Development of a Liquid Chromatography–High ... - ACS Publications

Jun 22, 2017 - The percent relative standard deviations between ion ratios of a ... screening was then created, merging m/z, RT, MS/MS, and ion ratio ...
0 downloads 0 Views 1MB Size
Subscriber access provided by TUFTS UNIV

Article

Development of an LC-HRMS metabolomics method with high specificity for metabolite identification using all ion fragmentation (AIF) acquisition Shama Naz, Hector Gallart-Ayala, Stacey Reinke, Caroline Mathon, Richard Blankley, Romanas Chaleckis, and Craig Edward Wheelock Anal. Chem., Just Accepted Manuscript • Publication Date (Web): 22 Jun 2017 Downloaded from http://pubs.acs.org on June 23, 2017

Just Accepted “Just Accepted” manuscripts have been peer-reviewed and accepted for publication. They are posted online prior to technical editing, formatting for publication and author proofing. The American Chemical Society provides “Just Accepted” as a free service to the research community to expedite the dissemination of scientific material as soon as possible after acceptance. “Just Accepted” manuscripts appear in full in PDF format accompanied by an HTML abstract. “Just Accepted” manuscripts have been fully peer reviewed, but should not be considered the official version of record. They are accessible to all readers and citable by the Digital Object Identifier (DOI®). “Just Accepted” is an optional service offered to authors. Therefore, the “Just Accepted” Web site may not include all articles that will be published in the journal. After a manuscript is technically edited and formatted, it will be removed from the “Just Accepted” Web site and published as an ASAP article. Note that technical editing may introduce minor changes to the manuscript text and/or graphics which could affect content, and all legal disclaimers and ethical guidelines that apply to the journal pertain. ACS cannot be held responsible for errors or consequences arising from the use of information contained in these “Just Accepted” manuscripts.

Analytical Chemistry is published by the American Chemical Society. 1155 Sixteenth Street N.W., Washington, DC 20036 Published by American Chemical Society. Copyright © American Chemical Society. However, no copyright claim is made to original U.S. Government works, or works produced by employees of any Commonwealth realm Crown government in the course of their duties.

Page 1 of 30

1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60

Analytical Chemistry

1

Development of an LC-HRMS metabolomics method with high specificity for metabolite

2

identification using all ion fragmentation (AIF) acquisition

3 4

Shama Naz‡ǂ, Hector Gallart-Ayala‡ǂ, Stacey N. Reinke‡, Caroline Mathon‡, Richard Blankley§,

5

Romanas Chaleckis‡ ll, and Craig E. Wheelock‡ ll *

6 7



8

Karolinska Institutet, Stockholm, Sweden

9

§

Division of Physiological Chemistry 2, Department of Medical Biochemistry and Biophysics,

Agilent Technologies, Cheadle, Cheshire, UK

ll

Gunma University Initiative for Advanced Research (GIAR), Gunma University, Gunma, Japan

11

ǂ

Authors equally contributed to the manuscript

12

*

13

Craig E. Wheelock, PhD

14

Division of Physiological Chemistry 2

15

Department of Medical Biochemistry and Biophysics

16

Karolinska Institutet, 17177 Stockholm, Sweden

17

Email: [email protected]

18

Phone: +46 8 524 87630, Fax: +46 8 736 0439

10

Correspondence to be addressed to:

19 20

1 ACS Paragon Plus Environment

Analytical Chemistry

1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60

Page 2 of 30

1

Abstract

2

High-resolution mass spectrometry (HRMS)-based metabolomics approaches have made

3

significant advances. However, metabolite identification is still a major challenge and significant

4

bottleneck in translating metabolomics data into biological context. In the current study, a liquid

5

chromatography (LC)–HRMS metabolomics method was developed using an all ion

6

fragmentation (AIF) acquisition approach. In order to increase the specificity in metabolite

7

annotation, four criteria were considered: i) accurate mass (AM), ii) retention time (RT), iii)

8

MS/MS spectrum, and iv) product/precursor ion intensity ratios. We constructed an in-house

9

mass spectral library of 413 metabolites containing AMRT and MS/MS spectra information at

10

four collision energies. The % relative standard deviations between ion ratios of a metabolite in

11

an analytical standard vs. sample matrix were used as an additional metric for establishing

12

metabolite identity. A data processing method for targeted metabolite screening was then

13

created, merging m/z, RT, MS/MS and ion ratio information for each of the 413 metabolites. In

14

the data processing method, the precursor ion and product ion were considered as the quantifier

15

and qualifier ion, respectively. We also included a scheme to distinguish co-eluting isobaric

16

compounds by selecting a specific product ion as the quantifier ion instead of the precursor ion.

17

An advantage of the current AIF approach is the concurrent collection of full scan data, enabling

18

identification of metabolites not included in the database. Our data acquisition strategy enables a

19

simultaneous mixture of database-dependent targeted and non-targeted metabolomics in

20

combination with improved accuracy in metabolite identification, increasing the quality of the

21

biological information acquired in a metabolomics experiment.

22

2 ACS Paragon Plus Environment

Page 3 of 30

1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60

Analytical Chemistry

1

High-resolution mass spectrometry (HRMS)-based metabolomics has become an integral method

2

for understanding health and disease 1,2. Metabolomics has been used to obtain insight into

3

multiple biochemical processes including biomarker discovery, food safety, and nutrition 1,3,4,

4

and is considered an important component of precision medicine initiatives 5. These studies are

5

often performed without a prior knowledge of metabolite identity, and compound identification

6

is frequently based upon database searches (e.g., HMDB, Metlin, KEGG) 6-8. The accurate

7

identification/annotation of metabolites is a vital component of transferring HRMS data into

8

biological information; however, it remains a significant bottleneck in non-targeted

9

metabolomics to correctly annotate the biological identity of a detected feature 9-12. Several

10

guidelines have been proposed for metabolite identification to aid in the ability to directly

11

compare data from different studies and laboratories 13-16, following upon the criteria proposed

12

by the Metabolite Standard Initiative (MSI) 17.

13

The MSI defined four levels of metabolite identification 17. The highest level (level 1) is

14

based on matching two or more orthogonal properties [e.g., accurate mass (AM), retention time

15

(RT)/index, isotopic pattern, MS/MS spectrum] of an authentic reference standard analyzed

16

under the same condition as the metabolite of interest. This level of structural information

17

provides a high level of confidence in metabolite identity, but is resource intensive. Attempts

18

have also been made to develop targeted screening methods for a large number of metabolites 18-

19

20

20

of targeted metabolites, which restricts discovery efforts to those metabolites included in the

21

targeted list. While generating MS/MS data for all metabolites is challenging, a number of tools

22

have been proposed for in silico MS/MS-based metabolite identification 21-23. However, even this

23

level of annotation does not address the issue of identifying co-eluting isobaric compounds,

. However, while useful, these screening methods are by definition limited to a selected group

3 ACS Paragon Plus Environment

Analytical Chemistry

1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60

Page 4 of 30

1

which traditionally requires chromatographic resolution. To improve metabolite identification

2

and reduce the requirement for multiple analytical runs for structural confirmation, two different

3

MS/MS strategies have been implemented to date in non-targeted metabolomics: i) with

4

selection of the precursor ion (data dependent acquisition), and ii) without selection of the

5

precursor ion [all ion fragmentation (AIF), data independent acquisition (DIA), and MSE] 24-31.

6

DIA-based MS generates MS/MS spectra that contain a mixed population of product ions

7

together with their precursor ions and the extracted ion chromatogram (EIC) of each product ion

8

needs to be mapped to its parent compound. This can be a challenging process; however, recent

9

software developments have addressed some of these issues 32,33. The AIF and MSE approaches

10

have been successfully used to conduct multiple fragmentation experiments in a single

11

acquisition 30,31,34,35. The difficulty in identifying co-eluting isobaric compounds has been

12

suggested to be solved using a DIA-based approach in combination with software deconvolution

13

algorithms that merge precursor ions from low energy experiments and product ions from high

14

energy experiments 32. This approach has been successfully applied in lipidomics 35,36.

15

The aim of the current work was to establish a comprehensive analytical workflow for the

16

application of liquid chromatography-mass spectrometry (LC-MS) to non-targeted

17

metabolomics, with a high level of accuracy in metabolite identification employing the all ion

18

fragmentation (AIF) approach. To accomplish this aim, we developed an HRMS-based

19

metabolomics method coupled to both reversed phase (RP) and hydrophilic interaction liquid

20

chromatography (HILIC) for metabolite screening. The applied AIF mode includes three

21

sequential full scans at 0, 10 and 30 eV collision energies. In the subsequent data analysis, EIC

22

from any precursor or associated product ions of interest can be extracted from the low or high

23

energy scan data. One EIC is chosen for relative quantification (the quantifier ion) of the

4 ACS Paragon Plus Environment

Page 5 of 30

1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60

Analytical Chemistry

1

metabolite and further product ions from the same compound are used as qualifier ions. The

2

ratios of qualifier/quantifier ion intensities are established from authentic analytical standards,

3

and should be preserved when measured in a biological sample, increasing the accuracy of the

4

identification. The same acquired data (0 eV) can be used in parallel for a global metabolite

5

profiling workflow, enabling a combined database-dependent targeted and non-targeted

6

metabolomics experiment. The combination of the AIF-based data acquisition with the ion ratio

7

confirmation and deconvoluted co-eluting isobaric pairs provides a useful method for increasing

8

the accuracy of metabolite identification in a metabolomics experiment.

9 10

Materials and methods

11 12 13

Reagents and Chemicals LC-MS grade water and formic acid were purchased from Sigma-Aldrich (St. Louis,

14

USA). Acetonitrile (Optima ®-LC/MS), methanol (Optima ®-LC/MS) and 2-propanol (Optima

15

®-LC/MS) were purchased from Fisher-Scientific (Loughborough, UK). The internal lock

16

masses (purine and HP-0921) and tune mix for calibrating the TOF-MS (ESI-low concentration

17

tuning mix) were purchased from Agilent Technologies (Santa Clara, USA). The analytical

18

standards used to construct the compound spectral database, as well as the internal standards

19

(Tables S1-S5), were purchased from Sigma Aldrich (St. Louis, USA), Cayman Chemical

20

Company (Michigan, USA), Toronto Research Chemicals (Ontario, Canada), Zhejiang Ontores

21

Biotechnologies Co., Ltd (Zhejiang, China) and Avanti Polar Lipids, Inc. (Alabama, USA)

22

depending upon availability. The internal standards and standards were prepared at 1 mM

5 ACS Paragon Plus Environment

Analytical Chemistry

1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60

1

concentrations in the appropriate solvent for dissolution, stored at −20 °C and diluted

2

appropriately on the day of analysis.

Page 6 of 30

3 4 5

LC-HRMS instrumentation All experiments used a 1290 Infinity II ultra-high performance liquid chromatography

6

(UHPLC) system coupled to a 6550 iFunnel quadrupole-time of flight (Q-TOF) mass

7

spectrometer equipped with a dual AJS electrospray ionization source (Agilent Technologies,

8

Santa Clara, CA, USA).

9

Metabolite separation was performed with two complementary stationary phases. Polar

10

metabolites were separated on a HILIC SeQuant® ZIC®-HILIC (Merck, Darmstadt, Germany)

11

column 100 Å (100 mm × 2.1 mm, 3.5 µm particle size) coupled to a guard column (2.1 mm × 2

12

mm, 3.5 µm particle size) and an inline-filter. Sample analysis in both positive and negative

13

ionization mode was performed using water with 0.1 % formic acid (solvent A) and acetonitrile

14

with 0.1 % formic acid (solvent B). The elution gradient used was as follows: isocratic step at 95

15

% B for 1.5 min, 95 % B to 40 % B in 12 min and maintained at 40 % B for 2 min, then

16

decreasing to 25% B at 14.2 min and maintained for 2.8 min, then returned to initial conditions

17

over 1 min, and the column was equilibrated at initial conditions for 7 min. The flow rate was 0.3

18

mL/min, injection volume was 2 µL and the column oven was maintained at 25 °C.

19

Non-polar metabolites were separated on a RP Zorbax Eclipse Plus C18, RRHD (Agilent

20

Technologies, USA) (100 mm × 2.1 mm, 1.8 µm particle size) column coupled to a guard

21

column (5 mm × 2 mm, 1.8 µm particle size) and an inline-filter. Sample analysis in both

22

positive and negative mode was performed using water with 0.1 % formic acid (solvent A) and

23

2-propanol: acetonitrile (90:10, v/v) with 0.1 % formic acid (solvent B). The gradient elution was

6 ACS Paragon Plus Environment

Page 7 of 30

1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60

Analytical Chemistry

1

set as follows: isocratic step of 5 % B for 3 min, 5 % to 30 % B in 2 min, then B was increased to

2

98 % in 13.5 min, maintained at 98 % B for 1.5 min, and returned to initial conditions over 0.5

3

min, then held for a further 4.5 min. The flow rate was 0.4 mL/min, injection volume was 2 µL,

4

and the column oven was maintained at 50 °C.

5

Both chromatographic separations were coupled to an Agilent 6550 Q-TOF-MS system.

6

The system was calibrated and tuned according to the protocols recommended by the

7

manufacturer. Nitrogen (purity >99.999%) was used as a sheath gas and drying gas at a flow of 8

8

L/min and 15 L/min, respectively. The drying and sheath gas temperature was set at 250 °C, with

9

the nebulizer pressure at 35 psig and voltage 3000 V (+/- for positive and negative ionization

10

mode). The fragmentor voltage was set at 380 V. The acquisition was obtained with a mass range

11

of 40-1200 m/z for HILIC and 50-1200 m/z for RP in AIF mode, where a single high-resolution

12

full scan is acquired including three sequential experiments at three alternating collision energies

13

(one full-scan at 0 eV, followed by one MS/MS scan at 10 eV and then followed by one MS/MS

14

scan at 30 eV). The data acquisition rate was 6 scans/s.

15

An internal lock mass mixture (Agilent Technologies, Santa Clara, USA) was prepared at

16

a final concentration of 2 µM purine (C5H4N4) and 2.5 µM HP-0921 (C18H18O6N3P3F24) in

17

acetonitrile:water (19:1, v/v). The internal lock mass mixture was constantly infused at a flow

18

rate of 1mL/min (split 1:100) using an isocratic pump together with the LC eluent for constant

19

mass correction [positive ionization mode: purine (m/z 121.0509), HP-0921 (m/z 922.0098); and

20

negative ionization mode: purine (m/z 119.0363), HP-0921 (m/z 966.0007, HP-0921+formate

21

adduct)]. Although observed mass accuracy will depend upon the resolution, potential metabolite

22

co-elution and isobaric compounds, a mass accuracy of