Neuronal ER-plasma membrane junctions couple excitation to Ca2+-activated PKA signaling

Abstract

Junctions between the endoplasmic reticulum (ER) and the plasma membrane (PM) are specialized membrane contacts ubiquitous in eukaryotic cells. Concentration of intracellular signaling machinery near ER-PM junctions allows these domains to serve critical roles in lipid and Ca²⁺ signaling and homeostasis. Subcellular compartmentalization of protein kinase A (PKA) signaling also regulates essential cellular functions, however, no specific association between PKA and ER-PM junctional domains is known. Here, we show that in brain neurons type I PKA is directed to Kv2.1 channel-dependent ER-PM junctional domains via SPHKAP, a type I PKA-specific anchoring protein. SPHKAP association with type I PKA regulatory subunit RI and ER-resident VAP proteins results in the concentration of type I PKA between stacked ER cisternae associated with ER-PM junctions. This ER-associated PKA signalosome enables reciprocal regulation between PKA and Ca²⁺ signaling machinery to support Ca²⁺ influx and excitation-transcription coupling. These data reveal that neuronal ER-PM junctions support a receptor-independent form of PKA signaling driven by membrane depolarization and intracellular Ca²⁺, allowing conversion of information encoded in electrical signals into biochemical changes universally recognized throughout the cell.

Fig. 1. SPHKAP clusters type I PKA near Kv2.1-associated ER-PM junctions.

a Mouse brain section immunolabeled for the Kv2 channel subunit AMIGO-1 (magenta) and SPHKAP (green) (representative of n = 8 mice). Scale bar, 1 mm. b, c SPHKAP (magenta), PKA-RIα/β (green), and Kv2.1 (blue) immunolabeling in hippocampal area CA1 in mouse brain (b). Scale bar, 10 µm. Lower panels, expanded views of regions marked by boxes (representative of n = 8 mice). Scale bar, 2 µm. Red line indicates region selected for intensity profile line scan (c). d Representative images of AMIGO-1 and SPHKAP immunolabeling in hippocampal area CA1 in brain sections from WT (n = 8), Kv2.1 KO (n = 4), Kv2.2 KO (n = 3), and Kv2 DKO (n = 3) mice. Scale bar, 5 µm. e Mean ± s.e.m. SPHKAP cluster size obtained in d. Each point represents 1 mouse. Statistical significance determined by one-way ANOVA followed by Dunnett’s multiple comparisons test (vs. WT); ****P < 0.0001, ***P = 0.0003, **P = 0.0055. f Representative images of hippocampal area CA1 neuron somata from WT, Kv2.1 KO, Kv2.2 KO, and Kv2 DKO brain sections immunolabeled for Kv2.1 (magenta), SPHKAP (green), and RI (blue). Region of interest (ROI) in which immunofluorescence was quantified shown in white (scale bar, 5 µm). g Frequency distribution and mean ± s.e.m. SPHKAP:RI PCC values in CA1 neurons from WT (n = 8), Kv2.1 KO (n = 6), Kv2.2 KO (n = 3), and Kv2 DKO (n = 3) mice. Each point represents 1 mouse; statistical significance determined using one-way ANOVA followed by Tukey’s multiple comparisons test (vs. WT); ****P < 0.0001, **P = 0.0076. h Mean ± s.e.m. immunofluorescence signal intensity of RI (left) and SPHKAP (right) in CA1 pyramidal neurons. Each point represents 1 mouse (as in g); statistical significance assessed using one-way ANOVA followed by Tukey’s multiple comparisons test (vs. WT); no significant differences were found. i Plot of normalized Kv2.1, SPHKAP, and RI (color gradient) immunofluorescence intensity in CA1 pyramidal neurons from n = 4 mice. Each point represents 1 cell (n = 429 cells). All source data are provided as a Source Data file.

Fig. 2. SPHKAP-type I PKA closely associates with Kv2.1 and VAPs.

Total internal reflection fluorescence (TIRF) images and ground-state depletion (GSD) reconstruction maps of cultured neurons immunolabeled for SPHKAP and a Kv2.1, c RI, or e VAPA/B (TIRF scale bars, 5 µm); representative of n = 5 [Kv2.1], n = 7 [RI], and n = 5 [VAPA/B] cells. Two magnified regions marked by boxes are shown for each panel; x-y fluorescence intensity profile from region indicated by dotted line show areas of proximity between SPHKAP and a Kv2.1 (scale bars i & ii = 1 µm), c RI (scale bar i = 1 µm; scale bar ii = 500 nm), and e VAPA/B (scale bar i = 1 µm; scale bar ii = 500 nm). b, d, f Histograms of nearest neighbor distances to SPHKAP centroids for b Kv2.1 (n = 4863 particles/3 cells), d RI (n = 3972 particles/4 cells), or f VAPA/B (n = 7140 particles/3 cells) centroids. Histograms were fit with a b Gaussian or d, f sum of two Gaussians function; mean values of distributions and R² values depicted on graphs. g Representative images of proximity ligation assay (PLA) puncta (magenta) in neurons expressing scrambled shRNA (scr) or one of two distinct anti-SPHKAP shRNAs (A or B) and immunolabled for Kv2.1 and SPHKAP. The neuron periphery is outlined in each image; scale bar, 10 µm. Mean ± s.e.m PLA signal density is presented below each image. Each point represents 1 cell; n = 31 (scr), 12 (A), 16 (B), 12 (Kv2.1 ab alone), and 11 (SPHKAP ab alone) cells; statistical significance determined using one-way ANOVA followed by Tukey’s multiple comparisons test (vs. scr); ****P < 0.0001, ***P = 0.0001, **P = 0.0004, *P = 0.0009. h As in g, but for SPHKAP-VAPA/B PLA signal density. Each point represents 1 cell; n = 19 (scr), 15 (A), 16 (B), 11 (SPHKAP ab alone), and 12 (VAP ab alone) cells; statistical significance determined using one-way ANOVA followed by Tukey’s multiple comparisons test (vs. scr); ****P < 0.0001. Source data are provided as a Source Data file.

Fig. 3. SPHKAP-type I PKA localizes to stacked ER cisternae at ER-PM junctions.

a EM images of SPHKAP-immunoperoxidase DAB reaction product acquired from the soma of a CA1 pyramidal neuron in a rat brain section. High-magnification images of insets are provided on right. Scale bars, 1 µm (left panel) and 500 nm (insets i & ii). Images are representative of results obtained from n = 2 rats. b EM image of SPHKAP-immunogold particles acquired from the soma of a CA1 pyramidal neuron in a mouse brain section. Scale bar, 200 nm. Images are representative of results obtained from n = 2 mice. c EM image of SPHKAP-immunogold particles and RI-immunoperoxidase reaction product acquired from the soma of a CA1 pyramidal neuron in a mouse brain section. Scale bar, 200 nm. Images are representative of results obtained from n = 2 mice.

Fig. 4. RI anchoring drives SPHKAP-RI co-clustering.

a Schematic of RIα-GFP (D/D dimerization and docking domain [black], pseudosubstrate inhibitor domain [purple], CNB-A and CNB-B cyclic nucleotide binding domains [green]) and SPHKAP-mScarlet constructs (p-FFAT phospho-FFAT motif [blue], AKB A-kinase binding domains [red]). b Representative images of HEK293T (HEK) cells quantified in c-d expressing full-length SPHKAP-mScarlet (magenta) and/or RIα-GFP and RIα-GFP D/D domain mutants (green). Scale bar, 10 µm. c Quantification of pooled SPHKAP-mScarlet mean ± s.e.m. cluster size in all HEK cells co-expressing indicated RIα-GFP constructs. Data are mean ± s.e.m. area of 1045 puncta from 87 cells (SPHKAP), 168 puncta from 97 cells (+RIα), 115 puncta from 35 cells (+D/D-linker), 588 puncta from 37 cells (+RIα-ΔD/D-linker), and 52 puncta from 15 cells (+D/D). Statistical significance determined by one-way ANOVA followed by Dunn’s multiple comparisons test (vs. SPHKAP); ****P < 0.0001. e Representative images of HEK cells expressing SPHKAP-mScarlet deletion mutants (magenta) and RIα-GFP (green) (scale bar, 10 µm); mean ± s.e.m. SPHKAP cluster size for each condition is quantified below the representative images for 31 (SPHKAP-N), 41 (SPHKAP-N + RIα), 40 (SPHKAP-C), and 37 (SPHKAP-C + RIα) cells; each point represents 1 cell. Statistical significance determined by one-way ANOVA followed by Dunn’s multiple comparisons test (vs. SPHKAP-C + RI); ****P < 0.0001. f Representative images of HEK cells expressing full-length SPHKAP-mScarlet (magenta), RIα-GFP (green), and the indicated interfering peptide (not shown) (scale bar, 10 µm); mean ± s.e.m. SPHKAP cluster size for each condition is quantified below the representative images for 100 (RIAD-scr), 65 (RIAD), 28 (sAKAP-IS-scr), 32 (sAKAP-IS), and 66 (PKI) cells; each point represents 1 cell. Statistical significance determined by one-way ANOVA followed by Dunn’s multiple comparisons test (vs. indicated pairs); ****P < 0.0001; ns not significant (P = 0.5189). PKI compared to either scr construct was not significant (P > 0.9999). Source data are provided as a Source Data file.

Fig. 5. SPHKAP-RI complexes associate with VAPs.

a Alignment of a conventional FFAT motif with phospho-FFAT motifs (7 core residues highlighted) found in SPHKAP, Kv2.1, and STARD3. Alcoholic residues are highlighted in green, acidic residues in red, and aromatic residues in blue; all other residues are in black. Critical F residue at position 2 (corresponding to SPHKAP F212) and T residue at position 4 (corresponding to SPHKAP T214) in these FFAT motifs are outlined with boxes. b Representative images of HEK cells (quantified in c) expressing WT SPHKAP-mScarlet, SPHKAP-C-mScarlet, or SPHKAP-mScarlet F212A, RIα-GFP, and immunolabeled for endogenous VAPs and PKA-C. Scale bar, 10 µm. c Quantification of SPHKAP-mScarlet overlap with endogenous VAPs (mean ± s.e.m.; Pearson’s correlation coefficient, PCC) in HEK cells transfected with the indicated SPHKAP-mScarlet constructs; each point represents 1 cell (n = 23 [WT], 15 [SPHKAP-C], 25 [F212A], 13 [T214A], and 9 [T214D] cells). Statistical significance determined by one-way ANOVA followed by Tukey’s multiple comparisons test (vs. WT); ****P < 0.0001. d Representative TEM images of OSER in HEK cells co-transfected with SPHKAP-mScarlet and RIα-GFP and histogram of number of OSER structures/cell in single EM sections acquired from HEK cells transfected with SPHKAP-mScarlet + RIα-GFP (n = 26 cells) or SPHKAP-C-mScarlet + RIα-GFP (n = 30 cells). Scale bars, 500 nm. e Representative images of neurons treated with control TAT-FFAT-scr or TAT-FFAT 472–484 peptides overnight and then immunolabeled for endogenous SPHKAP and VAPs (quantified in f). Scale bar, 10 µm. f Colocalization (Pearson’s colocalization coefficient, PCC) of VAPs with SPHKAP under indicated conditions. Each point represents 1 cell (n = 14 [scr], 13 [472–484], and 13 [465–494] cells). Statistical significance determined by one-way ANOVA followed Tukey’s multiple comparisons test (vs. scr); ***P = 0.0004, **P = 0.0038. Source data are provided as a Source Data file. g Schematic illustrating hypothesized mechanism of SPHKAP-RI organization at Kv2.1-associated ER-PM junctional domains: phosphorylation of the phospho-FFAT (p-FFAT) motif in Kv2.1 or SPHKAP leads to their physical association with VAPs, while RI association with SPHKAP enables formation of SPHKAP-RI oligomers.

Fig. 6. SPHKAP supports LTCC-dependent Ca²⁺ entry.

a Images of a hippocampal neuron (representative of n = 6 neurons) immunolabeled for SPHKAP (green), Kv2.1 (blue), Cav1.2 (magenta), and RyR (yellow). Scale bar, 10 µm. b Representative images of proximity ligation assay (PLA) puncta (magenta) detected in control neurons expressing a scrambled shRNA (scr) or in neurons expressing one of two distinct anti-SPHKAP shRNAs (A or B) and immunolabled for RyRs and RI. The neuron periphery is outlined in each image; scale bar, 10 µm. Quantification of mean ± s.e.m PLA signal density for each condition is presented below each image. Each point represents 1 cell; n = 26 (scr), 21 (A), 19 (B), 11 (RyR ab alone), and 11 (RI ab alone) cells; statistical significance was determined using one-way ANOVA followed by Tukey’s multiple comparisons test (vs. scr); ****P < 0.0001. c Mean ± s.e.m. jRGECO1a fluorescence in neurons expressing either control scr shRNA (black, n = 48 cells) or anti-SPHKAP shRNAs (shRNA A, red, n = 30 cells; shRNA B, blue, n = 34 cells) and depolarized with 45 mM external K⁺. d Mean ^± s.e.m. jRGECO1a fluorescence in neurons expressing either control scr shRNA (black, n = 28 cells) or anti-SPHKAP shRNAs (shRNA A, red, n = 18 cells; shRNA B, blue, n = 28 cells) and depolarized with 45 mM external K⁺ and 10 µM nimodipine. e Mean ± s.e.m peak jRGECO1a fluorescence (normalized to response in scr cells) in cells depolarized with 45 mM external K⁺; each point represents 1 cell (n cells as in c); statistical significance was determined using one-way ANOVA followed by Dunn’s multiple comparisons test (vs. scr); ***P = 0.0002, **P = 0.0004. f Mean ± s.e.m peak jRGECO1a fluorescence (normalized to response to 45 mM K⁺ in scr cells) in cells depolarized with 45 mM external K⁺ in the presence of 10 µM nimodipine; each point represents 1 cell (n cells as in d); statistical significance was assessed using one-way ANOVA followed by Dunn’s multiple comparisons test (vs. scr); P = 0.9192 (A), P > 0.9999 (B).

Fig. 7. SPHKAP promotes depolarization-induced transcription factor activation.

a Experimental paradigm to assess depolarization-induced transcription factor activation in cultured hippocampal neurons. b Images of phosphorylated CREB (pCREB) and c-Fos immunolabeling in control (4.7 mM K⁺ solution: 4.7 K) and depolarized (45 mM K⁺ solution: 45 K) neurons expressing scr or SPHKAP shRNA B. Scale bar, 10 µm. c Mean ± s.e.m pCREB and c-Fos immunofluorescence signal intensity; each point represents 1 cell (n = 10 [scr, 4.7 K pCREB], 12 [SPHKAP, 4.7 K pCREB], 15 [scr, 45 K pCREB], 14 [SPHKAP, 45 K pCREB], 25 [scr, 4.7 K c-Fos], 31 [SPHKAP, 4.7 K c-Fos], 20 [scr, 45 K c-Fos], and 30 [SPHKAP, 45 K c-Fos] cells. Statistical significance determined using one-way ANOVA followed by Tukey’s multiple comparisons test; *P = 0.0185, **P = 0.0051, ***P = 0.008 (pCREB) or 0.006 (c-Fos), ****P < 0.0001. Source data are provided as a Source Data file.

Fig. 8. SPHKAP underlies PKA-dependent excitation-phosphorylation coupling.

a Representative images of control (4.7 mM K⁺) and depolarized (45 mM K⁺) neurons (quantified in b) immunolabeled for Kv2.1 (magenta), PKA-C (green), SPHKAP (blue), and phosphorylated PKA (p-PKA) substrate (orange). Scale bar, 10 µm. b Mean ± s.e.m. p-PKA substrate immunofluorescence signal in neurons stimulated with elevated extracellular potassium ([K⁺]_o); n cells analyzed under each condition noted below graph (black: vehicle; red: nimodipine; blue: H89). c PKA-C cluster size (points) and spatial association with SPHKAP (Pearson’s correlation coefficient [PCC], color gradient) in neurons stimulated with 4.7 mM K⁺ (4.7 K), 45 mM K⁺ (45 K), 45 mM K⁺ + 10 µM nimodipine (45 K + nim), or 1 µM isoproterenol (ISO). GAD67⁺ cells: green symbols. Data are mean ±  s.e.m. (excluding GAD67^- cells); each point represents 1 cell; n cells analyzed under each condition is noted below the graph (black: GAD67^- cells, green: GAD67⁺ cells). Statistical significance of differences in PKA-C cluster size or PCC in GAD67^- cells assessed using one-way ANOVA followed by Dunnett’s multiple comparisons test (vs. 4.7 K); PKA-C cluster size: **P = 0.0072, ***P = 0.0004, ****P < 0.0001; PCC: ^#P = 0.0334^{, ‡}P < 0.0001. d Left, mean ± s.e.m. ExRai-AKAR2 fluorescence in neurons expressing PKI, RIAD, or RIAD-scr. Right, peak fluorescence from data in left panel. Each point represents 1 cell (n = 22 [RIAD-scr], 24 [RIAD], and 17 [PKI] cells). Statistical significance determined using one-way ANOVA followed by Tukey’s multiple comparisons test; *P = 0.0443, **P = 0.0051, ***P = 0.002. e As in d, in neurons stimulated with forskolin and IBMX (n = 10 [RIAD-scr], 12 [RIAD], and 12 [PKI] cells); **P = 0.0034 (RIAD-scr versus PKI) and **P = 0.0023 (RIAD versus PKI). f as in d, in neurons expressing scr or SPHKAP shRNA (n = 25 [scr, 45 K], 25 [SPHKAP, 45 K], 7 [scr, 45 K + nim], and 5 [SPHKAP, 45 K + nim] cells). Statistical significance determined using unpaired two-tailed Student’s t-test; ****P < 0.0001. g Hypothesized mechanism of depolarization-induced activation of PKA and regulation of Ca²⁺ signaling machinery. Source data are provided as a Source Data file.