1. Introduction
Glycosylation is the main post-translational modification (PTM) in which glycans are conjugated to a protein at asparagine (N-linked) or serine/threonine (O-linked) residue, contributing to protein functions such as cell-to-cell interaction, recognition, adhesion, and migration. Aberrant glycosylation is a dominant feature of all human cancers that influences tumor proliferation, invasion, and metastasis. Of the glycosylation aberrations, hypersialylation is the most widely occurring cancer-associated glycosylation [
1].
Sialic acid is a family of nine-carbon backbone glycans attached to the terminal position of glycoprotein/glycolipid on the cell surface. The process of sialic acid incorporation into glycoconjugates on the cell surface is known as sialylation, and it is catalyzed by sialyltransferases (STs). 20 STs have been categorized into four groups: ST3Gal1-5, ST6Gal1-2, ST6GalNAc1-6, and ST8SIA1-6. The four groups of STs are categorized according to (1) the linkage of carbon connecting the second carbon (C2) to the C3, C6, C8, and C9 positions, catalyzing the synthesis of α2,3-, α2,6-, α2,8-, and α2,9-linked sialic acids, and (2) the glycan substrate including Gal: sialic acid attached to the terminal position of galactose, GalNAc: sialic acid attached to the terminal position of N-acetyl galactose, and SIA: sialic acid attached to the terminal position of sialic acid [
1,
2,
3].
Sialylation is crucial for cell-to-cell interaction, recognition, and adhesion. Desialylation is the primary physiological function of sialic acid. For instance, desialylation of serum glycoproteins and aging red blood cells are absorbed by hepatocytes and are recognized and degraded by macrophages [
2,
4,
5]. On the contrary, hypersialylation on the cell surface gives malignant cells better survival, and this aberrant sialylation is accompanied by overexpression of STs. Aberrant sialylation has been well recognized in various human cancers with abnormal ST activity. For instance, ST6Gal1 has been reported to be involved in different aspects of tumorigenesis [
6]. Oncogenic Ras can increase ST6Gal1 and alter sialylation of integrin β1 and the adhesion of cancer cells [
7]. Sialylation of endothelial growth factor receptor (EGFR) was mediated by ST6Gal1 via the PI3K/Akt pathway to affect tumor proliferation [
8,
9]. In addition, sialylation of the Fas receptor (FasR) by ST6Gal1 has been shown to inhibit apoptotic signaling in colon cancer [
10]. In summary, hypersialylation by α2,3-, α2,6-, and α2,8-STs could significantly affect tumor proliferation, adhesion, metastasis, apoptosis, immune evasion, and angiogenesis [
11,
12,
13,
14].
In breast cancer, STs were reported to be associated with stage/progression in 1980 [
15]. Various α2,3-, α2,6-, and α2,8-STs were involved in breast cancer. For example, α2,3-STs have been shown to catalyze the sialylation of chemokine (C-C motif) receptor 7 (CCR7) to affect proliferation, invasion, and anoikis in breast cancer cells [
16]. The adhesion and invasion abilities of breast cancer cells have been shown to be affected by α2,6-STs [
17]. In addition, α2,8-STs were involved in tumor growth and metastasis of triple-negative breast cancer (TNBC) [
18]. However, there is still some controversy about the role of STs in breast cancer. There was little agreement on the percentage of STs in breast cancer measured using immunohistochemistry [
19]. Some α2,6-STs have been reported to function as metastasis suppressors in breast cancer [
20], and intriguingly, sialic acid production in an advanced TNBC cell line was lower than that in an early stage cell line [
21]. Comprehensive analysis of STs has only been conducted in cell line models [
21,
22,
23], and there has been little agreement about STs measured using immunohistochemistry [
19]. Thus, in this study, we aimed to investigate the expression of STs in breast cancer using RNA sequencing (RNA-Seq) data from The Cancer Genome Atlas Breast Invasive Carcinoma (TCGA-BRCA) of the Genomic Data Commons (GDC) data portal. We comprehensively analyzed expression of STs and its relation to the 10-year overall survival (OS) and disease-free survival (DFS) rates, which confirmed the importance of candidate STs and provided another therapeutic approach for breast cancer.
4. Discussion
Hypersialylation is a feature of various cancers in which aberrant expression of STs increases sialylation expression on tumor cell surfaces. Hypersialylation further facilitates several aspects of tumorigenesis including (1) immune evasion through immune inhibitory Siglecs (Immune inhibitory receptors), (2) enhancement of tumor proliferation and metastasis through cytoskeleton-related protein, (3) promotion of tumor angiogenesis through the interaction between VEGF and polysialic acid, and (4) resistance to apoptosis through anti-apoptosis/kinase inhibitors [
14]. The above effects of hypersialylation on cancer affect the survival of patients. Thus, it is crucial to elucidate survival-related STs in breast cancer and provide another therapeutic approach.
In our studies, we used RNA-Seq data from the clinical database of TCGA-BRCA of the GDC data portal to comprehensively analyze the role of STs in breast cancer. Prior to this study, a comprehensive analysis of STs had only been conducted in cell lines. In the cell line model, the author analyzed the transcript level of sialic acid metabolism and glycosylation (SAMG) genes including 20 STs in MCF10A, T-47D, and MDA-MB-231 cells [
21,
22]. In compartment 2 of 20 STs, MDA-MB-231 cells displayed higher levels of
ST3GAL5 and
ST8SIA1 than MCF10A/T-47D cells and no change was observed in
ST6GALNAC4. Intriguingly, sialylation was lower in MDA-MB-231 than in T-47D cells due to metabolic flux-based control of sialylation [
21,
22]. In a proteomic analysis of different breast cancer cell lines, lectin chromatography/mass spectrometry was used to compare glycosylation profiles between luminal and TNBC cell lines. In contrast to the luminal type, a number of TNBC-specific glycosites were enriched with sialic acids and a concomitant increase in STs gene expression was observed [
23]. The results of the proteomic analysis indicated that TNBC-specific sialylation might be a therapeutic target for this aggressive tumor subtype.
In our RNA-Seq analysis of the clinical data set, we identified three STs genes that were significantly different between TNBC and non-TNBC. ST3GAL5 was higher in the TNBC group as in MDA-MB-231 cells, whereas ST6GALNAC4 and ST8SIA1 were higher in the non-TNBC group, which was somewhat different from the cell line model, indicating the difference of transcript level between cell line and clinical patients. In order to clarify the potential of STs as a therapeutic target, their effect on OS was investigated, and we found that ST8SIA1 significantly impacted OS in all patients, the TNBC group, and the non-TNBC group. Moreover, ST8SIA1 specifically affected DFS in TNBC patients, indicating that ST8SIA1 might be a therapeutic target for this aggressive TNBC subtype.
ST8SIA1, also known as GD3 synthase gene (GD3S), is a key enzyme in the biosynthesis of b-/c-series gangliosides (GD3, GD2, and GT3) expressed on the cell surface. Gangliosides are glycosphingolipids carrying one or more sialic acid residues and are located on the outer leaflet of the plasma membrane to interact with receptor tyrosine kinases (RTKs) [
25,
26]. High expression of
ST8SIA1 or GD3S has been reported to be associated with poor histologic grade/survival in estrogen-receptor (ER)-negative breast cancer and with better prognosis in ER-positive breast cancer [
27,
28]. However, in our study, the impact of
ST8SIA1 on survival was not limited to ER-negative breast cancer. High expression of
ST8SIA1 was significantly correlated with poor OS in all patients, the TNBC group, and the non-TNBC group. In Cox proportional hazard regression, high expression of
ST8SIA1 was also significantly associated with higher risk of poor OS in all patients (HR = 8.57), the TNBC group (HR = 11.9), and the non-TNBC group (HR = 8.01). Thus,
ST8SIA1 is a crucial survival-related ST in breast cancer, regardless of the ER receptor status. Moreover, high expression of
ST8SIA1 was also significantly associated with poor DFS and a higher risk for poor DFS (HR = 4.30) in TNBC patients particularly. This TNBC-specific effect was similar to that found in other studies indicating an association between
ST8SIA1 and ER-negative breast cancer. The authors of several studies reported that
ST8SIA1 could promote tumor growth, metastasis, and chemoresistance through the FAK/Akt/mTOR and Wnt/β-catenin signaling pathways in TNBC cell lines [
18,
29,
30]. This specific effect of
ST8SIA1 in TNBC cell lines may indicate the potential mechanisms underlying the association of higher
ST8SIA1 levels with poor DFS in TNBC patients. Thus, from the literature and our comprehensive study, survival-related STs of
ST8SIA1 may be a novel and crucial therapeutic target in breast cancer, especially for TNBC patients.
Several groups of sialyltransferase inhibitors (STIs) have been developed, including (1) analogs of sialic acids/CMP-sialic acid/cytidine/lithocholic acid, (2) oligosaccharide derivatives, (3) aromatic compounds, and (4) flavonoids [
31,
32]. Most STIs have poor cell membrane permeability and bioavailability that fail to meet the criteria for clinical utility. Few STIs have great permeability across the cell membrane. For instance, lithocholic acid analogs of AL10 exert great permeability and lower cytotoxicity, and they inhibit adhesion, migration, proliferation, and invasion of lung and breast cancer cells [
11,
16]. However, most STIs were developed against α2,3/2,6-STs, and fewer were developed to inhibit polysialyltransferases of α2,8-STs, especially
ST8SIA1. Triptolide was isolated from the perennial vine
Tripterygium wilfordii and displayed several anticancer cell properties [
33,
34]. Triptolide was also found to indirectly inhibit
ST8SIA1 in melanoma cells [
35]. A specific STI against
ST8SIA1 is an unmet clinical need for breast cancer and TNBC treatment to address in a future study.
The present study still has some limitations. In our comprehensive transcriptomic analysis of STs from the TCGA-BRCA of the GDC data portal, 20 STs were included initially, but only 8 STs were mapped by the RNA-Seq expression. However, our studies have identified ST8SIA1 as a crucial survival-related ST that is associated with higher risk for poor OS in all patients and DFS in TNBC patients particularly. Further in vitro and clinical studies are required to verify the expression and function of ST8SIA1 beside the GO analysis and other members of STs to provide another therapeutic approach for breast cancer in the future.