QUANTITATIVE LC-MS/MS Analysis Of Proteins And Peptides

QUANTITATIVE LC-MS/MSAnalysis of proteins and peptidesKarolina Krasinska

Overview Introduction: Quantitative approach to proteinand peptide analysis Background: Instrumentation, workflow Assay development: Step by step Quantitative LC-MS/MS assay: method,results Common issues:– Matrix effect– Internal standard (IS)– Method validation– Analyte stability2

Absolute quantitation – targeted analysis Quantitation on peptide levelProteolytic digest(Trypsin, chymotrypsin, AspN etc.) Know what you are going to quantify:– well characterized protein– protein ID via LC-IT-MS/MS– in silico digest based on theoretical AA sequence3

Sample Protein – unlimited size (gel, solution,biological matrix) Peptides – ideally 6-12 aa (up to 30-40 aa);may be phosphorylated Biological matrices:– plasma, serum, erythrocytes– cerebrospinal fluid (CSF)– urine– bile– cell culture media– plant and animal tissues (e.g. leaf, brain, liver)4

Analysis and data processing overviewA) Analysis workflowXX ISSamplePrep digestHPLCQ1CollisioncellQ3ISB) Data processing5

Triple quadrupole analyzerMS1CollisionCell (w/Argon)1Vm1Precursor ionMS21Vm2Fragment ionScheme from Water Quattro Premiere Training6

Advantages of SRM scanning mode MS/MS provides higher sensitivity andselectivity, enabling– Less extensive sample preparation5.61100%– Greater sensitivity via increased selectivity0Time4.00– Use of shorter HPLC columns Translates into time and effort savings,plus a more sensitive method6.007.00178.0%100– Use of shorter run times and higher 00

Before we start – checklistQuestions: What is the protein of interest? (MW, sequence,PTMs) Number of peptides included in the assay? Sample types? (matrices) Desired/required LLOQ and calibration range? Standards availability? (purified protein, peptides andlabeled peptides) Purpose of developing the assay? (preliminary studies,confirmation of findings from different methods,publication) Available funding and timelines?8

Peptide selection rulesMain rules: 6-12 AA optimal (up to 30-40aa) No chemically reactive residues (Trp, Met, Cys) No ragged ends (2xR; RK; 2xK) No potential PTM unique sequenceAlso consider: Preferably containing P (dominant cleavages) If MS/MS data already available than look for peptides giving highintensity fragment ions of m/z higher than precursor ion (noisereduction) R in P proximity (potential missed tryptic cleavage) Select multiple peptides for each protein whenever possible9

Protein X - fragmentDigest Table# PosMassDigest TablepIPeptide# 79APASAAWR19 LR20146.198.75K51464.69 PVTAFGLGTWQVQR23 TETR9146.198.75K25 WHYR11 1157.334.53LNLIAELESR27804.924.37DLEAMAR12 1719.074.14VLAEPVPLPADPMELK28 2744.064.37ITGAEPIFIDANFQSTVPGGPIGGQTR13 0 33429.475.52GTPGV10

Protein X - fragmentDigest Table# PosMassDigest TablepIPeptide# 79APASAAWR19 LR20146.198.75K51464.69 PVTAFGLGTWQVQR23 TETR9146.198.75K25 WHYR11 1157.334.53LNLIAELESR27804.924.37DLEAMAR12 1719.074.14VLAEPVPLPADPMELK28 2744.064.37ITGAEPIFIDANFQSTVPGGPIGGQTR13 0 33429.475.52GTPGV11

Protein X - fragmentDigest Table# PosMassDigest TablepIPeptide# 79APASAAWR19 LR20146.198.75K51464.69 PVTAFGLGTWQVQR23 TETR9146.198.75K25 WHYR11 1157.334.53LNLIAELESR27804.924.37DLEAMAR12 1719.074.14VLAEPVPLPADPMELK28 2744.064.37ITGAEPIFIDANFQSTVPGGPIGGQTR13 0 33429.475.52GTPGV12

Protein X - fragmentDigest Table# PosMassDigest TablepIPeptide# 79APASAAWR19 LR20146.198.75K51464.69 PVTAFGLGTWQVQR23 TETR9146.198.75K25 WHYR11 1157.334.53LNLIAELESR27804.924.37DLEAMAR12 1719.074.14VLAEPVPLPADPMELK28 2744.064.37ITGAEPIFIDANFQSTVPGGPIGGQTR13 0 33429.475.52GTPGV13

Peptide selection – “good spectrum”y5100%1,120.90 AMU, 2 H (Parent Error: 250 ppm)LQLRelative IntensityRSFTGVTTVGTFSRQy7LL75%y4b5 Excellent sequence coverage High dispersion of energy Good for peptide IDb650%y6y8 2Hy8-H2O 2H25%b2b3b4y2y30200b8b7y10%y8y7 1parent 2H-H2O447.6?400600b9800y91000m/zy6100%1,548.30 AMU, 2 H (Parent Error: 310 ppm)Relative IntensityEKGFFILEQDLPIPIDQLELFIFKGE75%50% Good sequence coverage Low dispersion of energy Good for quantitationb7y6 0b111250b12150014

Assay development outline Mass spectrometry (MS) Liquid chromatography (LC) LC-MS/MS method optimizationand characterization Sample preparation15

Method development - mass spectrometry Acquisition of MS and MS/MS spectra forstandard solution of the peptideMS– 10-50µM,– direct infusion Precursor ion – fragment ion MS parametersoptimization– most efficient ionization; precursor ion preferably 2– most efficient fragmentation; fragment ions preferablywith m/z higher than precursor ion Method set up and test run– Choose three precursor ion – fragment ion pairs foreach peptide16

Liquid chromatography HPLC column– RP type column – C18– Retention– Separation – usually no need to have a basepeak separation– Gradient elution– MS compatible solvents and buffer modifiers(0.1% FA in water, 0.1% FA in acetonitrile)– Shortest runs possible (usually 4-8 min perinjection)LC17

LC-MS method optimization & characterization Linear calibration curve Limit of detection (LOD)– 3:1 signal-to-noise ratio Limit of quantitation (LOQ)– 10:1 signal-to-noise ratio Sample matrix interferencesLC-MS– analysis of blank sample matrix spikedwith pure standard Carryover Stability of the analyte18

Sample preparation Objectives:––––Isolating analyte from matrixRemoving contaminants, desaltingConcentrating analyte if necessaryReconstituting in appropriate LC-MS compatible reagent Extraction methods:– Prior to proteolytic digest Protein precipitation– Methanol– Acetonitrile Immunoprecipitation– After proteolytic digest Solid Phase Extraction (SPE)– C18– Ion exchange (e.g. SCX)– Combination of the above19

Quantitative LC-MS/MS assayEach set of analyzed samples contains the following: 6-8 point calibration curve sets (in triplicate).A calibration curve set is usually run at the beginning,at the end, and once or more during the sample set. 1 QC per 10 samples (in triplicate) n# of samples (in triplicate) blank injections if necessaryXHPLCQ1CollisioncellQ3IS20

Results – QuanLynx reportAnalyte and ISareaSRM alyteCalibrationcurveIS21

Most common challenges Internal standard (IS) Matrix effects Method validation22

Internal standard – to use or not to use? Benefit of IS:– If the analyte and IS suffer the same losses and the sameeffects in the matrix, matrix effects and sample losses cancelwhen we take the ratio of IS to analyte IS compensates for common analyte losses:–––––Analyte adsorption on surfacesExtensive sample manipulationDegradationEvaporationAutosampler variabilityXIS23

Peptide adsorption on the surfaceNo ISCompound name: PSMB2Correlation coefficient: r 0.994609, r 2 0.989247Calibration curve: 6.64019 * x -1.62103Response type: Internal Std ( Ref 10 ), Area * ( IS Conc. / IS Area )Curve type: Linear, Origin: Exclude, Weighting: 1/x, Axis trans: None30000025002500002000ResponseResponseCompound name: PSMB2Correlation coefficient: r 0.973449, r 2 0.947603Calibration curve: 746.428 * x -726.173Response type: External Std, AreaCurve type: Linear, Origin: Exclude, Weighting: 1/x, Axis trans: NoneWith IS20000015000015001000100000500500000050fmol/ul100 150 200 250 300 350 4000050100150200250300350fmol/ul40024

Extreme exampleCompound name: OligoACorrelation coefficient: r 0.852161, r 2 0.726178Calibration curve: 20.9105 * x -1.15707Response type: Internal Std ( Ref 2 ), Area * ( IS Conc. / IS Area )Curve type: Linear, Origin: Exclude, Weighting: 1/x, Axis trans: 00.600.801.001.201.401.601.80pmol/ul2.0025

Tricks to improve calibration curve linearity Isotopically labeled internal standard Sample buffer modification:– Sample matrix– Diluted sample matrix– Addition of neutral protein– Higher organic solvent content26

Matrix effects Matrix effects:– Ion suppression– Interferences from metabolites– Signal enhancement Assessment of matrix effect:– Post-column infusion of analyte– Comparison of analyte in matrix-free solution vs. spikedblank matrix – extraction efficiency assessment Minimizing matrix effects:––––Use isotopically labeled internal standardUse fragment ions of m/z higher than precursor ionGenerate “cleaner” extractOptimize HPLC method27

Method Validation Detection capability:– LOD – signal to noise ratio 3:1– LOQ – signal to noise ratio 10:1 Calibration curve – linear dynamic rangePrecision and accuracySelectivitySpecificityStability of analyte (and matrix):–––––Short-term, long-termLow and high concentrationsAnalyte in sample solvent and in raw matrixDry extract, reconstituted standard/extractFreeze/thaw cycles28

Analyte stabilityWith ISNo ISCompound name: ProgCorrelation coefficient: r 0.999252, r 2 0.998505Calibration curve: 0.440538 * x 0.0156705Response type: Internal Std ( Ref 2 ), Area * ( IS Conc. / IS Area )Curve type: Linear, Origin: Exclude, Weighting: 1/x, Axis trans: NoneCompound name: ProgCorrelation coefficient: r 0.970733, r 2 0.942323Calibration curve: 4511.2 * x 221.327Response type: External Std, AreaCurve type: Linear, Origin: Exclude, Weighting: 1/x, Axis trans: 1.002.003.004.00fmol/ul5.0029

Literature References1.Mass Spectrometry: Principles and Applications byEdmond de Hoffmann, Vincent Stroobant2.Trace Quantitative Analysis by Mass Spectrometry3.4.by Robert K. Boyd, Cecilia Basic, Robert A. BethemLC/MS: A practical User’s Guide by Marvin McMasterQuantitative Proteomics by Mass Spectrometry bySalavatore Sechi5.6.www.ionsource.comSUMS website: mass-spec.stanford.edu30

AcknowledgementsStanford University Mass Spectrometry Staff Allis Chien Chris Adams Pavel Aronov Theresa McLaughlinVincent and Stella Coates Foundation31

Introduction: Quantitative approach to protein and peptide analysis Background: Instrumentation, workflow Assay development: Step by step Quantitative LC-MS/MS assay: method, results Common issues: – Matrix effect – Internal standard (IS) – Method validation – Analyte stability