The human A2780 OC cell line [cat. no. iCell-h004, Mirror Qidian (Shanghai) Cell Technology Co., Ltd.] was cultivated in RPMI-1640 medium (Gibco; Thermo Fisher Scientific, Inc.), supplemented with 10% FBS and 1% antibiotic-antimycotic solution (all Gibco; Thermo Fisher Scientific, Inc.) at 37°C and 5% CO₂ atmospheric conditions. Cells were incubated with 75 µM RES (Beijing Solarbio; Beijing, China) at 37°C for 24 h, where indicated, a dose previously reported to approximate an IC₅₀ dose (21). The cells were confirmed to be free of Mycoplasma contamination and subjected to short tandem repeat identification.

A2780 cells were treated with or without 75 µM RES for 24 h, and TRIzol™ (Invitrogen™; Thermo Fisher Scientific, Inc.) were used to isolate total RNA from cells. RNA quality was quantified using RNase-free agarose gel electrophoresis using an Agilent 2100 Bioanalyzer (Agilent Technologies, Inc.). After eliminating ribosomal RNAs from total RNA, the samples were diluted in fragmentation buffer before reverse transcription of the mRNA/non-coding RNA transcripts with random primers using the VAHTS Universal V6 RNA-seq Library Prep Kit for Illumina (Vazyme Biotech Co., Ltd., Nanjing, China, Cat no. NR604). First-strand cDNA synthesis was performed at 25°C for 10 min, 42°C for 15 min, and 70°C for 15 min; second-strand cDNA synthesis was then performed at 16°C for 30 min, 65°C for 15 min. The cDNA fragments were purified using the QiaQuick PCR extraction kit (cat no. 28104, Qiagen, Inc.) and subsequently ligated to an Illumina sequencing adapter (Illumina, Inc.), to construct cDNA libraries. After digestion, PCR products were amplified and the library concentration was measured using the DNA 1000 assay kit (Agilent Technologies, Inc.), with quantification and pooling performed using the ABI StepOnePlus Real-Time PCR System (Thermo Fisher Scientific, Inc.). The libraries, with a loading concentration of 3 ng/µl, were then sequenced by Guangzhou Gene Denovo Biotechnology Co., Ltd. using the HiSeq 2500 Sequencing System (Illumina, Inc.) with a paired-end 150 bp sequencing strategy. Raw sequence reads were screened using fastp (version 0.18.0) to eliminate low-quality bases and adapter sequences, and high-quality clean reads were obtained (22). Once the reference genome index was constructed, clean reads from RNA sequencing were mapped to the reference genome using HISAT2 (version 2.1.0) (23). Transcript reconstruction was performed using StringTie (version 1.3.4) in combination with HISAT2 to identify the splice variants of known and novel genes (24–26). The protein-coding ability of the novel transcripts was evaluated using CNCI (version 2), CPC (version 0.9-r2) and FEELNC (version v0.2) with default parameters (27–29). The non-protein-coding ability results were intersected as lncRNAs, which were then divided into five types according to their comparison with protein-coding genes: Bidirectional, intergenic, antisense, intronic and sense-overlapping lncRNAs. For lncRNAs and mRNAs, the differential expression of their transcripts was assessed separately using DESeq2 (version 1.20.0) between groups and edgeR (version 3.20.9) between samples (30,31). Differential expression of genes/transcripts was determined using |fold change|≥2 and false discovery rate of <0.05.

RNAplex (version 0.2) and the Vienna RNA package were used to predict the complementary relationships between antisense lncRNAs and mRNA (32). Optimal base pairing was predicted using minimal free-energy prediction based on thermodynamics. lncRNAs can function to cis-regulate the genes surrounding an identical allele (33). Upstream lncRNAs that interact with promoters or additional cis-elements can modulate gene expression transcriptionally or post-transcriptionally (33). Typically, lncRNAs in the 3′untranslated region or downstream can show additional regulatory effects. Therefore, lncRNAs annotated as ‘unknown region’ were reannotated as cis-regulators at <10 kb up/downstream of one specific gene. lncRNAs can also trans-regulate long-distance genes (33). Pearson's correlation coefficients were calculated between lncRNA and protein-coding genes, with the pairs satisfying |correlation|>0.999, and Gene Ontology (GO) and Kyoto Encyclopedia of Genes and Genomes (KEGG) analyses were performed. Moreover, the functions of the lncRNA parental genes were evaluated using GO and KEGG analyses. GO annotation was performed using GO resources (http://www.geneontology.org/).

Based on the results of KEGG enrichment analysis, combined with published evidence supporting the anticancer mechanism pathways, including the p53 signaling pathway, oocyte meiosis and cell cycle pathways, differentially expressed lncRNAs (DELs) and mRNAs and in these pathways were screened for co-expression analysis (34–36). The lncRNA-mRNA co-expression network was visualized using Cytoscape software (version 3.9.1, cytoscape.org/).

RT-qPCR was performed to assess DELs and mRNA expression. Total RNA was extracted using the Ultrapure RNA kit (DNase I, Cat no. CW0597S, Jiangsu Cowin Biotech Co., Ltd). cDNA was synthesized using Easyscript one-step gDNA removal and cDNA synthesis supermix (Cat no. AE-311, TransGen Biotech Co., Ltd.) with incubation at 42°C for 1 h. Specific primers were designed for RT-qPCR assays in a 20 µl reaction (Table SI). Reactions were performed using a Light-Cycler96 instrument (Roche Diagnostics) with the TransStart green qPCR supermix containing SYBR GREEN (Cat no. AQ131-01, TransGen Biotech Co., Ltd.). The thermocycling conditions were as follows: pre-denaturation at 94°C for 30 sec, followed by 40 cycles with denaturation at 94°C for 5 sec and annealing at 61°C for 35 sec. Post-amplification, melting curve analysis was performed at 95°C for 10 sec, 65°C for 60 sec, and 97°C for 1 sec. Relative RNA expression subsequently calculated as 2^−ΔΔCq values based on GAPDH as the housekeeping gene (37). Each experiment was performed in triplicate to ensure reproducibility.

pcDNA3.1-SNHG15-203 overexpression plasmids (Nanjing GenScript Co., Ltd.) and small interfering (si)RNAs targeting SNHG15-203 (Table SII, Shanghai GenePharma Co., Ltd.) were constructed and transfected into A2780 cells. Specifically, 200 nM siRNA duplexes or the overexpression plasmid, and 2.5 µl Lipofectamine™ 2000 (Thermo Fisher Scientific, Inc.) were diluted in 100 µl Opti-MEM, mixed, and incubated at 25°C for 20 min before being added to each well of a 12-well plate containing target cells. An additional 800 µl Opti-MEM was introduced, followed by gentle shaking. Transfection was performed at 25°C. The transfection mixture was replaced with fresh medium without antibiotics after 6–8 h, and the cells were further cultured at 37°C for 48 h before harvesting. RT-qPCR was performed to validate the overexpression and knockdown efficiency, as aforementioned.

Following transfection, cell proliferation was assessed using a CCK-8 kit (Biosharp Life Sciences). Briefly, the CCK-8 reagent was added to the treated cells cultured in a 96-well plate. After incubation for 48 h, the optical density of each well was measured at 450 nm using a microplate reader (NanJing DeTie Laboratory Equipment Co., Ltd.). Each experiment was performed in triplicate.

Cell migration was assessed using Transwell chambers (Corning, Inc.) with a pore size of 8 µm. A total of 2×10⁵ cells in 1 ml serum-free medium was added to the upper Transwell chamber, and 600 µl complete medium containing 10% FBS was added to the lower chamber. After 24 h incubation at 37°C, cells were fixed in paraformaldehyde at room temperature for 15 min and stained with crystal violet at room temperature for 10 min. The number of migrating cells in random areas was counted under an inverted microscope (IX51; Olympus Corporation).

GraphPad Prism statistical software (version 8.0; Dotmatics) was used for experimental data processing. For comparisons between two groups, an unpaired two-tailed Student's t-test was used if the data were normally distributed and had equal variances; otherwise, the Mann-Whitney U test was applied. For comparisons among multiple groups, one-way analysis of variance was used for parametric data, followed by Tukey's HSD test for multiple comparisons; otherwise, the Kruskal-Wallis H-test was used, followed by Dunn's test for post hoc pairwise comparisons. Pearson's correlation coefficient was used for correlation analysis. Data are expressed as mean ± standard error of the mean of ≥3 independent experimental repeats. P<0.05 was considered to indicate a statistically significant difference.

The effects of RES treatment on the migration of the OC cell line, A2780, were assessed. Transwell assays revealed that cells in the control group had a strong migration ability. The number of cells passing through the membrane was high, and numerous cells were observed on the lower surface of the membrane in the field of view (Fig. 1). By contrast, the migration ability of cells treated with RES for 24 h was notably altered, with the number of cells crossing the membrane was greatly reduced. The difference was significant compared with that in the control group (P<0.01). These results indicate that RES significantly inhibited the migration of A2780 OC cells.

To evaluate the potential mechanism underlying the inhibitory effect of RES on the proliferation of OC cells, lncRNA-sequencing analysis was performed on the OC cell line A2780 in the control and RES-treated groups. Novel lncRNAs were first identified and categorized into the following five groups based on their positions in the genome: 32,988 sense lncRNAs, 10,687 antisense lncRNAs, 7,316 intronic lncRNAs, 4,552 bidirectional lncRNAs and 35,927 intergenic lncRNAs (Fig. 2A). Pearson's correlation analysis demonstrated that lncRNA expression among biological replicates within groups showed a high degree of similarity, with significant differences between groups (Fig. 2B). Further analysis revealed 721 DELs, of which 461 were upregulated and 260 were downregulated (Fig. 2C). Hierarchical clustering analysis demonstrated significant differences in the lncRNA expression patterns between the two groups (Fig. 2D). Notably, lncRNAs H19-203 and vacuole membrane protein 1 (VMP1)-217 were highly expressed in the RES-treated group, whereas lncRNAs enolase 1 (ENO1)-210, SNHG15-203 and SNHG1-266 were markedly downregulated (Table SIII). Meanwhile, 24 SNHG family lncRNAs were identified, including lncRNA SNHG15-203.

Previously, mRNA sequencing in the RES-treated OC cell line A2780 revealed significant changes in mRNA expression in RES-treated cells and clarified its key role in cell proliferation and immunomodulation. Based on this, the potential roles of lncRNAs and their synergistic regulation of mRNAs were further explored. Compared with mRNAs, both annotated and novel lncRNAs exhibited fewer exons, shorter lengths and lower coding potentials (Fig. 3A-C). Analysis of total lncRNA expression data demonstrated markedly lower levels than those of mRNA, and the abundances of lncRNAs and mRNAs were notably increased by RES treatment (Fig. 3D). Differences in transcript abundance were also observed between DELs and mRNAs (Fig. 3E). In addition, the coding potential scores of differentially expressed mRNAs were higher than those of the lncRNAs, and those of novel lncRNAs were significantly lower than those of the annotated lncRNAs (Fig. 3F).

To further assess the potential function of lncRNAs in the inhibition of OC cell proliferation by RES, the complementary binding relationship between lncRNAs and mRNAs was analyzed and the potential function of lncRNAs was inferred based on the functions of known mRNAs. The DELs were first predicted for cis and trans roles, and 5,503 and 1,780,009 lncRNA-mRNA target gene pairs were obtained, respectively. Subsequently, GO functional enrichment analysis of the DELs was performed. The results revealed that the target genes associated with these differentially expressed cis-acting lncRNAs were significantly enriched in several key biological processes, such as response to viruses, regulation of signal transduction by p53 class mediators, DNA metabolic processes and response to oxidative stress (Fig. 4A). KEGG enrichment analysis further indicated that these cis-acting lncRNAs were mainly involved in carcinogenesis, Notch, Hippo, cell cycle and other classical tumor-related signaling pathways (Fig. 4B). The Reactome pathways involved also covered cell cycle progression and antiviral activity, which corroborated the GO and KEGG analysis results (Fig. 4C). the GO function of the differentially expressed trans-acting lncRNAs were analyzed, which revealed that they were mainly enriched in signal transduction and response to stress and stimuli (Fig. 4D). The results of KEGG enrichment analysis demonstrated that the trans-acting lncRNAs were mainly involved in p53, Forkhead Box (Fox)O signaling pathway, cellular senescence, extracellular matrix-receptor interaction and cancer-related signaling pathways, whilst the Reactome pathways included deacetylation of histones, Signaling by Rho GTPases and phosphorylation (Fig. 4E and F).

Furthermore, certain genes were revealed to be involved in multiple important biological functions (Fig. 4A-F). For example, cyclin dependent kinase inhibitor 1A (CDKN1A) and sequestosome (SQSTM) were enriched in cellular entries, anatomical entities, binding, cellular processes and biological regulation, and were associated with cellular senescence and cell cycle signaling pathways. Reactome analysis revealed certain specific pathways, such as estrogen-responsive MYC gene expression, lysine demethylase 6B demethylation of H3K27me3 on the p16-INK4A promoter, CDK1 phosphorylation of PHD finger protein 8, binding of serum amyloid P to DNA and chromatin, and binding of methyl-CpG binding domain protein 2 to methylcytosine in chromatin. These specific pathways contain several members of the histone H2A family, such as H2AC7, H2AC20 and H2AC18. Therefore, we hypothesize that the regulatory network centered on CDKN1A, SQSTM1 and histones may serve a crucial role in the inhibition of A2780 cell proliferation by RES.

A total of five mRNAs associated with the p53 signaling pathway, oocyte meiosis, cellular senescence, cell cycle and their associated lncRNAs were selected and lncRNA-mRNA regulatory networks were constructed using Cytoscape. The results revealed that CDKN1A and SQSTM1 formed a large regulatory network, whereas the other genes, H2AC7, H2AC20 and H2AC18, formed a small network. Moreover, there was no overlap in the lncRNAs involved in the two networks (Fig. 5 and Table SIV). In the large network, CDKN1A and SQSTM1 demonstrated extensive regulatory associations, with shared regulatory links with 26 lncRNAs, including lnc-SQSTM1-206 and MSTRG.5444.1. It also formed unique regulatory relationships with MSTRG.22456.2 and MSTRG.3858.2. In the small network, H2AC7 and H2AC20 were targeted by five lncRNAs, including SNHG15-203 and cysteinyl-tRNA synthetase 1–212. A specific linkage between H2AC18, H2AC2 and MSTRG.7008.104 allowed the integration of the three histone genes into the same network.

To verify the accuracy of the data, five lncRNAs (ENO1-210, SNHG1-266, SNHG15-203, VMP1-217 and H19-203) were randomly selected and their expression levels were assessed using RT-qPCR. The expression levels of ENO1-210, SNHG1-266 and SNHG15-203 were significantly downregulated, and the expression levels of VMP1-217 and H19-203 were significantly upregulated in the RES-treated group compared with in the control group (Fig. 6A). These results are consistent with the RNA sequencing results (Table SIII).

The expression levels of key mRNAs associated with the lncRNAs were further evaluated. A total of five key genes, CDKN1A, SQSTM1, H2AC7, H2AC20 and H2AC18, were evaluated for changes in expression levels, confirming alterations in the mRNA levels of each gene after RES treatment. Among these, CDKN1A and SQSTM1 were significantly upregulated, whereas H2AC7, H2AC20 and H2AC18 were significantly downregulated, compared with in the control group (Fig. 6B).

To assess the predicted regulatory network, lnc-SNHG15-203 was selected for an in-depth study, as it was predicted to regulate both H2AC7 and H2AC20. First, four siRNAs and plasmids were designed to knock-down and overexpress lnc-SNHG15-203 in A2780 cells, respectively. The knockdown efficiencies of different siRNAs varied, with the knockdown efficiency of siRNA-2 at >70%, whereas that of siRNA-1 was ~30% (Fig. 7A). By contrast, the overexpression group demonstrated significant upregulation, with the expression level increasing by ~12-fold compared with that in the control group (Fig. 7B). For further validation, siRNA-1 and siRNA-2 were also used. The CCK-8 cell proliferation assay results demonstrated that both siRNA-1 and siRNA-2 significantly inhibited the proliferation of A2780 cells compared with the siRNA negative control, and lnc-SNHG15-203 overexpression notably promoted cell proliferation compared with the empty vector control group (Fig. 7C). Compared with the negative control group, the expression levels of its potential target genes, H2AC7 and H2AC20, were notably downregulated after lnc-SNHG15-203 knock-down (Fig. 7D and E). Conversely, compared with the empty vector control group, lnc-SNHG15-203 overexpression significantly upregulated H2AC7 and H2AC20, with their expression levels increased by ~2- and 3-fold, respectively. These results suggest that knockdown of lnc-SNHG15-203, a type of lncRNA that is downregulated by RES, could inhibit the expression of its target genes and cell proliferation, whereas its overexpression exerts the opposite effects.