Natural xanthones as α-Mangostin induce vasorelaxation involving key gating residues in the S6 domain of BK channels
- Institute of Physiology, Christian-Albrechts-University Kiel, Germany
- Department of Human Medicine, Health and Medical University Erfurt, Germany
Figures
Figure 1 with 1 supplement
Mangostin potently activates BKα and BKα/β1 channels compared to other potassium channel representatives.
(A) Current fold change ± SEM of currents of representatives from different potassium channel families upon application of 10 µM α-Mangostin. K2P and KCa channel currents were recorded with ramp protocols from –100 to +50 mV and analyzed at +40 mV; Kv1.1 and Kv1.3 channel currents were evoked using a rectangle pulse to +40 mV, and hERG currents were recorded with a rectangle pulse to +60 mV followed by hyperpolarization to –120 mV, which was analyzed. Kir channels were recorded using ramp protocols from –150 to +50 mV, and currents were analyzed at –140 mV. All measurements were made in transiently transfected HEK293 cells in physiological potassium gradients with 100 nM intracellular free Ca2+ with a holding potential of –80 mV. TWIK-1m and TWIK-2m denote channels where the retrieval motif was removed to improve membrane expression, and intracellular K+ was exchanged for Rb+ to enhance currents. (B) Representative current traces of channels that were inhibited more than 60% by 10 µM α-Mangostin. (C) Representative current traces of BKα and BKα/β1 channels activated by 10 µM α-Mangostin (α-M). (D) Current fold change ± SEM after application of α-Mangostin, γ-Mangostin, and a dietary supplement to TREK-1, BKα, and BKα/β1 channels. Currents were recorded and analyzed as in (A). (E) Representative time course of the dose-dependent activation of BKα channels by increasing concentrations of α-Mangostin (left) and the resulting dose–response relationships for BKα and BKα/β1 channels (right). Currents were recorded and analyzed as in (A); the grey data point at 12.5 µM was not included in the Hill fit. Data and statistics see Source data file 1.
Figure 1—figure supplement 1
Effect of α-Mangostin on the current of potassium channels from different families.
Representative whole-cell current traces recorded from the indicated channels before and after application of 10 µM α-Mangostin. K2P and Kir channels were recorded using a ramp protocol from –100 to +50 mV or –150 to +50 mV, respectively, Kv1.1 and Kv1.3 channels were recorded using a rectangle pulse to +40 mV, and hERG was recorded with a rectangle pulse to +60 mV followed by hyperpolarization to –120 mV. All measurements were made in transiently transfected HEK293 cells in physiological potassium gradients with 100 nM intracellular free Ca2+. Intracellular K+ was exchanged for Rb+ for TWIK channels. TWIK-2m denotes channels where the dibasic retrieval motif was removed to enhance membrane expression (I289A/L290A). Data and statistics see Source data file 1.
Figure 2 with 1 supplement
Effects of α-Mangostin on BKα and BKα/β1 channel gating.
(A) Representative current traces for BKα and BKα/β1 channels before and after activation by 10 µM α-Mangostin. Cells were measured in symmetrical 140 mM bi-ionic conditions with 140 mM Cs+ as the intracellular ion and 100 nM free Cai2+. Currents were elicited by a family of rectangle pulses from –100 to up to +300 mV in 20 mV increments from a holding potential of –80 mV, followed by repolarization to –50 mV to elicit inward tail currents. (B) GV relationships for BKα and BKα/β1 channels in the basal state and activated by 10 µM α-Mangostin, derived from tail current analysis of recordings as in (A). The gray arrow illustrates the shift of voltage activation toward more negative voltages caused by α-Mangostin and the inset shows the slope ± SEM of the Boltzmann fits. (C) Activation and deactivation kinetics of BKα and BKα/β1 channels in the basal state and after activation by 10 µM α-Mangostin. The τ ± SEM of activation/deactivation was determined from exponential fits to the current traces at +100 mV, as shown by the green and purple colored lines in panel (A). (D) GV relationships of BKα channels in different free Cai2+ concentrations before and after activation by 10 µM α-Mangostin, derived from tail current analysis as above. Pulse voltages were –100 to +300 mV/repolarization to –50 mV from a holding potential of –80 mV for 100 nM free Cai2+, –120 to +200 mV/repolarization to –50 mV from a holding potential of –80 mV for 1 µM free Cai2+, and –160 mV to +100 mV/repolarization to –80 mV from a holding potential of –120 mV for 10 µM free Cai2+, in symmetrical 140 mM bi-ionic conditions with 140 mM Cs+ as the intracellular ion. The V½ values ± SEM before and after α-Mangostin application and the resulting shifts (ΔV½ ± SEM) are shown in the bar graphs. Data and statistics see Source data file 1.
Figure 2—figure supplement 1
Change of activation and deactivation time course upon α-Mangostin activation of BKα channels measured with 10 µM Cai2+.
(A) Time constant (τ) of activation for the voltage range between 0 and +100 mV before and after activation by 10 µM α-Mangostin (n = 7). (B) Time constants of activation (n = 7) and deactivation (n = 6) at +20 mV are shown to illustrate their change within the physiological voltage range. Time constants were obtained by fitting a monoexponential function to the current traces after rectangle depolarization (for activation) and after a repolarization step to elicit tail currents (for deactivation) as shown in Figure 2A. Data and statistics see Source data file 1.
Figure 3
Activation mechanism of α-Mangostin.
(A) Exemplary current traces of a single BKα channel in the basal state and after activation by 10 µM α-Mangostin (1 min recordings; inset shows 1 s; O and C denote open and closed levels). Single-channel currents were recorded at +40 mV in inside-out patches from transiently transfected HEK293 cells in symmetrical potassium gradients with 100 mM free Cai2+ (n = 4–7). (B) All-points histograms with Gaussian fits for the basal and α-Mangostin-activated state, and bar graphs of the derived mean ± SEM open probabilities (Po) and amplitudes. The inset magnifies the open peak in the basal state. (C) Closed and open dwell time histograms with fits for channels in the basal and in the α-Mangostin-activated state derived after event detection in single-channel measurements as shown in (A). (D) Exemplary current traces of a single BKα channel recorded as in (A), but with 5 µM free Cai2+. (E) Burst duration, duration of long closed times, and number of openings per burst derived from burst analysis (n = 3 patches). (F) Closed and open dwell time distribution within bursts with fits for channels in the basal and in the α-Mangostin-activated state. (G) Normalized mean ± SEM currents of BKα channels before and after application of 3 or 10 µM α-Mangostin in the closed state. Channels were held closed at –80 mV and only very shortly pulsed to +60 mV after 5 min incubation to assess the current size/activation state of the first and the following pulses. Measurements were done in transiently transfected HEK293 cells in whole-cell mode in physiological potassium gradients with 100 nM Cai2+ as shown by the representative current traces to the right, and steady-state currents were analyzed (grey arrow). Data and statistics see Source data file 1.
Figure 4 with 3 supplements
Investigation of the binding site of α-Mangostin in BKα channels.
(A) Competition experiment showing a reduction of the block caused by THexA in the presence of α-Mangostin. Left, representative current traces for different THexA concentrations in the presence of 10 µM α-Mangostin; middle, competition experiment analysis showing the relative current of BKα channels in different THexA concentrations in the absence and in the presence of 10 µM α-Mangostin or 100 µM BC5, which does not bind in the pore; and right, estimated IC50 values of THexA alone and in the presence of α-Mangostin or BC5. Whole-cell currents were recorded from transiently transfected HEK293 cells with a ramp protocol (–100 to +50 mV) in a physiological potassium gradient with 100 nM free Cai2+ and data are shown as mean ± SEM at +40 mV. (B) Molecular docking of α-Mangostin to the human BK channel structure (PDB ID 6v3g, Ca2+-free state). The full-length structure (green) was reduced to the inner pore region (pink) for the docking, and the zoom-in shows the best pose for α-Mangostin with interacting residues in stick representation; green residues mark hits from the following functional assay. Protein chain B was removed for clarity. See Figure 4—figure supplement 3 for the Ca2+-bound state. (C) Voltage of half-maximal activation (V½) before and after activation by 10 µM α-Mangostin in different pH and the resulting shifts in V½ (ΔV½). The pH was changed intra- and extracellularly. (D) Voltage of half-maximal activation (V½) before and after activation by 10 µM α-Mangostin, and the resulting shifts in V½ (ΔV½). (E) GV relationships for the six BKα mutants in the S6 segment. (F) Voltage of half-maximal activation (V½) before and after activation by 1 µM GoSlo-SR-5-6, the resulting shifts in V½ (ΔV½), and the GV relationships for the wildtype and two BKα mutants. GV relationships were measured as in Figure 2, and all data represent mean ± SEM. Data and statistics see Source data file 1.
Figure 4—figure supplement 1
Activation of TREK-1 channels by α-Mangostin and investigation of the binding region.
(A) Dose-dependent activation of TREK-1 channels (left) and the derived dose–response relationship for the WT and for the L304C mutant located at the fenestration entrance. (B) Dose–response relationship for TPA in the absence or presence of 5 µM α-Mangostin (left) and the IC50 values derived from Hill fits (right). (C) Cysteine scanning mutagenesis of M2 and M4 residues belonging to the fenestration site revealed the fold change of their currents after application of 5 µM α-Mangostin. (D) Location of the hit residues (green) in the TREK-1 crystallographic structure (PDB ID 6cq6). All measurements were done in transiently transfected HEK293 cells in physiological potassium gradients with a ramp protocol from –100 to +50 mV (holding potential –80 mV) and currents were analyzed at +40 mV. Data and statistics see Source data file 1.
Figure 4—figure supplement 2
Activation of BKα channels by 10 µM α-Mangostin in different pH.
GV relationships for BKα channels in the basal state and activated by 10 µM α-Mangostin, derived from tail current analysis. HEK 293 cells were measured in whole-cell configuration in symmetrical 140 mM bi-ionic conditions with 140 mM Cs+ as the intracellular ion and 100 nM free Cai2+. Currents were elicited by a family of rectangle pulses from –100 to up to +300 mV in 20 mV increments from a holding potential of –80 mV, followed by repolarization to –50 mV to elicit inward tail currents. The pH was changed intra- and extracellularly. Data and statistics see Source data file 1.
Figure 4—figure supplement 3
Molecular docking of α-Mangostin to the Ca2+-free and Ca2+-bound human BKα channel.
Central pore region of the Ca2+-free (PDB ID 6v3g) and the Ca2+-bound structure (PDB ID 6v38) comprised of S5, pore helices, and S6 segments with the best pose of α-Mangostin, viewed from the intracellular side (A) and the membrane plane (B). (C) Visualization of the binding pocket comprising S6 segments (chain B removed for clarity), and (D) potentially interacting residues in stick representation with the mutants exhibiting a loss in the shift of voltage activation upon α-Mangostin application highlighted in green. (E) Overlay of the S6 helices in the Ca2+-free (blue) and the Ca2+-bound state (grey) illustrating the upwards movement that concomitantly shifts α-Mangostin to a more horizontal pose, and a table comparing the interactions in both states.
Figure 5 with 1 supplement
α-Mangostin activation of BK channels in physiological settings.
(A) Representative whole-cell current traces of BKα/β1 channels alone and BKα/β1 coexpressed with Cav1.2 channels before and after application of 10 µM α-Mangostin. Currents were measured in Cai2+-free conditions in a physiological potassium gradient with a family of voltage steps from –50 to +50 mV in 10 mV increments. The inset shows voltage activation with a family protocol up to +200 mV to show the presence of BKα/β1 channels. (B) Currents of BKα/β1 channels and BKα/β1–Cav complexes before and after application of 10 µM α-Mangostin plotted against voltage. The last panel shows the α-Mangostin-activated currents for the range –50 to 10 mV obtained by subtracting the current before α-Mangostin application from the current after application for each potential (mean ± SEM, n = 8–11 for each condition). (C) Representative contraction force recordings of aortic preparations from mice. 10 µM α-Mangostin were either applied directly to aortic preparations precontracted with 100 nM Noradrenaline (NA; left), or the precontracted preparations were incubated with 100 nM Iberiotoxin (IbTx) before α-Mangostin application (right). The contraction force was analyzed 10 min after α-Mangostin addition (dotted lines in recordings). The bar graph shows the normalized contraction force of preparations as mean ± SEM together with the median (orange). Data and statistics see Source data file 1. DMSO controls are shown in Figure 5—figure supplement 1.
Figure 5—figure supplement 1
Vehicle control for BKα and Cav1.2 channels and aortic tissue.
(A) GV relationships for BKα channels with and without DMSO, derived from tail current analysis (mean ± SEM; n = 4). (B) Current–voltage relationship of Cav1.2 channels with and without DMSO (mean ± SEM; n = 3). HEK 293 cells were measured in whole-cell configuration in symmetrical 140 mM bi-ionic conditions with 140 mM Cs+ as the intracellular ion and 100 nM free Cai2+. Currents were elicited from a holding potential of –80 mV by a family of rectangle pulses from –100 to +240 mV followed by repolarization to –20 mV to elicit inward tail currents for BKα channels, and by a family of rectangle pulses from –50 to +60 mV for Cav channels. Data and statistics see Source data file 1. (C) Representative measurement showing the application of 0.1% DMSO to precontracted mouse aortic tissue in the organ bath.
Author response image 1
Alphafold3 models of BK I308A, L312M, and A316P with α-Mangostin docked to the mutant structures.
The upper row shows an overview of the mutant pore helices (AF3 models) used for molecular docking. The lower row shows the binding region with the wildtype structure overlaid in gray. Only 3 helices are shown for clarity.
Author response image 2
Activation of BKα by 3-hydroxyxanthone.
(A) GV-relationship before and after application of 10 µM 3-hydroxyxanthone. (B) V½ before and after application of 10 µM 3-hydroxyxanthone compared to α-Mangostin and the resulting difference in V½ (ΔV½). Measurements were conducted as described in the main manuscript with 100 nM free Cai2+.
Author response image 3
Double Mutants I308A/L312M, I308A/A316G and L312M/A316G compared to the single mutations in the main manuscript.
The V½ before and after activation with 10 µM α-Mangostin, the resulting shift in V½, and the GV-relationships are shown (n=6-7), measurements were made as in Fig. 4.
Tables
Key resources table
| Reagent type (species) or resource | Designation | Source or reference | Identifiers | Additional information |
|---|---|---|---|---|
| Gene (Homo sapiens) | KCNK2 | GenBank | NM_001017425.2 | hTREK-1b |
| Gene (Homo sapiens) | KCNK16 | GenBank | NM_032115.3 | hTALK-1 |
| Gene (Homo sapiens) | KCNK18 | GenBank | NM_181840.1 | hTRESK |
| Gene (Homo sapiens) | KCNK1 | GenBank; Feliciangeli et al., 2010 | NM_002245.3 | hTWIK-1m; I293A, I294A |
| Gene (Homo sapiens) | KCNK6 | GenBank; Bobak et al., 2017 | NM_004823.1 | hTWIK-2m; I289A, L290A |
| Gene (Homo sapiens) | KCNK9 | GenBank | NM_001282534.1 | hTASK-3 |
| Gene (Homo sapiens) | KCNK13 | GenBank | NM_022054.3 | hTHIK-1 |
| Gene (Mus musculus) | Kcnma1 | GenBank | NM_001014797.3 | mBKα (Slo1.1); G/S-rich N-terminus removed, corresponding to residues 65–1236 of the native channel |
| Gene (Mus musculus) | Kcnmb1 | GenBank | NM_031169.4 | mβ1 |
| Gene (Homo sapiens) | KCNA1 | GenBank | NM_000217.3 | hKv 1.1 |
| Gene (Homo sapiens) | KCNA3 | GenBank | NM_002232.5 | hKv 1.3 |
| Gene (Homo sapiens) | KCNH2 | GenBank | AJ512214.1 | hKv 11.1 (hERG1b) |
| Gene (Homo sapiens) | KCNJ1 | GenBank | NM_000220.4 | hKir 1.1 (ROMK) |
| Gene (Mus musculus) | KCNJ2 | GenBank | NM_008425.4 | mKir 2.1 |
| Gene (Rattus norvegicus) | Cacna1c | GenBank | M67515.1 | Cavα 1C |
| Gene (Rattus norvegicus) | Cacnb1 | GenBank | NM_017346.1 | Cav β1 |
| Gene (Rattus norvegicus) | Cacna2d1 | GenBank | AF286488 | Cav α2δ1 |
| Gene (synthetic) | EYFP | Addgene | https://www.addgene.org/vector-database/2688/ | pEYFP |
| Strain, strain background (Escherichia coli) | DH5α | NEB | Cat. #: C2987 | Chemically competent cells |
| Strain, strain background (Mus musculus) | CD1/CHR2 Mice (1 female, 7 male) | IMSR | RRID:IMSR_CRL:022 | age 5–6 months |
| Cell line (Homo sapiens) | HEK293 | Sigma-Aldrich/Cellosaurus | RRID:CVCL_0045 | |
| Cell line (Cricetulus griseus) | CHO-K1 | Sigma-Aldrich/Cellosaurus | RRID:CVCL_0214 | |
| Peptide, recombinant protein | Iberiotoxin | Alomone Labs (Jerusalem, Israel) | Cat. #: STI-400 | |
| Commercial assay or kit | Lipofectamine 2000 | Invitrogen, Thermo Fisher, Schwerte, Germany | Cat. #: 11668019 | |
| Commercial assay or kit | FuGENE | Promega, Walldorf, Germany | Cat. #: E2311 | |
| Commercial assay or kit | Venor GEM OneStep | Minerva Biolabs | Cat. #: 11-8025 | Mycoplasm PCR test |
| Chemical compound, drug | α-Mangostin | Merck (Darmstadt, Germany) | Cat. #: M3824 | |
| Chemical compound, drug | γ-Mangostin | Merck (Darmstadt, Germany) | Cat. #: M6824 | |
| Chemical compound, drug | Mangosteen antioxidant support | https://Swanson.com | Cat. #: SWH259 | |
| Chemical compound, drug | BC-5 | MP Biomedicals (Irvine, USA) | Arg-4-methoxy-2-naphthylamine | |
| Chemical compound, drug | TPA | Merck (Darmstadt, Germany) | Cat. #: 241970 | |
| Chemical compound, drug | THexA | Merck (Darmstadt, Germany) | Cat. #: 263834 | |
| Chemical compound, drug | GoSlo-SR-5-6 | Courtesy of Mark Hollywood (Dundalk Institute of Technology) | Sodium 1-amino-4-((3trifluoromethylphenyl)amino)-9,10-dioxo-9,10-dihydroanthracene-2-sulfonate | |
| Software, algorithm | Patchmaster v2.78 | HEKA Elektronik (Lambrecht, Germany) | RRID:SCR_000034 | |
| Software, algorithm | Fitmaster v2.92 | HEKA Elektronik (Lambrecht, Germany) | RRID:SCR_016233 | |
| Software, algorithm | PyMOL | Schrodinger LLC, 2015 | RRID:SCR_000305 | |
| Software, algorithm | AutoDock4 v4.2.6/MGL Tools v1.5.7 | Morris et al., 2009 | RRID:SCR_012746 | |
| Software, algorithm | LabChart | AD Instruments, Mannheim, Germany | RRID:SCR_018833 | |
| Software, algorithm | IgorPro | WaveMetrics, Portland, USA | RRID:SCR_000325 | |
| Software, algorithm | pClamp Clampfit v11.2.2.17 | Molecular Devices, San Jose, USA | RRID:SCR_011323 | |
| Other | EPC10 | HEKA Elektronik (Lambrecht, Germany) | RRID:SCR_018399 |
Table 1
Channels and subunits used in this study.
| Name | Organism | Gene name | Genbank Acc. No. |
|---|---|---|---|
| hTREK-1b | Homo sapiens | KCNK2 | NM_001017425.2 |
| hTALK-1 | Homo sapiens | KCNK16 | NM_032115.3 |
| hTRESK | Homo sapiens | KCNK18 | NM_181840.1 |
| hTWIK-1m | Homo sapiens | KCNK1 | NM_002245.3 |
| hTWIK-2m | Homo sapiens | KCNK6 | NM_004823.1 |
| hTASK-3 | Homo sapiens | KCNK9 | NM_001282534.1 |
| hTHIK-1 | Homo sapiens | KCNK13 | NM_022054.3 |
| mBKα (Slo1.1)* | Mus musculus | Kcnma1 | NM_001014797.3 |
| mβ1 | Mus musculus | Kcnmb1 | NM_031169.4 |
| hKv 1.1 | Homo sapiens | KCNA1 | NM_000217.3 |
| hKv 1.3 | Homo sapiens | KCNA3 | NM_002232.5 |
| hKv 11.1 (hERG1b) | Homo sapiens | KCNH2 | AJ512214.1 |
| hKir 1.1 (ROMK) | Homo sapiens | KCNJ1 | NM_000220.4 |
| mKir 2.1 | Mus musculus | KCNJ2 | NM_008425.4 |
| Cav 1.2: | |||
| Cavα 1C | Rattus norvegicus | Cacna1c | M67515.1 |
| Cav β1 | Rattus norvegicus | Cacnb1 | NM_017346.1 |
| Cav α2δ1 | Rattus norvegicus | Cacna2d1 | AF286488 |
-
*
G/S-rich N-terminus removed, corresponding to residues 65–1236 of the native channel.
Author response table 1
Number of interactions of S6 residues in 20 analyzed α-Mangostin poses in the molecular dockings to the Ca2+-free and Ca2.
| Residue | Ca^(2+)-free structure (6v3g) | Ca^(2+)-bound structure (6v38) |
|---|---|---|
| 1308 | 17/20 poses | 0/20 poses |
| L312 | 20/20 | 20/20 |
| F315 | 20/20 | 19/20 |
| A316 | 5/20 | 19/20 |
| S317 | 0/20 | 0/20 |
| Y318 | 0/20 | 0/20 |
Author response table 2
Comparison of the V½ ± SEM and ΔV½ ± SEM before and after activation by 10 µM α-Mangostin or 10 µM 3-hydroxyxanthone in BKα channels.
Unpaired t-test, two-tailed P values (α=0.05)
| V_(1//2)(mV) basal | V_(1//2)(mV) activated | DeltaV_(1//2)(mV) | P-value | |
|---|---|---|---|---|
| alpha-Mangostin | 110.49+-2.69 | 57.37+-3.60 | 53.08+-4.9 | < 0.001 |
| 3-hydroxyxanthone | 111.82+-3.93 | 96.83+-1.01 | 14.99+-5.67 |
Author response table 3
Summary of the V½ before and after Mangostin activation and the resulting shifts in V½ for the double mutants compared to the single mutants shown in the main manuscript.
| V_((1)/(2))(mV) | V_(1//2)(mV) in 10muMalpha-Mangostin | DeltaV_((1)/(2))(mV) | |
|---|---|---|---|
| 1308A | 54.45+-3.86 | 34.48+-3.81 | 19.97+-3.12 |
| L312M | 126.29+-6.19 | 98.39+-4.46 | 27.89+-5.42 |
| A316G | 36.7+-2.96 | 13.13+-3.11 | 23.57+-2.05 |
| 1308A/L312M | 60.57+-3.77 | 47.88+-0.77 | 12.7+-3.37 |
| I308A/A316G | 30.98+-1.94 | 14.47+-2.83 | 16.50+-1.48 |
| L312M/A316G | 51.55+-1.99 | 22.69+-5.81 | 28.86+-6.65 |
Additional files
-
MDAR checklist
- https://cdn.elifesciences.org/articles/109479/elife-109479-mdarchecklist1-v1.docx
-
Source data 1
Summary of data used in figures and applied statistical tests with p values.
- https://cdn.elifesciences.org/articles/109479/elife-109479-data1-v1.docx
Download links
A two-part list of links to download the article, or parts of the article, in various formats.
Downloads (link to download the article as PDF)
Open citations (links to open the citations from this article in various online reference manager services)
Cite this article (links to download the citations from this article in formats compatible with various reference manager tools)
Natural xanthones as α-Mangostin induce vasorelaxation involving key gating residues in the S6 domain of BK channels
eLife 14:RP109479.
https://doi.org/10.7554/eLife.109479.3
