Hypoxic-preconditioned mesenchymal stem cell-derived small extracellular vesicles promote the recovery of spinal cord injury by affecting the phenotype of astrocytes through the miR-21/JAK2/STAT3 pathway

Abstract

Background: Secondary injury after spinal cord injury (SCI) is a major obstacle to their neurological recovery. Among them, changes in astrocyte phenotype regulate secondary injury dominated by neuroinflammation. Hypoxia-preconditioned mesenchymal stem cells (MSCs)-derived extracellular vesicle (H-EV) plays a multifaceted role in secondary injury by interacting with cellular components and signaling pathways. They possess anti-inflammatory properties, regulate oxidative stress, and modulate apoptotic pathways, promoting cell survival and reducing neuronal loss. Given the unique aspects of secondary injury, H-EV shows promise as a therapeutic approach to mitigate its devastating consequences. Our study aimed to determine whether H-EV could promote SCI repair by altering the phenotype of astrocytes.

Methods: Rat bone marrow MSCs (BMSCs) and EVs secreted by them were extracted and characterized. After the SCI model was successfully constructed, EV and H-EV were administered into the tail vein of the rats, respectively, and then their motor function was evaluated by the Basso-Beattie-Bresnahan (BBB) score, Catwalk footprint analysis, and electrophysiological monitoring. The lesion size of the spinal cord was evaluated by hematoxylin-eosin (HE) staining. The key point was to use glial fibrillary acidic protein (GFAP) as a marker of reactive astrocytes to co-localize with A1-type marker complement C3 and A2-type marker S100A10, respectively, to observe phenotypic changes in astrocytes within tissues. The western blot (WB) of the spinal cord was also used to verify the results. We also compared the efficacy differences in apoptosis and inflammatory responses using terminal deoxynucleotidyl transferase dUTP terminal labeling (TUNEL) assay, WB, and enzyme-linked immunosorbent assay (ELISA). Experiments in vitro were also performed to verify the results. Subsequently, we performed microRNA (miRNA) sequencing analysis of EV and H-EV and carried out a series of knockdown and overexpression experiments to further validate the mechanism by which miRNA in H-EV plays a role in promoting astrocyte phenotypic changes, as well as the regulated signaling pathways, using WB both in vivo and in vitro.

Results: Our findings suggest that H-EV is more effective than EV in the recovery of motor function, anti-apoptosis, and anti-inflammatory effects after SCI, both in vivo and in vitro. More importantly, H-EV promoted the conversion of A1 astrocytes into A2 astrocytes more than EV. Moreover, miR-21, which was found to be highly expressed in H-EV by miRNA sequencing results, was also demonstrated to influence changes in astrocyte phenotype through a series of knockdown and overexpression experiments. At the same time, we also found that H-EV might affect astrocyte phenotypic alterations by delivering miR-21 targeting the JAK2/STAT3 signaling pathway.

Conclusion: H-EV exerts neuroprotective effects by delivering miR-21 to promote astrocyte transformation from the A1 phenotype to the A2 phenotype, providing new targets and ideas for the treatment of SCI.

Keywords: astrocytes; extracellular vesicles; miR-21; neuroinflammation; spinal cord injury.

FIGURE 1

Identification of BMSCs. (A) Morphological image of BMSCs adherent growth under the light microscope. (B) Flow cytometric analysis of two positive (CD29 and CD90) and two negative surface protein markers (CD34 and CD45) of BMSCs. (C) Alizarin red, Orley red O, and Alcian blue staining were used to identify the osteogenic, lipogenic, and chondrogenic differentiation potential of BMSCs, respectively (Scale bar: 100 μm).

FIGURE 2

Identification of EV and H‐EV and astrocyte tracing in vitro. (A) Morphological images of EV and H‐EV under TEM. (B) Particle sizes of EV and H‐EV were analyzed by NTA. (C) Western blot analysis of three protein markers (TSG101, CD63, and CD9) of EV and H‐EV. (D) Quantitative analysis of TSG101, CD63, and CD9 protein expression levels relative to β‐Actin in (C), respectively (n = 3/group). (E) The uptake of CFSE‐labeled EV with green fluorescence by CTX‐TNA2 was observed under the fluorescence microscope (*P < 0.05, **P < 0.01, ***P < 0.001, ****P < 0.0001, ns: Non‐significant; Scale bar: 100 μm).

FIGURE 3

EV and H‐EV promote functional recovery after SCI in vivo. (A) BBB score of sham, SCI + PBS, SCI + EV, and SCI + H‐EV groups on days 1, 3, 7, 14, 21, and 28 (n = 6/group). (B) Representative Catwalk footprint analysis of sham, SCI + PBS, SCI + EV, and SCI + H‐EV groups at day 28 (n = 6/group). (C–E) Three parameters of footprint analysis were used to quantify the motion function. (C) Stride length. (D) Regularity index. (E) The base of support. (F) Representative MEP analysis of sham, SCI + PBS, SCI + EV, and SCI + H‐EV groups at day 28 (n = 6/group). (G–H) Two parameters of MEP analysis were used to quantify the muscle strength of hind limbs. (G) AUC. (H) Amplitude. (I) Representative the gross morphology of the spinal cord of sham, SCI + PBS, SCI + EV, and SCI + H‐EV groups at day 28 (n = 6/group). (J) Representative the HE staining of the spinal cord of sham, SCI + PBS, SCI + EV, and SCI + H‐EV groups at day 28 (n = 6/group). (K) The size of the lesion was quantified by HE staining (n = 6/group) (*P < 0.05, **P < 0.01, ***P < 0.001, ****P < 0.0001, vs SCI + PBS; #P < 0.05, ##P < 0.01, ###P < 0.001, ####P < 0.0001, vs SCI + EV; ns: non‐significant; scale bar: 250 μm).

FIGURE 4

EV and H‐EV play a role in promoting astrocyte conversion from A1 to A2 in vivo. (A) Representative the immunofluorescence images of A1 astrocytes of the spinal cord in sham, SCI + PBS, SCI + EV, and SCI + H‐EV groups on days 3, 7, and 14 (n = 6/group, DAPI: Blue, GFAP: Green, C3: Far red, white arrow: C3⁺GFAP⁺ A1 astrocytes). (B) Quantitative analysis of C3⁺GFAP⁺/GFAP⁺ in (A). (C) Representative the immunofluorescence images of A2 astrocytes of the spinal cord in sham, SCI + PBS, SCI + EV, and SCI + H‐EV groups at day 3, 7, and 14 (n = 6/group, DAPI (blue), GFAP (green), S100A10 (red), white arrow: S100A10⁺GFAP⁺ A2 astrocytes). (D) Quantitative analysis of S100A10⁺GFAP⁺/GFAP⁺ in (C). (E) Representative western blot analysis of the expression levels of three protein markers of the spinal cord in sham, SCI + PBS, SCI + EV, and SCI + H‐EV groups at day 7 (n = 6/group, C3: A1 Astrocyte, S100A10: A2 Astrocyte, β‐Actin: Internal reference protein). (F) Quantitative analysis of C3 and S100A10 expression levels relative to β‐Actin in (E), respectively (*P < 0.05, **P < 0.01, ***P < 0.001, ****P < 0.0001, scale bar: 50 μm).

FIGURE 5

EV and H‐EV play a role in anti‐inflammatory and anti‐apoptotic effects in vivo. (A) ELISA was used to measure the concentrations of pro‐inflammatory cytokines (TNF‐α and IL‐1β) and anti‐inflammatory cytokines (TGF‐β and IL‐10) of the spinal cord in sham, SCI + PBS, SCI + EV, and SCI + H‐EV groups at day 7 (n = 6/group). (B) Representative the immunofluorescence images of TUNEL staining in sham, SCI + PBS, SCI + EV, and SCI + H‐EV groups at day 7 (n = 6/group, DAPI: Blue, TUNEL: Green, TUNEL‐positive cells: Apoptotic nerve cells). (C) Quantitative analysis of TUNEL‐positive cells as a percentage of the total number of cells in (B). (D) Representative western blot analysis of the expression levels of pro‐apoptotic proteins (Bax and Caspase‐3) and anti‐apoptotic protein (Bcl‐2) of the spinal cord in sham, SCI + PBS, SCI + EV, and SCI + H‐EV groups at day 7 (n = 6/group). (E) Quantitative analysis of Caspase‐3, Bcl‐2, and Bax protein expression levels relative to β‐Actin in (D), respectively (*P < 0.05, **P < 0.01, ***P < 0.001, ****P < 0.0001, ns: Non‐significant, scale bar: 50 μm).

FIGURE 6

EV and H‐EV promote astrocyte phenotype switching from A1 to A2 and anti‐apoptotic effect in vitro. (A) Representative flow cytometric analysis of A1‐related C3 and A2‐related S100A10 markers of CTX‐TNA2 astrocytes in control, ACM + PBS, ACM + EV, and ACM + H‐EV groups (n = 3/group). (B) Quantitative analysis of C3 positive and S100A10 positive cells as a percentage of the total number of CTX‐TNA2 astrocytes in (A), respectively. (C) Representative western blot analysis of the expression levels of CTX‐TNA2 astrocytes A1‐related C3 and A2‐related S100A10 protein markers in control, ACM + PBS, ACM + EV, and ACM + H‐EV groups (n = 3/group, GAPDH: Internal reference protein). (D) Quantitative analysis of C3 and S100A10 protein expression levels relative to GAPDH in (C), respectively. (E) Representative the flow cytometric analysis of Annexin V‐FITC/PI double staining of CTX‐TNA2 astrocytes in control, UVR + PBS, UVR + EV, and UVR + H‐EV groups (n = 3/group, both Annexin V and PI positive cells in the Q2 quadrant: Apoptotic cells). (F) Quantitative analysis of apoptotic cells as a percentage of the total number of cells in (E). (G) Representative western blot analysis of the expression levels of pro‐apoptotic proteins (Bax and Caspase‐3) and anti‐apoptotic protein (Bcl‐2) of CTX‐TNA2 astrocytes in control, UVR + PBS, UVR + EV, and UVR + H‐EV groups (n = 3/group). (H) Quantitative analysis of Caspase‐3, Bcl‐2, and Bax protein expression levels relative to β‐Actin in (G), respectively (*P < 0.05, **P < 0.01, ***P < 0.001, ****P < 0.0001).

FIGURE 7

Mir‐21 is up‐regulated in H‐EV and H‐EV promote astrocyte transformation from A1 to A2 via miR‐21/JAK2/STAT3 pathway. (A) Heat map of the two up‐regulated and three down‐regulated miRNAs between EV and H‐EV (n = 3/group, the P < 0.05 and |log2FoldChange| ≥ 1 standard). (B) qRT‐PCR was used to verify the difference in miR‐21 expression levels between EV and H‐EV (n = 3/group). (C) qRT‐PCR was used to verify the difference in miR‐21 expression levels of CTX‐TNA2 astrocytes in NC‐H‐EV, miR‐21‐H‐EV, IN‐NC‐H‐EV, and miR‐21‐IN‐H‐EV groups (n = 3/group). (D) Representative western blot analysis and relative quantitative of the expression levels of CTX‐TNA2 astrocytes A1‐related C3 and A2‐related S100A10 protein markers in NC‐H‐EV and miR‐21‐H‐EV groups (n = 3/group, β‐Actin: Internal reference protein). (E) Representative western blot analysis and relative quantitative of the expression levels of CTX‐TNA2 astrocytes A1‐related C3 and A2‐related S100A10 protein markers in IN‐NC‐H‐EV and miR‐21‐IN‐H‐EV groups (n = 3/group, β‐Actin: Internal reference protein). (F) Representative western blot analysis and relative quantitative of the expression levels of JAK2, P‐JAK2, STAT3, and P‐STAT3 proteins in NC‐H‐EV and miR‐21‐H‐EV groups (n = 3/group, β‐Actin: Internal reference protein). (G) Representative western blot analysis and relative quantitative of the expression levels of JAK2, P‐JAK2, STAT3, and P‐STAT3 proteins in IN‐NC‐H‐EV and miR‐21‐IN‐H‐EV groups (n = 3/group, β‐Actin: Internal reference protein). (H) Representative western blot analysis and relative quantitative of the expression levels of JAK2, P‐JAK2, STAT3, and P‐STAT3 proteins in control, ACM + PBS, ACM + miR‐21‐H‐EV, and ACM + miR‐21‐H‐EV + AG490 groups (n = 3/group, β‐Actin: Internal reference protein) (*P < 0.05, **P < 0.01, ***P < 0.001, ****P < 0.0001).

FIGURE 8

H‐EV promotes SCI repair by affecting the phenotype of astrocytes through the miR‐21/JAK2/STAT3 pathway. (A) Schematic diagram of the experimental process. (B) The time chart of experiments in vivo.