miR-27b-Regulated TCTP as a Novel Plasma Biomarker for Oral Cancer: From Quantitative Proteomics to Post-Transcriptional Study
Wan-Yu Lo1,2,3, Huang-Joe Wang4,5, Chih-Wei Chiu6, Sung-Fang Chen6* 1 Graduate Institute of Integrated Medicine, China Medical University, Taichung, Taiwan
2. .Division of Surgery, Department of Medical Research, China Medical University Hospital, Taichung, Taiwan 3. Department of life science, National Chung Hsing University, Taichung, Taiwan
4. Department of Internal Medicine, School of Medicine, China Medical University, Taichung, Taiwan 5. Division of Cardiology, Department of Medicine, China Medical University Hospital, Taichung, Taiwan 6. Department of Chemistry, National Taiwan Normal University, Taipei, Taiwan
*Correspondence: Dr. Sung-Fang Chen, Department of Chemistry , National Taiwan Normal University, No. 88, Sec. 4, Ting-Chow Rd, Taipei, Taiwan, 11677 Taipei, Taiwan
Phone: 886-2-77346210 Fax: 886-2-9324249 Email: [email protected]
Keywords: Quantitative proteomic analysis, iTRAQ, oral cancer, translationally controlled tumor protein, TCTP, miR-27b 1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20
Abstract
We combined an iTRAQ-based quantitative proteomic analysis and the miRNA determination to profile potentially novel biomarker from oral cancer. There are 757 and 674 unique proteins were identified from proteomic analysis, and 13 proteins displayed consistent underexpression (<0.67 fold) in normal tissues in comparison with the corresponding tumor tissues. After the preliminary screening, the EGFR, OAT, TPT1, ITGA6, G3BP1 and CB39L were the six genes validated in the 37 oral cancer patients (T1, n=10; T2, n=10; T3, n=10 and T4, n=7). The TPT1, ITGA6 and CAB39L genes were displayed the higher transcriptions level in the tumor tissues and the TPT1, ITGA6 and CAB39L proteins were also shown overexpression in the tumor tissues from the same patients. The miR-19a, 19b, 27a, 27b, 186, 203 and 377 transcripts were predicted and the miRNA-27a and 27b level was shown significantly reduction in the tumor tissues and the plasma of OSCC patients. In the in vitro study, the overexpression of miR-27b significantly only decreased TCTP protein and gene levels in both of HSC-3 and Cal-27 cell lines. Our results demonstrate that human miR-27b regulates the expression of the TCTP tumor protein, and circulating miR-27b may be useful as a biomarker for oral cancer research. 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35
Introduction
An estimated 263,900 new cases and 128,000 deaths from oral cavity cancer (including lip cancer) occur annually worldwide [1]. Smoking, betel-quid and areca-nut chewing, alcohol use, smokeless tobacco products and Human Papillomavirus (HPV)infection are the major risk factors for oral cancer. Various treatment options including surgery, radiotherapy and chemotherapy are available for oral cancer. However, the five-year survival rate of oral cancer is one of the lowest among common malignant neoplasms [2]. In Taiwan, an endemic betel quid chewing area, a significantly increasing trend in oral cancer has been observed in males [3]. In 2006, oral cancer became the 6th most common cancer in Taiwan and the 4th most common cause of cancer deaths in Taiwanese men. Therefore, the development of a reliable, accurate, cost-effective and noninvasive test for oral cancer is highly desirable [4,5]. The alteration of genes has been traditionally revealed by the use of cytogenetics, immunohistochemistry, or molecular approaches based on one or a few genes that change the expression of many genes, e.g., oncogenes and tumor suppressor genes, which have been associated with oral carcinogenesis [6-8]. Genomics has been incorporated in oncology research, and now, in the post-genomic era, there is a strong drive to additionally incorporate proteomic technologies[9,10].
Proteomics can help us to better understand the changes in the levels of multiple proteins involved in oncogenesis and cancer progression and identify new diagnostic and prognostic biomarkers [11]. Quantitative proteomics is an important branch of proteomics that is applied to quantify and identify all the proteins expressed by a whole genome or in a complex mixture. The method of using isobaric tags for relative and absolute quantification (iTRAQ) was developed in 2004. This method uses global peptide labeling to preserve post-translational modification information and involves the simultaneous quantitative proteomic analysis of four samples under the same experimental conditions [12, 13].This unique approach labels samples with four independent isobaric tags of the same mass 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58
that, after fragmentation in MS/MS, give rise to four unique reporter ions (with m/z values ranging from 114 to 117) that provide quantitative information upon integration of the peak areas to quantify the four different samples [14, 15]. The advanced combination of proteomics and bioinformatics provides the opportunity to study changes in the global proteome and other molecular-level indicators of expression in cells at any time point or treatment. Unfortunately, only a few limited studies have been conducted in oral cancer research to date, and none have reported a miRNA-related approach.
The miRNAs are regulatory, non-coding RNAs approximately 21–23 nucleotides long that are expressed at specific stages of tissue development or cell differentiation and have large-scale effects on the expression of a variety of genes at the post-transcriptional level. Through base pairing with target mRNAs, miRNA induces RNA degradation or translational suppression of the targeted transcripts [16, 17]. Mature miRNAs result from the processing of pri-miRNAs in two sequential cleavage steps
mediated by two RNase III enzymes, Drosha and Dicer [18].Each mature 21- to 23-nt miRNA product contains a 2-nt 3′ overhang on each strand and acts as the functional intermediate of RNAi that direct mRNA cleavage and translational attenuation. Although their biological functions remain largely unknown, recent studies suggest that miRNAs contribute to the development of various cancers [19-23]. An important aspect of cancer biomarker identification is the development of simple, noninvasive tests that indicate cancer risk, allow for early detection, and enable the classification of tumors to ensure that patients receive the most appropriate therapy while monitoring disease progression, regression, and recurrence [24]. Serological biomarkers can be analyzed relatively easily and economically and therefore have the potential to greatly enhance screening acceptance [25]. Determinations of serum miRNA are more convenient and less costly than serum protein marker determinations. Thus, serum miRNA biomarkers are more suitable for future clinical applications in parallel with cancer-related examination and have the potential to be the reliable biomarkers that can be 59 60 61 62 63 64 65 66 67 68 69 70 71 72 73 74 75 76 77 78 79 80 81
utilized to develop an effective approach for the early diagnosis of oral cancer.
Our previous study exploited a high-throughput technique that yields results complementary to the traditional gel-based proteomics approaches for oral cancer and colon cancer [26, 27].In this study, we globally surveyed differential protein expression to identify potential diagnostic marker proteins from the paired tissues of patients with early stage oral cancer using a comprehensive iTRAQ analysis and then utilized bioinformatics to predict the associated target miRNAs and validated the correlations from the tissues and sera from patients at varying stages of disease. We determined that plasma miR-27b levels can provide such a biomarker for early cancer detection in clinical applications and provide new insight in cancer research.
82 83 84 85 86 87 88 89 90
Materials and methods
Patients and clinical samples for iTRAQ
We collected pairs of surgically archived specimens of primary oral squamous cell carcinoma (OSCC) and their matched adjacent normal surrounding mucosa specimens from 20 patients at the China Medical University Hospital (Taichung, Taiwan), the collection methods as described in our previous study.26 The 20 patients were classified into two categories: T1 (n=10) and T2 (n=10)
according to UICC TNM staging. The cancer stage was defined by the extent of the lesion as determined by physical examination, radiological studies and pathological examination. The histology for each patient was confirmed by two independent histopathologists following fixation, embedding, sectioning and H&E staining. Overall, the tumor specimens all contained 90% tumor cells. No tumor cells were detected in the surrounding mucosal tissue. Tissues were kept at −80 °C continuously until analysis. The collection protocols were approved by the Institutional Review Board of China Medical University Hospital (DMR98-IRB-77) and informed consent was obtained from each subject.
Sample preparations for iTRAQ
The extraction of total protein from oral cancer tissues (Tumor, n=20) and the corresponding normal tissues (Normal, n=20) were performed as previously described. Additionally, the protein supernatants were enriched using a 3-kDa centrifugal filter as described by the manufacturer (Millipore, Merck KGaA, Germany). This process was repeated twice using ddH2O for desalting and to
remove the protease inhibitor cocktail. The protein concentrations of the resulting supernatants were measured using the Bio-Rad Protein Assay (Bio-Rad, Hercules, CA) according to the manufacturer’s instructions and stored at -20 °C for subsequent processing. A total of 2000 µg of protein was collected from the paired tumor and normal tissues in the 20 patients (100 µg protein from the normal and tumor tissue in each patient; Normal group, n=20; Tumor group, n=20) for iTRAQ analysis.
91 92 93 94 95 96 97 98 99 100 101 102 103 104 105 106 107 108 109 110 111 112 113
Chemicals and reagents for iTRAQ
The acetonitrile solvent was chromatography grade from Merck KGaA. Bradford protein assay kits were purchased from Bio-Rad. Modified porcine trypsin (sequencing grade) was obtained from Promega (Madison, WI). The iTRAQ 4-plex reagent kits were purchased from Applied Biosystems (Framingham, MA). Desalting spin columns were purchased from Pierce (Thermo Scientific, San Jose, CA). Centrifugal filters were purchased from Millipore (Merck KGaA). Column packing materials for analyte column were purchased from Macherey Nagel (Düren, Germany). Packing materials for the trap column were purchased from Michrom Bioresources (Auburn, CA). All other chemicals and reagents were analytical grade from Sigma-Aldrich (St. Louis, MO) unless otherwise stated.
Reduction, alkylation, digestion, and labeling with iTRAQ
For our iTRAQ experiment, the two groups of proteins isolated from the 20 patients were polled. Protein samples were then reduced, alkylated, digested, and labeled with iTRAQ reagents according to the recommended protocol (Applied Biosystems, Framingham, MA). The protein pellets were resuspended in 0.5 M triethylammonium bicarbonate (TEAB), pH 8.5, and 2% sodium dodecyl sulfate (SDS), reduced with 5 mM Tris (2-carboxyethyl) phosphine (TCEP) for 1 h at 60 °C, and alkylated with 10 mM s-methyl methanethiosulfonate (MMTS) at room temperature for 10 min. A total of 60 µg of protein was digested overnight in tryptic solution (30/1, w/w) at 37 °C. Digested samples were labeled with the iTRAQ reagents, and ethanol and the corresponding iTRAQ reagent was added to each sample vial. The samples were labeled as follows: 114, 1; 115, Normal-1; 116, Tumor-2; and 117, Normal-2. Duplicate sets of iTRAQ samples were labeled to monitor the consistency of the results. After 1 h of iTRAQ labeling; the samples were then mixed and dried by centrifugal evaporation.
Strong cation exchange chromatography (SCX) and solution IEF 114 115 116 117 118 119 120 121 122 123 124 125 126 127 128 129 130 131 132 133 134 135 136
The dried, labeled peptide mixture was reconstituted with buffer A (10 mM KH2PO4 in 25%
ACN, pH 3). Peptide separation was performed using a Polysulfoethyl A Column (200 mm L× 2.1 mm i.d., 5 µm, 300 Å, PolyLC, Columbia, MD) on an Agilent 1100 binary HPLC (Agilent Technologies, Wilmington, DE) using a 90 min gradient. One-hundred twenty micrograms of iTRAQ-labeled peptides were eluted at a flow rate of 200 µL/min with a gradient of 2% buffer B (10 mM KH2PO4 in
25% ACN/ 350 mM KCl, pH 3) for 15 min, 2-40% buffer B for 38 min and 40-98% buffer B for 7 min and then maintained in 98% buffer B for 5 min before equilibrating with 2% buffer B for 20 min. A total of 24 fractions were collected, pooled and purified using C-18 Spin columns (Thermo Scientific, San Jose, CA) for further nano-LC-MS/MS analysis.
One-hundred twenty micrograms of dried, iTRAQ-labeled peptide was dissolved in 0.72 mL ddH2O and 2.88 mL IPG stock buffer (pH 3-10). The IPG strips (pH 3-10, 24 cm) were assembled on
the OGE trays and rehydrated for 30 min with a solution of 240 µL H2O and 0.96 mL IPG stock buffer.
The samples were loaded into 24 off-gel wells. Peptide separation was performed over 48 h using the 3100 OFFGEL fractionator (Agilent Technologies, Wilmington, DE) with a limiting current of 50 µA and a limit of 50 kV·h before holding the voltage at 500 V. The samples from the collected fractions were purified using C-18 Spin columns (Thermo Scientific, San Jose, CA) for further nano-LC-MS/MS analysis.
LC-ESI-MS/MS analysis
Each separated peptide fraction was reconstituted in buffer A (0.1% formic acid in 1% ACN), and 0.5 µg of the peptides from each fraction were loaded onto a lab-made 2 cm trap column (100 μm i.d., 5μm, 200 Å) at a flow rate of 10 µL/min in 2% buffer A. Separation was performed on a lab-made 10 cm analytical column (75 µm i.d., 3 µm, 100Å). Peptides were eluted at a flow rate of 250 nL/min across the analytical column with a linear gradient of 2-40% buffer B (0.1% formic acid in 99% ACN) 137 138 139 140 141 142 143 144 145 146 147 148 149 150 151 152 153 154 155 156 157 158 159
for 95 min and 40-80% buffer B for 5 min and then maintained in 80% buffer B for 5 min before equilibrating with 2% buffer B for 25 min. The LTQ-XL settings were as follows: spray voltage, 2.0 kV; capillary temperature, 200 °C; and full MS scan range, 350-2000 Da. Thermo Scientific LTQ-XL
was operated in a data-dependent mode, i.e., using one MS1 scan for precursor ions followed by three
data-dependent zoom scans for precursor ions above a threshold ion count of 200,000 followed by three data-dependent PQD-MS2 scans and CID-MS2 scans for precursor ions in zoom scan above a threshold
ion count of 100, with collisional energies of 29% for PQD and 35% for CID. Database search and iTRAQ quantification
Protein identification and quantification in the iTRAQ samples were performed using Protein Discoverer software (ThermoFisher, version 1.1) using MASCOT search algorithm. The search was performed against the SwissProt v.2011_02 database (525207 sequences) with the following search parameters: taxonomy, Homo sapiens; enzyme, trypsin; max. miss cleavages, 1; fixed modifications, methylthiolation, N-terminal iTRAQ 4plex, lysine iTRAQ 4plex; variable modifications, methionine oxidation, tyrosine iTRAQ 4plex; MS peptide tolerance, 1.5 Da; and MS/MS tolerance, 0.6 Da. Proteins identified with at least two distinct peptides matched with a probability of 0.95 or above are considered as correct identifications.
Patients and clinical samples for quantitative analysis of gene, tissue miRNA, plasma miRNA and protein expression
The 17 patients were classified into two categories: T3 (n=10) and T4 (n=7) according to UICC TNM stages, and samples were collected to evaluate the levels of gene, miRNA and protein expression in the 20 patients following the same IRB project. Total RNA, miRNA and protein extractions were isolated from pairs of surgically archived specimens and their matching adjacent normal surrounding 160 161 162 163 164 165 166 167 168 169 170 171 172 173 174 175 176 177 178 179 180 181
mucosa specimens in the total 37 patients. Following our previous study, the tumor and corresponding normal tissues were dissected by laser pressure catapulting [27].
Plasma was collected from the above 37 patients and 20 healthy volunteers from the same department during 2007–2008. Cell-free nucleic acids were isolated from the blood samples using a 3-spin protocol (1500 r.p.m. for 30 min, 3000 r.p.m. for 5 min, and 4500 r.p.m. for 5 min) to prevent contamination by cellular nucleic acids. All samples were then stored at −80 °C until further processing.
Quantitative analysis of candidate gene expression
Total RNA was extracted from cells using the RNeasy Mini kit (QIAGEN), and the RNA was reverse-transcribed with the SuperScript III First Strand Synthesis System (Invitrogen) according to the manufacturer's instructions. Real-time quantitative PCR was run on a LightCycler 480 with LightCycler 480 SYBR Green I Master kit (Roche). The primers used for real-time quantitative PCR are listed in Supplemental data 1. A reaction mixture containing the following components at the indicated final concentrations was prepared according to the manufacturer's instructions: 0.2 μL of forward primer (20 μmol), 0.2 μL of reverse primer (20 μmol), 0.1 μL of UPL probe and 5 μL master mix. Fifty nanograms of reverse-transcribed total RNA in a volume of 4.4 μL was added as the PCR template. The GAPDH gene was selected for data normalization. A negative control without a cDNA template was included to assess overall amplification specificity. The PCR cycle conditions were as follows: an initial denaturation for 10 min at 95 °C was followed by 40 cycles of amplification at 95 °C for denaturation, 60 °C for annealing and 72 °C for extension. After amplification, the temperature was slowly elevated above the melting temperature of the PCR product to measure the fluorescence and thereby determine the melting curve. The real-time PCR data were calculated by the 2−ΔΔCt method for
RNA quantification. 182 183 184 185 186 187 188 189 190 191 192 193 194 195 196 197 198 199 200 201 202 203 204
Immunoblotting
For the immunoblot assays, 40 µg of each protein extract from the in vivo (Tumor and normal tissues) and in vitro (transfected and control groups) studies were individually separated by 12% SDS-PAGE and treated as described in our pervious study. After blocking, the membranes were probed with one of the following monoclonal antibodies: anti-CAB39L (Santa Cruz Biotechnology Inc., Santa Cruz, CA), anti-TCTP or anti-ITGA6 (Cell Signaling Technology).
Quantization of miRNA expression
Total miRNA was isolated from tissues and plasma using the miRNeasy Mini Kit (Qiagen) according to the manufacturer's instructions. All aliquots of crude products were treated using the same procedure except for the use of 5U DNase. RNA was purified using the RNeasy MinElute Cleanup Kit (Qiagen) according to the manufacturer's instructions. The microRNA levels were determined by real-time quantitative PCR using the TaqMan MicroRNA Assays kit (Applied Biosystems, Foster City, CA) according to the manufacturer's instructions for the analysis of the human 19a (ID 000395), miR-19b (ID 000396), miR-27a (ID 000408), miR-27b (ID 000409), miR-186 (ID 000486), miR-203(ID 000507) and miR-377 (ID 000566). The relative expression levels of the mature miRNAs were calculated using the comparative CT (2−ΔΔCT) method with miRNA-U6 as an endogenous control for
data normalization. All experiments were performed in triplicate wells on an Applied Biosystems 7900 real-time PCR system. The significance of the differences between the plasma miR-27b levels of the control subjects and patients was determined using Student’s t-test.
Transfection of miR-27b
The human oral cancer cell lines HSC-3 and Cal-27 were gifts from Dr. Jing-Gung Chung (China Medical University) and grown in DME/F-12 medium (Gibco) supplemented with 10% FBS and 1% glutamine under humidified 5% CO2 at 37℃. HSC-3 and Cal-27 Cells were plated in six-well
205 206 207 208 209 210 211 212 213 214 215 216 217 218 219 220 221 222 223 224 225 226 227
plates at a density of 1×105 cells/well. The next day, the cells were transfected with 100 nM of
synthetic miR-27b (Ambion) using Oligofectamine (Invitrogen) following the manufacturer's instructions, a cohort transfected with miRNA-let7 as a negative control. The treatment was performed in triplicate, and cells treated with the transfection reagent only were used as a mock control.
Statistical analysis
All experiments were repeated at least in triplicate. All results are expressed as the means ±SD. Student’s t-test was used to evaluate differences between two groups. The data were analyzed using the SPSS 12.0 software package. P-values <0.05 were considered to be significant.
228 229 230 231 232 233 234 235
Results
iTRAQ analysis of the paired tumor and normal tissues from oral cancer patients
A schematic flowchart of the iTRAQ method is provided in Figure 1. Protein identifications were initially accepted when based on at least two peptide identifications with (Mascot scores of ≥ 34) and a false discovery rate of < 3%. The following parameters were used: precursor mass tolerance, 1.5 Da; fragment mass tolerance, 0.6 Da; static modification, methylthio (C), iTRAQ 4-plex (K) and iTRAQ 4-plex (N); and dynamic modification, iTRAQ 4-plex (Y) and oxidation (M). This resulted in the identification of 1959 and 2402 unique peptides, respectively, from the first and second iTRAQ experiments. The relative quantitation was measured as two pairwise ratios: the Normal group against the Tumor group (iTRAQ 115/114 or iTRAQ 117/116). From the two iTRAQ experiments 14,199 and 8,618 MS/MS spectra were identified, leading to identification of 2,402 and 1,959 unique peptides with ≥ 34 ion score and identification of 674 and 757 proteins for the respective groups (Supplemental data 2). A total of 512 and 473 proteins were quantified with at least 2 MS/MS spectra assigned. The following criteria were applied to obtain differentially expressed proteins that displayed consistent under-expression in the Normal group: (1) the proteins are identified by two biological replicates; (2) the proteins have at least 2 quantified MS/MS spectra; (3) the iTRAQ 115/114 or iTRAQ 117/116 ratios must be both lower than 0.67; (3) the variation of coefficient of the iTRAQ ratio must be less than 20%. The overall quality of the quantitation, and subsequent normalization, is revealed by a careful consideration of the ratios for the duplicate analyses at the peptide level. Among the identified proteins, 13 proteins displayed consistent under-expression in the Normal group (Table 1). To gain a better understanding of the 13 proteins in this study, we performed a functional annotation analysis of these proteins. The grouping and naming of the identified proteins in the functional annotation analysis were performed according to the Gene Ontology convention [27].
236 237 238 239 240 241 242 243 244 245 246 247 248 249 250 251 252 253 254 255 256 257 258
Quantification of the expression of the six candidate genes
Excluding ribosomal protein and the variability of both the 115/114 and 117/116 samples in over 10% of the candidates, among the 13 proteins, we selected the epidermal growth factor receptor
(EGFR, coding for receptor tyrosine-protein kinase ErbB-1, EGFR), ornithine aminotransferase (OAT,
coding for Ornithine aminotransferase, OAT), tumor protein, translationally controlled 1 (TPT1, coding for translationally controlled tumor protein, TCTP), integrin alpha-6 (ITGA6, coding for Integrin alpha-6 precursor, ITGA6), Ras GTPase-activating protein-binding protein 1 (G3BP1, coding for Ras GTPase-activating protein-binding protein 1, G3BP1) and calcium-binding protein 39-like
(CB39L, coding for Calcium-binding protein 39-like, CB39L). The transcript levels of these six genes
were validated in the paired tumor and normal tissues from 37 OSCC patients (T1=10, T2=10, T3, n=10 and T4, n=7) by real-time quantitative PCR. The dynamic results suggested that all of the candidate genes were overexpressed in the tumor tissues across all stages. Comparing the tumor and corresponding normal tissues from the 37 oral cancer patients, the EGFR, OAT, TPT1, ITGA6, G3BP1 and CB39L genes were overexpressed by an average of 2.74-fold (95% CI, 1.53-3.25), 1.07-fold (95% CI, 0.57-1.65), 3.42-fold (95% CI, 1.80-4.52), 5.65-fold (95% CI, 1.58-5.22), 1.33-fold (95% CI, 0.13-3.74) and 3.72-fold (95% CI, 1.08-4.22) in the tumor tissues, respectively (Figure 2A). These results indicate that the overexpression phenomena differentiating the tumor and normal tissues in the oral cancer patients were not only manifested at the tissue protein levels but also at the mRNA level. In addition, three genes, TPT1, CAB39L and ITGA6, displayed stable and higher expression levels related to gene transcription in the tumor tissues (with a 3-fold greater mean overexpression level).
Moreover, we organized the individual overexpression levels of TPT1, CAB39L and ITGA6 in the T1-4 groups. The overexpression of TPT1gene was observed not only in the early stage but also increased in the late stages from 4.51-fold (T1) and 4.3-fold (T2) to 4.62-fold (T3) and 5.74-fold (T4), 259 260 261 262 263 264 265 266 267 268 269 270 271 272 273 274 275 276 277 278 279 280 281
although the overexpression was not significantly different between T1 + 2 and T3 + 4 (p= 0.078). Additionally, CAB39L and ITGA6 consistently displayed greater overexpression levels in the T1 and T2 groups (with 6.9- and 6.6-fold increases on average for ITGA6 and 4.7- and 5.1-fold increases for
CAB39L) in the tumor tissues (Figure 2B). However, lower transcription levels of these genes were
observed in the tumor tissues of the T3 and T4 patients (with 3.8- and 3.1-fold increases for ITGA6 on average and 2.3- and 1.6-fold for CAB39L). These two genes displayed significant differences between their mean transcription levels in the T1+T2 and T3+T4 groups (P=0.02024 for ITGA6, and P=0.00038 for CAB39L). These results indicated that tissue overexpression of CAB39L and ITGA6 were dominant in the early stage and not correlated with oral cancer progression, which are important for early diagnosis.
Validation of the protein expressions in the tissues
The protein expression levels of TPT1, ITGA6 and CAB39L were validated in the paired tumor and normal tissues from 37 OSCC patients (T1=10, T2=10, T3, n=10 and T4, n=7). The results suggested that all of the three proteins were overexpressed in the tumor tissues across all stages (Figure 3A). Comparing the tumor and corresponding normal tissues from the 37 oral cancer patients, the TPT1, ITGA6 and CAB39L proteins were overexpressed by an average of 2.83-fold, 3.27-fold and 1.62-fold in the tumor tissues, respectively (Figure 3B).
Predictions of the candidate miRNAs
The candidate microRNA predictions were based on the 3′UTRs of the TPT1, CAB39L and
ITGA6 genes using two bioinformatics tools, the mirSVR predicted target site scoring method[28] and
the online tools at MicroRNA.org (http://www.mirbase.org). All the individual candidate miRNAs were reported (Table 2); we selected the seven overlapping miRNAs (miR-19a, miR-27a, miR-27b, miR-19b and miR-186 miR-203, miR-377) underlined in bold font as candidate targets for the 282 283 284 285 286 287 288 289 290 291 292 293 294 295 296 297 298 299 300 301 302 303 304
subsequent miRNA determinations.
Quantitative analysis of candidate miRNA expression
In the tissue study comparing the paired normal and tumor tissues, the target miRNAs miR-19a, miR-19b, miR-186, miR-203 and miR-377 were detected by real-time Q-PCR and showed overexpression levels averaging 7.93-, 1.18-, 1.64-, 14.57- and 33.21-fold, respectively, in the tumor tissues from the 20 patients. However, only the has-miR-27a and has-miR-27b were underexpressed in the tumor tissues, by 0.57- and 0.32-fold on average, respectively (Figure 4A). Following the introduction, the overexpressions of miRNA will reduce the target protein expressions after transcriptions, thus we selected the has-miR-27a and has-miR-27b for advance detection.
The plasma levels of miR-27a and miR-27b were determined and compared in 10 normal individuals and 20 oral cancer patients. Unlike the control, miR-16, which was found in all patients and normal individuals, plasma miR-27a was detected in only 11 % (4 of 37) of the OSCC patients and in only 25% (5 of 20) of the healthy volunteers under the same conditions. The data indicate that the major expression of miRNA-27a in tissues does not translated to higher plasma levels. However, the plasma miR-27b levels were detectable in 70 % (26 of 37) of the tumor patients and all the healthy volunteers; the levels tended to be significantly higher in the normal group than in the oral cancer group (P=0.00019; 95% confidence interval, 9.5-29.7; with a mean level 13.2-fold higher) (Figure 4B). This result indicated that decreased plasma miR-27b levels are a significant indicator for oral cancer and that the miR-27b suppression may involve the expression off the tumor-associated proteins TCTP, CAB39L and ITGA6 in oral cancer. We also investigated possible correlations between plasma miR-27b levels and sex, age and tumor size in the 37 patients; no significant associations were observed. Translational regulation of candidate proteins by miR-27b
305 306 307 308 309 310 311 312 313 314 315 316 317 318 319 320 321 322 323 324 325 326
To determine the effects of miRNA-27b on the expression of the TCTP, CAB39L and ITGA6 proteins, 100 nM of synthetic miR-27b was transfected into HSC-3 and Cal-27 cells. The overexpression of miR-27b significantly decreased the expression of TCTP mRNA and protein in both the HSC-3 and Cal-27 cell lines but did not significantly affect gene and protein expression in the CAB39L and ITGA6, as determined by both Q-PCR and immunoblot assays (Figure 5A and B). These results indicate that miRNA-27b expression is involved in the regulation of TCTP in oral cancer cells. 327 328 329 330 331 332
Discussion
In tumorigenesis and cancer progression, dynamic protein expression is always a critical target for initial investigation. Recently, functional proteomics combined with mass spectrometry (MS) has offered great promise for unveiling the complex molecular events of tumorigenesis and transforming the management of cancer by identifying new markers for screening, diagnosis, prognosis, and monitoring responses to therapy [29-31]. Following previous reports, the rapid advances in quantitative proteomic analysis based on isotope-dilution MS with mass-tagging reagents, such as iTRAQ, in addition to multidimensional liquid chromatography−tandem mass spectrometry (LC−MS/MS) have revolutionized the field of biomarker discovery and identification [32, 33]. Gel-free approaches LC-MS/MS have been widely used for clinical tissue and cell line analysis. iTRAQ reagents contain isobaric tags designed to bind specifically to the amine groups of peptides, enabling their quantitation by the measurement of relative intensities of the reporter ions generated upon MS/MS fragmentation. To avoid possible analytical and methodological biases, a technology different from the one used for biomarker discovery is typically employed. Real-time Q-PCR and immunoblots are robust, powerful techniques that are frequently selected for biomarker verification. These techniques permit not only the identification and verification of biomarkers but also their quantification in various tissues and body fluids [27]. Studies evaluating the prognostic utility of potential biomarkers identified using gel free-proteomic techniques are not rare and are also needed to substantiate advances in clinical utility and elucidate the potential mechanisms of new biomarkers.
The mature 20- to 23-nt miRNAs act as functional intermediates of RNAi that direct mRNA cleavage and translational attenuation, and recent studies suggest that miRNAs contribute to the development of hand and neck cancers [19, 34]. Generally, the miRNA can be easily and quickly determined using real-time Q-PCR and is more suitable as a potential biomarker than protein 333 334 335 336 337 338 339 340 341 342 343 344 345 346 347 348 349 350 351 352 353 354 355
biomarkers for clinical applications. Generally, mRNA profiling with microarrays has been widely used for miRNA target identification, but microarrays only detect the effects of miRNAs at the transcriptional level and will miss targets repressed solely at the protein level. However, methods based on the use of miRNA profiling techniques to predict proteins levels are difficult to implement in a high-throughput fashion. Thus, we designed a new platform to demonstrate the potential for the prognostic applications of candidate biomarkers identified using tissue proteomics by iTRAQ analysis and attempted to identify biomarkers related to target miRNAs to replace proteins as useful biomarkers for oral cancer.
In the iTRAQ analysis, we identified and quantified a total of 13 down-regulated and 6 regulated proteins with consistently different expression from the Normal tissue group. The 6 up-regulated proteins have been identified for future work in another study. In this study, we focus on the 13 down-regulated proteins and the characterization of their mRNA- and miRNA-associated expression levels in tissues and plasma. Moreover, we integrated in vivo and in vitro studies to demonstrate that the expression of miR-27b significantly affects the downregulation of the TCTP protein, one of the three candidate biomarkers. The developed platform represents the first combined application of a differentially expressed tumor proteomic study to the identification of miRNA biomarkers.Until now, there have been no reports of the identification of potential miRNA biomarkers based on protein data from iTRAQ analysis in cancer research utilizing two types of quantitation. This novel plateform can facilitate simultaneous screening for potentially correlated genes, miRNA and downstream proteins.
Among of the 13 underexpressed proteins, we selected 6 proteins for subsequent study. The expression of their coding genes, EGFR, OAT, TPT1, ITGA6, G3BP1 and CAB39L, were confirmed by real-time quantitative PCR analysis. The TPT1, ITGA6 and CB39L genes were found to be the most 356 357 358 359 360 361 362 363 364 365 366 367 368 369 370 371 372 373 374 375 376 377 378
consistently over-expressed in the tumor tissues, and we suggest that detecting the these three genes could contribute considerably to the early diagnosis of oral cancer (T1 or T2). Excluding TPT1, the QPCR results for the ITGA6 and CB39L genes demonstrated that their over-expression appears more significantly in the early stages of oral cancer (T1+T2) than in the late stages (T3+T4) (Figure 2). In previous studies, the TPT1, ITGA6 and CB39L genes were reported to be implicated in tumor growth, malignancy and poor survival rates in breast cancer and monocytic leukemia [35-37], but the studies have not reported the protein levels and correlated miRNAs associated with cancer. The three tumor-related proteins and genes were also demonstrated for the first time here to be useful as the biomarkers for oral cancer based on a clinical global quantitative proteomic analysis. The correlation of Q-PCR data with protein expression provides evidence that the iTRAQ labeling method for large-scale protein quantification is powerful and reliable for advanced miRNA biomarker profiling.
It is known the miRNAs can control the expression levels of particular genes by binding to the 3′UTRs of mRNAs. Thus, the dysregulation of miRNAs is expected to be found in diseases such as cancer that are attributed to deregulated gene expression, suggesting that miRNA alteration may initiate carcinogenesis [38]. However, there is limited information on the aberrant expression of miRNAs in oral cancer [39]. Among the 7 miRNAs that were predicted for the TPT1, CAB39L and ITGA6 genes in a bioinformatic analysis, miR-27a and miR-27b demonstrated consistent under-expression in the laser pressure catapulting dissected tumor tissues in comparison with the corresponding normal tissues; unfortunately, the miR-27a was found in too few plasma samples in the oral cancer patients to be statistically significant in this group. Specially, the tissue related determinations were carried out after the laser pressure catapulting dissection that has been reported is the best tool to define the malignant and benign sample in the translational medicine. Thus, we focused on miR-27b in this study. The aberrant expression of miRNAs has been implicated in the carcinogenesis of various cancers. In 379 380 381 382 383 384 385 386 387 388 389 390 391 392 393 394 395 396 397 398 399 400 401
previous studies, an up-regulation of miR-27b was observed in OSCC and breast cancers as the tumors suppress various factors in genomic experiments, but the effects of miR-27b on the regulation of intercellular proteins in oral tumorigenesis remains unclear [40, 41].
As established in previous studies, TCTP is a highly conserved protein expressed abundantly across a wide range of eukaryotes and is known to promote diverse cellular functions by interacting with various proteins, with expression levels varying depending on the tissue type, growth, stress factors and cytotoxic signals [42]. Knocking down TCTP protein expression in tumor cells inhibits
their growth, motility and invasive functions, and its expression levels have been demonstrated to be downregulated by the activation of the tumor suppressor protein p53 [43, 44]. It also has been suggested that several breast tumor cell lines, including MCF-7 and T47D, have high endogenous levels of TCTP protein, and siRNA transfection promotes the downregulation of TCTP, resulting in tumor reversion. However, the molecular mechanisms of tumor reversion remain to be defined. Recently, TCTP was demonstrated to induce the chronic activation of Src, and EGFR and downstream signaling molecules associated with cell transformation were significantly down-regulated by TCTP siRNA [45]. It thus appears that the down-regulation of TCTP quenches the activation of tumor-associated signals, causing tumor reversion [46, 47]. So far, it has not been investigated whether the TCPT protein or the TPT1 gene can be regulated by miRNAs.
Thus, we performed transient transfection to overexpress the human miR-27b in oral cancer cell lines to identify its regulatory role. This is the first report demonstrating that miR-27b is a negatively regulator of the expression the of TPT1 gene and TCPT protein. We further demonstrated that miR-27b overexpression not only reduced the level of TPT1 gene expressions and also downregulates cellular TCPT expression in the oral cancer cells (Figure 4), but also that gene and protein levels of CAB39L and ITGA6 are not significantly regulated by miR-27b overexpression, although the miR-27b sequence 402 403 404 405 406 407 408 409 410 411 412 413 414 415 416 417 418 419 420 421 422 423 424
was predicted from these three genes’ 3-UTR. These interesting and varied regulatory mechanisms require further study to clarify their roles in cancer cells.
Our results indicate that miR-27b may function as a tumor-suppressor gene, whereas another study suggested it may function as an oncogene in MDA-MB-231 cells [47]. They revealed that the introduction of miR-27b into cells expressing the target gene, ST14, did not suppress cell growth, suggesting that both miR-27b-dependent and independent functions of ST14 protein exist in breast cancer. Furthermore, the miR-19a, miR-19b, miR-20a and miR-27b sequences were identified as the major growth-sustaining micro-RNAs in the tumor cells in an assay of individual cell growth arrest [48].The complementary screening study unveiled only functional differences between homologous mi-RNAs; miR-27b appears to be involved in an important novel mechanism in tumorigenesis. These differences may occur due to differences in cancer types, patient organization, tumor status or histologic tumor grade, etc.
Conclusion
Our study introduced the efficient use of an iTRAQ proteomics study to identify three protein biomarkers for early stage oral cancer. We additionally discovered that tissue and plasma miR-27b levels are better biomarkers than protein biomarkers due to their convenient determination by Q-PCR, and that miR-27b overexpression in oral cancer cells significantly down-regulates the tumor associated TCTP protein. The integrated study of quantitative proteomics and molecular biology will provide new insights in both Proteomics and Translational medicine.
Acknowledgements 425 426 427 428 429 430 431 432 433 434 435 436 437 438 439 440 441 442 443 444 445 446
NSC 97-2320-B-039-016-MY3, and NSC 98-2113-M-003-007-MY2), the China Medical University Hospital (DMR-98-100) and China Medical University (CMU100-S-04).
448 449
References
1. Jemal A, Bray F, Center MM, Ferlay J, Ward E, Forman D. Global cancer statistics. CA Cancer J Clin 2011;61(2):69-90.
2. Seiwert TY, Cohen EE. State-of-the-art management of locally advanced head and neck cancer. Br J Cancer 2005;92(8):1341-8.
3. Ho PS, Ko YC, Yang YH, Shieh TY, Tsai CC. The incidence of oropharyngeal cancer in Taiwan: an endemic betel quid chewing area. J Oral Pathol Med 2002;31(4):213-9.
4. Chen YJ, Chang JT, Liao CT, Wang HM, Yen TC, Chiu CC, et al. Head and neck cancer in the betel quid chewing area: recent advances in molecular carcinogenesis. Cancer Sci 2008;99(8):1507-14.
5. Turhani D, Krapfenbauer K, Thurnher D, Langen H, Fountoulakis M. Identification of differentially expressed, tumor-associated proteins in oral squamous cell carcinoma by proteomic analysis. Electrophoresis 2006;27(7): 1417–23.
6. Gstaiger M, Jordan R, Lim M, Catzavelos C, Mestan J, Slingerland J, et al. Skp2 is oncogenic and overexpressed in human cancers. Proc Natl Acad Sci U S A 2001;98(9):5043–8.
7. Huang X, Gollin SM, Raja S, Godfrey TE. High-resolution mapping of the 11q13 amplicon and identification of a gene, TAOS1, that is amplified and overexpressed in oral cancer cells. Proc Natl Acad Sci U S A 2002;99(17):11369–74.
8. Sugimura T, Kumimoto H, Tohnai I, Fukui T, Matsuo K, Tsurusako S, et al. Gene–environment interaction involved in oral carcinogenesis: molecuar epidemiological study for metabolic and DNA repair gene polymorphisms. J Oral Pathol Med 2006;35(1):11–8.
9. Kallioniemi O. Medicine: profile of a tumour. Nature 2004;428(6981):379–82. 450 451 452 453 454 455 456 457 458 459 460 461 462 463 464 465 466 467 468 469 470 471
10. Geho DH, Petricoin EF, Liotta LA. Blasting into the microworld of tissue proteomics: a new window on cancer. Clin Cancer Res 2004;10(3):825–7.
11. Garbis SD, Tyritzis SI, Roumeliotis T, Zerefos P, Giannopoulou EG, Vlahou A, et al. Search for potential markers for prostate cancer diagnosis, prognosis and treatment in clinical tissue specimens using amine-specific isobaric tagging (iTRAQ) with two-dimensional liquid chromatography and tandem mass spectrometry. J Proteome Res 2008;7(8):3146-58.
12. Wu WW, Wang G, Baek SJ, Shen RF. Comparative study of three proteomic quantitative methods, DIGE, cICAT, and iTRAQ, using 2D gel- or LC-MALDI TOF/TOF. J Proteome Res 2006;5(3):651-8.
13. Ross PL, Huang YN, Marchese JN, Williamson B, Parker K, Hattan S, et al. Multiplexed protein quantitation in Saccharomyces cerevisiae using amine-reactive isobaric tagging reagents. Mol Cell Proteomics 2004;3(12):1154-69.
14. Ralhan R, Desouza LV, Matta A, Chandra TS, Ghanny S, Dattagupta S, et al. iTRAQ-multidimensional liquid chromatography and tandem mass spectrometry-based identification of potential biomarkers of oral epithelial dysplasia and novel networks between inflammation and premalignancy. J Proteome Res 2009;8(1):300-9.
15. Lewis BP, Burge CB, Bartel DP. Conserved seed pairing, often flanked by adenosines, indicates that thousands of human genes are microRNA targets. Cell 2005;120(1):15–20.
16. Hsieh HC, Chen YT, Li JM, Chou TY, Chang MF, Huang SC, Protein Profilings in Mouse Liver Regeneration after Partial Hepatectomy using iTRAQ Technology. J Proteome Res 2009;8(2):1004-13.
17. Farh KK, Grimson A, Jan C, Lewis BP, Johnston WK, Lim LP, et al. The widespread impact of mammalian MicroRNAs on mRNA repression and evolution. Science 2005;310(5755):1817–21. 472 473 474 475 476 477 478 479 480 481 482 483 484 485 486 487 488 489 490 491 492 493 494
18. Cullen BR. Transcription and processing of human microRNA precursors. Mol Cell 2004;16(6):861–5.
19. Calin GA, Croce CM. MicroRNA signatures in human cancers. Nat Rev Cancer 2006;6(11):857– 66.
20. Gomes CC, Gomez RS. MicroRNA and oral cancer: future perspectives. Oral Oncol 2008;44(10):910-4.
21. Lima RT, Busacca S, Almeida GM, Gaudino G, Fennell DA, Vasconcelos MH. MicroRNA regulation of core apoptosis pathways in cancer. Eur J Cancer 2011;47(2):163-74.
22. Kouwenhove MV, Kedde M, Agami R. MicroRNA regulation by RNA-binding proteins and its implications for cancer. Nat Rev Cancer 2011;11(9):644-56.
23. Uesugi A, Kozaki K, Tsuruta T, Furuta M, Morita K, Imoto I, et al. The tumor suppressive microRNA miR-218 targets the mTOR component Rictor and inhibits AKT phosphorylation in oral cancer. Cancer Res 2011;71(17):5765-78.
24. Hanash SM, Pitteri SJ, Faca VM. Mining the plasma proteome for cancer biomarkers. Nature 2008;452(7187):571–9.
25. Faca VM, Song KS, Wang H, Zhang Q, Krasnoselsky AL, Newcomb LF, et al. A mouse to human search for plasma Proteome changes associated with pancreatic tumor development. PLoS Med 2008;5(6):953–967.
26. Lo WY, Tsai MH, Tsai Y, Hua CH, Tsai FJ, Huang SY, et al. Identification of over-expressed proteins in oral squamous cell carcinoma (OSCC) patients by clinical proteomic analysis. Clin Chim Acta 2007;376(1-2):101-7.
27. Chen WT, Chang SC, Ke TW, Chiang HC, Tsai FJ, Lo WY. Identification of Biomarkers to Improve Diagnostic Sensitivity of Sporadic Colorectal Cancer in Patients with Low Preoperative 495 496 497 498 499 500 501 502 503 504 505 506 507 508 509 510 511 512 513 514 515 516 517
Serum Carcinoembryonic Antigen by Clinical Proteomic Analysis. Clin Chim Acta 2011;412(7-8):636-41.
28. Betel D, Koppal A, Agius P, Sander C, Leslie C. Comprehensive modeling of microRNA targets predicts functional non-conserved and non-canonical sites. Genome Biology 2010;11:R90.
29. Cho WC. Contribution of oncoproteomics to cancer biomarker discovery. Mol Cancer 2007;6:25. 30. Tanke HJ. Genomics and proteomics: the potential role of oral diagnostics. Ann. N.Y. Acad. Sci
2007;1098:330-334.
31. Najam-ul-Haq M, Rainer M, Trojer L, Feuerstein I, Vallant RM, Huck CW, et al. Alternative profiling platform based on MELDI and its applicability in clinical proteomics. Expert Rev of Proteomics 2007;4(4):447-52.
32. Zieske LR. A perspective on the use of iTRAQ reagent technology for protein complex and profiling studies. J Exp Bot 2006;57(7):1501-8.
33. DeSouza LV, Grigull J, Ghanny S, Dubé V, Romaschin AD, Colgan TJ, et al. Endometrial carcinoma biomarker discovery and verification using differentially tagged clinical samples with multidimensional liquid chromatography and tandem mass spectrometry. Mol Cell Proteomics 2007;6(7):1170-82.
34. Chen LH, Tsai KL, Chen YW, Yu CC, Chang KW, Chiou SH, et al. MicroRNA as a Novel Modulator in Head and Neck Squamous Carcinoma. J Oncol 2010;2010:135632.
35. Jung J, Kim HY, Kim M, Sohn K, Kim M, Lee K. Translationally controlled tumor protein induces human breast epithelial cell transformation through the activation of Src. Oncogene 2011;30(19):2264-74.
36. Cariati M, Naderi A, Brown JP, Smalley MJ, Pinder SE, Caldas C. Purushotham A. D. Alpha-6 integrin is necessary for the tumourigenicity of a stem cell-like subpopulation within the MCF7 518 519 520 521 522 523 524 525 526 527 528 529 530 531 532 533 534 535 536 537 538 539 540
breast cancer cell line. Int J Cancer 2008;122(2):298-304.
37. Zhang PY, Zhang WG, He AL, Wang JL, Li WB. Identification and functional characterization of the novel acute monocytic leukemia associated antigen MLAA-34. Cancer Immunol Immunother 2009;58(2):281-90.
38. Calin GA, Croce CM. MicroRNA signatures in human cancers. Nat Rev Cancer 2006;6(11):857– 66.
39. Komatsu S, Ichikawa D, Takeshita H, Tsujiura M, Morimura R, Nagata H, et al. Circulating microRNAs in plasma of patients with oesophageal squamous cell carcinoma. Br J Cancer 2011;105(1):104-11.
40. Kozaki K, Imoto I, Mogi S, Omura K, Inazawa J. Exploration of tumor-suppressive microRNAs silenced by DNA hypermethylation in oral cancer. Cancer Res 2008;68(7):2094-105.
41. Tsuchiya Y, Nakajima M, Takagi S, Taniya T, Yokoi T. MicroRNA regulates the expression of human cytochrome P450 1B1. Cancer Res 2006;66(18):9090-8.
42. Chen S. H.; Wu P. S.; Chou C. H.; Yan Y. T.; Liu H.; Weng S. Y.; Yang-Yen H. F. A knockout mouse approach reveals that TCTP functions as an essential factor for cell proliferation and survival in a tissue- or cell type-specific manner. Mol Biol Cell. 2007, 18(7), 2525-32
43. Tuynder M.; Fiucci G.; Prieur S.; Lespagnol A.; Géant A.; Beaucourt S.; Duflaut D.; Besse S.; Susini L.; Cavarelli J.; Moras D.; Amson R.; Telerman A. Translationally controlled tumor protein is a target of tumor reversion. Proc Natl Acad Sci U S A. 2004, 101(43), 15364-9
44. Tuynder M.; Susini L.; Prieur S.; Besse S.; Fiucci G.; Amson R.; Telerman A. Biological models and genes of tumor reversion: cellular reprogramming through tpt1/TCTP and SIAH-1. Proc Natl
Acad Sci U S A. 2002, 99(23), 14976-81
45. Jung J.; Kim H. Y.; Kim M.; Sohn K.; Kim M.; Lee K. Translationally controlled tumor protein 541 542 543 544 545 546 547 548 549 550 551 552 553 554 555 556 557 558 559 560 561 562 563
induces human breast epithelial cell transformation through the activation of Src. Oncogene. 2011, 30(19), 2264-74
46. ZOU G. M. Cancer initiating cells or cancer stem cells in the gastrointestinal tract and liver. J Cell
Physiol. 2008, 217(3), 598-604.
47. Wang Y.; Rathinam R.; Walch A.; Alahari S. K. ST14 (suppression of tumorigenicity 14) gene is a target for miR-27b, and the inhibitory effect of ST14 on cell growth is independent of miR-27b regulation. J Biol Chem. 2009, 284(34), 23094-106
48. Peric D.; Chvalova K.; Rousselet G. Identification of microprocessor-dependent cancer cells allows screening for growth-sustaining micro-RNAs. Oncogene. 2011, 1-10
564 565 566 567 568 569 570 571 572 573
