Apatinib inhibits cell proliferation and migration of osteosarcoma via activating LINC00261/miR-620/PTEN axis

ABSTRACT

Apatinib has been recently identified as a potential treatment option for osteosarcoma (OS). Nonetheless, the molecular mechanism of Apatinib in regulating OS progression remains unclear. To explore the downstream molecules that mediated the tumor-suppressive effect of Apatinib on OS. Expression levels of genes were detected by RT-qPCR and western blot assays. Functional assays including Transwell assay were applied to detect the proliferation, apoptosis and migration of OS cells. Molecular interactions were detected by luciferase reporter assay and RIP assay. Apatinib inhibited the proliferation and migration of OS cells. LINC00261 was down-regulated in OS cells but then up-regulated after the treatment by Apatinib. Silencing LINC00261 abrogated the suppressive effect of Apatinib on OS cell proliferation and migration. MicroRNA-620 (miR-620) could be sponged by LINC00261. Besides, miR-620 was up-regulated in OS cells and Apatinib treatment reduced miR-620 expression. Furthermore, LINC00261 acted as a competitive endogenous RNA (ceRNA) by sequestering miR-620 to up-regulate the expression of phosphatase and tensin homolog (PTEN). Moreover, Apatinib hindered in vitro cell proliferation and migration as well as the in vivo tumorigenesis of OS through LINC00261/miR-620/PTEN axis. Apatinib-enhanced LINC00261 restrained OS via miR-620/PTEN axis, indicating LINC00261 might promote the efficacy of Apatinib on OS.

Background

Osteosarcoma (OS) is a class of primary malignant tumor with high morbidity among children and adolescents, and its mortality is the second highest among the major pediatric tumors [1]. Although the past decades have witnessed a relative increase in the survival time of OS patients due to improved treatment strategies [2,3], the 5-year survival rate of OS patients remains less than 30% [4,5]. Thus, to further study the pathogenesis of OS is particularly important for identifying novel biological targets for the treatment of OS.

Apatinib (YN968D1) is known as a newly-discovered high-affinity selective inhibitor of vascular endothelial growth factor receptor-2 (VEGFR2) tyrosine kinase, which can inhibit the downstream signal transduction of VEGFR2 [6–8]. Numerous studies have proved that Apatinib harbors high anti-tumor potential in various tumors [9–11]. Also, recent studies have demonstrated that Apatinib can repress the progression of OS. For example, Apatinib represses OS cell migration and invasion through reducing PD-1 expression and inactivating STAT3 [12]. Also, Apatinib has been confirmed to induce the autophagy and apoptosis of OS cells through repressing VEGFR2/STAT3/BCL2 signaling [13]. Moreover, a case report suggests that Apatinib can be treated as a new option for the treatment of metastatic OS [14]. However, more efforts remain to be made in unveiling the precise molecular mechanism behind the tumor-suppressive effect of Apatinib on OS.

Long non-coding RNAs (lncRNAs), a set of non-coding transcripts with a length over 200 nucleotides, can exert pivotal functions in the pathogenesis of various cancers [15]. Studies have shown that lncRNAs influence the tumor-related cellular activities including cell proliferation, migration, apoptosis, as well as cell response to drugs [16–20]. Also, the regulatory functions of lncRNAs in OS have been unceasingly revealed during the past decades [21,22]. In addition, the effects of lncRNAs on the efficacy of chemo-therapies on OS have also been revealed. For example, LINC00161 can sensitize OS cells to cisplatin-induced apoptosis via miR-645/IFIT2 axis [23]. Moreover, lncRNA CTA can sensitize OS cells to doxorubicin through inhibiting autophagy [24]. Long intergenic non-protein coding RNA 261 (LINC00261) has been demonstrated as an anti-tumor lncRNA in several cancers such as endometriosis [25], hepatocellular cancer [26], choriocarcinoma [27], and gastric cancer [28]. Also, several reports have shown that LINC00261 may enhance chemo-sensitivity of cancers, such as esophageal cancer and colon cancer [29,30]. However, the role of LINC00261 in OS cells and its impact on Apatinib efficacy on OS have never been investigated.

Phosphatase and tensin homolog (PTEN) has been recognized as a tumor suppressor gene which functions through inactivating PI3K/AKT pathway [31]. Previous studies have also demonstrated that PTEN is associated with the chemosensitivity of cancer cells. For instance, miR-147 silencing sensitizes gastric cancer cells to 5-fluorouracil by up-regulating PTEN [32]. PTEN overexpression can sensitize leukemia cells to arsenious acid and doxorubicin [33,34]. Also, mounting findings have unveiled the tumor-suppressive property and chemo-sensitizing function of PTEN in OS [35–37]. However, the relation of PTEN to Apatinib efficacy in OS has never been established before.

Therefore, this study was conducted to detect the underlying molecular mechanism of Apatinib-exerted suppression on OS development. Consequently, we discovered that Apatinib retarded OS progression through activating LINC00261/miR-620/PTEN pathway. Apatinib Our study might serve as a reference for enhancing the efficacy of Apatinib on OS.

Materials and methods

Cell culture and treatment

Human normal osteoblast cell line (hFOB1.19) and OS cell lines (MG63 and KHOS) were obtained from American Type Culture Collection (ATCC; Manassas, VA, USA). Cells were all cultivated in Dulbecco’s Modified Eagle’s medium (DMEM; Thermo Fisher Scientific, Waltham, MA, USA) supplemented with 10% fatal bovine serum (FBS; Thermo Fisher Scientific) at 37°C with 5% CO₂. OS cells were separately treated with different doses of Apatinib (Sigma-Aldrich, St. Louis, MO, USA) or control DMSO (Sigma-Aldrich) for follow-up experiments. Before the experiments, we selected 48 h as the cell treatment time of Apatinib.

Cell transfection

MG63 or KHOS cells at 80–90% confluence treated with or without Apatinib were plated into 6-well plates and cultivated for one day. The short hairpin RNAs (shRNAs) targeting LINC00261 (sh-LINC00261) or PTEN and the corresponding negative control (sh-NC), along with miR-620 mimics/inhibitor and NC mimics/inhibitor, were purchased from RiboBio (Guangzhou, China). The PTEN-overexpressing plasmids and empty pcDNA3.1 vector (Invitrogen) were constructed by ZonHon Biopharma Institute, Inc. (Changzhou, Jiangsu, China). Transfection of cells was implemented via Lipofectamine 2000 (Invitrogen) and the cells were then harvested 48 h later.

RNA isolation and quantitative RT-PCR (RT-qPCR)

MG63 or KHOS cells under different conditions were processed with TRIzol reagent (Invitrogen) in order to extract total RNAs. Then, the first strand of cDNA was synthesized via Reverse Transcription Kit (Invitrogen). On a Bio-Rad Real-Time PCR System (Bio-Rad Laboratories, Hercules, CA, USA), RT-qPCR was performed through SYBR Green Real-Time PCR Master Mixes (Roche, Mannheim, Germany). GAPDH/U6 was applied as the normalized control. The fold change was evaluated via the 2^−ΔΔCt method.

Cell counting Kit 8 (CCK-8) assay

MG63 or KHOS cells treated with Apatinib (0, 5, 10, 15 or 20 μM) were cultured in 96-well plates. Then, 10 μl of CCK-8 solution (Dojindo, Tokyo, Japan) was added into each well. A microplate reader (Thermo Fisher Scientific) was utilized to measure the absorbance at 450 nm.

Colony formation assay

As for colony formation assay, MG63 or KHOS cells treated with 10 or 15 μM Apatinib or under different transfections together were seeded in 6-well plates and incubated in complete culture medium (Thermo Fisher Scientific) for 14 days. After that, cells were fixed through methanol (Sigma-Aldrich) and stained by crystal violet (Sigma-Aldrich). The stained colonies (≥ 50 cells) were counted manually. All the cells used for colony formation assay were stably transfected.

Flow cytometry analysis

The apoptosis of MG63 and KHOS cells under different conditions was analyzed applying the FITC Annexin V Apoptosis Detection Kit (BD Biosciences, Franklin Lakes, NJ, USA). After being digested by trypsin, cells were washed in pre-cooled PBS and then re-suspended in binding buffer. After that, 10 µL Annexin V-FITC and 5 µL propidium iodide (PI) were applied to double label the cells for 15 min in the dark room under room temperature. Ratio of apoptotic cells was analyzed with the application of FACS Calibur flow cytometer and Cell Quest software (BD Biosciences).

Caspase‐3 activity assay

Cell apoptosis was assessed with the help of Beyotime C1115 Caspase 3 Activity Assay Kit (Beyotime, Nantong, China). Total proteins were isolated from indicated MG63 or KHOS cells, and then added into 96-well plates with the reaction buffer (Invitrogen). Protein extracts were cultured with caspase-3 substrate in provided reaction buffer. Finally, the caspase-3 activity was examined via a microplate reader (Thermo Fisher Scientific) through evaluating the absorbance at 405 nm.

Transwell migration assay

MG63 or KHOS cells (2 × 10⁴) were planted in the upper chamber containing serum-free medium, while the lower chamber were added with DMEM medium (Thermo Fisher Scientific) plus 10% FBS. After cells were incubated for one day, a cotton swab was applied to remove cells from the upper surface of the chamber. Meanwhile, cells migrated to the lower chamber were fixed by methanol solution for 15 min. Crystal violet was adopted to stain the membranes for 10 min, and the migrated cells were observed and counted under an inverted microscope (Bio-Tek, Winooski, VT, USA) at a magnification of 10 × 10.

Subcellular fractionation

Subcellular fractionation was achieved by using a Cytoplasmic and Nuclear RNA Purification Kit (Norgen, Thorold, ON, Canada) as per the user guide. The levels of LINC00261, GAPDH (cytoplasmic control) and U6 (nuclear control) in different fractions were tested via RT-qPCR.

Dual-luciferase reporter assay

The sequence of full-length LINC00261 or PTEN 3ʹ-UTR with wild-type (WT) or mutant (Mut) miR-620 binding sites was synthesized and sub-cloned into pmirGLO dual-luciferase vector (Promega, Madison, WI, USA). LINC00261-WT/Mut or PTEN-WT/Mut vectors were co-transfected with miR-620 mimics or NC mimics into HEK-293 T cells. After 48 h of transfection, Dual Luciferase Report Assay System (Promega, Madison, WI, USA) was adopted to monitor the luciferase activity.

RNA Immunoprecipitation (RIP) assay

MG63 or KHOS cells were lysed by RIP lysis buffer (Millipore, Billerica, MA, USA) containing protease inhibitor (Thermo Fisher Scientific). Then, magnetic beads (Thermo Fisher Scientific) conjugated with anti-Ago2 antibody (Millipore) or control anti-IgG antibody (Millipore) were cultured with cell extracts. Proteinase K (Absin, Shanghai, China) was chosen to digest proteins, followed by the analysis of the precipitated RNAs via RT-qPCR.

Western Blot

The complete proteome was isolated from cells via lysis buffer (Millipore). Then the concentration of protein was determined by utilizing a BCA protein assay kit (Solarbio, Billerica, MA, USA). After being separated by sodium dodecyl sulfate polyacrylamide gel electrophoresis (SDS-PAGE) (Bio-Rad Laboratories, Hercules, CA, USA), proteins were transferred onto PVDF (Bio-Rad Laboratories) membranes. After being blocked with 5% nonfat milk, the membranes were probed overnight with primary antibodies from Abcam (Cambridge, USA) as following: anti-PTEN (ab32199, 1/1000 dilution), anti-cleaved caspase-3 (ab32042, 1/500 dilution), anti-total caspase-3 (ab13847, 1/500 dilution) and anti-GAPDH (ab8245, 1/10,000 dilution) at 4°C, followed by the incubation with corresponding secondary antibodies (ab205718 or ab205719, 1/20,000 dilution). GAPDH was used as the loading control. Blots were visualized via ECL (Pierce Biotechnology, Rockford, IL, USA).

In vivo tumorigenesis model

Eighteen BALB/C nude mice (aged four weeks old) were procured from the National Laboratory Animal Center (Beijing, China) and then randomly divided into six groups. Then, mice in different groups were treated as follows: group 1 and 2 were inoculated with parental MG63 cells, group 3 and 4 were respectively inoculated with sh-NC or sh-LINC00261-transfected MG63 cells, group 5 were injected with MG63 cells that were transfected with sh-LINC00261 and antagomiR-620, and group 6 were inoculated with MG63 cells that were transfected with sh-LINC00261, antagomiR-620 and sh-PTEN; besides, group 1 were treated with saline, while group 2–6 were all treated with Apatinib (50 mg/kg orally daily). Afterward, tumor volume was monitored every fourth day and the mice were sacrificed four weeks later. Then, the formed tumors were excised from mice, and then photographed and weighed. The in vivo study was performed under the approval of the Institutional Animal Care and Use Committee of the First Medical Center of General Hospital of PLA.

Statistical analysis

Results were assessed via GraphPad Prism 6.0 (GraphPad Software, Inc., La Jolla, CA, USA). Data were shown as the means ± SD. All experiments were conducted in triplicate. Student’s t-test or ANOVA (one-way or two-way) was performed to assay the differences. P < 0.05 was considered to be statistically significant.

Results

Apatinib dose-dependence led to inhibited cell proliferation and migration and enhanced cell apoptosis of OS

First, we examined the effect of Apatinib on the functions of OS cells. CCK-8 assay was carried out to determine the appropriate concentration of Apatinib. As a result, treatment of Apatinib at 10 μM resulted in a significant decrease in the viability of MG63 and KHOS at both 24 h and 48 h (Figure 1(a)). On this basis, 10 μM was considered as the appropriate dose of Apatinib treatment. Then, we treated OS cells with 10 or 15 μM Apatinib to detect the impact of Apatinib on OS cells. Through CCK-8 and colony formation assays, it was found that the treatment of Apatinib obstructed the proliferation of two OS cell lines in a dose-dependent manner (Figure 1(b-c)). Conversely, the apoptotic ratio and caspase-3 activity of OS cells were enhanced accompanied by elevated cleaved caspase-3 level under the treatment of Apatinib (Figure 1(d-e) and Supplementary Figure 1a), indicating that Apatinib enhanced the cell apoptosis of OS. In addition, the migration of OS cells was also weakened by Apatinib treatment (figure 1(f)). Altogether, it was confirmed that Apatinib inhibited the proliferation and migration of OS cells and enhanced the cell apoptosis of OS cells.

Figure 1.

Apatinib inhibited cell proliferation, induced cell apoptosis, and restrained cell migration of OS. (a) CCK-8 assay was used to detect the viability of MG63 and KHOS cells under the treatment of Apatinib at the concentration of 0, 5, 10, 15, and 20 μM. (b-c) The proliferation of OS cells with Apatinib treatment (10 or 15 μM) or DMSO treatment (control) was detected by CCK-8 assay and colony formation assay. (d-e) The apoptosis of OS cells under above conditions was assessed by flow cytometry analysis and caspase-3 activity assay. (f) The migration of indicated OS cells was determined by Transwell migration assay. *P < 0.05, **P < 0.01

LINC00261 silencing weakened the suppressive effect of Apatinib on OS cells

Next, we tried to sift out the lncRNAs involved in the regulation of Apatinib on OS. Microarray analysis was conducted in MG63 and KHOS cells treated with Apatinib or DMSO (negative control). Hierarchical clustering showed the differentially expressed lncRNAs upon Apatinib treatment in OS cells, and the top 5 up-regulated lncRNAs were TMPO-AS1, LINC00261, LINC00452, MIR503 HG, and LINC01089 (Figure 2(a)). Then, we found that only LINC00261 was obviously down-regulated in OS cell lines compared with that in the normal hFOB1.19 cells (Figure 2(b)). Hence, we speculated that Apatinib might exert anti-tumor functions through activating LINC00261 in OS cells. To testify such speculation, we performed subsequent assays in OS cells treated with 10 μM Apatinib. RT-qPCR data depicted that LINC00261 was up-regulated by Apatinib in OS cells, while such elevation of LINC00261 could be partially reversed by the transfection of sh-LINC00261 (Figure 2(c)). As expected, the knockdown of LINC00261 could partially restore the proliferation of OS cells attenuated by Apatinib (Figure 2(d-e)). Also, by carrying out flow cytometry assay and detecting the caspase-3 activity, it was found that LINC00261 silencing abrogated the enhancing effect of Apatinib on OS cell apoptosis (figure 2(f-g) and Supplementary Figure 1b). Moreover, the inhibitory effect of Apatinib on OS cell migration was partially counteracted by the silencing of LINC00261 (Figure 2(h)). Collectively, LINC00261 silencing weakened the suppressive effect of Apatinib on OS cells.

Figure 2.

LINC00261 silencing reversed the tumor-suppressive effect of Apatinib on OS cells. (a) Hierarchical clustering showed the differentially expressed lncRNAs under Apatinib treatment compared with DMSO control in MG63 and KHOS cells. The top 5 up-regulated lncRNAs were TMPO-AS1, LINC00261, LINC00452, MIR503HG, and LINC01089. (b) Expression of the above mentioned 5 lncRNAs in OS cell lines and normal cell line (hFOB1.19) was detected by RT-qPCR. (c) Expression of LINC00261 in MG63 and KHOS cells under indicated treatments was detected by RT-qPCR. (d-e) The proliferation of OS cells treated with DMSO or Apatinib or together with the transfection of sh-NC or sh-LINC00261 was detected by CCK-8 and colony formation assays. (f-g) The apoptosis of OS cells with indicated treatments was detected by flow cytometry analysis and caspase-3 activity assay. (h) The migration of OS cells with indicated treatments was detected by Transwell migration assay. *P < 0.05, **P < 0.01

LINC00261 sponged miR-620 in OS

As is largely reported, lncRNAs in cytoplasm can function in various cancers through sequestering microRNAs (miRNAs) [38,39], including in OS [23]. Here, we found that LINC00261 mainly distributed in the cytoplasm of OS cells based on the data of subcellular fractionation assay (Supplementary Figure 1c). Therefore, we speculated that LINC00261 might mediate the efficacy of Apatinib in OS through interacting with certain miRNAs. After using starBase v2.0 (http://starbase.sysu.edu.cn/), 233 miRNAs were identified to have putative interactions with LINC00261. After that, we detected their expressions in OS cell lines through RT-qPCR, finding that the most up-regulated 5 miRNAs in OS cell lines were miR-1270, miR-3611, miR-620, miR-6783-3p, and miR-5009-3p (Figure 3(a)). To probe the relationship between the 5 miRNAs and Apatinib, we detected their expressions in MG63 and KHOS cells under Apatinib treatment. Consequently, only miR-620 expression was apparently down-regulated in both OS cell lines (Figure 3(b)). Previously, multiple studies have suggested that miR-620 may serve as an oncogenic miRNA in different cancer cells [40–42]. Therefore, we chose miR-620 as the target of our following researches. We then identified the miR-620 binding sites in LINC00261 through starBase and also mutated such sites into complementary ones (Figure 3(c)). Then the results of luciferase reporter assays indicated that the luciferase activity of LINC00261 WT was reduced by miR-620 mimics, whereas that of LINC00261 Mut showed no significant variation (Figure 3(d)). RIP assay data further showed that LINC00261 and miR-620 were both enriched in the precipitates of anti-Ago2 (Figure 3(e)). According to the results presented above, our speculation that miR-620 could bind to LINC00261 was proved. Next, the interference efficiency of miR-620 was detected (figure 3(f)). By conducting RT-qPCR assay, it was found that the level of LINC00261 was elevated in OS cells transfected with miR-620 inhibitor (Figure 3(g)). Furthermore, the level of miR-620 reduced by Apatinib in OS cells was partially recovered by the transfection of miR-620 mimics (Figure 3(h)). However, the ectopic expression of miR-620 impaired the expression of LINC00261 induced by Apatinib in OS cells (Figure 3(i)). Importantly, the overexpression of miR-620 also partially reversed the influences of Apatinib treatment on OS cell proliferation, apoptosis and migration (Figure 3(j-m) and Supplementary Figure 1d). Hence, we concluded that LINC00261 sponged miR-620 to mediate the suppression of Apatinib on the malignancy of OS cells.

Figure 3.

LINC00261 sponged miR-620 in OS. (a) Heat map of the RT-qPCR results showed the expression of 233 miRNAs which could interact with LINC00261 in OS cell lines and normal cell line, and 5 most up-regulated miRNAs were miR-1270, miR-3611, miR-620, miR-6783-3p, and miR-5009-3p. (b) Expression of the 5 miRNAs in OS cells treated with Apatinib or DMSO was detected by RT-qPCR. (c) Binding sequences between LINC00261 WT and miR-620 and the sequence of LINC00261 Mut with mutated miR-620 binding sites were presented. (d-e) The interaction between LINC00261 and miR-620 was determined by luciferase reporter assay and RIP assay. (f) MiR-620 interference efficiency in OS cells was determined by RT-qPCR. (g) Expression of LINC00261 under miR-620 silencing in OS cells was detected by RT-qPCR. (h-i) MG63 and KHOS cells were treated with DMSO or Apatinib, or together with the transfection of NC mimics or miR-620 mimics for subsequent assays. Expression of miR-620 or LINC00261 under indicated treatments in OS cells was detected by RT-qPCR. (j-k) The proliferation of indicated OS cells was detected by CCK-8 and colony formation assays. (l) The apoptosis of OS cells in each group was detected by flow cytometry analysis and caspase-3 activity analysis. (m) The migration of OS cells of each group was assessed by Transwell migration assay. *P < 0.05, **P < 0.01, ***P < 0.001. No significance was represented as “n.s.”

Apatinib activated LINC00261/miR-620/PTEN axis in OS

It is well acknowledged that through sponging miRNAs, lncRNAs could protect the downstream target mRNAs from miRNA-caused post-transcriptional repression [43,44]. Therefore, we sought to identify the downstream target of LINC00261/miR-620 signaling in OS. Combining the prediction results of two bioinformatics tools (miRmap and PITA), we identified 493 genes potentially targeted by miR-620 (Figure 4(a)). To figure out which genes were under the regulation of Apatinib, we detected the expression of them by RT-qPCR in two OS cell lines treated with Apatinib or DMSO. As a result, top 5 most up-regulated genes in Apatinib-treated OS cells were sifted out, including SUFU, PTEN, HAUS5, PRKD3, and PTCD3 (Figure 4(a)). Furthermore, we found that among the 5 genes, only PTEN expression was dramatically up-regulated in OS cells transfected with miR-620 inhibitor (Figure 4(b)). PTEN has been demonstrated as a tumor-suppressive gene which could block PI3K/AKT signaling [31]. In addition, we confirmed that PTEN was down-regulated in OS cell lines relative to that in hFOB1.19 cells (Figure 4(c)). Thus, we speculated that PTEN was a downstream target of LINC00261/miR-620 axis in OS. Then the binding sequences between miR-620 and PTEN 3ʹUTR were predicted by starBase, and the mutant sequences of PTEN without miR-620 binding sites were also presented (Figure 4(d)). Expectedly, after conducting luciferase reporter assay, we found that the overexpression of miR-620 attenuated the luciferase activity of PTEN WT rather than PTEN Mut (Figure 4(e)). Meanwhile, RIP analysis validated the co-immunoprecipitation of miR-620, LINC00261 and PTEN by anti-Ago2, which indicated their coexistence in OS cells (figure 4(f)). Also, we found that the silencing of miR-620 increased PTEN mRNA and protein levels in OS cells (Figure 4(g)). Moreover, inhibiting miR-620 reversed the reduced expression of PTEN at mRNA and protein levels due to LINC00261 silencing in Apatinib-treated OS cells (Figure 4(h)). Together, those results suggested that LINC00261 induced PTEN expression in OS cells through sponging miR-620 and Apatinib activated LINC00261/miR-620/PTEN axis in OS.

Figure 4.

Apatinib-activated LINC00261 sponged miR-620 to boost PTEN in OS cells. (a) 493 genes were identified as the targets of miR-620 through the prediction of miRmap and PITA. The 5 most up-regulated genes were SUFU, PTEN, HAUS5, PRKD3, and PTCD3. (b) Expression of above 5 genes in OS cells under the transfection of NC inhibitor or miR-620 inhibitor was detected by RT-qPCR. (c) Expression of PTEN in OS cell lines and normal cell line was detected by RT-qPCR. (d) MiR-620 binding sites on PTEN 3ʹUTR were predicted by starBase, and the sites were mutated in the sequence of PTEN Mut. (e-f) Luciferase reporter assay and RIP assay were used to detect the interaction between miR-620 and PTEN and LINC00261. (g) The mRNA and protein levels of PTEN under miR-620 inhibition in OS cells were detected by RT-qPCR and western blot assays. (h) PTEN mRNA and protein levels in OS cells treated with Apatinib under indicated transfections were detected by RT-qPCR and western blot assays. **P < 0.01, ***P < 0.001

Apatinib inhibited OS cell growth and migration through LINC00261/miR-620/PTEN axis

Thereafter, we tried to detect whether LINC00261 could affect the tumor-suppressing effect of Apatinib through PTEN. Hence, under 10 μM Apatinib treatment, MG63 cells transfected with sh-NC, sh-LINC00261, sh-LINC00261+ pcDNA3.1 or sh-LINC00261+ pcDNA3.1/PTEN were applied for subsequent assays. Data from RT-qPCR and western blot assays depicted that the overexpression of PTEN partially restored the reduced PTEN expression and protein level caused by LINC00261 deficiency in Apatinib-treated MG63 cells (Figure 5(a)). Further, loss of LINC00261 facilitated the proliferation of Apatinib-treated MG63 cells, while such facilitating effect was then partially counteracted after PTEN overexpression (Figure 5(b-c)). Under the treatment of Apatinib, elevated expression of PTEN mitigated the inhibitory effect of LINC00261 silencing on the apoptosis of MG63 cells (Figure 5(d-e) and Supplementary Figure 1e). Furthermore, the migratory ability of Apatinib-treated MG63 cells was strengthened by the silencing of LINC00261 and such result was then partially reversed by the overexpression of PTEN (figure 5(f)). In conclusion, Apatinib inhibited OS progression through activating LINC00261/miR-620/PTEN axis.

Figure 5.

Apatinib suppressed OS cell growth and migration through LINC00261/miR-620/PTEN axis. (a) RT-qPCR and western blot assays were used to detect the mRNA and protein levels of PTEN in Apatinib-treated MG63 cells with indicated transfections. (b-c) Proliferation of Apatinib-treated MG63 cells with indicated transfections was detected by CCK-8 and colony formation assays. (d-e) Apoptosis of Apatinib-treated MG63 cells with indicated transfections was detected by flow cytometry analysis and caspase-3 activity assay. (f) Migration of Apatinib-treated MG63 cells with indicated transfections was detected by Transwell migration assay. *P < 0.05, **P < 0.01. No significance was represented as “n.s.”

Apatinib blocked OS tumorigenesis via LINC00261/miR-620/PTEN axis

Further, we tested above findings in vivo. Mice were inoculated with indicated MG63 cells and then treated with or without Apatinib. As anticipated, the in vivo growth of MG63 cells was hindered under Apatinib treatment while such inhibition could be partially reversed by LINC00261 silencing. Moreover, the knockdown of miR-620 could mitigate the rescuing impact of LINC00261 deficiency on in vivo tumor growth, while the growth was further promoted upon the silencing of PTEN (Figure 6(a-b)). Similar trend was seen in the changes of tumor weight (Figure 6(c)). In sum, Apatinib blocked the tumorigenesis of OS by regulating LINC00261/miR-620/PTEN axis.

Figure 6.

Apatinib blocked the in vivo tumorigenesis of OS by LINC00261/miR-620/PTEN axis. Mice were inoculated with parental MG63 cells or those under different transfections (with sh-NC, sh-LINC00261, sh-LINC00261+ antagomiR-620, or sh-LINC00261+ antagomiR-620+ sh-PTEN), and then treated with saline (control) or Apatinib. (a) Representative images of tumors derived from indicated MG63 cells in each group were presented. (b) The growth curve of tumors from indicated groups was presented. (c) The final weight of tumors from different groups was presented. *P < 0.05, **P < 0.01

Discussion

Recently, researchers have discovered that Apatinib (YN968D1), the newly-discovered high-affinity selective inhibitor of the VEGFR2 tyrosine kinase [6–8], could repress the progression of various cancers [9–11]. Besides, a previous report has unmasked the use of Apatinib in metastatic OS [14]. In addition, several works have unveiled that Apatinib repressed cell migration and invasion, but induced cell autophagy and apoptosis of OS [12,13]. Accordingly, our study validated that the treatment of Apatinib hindered cell proliferation, promoted cell apoptosis, and retarded cell migration of OS. Those findings suggested that Apatinib treatment could serve as a promising option for OS, and the in-depth exploration on the tumor-suppressing effect of Apatinib on OS progression might provide more references for the Apatinib-based treatment of OS.

Association of dysregulated lncRNAs with the pathogenesis of cancers have been well established over the past years [15]. In OS, studies on the role of lncRNAs have been emerging as well [21,22]. Notably, the implication of lncRNAs in the efficacy of drugs for OS has been delineated by a number of studies. For example, LINC00161 has been proved to strengthen the effect of cisplatin on inducing the apoptosis of OS cells via miR-645/IFIT2 axis [23]; lncRNA CTA may enhance the efficacy of doxorubicin on OS cells through inhibiting autophagy [24]. Through microarray analysis, our study firstly showed that Apatinib treatment induced LINC00261 in OS cells, and that LINC00261 level was down-regulated in OS cells compared with normal cells. According to previous researches, LINC00261 has been proved to present down-regulation and possess tumor-suppressing functions in a number of cancer types, such as endometriosis [25], hepatocellular cancer [26], choriocarcinoma [27], and gastric cancer [28]. Moreover, previous studies have revealed that LINC00261 can strengthen the chemo-sensitivity of cancer cells [29,30]. Herein, our study firstly demonstrated that knockdown of LINC00261 abrogated the restraining effect of Apatinib on OS cells, indicating that Apatinib suppressed OS progression through enhancing LINC00261.

Currently, the role of lncRNAs as a ceRNA in cancers has been demonstrated by mounting reports. In a ceRNA network, lncRNAs sequester miRNAs to protect the downstream target genes from silencing [43,44]. Herein, we identified miR-620 was the downstream of LINC00261 in OS, and we certified that miR-620 was up-regulated in OS cells and could be down-regulated by Apatinib. Former studies have shown that miR-620 was up-regulated and performed oncogenic roles in multiple cancers, such as lung adenocarcinoma, cervical cancer, lung cancer, and so on [40–42]. Herein, we firstly revealed that miR-620 was sponged by LINC00261 in OS cells. In addition, we validated that miR-620 overexpression abrogated the tumor-suppressive effect of Apatinib on OS. Furthermore, we found that PTEN was a downstream target of miR-620, and it was low-expressed in OS cells but significantly up-regulated under Apatinib treatment. PTEN is recognized as an anti-tumor gene which is responsible for the inactivation of PI3K/AKT pathway [31]. Several studies have revealed the participation of PTEN in the regulation of drug efficacy in cancer cells [32–34], including in OS cells [35–37]. Presently, we firstly validated that miR-620 targeted and inhibited PTEN in OS, and that LINC00261 up-regulated PTEN through absorbing miR-620. Finally, rescue assays suggested that Apatinib hindered in vitro cell growth and migration as well as in vivo tumorigenesis of OS through activating LINC00261/miR-620/PTEN axis.

Conclusion

In conclusion, our study firstly showed that Apatinib inhibited cell proliferation and migration whereas induced cell apoptosis of OS through activating LINC00261. LINC00261 functioned in OS cells by sponging miR-620 to induce PTEN expression. Those findings suggested that LINC00261 could be a promising target to improve the efficacy of Apatinib on OS.

Data availability statement

Not applicable.

Disclosure statement

This paper contains no competing interests.

Supplementary material

Supplemental data for this article can be accessed here