Fig. 4: Intracellular Mg²⁺ regulates synaptic configuration by inhibiting release probability.

a Time-courses of changes in synaptic configuration and \(\overline{[{{\mathrm{Mg}}}^{ 2+}]}_{ {{\mathrm{i}}}}\) (mM) in neurites after various treatments (n = 5, 8, 10, 11, 8, 8, 7, 7, 6, 7, 9, 8 repeats for configuration, 5, 8, 9, 8, 9, 7, 7, 6, 6, 6, 6, 6 repeats for Mg²⁺). Two-sided unpaired t-tests (time 4 h vs. 0 h). For ΣPr, P = 0.4068, 0.7444, 0.7391; for D, P < 0.0001, = 0.0305, 0.0003; for \(\overline{{{{{{\boldsymbol{Pr }}}}}}}\), P = 0.0084, 0.0799, 0.0044; for [Mg²⁺]_i, P = 0.0001, 0.0724, 0.0009. NS, no significance, *P < 0.05, **P < 0.01, ***P < 0.001, ****P < 0.0001. b Experimental design for measuring [Mg²⁺]_i and Pr in the same boutons. c FM4-64⁺ and MgGrn⁺ signals in the same boutons. d Measurement of Pr and [Mg²⁺]_i (mM) in boutons (n = 204, 444 from 3, 4 repeats). Inset, Pr versus 1/[Mg²⁺]_i. Double logarithmic fitting, R² = 0.54, P < 0.0001. e Measurement of \(\overline{{{{{{\boldsymbol{Pr }}}}}}}\) and \(\overline{[{{\mathrm{Mg}}}^{2+}]}_{{{\mathrm{i}}}}\) (mM) of the boutons at individual branches (n = 55, 56 branches from 4, 4 repeats). f–h Concurrent measurement of Pr and 1AP-evoked Ca²⁺ influx in the same boutons. f Left, images to show evoked Ca²⁺ influx (peak ΔF/F₀) and FM5-95⁺ puncta (masked by GCaMP⁺ axon morphology) in the same boutons. Top-right, peak ΔF/F₀, Ca²⁺ traces, FM5-95⁺ puncta, and Pr values in the same boutons. Bottom-right, percentage of boutons undergoing Ca²⁺ influx, referred to as Ca²⁺(+), and vesicle turnover (Pr ≥ 0.04) (n = 3 repeats). ΔF/F₀ puncta and Ca²⁺ traces were averaged from 30 repeats. Band over traces, SEM. g Plot of Pr against [Ca²⁺]_evoke (the amplitude of 1AP-evoked Ca²⁺ influx) in boutons (n = 243 boutons from 5 repeats, physiological Mg²⁺ condition). Inset, logarithmic plot. Line, double logarithmic fitting, R² = 0.19, P < 0.0001. h Similar to (g) but using working solution containing [Ca²⁺]_o 4.8 mM and [Mg²⁺]_o 1.2 mM (ratio 4:1) (n = 146 boutons from 3 repeats). Line, double logarithmic fitting, R² = 0.33, P < 0.0001. Data are mean ± SEM. Source data are provided as a Source Data file.