The GABAA receptor alpha 1 subunit Pro174-Asp191 segment is involved in GABA binding and channel gating.

The GABA-binding site undergoes structural rearrangements during the transition from agonist binding to channel opening. To define possible roles of the GABA(A) receptor alpha(1) subunit Pro(174)-Asp(191) segment in these processes, we used the substituted cysteine accessibility method to characterize this region. Each residue was individually mutated to cysteine, expressed with wild-type beta(2) subunits in Xenopus laevis oocytes, and examined using two-electrode voltage clamp. Most mutations did not alter GABA EC(50) values. The D183C mutation produced a 7-fold reduction in GABA sensitivity. There were no significant changes in the K(I) values for the competitive antagonist, SR-95531. N-Biotinylaminoethyl methanethiosulfonate modified P174C-, R176C-, S177C-, V178C-, V180C-, A181C-, D183C-, R186C- and N188C-containing receptors. The pattern of accessibility suggests that this protein segment is aqueous-exposed and adopts a random coil conformation. Both GABA and SR-95531 slowed covalent modification of V178C, V180C, and D183C, indicating that these residues may line the GABA-binding site. Further, pentobarbital-induced channel activation accelerated modification of V180C and A181C and slowed the modification of R186C, suggesting that this region of the alpha(1) subunit may act as a dynamic element during channel-gating transitions.

Allosteric transitions of neurotransmitter binding sites remain poorly understood, despite increased efforts in recent years to map protein domains important for ligand recognition and ion channel activation. It is likely that amino acid residues other than those that mediate initial contact with agonist will be important for inducing gating transitions. Characterization of receptor-ligand interactions using site-directed mutagenesis and photolabeling studies provides limited information as to the state-dependent nature of ligand binding domains (1). Nevertheless, identification of all residues lining the neurotransmitter binding-site, irrespective of the conformational state of the receptor, represents a critical step to understanding receptor-ligand interactions at allosteric proteins.
Identification of amino acid residues important in agonist/ antagonist binding at ␥-aminobutyric acid type A receptors (GABA A R) 1 reveals that the GABA-binding sites are located at ␤-␣ subunit interfaces. Consistent with the agonist-binding site of nicotinic acetylcholine receptors (nAChR), the GABA-binding site is formed by amino acid residues clustered in noncontinuous protein segments of the extracellular aminoterminal domains of adjacent subunits. Multiple residues have been implicated in the formation of this binding site using a variety of approaches, including site-directed mutagenesis, photoaffinity labeling, and the substitutedcysteine accessibility method (SCAM). These include Phe 64 , Arg 66 , Arg 119 , and Ile 120 of the ␣ 1 subunit (2)(3)(4)(5)(6)(7), in addition to Tyr 97 , Leu 99 , Tyr 157 , Thr 160 , Thr 202 , Ser 204 , Tyr 205 , Arg 207 , and Ser 209 of the ␤ 2 subunit (8 -10). Of these residues, it is likely that some contact agonist/antagonist molecules directly, some maintain the overall structure of the binding site, while others mediate conformational dynamics within the site during allosteric transitions among the resting, active, and desensitized states.
The GABA A R ␣ 1 subunit segment between Pro 174 and Asp 191 is homologous in position to the putative "loop F" of the nAChR (see Fig. 1) (11). Studies of this segment of the nAChR ␥/⑀ and ␦ subunits have identified negatively charged amino acid residues that influence acetylcholine binding, channel gating, and perhaps potassium ion interactions (see Fig. 1) (12)(13)(14)(15)(16). Based on the crystal structure of a soluble acetylcholine-binding protein (AChBP), a protein homologous to the extracellular domain of the nAChR, the secondary structure of the loop F region is predicted to be a random coil (17). Strikingly, the loop F protein sequence is poorly conserved among all GABA A R subunit isoforms and other related ligand-gated ion channel subunits and may represent a unique structural element that could account for differences in agonist affinity, dimensions of binding pockets, and access pathways important for receptorligand interactions. Therefore, an analysis of the structure and the role(s) of the ␣ 1 subunit Pro 174 -Asp 191 segment in ligand binding and ion channel activation is fundamental for understanding GABA A R function.
The development of SCAM has proved to be very powerful tool for identifying residues important for the pharmacology of both agonists and antagonists. Originally developed to identify the channel-lining residues of ligand-gated ion channels (18), SCAM has gained widespread use in the study of the ligand binding domains of these channels (2, 3, 9, 10, 19 -25). The method entails introduction of successive cysteine residues, one at a time, within a protein domain and expression of recombinant receptors in heterologous systems. Solvent accessibility of a given cysteine is determined by monitoring changes in function following application of a sulfhydryl-specific modifying reagent (18). The role of a given residue in the formation of a ligand binding site is determined by the ability of both agonists and antagonists to impede modification of the introduced cysteine by the sulfhydryl-specific reagent.
Here, we used SCAM to examine the structure, solvent accessibility, and dynamics of the GABA A receptor ␣ 1 subunit Pro 174 -Asp 191 region, which comprises the putative loop F of the GABA binding pocket. We demonstrate that this region is highly accessible and adopts a random coil/turn conformation. In addition, we identify several residues, Val 178 , Val 180 , and Asp 183 , that likely participate in forming part of the GABAbinding pocket. Moreover, we provide evidence that this region of the receptor undergoes conformational rearrangements during pentobarbital-mediated gating of the channel. The results are discussed in terms of a homology model of the GABA A R agonist binding site, based on the recently solved crystal structure of the AChBP (17).

EXPERIMENTAL PROCEDURES
Mutagenesis and Expression in Oocytes-Rat cDNAs for the ␣ 1 and ␤ 2 subunits of the GABA A receptor were used in this study. The ␣ 1 cysteine mutants were engineered using a recombinant polymerase chain reaction method, as described previously (2,10,26 (Fig. 1). Cysteine substitutions were verified by restriction endonuclease digestion and double-stranded DNA sequencing.
Two-electrode Voltage Clamp Analysis-Oocytes under two-electrode voltage clamp were perfused continuously with ND96 at a rate of ϳ5 ml/min. The holding potential was Ϫ80 mV. The volume of the recording chamber was 200 l. Standard two-electrode voltage clamp procedures were carried out using a GeneClamp500 Amplifier (Axon Instruments, Inc.). Borosilicate electrodes were filled with 3 M KCl and had resistances of 0.5-3.0 M⍀ in ND96. Stock solutions of GABA (Sigma) and SR-95531 (Sigma) were prepared in water, while N-biotinylaminoethyl methanethiosulfonate (100 mM) (MTSEA-biotin, Biotium, Hayward, CA) was prepared in dimethyl sulfoxide (Me 2 SO). All compounds were prepared fresh daily and MTSEA-biotin was diluted appropriately in ND96 such that the final concentration of Me 2 SO was Յ2%. This solvent concentration did not affect recombinant GABA A R.
To measure the sensitivity to GABA, the agonist (0.0001-1 mM) was applied via gravity perfusion or by pipettor application (ϳ5-8 s) with a 3-15-min washout period between each application to ensure complete recovery from desensitization. Peak GABA-activated current (I GABA ) was recorded. To correct for slow drift in the maximum amplitude of the response as a function of time, concentration-response data were normalized to a low concentration of GABA (EC 2 -EC 5 ). Concentrationresponse curves were generated for each recombinant receptor, and the data were fit by non-linear regression analysis using GraphPad Prism software (San Diego, CA; graphpad.com). Data were fit to the following where I is the peak amplitude of the current for a given concentration of GABA ([A]), I max is the maximum amplitude of the current, EC 50 is the concentration required for halfmaximal receptor activation, and n is the Hill coefficient.
To measure the sensitivity to SR-95531, GABA (EC 50 ) was applied via gravity perfusion followed by a brief (20 s) washout period before co-application of GABA (EC 50 ) and increasing concentrations of SR-95531. The response to the application of SR-95531 and GABA was normalized to the response elicited by the agonist alone. Concentration-inhibition curves were generated for each recombinant receptor, and the data were fit by non-linear regression analysis using GraphPad Prism software. Data were fit to the following equation: where IC 50 is the concentration of antagonist ([Ant]) that reduces the amplitude of the GABA-evoked current by 50% and n is the Hill coefficient. K I values were calculated using the Cheng-Prussof correction: is the concentration of GABA used in each experiment and EC 50 is the concentration of GABA that elicits a half-maximal response for each receptor (30).
Modification of Introduced Cysteine Residues by MTSEA-biotin-MTSEA-biotin was the sulfhydryl-specific reagent used in this study. It is a relatively impermeant compound (31) with dimensions (14.5 Å unreacted moiety; 11.2 Å reacted moiety) that are similar to SR-95531 (13.5 Å) but much longer than GABA (4.5 Å). Methanethiosulfonate reagents react 10 9 -10 10 times faster with the ionized thiolate (RS-) form of cysteine than the unionized form (32). Based on these properties, it is reasonable to assume that MTSEA-biotin can occupy the GABA-binding site and that this reagent will principally modify extracellular cysteine residues that are solvent-exposed.
Oocytes expressing either wild-type or mutant receptors were activated by GABA (EC 50 ) at regular intervals until the peak current amplitude varied by Յ10% on two consecutive applications. Oocytes were then allowed to fully recover, after which a high concentration of MTSEA-biotin (2 mM) was applied (2 min). Following MTSEA-biotin application, cells were washed (5 min) with ND96, after which GABA (EC 50 ) was again applied to determine the effect of MTSEA-biotin application on I GABA . The effect of MTSEA-biotin was calculated as the difference in the amplitude of the I GABA before and after MTSEA-biotin application as follows: (I GABApre Ϫ I GABApost /I GABApre ) ϫ 100, where post refers to the amplitude of I GABA following MTSEA-biotin application and pre refers to the amplitude of I GABA prior to exposure to MTSEA-biotin.
Rate of Modification of Introduced Cysteine Residues-Rates were measured only for those cysteine mutants that had a Ͼ40% change in I GABA following MTSEA-biotin treatment (2 min, 2 mM). The rate at which MTSEA-biotin modified introduced cysteine residues was measured using low MTSEA-biotin concentrations as described previously (3). In general, the concentration of MTSEA-biotin used was 50 M, with the exception of A181C (500 nM) and R186C (5 M). The experimental protocol is described as follows: GABA (EC 50 ) application (5 s); ND96 wash-out (25 s); MTSEA-biotin application (10 -20 s); ND96 washout (2.2-2.3 min). The sequence was repeated until I GABA no longer changed following the MTSEA-biotin treatment (i.e. the control reaction had proceeded to apparent completion). The individual abilities of GABA, SR-95531, and pentobarbital to alter the rate of cysteine modification by MTSEA-biotin were determined by co-applying either GABA (5 ϫ EC 50 ), SR-95531 (40 ϫ K I ), or an activating concentration of pentobarbital (500 M) during the MTSEA-biotin pulse. In all cases, the wash times were adjusted to ensure that currents obtained from test pulses of GABA (EC 50 ) following exposure to high concentrations of GABA, SR-95531, or pentobarbital were stabilized. This ensured complete wash-out of drugs and that any reductions in the current amplitude were the result of MTSEA-biotin application.
FIG. 1. The Pro 174 -Asp 191 segment (loop F) of the rat GABA A R ␣ 1 subunit is aligned with analogous regions of the rat GABA A R ␤ 2 and ␥ 2 subunits and rat nAChR ␥, ⑀, and ␦ subunits. The numbering reflects the position of the residues in the mature GABA A R ␣ 1 subunit. Residues implicated in acetylcholine binding are circled and include ␦Asp 180 , ␦Glu 189 , ␥Asp 174 , ⑀Asp 175 , and ⑀Asn 182 (12)(13)(14)16), while residues that line the GABA-binding site are boxed. Residues implicated in the interactions of divalent cations are underlined (15,46). The asterisks (*) indicate gaps in the amino acid sequence alignment.
For all rate experiments, the decrease in I GABA was plotted as a function of the cumulative time of MTSEA-biotin exposure and fit to a single-exponential decay function using GraphPad Prism software. A pseudo-first order rate constant (k 1 ) was determined and the second order rate constant (k 2 ) was calculated by dividing k 1 by the concentration of MTSEA-biotin used in the assay (33). Second order rate constants were determined using at least two different concentrations of MTSEA-biotin.
Statistical Analysis-log (EC 50 ) and log (K I ) values were analyzed using a one-way analysis of variance, followed by a post-hoc Dunnett's test to determine levels of significance between wild-type and mutant receptors. Differences among the second order (k 2 ) rates of covalent modification of the various mutants were assessed using the false positive discovery rate method (34). This method limited the expected percent of false positives to 5%. The false positive discovery rate is a more meaningful measure of error in large screening experiments than the more traditional approach of limiting the probability of one or more false positives (also known as experiment-wise error control). Before analysis, the rates were transformed to a log scale to obtain more normally distributed residuals. Results are reported in the original scale. Even using this approach, clear trends in the data did not always achieve significance as has been noted in other large assays using SCAM (35).
Structural Modeling-The mature protein sequences of the rat ␣ 1 and ␤ 2 subunits were homology-modeled with a subunit of the AChBP (17). The crystal structure of the AChBP was downloaded from the Research Collaboratory for Structural Bioinformatics Protein Data Bank (code 1I9B) and loaded into Swiss Protein Bank Viewer (SPDBV, ca.expasy.ord/spdbv). The ␣ 1 protein sequence from Thr 12 -Ile 227 and the ␤ 2 protein sequence from Ser 10 -Leu 218 were aligned with the AChBP primary amino acid sequence as depicted in Cromer et al. (36) and threaded onto the AChBP tertiary structure using the "Interactive Magic Fit" function of SPDBV. The threaded subunits were imported into SYBYL (Tripos, Inc., St. Louis, MO) where energy minimization was carried out (Ͻ0.5 kcal/Å). The first 100 iterations were carried out using Simplex minimization (37) followed by 1000 iterations using the Powell conjugate gradient method (38). A ␤ 2 /␣ 1 GABA-binding site interface was assembled by overlaying the monomeric subunits on the AChBP scaffold, the resulting structure was imported into SYBYL, and energy was minimized. Our model is quite similar to models recently published for the nAChR and GABA A R ligand binding domains (36,39). It is worth noting that positioning of the ␣ 1 subunit Pro 174 -Asp 191 (loop F) region is inexact since the ␣ 1 subunit has low homology to AChBP sequence in this region and contains three additional amino acid residues. This region and other regions with insertions were modeled by fitting structures from a loop data base.   Table I). The lack of functional expression of receptors carrying the L187C and Q189C mutations may indicate a role for these residues in receptor synthesis/assembly as they are conserved in all GABA A R and glycine receptor subunits. Expression of D183C produced a significant 7-fold rightward shift in EC 50 relative to wild-type values (EC 50 ϭ 1.6 M). However, the K I values for the competitive antagonist, SR-95531, for mutant receptors were not significantly different from wild-type values (K I ϭ 330 nM). Hill coefficients were not significantly different from wild type (Table I). In general, the maximum current amplitude was 1-10 A for wild-type and mutant receptors, with the exception of R186C (Ͻ300 nA).

Expression and Functional Characterization of GABA
These data suggest that cysteine substitution within this domain of the GABA A R ␣ 1 subunit protein is well tolerated. A major assumption of SCAM is that the side chain of the introduced cysteine is in a similar position as the side chain of the native residue. Since GABA and SR-95531 bind equally well to both mutant and wild-type receptors, it is likely that the structures of the receptors are similar.
Modification of Introduced Cysteine Residues by MTSEAbiotin-To define the surface accessibility of the ␣ 1 subunit P174C-D191C segment, wild type and mutant receptors were exposed to MTSEA-biotin (2 mM) for 2 min (Fig. 3). MTSEAbiotin had no effect on wild-type receptors. MTSEA-biotin significantly reduced I GABA at P174C (60.5 Ϯ 1.1%, n ϭ 3), R176C  3 Ϯ 10%, n ϭ 3). An apparent lack of reaction (as in the case of A175C, V179C, E182C, and D191C) may indicate that no reaction has occurred or that the outcome of modification is functionally silent. It should be noted that most residues in this region were modified, although the magnitude of the effect of modification did not always achieve statistical significance (e.g. G184C, S185C, and Y190C). The pattern of solvent accessibility is not indicative of either a ␤-strand or an ␣-helix, suggesting that this domain of the GABA A R ␣ 1 subunit adopts either a loop or a random coil conformation (Fig. 6).
MTSEA-biotin Rates of Reaction-The rate at which MTSEA-biotin reacts with a cysteine side chain depends mainly on the ionization of the thiol group and the access route to the engineered cysteine (18). A residue in a relatively open, aqueous environment will react faster than a residue in a relatively restrictive, non-polar environment. To gain insight into the physico-chemical environment of the loop F region of the GABA-binding site, we determined the reaction rate of MTSEA-biotin with several accessible cysteine mutants (Fig.  4). The rate MTSEA-biotin modified A181C was ϳ400-fold faster than the slowest reacting cysteine mutant, V180C. The rank order k 2 values were A181C Ͼ R186C ϭ R176C Ϸ S177C Ͼ D183C Ϸ V180C ϭ V178C (Table II).
Effects of GABA and SR-95531 on MTSEA-biotin Rate Constants-To determine whether a given cysteine residue lines the neurotransmitter binding pocket, the rate of MTSEAbiotin modification of an introduced cysteine is measured in the presence of GABA and the competitive antagonist, SR-95531. We identify a residue as being within or near the binding site if the rate of covalent modification of the introduced cysteine is slowed in the presence of both agonists and antagonists, which presumably promote different conformational changes within the site. SR-95531 slowed the rate of modification at V178C, V180C, and D183C by factors of 3.6, 1.9, and 3.5, respectively (Fig. 4, Table II). GABA slowed the rate of reaction at R176C, V178C, V180C, and D183C (2.4-, 1.9-, 1.8-, and 3.5-fold, respectively). Protection of V178C, V180C, and D183C from covalent modification by MTSEAbiotin by GABA and SR-95531 suggests that the slowing of the MTSEA-biotin reaction rate results from steric block rather than allosteric changes induced in the protein. It is interesting to note that R176C was protected only by GABA but not SR-95531. S177C was protected significantly only by SR-95531. While the effects of GABA failed to reach statistical significance for this mutant, there was a clear trend in the data to suggest that GABA also slowed the MTSEA-biotin reaction rate (Table II, Fig. 5).
Effects of Pentobarbital on MTSEA-biotin Rate Constants-At wild-type ␣ 1 ␤ 2 or ␣ 1 ␤ 2 ␥ 2 GABA A R, the apparent affinity for direct activation by pentobarbital ranges from 500 -700 M (8, 10). Further, the mean single channel conductances elicited by GABA and pentobarbital are not different, suggesting that the open states produced by both ligands is similar (40). Moreover, mutations that compromise the affinity of GABA have thus far not affected the affinity or efficacy of barbiturates (8, 10), suggesting that the actions of pentobarbital are mediated from a site distinct from the GABA-binding site. Therefore, pento-  EC 50 ). B, sequential application of MTSEA-biotin reduced the amplitude of subsequent GABA-mediated (EC 50 ) currents. Data were normalized to the current measured at t ϭ 0 for each experiment and plotted as a function of cumulative MTSEA-biotin exposure. Data were fit to a single exponential function to obtain a pseudo-first order rate constant (k). Second order rate constants (k 2 ) were calculated by dividing the pseudo-first order rate constant by the concentration of MTSEA-biotin used (50 M). Data points represent the mean Ϯ S.E. for control (f), GABA (E), SR95531(q) for at least three independent experiments. k 2 values are shown in Table II.   TABLE II  Second order rate constants  barbital can be used as a pharmacological tool to assess gatinginduced changes in the GABA-binding site. The rate of modification at R186C was slowed 3.2-fold in the presence of pentobarbital, while the rates of covalent modification at V180C and A181C were accelerated 1.4-and 2.3-fold, respectively (Table II, Fig. 5). Thus, these residues act as reporters of barbiturate-mediated channel gating.

Structure of the GABA Binding Pocket-Previous work has
shown that the GABA binding pocket is composed of aromatic (␣ 1 Phe 64 , ␤ 2 Tyr 97 , ␤ 2 Tyr 157 , ␤ 2 Tyr 205 ), hydroxylated (␤ 2 Thr 160 , ␤ 2 Thr 202 , ␤ 2 Ser 204 , ␤ 2 Ser 209 ), and charged amino acid residues (␣ 1 Arg 66 , ␤ 2 Arg 207 ). Here, our data demonstrating that GABA and SR-95531 protect V178C, V180C, and D183C also indicate that residues in loop F are near the agonist-binding site. An additional residue, Arg 176 , may be important for interactions with the agonist alone as modification of R176C was protected by GABA and not SR-95531. Barbiturate-mediated receptor activation did not alter MT-SEA modification of R176C, suggesting that the observed slowing of the derivatization of R176C by GABA was a function of steric block, as opposed to channel-gating phenomena. Ligands of divergent chemical structure such as GABA and SR-95531 likely have different contact points within the GABA-binding site (3). However, the amino acid residues identified here need not be contact points for agonist/antagonist molecules, but they may be important for stabilizing the structure of the GABA-binding site or mediating local movements important for activation and/or desensitization.
When mapped onto a homology model of the GABA binding site, these residues appear to be located at the putative entrance of the binding site (Fig. 6). Using this model, we measured distances between loop F GABA-binding site residues and core GABA-binding regions.  (39,41). In addition, the loop F region was not well defined in the AChBP structure (17).
Previous work has demonstrated that the nAChR loop F is involved in agonist binding. Using a chemical cross-linker, Czajkowski and Karlin identified several negatively charged residues in loop F (␦Asp 180 , ␦Glu 182 , and ␦Glu 189 ) within 9 Å of the Cys 192 /Cys 193 loop of the ␣ subunit (13). These data suggest that, at least in some cases, the loop F domain of the ␦ subunit is in close proximity to residues on the ␣ subunit that are within the core of the ACh-binding site. In addition, recent studies have shown that naturally occurring mutations in the loop F protein chain of the ⑀ subunit (D175N, N182Y) alter ACh microscopic binding affinity and channel gating (16). FIG. 5. Summary of the effects of GABA, SR-95531, and pentobarbital on MTSEA-biotin second order rate constants. Data were normalized to control second order rate constants (rate measured when no other compound was present). Co-application of GABA (5 ϫ EC 50 ) or SR95531 (40 ϫ K I ) slowed reaction of MTSEA-biotin at receptors containing the following mutations: R176C, V178C and D183C, suggesting that they line the GABA-binding site. Data represent the mean Ϯ S.E. for at least three experiments. R176C was protected only by GABA, and while not significant, there is a clear trend in the data to suggest that S177C was protected by both agonist and antagonist. The rate of covalent modification at V180C, A181C, and R186C is significantly altered by pentobarbital (500 M). *, p Ͻ 0.05.
FIG. 6. A, model of the GABA-binding site at the ␤Ϫ␣ subunit interface illustrating the random coil structure of the ␣ 1 subunit loop F protein segment. Regions colored cyan correspond to cysteine mutants that were not accessible to MTSEA-biotin modification, and those residues that were accessible are colored yellow. B, GABA-binding site residues Val 178 and Val 180 (red) and Asp 183 (blue) are illustrated. C, position of Asp 183 in relation to other core GABA-binding site residues ␣ 1 Phe 64 and ␣ 1 Arg 66 from loop D (yellow), in addition to ␤ 2 Arg 207 and ␤ 2 Tyr 205 of loop C (red). Shown also is the predicted theoretical distance (12.0 Å) between ␤ 2 Arg 207 and ␣ 1 Asp 183 . The predicted distances between ␣ 1 Asp 183 and other core binding site residues are summarized under "Discussion." Structural Rearrangements during Gating Transitions-Allosteric proteins such as ligand-gated ion channels cycle through a number of affinity states, including a low affinity resting state, an active open channel state of moderate affinity and two desensitized states of high and very high affinity, respectively (42). During these state transitions, a molecule of GABA likely contacts a number of different residues. Residues important in the initial docking of the ligand may be different than residues involved in stabilizing ligand binding in open and desensitized states. It is likely that the GABA-binding site undergoes a series of transitions in which alternate domains of the protein are brought into closer contact with the ligand during active and desensitized states. It is equally possible that ligand interactions with amino acids in the inactive state are entirely different from those in the active and desensitized states (1), further complicating analysis of agonist binding segments.
Methanethiosulfonate reagents can be used as reporter molecules to detect agonist-or drug-induced changes in protein regions that are distant from the agonist or modulator binding site. GABA-induced structural rearrangements have been reported in the benzodiazepine-binding site (19) and in the ␣ 1 subunit M2-M3 loop (43). The allosteric modulators, diazepam and propofol, induce changes in the ␣ 1 subunit M3-spanning segment (35,44). In addition, we have previously demonstrated movements within the GABA-binding site in response to pentobarbital gating of the channel (3,10).
To test the hypothesis that movement of loop F is a plausible ion channel activation mechanism (14), we measured the rate of covalent modification of accessible amino acid residues in the presence of pentobarbital (500 M). The ability of pentobarbital to alter the rates of modification of the loop F segment provides an indirect measure of changes that occur within this region of the binding cleft in the transition from the resting to the active/desensitized states. Co-application of pentobarbital and MTSEA-biotin should capture a receptor state that differs from that captured by application of MTSEA-biotin alone. Pentobarbital-mediated acceleration of the rate of modification at V180C (a GABA-binding site residue) and A181C and the concomitant slowing of the rate of modification of R186C indicate that Val 180 and Ala 181 move to a more accessible environment, while Arg 186 becomes less accessible. These data demonstrate that the loop F region of the GABA binding site undergoes conformational rearrangements during receptor activation and/or desensitization. Other movements within the binding site may also be needed to trigger channel gating. For example, rotations and/or tilting movements of the ␤ 2 subunit may move the loop C region of the GABA-binding site closer to ␣ 1 subunit binding segments (45). CONCLUSIONS SCAM analysis has enabled us to identify novel residues of the ␣ 1 subunit (Val 178 , Val 180 , and Asp 183 ) that contribute to forming the GABA-binding site. Further, we provide evidence that the domain defined by Pro 174 -Asp 191 adopts a random coil/turn conformation. Barbiturate-mediated channel activation suggests that this segment of the protein undergoes conformational movements during channel gating. We speculate that this loop of the protein is a dynamic element that may move closer to the core of the binding site during allosteric transitions to higher affinity states. While this is a plausible channel-gating mechanism, corroboration of these SCAM observations will require studies using chemical cross-linkers to understand the relative positions of amino acids in this domain during the transduction of agonist binding to channel opening and desensitization.