Structure of human CALHM1 reveals key locations for channel regulation and blockade by ruthenium red - PubMed (original) (raw)

Structure of human CALHM1 reveals key locations for channel regulation and blockade by ruthenium red

Johanna L Syrjänen et al. Nat Commun. 2023.

Abstract

Calcium homeostasis modulator 1 (CALHM1) is a voltage-dependent channel involved in neuromodulation and gustatory signaling. Despite recent progress in the structural biology of CALHM1, insights into functional regulation, pore architecture, and channel blockade remain limited. Here we present the cryo-EM structure of human CALHM1, revealing an octameric assembly pattern similar to the non-mammalian CALHM1s and the lipid-binding pocket conserved across species. We demonstrate by MD simulations that this pocket preferentially binds a phospholipid over cholesterol to stabilize its structure and regulate the channel activities. Finally, we show that residues in the amino-terminal helix form the channel pore that ruthenium red binds and blocks.

© 2023. The Author(s).

PubMed Disclaimer

Conflict of interest statement

The authors declare no competing interests.

Figures

Fig. 1

Fig. 1. The cryo-EM structure of human CALHM1.

a Cryo-EM density (top) and molecular models (bottom) of human CALHM1 viewed from the side of the membrane, the extracellular region, and the cytoplasm. The pore distance indicated by the double-headed arrow is measured between the Gln33 Cα positions of chains A and E. b Inter-subunit interactions between transmembrane helices of human CALHM1 (left). Neighboring subunits are colored red (TMD4 of subunit A) and green (TMD2 of subunit B). Inter-subunit interactions within the C-terminal helix (CTH) are shown (right) across four subunits in orange (subunit H), red (subunit A), green (subunit B), and splitpea (subunit C). Residues conserved across human, mouse, chicken, killifish, and zebrafish are indicated by asterisks. c Topology schematic of the human CALHM1 protomer (left) with TMDs, extracellular helix (ECH), and cytoplasmic C-terminal helix (CTH) indicated. The protomer of the chicken CALHM1∆ct model superposed with the human CALHM1∆ct model (middle) and with the chicken CALHM1 model (right; PBD code: 6VAM).

Fig. 2

Fig. 2. The conserved hydrophobic pocket preferentially binds to phospholipid.

a A cross-section of human CALHM1 shows the location of the hydrophobic pocket filled with a phospholipid. Residues that form the hydrophobic pocket are shown in stick representation. There are eight hydrophobic pockets per CALHM1 octamer. The inset shows a zoomed-in view of the hydrophobic pocket viewed from the pore. Residues are numbered in red (TMD3 and TMD4) or green (TMD2). b Coarse-grained PMF calculations suggest that binding of POPC into this pocket is thermodynamically favored over that of cholesterol with energy minimums of approximately −6.4 kcal/mol and −1.9 kcal/mol, respectively (right). The arrow indicates the direction of steering for which the reaction coordinate was generated (left). The steered POPC molecule is colored from yellow to red according to progress along the steered MD simulation, with light yellow beads representing POPC at the start, and red beads at the end of the steered MD. Two other bound POPC molecules are shown in blue, brown and green beads, representing different bead types. Error bands are 1 standard deviation generated from 200 rounds of bootstrap analysis.

Fig. 3

Fig. 3. The conserved hydrophobic pocket is a key locus for structural integrity and channel functions.

a Positions of residues (sticks) within the hydrophobic pocket, analyzed by site-directed mutagenesis. b Current density pA/pF at +100 mV for each point mutant (color coded as in panel a). Each data point represents a measurement on a different cell (wild-type hCALHM1Δct, n = 10; Ile109Trp hCALHM1Δct, n = 11; Ala116Trp hCALHM1Δct, n = 9; Ala199Trp hCALHM1Δct, n = 9; Thr196Trp hCALHM1Δct, n = 10; Val192Trp hCALHM1Δct, n = 6; vector, n = 9; Val112Trp hCALHM1Δct, n = 5; Leu67Trp hCALHM1Δct, n = 7). Boxes represent the median, 25th, and 75th percentile values, and the whiskers represent the minimum and maximum values. *** denotes p < 0.001 versus wild-type. An unpaired two-tailed _t_-test with Welch’s correction was used to analyze data. _P_-values are as follows: hCALHM1Δct vs vector, p = 0.0001; hCALHM1Δct vs Leu67Trp hCALHM1Δct, p = 0.9925; hCALHM1Δct vs Ile109Trp hCALHM1Δct, p = 0.003; hCALHM1Δct vs Val112Trp hCALHM1Δct, p = 0.4962; hCALHM1Δct vs Ala116Trp hCALHM1Δct, p = 0.4395; hCALHM1Δct vs Val192Trp hCALHM1Δct, p = 0.0006; hCALHM1Δct vs Thr196Trp hCALHM1Δct, p = 0.3810; hCALHM1Δct vs Ala199Trp hCALHM1Δct, p = 0.3253. WTΔct is used to refer to hCALHM1Δct in the graph. Source data are provided as a Source Data file. c Assessing surface expression of hCALHM1Δct wild-type and selected point mutations. The hCALHM1Δct (C-terminally tagged with the 1D4 epitope) trafficked to the plasma membrane are biotinylated at Lys123 by NHS-SS-biotin, followed by detergent extraction and purification by Streptavidin-Sepharose. The ‘pulled-down’ human CALHM1Δct is mixed with a FITC-labeled anti-1D4 antibody and analyzed by ab-FSEC for protein quantity (fluorescence intensity) and size (retention time). d Representative western blots of hCALHM1Δct wild-type and selected point mutations. The top blot shows samples with and without treatment of NHSSS-biotin, taken after detergent solubilization. The samples were probed with anti-1D4 antibody and anti-β-actin antibody, representing whole cell expression of CALHM1 (human CALHM1Δct) and the β-actin loading control, respectively. The bottom blot shows samples probed with anti-1D4 antibody, representing detection of human CALHM1Δct and point mutant samples expressed at the cell surface with and without treatment of NHS-SS-biotin. This assay was repeated independently with similar results two (Leu67Trp, Thr196Trp) to three (Ile109Trp, Ala199Trp, Val192Trp, Ala116Trp, Val112Trp and hCALHM1∆ct) times. Source data are provided as a source data file. WTΔct is used to refer to hCALHM1Δct in the blots. e Representative abFSEC traces using a superose 6 10/300 size-exclusion chromatography for the wild-type (WT) and the selected mutants. Peaks representing the void, the hCALHM1Δct- and point mutant-1D4 antibody complexes, and the free 1D4 antibody are observed. The zoom-in view of the hCALHM1-1D4 peak (inset) shows differences in the protein amount.

Fig. 4

Fig. 4. The cryo-EM structure of Ile109Trp hCALHM1WT∆ct (hCALHM1I109W∆ct) in the presence of RuR.

a, b The C8 symmetry imposed cryo-EM density (a) and the molecular model (b) of RuR-hCALHM1I109W∆ct viewed from the side of the membrane, the extracellular region, and the cytoplasm. c A zoomed-in view of subunit G showing the phospholipid (cryo-EM density and sticks), Ile109Trp, and the residues interacting with them (sticks; left). A schematic representation highlighting the positions of TMD1, the NTH, Ile109Trp (W), Phe19 (F), Asn21 (N), Ala28 (A; black lines) (right). The Ile109Trp mutation stabilizes the interaction between TMD3 and the pore composed of TMD1 and NTH via direct and phospholipid-mediated interactions. d PMF calculations of cholesterol or POPC in hCALHM1Δct (WT∆ct, blue) or hCALHM1I109WΔct (I109W, green). Both models contained NTHs without RuR. The hCALHM1Δct model was built based on the RuR-hCALHM1I109WΔct model but with I109W reverted to isoleucine. Bulk phase is reached at different points along the reaction coordinates for respective lipids, as initial steered MD was propagated along the _x_-axis along the bilayer normal from binding pockets located at different rotational angles in the simulation box. Error bands are 1 standard deviation generated from 200 rounds of bootstrap analysis.

Fig. 5

Fig. 5. All atom MD simulations of hCALHM1WT∆ct in the presence and absence of lipids.

Population density of Cα atom RMSDs for the hCALHM1Δct homology model based on the RuR-hCALHM1I109WΔct model with either cholesterol (orange) or POPC (blue) bound to each hydrophobic pocket or in apo state (red). Means of corresponding histograms are plotted as vertical dashed lines of corresponding color. Histograms are each comprised of 50 bins each.

Fig. 6

Fig. 6. Ruthenium red binding site in the central pore.

a A cross-section of the cryo-EM map (top) and model (bottom) of hCALHM1I109W∆ct without imposing symmetry (c1). The RuR density is shown as white surface (top) and gray mesh (bottom). The Cα atoms of the pore-lining residues, Gln10, Gln13, and Gln16, are shown as spheres. Asp121 is in the vicinity of NTH. b Representative current traces and the concentration-response of the hCALHM1I109W∆ct channel blockade by RuR at +60 mV (IRuR/I0) displayed as a bar chart. Data are represented as individual points, the bars show the mean and the whiskers indicate the standard error of the mean (n = 6, 4, and 7 cells for 2, 10, and 20 µM RuR application, respectively). Definitions of currents without RuR (I0) and with RuR (IRuR) are indicated in the middle trace. WTΔctI109W is used to refer to hCALHM1I109W∆ct in the chart and traces. Source data are provided as a source data file. c A cartoon of the central pore, highlighting the positions of Glu10, Glu13, and Glu16 (spheres) around the RuR binding site (cryo-EM density for RuR shown as mesh; RuR model in stick representation). d Representative current traces in the absence (black traces) and presence (red traces) of 20 µM RuR at +60 mV for the Glu10Arg, Glu13Arg, and Glu16Arg point mutants. The graph shows the extent of the channel blockade (IRuR/I0) calculated from the recordings. Boxes represent the median, 25th, and 75th percentile values, and the whiskers represent the minimum and maximum values (n = 7, 4, 4, and 3 cells for hCALHM1I109W∆ct, Glu10Arg, Glu13Arg, and Glu16Arg, respectively). ****, ***, and ** denote p < 0.0001, p < 0.001, and p < 0.01, respectively, versus basal conditions (absence of RuR) for each construct studied (two-tailed paired _t_-test: p < 0.0001 for WT, p = 0.0046 for Glu10Arg, p = 0.0001 for Glu13Arg, and p = 0.2369 for Glu16Arg). WTΔctI109W is used to refer to hCALHM1I109W∆ct in the chart and traces. Source data are provided as a source data file.

Similar articles

Cited by

References

    1. Syrjanen J, Michalski K, Kawate T, Furukawa H. On the molecular nature of large-pore channels. J. Mol. Biol. 2021;433:166994–166994. doi: 10.1016/j.jmb.2021.166994. - DOI - PMC - PubMed
    1. Ma Z, et al. Calcium homeostasis modulator 1 (CALHM1) is the pore-forming subunit of an ion channel that mediates extracellular Ca2+ regulation of neuronal excitability. Proc. Natl Acad. Sci. USA. 2012;109:E1963–E1971. doi: 10.1073/pnas.1204023109. - DOI - PMC - PubMed
    1. Dreses-Werringloer U, et al. A polymorphism in CALHM1 influences Ca2+ homeostasis, Aβ levels, and Alzheimer’s disease risk. Cell. 2008;133:1149–1161. doi: 10.1016/j.cell.2008.05.048. - DOI - PMC - PubMed
    1. Dreses-Werringloer U, et al. CALHM1 controls the Ca2+-dependent MEK, ERK, RSK and MSK signaling cascade in neurons. J. Cell Sci. 2013;126:1199–1206. doi: 10.1242/jcs.117135. - DOI - PMC - PubMed
    1. Taruno A, et al. CALHM1 ion channel mediates purinergic neurotransmission of sweet, bitter and umami tastes. Nature. 2013;495:223–226. doi: 10.1038/nature11906. - DOI - PMC - PubMed

Publication types

MeSH terms

Substances

LinkOut - more resources