| HOME | HELP | FEEDBACK | SUBSCRIPTIONS | ARCHIVE | SEARCH | TABLE OF CONTENTS |
|
|
|
|||||||
|
|||||||||
| ||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||
research-article |
AgResearch, Wallaceville Animal Research Centre, Upper Hutt 6007, New Zealand
ABSTRACT
The aims of these studies were to determine the abilities of antisera against different regions of ovine bone morphogenetic protein 15 (BMP15) and growth differentiation factor 9 (GDF9) to inhibit ovarian follicular activity, estrus (mating), and ovulation in sheep. The 9–15-mer peptides were conjugated to keyhole limpet hemocyanin (KLH) and used to generate antibodies against the flexible N-terminal regions of the mature protein as well as against regions in which dimerization of the protein or interaction with a type 1 BMP or a type 2 TGFB or BMP receptor was predicted to occur. Ewes (n = 10 per treatment group) were vaccinated with KLH or the KLH-BMP15 (n = 9 different peptides) or KLH-GDF9 (n = 10) peptides in Freund adjuvant at five consecutive monthly intervals. Overall, antisera generated against peptides that corresponded to amino acid residues 1–15 of the N-terminus of the BMP15 or GDF9 mature protein or GDF9 amino acid residues 21–34 were the most potent at inhibiting ovulation following primary and single booster vaccination. Several other BMP15 (8/9) or GDF9 (6/10) treatment groups, but not KLH alone, also produced significant reductions in the numbers of animals that ovulated, although 2, 3 or 4 booster vaccinations were required. Anovulation was commonly associated with the inhibition of normal ovarian follicular development and anestrus. The in vitro neutralization studies with IgG from the BMP15 or GDF9 immunized ewes showed that the mean inhibition of BMP15 plus GDF9 stimulation of 3H-thymidine uptake by rat granulosa cells was approximately 70% for animals without corpora lutea (CL), whereas for animals with one to three CL or more than three CL, the inhibition was 24%–33% or 27%–42%, respectively. In summary, these data suggest that reagents that block the biological actions of BMP15 or GDF9 at their N-termini have potential as contraceptives or sterilizing agents.
follicle, follicular development, granulosa cells, growth factors, ovary
Repeated monthly immunizations of sheep with a 15-mer amino acid (aa) sequence from either ovine bone morphogenetic protein (BMP15) or growth differentiation factor 9 (GDF9) conjugated to keyhole limpet hemocyanin (KLH) and administered in Freund adjuvant inhibited both normal follicular development and ovulation in most animals [1]. Passive immunization of ewes with either BMP15 or GDF9 antisera shortly before prostaglandin F2
-induced luteolysis inhibited either ovulation or normal corpus luteum (CL) function [1]. The peptides used to generate the antibodies used in the present studies spanned the mature BMP15 and GDF9 protein regions from aa residues 7 to 21 and 10 to 24, respectively. In contrast, when the aforementioned BMP15 or GDF9 peptide-KLH conjugates in a water-based DEAE-dextran adjuvant were administered to ewes as a primary and single booster vaccination, significant increases in the ovulation rate were observed with no obvious detrimental effects on fertilization, fetal development or the ability of immunized ewes to carry a pregnancy to term [2]. In follow-up studies with larger populations of ewes, using the above 15-mer BMP15 peptide conjugated to bovine serum albumin, increases of 25% in both the ovulation rate and lambing rate were achieved using the primary and single booster water-based vaccination regimen [3].
The ovarian phenotypes arising from the long-term or short-term immunization regimen were similar to those reported for ewes that are homozygous or heterozygous for point mutations in either the BMP15 [4–6] or GDF9 genes [6]. For example, animals homozygous for a BMP point mutation with a stop codon in the pro- or mature region of the protein or a nonconservative aa substitution at residue numbers 31, 53 or 99 [3] are anovulatory due to the inhibition of normal follicular development, whereas animals heterozygous for these mutations have higher than normal ovulation rates. A similar finding has been reported for the point mutation in GDF9 with a nonconservative aa substitution at residue 77 in the mature protein [6]. It has been proposed that the aforementioned point mutations in BMP15 and GDF9 affect the levels of protein synthesis or secretion [7] or the ability of the growth factor to dimerize [4] or to interact with either a type 1 or 2 receptor [3].
The aims of the present study were to test the antigenic effects of the ovine BMP15 and GDF9 peptides in sheep against the flexible N-termini of the mature proteins, as well as the peptide regions in which dimerization of the protein or interaction with a type 1 BMP or a type 2 TGFB or BMP receptor is predicted to occur. Ewes were immunized with the selected peptides, which were conjugated to KLH and administered in Freund adjuvant, on a monthly basis for 5 months, and the antibody response, estrus activity, ovulation rate, and ovarian follicular activity were assessed. Moreover, antisera from some of the treated ewes were tested for their abilities to neutralize ovine BMP15 or GDF9 in a rat granulosa cell bioassay [8].
All experiments were performed with the approval of the Animal Ethics Committee of the Wallaceville Animal Research Centre in accordance with the 1999 Animal Welfare Act Regulations of New Zealand. All animals had access to pasture and water ad libitum. The animals (n = 200) were 5- to 6-yr-old parous Romney ewes and the studies commenced before the onset of the breeding season. Vasectomized Dorset rams (n = 4) with marking harnesses were present with the ewes throughout the study, to aid the detection of estrus activity. Under pasture conditions in New Zealand, Dorset rams are sexually active for most of the year, whereas Romney ewes are in anestrus from July to February. Except where indicated, laboratory chemicals were obtained from BDH Chemicals (Palmerston North, New Zealand) or Roche Diagnostics (Auckland, New Zealand).
Generation of Antigens for Immunization of Sheep
The ovine BMP15 and GDF9 peptide treatment group numbers, peptide sequences, and predicted properties of each of the peptides are shown in Tables 1 and 2. The peptides were synthesized and conjugated to KLH through either an N- or C-terminal cysteine residue by Macromolecular Resources (Colorado State University, Fort Collins, CO).
|
|
Ewes were injected i.m. with 0.4 mg KLH (control; n = 10), 0.4 mg KLH-BMP15 peptides 1–9 (n = 10 per peptide) or 0.4 mg KLH-GDF9 peptides 1–10 (n = 10 per peptide) in 1 ml of Freund complete adjuvant for the initial immunization, which was given around 5–6 wks before the anticipated onset of the breeding season. Thereafter, the ewes were immunized once every 28 days on four consecutive occasions with 0.2 mg KLH, KLH-BMP15 peptide or KLH-GDF9 peptide conjugate in 1 ml of saline mixed with 1.25 ml STM [1]. The estrus cycle for each animal was determined from successive changes in color of the crayon in the marking harnesses on the vasectomized rams. Blood samples were collected for antibody titers before the primary immunization and 14 days after the third booster immunization (i.e., 98 days after the start of the experiment). The sera from these blood samples were collected and stored at –20°C until assessed for antibody titers. The ovaries of all ewes were examined by laparoscopy, 16–17, 8–9, 2–3, and 22–23 days after the first, second, third and third boosters, respectively. At 14–24 days after the final booster, all of the ewes were killed using a captive bolt and blood samples from animals in some of the treatment groups was collected to assess the abilities of the sera to neutralize the actions of BMP5 or GDF9 in an in vitro bioassay. Both ovaries were recovered, the number of surface-visible CL was recorded, and one ovary from each ewe was fixed in Bouin reagent for subsequent morphological assessment.
The ovaries were cut in half before processing. Two representative 5–6-µm sections were prepared from each half of the ovarian tissues and stained with hematoxylin and eosin. In total, four sections from each animal were examined. Each section was examined for the presence of types 1, 1a, 2, 3, and 4 (i.e., preantral) as well as type 5 (i.e., antral) follicles using the criteria of Lundy et al [13]. The oocyte diameters of the types 1–4 follicles were measured in the sections as described elsewhere [1]. Other structures observed in previous immunization studies [1] or in homozygous BMP15 or GDF9 mutants [4, 6] were also recorded, namely, follicles with abnormally large oocytes, grossly misshapen oocytes, oocyte-free follicles, luteinized follicles, as well as normal and abnormal looking CL.
Determination of Antibody Titers
Sera collected before and 14 days after the third booster vaccination were assessed for antibody titers by ELISA, as described previously [1]. The wells of the microtiter plates were coated overnight at 4°C with either Escherichia coli-derived ovine BMP15 or ovine GDF9 mature protein as the antigen, to assess the anti-BMP15 and anti-GDF9 antibody titers respectively. Serum samples from ewes immunized with KLH were also reacted with wells that contained the GDF9 or BMP15 mature protein. The sera from the KLH-, BMP15- or GDF9-immunized ewes were diluted 1:20 000 in GAB (phosphate-buffered saline that contained 0.05% Tween and 0.10% gelatin) and 0.2 ml aliquots of these dilutions were incubated at 37°C for 2 h in the wells of the microtiter plates previously coated with 100 ng/well of BMP15 or 200 ng/well of GDF9 mature protein. Wells were washed three times in wash buffer and incubated with 0.1 ml of HRP-conjugated rabbit anti-sheep IgG (1:10 000 dilution in GAB; Scientific Supplies, Auckland, New Zealand) for 1 h at 37°C. The wells were then washed three times and incubated with 0.1 ml of 50 mM citrate buffer that contained 0.4 mg/ml o-phenylenediamine and 0.04% H2O2 for 30 min at 37°C in the dark. Reactions were stopped by the addition of 50 µl of 2.5 M H2SO4 and the absorbance at 490 nm was determined in a Bio-Tek EL 311 microplate autoreader (Bio-Tek Instruments Inc., Winooski, VT). To assess the specificity of the antisera raised against the BMP15 or GDF9 peptides, the sera were diluted 1:5000 in GAB, and 0.2 ml aliquots of these dilutions were then incubated as described above, except that the diluted GDF9 antisera were incubated with 100 ng/well BMP15, the BMP15 antisera with 200 ng/well GDF9, and the KLH antisera with either 100 ng/well of BMP15 or 200 ng/well GDF9.
Abilities of Antibodies to Neutralize the Biological Effects of BMP15 Plus GDF9 on Rat Granulosa Cells In Vitro
The thymidine uptake assay has been described previously [8]. Ovaries were collected from Sprague-Dawley rats (23–26 days old; University of Otago, Dunedin, New Zealand) approximately 46 h after i.p. administration of 20 IU eCG (Intervet, Auckland, New Zealand). Granulosa cells were collected by syringe aspiration of all surface-visible follicles and suspended in Leibovitz L-15 media (Invitrogen, Auckland, New Zealand) that contained 0.1% BSA, 100 IU/ml penicillin, and 100 µg/ml streptomycin. Oocyte-cumulus cell complexes, isolated oocytes, and follicular debris were visualized with the aid of a dissecting microscope and most, if not all, were removed using a glass pipette. The remaining cells were recovered following centrifugation at 300 x g for 5 min at room temperature after one wash in 5 ml of Leibovitz L-15 medium. After determination of cell viability using trypan blue exclusion, 20 000 viable granulosa cells were cultured with 25 ng/ml GDF9 and 1.3 ng/ml BMP15 with or without IgG-purified polyclonal antibodies to the KLH, BMP15 or GDF9 peptides in a total volume of 125 µl per well. The GDF9- and BMP15-enriched media were generated in-house using ovine GDF9- or BMP15-expressing 293H cells, as described previously [8]. Control conditioned media were generated using cultures of 293H (untransfected) cells. The concentrations of GDF9 and BMP15 were chosen as representing a combination dose that caused a stimulation of thymidine incorporation that was between 0.5 and 0.8 of the maximum effect; neither BMP15 nor GDF9 alone stimulated or inhibited thymidine incorporation by rat granulosa cells [8]. The neutralizing abilities of individual or pooled samples of IgG-purified polyclonal antibodies to BMP15 peptides 1, 2, 4, and 8 and GDF9 peptides 1, 2, 3, 6, 9, and 10 were tested in this assay. The control consisted of IgG purified from sheep immunized with KLH. The IgGs from the sera collected from the blood samples at slaughter were purified by precipitation with ammonium sulfate (45% v/v) and hydrophobic charge induction chromatography using a MEP HyperCel column (Pall Corp., East Hills, NY), according to the manufacturers instructions. Briefly, 60 mg of ammonium sulfate precipitate was applied to a column that contained 2 ml Mep HyperCel gel (equilibrated with 10 mM PBS) and allowed to recirculate through the column at a flow rate of 1 ml/min for 15 mins. The column was washed with 10 mM PBS and ovine IgG was eluted using 50 mM acetate buffer (pH 4.0). The IgG fractions were pooled, dialyzed against 10 mM PBS (pH 7.4), and stored at –20°C until required. For each treatment, we added 100 µg of IgG, which was composed of IgG purified from the animals immunized with the BMP15 or GDF9 peptide with the balance of the IgG, where appropriate, being derived from sheep immunized with KLH. Each assay was performed in quadruplicate and two to four replicate experiments were performed.
To assess whether significant antibody responses occurred, differences between the postimmunization and preimmunization optical density readings were determined by paired t-test. Differences between the KLH (control) and BMP15 or GDF9 treatment groups in the proportions of ewes ovulating at each observation were determined by Chi-squared analysis. The overall differences in ovulation rates for all the animals in each treatment group compared to the KLH immunized group were determined by the Mann-Whitney test. The overall differences between the ovulation rates of animals in each treatment group and those of the KLH-immunized animals were determined by ANOVA after square root transformation of the data. The dose-response effects of the anti-GDF9 or anti-BMP15 IgGs on the inhibition of GDF9 plus BMP15 stimulation of 3H-thymidine uptake by rat granulosa cells were determined by exponential regression analysis.
All of the BMP15 peptides, with the exception of peptide group 7 and the KLH (control) group, produced significant antibody responses to BMP15 (Table 3). All of the GDF9 peptides, but not the KLH (control) group, produced significant antibody responses to GDF9 (Table 4). When the BMP15 peptide antisera were tested for their abilities to interact with GDF9, a positive correlation (P < 0.05) was found for the percentage aa identity between BMP15 and GDF9 in the region of the peptide and the percentage of animals in the group with cross-reacting antisera (Table 1 and Fig. 1). None of the animals in the BMP15 peptide groups 1, 2, 3, and 6 had sera that showed evidence of binding to GDF9. When the GDF9 antisera were tested for their abilities to bind to BMP15, a similar positive correlation (P < 0.05) was found for the percentage aa identity between GDF9 and BMP15 in the region of the peptide and the percentage of animals in the group with cross-reacting antisera (Table 2 and Fig. 1), and none of the animals in GDF9 peptide groups 1, 2, 3, 5, and 7 showed any evidence of binding.
|
|
|
Effects of BMP15 or GDF9 Peptide Immunization on Ovulatory Activity Assessed as the Presence of Surface-Visible Corpora Lutea
The data with respect to each observation period are summarized in Table 5 (BMP15) and Table 6 (GDF9). All of the KLH-immunized controls were observed to have at least one CL at each observation. It was evident over the five observations for the BMP15 and GDF9 treatment groups that there were a number of animals with ovulation rates greater than 3. However, by the end of the treatment period, there was a significant reduction in the number of ewes that ovulated compared to the controls for all of the BMP15 peptide vaccination groups, with the exception of group 7. Immunizations with BMP15 peptide group 1 caused a significant reduction in the number of animals that had ovulated by the first observation (i.e., 16–17 days after the first booster vaccination) and thereafter, no animals were observed to have ovulated. Peptide 2 caused a significant reduction by the second observation (i.e., 8–9 days after the second booster), with most animals thereafter showing no CL. The BMP15 groups 3, 4, 5, 6, 8, and 9 were only observed to have significant effects after the third, fourth, and fifth observations.
|
|
Immunization with GDF9 peptide groups 1 and 3 significantly reduced the numbers of animals that had ovulated by the first observation (i.e., 16–17 days after the first booster vaccination) and thereafter, most of the animals were anovulatory (Table 6). Immunization with GDF9 peptide group 6 also significantly reduced the number of animals ovulating at each observation, although 3–6 out of 9 or 10 animals continued to show evidence of having ovulated at these time-points. In addition, immunization with GDF9 peptide groups 2, 4, and 5 produced significant reductions in the numbers of ovulating ewes.
The overall mean ± SEM ovulation rates for all animals in the BMP15 and GDF9 peptide-immunized groups are shown in Table 7. BMP15 groups 1 and 2 and GDF9 groups 1–3 all caused significant reductions in the mean ovulation rate, whereas BMP15 groups 7 and 9 and GDF9 groups 8, 9, and 10 all produced significant increases in the ovulation rate compared to the controls. The overall mean (and 95% confidence limits) ovulation rates for those animals that ovulated are shown in Table 8. For the BMP15 treatment groups, all except group 1 were significantly higher than the KLH-immunized controls. For GDF9, the overall mean ovulation rates were higher for all treatment groups, except for groups 2, 7, and 8.
|
|
In most (70–100%) of the KLH-immunized animals, normal preantral and antral follicles and CL were observed in the four representative sections that were examined (Figs. 2 and 3). No abnormal CL or luteinized unruptured follicles were noted. Moreover, none of the ovarian sections contained abnormally enlarged oocytes or abnormal-looking preantral or antral follicles.
|
|
For the BMP15 peptide and GDF9 peptide groups, most, if not all, of the four representative sections (i.e., 60–100%) were observed to have normal types 1, 1a, and 2 follicles for each animal. The BMP15 peptide groups 1 and 2 and GDF9 peptide groups 1, 2, and 3 displayed the highest frequencies of abnormalities, with between 8/9 and 10/10 animals lacking normal types 3 and 4 preantral follicles or antral follicles (Figs. 2 and 3). Moreover, 10/10 BMP15 peptide group 1, 10/10 GDF9 peptide groups 1, 2, and 3, and 7/10 BMP15 group 2 animals showed evidence of abnormal-looking preantral or antral follicles (Figs. 2 and 3). Moreover, in none of these peptide groups were normal CL present (Fig. 3). In all other BMP15 peptide groups (i.e., 3–9) normal preantral and antral follicles were present in more than 66.6% of the animals, although the proportions with normal-looking CL were generally low at 3/9, 1/9, 7/10, 3/10, 5/9, 0/10, and 3/10 for groups 3–9, respectively. All the BMP15 peptide groups contained some animals (1–4 per group) with either luteinized unruptured follicles or abnormal CL-like structures (Fig. 3). Similarly for GDF9 peptide groups 4–10, 50–100% of animals showed evidence of normal preantral and/or antral follicles, although the proportions with normal-looking CL were generally low at 3/10, 1/0, 2/10, 5/10, 3/10, 0/10, and 1/10 for groups 4–10, respectively. All the GDF9 peptide groups, with the exception of group 5, contained animals (2–9 per group) with either luteinized unruptured or abnormal CL-like structures (Fig. 3).
The mean oocyte diameters with respect to follicle types 1–4 in the BMP15 and GDF9 peptide treatment groups are shown in Tables 9 and 10, respectively. No data are reported for the type 5 follicles, as there were insufficient numbers across all the groups. For the BMP15 peptide treatment groups 1–3, the mean oocyte diameters were significantly larger than the KLH-treated controls at the types 1 and 1a stages of growth, with the larger diameter phenotype evident up to the type 3 follicular stage for group 2. Significant effects of treatment were also observed at the type 1a and 2 stages for BMP15 groups 6 and 8, while no effects were observed for groups 4, 5, 6 or 9. For GDF9 treatment groups 1, 2, 3, and 5, significant increases in mean oocyte diameter were observed for the types 1 and 2 follicular stages but not thereafter. For the other GDF9 peptide groups, no treatment effects on mean oocyte diameter were observed.
|
|
Abilities of the Anti-BMP15 and Anti-GDF9 Antisera to Neutralize oBMP15 Plus oGDF9 stimulation of 3H-Thymidine Incorporation by Rat Granulosa Cells In Vitro
Dose-response studies using sera from BMP15 peptide group 1 or 4 (1–100 µg IgG/ml) from animals with no CL indicated that the level of inhibition increased with increasing doses of IgG (Fig. 4). When individual or pooled sera from animals that were immunized with BMP15 peptides and that had 0–7 CL were examined using 100 µg IgG antisera, the mean ± SEM (range) percentages of inhibition of 3H-thymidine incorporation were 70 ± 8 (12–100) %, 24 ± 10 (13–88) %, and 27 ± 3 (18–33)% for 0 (n = 15), 1–3 (n = 7), and more than 3 (n = 4) CL, respectively.
|
The IgG from KLH-immunized animals produced a 6.5 ± 0.6 (0–10)% inhibition of thymidine uptake relative to the controls when no IgG was added. For most of the sera tested from the animals with no CL (13/15), the level of inhibition of thymidine uptake was 40–100% relative to the control; in two animals, the sera inhibited thymidine uptake by 12% and 25%.
Dose-response studies with sera from GDF9 peptide groups 3 and 10 (0.1–100 µg IgG/ml) from animals without normal CL showed that the level of inhibition increased with increasing dose of IgG (Fig. 4), whereas no dose-response was evident with serum from a GDF9-immunized animal with apparently normal ovulation (Fig. 4). The mean ± SEM (range) percentages inhibition of 3H-thymidine incorporation for the antisera from the animals with 0 (n = 8), 1–3 (n = 5), or more than 3 (n = 2) CL were 72 ± 8 (38–100) %, 33 ± 4 (20–42)%, and 42 (39–44) %, respectively.
With the exception of the BMP15 peptide 7 group, all the BMP15 and GDF9 peptides generated significant antibody responses, as assessed by ELISA using an E. coli-derived mature ovine BMP15 or GDF9 protein as the antigen. In the case of BMP15 peptide group 7, antibodies were detected when BMP15 peptide 7 was used as the antigen in the ELISA (data not shown), which suggests that the antigen-binding site in the E. coli BMP15 protein is masked from the antibodies. The antibodies generated against the N-terminal peptides of BMP15 or GDF9 (i.e., peptides 1–3 for both groups) did not lead to any GDF9 binding to the sera from the BMP15-immunized animals or vice versa. However, in several other treatment groups, there was evidence for animals having antisera with the capacity to bind both BMP15 and GDF9. The positive correlations between the percentages of identical aa residues of BMP15 and GDF9 in the selected regions of the peptides and the percentages of animals in the groups with cross-reacting antisera indicate that the cross-reactivity is due to homology between BMP15 and GDF9.
These studies provide clear evidence that BMP15 peptide 1 and GDF9 peptides 1 and 3 are highly effective at inhibiting ovulation following a primary and single booster vaccination. Thus, the peptides directed against aa residues 1–15 at the N-terminus of the BMP15 mature region or GDF9, as well as GDF9 aa residues 21–34 are the most potent at inhibiting ovulation in sheep. BMP15 and GDF9 peptides 2, which correspond to aa residues 5–19 and 10–24, respectively, were also effective, although 2 or 3 booster vaccinations were required to inhibit significantly ovulation.
In all of these treatment groups, most of the animals showed evidence of the presence of abnormal preantral (type 3 or 4 follicles) or antral follicles with phenotypes similar to those reported previously for infertile sheep homozygous for mutations in BMP15 or GDF9 [4, 6] or following long-term immunization with a BMP15 or GDF9 peptide identical to that used in the present study [1]. For the other BMP15 peptide treatment groups, all but peptide group 7 eventually produced a significant reduction in the number of ovulating ewes. However, the peptides that were designed to target the putative type 1, type 2 receptors or dimerization sites were less effective overall than those targeting the flexible N-terminal region. In some instances, e.g., with BMP15 peptide groups 7 and 9 and GDF9 peptide groups 8, 9, and 10, which targeted either a putative dimerization site (BMP15 peptide group 7) or a type 2 BMP receptor binding site [3], the ovulation rates were significantly higher than the controls over the treatment interval, and most of these animals showed evidence of the presence of normal preantral and antral follicles. However, the proportion of animals with normal-looking CL was lower (0–50%) than in the KLH-immunized group (
90%). Therefore, it seems unlikely that long-term immunization of ewes with peptides directed against the type 2 BMP receptor binding site or putative dimerization site would have beneficial effects on fertility, notwithstanding the apparent increase in ovulation rate.
One of the phenotypic differences observed previously following BMP15 or GDF9 peptide immunization [1] was premature enlargement of the oocyte relative to the number of layers of granulosa cells observed in the small preantral follicles. In the present study, a similar phenomenon was observed in some but not all treatment groups. In the present study, a significant increase in mean oocyte diameter relative to the control was observed at the type 1, type 1a, and type 2 follicular stages for the N-terminal peptides of the BMP15 and GDF9 treatment groups 1, 2, and 3. Two other BMP15 treatment groups (groups 6 and 8) were also found to have significantly larger mean oocyte diameters at the types 1a and 2 stages of follicular growth. Interestingly, all of the above groups had fewer ovulating animals at the end of the study than the other BMP15 treatments. For GDF9, only one other treatment group (group 5) contained oocytes at the types 1–2 stages of follicular growth with significantly larger mean diameters than the control. As with BMP15, this group, along with groups 1–3, had fewer ovulating animals at the end of the study compared to the other GDF9 treatments or the control. In the previous study [1], immunization with BMP15 or GDF9 peptide influenced the mean oocyte diameter of only the type 1a stage of growth, whereas in the present study, the mean oocyte diameters were larger, in some instances, at the type 1 stage. This may be an artifact of the classification system, in that follicles with abnormally large oocytes are characterized based on the presence of either flattened granulosa cells or a mixture of flattened and cuboidal cells, which is a phenotype that is not observed in control animals.
The in vitro antibody neutralization studies indicate that antisera against either BMP15 or GDF9 from animals without CL are capable of inhibiting directly the stimulatory effects of added BMP15 plus GDF9 on thymidine incorporation by rat granulosa cells. These findings suggest that the in vivo effects of peptide immunization on ovarian activity are due, at least in part, to inhibiting access of the ligand (i.e., BMP15 or GDF9) to its receptor on granulosa cells. While most sera from animals immunized with either BMP15 or GDF9 peptides that had no CL showed an appreciable level of inhibition (i.e., 38%) of thymidine incorporation, there were two animals with only modest levels of inhibition (12% and 25%). Animals with either a normal number (1–3) or high number (>3) of CL demonstrated levels of inhibition that were in most instances (15/18) greater than that observed for the KLH IgG control. However, overall, the level of inhibition of biological activity of BMP15 or GDF9 antisera from the immunized animals that had no CL was more than twice that observed for animals that had CL. When averaged, no obvious differences were observed in the degree of inhibition between the sera from animals with normal or high numbers of CL following immunization with BMP15 or GDF9 peptides. However, it was noted that the variation in the ability to neutralize the biological activities of BMP15 and GDF9 was greater in the group of animals that had a normal ovulation rate (i.e., 1–3 follicles) than those that had a high ovulation rate (i.e., >3 follicles). One explanation for this variation may be that the normal group is actually composed of two types of animals, with one group experiencing only minor affects of the immunization and with follicular development proceeding as normal and the other group having antisera that neutralize most, but not all, of the GDF9 or BMP15 available in vivo. The evidence gathered in the present study indicates that when normal amounts of BMP15 and GDF9 are present, ewes normally ovulate between 1–3 follicles, whereas with decreasing amounts of available BMP15 or GDF9, the ovulation rate increases until a threshold is reached, after which the limited availability of either of these growth factors negatively affects the ovulation rate. Thereafter, with little or no available BMP15 or GDF9, normal follicular development to ovulation does not occur. Moreover, further explanation as to why no differences were observed in the neutralizing abilities of the antibodies in animals with normal or high ovulation rates requires a more detailed examination of the antibody characteristics in the individual animals. However, it is reasonable to conclude that most animals with the anovulatory phenotype had circulating antibodies capable of inhibiting the paracrine effects of BMP15 and GDF9 in vivo. In addition, most of the animals immunized with the BMP15 or GDF9 peptides contained antibodies that were capable of inhibiting to some degree the endogenous BMP15 and GDF9.
The antibodies generated against peptides that target type 1 and 2 receptor binding or dimerization sites were generally less effective at altering ovarian function, at least after 1–3 booster vaccinations, as compared to those peptides that target the N-terminal regions. It is of interest to note that some of the naturally-occurring mutations in GDF9 or BMP15 are in the regions of the putative receptor binding or dimerization sites. Overall, the phenotypes that followed long-term repeated immunization with peptide groups 4–9 for BMP15 and 4–10 for GDF9 differed from those observed in ewes with the GDF9 mutation in a putative type II receptor binding region or the BMP15 mutation against a putative type I receptor binding or dimerization site [3]. This may relate to the characteristics of the antibodies generated in the experimental animals or to a reduction in the levels of GDF9 or BMP15 processed and secreted in the mutant sheep. The latter explanation is consistent with that proposed by Shimasaki and colleagues [7, 14].
In summary, these studies provide evidence that antibodies generated against the N-terminal region of BMP15 or GDF9 potently inhibit the paracrine actions of these oocyte-secreted factors in vivo and in vitro. Although further studies are needed, it appears that only one or two booster vaccinations with the N-terminal BMP15 or GDF9 peptide are needed to cause anovulation in sheep. This raises the possibility that the N-terminal BMP15 and GDF9 peptides alone or in combination may have roles as contraceptive and/or sterilizing reagents.
ACKNOWLEDGMENTS
We thank Doug Jensen, Daniel Olliver, and Kim Wearne for assistance with the animal work, Lee-Anne Still for preparation of histological materials, and Sue Swaney for secretarial assistance.
FOOTNOTES
1Supported by grants from the New Zealand Foundation for Research, Science and Technology and Ovita Ltd., Dunedin, New Zealand. 
Correspondence: 2Kenneth P. McNatty, School of Biological Sciences, Victoria University of Wellington, PO Box 600, Wellington 6140, New Zealand. FAX: 64 4 4635331; e-mail: Kenneth.mcnatty{at}vuw.ac.nz
Received: 1 June 2006.
First decision: 6 July 2006.
Accepted: 2 November 2006.
REFERENCES
This article has been cited by other articles:
|
|
 |
|
  J. L Juengel, N. L Hudson, M. Berg, K. Hamel, P. Smith, S. B Lawrence, L. Whiting, and K. P McNatty Effects of active immunization against growth differentiation factor 9 and/or bone morphogenetic protein 15 on ovarian function in cattle Reproduction, July 1, 2009; 138(1): 107 - 114. [Abstract] [Full Text] [PDF]
|
|
|
|
 |
|
 C. J. McIntosh, S. Lun, S. Lawrence, A. H. Western, K. P. McNatty, and J. L. Juengel The Proregion of Mouse BMP15 Regulates the Cooperative Interactions of BMP15 and GDF9 Biol Reprod, November 1, 2008; 79(5): 889 - 896. [Abstract] [Full Text] [PDF]
|
|
|
|
 |
|
 C. J. Lupton ASAS CENTENNIAL PAPER: Impacts of animal science research on United States sheep production and predictions for the future J Anim Sci, November 1, 2008; 86(11): 3252 - 3274. [Abstract] [Full Text] [PDF]
|
|
|
|
 |
|
 R. B. Gilchrist, M. Lane, and J. G. Thompson Oocyte-secreted factors: regulators of cumulus cell function and oocyte quality Hum. Reprod. Update, March 1, 2008; 14(2): 159 - 177. [Abstract] [Full Text] [PDF]
|
|
| ||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||
| HOME | HELP | FEEDBACK | SUBSCRIPTIONS | ARCHIVE | SEARCH | TABLE OF CONTENTS |
