| HOME | HELP | FEEDBACK | SUBSCRIPTIONS | ARCHIVE | SEARCH | TABLE OF CONTENTS |
Other Original Article |
Department of Physiology and Cell Biology, University of Pittsburgh School of Medicine, Pittsburgh, Pennsylvania 15261
Address all correspondence and requests for reprints to: Anthony J. Zeleznik, Ph.D., 830 Scaife Hall, University of Pittsburgh School of Medicine, 3550 Terrace Street, Pittsburgh, Pennsylvania 15261. E-mail: zeleznik{at}pitt.edu.
Abstract
Although it is well established that IGF-I is able to amplify the actions of FSH and LH on ovarian cells in vitro, little information is available regarding the effects of IGF-I on ovarian function in vivo. To address this question, rhesus monkeys whose spontaneous gonadotropin secretion was interrupted with a GnRH antagonist received continuous iv infusions of saline, IGF-I (240 µg/kg·d), or IGF-I (240 µg/kg·d) plus human GH (hGH) (200 µg/kg·d) 7 d before and continuing throughout a 15-d iv infusion of hFSH and hLH during which serum LH concentrations were maintained at 7–10 mIU/ml and FSH concentrations were incrementally increased every 3 d from 7.5 to 17.5 mIU/ml. Serum estradiol concentrations in saline-treated control animals did not differ (P > 0.05) from animals treated with IGF-I + hGH. In contrast, serum estradiol levels in IGF-I-treated animals were significantly less (P < 0.05) than those of control or IGF-I + hGH-treated animals. Serum androstenedione levels did not differ among the three treatment groups. Analysis of follicular fluids on the final day of gonadotropin infusion indicated that intrafollicular IGF-I concentrations paralleled serum IGF-I concentrations in all treatment groups. Measurement of the ratio of IGF-I to IGF-binding protein-3 in follicular fluids indicated that there was not a disproportionate increase in I-binding protein-3 in animals infused with either IGF-I alone or IGF-I + hGH. Concentrations of GH in follicular fluids of IGF-I treated animals were less than control animals suggesting that the diminished responsiveness of ovaries to FSH in the IGF-I treatment group may have been due to reduced GH.
IT IS WELL ESTABLISHED that FSH and LH are essential for the maturation of the ovarian follicle and its secretion of estrogen (1). In addition, numerous in vitro studies have also revealed that other peptides and steroid hormones are able to modify the actions of FSH and LH on their respective ovarian target cells. This has led to the hypothesis that the normal regulation of ovarian function by FSH and LH could be affected, in either a positive or negative manner, by the prevailing milieu of other endocrine, paracrine, or autocrine factors (2, 3, 4). For example, because both insulin and IGF-I exert synergistic effects with FSH and LH on follicular steroid biosynthesis, it has been proposed that pathological disturbances in the secretion and/or actions of insulin and/or IGF-I could be causal to the onset of polycystic ovarian syndrome in humans (5, 6, 7, 8).
To date, however, virtually all our information on the biological actions of nongonadotropic regulators of ovarian function are based on in vitro studies and the extent to which results from these investigations can be extrapolated to the in vivo state remains uncertain. To address this important question, we initiated the present study to develop an experimental model in nonhuman primates that could be used to examine directly the effects of various autocrine and paracrine agents on responsiveness of the primate ovary to FSH and LH in vivo. We chose first to explore the effects of IGF-I because of its well-documented in vitro effects on follicular function in both primates and nonprimate species.
Materials and Methods
Animals
Female rhesus monkeys (Macaca mulatta), 5–8 kg body weight, with normal menstrual cycle histories were used in this study and were housed under standard husbandry conditions at the University of Pittsburgh Primate Research Laboratory. All experimental procedures were approved by the University of Pittsburgh Institutional Animal Care and Use Committee. Each animal was equipped with three catheters inserted into the jugular and femoral veins and exteriorized through a small incision in the scapular region. The catheters were protected by a vest and flexible stainless steel cable and were connected to a three-channel swivel device (Spalding Medical Products, Arroyo Grand, CA) mounted to the roof of the cage.
Effects of IGF-I and IGF-I+human (h) GH on ovarian steroidogenesis
On the day of catheterization, each animal began receiving im injections of a GnRH receptor antagonist (Acycline, 60 µg/kg in 5% mannitol) every 2 d until the completion of the study to suppress endogenous gonadotropin secretion. Acycline was synthesized by Bioqual (Rockville, MD) and was kindly provided by the Contraception and Reproductive Health Branch, Center for Population Research, NICHHD, NIH. Seven days later, animals (n = 3 per group) began receiving a continuous infusion of vehicle (saline containing 0.1% rhesus monkey serum and 100 U/ml penicillin and 100 µg/ml streptomycin), IGF-I (240 µg/kg·d), or IGF-I (240 g/kg·d plus hGH (200 µg/kg·d), which continued until the completion of the study. hIGF-I and hGH were generously provided by Genentech, Inc. (South San Francisco, CA).
Seven days thereafter an intermittent infusion of hFSH and hLH at a frequency of one 3-min pulse/h was initiated. This frequency was chosen because it corresponds to the spontaneous gonadotropin pulse frequency during the follicular phase of the macaque menstrual cycle (9). hFSH (AFP-8792B, 1685 IU/mg) and hLH (AFP 7572B, 4500 IU/mg) were generously provided by Dr. A. Parlow and the National Hormone and Pituitary Program, NIDDK, NIH. Stock solutions were diluted into saline containing 0.1% rhesus monkey serum and 100 U/ml penicillin and 100 µg/ml streptomycin and were infused in a total volume of 0.5 ml for each 3-min pulse. The hLH infusion was set at dosage of 75 ng/kg per pulse, which was estimated to produce a serum LH concentration of 6–8 mIU/ml (10). The hFSH infusion was set at 54 ng/kg per pulse, which was estimated to produce a serum FSH concentration of 7.5 mIU/ml (10). Approximately 48 h after initiating the gonadotropin infusion, serum concentrations of hFSH and hLH were determined by RIA and, if necessary, the amounts of FSH and LH administered were adjusted by direct proportion to achieve the desired serum concentrations. Thereafter, the amount of LH delivered per pulse remained constant throughout the entire 15-d treatment interval. The amount of FSH delivered per pulse was incrementally increased every 3 d to produce stepwise elevations in serum FSH concentrations of 7.5, 10, 12.5, 15, and 17.5 mIU/ml over the 15-d treatment interval. On the fifth day of gonadotropin infusion, the animals were anesthetized with 4% isflourane, and the left ovaries were removed through a midventral incision, weighed, and frozen. On the final day of gonadotropin infusion, the right ovaries were removed, weighed, and frozen.
Effect of IGF-I on GH secretion
Following the termination of the aforementioned experiment, animals were removed from the tether system and catheters were filled with saline and isolated in an sc pouch. Two months thereafter animals were reattached to the tethers for analysis of the effects of IGF-I on GH and gonadotropin secretion. All animals first received a 7-d infusion of vehicle (saline containing 0.1% rhesus monkey serum, 100 U/ml penicillin, and 100 µg/ml streptomycin). Thereafter control animals (n = 3) continued to receive vehicle for 7 d, and experimental animals (n = 3) received IGF-I (240 µg/kg·d) for 7 d. Animals continued to receive either vehicle or IGF-I for an additional 7 d and in addition received sc a Silastic (Dow Corning, Inc. Corp., Midland, MI) capsule containing crystalline estradiol, which elevated peripheral estrogen concentrations to approximately 50 pg/ml. On the final day of each of the 7-d treatment regimens, blood samples were collected at 10-min intervals for a period of 6 h. Serum was collected from blood samples and stored at -20C until analyzed for GH concentrations by RIA.
RIAs
Serum levels of hFSH, hLH, hGH, estradiol, and androstenedione were measured by kits purchased from Diagnostic Products (Los Angeles, CA). Total IGF-I was measured with a kit purchased from Diagnostic Systems Laboratories (Webster, TX).
Statistics
Statistical analyses were performed using Statview version 4.5 (Abacus Concepts, Inc., Berkeley, CA). Serum levels of estradiol and androstenedione during the interval of gonadotropin infusions were analyzed for statistical differences using ANOVA with repeated measures. Follicular fluid concentrations of IGF-I, GH, IGF-binding protein-3 (IGFBP-3), and the IGF-I/IGFBP-3 ratios of animals infused with IGF-I and IGF-I+hGH were compared with those of control animals by unpaired, two-tailed t tests.
Results
Serum concentrations of IGF-I and hGH in infused monkeys
Figure 1
illustrates serum concentrations of total IGF-I in the three groups of animals. Before the initiation of IGF-I or IGF-I+hGH treatment, there were no differences in total IGF-I concentrations in the three groups of animals (P > 0.5). Mean levels of total IGF-I in IGF-I-infused animals were approximately twice those of saline-treated controls throughout the infusion period, although these differences did not reach statistical significance (P = 0.075). Total IGF-I concentrations in animals that received IGF-I+hGH were significantly greater than both the control animals and the animals that received IGF-I alone (P < 0.05).
|
illustrates GH levels in the three groups of animals. Daily serum GH levels were comparable in control and IGF-I-treated animals. GH levels in the animals that received the combined IFG-I+hGH infusion were more than 25-fold greater than saline and IGF-I-infused animals (P < 0.05). Serum IGFBP-3 concentrations are illustrated in Fig. 3. To control for differences in individual animals, results are expressed as a ratio of preinfusion values. IGFBP-3 concentrations in IGF-I+hGH-treated animals were elevated above preinfusion values throughout the interval of infusion. IGF-I-infused animals displayed an initial rise in IGFBP-3 concentrations, which declined during the infusion interval. Control animals exhibited a slight rise in IGFBP-3 levels at the end of the study.
|
|
Figure 4 (top) illustrates serum FSH concentrations during the 15th day infusion interval during which the amount of FSH infused was incrementally increased every 3 d. At the initiation of the FSH infusion, FSH concentrations rose to approximately 7 mIU/ml and increased progressively throughout the infusion interval. Mean concentrations of FSH were similar in the three treatment groups throughout the duration of the gonadotropin infusion. Figure 4 (bottom) illustrates serum LH concentrations during the gonadotropin infusion. At the initiation of the LH infusion, serum LH concentrations rose to approximately 7 mIU/ml and remained near that level throughout the entire infusion interval. LH concentrations were comparable in the three groups of animals.
|
Figure 5 illustrates serum estradiol concentrations during the 15-d interval of gonadotropin infusion, (as shown in Fig. 4
) in control animals and animals that received a continuous infusion of either IGF-I alone or IGF-I+hGH before and during the gonadotropin infusion. As expected, estradiol secretion was stimulated in all three groups of animals. When these data were analyzed by ANOVA with repeated measures, there was no difference in the overall pattern of estrogen secretion between the control animals and the animals that received the IGF-I+hGH infusion. However, the daily concentrations of estrogen in the animals that received IGF-I alone were lower (P < 0.05) than both the control and IGF-I+hGH-infused animals. Although there was not a significant difference in the overall pattern of estrogen levels between control and IGF-I+hGH-infused animals, analysis of daily estrogen levels by one-way ANOVA indicated that the estrogen concentrations on the third day of FSH and LH infusion was greater (P < 0.05) in the animals that received IGF-I+hGH, compared with control animals and animals that received IGF-I, as indicated by the asterisk. Serum levels of androstenedione, used as an index for ovarian androgen production, were unaffected by either IGF-I or IGF-I+hGH during the gonadotropin infusion regimen (Fig. 6).
|
|
illustrates that the mean of total IGF-I concentrations in follicular fluids was greater in IGF-I- and IGF-I+hGH-infused animals in comparison with control animals. Likewise, mean levels of GH in follicular fluids collected from IGF-I+hGH-treated animals were significantly greater than control animals (P < 0.01). The mean concentration of GH in follicular fluids from animals that received IGF-I was significantly less than control animals (P < 0.01). We also determined the ratio of IGF-I to IGFBP-3 to assess the relative changes in IGF-I and IGFBP-3 in follicular fluids. The IGF-I/IGFBP-3 ratios in IGF-I- and IGF-I+hGH-infused animals were greater than those of control animals, indicating that IGF-I entered the follicle in a biologically active form.
|
illustrates ovarian weights collected on the eighth day of gonadotropin infusion when plasma concentrations of FSH were approximately 12.5 mIU/ml (ovary 1) and on the final day of gonadotropin infusion (ovary 2). Although there were no statistically significant differences among the groups because of the variations among individual animals, the ovarian weights paralleled the estrogen production rates in that the mean weight of ovaries collected from the IGF-I treatment group was less than that of both the control and the IGF-I+hGH-treated animals.
|
Figure 8 illustrates the temporal pattern of GH secretion in vehicle-treated animals and animals infused with IGF-I. Control animals (left) displayed pulsatile GH secretion during the 6-h sampling window during all three infusion periods (as described in Materials and Methods). By comparison, all experimental animals exhibited pulsatile GH secretion during the saline vehicle infusion interval, which, by subjective analysis, was suppressed in two of three animals during the infusion of IGF-I.
|
Herein we describe the development of a gonadotropin-clamp experimental model in rhesus monkeys in which the responsiveness of the ovary to FSH and LH can be assessed directly in the presence of other proposed nongonadotropic regulators of ovarian function. Using this model, we show that a sustained elevation of serum IGF-I 7 d before and continuing throughout a 15-d FSH and LH infusion regimen did not increase the sensitivity of the ovary to FSH or the maximal responsiveness of the ovary to FSH. In addition, neither IGF-I, alone or in combination with hGH, affected blood androstenedione levels, which we used as an index of LH-dependent theca cell androgen secretion. Collectively, these results indicate that, in contrast to its dramatic effects in vitro on both granulosa cell and theca cell steroidogenesis, IGF-I did not have a similar gonadotropin-amplifying actions on the primate ovary in vivo. The lack of a synergistic effect of IGF-I on ovarian steroidogenesis in monkeys is consistent with the observations that IGF-I-deficient humans (Laron dwarfs) achieve pregnancies spontaneously (11) and are able to respond to exogenous gonadotropins (12). In fact, our results indicate that chronic treatment with IGF-I actually diminished ovarian responses to FSH and LH because serum estradiol levels during FSH and LH infusions were significantly less than those observed in control and IGF-I+hGH-treated animals. Similarly, although not significantly different, ovarian weights in the IGF-I-treated animals were less than controls and IGF-I+hGH-treated animals.
The inability of IGF-I infusions to augment gonadotropin-stimulated ovarian function did not appear to be due to a restricted access of systemically infused IGF-I to enter the follicular compartment in a biologically active form. Our analysis of follicular fluids collected on the final day of gonadotropin infusion indicated that the immunoreactive IGF-I levels in follicular fluids paralleled the serum levels of IGF-I in all treatment groups. In addition, we analyzed the IGF-I/IGBP3 ratios in follicular fluids as an index of bioavailable IGF-I and found that this ratio was increased in animals infused either with IGF-I alone or in combination with hGH. Because IGFBP-3 is the major IGF-I binding protein in follicular fluid (13), it does not appear that a disproportionate increase in intrafollicular IGFBP-3 can account for the lack of effect of IGF-I on the ovary in vivo.
Serum concentrations of IGFBP-3 were also measured during the course of the study as an index of responsiveness to the IGF-I and IGF-I+hGH infusions. Animals infused with IGF-I+hGH exhibited a sustained elevation in serum IGFBP-3 concentrations throughout the study, consistent with the notion that GH appears to be the principal regulator of IGFBP-3 production with IGF-I playing a lesser role (14). Animals infused with IGF-I exhibited an initial rise in IGFBP-3 levels that gradually declined during the infusion interval. We hypothesized that the fall in IGFBP-3 levels in the IGF-I-treated animals was due to negative feedback inhibition of IGF-I on GH secretion. However, as shown in Fig. 2, our daily blood-sampling regimen was not sufficient to detect differences in GH concentrations between control and IGF-I-infused animals. In an experiment that was designed to determine the effect of IGF-I on gonadotropin secretion, we infused animals with either vehicle or IGF-I in the absence and presence of estradiol and withdrew blood samples every 10 min for 6 h during three infusion periods. These samples provided an opportunity to more completely assess GH secretion in control and IGF-I-treated animals. Subjective analysis of GH levels in these samples indicated that IGF-I suppressed GH secretion in two of three animals. In addition, follicular fluid GH levels were significantly less in IGF-I-infused animals, compared with control animals (Table 1). Because the follicle is permeable to small proteins, follicular fluid provides an index of the integrated concentration of plasma proteins over time as can be seen by the comparison of serum IGF-I levels in Fig. 1 with follicular fluid IGF-I concentrations in Table 1. Taken together, these observations indicated that IGF-I was biologically effective (i.e. suppression of GH secretion) and that the absence of synergistic actions of IGF-I on gonadotropin-stimulated ovarian steroidogenesis was not due to a systemic resistance to IGF-I.
In humans, it has been suggested that IGF-II rather than IGF-I may be the prominent ovarian ligand (15); hence, our failure to observe a response to exogenous IGF-I in vivo may reflect this difference. Although IGF-II may be the physiologically relevant ovarian IGF in primates, it appears in the human ovary that the actions of IGF-I and IGF-II are both mediated through the IGF-I receptor and that the potencies of IGF-I and IGF-II on a mass basis are similar (16). Therefore, the exogenously infused IGF-I would have occupied the physiologically meaningful receptor and presumably would have activated the same intracellular signaling pathways as IGF-II. It is possible that our 7-d pretreatment with IGF-I followed by continued IGF-I treatment during the 15-d period of gonadotropin infusion was not sufficient time for IGF-I to manifest its actions. However, from in vitro studies, it is apparent that IGF-I affects both FSH- and LH-simulated steroidogenesis within a period of 3–4 d (17, 18); thus, we expected to observe similar effects in our current study. However, we cannot rule out the possibility that a longer infusion of IGF-I would have increased the ovarian responsiveness to FSH and LH. Finally, because of our limited sample size, we cannot rule out the possibility that a small subset of animals could have responded differently to the IGF-I treatment.
To date, there have been very few studies in which IGF-I was administered long term in vivo and ovarian activity was assessed. Spicer et al. (19) infused IGF-I directly into ovaries of cows for 7 d and did not observe differences in ovarian weights, and the numbers of small, medium, and large follicles between control and IGF-I-treated animals. Analysis of follicular fluids revealed no differences in concentrations of androstenedione or progesterone between control and IGF-I-infused ovaries. Follicular fluid estrogen concentrations were slightly greater in small follicles from IGF-I-infused animals but estrogen concentrations in larger follicles did not differ between the two groups. In experiments designed to examine the effects if IGF-I on the onset of puberty, Wilson (20) treated rhesus monkeys with IGF-I at a dosage of 110 µg/kg·d from 18 to 36 months of age and found that serum estradiol levels during the first ovulatory cycles did not differ between control and IGF-I-treated animals, and serum progesterone levels were slightly greater in IGF-I-treated animals. It is uncertain if the elevated progesterone was due to a direct effect of IGF-I on the corpus luteum or was secondary to the elevated LH levels that were seen in IGF-I-treated animals. Klinger et al. (21) administered IGF-I (120 µg/kg·d) to humans with primary growth hormone resistance and observed signs of hyperandrogenism including acne, increases in serum androgens, and increased LH/FSH ratios in two of four women.
Presently we have no explanation for the difference between our findings and those of Klinger et al. However, it should be emphasized that in our studies, FSH and LH concentrations were clamped at defined levels, but the studies of Wilson (20) and Klinger et al. (21) were conducted in the presence of an intact hypothalamic-pituitary-ovarian axis in which IGF-I could exert effects on gonadotropin secretion. In this regard, Wilson (22) noted that prolonged IGF-I treatment of prepubertal rhesus monkeys reduced the ability of estrogen to suppress LH secretion. In preliminary studies, we have also seen that IGF-I appears to reduce the negative feedback actions of estrogen on FSH and LH secretion in ovariectomized adult female rhesus monkeys (data not shown). Blunted estrogen negative feedback would lead to elevations in LH levels that conceivably could alter ovarian function, although it is uncertain whether elevated LH concentrations alone could produce a polycystic ovarian syndromelike phenotype (23).
Although the analysis of the effects of GH on the ovary was not an original aim of the current studies, our results suggest that GH may in fact have synergistic effects on ovarian responsiveness to gonadotropins as proposed previously (24, 25, 26). Thus, in animals treated with IGF-I alone, ovarian responsiveness to FSH and LH appears to have been attenuated. Analysis of follicular fluid revealed that GH concentrations were less in IGF-I-treated animals when compared with control animals, most likely because of the negative feedback effects of IGF-I on pituitary GH secretion. Further, the reduced responsiveness of the ovary to FSH and LH in IGF-I-infused animals appears to have been overcome by cotreatment with hGH. It must be noted, however, that the blood concentrations of GH in animals in our current study were supraphysiological, and it is not known whether similar results would have been observed with GH concentrations within the normal physiological range. As discussed previously (11, 12), humans with primary GH deficiency are fertile and respond to exogenous gonadotropin, indicating that neither GH nor IGF-I is essential for follicular development or menstrual cyclicity. To address a role for a direct effect of GH on the ovary would require a GH and IGF-I clamped animal. This in turn would require a nonhuman primate with a primary IGF-I deficiency, which does not exist.
In summary, the goal of this research was to determine whether IGF-I is able to augment ovarian function in otherwise normal monkeys (i.e. in the absence of insulin resistance or other metabolic disorders). Our findings indicate that, unlike previous in vitro studies, we found no evidence for a synergistic effect of IGF-I on gonadotropin-simulated ovarian steroidogenesis in vivo.
Acknowledgments
We thank Drs. Tony M. Plant and Stephen Franks for helpful comments.
Footnotes
This work was supported by NICHD/NIH through cooperative agreement (U54 HD 08610) as part of the Specialized Cooperative Centers Program in Reproduction Research.
Abbreviations: h, Human; IGFBP-3, IGF-binding protein-3.
Received July 24, 2002.
Accepted September 12, 2002.
References
This article has been cited by other articles:
 |
|
 |
|
  E. Diamanti-Kandarakis and C. Piperi Genetics of polycystic ovary syndrome: searching for the way out of the labyrinth Hum. Reprod. Update, November 1, 2005; 11(6): 631 - 643. [Abstract] [Full Text] [PDF]
|
|
|
|
|
|
S. Jonard and D. Dewailly The follicular excess in polycystic ovaries, due to intra-ovarian hyperandrogenism, may be the main culprit for the follicular arrest Hum. Reprod. Update, March 1, 2004; 10(2): 107 - 117. [Abstract] [Full Text] [PDF]
|
|
|
|
 |
|
 A. J. Zeleznik, L. Little-Ihrig, and S. Ramasawamy Administration of Dihydrotestosterone to Rhesus Monkeys Inhibits Gonadotropin-Stimulated Ovarian Steroidogenesis J. Clin. Endocrinol. Metab., February 1, 2004; 89(2): 860 - 866. [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 |
