Knockdown of SOX2OT inhibits the malignant biological behaviors of glioblastoma stem cells via up-regulating the expression of miR-194-5p and miR-122

Abstract

Background: Accumulating evidence has highlighted the potential role of long non-coding RNAs (lncRNAs) in the biological behaviors of glioblastoma stem cells (GSCs). Here, we elucidated the function and possible molecular mechanisms of the effect of lncRNA-SOX2OT on the biological behaviors of GSCs.

Results: Real-time PCR demonstrated that SOX2OT expression was up-regulated in glioma tissues and GSCs. Knockdown of SOX2OT inhibited the proliferation, migration and invasion of GSCs, and promoted GSCs apoptosis. MiR-194-5p and miR-122 were down-regulated in human glioma tissues and GSCs, and miR-194-5p and miR-122 respectively exerted tumor-suppressive functions by inhibiting the proliferation, migration and invasion of GSCs, while promoting GSCs apoptosis. Knockdown of SOX2OT significantly increased the expression of miR-194-5p and miR-122 in GSCs. Dual-luciferase reporter assay revealed that SOX2OT bound to both miR-194-5p and miR-122. SOX3 and TDGF-1 were up-regulated in human glioma tissues and GSCs. Knockdown of SOX3 inhibited the proliferation, migration and invasion of GSCs, promoted GSCs apoptosis, and decreased TDGF-1 mRNA and protein expression through direct binding to the TDGF-1 promoter. Over-expression of miR-194-5p and miR-122 decreased the mRNA and protein expression of SOX3 by targeting its 3'UTR. Knockdown of TDGF-1 inhibited the proliferation, migration and invasion of GSCs, promoted GSCs apoptosis, and inhibited the JAK/STAT signaling pathway. Furthermore, SOX3 knockdown also inhibited the SOX2OT expression through direct binding to the SOX2OT promoter and formed a positive feedback loop.

Conclusion: This study is the first to demonstrate that the SOX2OT-miR-194-5p/miR-122-SOX3-TDGF-1 pathway forms a positive feedback loop and regulates the biological behaviors of GSCs, and these findings might provide a novel strategy for glioma treatment.

Keywords: Glioma; SOX2OT; SOX3; TDGF-1; miR-122; miR-194-5p.

Fig. 1

The SOX2OT expression and effects of SOX2OT in glioma. a The expression of SOX2OT in normal brain tissues (NBTs) and glioma tissues of different grades. Data are presented as the mean ± SD (NBTs (n = 5), Grade I(n = 5),Grade II(n = 5), Grade III (n = 8), Grade IV (n = 8)). ^** P < 0.01 vs. NBTs group; ^## P < 0.01 vs. Grade I group; ^△△ P < 0.01 vs. Grade II group; ^ΨΨ P < 0.01 vs. Grade III group. b The expression of SOX2OT in human astrocytes (HA), glioblastoma cell lines (U87 and U251) and glioblastoma stem cells (GSC-U87, GSC-U251). Data are presented as the mean ± SD (n = 5, each group). ^** P < 0.01 vs. HA group; ^## P < 0.01 vs. U87 group; ^ΨΨ P < 0.01 vs. U251 group. c The expression of SOX2OT after cells transfection with sh-SOX2OT plasmids. d CCK-8 assay was used to measure the effect of SOX2OT on the proliferation of GSC-U87 and GSC-U251. e The apoptotic percentages of GSC-U87 and GSC-U251 were detected after SOX2OT knockdown. f Transwell assays were used to measure the effect of SOX2OT on cell migration and invasion of GSC-U87 and GSC-U251. Data represent as the mean ± SD (n = 5,each group). ^** P < 0.01 vs. sh-NC group. Scale bars represent 40 μm

Fig. 2

MiR-194-5p and miR-122 exerted tumor-suppressive functions in GSCs. a The expression of miR-194-5p in normal brain tissues (NBTs) and glioma tissues of different grades. Data are presented as the mean ± SD (NBTs (n = 5), Grade I (n = 5), Grade II (n = 5), Grade III (n = 8), Grade IV (n = 8)). ^** P < 0.01 vs. NBTs group; ^## P < 0.01 vs. Grade I group; ^△△ P < 0.01 vs. Grade II group; ^ΨΨ P < 0.01 vs. Grade III group. The expression of miR-194-5p in human astrocytes (HA), glioblastoma cell lines (U87 and U251) and glioblastoma stem cells (GSC-U87, GSC-U251). Data are presented as the mean ± SD (n = 5,each group). ^** P < 0.01 vs. HA group; ^## P < 0.01 vs. U87 group; ^△△ P < 0.01 vs. U251 group. b The expression of miR-194-5p after SOX2OT knockdown in GSC-U87 and GSC-U251 cells. ^** P < 0.01 vs. sh-NC group. c The predicted miR-194-5p binding site in the SOX2OT sequence (SOX2OT-Wt) and the designed mutant sequence of miR-194-5p binding site (SOX2OT-Mut) are indicated. Relative luciferase activity was conducted after cells were transfected with SOX2OT-Wt or SOX2OT-Mut. Data were presented as the mean ± SD (n = 5, each group). ^** P < 0.01 vs. SOX2OT-Wt + Agomir-194-5p-NC. d CCK-8 assay was used to measure the effect of miR-194-5p on the proliferation of GSC-U87 and GSC-U251 cells. e The apoptotic percentages of GSC-U87 and GSC-U251 cells were detected after miR-194-5p over-expression or inhibition. f Transwell assays was used to measure the effect of miR-194-5p on the migration and invasion of GSC-U87 and GSC-U251 cells. Scale bars represent 40 μm. Data are presented as the mean ± SD (n = 5, each group).^** P < 0.01 vs. Agomir-194-5p-NC group; ^## P < 0.01 vs. Antagomir-194-5p-NC group. g The expression of miR-122 in normal brain tissues(NBTs) and glioma tissues of different grades. Data are presented as the mean ± SD (NBTs (n = 5), Grade I (n = 5), Grade II(n = 5), Grade III (n = 8), Grade IV (n = 8)). ^** P < 0.01 vs. NBTs group; ^## P < 0.01 vs. Grade I group; ^△△ P < 0.01 vs. Grade II group; ^ΨΨ P < 0.01 vs. Grade III group. The expression of miR-122 in HA cells, glioblastoma cell lines(U87 and U251) and glioblastoma stem cells (GSC-U87, GSC-U251). Data are presented as the mean ± SD (n = 5,each group). ^** P < 0.01 vs. HA group; ^## P < 0.01 vs. U87 group; ^△△ P < 0.01 vs. U251 group. h The expression of miR-122 with SOX2OT knockdown in GSC-U87 and GSC-U251 cells. ^** P < 0.01 vs. sh-NC group. i The predicted miR-122 binding sites in the SOX2OT sequence (SOX2OT-Wt) and the designed mutant sequence of miR-122 binding site (SOX2OT-Mut) are indicated. Relative luciferase activity was conducted after cells were transfected with SOX2OT-Wt or SOX2OT-Mut. Data were presented as the mean ± SD (n = 5, each group). ^** P < 0.01 vs. SOX2OT-Wt + Agomir-122-NC. j CCK-8 assay was used to measure the effect of miR-122 on the proliferation of GSC-U87 and GSC-U251 cells. k The apoptotic percentages of GSC-U87 and GSC-U251 cells were detected after miR-122 over-expression or inhibition. l Transwell assays were used to measure the effect of miR-122 on the migration and invasion of GSC-U87 and GSC-U251 cells. Scale bars represent 40 μm. Data are presented as the mean ± SD (n = 5, each group). ^** P < 0.01 vs. Agomir-122-NC group; ^## P < 0.01 vs. Antagomir-122-NC group

Fig. 3

MiR-194-5p and miR-122 mediated the tumor-suppressive effects of SOX2OT knockdown on GSCs. a CCK-8 assay was used to measure the effect of SOX2OT and miR-194-5p on the proliferation of GSC-U87 and GSC-U251 cells. b Flow cytometry analysis to evaluate the effect of SOX2OT and miR-194-5p on the apoptosis of GSC-U87 and GSC-U251 cells. c Transwell assays was used to measure the effect of SOX2OT and miR-194-5p on the migration and invasion of GSC-U87 and GSC-U251 cells. Data are presented as the mean ± SD (n = 5, each group).^** P < 0.01 vs. sh-NC + agomir-194-5p-NC. d CCK-8 assay was used to measure the effect of SOX2OT and miR-122 on the proliferation of GSC-U87 and GSC-U251 cells. e Flow cytometry analysis to evaluate the effect of SOX2OT and miR-122 on the apoptosis of GSC-U87 and GSC-U251 cells. f Transwell assays was used to measure the effect of SOX2OT and miR-122 on the migration and invasion of GSC-U87 and GSC-U251 cells. Scale bars represent 40 μm. Data are presented as the mean ± SD (n = 5,each group) . ^** P < 0.01 vs. sh-NC + agomir-122-NC

Fig. 4

SOX3 endogenous expression and its effect on the proliferation, migration, invasion and apoptosis in GSCs, as well as the expression of SOX2OT and TDGF-1. a The SOX3 protein expression levels in normal brain tissues (NBTs), low-grade glioma tissues (WHO I-II) and high-grade glioma tissues (WHO III-IV) are shown. Data are presented as the mean ± SD (n = 3,each group). ^** P < 0.01 vs. NBTs group; ^## P < 0.01 vs. low-grade glioma tissues group. bThe expression level of SOX3 in human astrocytes (HA) and glioblastoma cell lines (U87 and U251). c The expression of SOX3 in glioblastoma cell lines (U87 and U251) and glioblastoma stem cells (GSC-U87, GSC-U251). d CCK-8 assay was used to measure the effect of SOX3 on the proliferation of GSC-U87 and GSC-U251 cells. e The apoptotic percentages of GSC-U87 and GSC-U251 were detected after SOX3 over-expression or knockdown. f Transwell assays was used to measure the effect of SOX3 on cell migration and invasion of GSC-U87 and GSC-U251 cells. Scale bars represent 40 μm. Data are presented as the mean ± SD (n = 5, each group). ^** P < 0.01 vs. SOX3(+)NC group; ^## P < 0.01 vs. SOX3(−)NC group. g Relative expression of SOX2OT with the over-expressing or silencing SOX3. h SOX3 bound to the promoter of SOX2OT in GSC-U87 and GSC-U251 cells. Schematic representation of the human SOX2OT promoter region 3000 bp upstream of the transcription start site (TSS), which was designated as +1. Putative SOX3 binding site was indicated. PCR was conducted with the resulting precipitated DNA. i Weatern blot assay was used to detect the expression of TDGF-1 after SOX3 over-expression or knockdown. j SOX3 bound to the promoter of TDGF-1 in GSC-U87 and GSC-U251 cells. Schematic representation of the human TDGF-1 promoter region 3000 bp upstream of the transcription start site (TSS), which was designated as +1. Putative SOX3 binding site was indicated. PCR was conducted with the resulting precipitated DNA

Fig. 5

The SOX3 expression regulated by SOX2OT, miR-194-5p and miR-122. a Real-time PCR and (b) Weatern blot assay were used to detect the SOX3 expression after SOX2OT knockdown. c Real-time PCR and (d) Weatern blot assay were used to detect the SOX3 expression after miR-194-5p over-expression or knockdown. e Real-time PCR and (f) Weatern blot assay were used to detect the SOX3 expression regulated by SOX2OT and miR-194-5p. g The predicted miR-194-5p binding sites in the 3’UTR region of SOX3 (SOX3–3’UTR-Wt) and the designed mutant sequence (SOX3–3’UTR-Mut) are indicated. Relative luciferase activity was conducted after cells were transfected with SOX3–3’UTR-Wt or SOX3–3’UTR-Mut. Data were presented as the mean ± SD (n = 5,each group).^** P < 0.01 vs. SOX3–3’UTR-Wt + Agomir-194-5p-NC group. h Real-time PCR and (i) Weatern blot assay were used to detect the SOX3 expression after miR-122 over-expression or knockdown. j Real-time PCR and (k)Weatern blot assay were used to detect the SOX3 expression regulated by SOX2OT and miR-122. l The predicted miR-122 binding sites in the 3’UTR region of SOX3 (SOX3–3’UTR-Wt) and the designed mutant sequence (SOX3–3’UTR-Mut) are indicated. Data were presented as the mean ± SD (n = 5,each group). ^** P < 0.01 vs. SOX3–3’UTR-Wt + Agomir-122-NC group

Fig. 6

SOX3 mediated tumor-suppressive effects of miR-194-5p and miR-122. a CCK8 assay to evaluate the effect of miR-194-5p and SOX3 on cell proliferation of GSC-U87 and GSC-U251 cells. b Flow cytometry analysis to evaluate the effect of miR-194-5p and SOX3 on cell apoptosis of GSC-U87 and GSC-U251 cells. c Transwell assay to evaluate the effect of miR-194-5p and SOX3 on the cell migration and invasion of GSC-U87 and GSC-U251 cells. Data are presented as the mean ± SD (n = 5, each group). Scale bars represent 40 μm. ^** P < 0.01 vs. AgomiR-194-5p-NC + SOX3(+)NC group, ^## P < 0.01 vs. AgomiR-194-5p + SOX3(+)NC group. d CCK8 assay to evaluate the effect of miR-122 and SOX3 on cell proliferation of GSC-U87 and GSC-U251 cells. e Flow cytometry analysis to evaluate the effect of miR-122 and SOX3 on cell apoptosis of GSC-U87 and GSC-U251 cells. f Transwell assay to evaluate the effect of miR-122 and SOX3 on the cell migration and invasion of GSC-U87 and GSC-U251 cells. Data are presented as the mean ± SD (n = 5, each group). Scale bars represent 40 μm. ^** P < 0.01 vs. AgomiR-122-NC + SOX3(+)NC group, ^## P < 0.01 vs. AgomiR-122 + SOX3(+)NC group

Fig. 7

TDGF-1 endogenous expression and effect on proliferation, migration, invasion and apoptosis of GSCs. a TDGF-1 protein expression levels in normal brain tissues (NBTs), low-grade glioma tissues (WHO I-II) and high-grade glioma tissues (WHO III-IV) are shown. Data are presented as the mean ± SD (n = 3,each group). b The expression of TDGF-1 in human astrocytes (HA) and glioblastoma cell lines (U87 and U251). c The expression of TDGF-1 in glioblastoma cell lines (U87 and U251) and glioblastoma stem cells (GSC-U87, GSC-U251). d CCK-8 assay was used to measure the effect of TDGF-1 on the proliferation of GSC-U87 and GSC-U251 cells. e The apoptotic percentages of GSC-U87 and GSC-U251 were detected after TDGF-1 over-expression or knockdown. f Transwell assays were used to measure the effect of TDGF-1 on cell migration and invasion of GSC-U87 and GSC-U251 cells. Data are presented as the mean ± SD (n = 5, each group). ^** P < 0.01 vs. TDGF-1(+)NC, ^## P < 0.01 vs. TDGF-1(−)NC. g Western blot assay of the p-JAK-1/JAK-1 and p-STAT3/STAT3 expression regulated by TDGF-1. Data are presented as the mean ± SD (n = 5, each group). ^** P < 0.01 vs. TDGF-1(+)NC, ^## P < 0.01 vs. TDGF-1(−)NC. h Weatern blot assay were used to detect the TDGF-1 expression regulated by miR-194-5p and SOX3. Data are presented as the mean ± SD (n = 5, each group). ^** P < 0.01 vs. AgomiR-194-5p-NC + SOX3(+)NC group, ^## P < 0.01 vs. AgomiR-194-5p + SOX3(+)NC group. i Weatern blot assay were used to detect the TDGF-1 expression regulated by miR-122 and SOX3. Data are presented as the mean ± SD (n = 5, each group). ^** P < 0.01 vs. AgomiR-122-NC + SOX3(+)NC group, ^## P < 0.01 vs. AgomiR-122 + SOX3(+)NC group

Fig. 8

Tumor xenograft studies. a The nude mice carrying tumors from respective groups were shown. The sample tumors from respective group were shown. b Tumor growth curves were shown. Tumor volume was calculated every 5 days after injection, and the tumor was taken after 45 days. c Survival curves from representive nude mice injected into the right striatum were shown (n = 8, each group)

Fig. 9

The schematic diagram of the oncogenic role of SOX2OT in GSCs