Fig 2: BDNF-induced formation of ribosomal complexes can be visualized by proximity ligation assay. a Representative images show PLA signal from RPL24 and RPS6 proteins in growth cones of Smn+/+;SMN2tgtg and Smn-/-;SMN2tgtg motoneurons following 1-min BDNF stimulation. b Increased PLA signal is detectable in growth cones of BDNF-stimulated Smn+/+;SMN2tgtg (****P < 0.0001; n = 65–69 cells from 4 independent experiments), but not in Smn-/-;SMN2tgtg neurons (n.s., P = 0.9315; n = 39 cells from 3 independent experiments). As a negative control, Smn+/+;SMN2tgtg neurons were pretreated with puromycin before and during the BDNF pulse (a). Only a low PLA signal is detectable (n.s., P = 0.6882; n = 13 cells from 1 experiment). c Omitting either the RPL24 or the RPS6 primary antibody abolishes the PLA signal. Growth cones were outlined using GFP (see the method part for PLA assay). Data are presented in scatter dot plot; error bars represent mean ± SEM. Statistical analyses were done by Two-way ANOVA with Tukey multiple comparison post-hoc test

Fig 3: Ribosome responsiveness to the BDNF/TrkB signaling as well as ribosome/ER tethering are impaired in Smn-deficient axon terminals. a BDNF-stimulated Smn+/+;SMN2tgtg and Smn-/-;SMN2tgtg motoneurons expressing mCherry-KDEL were stained against RPL24, RPS6 and mCherry-KDEL. Growth cones were imaged by SIM. White boxes indicate enlarged ROIs within growth cones. In control but not Smn-/-;SMN2tgtg motoneurons, BDNF induces formation of RPL24/RPS6 co-clusters that colocalize with the axonal ER at 10 s and 1-min poststimulation. b Smn-/-;SMN2tgtg motoneurons do not exhibit increased number of RPL24/RPS6 co-clusters upon stimulation (n.s., P = 0.9999; n = 21–23 cells), in contrast to Smn+/+;SMN2tgtg (*P = 0.0147 for no BDNF vs. 10-s BDNF; *P = 0.0210 for no BDNF vs. 1-min BDNF; n = 13–16 cells from 3 independent experiments). In control experiments, either RPL24 or RPS6 antibodies were omitted to exclude the crosstalk between channels. Colocalization analysis detected almost no co-clusters in the growth cones of control neurons (see also Additional file 1: Fig. S3). c In Smn+/+;SMN2tgtg motoneurons, the number of RPL24/RPS6 co-clusters that colocalize with ER increases significantly within 10 s and 1-min of stimulation (**P = 0.0089 for no BDNF vs. 10 s BDNF; **P = 0.0092 for no BDNF vs. 1-min BDNF; n = 13–16 cells from 3 independent experiments). In Smn-/-;SMN2tgtg motoneurons the number of RPL24/RPS6 co-clusters that associate with ER does not increase upon stimulation (n.s., P = 0.7601; n = 21–24 cells from 3 independent experiments). In all graphs, the average number of co-clusters are shown per growth cone. All representative images are from maximum projection of five 0.12-µm z-stacks. Data are presented in scatter dot plot; error bars represent mean ± SEM. Statistical analyses were done by one-way ANOVA with Dunn’s post-test