08/07/2013

1

SWATH™ Acquisition; Accelerating Quantitative Metabolomics

Brigitte Simons, Academic Business Market Development Manager

AB SCIEX, Canada

2 © 2013 AB SCIEX

Metabolomics Workflow Strategies

MS Accelerating the Future of Metabolomics = Optimizing and combining the best of targeted and non-targeted strategies for comprehensive data collection with

scalability and throughput

08/07/2013

2

3 © 2013 AB SCIEX

Pave the Way Forward… Reducing Data Interpretation Bottlenecks…

Data Acquisition

Metabolite ID

Processing & Analysis

https://xcmsonline.scripps.edu/

4 © 2013 AB SCIEX

Pave the Way Forward…Reducing Data Interpretation Bottlenecks…

Quehenberger et al, J Lipid Res. 2010 Nov;51(11):3299-305

Design and Implementation of Quantitative Lipid Internal Standards

SWATH MS for Lipidomics

08/07/2013

3

5 © 2013 AB SCIEX

Pave the Way Forward… Reducing Data Interpretation Bottlenecks…

N

NN

N

N

O

O

122 (+O)

SWATH MS for Metabolite Fishing

Buspirone

6 © 2013 AB SCIEX

1 amu 1 amu

1amu 0.01 amu 0.01 amu0.01 amu

MRM/SRM

MRMHR

Precursors FragmentsCID

Triple quadrupole

QTOFs

Transition

X amu

TripleTOF™5600+

SWATH-MS

m/z m/z

Quantitative LC-MS Techniques

08/07/2013

4

7 © 2013 AB SCIEX

Cycle Time

< 1 s

Retention Time

Q1 width is user defined!

e.g 25 Da

SWATH = Sequential Windowed Acquisition of All Theoretical MS

Q1 Isolation Strategy

m/z

8 © 2013 AB SCIEX

400-425

Easy Method Development and Data Mining is Carried Out Retrospectively

424-450

450-475

475-500

500-525525-550

Time, min

Fra

gmen

t Ion

m/z

, Da

24 LCMS maps of fragment ions

Time, min

08/07/2013

5

9 © 2013 AB SCIEX

Why SWATH Analysis for Metabolomics? Data Independent Acquisition – no prioritization or bias in precursor data

collection given LC separations are not always ideal

Best Quant and Best Qual metabolite ID all in one run– Quant is MS/MS for best selectivity and signal to noise

– No pre-designed inclusion lists

– No pre-set compound specific information

All the information is extracted post LC-MS acquisition

Designing a balance between Specificity vs Sensitivity– Dependence on separation of isomers

– Association of Precursor and putative Products signals is critical

– Managing LC eluted m/z in dense regions of the chromatography

A Permanent Digital Record of MS/MS fragment of your samples– Come back to it for hypothesis-driven discovery

– Build an MRM method to transfer for targeted analysis easily.

10 © 2013 AB SCIEX

Theoretical Human MetabolomesHow Important is Relative Quantitation?

M mM M nM pM fM

Secondary endogenous metabolites

Secondary food metabolites

Secondary drug metabolites

Lipids/Lipid derivatives200,000 (Lipidome)

10,000 (Drug metabolome)

100,000 (Food metabolome)

10,000 (Secondome)

Wishart Bioinformatics Workshop; www.bioinformatics.ca

08/07/2013

6

11 © 2013 AB SCIEX

The Importance of MS/MS for Metabolite ID

11

25%

90%

Slide courtesy of Jiang Zhu; Analytical Chemistry, 2013, 85, 798-804

12 © 2013 AB SCIEX

SWATH MS and Data Independent Acquisitions

SWATH™ Acquisition

Quantitative

Technical repsBiological reps

IDA for General Unknown Screening

Qualitative

Metabolite Database

SWATH™ XIC Manager

Fast targeted or untargeted XIC

generation

08/07/2013

7

13 © 2013 AB SCIEX

AB SCIEX SWATH Quant/Qual Experimental Workflow

LipidView® Software

MS and SWATH MS/MS

Multivariate Statistical Analysis – Find the differential features– Link to raw data

Identify and confirm metabolite/lipid formula or structure 

FormulaFinderTM predicts elemental composition and MS/MS fragment assignment

ChemSpider for Structural information if not in your in‐

house library

toolsetstoolsets

14 © 2013 AB SCIEX

Metabolomics Study Design

• 12 plasma samples were consensually collected.

• 6 male and 6 female samples.

• Female samples were split into 2 groups, pre-menopausal (4) and post menopausal (2).

• Urine was diluted 1:4 with distilled water and centrifuged.

• A pooled sample was prepared using an aliquot from all samples.

08/07/2013

8

15 © 2013 AB SCIEX

Multivariate Statistical Analysis of MS Data

The PCA Scores (top) and Loadings (bottom) plots with Pareto scaling highlight the differences between the urine samples.

Principle Component Variable Grouping (PCVG) is used in the loadings plot to cluster ions with similar profiles (represented here in the same colour), accounting for 70 % of the variation.

m/z 655 is one of the peaks contributing to the difference in one of the urine samples

Differential sample

Component contributing to difference

16 © 2013 AB SCIEX

7-HYDROXYETHYLTHEOPHYLLINEC9H12N4O3

N-linolenoyl-glutamineC23H38N2O4

L-PhenylalanineC9H11NO2

N-Palmitoyl tryptophanFormula: C27H42N2O3

LC-MS Peak Alignment with Metabolite ID

08/07/2013

9

17 © 2013 AB SCIEX

Known Compound IdentificationScreen of SWATH MS/MS against in-house Libraries

Library Match Strategy for Known Compound Identification

• Identification criteria

1. Accurate mass TOF MS (ppm error)

2. Retention time (min)

Metabolite ID Confirmation criteria

3. TOF MS/MS spectral matching (purity score)

– Library MS/MS spectrum with experimental

4. Isotope pattern match

– theoretical vs. experimental

18 © 2013 AB SCIEX

Table showing name, formula, mass accuracy, RT, intensity, library hit & purity score

Arginine TOF MS spectrum

Mass accuracy: 0.1 ppm

Arginine TOF MS/MS spectrum

Purity score 85.3

Known Compound Identification- Library Match

08/07/2013

10

19 © 2013 AB SCIEX

Unknown Compound Identification

Steps for unknown compound ID using chemical logic

1. Elemental composition assignment using TOF MS

2. Elemental composition of fragment ion (TOF MS/MS)

3. Structure search for elemental composition in public database

( PubChem, ChemSpider, HMDB, MassBank, Metlin etc.)

4. Structural elucidation and fragmentation correlation

* All from a single injection

20 © 2013 AB SCIEX

162.0 162.5 163.0 163.5 164.0Mass/Charge, Da

0.0e0

5.0e4

1.0e5

1.5e5

2.0e5

2.5e5

Inte

nsity

163.1175 (1)163.1154 (1)

*162.1125 (1)

TOF MS Isotope pattern match

163.1154

1.2 ppm 163.1175

-0.6 ppm

Elemental Composition Assignment using TOF MS

08/07/2013

11

21 © 2013 AB SCIEX

60 70 80 90 100 110 120 130 140 150 160 170

Mass/Charge, Da

0

500

1000

1500In

ten

sit

y*162.1121*103.0388

*60.0806

*85.0283

*102.0914

*57.0335

*59.0725

Elemental Composition Assignment of Fragment Ions

TOF MS/MS Spectrum

Elemental composition

22 © 2013 AB SCIEX

Elemental Formula & Structure Search using Metabolite Databases including METLIN

08/07/2013

12

23 © 2013 AB SCIEX

Structural Elucidation in PeakView™ SoftwareMatch N-Acetylglutamic Acid Fragments from Structure with MS/MS Data

TOF MS spectra TOF MS/MS spectra

N-acetyl glutamic acidFragment assignment

24 © 2013 AB SCIEX

Hypothesis Driven Targeted Biochemical Pathway Workflow1. Targeted data processing using a list of known metabolites in any

biochemical pathway

2. Single injection workflow for quantitative and qualitative data

3. Validate hypothesis whether a specific pathway is effecting as a result of disease state or treatment or phenotype or genotype

4. Quick answers to biologists and fast turn around time

5. Ready to use list of metabolites in a given biochemical pathway like

– Glycolysis

– TCA cycle

– Acyl carnitines

– Phosphate pentose pathway

– Oxidative stress markers

– Eicosanoid pathway

– Leukotriene's

– Amino acids etc.

08/07/2013

13

25 © 2013 AB SCIEX

SWATH MS Quantitation of Acyl CarnitinesXIC Manager for Targeted Extraction of MS or MS/MS

Acyl carnitine C18C25H49NO4

MS/MS for confirmation

26 © 2013 AB SCIEX

Plasma

CSF

MarkerView™ SoftwareProfiling of Acyl Carnitines in CSF & Plasma Extracts

08/07/2013

14

27 © 2013 AB SCIEX

SWATH: 1 nM- 0.14 ng/mL SpermidineTOF MS: 1 nM- 0.14 ng/mL Spermidine

Quantitative Performance TOF MS Quant vs. SWATH Quant

28 © 2013 AB SCIEX

SWATH Acquisition: 1 nM- Amine-1 MW:153.05

High Selectivity with 5X less background noise

TOF MS: 1 nM- Amine-1 MW: 153.05

High background noise

* 8 scanned points across the peak * 8 scanned points across the peak

Quantitative Performance TOF MS Quant vs. SWATH Quant

08/07/2013

15

29 © 2013 AB SCIEX

TOF MS vs. SWATH™ Acquisition Quantification

SWATH Acquisition: 1 nM- Amine-2 MW : 177.1

High Selectivity with 10X less background noise

TOF MS: 1 nM- Amine-2 MW : 177.1

High background noise

* 8 scanned points across the peak * 8 scanned points across the peak

30 © 2013 AB SCIEX

cycle time~ 3.3 min

acquisition time

Q1 mass filter width = 0.7 Da

SWATH Acquisition Method Set-up for Lipidomics

200 ‐1200 (m/z)

700.550

701.551

702.552

703.553

704.554

705.555

1200.051

200.0550

08/07/2013

16

31 © 2013 AB SCIEX

Time (one cycle)

Pre

curs

or

mas

s m

/z

TOF MS Full Scan200 ms

1 ms

Total Cycle3.5 min

MS/MSALL covering 1000 MS/MS (200 ms each)

Q1 selection0.7 amu

Lipidomics by Flow Injection SWATHMethod Set-up

How to calculate cycle time?– Mass range

– Step size to cover mass range

– MS/MS accumulation time

– Looping in a TOF MS scan (~ 200 ms)

32 © 2013 AB SCIEX

0

100

200

conc

entr

atio

n, µ

M

TAG Lipid Profiling in Plasma

+MS TAG 52:3

Group A Group B

+MS/MS of TAG 52:3 NL 18:0

Group BGroup A

08/07/2013

17

33 © 2013 AB SCIEX

Conclusions

SWATH adopting the concept of Data Independent Acquisition is the future of Quant/Qual metabolomics workflows– MRM quantitative data mining retrospectively– 100% MS/MS coverage – “The Ultimate Safety Net”

Design a balance of sensitivity and selectivity– Lipidomics focuses data collection in MS/MS with narrow windows– Metabolomics by LC-SWATH is optimized for UPLC separations with wider

windows stepped quickly across a wider mass range– A totally generic method – no need to change the methods, CE is ramped

Follow up with a targeted MRMHR method, + CE optimized, and targeted data processing for quantitative reporting of the important metabolites

34 © 2013 AB SCIEX

Acknowledgments

University of Toronto & Toronto General Hospital

– Johanne Allard– Bianca Arendt– Elaheh Aghdassi– David Ma

VTT Technical Research Institute of Finland

– Kari Raino

AB SCIEX, USA– Fadi Abdi– Mark Cafazzo– Christie Hunter

AB SCIEX, CAN– Suma Ramagiri– Paul Baker– Eva Duchoslav– Dave Cox– Lyle Burton

AB SCIEX, UK– Vicki Gallant, PhD

– Baljit Ubhi, PhD

08/07/2013

18

35 © 2013 AB SCIEX

Thank You for Your Attention

36 © 2013 AB SCIEX

Trademarks/Licensing

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

The trademarks mentioned herein are the property of AB Sciex Pte. Ltd. or their respective owners. AB SCIEX™ is being used under license.

© 2013 AB SCIEX. All rights reserved. Information subject to change without notice.