A new panel of pancreatic cancer biomarkers discovered using a mass spectrometry-based pipeline

Xiaohui Liu‡1,2,3, Weimin Zheng‡3, Wansheng Wang4,Huali shen2, Linxiao Liu5,6,Wenhui Lou†7, Xiaolin Wang†5,6, Pengyuan Yang†1,2

1 The Fifth People’s Hospital of Shanghai,FudanUniversity,Shanghai 200240, P. R. China. 2 Institutes of Biomedical Sciences, Fudan University, Shanghai 200032, P. R. China.3Department of Chemistry, Fudan University, Shanghai 200433, P. R. China. 4 Department of Interventional Radiology, First Affiliated Hospital, Soochow University, Suzhou 215006, P. R. China.5 Department of Interventional Radiology, Zhongshan Hospital, Fudan University, Shanghai, 200032,P. R. China.6 Department of Interventional Radiology, Zhongshan Hospital, Fudan University, Shanghai, 200032, P. R. China. 7 Department of pancreatic surgery, Zhongshan Hospital, Fudan University, Shanghai, 200032, P. R. China.

Running title: A new panel of pancreatic cancer biomarkers

‡ Xiaohui Liu and WeiminZheng contributed equally to this work.

† Corresponding author.

Correspondence and requests for materials should be addressed to Pengyuan yang () or Wenhui Lou () or Xiaolin Wang ().

Supplementary Information

Supplemental Methods

Patients and Specimens

A total of 150 serum samples were collected at the Zhongshan Hospital between June 2010 and January 2012. All study participants provided written informed consent prior to inclusion. The research followed the tenets of the Declaration of Helsinki and was approved by the Ethics Committee of the Fudan University Shanghai ZhongshanHospital. The serum samples were categorised as follows: NC(n=40), BD(n=30, pancreatitis(4), pancreatic cysts(13), benign tumors(13)),and PC(n=80). Blood samples were collected in the morning after an overnight fast using Vacutainer tubes (Becton Dickinson, Franklin Lakes, NJ, USA) without anticoagulant and allowed to clot at room temperature for 1 h before centrifugation at 1500 × g for 10 min. The serum was removed, immediately aliquoted in sterile centrifuge tubes and stored at -80 °C for future analysis.

Protein preparation

Each serum sample was centrifuged at 14000g for 30 min at 4℃to discard lipids of serum before sample preparation.

In the discovery stage, every 10 serum samples were pooled together in each group for subsequent analysis. The high-abundant proteins from the pooledserum sample were depleted using a Human 14 Multiple Affinity Removal System Column (Agilent Technologies, USA) on a 20 AD HPLC systems (Shimadzu,Japan). Proteins were then prepared and digested, followed by iTRAQ labeling according to the manufacturer’s instructions. Briefly, 100ug protein of each pooled serum sample were reduced in 50 mMtris-(2-carboxyethyl)phosphine (TCEP)at 56°C for 1 h and alkylated in 200 mMmethylmethanethiosulfonate (MMTS) at room temperature. The reduced and alkylated protein mixtures were precipitated by adding 5× volume of chilled acetone at -20 °C overnight. After centrifugation at 4 °C, 30,000g, the pellet was dissolved in 0.5 M tetraethyl-ammonium bromide (TEAB) and sonicated in ice. Samples were then proteolyzed with 1 μg trypsin:20μg protein. Each digested sample was labeled with iTRAQ reagent at room temperature for 2 h.As a consequence, two sets of iTRAQ 8-Plex (NC with 113 and 114 tags, BD with115 and 116 tags, and PC with 117,118,119 and 121 tags) and 1 set of iTRAQ 5-Plex (NC with 113 tags, BD with 115 and 116 tags, and PC with117 tags) were constructed to provide multiple biological replicates. The labeled peptides were then fractionated withhigh pH reversed-phase liquid chromatography on a UPLC system(Waters, USA).

For P-MRM and SID MRM,protein concentrations were determined by Bradfordassay (Bio-Rad Protein Assay; Bio-Rad Laboratories, Hercules, CA, USA). 100ug protein of each serum was load on Microcon devices YM-30 (Millipore), and filter-aided samplepreparation(FASP) protocol was performed as described previously. Briefly, serum samples were washed with 8M Urea followed by reducing with 10 mMdithiothreitol(DTT) and alkylation with 50 mMiodoacetamide(IAA )in 8M Urea. After alkylation, devices were washed twice with 8 M urea and twice with 25mM ammonium bicarbonate. Trypsin was added to the concentration at a ratio of 1 μg trypsin:20μg protein and incubated overnight at 37 °C. Peptides were collected from the filter by centrifugation and two subsequent washes with 25mM ammonium bicarbonate. All the collected sample were dried on a rotation vacuum concentrators(Christ, Germany)andresuspended in 2% ACN /0.1% formic acid.

High pH RPLC separation of iTRAQ-labled peptides

All the three sets of iTRAQLabled peptides were fractionated separately by using high pH reversed phase liquidchromategraphy on a UPLC system(Waters, USA).The peptides were resuspended with loading buffer (5mM Ammonium formate containing 2% acetonitrile, pH 10). The gradient elution was performed by 0-25%B(5-35 min)and 25-45% B (35-48 min) on high pH RPLC column(Waters, Xbridge C18 3.5µm, 150×2.1 mm). A total of 20 fractions were collected andthen mixed to 10 fractions for each iTRAQ set.

LC-MS/MS analysis of iTRAQ-labled peptides

1 set of 8-Plex labeled sample(8-Plex set1)was analyzed by Q Exactive were performed on a Nanoeasy system (Thermo Fisher Scientific, USA) with a 50-cm-long column (75 um id × 50-cm-long, packed with 2 um id 100 Å pore size, C18 packing material，Thermo scientific, USA) connected to a Q Exactive hybrid quadrupole-Orbitrap mass spectrometer (Thermo Fisher Scientific, USA).Pepitdes were separated with a linear gradient, from 5% B(98% ACN with 0.1% formic acid) to 45% B in 90min. The electrospray voltage of 1.9 kV versus the inlet of the mass spectrometer was used.Q Exactive mass spectrometer was operated in the data-dependent mode to switch automatically between MS and MS/MS acquisition. Survey full-scan MS spectra (m/z 300-1200) were acquired with a mass resolution of 70K, followed by fifteen sequential high energy collisional dissociation (HCD) MS/MS scans with a resolution of 17.5K. In all cases, one microscan was recorded using dynamic exclusion of 30 seconds. For MS/MS, precursor ions were activated using 27% normalized collision energy.

The other 8-Plex and 5-Plex Labeled (8-Plex set2 and 5-Plex) sample were analyzed by 1D Plus nano LC system (Eksigent, USA) coupled with Triple TOF 5600 system (AppiedBiosystems, USA) with a 15-cm-long column (75 um id × 15-cm-long, packed with 2 um id 100 Å pore size, C18 packing material，Thermo scientific, USA). Peptides were subsequently eluted using the following gradient conditions with phase B (98% ACN with 0.1% formic acid): 5-7% B (0-5 min), 7-23% B (5-40min) and 23-45% B (40-75min), and total flow rate was maintained at 300 nL/min. Electrospray voltage of 2.3 kV versus the inlet of the mass spectrometer was used.Triple TOF 5600 mass spectrometer was operated in information-dependent mode to switch automatically between MS and MS/MS acquisition. MS spectra were acquired across the mass range of 350–1250 m/z in high resolution mode using 250 ms accumulation time per spectrum. Tandem mass spectral scanned from 100-1250 m/z in high sensitivity mode with rolling collision energy. The 20 most intense precursors were selected for fragmentation per cycle with dynamic exclusion time of 9s.

iTRAQ Protein identification and quantitation:

All the data was searched against Swissprot human database(20238 entries ).

For data generated on Q Exactive platform, Protein identification and quantitation was performed on Proteome Discovery version 1.3 (PD,Thermoscientific,USA)using MASCOT search engine. Parameters searching were set as following:Quantitation: iTRAQ-8plex; Enzyme:Trypsin;Variable modification: Oxidation (Met);Fixed modification: Methyl methanethiosulfonate (Cys).MS1 tolerance :20ppm; MS/MS tolerance: 20mmu;with One missed cleavage on trypsin was allowed.

For data generated on Triple TOF 5600system, database searching were performed on ProteinPilot(Version 4.5).search parameters were: Sample Type: iTRAQ 8plex(peptide labeled);Cys alkylation:MMTS; Digestion: Trypsin; Instrument: Triple TOF 5600; Special factors: none; Species: Homo sapiens; ID Focus: biological modifications; Search effort:Thorough ID; and false discovery rate(FDR) analysis was performed. Pro Group algorithm was used to calculate the reporter peak area, error factor (EF) and p value and thus quantitate peptides and proteins.

In this study, a false discovery rate (FDR) lower than 1% was used to control protein level identification based on the target-decoy strategy. Proteins with at least one unique peptide with confidence higher than 95% were used for quantitation. Student's t-test (PC vs. control (NC and BD)) was applied to compare the protein expression levels between the PC group and the control group. The mean value of the ratio of each group was used to calculate the fold change. Proteins with a fold change larger than 1.2 or less than 0.8 with aStudent's t-test p-value <0.05 were selected as differently expressed proteins.

1D-Targeted LC/MSMS

eferring to the spectrums generated above, the precursor-ion intensities of all the unique peptides of the 142 proteins were analyzed. The six most intense unique peptides of each protein were selected and their m/z values were set as the inclusion list for 1D-targeted LC-MS/MS detection on a Triple TOF 5600 system. A total of 2 µg of enzymatically digested crude serum sample without high abundant protein depleted was used for the analysis. A short list of 49 proteins was identified with high confidence (FDR<1%).

P-MRM and SID-MRM analysis

P-MRM: A total of 96 crude peptides were synthesized corresponding to the 49 proteins according to the identification of 1D-targeted LC MS/MS. 4 to 10 b- or y- fragment ions were chosen as the candidate transition according to their LC-MS/MS spectrum. A exogenous peptide (DIPVPKPK, K was labeled with 13C and 15N）was added to each digested serum samples to a final concentration of 20 fmol/uL as an internal standards.1 pmol of each crude peptides and 2ug of digested peptides from each of the 52 crude serum samples were analyzed separately.

SID-MRM: For absolute quantification of 4 candidate proteins, 4 stable isotope-labeled peptides corresponding to the best performance peptides of targeted proteins were synthesized (Arginine or Lysine of each peptide were isotope labeled at 13C and 15N,Bankpeptide, Ltd, China). SIS peptide of4-5 orders of magnitude were added to the digested serum proteins and tested in triplicate to construct a standard curve. SIS peptides were spiked-in each sample to a certain concentration(THLAPYSDELR(APOA1):1pmol/uL, VTEPISAESGEQVER(APOL1):10fmol/uL, WVQTLSEQVQEELLSSQVTQELR (APOE): 0.5 pmol/uL, DYIFGNYIER(ITIH3):5 fmol/uL ). 1 ugproteotypic peptides of each sample were applied for analysis.

All the MRM analysis was performed on a 6500 QRTAP hybrid triple quadrupole/liner ion trap mass spectrometry (AB Sciex, CA) interfaced with UPLC systems (Eksigent, CA ). Peptide of each sample was loaded and separated at a flow rate of 300nl/min using a 90 min gradient, consisting of 5% solution B( 0.1% formic acid in acetonitrile) for 5min, 5% to 25% solution B for 50 min, 25% to 35% solution B for 10 min, 35% to 80% solution B for 5 min, 80% solution B for 3 min, and 80% to 5% solution for 3 min, 5% solution B for 10 min to equilibrate the column. The instrument settings were: ion-spray capillary voltage: 2300 V, interface temperature :150℃,gas flow rate: 20L/min. All the MRM data were processed using Skyline software.

Immunoassay measurement of CA19-9 level

CA19-9 levels of the 100 serum samples tested in SID-MRM analysis were determined by electrochemiluminescence immunoassay (double-antibody sandwich ELISA) on a Roche cobas e 602 module according to the manufacturer’s instructions(Roche, USA).

Statistics Construction of a Diagnostic Model

The quantitative results from the P-MRM and SID-MRM analyses were compared and visualized using Prism 5.0. The peptide concentrations of APOE, ITIH3, APOA1, APOL1 and the expression level of CA19-9in the serum of the NC group (34 cases), BD (26 cases) and the PC group (40 cases) were used to construct the diagnostic model. The statistical analyses were performed using SPSS(v24.0) , and P<0.05 was considered statistically significant. Receiver operation characteristic curves (ROCs) were calculated to determine the specificity and sensitivity, as well as to compare the area under the curve (AUC) of single candidate biomarkers and their combinations using a binary logistic regression analysis.

Immunohistochemistry(IHC)

4 um whole sections of formalin-fixed and paraffin-embedded tissue(cancer and Para cancer tissues of pancreas) samples were prepared. These were deparaffinized with xylene, then rehydrated in four graded alcohols(70,80,90 and 100%).The rabbit polyclonal antibodies, APOL1 Polyclonal antibody (Proteintech, 1:300 dilution), Apolipoprotein AI Polyclonal antibody (proteintech, 1:50 dilution), ITIH3 Polyclonal antibody (Proteintech, 1:50 dilution), and Anti-Apolipoprotein E antibody (abcam,1:1000 dilution) were incubated for 1h at room temperature, detected with anti-rabbit IgImmPRESS-HRP reagent(Beyotime) and visualized using DAB+ substrate(Dako). AperioScanscopeXT(Leica Biosystems) was applied for digitally scanning the slides.