Advertisement
JBC

HOME HELP FEEDBACK SUBSCRIPTIONS ARCHIVE SEARCH TABLE OF CONTENTS
QUICK SEARCH:   [advanced]


     

Originally published In Press as doi:10.1074/jbc.M403545200 on May 6, 2004

J. Biol. Chem., Vol. 279, Issue 27, 28149-28158, July 2, 2004
This Article
Right arrow Abstract Freely available
Right arrow Full Text (PDF)
Right arrow All Versions of this Article:
279/27/28149    most recent
M403545200v1
Right arrow Submit a Letter to Editor
Right arrow Alert me when this article is cited
Right arrow Alert me when eLetters are posted
Right arrow Alert me if a correction is posted
Right arrow Citation Map
Services
Right arrow Email this article to a friend
Right arrow Similar articles in this journal
Right arrow Similar articles in PubMed
Right arrow Alert me to new issues of the journal
Right arrow Download to citation manager
Right arrowRequest Permissions
Citing Articles
Right arrow Citing Articles via HighWire
Right arrow Citing Articles via Google Scholar
Google Scholar
Right arrow Articles by Panicker, S.
Right arrow Articles by Slesinger, P. A.
Right arrow Search for Related Content
PubMed
Right arrow PubMed Citation
Right arrow Articles by Panicker, S.
Right arrow Articles by Slesinger, P. A.
Social Bookmarking
Add to CiteULike   Add to Complore   Add to Connotea   Add to Del.icio.us   Add to Digg   Add to Reddit   Add to Technorati  
What's this?

Minimal Structural Rearrangement of the Cytoplasmic Pore during Activation of the 5-HT3A Receptor*

Sandip Panicker{ddagger}§, Hans Cruz{ddagger}, Christine Arrabit{ddagger}, Ka Fai Suen{ddagger}, and Paul A. Slesinger{ddagger}§||

From the {ddagger}The Salk Institute for Biological Studies, La Jolla, California 92037 and §Neurosciences Graduate Program, University of California, San Diego, California 92093

Received for publication, March 31, 2004 , and in revised form, May 6, 2004.


   ABSTRACT
TOP
 ABSTRACT
INTRODUCTION
 EXPERIMENTAL PROCEDURES
 RESULTS
 DISCUSSION
 REFERENCES
 
Ligand-gated ion channel receptors mediate the response of fast neurotransmitters by opening in less than a millisecond. Here, we investigated the activation mechanism of a serotonin-gated receptor (5-HT3A) by systematically introducing cysteine substitutions throughout the pore-lining M1-M2 loop and M2 transmembrane domain. We hypothesized that multiple cysteines in the narrowest region of the pore, which together can form a high affinity binding site for metal cations, would reveal changes in pore structure during gating. Using cadmium (Cd2+) as a probe, two cysteine substitutions in the cytoplasmic selectivity filter, S2'C and, to a lesser extent, G-2'C, showed high affinity inhibition with Cd2+ when applied extracellularly in the open state. Cd2+ inhibition in S2'C was attenuated if applied in the presence of an open-channel inhibitor and showed voltage-dependent recovery, indicating a direct effect of Cd2+ in the pore. When applied intracellularly, Cd2+ appeared to bind S2'C receptors in the closed state. The ability of cysteine side chains at the 2' and –2' positions to coordinate Cd2+ in both the native open and closed states of the channel suggests that the cytoplasmic selectivity filter of 5-HT3A receptors maintains a narrow pore during channel gating.


   INTRODUCTION
TOP
 ABSTRACT
 INTRODUCTION
EXPERIMENTAL PROCEDURES
 RESULTS
 DISCUSSION
 REFERENCES
 
Ligand-gated ion channel (LGIC)1 receptors are responsible for rapid chemical transmission between neurons in the nervous system (1). Alterations in the response of ligand-gated receptors have profound effects on neuronal activity in the brain. For example, mutations in nicotinic acetylcholine-gated receptors (nAChRs) and glycine-gated receptors have been linked to certain forms of epilepsy and startle disease, respectively (2–4). LGIC receptors that respond to acetylcholine, {gamma}-aminobutyric acid, glycine, or serotonin (5-HT) possess a conserved pair of extracellular cysteines in the N-terminal domain ("Cys loop" receptors) and are composed of five subunits, each containing four putative transmembrane domains (5). Of these four domains, the second transmembrane domain (M2) lines the majority of the water-filled pore (Fig. 1A).



View larger version (28K):
[in this window]
[in a new window]
  		FIG. 1.
Extracellular Cd2+ persistently inhibits S2'C and G-2'C 5-HT3A receptors expressed in Xenopus oocytes. A, alignment of the M2 from the indicated Cys-loop ligand-gated cationic (mouse 5-HT3A, mouse 5-HT3B, mouse nAChR {alpha}7, Torpedo nAChR {alpha}1) and anionic (mouse glycine receptor {alpha}1, mouse {gamma}-aminobutyric acid, type A (GABAA) {alpha}4) receptors. The X' nomenclature is shown at the bottom. The S2' and G-2' positions are boxed. The schematic above shows the approximate boundaries of the M2 domain. Stars indicate three amino acids important for cationic/anionic selectivity (18–20, 22, 23). The predicted topology of the 5-HT3A receptor is shown on the right, highlighting the M2 and the approximate location of the S2'. B and C, two-electrode voltage clamp recordings show the effect of extracellular Cd2+ (200 µM) on wild-type (WT) 5-HT3A (B) and S2'C (C) receptors in their open states (+10 µM 5-HT for 60 s). Note that the inhibition of 5-HT-induced current persisted in the absence of extracellular Cd2+ for S2'C but not for wild-type receptors (1 ± 1% inhibition, n = 8). Re-application of 5-HT plus Cd2+ did not produce additional inhibition. The holding potential was –80 mV. D, extracellular Cd2+ does not inhibit S2'C receptors in the closed state. E, bar graphs show the average Cd2+ inhibition (%) for receptors in open (n = 4–14) and closed (n = 6–11) states. There was no significant inhibition of S2'C or G-2'C receptors in the closed state. The asterisk indicates significant difference from wild-type using one-way analysis of variance with Bonferroni post hoc test (p < 0.05). Other mutants studied that showed no persistent inhibition were D-4'C, F3'C, K4'C, I5'C, T6'C, L7'C, L9'C, S12'C, V13'C, L15'C, I16'C, I17'C, and V18'C.

 
The rapid response of LGIC receptors is achieved by the energetic coupling of agonist binding in the extracellular N-terminal domain with a "gate" located 60 Å away in the receptor pore (M2 domain) (6). Several studies with LGIC receptors suggest that the gate is situated in the middle of the M2 (6–12) (but see Ref. 13). The primary determinants of selectivity appear to be located below the gate, near the cytoplasmic membrane surface. Mutations in the cytoplasmic end of the M2 alter ion selectivity (14–17), and several recent studies show that by mutating sites in this region, cation-conducting LGIC can be converted to anion-conducting channels and vice versa (18–23). Because permeability of like-charged molecules in LGIC appears to be largely dependent on size, it is believed that the narrowest region of the open LGIC pore coincides with the channel selectivity filter, which is formed by a portion of the M1-M2 loop and the cytoplasmic side of the M2 helix (16, 24). Consistent with this region being narrow, site-specific mutations in the filter alter conductivity in a manner that depends upon side-chain volume (8, 24). Thus, the pore of the open receptor can be envisaged as a funnel, with a wide extracellular vestibule that tapers to a constriction at the cytoplasmic side of the membrane, where ion selectivity is determined (25).

What is known about the extent of movement in the cytoplasmic selectivity filter? The 9-Å three-dimensional structures of open and closed Torpedo nAChRs derived from cryoelectron microscopy appear to indicate a large conformational change upon activation in the region of the cytoplasmic selectivity filter (6, 8). The more recent 4-Å closed structure places the M2 helices closer (6, 8), although no comparable high resolution structure exists for the open state. Differences in the rates of methanethiosulfonate (MTS) modification of cysteine-substituted nAChRs in their open and closed states, on the other hand, suggest that the intracellular region of the pore at the level of the selectivity filter is narrow in both the open and closed states (13).

To better understand the coupling of ligand binding to receptor activation, we sought to determine which regions move in the pore. We took advantage of the metal binding properties of multiple cysteines; a high affinity binding site for divalent metal cations is created when two or more cysteines are in close proximity (~7 Å) and oriented appropriately (26–29). Such Cd2+ binding sites have been introduced into other proteins, such as KATP channels, voltage-gated K+ channels, and glutamate transporters (30–33). Previously, we determined the structure of the M2 and the location of the gate using cysteine-substituted 5-HT3A receptors (9). Using these cysteine-substituted receptors, we now demonstrate that the selectivity filter of the 5-HT3A receptor is sufficiently narrow to form high affinity Cd2+ binding sites. We then investigated changes in pore structure using Cd2+ as a probe in cysteine-substituted 5-HT3A receptors.


   EXPERIMENTAL PROCEDURES
TOP
 ABSTRACT
 INTRODUCTION
 EXPERIMENTAL PROCEDURES
RESULTS
 DISCUSSION
 REFERENCES
 
Molecular Biology—Cysteine mutants of 5-HT3A were constructed previously (9). Four mutants (L8'C, G10'C, Y11'C, F14'C) were not tested because they yielded little or no currents when expressed alone (see Ref. 9). In vitro methyl-capped cRNA was made from linearized cDNA by T3 or T7 RNA polymerase (Stratagene). The quality of cRNA was estimated using an ethidium-stained formaldehyde gel, and the concentration was measured by UV spectrophotometry. Xenopus oocytes were isolated as described previously (34). Stage V/VI oocytes were injected with a 46-nl solution containing 0.1–10 ng of receptor cRNA. Oocytes were incubated in 96 mM NaCl, 2 mM KCl, 1 mM CaCl2, 1 mM MgCl2, and 5 mM HEPES (pH 7.6 with NaOH) for 1–7 days at 16 °C.

For studies with HEK-293T cells, 5-HT3A receptor cDNAs were subcloned in pcDNA3 (Invitrogen). HEK-293T cells were cultured in Dulbecco's modified Eagle's medium supplemented with fetal bovine serum (10%), glutamine (2 mM), penicillin (50 units/ml), and streptomycin (50 µg/ml; Invitrogen) in a humidified 37 °C incubator with 95% air, 5% CO2. For electrophysiological recordings, cells were plated onto 12-mm glass coverslips (Warner Instruments) coated with poly-D-lysine (20 µg/ml; Sigma) and collagen (100 µg/ml; BD Biosciences) in 24-well plates. HEK-293T cells were transiently transfected with cDNA using the calcium phosphate method. Briefly, cDNA was mixed in sterile deionized water with 2.5 M CaCl2 to a final concentration of 2 ng/µl cDNA and 0.25 M CaCl2. This was combined 1:1 with HEPES-buffered saline (280 mM NaCl, 10 mM KCl, 1.5 mM Na2HPO4, 12 mM glucose, 50 mM HEPES, pH 6.9, with ~1 N NaOH) and then added to each well. HEK-293T cells were cultured for 16–32 h.

Electrophysiology—Macroscopic currents were recorded from oocytes with a two-electrode voltage clamp amplifier (Geneclamp 500, Axon Instruments), filtered at 0.05 kHz, digitized (0.1 kHz) with a Digidata 1200 A/D interface (Axon Instruments), and stored on a laboratory computer. Electrodes were filled with 3 M KCl and had resistances of 0.4–2 megaohms. Oocytes were perfused continuously with a solution containing 96 mM NaCl (Aldrich), 2 mM KCl, 4.69 mM MgCl2 (free Mg2+ = 3.3 mM), 0.5 mM EGTA, and 5 mM HEPES (pH 7.6 with NaOH) while was voltage clamped at –80 mV. EGTA was omitted in solutions containing Cd2+. Programmable pinch valves (Warner Instruments) were used to change the extracellular solution flowing through a small chamber (0.32-cm width). The extracellular solution was connected to ground via a 3 M KCl-agarose bridge.

The whole-cell patch clamp technique (35) was used to record macroscopic currents from HEK-293T cells voltage-clamped at –40 mV. Borosilicate glass (Warner; P6165T) electrodes had resistances of 1–3 megaohms and were coated with Sylgard to reduce capacitance. Membrane currents were recorded with an Axopatch 200 or 200B (Axon Instruments) amplifier adjusted electronically for cell capacitance and series resistance (80–100%) before every sweep (once every 60 s), filtered at 2 kHz with an 8-pole Bessel filter, digitized at 5 kHz with a Digidata 1200 series interface (Axon Instruments), and stored on a laboratory computer. Intracellular pipette solution contained 130 mM KCl, 20 mM NaCl, 5 mM EGTA (pH 7.4 with 10 mM KOH), 2.56 mM K2ATP, 5.46 mM MgCl2 (MgATP = 2 mM), and 10 mM HEPES (pH 7.4 with ~2 mM NaOH). EGTA was omitted when intracellular Cd2+ was added, and MgCl2 was reduced to 5.2 mM to keep free MgATP = 2 mM. The external bath solution contained 158 mM NaCl, 2 mM KCl, 0.5 mM EGTA (pH 7.4 with ~2 mM KOH), 4.69 mM MgCl2, and 10 mM HEPES (pH 7.4 with ~2mM NaOH) to approximate two-electrode voltage clamp solutions but with higher NaCl outside. EGTA was omitted in extracellular Cd2+-containing solutions. The osmolarity of all solutions was 310–330 mosmol. Extracellular solutions were delivered through a perfusion manifold (Warner MM-6) that was positioned ~50–100 µm away from the cell and regulated by computer-controlled pinch valves (Warner VC-8).

Analysis—The effect of extracellular Cd2+ on 5-HT-induced currents was examined in oocytes using the following protocol: two 5–10-s pulses of 10 µM 5-HT (I1 and I2), 2–3 min of wash, 1 min of Cd2+ (200 µM) in the absence or presence of 10 µM 5-HT, 2–3 min of wash, and finally two 5–10-s pulses of 5-HT (I3 and I4) (see Fig. 1, B and C). Depending upon the mutant studied, times for 5-HT activation (5–10 s, to ensure peak activation) and wash (75–180 s, to allow for complete recovery from desensitization) varied. The percentage inhibition was calculated using the expression (1 – I4/I2) x 100.

The effect of extracellular Cd2+ administration on receptors expressed in HEK-293T cells was studied by first establishing a base-line current response from receptors by exposure to a 2-s 5-HT pulse followed by a 58-s wash (1 sweep) to allow for recovery from desensitization. Receptors were then exposed to 2-s 5-HT + Cd2+ (200 µM). This was followed by repeated 5-HT sweeps. Data were normalized by dividing the Cd2+-modified current by the initial base-line current response (Fig. 3B). For diltiazem experiments (Fig. 5), 5-HT was applied for 2 s followed by 3 s 5-HT plus diltiazem (100 µM), 2 s 5-HT plus diltiazem plus Cd2+ (200 µM), 2 s 5-HT plus diltiazem, and finally 3 s 5-HT alone (followed by wash). This Cd2+ treatment was repeated 5 times. For 1 µM Cd2+ experiments (Fig. 4), 5-HT was applied initially for 0.75 s followed by 2 s of 5-HT and Cd2+ and a 58-s wash. This was repeated until maximal inhibition was achieved. The normalized 5-HT-induced current (In/I0) was plotted as a function of cumulative exposure time to Cd2+ plus 5-HT and fit with a single exponential, y0 + a x exp(–x/{tau}), where y0 is the current remaining, a is the extent of inhibition, and {tau} is the time constant for inhibition. The bimolecular rate constant equals 1/({tau} x [Cd2+]). For voltage-dependent recovery experiments (Fig. 6), maximal Cd2+ inhibition was elicited by applying 2-s pulses of 5-HT plus Cd2+ (200 µM) in 5 successive sweeps. Recovery was induced by repeatedly alternating a 2-s 5-HT sweep at a pre-pulse voltage (–40, 0, +40, and +80 mV) and a 2-s 5-HT sweep at the test-pulse voltage (–40 mV). The time course of recovery at –40 mV was plotted as a function of 5-HT time at the pre-pulse voltage and fit with a single exponential, y0 +a x (1 – exp(–x/{tau})), where y0 is the initial extent of Cd2+ inhibition, a = 1– y0 (for complete recovery), and {tau} is the time constant for recovery. For intracellular Cd2+ experiments (Fig. 7), receptors were stimulated with 2-s 5-HT pulses delivered every 60 s (Protocol 1). For Protocol 2, a 200-s wash time was inserted between the first and second pulse of 5-HT.



View larger version (17K):
[in this window]
[in a new window]
  		FIG. 3.
S2'C and G-2'C receptors expressed in HEK-293T cells show pronounced Cd2+ inhibition. HEK-293T cells were transfected with wild-type (WT), S2'C, E-1'C, or G-2'C cDNAs. A, the schematic shows whole-cell patch clamp recording from HEK-293T cells with Cd2+ (200 µM) and 5-HT applied rapidly through a local perfusion pipette. B–D, effect of extracellular Cd2+ on 5-HT3A receptors expressed in HEK-293T cells. Extracellular Cd2+ (arrow) was co-applied during the third 5-HT pulse for wild-type (B), E-1'C (C), and S2'C (D) receptors. Note the potentiation of 5-HT-induced current for wild-type receptors. The holding potential was –40 mV. E, comparison of the Cd2+-inhibited current (5-HT plus Cd2+) and recovery from Cd2+ inhibition. The 5-HT-induced current during and after Cd2+ administration is divided by the 5-HT-induced current before Cd2+ (second pulse) and plotted as a function of 5-HT exposure time. E-1'C shows a reduction in current when 5-HT is co-applied with Cd2+ but recovers rapidly. G-2'C is also inhibited and recovers by the fifth 5-HT pulse. S2'C shows ~80% inhibition and recovers slowly in the absence of Cd2+ (n = 3–6).

 



View larger version (18K):
[in this window]
[in a new window]
  		FIG. 5.
Occlusion of the S2'C receptor pore with diltiazem retards Cd2+ inhibition. HEK-293T cells were transfected with S2'C cDNA. A, whole-cell patch clamp recording shows the effect of extracellular Cd2+ after inhibiting the 5-HT-induced current nearly 100% with 100 µM diltiazem. B, example of 5-HT-induced current in the presence of diltiazem and Cd2+ (indicated by the asterisk in A) is shown on an expanded time scale. The holding potential was –40 mV. C, using the same stimulation protocol, extracellular Cd2+ (arrow) in the absence of diltiazem inhibits S2'C receptors in the open state. D, example of 5-HT-induced current inhibited by Cd2+ alone (asterisk in C) is shown on expanded time scale. E, the 5-HT-induced current is normalized to control (Cd2+-free) and plotted as a function of the 5-HT pulse number. Inhibiting the current through S2'C with diltiazem clearly attenuates the inhibition produced by extracellular Cd2+ (n = 4–6).

 



View larger version (19K):
[in this window]
[in a new window]
  		FIG. 4.
S2'C receptors are inhibited by 1 µM extracellular Cd2+. HEK-293T cells were transfected with S2'C cDNA. A, whole-cell patch clamp recording shows the progressive reduction of 5-HT-induced current with repetitive pulses of 1 µM extracellular Cd2+ in the presence of 5-HT for 2 s. 5-HT was applied for 0.75 s before each Cd2+ application to monitor current reduction. Inset, sweep (indicated by asterisk) is shown on an expanded time scale. The scale bar is 0.5 nA and 2 s. The smooth curve shows a single-exponential fit to the open state, having a time constant {tau} = 15 s. Assuming a single-state model for inhibition, the rate of Cd2+ inhibition = 1.26 ± 0.29 x 105 M–1 s–1. The holding potential was –40 mV. B, the 5-HT-induced current (peak at 0.75 s) is normalized to the pre-modified current level (second and third pulse of 5-HT) and plotted as a function of cumulative time exposed to 5-HT. For the control, S2'C receptors were repetitively pulsed with 5-HT alone (same protocol as in A but without Cd2+), and the current is plotted as a function of the cumulative time exposed to 5-HT. For S2'C receptors, the 5-HT-induced current decreases ~20% over the time course of the experiments (~15 min of recording).

 



View larger version (24K):
[in this window]
[in a new window]
  		FIG. 6.
Recovery from Cd2+ inhibition is enhanced at positive membrane potentials in the open but not the closed state. HEK-293T cells were transfected with S2'C cDNA. A, a whole-cell patch clamp recording shows the recovery from Cd2+ inhibition with a prepulse voltage of –40 mV and a test-pulse voltage of –40 mV. Cd2+ inhibition (not shown; gray bar) was produced by 5 repetitive 2-s pulse of 5-HT (10 µM) and Cd2+ (200 µM). The holding potential was –40 mV. B, current recording shows the recovery of S2'C receptors with a prepulse voltage of +80 mV and a test pulse voltage of –40 mV. Note that the recovery of 5-HT-induced current is faster at +80 mV than at –40 mV (A). The asterisk shows the inward current through S2'C receptors that occurs upon pulsing back to –40 mV immediately after 5-HT application at +80 mV. C, recovery from Cd2+ inhibition at positive potentials in the closed state with 5-HT. S2'C receptors were opened only at the holding potential of –40 mV. The average inhibition after the third pre-pulse at +80 mV in the closed state is 88 ± 3% (n = 3) compared with 9 ± 5% in the open state (n = 6). D, the 5-HT-induced current (at –40 mV) after Cd2+ treatment was normalized to the control (first 5-HT pulse) and plotted as a function of the cumulative open time at the pre-pulse voltage. Smooth curves show exponential fits having rate constants of 0.077 ± 0.016 s–1 (n = 4) for –40 mV, 0.158 ± 0.036 s–1 (n = 5) for 0 mV, 0.303 ± 0.068 s–1 (n = 7) for +40 mV, and 0.357 ± 0.098 s–1 (n = 6) for +80 mV.

 



View larger version (20K):
[in this window]
[in a new window]
  		FIG. 7.
Intracellular Cd2+ inhibits S2'C and G-2'C receptors in the predominantly closed state. A, the schematic shows the administration of intracellular 200 µM Cd2+ via whole-cell patch pipette. B and C, whole-cell patch clamp recordings with 200 µM Cd2+ in the intracellular solution show little inhibition of wild-type (WT) 5-HT3A receptors (B) or E-1'C receptors (C) opened with 10 µM 5-HT (1.5 s). D and E, current traces show two different stimulation protocols for S2'C. In Protocol 1, 5-HT was applied once every 60 s (D). In Protocol 2, 5-HT was applied once, washed for 200 s, then followed by 5-HT pulses delivered every 60 s (E). Holding potential was –40 mV. F, the 5-HT-induced current was normalized to the first 5-HT pulse and plotted as a function of the total time of the recording for Protocol 1 () and Protocol 2 ({circ}). Inset, the inhibition of current is plotted as a function of cumulative time exposed to 5-HT for both protocols. The smooth curve shows the exponential fit to the data. Note the superimposition when plotted as a function of total time, suggesting that S2'C and G-2C' also coordinate intracellular Cd2+ in the closed state (S2'C, {tau} = ~85 s with 97 ± 1% inhibition; G-2'C, {tau} = ~90 s and 82 ± 2% inhibition; n = 4–7). E-1'C and wild-type receptors showed a 2 ± 13% (n = 7) and 11 ± 9% (n = 7) decrease in 5-HT-induced current, respectively, with 200 µM intracellular Cd2+ (not shown).

 
All values are reported as the mean ± S.E. Data were analyzed for statistical significance (SigmaStat 2.0) using one-way analysis of variance followed by Bonferroni post hoc test with wild-type as control or two-tailed Student's t test. Values of p < 0.05 were considered significant.


   RESULTS
TOP
 ABSTRACT
 INTRODUCTION
 EXPERIMENTAL PROCEDURES
 RESULTS
DISCUSSION
 REFERENCES
 
High Affinity Cd2+ Binding Sites at the –2' and 2' Positions of M2—The Cd2+ sensitivity of cysteine-substituted 5-HT3A receptors was initially examined using the Xenopus oocyte expression system. To facilitate comparison among different LGIC receptors, we have used the X' nomenclature system to refer to mutations in the M1-M2 loop and the M2 transmembrane domain (see Ref. 16). Twenty-three mutant receptors were made (D-4'C through V18'C; see Fig. 1A), of which 19 gave rise to 5-HT-induced current in oocytes (oocytes injected with cRNA encoding L8'C, G10'C, Y11'C, and F14'C gave rise to little or no 5-HT-induced current). To test the effect of Cd2+ on the open state of the receptor, we co-applied Cd2+ (200 µM) and 5-HT (10 µM) from the extracellular side of the mutant receptors. Cd2+ partially inhibited wild-type 5-HT3A receptors, but this reversed completely within ~200 s of wash (the minimal time needed for full recovery from desensitization produced by 60-s stimulation with 5-HT; see Fig. 1B). Similar to wild-type, most cysteine mutants exhibited reversible inhibition by extracellular Cd2+ in the open state (not shown). However, two cysteine-substituted receptors, G-2'C and S2'C, appeared to "coordinate" Cd2+, which we define as the mechanism by which sulfhydryl groups trap Cd2+, resulting in the persistent inhibition of 5-HT-induced currents in the absence of applied Cd2+ (Fig. 1, C and E). Although S2'C receptors are inhibited nearly 100% during the co-application of 5-HT and Cd2+, the extent of persistent inhibition is only ~80%. This observation suggests that there are two mechanisms of Cd2+ block, a rapidly reversible component that is relieved upon wash and a much slower component (coordination). There was no detectable increase in basal current after Cd2+ treatment, suggesting the receptors closed completely. A second exposure to 5-HT and Cd2+ did not result in additional inhibition of the 5-HT-induced current for S2'C receptors (Fig. 1C). The EC50 for G-2'C was 1.18 ± 0.01 µM (n = 5) before and 1.07 ± 0.03 µM (n = 4) after Cd2+ administration and for S2'C was 2.21 ± 0.03 µM (n = 10) before and 1.91 ± 0.07 µM (n = 4) after Cd2+. Thus, the smaller 5-HT-induced current could not be explained by a large shift in the EC50 for activation.

Two native cysteines exist near the pore of 5-HT3A receptors, one in the M1 (C270) and one in the M3 transmembrane domains (C316). To examine whether these cysteines contributed to Cd2+ coordination, we studied the effect of extracellular Cd2+ on S2'C receptors containing a second mutation (C270A or C316A). Although the gating kinetics of the S2'C/C316A double mutant appeared to be somewhat slower than S2'C, both S2'C/C270A (80% ± 5%; n = 6) and S2'C/C316A (88% ± 1%; n = 6) mutants were significantly inhibited after exposure to Cd2+ in the open state. We also examined the possible contribution of the negatively charged glutamate at the –1' position (Glu-1') since it has been shown that acidic amino acids can participate in coordinating Cd2+ in metalloproteins (36). Both S2'C/E-1'N (79 ± 3%; n = 6) and G-2'C/E-1'N (67 ± 1%; n = 6) appeared to bind Cd2+ equally well as S2'C and G-2'C, respectively. These results suggest that neither the receptor endogenous transmembrane cysteines nor the glutamate at the –1' site contribute to persistent Cd2+ inhibition in S2'C and G-2'C receptors.

We next investigated whether extracellularly applied Cd2+ modifies receptors in their closed state. We examined the closed state sensitivity of cysteine substitutions S-3'C through S2'C by exposing each mutant to extracellular Cd2+ in the absence of 5-HT. None of the receptors, including S2'C and G-2'C, exhibited any significant inhibition of 5-HT-induced current after the 60-s exposure to Cd2+ alone (Fig. 1, D and E). Thus, S2'C and G-2'C receptors are not readily accessible to extracellular Cd2+ in the closed state.

Cd2+ Coordination Occurs Primarily in the Open State— With prolonged agonist stimulation, LGIC receptors enter a non-conducting, desensitized state that may be structurally distinct from the open and closed states (37). Although all recordings were performed in Ca2+-free solutions to slow the rate of desensitization (38), significant desensitization remained during the 1-min stimulation of 5-HT3A receptors expressed in oocytes. To examine the possible role of the desensitized state in coordinating Cd2+, Cd2+ was applied during the first 15 s ("early") or the last 15 s ("late") of the 1 min exposure to 5-HT (Fig. 2A). Co-application of 5-HT and Cd2+ for the initial 15 s showed robust inhibition during the time of Cd2+ application. For G-2'C, some inhibition is relieved as Cd2+ washes out (as evidenced by the "hook" upon removal of Cd2+), in contrast to S2'C receptors, which do not exhibit this phenomenon. For both G-2'C and S2'C receptors, Cd2+ applied early produced the same extent of inhibition as when applied during the entire 60-s co-application of 5-HT and Cd2+ (black arrows, Fig. 2B). By contrast, Cd2+ applied after a large population of receptors had desensitized resulted in significantly less inhibition (Fig. 2, A and B). In fact, the extent of inhibition due to the late Cd2+ application can be attributed to inhibition of the remaining population of open receptors. If desensitized receptors are protected from modification and we assume that the population of receptors open at the time of Cd2+ exposure are inhibited 80% (as in the early and 1' application), we can predict the amount of total inhibition following the late paradigm (white arrows, Fig 2B). The data agree very well with our predicted values, suggesting that the engineered cysteines in G-2'C and S2'C receptors are more accessible to extracellular Cd2+ in the open state than the desensitized or closed state.



View larger version (15K):
[in this window]
[in a new window]
  		FIG. 2.
S2'C and G-2'C receptors are less accessible to extracellular Cd2+ in the desensitized state. A, two-electrode voltage clamp recordings from oocytes injected with S2'C or G-2'C cRNA. Current traces show the effect of exposing S2'C or G-2'C receptors to extracellular Cd2+ (200 µM) during the first 15 s (early) or last 15 s (late) of the 60-s application of 5-HT (10 µM). B, bar graphs show the average Cd2+ inhibition during the early and late application, as compared with the inhibition produced with the full 60-s exposure of Cd2+ (black arrow). The white arrow indicates the predicted inhibition if only open receptors are inhibited by extracellular Cd2+. The asterisk indicates statistical significance using Student's t test (p < 0.05). n = 4–5.

 
Extracellular Cd2+ Inhibition of S2'C Expressed in HEK-293T Cells—For some LGIC receptors, the electrophysiological properties of the LGIC can be different depending on whether the receptors are expressed in Xenopus oocytes or in mammalian cells (for example, see Ref. 7). We, therefore, examined the robustness of Cd2+ sensitivity by studying cysteine-substituted 5-HT3A receptors via whole-cell patch clamp recordings from transiently transfected mammalian cells (HEK-293T, Fig. 3A). Wild-type 5-HT3A receptors expressed in HEK-293T cells exhibited three notable differences from oocytes, 1) faster desensitization (even in the absence of extracellular Ca2+), 2) stronger inward rectification (cf. Refs. 39–41), and 3) a transient potentiation of the 5-HT-induced current in the presence of Cd2+ (Fig. 3, B and E). Divalent cations (Zn2+ and Cd2+) have been reported previously to potentiate (42, 43) and inhibit (44) 5-HT-induced currents through 5-HT3 receptors. Although the reasons for these differences in 5-HT3A receptor properties when expressed in oocytes and HEK-293T cells are unknown, the lack of Cd2+ inhibition of wild-type receptors in HEK-293T cells greatly simplified the subsequent analyses and interpretation of data.

Because of the faster desensitization in HEK-293T cells, we developed a different stimulation protocol to study Cd2+ inhibition (see "Experimental Procedures"). As found in oocytes, a brief application of extracellular Cd2+ (200 µM) in the open state (10 µM 5-HT) inhibited E-1'C by ~40% but showed rapid recovery (Fig. 3, C and E). G-2'C initially showed ~60% inhibition that recovered during five pulses of 5-HT alone (Fig. 3E). S2'C, on the other hand, showed ~80% inhibition during exposure to extracellular Cd2+ in the open state and recovered slowly during subsequent 5-HT pulses (Fig. 3, D and E). S2'C receptors exposed to extracellular Cd2+ in the closed state did not inhibit the 5-HT-induced current (not shown), similar to oocytes. Thus, S2'C, E-1'C, and G-2'C receptors expressed in HEK-293T cells retained the same rank order sensitivity to extracellular Cd2+ as observed in oocytes (S2'C > G-2'C > E-1'C). S2'C was the only mutant to exhibit appreciable long-lasting inhibition (>5 min) in both expression systems. Based on this finding and the observation that Cd2+ showed little modulation of wild-type 5-HT3A receptors expressed in HEK-293T cells, we investigated in more detail the mechanism underlying Cd2+ inhibition of S2'C receptors expressed in HEK-293T cells.

To determine the sensitivity of the S2'C receptor to Cd2+, we studied the effect of 1 µM Cd2+ in the open state of the S2'C receptor (a dose 200 times lower than initially used to test all cysteine mutants). After a single pulse of 5-HT plus Cd2+, S2'C receptors were inhibited ~20%, in contrast to 80% inhibition seen with 200 µM Cd2+ (Fig. 3E). Therefore, to determine the maximal extent of inhibition achievable with 1 µM Cd2+, we used a repetitive pulsing protocol, in which channels were first opened with 5-HT (0.75 s) and then exposed to 1 µM Cd2+ in the continued presence of 5-HT for another 2 s. This stimulation paradigm was repeated once every minute until inhibition reached a steady state level (Fig. 4A). In this manner the decrease in 5-HT-induced current due to Cd2+ inhibition was monitored over time. We found that upon repetitive stimulation, 1 µM Cd2+ inhibited 5-HT-induced current to the same extent as 200 µM Cd2+, albeit much more slowly. Testing lower Cd2+ concentrations proved to be difficult since, over the time course of these experiments, we found that 5-HT-induced current decreased upon repeated 5-HT stimulation alone (see Fig. 4B, control). One possible explanation for this decrease could be disulfide bond formation between the engineered cysteines, although this is unlikely since we observed a similar trend with wild-type receptors (Fig. 3E), and DTT did not recover the current in S2'C receptors. We speculate that this decrease in current is due to the accumulation of receptors in the desensitized state over time. Nonetheless, the ability of S2'C to coordinate Cd2+ at low concentrations is consistent with the interpretation that the introduced cysteines form a high affinity metal cation binding site in the receptor pore.

The progressive decrease in 5-HT-induced current at 1 µM Cd2+ is well fit with a monoexponential function (Fig. 4A). If we assume a single-site model for inhibition (which may be an oversimplification of the inhibition, see "Discussion"), then the time constant can be used to estimate the forward rate of Cd2+ inhibition (k1) at S2'C. Adjusting for the decrease in current due to 5-HT alone, in five patches k1 = 1.26 x 105± 0.29 x 105 M–1 s–1, which is slower than diffusion (~109 M–1 s–1; see Ref. 5) but faster than MTS modification (9).

Cd2+ Coordination Occurs within the S2'C Pore—To ensure that Cd2+ was coordinated by the introduced cysteines in the pore of the S2'C receptor, we applied Cd2+ in the presence of a 5-HT3A open channel inhibitor. If Cd2+ binds to cysteines located within the cytoplasmic selectivity filter, then occluding the S2'C receptor pore before applying Cd2+ would be expected to attenuate Cd2+ inhibition. Diltiazem is a calcium channel inhibitor that also acts as an open-channel inhibitor of 5-HT3A receptors (45). Similar to its inhibition of wild-type 5-HT3A receptors, 100 µM diltiazem inhibited nearly 100% of the 5-HT-induced current in the S2'C receptor (Fig. 5, A and B). To examine the effect of diltiazem on Cd2+ inhibition, S2'C was opened first with 5-HT, exposed to 5-HT plus diltiazem (100 µM), and then exposed to Cd2+ (200 µM) in the continued presence of 5-HT plus diltiazem (Fig. 5, A and B). This protocol was repeated five times followed by a 5-HT pulse to determine the extent of Cd2+ inhibition (see Fig. 5, A and C). In the presence of diltiazem, the 5-HT-induced current decreased by only ~20% after treatment with Cd2+ compared with nearly 60% inhibition observed in the absence of the open-channel inhibitor (Fig. 5, C–E). Thus, occluding the pore of S2'C with diltiazem attenuates Cd2+ inhibition, most likely by hindering Cd2+ access to the 2' cysteine in the selectivity filter.

Voltage Dependence of Cd2+ Inhibition—Although it is unknown precisely where the voltage drop occurs along the M2 domain of the 5-HT3A receptor, Cd2+ situated in the pore is likely to experience some of the voltage drop across the membrane. Thus, if Cd2+ binds to S2'C within the receptor conduction pathway, we would expect Cd2+ modification of sites deep within the pore to be voltage-dependent (5). We, therefore, applied Cd2+ to S2'C receptors at different membrane potentials in the open state to study the voltage dependence of inhibition. Surprisingly, S2'C receptors exhibited weak voltage dependence for the onset of Cd2+ inhibition. Receptors appeared to be modified equally well whether Cd2+ was applied at –40 mV (89 ± 1% inhibition 1 min after Cd2+ administration; n = 4), 0 mV (88 ± 1%; n = 5), or +40 mV (86 ± 3%; n = 4). When applied at +100 mV, however, Cd2+ inhibited only 60 ± 2% (n = 4) of the pre-modified 5-HT-induced current (p < 0.05). Unfortunately, examining voltage dependence at more positive membrane potentials was not possible because the cell could not tolerate such large depolarizations.

Once Cd2+ is bound within the pore, S2'C receptors recover slowly from inhibition at –40 mV (Fig. 3E). We reasoned that if Cd2+ senses the voltage drop across the pore, then recovery from Cd2+ inhibition should be dependent on the absolute membrane potential (46). To examine this possibility, maximal Cd2+ inhibition was first induced at –40 mV (Figs. 6, A–C, gray bar; see "Experimental Procedures" for details). Recovery was studied by opening Cd2+-modified S2'C receptors at a pre-pulse voltage (–40, 0, +40, or +80 mV) and then measuring the amplitude of the 5-HT-induced current at –40 mV (test pulse) after each pre-pulse (Figs. 6, A, B, and D). We selected a test pulse of –40 mV to reduce potential contamination by voltage-gated conductances endogenous to the HEK-293T cells. Plotting the amplitude of 5-HT-induced current at –40 mV reveals the rate of recovery as a function of the pre-pulse potential (Fig. 6D). Clearly, opening Cd2+-modified receptors at more positive voltages speeds the rate of recovery from Cd2+ inhibition. Note that the recovery at –40 mV was slower than that at 0 or +40 mV (see "Discussion").

To determine whether recovery occurred in the closed state, we depolarized Cd2+-modified receptors to +80 mV in the absence of 5-HT followed immediately by a test pulse of 5-HT at –40 mV (Fig. 6C). Significantly, positive membrane potentials did not accelerate recovery from Cd2+ inhibition in the absence of 5-HT. Thus, voltage-dependent recovery occurred only when receptors were opened with 5-HT. The observation that the rate of recovery from Cd2+ inhibition depends on the pre-pulse voltage supports the conclusion that Cd2+ binds to the S2'C in the cytoplasmic selectivity filter, where it is driven out by voltage and current flow through the receptor pore.

Intracellular Application of Cd2+ to S2'C and G-2'C—Inhibition of 5-HT-induced current after application of extracellular Cd2+ in the open state suggests that sulfhydryl groups at the 2' position can coordinate Cd2+ in the presence of 5-HT and, thus, attain a distance suitable for coordinating Cd2+ (<7 Å) in the open state. Cd2+ applied in the closed state (in the absence of 5-HT), however, was ineffective at inhibiting S2'C receptors when applied from the extracellular side of the membrane (Figs. 1, D and E). There are two possible explanations for this outcome. First, the sulfhydryl side chains may not be in the proper orientation in the closed state to coordinate Cd2+ (either too far apart or face away from the pore). Alternatively, the gate for the 5-HT3A receptor could be situated on the extracellular side of the 2' site and, thus, prevents Cd2+ from accessing the cytoplasmic region of the pore. To distinguish between these two possibilities we investigated the effect of Cd2+ applied to the intracellular side of the closed receptor.

Having established that S2'C, G-2'C, and E-1'C receptors retained the same rank order sensitivity to extracellular Cd2+ when expressed in HEK-293T cells, we investigated the effect of intracellularly applied Cd2+ on the wild-type and these three mutant 5-HT3A receptors. Upon establishing a whole-cell configuration, the pipette solution, which contained Cd2+ (200 µM), began to exchange for the cell cytoplasmic solution (Fig. 7A). Therefore, to establish a time point at which most of the receptors were not exposed to Cd2+, 5-HT-induced responses were recorded immediately after establishing a whole-cell configuration. Both wild-type and E-1'C receptors did not show appreciable inhibition of the 5-HT-induced current with intracellular Cd2+ (Figs. 7, B and C). Using these recording conditions, however, it was not possible to study the inhibition with Cd2+ exclusively in the open state (because Cd2+ is present throughout the duration of the recording) or exclusively in the closed state (since receptors must be opened to record current). To discern between open and closed state modification, we therefore examined Cd2+ inhibition using two different stimulation protocols. In Protocol 1, cells were exposed to 5-HT (1.5 s) every 60 s until achieving steady state inhibition (Fig. 7D). In Protocol 2, we stimulated cells once with 5-HT, washed for ~200 s, and then stimulated with 5-HT every 60 s until steady state inhibition was attained (Fig. 7E). We reasoned that if the receptor coordinates Cd2+ in the closed state, then the rate of inhibition will depend on the total time that Cd2+ is exposed to the receptor. On the other hand, if Cd2+ coordination occurs only in the open state, then the rate of inhibition will depend on the cumulative time that the receptor was open (exposed to 5-HT).

When plotted as a function of cumulative time exposed to 5-HT, the time courses for the decrease in 5-HT-induced currents for Protocol 1 (Fig. 7F, closed circles) and Protocol 2 (open circles) do not overlap (Fig. 7F, inset). By contrast, the time course of inhibition for the two protocols is indistinguishable when 5-HT-induced currents are plotted as a function of the total time exposed to Cd2+ (Fig. 7F). Similar results were obtained with G-2'C (see the legend to Fig. 7). With both mutants, the 5-HT-induced current decreased along a single exponential time course with a time constant of ~85 s. The rate of Cd2+ entry as well as the concentration of Cd2+ at the pore is not known in these experiments; therefore, we cannot provide an estimate of the bimolecular rate of Cd2+ modification. However, for comparison, (2-aminoethyl)methanethiosulfonate, which is much larger than Cd2+, has a time constant of modification of ~400 s when applied intracellularly to the G-2'C receptor of the nAChR in the closed state (13). The ability of intracellularly applied Cd2+ to inhibit 5-HT-induced current in the absence of 5-HT provides evidence that S2'C and G-2'C can coordinate Cd2+ in their native closed states.


   DISCUSSION
TOP
 ABSTRACT
 INTRODUCTION
 EXPERIMENTAL PROCEDURES
 RESULTS
 DISCUSSION
REFERENCES
 
In this study we investigated the conformational changes that underlie opening of ligand-gated ion channels by taking advantage of the physiochemical properties of Cd2+ and cysteines engineered into the 5-HT3A receptor. By introducing a high affinity metal divalent binding site in a narrow region of the pore of the 5-HT3A receptor, we showed that both native open and closed states of the receptor can coordinate Cd2+. Based on these results, we propose a model for receptor activation in which the cytoplasmic pore remains narrow during receptor gating (Fig. 8).



View larger version (25K):
[in this window]
[in a new window]
  		FIG. 8.
Working model for gating of 5-HT3A receptors. 5-HT3A receptor is a homopentameric receptor with the pore lined predominantly by the M2 domain (only two of the five subunits are shown for clarity). The N-terminal and M3-M4 loop (dashed line) are not drawn to scale. The ligand binding domain (not shown) communicates to the M2 via the M2-M3 loop (see "Discussion" for details). The gate is hypothesized to be located in the middle to extracellular (~9' to 13') half of the M2 domain (shaded region). The narrowest region of the closed receptor lumen is formed by the cytoplasmic segments of M2 (~–2' to 2'). Upon activation, the extracellular half of the M2 exhibits greater movement (arrow), exposing amino acids previously inaccessible in the closed state, whereas the cytoplasmic portion of M2 moves little. Positions of transmembrane domains are based loosely on the nAChR structure (6). In this model, Cd2+ has access to S2'C receptors in the open state from the extracellular side and to S2'C receptors in the closed and open states from the intracellular side of the membrane.

 
Evidence That S2'C Coordinates Cd2+ in the Pore of the 5-HT3A Receptor—Of the 19 cysteine substitutions studied in the 5-HT3A receptor, extracellular application of Cd2+ in the open state led to the persistent inhibition of only two cysteine-substituted receptors, G-2'C and S2'C. The engineered cysteines are located in the M1-M2 loop (G-2'C) and on the cytoplasmic side of the M2 helix (S2'C), a region known to be important for permeability and ion selectivity (15–23). Because of the magnitude and persistence of inhibition in both oocyte and HEK-293T cell expression systems, we investigated in more detail the Cd2+ inhibition of the S2'C receptor. Fortunately, the serine to cysteine mutation does not dramatically alter the side-chain polarity, length, or volume at this site in the lumen. Thus, the S2'C receptor probably retains the native wild-type cytoplasmic pore structure.

How does Cd2+ coordination lead to an inhibition of 5-HT-induced current in the S2'C receptor? Divalent cations, such as Mg2+ and Ca2+, have been shown to accelerate the 5-HT-induced desensitization of wild-type 5-HT3A receptors (38). Divalent cations have also been shown to modulate LGIC current by interacting with binding sites on the extracellular side of the receptor, such as the N-terminal domain or M2-M3 loop. Specifically, Zn2+ has been shown to transiently modify LGIC current in a voltage-independent manner (47–49). Therefore, one possibility is that Cd2+ interacts with sites on the extracellular side of the receptor, where it inhibits current flow either by modulating gating (drawing the receptor into a desensitized state) or by affecting 5-HT binding at the ligand binding domain. However, recovery from Cd2+ inhibition of S2'C receptors occurs only in the presence of 5-HT (Fig. 6), indicating that Cd2+-bound channels must open and are, in fact, sensitive to 5-HT. Furthermore, recovery from Cd2+ inhibition is voltage-dependent, whereas recovery from desensitization should be independent of voltage. In addition, voltage-dependent recovery also shows that Cd2+ senses the voltage drop across the receptor pore. This, taken together with the finding that inhibition is attenuated when Cd2+ is applied in the presence of the open channel blocker diltiazem (Fig. 5), indicates that the site of Cd2+ inhibition is within the receptor pore.

Once in the pore, Cd2+ could induce disulfide bond formation between the engineered cysteines, similar to the effect of oxidation with copper phenanthroline. A similar oxidizing mechanism was proposed to occur with MTS reagent exposure to cysteine-substituted {gamma}-aminobutyric acid, type A receptors (7). This scenario, however, is inconsistent with the observation that the rate of recovery from Cd2+ inhibition was faster at more positive voltages. In addition, we were unable to inhibit 5-HT-induced current using the oxidizing agent copper phenanthroline, suggesting that the cysteine side chains are not in the proper orientation for disulfide bond formation. Thus, we propose that the inhibition of the S2'C receptor as well as the G-2'C receptor is caused by Cd2+ binding to a high affinity binding site created by the introduced cysteines, where Cd2+ sterically and/or electrostatically inhibits current flow.

With Cd2+ bound to the cytoplasmic pore, where it can experience the voltage drop across the membrane, we expected the rate of recovery to depend on the membrane potential (5). For S2'C receptors, the recovery from Cd2+ inhibition was slower at –40 mV than it was at +40 mV. This result suggests that the ability of Cd2+ to exit into the cytoplasm is less energetically favorable than exiting extracellularly (5). One possible explanation for a higher cytoplasmic energy barrier is that the structure underneath the M2 pore, which is formed by the M3-M4 loop, impedes Cd2+ exiting into the cytoplasm. Consistent with this, the rate of inhibiting cysteine substitutions located in the 2' region is slow with intracellularly applied Cd2+ (this study) or MTS reagents (13). Furthermore, crystallographic studies on the nAChR revealed the presence of small fenestrations in the cytoplasmic walls of the protein, which could regulate ion flow into and out of the conduction pathway (50). Although this suggests that something impedes Cd2+ exiting into the cytoplasm, it is unlikely that an activation gate is located on the cytoplasmic side of the S2' site because S2'C binds intracellularly applied Cd2+ in the closed state (cf. Ref. 13).

Mechanism of Cd2+ Inhibition—During the co-application of extracellular Cd2+ and 5-HT, Cd2+ abolished nearly 100% of the current flow of S2'C receptors in both oocytes (Fig. 1C) and HEK-293T cells (see Fig. 5D for expanded time scale). After Cd2+ exposure, however, the 5-HT-induced current was larger than that during the co-application of Cd2+ and 5-HT for S2'C expressed in both oocytes and HEK-293T cells. This suggests that during the co-application of Cd2+ and 5-HT, S2'C exhibits two forms of Cd2+ inhibition, a slowly reversible block (extremely slow, if at all, in oocytes) and a rapidly reversible block that is relieved upon wash. The rapidly reversible block may be due to Cd2+ acting as a reversible open channel blocker, similar to MTS reagents (9), or allosterically altering the response to 5-HT on the extracellular side of the receptor. The slowly reversible block, however, is most likely due to Cd2+ coordination at the site of the 2' cysteines. Although the difference in the rate of recovery from Cd2+ coordination is not clear, it is clear that S2'C receptors are potently inhibited with Cd2+ in both expression systems.

Crystallographic and mutagenesis studies on metalloproteins and channels reveal some of the details of the coordination of metal cations by sulfhydryls (26, 28, 31, 32). Cd2+ is a soft metal cation (Lewis acid) that preferentially interacts with soft bases such as deprotonated sulfhydryl groups (S–) (29). Cd2+ donates electrons from its {pi} orbital to the d or {pi} orbital of S–, thereby reducing the effective charge of Cd2+ and forming a covalent Cd-S bond (29). The coordination number for Cd2+ with thiol groups is 4.6 (29), although sometimes 2 cysteine or histidine combinations are sufficient to coordinate Cd2+ (26–28, 30, 32).

In the pentameric 5-HT3A receptor, Cd2+ coordination could be achieved with several different arrangements of cysteine side chains in the pore. A single Cd2+ in the pore could be coordinated by four-five cysteines, similar to that found in some metalloproteins and channels (26, 30, 32, 51). Alternatively, Cd2+ could be coordinated along the perimeter of the pore, one between each subunit, as was recently suggested for a cysteine-substituted hyperpolarization-activated cation channel (52). Finally, pairs of sulfhydryl groups on adjacent subunits may coordinate Cd2+, as was suggested in cysteine-substituted homomeric glutamate A receptors (53). In the present study we were unable to distinguish between the many possible coordination geometries in the pore. However, the ability of S2'C receptors to coordinate Cd2+ in both the native open and closed configurations of the receptor places a restriction on the maximal diameter the pore attains in both states. Using the length of the Cd-S bond (2.1–2.5 Å; (28, 29, 36, 54)), the ionic radius of Cd2+ (0.91 Å), and assuming 5-fold pseudosymmetry in the 5-HT3A homopentameric receptor, we can estimate a maximum pore diameter of ~7 Å if coordination occurs across the receptor pore. If Cd2+ is coordinated by sulfhydryl moieties on adjacent subunits, the pore diameter can be as wide as ~11 Å (using the Pythagorean theorem). These distances are in good agreement with the lower pore size limit of 7.6 Å for native 5-HT3 receptors (55) as well as 6.5 Å for the nAChR (5). Thus, the ability of S2'C sulfhydryl groups to coordinate Cd2+ is consistent with the known geometry of the open LGIC pore.

The finding that S2'C and G-2'C receptors, but not E-1'C receptors, coordinated Cd2+ applied from both the intra and extracellular sides of the pore suggests that 1) the mechanism of Cd2+ inhibition is the same regardless of which side of the membrane Cd2+ is applied, and 2) the side chains at the –2' and 2' sites can coordinate Cd2+ in the native closed and open states of the receptor. Although we cannot be certain that Cd2+-bound S2'C receptors conduct current, modified receptors retain the ability to open in response to 5-HT (Fig. 6), suggesting that Cd2+ does not interfere with the receptor's gating machinery. Because sulfhydryl groups must remain within 7 Å to coordinate Cd2+, these findings suggest that the M2 helices and part of the M1-M2 loop move little relative to one another during receptor gating.

Mutating E-1' in different types of LGIC produces changes in ion selectivity, inward rectification, and single-channel conductance (15, 19, 20, 23, 56). E-1' likely faces the lumen of the pore, similar to S2' and G-2', and can be modified by MTS reagents (9, 13, 57). Thus, we were surprised to find that the E-1'C receptor was not inhibited with either intracellularly or extracellularly applied Cd2+ with the same potency as S2'C or G-2'C receptors (Figs. 3 and 7). One possible explanation is the Cd2+ binding affinity of E-1'C is reduced because of a less optimal arrangement of cysteines in the pore. In the 4-Å structure of the closed nAChR, the E-1' is positioned at the apex of the M1-M2 loop with its side chains facing the intracellular vestibule, along the pore axis (6). By contrast, the S2' side chains are located at the cytoplasmic end of the M2 helix, facing each other and forming the narrowest region of the pore (6). This arrangement could lead to more potent Cd2+ inhibition at the S2' site. The ability of G-2'C to coordinate Cd2+ more effectively than E-1'C receptor, however, is not easily explained by the closed-state structure of the nAChR. Perhaps increasing the side-chain length, by substituting cysteine for glycine at the –2' position, brings the sulfhydryl groups closer to the central pore axis where they can coordinate Cd2+. A comparison of high resolution crystallographic structures of open and closed receptors is needed to further elucidate the precise position of side chains for amino acids in the selectivity filter.

Proposed Model for LGIC Gating; Minimal Structural Rearrangement in Selectivity Filter—The conformational changes that occur during LGIC activation are initiated by neurotransmitter binding to the ligand binding site of the N-terminal domain. This binding is thought to induce a rotation in the N-terminal domain that produces a conformational change in the M2 transmembrane domain, conveyed through the M2-M3 loop (58–60). The large conformational change within the pore presumably occurs at or near the receptor gate during activation (Fig. 8). Although the precise location of the gate has been equivocal (8, 13), the 4-Å density map of the nAChR (6) and recent MTS studies with glycine, {gamma}-aminobutyric acid, type A, and 5-HT3 receptors implicate movement between the 6' and 13' positions (7, 9, 10, 57). In support of this, linear free energy relationship analysis of mutant nAChRs suggested that the cytoplasmic side of M2 moves last upon activation after a wave of energy moving from the site of ligand binding down toward the cytoplasmic selectivity filter (61, 62). Although linear free energy relationship analysis does not reveal information concerning the extent of movement, these studies suggest that for residues below the 11' site, the structure of the transition state (the highest energy state the receptor attains during gating) closely resembles that of the closed state. In addition, the M2 helices are close enough in the native open and closed states to allow for Cd2+ coordination at the 2' site. Based on these studies, we propose that structural rearrangements are minimal at the 2' site while the receptor moves between open and closed states. A narrow pore is in accordance with the openstate structure of the cytoplasmic pore of the nAChR inferred by cysteine substitution studies of the M2 domain (13). In these studies the ability to modify cysteines between positions E-1' and T2' with MTSET, a compound much larger than Cd2+ (5.9-Å head group versus 1.82 Å), was slow in both the open and closed states, leading the authors to conclude that this region of the receptor pore is narrow in both states.

Based on initial structural studies of nAChRs, however, Unwin (8) proposed that the pore widens in the middle of M2, and the cytoplasmic selectivity filter rotates during activation. Thus, the outwardly splayed intracellular segments of M2 move closer together during receptor activation (8). This type of motion, however, is difficult to reconcile with the Cd2+ inhibition observed for S2'C 5-HT3A receptors as well as the MTS studies of nAChRs (13). Recently, Miyazawa et al. (6) describe a higher resolution structure of the closed nAChR, which places the side chains of the 2' amino acids closer than originally proposed. Although a significant conformational change is still postulated upon receptor activation, the position of the 2' site in the closed nAChR is remarkably compatible with the ability of S2'C 5-HT3A receptor to coordinate Cd2+. These findings suggest that the selectivity filter of the Cys-loop family of LGIC may undergo minimal structural rearrangement during gating.


   FOOTNOTES
 
* This work was supported by a National Institutes of Health training grant (to S. P.), the Legler Benbough Foundation (to S. P.), the Sloan Foundation (to P. A. S.), the McKnight Endowment for Neuroscience (to P. A. S.), and the Fritz-Burns Foundation (to P. A. S.). The costs of publication of this article were defrayed in part by the payment of page charges. This article must therefore be hereby marked "advertisement" in accordance with 18 U.S.C. Section 1734 solely to indicate this fact. Back

Current address: Dept. of Basic Neurosciences, University of Geneva, Switzerland, CH-1211. Back

|| To whom correspondence should be addressed: Peptide Biology Laboratory, The Salk Institute, 10010 N. Torrey Pines Rd. La Jolla, CA 92037. Tel.: 858-453-4100 (ext. 1560); Fax: 858-552-1546; E-mail: slesinger{at}salk.edu.

1 The abbreviations used are: LGIC, ligand-gated ion channel; nAChR, nicotinic acetylcholine-gated receptor; 5-HT, serotonin; MTS, methanethiosulfonate; HEK cells, human embryonic kidney cells. Back


   ACKNOWLEDGMENTS
 
We thank D. Julius for the 5-HT3A cDNA, S. Heinemann and E. Isacoff for helpful comments, and members of the Slesinger laboratory for comments on the manuscript.



   REFERENCES
TOP
 ABSTRACT
 INTRODUCTION
 EXPERIMENTAL PROCEDURES
 RESULTS
 DISCUSSION
 REFERENCES
 

  1. Kandel, E. R., Schwartz, J. H., and Jessell, T. M. (2000) Principles of Neural Science, 4th Ed., McGraw-Hill Health Professions Division, New York
  2. Kuryatov, A., Gerzanich, V., Nelson, M., Olale, F., and Lindstrom, J. (1997) J. Neurosci. 17, 9035–9047[Abstract/Free Full Text]
  3. Phillips, H. A., Favre, I., Kirkpatrick, M., Zuberi, S. M., Goudie, D., Heron, S. E., Scheffer, I. E., Sutherland, G. R., Berkovic, S. F., Bertrand, D., and Mulley, J. C. (2001) Am. J. Hum. Genet. 68, 225–231[CrossRef][Medline] [Order article via Infotrieve]
  4. Shiang, R., Ryan, S. G., Zhu, Y. Z., Hahn, A. F., O'Connell, P., and Wasmuth, J. J. (1993) Nat. Genet. 5, 351–358[CrossRef][Medline] [Order article via Infotrieve]
  5. Hille, B. (2001) Ion Channels of Excitable Membranes, pp. 170–171, Sinauer Associates, Inc., Sunderland, MA
  6. Miyazawa, A., Fujiyoshi, Y., and Unwin, N. (2003) Nature 423, 949–955[CrossRef][Medline] [Order article via Infotrieve]
  7. Shan, Q., Haddrill, J. L., and Lynch, J. W. (2002) J. Biol. Chem. 277, 44845–44853[Abstract/Free Full Text]
  8. Unwin, N. (1995) Nature 373, 37–43[CrossRef][Medline] [Order article via Infotrieve]
  9. Panicker, S., Cruz, H., Arrabit, C., and Slesinger, P. A. (2002) J. Neurosci. 22, 1629–1639[Abstract/Free Full Text]
  10. Horenstein, J., Wagner, D. A., Czajkowski, C., and Akabas, M. H. (2001) Nat. Neurosci. 4, 477–485[Medline] [Order article via Infotrieve]
  11. Filatov, G., and White, M. (1995) Mol. Pharmacol. 48, 379–384[Abstract]
  12. Labarca, C., Nowak, M. W., Zhang, H., Tang, L., Deshpande, P., and Lester, H. A. (1995) Nature 376, 514–516[CrossRef][Medline] [Order article via Infotrieve]
  13. Wilson, G. G., and Karlin, A. (1998) Neuron 20, 1269–1281[CrossRef][Medline] [Order article via Infotrieve]
  14. Bertrand, D., Galzi, J. L., Devillers-Thiéry, A., Bertrand, S., and Changeux, J. P. (1993) Proc. Natl. Acad. Sci. U. S. A. 90, 6971–6975[Abstract/Free Full Text]
  15. Imoto, K., Busch, C., Sakmann, B., Mishina, M., Konno, T., Nakai, J., Bujo, H., Mori, Y., Fukuda, K., and Numa, S. (1988) Nature 335, 645–648[CrossRef][Medline] [Order article via Infotrieve]
  16. Cohen, B. N., Labarca, C., Davidson, N., and Lester, H. A. (1992) J. Gen. Physiol. 100, 373–400[Abstract/Free Full Text]
  17. Villarroel, A., and Sakmann, B. (1992) Biophys. J. 62, 162–208
  18. Wotring, V. E., Miller, T. S., and Weiss, D. S. (2003) J. Physiol. (Lond.) 548, 527–540[Abstract/Free Full Text]
  19. Thompson, A. J., and Lummis, S. C. R. (2003) Br. J. Pharmacol. 140, 359–365[CrossRef][Medline] [Order article via Infotrieve]
  20. Corringer, P.-J., Bertrand, S., Galzi, J.-L., Devillers-Thiéry, A., Changeux, J.-P., and Bertrand, D. (1999) Neuron 22, 831–843[CrossRef][Medline] [Order article via Infotrieve]
  21. Jensen, M. L., Timmermann, D. B., Johansen, T. H., Schousboe, A., Varming, T., and Ahring, P. K. (2002) J. Biol. Chem. 277, 41438–41447[Abstract/Free Full Text]
  22. Gunthorpe, M. J., and Lummis, S. C. (2001) J. Biol. Chem. 276, 10977–10983[Abstract/Free Full Text]
  23. Keramidas, A., Moorhouse, A. J., French, C. R., Schofield, P. R., and Barry, P. H. (2000) Biophys. J. 79, 247–259[Medline] [Order article via Infotrieve]
  24. Lester, H. A. (1992) Annu. Rev. Biophys. Biomol. Struct. 21, 267–292[CrossRef][Medline] [Order article via Infotrieve]
  25. Unwin, N. (2000) Philos. Trans. R. Soc. Lond. B. Biol. Sci. 355, 1813–1829[Abstract/Free Full Text]
  26. Ueyama, N., Sugawara, T., Sasaki, K., Nakamura, A., Yamashita, S., Wakatsuki, Y., Yamazaki, H., and Yasuoka, N. (1988) Inorg. Chem. 27, 741–747
  27. Arnold, F. H., and Haymore, B. L. (1991) Science 252, 1796–1797[Free Full Text]
  28. Rauser, W. E. (1999) Cell Biochem. Biophys. 31, 19–48[CrossRef][Medline] [Order article via Infotrieve]
  29. Glusker, J. P. (1991) Adv. Protein Chem. 42, 1–76[Medline] [Order article via Infotrieve]
  30. Holmgren, M., Shin, K. S., and Yellen, G. (1998) Neuron 21, 617–621[CrossRef][Medline] [Order article via Infotrieve]
  31. Krovetz, H. S., VanDongen, H. M. A., and VanDongen, A. M. J. (1997) Biophys. J. 72, 117–126[Medline] [Order article via Infotrieve]
  32. Loussouarn, G., Makhina, E. N., Rose, T., and Nichols, C. G. (2000) J. Biol. Chem. 275, 1137–1144[Abstract/Free Full Text]
  33. Brocke, L., Bendahan, A., Grunewald, M., and Kanner, B. I. (2002) J. Biol. Chem. 277, 3985–3992[Abstract/Free Full Text]
  34. Slesinger, P. A., Patil, N., Liao, Y. J., Jan, Y. N., Jan, L. Y., and Cox, D. R. (1996) Neuron 16, 321–331[CrossRef][Medline] [Order article via Infotrieve]
  35. Hamill, O. P., Marty, A., Neher, E., Sakmann, B., and Sigworth, F. J. (1981) Pflügers Arch. 391, 85–100[CrossRef][Medline] [Order article via Infotrieve]
  36. Benning, M. M., Shim, H., Raushel, F. M., and Holden, H. M. (2001) Biochemistry 40, 2712–2722[CrossRef][Medline] [Order article via Infotrieve]
  37. Wilson, G., and Karlin, A. (2001) Proc. Natl. Acad. Sci. U. S. A. 98, 1241–1248[Abstract/Free Full Text]
  38. Yakel, J. L., Lagrutta, A., Adelman, J. P., and North, R. A. (1993) Proc. Natl. Acad. Sci. U. S. A. 90, 5030–5033[Abstract/Free Full Text]
  39. Maricq, A. V., Peterson, A. S., Brake, A. J., Myers, R. M., and Julius, D. (1991) Science 254, 432–437[Abstract/Free Full Text]
  40. Brown, A. M., Hope, A. G., Lambert, J. J., and Peters, J. A. (1998) J. Physiol. (Lond.) 507, 653–665[Abstract/Free Full Text]
  41. Gunthorpe, M., Peters, J., Gill, C., Lambert, J., and Lummis, S. (2000) J. Physiol. (Lond.) 522, 187–198[Abstract/Free Full Text]
  42. Uki, M., and Narahashi, T. (1996) Arch. Toxicol. 70, 652–660[CrossRef][Medline] [Order article via Infotrieve]
  43. Hubbard, P. C., and Lummis, S. C. (2000) Eur. J. Pharmacol. 394, 189–197[CrossRef][Medline] [Order article via Infotrieve]
  44. Lovinger, D. M. (1991) J. Neurophysiol. 66, 1329–1337[Abstract/Free Full Text]
  45. Gunthorpe, M. J., and Lummis, S. C. (1999) J. Physiol. (Lond.) 519, 713–722[Abstract/Free Full Text]
  46. Woodhull, A. M. (1973) J. Gen. Physiol. 61, 687–708[Abstract/Free Full Text]
  47. Laube, B., Kuhse, J., and Betz, H. (2000) J. Physiol. (Lond.) 522, 215–230[Abstract/Free Full Text]
  48. Hsiao, B., Dweck, D., and Luetje, C. W. (2001) J. Neurosci. 21, 1848–1856[Abstract/Free Full Text]
  49. Palma, E., Maggi, L., Miledi, R., and Eusebi, F. (1998) Proc. Natl. Acad. Sci. U. S. A. 95, 10246–10250[Abstract/Free Full Text]
  50. Miyazawa, A., Fujiyoshi, Y., Stowell, M., and Unwin, N. (1999) J. Mol. Biol. 288, 765–786[CrossRef][Medline] [Order article via Infotrieve]
  51. Dudev, T., and Lim, C. (2002) J. Am. Chem. Soc. 124, 6759–6766[CrossRef][Medline] [Order article via Infotrieve]
  52. Rothberg, B. S., Shin, K. S., and Yellen, G. (2003) J. Gen. Physiol. 122, 501–510[Abstract/Free Full Text]
  53. Sobolevsky, A. I., Yelshansky, M. V., and Wollmuth, L. P. (2004) Neuron 41, 367–378[CrossRef][Medline] [Order article via Infotrieve]
  54. Zhu, Z., DeRose, E. F., Mullen, G. P., Petering, D. H., and Shaw, C. F., III (1994) Biochemistry 33, 8858–8865[CrossRef][Medline] [Order article via Infotrieve]
  55. Yang, J. (1990) J. Gen. Physiol. 96, 1177–1198[Abstract/Free Full Text]
  56. Forster, I., and Bertrand, D. (1995) Proc. R. Soc. Lond. B Biol. Sci. 260, 139–148[Medline] [Order article via Infotrieve]
  57. Reeves, D. C., Goren, E. N., Akabas, M. H., and Lummis, S. C. R. (2001) J. Biol. Chem. 276, 42035–42042[Abstract/Free Full Text]
  58. Kash, T. L., Jenkins, A., Kelley, J. C., Trudell, J. R., and Harrison, N. L. (2003) Nature 421, 272–275[CrossRef][Medline] [Order article via Infotrieve]
  59. Campo-Caro, A., Sala, S., Ballesta, J. J., Vicente-Agulló, F., Criado, M., and Sala, F. (1996) Proc. Natl. Acad. Sci. U. S. A. 93, 6118–6123[Abstract/Free Full Text]
  60. Grosman, C., Salamone, F. N., Sine, S. M., and Auerbach, A. (2000) J. Gen. Physiol. 116, 327–340[Abstract/Free Full Text]
  61. Grosman, C., Zhou, M., and Auerbach, A. (2000) Nature 403, 773–776[CrossRef][Medline] [Order article via Infotrieve]
  62. Cymes, G. D., Grosman, C., and Auerbach, A. (2002) Biochemistry 41, 5548–5555[CrossRef][Medline] [Order article via Infotrieve]

Add to CiteULike CiteULike   Add to Complore Complore   Add to Connotea Connotea   Add to Del.icio.us Del.icio.us   Add to Digg Digg   Add to Reddit Reddit   Add to Technorati Technorati    What's this?


This article has been cited by other articles:


Home page
 JGPHome page
M. Bali and M. H. Akabas
The Location of a Closed Channel Gate in the GABAA Receptor Channel
J. Gen. Physiol., January 29, 2007; 129(2): 145 - 159.
[Abstract] [Full Text] [PDF]

Home page
 J. Biol. Chem.Home page
L. Chen, K. A. Durkin, and J. E. Casida
Spontaneous Mobility of GABAA Receptor M2 Extracellular Half Relative to Noncompetitive Antagonist Action
J. Biol. Chem., December 15, 2006; 281(50): 38871 - 38878.
[Abstract] [Full Text] [PDF]

Home page
 Proc. Natl. Acad. Sci. USAHome page
Y. Paas, G. Gibor, R. Grailhe, N. Savatier-Duclert, V. Dufresne, M. Sunesen, L. P. de Carvalho, J.-P. Changeux, and B. Attali
Pore conformations and gating mechanism of a Cys-loop receptor
PNAS, November 1, 2005; 102(44): 15877 - 15882.
[Abstract] [Full Text] [PDF]

Home page
 J. Biol. Chem.Home page
R. Hawthorne and J. W. Lynch
A Picrotoxin-specific Conformational Change in the Glycine Receptor M2-M3 Loop
J. Biol. Chem., October 28, 2005; 280(43): 35836 - 35843.
[Abstract] [Full Text] [PDF]

Home page
 J. Biol. Chem.Home page
A. K. Bera and M. H. Akabas
Spontaneous Thermal Motion of the GABAA Receptor M2 Channel-lining Segments
J. Biol. Chem., October 21, 2005; 280(42): 35506 - 35512.
[Abstract] [Full Text] [PDF]

Home page
 J. Biol. Chem.Home page
J. Horenstein, P. Riegelhaupt, and M. H. Akabas
Differential Protein Mobility of the {gamma}-Aminobutyric Acid, Type A, Receptor {alpha} and {beta} Subunit Channel-lining Segments
J. Biol. Chem., January 14, 2005; 280(2): 1573 - 1581.
[Abstract] [Full Text] [PDF]

This Article
Right arrow Abstract Freely available
Right arrow Full Text (PDF)
Right arrow All Versions of this Article:
279/27/28149    most recent
M403545200v1
Right arrow Submit a Letter to Editor
Right arrow Alert me when this article is cited
Right arrow Alert me when eLetters are posted
Right arrow Alert me if a correction is posted
Right arrow Citation Map
Services
Right arrow Email this article to a friend
Right arrow Similar articles in this journal
Right arrow Similar articles in PubMed
Right arrow Alert me to new issues of the journal
Right arrow Download to citation manager
Right arrowRequest Permissions
Citing Articles
Right arrow Citing Articles via HighWire
Right arrow Citing Articles via Google Scholar
Google Scholar
Right arrow Articles by Panicker, S.
Right arrow Articles by Slesinger, P. A.
Right arrow Search for Related Content
PubMed
Right arrow PubMed Citation
Right arrow Articles by Panicker, S.
Right arrow Articles by Slesinger, P. A.
Social Bookmarking
Add to CiteULike   Add to Complore   Add to Connotea   Add to Del.icio.us   Add to Digg   Add to Reddit   Add to Technorati  
What's this?

HOME HELP FEEDBACK SUBSCRIPTIONS ARCHIVE SEARCH TABLE OF CONTENTS
 All ASBMB Journals   Molecular and Cellular Proteomics 
 Journal of Lipid Research   ASBMB Today 
Copyright © 2004 by the American Society for Biochemistry and Molecular Biology.
Advertisement
spacer
Advertisement
Advertisement