6
Global Plasma Proteome Quantification Using Internal Standard Triggered Targeted Analyses POSTER NOTE 65534 Authors Sebastien Gallien 1,2 , Jing Wang 2 , Aaron S. Gajadhar 3 , Bhavin Patel 4 , Markus Kellmann 5 , Tabiwang N. Arrey 5 , Alexander Harder 5 , Romain Huguet 3 , Graeme McAlister 3 , Derek Bailey 3 , Shannon Eliuk 3 , Yue Xuan 5 , Andreas Huhmer 3 , Emily L. Chen 2 1 Thermo Fisher Scientific, Paris, France; 2 Thermo Fisher Scientific, Precision Medicine Science Center, Cambridge, MA; 3 Thermo Fisher Scientific, San Jose, CA; 4 Thermo Fisher Scientific, Rockford, IL; 5 Thermo Fisher Scientific, Bremen, Germany ABSTRACT Purpose: A novel global plasma proteome quantification workflow has been developed. It relies on a new high density targeted acquisition method, called ‘SureQuant’ method, implemented on next generation Orbitrap mass spectrometers: Thermo Scientific TM Orbitrap Exploris TM 480 mass spectrometer and Thermo Scientific TM Orbitrap Eclipse TM Tribrid TM mass spectrometer. In addition, the workflow leverages a large set of stable isotopically labeled (SIL) peptides, used as internal standards, to drive the systematic screening of more than 500 plasma proteins (PQ500 kit from Biognosys). Methods: The SureQuant PQ500 analyses of undepleted plasma samples were performed using Orbitrap Exploris 480 and Orbitrap Eclipse Tribrid mass spectrometers, coupled to Thermo Scientific TM UltiMate TM 3000 RSCLC system or Thermo Scientific TM EASY-nLC TM 1200 system. An acquisition variant of the internal standard triggered targeted acquisition, dedicated to higher throughput analyses, was executed through an application programming interface (API) controlling a Thermo Scientific TM Q Exactive TM HF-X hybrid quadrupole-Orbitrap TM mass spectrometer coupled to a Evosep One LC system. PQ500 kit from Biognosys was spiked into the plasma samples to drive the targeted acquisition. Results: The SureQuant method has been adapted from the internal standard triggered PRM (IS-PRM) method in order to improve its usability and robustness, while retaining its analytical performance, and has been implemented in the native instrument control software of next generation Orbitrap mass spectrometers. The application of the method to the analysis of plasma samples, leveraging spiked-in SIL peptides from PQ500 kit, allowed robust and precise quantification of around 560 endogenous peptides used as surrogate of 400 plasma proteins using a 70-min LC gradient. Therefore, the method combined data quality of targeted analyses with proteome coverage of state-of-the-art profiling experiments. In addition, the availability of pre-set PQ500 SureQuant methods embedded into the instrument control software enables a ‘load and play’ execution of such plasma proteomics experiments. INTRODUCTION The success of plasma proteomics studies to deliver useful protein biomarkers has remained lower than initial expectations, prompting a redefinition of the plasma biomarker development pipeline. One proposed measure consisted in the implementation of a “rectangular” approach, relying exclusively on broad proteome profiling (through DDA or DIA analyses) across large cohorts for both the discovery and validation/verification stages. We describe here an alternative global quantification workflow, relying on a high density targeted acquisition method implemented on next generation Orbitrap mass spectrometers. This adaptation of the IS-PRM method leverages spiked-in stable isotopically labeled (SIL) peptides to drive the systematic screening of more than 500 plasma proteins per analysis across large sample sets.

Global Plasma Proteome Quantification Using Internal Standard … · 2019. 12. 20. · Global Plasma Proteome Quantification Using Internal Standard Triggered Targeted Analyses POSTER

  • Upload
    others

  • View
    4

  • Download
    0

Embed Size (px)

Citation preview

Page 1: Global Plasma Proteome Quantification Using Internal Standard … · 2019. 12. 20. · Global Plasma Proteome Quantification Using Internal Standard Triggered Targeted Analyses POSTER

Global Plasma Proteome Quantification Using Internal Standard Triggered Targeted Analyses

POSTER NOTE 65534

AuthorsSebastien Gallien1,2, Jing Wang2, Aaron S. Gajadhar3, Bhavin Patel4, Markus Kellmann5, Tabiwang N. Arrey5, Alexander Harder5, Romain Huguet3, Graeme McAlister3, Derek Bailey3, Shannon Eliuk3, Yue Xuan5, Andreas Huhmer3, Emily L. Chen2

1Thermo Fisher Scientific, Paris, France; 2Thermo Fisher Scientific, Precision Medicine Science Center, Cambridge, MA; 3Thermo Fisher Scientific, San Jose, CA; 4Thermo Fisher Scientific, Rockford, IL; 5Thermo Fisher Scientific, Bremen, Germany

Sebastien Gallien1,2, Jing Wang2, Aaron S. Gajadhar3, Bhavin Patel4, Markus Kellmann5, Tabiwang N. Arrey5, Alexander Harder5, Romain Huguet3, Graeme McAlister3, Derek Bailey3, Shannon Eliuk3, Yue Xuan5, Andreas Huhmer3, Emily L. Chen2

1Thermo Fisher Scientific, Paris, France; 2Thermo Fisher Scientific, Precision Medicine Science Center, Cambridge, MA; 3Thermo Fisher Scientific, San Jose, CA; 4Thermo Fisher Scientific, Rockford, IL; 5Thermo Fisher Scientific, Bremen, Germany

SureQuant MethodologyThe developed SureQuant acquisition method includes different scan events and filters, with parameters optimized for thehighest data quality and sensitivity of triggering, but also to retain sufficient triggering specificity (Figure 2). DuringSureQuant analyses, a high resolution (120k) full scan MS (1) is acquired, using lock mass re-calibration, to monitor thepredefined optimal precursor ions of the IS, based on the list of associated m/z values and intensity thresholds (defined at1% of the expected MS1 intensity at the chromatographic peak apex) included in the ‘Targeted Mass’ filter (2). In case ofthe detection of a candidate precursor ion satisfying the filtering criteria (MS peak with an m/z value within 3-10 ppmtolerance and intensity exceeding the threshold), it is subjected to fast MS/MS acquisition (3) performed at low resolutionand short fill times (typically 7.5k/10 ms), which is used as a second layer of confirmation of the detection of the IS.Several fragment ions, predefined as associated with the IS (6 optimal frag.) are included in the ‘Targeted Mass Trigger’filter (4), and the detection of a subset of them (at least 5 frag., 10-20 ppm tolerance) confirms the actual elution of the IS.This two-step process interrogating MS1 and MS2 data to track IS elution is the ‘watch’ mode of the SureQuant method.The detection of an IS in watch mode triggers the ‘quant’ mode, and therefore MS/MS acquisition of the correspondingendogenous pept. (5) with parameters favoring data quality, i.e., high resolution and long fill times (typically 60k/116 ms).The process is repeated over the entire LC separation with a systematic cycle-to-cycle assessment of the elution of the IS.The development of a SureQuant assay to be associated with a custom peptide panels requires the preliminarydetermination of the optimal precursor ion of each peptide target together with their intensity response and optimalassociated fragment ions. This information is typically extracted from an initial directed DDA analysis of the correspondingIS peptide mixture spiked into a representative matrix. The target list of multiple theoretical precursor ions of each ISpeptide included in such survey runs enables comprehensive data collection, and therefore the refinement of analytesdefinition for subsequent SureQuant analyses. The process is supported by Skyline software and facilitated by the genericdata acquisition and processing templates provided. The transfer of the generated information can be achieved throughcopy/paste from Skyline reports to ‘Targeted Mass’ and ‘Targeted Mass Trigger’ tables of SureQuant method. The surveyanalysis of a SureQuant assay needs to be run one time as the onset of a project without further adjustment over time,unless a major change in the experimental design is applied (e.g., LC set up, spiked amount of SIL peptides). In the lattercase, only the intensity responses of predefined IS precursor ions need to be re-adjusted though an additional survey runperformed under the modified experimental conditions. Therefore, the ability of embedding pre-set methods associatedwith predefined kits of IS peptides expedites SureQuant assay development at the user sites. Such validated methodtemplates have already been included in the instrument control software for the PQ500 related methods (Figure 3).

ABSTRACTPurpose: A novel global plasma proteome quantification workflow has been developed. It relies on a new high densitytargeted acquisition method, called ‘SureQuant’ method, implemented on next generation Orbitrap mass spectrometers:Thermo ScientificTM Orbitrap ExplorisTM 480 mass spectrometer and Thermo ScientificTM Orbitrap EclipseTM TribridTM

mass spectrometer. In addition, the workflow leverages a large set of stable isotopically labeled (SIL) peptides, used asinternal standards, to drive the systematic screening of more than 500 plasma proteins (PQ500 kit from Biognosys).

Methods: The SureQuant PQ500 analyses of undepleted plasma samples were performed using Orbitrap Exploris 480and Orbitrap Eclipse Tribrid mass spectrometers, coupled to Thermo ScientificTM UltiMateTM 3000 RSCLC system orThermo ScientificTM EASY-nLCTM 1200 system. An acquisition variant of the internal standard triggered targetedacquisition, dedicated to higher throughput analyses, was executed through an application programming interface (API)controlling a Thermo ScientificTM Q ExactiveTM HF-X hybrid quadrupole-OrbitrapTM mass spectrometer coupled to aEvosep One LC system. PQ500 kit from Biognosys was spiked into the plasma samples to drive the targeted acquisition.

Results: The SureQuant method has been adapted from the internal standard triggered PRM (IS-PRM) method in orderto improve its usability and robustness, while retaining its analytical performance, and has been implemented in the nativeinstrument control software of next generation Orbitrap mass spectrometers. The application of the method to the analysisof plasma samples, leveraging spiked-in SIL peptides from PQ500 kit, allowed robust and precise quantification of around560 endogenous peptides used as surrogate of 400 plasma proteins using a 70-min LC gradient. Therefore, the methodcombined data quality of targeted analyses with proteome coverage of state-of-the-art profiling experiments. In addition,the availability of pre-set PQ500 SureQuant methods embedded into the instrument control software enables a ‘load andplay’ execution of such plasma proteomics experiments.

INTRODUCTIONThe success of plasma proteomics studies to deliver useful protein biomarkers has remained lower than initialexpectations, prompting a redefinition of the plasma biomarker development pipeline. One proposed measure consisted inthe implementation of a “rectangular” approach, relying exclusively on broad proteome profiling (through DDA or DIAanalyses) across large cohorts for both the discovery and validation/verification stages. We describe here an alternativeglobal quantification workflow, relying on a high density targeted acquisition method implemented on next generationOrbitrap mass spectrometers. This adaptation of the IS-PRM method leverages spiked-in stable isotopically labeled (SIL)peptides to drive the systematic screening of more than 500 plasma proteins per analysis across large sample sets.

MATERIALS AND METHODSSample PreparationA commercial human plasma sample pooled from several individuals purchased from BioreclamationIVT was processedusing Thermo ScientificTM EasyPepTM Mini MS Sample Prep Kit.A set of 804 SIL peptides from PQ500 kit (Biognosys PN#Ki-3019-96) was spiked at around 80 fmol (median value) into 1µg of undepleted plasma sample. One µL of the sample was injected for LC-MS/MS analyses.

LC-MS/MS AnalysisFor SureQuant analyses, chromatographic separations were performed on UltiMate 3000 RSLC system equipped with C18trap cartridge (5 µm, 0.3 x 5 mm) and analytical column (2 μm, 0.15 x 150 mm), and EASY-nLC 1200 system equippedwith a C18 analytical column (2 μm, 0.15 x 150 mm), using a 70-min gradient separation. The LC systems were coupled toOrbitrap Exploris 480 and Orbitrap Eclipse Tribrid mass spectrometers. In SureQuant analyses, Full MS scans wereacquired at a resolution of 120k with a maximum fill time of 50 ms, while ddMS2 of IS and endogenous peptides wereacquired sequentially at a resolution of 7.5k or 60k and maximum fill times of 10 ms and 116 ms, respectively.The analyses using the high throughput IS-PRM acquisition variant were performed on Evosep One LC system equippedwith Evosep C18 analytical column (3 μm, 0.1 x 80 mm), using a 21-min gradient separation. The LC system was coupledto a Q Exactive HF-X controlled by an API. In the analyses performed with the ‘Single Scan Msx IS-PRM’ method (Figure1, right panel), PRM scans in watch mode were acquired at a resolution of 7.5k and a maximum fill time of 10 ms. In thequant. mode, the pairs of IS and endogenous peptides were measured simultaneously by ‘multiplex’ acquisition of theirfragments in a single PRM scan generated at a resolution of 120k and non-isochronous maximum fill times of 20 or 124ms for the IS and endogenous peptides, respectively.

CONCLUSIONS A novel global plasma IS triggered targeted quantification workflow, based on ‘SureQuant’ method has been

successfully implemented in the native instrument control software of Orbitrap Exploris 480 and Orbitrap EclipseTribrid MS.

The global plasma quantification workflow combines the figures of merit of targeted acquisition method with thecoverage capabilities of profiling methods for plasma proteomics. In addition, it exhibits an efficient portability and ahigh usability, benefiting from embedded preset PQ500 related methods.

The developed IS-PRM acquisition variant was explored to carry out high throughput analyses. It exhibited slightlylower performance, but also the capability for a 5-fold increase in throughput/experiment scale. Further investigation isneeded to evaluate the pertinence of its native implementation under the form of a SureQuant method variant.

For Research Use Only. Not for use in diagnostic procedures.

TRADEMARKS/LICENSING© 2019 Thermo Fisher Scientific Inc. All rights reserved. All trademarks are the property of Thermo Fisher Scientific andits subsidiaries. This information is not intended to encourage use of these products in any manner that might infringe theintellectual property rights of others.

Global Plasma Proteome Quantification Using Internal Standard Triggered Targeted Analyses

Figure 5. Sampling rate in SureQuant PQ500 global plasma analysis

SureQuant PQ500 Plasma QuantificationThe PQ500 SureQuant acquisition workflow has been applied to the analyses of 1 µg of undepleted plasma samplesupplemented with PQ500 kit (median spiked-in amount of SIL peptides of 80 fmol) using a 70-min LC gradient onUltiMate 3000 RSLC system. The survey run revealed that 802 of 804 SIL peptides were compatible (intensity responseand hydrophobicity) with the LC-MS setup used (Figure 4), and were retained for subsequent SureQuant analyses. Thehigh overall intensity response of the SIL peptides (median MS1 intensity at chromatographic peak apex of 1e8) insuredhigh acquisition efficiency (>90%) in SureQuant analyses, benefiting from the median MS1 triggering intensity threshold at1e6. This translated into a good overall sampling rate over the LC separation, exceeding 8 data points collected over theelution profile of the vast majority of the peptides, which also remained acceptable in the most crowded time ranges, with6 to 8 data points collected per peak (Figure 5). The 802 retained internal standards were systematically detected in thetriplicated SureQuant analyses of plasma samples, and triggered high quality measurement of the correspondingendogenous peptides, translating into a broad coverage of the undepleted plasma proteome (Figure 6). Around 400proteins were detected in each replicate (based on around 560 surrogate endogenous peptides), while 90% of the totalsets were systematically quantified across triplicates.

In addition to delivering a broad plasma proteome coverage, rivalling that of state-of-the-art profiling methods, SureQuantPQ500 analyses still provide targeted quantification performance (Figure 7). The amounts of endogenous peptides wereestimated based on the known amounts of IS injected and single-point quantification strategy. This revealed aquantification range between 4 amol and 15 pmol, spanning more than 6 orders of magnitude, which is in theory sufficientto cover the vast majority of current FDA-approved plasma biomarkers. The fragment ion signals of the lowest abundantquantified endogenous peptide (PFLVFIR, 4 amol/µg of plasma) have been extracted (Figure 7, lower panel), illustratingthe acceptable data quality even obtained for this challenging analyte. The quantification precision obtained on the entiredataset was excellent, with 75% of the endogenous peptides quantified with a CV below 10% (median CV value of 6%).The experiments were reproduced at a second site by another operator, using a different LC set-up (EASY-nLC 1200system operated in a one-column setup instead of UltiMate 3000 RSLC system operated in a two-column setup) (Figure8). After initial survey analysis, enabling the adjustment of internal standards MS1 triggering intensity threshold,SureQuant PQ500 analyses of undepleted plasma sample were repeated. It turned out that the quantification of theendogenous peptides between sites / systems / operators was very consistent, as reflected by the quantitative valuestightly aligned along the theoretical ’perfect’ correlation line, and the absence of bias revealed by the equation of theregression. The acquisition scheme of SureQuant method, fully avoiding any dependency on time-scheduling, enablesefficient method portability and strong robustness.

The analyses of plasma sample supplemented with PQ500 kit were reproduced using the ‘Single Scan Msx IS-PRM’acquisition method on QE HF-X MS coupled to Evosep One LC system operated at a 60 samples/day analyticalthroughput (21-min LC gradient). The sampling rate under these conditions remained significantly higher than thepredefined (relaxed) limit of 4-5 data points collected per peptide chromatographic peak (Figure 9, upper panel). Incomparison with SureQuant analyses performed with longer LC gradient, a substantial number of IS peptides (81) werenot detected, corresponding to the most hydrophilic peptides, which are not compatible with the LC setup used (Figure 9,lower panel). The remaining 721 IS peptides were properly detected in the watch mode of the method and triggeredmultiplex PRM measurement in the quant. mode, enabling the detection of 421 endogenous peptides, surrogate of 301plasma proteins. In Figure 10, the data generated by SureQuant and Single Scan Msx IS-PRM analyses are displayed forone endogenous peptide quantified by both methods. The quantification results were very consistent across the methods.

Figure 3. Overall PQ500 SureQuant acquisition workflow.

Figure 6. Peptide detection/quantification in SureQuant PQ500 analyses of undepleted plasma

Figure 7. Quantification performance of SureQuant PQ500 analyses of undepleted plasma

Figure 8. Consistency of SureQuant PQ500 Quantification across sites/systems

Figure 9. High throughput plasma analyses using Single Scan Msx IS-PRM method.

Figure 10. Endogenous peptide quantification from SureQuant and Single Scan Msx IS-PRM

Figure 1. Sequential and multiplex acquisition modes

LC-MS Analysis

SIL peptides (internal standards)

#1#2

#n….

Sample SureQuant

ENDO Peptides Pseudo Spectral Matching

‘QUANT MODE’High Quality MS2 of Target

‘WATCH MODE’

Fast, Low Res MS2 of Trigger

1

4

5

2 3

MS1 Survey Scan

Mix

MS/MS IS

IS peptides (trigger)

1

2

3

4

5

Figure 2. SureQuant acquisition Scheme and method structure

00.5

11.5

22.5

33.5

4

0 10 20 30 40 50 60 70 80 90[min]

Cycle Time [s]

Sampling rate

6-8 data points /peak(20-s peak width)

≥8 data points /peak(20-s peak width)

Intensity response of precursor ions

PQ500 kit in plasma

Survey Run

SureQuant

LC-MS Analysis

PQ500 kit Plasma sample LC-MS Analysis

Survey Run Analysis

Export Targeted Mass report

Update Targeted Mass intensity threshold

Target Quantification

PQ500 Directed DDAInclusion list optimal IS precursor ions

Ready-to-use template

Ready-to-use template PQ500 SureQuant

PQ500 Skyline analysis template

PQ500 Skyline analysis template

Targeted Mass

Copy/Paste

589 total peptides

Rep 2Rep 1

Rep 3

526 peptides

421 total proteins

Rep 2Rep 1

Rep 3

368 proteins

0.001

0.01

0.1

1

10

100

1000

10000

100000

0.001 0.01 0.1 1 10 100 1000 10000 100000

Site 2-Endo. Pep. Conc. [fmol/µg plasma]

Site 1-Endo. Pep. Conc. [fmol/µg plasma]

Theoretical ‘perfect’ correlationActual correlation

R2 = 0.97y = 1.0009 x + 0.0881

0

1

2

3

4

0 5 10 15 20

0.E+00

2.E+07

4.E+07

6.E+07

4.45 4.65 4.850.E+00

2.E+07

4.E+07

6.E+07

23 23.2 23.4 23.60.E+00

2.E+07

4.E+07

4.45 4.65 4.85

FQASVATPR

FQASVATPRWATCH MODE QUANT MODE

0.E+00

1.E+05

2.E+05

3.E+05

4.45 4.65 4.850.E+00

1.E+05

2.E+05

3.E+05

23 23.2 23.4 23.6

Single-Scan Msx IS-PRM (QE HF-X) SureQuant (Exploris 480)

Number of spectra including ≥ 5 ref. fragments

Peptide# 1…………………………………………………………………………………………………….802

Detection endo. pep. / prot.

564 pep. / 397 prot.

560 pep. / 396 prot.

550 pep. / 390 prot.

Internal Standards peptides

Endogenous peptides

0 1-2 ≥ 3

802 peptides

802 peptides

802 peptides

Internal Standards peptides

Endogenous peptides

Rep. 1

Rep. 2

Rep. 3

Rep. 1

Rep. 2

Rep. 3

Peptide# 1………………………………………………………………………………………………….802

Sampling rate

≥ 4-5 data points /peak(12-s peak width)

Internal Standards peptides

Endogenous peptides

Number of spectra including ≥ 4 ref. fragments 0 1

Detection pep. / prot.

712 peptides

421 pep. / 301 prot.

Non-detectable IS Hydrophobicity scale

- +

Intensity ratio ENDO/SIL (Ss Msx IS-PRM): 0.0051

Area ratio ENDO/SIL (SureQuant): 0.0049

[min]

Cycle Time [s]

0

50

100

150

200

250

300

<5% 5-10% 10-20% >20%

CV

0.001

0.01

0.1

1

10

100

1000

10000

0 100 200 300 400 500 600

Endo. Pep. Conc. [fmol/µg plasma]

Abundance Rank

AlbuNumber of endo. Peptide.

56.4 56.6 56.8 5756.4 56.6 56.8 57

SIL (10fmol) ENDO (calc: 4amol) PFLVFIR

8e7 4e4

[min] [min]

I IQuantification range: 4 amol – 15 pmol(>6 orders of magnitude)

CV < 10% for more than 75% of the peptides (median CV: 5.7%)

Sequential acquisition Multiplex acquisition

Figure 4. Characterization of PQ500 kit through survey run analysis.

1.0E+06

1.0E+07

1.0E+08

1.0E+09Median intensity: 1e8

MS1 Intensity Apex IS

1.0

10.0

100.0

1,000.0

10,000.0 Median amount: 80 fmol

Injected Amount IS [fmol]

1………………………………………………………………………………………………………………………….802Peptide#

High Throughput IS Triggered Plasma AnalysesA variant of SureQuant acquisition, dedicated to higher throughput analyses, has been developed using API-based IS-PRM application, which is only compatible with QE MS platforms. This method has not been natively implemented in theinstrument control software of orbitrap instruments and does not benefit from the revised acquisition strategy ofSureQuant method, but relies on the acquisition scheme of the original IS-PRM method. More specifically, in this ‘SingleScan Msx IS-PRM’ acquisition variant, the detection of an IS peptide in watch mode triggers the concomitantmeasurement of the corresponding pair of IS and endogenous peptides in a single PRM scan in quant. mode close to theirelution profile apex, following their ‘multiplex’ isolation (Figure 1). This contrast with the regular sequential acquisition, inwhich data points are collected over the full elution profiles of the peptides. With this acquisition variant, quantification isperformed based on the fragment ion intensities extracted from single high quality PRM scan, and not on the area underthe curve of the corresponding extracted fragment traces. This quantification strategy alleviates sampling rate constraint.

[min]

I

PO65534-EN0519S

Page 2: Global Plasma Proteome Quantification Using Internal Standard … · 2019. 12. 20. · Global Plasma Proteome Quantification Using Internal Standard Triggered Targeted Analyses POSTER

Sebastien Gallien1,2, Jing Wang2, Aaron S. Gajadhar3, Bhavin Patel4, Markus Kellmann5, Tabiwang N. Arrey5, Alexander Harder5, Romain Huguet3, Graeme McAlister3, Derek Bailey3, Shannon Eliuk3, Yue Xuan5, Andreas Huhmer3, Emily L. Chen2

1Thermo Fisher Scientific, Paris, France; 2Thermo Fisher Scientific, Precision Medicine Science Center, Cambridge, MA; 3Thermo Fisher Scientific, San Jose, CA; 4Thermo Fisher Scientific, Rockford, IL; 5Thermo Fisher Scientific, Bremen, Germany

SureQuant MethodologyThe developed SureQuant acquisition method includes different scan events and filters, with parameters optimized for thehighest data quality and sensitivity of triggering, but also to retain sufficient triggering specificity (Figure 2). DuringSureQuant analyses, a high resolution (120k) full scan MS (1) is acquired, using lock mass re-calibration, to monitor thepredefined optimal precursor ions of the IS, based on the list of associated m/z values and intensity thresholds (defined at1% of the expected MS1 intensity at the chromatographic peak apex) included in the ‘Targeted Mass’ filter (2). In case ofthe detection of a candidate precursor ion satisfying the filtering criteria (MS peak with an m/z value within 3-10 ppmtolerance and intensity exceeding the threshold), it is subjected to fast MS/MS acquisition (3) performed at low resolutionand short fill times (typically 7.5k/10 ms), which is used as a second layer of confirmation of the detection of the IS.Several fragment ions, predefined as associated with the IS (6 optimal frag.) are included in the ‘Targeted Mass Trigger’filter (4), and the detection of a subset of them (at least 5 frag., 10-20 ppm tolerance) confirms the actual elution of the IS.This two-step process interrogating MS1 and MS2 data to track IS elution is the ‘watch’ mode of the SureQuant method.The detection of an IS in watch mode triggers the ‘quant’ mode, and therefore MS/MS acquisition of the correspondingendogenous pept. (5) with parameters favoring data quality, i.e., high resolution and long fill times (typically 60k/116 ms).The process is repeated over the entire LC separation with a systematic cycle-to-cycle assessment of the elution of the IS.The development of a SureQuant assay to be associated with a custom peptide panels requires the preliminarydetermination of the optimal precursor ion of each peptide target together with their intensity response and optimalassociated fragment ions. This information is typically extracted from an initial directed DDA analysis of the correspondingIS peptide mixture spiked into a representative matrix. The target list of multiple theoretical precursor ions of each ISpeptide included in such survey runs enables comprehensive data collection, and therefore the refinement of analytesdefinition for subsequent SureQuant analyses. The process is supported by Skyline software and facilitated by the genericdata acquisition and processing templates provided. The transfer of the generated information can be achieved throughcopy/paste from Skyline reports to ‘Targeted Mass’ and ‘Targeted Mass Trigger’ tables of SureQuant method. The surveyanalysis of a SureQuant assay needs to be run one time as the onset of a project without further adjustment over time,unless a major change in the experimental design is applied (e.g., LC set up, spiked amount of SIL peptides). In the lattercase, only the intensity responses of predefined IS precursor ions need to be re-adjusted though an additional survey runperformed under the modified experimental conditions. Therefore, the ability of embedding pre-set methods associatedwith predefined kits of IS peptides expedites SureQuant assay development at the user sites. Such validated methodtemplates have already been included in the instrument control software for the PQ500 related methods (Figure 3).

ABSTRACTPurpose: A novel global plasma proteome quantification workflow has been developed. It relies on a new high densitytargeted acquisition method, called ‘SureQuant’ method, implemented on next generation Orbitrap mass spectrometers:Thermo ScientificTM Orbitrap ExplorisTM 480 mass spectrometer and Thermo ScientificTM Orbitrap EclipseTM TribridTM

mass spectrometer. In addition, the workflow leverages a large set of stable isotopically labeled (SIL) peptides, used asinternal standards, to drive the systematic screening of more than 500 plasma proteins (PQ500 kit from Biognosys).

Methods: The SureQuant PQ500 analyses of undepleted plasma samples were performed using Orbitrap Exploris 480and Orbitrap Eclipse Tribrid mass spectrometers, coupled to Thermo ScientificTM UltiMateTM 3000 RSCLC system orThermo ScientificTM EASY-nLCTM 1200 system. An acquisition variant of the internal standard triggered targetedacquisition, dedicated to higher throughput analyses, was executed through an application programming interface (API)controlling a Thermo ScientificTM Q ExactiveTM HF-X hybrid quadrupole-OrbitrapTM mass spectrometer coupled to aEvosep One LC system. PQ500 kit from Biognosys was spiked into the plasma samples to drive the targeted acquisition.

Results: The SureQuant method has been adapted from the internal standard triggered PRM (IS-PRM) method in orderto improve its usability and robustness, while retaining its analytical performance, and has been implemented in the nativeinstrument control software of next generation Orbitrap mass spectrometers. The application of the method to the analysisof plasma samples, leveraging spiked-in SIL peptides from PQ500 kit, allowed robust and precise quantification of around560 endogenous peptides used as surrogate of 400 plasma proteins using a 70-min LC gradient. Therefore, the methodcombined data quality of targeted analyses with proteome coverage of state-of-the-art profiling experiments. In addition,the availability of pre-set PQ500 SureQuant methods embedded into the instrument control software enables a ‘load andplay’ execution of such plasma proteomics experiments.

INTRODUCTIONThe success of plasma proteomics studies to deliver useful protein biomarkers has remained lower than initialexpectations, prompting a redefinition of the plasma biomarker development pipeline. One proposed measure consisted inthe implementation of a “rectangular” approach, relying exclusively on broad proteome profiling (through DDA or DIAanalyses) across large cohorts for both the discovery and validation/verification stages. We describe here an alternativeglobal quantification workflow, relying on a high density targeted acquisition method implemented on next generationOrbitrap mass spectrometers. This adaptation of the IS-PRM method leverages spiked-in stable isotopically labeled (SIL)peptides to drive the systematic screening of more than 500 plasma proteins per analysis across large sample sets.

MATERIALS AND METHODSSample PreparationA commercial human plasma sample pooled from several individuals purchased from BioreclamationIVT was processedusing Thermo ScientificTM EasyPepTM Mini MS Sample Prep Kit.A set of 804 SIL peptides from PQ500 kit (Biognosys PN#Ki-3019-96) was spiked at around 80 fmol (median value) into 1µg of undepleted plasma sample. One µL of the sample was injected for LC-MS/MS analyses.

LC-MS/MS AnalysisFor SureQuant analyses, chromatographic separations were performed on UltiMate 3000 RSLC system equipped with C18trap cartridge (5 µm, 0.3 x 5 mm) and analytical column (2 μm, 0.15 x 150 mm), and EASY-nLC 1200 system equippedwith a C18 analytical column (2 μm, 0.15 x 150 mm), using a 70-min gradient separation. The LC systems were coupled toOrbitrap Exploris 480 and Orbitrap Eclipse Tribrid mass spectrometers. In SureQuant analyses, Full MS scans wereacquired at a resolution of 120k with a maximum fill time of 50 ms, while ddMS2 of IS and endogenous peptides wereacquired sequentially at a resolution of 7.5k or 60k and maximum fill times of 10 ms and 116 ms, respectively.The analyses using the high throughput IS-PRM acquisition variant were performed on Evosep One LC system equippedwith Evosep C18 analytical column (3 μm, 0.1 x 80 mm), using a 21-min gradient separation. The LC system was coupledto a Q Exactive HF-X controlled by an API. In the analyses performed with the ‘Single Scan Msx IS-PRM’ method (Figure1, right panel), PRM scans in watch mode were acquired at a resolution of 7.5k and a maximum fill time of 10 ms. In thequant. mode, the pairs of IS and endogenous peptides were measured simultaneously by ‘multiplex’ acquisition of theirfragments in a single PRM scan generated at a resolution of 120k and non-isochronous maximum fill times of 20 or 124ms for the IS and endogenous peptides, respectively.

CONCLUSIONS A novel global plasma IS triggered targeted quantification workflow, based on ‘SureQuant’ method has been

successfully implemented in the native instrument control software of Orbitrap Exploris 480 and Orbitrap EclipseTribrid MS.

The global plasma quantification workflow combines the figures of merit of targeted acquisition method with thecoverage capabilities of profiling methods for plasma proteomics. In addition, it exhibits an efficient portability and ahigh usability, benefiting from embedded preset PQ500 related methods.

The developed IS-PRM acquisition variant was explored to carry out high throughput analyses. It exhibited slightlylower performance, but also the capability for a 5-fold increase in throughput/experiment scale. Further investigation isneeded to evaluate the pertinence of its native implementation under the form of a SureQuant method variant.

For Research Use Only. Not for use in diagnostic procedures.

TRADEMARKS/LICENSING© 2019 Thermo Fisher Scientific Inc. All rights reserved. All trademarks are the property of Thermo Fisher Scientific andits subsidiaries. This information is not intended to encourage use of these products in any manner that might infringe theintellectual property rights of others.

Global Plasma Proteome Quantification Using Internal Standard Triggered Targeted Analyses

Figure 5. Sampling rate in SureQuant PQ500 global plasma analysis

SureQuant PQ500 Plasma QuantificationThe PQ500 SureQuant acquisition workflow has been applied to the analyses of 1 µg of undepleted plasma samplesupplemented with PQ500 kit (median spiked-in amount of SIL peptides of 80 fmol) using a 70-min LC gradient onUltiMate 3000 RSLC system. The survey run revealed that 802 of 804 SIL peptides were compatible (intensity responseand hydrophobicity) with the LC-MS setup used (Figure 4), and were retained for subsequent SureQuant analyses. Thehigh overall intensity response of the SIL peptides (median MS1 intensity at chromatographic peak apex of 1e8) insuredhigh acquisition efficiency (>90%) in SureQuant analyses, benefiting from the median MS1 triggering intensity threshold at1e6. This translated into a good overall sampling rate over the LC separation, exceeding 8 data points collected over theelution profile of the vast majority of the peptides, which also remained acceptable in the most crowded time ranges, with6 to 8 data points collected per peak (Figure 5). The 802 retained internal standards were systematically detected in thetriplicated SureQuant analyses of plasma samples, and triggered high quality measurement of the correspondingendogenous peptides, translating into a broad coverage of the undepleted plasma proteome (Figure 6). Around 400proteins were detected in each replicate (based on around 560 surrogate endogenous peptides), while 90% of the totalsets were systematically quantified across triplicates.

In addition to delivering a broad plasma proteome coverage, rivalling that of state-of-the-art profiling methods, SureQuantPQ500 analyses still provide targeted quantification performance (Figure 7). The amounts of endogenous peptides wereestimated based on the known amounts of IS injected and single-point quantification strategy. This revealed aquantification range between 4 amol and 15 pmol, spanning more than 6 orders of magnitude, which is in theory sufficientto cover the vast majority of current FDA-approved plasma biomarkers. The fragment ion signals of the lowest abundantquantified endogenous peptide (PFLVFIR, 4 amol/µg of plasma) have been extracted (Figure 7, lower panel), illustratingthe acceptable data quality even obtained for this challenging analyte. The quantification precision obtained on the entiredataset was excellent, with 75% of the endogenous peptides quantified with a CV below 10% (median CV value of 6%).The experiments were reproduced at a second site by another operator, using a different LC set-up (EASY-nLC 1200system operated in a one-column setup instead of UltiMate 3000 RSLC system operated in a two-column setup) (Figure8). After initial survey analysis, enabling the adjustment of internal standards MS1 triggering intensity threshold,SureQuant PQ500 analyses of undepleted plasma sample were repeated. It turned out that the quantification of theendogenous peptides between sites / systems / operators was very consistent, as reflected by the quantitative valuestightly aligned along the theoretical ’perfect’ correlation line, and the absence of bias revealed by the equation of theregression. The acquisition scheme of SureQuant method, fully avoiding any dependency on time-scheduling, enablesefficient method portability and strong robustness.

The analyses of plasma sample supplemented with PQ500 kit were reproduced using the ‘Single Scan Msx IS-PRM’acquisition method on QE HF-X MS coupled to Evosep One LC system operated at a 60 samples/day analyticalthroughput (21-min LC gradient). The sampling rate under these conditions remained significantly higher than thepredefined (relaxed) limit of 4-5 data points collected per peptide chromatographic peak (Figure 9, upper panel). Incomparison with SureQuant analyses performed with longer LC gradient, a substantial number of IS peptides (81) werenot detected, corresponding to the most hydrophilic peptides, which are not compatible with the LC setup used (Figure 9,lower panel). The remaining 721 IS peptides were properly detected in the watch mode of the method and triggeredmultiplex PRM measurement in the quant. mode, enabling the detection of 421 endogenous peptides, surrogate of 301plasma proteins. In Figure 10, the data generated by SureQuant and Single Scan Msx IS-PRM analyses are displayed forone endogenous peptide quantified by both methods. The quantification results were very consistent across the methods.

Figure 3. Overall PQ500 SureQuant acquisition workflow.

Figure 6. Peptide detection/quantification in SureQuant PQ500 analyses of undepleted plasma

Figure 7. Quantification performance of SureQuant PQ500 analyses of undepleted plasma

Figure 8. Consistency of SureQuant PQ500 Quantification across sites/systems

Figure 9. High throughput plasma analyses using Single Scan Msx IS-PRM method.

Figure 10. Endogenous peptide quantification from SureQuant and Single Scan Msx IS-PRM

Figure 1. Sequential and multiplex acquisition modes

LC-MS Analysis

SIL peptides (internal standards)

#1#2

#n….

Sample SureQuant

ENDO Peptides Pseudo Spectral Matching

‘QUANT MODE’High Quality MS2 of Target

‘WATCH MODE’

Fast, Low Res MS2 of Trigger

1

4

5

2 3

MS1 Survey Scan

Mix

MS/MS IS

IS peptides (trigger)

1

2

3

4

5

Figure 2. SureQuant acquisition Scheme and method structure

00.5

11.5

22.5

33.5

4

0 10 20 30 40 50 60 70 80 90[min]

Cycle Time [s]

Sampling rate

6-8 data points /peak(20-s peak width)

≥8 data points /peak(20-s peak width)

Intensity response of precursor ions

PQ500 kit in plasma

Survey Run

SureQuant

LC-MS Analysis

PQ500 kit Plasma sample LC-MS Analysis

Survey Run Analysis

Export Targeted Mass report

Update Targeted Mass intensity threshold

Target Quantification

PQ500 Directed DDAInclusion list optimal IS precursor ions

Ready-to-use template

Ready-to-use template PQ500 SureQuant

PQ500 Skyline analysis template

PQ500 Skyline analysis template

Targeted Mass

Copy/Paste

589 total peptides

Rep 2Rep 1

Rep 3

526 peptides

421 total proteins

Rep 2Rep 1

Rep 3

368 proteins

0.001

0.01

0.1

1

10

100

1000

10000

100000

0.001 0.01 0.1 1 10 100 1000 10000 100000

Site 2-Endo. Pep. Conc. [fmol/µg plasma]

Site 1-Endo. Pep. Conc. [fmol/µg plasma]

Theoretical ‘perfect’ correlationActual correlation

R2 = 0.97y = 1.0009 x + 0.0881

0

1

2

3

4

0 5 10 15 20

0.E+00

2.E+07

4.E+07

6.E+07

4.45 4.65 4.850.E+00

2.E+07

4.E+07

6.E+07

23 23.2 23.4 23.60.E+00

2.E+07

4.E+07

4.45 4.65 4.85

FQASVATPR

FQASVATPRWATCH MODE QUANT MODE

0.E+00

1.E+05

2.E+05

3.E+05

4.45 4.65 4.850.E+00

1.E+05

2.E+05

3.E+05

23 23.2 23.4 23.6

Single-Scan Msx IS-PRM (QE HF-X) SureQuant (Exploris 480)

Number of spectra including ≥ 5 ref. fragments

Peptide# 1…………………………………………………………………………………………………….802

Detection endo. pep. / prot.

564 pep. / 397 prot.

560 pep. / 396 prot.

550 pep. / 390 prot.

Internal Standards peptides

Endogenous peptides

0 1-2 ≥ 3

802 peptides

802 peptides

802 peptides

Internal Standards peptides

Endogenous peptides

Rep. 1

Rep. 2

Rep. 3

Rep. 1

Rep. 2

Rep. 3

Peptide# 1………………………………………………………………………………………………….802

Sampling rate

≥ 4-5 data points /peak(12-s peak width)

Internal Standards peptides

Endogenous peptides

Number of spectra including ≥ 4 ref. fragments 0 1

Detection pep. / prot.

712 peptides

421 pep. / 301 prot.

Non-detectable IS Hydrophobicity scale

- +

Intensity ratio ENDO/SIL (Ss Msx IS-PRM): 0.0051

Area ratio ENDO/SIL (SureQuant): 0.0049

[min]

Cycle Time [s]

0

50

100

150

200

250

300

<5% 5-10% 10-20% >20%

CV

0.001

0.01

0.1

1

10

100

1000

10000

0 100 200 300 400 500 600

Endo. Pep. Conc. [fmol/µg plasma]

Abundance Rank

AlbuNumber of endo. Peptide.

56.4 56.6 56.8 5756.4 56.6 56.8 57

SIL (10fmol) ENDO (calc: 4amol) PFLVFIR

8e7 4e4

[min] [min]

I IQuantification range: 4 amol – 15 pmol(>6 orders of magnitude)

CV < 10% for more than 75% of the peptides (median CV: 5.7%)

Sequential acquisition Multiplex acquisition

Figure 4. Characterization of PQ500 kit through survey run analysis.

1.0E+06

1.0E+07

1.0E+08

1.0E+09Median intensity: 1e8

MS1 Intensity Apex IS

1.0

10.0

100.0

1,000.0

10,000.0 Median amount: 80 fmol

Injected Amount IS [fmol]

1………………………………………………………………………………………………………………………….802Peptide#

High Throughput IS Triggered Plasma AnalysesA variant of SureQuant acquisition, dedicated to higher throughput analyses, has been developed using API-based IS-PRM application, which is only compatible with QE MS platforms. This method has not been natively implemented in theinstrument control software of orbitrap instruments and does not benefit from the revised acquisition strategy ofSureQuant method, but relies on the acquisition scheme of the original IS-PRM method. More specifically, in this ‘SingleScan Msx IS-PRM’ acquisition variant, the detection of an IS peptide in watch mode triggers the concomitantmeasurement of the corresponding pair of IS and endogenous peptides in a single PRM scan in quant. mode close to theirelution profile apex, following their ‘multiplex’ isolation (Figure 1). This contrast with the regular sequential acquisition, inwhich data points are collected over the full elution profiles of the peptides. With this acquisition variant, quantification isperformed based on the fragment ion intensities extracted from single high quality PRM scan, and not on the area underthe curve of the corresponding extracted fragment traces. This quantification strategy alleviates sampling rate constraint.

[min]

I

PO65534-EN0519S

Page 3: Global Plasma Proteome Quantification Using Internal Standard … · 2019. 12. 20. · Global Plasma Proteome Quantification Using Internal Standard Triggered Targeted Analyses POSTER

Sebastien Gallien1,2, Jing Wang2, Aaron S. Gajadhar3, Bhavin Patel4, Markus Kellmann5, Tabiwang N. Arrey5, Alexander Harder5, Romain Huguet3, Graeme McAlister3, Derek Bailey3, Shannon Eliuk3, Yue Xuan5, Andreas Huhmer3, Emily L. Chen2

1Thermo Fisher Scientific, Paris, France; 2Thermo Fisher Scientific, Precision Medicine Science Center, Cambridge, MA; 3Thermo Fisher Scientific, San Jose, CA; 4Thermo Fisher Scientific, Rockford, IL; 5Thermo Fisher Scientific, Bremen, Germany

SureQuant MethodologyThe developed SureQuant acquisition method includes different scan events and filters, with parameters optimized for thehighest data quality and sensitivity of triggering, but also to retain sufficient triggering specificity (Figure 2). DuringSureQuant analyses, a high resolution (120k) full scan MS (1) is acquired, using lock mass re-calibration, to monitor thepredefined optimal precursor ions of the IS, based on the list of associated m/z values and intensity thresholds (defined at1% of the expected MS1 intensity at the chromatographic peak apex) included in the ‘Targeted Mass’ filter (2). In case ofthe detection of a candidate precursor ion satisfying the filtering criteria (MS peak with an m/z value within 3-10 ppmtolerance and intensity exceeding the threshold), it is subjected to fast MS/MS acquisition (3) performed at low resolutionand short fill times (typically 7.5k/10 ms), which is used as a second layer of confirmation of the detection of the IS.Several fragment ions, predefined as associated with the IS (6 optimal frag.) are included in the ‘Targeted Mass Trigger’filter (4), and the detection of a subset of them (at least 5 frag., 10-20 ppm tolerance) confirms the actual elution of the IS.This two-step process interrogating MS1 and MS2 data to track IS elution is the ‘watch’ mode of the SureQuant method.The detection of an IS in watch mode triggers the ‘quant’ mode, and therefore MS/MS acquisition of the correspondingendogenous pept. (5) with parameters favoring data quality, i.e., high resolution and long fill times (typically 60k/116 ms).The process is repeated over the entire LC separation with a systematic cycle-to-cycle assessment of the elution of the IS.The development of a SureQuant assay to be associated with a custom peptide panels requires the preliminarydetermination of the optimal precursor ion of each peptide target together with their intensity response and optimalassociated fragment ions. This information is typically extracted from an initial directed DDA analysis of the correspondingIS peptide mixture spiked into a representative matrix. The target list of multiple theoretical precursor ions of each ISpeptide included in such survey runs enables comprehensive data collection, and therefore the refinement of analytesdefinition for subsequent SureQuant analyses. The process is supported by Skyline software and facilitated by the genericdata acquisition and processing templates provided. The transfer of the generated information can be achieved throughcopy/paste from Skyline reports to ‘Targeted Mass’ and ‘Targeted Mass Trigger’ tables of SureQuant method. The surveyanalysis of a SureQuant assay needs to be run one time as the onset of a project without further adjustment over time,unless a major change in the experimental design is applied (e.g., LC set up, spiked amount of SIL peptides). In the lattercase, only the intensity responses of predefined IS precursor ions need to be re-adjusted though an additional survey runperformed under the modified experimental conditions. Therefore, the ability of embedding pre-set methods associatedwith predefined kits of IS peptides expedites SureQuant assay development at the user sites. Such validated methodtemplates have already been included in the instrument control software for the PQ500 related methods (Figure 3).

ABSTRACTPurpose: A novel global plasma proteome quantification workflow has been developed. It relies on a new high densitytargeted acquisition method, called ‘SureQuant’ method, implemented on next generation Orbitrap mass spectrometers:Thermo ScientificTM Orbitrap ExplorisTM 480 mass spectrometer and Thermo ScientificTM Orbitrap EclipseTM TribridTM

mass spectrometer. In addition, the workflow leverages a large set of stable isotopically labeled (SIL) peptides, used asinternal standards, to drive the systematic screening of more than 500 plasma proteins (PQ500 kit from Biognosys).

Methods: The SureQuant PQ500 analyses of undepleted plasma samples were performed using Orbitrap Exploris 480and Orbitrap Eclipse Tribrid mass spectrometers, coupled to Thermo ScientificTM UltiMateTM 3000 RSCLC system orThermo ScientificTM EASY-nLCTM 1200 system. An acquisition variant of the internal standard triggered targetedacquisition, dedicated to higher throughput analyses, was executed through an application programming interface (API)controlling a Thermo ScientificTM Q ExactiveTM HF-X hybrid quadrupole-OrbitrapTM mass spectrometer coupled to aEvosep One LC system. PQ500 kit from Biognosys was spiked into the plasma samples to drive the targeted acquisition.

Results: The SureQuant method has been adapted from the internal standard triggered PRM (IS-PRM) method in orderto improve its usability and robustness, while retaining its analytical performance, and has been implemented in the nativeinstrument control software of next generation Orbitrap mass spectrometers. The application of the method to the analysisof plasma samples, leveraging spiked-in SIL peptides from PQ500 kit, allowed robust and precise quantification of around560 endogenous peptides used as surrogate of 400 plasma proteins using a 70-min LC gradient. Therefore, the methodcombined data quality of targeted analyses with proteome coverage of state-of-the-art profiling experiments. In addition,the availability of pre-set PQ500 SureQuant methods embedded into the instrument control software enables a ‘load andplay’ execution of such plasma proteomics experiments.

INTRODUCTIONThe success of plasma proteomics studies to deliver useful protein biomarkers has remained lower than initialexpectations, prompting a redefinition of the plasma biomarker development pipeline. One proposed measure consisted inthe implementation of a “rectangular” approach, relying exclusively on broad proteome profiling (through DDA or DIAanalyses) across large cohorts for both the discovery and validation/verification stages. We describe here an alternativeglobal quantification workflow, relying on a high density targeted acquisition method implemented on next generationOrbitrap mass spectrometers. This adaptation of the IS-PRM method leverages spiked-in stable isotopically labeled (SIL)peptides to drive the systematic screening of more than 500 plasma proteins per analysis across large sample sets.

MATERIALS AND METHODSSample PreparationA commercial human plasma sample pooled from several individuals purchased from BioreclamationIVT was processedusing Thermo ScientificTM EasyPepTM Mini MS Sample Prep Kit.A set of 804 SIL peptides from PQ500 kit (Biognosys PN#Ki-3019-96) was spiked at around 80 fmol (median value) into 1µg of undepleted plasma sample. One µL of the sample was injected for LC-MS/MS analyses.

LC-MS/MS AnalysisFor SureQuant analyses, chromatographic separations were performed on UltiMate 3000 RSLC system equipped with C18trap cartridge (5 µm, 0.3 x 5 mm) and analytical column (2 μm, 0.15 x 150 mm), and EASY-nLC 1200 system equippedwith a C18 analytical column (2 μm, 0.15 x 150 mm), using a 70-min gradient separation. The LC systems were coupled toOrbitrap Exploris 480 and Orbitrap Eclipse Tribrid mass spectrometers. In SureQuant analyses, Full MS scans wereacquired at a resolution of 120k with a maximum fill time of 50 ms, while ddMS2 of IS and endogenous peptides wereacquired sequentially at a resolution of 7.5k or 60k and maximum fill times of 10 ms and 116 ms, respectively.The analyses using the high throughput IS-PRM acquisition variant were performed on Evosep One LC system equippedwith Evosep C18 analytical column (3 μm, 0.1 x 80 mm), using a 21-min gradient separation. The LC system was coupledto a Q Exactive HF-X controlled by an API. In the analyses performed with the ‘Single Scan Msx IS-PRM’ method (Figure1, right panel), PRM scans in watch mode were acquired at a resolution of 7.5k and a maximum fill time of 10 ms. In thequant. mode, the pairs of IS and endogenous peptides were measured simultaneously by ‘multiplex’ acquisition of theirfragments in a single PRM scan generated at a resolution of 120k and non-isochronous maximum fill times of 20 or 124ms for the IS and endogenous peptides, respectively.

CONCLUSIONS A novel global plasma IS triggered targeted quantification workflow, based on ‘SureQuant’ method has been

successfully implemented in the native instrument control software of Orbitrap Exploris 480 and Orbitrap EclipseTribrid MS.

The global plasma quantification workflow combines the figures of merit of targeted acquisition method with thecoverage capabilities of profiling methods for plasma proteomics. In addition, it exhibits an efficient portability and ahigh usability, benefiting from embedded preset PQ500 related methods.

The developed IS-PRM acquisition variant was explored to carry out high throughput analyses. It exhibited slightlylower performance, but also the capability for a 5-fold increase in throughput/experiment scale. Further investigation isneeded to evaluate the pertinence of its native implementation under the form of a SureQuant method variant.

For Research Use Only. Not for use in diagnostic procedures.

TRADEMARKS/LICENSING© 2019 Thermo Fisher Scientific Inc. All rights reserved. All trademarks are the property of Thermo Fisher Scientific andits subsidiaries. This information is not intended to encourage use of these products in any manner that might infringe theintellectual property rights of others.

Global Plasma Proteome Quantification Using Internal Standard Triggered Targeted Analyses

Figure 5. Sampling rate in SureQuant PQ500 global plasma analysis

SureQuant PQ500 Plasma QuantificationThe PQ500 SureQuant acquisition workflow has been applied to the analyses of 1 µg of undepleted plasma samplesupplemented with PQ500 kit (median spiked-in amount of SIL peptides of 80 fmol) using a 70-min LC gradient onUltiMate 3000 RSLC system. The survey run revealed that 802 of 804 SIL peptides were compatible (intensity responseand hydrophobicity) with the LC-MS setup used (Figure 4), and were retained for subsequent SureQuant analyses. Thehigh overall intensity response of the SIL peptides (median MS1 intensity at chromatographic peak apex of 1e8) insuredhigh acquisition efficiency (>90%) in SureQuant analyses, benefiting from the median MS1 triggering intensity threshold at1e6. This translated into a good overall sampling rate over the LC separation, exceeding 8 data points collected over theelution profile of the vast majority of the peptides, which also remained acceptable in the most crowded time ranges, with6 to 8 data points collected per peak (Figure 5). The 802 retained internal standards were systematically detected in thetriplicated SureQuant analyses of plasma samples, and triggered high quality measurement of the correspondingendogenous peptides, translating into a broad coverage of the undepleted plasma proteome (Figure 6). Around 400proteins were detected in each replicate (based on around 560 surrogate endogenous peptides), while 90% of the totalsets were systematically quantified across triplicates.

In addition to delivering a broad plasma proteome coverage, rivalling that of state-of-the-art profiling methods, SureQuantPQ500 analyses still provide targeted quantification performance (Figure 7). The amounts of endogenous peptides wereestimated based on the known amounts of IS injected and single-point quantification strategy. This revealed aquantification range between 4 amol and 15 pmol, spanning more than 6 orders of magnitude, which is in theory sufficientto cover the vast majority of current FDA-approved plasma biomarkers. The fragment ion signals of the lowest abundantquantified endogenous peptide (PFLVFIR, 4 amol/µg of plasma) have been extracted (Figure 7, lower panel), illustratingthe acceptable data quality even obtained for this challenging analyte. The quantification precision obtained on the entiredataset was excellent, with 75% of the endogenous peptides quantified with a CV below 10% (median CV value of 6%).The experiments were reproduced at a second site by another operator, using a different LC set-up (EASY-nLC 1200system operated in a one-column setup instead of UltiMate 3000 RSLC system operated in a two-column setup) (Figure8). After initial survey analysis, enabling the adjustment of internal standards MS1 triggering intensity threshold,SureQuant PQ500 analyses of undepleted plasma sample were repeated. It turned out that the quantification of theendogenous peptides between sites / systems / operators was very consistent, as reflected by the quantitative valuestightly aligned along the theoretical ’perfect’ correlation line, and the absence of bias revealed by the equation of theregression. The acquisition scheme of SureQuant method, fully avoiding any dependency on time-scheduling, enablesefficient method portability and strong robustness.

The analyses of plasma sample supplemented with PQ500 kit were reproduced using the ‘Single Scan Msx IS-PRM’acquisition method on QE HF-X MS coupled to Evosep One LC system operated at a 60 samples/day analyticalthroughput (21-min LC gradient). The sampling rate under these conditions remained significantly higher than thepredefined (relaxed) limit of 4-5 data points collected per peptide chromatographic peak (Figure 9, upper panel). Incomparison with SureQuant analyses performed with longer LC gradient, a substantial number of IS peptides (81) werenot detected, corresponding to the most hydrophilic peptides, which are not compatible with the LC setup used (Figure 9,lower panel). The remaining 721 IS peptides were properly detected in the watch mode of the method and triggeredmultiplex PRM measurement in the quant. mode, enabling the detection of 421 endogenous peptides, surrogate of 301plasma proteins. In Figure 10, the data generated by SureQuant and Single Scan Msx IS-PRM analyses are displayed forone endogenous peptide quantified by both methods. The quantification results were very consistent across the methods.

Figure 3. Overall PQ500 SureQuant acquisition workflow.

Figure 6. Peptide detection/quantification in SureQuant PQ500 analyses of undepleted plasma

Figure 7. Quantification performance of SureQuant PQ500 analyses of undepleted plasma

Figure 8. Consistency of SureQuant PQ500 Quantification across sites/systems

Figure 9. High throughput plasma analyses using Single Scan Msx IS-PRM method.

Figure 10. Endogenous peptide quantification from SureQuant and Single Scan Msx IS-PRM

Figure 1. Sequential and multiplex acquisition modes

LC-MS Analysis

SIL peptides (internal standards)

#1#2

#n….

Sample SureQuant

ENDO Peptides Pseudo Spectral Matching

‘QUANT MODE’High Quality MS2 of Target

‘WATCH MODE’

Fast, Low Res MS2 of Trigger

1

4

5

2 3

MS1 Survey Scan

Mix

MS/MS IS

IS peptides (trigger)

1

2

3

4

5

Figure 2. SureQuant acquisition Scheme and method structure

00.5

11.5

22.5

33.5

4

0 10 20 30 40 50 60 70 80 90[min]

Cycle Time [s]

Sampling rate

6-8 data points /peak(20-s peak width)

≥8 data points /peak(20-s peak width)

Intensity response of precursor ions

PQ500 kit in plasma

Survey Run

SureQuant

LC-MS Analysis

PQ500 kit Plasma sample LC-MS Analysis

Survey Run Analysis

Export Targeted Mass report

Update Targeted Mass intensity threshold

Target Quantification

PQ500 Directed DDAInclusion list optimal IS precursor ions

Ready-to-use template

Ready-to-use template PQ500 SureQuant

PQ500 Skyline analysis template

PQ500 Skyline analysis template

Targeted Mass

Copy/Paste

589 total peptides

Rep 2Rep 1

Rep 3

526 peptides

421 total proteins

Rep 2Rep 1

Rep 3

368 proteins

0.001

0.01

0.1

1

10

100

1000

10000

100000

0.001 0.01 0.1 1 10 100 1000 10000 100000

Site 2-Endo. Pep. Conc. [fmol/µg plasma]

Site 1-Endo. Pep. Conc. [fmol/µg plasma]

Theoretical ‘perfect’ correlationActual correlation

R2 = 0.97y = 1.0009 x + 0.0881

0

1

2

3

4

0 5 10 15 20

0.E+00

2.E+07

4.E+07

6.E+07

4.45 4.65 4.850.E+00

2.E+07

4.E+07

6.E+07

23 23.2 23.4 23.60.E+00

2.E+07

4.E+07

4.45 4.65 4.85

FQASVATPR

FQASVATPRWATCH MODE QUANT MODE

0.E+00

1.E+05

2.E+05

3.E+05

4.45 4.65 4.850.E+00

1.E+05

2.E+05

3.E+05

23 23.2 23.4 23.6

Single-Scan Msx IS-PRM (QE HF-X) SureQuant (Exploris 480)

Number of spectra including ≥ 5 ref. fragments

Peptide# 1…………………………………………………………………………………………………….802

Detection endo. pep. / prot.

564 pep. / 397 prot.

560 pep. / 396 prot.

550 pep. / 390 prot.

Internal Standards peptides

Endogenous peptides

0 1-2 ≥ 3

802 peptides

802 peptides

802 peptides

Internal Standards peptides

Endogenous peptides

Rep. 1

Rep. 2

Rep. 3

Rep. 1

Rep. 2

Rep. 3

Peptide# 1………………………………………………………………………………………………….802

Sampling rate

≥ 4-5 data points /peak(12-s peak width)

Internal Standards peptides

Endogenous peptides

Number of spectra including ≥ 4 ref. fragments 0 1

Detection pep. / prot.

712 peptides

421 pep. / 301 prot.

Non-detectable IS Hydrophobicity scale

- +

Intensity ratio ENDO/SIL (Ss Msx IS-PRM): 0.0051

Area ratio ENDO/SIL (SureQuant): 0.0049

[min]

Cycle Time [s]

0

50

100

150

200

250

300

<5% 5-10% 10-20% >20%

CV

0.001

0.01

0.1

1

10

100

1000

10000

0 100 200 300 400 500 600

Endo. Pep. Conc. [fmol/µg plasma]

Abundance Rank

AlbuNumber of endo. Peptide.

56.4 56.6 56.8 5756.4 56.6 56.8 57

SIL (10fmol) ENDO (calc: 4amol) PFLVFIR

8e7 4e4

[min] [min]

I IQuantification range: 4 amol – 15 pmol(>6 orders of magnitude)

CV < 10% for more than 75% of the peptides (median CV: 5.7%)

Sequential acquisition Multiplex acquisition

Figure 4. Characterization of PQ500 kit through survey run analysis.

1.0E+06

1.0E+07

1.0E+08

1.0E+09Median intensity: 1e8

MS1 Intensity Apex IS

1.0

10.0

100.0

1,000.0

10,000.0 Median amount: 80 fmol

Injected Amount IS [fmol]

1………………………………………………………………………………………………………………………….802Peptide#

High Throughput IS Triggered Plasma AnalysesA variant of SureQuant acquisition, dedicated to higher throughput analyses, has been developed using API-based IS-PRM application, which is only compatible with QE MS platforms. This method has not been natively implemented in theinstrument control software of orbitrap instruments and does not benefit from the revised acquisition strategy ofSureQuant method, but relies on the acquisition scheme of the original IS-PRM method. More specifically, in this ‘SingleScan Msx IS-PRM’ acquisition variant, the detection of an IS peptide in watch mode triggers the concomitantmeasurement of the corresponding pair of IS and endogenous peptides in a single PRM scan in quant. mode close to theirelution profile apex, following their ‘multiplex’ isolation (Figure 1). This contrast with the regular sequential acquisition, inwhich data points are collected over the full elution profiles of the peptides. With this acquisition variant, quantification isperformed based on the fragment ion intensities extracted from single high quality PRM scan, and not on the area underthe curve of the corresponding extracted fragment traces. This quantification strategy alleviates sampling rate constraint.

[min]

I

PO65534-EN0519S

Page 4: Global Plasma Proteome Quantification Using Internal Standard … · 2019. 12. 20. · Global Plasma Proteome Quantification Using Internal Standard Triggered Targeted Analyses POSTER

Sebastien Gallien1,2, Jing Wang2, Aaron S. Gajadhar3, Bhavin Patel4, Markus Kellmann5, Tabiwang N. Arrey5, Alexander Harder5, Romain Huguet3, Graeme McAlister3, Derek Bailey3, Shannon Eliuk3, Yue Xuan5, Andreas Huhmer3, Emily L. Chen2

1Thermo Fisher Scientific, Paris, France; 2Thermo Fisher Scientific, Precision Medicine Science Center, Cambridge, MA; 3Thermo Fisher Scientific, San Jose, CA; 4Thermo Fisher Scientific, Rockford, IL; 5Thermo Fisher Scientific, Bremen, Germany

SureQuant MethodologyThe developed SureQuant acquisition method includes different scan events and filters, with parameters optimized for thehighest data quality and sensitivity of triggering, but also to retain sufficient triggering specificity (Figure 2). DuringSureQuant analyses, a high resolution (120k) full scan MS (1) is acquired, using lock mass re-calibration, to monitor thepredefined optimal precursor ions of the IS, based on the list of associated m/z values and intensity thresholds (defined at1% of the expected MS1 intensity at the chromatographic peak apex) included in the ‘Targeted Mass’ filter (2). In case ofthe detection of a candidate precursor ion satisfying the filtering criteria (MS peak with an m/z value within 3-10 ppmtolerance and intensity exceeding the threshold), it is subjected to fast MS/MS acquisition (3) performed at low resolutionand short fill times (typically 7.5k/10 ms), which is used as a second layer of confirmation of the detection of the IS.Several fragment ions, predefined as associated with the IS (6 optimal frag.) are included in the ‘Targeted Mass Trigger’filter (4), and the detection of a subset of them (at least 5 frag., 10-20 ppm tolerance) confirms the actual elution of the IS.This two-step process interrogating MS1 and MS2 data to track IS elution is the ‘watch’ mode of the SureQuant method.The detection of an IS in watch mode triggers the ‘quant’ mode, and therefore MS/MS acquisition of the correspondingendogenous pept. (5) with parameters favoring data quality, i.e., high resolution and long fill times (typically 60k/116 ms).The process is repeated over the entire LC separation with a systematic cycle-to-cycle assessment of the elution of the IS.The development of a SureQuant assay to be associated with a custom peptide panels requires the preliminarydetermination of the optimal precursor ion of each peptide target together with their intensity response and optimalassociated fragment ions. This information is typically extracted from an initial directed DDA analysis of the correspondingIS peptide mixture spiked into a representative matrix. The target list of multiple theoretical precursor ions of each ISpeptide included in such survey runs enables comprehensive data collection, and therefore the refinement of analytesdefinition for subsequent SureQuant analyses. The process is supported by Skyline software and facilitated by the genericdata acquisition and processing templates provided. The transfer of the generated information can be achieved throughcopy/paste from Skyline reports to ‘Targeted Mass’ and ‘Targeted Mass Trigger’ tables of SureQuant method. The surveyanalysis of a SureQuant assay needs to be run one time as the onset of a project without further adjustment over time,unless a major change in the experimental design is applied (e.g., LC set up, spiked amount of SIL peptides). In the lattercase, only the intensity responses of predefined IS precursor ions need to be re-adjusted though an additional survey runperformed under the modified experimental conditions. Therefore, the ability of embedding pre-set methods associatedwith predefined kits of IS peptides expedites SureQuant assay development at the user sites. Such validated methodtemplates have already been included in the instrument control software for the PQ500 related methods (Figure 3).

ABSTRACTPurpose: A novel global plasma proteome quantification workflow has been developed. It relies on a new high densitytargeted acquisition method, called ‘SureQuant’ method, implemented on next generation Orbitrap mass spectrometers:Thermo ScientificTM Orbitrap ExplorisTM 480 mass spectrometer and Thermo ScientificTM Orbitrap EclipseTM TribridTM

mass spectrometer. In addition, the workflow leverages a large set of stable isotopically labeled (SIL) peptides, used asinternal standards, to drive the systematic screening of more than 500 plasma proteins (PQ500 kit from Biognosys).

Methods: The SureQuant PQ500 analyses of undepleted plasma samples were performed using Orbitrap Exploris 480and Orbitrap Eclipse Tribrid mass spectrometers, coupled to Thermo ScientificTM UltiMateTM 3000 RSCLC system orThermo ScientificTM EASY-nLCTM 1200 system. An acquisition variant of the internal standard triggered targetedacquisition, dedicated to higher throughput analyses, was executed through an application programming interface (API)controlling a Thermo ScientificTM Q ExactiveTM HF-X hybrid quadrupole-OrbitrapTM mass spectrometer coupled to aEvosep One LC system. PQ500 kit from Biognosys was spiked into the plasma samples to drive the targeted acquisition.

Results: The SureQuant method has been adapted from the internal standard triggered PRM (IS-PRM) method in orderto improve its usability and robustness, while retaining its analytical performance, and has been implemented in the nativeinstrument control software of next generation Orbitrap mass spectrometers. The application of the method to the analysisof plasma samples, leveraging spiked-in SIL peptides from PQ500 kit, allowed robust and precise quantification of around560 endogenous peptides used as surrogate of 400 plasma proteins using a 70-min LC gradient. Therefore, the methodcombined data quality of targeted analyses with proteome coverage of state-of-the-art profiling experiments. In addition,the availability of pre-set PQ500 SureQuant methods embedded into the instrument control software enables a ‘load andplay’ execution of such plasma proteomics experiments.

INTRODUCTIONThe success of plasma proteomics studies to deliver useful protein biomarkers has remained lower than initialexpectations, prompting a redefinition of the plasma biomarker development pipeline. One proposed measure consisted inthe implementation of a “rectangular” approach, relying exclusively on broad proteome profiling (through DDA or DIAanalyses) across large cohorts for both the discovery and validation/verification stages. We describe here an alternativeglobal quantification workflow, relying on a high density targeted acquisition method implemented on next generationOrbitrap mass spectrometers. This adaptation of the IS-PRM method leverages spiked-in stable isotopically labeled (SIL)peptides to drive the systematic screening of more than 500 plasma proteins per analysis across large sample sets.

MATERIALS AND METHODSSample PreparationA commercial human plasma sample pooled from several individuals purchased from BioreclamationIVT was processedusing Thermo ScientificTM EasyPepTM Mini MS Sample Prep Kit.A set of 804 SIL peptides from PQ500 kit (Biognosys PN#Ki-3019-96) was spiked at around 80 fmol (median value) into 1µg of undepleted plasma sample. One µL of the sample was injected for LC-MS/MS analyses.

LC-MS/MS AnalysisFor SureQuant analyses, chromatographic separations were performed on UltiMate 3000 RSLC system equipped with C18trap cartridge (5 µm, 0.3 x 5 mm) and analytical column (2 μm, 0.15 x 150 mm), and EASY-nLC 1200 system equippedwith a C18 analytical column (2 μm, 0.15 x 150 mm), using a 70-min gradient separation. The LC systems were coupled toOrbitrap Exploris 480 and Orbitrap Eclipse Tribrid mass spectrometers. In SureQuant analyses, Full MS scans wereacquired at a resolution of 120k with a maximum fill time of 50 ms, while ddMS2 of IS and endogenous peptides wereacquired sequentially at a resolution of 7.5k or 60k and maximum fill times of 10 ms and 116 ms, respectively.The analyses using the high throughput IS-PRM acquisition variant were performed on Evosep One LC system equippedwith Evosep C18 analytical column (3 μm, 0.1 x 80 mm), using a 21-min gradient separation. The LC system was coupledto a Q Exactive HF-X controlled by an API. In the analyses performed with the ‘Single Scan Msx IS-PRM’ method (Figure1, right panel), PRM scans in watch mode were acquired at a resolution of 7.5k and a maximum fill time of 10 ms. In thequant. mode, the pairs of IS and endogenous peptides were measured simultaneously by ‘multiplex’ acquisition of theirfragments in a single PRM scan generated at a resolution of 120k and non-isochronous maximum fill times of 20 or 124ms for the IS and endogenous peptides, respectively.

CONCLUSIONS A novel global plasma IS triggered targeted quantification workflow, based on ‘SureQuant’ method has been

successfully implemented in the native instrument control software of Orbitrap Exploris 480 and Orbitrap EclipseTribrid MS.

The global plasma quantification workflow combines the figures of merit of targeted acquisition method with thecoverage capabilities of profiling methods for plasma proteomics. In addition, it exhibits an efficient portability and ahigh usability, benefiting from embedded preset PQ500 related methods.

The developed IS-PRM acquisition variant was explored to carry out high throughput analyses. It exhibited slightlylower performance, but also the capability for a 5-fold increase in throughput/experiment scale. Further investigation isneeded to evaluate the pertinence of its native implementation under the form of a SureQuant method variant.

For Research Use Only. Not for use in diagnostic procedures.

TRADEMARKS/LICENSING© 2019 Thermo Fisher Scientific Inc. All rights reserved. All trademarks are the property of Thermo Fisher Scientific andits subsidiaries. This information is not intended to encourage use of these products in any manner that might infringe theintellectual property rights of others.

Global Plasma Proteome Quantification Using Internal Standard Triggered Targeted Analyses

Figure 5. Sampling rate in SureQuant PQ500 global plasma analysis

SureQuant PQ500 Plasma QuantificationThe PQ500 SureQuant acquisition workflow has been applied to the analyses of 1 µg of undepleted plasma samplesupplemented with PQ500 kit (median spiked-in amount of SIL peptides of 80 fmol) using a 70-min LC gradient onUltiMate 3000 RSLC system. The survey run revealed that 802 of 804 SIL peptides were compatible (intensity responseand hydrophobicity) with the LC-MS setup used (Figure 4), and were retained for subsequent SureQuant analyses. Thehigh overall intensity response of the SIL peptides (median MS1 intensity at chromatographic peak apex of 1e8) insuredhigh acquisition efficiency (>90%) in SureQuant analyses, benefiting from the median MS1 triggering intensity threshold at1e6. This translated into a good overall sampling rate over the LC separation, exceeding 8 data points collected over theelution profile of the vast majority of the peptides, which also remained acceptable in the most crowded time ranges, with6 to 8 data points collected per peak (Figure 5). The 802 retained internal standards were systematically detected in thetriplicated SureQuant analyses of plasma samples, and triggered high quality measurement of the correspondingendogenous peptides, translating into a broad coverage of the undepleted plasma proteome (Figure 6). Around 400proteins were detected in each replicate (based on around 560 surrogate endogenous peptides), while 90% of the totalsets were systematically quantified across triplicates.

In addition to delivering a broad plasma proteome coverage, rivalling that of state-of-the-art profiling methods, SureQuantPQ500 analyses still provide targeted quantification performance (Figure 7). The amounts of endogenous peptides wereestimated based on the known amounts of IS injected and single-point quantification strategy. This revealed aquantification range between 4 amol and 15 pmol, spanning more than 6 orders of magnitude, which is in theory sufficientto cover the vast majority of current FDA-approved plasma biomarkers. The fragment ion signals of the lowest abundantquantified endogenous peptide (PFLVFIR, 4 amol/µg of plasma) have been extracted (Figure 7, lower panel), illustratingthe acceptable data quality even obtained for this challenging analyte. The quantification precision obtained on the entiredataset was excellent, with 75% of the endogenous peptides quantified with a CV below 10% (median CV value of 6%).The experiments were reproduced at a second site by another operator, using a different LC set-up (EASY-nLC 1200system operated in a one-column setup instead of UltiMate 3000 RSLC system operated in a two-column setup) (Figure8). After initial survey analysis, enabling the adjustment of internal standards MS1 triggering intensity threshold,SureQuant PQ500 analyses of undepleted plasma sample were repeated. It turned out that the quantification of theendogenous peptides between sites / systems / operators was very consistent, as reflected by the quantitative valuestightly aligned along the theoretical ’perfect’ correlation line, and the absence of bias revealed by the equation of theregression. The acquisition scheme of SureQuant method, fully avoiding any dependency on time-scheduling, enablesefficient method portability and strong robustness.

The analyses of plasma sample supplemented with PQ500 kit were reproduced using the ‘Single Scan Msx IS-PRM’acquisition method on QE HF-X MS coupled to Evosep One LC system operated at a 60 samples/day analyticalthroughput (21-min LC gradient). The sampling rate under these conditions remained significantly higher than thepredefined (relaxed) limit of 4-5 data points collected per peptide chromatographic peak (Figure 9, upper panel). Incomparison with SureQuant analyses performed with longer LC gradient, a substantial number of IS peptides (81) werenot detected, corresponding to the most hydrophilic peptides, which are not compatible with the LC setup used (Figure 9,lower panel). The remaining 721 IS peptides were properly detected in the watch mode of the method and triggeredmultiplex PRM measurement in the quant. mode, enabling the detection of 421 endogenous peptides, surrogate of 301plasma proteins. In Figure 10, the data generated by SureQuant and Single Scan Msx IS-PRM analyses are displayed forone endogenous peptide quantified by both methods. The quantification results were very consistent across the methods.

Figure 3. Overall PQ500 SureQuant acquisition workflow.

Figure 6. Peptide detection/quantification in SureQuant PQ500 analyses of undepleted plasma

Figure 7. Quantification performance of SureQuant PQ500 analyses of undepleted plasma

Figure 8. Consistency of SureQuant PQ500 Quantification across sites/systems

Figure 9. High throughput plasma analyses using Single Scan Msx IS-PRM method.

Figure 10. Endogenous peptide quantification from SureQuant and Single Scan Msx IS-PRM

Figure 1. Sequential and multiplex acquisition modes

LC-MS Analysis

SIL peptides (internal standards)

#1#2

#n….

Sample SureQuant

ENDO Peptides Pseudo Spectral Matching

‘QUANT MODE’High Quality MS2 of Target

‘WATCH MODE’

Fast, Low Res MS2 of Trigger

1

4

5

2 3

MS1 Survey Scan

Mix

MS/MS IS

IS peptides (trigger)

1

2

3

4

5

Figure 2. SureQuant acquisition Scheme and method structure

00.5

11.5

22.5

33.5

4

0 10 20 30 40 50 60 70 80 90[min]

Cycle Time [s]

Sampling rate

6-8 data points /peak(20-s peak width)

≥8 data points /peak(20-s peak width)

Intensity response of precursor ions

PQ500 kit in plasma

Survey Run

SureQuant

LC-MS Analysis

PQ500 kit Plasma sample LC-MS Analysis

Survey Run Analysis

Export Targeted Mass report

Update Targeted Mass intensity threshold

Target Quantification

PQ500 Directed DDAInclusion list optimal IS precursor ions

Ready-to-use template

Ready-to-use template PQ500 SureQuant

PQ500 Skyline analysis template

PQ500 Skyline analysis template

Targeted Mass

Copy/Paste

589 total peptides

Rep 2Rep 1

Rep 3

526 peptides

421 total proteins

Rep 2Rep 1

Rep 3

368 proteins

0.001

0.01

0.1

1

10

100

1000

10000

100000

0.001 0.01 0.1 1 10 100 1000 10000 100000

Site 2-Endo. Pep. Conc. [fmol/µg plasma]

Site 1-Endo. Pep. Conc. [fmol/µg plasma]

Theoretical ‘perfect’ correlationActual correlation

R2 = 0.97y = 1.0009 x + 0.0881

0

1

2

3

4

0 5 10 15 20

0.E+00

2.E+07

4.E+07

6.E+07

4.45 4.65 4.850.E+00

2.E+07

4.E+07

6.E+07

23 23.2 23.4 23.60.E+00

2.E+07

4.E+07

4.45 4.65 4.85

FQASVATPR

FQASVATPRWATCH MODE QUANT MODE

0.E+00

1.E+05

2.E+05

3.E+05

4.45 4.65 4.850.E+00

1.E+05

2.E+05

3.E+05

23 23.2 23.4 23.6

Single-Scan Msx IS-PRM (QE HF-X) SureQuant (Exploris 480)

Number of spectra including ≥ 5 ref. fragments

Peptide# 1…………………………………………………………………………………………………….802

Detection endo. pep. / prot.

564 pep. / 397 prot.

560 pep. / 396 prot.

550 pep. / 390 prot.

Internal Standards peptides

Endogenous peptides

0 1-2 ≥ 3

802 peptides

802 peptides

802 peptides

Internal Standards peptides

Endogenous peptides

Rep. 1

Rep. 2

Rep. 3

Rep. 1

Rep. 2

Rep. 3

Peptide# 1………………………………………………………………………………………………….802

Sampling rate

≥ 4-5 data points /peak(12-s peak width)

Internal Standards peptides

Endogenous peptides

Number of spectra including ≥ 4 ref. fragments 0 1

Detection pep. / prot.

712 peptides

421 pep. / 301 prot.

Non-detectable IS Hydrophobicity scale

- +

Intensity ratio ENDO/SIL (Ss Msx IS-PRM): 0.0051

Area ratio ENDO/SIL (SureQuant): 0.0049

[min]

Cycle Time [s]

0

50

100

150

200

250

300

<5% 5-10% 10-20% >20%

CV

0.001

0.01

0.1

1

10

100

1000

10000

0 100 200 300 400 500 600

Endo. Pep. Conc. [fmol/µg plasma]

Abundance Rank

AlbuNumber of endo. Peptide.

56.4 56.6 56.8 5756.4 56.6 56.8 57

SIL (10fmol) ENDO (calc: 4amol) PFLVFIR

8e7 4e4

[min] [min]

I IQuantification range: 4 amol – 15 pmol(>6 orders of magnitude)

CV < 10% for more than 75% of the peptides (median CV: 5.7%)

Sequential acquisition Multiplex acquisition

Figure 4. Characterization of PQ500 kit through survey run analysis.

1.0E+06

1.0E+07

1.0E+08

1.0E+09Median intensity: 1e8

MS1 Intensity Apex IS

1.0

10.0

100.0

1,000.0

10,000.0 Median amount: 80 fmol

Injected Amount IS [fmol]

1………………………………………………………………………………………………………………………….802Peptide#

High Throughput IS Triggered Plasma AnalysesA variant of SureQuant acquisition, dedicated to higher throughput analyses, has been developed using API-based IS-PRM application, which is only compatible with QE MS platforms. This method has not been natively implemented in theinstrument control software of orbitrap instruments and does not benefit from the revised acquisition strategy ofSureQuant method, but relies on the acquisition scheme of the original IS-PRM method. More specifically, in this ‘SingleScan Msx IS-PRM’ acquisition variant, the detection of an IS peptide in watch mode triggers the concomitantmeasurement of the corresponding pair of IS and endogenous peptides in a single PRM scan in quant. mode close to theirelution profile apex, following their ‘multiplex’ isolation (Figure 1). This contrast with the regular sequential acquisition, inwhich data points are collected over the full elution profiles of the peptides. With this acquisition variant, quantification isperformed based on the fragment ion intensities extracted from single high quality PRM scan, and not on the area underthe curve of the corresponding extracted fragment traces. This quantification strategy alleviates sampling rate constraint.

[min]

I

PO65534-EN0519S

Page 5: Global Plasma Proteome Quantification Using Internal Standard … · 2019. 12. 20. · Global Plasma Proteome Quantification Using Internal Standard Triggered Targeted Analyses POSTER

Sebastien Gallien1,2, Jing Wang2, Aaron S. Gajadhar3, Bhavin Patel4, Markus Kellmann5, Tabiwang N. Arrey5, Alexander Harder5, Romain Huguet3, Graeme McAlister3, Derek Bailey3, Shannon Eliuk3, Yue Xuan5, Andreas Huhmer3, Emily L. Chen2

1Thermo Fisher Scientific, Paris, France; 2Thermo Fisher Scientific, Precision Medicine Science Center, Cambridge, MA; 3Thermo Fisher Scientific, San Jose, CA; 4Thermo Fisher Scientific, Rockford, IL; 5Thermo Fisher Scientific, Bremen, Germany

SureQuant MethodologyThe developed SureQuant acquisition method includes different scan events and filters, with parameters optimized for thehighest data quality and sensitivity of triggering, but also to retain sufficient triggering specificity (Figure 2). DuringSureQuant analyses, a high resolution (120k) full scan MS (1) is acquired, using lock mass re-calibration, to monitor thepredefined optimal precursor ions of the IS, based on the list of associated m/z values and intensity thresholds (defined at1% of the expected MS1 intensity at the chromatographic peak apex) included in the ‘Targeted Mass’ filter (2). In case ofthe detection of a candidate precursor ion satisfying the filtering criteria (MS peak with an m/z value within 3-10 ppmtolerance and intensity exceeding the threshold), it is subjected to fast MS/MS acquisition (3) performed at low resolutionand short fill times (typically 7.5k/10 ms), which is used as a second layer of confirmation of the detection of the IS.Several fragment ions, predefined as associated with the IS (6 optimal frag.) are included in the ‘Targeted Mass Trigger’filter (4), and the detection of a subset of them (at least 5 frag., 10-20 ppm tolerance) confirms the actual elution of the IS.This two-step process interrogating MS1 and MS2 data to track IS elution is the ‘watch’ mode of the SureQuant method.The detection of an IS in watch mode triggers the ‘quant’ mode, and therefore MS/MS acquisition of the correspondingendogenous pept. (5) with parameters favoring data quality, i.e., high resolution and long fill times (typically 60k/116 ms).The process is repeated over the entire LC separation with a systematic cycle-to-cycle assessment of the elution of the IS.The development of a SureQuant assay to be associated with a custom peptide panels requires the preliminarydetermination of the optimal precursor ion of each peptide target together with their intensity response and optimalassociated fragment ions. This information is typically extracted from an initial directed DDA analysis of the correspondingIS peptide mixture spiked into a representative matrix. The target list of multiple theoretical precursor ions of each ISpeptide included in such survey runs enables comprehensive data collection, and therefore the refinement of analytesdefinition for subsequent SureQuant analyses. The process is supported by Skyline software and facilitated by the genericdata acquisition and processing templates provided. The transfer of the generated information can be achieved throughcopy/paste from Skyline reports to ‘Targeted Mass’ and ‘Targeted Mass Trigger’ tables of SureQuant method. The surveyanalysis of a SureQuant assay needs to be run one time as the onset of a project without further adjustment over time,unless a major change in the experimental design is applied (e.g., LC set up, spiked amount of SIL peptides). In the lattercase, only the intensity responses of predefined IS precursor ions need to be re-adjusted though an additional survey runperformed under the modified experimental conditions. Therefore, the ability of embedding pre-set methods associatedwith predefined kits of IS peptides expedites SureQuant assay development at the user sites. Such validated methodtemplates have already been included in the instrument control software for the PQ500 related methods (Figure 3).

ABSTRACTPurpose: A novel global plasma proteome quantification workflow has been developed. It relies on a new high densitytargeted acquisition method, called ‘SureQuant’ method, implemented on next generation Orbitrap mass spectrometers:Thermo ScientificTM Orbitrap ExplorisTM 480 mass spectrometer and Thermo ScientificTM Orbitrap EclipseTM TribridTM

mass spectrometer. In addition, the workflow leverages a large set of stable isotopically labeled (SIL) peptides, used asinternal standards, to drive the systematic screening of more than 500 plasma proteins (PQ500 kit from Biognosys).

Methods: The SureQuant PQ500 analyses of undepleted plasma samples were performed using Orbitrap Exploris 480and Orbitrap Eclipse Tribrid mass spectrometers, coupled to Thermo ScientificTM UltiMateTM 3000 RSCLC system orThermo ScientificTM EASY-nLCTM 1200 system. An acquisition variant of the internal standard triggered targetedacquisition, dedicated to higher throughput analyses, was executed through an application programming interface (API)controlling a Thermo ScientificTM Q ExactiveTM HF-X hybrid quadrupole-OrbitrapTM mass spectrometer coupled to aEvosep One LC system. PQ500 kit from Biognosys was spiked into the plasma samples to drive the targeted acquisition.

Results: The SureQuant method has been adapted from the internal standard triggered PRM (IS-PRM) method in orderto improve its usability and robustness, while retaining its analytical performance, and has been implemented in the nativeinstrument control software of next generation Orbitrap mass spectrometers. The application of the method to the analysisof plasma samples, leveraging spiked-in SIL peptides from PQ500 kit, allowed robust and precise quantification of around560 endogenous peptides used as surrogate of 400 plasma proteins using a 70-min LC gradient. Therefore, the methodcombined data quality of targeted analyses with proteome coverage of state-of-the-art profiling experiments. In addition,the availability of pre-set PQ500 SureQuant methods embedded into the instrument control software enables a ‘load andplay’ execution of such plasma proteomics experiments.

INTRODUCTIONThe success of plasma proteomics studies to deliver useful protein biomarkers has remained lower than initialexpectations, prompting a redefinition of the plasma biomarker development pipeline. One proposed measure consisted inthe implementation of a “rectangular” approach, relying exclusively on broad proteome profiling (through DDA or DIAanalyses) across large cohorts for both the discovery and validation/verification stages. We describe here an alternativeglobal quantification workflow, relying on a high density targeted acquisition method implemented on next generationOrbitrap mass spectrometers. This adaptation of the IS-PRM method leverages spiked-in stable isotopically labeled (SIL)peptides to drive the systematic screening of more than 500 plasma proteins per analysis across large sample sets.

MATERIALS AND METHODSSample PreparationA commercial human plasma sample pooled from several individuals purchased from BioreclamationIVT was processedusing Thermo ScientificTM EasyPepTM Mini MS Sample Prep Kit.A set of 804 SIL peptides from PQ500 kit (Biognosys PN#Ki-3019-96) was spiked at around 80 fmol (median value) into 1µg of undepleted plasma sample. One µL of the sample was injected for LC-MS/MS analyses.

LC-MS/MS AnalysisFor SureQuant analyses, chromatographic separations were performed on UltiMate 3000 RSLC system equipped with C18trap cartridge (5 µm, 0.3 x 5 mm) and analytical column (2 μm, 0.15 x 150 mm), and EASY-nLC 1200 system equippedwith a C18 analytical column (2 μm, 0.15 x 150 mm), using a 70-min gradient separation. The LC systems were coupled toOrbitrap Exploris 480 and Orbitrap Eclipse Tribrid mass spectrometers. In SureQuant analyses, Full MS scans wereacquired at a resolution of 120k with a maximum fill time of 50 ms, while ddMS2 of IS and endogenous peptides wereacquired sequentially at a resolution of 7.5k or 60k and maximum fill times of 10 ms and 116 ms, respectively.The analyses using the high throughput IS-PRM acquisition variant were performed on Evosep One LC system equippedwith Evosep C18 analytical column (3 μm, 0.1 x 80 mm), using a 21-min gradient separation. The LC system was coupledto a Q Exactive HF-X controlled by an API. In the analyses performed with the ‘Single Scan Msx IS-PRM’ method (Figure1, right panel), PRM scans in watch mode were acquired at a resolution of 7.5k and a maximum fill time of 10 ms. In thequant. mode, the pairs of IS and endogenous peptides were measured simultaneously by ‘multiplex’ acquisition of theirfragments in a single PRM scan generated at a resolution of 120k and non-isochronous maximum fill times of 20 or 124ms for the IS and endogenous peptides, respectively.

CONCLUSIONS A novel global plasma IS triggered targeted quantification workflow, based on ‘SureQuant’ method has been

successfully implemented in the native instrument control software of Orbitrap Exploris 480 and Orbitrap EclipseTribrid MS.

The global plasma quantification workflow combines the figures of merit of targeted acquisition method with thecoverage capabilities of profiling methods for plasma proteomics. In addition, it exhibits an efficient portability and ahigh usability, benefiting from embedded preset PQ500 related methods.

The developed IS-PRM acquisition variant was explored to carry out high throughput analyses. It exhibited slightlylower performance, but also the capability for a 5-fold increase in throughput/experiment scale. Further investigation isneeded to evaluate the pertinence of its native implementation under the form of a SureQuant method variant.

For Research Use Only. Not for use in diagnostic procedures.

TRADEMARKS/LICENSING© 2019 Thermo Fisher Scientific Inc. All rights reserved. All trademarks are the property of Thermo Fisher Scientific andits subsidiaries. This information is not intended to encourage use of these products in any manner that might infringe theintellectual property rights of others.

Global Plasma Proteome Quantification Using Internal Standard Triggered Targeted Analyses

Figure 5. Sampling rate in SureQuant PQ500 global plasma analysis

SureQuant PQ500 Plasma QuantificationThe PQ500 SureQuant acquisition workflow has been applied to the analyses of 1 µg of undepleted plasma samplesupplemented with PQ500 kit (median spiked-in amount of SIL peptides of 80 fmol) using a 70-min LC gradient onUltiMate 3000 RSLC system. The survey run revealed that 802 of 804 SIL peptides were compatible (intensity responseand hydrophobicity) with the LC-MS setup used (Figure 4), and were retained for subsequent SureQuant analyses. Thehigh overall intensity response of the SIL peptides (median MS1 intensity at chromatographic peak apex of 1e8) insuredhigh acquisition efficiency (>90%) in SureQuant analyses, benefiting from the median MS1 triggering intensity threshold at1e6. This translated into a good overall sampling rate over the LC separation, exceeding 8 data points collected over theelution profile of the vast majority of the peptides, which also remained acceptable in the most crowded time ranges, with6 to 8 data points collected per peak (Figure 5). The 802 retained internal standards were systematically detected in thetriplicated SureQuant analyses of plasma samples, and triggered high quality measurement of the correspondingendogenous peptides, translating into a broad coverage of the undepleted plasma proteome (Figure 6). Around 400proteins were detected in each replicate (based on around 560 surrogate endogenous peptides), while 90% of the totalsets were systematically quantified across triplicates.

In addition to delivering a broad plasma proteome coverage, rivalling that of state-of-the-art profiling methods, SureQuantPQ500 analyses still provide targeted quantification performance (Figure 7). The amounts of endogenous peptides wereestimated based on the known amounts of IS injected and single-point quantification strategy. This revealed aquantification range between 4 amol and 15 pmol, spanning more than 6 orders of magnitude, which is in theory sufficientto cover the vast majority of current FDA-approved plasma biomarkers. The fragment ion signals of the lowest abundantquantified endogenous peptide (PFLVFIR, 4 amol/µg of plasma) have been extracted (Figure 7, lower panel), illustratingthe acceptable data quality even obtained for this challenging analyte. The quantification precision obtained on the entiredataset was excellent, with 75% of the endogenous peptides quantified with a CV below 10% (median CV value of 6%).The experiments were reproduced at a second site by another operator, using a different LC set-up (EASY-nLC 1200system operated in a one-column setup instead of UltiMate 3000 RSLC system operated in a two-column setup) (Figure8). After initial survey analysis, enabling the adjustment of internal standards MS1 triggering intensity threshold,SureQuant PQ500 analyses of undepleted plasma sample were repeated. It turned out that the quantification of theendogenous peptides between sites / systems / operators was very consistent, as reflected by the quantitative valuestightly aligned along the theoretical ’perfect’ correlation line, and the absence of bias revealed by the equation of theregression. The acquisition scheme of SureQuant method, fully avoiding any dependency on time-scheduling, enablesefficient method portability and strong robustness.

The analyses of plasma sample supplemented with PQ500 kit were reproduced using the ‘Single Scan Msx IS-PRM’acquisition method on QE HF-X MS coupled to Evosep One LC system operated at a 60 samples/day analyticalthroughput (21-min LC gradient). The sampling rate under these conditions remained significantly higher than thepredefined (relaxed) limit of 4-5 data points collected per peptide chromatographic peak (Figure 9, upper panel). Incomparison with SureQuant analyses performed with longer LC gradient, a substantial number of IS peptides (81) werenot detected, corresponding to the most hydrophilic peptides, which are not compatible with the LC setup used (Figure 9,lower panel). The remaining 721 IS peptides were properly detected in the watch mode of the method and triggeredmultiplex PRM measurement in the quant. mode, enabling the detection of 421 endogenous peptides, surrogate of 301plasma proteins. In Figure 10, the data generated by SureQuant and Single Scan Msx IS-PRM analyses are displayed forone endogenous peptide quantified by both methods. The quantification results were very consistent across the methods.

Figure 3. Overall PQ500 SureQuant acquisition workflow.

Figure 6. Peptide detection/quantification in SureQuant PQ500 analyses of undepleted plasma

Figure 7. Quantification performance of SureQuant PQ500 analyses of undepleted plasma

Figure 8. Consistency of SureQuant PQ500 Quantification across sites/systems

Figure 9. High throughput plasma analyses using Single Scan Msx IS-PRM method.

Figure 10. Endogenous peptide quantification from SureQuant and Single Scan Msx IS-PRM

Figure 1. Sequential and multiplex acquisition modes

LC-MS Analysis

SIL peptides (internal standards)

#1#2

#n….

Sample SureQuant

ENDO Peptides Pseudo Spectral Matching

‘QUANT MODE’High Quality MS2 of Target

‘WATCH MODE’

Fast, Low Res MS2 of Trigger

1

4

5

2 3

MS1 Survey Scan

Mix

MS/MS IS

IS peptides (trigger)

1

2

3

4

5

Figure 2. SureQuant acquisition Scheme and method structure

00.5

11.5

22.5

33.5

4

0 10 20 30 40 50 60 70 80 90[min]

Cycle Time [s]

Sampling rate

6-8 data points /peak(20-s peak width)

≥8 data points /peak(20-s peak width)

Intensity response of precursor ions

PQ500 kit in plasma

Survey Run

SureQuant

LC-MS Analysis

PQ500 kit Plasma sample LC-MS Analysis

Survey Run Analysis

Export Targeted Mass report

Update Targeted Mass intensity threshold

Target Quantification

PQ500 Directed DDAInclusion list optimal IS precursor ions

Ready-to-use template

Ready-to-use template PQ500 SureQuant

PQ500 Skyline analysis template

PQ500 Skyline analysis template

Targeted Mass

Copy/Paste

589 total peptides

Rep 2Rep 1

Rep 3

526 peptides

421 total proteins

Rep 2Rep 1

Rep 3

368 proteins

0.001

0.01

0.1

1

10

100

1000

10000

100000

0.001 0.01 0.1 1 10 100 1000 10000 100000

Site 2-Endo. Pep. Conc. [fmol/µg plasma]

Site 1-Endo. Pep. Conc. [fmol/µg plasma]

Theoretical ‘perfect’ correlationActual correlation

R2 = 0.97y = 1.0009 x + 0.0881

0

1

2

3

4

0 5 10 15 20

0.E+00

2.E+07

4.E+07

6.E+07

4.45 4.65 4.850.E+00

2.E+07

4.E+07

6.E+07

23 23.2 23.4 23.60.E+00

2.E+07

4.E+07

4.45 4.65 4.85

FQASVATPR

FQASVATPRWATCH MODE QUANT MODE

0.E+00

1.E+05

2.E+05

3.E+05

4.45 4.65 4.850.E+00

1.E+05

2.E+05

3.E+05

23 23.2 23.4 23.6

Single-Scan Msx IS-PRM (QE HF-X) SureQuant (Exploris 480)

Number of spectra including ≥ 5 ref. fragments

Peptide# 1…………………………………………………………………………………………………….802

Detection endo. pep. / prot.

564 pep. / 397 prot.

560 pep. / 396 prot.

550 pep. / 390 prot.

Internal Standards peptides

Endogenous peptides

0 1-2 ≥ 3

802 peptides

802 peptides

802 peptides

Internal Standards peptides

Endogenous peptides

Rep. 1

Rep. 2

Rep. 3

Rep. 1

Rep. 2

Rep. 3

Peptide# 1………………………………………………………………………………………………….802

Sampling rate

≥ 4-5 data points /peak(12-s peak width)

Internal Standards peptides

Endogenous peptides

Number of spectra including ≥ 4 ref. fragments 0 1

Detection pep. / prot.

712 peptides

421 pep. / 301 prot.

Non-detectable IS Hydrophobicity scale

- +

Intensity ratio ENDO/SIL (Ss Msx IS-PRM): 0.0051

Area ratio ENDO/SIL (SureQuant): 0.0049

[min]

Cycle Time [s]

0

50

100

150

200

250

300

<5% 5-10% 10-20% >20%

CV

0.001

0.01

0.1

1

10

100

1000

10000

0 100 200 300 400 500 600

Endo. Pep. Conc. [fmol/µg plasma]

Abundance Rank

AlbuNumber of endo. Peptide.

56.4 56.6 56.8 5756.4 56.6 56.8 57

SIL (10fmol) ENDO (calc: 4amol) PFLVFIR

8e7 4e4

[min] [min]

I IQuantification range: 4 amol – 15 pmol(>6 orders of magnitude)

CV < 10% for more than 75% of the peptides (median CV: 5.7%)

Sequential acquisition Multiplex acquisition

Figure 4. Characterization of PQ500 kit through survey run analysis.

1.0E+06

1.0E+07

1.0E+08

1.0E+09Median intensity: 1e8

MS1 Intensity Apex IS

1.0

10.0

100.0

1,000.0

10,000.0 Median amount: 80 fmol

Injected Amount IS [fmol]

1………………………………………………………………………………………………………………………….802Peptide#

High Throughput IS Triggered Plasma AnalysesA variant of SureQuant acquisition, dedicated to higher throughput analyses, has been developed using API-based IS-PRM application, which is only compatible with QE MS platforms. This method has not been natively implemented in theinstrument control software of orbitrap instruments and does not benefit from the revised acquisition strategy ofSureQuant method, but relies on the acquisition scheme of the original IS-PRM method. More specifically, in this ‘SingleScan Msx IS-PRM’ acquisition variant, the detection of an IS peptide in watch mode triggers the concomitantmeasurement of the corresponding pair of IS and endogenous peptides in a single PRM scan in quant. mode close to theirelution profile apex, following their ‘multiplex’ isolation (Figure 1). This contrast with the regular sequential acquisition, inwhich data points are collected over the full elution profiles of the peptides. With this acquisition variant, quantification isperformed based on the fragment ion intensities extracted from single high quality PRM scan, and not on the area underthe curve of the corresponding extracted fragment traces. This quantification strategy alleviates sampling rate constraint.

[min]

I

PO65534-EN0519S

Page 6: Global Plasma Proteome Quantification Using Internal Standard … · 2019. 12. 20. · Global Plasma Proteome Quantification Using Internal Standard Triggered Targeted Analyses POSTER

Find out more at thermofisher.com/Orbitrap

© 2019 Thermo Fisher Scientific Inc. All rights reserved. All trademarks are the property of Thermo Fisher Scientific and its subsidiaries. This information is presented as an example of the capabilities of Thermo Fisher Scientific products. It is not intended to encourage use of these products in any manners that might infringe the intellectual property rights of others. Specifications, terms and pricing are subject to change. Not all products are available in all countries. Please consult your local sales representatives for details. PN65534-EN 0819M

Sebastien Gallien1,2, Jing Wang2, Aaron S. Gajadhar3, Bhavin Patel4, Markus Kellmann5, Tabiwang N. Arrey5, Alexander Harder5, Romain Huguet3, Graeme McAlister3, Derek Bailey3, Shannon Eliuk3, Yue Xuan5, Andreas Huhmer3, Emily L. Chen2

1Thermo Fisher Scientific, Paris, France; 2Thermo Fisher Scientific, Precision Medicine Science Center, Cambridge, MA; 3Thermo Fisher Scientific, San Jose, CA; 4Thermo Fisher Scientific, Rockford, IL; 5Thermo Fisher Scientific, Bremen, Germany

SureQuant MethodologyThe developed SureQuant acquisition method includes different scan events and filters, with parameters optimized for thehighest data quality and sensitivity of triggering, but also to retain sufficient triggering specificity (Figure 2). DuringSureQuant analyses, a high resolution (120k) full scan MS (1) is acquired, using lock mass re-calibration, to monitor thepredefined optimal precursor ions of the IS, based on the list of associated m/z values and intensity thresholds (defined at1% of the expected MS1 intensity at the chromatographic peak apex) included in the ‘Targeted Mass’ filter (2). In case ofthe detection of a candidate precursor ion satisfying the filtering criteria (MS peak with an m/z value within 3-10 ppmtolerance and intensity exceeding the threshold), it is subjected to fast MS/MS acquisition (3) performed at low resolutionand short fill times (typically 7.5k/10 ms), which is used as a second layer of confirmation of the detection of the IS.Several fragment ions, predefined as associated with the IS (6 optimal frag.) are included in the ‘Targeted Mass Trigger’filter (4), and the detection of a subset of them (at least 5 frag., 10-20 ppm tolerance) confirms the actual elution of the IS.This two-step process interrogating MS1 and MS2 data to track IS elution is the ‘watch’ mode of the SureQuant method.The detection of an IS in watch mode triggers the ‘quant’ mode, and therefore MS/MS acquisition of the correspondingendogenous pept. (5) with parameters favoring data quality, i.e., high resolution and long fill times (typically 60k/116 ms).The process is repeated over the entire LC separation with a systematic cycle-to-cycle assessment of the elution of the IS.The development of a SureQuant assay to be associated with a custom peptide panels requires the preliminarydetermination of the optimal precursor ion of each peptide target together with their intensity response and optimalassociated fragment ions. This information is typically extracted from an initial directed DDA analysis of the correspondingIS peptide mixture spiked into a representative matrix. The target list of multiple theoretical precursor ions of each ISpeptide included in such survey runs enables comprehensive data collection, and therefore the refinement of analytesdefinition for subsequent SureQuant analyses. The process is supported by Skyline software and facilitated by the genericdata acquisition and processing templates provided. The transfer of the generated information can be achieved throughcopy/paste from Skyline reports to ‘Targeted Mass’ and ‘Targeted Mass Trigger’ tables of SureQuant method. The surveyanalysis of a SureQuant assay needs to be run one time as the onset of a project without further adjustment over time,unless a major change in the experimental design is applied (e.g., LC set up, spiked amount of SIL peptides). In the lattercase, only the intensity responses of predefined IS precursor ions need to be re-adjusted though an additional survey runperformed under the modified experimental conditions. Therefore, the ability of embedding pre-set methods associatedwith predefined kits of IS peptides expedites SureQuant assay development at the user sites. Such validated methodtemplates have already been included in the instrument control software for the PQ500 related methods (Figure 3).

ABSTRACTPurpose: A novel global plasma proteome quantification workflow has been developed. It relies on a new high densitytargeted acquisition method, called ‘SureQuant’ method, implemented on next generation Orbitrap mass spectrometers:Thermo ScientificTM Orbitrap ExplorisTM 480 mass spectrometer and Thermo ScientificTM Orbitrap EclipseTM TribridTM

mass spectrometer. In addition, the workflow leverages a large set of stable isotopically labeled (SIL) peptides, used asinternal standards, to drive the systematic screening of more than 500 plasma proteins (PQ500 kit from Biognosys).

Methods: The SureQuant PQ500 analyses of undepleted plasma samples were performed using Orbitrap Exploris 480and Orbitrap Eclipse Tribrid mass spectrometers, coupled to Thermo ScientificTM UltiMateTM 3000 RSCLC system orThermo ScientificTM EASY-nLCTM 1200 system. An acquisition variant of the internal standard triggered targetedacquisition, dedicated to higher throughput analyses, was executed through an application programming interface (API)controlling a Thermo ScientificTM Q ExactiveTM HF-X hybrid quadrupole-OrbitrapTM mass spectrometer coupled to aEvosep One LC system. PQ500 kit from Biognosys was spiked into the plasma samples to drive the targeted acquisition.

Results: The SureQuant method has been adapted from the internal standard triggered PRM (IS-PRM) method in orderto improve its usability and robustness, while retaining its analytical performance, and has been implemented in the nativeinstrument control software of next generation Orbitrap mass spectrometers. The application of the method to the analysisof plasma samples, leveraging spiked-in SIL peptides from PQ500 kit, allowed robust and precise quantification of around560 endogenous peptides used as surrogate of 400 plasma proteins using a 70-min LC gradient. Therefore, the methodcombined data quality of targeted analyses with proteome coverage of state-of-the-art profiling experiments. In addition,the availability of pre-set PQ500 SureQuant methods embedded into the instrument control software enables a ‘load andplay’ execution of such plasma proteomics experiments.

INTRODUCTIONThe success of plasma proteomics studies to deliver useful protein biomarkers has remained lower than initialexpectations, prompting a redefinition of the plasma biomarker development pipeline. One proposed measure consisted inthe implementation of a “rectangular” approach, relying exclusively on broad proteome profiling (through DDA or DIAanalyses) across large cohorts for both the discovery and validation/verification stages. We describe here an alternativeglobal quantification workflow, relying on a high density targeted acquisition method implemented on next generationOrbitrap mass spectrometers. This adaptation of the IS-PRM method leverages spiked-in stable isotopically labeled (SIL)peptides to drive the systematic screening of more than 500 plasma proteins per analysis across large sample sets.

MATERIALS AND METHODSSample PreparationA commercial human plasma sample pooled from several individuals purchased from BioreclamationIVT was processedusing Thermo ScientificTM EasyPepTM Mini MS Sample Prep Kit.A set of 804 SIL peptides from PQ500 kit (Biognosys PN#Ki-3019-96) was spiked at around 80 fmol (median value) into 1µg of undepleted plasma sample. One µL of the sample was injected for LC-MS/MS analyses.

LC-MS/MS AnalysisFor SureQuant analyses, chromatographic separations were performed on UltiMate 3000 RSLC system equipped with C18trap cartridge (5 µm, 0.3 x 5 mm) and analytical column (2 μm, 0.15 x 150 mm), and EASY-nLC 1200 system equippedwith a C18 analytical column (2 μm, 0.15 x 150 mm), using a 70-min gradient separation. The LC systems were coupled toOrbitrap Exploris 480 and Orbitrap Eclipse Tribrid mass spectrometers. In SureQuant analyses, Full MS scans wereacquired at a resolution of 120k with a maximum fill time of 50 ms, while ddMS2 of IS and endogenous peptides wereacquired sequentially at a resolution of 7.5k or 60k and maximum fill times of 10 ms and 116 ms, respectively.The analyses using the high throughput IS-PRM acquisition variant were performed on Evosep One LC system equippedwith Evosep C18 analytical column (3 μm, 0.1 x 80 mm), using a 21-min gradient separation. The LC system was coupledto a Q Exactive HF-X controlled by an API. In the analyses performed with the ‘Single Scan Msx IS-PRM’ method (Figure1, right panel), PRM scans in watch mode were acquired at a resolution of 7.5k and a maximum fill time of 10 ms. In thequant. mode, the pairs of IS and endogenous peptides were measured simultaneously by ‘multiplex’ acquisition of theirfragments in a single PRM scan generated at a resolution of 120k and non-isochronous maximum fill times of 20 or 124ms for the IS and endogenous peptides, respectively.

CONCLUSIONS A novel global plasma IS triggered targeted quantification workflow, based on ‘SureQuant’ method has been

successfully implemented in the native instrument control software of Orbitrap Exploris 480 and Orbitrap EclipseTribrid MS.

The global plasma quantification workflow combines the figures of merit of targeted acquisition method with thecoverage capabilities of profiling methods for plasma proteomics. In addition, it exhibits an efficient portability and ahigh usability, benefiting from embedded preset PQ500 related methods.

The developed IS-PRM acquisition variant was explored to carry out high throughput analyses. It exhibited slightlylower performance, but also the capability for a 5-fold increase in throughput/experiment scale. Further investigation isneeded to evaluate the pertinence of its native implementation under the form of a SureQuant method variant.

For Research Use Only. Not for use in diagnostic procedures.

TRADEMARKS/LICENSING© 2019 Thermo Fisher Scientific Inc. All rights reserved. All trademarks are the property of Thermo Fisher Scientific andits subsidiaries. This information is not intended to encourage use of these products in any manner that might infringe theintellectual property rights of others.

Global Plasma Proteome Quantification Using Internal Standard Triggered Targeted Analyses

Figure 5. Sampling rate in SureQuant PQ500 global plasma analysis

SureQuant PQ500 Plasma QuantificationThe PQ500 SureQuant acquisition workflow has been applied to the analyses of 1 µg of undepleted plasma samplesupplemented with PQ500 kit (median spiked-in amount of SIL peptides of 80 fmol) using a 70-min LC gradient onUltiMate 3000 RSLC system. The survey run revealed that 802 of 804 SIL peptides were compatible (intensity responseand hydrophobicity) with the LC-MS setup used (Figure 4), and were retained for subsequent SureQuant analyses. Thehigh overall intensity response of the SIL peptides (median MS1 intensity at chromatographic peak apex of 1e8) insuredhigh acquisition efficiency (>90%) in SureQuant analyses, benefiting from the median MS1 triggering intensity threshold at1e6. This translated into a good overall sampling rate over the LC separation, exceeding 8 data points collected over theelution profile of the vast majority of the peptides, which also remained acceptable in the most crowded time ranges, with6 to 8 data points collected per peak (Figure 5). The 802 retained internal standards were systematically detected in thetriplicated SureQuant analyses of plasma samples, and triggered high quality measurement of the correspondingendogenous peptides, translating into a broad coverage of the undepleted plasma proteome (Figure 6). Around 400proteins were detected in each replicate (based on around 560 surrogate endogenous peptides), while 90% of the totalsets were systematically quantified across triplicates.

In addition to delivering a broad plasma proteome coverage, rivalling that of state-of-the-art profiling methods, SureQuantPQ500 analyses still provide targeted quantification performance (Figure 7). The amounts of endogenous peptides wereestimated based on the known amounts of IS injected and single-point quantification strategy. This revealed aquantification range between 4 amol and 15 pmol, spanning more than 6 orders of magnitude, which is in theory sufficientto cover the vast majority of current FDA-approved plasma biomarkers. The fragment ion signals of the lowest abundantquantified endogenous peptide (PFLVFIR, 4 amol/µg of plasma) have been extracted (Figure 7, lower panel), illustratingthe acceptable data quality even obtained for this challenging analyte. The quantification precision obtained on the entiredataset was excellent, with 75% of the endogenous peptides quantified with a CV below 10% (median CV value of 6%).The experiments were reproduced at a second site by another operator, using a different LC set-up (EASY-nLC 1200system operated in a one-column setup instead of UltiMate 3000 RSLC system operated in a two-column setup) (Figure8). After initial survey analysis, enabling the adjustment of internal standards MS1 triggering intensity threshold,SureQuant PQ500 analyses of undepleted plasma sample were repeated. It turned out that the quantification of theendogenous peptides between sites / systems / operators was very consistent, as reflected by the quantitative valuestightly aligned along the theoretical ’perfect’ correlation line, and the absence of bias revealed by the equation of theregression. The acquisition scheme of SureQuant method, fully avoiding any dependency on time-scheduling, enablesefficient method portability and strong robustness.

The analyses of plasma sample supplemented with PQ500 kit were reproduced using the ‘Single Scan Msx IS-PRM’acquisition method on QE HF-X MS coupled to Evosep One LC system operated at a 60 samples/day analyticalthroughput (21-min LC gradient). The sampling rate under these conditions remained significantly higher than thepredefined (relaxed) limit of 4-5 data points collected per peptide chromatographic peak (Figure 9, upper panel). Incomparison with SureQuant analyses performed with longer LC gradient, a substantial number of IS peptides (81) werenot detected, corresponding to the most hydrophilic peptides, which are not compatible with the LC setup used (Figure 9,lower panel). The remaining 721 IS peptides were properly detected in the watch mode of the method and triggeredmultiplex PRM measurement in the quant. mode, enabling the detection of 421 endogenous peptides, surrogate of 301plasma proteins. In Figure 10, the data generated by SureQuant and Single Scan Msx IS-PRM analyses are displayed forone endogenous peptide quantified by both methods. The quantification results were very consistent across the methods.

Figure 3. Overall PQ500 SureQuant acquisition workflow.

Figure 6. Peptide detection/quantification in SureQuant PQ500 analyses of undepleted plasma

Figure 7. Quantification performance of SureQuant PQ500 analyses of undepleted plasma

Figure 8. Consistency of SureQuant PQ500 Quantification across sites/systems

Figure 9. High throughput plasma analyses using Single Scan Msx IS-PRM method.

Figure 10. Endogenous peptide quantification from SureQuant and Single Scan Msx IS-PRM

Figure 1. Sequential and multiplex acquisition modes

LC-MS Analysis

SIL peptides (internal standards)

#1#2

#n….

Sample SureQuant

ENDO Peptides Pseudo Spectral Matching

‘QUANT MODE’High Quality MS2 of Target

‘WATCH MODE’

Fast, Low Res MS2 of Trigger

1

4

5

2 3

MS1 Survey Scan

Mix

MS/MS IS

IS peptides (trigger)

1

2

3

4

5

Figure 2. SureQuant acquisition Scheme and method structure

00.5

11.5

22.5

33.5

4

0 10 20 30 40 50 60 70 80 90[min]

Cycle Time [s]

Sampling rate

6-8 data points /peak(20-s peak width)

≥8 data points /peak(20-s peak width)

Intensity response of precursor ions

PQ500 kit in plasma

Survey Run

SureQuant

LC-MS Analysis

PQ500 kit Plasma sample LC-MS Analysis

Survey Run Analysis

Export Targeted Mass report

Update Targeted Mass intensity threshold

Target Quantification

PQ500 Directed DDAInclusion list optimal IS precursor ions

Ready-to-use template

Ready-to-use template PQ500 SureQuant

PQ500 Skyline analysis template

PQ500 Skyline analysis template

Targeted Mass

Copy/Paste

589 total peptides

Rep 2Rep 1

Rep 3

526 peptides

421 total proteins

Rep 2Rep 1

Rep 3

368 proteins

0.001

0.01

0.1

1

10

100

1000

10000

100000

0.001 0.01 0.1 1 10 100 1000 10000 100000

Site 2-Endo. Pep. Conc. [fmol/µg plasma]

Site 1-Endo. Pep. Conc. [fmol/µg plasma]

Theoretical ‘perfect’ correlationActual correlation

R2 = 0.97y = 1.0009 x + 0.0881

0

1

2

3

4

0 5 10 15 20

0.E+00

2.E+07

4.E+07

6.E+07

4.45 4.65 4.850.E+00

2.E+07

4.E+07

6.E+07

23 23.2 23.4 23.60.E+00

2.E+07

4.E+07

4.45 4.65 4.85

FQASVATPR

FQASVATPRWATCH MODE QUANT MODE

0.E+00

1.E+05

2.E+05

3.E+05

4.45 4.65 4.850.E+00

1.E+05

2.E+05

3.E+05

23 23.2 23.4 23.6

Single-Scan Msx IS-PRM (QE HF-X) SureQuant (Exploris 480)

Number of spectra including ≥ 5 ref. fragments

Peptide# 1…………………………………………………………………………………………………….802

Detection endo. pep. / prot.

564 pep. / 397 prot.

560 pep. / 396 prot.

550 pep. / 390 prot.

Internal Standards peptides

Endogenous peptides

0 1-2 ≥ 3

802 peptides

802 peptides

802 peptides

Internal Standards peptides

Endogenous peptides

Rep. 1

Rep. 2

Rep. 3

Rep. 1

Rep. 2

Rep. 3

Peptide# 1………………………………………………………………………………………………….802

Sampling rate

≥ 4-5 data points /peak(12-s peak width)

Internal Standards peptides

Endogenous peptides

Number of spectra including ≥ 4 ref. fragments 0 1

Detection pep. / prot.

712 peptides

421 pep. / 301 prot.

Non-detectable IS Hydrophobicity scale

- +

Intensity ratio ENDO/SIL (Ss Msx IS-PRM): 0.0051

Area ratio ENDO/SIL (SureQuant): 0.0049

[min]

Cycle Time [s]

0

50

100

150

200

250

300

<5% 5-10% 10-20% >20%

CV

0.001

0.01

0.1

1

10

100

1000

10000

0 100 200 300 400 500 600

Endo. Pep. Conc. [fmol/µg plasma]

Abundance Rank

AlbuNumber of endo. Peptide.

56.4 56.6 56.8 5756.4 56.6 56.8 57

SIL (10fmol) ENDO (calc: 4amol) PFLVFIR

8e7 4e4

[min] [min]

I IQuantification range: 4 amol – 15 pmol(>6 orders of magnitude)

CV < 10% for more than 75% of the peptides (median CV: 5.7%)

Sequential acquisition Multiplex acquisition

Figure 4. Characterization of PQ500 kit through survey run analysis.

1.0E+06

1.0E+07

1.0E+08

1.0E+09Median intensity: 1e8

MS1 Intensity Apex IS

1.0

10.0

100.0

1,000.0

10,000.0 Median amount: 80 fmol

Injected Amount IS [fmol]

1………………………………………………………………………………………………………………………….802Peptide#

High Throughput IS Triggered Plasma AnalysesA variant of SureQuant acquisition, dedicated to higher throughput analyses, has been developed using API-based IS-PRM application, which is only compatible with QE MS platforms. This method has not been natively implemented in theinstrument control software of orbitrap instruments and does not benefit from the revised acquisition strategy ofSureQuant method, but relies on the acquisition scheme of the original IS-PRM method. More specifically, in this ‘SingleScan Msx IS-PRM’ acquisition variant, the detection of an IS peptide in watch mode triggers the concomitantmeasurement of the corresponding pair of IS and endogenous peptides in a single PRM scan in quant. mode close to theirelution profile apex, following their ‘multiplex’ isolation (Figure 1). This contrast with the regular sequential acquisition, inwhich data points are collected over the full elution profiles of the peptides. With this acquisition variant, quantification isperformed based on the fragment ion intensities extracted from single high quality PRM scan, and not on the area underthe curve of the corresponding extracted fragment traces. This quantification strategy alleviates sampling rate constraint.

[min]

I

PO65534-EN0519S