Project PXD003644

PRIDE Assigned Tags:
Biomedical Dataset
Download Project Files
Project Protein Table
Project Peptide Table
Visualize in PRIDE Inspector
Follow the next three steps to open your selected project or assay in PRIDE Inspector:

  • 1.

    Download, uncompress and open PRIDE Inspector
  • 2.

    Click in the magnifier on the left top corner, paste the project or assay that you would like to open in the search box, and hit search
  • 3.

    Click in the corresponding "Download" button to download the files and visualize them

Summary

Title

Protein profiling and identification of new biomarkers of brain ischemia by MALDI-imaging-mass-spectrometry

Description

Introduction The identification of proteins involved in brain ischemia might allow the discovery of new putative biomarkers or potential therapeutic target of one of the most important neurological disorders. MALDI Imaging-Mass-Spectrometry (IMS) is a powerful technique that allows the visualization of protein distribution along a tissue without labeling. Our aim is to study the distribution of proteins along mouse brain after an ischemic insult and to identify relevant proteins involved in brain ischemia. Materials and methods We occluded the middle cerebral artery (60min) of C57BL/6J mice (n=4) with 24h of reperfusion. Brain slices were analyzed by MALDI-TOF and mass spectra from infarct (IC) and contralateral (CL) regions were compared using ClinProTools. Relevant m/z were selected after PCA analysis and their ion distribution maps were analyzed by FlexImagin3.0. Protein identification was conducted on mouse brain homogenates through a bottom-up approach consisting on complementary fractionations based on tricine gels and RP-HPLC. The identifications were confirmed by immunohistochemistry. Results We identified 102 m/z with different abundances between IC and CL (p<0.05), from which 21 m/z were selected by PCA as more relevant. Thirteen of them were found increased in the infarct region and 4 m/z showed a discrimination higher than 90% between IC and CL. After bottom-up identification, immunohistochemistry analysis confirmed increased ATP5i and COX6C expressions, and decreased UMP-CMP kinase in IC compared to CL. Conclusions We identified for the first time by MALDI-IMS several m/z peaks with different abundances between IC and CL. These proteins involved in brain ischemia might represent potential diagnostic biomarkers or target molecules for neurological recovery.

Sample Processing Protocol

Brain tissue processing, MALDI mass spectrometry and image analysis Brains (n=4) were isolated and immediately frozen in isopentane (Sigma, USA) at -40oC during 30 sec and stored at -80oC. Ten µm-thick coronal brain sections (+0.74 to +0.98 from bregma) were obtained using a criostat (Microm HM 505E, CBIS, USA) and mounted onto superconductive glass slides covered with ITO coating (Bruker Daltonics, Germany). Tissue preparations were dried with vacuum pump (Millipore, USA) at 600 mmHg for 15 minutes, washed for two minutes in 70% Ethanol and once in 96% Ethanol and then stored in a dessicator under vacuum. In order to be analyzed by MALDI TOF MS, samples werec overed with sinapinic acid (10mg/mL 60% acetonitrile and 0.2% trifluoroacetic acid (TFA)) as MALDI matrix sprayed using ImagePrep device (Bruker Daltonics). The optimal parameters of the device were set (dry time, incubation time and thickness). Finally the slides were mounted into MTP slide adapter (Bruker Daltonics) and transferred to the MALDI-TOF mass spectrometer. These four biologically independent samples were analyzed by MS. We acquired the IMS spectra in linear ion positive mode in an UltrafleXtreme MALDI-TOF/TOF mass spectrometer. MS data were averaged from 300 consecutive laser shots with a frequency of 2000 shot/s. The raster spatial resolution was 100 µm. After the MS analysis hematoxylin & eosin (H&E) staining was performed in each brain section and they were used as reference for colocalization to select the regions of interest (ROI). Laser Micro-Dissection and protein extracts We mounted consecutive sections of 12 µm thickness obtained from the brains analyzed by the MALDI imaging alternating PEN-membrane glass slides (Leica, Germany) and superfrost glass slides (Thermoscientific). From the samples mounted on PEN-membrane slides we cut sections of the infarct and the contralateral area by means of laser microdissection (Leica, Germany) and recovered them separately in 25 µL of 0.1% Rapigest SF surfactant (Waters). Samples mounted on glass slides were used for infarct core colocalization. Samples were sonicated at 80% amplitude for intervals of 10 sec for a total of 2 min at 4oC. Then, samples were centrifuged at 13,000g at 4ºC, the supernatants were recovered, supplemented with 4.5 µL of aprotinin (Sigma) and 10 µL of phenylmethylsulfonyl fluoride (Sigma) as protease inhibitors, and finally stored at -80oC until use. The obtained homogenates were fractionated on a 10-20% Tris-tricine gel (Biorad, USA) and 4 sections of each sample, infarct core and contralateral area, were cut from the gel. The gel bands, corresponding to the proteins with an apparent molecular weight bellow 25 kDa, were digested in-gel with trypsin and subsequently analyzed by MS/MS. Tissue homogenates and reverse phase HPLC For reverse phase HPLC separations, mechanical homogenates of infarct core and contralateral healthy areas of mouse brain were obtained in 0.2% of TFA and 60% of acetonitrile (ACN). Samples were sonicated using the previous parameters. Insoluble material was removed by centrifugation 13,000g at 4ºC, protease inhibitors were added into the recovered supernatant and samples were stored at -80oC until use. Proteins from the soluble material were separated using and Alliance HPLC system (Waters, USA) and a C4 widepore 3.6 um; 4.6x100 mm Aeris reverse phase (RP) column (Phenomenex). Solvent A was 0.1% (TFA) in water and solvent B was 0.1% TFA in acetonitrile. A linear gradient from 10 to 30% for 1 min, followed by a second step from 30 to 90% for 60 min was applied. Fractions were collected every 2 min, dried using SpeedVac (ThermoScientific, USA) and reconstituted in 30 µL of H2O 0.1 % TFA mixed at 1:1 volume ratio with sinapinic acid (SA) matrix (10 g/ml SA in 1:2 H2O:ACN, 0.2 % TFA). Samples were mounted onto a ground steel plate, let to cristalize and mass spectra were obtained by MALDI MS analysis (lineal mode, 24kV) to check for the presence of m/z of interest.

Data Processing Protocol

Data quantification and statistical MALDI-IMS data analysis After MALDI analysis, we used FlexImaging 3.0 (Bruker Daltonics, Germany) software to obtain ion density maps and to define the ROIs on each brain section. One ROI was defined on the infarct core at the cerebral cortex and another ROI on the contralateral healthy area based on H&E staining. Mass spectra from each ROI were combined to a single dataset and exported to ClinProtools 3.0 (Bruker Daltonics, Germany). The analysis of the mass spectra was focused on the mass range 2000-24000 m/z. A 10% of minimal baseline width was applied as top hat baseline substraction and a factor of 2 was used for data reduction. For recalibration, a maximal peak shift of 2000 ppm was defined and a 20% match to calibrant peaks was set. Not recalibratable spectra were excluded from analysis. Signal to noise ratio for data acquisition was set to 5. An average spectrum from each ROI and the average peak list were obtained. To check the homogeneity of the spectra set we carried out principal components analysis (PCA) with the Matlab algorithm and used the 3 primary PCAs for the 3-D scaling representation. To select the most relevant m/z peaks from the average peak list we used the loading plot coupled to each PCA graphic. We also obtained the area under the curve (AUC) to assess the discrimination of contralateral (CL) and infact (IC) of each m/z peak. Bottom-up nano LC-MS/MS The samples obtained after laser microdissection and resolved by tricine gels, as well as fractions from RP-HPLC that contained m/z of interest were digested with trypsin. Peptides obtained from in-gel and in-solution digestions were subsequently analyzed on a LTQ Velos-Orbitrap mass spectrometer (ThermoFisher Scientific, Bremen, Germany) coupled to a nano-HPLC system (Proxeon, Odense, Denmark). Briefly, peptide mixtures were initially concentrated on an EASY-column, 2 cm long, 100 μm internal diameter (id), and packed with ReproSil C18, 5 μm particle size (Proxeon, Denmark), and subsequently loaded onto an EASY-column, 75 μm id, 10 cm long, and packed with ReproSil-Pur C18-AQ, 3 μm particle size (Proxeon, Denmark). An ACN gradient (5− 35% ACN/0.1% formic acid in water, in 90 min, flow rate of 300 nL/min) was used to elute the peptides through a stainless steel nanobore emitter (Proxeon, Denmark) onto the nanospray ionization source of the LTQ Velos-Orbitrap mass spectrometer. MS/MS fragmentation spectra (200 ms, 100− 2800 m/z) of 20 of the most intense ions, as detected from a 500 ms MS survey scan (300−1500 m/z), were acquired using a dynamic exclusion time of 20 s for precursor selection and excluding single-charged ions. Precursor scans were acquired in the Orbitrap analyzer at a mass resolution of 30,000. MS/MS spectra were acquired at the LTQ Velos analyzer using a relative collision energy of 35. An intensity threshold of 1000 counts was set for precursor selection. Bioinformatics for protein identification In order to match the relevant identified m/z peaks after MALDI IMS with the molecular masses of the proteins identified by bottom-up analysis, we used the PeptideMass tool from ExPasy (http://web.expasy.org/peptide_mass/)[23]. This tool allows the generation of the theoretical peptide masses of the input proteins. By selecting the “no cutting” option as enzyme all the putative protein fragments resulting after their processing were displayed (i.e. pro-form, signal sequences, transit peptides, active peptide, etc). All known post-translational modifications and variants of the proteins were considered and all other parameters were used as default. The identified MALDI-IMS m/z peaks were matched with the masses of proteins and/or peptides retrieved by PeptideMass allowing a mass shift of 2000 ppm.

Contact

Víctor Llombart, Vall Hebrón Institute of Research
Joan Montaner, vall hebron institute of research ( lab head )

Submission Date

16/02/2016

Publication Date

14/12/2016

Tissue

brain

Quantification

Not available

Assay count

11

Publication

    Llombart V, Trejo SA, Bronsoms S, Morancho A, Feifei M, Faura J, García-Berrocoso T, Simats A, Rosell A, Canals F, Hernández-Guillamón M, Montaner J. Profiling and identification of new proteins involved in brain ischemia using MALDI-imaging-mass-spectrometry. J Proteomics. 2016 Nov 22;152:243-253 PubMed: 27888142

Assay

Page 1 2
Page size 10
Showing 1 - 10 of 11 results
# Accession Title Proteins Peptides Unique Peptides Spectra Identified Spectra View in Reactome
1 62059 fraction 7 (24 min RT) 325 2035 769 11805 1865
2 62061 tricine gel section IC4 62 284 93 2668 137
3 62060 tricine gel section IC3 270 817 459 3577 627
4 62053 fraction 8 (26 min RT) 533 3826 1569 14077 3389
5 62054 fraction 9 (28 min RT) 655 4270 1995 13731 3816
6 62051 tricine gel section IC2 632 1988 1158 5081 1401
7 62052 tricine gel section IC1 628 2001 1238 5176 1493
8 62057 tricine gel secgtion CL4 52 289 107 3454 165
9 62058 tricine gel section CL3 205 648 328 3558 495
10 62055 tricine gel section CL2 604 2316 1311 5618 1782