Fig 2: Measurement of synaptopodin (SYNPO) distribution along dendritic spines using line scanning(A) Representative Z-series projection images for analyzing SYNPO signal. From left to right: YFP; SYNPO; YFP+SYNPO; YFP (outlined)+SYNPO. The 20-μm dendrite segment is labeled by a red line. The region within the yellow rectangle has been further enlarged in (D) for analysis. Scale bar, 3.5 μm.(B) The steps for acquiring the scale of the image and adjusting the width of the segmented line to 0.5 μm.(C) Drag the segmented line from the spine top to the spine base within the defined 20-μm dendrite segment. Take spines 1,2 and 3 as examples, which are labeled pink.(D) Click the keyboard “T” in both the YFP and SYNPO channels of spine 1–3 to show the respective information in the ROI table.(E) “Multiplot” the data for spines 1–3 → “list” → export the SYNPO-signal distributed data to Excel. The SYNPO signal of spine 1 is < 20 pixels/μm, so it is a SYNPO– spine. The SYNPO signal of spines 2 and 3 is > 20 pixels/μm, thus representing candidate SYNPO+ spines, which are subjected to further analysis in Figure 6.

Fig 3: Determination of SYNPO+ dendritic spines from the Z-series images(A) Flowchart of the process to determine SYNPO+ dendritic spines. The Z-series projection image is identical to that shown in Figure 5C.(B) The inset of (A) has been further enlarged and is shown as Z-series images to validate SYNPO+ spines (Spines 2 and 3). The results indicate that Spine 2 is SYNPO+, whereas Spine 3 is SYNPO– (<20 pixels/μm).

Fig 4: SYNPO overexpression rescues the deficits of Klhl17-deficient neurons.(A–D) Ryanodine treatment alters calcium dynamics in wild-type neurons but not Klhl17–/–neurons. (A) Representative patterns of calcium events. Scale units: amplitude (ΔF/F) and time (1 min). Quantification of the frequency (B), interval time (C), and amplitude (D) of calcium events. (E–H) SYNPO overexpression restores calcium dynamics in Klhl17-deficient neurons. Cultured neurons were co-transfected with GCaMP6s and SYNPO (or “S”) or vector control, as indicated, at 12 DIV, before performing calcium recording by means of live-imaging at 18–19 DIV. (E) Representative patterns of the calcium events. Scale units: amplitude (ΔF/F) and time (1 min). Quantification of frequency (F), interval time (G) and amplitude (H). (I–L) SYNPO overexpression also rescues the spine enlargement of Klhl17-deficient neurons upon bicuculline stimulation. Cultured neurons were transfected with GFP and SYNPO or vector control at 12 DIV, as indicated, and performed activity-induced spine enlargement at 18 DIV. (I) Representative images of dendritic segments. (J) Quantification of density, (K) width and (L) length of dendritic protrusions. In (J), the density of dendritic spines was analyzed on both a per-neuron and per-dendrite basis. (M, N) SYNPO overexpression increases ERK phosphorylation of Klhl17-deficient neurons upon bicuculline treatment. Cultured neurons were co-transfected with GFP and SYNPO or vector control at 12 DIV and treated with bicuculline at 18 DIV. (M) Representative images. GFP images are shown in insets, with transfected cells highlighted by yellow arrows. (N) Quantification of the relative intensity of phospho-ERK signals. All data were collected from at least 2 independent experiments. The sample sizes of examined neurons (N), dendritic segements (n), and dendritic spines (n) are indicated. For the same set of experiments, the sample size is labeled only in 1 panel. The data represent mean ± SEM and cumulative curves (K, L: right). Individual data points are also shown. ** P < 0.01; *** P < 0.001; ns, not significant. Two-way ANOVA (B–D, F–H, J–L: left, N); Kolmogorov–Smirnov test for cumulative probability (K, L: right). Scale bars: (I) 5 μm; (M) 20 μm. The numerical value data and statistical results are available in S1 and S2 Data, respectively. DIV, day in vitro; ERK, extracellular signal-regulated kinase; KLHL17, Kelch-like protein 17.

Fig 5: The KLHL17-SYNPO association is required for activity-dependent dendritic spine enlargement and neuronal activation.(A–D) Overexpression of SYNPO-C impairs calcium events. Cultured neurons were transfected with GCaMP6s and the indicated plasmids at 12 DIV before performing calcium recording based on GCaMP6s signals using live-imaging at 18–19 DIV. (A) Representative patterns of calcium events. Scale units: amplitude (ΔF/F) and time (1 min). Quantification of frequency (B), interval time (C), and amplitude (D). (E–H) Overexpression of SYNPO-C impairs the spine enlargement induced by bicuculline. Cultured neurons were transfected with the indicated plasmids at 12 DIV and activity-induced spine enlargement was assessed at 18 DIV. (E) Representative images of dendritic segments. (F) Quantification of density, (G) width, and (H) length of dendritic protrusions. In (F), the density of dendritic spines was analyzed on both a per-neuron and per-dendrite basis. (I, J) Overexpression of SYNPO-C reduces bicuculline-induced ERK phosphorylation. Cultured neurons were transfected with the indicated plasmids at 12 DIV and treated with bicuculline at 18 DIV. (I) Representative images. GFP images are shown in insets, with yellow arrows indicating transfected neurons. (J) Quantification of the relative intensity of phospho-ERK signals. Images were collected from 2 independent experiments. All data were collected from at least 2 independent experiments. The sample sizes of examined neurons (N), dendritic segements (n), and dendritic spines (n) are indicated. For the same set of experiments, the sample size is labeled only in 1 panel. The data represent mean ± SEM and cumulative curves (G, H: right). Individual data points are also shown. *** P < 0.001; ns, not significant. One-way ANOVA (B–D, F–H: left, J); Kolmogorov–Smirnov test for cumulative probability (G, H: right). Scale bars: (E) 5 μm; (I) 20 μm. The numerical value data and statistical results are available in S1 and S2 Data, respectively. DIV, day in vitro; ERK, extracellular signal-regulated kinase; KLHL17, Kelch-like protein 17.