The Adapter Protein APPL1 Links FSH Receptor to Inositol 1,4,5-Trisphosphate Production and Is Implicated in Intracellular Ca²⁺ Mobilization

The adapter protein APPL1 interacts with the first intracellular loop of the FSH receptor and this association supports FSH-induced Ca²⁺ release from intracellular stores.

Abstract

FSH binds to its receptor (FSHR) on target cells in the ovary and testis, to regulate oogenesis and spermatogenesis, respectively. The signaling cascades activated after ligand binding are extremely complex and have been shown to include protein kinase A, mitogen-activated protein kinase, phosphatidylinositol 3-kinase/protein kinase B, and inositol 1,4,5-trisphosphate–mediated calcium signaling pathways. The adapter protein APPL1 (Adapter protein containing Pleckstrin homology domain, Phosphotyrosine binding domain and Leucine zipper motif), which has been linked to an assortment of other signaling proteins, was previously identified as an interacting protein with FSHR. Thus, alanine substitution mutations in the first intracellular loop of FSHR were generated to determine which residues are essential for FSHR-APPL1 interaction. Three amino acids were essential; when any one of them was altered, APPL1 association with FSHR mutants was abrogated. Two of the mutants (L377A and F382A) that displayed poor cell-surface expression were not studied further. Substitution of FSHR-K376A did not affect FSH binding or agonist-stimulated cAMP production in either transiently transfected human embryonic kidney cells or virally transduced human granulosa cells (KGN). In the KGN line, as well as primary cultures of rat granulosa cells transduced with wild type or mutant receptor, FSH-mediated progesterone or estradiol production was not affected by the mutation. However, in human embryonic kidney cells inositol 1,4,5-trisphosphate production was curtailed and KGN cells transduced with FSHR-K376A evidenced reduced Ca²⁺ mobilization from intracellular stores after FSH treatment.

FSH supports normal folliculogenesis and spermatogenesis by binding to the FSH receptor (FSHR) on ovarian granulosa cells and testicular Sertoli cells, respectively (1). FSHR is a G protein–coupled receptor, composed of a large N-terminal extracellular ligand-binding domain, a heptahelical transmembrane domain, and an intracellular C terminus (2). FSHR responds to FSH binding via activation of the heterotrimeric G protein Gαs, with subsequent activation of adenylate cyclase and the cAMP/protein kinase A (PKA) signaling pathway (1, 3). In the female, the successful maturation of a preovulatory follicle is marked by stages of granulosa cell proliferation and differentiation. FSH stimulates estrogen production in granulosa cells by inducing aromatase expression through activation of the cAMP/PKA pathway. In the male, FSH is essential for androgen production and spermatogenesis. Synthesis of the steroidogenic acute regulatory protein (StAR), the rate limiting enzyme in the process of steroidogenesis, is initiated by trophic hormone activation of the cAMP/PKA signaling pathway (4, 5).

The pleiotropic effects of FSH on its target cells suggest that the signaling pathways activated by FSHR are highly complex and nonlinear. Downstream effector proteins activated by cAMP (Epacs) have been identified, suggesting that FSHR can signal through cAMP in a PKA-independent manner (6–8). It has also been shown that FSH-mediated signaling branches out from the PKA pathway to use multiple downstream cytosolic effectors (9). Although FSH induction of aromatase is considered to be primarily a cAMP/PKA-mediated event, the phosphatidylinositol 3-kinase/protein kinase B (PKB) pathway has also been identified as an effector of FSH action (16, 17, 18) in granulosa cells (10) and Sertoli cells (11). Involvement of the MAPK family as an additional pathway in FSHR signaling has been demonstrated. FSH treatment activates p38 MAPK (12) and ERK1/2 (13–15) in granulosa cells, as well as ERK1/2 in ovarian surface epithelial cells (16).

Although the canonical pathway of FSH-mediated cellular activation is considered to be through the cAMP/PKA pathway, FSH stimulation of granulosa (17, 18) and Sertoli cells (19–23) also results in a rise in intracellular Ca²⁺ ([Ca²⁺]i). A consensus has not yet been reached on whether the rise in [Ca²⁺]i represents a Ca²⁺ release from intracellular stores or an influx through plasma membrane ion channels. In Sertoli cells, FSH appears to induce a Ca²⁺ influx through T-type Ca²⁺ channels (24) independent of the Gαs/adenylate cyclase pathway (21). In granulosa cells, the rise in [Ca²⁺]i has been shown to be a direct effect of cAMP, although it is independent of PKA (18). Interestingly, when FSHRs are overexpressed, FSH-mediated inositol phosphate production has been shown to dampen the expression of aromatase (25) in granulosa cells, likely through an EGF-like receptor transactivation event followed by delayed ERK1/2 activation (26).

The structure of the FSHR transmembrane domain includes three extracellular loops (ECL1, ECL2, ECL3) and three intracellular loops (ICL1, ICL2, ICL3). The ICLs are involved in the initiation and regulation of signaling pathways (27). Identification of cytosolic proteins that interact with the ICLs and/or C terminus of the FSHR is of particular importance in the identification of novel signaling pathways. To date, an assortment of additional signaling proteins have been identified as specific interactors with the FSHR, including 14-3-3τ (28), PKB (29), APPL2, and Foxo1a (30).

APPL1 (Adapter protein containing Pleckstrin homology domain, Phosphotyrosine binding domain and Leucine zipper motif; dip13α) is a cytoplasmic scaffold protein that interacts with the FSHR ICL1 (29). To define this interaction, it was necessary to identify critical residues in the FSHR ICL1 that are required for APPL1 binding and further, to identify potential functions of APPL1–FSHR interaction in the mediation of FSHR signal transduction.

Three residues were required for FSHR binding to the adapter protein APPL1, and lysine 376 was studied. Mutation of this residue did not affect FSHR-ligand affinity, agonist-stimulated induction of cAMP, or steroidogenesis. However, FSHR-K376A evidenced a defect in FSH-mediated inositol 1,4,5-trisphosphate (IP₃) production. Moreover, although wild type (wt)-FSHR can enable both external and internal FSH-mediated Ca²⁺ signaling, FSHR-K376A supported primarily FSH-mediated extracellular calcium flux.

Materials and Methods

Antibodies

The anti-FSHR monoclonal antibody (mAb) 106.105 has been previously described (31). A mouse isotype control antibody (IgG_2b) was kindly provided by Dr. Gary Winslow (Wadsworth Center, Albany, NY). Mouse anti-FLAG horseradish peroxidase (HRP) conjugate was purchased from Sigma (St. Louis, MO), and goat antimouse HRP conjugate was purchased from Invitrogen (Carlsbad, CA).

β-Galactosidase assay

Full-length APPL1 was inserted into the yeast library plasmid pJG4-5 and cotransformed into the yeast strain RFY206 (MATa, trp1Δ::hisG, his3Δ200, ura3-52, lys2Δ201, leu2-3), along with a β-galactosidase reporter plasmid (pSH18-34) and individual bait plasmids. The bait plasmid consisted of pJK202, in which lexA was fused to wt FSHR ICL1, FSHR ICL1 containing an alanine substitution mutation or lexA alone. Colonies were isolated on synthetic medium lacking histidine, tryptophan, and uracil; they were induced in the same medium with galactose and assayed for β-galactosidase activity according to the method of Miller (32). The experiment was repeated twice and the results analyzed by one-way ANOVA (P < 0.05) using Prism software (GraphPad Software Inc., San Diego, CA).

Mammalian expression-vector construction

The APPL1 plasmid, which contains an N-terminal FLAG epitope (FLAG-APPL1), has been described (29). A cDNA clone of human FSHR was generously provided by Ares Advanced Technologies (Randolph, MA) (33) and recloned into pShuttle-cytomegalovirus (Agilent Technologies, La Jolla, CA). This plasmid was used as a template in a PCR to introduce single alanine substitutions (K376A, L377A, F382A) into FSHR.

Adenovirus construction

Recombinant adenovirus expressing the K376A, L377A, or F382A mutation of FSHR was generated using the AdEasy Adenoviral Vector System (Agilent Technologies, La Jolla, CA) according to manufacturer's instructions. Briefly, mutant FSHR cloned into pShuttle-cytomegalovirus was cotransformed with pAdEasy-1 into BJ5183 Escherichia coli to generate recombinant pAdEasy-1. The recombinant pAdEasy-1 was then transfected into human embryonic kidney (HEK 293) cells, which express the adenovirus E1 protein missing from pAdEasy-1, thereby allowing the production of infectious virus. Viral stocks were amplified on HEK 293 cells to generate high-titer virus stocks which were quantitated by O.D. at 260 nm.

HEK 293 cell culture and transfection

HEK 293 cells were maintained in Eagle's minimum essential medium (EMEM) containing 10% fetal bovine serum, supplemented with penicillin and streptomycin, and incubated in a 5% CO₂-humidified atmosphere at 37 C. Subconfluent cells in 60-mm dishes were transfected with FLAG-APPL1 and FSHR constructs (1 μg/each), using Lipofectamine reagent according to the manufacturer's instructions (Invitrogen Corp., Carlsbad, CA). Cells were incubated for an additional 24 h before experiments were performed.

Cell culture and viral transduction of human cell line KGN

KGN cells (56) were maintained in Dulbecco's Modified Eagle's Medium/Ham's F-12 (DMEM/F12) containing 10% fetal bovine serum, supplemented with penicillin and streptomycin, and incubated in a 5% CO₂-humidified atmosphere at 37 C. Subconfluent cells were transduced with recombinant adenovirus (1 × 10⁵ viral particles per cell) to express wt FSHR (Ad-wt FSHR) or FSHR K376A (Ad-FSHR K376A). Transduced cells were incubated for 48 h before experiments were performed.

Radioreceptor assay

HEK 293 cells transiently transfected with 2 μg plasmid encoding wt FSHR or one of the FSHR ICL1 mutants were assayed for specific binding to FSH in a radioreceptor assay (RRA) as described previously (34). For the competition binding assay, HEK 293 cells grown in 24-well plates were transfected with plasmid encoding wt FSHR or FSHR K376A. After 24 h, the medium was removed, and serum-free medium containing 20 ng/ml ¹²⁵I-pituitary human FSH (hFSH) was added to each well in the presence of successively higher concentrations of unlabeled pituitary hFSH. Hormone was allowed to bind for 1 h at 37 C in a 5% CO₂-humidified incubator before plates were placed on ice, and cells were washed twice with ice-cold PBS. The cells were solubilized in 0.5 ml 2 N NaOH at room temperature for 45 min and samples were counted in a γ-counter (Wallac 1470 Wizard).

Cell lysate preparation and immunoprecipitation

Cells lysis and immunoprecipitations were performed as previously described (34).

Gel electrophoresis and Western immunoblotting

Samples (1/4 of each immunoprecipitate) were resolved by 7.5% SDS-PAGE in a discontinuous buffer system (35). After separation, the proteins were transferred to polyvinylidene fluoride membranes (Millipore, Billerica, MA) according to the method of Bjerrum and Schafer-Nielsen (36). Membranes were blocked in Tris-buffered saline with Tween 20 (TBST) [10 mm Tris-HCl (pH 7.2), 150 mm NaCl, 0.5% Tween 20] and 5% nonfat dry milk overnight at 4 C. Blots were washed briefly in TBST and probed with mAb 106.105 (5 μg/10 ml) for 1 h at room temperature. Blots were washed and incubated with goat antimouse Ab (1:10,000) conjugated to HRP (Biosource, Carlsbad, CA) for 1 h at room temperature. Alternatively, blots were probed with HRP-conjugated anti-FLAG mAb (1:5,000) (Sigma, St. Louis, MO) for 1 h at room temperature. Signal was developed using Supersignal West Pico chemiluminescent substrate (Pierce, Rockford, IL).

Induction and measurement of cAMP production

HEK 293 cells were grown in 24-well plates and transfected with plasmid encoding wt FSHR or FSHR K376A. In a parallel assay, KGN cells were grown in 24-well plates and transduced with recombinant adenovirus encoding wt FSHR or FSHR K376A. After 24 h, the cells were pretreated with 250 μl serum-free medium containing 1 mm isobutylmethylxanthine (IBMX) for 15 min. The cells were then stimulated with successively higher doses of FSH (in a final volume of 300 μl) for 1 h at 37 C in a 5% CO₂-humidified incubator. Plates containing cells and medium were frozen/thawed four times, an equal volume of 100% EtOH was added, and the medium was clarified by centrifugation at 13,000 × g for 10 min at room temperature. Supernatants were stored at −80 C. Total cAMP accumulation was determined by RIA as previously described (37).

Induction and measurement of progesterone and estradiol production

KGN cells were grown in 24-well plates and transduced with recombinant adenovirus (1 × 10⁵ viral particles per cell) to express wt FSHR (Ad-wt FSHR) or FSHR K376A (Ad-FSHR K376A). After 48 h, the cells were stimulated with successively higher doses of FSH in serum-free medium (in a final volume of 1 ml) for 72 h at 37 C in a 5% CO₂-humidified incubator. For estradiol production, the treatment medium was supplemented with 0.5 μm testosterone. Samples were heated to 100 C for 10 min and clarified by centrifugation at 13,000 × g for 15 min at room temperature. Supernatants were stored at −20 C. Total progesterone and estradiol accumulation was determined by RIA as previously described (38).

In vitro bioassay of FSH

Animal studies were approved by the Wadsworth Center Institutional Animal Care and Use Committee (Protocols 06-407 and 08-393). Sprague Dawley rats (Taconic Farms, Germantown, NY) were provided with a standard pellet diet and water ad libitum.

The in vitro bioassay for FSH was conducted as previously described with some modifications (39). Immature (21 ± 3 d old) female rats were injected sc once daily for 4 d with 1 mg diethylstilbesterol (DES) in sesame oil. DES was solubilized in sesame oil (10 mg/ml) in a sonicating water bath. Animals were injected once a day for 4 d with 100 μl of DES in the scruff of the neck. On the 5th day animals were killed and the ovaries and part of the uterus dissected, and placed in Modified Eagles Media (MEM) with 10% chicken serum. Collected tissues were kept at 37 C until the ovaries were freed from the bursa, trimmed, and placed into fresh medium. Granulosa cells were harvested by puncturing the ovaries with 27-gauge needles in warm MEM containing 500 nm testosterone, 10% chicken serum, gentamycin, and 0.01 m HEPES. Only two ovaries were processed at a time, keeping all other tissues at 37 C with 5% C0₂. When all tissues were processed, cells and ovaries were taken up into 10 ml pipettes and triturated into a 50 ml conical tube at room temperature. The triturated cells were collected allowing the ovary shells and large particles to sediment by gravity. An equivalent of cells harvested from two ovaries was plated in a 24-well plate and incubated at 37 C and 5% CO₂. This was roughly about 300,000 cells per well.

Forty-eight hours after plating, granulosa cells were transduced with wt FSHR or FSHR K376A adenovirus as described above before the addition of hFSH as indicated. After addition of hormone, the cells were cultured for 72 h. Media was collected, heated to 100 C for 10 min, and cleared at 13,000 × g for 10 min. Samples were frozen until assayed.

Induction and measurement of IP₃ production

HEK 293 cells were seeded in 24-well plates at a density of 1 × 10⁵ cells per well and were incubated in a 5% CO₂-humidified atmosphere at 37 C for 24 h. Cells were transfected with 0.4 μg per well of the wt FSHR or FSHR K376A receptor plasmid using 2 μl lipofectamine in 0.25 ml OPTI-MEM (Life Technologies, Inc., Grand Island, NY). After 5 h, 0.25 ml of DMEM containing 20% fetal calf serum was added to each well. Twenty-four hours after the start of transfection, the medium was replaced with fresh DMEM and the cells were allowed to grow for another 24 h before treatment. Cells were washed twice with DMEM/0.1% BSA and cellular inositol lipids were labeled in DMEM (inositol-free) supplemented with 4 μCi [³H] myo-inositol (PerkinElmer Life Sciences Inc., Boston, MA) for 18 h at 37 C. After preloading, the cells were washed twice in DMEM (inositol-free) containing 5 mm LiCl and were stimulated with medium or recombinant FSH (500 ng/ml) in 0.5 ml DMEM/LiCl for 2 h. The medium was removed and 1 ml 0.1 m formic acid was added to each well. The cells were frozen and thawed to disrupt the cell membranes. IP₃ accumulation was determined by Dowex anion exchange chromatography and liquid scintillation spectroscopy (40). The experiment was repeated three times, and statistical significance was determined by t test (P < 0.01).

Measurement of intracellular-free calcium concentration

KGN cells transduced with wt FSHR or FSHR K376A were loaded with the fluorescent calcium indicator dye Fluo-4-AM (Invitrogen Corp., Carlsbad, CA), and [Ca²⁺]i was monitored using time-lapse confocal laser scanning microscopy. KGN cells were incubated with 5 μm Fluo-4-AM in Krebs-Ringer solution (140 mm NaCl, 3 mm KCl, 1 mm MgCl₂, 2 mm CaCl₂, 10 mm HEPES, 10 mm glucose; pH 7.4) for 30 min at room temperature in the dark. Cells were then washed twice with Krebs-Ringer and the incubation continued for an additional 15 min at room temperature in the dark to permit Fluo-4-AM ester hydrolysis before stimulation with hFSH (2 μg/50 μl Krebs-Ringer solution).

Fluorescence images were collected on a Zeiss LSM510 META confocal microscope with a Plan-Neofluar 0.8-NA 25 × Imm Corr DIC objective (Carl Zeiss, Thornwood, NY). Cells were excited with the 488-nm line from an argon laser and fluorescence emission was collected with a BP 500–530 nm IR filter. Experiments were initiated by the pneumophoretic application of hFSH, using an Eppendorf Transjector 5246 (Eppendorf North America, Westbury, NY).

FSH was applied by pressure-ejection (7 ψ, 500 msec) from a 3-μm-diameter glass microelectrode placed ∼50 μm away from the center of a group of cells in the microscope field, and FSH-induced fluorescence emission from a 2.2-μm-thick optical slice was collected at the rate of 1 image per sec for 90 sec. Imaging was initiated, and after collection of six images of unstimulated cells hFSH (40 nm) was applied by pneumophoresis. In some experiments, to chelate extracellular calcium, EGTA (final concentration of 3 mm) was added to the solution that was bathing the cells and incubated for 15 min before FSH stimulation. To deplete intracellular calcium stores and prevent their refilling, cells expressing wt FSHR were incubated with 1 μm thapsigargin for 20 min before FSH stimulation.

Analysis of intracellular calcium changes was performed using the Zeiss AIM software (Carl Zeiss). Data were collected from four separate experiments. Within each experiment, five randomly selected regions of cells in each of two MatTek microwell dishes were stimulated with FSH. At least twenty cells were analyzed within each field of view unless otherwise indicated. Raw data were plotted as relative fluorescence intensity vs. time (in sec).

Results

Identification of residues in FSHR ICL1 that are critical for interaction with APPL1

To determine which amino acid residues in the wt FSHR ICL1 are necessary for its interaction with APPL1, an interaction assay was performed in yeast. A single alanine substitution mutation was made at each of the 15 ICL1 residues, and each was tested for interaction with APPL1. Several mutations significantly decreased the association of the loop with APPL1, as determined by measurement of β-galactosidase activity (Fig. 1). However, mutation of residues K376, L377, or F382 in the FSHR ICL1 bait reduced the β-galactosidase activity to the control pJK202 plasmid, providing a strong rationale for selecting these mutations for further experimentation. The lack of interaction observed for the P380A mutant, although statistically significant in the experiment shown, was not a consistent result.

Fig. 1.

Alanine substitution mutations in the FSHR ICL1 affect its association with APPL1 in yeast. The library plasmid pJG4-5 expressing full-length APPL1 was transformed into yeast RFY206 containing the pSH18-34 reporter plasmid and bait plasmids containing LexA fused to the following: wt ICL1, one of 15 ICL1 single alanine substitution mutations or lexA alone (pJK202). Cell extracts were assayed for β-galactosidase activity. This experiment was repeated twice. The results shown are from one experiment with each sample assayed in duplicate. *, Statistically different (P < 0.001) from wt ICL1 cell extract.

Expression, trafficking, and ligand binding of FSHR ICL1 mutants in mammalian cells

To study the interaction of APPL1 and the FSHR ICL1 in a mammalian cell system, constructs containing each one of the ICL1 mutants K376A, L377A, and F382A were made in a mammalian expression vector. To characterize the expression and trafficking of these mutants, a RRA was performed to assess binding to FSH and the protein expression profile was examined by immunoblotting. All mutants were expressed as the three forms of immature FSHR (Fig. 2A) (41). However, only the K376A mutant and, to a lesser extent, the L377A mutant were processed and trafficked to the cell surface comparably to wt FSHR (80 kDa; mature FSHR), as evidenced by immunoblotting and ¹²⁵I-FSH binding results (Fig. 2, A and B). Protein expression in the F382A mutant consisted principally of immature receptor forms at 75 and 62 kDa and the high molecular mass 175-kDa band; the latter possibly represents an irreversible FSHR association with a chaperone. Cell-surface expression correlated with expression of the mature 80-kDa form of the receptor.

Fig. 2.

Mature wt FSHR and FSHR mutant K376A are expressed, traffic to the cell surface, and bind ligand with similar affinity. A, Cell lysates from transiently transfected HEK 293 cells were resolved by SDS-PAGE, transferred to a polyvinylidene fluoride membrane and probed with anti-FSHR mAb 106.105. B, HEK 293 cells transiently transfected with wt FSHR, or one of the single alanine point mutations were analyzed in a radio receptor assay using ¹²⁵I-FSH. Data are expressed as specifically bound counts. C, HEK 293 cells transiently transfected with wt FSHR, or FSHR K376A were incubated in the presence of ¹²⁵I-FSH and increasing concentrations of unlabeled FSH. The experiment was repeated twice.

APPL1 does not modulate FSHR-ligand affinity

To compare the receptor affinity of wt FSHR and FSHR K376A, a competition RRA was performed on transfected HEK 293 cells in the presence of 20 ng/ml ¹²⁵I-FSH and successively increasing concentrations of unlabeled FSH. The K_d values calculated for wt FSHR (3.032 × 10⁻⁹) and for the K376A mutant (3.723 × 10⁻⁹) are similar, indicating no significant difference in ligand-binding affinity (Fig. 2C).

APPL1 interaction with FSHR is blocked by the K376A mutation in mammalian cells

To test whether the K376A mutation disrupts FSHR interaction with APPL1 in mammalian cells, the mutant or wt FSHR was co-expressed with N-terminally FLAG-tagged APPL1 in HEK 293 cells. Immunoprecipitations were performed with the FSHR-specific mAb 106.105 (31). Immunoprecipitates were assayed for the presence of APPL1 through immunoblotting with HRP-conjugated FLAG mAb. As previously demonstrated, APPL1 forms a constitutive association with FSHR even in the absence of FSH treatment (29). In agreement with the β-galactosidase assay results, APPL1 was detected in immune complexes with wt FSHR but not with FSHR K376A (Fig. 3).

Fig. 3.

APPL1 coimmunoprecipitates with wt FSHR but not with FSHR K376A mutant in HEK 293 cells. Cells were cotransfected with FLAG-tagged APPL1 and either wt FSHR or FSHR K376A. Cell lysates were immunoprecipitated with mAb 106.105 (anti-FSHR ECD) or IgG2b (isotype control) as indicated. A, The immunoblot was probed with HRP-conjugated FLAG mAb to detect FLAG-APPL1 and then re-probed (B) with mAb 106.105 to detect FSHR. The experiment was repeated twice.

APPL1 does not modulate FSH-induced cAMP induction

To investigate the possible involvement of APPL1 in PKA signaling, levels of the second messenger cAMP in response to FSH stimulation were measured in transfected HEK 293 cells and in KGN cells expressing wt FSHR or FSHR K376A. Binding of ¹²⁵I-FSH was measured on cells in parallel so that cAMP induction could be normalized to cell surface expression of FSHR. The interaction of APPL1 with FSHR is not essential for FSH-induced cAMP induction in either HEK 293 cells or KGN cells (Fig. 4, A and B).

Fig. 4.

APPL1 interaction with FSHR is not essential for cAMP production in HEK293 cells or cAMP, progesterone, or estradiol production in (KGN) granulosa cells. A, HEK 293 cells were transiently transfected with plasmid encoding wt FSHR or FSHR K376A and then treated with increasing doses of human pituitary FSH for 1 h. B, KGN cells were transduced with adenovirus to express wt FSHR or FSHR K376A and then treated with increasing doses of human pituitary FSH for 1 h. Total cAMP accumulation was determined by RIA. For the steroidogenesis experiments, KGN cells were transduced with adenovirus to express wt FSHR or FSHR K376A. After 24 h, the cells were treated with increasing doses of human pituitary FSH for an additional 72 h in medium ± 0.5 m testosterone (as a substrate for estradiol production). Total progesterone (C) and estradiol (D) accumulation in the medium was determined by RIA. The data in A and B are normalized to ¹²⁵I-FSH binding values, obtained in a parallel assay. The experiments were repeated three times; a representative result is shown.

APPL1 does not modulate FSH-induced production of progesterone or estradiol

To investigate further the possible role of APPL1 in FSHR signaling, steroidogenesis was evaluated in KGN cells transduced with wt or FSHR K376A adenovirus. FSH stimulation of KGN resulted in the induction of progesterone (Fig. 4C) and estradiol (Fig. 4D) whether cells were virally transduced to express wt FSHR or FSHR K376A. As expected, estradiol accumulation was detected only when the treatment medium was supplemented with testosterone; interestingly, testosterone also suppressed the progesterone production in these cells. APPL1 interaction with FSHR is not required for FSH-induced progesterone or estradiol production in KGN cells.

This observation was verified in rat granulosa cells transduced with wt FSHR or FSHR K376A adenovirus (Supplemental Fig. 1 published on The Endocrine Society's Journals Online web site at http://endo.endojournals.org/). Similar responses of the cells to FSH were noted.

APPL1 is essential for FSHR-mediated IP₃ induction

To determine whether FSHR interaction with APPL1 is necessary for agonist-induced IP₃ production, transfected HEK 293 cells expressing wt FSHR or FSHR K376A were treated with FSH. The net increase in IP₃ production that was observed with the wt FSHR was virtually abolished in the K376A mutant (Fig. 5A). Cell lysates analyzed by Western blotting showed similar expression of the wt and K376A mutant (Fig. 5B).

Fig. 5.

Interaction of FSHR with APPL1 is essential for IP₃ production. HEK 293 cells transiently transfected with wt FSHR or FSHR K376A were incubated in the presence of inositol-free DMEM containing 4 μCi/ml [³H]-inositol for 18 h. Cells were washed twice with DMEM containing 5 mm LiCl and incubated for 2 h in DMEM ± 500 ng/ml FSH. After the addition of 0.1 m formic acid, the cell lysates were resolved by Dowex anion-exchange chromatography. A, The fractions were measured in a liquid scintillation counter. The experiment was repeated four times, a representative result is shown. B, Cell lysates were analyzed by Western blotting with mAb 106.105 to compare wt and mutant FSHR expression. A representative immunoblot is shown.

FSH-induced calcium signaling

In KGN cells transduced with wt FSHR adenovirus, FSH induced a rapid increase in [Ca²⁺]i that began within 1 sec after stimulation, peaked at 12–18 sec poststimulation, and gradually decreased toward unstimulated levels near the end of the 90-sec collection period (Fig. 6A). The [Ca²⁺]i increased 1.4- to 3.1-fold with a mean of 1.7 ± 0.03 (sem). This experiment was repeated four times with an average of 46.8 cells analyzed per experiment (n = 187). This increase in [Ca²⁺]i could result from either extracellular Ca²⁺ influx or release from intracellular stores. To determine whether the source of the FSHR-stimulated [Ca²⁺]i was extracellular, EGTA was added to the bathing medium 10 min before FSH stimulation of KGN cells expressing wt FSHR, so as to chelate extracellular Ca²⁺. In the presence of extracellular EGTA, there was a small but detectable rise in [Ca²⁺]i that started almost immediately after stimulation (1–2 sec) and plateaued at ∼21 sec poststimulation (Fig. 6B). The mean fold increase in [Ca²⁺]i was 1.24 ± 0.03 (sem). This experiment was repeated four times with an average of 6.8 cells analyzed per experiment (n = 27). These data suggest that in the absence of an extracellular source of calcium, FSH stimulation can mobilize calcium from intracellular stores. To deplete intracellular stores of Ca²⁺ and prevent their refilling, KGN cells expressing wt FSHR were incubated with thapsigargin for 20 min before FSH stimulation. A small but detectable rise in [Ca²⁺]i resulted; however, the onset of the rise was delayed by ∼10 sec poststimulation and peaked at ∼18 sec poststimulation. The level then slowly decreased to near prestimulation values (Fig. 6C). The mean fold increase in [Ca²⁺]i was 1.23 ± 0.07 (sem). This experiment was performed once (n = 9). The occurrence of a lag time is consistent with the involvement of a secondarily-induced signaling pathway linked to channels in the plasma membrane; mobilization of calcium from intracellular stores displays more rapid kinetics. Overall, these results suggest that KGN cells expressing wt FSHR can mobilize calcium from both extracellular sources and intracellular stores, with the latter most likely being mediated by IP₃.

Fig. 6.

Calcium-mobilizing effect of hFSH on KGN cells expressing wt FSHR or FSHR K376A in culture. Representative profiles of [Ca²⁺]i in a group of cells in culture stimulated by application of hFSH at the time indicated by the solid arrow in each panel. A, In the presence of extracellular calcium in the bathing medium, the [Ca²⁺]i of KGN cells expressing wt FSHR increased almost immediately after FSH application, reaching peak response at 15–20 sec poststimulation (arrowhead) and slowly decreased toward baseline. [Ca²⁺]i increased 1.4- to 3.1-fold with a mean of 1.7 ± 0.03 (sem), n = 187. B, After chelation of extracellular calcium with EGTA, the rise in [Ca²⁺]i began almost immediately after application of FSH, reaching a plateau ∼21 sec poststimulation (arrowhead). The mean fold increase in [Ca²⁺]i was 1.24 ± 0.03 (sem), n = 27. C, When cells were preincubated with thapsigargin and stimulated with FSH in the presence of extracellular Ca²⁺, the rise in [Ca²⁺]i was delayed for ∼10 sec (dashed arrow), peaked at ∼18 sec poststimulation, and slowly decreased thereafter. The mean fold increase in [Ca²⁺]i was 1.23 ± 0.07 (sem), n = 9. D, In the presence of extracellular Ca²⁺ in the bathing medium, the [Ca²⁺]i of KGN cells expressing FSHR K376A increased almost immediately after FSH application and, as was the case for wt FSHR-expressing cells, decreased toward baseline levels. The responses and onset time of response were more variable than for wt FSHR-expressing cells. [Ca²⁺]i increased 1.1- to 2.1-fold with a mean of 1.27 ± 0.02 (sem), n = 88. E, Chelation of extracellular calcium with EGTA completely abolished the FSH-induced increase in [Ca²⁺]i, (n = 39). The data shown are representative of at least four experiments.

APPL1 is implicated in FSHR-mediated Ca²⁺ release from intracellular stores

When KGN cells transduced with the FSHR K376A adenovirus were stimulated with FSH in the presence of extracellular calcium, an increase in [Ca²⁺]i was observed; however, relative to what had been seen for the wt FSHR, the increase displayed variable time kinetics. The response time to peak [Ca²⁺]i ranged from 8–16 sec poststimulation. However, in a manner similar to the kinetics observed for wt FSHR-expressing KGN cells after thapsigargin incubation, [Ca²⁺]i levels decreased toward prestimulation values (Fig. 6D). The [Ca²⁺]i increase ranged from 1.1- to 2.1-fold, with a mean of 1.27 ± 0.02 (sem). This experiment was repeated four times with an average of 22 cells analyzed per experiment (n = 88). The FSH-mediated increase in [Ca²⁺]i was completely abolished when extracellular calcium was depleted in the medium bathing KGN cells that were transduced with adenovirus encoding the K376A mutant FSHR (Fig. 6E). This experiment was repeated five times with an average of 7.8 cells analyzed per experiment (n = 39).

Discussion

APPL1 interacts with the first ICL of FSHR (ICL1) (29, 30). To identify residues in ICL1 that are critical for APPL1 interaction, an alanine substitution mutation was generated at each of the fifteen amino acid residues in the FSHR ICL1 and the mutants were screened for interaction with APPL1 in yeast. The FSHR mutants K376A, L377A, and F382A exhibited a significant lack of interaction with APPL1 and were selected for further analysis.

By definition, a G protein–coupled receptor is coupled to at least one heterotrimeric G protein signaling system. It is well known that binding of FSH to its receptor in target cells results in activation of the Gαs isoform, leading to the induction of cAMP through adenylate cyclase. Acting as a second messenger, cAMP activates the downstream effector PKA (3, 6, 9). The K376A mutation in the FSHR ICL1 does not affect total FSH-induced cAMP production in HEK 293 cells. This is in agreement with the finding by Timossi et al. (42) that while other regions may be involved, the FSHR ICL2 is the loop that plays a central role in Gαs coupling. Note that although total cAMP accumulation appears to be unaffected by the K376A mutation, further investigation will be required to establish whether the temporal and/or spatial regulation of cAMP and its downstream targets are affected.

Human FSH induces cAMP production and steroidogenesis in KGN cells, a cell line generated from an ovarian granulosa cell carcinoma (43). However, ¹²⁵I-hFSH binding to KGN cells is not detectable and with prolonged culture the FSH response is lost (data not shown). Thus, KGN cells provide an appropriate model system for evaluating the competency of the FSHR K376A mutant to stimulate cAMP production and induce steroidogenesis by rescuing FSH responsiveness in these cells. Thus, the previous finding that the K376A mutant is competent to induce cAMP in HEK 293 cells was confirmed in the KGN cell line. In addition, FSH treatment of KGN cells virally transduced to express either wt FSHR or FSHR K376A induced production of progesterone and estradiol, further evidence that APPL1 is not essential for FSHR signaling via the cAMP/PKA signaling pathway. Moreover, similar results were obtained with primary cultures of rat granulosa cells transduced with adenovirus encoding both wt and mutant receptors (Supplemental Fig. 1).

The multiple effects of FSH stimulation on granulosa cell proliferation and maturation suggest that a more complex, nonlinear signaling program exists for FSHR. The traditional view of a linear cAMP/PKA relationship has changed. It is now well established that FSH, in addition to its induction of activation of PKA also induces activation of PKB (8, 10, 11, 29), p38 MAPK (12, 44), ERK1/2 (15, 26, 45), and Src family kinase (15), as well as mobilization of Ca²⁺ (17, 18, 23, 24).

The ability of FSHR to use Ca²⁺ signaling pathways has been well established. FSH stimulation induces an increase in [Ca²⁺]i concentration in both rat granulosa cells and Sertoli cells (17, 18, 20, 23). This increase in [Ca²⁺]i can be attributed to Ca²⁺ release from intracellular stores or from the influx of extracellular Ca²⁺ through plasma membrane channels (46). The results of the present study show that unlike wt FSHR, the FSHR K376A mutant is deficient for the induction of the lipid second messenger IP₃ in HEK 293 cells. Also, and in support of this finding, FSHR K376A did not mobilize Ca²⁺ from intracellular stores (an IP₃ mediated process) in KGN cells, but it did elicit a Ca²⁺ influx after FSH stimulation. While it remains unclear whether this [Ca²⁺]i is influenced by the cAMP-PKA cascade, strong evidence points to the involvement of alternate pathway(s) (18, 19, 21, 47).

It is likely that the mobilization of Ca²⁺ by FSHR signaling is mediated by one or more members of the PLC family of enzymes, the members of which respond to distinct stimuli that induce hydrolysis of PIP₂ to IP₃ and DAG. While the canonical signaling pathway by which IP₃ is generated occurs through Gαq activation of PLCβ, this is not always the case. To date, 13 mammalian isoforms of PLC have been identified. These isozymes may be functionally coupled to receptors by various intracellular regulators such as G-proteins (both Gα and βγ subunits), Epacs, Rap, or Ras (48). Whether the FSHR ICL1 is coupled to any of these effectors through APPL1 is currently unknown.

A recent finding is that FSHR can signal through Gαh, an upstream activator of PLCδ-1 in rat Sertoli cells to promote a Ca²⁺ influx (22). Gαh, also known as tissue transglutaminase (TGII), a multifunctional protein with GTP-binding and GTPase (GAP) activities, had previously been shown to interact with PLCδ-1 (49). Further investigation of this pathway in Sertoli cells demonstrated a noncapacitative Ca²⁺ influx mechanism through T-type Ca²⁺ channels (24). Inositol trisphosphate signaling has also been demonstrated for the LH receptor (LHR) (50) and thyroid hormone receptor (THR) (51). These results with the FSHR K376A mutant suggest that APPL1 interaction with FSHR facilitates activation of the PLC pathway by one or more G protein effectors such as Gαq/11 or Gαh, ultimately leading to mobilization of [Ca²⁺]i.

The physiological significance of APPL1 in fertility is an open question. Recently, APPL1 has been knocked out in the mouse without any ill effects on gross fertility (52, 53). The authors concluded that the homologous APPL2 protein may obfuscate a clear role for APPL1. APPL2 also associates with FSHR, suggesting that redundancy may be plausible (54). However, APPL1 appears to play a role in growth factor and metabolic signaling and differentiation (55). As such it may well be that ill effects of its deletion will not be apparent unless nutritional stress or other environmental factors come into play either during fetal development or following birth.

In conclusion, overexpression of FSHR in the present studies demonstrated that APPL1 participates in the FSH-induced IP₃ pathway and implicates it in intracellular calcium signaling. The major functions of granulosa cells, such as progesterone production and estradiol production, are unaffected when an FSHR mutant that cannot interact with APPL1 is expressed in KGN cells. Whether other granulosa cell functions are linked to additional signal transduction pathways (such as intracellular calcium release) needs to be examined further. In addition, future experiments in native granulosa cells will be required to confirm these results in the presence of native levels of FSHR expression.