Introduction
Cancer in most nations gradually displaces cardiovascular disease as the leading cause of premature mortality [
1]. According to statistics, there will be 19.3 million new cancer cases and almost 10 million cancer-related deaths in 2020. The global cancer burden is expected to reach 28.4 million patients by 2040 [
2]. Cancer treatment has steadily transitioned into the era of precision therapy because of the advanced high-throughput sequencing technology. Advanced cancer patients frequently receive targeted therapy and immunotherapy in the clinic. For instance, palbociclib (a CDK4/6 inhibitor) in conjunction with letrozole (24.8 months) resulted in considerably longer progression-free survival than letrozole alone (14.5 months) in patients with previously untreated ER-positive, HER2-negative advanced breast cancer [
3]. Due to the identification of molecular targets and the growing understanding of their cellular effects, small molecule inhibitors have been created as a primary therapeutic approach for cancer treatment. The discovery of new molecular targets may contribute to advancing combined therapies, eradicating drug resistance, and increasing therapeutic efficacy. Small molecule inhibitors have a bright future; thus, scientists should investigate novel biomarkers to find potential therapeutic targets.
The cell cycle is a tightly controlled process that supports genetic material replication and cell proliferation. One of the distinguishing characteristics of malignancies is abnormal cell proliferation brought on by cell cycle dysregulation [
4]. Dysregulation of the cancer cell cycle facilitates cell proliferation, which is driven by excessive mitotic signaling, inhibitory checkpoint failure, or both. Scientists have demonstrated that it is effective to target proteins associated with the cell cycle in order to limit tumor growth. The most notable successes in targeting cell cycle mechanisms have been inhibitors of CDK4 and CDK6 [
5‐
7]. The introduction of these CDK4/6 inhibitors for clinical use constitutes a milestone in the treatment of breast cancer and may have broad ramifications for the management of other tumor types [
8]. Despite the current success of CDK4/6 inhibitors, cancer therapy that targets cell cycle proteins is still in its infancy. An in-depth exploration of cell cycle regulatory mechanisms and their role in cancer will guide current cancer treatments and identify new therapeutic targets.
Establishment of sister chromatid cohesion N-acetyltransferase 2 (ESCO2) was identified as an effective target for cancer therapy, which is a pivotal protein in the cell division process [
9]. ESCO2, the human homolog of yeast ECO1, works on proliferating cell nuclear antigen (PCNA) to stimulate sister chromatid cohesion [
10,
11]. Acetylation of the SMC3 subunit of the adhesive protein via ESCO2 acetyltransferase facilitates sister chromatid cohesion, inhibiting cohesin release from chromatin [
12]. The cohesin is a multiprotein complex whose typical role is to bind sister chromatids from S-phase to anaphase to prevent premature segregation of sister chromatids and to ensure equal segregation of chromosomes [
13,
14]. Hence, the presence of ESCO2 ensures correct chromosomal segregation and makes recombinant DNA repair possible. Recent research has demonstrated that ESCO2 is strongly linked to the formation of several cancer types [
15‐
19]. ESCO2 knockdown in human gastric cancer cell lines in vitro significantly inhibited cell proliferation and induced apoptosis by regulating P53 [
17]. Meanwhile, ESCO2 can promote LUAD cell proliferation and metastasis by promoting aerobic glycolysis [
16]. Similarly, Fu et al. identified that 53BP1-MDS ring-like structure disruption caused by ESCO2 depletion in colorectal cancer cells, which reduced the effectiveness of non-homologous end joining (NHEJ) repair and made cancer cells more susceptible to chemotherapy [
20]. However, Guo et al. discovered that ESCO2 overexpression in colon cancer reduced MMP2’s transcriptional activity to limit tumor metastasis [
19]. ESCO2’s involvement in cancer is therefore controversial. Investigating ESCO2’s biological function in other malignancies is necessary. A pan-cancer analysis can investigate the oncogene in several malignancies, allowing oncogenes’ effector function and differential expression analyses.
We identified ESCO2’s abnormal expression value in tumor and normal samples by pan-cancer analysis and then validated its protein value in clinical tissues. Subsequently, we confirmed the value of ESCO2 in assessing prognosis and performed gene enrichment analysis, single cell analysis and immuno-infiltration analysis for ESCO2.Moreover, we conducted in vitro experiments to corroborate our bioinformatics findings. ESCO2 is a unique and useful biomarker and a prospective molecular target.
Materials and methods
Data preparation and differential expression analysis
The Cancer Genome Atlas (TCGA) database (
https://portal.gdc.cancer.gov/), which has 11,315 total samples, was employed to download the bulk mRNA sequencing profiles of 33 distinct cancer samples and their corresponding normal tissues. All gene expression data were standardized using the Transcripts Per Kilobase of Exon Model per Million Mapped Reads (TPM) for all samples. In the meanwhile, we collected the relevant clinical information on cancer patients. The Progression Free Interval (PFI), Disease Free Interval (DFI), and Disease-Specific Survival (DSS) timings were retrieved from UCSC Xena(
https://xena.ucsc.edu/), and the Overall Survival (OS) time was acquired from TCGA.
Then, using the TIMER [
21] and Xena Shiny databases (
https://shiny.hiplot.cn/ucsc-xena-shiny/)-powerful online platforms that incorporated mRNA sequencing data from the TCGA and GTEx datasets [
22]-we analyzed the ESCO2’s mRNA expression value between tumor and normal samples for 33 types of cancer. Furthermore, the ESCO2’s protein level was determined using the UALCAN online platform (
https://ualcan.path.uab.edu/index.html). Finally, three pairs of paraffin-embedded kidney renal clear cell carcinoma (KIRC), bladder urothelial carcinoma (BLCA), and adjacent samples were collected from the Second Hospital of Tianjin Medical University. The pathologic type of these individuals was clear cell renal cell carcinoma (ccRCC, i.e., KIRC) or BLCA, and none of them received preoperative chemotherapy. More crucially, all patients have approved the use of the surgical material for academic research and publications. All methods were approved by The Institutional Review Committee and the Medical Ethics Committee of the Second Hospital of Tianjin Medical University. Written informed consents have been obtained from all subjects.
Survival analysis and relationship with clinical stage
We filtered the cancer patients to assess the ESCO2’s impacts on the cancer cases’ prognosis. (a) The average value of several samples from the same patient was thought to represent the ESCO2 expression level. (b) cancer patients with a follow-up time of zero were excluded. After that, the survival R package was applied to run the survival analysis. The survfit function was utilized to model the Kaplan-Meier survival curve. The best-cutoff point was discovered using the “surv cutpoint” function in a survminer R package, which can repeatedly check all viable cutting points to get the highest rank statistic. We next divided each cancer type’s patients into two groups using the appropriate cutoff point for gene expression. The Kaplan-Meier survival curves were compared based on the two-sided long-rank test. Moreover, we performed Receiver operating characteristic curve (ROC) testing to assess the ESCO2’s ability to predict cancer across all types. Then, we investigated the relationship between ESCO2’s expression level and clinical parameters, including tumor size, stage, and grade.
Functional enrichment analysis of ESCO2
We extracted the RNA sequencing profiles of cancer types where ESCO2 has prognostic value to investigate ESCO2’s carcinogenesis in malignancies. Hence, 12 cancer types, including thymoma (THYM), colon adenocarcinoma (COAD), rectum adenocarcinoma (READ), adrenocortical carcinoma (ACC), kidney chromophobe (KICH), KIRC, kidney renal papillary cell carcinoma (KIRP), brain lower grade glioma (LGG), liver hepatocellular carcinoma (LIHC), lung adenocarcinoma (LUAD), mesothelioma (MESO), and pancreatic adenocarcinoma (PAAD), were rolled into the next analysis. For each cancer type, individuals were grouped into ESCO2-high and ESCO2-low subgroups relied on the median value of ESCO2, respectively. After that, gene set enriched in two groups were detected by gene set enrichment analysis (GSEA). Gene sets, the 50 hall markers, were obtained from MSigDB [
23] (
https://www.gsea-msigdb.org/gsea/msigdb).
The above data shows that ESCO2 is a negative oncogene in most cancers. Therefore, we extracted the RNA-seq matrix from KICH, KIRC, LGG, LIHC, LUAD, and PAAD to study ESCO2-related regulatory genes. ESCO2 had a negative impact on these cancer types, which also accounted for more than 150 cases. Similarly, we split patients into ESCO2-high and ESCO2-low groups based on the median value of ESCO2 in each cancer type. Following that, differentially expressed genes (DEGs) between ESCO2-high and the ESCO2-low group were performed using limma R packge [
24](
p < 0.05, Log FC > 1). After identifying the intersecting genes of DEGs in each cancer, we calculated the association between ESCO2 and these DEGs by spearman’s correlation analysis for each cancer type, respectively (
p < 0.05, Cor > 0.4). Intersecting genes for these highly correlated genes were considered ESCO2-related regulatory genes. Then, using the clusterProfiler R [
25] package, the Kyoto Encyclopedia of Genes and Genomes (KEGG) [
26‐
28] and Gene Ontology (GO) pathway enrichment analyses were performed on ESCO2-related regulatory genes. Incorporating ESCO2-related regulatory genes into the String database (
https://cn.string-db.org/) to form a PPI network enables the identification of core genes.
Single-cell analysis of ESCO2 in cancers
Tumor Immune Single-cell Hub (TISCH,
http://tisch.comp-genomics.org/home/) is a robust database created for multiple single-cell analyses containing nearly 190 single-cell databases. We estimated the ESCO2 expression level across multiple cancer types in each cell type. Our study focused on the distribution of ESCO2 in KIRC cell subpopulations. The expression levels of ESCO2 in each cell type were quantified and visualized by a heatmap, scatter diagrams, and violin plots. In addition to the single-cell dataset of KIRC contained in the TISCH database, we also downloaded additional single-cell sequencing data from GSE156632 [
29] in the Gene Expression Omnibus (GEO) database to validate ESCO2 expression level in each cell types.
Cell culture and siRNA transfection
The Cell Resource Center Affiliated with the Chinese Academy of Medical Sciences provided the human ccRCC cell lines A498 and human BLCA cell lines T24. In a 5% CO2 humidified cell incubator at 37 °C, A498 and T24 cells were grown in Dulbecco’s modified Eagle medium (DMEM) (Gibco BRL Life Technologies Inc., USA) with 10% FBS (Gibco BRL Life Technologies Inc., USA) and 1% penicillin-streptomycin (Hy-clone, CA, USA).
Small interfering RNAs (siRNAs) were ordered from GenePharma. The si-NC and si-ESCO2 siRNA were transfected into A498 and T24 cells with a Transfection reagent. When cells occupied 70–80% of a 50 mm Petri dish, transfection experiments were conducted; the whole medium was aspirated, and 2 mL of opti-MEM was applied to each well. The siRNA (GenePharma, Shanghai, China) and Transfection reagent (Cat#:L3000075, Thermo Fisher Scientific, United States) were added to 500 µL opti-MEM, mixed, and incubated for 20 min at room temperature. After that, the siRNA-Transfection reagent mixture was added to each well. Each well was switched to a complete medium 6 h after the transfection assay. The sequences for ESCO2 siRNA are as follows: si-ESCO2-1 (forward: GCAAAUCAAGGCUCACCAUTT; reverse: AUGGUGAGCCUUGAUUUGCTT) and si-ESCO2-2 (forward: CUCUUAGACCAGGAUUAUCTT; reverse: GAUAAUCCUGGUCUAAGAGUG).
qPCR and western blotting analysis
After the cells were transfected for 48 and 72 h, we harvested cells for RNA and protein extraction, respectively. Following the protocol, the total RNAs of cells were extracted using TRIzol reagents (Invitrogen). We then used cDNA reverse transcription kits to synthesize cDNA, and the mRNA expression level was calculated through qPCR assay in q225. We used the 2−ΔΔCt method to calculate GAPDH as an internal reference for mRNA. The ESCO2’s sequences of primers are as follows: F, CACTGGGACGCACCCAAAA, R, CACTTGCCTTGTCGCAAAAG (Sangon biotech). The following are the primer sequences for the GAPDH gene.: F, CAGGAGGCATTGCTGATGAT, R, GAAGGCTGGGGCTCATTT (Sangon biotech).
The tissues or cells were collected and lysed in RIPA lysis buffer (Cat#:89,901, Thermo Fisher Scientific, USA) with supplementation of a protease inhibitor cocktail (Cat#:539,131, Roche, Switzerland) in the ice for 30 min. The total protein extracts were measured by the ABC method (Cat#:23,227, Thermo Fisher Scientific, USA), separated by 10% sodium dodecyl sulfate-polyacrylamide gel electrophoresis (SDS-PAGE), and transferred to the nitrocellulose membrane. The membrane was blocked using 5% non-fat milk or 5% BSA (Solarbio, Beijing, China), then incubated with primary antibodies anti- (ESCO2,Cat#:23525-1-AP, diluted 1:2000, Proteintech, USA; CDK1, Cat#:23525-1-AP, diluted 1:2000, Proteintech, USA). The proper secondary antibody is chosen based on the primary antibody’s type, including goat anti-mouse HRP-conjugated IgG (Cat#:SA00001-1, 1:2000 dilutions, Proteintech, USA) and goat anti-rabbit HRP-conjugated IgG (Cat#:SA00001-2, 1:2000 dilutions, Proteintech, USA). The target band signals were captured with the help of a BIO-RAD ChemiDoc XRS chemiluminescence system (Bio-Rad Inc., CA, USA). The densitometry of all Western blots was analysed using a Gel-Pro analyzer (Media Cybernetics, USA). Imagej software was used to perform quantitative analysis with β-actin as a control (1.52a, National Institutes of Health, Bethesda, MD, USA).
Cell proliferation and colony formation assay
Briefly, 2000 cells per well were planted onto 96-well plates for the cell proliferation assay, and transfection assay was conducted after cell fixation. Cells were cultured in the incubator for 12, 24, 48, and 72 h. After that, each well received 10 µL of Cell Counting Kit-8 reaction solution (CCK8, Cat#:CA1210, Solarbio Science & Technology Co., Ltd., Beijing, China) and was incubated for 1 h. The light absorption values (OD) for si-NC and si-ESCO2 groups were captured at 450 nm on a Varioskan Flash Multimode plate reader (Thermo Fisher Scientific Inc., USA).
Cancer cells per well were seeded onto a 12-well plate, and the medium was changed every two days. After being stained with 0.5% crystal violet (Cat#: C0121, Beyotime) for 20 min and fixed with 4% paraformaldehyde at room temperature for 10 min, colonies were counted 14 days later.
Ethynyl-2-Deoxyuridine (EdU) incorporation assay
Cancer cells were seeded in 3.5 cm confocal dishes (Thermo Fisher Scientific Inc., USA) with 50,000 cells per dish. An appropriate amount of 50 µM EdU (Cat#:K1077, ApexBio Technology LLC) medium was prepared by diluting the EdU solution with cell culture medium at a ratio of 1000:1.1 ml of 50 µM EdU medium was added after the entire medium had been aspirated, and it was then incubated for 2 h. After discarding the medium, it was washed with PBS, fixed with 4% paraformaldehyde, and then permeabilized with 0.5% TritonX-100 (Cat#:X100, Sigma-Aldrich, USA). 200 μm of 1x Apollo (Cat#:K1077, ApexBio Technology LLC) staining reaction solution was added to each dish, incubated in the dark for 30 min. Nuclei were stained with DAPI (Cat#:MBD0015, Sigma-Aldrich, USA) for 10 min. Using a Nikon ECLIPSE 80i fluorescence microscope, images were captured, and ImageJ was used to count the cell number (1.52a, National Institutes of Health, Bethesda, MD, USA).
Transwell migration and wound healing assay
Matrigel (BD Biosciences, Franklin Lakes, NJ, USA) was evenly spread in transwell chambers (Cat#: 3422, Corning Inc., Corning, NY, USA), and cells starved for 24 h were seeded in a serum-free medium in the upper chamber. Medium containing 10% FBS was added to the lower chamber as a catalyst. The upper chamber was removed and separated after roughly 24 h, then fixed with methanol for 10 min, rinsed with PBS, and stained with 0.1% crystal violet (Cat#:G1064, Solarbio Science & Technology Co., Ltd., Beijing, China). Under a microscope, cells that had been pierced were viewed and counted (CKX41, Olympus, Tokyo, Japan).
Cells were seeded uniformly in micro-insert 4-wells in a µ-dish35 mm, high ibiTreat (ibidi GmbH, Germany). The culture inserts were removed when the cell density reached 90%. The 500 mm wide scratching gaps were washed with phosphate-buffered solution (PBS, Cat#:D1020, Solarbio Science & Technology Co., Ltd., Beijing, China) and were replaced with fresh medium. The dynamic changes of wells were recorded at 0, 12, 24 and 48 h on a microscope (Olympus, Japan).
Statistical analysis
Using the Wilcox test, comparisons between the two groups were made. For analyzing three or more groups, the one-way ANOVA test was performed. For all statistical calculations, R studio and GraphPad software were used. The threshold for significance was defined at P < 0.05.
Discussion
Cancer treatment has evolved significantly, with the introduction of immunotherapy and targeted medicines increasing patient survival rates for those with advanced or metastatic cancer, but overall clinical outcomes remain disappointing. The challenge of treating advanced cancers has inspired researchers to explore the underlying mechanisms that contribute to cancer growth, which will help identify potentially effective therapeutic targets. Unquestionably, one of the most promising therapeutic targets is cell cycle related protein. Palbociclib, the first CDK4/6-specific inhibitor, was created in 2004 and has proven effective against a variety of human cancer cell lines [
34]. All three CDK4/6 inhibitors (palbociclib, ribociclib, and abemaciclib) are currently approved by the United States Food and Drug Administration (FDA) for the treatment of breast cancer. In addition, numerous inhibitors of cell-cycle proteins, including CDK9, CDK2, and CDK5, are in clinical trials [
35,
36]. Cell cycle regulating proteins such as CDCA4, CDCA8, and KIF2C have been linked to tumor growth and progression, affecting the proliferation, migration, invasion and metastasis of cancer cell lines [
37‐
39]. ESCO2 was first reported in Roberts syndrome, whose inactivation mutation causes Roberts Syndrome [
11]. Subsequent researchers have identified that ESCO2 is crucial in controlling cell mitosis and preserving genomic stability because it bridges two sister chromatids and is involved in the mitotic S-phase adhesins acetylation [
40,
41]. As described in the context, current researchers consider ESCO2 an excellent therapeutic target as an oncogene that promotes cancer development. However, current cancer research on ESCO2 is limited to a few cancers, such as lung, stomach, and colon cancers. No studies are focusing on ESCO2 in multiple cancers that can shed light on the similarities of ESCO2 in cancer. In this study, we identified that ESCO2 was a reliable biomarker for cancer patients and could accurately predict the cancer patient’s prognosis. Furthermore, ESCO2 participated in mitosis, the cell cycle, DNA damage repair, other processes, and tumor immune infiltration. Finally, we confirmed that ESCO2 is essential for the proliferation and invasion of human ccRCC and BLCA cells in vitro.
First, we evaluated the ESCO2’s mRNA expression value in 33 cancers by TCGA and GTEx databases. The analysis showed that ESCO2 expression was upregulated in most cancer tissues except LAML, ACC and PCPG. Due to post-transcriptional processing, protein and mRNA gene expression levels may differ. Thus, we validated ESCO2 protein levels in KIRC and BLCA clinical samples. ESCO2 protein values were considerably greater in malignant tissues than in para-cancerous ones, consistent with the mRNA analysis. These results confirmed that ESCO2 was upregulated in various cancers, suggesting a promising future for ESCO2 in cancer diagnosis. Additionally, OS, DSS, DFI, and PFI analyses all revealed that ESCO2 was significantly associated with cancer patient’s prognosis. Evaluated ESCO2 resulted in poorer prognosis in ACC, BLCA, BRCA, CESC, COAD, GBM, HNSC, KIRC, LGG, LIHC, LUAD, LUSC, and PAAD. However, in THYM, COAD and READ, ESCO2 was a favorable prognostic factor. ESCO2 has been shown to prevent cancer metastasis in a recent trial on colon cancer, which is consistent with the results of our bioinformatics research [
19]. According to the findings above, ESCO2 is crucial in determining a cancer patient’s prognosis and can serve as a reliable prognostic biomarker. To some extent, tumor grading and clinical staging can show how tumors progress. Our results show that ESCO2 expression was positively associated with tumor stage and size in LIHC, KIRC, KIRP and LUAD. Furthermore, ESCO2 was crucial for the metastasis of COAD, KIRC, and LUAD. Our research identified that ESCO2 is a proto-oncogene that is linked to the development and progression of tumors.
Emerging research has confirmed that ESCO2 participated in apoptosis and cell proliferation-related pathways, such as the P53 and mTOR pathway [
15,
17]. Furthermore, ESCO2 could stimulate aerobic glycolysis in LUAD cells by upregulating PKM2 and downregulating PKM1 expression [
16]. The GSEA results in our study were unexpectedly uniform among the 12 different cancers, indicating the validity of our findings. The cell cycle and proliferation-related E2F targets, G2M checkpoint, mitotic spindle, and Mtorc1 signaling were concentrated in the genes upregulated in the ESCO2-high group. These ESCO2-high patients also displayed greatly enhanced DNA repair. In sum up, our bioinformatics analysis confirmed ESCO2’s contribution to the control of cell proliferation and cycle. Therefore, we conducted experimental validation to verify the results of the bioinformatics analysis. We knocked down ESCO2 in A498 and T24 cells, and ESCO2’S knockdown significantly inhibited the proliferation, invasion, and migration of A498 and T24 cells. This result demonstrated that ESCO2’s role in cancer might be unified, further indicating the possibility that it could be an effective therapeutic target. Furthermore, we investigated its potential regulatory mechanisms. We found the core of the ESCO2-CDK1 regulatory network. When ESCO2 was inhibited, CDK1 expression decreased, demonstrating that CDK1 is a downstream gene controlled by ESCO2. Cell cycle progression is the central event in all proliferating cells and is primarily regulated by cell cycle-dependent kinases (CDK) [
42].
CDK1 is the only CDK required for the G2-M transition and regulates G1 progression and G1-S transition [
43]. In recent years, CDK1 has been suggested as a therapeutic target for cancer. CDK1 overexpression has been found in many cancers, including gastric cancer, ovarian cancer, oral squamous cell carcinoma, liver cancer, and breast cancer [
44]. CDK1 inhibitors could be a potential small-molecule drug. CDK1 inhibitor RO3306 could improve the efficacy of sorafenib treatment by targeting cancer stem cells in a preclinical model of hepatocellular carcinoma [
45]. In addition, Several CDK1 inhibitors, including Rigosertib (phase II/III) and Zotaraciclib, have begun phase I clinical trials for treating pancreatic cancer and glioma [
46,
47]. As a CDK1’s upstream regulatory gene, ESCO2 took part in the control of the cell cycle. Hence, ESCO2 is likely to be a promising therapeutic target in the future. Finally, single-cell analysis observed that ESCO2 was expressed not only in malignant cells but also in endothelial and immune cells, especially proliferating T cells. The GSEA results also indicate that ESCO2 is involved in cancer inflammation control, including IL2 STAT5 signaling, which raises the possibility that ESCO2 also controls the tumor immune microenvironment. Studies have shown that inhibition of cell cycle protein not only induce tumor cell cycle arrest, but also promote anti-tumor immunity. Inhibition of CDK4 /6, for example, increases levels of PD-L1 protein. More excitingly, combining CDK4/6 inhibitor with anti-PD-1 immunotherapy promotes tumor regression and significantly increases overall survival in mouse tumor models [
48]. Therefore, research into ESCO2’s function in the tumor immune microenvironment is essential.
This study has the following limitations, though. First, this work does not apply any unique clinical cohorts to evaluate the diagnostic and prognostic significance of ESCO2 in cancer, despite its extensive use of sequencing data from public databases. Second, this study only performed ESCO2’s functional experiment in ccRCC cells; the upstream and downstream pathways of ESCO2 were not comprehensively examined, and the precise molecular mechanism of ESCO2 regulation is still unknown. Third, we did not conduct experiments in vivo, which is an important issue. We should attempt to resolve this issue in the future.
In conclusion, ESCO2 is a potential biomarker and oncogene for pan-cancer that can accurately predict the cancer patient’s prognosis in ACC, KICH, KIRC, KIRP, LGG, LIHC, LUAD, MESO, PAAD, THYM, COAD, and READ. Furthermore, our bioinformatics results discovered that ESCO2 is involved in cell division and cell cycle regulation and verified that ESCO2 is essential for the proliferation and invasion of human ccRCC and BLCA cells in vitro. In a nutshell, ESCO2 is a potential and reliable therapeutic target.
Publisher’s Note
Springer Nature remains neutral with regard to jurisdictional claims in published maps and institutional affiliations.