QUICK SEARCH:   [advanced] Author: Keyword(s): Year:  Vol:  Page:

This Article Abstract Full Text (PDF) All Versions of this Article: mol.108.048710v1 74/4/941    most recent Submit a response Alert me when this article is cited Alert me when eLetters are posted Alert me if a correction is posted Services Similar articles in this journal Similar articles in PubMed Alert me to new issues of the journal Download to citation manager Citing Articles Citing Articles via Google Scholar Google Scholar Articles by Zhang, J. Articles by Chang, Y. Search for Related Content PubMed PubMed Citation Articles by Zhang, J. Articles by Chang, Y.

Structural Determinants for Antagonist Pharmacology That Distinguish the ₁ GABA_C Receptor from GABA_A Receptors

Division of Neurobiology, Barrow Neurological Institute, St. Joseph's Hospital and Medical Center, Phoenix, Arizona

Received for publication May 6, 2008.

Accepted for publication July 2, 2008.

	Abstract

Top Abstract Materials and Methods Results Discussion References

 
GABA receptor (GABAR) types C (GABA_CR) and A (GABA_AR) are both GABA-gated chloride channels that are distinguished by their distinct competitive antagonist properties. The structural mechanism underlying these distinct properties is not well understood. In this study, using previously identified binding residues as a guide, we made individual or combined mutations of nine binding residues in the ₁ GABA_CR subunit to their counterparts in the ₁β₂₂ GABA_AR or reverse mutations in ₁ or β₂ subunits. The mutants were expressed in Xenopus laevis oocytes and tested for sensitivities of GABA-induced currents to the GABA_A and GABA_C receptor antagonists. The results revealed that bicuculline insensitivity of the ₁ GABA_CR was mainly determined by Tyr106, Phe138 and Phe240 residues. Gabazine insensitivity of the ₁ GABA_CR was highly dependent on Tyr102, Tyr106, and Phe138. The sensitivity of the ₁ GABA_CR to 3-aminopropyl-phosphonic acid and its analog 3-aminopropyl-(methyl)phosphinic acid mainly depended on residues Tyr102, Val140, FYS240-242, and Phe138. Thus, the residues Tyr102, Tyr106, Phe138, and Phe240 in the ₁ GABA_CR are major determinants for its antagonist properties distinct from those in the GABA_AR. In addition, Val140 in the GABA_CR contributes to 3-APA binding. In conclusion, we have identified the key structural elements underlying distinct antagonist properties for the GABA_CR. The mechanistic insights were further extended and discussed in the context of antagonists docking to the homology models of GABA_A or GABA_C receptors.

Molecular cloning has identified at least 18 GABA receptor subunits in the nervous system (Barnard et al., 1998). They all belong to the cys-loop receptor family of the ligand-gated ion channels (Lester et al., 2004). A typical GABA_AR can be formed by exogenously coexpressing , β, and subunits with two subunits, two β subunits, and one subunit in a receptor (Chang et al., 1996). The ₁β₂₂ is the most abundant subtype of GABA_ARs in the central nervous system (Whiting et al., 2000). The GABA_CRs seemed to be mainly formed by subunits (Zhang et al., 2001). When exogenously expressed, the ₁ GABA receptor subunit can form functional channels with the GABA_CR pharmacological properties (Cutting et al., 1991).

Studies in the past 2 decades with photoaffinity labeling, site-directed mutagenesis, and the substituted cysteine accessibility method have shaped relatively complete models for the extracellular agonist/antagonist binding pockets of β GABA_AR (Sigel et al., 1992; Amin and Weiss, 1993; Smith and Olsen, 1994; Westh-Hansen et al., 1997, 1999; Boileau et al., 1999; Boileau et al., 2002; Holden and Czajkowski, 2003; Newell and Czajkowski, 2003) and the ₁ GABA_CR (Amin and Weiss, 1994; Lummis et al., 2005; Sedelnikova et al., 2005; Harrison and Lummis, 2006). In the structural model of the GABA_AR, residues in six loops (segments) designated A through F have been identified to form the agonist/antagonist binding pocket in the subunit interface between β and subunits. The β subunit contributes the binding loops A, B, and C. The subunit contributes the binding loops D, E, and F. In contrast, the agonist/antagonist binding pocket of the ₁ GABA_CR is formed in the subunit interface between two ₁ subunits with five binding loops (A-E) identified (Sedelnikova et al., 2005).

View larger version (32K): [in this window] [in a new window]   Fig. 1. Design of mutations based on the distinct binding residues in the GABAA and GABACRs. Sequence alignment of the amino-terminal domains of the GABAA (1, β2) and GABAC (1) receptor subunits with binding sites in bold. Except for an extended region of loop E (first three binding residues in loop E) that appears to not face the binding pocket in the structure model (Sedelnikova et al., 2005), all other distinct binding residues are potential binding residues underlying distinct antagonist pharmacological properties between the GABAA and GABACRs. The residues pointed by arrowheads are the residues under investigation.

In this study, we substituted the distinct residues of the ₁ GABA_CR subunit, indicated by arrowheads in Fig. 1, with the corresponding residues in the GABA_AR β₂ (for loops A and C) or ₁ (for loops D, E, and F) subunits individually or in combinations. When these mutants were expressed in Xenopus laevis oocytes, they formed functional channels with altered sensitivities to agonists and antagonists. By testing antagonist sensitivity of agonist-induced current in these mutants, we have identified key structural elements underlying distinct antagonist properties for the GABA_A and GABA_CRs. The mechanistic insights for the selective interactions between the antagonists and two types of receptors were further discussed in the context of antagonist dockings to the homology models of the GABA_A or GABA_CRs.

	Materials and Methods

Top Abstract Materials and Methods Results Discussion References

 
Mutagenesis and cRNA Preparation. The cDNA encoding human ₁ GABA receptor subunit and rat ₁, β₂, and ₂ GABA receptor subunits were kindly provided by Dr. David S. Weiss. Note that the rat GABA receptor subunits are highly homologous (98-99%) to their human counterparts. In fact, in all binding loops, these rat and human GABA receptor subunits are virtually identical. All subunits were cloned into the oocyte expression vector pGEMHE with T7 orientation. The residues in the amino-terminal segments corresponding to loops A, C, D, E, and F in the ₁ GABA receptor subunit were mutated to their homologous residues in the ₁ or β₂ GABA receptor subunits, individually or in combination, using the polymerase chain reaction-based QuikChange method of site-directed mutagenesis following the manufacturer's protocol (Stratagene, La Jolla, CA). The mutations were confirmed by automated DNA sequencing. The wild-type and mutant cDNAs were then linearized by NheI digestion. The cRNAs were transcribed with a standard in vitro transcription protocol as described previously (Sedelnikova et al., 2005). The cRNA yield and integrity were examined on a 1% agarose gel. cRNA concentration was further quantitated with an Eppendorf BioPhotometer (Eppendorf North America, New York, NY).

Oocyte Preparation and RNA Injection. Female X. laevis frogs (Xenopus I, Ann Arbor, MI) were anesthetized by 0.2% MS-222. The ovarian lobes were surgically removed from the frog and placed in the incubation solution, consisting of 82.5 mM NaCl, 2.5 mM KCl, 1 mM MgCl₂, 1 mM CaCl₂, 1 mMNa₂HPO₄, 0.6 mM theophylline, 2.5 mM sodium pyruvate, 5 mM HEPES, 50 µg/ml gentamicin, 50 U/ml penicillin, and 50 µg/ml streptomycin, pH 7.5. The frog was then given the analgesic xylazine hydrochloride (10 mg/kg i.p.) and allowed to recover from surgery in shallow water before being returned to the incubation tank. The lobes were cut into small pieces and digested with 1 Wünsch unit/ml Liberase Blendzyme 3 (Roche Applied Science, Indianapolis, IN) with constant stirring at room temperature for 1.5 to 2 h. The dispersed oocytes were thoroughly rinsed with the above solution. The stage VI oocytes were selected and incubated at 16°C before injection. Micropipettes for injection were pulled from borosilicate glass (Drummond Scientific, Broomall, PA) on a horizontal puller (P87; Sutter Instrument Company, Novato, CA), and the tips were cut with forceps to 40 µm in diameter. The cRNA was drawn up into the micropipette and injected into oocytes with a Nanoject microinjection system (Drummond Scientific) at a total volume of 2060 nl.

Two-Electrode Voltage-Clamp. One to 3 days after injection, the oocyte was placed in a homemade small-volume chamber with continuous perfusion with oocyte Ringer's solution, which consisted of 92.5 mM NaCl, 2.5 mM KCl, 1 mM CaCl₂, 1 mM MgCl₂, and 5 mM HEPES, pH 7.5. The chamber was grounded through an agar bridge. The oocytes were voltage-clamped at -60 mV to measure GABA-induced currents using a GeneClamp 500B (Axon Instruments, Foster City, CA). The current signal was low-pass filtered at 10 Hz with the built-in low-pass Bessel filter in the GeneClamp 500B and digitized at 20 Hz with Axon Digidata1320 and pClamp9 (Molecular Devices, Sunnyvale, CA) in a Dell desktop computer. For the antagonist sensitivity test, GABA-induced current with an EC₂₀ concentration, the concentration that induces 20% of maximum current, for each mutant was inhibited with coapplication of an antagonist with increasing concentrations. The antagonist IC₅₀ (the concentration that inhibits 50% of GABA-induced current) was then determined by fitting concentration-dependent inhibition data with a Hill inhibition equation using Prism 4 software (GraphPad Software, San Diego, CA). The IC₅₀ values were further used to calculate antagonist apparent affinity K_i for different mutants by the following equation: K_i = IC₅₀/(1 + [GABA]/EC₅₀) (Newell and Czajkowski, 2003).

Drug Preparation. GABA (SigmaAldrich, St. Louis, MO) stock solution (100 mM) was prepared daily from solid. (-)-Bicuculline methiodide (Tocris Bioscience, Ellisville, MO), gabazine (SR95531; Tocris), 3-APMPA (SKF97541; Tocris), and 3-APA (SigmaAldrich) stock solutions (20, 25, 100, and 100 mM, respectively) were prepared and stored at -20°C in aliquots before use.

Data Analysis. The dose-response relationship of the GABA-induced current in recombinant GABA_A/C receptors was least-squares fit to a Hill equation with Prism 4.0 (GraphPad Software) to derive EC₅₀, Hill coefficient (the slope factor), and maximum current. The dose-dependent inhibition by competitive antagonists was fitted to a Hill inhibition equation to derive IC₅₀, Hill coefficient, and maximal current. The maximum current was then used to normalize the dose-response/-inhibition curve for each individual oocyte. The averages of the normalized currents were used to plot the data. All the data were presented as mean ± S.E.M. (standard error of the mean).

Homology Modeling. The three-dimensional model of the pentameric extracellular domains of the ₁ GABA receptor was made previously (Sedelnikova et al., 2005). The three-dimensional model of ₁β₂₂ GABA receptor extracellular domain was built using Discovery Studio 1.7 software (Accelrys, San Diego, CA) running in a Dell Precision 690 computer (Dell, Austin, TX) with the homology model of the ₁ GABA receptor as the template to ensure that two homology models converge in a similar way for better comparison. In brief, amino-terminal domains of the rat GABA receptor subunits (β₂₁β₂₁₂) were aligned to the human ₁ sequences (chains A to E) with the modeler in the Discovery Studio Modeler 9 using "Align Sequence with Structure" protocol (Sali and Blundell, 1993) with blosum62 scoring matrix, gap open penalty of -100, gap extension penalty of -10, and default two-dimensional gap weights. The homology models were then built using "Building Homology Models" (Sali and Blundell, 1993). The pentameric model was further energy minimized for 400 steps of the "Steepest Descent" minimization followed by 1000 steps of "Conjugated Gradient" minimization using "Minimization" protocol with CHARMm force field. The model of mutant receptor with mutation(s) in the binding regions was generated with "Build Mutant" protocol and energy minimized as above.

Ligand Docking. The three-dimensional ligand structures of bicuculline, gabazine, 3-APA, and 3-APMPA were downloaded, as MDL molecule files, from the ChemIDplus National Institutes of Health web site (http://chem.sis.nlm.nih.gov/chemidplus/). The receptor subunit dimers were saved from original pentameric models. The docking of flexible ligands to the putative binding pockets of the GABA_A (in the interface between β₂ and ₁ subunits) or GABA_C (in the interface between two ₁ subunits) receptors was performed with "Dock Ligands (LigandFit)" protocol (Venkatachalam et al., 2003) in the Discovery Studio 1.7 software (Accelrys, San Diego, CA). The docking results were scored with all available scoring functions, which include DockScore [=-(ligand/receptor interaction energy + ligand internal energy)], LigScore1 and LigScore2 (Krammer et al., 2005), Ludi1 (Böhm, 1994) and Ludi2 (Böhm, 1998), Piecewise Linear Potential 1 (Gehlhaar et al., 1995) and Piecewise Linear Potential 2 (Gehlhaar et al., 1999), potential of mean force (Muegge and Martin, 1999), and Jain (Jain, 1996). The poses with the highest DockScores tended to have highest scores in other functions, although the scores from these scoring functions in different poses were only partially correlated (data not shown). Thus, the poses with highest DockScores and the lowest ligand internal energy were used for presentation unless specified otherwise. The docking success [with output pose(s)] was determined by pose saving thresholds. We used default values of Pose Saving Dockscore Threshold (0.0), Pose Saving RMS Threshold for Diversity (1.50), and Pose Saving Score Threshold for Diversity (20.0). A docking without any output pose is considered as a failure.

	Results

Top Abstract Materials and Methods Results Discussion References

 
Bicuculline Sensitivity. Bicuculline insensitivity is the major distinction of GABA_CRs from GABA_ARs. To search for the structural basis underlying this difference, we first made individual mutations, except for Tyr241 (FYS240VF), in nine distinct binding site residues in the ₁ GABA_CR to their counterparts in the ₁ or β₂ GABA_AR subunits. When expressed in X. laevis oocytes, all nine individual mutant ₁ GABA receptors resulted in functional channels with slightly altered GABA sensitivity (with maximum of 24-fold reduction). The EC₅₀ values and maximal currents (I_maxs) derived from GABA dose-response relationships of these mutants are listed in Table 1. The GABA-induced current with an EC₂₀ GABA concentration was then used to test bicuculline sensitivity of these mutants. Figure 2A represents the dose-dependent inhibition of the GABA-induced currents by bicuculline in the wild-type and single mutant ₁ GABA receptors. Note that although the wild-type ₁ GABA_CR was essentially insensitive to bicuculline, three (of nine) mutants (Y106S, F138Y and FYS240VF) exhibited slightly increased sensitivity to bicuculline (Fig. 2A). At the highest concentrations tested, bicuculline blocked nearly one half of the GABA-induced currents in these three mutants. Due to incomplete inhibition (<50%) at the highest concentrations tested, the IC₅₀ values could not be reliably determined. They were clearly slightly higher than the highest concentrations tested in these three mutants. The range of IC₅₀ values and derived apparent affinity K_i are listed in Table 2. The result suggests that these three mutants are potential candidates for further investigation.

View larger version (33K): [in this window] [in a new window]   Fig. 2. Effect of nine individual mutations and their combinations of the 1 GABACR on the sensitivity to bicuculline. A, effect of individual mutations. Top, examples of GABA (EC20)-induced current traces blocked by increasing concentrations of bicuculline. Bottom, normalized and averaged bicuculline dose-inhibition of the wild type and nine mutant receptors (n 3 for each construct). Note that at the highest concentration, bicuculline blocked only <10% of the GABA-induced current in the wild-type receptor. In contrast, bicuculline blocked >40% of the GABA-induced currents in three mutants (Y106S, F138Y, and FYS240VF). Because the maximal inhibitions were less than 50%, dose-inhibition fitting could not generate reliable results and was not performed. Instead, straight lines are used to link the points for each construct (WT or mutant). B, effect of double and triple mutations on bicuculline sensitivity (n 3 for each construct). Top, examples of GABA(EC20)-induced current traces blocked by increasing concentrations of bicuculline. Bottom, normalized and averaged bicuculline dose-inhibition of the double or triple mutants. Continuous lines are best fits of the data to a Hill inhibition equation, and the resulting IC50 values are listed in Table 2. Note that the triple mutant containing Y106S, F138Y, and FYS240VF exhibited highest sensitivity to bicuculline.

By making single mutations at these nine distinct binding residues in the ₁ GABA_CR subunit, we were clearly unable to dramatically increase binding affinity to bicuculline. However, we could potentially achieve this goal by combining several promising individual mutants. To test this, we made double and triple mutants with combinations of the three promising individual mutants. Figure 2B represents bicuculline dose-inhibition for the combinations of these promising mutants along with one nonpromising mutant as a control. Double mutations (Y106S+ FYS240VF, Y106S+F138Y, F138Y+FYS240VF) clearly increased bicuculline sensitivity of the receptor. The triple mutant Y106S+F138Y+FYS240VF exhibited the highest bicuculline sensitivity (K_i = 28.85 ± 1.96 µM, Table 2), which was only severalfold lower than the bicuculline sensitivity of the wild-type GABA_AR(K_i = 5.21 ± 0.74 µM, Table 2). Thus, mutations of these residues to their homologous residues in the GABA_AR are enough to confer bicuculline sensitivity to the ₁ GABA_CR. These residues must synergistically contribute to bicuculline affinity, although we cannot role out the contribution of other binding and "nonbinding" residues to bicuculline binding. For individual contributions of the residues Phe240 and Tyr241 in FYS240VF mutant, please see Discussion.

Gabazine Sensitivity. Like bicuculline, gabazine is also a relatively selective GABA_AR antagonist, although their structures are quite different. In fact, gabazine has a nanomolar affinity for GABA_AR. However, the wild-type ₁ GABA_CR was still sensitive to gabazine but with a much lower affinity (K_i = 57.84 ± 7.66 µM) compared with K_i = 0.12 ± 0.01 µM in the wild type GABA_AR (Table 3). Figure 3A represents gabazine dose-inhibition relationships in nine individual mutants of the ₁ GABA receptor. Note that whereas FYS240VF mutation increased apparent affinity by only 2-fold, the other three mutants (Y102F, Y106S, and F138Y) exhibited a larger increase in gabazine affinity (with decreased K_i; Table 3). Combination of Y102F, Y106S, and F138Y resulted in a receptor with a much lower K_i (1.61 ± 0.03 µM) despite a dramatic increase in GABA EC₅₀ (156.46 ± 10.73 µM) for this mutant receptor. Thus, residues Tyr102, Tyr106, and Phe138 are major contributors to gabazine binding. Other minor contributors include Ser168 and Thr218. Mutations of them increased gabazine affinity by approximately 5-fold.

View larger version (33K): [in this window] [in a new window]   Fig. 3. Effect of nine individual mutations and their combinations of the 1 GABACR on the sensitivity to gabazine. A, effect of individual mutations. Top, examples of GABA (EC20)-induced current traces blocked by increasing concentrations of gabazine. Bottom, normalized and averaged gabazine dose-inhibition of the wild-type and nine mutant receptors (n 3 for each construct). The continuous lines are best fits of the data to a Hill inhibition equation, and the resulting IC50 values are listed in Table 3. B, effect of double and triple mutations on gabazine sensitivity (n 3 for each construct). Top, examples of GABA (EC20)-induced current traces blocked by increasing concentrations of gabazine. Bottom, normalized and averaged gabazine dose inhibition of the GABA-induced currents in the double or triple mutants. The continuous lines are best fits of the data to a Hill inhibition equation. The dashed line is the fit of gabazine inhibition of the GABA-induced current in the wild type receptor.

3-APA and 3-APMPA Sensitivity. 3-APA and its methylated analog 3-APMPA are selective competitive antagonists for GABA_C over GABA_ARs (Johnston, 1996), although they are also GABA_B receptor agonists (Froestl et al., 1995). Mutants Y102F, V140L, and FYS240VF exhibited the most dramatic reduction of apparent affinity to the GABA_CR competitive antagonist 3-APA (11-, 32-, and 25-fold reduction, respectively, Fig. 4A and Table 4). Thus, bicuculline and 3-APA both interact with three binding loops (D, A, and C) but with slightly different residues in loops D and A. We predicted that when all three residues are mutated, the sensitivity to 3-APA should be further reduced. Indeed, the triple mutant Y102F+Val140L+FYS240VF exhibited 77-fold reduction in sensitivity to 3-APA (Fig. 4B and Table 4).

View larger version (31K): [in this window] [in a new window]   Fig. 4. Effect of mutations of the 1 GABACR on the sensitivity to 3-APA. A, effect of individual mutations on 3-APA inhibition of the receptor. Top, an example of GABA-induced current traces blocked by increasing concentrations of 3-APA. Bottom, normalized and averaged 3-APA dose-inhibition of the wild type and nine single mutant receptors. The continuous lines are best fits of the data to a Hill inhibition equation, and the resulting IC50 values are listed in Table 4. B, effect of double and triple mutations on 3-APA sensitivity. The continuous lines are best fits of the data to a Hill inhibition equation, and the resulting IC50 values are listed in Table 4. C, effect of quintuple mutations on 3-APA sensitivity. Because of limited inhibition, fitting was not performed. Straight lines are used to link multiple points.

Because the triple mutant is still slightly sensitive to 3-APA (K_i = 844.88 ± 100.67 µM), we then combined two additional mutations (Y106S and F138Y) to the receptor. When the quintuple mutant was expressed, it exhibited bicuculline sensitivity (although reduced compared with the triple mutant; Table 2) and 3-APA insensitivity (Fig. 4C). Further reduction of 3-APA sensitivity in the quintuple mutant could be due to the contribution of F138Y mutation, which reduced 3-APA sensitivity by 7.5-fold when singly mutated. Thus, we have identified five residues in the binding site that conferred GABA_C properties to the ₁ GABA receptor, and the combined mutation of these residues converted the ₁ GABA receptor to the GABA_AR antagonist pharmacology.

3-APMPA has a structure similar to that of 3-APA. Using high concentrations of 3-APMPA to test all mutants is cost-prohibitive. Thus, we tested only its sensitivity in the quintuple mutant. Indeed, the quintuple mutant also exhibited insensitivity to 3-APMPA. At the concentration of 1500 µM, 3-APMPA only inhibited the GABA-induced current by approximately 30% (data not shown). Thus, residues Tyr102, Phe138, Val140, and potentially Tyr241 are important determinant for 3-APA and 3-APMPA binding.

Corresponding Mutants in GABA_AR Partially Converted the Receptor to GABA_C Pharmacology. If the identified residues in GABA_CR are major determinants for its antagonist specificity, we should expect that mutations of these residues in GABA_AR also convert its pharmacological properties to GABA_CR. In the GABA_AR, loops A and C are in the β subunit, whereas loop D is in the subunit. The three mutants corresponding to the Y106S, F138Y, and FYS240VF in the ₁ GABA_CR for bicuculline sensitivity are ₁(S68Y), β₂(Y97F), and β₂(VF199FYS). In fact, single mutations of ₁(S68Y) or β₂(Y97F) reduced the receptor sensitivity to bicuculline (Table 2 and Fig. 5A). Thus, ₁ Ser68 and β₂ Tyr97 are important determinants of GABA_AR bicuculline sensitivity. In contrast, the mutant β₂VF199FYS in the loop C slightly increased bicuculline sensitivity (Table 2). This opposite effect will be discussed in more detail under Discussion. When the quintuple mutant (triple mutant β₂ subunit coexpressed with the double mutant ₁ subunit) in the GABA_AR, corresponding to the quintuple mutant of the GABA_CR in Fig. 4, C and D, the receptor exhibited a significant reduction in bicuculline (Table 2) and gabazine (Fig. 5B) sensitivity and became slightly sensitive to 3-APA (data not shown). The results further support the importance of these five binding residues for the GABA_A and GABA_CR antagonist properties.

View larger version (23K): [in this window] [in a new window]   Fig. 5. Mutations of the GABAAR in the corresponding residues partially converted the receptor to GABAC antagonist properties. A, effect of mutations of the GABAAR on bicuculline sensitivity. The GABA receptor β2Y97F mutation resulted in more than 10-fold decrease in bicuculline sensitivity. Coexpression of this mutant with the 1S68Y mutant resulted in further reduction in bicuculline sensitivity. This further reduction of bicuculline sensitivity was counteracted by adding the third mutation β2VF199FYS to the receptor. The dashed line represents the normalized dose-inhibition of GABA-induced current by bicuculline in the wild-type GABAAR. B, the quintuple mutant of the GABAAR to their homologous residues in the 1 GABACR reduced gabazine sensitivity. Top, an example of a GABA-induced current inhibited by increasing concentrations of gabazine. Bottom, normalized and averaged dose-inhibition of the GABA-induced current by gabazine. Continuous line is the best fit of the data to a Hill inhibition equation. The resulting IC50 is listed in Table 3. Dashed line represents the normalized dose-inhibition of GABA-induced current by gabazine in the wild type GABAAR.

	Discussion

Top Abstract Materials and Methods Results Discussion References

Bicuculline Sensitivity. We have successfully docked bicuculline into the putative GABA_AR binding pocket but failed in docking bicuculline into the GABA_CR binding pocket. Figure 6A shows the docked bicuculline in the GABA_AR binding pocket. Note that three putative hydrogen bonds are formed between the docked bicuculline molecule and residues β₂ Tyr97 (loop A), β₂ Tyr157 (loop B), or ₁ Arg66 (loop D). Interaction of bicuculline to β₂ Tyr97 is supported by that a single β₂ Y97F mutation dramatically reduced the GABA_AR sensitivity to bicuculline (Table 2, Fig. 5A). The homologous residue in the GABA_CRis ₁ Phe138. However, the F138Y mutation only slightly increased the GABA_CR affinity to bicuculline, presumably by forming a hydrogen bond with bicuculline. Thus, other residues must make substantial contributions to the bicuculline insensitivity of GABA_CR. β₂ Tyr157 and ₁ Arg66 of GABA_AR are binding residues (Amin and Weiss, 1993; Harrison and Lummis, 2006). They are the same as their homologs ₁ Tyr198 and ₁ Arg104 in the GABA_CR and thus do not contribute distinct bicuculline sensitivity.

View larger version (54K): [in this window] [in a new window]   Fig. 6. Structural models of the GABAA or GABACR amino-terminal domains docked with antagonists. A, bicuculline docked to the GABAAR binding pocket. Note that the docked bicuculline formed hydrogen bonds with three residues in loops A (β2 Tyr97), D (1 Arg66), and B (β2 Tyr157) (not visible in this view), and was also in close vicinity to 1 Ser68. B, gabazine docked to the GABAAR binding pocket. Note that the docked gabazine molecule formed hydrogen bonding with four residues in loops A (β2Y97), B (β2 Glu155 and β2 Tyr157), and C (β2 Ser201). C, 3-APA docked to GABACR binding pocket. The docked 3-APA molecule can form a hydrogen bond with Glu196 in loop B of the 1 GABA receptor subunit, homologous to β2 Glu155 in the GABAAR. Note that the docked 3-APA resided in the aromatic box formed by Phe138, Phe240, Tyr241, and Tyr102 [and Tyr247 and Tyr198 (behind)].

To further dissect contributions of individual residues in the FYS240FY mutant in loop C, we first examined the homology model of GABA_CR. The loop C of the GABA_CR has two aromatic residues, Phe240 and Tyr241 (a binding residue). The corresponding region in GABA_AR has only one phenylalanine aligned to Tyr241. In addition, the position of these two aromatic residues could be altered by one insertion, Ser242. In the model, both residues (Phe240 and Tyr241) are lining the binding pocket. Because bicuculline is a large molecule, it is possible that this additional aromatic residue, Phe240, provides a steric hindrance for bicuculline binding. In fact, virtual F240V mutation in the GABA_CR homology model increased the size of the binding pocket. Consequently, we were able to dock the bicuculline into the binding pocket of this mutant GABA_CR model (data not shown). This effect was confirmed experimentally; whereas GABA sensitivity of the ₁ F240V mutant was similar to that of the wild type (but with a substantially reduced efficacy), this mutant did exhibit bicuculline sensitivity (K_i = 67.4 ± 0.5 µM). This further supports the notion that Phe240 in GABA_C provides steric hindrance for bicuculline binding. In contrast, Y241F mutant remained insensitive to bicuculline (data not shown), suggesting that ₁ Tyr241 does not contribute to bicuculline insensitivity. Note that the reverse mutation in the GABA_AR (β₂VF199FYS) did not reduce bicuculline sensitivity. In the GABA_CR binding pocket, a residue near the Phe240 is the Asp219 in loop F. This negatively charged residue may provide electrostatic repulsion to the bicuculline, because the homologous residue in GABA_ARis ₁ Ala181, a small and noncharged residue. Thus, it may need both Phe240 and Asp219 acting collaboratively to narrow the bottom of the binding pocket, preventing bicuculline binding.

As for Tyr106 in loop D, ₁ Y106S mutation improved sensitivity to bicuculline. The reverse mutation in the GABA_AR(₁ S68Y) reduced bicuculline sensitivity (Table 2). ₁ Tyr106 is next to Arg104, which is equivalent to ₁ R66 (making hydrogen bonding to bicuculline). It is possible that steric hindrance of loop C in the GABA_CR pushes bicuculline to an upper position (Y106S), shifting hydrogen bonding from Arg104 to Y106S. The bulky Tyr106 may protrude too far so that it cannot form a hydrogen bond to bicuculline. However, when it was mutated to serine, the distance to bicuculline became closer.

Influence of Other Nearby Residues Although addition of Y102F and Val140L mutations to the triple mutant further decreased 3-APA sensitivity, it also unexpectedly resulted in a decrease in bicuculline sensitivity. This reduction could partially due to gating effect because the quadruple and quintuple mutants exhibited significantly reduced maximal current (Table 1). We re-tested I_max values in the quadruple and quintuple mutants and compared them with the wild type (Table 1 legend). Both mutants exhibited approximately a 7-fold reduction of I_max. Three-binding-to-open model (Amin and Weiss, 1996; Chang et al., 2000) predicts that EC₅₀ shift due to this I_max reduction could account for 2.7-fold increase in EC₅₀ value. Thus, the calculated K_i could be underestimated by 2-fold as a result of reduction of gating efficiency. In addition, trivial contributions of many other nearby nonbinding residues may influence the conformation in the binding site, making the GABA_AR more sensitive to bicuculline.

Gabazine Sensitivity. Our results suggest that Tyr102, Tyr106, and Phe138 are important residues that make the GABA_CR less sensitive to gabazine, and FYS240VF is not important for gabazine sensitivity. The result is consistent with a previous finding in the GABA_AR: ₁ F64 (homologous to ₁ Tyr102) and β₂ Tyr97 (homologous to ₁ Phe138) are binding residues for gabazine (Boileau et al., 2002; Holden and Czajkowski, 2002). To get structural insights into the mechanism, we successfully docked gabazine molecule into the binding pockets of both GABA_A and GABA_C receptors with higher docking score in the GABA_AR. Figure 6B is the docked gabazine to the GABA_AR binding pocket. Four putative hydrogen bonds were identified between the docked gabazine and four residues in loops A (β₂ Tyr97), B (β₂ Glu155 and β₂ Tyr157) and C (β₂ Ser201). β₂ Tyr97 is homologous to ₁ Phe138. Thus, this conserved YF mutation potentially eliminates one hydrogen bonding, making the GABA_CR less sensitive to gabazine. β₂ Glu155 and β₂ Tyr157 are important binding residues in the GABA_AR (Amin and Weiss, 1993; Newell et al., 2004). The residues corresponding to β₂ Glu155, β₂ Tyr157, and β₂ Ser201 in the ₁ subunit are Glu196, Tyr198, and Ser243 (or Ser242 because of an insertion of a serine). Because of identical residues at these positions in both GABA_AR and GABA_CR, they are not under our consideration. It is noteworthy that ₁ Y102F significantly increased the apparent affinity to gabazine. Using the structural model of GABA_AR as a reference, we speculate that the nature of interaction between ₁ F64 and gabazine is most likely a hydrophobic interaction. Thus, a more hydrophilic tyrosine at this position of the ₁ GABA_CR would substantially weaken this interaction. As for ₁ Tyr106 (₁ Ser68), the docked gabazine in the GABA_AR model could not reach ₁ Ser68 in all poses. However, the docked gabazine in the GABA_CR binding pocket exhibited a clockwise rotation, bringing the top of the molecule to the vicinity of Tyr106 or its mutant Y106S (data not shown) while maintaining the contact with Tyr102. Thus, in the complex interacting network, ligand docking position can be altered by other available interactions. Finally, FYS240VF is not important for gabazine sensitivity, probably because phenylalanine no longer provides steric hindrance to the smaller-sized gabazine.

3-APA and 3-APMPA Sensitivity. Compared with bicuculline and gabazine, 3-APA and 3-APMPA are much smaller molecules. Our results suggest that 3-APA apparent affinity was reduced substantially by four mutations of Y102F, V140L, FYS240VF, and F138Y (Table 4). Thus, the binding site must be located in the vicinity of these four positions. Figure 6C represents a pose with 3-APA docked into an aromatic box formed by Phe138, Phe240, Tyr241, and Tyr102 [and Tyr247 and Tyr198 (behind)]. However, Val140 is not in direct contact with 3-APA. It is possible that mutation of this valine to a larger residue leucine would provide a steric hindrance to the binding and result in a decreased binding affinity for 3-APA. The amino group of the 3-APA also potentially forms a hydrogen bond with Glu196, which is homologous to β₂ Glu155 in the GABA_AR. With the aromatic box surrounding the docked 3-APA, it is possible that there is a -cation interaction between 3-APA and a nearby aromatic residue(s). Thus, the binding of 3-APA to the GABA_CR could be very similar to the binding of GABA to the receptor (Lummis et al., 2005). Lacking an aromatic residue in the GABA_AR at the position homologous to Phe240 could reduce 3-APA affinity. 3-APMPA has structure similar to that of 3-APA. Parallel affinity reduction with these mutations for 3-APA and 3-APMPA also suggests that the structural requirements of their binding to the GABA_CR are similar. Reverse mutation of the GABA_AR only slightly increased 3-APA affinity suggesting other residues also make a significant contribution to the 3-APA binding.

In summary, we have identified important residues in three binding loops responsible for the GABA_CR antagonist properties distinct from those of GABA_AR. The insights gained from this study can aid design of new antagonists for the GABA_A and GABA_C receptors. The approach we used in this study can also be applied to examine the mechanisms of agonist/antagonist specificity among the GABA_AR subtypes.

	Acknowledgements

 
We thank Dr. David S. Weiss from the Department of Neurobiology at the University of Alabama at Birmingham (currently in the Department of Physiology at the University of Texas at San Antonio) for kindly providing the wild type human ₁ and rat a₁, b₂, and ₂ subunit constructs. We also thank Dr. Alan Gibson in the Barrow Neurological Institute for his help in proofreading the manuscript.

	Footnotes

 
This work was supported by Arizona Biological Research Commission grant (ABRC0702) and by Barrow Neurological Foundation (to Y.C.).

ABBREVIATIONS: GABA_AR, GABA_A receptor; GABA_CR, GABA_C receptor; 3-APA, 3-aminopropyl-phosphonic acid; 3-APMPA, 3-aminopropyl-(methyl)phosphinic acid.

Address correspondence to: Dr. Yongchang Chang, Division of Neurobiology, Barrow Neurological Institute, 350 West Thomas Road, Phoenix, AZ 85013. E-mail: yongchang.chang{at}chw.edu

	References

Top Abstract Materials and Methods Results Discussion References

 
Amin J and Weiss DS (1993) GABA_A receptor needs two homologous domains of the β subunit for activation by GABA, but not by pentobarbital. Nature 366: 565-569.[CrossRef][Medline]

Amin J and Weiss DS (1994) Homomeric ₁ GABA channels: activation properties and domains. Receptors Channels 2: 227-236.[Medline]

Amin J and Weiss DS (1996) Insights into the activation mechanism of 1 GABA receptors obtained by coexpression of wild type and activation-impaired subunits. Proc Biol Sci 263: 273-282.[Abstract/Free Full Text]

Barnard EA, Skolnick P, Olsen RW, Mohler H, Sieghart W, Biggio G, Braestrup C, Bateson AN, and Langer SZ (1998) International Union of Pharmacology. XV. Subtypes of -aminobutyric acidA receptors: classification on the basis of subunit structure and receptor function. Pharmacol Rev 50: 291-313.[Abstract/Free Full Text]

Böhm HJ (1994) The development of a simple empirical scoring function to estimate the binding constant for a protein-ligand complex of known three-dimensional structure. J Comput Aided Mol Des 8: 243-256.[CrossRef][Medline]

Böhm HJ (1998) Prediction of binding constants of protein ligands: A fast method for the prioritization of hits obtained from the de novo design or 3D database search programs. J Comput Aided Mol Des 12: 309-323.[CrossRef][Medline]

Boileau AJ, Evers AR, Davis AF, and Czajkowski C (1999) Mapping the agonist binding site of the GABA_A receptor: evidence for a β-strand. J Neurosci 19: 4847-4854.[Abstract/Free Full Text]

Boileau AJ, Newell JG, and Czajkowski C (2002) GABA_A receptor β2 Tyr97 and Leu99 line the GABA-binding site: Insights into mechanisms of agonist and antagonist actions. J Biol Chem 277: 2931-2937.[Abstract/Free Full Text]

Chang Y, Covey DF, and Weiss DS (2000) Correlation of the apparent affinities and efficacies of -aminobutyric acidC receptor agonists. Mol Pharmacol 58: 1375-1380.[Medline]

Chang Y, Wang R, Barot S, and Weiss DS (1996) Stoichiometry of a recombinant GABA_A receptor. J Neurosci 16: 5415-5424.[Abstract/Free Full Text]

Cutting GR, Lu L, O'Hara BF, Kasch LM, Montrose-Rafizadeh C, Donovan DM, Shimada S, Antonarakis SE, Guggino WB, and Uhl GR (1991) Cloning of the -aminobutyric acid (GABA) ₁ cDNA: a GABA receptor subunit highly expressed in the retina. Proc Natl Acad Sci U S A 88: 2673-2677.[Abstract/Free Full Text]

Drew CA, Johnston GA, and Weatherby RP (1984) Bicuculline-insensitive GABA receptors: studies on the binding of (-)-baclofen to rat cerebellar membranes. Neurosci Lett 52: 317-321.[CrossRef][Medline]

Froestl W, Mickel SJ, Hall RG, von Sprecher G, Strub D, Baumann PA, Brugger F, Gentsch C, Jaekel J, and Olpe HR (1995) Phosphinic acid analogs of GABA. 1. New potent and selective GABA_B agonists. J Med Chem 38: 3297-3312.[CrossRef][Medline]

Gehlhaar D, Bouzida D and Rejto P (1999) Reduced dimensionality in ligand-protein structure prediction: covalent inhibitors of serine proteases and design of site-directed combinatorial libraries, in Rational Drug Design: Novel Methodology and Practical Applications (ACS Symposium Series) (Parrill L and Rami Reddy M eds) pp 292-311, American Chemical Society, Washington, DC.

Gehlhaar DK, Verkhivker GM, Rejto PA, Sherman CJ, Fogel DB, Fogel LJ, and Freer ST (1995) Molecular recognition of the inhibitor AG-1343 by HIV-1 protease: conformationally flexible docking by evolutionary programming. Chem Biol 2: 317-324.[CrossRef][Medline]

Harrison NJ and Lummis SC (2006) Locating the carboxylate group of GABA in the homomeric rho GABA_A receptor ligand-binding pocket. J Biol Chem 281: 24455-24461.[Abstract/Free Full Text]

Holden JH and Czajkowski C (2002) Different residues in the GABA_A receptor 1T60-1K70 Region mediate GABA and SR-95531 actions. J Biol Chem 277: 18785-18792.[Abstract/Free Full Text]

Holden J and Czajkowski C (2003) 1G124-1L132: a novel binding site region on the GABA_A receptor that undergoes distinct conformational rearrangements during ligand binding and allosteric modulation. Soc Neurosci Abstr 29: 50.10.

Jain AN (1996) Scoring noncovalent protein-ligand interactions: A continuous differentiable function tuned to compute binding affinities. J Comput Aided Mol Des 10: 427-440.[CrossRef][Medline]

Johnston GA (1996) GABA_C receptors: relatively simple transmitter-gated ion channels? Trends Pharmacol Sci 17: 319-323.[CrossRef][Medline]

Krammer A, Kirchhoff PD, Jiang X, Venkatachalam CM, and Waldman M (2005) LigScore: a novel scoring function for predicting binding affinities. J Mol Graph Model 23: 395-407.[CrossRef][Medline]

Lester HA, Dibas MI, Dahan DS, Leite JF, and Dougherty DA (2004) Cys-loop receptors: new twists and turns. Trends Neurosci 27: 329-336.[CrossRef][Medline]

Lummis SC, L Beene D, Harrison NJ, Lester HA, and Dougherty DA (2005) A cation- binding interaction with a tyrosine in the binding site of the GABA_C Receptor. Chem Biol 12: 993-997.[CrossRef][Medline]

Muegge I and Martin YC (1999) A general and fast scoring function for protein-ligand interactions: a simplified potential approach. J Med Chem 42: 791-804.[CrossRef][Medline]

Murata Y, Woodward R, Miledi R, and Overman L (1996) The first selective antagonist for a GABA_C receptor. Bioorg Med Chem Lett 6: 2073-2076.[CrossRef]

Newell JG and Czajkowski C (2003) The GABA_A receptor 1 subunit Pro¹⁷⁴-Asp¹⁹¹ segment is involved in GABA binding and channel gating. J Biol Chem 278: 13166-13172.[Abstract/Free Full Text]

Newell JG, McDevitt RA, and Czajkowski C (2004) Mutation of glutamate 155 of the GABA_A receptor β2 subunit produces a spontaneously open channel: a trigger for channel activation. J Neurosci 24: 11226-11235.[Abstract/Free Full Text]

Ragozzino D, Woodward RM, Murata Y, Eusebi F, Overman LE, and Miledi R (1996) Design and in vitro pharmacology of a selective -aminobutyric acid C receptor antagonist. Mol Pharmacol 50: 1024-1030.[Abstract]

Sali A and Blundell TL (1993) Comparative protein modeling by satisfaction of spatial restraints. J Mol Biol 234: 779-815.[CrossRef][Medline]

Sedelnikova A, Smith CD, Zakharkin SO, Davis D, Weiss DS, and Chang Y (2005) Mapping ₁ GABA_C receptor agonist binding pocket: constructing a complete model. J Biol Chem 280: 1535-1542.[Abstract/Free Full Text]

Sigel E, Baur R, Kellenberger S, and Malherbe P (1992) Point mutations affecting antagonist affinity and agonist dependent gating of GABA_A receptor channels. EMBO J 11: 2017-2023.[Medline]

Smith GB and Olsen RW (1994) Identification of a [³H]muscimol photoaffinity substrate in the bovine -aminobutyric acidA receptor subunit. J Biol Chem 269: 20380-20387.[Abstract/Free Full Text]

Venkatachalam CM, Jiang X, Oldfield T, and Waldman M (2003) LigandFit: a novel method for the shape-directed rapid docking of ligands to protein active sites. J Mol Graph Model 21: 289-307.[CrossRef][Medline]

Westh-Hansen SE, Rasmussen PB, Hastrup S, Nabekura J, Noguchi K, Akaike N, Witt MR, and Nielsen M (1997) Decreased agonist sensitivity of human GABA_A receptors by an amino acid variant, isoleucine to valine, in the 1 subunit. Eur J Pharmacol 329: 253-257.[Medline]

Westh-Hansen SE, Witt MR, Dekermendjian K, Liljefors T, Rasmussen PB, and Nielsen M (1999) Arginine residue 120 of the human GABA_A receptor 1 subunit is essential for GABA binding and chloride ion current gating. Neuroreport 10: 2417-2421.[Medline]

Whiting P, Wafford K and McKernan R (2000) Pharmacological subtypes of GABA_A receptors based on subunit composition, in GABA in the Nervous System: The View at Fifty Years (Martin D and Olsen R eds) pp 113-126, Lippincott Williams & Wilkins, Philadelphia.

Zhang D, Pan ZH, Awobuluyi M, and Lipton SA (2001) Structure and function of GABA_C receptors: a comparison of native versus recombinant receptors. Trends Pharmacol Sci 22: 121-132.[CrossRef][Medline]

This Article Abstract Full Text (PDF) All Versions of this Article: mol.108.048710v1 74/4/941    most recent Submit a response Alert me when this article is cited Alert me when eLetters are posted Alert me if a correction is posted Services Similar articles in this journal Similar articles in PubMed Alert me to new issues of the journal Download to citation manager Citing Articles Citing Articles via Google Scholar Google Scholar Articles by Zhang, J. Articles by Chang, Y. Search for Related Content PubMed PubMed Citation Articles by Zhang, J. Articles by Chang, Y.