Circ_0049271 targets the miR-1197/PTRF axis to attenuate the malignancy of osteosarcoma

2.1Sources of reagents, cells, and animals

hFOB 1.19 osteoblasts and human-derived OS cell lines (HOS and Saos-2) were purchased from Procell Life Science & Technology Co., Ltd. (Wuhan, China). Another human SW1353-derived OS cell line was provided by the BeNa Culture Collection (Beijing, China). Lipofectamine 3000, fetal bovine serum (FBS), McCoy’s 5A medium, Dulbecco’s Modified Eagle’s Medium (DMEM), and L-15 medium were purchased from Procell Life Science & Technology Co., Ltd. Cytoplasmic and nuclear RNA purification kits were acquired from Norgen BioTek (Thorold, Canada). RNase R was purchased from Geneseed Biotech (Guangzhou, China). An RNA extraction Kit (Cwbio, Beijing, China), an Evo m-mlv reverse transcription reagent, and a SYBR green Kit (Accyrate Biology, Changsha, China) were used for quantitative reverse transcription-polymerase chain reaction (qRT-PCR). The Magna RIP RNA-Binding Protein Immunoprecipitation Kit was purchased from Millipore (Billerica, MA, USA). Cell counting kit-8 (CCK-8), crystal violet, and Matrigel were obtained from Solarbio (Beijing, China). Primary antibodies against PTRF (Cat# ab76919) and glyceraldehyde 3-phosphate dehydrogenase (GAPDH; Cat# ab8245) and horseradish peroxidase (HRP)-conjugated secondary antibodies (Cat# ab6759) for western blotting were purchased from Abcam (Cambridge, UK). Genomeditech (Shanghai, China) synthesized the pcDNA3.1-circ-circ_0049271 overexpression vector (OE-circ), pcDNA3.1-PTRF, pcDNA3.1 empty vector (OE-NC), miR-1197 mimic (mimic), and mimic-NC. Tumor xenograft models were established using four-week-old BALB/c nude mice (20–26 g) obtained from Charles River Laboratories (Sulzfeld, Germany).

2.2Cell culture

HOS and hFOB 1.19 cells were cultivated in DMEM containing 1% penicillin/streptomycin and 10% FBS. Saos-2 cells were cultured in McCoy’s 5A medium supplemented with 1% penicillin/streptomycin and 15% FBS. SW1353 cells were cultured in L-15 medium supplemented with 1% penicillin/streptomycin and 10% FBS. The cultures were all maintained in an incubator at 37^∘C with 5% CO₂. The cells were cultured until the logarithmic growth phase and then collected for subsequent experiments.

2.3Cell transfection

Lipofectamine 3000 was used to transfect 2 μg OE-circ, 2 μg OE-PTRF, 2 μg OE-NC, 50 nM mimic, and 50 nM mimic-NC into 1 × 10⁶ HOS and SW1353 cells when cell confluency was more than 50%. The cells were incubated for 48 h and collected for subsequent experiments.

2.4Measurement of cell proliferation

Next, the viabilities of SW1353 and HOS cells were assessed via the CCK-8 assay. Briefly, 1 × 10⁴ cells transfected with SW1353 or HOS were inoculated into a 96-well plate and incubated for 0, 24, 48, and 72 h at 37^∘C with 5% CO₂. Approximately 15 μL of CCK-8 solution was added to the corresponding wells, and the cells were incubated for 3 h at 37^∘C. OS cell viability was analyzed at 450 nm using a microplate reader (DeTie Laboratory Equipment, Nanjing, China).

Cell numbers were estimated using a cell colony experiment. HOS and SW1353 cells in a 6-well plate (3 × 10² cells/well) were incubated for two weeks. The medium was replaced every two days. Subsequently, the cells were washed, immobilized, stained, and kept at 25^∘C. Finally, photographs of cell colonies were captured using the Olympus light microscope (Tokyo, Japan).

2.5Transwell invasion assay

SW1353 and HOS cells (2 × 10⁵ cells/well) were cultured in the Matrigel-precoated upper chambers filled with DMEM without FBS. The bottom chambers were filled with DMEM with 10% FBS. After incubation for 24 h, invasive OS cells were stained with 0.5% crystal violet, and three randomly selected fields were photographed using the Olympus light microscope.

2.6Western blotting assay

Total protein was extracted from 2 × 10⁶ HOS and SW1352 cells after cell lysis with the radioimmunoprecipitation assay buffer, followed by total protein quantification using the BCA kit (Beyotime, China). Proteins (20 μg) were subjected to electrophoresis on a 10% sodium dodecyl sulfate-polyacrylamide gel and transferred to polyvinylidene fluoride membranes using a protein marker (10–200 kD; Beyotime). Subsequently, 5% fat-free milk was used to block the membranes, which were incubated with anti-PTRF (1:5000) and anti-GAPDH (1:2000) antibodies overnight at 4^∘C. Then, the membranes were incubated with HRP-conjugated secondary antibodies (1:20000) at 25^∘C for 2 h. An enhanced chemiluminescence detection kit (Beyotime) was used for band visualization, and GAPDH was used for normalization.

2.7qRT-PCR

Total RNAs from OS samples and cell lines (1 × 10⁶ cells) were isolated using an RNA extraction kit. The First-Strand cDNA Synthesis Kit was used to synthesize cDNA. PCR was performed using the Applied Biosystems (ABI) 7500 Real-Time PCR System with the SYBR Green FAST Mastermix Kit. Table 1 provides all primers used in this study based on the previous literature [33, 34, 35, 36] and the National Center of Biotechnology Information database. Then, expression levels of circ_0049271, miR-1197, and PTRF were computed via the 2 ^{- ΔΔCt} approach. U⁢6 and GAPDH were used for normalization.

2.8Subcellular location analysis

The cytoplasmic and nuclear RNAs of SW1353 and HOS cells (1 × 10⁶ cells) were separated using the Cytoplasmic and Nuclear RNA Purification Kit. Subsequently, purified RNAs were subjected to qRT-PCR. U⁢6 and GAPDH were used as the controls for nuclei and cytoplasm, respectively.

2.9RNase R treatment

RNAs extracted from the SW1353 and HOS cells (1 × 10⁶ cells) were mixed with RNase R (3 U/μg) for 1 h at 37^∘C. Then, expression levels of circ_0049271 and its linear gene, kelch-like ECH-associated protein 1 (KEAP1), were assessed via qRT-PCR.

2.10RNA immunoprecipitation (RIP) assay

HOS and SW1353 cells (1 × 10⁶ cells) were mixed with the RIP lysis buffer in a magnetic bead suspension and conjugated with anti-Ago2 or anti-IgG (5 μg/μL) overnight. After protein digestion and RNA purification, the levels of circ_0049271, miR-1197, miR-620, and miR-1270 were determined via qRT-PCR.

2.11Bioinformatics analysis

Two online databases, starBase (https://rnasysu.com/ encori/) and CircInteractome (https://circinteractome. nia.nih.gov/index.html), were used to predict the miRNA binding to circ_0049271. Another database, TargetScan (https://www.targetscan.org/vert_71/), was used to predict the target genes of miR-1197. In addition, an mRNA microarray (GSE12865) from Gene Expression Omnibus was used to confirm the downregulated genes in OS samples with adj. P< 0.01 and logFC <–2. Finally, Venny 2.1.0 (https://bioinfogp.cnb.csic.es/tools/venny/) was used to overlap the common miRNAs or genes from different databases.

2.12Dual-luciferase reporter assay

To create mutant (MUT) and wild-type (WT) circ_0049271/PTRF, we inserted miR-1197-binding sites on circ_0049271/PTRF sequences into pGL3 vectors. Recombinant vectors and mimic/mimic-NC were introduced into HOS and SW1353 cells (> 50% cell confluency). Then, the Promega Dual-Glo Luciferase Assay System (Madison, WI, USA) was used to measure the luciferase activity.

2.13Murine xenograft model

OE-circ or OE-NC lentiviral vector was introduced into SW1353 cells (1 × 10⁵ cells), which were subcutaneously injected into the right flank of each mouse (n= 5/group). Tumor volume was computed every four days. After four weeks, euthanasia was performed before extracting and weighing the tumor xenografts.

2.14OS sample collection

Nineteen individuals diagnosed with OS by two pathologists using computed tomography and a surgical biopsy were enrolled in this study. Tumors and matched normal tissues (> 5 cm from the tumor) were acquired via surgery. The inclusion criteria for patients with OS were as follows: 1) patients initially diagnosed with OS in our hospital; 2) patients with OS who did not receive any treatment before surgery; and 3) patients who signed the informed consent forms for inclusion in this study. The exclusion criteria for patients with OS were as follows: 1) patients with recurrent OS; and 2) patients with other diseases. The collected OS specimens were stored at -80^∘C until use. All participants or their guardians provided written informed consent for inclusion in this study. All experimental procedures were approved by the ethics committee of the Fifth Hospital of Wuhan (approval number: 2022-11-07).

2.15Statistical analyses

Data are represented as the mean ± standard deviation. All data were analyzed using the SPSS v.22.0 software. Pearson’s correlation coefficient was used for linear correlation analysis. Differences between groups were analyzed using the Student’s t-test. In contrast, those among more than two groups were analyzed using the analysis of variance followed by Tukey’s post-hoc or Sidak’s multiple comparisons tests. Statistical significance was set at P< 0.05.

Figure 1.

Circ_0049271 is under-expressed in OS. (A) Circ_0049271 levels in OS normal and tumoral tissues were estimated through qRT-PCR. P< 0.0001 vs. normal. (B) Circ_0049271 levels in hFOB1.19 cells and OS cell lines were quantified through qRT-PCR. ^∗∗ P < 0.01 vs. hFOB1.19 cells. (C) Relative circ_0049271 expressions within the nuclei and cytoplasm of SW1353 and HOS cells as quantified through subcellular localization analysis. (D) The levels of circ_0049271 and the corresponding linear KEAP1 in total cellular RNA incubated with RNase R. ^∗∗ P < 0.01 vs. control.

3.1Circ_0049271 is underexpressed in OS

Here, circ_0049271 expression levels were considerably lower in tumor tissues than in the corresponding normal tissues (Fig. 1A). Moreover, circ_0049271 was underexpressed in SW1353, HOS, and Saos-2 OS cells relative to its expression in hFOB1.19 cells (Fig. 1B). Therefore, SW1353 and HOS cells were selected for subsequent experiments. A subcellular localization assay revealed that circ_0049271 was mainly localized in the cytoplasm (Fig. 1C). To corroborate its circular features, circ_0049271 and its linear transcript, KEAP1, were treated with RNase R. As shown in Fig. 1D, unlike KEAP1, circ_0049271 exhibited resistance to RNase R digestion. These findings suggest that circ_0049271 is underexpressed and functions as a competing endogenous RNA (ceRNA) in OS.

Figure 2.

Enforced circ_0049271 suppresses the invasion and proliferation of OS cells in vitro as well as the growth of tumors in vivo. (A) Circ_0049271 levels in HOS and SW1353, following their OE-circ/NC transfection, were estimated through quantitative RT-PCR. ^∗∗P < 0.01 vs. OE-NC. The viability (B) and colony numbers (C) of SW1353 and HOS cells harboring either OE-circ/NC were assessed through the CCK-8 and colony formation experiments. ^∗P < 0.05, ^∗∗P < 0.01 vs. OE-NC. (D) The invasive capacity of SW1353 and HOS as assessed in the transwell experiment. ^∗∗P < 0.01 vs OE-NC. (E) Representative image, volume, and weight of xenograft tumors after injecting the mice with SW1353 cells stably expressing OE-circ/NC. ^∗∗P < 0.01 vs. OE-NC.

3.2Enforced circ_0049271 suppresses the invasion and proliferation of OS cells in vitro and tumor growth in vivo

To verify the effects of circ_0049271 on OS cells, it was overexpressed by transfecting SW1353 and HOS cells with OE-circ (Fig. 2A). As shown in Fig. 2B, the viability of OS cells decreased markedly after circ_0049271 overexpression. Moreover, circ_0049271 overexpression reduced the number of OS cell colonies (Fig. 2C) and suppressed the invasion of SW1353 and HOS cells (Fig. 2D). In addition, analysis of the influence of circ_0049271 on solid tumor growth revealed that the injection of stably overexpressing circ_0049271 drastically diminished the tumor volume and weight in SW1353 cells (Fig. 2E). Therefore, circ_0049271 represses OS tumorigenesis in vivo and in vitro.

Figure 3.

Circ_0049271 binds to miR-1197. (A) starBase and circInteractome predicted the miRNAs binding to circ_0049271. (B) RIP experiment was accomplished to verify the interaction of miR-1197/miR-620/miR-1270 with circ_0049271. ^∗∗P < 0.01 vs Anti-IgG. (C) The binding sites between miR-1197 and circ_0049271. (D) The luciferase activities in HOS and SW1353 transfected with a combination of a pGL3-circ_0049271 WT/MUT and either a miR-1197 mimic/NC were gauged through the dual luciferase experiment. ^∗∗P < 0.01 vs. circ_0049271 WT + mimic-NC. (E) The miR-1197 levels in OS normal and tumoral tissues were analyzed via quantitative RT-PCR. P< 0.0001 vs. normal. (F) MiR-1197 levels in hFOB1.19 cells and OS cell lines were estimated through quantitative RT-PCR. ^∗∗P < 0.01 vs hFOB1.19 cells. (G) The correlation of miR-1197 expression with that of circ_0049271 in OS tissues was ascertained through the Pearson correlation coefficient. (H) MiR-1197 expression in SW1353 and HOS cells, after their OE-circ, OE-NC, mimic, mimic-NC, or OE-circ + mimic transfection, were quantified via qRT-PCR. ^∗∗P < 0.01 vs. OE-NC. ^{# #} P < 0.01 vs. mimic-NC. ^{& &} P < 0.01 vs. OE-circ + mimic.

3.3Circ_0049271 binds to miR-1197

To confirm miRNA binding to circ_0049271, starBase and circInteractome databases were used to predict the binding of three miRNAs (miR-1197, miR-620, and miR-1270) to circ_0049271 (Fig. 3A). RIP assays of SW1353 and HOS cells revealed that, compared to miR-620 and miR-1270, circ_0049271 and miR-1197 were principally enriched by Ago2 antibodies (Fig. 3B). As shown in Fig. 3C, the circInteractome revealed binding sites between circ_0049271 and miR-1197. Moreover, the luciferase reporter assay indicated no obvious changes in the luciferase activity of the circ_0049271 MUT plus mimic and circ_0049271 MUT plus mimic-NC groups. Luciferase activity in the circ_0049271 WT plus mimic group was significantly lower than that in the circ_0049271 WT plus mimic-NC group (Fig. 3D). Relative to the respective NCs, elevated levels of miR-119 were detected in OS tissues (Fig. 3E) and cell lines (Fig. 3F). Circ_0049271 expression exhibited an inverse linear correlation with miR-1197 expression in OS tissues (Fig. 3G). Mimic/mimic-NC was introduced into SW1353 and HOS cells to verify this regulatory relationship. As illustrated in Fig. 3H, miR-1197 expression levels increased after mimic transfection and decreased after OE-circ transfection. Furthermore, transfection with a mimic partially attenuated the repressive effect of OE-circ on miR-1197 expression. Therefore, circ_0049271 targets miR-1197, and these two exhibit a negative regulatory relationship with each other in OS.

Figure 4.

Circ_0049271 overexpression prohibits the invasion and proliferation of cells through miR-1197 modulation in OS. The viability (A) and colony numbers (B) of SW1353 and HOS cells transfected with OE-circ, OE-NC, mimic, mimic-NC, or OE-circ + mimic were measured by means of the CCK-8 and colony formation experiments. (C) SW1353 and HOS cell invasion was evaluated through the transwell experiment. ^# P < 0.05; ^{# #} P < 0.01 vs. mimic-NC. ^& P < 0.05; ^{& &} P < 0.01 vs. OE-circ + mimic. ^∗P < 0.05; ^∗∗P < 0.01 vs. OE-NC.

3.4Circ_0049271 overexpression abates the proliferation and invasion of OS cells via miR-1197 modulation

Rescue experiments investigated the interaction between circ_0049271 and miR-1197 in SW1353 and HOS cells. Overexpression of miR-1197 distinctly enhanced the viability (Fig. 4A), number (Fig. 4B), and invasion (Fig. 4C) of SW1353 and HOS cells. Impaired proliferation and invasion of OS cells caused by enforced circ_0049271 were considerably reversed by miR-1197 upregulation (Fig. 4A–C). These findings imply that circ_0049271 acts as a ceRNA of miR-1197 in OS progression.

Figure 5.

MiR-1197 targets PTRF. (A) Three genes were overlapped from TargetScan and GSE12865. (B) The levels of SPOCK1, PTRF, and TIMP3 in OS normal and tumoral tissues were estimated via quantitative RT-PCR. P< 0.0001 vs. normal. (C) The correlation between the miR-1197 and PTRF expressions in OS tissues was ascertained through Pearson correlation coefficient. (D) TargetScan predicted a miR-1197 – PTRF binding site. (E) The target relationship of miR-1197 with PTRF was verified through the dual luciferase experiment. ^∗∗P < 0.01 vs PTRF WT + mimic-NC. (F) PTRF levels in hFOB1.19 cells and OS cell lines were gauged through qRT-PCR. ^∗∗P < 0.01 vs. hFOB1.19 cells. (G) The levels of PTRF proteins in HOS and SW1353, after their OE-PTRF, OE-NC, mimic, mimic-NC, or OE-PTRF + mimic transfection, were analyzed via western blotting. ^# P < 0.05, ^{# #} P < 0.01 vs. mimic-NC. ^∗∗P < 0.01 vs. OE-NC. ^& P < 0.05, ^{& &} P < 0.01 vs. OE-PTRF + mimic.

3.5PTRF is a target of miR-1197

TargetScan was used to identify the target genes of miR-1197. An mRNA microarray (GSE12865) was used to identify the downregulated genes in OS samples. Using Venny 2.1.0, three genes (SPOCK1, PTRF, and tissue inhibitor of metalloproteinase 3 [TIMP3]) overlapped between the TargetScan and GSE12865 datasets (Fig. 5A). qRT-PCR revealed that only PTRF expression was reduced and inversely correlated with miR-1197 expression in OS tissues (Fig. 5B–C). Therefore, PTRF was selected for further investigation. Figure 5D shows the TargetScan-predicted binding site of miR-1197 on PTRF. Luciferase reporter assays revealed that combining PTRF WT and mimic markedly diminished the luciferase activity in SW1353 and HOS cells. However, luciferase activity was barely affected by the combination of PTRF MUT and mimic or mimic-NC (Fig. 5E). Reduced PTRF expression was detected in OS cell lines (Fig. 5F). Western blotting revealed that OE-PTRF transfection markedly elevated PTRF protein levels. In contrast, mimic transfection reduced PTRF production in SW1353 and HOS cells (Fig. 5G). Moreover, decreased PTRF protein levels via mimic transfection were reversed by transfecting OE-PTRF into SW1353 and HOS cells (Fig. 5G). These data suggest that miR-1197 targets PTRF and inversely modulates its expression.

Figure 6.

MiR-1197/PTRF axis is associated with the abilities of OS cells to invade and proliferate. The viability (A) and colony numbers (B) of SW1353 and HOS cells carrying OE-PTRF, OE-NC, mimic, mimic-NC, or OE-PTRF + mimic were evaluated through the CCK-8 and colony formation experiments. (C) SW1353 and HOS cell invasion was assessed by means of the transwell experiment. ^# P < 0.05, ^{# #} P < 0.01 vs. mimic-NC. ^∗ P < 0.05, ^∗∗ P < 0.01 vs. OE-NC. ^& P < 0.05, ^{& &} P < 0.01 vs. OE-PTRF + mimic.

3.6miR-1197/PTRF axis is associated with the proliferation and invasion of OS cells

Next, the interaction between miR-1197 and PTRF during OS progression was verified in vitro. High levels of PTRF decreased the viability (Fig. 6A) and colony number (Fig. 6B) of OS cells. Moreover, PTRF overexpression inhibited OS cell invasion (Fig. 6C). PTRF overexpression further attenuated the promotive effect of miR-1197 upregulation on the proliferation and invasion of OS cells (Fig. 6A–C). Our data suggest that miR-1197 interacts with PTRF during OS progression.