Jaume C. Morales

LC/MS Product Specialist

Agilent Technologies

From Biomarker Discovery to Validation:

Routine Nano-LC/QQQ for Quantitation of

Peptides

Page 1

Agilent Proteomics Biomarker Workflow

Candidates

Extraction

Depletion

6520 QTOF

Candidate

Biomarker

Identification

SAMPLE1 DATA

SAMPLE2

Page 2

Proteolytic Digest

Fractionation

Extraction

Identification

6410 QQQ

Validation

• Perform statistical analysis for identification of biomarker candidates

Biomarker validation workflow

• Run samples on Q-TOF for protein ID in data-dependent MS/MS mode

Step 1

Q-TOF

• Search QTOF data using Spectrum Mill

• Use Spectrum Mill MRM Builder to create a list of MRM transitions with RT

Step 3

Spectrum Mill

Step 2 and 5

Mass Profiler Pro• Run samples on

QQQ in Dynamic MRM mode

Step 4

QQQ

• Integrate the MRM chromatograms

• Import quantitation results into MPP to perform statistical analysis

2/15/2011Page 3

MRM

acquisition

QTOF

acquisition

MRM

optimization

Workflow for quantitative peptide analysis using MRM

1. Creation of a (D)MRM acquisiton method from

discovery data

SpectrumMill MRM Builder

DMRM

acquisitionExport MRM transitions and optional retention time information

for optimization or direct (D)MRM acquisition analysis

1. Export transitions from SpectrumMill.wmv

2. Import SpectrumMill results into Optimizer for Peptides.wmv

MRM acquisition

MRM optimization

Workflow for quantitative peptide analysis using MRM

2. In-silico prediction of MRM transitions

SpectrumMillPeptide Selector

1. + 2. 3.

Predict proteotypic peptides (BLAST search) and import optimized transitions into MRM method

1. Peptide selector import into MassHunter Qual.wmv2. Import transitions from MassHunter Qual.wmv3. Import transitions into MRM method.wmv

DMRM acquisition

MRM acquisiton

Workflow for quantitative peptide analysis using MRM

3. Generation of a (D)MRM acquisition method

3.MRM optimization

1. + 2.

Verify results from Optimizer for peptides and perform (D)MRM analysis

1. Optimizer for peptides data.wmv2. Import transitions into MRM method.wmv3. Update RT information for DMRM method.wmv

Prerequisites for successful profiling and quantitation of biomarkers

Signal Response

• Sensitivity

• Linearity

• Reproducibility

• Separation Speed

• Peak Capacity

Mass Spectrum

• Selectivity

• Acquisition Rate

• Peak Capacity

• Reproducibility

Page 7 2/15/2011

HPLC-Chip/QQQ LCMS TechnologyNanospray chip configuration brings new era in high sensitivity quantitation

NanoLC system foranalytical chromatography

HPLC Chip Cube system

QQQ LCMS

CapLC pump for sampleloading on enrichment column

Page 8

Sensitivity: low -mid amol

Dynamic range: 103 -105

Triple Quadrupole Mass SpectrometerExtending Outstanding Performance

6400 Series – Triple Quad – NEW Functionality

� No Cross Talk collision cell – commonality with QTOF

� Peptide Optimizer

� Dynamic MRM – 4000 MRMs

Page 9

� High Sensitivity – 10attomols on peptides

HPLC-Chip/MS Interface:Fluid Connections to the HPLC-Chip

Stator

Rotor

Side View

Nanopump

Autosampler

Waste

Page 10

Rotor

Stator

inner rotor

outer rotor

Microvalve

HPLC-Chip

Retention Time Reproducibility

RT SD %RSD

EIC 487.8 3.618 0.014 0.40

EIC 752 3.788 0.011 0.29

EIC 740.6 5.018 0.010 0.20

EIC 874.4 3.968 0.012 0.31

EIC 653.6 4.289 0.012 0.28

EIC 511.7 3.681 0.012 0.31

EIC 722.7 3.547 0.012 0.35

Extracted ion chromatograms for 17 peaks from a EIC 778 4.143 0.010 0.23

EIC 526.3 4.399 0.015 0.34

EIC 547.5 4.472 0.011 0.25

EIC 746.7 5.196 0.011 0.20

EIC 519.1 4.142 0.011 0.26

EIC 508.2 4.972 0.011 0.23

EIC 582.4 4.679 0.011 0.23

EIC 461.9 3.905 0.012 0.30

EIC 474 4.759 0.011 0.22

EIC 628 4.584 0.010 0.22

Extracted ion chromatograms for 17 peaks from a BSA tryptic digest (50 fmol on-column)

RT reproducibility evaluated using 69 repeat injections

1-3%RSD Interchip RT

differences – Applied

Proteomics reference

Ion Intensity Reproducibility

Scatter plot of ions intensities

3,5

4

4,5

5

5,5

log

in

div

idu

al in

ten

sit

y

Replicate 1

Replicate 2

Replicate 3

Replicate 4

Replicate 5

1

Scatter plot of intensities: Run 1 vs, Run 10

� 234 ions were monitored over 5 replicate runs

� ~ 94 % of all ions show less than 20% variation in intensity across all 5 replicates.

� HPLC-Chip/TOF

N =234

Data courtesy of Dr. Pierre Thibault University of Montréal Immunology and Cancer Research Institute

2,5

3

2,5 3 3,5 4 4,5 5 5,5

log average intensity

Replicate 5

2

1 2 Data courtesy of Dr. H. Zhang, X.J. Li and Dr. R. Aebersold, Institut für Molekulare Systembiologie, Switzerland

� 5000 glycopetides monitored over 10 replicate runs.

� HPLC-Chip/TOF

Rapid Nanoflow LC/MS – 3 Minute Analysis21 MRM Transitions From 7 Peptides

MRM Optimizer for QQQ

Programa que permite encontrar los méjores valores de

Fragmentor y/o Energía de colisión

para aquellas transiciones que nos interesan en base a su abundancia.

Es necesario especificar la m/z de precursor y product ion.

Peptide Optimizer for QQQPeptide Optimizer for QQQ

Caso particular del MRM optimizer.

Las transiciones se especifican tan sólo detallando el péptido.

El péptido escogido puede provenir de SM Peptide Selector

o bien de MH Qual

¿Que es diferente en Peptide Optimizer?

Secuencia de péptidos en lugar de fórmulas químicas

Precursor generalmente +2 y/o +3 de estado de carga

El espectro MS/MS muestra varios product ions

Algunos product ions tienen m/z > precursor m/z

Es predecible la fragmentación de los deseados iones “b” y “y”

aunque algunos product ions pueden presentar también

multicarga

Típicamente se buscan 2 -3 péptidos por proteína y 2-3

transiciones por péptido.

Desarrollo y Optimización de métodos SRM para Péptidos

Selección de transiciones basada en

• Observaciones de MS/MS durante el descubrimiento ó

predicciones de iones “b” y “y”.

• Predicción de exclusividad de la secuencia peptídica en la base

de datos.de datos.

Optimización de las transiciones

• El parámetro más importante en QQQ es la energía de colisión.

• Debe seleccionar las transiciones que ofrezcan una mayor

relación S/N y menor interferencias en la matriz.

MRM Method Optimizer for Peptides

Ejecutar Optimizer: Importar las secuencias de péptidos, luego ver las predicciones de fragmentos b/y y seleccionar iones para la optimización.

Run 1 (MRM): Optimiza la CE en modo MRM utilizando la fórmula del Q-TOFcomo punto de partida, en un sólo análisis.

Ventajas:

• Reduce considerablemente la cantidad de muestra utilizada. • Reduce considerablemente la cantidad de muestra utilizada.

• Puede optimizar en más de una carga para el mismo péptido.

• Pre-selección de iones b/y (incluyendo multicarga)

• El único límite está en el numero total de transiciones por inyección de optimización. El propio software nos da información basado en la anchura de los picos y Cycle Time. Conviene adquirir ≥10 puntos por pico.

Enter Sequences and Select b/y Ions For Optimization

Results of Peptide Optimization

Precursor

Peptide sequence

19

Abundance of each transition allows customer to choose best transitions for the final method

Optimized product ions

� Comparison of MRM and Dynamic MRM

Time (min) 1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25

Compounds (10/block)

Cycle Time (sec)

MRM

50 80

0.5 0.8 1

Time Segment 1 Time Segment 2 Time Segment 3 Time Segment 4

0.7

100 70

Dynamic MRM

Page 20

Cycle Time (sec)

Max Coincident

Cycle Time (sec) 0.4 0.4 0.4 0.4

20 40 40

Dynamic MRM

0.5 0.8 1

30

0.7

• 2 x shorter cycle times supports narrow chromatographic peaks, more analytes or longer dwell per

analyte.

Dynamic MRM – what happens?

Increases

Sensitivity

Improves cycle

time

Page 21

4 ions – not 11 7 ions - not 11

Provides better

chromatographic

definition

Dynamic MRM: Retention Time Based Scheduling of MRM for Optimal Sensitivity

Absolute Protein Quantification in the

Context of Non-clinical Drug Safety Evaluation

UCD Conway Institute

University college Dublin

And

Agilent Technologies

Page 23

Vehicle Low Dose High Dose

Day 2,4,15 Day 2,4,15 Day 2,4,15 Day -3/-4, 1/2, 3/4, 12/13

Histopathology

Transcriptomics

Proteomics

Histopathology

Transcriptomics

Proteomics

Transcriptomics

Proteomics

Metabonomics

Clinical Biochemistry

Metabonomics

Clinical Biochemistry

VehicleVehicle Low DoseLow Dose High DoseHigh Dose

Day 2,4,15 Day 2,4,15 Day 2,4,15 Day -3/-4, 1/2, 3/4, 12/13

Histopathology

Transcriptomics

Proteomics

Histopathology

Transcriptomics

Proteomics

Transcriptomics

Proteomics

Metabonomics

Clinical Biochemistry

Metabonomics

Clinical Biochemistry

Collins B. C. et al. ASMS 2008 MPQ 477

Experimental Design

Rat liver lysate were prepared from

rats treated with troglitazone or

vehicle control

Peptides and MRM transitions were

selected using Peptide Selector in

Catalase was selected based on

previous 2D-DIGE data

1 mg of soluble protein extract was

Page 24

selected using Peptide Selector in

Spectrum Mill and 13C, 15N labeled

peptides were synthesized

1 mg of soluble protein extract was

reduced, alkylated, acetone

precipitated and trypsin digested

The liver digest were spiked with the isotope-labeled peptides

and analyzed by Agilent 6410 QQQ system

Using Spectrum Mill Peptide Selector for Optimising MRM Transitions

Chemically reactive residues

(Cys = C, Met = M, Trp = W)

Peptides adjacent to multiple cleavage site

Chemically unstable residues (Asp-Gly

Page 25

Chemically unstable residues (Asp-Gly = D-G; Asn-Gly = N-G; N-term Glu = E; N-term Asn = N)

Eliminate “LC-incompatible” peptides

Uniqueness

Spectrum Mill – Peptide Selector

Page 26

Peptide Selector – Catalase Results

Page 27

Catalase Peptide LAQ – Peptide Selector

Page 28

Catalase Peptide EAE – Peptide Selector

Page 29

External Calibration on Catalase PeptidesLinearity : five order of magnitude

External quantitation curve of catalase peptide L*AQEDPDYGLR from 78 amol to 7800 fmol

78aMol

78fMol

780 fMol

0.9965

Page 30

RSD < 6%

Catalase Quantitation Results

Page 31

Quantitation of protein phosphorylation using

MRM

Erk1 protein was quantified

from depleted human serum.

+3 PO4

Peptide MRMLorne

Page 32

Erk1

intact

protein+2 PO4

+3 PO4

+4 PO4

Selection of MRM transitions

TY: IADPEHDHTGFLTEYVATR

y5

b14

t202: IADPEHDHTGFLTEYVATR

y16

y5

P

Precursor ion Product ions

TY

545.3 615.3 782.5

[M+3H] 3+ y5 b142+

t202

753.3 615.3 979.9

Page 33

t202y204:

P

IADPEHDHTGFLTEYVATR

y16

y5

P

y204: IADPEHDHTGFLTEYVATR

y16

y5

P

y5 t202

[M+2H] 2+ y5 y162+

y204

753.3 979.9 695.3

[M+2H] 2+ y162+ y5

t202y204

780.0 647.6 695.3

[M+2H] 2+ y163+-H3PO4 y5

Chromatographic Separation of the Four Peptide Standards allowed the selection of the same Q1 and Q3 transitions for two different peptides

4x10

3.5

4

4.5

5

5.5

6

+ MRM (545.29999 -> 615.29688) mix500f-r001.d

TY

t202t202y204

Page 34

0

0.5

1

1.5

2

2.5

3

Counts vs. Acquisition Time (min)

9.5 10 10.5 11 11.5 12 12.5 13 13.5 14 14.5 15 15.5 16 16.5 17 17.5 18 18.5 19 19.5

y204

Y434y13 - 5 Levels, 5 Levels Used, 15 Points, 15 Points Used, 0 QCs

T432tY434y13 - 5 Levels, 5 Levels Used, 15 Points, 15 Points Used, 0 QCs

Concentration (fmol/ul)

-25 0 25 50 75 100 125 150 175 200 225 250 275 300 325 350 375 400 425 450 475 500 525

Relative Responses 1x10

-0.2

0

0.2

0.4

0.6

0.8

1

1.2

1.4

1.6

1.8

2

2.2

2.4

2.6

2.8

3

3.2

3.4

3.6

3.8

4y = 0.0764 * x + 0.2516R^2 = 0.99673108

t202y204

R2 =0.9967Relative Responses

TY13 - 5 Levels, 5 Levels Used, 15 Points, 15 Points Used, 0 QCs

Concentration (fmol/ul)

-25 0 25 50 75 100 125 150 175 200 225 250 275 300 325 350 375 400 425 450 475 500 525

Relative Responses 2x10

-0.1

0

0.1

0.2

0.3

0.4

0.5

0.6

0.7

0.8

0.9

1

1.1

1.2

1.3

1.4

1.5

1.6

1.7

1.8

1.9

y = 0.3630 * x + 0.0337R^2 = 0.99816859

TY

R2 =0.9981Relative Responses

T432t13 - 5 Levels, 5 Levels Used, 15 Points, 15 Points Used, 0 QCs

0.5fm

2.5fm

5fm

0.5fm2.5fm

5fm

Page 35

Y434y13 - 5 Levels, 5 Levels Used, 15 Points, 15 Points Used, 0 QCs

Concentration (fmol/ul)

-25 0 25 50 75 100 125 150 175 200 225 250 275 300 325 350 375 400 425 450 475 500 525

Relative Responses 2x10

-0.1

0

0.1

0.2

0.3

0.4

0.5

0.6

0.7

0.8

0.9

1

1.1

1.2

1.3

1.4

1.5

1.6

1.7

1.8

y = 0.3447 * x + 0.1682R^2 = 0.99564716

y204

R2 =0.9956

Relative Responses

T432t13 - 5 Levels, 5 Levels Used, 15 Points, 15 Points Used, 0 QCs

Concentration (fmol/ul)

-25 0 25 50 75 100 125 150 175 200 225 250 275 300 325 350 375 400 425 450 475 500 525

Relative Responses 1x10

-0.5

0

0.5

1

1.5

2

2.5

3

3.5

4

4.5

5

5.5

y = 0.1069 * x + 0.2840R^2 = 0.99593214

t202

R2 =0.9959

Relative Responses

0.5f

m

2.5f

m

5f

m

0.5fm

2.5fm

5fm

Quantitation of the degree of phosphorylation at T202 and Y204 in active Erk1 protein

peptide% Molar

ratioRSD (n=9)

TY 20% 0.13

t202 25% 0.15

Page 36

y204 21% 0.12

t202y204 34% 0.08

In this batch of active Erk1 sample, 59% of T202

and 55% of Y204 were phosphorylated

Peroxidase in Human PlasmaFrom Discovery Mode to Validation Step

Page 37

Step-2

Spectrum Mill • Import the MRM list

Step

Mass Prof

Biomarker validation workflow

Discovery phase to Validation : MRM Selector

Page 38

• Run samples on Q-TOF for protein ID in data-dependent MS/MS mode.

Step-1

Q-TOF

• Search QTOF data using Spectrum Mill

• Use Spectrum Mill MRM Selector to create a list of MRM transitions with RT

Spectrum Mill • Import the MRM list into QQQ Acquisition software

• Run samples on QQQ in Dynamic MRM mode

Step-3

QQQ

• Integrate the MRM chromatograms

• Import quantitation results into MPP to perform statistical analysis

Mass Prof

Depleted Human Plasma Sample analysis

Replicate LC/MS runs

HPLC – Chip / QTOFSpiked with 0,0.5 and 5fmol

per 0.5ug plasma

Page 39

Data Dependent

Protein IDs from

Spectrum Mill

MRM SelectorGenerates MRM method from discovery QTOF data

Page 40

Dynamic MRM

Page 41

Dynamic MRMOverlaid 2000 MRM chromatograms acquired in a single run using Dynamic MRM

Page 42

Sensitivity : Peroxidase in plasma matrix (per 0.5ug)

M A B

matrix 5fmol 500amole

Page 43

# MRM

RT window

(min)

Cycle time (ms)

Min. dwell (ms)

Max. # concurrent

MRM

%RSDArea

RSDRT

(min)

443 2 1000 16.5 50 2.5 0.038

443 1 1000 29.83 30 3.2 0.016

2000 2 1000 2.75 160 4.5 0.030

3293 1 1050 2.18 185 4.7 0.025

Reproducibility

of MS response

and RT

Mass Profiler ProfessionalStatistical Processing of MRM Data

Four peptides from

peroxidase were

highlighted in green.

The mean of 443 MRM

abundances is

displayed (black) to

show the peptides from

Page 44

All Samples

B1 B2 B3 A1 A2 A3 M1 M2

show the peptides from

plasma did not vary

from sample to sample.

Principle Component AnalysisMatrix and 2 different peroxidase levels

Samples at

Page 45

Samples at

different

peroxidase

concentrations

were correctly

grouped

together.

Hierarchical Clusteringcomparing different peroxidase concns.

A condition was generated

with peroxidase

concentration color-coded

on the tree branches, along

with the peptide features

labeled on each row. The

Page 46

M1 M2 A1 A2 A3 B1 B2 B3

labeled on each row. The

heat map is colored from

blue to red, where blue is

low abundance and red is

high abundance. The full

view of all the features is on

the left. The zoom view is

on the right.