Oxidation-assisted structural elucidation of compounds containing a tertiary amine side chain using liquid chromatography mass spectrometry

Article abstract of DOI:10.1002/jms.4081

A novel analytical technique for the structural elucidation of compounds bearing a tertiary amine side chain via “in vial” instantaneous oxidation and liquid chromatography mass spectrometry (LC-MS) was developed. A series of lidocaine homologs and benzimidazole derivatives with a major/single amine representative base peak in both their EI-MS and ESI-MS/MS spectra were subjected to oxidation by a 0.1% solution of hydrogen peroxide (including several ¹⁶O/¹⁸O exchange experiments), followed by LC-ESI-MS/MS analysis. The N-oxide counterparts promoted extensive fragmentation with complete coverage of all parts of the molecule, enabling detailed structural elucidation and unambiguous identification of the unoxidized analytes at low nanogram per milliliter levels.

Full text of DOI:10.1002/jms.4081

Oxidation-assisted structural elucidation of compounds containing a tertiary amine side
chain using liquid chromatography-mass spectrometry
Short title:
Structure elucidation via oxidation and LC-MS
Tamar Shamai Yamin, Hagit Prihed, Avital Shifrovitch, Shai Dagan, Avi Weissberg*
*
Corresponding author:
Avi Weissberg, Department of Analytical Chemistry, Israel Institute for Biological Research
IIBR) P.O.B. 19, Ness Ziona, Israel, Email: aviwe@iibr.gov.il, Tel.: 972-8-9385852, Fax:
72-8-9381688
(
9
This article has been accepted for publication and undergone full peer review but has not
been through the copyediting, typesetting, pagination and proofreading process which may
lead to differences between this version and the Version of Record. Please cite this article as
doi: 10.1002/jms.4081
This article is protected by copyright. All rights reserved.

Abstract
A novel analytical technique for the structural elucidation of compounds bearing a tertiary
amine side chain via "in vial" instantaneous oxidation and liquid chromatography-mass
spectrometry (LC-MS) was developed. A series of lidocaine homologues and benzimidazole
derivatives with a major/single amine-representative base peak in both their EI-MS and ESI-
MS/MS spectra were subjected to oxidation by a 0.1% solution of hydrogen peroxide
1
6
18
)
including several O/ O exchange experiments(, followed by LC-ESI-MS/MS analysis.
The N-oxide counterparts promoted extensive fragmentation patterns with complete coverage
of all parts of the molecule, enabling detailed structural elucidation and unambiguous
identification of the un-oxidized analytes at low ng/ml levels.
Keywords: Structural elucidation, Tertiary amine side chain, Oxidation, LC-MS/MS, Water
analysis
This article is protected by copyright. All rights reserved.

1
.
Introduction
Mass spectrometry is commonly applied to qualitatively and quantitatively profile
small molecules. The increased availability of high-resolution mass spectrometry (HR-MS)
for chemical analysis has dramatically improved the detection and identification of
compounds in the fields of metabolomics, drug discovery, forensics, and environmental
1
monitoring. Modern mass spectrometers provide accurate measurements of the mass-to-
charge ratios of ions with errors as low as sub-ppm depending on the instruments and
2
calibration. In addition to the mass accuracy, the ability of a mass spectrometer to faithfully
measure the isotopic distribution of an ion is also important to determine the elemental
composition of a molecule. Even a high mass accuracy and isotope distribution are not
sufficient to determine the unique chemical structure of a molecule based on the mass value
3
alone. Evaluation of tandem MS (MS/MS) fragmentation is useful, however not always
4
sufficient to propose structures. For structure elucidation, which is the main bottleneck in the
5
,6
identification of unknown compounds, MS/MS spectra with diagnostic and informative
product ions representing all parts of the molecule are needed. The most common approach to
determining the structure of unknown small organic molecules is searching the molecular
formula against databases, such as Chemindex, NIST library or even Scifinder, for possible
7
structures. Structure-fragmentation relations have been widely studied by Niessen, Levsen et
8
9
10
al., Holcapek et al., and our group. Various algorithms that evaluate the structures of
organic compounds and predict mass spectral fragments based on these rules have been
developed (Mass Frontier, Mass Fragmenter, Advanced Chemistry Development, EPIC).
However, difficulties arise in cases where the ESI-MS/MS spectrum is uninformative or
1
1-14
exhibits poor quality, leading to insufficient evidence to match the ESI-MS/MS spectra to the
proposed structure. Several compounds containing an amine side chain, particularly tertiary
This article is protected by copyright. All rights reserved.

amines, present information-poor mass spectra with product ions mostly/only representing
1
5-18
the amine-containing residue.
This lack of information is attributed to the strong tendency
of the amine side chain to stabilize the positive charge and leads to unavoidable ambiguity in
the identification process. In order to address this challenge, we have recently demonstrated a
method for the structural determination of benzimidazoles containing a tertiary amine side
chain employing multi-stage tandem MS, in combination with fragmentation pattern
1
6
matching to an analogue utilizing LC-MS analysis.
The objective of the current study was to develop a novel analytical technique based
on oxidation for the structural elucidation of compounds containing tertiary amine side chains
in aqueous solutions based on LC-MS analysis. The outcome of this study is discussed as a
helpful tool when encountering such challenges. To the best of our knowledge, oxidation-
assisted structure elucidation is demonstrated here for the first time.
2
.
Experimental
2
.1. Materials and reagents
Lidocaine hydrochloride monohydrate, L5647 (A1); 6-chloro-2-diethylamino-ortho-
acetotoluidide hydrochloride, S423599 (A2); N-(4-tert-bu-2,6-dimethyl-ph)-2-
diethylaminoacetamide, hydrochloride, S243205 (A3); ropivacaine hydrochloride
monohydrate, R0283 (A4); bupivacaine hydrochloride, SML1092 (A5); amino-1-(2-(MeO-
ph)-2-oxo-et)-3-(2-diethylamino-et)-benzimidazol-1-ium bromide, S20401 (A7); 2-amino-1-
(2-(4-Cl-ph)-2-oxo-et)-3-(2-diethylamino-et)-benzimidazole-1-ium bromide, S20398 (A8);
amino-1-(2-(di-MeO-ph)-2-oxo-et)-3-(2-morpholin-4-yl-et)-benzimidazol-1-ium bromide,
S20320 (A9); 2-amino-1-(2-(4-Br-ph)-2-oxo-et)-3-(2-morpholin-4-yl-et)-benzimidazol-1-ium
bromide, S20312 (A10), ammonium formate, 70221; 50 wt. % hydrogen peroxide solution in
This article is protected by copyright. All rights reserved.

1
8
18
H O, stabilized, 516813; and hydrogen peroxide O solution, 2.2% in H O, 90% O atoms,
2
2
2
6
09978 were obtained from Sigma-Aldrich (St. Louis, MO, USA). Etonitazene (A6) was
purchased from Tzamal D-chem Company (Israel). Water (LC-MS grade) and methanol (LC-
MS grade) were obtained from Biolab Company (Israel).
2
.2. Sample preparation
Stock solutions were prepared in water (1 mg/ml) and diluted with water to
concentrations in the range of 1 ng/ml- 1 µg/ml.
Oxidation of A1-A10
Hydrogen peroxide (2 µl, 50% in water) was added directly into an injection vial
containing 998 µl of water with 0.001-1 µg of A1-A10 to obtain a 0.1% solution of hydrogen
peroxide. The analysis vial was stirred for 1 min prior to the LC-MS analysis.
1
6
18
O/ O oxygen isotope exchange
1
6
18
The O/ O exchange experiments were performed by exchanging the H O for
2
2
18
H O . The experiment was then run via the same method described earlier.
2
2
2
.3. Instrumentation
An HPLC-QTRAP-MS system and an HPLC-Q-Exactive Plus Orbitrap MS system
were used in this study and data from both instruments is presented.
The analytes were separated using an Agilent 1290 high-performance LC system
(
Palo Alto, CA, USA) containing a 1290 infinity binary pump with a jet weaver V35 mixer, a
290 infinity autosampler and a 1290 infinity thermostatted column compartment (TCC).
HPLC conditions: gradient elution was performed on a reverse phase separation column
Gemini C18, 3.0 µm, 150 mm, 2.1 mm ID by Phenomenex, Switzerland) with a flow of 0.3
1
(
ml/min. The column was maintained at 40 ℃ in all the experiments. The gradient (solvent A,
This article is protected by copyright. All rights reserved.

water with 5% MeOH containing 1 mM ammonium formate; solvent B, MeOH containing 1
mM ammonium formate) was as follows: 0-10 min, linear increase from 0% B to 95% B,
hold of 6 min at 95% B, followed by a 4 min equilibration period at 100% A.
MS and MS/MS experiments were carried out using two mass spectrometers in
positive-ion mode: an Applied Biosystems 5500 QTRAP LIT quadrupole MS (AB SCIEX,
Foster City, CA, USA) with ―Analyst‖ software (version 1.6) equipped with a Turbo V ESI
source and a Thermo Scientific Q-Exactive Plus Orbitrap MS (Thermo Fisher Scientific,
Bremen, Germany) operated with a heated-ESI (HESI) source.
The QTRAP ESI operating conditions were as follows: gas 1, air (40 psi); gas 2, air
(70 psi); ion spray voltage, 4500 V; ion source temperature, 600 ℃; curtain gas, nitrogen (35
psi). MS/MS experiments: The settings for the enhanced product ion (EPI) scans were as
follows: collision gas ‗high‘, and the collision energy was between 20 and 60 V. The fixed
LIT fill time was set to 50 ms.
The orbitrap HESI operating parameters were as follows: electrospray voltage, 1.25
kV; sheath gas flow rate, 45 (arbitrary units); auxiliary gas flow rate, 10 (arbitrary units);
sweep gas flow rate, 2 (arbitrary units); aux gas heater temperature, 400 ℃; capillary
temperature, 275 ℃. The instrument was calibrated using a positive ESI calibration solution
prepared according to the operating manual. All samples were analyzed using two alternating
6
experiment types: full scan mode from m/z 60-900 at a resolving power of 70,000 and 1x10
automatic gain control (AGC) target, and a data independent acquisition (DIA) experiment
5
with an inclusion list at a resolving power of 35,000 and 5x10 AGC target. The normalized
collision energy (NCE) was set between 20 and 100 V.
This article is protected by copyright. All rights reserved.

3
.
Results and discussion
3
.1. Chemical derivatization
19,20
Because N-oxides undergo thermal deoxygenation during a GC-MS analysis,
we
focused on an LC-MS-based method. To address the problem of information-poor MS/MS
spectra, we sought a derivatizing agent that would react with tertiary amines, reduce their
stability as protonated product ions in preference to the main core, and as a result, improve
the mass spectral quality in LC-MS/MS analysis. The absence of an active hydrogen on
tertiary amines renders traditional derivatization reactions, such as acylation, ineffective.
Moreover, most derivatization reactions, such as acylation or silylation, usually require
anhydrous reaction conditions because these derivatization reagents are very sensitive to
2
1
moisture and may hydrolyze. Because our investigated compounds may be found in
aqueous solutions, it is desirable that these amine transformations be carried out in water. Our
goal was to develop a new derivatization method that allows structural elucidation of
compounds containing a tertiary amine side chain. The reaction between hydrogen peroxide
2
2
and tertiary amines has been documented to lead to the formation of N-oxides. Mass
spectral analysis of various N-oxide compounds, including nitrogen mustards, N,N-
dialkylaminoethyl-2-chlorides and N,N-dialkylaminoethanols etc., demonstrated a distinctive
2
3,24,25
fragmentation pattern compared with the intact parent amine.
Therefore, we decided to
explore the oxidation reactions of various tertiary amines and their effect on the MS/MS
fragmentation behavior.
3
.2. Analyte selection
A series of lidocaine-related structures bearing a tertiary amine side chain with
varying aromatic cores (A1, A2, A3) or tertiary amine side chains (ropivacaine (A4),
This article is protected by copyright. All rights reserved.

bupivacaine (A5)), and a series of benzimidazole derivatives containing a tertiary amine side
2
chain (A6-A10), which all present information-poor MS spectra with insufficient
information about the core, were chosen to be analyzed and evaluated (Fig. 1).
3
.3. Reaction conditions (reaction time, oxidant concentration, base addition)
Several reaction parameters were investigated, including reaction time (1 min- 24 h),
H O concentration (0.01%- 1%) and effect of base addition (potassium carbonate). We
2
2
noticed that the oxidation rate of A1-A10 increased with H O concentration up to 0.1%,
2
2
while no substantial increase was further observed in the concentration range from 0.1% to
%. Base addition only slightly improved the signal intensity. Extending the reaction time
1
improved the signal intensity to some extent; after 1 min stirring, between 10-20% of the
analytes were converted to the oxide form. The analysis after 3 h revealed up to a 40%
conversion with a higher signal intensity, and a 40-60% conversion was obtained after 24 h.
Even though the N-oxide counterparts were less sensitive to detect than the non-oxidized
compounds, their MS/MS spectra provided the information needed to enable the
identification of the analytes. Hence, utilizing 0.1% H O with only 1 min of stirring enables
2
2
the formation of adequate amounts of N-oxide counterparts, allowing unambiguous
identification of the analytes at low ng/ml levels. In cases where the analyte concentration is
at the detection limit, prolonged reaction time of up to 24 h may improve signal intensity by a
factor of 3-5.
3
.4. MS/MS spectra of aromatic amides containing a tertiary amine side chain [R -NH-
1
CO-R ] (Lidocaine homologues, A1-A5)
2
This interesting class of aromatic amides contains some local anesthetic agents. Their
general structures are presented in Fig. 1, which shows that these molecules can be
This article is protected by copyright. All rights reserved.

characterized by two sub-structures: the ―core‖ substituted ring (R ) and the tertiary amine
1
side chain (R ). In general, the tertiary amine side chain of these compounds determines the
2
fragmentation pattern for both EI and ESI, regardless of the substituent present on the core
ring.
A1-A3 have the structure of R -NH-CO-CH -N(C H ) with different aromatic cores,
1
2
2
5 2
where R =2,6-dimethyl phenyl (A1), R =6-chloro-2-methyl phenyl (A2) and R =4-tert-butyl-
1
1
1
2
,6-dimethyl phenyl (A3). The only product ions observed in their MS/MS spectra were at
+
+
m/z 86 (corresponding to CH =N (C H ) ) and m/z 58, corresponding to CH =N H(C H )
2
2
5 2
2
2
5
(sequential loss of ethylene). Lidocaine (A1), ropivacaine (A4) and bupivacaine (A5) have an
identical core with different tertiary amine side chains, and their EI-MS and ESI-MS/MS
spectra exhibit fragments/product ions representing only the tertiary amine side chain (m/z
8
1
6, 126, and 140, respectively) without core-representative product ions. For example, m/z
26 and 140 in A4 and A5 correspond to the cleavage between the amide carbon and
piperidyl ring. The MS/MS spectra of A1-A3 and A5 are shown in Fig. 2 (left) (A4, not
shown).
The amide bond in all the derivatives A1-A5 appears to not be cleaved, unlike that in
peptides. This is probably due to amide bond stabilization by virtue of resonance with the
2
6
aromatic group. It is noteworthy that our attempt to obtain more abundant ions related to the
main core utilizing EPI (MS/MS) experiments with a broad range of collision induced
dissociation (CID) energies failed, even when we increased the analyte concentration up to
µg/ml levels.
This article is protected by copyright. All rights reserved.

3
.5. Oxidation effect on MS/MS spectra of the aromatic amides containing a tertiary
amine side chain [R -NH-CO-R ] (Lidocaine homologues, A1-A5)
1
2
Oxidation of A1-A5 employing a 0.1% hydrogen peroxide solution resulted in the
addition of an oxygen atom. The N-oxide counterparts with a 16 Da increase in mass, as
expected, were characterized by LC-ESI-MS/MS, revealing a different CID-MS/MS pattern
with product ions indicative of both the core ring and tertiary amine side chain, though the
main peak, that seems to exclude the formation of N-oxides, was still at m/z 86 for A1-A3
and m/z 126 and 140 for A4 and A5, respectively, for all collision energies. These peaks can
2
7
be rationalized however by OH-group migration in the protonated N-oxide or by loss of an
OH radical, followed by a homolytic carbon-carbon cleavage via a concerted process, as
reported in the fragmentation of N-oxide of N,N-dialkylaminoethyl-2-chlorides and N,N-
2
3,24
dialkylaminoethanols.
The ESI-MS/MS spectra of A1-A3 and A5 after the oxidation are
depicted in Fig. 2 (right). At a low collision energy (20 eV, not shown), cleavage at the C-N
bond in the amide position with charge retention on the tertiary N-oxide side chain was
observed in the ESI-MS/MS spectra (m/z 130 in A1-A3, m/z 170 in A4 and m/z 184 in A5).
At a higher collision energy (40 eV), an ion at m/z 88 (loss of CH =CO from m/z 130) was
2
observed for A1-A3, and an ion at m/z 100 (loss of CO and propylene or alternatively
butylene) was observed for A4 and A5. The ions attributed to the N-oxide of the tertiary
amine side chain strengthened our assumption that oxygen was incorporated in the tertiary
amine rather than the aromatic or benzylic sites. The formation of the ions at m/z 130 and m/z
2
7
8
8 in lidocaine has been discussed in the literature, demonstrating the incorporation of
oxygen on the right-hand side of the amide bond and supporting the assumption of N-oxide
1
8
formation. Furthermore, when the reaction was performed using a solution of 0.1% O-
labeled hydrogen peroxide, the expected 2 Da mass increase was observed for those ions (m/z
9
0 and 132 in A1-A3, m/z 102 and 172 in A4 and m/z 102 and 186 in A5). The MS/MS
This article is protected by copyright. All rights reserved.

18
spectra of A1 after oxidation with O-labeled hydrogen peroxide are depicted in the
supporting information, Fig. S1. At a collision energy of 40 eV, some cleavages with charge
retention on the core were observed in the ESI-MS/MS spectra. For A1 and A2, product ions
were observed at m/z 164 and 184, respectively. These ions are attributed to cleavage at the
C-N position (α to the tertiary N-oxide) with a proton transfer to the main core. The
equivalent ion for A3 at m/z 220 was not observed; however, an ion at m/z 164 was observed,
which apparently is the secondary product of the m/z 220 ion after a sequential loss of the
tert-butyl group. In addition, ions at m/z 148, 168, and 204 were observed for A1, A2 and A3,
respectively, and correspond to cleavage between the carbonyl and the carbon α to the tertiary
N-oxide. These ions sequentially lose the carbonyl group, resulting in the formation of ions at
m/z 120, 140 and 176, respectively. Another common dissociation process was observed at
the C-N amide bond, which resulted in the formation of ions at m/z 122, 142 (for A1 and A2)
and m/z 122 (A3) (C-N cleavage with sequential loss of the tert-butyl group). A comparison
between the ESI-MS/MS spectra of the N-oxides of lidocaine (A1), ropivacaine (A4) and
bupivacaine (A5), all with an identical core, reveals a repetitive fragmentation pattern with
ions at m/z 148, 120 and 105, which is indicative of the aromatic amide core. The ion at m/z
1
05 corresponds to cleavage at the aromatic C-N bond, leading to the formation of a 2,6-
dimethyl phenyl cation, in addition to the aforementioned ions at m/z 148 and 120. However,
the ion at m/z 164 was only observed in lidocaine and not in A4 and A5 due to their ring
system, preventing cleavage at the C-N position (α to the tertiary N-oxide). All the formulae
2
of the discussed ion structures were supported by orbitrap-MS exact mass measurements (<
2
ppm error) [Table 1]. It is noteworthy that the formation of the N-oxide counterparts leads
to an increase in polarity, and therefore, shorter LC retention times of all the oxide
derivatives were observed utilizing reverse-phase C18 chromatography.
This article is protected by copyright. All rights reserved.

3
.6. MS/MS spectra of benzimidazole derivatives containing a tertiary amine side chain
A6-A10)
The dissociation pathways of five benzimidazoles containing a tertiary amine side
(
1
6
chain, A6-A10, were previously studied. The most abundant product ions in the ESI-
MS/MS spectra of A6-A8 are at m/z 100, which corresponds to the charge-favorable N,N-
diethylaziridinium ion and m/z 72 (sequential loss of ethylene). For A9-A10, the dominant
ion is at m/z 114, corresponds to the charge-favorable vinyl morpholinium ion, which further
dissociates to m/z 84 (sequential loss of formaldehyde) and m/z 70 (sequential loss of
acetaldehyde). The MS/MS spectrum of A6 is depicted in Fig. 3 (left), and the MS/MS
spectra of A7-A10 are available in the supporting information, Figs. S2 and S3 (left).
All the above mentioned product ions are representative of only the amine-containing
residue. As mentioned before, this is attributed to the strong tendency of the amine side chain
to stabilize the positive charge rather than the benzimidazole core. A cleavage of the C-N
bond (α to the tertiary amine) that leads to an ion resulting from dimethylamine loss, which is
1
0
typical of alkyl amine groups, is not favorable here, as we previously reported. This type of
cleavage may have provided structural information about the benzimidazole core. However,
only low-abundance, core-indicative product ions were observed. Therefore, an interpretation
based on these low-abundance ions was impractical.
3
.7. Oxidation effect on MS/MS spectra of benzimidazole derivatives containing a
tertiary amine side chain (A6-A10)
In general, benzimidazole derivatives are very stable towards heat and oxidizing
2
8
agents. The two lone pair electrons contribute to the aromatic sextet by an imino group, and
the other nitrogen is a pyridine-like base with a pKa value of the conjugate acid ~5, versus
~11 for the tertiary amine side chain. Thus, we assumed oxidation would preferably occur on
This article is protected by copyright. All rights reserved.

the amine side chain moiety (similarly to A1-A5). Employing an oxidation reaction resulted
in high-intensity, rich and informative ESI-MS/MS spectra of all the N-oxide derivatives with
product ions representing all parts of the molecule. The ESI-MS/MS spectrum of the N-oxide
of etonitazene (A6) is presented in Fig. 3 (right).
For etonitazene, ions representative of the amine side chain (m/z 100, 72) were mainly
observed, in addition to trace-abundance ions indicative of the benzimidazole core and the
ethoxybenzyl side residue (m/z 278, 135, 107). For the N-oxide counterpart, product ions
indicative of the core were primarily observed. This can be attributed to the less favorable
tendency of the N-oxide group to stabilize a positive charge, leading to the more favorable
formation of core-structure indicative product ions. The ions at m/z 324 (loss of diethyl
amine), m/z 278 (sequential loss of NO ), m/z 250 (loss of ethylene from m/z 278), m/z 202
2
(loss of ethoxy benzene from m/z 324), and m/z 156 (loss of phenol from m/z 250) are all
indicative of the benzimidazole core. Comparison of the MS/MS fragmentation of the
3
oxidized etonitazene to the MS fragmentation of the un-oxidized etonitazene, as we
1
6
previously reported, revealed identical product ions.
The oxidation-assisted method was applied to 4 additional, different benzimidazole-
containing tertiary amine side chain compounds (A7-A10) to reveal their core structure (Figs.
S2 and S3 in the supporting information). In contrast to A1-A5, no product ions exhibiting
1
8
the incorporation of oxygen were observed for A6-A10, hence O-labeled oxidation
experiments have no added value and were not performed. The generated N-oxide
counterparts of A6-A10, completely altered the charge distribution during MS/MS
fragmentation. In the non-oxidized compounds (A6-A10), the C-N bond cleavage (α to the
tertiary amine side chain), resulted in extremely low abundance ions indicative of the
benzimidazole core, with MS/MS spectra dominated by product ions representing the amine
side chain. However, in the N-oxide counterparts of A6-A10, the same C-N bond cleavage (α
This article is protected by copyright. All rights reserved.

to the tertiary amine-oxide side chain), resulted in high abundance ions indicative of the
benzimidazole structure. From all investigated benzimidazole derivatives (A6-A10), only the
MS/MS spectra of the N-oxide of A6 show, in addition, ions at m/z 100 (representing the
tertiary amine side chain N,N-diethylaziridinium ion) and m/z 72 (sequential loss of
acetaldehyde), as supported by HRMS data, and not the expected 16 Da oxygen mass
2
7
increase. As mentioned before for A1-A5, this can be rationalized by OH-group migration
2
3,24
or by loss of an OH radical, followed by C-N bond cleavage in a concerted process.
All
2
the observed product ions were supported by ESI-orbitrap-MS exact mass measurements (<
2
ppm error), enabling detailed structural elucidation and unambiguous identification of the
analytes. For the exact mass measurements of the product ions of A6, see Table 1, and for
A7-A10, see Figs. S2 and S3 in the supporting information.
This article is protected by copyright. All rights reserved.

4
.
Conclusions
We have demonstrated a new, convenient and very simple analytical technique for
rapid structural elucidation of compounds containing a tertiary amine side chain (lidocaine
homologues and benzimidazole derivatives). The method is based on instantaneous oxidation
with a 0.1% solution of hydrogen peroxide in water that forms an N-oxide, altering the
charge distribution during ESI-MS/MS fragmentation and revealing structural information
that is otherwise not available and can easily be traced back to the original, un-oxidized
compound. Although a sensitivity loss of approximately one order of magnitude was
observed with the N-oxide derivatives, extensive fragmentation patterns with complete
representation of both sides of the molecule were achieved, allowing unambiguous
identification of such compounds at trace levels. It is noteworthy, that the mild oxidation
carried out in this study, generally resulted in one significant peak of the N-oxide product,
with no measurable amount of other oxidation products. However, in some derivatives, a few
small LC peaks of other oxidized species were observed (3-10% of the main peak), with
MS/MS spectra similar to the original un-oxidized compounds. We assume that oxidation in
other positions, distant from the tertiary amine side chain site (for example, on the
benzimidazole ring) would not significantly change the charge distribution during MS/MS
fragmentation. Several substrates that can potentially be oxidized in a competitive reaction to
the tertiary amine (for example, compounds containing a sulfur group), may result in the
formation of positional isomers. LC separation prior to ESI-MS analysis is usually beneficial
to separate such positional isomers and their distinction can be achieved based on differences
in their MS/MS spectra. Most challenging are rare cases in which two isomers co-elute,
leading to ambiguity in the identification process. Careful LC analysis with improved
separation and MS/MS interpretation should be performed in such cases. The method
This article is protected by copyright. All rights reserved.

described here can be elaborated for different cores attached to various aliphatic and cyclic
amines and may facilitate the identification of ―unknowns‖.
This article is protected by copyright. All rights reserved.

References
[
1]
E.L. Schymanski, J. Jeon, R. Gulde, K. Fenner, M. Ruff, H.P. Singer, J. Hollender.
Identifying small molecules via high resolution mass spectrometry: communicating
confidence. Environ. Sci. Technol. 2014, 48, 2097-2098.
[
[
2]
3]
A.G. Marshall, C.L. Hendrickson. High-resolution mass spectrometers. Annu. Rev.
Anal. Chem. 2008, 1, 579-599.
T. Kind, O. Fiehn. Metabolomic database annotations via query of elemental
compositions: mass accuracy is insufficient even at less than 1 ppm. BMC
Bioinformatics. 2006, 7, 234.
[
[
4]
5]
T. Kind, O. Fiehn. Advances in structure elucidation of small molecules using mass
spectrometry. Bioanal. Rev. 2010, 2, 23-60.
S. Wolf, S. Schmidt, M. Muller-Hannemann, S. Neumann. In silico fragmentation for
computer assisted identification of metabolite mass spectra. BMC Bioinformatics.
2
010, 11, 148.
[
[
6]
7]
A. Vaniya, O. Fiehn. Using fragmentation trees and mass spectral trees for identifying
unknown compounds in metabolomics. Trends Analyt. Chem. 2015, 69, 52-61.
W.M. Niessen. Group-specific fragmentation of pesticides and related compounds in
liquid chromatography-tandem mass spectrometry. J. Chromatogr. A 2010, 1217,
4
061-4070.
[
[
8]
9]
K. Levsen, H.M. Schiebel, J.K. Terlouw, K.J. Jobst, M. Elend, A. Preiß, H. Thiele, A.
Ingendoh. Even-electron ions: a systematic study of the neutral species lost in the
dissociation of quasi-molecular ions. J. Mass Spectrom. 2007, 42, 1024-1044.
M. Holčapek, R. Jirásko, M. Lísa. Basic rules for the interpretation of atmospheric
pressure ionization mass spectra of small molecules. J. Chromatogr. A 2010, 1217,
3
908-3921.
This article is protected by copyright. All rights reserved.

[
[
10] A. Weissberg, S. Dagan. Interpretation of ESI(+)-MS-MS spectra—towards the
identification of ―unknowns‖. Int. J. Mass Spectrom. 2011, 299, 158-168.
11] D.W. Hill, T.M. Kertesz, D. Fontaine, R. Friedman, D.F. Grant. Mass spectral
metabonomics beyond elemental formula: chemical database querying by matching
experimental with computational fragmentation spectra. Anal. Chem. 2008, 80, 5574-
5
582.
[
12] A. Pelander, E. Tyrkkö, I. Ojanperä. In silico methods for predicting metabolism and
mass fragmentation applied to quetiapine in liquid chromatography/time-of-flight
mass spectrometry urine drug screening. Rapid Commun. Mass Spectrom. 2009, 23,
5
06-514.
[
[
13] ACD Labs. Available at http://www.acdlabs.com/.
14] A. Kerber, R. Laue, M. Meringer, K. Varmuza. MOLGEN-MS: evaluation of low
resolution electron impact mass spectra with MS classification and exhaustive
structure generation. Adv. Mass Spectrom. 2001, 15, 939-940.
[
[
15] A. Koehler, R. Oertel, W. Kirch. Simultaneous determination of bupivacaine,
mepivacain, prilocaine and ropivacain in human serum by liquid chromatography-
tandem mass spectrometry. J. Chromatogr. A 2005, 1088, 126-130.
16] A. Weissberg, M. Madmon, S. Dagan. Structural identification of compounds
3
containing tertiary amine side chains using ESI-MS combined with fragmentation
pattern matching to chemical analogues – benzimidazole derivatives as a case study.
Int. J. Mass Spectrom. 2016, 394, 9-21.
[
17] A. Weissberg, N. Tzanani, S. Dagan. Specificity enhancement by electrospray
ionization multistage mass spectrometry – a valuable tool for differentiation and
identification of ‗V‘-type chemical warfare agents. J. Mass Spectrom. 2013, 48, 1340-
1
348.
This article is protected by copyright. All rights reserved.

[
18] E. ter Weijden, M.P. van den Broek, F.F. Ververs. Easy and fast LC-MS/MS
determination of lidocaine and MEGX in plasma for therapeutic drug monitoring in
neonates with seizures. J. Chromatogr. B Analyt. Technol. Biomed. Life Sci. 2012,
8
81-882, 111-114.
[
[
[
[
[
19] E.E. Essien, D.A. Cowan, A.H. Beckett. Metabolism of phenothiazines: identification
of N-oxygenated products by gas chromatography and mass spectrometry. J. Pharm.
Pharmacol. 1975, 27, 334-342.
20] J.S. Markowitz, K.S. Patrick. Thermal degradation of clozapine-N-oxide to clozapine
during gas chromatographic analysis. J. Chromatogr. B Biomed. Appl. 1995, 668,
1
71-174.
21] H. Kataoka. Derivatization reactions for the determination of amines by gas
chromatography and their applications in environmental analysis. J. Chromatogr. A
1
996, 733, 19-34.
22] A.A. Oswald, D.L. Guertin. Organic nitrogen compounds. I. Peroxide intermediates
of tertiary alkylamine oxidation by hydrogen peroxide. J. Org. Chem. 1963, 28, 651-
6
57.
23] L. Sridhar, R. Karthikraj, M. R. V. S. Murty, N. Prasada Raju, M. Vairamani, S.
Prabhakar. Mass spectral analysis of N-oxides of nitrogen mustards, and N,N-
dialkylaminoethyl-2-chlorides under electrospray ionization conditions. Int. J. Mass
Spectrom. 2013, 333, 15-20.
[
24] L. Sridhar, R. Karthikraj, M. R. V. S. Murty, N. Prasada Raju, M. Vairamani, S.
Prabhakar. Mass spectral analysis of N-oxides of chemical weapons convention
related aminoethanols under electrospray ionization conditions. Rapid Commun. Mass
Spectrom. 2011, 25, 533-542.
This article is protected by copyright. All rights reserved.

[
25] R. Ramanathan, A.-D. Su, N. Alvarez, N. Blumenkrantz, S. K. Chowdhury, K. Alton,
J. Patrick. Liquid chromatography/mass spectrometry methods for distinguishing N-
oxides from hydroxylated compounds. Anal. Chem. 2000, 72, 1352-1359.
[
26] C. Joyce, W.F. Smyth, V.N. Ramachandran, E. O'Kane, D.J. Coulter. The
characterisation of selected drugs with amine-containing side chains using
electrospray ionisation and ion trap mass spectrometry and their determination by
HPLC-ESI-MS. J. Pharm. Biomed. Anal. 2004, 36, 465-476.
[
[
27] E. Nouri-Nigjeh, H.P. Permentier, R. Bischoff, A.P. Bruins. Lidocaine oxidation by
electrogenerated reactive oxygen species in the light of oxidative drug metabolism.
Anal. Chem. 2010, 82, 7625-7633.
28] J.A. McCleverty, T.J. Meyer. Comprehensive Coordination Chemistry II: From
Biology to Nanotechnology. Elsevier Science, United Kingdom, 2003.
This article is protected by copyright. All rights reserved.

Table 1. Exact mass measurements of product ions of 10 ng/ml solutions of A1-A6 and their
N-oxide counterparts (mass errors < 2 ppm).
Compound RT
Measured +
Product ion masses
(min) mass (MH )
A1
A1 oxide
7.6
6.2
235.1806
251.1754
86.0964, 58.0654
164.1069, 148.0755, 130.0862, 122.0964, 120.0808, 105.0698,
8
8.0757, 86.0964
A2
A2 oxide
8.0
5.9
255.1260
271.1206
86.0964, 58.0654
184.0524, 168.0211, 142.0416, 140.0261, 130.0863, 88.0757,
8
6.0965
A3
A3 oxide
9.6
8.5
291.2432
307.2382
86.0964, 58.0654
204.1383, 176.1434, 164.1070, 130.0863, 122.0964, 88.0757,
8
6.0964
A4
A4 oxide
8.8
7.5
275.2120
291.2066
126.1277, 98.0964, 84.0808
170.1176, 148.0756, 126.1277, 120.0808, 105.0699, 100.0757,
9
8.0964, 84.0808
A5
A5 oxide
8.6
8.0
289.2276
305.2223
140.1434, 98.0965, 84.0808
184.1332, 148.0756, 140.1435, 120.0808, 105.0700, 100.0757,
9
8.0965, 84.0809, 70.0652
A6
A6 oxide
9.3
7.8
397.2234
413.2183
278.1409, 135.0803, 107.0491, 100.1121, 72.0809
324.1342, 296.1028, 278.1417, 250.1096, 202.0610, 156.0683,
1
35.0805, 107.0492, 100.1121, 72.0809
This article is protected by copyright. All rights reserved.

Fig. 1. Structures of the amino amide (A1-A5) and benzimidazole (A6-A10) compounds
studied in this work.
This article is protected by copyright. All rights reserved.

Fig. 2. Enhanced product ion (EPI) MS/MS spectra of 10 ng/ml solutions of A1-A3, A5 (left)
and their N-oxide counterparts (right).
This article is protected by copyright. All rights reserved.

Fig. 3. Enhanced product ion (EPI) MS/MS spectra of a 10 ng/ml solution of A6 (left) and its
N-oxide counterpart (right).
This article is protected by copyright. All rights reserved.