Sage Journals: Discover world-class research

Abstract

BACKGROUND AND AIM:

Gastric cancer (GC) is a common and fatal malignancy with a worldwide occurrence. There still lacks effective biomarkers for precisely evaluating GC. Saliva is a biological fluid with enormous diagnostic potentials which emerged many advantages. We aimed to discover the novel biomarkers for accurately distinguishing early GC based on saliva glycopatterns.

METHODS:

We used Aleuria Aurantia Lectin (AAL)-magnetic particle conjugates to isolate fucosylated glycoproteins in the pooled saliva of healthy volunteers (HV, $n=$ 51) and patients with atrophic gastritis (AG, $n=$ 51) or GC ( $n=$ 51), following to release the $N$ - and $O$ -linked glycans from the isolated proteins with PNGase F and NaClO, and further identified the released glycans by MALDI-TOF/TOF-MS, respectively.

RESULTS:

A total of 9/9, 8/11, and 9/9 fucosylated $N$ -/ $O$ -linked glycans were annotated in the isolated salivary proteins from HV, AG, and GC, respectively. Among these, six fucosylated $N$ -linked glycansand four $O$ -linked glycans exhibited significantly increased expression levels in GC, while five fucosylated $N$ -linked glycans and ten fucosylated $O$ -linked glycans exhibited significantly decreased expression levels in GC. The proportion of fucosylated $N$ -linked glycans was decreased in GC (41.66%) compared with AG (43.63%) and HV (52.57%), as well as the fucosylated $O$ -linked glycans was apparently decreased in GC (19.58%) compared with AG (25.43%) and HV (55.54%).

CONCLUSIONS:

This study could provide pivotal information to distinguish among HV, AG, and GC, and facilitate the discovery of biomarkers for GC diagnosis based on precise alterations of $N$ - and $O$ -linked glycans in saliva.

Keywords

Saliva gastric cancer MALDI-TOF/TOF-MS/MS N- and O-Linked glycan

1. Introduction

Although the incidence and mortality have declined in recent years, gastric cancer (GC) is still a major public health problem, particularly in the eastern countries [1]. Helicobacter pylori (H. pylori) infection is known as the major trigger of atrophic gastritis (AG), and AG is the precursor of gastric cancer [2, 3]. The existing biomarkers such as CEA, CA125, CA19-9 and CA72-4 are not sensitive or specific enough to characterize the early GC [4, 5]. These reasons result in that about 80% patients are not diagnosed until advanced stages, and inducing the high mortality [4, 5]. Thus, identification of the difference among HV, AG, and GC to discover the novel biomarkers is an urgent demand for accurately distinguishing early GC.

As the two major types of protein glycosylation, $N$ -glycosylation and $O$ -glycosylation involved in a variety of critical biological processes and closely associated with human health and diseases [6]. Many reports indicated that glycosylation is directly related to GC generation and development. H. pylori recognized the Lewis b and H-type 1 carbohydrate structures to adhere gastric mucosa, which lead the increasing of sialyl-Lewis x level and induce sustained inflammation in stomach [7]. Sialyl-Tn antigen showed a significant increase in AG [8]. Moreover, decreased core-fucosylation and increased $\beta$ 1, 6-GlcNAc branched $N$ -linked glycan contributed to malignancy in GC [4, 9].

Saliva is a kind of mixed fluid secreted by parotid, submandibular, sublingual and minor glands, which contains several molecular profiles that could reflect the condition of body [10, 11]. Using saliva in diagnosis has many advantages such as noninvasive and convenient, which might facilitate early detection of many diseases [11]. Recent studies have elucidated that many diseases could be detected by saliva, such as familial juvenile systemic lupus erythematosus [12], type II diabetes [13], laryngeal cancer [14], oral cancer [15], breast cancer [16], lung cancer [17], pancreatic cancer [18], and GC [19]. Salivary glycopatterns could also be used for detecting of disease and cancer [20, 21]. Our previous study revealed that salivary glycopatterns recognized by Aleuria Aurantia Lectin (AAL) exhibited significantly decreased expression levels in GC compared with HV and AG (all $p\leqslant$ 0.001), which were negatively correlated with the development of GC [22].

In this study, we further investigated the alterations of fucosylated $N$ - and $O$ -linked glycans especially recognized by AAL in saliva from 153 subjects (51 healthy volunteers (HV), 51 patients with AG, and 51 patients with GC) using MALDI-TOF/TOF-MS. This study could provide pivotal information to distinguish among HV, AG, and GC, and facilitate the discovery of biomarkers for GC diagnosis based on precise alterations of $N$ - and $O$ -linked glycans in saliva. The overall strategy applied is schematically represented in Fig. 1.

Table 1
Clinical characteristics of healthy volunteers and patients with atrophic gastritis and gastric cancer

Subject groups	HV	AG	GC
Males, (Total)	26 (51)	26 (51)	26 (51)
Age, y, mean $\pm$ SD	53.7 $\pm$ 5.5	53.3 $\pm$ 6.2	54.1 $\pm$ 7.4
Pathological (AJCC) ${}^{\text{a}}$
I			17
II			15
III			19
IV
Tumor (T), $n$ ,
T1			16
T2			11
T3			3
T4			21
Node (N), $n$ ,
N0			17
N1			9
N2			15
N3			10
Metastasis (M), $n$ ,
M0			51
M1			0

${}^{\text{a}}$ AJCC, American Joint Committee on Cancer staging system (7 ${}^{\text{th}}$ edition).

Figure 1.

The overall strategy applied in this study.

2. Materials and methods

2.1 Study approval and population

The collection and use of human whole saliva for research presented here were approved by the Ethical Committee of Northwest University (Xi’an, China), First People’s Hospital of Chenzhou (Chenzhou, China). Written informed consent was received from participants for the collection of their whole saliva. This study was conducted in accordance with the ethical guidelines of the Declaration of Helsinki. In total, 51 patients of AG, 51 patients of GC and 51 age- and sex-matched HV were enrolled for this study. All the patients were histopathologically confirmed in the First People’s Hospital of Chenzhou and who received preoperative radiotherapy, chemotherapy, chemoradiotherapy or curative were excluded from the study. The clinical characteristics of HV, AG, and GC were summarized in Table 1.

2.2 Whole saliva collection and preparation

The collection of all the saliva samples were according to the protocol [20, 21]. Saliva was collected without stimulation between 9 a.m. and 10 a.m. at least 2 h after intake of food. Before collecting, the mouth was rinsed with physiological saline immediately. Whole saliva (about 1 mL) was collected and placed on ice. Protease Inhibitor Cocktail (1 $\mu$ L/mL of whole saliva) was added to the saliva immediately after collection to minimize protein degradation. Then the supernatant of crude saliva protein was separated by centrifuging at 1,2000 $\times$ g at 4 ${}^{\circ}$ C for 30 min in order to remove the insoluble materials. The supernatant was collected and filtered through a 0.22 $\mu$ m pore size against bacteria and microorganism, and then was immediately used or stored at $-$ 80 ${}^{\circ}$ C. To normalize the differences between subjects and to tolerate individual variation, samples in each group were pooled respectively.

2.3 Preparation of AAL-magnetic particle conjugates

AAL was purchased from Vector Laboratories (Burlingame, CA, USA). The epoxy-coated magnetic particles (2 mg) were rinsed three times with ethanol and then rinsed three times with coupling buffer (5 mM NaB ${}_{4}$ O ${}_{7}$ , 180 mM H ${}_{3}$ BO ${}_{4}$ , 150 mM Na $+$ , pH 7.4), and reacted with 600 uL of 0.5 mg/mL AAL solution (AAL dissolve in coupling buffer) at room temperature for 6 h under gentle shaking according to the protocol [23]. The unbound lectins were removed from the conjugates by washing six times with the coupling buffer.

2.4 Selective isolation of glycoprotein from human saliva by AAL-magnetic particle conjugates

The prepared AAL-magnetic particle conjugates were blocked by 600 $\mu$ L blocking buffer (2% ethanol amine, 0.1% BSA, pH 9.0) at room temperature for 6 h under gentle shaking, and then rinsed 3 times with the binding buffer (100 mM Tris-HCl, 150 mM NaCl, 1 mM CaCl ${}_{2}$ , 1 mM MgCl ${}_{2}$ , 1 mM MnCl ${}_{2}$ , pH 7.4). Two mg proteins dissolved in binding buffer, and then added to AAL-magnetic particle conjugates at room temperature for 3 h under gentle shaking. The unbound proteins were removed from the conjugates by washing six times with the washing buffer (0.1% Tween-20, 100 mM Tris-HCl, 150 mM NaCl, 1 mM CaCl ${}_{2}$ , 1 mM MgCl ${}_{2}$ , 1 mM MnCl ${}_{2}$ , pH 7.2). And the selective isolated glycoproteins were eluted with 400 $\mu$ L eluting buffer (8 M Urea, 40 mM NH ${}_{4}$ HCO ${}_{3}$ ) for 1 h. The AAL-isolated glycoproteins were quantitated by BCA.

2.5 Trypsin digestion of the AAL-isolated glycoproteins and PNGase F release of the N-linked glycans

The AAL-isolated glycoproteins (200 $\mu$ g) were concentrated by 10 k Da molecular weight cutoff (MWCO) ultrafiltration unit (Millipore), and rinsed three times with 400 $\mu$ L urea solution (8 M urea, 40 mM NH ${}_{4}$ HCO ${}_{3}$ ). 10 $\mu$ L DTT solution (200 mM DTT, 40 mM NH ${}_{4}$ HCO ${}_{3}$ ) were added in the 10 k Da molecular weight cutoff (MWCO) ultrafiltration unit and then heated at 95 ${}^{\circ}$ C for 20 min. After glycoproteins denaturation, 20 $\mu$ L IAM solution (200 mM IAM, 40 mM NH ${}_{4}$ HCO ${}_{3}$ ) were added and reacted at 56 ${}^{\circ}$ C for 20 min in the dark. The mixture was rinsed 3 times with 50 mM NH ${}_{4}$ HCO ${}_{3}$ in the 10 k Da MWCO ultrafiltration unit, and 2 ug trypsin was added to digest the glycoproteins overnight at 37 ${}^{\circ}$ C. The mixture was heated at 80 ${}^{\circ}$ C for 5 min to deactivate the activity of trypsin. After the temperature of mixture restored to room temperature, 2 $\mu$ L PNGase F (New England BioLabs) was added and incubated at 37 ${}^{\circ}$ C overnight to release the $N$ -linked glycans from the glycopeptides. Finally, the filtrates that contained peptides and $N$ -linked glycans were collected by centrifugation, and the ultrafiltration unit was rinsed with 200 $\mu$ L of 40 mM NH ${}_{4}$ HCO ${}_{3}$ and centrifuged. This step was repeated 2 times. The filtrates that contained peptides and $N$ -linked glycans were lyophilized by CentriVap (Labconco, America).

2.6 NaClO release of the O-linked glycans

The protocol of $O$ -linked glycans releasing was based on the protocol [24]. And it was adjusted to small scale release of $O$ -linked glycan in this study. The selective isolation of glycoproteins (200 $\mu$ g) were concentrated by the new 10 k Da MWCO ultrafiltration unit, and rinsed 6 times with ultrapure water. The volume of glycoproteins was adjusted about 200 $\mu$ L, and 100 $\mu$ L of 6% NaClO was added under gentle shaking and shook for another 30 min. 15 $\mu$ L of 10% formic acid was added slowly to the reaction mixture and cooled on ice for 10 min. The mixture was centrifuged and rinsed 6 times with ultrapure water to remove formic acid and other residues. The mixture was adjusted to a volume of 500 $\mu$ L by addition of water and adjusted to pH 7.6 by addition of NaOH. 6.64 $\mu$ L of 6% NaClO was added in the mixture and reacted at room temperature for 24 h under gentle shaking. 8 $\mu$ L of 10% formic acid was added slowly to the reaction mixture and cooled on ice for 10 min. Finally, the filtrates that contained $O$ -linked glycans were collected by centrifugation, and the ultrafiltration unit was rinsed with 200 $\mu$ L of 40 mM NH ${}_{4}$ HCO ${}_{3}$ and centrifuged. This step was repeated 2 times, and the collected filtrates were lyophilized.

2.7 Purification of $N$ -/ $O$ -linked glycans

Purification of glycans was performed as described previously [23, 25]. The C ${}_{18}$ SepPak column (Waters) was conditioned by washing two times with ACN, 0.1% TFA in 50% ACN, and 0.1% TFA, sequentially. The samples were loaded and pipetted into the column bed. The columns were eluted three times with 0.5 mL of 0.1% TFA to obtain the glycans. The eluted glycans were pooled and lyophilized. And Sepharose 4B was used for further purification and desalting [23]. The Sepharose 4B was washed with methanol/water (MW; 1:1, v/v) and 1-butanol/methanol/water (BMW; 5:1:1, v/v) under centrifugation. The glycans were dissolved in 500 $\mu$ L BMW and mixed with the Sepharose 4B. The mixture was gently shaken for 45 min and washed 3 times with BMW. Glycans were eluted with MW, and the eluent was collected and lyophilized.

2.8 Characterization of the $N$ - and $O$ -linked glycans by MALDI-TOF/TOF-MS

The purified glycans were characterized by MALDI-TOF/TOF-MS (UltrafleXtreme, Bruker Daltonics). Glycans were resuspended in 10 $\mu$ L of MW, and 1 $\mu$ L was spotted directly on a MTP AnchorChip sample target and dried. Then 1 $\mu$ L of 10 mg/mL DHB in 50% (v/v) methanol solution was spotted to the recrystallize glycans. Mass calibration was performed using peptide calibration standards (250 calibration points; Bruker). Measurements were taken in positive-ion mode, m/z data were generated using FlexAnalysis software (Bruker Daltonics), analyzed and annotated using the GlycoWorkbench software program. Relative intensity (RI) was calculated by dividing the intensity of a given type of glycan by the total glycans intensity [23, 25, 26].

3. Results

3.1 The $N$ -linked glycan profiles of the isolated salivary glycoprotein

The identified $N$ -linked glycans and their proposed structures from the pooled salivary samples of HV, AG, and GC were shown in Fig. 2. A total of 14, 13, and 13 $N$ -linked glycans were identified and annotated, respectively. Of these, 7 $N$ -linked glycans (m/z 1485.534 ((Fuc) ${}_{1}$ (GlcNAc) ${}_{2}$ $+$ (Man) ${}_{3}$ (GlcNAc) ${}_{2}$ ), 1743.581 ((Man) ${}_{5}$ $+$ (Man) ${}_{3}$ (GlcNAc) ${}_{2}$ ), 1752.610 ((Fuc) ${}_{2}$ (Gal) ${}_{1}$ (Man) ${}_{1}$ (GlcNAc) ${}_{1}$ $+$ (Man) ${}_{3}$ (GlcNAc) ${}_{2}$ ), 1776.629 ((Fuc) ${}_{1}$ (Sia) ${}_{1}$ (GalNAc) ${}_{1}$ (GlcNAc) ${}_{1}$ $+$ (Man) ${}_{3}$ (GlcNAc) ${}_{2}$ ), 1787.657 ((Fuc) ${}_{1}$ (Gal) ${}_{2}$ (GlcNAc) ${}_{2}$ $+$ (Man) ${}_{3}$ (GlcNAc) ${}_{2}$ ), 2158.777 ((Fuc) ${}_{2}$ (Gal) ${}_{2}$ (GlcNAc) ${}_{3}$ $+$ (Man) ${}_{3}$ (GlcNAc) ${}_{2}$ ), 2161.823 ((Fuc) ${}_{3}$ (GalNAc) ${}_{2}$ (GlcNAc) ${}_{2}$ $+$ (Man) ${}_{3}$ (GlcNAc) ${}_{2}$ )) observed in all groups of HV, AG, and GC. However, one $N$ -linked glycan (m/z 1663.581 ((Gal) ${}_{2}$ (GlcNAc) ${}_{2}$ $+$ (Man) ${}_{3}$ (GlcNAc) ${}_{2}$ )) was observed only in HV, one $N$ -linked glycan (m/z 1792.624 ((Sia) ${}_{1}$ (Gal) ${}_{1}$ (GlcNAc) ${}_{2}$ $+$ (Man) ${}_{3}$ (GlcNAc) ${}_{2}$ )) was observed only in AG, and five $N$ -linked glycans (m/z 1460.500 ((Gal) ${}_{1}$ (GlcNAc) ${}_{1}$ (Man) ${}_{1}$ $+$ (Man) ${}_{3}$ (GlcNAc) ${}_{2}$ ), 1501.529 ((Gal) ${}_{1}$ (GlcNAc) ${}_{2}$ $+$ (Man) ${}_{3}$ (GlcNAc) ${}_{2}$ ), 1751.597 ((Sia) ${}_{1}$ (Gal) ${}_{1}$ (GlcNAc) ${}_{1}$ (Man) ${}_{1}$ $+$ (Man) ${}_{3}$ (GlcNAc) ${}_{2}$ ), 2141.761 ((Fuc) ${}_{1}$ (Sia) ${}_{1}$ (Gal) ${}_{1}$ (GalNAc) ${}_{1}$ (GlcNAc) ${}_{2}$ $+$ (Man) ${}_{3}$ (GlcNAc) ${}_{2}$ ), 2180.692 ((Fuc) ${}_{1}$ (Sia) ${}_{1}$ (Gal) ${}_{1}$ (Sulf-Gal) ${}_{1}$ (GlcNAc) ${}_{2}$ $+$ (Man) ${}_{3}$ (GlcNAc) ${}_{2}$ )) were observed only in GC. The MS/MS spectra of the precursor ions m/z 1663.581, 1752.618, and 1809.639 were illustrated in Fig. 3A–C.

Figure 2.

MALDI-TOF/TOF-MS spectra of the purified $N$ -linked glycans from the pooled salivary samples of HV, AG, and GC, respectively. Detailed glycan structures were analyzed using the GlycoWorkbench software. Proposed structures and their m/z values were shown for each peak. Blue square $=$ GlcNAc, green circle $=$ Man, yellow circle $=$ Gal, yellow square $=$ GalNAc, purple diamond $=$ NeuAc, red triangle $=$ Fuc.

Figure 3.

MALDI-TOF/TOF MS/MS analyzing the $N$ -glycan precursor ion from MS spectra. The three $N$ -glycan peaks (A) m/z 1663.581, (B) 1752.618, and (C) 1809.639 subjected to MS/MS analysis.

Table 2

The proposed fucosylated structures for 11 $N$ -linked glycan peaks detected in the present study

NO	Calculated m/z	Charge	% N-glycan			Fold-Change
			HV	AG	GC	AG/HV	GC/HV	GC/AG
1	1485.534	[M+Na] ${}^{+}$	1.05	0.84	3.37	0.80	3.22	4.01
2	1647.587	[M+Na] ${}^{+}$	3.12	2.83	ND	0.91	0	0
3	1730.636	[M+H] ${}^{+}$	6.21	4.04	11.01	0.65	1.77	2.73
	1752.610	[M+Na] ${}^{+}$	3.89	2.22	6.17	0.57	1.59	2.78
4	1768.570	[M+Na] ${}^{+}$	3.35	ND	2.35	0	0.70	–
5	1776.629	[M+Na] ${}^{+}$	19.29	19.78	3.55	1.03	0.18	0.18
6	1787.657	[M+H] ${}^{+}$	6.70	6.75	2.98	1.01	0.45	0.44
	1809.639	[M+Na] ${}^{+}$	4.70	3.76	ND	0.80	0	0
7	1834.671	[M+Na] ${}^{+}$	2.72	2.04	ND	0.75	0	0
8	2136.795	[M+H] ${}^{+}$	ND	ND	2.27	–	–	–
	2158.777	[M+Na] ${}^{+}$	0.76	0.79	4.31	1.04	5.68	5.46
9	2141.761	[M+Na] ${}^{+}$	ND	ND	1.77	–	–	–
10	2161.823	[M+H] ${}^{+}$	0.78	0.58	2.33	0.74	2.99	4.02
11	2180.692	[M+Na] ${}^{+}$	ND	ND	1.55	0	–	–
			52.57	43.63	41.66

*Monosaccharides are represented according to MS-tools from the GlycoWorkbench software (GlcNAc, blue square; Man, green circle; Gal, yellow circle; Fuc, red triangle; NeuAc, purple diamond; ND, not detected in the sample. The Relative Ratios of each peak from HV, AG, and GC groups are compared with each other based upon fold change in pairs. –, the denominator of the fold-change is zero.

Figure 4.

MALDI-TOF/TOF-MS spectra of purified $O$ -linked glycans from the pooled salivary samples of HV, AG, and GC, respectively. Detailed glycan structures were analyzed using the GlycoWorkbench software. Proposed structures and their m/z values were shown for each peak. Blue square $=$ GlcNAc, green circle $=$ Man, yellow circle $=$ Gal, yellow square $=$ GalNAc, purple diamond $=$ NeuAc, red triangle $=$ Fuc.

Table 3

The proposed fucosylated structures for 19 $O$ -linked glycan peaks detected in the present study

NO.	Calculated m/z	Charge	% O-glycan			Fold-change
			HV	AG	GC	AG/HV	GC/HV	GC/AG
1	602.208	[M+H] ${}^{+}$	2.99	0	0	–	ND	ND
	624.190	[M+Na] ${}^{+}$	2.62	0	0	–	ND	ND
2	616.224	[M+H] ${}^{+}$	16.42	0	0	–	ND	ND
	638.206	[M+Na] ${}^{+}$	22.72	0.34	0	0	7.07	ND
3	652.221	[M+Na] ${}^{+}$	1.83	0	0	–	ND	ND
4	715.221	[M+Na] ${}^{+}$	1.66	1.35	2.84	2.11	2.24	4.72
5	1039.284	[M+Na] ${}^{+}$	3.88	0.56	0.60	1.05	2.19	2.32
6	1069.279	[M+Na] ${}^{+}$	ND	–	–	1.20	1.30	1.56
7	1113.398	[M+H] ${}^{+}$	ND	–	–	0	1.70	ND
8	1156.420	[M+H] ${}^{+}$	ND	–	–	0	1.40	ND
9	1201.337	[M+Na] ${}^{+}$	ND	–	–	0	1.48	ND
10	1244.436	[M+H] ${}^{+}$	ND	–	–	1.00	1.97	1.97
11	1338.459	[M+Na] ${}^{+}$	2.03	0	0.96	–	1.98	1.94
12	1354.454	[M+Na] ${}^{+}$	1.60	0	0	–	ND	ND

Table 3, continued
NO.	Calculated m/z	Predicted structure*	Charge	% O-glycan			Fold-change
				HV	AG	GC	AG/HV	GC/HV	GC/AG
13	1377.390		[M+Na] ${}^{+}$	ND	–	–	0	2.09	ND
14	1381.481		[M+Na] ${}^{+}$	ND	–	–	–	ND	1.42
15	1420.488		[M+H] ${}^{+}$	ND	–	–	0	2.01	ND
16	1426.475		[M+Na] ${}^{+}$	ND	–	–	–	ND	2.10
17	1470.453		[M+H] ${}^{+}$	ND	–	–	–	ND	2.01
18	1500.512		[M+Na] ${}^{+}$	1.50	0	0	–	ND	ND
19	1516.507		[M+Na] ${}^{+}$	1.28	0	1.20	–	ND	1.54
Sum				55.54	25.43	19.58

3.2 Alterations of fucosylated

N

-linked glycan profiles among healthy volunteers, gastric cancer, and atrophic gastritis patients

$N$ -linked glycan with the fucosylated structures (Fuc $\alpha$ 1-6GlcNAc and Fuc $\alpha$ 1-3(Gal $\beta$ 1-4)GlcNAc) could be specifically recognized by AAL. It is noticeable that there were 9, 8, and 9 fucosylated $N$ -linked glycan identified and annotated in the pooled salivary samples of HV, AG, and GC, respectively (Table 2). Of these, six fucosylated $N$ -linked glycans (m/z 1485.534, 1752.610, 1776.629, 1787.657, 2158.777, and 2161.823) were observed in all groups of HV, AG, and GC, while two fucosylated $N$ -linked glycans (m/z 1647.587 and 1834.671) were simultaneously observed in the groups of HV and AG, one fucosylated $N$ -linked glycan (m/z 1768.570) was simultaneously observed in the groups of HV and GC. And two fucosylated $N$ -linked glycans (m/z 2141.761 and 2180.692) were observed only in GC.

The proportion of the fucosylated $N$ -linked glycans was decreased in GC (41.66%) compared with AG (45.37%) and HV (53.60%), which was consistent with the results of our previous study [22]. To further analyze the alteration of the expression levels of salivary fucosylated $N$ -linked glycans with the development of GC, the relative intensities (RIs) of each fucosylated $N$ -linked glycan structure in the pooled salivary samples of HV, AG, and GC were compared each other based upon fold-changes. The results showed that the RIs of 6 fucosylated $N$ -linked glycans (m/z 1485.534, 1752.610, 2141.761, 2158.777, 2161.823, 2180.692) were increased in GC compared with AG and HV (all fold-changes $>$ 1.5), while the RIs of four fucosylated $N$ -linked glycans (m/z 1647.587, 1776.629, 1809.639, and 1834.671) were decreased in GC compared with AG and HV (all fold-changes $<$ 0.67).

3.3 The $O$ -linked glycan profiles of the isolated salivary glycoprotein

The $O$ -glycan-acids were released from the isolated glycoproteins by NaClO oxidative method, which was 58-Da or 72-Da increased in molecular mass over the corresponding free reducing glycans based on the glycan linked to the serine or threonine residues [24]. The identified $O$ -linked glycans and their proposed structures from the pooled salivary samples of HV, AG, and GC were shown in Fig. 4. A total of 14, 16, and 14 $O$ -linked glycans were identified and annotated, respectively. Of these, there were six $O$ -linked glycans (m/z 536.107 (Sulf-Gal) ${}_{1}$ (GalNAc) ${}_{1}$ , 698.160 (Sulf-Gal) ${}_{1}$ (Gal) ${}_{1}$ (GalNAc) ${}_{1}$ , 715.221 (Fuc) ${}_{2}$ (Gal) ${}_{2}$ , 876.211 (Sia) ${}_{1}$ (Sulf-GlcNAc) ${}_{1}$ (GalNAc) ${}_{1}$ , and 1039.284 (Fuc) ${}_{2}$ (Gal) ${}_{1}$ (Sulf-GlcNAc) ${}_{1}$ (GalNAc) ${}_{1}$ , 1338.459 (Fuc) ${}_{2}$ (Gal) ${}_{2}$ (GlcNAc) ${}_{2}$ (GalNAc) ${}_{1}$ ) presented in all samples of HV, AG, and GC. However, 6 $O$ -linked glycans (m/z 585.193 (Sia) ${}_{1}$ (GalNAc) ${}_{1}$ , 604.203 (Gal) ${}_{2}$ (GalNAc) ${}_{1}$ , 624.190 (Fuc) ${}_{1}$ (Gal) ${}_{1}$ (GalNAc) ${}_{1}$ , 652.221 (Met-Fuc) ${}_{1}$ (Met-Gal) ${}_{1}$ (GalNAc) ${}_{1}$ , 1354.454 (Fuc) ${}_{1}$ (Gal) ${}_{3}$ (GlcNAc) ${}_{2}$ (GalNAc) ${}_{1}$ , and 1500.512 (Fuc) ${}_{2}$ (Gal) ${}_{3}$ (GlcNAc) ${}_{2}$ (GalNAc) ${}_{1}$ ) were presented only in HV, and five $O$ -linked glycans (m/z 1113.398 (Fuc) ${}_{2}$ (Gal) ${}_{2}$ (GlcNAc) ${}_{1}$ (GalNAc) ${}_{1}$ , 1156.420 (Fuc) ${}_{1}$ (Gal) ${}_{2}$ (GlcNAc) ${}_{2}$ (GalNAc) ${}_{1}$ , 1201.337 (Fuc) ${}_{2}$ (Sulf-Gal) ${}_{1}$ (Gal) ${}_{1}$ (GlcNAc) ${}_{1}$ (GalNAc) ${}_{1}$ , 1377.390 (Fuc) ${}_{2}$ (Gal) ${}_{3}$ (Sulf-GlcNAc) ${}_{1}$ (GalNAc) ${}_{1}$ and 1420.488 (Sia) ${}_{1}$ (Fuc) ${}_{1}$ (Gal) ${}_{3}$ (GlcNAc) ${}_{1}$ (GalNAc) ${}_{1}$ ) were presented only in AG, while three $O$ -linked glycans (m/z 1381.481 (Fuc) ${}_{1}$ (Gal) ${}_{2}$ (GlcNAc) ${}_{1}$ (GalNAc) ${}_{3}$ , 1426.475 (Sia) ${}_{1}$ (Fuc) ${}_{2}$ (Gal) ${}_{2}$ (GlcNAc) ${}_{1}$ (GalNAc) ${}_{1}$ , and 1470.453 (Sia) ${}_{1}$ (Fuc) ${}_{2}$ (Gal) ${}_{2}$ (Sulf-GlcNAc) ${}_{1}$ (GalNAc) ${}_{1}$ ) were presented only in GC.

3.4 Alterations of fucosylated O-linked glycan profiles among healthy volunteers, gastric cancer, and atrophic gastritis patients

It is also noticeable that there were 9, 11, and 9 fucosylated $O$ -linked glycans identified and annotated in the pooled salivary samples of HV, AG, and GC, respectively (Table 3). Of these, three fucosylated $O$ -linked glycans (m/z 715.221, 1039.284, and 1338.459) were observed in all groups of HV, AG, and GC. One fucosylated $O$ -linked glycan (m/z 638.206) was simultaneously observed in HV and AG, one fucosylated $O$ -linked glycans (m/z 1516.507) was simultaneously observed in HV and GC, two fucosylated $O$ -linked glycans (m/z 1069.279 and 1244.436) were simultaneously observed in AG and GC, and four fucosylated $O$ -linked glycans (m/z 624.190, 652.221, 1354.454, and 1500.512) were observed only in HV, while five fucosylated $O$ -linked glycans (m/z 1113.398, 1156.420, 1201.337, 1377.390, and 1420.488) were observed only in AG, and three fucosylated $O$ -linked glycan peaks (m/z 1381.481, 1426.475, and 1470.453) were observed only in GC.

The proportion of the fucosylated $O$ -linked glycan was decreased in GC (19.58%) compared with AG (25.43%) and HV (55.54%). And the results showed that the RIs of four fucosylated $O$ -linked glycans (m/z 715.221, 1381.481, 1426.475, and 1470.45) were increased in GC compared with HV and AG (all fold-changes $>$ 1.5), while 10 $O$ -linked fucosylated glycans (m/z 624.190, 638.206, 652.221, 1113.398, 1156.420, 1201.337, 1354.454, 1377.390, 1420.488, and 1500.512) were decreased in GC compared with HV or GC (fold-changes $<$ 0.67).

4. Discussion

There are many studies indicated that the alteration of glycans affects the function of glycoproteins as well as depicted that the association between glycosylation and cancer. E-cadherin aberrant $N$ -glycosylation at Asn-554 was demonstrated to affect its critical functions, and the aberrant glycans could be used as potential biomarkers [9]. Besides, the expression levels of core-fucosylated glycans in GC tissue and serum is decreased, which showed contribute to malignant transformation in GC and have great potential to be biomarkers [4, 27]. And other researches indicated that serum glycan profiles may provide potential biomarkers to differentiate GC cases from control [28, 29]. In our previous study, the fucosylated glycan structures recognized by AAL exhibited significantly decreased expression levels in GC [22]. And we further investigated the detailed structural and alteration of $N$ - and $O$ -linked glycans from AAL-isolated glycoproteins in the saliva with the development of GC.

A total of 14/14, 13/16, and 13/14 $N$ -/ $O$ -linked glycans were identified and annotated from the pooled salivary samples of HV, AG, and GC, respectively. Among these, there were 9/9, 8/11, and 9/9 fucosylated $N$ -/ $O$ -linked glycans in the pooled salivary samples of HV, AG, and GC. The proportion of fucosylated $N$ -linked glycans was decreased in GC (41.66%) compared with AG (43.63%) and HV (52.57%), and the proportion of $O$ -linked glycans was apparently decreased in GC (19.58%) compared with AG (25.43%) and HV (55.54%). Interestingly, the fucosylated $N$ -linked glycan peak (m/z 1776.629 ((Fuc) ${}_{1}$ (Sia) ${}_{1}$ (GalNAc) ${}_{1}$ (GlcNAc) ${}_{1}$ $+$ (Man) ${}_{3}$ (GlcNAc) ${}_{2}$ )) was the highest peak in the $N$ -linked glycan profiles of HV (19.29%) and AG (19.78%), but it decreased apparently in GC (3.55%). Conversely, the glycan peaks of m/z 1730.636 ([M+H] ${}^{+}$ ) and 1752.610 ([M+Na]+) with the same structure ((Fuc) ${}_{2}$ (Gal) ${}_{1}$ (Man) ${}_{1}$ (GlcNAc) ${}_{1}$ $+$ (Man) ${}_{3}$ (GlcNAc) ${}_{2}$ ), were the first and second highest peaks in the $N$ -linked glycan profiles of GC (11.01% and 6.17%), but decreased in HV (6.21% and 3.89%) and AG (4.04% and 2.22%). Notably, the structure ((Fuc) ${}_{1}$ (Gal) ${}_{1}$ (GalNAc) ${}_{1}$ ) of m/z 602.208 ([M+H] ${}^{+}$ ) and 624.190 ([M+Na] ${}^{+}$ ) were identified only in HV (2.99% and 2.62%), while their methylated structure ((Met-Fuc) ${}_{1}$ (Gal) ${}_{1}$ (GalNAc) ${}_{1}$ ) of m/z 616.224 ([M+H] ${}^{+}$ ) and 638.206 ([M+Na] ${}^{+}$ ) showed the first and second highest intensity HV (16.42% and 22.72%), however, they all significantly decreased in AG (ND, 7.07%), and disappeared in GC. Besides, the unusual structure of GalNAc $\beta$ 1, 4GlcNAc (LacdiNAc) was found in the saliva samples from HV, AG and GC in this study. The LacdiNAc glycans were identified in many human glycoproteins, such as prostate-specific antigen, MUC1, trefoil factor family 2, RNase I [30, 31, 32, 33]. And some reports indicated that LacdiNAc glycans significantly decreased in human breast cancer, but increased in human prostate, ovarian, and pancreatic, and liver cancers, which have the potential to be cancer biomarkers [34, 35]. The carrier proteins of the LacdiNAc glycans and their correlation with gastric cancer are still need for further research.

In conclusion, the present study investigated the correlation of the fucosylated glycans from the salivary proteins isolated by the AAL-magnetic particle conjugates, and systematically compared their alterations of the fucosylated $N$ - and $O$ -linked glycans among HV, AG and GC. The results showed that there were six fucosylated $N$ -linked glycans (m/z 1485.534, 1752.610, 2141.761, 2158.777, 2161.823, 2180.692) and four $O$ -linked glycans (m/z 715.221, 1381.481, 1426.475, and 1470.453) that exhibited significantly increased expression levels in GC, while four fucosylated $N$ -linked glycans (m/z 1647.587, 1776.629, 1809.639, and 1834.671) and ten fucosylated $O$ -linked glycans (m/z 624.190, 638.206, 652.221, 1113.398, 1156.420, 1201.337, 1354.454, 1377.390, 1420.488, and 1500.512) that exhibited significantly decreased expression levels in GC. Our data provide pivotal information to distinguish among HV, AG and GC, and facilitate the discovery of biomarkers for GC based on precise alterations of fucosylated $N$ - and $O$ -linked glycans in saliva.

Footnotes

Acknowledgments

This work is supported by the National Natural Science Foundation (Grant No. 81372365) and the training Fund of Northwest University (Grant No. YYB17018).

Conflict of interest

The authors declare no competing financial interests.

Abbreviations

References

Torre

L.A.

Bray

Siegel

R.L.

Ferlay

Lortet-Tieulent

and Jemal

, Global cancer statistics, 2012, CA Cancer J Clin 65 (2015), 87–108.

Sheu

B.S.

Yang

H.B.

Yeh

Y.C.

and Wu

J.J.

, Helicobacter pylori colonization of the human gastric epithelium: A bug’s first step is a novel target for us, J Gastroenterol Hepatol 25 (2010), 26–32.

Graham

D.Y.

, Helicobacter pylori update: Gastric cancer, reliable therapy, and possible benefits, Gastroenterology 148 (2015), 719.

Liu

Yan

Huang

Wang

Fang

Jiao

and Yue

, The identification and characterization of novel N-glycan-based biomarkers in gastric cancer, PloS One 8 (2013), e77821.

Layke

J.C.

and Lopez

P.P.

, Gastric cancer: Diagnosis and treatment options, American Family Physician 69 (2004), 1133–40.

Pinho

S.S.

Carvalho

Marcos-Pinto

Magalhaes

Oliveira

Dinis-Ribeiro

Carneiro

Seruca

and Reis

C.A.

, Gastric cancer: Adding glycosylation to the equation, Trends Mol Med 19 (2013), 664–76.

Magalhães

and Reis

C.A.

, Helicobacter pylori adhesion to gastric epithelial cells is mediated by glycan receptors, Brazilian Journal of Medical and Biological Research 43 (2010), 611–618.

Gomes

Almeida

Ferreira

J.A.

Silva

Santos-Sousa

Pinto-de-Sousa

Santos

L.L.

Amado

Schwientek

Levery

S.B.

Mandel

Clausen

David

Reis

C.A.

and Osorio

, Glycoproteomic analysis of serum from patients with gastric precancerous lesions, J Proteome Res 12 (2013), 1454–66.

Carvalho

Catarino

T.A.

Dias

A.M.

Kato

Almeida

Hessling

Figueiredo

Gärtner

Sanches

J.M.

and Ruppert

, Preventing E-cadherin aberrant N-glycosylation at Asn-554 improves its critical function in gastric cancer, Oncogene 35 (2016), 1619.

10.

Humphrey

S.P.

and Williamson

R.T.

, A review of saliva: Normal composition, flow, and function, Journal of Prosthetic Dentistry 85 (2001), 162–169.

11.

Bonne

N.J.

and Wong

D.T.W.

, Salivary biomarker development using genomic, proteomic and metabolomic approaches, Genome Medicine 4 (2012), 82.

12.

Abrão

A.L.

Falcao

D.P.

de Amorim

R.F.

Bezerra

A.C.

Pombeiro

G.A.

Guimarães

L.J.

Fregni

Silva

L.P.

and Da

M.L.

, Salivary proteomics: A new adjuvant approach to the early diagnosis of familial juvenile systemic lupus erythematosus, Medical Hypotheses 89 (2016), 97–100.

13.

Mascarenhas

Fatela

and Barahona

, Effect of diabetes mellitus type 2 on salivary glucose – a systematic review and meta-analysis of observational studies, Plos One 9 (2014), e101706.

14.

Trapasso

Garozzo

Belfiore

and Allegra

, Evaluation of the CD44 isoform v-6 (sCD44var, v6) in the saliva of patients with laryngeal carcinoma and its prognostic role, Cancer Biomarkers 16 (2015).

15.

Arellano

Boontheung

Wang

Zhou

Jiang

Elashoff

Wei

Loo

J.A.

and Wong

D.T.

, Salivary proteomics for oral cancer biomarker discovery, Clinical Cancer Research An Official Journal of the American Association for Cancer Research 14 (2008), 6246–6252.

16.

Porto-Mascarenhas

E.C.

Assad

D.X.

Chardin

Gozal

L.C.G.

Acevedo

A.C.

and Guerra

E.N.

, Salivary biomarkers in the diagnosis of breast cancer: A review, Critical Reviews in Oncology/Hematology 110 (2017), 62.

17.

Zhang

Xiao

Zhou

Santiago

Lee

J.M.

Garon

E.B.

Yang

Brinkmann

Yan

and Akin

, Development of transcriptomic biomarker signature in human saliva to detect lung cancer, Cellular and Molecular Life Sciences Cmls 69 (2012), 3341.

18.

Farrell

J.J.

Zhang

Sugimoto

Hirayama

Soga

Zhou

Elashoff

Gao

Paster

and Tomita

, 964 Multiple salivary biomarkers for pancreatic cancer detection, Gastroenterology 136 (2009), A-147–A-147.

19.

Xiao

Zhang

Kim

J.J.

Kim

K.M.

Yoshizawa

Fan

L.Y.

Cao

C.X.

and Wong

D.T.

, Differential proteomic analysis of human saliva using tandem mass tags quantification for gastric cancer detection, Scientific Reports 6 (2016), 22165.

20.

Qin

Zhong

Zhu

Dang

Chen

Wang

Zhang

and Li

, Age- and sex-associated differences in the glycopatterns of human salivary glycoproteins and their roles against influenza a virus, J Proteome Res 12 (2013), 2742–54.

21.

Zhong

Qin

Chen

Zhang

Wang

Jia

Guo

Zhang

Shan

Wang

Xie

and Li

, Avian influenza virus infection risk in humans with chronic diseases, Sci Rep 5 (2015), 8971.

22.

Shu

Zhang

Liu

Zhang

Yang

Xie

and Li

, Salivary glycopatterns as potential biomarkers for diagnosis of gastric cancer, Oncotarget 8 (2017), 35718–35727.

23.

Yang

Cui

Wang

Sun

Wang

Chen

and Li

, Selective isolation and analysis of glycoprotein fractions and their glycomes from hepatocellular carcinoma sera, Proteomics 13 (2013), 1481–98.

24.

Song

Lasanajak

Kudelka

M.R.

Smith

D.F.

and Cummings

R.D.

, Oxidative release of natural glycans for functional glycomics, Nat Methods 13 (2016), 528–34.

25.

Qin

Zhong

Zhang

Yang

Feng

and Li

, A pilot study of salivary N-glycome in HBV-induced chronic hepatitis, cirrhosis, and hepatocellular carcinoma, Glycoconjugate Journal 34 (2017), 523–535.

26.

Tan

Yang

Guo

and Guan

, Altered N-glycan expression profile in epithelial-to-mesenchymal transition of NMuMG cells revealed by an integrated strategy using mass spectrometry and glycogene and lectin microarray analysis, Journal of Proteome Research 13 (2014), 2783–95.

27.

Zhao

Y.P.

X.Y.

Fang

Wang

You

C.H.

Zhou

P.T.

Cheng

and Gao

C.F.

, Decreased core-fucosylation contributes to malignancy in gastric cancer, PLoS One 9 (2014), e94536.

28.

Ozcan

Barkauskas

D.A.

Renee

R.L.

Torres

Cooke

C.L.

H.J.

Hua

Williams

C.C.

Dimapasoc

L.M.

and Han

K.J.

, Serum glycan signatures of gastric cancer, Cancer Prevention Research 7 (2014), 226–35.

29.

Bones

Byrne

J.C.

O’Donoghue

Mcmanus

Scaife

Boissin

Nastase

and Rudd

P.M.

, Glycomic and glycoproteomic analysis of serum from patients with stomach cancer reveals potential markers arising from host defense response mechanisms, Journal of Proteome Research 10 (2011), 1246.

30.

Fukushima

Satoh

Baba

and Yamashita

, α12-Fucosylated, and β-N-acetylgalactosaminylated prostate-specific antigen as an efficient marker of prostatic cancer, Glycobiology 20 (2010), 452–460.

31.

Yamaguchi

Yokoyama

Ebata

Matsuda

Kuno

Ikehara

Shoda

Narimatsu

and Nagino

, Verification of WFA-sialylated MUC1 as a sensitive biliary biomarker for human biliary tract cancer, Annals of Surgical Oncology 23 (2015), 1–7.

32.

Hanisch

F.G.

Ragge

Kalinski

Meyer

Kalbacher

and Hoffmann

, Human gastric TFF2 peptide contains an N-linked fucosylated N,N’-diacetyllactosediamine (LacdiNAc) oligosaccharide, Glycobiology 23 (2013), 2.

33.

Peracaula

Royle

Tabares

Mallorqui-Fernández

Barrabés

Harvey

D.J.

Dwek

R.A.

Rudd

P.M.

and De

L.R.

, Glycosylation of human pancreatic ribonuclease: Differences between normal and tumor states, Glycobiology 13 (2003), 227.

34.

Hirano

Matsuda

Shirai

and Furukawa

, Expression of LacdiNAc groups on N-glycans among human tumors is complex, Biomed Res Int 2014 (2014), 981627.

35.

Hagiwara

Tobisawa

Kaya

Kaneko

Hatakeyama

Mori

Hashimoto

Koie

Suda

and Ohyama

, Wisteria floribundaAgglutinin and its reactive-glycan-carrying prostate-specific antigen as a novel diagnostic and prognostic marker of prostate cancer, International Journal of Molecular Sciences 18 (2017), 261.

Identification of N - and O -linked glycans recognized by AAL in saliva of patients with atrophic gastritis and gastric cancer

Abstract

BACKGROUND AND AIM:

METHODS:

RESULTS:

CONCLUSIONS:

Keywords

1. Introduction

Table 1 Clinical characteristics of healthy volunteers and patients with atrophic gastritis and gastric cancer

2.1 Study approval and population

2.2 Whole saliva collection and preparation

2.3 Preparation of AAL-magnetic particle conjugates

2.4 Selective isolation of glycoprotein from human saliva by AAL-magnetic particle conjugates

2.5 Trypsin digestion of the AAL-isolated glycoproteins and PNGase F release of the N-linked glycans

2.6 NaClO release of the O-linked glycans

2.7 Purification of N -/ O -linked glycans

2.8 Characterization of the N - and O -linked glycans by MALDI-TOF/TOF-MS

3. Results

3.1 The N -linked glycan profiles of the isolated salivary glycoprotein

3.3 The O -linked glycan profiles of the isolated salivary glycoprotein

3.4 Alterations of fucosylated O-linked glycan profiles among healthy volunteers, gastric cancer, and atrophic gastritis patients

4. Discussion

Footnotes

Acknowledgments

Conflict of interest

Abbreviations

References

Table 1
Clinical characteristics of healthy volunteers and patients with atrophic gastritis and gastric cancer

2.7 Purification of $N$ -/ $O$ -linked glycans

2.8 Characterization of the $N$ - and $O$ -linked glycans by MALDI-TOF/TOF-MS

3.1 The $N$ -linked glycan profiles of the isolated salivary glycoprotein

3.3 The $O$ -linked glycan profiles of the isolated salivary glycoprotein