The present address of Dr. J. Kirsch is Department of Anatomy8 and Cellular Neurobiology, University of Ulm, Albert-Einstein-Alle 11, 89069 Ulm, Germany.
Hydrophobic Interactions Mediate Binding of the Glycine Receptor β-Subunit to Gephyrin
Article first published online: 7 JUL 2008
Journal of Neurochemistry
Volume 72, Issue 3, pages 1323–1326, March 1999
How to Cite
Kneussel, M., Hermann, A., Kirsch, J. and Betz, H. (1999), Hydrophobic Interactions Mediate Binding of the Glycine Receptor β-Subunit to Gephyrin. Journal of Neurochemistry, 72: 1323–1326. doi: 10.1046/j.1471-4159.1999.0721323.x
Abbreviations used : FITC, fluorescein isothiocyanate ; GFP, green fluorescent protein ; GlyR, glycine receptor ; HEK human embryonic kidney.
- Issue published online: 7 JUL 2008
- Article first published online: 7 JUL 2008
- Glycine receptor;
- Receptor clustering;
- Hydrophobic interaction;
- Green fluorescent protein fusion protein
Abstract : Glycine receptors (GlyRs) are ligand-gated chloride channel proteins composed of α- and β-subunits. GlyRs are located to and anchored at postsynaptic sites by the receptorassociated protein gephyrin. Previous work from our laboratory has identified a core motif for gephyrin binding in the cytoplasmic loop of the GlyR β-subunit. Here, we localized amino acid residues implicated in gephyrin binding by site-directed mutagenesis. In a novel transfection assay, a green fluorescent protein-gephyrin binding motif fusion protein was used to monitor the consequences of amino acid substitutions for β-subunit interaction with gephyrin. Only multiple, but not single, replacements of hydrophobic side chains abolished the interaction between the two proteins. Our data are consistent with gephyrin binding being mediated by the hydrophobic side of an imperfect amphipathic helix.
Gephyrin is a peripheral membrane protein that was originally identified by copurification with the inhibitory glycine receptor (GlyR) (Schmitt et al., 1987). Gephyrin codistributes with the GlyR at central synapses (Triller et al., 1985 ; Altschuler et al., 1986). It serves as an anchor molecule linking GlyRs to the subsynaptic cytoskeleton and binds polymerized tubulin (Kirsch et al., 1991 ; Kirsch and Betz, 1995). In situhybridization (Kirsch et al., 1993a) and immunocytochemical (Triller et al., 1985 ; Altschuler et al., 1986 ; Araki et al., 1988 ; Kirsch and Betz, 1993) studies have revealed a widespread expression of gephyrin in embryonic and adult brain. Furthermore, gephyrin is found in close association with GABAA receptor subunits at GABAergic synapses in retina (Sassoe-Pognetto et al., 1995), spinal cord (Todd et al., 1996), and hippocampus (Craig et al., 1996), indicating that gephyrin also plays a role in clustering GABAA receptors at synaptic sites (Essrich et al., 1998).
Studies on spinal neurons in culture indicate that the formation of membrane-associated gephyrin clusters precedes the postsynaptic localization of GlyRs (Kirsch et al., 1993b ; Bechade et al., 1996). Moreover, gephyrin depletion by antisense treatment (Kirsch et al., 1993b) and gene targeting (Feng et al., 1998) has been found to prevent the synaptic accumulation of GlyRs. Gephyrin is therefore believed to act not only as an anchoring protein (Kirsch and Betz, 1995), but also as an organizer protein (Kirsch and Betz, 1998) at inhibitory postsynaptic membrane specializations.
Postsynaptic GlyRs are pentameric chloride channel proteins composed of α- and β-subunits (reviewed by Kuhse et al., 1995). Presently four α-subunit (α1-α4) genes and a single β-subunit gene are known. The α-subunits harbor the GlyR binding sites for agonists and antagonists and form functional glycine-gated chloride channels on heterologous expression. The β-subunit alone does not assemble into functional GlyRs, but its incorporation into recombinant receptors changes the properties of the ion channel. In addition, it mediates binding to gephyrin via its cytoplasmic loop region between the third and fourth transmembrane segments (Meyer et al., 1995). In overlay and transfection experiments, this binding was shown to involve a motif of 33 amino acids in the central region of the M3-M4 loop, with an 18-amino acid core sequence harboring the dominant binding determinants (Meyer et al., 1995). Engineering this 18-amino acid sequence into specific GABAA (Meyer et al., 1995) and NMDA (Kins et al., 1999) receptor subunits has been found to confer gephyrin binding onto these non-GlyR membrane proteins. Here, we present an improved transfection assay for studying GlyR-gephyrin interactions that uses a green fluorescent protein (GFP) binding motif fusion protein. Site-directed mutagenesis of the binding motif revealed that the gephyrin binding activity of the GlyR β-subunit can be assigned to hydrophobic amino acid residues located on one side of a potential imperfect amphipathic helix.
MATERIALS AND METHODS
Constructs and mutagenesis
Amino acids 378-426 of the GlyR β-subunit containing the gephyrin binding motif (Meyer et al., 1995) were fused to the C terminus of recombinant GFP by cloning a 150-bp BamHI-EcoRI fragment of the GST-49 plasmid DNA (Meyer et al., 1995), into the Bg/II and EcoRI sites of the pEGFP-C2 vector (Clontech, Heidelberg, Germany). This construct encompassing 49 amino acids of the β-subunit intracellular loop region including the 18-amino acid core gephyrin binding motif, termed GFP-49, was routinely used in all “wild-type” transfection experiments. Mutants of GFP-49 were generated using the Quik-Change site-directed mutagenesis kit (Stratagene, Heidelberg). All substitutions were verified by dideoxy sequencing.
The monoclonal antibody mAb 5a specific for gephyrin (Pfeiffer et al., 1984) was used at a dilution of 1:100 in phosphate-buffered saline. GFP-polyclonal antibody (Clontech) was used for detection of GFP at a dilution of 1:2,000. As secondary antibodies, Texas red-anti-mouse IgG (Dianova, Hamburg, Germany) was used for mAb 5a, and fluorescein isothiocyanate (FITC)-anti-rabbit IgG (Dianova) was used for GFP-polyclonal antibody.
Cell culture, immunochemistry, and confocal microscopy
Human embryonic kidney (HEK) 293 cells were cultured and processed for immunofluorescence as described previously (Kirsch et al., 1995). Confocal microscopy was performed using the fluorochromes FITC and Texas red for double-labeling studies and CY3 for single-channel detection as described (Kirsch et al., 1995). All data are displayed along a linear intensity scale ; primary data sets were analyzed using Image Space Software (Molecular Dynamics, Sunnyvale, CA, U.S.A.).
Secondary structure and hydrophilicity prediction
RESULTS AND DISCUSSION
To investigate the sequence requirements for the interaction of the GlyR β-subunit with gephyrin, we first developed a novel transfection assay. The latter was based on our previous observation that coexpression in HEK 293 cells of the β-subunit with gephyrin causes its retention on intracellular gephyrin aggregates (Kirsch et al., 1995 ; Meyer et al., 1995). To facilitate the optical detection of this retention reaction, we fused the 18-amino acid core gephyrin binding motif together with its flanking sequences (amino acids 378-394 and 412-426) to the C-terminal region of GFP. Coexpression of the resulting construct, termed GFP-49, with gephyrin in HEK 293 cells allowed the rapid visualization of gephyrin binding motif interactions by fluorescence microscopy. As reported previously, single expression of gephyrin produced intracellular aggregates (Meyer et al., 1995), whereas GFP was diffusely distributed throughout the cytoplasm of HEK 293 cells (data not shown). Coexpression of gephyrin with GFP did not change these characteristic distributions, either for GFP (Fig. 1A) or for gephyrin (Fig. 1B). In some experiments with high expression levels, GFP formed cytoplasmic aggregates on its own ; however, these aggregates never overlapped with gephyrin immunoreactivity. In contrast, GFP-49 accumulated at the gephyrin aggregates (Fig. 1C). This finding corroborates our previous suggestion (Meyer et al., 1995) that the binding motif alone is sufficient for high-affinity binding to gephyrin.
Applying site-directed mutagenesis we then mutated specific amino acid residues in the core binding motif of GFP-49 between β-subunit residues 394 and 411 (Fig. 2). Group mutagenesis of polar residues to alanine generated mutant 1, containing R394A and D397A substitutions, and mutant 2, containing R406A, D407A, and E409A substitutions ; mutant 3 had all charged amino acid residues within the core binding motif replaced (R394A, D397A, R406A, D407A, and E409A). All of these mutant constructs retained the ability to interact with gephyrin as evidenced by their accumulation at gephyrin aggregates (Figs. 1D and 2). A similar result was observed when exchanging Pro405 (mutant 4, P405A ; (Figs. 1 E and 2). These data indicate that charged amino acids are not important for gephyrin binding.
In a second round of substitutions, we generated group mutations of hydrophobic residues. Mutant 5 had all hydrophobic side chains replaced (F398A, I400A, V401A, L404A, F408A, and L410A) ; this abolished the colocalization with gephyrin (Fig. 1F). In contrast, substitution of individual hydrophobic residues in mutant 6 (F398A), mutant 7 (I400A), mutant 8 (V401A), and mutant 9 (L404A) did not prevent the GFP fusion protein-gephyrin interaction (Fig. 2). Similarly, a double mutant with neighboring hydrophobic amino acid exchanges I400A and V401A (mutant 10) still bound gephyrin (Fig. 2). Furthermore, substitution of the isoleucine residue in position 400 by a basic lysine (mutant 11 ; I400K) failed to prevent gephyrin binding. Collectively, these data indicate that the interaction between the GlyR β subunit and gephyrin is mediated by multiple hydrophobic side chains in the binding region of the GlyR β-subunit.
Hydrophilicity analysis of residues 384-419 of the β-subunit indicates a modestly hydrophobic segment between Phe398 and Leu404 in a background of residues with positive hydrophilicity values (Fig. 3A). This suggests that the four hydrophobic residues in this cluster (F398, I400, V401, and L404) might play a decisive role for interaction of the GlyR β-subunit with gephyrin. As group mutation of all hydrophobic residues in the core binding motif (mutant 5) including these four candidate side chains was required to abolish the interaction with gephyrin, no critical residues seem to exist within this motif. This is consistent with our observation that single substitutions of these residues (mutants 6-9 and 11) did not destroy the GFP-49—gephyrin interaction. We therefore speculate that a higher-order structure, such as an α-helix, may be involved. Secondary structure prediction with different algorithms gave inconsistent results about the propensity of this region to form a helical structure, but helical wheel projections including this hydrophobic stretch (398-404) indicate a segregation of polar and hydrophobic residues to opposite sides of a putative α-helix (Fig. 3B). This type of residue segregation has been described for the calmodulin binding domain interacting with Ca2+/calmodulin-dependent kinase II (James et al., 1995). We therefore propose that an amphipathic structure mediates the interaction of the GlyR β-subunit with gephyrin.
In conclusion, we have identified several hydrophobic amino acids in the binding motif of the GlyR β-subunit that appear to be involved in the interaction with gephyrin. Because gephyrin shares structural (Prior et al., 1992) and functional (Feng et al., 1998) homology with proteins involved in the biosynthesis of the molybdenum cofactor in Escherichia coli, Drosophila, and plants and interacts with other cytoskeletal (Kirsch et al., 1991) and receptor subunit (Kirsch et al., 1995 ; Essrich et al., 1998) proteins, this protein must have pleiotropic functions in eukaryotic cells (Betz, 1998). Selective inactivation of gephyrin interaction motifs and its interaction partners via gene targeting may facilitate attempts to eliminate selectively subfunctions of gephyrin, such as inhibitory receptor clustering and/or coenzyme biosynthetic functions. Moreover, such approaches may be exploited to alter the receptor composition and/or density within defined postsynaptic membrane specializations or selected neuronal subpopulations. In other words, the mutants described here might open an avenue to synaptic engineering at the postsynaptic level.
We thank Ina Bartnik for excellent technical assistance and Masahisa Horiuchi for help with the prediction software. This work was funded by grants from the Bundesministerium für Bildung und Forschung, the Deutsche Forschungsgemeinschaft (SFB 474), and the Fonds der Chemischen Industrie.
- 101999) Incorporation of a gephyrin-binding motif targets NMDA receptors to gephyrin-rich domains in HEK 293 cells.Eur. J. Neurosci. (in press)., , , & (