For Sp1 knockdown, the Sp1 shRNA plasmid (h) was purchased from Santa Cruz Biotechnology, Inc. (Dallas, TX, USA). For ERα knockdown, the short hairpin RNA (shRNA)-expressing lentiviral vector (pLKO.1-shERα) was constructed by targeting to 5′-GTACCAATGACAAGGGAAGT-3′; scrambled control sequence, 5′-AGCCAGGCAGTATAGAGAAT-3′. For constructing the ERα expressing vector, an approximately 1788 base pairs (bp) coding sequence was polymerase chain reaction (PCR)-amplified from cDNA reverse transcribed using 1 µg MCF-7 total RNA (MCF-7 cells from the American Type Culture Collection, Manassas, VA, USA) with the First Strand Synthesis kit (Guangzhou RiboBio Co., Ltd., Guangzhou, China) with a poly(dT) oligonucleotide. MCF-7 RNA was used due to high levels of ERα mRNA expression. For the following synthesis of the coding sequence of ERα, 0.5 µl cDNA template was amplified using PrimeSTAR^® Max DNA Polymerase (Takara Bio, Inc., Shiga, Japan). For PCR amplification, the following primers were used: ERα, forward 5′-GCCGGCGCTAGCATGCCATGACCCTCCACACCA−3′ and reverse 5′-CCTTAACTTAAGCAGACCGTGGCAGGGAAACCC−3′. Subsequent to 5 min initial denaturation at 98°C, the mixture was amplified for a total 30 cycles with a two-step cycle process that began with the denaturation at 98°C for 15 sec and annealing and extension at 65°C for 2 min. Subsequent to digestion with Nhe I and Hind III, the fragment was inserted into pcDNA3.1. For construction of the different expression vectors, the same procedure was followed described above with the following primer pairs: Hemagglutinin (HA)-ERα, forward 5′-GCCGGCGCTAGCTACCCATACGACGTCCCAGACTACGCTATGCCATGACCCTCCACACCA−3′ and reverse 5′-CCTTAACTTAAGCAGACCGTGGCAGGGAAACCC−3′; Flag-Sp1 forward 5′-GCCGGCGCTAGCGATTACAAGGATGACGACGATAAGATGGATGAAATGACAGCTGTGGTGA and reverse 5′-CCTTAACTTAAGTCAGAAGCCATTGCCACTGATA.

The human osteosarcoma cell line U2OS was purchased from the American Type Culture Collection. The cells were cultured in Dulbecco's modified Eagle's medium (DMEM; Gibco; Thermo Fisher Scientific, Inc., Waltham, MA, USA) supplemented with 10% fetal bovine serum (FBS; Life technologies; Thermo Fisher Scientific, Inc.) and 100X antibiotic-antimycotic mixed stock solution (Life technologies; Thermo Fisher Scientific, Inc.).

For transfection, Lipofectamine 2000 (Life technologies; Thermo Fisher Scientific, Inc.) was employed following the manufacturer's instructions. These methods were used to establish U2OS-shER, U2OS-shSp1, U2OS-ERα, U2OS-Sp1 and U2OS-shScrambled.

Total RNA of the target cells were isolated using TRIzol (Life technologies; Thermo Fisher Scientific, Inc.). First-strand cDNA was synthesized from 1 µg total RNA using the First Strand Synthesis kit (Guangzhou RiboBio Co., Ltd.). qPCR was performed with a SoEva Green Super Mix (Bio-Rad Laboratories, Inc., Hercules, CA, USA), and the Applied Biosystems 7500 Real-time system (ABI 7500HT instrument; Applied Biosystems; ThermoFisher Scientific, Inc.) was used for measurement. The reaction mixture contained 0.2 µl cDNA, 10 µl SoEva Green Super Mix (Bio-Rad Laboratories, Inc., Hercules, CA, USA), 0.5 µl of each primer and 8.8 µl deionized water. The primers used in the current study were as follows: MALAT-1, forward 5′-GACTTCAGGTCTGTCTGTTCT−3′ and reverse 5′-CAACAATCACTACTCCAAGC-3′; NEAT1, forward 5′-GCTCTGGGACCTTCGTGACTCT-3′ and reverse 5′-CTGCCTTGGCTTGGAAATGTAA-3′; HN1, forward 5′-CACAGCAAGACGAGAAGACCCTATGGAGC-3′ and reverse 5′-GTCAAGTTATTGGATCAA−3′; HOTAIR forward 5′-GACAGGGTCTGGGACAGAAG-3′ and reverse 5′-GAGTCAGAGTTCCCCACTGC-3; GAPDH, forward 5′-CTTTGGTATCGTGGAAGGACTC-3′ and reverse 5′-GTAGAGGCAGGGATGATGTTCT-3′.

Samples were separated by 10% SDS-PAGE and transferred to PVDF membranes (EMD Millipore, Billerica, MA, USA) pre-blocked in 5% powdered milk + 5% bovine serum albumin (Sigma-Aldrich; Merck Millipore, Darmstadt, Germany) in PBS containing 0.3% Tween 20 for 30 min at room temperature. Subsequently, the blot-transferred membrane was incubated at 4°C overnight with the following primary antibodies from Abcam (Cambridge, UK): Anti-SP1 (ab13370; 1:1,000), anti-ERα (ab32063; 1:1,000), anti-GAPDH (ab8245; 1:1,000) and anti-TIM antibody (ab66062; 1:1,000). After washing three times with PBS containing 0.3% Tween 20, the membrane was incubated with goat anti-rabbit horseradish peroxidase (HRP)-labeled IgG secondary antibodies (ab6721; 1:5,000; Abcam) for 1 h at room temperature. Then membrane was then developed by using ECL detection systems (Thermo Fisher Scientific, Inc.).

For the ChIP assay, 5×10⁶ cells were cross-linked with ice-cold 1% formaldehyde in 4°C for 15 min and washed three times with ice-cold PBS. Cells were collected with lysis buffer. DNA was then sheared to an approximate length of 200–500 bp. Sheared DNA (100 µl) was incubated with 10–15 µg primary antibody or rabbit IgG (negative control) followed by IP with 50 µl protein A agarose beads (Life Technologies; Thermo Fisher Scientific, Inc.) during an overnight incubation at 4°C with rotation. Enriched DNA was extracted from the DNA/antibody/protein A bead complexes by proteinase K digestion, reverse crosslinking process at 65°C for 4 h and purification via centrifugation at 1,000 × g for 10 min at 4°C. The IP product was amplified using the MALAT-1 promoter region: Forward GGAAGTTGGGCAGCAGCTCCACG and reverse CCACTGGTTCTAACCGGCTCTAG. Dihydrofolate reductase 5′untranslated region was included as a negative control; forward ACCTGGTCGGCTGCACCT and reverse TTGCCCTGCCATGTCTCG.

Protein A agarose bead slurry (50 µl) was incubated with 10 µg anti-Flag or anti-HA antibody overnight at 4°C with rotation. The beads were then centrifuged at 1,000 × g for 10 min at 4°C and washed three times with PBS. The protein A/antibody complex was then incubated with 500 µg lysate protein overnight with rotation, followed by three washes with PBS. The complex was then heated at 100°C for 10 min, and 15 µl eluted sample was loaded for the following semiquantitative western blot assay.

Intact nuclei were isolated from target cells using a nuclei isolaton kit (cat. no. NUC201; Sigma-Aldrich; Merck Millipore) following the manufacturer's instructions.

The Sp1-binding fragment of the MALAT-1 promoter region was 5′-biotinylated and annealed. The sequence was as follows: 5′-CAGGCGTTAGGGCGGGGCGCGCGTGC-3′. Either 10 or 30 ng target protein was incubated with 10 pM biotinylated DNA fragment for 30 min at 4°C. The complexes were fractionated using 4% PAGE gel in 0.5X TBE and transferred onto Hybond-N+ membranes (Life Technologies; Thermo Fisher Scientific, Inc. Subsequently, the assay was conducted and detected using HRP-conjugated streptavidin (LightShift™ Chemiluminescent EMSA kit; Thermo Fisher Scientific, Inc.) according to the manufacturer's instructions.

Target cells were seeded into a 96-well plate at a concentration of 5,000 cells/well. After 24 h, E2 was added if required. At 1, 2, 3 and 4 days after transfection, the cell proliferation assay was conducted by the addition of 10 µl CCK-8 solution [Sangon Biotech (Shanghai) Co., Ltd., Shanghai, China] to each well, followed by incubation at 37°C for 2 h. Absorbance was measured at a wavelength of 450–620 nm using a Multiskan spectrum microplate reader (Thermo Fisher Scientific, Inc.).

Cell migration was analyzed using a scratch assay. The cell layer that reached confluence was scratched by a 200 µl pipette tip and incubated at 37°C for 24 h. The average extent of wound closure was imaged. For analyzing invasion, the under surface of the membrane was coated with Matrigel (0.01%) at 37°C for 2 h. The lower chamber was filled with 1 ml DMEM supplemented with 10% FBS. A total of 1×10⁶ cells in a volume of 0.2 ml were added to the upper chamber. Following incubation at 37°C for 24 h, the cells on the upper surface of the transwell membrane were removed. The cells attached to the lower surface of membrane were stained with 1% crystal violet and imaged.

The results were expressed as the means ± standard deviation. The statistical analysis involving two groups was performed using the means of Student's t-test. All data were processed using SPSS software, version 19. P<0.05 was considered to indicate a statistically significant difference.

Due to the fact that ERα is strongly expressed in U2OS cells (data not shown), the effects of E2 treatment on the expression of lncRNAs was analyzed in U2OS cells. MALAT-1, NEAT1 and HOX transcript antisense RNA (HOTAIR), which are all associated with malignancy and upregulated in osteosarcoma, were detected quantitatively. As presented in Fig. 1A, 10–100 pM E2 treatment specifically and significantly upregulated the MALAT-1 RNA levels, however did not affect NEAT1 or HOTAIR. To determine whether the E2 treatment is mediated by the stimulation of ERα, ERα was efficiently knockdown by a plasmid that encodes a shERα sequence (Fig. 1B; knockdown efficiency ~65±4.5%). As expected, knockdown of ERα desensitized U2OS-shERα to E2, indicating the necessity of ERα. As mentioned previously, the direct interaction between ERα and Sp1, and the transcriptional regulatory role of Sp1 on MALAT-1 suggests a role of Sp1 in E2-induced MALAT-1 upregulation. Similar to the results of ERα knockdown, Sp1 knockdown also desensitized U2OS-shSp1 to E2. In order to establish whether ERα may regulate MALAT-1 expression without E2 stimulation, ERα was overexpressed in U2OS. It was identified that, compared with the U2OS-vector, MALAT-1 RNA levels were not significantly different without E2 treatment (Fig. 1E, left panel), however were significantly upregulated with E2 treatment (Fig. 1E, right panel).

Due to the fact that Sp1 is reported to bind to MALAT-1 promoter region, its binding activity in ERα-knockdown U2OS cells was analyzed. ChIP was performed with an anti-Sp1 antibody. IgG was employed as a negative control. As presented in Fig. 2A, the binding activity of Sp1 to the MALAT-1 promoter region increased in U2OS-shScrambled, however not in U2OS-shERα. For the quantitative assay, IP products were included in the qPCR reaction. Consistently, the stimulation of MALAT-1 promoter binding activity of Sp1 by E2 treatment was abolished when ERα was knocked down (Fig. 2B). As it has been previously demonstrated, E2 treatment will translocate ERα from the cell membrane into the nucleus, it was hypothesized that the increase of Sp1/DNA binding activity may be associated with the subcellular location. The immunofluorescence assay indicated that, following E2 treatment, Sp1 accumulated in the nuclei instead of the cytoplasm (Fig. 2C). For further confirmation, nuclear fractions from E2-treated U2OS cells were isolated. GAPDH (cytoplasm specific protein) and translocase of the inner membrane (mitochondrial specific protein) were employed as indicators of purity. Western blotting identified nuclear fractions, indicating acceptable cytoplasm and mitochondria contamination (Fig. 2D). Following E2 treatment, ERα and Sp1, particularly Sp1, were significantly accumulated in the nuclear fraction.

Plasmids containing the HA-tagged ERα or Flag-tagged Sp1 coding sequences were established, and they were introduced into HEK293 cells for transient expression. Cell lysates were immunoprecipitated with anti-HA or anti-Flag antibodies individually. Subsequently, the IP products were assayed by immunoblotting using anti-HA or anti-Flag antibodies separately. As presented in Fig. 3A, Flag-tagged and HA-tagged proteins were detectable in the IP products, suggesting the direct binding of Sp1/ERα. It was additionally investigated whether ERα binds to the MALAT-1 promoter by itself. In order to establish this, the EMSA assay was conducted. The MALAT-1 promoter region DNA sequence was labeled with biotin was incubated with purified HA-tagged ERα or Flag-tagged Sp1 separately and fractionated by 4% 0.5X TBE PAGE gel. The binding bands were observed only in Sp1 involved lanes. For further confirmation, E2 treated U2OS cells underwent a ChIP assay. Following E2 treatment, anti-ERα and anti-Sp1 antibodies enriched the MALAT-1 promoter region. However, without E2 treatment, the anti-ERα antibody failed to enrich the MALAT-1 promoter region, while the anti-Sp1 antibody was observed to exhibit no difference to the E2-treated sample (Fig. 3C). Taken together, ERα binds to the Sp1 protein, however not to the MALAT-1 promoter.

MALAT-1 tightly regulates the physiological processes in osteosarcoma, and aids in the determination of the effects of E2 treatment on these processes in an ERα-dependent manner. For testing proliferation of E2 treated cells, U2OS, U2OS-shScrambled and U2OS-shERα were seeded into 12-wells and were stained with CCK-8 on days 1, 2, 3 and 4. As presented in Fig. 4A, E2 treatment significantly promoted proliferation, however not in U2OS-ERα. Subsequently, cells were seeded in 0.3% soft agar for testing colony formation for 14 days. Consistently, E2 promoted the colony formation, however not in ERα-knockdown U2OS cells (Fig. 4B). In order to further determine the effects of E2 treatment on migration and invasion, the scratch assay and transwell assay were performed with the above mentioned cell lines. It is observed that E2 treatment affected the migration and invasion independent of the presence of ERα. (Fig. 4C).