| HOME | HELP | FEEDBACK | SUBSCRIPTIONS | ARCHIVE | SEARCH | TABLE OF CONTENTS |
| ||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||
Department of Obstetrics and Gynecology (M.B.S., J.M.L., P.E.P.), Oregon Health & Science University, Portland, Oregon 97239; and Division of Reproductive Sciences (S.M.B., T.A.M., R.L.S.), Oregon National Primate Research Center, Beaverton, Oregon 97006
Address all correspondence to: Phillip E. Patton, M.D., OHSU Fertility Consultants, Center for Health and Healing, CH10F, 3303 SW Bond Avenue, Portland, Oregon 97239-4501. E-mail: pattonp{at}ohsu.edu.
 |
Abstract |
|---|
 Top Abstract IntroductionSubjects and MethodsResultsDiscussionReferences |
|---|
Objective: The aim of this study was to determine whether there were differences in the dose- and time-dependent effects of insulin and IGFs on VEGF-A production by luteinized granulosa cells (LGCs) from women with and without PCOS.
Design and Setting: A prospective comparative experimental study was conducted at an institutional practice.
Patients: Patients included six PCOS and six non-PCOS women undergoing COS and IVF.
Interventions: Interventions included COS for IVF.
Main Outcome Measures: VEGF-A levels in culture media were collected daily for 3 d from LGCs after incubation with variable doses of insulin, IGF-I, and IGF-II in the presence and absence of LH.
Results: In both study groups, exposure to LH alone did not alter VEGF-A levels. However, insulin or IGF increased VEGF-A levels within 1 d and appeared to synergize with LH at 3 d. VEGF-A production by non-PCOS LGCs was more sensitive to IGF exposure, whereas PCOS cells were more sensitive to insulin. Although an increase in DNA content (P < 0.05) was noted in cultures of PCOS cells, progesterone levels were lower compared with non-PCOS LGCs.
Conclusion: Insulin and IGFs promote VEGF-A production in LGCs, but the response patterns are different when cells from PCOS and non-PCOS women are compared.
|
Introduction |
|---|
|
TopAbstractIntroductionSubjects and MethodsResultsDiscussionReferences |
|---|
LH/human chorionic gonadotropin (hCG) stimulates VEGF-A expression and production by luteinizing granulosa cells in both a dose- and time-dependent manner (5, 6). However, other regulators of VEGF, including insulin and the IGFs, can act synergistically with LH/hCG or act alone to stimulate VEGF production (7, 8). Although it is conceivable that the exposure of ovarian tissues to elevated levels of insulin and/or IGFs result in altered VEGF production, no studies to date have rigorously examined this hypothesis. Furthermore, little is known concerning the differential effects of these agents on VEGF production in vitro from diverse infertility populations [polycystic ovarian syndrome (PCOS) and non-PCOS women]. Therefore, we investigated whether there were differences in the dose- and time-dependent effects of insulin and IGFs in the presence and absence of LH on VEGF-A production by cultured luteinized granulosa cells (LGCs) from women with and without PCOS.
|
Subjects and Methods |
|---|
|
TopAbstractIntroductionSubjects and MethodsResultsDiscussionReferences |
|---|
A total of 12 women undergoing controlled ovarian stimulation (COS) for in vitro fertilization-embryo transfer (IVF-ET) were enrolled into the study. Six women had regular menstrual cycles with no evidence of PCOS or other endocrinopathies, whereas six others were diagnosed as having PCOS based on the European Society of Human Reproduction and Embryology-Rotterdam criteria (9). All six of the PCOS women were oligoovulatory (fewer than eight menstrual cycles per year), and five of the six had evidence of clinical hyperandrogenism. The Institutional Review Board of the Oregon Health & Science University approved the study and its design.
The COS protocol was described previously (10, 11). Briefly, COS was achieved using either a GnRH antagonist (250 µg/0.5 ml Antagon, twice daily; Organon, West Orange, NJ) or a long-acting GnRH agonist (0.5 mg Lupron, twice daily; TAP Pharmaceutical, Deerfield, IL) administered after a minimum of 2 wk of oral contraceptive treatment. FSH or FSH/human menopausal gonadotropin was then given (twice daily, three to six ampoules of 75 IU/d; Serono, Randolph, MA). Seven of the 12 women in the study received pretreatment with a GnRH agonist (five in the PCOS group and two in the non-PCOS group). There were no differences in serum or granulosa cell data from agonist vs. antagonist treatment protocols. Follicular response was monitored by serial pelvic ultrasonography and daily serum estradiol (E) measurements. When at least two follicles were more than 17 mm diameter, hCG (10,000 IU, im; Serono) was administered, and transvaginal ultrasound-directed follicle aspiration was performed 36 h later. After oocyte retrieval, follicular aspirates containing LGCs from each patient were pooled. Peripheral blood samples were drawn from the patients on the day of hCG administration, at the time of follicular aspiration, and on the day of pregnancy testing 12 d after ET. Serum was isolated after centrifugation and stored at –80 C until subsequent assays to determine the concentrations of E (picograms per milliliter) and progesterone (P, nanograms per milliliter).
Preparation and culture of LGCs
LGCs were isolated using density-gradient centrifugation in Percoll (Sigma Chemical Co., St. Louis, MO) as previously described (5). Approximately 2 x 106 to 19 x 106 cells were obtained from non-PCOS and PCOS patients. Cellular viability was assessed after incubation of a 10-µl cell suspension with 10 µl of Trypan blue as previously described (5) and ranged from 50–80% for both non-PCOS and PCOS groups. LGCs were plated at 40,000 cells per well on fibronectin-coated plates (Fisher Scientific, Pittsburgh, PA) in DMEM-Ham’s F-12 medium with 5 µg/ml transferrin (Sigma), 5 ng/ml selenium (Sigma), 10 µg/ml aprotinin (Sigma), and 25 µg/ml human low-density lipoprotein (Sigma). Cells were cultured in triplicate in the presence or absence of 100 ng/ml recombinant human LH (Serono Reproductive Biology Institute, Rockland, MA) and 0, 1, 10, or 100 ng/ml recombinant human IGF-I, IGF-II, or insulin (Sigma) at 37 C in a humidified 5% CO2/95% air environment. Culture medium was collected at 24, 48, and 72 h and frozen at 20 C until assay. DNA content was determined in each well after 3 d by previously described methods (12).
Immunoassays for measurement of VEGF and P
Medium collected daily was assayed for free VEGF-A and P. VEGF-A concentrations in culture media were determined using a human ELISA kit (Quantikine VEGF ELISA; R&D Systems, Minneapolis, MN) as described previously (5). The lower limit of detectability was 5 pg/ml. The inter- and intraassay coefficients of variation for VEGF were 9.5 and 15.1%, respectively. The concentration of P in the same media samples and serum concentrations of E and P were determined by specific electrochemoluminescent assay using a Roche Elecsys 2010 analyzer by the Endocrine Services Core Laboratory, Oregon National Primate Research Center (13). Hormone concentrations were validated against previous RIAs in this laboratory (14, 15). Inter -and intraassay coefficients of variation for the steroid assays were 7.2 and 11.6% for E and 9.7 and 8.9% for P, respectively.
Statistical analysis
The data were normalized to DNA content and analyzed using a linear regression model, a mixed-effects model, and adjusted using the Bonferroni method for multiple comparisons. Data are presented as means ± SEM for experimental studies and means ± SD for demographic data (Table 1
). A Fisher exact test was used to determine differences in pregnancy rates. Using SigmaStat (Jandel Corporation, San Rafael, CA) for statistical analysis, a two-way repeated ANOVA, with a randomized block design, was used to estimate the statistical difference between VEGF-A or P concentration in the dose range. Student’s t test was used to determine the statistical difference between control and PCOS treatments for serum (E and P), culture media (VEGF-A and P), and demographic data. A value of P < 0.05 was considered significant.
|
|
Results |
|---|
|
TopAbstractIntroductionSubjects and MethodsResultsDiscussionReferences |
|---|
The study groups, non-PCOS and PCOS women undergoing COS, were comparable in regards to demographic data (Table 1
). With the exception of a higher body mass index (BMI) (P < 0.05) and lower amount of administered gonadotropins used (P < 0.05), there were no differences in any of the other variables examined. Although there was a trend toward increased pregnancy rates in the non-PCOS group (66 vs. 40%), this difference was not statistically significant (P = 1.0).
On the days of hCG administration and oocyte retrieval, circulating concentrations of E were significantly higher (P < 0.05) in PCOS patients, but no difference was noted on the day of pregnancy test (Fig. 1A). Conversely, no significant difference was detected for P levels between non-PCOS and PCOS patients on the day of hCG or oocyte retrieval (Fig. 1B). However, on the day of the pregnancy test, non-PCOS women had significantly (P < 0.05) higher P levels when compared with PCOS women (Fig. 1B).
|
The levels of VEGF-A attained after 1 d of culture in control conditions were similar to those observed after 3 d of culture (Fig. 2). Although LH exposure did not alter VEGF levels at either 1 or 3 d, addition of insulin or IGFs (IGF-II; Fig. 2; others not shown) increased VEGF levels within 1 d and appeared to synergize with LH at 3 d. Therefore, all subsequent data will be presented for 3 d of culture. After 3 d, no differences in DNA content were noted in cultures of non-PCOS cells regardless of treatment with insulin, IGF-I, and IGF-II or concentration (Fig. 3). However, when the PCOS cells were analyzed, an increase in DNA content (P < 0.05) was noted in the presence of insulin at 10–100 ng/ml and IGF-II at 100 ng/ml. Thus, all subsequent data are normalized to DNA content (0.1 absorbance) per well.
|
|
). In contrast, cells from non-PCOS patients responded to a lower concentration of IGF-I (10–100 ng/ml), whereas cells from PCOS patients responded only to the highest dose (100 ng/ml; Fig. 4B). Cells from non-PCOS and PCOS patients were relatively insensitive to IGF-II, with only a modest increase in VEGF-A production at the highest dose tested (100 ng/ml; Fig. 4C).
|
). When cells from either non-PCOS or PCOS patients were tested, LH alone did not increase VEGF-A levels during up to 3 d of culture (see also Fig. 2). But in the non-PCOS population, the greatest effects on VEGF-A production were found in the presence of LH across all treatment arms. The most sensitive response was found when the non-PCOS LGCs were incubated in the presence of IGF-I or IGF-II. VEGF-A production significantly increased in the presence of as little as 1 ng IGF-I/ml or 10 ng IGF-II/ml plus LH, compared with IGFs alone. In contrast, the dose-response curves for insulin with and without LH were similar, although VEGF-A levels were greater at 100 ng insulin/ml plus LH. However, in the PCOS population, the presence of LH influenced only the insulin treatment arm. As little as 1 ng insulin/ml increased VEGF-A levels during LH exposure, and VEGF-A concentrations were highest at 10–100 ng insulin/ml plus LH. In this same population, the addition of LH did not alter the dose-response curve or VEGF-A levels after IGF-I or IGF-II stimulation.
|
). In the non-PCOS group, the exposure to LH alone typically increased P levels above control. In the PCOS group, LH alone did not increase P levels within any treatment group. In the absence of LH, exposure to insulin or the IGFs at any dose (100 ng/ml; Fig. 6) did not increase P levels with LGCs from either study population. Likewise, in the presence of LH, P levels after exposure to either insulin or IGFs were not different from that in LH alone.
|
|
Discussion |
|---|
|
TopAbstractIntroductionSubjects and MethodsResultsDiscussionReferences |
|---|
In contrast to the insensitivity to LH, we observed a dose-dependent increase in VEGF-A after incubation of LGCs with IGF-I, IGF-II, or insulin. LGCs were relatively more sensitive to the effects of IGF-I when compared with IGF-II, consistent with previous studies (17). The addition of insulin or IGFs increased VEGF levels within 1 d and appeared to synergize with LH at 3 d of culture. These results support and extend earlier observations in nonhuman primates showing a stimulatory and synergistic effect of LH and IGFs/insulin on VEGF secretion by nonluteinized as well as luteinized granulosa cells from macaque follicles (7). Collectively, the data argue that insulin/growth factors act synergistically with LH to promote VEGF production and supports the hypothesis that circulating IGFs and insulin contribute to the regulation of VEGF during the luteal phase (7).
We found an increase in DNA content at 72-h exposure to insulin or IGF for PCOS LGCs but not for non-PCOS cells. Because DNA content is unchanged in cultured LGCs obtained from COS protocols in the macaque (7, 18), the finding of differences in DNA content was unexpected. Increased cell survival after plating and/or proliferation during culture are two potential explanations. During the follicular phase, granulosa cells proliferate, but mitosis is markedly inhibited (but not abolished) after exposure to the hCG/LH ovulatory bolus (19). Recent work suggests that less mature antral follicles in PCOS women undergo incomplete or impaired luteinization after an hCG bolus, although intrafollicular P levels are normal in follicles containing mature oocytes (20). It is possible, therefore, that within some PCOS follicles after an ovulatory bolus, undifferentiated and nonluteinized granulosa cells persist and maintain the capacity to proliferate. Alternatively, cellular proliferation could be maintained under conditions where cells secrete low levels of P, as found in our study. In the macaque, steroid depletion increases cyclin B1 mRNA, which is reversible by progestin replacement, suggesting that intrafollicular steroids, notably P, can exhibit antiproliferative actions during the luteal phase (19).
We also found that VEGF-A levels produced by non-PCOS LGCs were more sensitive to IGF exposure, whereas PCOS cells were more sensitive to insulin. Previous work has shown that insulin enhances FSH-stimulated and LH-stimulated steroid production by PCOS granulosa and theca cells, respectively (21, 22). In contrast, Agrawal et al. (8) reported that PCOS LGCs exposed to LH or the combination of hCG and insulin stimulated VEGF production, whereas insulin alone had no effect; however, their diagnosis of PCOS was based on PCO morphology, plus mean BMIs and testosterone levels were no different compared with a non-PCOS control group. Thus, it is possible that response patterns by LGCs from their population may be different when compared with cells obtained in the current study from an obese, hyperandrogenic, hyperinsulinemic PCOS population. Our data support and extend the concept that insulin plus gonadotropin augments not only steroidogenic (23) but also nonsteroidogenic responses in ovarian cells from PCOS patients compared with those in non-PCOS women.
Finally, we observed that in the presence or absence of LH, P levels produced by PCOS LGCs were lower than non-PCOS LGCs. Our results substantiate previous studies showing that P levels are lower and androgen levels higher within PCOS follicles from women undergoing IVF (20) and that P production in PCOS LGCs is unchanged after hCG exposure (24). Whether P modulates VEGF secretion is unknown. In differentiated human luteal cells, hCG stimulates P production without changing VEGF mRNA or protein levels (17), presumably because VEGF-producing cells are maximally stimulated and thus refractory to repeated gonadotropin exposure. Alternatively, P could act as an inhibitor of VEGF secretion during the luteal phase (17). Although it is interesting to speculate the effect of low P on VEGF secretory activity, particularly in the context of PCOS, blockade of steroid synthesis during the periovulatory period does not alter VEGF production by LGCs in the macaque follicle (18).
Another factor that may influence LGC function, e.g. P production, is the degree of adiposity or BMI, which was higher in PCOS women. Previous observations that intrafollicular insulin levels correlate directly with the degree of adiposity suggest that BMI can alter granulosa cell luteinization (25). Additionally, serum hCG levels correlate negatively with BMI (26), a metabolic effect that could potentially also alter LGC steroid production, especially in obese PCOS women. Differences in FSH dose requirements between the two study groups during COS could also contribute to our findings (27). Based on these works, impaired luteinization and P production by PCOS granulosa cells could be partially attributed to BMI-related and FSH dose effects coupled with reduced systemic hCG levels.
The use of pooled follicular aspirates, which contain a heterogeneous population of LGCs from multiple antral follicles, is a recognized limitation of this and other studies. It is possible that PCOS follicle aspirates could contain a higher percentage of immature cells; however, the distribution of large, medium, and small follicles as measured by ultrasound just before follicular aspiration was similar between PCOS and non-PCOS women. Furthermore, there were no differences in the percentage of mature (Metaphase II) oocytes obtained at oocyte retrieval between the two study groups. Although in general, aspirates from small follicles are relatively acellular, it is still possible that small follicles from PCOS patients contain more undifferentiated elements compared with small follicles from non-PCOS patients.
OHSS is a potential life-threatening condition associated with an exaggerated response to gonadotropins, multiple preovulatory follicles, and PCOS. Several lines of evidence indicate that excessive VEGF production is implicated in the pathogenesis of this disorder (4, 6, 11, 28). We reported previously an exaggerated VEGF response during COS cycles, which exceeded the capacity of its circulating binding proteins, could predispose women to the risk of OHSS (4). In the current study, we observed not only that VEGF secretion in PCOS LGCs was responsive to insulin but also that VEGF secretion was augmented by the coincubation of insulin and LH. Based on these observations, the increased risk for OHSS in PCOS women undergoing COS could be partially attributed to the effects of chronically elevated insulin and LH levels and the effects of an hCG bolus during COS cycles. Whether the risk of OHSS in PCOS women can be reduced with therapies targeted at minimizing the effects of underlying metabolic (hyperinsulinemia) (29) and endocrine (LH hypersecretion) alterations (30) is uncertain but holds promise.
In summary, both insulin and IGFs promote VEGF-A production by human LGCs, which appears to synergize with LH exposure. However, LGCs from PCOS women are more sensitive to insulin, whereas cells from non-PCOS patients are more sensitive to IGFs. The VEGF-A response of PCOS LGCs after exposure to insulin and LH could play a role in the increased risk of OHSS in PCOS women. In contrast to non-PCOS cells, PCOS cells may undergo impaired luteinization after an hCG bolus resulting in cellular aspirates with diminished P production and sustained capacity to proliferate. Whether these response patterns reflect inherent differences in maturing follicles or occur as a result of COS treatment remains to be established.
|
Acknowledgments |
|---|
|
Footnotes |
|---|
Disclosure Summary: All authors have nothing to disclose.
First Published Online May 8, 2007
Abbreviations: BMI, Body mass index; COS, controlled ovarian stimulation; E, estradiol; hCG, human chorionic gonadotropin; IVF-ET, in vitro fertilization-embryo transfer; LGC, luteinized granulosa cell; OHSS, ovarian hyperstimulation syndrome; P, progesterone; PCOS, polycystic ovarian syndrome; VEGF-A, vascular endothelial growth factor A.
Received December 21, 2006.
Accepted April 30, 2007.
|
References |
|---|
|
TopAbstractIntroductionSubjects and MethodsResultsDiscussionReferences |
|---|
This article has been cited by other articles:
 |
|
 |
|
  S.-U. Chen, C.-H. Chou, H. Lee, C.-H. Ho, C.-W. Lin, and Y.-S. Yang Lysophosphatidic Acid Up-Regulates Expression of Interleukin-8 and -6 in Granulosa-Lutein Cells through Its Receptors and Nuclear Factor-{kappa}B Dependent Pathways: Implications for Angiogenesis of Corpus Luteum and Ovarian Hyperstimulation Syndrome J. Clin. Endocrinol. Metab., March 1, 2008; 93(3): 935 - 943. [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 |
