synthesis is enhanced [13] by GDF9. GDF9 also modulates 8-Br-cAMP-stimulated steroid production by cultured human granulosa and theca cells [9]. Furthermore, attenuation of gonadotropin-induced steroid production by GDF9 led Vitt et al. [14] to suggest that the oocyte factor may stimulate proliferation but also modulate differentiation.
A mammalian preovulatory follicle is characterized by a fluid-filled antrum with functional classification of the granulosa cells surrounding the oocyte into those closest (cumulus) and those more peripheral (mural) to the oocyte. In oviparous species, like the hen, a fluid-filled antrum is not observed. Instead, yolk accumulation by the rapidly growing oocyte is the most obvious aspect associated with follicle development. In this case, no follicular fluid accumulates and all granulosa cells are in contact with the oocyte [15]. The germinal disk exists as a small, white area on the surface of the oocyte [16] and contains the nuclear material and most cellular organelles. The granulosa cells of the hen oocyte are functionally compartmentalized into those closest to the germinal disk and those further removed [17]. The granulosa cells overlying the germinal disk are more rapidly dividing than peripheral granulosa cells, while the peripheral granulosa cells are more differentiated and produce more progesterone [17–19]. Therefore, although avian granulosa cells are arranged in a different way relative to the oocyte compared with mammalian granulosa cells, there are functional specializations of the granulosa cells in birds similar to mammals.
From a variety of studies, we know a great deal about follicle development in the hen [20] and have accumulated significant information on the mRNA [21] and protein [22] of TGFß family members expressed in the hen ovary. Comparative biology would suggest that the hen oocyte, although actively accumulating yolk, may express GDF9 and that this protein may be expressed in a similar manner and have a role in follicle development, as has been shown in mammals. Therefore, we investigated the occurrence, relative abundance, and localization of GDF9 in the hen. We hypothesized that there may be a regional difference in expression of GDF9 between the germinal-disk and non-germinal-disk area of the oocyte in light of previous data cited above. In addition to these studies, we examined the functional activity of a presumed GDF9 protein on hen granulosa cell proliferation.
|
MATERIALS AND METHODS |
|---|
|
TOPABSTRACTINTRODUCTIONMATERIALS AND METHODSRESULTSDISCUSSIONREFERENCES |
|---|
Single-comb White Leghorn hens of the Babcock B300-strain were individually caged, with egg records maintained at 2-h intervals during daylight. The hens had free access to water and feed and were maintained on a 15L:9D schedule (lights-on at 0600 h). Hens between 22 and 74 wk of age and laying regular sequences were selected and killed at 1.5–2 h after oviposition for collection of follicles. All animal procedures were approved by the Institutional Animal Care and Use Committee of Cornell University.
GDF9 Cloning
A partial sequence for a chicken GDF9 cDNA was identified in the National Center for Biotechnology Information database from a chicken expressed sequence tag (EST) (#BM439817; http://udgenome.ags.udel.edu/
cogburn/index.html) deposited by Dr. L. Cogburn (University of Delaware) and reported to be most similar to mammalian GDF9. Primers were designed (forward: 5'TAAGATGTACGCCACCAAGG3'; reverse: 5'ACACGATCCAGGTTGAAGAG3') on the basis of this sequence and used in real-time-polymerase chain reaction (RT-PCR) with total RNA isolated from hen ovary to generate a 162-base pair (bp) chicken cDNA for GDF9. This fragment was then subcloned in pGEM-TEasy Vector (Promega, Madison, WI) and XL1 Blue competent cells (Stratagene, LaJolla, CA), labeled and used in Northern blot analysis. On the basis of ovarian-specific expression, two clones (pgr1n.pk001.f24 and pgr1n.pk001.o6), from which the EST was derived, were obtained from Dr. Joan Burnside (University of Delaware) and repeated rounds of PCR-based amplification resulted in the sequence deposited in GenBank (accession #AY672110).
RNA Extraction
For Northern blot analysis, the liver, testis, granulosa cell layer from large (F1–F3) follicles, pituitary, brain, and the complete ovary (without the large yolky follicles) were removed from birds after being killed, immediately placed in a guanidine isothiocyanate solution and frozen at –80°C until RNA extraction. Tissues were homogenized and RNA was extracted using the guanidine isothiocyanate/phenol-chloroform method as previously described by Chomczynski and Sacchi [23]. For quantitative real-time PCR (n = 5 replicates), the liver and ovary were removed from killed hens. One healthy-appearing follicle from each size class of 6, 8, 10, and 12 mm was collected, the granulosa layer was separated from the theca layer, and the granulosa layer was bisected so that the germinal disk (G) was in one half and the opposite pole (N; nongerminal disk) in the other. Similarly, the granulosa layer was removed from the F1 and F4 follicles and an equivalent-sized piece of granulosa layer was taken from the G and N regions of the layer. The tissues were collected in Buffer RLT (supplied in Qiagen RNeasy Micro Kit) and extracted immediately or homogenized and stored at –80°C until later extraction. Smaller intact follicles of 1–3-mm diameter (7–15 per hen) and less than 1-mm diameter (45–60 per hen) were also collected as well as sections of the whole ovary and liver. Tissue was homogenized by vortexing and by aspirating the tissue through a 25-gauge needle approximately 40 times. RNA was extracted using the Qiagen RNeasy micro kit according to the manufacturer's instructions.
Northern Blot Analysis
Approximately 25 µg of total RNA were prepared, separated on a 1.5% denaturing formaldehyde gel, blotted by capillary transfer, and ultraviolet cross-linked to GeneScreen Plus nylon membrane (NEN, Boston, MA). Northern analysis was done with the chicken GDF9 cDNA (162-bp piece) labeled with 5 µCi 32P-dCTP (DuPont, Boston, MA) by a random primer kit (Prime-It II, Stratagene, La Jolla, CA) to a specific activity of approximately 1.0 x 109 cpm/µg. The membranes were hybridized with the chicken GDF9 cDNA probe at 42°C overnight, stringently washed, exposed to film (Kodak BioMax MS, Eastman Kodak, Rochester, NY), and the film developed as previously described [21].
Real-Time PCR
Probes and primers were designed using Primer Express Software v2.0 from Applied Biosystems. The primers defined a cDNA (within the 162-bp cDNA for chicken GDF9) of 88 bp that spanned an intron (forward: CTTTTCACCCCGTGTTCTGAGT; reverse: CCAGGTTGAAGAGCAATTCCA) and the probe (6FAMACCCGATTACAGGAGACTMGBNFQ) was designed to anneal at the exon/intron junction. Control reagents were TaqMan ribosomal RNA (18S) primers and probe and a control reaction was run for each sample. Preliminary Northern blot analysis had shown that GDF9 was abundantly expressed in total ovarian RNA, and therefore, RNA was extracted from five ovaries, subjected to reverse transcription to cDNA, and used to establish a standard curve for the assay. Serial dilutions of the cDNA were made to establish five points on the standard curve, which was run in triplicate. Samples of RNA (as described above) from the various tissues were also subjected to reverse transcription to cDNA. The reactions were performed (using an ABI Prism 7000 Sequence Detection System) in a 50-µl volume of reaction buffer containing 1x TaqMan Universal PCR Master Mix (Applied Biosystems, Foster City, CA) and 900 nM of the GDF9 primer pair or 50 nM of the 18S primer pair and each respective (GDF9 at 250 nM and 18S at 200 nM) TaqMan probe. Control reactions containing no template and reactions lacking reverse transcriptase were also used. All PCR amplifications were carried out in duplicate for each sample and the mean value was calculated relative to 18S reactions (also run in duplicate). The amount of mRNA and the estimated crossing point in a particular sample was determined by the comparative threshold cycle (CT) method using Sequence Detection System Software (Applied Biosystems).
Immunocytochemistry
Ovaries (with the large follicles removed) were isolated from hens and fixed in 10% buffered formalin. The ovaries were then embedded in paraffin and sectioned at 7 µm by the Histology Lab at the Cornell Veterinary School. Prior to immunocytochemistry, the slides were deparaffinized and rehydrated. Antigen retrieval was accomplished by boiling the slides in citrate buffer (0.01 M at pH 6.0) for 20 min. The tissues were blocked with 10% goat serum in PBS and incubated overnight at room temperature in a humidified chamber. The primary antiserum was made in a rabbit against the C-terminal portion of mouse GDF9 (JH131; obtained as a gift from Dr. S.-J. Lee at Johns Hopkins University) and was used at a dilution of 1:100 overnight at 5°C. Control slides were either incubated without the primary antibody or in the presence of normal rabbit serum. The second antibody was Alexa Fluor 488 goat anti-rabbit IgG (Molecular Probes, Inc., Eugene, OR) used at a dilution of 1:2000 for 1 h at 37°C. To identify nuclei, sections were incubated with propidium iodide (PI; 1 µg/ml) in PBS for 15 min at room temperature following secondary antibody incubation. Slides were examined with a Nikon Eclipse E600 microscope with fluorescence capability and the images captured with a Spot RT Slider camera.
Oocyte-Conditioned Medium
The large pedunculated follicles were removed from the ovary of a laying hen and discarded. Follicles less than or equal to 1 mm in size were dissected from the ovarian stroma, pooled, and incubated in 3 ml of M199 plus 0.1% BSA at 37°C for 2–3 days. At the end of the incubation period, the tubes were gently centrifuged (for 5 min at 2000 rpm) and the medium was removed, filter sterilized (0.2-µm membrane), and used immediately or frozen for subsequent use in cell culture (four different preparations were used in the cultures described below). Aliquots of medium were removed for Western blot analysis using the anti-mouse GDF9 antiserum (at a dilution of 1:5000) previously described (JH131). The medium was concentrated with acetone or a Micron-30 filter (Amicon) and run on a 10% SDS-polyacrylamide gel and transferred to nitocellulose. Detection of signal was with LumiGLO Western blot chemiluminescence (KPL, Gaithersburg, MD).
Effect of Oocyte-Conditioned Medium on Granulosa Cell Proliferation
To begin to investigate the function of GDF9 in the hen, we examined the effect of oocyte-conditioned medium on granulosa cell proliferation. Prehierarchical follicles (n = 12–16 follicles of 3- to 8-mm size) were removed from one hen for each replicate experiment (n = 4 experiments). The granulosa layers were isolated from each follicle, pooled, and dispersed as previously described [24]. Number of cells and viability (trypan blue exclusion) were estimated using a hemocytometer. Cell viability was 95% or greater at the start of an experiment. Cells were plated in 96-well plates (in M199 plus 5% fetal bovine serum) at a density of 1.2 x 105 cells/well and incubated for 24 h under the conditions that have been previously described [24], except that they were cultured in 100-µl volume without LipoMax Bovine Lipoprotein Supplement. After 24 h, the medium (which contained serum) was removed and replaced with M199 with 0.1% BSA and various doses of oocyte-conditioned medium. Oocyte-conditioned medium (OCM) was added in amounts of 0, 1, 5, 10, 25, or 50 µl out of 100 µl total volume. The plates were then returned to the incubator and cultured for an additional 48 h. Granulosa cell proliferation was quantified with Cell Titer 96 (Aqueous One Solution Cell Proliferation Assay; Promega) and visual inspection of plate confluence confirmed proliferation.
Specificity of OCM on Granulosa Cell Proliferation
In a second approach (n = 3 experiments), granulosa cells (after attachment as previously described) were treated with control medium, OCM (50 µl), GDF9 antiserum (50 µl of 1:100 dilution of JH131 antiserum), or GDF9 antiserum preincubated for 30 min at room temperature with OCM (50 µl). Granulosa cell proliferation was assessed at the termination of culture by the Cell Titer 96 assay.
Statistics
All data were analyzed with SAS using Proc GLM with protected least-significant difference. The significance level was P < 0.05.
|
RESULTS |
|---|
|
TOPABSTRACTINTRODUCTIONMATERIALS AND METHODSRESULTSDISCUSSIONREFERENCES |
|---|
|
In Figure 2, we have compared the chicken GDF9 amino acid sequence in the biologically active C-terminal region with that of several different mammals and show that it is highly conserved, with approximately 80% identity in sequence. There is about 65% similarity of the chicken sequence to full-length cDNA mammalian GDF9 sequences. In addition, there are six cysteine residues (indicated in Fig. 2) and three of the four potential N-glycosylation sites that appear to be conserved across species. Only one N-glycosylation site is indicated in Figure 2.
|
Quantitative real-time PCR was performed to assess expression of GDF9 mRNA with respect to follicle development (Fig. 3). After normalization to 18S expression, real-time PCR analysis indicated that GDF9 mRNA abundance differed with respect to the source of tissue (P < 0.001). Messenger RNA for GDF9 was most abundant in the very small follicles (<1 mm), with significant amounts in the 1- to 3-mm follicles. In addition, we found that GDF9 mRNA was expressed in total ovary RNA as well as in the granulosa layer (both G and N) of 6-mm follicles, but these values were not significantly different from those observed for larger follicles. There was no difference in expression of GDF9 mRNA between the G and N portions of the granulosa layer of follicles. Negligible mRNA for GDF9 was present in the granulosa layer from larger follicles and in the liver.
|
Immunocytochemistry revealed specific staining of the GDF9 antiserum in the oocyte within hen ovaries. Staining was not observed when normal rabbit serum was substituted for the primary antiserum or when the primary antibody was omitted. A representative picture is shown in Figure 4A. Strong fluorescence is seen in all oocytes in this section, the largest of which is estimated at about 300 µm. Smaller follicles are also observed in this section, all with strong staining in the oocyte. PI staining (Fig. 4B) of the same section permits the granulosa cell nuclei to be distinguished. Figure 4C represents an overlay of staining observed in Figure 4, A and B. In additional sections that were also stained with PI, cytoplasmic GDF9 staining could be clearly observed (white arrow) in the granulosa cell layer (Fig. 4D). GDF9 staining of a 6-mm small, yellow follicle is shown in Figure 4E. Strong staining is observed under the vitelline membrane (and therefore, adjacent to the granulosa cell layer), and only scattered staining is evident throughout the oocyte. Membrane-bound yolk platelets are faintly outlined in this oocyte (Fig. 4E), which was at the stage of accumulating yellow yolk and displacing the oocyte cytoplasm (with the GDF9) to the periphery. The negative control (normal rabbit serum) of the 6-mm follicle stained with PI is depicted in Figure 4F.
|
Western blot analysis showed a band at approximately 30 kDa in concentrated OCM (Fig. 5A). The effect of OCM on granulosa cell proliferation is seen in Figure 5B. There was a significant dose-related effect of OCM on granulosa cell proliferation (P < 0.001). The greatest proliferation was observed at the highest dose of OCM with less proliferation at lower doses. Because the proliferation assay (Promega Cell Titer 96) measures metabolic activity, we also observed the cultures for an increase in confluence, and this was consistent with the results of the proliferation assay (Fig. 6). The specificity of OCM on granulosa cell proliferation is seen in Figure 5C. Proliferation of granulosa cells in response to treatment of the cells with antiserum to GDF9 was not different from the control, whereas treatment with 50 µl of OCM significantly increased proliferation (P < 0.01), as expected from the dose-response study. Preincubation of OCM with the GDF9 antiserum inhibited the effect of OCM on granulosa cell proliferation.
|
|
|
DISCUSSION |
|---|
|
TOPABSTRACTINTRODUCTIONMATERIALS AND METHODSRESULTSDISCUSSIONREFERENCES |
|---|
Interestingly, an alternate sequence for chicken GDF9 has been posted on GenBank (accession #AY566700). This sequence has an insertion at position 725, which causes a frame shift. A deletion occurs between 761 and 762, which brings it back in line with our reported sequence (accession #AY672110) and with the chicken genomic sequence (chromosome 13, bp 16091112..0.16093510 from www.ensembl.org/Gallus_gallus). Significantly, all chicken sequences contained the same serine residue at position 77 of the mature protein. It has recently been reported that mutation of this site is associated with increased ovulation rate and sterility in certain breeds of sheep [25, 26].
The quantitative PCR analysis of the mRNA for chicken GDF9 indicates that expression is highest (of follicle sizes examined) in isolated, small follicles <1 mm in diameter. We also found GDF9 mRNA in samples of the total ovary (which had the largest follicles removed). This material contained follicles, many of which were much smaller than 1 mm, but the tissue was likely diluted by stroma and other ovarian cellular components. The major component of avian follicles is the oocyte. Growth from the 1-mm to 3-mm stage (not yet accumulating yellow yolk) is largely the result of increase in oocyte size, suggesting that GDF9 may be diluted as the oocyte grows. It is important to note that GDF9 mRNA analysis in the small (<1- and 1- to 3-mm) follicles includes the whole follicle because it is difficult to cleanly isolate granulosa cells from the oocyte at this early stage before the accumulation of yellow yolk. In contrast, mRNA analysis in the follicles with yellow yolk (6 mm and larger) involved only the granulosa layer because the cytoplasm of the oocyte is minimal compared with yolk at this time. For these reasons, we have used both RNA analysis and immunocytochemistry to localize GDF9 in the hen.
Our immunocytochemistry results show strong GDF9 staining in the follicles embedded within the ovary (<300 µm; Fig. 4C) and suggest functional importance of GDF9 from an early stage. High concentrations of GDF9 in follicles <1 mm may indicate a critical role for this hormone in early follicle organization and progression in oviparous species, as has been demonstrated in mammals [5, 27]. The peripheral localization of GDF9 protein in oocytes from small, yellow follicles (6 mm; Fig. 4E) suggests that the cytoplasmic GDF9, like many other cellular proteins, is displaced to the periphery [28] as yolk accumulates. That is, although the oocyte increases dramatically in size due to the accumulation of centrally located yolk platelets, the cytoplasmic layer becomes narrowed between the yolk and the vitelline membrane. With respect to GDF9, this may be functionally important because, as yolk accumulates, GDF9 is still localized in high concentrations adjacent to the granulosa cell layer, separated only by the vitelline membrane. The displacement of GDF9 to the periphery of the oocyte in the hen is somewhat different from a report on primate ovaries [10]. This study indicated no difference in oocyte localization correlated to follicle size and is likely related to the process of yolk accumulation in oviparous species and the possible evolutionary pressure for confinement of GDF9 close to the granulosa layer.
Our evidence of granulosa cell expression of GDF9, although low, is similar to reports on primates [9, 10] but contradicts reports from other species [4, 8] localizing GDF9 exclusively to the oocyte. Interestingly, we found a pattern of granulosa cell expression of GDF9 that was roughly inversely correlated with follicle size, although there was no difference in expression between the germinal- and non-germinal-disk areas. We cannot exclude the possibility that a small amount of oocyte RNA may have adhered to the vitelline membrane and surrounding granulosa cell layer. Our immunohistochemistry results (Fig. 4D), however, supported the finding of granulosa cell expression of GDF9 mRNA. We can only speculate that granulosa cell production of GDF9 may augment oocyte GDF9 as the oocyte cytoplasm becomes reduced to less than 0.1% of the cell's volume with increasing yolk accumulation [28]. In spite of these observations, it is clear that GDF9 is predominantly expressed by the oocyte in the hen.
Although GDF9 is peripherally located in the rapidly growing (accumulating yellow yolk) oocyte, it is localized adjacent to the granulosa cell layer. One of the clear effects of GDF9 in mammals is on granulosa cell proliferation [14]. Because the isolated oocytes from follicles <1 mm were a rich source of GDF9 mRNA in the quantitative PCR analysis, we used these as a source for oocyte-conditioned medium. To test the bioactivity on isolated granulosa cells, we used granulosa cells isolated from somewhat larger follicles (3–5 mm) because we can cleanly isolate granulosa cells at this size, while the population is less pure from smaller follicles. Conditioned medium from the very small oocytes was capable of stimulating granulosa cell proliferation in a dose-responsive manner. More importantly, the stimulation of proliferation was blocked by antiserum to GDF9, indicating a specific effect of GDF9 in the hen oocyte. It was important to examine the specificity of the effect because other oocyte factors, such as EGF [19] or possibly BMP15 [7, 29], may influence hen granulosa cell proliferation. In conclusion, this study supports the broad importance and conservation of GDF9 in follicle development because this is the first demonstration of this growth factor in an oviparous species. Future studies will examine the activity of GDF9 on other aspects of granulosa and thecal cell function because a coordinated regulation of the two cell types is essential in early follicle development [30].
|
ACKNOWLEDGMENTS |
|---|
|
FOOTNOTES |
|---|
2 Correspondence: Patricia A. Johnson, Cornell University, Ithaca, New York 14853. FAX: 607 255 9829; paj1{at}cornell.edu
Received: 4 October 2004.
First decision: 28 October 2004.
Accepted: 21 December 2004.
|
REFERENCES |
|---|
|
TOPABSTRACTINTRODUCTIONMATERIALS AND METHODSRESULTSDISCUSSIONREFERENCES |
|---|
-subunit mRNA and immunoreactive protein in cultured chicken granulosa cells. Gen Comp Endocrinol 1999 116:90-103[CrossRef][Medline]This article has been cited by other articles:
|
|
 |
|
  P.A Johnson, T.R Kent, M.E Urick, and J.R Giles Expression and Regulation of Anti-Mullerian Hormone in an Oviparous Species, the Hen Biol Reprod, January 1, 2008; 78(1): 13 - 19. [Abstract] [Full Text] [PDF]
|
|
|
|
 |
|
 S Elis, J Dupont, I Couty, L Persani, M Govoroun, E Blesbois, F Batellier, and P Monget Expression and biological effects of bone morphogenetic protein-15 in the hen ovary J. Endocrinol., September 1, 2007; 194(3): 485 - 497. [Abstract] [Full Text] [PDF]
|
|
| ||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||
| HOME | HELP | FEEDBACK | SUBSCRIPTIONS | ARCHIVE | SEARCH | TABLE OF CONTENTS |
