点击显示 收起
Pharmazentrum, University of Basel, Department of Clinical-Biological Sciences, Institute of Physiology, Basel, Switzerland (M.G., C.H., Y.S., B.B., S.A.A., B.B.)
Novartis Institutes for Biomedical Research, Novartis Pharma AG, Basel, Switzerland (J.M., K.K.)
Abstract
Functional -aminobutyric acid type B (GABAB) receptors are normally only observed upon coexpression of GABAB1 with GABAB2 subunits. A C-terminal arginine-based endoplasmic reticulum (ER) retention/retrieval signal, RSRR, prevents escape of unassembled GABAB1 subunits from the ER and restricts surface expression to correctly assembled heteromeric receptors. The RSRR signal in GABAB1 is proposed to be shielded by C-terminal coiled-coil interaction of the GABAB1 with the GABAB2 subunit. Here, we investigated whether the RSRR motif in GABAB1 remains functional when grafted to ectopic sites. We found that the RSRR signal in GABAB1 is inactive in any of the three intracellular loops but remains functional when moved within the distal zone of the C-terminal tail. C-terminal deletions that position the RSRR signal closer to the plasma membrane drastically reduce its effectiveness, supporting that proximity to the membrane restricts access to the RSRR motif. Functional ectopic RSRR signals in GABAB1 are efficiently inactivated by the GABAB2 subunit in the absence of coiled-coil dimerization, supporting that coiled-coil interaction is not critical for release of the receptor complex from the ER. The data are consistent with a model in which removal of RSRR from its active zone rather than its direct shielding by coiled-coil dimerization triggers forward trafficking. Because arginine-based intracellular retention signals of the type RXR, where X represents any amino acid, are used to regulate assembly and surface transport of several multimeric complexes, such a mechanism may apply to other proteins as well.
GABAB receptors are the G protein-coupled receptors for GABA, the predominant inhibitory neurotransmitter in the mammalian central nervous system. GABAB receptors modulate synaptic transmission by controlling neurotransmitter release and by causing postsynaptic hyperpolarization (Bowery et al., 2002; Calver et al., 2002; Bettler et al., 2004). They are broadly expressed in the nervous system and have been implicated in a variety of neurological and psychiatric conditions. In heterologous cells, functional GABAB receptors are usually only observed upon coexpression of GABAB1 with GABAB2 subunits, which provided compelling evidence for heteromerization among G protein-coupled receptors (Kaupmann et al., 1997, 1998; Jones et al., 1998; White et al., 1998; Kuner et al., 1999; Marshall et al., 1999; Ng et al., 1999). Two GABAB1 subunit isoforms, GABAB1a and GABAB1b, arise from the GABAB1 gene by differential promoter use (Kaupmann et al., 1997; Bettler et al., 2004). The data therefore support the existence of two predominant GABAB receptors in the nervous system, the heteromeric GABAB(1a,2) and GABAB(1b,2) receptors. However, knockout studies also suggest that GABAB1a and GABAB1b could be functional in neurons that naturally lack GABAB2 expression (Gassmann et al., 2004).
In the GABAB heteromer, the GABAB1 subunit binds GABA and all competitive GABAB ligands (Kaupmann et al., 1998), whereas the GABAB2 subunit is predominantly responsible for activating the G protein (Galvez et al., 2001; Margeta-Mitrovic et al., 2001; Robbins et al., 2001; Grunewald et al., 2002; Havlickova et al., 2002). Trafficking of unassembled GABAB1 subunits to the plasma membrane is prevented by an arginine-based ER-retention/retrieval signal, the four amino acids RSRR, in the cytoplasmic tail of GABAB1 (Couve et al., 1998; Margeta-Mitrovic et al., 2000; Pagano et al., 2001). This ER-retention/retrieval signal is proposed to be shielded by C-terminal coiled-coil interaction of the GABAB1 with the GABAB2 subunit. Within the RSRR motif the serine residue and the third arginine are not absolutely critical for function, because they can be substituted by other amino acids (Margeta-Mitrovic et al., 2000; Pagano et al., 2001). More recently, it was shown that the sequence context of the RSRR signal in GABAB1 influences its function (Grunewald et al., 2002). Thus, the full ER-retention/retrieval motif in GABAB1 was extended to the sequence QLQSRQQLRSRR, which includes part of the coiled-coil domain. Arginine-based ER-retention/retrieval signals were observed in a number of other multisubunit proteins [e.g., the KATP channels (Zerangue et al., 1999) and N-methyl-D-aspartate receptors (Scott et al., 2001)], where they control stoichiometry and surface expression of the channel complex. From the available data, it emerges that the core ER-retention/retrieval motif is RXR, consisting of two arginines that are separated by any amino acid (X).
Dilysine ER-retention/retrieval signals require a strict spacing relative to the C terminus. In contrast to KK-signals, functional RXR signals are found in a variety of cytosolic positions, including intracellular loops and the N and C termini in type II and type I membrane proteins, respectively (Schutze et al., 1994; Zerangue et al., 1999). This broad distribution initially suggested that many proteins that harbor the consensus sequence RXR are retained in the ER. This was recently challenged in a study that showed that the RXR-dependent ER-retention/retrieval machinery is sensitive to the length of the spacer that separates the RXR motif and the receptor-anchored membrane (Shikano and Li, 2003). Here, we studied whether the RSRR signal in GABAB1 can still function when grafted to ectopic cytoplasmic positions and whether it can be masked by GABAB2 regardless of its position. The data let us propose a new mechanism to explain RSRR inactivation upon GABAB subunit dimerization.
Materials and Methods
Generation of Mutant Expression Plasmids. All constructs were subcloned into the cytomegalovirus-based eukaryotic expression vector pCI (Promega, Madison, WI). Overlap extension polymerase chain reaction (Horton et al., 1990) was used to introduce ectopic RSRR and LRSRR motifs into a GABAB1a mutant (R1) where the endogenous RSRR was inactivated by substitution of arginine with alanine residues (Pagano et al., 2001). Overlap extension polymerase chain reaction was also used to construct GABAB1a deletion mutants, leaving the wild-type RSRR unchanged.
Cell Surface Labeling. HEK293 cells for transient transfection of expression constructs were purchased from the American Type Culture Collection (Manassas, VA) and cultured in Dulbecco's minimum Eagle's medium (Invitrogen, Basel, Switzerland) supplemented with 10% fetal calf serum and 2 mM L-glutamine. The photoaffinity ligand [125I]CGP71872 specifically binds to the GABA-binding site of GABAB1 subunits and does not permeate the plasma membrane (Pagano et al., 2001). [125I]CGP71872 labeling of intact cells therefore reveals GABAB1 protein at the cell surface, whereas labeling of lysed cells reveals total GABAB1 protein, independent of where in the biosynthetic pathway it is present. Six hours after transfection of expression plasmids using Lipofectamine 2000 transfection reagent (Invitrogen), HEK293 cells were transferred to six-well plates. After an additional 24-h incubation, cells were washed twice with ice-cold HEPES, pH 7.6. Half of the cells were then used for photoaffinity labeling of surface receptors (S in Figs. 2, 6, and 7), and the other half were used for labeling of total receptors in the cell homogenates (H in Figs. 2, 6, and 7). For surface labeling, intact cells were incubated in the dark for 1 h at room temperature with 0.8 nM [125I]CGP71872. Thereafter, cells were washed twice with ice-cold Krebs-Tris buffer (118 mM NaCl, 4.7 mM KCl, 1.8 mM CaCl2, 1.2 mM MgSO4, 1.2 mM KH2PO4, 5.6 mM glucose, and 20 mM Tris-Cl, pH 7.4) to remove unbound ligand. Bound [125I]CGP71872 was cross-linked to the receptor using UV light (Kaupmann et al., 1997). Photoaffinity-labeled cells were then harvested, and the radioactivity was determined in a gamma counter (PerkinElmer Life and Analytical Sciences, Zurich, Switzerland). For [125I]CGP71872 labeling of total GABAB1 protein, cells were harvested and lysed before incubation with the photoaffinity ligand. Preparation of lysates and [125I]CGP71872 binding was as described previously (Kaupmann et al., 1997). For 10% SDS-PAGE, cell pellets and homogenates were resuspended in Krebs-Tris buffer containing 0.1% SDS. An aliquot was used for determination of protein concentration (Micro Protein Assay; Bio-Rad, Munich, Germany). Equal amounts of total protein were used when comparing S receptors and total receptors in cell Hs. We normalized the input of radiolabeled protein in the SDS-PAGE by using equal counts of the H samples for each set of transfections (expression with and without GABAB2). Photoaffinity-labeled protein was detected using autoradiography. The S/H ratio of the radioactivity incorporated into the cell surface and the homogenate fraction was determined from the autoradiograms. Because of the differences in the radiolabeling procedure for surface and homogenate receptors, the percentage S/H sometimes exceeds the theoretical value of 100%. Loading was controlled for by Western blot analysis with the polyclonal GABAB1 antibody Ab174.1 that is directed against the C-terminal tail of GABAB1 (Malitschek et al., 1998). Surface labeling with [125I]CGP71872 was compared with surface biotinylation (Fig. 3). For the biotinylation experiments, we used membrane-impermeable EZ-link Sulfo-NHS-SS-biotin (Pierce Chemical, Rockford IL). Forty-eight hours after transfection, HEK293 cells were washed three times in PBS and then incubated with 1 mg/ml Sulfo-NHS-SS-biotin for 30 min at 4°C on a rocking table. To quench the biotinylation reaction, the cells were then washed in PBS and incubated in 50 mM glycine in PBS for 5 min. After three washes in PBS, the cells were lysed in radioimmunoprecipitation assay buffer (150 mM NaCl, 1% Nonidet-40, 0.5% sodium deoxycholate, 0.1% SDS, and 50 mM Tris-Cl, pH 7.5). The lysates were cleared by centrifugation at 10,000g for 10 min. Aliquots were taken and mixed with 2x SDS loading buffer to detect total GABAB1 protein expressed. The remaining cleared lysates were incubated with avidin beads (Pierce Chemical) at 4°C overnight. After five washes in radioimmunoprecipitation assay buffer, biotinylated proteins were eluted from the avidin beads using SDS loading buffer. Finally, total and eluted GABAB1 proteins were separated on SDS-PAGE and analyzed on Western blots.
Western Blots. After SDS-PAGE, proteins were blotted onto a nitrocellulose membrane (Immobilon-P; Millipore Corporation, Billerica, MA) by standard electrophoretic transfer. After blotting, the membrane was blocked with 5% nonfat milk powder in PBS for 1 h at room temperature. Rabbit antiserum Ab174.1 (1:2500; Malitschek et al., 1998), the monoclonal anti--tubulin antibody MAB3408 (1: 500; Chemicon International, Temecula, CA), and peroxidase-coupled secondary antibodies (donkey anti-rabbit or anti-mouse conjugates, 1:2500; Amersham Biosciences UK, Ltd., Little Chalfont, Buckinghamshire, UK) were incubated in PBS containing 2.5% nonfat milk powder and 0.1% Tween 20 for 1 h at room temperature. After antibody incubation, three wash steps with PBS containing 0.1% Tween 20 were carried out for 10 min. The blots were developed using the enhanced chemiluminescence chemiluminescent detection system (Amersham Biosciences UK, Ltd.) and exposed to Kodak Bio-Max maximum resolution X-ray films (Sigma-Aldrich, St. Louis, MO).
Fluorimetric Measurement of Changes in the Intracellular Ca2+ Concentration ([Ca2+]i). For measurement of [Ca2+]i, all transfections included GqzIC to artificially couple GABAB receptors to PLC (Franek et al., 1999). Transfected HEK293 cells were plated into poly-D-lysine-coated 96-well plates (BD Biosciences, Erembodegem, Belgium). After transfection (48eC72 h), cells were loaded for 45 min with 2 e fluo-4 acetoxymethyl ester (Molecular Probes, Leiden, The Netherlands) in Hanks' balanced salt solution (Invitrogen) supplemented with 20 mM HEPES buffer and 50 e probenecid (Sigma, Buchs, Switzerland). Plates were washed and transferred to a fluorimetric image plate reader (Molecular Devices, Crawley, UK). Fluorescence changes F upon addition of GABA (final concentration of 0.1 mM) were recorded as a function of time, as described previously (Pagano et al., 2001). No quantitative comparison between experiments was made, because the signal amplitude depends on the transfection efficiency.
Results
Generation and Characterization of GABAB1 Mutants with Ectopic RSRR Signals. To study whether ectopic RSRR motifs are functional in GABAB1, we introduced the RSRR motif into a GABAB1 protein where the endogenous RSRR motif is inactivated by substitution of arginine with alanine residues. This protein, R1, is efficiently transported to the cell surface in the absence of GABAB2 (Pagano et al., 2001). Whenever possible, we inserted the RSRR motif at positions that already harbored an arginine or a serine residue, which is expected to minimize interference with the wild-type amino acid sequence. A scheme depicting the insertion sites of ectopic RSRR motifs in R1 is shown in Fig. 1A. The positions of the ectopic RSRR motifs in the primary sequence of GABAB1a are listed in Fig. 1B. We confirmed expression of mutant GABAB1 proteins in transiently transfected HEK293 cells by Western blot analysis, using an antibody directed against a C-terminal epitope (Fig. 1C). In general, the expression levels of mutant GABAB1 proteins are comparable with those of the wild-type GABAB1a (R1) and R1 proteins (Fig. 1C, top). The only exception is R1[R862SRR], which harbors the ectopic RSRR motif in the C-terminal tail and for unknown reasons is poorly expressed. On the other hand, it is also possible that some of the C-terminal epitopes in R1[R862SRR] are affected by the mutation and are no longer recognized by the antibody. Equal loading was controlled for by Western blot analysis with a -tubulin antibody (Fig. 1C, bottom).
RSRR Remains Functional at the C Terminus but Not in Any of the Intracellular Loops. To examine the functionality of ectopic RSRR motifs, we expressed GABAB1 mutants either in isolation or together with GABAB2. We determined the ratio of surface and total GABAB1 protein levels by photoaffinity labeling of intact and lysed cells, respectively, with the membrane-impermeable antagonist [125I]CGP71872. After SDS-PAGE, labeled proteins were visualized by autoradiography. We consistently observed that wild-type and mutant GABAB1 proteins bind significantly more [125I]CGP71872 when coexpressed with GABAB2, suggesting that GABAB2 assists GABAB1 in reaching a binding-competent conformation. To correct for this as well as variability in transfection efficiency, the amount of protein sample subjected to gel electrophoresis was normalized to the respective amount of radioactivity incorporated into the cell Hs. Therefore, for the reason mentioned above, substantially less immunostained GABAB1 protein is seen on all Western blots of samples where GABAB2 was coexpressed (Fig. 2). For each transfection, photoaffinity-labeled GABAB1 protein at the cell S was compared with total GABAB1 protein labeled in the cell Hs. We investigated whether the binding-incompetent form of GABAB1, which is more abundant in the absence of GABAB2, is able to reach the cell surface. We used biotinylation of intact cells and precipitation with avidin-Sepharose as an alternative method to [125I]CGP71872 labeling to detect proteins expressed at the cell surface (Fig. 3). We failed to detect significant amounts of GABAB1 protein expressed at the cell surface of HEK293 cells transfected with GABAB1 alone (R1), indicating that the binding-incompetent form of GABAB1 fails to reach the cell surface in the absence of GABAB2. This is also supported by recent studies that show that ligand binding is a critical requirement for plasma membrane expression (Mah et al., 2005; Valluru et al., 2005). In all our experiments, we therefore used photoaffinity labeling with [125I]CGP71872 to quantify GABAB1 protein at the cell surface.
As shown in Fig. 2, wild-type GABAB1 (R1) is retained in the ER and therefore does not bind the photoaffinity ligand at the cell surface. However, upon coexpression with GABAB2 (R1 + R2), or inactivation of the RSRR motif (R1), GABAB1 is released to the cell surface, in agreement with previous reports (Margeta-Mitrovic et al., 2000; Pagano et al., 2001). Insertion of RSRR motifs into any of the three intracellular loops (mutant proteins R1[RS616RR], R1[RS624RR], R1[RV690RSRR], R1[E699RSRR], and R1[E796RSRR]) failed to confer detectable intracellular retention in our assay. Likewise, mutants with an ectopic RSRR motif in the C-terminal tail at positions R862 (R1[R862SRR]), S877 (R1[RS877RR]), or S917 (R1[RS917RR]) were efficiently transported to the cell surface, no matter whether they were expressed alone or in combination with GABAB2. In contrast, insertion of an ectopic RSRR motif in the C-terminal tail at positions S887 (R1[RS887RR]) and R939 (R1[R939SRR]) resulted in partial intracellular retention. In summary, transposing the RSRR ER-retention/retrieval motif of GABAB1 to ectopic positions indicates that it can be functional in preventing transport to the cell surface in the cytoplasmic tail but not in any of the intracellular loops. Functional RSRR signals are efficiently masked at ectopic sites by heterodimerization with GABAB2, as shown by the release of the R1[RS887RR] and R1[R939SRR] proteins to the cell surface in the presence of GABAB2.
Ectopic RSRR Motifs Do Not Interfere with Receptor Function. The experiments described above show that all GABAB1 subunits with ectopic RSRR motifs can reach the cell surface when coexpressed with GABAB2. This suggests that the mutated GABAB1 proteins fold correctly and assemble into heterodimers. When expressed in heterologous cells, GABAB1 is not functional by itself, even when artificially targeted to the cell surface by inactivation of the RSRR signal or by shielding it with a C-terminal GABAB2 peptide (Margeta-Mitrovic et al., 2000; Pagano et al., 2001). To confirm heteromeric assembly between mutated GABAB1 and wild-type GABAB2 subunits, we examined whether coexpression of the subunits yielded functional receptors. Upon transient coexpression of the subunits with a chimeric G subunit, GqzIC (Franek et al., 1999), in HEK293 cells, we measured GABA-induced increases in intracellular Ca2+ levels by fluorimetry. As illustrated in Fig. 4, all GABAB1 mutants can be activated with 0.1 mM GABA upon coexpression with GABAB2, similarly to wild-type GABAB1 (R1 + R2). Hence, insertion of ectopic RSRR motifs does not interfere with G protein coupling of the mutant proteins.
Appropriate Spacing to the Plasma Membrane Is Required for ER-Retention/Retrieval of GABAB1. RXR-type motifs were proposed to have an operating range and to be sensitive with regard to their spacing from the plasma membrane (Shikano and Li, 2003). This could explain why in GABAB1 ectopic RSRR motifs are only functional when located within the distal C-terminal tail (Fig. 2). Conflicting with this explanation, the ectopic RSRR motif at S917, in between the functional motifs at S887 and R939, is unable to confer intracellular retention (Fig. 2, construct R1[RS917RR]). Small changes in the local sequence context can alter the signal strength of arginine-based ER-retention motifs (Zerangue et al., 2001). For example, the functionality of RXR signals is described to improve when a hydrophobic amino acid, in particular leucine, precedes the arginine cluster. We therefore investigated whether insertion of a leucine preceding the RSRR in R1[RS917RR] rescues intracellular retention. We additionally tested whether including a leucine in the R1[RS887RR] and R1[R939SRR] proteins, which are less well retained than R1, improves retention. Indeed, insertion of a leucine preceding the RSRR at position S917 renders the otherwise nonfunctional ectopic motif functional (Fig. 5, R1[LRS917RR] versus R1[RS917RR]). In contrast, insertion of leucine in R1[RS887RR] or R1[R939SRR] does not improve intracellular retention of these proteins. Intracellular retention of the R1[LRS917RR] protein further supports that the distal cytoplasmic tail has the potential to harbor functional RSRR signals.
We next tested whether the spacing to the plasma membrane affects the functionality of the ER-retention/retrieval motif in GABAB1. To that aim, we constructed three deletion mutants that gradually move the endogenous RSRR motif closer to the plasma membrane (Fig. 6). Deletion of nine amino acid residues has no effect on the functionality of the RSRR motif, whereas deletion of 30 or 52 amino acids increasingly boosts cell surface expression of GABAB1. This gradual increase in surface expression clearly shows that the spacing to the plasma membrane is critical for RSRR function.
Masking of Ectopic RSRR Signals in GABAB1 Does Not Involve C-Terminal Coiled-Coil Domain Interaction. Two reports indicate that surface trafficking is not entirely dependent on coiled-coil domain interaction between the GABAB1 and GABAB2 subunits (Pagano et al., 2001; Grunewald et al., 2002). For example, GABAB2 mutants lacking the C-terminal coiled-coil domain (R2LZ2) are able to traffic GABAB1 to the cell surface. We therefore investigated whether coiled-coil interaction is necessary for masking the functional ectopic RSRR motifs in R1[RS887RR] and R1[R939SRR] proteins by cotransfecting them with R2LZ2 (Pagano et al., 2001). As shown in Fig. 7 and in agreement with previous reports, R2LZ2 is able to traffic wild-type GABAB1 (R1) to the cell surface, but to a smaller extent than wild-type GABAB2 (R2). Both wild-type GABAB2 and R2LZ2 are able to traffic the R1[RS887RR] and R1[R939SRR] proteins with functional ectopic RSRR motifs to the cell surface. In addition R1[LRS917RR], which is efficiently retained in the absence of GABAB2 (Fig. 5) is translocated to the cell surface by coexpression with R2LZ2 (S/H ratio 69%; not shown). This indicates that coiled-coil domain interaction between the cytoplasmic tails of GABAB1 and GABAB2 is not crucial for masking the ectopic RSRR motifs in the mutant GABAB1 subunits. Additional interaction sites between GABAB1 and GABAB2 obviously mediate heterodimerization and compensate for the lack of coiled-coil domain interaction, thereby presumably preventing the ectopic RSRR motifs from binding to protein(s) that localize it in the ER.
Discussion
The generic membrane trafficking signals RXR and KK are part of quality control mechanisms that prevent incorrectly folded and/or assembled membrane proteins from reaching the cell surface. Signals of the RXR-type are generally used to control assembly of multimeric protein complexes. It is assumed that the RXR motif is masked upon association with an appropriate partner subunit and consequently only correctly assembled complexes are able to exit the ER. In contrast to the carboxyl-terminal dilysine signal KK, which exhibits a strict spacing relative to the C terminus, RXR-type signals are found in a variety of sequence positions. In octameric KATP channels they are localized in the cytoplasmic tail of the pore forming subunit (Kir6.1/2) as well as in a cytoplasmic loop of the regulatory subunit SUR1 (Zerangue et al., 1999). In addition, the related ER-retention/retrieval motif RR was identified in the cytosolic N terminus of the myosin heavy chain class II invariant chain isoform lip33, a type II membrane protein (Schutze et al., 1994). In the experiments presented herein, we transposed the RSRR ER-retention/retrieval signal of GABAB1 from its normal position adjacent of the coiled-coil domain to ectopic positions within the cytoplasmic tail or within the three intracellular loops. We show that the RSRR motif is not functional in any of the intracellular loops but that it is partially functional at two ectopic positions within the cytoplasmic tail (Fig. 2). A previous study suggested that the functionality of the RSRR motif of GABAB1 depends on surrounding sequences (Grunewald et al., 2002). In particular, amino acid residues that are part of the coiled-coil domain and neighbor the RSRR motif N-terminally were proposed to be important for recognition of the RSRR motif. From these previous experiments, it was concluded that the minimal ER retention sequence in GABAB1 is comprised of the amino acids QLQXRQQLRSRR, where X can be either S or D (Grunewald et al., 2002). Our data demonstrate that there is not a strict requirement for the RSRR motif to be in its normal sequence context to be functional, because the motif mediates retention when moved N-terminally of QLQXRQQLRSRR to position S887 (R1[RS887RR]) or C-terminally to position R939 (R1[R939SRR]) (Fig. 2). However, the R1[RS917RR] protein, harboring an RSRR motif positioned in between the motifs in R1[RS887RR] and R1[R939SRR], is not retained. This suggests that the sequence environment and/or the secondary structure of the area where the ectopic RSRR motif has been inserted are nevertheless of some influence. It was proposed that small changes in the local sequence context can alter the signal strength of arginine-based ER-retention motifs and that it is favorable when a hydrophobic amino acid, in particular leucine, precedes the arginine cluster (Zerangue et al., 2001). This sequence configuration is also observed for the ER-retention/retrieval signal in wild-type GABAB1. R1[RS917RR] and the partly retained R1[RS887RR] and R1[R939SRR] proteins violate this rule. Insertion of a leucine preceding the RSRR rescues intracellular retention of R1[RS917RR] but does not increase retention of R1[RS887RR] and R1[RS939RR] (Fig. 5). This reinforces that the local sequence context can influence RSRR functionality and supports that the distal cytoplasmic tail is accessible for intracellular retention at various sites.
It is emerging that different types of ER-retention/retrieval motifs have characteristic operating ranges with respect to the distance to the plasma membrane. Whereas carboxyl-terminal KK motifs are operational proximal to the membrane, RXR-type motifs are most effective at a certain distance away from the intracellular plasma membrane (Shikano and Li, 2003). In our experiments the ectopic RSRR motifs in the intracellular loops may therefore be positioned too close to the plasma membrane to be in the active zone. It is also conceivable that the binding of a putative RSRR-interacting protein involved in ER retention depends on additional sequence elements within GABAB1. Appropriate spacing between the RSRR motif and such additional sequence elements may be lost in GABAB1 proteins with mutations in the intracellular loops. On the other hand, in certain ectopic positions the RSRR motif might be inaccessible because of simple steric hindrance. We show that C-terminal deletions that progressively move the wild-type RSRR motif closer to the membrane gradually reduce its signal strength, favoring that primarily the spacing to the plasma membrane is important for RSRR function (Fig. 6).
Functional ectopic RSRR signals in GABAB1 are efficiently masked by the GABAB2 subunit in the absence of coiled-coil dimerization (Fig. 7). This agrees with previous findings that coiled-coil interaction is not absolutely necessary for shielding (Pagano et al., 2001). The mechanism by which GABAB2 prevents intracellular retention of GABAB1 therefore remains unclear. The data presented herein suggest a model in which global conformational changes associated with heteromeric assembly remove the RSRR signal from the active zone, thereby restricting its access and triggering surface delivery of the complex. COPI and 14-3-3 are prime candidates for regulating aspects of GABAB receptor trafficking. COPI components can interact with arginine-based motifs and compete for binding with proteins of the 14-3-3 family (Yuan et al., 2003). It is thought that 14-3-3 binding overcomes ER-retention by preventing recycling of correctly assembled proteins from the ER-Golgi intermediate compartment to the ER via COP1 vesicles (O'Kelly et al., 2002; Nufer and Hauri, 2003). 14-3-3 proteins are known to associate with the C terminus of GABAB1 through a domain partially overlapping with the coiled-coil domain (Couve et al., 2001). It is conceivable that COP1 components bind to RSRR when GABAB1 in unassembled, which recycles GABAB1 back to the ER. After heteromeric assembly and removal of the RSRR motif from its active zone, COP1 could then be replaced by 14-3-3, which avoids recycling and allows for surface trafficking.
In conclusion, our results support that the RSRR ER-retention/retrieval signal of GABAB1 is only functional within the distal C-terminal tail. Moreover, coiled-coil interaction is not crucial for inactivation of wild-type (Pagano et al., 2001) and ectopic RSRR motifs. In the light of these data, we propose that removal of the RSRR motif from its active zone rather than direct coiled-coil shielding may trigger surface delivery of the receptor complex. On a broader scope, the data suggest that many proteins featuring the RXR consensus sequence in proximity of the membrane escape intracellular retention because the motif does not reach into its operational zone.
Acknowledgements
We thank Dr. A. Pagano for critical comments on the manuscript and advice for the biotinylation experiments and D. Ristig and C. Lampert for excellent technical assistance.
1 M.G., C.H., and Y.S. contributed equally to this work.
doi:10.1124/mol.104.010256.
References
Bettler B, Kaupmann K, Mosbacher J, and Gassmann M (2004) Molecular structure and physiological functions of GABA(B) receptors. Physiol Rev 84: 835eC867.
Bowery NG, Bettler B, Froestl W, Gallagher JP, Marshall F, Raiteri M, Bonner TI, and Enna SJ (2002) International Union of Pharmacology. XXXIII. Mammalian -aminobutyric acidB receptors: structure and function. Pharmacol Rev 54: 247eC264.
Calver AR, Davies CH, and Pangalos MN (2002) GABAB receptors: from monogamy to promiscuity. Neurosignals 11: 299eC314.
Couve A, Filippov AK, Connolly CN, Bettler B, Brown DA, and Moss SJ (1998) Intracellular retention of recombinant GABAB receptors. J Biol Chem 273: 26361eC26367.
Couve A, Kittler JT, Uren JM, Calver AR, Pangalos MN, Walsh FS, and Moss SJ (2001) Association of GABAB receptors and members of the 14-3-3 family of signaling proteins. Mol Cell Neurosci 17: 317eC328.
Franek M, Pagano A, Kaupmann K, Bettler B, Pin JP, and Blahos J (1999) The heteromeric GABAB receptor recognizes G-protein alpha subunit C-termini. Neuropharmacology 38: 1657eC1666.
Galvez T, Duthey B, Kniazeff J, Blahos J, Rovelli G, Bettler B, Prezeau L, and Pin JP (2001) Allosteric interactions between GB1 and GB2 subunits are required for optimal GABAB receptor function. EMBO (Eur Mol Biol Organ) J 20: 2152eC2159.
Gassmann M, Shaban H, Vigot R, Sansig G, Haller C, Barbieri S, Humeau Y, Schuler V, Muller M, Kinzel B, et al. (2004) Redistribution of GABAB(1) protein and atypical GABAB responses in GABAB(2)-deficient mice. J Neurosci 24: 6086eC6097.
Grunewald S, Schupp BJ, Ikeda SR, Kuner R, Steigerwald F, Kornau HC, and Kohr G (2002) Importance of the gamma-aminobutyric acid(B) receptor C-termini for G-protein coupling. Mol Pharmacol 61: 1070eC1080.
Havlickova M, Prezeau L, Duthey B, Bettler B, Pin JP, and Blahos J (2002) The intracellular loops of the GB2 subunit are crucial for G-protein coupling of the heteromeric -aminobutyrate B receptor. Mol Pharmacol 62: 343eC350.
Horton RM, Cai ZL, Ho SN, and Pease LR (1990) Gene splicing by overlap extension: tailor-made genes using the polymerase chain reaction. Biotechniques 8: 528eC535.
Jones KA, Borowsky B, Tamm JA, Craig DA, Durkin MM, Dai M, Yao WJ, Johnson M, Gunwaldsen C, Huang LY, et al. (1998) GABAB receptors function as a heteromeric assembly of the subunits GABAB(R1) and GABAB(R2). Nature (Lond) 396: 674eC679.
Kaupmann K, Huggel K, Heid J, Flor PJ, Bischoff S, Mickel SJ, McMaster G, Angst C, Bittiger H, Froestl W, and Bettler B (1997) Expression cloning of GABAB receptors uncovers similarity to metabotropic glutamate receptors. Nature (Lond) 386: 239eC246.
Kaupmann K, Malitschek B, Schuler V, Heid J, Froestl W, Beck P, Mosbacher J, Bischoff S, Kulik A, Shigemoto R, et al. (1998) GABAB-receptor subtypes assemble into functional heteromeric complexes. Nature (Lond) 396: 683eC687.
Kuner R, Kohr G, Grunewald S, Eisenhardt G, Bach A, and Kornau HC (1999) Role of heteromer formation in GABAB receptor function. Science (Wash DC) 283: 74eC77.
Mah SJ, Cornell E, Mitchell NA, and Fleck MW (2005) Glutamate receptor trafficking: endoplasmic reticulum quality control involves ligand binding and receptor function. J Neurosci 25: 2215eC2225.
Malitschek B, Re筫gg D, Heid J, Kaupmann K, Bittiger H, Frstl W, Bettler B, and Kuhn R (1998) Developmental changes in agonist affinity at GABAB(1) receptor variants in rat brain. Mol Cell Neurosci 12: 56eC64.
Margeta-Mitrovic M, Jan YN, and Jan LY (2000) A trafficking checkpoint controls GABAB receptor heterodimerization. Neuron 27: 97eC106.
Margeta-Mitrovic M, Jan YN, and Jan LY (2001) Function of GB1 and GB2 subunits in G protein coupling of GABAB receptors. Proc Natl Acad Sci USA 98: 14649eC14654.
Marshall FH, Jones KA, Kaupmann K, and Bettler B (1999) GABAB receptors - the first 7TM heterodimers. Trends Pharmacol Sci 20: 396eC399.
Ng GYK, Clark J, Coulombe N, Ethier N, Hebert TE, Sullivan R, Kargman S, Chateauneuf A, Tsukamoto N, McDonald T, et al. (1999) Identification of a GABAB receptor subunit, gb2, required for functional GABAB receptor activity. J Biol Chem 274: 7607eC7610.
Nufer O and Hauri HP (2003) ER export: call 14-3-3. Curr Biol 13: R391eCR393.
O'Kelly I, Butler MH, Zilberberg N, and Goldstein SA (2002) Forward transport. 14eC3-3 binding overcomes retention in endoplasmic reticulum by dibasic signals. Cell 111: 577eC588.
Pagano A, Rovelli G, Mosbacher J, Lohmann T, Duthey B, Stauffer D, Ristig D, Schuler V, Meigel I, Lampert C, et al. (2001) C-Terminal interaction is essential for surface trafficking but not for heteromeric assembly of GABAB receptors. J Neurosci 21: 1189eC1202.
Robbins MJ, Calver AR, Filippov AK, Hirst WD, Russell RB, Wood MD, Nasir S, Couve A, Brown DA, Moss SJ, et al. (2001) GABAB(2) is essential for G-protein coupling of the GABAB receptor heterodimer. J Neurosci 21: 8043eC8052.
Schutze MP, Peterson PA, and Jackson MR (1994) An N-terminal double-arginine motif maintains type II membrane proteins in the endoplasmic reticulum. EMBO (Eur Mol Biol Organ) J 13: 1696eC1705.
Scott DB, Blanpied TA, Swanson GT, Zhang C, and Ehlers MD (2001) An NMDA receptor ER retention signal regulated by phosphorylation and alternative splicing. J Neurosci 21: 3063eC3072.
Shikano S and Li M (2003) Membrane receptor trafficking: evidence of proximal and distal zones conferred by two independent endoplasmic reticulum localization signals. Proc Natl Acad Sci USA 100: 5783eC5788.
Valluru L, Xu J, Zhu Y, Yan S, Contractor A, and Swanson GT (2005) Ligand binding is a critical requirement for plasma membrane expression of heteromeric kainate receptors. J Biol Chem 280: 6085eC6093.
White JH, Wise A, Main MJ, Green A, Fraser NJ, Disney GH, Barnes AA, Emson P, Foord SM, and Marshall FH (1998) Heterodimerization is required for the formation of a functional GABAB receptor. Nature (Lond) 396: 679eC682.
Yuan H, Michelsen K, and Schwappach B (2003) 14-3-3 dimers probe the assembly status of multimeric membrane proteins. Curr Biol 13: 638eC646.
Zerangue N, Malan MJ, Fried SR, Dazin PF, Jan YN, Jan LY, and Schwappach B (2001) Analysis of endoplasmic reticulum trafficking signals by combinatorial screening in mammalian cells. Proc Natl Acad Sci USA 98: 2431eC2436.
Zerangue N, Schwappach B, Jan YN, and Jan LY (1999) A new ER trafficking signal regulates the subunit stoichiometry of plasma membrane K(ATP) channels. Neuron 22: 537eC548.