The human SK-N-AS NB cell line was purchased from Heilongjiang Jiufeng Bioengineering Co., Ltd. The human SH-SY5Y NB cell line was purchased from Haixing Biosciences Co., Ltd. (Suzhou, China) and was authenticated by STR profiling. The human MO3.13 oligodendrocytic cell line (normal cells) was purchased from Cyagen Biosciences, Inc. These cells were initially cultured in Dulbecco's Modified Eagle Medium (DMEM; Gibco; Thermo Fisher Scientific, Inc.) containing 10% fetal bovine serum [FBS; Cellmax Cell Technology (Beijing) Co., Ltd.] and 1% penicillin-streptomycin (Gibco; Thermo Fisher Scientific, Inc.) in an incubator maintained at 37°C with a humidified environment of 5% CO₂ and 95% air. Afterwards, cells were transferred every 2–3 days depending on their density, and experiments began once they achieved 80–90% confluency.

Cell viability was measured using the MTT assay following the protocol by Cheng and Ying (23). In brief, a cell suspension (SK-N-AS/SH-SY5Y/MO13.3) in DMEM containing 10% FBS (10⁶ cells/ml) was inoculated into 96-well plates with 100 µl per well and cultured at 37°C for 24 h with 5% CO₂ and 95% air. The cells were then treated with SSa (purity: HPLC ≥98%, Baoji Herbest Bio-Tech Co., Ltd.) at various concentrations [0 µM (control), 5, 10, 15, 20 and 25 µM] for 24 h. Following the treatment, 10 µl of MTT solution (5 mg/ml) was pipetted into each well, allowing the cells to incubate for an additional 4 h at 37°C. The medium was discarded and 100 µl of DMSO was added to each well to dissolve the purple formazan while the absorbance was measured at 490 nm using a Microplate Reader (RT-6000; Rayto Life and Analytical Sciences Co., Ltd.). Cell viability was calculated as follows: Cell viability=[OD₄₉₀ (SSa)/OD₄₉₀ (control)] ×100%.

The methods were performed following the protocol by Cheng and Ying (23). In brief, SK-N-AS cells were digested and suspended in serum-free DMEM containing SSA (0 or 5 µM) at a final concentration of 10⁶ cells/ml. For the invasion assay, 50 µl diluted Matrigel (Matrigel and serum-free DMEM were prepared at a ratio of 1:3 at 4°C) was added to the upper Transwell chamber (24-well; pore size, 8 µm; Corning, Inc.) and incubated at 37°C for 5 h. Subsequently, DMEM containing 20% FBS was added to the lower Transwell chamber. Subsequently, 100 µl of SK-N-AS cell suspension was plated in the upper chamber and incubated at 37°C for 24 h. The cells that invaded into the lower surface of the Transwell membrane were fixed with 100% methanol (room temperature for 30 min) and stained with 0.1% crystal violet (room temperature for 20 min). Stained cells were observed by fluorescence inverted microscopy (Leica Microsystems GmbH). A total of five random fields were imaged to calculate the relative invasion rates based on the ratio of the number of invaded cells in the SSa group to the control group. For the migration assay, all the steps were the same as in the invasion assay but without the addition of Matrigel.

To isolate RNA, ~4 ml of SK-N-AS cells (10⁶ cells/ml) were seeded into a T25 flask. The cells were cultured for ~2 days at standard conditions (37°C for 48 h in humidified atmosphere of 5% CO₂ and 95% air) before being divided into two groups and treated for another 24 h: Those receiving SSa treatment (n=3) were incubated in FBS-free DMEM containing 15 µM SSa, and those in the control group did not receive any substance addition (n=3). Post-treatment, total RNA extraction from the SN-K-AS cells was performed using TRIzol Reagent (Invitrogen; Thermo Fisher Scientific, Inc.). NanoDrop spectrophotometer-based measurements (Thermo Fisher Scientific, Inc.) confirmed the concentration and purity levels alongside LabChip GX Touch HT Nucleic Acid analysis (PerkinElmer, Inc.).

The collected RNA samples (a total of 6, 3 for SSa treatment and 3 for control) were then transported to Wuhan Boyue Zhihe Biotechnology Co., Ltd., where library construction and sequencing tasks employing KAPA Stranded RNA-Seq Kit protocols plus multiplexing primer guidance were undertaken. In brief, mRNA was enriched by oligo(dT) beads and sequencing results were obtained using NovaSeq^™ 6000 systems (Illumina, Inc.).

Raw data (raw reads) were filtered into clean data (clean reads) using Trimmomatic v0.36 (24). This involved removing reads with adapters, reads containing ploy-N and low-quality reads from the raw data. Clean data were then aligned to the human reference genome (http://asia.ensembl.org/Homo_sapiens/Info/Index) using Hisat 2 (25). The number of perfect clean tags for each gene was calculated and normalized as fragments per kilo bases per million mapped reads using featureCounts v1.5.0 (26). DEGs were identified using the DESeq R package (1.10.1) (27), with the following criteria: Fold-change ≥1.5 or ≤0.667 and adjusted P<0.05 (28).

Gene Ontology (GO) enrichment analysis was conducted by Database for Annotation, Visualization, & Integrated Discovery (DAVID) to reveal the biological function of DEGs (29,30). Kyoto Encyclopedia of Genes and Genomes (KEGG) was utilized to analyze the metabolic and signaling pathways inherent in the DEGs; enrichment and network construction were performed by Metascape 3.5 (31,32). Protein-protein interaction (PPI) networks of putative proteins encoded by DEGs were constructed with STRING 12.0 (33).

Total RNA was isolated from SK-N-AS cells using the SV Total RNA Isolation System (Promega Corporation), then reverse transcribed into cDNA using the GoScript^™ kit (Promega Corporation) according to the manufacturer's protocol. qPCR was employed to assess DEG expression using the LineGene 9620 qPCR system (Hangzhou Bioer Co., Ltd.) following the GoTaq^® qPCR protocol (Promega Corporation). The primer sequences for amplifying the DEGs are listed in Table I. Relative gene expression level was calculated based on the 2^−∆∆Cq method, with GAPDH as the internal control gene (34).

Data are presented as mean ± standard deviation. Student's unpaired t-test and one-way analysis of variance with Dunnett's post hoc test were applied for the statistical analyses of the cell viability and qPCR results. Fisher's exact test was used for bioinformatic analysis. P<0.05 was considered to indicate a statistically significant difference. Statistical analysis was performed using SPSS 26.0 (IBM Corp.).

As depicted in Fig. 1A and B, after 24 h, SSa reduced the viability of human SK-N-AS and SH-SY5Y NB cells in a dose-dependent manner. The half maximal inhibitory concentration values for SSa in SK-N-AS and SH-SY5Y cells after 24 h were 14.5 and 15.8 µM, respectively. SSa had no inhibitory effect on normal MO3.13 cells at doses <25 µM (Fig. 1C). This indicated that SSa was cytotoxic to SK-N-AS and SH-SY5Y cells and was non-toxic to normal cells at the effective doses, which aligned with the results of a previous study (23). In subsequent experiments, 15 µM was selected for transcriptomic analysis.

Transwell assays were performed to verify the inhibitory effect of SSa on the migration and invasion of SK-N-AS cells. To prevent interference from the inhibitory effect of SSa on SK-N-AS cell viability, a concentration of SSa (5 µM) was used in these assays. As shown in Fig. 2, SSa significantly inhibited the migration and invasion of SK-N-AS cells, consistent with previous literature (23).

RNA-Seq was carried out to analyze the transcriptomic profile of the SK-A-AS cell samples, producing means of 42.5 million and 46.9 million clean reads for the control group and SSa treatment group, respectively. The overall mapped rates of clean reads to the human reference genome in all samples ranged from 92.41 to 94.68% (Table II). Compared with the control group, there were 297 DEGs in the SSa treatment group; among these genes, 23 were upregulated while 274 were downregulated (Table SI). The volcano plot in Fig. 3A illustrates the DEGs between the SSa treatment and control groups; genes were arranged based on their log2 fold change value along the x-axis and their-log₁₀ adjusted P-value along the y-axis. A hierarchical clustered heatmap was created to visualize the gene expression relationship patterns between the samples. The 3 samples from each group were clustered on the same branch, indicating that SSa treatment significantly changed the gene expression patterns in SK-N-AS cells (Fig. 3B).

GO annotation enrichment analyses within the DEGs were assessed to determine the significant biological function changes. A total of 717 GO terms were enriched including 610 biological processes, 59 cellular components and 48 molecular functions. In the biological process category, ‘cell adhesion’ (GO: 0007155), ‘biological adhesion’ (GO: 0022610), ‘neurogenesis’ (GO: 0022008), ‘extracellular matrix organization’ (GO: 0030198) and ‘extracellular structure organization’ (GO: 0043062) were the top five significantly enriched processes. Biological processes related to the antitumor activity of SSa on human NB cells were also enriched, such as ‘apoptotic process’ (GO: 0006915), ‘angiogenesis’ (GO: 0001525) and ‘epithelial to mesenchymal transition’ (GO: 0001837). In the molecular function category, ‘extracellular matrix structural component’ (GO: 0005201), ‘integrin binding’ (GO: 0005178), ‘glycosaminoglycan binding’ (GO: 0005539), ‘receptor binding’ (GO: 0005102) and ‘heparin binding’ (GO: 0008201) represented the major functions encompassed by the DEGs. Moreover, in the cellular component category, the DEGs primarily contributed to ‘extracellular matrix’ (GO: 0031012), ‘extracellular region’ (GO: 0005576), ‘extracellular region part’ (GO: 0044421), ‘cell surface’ (GO: 0009986) and ‘extracellular space’ (GO: 0005615). Fig. 4 illustrates the top 20 GO terms within each category.

The metabolic and signaling pathways contributing to SSa inhibition of SK-N-AS cells were also analyzed using the KEGG database. The results showed that the DEGs were enriched in 55 KEGG pathways (Table SII). Pathways related to the invasion and migration of cancerous cells were significantly enriched, such as ‘ECM-receptor interaction’ (hsa04512), ‘Focal adhesion’ (hsa04510) and ‘Cell adhesion molecules’ (hsa04514). In addition, 14 pathways belonged to signal transduction and 7 pathways were related to cancer. The signaling pathways included ‘PI3K-Akt signaling pathway’ (hsa04151), ‘TNF signaling pathway’ (hsa04668) and ‘Apelin signaling pathway’ (hsa04371). Cancer-related pathways included Metabolism of central carbon in cancer (hsa05230), ‘Proteoglycans in cancer’ (hsa05205) and Non-small cell lung cancer (hsa05223). Other pathways mainly encompassed human disease and metabolic pathways. The top 20 significantly enriched pathways are shown in Fig. 5A, with connections among the enriched pathways shown as a network in Fig. 5B.

The list of gene symbols was submitted to the STRING database to construct the PPI network of putative proteins encoded by the DEGs, with ‘organisms’ set to Homo sapiens and the ‘minimum required interaction score’ set to high confidence. As shown in Fig. 6, the largest network contained 98 nodes and 201 edges, while 139 isolated nodes without any connections were hidden. Genes such as FN1, COL1A1, DDX58, PTPRC, THBS1, CCL2, IFIH1, IL1B, MX1 and VCAM1 ranked among the 10 most-connected genes and acted as hubs in the network. A total of 29 clusters were obtained based on Markov clustering, with the largest cluster containing 21 genes whose functions mainly referred to cell adhesion (hsa04512, hsa04514, hsa04510, GO:0007155, GO:0030155 and GO:0016477).

To verify the transcriptomic sequencing results, genes related to the SSa inhibition of SK-N-AS cells were determined by qPCR. These genes included IL24, EGR1, RET, MDK, PDGFRA, HGF, VCAM1, SLIT3, CD34, FN1, COL1A1 and NCAM1. In the group treated with SSa, certain gene expression levels increased (IL24 and EGR1) or decreased (RET, MDK, PDGFRA, HGF, VCAM1, SLIT3, CD34, FN1, COL1A1 and NCAM1) significantly compared with the control group (Fig. 7). Therefore, the RNA-seq and qPCR data were consistent.