| HOME | HELP | FEEDBACK | SUBSCRIPTIONS | ARCHIVE | SEARCH | TABLE OF CONTENTS |
|
|
|
|||||||
|
|||||||||
| ||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||
Ovary |
Reproductive Science Group, Discipline of Biological Sciences3
ARC Centre of Excellence in Biotechnology and Development,4 School of Environmental & Life Sciences, University of Newcastle, Callaghan, New South Wales 2308, Australia
 |
ABSTRACT |
|---|
 TOP ABSTRACT INTRODUCTIONMATERIALS AND METHODSRESULTSDISCUSSIONREFERENCES |
|---|
-mannose, and terminal sialic acid residues in a developmentally specific manner. Riboprobes for canine ZPA and ZPC genes were produced and used for in situ hybridization studies of mRNA expression in canine folliculogenesis. In addition, we isolated a partial cDNA transcript from total ovarian RNA for the canine ZPB gene having a high degree of sequence identity with the felid and porcine ZPB homologues. Subsequently, the ZPA gene transcripts were localized to the cytoplasm of oocytes in primordial, primary, and early secondary follicles. We then localized expression of ZPB and ZPC gene transcripts to the granulosa cells of growing follicles, but not in squamous granulosa cells of primordial follicles or oocytes. These observations indicate that in the juvenile canine ovary, the oocyte is responsible for synthesis of the ZPA protein and directing synthesis of the ZPB and ZPC proteins by the granulosa cells and that ZP gene transcription occurs in a sequential manner during folliculogenesis.
follicle, granulosa cells, oocyte development, ovary
|
INTRODUCTION |
|---|
|
TOPABSTRACTINTRODUCTIONMATERIALS AND METHODSRESULTSDISCUSSIONREFERENCES |
|---|
The ZP is a heavily glycosylated structure, and the oligosaccharide side chains of the constituent glycoproteins play a key role in defining the ability of the ZP to recognize and bind capacitated mammalian spermatozoa. In particular, the monosaccharide residues ß-N-acetyl glucosamine, mannose, fucose, galactose, and sialic acid have all been proposed to play a crucial role in the sperm-zona recognition event that initiates fertilization [5]. To determine the location and ontogeny of these carbohydrate residues during folliculogenesis, lectin expression patterns have been conducted for a diverse array of mammals, including rodents [6–9], domestic animals [6, 10, 11], as well as lagomorphs and various other species [6, 12–14], including the dog [6, 15, 16]. These studies have revealed a remarkable array of species-specific carbohydrate structures associated with the ZP and have variously suggested that the cellular origins of this structure lie within the oocyte or the granulosa cells. The use of additional techniques to address the cellular and temporal expression of ZP proteins during folliculogenesis, however, has generated conflicting results, with distinct differences being reported both between and within species [17–21]. Most of these studies have employed antibodies raised against homologous native recombinant proteins or heterologous whole-porcine zonae [21–23]. Issues of antigen retrieval, fixation procedures, and cross-reactivity with homologues raised against heterologous immunogens have all contributed to the debate [24, 25].
In the most well-characterized system, the mouse, the ZP proteins appear to be synthesized and secreted solely by the growing oocyte during folliculogenesis [26]. Conversely, other researchers have shown that in domestic animals, such as the bull and rabbit, the zona proteins are differentially expressed by both the oocyte and the granulosa cells at different stages of follicular development and that granulosa cells can even synthesize some ZP proteins in the absence of oocytes [19, 23, 27]. In situ hybridization studies of porcine and bovine ZPC transcript localization have suggested that ZPC mRNA is mainly located in the oocyte of primordial and primary follicles. However, in secondary follicles, both oocytes and granulosa cells contain the ZPC mRNA, whereas in tertiary and preovulatory follicles, the transcripts are primarily located in the cytoplasm of corona radiata cells [19, 23]. Similarly, expression of the ZPB (ZP1) transcript, the major constituent of the rabbit ZP, has also been localized to both oocytes and granulosa cells of primordial and developing follicles [27]. In contrast, studies in the domestic cat suggest that the ZP is produced exclusively by granulosa cells, not by oocytes, and that the synthesis of ZP takes place at every stage during follicular development [28].
In the dog, attempts to localize the expression of ZP proteins during folliculogenesis have failed to demonstrate conclusively whether any or all of the zona proteins are synthesized in the oocyte or the granulosa cells [21]. In the present study, we have established the existence of the hitherto unidentified canine ZPB gene and confirmed exclusive expression of the canine ZPB and ZPC genes within the granulosa cells of actively growing ovarian follicles, which is in keeping with the lectin-binding patterns recorded for this tissue. In contrast, ZPA gene expression in the canine ovary was confined exclusively to the oocyte in primordial, primary, and secondary follicles. Thus, notwithstanding the highly conserved nature of the glycoproteins that comprise the ZP, important species-specific differences exist in the cellular mechanisms that underpin the creation of this pivotal structure.
|
MATERIALS AND METHODS |
|---|
|
TOPABSTRACTINTRODUCTIONMATERIALS AND METHODSRESULTSDISCUSSIONREFERENCES |
|---|
Ovaries from prepubertal bitches (age, 12–24 wk) were obtained following routine desexing by local veterinarians following approval from the University of Newcastle (Australia) Animal Care and Ethics Committee. Ovaries were dissected free of surrounding connective tissue and either snap-frozen in liquid nitrogen or fixed in ice-cold, 4% paraformaldehyde before further analysis. All chemicals and reagents were obtained from Sigma (St. Louis, MO) or Research Organics (Cleveland, OH) unless otherwise stated.
Ovarian Histology and Lectin Histochemistry
Entire ovaries were fixed in 4% paraformaldehyde in PBS at 4°C overnight, dehydrated, and embedded in paraffin. Following dewaxing and rehydration, sections (thickness, 4 µm) were stained with hematoxylin and eosin, mounted, and examined using bright-field microscopy. Folliculogenesis was categorized according to previously described criteria [29, 30].
A panel of fluorescently labeled lectins was used to probe glycoprotein expression on oocytes and granulosa cells. Lectins were conjugated with fluorescein isothiocyanate (FITC) or tetramethylrhodamine isothiocyanate (TRITC) fluorochromes and included PSA, s-WGA, Con-A, GSL II, ECL and LEL (see Table 1 for definitions of lectins). The complete complement of lectins investigated and the oligosaccharide specificity associated with these molecules are detailed in Table 1. Ovarian sections (thickness, 4 µm) were dewaxed before incubation for 2 h at room temperature with each lectin (5–20 µg/ml) in a darkened, humidified chamber. After incubation, sections were gently washed three times with PBS, mounted with Mowiol (Sigma), and viewed using a Zeiss AxioPlan epifluorescent microscope (Carl Zeiss Pty, Sydney, NSW, Australia); photomicrographs were subsequently taken with a Kodak digital camera (Eastman Kodak, Rochester, NY).
|
RNA Extraction and Polymerase Chain Reaction
Total RNA was isolated from juvenile canine ovaries (age, 16 wk) using a TRIzol extraction protocol according to the manufacturer's instructions (Invitrogen, Carlsbad, CA). Oligonucleotide primers were designed, based on the published canine ZPA and ZPC sequences [2], to give product sizes in the region of 0.4–1.0 kilobase (Table 2). In addition, homology analysis of known ZPB sequences deposited in NCBI GenBank (bovine, accession no. AB042652; feline, accession no. U05777; and porcine, accession no. L11000) aided in the design of novel primers to amplify a portion of canine ZPB, an as-yet-uncharacterized transcript [2].
|
Reverse transcription (RT) was performed with 2 µg of total RNA, 500 ng of oligo-dT primer, 40 U of RNasin (Fermentas, Hanover, MD), 0.5 mM dNTPs, and 20 U of RevertAid M-MuLV-Reverse Transcriptase (Fermentas). Polymerase chain reactions (PCRs) contained cDNA equivalent to 100 ng of RNA, 200 nmol of each primer, 0.5 mM dNTPs, and 1 U of Taq polymerase (Fermentas). The RT reactions were tested with ß-actin primers (Table 2) before analysis of gene expression. The PCR fragments were cloned into pGEM-T Easy (Promega, Madison, WI) according to the manufacturer's instructions and sequenced by Newcastle DNA (Newcastle, NSW, Australia) to confirm identity.
In Situ Hybridization
Following isolation, ovaries were fixed as described above, and tissue sections (thickness, 5 µm) were processed for in situ hybridization. The pGEM-T Easy plasmids containing the canine ZPA, ZPB, and ZPC gene fragments generated above were linearized following incubation with either NcoI or NdeI restriction enzymes (for sense and antisense probes, respectively). Following phenol/chloroform extraction, the linearized plasmids were ethanol precipitated and resuspended in diethyl pyrocarbonate-treated H2O.
The digoxygenin (DIG) labeling of linearized probes was carried out using the Roche (Roche Molecular Biochemicals, Mannheim, Germany) DIG RNA Labeling Kit (SP6/T7) according to the manufacturer's instructions. In brief, the linearized plasmid templates were incubated with RNA polymerases (either SP6 or T7) in the presence of labeled NTPs to generate a labeled RNA probe for the in situ hybridization. Following labeling, dot blotting and comparison with serial dilutions of a control RNA sample were used to determine the concentration of labeled ZPA, ZPB, and ZPC probes that were present.
Following isolation, ovaries were fixed as described above, and tissue sections (thickness, 5 µm) were processed for in situ hybridization. The ovarian tissue sections were incubated twice for 5 min each in PBS, followed by a 15-min incubation in PBS/0.3% Triton-X and two 5-min incubations in PBS. Tissue was then incubated in either 5 µg/ml of proteinase K or PBS for 30 min at 37°C. Sections were covered with 4% paraformaldehyde and placed at 4°C for 5 min, washed twice for 5 min in 0.1 M triethanolamine buffer (pH 8)/0.25% acetic anhydride, and then rinsed in PBS.
The sections were next covered with prehybridization buffer (4x SSC [1x SSC: 0.15 M sodium chloride and 0.015 M sodium citrate] and 50% deionized formamide) and incubated at 37°C for 10 min. Probe (10 ng) in hybridization buffer (40% formamide, 10% dextran sulfate, 1x Denhardt solution, 4x SSC, 10 mM dithiothreitol, and 1 mg/ml of tRNA) was denatured at 80°C for 5 min and then placed on ice for 5 min.
Approximately 10–20 µl of either sense or antisense denatured probe in hybridization buffer were incubated with each ovarian tissue section; the sections were covered with Surfasil-coated coverslips (Pierce, Rockford, IL) and sealed with rubber cement. Following overnight incubation at the annealing temperature (range, 42–52°C), the rubber cement was removed and the slides washed twice for 15 min each with 2x SSC/50% formamide, each slide in a separate wash, at 40–45°C. The slides were then washed separately, twice for 15 min in 1x SSC at 23°C.
Probe detection was accomplished using an anti-DIG, alkaline phosphatase (AP)-conjugated antibody and the HNPP/Fast Red TR Fluorescent Detection set (Roche). Briefly, sections were blocked (100 mM Tris-HCl [pH 7.5], 150 mM NaCl, and 0.5% BSA) for 1 h at 4°C. Anti-DIG-AP conjugate was diluted 1:500 in blocking buffer on ice, 50 µl applied to each section and covered with a coverslip, and the sections incubated for 1 h at 37°C in a humid chamber. The slides were then washed three times for 10 min each time in Wash Buffer (100 mM Tris-HCl, 150 mM NaCl, and 0.05% Tween 20) at room temperature and then twice for 10 min each in Detection Buffer (100 mM Tris-HCl, 100 mM NaCl, and 10 mM MgCl2 [pH 8.0]) at room temperature. The slides were washed in distilled H2O and mounted with Mowiol, and images were collected with a LSM510 laser-scanning microscope (Carl Zeiss Pty) at excitation wavelengths of 488/543 nm and emission spectra of greater than 560 nm.
|
RESULTS |
|---|
|
TOPABSTRACTINTRODUCTIONMATERIALS AND METHODSRESULTSDISCUSSIONREFERENCES |
|---|
Examination of ovarian sections from 12- to 16-wk-old animals revealed the presence of all stages of follicular development, including nests of quiescent primordial follicles (Fig. 1a); activated primary, early secondary, and preantral follicles (Fig. 1, b–d); as well as early and late antral follicles containing oocytes with a distinct germinal vesicle (Fig. 1e). This is consistent with the observations of other researchers [29–31], who concluded that the initiation of folliculogenesis in the dog takes place 2–12 wk postnatally, similar to that observed in the rabbit [32].
|
In addition, several follicles contained multiple oocytes (Fig. 1f), a phenomenon that has previously been reported as a common occurrence in juvenile dogs. Surprisingly, in these polyovular follicles, oocytes can be at varying stages of development or even contain viable and atretic oocytes simultaneously [30]. No corpora lutea were noted, suggesting that ovulation had not yet occurred in these prepubertal animals [33]. The ZP was readily identifiable in developing secondary follicles (Fig. 1, c–e), though less obvious in the newly activated primary follicle (Fig. 1b) [33] and absent in the primordial follicle (Fig. 1a) [34].
Lectin Histochemistry
Several specific lectin-binding patterns were observed in developing canine follicles, and these are summarized in Table 1. In particular, six lectins (PSA, s-WGA, ECL, GSL-II, LEL, and STL) consistently labeled the ZP and adjacent granulosa cells of the developing follicle (Fig. 2, a–d and f). The PSA, which is specific for -mannose, weakly labeled the granulosa cells of primary follicles and strongly labeled the surrounding granulosa cells and ZP of secondary follicles (Fig. 2a). The s-WGA, which is specific for N-acetyl glucosamine, strongly labeled ZP, adjacent granulosa cells, and oolemma in secondary follicles (Fig. 2b). The ECL, which binds D-galactose and N-acetyl glucosamine, gave strong labeling of adjacent granulosa cells and ZP in secondary follicles only (Fig. 2c). Furthermore, GSL-II and LEL, which also bind N-acetyl glucosamine (Fig. 2, d and f), gave strong labeling of ZP and surrounding granulosa cells within secondary follicles and weak labeling in primary follicles.
|
In contrast, the lectins WGA, DSL, GSL-I, and Con-A (Fig. 2e) labeled granulosa cells and/or the oolemmal membrane or oocyte cytoplasm but not the ZP in either primary or secondary follicles (Table 1). Finally, the lectins SJA, UEA-I, PVL, and PVE did not bind to canine ovarian follicles (Table 1).
In addition, dual-staining studies with fluorescein-conjugated Con-A, which binds to granulosa cells only, and with rhodamine-conjugated s-WGA (Fig. 2, g and h) suggested that ZP glycoproteins might originate in adjacent granulosa cells. To confirm the origin and temporal expression of members of the canine ovarian ZP family, identification and localization of the mRNA transcripts for all three ZP gene families were then undertaken.
Identification and Characterization of ZP Gene Family Members in Juvenile Canine Ovary
Using RT-PCR, amplicons were obtained for canine ZPA and ZPC of approximately 305 and 650 base pairs (bp), respectively (Fig. 3, a and b), which are consistent with predicted sizes based on the published canine ZPA and ZPC nucleotide sequences. Following cloning into the vector pGEMT-Easy and subsequent sequencing of the inserts, these PCR products were confirmed as canine ZPA and ZPC by BLASTN analysis and comparison with the publicly available data (GenBank accession no. U05779 and U05780, respectively) [2, 4]. Similarly, RT-PCR with oligonucleotide primers based on the known domestic animal ZPB sequences (feline, porcine, and bovine) also produced an amplicon approximately 750 bp in length (Fig. 3b). This PCR product was cloned into pGEMT-Easy, and following sequencing, the predicted cDNA sequence (Fig. 4) was aligned using BLASTN and found to exhibit a high degree of identity (86%) with the published feline ZPB mRNA sequence (Fig. 4) [2, 4]. This partial cDNA sequence has been submitted to GenBank (accession no. AY573930).
|
|
As a consequence of these PCR analyses, it was concluded that all three ZP gene families were expressed in the juvenile (16-wk-old) canine ovary, including an as-yet-unrecorded homologue of ZPB [35]. Riboprobes produced from these RT-PCR products were then used to localize ZP transcript expression in canine ovarian follicles.
In Situ Localization of ZPA, ZPB, and ZPC mRNA Expression in the Juvenile Canine Ovary
Figure 5 presents images of ZP gene mRNA expression patterns during folliculogenesis in the canine ovary. The ZPA gene transcripts were localized to the cytoplasm of oocytes in primordial, primary, and early secondary follicles (Fig. 5a). However, ZPA gene expression was absent in later developmental stages, particularly in antral follicles (Fig. 5b). Specificity of the riboprobes for ZPA mRNA was confirmed by the complete absence of labeling with the ZPA sense control (Fig. 5c), whereas no signal was detected with either probe in the granulosa cell population.
|
No ZPB mRNA expression was identifiable in either the squamous granulosa cells or the oocyte of quiescent primordial follicles (Fig. 5d). In contrast, ZPB mRNA expression was easily identifiable in the cuboidal granulosa cells of activated primary (Fig. 5e) and developing secondary follicles (Fig. 5, f–h). At no stage of follicular development was it possible to identify ZPB gene transcripts in the oocyte. Labeling was absent with the ZPB mRNA sense control (Fig. 5i).
A similar, but not identical, pattern of gene expression emerged following probing of ovarian tissue for ZPC mRNA. Again, no expression of the ZPC gene was identifiable in quiescent primordial follicles (Fig. 5j). Interestingly, there appeared to be heterogeneous expression of ZPC mRNA expression in subpopulations of granulosa cells of activated primary and early secondary follicles (Fig. 5k) and consistent, widespread expression in late secondary follicles. As expected, no staining was observed with the ZPC mRNA sense control (Fig. 5l).
|
DISCUSSION |
|---|
|
TOPABSTRACTINTRODUCTIONMATERIALS AND METHODSRESULTSDISCUSSIONREFERENCES |
|---|
Evidence from the mouse suggests that GluNAc residues found on the ZP glycoproteins are recognized by a sperm surface ß-1,4-galactosyltransferase [37–39]. The combination of O-glycans expressed by the ZP appears to confer some taxon specificity to the sperm-zona interaction, both because "humanized ZPC" transgenic mice display glycosylation patterns typical of the mouse ZP [5] and because their ZP bind exclusively to murine and not human spermatozoa [3, 37, 39]. Previous studies have used a variety of lectins to characterize the glycoside residue expression patterns in the canine ovary [6, 15, 16]. Here, we confirm that the canine ZP expresses ß-gal(1,4)glcNac and ß-gal(1,3)galNac and terminal sialic acid residues and that the presence and intensity of staining depend on the developmental stage of folliculogenesis [15, 16], suggesting that ZP proteins are synthesized and expressed in a sequential manner. In contrast to the findings of Parillo et al. [15], we were not always able to identify costaining of the ooplasm or oolemma with lectins specific for GlcNAc (compare Fig. 2, g vs. h), suggesting an extraoocytic origin for some of these glycoproteins. In addition, close examination of fluorescently labeled glycoproteins localized these residues to the extracellular spaces between granulosa cells (e.g., Fig. 2, a and f) rather than the oolemma (Fig. 2b). These intriguing observations indicated that in the dog, ZP proteins might originate in the granulosa cells immediately surrounding the oocyte as well as in the oocyte.
To identify the site of ZP gene mRNA synthesis, we undertook production of suitable riboprobes for in situ hybridization localization experiments. The cDNA sequences of the canine ZPA and ZPC genes were previously available through public databases and readily amplified from total ovarian RNA isolated from prepubertal animals (Fig. 3). The existence of the canine ZPB gene has long been postulated [2] following Northern blot analysis of canine ovarian RNA from a 6-mo-old bitch using a heterologous porcine ZP3 (ZPB) probe [2]. However, Harris et al. [2] were unable to clone the ZPB gene from a canine cDNA library prepared from a younger (age, 16 wk) animal, and they proposed that the canine ZPB gene may, in fact, be expressed later than the ZPA and ZPC genes. Using oligonucleotide primers designed from conserved regions of the ZPB gene in domestic species, we were able to amplify a partial cDNA sequence from 16-wk-old canine RNA and, following alignment with known ZPB sequences, to confirm the existence of the canine ZPB gene with a high degree of homology to other mammalian homologues [2, 4]. These results indicated that all three ZP gene families appeared to be coexpressed in the prepubertal canine ovary.
Probing of canine ovarian material with specific riboprobes for all three ZP genes indicated that the expression pattern for each gene is quite distinct. In line with the results of studies in the mouse [3, 25], the ZPA gene is expressed exclusively in the oocyte, with peak expression being observed in early developing primary and in secondary follicles. In contrast, a similar, but not identical, expression pattern was observed for the ZPB and ZPC genes, with both mRNAs being localized exclusively in the granulosa cells of activated follicles and not in the quiescent or growing primary oocyte (Fig. 5). At no developmental stage in folliculogenesis were we able to identify any ZPB or ZPC transcripts in primordial follicles or in the oocytes. These findings are consistent with the observations of two research groups who hypothesized that canine ZP proteins may be synthesized in and secreted from the follicular granulosa cells immediately surrounding the oocytes [21, 36], and they are similar to findings in the domestic cat indicating that granulosa cells are the sole source of ZP proteins in this species [28].
In conclusion, synthesis of the canine ZP involves the integrated action of both the oocyte and the granulosa cells, with the latter contributing substantially to the overall creation of this pivotal structure. This is different from the ZP synthesis observed in the mouse, in which the oocyte is exclusively responsible for elaborating the ZP. The production of the substantial canine ZP represents a considerable burden to the oocyte and could represent a serious challenge to the protein synthetic machinery of this cell if not for the participation of the granulosa cells. It is even possible that the priority given by granulosa cells to the production of ZP glycoproteins over follicular fluid secretion explains why antrum formation is such a late event in canine ovarian folliculogenesis. Therefore, we postulate that the oocyte within the activated canine primary follicle, in a manner similar to the bovine, lagomorph, and felid, directs the synthesis of the ZP by granulosa cells, possibly through secretion of signaling molecules such as GDF-9, BMP-15, or activin, which are members of the transforming growth factor ß superfamily known to act in the coordination of folliculogenesis.
|
ACKNOWLEDGMENTS |
|---|
|
FOOTNOTES |
|---|
2 Correspondence. FAX: 61 2 4921 6923; eileen.mclaughlin{at}newcastle.edu.au
Received: 19 February 2004.
First decision: 8 March 2004.
Accepted: 15 April 2004.
|
REFERENCES |
|---|
|
TOPABSTRACTINTRODUCTIONMATERIALS AND METHODSRESULTSDISCUSSIONREFERENCES |
|---|
| ||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||
| HOME | HELP | FEEDBACK | SUBSCRIPTIONS | ARCHIVE | SEARCH | TABLE OF CONTENTS |
