Fig. 4: Intracellular K⁺ release from AISa-MGs through THIK-1 promotes AP firing of the associated PNs.

a Schematic of the experimental procedure. b Representative image of a dual whole-cell recording of one MG-PN pair that interacted physically at the AIS. Scale bar, 10 μm. c AP numbers of PNs (+K⁺: n = 8 pairs from 3 mice vs –K⁺: n = 10 pairs from 4 mice; two-sided two-way repeated-measures ANOVA). Insets are sample traces of paired AISa-MGs and PNs with the AISa-MG recorded using +K⁺ or –K⁺ internal solution. d Rheobase values of AP firing in PNs (+K⁺: 46.25 pA ± 5.65 pA vs –K⁺: 85.00 pA ± 6.01 pA; two-sided unpaired t-test). e Voltage thresholds of AP firing in PNs (+K⁺: –47.54 mV ± 1.40 mV vs –K⁺: –46.13 pA ± 1.65 pA; two-sided unpaired t-test). f Representative image of a whole-cell recording of one PN with 5 mM K⁺ puffed onto its AIS. Scale bar, 10 μm. g Sample trace showing the MP change of the recorded PN. h Sample traces of AP firing of the recorded PN in g. i Peak MP change of PNs (K⁺: –61.00 mV ± 1.94 mV vs Ctrl: –66.31 mV ± 1.85 mV, n = 11 cells from 4 mice; two-sided paired t-test). j AP numbers of PNs (K⁺: 9.73 ± 0.47 vs Ctrl: 6.91 ± 0.34, n = 11 cells from 4 mice; two-sided paired t-test). k mRNA copy numbers of the THIK-1 coding gene Kcnk13 in AISa-MGs vs AISn-MGs (AISa-MG: 258.10 ± 42.86, n = 8 mice vs AISn-MG: 83.57 ± 24.53, n = 7 mice; two-sided unpaired t-test). l Confocal images showing THIK-1 expression on the microglial primary process that interacted with the neuronal AIS. Scale bar, 1 μm. m Quantification of THIK-1 protein expression on the processes of AISa-MGs vs AISn-MGs (AISa-MG: 0.30 ± 0.02, n = 37 cells vs AISn-MG: 0.21 ± 0.02, n = 45 cells from 5 mice; two-sided Mann–Whitney test). n Representative confocal images showing the physical interactions between AISs and MGs, with or without C101248 application; Scale bar, 10 μm. o Percentages of MGs interacting with neuronal AISs (C101248: 16.40% ± 1.31%, n = 4 mice vs Ctrl: 17.60% ± 2.20%, n = 4 mice; two-sided unpaired t-test). p Representative confocal images showing physical interactions between AISs and MGs in WT vs THIK-1 cKO mice. Scale bar, 10 μm. q Percentages of MGs interacting with neuronal AISs (cKO: 18.30% ± 1.29%, n = 4 mice vs WT: 17.70% ± 1.96%, n = 4 mice; two-sided Mann–Whitney test). r Representative image of a dual whole-cell recording of one MG-PN pair that interacted physically at the AIS. Scale bar, 10 μm. s Quantification of microglial MPs (C101248: –17.48 mV ± 1.26 mV, n = 17 cells from 4 mice vs Ctrl: –31.78 mV ± 1.76 mV, n = 16 cells from 4 mice; two-sided unpaired t-test). t AP numbers of PNs (Ctrl, n = 16 pairs from 4 mice vs C101248, n = 17 pairs from 4 mice; two-sided two-way repeated-measures ANOVA). Insets are sample traces of dual recordings. u Rheobase values of AP firing in PNs (C101248: 84.12 pA ± 2.98 pA, n = 17 cells from 4 mice vs Ctrl: 70.00 pA ± 4.56 pA, n = 16 cells from 4 mice; two-sided unpaired t-test). v Voltage thresholds of AP firing in PNs (C101248: –48.38 mV ± 0.74 mV, n = 17 cells from 4 mice vs Ctrl: –49.33 mV ± 0.90 mV, n = 16 cells from 4 mice; two-sided unpaired t-test). w Representative images of a dual whole-cell recording of one MG-PN pair from a THIK-1 cKO mouse. Scale bar, 10 μm. x Quantification of microglial MP (cKO: –18.06 mV ± 1.77 mV, n = 17 cells from 4 mice vs WT: –33.25 mV ± 1.25 mV, n = 12 cells from 4 mice; two-sided unpaired t-test). y AP numbers of PNs (n = 18 pairs from 4 mice in both cKO and WT groups; two-sided two-way repeated-measures ANOVA). Insets are sample traces of dual recordings. z Rheobase values of AP firing in PNs (cKO: 97.78 pA ± 3.67 pA vs WT: 67.78 pA ± 3.67 pA, n = 18 cells from 4 mice; two-sided Mann–Whitney test). aa Voltage thresholds of AP firing in PNs (cKO: –48.81 mV ± 1.10 mV vs WT: –50.37 mV ± 0.78 mV, n = 18 cells from 4 mice; two-sided unpaired t-test). Error bars indicate SEM. Whiskers of box plots extend to the 2.5th and 97.5th percentiles of the data.