Alzheimer's disease biomarkers discovery by high resolution mass spectrometry: a challenging task of the Diatral project

Transcription

1 Franco-British bilateral symposium on clinical research, October 2012 Alzheimer's disease biomarkers discovery by high resolution mass spectrometry: a challenging task of the Diatral project Franck Vandermoere, CNRS investigator, Institute for Functional Genomics, CNRS UMR5203, INSERM U661 Montpellier

2 SP1 Diatral task: four partner teams Audrey Gabelle Jacques Touchon Sylvain Lehmann Justine Boulanghien Biological samples Séverine Chaumont-Dubel Ramzi El Feghali Nathalie Galéotti Philippe Marin Madeleine Sugano Franck Vandermoere Biological samples Biological samples Identification/validation by LC-HRMS/MS (potential biomarkers) Denis Loyaux Catherine Pech Florence Oury-Donat Laurent Pradier Stephan Kirsch Halim Guerraoui FUI Eurobiomed Identification/validation by LC-HRMS/MS (potential biomarkers) Jeannette Fareh Daniel Laune Franck Molina Biomarkers validated (Ab assays)

3 Introduction on MS peptide mapping MS1 spectrum Consecutive MS1 spectra Isolation, fragmentation Mass/Charge MS intensity MS2 spectrum Quantification Protein sequence database Identification

4 Retention Time Introduction on MS peptide mapping m/z Features characterized by m/z, RT coordinates: accurate mass tag More than peptide signatures in a CSF

5 Method for biomarker discovery in AD: high resolution peptide profiling of CSF samples by FT-MS Log (P values) p-value 1e-01 1e-04 1e-07 1e-10 1e dow n, 487 up 2639 total ( NA) Replicates/sp Disease : 14 AD CSF 2D peptide map Statistical Analyses CSF Depletion of major plasma protein 1D-NanoLC-MS/MS-MS Volcano plot (p-value vs. log fold change) MiceTypeMRL Trypsin Digestion 1D-NanoLC-MS/MS-MS Control : 5DFT/5MS/5ALS Log fold change Log 2 fold change LC-FT MS/MS Triplicate injections /sample DIFFTAL / Progenesis LCMS Detection, Alignment/Match & sequence LC-FT MS/MS Triplicate injections /sample Differential at the features level : m/z, RT, Isot. Pattern, Intensity, Identification/not 2D LC Disease : 28 samples

6 SP1: CSF / Plasma samples dedicated to the project Collection of CSF & plasma samples (CHRU, 24 months) - Biomarker identification HR MS peptide profiling: 15 AD (homogenous group) and 15 age-matched controls (FTD, ALS & MS): 1 ml each (split between IGF and SANOFI): 29 samples (14 AD (1 with 0.5 ml) and 15 Controls 2D peptide map of human CSF: CSF samples with maximal heterogeneity (age, sex, pathologies, but including AD, FTD, ALS & MS): 0.5 ml each 28 samples - Biomarker validation by LC/QqQ/MRM On going. 24 AD at different stage of the disease (MCI, moderate & severe) and 24 age-matched controls (FTD, ALS & MS) - Biomarker validation by Ab assays Scheduled samples with clinical data - Preclinical study / biomarker final selection Scheduled samples with clinical data

7 Sample preparation Standardized and reproducible method for CSF sample preparation Depletion of 20 major proteins 80% of CSF proteins are common with the plasma [97% of total protein content] Immunodepletion kit ProteoPrep20 (Sigma) optimized Only 3 proteins (Apo-A1; Apo-A2 and Transthyretin are not well depleted Proteins quantification after depletion (for 200 µl CSF Natif) 88% depletion in protein amount Reproducible protein digestion on filter (FASP) We now start with 200 µl of CSF

8 1D-RP Populating CSF MS peptide map 1D-HILIC Bank strategy : Results 2D-LC/MS RP-fractionations: Global comparison Sample used : CSF (Hydrocephalus + MS + ALS) Polarendcapped RP RP1 classical RP RP2 Fractionation Cumulative Unique peptides RP1 (PAII) 4914 TSK Gel Amide 80 TSK Zwitterionic HILIC ZIC RP2 (LiChrosorb) 7137 ZIC 8206 TSK 9305

9 Comparison of our master map with the litterature Pool of Human CSF (hydrocephalus, ALS, MS) 2246 CSF proteins Plasma vs CNS proteins (DB plasma proteins: HuPO/PPP & Liu and al, 2006) 4563 proteins CNS only CNS proteins: vs CSF proteome (CSF proteome: Schutzer and al, 2007 and 2011) CSF: 2624 proteins New CNS proteins 1026 (46 %) SA-CSF Plasma protein DB SA CSF/CNS DB Schutzer CSF/CNS

10 How deep do we dig in CSF proteome? TF Macrophage Migration Inhibitory Factor : 145 pg/ml Heparin-binding epidermal growth factor-like growth factor : 25 pg/ml Proteins matched1d Titration by Multiplex immunoassay of 125 proteins or peptides in 333 CSF samples (PLoS One. 2011, 12;6(1):e16032)

11 Multiple reaction monitoring (MRM) validation Classicaly performed on a QQQ mass spectrometer We know what we are looking for: One precursor mass + one fragment mass = one MRM transition Very specific Low background noise (LOQ is very low) Very fast (80 per second) so multiplexing is possible Several peptides per protein and several transitions per peptides

12 Conclusion and perspectives 1 Collection of CSF & plasma samples (CHRU, 24 months) - For biomarker identification (CSF 2D peptide map + HR profiling) - For biomarker validation by LC/QqQ/MRM - For biomarker validation by Ab assays and preclinical study 2 Method development (SANOFI & IGF, 16 months) - Standardized and reproducible method for CSF sample preparation and MS analysis 3 Candidate Biomarkers identification (SANOFI & IGF, 18 months) - Establishment of 2D peptide of human CSF - HR-MS peptide profiling of CSF from AD patients and age-matched controls : 4 Candidate Biomarkers Validation by LC/QqQ/MRM (SANOFI & IGF, 18 months) On going 5 Candidate Biomarkers Validation by Ab assays (BioRad, 12 months) Scheduled 6 Preclinical study and final biomarker selection (BioRad, 18 months) Scheduled Bonus: 2D HR MS map of LCR is now established and continously enriched with new experiments A new project on multiple sclerosis was started Technology applicable to any type of protein samples

13 Thank you for your attention