HuR expression is significantly reduced in P10 nerves, compared to NB and P5, as shown by (a) qPCR, (b) Western blotting of total protein extracts and (c) cytosolic and nuclear protein fractions. (d) IHC showing HuR expression (red) in Schwann cells from P5 teased sciatic nerves (arrows). TuJ1 labels axons (green) and DAPI labels nuclei (blue). (b,c) Gapdh/β-actin: loading control. (c) Gapdh: cytoplasmic marker; Histone H3: nuclear marker. Data: mean ± s.e.m; *p<0.01.

(a) Cell scratch assay showing that migration on laminin-coated coverslips is significantly reduced in sh HuR-infected cells (**) compared to sh Control-infected cells (*). (b) Graphs showing the closure of the gap distance with time after culture on laminin-coated coverslips. (c) Talin ICC (red) showing morphological differences between sh HuR-infected and sh Control-infected cells plated onto laminin. (d) HuR-silenced cells have a smaller surface area and fewer lamellipodia on laminin substrate, but not on PDL. (e) HuR-silenced cells have shorter axial lamellipodia and fewer radial lamellipodia than sh Control-infected cells on laminin. (f) qPCR showing expression of different transcripts after culture on laminin with time, expressed as fold change relative to cells replated onto PDL dishes. (g) RIP-qPCR analysis shows significant enrichment of several motility/morphology-associated genes bound to HuR in cells plated onto laminin compared to cells plated onto PDL. (h) HuR silencing leads to a significant reduction in expression of these genes.

(a, b) HuR silencing significantly reduces (a) NRG1- and (b) TGFβ-induced proliferation compared to sh Control-infected cells as determined by BrdU incorporation (BrdU⁺ cells: red; Dapi⁺ Nuclei: blue). (c) qPCR showing expression of different transcripts after NRG1 or TGFβ treatment with time, expressed as fold change relative to untreated cells. (d) RIP-qPCR analysis shows significant enrichment of proliferation-associated genes bound to HuR in NRG1- and TGFβ-treated cells compared to cells cultured in MM. (e, f) HuR silencing leads to a significant reduction in expression of proliferation-associated genes after (e) NRG1 and (f) TGFβ treatment. Data: mean ± s.e.m; *p<0.01

(a, b) HuR silencing leads to a significant reduction in TGFβ-apoptosis as seen by (a) number of surviving cells identified by nuclear condensation viewed by DAPI (blue) and number of apoptotic cells viewed by Tunel labelling (green), and (b) Western blot of PARP cleavage and active caspase-3, 48h after treatment. (c) qPCR showing expression of different transcripts after TGFβ treatment with time, expressed as fold change relative to untreated cells. (d) RIP-qPCR analysis shows significant enrichment of apoptosis-associated genes bound to HuR in TGFβ-treated cells compared to cells cultured in SM. (e) HuR silencing leads to a significant reduction in expression of these genes. (f) Western blot showing nucleo-cytoplasmic translocation of HuR after TGFβ treatment. (g, h) Treatment with SB203580, but not with UO126, prevents TGFβ-induced (g) nucleo-cytoplasmic translocation of HuR and (h) apoptosis. (f,g) β-actin; loading control. (f) Gapdh: cytoplasmic marker; Histone H3: nuclear marker. Cyto: cytoplasmic fractions; Nucl: Nuclear fractions. Data: mean ± s.e.m; *p<0.01

(a, b) HuR silencing increases myelination in Schwann cell-DRG co-cultures, as seen by (a) number of MBP⁺ myelin segments (red), and (b) qPCR analysis of myelination-related genes. (c, d) Enforced HuR expression induced a small but significant decrease in myelination in Schwann cell-DRG co-cultures, as seen by (c) number of MBP⁺ myelin segments (red), and (d) qPCR analysis of myelination-related genes. Gapdh; loading control. Data: mean ± s.e.m; *p<0.01

(a) Heatmap showing expression of the top transcripts most significantly bound to HuR compared to control IgG in NB nerves (upper panel) and P5 nerves (bottom panel) in 4 different replicates (1–4). (b) Heatmap showing relative enrichment of top 50 mRNA targets of HuR, compared to control IgG (HuR IP/IgG IP, Column 1) and input mRNA (HuR IP/Input, Column 2). The colour scale indicates the degree of enrichment (green-red ratio scale). (c) Venn diagram showing the overlap of HuR- bound transcripts with a fold change ≥ 1.5 in lysates isolated from NB and P5 nerves. (d) Enriched Gene Ontology (GO) classification of the genes identified by the RIP-chip analyses in NB (blue histograms) and P5 nerves (red histograms).

(a) Western blot showing nucleo-cytoplasmic translocation of HuR after replating of Schwann cells onto laminin. (b, c) Western blot showing that treatment with SB203580 (to block p38 phosphorylation) (b) prevents the nucleo-cytoplasmic translocation of HuR after 4h culture onto laminin and (c) reduces the migration rate on laminin . (a,b) β-actin: loading control; (a) Gapdh: cytoplasmic marker; Histone H3: nuclear marker. Cyto: cytoplasmic fractions; Nucl: Nuclear fractions. Data: mean ± s.e.m; *p<0.01

(a) Western blot showing nucleo-cytoplasmic translocation of HuR after NRG1 and TGFβ treatment. (b) Treatment with UO126 (to block ERK 1/2 phosphorylation) or LY-294002 (to block AKT phosphorylation) prevents NRG1-induced nucleo-cytoplasmic translocation of HuR (left panel) and proliferation (right panel). (c) Treatment with UO126 or SB203580 prevents TGFβ-induced nucleo-cytoplasmic translocation of HuR (left panel) and proliferation (right panel). (a-c) β-actin; loading control. (a) Gapdh: cytoplasmic marker; Histone H3: nuclear marker. Cyto: cytoplasmic fractions; Nucl: Nuclear fractions. Data: mean ± s.e.m; *p<0.01

(a) RIP-qPCR analysis shows binding of HuR to Pmp22 mRNA both in NB and P5 nerves. (b) Schematic of Pmp22 mRNA depicting the 5′ UTR, Coding Region and 3′UTR. In silico analysis revealed several computationally predicted HuR motifs (arrowheads), and biotinylated transcripts spanning different mRNA regions (A-E) were incubated with P5 sciatic nerve lysates and the interaction assessed by biotin pull-down assays followed by Western blot analysis. HuR only formed complexes with a specific region of the 3′ UTR, but not with the 5′ UTR or coding region. No interaction was seen with a control biotinylated RNA corresponding to the 3′UTR of GAPDH mRNA, which is not a target of HuR (data not shown). (c) RIP-qPCR analysis shows binding of HuR to several myelination-related genes in postnatal nerves. (d) Western blot showing an increased expression of several myelination genes after HuR silencing under basal (MM) and myelinogenic conditions (dibutryl cAMP treatment). (e, f) qPCR showing an increased expression of myelination- related genes after HuR silencing (e) in MM, and (f) after cAMP treatment. (g) qPCR showing the half-life of myelination-related genes in control and HuR-silenced cAMP-treated cells.

(a, b) Schematic diagram of the promoter region of HuR showing (a) NFkB consensus binding sites (green ovals) and regions analysed (A1 and A2) and (b) SMAD 2/3 consensus binding sites (yellow ovals) and regions analysed (B1 and B2). ChIP analysis shows a decreased binding of p65 and SMAD 2/3 to the regions analysed, in P10 nerves compared to NB and P5 nerves. (c) Western blot showing reduced p65 and SMAD 2/3 expression in nuclear fractions of P10 nerves compared to NB and P5 nerves. (d-f) NRG1 increases HuR transcription through NFkB activation. (d) Treatment with NRG1 increases HuR mRNA (upper panel) and HuR protein levels (bottom panel), as seen by qPCR and Western blotting respectively. (e) ChIP analysis shows that p65 is recruited to the HuR promoter regions (indicated in a) after treatment with NRG1 for 1 hr, but not in cells cultured in MM, or with NRG1 for 12 hrs. (f) Western blot showing that treatment with the NFkB inhibitor BAY11-7082, which prevents the nuclear translocation of p65, inhibits the NRG1-induced increase in HuR levels. (g-i) TGFβ increases HuR transcription through SMAD2/3 activation. (g) Treatment with TGFβ increases HuR mRNA (upper panel) and HuR protein levels (bottom panel), as seen by qPCR and Western blotting respectively. (h) ChIP analysis shows that SMAD2/3 is recruited to the HuR promoter regions (indicated in b) after treatment with TGFβ for 1 hr, but not in cells cultured in MM, or with TGFβ for 12 hrs. (i) Western blot showing that adenoviral infection of cells with SMAD7 adenovirus (Ad-SMAD7), which prevents the nuclear translocation of SMAD2/3, inhibits the TGFβ-induced increase in HuR levels. (j) Combined treatment of NRG1 and TGFβ of Schwann cells plated onto laminin did not lead to an enhanced upregulation of HuR, induced by the two growth factors alone, as seen by qPCR (upper panel) and Western blot (bottom panel), 2h after treatment. (k, l) Western blots showing that adenoviral infection of cells with Egr2 adenovirus (Ad-Egr2), (k) decreases HuR levels after 48 hrs, (l) an effect reduced with treatment with the proteosome inhibitor MG132. β-actin/Gapdh; loading control. Nuc: Nuclear fractions. Data: mean ± s.e.m; *p<0.01