Electrospray ionization tandem mass spectrometry of monoketone curcuminoids

Article abstract of DOI:10.1002/rcm.8699

Rationale: Although monoketone curcuminoids (MKCs) have been largely investigated due to their biological activities, data on the gas-phase fragmentation reactions of protonated MKCs under collision-induced dissociation (CID) conditions are still scarce. Here, we combined electrospray ionization tandem mass spectrometry (ESI-MS/MS) data, multiple-stage mass spectrometry (MSⁿ), deuterium exchange experiments, accurate-mass data, and thermochemical data estimated by computational chemistry to elucidate and to rationalize the fragmentation pathways of eleven synthetic MKCs. Methods: The MKCs were synthesized by Claisen-Schmidt condensation under basic (1–9) or acidic (10–11) conditions. ESI-CID-MS/MS analyses and deuterium-exchange experiments were carried out on a triple quadrupole mass spectrometer. MSⁿ analyses on an ion trap mass spectrometer helped to elucidate the fragmentation pathways. Accurate-mass data and thermochemical data, obtained at the B3LYP/6–31+G(d,p) level of theory, were used to support the ion structures. Results: The most intense product ions were the benzyl ions ([C₇H₂R₁R₂R₃R₄R₅]⁺) and the acylium ions ([M + H ? C₈H₃R₁R₂R₃R₄R₅]⁺), which originated directly from the precursor ion as a result of two competitive hydrogen rearrangements. Product ions [M + H – H₂O]⁺ and [M + H ? C₆HR₁R₂R₃R₄R₅]⁺, which are formed after Nazarov cyclization, were also common to all the analyzed compounds. In addition, ?Br and ?Cl eliminations were diagnostic for the presence of these halogen atoms at the aromatic ring, whereas ?CH₃ eliminations were useful to identify the methyl and methoxy groups attached to this same ring. Nazarov cyclization in the gas phase occurred for all the investigated MKCs and did not depend on the presence of the hydroxyl group at the aromatic ring. However, the presence and the position of a hydroxyl group at the aromatic rings played a key role in the Nazarov cyclization mechanism. Conclusions: Our results reinforce some aspects of the fragmentation pathways previously published for 1,5-bis-(2-methoxyphenyl)-1,4-pentadien-3-one and 1,5-bis-(2-hydroxyphenyl)-1,4-pentadien-3-one. The alternative fragmentation mechanism proposed herein can explain the fragmentation of a wider diversity of monoketone curcuminoids.

Full text of DOI:10.1002/rcm.8699

Received: 16 August 2019
DOI: 10.1002/rcm.8699
Revised: 20 November 2019
Accepted: 12 December 2019
S P E C I A L I S S U E R E S E A R C H A R T I C L E
Electrospray ionization tandem mass spectrometry of
monoketone curcuminoids
Tatiana M. Vieira1
Saulo S.P. Furtado2 | Ricardo Vessecchi1
Antônio E.M. Crotti1
| Renato P. Orenha2
| Eduardo J. Crevelin1
| Renato L.T. Parreira2
|
|
1Departamento de Química, Faculdade de
~
Filosofia, Ciências e Letras de Ribeirao Preto,
Rationale: Although monoketone curcuminoids (MKCs) have been largely
investigated due to their biological activities, data on the gas-phase fragmentation
reactions of protonated MKCs under collision-induced dissociation (CID) conditions
are still scarce. Here, we combined electrospray ionization tandem mass
spectrometry (ESI-MS/MS) data, multiple-stage mass spectrometry (MSn), deuterium
exchange experiments, accurate-mass data, and thermochemical data estimated by
computational chemistry to elucidate and to rationalize the fragmentation pathways
of eleven synthetic MKCs.
~
Universidade de Sao Paulo, Av. Bandeirantes,
~
3900, Monte Alegre, CEP 14040-901Ribeirao
Preto, SP, Brazil
2Núcleo de Pesquisas em Ciências Exatas e
Tecnológicas, Universidade de Franca, CEP
14404-600Franca, SP, Brazil
Correspondence
A. E. M. Crotti, Departamento de Química,
Faculdade de Filosofia, Ciências e Letras de
~
~
Ribeirao Preto, Universidade de Sao Paulo,
Av. Bandeirantes, 3900, Monte Alegre, CEP
Methods: The MKCs were synthesized by Claisen-Schmidt condensation under basic
(1–9) or acidic (10–11) conditions. ESI-CID-MS/MS analyses and deuterium-
exchange experiments were carried out on a triple quadrupole mass spectrometer.
MSn analyses on an ion trap mass spectrometer helped to elucidate the
fragmentation pathways. Accurate-mass data and thermochemical data, obtained at
the B3LYP/6–31+G(d,p) level of theory, were used to support the ion structures.
Results: The most intense product ions were the benzyl ions ([C7H2R1R2R3R4R5]+)
and the acylium ions ([M + H − C8H3R1R2R3R4R5]+), which originated directly from
the precursor ion as a result of two competitive hydrogen rearrangements. Product
ions [M + H – H2O]+ and [M + H − C6HR1R2R3R4R5]+, which are formed after
Nazarov cyclization, were also common to all the analyzed compounds. In addition,
•Br and •Cl eliminations were diagnostic for the presence of these halogen atoms at
the aromatic ring, whereas •CH3 eliminations were useful to identify the methyl and
methoxy groups attached to this same ring. Nazarov cyclization in the gas phase
occurred for all the investigated MKCs and did not depend on the presence of the
hydroxyl group at the aromatic ring. However, the presence and the position of a
hydroxyl group at the aromatic rings played a key role in the Nazarov cyclization
mechanism.
~
14040-901, Ribeirao Preto, SP, Brazil.
Email: millercrotti@ffclrp.usp.br
Funding information
Conselho Nacional de Desenvolvimento
Científico e Tecnológico, Grant/Award
~
Number: 313648/2018-2; Coordenaçao de
Aperfeiçoamento de Pessoal de Nível Superior,
Grant/Award Number: Finance Code 001;
~
Fundaçao de Amparo à Pesquisa do Estado de
~
Sao Paulo, Grant/Award Numbers:
2011/07623-8, 2016/19272-9,
2017/24856-2
Conclusions: Our results reinforce some aspects of the fragmentation pathways
previously published for 1,5-bis-(2-methoxyphenyl)-1,4-pentadien-3-one and 1,5-bis-
(2-hydroxyphenyl)-1,4-pentadien-3-one. The alternative fragmentation mechanism
proposed herein can explain the fragmentation of a wider diversity of monoketone
curcuminoids.
Rapid Commun Mass Spectrom. 2020;e8699.
wileyonlinelibrary.com/journal/rcm
© 2019 John Wiley & Sons, Ltd.
1 of 17
https://doi.org/10.1002/rcm.8699

2 of 17
VIEIRA ET AL.
1
|
INTRODUCTION
cyclization of protonated chalcones in the gas phase is analogous to
Nazarov cyclization in solution, but its occurrence in the gas phase
depends on the employed protonation method and the nature and
position of the substituents at the aromatic ring. More recently,
Cyriac and co-workers also verified that the gas-phase Nazarov
cyclization of protonated 2-methoxychalcone is analogous to the
corresponding process in solution. The authors extended their studies
to 1,5-bis- (2-methoxyphenyl)-1,4-pentadien-3-one to demonstrate
that the methoxy group acts as an effective “catalyst” for the transfer
of protons involved in the cyclization step.11
Curcumin
(1,7-bis(4-hydroxy-3-methoxyphenyl)-hepta-1,6-dien-
3,5-dione], I, Figure 1) is a hydrophobic phenolic compound isolated
from turmeric (Curcuma sp.).1-3 The term “curcuminoids” is used to
describe curcumin and other α,β-unsaturated bis-β-diketone
analogues (e.g., demethoxycurcumin (II) and bis-demethoxycurcumin
(III)) and structurally related compounds, such as monoketone
curcuminoids (IV), 4-hydroxymethylene curcuminoids (V), and
4-arylidene curcuminoids (VI, Figure 1).
Monoketone curcuminoids (MKCs), also named mono-carbonyl
As part of our ongoing project on the gas-phase fragmentation
reactions of biologically active synthetic and natural compounds,13-21
and given the scarcity of ESI-MS/MS data of MKCs, here we
investigate the gas-phase fragmentation pathways of selected
curcuminoids by ESI-MS/MS combined with thermochemical data
obtained at the B3LYP/6-31+G(d) level of theory.
curcuminoids,
are
1,5-diaryl-penta-1,4-diene-3-ones.
These
compounds have attracted interest due to their structural similarity
with curcumin (MKCs have one less methylene and one less carbonyl
group than curcumin) and to their various biological activities,
including antimicrobial,4 cytotoxic,2 anti-inflammatory,1 antiparasitic,5
and anti-HIV actions.6
Over the last decade, electrospray ionization tandem mass
spectrometry (ESI-MS/MS) has been used to identify natural and
synthetic organic compounds mainly in combination with liquid
chromatography (LC/MS/MS). Because MS is highly sensitive, it has
been widely employed in in vitro and in vivo metabolism studies.7 In
this context, the MS/MS data of the class of compounds under
study play a key role.8 The gas-phase fragmentation reactions of a
series of protonated and deprotonated β-diketone curcuminoids
have been previously investigated.9,10 However, in the case of the
MKCs, data on their fragmentation are still scarce. To date, a
systematic study of only three compounds has been performed by
Cyriac and co-workers, who proposed the occurrence of a Nazarov
cyclization to explain the eliminations of anisole and/or ketene
from protonated 1,5-bis-(2-methoxyphenyl)-1,4-pentadien-3-one and
1,5-bis-(2-hydroxyphenyl)-1,4-pentadien-3-one.11 In solution, the
Nazarov cyclization is an acid-catalyzed cyclization reaction of divinyl
ketones to cyclopentanones. The mechanism involves the conrotatory
electrocyclic ring closure of a protonated divinyl ketone followed by
deprotonation and double-bond reorganization.12 In solution,
sufficiently basic substituents like OCH3 or OH in the ortho position
may cause deprotonation during Nazarov cyclization or aid proton
transfer. This reaction has aroused great interest in mass
spectrometry because these groups can catalyze proton migration in
the gas phase.12 George and co-workers have shown that Nazarov
2
|
EXPERIMENTAL
2.1
|
Synthesis of monoketone curcuminoids 1–11
Compounds 1–9 were obtained by aldol condensation between
acetone and different aromatic aldehydes according to a previously
published methodology (Scheme 1).4 Briefly, a solution of NaOH
(2.5 mol/L) and ethanol (20 mL) was added to a flask (100- mL)
containing an aromatic aldehyde (26 mmol) and acetone (0.95 mL,
13 mmol). The reaction mixture was stirred at 0ꢀC for 1–24 h, washed
with cold water (to remove the excess base), and filtered under
reduced pressure. The resulting solids were dried under vacuum, to
yield compounds 1 (24% yield), 2 (64% yield), 3 (69% yield), 4 (83%
yield), 5 (20% yield), 6 (72% yield), 7 (45% yield), 8 (97% yield), and
9 (44% yield). Compounds 10 and 11 were synthesized by Claisen-
Schmidt condensation in acidic conditions according to the
methodology described in the literature.6 Initially, acetic acid (50 mL)
was saturated with hydrogen chloride at 0ꢀC. Next, a mixture of
vanillin (3.95 g, 26 mmol) and acetone (0.75 g, 13 mmol) was slowly
added to the acid. The mixture was brought to room temperature and
stirred for 24 h. Subsequently, the crude mixture was poured into
ice-cold water (200 mL) and extracted with ethyl acetate (3 × 30 mL);
the organic phase was dried over MgSO4. After the solvent had been
FIGURE 1 Chemical structures of curcumin (I),
demethoxycurcumin (II), bis-demethoxycurcumin (III), and
structurally related monoketone curcuminoids (IV),
4-hydroxymethylene curcuminoids (V), and 4-arylidene
curcuminoids (IV)

VIEIRA ET AL.
3 of 17
SCHEME 1 Synthesis of
monoketone curcuminoids 1–11
removed by evaporation under reduced pressure in
a
rotary
2.3
|
Computational methods
evaporator, the resulting mixture was purified by flash column
chromatography; isocratic elution with hexane/ethyl acetate 1:1 (v/v)
was employed. Finally, compounds 10 and 11 (21% and 44% yield,
respectively) were obtained as yellow solids. All the compounds were
identified on the basis of their 1H and 13C NMR, ESI-MS, UV and IR
data (see supporting information).
The geometry of all the structures was optimized by the B3LYP
computational method22,23 and the 6-31+G(d,p)24 basis set with
Gaussian09 software.25 The vibrational frequency analysis
demonstrated that all the studied compounds were in a minimum on
their respective potential energy surfaces. The Gibbs energies and the
enthalpies calculated at 298.15 K of the most stable ion structures
were used to investigate the fragmentation pathways. The gas-phase
proton affinity (PA) values were calculated from the protonation
reaction Gibbs energy and enthalpy: M + H+ ! MH+, as has usually
been done in recent studies.15,23 The experimental enthalpy and the
Gibbs energy values that were considered for the proton were 1.48
and −6.28 kcal mol−1, respectively.22
2.2
|
Mass spectrometry analyses
Electrospray ionization tandem mass spectrometry (ESI-MS/MS)
analyses of curcuminoids 1–11 were carried out on a Xevo TQS
tandem quadrupole mass spectrometer (QqQ; Waters, Milford, MA,
USA) equipped with a Z-spray ionization source operating in the
positive ion mode. The samples were dissolved in MeOH/H2O 4:1
(v/v) to achieve
a
final concentration of curcuminoids of
3
|
RESULTS AND DISCUSSION
1.0 μg/mL. Next, traces of 0.1% formic acid were added, and the
sample was infused directly into the ionization source at a flow rate of
0.1 mL/min. In the case of deuterium-exchange experiments,
compounds 1–11 were dissolved in MeOH/D2O 1:1 (v/v). To
maximize the relative intensity of the peaks corresponding to the
protonated MKCs, the capillary and the cone voltages were optimized
to 3.2 kV and 40 V, respectively. Electrospray ionization collision-
induced dissociation tandem mass spectrometry (ESI-CID-MS/MS)
experiments were carried out by using argon (99.999% purity) as the
collision gas on the selected precursor ion ([M + H]+) at collision
energy values ranging from 5 to 50 eV. The ionization source
temperature was 150ꢀC and the desolvation gas temperature (N2)
was 250ꢀC. The injected sample volume was 5 μL.
Multiple-stage ESI mass spectrometry (ESI-MSn) analyses were
performed on an ion trap (IT) mass spectrometer (Amazon Speed,
Bruker Daltonics, Bremen, Germany) operating in the positive ion
mode. The ion trap amplitude was varied from 40 to 60 and
optimized for each compound. N2 was used as nebulizing (6 psi
pressure) and drying gas (8 mL/min flow rate, 220ꢀC temperature),
and the capillary voltage and the end plate offset voltage were set
to 3.5 kV and 500 V, respectively. The analytical mass range was
m/z 50–450.
3.1
|
Structure–fragmentation correlations
When a series of compounds that share the same structural core is
available, comparison between their product ion spectra obtained
at an optimum collision energy (Elab) can provide useful structural
fragmentation correlations to elucidate their gas-phase fragmentation
pathways.13,15 The optimum value of Elab is usually expected to
reduce the relative intensity of the protonated molecule in the
product ion spectrum to below 50% without promoting extensive
fragmentation. Thus, on the basis of the relative intensity (RI, %)
versus Elab (eV) plots, the optimum Elab for MKCs 1–11 was 20 eV
(Figure 2 for compound 1, and supporting information for compounds
2–11).
Figure 3 shows the product ion spectra of the protonated
compounds obtained at Elab = 20 eV; Table 1 summarizes their main
product ions (relative intensity higher than 5%), where H is the most
intense product ion in the spectra of most of the analyzed
compounds. This ion results from direct elimination of
C10H4OR1R2R3R4R5 from the protonated molecule (A). Formation of
D from A by elimination of C8H3OR1R2R3R4R5 competes with
formation of H, as shown in Scheme 2. Other common product ions

4 of 17
VIEIRA ET AL.
FIGURE 2 Correlation between the relative
intensity (%) and the collision energy (Elab, in eV)
for compound 1 [Color figure can be viewed at
wileyonlinelibrary.com]
include B (A–H2O), C (A–C6HR1R2R3R4R5), and G (D–CO). F (C–
C2H2O) and E (C–CO) also appear in the spectra of most of the
investigated MKCs, except for compound 7. The MS3 spectra support
formation of E and F from C (see supporting information).
discussed.14,15. In general, two possibilities are considered. The first,
known as the “mobile proton theory”,24 assumes that the proton
initially binds to the most basic site of the molecule and then migrates
to more basic sites under CID conditions, with fragmentation
occurring from species resulting from this migration. The second
assumes that the proton remains bound to the most basic site of the
molecule in the post-collision state. In this case, the translational
energy is converted into internal energy, which triggers the
fragmentation reactions.26
A detailed analysis of the product ion spectra of protonated
compounds 1–11 allowed us to identify some diagnostic ions that
could help to distinguish the nature of the substituents at the
aromatic ring of these compounds (Scheme 2). Consecutive
eliminations of two methyl radicals (15 Da each) from B to produce
Q (m/z 230) and R (m/z 215) are diagnostic for compound 3, which
displays a methyl group at the aromatic ring. Similarly, formation of
T (m/z 250 for compounds 5 and 9 and m/z 294 for compound 7)
from B and formation of V (m/z 215) from T due to radical X•
eliminations (a chlorine for compounds 5 and 9 and a bromine for
compound 7) are diagnostic for compounds 5, 7, and 9. Radical X•
eliminations can also occur directly from A, to produce S (m/z 268 and
m/z 312) and U (m/z 233), or from G, to form P (m/z 102). These
eliminations are diagnostic for compounds 5, 7, and 9. Methyl radical
eliminations from other product ions also occur and produce product
ions that are diagnostic for the methoxy group at the aromatic ring
and specific for each compound. For example, •CH3 elimination
from H, G, D, and E yields O (m/z 136 for compound 4 and
m/z 166 for compound 6), K (m/z 206 for compounds 6 and 8), J (m/z
Figure 4 displays the proton affinity (PA) values for different sites
of the structures of compounds 1–11, as estimated by computational
chemistry. These data indicate that the carbonyl oxygen is the most
susceptible site to protonation for all the investigated MKCs, as
previously reported by Cyriac and co-workers for compounds
1 and 2.11 Recently, Hu and co-workers studied the reactions of
α,β-unsaturated aromatic ketones (chalcones) during mass
spectrometry. As in the case of MKCs, the investigated chalcones
bear a vinyl unit between the carbonyl and the phenyl groups. The
authors showed that the carbonyl oxygen in the studied chalcones is
the most susceptible site to protonation; however, protonation at the
carbon α to the carbonyl may also occur. The authors described
proton migration from the carbonyl oxygen to less favorable sites,
such as the α-carbon, as an energy-dependent process involving a
certain energy barrier.25 In the present study, we only consider the
species protonated at the carbonyl oxygen as the precursor ion.
118 for compound
2 and m/z 148 for compound 4), and
Y (m/z 160 for compound 10), respectively, which are diagnostic for
compounds 2, 4, 6, 8, and 10. The MS3 and MS4 experiments
also support formation of J, K, L, O, Q, R, S, T, U, and V (see
supporting information).
3.3
|
Formation of product ions B–G
H (A – C10H4OR1R2R3R4R5) and D (A – C8H3OR1R2R3R4R5) are the
most intense product ions in almost all the spectra of compounds
1–11 (Figure 3, Table 1). These ions originate directly from the
precursor ion (protonated molecule) by two competitive hydrogen
rearrangements, as depicted in Scheme 3. Formation of acylium ion
3.2
|
Protonation sites and gas-phase reactivity
The importance of determining the protonation site in gas-phase
fragmentation studies under CID conditions has recently been

VIEIRA ET AL.
5 of 17
FIGURE 3 Product ion spectra of
protonated monoketone curcuminoids
1–11 (QqQ, Ar, Elab = 20 eV)

6 of 17
VIEIRA ET AL.

VIEIRA ET AL.
7 of 17

8 of 17
VIEIRA ET AL.
SCHEME 2 Structure–fragmentation correlations in protonated MKCs 1–11
D (path II, Scheme 3) by means of a hydrogen rearrangement is similar
to the hydrogen rearrangement that has been previously reported for
2-aroylbenzofurans15 and chalcones.25 This mechanism is fully
supported by data from the deuterium-exchange experiments
(Table 1). For compounds 1 and 2, the D structure has been
proposed.11 Experimental data revealed that H is more intense than D
in the product ion spectra of compounds 1–3, 5, 7, and 9–11,
whereas D is more intense than H in the product ion spectra of
compounds 4, 6, and 8 at Elab = 20 eV. ΔH values for H and
D formation from A revealed that A' formation (which leads to
H formation) is energetically more favored (−3.2–5.5 kcal.mol−1) than
Consequently, most of the precursor ions A are converted into A'
instead of D, which could be explained by the higher intensity of
H than that of D. D is further converted into G by elimination of CO
resulting from heterolytic C1–C2 bond cleavage. However, our
calculations revealed that the vinylic ion
G
is unstable
(it spontaneously rearranged to G1 during the geometry optimization).
The calculated ΔH values indicated that the D ! G1 conversion is
less
favored
for
compounds
4
(ΔH = 50.5 kcal.mol−1),
6 (ΔH = 51.0 kcal.mol−1), and 8 (ΔH = 38.6 kcal.mol−1) than for the
other MKCs. This can be related to the presence of more than one
methoxyl group at the aromatic rings, which stabilizes the D structure
compared with G1 due to their electron-release mesomeric effect.
D
formation (ΔH values = 31.1–38.5 kcal.mol−1
,
Scheme 3).

VIEIRA ET AL.
9 of 17
Because of the higher stability of D derived from compounds 4, 6, and
8, a reduced number of D ions are converted into G/G1, so that a
high number of intact ions D reach the detector. This can explain, at
least in principle, why D is the base peak in the product ion spectra of
compounds 4, 6, and 8 (Table 1).
ΔH values estimated by computational calculations. The authors
concluded that the oxygen atom (a hydroxyl or a methoxyl group
at C9 and C9’) plays a crucial part in these steps. They also
reported losses of arene and ketene for compounds 1 and 2;
however, they postulated that these processes must involve
different fragmentation pathways without providing further details.
Here, we provide an alternative mechanism for [MKC + H − arene]+
and [MKC + H − ketene]+ formation that operates for various
compounds and which does not depend on the presence of the
oxygen atom at the aromatic ring.
For compounds 2, 3, 5, 7, and 9, the nature of the substituent at
the aromatic ring also plays an essential role in formation of G from
D (Scheme 4). For compounds 2 and 3, which bear a methyl and a
methoxy substituent at the aromatic ring p-position, respectively, the
relative intensity of G is higher than 70%. On the other hand, for
compounds 5, 7, and 9, which have a halogen (F or Cl) attached to the
aromatic ring, the G/G1 intensity is lower than 35%. These relative
intensities are consistent with the estimated ΔH values for the
D ! G/G1 decomposition; these values are lower for compounds
For compounds 1–9, Nazarov cyclization involves the most stable
conformer A1 and produces A1’ (pathway III, Scheme 4). Water
elimination from A1 to produce B is preceded by aryl migration from
C3 to C2 and consequent A1” formation. The aryl migration
mechanism has been extensively investigated by Cyriac and
co-workers.11 Nevertheless, although the authors reported water
elimination from protonated 1 and 2, they did not discuss the
underlying mechanisms or the ion structures. B (A1–H2O) and
2
(ΔH = 27.1 kcal.mol−1 (ΔH = 29.2 kcal.mol−1
) and 3 ) than for
compounds 5 (ΔH = 30.7 kcal.mol−1), 7 (ΔH = 30.2 kcal.mol−1), and
9 (ΔH = 30.1 kcal.mol−1). This increase in the relative intensity of G in
compounds 2 and 3 compared with compounds 5, 7, and 9 can be due
to the methyl group hyperconjugative effect in compound 3 and to
the methoxy group electron-release mesomeric effect in compound 2,
which contribute to stabilizing the vinylic cation G1.
C
(A1–C6HR1R2R3R4R5), which are stabilized by resonance, are
competitively formed from A1’. However, the peak corresponding to
C is more intense than that corresponding to B in the product ion
spectra of protonated MKCs 1–9. This difference between the
intensities of B and C can be interpreted in terms of the ΔH values
estimated for pathways IIIa and IIIb, which revealed that formation of
C from A1’ (pathway IIIa, ΔH between −8.2 and 5.6 kcal.mol−1) is
energetically more favored than formation of A1” (pathway IIIb, ΔH
between 0.0 to 5.2 kcal.mol−1) for all the analyzed MKCs.
Nazarov cyclization in protonated curcuminoids has been
reported to occur in solution and in the gas phase.11 Cyriac and
co-workers established that ketene and arene losses from the
protonated MKCs 1,5-bis-(2-methoxyphenyl)-1,4-pentadien-3-one
and 1,5-bis-(2-hydroxyphenyl)-1,4-pentadien-3-one are preceded by
Nazarov cyclization and further aryl migration. The authors proposed
some possible transition states (TSs) involved in [MKC + H–arene]+
and [MKC + H–ketene]+ formation and supported their hypothesis by
For MKCs 1–9, E and F originate from C, as evidenced by the
MS3 experiments (see supporting information). E and F result from
FIGURE 4 Proton affinity (PA) values for monoketone curcuminoides 1–11. The PA values were calculated at the B3LYP/6-31+G(d,p) level.
All the PA values are in kcal.mol−1

10 of 17
VIEIRA ET AL.
two competitive ring contractions of cyclic ketone C1 (the most stable
tautomer in the equilibrium with C), which lead to elimination of CO
E
and 50.1–58.2 kcal.mol−1 and 33.6–43.6 kcal.mol−1 for F1,
respectively. These values fully agree with the experimental results, as
the intensity of E is greater than the intensity of F at Elab = 20 eV. In
the case of compound 7, ions E and F do not apear in the product ion
spectrum. A possible explanation for this is that formation of C, which
(28 Da) and ketene (42 Da), respectively.25
F has an unstable
structure and is spontaneously converted into F1 during geometry
optimization. The estimated ΔH and ΔG values for C1 conversion into
E and F1 are 25.9–30.7 kcal.mol−1 and 14.5–19.2 kcal.mol−1 for
leads
to
formation
of
E
(ΔH = 28.67 kcal.mol−1
)
and
SCHEME 3 Formation of product ions D, G, and H from protonated MKCs 1–11. Relative enthalpies (ΔH) and Gibbs energies (ΔG) values are
in kcal.mol−1

VIEIRA ET AL.
11 of 17
SCHEME 4 Formation of product ions B, C, E, and F from protonated MKCs 1–9. Relative enthalpies (ΔH) and Gibbs energies (ΔG) values are
in kcal.mol−1

12 of 17
VIEIRA ET AL.
SCHEME 5 Formation of product ions B, C, E, and F from compounds 10 and 11
F
(ΔH = 58.35 kcal.mol−1), competes with the formation of the
diagnostics ions and
(ΔH = 0.13 kcal.mol−1), which are
formation of B, C, E, and F for these compounds is due to a different
fragmentation pathway from that in compounds 1–9, as shown in
Scheme 5. In these cases, cyclization is followed by an 1,2-aryl
migration from C3 to C2 to produce A2. However, the A2 structure is
unstable and is spontaneously converted into A2’ by means of a
second 1,2-aryl migration from C3 to C3’ (Scheme 5). The A2’
structure has also been reported by Cyriac and co-workers to explain
S
U
energetically more favored (Scheme 6). This result also agrees with
the intensities in the product ion spectrum.
In the case of MKCs 10 and 11, the expected cyclic structure
resulting from Nazarov cyclization (A1’) is spontaneously converted
into A1 during geometry optimization. This led us to conclude that
SCHEME 6 Formation of diagnostic
product ions P, S, T, U, and V from
compounds 5, 7, and 9. Relative
enthalpies (ΔH) and Gibbs energies (ΔG)
values are in kcal.mol−1

VIEIRA ET AL.
13 of 17
SCHEME 7 Formation of diagnostic product ions
Q and R from compound 3. Relative enthalpies (ΔH) and
Gibbs energies (ΔG) values are in kcal.mol−1
SCHEME 8 Formation of diagnostic product ions J, L, O, K, and I. Relative enthalpies (ΔH) and Gibbs energies (ΔG) values are in kcal.mol−1
1,5-bis-(2-methoxyphenyl)-1,4-pentadien-3-one
fragmentation.11
Cyriac and co-workers suggested that Nazarov cyclization and
subsequent arene and ketene losses from protonated MKC 2 and
1,5-bis-(2-hydroxyphenyl)-1,4-pentadien-3-one occur. The latter
compounds bear methoxyl and hydroxyl groups at the para (C8 and
C7) and ortho positions, respectively.11 Because the CID energy that
is necessary to dissociate compound 2 is higher than that required to
fragment its ortho isomer, the authors proposed that the mechanisms
involved in the Nazarov cyclization of these two compounds differ
and depend on the position of the methoxy group. Here, we revisit
the MKC 2 fragmentation and investigate the fragmentation of
Formation of B1 ([M + H – H2O]+ is preceded by A2’ isomerization
into A2”, which is an energetically favored process (ΔH = 13.2 kcal.
mol−1, ΔG = 3.6–4.32 kcal.mol−1). On the other hand, formation of C,
D, E, and F involves arene elimination (Scheme 5). Formation of C1.1
from A2’ is energetically more favored for compound 10
(ΔH = 12.6 kcal.mol−1) than for compound 11 (ΔH = 25.1 kcal.mol−1).
Consequently, peaks corresponding to E and F, which are derived
from C, are more intense in the product ion spectrum of compound
10 than in that of compound 11 (Table 1).
SCHEME 9 Formation of diagnostic product ions
M from protonated MKCs 1, 3, 5, 7, and 9 and N from
compound 3. Relative enthalpies (ΔH) and Gibbs energies
(ΔG) values are in kcal.mol−1

14 of 17
VIEIRA ET AL.
SCHEME 10 Formation of diagnostic
product ions W from compound 10 and
Z from compounds 10 and 11
compound 11, which is the para isomer of 1,5-bis-(2-hydroxyphenyl)-
1,4-pentadien-3-one. Our results indicated that Nazarov cyclization
and subsequent ketene and arene losses from protonated 2 follow the
same fragmentation pathway as in compounds 1 and 3–9, thus
revealing that formation of B, C, E, and F does not depend on the
presence of an oxygen atom at the ortho position of the aromatic ring.
In contrast, calculations carried out at the B3LYP level of theory
demonstrated that the A1’ ion structure of compounds 10 and 11,
which display a hydroxyl group at the para position, is spontaneously
converted back into A1. This implies that, compared with compounds
1–9, formation of B, C, E, and F follows a different fragmentation
pathway in the case of compounds 10 and 11. On the other hand, our
theoretical calculations revealed that the A1’ structure corresponding
to 1,5-bis-(2-hydroxyphenyl)-1,4-pentadien-3-one, which displays a
hydroxyl group at the ortho position, is stable and is not converted
back into A1. These data evidenced that the position of the hydroxyl
group plays a key part in the fragmentation of MKCs. However, the
hydroxyl effect on the mechanism appears to be more related to the
formation of an intramolecular hydrogen bond than to the hydroxyl
hydrogen, as was previously suggested by Cyriac and co-workers.11 In
the case of 1,5-bis-(2-hydroxyphenyl)-1,4-pentadien-3-one, the
intramolecular hydrogen bond between the two hydroxyl groups at
the ortho position stabilizes the A1’ structure resulting from Nazarov
cyclization, so that its fragmentation is similar for MKCs 1–9. In the
case of compounds 10 and 11, this stabilizing interaction does not
occur – the hydroxyl group is now involved in an intramolecular
hydrogen bond with the methoxyl group at the adjacent carbon,
which causes A1’ to be converted back into A1.
from A, formation of U (m/z 233) from S, and formation of P from
G occur through a similar mechanism (Scheme 6). The ΔH values
involved in formation of U are 86.9, 88.8, and 89.2 kcal.mol−1 for
compounds 5, 7, and 9, respectively. The ΔH values for formation of
P are 60.5, 100.4, and 90.6 kcal.mol−1 for compounds 5, 7, and 9,
respectively. However, formation of P, S, and T violates the even-
electron rule, so it is disfavored by enthalpy.27–29. On the other hand,
formation of V and U from T and S agrees with the odd-electron rule.
This could explain, at least in principle, the higher relative intensity of
V and U than that of their precursors T and S. However, although
formation of U and V is in accordance with the odd-electron rule, the
ΔH values are still high because of the formation of bi-radicals. In the
case of compound 7, which has a bromine atom attached to the
aromatic ring, the higher relative intensity of S and U than that of
compounds 5 and 9 is probably due to the lower C–Br bond energy
than that of C–Cl.
Q (m/z 230) and R (m/z 215) are also diagnostic for a methyl
group at the aromatic ring for compound 3. These ions originate from
a mechanism that resembles that involved in T and V, as well as in
S and U formation, as previously discussed in this paper (Scheme 7).
The ΔH and ΔG values involved in formation of Q and R are 116.3
and 105.48 for Q, respectively, and 100.9 and 89.18 kcal.mol−1 for R,
respectively. Formation of J (m/z 118) from G (compounds 2 and 4),
formation of O (m/z 136 and 166) from H (compounds 4 and 6), and
formation of K from D (compounds 6 and 8) are diagnostic for the
methoxy groups at the aromatic rings of MKCs 2, 4, 6, and 8,
respectively. They originate from homolytic C10–O bond cleavage
and consequent elimination of a methyl radical (Scheme 8).
For compounds 6 and 8, CO elimination from the aromatic ring to
produce I (m/z 178) involves a ring contraction that produces K and
3.4
|
Formation of diagnostic ions
Product ions produced by homolytic C–X bond cleavage (X = Cl or Br)
and consequent X• elimination are diagnostic for the presence of Cl
and Br in the structures of compounds 5, 7, and 9. Schemes 2 and 6
illustrate formation of T from B, formation of V (m/z 215) from T, and
formation of P (m/z 102) from G. The ΔH values estimated for the
conversion of B into T for compounds 5, 7, and 9 are 90.7, 100.6, and
86.1 kcal.mol−1, respectively. Formation of S (m/z 268 and m/z 312)
SCHE ME 11 Formation of diagnostic ion Y from compound 10

VIEIRA ET AL.
15 of 17
consequently expels CO from the phenoxyl radical ketone form
(Scheme 8). Methyl radical eliminations followed by decarbonylation
have been previously reported for many compounds displaying
aromatic methoxy groups.8,27 Elimination of the methyl radical from
D to form K violates the even-electron rule.27-29 and is energetically
disfavored (ΔH = 95.6 kcal.mol−1 for compound 6 and 103.8 kcal.
mol−1 for compound 8) due to loss of aromaticity, which agrees with
its relative intensity in the product ion spectra. On the other hand,
formation of L (m/z 91) from H, which is diagnostic for the presence
of a methoxy group at the aromatic ring of compound 2, results from
a remote hydrogen rearrangement and consequent elimination of
CH2O (30 Da), as shown in Scheme 8. The estimated ΔH and ΔG
the distonic ion (m/z 76), which is diagnostic for compound 3.
Although the low ΔH and ΔG values favor formation of
N (ΔH = 103.5 kcal.mol−1 and ΔG = 91.0 kcal.mol−1), its relatively low
intensity in the product ion spectrum of curcuminoid 3 can be
attributed to competition with the formation of diagnostic ions Q and
R. In addition, M comes from D, whose formation from A is favored
compared with formation of H only for compounds 4, 6, and 8.
Consequently, most of the product ions originating from G present
low relative intensities in the product ion spectra (Table 1).
In the case of compound 10, formation of diagnostic ions W (m/z
145) from D and Z (m/z 117) from W involves the hydroxyl group at
C7, as shown in Scheme 10. For compound 10, conversion of D into
values for conversion of H into L are 24.6 and 12.3 kcal.mol−1
,
W ) and conversion of W into Z are
(ΔH = 35.6 kcal.mol−1
respectively. For compounds 4, 6, and 8, which bear more than one
methoxy in the aromatic ring, CH2O loss and consequent formation of
L do not occur. This is probably due to the presence of more than one
methoxy group at specific positions on the aromatic ring of these
compounds, which tends to favor other pathways (e.g., K and
O formations) in competition with formation of L.
endothermic (ΔH = 39.9 kcal.mol−1), which agrees with their relative
low intensity in the product ion spectrum. For compound 11, Z (m/z
117) is formed directly from D. However, the ΔH value involved in
formation of Z from compound 11 is lower (ΔH = −38.0 kcal.mol−1
)
than for the same ion in compound 10 (ΔH = 39.9 kcal.mol−1). This
difference may be associated with the difference between the
structures of the D2 intermediates for the two compounds. On the
other hand, product ion Y, which is diagnostic for compound 10,
Acetylene (C2H2) elimination concomitant to CO loss from
D
produces M, as shown in Scheme 9. Although no specific
substituent participates in formation of M, this ion arises only for
compounds 1, 3, 5, 7, and 9. This is possibly because formation of
M competes with formation of G. The ΔH values revealed that
formation of G (ΔH = 27.1–52.8 kcal.mol−1) is more energetically
originates from methyl radical removal from D (scheme 11).
Formation of Y from E is disfavored by enthalpy (ΔH = 75.5 kcal.
mol−1), which agrees with the low intensity of Y in the product ion
spectrum of compound 10.
favored than formation of
M
(ΔH = 118.09–124.54 kcal.mol−1).
Finally, X results from direct ketene (42 Da) elimination from A.
Formation of this product ion is preceded by conversion of A1 into
A2, which results from Nazarov cyclization, as shown in Scheme 12.
As with formation of B, migration of an aryl from C3 to C2
consequently forms A''' (Scheme 12). The ΔH values revealed that
formation of X is more energetically favored for compound 2 (ΔH =
−16.0 kcal/mol−1) than for compounds 1 (ΔH = 1.6 kcal/mol−1) and
3 (ΔH = 2.3 kcal/mol−1), which is consistent with the higher relative
intensity of X in the product ion spectrum of compound 2 than in
those of compounds 1 and 3. In this case, A2 is unstable and
therefore rearranges spontaneously to X for the three compounds, as
Conversion of D into M is more endothermic (ΔH = 118.6–124.5 kcal.
mol−1) and endergonic (ΔG = 96.3–103.8 kcal.mol−1) than formation
of G from D, which is consistent with the relative intensity of M in
the product ion spectrum. In addition, for the other compounds, the
substituents attached to the aromatic ring yield other product
ions in competition with formation of M. For example, formation of
K
(m/z 206) from D is diagnostic for compounds 6 and 8,
while formation of W (m/z 145) for compound 10 and formation of
Z (m/z 117) for compounds 10 and 11 compete with formation of
M from D. Further methyl radical (•CH3) elimination from M produces
SCHEME 12 Formation of
diagnostic product ion X from
compounds 1, 2, and 3. Relative
enthalpies (ΔH) and Gibbs energies
(ΔG) values are in kcal.mol−1

16 of 17
VIEIRA ET AL.
3. Park SJ, Garcia CV, Shin GH, Kim JT. Improvement of curcuminoid
can be seen in Scheme 12. For compounds 4, 6, and 8, which display
more than one methoxy group at the aromatic ring, the electron-
releasing mesomeric effect of these groups further stabilizes A'''
compared with the corresponding A2 of the other compounds, thus
increasing the energy barrier for its conversion into X. Consequently,
X does not appear in the product ion spectra of compounds 4, 6, and
8 obtained at Elab = 20 eV.
bioaccessibility from turmeric by
a nanostructured lipid carrier
system. Food Chem. 2018;12:25151-25157. https://doi.org/10.1016/
j.foodchem.2018.01.071
4. Vieira TM, Santos IA, Silva TS, Martins CHG, Crotti AEM.
Antimicrobial activity of monoketone curcuminoids against cariogenic
bacteria. Chem Biodivers. 2018;15(8):5-12. https://doi.org/10.1002/
cbdv.201800216
5. Ud Din Z, Fill TP, de Assis FF, et al. Unsymmetrical 1,5-diaryl-3-oxo-
1,4-pentadienyls and their evaluation as antiparasitic agents. Bioorg
Med Chem. 2013;22(3):1121-1127. https://doi.org/10.1016/j.bmc.
2013.12.020
4
|
CONCLUSIONS
6. Kumari N, Kulkarni A, Lin X, et al. Inhibition of HIV-1 by curcumin a, a
novel curcumin analog. Drug des Devel Ther. 2015;2015(9):5051-
5060. https://doi.org/10.2147/DDDT.S86558
Our results reinforce, at least in part, the fragmentation pathways
previously reported in the literature. In addition, we propose an
alternative fragmentation mechanism that can explain the
fragmentation of a wider diversity of curcuminoids that do not display
a hydroxyl group in their structures. The most intense product ions
are acylium ions resulting from hydrogen rearrangement or product
ions derived from Nazarov cyclization of the protonated molecule.
The presence and the position of a hydroxyl group at the aromatic
~
7. Carrao DB, Albuquerque NCP, Marques LMM, et al. In vitro
metabolism of artepillin C by rat and human liver microsomes. Planta
Med. 2017;83(8):737-745. https://doi.org/10.1055/s-0042-124359
8. Demarque DP, Crotti AEM, Vessecchi R, Lopes JL, Lopes NP.
Fragmentation reactions using electrospray ionization mass
spectrometry: An important tool for the structural elucidation and
characterization of synthetic and natural products. Nat Prod Rep.
2016;33(3):432-455. https://doi.org/10.1039/c5np00073d
ꢀ
9. Jiang H, Somogyi A, Jacobsen NE, Timmermann BN, Gang DR.
rings play
a key role in the Nazarov cyclization mechanism.
Analysis of curcuminoids by positive and negative electrospray
ionization and tandem mass spectrometry. Rapid Commun
Mass Spectrom. 2006;20(6):1001-1012. https://doi.org/10.1002/rcm.
2401
Thermochemical data estimated by computational chemistry helped
to confirm the identity of the ion structures and to understand the
experimental data, mainly with respect to the relative intensities of
the peaks in the product ion spectra. These data, in combination with
multiple-stage mass spectrometry (MSn) and accurate mass data, can
be further used to identify in vitro or in vivo metabolites obtained
from curcuminoids in biological samples by LC/ESI-MS/MS.
10. Ostrowski W, Sniecikowska L, Hoffmann M, Franski R.
Demethoxycurcumin-metal
comparison, with curcumin-metal complexes, as studied by ESI-
MS/MS. Spectros. 2013;2913(8): 1-8. https://doi.org/10.1155/
complexes:
Fragmentation
and
J
2013/749641
11. Cyriac J, Paulose J, George M, et al. The role of methoxy group in the
Nazarov cyclization of 1,5-bis-(2-methoxyphenyl)-1,4-pentadien-
3-one in the gas phase and condensed phase. J Am Soc Mass
Spectrom. 2014;25(3):398-409. https://doi.org/10.1007/s13361-
013-0785-8
ACKNOWLEDGEMENTS
The authors acknowledge grant numbers 2016/19272-9, 2011/
~
07623-8, and 2017/24856-2 provided by the Fundaçao de Amparo à
12. George M, Sebastian VS, Reddy PN, Srinivas R, Giblin D, Gross ML.
Gas-phase Nazarov cyclization of protonated 2-methoxy and
2-hydroxychalcone: An example of intramolecular proton-transport
catalysis. J Am Soc Mass Spectrom. 2009;20(5):805-818. https://doi.
org/10.1016/j.jasms.2008.12.017
~
Pesquisa do Estado de Sao Paulo (FAPESP). This study was
~
partially financed by Coordenaçao de Aperfeiçoamento de Pessoal de
Nível Superior - Brasil (CAPES) - Finance Code 001. RLTP thanks
the Conselho Nacional de Desenvolvimento Científico
e
13. Crotti AEM, Bronze-Uhle ES, Nascimento PG, et al. Gas-phase
fragmentation of gamma-lactone derivatives by electrospray
ionization tandem mass spectrometry. J Mass Spectrom. 2009;44(12):
1733-1741. https://doi.org/10.1002/jms.1682
Tecnológico (CNPq, Brazil, grant 313648/2018-2) for a research
fellowship.
14. Dias HJ, Baguenard M, Crevelin EJ, et al. Gas-phase fragmentation
reactions of protonated benzofuran- and dihydrobenzofuran-type
neolignans investigated by accurate-mass electrospray ionization
ORCID
Tatiana M. Vieira
Renato P. Orenha
Eduardo J. Crevelin
Ricardo Vessecchi
Renato L.T. Parreira
Antônio E.M. Crotti
https://orcid.org/0000-0002-9469-8053
https://orcid.org/0000-0001-7285-8098
https://orcid.org/0000-0001-8708-147X
https://orcid.org/0000-0001-8472-3190
https://orcid.org/0000-0002-5623-9833
https://orcid.org/0000-0002-1730-1729
tandem mass spectrometry.
J Mass Spectrom. 2018;54(1):35-46.
https://doi.org/10.1002/jms.4304
15. Dias HJ, Vieira TM, Crevelin EJ, Donate PM, Crotti AEM.
Fragmentation of 2-aroylbenzofuran derivatives by electrospray
ionization tandem mass spectrometry. J Mass Spectrom. 2017;52(12):
809-816. https://doi.org/10.1002/jms.4024
16. Aguiar GP, Crevelin EJ, Dias HJ, et al. Electrospray ionization tandem
mass spectrometry of labdane-type acid diterpenes. J Mass Spectrom.
2018;53(11):1086-1096. https://doi.org/10.1002/jms.4284
17. Dias HJ, Bento MVB, da Silva EH, et al. Gas-phase fragmentation
reactions of protonated cocaine: New details to an old story.
J Mass Spectrom. 2018;53(3):203-213. https://doi.org/10.1002/jms.
4053
REFERENCES
1. Liang G, Li X, Chen L, et al. Synthesis and anti-inflammatory activities
of mono-carbonyl analogues of curcumin. Bioorg Med Chem Lett.
2008;18(4):1525-1529. https://doi.org/10.1016/j.bmcl.2007.12.068
2. Zamrus SNH, Akhtar MN, Yeap SK, et al. Design, synthesis and
cytotoxic effects of curcuminoids on HeLa, K562, MCF-7 and
MDA-MB-231 cancer cell lines. Chem Cent J. 2018;12(1):1-10.
https://doi.org/10.1186/s13065-018-0398-1
18. Dias HJ, Stefani R, Tomaz JC, Vessecchi R, Crotti AEM.
Differentiation between 3,4- and 4,15-epoxyeudesmanolides by

VIEIRA ET AL.
17 of 17
electrospray ionization tandem mass spectrometry. J Anal Methods
in mass spectrometry. J Org Chem. 2008;73(9):3369-3376. https://
doi.org/10.1021/jo702464b
Chem. 2017;2017: 1-9. https://doi.org/10.1155/2017/7921867
19. Aguiar GP, Wakabayashi KAL, Luz GF, et al. Fragmentation of
plumeran indole alkaloids from Aspidosperma spruceanum by
electrospray ionization tandem mass spectrometry. Rapid Commun
Mass Spectrom. 2010;24(3):295-308. https://doi.org/10.1002/rcm.
4389
26. Sleno L, Volmer DA. Ion activation methods for tandem mass
spectrometry. J Mass Spectrom. 2004;39(10):1091-1112. https://doi.
org/10.1002/jms.703
27. Karni MA. The “even-electron rule”. Org Mass Spectrom. 1980;15(2):
53-64. https://doi.org/10.1002/oms.1210150202
20. Vessecchi R, Carollo CA, Lopes JNC, Crotti AEM, Lopes NP,
Galembeck SE. Gas-phase dissociation of 1,4-naphthoquinone
derivative anions by electrospray ionization tandem mass
spectrometry. J Mass Spectrom. 2009;44(8):1224-1233. https://doi.
org/10.1002/jms.1600
28. McLafferty FW, Turecek F. Interpretation of Mass Spectra. 4th
ed. Sausalito: University Science Books; 1993.
29. Schäfer M, Dray M, Springer A, Zacharias P, Meerholz K. Radical
cations in electrospray mass spectrometry: Formation of open-shell
species, examination of the fragmentation behavior in ESI-MSn and
reaction mechanism studies by detection of transient radical cations.
Eur J Org Chem. 2007;2007:5162-5174. https://doi.org/10.1002/
ejoc.20070019
21. Crevelin EJ, Possato B, Lopes JLC, Lopes NP. Precursor ion scan
mode-based strategy for fast screening of polyether ionophores by
copper-induced gas-phase radical fragmentation reactions. Anal
Chem.
2017;89(7):3929-3936.
https://doi.org/10.1021/acs.
analchem.6b02855
22. Bouchoux G. Gas-phase basicities of polyfunctional molecules. Part 1:
Theory and methods. Mass Spectrom Rev. 2007;26(6):775-835.
https://doi.org/10.1002/mas.20151
SUPPORTING INFORMATION
Additional supporting information may be found online in the
Supporting Information section at the end of this article.
23. Range K, Riccardi D, Cui Q, Elstner M, York DM. Benchmark
calculations of proton affinities and gas-phase basicities of molecules
important in the study of biological phosphoryl transfer. Phys Chem
Chem
Phys.
2005;7(16):3070-3079. https://doi.org/10.1039/
How to cite this article: Vieira TM, Orenha RP, Crevelin EJ,
et al. Electrospray ionization tandem mass spectrometry of
monoketone curcuminoids. Rapid Commun Mass Spectrom.
2020;e8699. https://doi.org/10.1002/rcm.8699
b504941e
ꢀ
24. Boyd R, Somogyi A. The mobile proton hypothesis in fragmentation
of protonated peptides: A perspective. J Am Soc Mass Spectrom.
2010;21(8):1275-1278. https://doi.org/10.1016/j.jasms.2010.04.017
25. Hu N, Tu YP, Liu Y, Jiang K, Pan Y. Dissociative protonation and
proton transfers: Fragmentation of α,β-unsaturated aromatic ketones