Fig 2: Ectopic expression of brevican in the medial nucleus of the trapezoid body (MNTB) of Hapln4-KO mice. To clarify the spatial location of perineuronal net components, the immunoreactivity of HAPLNs (A,B,E,F), brevican or aggrecan (C,D,G,H), and tenascin-R (I,J) (green) is examined in the MNTB of 4-month-old adult mice using VGLUT1 as a marker of the calyces of Held (magenta). The intensity profiles of the fluorescence signals along the white lines are shown in the lower panels. The fluorescence intensities are normalized by the highest intensities of individual proteins within the lines except in (A,B). Since these have a relatively low signal-to-noise ratio of the anti-HAPLN4 antibodies, the background is high when the intensities are normalized as previously described. Of note is the observed ectopic shift of brevican immunolocalization from the perisynaptic space in the wild type (arrows) to the surrounding extracellular space in the neuropil in Hapln4-KO (arrowheads). Acan, aggrecan; Bcan, brevican; TN-R, tenascin-R. The scale bar represents 10 µm.

Fig 3: Schematic representation of HAPLN-dependent perineuronal net (PNN) micro-organization at the calyx of Held synapses. (A,C,D) In wild-type mice, HAPLN4-brevican-type PNN is enriched in the perisynaptic space between the calyx of Held terminals and the principal neuron, and HAPLN1–aggrecan–tenascin-R-type PNN surrounds the whole calyx of Held terminals. (B,E,F) In Hapln4-KO, the ectopic expression of brevican is detectable with HAPLN1 association at the surrounding extracellular space in the neuropil in addition to the HAPLN1–aggrecan–tenascin-R complex. Acan, aggrecan; Bcan, brevican; TN-R, tenascin-R; HA, hyaluronan; PSD, postsynaptic density.

Fig 4: Detection of specific HAPLN–lectican molecular interactions by an in situ proximity ligation assay (PLA). In situ PLA to visualize in situ endogenous HAPLN–lectican interactions (green) in the medial nucleus of the trapezoid body of 5-month-old adult WT (A–D) and KO (E–G) mice using VGLUT1 as a marker of the calyx of Held (magenta). A pair of primary antibodies of HAPLN4 and brevican shows dot signals at the edge or perisynaptic space of the calyx of Held terminals (C, arrowheads). Notably, in the absence of HAPLN4, brevican shows an ectopic shift of localization to the surrounding neuropil, where HAPLN1 might potentially stabilize brevican in the aberrant condition (F, arrowheads). The specificity of the PLA reaction is validated with two different negative controls: one is a reaction that omits the anti-HAPLN4 antibody on wild-type tissue (D); the other is the PLA reaction of HAPLN4 and brevican using Hapln4-KO tissue (G). The non-specific reactions in the negative controls are indicated by asterisks (D,G). Acan, aggrecan; Bcan, brevican. The scale bar represents 10 µm.

Fig 5: Distribution of HAPLNs and lecticans in the medial nucleus of the trapezoid body (MNTB) of adult wild-type (WT) mice. Immunohistochemical detection of HAPLNs (green) and lecticans (magenta) in the MNTB of 4-month-old adult WT mice using HAPLN4 (A–D), HAPLN1 (E–H), aggrecan (A,B,E,F), and brevican (C,D,G,H) immunostaining. The intensity profiles of the fluorescence signals along the white lines are shown in the rightmost panels. The fluorescence intensities are normalized by the highest intensities of individual proteins within the lines. Of note are the well-segregated signals between HAPLNs and CSPGs that are indicated by arrows (HAPLNs in green and CSPGs in magenta, respectively). The colocalized signals are indicated by arrowheads. Acan, aggrecan; Bcan, brevican. Scale bar = 30 µm in the left panels and 10 µm in the right panels.