Protein kinase CK2 contributes to the organization of sodium channels in axonal membranes by regulating their interactions with ankyrin G

Abstract

In neurons, generation and propagation of action potentials requires the precise accumulation of sodium channels at the axonal initial segment (AIS) and in the nodes of Ranvier through ankyrin G scaffolding. We found that the ankyrin-binding motif of Na(v)1.2 that determines channel concentration at the AIS depends on a glutamate residue (E1111), but also on several serine residues (S1112, S1124, and S1126). We showed that phosphorylation of these residues by protein kinase CK2 (CK2) regulates Na(v) channel interaction with ankyrins. Furthermore, we observed that CK2 is highly enriched at the AIS and the nodes of Ranvier in vivo. An ion channel chimera containing the Na(v)1.2 ankyrin-binding motif perturbed endogenous sodium channel accumulation at the AIS, whereas phosphorylation-deficient chimeras did not. Finally, inhibition of CK2 activity reduced sodium channel accumulation at the AIS of neurons. In conclusion, CK2 contributes to sodium channel organization by regulating their interaction with ankyrin G.

Figure 1.

Abrogation of E1111 of the Na_v1.2 ankyrin-binding motif did not impair ion channel segregation at the AIS of hippocampal neurons. Cell surface distribution of K_v2.1 (A), K_v2.1-ΔCter (B), K_v2.1-Na_v1.2 (C), and K_v2.1-Na_v1.2 E1111A mutant (D–F) in cultured hippocampal neurons. The addition of the Na_v1.2 ankyrin-binding motif to the C terminus of K_v2.1-ΔCter segregated K_v2.1 at the AIS (C, arrow). K_v2.1-Na_v1.2 compartmentalization at the AIS was not impaired by the Na_v1.2 E1111A mutation (D and F, arrow). The indicated constructs were immunodetected with an antibody to myc (green) before cell permeabilization; the somatodendritic domain was subsequently identified by MAP2 staining. Bars, 10 μm.

Figure 2.

Requirement for glutamate 1111 and serine residues for K_v2.1-Na_v1.2 compartmentalization at the AIS of hippocampal neurons. (A) Schematic representation of the ankyrin-binding motif of sodium channels predominantly expressed in the CNS. Note the presence of four conserved serines (black boxes) and three potential serine phosphorylation sites for CK2 (Na_v1.2 S1112, S1124, and S1126). Numbers refer to the position of the amino acid residues in the corresponding Na_v1 types. (B–E) Cell surface distribution of K_v2.1-Na_v1.2 mutants. Cultured hippocampal neurons were transfected with the indicated constructs. 1 d after transfection, K_v2.1-Na_v1.2 was detected with an antibody to myc (green) before permeabilization, and the somatodendritic domain was subsequently identified by MAP2 staining (red). Mutations of different serine residues in the ankyrin-binding motif did not perturb K_v2.1-Na_v1.2 segregation at the AIS (B and C). In contrast, co-mutation of E1111 with serine residues induced a loss of K_v2.1-Na_v1.2 compartmentalization (D and E). (C and E) Histograms of the cell surface distribution of K_v2.1-Na_v1.2 mutants. The percentage of myc-positive neurons were classified into four categories: myc staining segregated at the AIS ([AIS]); distributed at the cell surface of the soma and proximal dendrites with an enrichment at the AIS (SD-[AIS]); enriched at the AIS with a distribution in soma, proximal dendrites, and axons (SD-[AIS]-PA); and uniformly distributed at the cell surface (non polarized; SD-A). 100% represents the total population of transfected neurons. Data are means ± SD from two to three different experiments; n denotes the total number of cells analyzed for each construct. Bars, 10 μm.

Figure 3.

CK2 phosphorylates the Na_v1.2 ankyrin-binding motif and regulates its interaction with MBD-ank. (A–C) The Na_v1.2 ankyrin-binding motif is phosphorylated by CK2 in vitro. (A) Schematic representation of GST-Na_v1.2 linker II-III constructs. The position of the ankyrin-binding motif is indicated by gray boxes. Numbers refer to the position of corresponding amino acid residue in Na_v1.2. (B) The constructs were subjected to in vitro CK2 phosphorylation followed by SDS-PAGE. Coomassie brilliant blue staining (CBB) and ³²P incorporation revealed by autoradiography are shown. (C) The effect of site-directed serine mutations on CK2 phosphorylation of GST-Na_v1.2 II-III. CBB staining (bottom) and ³²P incorporation revealed by autoradiography (top) are shown. Numbers to the right of the gel blots indicate the molecular mass of standard markers (in kD). (D–F) SPR analysis of the interaction between the Na_v1.2 ankyrin-binding motif and MBD-ank. Typical sensorgrams are illustrated. Increasing concentrations of MBD-ankB ranging from 0.1 to 250 nM were injected over immobilized GST-Na_v1.2 1,080–1,203 (D), GST-Na_v1.2 989–1,133 (E), and GST-Na_v1.2 1,080–1,203 E1111A (F) before (top) and after in situ CK2 phosphorylation (bottom) of the immobilized constructs.

Figure 4.

CK2α is concentrated at AISs. (A) A cultured hippocampal neuron was immunostained for CK2α (A1) and ankyrin G (A2). CK2α immunoreactivity is highly expressed in the initial segment (box). Enlarged views of the initial segment are visualized in A1–A3, bottom. (B and C) In the CA1 area of the hippocampus, CK2α immunoreactivity is visible within nuclei (B1, arrowheads) of putative pyramidal cells and interneurons. As demonstrated by colabeling with ankyrin G (B2, arrows), CK2α is present in initial segments all over the different strati (B1 and B3, arrows). CK2α shows light and diffuse additional staining within the neuropil. CK2α immunoreactivity (C1 and C3, arrows) and ankyrin G labeling (C2, arrows) are concentrated along the cytoplasmic membrane of an initial segment in the stratum pyramidale next to a CK2α-labeled nucleus (C1 and C3, asterisks). (D) In cerebellum, nuclei (D1, asterisk) of Purkinje cells and interneurons of the molecular layer (ml) are labeled for CK2. Coimmunolabeling for CK2α (D1) and for ankyrin G (D2) is visible within initial segments of a putative Purkinje cell (D1–D3, arrows), granular cells (D1–D3, double arrowheads), and interneurons (D1–D3, single arrowheads). gl, granular layer; Pcl, Purkinje cell layer, so, stratum; sp, stratum pyramidale; sr, stratum radiatum. Bars: (A) 23 μm; (B) 26.8 μm; (C and D) 6.82 μm.

Figure 5.

CK2α is concentrated in CNS and PNS nodes of Ranvier. (A and B) CK2α immunoreactivity is visible as numerous punctiform structures (B1 and B3, arrows) colabeled for ankyrin G (B2 and B3, arrows) within the alveus and cortical white matter that probably correspond to nodes. B1–B3 are enlarged views of the boxed areas in A1–A3. (C and D) Ranvier nodes are labeled for CK2α and ankyrin G (C1–C3, arrowheads) in teased fibers of the sciatic nerve. CK2α labeling is concentrated at the membrane of the nodal region (D1, arrows). CK2α immunoreactivity is also visible in the nucleus of a Schwann cell (e.g., C1 and C3, arrows). al, alveus; wm, white matter. Bars: (A) 26.8 μm; (B) 6. 71 μm; (C) 23.2 μm; (D) 3 μm.

Figure 6.

Impact of K_v2.1-Na_v1.2 and of phosphorylation-deficient K_v2.1-Na_v1.2 constructs on the accumulation of sodium channels (Na_v1) and KCNQ2/KCNQ3 potassium channels at the AIS of cultured hippocampal neurons. (A, top and middle) K_v2.1-Na_v1.2 expression perturbed Na_v1 accumulation at the AIS, unlike the phosphorylation-deficient K_v2.1-Na_v1.2 4SA mutant. Hippocampal neurons were transfected with either K_v2.1-Na_v1.2 or phosphorylation-deficient K_v2.1-Na_v1.2 mutants. Then cells were stained for myc (gray), ankyrin G (red), and sodium channels (green). (B) Quantification of Na_v1 and ankyrin G staining intensity in untransfected cells (A, open arrowheads) and in transfected cells (closed arrowheads). (A, bottom) K_v2.1-Na_v1.2 expression did not perturb KCNQ3 accumulation at the AIS. Cells transfected with K_v2.1-Na_v1.2 were subsequently stained for myc (gray), ankyrin G (red), and KCNQ3 potassium channels (green). (C) Quantification of KCNQ3 and ankyrin G staining intensity. Fluorescence intensity measured in transfected cells, identified by myc staining, was normalized by taking as 100% the staining intensity measured in nontransfected cells (arrowheads). Numbers at the base of the bars denote the number of quantified cells. Error bars indicate mean ± SEM. Mann-Whitney test: *, P < 0.05; **, P < 0.01; ***, P < 0.001. WT, wild type. Bars, 10 μm.

Figure 7.

The effect of a CK2 inhibitor on the accumulation of sodium channels at the AIS of cultured hippocampal neurons. Cultured hippocampal neurons were treated with either DMAT (50 μM) or with DMSO (control cells) either from 2 to 4 DIV (A and B) or from 9 to 10 DIV (C and D). They were subsequently immunostained for Na_v1 channels (green), ankyrin G (red), and MAP2 (blue). (B and D) Quantification of Na_v1 and ankyrin G staining intensity. Fluorescence intensity was normalized by taking as 100% the staining intensity measured in control cells. The results are from three independent experiments, the number of quantified AIS ranged from 70 to 230. Error bars indicate mean ± SEM. t test: ***, P < 0.0001. Bars, 10 μm.