Toward the Rational Design of Universal Dual Polarity Matrix for MALDI Mass Spectrometry

Article abstract of DOI:10.1021/acs.analchem.0c00570

A series of novel anthranilic acid derivatives I-IV, of which COOH-NH2 (I) and COOH-NHMe (IV) are endowed with acid and base bifunctionality, were designed and synthesized for matrix-assisted laser desorption/ionization time-of-flight (MALDI-TOF) mass spectrometry applications in dual polarity molecular imaging of biological samples, particularly for lipids. The heat of protonation, deprotonation, and proton transfer reaction as well as the capability of analyzing biomolecules in both positive and negative ion modes for I-IV were systematically investigated under standard 355 nm laser excitation. The results indicate correlation between dual polarity and acid-base property. Further, COOH-NHMe (IV) showed a unique performance and was successfully applied as the matrix for MALDI-TOF mass spectrometry imaging (MSI) for studying the mouse brain. Our results demonstrate the superiority of COOH-NHMe (IV) in detecting more lipid and protein species compared to commercially available matrices. Moreover, MALDI-TOF MSI results were obtained for lipid distributions, making COOH-NHMe (IV) a potential next generation universal matrix.

Full text of DOI:10.1021/acs.analchem.0c00570

Subscriber access provided by BIU Pharmacie | Faculté de Pharmacie, Université Paris V
Article
Towards the Rational Design of Universal Dual
Polarity Matrix for MALDI Mass Spectrometry
Penghsuan Huang, Chun-Ying Huang, Ta-Chun Lin, Li-En Lin,
Ethan Yang, Chuping Lee, Cheng-Chih Hsu, and Pi-Tai Chou
Anal. Chem., Just Accepted Manuscript • DOI: 10.1021/acs.analchem.0c00570 • Publication Date (Web): 21 Apr 2020
Downloaded from pubs.acs.org on April 23, 2020
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 8
Analytical Chemistry
1
2
3
4
5
6
7
8
9
Towards the Rational Design of Universal Dual Polarity Matrix for
MALDI Mass Spectrometry
Penghsuan Huang,+† Chun-Ying Huang,+† Ta-Chun Lin,+† Li-En Lin,† Ethan Yang,†, § Chuping Lee, *†, ‡
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
Cheng-Chih Hsu, *† and Pi-Tai Chou *†
† Department of Chemistry, National Taiwan University, Taipei 10617, Taiwan (R.O.C.)
‡ Department of Applied Chemistry, National Chiayi University, Chiayi City 60004, Taiwan (R.O.C.)
§ Department of Chemistry, Université de Montréal, Montreal, Quebec, Canada H3T 1J4
ABSTRACT: A series of novel anthranilic acid derivatives I-IV, of which COOH-NH2 (I) and COOH-NHMe (IV) are endowed
with acid and base bifunctionality, were designed and synthesized for matrix-assisted laser desorption/ionization time-of-
flight (MALDI-TOF) mass spectrometry applications in dual polarity molecular imaging of biological samples, particularly
for lipids. The heat of protonation, deprotonation and proton transfer reaction as well as the capability of analyzing
biomolecules in both positive and negative ion modes for I-IV were systematically investigated under standard 355-nm
laser excitation. The results indicate correlation between dual polarity and acid-base property. Further, COOH-NHMe (IV)
showed a unique performance and was successfully applied as the matrix for MALDI-TOF mass spectrometry imaging (MSI)
for studying the mouse brain. Our results demonstrate the superiority of COOH-NHMe (IV) in detecting more lipid and
protein species compared to commercially available matrices. Moreover, MALDI-TOF MSI results were obtained for lipid
distributions, making COOH-NHMe (IV) a potential next generation universal matrix.
Lipids are essential building blocks for cells, storing
energy as well as many biological functions, such as cell
signaling.1,2 Their accumulation is highly correlated with
many diseases.3,4 Changes in lipid localization and profile
are key features of diseases such as cancer,3,4 metabolic
disorders5 and neurodegenerative disorders.6 Molecular
imaging techniques of lipid species in tissues are
important for researchers to gain insight into the
biochemical mechanisms involved and possibly provide
clinical evaluations of the disease status.5 Among these,
mass spectrometry imaging (MSI) is a powerful tool for
discerning the spatial localization of biomolecules.7–9
Matrix-assisted laser desorption/ionization time-of-flight
(MALDI-TOF) MSI is a high spatial resolution technique
capable of analyzing a wide variety of biomolecules
simultaneously10–14 and have been applied in determining
biomarkers of diseases and tracking target molecule such
as administered drugs.3,15–19 Imaging of lipid species is one
of the most widely utilized applications using MALDI-TOF
MSI.14,20–22
Lipids have diverse chemical structures and many of
them are more easily detected in one ionization polarity
than the other, positive or negative ion mode. For example,
phosphatidylcholines (PC) are typically observed as
positive ions, whereas many other lipids, such as
phosphatidic acids (PA), phosphatidylethanolamine (PE),
phosphatidylserines (PS), phosphatidylinositols (PI) and
sulfatides (ST) are usually observed as negative ions in
MALDI-TOF MSI.20 However, most commercial MALDI
matrices are only suitable for single polarity detection. For
example, 2,5-dihydroxybenzoic acid (2,5-DHB) and α-
cyano-4-hydroxybenzoic acid (CHCA) are most often used
in positive ion detection because their carboxyl group acts
as a proton donor, while 9-aminoacridine (9-AA) is favored
in negative ion detection because its amine group acts as
the proton acceptor.23, 24 A dual polarity MALDI matrix is
one approach that allows universal mass spectrometric
analysis of both positively- and negatively-charged lipid
species on one sample.
Researchers have discovered
a few dual-polarity
matrices suitable for MALDI MSI at high spatial resolution.
One early example is 1,5-diaminonaphthalene (1,5-DAN)
proposed by Thomas et al25-27 and more recently 3-
aminophthalhydrazide (luminol) proposed by Li et al.28
However, in both cases, only MSI results of lipids and small
metabolites are shown. Their effectiveness for protein and
peptide detection not demonstrated. More interestingly,
neither of the two compounds possess a carboxylic group
for enhanced ionization efficiency in positive ion mode.
To discover a truly universal dual-polarity matrix, we
used a rational design approach to determine the most
probable matrix candidates before synthesis and
experimentation. In essence, an ideal matrix for would
have: (i) a high absorption coefficient at the commercial
laser wavelength, (ii) a simple and low matrix background
below 1000 Da and (iii) the ability to observe both positive
and negative lipid ions. Based on these criteria and the
dual acid-base functionality, we strategically designed and
synthesized a series of new MALDI matrices I-IV (see
Scheme 1) containing a simple aniline derivative anchored
by a carboxyl group for COOH-NH2 (I) and COOH-NHMe
(IV) , and formyl group for CHO-NH2 (II) and CHO-NHAc
ACS Paragon Plus Environment

Analytical Chemistry
(III). Though I-IV have been synthesized previously, their
Page 2 of 8
(v/v/v) and sonicated for 15 minutes to extract the lipid
species. One milliliter of the supernatant was transferred
to a clean Eppendorf tube and concentrated to 250 µL
under vacuum. The lipid extract was stored under -20°C
until needed. Like the standard profile, the matrix and
extracts were mixed 1:1 (v/v) and 1.4 µL of the mixture
was spotted onto the sample plate and air-dried. The
remaining pellet after lipid removal was mixed with 1 mL
water/ACN/TFA 50:50:1 (v/v/v) solution and sonicated
for 2 minutes to extract the protein species34. In this case,
the supernatant was transferred to a clean Eppendorf tube
and 1 mL of the supernatant was mixed with 10 mL of one
of four saturated matrix solutions (COOH-NHMe, 2,5-DHB,
2,5-DHAP and 1,5-DAN) for analysis.
1
2
3
4
5
6
7
8
related application as a MALDI matrix as part of a
rationally designed strategy have not been proposed,
especially for II-IV. 29-32 The amphoteric property of COOH-
NH2 (I) and COOH-NHMe (IV)33 could allow the detection
of both positive and negative ions in MALDI. In this study,
we demonstrated that these dual polarity MALDI matrix,
COOH-NH2 (I) and COOH-NHMe (IV), allowed
complementary analysis of lipids and proteins, including
the positively charged and negatively charged species,
within one experiment. Moreover, this novel matrix,
COOH-NHMe (IV), enables the MSI of phospholipids with
enhanced sensitivity as demonstrated from mouse brain
sections using MALDI-TOF.
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
Tissue preparation for MSI. A mouse brain was
sectioned at 12 µm thickness using a Leica CM1950
cryostat following a previously described protocol.35 In
short, the tissue was sectioned at -20°C and thaw-mounted
onto a ITO-coated glass slide. The slides were then stored
under -80°C until needed. After desiccation for >30mins,
EXPERIMENTAL SECTION
Materials and Instruments. 2,5-DHB and 1,5-DAN
were purchased from Tokyo Chemical Industry Co., Ltd.
(Tokyo, Japan), while sinapic acid (SA), CHCA and 9-AA
were purchased from Sigma-Aldrich (St. Louis, MO, USA).
2’,5’-dihydroxyacetophenone (2,5-DHAP) was purchased
from Alfa Aesar (Ward Hill, MA, USA), and 2’,6’-
dihydroxyacetophenone (2,6-DHAP) from Acros Organics
sublimation was carried out using
a
home-built
sublimation system where 0.5mg/cm2 of matrix was
deposited onto the sample at 60 °C with a 0.7 Torr vacuum
for 10 seconds.
(Geel,
Belgium).
PC(14:0/14:0),
PE(14:0/14:0),
MALDI-TOF mass spectrometry. Both MALDI-TOF MS
and MALDI-TOF MSI were performed on a Bruker Autoflex
Speed mass spectrometer in reflectron geometry at 1 kHz
repetition rate. For profiling, mass spectra were obtained
by accumulating 1500 laser shots with the “random walk”
setting. The detector gain was set at 2016 V for positive
mode and 2012 V for negative mode. Mass spectrometry
images (MSI) were acquired at 150 µm spatial resolution
with 1000 laser shots per pixel, with the laser operating
power at 50% and a 1 kHz repetition rate. The laser spot
size was set at “medium”, ~80 µm in diameter. The
measurement order was set at Fast (from top to bottom).
The imaging results shown were normalized by total ion
counts (TIC). Image acquisition was carried out using
FlexImaging 3.0, and spectral analysis was performed with
FlexAnalysis 3.4.
PA(14:0/14:0), PS(14:0/14:0) and phosphatidylglycerol
PG(14:0/14:0) were purchased from Avanti Polar Lipids,
Inc. (Alabaster, AL, USA) Mouse brains were purchased
from BioLASCO Taiwan Co., Ltd (Taipei, Taiwan). Indium
tin oxide (ITO)-coated glass slides were purchased from
Ruilong (Taichung, Taiwan). Protein calibration I kit was
purchased from Bruker Daltonics (Bremen, Germany).
Methanol (MeOH), acetonitrile (ACN), chloroform and
ethyl acetate (EA) used as extraction solvent were
purchased from Duksan (Korea), J.T. Baker (USA) and
Macron (USA), respectively.
The Bruker Autoflex Speed mass spectrometer (Bruker
Daltonics, Bremen, Germany) equipped with a Smartbeam
II Nd:YAG laser (355 nm) was used in this study. The LC-
MS/MS system was a hybrid LTQ-Orbitrap Elite Mass
Spectrometer (Thermo Fisher Scientific, Waltham,
Massachusetts, US) couple with Vanquish™ Flex
Quaternary UHPLC. A sublimation apparatus (Singlong,
Taichung, Taiwan) and Leica CM1950 cryostat (Leica
Microsystem Inc., Wetzlar, Germany) were used in tissue
preparation.
RESULTS AND DISCUSSION
Sample preparation. A 0.1 M solution of each matrix
was prepared in acetonitrile and water (1:1, v/v).
Individual lipid solutions of PS(14:0/14:0), PE(14:0/14:0)
and PC(14:0/14:0) at 0.2 mg/mL was prepared in
chloroform and methanol (1:1, v/v). A lipid mixture of
PS(14:0/14:0),
PE(14:0/14:0),
PC(14:0/14:0),
PA(14:0/14:0) and PG(14:0/14:0) was prepared at the
same concentration and in the same solvent system. The
protein calibration I kit was dissolved in 150 µL of
deionized water to make the protein mixture. The matrix
and standard solutions were mixed 1:1 (v/v) and 1.4 µL of
each mixture was spotted onto the sample plate and left to
air dry. MALDI samples with different pH value were
evaluated in Figure S1-S2.
Scheme 1. (A) Synthesis pathway for the anthranilic acid
derivatives and X-ray structure of III and IV. Molecular structure
of (B) III and (C) IV resolved from single crystal X-ray analysis.
Tissue preparation for mouse brain extract. Mouse
brains were grounded into powder in liquid nitrogen. The
powder was then mixed with MeOH/ACN/EA 10:7:3
ACS Paragon Plus Environment

Page 3 of 8
Analytical Chemistry
The synthetic routes of the title compounds are depicted
For COOH-NH2 (I) and COOH-NHMe (IV), the carboxylic
and amino groups that serve as proton donor and proton
acceptor, respectively, are expected to effectively produce
positive and negative ions. CHO-NH2(II) and CHO-NHAc
(III), lacking the potent proton donor site, were compared
with COOH-NH2 (I) and COOH-NHMe (IV) in terms of their
dual polarity properties. These compounds were rationally
designed for the carboxylic/carbonyl group and amino
derivatives to be in the ortho position in order to form an
intramolecular hydrogen bond (H-bond) between the
carboxylic/carbonyl and amino hydrogen. This strategy
has two major characteristics. Firstly, the intramolecular
H-bond induced -conjugation, which decreases the
HOMO-LUMO energy gap to reach the absorption
maximum suitable for commercial laser excitation (third
harmonic Nd:YAG laser, 355 nm).38,39 In addition, the
formation of intramolecular H-bond is expected to reduce
the intermolecular H-bond interaction and hence to
increase the efficiency of the desorption.
Figure 1A shows the absorption spectra of compounds
I-IV in dichloromethane (DCM), which all exhibit the
lowest lying absorption bands in the range of 300 to 400
nm. Nevertheless, appreciable absorption extinction of >
2000 cm-1 M-1 was observed for all I-IV at 355 nm of the
third harmonic Nd:YAG laser, the most popular
commercial excitation source for MALDI.29,30 Among the
title compounds, COOH-NHMe (IV) exhibits the highest
extinction coefficient at 355 nm. It is also worth noting that
the formation of intramolecular H-bond is critical to
enhance the charge transfer character and hence the red
shift of the absorption. For example, anchoring carboxyl
and amino groups in meta- and para- positions with no
intramolecular H-bond formation causes the blue shift of
the absorption with peak wavelength being far blue shifted
from 355 nm (See Table S1 for the detailed information of
the references).40–42 In fact, aside from I-IV, other
derivatives have also been synthesized (see compounds V-
IX in SI); however, those compounds show either
unsuitable absorption band (V-VIII) or low melting points
(IX) and hence were not investigated further in this study.
Figure 1B shows the absorption and emission spectra of
matrix I-IV measured in solid state. The different
absorption pattern of matrix I-IV between solution and
solid phase suggests slightly intermolecular interaction
existing in solid state.
1
in Scheme 1A. Commercially available COOH-NH2 (I) was
the initial reactant for the synthesis. COOH-NH2 (I) is first
reduced by lithium aluminum hydride, followed by an
oxidation reaction with MnO2, yielding aldehyde CHO-
NH2(II). CHO-NH2 (II) was then strategically modified into
acetamide derivative CHO-NHAc (III). Meanwhile, the
amino group of COOH-NH2 (I) is also functionalized so that
one of the amino hydrogen has been replaced with alkyl
substituent (-CH3) to obtain derivative COOH-NHMe (IV).
Notably, the reactions adopted in this study are facile and
produces title molecules with good yields (> 50%).
Detailed synthetic procedures and characterization of
these compounds are provided in the experimental section
of the Supporting Information (SI). The crystal structures
of COOH-NH2 (I) and COOH-NHMe (IV) have been resolved
in the literature; however, for readers’ convenience the
prerequisite structural information of I and IV was shown
in this study, as well.36,37 Scheme 1B and 1C depict the X-
ray structure of CHO-NHAc (III) and COOH-NHMe (IV) (see
Figure S3 for COOH-NH2, I), respectively, both of which
are in a planar configuration. Although the N-H hydrogen
atom is artificially added in solving the X-ray structure, the
N-H---O=C intramolecular H-bond formation can be
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
supported by the H NMR measurement, in which COOH-
NH2 (I) in DMSO-d6 and II-IV in benzene-d6 show
downfield shift of the N-H proton in the range of 5.78-
11.03 ppm (see Figures S4-7 in SI).
We then used I-IV as the matrix and PS(14:0/14:0) as
the analyte to demonstrate the 355-nm laser desorption
property. COOH-NHMe (IV) clearly shows the highest ion
signal (see Figure S8) due to its strongest absorption at
355 nm among I-IV (Table S2), suggesting that the ion
efficiency is highly related to the extinction coefficient.
Moreover, the characteristic signals for PC(14:0/14:0),
PS(14:0/14:0) and PE(14:0/14:0) were most strongly
detected by COOH-NH2 (I) and COOH-NHMe (IV) (Figure
S9). In fact, only COOH-NH2 (I) and COOH-NHMe (IV)
yielded signals for PS(14:0/14:0), while no signal could be
observed when applying CHO-NH2(II) and CHO-NHAc (III)
as the matrix (see Figure S8).
Figure 1. (A) Absorption spectra of matrix I-IV measured in
dichloromethane (DCM). (B) Absorption (solid line) and
emission spectra (dash line) of matrix I-IV measured in solid
state. ex = 340 nm for COOH-NH2 (I), CHO-NH2 (II), COOH-
NHMe (IV) and ex = 320 nm for CHO-NHAc (III).
These experimental results do not align with the
extinction coefficient of the matrix: COOH-NHMe (IV) >
CHO-NH2 (II) > CHO-NHAc (III) > COOH-NH2 (I) (see Table
S2). This is not surprising because the extinction
ACS Paragon Plus Environment

Analytical Chemistry
coefficient merely accounts for one of multiple factors for
Page 4 of 8
comparison, compounds I-IV, having a simple aromatic
ring endowed with acid-base bifunctionality are great
candidates of MALDI matrices for dual polarity detection.
Figure 2 compares the signal intensities of lipid and
protein standards detectable with the COOH-NH2 (I) and
COOH-NHMe (IV) (commercially available) and commonly
used matrices. Among them, lipids analyzed using COOH-
NHMe (IV) as the matrix (Figure 2) exhibited the highest
ion signal compared to other compounds, even though the
extinction coefficients of all studied matrices were within
the same order of magnitude. The most comparative
commercially available matrix was 1,5-DAN for lipid
detection, though its protein signals were less impressive
(Figure 2 and 3). Additional experiments were performed
to analyze proteins extracted from mouse brains in both
positive and negative ion mode, with results from triplicate
analyses shown in Figure S10-S12.
1
2
3
4
5
6
7
8
analyte desorption/ionization. Other factors, such as the
difference in fluorescence lifetime among I-IV, also need be
considered. Since the applied 355-nm laser has a pulse
duration time of ~5 ns, matrices with longer fluorescence
lifetime such as COOH-NH2 (I) and COOH-NHMe (IV) (see
Table S3) are expected to have a higher chance for
multiphoton events. More importantly, matrix interaction
with analyte is important when it comes to the MALDI
process. Commonly, positive ions are observed as
protonated, sodiated or potassiated ions ([M+H]+, [M+Na]+
and [M+K]+), where negative ions are observed as
deprotonated ions ([M-H]-).1,4,27,38,39,43–45 Therefore, in order
to quantify the feasibility of the synthesized matrix,
theoretical calculation based on density function theory
(DFT) was conducted to acquire the corresponding proton
affinity (PA) and deprotonation energy (DE), whose values
can provide insights into the tendency towards producing
[M+H]+ and [M-H]-, respectively. Among the synthesized
matrices, COOH-NHMe (IV) showed relatively high PA and
low DE (Table S2),46 and thus is expected to be a suitable
matrix in both positive and negative modes. Importantly,
the values of PA and DE of COOH-NHMe (IV) are
comparable to those of commercial matrices that are solely
suitable in positive or negative mode (see Table S2),
showing its high potential in practical matrix applications.
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
Figure 3. Mass spectra of the brain protein extracts obtained
from samples prepared by COOH-NHMe (IV) and commonly
used commercial matrices for protein analysis, 2,5-DHB, 2,5-
DHAP and 1,5-DAN in positive ion mode.
As shown in Figure 3 and S13, a great number of
protein species were observed with COOH-NHMe (IV) and
2,5-DHAP relative to 2,5-DHB. Moreover, relatively few
protein species can be observed from 1,5-DAN, implying
that this previously-reported dual polarity matrix may
have limited mass coverage for biomolecules. MSI of
protein distribution in Mouse brains section was
performed by using COOH-NHMe (IV) (Figure S14).
We further demonstrated the superiority of COOH-
NHMe (IV) in terms of the molecular coverage and dual-
polarity compatibility in the MALDI-TOF measurement on
mouse brains lipid extract. The comparison between
COOH-NHMe (IV) and two of the most commonly used
matrices for lipid analysis, 2,5-DHB (standard matrix for
positive ion mode detection) and 9-AA (standard matrix
for negative ion mode detection), is shown in Figure 4.
Remarkably, COOH-NHMe (IV) showed a rich lipid profile
in both positive and negative ion mode, whereas 2,5-DHB
and 9-AA showed only limited coverage due to their single-
polarity character.24 Moreover, the comparison between
COOH-NHMe (IV) and early reported matrices for dual
polarity has been made. As a result, the performance of
COOH-NHMe (IV) is as good as 2,6-DHAP and 1,5-DAN in
Figure 2. Comparison of the relative signal intensities of lipid
species and proteins species when using the synthesized
COOH-NH2 (I), COOH-NHMe (IV) or some of the most
representative commercially available matrices. The laser
intensities used for different matrices were optimized to
achieve the most optimal signal. The analytes written in red
indicates signals were obtained in positive ion mode and blue
in negative ion mode. COOH-NHMe (IV) had superior
performance in lipid analysis and was also capable of
analyzing proteins in both positive and negative mode.
To gain more insight into the efficacy of our synthesized
compounds, we chose some of the most representative
commercially available matrices to compare their
performances. Figure S9 shows the relative signal
intensities of PC(14:0/14:0), PS(14:0/14:0) and
PE(14:0/14:0) from the typical commercially available
matrices. 2,5-DHB and CHCA provide lower ion abundance
of negative ion relative to positive ion, and are commonly
used in positive ion detection.47 On the other hand, 9-AA
and 1,5-DAN are suitable for negative ion detection.27,48 In
ACS Paragon Plus Environment

Page 5 of 8
Analytical Chemistry
both positive and negative ion mode (Figure S15). Note
that due to the limited vacuum stability, 2,6-DHAP is not
commonly applied for vacuum MALDI MSI, but instead
used for atmospheric pressure (AP)-MALDI.49,50
phosphatidylethanolamines (PE), phosphatidic acids (PA),
phosphatidylserines (PS), phosphatidylinositols (PI),
sulfatides (ST) and hydroxylated sulfatides (ST(OH)). LC-
MS/MS was conducted on the sample mouse brain extract
to and the species detected were confirmed by
LipidSearch™ (Thermo Fisher Scientific). Details of the
assignments of mass spectra are listed in Table S4 and S5.
The results obtained by using COOH-NHMe (IV) as a matrix
revealed more lipid species than using 2,5-DHB and 9-AA.
For example, [PE(40:6)-H]-, [PI(36:4)-H]- and [PI(40:6)-H]-
were observed when COOH-NHMe (IV) was used as the
matrix but missing when using 2,5-DHB. Also, the spatial
distribution of the three species only detectable by COOH-
MeOH, spatial distributions of [PE(40:6)-H]-, [PI(36:4)-H]-
and [PI(40:6)-H]- are shown in Figure 5. Distinct
distributions of lipids were thus obtained from COOH-
NHMe (IV).
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
Figure 4. Mass spectra of the brain lipid extract obtained from
samples prepared by COOH-NHMe (IV) and commonly used
commercial matrices for lipid analysis, 9-AA and 2,5-DHB, in
(A) positive and (B) negative ion mode respectively. COOH-
NHMe (IV) outperformed the widely used commercial
matrices for lipid analysis in both positive and negative mode.
The ions marked in red indicate that the ions observed only
using COOH-NHMe (IV) as matrix. The ions observed using 9-
AA as matrix but not 2,5-DHB were marked in blue.
The positive (deprotonation of the matrix) and negative
(protonated of the matrix) can be explained by their
proton affinity and deprotonation energy (Table S2). The
deprotonation energy of 2,5-DHB is relatively low, so the
proton transfers easily from matrix to analyte, producing
protonated analyte ion. On the contrary, the proton affinity
of 9-AA is relatively high and the proton easily transfers
from analyte to matrix, yielding deprotonated analyte ions.
Figure 4A shows the signals of the positive ions, [M+H]+,
[M+Na]+ and [M+K]+. In Figure 4B, signals of negative ion
Figure 5. MSI of mouse brains sections sublimated with
COOH-NHMe (IV) and 9AA in negative ion mode. The color
scale for each ion species were under the same intensity
threshold. The lipid species scan revealed the deprotonated
ion at m/z 790.5, 857.5 and 909.6, corresponding to
[PE(40:6)-H]-, [PI(36:4)-H]- and [PI(40:6)-H]-, respectively.
represent
[M-H]-
correspond
to
Figure 6. MSI of mouse brain sections sublimated with COOH-NHMe (IV) in positive and negative ion mode at 150 µm
spatial resolution. We observed glycerophospholipids in both (A) positive and (B) negative ion modes, (C)
ACS Paragon Plus Environment
phosphosphingolipids in positive ion mode, and (D) glycosphingolipids in negative ion mode. Both positively charged and
negatively charged lipid species were observed using common sublimation method.

Analytical Chemistry
To demonstrate the capability of COOH-NHMe (IV) in
Page 6 of 8
resolved in the negative ion mode. As such, similar to the
analysis of the brain lipid extract, untargeted molecular
imaging of lipids can be successfully obtained by MALDI-
TOF MSI with both positive ions and negative ions
simultaneously within one experiment. Also, this matrix is
potentially capable of performing high spatial resolution
MSI, although most of our tests, due to the limit of facility,
are limited at 80 µm within 2 hours under vacuum. Given
that sublimation produces sub-nanometer crystals, with a
more advanced system such as the Bruker rapifleX
equipped with a 10 kHz laser, it would be possible to
accomplish 5 µm MSI on an entire mouse brain section
1
2
3
4
5
6
7
8
mapping the lipid distribution, we applied it as the matrix
for MALDI-TOF MSI on a coronal mouse brain section
(Figure 6 and Figure S15). In the positive ion mode, we
observed PC and sphingomyelin (SM) species, in which PC
(36:1) was largely observed in the white matter of the
brain, while PC (32:0) was revealed in the gray matter.
These results were in agreement with the previous report
using the other universal matrices.51-53 In the negative ion
mode, we observed several glycerophospholipids, e.g. PA
(34:1), PE (36:1), PS (36:0), PI (38:4) in the gray matter
and the dimethyl-phosphatidylethanolamine DMPE
(34:1p), PG (40:6) in the white matter. ST was also
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
within
2
hours.
Notes The authors declare no competing financial
interest.
CONCLUSIONS
In this work, we synthesized a series of new anthranilic
acid derivatives I-IV, of which COOH-NH2 (I) and COOH-
NHMe (IV) are endowed with acid/base bifunctional
properties and have demonstrated their potential for dual
polarity detection in MALDI-TOF mass spectrometry. The
dual polarity character can be rationalized by its proton
affinity and deprotonation energy. Of them, COOH-NHMe
(IV) exhibits an outstanding efficiency in generating ions of
a wide variety of biomolecules, including lipids and
proteins, upon laser excitation. High quality molecular
imaging of lipids in mouse brains sections was readily
obtained by a commercial MALDI-TOF mass spectrometer
using our newly designed organic matrix. As many new
candidate compounds have been recently proposed as
suitable dual polarity matirces,27, 54-56 we aim to compare
our proposed matrix with these compounds in the near
future. We believe that via a more thorough and systematic
investigation that integrates the photophysical and
chemical insights, a universal matrix for MALDI mass
spectrometry shall soon be realized.
ACKNOWLEDGMENT
This research was supported by Ministry of Science and
Technology (MOST), R.O.C. (Grant nos.: 109-2636-M-002-005-
), and Center for Emerging Materials and Advanced Devices,
National Taiwan University. The authors acknowledged the
mass spectrometry support from NTU Consortia of Key
Technologies.
REFERENCES
(1) Berry, K. A. Z.; Hankin, J. A.; Barkley, R. M.; Spraggins, J. M.;
Caprioli, R. M.; Murphy, R. C. Chem. Rev. 2011, 111, 6491–6512.
(2) McDonnell, L. A.; Piersma, S. R.; MaartenAltelaar, A. F.; Mize, T.
H.; Luxembourg, S. L.; Verhaert, P. D. E. M.; van Minnen, J.; Heeren,
R. M. A. J. Mass Spectrom. JMS 2005, 40, 160–168.
(3) Eberlin, L. S.; Norton, I.; Dill, A. L.; Golby, A. J.; Ligon, K. L.;
Santagata, S.; Cooks, R. G.; Agar, N. Y. R. Cancer Res. 2012, 72, 645–
654.
(4) Guo, S.; Wang, Y.; Zhou, D.; Li, Z. Sci. Rep. 2014, 4, 5959.
(5) Mehlem, A.; Hagberg, C. E.; Muhl, L.; Eriksson, U.; Falkevall, A.
Nat. Protoc. 2013, 8, 1149–1154.
(6) Yadav, R. S.; Tiwari, N. K. Mol. Neurobiol. 2014, 50, 168–176.
(7) Norris, J. L.; Caprioli, R. M. Chem. Rev. 2013, 113, 2309–2342.
(8) Hsu, C.-C.; Chou, P.-T.; Zare, R. N. Anal. Chem. 2015, 87, 11171–
Supporting Information
The supporting information is available free of charge via
the Internet. Synthetic methods for the compounds, NMR,
and computational details are provided.
11175.
(9) Hsu, C.-C.; Dorrestein, P. C. Curr. Opin. Biotechnol. 2015, 31,
24–34.
(10) Caprioli, R. M.; Farmer, T. B.; Gile, J. Anal. Chem. 1997, 69,
Crystallographic data. The X-ray crystal structure for the
anthranilic acid derivatives reported in this article have been
submitted to the Cambridge Crystallographic Data Centre
(CCDC), with the deposition number CCDC 1948704 (COOH-
NH2, I), 1947215 (CHO-NHAc, III), 1947199 (COOH-NHMe,
IV), 1947202 (COOH-NHAc, V), 1947210 (COOH-NHCOCF3,
VI), 1947216 (CHO-NHCOCF3, VII) and 1947201 (COOH-
NMe2, VIII), respectively. These data can be obtained free of
charge from The Cambridge Crystallographic Data Centre via
www.ccdc.cam.ac.
4751–4760.
(11) Lanni, E. J.; Rubakhin, S. S.; Sweedler, J. V. J. Neurochem.
2013, 124, 581–583.
(12) Bednařík, A.; Bölsker, S.; Soltwisch, J.; Dreisewerd, K. Angew.
Chem. Int. Ed Engl. 2018, 57, 12092–12096.
(13) Paine, M. R. L.; Poad, B. L. J.; Eijkel, G. B.; Marshall, D. L.;
Blanksby, S. J.; Heeren, R. M. A.; Ellis, S. R. Angew. Chem. Int. Ed.
2018, 57, 10530–10534.
(14) Neumann, E. K.; Comi, T. J.; Rubakhin, S. S.; Sweedler, J. V.
Angew. Chem. Int. Ed. 2019, 58, 5910–5914.
(15) McDonnell, L. A.; Corthals, G. L.; Willems, S. M.; van
Remoortere, A.; van Zeijl, R. J. M.; Deelder, A. M. J. Proteomics
2010, 73, 1921–1944.
AUTHOR INFORMATION
Corresponding Author
(16) McDonnell, L. A.; Heeren, R. M. A. Mass Spectrom. Rev. 2007,
26, 606–643.
(17) Mirnezami, R.; Spagou, K.; Vorkas, P. A.; Lewis, M. R.; Kinross,
J.; Want, E.; Shion, H.; Goldin, R. D.; Darzi, A.; Takats, Z.; et al. Mol.
Oncol. 2014, 8, 39–49.
(18) Wang, F.; Yang, P.; Choi, J.; Antovski, P.; Zhu, Y.; Xu, X.; Kuo,
T.-H.; Lin, L.-E.; Kim, D. N. H.; Huang, P.-C.; et al. ACS Nano 2018,
12, 6851–6859.
* E-mail: ccrhsu@ntu.edu.tw
* E-mail: chop@ntu.edu.tw
* E-mail: cplee@mail.ncyu.edu.tw
Author Contributions
+ These authors contributed equally.
ACS Paragon Plus Environment

Page 7 of 8
Analytical Chemistry
(19) Soltwisch, J.; Kettling, H.; Vens-Cappell, S.; Wiegelmann, M.;
(39) Karas, M.; Bachmann, D.; Bahr, U.; Hillenkamp, F. Int. J. Mass
Spectrom. Ion Process. 1987, 78, 53–68.
(40) Hmida, F.; Ayed, B.; Haddad, A. J. Mol. Struct. 2016, 1116,
218–225.
Müthing, J.; Dreisewerd, K. Science 2015, 348, 211–215.
(20) Murphy, R. C.; Hankin, J. A.; Barkley, R. M. J. Lipid Res. 2009,
50, S317–S322.
1
2
3
4
5
6
7
8
9
(21) Jackson, S. N.; Ugarov, M.; Egan, T.; Post, J. D.; Langlais, D.;
Schultz, J. A.; Woods, A. S. J. Mass Spectrom. 2007, 42, 1093–1098.
(22) Sparvero, L. J.; Amoscato, A. A.; Dixon, C. E.; Long, J. B.;
Kochanek, P. M.; Pitt, B. R.; Bayır, H.; Kagan, V. E. Chem. Phys.
Lipids 2012, 165, 545–562.
(23) Demeure, K.; Quinton, L.; Gabelica, V.; De Pauw, E. Anal.
Chem. 2007, 79, 8678–8685.
(24) Leopold, J.; Popkova, Y.; Engel, K. M.; Schiller, J. Biomolecules
2018, 8, 173.
(25) Molin, L.; Seraglia, R.; Dani, F. R.; Moneti, G.; Traldi, P. Rapid
Commun. Mass Spectrom. 2011, 25, 3091–3096.
(26) Thomas, A.; Déglon, J.; Lenglet, S.; Mach, F.; Mangin, P.;
Wolfender, J.-L.; Steffens, S.; Staub, C. Anal. Chem. 2010, 82, 6687–
6694.
(27) Thomas, A.; Charbonneau, J. L.; Fournaise, E.; Chaurand, P.
Anal. Chem. 2012, 84, 2048–2054.
(28) Li, B.; Sun, R.; Gordon, A.; Ge, J.; Zhang, Y.; Li, P.; Yang H.,
Anal. Chem. 2019, 91, 8221-8228.
(29) Jia, M.-Q.; You, S.-L. ACS Catal. 2013, 3, 622-624.
(30) Chiurato, M.; Boulahjar, R.; Routier, S.; Troin, Y.; Guillaumet,
G., Tetrahedron, 2010, 66, 4647-4653.
(31) Geddis, S. M.; Hagerman, C. E.; Galloway, W. R. J. D.; Sore, H.
F.; Goodman, J. M. and Spring D. R. Beilstein J. Org. Chem. 2017, 13,
323-328.
(32) Aitken, R. A.; Murray, L. J. Org. Chem. 2008, 73, 9781-9783.
(33) Zapała, L.; Woz´nicka, E; Kalembkiewicz, J. J. Solution Chem.
2014, 43, 1167–1183.
(34) Böhme, K.; Fernández-No, I. C.; Barros-Velázquez, J.; Gallardo,
J. M.; Cañas, B.; Calo-Mata, P. Anal. Methods 2010, 2, 1941–1947.
(35) Lin, L.-E.; Su, P.-R.; Wu, H.-Y.; Hsu, C.-C. J. Am. Soc. Mass
Spectrom. 2018, 29, 796–799.
(36) Hardy, G. E.; Kaska, W. C.; Chandra, B. P.; Zink, J. I. J. Am.
Chem. Soc. 1981, 103, 1074-1079.
(37) Dhaneshwar N. N.; Pant L. M. Acta Cryst. 1972, B28, 647-649.
(38) Karas, Michael.; Bachmann, Doris.; Hillenkamp, Franz. Anal.
Chem. 1985, 57, 2935–2939.
(41) Boroff, J. A.; Matesich, Z. D.; Stuetzer, D. C.; Sobeck, S. J. S. J.
Photochem. Photobiol. Chem. 2015, C, 60–66.
(42) Thayer, M. P.; McGuire, C.; Stennett, E. M. S.; Lockhart, M. K.;
Canache, D.; Novak, M.; Schmidtke, S. J. Spectrochim. Acta. A. Mol.
Biomol. Spectrosc. 2011, 84, 227–232.
(43) Lu, I.-C.; Lee, C.; Chen, H.-Y.; Lin, H.-Y.; Hung, S.-W.; Dyakov, Y.
A.; Hsu, K.-T.; Liao, C.-Y.; Lee, Y.-Y.; Tseng, C.-M.; et al. J. Phys.
Chem. B 2014, 118, 4132–4139.
(44) Lu, I.-C.; Lee, C.; Lee, Y.-T.; Ni, C.-K. Annu. Rev. Anal. Chem.
2015, 8, 21–39.
(45) Mirabelli, M. F.; Zenobi, R. J. Am. Soc. Mass Spectrom. 2017,
28, 1676–1686.
(46) Lee, C.; Inutan, E. D.; Chen, J. L.; Mukeku, M. M.; Weidner, S.
M.; Trimpin, S.; Ni, C.-K. Rapid Commun. Mass Spectrom. [Online]
2019.
(47) Schröter, J.; Fülöp, A.; Hopf, C.; Schiller, J. Anal. Bioanal. Chem.
2018, 410, 2437–2447.
(48) Fuchs, B.; Bischoff, A.; Süß, R.; Teuber, K.; Schürenberg, M.;
Suckau, D.; Schiller, J. Anal. Bioanal. Chem. 2009, 395, 2479.
(49) Wang, H.-Y. J.; Jackson, S. N.; Woods, A. S. J. Am. Soc. Mass
Spectrom. 2007, 18(3), 567–577.
(50) Sakai,M.; Martinez-Arguelles, D. B.; Patterson, N. H.;
Chaurand, P.; Papadopoulos, V. PLoS One. 2015, 10(3), e0120670.
(51) Li, S.; Zhang, Y.; Liu, J.; Han, J.; Guan, M.; Yang, H.; Lin, Y.;
Xiong, S.; Zhao, Z. E Sci. Rep. 2016, 6, 37903.
(52) Wiseman, J. M.; Ifa, D. R.; Song, Q.; Cooks, R. G. Angew. Chem.
Int. Ed. 2006, 45, 7188-7192.
(51) Eberlin, L. S.; Ifa, D. R.; Wu, C.; Cooks, R. G. Angew. Chem. Int.
Ed. 2010, 49, 873–876.
(52) Wu, Q.; Chu, J. L.; Rubakhin, S. S.; Gillette, M. U.; Sweedler, J.
V. Chem. Sci. 2017, 8, 3926−3938.
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
(53) Horatz, K.; Giampa, M.; Karpov, Y.; Sahre, K.; Bednarz, H.;
̀
al. J. Am. Chem. Soc. 2018, 140, 11416-11423.
(54) Ellis, S. R.; Cappell, J.; Potočnik, N. O.; Balluff, B.; Hamaide, J.;
Van der Linden, A.; Heeren, R. M. A. Analyst 2016, 141, 3832-3841.
ACS Paragon Plus Environment

Analytical Chemistry
Page 8 of 8
1
2
3
Insert Table of Contents artwork here
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
8
ACS Paragon Plus Environment