Fig. 4: VGLUT1_PM functions as H⁺-substrate exchanger with variable stoichiometry.

a, c Current–voltage relationships of glutamate (a) or aspartate (c) currents for various internal pHs after correction for background currents. The external solution contained 40 mM Cl⁻ and 100 mM glutamate (a)/gluconate (c) and was titrated to external pH 5.5. Small symbols show background currents (transbg) from cells expressing H191K-H426K-D428Q VGLUT1_PM under the same conditions. All data are given as means ± CI (n = 10 cells). b Changes of glutamate (filled circle, with an external solution containing 40 mM Cl⁻ and 100 mM gluconate (n = 8, 7, 29, 4, 10 cells), open triangle, with an external solution containing 40 mM Cl⁻ and 100 mM glutamate (n = 10, 10, 11 cells) and aspartate, filled square, with external solution containing 40 mM Cl⁻ and 100 mM gluconate (n = 8, 5, 22, 6, 9 cells) current reversal potentials with the Nernst potential for H⁺ calculated as E_H = \(\frac{{RT}}{F}{{{{{\mathrm{ln}}}}}}\frac{[{H}^{+}]_{o}}{[{H}^{+}]_{i}}\). The external solution was titrated to external pH 5.5. Box plots show single data as symbols, mean values as dashed lines, medians as solid line, upper and lower quartiles as box borders, and 95% CI as whiskers. Solid and dashed lines represent weighted fitting to linear functions. d E_revs as functions of relative Cl⁻ currents calculated as ratios of current amplitude at +60 mV by −140 mV. Closed symbols denote cells dialyzed with glutamate-based internal solution at pH_i = 6.5, open symbols at pH_i = 8.5. Solid lines represent the bootstrapped mean of non-parametric fits, and shaded area show bootstrapped 95% confidence intervals for those fits. Non-overlapping confidence intervals indicate a s_ignificant d_ifference between pH_i = 6.5 and pH_i = 8.5 curves. e E_revs as functions of relative Cl⁻ currents calculated as ratios of current amplitude at +60 mV by −140 mV. Closed black symbols denote cells dialyzed with aspartate-based internal solutions, open symbols with glutamate based solutions, all experiments were performed at pH_i = 8.5. At 40 mM [Cl⁻]_o, aspartate reversal potentials were negative to glutamate reversal potentials (p = 1.4 × 10⁻⁸, two^-tailed t-test), while relative Cl⁻ currents were not different. Solid lines represent the bootstrapped means of non-parametric fits, and shaded area show bootstrapped 95% confidence intervals for those fits. Non-overlapping confidence intervals indicate a significant difference between aspartate and glutamate curves. Source data are provided in the Source Data file.