LC/MS Application Note

Automated Parallel Derivatization Strategy with Broad Metabolite Coverage Coupled to SWATH/MS Data Acquisition for Qualitative and Quantitative Analysis

O

HO

H

H

H

S

NCH3

H3CO

ON

OH

NH3C

H3C

SO

O NH

ww

w.pa

lsyste

m.co

m

2 IngeniousNews 04/2018

Automated Parallel Derivatization Strategy with Broad Metabolite Coverage Coupled to SWATH/MS Data Acquisition for Qualitative and Quantitative AnalysisDavid Ruskic 1, Maria Fernanda Cifuentes Girard 1, Renzo Picenoni 2 , Günter Böhm 2, Gerard Hopfgartner 1

1) University of Geneva, Department of Analytical and Inorganic Chemistry, Life Sciences Mass Spectrometry, Geneva, Switzerland 2) CTC Analytics AG, Zwingen, Switzerland

Overview

• Parallel derivatization with Dansyl-Cl and Dansyl-N2H3 covers broad range of polar metabolites

• Improved retention on reverse phase LC by ~10 min for a total run time of 25 minutes

• Enhanced signal to noise (S/N) by a factor of 10 to 200• Automated derivatization with on-line C18 SPE based

fractionation improves reaction kinetics control for quantitative analysis

• QUAL/QUANT SWATH/MS data analysis using (13C) labelled derivatization agents

Introduction

Due to the vast variety of metabolite physicochemical properties there is no single workflow which can provide a large scale metabolomic coverage in particular for polar metabolites. With LC-MS several limitations have to be addressed including:poor MS response, chromatographic retention and MS/MS fragmentation. Among different sample preparation approaches, chemical derivatization has proven to be a suitable workflow to address these issues. We propose a parallel derivatization approach based on two well establishedprocedures covering a broad range of metabolites with respect to the different functional groups. By derivatization of the metabolites, highly polar analytes are converted to more hydrophobic products, thus improving retention on reverse phase LC by ~10 min for a total run time of 25 minutes. Due to the presence of a basic heteroatom on Dansyl-Cl and Dansyl-N2H3, increased chromatographic retention and enhancedsignal of analytes with a normally very low ESI response, the S/N can be improved by a factor of 10 to 200 times. Dansyl-Cl derivatizes primary alcohols, phenols, primary and secondary amines while Dansyl-N2H3 transforms ketones and aldehydesto hydrazones. To cover both classes of analytes two independent sample preparation batches would be required. With the automated parallel derivatization all analytes could

be analyzed in a single run with sufficient detection limits. Automation was found to be essential to allow reproducible derivatization as the reaction kinetics of individual analytes are highly variable within a short time scale (15 min).Using labelled (13C) Dansyl-Cl and Dansyl-N2H3 for derivatization of a mix of suitable internal standards with a parallel non-labelled derivatization of the sample (pooled urine sample) can facilitate SWATH data processing by following derivatizationspecific tag-ion fragments (12C m/z 170 and 13C m/z 172) for all derivatized analytes.An on-line SPE based fractionation using a C18 guard column was used to remove the excess of reagents and highly lipophilic background compounds which can significantly affect the ionization efficiency and chromatographic performance.The mix of heavy (13C) derivatized standards can be used for quantitative analysis of metabolites as individual internal standards. Due to the high dynamic range of different metabolites, SWATH/MS data acqusition allows the quatification on both precursor and product ion level including flexibility of the quantifier choice.

3IngeniousNews 04/2018

Methods

Fig 1: Schematics of the RTC autosampler used for automated parallel derivatization with hyphenation to two quaternary LPG pumps and MS QTOF instru-ment.

Fig 2: Schematics of the parallel derivatization of the sample with Dansyl-Cl and Dansyl-N2H3 and the metabolite standard mix with (13C) labelled Dansyl-Cl and Dansyl-N2H3. The incubation is performed in separate thermally stable compartments. Automated sample preparation is stopping both reactions by having an exact control of the reaction time between injections.

Q1 filter[20 u]

CIDramp

LC 30 AD Pumps

Phenomenexsecurity guard cartridge

C18 4x2 mm as trap

XSelect HSS T3 2.5 um2.1x150 mm Column XP 6600 QTOF (SCIEX)

CTC RTC (Robotic Tool Changer)Shimadzu

Agitator I Agitator II

Injecor valves

1 Agitator I Agitator II

Sample light labeling

Dansyl-hydrazineDansyl-chloride

10 s mix and inject

Metabolite standard mix heavy labeling

15 min mixing at 750 rpm

S

N

OOCl

C(12)H3H3(12)C

S

N

OOCl

C(13)H3H3(13)C

S

N

OON2H3

C(12)H3H3(12)C

S

N

OON2H3

C(13)H3H3(13)C

pH = 9.5T= 60°C

pH = 2T= 30°C

Derivatizatin of primary and secondary amines, phenols

and primary alcohols

Derivatization of aldehydes and ketones (hydrazones)

( )( )

O

HO

H

H

HS

NCH3

H3C O

O

N OH

N

H3C

H3C

SO

ONH

4 IngeniousNews 04/2018

Fig 4: SWATH data acquisition. An independent data acqui-sition that gives information of precursors and fragments within selected Q1 mass windows illustrated as a red frame. The window size and position is optimized for a medium mass range due to the mass shift resulting from metabolite derivatization.

The cycle time of 0.75 s (40 SWATH windows including TOF MS (100-1000 Da)) gives about 15 points across the chromatographic peak suitable for quantitative analysis. The collision energy of 50 eV with an energy spread of 40 eV allows an optimum compromise between sensitivity and fragmentation efficiency.

Fig 3: Time frame for the automatized sample derivatization and on-line SPE based sample clean up overlapped with the LC run time. At the start, Pump P2 is used for analytical column conditioning and for backflushing the trap after 13 min of the LC run. Pump P1 is gradient pump and at the start, it is trapping and washing out salt from the trap. All events are controlled with valve positions depicted on the graph.

2 Analyte elution from trap Trapping Trap Backflush

Gradient pump5% ACN, 0.3 ml/min

Backflush pumpFlow control

5% ACN, 0.3 ml/min

0 4 13 25 t (min)

95% ACN

95% IPA1.5 ml/min0.05 ml/min0.3 ml/min

Inject 1

Inject 2

11Valve 2_1

Valve 3_2 Valve 3_1 Valve 3_1

Valve 2_1 Valve 2_2

Sample prep overlaps with LC run timeLiquid

transfer

1

2

3

4

5

6

1 1

2

3

4

5

6

Injector

Valve 2 Valve 3

Waste

Waste

Analytical column

P2

P1

1 µl

3

010

0040

SW

ATH

Win

dow

s In

tens

ity (c

ps)

0 - 200 Da

7.5 7.52

199.5 - 300 Da

299.5 - 400 Da 399.5 - 420.5 Da

420 - 440 Da

800 - 1000 Da

TOF MS SWATH MS/MS

Cycle time (0.75 s)

5IngeniousNews 04/2018

Results

SWATH QUAL ANALYSIS

Fig 6: The graph represents the chromatography obtained from the trap without column switching. On-line SPE based sample fractionation and analyte are illustrated by the blue frame. XIC of the most abundant background components are on the left and right side of the chromatogram. The m/z = 226 and m/z = 252 are reagent related fragments (1 mg/ml was used for derivatization). The m/z = 352, on the right side repre-sents one of the highly lipophilic background components that is removed by the trap column backflush. Analytes from blue region (4 – 13 min) are eluted on the analytical column and significantly improving the chromatographic performance and column lifetime.

Fig 5: An example of fragmentation pattern of (12C/13C) derivatized L-phenylalanine in urine sample. In the spectrum on the top, three tag-fragments that represent common fragments of derivatized metabolites that belong to same chemical group containing amino, phenol or primarily alcohol group can be seen. The parallel derivatization with heavy/light isotope labelling helps the assignment of fragments that are related to tagging molecule and the identification of analyte-selective fragments. XIC offset traces of the precursor ion m/z = 399.1371, the tag-fragment m/z = 170.0964 and the analyte-selective fragment m/z = 120.0805 demonstrate that the tag-fragment is not a selective fragment and can therefore not be used as a quantifier.

150 200 250 300 350 400Mass/Charge, Da

500010000150002000025000

Inte

nsity

(cps

) SWATH Q1 isolation winodow 395.5-420.0 Da, from 13.381 to 13.541 min

13.45

13.1 13.3 13.5 13.7 13.9Time, min

10%

50%

90%

% In

tens

ity (o

f 908

3.4) XIC 399.1371 Da

XIC 120.0805 DaXIC 170.0964 Da

13.4513.45

Analyte selective fragment

NH

HO OH

S

N

O

O

+170.096172.103

234.058

(12C/13C)

(12C/13C)

399.137401.146

384.113385.116

236.063

385.1162384.1134

387.0761386.1174

399.1371

401.1435

156.0807170.0964158.0922

172.1029115.0545

120.0805 236.0625234.0581 353.1307339.1108

N H

HN H

H

120.0 120.5 121.0Mass/Charge, Da

120.0805

121.0833

2 4 6 8 10 12 14 16 18 20 22 24Time, min

1e5

3e5

5e5

7e5

9e5

Inte

nsity

(cps

) XIC +TOF MS: 226.9625 Da3-Cresotinic acid L-Proline 5-Aminopentanoic acid 3-Aminoisobutanoic acid Aminocaproic acid XIC +TOF MS: 352.3370 Da XIC +TOF MS: 252.0685 Da

Analyte elution from trap

2 4 6 8 10Time, min

5.0e4

1.0e5

1.5e5

Inte

nsity

(cps

)

Inte

nsity

(cps

)

11 12 13 14 15Time, min

1e5

2e5

3e5

4e5

L-Proline / 1.52

Homovanillic acid / 2.68

L-Histidine / 1.06

L-Phenylalanine / 3.72

L-Proline, 80 x

L-Histidine, 51 x

L-Phenylalanine, 10 xHomovanillic acid, 84 x

XIC from +TOF MS XIC from +TOF MS

Fig 7: XIC from precursor ions depicted in different colours for each analyte. The derivatized amino acids have a significant retention time shift of ~10 min on the reverse phase LC, moving to a retention time window with less interferences. Whereas the effect of the derivatization on the absolute signal is not significant, the S/N ratio is increased by up to 84 x (labelled on the right chromatogram).

6 IngeniousNews 04/2018

SWATH QUANT ANALYSIS

10 11 12 13 14 15 16 17 18 19Time, min

2e5

4e5

6e5

Inte

nsity

(cps

)In

tens

ity (c

ps)

x5.0 x10.0XIC from +TOF MS^2: 170.0957 +/- 0.0036 Da

XIC from +TOF MS^2: 172.105 +/- 0.006 Da Da

10 11 12 13 14 15 16 17 18 19Time, min

2e5

4e5

6e5

x10.0 x3.0 x3.0

14.28 18.03

18.56

12.4911.0914.57

10.47

10.26

11.13

16.49

18.03

12.49

11.37

12.4814.48

16.49

12.7011.09

18.56

10.47

10.26

18.03

NC(13)H3H3(13)C

NC(12)H3H3(12)C

Fig 8: Overlayed XICs from all SWATH Q1 isolation windows of tag-fragment m/z = 170 (top) and m/z = 172 (bottom). The urine sample was derivatized with non-labelled Dansyl-Cl II Dansyl-N2H3 whereas the metabolite standard mix (45 compounds) was derivatized with C13 labelled Dansyl-Cl II Dan-syl-N2H3. The 12C and 13C analogues match in retention time (10.26 min, 10.47 min, 11.09 min, 12.49 min, 16.49 min, 18.03 min) and are located within the same Q1 SWATH isolation windows. Chemically related metabolites are in similar SWATH windows and have similar retention what eases SWATH data processing and metabolite profiling.

5.00E+05

1.50E+06

2.50E+06

3.50E+06

n=10RSD ≤ 7 %

y = 9226.5x - 7964.9R² = 0.9991

0.0E+002.0E+054.0E+056.0E+058.0E+051.0E+06

0 50 100Dei

soto

ped

peak

are

a of

IS

ng/ml

y = 0.0249x - 0.0105R² = 0.9999

0

5

10

15

20

25

0 500 1000

Are

a ra

tio (A

IS/A

s)

ng/ml

Vanilic acid

4-Hydroxybenzoic acid

y = 0.0319x - 0.0325R² = 0.9999

02468

1012

0 100 200 300

Are

a ra

tio (A

IS/A

s)

ng/ml

2-Pyrocatechuic acid

y = 0.0203x + 0.1692R² = 0.9989

0

5

10

15

20

0 500 1000

Are

a ra

tio (A

IS/A

s)

ng/ml

L-Methionine

Fig 9: Automatized parallel light/heavy derivatization of sample/metabolite standard mix shows very high accuracy with low LOD and LOQ. All heavy deriva-tized analytes are individual internal standards for the corresponding identified metabolite. In combination with SWATH data acquisition the method has a high dynamic range due to the possibility for quantification at precursor or fragment level.

ConclusionsAutomated parallel light II heavy derivatization of sample II metabolite standard mix with Dansyl-Cl and Dansyl-N2H3 with SWATH independent data acquisition allows qualitative and quantitative analysis within same batch. SWATH data acquisition gives information about precursor and fragment ions of both light and heavy labelled analytes that eases fragmentation path assignments based on isotope pattern. XIC based on tag-ion fragment facilitate data processing giving information about chemically related analytes in addition to retention time match and accurate mass between heavy/light labelling. Fragmentation profile of derivatized analytes gives information about compound specific fragments (QUANT) and tag-fragment (QUAL)

Imprint

Date of print: 01.2019

CTC Analytics AGIndustriestrasse 20CH-4222 ZwingenSwitzerland T +41 61 765 81 00Contact: info@ctc.ch

Legal Statements

CTC Analytics AG reserves the right to make improvements and/or changes to the product(s) described in this document at any time without prior notice.

CTC Analytics AG makes no warranty of any kind pertaining to this product, including but not limited to implied warranties of merchantability and suitability for a particular purpose.

Under no circumstances shall CTC Analytics AG be held liable for any coincidental damage or damages arising as a consequence of or from the use of this document.

© 2019 CTC Analytics AG. All rights reserved. Neither this publication nor any part hereof may be copied, photocopied, reproduced, translated, distributed or reduced to electronic medium or machine readable form without the prior written permission from CTC Analytics AG, except as permitted under copyright laws.

CTC Analytics AG acknowledges all trade names and trademarks used as the property of their respective owners.

PAL is a registered trademark of CTC Analytics AG | Switzerland

www.palsystem.comVisit our homepage for more information.