Detection of Neutral CO Lost During Ionic Dissociation Using Atmospheric Pressure Thermal Dissociation Mass Spectrometry (APTD-MS)

Article abstract of DOI:10.1007/s13361-018-2055-2

Elucidation of ion dissociation patterns is particularly important to structural analysis by mass spectrometry (MS). However, typically, only the charged fragments from an ion dissociation event are detected in tandem MS experiments; neutrals are not identified. In recent years, we have developed an atmospheric pressure thermal dissociation (APTD) technique that can be applied to dissociate ions at atmosphere pressure and thus provide one way to characterize neutral fragments. In this paper, we focus on the detection of neutral CO resulting from amino acid and peptide ion dissociation. In the first set of experiments, several protonated amino acids (e.g., + 1 ion of phenylalanine) were found to undergo loss of a neutral (s) of total mass 46?Da, a process leading to iminium ion formation. We successfully detected the neutral species CO by using a CO sensor, UV-Vis and MS analysis following selective CO trapping with a rhodium complex. The capture of CO from dissociation of protonated amino acids supports the assignment of the loss of 46?Da to neutral losses of CO and H₂O, rather than loss of formaldehyde or dihydroxycarbene, other possible fragmentation pathways that have been subject of debate for a long time. In a second experiment, we used the APTD method in combination with the CO detection technique, to demonstrate the formation of CO in the conversion of b ions to a ions during peptide ion dissociations. These results showed the potential of APTD in the elucidation of ion dissociation mechanisms, using simple home-built apparatus. [Figure not available: see fulltext.].

Full text of DOI:10.1007/s13361-018-2055-2

B American Society for Mass Spectrometry, 2018
J. Am. Soc. Mass Spectrom. (2018)
DOI: 10.1007/s13361-018-2055-2
RESEARCH ARTICLE
Detection of Neutral CO Lost During Ionic Dissociation
Using Atmospheric Pressure Thermal Dissociation Mass
Spectrometry (APTD-MS)
Pengyi Zhao,1 Travis White,1 R. Graham Cooks,2 Qinghao Chen,3 Yong Liu,3 Hao Chen1
1Department of Chemistry and Biochemistry, Ohio University, Athens, OH 45780, USA
2Department of Chemistry, Purdue University, 560 Oval Drive, West Lafayette, IN 47907, USA
3Department of Process and Analytical Chemistry and Department of Structural Chemistry, Merck Research Laboratories, Merck
& Co., Inc., Rahway, NJ 07065, USA
Abstract. Elucidation of ion dissociation patterns
is particularly important to structural analysis by
mass spectrometry (MS). However, typically, only
the charged fragments from an ion dissociation
event are detected in tandem MS experiments;
neutrals are not identified. In recent years, we
have developed an atmospheric pressure ther-
mal dissociation (APTD) technique that can be
applied to dissociate ions at atmosphere pressure
and thus provide one way to characterize neutral
APTD Coiled Tube
CO
Hot
•
•
Fragmented Ions
Carbon Monoxide
fragments. In this paper, we focus on the detection of neutral CO resulting from amino acid and peptide ion
dissociation. In the first set of experiments, several protonated amino acids (e.g., + 1 ion of phenylalanine) were
found to undergo loss of a neutral (s) of total mass 46 Da, a process leading to iminium ion formation. We
successfully detected the neutral species CO by using a CO sensor, UV-Vis and MS analysis following selective
CO trapping with a rhodium complex. The capture of CO from dissociation of protonated amino acids supports
the assignment of the loss of 46 Da to neutral losses of CO and H2O, rather than loss of formaldehyde or
dihydroxycarbene, other possible fragmentation pathways that have been subject of debate for a long time. In a
second experiment, we used the APTD method in combination with the CO detection technique, to demonstrate
the formation of CO in the conversion of b ions to a ions during peptide ion dissociations. These results showed
the potential of APTD in the elucidation of ion dissociation mechanisms, using simple home-built apparatus.
Keywords: Neutral fragment detection, Carbon monoxide, peptide and amino acid, Mass spectrometry
Received: 3 July 2018/Revised: 8 August 2018/Accepted: 9 August 2018
pay attention to the “lost” neutral fragments several decades
Introduction
ago. Analysis of neutral species by mass spectrometric tech-
niques became possible when neutralization-reionization
mass spectrometry (NRMS) [9–11] and neutral fragment
reionization (NfR) [12] were introduced, based on reionization
of the resulting neutrals by collision with another target (e.g.,
O2) via electron transfer or charge stripping. However, al-
though this type of technique is very powerful, it requires
specialized instrumentation and it is implemented in vacuum,
which limits its utility. Recently, we reported our efforts to
solve this problem by developing atmospheric pressure neutral
reionization mass spectrometry (APNR) [13–17]. In APNR,
analyte ions are thermally dissociated using atmospheric
onic dissociation is the usual basis for structural analysis by
I
MS [1–8]. However, a significant limitation is that only
charged fragments from ion dissociation events are detected
while neutral fragments are simply lost. Researchers began to
Electronic supplementary material The online version of this article (https://
doi.org/10.1007/s13361-018-2055-2) contains supplementary material, which
is available to authorized users.
Correspondence to: R. Graham Cooks; e-mail: cooks@purdue.edu, Qinghao
Chen; e-mail: qinghao_chen@merck.com, Hao Chen;
e-mail: chenh2@ohio.edu

P. Zhao et al.: Detection of Neutral Fragments from Ion Dissociation
pressure thermal dissociation (APTD) followed by online am-
anism through a computational study. Based on ab initio
and density functional theory calculations, the losses of
CO and H2O neutrals were shown to be both thermody-
namically and kinetically favored over the alternative
HCOOH or C(OH)2 loss process [31]. Furthermore, the
unimolecular chemistry of protonated amino acids was also
explored in detail by Wesdemiotis et al [1]. However,
direct characterization of CO neutral resulting from amino
acid ion dissociation has not previously been achieved and
that is the focus of the current study.
bient reionization using electrosonic spray ionization (ESSI) or
corona discharge. In this study, we report that neutral CO
species lost during fragmentation can be characterized by sev-
eral techniques including online CO sensor monitoring, and
offline UV-Vis or MS detection following complexation with a
binuclear rhodium complex. Detection of CO is significant as
CO was proposed to be produced from two important ion
dissociation pathways: the dissociation of protonated amino
acids to form iminium ions (loss of 46 Da) and the conversion
from peptide b ions to a ions (loss of 28 Da).
In this paper, we applied the APTD technique to disso-
ciate protonated amino acids such as + 1 ion of phenylal-
anine, generated by electrosonic spray ionization (ESSI, a
soft ionization method similar to electrospray ionization
ESI [34]), to generate both ionic and neutral fragments at
atmospheric pressure. As APTD operates outside the mass
spectrometer rather than in vacuum, it should be possible
to access the neutrals and characterize them. As shown in
Scheme 2, an amino acid (e.g., phenylalanine) was sprayed
by ESSI into a coiled and heated tube outside the mass
spectrometer where it underwent thermal dissociation to
produce fragment ions and neutrals. The coiled shape
increases gas turbulence, facilitating ion dissociation. Our
results showed that neither formic acid (HCOOH) nor
dihydroxycarbene C(OH)2 was detected during APTD of
phenylalanine. On the other hand, CO was successfully
detected in the process. This study provides direct evi-
dence to support the production of carbon monoxide in
the process for the formation of iminium ion from amino
acid ion dissociation. In addition, upon ion activation such
as collision-induced dissociation (CID), b ions are often
produced as a result of peptide amide backbone cleavage.
It is also well accepted that a ions are generated from b
ions via further dissociation by loss of CO (Scheme 3)
[35–38]. However, CO was not characterized previously
in this ion dissociation process. This experiment also was
Although gas-phase ion dissociation of protonated amino
acids has been extensively studied under a variety of ioni-
zation conditions and methods, [18–27] the mechanism for
the formation of iminium ion during amino acid or peptide
ion dissociation [28–30] has been the subject of debate.
Three main mechanisms have been previously proposed in
different studies (see Scheme 1 (a) losses of H2O and CO;
(b) loss of dihydroxycarbene, C(OH)2, and (c) loss of
formic acid, HCOOH [31]. However, none of these studies
have given unambiguous experimental evidence regarding
the structure of the lost neutrals. In the report by Harrison
et al., which proposed pathway a (Scheme 1), the authors
demonstrated that the reaction was analogous to the retro-
Koch reaction and required a translocation of the proton
from nitrogen to the OH group with a subsequent loss of CO
and H2O [27]. Pathway b (Scheme 1), proposed to produce
C(OH)2, was supported by Kulik and coworkers via meta-
stable ion and collisional activation fast atom bombardment
(FAB) technique [32]. Unfortunately, the result showed no
evidence for an intact C(OH)2 neutral product. Pathway c
(Scheme 1) was proposed by Meot-Ner and Field, and
formic acid was suggested to be produced. However, this
mechanism requiring a four-centered transition state was
expected to be thermodynamically unfavorable [33]. Later,
O’Hair and coworkers explored this ion dissociation mech-
a)
b)
- C(OH)2
c)
Scheme 1. Three proposed amino acid fragmentation pathways toward the formation of iminium ion (illustrated using phenylal-
anine as an example): (a) losses of CO and H2O molecules, (b) loss of dihydroxycarbene C(OH)2, and (c) loss of formic acid HCOOH

P. Zhao et al.: Detection of Neutral Fragments from Ion Dissociation
1
ESSI Probe
APTD Coiled Tube
Hot
Sample
soluꢀon
2
3
UV-Vis
MS
Fragment Ion
Carbon Monoxide
Rhodium
Complex
Scheme 2. Schematic drawing of the APTD apparatus and processes: (1) direct detection of ionic fragments from the APTD coiled
tube by MS, (2) use of CO sensor to directly detect carbon monoxide emerging from the APTD coiled tube, and (3) use a binuclear
rhodium complex solution to trap CO molecules followed with offline characterization by MS and/or UV/Vis
attempted to provide evidence for this widely accepted
assumption.
stainless steel tube was heated to 300 °C. The temperature
of the APTD tube was measured using an infrared thermom-
eter (Tenma, Taiwan). The sample solution was sprayed
from an ESSI source operated at the flow rate of 10 μL/
min with the assistance of + 5 kV voltage and 170 psi
nebulizing N2 gas.
Experimental
General Materials
Phenylalanine, leucine, histidine, tryptophan, 13C-labeled phe-
nylalanine, phenethylamine, and rhodium acetate dimer were
purchased from Sigma-Aldrich (St. Louis, MO). Gly-His-Gly
and Gly-Trp-Gly peptides were bought from MP Biomedicals,
LLC (Solon, OH) and Chem-Impex Int’l Inc., (Wood Dale, IL),
respectively. HPLC-grade acetonitrile was purchased from
Fisher Scientific (Hampton, NH). Acetic anhydride was pur-
chased from Spectrum Chemical (New Brunswick, NJ). Am-
monium hydroxide was purchased from GFS Chemicals (Pow-
ell, OH). Deionized water for sample preparation was obtained
from a Nanopure Diamond Barnstead purification system
(Barnstead International, Dubuque, IA).
Synthesis of Binuclear Rhodium Complex Cis-[-
Rh2(C6H4PPh2)2(O2CCH3)2](HOAc)2
The synthesis followed a reported procedure [43]. A 100-mg
quantity of rhodium acetate dimer Rh2(O2CCH3)4 and 150 mg
of triphenylphosphine were added into 10 mL of acetic acid in a
flask under a nitrogen gas atmosphere. The mixture solution
was refluxed for 2 h during which the blue coloration was
replaced by red-brown, and then a dark purple color. The
solution was concentrated to half of its initial volume and the
purple solid deposit was filtered off and washed by ACN. The
1H-NMR for this compound was δ 1.17 (s, [O2CCH3], 6H),
2.15 (S, CH3COOH, 6H), 6.4–7.7 (m, [(C6H5)2P(C6H4)], 28H)
(Signal of the carboxylic acid proton of CH3COOH was not
observed due to proximity to Rh).
APTD-MS Setup
The APTD device was home-made, as a one-loop stainless
steel tube (3.2 mm o.d., 1.6 mm i.d., 20 cm length, and
2.5 cm loop diameter) wrapped by heating tape [17]. Either
a LCQ DECA XP MAX (Thermo Finnigan, San Jose, CA)
or a Thermo Fisher Orbitrap QE Plus mass spectrometer
(San Jose, CA) was used to analyze the APTD-produced
ions. A 90-V voltage was applied to the heating tape and the
MS Analysis of Carbon Monoxide Trapped
by Binuclear Rhodium Complex Solution
A 25-mM phenylalanine in ACN/H2O/HOAc solution was
sprayed by an ESSI source and dissociated in the APTD tube,
and the species flowing out of the tube was then directed into a
+ CO
bn
an
Scheme 3. Scheme showing dissociation from peptide (b) ion to a ion by the loss of neutral CO. Note that a and b ions are drawn in
the way shown in their original definitions; their true structures may well be cyclic [39–42]

P. Zhao et al.: Detection of Neutral Fragments from Ion Dissociation
trap containing 5 mL of 3 mM dinuclear rhodium complex
at this temperature. Only a small peak at m/z 120 was seen,
corresponding to the iminium ion [Phe + H-CO2H2]+ from dis-
sociation of [Phe + H]+ via loss of a neutral (s) with chemical
formula CO2H2 (total mass 46 Da). When the temperature was
increased, the intensity of [Phe + H-CO2H2]+ peak (m/z 120)
increased as well. At 300 °C, the [Phe + H-CO2H2]+ peak (m/z
120) was abundant and the [Phe + H]+ peak (m/z 166) had
disappeared from the spectrum. As expected, when phenylala-
nine was replaced by 13C-labled phenylalanine sample (the
carboxylic carbon is labeled with 13C), a major fragment ion
[13C-Phe + H-13CO2H2]+ (m/z 120) was also observed, resulting
from dissociation of [13C-Phe + H]+ via loss of a neutral (s) with
chemical formula 13CO2H2 (total mass 47 Da, Fig. 1S,
Supporting Information). Indeed, in this ion dissociation exper-
iment, although the fragment ion of m/z 120 was observed, the
corresponding neutrals were not detected, and they could be
either CO + H2O, HCOOH or dihydroxycarbene C(OH)2.
solution in chloroform for 3 min. The amino acid sample
injection flow rate was 10 μL/min and the nebulizing gas
pressure used was 100 psi. After trapping, the solution was
diluted by ACN with 0.1% HOAc (1:1 dilution), and then
subject to ESI-MS analysis. The ESI-MS spectra were collect-
ed by using a Thermo Fisher Orbitrap QE Plus mass spectrom-
eter (San Jose, CA) in the positive ion mode.
UV-Vis Analysis of Carbon Monoxide Trapped
by Binuclear Rhodium Complex Solution
The dinuclear rhodium complex solution (3 mL, 1 mM) was
used to capture the carbon monoxide from the phenylalanine
APTD dissociation. The trapping period was 3 min for each
sample. The amino acid sample injection flow rate for APTD
was 10 μL/min and the nebulizing gas pressure used was
100 psi. The UV spectra were collected by using an Agilent
8453 UV-visible spectrophotometer (Santa Clara, CA).
Detection of Carbon Monoxide by Sensor
Loss of a neutral fragment (s) with mass 46 Da from [Phe + H]+
(m/z 166) yielded the fragment ion [Phe + H-CO2H2]+ (m/z
120) upon APTD of phenylalanine at 300 °C. O’Hair et al.
[31] proposed that protonated aliphatic amino acids could lose
neutral H2O and CO (the combined mass is 46 Da) upon
fragmentation in the gas phase, based on computation. Previ-
ously established methods for carbon monoxide detection in-
clude gas chromatography [44], carboxy-myoglobin (Mb-CO)
assay [45, 46], electrochemical assays [47, 48], colorimetric
Results and Discussion
APTD of Phenylalanine
The dissociation of the amino acid phenylalanine (Phe) by
APTD was first investigated at several different temperatures
(Fig. 1). At relatively low temperature (45 °C, APTD coiled
tube temperature), the [Phe + H]+ peak (m/z 166) was domi-
nant, which meant that most of [Phe + H]+ had not dissociated
100
(a) 45 °C
80
[Phe+H]+
166.07
60
40
20
[Phe+H-CO2H2]+
120.13
0
100
(b) 180 °C
80
60
40
20
[Phe+H]+
166.07
[Phe+H-CO2H2]+
120.07
0
100
80
60
40
20
0
(c) 300 °C
[Phe+H-CO2H2]+
120.07
40
60
80
100
120
140
160
180
200
m/z
Figure 1. APTD–MS spectra for phenylalanine at (a) 45 °C, (b) 180 °C, and (c) 300 °C. 0.25 mM phenylalanine solution in ACN/H2O/
HOAc (50:50:0.1 by volume) was sprayed by ESSI (spray voltage: + 5 kV) into a hot coiled tube at a flow rate of 10 μL/min

P. Zhao et al.: Detection of Neutral Fragments from Ion Dissociation
CO sensing [49, 50], and laser infrared absorption methods
[51]. In this study, a Sensorcon carbon monoxide detector, a
household product for monitoring the presence of CO due to
safety concerns, was used to detect CO generated in the amino
acid APTD process. The detection mechanism of the CO
sensor is that CO is oxidized on a platinum working electrode.
The current generated from the electrochemical reaction gives a
measure of the concentration of CO in the atmosphere [52–54].
In this experiment, the CO detector was held about 1 cm from
the outlet of APTD tube. All of the blank solvents and samples
were measured in triplicates (Fig. 2). While CO was not
detected by the sensor from APTD of the blank solvent
(Fig. 2(1)), phenylalanine sample did show a positive reading
from CO detector (15 ppm on average, Fig. 2(3)). In addition,
phenethylamine, structurally similar to phenylalanine except
for the absence of a carboxylic acid group, was also sprayed by
ESSI and underwent dissociation in APTD tube. As expected,
no CO was detected by the sensor reading (Fig. 2(2)). As
phenethylamine and phenylalanine only differ in a terminal
carboxylic group, the CO sensor detection result strongly sug-
gested that CO was produced from the loss of the carboxylic
group of phenylalanine.
·(HOAc)2
·(CO, HOAc)
Scheme 4. Selective binding of CO by the binuclear rhodium
complex 1∙(HOAc)2. The complex product 1∙(CO, HOAc) forms
after CO binding
UV-Vis absorption peak at 480 nm can be seen. This complex is
highly selective to CO, and does not bind other gases like CO2,
N2, O2, or Ar [49].
In this experiment, the UV-Vis spectrum of the pure dinu-
clear rhodium complex 1∙(HOAc)2 in chloroform (1 mM) was
measured first and the absorption at 480 nm was low (line a,
Fig. 3). When 25 mM phenylalanine in ACN/H2O/HOAc
(50:50:0.1) was sprayed into the coiled tube at 300 °C for
APTD and the gaseous species flowing out of the APTD tube
was bubbled through the trapping solution of dinuclear rhodi-
um complex 1∙(HOAc)2 in chloroform (1 mM), the UV-Vis
spectrum of the resulting solution showed a pronounced in-
crease in absorption at 480 nm (line d, Fig. 3). The duration for
the CO collection was 3 min. The result indicates the formation
of CO during APTD of phenylalanine. Several control exper-
iments were also conducted. When no phenylalanine was in the
UV-Vis Analysis of Carbon Monoxide after Binding
with a Binuclear Rhodium Complex
In order to more thoroughly examine our hypothesis, a binuclear
rhodium complex cis-[Rh2(C6H4PPh2)2(O2CCH3)2](HOAc)2,
1
(HOAc)2, was chosen to trap the APTD-generated carbon mon-
oxide gas, and the resulting complex 1 (CO, HOAc) was then
analyzed by UV-Vis spectroscopy. This binuclear rhodium com-
plex was first synthesized by Cotton et al. [43], and further
developed into a colorimetric sensing method by Esteban et al
[49, 50]. The molecule contains two cyclometalated phosphine
ligands and two acetic acid molecules attached to the two rhodium
atoms. When this binuclear rhodium complex is exposed to
carbon monoxide gas, one acetic acid molecule can be replaced
by a CO molecule (Scheme 4). Upon CO binding, a characteristic
2.0
Rh
a
Rh + solvent
b
Rh + amine
c
Rh + CO
d
1.5
1.0
0.5
0.0
Avg. Readings
#
1
Stucture
Triplicated Trials
(ppm)
400
500
600
700
800
blank solvent
0
λ
(nm)
Figure 3. UV-vis spectra of (a) 1 mM dinuclear rhodium com-
plex 1∙(HOAc)2 in chloroform (black line a), (b) 1 mM dinuclear
rhodium complex 1∙(HOAc)2 in chloroform after trapping gas-
eous species flowing out of the APTD when a blank solvent
ACN/H2O/HOAc (50:50:0.1) was sprayed into the APTD tube at
300 °C (red line b), (c) 1 mM dinuclear rhodium complex
1∙(HOAc)2 in chloroform after trapping species flowing out of
the APTD when 25 mM phenethylamine in ACN/H2O/HOAc
(50:50:0.1) was sprayed into the APTD tube at 300 °C (blue line
c), and (d) 1 mM dinuclear rhodium complex 1∙(HOAc)2 in chlo-
roform after trapping species flowing out of the APTD when
25 mM phenylalanine in ACN/H2O/HOAc (50:50:0.1) was
sprayed into the APTD tube at 300 °C (orange line d)
2
3
0
15
Figure 2. Measurements of carbon monoxide produced from
APTD of (1) ACN/H2O/HOAc (50:50:0.1) blank solvent, (2)
25 mM phenethylamine in ACN/H2O/HOAc (50:50:0.1), (3)
25 mM phenylalanine in ACN/H2O/HOAc (50:50:0.1). All of
measurements were made in triplicate as shown

P. Zhao et al.: Detection of Neutral Fragments from Ion Dissociation
spray solvent for APTD, no increase in 480 nm absorption was
786.99 corresponding to the binuclear rhodium complex cation
[1-OAc]+ (theoretical mass 786.9909, observed mass
786.9882, mass error 3.43 ppm), was detected. Notably, when
the gaseous species flowing out of the heated tube after APTD
of phenylalanine was trapped by the dinuclear rhodium com-
plex 1∙(HOAc)2, analysis of the rhodium complex revealed a
new peak at m/z 814.99 (Fig. 4d), corresponding to [1-OAc +
CO]+ (theoretical mass 814.9859, observed mass 814.9852,
mass error 0.86 ppm). CID of m/z 814.99 gave rise to fragment
ions at m/z 786.99 and 726.97 (Fig. 2S-a, Supporting Informa-
tion), by consecutive losses of CO and HOAc, consistent with
its structural assignment. Furthermore, when standard carbon
monoxide gas was bubbled into the dinuclear rhodium complex
1∙(HOAc)2, the adduct ion [1-OAc + CO]+ (m/z 814.99) was
also observed (Figure 4e). These results show that APTD of
phenylalanine does produce CO. Meanwhile, when only solvent
or the structurally similar compound phenethylamine was
sprayed for APTD and subsequent trapping by the rhodium
complex, no peak at m/z 814.99 was detected (Fig. 4b, c).
noted (line b, Fig. 3). Furthermore, when phenethylamine
solution was sprayed for APTD, no change in 480 nm absorp-
tion was detected (line c, Fig. 3). Similar to the CO sensor
measurement result mentioned above, these UV-Vis data
showed that CO is formed from APTD of phenylalanine and
originates from the carboxylic acid group of the amino acid.
MS Analysis of Carbon Monoxide after Binding
with Binuclear Rhodium Complex
MS should be an accurate, sensitive, and reliable tool to detect
small molecules like carbon monoxide. However, the mass of
CO is smaller than the cutoff mass of our mass spectrometers.
Therefore, direct online analysis of CO from APTD was not an
option in this case. We reasoned that CO could be detectable
after selective trapping with the binuclear rhodium complex.
Figure 4a displays the ESI-MS spectrum of 3 mM dinuclear
rhodium complex 1∙(HOAc)2 in chloroform. The peak of m/z
786.99
100
[1-OAc]+
50
(a)
(b)
0
786.99
100
[1-OAc]+
50
0
786.99
[1-OAc]+
100
50
(c)
0
786.99
100
[1-OAc]+
[1-OAc+CO]+
50
814.99
(d)
0
814.99
100
[1-OAc+CO]+
[1-OAc]+
786.99
50
(e)
0
785
790
795
800
805
810
815
820
m/z
Figure 4. ESI-MS spectra of (a) 3 mM dinuclear rhodium complex 1∙(HOAc)2 in chloroform, (b) 3 mM dinuclear rhodium complex
1∙(HOAc)2 in chloroform after trapping gaseous species flowing out of the APTD tube when a blank solvent of ACN/H2O/HOAc
(50:50:0.1) was sprayed into the APTD tube at 300 °C, (c) 3 mM dinuclear rhodium complex 1∙(HOAc)2 in chloroform after trapping
gaseous species flowing out of the APTD tube when 25 mM phenethylamine in ACN/H2O/HOAc (50:50:0.1) was sprayed into the
APTD tube at 300 °C, (d) 3 mM dinuclear rhodium complex 1∙(HOAc)2 in chloroform after trapping gaseous species flowing out of the
APTD tube when 25 mM phenylalanine in ACN/H2O/HOAc (50:50:0.1) was sprayed into the APTD tube at 300 °C, (e) 3 mM dinuclear
rhodium complex 1∙(HOAc)2 in chloroform after trapping standard carbon monoxide gas (purity 99.99%) for 3 min

P. Zhao et al.: Detection of Neutral Fragments from Ion Dissociation
In addition, a 13C-labled phenylalanine sample (the carbox-
ylic carbon is labeled with 13C) was subjected to APTD, and
the rhodium complex trapping, a peak of m/z 815.99, corre-
sponding to [1-OAc + 13CO]+ (theoretical mass 815.9879,
observed mass 815.9883, mass error 0.49 ppm) was detected,
as expected (Fig. 3S, Supporting Information). CID of m/z
815.99 produced fragment ions at m/z 786.99 and 726.97
(Fig. 2S-b, Supporting Information), by consecutive losses of
13CO and HOAc, consistent with its structural assignment. This
result supported the hypothesis that APTD of phenylalanine
produces CO and the CO stems from the carboxylic group of
the amino acid.
solution. After trapping and subsequent ESI-MS analysis, no
peak was detected at m/z 814.99 (Fig. 6S, Supporting
Information).
APTD of Other Amino Acids
Other amino acids (leucine, histidine, and tryptophan) were
also tested for APTD in our study. The amino acids were
dissociated first by APTD and the loss of 46 Da from the
protonated amino acid was seen in all cases (Fig. S7,
Supporting Information). Then the released neutral CO mole-
cules were trapped by rhodium complex solution and success-
fully detected by MS (Fig. S8, Supporting Information). This
information further revealed that CO released from the disso-
ciation of protonated amino acid into iminium ion is a quite
general pathway.
Investigation of Other Possible Neutral Species
from APTD
At the same time, the possibility of producing
dihydroxycarbene C(OH)2 (pathway b, Scheme 1) was also
investigated. Due to the high reactivity of carbene, the
dihydroxycarbene would react with water from the spray sol-
vent via insertion into the O-H bond of H2O, [55] and a product
peak at m/z 65, corresponding to [HC (OH)3 + H]+, should be
detected by the APTD-MS in the positive ion mode. However,
there was no such peak in the acquired MS spectrum, which
suggested that dihydroxycarbene was not formed from APTD
of phenylalanine (Fig. 4S, Supporting Information).
Furthermore, the possibility of producing formic acid
(HCOOH) following APTD of the amino acid (pathway
c, Scheme 1) was also examined. To detect neutral
species HCOOH, we reasoned that HCOOH could be
re-ionized into formate ion HCOO− (m/z 45) in the
negative ion mode, by using the atmospheric pressure
neutral re-ionization (APNR) technique. [17] Thus, in
our APTD experiment of phenylalanine, a second ESSI
sprayer was added downstream so as to spray solvent
ACN/H2O/ammonia (50:50:0.1). The generated
microdroplet could capture neutral species flowing out
of the ATPD tube and get them ionized. The re-ionized
neutrals could be detected by the nearby mass spectrom-
eter (Scheme 1S, Supporting Information). When only
solvent ACN/H2O/HOAc (50:50:0.1) was sprayed into
the APTD tube at 300 °C, APNR–MS spectrum
(Fig. 5S-a) shows a background peak at m/z 45 with
intensity of 5.09 × 103 (manufacturer’s arbitrary unit of
counts). When phenylalanine was added into the solvent
for APTD, the peak of m/z 45 did not show a significant
increase (peak intensity 5.29 × 103, Fig. 5S-b). This result
suggests that no HCOOH was formed from APTD of
phenylalanine.
APTD of Peptides
The APTD method can also be used to dissociate peptides.
Two peptides, Gly-His-Gly (GHG), and Gly-Trp-Gly (GWG),
were chosen as test samples in this experiment. Compared to
APTD dissociation of amino acids, loss of 46 Da from the two
peptide ions was not observed (Fig. 5). However, both a and b
ions are seen in the two cases. It has been long regarded that a
ions originate from b ion, by facile loss of CO. However, no
direct experimental evidence is reported in the literature, to our
knowledge. Thus, by simply carrying out trapping of gaseous
species from APTD of these two peptides using the dinuclear
rhodium complex, we were able to detect the characteristic
adduct ion [1-OAc + CO]+ (m/z 814.99, Fig. 6). This data
provides evidence to corroborate the formation of CO during
the dissociation of these peptide ions, which is most likely
ascribed to the conversion of b into a ions. Therefore, a ions
should arise from further dissociation of b ions, rather than
from direct dissociation of precursor peptide ions, as no CO
would be formed in the latter case. (Note that under some
circumstances, direct formation of CO during a ion and y ion
formation is possible [57, 58]; however, this appears not to be
the case for the peptides that were examined in this study, as
neither [peptide-CO]+ or y1 ions was detected during the APTD
process (Fig. 5)).
Furthermore, the approach of CO neutral characteriza-
tion of this study can be used in a more quantitative
manner. A quantitative comparison experiment was run
to compare the amount of CO produced from APTD of
peptides GHG and GWG. As measured by the CO sen-
sor, the averaged values from APTD of GHG and GWG
were 7 ppm and 8.3 ppm, respectively (Table 1S,
Supporting Information). The result showed that the dis-
sociation of GWG generated more CO than GHG under
the same experimental conditions. Likewise, as shown in
Fig. 6b, c, the relative intensity of the adduct ion [1-
OAc + CO]+at m/z 814.99 (relative to the binuclear rho-
dium complex cation [1-OAc]+ at m/z 786.99) was
higher in the case of GWG than that for GHG, which
is in line with the CO sensor measurement result. As
As formic acid HCOOH could decompose into CO and
H2O [56] at elevated temperature, we also performed an exper-
iment to exclude the possibility that the detected CO from
APTD of phenylalanine was a HCOOH decomposition prod-
uct. In a separate experiment, HCOOH was directly sprayed
into APTD tube at 300 °C and the gaseous species coming out
of the tube was trapped by the dinuclear rhodium complex

P. Zhao et al.: Detection of Neutral Fragments from Ion Dissociation
235.08
[GHG+H-H2O-NH3]+
100
(a)
50
b2
195.09
b -NH3
2
[GHG+H]+
270.12
a2
167.09
178.06
0
160
180
200
220
240
260
280
300
m/z
a2
216.11
100
(b)
50
b2
244.27
[GWG-H2O+H]+
301.13
[GWG+H]+
319.14
0
220
240
260
280
300
320
m/z
Figure 5. APTD-MS spectra of (a) 5 mM GHG in ACN/H2O/HOAc (50:50:0.1) and (b) 5 mM GWG in ACN/H2O/HOAc (50:50:0.1). The
temperature of the APTD tube was 300 °C
shown in Fig. 5, the a ion appears to be dominant in the
supports the assumption that CO was produced along
case of GWG upon APTD; therefore, this result also
with generation of the a ion for the peptides studied.
x10
100
80
786.99
787.99
(a)
60
40
20
0
100
786.99
787.99
(b)
80
60
40
20
814.99
0
100
786.99
787.99
(c)
80
60
40
20
0
814.99
785
795
805
815
m/z
Figure 6. ESI-MS spectra of 3 mM dinuclear rhodium complex 1∙(HOAc)2 solution after trapping neutral species flowing out of the
APTD tube when (a) ACN/H2O/HOAc (50:50:0.1), (b) 5 mM GHG in ACN/H2O/HOAc (50:50:0.1), and (c) 5 mM GWG in ACN/H2O/
HOAc (50:50:0.1) was sprayed into the tube for APTD at 300 °C

P. Zhao et al.: Detection of Neutral Fragments from Ion Dissociation
H) ions in the gas phase and the generation of their neutral counterparts by
neutralization-reionization mass spectrometry. J. Am. Soc. Mass
Spectrom. 6, 1143–1153 (1995)
Conclusions
This study demonstrates a new application of APTD in inves-
tigating the amino acid and peptide ion dissociation mecha-
nisms, specifically identification of the neutral loss species CO
involved in dissociation of protonated amino acid into iminium
ion and peptide ion b ➔a conversion. Due to the fact that ion
dissociation in APTD occurs at atmospheric pressure, charac-
terization of neutral species becomes possible, as demonstrated
using either online CO sensor or offline spectroscopic/mass
spectrometric tools. Our results demonstrate that the fragmen-
tation of protonated amino acids into iminium ions involves
losses of CO and H2O rather than loss of HCOOH or C(OH)2.
Detection of CO from the tested peptide ions demonstrates that
a ion is most likely to come from b ion dissociation by loss of
CO. Our study once again suggests that APTD has utility for
structure analysis and elucidation of ion dissociation mecha-
nism. Due to the neutral fragment characterization could pro-
vide increased structural information, this method could be
used to elucidate complicated ion dissociation mechanisms,
such as the investigation of cyclic peptide or glycan ion disso-
ciation behaviors, which could be helpful to cyclic peptide
sequencing and glycan structural analysis.
12. Polce, M.J., Nold, M.J., Wesdemiotis, C.: Characterization of neutral
fragments in tandem mass spectrometry: a unique route to mechanistic
and structural information. J. Mass Spectrom. 31, 1073–1085 (1996)
13. Chen, H., Eberlin, L.S., Nefliu, M., Augusti, R., Cooks, R.G.: Organic
reactions of ionic intermediates promoted by atmospheric-pressure ther-
mal activation. Angew. Chemie Int. Ed. 47, 3422–3425 (2008)
14. Eberlin, L.S., Xia, Y., Chen, H., Cooks, R.G.: Atmospheric pressure
thermal dissociation of phospho- and sulfopeptides. J. Am. Soc. Mass
Spectrom. 19, 1897–1905 (2008)
15. Chen, H., Eberlin, L.S., Cooks, R.G.: Neutral fragment mass spectra via
ambient thermal dissociation of peptide and protein ions. J. Am. Chem.
Soc. 129, 5880–5886 (2007)
16. Liu, P., Cooks, R.G., Chen, H.: Nuclear magnetic resonance structure
elucidation of peptide b 2 ions. Angew. Chemie Int. Ed. 54, 1547–1550
(2015)
17. Liu, P., Zhao, P., Cooks, R.G., Chen, H.: Atmospheric pressure neutral
reionization mass spectrometry for structural analysis. Chem. Sci. 8,
6499–6507 (2017)
18. Thomas, D.A., Sohn, C.H., Gao, J., Beauchamp, J.L.: Hydrogen bonding
constrains free radical reaction dynamics at serine and threonine residues
in peptides. J. Phys. Chem. A. 118, 8380–8392 (2014)
19. Sohn, C.H., Gao, J., Thomas, D.A., Kim, T.-Y., Goddard III, W.A.,
Beauchamp, J.L.: Mechanisms and energetics of free radical initiated
disulfide bond cleavage in model peptides and insulin by mass spectrom-
etry. Chem. Sci. 6, 4550–4560 (2015)
20. Tureček, F., Julian, R.R.: Peptide radicals and cation radicals in the gas
phase. Chem. Rev. 113, 6691–6733 (2013)
21. Ly, T., Julian, R.R.: Residue-specific radical-directed dissociation of
whole proteins in the gas phase. J. Am. Chem. Soc. 130, 351–358 (2007)
22. Yang, L., Mazyar, O.A., Lourderaj, U., Wang, J., Rodgers, M.T., Martí-
nez-Núñez, E., Addepalli, S.V., Hase, W.L.: Chemical dynamics simula-
tions of energy transfer in collisions of protonated peptide−ions with a
perfluorinated alkylthiol self-assembled monolayer surface. J. Phys.
Chem. C. 112, 9377–9386 (2008)
23. Benninghoven, A., Sichtermann, W.K.: Detection, identification, and
structural investigation of biologically important compounds by second-
ary ion mass spectrometry. Anal. Chem. 50, 1180–1184 (1978)
24. Milne, G.W.A., Axenrod, T., Fales, H.M.: Chemical ionization mass
spectrometry of complex molecules. IV. Amino acids. J. Am. Chem.
Soc. 92, 5170–5175 (1970)
25. Bouchonnet, S., Denhez, J.P., Hoppilliard, Y., Mauriac, C.: Is plasma desorp-
tion mass spectrometry useful for small-molecule analysis? Fragmentations of
the natural.alpha.-amino acids. Anal. Chem. 64, 743–754 (1992)
26. Leclercq, P.A., Desiderio, D.M.: Chemical ionization mass spectra of
amino acids and derivatives. Occurrence and fragmentation of ion-
molecule reaction products. Org. Mass Spectrom. 7, 515–533 (1973)
27. Tsang, C.W., Harrison, A.G.: Chemical ionization of amino acids. J. Am.
Chem. Soc. 98, 1301–1308 (1975)
Funding Information
This work was supported by the fund from the Merck Research
Laboratories New Technologies Review & Licensing Commit-
tee (NT-RLC) and also funded by NSF (CHE-1709075) and
NSF IDBR (CHE-1455554).
References
1. Beranová, S., Cai, J., Wesdemiotis, C.: Unimolecular chemistry of pro-
tonated glycine and its neutralized form in the gas phase. J. Am. Chem.
Soc. 117, 9492–9501 (1995)
2. Steen, H., Mann, M.: The abc’s (and xyz’s) of peptide sequencing. Nat.
Rev. Mol. Cell Biol. 5, 699–711 (2004)
3. Medzihradszky, K.F., Chalkley, R.J.: Lessons in de novo peptide se-
quencing by tandem mass spectrometry. Mass Spectrom. Rev. 34, 43–
63 (2015)
4. Seidler, J., Zinn, N., Boehm, M.E., Lehmann, W.D.: De novo sequencing
of peptides by MS/MS. Proteomics. 10, 634–649 (2010)
5. Marzluff, E.M., Campbell, S., Rodgers, M.T., Beauchamp, J.L.: Low-
energy dissociation pathways of small deprotonated peptides in the gas
phase. J. Am. Chem. Soc. 116, 7787–7796 (1994)
28. Ambihapathy, K., Yalcin, T., Leung, H.-W., Harrison, A.G.: Pathways to
immonium ions in the fragmentation of protonated peptides. J. Mass
Spectrom. 32, 209–215 (1997)
29. Pingitore, F., Polce, M.J., Wang, P., Wesdemiotis, C., Paizs, B.: Intramo-
lecular condensation reactions in protonated dipeptides: carbon monox-
ide, water, and ammonia losses in competition. J. Am. Soc. Mass
Spectrom. 15, 1025–1038 (2004)
6. Ly, T., Julian, R.R.: Ultraviolet photodissociation: developments towards
applications for mass-spectrometry-based proteomics. Angew. Chemie
Int. Ed. 48, 7130–7137 (2009)
7. Butcher, D.J., Asano, K.G., Goeringer, D.E., McLuckey, S.A.: Thermal
dissociation of gaseous bradykinin ions. J. Phys. Chem. A. 103, 8664–
8671 (1999)
8. Zubarev, R.A., Kelleher, N.L., McLafferty, F.W.: Electron Capture Dis-
sociation of Multiply Charged Protein Cations. A Nonergodic Process. J
Am Chem Soc. 120, 3265–3266 (1998)
9. Wesdemiotis, C., McLafferty, F.W.: Neutralization-reionization mass
spectrometry (NRMS). Chem. Rev. 87, 485–500 (1987)
30. Reid, G.E., Simpson, R.J., O’Hair, R.A.J.: A mass spectrometric and ab
initio study of the pathways for dehydration of simple glycine and
cysteine-containing peptide [M+H]+ ions. J. Am. Soc. Mass Spectrom.
9, 945–956 (1998)
31. O’Hair, R.A.J., Broughton, P.S., Styles, M.L., Frink, B.T., Hadad, C.M.:
The fragmentation pathways of protonated glycine: a computational
study. J. Am. Soc. Mass Spectrom. 11, 687–696 (2000)
32. Kulik, W., Heerma, W.: Fast atom bombardment tandem mass spectrom-
etry for amino acid sequence determination in tripeptides. Biol. Mass
Spectrom. 18, 910–917 (1989)
33. Meot-Ner, M., Field, F.H.: Chemical ionization mass spectrometry. XX.
Energy effects and virtual ion temperature in the decomposition kinetics
of amino acids and amino acid derivatives. J. Am. Chem. Soc. 95, 7207–
7211 (1973)
10. Goldberg, N., Schwarz, H.: Neutralization-reionization mass spectrome-
try: a powerful “laboratory” to generate and probe elusive neutral mole-
cules. Acc. Chem. Res. 27, 347–352 (1994)
11. Zagorevskii, D.V., Holmes, J.L., Zverev, D.V., Orlova, T.Y., Nekrasov,
Y.S.: The structure of C5H5RFe+ (R = F, cl, Br, I, O, OH, OCH3, C6H5,
34. Zoltán Takáts, Justin M. Wiseman, Bogdan Gologan, and, Cooks, R.G.:
Electrosonic Spray ionization. A gentle technique for generating folded

P. Zhao et al.: Detection of Neutral Fragments from Ion Dissociation
proteins and protein complexes in the gas phase and for studying ion−mole-
CO-release rates from organometallic carbonyl complexes. Dalt. Trans.
40, 5755 (2011)
cule reactions at atmospheric pressure. Anal. Chem. 76, 4050–4058 (2004)
35. Clipston, N.L., Jai-nhuknan, J., Cassady, C.J.: A comparison of negative
and positive ion time-of-flight post-source decay mass spectrometry for
peptides containing basic residues. Int. J. Mass Spectrom. 222, 363–381
(2003)
47. Park, S.S., Kim, J., Lee, Y.: Improved electrochemical microsensor for
the real-time simultaneous analysis of endogenous nitric oxide and carbon
monoxide generation. Anal. Chem. 84, 1792–1796 (2012)
48. Hasegawa, U., van der Vlies, A.J., Simeoni, E., Wandrey, C., Hubbell,
J.A.: Carbon monoxide-releasing micelles for immunotherapy. J. Am.
Chem. Soc. 132, 18273–18280 (2010)
49. Esteban, J., Ros-Lis, J.V., Martínez-Máñez, R., Marcos, M.D., Moragues,
M., Soto, J., Sancenón, F.: Sensitive and selective chromogenic sensing of
carbon monoxide by using binuclear rhodium complexes. Angew.
Chemie Int. Ed. 49, 4934–4937 (2010)
50. Moragues, M.E., Esteban, J., Ros-Lis, J.V., Martínez-Máñez, R., Marcos,
M.D., Martínez, M., Soto, J., Sancenón, F.: Sensitive and selective
chromogenic sensing of carbon monoxide via reversible axial CO coor-
dination in binuclear rhodium complexes. J. Am. Chem. Soc. 133,
15762–15772 (2011)
51. Morimoto, Y., Durante, W., Lancaster, D.G., Klattenhoff, J., Tittel, F.K.:
Real-time measurements of endogenous CO production from vascular
cells using an ultrasensitive laser sensor. Am. J. Physiol. Heart Circ.
Physiol. 280, H483–H488 (2001)
52. Wang, J.: Electroanalytical techniques in clinical chemistry and laborato-
ry medicine. VCH, Weinheim (1988)
53. Woermann, D., Janata, J.: Principles of Chemical Sensors. Plenum Press,
New York and London 1989.317 Seiten, Preis in Europa: US $ 47.40.
Ber. Bunsenges. Phys. Chem. 94, 543–543 (1990)
54. Wang, J.: Analytical electrochemistry. Wiley-VCH, Weinheim (2006)
55. Brinker, U.H.: Advances in carbene chemistry, vol. 1. JAI Press, Green-
wich (1994)
36. Harrison, A.G.: Sequence-specific fragmentation of deprotonated pep-
tides containing H or alkyl side chains. J. Am. Soc. Mass Spectrom. 12,
1–13 (2001)
37. Johnson, R.S., Martin, S.A., Biemann, K., Stults, J.T., Watson, J.T.:
Novel fragmentation process of peptides by collision-induced decompo-
sition in a tandem mass spectrometer: differentiation of leucine and
isoleucine. Anal. Chem. 59, 2621–2625 (1987)
38. Roepstorff, P., Fohlman, J.: Letter to the editors. Biol. Mass Spectrom. 11,
601–601 (1984)
39. Verkerk, U.H., Siu, C.-K., Steill, J.D., El Aribi, H., Zhao, J., Rodriquez,
C.F., Oomens, J., Hopkinson, A.C., Siu, K.W.M.: a2 Ion derived from
triglycine: An N1 -Protonated 4-imidazolidinone. J. Phys. Chem. Lett. 1,
868–872 (2010)
40. Bythell, B.J., Maître, P., Paizs, B.: Cyclization and rearrangement reac-
tions of a n fragment ions of protonated peptides. J. Am. Chem. Soc. 132,
14766–14779 (2010)
41. Paizs, B., Suhai, S.: Fragmentation pathways of protonated peptides.
Mass Spectrom. Rev. 24, 508–548 (2005)
42. Polfer, N.C., Oomens, J., Suhai, S., Paizs, B.: Infrared spectroscopy and
theoretical studies on gas-phase protonated leu-enkephalin and its frag-
ments: direct experimental evidence for the mobile proton. J. Am. Chem.
Soc. 129, 5887–5897 (2007)
43. Chakravarty, A.R., Cotton, F.A., Tocher, D.A., Tocher, J.H.: Structural
and electrochemical characterization of the novel ortho-metalated
dirhodium (II) compounds Rh2(O2CMe)2[Ph2P(C6H4)]2.cntdot.2L. Or-
ganometallics. 4, 8–13 (1985)
56. Yasaka, Y., Yoshida, K., Wakai, C., Matubayasi, N.: A., Nakahara, M.:
kinetic and equilibrium study on formic acid decomposition in relation to
the water-gas-shift reaction. J. Phys. Chem. A. 110, 11082–11090 (2006)
57. Bythell, B.J., Barofsky, D.F., Pingitore, F., Polce, M.J., Wang, P.,
Wesdemiotis, C., Paizs, B.: Backbone cleavages and sequential loss of
carbon monoxide and ammonia from protonated AGG: a combined
tandem mass spectrometry, isotope labeling, and theoretical study. J.
Am. Soc. Mass Spectrom. 18, 1291–1303 (2007)
44. Marks, G.S., Vreman, H.J., McLaughlin, B.E., Brien, J.F., Nakatsu, K.:
Measurement of endogenous carbon monoxide formation in biological
systems. Antioxid. Redox Signal. 4, 271–277 (2002)
45. McLean, S., Mann, B.E., Poole, R.K.: Sulfite species enhance carbon
monoxide release from CO-releasing molecules: implications for the
deoxymyoglobin assay of activity. Anal. Biochem. 427, 36–40 (2012)
46. Atkin, A.J., Lynam, J.M., Moulton, B.E., Sawle, P., Motterlini, R., Boyle,
N.M., Pryce, M.T., Fairlamb, I.J.S.: Modification of the deoxy-myoglo-
bin/carbonmonoxy-myoglobin UV-vis assay for reliable determination of
58. Paizs, B., Suhai, S.: Theoretical study of the main fragmentation pathways
for protonated glycylglycine. Rapid Commun. Mass Spectrom. 15, 651–
663 (2001)