Expression and localization of βII-spectrin in cross sections of the sciatic nerve.

(A) Low magnification confocal image of a cross-section of an adult mouse sciatic nerve immunostained for βII-spectrin (cyan). The characteristic concentric organization of myelinated axons is observed: an outer ring corresponding to the soma of the ensheating Schwann cell, an unlabeled region corresponding to compact myelin, and an inner ring surrounding the axon. Scale bar: 20 μm. (B-C) Magnification of individual myelinated axons double immunostained for βII-spectrin (cyan) and MBP (magenta, B) or βIII-tubulin (orange, C). βII-spectrin is located at both sides of the MBP signal, and outside of the βIII-tubulin signal. sc: Schwann cell soma; cm: compact myelin compartment; ax: axon. Scale bars: 5 μm. (D-E) STED microscopy images showing the distribution of βII-spectrin (cyan) combined with MBP (D), or protein 4.1B (E), and NF-H (gray, confocal channel). Yellow lines exemplify where intensity profiles were collected. Scale bars: 1μm. (F) Magnified regions of STED images exemplifying where the intensity profiles for βII-spectrin, MBP, protein 4.1B, TrkB-ECD and NF-H were obtained. cm: compact myelin compartment; ax: axon. Scale bar: 0.5 μm. (G) Normalized mean intensity profiles aligned to the βII-spectrin peak (arbitrary zero). Left: MBP and NF-H are distributed on opposite sides of βII-spectrin. Center: Protein 4.1B presents a profile almost identical to that of βII-spectrin. Right: The TrkB-ECD peak is slightly more external relative to that of βII-spectrin. Solid lines represent the mean, and shaded areas indicate the ± SEM. (H) Graph showing the linear distances between each protein to βII-spectrin. Negative values are arbitrarily located toward the exterior of the axon, and positive values toward its interior. Mean ± SEM. Pair-wise comparisons reflect one-way ANOVA followed by Tukey’s post-hoc tests (***: p<0.0001, *: p<0.05, ns: non significant). All groups were significantly different from zero (i.e. different from βII-spectrin, one-sample t test, p<0.05) (I) Representative diagram summarizing the inferred localization of βII-spectrin with respect to the other proteins evaluated.

3D-dSTORM imaging of βII-spectrin reveals a non-uniform distribution along the axonal perimeter consistent with the MPS.

A) dSTORM image of individual axons stained for βII-spectrin (cyan) and βIII-tubulin (magenta) with the z axis (∼800 nm) collapsed to show only two dimensions. Scale bars: 1 μm. (B) Selected regions from the reconstructed axons, indicated by orange squares in (A), used for quantitative analysis of the axial (z) distribution of localizations. Peaks in the axial localization profiles were identified by fitting a Gaussian mixture model (GMM; orange dashed curves). The inter-peak distance was calculated as the separation between the estimated means of consecutive Gaussian components. The corresponding kernel density estimation (KDE) is shown in black. (C) Distribution of inter-peak distances measured along the z axis, pooling all values obtained from the analyzed axonal cross-sections. The mean inter-peak distance measured is 170 ± 15 nm.

βII-Spectrin localizations form clusters of varying sizes along the axonal cortex.

A) Example images obtained by rendering the single-molecule localizations of the βII-spectrin clusters. (B) Estimation of axon perimeter and cluster area. The areas shown correspond to the clusters rendered in the first example image in (A). (C) Distribution of βII-spectrin clusters area. Median: 1965 nm2 (D) Scatter plot of number of clusters per axon vs axon perimeter. The linear regression (y = 4.08x −12.62), shown in brown, with a Pearson correlation coefficient of 0.81, indicates a significant positive correlation between the two variables.

Spatial analysis of βII-spectrin clusters distribution along the axon perimeter.

A) First nearest-neighbor (1NN) cluster distance distributions. (B) Boxplots for individual axons showing the relationship between axonal perimeter and the 1NN distances between βII-spectrin cluster mass centers. The median value for each axon is highlighted in orange. The horizontal light orange line in the background represents the median of all values (260 nm). (C) Distribution of axon occupancy values by βII-spectrin clusters across all analyzed axons. (D) Representative scheme of a randomized distribution of βII-spectrin clusters along an axonal perimeter. The axonal contour defined the inner (red) and outer (green) boundaries of a 100 nm-wide annular region. For each axon, cluster centers were randomly assigned within this region under a minimum inter-cluster distance constraint (based on the smallest experimental 1NN distance), with internal cluster structure (blue) preserved via rigid-body translation. (E) Left: histogram of 1NN distances collected from the experimental data and from randomized βII-spectrin clusters. N = 1000 randomizations were performed for each axon. Right: cumulative density functions of experimental and random distances (Inset: magnification of plot. Kolmogorov-Smirnov D = 0.054, p-value<10-3).