| HOME | HELP | FEEDBACK | SUBSCRIPTIONS | ARCHIVE | SEARCH | TABLE OF CONTENTS |
Women’s Health Research Institute Wyeth-Ayerst Research Radnor, Pennsylvania 19087
 |
ABSTRACT |
|---|
 TOP ABSTRACT INTRODUCTIONRESULTSDISCUSSIONMATERIALS AND METHODSREFERENCES |
|---|
|
INTRODUCTION |
|---|
|
TOPABSTRACTINTRODUCTIONRESULTSDISCUSSIONMATERIALS AND METHODSREFERENCES |
|---|
-subunit that is linked in a noncovalent manner
with a hormone-specific ß-subunit. Despite their similar structures
and sequence homologies, there is only marginal cross-reactivity among
the receptors (Ref. 2 and B. J. Arey, D. C. Deecher, E. S. Shen, and F.
J. López, unpublished observations). The presence of multiple
glycosylation sites on both subunits is one of the common structural
features of these hormones. FSH, for example, contains four such
glycosylation sites, two each on the - and ß-subunits (3). The
action of FSH in the gonad involves an initial binding event with a
specific FSH receptor. The human (h) FSH receptor (hFSH-R) is a large,
integral membrane protein comprised of multiple hydrophobic regions
consistent with the seven transmembrane-spanning domains identified in
G protein-coupled receptors (4). Indeed, the FSH receptor has been
shown to be positively coupled to Gs, resulting in
activation of adenylate cyclase and cAMP production (5, 6). In vivo, FSH is secreted into and maintained in serum as a series of isoforms of differing isoelectric points (pI) (for review see Refs. 7 and 8). These charge differences have been attributed to the presence of differently glycosylated forms of the hormone (9, 10). It is clear that the relative abundance of FSH isoforms in the circulation is dependent on the physiological status of the subject (11, 12, 13, 14). For example, in the rat, a shift in the pituitary content of FSH isoforms is observed on proestrus from those of lower pI to those of higher pI (12, 13). Because forms of FSH possessing greater pI values have a decreased sialic acid content, this represents a shift in circulating FSH from those species having a more complex glycosylated structure (more acidic) and lesser bioactivity to those of higher pI and greater bioactivity (15, 16). These observations suggest that feedback from the ovary signals an increase in bioactive FSH before ovulation. Therefore, the effect of a given stimulus on the gonad depends not only upon the amount of circulating gonadotropins but perhaps, more importantly, on the qualitative aspects of the stimulus such as the relative distribution of diverse hormone isoforms under various physiological conditions.
Several studies have provided evidence to indicate that FSH isoforms exhibit different properties, which, in turn, modify their biological activity. The data available suggest that hormone glycosylation is important for both serum half-life (17, 18, 19) and signal transduction (7, 20, 21, 22, 23). Indeed, it has been shown that chemically deglycosylated FSH acts as an antagonist at the FSH-R (24, 25). Because FSH binding to its receptor involves multiple interactions between the proteins, one possible explanation for these observations is that deglycosylation of the hormone decreases its intrinsic activity, producing an isoform that retains affinity for the receptor. In this respect, isoforms of FSH are known to contain differing degrees of glycosylation and also a range of receptor-binding characteristics (for review see Ref.8). Therefore, the deglycosylated hormone would act as a competitive blocker of the native hormone. Under this hypothesis then, the trophic signal to the gonad consists of a mix of agonists and antagonists with varying affinities and intrinsic activities that would determine gonadal response. Alternatively, it is plausible that certain FSH isoforms may bind to the FSH-R in a unique manner that induces ligand-specific conformational changes to the ligand-receptor complex. Distinct ligand-receptor complex conformations, in turn, could provide the molecular foundation for either activation or deactivation of alternative signaling pathways. These, therefore, may serve as a means to provide pleiotropic responses to complex signals consisting of multiple or varying glycosylated forms of FSH. In this study, we have evaluated whether different glycosylated forms of hFSH are capable of activating multiple different signal transduction pathways. For this purpose, dose-effect relationships for differently glycosylated forms of FSH were evaluated, not only in recombinant cell lines expressing the hFSH-R, but also in some physiologically relevant in vitro models of hFSH action.
|
RESULTS |
|---|
|
TOPABSTRACTINTRODUCTIONRESULTSDISCUSSIONMATERIALS AND METHODSREFERENCES |
|---|
/ß hFSH. Purified hFSH induced a
sigmoidal dose-dependent increase in cAMP accumulation that reached
maximal levels more than 300-fold greater than control (Fig. 1
, upper left panel), consistent with the
coupling of the FSH-R to Gs (5, 6). The estimated
ED50 for phFSH in 3D2 cells was 9.55 ng/ml. Similarly,
DeGly-phFSH induced a dose-dependent elevation in cAMP accumulation
apparent at doses ranging from 1–10,000 ng/ml (ED50 = 9.61
ng/ml; Fig. 1, upper right panel). However, unlike the
native hormone, DeGly-phFSH produced a biphasic (bell-shaped)
dose-response curve. Concentrations of DeGly-phFSH greater than 100
ng/ml were less efficacious in bioactivity when compared with that
elicited by this dose (Fig. 1, upper right panel). Maximal
stimulation of cAMP accumulation was approximately 20-fold greater than
control. Thus, DeGly-phFSH is approximately 10-fold less efficacious
than the native hormone. The IC50 for DeGly-phFSH, as
estimated using a seven-parameter logistic model, was 7.62 µg/ml.
|
, lower left panel). Maximal stimulation of cAMP
accumulation by BV-hFSH was approximately 200-fold greater than control
at a dose of 100 ng/ml. Higher doses of BV-hFSH were less bioactive,
leading to a bell-shaped dose-response curve. The ED50 for
the ascending phase of the BV-hFSH bioactivity was 1.37 ng/ml, whereas
the ID50 for the descending phase of the BV-hFSH curve was
532.59 ng/ml. Moreover, in a cell-free adenylate cyclase assay using
isolated 3D2 cell membranes, phFSH induced a sigmoidal dose-dependent
stimulation in adenylate cyclase activity (Table 1). In
contrast, increasing concentrations of BV-hFSH elicited a biphasic
effect on adenylate cyclase activity with distinct ascending and
descending phases to the dose-response relationship (Table 1).
Supernatants of Hi5 cells that were infected with baculovirus without
the coding region of hFSH did not increase cAMP accumulation as
compared with control in the whole-cell assay (data not shown).
Furthermore, purification of BV-hFSH from Hi5 cell supernatants by ion
exchange and subsequent size exclusion chromatography revealed that a
single peak was responsible for the biphasic dose-responses induced by
crude Hi5 cell supernatants (data not shown). In contrast, FSH
expressed in a mammalian cell line (CHO cells) induced a sigmoidal
dose-dependent increase in cAMP accumulation that was nearly
indistinguishable from that observed for phFSH (ED50 = 7.29
ng/ml; Fig. 1, lower right panel). Maximal induction of cAMP
production by CHO-hFSH was approximately 300-fold greater than that
observed in control wells.
|
). Purified hFSH competed for
[125I]hFSH binding with an ID50 = 869 ±
7 pM. Deglycosylated and BV-hFSH were slightly more potent
with an estimated ID50 of 395 ± 5 pM and
207 ± 7.8 pM, respectively. The slopes of competition
curves for DeGly-phFSH and BV-hFSH, however, were not parallel to that
of phFSH. Deglycosylated-phFSH and BV-hFSH competed with
[125I]hFSH for the hFSH-R with slope factors of 2.25 and
1.56, respectively. Purified hFSH competed for binding to the receptor
with a slope factor of 0.77. Thus, these data suggest that differently
glycosylated forms of FSH may bind to the receptor in a
hormone-specific fashion and, perhaps, elicit diverse conformations to
the ligand-receptor complex.
|
, top panel). In contrast, treatment of granulosa cells for
72 h with BV-hFSH resulted in a bell-shaped dose-response
relationship (Fig. 3, bottom panel) similar to that observed
in 3D2 cells. BV-hFSH stimulated estradiol secretion above control at
all but the lowest dose tested (Fig. 3, bottom panel).
Maximal stimulation of estradiol secretion was apparent at a dose of 10
ng/ml with an ED50 for the ascending portion of the BV-hFSH
dose-response curve of 0.97 ng/ml. Concentrations of BV-hFSH greater
than 10 ng/ml were less efficacious such that the highest dose of
BV-hFSH induced only a 1.4-fold increase in estradiol secretion over
control levels. The estimated ID50 for the descending phase
of the BV-hFSH dose-response curve was 295.44 ng/ml. A similar
dichotomy between phFSH and BV-hFSH in terms of the qualitative
dose-response profile on progesterone secretion was evident in an
adrenal cell line (Y1) stably expressing the hFSH-R (data not
shown).
|
, top
panel). However, when the ED50 of phFSH was incubated in
the presence of the same doses of BV-hFSH, an intriguing difference in
biological activity was observed (Fig. 4, middle panel). The
recombinant hormone had no effect on cAMP accumulation except at the
highest dose tested, where it inhibited cAMP accumulation induced by
the ED50 of phFSH by 40%. In contrast, when the
ED80 of phFSH (representing high receptor-transducer
activation) was incubated in the presence of varying concentrations of
BV-hFSH, the recombinant hormone was strictly inhibitory, with a
maximal inhibition of phFSH-induced cAMP accumulation of 60% at a dose
of 1.0 µg/ml (Fig. 4, bottom panel). Thus, when the
receptor is only partially activated (low concentrations of phFSH)
BV-hFSH displays agonistic activity, and accumulation of cAMP is
observed, which is consistent with the coupling of the FSH-R to
Gs. However, when the receptor is more fully activated
(higher concentrations of phFSH), BV-hFSH becomes antagonistic,
suggesting that at high receptor-transducer activation this hormone is
either capable of uncoupling the receptor from the stimulatory
signaling pathway and/or coupling it to an inhibitory cascade.
|
, top panel, solid
symbols). Low concentrations of BV-hFSH (1–10 ng/ml) had no
effect on cAMP accumulation as compared with cells treated with the
ED20 alone. However, higher concentrations of BV-hFSH led
to a decrease in basal cAMP accumulation (ID50 = 2.02
µg/ml). Cells not pretreated with CTX responded to BV-hFSH in a
biphasic manner in the presence of the ED20 of phFSH (Fig. 5, top panel, open symbols). Thus, CTX
pretreatment had no effect on the descending phase of the dose-response
curve, suggesting that this portion of the bioactivity of BV-hFSH is
independent of Gs-mediated signaling.
|
, bottom panel, solid symbols). In cells
not treated with PTX, BV-hFSH was inhibitory to cAMP accumulation
induced by the ED80 of phFSH (Fig. 5, bottom
panel, open symbols). As a confirmation of these
observations, we also tested the effect of PTX on BV-hFSH and phFSH
bioactivity in the absence of additional phFSH. As observed in Fig. 6, in the absence of PTX (solid symbols),
phFSH (top left panel) produced a sigmoidal dose-dependent
increase in cAMP accumulation, while BV-hFSH (top right
panel) induced its typical biphasic dose-response relationship.
However, PTX pretreatment completely abolished the descending phase of
BV-hFSH bioactivity. In the presence of PTX (Fig. 6, open
symbols), BV-hFSH (bottom right panel) induced a
sigmoidal dose-dependent increase in cAMP accumulation that was nearly
identical to that of phFSH (bottom left panel).
Interestingly, PTX treatment decreased the efficacy of both phFSH and
BV-hFSH, as well as forskolin (data not shown), to increase cAMP levels
approximately 6-fold when compared with control cells. Taken together,
these data suggest that the biphasic activity of BV-hFSH is mediated
through recruitment of differing G proteins.
|
|
DISCUSSION |
|---|
|
TOPABSTRACTINTRODUCTIONRESULTSDISCUSSIONMATERIALS AND METHODSREFERENCES |
|---|
Many investigators have shown the importance of secondary protein processing for bioactivity in several homeostatic systems (7, 24, 25). Glycosylation patterns found in FSH and the FSH-R are likely to play a significant role in ligand binding and possibly G protein coupling. Under physiological conditions, FSH is found in serum of many species as a myriad of different isoforms with differing pI values (7). The differences in pI have been shown to be primarily due to secretion of alternatively glycosylated forms of the hormone (10, 27). Therefore, the observed biological effects of FSH could be the result of a highly complex interaction between many different factors. For example, FSH bioactivity could depend upon the number of receptors present on the cell surface, the concentration of FSH in the serum, as well as the ratio of the isoforms of the hormone secreted. However, these factors would transduce signals solely as positive inputs to the target organ, since the FSH-R is thought to be coupled to a stimulatory G protein pathway (5, 6). An alternative mechanism to provide plasticity to the responsiveness of the target organ could conceivably involve coupling of the FSH-R with inhibitory transduction pathways (i.e. G proteins different than Gs). If such a mechanism operates within the FSH-R-transducer system, then the ability of the FSH-R to couple to different G proteins provides a means to respond in a pleiotropic manner to a complex ligand, a mix of multiple isoforms of FSH.
In our studies, we have taken advantage of chemically deglycosylated hFSH and the altered secondary processing of newly synthesized proteins in insect cells (26) to produce alternatively glycosylated forms of hFSH. Studies of the expression of other proteins have shown that insect cells have the ability to produce small truncated sugar side chains in place of the more complex oligosaccharide structures produced by cells of higher organisms (26). Using this paradigm, we demonstrate that such alterations in glycosylation of FSH can have profound effects on its biological activity. Whereas phFSH and mammalian cell-expressed hFSH (CHO-hFSH) induced sigmoidal stimulation in cAMP accumulation in 3D2 and granulosa cells, DeGly-phFSH and BV-hFSH induced bell-shaped dose-response curves. Furthermore, the differences in bioactivity between phFSH and alternatively glycosylated FSH is evident at the membrane level, since adenylate cyclase activity in a cell-free paradigm revealed similar differences in bioactivity between phFSH and BV-hFSH. This may be due to an altered receptor-ligand conformation, since we have also shown that deglycosylation of the hormone had a notable effect on receptor binding. Both BV-hFSH and DeGly-phFSH competed for the hFSH-R with similar slope factors to each other, but different from that of phFSH, as has been described by others (28). It has been proposed that isoforms of hFSH or hLH could act in both a competitive or noncompetitive manner with native hormone for interaction with the receptor. In fact, chemically deglycosylated hFSH can antagonize the effects of native FSH (24, 25). Similarly, deglycosylated hCG acts as a noncompetitive antagonist (with equimolar affinity) to native LH, suggesting that these two ligands do not share the same binding site(s) (29). We have observed that, depending on the degree of receptor-transducer system activation, the underglycosylated BV-hFSH could behave either in a strictly inhibitory or stimulatory/inhibitory manner simultaneously. Thus, our data have extended earlier observations to clearly demonstrate that alternatively glycosylated hFSH is not a true antagonist of the hFSH-R, but actually an analog with partial agonistic activity capable of inducing the FSH-R to activate other signaling pathways than those already established for this receptor.
We hypothesized that the mechanism for the dual bioactivity of these hormones is related to the ability of the ligands to stabilize certain receptor conformations that permit interaction with multiple G proteins. This was evaluated by the use of two ADP-ribosylating toxins, PTX and CTX. These two toxins block either Gs- or Gi/Go-activated signaling pathways, respectively. Treatment of 3D2 cells with CTX specifically blocked the ascending phase of BV-hFSH bioactivity, but did not abolish the descending phase. In contrast, PTX completely and selectively blocked the descending phase of BV-hFSH bioactivity. The two toxins, therefore, differentially affected BV-hFSH-induced responses in 3D2 cells. Interestingly, PTX treatment also altered the efficacy of both phFSH- and BV-hFSH-induced cAMP accumulation. These observations imply that a PTX-sensitive mechanism participates in maintenance of a fully responsive transduction system. The reason(s) for these findings is not readily apparent; however, two possible mechanisms could be invoked. First, it is possible that either multiple PTX-sensitive G proteins (e.g. Go) or another PTX-sensitive pathway confers full responsiveness to the system. Alternatively, PTX-dependent chronic activation of adenylate cyclase (due to the removal of tonic Gi-dependent inhibitory inputs) could lead to densensitization of the enzyme. The latter hypothesis is supported by the observation that PTX treatment reduced forskolin responsiveness in terms of cAMP production (data not shown); however, this observation does not completely refute the first mechanism. Experiments are currently underway in our laboratory to address these possibilities in an attempt to discern whether other signaling pathways are involved in maintaining the gain of the FSH/FSH-R transduction system.
Our data provide strong evidence that the FSH-R is capable of activating alternate signaling cascades other than those activated through Gs. Moreover, the ability of the FSH-R to associate with alternative signaling molecules was dependent upon the degree of receptor-transducer system activation. At low levels of activation, BV-hFSH was capable of inducing both an ascending (stimulatory) or descending (reduced activity) bioactivity profile. At a midrange of receptor-transducer activation, addition of BV-hFSH did not increase cAMP accumulation over levels observed with phFSH alone. Furthermore, an inhibitory component was only identifiable at high concentrations of BV-hFSH. Similarly, at high receptor-transducer activation, BV-hFSH was inhibitory at all doses tested. Because in these studies cells were exposed to a mix of differently glycosylated forms of FSH, these conditions could resemble what occurs in vivo, i.e. circulating FSH isoforms with different glycosylation patterns. Taken together with the fact that biphasic responses were also evident in primary granulosa cells, these data suggest that this phenomenon may occur under physiological conditions. It is apparent from our data that high ratios of fully to incompletely glycosylated hormone result in positive intracellular signals, whereas lower ratios convey inhibition. Overall, it is tempting to speculate that, while the FSH receptor has an affinity for other G proteins, it has a higher affinity for Gs. Therefore, our data would define the FSH-R as a preferentially Gs-coupled receptor, but with capacity to associate with other (Gi/Go) proteins as well. This so-called promiscuity in receptor-signal transducer coupling is well documented for catecholamine and adenosine receptors (30, 31, 32). However, unlike the catecholamine and adenosine receptors in which different ligands induce promiscuity, the ability of the FSH-R to couple to multiple G protein signaling pathways appears to be dependent on different physiological statuses (glycosylation) of the same ligand (i.e. the FSH molecule). Promiscuity of the FSH-R for coupling would be a physiologically relevant event by empowering the signal transduction system to respond in a positive or negative fashion, depending on the prevailing gonadotropic stimulus interacting with the system.
In conclusion, these data provide evidence that the FSH-R is capable of coupling with more than one G protein subtype and that its association with other subtypes is dependent on the glycosylation pattern of the ligand bound and the degree of receptor-transducer activation. Perhaps, more importantly, our data provide some basis for a physiological role of alternatively glycosylated isoforms of circulating FSH. That is, depending on the prevailing physiological status of the subject, the ovary may be presented with and respond to differing pleiotropic signals from the pituitary that are sensed by the FSH-R and perceived as activation of alternative signaling pathways.
|
MATERIALS AND METHODS |
|---|
|
TOPABSTRACTINTRODUCTIONRESULTSDISCUSSIONMATERIALS AND METHODSREFERENCES |
|---|
and ß were isolated from human pituitary
poly-A+ RNA (Clontech, Palo Alto, CA) by RT-PCR using
gene-specific primers: FSH1,
5'CGCGGATCCGCCATGGATTACTACAGAAAATATGC3'; FSH2,
5'CGCAAGCTTAGCAGTCATCAAGACAGCAC3'; FSHß1,
5'CGCGGATCCCA-GGATGAAGACACTCCAG3'; FSHß2, 5'CGCAAGCT
TCAGGACAAGGGTATGTGGC3'. The cDNA fragments were cloned into pCRII
(Invitrogen, San Diego, CA) and sequenced to verify identity and
fidelity of the cloned fragments. The fragments included restriction
digestion sites (BamHI-hFSH-HindIII) for use in
subcloning into the corresponding sites of the baculovirus transfer
vector, pBluBacIII (Invitrogen). The resulting transfer vectors
containing hFSH and hFSHß were cotransfected into Sf9 insect cells
with linearized AcNPV (Baculogold, Pharmingen, San Diego, CA) to
generate recombinant virus. The latter was purified by successive
rounds of plaque purification. A high titer viral stock was generated
by two rounds of infection with the purified viral stock. For
expression experiments, the high titer viral stock was used to infect
Hi5 cells (Invitrogen), cultured in serum-free medium using a
multiplicity of infection of 5 for each recombinant virus. The culture
supernatant was collected at 96 h postinfection and analyzed for
production of immunoreactive hFSH/ß dimer by a hFSH IRMA and by
Western blot using antibodies specific for FSH or FSHß. The
expression level of intact hFSH/ß dimer ranged from 1 to 5
mg/liter.
Deglycosylation of phFSH
Deglycosylated phFSH was obtained from Dr. P. M. Sluss
(Massachusetts General Hospital, Charlestown, MA). Purified hFSH (2 mg,
Cortex Biochem, San Leandro, CA) was deglycosylated by a 60-min
exposure to hydrogen fluoride gas at room temperature. After exposure
to hydrogen fluoride, the deglycosylated (all but the N-linked sugar)
hormone was separated from all other carbohydrates by gel filtration
chromatography. The mass of the purified material was determined by
amino acid analysis and found to be consistent with acid hydrolysis of
phFSH. The deglycosylated phFSH had an apparent molecular mass of 29
kDa, with no larger forms of the protein detectable by SDS-PAGE
analysis. Dubois assay of phFSH revealed no detectable carbohydrate
beyond the N-linked sugar of a 20-µg aliquot. The remaining material
was lyophilized and reconstituted in 0.1 M acetic acid
before use in bioactivity studies.
Primary Culture of Granulosa Cells and Aromatase Bioassay
All procedures using animals were approved by the Radnor Animal
Care and Use Committee.Twenty one-day-old immature female
Sprague-Dawley rats (Charles River Laboratories, Wilmington, MA) were
housed under controlled light (12-h light, 12-h dark) and temperature
(25 C) conditions. Food and water were available ad libitum.
Animals were treated by single daily injections of 100 µg/kg
diethylstilbestrol (DES) in olive oil for 3 days. On the fourth
day, animals were euthanized by rapid CO2 asphyxiation, and
the ovaries were removed. Ovaries were washed three times in 50 ml of
sterile HEPES-buffered saline (pH 7.4). Granulosa cells were harvested
by incubating ovaries in a serum-free hypertonic medium consisting of
McCoy’s 5A medium (GIBCO Life Sciences, Grand Island, NY) supplemented
with 5 µg/ml insulin, 5 µg/ml transferrin, 5 ng/ml sodium selenite
(ITS, Sigma Chemical Co, St Louis, MO), 146 µg/ml
L-glutamine, 100 nM testosterone, 100
nM DES, and 100 U/ml penicillin/10 mg/ml streptomycin/250
ng/ml amphotericin B (antibiotic/antimycotic, GIBCO) containing
0.5 M sucrose (Sigma) and 0.1 mM EGTA (Sigma).
Ovaries were then incubated for 45 min at 37 C in a humidified
incubator gassed with 95% air/5% CO2. Subsequently,
ovaries were washed three times with 10 ml isotonic medium (hypertonic
medium without sucrose and EGTA) and incubated for an additional 45 min
in isotonic medium at 37 C. Granulosa cells were harvested by puncture
of swollen follicles using a 23 gauge needle. Isolated granulosa cells
were placed in a 50-ml centrifuge tube and washed two times by the
addition of 50 ml serum-free McCoy’s 5A medium followed by
centrifugation at 700 x g for 5 min. The final cell
pellet was resuspended by gentle trituration in 25 ml serum-free
isotonic medium. Cell number was determined using a hemocytometer, and
viability was estimated by trypan blue exclusion. Cells were plated
into 24-well Nunc (Naperville, IL) tissue culture plates at 100,000
viable cells per well.
The aromatase bioassay was performed according to the method of Hsueh et al. (33). Briefly, cells were challenged with test substances in isotonic McCoy’s 5A medium supplemented with 0.1% BSA (Fraction V, Sigma), ITS, testosterone, DES, glutamine, and antibiotic/antimycotic mix in a total incubation volume of 500 µl. The cells were incubated for 72 h at 37 C with the test substances. At the end of the challenge period, the medium was assayed for estradiol concentration by RIA.
CHO Cell Line and cAMP Accumulation Assay
A CHO cell line expressing the hFSH-R was used to study effects
of FSH on receptor activation (kindly provided by Dr. Kerry Koller,
Affymax Inc., Palo Alto, CA). CHO cells were stably transfected with
the cDNA for the hFSH-R, which was cloned by RT-PCR from human ovarian
RNA. One clone (3D2) was found to express the hFSH-R and to respond to
phFSH with a dose-dependent stimulation in cAMP production. 3D2 cells
were maintained at 37 C in 1:1 DMEM/F12 medium supplemented with 10%
FBS (GIBCO), 146 µg/ml L-glutamine, and 100 U/ml
penicillin/10 µg/ml streptomycin. The cells were plated 1 day before
each experiment into 24-well or 96-well Nunc tissue culture plates at
200,000 or 30,000 cells per well, respectively.
FSH activation of the FSH-R was studied by monitoring cAMP accumulation. Cells were washed twice with Optimem (GIBCO)/0.1% BSA. After the second wash, cells were preincubated in either 500 µl (24-well format) or 100 µl (96-well format) Optimem/0.1% BSA for 30 min at 37 C. The medium was removed from the wells, and the cells were challenged for 30 min at 37 C in Optimem/0.1% BSA containing test substances in a total incubation volume of 250 µl or 50 µl for the 24-well and 96-well formats, respectively. Experiments were terminated by the addition of an equal volume of 0.2 N HCl, and cAMP accumulation was measured by RIA.
Adenylate Cyclase Activity Assay
Adenylate cyclase assays were performed on isolated 3D2
cell membranes. 3D2 cells were grown to 90% confluency on 15-cm Nunc
tissue culture dishes in growth medium. The medium was removed and
cells were scraped from the plate into 30 ml FSH-binding buffer
(10 mM Tris-HCl, 1 mM MgCl2, 1
mM CaCl2, 0.1% BSA, and 0.025% sodium azide,
pH 7.2), and the cells were homogenized. The homogenate was centrifuged
at 15,000 x g for 10 min, and the pellet was
resuspended in binding buffer and centrifuged again. The supernatant
was discarded and the pellet resuspended to 100–150 µg/ml protein in
binding buffer. At the start of the assay, 3D2 membranes were pelleted
as above and resuspended in a volume of membrane buffer (50
mM Tris-HCl, pH 7.2, 10 mM MgCl2,
and 2 mM EGTA) to give 2.5 mg membrane protein/ml. Assays
were performed in 96-well plates (Nunc). The following additions were
made to each well in order: 20 µl 0.25% BSA (Sigma), 20 µl of each
concentration of hormone in 0.25% BSA, 20 µl of a solution
containing 2,500 U/ml phosphocreatine kinase (Sigma), 50 mM
creatine phosphate (Sigma) and 0.5% BSA, 40 µl of buffer containing
450 mM Tris-HCl, pH 7.4, 40 mM
MgCl2 and 5 mM isobutylmethylxanthine (Sigma),
10 µl 2 mM GTP (Sigma), and 50 µl 4 mM ATP
(Sigma). Each hormone concentration was assayed in quadruplicate. The
plates were incubated for 10 min at 37 C. After the incubation, the
content of each well was rapidly transferred to a 1.5-ml centrifuge
tube and spun at 12,000 x g for 5 min at room
temperature. The supernatants were placed into fresh tubes and stored
at -20 C until assayed for cAMP by RIA.
Radioligand-Binding Assay
Binding assays were performed using the same 3D2 cell membrane
preparations used in the adenylate cyclase assays. To perform the
binding assay, 100 µl/well (100 µg membrane protein) of the 3D2
membrane homogenate were added to a 96-well microtiter plate followed
by the addition of 50 µl of either binding buffer (total binding),
phFSH, DeGly-phFSH, or BV-hFSH at varying concentrations. Nonspecific
binding was determined in the presence of 1 µM phFSH.
Reactions were initiated by the addition of 50 µl
[125I]phFSH (50 pM; 55,000 cpm, 3500–4500
Ci/mmol; NEN, Boston, MA) in binding buffer, for a final reaction
volume of 200 µl. Plates were incubated on an orbital shaker for
2 h at 25 C.
The binding assay was terminated by harvesting the cell membranes using
a 96-well vacuum harvester (Skatron Instruments, Inc, Sterling, VA)
onto presoaked (30 min in 50 mM Tris/1% BSA, pH 7.2)
Skatron Blue mat 11740 glass fiber filters. Harvesting was completed by
washing unbound radioactivity from the mats with five cycles of 3.5 ml
of 50 mM Tris-HCl (4 C). Filters were individually punched
out and the bound radioligand was determined by counting single disks
for 1 min in a 
-counter (ICN Biomedical, Costa Mesa, CA).
RIAs
Estradiol levels in medium samples from the aromatase bioassay
were measured using a commercially available Coat-a-Count kit with
modifications (Diagnostic Products Corp., Los Angeles, CA). Medium
samples were preincubated in the presence of assay buffer (100 µl
total volume) in antibody-coated tubes for 1 h at 37 C and after
the addition of [125I]estradiol (1 ml), tubes were
incubated for 2 h at room temperature. The assay was terminated by
draining the tubes, and bound radioactivity was counted in an ICN
-counter for 1 min. This assay has a sensitivity of 0.25 pg/tube.
Intra- and interassay variability is 4.3% and 6.8%, respectively.
Cyclic AMP accumulation in the 3D2 cells was measured using a
commercially available double-antibody RIA kit with some modifications
(Amersham, Arlington Heights, IL). Medium samples were incubated in the
presence of tracer and primary antibodies for 1 h at room
temperature. Secondary antibodies were added, and the tubes were
incubated for 10 min at room temperature and centrifuged at 1000
x g for 15 min. The supernatant was drained and the pellets
counted in an ICN -counter for 1 min. This assay has a sensitivity
of 2 fmol/tube. The intra- and interassay variation for this assay is
approximately 6.7% and 10.8%, respectively.
For the adenylate cyclase activity assay, an acetylation step was performed on samples and standards before assay for cAMP as above (Amersham). Using this protocol, the assay did not detect ATP at the concentrations used in the adenylate cyclase activity reactions. This assay has a sensitivity of 0.25 fmol/tube. The intra- and interassay variation for the acetylated protocol is approximately 4.8 and 6.6%, respectively.
To normalize for FSH concentration of phFSH, CHO-hFSH, and
DeGly-phFSH, as well as crude and purified Hi5 cell supernatants,
hormones were assayed using a hFSH IRMA. This assay was performed using
commercially available reagents (Diagnostic Products Corp.). Aliquots
of the hormones were serially diluted in assay buffer and assayed as
5-µl aliquots in 95 µl assay buffer in tubes coated with primary
hFSH antibodies. One milliliter of 125I-labeled secondary
hFSH antibodies was added to the tubes, and they were incubated on an
orbital shaker at room temperature for 1 h. Thereafter, the tube
content was discarded, and the tubes were washed twice by the addition
of 500 µl assay wash buffer per wash and then counted in an ICN
-counter for 1 min. This assay has a sensitivity of 0.019 mIU/tube.
The intra- and interassay variation is approximately 2.4% and 4.0%,
respectively.
Statistical Analysis
Statistical analyses were performed for bioassay data
using the SigmaStat software package (Jandel Scientific, San Raphael,
CA). Differences between treatment groups were analyzed by ANOVA.
Differences vs. the control group were analyzed after a
significant ANOVA by the Dunnet’s test. In some instances, data were
found to be skewed from normality or to have heterogeneous variance. In
such cases, log transformation of the data was performed. Differences
between treatment groups were considered significant if
P < 0.05.
Sigmoidal dose-response curves were fitted and ED50s determined mathematically using a four-parameter logistic equation and the SigmaPlot software (Jandel Scientific). Bell-shaped dose-response curves were fitted and ED50s and ID50s determined mathematically using a seven-parameter logistic equation as described by Rovati et al. (34) with some modifications: {((a - d1)/(1 + (x/c1)b1) + d1) - ((a - d2)/(1 + (x/c2)b2) + d2)}; where a = asymptotic maximum, b1 = ascending slope factor, c1 = ED50 of ascending portion of the curve, d1 = asymptotic minimum for the ascending portion of the curve, b2 = slope factor of the descending portion of the curve, c2 = ID50 of descending portion of the curve, and d2 = asymptotic minimum for the descending portion of the curve.
Data from radioligand binding studies were analyzed using the JMP software package (SAS Inc, Cary, NC). Square-root transformation of the data was performed in conjunction with a Huber weighting procedure. A four-parameter logistic equation was used to fit competition curves and calculate ID50s from the transformed, weighted data.
|
ACKNOWLEDGMENTS |
|---|
|
FOOTNOTES |
|---|
1 Current Address: Ligand Pharmaceuticals, Inc., 10255 Science Center
Drive, San Diego, California 92121. 
Received for publication December 4, 1996. Revision received February 6, 1997. Accepted for publication February 18, 1997.
|
REFERENCES |
|---|
|
TOPABSTRACTINTRODUCTIONRESULTSDISCUSSIONMATERIALS AND METHODSREFERENCES |
|---|
subunit in transduction of biological signal in glycoprotein hormones.
Science 229:65–67This article has been cited by other articles:
 |
|
 |
|
  K. S. Monkkonen, R. Aflatoonian, K.-F. Lee, W. S.B. Yeung, S.-W. Tsao, J. T. Laitinen, and A. Fazeli Hormonal regulation of G{alpha}i2 and mPR{alpha} in immortalized human oviductal cell line OE-E6/E7 Mol. Hum. Reprod., December 1, 2007; 13(12): 845 - 851. [Abstract] [Full Text] [PDF]
|
|
|
|
 |
|
 K. S. Monkkonen, R. Aflatoonian, K.-F. Lee, W. S.B. Yeung, S.-W. Tsao, J. T. Laitinen, E. M. Tuckerman, T.C. Li, and A. Fazeli Localization and variable expression of G{alpha}i2 in human endometrium and Fallopian tubes Hum. Reprod., May 1, 2007; 22(5): 1224 - 1230. [Abstract] [Full Text] [PDF]
|
|
|
|
 |
|
 S. D. Yanofsky, E. S. Shen, F. Holden, E. Whitehorn, B. Aguilar, E. Tate, C. P. Holmes, R. Scheuerman, D. MacLean, M. M. Wu, et al. Allosteric Activation of the Follicle-stimulating Hormone (FSH) Receptor by Selective, Nonpeptide Agonists J. Biol. Chem., May 12, 2006; 281(19): 13226 - 13233. [Abstract] [Full Text] [PDF]
|
|
|
|
 |
|
 S. Maudsley, B. Martin, and L. M. Luttrell The Origins of Diversity and Specificity in G Protein-Coupled Receptor Signaling J. Pharmacol. Exp. Ther., August 1, 2005; 314(2): 485 - 494. [Abstract] [Full Text] [PDF]
|
|
|
|
 |
|
 B. J. Arey, R. Seethala, Z. Ma, A. Fura, J. Morin, J. Swartz, V. Vyas, W. Yang, J. K. Dickson Jr., and J. H. M. Feyen A Novel Calcium-Sensing Receptor Antagonist Transiently Stimulates Parathyroid Hormone Secretion in Vivo Endocrinology, April 1, 2005; 146(4): 2015 - 2022. [Abstract] [Full Text] [PDF]
|
|
|
|
|
|
N. H. Moniri, D. Covington-Strachan, and R. G. Booth Ligand-Directed Functional Heterogeneity of Histamine H1 Receptors: Novel Dual-Function Ligands Selectively Activate and Block H1-Mediated Phospholipase C and Adenylyl Cyclase Signaling J. Pharmacol. Exp. Ther., October 1, 2004; 311(1): 274 - 281. [Abstract] [Full Text] [PDF]
|
|
|
|
 |
|
 A. Ulloa-Aguirre, C. Timossi, J. Barrios-de-Tomasi, A. Maldonado, and P. Nayudu Impact of Carbohydrate Heterogeneity in Function of Follicle-Stimulating Hormone: Studies Derived from in Vitro and in Vivo Models Biol Reprod, August 1, 2003; 69(2): 379 - 389. [Abstract] [Full Text] [PDF]
|
|
|
|
|
|
B. J. Arey, D. C. Deecher, E. S. Shen, P. E. Stevis, E. H. Meade Jr, J. Wrobel, D. E. Frail, and F. J. Lopez Identification and Characterization of a Selective, Nonpeptide Follicle-Stimulating Hormone Receptor Antagonist Endocrinology, October 1, 2002; 143(10): 3822 - 3829. [Abstract] [Full Text] [PDF]
|
|
|
|
|
|
J. A. Dias Is there any physiological role for gonadotrophin oligosaccharide heterogeneity in humans?: II. A biochemical point of view Hum. Reprod., May 1, 2001; 16(5): 825 - 830. [Abstract] [Full Text] [PDF]
|
|
|
|
 |
|
 M. P. Mayorga, J. Gromoll, H. M. Behre, C. Gassner, E. Nieschlag, and M. Simoni Ovarian Response to Follicle-Stimulating Hormone (FSH) Stimulation Depends on the FSH Receptor Genotype J. Clin. Endocrinol. Metab., September 1, 2000; 85(9): 3365 - 3369. [Abstract] [Full Text]
|
|
|
|
|
|
K. Gen, K. Okuzawa, B. Senthilkumaran, H. Tanaka, S. Moriyama, and H. Kagawa Unique Expression of Gonadotropin-I and -II Subunit Genes in Male and Female Red Seabream (Pagrus major) During Sexual Maturation Biol Reprod, July 1, 2000; 63(1): 308 - 319. [Abstract] [Full Text]
|
|
|
|
|
|
S. Krantic and M. Benahmed Somatostatin Inhibits Follicle-Stimulating Hormone-Induced Adenylyl Cyclase Activity and Proliferation in Immature Porcine Sertoli Cell via sst2 Receptor Biol Reprod, June 1, 2000; 62(6): 1835 - 1843. [Abstract] [Full Text]
|
|
|
|
|
|
G. M. Lanuza, N. P. Groome, J. L. Barañao, and S. Campo Dimeric Inhibin A and B Production Are Differentially Regulat |
