A T-type channel-calmodulin complex triggers αCaMKII activation

Fig. 1. Cav3.1 channels exhibit a calcium-dependent association with CaM. Biochemical analysis of binding interactions between Cav3.1 and CaM using the indicated antibodies to immunoprecipitate (IP) or immunoblot (IB). The experiments are representative of at least 3 repetitions. Input lanes reflect controls to verify the efficiency of the IP step should IB antibodies fail to reveal co-IPs. a, b Cav3.1 channels co-IP with CaM from rat brain lysate (a) and homogenates of tsA-201 cells coexpressing Cav3.1 and CaM (b). Co-IPs were conducted in the presence of 100 nM calcium. c The Cav3.1-CaM co-IP is lost for a construct lacking the Cav3.1 C-terminus (Cav3.1ΔCT). Co-IPs were conducted in the presence of 100 nM calcium. d A GST-pull down experiment between the C-terminus of Cav3.1 (CT-Cav3.1) onto CaM beads. GST-CT-Cav3.1 was grown in bacteria, and the GST tag subsequently cleaved off as the GST non specifically bound to CaM beads. CaM beads were instead incubated with purified recombinant Cav3.1 C-terminal peptides in the presence of 100 nM calcium and the sample eluted and run on a Western blot. The blot was probed with an antibody targeting the Cav3.1 C-terminus. e, f Co-IP tests…

Fig. 2. Cav3.1 and CaM associate at a level that supports FRET. Fluorescence spectral confocal images of tsA-201 cells expressing constructs of GFP-Cav3.1 (donor molecule) or mKate-CaM (acceptor molecule) excited at either 457 nm or 561 nm. Plots show the excitation (vertical line) and the associated emission spectrum for each condition. a, b Applying excitation at 457 nm to cells expressing either GFP-Cav3.1 alone (a) or in conjunction with the mKate acceptor molecule (b) results in a typical GFP emission spectrum. c, d Applying excitation at 561 nm to cells expressing either mKate-CaM alone (c) or in conjunction with GFP-Cav3.1 (d) results in a pure mKate emission spectrum. e Coexpressing GFP-Cav3.1 and mKate-CaM results in a double emission profile when excitation at 457 nm promotes FRET in a subset of cells (arrows). A control image of the same cells excited by a 561 nm laser line at right reveals that a GFP-Cav3.1 expressing cell (arrowhead) that does not exhibit FRET failed to express the mKate-CaM construct. All results were derived from at least 3 separate experiments. Scale bars 10 μm

Fig. 3. The Cav3-CaM association is lost upon Cav3 channel-mediated calcium influx. a, b Plots of FRET over time in tsA-201 cells coexpressing GFP-Cav3.1, mKate-CaM, and Kir2.1. Cells were exposed to Low [K]o (1 mM) or High [K]o (50 mM). a The FRET signal is stable in low [K]o for 11 min (upper plot) but is lost within 50 s upon perfusion of high [K]o (lower plot). b The decrease in FRET signal in high [K]o is blocked by expressing a Cav3.1 pore mutant that does not conduct calcium (upper plot), or 20 μM TTA-P2 (lower plot). c Bar plots of mean measures of FRET in the indicated conditions derived from 3 to 5 plates with 8–25 ROIs at time point 140 s. Mibefradil (Mib) 1 μM, Ni2+ 300 μM, BAPTA-AM 100 μM. Average values are mean ± SEM. * p < 0.05, ** p < 0.01, ns, not significant

Fig. 4. Cav3 channel-mediated calcium influx leads to αCaMKII aggregation. tsA-201 cells are cotransfected with GFP-αCaMKII, CaM, Cav3.1, and Kir2.1 and exposed to low [K]o (1.0 mM) or high [K]o (50 mM). Plots of the mean pixel variance of GFP-αCaMKII fluorescence in ROIs in the cytoplasm and nuclear regions shown at right. a, b In cells coexpressing Cav3.1 GFP-αCaMKII exhibits a diffuse cytoplasmic distribution in low [K]o that changes to aggregates in high [K]o, a pattern that is fully reversible upon returning to low [K]o. c, d Magnified images of the distribution of GFP-αCaMKII in low [K]o and after perfusing high [K]o, with plots indicating a restriction of clusters primarily to cytoplasm and peri-nuclear regions. e-i Formation of GFP-αCaMKII aggregates induced by high [K]o is prevented in cells lacking Cav3.1 (e, f), and blocked by substitution of CaM with CaM1234 (g, h), or by coexpression of CaMKIIN (0.3 μg) as an inhibitor of αCaMKII phosphorylation (i, j). Values are mean ± SD derived from n = 3 plates with 12–29 ROIs. Scale bars 10 μm.* p < 0.05

Fig. 5. Cav3.1-mediated activation of αCaMKII depends on an intact Cav3.1 C-terminus. tsA-201 cells are transfected with Cav3.1 or a Cav3.1 mutant lacking the C-terminus (Cav3.1ΔCT) to remove a site for preassociation of CaM with the Cav3.1 channel and preloaded with 2 μM X-rhod-1 to detect changes in [Ca]i. Cells are maintained at rest in low [K]o (1 mM) or exposed to 10 min of high [K]o (50 mM), with all medium containing 30 μM Cd2+ to block HVA calcium channels. The extent and timecourse of a shift in GFP-αCaMKII distribution from diffuse to aggregates in the cytoplasm is compared to that of a change in fluorescent intensity of X-rhod-1 relative to the mean of control baseline (ΔF/F0). High [K]o increased [Ca]i to an equivalent relative level in cells expressing either Cav3.1 or Cav3.1ΔCT but GFP-αCaMKII aggregation is blocked in cells expressing the Cav3.1ΔCT mutant. The distribution of GFP-αCaMKII is measured as density for at least 25–34 ROIs from 4 to 5 plates. Values are mean ± SEM derived from n = 4–5 plates with 25–34 ROIs

Fig. 6. GFP-αCaMKII clustering in transfected neurons. Neurons are cotransfected with GFP-αCaMKII and Cav3.1 (unlabeled) and exposed to low [K]o (1.0 mM) or high [K]o (50 mM). a-d A diffuse distribution of αCaMKII-GFP is stable in low [K]o over time (a, b), with high [K]o promoting aggregation of GFP-αCaMKII (c, d). e, f Formation of GFP-αCaMKII aggregates induced by high [K]o is blocked in the presence of mibefradil (1 μM) and Ni2+ (300 μM). g Bar plots of the mean pixel variance of GFP-αCaMKII fluorescence for the tests in (a-f). Values are mean ± SD derived from n = 3 plates with 5–10 ROIs. Scale bars 10 μm.* p < 0.05

Fig. 7. Cav3 calcium influx triggers αCaMKII phosphorylation in neurons. a-f Confocal images of cultured mouse hippocampal cells (a-c), and rat cerebellar tissue slices (d-f). Activated αCaMKII is identified by using a phospho-αCaMKII (pαCaMKII) antibody with additional DAPI label for nuclear staining in (a-c). Each preparation was maintained at rest in low [K]o (1 mM) or exposed to 10 min of high [K]o (50 mM) prior to fixation within 10 min of the end of high [K]o exposure. All results were derived from at least 3 separate experiments. In all neuronal cell tests calcium influx is restricted to Cav3 channels by maintaining cells in 30 μM Cd2+, 10 μM DNQX, 25 μM DL-AP5, and 1 μM TTX. a, b In hippocampal cell cultures high [K]o induces pαCaMKII labeling. d, e In cerebellar tissue slices high [K]o induces pαCaMKII labeling in Purkinje cell bodies. c, f High [K]o-induced labeling for pαCaMKII is blocked in both hippocampus (c) and cerebellum (f) in the presence of Cav3 channel blockers 1 μM mibefradil and 300 μM Ni2+. Dashed lines in (d-f) depict the boundaries of the Purkinje cell body layer. g Bar plots of the mean intensity of pαCaMKII fluorescence in hippocampal cell cultures and…