α2δ-2 is required for depolarization-induced suppression of excitation in Purkinje cells

Abstract

α2δ proteins (CACNA2D1-4) are required for normal neurological function and contribute to membrane trafficking of voltage-gated calcium channels, through which calcium entry initiates numerous physiological processes. However, it remains unclear how α2δ proteins influence calcium-mediated signalling to control neuronal output. Using whole-cell recordings of mouse Purkinje cells, we show that α2δ-2 is required for functional coupling of postsynaptic voltage-dependent calcium entry with calcium-dependent effector mechanisms controlling two different outputs, depolarization-induced suppression of excitation and spike afterhyperpolarization. Our findings indicate an important role for α2δ-2 proteins in regulating functional postsynaptic calcium channel coupling in neurons, providing new context for understanding the effects of α2δ mutations on neuronal circuit function and presenting additional potential avenues to manipulate α2δ-mediated signalling for therapeutic gain. KEY POINTS: Calcium influx, via voltage-dependent calcium channels, drives numerous neuronal signalling processes with precision achieved in part by tight coupling between calcium entry and calcium-dependent effectors. α2δ proteins are important for neurological function and contribute to calcium channel membrane trafficking, although how α2δ proteins influence postsynaptic calcium-dependent signalling is largely unexplored. Here it is shown that loss of α2δ-2 proteins disrupts functional calcium coupling to two different postsynaptic calcium-dependent signals in mouse Purkinje cell neurons, retrograde endocannabinoid signalling and the action potential afterhyperpolarization. The findings provide new insights into the control of calcium coupling as well as new roles for α2δ-2 proteins in neurons.

Keywords: calcium channels; depolarization-induced suppression of excitation; endocannabinoid signalling; nanodomains; purkinje cells; α2δ-2 proteins.

Figure 1. Depolarization-Induced Suppression of Excitation (DSE) reduces climbing fiber-EPSC amplitude in WT but not α2δ−2 KO Purkinje cells.

A) Experimental schematic. Purkinje cell (PC) is held at −70 mV while the climbing fiber axon (red) is stimulated at 0.2 Hz (arrows). A depolarization step to 0 mV is delivered to the PC between EPSC recordings. B) Overlay of peak-scaled EPSC traces during baseline (grey) and 5 s post-depolarization (dotted line) from WT (black) and KO (blue) PCs; 250 ms step duration. C) Summary of WT and KO timecourse responses to 250 ms depolarization shown as % of baseline EPSC. Each point averaged two consecutive EPSCs. WT (black, filled circle): 24.9 ± 18.5% (n = 6 cells from 3 mice; KO (blue, filled circle): 3.6 ± 2.8%; n = 4 cells from 3 mice; data are reported as the mean ± SD. D) Overlay of peak-scaled EPSCs during baseline (grey) and 5 s after 250 ms (dotted line), 500 ms (dashed line), and 1 s (solid line) depolarization from WT (black) and KO (blue) PCs. E) DSE timecourse from WT (black, open circle) and KO (blue, open circle) experiments using 1 s depolarizing step lengths shown as % of baseline EPSC. Each point averaged two consecutive EPSCs. WT_1s = 32.4 ± 15.1%, n = 12 cells (8 mice); KO_1s = 5.4 ± 4.4%, n = 10 cells (9 mice); data are reported as the mean ± SD. F) EPSC depression normalized to baseline (Magnitude of DSE) in WT (black) and KO (blue) after 250 ms (fine stripe), 500 ms (wide stripe) and 1 s (solid) depolarization steps: WT_250ms = 24.9 ± 18.5%, n = 6 cells (3 mice); WT_500ms = 20.4 ± 9.4%, n = 9 cells (4 mice); p = 0.951; WT_1s = 32.4 ± 15.1%, n = 12 cells (8 mice); p = 0.657; KO_250ms = 3.6 ± 2.8%, n = 4 cells (3 mice); p = 0.0283; KO_500ms = 3.4 ± 0.9%, n = 5 cells (5 mice); p = 0.0156; KO_1s = 5.4 ± 4.4%, n = 10 cells (9 mice); p = 0.00930; Two-way ANOVA compared to WT_250ms, Sidak’s correction for multiple comparisons; data are reported as the mean ± SD; * p < 0.05, ** p < 0.01.

Figure 2. Endocannabinoid-producing enzymes are present near climbing fiber synapses in Purkinje cells of both genotypes.

Purkinje cells from WT (top) and KO (bottom) mice express the eCB-producing enzyme, diacylglycerol lipase α (DGLα; green), along dendritic membranes (calbindin; blue) adjacent to presynaptic climbing fiber synapses (VGLUT2; magenta), indicated by arrows. Scale bar 5 μm. Representative images shown are max projection confocal images, 63×1.4 NA objective.

Figure 3. Reduced intracellular [EGTA] reveals DSE in the α2δ−2 KO.

A) Overlay of peak-scaled EPSCs during baseline (grey) and 5 s after 250 ms (dotted line), 500 ms (dashed line), and 1 s (solid line) depolarization using intracellular solution containing 2 mM EGTA from WT (black) and KO (blue) PCs. B) DSE timecourse from WT (black) and KO (blue) experiments using 2 mM EGTA and depolarization steps of 250 ms (filled circle), 500 ms (triangle) and 1 s (open circle). Data are reported as the mean ± SD. C) DSE Magnitude in WT (black) and KO (blue) after 250 ms (fine stripe), 500 ms (wide stripe) and 1 s (solid) depolarization steps using a 2 mM EGTA: WT_250ms = 16.8 ± 15.8%, 5 cells (4 mice); WT_500ms = 23.2 ± 18.1%, 7 cells (4 mice); p = 0.958; WT_1s = 53.2 ± 23.2%, n = 8 cells (5 mice); p = 0.000992; KO_250ms = 1.6 ± 0.8%, n = 6 cells (5 mice); p = 0.432; KO_500ms = 3.2 ± 2.2%, n = 5 cells (5 mice); p = 0.597; KO_1s = 12.2 ± 4.9%, n = 5 cells (5 mice); 0.994; Two-way ANOVA compared to WT_250ms, Sidak’s correction for multiple comparisons; data are reported as the mean ± SD. *** p < 0.001. D) Overlay of peak-scaled EPSCs during baseline (grey) and 5 s after 250 ms (dotted line), 500 ms (dashed line), and 1 s (solid line) depolarization using 0.2 mM EGTA from WT (black) and KO (blue) PCs. E) DSE timecourse from WT (black) and KO (blue) experiments using 0.2 mM EGTA and depolarization steps of 250 ms (filled circle), 500 ms (triangle) and 1 s (open circle). Data are reported as the mean ± SD. F) DSE Magnitude in WT (black) and KO (blue) after 250 ms (fine stripe), 500 ms (wide stripe) and 1 s (solid) depolarization steps using a 0.2 mM EGTA: WT_250ms = 38.0 ± 15.9%, n = 6 cells (4 mice); WT_500ms = 46.2 ± 16.5%, n = 6 cells (4 mice); p = 0.900; WT_1s = 51.6 ± 15.0%, n = 7 cells (4 mice); p = 0.493; KO_250ms = 4.8 ± 1.9%, n = 4 cells (3 mice); p = 0.0123; KO_500ms = 9.4 ± 7.8%, n = 5 cells (3 mice); p = 0.0250; KO_1s = 29.5 ± 19.7%, n = 11 cells (8 mice); p = 0.826; Two-way ANOVA compared to WT_250ms, Sidak’s correction for multiple comparisons; data are reported as the mean ± SD. * p < 0.05.

Figure 4. Spontaneous firing frequency, afterhyperpolarization (AHP) amplitude and Ca²⁺ coupling are reduced in the α2δ−2 KO.

A) Spontaneous spikes in WT (black) and KO (magenta) PCs; scale: 50 mV, 0.5 s. B) Spontaneous firing frequency in PCs: WT = 36.5 ± 15.8 Hz, n = 12 cells (10 mice); KO = 12.9 ± 11.4 Hz, n = 11 cells (8 mice); p = 0.0006; Student’s unpaired t-test; data are reported as the mean ± SD; *** p < 0.001. C) Averaged spontaneous spikes from WT and KO PCs during tonic firing (0.5 mM EGTA, black/magenta; 5 mM EGTA grey/light purple); scale: 10 mV, 2 ms. Grey dotted lines indicate V_thres and minimum voltage during AHP. Below, enlarged overlay demonstrating differences in AHP amplitude; scale: 5 mV, 1 ms. D) AHP amplitudes recorded using 0.5 mM EGTA or 5 mM EGTA intracellular solution: WT_0.5 = 12.8 ± 3.18 mV, n = 12 cells (10 mice); WT₅ = 9.93 ± 2.83 mV, n = 8 cells (6 mice); p < 0.0001; KO_0.5 = 8.18 ± 2.45 mV, n = 11 cells (8 mice); p < 0.0001; KO₅ = 8.98 ± 3.44 mV, n = 8 cells (5 mice); p < 0.0001; Two-way ANOVA comparison to WT 0.5 mM EGTA response (WT_0.5) with Sidak’s correction for multiple comparisons; data are reported as the mean + SD; **** p < 0.0001. E) No correlation of AHP amplitude vs. resting membrane potential (V_m) in WT (black) and KO (magenta) PCs using 0.5 mM EGTA. Linear regression, slope mean and 95% confidence interval; WT R² = 0.00185; −0.0673, 95% CI [−1.17, 1.03]; KO R² = 0.393; 2.16, 95% CI [0.137, 4.19]. F-I) No differences in spike waveform parameters in WT and KO PCs. (F) Maximum dV/dt: WT_0.5 = 169 ± 42.6 dV/dt; KO_0.5 = 196 ± 58.2 dV/dt; p = 0.217; (G) Minimum dV/dt: WT_0.5 = −130 ± 40.9 dV/dt; KO_0.5 = −121 ± 31.7 dV/dt; p = 0.594; (H) Spike threshold (V_thres): WT_0.5 = −41.1 ± 5.19 mV; KO_0.5 = −39.5 ± 8.50 mV; p = 0.591; (I) Spike height: WT_0.5 = 43.4 ± 11.6 mV; KO_0.5 = 46.4 ± 7.42 mV; p = 0.462; unpaired Student’s t-tests; data are reported as the mean ± SD; WT_0.5 n = 12 cells (10 mice); KO_0.5 n = 11 cells (8 mice). J) Phase plane plots of spontaneous spikes in WT and KO PCs (0.5 mM EGTA, black/magenta; 5 mM EGTA grey/light purple). Traces aligned by spike threshold (V_thresh) for comparison; scale: 100 mV/ms, 20 mV. Below, enlarged inset to illustrate differences in AHP minimum amplitude (arrows; 0.5 mM EGTA, black/magenta; 5 mM EGTA, grey/light purple); scale: 100 mV/ms, 5 mV. K) Immunohistochemistry of WT and KO cerebellar slices stained for parvalbumin (PV; cyan) and the BK channel (magenta); scale: 5 μm. Below, merged higher power image; scale: 2 μm. Images shown are max projection confocal images, 63×1.4 NA objective.