| HOME | HELP | FEEDBACK | SUBSCRIPTIONS | ARCHIVE | SEARCH | TABLE OF CONTENTS |
Original Article |
Developmental Endocrinology Branch (J.A.A., C.B., J.Z.), National Institute of Child Health and Human Development, National Institutes of Health, Bethesda, Maryland 20892; and Institut National de la Recherche Agronomique (P.M.), Physiologie de la Reproduction des Mammifères Domestiques, Unité de Recherche Associée Centre National de la Recherche Scientifique 1291, 37380 Nouzilly, France
Address all correspondence and requests for reprints to: Jose A. Arraztoa, M.D., Building 10, Room 10N262, National Institutes of Health, Bethesda, Maryland 20892. E-mail: arrazj{at}mail.nih.gov.
Abstract
IGFs and IGF-binding proteins (IGFBPs) are thought to play important roles in ovarian follicular growth and selection. To elucidate the role of IGFBPs in primate ovarian function, we analyzed IGFBP mRNA expression patterns in ovaries from mid-cycle rhesus monkeys using in situ hybridization. IGFBP-1 mRNA was concentrated in theca-interstitial cells and was present at low levels in granulosa cells of atretic follicles. IGFBP-2 mRNA was expressed in the ovarian surface epithelium and granulosa cells of all antral follicles, including obviously atretic as well as dominant follicles. IGFBP-3 mRNA was localized in oocytes and in the ovarian vascular endothelium; this mRNA was also concentrated in the superficial cortical stroma in which it was distinctly more abundant in the nondominant ovary. Granulosa cells of mature dominant and ovulatory follicles selectively expressed IGFBP-5 mRNA. IGFBP-5 mRNA was also widely expressed in the ovarian stroma, in which, in contrast to IGFBP-3, it was distinctly more abundant in dominant, compared with nondominant, ovary. IGFBP-6 mRNA was present at low levels in the ovary interstitium and theca externa and was more abundant in the ovary surface epithelium. These novel data reveal distinctive cellular expression patterns for IGFBPs 1, 2, 3, 5, and 6 in the nonhuman primate ovary, suggesting distinct roles for each binding protein in ovarian function.
IGFs AND IGF-BINDING PROTEINS (IGFBPs) are thought to play important roles in ovarian follicular growth and selection. IGF1 expression is correlated with ovarian follicular growth and selection in the rodent (1, 2, 3, 4) and IGF1 gene deletion results in arrested follicular development at the preantral/early antral stage (5). IGF1-null ovaries demonstrate resistance to exogenous gonadotropins and reduced follicular FSH receptor expression and reduced granulosa cell proliferation (6, 7, 8). Although IGF1 has an essential role in murine ovarian follicular development, IGF2 is more abundant and likely more important in the human and nonhuman primate ovary (9, 10, 11). It is unknown whether the role of IGF2 in the human ovary replicates that of IGF1 in the murine ovary. In vitro studies have shown that IGF2 promotes human granulosa cell proliferation and steroidogenesis (12); however, the cellular expression patterns of IGF2 in the human ovary appear different from that of IGF1 in the rodent (11, 13, 14).
IGFBPs bind both IGFs with high affinity and are abundant in ovarian follicles (15, 16, 17), in which their function remains a matter of intense investigation. IGFBPs bind IGFs in the bloodstream and other body fluids and thereby increase their half-life and may also stabilize or neutralize the ligands in extracellular spaces (18). Because of their ability to inhibit IGF effects, it has been suggested that IGFBPs may be involved in follicular atresia. However, these IGFBPs may also have IGF-independent effects (18). Detailed studies of cellular patterns and hormonal and/or cyclic regulation of IGFBP expression in the human ovary have been limited for obvious reasons. To further elucidate the potential roles/interactions of these factors in human follicular development, we turned to the rhesus monkey, which has an ovarian cycle identical to the human. Preliminary studies showed similar patterns of IGF system expression in the human and rhesus ovary (13). Hence, in the present study, we have carried out a careful examination of cellular patterns of IGFBP-1, -2, -3, -5, and -6 mRNA expression in both dominant and nondominant ovaries of the late-follicular Rhesus macaque monkey. IGF1 and 2 and IGFBP-4 expression in the monkey ovary is detailed in another study.
Materials and Methods
Animals
Female rhesus monkeys (Macaca mulatta) 6–13 yr of age from the NIH colony were used in accordance with a protocol approved by the NICHD Animal Care and Use Committee. Menstrual cycles were charted for 4 months with the first day of vaginal bleeding counted as cycle d 1. Three animals that demonstrated regular cycles over the 4-month observation period (average cycle length was 28 ± 3 d) underwent ovariectomy under ketamine anesthesia via a midventral laparotomy on cycle d 14. Ovaries were snap frozen on dry ice and stored at -70 C. Serial sections of 10-µm thickness were cut at -15 C and thaw mounted onto poly-L-lysine-coated slides for in situ hybridization and immunohistochemistry.
In situ hybridization
The IGFBP cDNA clones used for RNA probe synthesis have been described previously (11). 35S-labeled RNA probes were synthesized to a specific activity of about 2 x 108 dpm/µg in a protocol that has been previously described (19). Briefly, the sections were fixed; soaked for 10 min in 0.25% acetic anhydride, 0.1 mol/liter triethanolamine hydrochloride, and 0.9% NaCl; washed; and dehydrated. The 35S-labeled probes (107 cpm/ml) were added to hybridization buffer composed of 50% formamide, 0.2 mol/liter NaCl, 50 mmol/liter Tris HCl (pH 8), 2.5 mmol/liter EDTA, 250 µg transfer RNA/ml, 10% dextran sulfate, 10 mmol/liter dithiothreitol, and 0.02% each of BSA, Ficoll, and polyvinylpyrrolidone. Coverslips were placed over the sections, and the slides were incubated in humidified chambers overnight (14 h) at 55 C. Slides were washed several times in 4x NaCl and sodium citrate (Biofluids Inc., Rockville, MD) to remove coverslips and hybridization buffer; dehydrated; and immersed in 0.3 mmol/liter NaCl, 50% formamide, 20 mmol/liter Tris HCl, and 1 mmol/liter EDTA at 60 C for 15 min. Sections were then treated with ribonuclease A (20 µg/ml) for 30 min at room temperature, followed by a 15-min wash in 0.1x NaCl and sodium citrate at 50 C. Slides were air dried and apposed to Hyperfilm-ß Max (Amersham, Arlington Heights, IL) for 3–7 d and then dipped in NTB2 nuclear emulsion (Kodak, Rochester, NY), stored with desiccant at 4 C for 3–30 d, developed, and stained with Mayer’s hematoxylin and eosin for microscopic evaluation.
The specificity of the in situ hybridization results was confirmed by the fact that each of the antisense probes yielded a unique spatiotemporal pattern in the ovary section and the hybridization of sections with sense probe produced a nearly undetectable radioactive signal (Fig. 1
). Parallel sections were hybridized to sense probes and exposed together with antisense hybridized sections. The quantification of hybridization signal was carried out in a blinded fashion. The hybridization signal was captured at x100 using a monochrome video camera. Signal overlying interstitial tissue (excluding follicles) of the ovary was captured using NIH image version 1.57 software. Both cortical and central stromal regions were sampled. Ten measurements were obtained and meaned for each region in each animal. The values were compared between right and left ovaries using ANOVA. Significant differences among means were determined by Fisher’s least significant difference test.
|
Immunocytochemistry for CD45 surface antigen was performed by the avidin-biotin-immunoperoxidase technique. The mouse antihuman CD45 (Chemicon International, Temecula, CA) detects the major cell surface glycoprotein analog of the murine and rat T200 molecules, which is found on virtually all hemopoietic cells. Fresh-frozen tissue sections were fixed in 4% formalin/PBS for 10 min. After blocking in 2% hydrogen peroxide/10% methanol/PBS, avidin-biotin blocking reagent, and 2% normal goat serum, tissue sections were incubated overnight at 4 C with a 1:50 and 1:100 dilution of mouse antihuman CD45 antibody or with 1% PBS/BSA as negative control. Frozen sections of monkey spleen were used as a positive control with a 1:100 dilution. Thereafter, tissue sections were treated with biotinylated goat antimouse IgG (1:200) for 45 min at room temperature, followed by a 30-min incubation with the avidin biotin peroxidase complex (Vectastin ABC Elite Peroxidase Kit; Vector Laboratories, Burlingame, CA). The antigen-antibody complex was visualized by incubation with freshly prepared 3,3'-diaminobenzidine (DAB substrate kit; Vector Laboratories), and the tissue was counterstained with methyl green.
Results
Both ovaries from each of three mid-cycle monkeys were examined in this study. In each case a single, large (maximum diameter, >2 mm) dominant follicle (DF) was located in one of the ovaries, with multiple smaller antral follicles present in both ovaries. One and in some cases two to three residual corpora lutea (CL) were present in these ovaries. Sequential sections from both ovaries for each animal were evaluated for expression of IGFBP-1, -2, -3, -5, and -6 mRNAs, providing an abundant number of follicles in all anatomic stages of development for analysis. These same follicles were typed in sequential sections for expression of the Ki67 proliferation antigen, aromatase, and FSH and LH receptors (Zhou, J., J. Wang, D. Penny, P. Monget, J. A. Arraztoa, L. J. Fogelson, and C. A. Bondy, submitted for publication) to characterize their functional status. Typical patterns of IGF2 and the IGF1 receptor expression are illustrated in Fig. 2.
|
IGFBP-1 demonstrates a rather obscure expression pattern in the primate ovary (Figs. 2A
and 3). IGFBP-1 mRNA was not detected in healthy granulosa but was present at relatively low levels in granulosa cells of atretic follicles (Fig. 3, A and B). IGFBP-1 mRNA was abundant in stromal cells clustered around degenerating follicles (Figs. 2A and 3, A–D). Little IGFBP-1 was detected in residual CL in the mid-cycle ovary (Fig. 3, E and F). We considered the possibility that the intensely IGFBP-1-positive stromal cells could belong to the lymphoreticular system. Therefore, we compared IGFBP-1 mRNA distribution with immunohistochemical detection of the lymphoid CD45 antigen, which is expressed by lymphocytes and tissue macrophages, in sequential ovary sections. Although numerous cells in the monkey spleen were CD45 positive, very little staining was detected in the ovary, and there was no correlation with IGFBP-1-positive cell distribution (data not shown).
|
IGFBP-2 mRNA was very low in primary and early antral follicles but was abundant in granulosa cells in all antral follicles (Figs. 2C and 4, A and B). We did not appreciate any reduction in IGFBP-2 mRNA level in DF granulosa, compared with other antral follicles, and it was in fact very abundant in the ovulatory granulosa (Fig. 5C). IGFBP-2 mRNA was not detected in the mid-cycle CL (data not shown). This mRNA was also concentrated in the ovary surface epithelium (Fig. 2C).
|
|
IGFBP-3 mRNA was heavily concentrated in the superficial cortical stromal (Figs. 6 and 7). Interestingly, stromal IGFBP-3 mRNA levels were significantly higher in the nondominant, compared with the dominant, ovary (Fig. 6, A and B). This pattern was observed in each of the three mid-cycle monkeys. Quantification of hybridization signal in cortical stroma revealed approximately 2-fold higher IGFBP-3 mRNA in the nondominant ovary (167 ± 23 vs. 76 ± 18 grains per high-power field, P = 0.04). IGFBP-3 mRNA was also localized in oocytes of preantral follicles with more than one layer of granulosa, and early antral and antral follicles, and the endothelium of ovarian blood vessels (Fig. 7). IGFBP-3 mRNA was detected in the theca externa of larger antral follicles, but this signal appears to originate in perifollicular capillaries. IGFBP-3 mRNA was present at low levels in the CL, in which it also appeared to be localized mainly in microvessels (Fig. 6A).
|
|
IGFBP-5 mRNA was abundant throughout the stroma of the dominant ovary but was distinctly less abundant in the nondominant ovary (Figs. 6, C and D, and 7), and this pattern was consistent for all three monkeys. Quantification of hybridization signal in cortical and central stroma revealed approximately 6-fold higher IGFBP-5 mRNA in the dominant, compared with the nondominant, ovary (243 ± 55 vs. 36 ± 5 grains per high-power field, P = 0.02). IGFBP-5 mRNA was localized in granulosa and theca cells of dominant (Fig. 6D) and ovulatory follicles (Fig. 5F) but not in the granulosa of less mature antral follicles (Fig. 7, C and D).
IGFBP-6
IGFBP-6 mRNA was present at low levels throughout the ovarian interstitium, in some cases coalesced in the theca externa. It was also detected at very low levels in the granulosa cells of antral follicles without apparent selectivity for healthy or atretic stages. It was most abundant in the surface epithelium of the ovary (Fig. 5F).
The findings on IGFBP-1, -2, -3, -5, and -6 mRNA localization from this study are summarized in Table 1. All five binding protein mRNAs are also compared in sequential sections through the same ovulatory follicle in Fig. 5.
|
This work shows that IGFBP-1, -2, -3, -5, and -6 mRNAs are each expressed in a distinct, cell-specific manner in the monkey ovary. Moreover, there is a difference in IGFBP-3 and five mRNA expressions in dominant vs. nondominant ovary, which is an intriguing and, to our knowledge, novel observation. These distinctive expression patterns suggest that these IGFBPs have distinct roles in diverse aspects of ovarian function.
IGFBP-1 mRNA is expressed in two histologically distinct cell populations in the primate ovary. It is present at relatively low levels in granulosa cells of atretic follicles and is abundant in isolated stromal cells. IGFBP-1 mRNA was homogeneously expressed in the granulosa cells of obviously atretic follicles, in which it may, as a matter of speculation, be involved in granulosa cell apoptosis. Consideration of the potential role of IGFBP-1 in the ovarian stroma is limited by uncertainty about the identity or function of the IGFBP-1-expressing cells. Because of the distribution of IGFBP-1 around degenerating follicles and morphological appearance, we entertained the possibility that these were elements of the lymphoreticular system, e.g. tissue macrophages or lymphocytes. This seems unlikely, however, in view of the failure of CD45 immunostaining to correlate with IGFBP-1 mRNA expression in the ovary. In our study of the IGF system in the human ovary (11), we found IGFBP-1 mRNA in scattered stromal and CL cells but not in developing follicles of any type, consistent with the present findings in the monkey. The difficulty in clearly identifying DFs in our human collection precluded any conclusions on DF expression patterns (11). El-Roeiy et al. (14), however, reported that IGFBP-1 mRNA was abundant in DF granulosa cells in the human ovary. This latter study identified DFs based on size and a healthy appearance of the granulosa cells (14), but only two such follicles were identified. In the present study, DFs were identified as the single largest follicle in either of the animal’s two ovaries that also demonstrates FSH receptor and aromatase expression (Zhou, J., J. Wang, D. Penny, P. Monget, J. A. Arraztoa, L. J. Fogelson, and C. A. Bondy, submitted for publication). In one case the DF was actually in the process of ovulation (Fig. 5). IGFBP-1 mRNA was not detected in any of these DFs. It seems possible that the El-Roeiy criteria for identification of DFs may have inadvertently included luteinized persistent follicles or cysts. Supporting this possibility, IGFBP-1 expression is abundant in luteinized human granulosa cells and corpora lutea (18, 19) and is robustly increased after human chorionic gonadotropin treatment in the monkey (our unpublished data).
IGFBP-2 mRNA is abundant in the ovary of most if not all mammals, but remarkable interspecies differences in expression patterns are observed. For example, in the rat, IGFBP-2 mRNA is confined to the theca-interstitial compartment, whereas in the mouse it is highly expressed in granulosa cells (20, 21, 22, 23). In large domestic animals, IGFBP-2 mRNA is localized in granulosa cells and appears to decrease with follicular growth and FSH effect (24, 25, 26, 27) and may follow a similar pattern in humans (14). In the present study of the rhesus monkey, little IGFBP-2 mRNA was detected in primordial, primary, and small antral follicles, with increasing expression noted in the granulosa cells of larger antral and DFs. IGFBP-2 mRNA was present in the granulosa cells of some but not all atretic follicles. Despite its highly abundant expression in the ovary, there is little understanding of the role IGFBP-2 may play in ovarian function. IGFBP-2-null mice are fertile and appear to have normal ovarian function (28), so it may be that its role is nonessential under normal conditions. Interestingly, there is evidence for a correlation between ovarian and serum levels of IGFBP-2 and epithelial ovarian cancer (28, 29, 30). The present observation that IGFBP-2 mRNA is highly concentrated in the normal ovarian epithelium suggests that IGFBP-2 synthesis is a marker for ovarian epithelial cells and continues to be expressed by cancerous epithelium.
IGFBP-3 demonstrates a complex pattern of mRNA expression in the monkey ovary, distributed in at least three distinct cell types. IGFBP-3 mRNA is strongly expressed in the superficial cortex of the nondominant ovary and is expressed at much lower levels in the dominant ovary cortex. This hybridization signal is localized in stromal cells of the superficial cortex, not in the surface epithelium or tunica. This asymmetry between nondominant and dominant ovary IGFBP-3 expression in normal follicular-phase monkeys is a novel observation of unclear significance. IGFBP-3 mRNA was concentrated in the ovarian vascular endothelium including perifollicular and CL capillaries. This observation agrees with previous reports of IGFBP-3 expression in ovary endothelium in multiple species (11, 23, 31, 32, 33). There has been much discussion of the potential role of IGFBP-3 in CL function (31, 32, 33), but given that IGFBP-3 is also expressed by vascular endothelium in other tissues undergoing neovascularization such as placenta (33) and uterus (34), it seems possible that endothelial cell IGFBP-3 expression in the CL is primarily a feature of vascular biology rather than a factor specifically involved in luteal function. We have also noted in this study the localization of IGFBP-3 mRNA in oocytes of antral follicles, which is a novel finding to our knowledge. IGFBP-3 is also expressed by some populations of steroidogenic cells, as seen in the mature, possibly luteinized granulosa of porcine (35) and human follicles (14).
IGFBP-5 mRNA is quite abundant and homogeneously distributed throughout the ovarian stroma in humans (11) and monkeys (this study). Stromal IGFBP-5 mRNA expression is distinctly asymmetrical, however, being much higher in the dominant, compared with the nondominant, ovary, opposite the pattern for IGFBP-3. It is unknown whether asymmetric IGFBP-3 and -5 expression is characteristic of other primates because, to our knowledge, no other studies have compared expression in both ovaries from the same individual. One potential explanation for the asymmetric expression patterns is that higher blood flow to the dominant ovary (note the massive vascularization seen in Fig. 5A) somehow enhances IGFBP-5 and suppresses IGFBP-3 expression. IGFBP-5 expression in granulosa cells appears to be quite species dependent. In the rodent it is concentrated in a subpopulation of small follicles thought to be headed for selection [mouse (23)] or atresia [rat (36)], although in our view it is simply not possible to determine on morphological ground alone which fate awaits these IGFBP-5-expressing follicles (22). In the monkey, in contrast, IGFBP-5 is not detected in smaller follicles but is expressed at high levels in the DF. El-Roeiy et al. (14) also report IGFBP-5 mRNA in the human DF. In the porcine ovary, however, IGFBP-5 is not detected in the granulosa of any follicles, small or large (35). In the sheep, cow, and goat as well as in the rat, IGFBP-5 mRNA is detected in the granulosa of atretic follicles (25, 36). This marked species-specific variability in follicular IGFBP-5 expression suggests variable and probably dispensable functions for this binding protein in ovarian biology.
IGFBP-6 is the least abundant of all the IGFBPs in the ovary and demonstrates a rather nondescript expression pattern, not strongly suggestive of specific functions. These findings in the rhesus monkey are consistent with the reported scarcity of this IGFBP in the human ovary (14, 37).
In summary, this study has documented distinctive cellular expression patterns for IGFBPs 1, 2, 3, 5, and 6 in the rhesus monkey ovary. This detailed information provides an essential context or infrastructure for the elaboration of specific hypotheses about the function of the IGF system in the primate ovary. Identification of similar or identical expression patterns in the primate and other species may provide insight into the role of the IGF system in ovarian function. For example, the very restricted or absent IGFBP-1 expression in developing and DFs of both human and monkey suggests that this factor does not play a role in human follicular growth or ovulation but may have a role in the regression or luteinization of granulosa cells. The concentration of IGFBP-3 in perifollicular blood vessels of growing follicles and CL in primate and other species implies a role in angiogenesis or possibly transfer of circulating IGFs into the ovary. Variable ovarian expression patterns for IGFBP-2 and -5 in different species do not lend themselves to obvious functional correlation and may indicate redundancy/dispensability.
Acknowledgments
We thank Ricardo Dreyfuss for expert photomicrography.
Footnotes
Abbreviations: CL, Corpora lutea; DF, dominant follicle; IGFBP, IGF-binding protein.
Received March 14, 2002.
Accepted July 22, 2002.
References
This article has been cited by other articles:
 |
|
 |
|
  J. A. Arraztoa, J. Zhou, D. Marcu, C. Cheng, R. Bonner, M. Chen, C. Xiang, M. Brownstein, K. Maisey, M. Imarai, et al. Identification of genes expressed in primate primordial oocytes Hum. Reprod., February 1, 2005; 20(2): 476 - 483. [Abstract] [Full Text] [PDF]
|
|
|
|
 |
|
 E. M. Ongeri, Q. Zhu, M. F. Verderame, and J. M. Hammond Insulin-Like Growth Factor-Binding Protein-3 in Porcine Ovarian Granulosa Cells: Gene Cloning, Promoter Mapping, and Follicle-Stimulating Hormone Regulation Endocrinology, April 1, 2004; 145(4): 1776 - 1785. [Abstract] [Full Text] [PDF]
|
|
| |||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||
| HOME | HELP | FEEDBACK | SUBSCRIPTIONS | ARCHIVE | SEARCH | TABLE OF CONTENTS |
| Endocrinology | Endocrine Reviews | J. Clin. End. & Metab. |
| Molecular Endocrinology | Recent Prog. Horm. Res. | All Endocrine Journals |
