Clinica Chimica Acta 436 (2014) 121–129
Contents lists available at ScienceDirect
Clinica Chimica Acta journal homepage: www.elsevier.com/locate/clinchim
S100A8 as potential salivary biomarker of oral squamous cell carcinoma using nanoLC–MS/MS Yu-Jen Jou a,b, Chun-Hung Hua c, Chia-Der Lin c, Chih-Ho Lai d, Su-Hua Huang e, Ming-Hsui Tsai c, Jung-Yie Kao b,⁎, Cheng-Wen Lin a,e,⁎⁎ a
Department of Medical Laboratory Science and Biotechnology, China Medical University, Taichung, Taiwan Department of Biochemistry, College of Life Sciences, National Chung Hsing University, Taichung, Taiwan Department of Otolaryngology, China Medical University Hospital, Taichung, Taiwan d Department of Microbiology, School of Medicine, China Medical University, Taichung, Taiwan e Department of Biotechnology, College of Health Science, Asia University, Wufeng, Taichung, Taiwan b c
a r t i c l e
i n f o
Article history: Received 7 December 2013 Received in revised form 1 April 2014 Accepted 6 May 2014 Available online 24 May 2014 Keywords: Oral squamous cell carcinoma Saliva LC–MS/MS S100A8 Biomarker
a b s t r a c t Background: Oral squamous cell carcinoma (OSCC) shows low 5-year survival; early treatment greatly reduces mortality and morbidity. Saliva is a non-invasive sample, with good potential to discover biomarkers for early detection. Methods: NanoLC–MS/MS served to analyze saliva proteome from control subjects (n = 35) and OSCC patients T1 (n = 29), T2 (n = 36), T3 (n = 14) and T4 (n = 21) stages. Identified biomarkers were verified by Western blot and ELISA assays. Results: NanoLC–MS/MS analysis of salivary proteins between 10 and 15 kDa identified S100A8, hemoglobin delta and gamma-G globin in T3 and T4 stage OSCC as well as S100A7 in T1 and T2 stage OSCC. Western blot and ELISA indicated positive correlation between salivary S100A8 increment and tumor size stage. High level of S100A8 appeared in 3.4, 13.9, 92.9, and 100% of saliva OSCC patients with T1, T2, T3, and T4 stages, respectively. Significant increase of salivary S100A7 was observed in 20.7% and 11.1% of those with T1 and T2, respectively. AUROC curve indicated high sensitivity, specificity and accuracy of S100A8-based ELISA as a detector. Conclusions: NanoLC–MS/MS, Western blot and ELISA manifested salivary S100A8 as a specific and sensitive marker for detection of OSCC patients. Salivary S100A8 protein could be applicable in developing OSCC diagnostics. © 2014 Elsevier B.V. All rights reserved.
1. Introduction Oral squamous cell carcinoma (OSCC) is predominant in oral cancer, with N300,000 cases annually worldwide that account for 3% of malignancies in men and 2% in women [1–3]. Tobacco, alcohol, and betel quid are common risk factors for causing large areas of mucosal change, synergistically triggering oral carcinogenesis, even developing a secondary upper aerodigestive tract cancer [4–9]. Histological and clinical data indicate multi-step changes: leukoplakia, erythroplakia, hyperkeratosis, dysplasia, even carcinoma [10]. Despite advances in surgery, radiotherapy, and chemotherapy, overall average 5-year survival rate for patients has not improved significantly (still approximately 50%) over the past 30 years, far lower than that of laryngeal or nasopharyngeal carcinoma Abbreviations: OSCC, oral squamous cell carcinoma; NanoLC–MS/MS, automated nano liquid chromatography tandem mass spectrometry. ⁎ Corresponding author. ⁎⁎ Correspondence to: C.-W. Lin, Department of Medical Laboratory Science and Biotechnology, China Medical University, No. 91, Hsueh-Shih Road, Taichung 404, Taiwan. Tel.: +886 4 22053366x7210; fax: +886 4 22057414. E-mail address:
[email protected] (C.-W. Lin).
http://dx.doi.org/10.1016/j.cca.2014.05.009 0009-8981/© 2014 Elsevier B.V. All rights reserved.
[11,12]. Appropriate treatment for those with pre-malignant oral lesions proves more effective, significantly raising the survival rate to 80–90% [12]. Consequently, early diagnosis of oral cancer, distinguishing between malignant or premalignant lesions, is crucial to reduce the mortality and morbidity. Biopsy of suspicious lesions offers the gold specimens for the discovery of molecular biomarkers, but non-uniform appearance of cancerous and precancerous lesions allows the difficulty in choosing the location of biopsy, affecting the accuracy of potential OSCC markers [13,14]. Developing credible, accurate, cost-effective, and noninvasive techniques for early detection is essential. S100 A1-14 and B, a group of small acidic proteins, contain EF-hand calcium-binding motifs [15], modulating multiple biological properties in distinct cell- and tissue-types via binding with Ca2 +, Zn2 +, and Cu2+. S100 proteins involve in calcium homeostasis and cytoskeletal dynamics, as well as regulate cell proliferation and transcriptional factor activity [16]. S100A4 protein regulates myosin dynamics by inhibiting protein kinase C (PKC)-mediated phosphorylation on C-terminus of myosin heavy chain [17,18]. Secreted S100A4 is a candidate maker predicting metastatic and prognostic potential in breast cancer [19]. S100A7 is up-regulated in inflammatory epidermis, correlating with
122
Y.-J. Jou et al. / Clinica Chimica Acta 436 (2014) 121–129
epithelial malignancies: e.g., breast, skin, esophagus, head and neck [20]. S100A8 and S100A9 can be synthesized and secreted by granulocytes, monocytes, and macrophages, identified as cytokine-like and transcriptional factor-like molecules affecting expression of tumor necrosis factor-α, interleukin-1, and matrix metalloproteinases [21–24]. Upregulation of S100A8 and S100A9 is found in gastric, colorectal, breast, lung, pancreatic, and prostate cancer, correlating with inflammation cell proliferation and metastatic processes in tumor development [25–27]. S100B also inhibits PKC-mediated phosphorylation on p53, reducing tumor suppressor activity by suppressing p53-dependent transcription activation [28,29]. Altered expression of S100 proteins is thereby associated with cancer development; secreted form of S100 proteins could act as potential cancer markers. Tissue microarray indicated significant up-expression of S100A8 in severe oral dysplasias and OSCC tissue [30,31]. Proteomic analysis of normal and OSCC tissues suggested S100A7 as a positive marker for OSCC carcinogenesis and early tumor progression that can be confirmed by immunofluorescence and quantitative RT-PCR analyses [32]. Sharp decrease of S100A4 mRNA was evident in OSCC tissues [33]. Earlier we identified S100A8, transferrin, and zinc finger protein 497 as salivary biomarkers, using two-dimensional gel electrophoresis (2DE) and mass spectrometry (MS) [30]. Salivary samples from such patients showed elevated S100A8, necessary for further probe or correlation with oral cancer tumor grade. This study rated the potential of S100 proteins as salivary markers via proteomic analysis of low molecular weight salivary proteins, using nanoLC–MS/MS. Protein profile indicated change of S100A7 and S100A8 as unique markers in saliva of oral cancer patients. Western blot and direct binding ELISA further examined levels of S100A7 and S100A8 in their saliva while evaluating potency of these markers.
DMR96-IRB-80). Exclusion criteria for OSCC patients and control individuals were followed, as in our prior studies [30,34]. Table 1 lists personal information (age, gender, and clinical features) of subjects. Tumor size (T) and nodal metastasis (N) staging for OSCC patients, as verified by pathological examination, was reviewed according to the universal TNM staging system of the International Union against Cancer (UICC) [35]. Collection protocol of salivary samples from patients and controls were performed as in prior studies [30,34]: 5 ml collected in 15-ml centrifuge tube, mixed 5 μl Complete™ Protease inhibitor Cocktail (Roche), then centrifuged at 12,000 rpm (~13,400 × g) for 10 min at 4 °C. The resulting supernatants were stored at −80 °C. 2.2. SDS-PAGE and Western blot Preparation of each 500 μl sample for Western blot was performed according to the protocol of the 2-D clean up kit (Amersham). Salivary proteins were precipitated by a combination of precipitant and coprecipitant, then rehydrated in 100 μl of rehydration buffer (8 mol/l urea, 4% CHAPS, 0.002% bromophenol blue). Afterward, 5 μg of salivary proteins from each sample was mixed with sample buffer, heated at 100 °C for 8 min, then loaded onto 12% SDS-PAGE gels stained with Coomassie blue after running. For Western blot, the dissolved proteins in gels were transferred to nitrocellulose membranes. The resultant nitrocellulose membrane was blocked with 5% skim milk in Tris buffered saline (TBS) buffer containing 0.1% Tween 20 (TBST) at 4 °C for 2 h, reacted with primary monoclonal antibodies anti-S100A7 and -S100A8 (Abnova) overnight, then incubated with HRP-conjugated anti-mouse IgG antibodies (Invitrogen). Also, human sIgA as control was detected in salivary samples. Immune-reactive bands of interest were detected by ECL™ Western Blotting Detection Reagents (Amersham).
2. Materials and methods 2.3. In-gel digestion of low molecular weight salivary proteins 2.1. Human subjects and saliva collection In all, 35 subjects without and 100 with OSCC were enrolled for study from February 2007 to March 2014; OSCC patients were diagnosed via biopsy at China Medical University Hospital in Taichung. Subjects gave informed consent prior to saliva collection approved by the Institutional Review Board of China Medical University Hospital (permission number
Since S100 were low molecular weight proteins, salivary protein bands between approximate 10–15 kDa solved by 12% SDS-PAGE were excised, washed twice with buffer (25 mM ammonia bicarbonate (ABC) in 50% acetonitrile (ACN)) for 5 min and 100% ACN for 5 min, then dried by speed vacuum concentrator. Proteins embedded in gels were reduced by reduction buffer (10 mmol/l dithiothreitol (DTT), 25 mM
Table 1 Clinicopathological features of controls oral cancer patients.
Gender Male Female Mean age (year) Male Female Cancer sites Oral cavity Mouth Oropharynx Hypopharynx Laryngeal Node stage N0 N1 N2 Histology⁎ WD MD PD Keratinizing
Controls
OSCC patients (100)
(35)
T1 (29)
T2 (36)
T3 (14)
T4 (21)
27 8 50.5 ± 13.6 50.2 ± 14.5 49.6 ± 10.3
26 3 51.2 ± 12.7 50.8 ± 13.1 54.7 ± 10.8
36 0 52.9 ± 11.0 52.9 ± 11.0
13 1 56.9 ± 10.3 56.9 ± 10.7 57.0
21 0 52.1 ± 11.1 52.1 ± 11.1
27 1 1 0 0
30 5 0 1 0
11 0 0 2 1
16 0 0 4 1
25 1 3
17 8 11
8 2 4
8 2 11
10 16 2 1
7 25 4 0
4 9 1 0
9 9 3 0
⁎ WD, Well-differentiated; MD, Moderately differentiated; PD, Poorly-differentiated.
Y.-J. Jou et al. / Clinica Chimica Acta 436 (2014) 121–129
A
kDa
123
subjects
15
Control group 10
15
T1 group 10
15
T2 group 10
15
T3 group 10
15
T4 group 10
B
y11
y10 y9
y8
y7
y6
y5
y4
y3
y2
y1
K G T N Y L AD V F E K b1
b2
b3
b4
b5
b6
b7
b8
b9 b10 b11
Fig. 1. Peptide mass fingerprinting and peptide sequencing of salivary protein using nanoLC–MS/MS. Salivary samples from controls and OSCC patients were dissolved in 12% SDS PAGE (A), protein bands between 10 and 15 kDa subjected to nanoLC–MS/MS for identification. Representative LC–MS/MS spectra of S100A7 (B), S100A8 (C), and S100A9 (D) are shown with the calculated molecular weight (m/z values) along x-axis and relative intensity along y-axis. Amino acid sequence (Upper panel) was determined from mass differences in y- and bfragment ions series.
124
Y.-J. Jou et al. / Clinica Chimica Acta 436 (2014) 121–129
y10
C
y9
y8 y7 y6
y5
y4
y3
y2
y1
A L N S I I D V Y N K b1
b2
y14 y13 y12 y11 y10 y9
D
b3
y8
b4 b5 b6 b7
y7
y6
y5
b8
b9
y4 y3
b10
y2
y1
N I E T I I N T F H Q Y S VK b1
b2 b3 b4
b5 b6 b7
b8
b9 b10
b11 b12 b13 b14
Fig. 1 (continued).
ABC) at 56 °C for 15 min, alkylated by solution (55 mmol/l iodoacetamide (IAA), 25 mM ABC) at room temperature for 20 min in the dark, and digested with fresh trypsin solution (2 ng/μl trypsin,
25 mmol/l ABC) at 37 °C overnight. Peptides were extracted by sonication with 50–100% ACN and 0.1% formic acid (FA); released peptides in supernatants were dried by speed vacuum concentrator.
Y.-J. Jou et al. / Clinica Chimica Acta 436 (2014) 121–129
125
Table 2 NanoLC–MS/MS identification of low molecular weight salivary proteins in samples from controls and OSCC patients at T1, T2, T3 and T4 stages. Accession no.
gi|235948 gi|183817 gi|4505821 gi|4506773 gi|190668 gi|229172 gi|31725 gi|957232
Protein name
Cystatin SA-III = potential precursor of acquired enamel pellicle Beta-globin Prolactin-inducible protein precursor Protein S100-A9 S100-A7 (psoriasin) Hemoglobin delta Gamma-G globin S100-A8 (oncodevelopmental protein)
mol. weight⁎
Control Score
Peptides matched
T1
14,181
87
5
15,870 16,562 13,234 11,450 15,897 16,959 13,086
80 72 71
3 7 12
Score
104 133 145 165
T2 Peptides matched
3 5 8 8
T3
Score
Peptides matched
162
6
112 249 76
6 9 3
Score
T4 Peptides matched
65 50
7 9
123 58 57
5 3 2
Score
Peptides matched
93
4
103 171 134
7 5 5
197
7
110
5
⁎ mol. weight refers to molecular weight given in database, as inferred from gene sequence.
2.4. NanoLC–MS/MS analysis of digested salivary proteins Eluted peptides were subjected to nanoLC–MS/MS analysis, an integrated system (QSTAR XL) comprising LC Packings NanoLC system with autosampler and QSTAR XL Q-Tof mass spectrometer (AB Sciex) fitted with nano-LC sprayer. The eluted samples were desalted on a LCPackings PepMap™ C18 μ-Precolumn™ Cartidge (5 μm, 30 μm I.D. × 5 mm; Dionex, Sunnyvale, CA), separated on an LC-Packings PepMap C18 column (3 μm, 15 cm × 75 μm i.D.,) at 200 nl/min, using 45 min gradient of 5–60% ACN in 0.1% FA and analyzed by connecting inline to a mass spectrometer. NanoESI–MS and CID MS/MS sequencing of peptides were fully automated, synchronized with nanoLC runs under AnalystQS software. To identify proteins, 1 s survey scans were set at mass range m/z 400–1600 and 6 s MS/MS spectral acquisitions of multiple-charged precursor ions were detected at intensity above predefined threshold. Acquired individual MS/MS spectra within a single LC run were combined as a single Mascot-searchable peak list file. Peak list files served to query Swiss-Prot database via Mascot program with these parameters: peptide mass tolerance of 150 ppm, MS/MS ion mass tolerance of 0.15 Da, and allowing up to one missed cleavage. Minimum score above 20 was randomly set as acceptance threshold. 2.5. Relative quantitative analysis of salivary markers by ELISA Each sample containing 5 μg/ml of protein was quadruplicate procoated in 96-well plates at 4 °C overnight, unbound substrates washed out with TBST thrice. After blocking with 5% skim milk in TBST, relative amount of salivary markers was detected by monoclonal antibodies anti-S100A7 and anti-S100A8. Anti-mouse IgG conjugated HRP was
added into each well for 2-h incubation at room temperature, immune-reactive complexes measured with ABTS/H2O substrates. Optical absorbance was recorded at 405 nm by ELISA plate reader (ELX 808, BioTek).
2.6. Immunohistochemical staining of S100A7 and S100A8 in OSCC tissues Tissue sections from control, T1, and T3 cases underwent immunohistochemical staining. Sections were deparaffined and rehydrated. After blocking with 1% normal goat serum, diluted monoclonal antibodies anti-S100A7 or anti-S100A8 were added to tissue sections and incubated for 1 h, followed by HRP-conjugated anti-mouse IgG antibodies and then 3,3′-diaminobenzidine (Sigma-Aldrich) as substrate.
3. Results 3.1. Clinical parameters of OSCC patients and control subjects A total of 35 controls and 100 OSCC cases were recruited, the latter grouped as T1, T2, T3 and T4, based on tumor size stage of UICC TNM staging system (Table 1). Males formed a majority in all groups; mean age of OSCC cases was over 50 years, with controls (50.2 years) slightly younger. Among OSCC cases, tongue and buccal sites showed most frequent oral cancer lesions; nearly half were N0 stage without tumor cells from regional lymph nodes, 13% N1 and 29% N2 stage. Histologic examination of lesions showed moderately differentiated OSCC in 59% and well differentiated OSCC in 30% of the cases. Table 1 lists other clinical parameters of patients and controls.
OSCC
S100A7 S100A8 sIgA Fig. 2. Western blot analysis of S100A7 and S100A8 in saliva from controls and OSCC patients. Salivary samples from each group were analyzed by 12% SDS-PAGE, then electrophoretically transferred onto nitrocellulose membrane probed with monoclonal antibodies to S100A7, S100A8 and sIgA, and developed with HRP-conjugated secondary antibody and chemiluminescent HRP substrates. Lane 1: control group; Lane 2: T1 stage; Lane 3: T2 stage; Lane 4: T3 stage; Lane 5: T4 stage.
126
Y.-J. Jou et al. / Clinica Chimica Acta 436 (2014) 121–129
3.2. NanoLC–MS/MS analysis of low molecular weight salivary proteins To examine potential of salivary S100 as OSCC markers, samples from control subjects and patients were analyzed by 12% SDS-PAGE, protein bands between 10 and 15 kDa excised and identified by NanoLC–MS/MS analysis system (Fig. 1). Table 2 details proteins from in-gel trypsin digestion of low molecular weight salivary proteins in samples of each group by LC–MS/MS. Fig. 1B–D show peptide mass fingerprinting and peptide sequencing of S100A7, S100A8 and S100A9. Comparison of protein profile among groups indicated S100A7 as a unique marker of T1 and T2 stages and S100A8 as a potential marker of T3 and T4 stage in OSCC patients (Table 2). S100A9 appeared in salivary samples from control and all patient groups. Western blot analysis indicated significant rise of S100A8 in salivary samples from T3 and T4 groups, S100A7 elevated in T1 but falling dramatically in T3 and T4 stages (Fig. 2). 3.3. Relative levels of S100A7 and S100A8 in saliva from OSCC cases and controls To correlate between S100A7/A8 and OSCC stage, relative levels of S100A7 and S100A8 in the saliva of OSCC patients were measured via
A
1
S100A7
0.9
Table 3 Number of positive ELLISA reaction of S100A7 and S100A8 in salivary samples from controls and OSCC patients with T1, T2, T3 and T4 stages. S100 family⁎
S100A7 S100A8 S100A7/S100A8#
Controls
OSCC patients (100)
(35)
T1 (29)
T2 (36)
T3 (14)
T4 (21)
0 1†(2.9%) 1 (2.9%)
6 (20.7%) 1 (3.4%) 7 (24.1%)
4 (11.1%) 5 (13.9%) 9 (25%)
0 13 (92.9%) 13 (92.9%)
0 21 (100.0%) 21 (100.0%)
⁎ Cut-off value set at OD405nm of 0.3. † Case diagnosed as OSCC patient 1 year post sampling. # A positive from either one or both markers.
binding ELISA, compared to those in the saliva of controls (Fig. 3, Supplemental Figs. 1 and 2). Binding ELISA indicated the significant change for S100A7 and S100A8 in tumor size stage (Fig. 3), but no meaningful differences in lymph node status and age (Supplemental Figs. 1 and 2). The mean of OD405 nm indicated that S100A8 is significantly higher in the saliva of patients with T2-4 stages compared to T1 and control subjects (Fig. 3B). By contrast, salivary S100A7 peaked at T1 stage and reached its nadir in T3 and T4 stages (Fig. 3A). When cutoff OD index of 0.3 was recommended, positive cases of S100A8 were 1 (2.9%) in controls, 1 (3.4%) in T1, 5 (13.9%) in T2, 13 (92.9%) in T3, and 21 (100%) in T4 (Table 3). Meanwhile, S100A7-positive cases tallied 6 (20.7%) in T1, 4 (11.1%) in T2, but negative in other groups. Combination of the
0.8
OD 405nm
0.7
A
0.6 0.5 0.4
0.8
Sensitivity
0.3 0.2 0.1 0
Control
T1
T2
T3
T4
B
0.6
0
1
S100A8
0.9 0.8
T1 T2
0.4 0.2
Stage
0
0.2
0.4
0.6
0.8
1
1-Specificity
S100A8
B
0.7
1
0.6 0.8
0.5 0.4
Sensitivity
OD 405nm
S100A7
1
0.3 0.2
0.6 T3 T4
0.4
0.1 0
0.2 Control
T1
T2
T3
T4
Stage Fig. 3. Direct binding ELISA to salivary S100A7 and S100A8 proteins. 2.5 μg of salivary protein from each individual of control, T1, T2, T3, and T4 groups were pre-coated onto microtiter plates, then incubated with 2000-fold dilution of monoclonal antibodies anti-S100A7 or anti-S100A8 at room temperature for 1 h. Immune complexes were analyzed by ELISA procedures.
0
0
0.2
0.4
0.6
0.8
1
1-Specificity Fig. 4. Receiver-operating characteristic curves of salivary biomarkers for predicting tumor size stages. Salivary samples were analyzed via direct binding ELISA of S100A7 (A) and S100A8 (B).
Y.-J. Jou et al. / Clinica Chimica Acta 436 (2014) 121–129
positivity from either one or both markers significantly improved the sensitivity in T1 and T2 stages. 3.4. ROC curve analysis of S100A8 as the potential salivary marker of OSCC To rate sensitivity and specificity of S100A8 for OSCC detection, the area under receiver-operating characteristic curves (AUROC) discriminates between OSCC-free subjects and OSCC cases at each stage, using ELISA (Fig. 4). AUROC of S100A7 for predicting OSCC was 0.71 for T1 (95% CI: 0.88–1.04), and 0.68 for T2 (95% CI: 0.89–1.01) (Fig. 4A), that of salivary S100A8 0.99 for T3 (95% CI: 0.58–1.00), and 0.98 for T4 (95% CI: 0.63–1.06), respectively (Fig. 4B). This indicated S100A8based ELISA as highly accurate in detecting OSCC at T3 and T4 stages. 3.5. Overexpression of S100A8 in OSCC tissues Immunohistochemical staining further validated levels of S100A7 and S100A8 in oral tissues from OSCC and control groups (Fig. 5). Upregulation of S100A7 was observed in OSCC tissues from T1, but not control and T3 groups. By contrast, strong positive reaction for S100A8 was observed in tissues from the patients, not the controls. The results indicated that S100A8 is moderately expressing in T1 and overexpressing in T3 stage tissue, in agreement with Western blot and ELISA analyses of the saliva from OSCC patients and controls. 4. Discussion This study was the first report that nanoLC–MS/MS utilized to identify salivary protein filing of OSCC patients and controls. NanoLC–MS/ MS served as powerful tools to discover potential serum, urine, and tissue biomarkers for colon, lung, breast, colorectal, and gastric cancers [36–39]. In the present study, nanoLC–MS/MS analysis of salivary proteins between 10 and 15 kDa identified S100A8 as a potential salivary
Control
127
biomarker for oral cancer (Fig. 1 and Table 2), in accordance with our prior report: up-regulation of S100A8 in OSCC patients' saliva identified by two-dimensional gel electrophoresis (2DE) and matrix-assisted laser desorption/ionization time-of-flight mass spectrometry (MALDI-TOF MS) analyses [30]. Western blot showed that salivary S100A8 significantly elevated in patients compared to controls (Fig. 2). Further ELISA evaluation indicated the mean amount of S100A8 in the saliva of OSCC patients as higher than in healthy controls (Fig. 3), while AUROC of S100A8-based ELISA for the diagnosis of OSCC revealed that salivary S100A8 serves as OSCC biomarker (Fig. 4). The results indicated high sensitivity, specificity, and accuracy of S100A8 in detecting OSCC. In the healthy control group, one case (2.9%) with high level of salivary S100A8 was followed up and diagnosed as OSCC 1 year post sampling. Therefore, S100A8 proves valuable as the target of OSCC diagnostics. Comparison of salivary protein profiling between OSCC and control indicated increment of S100A8 and hemoglobin delta in T3 and T4 as well as that of S100A7 in T1 and T2 groups. S100A8, a calciumbinding and pro-inflammatory protein, was overexpressed in prostate, breast, lung, gastric, pancreatic and colorectal cancers [40], hence suggested as one critical tissue marker for aggressive breast cancer phenotype [41], discriminatory ovarian cyst fluid marker for ovarian tumors [42], and serological marker for colorectal and gastric cancers [43,44]. In this study, quantity of S100A8 in salivary samples was validated by direct binding ELISA (Fig. 3). NanoLC–MS/MS and ELISA analyses revealed positive correlation between tumor size stage of OSCC and salivary S100A8 protein level. Comparison of salivary S100A8 with the reported candidate marker transferrin [30] indicated that S100A8 had higher specificity but lower sensitivity in detecting T1 and T2 stage OSCC than transferrin. Salivary transferrin elevated in patients with parotid mixed tumor, alcoholic liver cirrhosis, nasal polyposis and bronchial asthma [30]. The combination of salivary S100A8 and transferrin could rule out the false positive and confirm the diagnosis of OSCC. Meanwhile, AUROC curve indicated that specificity, sensitivity, and
T1 stage
T3 stage
S100A7
S100A8
Fig. 5. Immunohistochemical analysis of S100A7 and S100A8 expression in OSCC. Tissue sections from healthy controls and OSCC patients with T1 or T3 stage, analyzed by immunohistochemical staining with monoclonal antibodies anti-S100A7 and anti-S100A8. Immunoreactivity complex was developed by HRP-conjugated anti-mouse IgG antibodies and 3,3′-diaminobenzidine as substrate, figures photographed under ×100 magnification.
128
Y.-J. Jou et al. / Clinica Chimica Acta 436 (2014) 121–129
accuracy of salivary S100A8 in detecting oral cancer were similar to those of the reported candidate OSCC markers [13,14]. Immunohistochemical staining also demonstrated that S100A8 moderately upregulated in OSCC tissues from T1 and overexpressed in T3 stage (Fig. 5). Immunohistochemical staining implied increment of salivary S100A8 derived from secreted proteins of OSCC tissues in OSCC patients. Previous reports demonstrated S100A7 overexpressing in tissues of oral lesions [45] as well as regulating invasion and growth of breast and prostate cancers [46,47]. Our study showed low increment of salivary S100A7 in OSCC cases at T1 stage as well as low expression of it in T1 stage tissues (Figs. 3 and 5). Also, hemoglobin delta and gamma globin were extracted from T3- and/or T4-stage patients' saliva (Table 2). Embryonic and fetal hemoglobin were expressed in human glioblastoma multiform cells [48]. The role of S100A7, hemoglobin delta, and gamma globin in OSCC diagnosis merits further examination. Salivary biomarkers have great potential for detection and surveillance of OSCC progression or recurrence. This study explored the profiling of low-weight salivary proteins between 10 and 15 kDa, using nanoLC–MS/MS; elucidating whole saliva proteome can identify potential biomarkers for detection or progression of OSCC as well as pathophysiology. In summary, nanoLC–MS/MS analysis of the saliva from OSCC patients and controls identified S100A8 as a potential salivary biomarker for the diagnosis of human oral cancer in humans. Salivary protein level of S100A8 in OSCC patients and controls was confirmed by Western blot and ELISA. In addition, salivary S100A8 levels strongly correlated with tumor size stage. AUROC curve exhibited high specificity, sensitivity, and accuracy of S100A8-based ELISA for detecting oral cancer.
Acknowledgments This work was supported by the National Science Council of Taiwan (NSC101-2320-B-039-036-MY3) and China Medical University (CMU101-ASIA-05 and CMU101-S-24).
Appendix A. Supplementary data Supplementary data to this article can be found online at http://dx. doi.org/10.1016/j.cca.2014.05.009.
References [1] Greenlee RT, Murray T, Bolden S, Phyllis A. Cancer statistics, 2000. CA Cancer J Clin 2000;50:7–33. [2] Parkin DM, Bray F, Ferlay J, Pisani P. Global cancer statistics, 2002. CA Cancer J Clin 2005;55:74–108. [3] Cancer facts and figures 2006. Atlanta: American Cancer Society; 2006. [4] Ho PS, Yang YH, Shieh YT, et al. Consumption of areca quid, cigarettes, and alcohol related to the comorbidity of oral submucous fibrosis and oral cancer. Oral Surg Oral Med Oral Pathol Oral Radiol Endod 2007;104:647–52. [5] Zhang X, Reichart PA. A review of betel quid chewing, oral cancer and precancer in Mainland China. Oral Oncol 2007;43:424–30. [6] Haughey BH, Gates GA, Arfken CL, Harvey J. Meta-analysis of second malignant tumors in head and neck cancer: the case for an endoscopic screening protocol. Ann Otol Rhinol Laryngol 1992;101:105–12. [7] Jones AS, Morar P, Phillips DE, Field JK, Husband D, Helliwell TR. Second primary tumors in patients with head and neck squamous cell carcinoma. Cancer 1995;75:1343–53. [8] Jovanovic A, van der Toll G, Kostense JP, et al. Second respiratory and upper digestive tract cancer following oral squamous cell carcinoma. Eur J Cancer B Oral Oncol 1994;30B:225–9. [9] Tepperman BS, Fitzpatrick PJ. Second respiratory and upper digestive tract cancers after oral cancer. Lancet 1981;2:547–9. [10] Tsantoulis PK, Kastrinakis NG, Tourvas AD, Laskaris G, Gorgoulis VG. Advances in the biology of oral cancer. Oral Oncol 2007;43:523–34. [11] Canto MT, Devesa SS. Oral cavity and pharynx cancer incidence rates in the United States, 1975–1998. Oral Oncol 2002;38:610–7. [12] Silverman SJ. Demographics and occurrence of oral and pharyngeal cancers. The outcomes, the trends, the challenge. J Am Dent Assoc 2001;132:7S–11S.
[13] Yen CY, Huang CY, Hou MF, et al. Evaluating the performance of fibronectin 1 (FN1), integrin α4β1 (ITGA4), syndecan-2 (SDC2), and glycoprotein CD44 as the potential biomarkers of oral squamous cell carcinoma (OSCC). Biomarkers 2013:63–72. [14] Yen CY, Chen CH, Chang CH, et al. Matrix metalloproteinases (MMP) 1 and MMP10 but not MMP12 are potential oral cancer markers. Biomarkers 2009;14:244–9. [15] Hsieh HL, Schäfer BW, Sasaki N, Heizmann CW. Expression analysis of S100 proteins and RAGE in human tumors using tissue microarrays. Biochem Biophys Res Commun 2003;307:375–81. [16] Salama I, Malone PS, Mihaimeed F, Jones JL. A review of the S100 proteins in cancer. EJSO 2008;34:357–64. [17] Kriajevska M, Tarabykina S, Bronstein I, et al. Metastasis-associated Mts1 (S100A4) protein modulates protein kinase C phosphorylation of the heavy chain of nonmuscle myosin. J Biol Chem 1998;273:9852–6. [18] Murakami N, Kotula L, Hwang YW. Two distinct mechanisms for regulation of nonmuscle myosin assembly via the heavy chain: phosphorylation for MIIB and mts 1 binding for MIIA. Biochemistry 2000;39:11441–51. [19] Rudland PS, Platt-Higgins A, Renshaw C, et al. Prognostic significance of the metastasis-inducing protein S100A4 (p9Ka) in human breast cancer. Cancer Res 2000;60:1595–603. [20] Sneh A, Deol YS, Ganju A, et al. Differential role of psoriasin (S100A7) in estrogen receptor α positive and negative breast cancer cells occur through actin remodeling. Breast Cancer Res Treat 2013;138:727–39. [21] Odink K, Cerletti N, Brüggen J, et al. Two calcium-binding proteins in infiltrate macrophages of rheumatoid arthritis. Nature 1987;330:80–2. [22] Roth J, Teigelkamp S, Wilke M, Grün L, Tümmler B, Sorg C. Complex pattern of the myelo-monocytic differentiation antigens MRP8 and MRP14 during chronic airway inflammation. Immunobiology 1992;186:304–14. [23] van Lent PL, Grevers LC, Blom AB, et al. Stimulation of chondrocyte-mediated cartilage destruction by S100A8 in experimental murine arthritis. Arthritis Rheum 2008;58:3776–87. [24] Youssef P, Roth J, Frosch M, et al. Expression of myeloid related proteins (MRP) 8 and 14 and the MRP8/14 heterodimer in rheumatoid arthritis synovial membrane. J Rheumatol 1999;26:2523–8. [25] Koon N, Zaika A, Moskaluk CA, et al. Clustering of molecular alterations in gastroesophageal carcinomas. Neoplasia 2004;6:143–9. [26] Sy SM, Wong N, Lai PB, To KF, Johnson PJ. Regional over-representations on chromosomes 1q, 3q and 7q in the progression of hepatitis B virus-related hepatocellular carcinoma. Mod Pathol 2005;18:686–92. [27] Moinzadeh P, Breuhahn K, Stützer H, Schirmacher P. Chromosome alterations in human hepatocellular carcinomas correlate with aetiology and histological grade-results of an explorative CGH meta-analysis. Br J Cancer 2005;92:935–41. [28] Grigorian M, Andresen S, Tulchinsky E, et al. Tumor suppressor p53 protein is a new target for the metastasis-associated Mts1/S100A4 protein: functional consequences of their interaction. J Biol Chem 2001;276:22699–708. [29] Lin J, Blake M, Tang C, et al. Inhibition of p53 transcriptional activity by the S100B calcium-binding protein. J Biol Chem 2001;276:35037–41. [30] Jou YJ, Lin CD, Lai CH, et al. Proteomic identification of salivary transferrin as a biomarker for early detection of oral cancer. Anal Chim Acta 2010;681:41–8. [31] Driemel O, Escher N, Ernst G, Melle C, von Eggeling F. S100A8 cellular distribution in normal epithelium, hyperplasia, dysplasia and squamous cell carcinoma and its concentration in serum. Anal Quant Cytol Histol 2010;32:219–24. [32] Kesting MR, Sudhoff H, Hasler RJ, et al. Psoriasin (S100A7) up-regulation in oral squamous cell carcinoma and its relation to clinicopathologic features. Oral Oncol 2009;45:731–6. [33] Sapkota D, Bruland O, Bøe OE, et al. Expression profile of the S100 gene family members in oral squamous cell carcinomas. J Oral Pathol Med 2008;37:607–15. [34] Jou YJ, Lin CD, Lai CH, et al. Salivary zinc finger protein 510 peptide as a novel biomarker for detection of oral squamous cell carcinoma in early stages. Clin Chim Acta 2011;412:1357–65. [35] Sobin LH, Wittekind CH. TNM classification of malignant tumors. New York: Wiley; 2002. [36] Li Y, Wen T, Zhu M, et al. Glycoproteomic analysis of tissues from patients with colon cancer using lectin microarrays and nanoLC–MS/MS. Mol Biosyst 2013;9:1877–87. [37] Schaaij-Visser TB, de Wit M, Lam SW, Jiménez CR. The cancer secretome, current status and opportunities in the lung, breast and colorectal cancer context. Biochim Biophys Acta 1834;2013:2242–58. [38] Opstal-van Winden AW, Krop EJ, Kåredal MH, et al. Searching for early breast cancer biomarkers by serum protein profiling of pre-diagnostic serum; a nested case–control study. BMC Cancer 2011;11:381. [39] Zhang Z, Li M, Zhang G, et al. Identification of human gastric cacinoma biomarkers by differential protein expression analysis using 18O labeling and nanoLC–MS/MS coupled with laser capture microdissection. Med Oncol 2010;27:296–303. [40] Grebhardt S, Müller-Decker K, Bestvater F, Hershfinkel M, Mayer D. Impact of S100A8/A9 expression on prostate cancer progression in vitro and in vivo. J Cell Physiol 2013. http://dx.doi.org/10.1002/jcp.24489. [41] Parris TZ, Kovács A, Aziz L, et al. Additive effect of the AZGP1, PIP, S100A8, and UBE2C molecular biomarkers improves outcome prediction in breast carcinoma. Int J Cancer 2013. http://dx.doi.org/10.1002/ijc.28497. [42] Skaggs HS, Saunders BA, Miller RW, et al. Ovarian cyst fluids are a cache of tumor biomarkers that include calgranulin A and calgranulin B isoforms. Cancer Invest 2013;31:433–53. [43] Kim HJ, Kang HJ, Lee H, et al. Identification of S100A8 and S100A9 as serological markers for colorectal cancer. J Proteome Res 2008;8:1368–79. [44] Wang L, Chang EW, Wong SC, Ong SM, Chong DQ, Ling KL. Increased myeloid-derived suppressor cells in gastric cancer correlate with cancer
Y.-J. Jou et al. / Clinica Chimica Acta 436 (2014) 121–129 stage and plasma S100A8/A9 proinflammatory proteins. J Immunol 2013;190:794–804. [45] Kaur J, Matta A, Kak I, et al. S100A7 overexpression is a predictive marker for high risk of malignant transformation in oral dysplasia. Int J Cancer 2013. http:// dx.doi.org/10.1002/ijc.28473. [46] Liu H, Wang L, Wang X, Cao Z, Yang Q, Zhang K. S100A7 enhances invasion of human breast cancer MDA-MB-468 cells through activation of nuclear factor-κB signaling. World J Surg Oncol 2013;11:93.
129
[47] Ye L, Sun PH, Martin TA, Sanders AJ, Mason MD, Jiang WG. Psoriasin (S100A7) is a positive regulator of survival and invasion of prostate cancer cells. Urol Oncol 2013;31:1576–83. [48] Emara M, Turner AR, Allalunis-Turner J. Adult, embryonic and fetal hemoglobin are expressed in human glioblastoma cells. Int J Oncol 2013. http://dx.doi.org/10.3892/ ijo.2013.2186.