Mizutani N , Kawanabe A , Jinno Y , Narita H , Yonezawa T , Nakagawa A , Okamura Y .

JP15H05901 Ministry of Education, Culture, Sports, Science and Technology (MEXT), JP19K06585 MEXT | Japan Society for the Promotion of Science (JSPS), JP19H03401 MEXT | Japan Society for the Promotion of Science (JSPS), JP21H02444 MEXT | Japan Society for the Promotion of Science (JSPS), JPMJCR14M3 MEXT | JST | Core Research for Evolutional Science and Technology (CREST), JP20H05791 Ministry of Education, Culture, Sports, Science and Technology (MEXT)

             potassium channel activity

	Fig. 1. Effect of the hydrophobicity at I233 and F234 within the C-terminal end of S4 on the phosphatase activity of Ci-VSP. (A) Top: Schematic of VSP and PTEN. Blue, VSD; yellow, VSD-PD linker; magenta, hydrophobic spine (HS); cyan, HCX5R motif in the PD (gray); orange, gating loop; green, C2 domain; and purple, disordered C-terminal tail (C-tail). Bottom: S4 sequence alignment among VSP orthologs. Dr; Danio rerio, Hs; Homo sapiens. R1 to R4 denote positively charged residues. Magenta, conserved isoleucine and phenylalanine. (B) Representative GIRK2d channel traces in oocytes expressing WT, I233Q, and F234Y Ci-VSP. The pulse protocol (Left) composed of a 50-ms ramp, 50-ms test step, and 300-ms depolarization was repeated 21 times, and all traces recorded during the test pulse (surrounded by a dotted square in the Left panel) are superimposed. Dotted lines indicate zero current level. (C) Time course of GIRK2d current in oocytes expressing WT, I233Q, and F234Y mutants. The current amplitudes at the end of the test pulse were normalized to that in the first trace, and mean normalized amplitudes were plotted against time. Data are shown as mean ± SD; n = 22, 8, and 7 for WT, I233Q, and F234Y, respectively. The rate constant of the current decay determined by fitting it with a single exponential function was taken as the phosphatase activity (see also SI Appendix, Fig. S1). (D) Representative TMRM fluorescence traces of G214C*, I233Q/G214C*, and F234Y/G214C* in response to depolarizing pulses (Top) to potentials indicated by colors. (E) Top: F-V curves for G214C*, I233Q/G214C*, and F234Y/G214C*; mean ± SD (n = 13, 7, and 8, respectively). Bottom: Phosphatase activity of WT, I233Q, and F234Y mutants at different potentials; mean ± SD (n ≥ 5 at each potential). The fluorescence values at the end of depolarization were normalized to that at 200 mV. Curves are Boltzmann fits, and values of (V1/2, z) are (69.3 ± 0.7 mV, 1.0), (103.7 ± 0.6 mV, 1.0), and (80.5 ± 1.1 mV, 0.7) for G214C*, I233Q/G214C*, and F234Y/G214C*, respectively. Dotted arrows in the Bottom panel indicate the phosphatase activity at V1/2 of the F-V curve. (F) Plots for phosphatase activities at 50 mV against the hydrophobicity of the amino acid side chain at I233 (Left) and F234 (Right). Data are shown as mean ± SD; n ≥ 6. An asterisk indicates the original amino acid residue, and letters indicate the substituted amino acid. Black dotted line in the Left panel represents phosphatase activity equal to 0.01. Red dotted lines are regression lines. R2, coefficient of determination.
	Fig. 2. Fluorometric analysis of voltage-driven conformational changes in S4 and the CCR with mutations at the C-terminal end of S4. (A) Representative TMRM fluorescence traces of Q208C* (cartoon Inset), I233Q/Q208C*, and F234Y/Q208C* in response to depolarizing pulses (Top) to potentials indicated by colors. (B) F-V curves for Q208C* (Left), I233Q/Q208C* (Center), and F234Y/Q208C* (Right). The fluorescence values measured at arrows in A (average values between 200 ms and 210 ms) were normalized to that at 0 mV. Data are shown as mean ± SD; n = 10, 6, and 8 for Q208C*, I233Q/Q208C*, and F234Y/Q208C*, respectively. (C) Representative fluorescence traces of K555Anap (cartoon Inset), I233Q/K555Anap, and F234Y/K555Anap during depolarization (Top) to potentials indicated by colors. The fluorescence was detected using a 460- to 510-nm band-pass emission filter (Em). (D) F-V curves for K555Anap (Left), I233Q/K555Anap (Center), and F234Y/K555Anap (Right). The fluorescence values before depolarization were subtracted from those at the beginning of depolarization (ΔF), and then ΔF was normalized to that at −60 mV. Data are shown as mean ± SEM; n = 3, 10, and 12 for K555Anap, I233Q/K555Anap, and F234Y/K555Anap, respectively.
	Fig. 3. Analysis of voltage-driven changes in the fluorescence of Anap introduced at I233, F234, and H237 and the helical nature of the VSD-PD linker. (A, C, and E) Representative fluorescence traces of I233Anap (A), F234Anap (C), and H237Anap (E) with or without W182A mutation during depolarization (Top) to potentials indicated by colors. The fluorescence was detected using a 460- to 510-nm band-pass emission filter (Em). W182A mutation suppressed fluorescence decrease during membrane depolarization. (B, D, and F) F-V curves for I233Anap (B), F234Anap (D), and H237Anap (F). Red lines show the plots from W182A mutants. The fluorescence values before depolarization were subtracted from those at the beginning of depolarization (ΔF), and then ΔF was normalized to that at −60 mV. Data are shown as mean ± SEM; n = 7 for both I233Anap and W182A/I233Anap, n = 6 for both F234Anap and W182A/F234Anap, n = 4 for H237Anap, and n = 3 for W182A/H237Anap. (G) Representative Anap emission spectra at −60 mV (black) and 100 mV (green) for H237Anap Ci-VSP fused with a C-terminal mCherry (cartoon is shown in Inset). Spectra were recorded from the same oocyte and normalized to the spectrum at −60 mV. (H) Anap emission peak values of H237Anap at four indicated membrane potential levels. Thin lines, individual oocytes (n = 5); thick line, mean ± SD across oocytes. *P < 0.05, Tukey-Kramer test. (I–K) Summary of phosphatase activities at 50 mV for WT and mutants with amino acid insertion (I and J) or deletion (K). Data are shown as mean ± SD; n ≥ 3. Phosphatase activity gradually decreased in accordance with an increase in the number of inserted alanines and deleted residues; however, it partially recovered at A3, A7, and A11 (red arrows in I) and del3 (red arrow in K). Glycine inserted mutants did not show such a recovery (see also SI Appendix, Fig. S14).
	Fig. 4. Interaction of the C-terminal end of S4 with the hydrophobic spine revealed by Anap fluorescence. (A and C) Representative Anap fluorescence traces of W182A/L284W/I233Anap (A) and W182A/L284W/F234Anap (C) in response to depolarizing steps (Top) to indicated potential levels. The fluorescence was detected using a 460- to 510-nm band-pass emission filter (Em). (A, Inset) Schematics of Ci-VSP with Anap incorporated in I233 or F234. W182 on S3 and L284 in the hydrophobic spine were mutated to Ala and Trp, respectively. (B and D) F-V curves for W182A/L284W/I233Anap (B) and W182A/L284W/F234Anap (D). The fluorescence values before depolarization were subtracted from those at the beginning of depolarization (ΔF), and then ΔF was normalized to that at −60 mV. Data are shown as mean ± SEM; n = 7 and 9 for W182A/L284W/I233Anap and W182A/L284W/F234Anap, respectively. (E) Representative Anap emission spectra at −60 mV (black), 50 mV (blue), 100 mV (green), and 160 mV (red) for W182A/L284W/I233Anap. The spectra at −60, 50, 100, and 160 mV were recorded from the same oocyte and normalized to the spectrum at −60 mV. Lower: Panels are magnification of the spectra shown in Upper panels. Curves are Gaussian fits. (F) Changes in Anap emission peak values of W182A/L284W/I233Anap in response to membrane depolarization. Thin lines, individual oocytes (n = 6); thick line, mean ± SD across oocytes. **P < 0.01, Tukey-Kramer test.
	Fig. 5. Voltage-dependent disulfide locking with a pair of cysteines in the C-terminal end of S4 and the hydrophobic spine. (A) Summary of a decrease in QOFF by a disulfide locking with a pair of cysteines in I233-H237 and L284 during a repetitive sensing current measurement under TEVC. QOFF, first means QOFF at 0 s, and QOFF, last means QOFF at 210 s except for H237C/L284C. QOFF, last in H237C/L284C and DTT-pretreated H237C/L284C was defined as QOFF at the last point and at 435 s, respectively. Bars are means ± SD; n = 12, 11, 11, 9, 13, and 9 for I233C/L284C, F234C/L284C, Y235C/L284C, S236C/L284C, H237C/L284C, and H237C/L284C (DTT), respectively; *P < 0.05, ***P < 0.001, ****P < 0.0001, Tukey-Kramer test. (B and D) Representative TMRM fluorescence traces of G214C* (B) and Q208C* (D) with or without H237C/L284C mutation in response to step to 150 mV. The traces recorded before (black, pre) and after (red, post) a repetitive sensing current measurement were superimposed. For comparison, the trace of H237C/L284C/Q208C* recorded at 100 mV is shown (gray). Six and seven oocytes from more than two batches were recorded for Q208C* and H237C/L284C/Q208C*, respectively, and reproducible results were obtained. (C) Summary of G214C* fluorescence change after a repetitive sensing current measurement. Bars are means ± SD; n = 6, 3, 6, and 7 for G214C*, H237C/G214C*, L284C/G214C*, and H237C/L284C/G214C*, respectively; ****P < 0.0001, Tukey-Kramer test. (E) Representative traces of WT and H237C/L284C recorded in the on-cell patch configuration. Sensing currents were evoked repeatedly (Top), and timing of recording is shown at each trace. Off sensing currents of H237C/L284C are indicated by black arrows. (F) On (Left) and off (Right) sensing currents of WT and H237C/L284C recorded at 0 s (black) and 135 s (red, WT) or 120 s (red, H237C/L284C) were superimposed at an enlarged scale. In E and F, dotted lines indicate zero current level. (G) The time course of normalized QOFF of eight oocytes expressing H237C/L284C. Red line indicates the result shown in E. In all oocytes, decrease in QOFF started within 90 s.
	Fig. 6. Structural and proposed models of VSD-CCR coupling mechanism in Ci-VSP. (A) Ribbon diagrams of the full-length Ci-VSP model. The bold gray horizontal line represents the cytoplasmic side of the plasma membrane. The S4-linker helix is colored blue. Side chains are shown as sticks for the positive charges on S4: R217 (R0), R223 (R1), R226 (R2), R229 (R3), and R232 (R4) (blue); for the hydrophobic gasket: F161 (light blue); for I233, F234, and H237 (orange); for W182 (magenta); and for the hydrophobic spine: L284 and F285 (red). Cα atom of the active site (C363) is shown as a sphere (green). Inset: Close-up top view of the region surrounded with a black box. (B) Schematic model for coupling in Ci-VSP. S3, S4, and the initial part of the PD are shown for clarity. The bold gray horizontal line indicates the cytoplasmic side of the plasma membrane. Dotted circles indicate that these four residues might be located behind S4 from this viewpoint. (C) Superposition of structural models of human PTEN (hPTEN, cyan) and Ci-VSP (orange) predicted using ColabFold (see Materials and Methods). The hydrophobic spine of both PTEN (V45 and Y46) and VSP (L284 and F285) is shown as a stick and colored green and red, respectively. Insets: Close-up views of the hydrophobic spine, the N-terminal helix of PTEN, and the VSD-PD linker helix surrounded with a black box.
	Fig. S1. Phosphatase activity of Ci-VSP assessed by coexpressing with GIRK2d in Xenopus oocyte. (A, Upper) Pulse protocol (from Fig. 1B). A hyperpolarizing test step pulse (50 ms to -120 mV) was for quantification of PI(4,5)P2-sensitive GIRK2d activity. The holding potential was -60 mV. (A, Lower) Representative GIRK2d current traces with activated WT Ci-VSP in an oocyte. All 21 traces are superimposed. Dotted line indicates zero current level. Since repeated activation of Ci-VSP by membrane depolarization to 50 mV for 300 ms decreases the plasma membrane PI(4,5)P2 which is required for GIRK2d activity, an inward current evoked by the test step pulse showed a rapid decrease. (B) Depolarization-induced decrease of GIRK2d current amplitudes measured at the end of the test pulse (arrow in A) was plotted against time. The current amplitudes were normalized to that in the first trace. The rate constant determined by fitting of the normalized current decay with a single exponential function (red curve) was taken as the phosphatase activity of Ci-VSP. In this case, the phosphatase activity at 50 mV was 4.34 s-1. (C) Representative GIRK2d current traces recorded during the test pulse from the same oocyte coexpressig I233Q Ci-VSP. The membrane was depolarized to 50 mV, 100 mV, and 150 mV for 300 ms repeatedly. The first and last traces are colored blue and red, respectively. The current traces at 50 mV are the same as shown in Figure 1B. These indicate that I233Q shows phosphatase activity at 150 mV.
	Fig. S2. Electrophysiological properties of mutants at I233 and F234 residues. (A) Depolarization-induced decrease of current amplitudes of GIRK2d recorded from oocytes expressing I233 (left) and F234 (right) mutants of Ci-VSP. The pulse protocol was the same as shown in Figure 1B. The current amplitudes were normalized to that in the first trace. Phosphatase activities plotted in Figure 1F were calculated by fitting the current changes with a single exponential function. (B) Phosphatase activities at 50 mV are plotted as a function of the side chain van der Waals (vdW) volume (1) at I233 (left) and F234 (right). Blue dotted lines indicate the linear relationship between the log of the phosphatase activity and the side chain vdW volume. R2 , coefficient of determination. Phosphatase activity has no relationship to the side chain vdW volume both in I233 and F234. (C) Representative sensing current traces of WT Ci-VSP evoked by depolarizing steps are shown in inset. On and off sensing currents are derived from the upward and downward movements of positively charged amino acids within S4 upon membrane depolarization and repolarization, respectively. Superimposed traces are shown every 30-mV increment from -80 mV to 130 mV. Scale bar: 50 ms, 1 μA. Moving charges (Q) can be calculated 4 by integration of the on or off sensing currents (QON or QOFF, respectively). QOFF-voltage (QOFF-V) curves for I233 (left) and F234 (right) mutants of Ci-VSP. Curves are Boltzmann fits (see Table S1 and Materials and Methods). (D) Phosphatase activities at 50 mV are plotted as a function of the V1/2 values calculated from each QOFF-V curve [V1/2 (QOFF)] for I233 (left) and F234 (right) mutants (Table S1); n ≥ 6 for QOFF-V curve of all I233 and F234 mutants. (E) Phosphatase activities of F234 mutants shown on an enlarged scale of the V1/2 (QOFF) indicated by the double arrow in D. In B, D, and E, an asterisk indicates the original amino acid residue. A shift of the V1/2 (QOFF) was greater in I233 mutants than F234 mutants when a hydrophilic amino acid was introduced (Table S1, e.g., V1/2 (QOFF) of I233Q = 174.1 mV and V1/2 (QOFF) of F234Q = 100.4 mV).
	Fig. S3. Phosphatase activity of I233 mutants. (A) QOFF-V curve (from Fig. S2C) and voltage-dependent phosphatase activity of I233 mutants (gray, WT; color, mutants); n ≥ 3 for phosphatase activities of all the mutants at each membrane potential except for I233V at 125 mV (n = 2), I233V at 150 mV (n = 1), and I233S at 175 mV (n = 2). Phosphatase activities were measured from individual cell at from 0 mV to 175 mV in 25-mV increment. (B and C) Phosphatase activities at a membrane potential where a shift of the V1/2 of the QOFF-V curve is considered are plotted as a function of (B) the hydrophobicity of the amino acid side chain using the Kyte-Doolittle hydrophobicity scale (2) and (C) the side chain vdW volume (1). The adjusted membrane potential was calculated as follows: 50 + ΔV1/2 (QOFF), where ΔV1/2 (QOFF) is a difference of the V1/2 of the QOFF-V curve between WT and a mutant, and was calculated with the equation: ΔV1/2 (QOFF) = [V1/2 (QOFF) of a mutant] – [V1/2 (QOFF) of WT]. Within phosphatase activities at from 0 mV to 175 mV in 25-mV increments, those at a membrane potential [approx. 50 + ΔV1/2 (QOFF)] closest to the adjusted one were plotted in B and C. An asterisk indicates the original amino acid residue. Red and blue dotted lines indicate the linear relationship between the log of the phosphatase activity and the hydrophobicity of the side chain (B), and the side chain vdW 6 volume (C), respectively. R2 value is stated in each graph. Even if the effect of the shift of the V1/2 (QOFF) was removed, the phosphatase activity still correlated with the hydrophobicity of the side chain at I233, but not with the side chain vdW volume. (D) Differences between the selected membrane potential [approx. 50 + ΔV1/2 (QOFF)] and the adjusted one [50 + ΔV1/2 (QOFF)]. ΔV < 0 means that the approx. 50 + ΔV1/2 (QOFF) is lower than the 50 + ΔV1/2 (QOFF).
	Fig. S4. Phosphatase activity of F234 mutants. (A) QOFF-V curve (from Fig. S2C) and voltage-dependent phosphatase activity of F234 mutants (gray, WT; color, mutants); n ≥ 3 for phosphatase activities of all the mutants at each membrane potential except for F234I at 175 mV (n = 1), F234V at 175 mV (n = 2), F234M at 150 mV (n = 2), F234W at 125 mV (n = 2), and F234Q at 150 mV (n = 1). Phosphatase activities were measured from individual cell at from 0 mV to 175 mV in 25-mV increment. (B and C) Phosphatase activities at a membrane potential where a shift of the V1/2 of the QOFF-V curve is considered are plotted as 8 a function of (B) the hydrophobicity of the amino acid side chain using the Kyte-Doolittle hydrophobicity scale (2) and (C) the side chain vdW volume (1). The membrane potential was selected in the same way as in Figure S3B and C. An asterisk indicates the original amino acid residue. Red and blue dotted lines indicate the linear relationship between the log of the phosphatase activity and the hydrophobicity of the side chain (B), and the side chain vdW volume (C), respectively. R2 value is stated in each graph. The phosphatase activity correlated with the hydrophobicity of the side chain at F234, but not with the side chain vdW volume. (D) Differences between the selected membrane potential [approx. 50 + ΔV1/2 (QOFF)] and the membrane potential adjusted by the shift of the V1/2 of the QOFF-V curve [50 + ΔV1/2 (QOFF)].
	Fig. S5. Effects of hydrophilic mutations in S4 residues upstream from R4 on the phosphatase activity. (A) Amino acid sequence of S4 and the proximal part of the VSD-PD linker of Ci-VSP. R1 to R4 denote the first to the fourth arginine on S4, respectively. I233 and F234 are highlighted in magenta. L225, V228, L230, and A231 are pointed by arrows. (B) Phosphatase activities at 50 mV are plotted as a function of the hydrophobicity of the amino acid side chain at positions L225, V228, L230, and A231 using the Kyte-Doolittle hydrophobicity scale (2); n ≥ 6 for both L225 and L230 mutants, n ≥ 7 for both V228 and A231 mutants. (C) Phosphatase activities at 50 mV are plotted as a function 10 of the V1/2 values calculated from each QOFF-V curve [V1/2 (QOFF)] for L225, V228, L230, and A231 mutants (Table S1); n ≥ 7 for QOFF-V curve of both L225 and V228 mutants, n ≥ 6 for QOFF-V curve of L230 mutants, n ≥ 10 for QOFF-V curve of A231 mutants. (D) Phosphatase activities at a membrane potential where a shift of the V1/2 of the QOFF-V curve is considered are plotted as a function of the hydrophobicity of the amino acid side chain using the Kyte-Doolittle hydrophobicity scale (2); n ≥ 6 for both L225 and L230 mutants, n ≥ 7 for both V228 and A231 mutants. The membrane potential was selected in the same way as in Figures S3 and S4. In B to D, an asterisk indicates the original amino acid residue. (E) Differences between the selected membrane potential [approx. 50 + ΔV1/2 (QOFF)] and the membrane potential adjusted by the shift of the V1/2 of the QOFFV curve [50 + ΔV1/2 (QOFF)].
	Fig. S6. Effects of mutations in the initial part of the VSD-PD linker on the phosphatase activity. (A) Amino acid sequence of the C-terminal end of S4 and the proximal part of the VSD-PD linker. R4 denotes the fourth arginine on S4. I233 and F234 are highlighted in magenta. Y235, S236, and H237 are pointed by arrows. (B) Phosphatase activities at 50 mV are plotted as a function of the hydrophobicity of the amino acid side chain at Y235, S236, and H237 using the Kyte-Doolittle hydrophobicity scale (2); n ≥ 6 for Y235 mutants, n ≥ 7 for S236 mutants, n ≥ 4 for H237 mutants. Red dotted lines indicate the linear relationship between the log of the phosphatase activity and the hydrophobicity of the side chain. R2 value is stated in each graph. (C) Phosphatase activities at 50 mV are plotted as a function of the V1/2 values calculated from each QOFF-V curve [V1/2 (QOFF)] for Y235, S236, and H237 mutants (Table S1); n ≥ 9 for QOFF-V curve of Y235 mutants, n ≥ 7 for QOFF-V curve of S236 mutants, n ≥ 5 for QOFF-V curve of H237 mutants. (D) Phosphatase activities at a membrane potential where a shift of the V1/2 of the QOFF-V curve is considered are plotted as a function of the hydrophobicity of the amino acid side chain using the Kyte-Doolittle hydrophobicity scale (2); n ≥ 6 for Y235 mutants, n ≥ 7 for S236 mutants, n ≥ 3 for H237 mutants. The membrane potential was selected in the same way as in Figures S3 and S4. Red dotted lines indicate the linear relationship between the log of the phosphatase activity and the hydrophobicity of the side chain. R2 value is stated in each graph. In B to D, an asterisk indicates the original amino acid 12 residue. (E) Differences between the selected membrane potential [approx. 50 + ΔV1/2 (QOFF)] and the membrane potential adjusted by the shift of the V1/2 of the QOFF-V curve [50 + ΔV1/2 (QOFF)].
	Fig. S7. Expression levels of Ci-VSP mutants in measurements of the phosphatase activity. Plasma membrane expression levels of Ci-VSP were assessed from QOFF of sensing currents which were measured with depolarization from the holding potential of -60 mV to 150 mV from oocytes coexpressing GIRK2d with Ci-VSP in each recording corresponding to Figure 1 and Figures S2, S5, and S6, so that the number of cells is the same as that in phosphatase activity measurements. Bars are means ± SD and the black dots represent the individual cells.
	Fig. S8. Sensing currents of all Ci-VSP mutants. Representative sensing current traces of all Ci-VSP mutants recorded from Xenopus oocytes under TEVC. Traces are shown every 30-mV increment ranging from -70 mV to 140 mV or 200 mV. Scale bars except for magnifications of L225N and Y235Q mutants: 100 ms, 1 μA.
	Fig. S9. Charge-voltage relationship of K555Anap mutants. (A) QOFF-V curves for K555Anap (black, n = 3), I233Q/K555Anap (blue, n = 5), and F234Y/K555Anap (red, n = 3). Curves are Boltzmann fits (see Table S1 and Materials and Methods). (B) Plasma membrane expression levels were assessed from QOFF of sensing currents measured with depolarization from the holding potential of -60 mV to 160 mV. Bars are means ± SD and the black dots represent the individual cells (n = 3 for K555Anap, n = 9 for I233Q/K555Anap, n = 12 for F234Y/K555Anap).
	Fig. S10. Voltage-dependent fluorescence changes of Anap in I233, F234, and H237 at the 420-460 nm bandwidth. (A) Emission spectra of Anap solution with increase in a tryptophan concentration. The emission spectrum of Anap with 20 mM alanine is shown in red. Alanine lacked ability of quenching of Anap fluorescence. (B) Stern-Volmer plot of the Anap fluorescence quenching by tryptophan. The SternVolmer quenching constant (KSV) is 0.0468 mM-1. (C, E, and G) Representative fluorescence traces of I233Anap (C), F234Anap (E), and H237Anap (G) with or without W182A mutation during depolarization (top) to indicated potential levels. The fluorescence was detected using a 420-460 nm band-pass emission filter. (D, F, and H) F-V curves for I233Anap (D), F234Anap (F), and H237Anap (H). Red lines show the plots from W182A mutants. The fluorescence values before depolarization were subtracted from those at the beginning of depolarization (ΔF) and then ΔF was normalized to that at -60 mV. (I) QOFF-V curves for I233Anap (left), F234Anap (center), and H237Anap (right); n = 3 for I233Anap and W182A/H237Anap, n = 5 for W182A/I233Anap, n = 6 for F234Anap, n = 4 for W182A/F234Anap, n = 8 for H237Anap. Curves are Boltzmann fits and those of W182A mutants are shown in red (see Table S1).
	Fig. S11. Voltage-dependent fluorescence changes of Anap in H237 in response to membrane hyperpolarization. (A and C) Representative fluorescence traces of H237Anap (left) and W182A/H237Anap (right) in response to hyperpolarization (top left) to indicated potential levels. The fluorescence was detected using a 420-460 nm (A) and a 460-510 nm (C) band-pass emission filters. (B and D) F-V curves for H237Anap within 420-460 nm (B) and 460-510 nm (D) bandwidths (n = 3 for both H237Anap and W182A/H237Anap). The fluorescence values before hyperpolarization were subtracted from those at the beginning of hyperpolarization (ΔF) and then ΔF was normalized to that at -20 mV.
	Fig. S12. Shift of H237Anap emission spectrum. (A) Representative Anap fluorescence image of an oocyte expressing H237Anap Ci-VSP fused with a C-terminal mCherry (cartoon is shown in inset). A white dotted curve indicates the edge of the oocyte. Scale bar, 100 μm. (B) Representative spectral images of Anap emission from the oocyte shown in A at -60 mV and 160 mV. Images were produced by passing the emission through a slit (a black dotted line in A). Scale bar, 40 nm. (C, E, and F) Representative Anap (C and F) and mCherry (E) emission spectra at -60 mV (black, from Fig. 3G), 50 mV (blue), 100 mV (green, from Fig. 3G), 160 mV (red), -20 mV (grey), and -180 mV (purple) for H237Anap Ci-VSP fused with a Cterminal mCherry. Spectra at -60 mV and 160 mV were extracted from the spectral images shown in B. Spectra were normalized to those at -60 mV (C and E) or -20 mV (F). Both Anap and mCherry emission spectra were recorded from the same oocyte shown in A. (D and G) Gaussian fits of Anap emission spectra shown in C (D) and F (G). (H) Anap emission peak values of H237Anap in response to membrane hyperpolarization (those in response to membrane depolarization are shown in Figure 3H). Thin lines, individual oocytes (n = 4); thick line, mean ± SD across oocytes. One oocyte was not included in the hyperpolarizing result because of its damage after depolarizing recordings. **p < 0.01, two-tailed paired t-test.
	Fig. S13. Spectral measurement from oocytes under TEVC. (A) Representative emission spectra of endogenous animal pole background fluorescence of uninjected oocyte at -60 mV (black) and 160 mV (red). Spectra were normalized to that at -60 mV. Similar spectra were observed in 9 independent oocytes from 3 batches (3 oocytes from each batch). The spectra had a peak similar to that of Anap, but a shape different from that of Anap. (B) Emission peak values of A. Emission peak of endogenous animal pole background fluorescence of uninjected oocyte did not show a shift upon membrane depolarization. Thin lines, individual oocytes (n = 9); thick line, mean ± SD across oocytes. (C and E) Representative Anap (C) and mCherry (E) emission spectra at -60 mV (black), 50 mV (blue), 100 mV (green), and 160 mV (red) for WT Ci-VSP fused with a C-terminal mCherry (cartoon is shown in inset in C). Spectra were normalized to those at -60 mV. Both Anap and mCherry emission spectra were recorded from the same oocyte coinjected with Ci-VSP cRNA, pAnap plasmid, and Anap. (D) Anap emission peak values of C. Emission peak of unincorporated Anap did not show a shift upon membrane depolarization. Thin lines, individual oocytes (n = 4); thick line, mean ± SD across oocytes.
	Fig. S14. Effects of alanine and glycine insertion into the VSD-PD linker and VSD-PD linker partial deletion on the phosphatase activity. (A and C) Amino acid sequence alignment of Ci-VSP WT and alanine-inserted mutants (A) and VSD-PD linker-deleted mutants (C). One to twelve alanines were inserted between Y235 and S236, and these mutants were named A1 to A12, respectively. A3 to A11 mutants were missed. One to seven glycines were inserted in the same way (G1 to G7). In the mutants named del1 to del4, one to four amino acids were deleted from S236, respectively. (B and D) Plasma membrane expression levels were assessed from QOFF of sensing currents measured with depolarization from the holding potential of -60 mV to 150 mV from oocytes coexpressing GIRK2d with Ci-VSP in each recording. Bars are means ± SD and the black dots represent the individual cells (n ≥ 3).
	Fig. S15. Voltage-dependent fluorescence changes of I233Anap and F234Anap with W182A/L284W mutations in the 420-460 nm bandwidth. (A and C) Representative fluorescence traces of W182A/L284W/I233Anap (A) and W182A/L284W/F234Anap (C) under TEVC during depolarization (top) to indicated potential levels. The fluorescence was detected using a 420-460 nm band-pass emission filter. (B and D) F-V curves for W182A/L284W/I233Anap (B) and W182A/L284W/F234Anap (D). The fluorescence values before depolarization were subtracted from those at the beginning of depolarization (ΔF) and then ΔF was normalized to that at -60 mV. (E) QOFF-V curves for W182A/L284W/I233Anap (upper, n = 11) and W182A/L284W/F234Anap (lower, n = 11). Curves are Boltzmann fits (see Table S1).
	Fig. S16. Electrophysiological screening of voltage-dependent disulfide locking with a pair of cysteines in the C-terminal end of S4 and the hydrophobic spine. (A and B) Representative sensing current traces recorded under TEVC (A) and the time course of normalized QOFF (B) of I233C/L284C (n = 12), F234C/L284C (n = 11), Y235C/L284C (n = 11), and S236C/L284C (n = 9). Sensing currents were evoked repeatedly 15 times every 15 s by depolarization from the holding potential of -60 mV to 150 mV (top in A). The series of sensing current traces were aligned horizontally with spaces from each other and offset to zero current line. (C) Representative sensing current traces of H237C/L284C, WT, H237C, and L284C. Only H237C/L284C exhibited a decrease in the amplitude of the sensing current. Inset shows off sensing currents of H237C/L284C recorded at 0 s (black), 75 s (blue), and 150 s (red). The recording condition was the same as in A. (D and E) The time course of normalized QOFF of (D) H237C/L284C (n = 13) and (E) WT, H237C, and L284C (n = 11, 9, and 9, respectively). Red line in D indicates the result shown in C. In all oocytes expressing H237C/L284C, decrease in QOFF started within 210 s. (F and G) Representative sensing current traces (F) and the time course of normalized QOFF (G) of H237C/L284C pre-treated with 5-min incubation into 100 mM DTT (n = 9). Sensing currents were evoked repeatedly 30 times every 15 s and traces are shown every 30 s. (H) The time course of normalized QOFF of H237C/L284C. Sensing currents were evoked by depolarization from the holding potential of -60 mV to 130 mV (left) and 100 mV (right). Both panels show the same color symbols but recording was not performed from the same oocytes (n = 4 each). (I and J) Representative sensing current traces evoked by depolarizing steps from the holding potential of - 60 mV to 150 mV in 10-mV increments under TEVC (I) and QOFF-V curves of WT (black, n = 10), H237C (blue, n = 11), and L284C (red, n = 9) Ci-VSP (J). Traces are shown every 30-mV increment. Curves are Boltzmann fits (see Table S1). (K and L) Representative sensing current traces (K) and the time course of normalized QOFF (L, n = 10) of H237C/F285C. The recording condition was the same as in A. QOFF in B, D, E, G, H, and L was normalized to that at 0 s.
	Fig. S17. Analysis of single cysteine mutants as a control of voltage-dependent disulfide locking. (A and C) Representative fluorescence traces of G214C* (A) and Q208C* (C) with H237C (left) and L284C (right) in response to depolarization from the holding potential of -60 mV to 150 mV (top). Traces recorded before (black, pre) and after (red, post) a 15-times repetitive sensing current measurement were superimposed. 4 and 5 oocytes from more than 2 batches were recorded for H237C/Q208C* and L284C/Q208C*, respectively, and reproducible results were obtained. (B and D) QOFF-V curves of G214C* (B) and Q208C* (D) mutants in comparison with G214C* and Q208C*, respectively; n = 3 for H237C/G214C*, n = 5 for G214C* and L284C/G214C*, n = 4 for H237C/Q208C*, n = 6 for L284C/Q208C*, n = 8 for Q208C*. Curves are Boltzmann fits (G214C* and Q208C*, black; H237C/G214C* and H237C/Q208C*, blue; L284C/G214C* and L284C/Q208C*, red; see Table S1). (E) Representative sensing current traces of WT Ci-VSP fused with a C-terminal mCherry recorded in the on-cell patch configuration (cartoon inset). Currents were evoked by depolarizing steps from the holding potential of -60 mV to 180 mV in 30-mV increments. (F) QOFF-V curves of WT Ci-VSP fused with a C-terminal mCherry under different recording conditions in comparison with WT (grey); n = 6 for WT (TEVC), n = 14 for TEVC, n = 5 for on-cell 26 in high K+ bath solution, n = 6 for on-cell in NMDG-based bath solution. Curves are Boltzmann fits (TEVC, black; on-cell in high K+ bath solution, blue; on-cell in NMDG-based bath solution, red; see Table S1). Since there was little shift of the QOFF-V curves between the two bath solutions, we estimated resting membrane potential of oocytes as zero and assumed that the transmembrane potential was controlled by a patch electrode. (G) Representative current trace of an uninjected oocyte evoked by depolarization from the holding potential of -60 mV to 150 mV recorded in the on-cell patch configuration. (H) Representative sensing current traces of H237C and L284C recorded in the on-cell patch configuration. Sensing currents were evoked repeatedly 10 times every 15 s by depolarization from the holding potential of -60 mV to 150 mV (top) and the currents recorded at 0, 75, and 135 s are shown. (I) On (left) and off (right) sensing currents of H237C and L284C recorded at 0 s (black) and 135 s (red) were superimposed at an enlarged current scale. (J) The time course of normalized QOFF of WT (n = 4), H237C (n = 5), and L284C (n = 4). QOFF was normalized to that at 0 s. Horizontal black dotted lines in E, G, H, and I indicate zero current level.
	Fig. S18. Structural comparison between Dr-VSP and Ci-VSP. (A) Comparison of the activated state X-ray crystal structure (gray, PDB 4G7V) (3) and structural model predicted using ColabFold (4) (black, see Materials and Methods) of VSD of Ci-VSP. Cα atoms of the hydrophobic gasket and S4 arginines are shown as spheres. (B) Superposition of predicted Dr-VSP (green) and Ci-VSP (orange) using ColabFold. Cα atoms of positively charged residues of the S4-linker helix and the active site are shown as spheres. The hydrophobic gasket of Dr-VSP (F101) and Ci-VSP (F161) is shown as a stick. (C) Expanded image for the S4 helix and the hydrophobic spine. The C-terminal end of S4 and the hydrophobic spine are labeled.
	Fig. S19. Surface-exposed, hydrophobic residues surrounding the active site well conserved among various phosphoinositide phosphatases. Surface models for Ci-VSP CCR (PDB 3V0H) (5), human PTEN (PDB 1D5R) (6), Legionella pneumophila SidF (PDB 4FYG) (7), human INPP5B (PDB 4CML) (8), human OCRL (PDB 4CMN) (8), human Synaptojanin1 (Synj1, PDB 7A17) (9), and human myotubularin-related protein-2 (MTMR2, PDB 1ZVR) (10). Surface-exposed, hydrophobic residues are colored green. The active site and Mg2+ ion is shown as a yellow and magenta sphere, respectively. The co-crystallized substances bound in the active site are shown as sticks colored yellow (VSP, IP3; PTEN, L(+)- tartrate; SidF, diC4-PI(3,4)P2; INPP5B, diC8-PI(3,4)P2; OCRL, phosphate ion; Synj1, diC8- 29 PI(3,4,5)P3; MTMR2, diC4-PI(3,5)P2). Ci-VSP CCR, PTEN, SidF, INPP5B, and OCRL have been also shown in our previous paper (11).
	Fig. S20. Comparison of human PTEN and Ci-VSP structures. X-ray crystal structure of human PTEN (hPTEN, purple, PDB 7JUL) (12) is superposed on the structural model of Ci-VSP predicted using ColabFold (orange). The hydrophobic spine of both PTEN (V45 and Y46) and VSP (L284 and F285) is shown as a stick and colored green and red, respectively. Close-up view of the hydrophobic spine, the N-terminal helix of PTEN, and the VSDPD linker helix surrounded with a black box is shown in insets.

Begley, Molecular basis for substrate recognition by MTMR2, a myotubularin family phosphoinositide phosphatase. 2006, Pubmed

Begley, Molecular basis for substrate recognition by MTMR2, a myotubularin family phosphoinositide phosphatase. 2006, Pubmed
Bezanilla, The voltage sensor in voltage-dependent ion channels. 2000, Pubmed
Byron, Inhibition of activated fibroblast growth factor receptor 2 in endometrial cancer cells induces cell death despite PTEN abrogation. 2008, Pubmed
Campbell, Allosteric activation of PTEN phosphatase by phosphatidylinositol 4,5-bisphosphate. 2003, Pubmed
Careaga, Thermal motions of surface alpha-helices in the D-galactose chemosensory receptor. Detection by disulfide trapping. 1992, Pubmed
Castle, Voltage-sensing phosphatase modulation by a C2 domain. 2015, Pubmed
Chatterjee, A genetically encoded fluorescent probe in mammalian cells. 2013, Pubmed
Dan, Correlating phosphatidylinositol 3-kinase inhibitor efficacy with signaling pathway status: in silico and biological evaluations. 2010, Pubmed
Dempsey, The structural basis of PTEN regulation by multi-site phosphorylation. 2021, Pubmed
Denning, A short N-terminal sequence of PTEN controls cytoplasmic localization and is required for suppression of cell growth. 2007, Pubmed
Furukawa, Differential interactions of the C terminus and the cytoplasmic I-II loop of neuronal Ca2+ channels with G-protein alpha and beta gamma subunits. II. Evidence for direct binding. 1998, Pubmed , Xenbase
Grimm, Allosteric substrate switching in a voltage-sensing lipid phosphatase. 2016, Pubmed , Xenbase
Hobiger, Coupling of Ci-VSP modules requires a combination of structure and electrostatics within the linker. 2012, Pubmed , Xenbase
Hou, Two-stage electro-mechanical coupling of a KV channel in voltage-dependent activation. 2020, Pubmed
Hsu, Structural basis for substrate recognition by a unique Legionella phosphoinositide phosphatase. 2012, Pubmed
Inanobe, Molecular cloning and characterization of a novel splicing variant of the Kir3.2 subunit predominantly expressed in mouse testis. 1999, Pubmed , Xenbase
Iwasaki, A voltage-sensing phosphatase, Ci-VSP, which shares sequence identity with PTEN, dephosphorylates phosphatidylinositol 4,5-bisphosphate. 2008, Pubmed , Xenbase
Johnston, Catalysis by the tumor-suppressor enzymes PTEN and PTEN-L. 2015, Pubmed
Jumper, Highly accurate protein structure prediction with AlphaFold. 2021, Pubmed
Kalstrup, S4-S5 linker movement during activation and inactivation in voltage-gated K+ channels. 2018, Pubmed , Xenbase
Kawai, Polarized PtdIns(4,5)P2 distribution mediated by a voltage-sensing phosphatase (VSP) regulates sperm motility. 2019, Pubmed
Kawanabe, The hydrophobic nature of a novel membrane interface regulates the enzyme activity of a voltage-sensing phosphatase. 2018, Pubmed
Kawanabe, Engineering an enhanced voltage-sensing phosphatase. 2020, Pubmed
Kohout, Electrochemical coupling in the voltage-dependent phosphatase Ci-VSP. 2010, Pubmed
Kohout, Subunit organization and functional transitions in Ci-VSP. 2008, Pubmed
Kurokawa, 3' Phosphatase activity toward phosphatidylinositol 3,4-bisphosphate [PI(3,4)P2] by voltage-sensing phosphatase (VSP). 2012, Pubmed
Kyte, A simple method for displaying the hydropathic character of a protein. 1982, Pubmed
Lacroix, Controlling the activity of a phosphatase and tensin homolog (PTEN) by membrane potential. 2011, Pubmed , Xenbase
Lee, Genetic incorporation of a small, environmentally sensitive, fluorescent probe into proteins in Saccharomyces cerevisiae. 2009, Pubmed
Lee, Methodological improvements for fluorescence recordings in Xenopus laevis oocytes. 2019, Pubmed , Xenbase
Lee, Crystal structure of the PTEN tumor suppressor: implications for its phosphoinositide phosphatase activity and membrane association. 1999, Pubmed , Xenbase
Li, Structural mechanism of voltage-dependent gating in an isolated voltage-sensing domain. 2014, Pubmed , Xenbase
Liu, A glutamate switch controls voltage-sensitive phosphatase function. 2012, Pubmed
Long, Voltage sensor of Kv1.2: structural basis of electromechanical coupling. 2005, Pubmed
Mannuzzu, Direct physical measure of conformational rearrangement underlying potassium channel gating. 1996, Pubmed , Xenbase
Mansoor, Distance mapping in proteins using fluorescence spectroscopy: the tryptophan-induced quenching (TrIQ) method. 2010, Pubmed
Mansoor, Mapping proximity within proteins using fluorescence spectroscopy. A study of T4 lysozyme showing that tryptophan residues quench bimane fluorescence. 2002, Pubmed
Matsuda, Crystal structure of the cytoplasmic phosphatase and tensin homolog (PTEN)-like region of Ciona intestinalis voltage-sensing phosphatase provides insight into substrate specificity and redox regulation of the phosphoinositide phosphatase activity. 2011, Pubmed
Mavrantoni, A method to control phosphoinositides and to analyze PTEN function in living cells using voltage sensitive phosphatases. 2015, Pubmed
Mirdita, ColabFold: making protein folding accessible to all. 2022, Pubmed
Murata, Depolarization activates the phosphoinositide phosphatase Ci-VSP, as detected in Xenopus oocytes coexpressing sensors of PIP2. 2007, Pubmed , Xenbase
Murata, Phosphoinositide phosphatase activity coupled to an intrinsic voltage sensor. 2005, Pubmed , Xenbase
Nguyen, Engineering ePTEN, an enhanced PTEN with increased tumor suppressor activities. 2014, Pubmed
Okamura, Voltage-Sensing Phosphatases: Biophysics, Physiology, and Molecular Engineering. 2018, Pubmed
Paesmans, A structure of substrate-bound Synaptojanin1 provides new insights in its mechanism and the effect of disease mutations. 2020, Pubmed
Rahdar, A phosphorylation-dependent intramolecular interaction regulates the membrane association and activity of the tumor suppressor PTEN. 2009, Pubmed
Sakata, Coupling of the phosphatase activity of Ci-VSP to its voltage sensor activity over the entire range of voltage sensitivity. 2011, Pubmed , Xenbase
Sakata, Voltage-dependent motion of the catalytic region of voltage-sensing phosphatase monitored by a fluorescent amino acid. 2016, Pubmed , Xenbase
Sanguinetti, Mutations of the S4-S5 linker alter activation properties of HERG potassium channels expressed in Xenopus oocytes. 1999, Pubmed , Xenbase
Slesinger, The S4-S5 loop contributes to the ion-selective pore of potassium channels. 1993, Pubmed , Xenbase
Trésaugues, Structural basis for phosphoinositide substrate recognition, catalysis, and membrane interactions in human inositol polyphosphate 5-phosphatases. 2014, Pubmed
Tsou, The role of MMAC1 mutations in early-onset breast cancer: causative in association with Cowden syndrome and excluded in BRCA1-negative cases. 1997, Pubmed
Tsutsui, Engineering voltage sensing phosphatase (VSP). 2021, Pubmed
Vazquez, Phosphorylation of the PTEN tail regulates protein stability and function. 2000, Pubmed
Villalba-Galea, Coupling between the voltage-sensing and phosphatase domains of Ci-VSP. 2009, Pubmed , Xenbase
Wang, Understanding the stereospecific interactions of 3-deoxyphosphatidylinositol derivatives with the PTEN phosphatase domain. 2010, Pubmed
Wisedchaisri, Resting-State Structure and Gating Mechanism of a Voltage-Gated Sodium Channel. 2019, Pubmed