Abstract
Despite the use of targeted drugs for human epidermal growth factor receptor type 2 (HER2)‐positive salivary duct carcinoma (SDC) treatment, the overall prognosis for SDC remains poor. In this study, we aimed to investigate whether the glycocalyx of SDC cells serves as a potential cell surface marker. To understand the complex structure of the glycocalyx of salivary gland tumors, we used a lectin with glycan‐specific binding properties. Lectin staining was performed for five types of common salivary gland tumors. We determined the expression of the transmembrane glycoprotein mucin 1 with Tn antigen (GalNAc), Tn‐MUC1, using 20 clinical SDC specimens. Vicia villosa lectin (VVL) was not stained in the vascular endothelium but specifically stained in the SDC tumor cells. GalNAc, to which VVL binds, formed Tn‐MUC1 via glycosylation with N‐acetylgalactosaminyltransferases (GALNTs). GALNT7 was highly expressed in SDC. Analysis of clinical SDC specimens revealed that Tn‐MUC1 was also positive in the SDC tumor cells, suggesting its potential as a cell surface target for SDC.
Keywords: glycocalyx, glycan, lectin, Tn‐MUC1, salivary gland cancer, salivary duct carcinoma
Introduction
Glycocalyx is a complex structure composed of mucopolysaccharides, such as glycoproteins, proteoglycans, and glycosaminoglycans. Both endothelial and tumor cells possess glycocalyx on their surface, and the glycocalyx structure was first observed via electron microscopy in the 1960s [1, 2]. Glycocalyx plays several roles in the vascular endothelium, including blood circulation, adhesion of inflammatory cells, and maintenance of homeostasis in vascular endothelial cells [3]. Cancer cells also have glycan in the form of glycocalyx and produce abundant glycocalyx, changing their phenotype due to the physical stress of their surroundings in the tumor microenvironment [4, 5]. Therefore, glycocalyx acts as a barrier and sensor on the surface of cancer cells that interacts with host bioactive substances, such as cytokines, and immune cells, such as lymphocytes [6].
Glycans are difficult to analyze owing to their complex structure and various components. Lectins that specifically bind to glycans are widely used to analyze their complex structure [7, 8]. Lectin staining is used to evaluate both the thickness of the glycocalyx on the cell surface and the specific composition and quantity of glycans in the glycocalyx [9]. Tissue fixation is crucial as glycans are unstable under chemical stimulation and hydration, and formalin‐fixed paraffin‐embedded (FFPE) tissues have lower lectin responsiveness than frozen specimens [10, 11].
In O‐glycosylation, a post‐translational modification, N‐acetylgalactosamine (GalNAc) is generally added to the protein transported to the Golgi apparatus via catalysis with N‐acetylgalactosaminyltransferases (GALNTs) [12]. GalNAc is added to serine/threonine residues via O‐glycosylation to form tumor‐associated mucin 1 (TA‐MUC1), which plays key roles in tumor invasion, metastasis, angiogenesis, proliferation, and apoptosis [13, 14, 15, 16]. GalNAc, to which Vicia villosa lectin (VVL) binds, forms Tn‐MUC1 via glycosylation. MUC1 in normal cells is modified with highly complex O‐glycans; however, in cancer cells, glycan elongation stops, resulting in the formation of abnormally short glycans and an increase in tumor‐associated glycans, such as Tn, T, STn, and ST antigens [17, 18]. Various diagnostic markers targeting these antigens, such as CA15‐3 for breast cancer, are currently used in clinical settings [19].
Salivary gland tumors are rare neoplasms of the head and neck, accounting for approximately 5% of all head and neck tumors. The overall incidence of salivary gland tumors is 7.03–8.58/100,000 [20], and the incidence of malignant tumors is even lower, 16/1,000,000 [21]. Salivary gland tumors are of various histological subtypes and are classified into 20 types of malignant tumor and 11 types of benign tumor according to the 5th edition of the World Health Organization (WHO) classification [22]. Based on histology, the incidence of malignant tumors is 20% for mucoepidermoid carcinoma (MEC), 12% for adenoid cystic carcinoma (ACC), and 10% for salivary duct carcinoma (SDC) [23]. In this study, we focused on two common benign and three malignant tumors with a particular focus on SDC. SDC is a high‐grade tumor with poor prognosis. For the treatment of salivary gland carcinoma, surgery is the first choice for resectable cases, and postoperative radiation therapy may also be required depending on the malignancy and histology. As salivary gland carcinomas have various histological subtypes and characteristics, a standard chemotherapy regimen has not yet been established. Recent studies have reported the effectiveness of molecular targeted drugs for human epidermal growth factor receptor type 2 (HER2)‐ and androgen receptor (AR)‐positive tumors [24, 25]. However, HER2 positivity in SDC is limited (30–40%) [24, 26, 27, 28] and the prognosis, especially for HER2‐negative SDC, is still poor. Salivary glands secrete mucins composed of glycoproteins; therefore, tumors that arise from the salivary glands are suitable for glycan analysis. However, there are no reports that comprehensively analyze glycans in salivary gland tumors or that verify the usefulness of tumor‐associated glycans. Here, we aimed to analyze the glycan structure in different histological types of salivary gland tumor, including two types of representative benign [pleomorphic adenoma (PA) and Warthin tumor (WT)] and three types of representative malignant (MEC, ACC, and SDC) tumors via lectin staining.
Materials and methods
Patients and sampling of fresh frozen specimens and FFPE material
This study was conducted in accordance with the Declaration of Helsinki and approved by the Ethics Committee of Gifu University (Gifu City, Japan). Informed patient consent and prior approval from the Gifu University Hospital affiliated with the Gifu University Graduate School of Medicine Ethics Committee (approval no. 2020‐234) were obtained before using the clinical materials in experiments. Informed patient consent was obtained before using the new clinical materials in experiments. If past clinical samples were used, informed consent was not sought because a waiver of consent was approved by the institutional review board due to the opt‐out approach.
The study population using fresh frozen specimens consisted of 12 patients with different salivary gland tumors (PA, n = 4; WT, n = 2; SDC, n = 2; MEC, n = 2; and ACC, n = 2) who underwent surgery at the Gifu University Hospital. The most common histological types of both benign and malignant tumors were selected. As glycan is water‐soluble and its structure is easily broken by formalin fixation, freshly frozen specimens were used in all experiments. Due to the lack of archived fresh‐frozen specimens from previous cases, new cases were prospectively collected, and this experiment was conducted as an initial screening. Immediately after excision, a sample of approximately 5 mm was collected, protected using Tissue OCT compound (Sakura Finetek, Japan), snap‐frozen in liquid nitrogen, and stored at −80 °C.
The study population using FFPE comprised 20 patients with SDC who underwent treatment at the Gifu University Hospital between May 2007 and September 2021. Regardless of clinical stage, only initial treatment cases were selected, and recurrent cases were excluded. Specimens from the initial surgery or initial biopsy were used.
All specimens were confirmed by two pathologists based on the 5th edition of the WHO Health Organization classification. All relevant clinical data were acquired from the patient medical records.
Lectin histochemistry
In total, 20 types of lectins were used in this study. Screening kit I (Vector Laboratories, Burlingame, CA, USA) consisted of seven types of lectins: concanavalin A (Con A), Dolichos biflorus agglutinin (DBA), peanut agglutinin (PNA), Ricinus communis agglutinin (RCA) I, soybean agglutinin I (SBA), Ulex europaeus agglutinin (UEA) I, and wheat germ agglutinin (WGA). Screening kit II (Vector Laboratories) consisted of six types of lectins: Griffonia simplicifolia lectin (GSL) I, Pisum sativum agglutinin (PSA), Len culinaris agglutinin (LCA), Phaseolus vulgaris erythroagglutinin (PHA‐E), Phaseolus vulgaris leucoagglutinin (PHA‐L), and succinylated WGA. Screening kit III (Vector Laboratories) consisted of seven types of lectins: Datura stramonium lectin (DSL), Erythrina cristagalli lectin (ECL), GSL II, Jacalin, Lycopersicon esulentum lectin (LEL), Solanum tuberosum lectin (STL), and Vicia villosa lectin (VVL; supplementary material, Table S1). In this study, a comprehensive analysis of diverse glycan specificities was performed by utilizing all 20 lectins from Vector Laboratories' lectin screening kits I, II, and III. Previous research has validated the combined use of these kits as an effective approach for comprehensive glycan profiling in human and mammalian tissues [29, 30, 31].
First, 5‐μm sliced sections were fixed with 4% paraformaldehyde (FujiFilm, Tokyo, Japan) for 15 min. All sections were incubated in a 10% carbon‐free blocking solution (Vector Laboratories) for 1 h at approximately 20–25 °C to block non‐specific labeling. After washing with phosphate‐buffered saline (PBS, Wako, Osaka, Japan), sections were incubated with various lectins (0.5%) and 0.5% anti‐CD31 antibody (ab76533; dilution 1:200; Abcam, Cambridge, UK) to stain the vascular endothelial cells overnight at 4 °C. The following day, all sections were incubated with 0.5% Streptavidin DyLight®594 (SA‐5594‐1; dilution 1:200; Vector Laboratories) and 0.5% pre‐adsorbed Goat Anti‐Rabbit IgG H&L (DyLight®488; ab96899; dilution 1:200; Abcam) for 60 min at room temperature. After washing with PBS, all sections were incubated with 0.1% 4,6‐diamidino‐2‐phenylindole (DAPI, Dojindo, Kumamoto, Japan) solution for 5 min at room temperature.
Fluorescence microscopy and image evaluation
Slides were evaluated using fluorescence microscopy and imaging software (CellSens, Olympus, Tokyo, Japan). Lectins, vascular endothelial cells, and cell nuclei were evaluated using GNA (red), fluorescein isothiocyanate (green), and DAPI (blue) staining, respectively. Images were taken at the same exposure time and labeled as ‘positive (+)’, ‘weakly positive (±)’, and ‘negative (−)’ according to the fluorescence intensity.
RNA extraction and RT‐PCR
After cutting and macrodissecting 10‐μm sections, total RNA of SDC cells was extracted using the RNeasy Kit (QIAGEN, Venlo, the Netherlands). cDNA was synthesized using the SuperScript III First‐Strand Synthesis Kit (Life Technologies, Carlsbad, CA, USA). Reverse transcription‐polymerase chain reaction (RT‐qPCR) was performed using the StepOnePlus system (Applied Biosystems, Foster City, CA, USA). GALNT primers were designed using Primer‐BLAST (supplementary material, Table S2). The comparative Ct method was used to analyze the relative gene expression. RT‐qPCR was performed in triplicate for each primer. Beta‐actin was used as an internal control.
Public RNA‐seq data
Here, mRNA expression datasets (GSE138581) for SDC were obtained from the National Center for Biotechnology Information database (https://www.ncbi.nlm.nih.gov/geo/). RNA‐seq data for GALNT was extracted and validated in SDC.
Immunohistochemistry
Paraffin blocks from 20 SDC tumors were cut into 3‐μm thick sections and subjected to hematoxylin–eosin staining. Adjacent serial sections were subjected to immunohistochemistry for Tn‐MUC1 [anti‐human MUC1 therapeutic antibody (5E5); dilution 1:1000; Creative Biolabs, Shirley, NY, USA], VVL (Vector Laboratories, Newark, CA, USA), HER2 (A0485; dilution 1:800; Dako, Santa Clara, CA, USA), Ki‐67 (M7240; dilution 1:150; Dako), and GALNT7 (Catalog Number: 13962‐1‐AP, dilution 1:200; Proteintech, Rosemont, IL, USA). Negative controls were used to confirm that the reaction was not non‐specific.
Deparaffinized sections were subjected to autoclave boiling in 0.015 M sodium citrate buffer solution (pH 6.0, Sigma, St Louis, MO, USA) for 10 min at 110 °C for antigen retrieval before incubation with 3% H2O2 diluted in methanol for 10 min and blocked with 2% normal bovine serum (Sigma). For GALNT7 immunostaining, EDTA buffer, pH 8, was specifically used for antigen retrieval. Sections were incubated with primary antibodies overnight at 4 °C, followed by incubation with conjugated secondary antibodies for 60 min at 37 °C. The immunoreaction was visualized using 3,3'‐diaminobenzidine tetrahydrochloride (Sigma).
Tn‐MUC1 and VVL were quantified using ImageJ software (version: 2.1.0/1.53c, National Institutes of Health, Bethesda, MA, USA; 16‐bit, threshold; 150). In brief, using three micrographs of each case with a ×400 field of view, the threshold value of the staining intensity was set, and the average value of the positive area was evaluated. HER2 expression was determined to be negative (0, 1+) or positive (2+, 3+) according to the H score [32]. The number of Ki‐67‐positive cells was measured using a ×400 field of view. All staining evaluations were performed by one pathologist (HT) and one otolaryngologist (MK).
Statistical analyses
Differences in categorical data between the two groups were analyzed using the chi‐square test. Correlations of staining area of VVL and TnMUC‐1 were examined using Pearson correlation coefficients. Survival time analysis was performed using the Kaplan–Meier method, and differences between groups were analyzed using the log‐rank test. All statistical analyses were conducted using the GraphPad Prism 8 software (GraphPad Software, Inc., San Diego, CA, USA). Statistical significance was set at p < 0.05.
Results
Representative glycans GalNAC and fucose are only expressed in malignant tumors
To investigate glycan expression in salivary gland tumors, lectin staining was performed on surgically resected human salivary gland tumors. We then analyzed PA and WT as representative benign salivary gland tumors. SDC, MEC, and ACC were analyzed as representative malignant salivary tumors (Figure 1A). All staining data are presented in Table 1. Typical staining patterns for lectins positive in both benign and malignant tumors (WGA), negative for both (GSL I), and positive only for malignant tumors (UEA I, SBA) are shown in Figure 1B. Lectins that detect mannose, galactose, and GlcNAc did not differ between the benign and malignant tumors. Notably, GalNAc and fucose were detected only in malignant tumors, suggesting their potential as malignant tumor cell surface markers (Figure 1C).
Figure 1.
Differences in the staining patterns of benign and malignant tumors. (A) Representative hematoxylin–eosin (HE) staining of fresh frozen specimens in this study. (B) Some lectins showed positive staining in both benign and malignant tumors, negative staining in both tumors, and positive staining only in malignant tumors. Nuclei were stained with DAPI. (C) GalNAc and fucose were positive only in malignant tumors. *p < 0.05; n.s., not significant.
Table 1.
Staining patterns in all samples
| Kit | No | Lectin | PA1 | PA2 | PA3 | PA4 | WT1 | WT2 | SDC1 | SDC2 | MEC1 | MEC2 | ACC1 | ACC2 |
|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|
| I | 1 | Con A | + | + | + | + | + | + | + | + | + | + | + | + |
| 2 | DBA | − | − | − | − | − | − | − | − | ± | − | − | − | |
| 3 | PNA | ± | + | + | + | − | − | ± | + | − | − | ± | + | |
| 4 | RCA I | + | + | + | + | + | + | + | + | + | + | + | + | |
| 5 | SBA | − | − | − | − | − | − | ± | − | ± | − | − | − | |
| 6 | UEA I | − | − | − | − | − | − | − | − | − | + | − | − | |
| 7 | WGA | + | + | + | + | + | + | + | + | + | + | + | + | |
| II | 1 | GSL‐I | − | − | − | − | − | − | + | + | − | − | − | − |
| 2 | PSA | + | ± | ± | ± | + | + | + | + | + | + | + | + | |
| 3 | LCA | ± | − | ± | ± | ± | + | + | + | ± | + | + | ± | |
| 4 | PHA‐E | + | + | + | + | + | + | + | + | + | + | + | + | |
| 5 | PHA‐L | + | + | + | + | + | + | ± | ± | + | + | + | + | |
| 6 | WGA | − | − | − | − | − | − | − | − | − | − | − | − | |
| III | 1 | DSL | + | + | + | ± | + | + | + | ± | + | + | ± | + |
| 2 | ECL | + | + | ± | + | + | + | + | ± | + | + | ± | + | |
| 3 | GSL‐II | − | − | − | ± | + | ± | − | − | ± | ± | − | − | |
| 4 | Jacalin | + | + | + | ± | ± | + | ± | ± | ± | ± | + | + | |
| 5 | LEL | + | + | + | + | ± | ± | + | + | ± | ± | + | + | |
| 6 | STL | ± | ± | ± | + | ± | ± | + | + | + | ± | + | + | |
| 7 | VVL | − | − | − | − | − | − | + | + | ± | − | − | − |
With screening kit I, concanavalin A (Con A), Ricinus communis agglutinin (RCA) I, and wheat germ agglutinin (WGA) were positive in all tumors. Dolichos biflorus agglutinin (DBA) was weakly positive only for mucoepidermoid carcinoma (MEC)‐1, soybean agglutinin I (SBA) was weakly positive only for salivary duct carcinoma (SDC)‐1 and MEC1, and Ulex europaeus agglutinin (UEA) I was weakly positive only for MEC‐2. With screening kit II, Pisum sativum agglutinin (PSA), Len culinaris agglutinin (LCA), and Phaseolus vulgaris erythroagglutinin (PHA‐E) and leucoagglutinin (PHA‐L) were weakly positive‐to‐positive in most tumors, whereas WGA was negative in all tumors. Griffonia simplicifolia lectin (GSL) I was positive only in SDC‐1 and SDC‐2. With screening kit III, all tumors were weakly positive‐to‐positive for Datura stramonium lectin (DSL), Erythrina cristagalli lectin (ECL), jacalin, Lycopersicon esculentum lectin (LEL), and Solanum tuberosum lectin (STL). Vicia villosa lectin (VVL) was positive only for SDC‐1 and SDC‐2 and weakly positive for MEC‐1. UEA I was specifically positive for MEC, whereas GSL I and VVL were specifically positive for SDC.
PNA was positive in all PA cases but negative in all WT cases (supplementary material, Figure S1). GSL II was positive or weakly positive in all WT cases, whereas PA was negative in all but one case.
VVL was negative in the vascular endothelial cells of SDC, and positive only in the SDC tumor cells
To investigate GalNAC expression in SDC tumors, VVL staining was performed. VVL was strongly stained in SDC (Figure 2A). VVL was more positive in malignant tumors than in benign tumors, particularly SDC (Figure 2). Glycocalyx is abundantly expressed in vascular endothelial cells; in some cases, it is not stained in vascular endothelial cells, but only in tumor cells. Notably, VVL was specifically positive in SDC tumor cells but negative in vascular endothelial cells in SDC tumor tissue, suggesting its mutually exclusive expression (Figure 2C,D). Comparing tumor cells and vascular endothelium in other tumor types, neither the vascular endothelium nor the SDC tumor cells were strongly stained.
Figure 2.
Vicia villosa lectin (VVL) only stains the tumor cells in salivary duct carcinoma (SDC). (A, B) VVL only stained malignant tumors, specifically SDC. CD31‐positive area indicates the vascular endothelium. (C, D) VVL did not stain the vascular endothelium but specifically stained the SDC tumor cells. Nuclei were stained with DAPI.
GALNT7 is highly expressed in SDC
To analyze the GALNT subtype involved in O‐glycosylation in SDC, we performed RT‐PCR using three freshly frozen specimens. GALNT7 was highly expressed in SDC (Figure 3A). To confirm the results, we analyzed a public RNA‐seq dataset of 24 SDC cases. Notably, similar results were obtained. These data indicate that GALNT7 influences O‐glycosylation in SDC more than the other GALNT family genes (Figure 3B).
Figure 3.
N‐acetylgalactosaminyltransferase (GALNT)‐7 is highly expressed in SDC. (A) Reverse transcription‐polymerase chain reaction (RT‐PCR) revealed that GALNT7 was highly expressed in SDC. (B) Public RNA‐sequencing dataset showed similar results. (C) Representative images of GALNT7 immunohistochemical staining in SDC tissues, categorized by staining intensity (0, 1+, 2+, 3+). The insets show magnified views. Scale bar = 200 μm. (D) Quantification of GALNT7 staining intensity in 24 SDC cases. The table indicates the number of cases for each staining grade. Positive staining was defined as 1+, 2+, or 3+, while 0 indicated negative staining.
GALNT7 is highly expressed at the protein level in SDC
To validate the high expression of GALNT7 at the protein level in SDC, we performed immunohistochemistry on 24 FFPE SDC samples using a GALNT7‐specific antibody. The staining intensity was graded on a scale of 0, 1+, 2+, and 3+, with 0 indicating negative staining and 1+, 2+, and 3+ indicating positive staining (Figure 3C). Out of 24 SDC cases, 12 cases (50%) showed strong positive (3+) staining, 6 cases (25%) showed moderate positive (2+) staining, 5 cases (20.8%) showed weak positive (1+) staining, and only 1 case (4.2%) showed negative (0) staining for GALNT7 (Figure 3D). These results demonstrate that GALNT7 is frequently and highly expressed at the protein level in SDC, consistent with our mRNA expression data. This finding further supports the notion that GALNT7 plays a significant role in O‐glycosylation and potentially in Tn‐MUC1 formation in SDC.
VVL and Tn‐MUC1 staining is only positive in the tumor cells of SDC
Based on the fact that VVL lectin staining specifically recognizes Tn‐MUC1, Tn‐MUC1 and VVL lectin expression were evaluated using clinical specimens from 20 SDC cases. Patient characteristics are shown in Table 2. Notably, in all cases, Tn‐MUC1 and VVL were positive only in the same tumor regions. A typical example is presented in Figure 4A. VVL and Tn‐MUC1 were strongly positive, as in case 12, and weakly positive, as in case 6 (Figure 4B). The staining intensities of Tn‐MUC1 and VVL were moderately correlated (r = 0.435, p = 0.056) (Figure 4C). These data suggest that VVL expression is a surrogate of Tn‐MUC1 expression in SDC tumors.
Table 2.
SDC patient characteristics
| Case | Age | Sex | Primary | T | N | M | Initial treatment | Recurrence | Outcome |
|---|---|---|---|---|---|---|---|---|---|
| 1 | 61 | M | Parotid | 4a | 2b | 0 | Surgery | + | Death |
| 2 | 63 | M | Parotid | 4 | 2b | 0 | Surgery | + | Death |
| 3 | 77 | F | Parotid | 2 | 0 | 0 | Surgery | − | Alive |
| 4 | 62 | M | Parotid | 4a | 2b | 0 | Surgery | − | Death |
| 5 | 81 | M | Parotid | 4a | 1 | 0 | Surgery | + | Death |
| 6 | 45 | M | Submandibular | 3 | 0 | 0 | Surgery | + | Death |
| 7 | 68 | M | Oral cavity | 4a | 2b | 0 | Surgery | + | Alive |
| 8 | 75 | M | Parotid | 4a | 2b | 0 | Surgery | + | Death |
| 9 | 64 | M | Parotid | 2 | 0 | 0 | Surgery | + | Alive |
| 10 | 84 | M | Parotid | 2 | 2b | 0 | Surgery | − | Alive |
| 11 | 69 | M | Parotid | 3 | 2b | 0 | Surgery | − | Death |
| 12 | 71 | M | Parotid | 4a | 2b | 0 | Surgery | + | Alive |
| 13 | 82 | M | Submandibular | 2 | 2b | 0 | Surgery | + | Death |
| 14 | 62 | M | Parotid | 4a | 2b | 0 | Surgery | + | Death |
| 15 | 81 | M | Parotid | 2 | 0 | 0 | Surgery | − | Alive |
| 16 | 72 | M | Parotid | 2 | 2b | 0 | Surgery | − | Alive |
| 17 | 75 | M | Parotid | 4a | 2c | 1 | Chemotherapy | + | Alive |
| 18 | 40 | M | Parotid | 1 | 0 | 0 | Surgery | − | Alive |
| 19 | 59 | M | Parotid | 3 | 2b | 1 | Chemotherapy | + | Death |
| 20 | 68 | M | Parotid | 3 | 0 | 0 | Surgery | − | Alive |
A total of 20 cases are included. The mean age was 68 years, 95% were male, and 85% had parotid gland origin; 60% of cases had recurrence, and 50% died.
Figure 4.
VVL and mucin 1 with Tn antigen (Tn‐MUC1) are only positive in SDC tumor cells. (A) Typical staining patterns for Tn‐MUC1 and VVL. (B) Case 12 was strongly positive, whereas case 6 was weakly positive for both Tn‐MUC1 and VVL. (C) Tn‐MUC1 and VVL correlate, with a correlation coefficient of 0.435. (D) Cases 7 and 16 were human epidermal growth factor receptor type 2 (HER2)‐negative but Tn‐MUC1‐positive.
To investigate whether Tn‐MUC1 is a notable cell surface marker in the HER2‐negative SDC subgroup, HER2 immunohistochemical staining was performed and compared with Tn‐MUC1 expression. It is known that some SDC cases express HER2, which serves as a therapeutic target using an anti‐HER2 molecular targeted antibody (trastuzumab). In this study, 10 cases (50%) were HER2‐positive. In other words, Tn‐MUC1 may be positive even in HER2‐negative SDC (Figure 4D). Thus, Tn‐MUC1 may be a therapeutic target for HER2‐negative SDC.
Correlation between Tn‐MUC1 and prognosis in SDC
We examined the prognosis of 20 patients with SDC treated at the Gifu University Hospital. Patient characteristics are shown in Figure 5A. The 5‐year overall survival (OS) was 45.6% (Figure 5B). In general, Tn‐MUC1 expression is associated with poor prognosis and high malignancy. However, in this study, Tn‐MUC1 staining intensity was not associated with the poor prognosis of SDC (Figure 5C). Furthermore, the staining intensity of Tn‐MUC1 was not related to tumor size, proliferation ability (Ki‐67), or HER2 staining intensity (Figure 5D).
Figure 5.
Correlations between Tn‐MUC1 and various clinical indicators. (A) Patient clinical characteristics. (B) Five‐year overall survival (OS) was 45.6%. (C) Tn‐MUC1 staining intensity was not associated with prognosis. Log‐rank test was used for the analysis; n.s., not significant. (D) Staining intensity of Tn‐MUC1 was not related to tumor size, proliferation (Ki‐67), or HER2 staining intensity; n.s., not significant.
Discussion
Many reports on lectin staining for head and neck cancer were published in the 1980s and the 1990s, and most of these reports used FFPE specimens instead of freshly frozen specimens. However, formalin fixation destroys the glycan structure; therefore, it may not reflect the actual expression of glycans in vivo. To date, five reports have used any of the 20 types of lectins in salivary gland tumors [33, 34, 35, 36, 37], but none have been investigated as comprehensively as in this study. Yang et al reported that Con A, RCA, and WGA are positive in MEC [34], similar to the present study. In addition, Sobral et al reported that all histological grades of MEC are positive for Con A, whereas UEA I shows differences in staining intensity depending on the tumor grade [33].
The utility of lectins as potential cancer therapeutic tools and carrier proteins has been reported in various carcinomas [38, 39, 40, 41]. However, there is a problem with lectins as drug treatments; Con A is known to cause liver failure due to its toxicity. PHA‐L can also cause gastrointestinal symptoms, such as vomiting and diarrhea [42]. In this study, GSL I of SDC2, VVL of SDC1, and GSL II of MEC2 were stained only in tumor cells but not in vascular endothelial cells. However, further studies, including animal experiments, are needed to show that these lectins are safe carrier proteins.
This study also suggested that GALNT7 is involved in the formation of Tn‐MUC1 in SDC. It has been reported that GALNT7 affects prognosis in several cancers, including prostate cancer, breast cancer, and glioma [43, 44, 45]. GALNT7 is the only gene effector that controls the height of the glycocalyx, and upregulation of GALNT7 alters O‐glycosylation and promotes tumor growth [43]. This study also revealed that GALNT7 is the most highly expressed member of the GALNT family, and GALNT7 may be correlated with prognosis in SDC as in other cancers.
Tn‐MUC1 antibody used in this study was developed to recognize only the Tn antigen, a representative molecule of TA‐MUC1 [46, 47]. Previous studies have analyzed the expression of Tn‐MUC1 in various cancers, but none have explored its roles in salivary gland tumors [48]. Tn‐MUC1 is highly expressed in most cancers, accounting for 100% of stomach, 71% of colon, 75% of lung, 83% of breast, and 80% of ovarian cancers, although the number of such cases is small. In head and neck cancer, the expression of TA‐MUC1 in laryngeal cancer has been analyzed with a positivity rate of 54% [49].
Owing to its expression in many cancer types, various treatments targeting Tn‐MUC1, including chimeric antigen receptor‐T cells (CAR‐T) [50, 51, 52, 53] and vaccines, have been developed [54, 55, 56]. Furthermore, various antibody–drug conjugates (ADCs) such as ENHERTU® (trastuzumab deruxtecan) targeting HER2 in breast and gastric cancers have been developed [57, 58, 59]. ADCs are complex anticancer drugs consisting of an antibody, a binding substance called a linker, and a payload [60]. Currently, six ADCs have been approved by the European Medicines Agency and US Food and Drug Administration for the treatment of solid tumors, and many new ADCs including TA‐MUC1 are in clinical trials. Thus, Tn‐MUC1 has the potential to be a new therapeutic application for other cancers.
This study revealed that Tn‐MUC1 was expressed in SDC tumor cells, suggesting its potential as a new diagnostic marker for SDC. A limitation of this study is that salivary gland tumors are rare cancers, so the number of cases that can be analyzed in a single‐center study is limited. Therefore, it is possible that no correlation was observed between the staining intensity of Tn‐MUC1 and prognosis. Secondly, frozen samples are more suitable for accurate glycan analysis, but FFPE samples were used in this study. In the future, it is necessary to increase the number of cases by conducting prospective studies at multiple centers.
Conclusion
To the best of our knowledge, this is the first report on the comprehensive glycan analysis of salivary gland tumors via lectin staining of fresh frozen sections. Lectins are useful for differentiating salivary gland tumors, and VVL was specifically expressed in tumor cells of SDC. Tn‐MUC1, which is formed by GalNAc specifically bound to VVL, was also highly expressed in SDC. In addition, it was suggested that GALNT7 is involved in the formation of Tn‐MUC1 in SDC. These results suggest that Tn‐MUC1 may be a novel potential marker for the diagnosis of SDC.
Author contributions statement
MK, RK and HT conceived the study and performed the experiments. MK, RK, HS and HT analyzed the data and wrote the first draft of the manuscript. T Ohashi, TI, HO, AH and T Ogawa checked the data and edited the manuscript. All authors wrote the manuscript and provided final approval for the submitted and published versions of the manuscript.
Supporting information
Figure S1. Salivary gland benign tumors can be differentiated using peanut agglutinin (PNA) and Griffonia simplicifolia lectin (GSL) II
Table S1. Lectin screening kits
Table S2. N‐acetylgalactosaminyltransferase (GALNT) primers
Acknowledgements
We thank Kyoko Takahashi, Ayako Suga, Reiko Kitazumia, and Izumi Toshima for their support in all aspects of the experiments. This study was supported by JSPS KAKENHI JP20K0758723 (HT), JP23H03326 (HT), JP21K16841 (HS), JP22K09703 (T. Ogawa), and JST FOREST Program JPMJFR220W (HT).
No conflicts of interest were declared.
Contributor Information
Hiroyuki Tomita, Email: tomita.hiroyuki.y6@f.gifu-u.ac.jp.
Hirofumi Shibata, Email: shibata.hirofumi.b3@f.gifu-u.ac.jp.
Data availability statement
RNA‐seq data were downloaded from the Gene Expression Omnibus database (https://www.ncbi.nlm.nih.gov/geo/; accession no. GSE138581). The data that support the findings of this study are available from the corresponding author (HS), upon reasonable request.
References
- 1. Rambourg A, Neutra M, Leblond CP. Presence of a ‘cell coat’ rich in carbohydrate at the surface of cells in the rat. Anat Rec 1966; 154: 41–71. [DOI] [PubMed] [Google Scholar]
- 2. Luft JH. Fine structures of capillary and endocapillary layer as revealed by ruthenium red. Fed Proc 1966; 25: 1773–1783. [PubMed] [Google Scholar]
- 3. Okada H, Takemura G, Suzuki K, et al. Three‐dimensional ultrastructure of capillary endothelial glycocalyx under normal and experimental endotoxemic conditions. Crit Care 2017; 21: 261. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 4. Paszek MJ, DuFort CC, Rossier O, et al. The cancer glycocalyx mechanically primes integrin‐mediated growth and survival. Nature 2014; 511: 319–325. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 5. Barnes JM, Kaushik S, Bainer RO, et al. A tension‐mediated glycocalyx‐integrin feedback loop promotes mesenchymal‐like glioblastoma. Nat Cell Biol 2018; 20: 1203–1214. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 6. Ghasempour S, Freeman SA. The glycocalyx and immune evasion in cancer. FEBS J 2023; 290: 55–65. [DOI] [PubMed] [Google Scholar]
- 7. Hashim OH, Jayapalan JJ, Lee CS. Lectins: an effective tool for screening of potential cancer biomarkers. PeerJ 2017; 5: e3784. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 8. Islam MK, Khan M, Gidwani K, et al. Lectins as potential tools for cancer biomarker discovery from extracellular vesicles. Biomark Res 2023; 11: 85. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 9. Roth J. Lectins for histochemical demonstration of glycans. Histochem Cell Biol 2011; 136: 117–130. [DOI] [PubMed] [Google Scholar]
- 10. Tachi M, Okada H, Matsuhashi N, et al. Human colorectal cancer infrastructure constructed by the glycocalyx. J Clin Med 2019; 8: 1270. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 11. Brandon DM, Nayak SK, Binder PS. Lectin binding patterns of the human cornea. Comparison of frozen and paraffin sections. Cornea 1988; 7: 257–266. [PubMed] [Google Scholar]
- 12. Bard F, Chia J. Cracking the glycome encoder: signaling, trafficking, and glycosylation. Trends Cell Biol 2016; 26: 379–388. [DOI] [PubMed] [Google Scholar]
- 13. Nath S, Mukherjee P. MUC1: a multifaceted oncoprotein with a key role in cancer progression. Trends Mol Med 2014; 20: 332–342. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 14. Kufe DW. Mucins in cancer: function, prognosis and therapy. Nat Rev Cancer 2009; 9: 874–885. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 15. Ahmad R, Raina D, Joshi MD, et al. MUC1‐C oncoprotein functions as a direct activator of the nuclear factor‐kappaB p65 transcription factor. Cancer Res 2009; 69: 7013–7021. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 16. Behrens ME, Grandgenett PM, Bailey JM, et al. The reactive tumor microenvironment: MUC1 signaling directly reprograms transcription of CTGF. Oncogene 2010; 29: 5667–5677. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 17. Pinto D, Parameswaran R. Role of truncated O‐GalNAc glycans in cancer progression and metastasis in endocrine cancers. Cancers (Basel) 2023; 15: 3266. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 18. Stowell SR, Ju T, Cummings RD. Protein glycosylation in cancer. Annu Rev Pathol 2015; 10: 473–510. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 19. Seregni E, Coli A, Mazzucca N. Circulating tumour markers in breast cancer. Eur J Nucl Med Mol Imaging 2004; 31: S15–S22. [DOI] [PubMed] [Google Scholar]
- 20. Bradley PJ, McGurk M. Incidence of salivary gland neoplasms in a defined UK population. Br J Oral Maxillofac Surg 2013; 51: 399–403. [DOI] [PubMed] [Google Scholar]
- 21. Del Signore AG, Megwalu UC. The rising incidence of major salivary gland cancer in the United States. Ear Nose Throat J 2017; 96: E13–e16. [DOI] [PubMed] [Google Scholar]
- 22. Skálová A, Hyrcza MD, Leivo I. Update from the 5th edition of the World Health Organization classification of head and neck tumors: salivary glands. Head Neck Pathol 2022; 16: 40–53. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 23. Eveson JW, Nagao T. Diseases of the salivary glands. In: Surgical pathology of the head and neck. CRC Press: Boca Raton, FL, 2009; 475–648. [Google Scholar]
- 24. Takahashi H, Tada Y, Saotome T, et al. Phase II trial of trastuzumab and docetaxel in patients with human epidermal growth factor receptor 2‐positive salivary duct carcinoma. J Clin Oncol 2019; 37: 125–134. [DOI] [PubMed] [Google Scholar]
- 25. Fushimi C, Tada Y, Takahashi H, et al. A prospective phase II study of combined androgen blockade in patients with androgen receptor‐positive metastatic or locally advanced unresectable salivary gland carcinoma. Ann Oncol 2018; 29: 979–984. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 26. Boon E, Bel M, van Boxtel W, et al. A clinicopathological study and prognostic factor analysis of 177 salivary duct carcinoma patients from the Netherlands. Int J Cancer 2018; 143: 758–766. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 27. Gilbert MR, Sharma A, Schmitt NC, et al. A 20‐year review of 75 cases of salivary duct carcinoma. JAMA Otolaryngol Head Neck Surg 2016; 142: 489–495. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 28. Masubuchi T, Tada Y, Maruya S, et al. Clinicopathological significance of androgen receptor, HER2, Ki‐67 and EGFR expressions in salivary duct carcinoma. Int J Clin Oncol 2015; 20: 35–44. [DOI] [PubMed] [Google Scholar]
- 29. Ibrahim D, Nakamuta N, Taniguchi K, et al. Histological and lectin histochemical studies on the olfactory and respiratory mucosae of the sheep. J Vet Med Sci 2014; 76: 339–346. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 30. Wi GR, Moon BI, Kim HJ, et al. A lectin‐based approach to detecting carcinogenesis in breast tissue. Oncol Lett 2016; 11: 3889–3895. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 31. Enomoto T, Okada H, Tomita H, et al. Glycocalyx analysis of bladder cancer: three‐dimensional images in electron microscopy and vicia villosa lectin as a marker for invasiveness in frozen sections. Front Cell Dev Biol 2023; 11: 1308879. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 32. Barnes DM, Harris WH, Smith P, et al. Immunohistochemical determination of oestrogen receptor: comparison of different methods of assessment of staining and correlation with clinical outcome of breast cancer patients. Br J Cancer 1996; 74: 1445–1451. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 33. Sobral AP, Rego MJ, Cavalacanti CL, et al. ConA and UEA‐I lectin histochemistry of parotid gland mucoepidermoid carcinoma. J Oral Sci 2010; 52: 49–54. [DOI] [PubMed] [Google Scholar]
- 34. Yang LJ. [Immunohistochemical localization of lectins in normal salivary glands and mucoepidermoid carcinoma]. Zhonghua Kou Qiang Yi Xue Za Zhi 1990; 25: 96–98, 126. [PubMed] [Google Scholar]
- 35. Born IA, Zimmer KP, Schwechheimer K, et al. Binding sites of Ulex europaeus‐lectin I in human parotid gland. A light‐microscopic and ultrastructural study using the immunoperoxidase technique and immunocryoultramicrotomy. Cell Tissue Res 1987; 248: 455–461. [DOI] [PubMed] [Google Scholar]
- 36. Takai Y, Hyun KH, Hosaka M, et al. Histochemical studies on concanavalin A‐binding in experimental carcinoma of the mouse submandibular gland. J Oral Pathol 1984; 13: 429–437. [DOI] [PubMed] [Google Scholar]
- 37. Hsu SM, Raine L. Warthin's tumor‐‐epithelial cell differences. Am J Clin Pathol 1982; 77: 78–82. [DOI] [PubMed] [Google Scholar]
- 38. Mazalovska M, Kouokam JC. Plant‐derived lectins as potential cancer therapeutics and diagnostic tools. Biomed Res Int 2020; 2020: 1631394. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 39. Kitaguchi D, Oda T, Enomoto T, et al. Lectin drug conjugate therapy for colorectal cancer. Cancer Sci 2020; 111: 4548–4557. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 40. Zhou SM, Cheng L, Guo SJ, et al. Lectin RCA‐I specifically binds to metastasis‐associated cell surface glycans in triple‐negative breast cancer. Breast Cancer Res 2015; 17: 36. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 41. Bhutia SK, Panda PK, Sinha N, et al. Plant lectins in cancer therapeutics: targeting apoptosis and autophagy‐dependent cell death. Pharmacol Res 2019; 144: 8–18. [DOI] [PubMed] [Google Scholar]
- 42. He S, Simpson BK, Sun H, et al. Phaseolus vulgaris lectins: a systematic review of characteristics and health implications. Crit Rev Food Sci Nutr 2018; 58: 70–83. [DOI] [PubMed] [Google Scholar]
- 43. Scott E, Hodgson K, Calle B, et al. Upregulation of GALNT7 in prostate cancer modifies O‐glycosylation and promotes tumour growth. Oncogene 2023; 42: 926–937. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 44. Berkel C, Cacan E. The expression of O‐linked glycosyltransferase GALNT7 in breast cancer is dependent on estrogen‐, progesterone‐, and HER2‐receptor status, with prognostic implications. Glycoconj J 2023; 40: 631–644. [DOI] [PubMed] [Google Scholar]
- 45. Hua S, Li H, Liu Y, et al. High expression of GALNT7 promotes invasion and proliferation of glioma cells. Oncol Lett 2018; 16: 6307–6314. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 46. Sørensen AL, Reis CA, Tarp MA, et al. Chemoenzymatically synthesized multimeric Tn/STn MUC1 glycopeptides elicit cancer‐specific anti‐MUC1 antibody responses and override tolerance. Glycobiology 2006; 16: 96–107. [DOI] [PubMed] [Google Scholar]
- 47. Natarajan A, Jaroentomeechai T, Cabrera‐Sánchez M, et al. Engineering orthogonal human O‐linked glycoprotein biosynthesis in bacteria. Nat Chem Biol 2020; 16: 1062–1070. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 48. Pinto R, Carvalho AS, Conze T, et al. Identification of new cancer biomarkers based on aberrant mucin glycoforms by in situ proximity ligation. J Cell Mol Med 2012; 16: 1474–1484. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 49. Wiest I, Alexiou C, Friese K, et al. Expression of the tumor‐associated mucin 1 epitope analyzed with the humanized PankoMab‐GEX™ antibody in malignant and normal tissues of the head and neck. Anticancer Res 2016; 36: 3179–3184. [PubMed] [Google Scholar]
- 50. Posey AD Jr, Schwab RD, Boesteanu AC, et al. Engineered CAR T cells targeting the cancer‐associated Tn‐glycoform of the membrane mucin MUC1 control adenocarcinoma. Immunity 2016; 44: 1444–1454. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 51. Mao L, Su S, Li J, et al. Development of engineered CAR T cells targeting tumor‐associated glycoforms of MUC1 for the treatment of intrahepatic cholangiocarcinoma. J Immunother 2023; 46: 89–95. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 52. Sharma P, Marada V, Cai Q, et al. Structure‐guided engineering of the affinity and specificity of CARs against Tn‐glycopeptides. Proc Natl Acad Sci U S A 2020; 117: 15148–15159. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 53. Posey AD Jr, Clausen H, June CH. Distinguishing truncated and normal MUC1 glycoform targeting from Tn‐MUC1‐specific CAR T cells: specificity is the key to safety. Immunity 2016; 45: 947–948. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 54. Pifferi C, Aguinagalde L, Ruiz‐de‐Angulo A, et al. Development of synthetic, self‐adjuvanting, and self‐assembling anticancer vaccines based on a minimal saponin adjuvant and the tumor‐associated MUC1 antigen. Chem Sci 2023; 14: 3501–3513. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 55. Scheid E, Major P, Bergeron A, et al. Tn‐MUC1 DC vaccination of rhesus macaques and a phase I/II trial in patients with nonmetastatic castrate‐resistant prostate cancer. Cancer Immunol Res 2016; 4: 881–892. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 56. Gabba A, Attariya R, Behren S, et al. MUC1 glycopeptide vaccine modified with a GalNAc glycocluster targets the macrophage galactose C‐type lectin on dendritic cells to elicit an improved humoral response. J Am Chem Soc 2023; 145: 13027–13037. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 57. Swain SM, Baselga J, Kim SB, et al. Pertuzumab, trastuzumab, and docetaxel in HER2‐positive metastatic breast cancer. N Engl J Med 2015; 372: 724–734. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 58. Liu F, Yin G, Xue S, et al. Adverse event profile differences between trastuzumab emtansine and trastuzumab deruxtecan: a real‐world, pharmacovigilance study. J Cancer 2023; 14: 3275–3284. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 59. Dowling GP, Daly GR, Keelan S, et al. Efficacy and safety of trastuzumab deruxtecan in breast cancer: a systematic review and meta‐analysis. Clin Breast Cancer 2023; 23: 847–855.e842. [DOI] [PubMed] [Google Scholar]
- 60. Bosi C, Bartha Á, Galbardi B, et al. Pan‐cancer analysis of antibody‐drug conjugate targets and putative predictors of treatment response. Eur J Cancer 2023; 195: 113379. [DOI] [PubMed] [Google Scholar]
Associated Data
This section collects any data citations, data availability statements, or supplementary materials included in this article.
Supplementary Materials
Figure S1. Salivary gland benign tumors can be differentiated using peanut agglutinin (PNA) and Griffonia simplicifolia lectin (GSL) II
Table S1. Lectin screening kits
Table S2. N‐acetylgalactosaminyltransferase (GALNT) primers
Data Availability Statement
RNA‐seq data were downloaded from the Gene Expression Omnibus database (https://www.ncbi.nlm.nih.gov/geo/; accession no. GSE138581). The data that support the findings of this study are available from the corresponding author (HS), upon reasonable request.