Structure and unimolecular chemistry of protonated sulfur betaines, (CH3)2S+(CH2)nCO 2H (n = 1 and 2)

Article abstract of DOI:10.1039/c0ob00770f

The fixed charge zwitterionic sulfur betaines dimethylsulfonioacetate (DMSA) (CH₃)₂S⁺CH₂CO ₂^- and dimethylsulfoniopropionate (DMSP) (CH ₃)₂S⁺(CH₂)₂CO ₂^- have been synthesized and the structures of their protonated salts (CH₃)₂S⁺CH₂CO ₂H...Cl^- [DMSA.HCl] and (CH₃) ₂S⁺(CH₂)₂CO₂H... Pcr^- [DMSP.HPcr] (where Pcr = picrate) have been characterized using X-ray crystallography. The unimolecular chemistry of the [M+H]⁺ of these betaines was studied using two techniques; collision-induced dissociation (CID) and electron-induced dissociation (EID) in a hybrid linear ion trap Fourier transform ion cyclotron resonance mass spectrometer. Results from the CID study show a richer series of fragmentation reactions for the shorter chain betaine and contrasting main fragmentation pathways. Thus while (CH ₃)₂S⁺(CH₂)₂CO ₂H fragments via a neighbouring group reaction to generate (CH ₃)₂S⁺H and the neutral lactone as the most abundant fragmentation channel, (CH₃)₂S⁺CH ₂CO₂H fragments via a 1,2 elimination reaction to generate CH₃S⁺CH₂ as the most abundant fragment ion. To gain insights into these fragmentation reactions, DFT calculations were carried out at the B3LYP/6-311++G(2d,p) level of theory. For (CH₃) ₂S⁺CH₂CO₂H, the lowest energy pathway yields CH₃S⁺CH₂via a six-membered transition state. The two fragment ions observed in CID of (CH₃) ₂S⁺(CH₂)₂CO₂H are shown to share the same transition state and ion-molecule complex forming either (CH₃)₂S⁺H or (CH₂) ₂CO₂H⁺. Finally, EID shows a rich and relatively similar fragmentation channels for both protonated betaines, with radical cleavages being observed, including loss of CH₃.

Full text of DOI:10.1039/c0ob00770f

View Article Online / Journal Homepage / Table of Contents for this issue
Organic &
Biomolecular
Chemistry
Dynamic Article Links
Cite this: Org. Biomol. Chem., 2011, 9, 2751
www.rsc.org/obc
PAPER
Structure and unimolecular chemistry of protonated sulfur betaines,
(CH3)2S+(CH2)nCO2H (n = 1 and 2)†‡
Ellie Jung-Hwa Yoo,a,b,c Linda Feketeova´,a,b,c George N. Khairallah,a,b,c Jonathan M. Whitea,b and
Richard A. J. O’Hair*a,b,c
Received 23rd September 2010, Accepted 6th January 2011
DOI: 10.1039/c0ob00770f
The fixed charge zwitterionic sulfur betaines dimethylsulfonioacetate (DMSA) (CH3)2S+CH2CO2 and
-
-
dimethylsulfoniopropionate (DMSP) (CH3)2S+(CH2)2CO2 have been synthesized and the structures of
their protonated salts (CH3)2S+CH2CO2H ◊ ◊ ◊ Cl- [DMSA.HCl] and (CH3)2S+(CH2)2CO2H ◊ ◊ ◊ Pcr-
[DMSP.HPcr] (where Pcr = picrate) have been characterized using X-ray crystallography. The
unimolecular chemistry of the [M+H]+ of these betaines was studied using two techniques;
collision-induced dissociation (CID) and electron-induced dissociation (EID) in a hybrid linear ion trap
Fourier transform ion cyclotron resonance mass spectrometer. Results from the CID study show a richer
series of fragmentation reactions for the shorter chain betaine and contrasting main fragmentation
pathways. Thus while (CH3)2S+(CH2)2CO2H fragments via a neighbouring group reaction to generate
(CH3)2S+H and the neutral lactone as the most abundant fragmentation channel, (CH3)2S+CH2CO2H
fragments via a 1,2 elimination reaction to generate CH3S+ CH2 as the most abundant fragment ion.
To gain insights into these fragmentation reactions, DFT calculations were carried out at the
B3LYP/6-311++G(2d,p) level of theory. For (CH3)2S+CH2CO2H, the lowest energy pathway yields
CH3S+ CH2 via a six-membered transition state. The two fragment ions observed in CID of
(CH3)2S+(CH2)2CO2H are shown to share the same transition state and ion-molecule complex forming
either (CH3)2S+H or (CH2)2CO2H+. Finally, EID shows a rich and relatively similar fragmentation
channels for both protonated betaines, with radical cleavages being observed, including loss of ∑CH3.
Introduction
Betaines are fixed charge zwitterions that play key roles in biology.1
Examples of betaines containing a carboxylate group are shown
in Scheme 1. Glycine betaine (GB, 1), is a biologically important
methyl donor in a metabolic pathway linked to the synthesis of neu-
rotransmitters and homocysteine remethylation2 and also serves as
Scheme 1 Examples of fixed charge zwitterionic betaines possessing a
carboxylate group: 1, Glycine betaine (GB); 2, Dimethylsulfonioacetate
osmoprotectant in many osmotolerant bacteria like lactobacillus
acidophilus.3 The sulfur analogues of GB, dimethylsulfonioacetate
(DMSA, 2) and its higher homologue, dimethylsulfoniopropionate
(DMSP, 3), are produced by a wide variety of photosynthetic
organisms including marine algae, phytoplankton, cyanobacteria
(DMSA); 3, Dimethylsulfoniopropionate (DMSP).
and/or nearshore halophilic higher plants4,5 and have been shown
to be potential osmoprotectants for many bacteria.4,6 DMSA is
also a potent biological methyl donor,7 while Seymour and co-
workers have recently highlighted the potential influence that
phytoplankton-produced DMSP could have on the global climate
change.8
Since amino acids form zwitterions in solution but are in their
canonical forms in the gas phase,9 GB has attracted considerable
interest as a model to study the fundamental gas phase chemistry
of zwitterions. The presence of the permanent fixed charges
can have a profound effect on solvation,10 cluster chemistry,11–13
thermochemical properties14 and unimolecular reactivity.13,15 For
example, Patrick et al. measured gas-phase basicity of GB and
showed that the permanent charge decreases the proton affinity
of the anionic site by over 4 eV in comparison to glycine.14
aSchool of Chemistry, The University of Melbourne, Victoria, 3010, Aus-
tralia. E-mail: rohair@unimelb.edu.au; Fax: 61 3 9347 5180; Tel: 61 3 8344
2452
bBio21 Institute of Molecular Science and Biotechnology, The University of
Melbourne, Victoria, 3010, Australia
cARC Centre of Excellence for Free Radical Chemistry and Biotechnology,
Victoria, 3010, Australia
† Electronic supplementary information (ESI) available: Cartesian co-
ordinates of all DFT calculated species; Potential energy diagram for
all competing reactions calculated relevant to the [DMSA+H]+. CCDC
reference numbers 794852 and 794853. For ESI and crystallographic data
in CIF or other electronic format see DOI: 10.1039/c0ob00770f
‡ Part 77 of the series “Gas-Phase Ion Chemistry of Biomolecules”.
This journal is
The Royal Society of Chemistry 2011
Org. Biomol. Chem., 2011, 9, 2751–2759 | 2751
©

View Article Online
Proton-bound dimers of GB with other molecules form ion-
zwitterion structure in which the binding energy depends on
the strength of ion-dipole interaction.11 Electrons can be bound
to GB through anion dipole bound state.16 The important
fragmentation pathway of protonated betaine and its neutralized
form involved cleavage of the N–C bond between nitrogen and
CH2 of the carboxyl group.13 The gas phase fragmentation
chemistry of protonated betaine clusters are size dependant:
while neutral evaporation dominates for the larger clusters,13
betaine evaporation competes with intracluster SN2 reactions for
the protonated dimer under collision-induced dissociation (CID)
conditions.12c Electronic excitation of multiply protonated betaine
clusters under electron-induced dissociation (EID) conditions can
lead to a number of new C–C and C–N bond cleavages, including
decarboxylation and CH3 group transfer.12e
of chloroform. Ether diffusion resulted in formation of crystals
after 5 days. Intensity data for [DMSP.HPcr] was collected with
an Oxford Diffraction Sapphire CCD diffractometer using Cu-
Ka radiation (graphite crystal monochromator l = 1.54184),
intensity data for DMSA.HCl was collected with a Bruker
SMART Apex CCD detector using Mo-Ka radiation (graphite
crystal monochromator l = 0.71073). The temperature during data
collections was maintained at 130.0(1) K.
Crystal data for [DMSA·HCl]. C4H9ClO2S, M = 156.62, T =
˚
130.0(2)K, l = 0.71073 A, Monoclinic, space group P21/n a =
◦
˚
7.0225(8), b = 7.3868(8) c = 13.854(1) A, b = 100.912(2) . V 705.7(1)
A , Z = 4, Dc = 1.474 mg M m(Mo-Ka) 0.752 mm-1, F(000) =
328, crystal size 0.50 ¥ 0.30 ¥ 0.15 mm. 3359 reflections measured,
1608 independent reflections (Rint = 0.0165) the final R was 0.0269,
[I > 2s(I)] and wR(F2) was 0.0768 (all data).
3
-3
˚
Here we: (i) synthesize the sulfur betaines DMSA, 2, and DMSP,
3; (ii) provide the first structures of the protonated salts of 2 and
3 via X-ray crystallography; (iii) use mass spectrometry and DFT
calculations to examine the gas phase unimolecular chemistry of
protonated DMSA and DMSP under CID and EID conditions.
This approach allows us to evaluate how changing the number
of methylene “spacers” between the charge sites influences the
structure and gas phase chemistry of sulfur betaines.
Crystal data for [DMSP·HPcr]. C11H13N3O9S, M = 363.30, T =
¯
˚
130.0(2)K, l = 1.54184 A, Triclinic, space group P1a = 7.3627(9),
◦
˚
b = 9.5890(9) c = 11.7822(19) A, a = 69.90(1) b = 73.70(1) g =
3
-3
˚
79.988(9). V 747.0(2) (6) A , Z = 2, Dc = 1.615 mg M m(Cu-Ka)
2.469 mm-1, F(000) = 376, crystal size 0.1 ¥ 0.08 ¥ 0.01 mm. 5047
reflections measured, 2860 independent reflections (Rint = 0.0636)
the final R was 0.0543, [I > 2s(I)] and wR(F2) was 0.1200 (all
data).
Mass spectrometry experiments
Experimental
All experiments were carried out using a hybrid Finnigan
LTQ FT (Thermo, Bremen) mass spectrometer equipped with a
Finnigan electrospray ionization (ESI) source described in detail
previously.12c,19 The samples were prepared as a 1 mM in 50 : 50 : 1
methanol : water : acetic acid solution, which was introduced into
the ESI source of the mass spectrometer at a flow rate of 5.0 mL
min-1. Typical ESI conditions used were; spray voltage, 3.0–4.0
kV, capillary temperature, 250 ◦C, nitrogen sheath gas, 10–30. The
capillary voltage and the tube lens offset were tuned to maximize
the yield of the desired ion. The injection time was set using the
automatic gain control function. The LTQ-FT mass spectrometer
consists of: (i) Linear ion Trap (LTQ); (ii) ion transfer optics; and
(iii) FT-ICR mass analyzer. For the tandem mass spectrometry
experiments, the desired ions produced via ESI were trapped in
the LTQ and subjected to CID at a He bath gas pressure of ca 5 ¥
10-3 Torr. The CID experiments were performed by mass selecting
the desired precursor ion with an activation window of m/z 1–2,
and then subjecting it to CID using normalized collision energies
between 40%–50% and an activation Q of 0.25 for a period of
30 ms. For high resolution mass analysis and EID, the ions were
transferred via the ion optics transfer region (~2 ¥ 10-7 Torr) into
an FT-ICR cell at a pressure below 1.5 ¥ 10-9 Torr. The FT-ICR
cell is supplied with low energy electrons produced by an indirectly
heated emitter cathode located downstream of the FT-ICR cell.
The energy of electrons is given by the potential difference between
the emitter cathode with ECD offset of -4.36 V and the grid
positioned infront of the cathode, which is variable. Ions were
bombarded with electrons of 30 eV energy, i.e. corrected 25.6 eV,
for 50 ms.
Materials
All purchased materials were used without further purification:
chloroacetic acid 99%, dimethyl sulfide 99%, acrylic acid 99%
(Bio Scientific, USA), acetic acid glacial 100% (BDH, AnalaR),
methanol (MERCK), picric acid in methanol (MERCK).
Synthesis of DMSA and DMSP
DMSA and DMSP were prepared following the procedures
described by Vasudevamurthy et al.17 and Howard et al.,18
respectively. They were characterized by: high resolution mass
spectrometry (HRMS) as described below and via 1H NMR and
13C NMR spectroscopy carried out on an Inova 400 or Inova 500
spectrometer.
Data for DMSA: HRMS (ESI): [DMSA+H]+ 121.03178 (exptl),
1
121.03178 (theory). NMR of [DMSA.HCl] salt: H NMR (500
MHz, CD3OD) d: 5.07 (1H, S), 4.58 (2H,S), 3.06 (6H, S). 13C
NMR (125 MHz, CD3OD) d: 167.9 (COOH), 47.6(–CH2–), 26.5
(Me2S).
Data for DMSP: HRMS (ESI): [DMSP+H]+ 135.04750 (exptl),
135.04743 (theory). NMR of crude DMSP: 1H NMR (500 MHz,
CD3OD) d: 4.42 (2H, t), 2.71 (2H, t), 2.10 (6H, S). 13C NMR (125
MHz, CD3OD) d: 170.3 (COO), 132.2 (–CH2–), 130.9 (–CH2–),
18.8 (Me2S).
X-ray crystallography
DMSA was converted to its hydrochloride salt within 10 days
from the mixture of chloroacetic acid and dimethyl sulfide. DMSP
was converted to its picrate salt by mixing liquid DMSP with
an equivalent amount of picric acid dissolved in a minimum
amount of MeOH. Evaporation under reduced pressure afforded
crude yellow crystals which were dissolved in a minimum amount
Theoretical calculations
Density functional theory (DFT) calculations were used to exam-
ine the structure and the stability of molecules using the Gaussian
2752 | Org. Biomol. Chem., 2011, 9, 2751–2759
This journal is
The Royal Society of Chemistry 2011
©

View Article Online
03 package.20 Optimised structures were calculated using the B3-
LYP level of theory with 6-311++G(2d,p) basis set for all atoms
(C,S,O,H). Transition states for important reaction pathways were
connected to the appropriate reactant and product ion minima
via intrinsic reaction coordinate (IRC) runs. All energies were
corrected for zero-point kinetic energies.21 (Ereported = Eelctronic
zpve). Vibrational frequency analysis was included to confirm that
+
E
the structures are local minima or transition states. All calculated
structures are available in the ESI.†
Strategies for calculating the global minimum of protonated
DMSA and DMSP
In order to calculate the global minima for protonated DMSA
and DMSP, and due to the large number of potential structures,
we followed a strategy to generate different likely input structures
using the Chem3D pro software package.22 By rotating the main
bonds centred around the backbone by 90◦ and thus changing the
corresponding angles, we were able to generate 16 possible input
structures for DMSA and 64 for DMSP. Note that this strategy
makes no assumption on the conformation of the carboxylic acid
group, which generally prefers to be in the cis rather than the trans
form for simple carboxylic acids such as acetic acid.23 Additionally,
we have used the experimentally generated X-ray crystal structures
minus the counter anions as input files for optimization. The
Initial 16 structures for [DMSA+H]+ were optimised at the HF/3-
21G level of theory followed by B3-LYP/6-31G(d). The 7 most
stable structures (ESI†) and the X-ray structure were then further
optimised, at B3-LYP/6-311++G(2d,p) level of theory.
Fig. 1 Thermal ellipsoid plot for: (a) [DMSA·HCl]; (b) [DMSP·HPcr].
Ellipsoids are at the 20% probability level.
Similarly, the initial 64 structures found for [DMSP+H]+ were
optimised at the same level of theory and the 9 most stable
structures as well as the X-ray structure were further optimised
at B3-LYP/6-311++G(2d,p) level of theory. The optimization of
cation portion of the X-ray structures yielded minima that closely
resembled those structures obtained with the aforementioned
search as discussed below (also see ESI†).
Results and Discussion
We first compare the X-ray and DFT calculated structures of
protonated DMSA and DMSP. The gas phase unimolecular
chemistry of protonated DMSA and DMSP are then described
using a combination of mass spectrometry experiments and DFT
calculations.
Fig. 2 DFT calculated structures of DMSA, Structure 4 is the global
minimum and 5 is the structure resulting from optimization of the isolated
cation part of the X-ray salt structure.
Structures of protonated DMSA and DMSP: X-ray versus DFT
calculations
structures show carboxyl groups that are in the cis form, except
for the higher energy conformation 14, which has its carboxyl
group in the trans form. Full details (Cartesian coordinates
and energies) are available in the ESI (Fig. S1†). The DFT
predicted global minimum structure of [DMSA+H]+, found from
the conformational search, is shown as 4 in Fig. 2. The S–C–
The thermal ellipsoid plots for [DMSA.HCl] and [DMSP.HPcr]
are presented in Fig. 1, while the DFT calculated structures of
protonated DMSA and DMSP are given in Fig. 2 and 3. The solid
state and DFT calculated gas phase structures are now compared
individually.
C
O bonds lie in the plane of symmetry. The (CH3)2S group
and the C O of the carbonyl have a syn arrangement. DFT
reoptimization of the [DMSA+H]+ portion of the X-ray structure
of the chloride salt (Fig. 1a) gives structure 5, which has undergone
minor structural changes compared to the X-ray structure. 5 is the
next lowest energy structure (+0.004 eV), almost isoenergetic with
rotamer 4, found in the conformational search. The comparison
between structure 5 and the X-ray structure in bond lengths and
Protonated DMSA
Seven low lying structures were found for protonated DMSA.
The structures were optimized at the B3-LYP/6-311++G(2d,p)
level of theory, four of which are shown in Fig. 2. The remaining
three rotamers (5¢, 7¢ and 14) are available in the ESI.† All
This journal is
The Royal Society of Chemistry 2011
Org. Biomol. Chem., 2011, 9, 2751–2759 | 2753
©

View Article Online
Table 1 Selected structural details of: the DMSA salt; DFT B3-LYP+6-
311++G(2d,p) optimised gas phase structure 5 of the isolated cation part
of the X-ray salt structure
X-ray
Opt (5)
˚
Bond distance (A)
S1–C2
S1–C3
S1–C4
C1–C2
C1–O1
C1–O2
1.794(3)
1.789(5)
1.789(7)
1.507(2)
1.208(4)
1.316(2)
1.829(4)
1.818(6)
1.818(1)
1.522(9)
1.201(8)
1.328(0)
Bond Angles (deg)
C3–S1–C2
C4–S1–C2
S1–C2–C1
O1–C1–C2
O2–C1–C2
O1–C1–O2
Dihedral Angles
S1–C2–C1–O1
S1–C2–C1–O2
C3–S1–C2–C1
C4–S1–C2–C1
98.9(7)
103.1(4)
110.2(6)
124.0(3)
110.3(2)
125.6(5)
101.0(2)
103.6(7)
108.3(9)
123.0(3)
110.4(0)
126.5(7)
4.3(2)
-176.4(6)
-178.2(9)
-74.4(3)
8.6(0)
-171.9(2)
-179.5(3)
-73.8(9)
angles is shown in Table 1. One of the CH3 groups is located in
the S–C–C O plane, while the second CH3 group is pointing
out of the plane, at an angle of 137◦ with the plane. The other
structure (5¢) shown in the ESI† is isoenergetic with structure
5 and is its mirror image (through the C–S–C–C O plane).
Structure 6 is similar to structure 4 through rotation of the COOH
group around the C–C bond, but with a higher relative energy
(0.15 eV). Similarily, structure 7 resembles structure 5 with the
COOH rotated around the C–C bond and with a relative energy
of +0.13 eV.
In the crystal structure, [DMSA+H]+ adopts a conformation
which is similar to that for 5 (see Table 1) with a small difference
that can be attributed to the chloride counter anion and/or crystal
packing forces. The non-bond distance between sulfur and oxygen
Fig. 3 DFT calculated structures of DMSP, Structure 8 is the global
minimum and 11 is the structure resulting from optimization of the isolated
cation part of the X-ray salt structure (0.22 eV higher than the global
minimum).
as 8 in Fig. 3. Similar to protonated DMSA, the global minimum is
involved in an electrostatic interaction arising between the sulfur
and oxygen atoms. The non bond distance between sulfur and
˚
oxygen atom is 2.893(4) A, which is within the van der Waals
radii, suggesting that there is s-hole interaction occurring. The
mirror image (through C–C–C plane) of structure 8 is given in the
ESI.† 9 is the next lowest energy structure (+0.14 eV) found in the
conformational search and shows a similar type of stabilization
found in structure 4 of [DMSA+H]+, involving hydrogen bonding
between the oxygen of the C O group and the hydrogens of
the two methyl groups on the sulfur atom. 10 is lowest lying
structure adopting a “zig-zag” conformation (+0.16 eV). DFT
reoptimization of the [DMSP+H]+ portion of the X-ray structure
of the picrate salt (Fig. 1b) gives structure 11, which has undergone
minor structural changes compared to the X-ray structure. The
comparison between structure 11 and the X-ray structure in bond
lengths and angles is shown in Table 2. However, the conformation
of DMSP in the crystal does not correspond to the calculated
global minimum, but rather to a higher energy local minima (see
Fig. 3 and Table 2). The adoption of this conformation in the solid
state presumably reflects the packing demands associated with the
large picrate anion. In contrast to the case of the isoenergetic
rotamers 4 and 5 of [DMSA+H]+, 11 is 0.22 eV higher in energy
than the global minimum 8. Structures 12 (+0.26 eV) and 13
(+0.27 eV) both have COOH group that have been rotated around
the C–C bond. The “zig-zag” formation is also present in the
structure 13. Since the “zig-zag” structures (10,13) are much
higher in energies than all other structures of [DMSP+H]+, the
interaction between the carbonyl group and the methyl groups on
˚
atom in the crystal structure is 2.888(2) A, and this distance is
within the van der Waals radii (3.32 A),24 suggesting that there is an
˚
attractive interaction between the sulfur and oxygen atoms that is
involved in stabilising this conformation. The natural bond orbital
(NBO)25 calculation reveals that there is no significant interaction
between the lone pair of oxygen and the anti-bonding orbital
located on the sulfur (For charge and NBO analysis see ESI Fig.
S2†). Thus the interaction is mainly electrostatic. In a recent paper,
Clark et al.26 have introduced the term s-hole with reference to
Group VII interactions, and Murry et al.27 have reported that this
interaction can be extended to Group V and VI. According to the
s-hole bonding concept, the electrostatic interaction in protonated
DMSA can be attributed to s-hole bonding, as the electron-
deficient outer lobe of a half-filled p bonding orbital of group
VI (sulfur) interacts attractively to the nucleophilic site (oxygen).
Protonated DMSP
Fig. 3 shows six low level lying structures found for protonated
DMSP at at the B3-LYP/6-311++G(2d,p) level of theory. The
remaining three structures are given in the ESI (8¢, 10¢ and 12¢).†
Once again, all structures had carboxyl groups which were in the
cis form (see ESI†). The DFT predicted global minimum structure
of [DMSP+H]+, found from the conformational search, is shown
2754 | Org. Biomol. Chem., 2011, 9, 2751–2759
This journal is
The Royal Society of Chemistry 2011
©

View Article Online
∑
→ (CH3)2S+CH2 + ∑CO2H
(6)
(7)
Table 2 Selected structural details of: the DMSP salt; DFT B3-LYP+6-
311++G(2d,p) optimised gas phase structure 11 of the isolated cation part
of the X-ray salt structure
→ CH3S∑+CH2CO2H + ∑CH3
X-ray
Opt (11)
Since our experiments do not detect the neutral(s) lost, we
turned to DFT calculations to examine the nature of these species.
In the case of the major fragment ion CH3S+ CH2, and as noted
in our previous study on protonated GB, there are a number of
possible mechanisms that give rise to different losses with an
overall stoichiometry of [C2,H4,O2].13 The mechanims with the
lowest barrier (2.02 eV) takes place via a six-centred transition
state (Fig. 5b, TS4–15) leading to the loss of the enol of acetic acid,
CH2C(OH)2. The products of this reaction are calculated to be
higher in energy than their associated transition state suggesting
that the product ion-molecule complex undergoes further heating
to induce dissociation. This transition state TS4–15 is 0.83 eV and
1.04 eV more favoured than two other transition states found (see
ESI†) that could lead respectively to the formation of the same
˚
Bond distance (A)
S1–C3
S1–C4
S1–C5
C2–C3
C1–C2
C1–O1
C1–O2
1.804(3)
1.781(0)
1.780(3)
1.513(7)
1.494(9)
1.327(7)
1.206(7)
1.834(3)
1.816(1)
1.816(3)
1.521(5)
1.524(9)
1.337(8)
1.201(0)
Bond Angles (deg)
C3–S1–C4
C3–S1–C5
S1–C3–C2
C1–C2–C3
O1–C1–C2
O2–C1–C2
O1–C1–O2
Dihedral Angles
S1–C3–C2–C1
C4–S1–C3–C2
C5–S1–C3–C2
O1–C1–C2–C3
O2–C1–C2–C3
105.7(5)
101.8(6)
116.1(5)
111.9(4)
112.0(0)
124.9(2)
123.0(6)
105.2(2)
104.9(9)
117.5(0)
108.9(0)
110.7(3)
123.9(7)
125.3(1)
fragment ion through either the loss of H2O + CH2
C O or
169.5(5)
40.1(9)
-65.6(4)
170.2(3)
-11.4(8)
178.9(6)
51.7(1)
-56.2(3)
177.5(2)
-2.6(0)
loss of acetic acid (ESI Fig. S6†). Thus the favoured mechanism is
+
similair to that found for the formation of (CH3)2N CH2 from
protonated glycine betaine.13
The second most abundant fragment ion (CH2)2S+CH3 at m/z
75 is due to the loss of formic acid (HCO2H) (eqn (2)) and most
likely posseses a three membered ring structure. Considering the
thermochemistry of these products, eqn (2) is found to be the
least endothermic (1.03 eV), well below all other fragmentation
channels considered (see Fig. 5). However, if (CH2)2S+CH3 does
not form a ring structure, but instead froms an open structure (see
ESI†), the associated transition state energy (TS4–18) has a barrier
of 4.27 eV and is endothermic by 3.8 eV. The energetics for this
latter reaction are too high for it to generate an ion of m/z 75 as the
second most abundant in the low energy CID spectrum. The other
minor fragments observed in the CID spectrum are calculated to
the sulfur most likely plays a role in stabilization of the low energy
conformation of [DMSP+H]+.
Gas phase unimolecular chemistry of protonated DMSA
Protonated DMSA, [DMSA+H]+ (m/z 121), formed via ESI of a
solution of DMSA, was mass selected and subjected to low energy
CID in the linear ion trap (Fig. 4a) and EID in the FT-ICR cell
(Fig. 4b). Eqn (1)–(7) summarize the fragment ions observed and
the inferred neutrals formed, but do not detail the mechanism(s)
of the fragmentation reactions. An examination of Fig. 4a reveals
that the CID spectrum of [DMSA+H]+ is dominated by the
formation of the CH3S+ CH2 fragment ion at m/z 61 (eqn
(1)), which has been observed earlier in the secondary ion mass
spectra of sulfonium ions.28 The second most abundant fragment
ion (CH2)2S+CH3 at m/z 75 is likely due to the loss of formic
acid (HCO2H) (eqn (2)). The other minor fragments observed
in the CID spectrum are due to loss of H2O to form (CH3)2S+–
be due to loss of H2O to form (CH3)2S+–CH
C
O ion at m/z
103, (eqn (3)), loss of a-lactone (CH2CO2) forming (CH3)2S+H at
m/z 63 (eqn (4)), and homolytic cleavage leading to the formation
of a radical ion (CH3)2S∑+ at m/z 62 (eqn (5)). The potential energy
surfaces are shown in Fig. 5. Formation of (CH3)2S+–CH
C O
(eqn (3)) is endothermic by 1.3 eV and has a transition state
barrier (TS5¢–17) of 2.8 eV, placing it ca 0.5 eV higher in energy
than the main fragmentation channel. The transition state for the
formation of (CH3)2S+H (eqn (4)) (TS14–16) was calculated to be
2.4 eV (> 0.1 eV higher than the main fragmentation channel) and,
thermochemically it is endothermic by 2.26 eV. The barrierless
homolytic cleavage to form (CH3)2S∑+ was calculated to be 2.78 eV.
Further mass spectrometry experiments have proven that the ions
observed are due to primary fragmentation channels (data not
shown). From these observations, it appears that there exists an
interplay between thermodynamics and kinetics in generating the
last three least abundant ions.
CH
C
O ion at m/z 103, (eqn (3)); formation of (CH3)2S+H
at m/z 63, most likely via loss of a-lactone (CH2CO2) (eqn (4)),
and homolytic cleavage leading to the formation of a radical ion
(CH3)2S∑+ at m/z 62 (eqn (5)). In contrast, the CID spectrum of the
analogous protonated nitrogen betaine, [GB+H]+, is much simpler,
yielding only two products: (CH3)3N∑+ as the major fragment
ion formed via homolysis (cf. eqn (5)) and the immonium ion
(CH3)2N+ CH2 (cf. eqn (1)).13
[DMSA+H]+ → CH3S+ CH2 + [C2,H4,O2]
→ (CH2)2S+CH3 + HCO2H
(1)
(2)
(3)
(4)
(5)
The EID spectrum shown in Fig. 4b is dominated by fragment
ions resulting from radical cleavage reactions leading to the
∑
formation of (CH3)2S∑+ (m/z 62) (eqn (5)), (CH3)2S+CH2 (m/z
→ (CH3)2S+–CH
C O + H2O
76) (eqn (6)) and CH3S∑+CH2CO2H (m/z 106) (eqn (7)). The
fragment at m/z 61 (CH3S+ CH2), which appeared in the CID
spectrum as the most abundant ion, is also observed in the EID
spectrum. The EID spectrum of [DMSA+H]+ bears similarities to
that of protonated glycine betaine13 except for one difference: while
→ (CH3)2S+H + CH2CO2
→ (CH3)2S∑+ + ∑CH2CO2H
This journal is
The Royal Society of Chemistry 2011
Org. Biomol. Chem., 2011, 9, 2751–2759 | 2755
©

View Article Online
Fig. 4 LTQ-FT MS2 of sulfur betaines: (a) CID spectrum of [DMSA+H]+ in the linear ion trap (activation time of 30 ms, normalised collision energy
48%); (b) EID spectrum of [DMSA+H]+ in the FT-ICR cell (activation time of 50 ms, corrected electron energy 25.6 eV); (c) CID spectrum of [DMSP+H]+
in the linear ion trap (activation time of 30 ms, normalised collision energy 40%); (d) EID spectrum of [DMSP+H]+ in the FT-ICR cell (activation time
of 50 ms, corrected electron energy 25.6 eV).
∑
[DMSA+H]+ loses a neutral radical CH3, protonated glycine
betaine loses CH4.
the latter instance, they most likely involve the a-amino group
rather than the carboxyl group, as this is expected to be the better
neighbouring group.29
Gas phase unimolecular chemistry of protonated DMSP
[DMSP+H]+ → (CH3)2S+H + (CH2)2CO2
→ (CH3)2S + (CH2)2CO2H+
(8)
(9)
ESI of a solution of DMSP in methanol generated protonated
DMSP, [DMSP+H]+ (m/z 121), which was mass selected and
subjected to low energy CID in the linear ion trap (Fig. 4c) and
EID in the FT-ICR cell (Fig. 4d). Eqn (8)–(13) summarize the
fragment ions observed and the inferred neutrals formed, but
do not detail the mechanism(s) of the fragmentation reactions.
The CID spectrum of [DMSP+H]+ (Fig. 4c) is much simpler
than that of [DMSA+H]+ (Fig. 4a), yielding only two fragment
ions: (CH3)2S+H at m/z 63 via loss of [C3,H4,O2] (eqn (8)) and
(CH2)2CO2H+ at m/z 73, formed via the loss of (CH3)2S (eqn (9)).
Both of these product ions are likely to arise from a neighbouring
group reaction29 in which the carboxylic acid group acts as a
nucleophile to displace the (CH3)2S leaving group to form an
ion-neutral complex (INC)30 between the protonated lactone and
(CH3)2S. Departure of (CH3)2S yields the protonated lactone (eqn
(9)) while proton transfer within the INC yields (CH3)2S+H (eqn
(8)). Related losses of CH3SR involving neighbouring group reac-
tions have been reported for alkylated methionine residues31 and
methylated S-methylcysteine, (CH3)2S+CH2CH(NH2)COOH.32 In
→ CH3S+ CH2 + (CH2)2C(OH)2
→ (CH3)2S∑+ + ∑(CH2)2CO2H
(10)
(11)
(12)
(13)
∑
→ (CH2)2S+CH2 + ∑[C2,H5,O2]
→ CH3S∑+(CH2)2COOH + ∑CH3
Calculation shows that these two fragmentation pathways are
possible through the same transition state and ion-molecule
complex (IMC, 19) (Fig. 6). In pathway 1, the further dissociation
of IMC (19) upon heating leads to the formation of the fragment
ion (CH2)2CO2H+, this channel was endothermic by 2.12 eV. In
pathway 2, IMC (19) reorganizes to a different IMC, 20 (we
have not calculated the energetic barrier associated with this
reorganisation, but it is expected to be small - see ref. 30). The
2756 | Org. Biomol. Chem., 2011, 9, 2751–2759
This journal is
The Royal Society of Chemistry 2011
©

View Article Online
Fig. 5 B3-LYP+6-311++G(2d,p) DFT calculations of the competing reactions shown in eqn (1)–(7) for DMSA. (a) Relative potential energy diagram;
(b) optimised geometries of transition states associated with reactions shown in eqn (1), (2), (3) and (4) from left to right.
dissociation of IMC (20) leads to (CH3)2S+H via a H transfer
and loss of (CH2)2CO2 (eqn (8)). This channel was endothermic
by 1.7 eV. Thus, in the case of DMSP, the DFT calculations are
consistent with the experimental observation (Fig. 4c) where the
ion at m/z 63 ((CH3)2S+H) was more abundant than (CH2)2CO2H+
at m/z 73.
did not appear in the [DMSA+H]+ case may be due to the longer
chain length that gives DMSP more flexibility.
Conclusions
The gas phase low energy CID chemistry of the protonated
betaines 1–3 depend upon both the cationic site (ammonium (in
1) versus sulfonium (in 2)) as well as the number of methylene
“spacers” between the cationic site and the carboxylic acid group
(CH2 for 2 and CH2CH2 for 3). Whereas CID of [GB+H]+ gener-
ates only (CH3)3N∑+ (major) and (CH3)2N+ CH2, is observed a
much richer CID spectrum for the sulfur analogue [DMSA+H]+
with the following fragment ions occuring: CH3S+CH2 (major),
EID of [DMSP+H]+ leads to several new fragments in compar-
ison to its CID spectrum, dominated by radical fragment ions.
This is not surprising due to the higher energy processes involved.
This EID is similar to that of [DMSA+H]+ forming two most
abundant ions; at m/z 61 (CH3S+ CH2) due to a H transfer from
the methyl group (eqn (10)) and the homolytic cleavage product
at m/z 62 ((CH3)2S∑+) (eqn (11)) (cf. eqn (1) and 5). The other
∑
fragment ions observed include; an ion at m/z 74 (CH2)2S+CH2 ,
(CH2)2S+CH3, (CH3)2S+CH
C
O, (CH3)2S+H and (CH3)2S∑+.
formed through the loss of ∑[C2,H5,O2] (eqn (12)), and might likely
accommodate a ring structure as determined via DFT calculations
as being lowest in energy in comparison to other open structures
(ESI†). The possible explanation for the loss of ∑[C2,H5,O2] could
be a concerted loss of a radical ∑CH3 from the parent ion forming
the observed fragment ion at m/z 120 (CH3S∑+(CH2)2CO2H) (eqn
(13)), this is followed by the H transfer from the second methyl
group for the further loss of HCO2H. The fact that this pathway
In contrast, CID of [DMSP+H]+ generates a simpler spectrum,
yielding only (CH3)2S+H (major) and (CH2)2COOH+. Both of
these ions arise from the same neighbouring group reaction. Thus
increasing the number of methylene “spacers” has a profound
effectonthe unimolecular chemistryof these sulfur betaines, which
is directly related to the the size of the lactone rings formed: the
a-lactone (CH2CO2) formed from [DMSA+H]+ has an activation
energy of 2.4 eV (TS14-16 in Fig. 5b); the b-lactone (CH2CH2CO2)
This journal is
The Royal Society of Chemistry 2011
Org. Biomol. Chem., 2011, 9, 2751–2759 | 2757
©

View Article Online
Fig. 6 B3-LYP+6-311++G(2d,p) DFT calculations of the competing reactions shown in eqn (8)–(13) for DMSP. (a) Relative energy diagram in eV; (b)
optimised geometries of transition state TS10¢-19 and ion-molecule complexes 19 and 20 associated with reactions shown in eqn (8) and (9).
formed from [DMSP+H]+ has an activation energy of 1.74 eV
(TS10¢–19 in Fig. 6b). EID of [DMSA+H]+ and [DMSP+H]+
provides complimentary structural information to CID. Given the
role of DMSA as a potent biological methyl donor, future studies
will compare the gas phase alkylation reactions of [DMSA+H]+
and [DMSP+H]+.
Lief grant and funding from the Victorian Institute for Chemical
Sciences are acknowledged for the purchase of the LTQ-FT mass
spectrometer.
Notes and references
1 J. Gorham, in Amino Acids and their Derivatives in Higher Plants, ed.
R. M. Wallsgrove, Cambridge, 1995, pp. 173–203; T. H. H. Chen and
N. Murata, Curr. Opin. Plant Biol., 2002, 5, 250; S. M. Joyce, A. C.
Cassells and S. M. Jain, Plant Cell, Tissue and Organ Cult., 2003, 74,
103.
2 P. I. Holm, P. M. Ueland, G. Kvalheim and E. A. Lien, Clin. Chem.,
2003, 49, 286.
3 J. B. Jewell and E. R. Kashket, Appl. Environ. Microbiol., 1991, 57,
2829.
4 A. Cosquer, V. Pichereau, J. A. Pocard, J. Minet, M. Cormier and
T. Bernard, Appl. Environ. Microbio., 1999, 65, 3304 and references
therein.
Acknowledgements
We thank the ARC for financial support via the ARC Centre of
Excellence for Free Radical Chemistry and Biotechnology. We
thank Dr Ina Ritsner-Sambor for discussions on NPA charge
analysis. LF and GNK thank the ARC for the award of an APD
and an ARF respectively. The authors gratefully acknowledge
the generous allocation of computing time from the Victorian
Partnership for Advanced Computing (VPAC) Facility. An ARC
5 G. Jackson and J. Stuckey, Aquat. Sci., 2007, 69, 419.
2758 | Org. Biomol. Chem., 2011, 9, 2751–2759
This journal is
The Royal Society of Chemistry 2011
©

View Article Online
6 V. Pichereau, J. A. Pocard, J. Hamelin, C. Blanco and T. Bernard, Appl.
Environ. Microbio., 1998, 64, 1420; A. Baliarda, H. Robert, M. Jebbar,
C. Blanco, A. Deschamps and C. Le Marrec, Int. J. Food Microbiol.,
2003, 84, 13; R. P. Kiene, L. P. H. Williams and J. E. Walker, Aquat.
Microb. Ecol., 1998, 15, 39.
7 V. Du Vigneaud, A. W. Moyer and J. P. Chandler, J. Biol. Chem., 1948,
174, 477.
20 Gaussian 03, Revision E.01: M. J. Frisch, G. W. Trucks, H. B. Schlegel,
G. E. Scuseria, M. A. Robb, J. R. Cheeseman, J. A. Montgomery, Jr.,
T. Vreven, K. N. Kudin, J. C. Burant, J. M. Millam, S. S. Iyengar, J.
Tomasi, V. Barone, B. Mennucci, M. Cossi, G. Scalmani, N. Rega,
G. A. Petersson, H. Nakatsuji, M. Hada, M. Ehara, K. Toyota,
R. Fukuda, J. Hasegawa, M. Ishida, T. Nakajima, Y. Honda, O.
Kitao, H. Nakai, M. Klene, X. Li, J. E. Knox, H. P. Hratchian, J.
B. Cross, V. Bakken, C. Adamo, J. Jaramillo, R. Gomperts, R. E.
Stratmann, O. Yazyev, A. J. Austin, R. Cammi, C. Pomelli, J. W.
Ochterski, P. Y. Ayala, K. Morokuma, G. A. Voth, P. Salvador, J. J.
Dannenberg, V. G. Zakrzewski, S. Dapprich, A. D. Daniels, M. C.
Strain, O. Farkas, D. K. Malick, A. D. Rabuck, K. Raghavachari, J. B.
Foresman, J. V. Ortiz, Q. Cui, A. G. Baboul, S. Clifford, J. Cioslowski,
B. B. Stefanov, G. Liu, A. Liashenko, P. Piskorz, I. Komaromi, R.
L. Martin, D. J. Fox, T. Keith, M. A. Al-Laham, C. Y. Peng, A.
Nanayakkara, M. Challacombe, P. M. W. Gill, B. Johnson, W. Chen, M.
W. Wong, C. Gonzalez, and J. A. Pople, Gaussian, Inc., Wallingford CT,
2004.
8 J. R. Seymour, R. Simo, T. Ahmed and R. Stocker, Science, 2010, 329,
342.
9 C. M. Aikens and M. S. Gordon, J. Am. Chem. Soc., 2006, 128, 12835.
10 R. A. Jockusch, A. S. Lemoff and E. R. Williams, J. Am. Chem. Soc.,
2001, 123, 12255; A. S. Lemoff, M. F. Bush and E. R. Williams, J. Am.
Chem. Soc., 2003, 125, 13576; A. S. Lemoff and E. R. Williams, J. Am.
Soc. Mass Spectrom., 2004, 15, 1014.
11 W. D. Price, R. A. Jockusch and E. R. Williams, J. Am. Chem. Soc.,
1998, 120, 3474.
12 (a) D. Ren, M. J. Polce and C. Wesdemiotis, Int. J. Mass Spectrom.,
2003, 228, 933; (b) H. A. Cox, R. Hodyss and J. L. Beauchamp, J. Am.
Chem. Soc., 2005, 127, 4084; (c) L. Feketeova´, G. N. Khairallah and R.
A. J. O’Hair, Eur. J. Mass Spectrom., 2008, 14, 107; (d) L. Feketeova´
and R. A. J. O’Hair, Chem. Commun., 2008, 4942; (e) L. Feketeova´ and
R. A. J. O’Hair, Rapid Commun. Mass Spectrom., 2009, 23, 3259.
13 J. A. Wyer, L. Feketeova´, S. B. Nielsen and R. A. J. O’Hair, Phys. Chem.
Chem. Phys., 2009, 11, 8752.
21 P. Scott and L. Radom, J. Phys. Chem., 1996, 100, 16502.
22 www.cambridgesoft.com.
23 K. B. Wiberg and K. E. Laidig, J. Am. Chem. Soc., 1987, 109, 5935.
24 A. Bondi, J. Phys. Chem., 1964, 68, 441.
25 F. Weinhold and C. R. Landis, Chem. Educ.: Res. Pract. Eur., 2001, 2,
91.
14 J. S. Patrick, S. S. Yang and R. G. Cooks, J. Am. Chem. Soc., 1996, 118,
26 T. Clark, M. Hennemann, J. Murray and P. Politzer, J. Mol. Model.,
231.
2007, 13, 291.
15 K. V. Wood, C. C. Bonham, D. Miles, A. P. Rothwell, G. Peel, B. C.
Wood and D. Rhodes, Phytochemistry, 2002, 59, 759; P. I. Holm, P. M.
Ueland, G. Kvalheim and E. A. Lien, Clin. Chem., 2003, 49, 286; R.
A. J. O’Hair, T. Waters and B. Cao, Angew. Chem., Int. Ed., 2007, 46,
7048.
16 J. Rak, P. Skurski and M. Gutowski, J. Chem. Phys., 2001, 114, 10673;
W. Zheng, S. Xu, D. Radisic, S. Stokes, X. Li and K. H. Bowen, Jr., J.
Chem. Phys., 2005, 122, 101103.
17 M. Vasudevamurthy, L. Weatherley and M. Lever, Biocatal. Biotrans-
form., 2005, 23, 285.
18 A. G. Howard and D. W. Russell, Anal. Chem., 1995, 67, 1293.
19 S. Horning, R. Malek, A. Wieghaus, M. W. Senko, and J. E. P. Syka,
Proceedings of the 51st ASMS Conference on Mass Spectrometry and
Allied Topics, Montreal, 2003.
27 J. S. Murray, P. Lane and P. Politzer, Int. J. Quantum Chem., 2008, 108,
2770.
28 K. L. Duffin and K. L. Busch, Int. J. Mass Spectrom., 1986, 74,
141.
29 B. Capon and S. P. McManus, in Neighbouring group participation,
Plenum Press, New York, 1976; R. A. J. O’Hair, J. Mass Spectrom.,
2000, 35, 1377.
30 R. D. Bowen, Acc. Chem. Res., 1991, 24, 364; R. D. Bowen, Org. Mass
Spectrom., 1993, 28, 1577.
31 R. A. J. O’Hair and G. E. Reid, Eur. J. Mass Spectrom., 1999, 5, 325;
G. E. Reid, K. D. Roberts, R. J. Simpson and R. A. J. O’Hair, J. Am.
Soc. Mass Spectrom., 2005, 16, 1131.
32 G. E. Reid, R. J. Simpson and R. A. J. O’Hair, J. Am. Soc. Mass
Spectrom., 2000, 11, 1047.
This journal is
The Royal Society of Chemistry 2011
Org. Biomol. Chem., 2011, 9, 2751–2759 | 2759
©