Gene expression profiling
We downloaded the publicly available GSE41804 gene expression datasets from NCBI-GEO. These studies compared gene expression profiles in HCC and non-cancer liver tissue samples. The data were generated using the GPL570 platform based on 40 tumour and 40 non-cancer tissue samples.
HCCDB database analysis
The online Integrative Molecular Database of Hepatocellular Carcinoma (HCCDB) database was used to confirm the expression of TOP2A. HCCDB (http://lifeome.net/database/hccdb) contains 15 public gene expression datasets of HCC containing 3917 samples16. We used it to analyse the TOP2A expression status in HCC and the corresponding adjacent tissues. The expression levels of TOP2A in different datasets were also detected. Samples were classified into high/low expression groups by median expression value of TOP2A.
Oncomine database analysis
To further evaluate the expression level of TOP2A in the HCC cohort, the Oncomine 4.5 database (https://www.oncomine.org) was used. Oncomine is a cancer microarray database and web-based data-mining platform 17. The analysis of mRNA expression fold change was filtered by selecting HCC vs. non-cancer tissue analysis. The standardised normalisation and parameters provided on the Oncomine platform were as follows: P-value < 0.001, fold change > 2, and gene ranking in the top 10%.
Human Protein Atlas
The Human Protein Atlas (HPA) database (www.proteinatlas.org) provides access to 32 human tissues and their protein expression profiles and uses antibody profiling to accurately assess protein localisation 18. Immunohistochemistry results of complement TOP2A in HCC tissues and non-cancer tissues were acquired from the HPA and were used to generate the images of cell cycle progression in 97H cells with the Premo™ FUCCI Cell Cycle Sensor.
UALCAN database analysis
UALCAN (http://ualcan.path.uab.edu) is an interactive web resource based on 31 cancer types available from TCGA database. It can be used to analyse the relative transcriptional expression of potential genes of interest between tumour and non-cancer samples, as well as in various tumour subgroups based on individual cancer stages, tumour grade, or other clinicopathological features 19. Here, UALCAN was used to analyse TOP2A expression in primary HCC tissues and their association with clinicopathologic parameters. The difference in transcriptional expression was compared using Students’ t-test, and P < 0.01 was considered statistically significant.
Kaplan-Meier plotter analysis
As a tool for assessing biomarkers, the Kaplan-Meier plotter (http://kmplot.com) can be used to assess the functions of 54,675 genes and 10,188 tumour tissue samples, including breast, liver, lung and gastric cancer 20–23. In this study, Kaplan-Meier curves were used to compare the differences in OS, PFS, RFS, and DSS in HCC patients separated by the median expression level of TOP2A. The prognostic values of high and low expression groups were evaluated according to the hazard ratios and log-rank P-values. A log-rank P-value of < 0.05 indicated the significance of survival time differences.
LinkedOmics database analysis
The LinkedOmics database (http://www.linkedomics.org/login.php) is a web-based platform for analysing 32 TCGA cancer-associated multidimensional datasets 24. TOP2A co-expression was analysed statistically using Pearson’s correlation coefficient and presented in volcano plots, heat maps, or scatter plots. Afterwards, the Function module of LinkedOmics, which performs GSEA results, was carried out to identify the potential underlying mechanisms of TOP2A expression on the pathogenesis and prognosis of HCC. The nominal P-value (NOM P < 0.05) and false discovery rate (FDR q < 0.25) were used to select significantly enriched gene sets.
STRING database
STRING (https://string-db.org/cgi/input.pl) is a database of known and predicted protein-protein interactions. We utilised STRING to create an interaction network between TOP2A and other important proteins involved in HCC pathology 25.
Gene function analysis
The GO; www.geneontology.org/) database describes the cellular components, molecular functions, and biological processes of TOP2A 26. The KEGG (www.genome.ad.jp/kegg/) database, which includes information on TOP2A and its functions, was used for pathway mapping 27.
CancerSEA
CancerSEA (http://biocc.hrbmu.edu.cn/CancerSEA/) was used to explore the potential roles of TOP2A in cancer 28. Particularly, CancerSEA was used to explore whether the module was associated with the carcinogenic process. A P < 0.05 was considered statistically significant.
Cistrome DB toolkit
The Cistrome DB (http://dbtoolkit.cistrome.org) is a resource of human and mouse cis-regulatory information derived from ChIP-seq, DNase-seq, and ATAC-seq chromatin profiling assays. In this study, the Cistrome DB was used to map the genome-wide locations of TF binding sites, histone post-translational modifications, and the regions of chromatin accessible for TOP2A regulation 29.
cBioPortal database analysis
The cBioPortal for Cancer Genomics (http://cbioportal.org) has multidimensional cancer genomics data sets 30. We studied the alterations of TOP2A in the hepatocellular carcinoma (TCGA, Firehose Legacy) case set using cBioPortal, in which mutations, CNV and gene co-occurrence of TOP2A in HCC were analysed. In addition, the tab OncoPrint displayed an overview of genetic alterations per sample in TOP2A.
TIMER database analysis
The TIMER correlation module was used to evaluate potential relationships between TOP2A expression and tumour-infiltrating immune cells (TIICs). TIMER is a comprehensive resource for systematic analysis of immune infiltrates across diverse cancer types (https://cistrome.shinyapps.io/timer/), which includes 10,897 samples across 32 cancer types 31. TIMER employed a recently published statistical method known as deconvolution 32 to deduce the prevalence of TIICs from gene expression profiles. We evaluated TOP2A expression in HCC and its correlation with the abundance of TIICs, including CD4+ T cells, dendritic cells, B cells, CD8+ T cells, B cells, neutrophils and macrophages as well as the tumour purity via gene modules.
TISIDB database analysis
The TISIDB database integrates 988 reported immune-related anti-tumour genes, high-throughput screening techniques, molecular profiling and paracancerous multi-omics data as well as various resources for immunological data retrieved from seven public databases 33. In this study, the TISIDB was employed to analyse the relationship between the abundance of both tumour-infiltrating lymphocytes (TILs) and immunoinhibitors and the expression levels of TOP2A.
CIBERSORT
We applied an established computational approach (CIBERSORT) (https://cibersort.stanford. edu/) to estimate the relative proportions of 22 immune cell types. Using CIBERSORT 34, we measured the immune response of 22 TIICs to evaluate their association with TOP2A expression in HCC and to uncover correlations between TIICs. The results were filtered by setting the maximum P-value to 0.05. Comparisons of relative fractions were performed using the Mann–Whitney U test.
Small molecule drugs screening
To analyse the correlation of TOP2A expression with drug sensitivity, we collected 481 small molecules from the Cancer Therapeutics Response Portal (CTRP) 35 and 265 small molecules from the Genomics of Drug Sensitivity in Cancer (GDSC) 36. The expression of TOP2A was assessed using Spearman’s correlation analysis of small molecule/drug sensitivity (IC50). To further explore the potential mechanism of prospective drugs, the information of drugs with significant scores as listed in the results and their 2D molecular structures as provided by DrugBank (https://www.drugbank.ca/) was displayed.
Validation of cell experiment
The HCC cell lines MHCC97H, Bel7404, SMCC7721, Hep3B, Huh7, and HepG2 and non-cancer L02 cells were cultured in DMEM medium with 10% foetal bovine serum (Gibco, NY, USA). Culture conditions were carried out at 37°C in an incubator with 5% CO2. Thereafter, total RNA extraction, reverse transcription, and quantitative real-time polymerase chain reaction (qRT-PCR) were performed as previously described 37. The primers used in this article are listed as follows: forward primer: 5'-ACCATTGCAGCCTGTAAATGA-3'; reverse primer: 5'-GGGCGGAGCAAAATATGTTCC-3'. For WB, the protein was extracted from cells using RIPA lysis buffer and quantified using the BCA protein assay kit (Thermo Scientific, USA). Equal amounts of protein were then separated through SDS-PAGE and transferred to PVDF membranes. Blots were blocked with 5% skim milk in TBST buffer, followed by overnight incubation with primary antibodies at 4°C. Finally, the blots were probed for 1 h with appropriate secondary antibodies (1:5000), and chemiluminescence was then used for protein visualisation. Immunofluorescence: cells were grown on coverslips in 12-well plates. Then, the cells were blocked with 5% BSA in PBS for 1 h at room temperature. Coverslips were incubated with anti-TOP2A (1:100/Beyotime) primary antibodies overnight at 4°C. The coverslips were washed with PBST and incubated for 1 h with Alexa Fluor 488-labelled secondary antibody. DAPI stained the nuclei. Images were taken under a Zeiss780 confocal laser-scanning microscope (Germany).