Rapid Assay Development and Refinement for Targeted Protein Quantitation Using an Intelligent SRMQuantitation Using an Intelligent SRM (iSRM) Workflow

September 2, 2010Reiko Kiyonami, Ph. D.Senior Strategic Marketing Specialist – ProteomicsThermo Fisher Scientific

What is Targeted Quantitative Proteomics?

DiscoveryFishing for differentially

Targeted Measuring proteins inFishing for differentially

expressed proteinsMeasuring proteins inknown pathways

2

Why Targeted Quantitative Proteomics? Need to erif potential biomarkers

Control Knockdown

SPAcp62F

Need to verify potential biomarkers which have been found in the discovery experiments to narrow down the long biomarker candidate list to a few of real

i i f lid ti tAcp62F

5e+55e+5

promising ones for validation stage.

Need to acquire consistent quantitative data for large protein sets t f ilit t d li i t

Control KnockdownUAS-IR(SP)/TM3, Sb UAS-IR(SP)/tubulin-GAL4

Inte

nsity

1e+5

2e+5

3e+5

4e+5

520.7 > 927.5520.7 > 813.4520.7 > 700.3520.7 > 643.3520.7 > 546.3520.7 > 485.2520.7 > 289.1

Inte

nsity

1e+5

2e+5

3e+5

4e+5520.8 > 927.5 520.8 > 813.4520.8 > 700.3520.8 > 643.3520.8 > 546.3520.8 > 475.2 520.8 > 289.1

to facilitate modeling in systems biology studies.

The traditional immunoassay th d i t i t i l t

FPIPSPNPRDK (from SP)

FPIPSPNPRDK (from SP)

Time 0

ty

6e+4

8e+4

690.9 > 1252.6690.9 > 1123.6690.9 > 1008.6690.9 > 877.5 690.9 > 764.4 690.9 > 584.3

Time

0

ty

6e+4

8e+4

690.9 > 1252.6690.9 > 1123.6690.9 > 1008.6 690.9 > 877.6690.9 > 764.4690.9 > 584.3

method is too expensive to implement on such a large scale.

Alternatively, a multiple SRM assay t i l d l

KEDMLLGVSNFK(from Acp62F)

KEDMLLGVSNFK(from Acp62F)

Time

Inte

nsit

0

2e+4

4e+4 690.9 > 495.2

Time

Inte

nsit

0

2e+4

4e+4690.9 > 495.2 on a triple quadrupole mass

spectrometer can be used for targeting hundreds of biomarker candidates from hundreds of complex samples in a hi h th h t

3

high throughput way.Data provided by M. J. MacCoss from Uni. of Washington

SRM (Selected Reaction Monitoring) assay

Q2 Q3Q1

Precursor IonPre set

Product IonPre set

CIDPre set Pre set

1. A “proteotypic” peptide is selected as being quantitative surrogate of a1. A proteotypic peptide is selected as being quantitative surrogate of a targeted protein.

2. Q1 is set to transmit only the precursor ion of the selected peptide.

3 Q2 i d t i d f t ti f th i3. Q2 is used to induce fragmentation of the precursor ion.

4. Q3 is set to transmit a specific product ion,

Each precursor to product ion set is called an SRM transition

4

Each precursor to product ion set is called an SRM transition.

Why SRM Analysis for Targeted Protein Quantitation?

Highest specificity/sensitivity – necessary for l b d t tid i l t i liklow abundant peptides in complex matrices like plasma

Largest linear dynamic range for quantitation

Well-established as the quantitative method of choice for small molecule applications

5

TSQ Vantage allows Low Level Quantitation in Background Matrix

Linear response over wide dynamic range

Stable isotope labeled peptides 2000000

spiked into whole yeast lysate

1200000

1600000

LOD in low attomole range

400000

800000

12000

0 20000 40000 60000 80000 100000 [amol]

800

3 amol

GILFVGSGVSGGEEGAR 801.4 1072.5LTILEELR 498.8 782.5LVEDPQVIAPFLGK 767.4 569.4

0

400

6

0 20 40 60 80 100 [amol]

Peptide Selection and SRM Assay Design

The success and confidence of an SRM experiment directly depends on the selected proteotypic peptides and the specific transitionsand the specific transitions.

Peptides have to be unique to the targeted protein

Peptides should not contain Cys, Met or other commonly

modified residues

Peptides should yield excellent mass spectral signal

Selected product ions used for SRM analysis should have higher

mass over charge values for increased selectivity

7

From Protein…to SRM – How can you select your peptides/transitions?

MNSFSTSAFGPVAFSLGLLLVLPAAFPAPVPPGEDSKDVAAPHRQPLTSSERIDKQIRYILDGISALRKETCNKSNMCESSKEALAENNLNLPKMAEKDGCFQSGFNEETCLVKIITGLLEFEVYLEYLQNRFESSEEQARAVQMSTKVLIQFLQKKAKNLDAITTPDPTTNASLLTKLQAQNQWLQDMTTHLILRSFKEFLQSSLRALRQM

P i l d t t d i di i t ?Previously detected in discovery experiments?

Yes No

SRMs chosen based on peptide MS/MS spectra acquired from previous

SRMs designed insilico based on protein q p

experiments.p

sequence

Bio-Informatics Hypothesis

8

Bio InformaticsBased

HypothesisBased

Can we use CID MS/MS data for selecting SRM transitions?

9

Comparing Product Ion Rankings – Linear Ion Trap CID vs. Triple Quadrupole SRM Transitions

12

ASGAFTGENSVDQIK (+2)LTQ CID MS/MS

3 456

78 9101111 12

Rank Correlation Coefficient = 0.988p-Value = 5e-5

TSQ SRM9095

100 989.49 1090.54y9 yC i S/ S1

23

4 567

8

p Value 5e

455055606570758085

503.28y3

y4

y7

y9 y10

y11

Abu

ndan

ce

Composite MS/MSSpectrum

89

1011 12

51015202530354045 388.26

602.35

689.38

803.42

932.47

1237.61

1308.64 1365.66 1452 670

y5

y6y8

y12 y13y14

Rel

ativ

e A

10MacCoss et al. J. Proteome Research 2008, 8(6), 2733-2739

m/z m/z m/z m/z m/z m/z m/z m/z m/z m/z m/z m/z05 1452.670

Evaluation of Rank Overlap: LTQ CID to SRM

100%

SRM

80%

SRM

/CID

nk # 6

SRM Transition

Rank

40%

60%

of M

atch

ing

duct

Ion

Ran # 5

# 4

# 3

20%

40%

Freq

uenc

y o

Prod # 2

# 1

0%1 2 3 4

F

11

LTQ CID Product Ion Rank

MacCoss et al. J. Proteome Research 2008, 8(6), 2733-2739

127 peptides

Automatic Peptide and Transition Selection based on MS/MS spectral library by using Pinpoint

MS/MS data fromprevious discoveryexperimentsexperiments

A t t d E S l ti R b t

12

Automated, Easy, Selective, Robust

Automated Data Processing and Peptide Verification using Pinpoint

13

Biomarker Workflow

ClinicalDiscovery ValidationVerification

List of candidatemarkers generated

Need to verify and validate in expanded population

TSQ VantageLTQ Velos

LTQ Orbitrap Velos

Ion Trap-based mass specs

LTQ Orbitrap Velos

14

Unbiased Targeted Analysis: SRM Assays

Challenge for Peptide Quantitation using SRMHow to verify if the detected peak is the right targeted peptide

Which peak is my peptide?Samples are very complex and background interferences require

verification of the targeted peptide peak.

RT: 7 .00 - 8 .97

85

90

95

10 0 8 .5 3

8 .477 .0 4

7 .8 5

NL : 2 .33 E 2TIC F : + c E S I S RM m s2 43 3 .7 36 [3 1 3 .7 8 3 -3 1 3 .7 8 5 , 35 0 .9 14 -35 0 .9 16 , 4 3 2 .2 3 4 -4 3 2 .2 3 6 , 51 9 .5 38 -51 9 .5 40 , 6 3 2 .4 9 2 -6 3 2 .4 9 4 , 71 9 .5 37 -71 9 .5 39 ] M S S am p le 8

60

65

70

75

80

nce

3 0

35

40

45

50

55

Rel

ativ

e A

bund

an

8 .8 78 .6 5

7 .43 7 .7 6 8 .2 47 .30 7 .9 9

7 .6 0

5

10

15

20

257 .2 4 8 .178 .12

15

7 .0 7 .2 7 .4 7 .6 7 .8 8 .0 8 .2 8 .4 8 .6 8 .8Tim e (m in)

0

Peptide verification using isotopically labeled peptides as internal standards

N t ti l f l l tit ti t i id i th t d ti

498.8 > 784.4 VIFQGFTGK-14N7

• Not practical for large scale quantitative proteomics considering the cost and time to make the internal standards

498 8 > 509 3

305 509

784% y3

y4

y5

y6

y7

23 25 [min]

***

498.8 > 509.3255

452 637897

m/z

y4y8

*

504.3 > 793.4VIFQGFTGK-15N

309

515 793y7y5

y3

*504.3 > 515.3

258

457645

907

m/z

y8y6y4

16

23 25 [min]

Peptide verification using SRM triggered MS/MS spectrum

R i b d l i i d• Requires a broad mass selection window• MS/MS spectra often contain multiple components for complex biological samples • Need longer cycle time and less sensitive

C:\Xcalibur\serum_redesign\071707_qed_c 7/18/2007 9:48:55 PMSRM & QED, 1.5s, 2usRT: 9.83 - 59.86

80

90

10030.86 NL: 3.65E6

TIC F: + c ESI sid=10.00 SRM ms2 656.310@cid25.60 [753.379-753.381, 909.469-909.471, 1081.519-1081.521] MS 071707 d

Targeted Peptide:TEGDGVYTLNDKSRM traces of m/z 656 3 +2

Sequest search resultsagainst database of human_ref.fasta.

10

20

30

40

50

60

70

Rel

ativ

e A

bund

ance

071707_qed_cSRM traces of m/z 656.3 +2

10 15 20 25 30 35 40 45 50 55Time (min)

0

10

071707_qed_c #32354 RT: 30.78 AV: 1 NL: 2.71E6F: + c ESI sid=-10.00 d Full ms2 656.310@cid25.31 [30.000-1322.620]

80

90

100202.96

230.96

753.31

20

30

40

50

60

70

80

Rel

ativ

e A

bund

ance

1081.49

590.32167.84 909.41376.24139.98

403.08 489.31 1024.40541.18288.15442 10

QED MS/MS of m/z 656.3 +2

17

100 200 300 400 500 600 700 800 900 1000 1100 1200 1300m/z

0

10 442.10340.85 647.21 852.56686.5584.13 1210.331165.631007.70953.59763.67 1264.60

Peptide verification using composite MS/MS spectra(constituted with a large number of transitions per peptide)

• Measurements of multiple transitions per peptide a composite tandem mass• Measurements of multiple transitions per peptide - a composite tandem mass spectrum• Narrow mass selection window (higher selectivity; Higher sensitivity• Co-elution of all ions allows to eliminate false positive calls• Limited throughput if constantly monitoring large number of transitions per peptide

772 4

1072.6GILFVGSGVSGGEEGAR

539.3

569.3

640.4753.5

LVEDPQVIAPFLGK

Heavy peptide

• Limited throughput if constantly monitoring large number of transitions per peptide

TSQ SRMComposite MS/MS

TSQ SRMComposite MS/MS

628.4

685.4

772.4

871.5928.51015.5

1171.6

0

852.5

980.6

1077.7

1192.7

0

Heavy peptide

00000

20000

40000

60000

539.3569.3

640.40000

0000

772.4

1072.6

0600 800 1000 1200 600 800 1000 1200

LTQ CID MS/MS LTQ CID MS/MS

0

20000

40000

60000

80000

00000

753.5

852.5980.6

1077.7

1192.70

0000

628.4

685.4

871.5

928.51015.5

1171.6

Heavy peptide

18

0400 500 600 700 800 900 1000 1100 1200500 600 700 800 900 1000 1100 1200

Introducing intelligent SRM (iSRM)PLarge number of transitions are split into two sets

1. Primary SRM AcquisitionTIC of two

i i

P

PPrimary ion setgeneratesquantitative Information

Large number of transitions are split into two sets

2. Trigger DDSRM Acquisition

transitions

P

P

Secondary ion set generatescomposite MS/MS

3. Acquire Secondary SRMAcquisition

PS

S

S

spectrum for peptideverification

SS

SS

19

iSRM – An advantage for verifying low level targets in biological matrices

Primary SRM Transitionm/z 680.37 → 789.44NL: 2.48E2

S

50 atmolon column

Primary SRM Transitionm/z 680.37 → 959.54NL: 1.50E2

Data Dependent SRMPrimary and Secondary SRMPrimary and Secondary SRMTransitionsNL: 1.12E3

y3

y4 y6y9 y10

E L A S G L F P V G F K

y5 y7 y8

y10

20

Pinpoint software is integrated with iSRM workflow for automate assay design and data processing

SpectralLibraries

TargetedCandidate

Proteins

A

SRMLibrariesPeptide Digest

Method

B

LC-MS/MS: iSRM + T-SRM

690iSRM

Int.Identification

(Composite-spectra)

472

543

690

803 931

CiSRM

[min]

Quantification(Peak areas)

21

What can iSRM and Pinpoint software provide you?

The combination of Pinpoint and iSRM provides the capability to simultaneously verify and quantitate up to 1000 targeted peptides in a single HPLC/MS/MS experimentpeptides in a single HPLC/MS/MS experiment.

Rapid SRM assay design and refinement for the targeted protein quantification based on the discoverytargeted protein quantification based on the discovery data.

Provide high throughput screening methods toProvide high throughput screening methods to narrow down a large number of candidate biomarkers coming out of discovery to the most promising few.

Provide high throughput screening methods for signaling pathway studies.

22

Rapid SRM assay development and p y prefinement by using iSRM- Targeting 40 known yeast proteins

Initial peptide selection based on previous proteomics discovery data

1

2

Peptides detected inthe discovery experimentare selected

24

Automatic transition selection based on the MS/MS spectral library

3

} Primary transitions

3

Two primary and additional six secondary transitions per peptide are selected based on the fragment ion gintensity.

25

Export initial transitions to csv file for generating the initial iSRM method

Collision energy is calculated automaticallycalculated automatically by using formula of 0.034 x m/z of peptide +2

4

26

iSRM instrument method set up

1

24

3

5

27

Peptide sequence verification and quantitation from the initial SRM assay

Correlation p-value between compositeSRM spectra and MS/MS library spectra

28

Composite SRM spectra comparedwith MS/MS library spectra

Chromatogram of the primary and secondary transitions. Area calculated based on primary

SRM method refinement based on the initial SRM assay data

153 green colored peptidesWith confirmed nice LC peak shapes were selected fromthe initial targeted 368 peptidesfor the final iSRM methodfor the final iSRM method.

33

29

Peptide sequence verification and quantitative precision from the refined iSRM assay

30

CV percentage for targeting 153 verified peptides using iSRM

90

100

The final SRM assay for targeting 40 yeast

60

70

80

pept

ides

The final SRM assay for targeting 40 yeastproteins is very robust with good analytical precision

40

50

60

tota

l tar

ted

p

10

20

30

% o

f

0

10

0-5 6-10 11-15

% CV

31

% CV

A robust SRM assay for quantitation of 40 targeted yeast proteins was developed in a few short hours

RT:9.94 - 40.89

95

100

20.70588.309

Step 1- Select Target Peptides:369 peptides/2952 transitions (738 primary, 2214 secondary) were selected by Pinpoint based on the discovery data (20 min).

65

70

75

80

85

90

19.51718.372

Step 2- Acquire data:Data acquisition using above initial SRM assay (60 min).

35

40

45

50

55

60

Rel

ativ

e A

bund

ance 26.70

713.45519.43

718.37226.74

713.45520.81588.30919.39

718.37224.09

659.354 25.93702.40321.60

720.33419.34718.372 24.15

659354 3213

Step 3 – Verify data & optimize conditions:Data verification and generate optimized final SRM assay which kept only peptides that were confidently identified by composite MS/MS data and gave strong signal intensity 153

10

15

20

25

30

35 659.354 32.13474.263

26.82713.455

21.67720.334

28.67758.451

22.45244.165 32.02

474.263 32.28474.263 35.47

854.50919.24

718.37216.9416.17 34.00 3748 38911274

data and gave strong signal intensity. 153 peptides/1224 transitions (306 primary and 918 secondary) were retained in the final SRM assay (40 min) .

10 12 14 16 18 20 22 24 26 28 30 32 34 36 38 40Time (min)

0

5 873.442739.44634.00

805.40437.48

315.20238.91

1095.05512.74

704.357 Step 4 – Validate final assay:Validate the robustness and precision of final SRM assay by triplicate acquisition(60 min each, total 180 min).

32

Additional sensitivity can be obtained by optimizing collision energy in a single injection

33

Retention time shift correction on-the-fly provides robust quantitative results across different columns and instruments

The TSQ series of triple quandrupoleinstruments allows user to assign single or multiple peaks as RT reference compounds. With thereference compounds. With the changes of chromatographic conditions, the retention time shifts from these reference compounds are captured and a linear regressioncaptured and a linear regression curve will be generated on-the-fly that provides an offset and slope corredction to the tSRM windows of the compounds eluted afterwards. t e co pou ds e uted a te a dsThis approach enabels the use of narrow timed-SRM window without missing peaks due to retention time shifts, caused by column/mobile , yphase replacement, switching instrument, column clog or changing gradient conditions.

34

Targeted peak was detected by correcting the RT shift on the flyThe time shift was created by increasing the flow rate from 300 nl/min to 400 nl/min

heavypep 25fmolmatrix 2minwindow 9/1/2009 4:14:00 PMheavypep_25fmolmatrix_2minwindow 9/1/2009 4:14:00 PMheavy peptides,25fmol in matrix, 400nlheavypep_25fmolmatrix_2minwindowTSQ Vantage 613.317 854.445 24 11.65 23.65 + 1.000e+03 Reference 613.317 968.488 24 11.65 23.65 + 1.000e+03 Secondary 613.317 1055.520 24 11.65 23.65 + 1.000e+03 Secondary 521.793 451.266 22 21.33 23.33 + 1.000e+03 Secondary 521.793 588.325 22 21.33 23.33 + 1.000e+03 Secondary

RT: 23.73 - 27.88

70

80

90

100

e

25.56

25.5925.53

25.632549 521.793 659.362 22 21.33 23.33 + 1.000e+03 Secondary 521.793 730.399 22 21.33 23.33 + 1.000e+03 Secondary 521.793 801.436 22 21.33 23.33 + 1.000e+03 Primary 521.793 914.520 22 21.33 23.33 + 1.000e+03 Primary 521.793 985.557 22 21.33 23.33 + 1.000e+03 Secondary 526.797 461.274 22 21.33 23.33 + 1.000e+03 Secondary 526.797 598.333 22 21.33 23.33 + 1.000e+03 Secondary 526.797 669.370 22 21.33 23.33 + 1.000e+03 Secondary 526.797 740.407 22 21.33 23.33 + 1.000e+03 Secondary 30

40

50

60

70

Rel

ativ

e A

bund

ance

25.6325.49

25.6625.46

2569 526.797 811.444 22 21.33 23.33 + 1.000e+03 Reference 526.797 924.528 22 21.33 23.33 + 1.000e+03 Reference 526.797 995.565 22 21.33 23.33 + 1.000e+03 Secondary 769.888 376.219 30 25.75 27.75 + 1.000e+03 Secondary 769.888 489.303 30 25.75 27.75 + 1.000e+03 Secondary 769.888 652.366 30 25.75 27.75 + 1.000e+03 Secondary 769.888 753.414 30 25.75 27.75 + 1.000e+03 Secondary 769.888 868.441 30 25.75 27.75 + 1.000e+03 Primary 769.888 967.509 30 25.75 27.75 + 1.000e+03 Secondary 769 888 1024 530 30 25 75 27 75 + 1 000 +03 P i

90

1000

10

2025.69

25.43

25.7325.39

25.76 26.03 26.3225.33 26.4624.9624.23 24.4025.48

769.888 1024.530 30 25.75 27.75 + 1.000e+03 Primary 769.888 1111.562 30 25.75 27.75 + 1.000e+03 Secondary 769.888 1296.642 30 25.75 27.75 + 1.000e+03 Secondary 773.896 384.233 30 25.75 27.75 + 1.000e+03 Secondary 773.896 497.317 30 25.75 27.75 + 1.000e+03 Secondary 773.896 660.380 30 25.75 27.75 + 1.000e+03 Secondary 773.896 761.428 30 25.75 27.75 + 1.000e+03 Secondary 773.896 876.455 30 25.75 27.75 + 1.000e+03 Primary 773.896 975.523 30 25.75 27.75 + 1.000e+03 Secondary 773 896 1032 545 30 25 75 27 75 + 1 000e+03 Primary

50

60

70

80

Expected RT: 26.7 Observed RT: 25 5 773.896 1032.545 30 25.75 27.75 + 1.000e+03 Primary

773.896 1119.577 30 25.75 27.75 + 1.000e+03 Secondary 773.896 1304.657 30 25.75 27.75 + 1.000e+03 Secondary 796.407 675.305 30 29.46 31.46 + 1.000e+03 Secondary 796.407 762.337 30 29.46 31.46 + 1.000e+03 Secondary 796.407 861.406 30 29.46 31.46 + 1.000e+03 Secondary 796.407 918.427 30 29.46 31.46 + 1.000e+03 Primary 796.407 1005.459 30 29.46 31.46 + 1.000e+03 Secondary 796.407 1062.480 30 29.46 31.46 + 1.000e+03 Primary 796 407 1161 549 30 29 46 31 46 + 1 000e+03 Secondary 0

10

20

30

40Observed RT: 25.5

35

796.407 1161.549 30 29.46 31.46 + 1.000e+03 Secondary 24.0 24.5 25.0 25.5 26.0 26.5 27.0 27.5

Time (min)

Scaling Options using iSRM and Pinpoint

Regular multiple

Scheduled SRM

(tSRM) up to

tSRM

+

tSRM

+SRM

(tSRM) up to 3000 transitions iSRM (up to 16000

transitionsiSRM

+

Split-n-stitchAutomated scoring schemes to help prioritize large analysisAutomated scoring schemes to help prioritize large analysis into high, medium, low quality bins

And more…• Single software to help iterative method building to go from protein list to absolute abundance• Multi-threaded• Extremely easy data and results sharing• Customers can give video feedback

36

Customers can give video feedback• Video help tutorials to get you started

Optimizing assays for specificity & sensitivityBioworks

Acquire the Raw Files

SRMH-SRM

BioworksProteome DiscovererMascotSIEVEBiblioSpecPeptideAtlas

Build the SRM Assay

Pinpoint

P th

H-SRMT-SRMiSRM

PeptideAtlas Pinpoint

Process the Raw Files

Refine the Instrument

Verify Targeted Peptide Information

Ion Ratio CalcInstrumentMethod

T-SRM, CE optimizationHeavy peptidesProtein Panels

GenerateReports

P t i L l

Ion Ratio Calc.Costas-SorasP-ValueHeavy Labeled

Peptides

37

Protein Panels Protein LevelPeptide Level

Summary

Targeted protein quantitaiton using triple quadrupole mass spectrometers is complimentary to traditional western blots

Thermo Fisher Scientific’s workflow incorporates iSRM to facilitate rapid and robust method development for multiplexed assays.p y

Pinpoint software allows initial SRM/iSRM assay to be designed automatically by using discovery MS/MS data or th ti l di ti It l ff i tidtheoretical prediction. It also offers unique peptide verification scheme for SRM assay refinement.

Thermo Fisher Scientific’s workflow provides high-Thermo Fisher Scientific s workflow provides highthroughput peptide screens by simultaneously verifying and quantitating large number of peptides in a single HPLC MS/MS run

38

MS/MS run.