Role of the intracellular domains of the human FSH receptor in G_aS protein coupling and receptor expression

Abstract

The human (h) follicle-stimulating hormone receptor (FSHR) belongs to the superfamily of G protein-coupled receptors (GPCRs). This receptor consists of 695 amino acid residues and is preferentially coupled to the G_s protein. This receptor is highly conserved among species (overall homology, 85%), with a 25 %-69 % homology drop when compared to the human LH and TSH receptors. Although studies in prototypical rhodopsin/β-adrenergic receptors suggest that multiple domains in the intracellular loops (iL) and the carboxyl-terminus (Ctail) of these receptors contribute to G protein coupling and receptor expression, there is a paucity of structure/function data on the role of these domains in FSHR function. Employing point mutations we have found that several residues present in the iL2 of the hFSHR are important for both coupling the receptor to the G_s protein and maintaining the receptor molecule in an inactive conformation. In fact, HEK-293 cells expressing several hFSHR mutants with substitutions at R⁴⁵⁰ (central to the highly conserved ERW triplet motif) and T⁴⁵³ (a potential target for phosphorylation) failed to mediate ligand-provoked G_s protein activation but not agonist binding, whereas substitutions at the hydrophobic L⁴⁶⁰ (a conserved residue present in all glycoprotein hormone receptors) conferred elevated basal cAMP to the transfected cells. Thus, this particular loop apparently acts as a conformational switch for allowing the receptor to adopt an active conformation upon agonist stimulation. Residues in both ends of the iL3 are important for signal transduction in a number of GPCRs, including the FSHR. We have recently explored the importance of the reversed BBXXB motif (BXXBB; where B represents a basic residue and X a non-basic residue) present in the juxtamembrane region of the hFSHR iL3. A hFSHR mutant with all basic amino acids present in the iL3 BXXBB motif replaced by alanine failed to bind agonist and activate effector, and was expressed as an immature =62 kDA form of the receptor. Individual substitutions of basic residues resulted in mutants that bound agonist normally but failed to activate effector when replaced at R⁵⁵² or R⁵⁵⁶. Triple mutations in the same motif located in the NH₂-end of the Ctail resulted in a complete inability of the receptor to bind agonist and activate effector, whereas individual substitutions resulted in decreased or virtually abolished agonist binding and cAMP accumulation, with both functions correlating with the detected levels of mature (80 kDa) forms of the receptor. Thus, the BXXBB motif at the iL3 of the FSHR is essential for coupling the activated receptor to the G_s protein, whereas the same motif in the Ctail is apparently more important for membrane expression. The role of cysteine residues present in the Ctail of the FSHR is an enigma since there are no conserved cysteines amongst LHR, FSHR and TSHR. C⁶²⁹ and C⁶⁵⁵ are conserved in the gonadotropin receptors but not in the TSHR. Alanine replacement of C⁶²⁷ had no effect on hFSHR expression and function, whereas the same mutation at C⁶²⁹ altered membrane expression and signal transduction. Serine or threonine substitutions of C⁶⁵⁵ did not modify any of the parameters analyzed. In the hFSHR, C⁶²⁹ may be a target for palmitoylation, and apparently it is the only cysteine residue in the Ctail domain that might play an important role in receptor function.

Introduction

Follicle-stimulating hormone (FSH) or follitropin belongs to a family of closely related glycoproteins composed of FSH, luteinizing hormone (LH) choriogonadotropin hormone (CG) and thyroid-stimulating hormone (TSH). This gonadotropin is produced by the anterior pituitary gland and co-regulates with LH the maintenance of several essential reproductive processes including gametogenesis, follicular development, and ovulation. FSH evokes its biological effects by interacting with a high affinity receptor (R) located on the plasma membrane of its target cells in the ovary and the testis. The FSHR is a glycoprotein that belongs to the superfamily of G protein-coupled receptors (GPCRs), specifically the family of rhodopsin-like receptors, which consist of a single polypeptide chain of variable length, that threads back and forth across the lipid bilayer seven times forming characteristic a-helical transmembrane-spanning domains (TM) connected by alternating extracelullar and intracellular loops (iL), with an extracellular NH₂-terminal domain and an intracellular COOH-terminal segment (Ctail) (Ulloa-Aguirre and Conn, 1998; Lu et al., 2002). These receptors mediate their intracellular actions via activation of one or more guanine-nucleotide-binding signal transducing proteins (G proteins). The human FSHR consists of 695 amino acids (the first 17 amino acids encoding the signal sequence) (Simoni et al., 1997; Ulloa-Aguirre and Timossi 1998; Dias et al., 2002); upon activation by agonist, the activated receptor may trigger activation of a number of intracellular signaling pathways. In the classical, linear signaling cascade, occupancy of the FSHR causes activation of the heterotrimeric G_s protein, which in turn stimulates the effector adenylyl cyclase with the consequent increase in the synthesis of the second messenger cAMP, activation of PKA, phosphorylation of cAMP response element-binding protein, and activation of transcription (Reichert and Dattatreyamurty, 1989). Nevertheless, increasing evidence indicates that in addition to the adenylyl cyclase/cAMP/PKA signaling pathway, activation of the FSH receptor by its cognate ligand also triggers activation of other intracellular signaling cascades, including the MAPK and PI3-K/Akt pathways (Cameron et al., 1996; Maizels et al., 1998; González-Robayna et al., 2000; Richards 2002; Seger et al., 2001).

Site-directed mutagenesis and chimeric studies of several GPCRs belonging to the family of rhodopsin-like receptors indicate that several cytoplasmic domains of these receptors, particularly the iL2, the juxtamembrane portions of the iL3 and the Ctail, are involved in signal transduction (Ulloa-Aguirre and Conn 1998). In contrast to the related TSHR and LHR, there is a paucity of structure-function data on the role of the intracellular domains in FSHR-mediated signal transduction and receptor expression. Here we summarize some recent information on the role of the intracytoplasmic domains in function of the hFSHR.

Intracellular loop 1

This relatively short FSHR domain encompasses 13 amino acid residues [including those residues in the juxtamembrane portions of the TM1 and TM2 (Nakamura et al., 1998a, Ulloa-Aguirre and Timossi, 1998; Dias et al., 2002)] with 100% amino acid sequence homology between the human and the rat FSHRs. Replacement of all but one of the serine and threonine residues present in the iL1 of the rat FSHR did not affect agonist induced cAMP accumulation but completely abolished agonist-induced, cAMP-independent protein kinase phosphorylation and impaired agonist-induced uncoupling and arrestin 3-associated receptor internalization (Nakamura et al., 1998a,b; Krishnamurthy et al., 2003). Apparently, these mutations not only impair iL1 phosphorylation but also result in a more generalized conformational change that prevents receptor phosphorylation. Interesting, the triple rat FSHR iL1 mutation (T³⁶⁹I/S³⁷¹I, and T³⁷⁶N) also resulted in constitutive activation of the receptor, whereas replacement of T³⁷⁰ (T³⁷⁸ in the hFSHR) prevented cell surface expression (Nakamura et al., 1998a). As noticed previously (Perez et al., 1996; Hwa et al., 1997; Nechamen and Dias, 2003, Ulloa-Aguirre et al., 2004), single mutations may result in the production of misfolded, trafficking-defective proteins or in constitutively activated conformers. In the case of the constitutively active rFSHR iL1 mutant, the structural alteration may disturb necessary interhelical and/or intracellular loop interactions required for dampening of basal activity.

Intracellular loop 2

The importance of the iL2 in signal transduction has been documented for a number of GPCRs (Ulloa-Aguirre and Conn, 1998). In fact, peptides mimicking the iL2 sequence of some GPCRs, may efficiently block receptor-G protein coupling (Varrault et al., 1994; Qian et al., 1998). In particular, the Glu/Asp-Arg-Tyr motif (E/DRY motif, permutated to ERW in the TSH, LH and FSH receptors) present in the majority of rhodopsin-like receptors has been implicated in G protein interaction and receptor activation (Gersherngorn and Osman, 2001). Mutation of the central R has resulted in either partial or complete abolition of agonist-stimulated signal transduction (Chazenbalk et al., 1991; Franke et al., 1992; Rosenthal et al., 1993; Zhu et al., 1994; Jones et al., 1995; Acharya and Karnik, 1996; Arora et al., 1997; Seibold et al., 1998; Schultz et al., 1999) or in variable degrees of constitutive activation of the altered receptor, depending on the nature of the replacing amino acid residue (Fanelli et al., 1999; Scheer et al., 2000). Other residues, such as hydrophobic leucines located in the middle of the iL2 (Moro et al., 1993; Arora et al., 1995) and residues present in the carboxyl-terminal end (Okamoto and Nishimoto, 1992; McClue et al., 1994; Burnstein et al., 1998) (Fig. 1A) of several GPCRs, including the LH and TSH receptors (Chazenbalk et al., 1990; Schultz et al., 1999), have been also implicated in G protein activation.

Figure 1.

Schematic of the second (iL2) (A) and third (iL3) (B) intracellular loops, and the carboxyl-terminus (Ctail) (C) of the human FSH receptor. The sequence alignment of the corresponding domains in other glycoprotein hormone receptors is also shown; conserved amino acids analyzed in the human FSHR are shaded in grey. Brackets in the carboxyl-terminal end of the iL3 and the amino-terminal end of the Ctail (B and C) delimit the BBXXB motif reversed which has been shown to be associated with G protein activation in other G protein-coupled receptors (Okamoto and Nishimoto, 1992). The square in the amino-terminal end of the Ctail delimits the F(x)₆LL motif required for receptor transport to the cell surface membrane (Duvernay et al., 2004). Figure 1A is reproduced from Timossi et al. (2002), with permission from Elsevier Ireland Ltd.

Transient transfection of several plasmids containing a complementary cDNA construct encoding either the free Wt hFSHR iL2 or the free hFSHR iL2 mutated at R⁴⁵⁰, T⁴⁵³ [a potentially phosphorylable residue in the Wt rat FSHR (Nakamura et al., 1998b)] and/or L⁴⁶⁰ to human embryonic kidney 293 (HEK-293) cells stably expressing the hFSHR, resulted in distinct effects on FSH-stimulated intracellular signaling. Expression of the free Wt iL2 as well as the R⁴⁵⁰A, R⁴⁵⁰K and L⁴⁶⁰A free iL2 mutants led to a significant (∼40-50%) attenuation in maximum FSH-stimulated cAMP accumulation, whereas expression of R⁴⁵⁰H, T⁴⁵³A and R⁴⁵⁰A/T⁴⁵³A/L⁴⁶0A free iL2 mutants did not affect FSH-provoked intracellular signaling, thus suggesting a specific role of iL2 residues in hFSHR function (Timossi et al., 2002). Further studies on modified hFSHRs corroborated the role of particular iL2 residues in receptor function. Agonist-stimulated cAMP production was virtually abolished by changing the hydrophilic R⁴⁵⁰ to A or H (a basically charged residue with a more rigid side chain) in the hFSHR, whereas the conservative substitution R? K did not affect FSH-stimulated intracellular signaling (Fig. 2A) (Timossi et al., 2002). None of the mutations introduced altered either cell surface membrane expression or ligand binding affinity of the modified hFSHR. These observations are in line with the results of studies on the related LH receptor, in which similar substitutions impaired the ability of the mutants to mediate a signaling response (Dhanawada et al., 1996; Schultz et al., 1999), indicating that the mechanisms underlying gonadotropin receptor activation through residues located in the amino-terminal end of the iL2 appear to be similar. Thus, a flexible, positively charged residue at this position is apparently required in both gonadotropin receptors to allow the iL2 for proper interactions with other receptor domains that may yield an active receptor configuration upon activation by agonist. In agreement with studies on altered rat FSHRs (Nakamura et al., 1998b), replacement of T for A at position 453 of the hFSHR virtually abolished the ability of the mutated receptor to mediate FSH-provoked cAMP accumulation, despite proper membrane expression (Fig. 2B) (Timossi et al., 2002). It is possible that this particular residue may be important to form the active structure of the FSHR via interaction with other iL2 residues, as suggested for bovine rhodopsin (Yamashita et al., 2000) in which replacement of those NH₂-terminal iL2 amino acid residues expected to cluster on the same face forming the active structure of rhodopsin, prevented G protein activation.

Figure 2.

Concentration-dependent FSH-stimulated cAMP accumulation (minus basal cAMP concentrations) in HEK-293 cells expressing the Wt and the R⁴⁵⁰A, R⁴⁵⁰K, R⁴⁵⁰H (A), and T⁴⁵³A (B) mutant hFSHRs. The inset shows the relative binding of the mutant hFSHR receptors. Reproduced from Timossi et al. (2002), with permission from Elsevier Ireland Ltd.

The importance of the iL2 on FSHR activation is emphasized by studies on mutant hFSHRs bearing replacements at the conserved L⁴⁶⁰ residue; upon expression in HEK-293 cells, the L⁴⁶⁰D, L⁴⁶⁰P and L⁴⁶⁰A mutants exhibited variable degrees of constitutive activation and altered responses to agonist stimulation; replacement for K or R did not affect basal cAMP production but blunted or reduced the receptor response to further FSH stimulation (Fig. 3) (Timossi et al., 2002). Thus, it seems that disruption of the bonds stabilized by the hydrophobic L⁴⁶⁰ may either promote varying degrees of receptor isomerization to an active, unconstraint state or compromise receptor/G_s-protein activation, depending on the nature of the replacing amino acid residue. These findings strongly suggest that the iL2 of the FSHR may actually act as a conformational switch for allowing the receptor to adopt an active conformation upon agonist stimulation, rather than as a G protein recognition domain.

Figure 3.

A: Concentration-dependent FSH-stimulated cAMP accumulation in HEK-293 cells expressing the Wt and the L⁴⁶⁰A constitutively active hFSHR; bars: basal cAMP production. B: Basal cAMP accumulation in HEK-293 expressing the Wt and the L⁴⁶⁰D, L⁴⁶⁰P, L⁴⁶⁰K, L⁴⁶⁰R, and L⁴⁶⁰A mutant hFSHRs. Black-filled bars: constitutively active mutants. *P <0.05 vs. Wt. C: FSH-stimulated cAMP accumulation (minus basal concentrations) in HEK-293 cells expressing the Wt and the L⁴⁶⁰ mutated hFSHRs. The inset shows the relative binding of the mutant hFSHR receptors. The results are representative of three separate experiments (A and C) or represent the mean ± SD of three independent experiments (B and inset). Reproduced from Timossi et al. (2002), with permission from Elsevier Ireland Ltd.

Intracellular loop 3

Several structural determinants present in this loop govern receptor cell surface residency and receptor-G protein interactions. In fact, this loop is considered pivotal in attenuation of agonist-stimulated, receptor-mediated intracellular signaling via internalization (Nakamura et al., 1998b; Krishnamurthy et al., 2003; Bhaskaran et al., 2003; Cohen et al., 2003). Naturally occurring point mutations and studies employing synthetic peptide strategies have provided some information on the potential role of this intracytoplasmic region in FSHR-mediated signal transduction. A point mutation involving the highly conserved basic residue R⁵⁵⁶ (Fig. 1B) led to decreased FSH-stimulated cAMP accumulation (Beau et al., 1998), while mutations in the conserved aspartate (D⁵⁵⁰) provoked constitutive activation of the hFSHR probably by disrupting the helix capping structure that presumably helps to maintain the receptor in an inactive conformation (Gromoll et al., 1996; Schultz et al., 2000; Haywood et al., 2002). Studies using synthetic peptides homologous to the rat FSHR iL3 showed that the peptide corresponding to residues 533 to 555 promoted guanine nucleotide exchange in testes light membranes and reduced FSH-stimulated cAMP production and estrogen production from primary Sertoli cell cultures (Grasso et al., 1995b); a more restricted peptide (residues 551 to 555) bearing the BBXXB motif reversed (BXXBB, where B is a basic amino acid) involved in G protein activation of several cell surface membrane receptors (Okamoto and Nishimoto, 1992; Wu et al., 1995; Pauwels et al., 1999, 2001; Cheung et al., 1991; Murthy and Makhlouf, 1999), provoked similar effects (Grasso et al., 1995a).

The role of the BXXBB motif present in the juxtamembrane region of the hFSHR on receptor function has been recently analyzed in more detail in our laboratories (Cohen et al., 2003; Timossi et al., 2004). Replacement of the three basic residues by alanine yielded a =62 kDa under glycosylated form of the receptor, which was unable to bind agonist and activate effector (Figs. 4A and D). Nevertheless, the functional role of each basic residue appears to differ depending on its position within the minimal motif. As observed in the LH and TSH receptors (Chazenbalk et al., 1991; Schultz et al., 2000), replacement of K⁵⁵⁵ (corresponding to K⁵⁵² and K⁶²⁴ in the LH and TSH receptors) with alanine did not alter signal transduction (Fig. 4B). Nevertheless, this particular hFSHR mutant exhibited an increased turnover probably due to increased ubiquitination of the altered receptor (Cohen et al., 2003; Timossi et al., 2004). In contrast, mutation of R⁵⁵² or R⁵⁵⁶ profoundly altered intracellular signaling (Fig. 4B) without affecting the ligand binding ability (Fig. 4E) or internalization rate of the receptor (Timossi et al., 2004). These latter effects clearly differ from the scenario prevailing in the other glycoprotein receptors since replacement of the basic K⁵⁶⁶ in the hLHR or K⁶²¹ in the hTSHR (corresponding to R⁵⁵² in the hFSHR), had no effect on agonist-provoked cAMP accumulation, whereas substitution of the comparable hFSHR R⁵⁵⁶ residue in these receptors led to either partial (in the hLHR) or complete (in the hTSHR) functional impairment (Chazenbalk et al., 1991; Schulz et al., 2000). Further, in the human FSH and TSH receptors, simultaneous substitution of all three basic residues completely abolished agonist-stimulated cAMP production and severely impaired intracellular trafficking and membrane expression of the modified receptors (Chazenbalk et al., 1990; Timossi et al., 2004), whereas in the rat LHR (92% homology with the human counterpart in the serpentine region of the receptor) simultaneous replacement of two out of the three cationic residues present in this motif had no functional consequence (Wang et al., 1993). Therefore, the ultimate basic residue of the BXXBB motif in all glycoprotein receptors appears to play a major role in the interaction of the receptor with the G_s protein. In the hFSHR, this motif may be one of the sites involved in receptor-G_s protein coupling (Strader et al., 1987; Cheung et al., 1992; Wu et al., 1995; Kohen et al., 2001) or alternatively it may be important to preserve the packing and interactions of the transmembrane domains 5 and 6 (Kudo et al., 1996), thereby favoring the accessibility of the G protein coupling sites or the formation of an active conformation upon agonist binding.

Figure 4.

Concentration-dependent basal and FSH-stimulated cAMP accumulation (A), specific [¹²⁵I]-labeled FSH binding (D) and Western blot analysis of hFSHR expression (inset) in HEK-293 cells expressing the Wt hFSHR and the mutant hFSHR in which all basic residues present in the BXXBB motif at the iL3 (iL3-AXXAA) or the Ctail (Ctail-AXXAA) were replaced with alanine. B and E show the functional analysis of HEK-293 cells expressing the Wt hFSHR or the iL3 hFSHR mutants in which R⁵⁵² (iL3-AXXBB), K⁵⁵⁵ (iL3-BXXAB) or R⁵⁵⁶ (iL3-BXXBA) were replaced with alanine. C and F show the concentration-dependent basal and FSH-stimulated cAMP accumulation, specific agonist binding, and Western blot analysis of hFSHR expression (inset) in HEK-293 cells expressing either the Wt hFSHR or the Ctail hFSHR mutants in which K⁶¹⁴ (Ctail-AXXBB), R⁶¹⁷ (Ctail-BXXAB) or R⁶¹⁸(Ctail-BXXBA) were replaced with alanine. A-C are representative of 3-4 independent experiments, whereas D-F are the mean ± S.E.M. from three experiments in triplicate incubations. Reproduced from Timossi et al. (2004), with permission from Elsevier Ireland Ltd.

Carboxyl-terminal domain

Several structural and functional features characterize this particular hFSHR domain (Fig. 1C): i) The domain is rich in serine and threonine residues which may be potential sites for phosphorylation; nevertheless, studies on its rat counterpart have clearly shown that in contrast to the LHR, the Ctail of the FSHR is not phosphorylated (Nakamura et al., 1995a,b); ii) The hFSHR Ctail contains the minimal BBXXB motif reversed in its juxtamembrane region, whereas the rat receptor does not bear this motif but rather a BBXB motif [also involved in G protein coupling of other receptors (Okamoto and Nishimoto 1992)] located in the segment encompassed by residues 650-653. The last two residues of the BXXBB motif (R⁶¹⁷ and R⁶¹⁸) and the preceding F⁶¹⁶ constitute the NH₂ end of the highly conserved F(x_s)LL motif required for receptor transport to the cell membrane (Duvernay et al., 2004); iii) The domain exhibits at least 2 cysteine residues (at positions 627 and 629) which are potential sites for palmitoylation; however, evidence that this domain is esterified by palmitoylation is lacking. Two of the 3 cysteine residues present in the hFSHR (C⁶²⁹ and C⁶⁵⁵) are also present in the LHR but not in the TSHR (Fig. 1C).

The hFSHR Ctail-BXXBB motif appears to be more important for receptor expression than for G protein coupling, as suggested by studies in which the basic residues present in this motif were replaced by alanine (Timossi et al., 2004). Substitution of K⁶¹⁴ reduced both ligand binding (Fig. 4F) and membrane localization of the receptor (insert of Fig. 4B) to ∼60% of the levels exhibited by the Wt receptor; consequently, cAMP production was also reduced (Fig. 4C). Meanwhile, replacement of R⁶¹⁷ or R⁶¹⁸ further reduced the ligand binding capacity and membrane expression of the mutant receptor, and virtually abolished FSH-stimulated cAMP production. Thus, mutations in these latter basic residues may provoke conformational changes in the NH₂-terminal end of the Ctail, which includes the above mentioned F(x)₆LL motif (F⁶¹⁶, L⁶²³, and L⁶²⁴), and thereby impair both receptor trafficking and cell membrane localization of the receptor.

Generation of a hFSHR Ctail structural model yielded a family of nine all-a helix globule-like models showing two different topologies that preserve the spatial arrangement of the helical structures. The predicted structures showed a bundle of a-helices connected by loop regions; as other all-α proteins, these a helices are packed pairwise against each other in an antiparallel fashion so that each helix provides the hydrophobic side chains for packing interactions (Branden and Tooze, 1991). The best-ranked model is composed of a three-a helices bundle connected by loop regions and a large loop of non-conventional secondary structure, and is completed by a small a helix non packed to the other a helices (Fig. 5). The region of non-conventional structure (residues 648 to 670) was also predicted to be a natively disordered region by the program DISOPRED2 (Jones and Ward, 2003; Ward et al., 2004). These natively unstructured regions have been described in many binding, assembler, scavenger and regulatory proteins (Tompa, 2002). Thus, this region in the hFSHR may participate in G protein recognition and binding. In nine of the models, included the best-ranked, the loop containing Cys⁶²⁹ is in close contact with the loop containing Cys⁶⁵⁵. Although the sulfur atom of Cys⁶²⁹ is at only 3.08 Å from the 2Hβ atom of Cys⁶⁵⁵, the formation of a disulfide bond between these cysteine residues would be otherwise extremely rare due to the reducing environment inside cells. Nevertheless, cysteine residues may play a key structural role within the intracellular environment; in GPCRs, cysteines in the Ctail are particularly important since they are potential sites for receptor palmitoylation. Among some other functions, this lipid modification appears necessary to form a fourth cytoplasmic loop (Ulloa-Aguirre and Conn, 1998). In some GPCRs, such as the β₂-adrenergic receptor, mutation of the palmitoylated cysteine leads to greatly decreased interaction of the mutant receptor with the G_s protein (O'Dowd et al., 1989), whereas in others it has little or no effect in receptor function (van Koppen et al., 1991; Kennedy et al., 1993). As the TSH and LH receptors, the hFSHR exhibits highly conserved cysteine residues with a neighboring basic residue (K⁷⁰⁰, K⁶⁴⁵ and K⁶²⁶ in the hTSHR, hLHR and hFSHR, respectively) that appears to be essential for palmitoylation (Belanger et al., 2001). In fact, it has been shown that the human and rat TSHRs as well as the rat LHR are palmitoylated in their Ctail and that mutation of their corresponding palmitoylated cysteines results in either abnormal trafficking or intracellular trapping of the mutant receptor (Zhu et al., 1995; Kosugi and Mori, 1996; Tanaka et al., 1998). In the hFSHR, replacement of Cys⁶²⁷ with alanine did not alter receptor function, agonist binding or membrane surface expression, whereas alanine substitution at Cys⁶²⁹ or at both Cys⁶²⁷ and Cys⁶²⁹ decreased FSH-stimulated cAMP accumulation by 20 to 30% (Fig. 6), and impaired both agonist binding and receptor expression (not shown). Meanwhile, replacement of Cys⁶⁵⁵ by alanine but not by threonine or serine, decreased agonist-stimulated cAMP production without altering cell surface membrane expression. These results suggests that only Cys⁶²⁹ plays an important role in hFSHR function and that as shown for the LHR (Zhu et al., 1995), this particular residue may be involved in receptor palmitoylation and cell surface membrane expression.

Figure 5.

Computational model of the hFSHR Ctail as disclosed by de novo modeling. The model was generated on Robetta server (Kim et al., 2004), which is based on the Rosetta software for protein prediction (Bradley et al., 2003; Simons et al., 1997, 1999). Robetta generates three- and nine-residue fragment libraries representing local conformations of amino acids as found in the RCSB Protein Data Bank. This library is used to assemble models by fragment insertion using a scoring function. The program eliminates the models showing too many local contacts and unlikely strand topologies, producing the best ranking models based on its scoring function. These models are finally compared with chains in the protein data bank to produce a Z-score, which represents the probability of getting a similar-length match by chance. The side chains of the Ctail hFSHR models were repacked using a Monte Carlo algorithm based on a back-bone-dependent side-chain rotamer library. The best-ranked model, with the three cysteine residues (Cys ⁶²⁷, Cys⁶²⁹ and Cys⁶⁵⁵) in color sticks is shown.

Figure 6.

Concentration-dependent basal and FSH-stimulated cAMP production (representative of 3 independent experiments), and percent maximal cAMP production (inset) (mean ± S.E.M. from 3 experiments) in cells transiently expressing the Wt hFSHR or the Ctail-cysteine mutant hFSHRs.

Conclusions

We have summarized herein some recent information on the role of the intracytoplasmic domains of the hFSHR on G protein coupling and receptor expression. Although the hFSHR preferentially couples to the G_sa-subunit, there is some experimental evidence suggesting that the FSHR may additionally signal through the pertusis toxin-sensitive G_i/o– mediated pathways (Eskola et al., 1994; Arey et al., 1997; Timossi et al., 1998). In this regard, alternative spliced variants of the receptor may be one of the mechanisms by which particular intracytoplasmic domains of the FSHR may signal through these G proteins (Sairam et al., 1996). As mentioned above, the FSHR also signals through cAMP- dependent, but PKA-independent alternate signaling cascades (Richards et al., 2002); in addition, it has been shown that the adapter protein 14-3-3τ, a member of the 14-3-3 protein family which play a key role in signal transduction pathways, cell division and apoptosis (Tzivion and Avruch, 2002), interacts with the iL2 of the hFSHR, suggesting a role for this cytoplasmic protein in FSH-mediated cAMP independent signaling (Cohen et al., 2004). APPL1, another adapter protein that interacts with the p110a catalytic subunit of PI3K and with inactive Akt (Mitsuuchi et al., 1999), has been more recently identified as an hFSHR iL1-interacting partner, providing a potential link between the FSHR and the PI3K/Akt signaling pathway (Nechamen et al., 2004). One thus may envision a complex system of multiple, pleiotropic signals triggered by the activated FSHR, in which compartmentalization and oligomerization of particular receptor populations may potentially play fundamental roles.