Zimmermann I et al. (2012), Inhibition of the prokaryotic pentameric ligand...

XB-ART-46953

PLoS Biol 2012 Jan 01;1011:e1001429. doi: 10.1371/journal.pbio.1001429.

Show Gene links Show Anatomy links

Inhibition of the prokaryotic pentameric ligand-gated ion channel ELIC by divalent cations.

Zimmermann I , Marabelli A , Bertozzi C , Sivilotti LG , Dutzler R .

Abstract
The modulation of pentameric ligand-gated ion channels (pLGICs) by divalent cations is believed to play an important role in their regulation in a physiological context. Ions such as calcium or zinc influence the activity of pLGIC neurotransmitter receptors by binding to their extracellular domain and either potentiate or inhibit channel activation. Here we have investigated by electrophysiology and X-ray crystallography the effect of divalent ions on ELIC, a close prokaryotic pLGIC homologue of known structure. We found that divalent cations inhibit the activation of ELIC by the agonist cysteamine, reducing both its potency and, at higher concentrations, its maximum response. Crystal structures of the channel in complex with barium reveal the presence of several distinct binding sites. By mutagenesis we confirmed that the site responsible for divalent inhibition is located at the outer rim of the extracellular domain, at the interface between adjacent subunits but at some distance from the agonist binding region. Here, divalent cations interact with the protein via carboxylate side-chains, and the site is similar in structure to calcium binding sites described in other proteins. There is evidence that other pLGICs may be regulated by divalent ions binding to a similar region, even though the interacting residues are not conserved within the family. Our study provides structural and functional insight into the allosteric regulation of ELIC and is of potential relevance for the entire family.

PubMed ID: 23185134
PMC ID: PMC3502511
Article link: PLoS Biol
Grant support: [+]

Species referenced: Xenopus laevis
Genes referenced: cdkn1a nsg1

Article Images: [+] show captions

	Figure 2. Inhibition of ELIC by different divalent cations.Doseâresponse relationships of ELIC activated by cysteamine at different concentrations of Ba2+ (A), Sr2+ (B), and Mg2+ (C) and ELIC activated by propylamine at different concentrations of Zn2+ (D). (E) Schild plot quantifying the inhibition by different divalent cations. EC50 values were obtained from fits to data shown in panels (AâD). Potencies of the Antagonists (pA values) were obtained by linear regression, and the intersection with the x-axis is indicated. The data presented in panels (AâD) are averages from at least 5 oocytes; errors are SD. The solid lines show fits to a Hill equation. Currents were recorded at â40 mV.
	Figure 3. Structure of ELIC in complex with divalent cations.(A) Sequence of ELIC with secondary structure elements indicated below. Residues contributing to ion coordination in different sites are highlighted (Sin, yellow; Sout, red; Spore, cyan; Arg 91 in the ligand-binding pocket, Slig, grey). (B) Anomalous difference electron density of ELIC in complex with Ba2+ superimposed on the structure of the ELIC pentamer (shown as ribbon representation). Ion-binding sites are labeled. Close-up of Spore (C) and Sin (D). The protein is shown as CÎ±-trace with selected side-chains close to Ba2+ (red sphere) shown as sticks. (E) Close-up of Sout. The protein is shown as CÎ±-trace with selected side-chains close to Ba2+ (red sphere) shown as sticks. The 2Fo-Fc electron density of a dataset from a crystal of space group P21 was calculated at 3.8 Ã and contoured at 1 Ï (shown in cyan). The refined model used to calculate phases did not contain Ba2+-ions. The anomalous difference electron densities shown in (BâE) (red mesh) were calculated from the same dataset at 5 Ã and contoured at 5 Ï. Crystals of space group P43 showed a qualitatively similar picture. Structures in Figures 3â7 were prepared with DINO (www.dino3d.org).
	Figure 4. Divalent ion inhibition in mutants of the nonregulatory sites Spore and Sin.Doseâresponse relationships of the ELIC point mutant R91A (A) activated by cysteamine at different concentrations of Ca2+. (B) Anomalous difference electron density (calculated at 5 Ã and contoured at 5 Ï) from data of the mutant N251A in complex with Ba2+ superimposed on a model of ELIC in ribbon representation. Ion-binding sites are indicated. Doseâresponse relationships of the ELIC point mutants N251A (C), S84A (D), and D86A (E) activated by cysteamine at different concentrations of Ca2+. The data presented in panels (A), (B), (D), and (E) are averages from at least 5 oocytes; errors are SD. The solid lines show fits to a Hill equation. Currents were recorded at â40 mV. A doseâresponse curve of WT in the absence of Ca2+ (dashed line) is shown for comparison. (F) Schild plots quantifying the inhibition of ELIC mutants by Ca2+. EC50 values were obtained from fits to data shown in panels (A), (B), (D), and (E). Potencies of the antagonists (pA values) were obtained by linear regression; the intersection with the x-axis is indicated (â¢). WT is shown for comparison.
	Figure 5. Divalent ion inhibition in mutants of the regulatory site Sout.Doseâresponse relationships of the ELIC mutants D113A (A) and D158A (B), E150A (C), and the double mutant D113A/D158A (D) activated by cysteamine at different concentrations of Ca2+ are shown. (E) Schild plots quantifying the inhibition of ELIC mutants by Ca2+. EC50 values were obtained from data shown in panels (AâD). Potencies of the antagonists (pA values) were obtained by linear regression; the intersection with the x-axis is indicated (â¢). WT is shown for comparison. (F) Graphical depiction of potencies for Ca2+ inhibition in different mutants. (G) Maximum current response of the double mutant D113A/D158A at different Ca2+ concentrations. The currents are normalized to the maximum response in the absence of Ca2+. Whole cell currents measured at â40 mV with the two-electrode voltage clamp technique are shown in red (the averages of at least 5 oocytes are shown; errors are SD). Single channel currents from the double mutant D113A/D158A were measured in the outside-out configuration and are shown in green. WT macroscopic and single channel currents are shown as dashed lines for comparison. (H) Anomalous difference electron density (calculated at 5 Ã and contoured at 5 Ï) from data of the double mutant D113A/D158A in complex with Ba2+ is superimposed on a model of ELIC in ribbon representation. Ion-binding sites are indicated. (I) Doseâresponse relationships of the double mutant D113A/D158A activated by propylamine at different concentrations of Zn2+. The data presented in panels (AâD) and (I) are averages from at least 5 oocytes; errors are SD. The solid lines show fits to a Hill equation. Currents were recorded at â40 mV. A doseâresponse curve of WT in the absence of Ca2+ (dashed line) is shown for comparison.
	Figure 6. Inhibition by ACh.(A) Anomalous difference electron density (calculated at 5 Ã and contoured at 6 Ï) from data of WT in complex with TMAs is superimposed on a model of ELIC in ribbon representation. The ligand-binding site is indicated. Doseâresponse relationships upon activation with cysteamine of the ELIC mutants R91A (B) and the double mutant D113A/D158A (C) at different concentrations of ACh. (F) Schild plots quantifying the inhibition of ELIC mutants by ACh. EC50 values were obtained from data shown in panels (BâC). Doseâresponse relationships upon activation with cysteamine of WT in the presence of either 0.25 (E), 0.5 (F), or 1 mM Ca2+ (G). (H) Schild plots quantifying the inhibition of ELIC mutants by ACh in the presence of Ca2+. EC50 values were obtained from data shown in panels (EâG). Potencies of the antagonists (pA values) were obtained by linear regression; the intersection with the x-axis is indicated (â¢). The data presented in panels (BâC) and (EâG) are averages from at least 5 oocytes; errors are SD. The solid lines show fits to a Hill equation. Currents were recorded at â40 mV. A doseâresponse curve of WT in the absence of ACh and Ca2+ (dashed line) is shown for comparison.
	Figure 7. Potential mechanisms.(A) Interactions in the regulatory divalent ion-binding site of ELIC (left) in comparison with a regulatory Ca2+ binding site of the BK-channel (middle) and the Zn2+-transporter YiiP (right). (B) Schematic model of a potential mechanism for the inhibitory effect of divalent ions. The two rows show simplified schemes for channel activation in control conditions (top) and in the presence of divalent ions. From left to right, the schemes show that binding of agonist molecules (red ovals) to the extracellular domain (with microscopic affinity Kd) is followed by conformational changes (yellow background) that result in channel opening. Channel gating (described by the efficacy equilibrium constant E) is impaired when the channel is bound to divalent ions (yellow circles, ECa2+). The decrease in agonist efficacy is likely to be due to a change in the rate of opening, as shown by the size of the arrows in the last step of the reaction. (C) Schematic mechanism of how binding sites located on similar places of an oligomeric channel could alternately stabilize the closed or open conformation of the channel.
	Figure 1. Inhibition of ELIC by calcium.(A) ELIC single channel currents in the presence of different extracellular concentrations of Ca2+ and all-points amplitude histograms (recordings were from oocyte outside-out patches at â80 mV holding potential). (B) Maximum ELIC responses to saturating concentrations of the agonist cysteamine in the absence and presence of Ca2+. Currents were measured from oocytes held at â40 mV under two-electrode voltage clamp. Agonist application is indicated by a bar. Responses to cysteamine in the absence of extracellular Ca2+ are followed by responses in the presence of Ca2+ (concentration in mM as shown) and by recovery application of cysteamine in the absence of Ca2+. (C) Plot of maximum agonist responses and single channel currents at different Ca2+ concentrations. The currents are normalized to the control values (in the absence of Ca2+). Maximum cysteamine currents (blue symbols) were measured with the two-electrode voltage clamp technique in oocytes as in panel B. Single channel currents (green symbols) were measured in the outside-out configuration as in panel A. (D) Cysteamine doseâresponse relationships for ELIC in the presence of different concentrations of Ca2+. (E) Equilibrium cysteamine binding isotherms determined by ITC for ELIC in the absence (left) and presence of Ca2+. Solid curves represent fits to a single-site binding isotherm with Keffâ=â0.7 mM (no Ca2+) and 1.5 mM (0.6 mM Ca2+), respectively. (F) Cysteamine doseâresponse relationships for ELIC in the presence of different concentrations of ACh. (G) Schild plot quantifying the inhibition by Ca2+ and ACh. EC50 values were obtained from fits to data shown in panels (D) and (F). Potencies of the antagonists (pA values) were obtained by linear regression, and the intersection with the x-axis is indicated (â¢). (H) Fraction of the maximum current response at different Ca2+ concentrations. The solid line shows a fit to a Langmuir equation with a Ki of 6 mM. The data presented in panels (C), (D), (F), and (H) are averages from at least five oocytes; errors are SD. The solid lines in (D) and (F) show fits to a Hill equation. Currents were recorded at â40 mV. (I) Activation kinetics of macroscopic currents of ELIC activated by propylamine in response to fast solution exchange at different Ca2+ concentrations. ELIC was expressed in HEK 293 cells, and currents were recorded from excised patches in the outside-out configuration at â100 mV.

References [+] :

Adams, The permeability of endplate channels to monovalent and divalent metal cations. 1980, Pubmed