Mass spectrometry and theoretical studies on N-C bond cleavages in the N-sulfonylamidino thymine derivatives

Article abstract of DOI:10.1007/s13361-014-1068-8

Abstract The reactivity of new biologically active thymine derivatives substituted with 2-(arylsulfonamidino)ethyl group at N1 and N3 position was investigated in the gas phase using CID experiments (ESI-MS/MS) and by density functional theory (DFT) calculations. Both derivatives show similar chemistry in the negative mode with a retro-Michael addition (Path A^-) being the most abundant reaction channel, which correlate well with the fluoride induced retro-Michael addition observed in solution. The difference in the fragmentation of N-3 substituted thymine 5 and N-1 substituted thymine 1 in the positive mode relates to the preferred cleavage of the sulfonyl group (m/z 155, Path B) in N-3 isomer and the formation of the acryl sulfonamidine 3 (m/z 309) via Path A in N-1 isomer. Mechanistic studies of the cleavage reaction conducted by DFT calculations give the trend of the calculated activation energies that agree well with the experimental observations. A mechanism of the retro-Michael reaction was interpreted as a McLafferty type of fragmentation, which includes H_β proton shift to one of the neighboring oxygen atoms in a 1,5-fashion inducing N1(N3)-C_α bond scission. This mechanism was found to be kinetically favorable over other tested mechanisms. Significant difference in the observed fragmentation pattern of N-1 and N-3 isomers proves the ESI-MS/MS technique as an excellent method for tracking the fate of similar sulfonamidine drugs. Also, the observed N-1 and/or N-3 thymine alkylation with in situ formed reactive acryl sulfonamidine 3 as a Michael acceptor may open interesting possibilities for the preparation of other N-3 substituted pyrimidines.

Full text of DOI:10.1007/s13361-014-1068-8

B American Society for Mass Spectrometry, 2015
J. Am. Soc. Mass Spectrom. (2015)
DOI: 10.1007/s13361-014-1068-8
RESEARCH ARTICLE
Mass Spectrometry and Theoretical Studies on N–C Bond
Cleavages in the N-Sulfonylamidino Thymine Derivatives
Renata Kobetić,1 Snježana Kazazić,2 Borislav Kovačević,3 Zoran Glasovac,4
Luka Krstulović,5 Miroslav Bajić,5 Biserka Žinić1
1Laboratory of Supramolecular and Nucleoside Chemistry, Division of Organic Chemistry and Biochemistry, Ruđer Bošković
Institute, 10000, Zagreb, Croatia
2Laboratory for Chemical Kinetics and Atmospheric Chemistry, Division of Physical Chemistry, Ruđer Bošković Institute, 10000,
Zagreb, Croatia
3Group for Quantum Organic Chemistry, Division of Organic Chemistry and Biochemistry, Ruđer Bošković Institute, 10000,
Zagreb, Croatia
4Laboratory for Physical-organic Chemistry, Division of Organic Chemistry and Biochemistry, Ruđer Bošković Institute, 10000,
Zagreb, Croatia
5Department of Chemistry and Biochemistry, Faculty of Veterinary Medicine, University of Zagreb, 10000, Zagreb, Croatia
Abstract. The reactivity of new biologically active thymine derivatives substituted with
2-(arylsulfonamidino)ethyl group at N1 and N3 position was investigated in the gas
phase using CID experiments (ESI-MS/MS) and by density functional theory (DFT)
–
calculations. Both derivatives show similar chemistry in the negative mode with a
retro-Michael addition (Path A ) being the most abundant reaction channel, which
correlate well with the fluoride induced retro-Michael addition observed in solution.
The difference in the fragmentation of N-3 substituted thymine 5 and N-1 substituted
thymine 1 in the positive mode relates to the preferred cleavage of the sulfonyl group
(m/z 155, Path B) in N-3 isomer and the formation of the acryl sulfonamidine 3 (m/z
309) via Path A in N-1 isomer. Mechanistic studies of the cleavage reaction conduct-
ed by DFT calculations give the trend of the calculated activation energies that agree well with the experimental
observations. A mechanism of the retro-Michael reaction was interpreted as a McLafferty type of fragmentation,
which includes Hβ proton shift to one of the neighboring oxygen atoms in a 1,5-fashion inducing N1(N3)–Cα bond
scission. This mechanism was found to be kinetically favorable over other tested mechanisms. Significant
difference in the observed fragmentation pattern of N-1 and N-3 isomers proves the ESI-MS/MS technique as
an excellent method for tracking the fate of similar sulfonamidine drugs. Also, the observed N-1 and/or N-3
thymine alkylation with in situ formed reactive acryl sulfonamidine 3 as a Michael acceptor may open interesting
possibilities for the preparation of other N-3 substituted pyrimidines.
Key words: N-sulfonylamidines, Thymine, Retro-Michael addition, ESI-MS, DFT calculations
Received: 23 October 2014/Revised: 10 December 2014/Accepted: 12 December 2014
antitumor activity under in vitro [1, 2] and in vivo conditions
[3, 4]. To further explore the biological potential of these types
Introduction
e have recently shown that N-sulfonylpyrimidine
derivatives of type I (Figure 1) possess strong
of molecules, their structure was modified by combining them
with another potent anticancer pharmacophore—the amidine.
The amidine group is present in many compounds capable of
interacting with a wide range of biological targets [5], resulting
in anti-degenerative [6], anticancer [7, 8] and antimicrobial
activities [9]. These findings encouraged us to explore the
synthesis and chemistry of novel N-sulfonylamidino deriva-
tives containing pyrimidine moiety II (Figure 1) and to study
their biological properties [10].
W
Electronic supplementary material The online version of this article
(doi:10.1007/s13361-014-1068-8) contains supplementary material, which is
available to authorized users.
Correspondence to: Borislav Kovačević; e-mail: boris@irb.hr, Zoran
Glasovac; e-mail: glasovac@irb.hr, Biserka Žinić; e-mail: bzinic@irb.hr

R. Kobetić et al.: N-Sulfonylamidino Thymine Derivatives
X
N
methods. Hydrogen/deuterium exchange experiments were
performed in order to confirm the number of exchangeable
protons and additionally support fragmentation pathways.
X
R'
NH
R'
NH
O
O
S
N
S
O
N
R1
O
O
X = O, NH
R' = H, Me
R
O
N
Experimental
I
II
R2
R3
Preparation of N-Alkyl N-Sulfonylamidino Thymine
Derivatives
Figure 1. N-sulfonylpyrimidine I and N-sulfonylamidinopyrimidine
II derivatives
N-1 substituted thymine 1 [10] was reacted with
tetrabutylammonium fluoride (TBAF) in dry THF producing
new N-3 substituted thymine 5 and N-1, N-3-disubstituted
thymine 4 along with unreacted starting compound 1.
Additionally, acryl sulfonamidine 3 and free thymine 2 were
also identified, confirming that the reaction proceeds through
the retro-Michael/Michael addition reaction sequence. Details
of the reaction are given in Supplemental Material. The reac-
tion products, five well separated spots on silica gel plates,
were scraped off and extracted with methanol. The structures of
Application of these derivatives as drugs asks for a method
that can successfully track the fate of these compounds in
organism. HPLC coupled with ESI-MS/MS is often an instru-
mental technique of choice for this purpose [11, 12], and
knowledge of the fragmentation pattern of the starting material
is required. Until now, many papers dealing with usual and
unusual fragmentation of sulfonamides have been published
[13–17], while, to the best of our knowledge, fragmentation of
sulfonamidines has not been investigated. Upon CID experi-
ments conducted on simple alkyl-benzenesulfonamides,
Attygalle and coworkers noticed a loss of benzenesulfonamide
or its protonated form producing alkyl cation or neutral alkene
depending on the structure of alkyl substituent [18]. They also
observed a low abundance of benzenesulfonyl cation and some
other second generation ions. One could assume that these
fragments will also appear upon CID experiments conducted
on sulfonamidines. Additionally, product ions resulting from the
retro-Michael addition of the sulfonamidine II are also expected.
Several years ago, Walczak and coworkers observed Michael/
retro-Michael reaction equilibrium on the N-1- and N-3-(2-
(methylcarboxy)ethyl)thymine under basic conditions [19].
They found significant difference in rates of retro-Michael addi-
tion depending on whether substituent is attached to N1 or N3
position of thymine subunit. Thus, the cleavage of C–N1 bond in
a retro-Michael reaction occurs under mild conditions, whereas
the same reaction starting from N-3 substituted derivative de-
mands significantly higher temperatures. At the same time, the
Michael reagent formed upon cleavage of C–N1 bond reacts
with thymine producing significant amounts of N-3 substituted
derivatives. Previous theoretical and experimental gas-phase
investigations of retro-Michael addition also showed that this
reaction is a low barrier process and therefore favorable in the
gas-phase regardless of the charge state [20].
1
the products were assigned according their H and 13C NMR
spectra and additionally confirmed by the ESI-MSn studies.
Both methods proved the presence of the products 1–5 in the
reaction mixture (Scheme 1, and Supplemental Material,
Figures S1–S5).
ESI-MS Analysis
The mass spectral data (ESI-MS and ESI-MS/MS) were ac-
quired on the Agilent 6410 Triple Quad mass spectrometer
from Agilent Technologies Inc. (Palo Alto, CA, USA)
equipped with electrospray ionization (ESI) interface operated
in positive and negative ion mode. The samples were prepared
in CH3OH to a concentration of about 0.05 mg/mL and directly
infused. The infusion into the mass spectrometer was per-
formed at a flow rate of 3 μL/min. Nitrogen was used as
auxiliary and sheath gas. Spray voltage was set at 4.0 kV, the
capillary temperature was 300°C, and the voltage range of the
collision cell was 80–180 eV. Full mass spectra were acquired
over the mass range m/z 10–2000.
MSn (n 9 2) were recorded on the amaZon ETD mass
spectrometer (Bruker Daltonik, Bremen, Germany) equipped
with the standard ESI ion source (Nebulizer pressure: 8 psi;
drying gas flow rate: 5 L/min; drying gas temperature: 250°C).
The mass spectrometer was operated in positive and negative
polarity mode; the potential on capillary cap was 4500 V.
Helium was used as collision gas. The compounds were dis-
solved in methanol to a concentration of approximately
0.5 × 10–6 mol/dm3 and injected into the ESI source of
the mass spectrometer by syringe pump at a flow rate of
1 μL/min.
Thymine derivatives employed by Walczak and coworkers
are structurally similar to our sulfonamidine II and, having this
in mind, one can ask a question whether there is a significant
difference in the fragmentation pattern of the N-1 and N-3
substituted thymine. Our aim was, therefore, to investigate
reactivity of the N-1 and N-3 subsitututed sulfonamidine in
the gas phase and solution. Obtained results will provide useful
information that can be used to track these compounds during
biological testing. Additionally, observed reactivity of these
sulfonamidines will help us to tune desired properties of the
second generation sulfonamidine drugs more easily. MS stud-
ies were performed in both ES+ and ES– modes and the mech-
anisms of the fragmentations were investigated using DFT
Calculation Methods
All calculations were performed by Gaussian09 [21].
Structures of each stationary point along the reaction pathways
were optimized at the B3LYP/6-31G(d) level of theory [22–26]
and the electronic energies were obtained from the single point

R. Kobetić et al.: N-Sulfonylamidino Thymine Derivatives
N
N
p-Ts
H3C
SO2
O
p-Ts =
O
N
O
N
N
N
p-Ts
3
NH
O
O
1
N
p-Ts
N
O
N
NH
N
TBAF
THF
p-Ts
+
+
+
N
N
N
H
O
N
O
p-Ts
N
H
N
(19 %)
(15 %)
(4 %)
(13 %)
4
(49 %)
5
1
2
3
Scheme 1. Reaction of 1 with tetrabutylammonium fluoride (TBAF). Approximate relative abundances (molar ratios) of each
compound are given in parentheses
energy calculations using the M06-2X/6-311+G(3df,2p) [27]
method. Nature of each stationary point was verified by vibra-
tional analyses (NImag = 0 for minima and NImag = 1 for
transition state structures). Correctness of each transition state
structure (TS) was confirmed by the intrinsic reaction coordi-
nate following until reactant or product region on the PES was
reached without a doubt. Gibbs energies of all molecules were
calculated by using thermochemical data directly obtained by
calculations without any scaling or corrections for the low
vibrations. Structures were visualized and plotted by
MOLDEN 5.0 program [28]. All energies are given in kilocal-
ories per mole where 1 kcal = 4.184 kJ.
Considering the structures of the products, it appears that the
fluoride, as a sufficiently strong base in THF, induced elimination
of thymine 2, giving conjugated acryl sulfonamidine 3 as a
potential Michael acceptor. The formation of the N-1,N-3-
dialkylated thymine derivative 4 could be explained by the
Michael-type addition of starting compounds 1 (F– assists the
creation of N3 anion in 1) to acrylic derivative 3, whereas the
Michael-type addition of thymine 2 anion to acrylic derivative 3
could lead to the formation of the N-1-alkylated 1 and N-3-
alkylated product 5 as reported for similar alkylation reactions
with other pyrimidine derivatives [30, 31]. Observed product
composition is in qualitative agreement with the known reactivity
of N-1 and N-3 substituted thymine derivatives [19].
We also wanted to test possibility of S–N bond cleavage in
the sulfonamidine 1 under different conditions since this process
was identified as one of two dominant processes in the gas phase
(see ESI-MS results below). Sulfonyl groups are used to protect
amines or other nitrogen containing functionalities and they are
difficult to remove under standard hydrolytical conditions [32].
Instead, reductive methods are usually employed, such as the use
of sodium amalgam [33, 34], Na-naphthalenide [35–37], or
magnesium in methanol [38, 39]. In our hands, application of
Na-naphthalenide or magnesium in methanol to remove the
sulfonyl group in N-1 substituted thymine 1 appeared unsuc-
cessful and mostly the starting material was isolated.
Results and Discussion
Driven by the results of Walczak and coworkers [19], we
decided to investigate the reaction of sulfonamidine 1
(Scheme 1) with fluoride anion as a base. Efficient removal
of the arylsulfonyl group from the sulfonamidino compounds
by TBAF in THF has been described in literature [29] and,
therefore, this reagent was used as a source of fluoride anion
expecting amidine deprotection as the significant competitive
side-reaction. Using tetrabutlyammonium fluoride (TBAF)
[29] to remove the sulfonyl group in 1 gave a complex reaction
mixture containing starting material 1, thymine 2, and product
compounds 3–5 (Scheme 1) while no products of desulfonation
reaction were found.
Since the retro-Michael addition was found to be the only
process in the solution that encompasses bond cleavage, we
conducted a detailed mechanistic investigation of the
fragmentation of 1 (N-1) and 5 (N-3) compounds in
Path A
Path A
+
+
H
H
m/z 309
very weak
major
O
N
Path B
N
Path B
strong
NH
m/z 155
m/z 264
O
major
O
S
O
O
N
S
N
N
O
Path C
weak
Path C
N
O
medium
N
H
O
Path D
weak
Path D
m/z 334
very weak
[1H]+
[5H]+
m/z 435
m/z 435
Scheme 2. Product ions obtained via four primary fragmentation pathways starting from signal at m/z 435 assigned as [MH]+
adduct of the compounds 1 or 5. Path D was observed at Elab ≥ 10 eV. Qualitative description of the intensities is given for the spectra
at the collision energy Elab = 10 eV

R. Kobetić et al.: N-Sulfonylamidino Thymine Derivatives
Figure 2. Full (bottom) and stretched out (top) spectra of product ion spectra of the protonated N-1 substituted thymine [1H]+ and
its N-3 substituted isomer [5H]+ in ES+ mode at the collision energy Elab = 10 eV
order to clarify, understand, and explain the preference
of the retro-Michael reaction against other possible frag-
mentation mechanisms.
for ESI-MS/MS studies. The spectra were obtained at the same
concentrations (in MeOH) and conditions in both ES+ and ES–
modes. The fragmentation pathways for all analyzed com-
pounds were proposed based on MS/MS and MSn spectra of
the ions obtained by protonation ([MH]+) or deprotonation
([M–H]–) of the starting molecule.
ESI MS Results
Transformation of compound 1 (Scheme 1) was studied by
Our first observation was that N-1 substituted thymine 1 and
ESI, using the ion trap for MSn studies, and triple-quadrupole
N-3 substituted thymine 5 formed protonation products 1H+
Figure 3. Product ion spectra of the monodeuterated N-1 substituted thymine [1D]+ and its N-3 isomer [5D]+ in ES+ mode at Elab = 5,
10 and 15 eV

R. Kobetić et al.: N-Sulfonylamidino Thymine Derivatives
and 5H+, respectively, as well as several adducts with Na+ ions
in the gas phase with ligand to sodium ratio varying from 1:1 to
4:1. We have noticed that CID experiments conducted on the
[1H]+ and [1Na]+ produced similar fragmentation scheme.
Significant difference between fragmentation patterns starting
from 1 (N-1) and 5 (N-3) has been observed in both positive
and negative mode. Although the results obtained in negative
mode could be better correlated with our experimental results,
we shall start with discussion of the results obtained in positive
mode since the differences in fragmentation channels are
richer. Four fragmentation channels for compound 1 (N-1)
were observed and identified in ES+ mode (Scheme 2) as
follows: (1) signal at m/z 309 assigned as [MH]+ acryl
sulfonamidine 3, gained because of cleavage of N1–CH2 bond
and elimination of thymine base (Path A); (2) signal at m/z 155
assigned as [C7H7SO2]+ sulfonyl group (Path B); (3) signal at
m/z 264 obtained because of elimination of the neutral sulfon-
amide [C7H7SO2NH2], (Path C), and (4) signal located at m/z
334 corresponding to the elimination of di-isopropylamine
(Path D). As it can be seen clearly in Figure 2, the signal at
m/z 309 assigned as [MH]+ of acryl sulfonamidine 3 is the most
abundant in the product ion spectra starting from [1H]+, where-
as signal at m/z 155 assigned as [C7H7SO2]+ sulfonyl group
(Path B) dominates in the product ion spectra for [5H]+. It
should be noted that the signal at m/z 155 is also observed in
the spectra of 1 and the relative abundance of this ion is similar
in CID product spectra of both N-1 and N-3 isomers. Several
other peaks (m/z 266, 222, 180, etc.) are also visible in the CID
spectra that are the product ions of two or more consecutive
fragmentation processes that have been confirmed by MS2 and
MS3 experiments (Figures S6 and S7 in Supplemental
Material) and mechanisms of their formation will not be
discussed any further.
By the increase of the collision energy Elab from 5 to 15 eV,
the fragmentation channels starting from [1H]+ switched prior-
ities and the loss of the tosyl group (Path B, m/z 155; at Elab
(a)
O
N
O
S
O
N
N
O
O
S
N
N
H
O
N
N
N
major path
(Path A-)
O
+
O
m/z 125
[1-H]-
O
at higher eVs
N
O
-
N
N
S
N
C
O
+
O
m/z 42
(b)
O
O
S
N
N
S
O
N
N
N
O
major path
H
O
O
N
+
(Path A-)
N
O
N
O
m/z 125
[5-H]-
O
N
O
H
N
O
minor path
(Path D-)
S
N
O
m/z 332
Scheme 3. Proposed structures of signals observed for fragmentation of compounds: a) [1-H]– and b) [5-H]–, in ES– mode

R. Kobetić et al.: N-Sulfonylamidino Thymine Derivatives
5 eV this signal was below threshold) became the main frag-
mentation channel (Figure 3 and Figure S8 in Supplemental
Materials).
the product ions spectra for [1-H]– also showed departure of
NCO– (m/z 42) as the minor path at higher collision energies
(Scheme 3a). Appearance of NCO– fragment in the negative
mode CID spectra of the ion [1-H]– indicates substitution
position, additionally confirming the structure of the starting
ion [40]. This could be rationalized by analogy to the fragmen-
tation of a thymine anion. Thymine N1-H deprotonation led to
an anion, which was stabilized by resonance through the ring
where O4 accepted charge what was confirmed by the N1-H
acidity decrease in O4 substituted derivative, Scheme 3 [41].
Owing to the higher gas-phase acidity (GA) of N1-H compared
with N3-H the corresponding anion with the charge located on
N1 is more stable and ring cleavage does not proceed through
deprotonation at N1. In line with this, the fragmentation of [5-
H]– did not show formation of NCO– fragment even under the
higher energy CID conditions. Our results agree well with
previous experimental and theoretical results for thymine, N-
1-methylthymine, N-3-methylthymine, O-2-methylthymine,
and O-4-methylthymine [42].
Additionally, H/D exchange experiments were conducted as
follows: the samples for H/D exchange were dissolved in D2O
to a concentration of about 0.05 mg/mL and directly infused. In
this way, the exchange of the most acidic protons (NH) was
expected. Thus, the parent ions [1H]+ and [5H]+ showed two
additional signals above the threshold of the measurements
corresponding to the mono- ([1D]+ and [5D]+) and di-
deuterated ([1D2]+ and [5D2]+) ions. The gas fragmentation
experiments on these none-, mono- and dideuterated ions
showed that the H/D exchange does not affect the fragmenta-
tion pattern showing, in the case of [1H]+, the same switch it
the preference of the fragmentation channels upon increase of
the collision energy. Fragmentation of both the mono-
deuterated ([1D]+ and [5D]+) derivatives led to the extrusion
of either mono- or non-deuterated thymine (product ions at m/z
= 309 and 310). Additionally, presence of the product ion with
m/z 311 in the CID spectra of dideuterated ions ([1D2]+ and
[5D2]+, Figure S8 in Supplementary Material) indicate the
presence of the highly enolizable CH hydrogen atoms, which
could undergo H/D exchange.
The apparent difference in the fragmentation of N-3 substitut-
ed [5H]+ and N-1 substituted [1H]+ is the preferred cleavage of
the sulfonic group (m/z 155) in the former and the formation of
the acryl sulfonamidine [3H]+ (m/z 309) in the latter. This
suggests that the substituent at N1 undergoes easier cleavage
compared with the N3 substituent. In accord with the latter
stands the observation that the fragmentation channels for the
protonated N-1, N-3-disubstituted thymine [4H]+ show forma-
tion of acryl sulfonamidine [3H]+ by preferred cleavage at the
N1 thymine position signifying that the N3–C bond is stronger
than the N1–C bond in [MH]+. Afore-mentioned experimental
findings observed in the positive mode were rationalized using
DFT calculations as discussed in the next section.
Calculation Results
To explain the observed fragmentation patterns in positive and
negative mode described in the previous section, we calculated
reaction Gibbs energies (ΔrG) and the activation energies (Ea)
for the primary reaction pathways leading to a fragmentation of
the starting ion into two fragments. Again, we shall start our
discussion by analyzing the reaction paths at the cationic po-
tential energy surface. Optimized structure of the starting ion
[1H]+ is presented in Figure 4.
Apart from Path B, all other observed processes are initiated
by the Hβ proton shifts to either thymine subunit (Path A) or
toward one of the amidine nitrogen atoms (NSA or NiPr, Paths C
and D, respectively, structures 9 or 12 in Scheme 4), and our
A difference in fragmentation of the N-3 substituted 5 versus
N-1 substituted 1 was also observed by the CID experiments in
the ES– mode. Fragmentation of [1H]+ and [5H]+ proceeded
with few product ions formed (Scheme 3, and Figure S9 in
Supplemental Materials).
Retro-Michael addition process leading to the extrusion of
free thymine base anion (m/z 125) was found to be the domi-
nant reaction channel for both isomeric starting anions
(Scheme 3b). This process corresponds to the Path A described
also as the most abundant reaction channel for [1H]+ in positive
mode (Scheme 2). To emphasize similarity of these fragmen-
tation paths, we shall term the negative mode path as Path A–.
On the other hand, no N–S bond cleavage products (Path B,
Scheme 2) or the sulfonamidine extrusion (Path C, Scheme 2)
were obtained upon CID of any of the considered isomeric
anions. The low energy product ion spectra of [5-H]– anion in
negative mode showed the departure of di-isopropylamine
(Scheme 3b), as the minor process with the ca 30% abundance
relative to the Path A–. The same was not observed in CID
experiments conducted on [1-H]–. It should be mentioned that
Figure 4. Optimal structure of the protonated N-1 substituted
thymine 1H+ calculated at B3LYP/6-31G(d) level of theory

R. Kobetić et al.: N-Sulfonylamidino Thymine Derivatives
H
H
S
CH2
iPr
CH2
iPr
CH3
O
C
CH3
O
N
N
O
N
O
Path A
N
iPr
iPr
+
H
N+
N+
H
H
O
N
H
H
O
N
H
S
O
O
pTol
pTol
[3H]+
6
Ea(A)
iPr
iPr
+
H
H
H
H
SO2
N
N
CH3
O
CH3
O
iPr
iPr
N
N
Path B
+
O
N+
Ea(B)
N
O
H
O
N
H
O
N
H
S
H
pTol
CH3
[1H]+
7
8
H
C
SO2NH2
iPr
H
CH3
O
N
N
CH3
O
iPr
N
Path C
+
iPr
H
N+
O
O
N
N+
Ea(C)
O
H
O
N
H
S
H
iPr
CH3
pTol
10
11
9
iPr
H
H
CH3
O
N+
CH3
O
N
iPr
O
N
Path D
Ea(D)
O
(iPr)2NH
+
O
N
N
O
S
N
H
O
H
O
N
H
S
pTol
H
pTol
13
iPr
H
H
H
H
CH2
iPr
CH2
iPr
CH3
C
N
CH3
O
CH3
O
Path A-
N
N
O
N
O
iPr
N
N
iPr
iPr
+
O
H
N+
N
N
H
H
O
N
O
O
O
N
S
O
N
S
S
O
O
pTol
pTol
pTol
[1-H]-
Ea(A-)
[6-H]-
3
Scheme 4. Schematic representation of four investigated fragmentation paths starting from the N-1 isomer [1H]+ and the
McLafferty-type fragmentation path (Path A–) starting from the anion [1-H]–
calculations predict that these proton shifts are the rate-
determining steps along the corresponding reaction paths. In
the case of Path B, fragmentation proceeds by direct S–NSA
bond cleavage without any proton shift associated with this
process.
The lowest energy transition state structures calculated for
all four primary fragmentation paths are shown in Figure 5, and
the structures of the cumulative results of kinetic and thermo-
dynamic investigations of the four fragmentation pathways are
shown in Table 1.
The McLafferty type of fragmentation was found to be
kinetically favorable over the other tested mechanisms (see
Supplemental Materials, Figure S11). This mechanism in-
cludes Hβ proton shift to one of the neighboring oxygen atoms
in a 1,5-fashion inducing N1(N3)–Cα bond scission and pro-
ceeding further to the products through the formation of the six-
membered transition state (TSA, Figure 5). A six-membered
cyclic transition state can also be formed by the amidine proton
shift toward N1, which would also lead to the N1–Cα bond
scission. However, the barrier calculated for this process is
significantly higher (63.8 and 72.7 kcal mol–1 in [1H]+ and
[5H]+, respectively; see Supplemental Material, Figure S11).
The reason for this lies in a direct formation of the acrylic
double bond in the McLafferty type of rearrangement, whereas
Locating a minimum energy reaction path for Path A ap-
peared to be the most demanding task. Four different routes
were found, each of them initiated by a proton shift from
ethyleneamidine moiety toward thymine subunit.

R. Kobetić et al.: N-Sulfonylamidino Thymine Derivatives
Figure 5. Optimized transition state structures for the fragmentation paths A-D of N-1 substituted derivative [1H]+
in the case of amidine (NH) proton migration, unstable primary
carbocation needs to be formed. Another considered possibility
was Hβ-to-N1 1,3-proton shift, which would also lead to a
direct formation of acrylic bond. Such reaction involves the
tight four-membered transition state, which is connected to the
relatively high activation energy (see Supplemental Materials,
Figure S11). The fourth possibility is direct C–N1(N3) bond
cleavage, which ends up in formation of the four-membered
ring instead of acrylamidine. Calculated barrier of
77.3 kcal mol–1 for [1H]+ (72.3 kcal mol–1 for [5H]+) for this
process also rules it out as the least favorable possibility.
Our calculations indicate that migration of the same Hβ
proton also triggers Paths C and D (Scheme 4 and TSC and
TSD in Figure 5) where 1,3-proton shift toward either of two
amidine nitrogen atoms leads to a formation of the less stable
intermediates 9 and 12 (Scheme 4), which undergo extrusion of
the neutral sulfonamide or diisopropylamine in the energetical-
ly favorable processes. Again, an alternative pathway for a
migration of the proton suited at thymine subunit (HT,
Figure 4) toward the amidine group was tested. This pathway
involves a three-step process in which carbonyl oxygen serves
as a mediator for the proton shift (see Supplemental Material
Figure S12). The latter mechanism is calculated to have a
slightly higher barrier (75.7 kcal mol–1 in the case of N-1
derivative) than Hβ proton shift. Thus, the one-step Hβ proton
shift is considered to be more probable if compared with the
three-step alternative. Finally, extrusion of the tolylsulfonyl
cation 8 (Path B in Scheme 5 and TSB in Figure 5) is a
consequence of direct N–S bond cleavage and is not activated
by any proton shift.
Such processes usually occur through a late transition state
possessing large density of states. Consequently, these frag-
mentation channels can be preferred regardless of their higher
activation energies [43].
Paths C and D are strongly disfavored over Paths A and B
because of their high activation energies (ca 70 kcal mol–1).
This is particularly true for the extrusion of diisopropylamine
(Path D), which starts to appear at higher collision energies
with respect to other paths as commented earlier in the text.
However, a slightly higher abundance of Paths C and D over
Path A (Figure 2) in the case of the N-3 isomer is still surprising
and may indicate another channel leading to the loss of the
signal with m/z = 309. Indeed, fragmentation of the signal with
m/z = 309 led to desulfonation of the amidine group producing
dominantly an ion with m/z = 155 (Figure S6 in Suppl. Mat.)
which, in conjunction with higher barrier for the Path A, could
be responsible for the very weak intensity of [3H]+ in the
product ion spectra of [5H]+.
Finally, a brief analysis of the Path A– on the anionic
potential energy surface was done. McLafferty-like transition
state analogous to the one found for the positive mode frag-
mentation was identified. The calculations predict the even
lower activation energies amounting to 26 and 34 kcal mol–1
Table 1. Activation Energies (Ea) of the Rate-Determining Reaction Steps and
the Gibbs Energies for the Overall Fragmentation Process (ΔrG) Along the
Investigated Fragmentation Pathsa,b
Precursor ions: [1H]+ or [1-H]–
Precursor ions: [5H]+ or [5-H]–
Paths
Ea(Path)
ΔrG(Path)
Ea(Path)
ΔrG(Path)
In the case of N-1 derivative, experimentally observed re-
sults agree well with the trend in the calculated activation
energies. Activation energy for the fragmentation via Path A
is the lowest one and consequently it is the most populated
fragmentation channel. According to the calculations, Path B is
disfavored with respect to Path A by 13.5 and 6.4 kcal mol–1
for N-1 and N-3 derivative, respectively. A smaller difference
in the activation energy is apparently sufficient to induce a
switch in the preference of the fragmentation channels and an
almost complete retardation of Path A. A fragmentation via the
Path B arises from a simple stretching of the sulfonamide N–S
bond and does not require to be activated by the proton shift.
A
B
C
35.5
49.0
69.2
74.3
26.0
17.7
62.3
20.4
31.8
19.7
41.4
48.0
71.3
76.9
33.9
28.9
65.3
20.7
34.9
28.1
D
A–
aAll energies are given in kcal mol–1 and were calculated using M06-2X/6-
311+G(3df,2p) //B3LYP/6-31G(d) level of theory
bStructures of the precursor ions [1H]+ and [1-H]–, their product ions, and
calculated intermediates along the investigated reaction paths are schematically
presented in Scheme 4 whereas the corresponding structures derived from [5H]+
and [5-H]– are given in Scheme S1 in the Supplemental Material. Complete
energy profiles for the fragmentation processes are given in Figure S10 in
Supplemental Material

R. Kobetić et al.: N-Sulfonylamidino Thymine Derivatives
for N-1 and N-3 isomers, respectively. Difference in the acti-
vation energies of 8 kcal mol–1 is also higher than the corre-
sponding difference obtained for the same processes in the
positive mode. In the negative mode, no mobile protons were
present at either thymine or amidine subunits, additionally
supporting our finding that shift of Hβ proton actually triggers
the elimination of thymine. It is reasonable to assume that the
presence of base additionally promotes the mobility of Hβ
proton, leading to the further lowering of the barrier in the
basic solution. These findings are in full accordance with the
experiment in solution conducted in this work (Scheme 1) and
explain strong preference of the retro-Michael reaction over
desulfonation and other fragmentation reactions.
N-3,N-1-disubstituted thymine 4, acryl sulfonamidine 3, thy-
mine 2, and starting N-1 substituted thymine 1 as a result of the
Michael/retro-Michael addition reactions. On the other hand,
expected removal of the sulfonyl group from the amidine
nitrogen atom was not obtained. The observed in situ N-1
and/or N-3 thymine alkylation with reactive acryl
sulfonamidine 3 as the Michael acceptor may open interesting
possibilities for the preparation of other N-3 substituted
pyrimidines.
Acknowledgments
This work was supported by the Ministry of Science, Educa-
tion, and Sports of the Republic of Croatia through grants no.
098-0982914-2935; 053-0982914-2965; 098-0982915-2945.
Z.G. and B.K. gratefully acknowledge support of the Comput-
ing Center of the University of Zagreb (SRCE) for granting
computational time on ISABELLA cluster.
Conclusions
Our investigations show that the N-1 and N-3
tolylsulfonamidino substituted thymine derivatives show sig-
nificant difference in fragmentation patterns and could be eas-
ily differentiated under ESI-MS/MS conditions. Thus, ESI-MS
experiments conducted on the N-1 derivative showed elimina-
tion of acrylamidine (retro-Michael addition) as the most abun-
dant reaction channel in the gas phase under applied conditions.
Besides other less important fragmentation pathways, elimina-
tion of the sulfonyl group proved to be a competitive process in
the positive mode but only under higher energy conditions. On
the other hand, N-3 derivative 5 is found to be more inert
toward the elimination of acrylamidine, which is ascribed to a
stronger N–Cα bond and somewhat higher barrier. In this case,
removal of the sulfonyl group was found to be a dominant
process. Detailed quantum chemical investigations of the ob-
served processes in the gas-phase reveal H(Cβ) proton as suf-
ficiently labile, triggering most of the observed primary frag-
mentation paths. This is particularly true for the formation of
acryl sulfonamidine 3, which is formed in a kinetically favor-
able process via six-membered McLafferty type transition
state. Our calculations also reveal that the same process pro-
ceeds even more easily on the anionic potential energy surface,
thus providing a basis for explanation of the experimental
results. In contrast to the situation in alkyl sulfonamides, reac-
tion channel leading to the elimination of the tolylsulfonamide
(Path C) was found to be less populated than the one producing
sulfonyl cation (Path B). According to the theoretical and
experimental investigations in the gas phase, formed N-3 de-
rivative 5 is less prone to the C–N bond cleavage with respect
to the derivative 1. A significant difference in the observed
fragmentation pattern of N-1 and N-3 isomers proves the ESI-
MS technique as an excellent method for tracking the fate of
similar sulfonamidine drugs during the applications.
References
1. Žinić, B., Žinić, M., Krizmanić, I.: Synthesis of the sulfonylpyrimidine
derivatives with anticancer activity. EP 0 877 022 B1 (2003)
2. Supek, F., Kralj, M., Marjanović, M., Šuman, L., Šmuc, T., Krizmanić, I.,
Žinić, B.: Atypical cytostatic mechanism of N-1-sulfonylcytosine deriva-
tives determined by in vitro screening and computational analysis. Investig.
New Drugs 26, 97–110 (2008)
3. Pavlak, M., Stojković, R., Radačić-Aumiler, M., Kašnar-Šamprec, J., Jerčić,
J., Vlahović, K., Žinić, B., Radačić, M.: Antitumor activity of novel N-
sulfonylpyrimidine derivatives on the growth of anaplastic mammary carci-
noma in vivo. J. Cancer Res. Clin. Oncol. 131, 829–836 (2005)
4. Kašnar-Šamprec, J., Ratkaj, I., Mišković, K., Pavlak, M., Baus-Lončar, M.,
Kraljević Pavelić, S., Glavaš-Obrovac, L., Žinić, B.: In vivo toxicity study
of N-1-sulfonylcytosine derivatives and their mechanisms of action in
cervical carcinoma cell line. Investig. New Drugs 30, 981–990 (2012)
5. Greenhill, J.V., Lue, P.: Amidines and guanidines in medicinal chemistry.
Prog. Med. Chem. 30, 203–326 (1993)
6. Panico, A., Vicini, P., Incerti, M., Cardile, V., Gentile, B., Ronsisvalle, G.:
Amidinobenzisothiazole derivatives with antidegenerative activity on car-
tilage. Il Farmacol 57, 671–675 (2002)
7. Bielawski, K., Bielawska, A., Sosnowska, K., Miltyk, W., Winnicka, K.,
Palka, J.: Novel amidine analogue of melphalan as a specific multifunc-
tional inhibitor of growth and metabolism of human breast cancer cells.
Biochem. Pharmacol. 72, 320–331 (2006)
8. Stolić, I., Mišković, K., Piantanida, I., Baus Lončar, M., Glavaš-Obrovac,
L., Bajić, M.: Synthesis, DNA/RNA affinity and antitumour activity of new
aromatic diamidines linked by 3, 4-ethylenedioxythiophene. Eur. J. Med.
Chem. 46, 743–755 (2011)
9. Wilson, W.D., Tanious, F.A., Mathis, A., Tevis, D., Hall, J.E., Boykin,
D.W.: Antiparasitic compounds that target DNA. Biochimie 90, 999–1014
(2008)
10. Krstulović, L., Ismaili, H., Bajić, M., Višnjevac, A., Glavaš-Obrovac, L.,
Žinić, B.: Synthesis of novel aliphatic N-sulfonylamidino thymine deriva-
tives by Cu(I)-catalyzed three-component coupling reaction. Croat. Chem.
Acta 85, 525–534 (2012)
11. Tiller, P.R., Land, A.P., Jardine, I., Murphy, D.M., Sozio, R., Ayrton, A.,
Schaefer, W.H.: Application of liquid chromatography-mass spectrometry
analyses to the characterization of novel glyburidemetabolites formed
in vitro. J. Chromatogr. A 794, 15–25 (1998)
The mass spectral results are in a good agreement with the
experiments in solution conducted in this work, where the
reaction of N-1-alkyl-N-sulfonylamidino thymine 1 with F–
(TBAF) in THF proceeded by the cleavage of thymine N1–
CH2 bond resulting in the formation of the mixture of the
products identified as the N-3 substituted thymine 5, the
12. Volmer, D.A.: Multiresidue determination of sulfonamide antibiotics in
milk by short-column liquid chromatography coupled with electrospray
ionization tandem mass spectrometry. Rapid Commun. Mass Spectrom. 10,
1615–1620 (1996)
13. Hu, N., Liu, P., Jiang, K., Zhou, Y., Pan, Y.: Mechanism study of SO2
elimination from sulfonamides by negative electrospray ionization mass
spectrometry. Rapid Commun. Mass Spectrom. 22, 2715–2722 (2008)

R. Kobetić et al.: N-Sulfonylamidino Thymine Derivatives
14. Xu, G., Huang, T., Zhang, J., Huang, J.K., Carlson, T., Miao, S.:
Investigation of collision-induced dissociations involving odd-electron
ion formation under positive electrospray ionization conditions using ac-
curate mass. Rapid Commun. Mass Spectrom. 24, 321–327 (2010)
15. Sun, M., Dai, W., Liu, D.Q.: Fragmentation of aromatic sulfonamides in
electrospray ionization mass spectrometry: elimination of SO2 via rear-
rangement. J. Mass Spectrom. 43, 383–393 (2008)
16. Wang, Z., Hop, C.E.C.A., Mi-Sook, K., Huskey, S.-E.W., Baillie, T.A.,
Guan, Z.: The unanticipated loss of SO2 from sulfonamides in collision-
induced dissociation. Rapid Commun. Mass Spectrom. 17, 81–86 (2003)
17. Hibbs, J.A., Jariwala, F.B., Weisbecker, C.S., Attygalle, A.B..: Gas phase
fragmentations of anions derived from N-phenyl benzenesulfonamides. J.
Am. Soc. Mass Spectrom. 24, 1280–1287 (2013)
18. Bialecki, J.B., Weisbecker, C.S., Attygalle, A.B..: Low-energy collision-
induced dissociation mass spectra of protonated p-toluenesulfonamides
derived from aliphatic amines. J. Am. Soc. Mass Spectrom. 25, 1068–
1078 (2014)
19. Boncel, S., Maczka, M., Walczak, K.Z.: Michael versus retro-Michael
reaction in the regioselective synthesis of N-1 and N-3 uracil adducts.
Tetrahedron 66, 8450–8457 (2010)
27. Zhao, Y., Truhlar, D.G.: The M06 suite of density functionals for main
group thermochemistry, thermochemical kinetics, noncovalent interactions,
excited states, and transition elements: two new functionals and systematic
testing of four M06-class functionals and 12 other functionals. Theor.
Chem. Accounts 120, 215–241 (2008)
28. Schaftenaar, G., Noordik, J.H.: Molden: A Pre- and post-processing pro-
gram for molecular and electronic structures. J. Comput. Aided Mol. Des.
14, 123–134 (2000)
29. Yasuhara, A., Sakamoto, T.: Deprotection of N-sulfonyl nitrogen-
heteroaromatics with tetrabutylammonium fluoride. Tetrahedron Lett. 39,
595–596 (1998)
30. Boncel, S., Gondela, A., Walczak, K.: Uracil as a target for nucleophilic and
electrophilic reagents. Curr. Org. Synth. 5, 365–396 (2008)
31. Boncel, S., Walczak, K.: The Influence of base on regioselectivity of 5-
substituted uracils addition to michael acceptors. Lett. Org. Chem. 3, 534–
538 (2006)
32. Green, T.W., Wuts, P.G.M.: Protective Groups in Organic Synthesis, pp.
604–607; 744–747. Wiley-Interscience, New York (1999)
33. Ranganathan, R., Larwood, D.: Facile conversion of adenosine into new 2′-
substituted-2′-deoxy-arabinofuranosyladenine derivatives: stereospecific
syntheses of 2′-azido-2′-deoxy- 2′-amino-2′-deoxy-, and 2′-mercapto-2′-
deoxy-β-D-arabinofuranosyladenines. Tetrahedron Lett. 19, 4341–4344
(1978)
34. Tanner, D., Ming, H.H., Bergdahl, M.: Stereo-controlled synthesis of the
spirocyclic alkaloid ( )-nitramine. Tetrahedron Lett. 29, 6493–6496 (1988)
35. Suzuki, H., Unemoto, M., Hagiwara, M., Ohyama, T., Yokoyama, Y.,
Murakami, Y.: Synthetic studies on indoles and related compounds. Part
46.1 First total syntheses of 4,8-dioxygenated β-carboline alkaloids. J.
Chem. Soc. Perkin Trans. 1, 1717–1724 (1999)
36. Ji, S., Gortler, L.B., Waring, A., Battisti, A.J., Bank, S., Closson, W.D.,
Wriede, A.P.: Cleavage of sulfonamides with sodium naphthalene. J. Am.
Chem. Soc. 89, 5311–5312 (1967)
37. Yoo, E.J., Bae, I., Cho, S.H., Han, H., Chang, S.: A Facile access to N-
sulfonylimidates and their synthetic utility for the transformation to
amidines and amides. Org. Lett. 8, 1347–1350 (2006)
38. Sridhar, M., Kumar, B.A., Narender, R.: Expedient and simple method for
regeneration of alcohols from toluenesulfonates using Mg-MeOH.
Tetrahedron Lett. 39, 2847–2850 (1998)
39. Murase, M., Watanabe, K., Yoshida, T., Tobinaga, S.: A new concise
synthesis of arcyriacyanin A and its unique inhibitory activity against a
panel of human cancer cell line. Chem. Pharm. Bull. 48, 81–84 (2000)
40. Bald, I., Flosadóttir, H.D., Ómarsson, B., Ingólfsson, O.: Metastable frag-
mentation of a thymidine-nucleotide and its components. Int. J. Mass
Spectrom. 313, 15–20 (2012)
41. Tehrani, Z.A., Fattahi, A.: Comparison of gas phase intrinsic properties of
cytosine and thymine nucleobases with their O-alkyl adducts: different
hydrogen bonding preferences for thymine versus O-alkyl thymine. J.
Mol. Model. 8, 2993–3005 (2013)
42. Kurinovich, M.A., Lee, J.K.: The acidity of uracil and uracil analogs in the
gas phase: four surprisingly acidic sites and biological implications. J. Am.
Soc. Mass Spectrom. 13, 985–995 (2002)
20. Wang, H.-Y., Zhang, X., Guo, Y.-L., Lu, L.: Mass spectrometric studies of
the gas phase retro-Michael type fragmentation reactions of 2-
hydroxybenzyl-N-pyrimidinylamine derivatives. J. Am. Soc. Mass
Spectrom. 16, 1561–1573 (2005)
21. Kitao, O., Nakai, H., Vreven, T., Montgomery Jr., J.A., Peralta, J.E.,
Ogliaro, F., Bearpark, M., Heyd, J.J., Brothers, E., Kudin, K.N.,
Staroverov, V.N., Kobayashi, R., Normand, J., Raghavachari, K.,
Rendell, A., Burant, J.C., Iyengar, S.S., Tomasi, J., Cossi, M., Rega, N.,
Millam, J.M., Klene, M., Knox, J.E., Cross, J.B., Bakken, V., Adamo, C.,
Jaramillo, J., Gomperts, R., Stratmann, R.E., Yazyev, O., Austin, A.J.,
Cammi, R., Pomelli, C., Ochterski, J.W., Martin, R.L., Morokuma, K.,
Zakrzewski, V.G., Voth, G.A., Salvador, P., Dannenberg, J.J., Dapprich,
S., Daniels, A.D., Farkas, Ö., Foresman, J.B., Ortiz, J.V., Cioslowski, J.,
Fox, D.J.: Gaussian 09, Revision A.1. Gaussian, Inc, Wallingford (2009)
22. Ditchfield, R., Hehre, W.J., Pople, J.A.: Self‐consistent molecular‐orbital
methods. IX. An wxtended Gaussian‐type basis for molecular‐orbital stud-
ies of organic molecules. J. Chem. Phys. 54, 724–728 (1971)
23. Hehre, W.J., Ditchfield, R., Pople, J.A.: Self-consistent molecular orbital
methods. XII. Further extensions of Gaussian-type basis sets for use in
molecular orbital studies of organic molecules. J. Chem. Phys. 56, 2257–
2261 (1972)
24. Francl, M.M., Pietro, W.J., Hehre, W.J., Binkley, J.S., Gordon, M.S.,
DeFrees, D.J., Pople, J.A.: Self‐consistent molecular orbital methods.
XXIII. A polarization‐type basis set for second‐row elements. J. Chem.
Phys. 77, 3654–3665 (1982)
25. Clark, T., Chandrasekhar, J., Spitznagel, G.W., Schleyer, P.V.R.: Efficient
diffuse function-augmented basis-sets for anion calculations. 3. The 3-
21+G Basis Set for 1st-row Elements, Li-F. J. Comput. Chem. 4, 294–
301 (1983)
26. Frisch, M.J., Pople, J.A., Binkley, J.S.: Self‐consistent molecular orbital
methods 25. Supplementary functions for Gaussian basis sets. J. Chem.
Phys. 80, 3265–3269 (1984)
43. Vekey, K.: Internal energy effects in mass spectrometry. J. Mass Spectrom.
31, 445–463 (1996)