Subscriber access provided by Karolinska Institutet, University Library
Article
Hydrogen/Deuterium exchange aids compounds identification for LC-MS and MALDI imaging lipidomics Yury I. Kostyukevich, Gleb Vladimirov, Elena Stekolschikova, Daniil G. Ivanov, Arthur Yablokov, Alexander Ya. Zherebker, Sergey Sosnin, Alexey Orlov, Maxim Fedorov, Philipp Khaitovich, and Evgeny N. Nikolaev Anal. Chem., Just Accepted Manuscript • Publication Date (Web): 06 Sep 2019 Downloaded from pubs.acs.org on September 6, 2019
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 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 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.
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 29 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
Hydrogen/Deuterium exchange aids compounds
2
identification for LC-MS and MALDI imaging
3
lipidomics.
4
5
AUTHOR NAMES
6
Yury Kostyukevicha,b, Gleb Vladimirova, Elena Stekolschikovaa, Daniil Ivanovb,c,
7
Yablokovd, Alexander Zherebkera, Sergey Sosnina, Alexey Orlova, Maxim Fedorova, Philipp
8
Khaitovicha and Evgeny Nikolaev*a
9
AUTHOR ADDRESS
10
a
11
Federation
12
b
13
c Emanuel
14
4. Moscow, Russia
Arthur
Skolkovo Institute of Science and Technology Novaya St., 100, Skolkovo 143025 Russian
Moscow Institute of Physics and Technology, 141700 Dolgoprudnyi, Moscow Region, Russia Institute of Biochemical Physics, Russian Academy of Sciences, 119334 Kosygina st.,
15
ACS Paragon Plus Environment
1
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
dInstitute
2
38 k.2, 119334 Moscow, Russia
Page 2 of 29
for Energy Problems of Chemical Physics Russian Academy of Sciences Leninskij pr.
3
4
KEYWORDS
5
Mass spectrometry imaging, lipidomics, MALDI, LCMS, Orbitrap, H/D exchange, ionization.
6
ABSTRACT
7
We present the novel approach for the increasing reliability of the compound identification for
8
LC-MS and MALDI imaging lipidomics. Our approach is based on the characterization of
9
compounds not only by the elution time, accurate mass and fragmentation spectra, but also by
10
the number of labile hydrogens which can be measured using Hydrogen/Deuterium exchange
11
approach. Number of labile hydrogens (those from –OH, -NH groups) serves as an additional
12
structural descriptor used when performing database search. For LC-MS experiment the H/D
13
exchange was performed in the heating capillary of the modified ESI source, while for MALDI-
14
imaging the exchange was performed in the ion funnel at the 10 torr pressure. It was observed
15
that such approach allowed achieving a considerable degree of deuteration enough to
16
unambiguously distinguish between different classes of lipids. The proposed analytical approach
17
may be successfully used not only for identification of lipids but also for peptides and
18
metabolites. A special software for automatic filtration of molecules based on the number of
19
functional groups was also developed.
20
21
ACS Paragon Plus Environment
2
Page 3 of 29 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
2
Analytical Chemistry
INTRODUCTION Accurate
and
reliable
identification
of
unknown
compounds
and
comprehensive
3
characterization of biological systems on the molecular level are the ultimate goals of the mass
4
spectrometry. Mass spectrometry imaging (MSI) is a powerful tool for direct determination of
5
the distribution of proteins, peptides, lipids, neurotransmitters, metabolites and drugs in neural
6
tissue sections in situ. Since its first introduction in 1962 by Castaing and Slodzian1 MSI evolved
7
from the secondary ion mass spectrometry (SIMS) imaging of elements2, to the molecular
8
imaging of biological objects3. Currently, desorption electrospray ionization (DESI)4 and matrix
9
assisted laser desorption ionization (MALDI)5 imaging techniques are most popular in the
10
biological and clinical applications of MSI, allowing visualization of the distribution of
11
thousands of lipids and proteins. Recently the possibility to reconstruct a 3D image from 2D
12
images of tissues was demonstrated6, 7. It is worth mentioning, that many alternative ionization
13
approaches were developed to perform MSI, such as electrospray laser desorption ionization
14
(ELDI) imaging8, direct liquid extraction imaging9 and many others10. A major disadvantage of
15
the MSI is the difficulty of the identification of molecules. Currently the time-of-flight mass
16
spectrometers are the main instruments for MSI and such devises not only have insufficient
17
resolving power ( 50 000) but also have large isolation windows what complicates the compound
18
identification when performing MS/MS.
19
Additional structural information about individual molecules that could assist identification
20
can be obtained by using selective chemical modifications such as H/D exchange (HDX) or
21
16O/18O
22
structure of the molecule (SIMLES or SMARTS) is known. So, applying modern tools of
23
chemical informatics, the number of exchanges observed experimentally for unknown molecule
24
can be used as structural descriptor or filter parameter when performing database search11, 12.
exchange. Under the mild conditions, the number of exchanges can be predicted if the
ACS Paragon Plus Environment
3
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
The first Hydrogen/Deuterium and 14.
16O/18O
Page 4 of 29
isotopic exchange reactions were observed by
2
Lewis in 193313,
3
exchange of protons from all functional groups such as –OH, -COOH, -NH, -SH. Addition of
4
acid and base or performing the reaction under elevated temperature and pressure leads to the
5
HDX in the electron enriched sites of aromatic ring. In a few years, the isotopic exchange
6
reaction attracted sufficient interest and great amount of research were dedicated to the
7
investigation of isotopic exchange reaction of all known substances from small molecules15-17 to
8
peptides and proteins18-21. Rapidly the isotopic exchange reaction became important for structural
9
characterization of unknown structures and gaining knowledge about the mechanism of many
Simple dissolving of sample in D2O instead of H2O leads to the instant
22.
10
chemical or biological reactions
With the introduction of the chemical ionization (CI)
11
methods in 196623-25 it was suggested by Donald Hunt to use the gas phase H/D exchange
12
approach in the combination with the CI method for the enumeration of active hydrogen atoms in
13
the molecular ions26, 27. In a few years the CI H/D exchange was combined with Ion Cyclotron
14
Resonance (ICR) mass spectrometry by J.L. Beauchamp and co-workers28 and the gas-phase
15
H/D exchange reactions of many organic cations and anions were investigated29-34. New
16
ionization methods (ESI and MALDI) instantly opened new possibilities of the H/D exchange
17
application for the investigation of the conformations of macromolecules in solution 35, 36 and gas
18
phases37. The use of gas-phase H/D exchange for the investigation of the gaseous conformations
19
of proteins was also demonstrated38. Recently it was demonstrated that HDX could be coupled to
20
the GC-ESI-MS systems39 and IMS systems40. It was also shown that Hydrogen/Deuterium
21
exchange can be performed in the gas phase at atmospheric pressure41-51, low52 or high vacuum53
22
conditions.
23
Robust identification of lipids, which are the major aim of MSI experiments, can be
24
achieved by identification of a number and the position of double bonds in hydrocarbon chains.
ACS Paragon Plus Environment
4
Page 5 of 29 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
Previously Pain et al54 have developed an approach for isomeric resolution using ozone-induced
2
dissociation55. Impact of ozone result in a cleavage of double bonds in molecules, which
3
facilitates the identification of isomeric compounds. The Paternò–Büchi56 reaction also can be
4
used to enumerate number of double bond position in lipids.
5
Here we present an integration of the H/D exchange approach to the low-pressure MALDI-
6
Orbitrap imaging57 and to the conventional LCMS lipidomics. For LCMC experiment the H/D
7
exchange was performed in the heating capillary of the modified ESI source, while for MALDI-
8
imaging the exchange was performed in the ion funnel at the 10 torr pressure.
9 10
A special software for automatic filtration of molecules based on the number of functional groups was also developed.
11
12
METHODS
13
Samples. Standard lipids 1,2-diheptadecanoyl-sn-glycero-3-phosphoethanolamine (17:0 PE)
14
and 1,2-diheptadecanoyl-sn-glycero-3-phosphocholine (17:0 PC) were used for a calibration and
15
verification of HDX results. Brain tissues frozen by liquid nitrogen was sectioned into 9-μm
16
slices using a cryostat Leica CM1950 at –18°C. Slices of tissue sections were placed on ITO
17
slides. Inbred mouse (C57BL/6 strain, males 3-4 months) was killed by the damage of the spinal
18
cord. The brain was opened and removed immediately. The resulting material (brain) was frozen
19
with liquid nitrogen and then fixed on the holder of the microtome-cryostat Leica CM1950 with
20
deionized water and cut in the following conditions: the temperature of the chamber -18oC, the
21
temperature of the sample -15oC. The thickness of the sections was 9 microns. Mouse brain
22
horizontal section in region with coordinates Bregma: ~-7mm, Interaural:~+3 mm was taken for
23
analysis. We also used fresh frozen postmortem rhesus macaque (Macaca mulatta) brain
ACS Paragon Plus Environment
5
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 6 of 29
1
sections. Macaque brain samples were obtained from the Suzhou Experimental Animal Center,
2
China. Posterior lobe of cerebellum from macaque male aged 26 years old was used to prepare
3
serial 20 um thick sections by Leica CM1950 cryomicrotome. Saturated solution of α-Cyano-4-
4
hydroxycinnamic acid (HCCA) in 50:50 water:acetonitril with 0.1% TFA was diluted two times.
5
This solution was applied using an airbrush (iwata мicron cm-b2) during 2 seconds and left to
6
dry during 2.5 minutes, 20 repetitions were performed. Lipid extraction for LCMS was
7
performed as follows: Brain tissue from adult mouse (10-15mg) was dissected and transferred to
8
a pre-cooled 2 ml Precellys tubes (Bertin Technologies). After addition of 0.5 ml of cold
9
extraction buffer MTBE:methanol (3:1), tissue was subjected to homogenization using a
10
Precellys Evolution homogeniser (Bertin Technologies, Montigney-le-Bretonneux, FR).Then 0.5
11
mL of cold extraction buffer was added, mixture was sonicated in an ice-cooled sonication bath
12
for 10 min and incubated for 30 min at 4 °C in a shaker. Phase separation was induced by adding
13
700 μL of H2O:methanol mixture (3:1) followed by vigorous shaking for 10 min and
14
centrifugation for 5 min at 13,000 rpm. Finally, 540 μl of the upper organic phase was
15
evaporated in a Speed Vac concentrator at room temperature. Dried lipid pellet was stored at -80
16
°C and resuspended in 200 µl ice-cold acetonitrile:isopropanol mixture (7:3). After brief rigorous
17
vortexing, the sample was shaken for 10 min, sonicated in an ice-cooled sonication bath for 10
18
min, and centrifuged 5 min at 13,000 rpm. For mass spectrometry analysis, sample was diluted
19
1:10 with acetonitrile:isopropanol (7:3).
20
Mass spectrometry imaging. Images were obtained using modified MALDI-Orbitrap mass
21
spectrometer57 (Thermo Scientific Q-Exactive orbitrap with MALDI/ESI Injector from
22
Spectroglyph, LLC) equipped with an 355nm Nd:YAG Laser Garnet (Laser-export. Co. Ltd,
23
Moscow, Russia). For producing of positive ions laser power was set to 20 uJ operating at
24
repetition rate of 1.7kHz. In IMS experiments, a sample was placed on a coordinate table 5 cm
ACS Paragon Plus Environment
6
Page 7 of 29 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
from ion funnel. Produced ions were captured by ion funnel and transferred to QExactive
2
Orbitrap mass spectrometer (Thermo). Mass-spectra where obtained in mass range of m/z 100–
3
1000, and a mass resolution was 140,000. The tissue region to be imaged and the raster step size
4
were controlled using the Spectroglyph MALDI Injector Software. To generate images, the
5
spectra were collected at 35-μm intervals in both the x and y dimensions across the surface of the
6
sample. Ion images were generated from raw files (obtained from Orbitrap tune software) and
7
coordinate files (obtained from MALDI Injector Software) by Image Insight software from
8
Spectroglyph LLC. Details of the experimental setup are presented below. MALDI-MS/MS
9
experiments were also performed using Ultraflex time-of-flight mass spectrometer (Bruker,
10
Germany).
11
Liquid chromatography/mass spectrometry
12
The liquid chromatography/mass spectrometry system consisted of a Waters Acquity UPLC
13
system (Waters, Manchester, UK) and a Q Exactive orbitrap mass spectrometer (Thermo Fisher
14
Scientific, USA) equipped with a heated electrospray ionization (HESI) probe. Separation of
15
lipids was performed at 60 °C using a reverse phase ACQUITY UPLC BEH C8 column (2.1 ×
16
100 mm, 1.7 μm, Waters co., Milford, MA, USA)equipped with Vanguard precolumn at a flow
17
rate of 0.4 mL/min. The mobile phases consisted of water containing 10 mM ammonium acetate,
18
0.1% formic acid (Buffer A), and a mixture of acetonitrile and isopropanol (7:3) containing 10
19
mM ammonium acetate, 0.1 % formic acid (Buffer B). Separation was carried out by gradient
20
elution according to the following profile: 1 min 55% B, 3 min linear gradient from 55% to 80%
21
B, 8 min linear gradient from 80% B to 85% B, and 3 min linear gradient from 85% B to 100%
22
B. After 4.5 min washing with 100% B the column was re-equilibrated with 55% B for 4.5 min.
23
The injection volume was 3 μL. Mass spectra were acquired in full-scan mode followed by
ACS Paragon Plus Environment
7
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 8 of 29
1
scheduled PRM mode. The resolution for the full scan mode was set as 75,000 (at m/z 200), the
2
AGC target at 1e6 and monitoring the m/z range 200 to 2000. Targeted MS2 scan was performed
3
at a resolution of 35 000 (AGC target: 5e5, maximum IT: 100 ms) with isolation window of 1.2
4
m/z and scheduled for 4 min around the expected RT. Stepped normalized collision energy was
5
set as 15, 25 and 30 for fragmentation. Data was acquired on the profile mode. All molecular
6
identifications were based on MS2 spectra, with a MS1 mass error of
