Fig 2: Direct cold stress exposure increases RNA-binding motif 3 protein expression in primary mouse skeletal muscle endothelial cells and inhibits cell proliferation. (A) Representative immunoblots and densitometry analyses for cellular protein levels of the cold-shock RNA-binding motif 3 protein (RBM3) in primary mouse skeletal muscle endothelial cells (SMECs) exposed for 24 h to cold stress (28°C) or maintained at 37°C. ß-actin is used as a loading control. Significant difference between groups: **, p = 0.01. Data is represented as means ± SEM from three independent experiments, each with n = 6 samples per condition of exposure and temperature. (B) SMECs proliferation at 37°C and 28°C was quantified at Days 0, 1 and 2. Data illustrates one representative proliferation assay out of three independent experiments with n = 4–8 samples per time-point in each. Significantly different from Day 0: ****, p = 0.0001. Significantly different from the corresponding 28°C time-point measurement: ##, p = 0.01; ####, p = 0.0001.

Fig 3: Knocking-down RBM3 alters the neuronal activity pattern throughout 24 h. A, We knocked down RBM3 in the cultures by shRNA expression (see Materials and Methods). Alternatively, a scrambled shRNA was expressed (Scr), as a control. Exemplary RBM3 stainings are shown. The reporter for the shRNA virus is GFP. Scale bar, 10 µm. B, To reveal how RBM3 shRNA is affecting the RBM3 abundance, we performed immunostaining on shRNA treatments. Each dot represents the mean of an image. Error bar indicates mean ± SEM. N = 3 independent experiments; n = 15 images. A significant difference was detected (Mann–Whitney test): p = 0.0021. **p < 0.005. C, To determine the firing patterns, we performed Ca2+ experiments on RBM3 knocked down neurons, exactly as in Figure 1. Scale bar, 50 μm. D, The signals over the entire 45 s recordings from the cells shown in C are plotted, at the four different time points. E, The activity score is plotted for the two exemplary neurons, normalized to their respective medians. F, To reveal the average firing patterns, 56 neurons were measured in each condition, from four independent experiments (with 2-4 different wells measured per experiment; mean ± SEM). To calculate the correlation between two firing patterns, we performed a Pearson's correlation test. A significant anticorrelation was observed (r = −0.870, p = 0.024).

Fig 4: Synaptic activity and morphology are modified on RBM3 shRNA treatment. A, E, To measure synaptic vesicle recycling, we relied on the Syt1 uptake assay, applying the antibodies for 15 or 60 min, in different experiments. The assays were performed on DIV18 neurons, at 18:00. GFP is used as a reporter for the expression of shRNA or a scrambled control sequence. Typical images are shown. Scale bar, 20 µm. B, F, The Syt1 intensity in 15 and 60 min incubations were measured. Symbols represent the mean of each image. Error bar indicates mean ± SEM. N = 4 independent experiments; n = 20 images. The recycling vesicle pool size (measured with 60 min incubations) as well the synaptic activity (measured with 15 min incubations) are significantly larger for the RBM3 knockdowns (Mann–Whitney tests). B, p = 0.0138. F, p = 0.0350. C, G, We blocked network activity using TTX and then performed the same vesicle-labeling assay as in A and E. Typical images are shown. Scale bar, 20 µm. D, H, The Syt1 intensity was measured in the TTX condition in 15 and 60 min incubations. Symbols represent the mean of each image. Error bar indicates mean ± SEM. N = 4 independent experiments; n = 20 images. A quantification of the signals revealed no significant differences (Mann–Whitney tests). D, p = 0.8065. H, p = 0.6362. I, To monitor changes in the size or morphology of synapses, neurons were immunostained for Syph (presynaptic marker) and Homer1 (postsynaptic marker). Scale bar, 2.5 µm. White arrows indicate the synapses. J, K, The intensity of the Syph and Homer1 stainings, respectively. Symbols represent the mean intensity of each image. Error bar indicates mean ± SEM. N = 4 independent experiments; n = 20 images. The Homer1 intensity of RBM3 KD is significantly lower than in the controls (Mann–Whitney tests). J, p = 0.6980. K, p = 0.0402. *p < 0.05. ns, not significant.

Fig 5: RBM3 protects against ROT-induced apoptosis. SH-SY5Y cells were transfected with the empty vector (Veh) or the RBM3-expressing vector (RBM3) and treated with 0.5 µmol/L ROT. A, An MTT assay was performed to examine the effects of RBM3 overexpression ROT cytotoxicity after ROT treatment for a time course of 0, 24, 36 and 48 h. B, Cells were treated with ROT for 24 h before Western blotting for the indicated proteins. The black and red arrows indicate endogenous RBM3 and myc-tagged RBM3, respectively. C, D, The levels of cl. PARP and Bcl-2 were quantified by densitometry and normalized to ß-actin and Bax, respectively. E, TUNEL staining was performed after 24 h ROT exposure to evaluate the protective effects of RBM3 overexpression on ROT-induced neurotoxicity. ***P < 0.001 vs vehicle-transfected cells