SRSF10 inhibits biogenesis of circ-ATXN1 to regulate glioma angiogenesis via miR-526b-3p/MMP2 pathway

Abstract

Background: Angiogenesis plays an important role in the progress of glioma. RNA-binding proteins (RBPs) and circular RNAs (circRNAs), dysregulated in various tumors, have been verified to mediate diverse biological behaviors including angiogenesis.

Methods: Quantitative real-time PCR (qRT-PCR) and western blot were performed to detect the expression of SRSF10, circ-ATXN1, miR-526b-3p, and MMP2/VEGFA. The potential function of SRSF10/circ-ATXN1/miR-526b-3p axis in glioma-associated endothelial cells (GECs) angiogenesis was further studied.

Results: SRSF10 and circ-ATXN1 were significantly upregulated in GECs compared with astrocyte-associated endothelial cells (AECs). Knockdown of SRSF10 or circ-ATXN1 significantly inhibited cell viability, migration and tube formation of GECs where knockdown of SRSF10 exerted its function by inhibiting the formation of circ-ATXN1. Moreover, the combined knockdown of SRSF10 and circ-ATXN1 significantly enhanced the inhibitory effects on cell viability, migration and tube formation of GECs, compared with knockdown of SRSF10 and circ-ATXN1, respectively. MiR-526b-3p was downregulated in GECs. Circ-ATXN1 functionally targeted miR-526b-3p in an RNA-induced silencing complex. Up-regulation of miR-526b-3p inhibited cell viability, migration and tube formation of GECs. Furthermore, miR-526b-3p affected the angiogenesis of GECs via negatively regulating the expression of MMP2/VEGFA.

Conclusion: SRSF10/circ-ATXN1/miR-526b-3p axis played a crucial role in regulating the angiogenesis of GECs. The above findings provided new targets for anti-angiogenic therapy in glioma.

Keywords: Angiogenesis; Circ-ATXN1; Glioma; Glioma associated endothelial cells; SRSF10; miR-526b-3p.

Fig. 1

Silencing of SRSF10 affected the proliferation, migration and tube formation of GECs. a The relative protein expression of SRSF10 in NBTs and LGGTs, HGGTs were evaluated by Western-blot assay. GAPDH was used as an endogenous control. Data represent mean ± SD (n = 5, each group; **P < 0.01). b The relative protein expression of SRSF10 in AECs and GECs were evaluated by Western-blot assay. GAPDH was used as an endogenous control. Data represent mean ± SD (n = 3, each group; **P < 0.01). c The effect of SRSF10 on the viability of GECs was determined by CCK-8 assay. Data represent mean ± SD (n = 3, each group; **P < 0.01). d The effect of SRSF10 on the migration of GECs was assessed by Transwell assay. Data represent mean ± SD (n = 3, each group; **P < 0.01). Scale bar represents 30 μm. e The effect of SRSF10 on the tube formation of GECs was evaluated by Matrigel tube formation assay (Black arrow, tube structures and grey arrow, tube branches). Data represent mean ± SD (n = 3, each group; **P < 0.01). Scale bar represents 30 μm. f The effect of SRSF10 on the expression of MMP2. Data represent mean ± SD (n = 3, each group; **P < 0.01). g The effect of SRSF10 on the expression of VEGFA. Data represent mean ± SD (n = 3, each group; **P < 0.01). h The activity of MMP2 in the GECs after downregulation of SRSF10 was detected by gelatin zymography. Data represent mean ± SD (n = 3, each group; *P < 0.05)

Fig. 2

Silencing of circ-ATXN1 affected the proliferation, migration and tube formation of GECs. a A schematic representation of how circ-ATXN1 arosed from the ATXN1 gene as determined by scanning ATXN1 genomic DNA and circBase. Sanger sequencing validated the sequence on the junction sites of circRNA-ATXN1. The black arrow indicated the head-to-tail splicing sites of circRNA-ATXN1. b The existence of circ-ATXN1 in GECs. c The expression of circ-ATXN1 in GECs treated with RNase R. Data represent mean ± SD (n = 3, each group; **P < 0.01, ^##P < 0.01). d The expression of line-ATXN1 in GECs treated with RNase R. Data represent mean ± SD (n = 3, each group; **P < 0.01, ^##P < 0.01). e The relative expression of circ-ATXN1 in AECs and GECs was detected by qRT-PCR. GAPDH was used as an endogenous control. Data represent mean ± SD (n = 3, each group; *P < 0.05) f The relative expression of line-ATXN1 was detected in AECs and GECs by qRT-PCR. GAPDH was used as an endogenous control. Data represent mean ± SD (n = 3, each group). g FISH was performed to investigate expression and location of circ-ATXN1 in AECs and GECs (green, circ-ATXN1; blue, DAPI nuclear staining). h Effect of circ-ATXN1 on the cell viability of GECs was detected by CCK-8 assay. Data represent mean ± SD (n = 3, each group; **P < 0.01). i Effect of circ-ATXN1 on the migration of GECs was detected by Transwell assay. Data represent mean ± SD (n = 3, each group; **P < 0.01). j Effect of circ-ATXN1 on tube formation of GECs was measured by Matrigel tube formation assay (Black arrow, tube structures; Grey arrow, tube branches). Data represent mean ± SD (n = 3, each group, **P < 0.01). Scale bar represents 30 μm. k-l Effect of circ-ATXN1 knockdown on expression of MMP2 and VEGFA was detected by Western blot analysis. Data represent mean ± SD (n = 3, each group; **P < 0.01). m The activity of MMP2 in the GECs after downregulation of circ-ATXN1 was detected by gelatin zymography. Data represent mean ± SD (n = 3, each group; *P < 0.05)

Fig. 3

Circ-ATXN1 was involved in SRSF10-mediated angiogenesis of GECs. a Relative enrichment of circ-ATXN1 in anti-IgG and anti-SRSF10 were detected by RNA immunoprecipitation assay. Data represent mean ± SD (n = 3, each group; **P < 0.01, ^##P < 0.01). b Effect of SRSF10 knockdown on expression of circ-ATXN1 was detected by qRT-PCR. Data represent mean ± SD (n = 3, each group; ** P < 0.01). c Effects of circ-ATXN1 and SRSF10 on the cell viability of GECs were detected by CCK-8 assay. Data represent mean ± SD (n = 3, each group; **P < 0.01, ^##P < 0.01, ^&&P < 0.01). d Effects of circ-ATXN1 and SRSF10 on the migration of GECs were detected by Transwell assay. Data represent mean ± SD (n = 3, each group; **P < 0.01, ^##P < 0.01, ^&&P < 0.01). e Effects of circ-ATXN1 and SRSF10 on tube formation of GECs were measured by Matrigel tube formation assay. (Black arrow, tube structures; Grey arrow, tube branches). Data represent mean ± SD (n = 3, each group; **P < 0.01, ^##P < 0.01, ^&&P < 0.01). Scale bar represents 30 μm. f-g Effects of circ-ATXN1 and SRSF10 on expressions of MMP2 and VEGFA were detected by western blot. Data represent mean ± SD (n = 3, each group; **P < 0.01, ^##P < 0.01, ^&&P < 0.01). h Effects of circ-ATXN1 and SRSF10 on activity of MMP2 were detected by gelatin zymography. Data represent mean ± SD (n = 3, each group; *P < 0.05)

Fig. 4

MiR-526b-3p affected the proliferation, migration and tube formation of GECs via binding to MMP2. a Relative expression levels of miR-526b-3p in AECs and GECs were determined by qRT-PCR. Data represent mean ± SD (n = 3, each group; **P < 0.01). b FISH was performed to investigate expression and location of miR-526b-3p in AECs and GECs (red, miR-526b-3p; blue, DAPI nuclear staining). c The effect of miR-526b-3p on the viability of GECs was determined by CCK-8 assay. Data represent mean ± SD (n = 3, each group; **P < 0.01, ^##P < 0.01). d The effect of miR-526b-3p on the migration of GECs was assessed by Transwell assay. Data represent mean ± SD (n = 3, each group; **P < 0.01, ^##P < 0.01). e The effect of miR-526b-3p on the tube formation of GECs was evaluated by Matrigel tube formation assay (Black arrow, tube structures; Grey arrow, tube branches). Data represent mean ± SD (n = 3, each group; **P < 0.01, ^##P < 0.01). Scale bar represents 30 μm. f-g The effect of miR-526b-3p on the expression of MMP2 and VEGFA. Data represent mean ± SD (n = 3, each group; **P < 0.01, ^##P < 0.01). h Effects of miR-526b-3p on the activity of MMP2. Data represent mean ± SD (n = 3, each group; **P < 0.01, ^##P < 0.01). i The putative miR-526b-3p binding site in MMP2 (MMP2-Wt) and the designated mutant sequence (MMP2-Mut) were illustrated. j Luciferase reporter assay of HEK293T cells co-transfected with MMP2-Wt or MMP2-Mut and miR-526b-3p or miR-526b-3p-NC. Data represent mean ± SD (n = 3, each group; **P < 0.01). k The putative miR-526b-3p binding site in VEGFA (VEGFA -Wt) and the designated mutant sequence (VEGFA-Mut) were illustrated. l Luciferase reporter assay of HEK293T cells co-transfected with VEGFA -Wt or VEGFA -Mut and miR-526b-3p or miR-526b-3p-NC. Data represent mean ± SD (n = 3, each group; **P < 0.01)

Fig. 5

Circ-ATXN1 facilitated angiogenesis of GECs via binding to miR-526b-3p. a The putative miR-526b-3p binding site in circ-ATXN1 (circ-ATXN1-Wt) and the designated mutant sequence (circ-ATXN1-Mut) were illustrated. b Luciferase reporter assay of HEK293T cells co-transfected with circ-ATXN1-Wt or circ-ATXN1-Mut and miR-526b-3p or miR-526b-3p-NC. Data represent mean ± SD (n = 3, each group; **P < 0.01). c-d The interaction between circ-ATXN1 and miR-526b-3p was determined by RIP assay. Circ-ATXN1 expression and miR-526b-3p enrichment were measured using real-time qPCR. Data represent mean ± SD (n = 3, each group; **P < 0.01, ^##P < 0.01). e The co-effects of circ-ATXN1 and miR-526b-3p on the viability of GECs were evaluated by CCK-8 assay. f The co-effects of circ-ATXN1 and miR-526b-3p on the migration of GECs were evaluated by Transwell assay. g The co-effects of circ-ATXN1 and miR-526b-3p on the tube formation of GECs were evaluated by Matrigel tube formation assay (Black arrow, tube structures; Grey arrow, tube branches). Data are presented as the mean ± SD (n = 3, each group; **P < 0.01 vs sh-circ-ATXN1-NC + pre-miR-526b-3p group. Scale bar represents 30 μm). h-i The co-effects of circ-ATXN1 and miR-526b-3p on the expression of MMP2 and VEGFA. Data represent mean ± SD (n = 3, each group; **P < 0.01). j The co-effects of circ-ATXN1 and miR-526b-3p on the activity of MMP2. Data represent mean ± SD (n = 3, each group; **P < 0.01)

Fig. 6

SRSF10 knockdown combined with circ-ATXN1 knockdown, and miR-256-3p overexpression suppressed the angiogenesis in vivo. a Matrigel plug assay was used to measure angiogenesis. b The hemoglobin content was measured. Data are presented as mean ± SD (n = 3, each group), *P < 0.05, **P < 0.01vs. Control group; ^##P < 0.01 vs. sh-SRSF10 group; ^&&P < 0.01 vs. sh-circ-ATXN1 group; ^@@P < 0.01 vs. pre-miR-526-3p group. c Schematic illustration of the mechanism of SRAF10/circ-ATXN1/miR-526b-3p regulating glioma angiogenesis