| HOME | HELP | FEEDBACK | SUBSCRIPTIONS | ARCHIVE | SEARCH | TABLE OF CONTENTS |
|
|
|
|||||||
|
|||||||||
| ||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||
Research Article |
Pest Animal Control Cooperative Research Centre,4 CSIRO Sustainable Ecosystems, Canberra, Australian Capital Territory 2615, Australia
ARC Centre of Excellence in Biotechnology & Development,5 Reproductive Science Group, University of Newcastle, Callaghan, New South Wales 2308, Australia
Monash Institute of Reproduction and Development,6 Monash University, Clayton, Victoria 3168, Australia
ABSTRACT
In rodent ovaries Kit ligand (KITL) and its receptor KIT have diverse roles, including the promotion of primordial follicle activation, oocyte growth, and follicle survival. Studies were undertaken to determine whether KITL and KIT carry out similar activities in rabbits.KitlandKitmRNA and protein were localized to oocytes and granulosa cells, respectively, in the rabbit ovary. Ovarian cortical explants from juvenile rabbits and neonatal mouse ovaries were subsequently cultured with recombinant mouse KITL and/or KITL neutralizing antibody. Indices of follicle growth initiation were compared with controls and between treatment groups for each species. Recombinant mouse KITL had no stimulatory effect on primordial follicle recruitment in cultured rabbit ovarian explants. However, the mean diameter of oocytes from primordial, early primary, primary, and growing primary follicles increased significantly in recombinant mouse KITL-treated explants compared with untreated tissues. In contrast, recombinant mouse KITL promoted both primordial follicle activation and an increase in the diameter of oocytes from primordial and early primary follicles in the mouse, and these effects were inhibited by coculture with KITL-neutralizing antibody. Recombinant mouse KITL had no effect on follicle survival for either species. These data demonstrate that KITL promotes the growth of rabbit and mouse oocytes and stimulates primordial follicle activation in the mouse but not in the rabbit. We propose that the physiologic roles of KITL and KIT may differ between species, and this has important implications for the design of in vitro culture systems for folliculogenesis in mammals, including the human.
follicle, folliculogenesis, granulosa cells, KIT, KIT ligand, oocyte development, ovary, primordial follicle, signal transduction
INTRODUCTION
Kit ligand (KITL) is a pleiotropic growth factor that exerts an influence on target cells through binding the tyrosine kinase receptor, KIT. KITL and KIT are expressed by a variety of developmentally distinct cell lineages during both embryogenesis and adult life, and roles for KITL and KIT in gametogenesis, melanogenesis, and hematopoiesis have been described [1–3]. In mice KITL and KIT are products of the steel and white spotting loci, respectively, and mutations at these loci are associated with reduced fertility or sterility, depending on the nature of the defect [4–7]. Numerous studies have shown that in the ovaries of rodents KITL and KIT are important for the migration, proliferation, and survival of primordial germ cells [8–15]; primordial follicle activation [4, 16, 17]; oocyte growth and survival [6, 18–22]; granulosa cell proliferation [20, 23]; the maintenance of meiotic arrest [24, 25]; theca cell recruitment; and the regulation of ovarian steroidogenesis [20, 26–28]. Therefore, KITL and KIT have diverse roles during oogenesis and folliculogenesis, and they are fundamentally important to fertility in rodents. However, many of these effects are yet to be confirmed in other species, including the human.
The expression of KITL and KIT is well characterized in the developing mouse ovary [29–36]. It has been reported that with the exception of oocytes undergoing the first stages of meiosis, in which Kit mRNA expression is downregulated, oocytes at all stages of development express Kit mRNA and protein [29–31, 33–36]. Similarly, the granulosa cells of follicles at all stages of development express Kitl mRNA and protein, although expression is very low in the squamous granulosa cells of primordial follicles [29, 31, 32, 34, 35]. Given that KITL is derived from granulosa cells and KIT is expressed by oocytes and theca cells, this receptor–ligand pair is an excellent example of how paracrine interactions between oocytes and their companion somatic cells control many aspects of follicle formation and development.
The controlled recruitment of primordial follicles into the growing follicle pool is a prerequisite for the development of mature oocytes and is a basic determinant of female reproductive fitness in all mammalian species. Despite this importance, the specific mechanisms that regulate primordial follicle survival, follicle activation, and the initiation of oocyte growth are largely unknown. Although a role for KITL and KIT in these processes is well established in rodents [4, 6, 16–22], there is a paucity of information regarding the functional significance of KITL and KIT during primordial follicle activation in other species. The limited information available for the human indicates that, similar to mice, the oocytes of primordial follicles in human fetal ovaries express KIT, and expression initially intensifies in growing preantral follicles, only to decrease again with antrum formation [37–39]. In contrast to the mouse, KIT protein also is expressed by the granulosa cells of newly formed human primordial follicles [37]. Additionally, KITL protein has been localized to the granulosa cells of primordial, preantral, and early antral follicles in human fetal ovaries [37]. However, data from Abir and colleagues suggested that KITL and KIT proteins were present in oocytes from both fetal and adult human ovaries, but neither protein was detected in granulosa cells [40]. Collectively, these expression profiles support both autocrine and paracrine roles for KITL and KIT during primordial follicle assembly and throughout folliculogenesis in the human.
Given the limited accessibility of human ovarian tissue and the lack of information available for the activity of KITL in species other than the mouse, we identified the rabbit as an alternative animal model for the study of the roles of KITL and KIT during the early stages of folliculogenesis. The initial objective of this research was to use in situ hybridization and immunohistochemistry to provide a detailed analysis of Kitl and Kit mRNA and protein expression within the ovaries of rabbits during the assembly of primordial follicles and the first wave of folliculogenesis, which occurs between Weeks 2 and 12 of postnatal development in this species [41–43]. These studies confirmed that the developmental expression patterns of KITL and KIT were consistent with those previously reported for mice and humans and suggest a role in primordial follicle activation and follicle growth in the rabbit.
Subsequently, using a novel organ culture system capable of supporting the survival and activation of primordial follicles, we found that incubation with recombinant mouse KITL significantly increased oocyte diameter and stimulated the recruitment of primordial follicles in neonatal mouse ovaries. In contrast, recombinant mouse KITL increased the diameter of oocytes but did not have a significant effect on primordial follicle recruitment in cultured rabbit ovarian cortical explants. On the basis of these findings we propose that the physiologic roles of KITL and KIT may differ between these two species. To the best of our knowledge this is the first identification of KITL/KIT in the rabbit and is the first study using a novel in vitro system suitable for the culture of rabbit primordial follicles to compare the roles of KITL during early folliculogenesis between two species.
MATERIALS AND METHODS
Materials
Unless otherwise stated, reagents were obtained from Sigma Chemical Co. (Australia).
Animals
All animal experimental procedures were approved by the CSIRO, Division of Sustainable Ecosystems, Gungahlin, Animal Experimentation Ethics Committee in accordance with National Health and Medical Research Council/CSIRO guidelines. Experiments were conducted using female BALB/C mice and New Zealand White rabbits (Oryctolagus cuniculus) bred and housed in temperature- and light-controlled rooms and given food and water ad libitum. Mice were killed by halothane inhalation (Halocarbon Laboratories, River Edge, NJ), and rabbits were killed by barbiturate (150 mg/kg body weight) overdose (Lethabarb; Jurox Pty. Ltd., Australia).
Molecular Cloning of Rabbit Kitl and Kit
Total RNA for cDNA synthesis was isolated from rabbit ovaries using TRIzol Reagent (Invitrogen Corp., Australia) in accordance with the manufacturer's instructions. First-strand cDNA was synthesized from 1 µg total RNA using Expand Reverse Transcriptase (Roche Diagnostics, Australia) and oligo(dT)15 primer. Kitl and Kit sequences were amplified by PCR using High Fidelity Taq Polymerase (Roche Diagnostics) and primers (see Supplemental Table 1, available online at: http://www.biolreprod.org/) based on conserved sequences between mouse and human Kitl and Kit. Conditions for PCR were as follows: 1x (3 min at 94°C); 30x (94°C for 45 sec, 55°C for 1 min, 72°C for 1 min); 1x (72°C for 5 min). PCR products were ligated into the pGEM-T Easy vector (Promega Corp., Australia) and cloned into XL-1 Blue Escherichia coli (Stratagene, La Jolla, CA). Cloned cDNAs were sequenced (Biomolecular Resource Facility, JCSMR, ANU), and nucleotide and amino acid sequence analyses were performed using Lasergene v6 software (DNASTAR Inc., Madison, WI). Sequences were deposited in GenBank, and accession numbers for rabbit Kitl DQ356265, rabbit Kit (cytoplasmic domain) DQ356267, and rabbit Kit (extracellular domain) DQ356266 were assigned respectively.
|
Recombinant Protein Production
Two bacterial recombinant proteins were generated for the production of antisera: 1) KITL1–150 comprising amino acids 1–150 of rabbit KITL and 2) KIT152 comprising 152 amino acids of rabbit KIT (located in the extracellular domain corresponding to amino acids 291–442 of human KIT, NP_000213). Additionally, a third recombinant protein KITL1–189, which comprises amino acids 1–189 of the mature KITL, was generated for use in functional studies. Expression plasmids were constructed using Gateway Cloning Technology (Invitrogen Corp.), and His-tagged recombinant proteins were expressed in bacterial hosts in accordance with the manufacturer's instructions. PCR primers, plasmids, and bacteria used for cloning and recombinant protein expression are described (see Supplemental Table 2, available online at: http://www.biolreprod.org/). Recombinant proteins were affinity purified with Ni-NTA resin (Qiagen, Australia) and renatured by dialysis using protocols described in The QIAexpressionist (Fifth edition, Qiagen; http://www1.qiagen.com/literature/handbooks/PDF/Protein/Expression/QXP_QIAexpressionist/1024473_QXPHB_0603.pdf).
|
Protein expression was monitored by SDS-PAGE followed by Coomassie Brilliant Blue staining (BioRad). Bands of predicted size showing increased intensity in induced compared with uninduced samples were assumed to represent recombinantly expressed protein. The identities of KITL1–150 and KITL1–189 also were verified by Western blot using a commercially obtained antibody directed against mouse KITL (recombinant mouse SCF; Sigma Chemical). Commercial antibodies directed against the expressed region of rabbit KIT were not available.
Polyclonal Antisera Production
For the production of polyclonal antibodies, female rats (n = 12) were immunized sub-cutaneously with 100 µg purified KITL1–150 or KIT152 dissolved in PBS (50 µl) and emulsified in an equal volume of Freunds Complete Adjuvant. Two booster injections of protein (50 µg), dissolved in PBS (50 µl) and emulsified in an equal volume of Freunds Incomplete Adjuvant, were given at 2-wk intervals, commencing 4 wk after priming injections. Fourteen days following the final boost, rats were anesthetized with carbon dioxide and then exsanguinated by cardiac puncture. Blood sera were collected and tested for specificity against the respective immunogens and commercially obtained protein (recombinant mouse SCF; Sigma Chemical Co.) by Western blot (not shown).
Immunohistochemistry
Ovaries for immunohistochemistry were obtained from female rabbits (four to five in each age group) aged 2, 4, 6, 8, and 12 wk. Ovaries were fixed overnight in 4% (w/v) paraformaldehyde at 4°C, embedded in paraffin, and sectioned at 7 µm. Sections (four for each ovary) were deparaffinized, rehydrated, and microwaved for 10 min in 10 mM sodium citrate (pH 6), followed by incubation with 0.01% (w/v) trypsin for 5 min at 37°C for antigen retrieval. Sections were blocked for 1 h in 3% (w/v) BSA in PBS at 37°C before polyclonal antisera (rat anti-rabbit KIT152 and rat anti-rabbit KITL1–150), diluted 1:50 to 1:200 in PBS containing 1% BSA, and they were applied to sections for 1 h at 37°C. The tissues were then incubated with biotin-conjugated goat anti-rat immunoglobulin G (IgG; Sigma Chemical Co.) diluted 1:1000 in PBS for 1 h at 37°C. Antibody binding was visualized using the ImmunoPure ABC Peroxidase Staining Kit (Pierce, Rockford, IL) and the Liquid DAB-Plus Substrate Kit (Zymed, San Francisco, CA). Sections were counterstained with hematoxylin and visualized by light microscopy using an Olympus BH-2 microscope. Negative controls, in which the primary antibody was replaced with preimmune serum, were routinely performed on adjacent serial sections, and no specific staining was observed.
In Situ Hybridization
Rabbit Kitl- and Kit-specific sense and antisense dioxigenin (DIG)-labeled riboprobes were generated from plasmid templates (see Supplemental Table 1, available online at: http://www.biolreprod.org/) using a DIG RNA Labeling Kit (SP6/T7) (Roche Diagnostics) in accordance with the manufacturer's instructions. Ovaries for in situ hybridization were obtained from female rabbits (four to five in each age group) aged 2, 4, 6, 8, and 12 wk. Ovaries were fixed overnight in 4% (w/v) paraformaldehyde at 4°C, embedded in paraffin, and sectioned at 7 µm. The tissue sections (four from each ovary) were deparaffinized in xylene and then rehydrated in alcohol. Pretreatment, hybridization, and posthybridization washes were performed according to protocols described in the Non-radioactive In Situ Hybridization Application Manual (Roche Diagnostics). Kitl sense and antisense probes were hybridized overnight at 50°C. Kit sense and antisense probes were hybridized overnight at 68°C. Probes were detected immunohistochemically using the DIG Nucleic Acid Detection Kit (Roche Diagnostics) according to protocols described by the manufacturer. The development of a dark purple-brown precipitate with the antisense but not the sense probes was interpreted as a positive reaction.
Organ Culture
Ovaries were obtained from eighteen 3-day-old BALB/c mice and four 8-wk-old domestic rabbits. Each rabbit ovary was dissected into six to eight explants (1 mm3), and the underlying stroma was trimmed to leave the primordial follicle-rich cortical region. Four mouse ovaries and four rabbit ovarian explants were randomly selected for immediate fixation and were then serially sectioned at 7 µm for baseline counts of every third section (Preculture). The remaining ovaries and explants were used for culture.
Ovaries and explants were cultured on collagen I- and fibronectin-coated nitrocellulose membranes (0.45-µm pore size; Millipore Corp., Bedford, MA) and mounted on triangular metal grid supports in Falcon Organ Culture Dishes (Becton Dickinson Co., Franklin Lakes, NJ). Coating was performed by incubating the membranes in 0.75 mg/ml collagen and 0.1 mg/ml fibronectin for 1 h at 37°C. Ovaries and ovarian explants were cultured in groups of three to five for 8 days in Dulbecco Modified Eagle medium (DMEM)/Ham F12 1:1 (v/v) (GibcoBRL, MD) supplemented with 10% fetal bovine serum, 0.03 IU recombinant FSH (Gonal-F 75; Serono, Italy), 5 µg/ml insulin, 5 µg/ml transferrin, 5 ng/ml selenium, and 0.1 µM retinoic acid.
Ovaries and ovarian cortical explants were maintained in organ culture for 8 days and were treated with either no treatment (0 KL), 50 ng/ml recombinant mouse KITL (50 KL), 150 ng/ml recombinant mouse KITL (150 KL), a combination of 150 ng/ml recombinant mouse KITL and 50 µg/ml recombinant mouse KITL-neutralizing polyclonal antibody (150 KL + anti), or 50 µg/ml neutralizing antibody alone (0 KL + anti). Both the recombinant mouse KITL protein and neutralizing antibody were obtained from Sigma Chemical Co. Doses of recombinant mouse KITL were selected on the basis of concentrations previously shown to promote oocyte growth and follicle activation in the mouse and rat, respectively [16, 18, 19]. The dose of neutralizing antibody was determined using the ED50 provided by the manufacturer. In similar experiments rabbit ovarian explants were incubated with 150 ng/ml of recombinant rabbit KITL (see Supplemental Table 2, available online at: http://www.biolreprod.org/).
Media containing supplements were refreshed every other day, and tissues were maintained in vitro at 37°C in an atmosphere of 5% CO2 in air. At least five rabbit ovarian cortical explants and six mouse ovaries were used within each cultured treatment group, and two independent experiments for each treatment were performed.
Classification of Follicles and Assessment of Survival and Growth
To enumerate follicle populations, tissues were fixed in Bouin solution, embedded in paraffin, serially sectioned at a thickness of 7 µm, and then stained with hematoxylin and eosin. In every third section follicles were counted and scored for stage of development, and the diameter of each oocyte was measured across the largest cross-section. Of follicles deemed healthy, only those with a visible germinal vesicle were counted or measured. The mean diameter of germinal vesicles for the mouse ranged from 9.6 ± 0.01 µm to 19.8 ± 2.6 µm (primordial to preantral follicles), and for the rabbit from 19.2 ± 0.01 µm to 22.2 ± 0.03 µm (primordial to primary follicles). Therefore, this strategy ensured that follicles were not counted twice. Additionally, when follicles with larger germinal vesicles (e.g., rabbit early primary and preantral follicles) were present, adjacent sections were checked to make certain the same follicle was not counted twice. Measurements were determined using an eyepiece graticule and an Olympus BH-2 microscope. Follicles were assessed blindly in all experiments.
A follicle was classified as being primordial if all the surrounding granulosa cells were flattened. If one or more granulosa cells were cuboidal then the follicle was considered to be at the early primary stage. Primary follicles were those surrounded by a complete layer of cuboidal granulosa cells. Growing primary follicles had multiple layers of granulosa cells along one side and a single layer on another. Preantral follicles had two or more complete layers of granulosa cells in the absence of an antrum. Small antral follicles had patchy fluid-filled zones between the granulosa layers, whereas in large antral follicles the fluid patches had merged into a single large cavity. Antral follicles were not observed in cultured ovaries and explants.
Follicles were classified as atretic if the oocyte was eosinophilic or had an irregular membrane. Most of the atretic follicles contained eosinophilic oocytes, and their developmental stages could not be determined due to advanced degeneration. Given that germinal vesicles often could not be visualized, all atretic follicles were counted; consequently, total atretic follicle numbers may be overestimated. In the case of mouse ovaries, the dying follicles were located toward the center of the ovary. In cases involving rabbit ovarian cortical explants, the dying follicles were most often located at the edge of the explant where manual dissection had occurred. A considerably higher proportion of follicles in rabbit ovarian cortical explants degenerated in culture compared with follicles in mouse ovaries, possibly due to the extensive dissection undertaken to generate rabbit ovarian cortical explants which was not required for the preparation of mouse ovaries for culture. It is also possible that the culture system employed did not adequately support the development of later-stage rabbit follicles. However, given that at the initiation of culture the majority of follicles were in the primordial and early primary stage, whereas at the end of the culture period healthy multilayered growing follicles were often observed, we suggest that the growth and development of early rabbit follicles in vitro was supported by this culture system. A definite role for KITL in later follicle development would need to be addressed by the use of an isolated follicle rather than the explant culture system described above.
Statistical Analysis and Presentation of Data
Statistical analysis was performed using Genstat software version 6.1 (VSN International, Herts, UK). A total of 22 594 healthy mouse follicles and 14 157 healthy rabbit follicles were evaluated. The results were obtained by pooling counts and measurements from the ovaries and explants within each treatment group from two independent experimental replicates. For continuous measurements such as oocyte diameter, generalized linear models (GLMs) were fitted by least squares, and generalized linear mixed models (GLMMs) were fitted using residual maximum likelihood [44, 45]. Binomial GLMs and GLMMs with a logistic link function were fitted to data in the form of counts out of a set total, and ordinal regression [45] was used to analyze data consisting of counts at various stages of development where there is a natural ordering of the data. F-tests, Wald tests, and chi-square tests were used to assess the significance of observed effects, and differences among means were determined by the analysis of 95% confidence intervals. Means with confidence intervals that do not overlap were considered statistically significant at the 5% level (P < 0.05). Growing primary follicle and preantral follicle data were pooled for the plots expressing the percentage of each follicle stage present in the ovaries and explants.
RESULTS
Characterization of Rabbit Kitl cDNA
The nucleotide sequence for the coding region of rabbit Kitl was 91%, 90%, 89%, and 89% identical at the nucleotide level to human [46], mouse [47], rat [46], and bovine [48] Kitl, respectively. The deduced amino acid sequence of the cDNA was 86%, 86%, 84%, and 84% identical to human, mouse, rat, and bovine KITL, respectively (see Supplemental Fig. 1, available online at: http://www.biolreprod.org/). These data indicate substantial interspecies conservation at gene and protein levels. Based on comparison with these highly conserved homologs, the deduced rabbit KITL amino acid sequence corresponds to a pro-protein of 273 amino acids, with an N-terminal signal sequence of 25 amino acids and a 248-amino acid mature protein. The mature protein comprises an extracellular, transmembrane, and short hydrophobic cytoplasmic domain. Exon six, encoding the proteolytic cleavage site for release of soluble protein, was highly conserved within rabbit KITL, with only one amino acid substitution (Arg177
Ser177) detected. Four cystine residues (amino acids 4, 43, 89, and 138) are required for the formation of KITL's intramolecular disulphide bonds [49] and also were conserved within the deduced amino acid sequence. Moreover, a series of three lysine residues (amino acids 213, 214, and 215) adjacent to the transmembrane domain of KITL are present in the rabbit, mouse, rat, and bovine amino acid sequence. These charged residues may serve to anchor the mature protein in the membrane [50] (see Supplemental Fig. 1, available online at: http://www.biolreprod.org/).
|
Localization of Kitl mRNA by In Situ Hybridization
In 2-wk-old rabbit ovaries, Kitl mRNA expression was very low close to the medulla of the ovary, where newly formed primordial follicles were located (Fig. 1a). In contrast, expression was high in the periphery of the ovarian cortex, where naked oocytes were clustered together in germ cell nests (Fig. 1, a and b). Although the naked oocytes themselves did not express Kitl mRNA, expression was detected within the somatic cells associated with germ cell nests (Fig. 1b). Somatic cells in close proximity to the ovarian epithelium exhibited particularly high levels of expression (Fig. 1b). High levels of Kitl mRNA also were observed within the cells of the ovarian epithelium (Fig. 1b). Kitl mRNA was rarely detected within the granulosa cells enclosing newly forming and developing follicles at this age (Fig. 1a).
The pattern of staining in ovarian tissue from 4- to 12-wk-old rabbits remained the same among age groups, with comparatively low levels of expression in primordial follicles, high expression levels in growing preantral follicles, and variable expression levels in antral follicles (Fig. 1, c–e). Kitl mRNA occasionally was observed in the flattened granulosa cells surrounding some primordial follicles (Fig. 1, c and d). In early primary follicles the cuboidal granulosa cells nearly always expressed Kitl (Fig. 1d), whereas the flattened granulosa cells within the same follicle often remained free of staining (Fig. 1d). Growing follicles from the primary stage onward expressed high levels of Kitl in almost every granulosa cell (Fig. 1e). In some large antral follicles (not shown), low levels of Kitl were expressed in mural granulosa cells, and the level of Kitl expressed in cumulus granulosa cells varied between different follicles from moderate to negligible. The ovarian epithelial cells also expressed high levels of Kitl mRNA. All follicles exhibited little or no staining, regardless of the developmental stage and rabbit age, when the antisense probe was substituted with the sense probe and hybridizations were carried out under the same conditions (Fig. 1f).
KITL Protein Expression in Juvenile Rabbit Ovaries
Similar to the results obtained by in situ hybridization, in the 2-wk-old rabbit ovary intense KITL immunostaining was observed in the outer cortex, with considerably less staining evident toward the medulla (not shown). KITL protein was strongly localized to the somatic cells associated with nests of naked oocytes, whereas the oocytes within the nests exhibited very diffuse staining. The somatic cells of the stroma located between the clusters of germ cells did not stain for KITL protein. KITL immunoreactivity was weakly detected in the granulosa cells and oocytes of the newly formed follicles located toward the medulla of the 2-wk-old ovary. KITL protein was detected in some cells of the stroma at the very center of the ovary, as well as in ovarian epithelial cells.
The distribution and intensity of KITL immunostaining did not change between ovaries from 4- to 12-wk-old rabbits. In these ovaries KITL protein was detected in some of the squamous granulosa cells surrounding primordial follicles, but not all primordial follicles exhibited KITL-positive granulosa cells (Fig. 2, a and b). In some primordial follicles clumps of strongly immunoreactive protein could be detected within the oocyte and at the oolemma (Fig. 2b). The cuboidal granulosa cells of most activated and growing follicles expressed high levels of KITL protein (Fig. 2, c and d). No difference in granulosa cell staining intensity was detected within or between preantral follicles. In contrast, the granulosa cells of large antral follicles appeared to express varying levels of KITL (not shown). In most large antral follicles, granulosa cell KITL expression was very low, whereas in very few follicles at this stage low to moderate KITL protein expression was evident in both mural and cumulus granulosa cells (not shown). Protein also was weakly detected within the oocytes of some follicles at all stages of follicular development (Fig. 2, b–d). The occasional stromal cell, as well as the ovarian epithelial cells, also stained positively for KITL. Specific staining was not detected when the antiserum was substituted with preimmune serum.
|
Localization of Kit mRNA by In Situ Hybridization
In 2-wk-old ovarian tissue, Kit mRNA expression was detected within the naked oocytes located closest to the medulla and in the oocytes of newly formed primordial follicles (Fig. 3, a and b). However, Kit mRNA expression was not detected in naked oocytes in the outermost region of the cortex (Fig. 3, a and b), which is consistent with oocytes entering the early stages of meiosis. While some of the germ cell clusters in the outer cortex were entirely devoid of staining, others contained a mixture of Kit-positive and Kit-negative oocytes (Fig. 3b). Kit mRNA was not detected in any somatic cells within 2-wk-old rabbit ovarian tissue (Fig. 3, a and b).
|
The pattern and intensity of staining following hybridization of the KIT riboprobe did not change between ovaries from 4- to 12-wk-old rabbits. High levels of Kit mRNA were detected in all of the oocytes of primordial and growing preantral follicles (Fig. 3, c–e). The oocytes from antral follicles appeared to express reduced levels of Kit mRNA compared with earlier stage follicles (Fig. 3f). Although the ovarian epithelium and somatic cells of the stroma remained free of staining, Kit mRNA expression was detected at moderate levels in the theca cells of antral follicles but not preantral follicles (Fig. 3, d–f). All follicles exhibited little or no staining, regardless of the developmental stage and age, when the antisense probe was substituted with the sense probe, and hybridizations were carried out under the same conditions (Fig. 3, g and h).
KIT Protein Expression in Juvenile Rabbit Ovaries
At 2 wks of age, KIT protein was detected at very low levels in the outer cortex of the ovary, whereas higher concentrations of the protein were localized toward the inner region of the ovary (not shown). Very weak KIT immunostaining was observed within naked oocytes located closest to the ovarian epithelium. Closer to the medulla the intensity of KIT immunoreactivity was variable, with KIT protein expression detected in some naked oocytes. Within these oocytes, KIT was localized both to the cytoplasm as diffuse staining and to the oolemma, where staining tended to be more intense. The oocytes of the newly formed primordial and developing follicles located toward the center of the 2-wk-old rabbit ovary expressed high levels of KIT protein in their cytoplasm and at the oolemma. KIT protein was not detected in the ovarian epithelia, stroma, or granulosa cells.
In ovaries from 4- to 12-wk-old rabbits KIT protein was detected in the oocytes of every follicle at all stages of development (Fig. 4, b–d). KIT was localized as a discrete ring at the membrane surface and also in the cytoplasm of oocytes, comprising both primordial and developing follicles (Fig. 4, b and c). Many oocytes of antral follicles appeared to express reduced levels of KIT protein compared with those of smaller growing follicles (Fig. 4b). Intensely stained granules also were detected at the oolemma and in the cytoplasm of oocytes from primordial through antral follicles (Fig. 4, b–d). KIT protein was weakly detected in the theca cells of some preantral and antral follicles (not shown). All other somatic cells within the ovary were free of KIT immunoreactivity. Specific staining was not detected when antiserum was substituted with preimmune serum (Fig. 4a).
|
Role of KITL During Early Folliculogenesis in the Rabbit
In situ hybridization and immunolocalization studies demonstrated that KITL is expressed by the granulosa cells of some primordial follicles, and KITL expression increases during follicle activation and the formation of cuboidal granulosa cells. In contrast, KIT is expressed by all oocytes. The selective distribution of KITL in primordial follicles is consistent with the hypothesis that KITL availability is a limiting factor in follicle activation. Therefore, an in vitro approach was employed to investigate the effect of recombinant mouse KITL on follicle survival, follicular recruitment, and oocyte growth in rabbit ovarian cortical explants.
Effects of Culture and Recombinant Mouse KITL Treatment on the Survival of Follicles in Rabbit Ovarian Cortical Explants
In preculture rabbit ovarian cortical explants, follicles were generally healthy and showed no signs of atresia. Following 8 days in vitro, approximately 50% of follicles exhibited signs of degeneration (Fig. 5). However, the proportion of total follicles that were healthy in the rabbit ovarian cortical explants did not differ significantly between the cultured treatment groups (Fig. 5). These results indicate that follicle survival was not enhanced by recombinant mouse KITL (50 ng/ml and 150 ng/ml), nor was follicular atresia promoted when the KITL/KIT interaction was inhibited by KITL-neutralizing antibody.
|
Effects of Culture and Recombinant Mouse KITL Treatment on the Distribution of Follicles in Rabbit Ovarian Cortical Explants
The proportion of healthy follicles at each stage of development in preculture and cultured rabbit ovarian cortical explants was calculated in order to assess the effects of both culture itself and recombinant mouse KITL treatment on follicle activation and development (Fig. 6). These analyses found that preculture rabbit ovarian cortical explants contained 60.4% ± 3.2% primordial follicles, 35.4% ± 2.7% early primary follicles, 3.1% ± 0.5% primary follicles, and 1.2% ± 0.2% preantral follicles (Fig. 6). Following 8 days of culture in vitro, the developmental distribution of follicles in cortical explants changed significantly (P < 0.05). In untreated cultured explants, the primordial follicle population decreased to comprise only 25.2% ± 2.7% of all of the follicles present, and the percentages of early primary, primary, and preantral follicles increased significantly (P < 0.05) to 58.1% ± 1.4%, 11.6% ± 1.4%, and 5.2% ± 0.8%, respectively (Fig. 6). These data indicate that a large proportion of rabbit primordial follicles spontaneously activate and develop in vitro.
|
The effect of exogenous KITL on follicle recruitment was investigated by treating explants with 50 ng/ml and 150 ng/ml recombinant mouse KITL. Neither dose of recombinant mouse KITL induced a detectable effect on the distribution of follicles in the resting or growing pools compared with untreated cultured explants (Fig. 6). Therefore, recombinant mouse KITL did not promote the activation of primordial follicles in rabbit ovarian cortical explants. Similarly, treatment of rabbit ovarian explants with 150 ng/ml recombinant rabbit KITL did not promote the activation of primordial follicles (see Supplemental Fig. 2, available online at: http://www.biolreprod.org/).
To determine whether endogenous KITL was responsible for the spontaneous activation of primordial follicles, rabbit ovarian cortical explants were treated with KITL-neutralizing antibody. There was no change in the developmental distribution of follicles in explants treated with neutralizing antibody compared with untreated cultured explants (Fig. 6). Therefore, neutralizing antibody did not prevent spontaneous follicle activation. There was, however, a significant (P < 0.05) decrease in the proportion of primordial follicles and an increase in primary follicles in neutralizing antibody-treated explants compared with explants treated with 150 ng/ml recombinant mouse KITL but not with any other cultured treatment group (Fig. 6).
Surprisingly, the addition of a combination of 150 ng/ml recombinant mouse KITL and neutralizing antibody resulted in a small but significant (P < 0.05) decrease in the percentage of primordial follicles compared with untreated explants and explants treated with recombinant mouse KITL (50 ng/ml and 150 ng/ml) alone (Fig. 6). This was not paralleled by a change in the percentage of early primary follicles, which remained the dominant follicle population. However, there was an increase in the proportions of primary and preantral follicles compared with untreated and recombinant mouse KITL-supplemented explants, although this difference was not statistically significant in all pairwise comparisons (Fig. 6). These data suggest that the inhibitory antibody possibly enhances receptor signaling.
Effects of Culture and Recombinant Mouse KITL Treatment on Oocyte Growth in Rabbit Ovarian Cortical Explants
We next measured and scored oocytes for developmental stage in order to determine the effects of culture and recombinant mouse KITL treatment on the growth of rabbit oocytes.
The culture process itself was shown to increase the size of oocytes from follicles at the primordial and early primary stages. In untreated cultured tissue, the mean diameters of oocytes of primordial and early primary follicles were significantly (P < 0.05) larger than their preculture counterparts (Table 1). Conversely, the mean oocyte diameters of primary, growing primary, and preantral follicles were either unchanged or smaller (P < 0.05) than in preculture tissues (Table 1).
In order to determine whether endogenous KITL had a role in promoting the growth of oocytes from primordial and early primary follicles that was observed during in vitro culture, neutralizing antibody to recombinant mouse KITL was added to the culture media. It was anticipated that if endogenous KITL were inducing growth in follicles maintained in vitro, then neutralization of the ligand would result in primordial and early primary follicle diameters comparable to those in preculture tissues. However, antibody added alone (i.e., in the absence of exogenous KITL) resulted in increased oocyte diameters (P < 0.05) in primordial, early primary, and primary follicles compared with those in untreated cultured explants (Table 1). The stimulatory effect was lost at the growing primary and preantral follicle stages. The mechanisms responsible for the induction of early follicle growth in the presence of antibody are unknown. Possibly, the neutralization of endogenous KITL permits other factors present in the culture media that affect follicle and oocyte growth to exert effects which would normally be masked by the presence of KITL.
Comparisons of oocyte sizes between untreated and recombinant mouse KITL-treated cultured explants were made in order to assess the effect of exogenous recombinant mouse KITL on oocyte growth in vitro. Recombinant mouse KITL was shown to have a major effect on the size of oocytes at the earliest stages of development (Table 1). The mean diameter of primordial follicle oocytes increased significantly (P < 0.05), from 31.7 ± 0.2 µm in untreated cultured explants, to 34.6 ± 0.3 µm in explants treated with 50 ng/ml recombinant mouse KITL, to a final diameter of 36.3 ± 0.1 µm in 150 ng/ml recombinant mouse KITL-treated explants (Table 1). In addition, KITL-neutralizing antibody antagonized (P < 0.05) the increase in oocyte diameter caused by 150 ng/ml recombinant mouse KITL (Table 1).
Similar to primordial follicles, the mean oocyte diameters of early primary, primary, and growing primary follicles increased with exposure to rising doses of recombinant mouse KITL (Table 1). Surprisingly, the stimulatory effects of recombinant mouse KITL at these stages of development were not mitigated by the addition of neutralizing antibody to the culture media (Table 1). Why antiserum should partially inhibit the effects of recombinant mouse KITL in primordial follicles but not at later developmental stages is unclear.
Recombinant mouse KITL (50 ng/ml and 150 ng/ml) had no stimulatory effect on the diameter of oocytes in the preantral follicle population compared with untreated cultured explants (Table 1).
Role of KITL During Early Folliculogenesis in the Mouse
Neither recombinant mouse KITL (Fig. 6) nor recombinant rabbit KITL (see Supplemental Fig. 2, available online at: http://www.biolreprod.org/) promoted the activation of primordial follicles in rabbit ovarian cortical explants. Therefore, we sought to confirm reports in the literature that KITL directly promotes primordial follicle activation and supports continued follicle development in rodents [16–18], thus validating our rabbit explant culture system.
Effects of Culture and Recombinant Mouse KITL Treatment on the Survival of Follicles in Mouse Ovaries
A total of 97.1% ± 1.5% of all follicles in preculture mouse ovaries were healthy (Fig. 7). Following 8 days of culture, the proportion of healthy follicles declined significantly (P < 0.05) to 82.5% ± 3.7% (Fig. 7). The proportions of healthy follicles within cultured mouse ovaries treated with recombinant mouse KITL (50 ng/ml and 150 ng/ml), a combination of 150 ng/ml recombinant mouse KITL and anti-recombinant mouse KITL antibody, and anti-recombinant mouse KITL antibody alone did not vary significantly from the untreated culture group (Fig. 7). Therefore, recombinant mouse KITL did not enhance the survival of follicles in neonatal mouse ovaries maintained in vitro, nor did KITL-neutralizing antibody promote follicular atresia.
|
Effects of Culture and Recombinant Mouse KITL Treatment on the Distribution of Follicles in Mouse Ovaries
The proportion of healthy follicles at each stage of development in preculture and cultured explants was quantified. Sections from preculture mouse ovaries contained 82.2% ± 3.3% primordial follicles, 17.0% ± 3.1% early primary follicles, 0.6% ± 0.2% primary follicles, and 0.3% ± 0.1% preantral follicles (Fig. 8). The culture process itself induced a considerable (P < 0.05) redistribution of follicles from the resting into the growing follicle pool. After 8 days of culture, the proportion of primordial follicles decreased to 33.0% ± 3.2%, and there was an increase in the number of early primary, primary, and preantral follicles to 59.7% ± 2.5%, 5.0% ± 0.9%, and 2.3% ± 0.5%, respectively (P < 0.05) (Fig. 8).
|
To determine whether the observed spontaneous activation was induced by endogenously produced KITL, the proportions of follicles in untreated and KITL-neutralizing antibody-treated cultured ovaries were compared. The addition of neutralizing antibody to the culture media partially inhibited the spontaneous activation of primordial follicles. This was manifested by a significant (P < 0.05) increase in the proportion of primordial follicles in neutralizing antibody-treated ovaries (47.2% ± 3.6%) compared with untreated ovaries (33.0% ± 3.2%) (Fig. 8). This increase in the proportion of primordial follicles was reflected by a decline in the proportions of early primary, primary, and preantral follicles in the antibody-treated group compared with the untreated culture control group, although the differences between these two treatment groups within individual follicle stages were not statistically significant.
To investigate the effect of exogenous KITL on follicle recruitment, ovaries were treated with 50 ng/ml or 150 ng/ml recombinant mouse KITL. Treatment of mouse ovaries with recombinant mouse KITL induced a dramatic decrease in the proportion of primordial follicles (Fig. 8). In untreated cultured ovaries, primordial follicles comprised 33.0% ± 3.2% of all follicles. In ovaries treated with 50 ng/ml recombinant mouse KITL only 20.8% ± 2.7% of follicles remained in the primordial stage of development, and this proportion decreased to 10.1% ± 1.5% in the presence of 150 ng/ml recombinant mouse KITL. The proportions of follicles in the early primary stage of development in untreated and recombinant mouse KITL-treated (50 ng/ml and 150 ng/ml) ovaries were not significantly different from each other. Despite a trend toward increased number, there was no significant difference in the proportions of primary and preantral follicles with 50 ng/ml recombinant mouse KITL. However, treatment with 150 ng/ml recombinant mouse KITL significantly (P < 0.05) increased the proportions of primary and preantral follicles (15.7% ± 1.8% and 8.6% ± 1.4%, respectively), compared with untreated cultured ovaries (5.0% ± 0.9% and 2.3% ± 0.5%, respectively) (Fig. 8). The addition of KITL-neutralizing antibody completely negated the follicle development induced by recombinant mouse KITL (Fig. 8). These data are consistent with KITL having an effect on both primordial follicle recruitment and the progression of follicles into later developmental stages in mice.
Effects of Culture and Recombinant Mouse KITL Treatment on Oocyte Growth in Mouse Ovaries
Mouse oocytes were measured and scored for developmental stage in order to determine the effects of culture and recombinant mouse KITL treatment on oocyte growth.
The culture process itself induced oocyte growth in mouse ovaries, but only at the primordial stage of development. In cultured untreated ovaries, oocytes from primordial follicles were considerably larger (P < 0.05) than those of primordial follicles in preculture tissues, whereas the mean diameters of oocytes from follicles at all other stages of development were unaffected by culture (Table 2). The spontaneous enlargement of oocytes from primordial follicles was not antagonized by KITL-neutralizing antibody (Table 2), suggesting that some unknown aspect of culture other than endogenous KITL induced the observed growth.
Supplementation of culture media with recombinant mouse KITL promoted a significant (P < 0.05) increase in the mean diameter of oocytes from mouse primordial and early primary follicles compared with those in untreated cultured ovaries (Table 2), but only with the higher dose. Furthermore, the growth-promoting effects of 150 ng/ml recombinant mouse KITL on primordial and early primary follicle oocytes were completely inhibited by the addition of KITL-neutralizing antibody (Table 2).
In contrast, recombinant mouse KITL, either alone or in combination with neutralizing antibody, had little effect on the oocytes of follicles at the later stages of development. In general, ovaries treated with 50 ng/ml or 150 ng/ml recombinant mouse KITL contained primary, growing primary, and preantral follicles with oocyte diameters similar to their counterparts in untreated cultured ovaries. The exception to this trend was growing primary follicles, for which the mean oocyte diameter was significantly (P < 0.05) larger in ovaries treated with 50 ng/ml recombinant mouse KITL compared with untreated ovaries.
These data support a role for KITL in promoting murine oocyte growth during the earliest stages of follicle development, whereas other growth stimuli may become more influential in primary, growing primary, and preantral follicles.
DISCUSSION
Primordial follicles are the stores of female gametes from which all mature eggs for ovulation and fertilization arise. The recruitment of primordial follicles into the growth phase is tightly controlled and likely involves the action of both stimulatory and inhibitory molecules. Although the exact mechanisms that regulate primordial follicle activation and the initiation of oocyte growth remain elusive, the study of rodent models has shown KITL and KIT to be involved in these processes. In order to determine whether KITL and KIT also are important during primordial follicle activation in another species, we first characterized the expressions of KITL and KIT during the initial wave of folliculogenesis in the rabbit. We then used an in vitro culture system for neonatal mouse ovaries and rabbit ovarian cortical explants to investigate the effect of recombinant mouse KITL on the regulation of follicle activation in these two animal species.
The temporal expressions of KITL and KIT during follicle assembly and folliculogenesis has been investigated in primates, including the human [40, 51–53], sheep [54, 55], and mouse [29–36]. However, this is the first report of their localization within the rabbit ovary. In the rabbit, Kitl mRNA and protein were detected in the squamous granulosa cells of some primordial follicles. The cuboidal granulosa cells surrounding growing follicles uniformly expressed high levels of KITL until the antral stage of development, when the presence of KITL became sporadic. Both mRNA and protein for KIT were detected within all oocytes, at every stage of development. In general, the amount of KIT protein localized to the oolemma increased with follicle growth and then decreased within the fully grown oocytes of antral follicles. The localization of both KITL and KIT protein within the oocytes of some primordial and growing follicles implies receptor activation and endocytosis, as well as receptor replacement [32, 56]. These findings suggest that the KITL/KIT signaling pathway is active from the earliest stages of follicle development. That only some primordial follicles expressed KITL, whereas KIT protein was expressed within all oocytes, is consistent with a role for KITL and KIT in primordial follicle activation and follicle growth in rabbits.
We, therefore, employed an in vitro approach to investigate the effect of KITL on follicle survival, oocyte growth, and follicle recruitment in rabbit ovarian cortical explants and neonatal mouse ovaries, and we found striking differences between the rabbit and mouse model systems regarding the effect of exogenous KITL supplementation on these parameters. In rabbit ovarian cortical explants treated with recombinant mouse KITL there was a significant and dose-dependent increase in oocyte diameter. The growth-promoting effects of KITL appeared to be particularly effective in follicles up to the preantral stage of development, at which point the influence of other factors may become more important. Alternatively, it is possible that the explant culture system employed did not fully support the development of later stage follicles, and, therefore, a role for KITL during later folliculogenesis cannot be ruled out. Since no increase in the activation of primordial follicles was observed in recombinant mouse KITL-treated explants, these data imply that another signal independent of KITL is required to coordinate primordial follicle activation in rabbits. Therefore, the primary role of KITL in this species may be to promote oocyte growth in follicles that have already received the signal(s) to activate.
Similar to rabbits, exogenous KITL stimulated the growth of oocytes in cultured neonatal mouse ovaries, but the growth-promoting effects were more stage restricted, with only primordial and early primary follicles affected. Unlike rabbit ovarian cortical explants, recombinant mouse KITL induced primordial follicle activation and supported the development of multilayered follicles in mouse ovaries. All of these effects were antagonized by KITL-neutralizing antibody. These findings are consistent with earlier reports for rats and mice [16–18] and support the hypothesis that KITL directly promotes primordial follicle activation and continued follicle development in rodents. The increased occurrence of secondary follicles in recombinant mouse KITL-treated ovaries also suggests that KITL may stimulate granulosa cell proliferation, as has been previously reported [20]. The KITL/KIT interaction may be of greater reproductive significance in rodents than in other species. Consistent with this view is the observation that a number of mutations at both the white spotting and steel loci have severe reproductive phenotypes in mice [4, 6, 57, 58], whereas even though a number of human diseases have been associated with defects in these genes [1], none have been described with reproductive phenotypes. Furthermore, Fortune et al. in a review quote data to indicate that KITL does not promote follicle activation in bovine ovarian cortical explants [59].
Primordial follicle activation is characterized by granulosa cell proliferation and morphogenesis, as well as oocyte growth. Given that granulosa cells do not express the KIT receptor, the signal to coordinate these activities in recombinant mouse KITL-stimulated mouse ovaries must therefore be transmitted via the oocyte, although the identity of such a signal is unknown. Studies describing the temporal expression patterns for KITL and KIT suggest that under normal in vivo conditions KITL availability may be a limiting factor in follicle activation. Only minimal expressions of Kitl mRNA and protein are detectable in murine primordial follicle granulosa cells, and increased expression is observed in follicles enclosed by cuboidal granulosa cells [29, 32]. In contrast, all mouse primordial follicles express KIT mRNA and protein [29, 30], indicating that every oocyte is potentially responsive to KITL-induced growth and activation. Therefore, an upstream signal may be required to induce the granulosa cells to generate threshold KITL levels before follicle activation can proceed. This signal may be a reduction in inhibitors, an increase in activators, or a combination of both.
KITL has been reported to be a survival factor for a number of different cell types, including hematopoietic stem cells [60], spermatogonia [61], and primordial germ cells [11]. However, recombinant mouse KITL did not enhance the survival of oocytes in cultured neonatal mouse ovaries or rabbit ovarian cortical explants. Similarly, the addition of KITL-neutralizing antibody did not increase the proportion of atretic follicles in these tissues. Definitive evidence in support of an antiapoptotic role for KITL during folliculogenesis is currently lacking, and the published data are contradictory. Yoshida et al. [17] reported that follicle survival was not affected in mice injected with the KIT-blocking antibody ACK2. In contrast, inhibition of the KITL/KIT interaction with ACK2 has been shown to decrease the survival of preantral mouse oocytes maintained in vitro [20]. Recently, KITL has been shown to promote primordial follicle survival in cultured mouse ovaries [22]. The results of the current study, however, were unable to support these latter findings.
The spontaneous activation of primordial follicles in cultured ovarian tissue has been observed for a number of species, including cattle [62], primates [63], and, to a much lesser extent, rats [16]. It is unknown why primordial follicles are triggered to spontaneously activate when maintained in vitro, although clues as to the mechanism are beginning to emerge. In vivo, primordial follicles exit the resting pool in a gradual manner. It has therefore been hypothesized that an inhibitor of activation that is absent in vitro regulates follicle recruitment. It is noteworthy that factors secreted by later-stage follicles, such as anti-Mullerian hormone and activin A, are inhibitory to primordial follicle activation [64, 65]. The suggestion that growing follicles negatively regulate the earliest stages of development also is corroborated by studies demonstrating that primordial follicle activation is accelerated during the initial waves of folliculogenesis, when growing follicles are yet to form in the ovary [66]. Therefore, a reduced number of later-stage growing follicles in rabbit ovarian cortical explants and in neonatal mouse ovaries may be responsible for at least some of the observed spontaneous activation in vitro. Alternatively, it has been proposed that the ovarian medullar may produce a factor that retards follicle activation [63].
Further insight into the phenomenon of spontaneous follicle activation has been generated from studies demonstrating that primordial follicles in bovine and baboon cortical explants grafted to chick embryonic membranes are not spontaneously recruited [67, 68]. These findings argue against the requirement for inhibitory factors originating from the ovarian stroma, or growing follicles, to prevent spontaneous activation. The grafted ovarian tissue was vascularized rapidly in ovo, and, consequently, the authors suggested that a factor of systemic origin may regulate global follicle recruitment. This study [67] also suggests that in vitro conditions intrinsic to the culture system itself may in some way promote activation. In this regard, it has been suggested that in vitro spontaneous follicle activation is a consequence of exposure to an environment that is richer than the normal in vivo situation [62, 63].
Parrott and Skinner [16] attributed some of the spontaneous activation of primordial follicles in in vitro-cultured whole rat ovaries to endogenous KITL. However, in contrast to the situation described for rats, the spontaneous activation of rabbit primordial follicles was not inhibited by KITL-neutralizing antibody, suggesting that endogenous KITL does not contribute to the spontaneous activation of primordial follicles in rabbit ovarian cortical explants in vitro. Caution must be applied in the interpretation of these results, as it is possible that this antibody does not inhibit KIT signal transduction in rabbits, as is suggested by our results showing increased rabbit oocyte growth in the presence of combined recombinant KITL protein and neutralizing KITL antibody. In parallel studies with mice, the spontaneous activation of primordial follicles was only partially inhibited by KITL-neutralizing antibody. Therefore, although the endogenous production of KITL may contribute to some spontaneous activation, in the mouse the majority of spontaneous activation appears to be related to some other aspect of the culture system. Possibilities include the provision of exogenous growth factors that follicles are not exposed to under normal physiologic conditions or the absence of an inhibitory substance that is normally present in vivo.
In summary, these data indicate that in the rabbit the primary role of KITL is to promote the growth of oocytes during the early stages of development. In contrast, in the mouse, KITL may have a direct role in both primordial follicle activation and promoting oocyte growth and continued follicle development. Only a small percentage of mouse primordial follicles were refractory to both the mediators of spontaneous activation and the recruitment promoting effects of KITL. Why all mouse follicles did not respond in the same way to KITL remains to be elucidated. KITL had no effect on follicle survival for either species. The findings reported here have important implications for the extrapolation of rodent culture systems for the in vitro development of human ovarian follicles.
ACKNOWLEDGMENTS
The authors wish to thank Dr. Jeff Woods and Christine Donnelly of the Australian National University Statistical Consulting Group for assistance in performing the statistical analyses described in this study. We thank Dr. Peter Kerr and Dr. Kate Loveland for reading the manuscript and giving constructive comments.
FOOTNOTES
1 Supported by the Pest Animal Control CRC and Post Graduate scholarships from the Australian National University to K.J.H. 
2 Correspondence. FAX: 61 2 4921 6308; Eileen.McLaughlin{at}newcastle.edu.au
3 Current address: Molecular and Integrative Physiology, University of Kansas Medical Center, 3901 Rainbow Boulevard, Kansas City, KS 66160.
Received: 6 February 2006.
First decision: 1 March 2006.
Accepted: 1 June 2006.
REFERENCES
