The human ovarian cancer cell lines OVCAR-3, SKOV3, A2780, 3AO, COC1, OV-90, and human ovarian surface epithelial cells (HOSEpiC) were purchased from Type Culture Collection of Chinese Academy of Sciences (Shanghai, China). OVCAR-3, SKOV3, A2780, and COC1 cells were maintained in RPMI1640 medium (Corning, United States) supplemented with 10% fetal bovine serum (FBS; Gibco; Thermo Fisher Scientific, Inc.), 100 U/ml penicillin and 100 μg/ml streptomycin (Thermo Fisher Scientific, Inc.) and cells were cultured at 37°C with 5% CO2. 3AO and OV-90 cells were maintained in Dulbecco’s Modified Eagle Medium (DMEM) (Corning, United States) supplemented with 10% fetal bovine serum (FBS; Gibco; Thermo Fisher Scientific, Inc.), 100 U/ml penicillin and 100 μg/ml streptomycin (Thermo Fisher Scientific, Inc.). HOSEpiC cells were maintained in ovarian epithelial cell medium (ScienCell, United States).

Ketamine was supplied by Sigma-Aldrich (United States) and dissolved in DMSO.

The whole data of ovarian cancer patients were downloaded from the GEO dataset (GSE38666)¹ (Lili et al., 2013). Data mining is implemented in the R programing language. Data were normalized by z-score in different samples. Heatmap was generated using clustering method and was used to reveal the differentially expressed ovarian cancer related lncRNAs when comparing that in normal tissues or ovarian cancer tissues.

Cell proliferation was assessed using Sulforhodamine B (SRB) assay. Briefly, OC cells were seeded in 96-well (3,000 cells per well) and treated with indicated reagents. The cell proliferation was measured by SRB assay after 3 days treatment (Vichai and Kirtikara, 2006). Cell survival was assessed using trypan blue staining, in which dead cells were blue stained, and counted manually using hemocytometer.

For the colony formation assay, OC cells (1,500 cells/well) were seeded in 6-well plate and maintained in medium for 10–14 days. Subsequently, the colonies were fixed with 4% paraformaldehyde and stained with 0.1% crystal violet, and the number of clones was counted using an inverted microscope.

Total RNA from OC cells was isolated using RNA isolation kit (Qiagen, United States) according to the manufacturer’s protocol. iScript^TM Reverse Transcription Super mix kit (Bio-Rad, United States) was used for cDNA synthesis, and the samples were analyzed using SYBR Green Master Mix on a real-time PCR system (Bio-Rad). GAPDH was utilized as an endogenous calibrator control. The primer sequences used were as follows: PVT1, forward 5′-TGAGAACTGTCCTTACGTGACC-3′, Reverse 5′-AGAGCACCAAGACTGGCTCT-3′; MALAT1, forward 5′-AGAGCACCAAGACTGGCTCTGTAAC-3′, Reverse 5′-GAACAGAAGGAAGAGCCAAG-3′; LINC00092, forward 5′-CCTATGATTTGGCCTCTGGA-3′, reverse 5′-GAGAGCA GCGTTCAGGAAAC-3′; PTAR, forward 5′-ACAGATGTAAAC CAACCAGA-3′, reverse 5′-ATGCTACTGGAGACTTTAGG-3′; SnaR, forward 5′-TGGAGCCATTGTGGCTCCGGCC-3′, reverse 5′-CCCATGTGGACCAGGTTGGCCT-3′; Meg3, forward 5′-CTGCCCATCTACACCTCACG-3′, reverse 5′-CTC TCCGCCGTCTGCGCTAGGGGCT-3′; ZFAS1, forward 5′-ACG TGCAGACATCTACAACCT-3′, reverse 5′-TACTTCCAACAC CCGCAT-3′; UCA1, forward 5′-CTCTCCTATCTCCCTTCAC TGA-3′, reverse 5′-CTTTGGGTTGAGGTTCCTGT-3′; MIR4697HG, forward 5′-GAAGTGTGTGTGCAGGCTTG-3′, reverse 5′-GGAAAAGGCTCTGTCGTGGA-3′; TUG1, forward 5′-TAGCAGTTCCCCAATCCTTG-3′, reverse 5′-CACAAATTC CCATCATTCC-3′; DNM3OS, forward 5′-GGTCCTAAATTCA TTGCCAGTTC-3′, reverse 5′-ACTCAAGGGCTGTGATTT CC-3′; EWSAT1, forward 5′-GTGTCTGGCAAGGAACAC TA-3′, reverse 5′-GGTGGAGAAGAGGGACAAT-3′; HOTAIR, forward 5′-GGCAAATGTCAGAGGGTTCT-3′, reverse 5′-TT CTTAAATTGGGCTGGGTC-3′; GAS5, forward 5′-TGGTTCT GCTCCTGGTAACG-3′, reverse 5′-AGGATAACAGGTCTGC CTGC-3′; and GAPDH, forward 5′-CACCCACTCCTCCACC TTTG-3′ and reverse 5′-CCACCACCCTGTTGCTGTAG-3′. The 2-ΔΔCq method was used to calculate the relative expression levels.

Cells were lysed by radioimmunoprecipitation buffer, and 20 μg cellular protein extracts were separated in SDS-PAGE gel and was then transferred to nitrocellulose membranes (Millipore, United States). Membrane was blocked with 5% non-fat milk and incubated with antibodies against cytocrome C (1: 1,000, Abcam Biotechnology, United States), VDAC (1: 1,000, Thermo Scientific, United States), PARP1 (1: 1,000, Cell Signaling Technology, United States), or Actin (1: 5,000, Santa Cruz Biotechnology, United States) overnight at 4°. Then, the membranes were incubated with secondary antibody and the proteins were visualized using Super Signal West Pico Chemiluminescent Substrate (Thermo Scientific).

Caspase-3/7 activity was assessed by using Apo-ONE^TM Homogeneous Caspase-3/7 Assay (Promega Corporation, United States) according to the manufacturer’s instruction.

After treated with vehicle or indicated ketamine, the OC cells were harvested by trypsinization, fixed with 70% ethanol, and then retained at −20°C overnight. After washed with PBS three times, cells were resuspended in propidium iodide (PI) solution that contains RNase (100 μg/mL), and incubated in dark at room temperature for 30 min followed by a flow cytometer study.

P300 siRNA was purchased from Sigma Aldrich (United States). Lipofectamine RNAiMAX (Invitrogen, United States) was used for transfection according to the instruction.

SimpleChIP^® Enzymatic Chromatin IP Kit (Cell Signaling Technologies, United States) was used for ChIP assays according to manufacturer’s instructions; antibodies against H3K27ac (Cell Signaling Technology, United States), P300 (Cell Signaling Technology, United States) and EZH2 (Cell Signaling Technology, United States) were used for immunoprecipitation. Immunoprecipitated DNA was analyzed by QRT-PCR using the following primers: PVT1 promoter fragment 1: F: 5′-GCA GGAGAATCGCTTGAAC-3′ and R: 5′-ACAGATGTAAGAG CTGCCC-3′; fragment 2: F: 5′-GAACAAGATAACCACA TCCCAC-3′ and R: 5′-TTTCCAGAAGCCGAGTTGC-3′; fragment 3: F: 5′-TCTGGCCCTCCTATTTCAC-3′ and R: 5′-TTTCCCTGAGCCCTCTTAC-3′; fragment 4: F: 5′-CAGA GCCTACCCTCCGCT-3′ and R: 5′-CGGGGCTGGCGGGTT-3′; fragment 5: F: 5′-TCCTCCCCAATCTAAGTGCC-3′ and R: 5′-GCCAGTCACTTTCCCGTTTC-3′. P57 promoter primer: F: 5′-GGTGTCTAGGTGCTCCAGGT-3′ and R: 5′-GCACTCTCCAGGAGGACACA-3′.

RNA immunoprecipitation (RIP) was conducted by using RNA Binding Protein Immunoprecipitation Kit (Millipore, United States) following the manufacturer’s instructions. Antibodies against EZH2 and IgG (control) (Cell Signaling Technology, United States) were used for immunoprecipitation. Immunoprecipitated RNAs were then determined by QRT-PCR analysis.

Data were presented as mean ± SD from three independent experiments. P-value was determined using Two-tailed Student’s t-test and ANOVA test. The results were illustrated with GraphPad 7. P < 0.05 was deemed to indicate statistical significance.

o investigate the inhibitory effect of ketamine on OC cells growth, six OC cell lines (OVCAR-3, SKOV3, A2780, 3AO, COC1, OV90) were treated with indicated concentration of ketamine. As shown in Figures 1A–D, ketamine treatment significantly inhibited OC cells proliferation and survival in dose- and time-dependent manner. We next evaluated the effect of ketamine on normal human ovarian surface epithelial cells (HOSEpiC). No obviously inhibitory effect was observed after ketamine treatment (Supplementary Figure S1).

The inhibitory effect of ketamine was further investigated in terms of possible mechanisms and cell cycle. As shown in Figure 2A, flow cytometry results indicated that ketamine treatment caused a distinct increase in the cells arrested at G2-M phase in OCVAR-3 and SKOV3 cells. In addition, the activity of caspase3/7 was higher after ketamine treatment (Figure 2B). The ketamine treatment also elevated the disassociation of cytochrome C from mitochondria and PARP 1 cleavage in both cell lines (Figures 2C,D). Furthermore, colony formation assay results showed that the colony formation capacity of OCVAR-3 and SKOV3 cells decreased after ketamine treatment (Figure 2E). These data indicated that ketamine elevated cell cycle arrest and cell apoptosis, but decreased colony formation capacity in OCVAR-3 and SKOV3 cells.

ncRNAs have been reported to play important role in controlling ovarian cancer cell proliferation and survival (Wang et al., 2019). To investigate the differential lncRNA expression in ovarian cancer patients, we did bioinformatics analysis of a GEO dataset (GSE38666). The expression level of lncRNAs that have been reported to relate to ovarian cancer was compared between ovarian cancer and normal tissues, and demonstrated using heatmap and volcano plot as shown in Figure 3A and Supplementary Figure S2. Fourteen lncRNAs (PVT1, LINC00092, PTAF, SnaR, Meg3, MALAT1, ZFAS1, UCA1, MIR4697HG, TUG1, GAS5, DNM3OS, HOTAIR, and EWSAT1) were found significantly dysregulated, including several lncRNAs that has been shown to play important role in other tumors, such as PVT1, MALAT1, TUG1, GAS5, and HOTAIR (Figure 3B; Wang et al., 2019).

To further confirm the results from bioinformatics analysis, we evaluated the expression level of nine lncRNAs (PVT1, SnaR, Meg3, HOTAIR, MIR4697HG, TUG1, DNM3OS, UCA1, and EWSAT1), which showed more obvious differences in expression based on Figure 3B, in ovarian cancer cell lines. Compared with parental lines, lncRNAs PVT1, SnaR, Meg3, HOTAIR, and TUG1 were significantly overexpressed in OCVAR-3 and SKOV3 cells (Figure 4A), which is consistent with previous report (Wang et al., 2019). Interestingly, ketamine significantly decreased the expression level of lncRNA PVT1, but had no effect on other lncRNAs in OCVAR-3 and SKOV3 cells (Figures 4B,C). Results in Figures 4D,E further confirmed that ketamine regulated the expression of lncRNA PVT1 in a time-dependent manner.

To explore the mechanism of PVT1 overexpression in OC, we first analyzed the modification in the promoter of PVT1. Abundant H3K27 acetylation (H3K27ac) signals were found in the promoter region of PVT1, which suggested that PVT1 might be regulated by histone acetylation (Figure 5A). To confirm that, we performed chromatin-immunoprecipitation assay (ChIP) using H3K27ac antibody and primers covering 5 regions within∼1kb promoter region (Figures 5B,C). The results demonstrated that H3K27ac marks are highly enriched at the PVT1 promoter regions 3–5, and this enrichment was significantly decreased after treatment with ketamine (Figure 5C). H3K27ac is known to be catalyzed by the P300/CBP complex (Raisner et al., 2018). We then treated the OCVAR-3 with P300 specific siRNA, and the results showed that H3K27ac marks and PVT1 expression was significantly decreased (Figure 5D). Consistent with these findings, P300 was recruited to PVT1 promoter regions 3–5, and this recruitment was significantly decreased after ketamine treatment (Figure 5E). These data indicated that ketamine regulated lncRNA PVT1 expression via P300 mediated histone acetylation.

Enhancer of zeste homolog 2 (EZH2), a subunit of the polycomb repressive complex 2, was reported to contribute to the deregulation of OC cell growth. In addition, PVT1 was shown to bind EZH2 and improve its stability in hepatocellular carcinoma (Guo et al., 2018). We next examine the association of PVT1 and EZH2 in OC cell by performing RIP assay. As shown in Figure 6A, PVT1 bound EZH2 in OCVAR-3 and SKOV3 cells, and this interaction was significantly decreased after ketamine treatment. More importantly, the recruitment of EZH2 to the target gene-p57 promoter region was significantly inhibited by ketamine, and consequently, the expression level of p57 was significantly increased in OCVAR-3 and SKOV3 cells (Figures 6B,C).