Fig 2: Syndecan-4 localizes to the nuclear envelope during muscle cell differentiation.A) Muscle cells were transiently transfected with SDC4-HA, fixed with 4% PFA and immunostained with mouse anti-HA, followed by Alexa 546-conjugated goat anti-mouse (red) before confocal microscopy analysis. Labelling was found throughout the cytosol, as well as perinuclear along the nuclear envelope. Scale bar: 10 µm. B-I) Immuno-EM localization of syndecan-4 HA in reticular compartments (B-C) and the nuclear envelope (D-I). Muscle cells transiently transfected cells with SDC4-HA were induced to differentiation before preparation for immuno EM. Thawed cryosections were labelled using anti-HA antibody and 15 nm protein A gold. B-C) Transfected cells showed strong labelling in the membrane of reticular compartments (R) which, as shown in C), often was localized close to the cell nucleus (N). Scale bars: 500 nm. D) Overview of a differentiated cell showing multiple nuclear profiles. Scale bar as indicated. E-F) Higher magnifications of framed areas in D. Arrow heads indicate syndecan-4 labelling localized to the nuclear envelope. N: Nucleus. Scale bars as indicated. G) Overview of parts of a multinucleated cell. H-I) Higher magnifications of the framed areas demonstrate labelling associated with the nuclear envelope. Arrow heads in I indicate the inner and outer nuclear membrane. N: Nucleus. Scale bars as indicated.

Fig 3: The inhibitory effect of astrocytes on neurite extension requires Syndecan-4. Cell tracker green-labeled CAD cells (5 × 104 cells/cm2) were seeded onto a plate or co-cultured on top of a fixed-monolayer of DITNC1 astrocytes. Neurite extension was then induced by serum deprivation for 24 h (1 × 105 cells/cm2). To evaluate the participation of heparan sulfate chains in the inhibition of neurite outgrowth, DITNC1 cells were pre-treated with Heparitinase III (Hase III; 0.5 mU; 3 h at 37°C) or pre-incubated with Heparin (Hep; 400 µg/ml; 30 min). To block aVß3-integrin, astrocytes were incubated with anti-ß3 integrin antibodies (anti-ß3; 5 µg/ml; 1 h; 37°C). (A) Representative microphotographs of different conditions. Quantification of neurite length (µm) after (B) Hase III treatment or (C) Hep incubation. (D) siRNA silencing of Syndecan-4 protein in whole cell lysates. DITNC1 cells that were either non-transfected (NT), transfected with siRNA control (siCTRL) or with siRNA targeting Syndecan-4 (siSDC4) were evaluated by immunoblotting. Actin was used as a loading control. The band intensities were quantified by ImageJ software and normalized to actin. (E) Quantification of neurite length (µm) after Syndecan-4 silencing. For each quantification (B,C,E), neurites of at least 100 cells per condition were evaluated by using NeuronJ plug-in for ImageJ. Arrowheads in (A) indicate neurites growing over the DITNC1 astrocytes. In all graphs data are expressed as mean ± SEM (n = 3; *p < 0.05; **p < 0.01; n.s. non-significant, assessed by Mann–Whitney’s test).

Fig 4: Syndecan-4 is a negative regulator of myogenesis.Confluent cells grown in differentiation media were either I) incubated for 24 h, II) treated with the cell-penetrating peptide Arg9-Syn-4 cyt for 24 h, III) transfected with SDC4 RNAi and incubated 24 h, and IV) transfected with SDC4 RNAi and treated with the cell-penetrating peptide Arg9-Syn-4 cyt for 24 h. A) Bars show the relative mRNA expression of syndecan-4 compared to negative control cells measured after 24h in differentiation media. The data in A is presented as the average mean of at least two independent experiments seeded out in duplicates, performed in technical triplicates, ± SD. Asterisks denote significant differences (***p<0.001). B) The fusion index (FI) (the number of cells with more than 2 nuclei) was calculated based on scoring at least six randomly chosen regions with nuclei and myotubes stained as in Fig 8A, in two independent experiments. Asterisk denote significant differences between negative control cells, compared to SDC4 RNAi and SDC4 RNAi + Arg9Syn-4 cyt treated cells (***p<0.001, n>100 cells).

Fig 5: Sdc4 and Vangl2 expression during mouse development. (A-E) Whole-mount in situ hybridization with an antisense RNA probe for Sdc4. Embryos at E8.5, dorsal view (A), and E9.5, lateral view (B) and rostral view (C). Transverse sections (D,E) are at the levels indicated on embryos in A and B, respectively. (F-H) ß-Galactosidase activity in Sdc4lacZ/+ embryos at E9.0. Lateral view (F) shows expression in otic vesicle (arrowhead) and hindgut (arrow). Dorsal view of caudal region (G) shows expression in non-neural ectoderm on the outside of the neural fold of the posterior neuropore (arrowhead) and overlying recently closed neural tube (arrow). Section (H) is from the level indicated by the line in G. Arrowhead indicates the non-neural ectoderm. (I,J) Immunostaining of transverse sections through wild-type E9.0 embryos at the hindbrain level using anti-Sdc4 (green, arrows) and anti-pan-cadherin (red). Expression is detected in the neural folds (I, arrow), foregut (J, arrowhead) and hindgut (J, arrow). (K) X-Gal staining of the whole cochlea from Sdc4lacZ/+ embryo at E18.5. Inset: section showing lacZ expression in the outer hair cells (bracket) and in the inner hair cell (arrow). (L) Immunostaining of Sdc4 (green) on transverse section through the otic vesicle in wild-type E9.0 embryo. (M,N) Sections through the closed neural tube (M) and open PNP (N) of an E9.5 embryo subjected to whole-mount in situ hybridization using an antisense RNA probe for Vangl2. cm, cephalic mesenchyme; hf, head folds; hg, hindgut; nt, neural tube; ov, otic vesicle; PNP, posterior neuropore. Scale bars: 500 µm in A,B; 300 µm in G; 10 µm in I,L; 250 µm in K.