Annexin A3 is a mammary marker and a potential neoplastic breast cell therapeutic target.

Breast cancers are the most common cancer-affecting women; critically the identification of novel biomarkers for improving early detection, stratification and differentiation from benign tumours is important for the reduction of morbidity and mortality.To identify and functionally characterise potential biomarkers, we used mass spectrometry (MS) to analyse serum samples representing control, benign breast disease (BBD) and invasive breast cancer (IDC) patients. Complementary and multidimensional proteomic approaches were used to identify and validate novel serum markers.Annexin A3 (ANX A3) was found to be differentially expressed amongst different breast pathologies. The diagnostic value of serum ANX A3 was subsequently validated by ELISA in an independent serum set representing the three groups. Here, ANX A3 was significantly upregulated in the benign disease group sera compared with other groups (P < 0.0005).In addition, paired breast tissue immunostaining confirmed that ANX A3 was abundantly expressed in benign and to a lesser extent malignant neoplastic epithelium. Finally, we illustrated ANX A3 expression in cell culture lysates and conditioned media from neoplastic breast cell lines, and its role in neoplastic breast cell migration in vitro.This study confirms the novel role of ANX A3 as a mammary biomarker, regulator and therapeutic target.


Serum sample collection and preparation
Serum samples were collected as part of the proteomic analysis in breast screening study involving breast disease and healthy volunteers from the Wessex region (UK) prior to any intervention (Table 1). Sample collection was approved by the Southampton General Hospital NHS Trust Ethics Committee (Ethical Approval 05/Q1702/13, R&D reference no. RHMCAN0392) and informed consent was obtained from all participants in the study.
Serum samples were fractionated using solid anion exchange (SAX) chromatography and pH gradient elution (pH between 9 and 2) as per manufacturer's instructions (BioRad,Hercules,CA,USA;. Protein array preparation for MS profiling was carried out using a BioMek3000 (Beckman Coulter, High Wycombe, UK) liquid handling robot where samples were applied to protein arrays using two technical replicates in a randomised fashion. Weak cation exchange (WCX) protein arrays were treated with 150 μl aqueous sodium acetate (50 mM, pH 4.0) twice for a total of 10 minutes at room temperature. An aliquot of 10 μl -mixed with 90 μl sodium acetate (50 mM pH 4.0) buffer to generate 1:10 dilutionsfrom each fraction was added to each spot. Samples were incubated for 60 minutes in a humid chamber to prevent drying at room temperature (25°C). Following incubation, arrays were washed with 150 μl ammonium acetate (50 mM, pH 4.0) buffer twice, rinsed with de-ionised water and allowed to air dry. Two applications of 1 μl of 50% sinapinic acid (SPA) in 50% acetonitrile (ACN) 0.5% trifluoroacetic acid (TFA) were added to each protein sample on the chip and allowed to air dry for 15 minutes at room temperature (25°C).

SELDI and MALDI-TOF MS-based proteomic profiling analysis
Both the SELDI and MALDI-TOF MS profiling platforms were used for the MS profiling analysis. For the SELDI-TOF MS analysis; protein arrays were analyzed using the Enterprise 4011 SELDI platform (BioRad, Hercules, CA). Data was analyzed using ProteinChip Data Manager v. 3.0.7 Software.
For the MALDI-TOF MS-MS analysis; protein arrays were analysed by MALDI-TOF MS Ultraflex III (Bruker Daltonics, Bremen, Germany). Spectra were acquired on the Ultraflex mass spectrometer in the linear mode at 80% laser power over a range of 2.5-100 kDa for a total of 200 laser shots using FlexControl 2.4 software (Bruker Daltonics, Bremen, Germany). Data was then analysed using Lucid system software (BioRad, Hercules, CA, USA).
The following settings were followed in both MS readers: Target m/z 5 kDa, matrix attenuation at 2.5 kDa and mass range between 0-100 kDa. External calibration was performed using protein standards comprised of recombinant hirudin (6.96 kDa), equine cytochrome c (12.23 kDa), equine myoglobin (16.95 kDa), and carbonic anhydrase (29.00 kDa). Mass accuracy (m/ßm) was calculated at ≤ 0.02% throughout the entire experimental mass range. Noise definitions were adjusted to eliminate chemical noise in the low mass range, the area below the detector blinding setting (m/z 2, 500) was excluded. Only peaks with a signal-to-noise (S/N) ratio of ≥ 5 and a valley depth ≥ 3 were considered for clustering. Qualified peaks which were present in ≥ 10% of the spectra were used to generate peak clusters. The mass window for each cluster was set at 0.3% of the peak mass for spectra optimized for low mass (0-30 kDa) and at 2% of the peak mass for spectra optimized for high mass (30-100 kDa). Qualified mass peaks (S/N > 5) within m/z range of 2.5-100 kDa were auto detected. Peak clustering was completed using a second-pass peak selection (S/N > 2, within 0.3% mass window) and estimated peaks added.

Biomarker purification
For marker purification, pooled fractions containing the biomarker of interest were subjected to further 2D fractionation by combining Free Flow Electrophoresis (FFE) and GE.
FFE separation was conducted in the IEF mode (IEF-FFE) using a BD FFE System (BD Diagnostics, Sparks, MD, USA) as described previously (33;34). Separation of pooled SAX-fractionated serum samples was performed across a pH range of 3-10. The separated sample was collected (3.5 ml/fraction) to a 96-well plate with a separation time of 1.5 h. Of the total fractions (n = 96) collected from FFE separations, 88 were analyzed for total protein separation by GE.
Following electrophoresis, the gel was silver stained using a Silverquest staining kit (Invitrogen, Paisley, UK; LC6070) and bands near the size of the putative biomarker APPENDIX A SUPPLEMENTARY DATA www.impactjournals.com/oncotarget/

In-gel Trypsin Digestion
Candidate silver stained bands representing the marker of interest, blank gel pieces from a spot-free region, and reference spots (known marker proteins from the gel ladder) were excised and placed in a 96-well plate. In-gel digestion was performed automatically using a ProGest robotic system (DIGILAB, Holliston, MA, USA) following standard digestion protocols (35). The gel plugs were digested for 4 hours at 37°C and the reaction was stopped by adding 7 μL of 3% formic acid to each well. The supernatant was removed. Extracted peptides were reconstituted in 25 μL 0.1% TFA for tandem mass analysis.

Infusion analysis with high resolution FT-Orbitrap MS 2 Analysis
Infusion experiments were performed on an LTQ Orbitrap Elite system (Thermo Fisher Scientific, Bremen, Germany). The isolated lyophilized tryptic peptide fractions were freshly reconstituted in 1 mL solution containing 2% ACN, 0.5% Formic acid and infused at 150 nL/min onto the Nanospray Flex nESI source (Thermo Fisher Scientific, Wien, Austria) and connected to a 1P-4P coated, 8 m tip × 360 μm OD × 75 μm ID PicoTip™ nESI emitter (New Objective, Dingoes, NJ, USA). The LTQ-Orbitrap settings were as follows: spray voltage 1.5 kV; full MS mass range: m/z 300 to 2 000, operated in positive ion mode with data-dependent acquisition. A single full-scan MS in the Orbitrap (120 000 resolution, 300-2 000 m/z) was followed by six data-dependent MS 2 scans for precursor ions above a threshold ion count of 10 000, using the multipole device (HCD; higher collision energy dissociation) with the resolution set to 7 500 and 45% normal collision energy. The data files (x.raw) were converted into mascot generic files using the MassMatrix File Conversion Tool (Version 2.0; http:// www.massmatrix.net) for input into the Mascot searching algorithm (Matrix Science). The data files were searched against SwissProt (v. 2010_06) NCBInr (v. 20080527) using the following search criteria: tryptic peptides with up to one missed cleavage and carbamidomethylation of cysteines and oxidation of methionines, which were set as variable modifications.

LC MS 2 analysis
Peptides were analyzed by LC MS/MS using a Surveyor LC system and LCQ Deca XP Plus (ThermoScientific). Briefly, peptides were resolved by reverse phase chromatography (Biobasic column, ThermoScientific; 180 μM × 15 mm) over a 30 minute ACN gradient at a flow rate of 2 μL/minute. Peptides were ionized by electrospray ionization (ESI) and were subjected to product ion MS 2 analysis. Mass accuracy, resolution and sensitivity of the ESI-MS 2 component were confirmed with the direct infusion of glufibrinopeptide (2.5 pMol/μL). Optimum LC MS/MS system performance for retention time reproducibility, mass accuracy, sensitivity, and protein sequence coverage were verified by the peptide analysis of a BSA tryptic digest (500 pg/μL). BSA quality control checks were performed before and after analysis of experimental samples. The data extraction and processing tasks were the same with those of the infusion FT-Orbitrap experiment.

Biomarker verification by immunoprecipitation (IP) MS
Serum fractions containing ANX A3 were immunodepleted using the μMACS protein A/G microbeads kit (Miltenyi Biotec, Bergisch Gladbach, Germany; 130-042-601) according to the manufacturer instructions. In brief, 2 μL of monoclonal ANX A3 antibody (Santa Cruz Biotechnology, Santa Cruz, CA, USA; sc-134260) was added to 100 μL of protein A/G and then mixed with 200 μL of sample before incubation on ice for 30 minutes. The mix was loaded onto a magnet mounted μMACS column followed by 4 washes with PBS-T. Purified protein was eluted by ACN/TFA/ Isopropranol (16.7/0.1/33.3%-50%).

ANX A3 ELISA validation
ANX A3 measurements in serum fractions and cell lysate/media were performed using commercially available ANX A3 sandwich ELISA kits according to the manufacturers' protocol (USCN Life Science, E94786Hu and CUSABIO Biotech Co. CSB-E12157 h). The total protein concentration of each cell line was determined using a BCA protein assay (Pierce, Thermo Scientific, Rockford, IL, USA) and equal protein quantities were used in the analyses. All samples and standards were run in duplicate with absorbance measured on a ThermoMax plate reader (Molecular Devices, Sunnyvale, CA, USA) and data analyzed using GraphPad Prism 5 software. Mann Whitney U-Tests were used to analyze all ELISA results and a P < 0.05 reflected a significant difference between groups.

Breast tissue Immunohistochemistry
Immunohistochemistry was carried out on 4 μm thick tissue sections cut from formalin fixed paraffin embedded tissue blocks using a Microm HM 325 microtome (Thermo Scientific, Walldorf, Germany) and mounted onto Superfrost ® Plus coated slides (Thermo Scientific). Sections were incubated with monoclonal; ANX A3 antibody (tgc Biomics, Mainz, Germany; clone tgc7 ProVII5C5). Immunohistochemistry was performed using a Bond-Max fully automated immunostaining machine with Bond reagents (Leica Microsytems, Milton Keynes, UK). Sections were dewaxed, pretreated, stained and counterstained using the standard Bond protocol. Slides were pretreated using the Bond ER2 protocol (20 minutes of EDTA). Primary antibody was then applied at a working dilution of 1:1 000, with an incubation period of 20 minutes. A peroxidase blocking step was included following primary antibody incubation. The Bond autostainer used a biotin-free, polymeric horseradish peroxidase-linked antibody conjugate detection system (Bond™ Polymer Refine Detection System, Leica Microsystems, Milton Keynes, UK) with a DAB-chromogen. The slides were counterstained with haematoxylin and mounted under glass coverslips in Pertex (Histolab® products, Gothenburg, Sweden) and imaged using a Leica CV5330 machine (Leica Biosystems, UK).
The frequency of positive ANX A3 staining was assessed using a four point scoring system: 0 for no positive cells, 1 for 1-25% of positive cells, 2 for 26-50% of positive cells and 3 for 51-100% of positive cells. In addition, an intensity score of positive staining was also used: 0 (negative staining), 1 (weak, light beige staining), 2 (moderate, brown staining), and 3 (strong, brown staining). An ANX A3 expression index was calculated as follows: expression index = Positive staining score X Intensity score. High ANX A3 expression was defined as an expression index score of > 4, and low expression defined as an expression index score of < 4. Two independent pathologists, blinded with respect to the clinic-pathological information, performed the scoring. The X 2 test was used to evaluate differences in ANX A3 expression between two groups of tissues.

Breast cells, lysates and media analysis
The six neoplastic human breast cell lines (HUMA121, MCF7, T47D, MDA MB231, SKBR3 and ZR 75-1) that were used in this study were cultured in appropriate media and incubated in a humidified atmosphere containing 5% CO2 at 37°C. The HUMA121 cells were cultured in RPMI (Gibco BRL, NY, USA) with 10% FBS, the remaining cell lines were cultured in Dulbecco's Modified Eagle Medium (DMEM; Gibco BRL, NY, USA) supplemented with 10% FBS. The media was removed once cells reached 70-80% confluence; cells were rinsed twice with 10 ml of PBS and then 30 ml of chemically defined complete hamster ovary (Gibco BRL, NY, USA) media was added followed by further 24 hours incubation. The conditioned media were collected and dialyzed for 48 hours in ultrapure water (with one water change after 24 hours) using a 2 kDa cut off dialysis cassette (Pierce, Thermo Scientific, Rockford, IL, USA). The cells were trypsinized and counted before being lysed with RIPA buffer. Protein concentrations were determined using a BCA kit (Pierce. Cat. No 23227). Equal total protein concentrations from conditioned media and cell lysates were analyzed by ANX A3 ELISA.

Transwell migration assay
BD bioscience transwell assays were used as per the manufacturer's instructions. Briefly, wells and inserts were equilibrated by adding DMEM medium supplemented with 10% FCS at least 30 minutes prior to cell seeding. MCF7 cells were seeded into inserts in DMEM medium supplemented with 10% FCS at a density of 5 × 10 5 cells per transwell insert so that a fully confluent cell layer was formed. After cells had adhered (approximately 6 hours post seeding) medium in inserts was removed and replaced with DMEM with no FCS. At indicated time points cells within the transwell and those that had migrated to the underside of the cell permeable membrane were washed with PBS and fixed with ice cold acetone:methanol 1:1 fixative. The cells within the transwell were stained with eosine solution and removed using a cotton bud and thorough removal was confirmed under a microscope. Cells on the underside of the of the cell permeable membrane were washed, stained with DAPI and representative fields of view were imaged using an Olympus CKX41 fluorescence microscope and cells counted using ImageJ software.

Proliferation assay
Cells were seeded at density of 5000 per well of 24 well plates at indicated time points media was removed, cells were washed once with PBS and then fixed with ice cold acetone:methanol 1:1 fixative. Cells were then stained with DAPI, representative fields of view were imaged using an Olympus CKX41 fluorescence microscope and cells counted using ImageJ software.