Fibroblast exosomal TFAP2C induced by chitosan oligosaccharides promotes peripheral axon regeneration via the miR-132-5p/CAMKK1 axis

Abstract

Chitosan and its degradation product, oligosaccharides, have been shown to facilitate peripheral nerve regeneration. However, the underlying mechanisms are not well understood. In this study, we analyzed the protein expression profiles in sciatic nerves after injury using proteomics. A group of proteins related to exosome packaging and transport is up-regulated by chitosan oligosaccharides (COS), implying that exosomes are involved in COS-induced peripheral nerve regeneration. In fact, exosomes derived from fibroblasts (f-EXOs) treated with COS significantly promoted axon extension and regeneration. Exosomal protein identification and functional studies, revealed that TFAP2C is a key factor in neurite outgrowth induced by COS-f-EXOs. Furthermore, we showed that TFAP2C targets the pri-miRNA-132 gene and represses miR-132-5p expression in dorsal root ganglion neurons. Camkk1 is a downstream substrate of miR-132-5p that positively affects axon extension. In rats, miR-132-5p antagomir stimulates CAMKK1 expression and improves axon regeneration and functional recovery in sciatic nerves after injury. Our data reveal the mechanism for COS in axon regeneration, that is COS induce fibroblasts to produce TFAP2C-enriched EXOs, which are then transferred into axons to promote axon regeneration via miR-132-5p/CAMKK1. Moreover, these results show a new facet of fibroblasts in axon regeneration in peripheral nerves.

Keywords: TFAP2C; axon regeneration; chitosan oligosaccharides; fibroblast exosomes; peripheral nerves.

Graphical abstract

Fig. 1

Fibroblasts derived exosomes are involved in COS-induced axon elongation. a The workflow for iTRAQ analysis of sciatic nerve samples from rats. 10-mm sciatic nerve defects were bridged with COS- or saline-filled silicone tubes. The proximal regenerated nerves at lesion sites were harvested at 6 h, 12 h, 1 d, 4 d, and 7 d post-surgery (n = 3). The samples were resolved by SDS-PAGE, followed by trypsin digestion and iTRAQ labeling, and then subjected to strong cation exchange fractionation and LC-MS/MS analysis.b The number of differentially expressed proteins (p < 0.05) induced by COS at different time points.c Significant Gene Ontology analysis of differentially expressed proteins induced by COS. The dot size represents functional significance. Gene number represents the number of proteins in each term. Red is a strong positive correlation, and blue is a strong negative correlation.d Heatmap showing relative expression patterns of proteins with selected enriched cluster terms.e Immunofluorescence staining for Vimentin, S100, NF, CD68, RTN3, and RAB21 in the cross sections of sciatic nerves at 4 days post-bridging with silicon tubes. Scar bar = 200 μm.

Fig. 2

Exosomes derived from COS-treated fibroblasts facilitate neurite outgrowth. a Transmission Electron Microscopy (TEM) analysis of exosomes secreted by fibroblasts. Scale bar = 500 nm b Particle size distribution of the vesicles secreted from fibroblasts was measured by Nanoparticle Tracking Analysis (NTA).c Exosome markers CD9 and CD81 were analyzed by western blot. Calnexin was used as a negative control. EXOs: exosomes; TCL: total cell lysates.d Fluorescent microscopy analysis showing fibroblasts derived exosomes were internalized by cultured DRG neurons. Exosomes derived from fibroblasts were labeled by PKH26 (red) and co-cultured with DRG neurons for 24 h, then DRG neurons were subjected to immunofluorescence analysis using an anti-NF antibody (green). Scale bar = 100 μm.e Neurite outgrowth was stimulated by exosomes from COS-treated fibroblasts. Exosomes from fibroblasts (f-EXOs) treated with or without COS were prepared and co-cultured with DRG neurons for 24 h or 48 h. The neurite outgrowth was evaluated by immunofluorescence analysis by using an anti-NF antibody. Scale bar = 200 μm.f Quantification of neurite length as shown in (e). n = 8.g Exosomes derived from COS-treated fibroblasts promote neurite outgrowth. DRG neurons were incubated with exosomes at different concentrations as indicated for 24 h. Neurite outgrowth was measured by immunofluorescence analysis by using an anti-NF antibody. Scale bar = 200 μm h Quantification of neurite length as shown in (g). n = 8.Error bar represents ±SD. **P < 0.01, two-way ANOVA analysis.

Fig. 3

Exosomes derived from COS-treated fibroblasts promote axon regeneration and functional recovery in rats with injured sciatic nerves. a Axon regeneration in sciatic nerves was promoted by exosomes from COS-treated fibroblasts (COS-f-EXOs). The transected sciatic nerves were bridged with silicone tubes filled with saline, f-EXOs, or COS-f-EXOs. 14 days post-surgery, the proximal nerve stumps were subjected to immunofluorescence analysis by using an anti-NF antibody. Red arrows indicate the start sites of axon regeneration, and white arrows indicate the axon growth cones. Scale bar = 500 μm b The length of regenerated axons as shown in (a). n = 5.c Walking track analysis showing COS-f-EXOs promote functional recovery in rats with injured sciatic nerves. RH means right hind paw. LH means left hind paw. n = 6.d-e The maximum contact mean area (d) and the mean intensity (e) were measured according to the walking track data.Error bar represents ± SD.*P < 0.05, **P < 0.01, ***P < 0.001, one-way ANOVA test (b) and two-way ANOVA test (d, e).

Fig. 4

Exosomal protein identification. Proteins were isolated from exosomes and subjected to mass spectrometry analysis. The up-regulated proteins enriched in exosomes from COS-treated fibroblasts were further analyzed by bioinformatics analysis. Differentially expressed proteins were defined as P < 0.05 with at least 1.2-fold changes. a Cellular components. b Heat map and functional annotations. c Network analysis for Tfap2c.

Fig. 5

TFAP2C is a positive regulator for neurite outgrowth. a TFAP2C induces neurite outgrowth. Cultured DRG neurons were transfected with the plasmid expressing Tfap2c by electroporation. 24 h post-transfection, neurons were subjected to immunofluorescence analysis using an anti-NF antibody. Scale bar = 500 μm. n = 8.b-c Knockdown of Tfap2c. Cultured DRG neurons were transfected with siRNA-Con, or siRNA-1, or -2, or -3 by electroporation. 48 h post-transfection, neurons were harvested for gene expression analysis by qRT-PCR (b) or western blot (c). Gapdh was used as an internal control in qRT-PCR analysis, while Actin was used a loading control in western blot analysis. n = 3.d Knockdown of Tfap2c represses neurite outgrowth. Cultured DRG neurons were transfected with siRNA-2 by electroporation. 48 h post-transfection, DRG neurons were subjected to immunofluorescence analysis using an anti-NF antibody. Scale bar = 500 μm. n = 8.e Preparation of TFAP2C-enriched exosomes from fibroblasts. Primary fibroblasts were transfected with the plasmid expressing Tfap2c by electroporation. 48 h post-transfection, cell culture medium was collected for preparing exosomes. Exosomal TFAP2C was analyzed by western blot. CD9 was used as a loading control. n = 3.f TFAP2C-enriched exosomes stimulate neurite outgrowth. Cultured DRG neurons were treated with TFAP2C-enriched exosomes for 24 h, and neurite outgrowth was analyzed by immunofluorescence staining with an anti-NF antibody. Scale bar = 500 μm. n = 8.g Preparation of TFAP2C-deficient exosomes from fibroblasts. Primary fibroblasts were transfected with Tfap2c siRNA-2. 72 h post-transfection, cell culture medium was collected for preparing exosomes. Exosomal TFAP2C was analyzed by western blot. CD9 was used as a loading control. n = 3.h TFAP2C-deficient exosomes impede neurite outgrowth. Cultured DRG neurons were treated with TFAP2C-knockdown exosomes for 24 h, and neurite outgrowth was analyzed by immunofluorescence staining with an anti-NF antibody. Scale bar = 500 μm. n = 8.EV: empty vector. Error bar represents ±SD.*P < 0.05, ***P < 0.001, Student's t-test.

Fig. 6

TFAP2C targets miR-132-5p to regulate neurite extension. a Identification of pri-miRNA-132 as a potential downstream effector of TFAP2C by bioinformatics analysis.b TFAP2C inhibits luciferase activity drove by the sequences in pri-miRNA-132 gene. 4 sequences containing TFAP2C binding sites were synthesized and incorporated into luciferase vector. 293T cells were co-transfected with the constructed luciferase plasmids and the plasmid expressing Tfap2c. 24 h post-transfection, cells were harvested for luciferase activity assay. EV: empty vector. n = 3.c The expression profiles of miR-132-3p and -5p in cultured DRG neurons. The expression of miRNA was analyzed by qRT-PCR, U6 was used as an internal control. n = 4.d Overexpression of TFAP2C represses miR-132-5p expression. Cultured DRG neurons were transfected with the plasmid expressing Tfap2c by electroporation. 24 h post-transfection, total RNA was extracted for analyzing the expression of Tfap2c and miR-132-5p by qRT-PCR. Gapdh and U6 were used as internal controls. EV: empty vector. n = 3.e-f The effects of miR-132-5p on neurite outgrowth. Cultured DRG neurons were transfected with miR-132-5p mimic (e) or inhibitor (f) by electroporation. 24 h post-transfection, neurite outgrowth was analyzed by immunofluorescence using an anti-NF antibody. Scale bar = 500 μm. n = 8.Error bar represents ±SD. **P < 0.01, ***P < 0.001, Student's t-test.

Fig. 7

Camkk1 is a substrate of miR-132-5p for its repression on neurite outgrowth. a Bioinformatics analysis showing miR-132-5p interacts with the 3′-untranslated region (UTR) of Camkk1.b-c The effects of miR-132-5p on Camkk1 expression. Cultured DRG neurons were transfected with miR-132-5p mimic or inhibitor by electroporation. 36 h post-transfection, Camkk1 expression was analyzed by qRT-PCR (b) or western blot (c). Gapdh was used as an internal control. Actin was used as a loading control. n = 3.d CAMKK1 stimulates neurite outgrowth. Cultured DRG neurons were transfected with the plasmid expressing Camkk1 by electroporation. 24 h post-transfection, the neurite length was analyzed by immunostaining for NF. Scale bar = 200 μm. n = 8.e-fCamkk1 knockdown. DRG neurons were transfected with the Camkk1 siRNAs by electroporation. 48 h post-transfection, the mRNA levels and the protein levels of Camkk1 were analyzed by qRT-PCR (e) and western blot (f), respectively. Gapdh was used as an internal control in qRT-PCR analysis and Actin was used as a loading control in western blot analysis.gCamkk1 knockdown retards neurite outgrowth. DRG neurons were transfected with the Camkk1 siRNAs by electroporation. 36 h post-transfection, the neurite outgrowth was analyzed by immunostaining for NF. Scale bar = 200 μm. n = 8.hCamkk1 knockdown abolishes miR-132-5p inhibitor induced neurite outgrowth. DRG neurons were transfected with miR-132-5p inhibitor and Camkk1 siRNAs as indicated by electroporation. 36 h post-transfection, the neurite length was evaluated by immunostaining for NF. Scale bar = 200 μm. n = 8.Data are expressed as means ± SD. **P < 0.01, **P < 0.001, Student's t-test was used to analyze statistical difference in (b, d, e, g). One-way ANOVA was used to analyze statistical difference in (h).

Fig. 8

Inhibition of miR-132-5p promotes sciatic nerve functional recovery after injury. 10-mm sciatic nerve defects were constructed in rats and bridged with silicone tubes filled with miR-132-5p antagomir or control.a The expression of miR-132-5p in sciatic nerves was reduced by miR-132-5p antagomir. qRT-PCR was used to analyze miR-132-5p expression and U6 was used as an internal control. n = 3.b-c The expression of Camkk1 in sciatic nerves was increased by miR-132-5p antagomir. Camkk1 expression was analyzed by qRT-PCR (b; n = 4–6) or western blot (c; n = 3). Gapdh was used as an internal control in qRT-PCR, Actin was used as a loading control in western blot.d miR-132-5p antagomir promotes axon regeneration in sciatic nerves after injury. Sciatic nerves were harvested at 14 d post-surgery and subjected to immunofluorescence analysis by using an anti-NF antibody. Scale bar = 1000 μm.e Sciatic functional index (SFI) calculated from the walking track analysis at different timepoints after surgery. 14D and 1 M mean 14 days, and 1 month, respectively. n = 6.f-h Electrophysiological assay in sciatic nerves. The assay was carried out at 4 weeks post-surgery and the representative recordings were shown in (f). The motor nerve conduction velocity (g) in sciatic nerves and the compound muscle action potential (CMAP) recordings (h) were detected in the injured side of animals. n = 6.Data are expressed as means ± SD. **P < 0.01, ***P < 0.001, Student's t-test.

Fig. 9

The in vivo effects of f-EXOs and COS-f-EXOs on the expressions of miR-132-5p and CAMKK1. Transected sciatic nerves were bridged with silicone tubes filled with saline, f-EXOs, or COS-f-EXOs. 7 days post-surgery, the proximal nerve stumps were collected for further analysis.a COS-f-EXOs repress the expression of miR-132-5p. qRT-PCR was used to analyze miR-132-5p expression and U6 was used as an internal control. n = 3.b COS-f-EXOs stimulate the expression of Camkk1. Camkk1 expression was analyzed by qRT-PCR and Gapdh was used as an internal control. n = 3.c COS-f-EXOs increase CAMKK1 expression and phosphorylated Akt (p-Akt). Proteins were analyzed by western blot and Actin was used as a loading control. n = 3.d-e Quantification of CAMKK1 (d) and p-Akt (e). n = 3.f Schematic model for elucidating the molecular mechanism of COS-f-EXOs induced neurite outgrowth.Data are expressed as means ± SD. *P < 0.05, **P < 0.01, ***P < 0.001, one-way ANOVA.