| HOME | HELP | FEEDBACK | SUBSCRIPTIONS | ARCHIVE | SEARCH | TABLE OF CONTENTS |
Original Studies |
Division of Reproductive Biology and Medicine (D.R.S.), Department of Obstetrics and Gynecology; and California Regional Primate Research Center (C.A.V.), University of California Davis, Davis, California 95616
Address all correspondence and requests for reprints to: Dennis R. Stewart, Medicine:Reproductive Biology, Suber House, University of California Davis, Davis, California 95616.
 |
Abstract |
|---|
 Top Abstract IntroductionMaterials and MethodsResultsDiscussionReferences |
|---|
This system was used to observe the endocrine responses of the granulosa lutein cells to three different protocols of CG administration designed to mimic the normal luteal phase, early pregnancy, and early pregnancy followed by pregnancy loss. The normal luteal phase was simulated by a constant baseline (0.02 IU/mL) CG model to simulate a nonconceptive cycle (baseline). The second model was baseline CG until day 8 of culture, followed by daily doubling from days 9–17 to simulate an early pregnancy (rescue-plateau). CG concentrations were then held constant from days 17–20 (5.12 IU/mL). A third model (rescue-drop) was used that was identical to the early pregnancy model except that on day 17 CG was returned to baseline concentrations (0.02 IU/mL) to simulate an early pregnancy loss.
Baseline CG stimulation resulted in profiles of estrogen, progesterone, and relaxin secretion in culture that were closely related to secretory profiles previously reported in serum during the nonconceptive luteal phase. The timing of appearance of relaxin secretion and later declines in steroid and relaxin secretion paralleled that observed in serum. In the CG rescue protocols, ovarian steroids rose in response to daily doubling of CG and fell when CG either plateaued or fell. Relaxin did not show an increase in response to increasing CG, but its secretion did not drop when CG concentrations plateaued or dropped. This cell culture system model mimics the profile of ovarian steroids and relaxin seen in serum during the nonconceptive luteal phase, although the relative magnitude of the hormones was not the same as seen in vivo. It was also used to investigate responses to luteal rescue protocols designed to simulate early pregnancy and pregnancy loss. This culture system may be useful to study differences in endocrine response in granulosa cells collected from different patients and to provide information of clinical relevance. This culture system provides a model to study luteal function and its response to different protocols of luteal rescue and thus may provide insight into early pregnancy and pregnancy loss.
|
Introduction |
|---|
|
TopAbstractIntroductionMaterials and MethodsResultsDiscussionReferences |
|---|
Most in vitro culture systems for human GCs have used cells plated on plastic. However, the use of a proteinaceous matrix to keep cells from being in contact with the plastic has been used by some investigators (7, 8) to provide a more physiological environment. Human GCs grown on an extracellular matrix have been shown to have a higher production of progesterone (9, 10) and estradiol (10, 11), form more gap junctions with neighboring cells (9), have more LH/CG receptors (9), and have a greater cAMP response to CG (12) than cells grown on plastic. It has been suggested that a cell adhesion receptor (an integrin) and laminin and fibronectin play important roles in the differentiation of GCs to luteal cells in the rat (8). Laminin and fibronectin are major glycoprotein components of the extracellular matrix (Matrigel matrix, Becton-Dickinson Labware, Franklin Lakes, NJ). We have utilized this matrix to try to develop a granulosa lutein cell culture system that produces relaxin and ovarian steroids in a physiological manner.
CG is administered to the GCs to cause luteinization and maintain functionality of the granulosa lutein cells during culture. The common CG dose of 1 IU/mL appears to date from early studies when human IVF clinics first became a source of GCs, and the potential of this model was developing. However, the literature does not contain a systematic study of responses to lower CG doses, and the use of 1 IU/mL may represent more CG than necessary to simulate a normal luteal phase in which gonadotropin support is low.
The human granulosa lutein cell culture system used in this study combined the use of an extracellular matrix, lower concentrations of CG than generally used by other investigators, and a sensitive relaxin assay. This system was used to simulate a normal luteal phase profile of steroids and relaxin. These profiles can be altered by the use of CG protocols to simulate early pregnancy and early pregnancy loss. This granulosa lutein cell model could prove useful to study patterns of early pregnancy loss by the design of different CG stimulation protocols that mimic patterns of CG observed in vivo. It could also be useful for analysis of GCs from assisted reproduction patients to determine whether abnormalities exist in their normal and CG-stimulated endocrine production.
|
Materials and Methods |
|---|
|
TopAbstractIntroductionMaterials and MethodsResultsDiscussionReferences |
|---|
MEM (Gibco, Grand Island, NY) was modified with the following additions: sodium bicarbonate, 4.4 mg/100 mL MEM (Sigma, St. Louis, MO); fungizone, 1 mL/100 mL (Gibco); penicillin G, 6 mg/100 mL (Sigma); streptomycin sulfate, 6 mg/100 mL (Sigma), and 10% FCS (Hyclone, Logan, UT). Media was filtered through a 0.22-µm sterile syringe filter (Fisher, Santa Clara, CA) and equilibrated at 37 C and 5% CO2 in air before use. CG (Pregnyl; Organon, West Orange, NJ) was added to the culture media in amounts as described below. Extracellular matrix was applied to culture dishes on the same day cells were collected according to the manufacturer’s directions.
Cell collection
Human GCs were obtained by ultrasound-guided follicle aspiration from women receiving assisted reproduction treatment at Pacific Fertility Center (Sacramento, CA). The cells were a by-product of the IVF/embryo transfer (ET) procedure and are normally discarded. They were provided to this study as coded samples with the identities of the women unavailable, so this research was exempt from review by the university Human Subjects Review Committee (HSRC) under Federal exemption category number 4. Exemption from full HSRC review was approved by the HSRC coordinator. The patients received varying doses of Metrodin (Serono Laboratories, Randolph, MA) and Pergonal (Serono) and received 10,000 CG 36 h before follicular aspiration. Individual follicles were not distinguished, because all GCs from an individual were pooled. Cells from different subjects were not pooled.
Culture preparation
Cells were prepared by initial centrifugation followed by layering onto a 40% Percoll (Sigma). The GC layer was washed twice with 5–10 mL fresh MEM and centrifuged for 10 min at 300 x g. The supernatant was discarded, and the pellet was resuspended in 2–4 mL MEM. GCs were filtered through an 89-µm polyester filter (Spectra/Mesh, Spectrum Medical, Laguna Hills, CA) just before being counted and plated. Cells were brought to a final concentration of 1 x 105 cells/mL in MEM and plated on 4-well plates (1.9 cm diameter wells) at 5 x 104 cells/well (0.5 mL). Cells had attached after 24 h, and media was changed to remove remaining debris. Media was changed daily in all experiments and stored frozen until assay for hormone concentrations. Because multiple wells were obtained from each subject, different wells from each subject were used for the various treatment protocols (viability and either CG dose response or CG stimulation).
Verification of viability and cell number during culture
Estimates of viability were obtained using trypan blue (0.4%, Gibco) exclusion on an Olympus CK2 microscope (Olympus Optical, Tokyo, Japan) at 200 times magnification. One milliliter Matrisperse (Fisher) was added to each well to free cells from the Matrigel, and cells were scraped into a centrifuge tube. The well was rinsed with an additional 1 mL Matrisperse, which was placed in the tube and kept on ice for 1 h. Cells were centrifuged for 5 min at 500 x g, and the pellet was resuspended in 100 µL PBS. Cells were counted on a hemacytometer.
CG dose response
To determine the lowest dose of CG needed to effectively maintain luteal support in terms of steroid and relaxin secretion, GCs collected from six patients were plated and cultured with seven different constant concentrations of CG. The doses of CG ranged from 0.002–0.2 IU/mL culture fluid, and these doses were maintained throughout the 20-day culture period. Estradiol, progesterone, and relaxin concentrations were determined in the conditioned media.
CG stimulation protocols
Three protocols of CG administration to the culture media were used to simulate three different luteal phase events using replicate wells from 10 subjects. The first protocol (baseline) was a constant baseline dose of CG to simulate a normal nonconceptive luteal phase. A baseline concentration of 0.02 IU/mL was selected from the dose-response study based on its ability to maintain physiological profiles of steroid and relaxin secretion. CG concentrations were held at baseline CG for each of the 20 days of culture. A second protocol (rescue-plateau) was used to simulate early pregnancy during the middle of the culture period. CG was maintained at baseline concentrations (0.02 IU/mL) for days 1–8 of culture and were then doubled each day until day 16 of culture. On days 17–20 of culture, CG concentrations were maintained at the highest CG concentrations (5.12 IU/mL) to determine the effect of plateaued CG concentrations for comparison with the third protocol. The third protocol (rescue-drop) was designed to simulate an early pregnancy followed by pregnancy loss. In this protocol, CG concentrations were identical to the previous protocol until day 16. On day 17 and thereafter CG concentrations were returned to baseline (0.02 IU/mL).
Assays
Estradiol and progesterone were measured by commercial kits (Diagnostic Products Corp., Los Angeles, Ca) as previously reported (13). Relaxin was measured by an enzyme immunoassay as previously reported for serum relaxin (5). The assay was modified by dilution of human relaxin using culture fluid instead of human serum for preparation of standards.
Data analysis
To normalize the endocrine data, the values were converted to the common logarithm for statistical analysis and averaging. Data were converted to arithmetic scale for graphing (geometric mean). Hormone values for the three CG protocols were compared by two-way repeated measures ANOVA, and significance was followed up by Student-Newman-Keuls multiple comparisons test using a 0.05 significance level.
|
Results |
|---|
|
TopAbstractIntroductionMaterials and MethodsResultsDiscussionReferences |
|---|
Cell number was verified on day 16 of culture (n = 6 patients). Mean cell counts per well were 50,000 ± 10,488 and 52,000 ± 12,000 for CG baseline and CG rescue protocols, respectively. Day 16 viability ranged from 95–98%.
Endocrine responses
In analysis of the results of CG dose response and CG stimulation protocols, cells from different patients resulted in two distinct patterns of relaxin response (not shown). The first, termed a nonresponder, was characterized low relaxin production. The second pattern, termed a responder, was characterized by robust relaxin concentrations on day 10 of culture. A responder was defined as a patient with >150 pg/mL relaxin on day 10 of culture using the standard baseline dose of CG (0.02 IU/mL). There was a clear distinction between responders and nonresponders, because the highest relaxin concentration from a nonresponder on day 10 was 46 pg/mL and the lowest relaxin concentration from a responder was 464 pg/mL. Four of six patients in the CG dose-response protocols and six of ten patients in the CG stimulation protocols were responders, and only the results from these were used for further analysis.
CG dose response
The cells showed a dose-response relationship to CG in terms of
steroid and relaxin production with the CG concentrations used in this
study (Fig. 1
). A dose of CG of 0.02
IU/mL was chosen as the standard baseline dose of CG. This dose gave
adequate steroid and relaxin production as well as a timely profile of
relaxin secretion. Higher amounts of CG tended to give prolonged
relaxin secretion past day 15 of culture, and thus might indicate that
the CG concentrations were higher than desired to simulate a
nonconceptive cycle.
|
Estradiol concentrations produced by the baseline CG protocol
cells were extremely high on the first day of culture (data not shown).
Concentrations significantly declined on days 3 and 5 of culture (Fig. 2), increased from days 5–11, and then
declined between days 11–19 (P < 0.05). Progesterone
secretion significantly increased on alternate days from days 1–7 of
culture. Progesterone then declined between days 7–11, with a
significant drop between days 15–17 of culture. Relaxin secretion was
first significantly elevated on day 5 of culture and then showed
significant daily increases through day 8. Relaxin levels plateaued on
day 9 and remained elevated until day 15 of culture. There was a
significant fall in relaxin secretion from days 15–18 and
thereafter.
|
) were similar to those
from baseline CG cells before the increase in CG. During the period of
culture that the CG was increasing, estradiol concentrations were
prevented from dropping. There was a significant increase in estradiol
from days 9–11 and 13–15 of culture. Estradiol concentrations began a
significant drop from days 15–19 of culture, during the time CG
concentrations were plateaued. Progesterone concentrations followed a
nearly identical profile to estradiol concentrations in response to the
increase in CG. There was a significant increase in progesterone from
days 9–11 and 13–15 of culture. Plateaued CG resulted in a
significant drop from days 15–17 and 19 of culture. Before day 16 of
culture, relaxin concentrations in the rescue-plateau protocol were not
different from the baseline protocol. However, the increasing
concentrations of CG prevented the decline of relaxin, because days
16–20 were not different from day 15 of culture.
|
),
estradiol, progesterone, and relaxin concentrations and profiles were
not different from the rescue-plateau protocol.
|
. The divergence of
estradiol concentrations in the two CG rescue protocols from the
concentrations in the baseline protocol was significant on day 13 and
thereafter. A difference was noted for progesterone concentrations on
day 11 from rescue protocols compared with the baseline protocol, but
this difference was not significant until day 13 of culture. The
difference in relaxin concentrations between CG baseline and the two
rescue protocols was significantly different after day 16.
|
|
Discussion |
|---|
|
TopAbstractIntroductionMaterials and MethodsResultsDiscussionReferences |
|---|
) are
similar in shape and timing to our in vivo profiles of
steroids (14) and relaxin (4, 6) from women with natural cycles sampled
on a daily basis. This is the first report of granulosa lutein cell
culture in which relaxin secretion mimics that observed in the normal
luteal phase, and suggests that this is a useful model for the study of
the control of luteal function. Although one other group has detected
relaxin secretion in culture as early as was observed in this study
(2), they only reported data for 9 days of culture, so long-term
profiles were not provided. The absence of the appearance of relaxin
secretion before day 5 of culture is consistent with the absence of
circulating relaxin early in the luteal phase (4) and the absence of
relaxin messenger RNA in the early luteal phase corpus luteum (15). The
reason that relaxin is not immediately produced from the beginning of
culture or early in the luteal phase of the cycle is not known. The
finding of a similar delay in culture compared with that observed
in vivo suggests that this is an inherent feature of
granulosa lutein cells and not caused by the influence of factors
arising outside the ovary. Under these culture conditions, two types of endocrine patterns were observed, one termed nonresponder because of an absence or minimal relaxin secretion and the other termed a responder because of a more robust relaxin response. Although steroid responses tended to be lower in nonresponders, they were not as strikingly different as relaxin concentrations, which were well separated in these two groups. It is not known whether a larger number of subjects would give a continuum in terms of relaxin secretion or whether these responses represent two distinct groups of subjects. Mayerhofer et al. (2) did not report on variations in relaxin secretion between patients, so it is not known whether this was observed in their system. Gagliardi et al. (1) reported a wide variation in the days of culture before relaxin was detected (10–22 days), but once secretion began 9 of 10 patients had relaxin production >200 pg/mL. The reason we saw 37% nonresponders (6/16) in the CG dose-response and CG stimulation protocols may because of the much smaller basal amounts of CG used in this culture system. We are continuing to collect information from nonresponders and are actively investigating the possible reasons for differences in relaxin secretion of GCs collected from different patients. It is speculated that endocrine production in vitro may be related to in vivo endocrine production. We have shown that serum luteal phase relaxin concentrations are highly variable between subjects, more so than progesterone concentrations (16). It was also found that relaxin concentrations were better correlated with the stage of the endometrial biopsy than progesterone (16), so relaxin concentrations may have physiological relevance. We are currently investigating the relationship of the profile of endocrine production in culture with serum values of these hormones in the same cycle from which cells were collected.
The decline in endocrine function during the latter days of culture is similar to the decline in luteal function late in the cycle. This may indicate an inherent pattern of endocrine secretion in the granulosa lutein cells, because this secretion pattern was in response to constant CG concentrations. Because cell number and viability were not changing, the change in endocrine concentrations must reflect different rates of endocrine production and secretion. The cell number did not appear to change during the 20 days of culture for human granulosa lutein cells grown on Matrigel. This is unlike GCs grown on plastic, which can double after 3 days in culture (17) or multiply 2- to 7-fold during a 24-day culture (3). It is possible that Matrigel contains some growth inhibitory factors that would be absent for cells grown on plastic, but if these factors were soluble they would be rapidly removed by the daily changes in culture fluid. Viability remained high throughout the culture period and thus did not appear to be the reason for differences in hormone production in response to CG.
There were at least two major differences between steroid and relaxin secretion in response to CG that were observed in these cell cultures. The first difference was that daily doubling of CG concentrations resulted in significantly increased steroid but not relaxin concentrations. The only significant effect of increasing CG on relaxin concentrations was prolonged secretion. This is different from the response of relaxin to CG in vivo, in which relaxin increases rapidly in parallel with trophoblastic CG in early pregnancy (5) or to the administration of exogenous CG in the nonconceptive luteal phase (18). Others have observed significant increases in relaxin in culture in response to an increase in CG, but the dose was 100 IU/mL (2), which probably represents a nonphysiological increase. It may be that the increase in CG used in this study was too gradual to give a significant increase in the amount of relaxin secreted. There appeared to be a small, although nonsignificant, increase in relaxin secretion in response to CG. A more vigorous rise in CG may be required to enhance relaxin secretion, and we are experimenting with different gradients of CG to test this. Alternately, there may be other factors in early pregnancy, in addition to CG, that also stimulate relaxin secretion but are absent from this culture system.
The second difference between granulosa lutein cell production of steroids and relaxin was in response to plateaued or dropping CG concentrations. A halt to daily doubling of CG, either in the rescue-plateau or rescue-drop protocol, resulted in an immediate fall in both estradiol and progesterone concentrations. The profile of steroids was contrasted by that of relaxin, which remained elevated with either a plateau or a drop in gonadotropin support. Thus, the granulosa lutein cells appear to require continually increasing CG to maintain steroid secretion, whereas they only require the prior elevation of CG above baseline to maintain enhanced relaxin secretion. The fall in steroid concentrations, but not relaxin, in response to plateaued CG concentrations is similar to profiles of circulating steroids and relaxin observed in women with early pregnancy loss (5). In cases of early pregnancy loss, CG can be observed to plateau and then fall. As observed in vitro, circulating steroids began to fall immediately as soon as CG concentrations plateaued, whereas relaxin concentrations remained elevated as long as CG was present in circulation (5). The differences between the steroid and relaxin response to CG could have several causes. There may be a differential sensitivity of the relaxin and steroid synthetic machinery to CG stimulation. Although the initial pathways for CG response are through the CG receptor and second-messenger systems, later events in their stimulation may have different regulation and responsiveness. Alternately, different granulosa lutein cells may have different sensitivities to CG stimulation and differential production of steroids and relaxin. It is possible that some cells produce steroids, whereas others produce relaxin, each with a different responsiveness to CG.
This GC culture system may be useful for the study of the cell types that produce relaxin and ovarian steroids and the control of endocrine secretion. This system produces steroid and protein markers with timing and profiles similar to that seen in vivo and allows the application of CG rescue protocols to be explored. With the ability to rescue the cell culture, much as the corpus luteum of early pregnancy is rescued, this system provides the ability to study the effects of different CG protocols on luteal function. This system might also prove useful for the study of altered forms of CG, which have been implicated in early fetal loss (19).
|
Acknowledgments |
|---|
|
Footnotes |
|---|
Received December 5, 1996.
Revised February 11, 1997.
Revised June 9, 1997.
Accepted June 14, 1997.
|
References |
|---|
|
TopAbstractIntroductionMaterials and MethodsResultsDiscussionReferences |
|---|
This article has been cited by other articles:
 |
|
 |
|
  J. K. Nyholt de Prada, Y. S. Lee, K. E. Latham, C. L. Chaffin, and C. A. VandeVoort Role for cumulus cell-produced EGF-like ligands during primate oocyte maturation in vitro Am J Physiol Endocrinol Metab, May 1, 2009; 296(5): E1049 - E1058. [Abstract] [Full Text] [PDF]
|
|
|
|
 |
|
 V. Drouineaud, P. Sagot, C. Garrido, E. Logette, V. Deckert, P. Gambert, C. Jimenez, B. Staels, L. Lagrost, and D. Masson Inhibition of progesterone production in human luteinized granulosa cells treated with LXR agonists Mol. Hum. Reprod., June 1, 2007; 13(6): 373 - 379. [Abstract] [Full Text] [PDF]
|
|
|
|
 |
|
 K.N. Fru, M. Cherian-Shaw, M. Puttabyatappa, C.A. VandeVoort, and C.L. Chaffin Regulation of granulosa cell proliferation and EGF-like ligands during the periovulatory interval in monkeys Hum. Reprod., May 1, 2007; 22(5): 1247 - 1252. [Abstract] [Full Text] [PDF]
|
|
|
|
 |
|
 M. Cherian-Shaw, R. Das, C. A. VandeVoort, and C. L. Chaffin Regulation of Steroidogenesis by p53 in Macaque Granulosa Cells and H295R Human Adrenocortical Cells Endocrinology, December 1, 2004; 145(12): 5734 - 5744. [Abstract] [Full Text] [PDF]
|
|
|
|
|
|
C. L. Chaffin, R. S. Brogan, R. L. Stouffer, and C. A. VandeVoort Dynamics of Myc/Max/Mad Expression during Luteinization of Primate Granulosa Cells in Vitro: Association with Periovulatory Proliferation Endocrinology, April 1, 2003; 144(4): 1249 - 1256. [Abstract] [Full Text] [PDF]
|
|
|
|
|
|
F. M. Moran, C. A. VandeVoort, J. W. Overstreet, B. L. Lasley, and A. J. Conley Molecular Target of Endocrine Disruption in Human Luteinizing Granulosa Cells by 2,3,7,8-Tetrachlorodibenzo-p-Dioxin: Inhibition of Estradiol Secretion Due to Decreased 17{alpha}-Hydroxylase/17,20-Lyase Cytochrome P450 Expression Endocrinology, February 1, 2003; 144(2): 467 - 473. [Abstract] [Full Text] [PDF]
|
|
|
|
 |
|
 F.M. Moran, P. Lohstroh, C.A. VandeVoort, J. Chen, J.W. Overstreet, A.J. Conley, and B.L. Lasley Exogenous Steroid Substrate Modifies the Effect of 2,3,7,8-Tetrachlorodibenzo-p-Dioxin on Estradiol Production of Human Luteinized Granulosa Cells In Vitro Biol Reprod, January 1, 2003; 68(1): 244 - 251. [Abstract] [Full Text] [PDF]
|
|
|
|
|
|
F.M. Morán, A.J. Conley, C.J. Corbin, E. Enan, C. VandeVoort, J.W. Overstreet, and B.L. Lasley 2,3,7,8-Tetrachlorodibenzo-p-Dioxin Decreases Estradiol Production Without Altering the Enzyme Activity of Cytochrome P450 Aromatase of Human Luteinized Granulosa Cells In Vitro Biol Reprod, April 1, 2000; 62(4): 1102 - 1108. [Abstract] [Full Text]
|
|
|
|
|
|
C. A. VandeVoort, J. W. Overstreet, B. L. Lasley, and D. R. Stewart Effects of Progesterone Receptor Blockers on Human Granulosa-Luteal Cell Culture Secretion of Progesterone, Estradiol, and Relaxin Biol Reprod, January 1, 2000; 62(1): 200 - 205. [Abstract] [Full Text]
|
|
|
|
|
|
D. R. Stewart and C. A. VandeVoort Relaxin secretion by human granulosa cell culture is predictive of in-vitro fertilization–embryo transfer success Hum. Reprod., February 1, 1999; 14(2): 338 - 344. [Abstract] [Full Text] [PDF]
|
|
|
|
|
|
K. M. Ohleth and C. A. Bagnell Relaxin Secretion and Gene Expression in Porcine Granulosa and Theca Cells Are Stimulated during In Vitro Luteinization Biol Reprod, February 1, 1999; 60(2): 499 - 507. [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 |
