| HOME | HELP | FEEDBACK | SUBSCRIPTIONS | ARCHIVE | SEARCH | TABLE OF CONTENTS |
|
|
|
|||||||
|
|||||||||
| ||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||
research-article |
Departments of Animal Sciences3 and Obstetrics and Gynecology,4 The Ohio State University, Columbus, Ohio 43210
ABSTRACT
The regulation of early follicular growth and development involves a complex interaction of autocrine, paracrine, and endocrine signals. The ability of these factors to regulate follicle growth may depend in part on the extent of vascular delivery to and perfusion of the ovary. Vascular endothelial growth factor A (VEGFA) is a major regulator of vascular physiology in the ovary. VEGFA is produced in numerous ovarian compartments and likely plays a role in the regulation of all phases of follicular growth, from preantral through preovulatory. The aim of the present study was to further evaluate the role of VEGF in early follicle growth by neutralization of endogenous VEGF or VEGF receptors. Adult mice were injected systemically and prepubertal mice were injected directly under the ovarian bursa with antibodies designed to neutralize VEGF or block interaction with its receptors in the ovary. Both systemic and intrabursal injections of VEGF antibody significantly reduced the number of primordial follicles within 1–3 days after administration without affecting primary or secondary follicle numbers. Primordial follicle numbers were not different from control levels by 30 days after VEGFA antibody administration. Administration of antibodies to the kinase domain receptor (KDR), but not the FMS-like tyrosine receptor (FLT1), for VEGF also resulted in a significant decrease in primordial follicles. These data suggest that VEGF plays a vital role in the maintenance and growth of the primordial follicle pool.
follicle, follicular development, ovary
The formation, survival, and growth of the preantral follicle pool in the mammalian ovary is a dynamic process regulated by a variety of autocrine, paracrine, and endocrine signals [1]. The complex interaction of these factors may orchestrate the movement of primordial follicles into the growing pool or destine them for atresia. Central to the function of endocrine and paracrine signals is their ability to reach their target cell via vascular delivery and/or diffusion.
One of the most important local regulators of ovarian vascular physiology is vascular endothelial growth factor A (VEGFA) [2–4]. Along with stimulating neovascularization, VEGFA also regulates vascular permeability. VEGFA production and increased blood vessel extension are associated with follicle activation [5]. VEGFA is produced by the thecal and/or granulosa cells in the ovary [6–10], and granulosa cells secrete VEGFA in response to stimulation by gonadotropins [6, 11–13]. After ovulation, luteal cells continue expression of VEGFA [14, 15]. Neutralization of VEGFA activity with antibodies [16] or a soluble form of the VEGF receptor [17] disrupts follicular development and corpus luteum function [18–20].
In addition to its role in the later stages of follicle growth, VEGFA also may play an important role in the activation and development of preantral follicles. Although immunohistochemical studies generally confirm that follicular VEGFA expression increases as follicles mature [21–23], several reports reveal expression of VEGF in preantral follicular compartments. VEGFA protein has been identified in the oocytes of human primordial [24, 25] and human and rat primary [25, 26] follicles. Kezele et al. [27] identified VEGFA as one of the important genes upregulated during primordial follicle development in the mouse. We have shown previously that VEGFA administration increases the number of primary and small secondary follicles in the rodent ovary [28]. Administration of VEGFA directly to the ovary results in an increase in preantral follicle numbers in a dose- and time-dependent manner. In addition, estrogen upregulates VEGFA expression in the rodent ovary similarly to its effects on early follicle growth [28]. These data suggest a potential role for VEGFA in primordial follicle activation. As such, we hypothesized that neutralization of endogenous VEGFA might alter initial follicle growth and perhaps disrupt maintenance of the primordial follicle pool.
Antigen selection and generation was based on previous work in our laboratory with antibodies against HER2 [29]. In this approach, an antigenic B-cell epitope of VEGFA (residues 127–144, common to VEGF A–G) was selected using Peptide Companion software (CSPS Pharmaceuticals Inc., San Diego, CA). This peptide sequence is located in the carboxy terminal domain, which is important for the mitogenic actions of VEGFA [30]. The B-cell epitope was colinearly synthesized with a promiscuous TH epitope (measles virus fusion protein [MVF] 288–302), using a four-residue amino acid linker, GPSL. All peptides were synthesized by solid-phase peptide synthesis and were purified by semipreparative reversed-phase high-performance liquid chromatography as described previously [31]. The identity of the peptide MVF-VEGFA was confirmed by matrix-assisted laser desorption/ionization time of flight mass spectroscopy. A schematic representation of the peptide construct is provided in Figure 1. Throughout this manuscript, we have used VEGFA to describe that specific isoform and elsewhere used VEGF when describing our data, since our antibody will recognize various forms of VEGF (A, B, C, D, etc.).
|
Adult Female New Zealand White rabbits (Harlan, Indianapolis, IN) received an initial injection of 1 mg/ml MVF-VEGFA, which was dissolved in H2O with 100 µg nor-MDP (N-acetylglucosamine-3 yl-acetyl-L-alanyl-D-isoglutamine). Montanide ISA-720 was used to emulsify the peptide solution. Booster injections (0.5 mg/ml) were given twice, 3 wk apart. Blood samples were collected prior to the first injection and once a week thereafter. VEGF antibodies were purified from heat-inactivated serum using a protein A column (Pierce, Rockford, IL). Antibody titers were determined using ELSA as previously described [29]. Since MVF by itself is not immunogenic [29], we chose to use preimmune serum from the same rabbits used for VEGF antibody generation as controls for our experiments.
All studies were approved by the Institutional Laboratory Animal Care and Use Committee at The Ohio State University and were in accordance with the National Institutes of Health guide for the care and use of laboratory animals.
Twenty-four adult (6–8 wk old) female C57Bl/6 mice (Harlan) were injected i.p. with 55 µg VEGF antibody in 100 µl PBS. Control mice received a corresponding amount of preimmune serum IgG. Injections were given every 3 days for 2 wk. Mice were killed at various time points, and both ovaries were removed and fixed for tissue processing. Vaginal smears were obtained via vaginal lavage every day for 30 days and were evaluated for estrous cycle status [32].
For some experiments, VEGF antibody was administered under the bursa of the ovary as previously described [28]. Prepubertal female C57Bl/6 mice were anesthetized, and each ovary was injected under the bursa with 1 µl VEGF antibody or a corresponding amount of preimmune rabbit IgG. Mice were killed at various time points (8 h, 24 h, 48 h, 72 h, 30 days), and both ovaries were removed and prepared for histological analysis. We used adult mice for the systemic (longer-term) study to avoid the potential confounding effects of VEGF neutralization on pubertal development. Prepubertal mice were used for the intrabursal experiments to provide a more homogeneous ovarian physiology and histological architecture.
Tissues were processed as previously described [28]. Briefly, ovaries were removed from fixative, dehydrated, and embedded in Paraplast (Fisher Scientific, Houston, TX) before being sectioned (7-µm thickness) and stained with Lillies allochrome. To avoid counting a follicle more than once, follicles were counted on every fifth section throughout the ovary (approximately 20–30 sections per ovary were counted), and only follicles containing an oocyte were counted. Primordial follicles were described as those having a small oocyte with a single layer of squamous granulosa cells. Primary follicles had an intact enlarged oocyte with a visible nucleus and a single layer of cuboidal granulosa cells. Secondary follicles had two or more layers of cuboidal granulosa cells. Secondary follicles were further classified as small if they contained fewer than four layers of granulosa cells and large if they contained four or more layers of granulosa cells.
Western blots characterizing VEGF antibodies were performed as previously described [28]. Briefly, 10 ng recombinant human (hu) VEGFA, 50ng of recombinant mouse (mou) VEGFA (R&D Systems, Minneapolis, MN), and 5 µg pancreatic tumor lysate were electrophoresed on a 12% polyacrylamide gel and transferred to nitrocellulose. The pancreatic tumor was obtained from the Rip1-Tag2 transgenic mouse, which spontaneously develops pancreatic tumors expressing VEGFA [33]. Nitrocellulose membranes were probed with commercially available rabbit polyclonal VEGFA antibody (A-20; Santa Cruz Biotechnologies), VEGF peptide antibody, or purified IgG from preimmune serum. Primary antibodies were diluted 1:500 and detected with goat anti-rabbit IgG horseradish peroxidase-labeled secondary antibody (Santa Cruz Biotechnologies).
Results are depicted as the mean ± SEM. Potential differences in estrous cycle lengths and follicle numbers were analyzed by analysis of variance followed by a least significant difference test or by a paired Student t-test when appropriate. A P value of less than 0.05 was considered to be significant for all analyses
Western blotting experiments confirmed that our VEGF peptide antibodies recognized both mouse and human recombinant VEGFA proteins (Fig. 2). It is unclear why the recombinant human VEGFA protein migrates slightly faster in our system than recombinant mouse VEGFA. In addition, these antibodies recognized a predominant 23-kDa band (presumably VEGFA165) in pancreatic tumor lysate, along with a slightly higher molecular weight protein (~32 kDa) in this biological sample. These data are similar to the results obtained with a commercially available antibody specific for VEGFA. Immunoblotting with purified IgG from preimmune serum revealed no specific binding to pure VEGFA proteins or tumor lysate.
|
Figure 3 reveals the effects of VEGF neutralization following i.p. antibody administration on preantral follicles in adult mice. Within 3 days after a single VEGF peptide antibody injection, the number of primordial follicles had decreased by more than 50%. Additional VEGF peptide antibody injections maintained but did not increase the destruction of primordial follicles seen after the first injection. In contrast, VEGF antibodies had no effect on primary or secondary follicle numbers throughout the study. Six months after the initial injection (and 5.5 mo after the last injection) primordial follicle numbers were not different from those observed in control mice. There were no significant changes in primordial follicles within either the control or VEGF antibody treatment groups in this experiment.
|
Systemic VEGF neutralization also disrupted estrous cyclicity in these mice. Within 3 days following the initial antibody injection, mice began to display persistent vaginal cornification; the interval between estrous vaginal smears was significantly reduced from 5.9 ± 0.2 days before treatment to 1.7 ± 0.2 days after VEGF antibody injection (P < 0.05, n = 3 per group).
It is possible that systemic administration of VEGF antibody could have effects outside the ovary that could have an impact on primordial follicle growth and survival. Therefore, we investigated whether direct ovarian administration of VEGF peptide antibodies would affect primordial follicle numbers. VEGF peptide antibody was injected under the bursa of one ovary, and a corresponding amount of preimmune IgG was injected under the bursa of the contralateral ovary. Thus, the effects of neutralization of ovarian VEGF could be determined with each animal serving as its own control. Direct ovarian administration of VEGF peptide antibody resulted in a dose-dependent decrease in the number of primordial follicle numbers compared with control (preimmune IgG) ovaries, with a maximal effect at 5–25 µg VEGF antibody (P < 0.05, n = 3 to 4 per group; Fig. 4).
|
We compared our VEGF peptide antibodies against a commercially available VEGF antibody preparation (AF-493-NA) from R&D Systems (Fig. 5). Both antibody preparations significantly depleted primordial follicles in the mouse ovary (P < 0.05, n = 4 to 6 per group). Similarly to the results obtained with systemic antibody administration, intrabursal administration of VEGF antibodies had no effect on primary or secondary follicle numbers (data not shown).
|
We also investigated the time course of primordial follicle destruction by VEGF antibody treatment (Fig. 6). Intrabursal administration of VEGF peptide antibody resulted in a relatively rapid destruction of primordial follicles within 24 h after injection. Follicle depletion was still evident at 3 days; however, primordial follicle numbers were not different from control by 30 days after injection.
|
Since our antibody preparation binds VEGF directly, it is not useful in determining which VEGF receptor (KDR or FLT1) might be important for primordial follicle growth and survival. Therefore, we performed an additional experiment examining the effects of direct ovarian administration of commercially available antibodies that bind KDR (VEGFR2) or FLT1 (VEGFR1; R&D Systems). Figure 7 reveals that treatment with antibodies that bind KDR but not FLT1 resulted in a statistically significant depletion of primordial follicles similar to that observed with anti-VEGF antibodies
|
The data presented here provide the first evidence that endogenous VEGF is important for primordial follicle survival in the rodent ovary. Administration of antibodies directed against an antigenic peptide region of the VEGF molecule resulted in a rapid and profound loss of primordial follicles in the ovary. Similarly, antibodies that specifically block the KDR receptor also resulted in primordial follicle depletion. Direct ovarian administration of antibodies also depleted primordial follicles, indicating that neutralization of ovarian VEGF, as opposed to peripheral neutralization, is key to primordial follicle destruction.
The factors responsible for the formation, growth, and survival of the primordial follicle pool are largely unknown. Although primordial follicle activation can occur in the absence of LH, FSH, or estrogen, it is clear that under normal physiological conditions primordial follicle growth occurs amid a rich milieu of endocrine, paracrine, and autocrine factors [34, 35]. It is unknown if or how these factors might interact to influence the primordial follicle population. We and others have hypothesized that that the availability of an adequate vascular supply to provide endocrine and paracrine signals may play a key role in the regulation of early follicle growth [1, 36]. The data presented here on the importance of endogenous VEGF for primordial follicle survival are consistent with this hypothesis.
Although the majority of research on the ovarian VEGF system has focused on the latter (antral) stages of follicle growth and corpus luteum function, previous data from our laboratory have suggested an important role for VEGFA in early follicle growth as well [28]. Administration of VEGFA directly to the rat ovary resulted in an increase in preantral follicle numbers in a dose- and time-dependent manner. In addition, estrogen upregulated VEGFA expression in the rodent ovary is similar to its effects on early follicle growth [28]. VEGFA also has been shown to be localized in the granulosa and thecal cells of preantral follicles [21], as well as the oocytes of primordial and primary follicles [24, 25]. These findings, along with our present data, provide important evidence that the ovarian VEGF system may be a critical component of early follicle physiology in the rodent ovary. Interestingly, however, in the present study VEGF neutralization only depleted primordial follicles; there were no differences in primary or secondary follicle numbers after either systemic or intrabursal antibody administration. This may reflect the especially labile nature of the primordial follicle pool: this population is exquisitely sensitive to radiation- or chemotherapy-induced follicle destruction. Alternatively, higher doses or more prolonged exposure to neutralizing antibodies might be required for effects on primary and secondary follicles to be manifest.
In addition to depleting primordial follicles in the ovary, systemic administration of VEGF peptide antibodies also disrupted estrous cyclicity in the mouse. Shortly after VEGF antibody administration, mice displayed persistent vaginal cornification similar to that observed in aging mice just prior to ovarian senescence [37]. Neutralization of endogenous VEGF has been demonstrated to inhibit ovulation [38], and it is possible that the presence of cornified cells in the vaginal smears is due to continued estrogen production by unovulated antral follicles persisting in the ovary for a few days after antibody administration. Indeed, four of six VEGF antibody-treated mice had not ovulated by 7 days after systemic antibody administration, whereas six of six control mice had ovulated. Alternatively, systemic administration of VEGF antibodies also might alter uterine and vaginal function, which might manifest in altered vaginal cytology. Other investigators have demonstrated effects of VEGF neutralization on uterine function in rodents [39, 40].
The depletion of primordial follicles by VEGF neutralization is rapid; however, over time this depletion of the follicular pool appeared to abate. Primordial follicle numbers were decreased within 24 h following intrabursal administration of VEGF peptide antibodies and remained below control levels for at least 3 days. Within 30 days the effect of VEGF neutralization was no longer apparent, and primordial follicle numbers were not different from control. It is important to note, however, that the number of primordial follicles did not increase over time after either systemic or intrabursal antibody administration.
It is unknown whether primordial follicle depletion following VEGF neutralization is due to a direct effect of VEGF on follicular units or is mediated via changes in the vascular compartment of the ovary. Certainly, neutralization of endogenous VEGF would likely affect angiogenesis and vascular permeability, as these are two of the primary effects of VEGFA in most tissues studied to date. It is perhaps less likely that changes in new vessel formation are involved, since the effects of VEGF antibody treatment are observed within 24 h, although neutralization of VEGF could result in endothelial cell apoptosis with resultant vascular compromise. We did not quantitate potential changes in angiogenesis in the present study. Changes in vascular permeability and, as such, delivery of important endocrine and paracrine modulators of early follicle growth may be a more important component of the effects observed.
Alternatively, increasing evidence suggests that VEGFA may have direct effects on nonvascular elements in a variety of tissues, including the ovary, although many of these studies are somewhat limited. VEGFA and its receptors have been identified tentatively in granulosa cells, thecal cells, oocytes, and zona pellucida [21, 24–26, 41]. VEGFA also has been shown to directly stimulate granulosa cell function in vitro [22]. As such, it is possible that the effects of VEGFA on the preantral follicle might be mediated by direct actions on follicular components rather than indirectly via alterations in vascular physiology.
Our data indicate that inhibition of VEGF binding to the KDR but not FLT1 receptor is important for primordial follicle survival in the mouse ovary. Ovarian administration of antibodies to KDR resulted in significant depletion of primordial follicle numbers, whereas anti-FLT1 antibodies did not. This is consistent with previous data in the literature indicating that the KDR receptor is the primary mediator of VEGF action and that the FLT1 receptor might serve as a decoy for VEGF [3]. Indeed, Zimmermann et al. [42] demonstrated that antibodies against KDR delayed follicular selection and development in the Rhesus monkey. However, it should be noted that we only used one dose of VEGF receptor antibodies, and it is possible that higher doses, longer treatment, or different FLT1 antibody preparations might also result in primordial follicle depletion similar to our results with KDR antibodies.
A recent report by Johnson et al. [43] proposed that the follicular pool is a dynamic rather than stagnant population. These authors suggested that a constant vascular influx of bone marrow-derived stem cells is required to maintain the primordial follicle population. VEGF neutralization might interfere with the delivery of stem cells to the ovary or disrupt new primordial follicle formation. Kezele et al. [27] has shown that VEGF expression is upregulated during primordial follicle development. While the importance of bone marrow stem cell migration remains controversial, its potential role in the loss of primordial follicles following neutralization of endogenous VEGF warrants further investigation.
We chose to develop our own antibodies against VEGF in order to have an ample supply to conduct long-term in vivo studies and to investigate potential applications of active immunotherapy for other (cancer) studies. In order to develop a pan-specific antibody capable of neutralizing the majority of VEGF isoforms, we chose a conserved peptide sequence found in the majority of VEGF species (VEGF A–G). Coupling antigenic regions of the VEGF molecule with a promiscuous T-cell epitope is a useful strategy for development of novel antigens for generation of polyclonal antibodies and for exploring active immunization for a variety of therapies. This approach resulted in specific VEGF antisera comparable to commercially available sources. Although we have not determined the affinity of this particular antibody, other antibodies developed using this technology displayed affinities similar to traditional polyclonal antibodies generated against intact proteins [31, 44, 45]. Our VEGF peptide antibody did detect an additional higher-molecular weight protein in the pancreatic tumor cell lysate, and the identity of this protein is unknown at present, although its molecular weight is roughly consistent with that of VEGFA189. It is noteworthy that our protein A-purified antibody preparation is much less pure than the affinity-purified antibody preparation we used for comparison from Santa Cruz Biotechnologies, yet the specificities are remarkably similar. Indeed, the data obtained with commercially available antibodies to VEGFA and KDR were nearly identical to those obtained with our VEGF peptide antibodies, suggesting that the effects on primordial follicles we observed were due to specific inhibition of VEGF interaction with its receptor in the ovary and not to nonspecific toxic effects of the antibody preparation. Moreover, the similar response to acute versus long-term antibody treatment argues against nonspecific toxic effects on the ovary. Finally, the lack of effect on primary and secondary follicles suggests that specific neutralization of VEGF is responsible for primordial follicle destruction in the ovary.
In summary, these studies provide compelling evidence that endogenous VEGF is essential for primordial follicle survival in the rodent. Disruption of VEGF interaction with its receptors results in a rapid and pronounced loss of primordial but not other preantral follicles in the ovary. Along with previous studies supporting the role of VEGF in early follicle growth, we conclude that the ovarian VEGF system is an integral component of early follicle growth and survival in the rodent ovary.
FOOTNOTES
1Supported in part by the Ohio Agriculture Research and Development Center (Proposal ID: 2003-094). 
Correspondence: 2Douglas R. Danforth, Department of Obstetrics and Gynecology, 400 Tzagournis Medical Research Facility, 420 W.12th Ave., Columbus, OH 43221. FAX: 614 688 3551; e-mail: danforth.2{at}osu.edu
Received: 11 January 2006.
First decision: 1 February 2006.
Accepted: 16 October 2006.
REFERENCES
| ||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||
| HOME | HELP | FEEDBACK | SUBSCRIPTIONS | ARCHIVE | SEARCH | TABLE OF CONTENTS |
