Neuronal α7 nicotinic ACh receptors (nAChRs) are permeable to and modulated by Ca2+, Ba2+, and Sr2+. These permeant divalent cations interact with slowly desensitizing L247T α7nAChRs to increase the potency and maximal efficacy of ACh, increase the efficacy of dihydro-β-erythroidine (DHβE), and increase agonist-independent activity. Mutation of glutamate 172 (E172) to glutamine or cysteine eliminated these effects of permeant divalent cations. 2-(Trimethylammonium)ethyl methanethiosulfonate (MTSET), a cysteine-modifying reagent directed at water-accessible thiols, inhibited ACh-evoked currents of E172C/L247T α7 nAChRs by >90%, demonstrating that E172 was accessible to permeant ions. The data are consistent with a model of α7receptors, derived from the crystal structure of the ACh binding protein (AChBP) from Lymnaea stagnalis, in which E172 projects toward the lumen of the extracellular vestibule. The observations that E172 was essential for divalent cation modulation of L247T α7 nAChRs and was accessible to permeating ions suggest that this residue participates in coupling ion permeation with modulation of receptor activity.
- acetylcholine receptor
- Xenopus oocyte
- cysteine modification
the potent psychological effects of nicotine underscore the importance of neuronal nicotinic acetylcholine receptors (nAChRs) in the central nervous system (CNS) (11, 14, 21, 26, 29, 43). Neuronal nAChRs are ligand-gated ion channels formed by heteromeric complexes of α- and β-subunits or by homomeric assemblies of α-subunits. The most prevalent example of homomeric assembly is the homopentamer of α7-subunits (7, 12, 36). The functional expression of α7 nAChRs in Xenopus oocytes does not require coexpression of other α- or β-subunits, although α7-subunits may coassemble with other subunits in vivo (45).
Neuronal α7 nAChRs are widely distributed in the CNS and serve several different functions (7, 21). They mediate postsynaptic responses of hippocampal interneurons and other neurons (2, 18, 21, 22, 31). They modulate the release of excitatory and inhibitory neurotransmitters in the hippocampus and other regions (1, 30, 34). They are important in the early development and proliferation of neurons (34). In addition, α7 nAChRs promote cell survival in several experimental models of neurological damage, including hypoxia (40), β-amyloid exposure (24), deprivation of nerve growth factor (NGF) (27), andN-methyl-d-aspartate (NMDA) receptor-mediated excitotoxicity (13). The high Ca2+permeability of α7 nAChRs (8, 28, 37, 41,42) is likely to play a role in many of their functions.
In addition to Ca2+, α7 nAChRs are highly permeable to Ba2+ and Sr2+ (15,35). These permeant divalent cations modulate the activity of wild-type and L247T α7 nAChRs by increasing the potency and maximal efficacy of ACh (15, 19). In addition, the antagonist dihydro-β-erythroidine (DHβE) (11) is a strong partial agonist of L247T α7 nAChRs in the presence but not the absence of permeant divalent cations (15). Mutation of E237, located at the intracellular end of the transmembrane pore of α7 nAChRs, eliminates Ca2+ influx (4) but does not alter Ca2+-dependent increases in ACh potency (15). This suggests that the site(s) required for modulation by extracellular Ca2+ is in the extracellular or pore-lining regions. Because permeant but not impermeant divalent cations modulate L247T α7nAChRs (15), it is possible that sites necessary for modulation of α7 nAChRs by divalent cations are in the ion permeation pathway.
Another site shown to play a role in the modulatory effects of Ca2+ on α7 receptors is glutamate 172 (E172), located in the extracellular NH2-terminal domain (19). In α7-V201–5-HT3 receptors (a chimeric receptor consisting of the NH2-terminal extracellular domain of α7 nAChRs and the transmembrane and COOH-terminal domains of 5-HT3 receptors), mutation of E172 to glutamine (E172Q) abolishes Ca2+-induced increases in ACh potency and maximal efficacy (19).
The crystal structure of the ACh binding protein (AChBP), a soluble protein secreted by glial cells of the snail Lymnaea stagnalis (6), provides further insight into the role of E172 in modulating the activity of α7nAChRs. Computational mapping of the α7 nAChR sequence onto the crystal structure of the AChBP can be used to generate a model of the extracellular domain of α7 receptors. In this model, E172 lines the “bottom” of the outer vestibule, near the interface of the extracellular NH2 terminus and the transmembrane domains. Thus the model suggests that E172 is accessible to permeant divalent cations in the water-filled vestibule.
In this study, we characterized the role of E172 in the divalent cation permeation and modulation of L247T α7 nAChRs. We used the L247T α7receptor (3, 33) as the “background” for the introduction of E172 mutations because its high potency, high efficacy, slow desensitization, and large Ca2+-dependent shifts in potency (15) increased the resolution of effects of mutations and ionic conditions. We report that the potency and efficacy of ACh on α7receptors containing both E172Q and L247T mutations (E172Q/L247T α7 nAChRs) were unchanged in the presence of permeable divalent cations (Ca2+, Ba2+, and Sr2+). In addition, the E172Q mutation eliminated the high level of agonist-independent activity of L247T α7nAChRs (5) and abolished the agonist behavior of the antagonist DHβE on these receptors (3). We show that E172 was accessible to permeant ions, as evidenced by the inhibition of E172C/L247T α7nAChRs by 2-(trimethylammonium)ethyl methanethiosulfonate (MTSET), a cysteine-modifying reagent directed at water-accessible thiols (23). Thus E172 was essential for the modulation of L247T α7 nAChRs by divalent cations and was accessible to extracellular ions, suggesting that it is part of a mechanism that couples ion permeation with modulation of receptor gating.
Divalent cation salts were obtained from Fluka. DHβE was from RBI (Natick, MA). Gentamicin was from GIBCO-BRL (Gaithersburg, MD). Methanethiosulfonate (MTS) reagents were from Toronto Research Chemical (Toronto, ON, Canada). All other reagents were from Sigma.
Chick wild-type and L247T α7 nAChR cDNAs were obtained as previously described (15). E172Q and E172C mutations were introduced into L247T α7 receptors with the QuickChange method (Stratagene). All mutations were verified by DNA sequencing.
Oocytes were removed surgically and dissociated enzymatically following standard protocols (20, 38) as described previously (15). cRNA was transcribed in vitro with the T7 RNA polymerase using the mMessage mMachine kit from Ambion. Oocytes were injected with 20 ng of cRNA and incubated for 2–3 days at 18°C.
Currents were recorded under two-microelectrode voltage clamp (GeneClamp and pCLAMP6, Axon Instruments). Typically, oocytes were bathed in a normal extracellular solution (in mM: 96 NaCl, 2 KCl, 1 MgCl2, 10 HEPES, pH 7.4) containing 1 mM EGTA (ES-EGTA), 2.5 mM CaCl2 (ES-Ca2+), 2.5 mM BaCl2 (ES-Ba2+), or 2.5 mM SrCl2(ES-Sr2+). Solution flow was continuous throughout an experiment (∼4 ml/min in a chamber of ∼100 μl). To prevent activation of the endogenous Ca2+-activated Cl− channels, the oocytes were injected with 46 nl of 50–100 mM BAPTA 15–60 min before recording, as described previously (15). Oocytes were usually held at a voltage of −60 mV, and responses were measured during superfusion with ACh. Oocytes were rinsed in the extracellular solutions for 3–4 min between ACh applications.
To measure currents carried by Ca2+ and other divalent ions, responses were recorded in bathing solutions containingN-methyl-d-glucamine methanesulfonate (NMG-MeS; 90 mM NMG, 10 mM HEPES, methanesulfonic acid to pH 7.4). NMG-MeS was supplemented with 10 mM Ca2+-gluconate, Ba(OH)2, or Sr(OH)2.
Agonist-independent activity was obtained from measurements of the block of basal “leak” currents by 1 μM methyllycaconitine (MLA). For each oocyte tested, the maximal response to ACh was obtained, an extensive wash was performed until baseline leak was stable for >1 min, and then the MLA-blocked leak was recorded. The MLA-blocked leak was normalized by the maximal ACh response.
MTS reagents were dissolved in water and kept at 4°C on the day of the experiment. Because these reagents are labile in physiological buffers (Toronto Research Chemicals), they were diluted in ES-Ca2+ to the final treatment concentration within 1–2 min of their application to an oocyte. Oocytes were exposed to the reagents for 0.5–5 min, superfused with ES-Ca2+, and retested for ACh responses.
Analysis of dose-response curves.
Dose-response relationships were fitted to the Hill equation with Prism software (GraphPad Software, San Diego, CA). To control for rundown during the acquisition of the data, responses were normalized to the peak currents from repeated applications of a standard dose of ACh (50 and 30 μM for E172Q/L247T and E172C/L247T α7 receptors, respectively). An F-test was performed to determine whether there were statistically significant differences in the EC50 values determined under different conditions. Data plotted and values reported are means ± SE. Statistical significance was assumed at P < 0.05.
Structural model of α7.
A model of the chick α7 nicotinic receptor was based on the coordinates of the AChBP (6). The BIOINBGU fold-recognition server (http://www.cs.bgu.ac.il/∼bioinbgu/; Ref.17) identified the monomer of the AChBP homopentamer (PDB ID 1I9B.pdb) as a structure compatible with the chick α7sequence. A model of the α7 monomer was based on chain A of the AChBP pentamer by using the Modeler module of the InsightII system (www.accelrys.com). A homopentamer of α7 was generated by superimposing five different copies of the α7 homology model on each of the monomers in the AChBP homopentamer. The Profiles-3D module of the InsightII molecular modeling system was used to evaluate the compatibility of the model of the α7 pentamer with the α7 sequence. The imaging program SPOCK (http://mackerel.tamu.edu/spock/) was used to create figures of the homology model.
ACh-evoked responses of E172Q/L247T are slowly desensitizing.
Figure 1 shows some of the key functional properties of the E172Q/L247T α7 receptor. Figure 1 A shows that E172Q/L247T α7 nAChRs, like L247T α7 receptors (33), displayed slow desensitization kinetics. Prolonged application of ACh produced sustained responses similar to those of the L247T α7 receptor. ACh dose-response curves (Fig.1 B) reveal that the EC50 of E172Q/L247T α7 receptors fell between those of L247T α7 nAChRs and wild-type α7 receptors. In the presence of 2.5 mM Ca2+, the EC50s for wild-type, E172Q/L247T, and L247T receptors were 180, 38, and 0.34 μM, respectively (see also Refs.12, 15, 33). The results show that the slow desensitization of L247T α7receptors was maintained in E172Q/L247T α7 nAChRs even though E172Q/L247T α7 receptors did not display the extremely low EC50 for ACh that was seen for L247T α7 receptors (33).
Ca2+, Ba2+, and Sr2+ do not stimulate E172Q/L247T α7 nAChRs.
Previous reports demonstrated that permeant divalent cations modulate agonist responses of L247T α7 ACh receptors (15, 19). In the presence of permeant divalent cations, ACh activates the L247T α7 receptor with an EC50 that is ∼10-fold lower than in the presence of EGTA (15). In addition, the maximal efficacy of ACh is increased in the presence of permeant divalent cations (15,19). In α7-5-HT3 chimeric receptors, the E172Q mutation abolishes the Ca2+-dependent increase in potency (19).
To evaluate the contribution of E172 to the modulation of L247T α7 receptors by permeant divalent cations, dose-response curves of the E172Q/L247T α7 receptors were compared in normal extracellular solutions containing 1 mM EGTA or 2.5 mM Ca2+, Ba2+, or Sr2+. E172Q/L247T α7 receptors displayed activation and slow desensitization kinetics that were indistinguishable from those of L247T α7receptors and were the same in the presence or absence of Ca2+ (Fig. 2,inset) or other permeant divalent cations. In the presence of Ca2+, Ba2+, or Sr2+, ACh had higher EC50 values and lower maximal efficacies on E172Q/L247T receptors than in the absence of these permeant divalent cations (Fig. 2). Specifically, the receptors had an EC50 of 23 μM in ES-EGTA, compared with 35, 35, and 39 μM in the presence of Ca2+, Ba2+, and Sr2+, respectively. The maximal efficacies of ACh were 87%, 75%, or 56%, in the presence of Ba2+, Ca2+, or Sr2+, respectively, compared with those obtained in the presence of EGTA. Thus, as in the α7-5-HT3 chimeric receptor, the E172Q mutation eliminated the divalent cation-dependent enhancement of L247T α7 receptor responses. In fact, E172Q/L247T α7 nAChRs were inhibited by Ca2+, Ba2+, and Sr2+ in that EC50 values were increased and maximal efficacies were decreased. These observations underscore the complexity of interactions between a divalent ion modulation site (E172), the ligand binding site, and the permeation pathway.
Ca2+, Ba2+, and Sr2+ carry inward currents through E172Q/L247T nAChRs.
To determine whether the loss of modulation by permeant divalent cations was due to the elimination of divalent permeation, we measured the currents carried by Ca2+, Ba2+, and Sr2+ in NMG-MeS solutions (Fig.3). In an NMG-MeS bathing solution, ACh evoked a small outward current at −60 mV that was probably carried by intracellular K+, indicating that NMG+ does not carry any inward current (Fig. 3, top trace). In NMG-MeS solutions containing 10 mM Ca2+, Ba2+, or Sr2+, ACh evoked clear inward currents. The activation and desensitization kinetics of ACh-evoked currents obtained in Ca2+-, Ba2+-, or Sr2+-containing NMG-MeS solutions were indistinguishable from those recorded in the normal Na+-containing extracellular solution. Thus the loss of Ca2+-, Ba2+-, and Sr2+-dependent changes in ACh EC50 in E172Q α7receptors was not due to an inability of these ions to permeate the receptors. Quantitatively, the flux of divalent ions through E172Q/L247T α7 nAChRs was significantly smaller than that of L247T α7receptors (15). The currents of E172Q/L247T α7 nAChRs carried by 10 mM Ca2+, Ba2+, or Sr2+ were ∼10% as large as those carried by 96 mM Na+ (Fig. 3), whereas the currents of L247T α7 receptors carried by the divalent cations were the same size as those carried by Na+ (15). These data suggest a role for E172 in divalent ion permeation.
Mutation of E172 reduces basal activity of L247T α7 nAChRs.
In addition to having an extremely low EC50 for ACh, L247T α7 receptors have a detectable level of activity in the absence of agonist (5). This activity is evident as an inhibition of basal current by MLA (32), a potent and selective blocker of α7 neuronal nAChRs (32, 44), including L247T α7receptors. Permeant divalent cations cause a 10-fold increase in the basal activity of L247T α7 receptors (15). To determine whether this agonist-independent activity of L247T α7 receptors was altered by mutations of E172, we measured the MLA-inhibited basal currents of oocytes expressing E172Q/L247T α7 receptors.
Figure 4 shows MLA-sensitive basal currents of L247T and E172Q/L247T α7 receptors in ES-Ca2+ and ES-EGTA. In ES-Ca2+, L247T α7 receptors had a significantly higher level of basal activity than did E172Q/L247T α7 receptors (Fig.4). In ES-EGTA, both receptor types had low levels of basal activity (Fig. 4). E172Q/L247T α7receptors also had low basal activity in the presence of Ba2+ and Sr2+ (not shown). These data show that E172 is essential for the stimulation of basal, agonist-independent activity of L247T α7nAChRs by permeant divalent cations.
Mutation of E172 eliminates activation of L247T α7 nAChRs by DHβE.
L247T α7 nAChRs are activated by DHβE, an antagonist of wild-type neuronal nAChRs (3). In the presence of permeant divalent cations, DHβE is a full agonist of L247T α7 receptors (3, 15). To determine whether E172 is critical for the activation of L247T α7 receptors by antagonists, we tested the efficacy of DHβE on E172Q/L247T α7 nAChRs. Figure 5,left, confirms that activation and desensitization of ACh- and DHβE-evoked responses of L247T α7nAChRs were virtually identical. In contrast, DHβE failed to activate E172Q/L247T α7 receptors (Fig. 5,right), even at concentrations as high as 300 μM. In ES-Ca2+, the maximal DHβE-evoked responses of E172Q/L247T α7 receptors were <0.04% of ACh-evoked responses (n = 9). In ES-EGTA, no responses to DHβE were recorded (n = 4). In the presence of an EC50 dose of ACh in ES-Ca2+ (38 μM), DHβE was an inhibitor of E172Q/L247T α7 nAChRs with an IC50 of ∼1 μM (not shown). This IC50 is similar to that recorded for wild-type α7 nAChRs (IC50 ∼0.7 μM; see Ref.9). DHβE also failed to activate E172Q/L247T α7 nAChRs in the presence of 2.5 mM Ba2+ or Sr2+ (n = 9).
E172 is accessible to the aqueous milieu.
A model of the extracellular domain of α7 receptors, based on the crystal structure of AChBP, suggests that E172is at a position near the bottom of the outer vestibule of nAChRs (Fig.6). E172 is located at the interface between two adjacent subunits, and the side chain appears to extend toward the lumen of the vestibule. Thus E172 may be positioned to interact directly with permeant ions, congruent with the observations that E172 is important for mediating the increase in agonist potency of α7 receptors by Ca2+ and other permeant divalent cations.
To test whether E172 was accessible to the aqueous milieu as suggested by the model, we introduced a cysteine at position 172 and tested whether MTSET, a cysteine-modifying reagent that reacts preferentially with water-accessible thiols, could alter ACh-evoked currents. E172C/L247T α7receptors, like E172Q/L247T receptors, had an EC50 for ACh that was insensitive to the presence of Ca2+ (Fig. 7 A) and showed activation and desensitization kinetics that were similar to those of L247T α7 nAChRs (Fig. 7 A, inset; ES-CA2+). MTSET, which covalently couples bulky, charged thioethyltrimethylammonium groups (6.9 Å in length; Ref. 39) to free sulfhydryls, caused complete inhibition of E172C/L247T α7 receptors (Fig.7 B). In contrast, MTSET had no effect on L247T α7 receptors (Fig. 7 C). Methylmethanethiosulfonate (MMTS), a reagent that covalently couples a thiomethyl group to free sulfhydryls, caused partial inhibition of E172C/L247T α7 receptors (Fig.7 D). The partial inhibition by MMTS occluded any further inhibition by MTSET, indicating that the modification by MMTS was complete. ACh dose responses of E172C/L247T α7 receptors before and after modification by MMTS indicated that the inhibition was due to both a decrease in maximal efficacy by ∼50% and an increase in EC50 from ∼20 to ∼130 μM (not shown). Thus MMTS modification of the cysteine at position 172 changed the activation of the receptor and did not simply alter the permeation pathway. Nevertheless, the inhibition by MTSET shows that position 172 is accessible to the extracellular aqueous milieu, where it could interact with extracellular divalent cations.
We have shown that the E172Q and E172C mutations of L247T α7 nicotinic receptors eliminate receptor stimulation by the presence of permeable divalent cations. ACh dose responses of E172Q/L247T α7 nAChRs were inhibited by the presence of Ca2+, Ba2+, or Sr2+, and E172C/L247T α7 nAChRs were insensitive to the presence of these ions. E172Q/L247T α7 nAChRs did not display the high level of basal activity that is a characteristic of L247T α7 receptors and were not activated by the antagonist DHβE. E172C/L247T α7 receptors were blocked by the thiol-modifying reagent MTSET, indicating that E172 is accessible to permeant ions. These data support the conclusion drawn from a chimera of α7 nicotinic receptors and 5-HT3 receptors that E172 is essential for the modulation of α7 receptors by Ca2+ and other permeant divalent cations (19). The data are also consistent with a model of α7 receptors based on the crystal structure of AChBP (6) that places E172 near the inner surface of the vestibule, where it can come in contact with the extracellular solution.
Our results do not support a kinetic model for the gating of L247T α7 receptors in which the receptors enter a conducting desensitized state (3). This model is based on the observations that slowly desensitizing L247T α7 receptors are activated by agonists at very low concentrations (suggesting binding to the high-affinity desensitized states), are activated by antagonists, and have a high level of agonist-independent basal activity (attributed to resting desensitization). However, E172Q/L247T α7 nAChRs have a slow desensitization rate similar to that of L247T α7 receptors, but E172Q/L247T α7 nAChRs are not activated by very low concentrations of ACh, are not activated by DHβE, and do not have a high level of resting activity. Thus in E172Q/L247T α7 nAChRs the phenomenon of slow desensitization is not accompanied by other characteristics suggesting that a desensitized state is conducting.
The data are consistent with an alternative model in which L247T α7 receptors have altered activation and desensitization kinetics (15, 16, 25). In this model, the prolonged response of the receptor to agonists arises from a decrease in the rate of entry into the desensitized state. Activation by low concentrations of agonists, agonist activity of antagonists, and elevated resting activity all derive from a dramatically increased opening rate, increasing the intrinsic efficacy of the receptor (10). In this interpretation, the E172Q/L247T α7 receptors have a slow desensitization rate similar to that of L247T α7 receptors but have an opening rate that is closer to that of wild-type α7 nAChRs. We conclude that L247T α7 nAChRs and other mutants derived from L247T α7 receptors are a reasonable model for wild-type α7 receptors because the mutations alter the rates of transitions between closed, open, and desensitized states but do not critically alter the nature of those states.
The observation that E172C/L247T α7 receptors were inhibited by extracellular MTSET is consistent with a structural model of the vestibule of α7nAChRs based on the crystal structure of AChBP (6). In the model, E172 in chick α7 nAChRs (R170 of AChBP) is located at the NH2 terminus of the β9 strand, extending toward the central pore, where it could contact the aqueous environment in the lumen of the pentamer (Fig. 6; Ref. 6). The accessibility of cysteine at position 172 to extracellular MTSET supports this vestibule-lining location of E172. This glutamate is conserved in ACh receptors and 5-HT3 receptors, but not in AChBP, GABA receptors, or glycine receptors (6), further supporting the view that it may have a role in cation permeation and/or modulation. The modification of the cysteine at position 172 by MMTS may inhibit the receptor by altering interactions at the subunit-subunit interface. In addition, it is possible that modification by MTSET and MMTS also partially interferes with ion permeation.
Ca2+ interacts with one or more low-affinity binding sites during permeation through α7 receptors (28). Even though E172Q/L247T α7receptors conduct divalent cations, they display divalent current amplitudes that are smaller than those of L247T α7 receptors (15). Thus E172 may be part of Ca2+-binding sites that are important in both ion permeation and modulation. Additional binding sites that do not involve E172 are also likely. Our results support the view that the outer vestibule of nicotinic receptors, especially near the junction between the vestibule and the transmembrane pore, plays an essential role in the modulation of α7 receptors by Ca2+ and other permeant divalent cations.
We thank M. Ballivet (University of Geneva) for the α7 nAChR cDNA and C. Labarca and H. A. Lester (California Institute of Technology) for the pAMV vector. We also thank B. Temple (University of North Carolina at Chapel Hill) for assistance with molecular modeling.
This work was supported by National Institute of Neurological Disorders and Stroke Grant NS-37317.
Address for reprint requests and other correspondence: R. L. Rosenberg, Dept. of Pharmacology, CB# 7365, Univ. of North Carolina at Chapel Hill, Chapel Hill, NC 27599-7365 (E-mail:).
The costs of publication of this article were defrayed in part by the payment of page charges. The article must therefore be hereby marked “advertisement” in accordance with 18 U.S.C. Section 1734 solely to indicate this fact.
July 17, 2002;10.1152/ajpcell.00204.2002
- Copyright © 2002 the American Physiological Society