Article pubs.acs.org/jpr
Comprehensive Comparative and Semiquantitative Proteome of a Very Low Number of Native and Matched Epstein−Barr-VirusTransformed B Lymphocytes Infiltrating Human Melanoma Margarita Maurer,† André C. Müller,‡ Katja Parapatics,‡ Winfried F. Pickl,§ Christine Wagner,† Elena L. Rudashevskaya,‡,∥ Florian P. Breitwieser,‡ Jacques Colinge,‡ Kanika Garg,† Johannes Griss,† Keiryn L. Bennett,*,‡ and Stephan N. Wagner*,† †
Division of Immunology, Allergy and Infectious Diseases, Department of Dermatology, Medical University of Vienna, Waehringer Guertel 18-20, A-1090 Vienna, Austria ‡ CeMM Research Center for Molecular Medicine of the Austrian Academy of Sciences, Lazarettgasse 14, AKH BT 25.3, A-1090 Vienna, Austria § Division of Cellular Immunology and Immunohematology, Institute of Immunology, Center for Pathophysiology, Infectiology and Immunology, Medical University of Vienna, Kinderspitalgasse 15, A-1090 Vienna, Austria S Supporting Information *
ABSTRACT: Melanoma, the deadliest form of skin cancer, is highly immunogenic and frequently infiltrated with immune cells including B cells. The role of tumor-infiltrating B cells (TIBCs) in melanoma is as yet unresolved, possibly due to technical challenges in obtaining TIBCs in sufficient quantity for extensive studies and due to the limited life span of B cells in vitro. A comprehensive workflow has thus been developed for successful isolation and proteomic analysis of a low number of TIBCs from fresh, human melanoma tissue. In addition, we generated in vitro-proliferating TIBC cultures using simultaneous stimulation with Epstein−Barr virus (EBV) and the TLR9 ligand CpG-oligodesoxynucleotide (CpG ODN). The FASP method and iTRAQ labeling were utilized to obtain a comparative, semiquantitative proteome to assess EBV-induced changes in TIBCs. By using as few as 100 000 B cells (∼5 μg protein)/sample for our proteomic study, a total number of 6507 proteins were identified. EBV-induced changes in TIBCs are similar to those already reported for peripheral B cells and largely involve changes in cell cycle proliferation, apoptosis, and interferon response, while most of the proteins were not significantly altered. This study provides an essential, further step toward detailed characterization of TIBCs including functional in vitro analysis. KEYWORDS: tumor-infiltrating B cells, TIBC, EBV-transformed B cells, iTRAQ, FASP, ultrasensitive proteomics, melanoma, Orbitrap
■
melanoma.8−13 In contrast, B cells can also engage in negative regulation of immune responses through secretion of IL-10 or TGF-β1.14 Moreover, TIBCs are also an important source of tumor-promoting cytokines such as lymphotoxin β (LT β), as recently exemplified in a mouse model of castration-resistant prostate cancer.15 In addition, B cells may also promote breast cancer metastasis by converting resting CD4+ T cells to Foxp3+ T-regulatory (Treg) cells by secretion of TGF-β,16 and B1 lymphocytes are suggested to have metastatic-promoting functions in melanoma.17 Recently, Lindner et al. identified Granzyme B expressing regulatory tumor-infiltrating B cells that are able to suppress antitumor responses.18 In addition, a seven biomarker signature consisting of Bax, Bcl-X, PTEN, COX-2, loss of β-Catenin, loss of MTAP, and presence of CD20+ Blymphocytes was found to be an independent negative
INTRODUCTION Solid tumors, including melanoma, are frequently infiltrated by various types of immune cells, including cytotoxic T cells (CD8+), regulatory T cells (Tregs), T-helper cells (CD4+), macrophages, natural killer (NK) cells, dendritic cells (DC), myeloid-derived suppressor cells (MDSC), and B cells.1,2 Whereas the tumor-promoting or antitumor activity of most of these infiltrating immune cells has been extensively studied in various types of cancer (for a review, see ref 2), the role of B cells in the context of tumor biology has long been neglected and is far from resolved.3 In general, B cells are considered to be positive regulators of immune responses. This includes the response against tumors by secreting tumor-specific antibodies (Ab) by serving as local antigen presenting cells (APCs) and by enhancing T-cell-mediated immune responses.4−7 Indeed, several studies report a positive prognostic value of tumorinfiltrating B cells (TIBCs) in various cancer types, including breast, ovarian, head and neck cancer and cutaneous © 2014 American Chemical Society
Received: December 19, 2013 Published: April 25, 2014 2830
dx.doi.org/10.1021/pr401270y | J. Proteome Res. 2014, 13, 2830−2845
Journal of Proteome Research
Article
Figure 1. Schematic of the experimental workflow. Two independent experiments were performed with tumor-infiltrating B cells (TIBCs) obtained from fresh, human melanoma tumor biopsies from two different patients. TIBCs were isolated using CD19-antibody-coupled magnetic beads and counted. An aliquot of the cells was immediately used for in vitro transformation using the Epstein−Barr virus (EBV). Another aliquot of the cells was immediately snap frozen as a dry pellet. For comparative and semiquantitative proteomic analysis of native and EBV-transformed TIBCs, cells were lysed and the same amount of protein was processed using the filter-aided sample preparation (FASP) technique. Tryptic peptides were chemically labeled with the iTRAQ reagents and analyzed using two-dimensional gel-free liquid chromatography mass spectrometry on a hybrid LTQ Orbitrap Velos mass spectrometer. Two independent iTRAQ experiments were performed using native and matched EBV-transformed TIBCs from two different patients.
studies.38−42 Almost all of these studies, however, have been performed with EBV-transformed peripheral blood B cells or with lymphoma cell lines. To the best of our knowledge, there are only two studies describing successful in vitro expansion of tumor-infiltrating B cells using EBV transformation.4,43 It is, however, as yet unclear if or how closely EBV-transformed TIBCs resemble the phenotype of native (i.e., not EBVtransformed) TIBCs. For any study employing EBV-transformed B cells, it has to be taken into consideration that upon EBV infection B cells undergo transformation, which might significantly alter the biology of the cell. Indeed, there are several studies that question the use of LCLs as a surrogate for freshly isolated B cells.44−46 In this study we designed a comprehensive approach allowing extensive characterization of TIBCs using high-end proteomic techniques as well as functional in vitro studies. In particular, we have developed a workflow that entailed: (i) efficient isolation of tumor-infiltrating B cells from human melanoma tissue that is directly compatible with subsequent comparative and semiquantitative MS-based proteomics and (ii) efficient in vitro EBV transformation of a very low number of isolated tumor-infiltrating B cells enabling comprehensive functional studies. We have recently shown that the FASP protocol is the method of choice for proteomic analysis of a very low number of cells.47 TIBCs are difficult to obtain in large quantities. Thus, we have (iii) applied FASP in combination with iTRAQ labeling to obtain a semiquantitative and comparative proteome of native versus EBV-transformed TIBCs using as few as 105 TIBCs (∼5 μg protein). We show that TIBCs are present in human melanoma tissue and that these cells can be isolated from fresh suspensions of melanoma tumors using CD19 antibody-coupled magnetic beads with subsequent removal of the beads and antibodies. This approach enabled MS-based proteomics without the presence of disturbing antibodies, which may be present in high quantities and would therefore mask low-abundance proteins present in TIBCs. In addition, we provide evidence that the isolated TIBCs can be efficiently expanded in vitro by EBV transformation using as few as 1000 tumor-infiltrating B cells. We analyzed native and matched EBV-transformed TIBCs by comparative and semiquantitative proteomics to assess how closely EBV-transformed TIBCs resemble the phenotype of
predictor for overall and recurrence-free survival in patients with metastatic melanoma.19 Consistent with these contrary findings are observations on antitumor20−22 and tumor-promoting effects23 of anti-CD20 treatment in vivo. Thus, whether tumor-infiltrating B cells either promote or inhibit cancer-initiation, -progression, or -metastases remains controversial. Such effects may depend on various factors including the cancer type, the presence of specific B cell subsets,18 the interaction with other tumorinfiltrating (immune) cells, and the interaction with the tumor cells themselves.6 Despite the apparent crucial role of TIBCs in cancer biology, there are few functional studies employing tumor-infiltrating B cells. Such studies would have tremendous translational impact and could lead to the development of novel (immuno-) therapies for cancer, including melanoma. Obvious reasons for the lack of such functional studies may be that TIBCs are difficult to obtain in large quantities and that the life-span is limited when isolated from the tumor and cultured in vitro without stimulation (unpublished observations). It is therefore necessary to develop methods that allow the efficient isolation and in vitro expansion of even very low numbers of tumorinfiltrating B cells to allow a comprehensive phenotypic and functional analysis of these cells. Epstein−Barr virus (EBV) is a ubiquitous human gammaherpes virus infecting more than 90% of the human population worldwide.24 The virus persists in the host throughout life and normally resides in memory B cells as a harmless passenger.25 EBV can, however, transform B cells, leading to certain malignancies including Burkitt’s lymphoma (BL) and Hodgkin’s lymphoma.26 In vitro, EBV-mediated transformation of B cells is a long-known and efficient method to expand B cells, leading to continuous proliferation of the cells as lymphoblastoid cell lines (LCLs).27−32 Efforts have been made to optimize the method for transformation of B cells using low amounts of cryo-preserved blood33 or very low B-cell numbers using simultaneous B cell stimulation by the toll-like receptor 9 (TLR9) ligand CpG-oligodesoxynucleotide (CpG ODN 2006) and EBV.34 Numerous studies have been performed in vitro with EBV-transformed B cells to study the cellular targets of EBV, EBV latency, virus-mediated B cell proliferation and tumorigenesis,35 DNA damage/repair and apoptosis.36,37 In addition, LCLs have already been used in proteomic 2831
dx.doi.org/10.1021/pr401270y | J. Proteome Res. 2014, 13, 2830−2845
Journal of Proteome Research
Article
Figure 2. EBV-transformed TIBCs in culture 3 weeks after EBV transformation. After isolation of CD19+ TIBCs from human melanoma tissue, an aliquot of the cells was used for simultaneous stimulation with CpG ODN and EBV transformation in vitro (as described34). One ×103 CD19+ TIBCs were seeded in a single well of a 96-well U-bottom microtiter plate together with 1 × 105 gamma irradiated (2 × 30 Gy) autologous PBMCs, 200 ng/mL cyclosporin A, 1 μg/mL CpG ODN and mixed with an equal volume of EBV-containing supernatant. On day 14, the cells were restimulated with 1 μg/mL CpG ODN and 50 U/mL IL-2. After 21 days, the cells were cultivated without further stimulation. Note the clump formation of the proliferating TIBCs 3 weeks after successful transformation with EBV (shown insert). (A) 40 × magnificaton; (B) 100 × magnification.
passed through a 70 μm cell strainer (BD Falcon, Bedford, MA) to remove any large particles from the single cell suspension. Sorting of the tumor-infiltrating B cells was performed using Dynabeads CD19 Pan B (Life Technologies, Grand Island, NY) according to the instructions supplied by the manufacturer. In brief, 50 μL of Dynabeads were washed with PEB buffer (phosphate-buffered saline (PBS), pH 7.2, 2 mM EDTA, and 0.5% bovine serum albumin (BSA)) and added to 1 mL of the tumor single cell suspension in RPMI + 10% FCS and incubated at 4 °C for 20 min under continuous rotation. The CD19+ B cells were separated from the negative fraction by placing the whole cell suspension in a magnetic particle concentrator (Dynal MPC-S, Invitrogen) for 2 min. The beadbound cells were washed three times to increase the purity of the CD19+ fraction and resuspended in 500 μL of medium (RPMI + 1% FCS). To remove the Dynabeads and the antibodies from the sorted B cell fraction, the cells were incubated at RT for 45 min under continuous rotation with DETACHaBEAD CD19 reagent (Life Technologies, Grand Island, NY) according to recommendations of the manufacturer. The sorted B cells were counted using a hemocytometer, and aliquots of the sorted tumor-infiltrating B cells were used for: (i) EBV transformation (Figure 2 and Table 1); (ii) flow
native TIBCs. In addition to the expected EBV-mediated induction of pathways associated with cell cycle progression and downregulation of apoptosis pathways, we found that EBV transformation may also impact on other processes/pathways such as glycolysis, cell adhesion, migration, and differentiation. Most of the proteins, however, were not significantly affected by EBV transformation, suggesting that EBV-transformed TIBCs may indeed closely match the phenotype of native TIBCs. In summary, we present a comprehensive workflow that enables a deep and thorough characterization of tumorinfiltrating B cells.
■
MATERIAL AND METHODS The workflow of the experiments is shown in Figure 1. Isolation of Tumor-Infiltrating B Cells from Human Melanoma Tissue
For the proteomic experiments, tumor-infiltrating B cells were isolated in two independent experiments from human melanoma tissues from two different patients. The melanoma samples analyzed in this study were collected and prepared under a local ethics committee-approved protocol (969-2010). The patient provided written, informed consent. Immediately after surgical resection, the metastatic melanoma tissue was dissociated into single cells using the tumor dissociation kit (Miltenyi Biotec, Bergisch Gladbach, Germany) in conjunction with the gentleMACS dissociator (Miltenyi Biotec), enabling both mechanical and enzymatic dissociation of the tumor tissue. In brief, the tissue was minced into 2−4 mm pieces and transferred into a gentleMACS C tube (Miltenyi Biotec) containing 4.7 mL of RPMI (Invitrogen, Carlsbad, CA) and an enzymatic mixture consisting of 200 μL of solution I, 100 μL of solution II, and 25 μL of solution III. (Solutions I−III are provided with the tumor dissociation kit.) The C tube containing the tissue and the enzymatic mixture was attached to the sleeve of the gentleMACS dissociator and sequentially run with the preinstalled programs h_tumor_1, h_tumor_2, and h_tumor_3. Between each program the C tube containing the tumor tissue and the enzymatic mixture was incubated at 37 °C for 30 min under continuous rotation. After dissociation, the cell suspension including the tumor-infiltrating B cells was
Table 1. Number of Sorted TIBCs and Number of TIBCs Used for EBV Transformation
patient
localization
no. of CD19 sorted TIBCs
PA/7 experiment 1 PA/8 experiment 2
s.c. s.c.
2.4 × 105 5.5 × 105
no. of CD19+ TIBCs/ well used for EBV transformation 2.5 × 104 2.5 × 104
cytometric analysis (Figure 3); and (iii) immediate preparation of cell pellets for proteomic analysis by centrifugation using an Eppendorf centrifuge (5415R) at 400g for 4 min at RT. The supernatant was discarded, and cell pellets were snap-frozen and stored at −80 °C until required. 2832
dx.doi.org/10.1021/pr401270y | J. Proteome Res. 2014, 13, 2830−2845
Journal of Proteome Research
Article
Figure 3. Flow cytometric analysis of CD19+ sorted TIBCs used for the proteomic analyses. After isolation of TIBCs from human tumor biopsies of (A) PA/7 and (B) PA/8, an aliquot of the cells was stained using a PE-conjugated anti-CD19+ antibody. Using a detachabead solution, the beads and antibodies were removed from the cells after sorting. An anti-CD19 antibody was used for subsequent FACS staining. Left panel: Dot plot showing SSC-H/FSC-H and the gating excluding cell debris and apoptotic cells. Right panel: blue: CD19+-staining, red: matched isotype control staining. PE, phycoerythrin; CD19, cluster of differentiation 19 (a pan B cell marker).
EBV Transformation of B Cells Isolated from Human Melanoma Tissue
Isolation of Peripheral Blood Mononuclear Cells from Whole Blood
EBV transformation of tumor-infiltrating B cells was performed essentially according to the method described by Fraussen et al.34 A number of 2.5 × 104 CD19+ tumor-infiltrating B cells were resuspended in 100 μL of RPMI supplemented with 10% FCS and penicillin/streptomycin and cultured in U-bottom 96well plates (Nunc, Roskilde, Denmark). The tumor-infiltrating B cells were transformed for 2 weeks in the presence of 1 × 105 irradiated (2 × 30 Gy) autologous peripheral blood mononuclear cells (PBMCs) serving as feeder cells, 1 μg/mL CpG (ODN 2006, 50-tcgtcgttttgtcgttttgtcgtt-30, InvivoGen, France), EBV-containing supernatant from the marmoset B cell line B95-8, and cyclosporine A (200 ng/mL). The cells were cultivated at 37 °C, 5% CO2 for 14 days. At day 14, the cells were restimulated with 1 μg CpG ODN2006 and 50 units/mL IL-2. At day 21, medium was changed, and the cells were further cultivated without any stimuli. The EBV-transformed B cells were passaged several times and pelleted by centrifugation using an Eppendorf centrifuge (5415R) at 400g for 4 min at RT. The supernatant was discarded, and cell pellets were snapfrozen and stored at −80 °C until required.
The blood samples were collected in BD Vacutainer CPT cell preparation tubes. PBMC preparation was performed according to the manufacturers instructions. In brief, tubes were inverted 8−10 times and centrifuged at RT at 1500g for 20 min. PBMCs were transferred to a new tube and washed twice in PBS and centrifuged at 400g for 15 min at RT. The number of isolated PBMCs was counted using a hemocytometer. Isolated PBMCs were irradiated (2× 30 Gy), and 105 PBMCs per well of a 96well plate were used as feeder cells for EBV transformation of tumor-infiltrating B cells as described in ref 34. Flow Cytometric Analysis
The purity of the CD19-sorted tumor-infiltrating B cells was assessed by flow cytometry. The cells were stained with an antibody against CD19, a well-established pan B cell marker. In brief, 5 × 104 cells were resuspended in 100 μL of PEB buffer and incubated for 10 min at 4 °C in the dark with an phycoerythrin (PE)-conjugated CD19 antibody (Miltenyi Biotec) or with the corresponding isotype control (Miltenyi Biotec). Cells were washed by adding 1 mL of PBS and centrifuged at 300g for 10 min at 4 °C. The supernatant was discarded, and the cells were resuspended in 100 μL of PBS before analysis on the FACS Calibur (Becton Dickinson, San Jose, CA). 2833
dx.doi.org/10.1021/pr401270y | J. Proteome Res. 2014, 13, 2830−2845
Journal of Proteome Research
Article
Cell Lysis
2D-RP/RP Liquid Chromatography Mass Spectrometry
Samples were lysed in 60 μL of freshly prepared lysis buffer. Ten microliters were used for the protein assay. Lysis buffer contained 50 mM HEPES (pH 8.0), 2% SDS, 0.1 M DTT, 1 mM PMSF, and protease inhibitor cocktail (SIGMA-Aldrich, St. Louis, MO), and lysis occurred at RT for 20 min. After heating to 99 °C for 5 min and cooling to RT, the cell lysate was sonicated using a Covaris S2 high-performance ultrasonicator. The lysate was centrifuged at 20 000g for 15 min at 20 °C, and the protein extract was collected from the supernatant. Total protein content of the whole cell lysates was determined using the BCA protein assay kit (Pierce Biotechnology, Rockford, IL) following the recommendations of the manufacturer. The limited size of the samples necessitated that the assay was performed in a 96-well plate format using 10 μL of each lysate and standard protein. BSA (Pierce Biotechnology) was used as the standard protein. On the basis of sample availability, the amount of protein used for the proteomic analyses was 7 μg/sample (i.e., per iTRAQ label) in experiment 1 and 5 μg/sample (i.e., per iTRAQ label) in experiment 2, respectively (Figure 1).
Two-dimensional liquid chromatography was performed by reverse-phase chromatography at high and low pH, as previously described. In the first dimension, peptides were separated on a Gemini-NX C18 (150 × 2 mm, 3 μm, 110 Å, Phenomenex, Torrance, US) using a 45 min gradient from 5 to 70% acetonitrile containing 20 mM ammonia formate buffer, pH 10, at a flow rate of 100 μL/min, run on an Agilent 1200 HPLC system (Agilent Biotechnologies, Palo Alto, CA). Seventy two time-based fractions were collected and pooled into 50 HPLC vials based on the UV-trace at 214 nm. Samples were acidified by the addition of 5 μL of 5% formic acid, solvent was removed in a vacuum concentrator, and samples were reconstituted in 5% formic acid for a single injection. Mass spectrometry was performed on a hybrid linear trap quadrupole (LTQ) Orbitrap Velos mass spectrometer (ThermoFisher Scientific, Waltham, MA) using the Xcalibur version 2.1.0 coupled to an Agilent 1200 HPLC nanoflow system (dual pump system with one trap-column and one analytical column) via a nanoelectrospray ion source using liquid junction (Proxeon, Odense, Denmark). Technical details of the nanoHPLC and MS conditions used in this study are described in detail elsewhere.50 In brief, the analyses were performed in a data-dependent acquisition mode using a top-10 high-energy collision-induced dissociation (HCD) method for peptide identification plus relative quantitation of iTRAQ reporter ions. Dynamic exclusion for selected ions was 60 s. A single lock mass at m/z 445.120024 was employed.51 The maximal ion accumulation time allowed for MS mode in the Orbitrap was 500 ms, and for HCD, the accumulation time was 200 ms. Automatic gain control (AGC) was used to prevent overfilling of the ion traps. In MS and MS2 modes, AGC was set to 106 and 105 ions, respectively. Peptides were detected in MS and MS2 mode at 30 000 (at m/z 400) and 7500 resolution, respectively. The threshold for switching from MS to MS2 was 2000 counts.
Filter-Aided Sample Preparation (FASP)
FASP was performed using a 30 kDa molecular weight cutoff filter (VIVACON 500; Sartorius Stedim Biotech, Goettingen, Germany) essentially according to the procedure described by Wisniewski et al.48 Fifty microliters of each protein extract was mixed with 180 μL of 8 M urea in 100 mM Tris-HCl (pH 8.5) (UA) in the filter unit and centrifuged at 14 000g for 15 min at 20 °C to remove SDS. Any remaining SDS was exchanged by urea in a second washing step with 200 μL of UA. The proteins were alkylated with 100 μL of 50 mM iodoacetamide for 30 min at RT. Afterward, three washing steps with 100 μL of UA solution were performed, followed by three washing steps with 100 μL of 50 mM TEAB buffer (SIGMA-Aldrich). Proteins were digested with trypsin overnight at 37 °C. Peptides were recovered using 40 μL of 50 mM TEAB buffer, followed by 50 μL of 0.5 M NaCl (SIGMA-Aldrich). Acidified tryptic peptides were concentrated and desalted using C18 spin columns (The Nest group, Southborough, MA).
Data Analysis
The acquired raw MS data files were processed with msconvert (ProteoWizard Library v2.1.2708) and converted into Mascot generic format (mgf) files. The resultant peak lists was searched against the human SwissProt database version v2013.01_20130110 (37 398 sequences, including isoforms as obtained from varsplic.pl) with the search engines Mascot (v2.3.02, MatrixScience, London, U.K.) and Phenyx (v2.5.14, GeneBio, Geneva, Switzerland).52 Submission to the search engines was via a Perl script that performs an initial search with relatively broad mass tolerances (Mascot only) on both the precursor and fragment ions (±10 ppm and ±0.6 Da, respectively). High-confidence peptide identifications were used to recalibrate all precursor and fragment ion masses prior to a second search with narrower mass tolerances (±4 ppm and ±0.025 Da). One missed tryptic cleavage site was allowed. Carbamidomethyl cysteine, N-terminal, and lysinemodified iTRAQ 4-plex were set as fixed modifications, and oxidized methionine was set as a variable modification. To validate the proteins, Mascot and Phenyx output files were processed by internally developed parsers. Proteins with ≥2 unique peptides above a score T1 or with a single peptide above a score T2 were selected as unambiguous identifications. Additional peptides for these validated proteins with score >T3 were also accepted. For Mascot and Phenyx, T1, T2, and T3 were equal to 16, 40, 10 and 5.5, 9.5, 3.5, respectively (p value
Real-Time PCR
RNA isolation was performed using the TRI Reagent (Sigma) according to the recommendation of the manufacturer and as described recently.49 In brief, complementary DNA (cDNA) was synthesized from 1 μg total RNA using Superscript II Reverse Transcriptase (Invitrogen) according to the instructions provided by the manufacturer. qPCR was performed on an ABI PRISM 7700 using TaqMan Gene Expression Assays (Applied Biosystems) according to the protocols provided by the manufacturer (assay IDs: FCER2: Hs00233627_m1, EBI3: Hs01057148_m1, IFI44: Hs00951349_m1, HMGN5: Hs01043329_m1, NDRG1: Hs00608387,). The mRNA of βactin was used as internal control. iTRAQ Derivatization
iTRAQ labeling was performed according to the instructions provided by the manufacturer. Pooled samples were concentrated and desalted with C18 microspin columns (5−60 μg, The Nest Group, Southborough, MA). Eluates were dried in a vacuum concentrator and reconstituted in 2 mM ammonia formate buffer, pH 10 before fractionation at basic pH. 2834
dx.doi.org/10.1021/pr401270y | J. Proteome Res. 2014, 13, 2830−2845
Journal of Proteome Research
Article
