Full text of DOI:10.1002/1097-461X(2001)82:1<44::AID-QUA1020>3.0.CO;2-6

Tautomeric Equilibria of 2- and
4-Thiouracil in Gas Phase and in Solvent:
A Density Functional Study
T. MARINO, N. RUSSO, E. SICILIA, M. TOSCANO
Dipartimento di Chimica, Università della Calabria-I-87030 Arcavacata di Rende (CS), Italy
Received 22 May 2000; revised 12 October 2000; accepted 13 November 2000
ABSTRACT: The relative stabilities of the five favored tautomers of 2- and
4-thiouracil in gas phase and in water solution were determined by density functional
theory employing the Becke, Lee, Yang, and Parr (B3LYP) exchange–correlation potential
and the three 6-31G(d,p), 6-311++G(d,p), and triple-zeta valence (TZVP) basis sets.
Zero-point vibrational corrections were also computed. Bulk solvent effects were studied
in the framework of the self-consistent reaction field approach by the polarizable
continuum model. All calculations indicate that the most stable tautomer for both species,
in the gas phase as well as in solution, has the oxo-thione form, in full agreement with the
previous ab initio and experimental studies. The tautomeric stability orders obtained
in the aqueous solution are sensibly different from that in the gas phase. At
B3LYP/6-311++G(d,p) level in the gas phase, the following orders of stability for 2- and
4-thiouracil tautomers were observed, respectively: S2U1 > S2U2 > S2U4 > S2U5 > S2U3
and S4U1 > S4U2 > S4U3 > S4U4 > S4U5. The corresponding trends in the
aqueous phase are S2U1 > S2U3 > S2U2 > S2U5 > S2U4 and S4U1 > S4U2 > S4U3 >
S4U5 > S4U4. On the basis of the computed energy differences we can hypothesize that
only the oxo-thione forms of 2- and 4-thiouracil should exist in the gas phase and in water
c
solution. ꢀ 2001 John Wiley & Sons, Inc. Int J Quantum Chem 82: 44–52, 2001
Key words: thiouracil; tautomeric equilibria; density functional theory; solvent effects
Correspondence to: M. Toscano; e-mail: m.toscano@unical.it.
Contract grant sponsors: MURST (Progetto Nazionale Sistemi
a Grandi Interfasi); Università della Calabria.
International Journal of Quantum Chemistry, Vol. 82, 44–52 (2001)
c
ꢀ 2001 John Wiley & Sons, Inc.

TAUTOMERIC EQUILIBRIA OF 2- AND 4-THIOURACIL
The interest in the gas-phase equilibria of such
molecular systems can be helpful for modeling the
real situation in the condensed phase. Solvent ef-
fects of biomolecules are extremely important in
DNA base-pair interactions because several biolog-
ical processes occur following the displacement of
water of hydration [18]. Although a modulation of
the tautomeric equilibria is possible upon solva-
tion especially in heterocyclic systems, recent cal-
culations, concerning DNA and RNA bases deriv-
atives [19], have pointed out that the enol forms
are very minor in water with respect to the dioxo
ones. The tautomeric equilibria of uracil and its
derivatives have been extensively studied at exper-
imental [20– 23] and quantum mechanical [2, 12,
24 – 30] level, but no comparison work, devoted to
the stability order of the tautomers of both 2- and
4-thiouracils in water, is known until now.
The analysis of the relative population of all
4-thiouracil tautomers in solvent without geom-
etry optimization has been performed theoreti-
cally in a work including the MP4(SDQ/6-311G
(2d,2p)//MP2/6-31G(d,p) computations [28]. The
interaction of the most stable tautomer of 2- and
4-thiouracil with few water molecules has been ex-
amined at B3LYP/6-31+G(d,p) level by Kryachko
et al. [12, 31]. The stability order of 2-thiouracil
tautomers into aqueous solution was never exam-
ined.
Introduction
he protomeric equilibria of heterocyclic mole-
cules are the subject of continuing theoretical
T
and experimental studies because of the biologi-
cal implications that these phenomena cause on
DNA and RNA base-pairs interactions [1 – 5]. In
particular, the study of tautomeric equilibria of the
nucleobases and of some of their analogs constitutes
the starting point to explain the relationships be-
tween the contemporary occurrence of the rare enol
tautomeric forms and the point mutations during
DNA replication due to the ability of nucleic acids to
accommodate noncanonical hydrogen bonds [3, 6].
Thiated derivatives can influence the structure
of DNA and, although the picture of such changes
is not completely known at the molecular level,
some attempts to clarify their unusual functioning
have already been made [7 – 12]. The introduction
of the sulfur atom in the molecule of the uracil
may induce changes in the properties of the base
and modify its interactions, even if it retains the
same distribution of hydrogen donors and accep-
tors as the standard base [7]. In particular, in a recent
work of Kryachko and Nguyen [12], these changes
are substantially attributed to the enhancement of
the polarizability of nucleobases that, in turn, influ-
ences the induction and dispersion contributions to
the stabilization energy of base pairs. No less sig-
nificant are the structural changes mainly due to
the C=S bonds that are longer than the C=O ones.
Furthermore, because the sulfur atom is a weaker
acceptor of H bonds than the oxygen one [13],
the resulting thiobase pairings are influenced both
from structural and energetic points of view. The
structural and energetic variations depend on the
formation of hydrogen bonds that are longer and
weaker when sulfur replaces the oxygen atom in the
uracil.
For these reasons and taking into account their
biological relevance, we have considered it inter-
esting to investigate the tautomeric behavior of
2- and 4-thiouracil in gas phase and mainly in wa-
ter. Our density functional study, performed using
the B3LYP hybrid functional and different basis sets,
should represent the first theoretical study in which
the five lower lying tautomers of both thiobases are
contemporarily investigated at the same level of the-
ory and their behavior compared.
Thiouracils have been identified as minor com-
ponents of transfer-RNA, and they are used as
anticancer and antithyroid drugs, thus playing an
important role in both the biological and pharma-
cological activities [13, 14]. Actually the 2-thiouracil
is used as highly specific melanoma seeker, and it
shows a mechanism of selective incorporation into
growing melanins both in vitro and in vivo [15].
Moreover, the triorganophosphinegold (I) com-
plexes of 2-thiouracil display promising antiarthritic
activity [16]. 4-thiouracil presents cytostatic proper-
ties and is used as the cross-linking agent in RNA
transcriptional regulation [17].
Method
All calculations were carried out using the
GAUSSIAN 98 [32] code. Geometry optimizations,
without imposing any symmetry constraints, were
performed using the Becke [33] and Lee, Yang, and
Parr [34] (B3LYP) functional.
Triple-zeta valence (TZVP) all-electron or-
bital and the corresponding auxiliary [35], 6-
31G(d,p) [36], and 6-311++G(d,p) [37] basis sets
were employed for the computations. Harmonic
vibrational frequencies were computed using
INTERNATIONAL JOURNAL OF QUANTUM CHEMISTRY
45

MARINO ET AL.
the analytical B3LYP/6-311++G(d,p) second
derivatives. The polarizable continuum model
(PCM) approach [38, 39] was used for the full
optimized calculations in water, employing the
same functional and basis sets as in the gas-phase
study. The PCM model [38 – 41] allows one to
work with cavities of realistic molecular shape
with the surface of the cavity subdivided in small
portions (tesserae). The solute–solvent electrostatic
interaction is represented by a set of polarization
point charges, placed in the center of each tessera.
FIGURE 1. Schematic drawing of the 2- and
4-thiouracil tautomers.
Results and Discussion
2-THIOURACIL
From previous theoretical studies [2, 6, 12, 14,
23 – 30], only the five lower-lying of the six possible
tautomers of the two 2- and 4-thiouracil derivatives
were considered in our work: the di-keto (U1), the
keto-enol (U2 and U3), and the di-enolic (U4 and
U5) forms depicted in Figure 1. Results obtained for
the 2- and 4-thiouracil tautomers will be discussed
separately.
Gas-phase stabilities of the 2-thiouracil tau-
tomers are reported in Table I. The oxo-thione S2U1
and the keto-enol S2U3 tautomers are the most and
the less stable ones at all levels of theory, respec-
tively.
In our and in previous studies, it is noteworthy
that the relative stability values of S2U2, S2U4, and
S2U5 forms fall always in a narrow range of energy.
TABLE I
Gas phase total (in a.u.) and relative energies (1E in kcal/mol) at 0 K, for the five tautomers of 2-thiouracil,
obtained at B3LYP level with different basis sets.
S2U1
S2U2
S2U3
S2U4
S2U5
B3LYP/6-31G(d,p)
Total energy
1E
−737.695862
−737.679524
−737.676719
−737.679990
−737.680458
0.0
10.25
12.01
9.96
9.66
B3LYP/6-311++G(d,p)
Total energy
1E
−737.815017
−737.800434
−737.795114
−737.799617
−737.799022
0.0
9.15
12.48
9.66
10.03
B3LYP/TZVP
Total energy
1E
−737.800577
−737.786839
−737.781602
−737.786888
−737.786252
0.0
8.62
11.91
8.58
8.99
SCF/6-31G(d,p)+MBPTa
1E
1E
0.0
8.64
11.93
6.46
—
B3LYP/6-31+G(d,p)b
0.0
0.0
9.34
12.03
9.38
9.30
9.76
9.80
MP2/6-31G(d)c
13.6
1E
9.80
a
From Ref. [2].
From Ref. [12].
From Ref. [30].
b
c
46
VOL. 82, NO. 1

TAUTOMERIC EQUILIBRIA OF 2- AND 4-THIOURACIL
This can be the reason for the slight discrepancies in
the stability order obtained by using different basis
sets as a consequence of their different reliability in
the orbital description. Starting from the B3LYP/6-
31(d,p) values, the introduction of only one diffuse
function (see B3LYP/6-31+(d,p) of Ref. [12]) pro-
duces the “most relevant” changes in the stabil-
ity order. On the contrary, a different contraction
scheme and a further diffuse function introduction
[set B3LYP/6-311++(d,p)] leaves the stability order
unaffected with respect to that of Kryachko and
Nguyen [12]. B3LYP/TZVP data show that S2U2
and S2U4 forms are practically degenerate and, ex-
cept for the relative position of S2U2, yield results
very similar to those of Lamsabhi et al. [30] (see
MP2/6-31G∗ results in Table I). The convergence
of our B3LYP/6-311++(d,p) and B3LYP/6-31+(d,p)
[12] results allows us to assume, with sufficient con-
fidence, the S2U1 > S2U2 > S2U4 > S2U5 > S2U3
order of stability as the most probable one in gas
phase.
data [42]. The only remarkable difference can be ob-
served in the case of N3C4 distance that, in all our
computations, is longer than that obtained at the
Hartree–Fock (HF) level (1.404 Å) [2] and by the ex-
perimental measurement [42]. It is worth noting that
experimental data are referred to the 2-thiouridine
(i.e., the respective nucleoside), and gas-phase the-
oretical calculations cannot take into account the
crystal forces, thus, small differences in the bond
lengths are justified. In any case, considering the
electronic situation of the S2U1 tautomer, the N3C4
bond should have preferentially a single-bond char-
acter. The slight discrepancy can be also ascribed to
the fact that experimental measurement is the result
of an average length assignable to all the tautomers
with very similar stability. Unscaled vibrational fre-
quencies computed at the B3LYP/6-311++G(d,p)
level, available upon request from the authors, are
slightly smaller (less than the 10%) than the experi-
mental ones [14] and confirm indirectly the reliabil-
ity of our geometrical parameters.
The consideration of the solvent effects can be
very important because, sometime, the solvent can
introduce significant changes in the molecular struc-
ture both in terms of nuclear and electronic distri-
butions. In fact a polar solvent, such as water, has
the tendency to displace the tautomeric equilibria
so as to increase the population of the most polar
tautomer. This effect can reverse the stability order
in gas phase.
Dipole moments (µ) of 2-thiouracil tautomers ob-
tained at the various levels of theory are practically
the same. Thus, in the course of our discussion, only
the B3LYP/6-311++G(d,p) values will be explicitly
reported. The results of the calculations performed
in water solution for 2-thiouracil tautomers are col-
lected in Table III.
Although the stability orders in solution are dif-
ferent from those in the gas phase, the data show
that the S2U1 remains the most stable tautomer, at
all levels of theory. The geometry optimization does
not introduce sensible changes with respect to the
gas-phase structures as can be concluded by the val-
ues of geometrical parameters reported in Table II
for the most stable S2U1 tautomer.
The equilibrium geometry of the S2U1 absolute
minimum obtained at B3LYP/6-311++(d,p) is re-
ported in Table II. The computed structural para-
meters are quite in agreement with experimental
TABLE II
Optimized geometry of the most stable
tautomer (S2U1) of 2-thiouracil at
B3LYP/6-311++G(d,p) level in vacuum
and in solvent.a
Parameter
Gas phase
Water
Expb
N1C2
C2N3
N3C4
C4C5
C5C6
C6N1
C2S8
C4O7
1.378
1.369
1.417
1.456
1.345
1.374
1.661
1.214
1.367
1.366
1.407
1.446
1.351
1.371
1.675
1.224
1.368
1.360
1.388
1.435
1.337
1.381
1.677
1.228
N1C2N3
C2N3C4
N3C4C5
C4C5C6
C5C6N1
C6N1C2
N1C2S8
N3C4O7
113.2
128.1
113.4
119.6
121.8
123.8
122.5
120.0
114.1
127.3
114.1
119.2
121.8
123.4
122.7
119.5
116.0
126.6
114.4
119.5
122.6
120.8
123.4
119.2
Aside from the used basis set, the presence of the
solvent stabilizes the S2U3 form having the larger
dipole moment (µ = 5.708 D) but is not able to re-
verse the stability of S2U1 (µ = 4.746 D) and S2U3
forms notwithstanding the difference in their dipole
values. The contributions of cavitation, repulsion,
and dispersion energy are not decisive because they
are very similar for all tautomers. The total of non-
a
Distances are in Å and angles in degrees.
From Ref. [42].
b
INTERNATIONAL JOURNAL OF QUANTUM CHEMISTRY
47

MARINO ET AL.
TABLE III
Relative stabilities (1E in kcal/mol) and free hydration energies (1G in kcal/mol) for the five tautomers
of 2-thiouracil in water, obtained by the PCM model.
Method
1E(S2U1-S2U2)
1E(S2U1-S2U3)
1E(S2U1-S2U4)
1E(S2U1-S2U5)
B3LYP/6-31G(d,p)
B3LYP/6-311++G(d,p)
B3LYP/TZVP
17.40
15.24
17.47
9.13
10.11
9.02
17.98
17.25
17.95
19.45
17.21
18.00
1G(S2U1-S2U2)
1G(S2U1-S2U3)
1G(S2U1-S2U4)
1G(S2U1-S2U5)
B3LYP/6-31G(d,p)
B3LYP/6-311++G(d,p)
B3LYP/TZVP
6.02
5.67
6.57
−4.69
−3.71
−2.90
7.51
7.91
7.19
6.56
6.94
6.80
electrostatic interactions range from 2.05 (S2U1) to
1.70 (S2U2) to 1.89 (S2U3) to 1.52 (S2U4) to 1.54
(S2U5) kcal/mol in the B3LYP/6-311++G(d,p) com-
putations, but it is very similar when the 6-31G(d,p)
or TZVP basis sets are used. The large energy dif-
ference between the S2U1 and S2U3 isomers in gas
phase probably accounts for keeping this stability
order.
and 4-thiouracil. In their study, the authors con-
clude that, although the S4U1 is more stable by
1.67 kcal/mol than S2U1 tautomer in gas phase,
and that the situation remains almost the same
when these tautomers interact with only one wa-
ter molecule, their trihydrated complexes have a
reversed stability order with the S2U1–(H2O)3 sys-
tem, favored by 0.7 kcal/mol over the S4U1–(H2O)3
one. Extracting the same information from our PCM
computations we obtain that the S4U1 tautomer is
always more stable than S2U1 both in vacuo and in
solution, at all levels of theory. The disagreement is
certainly due to the different approaches used in the
treatment of the interaction with the solvent. While
the PCM model does not allow the estimation of the
specific interactions with water, on the other hand, a
hydration shell consisting of three water molecules
cannot be considered as the complete first hydration
shell in the thiouracils.
The 6-31G(d,p) and TZVP basis sets propose the
same trend of relative energy with S2U2, S2U4,
and S2U5 tautomers all present within 2 kcal/mol.
On the contrary, the more extended 6-311++G(d,p)
set stabilizes the S2U2 form with respect to the
other two, which appear almost degenerate. This
latter result seems to be the most reliable, tak-
ing into account both the dipole moment values
of the three systems [4.746 (S2U2), 1.337 (S2U4),
and 2.105 (S2U5) D] and the fact that a net effect
of the polar solvent, such as water, is the destabi-
lization of the conformations with intramolecular
hydrogen bonds [41]. In S2U4 as well as in S2U5
tautomer, two intramolecular H bonds between the
—SH and the —OH groups and the N1 and N3
lone pairs (see Fig. 1) are present, thus, they are
the less hydrated forms with free hydration energy
values of about 6.5–7.0 kcal/mol lower than that of
S2U1 system taken as reference. Only in the case of
S2U3 tautomer the free hydration energy assumes
a more negative sign with respect ot S2U1 (see Ta-
ble III).
4-THIOURACIL
Gas-phase stabilities of the five 4-thiouracil tau-
tomers are reported in Table IV together with pre-
vious theoretical data [2, 12, 28, 30]. A first glance
at the table shows that, except for the most S4U1
and the less S4U5 stable systems, our computations
give slightly different orders. B3LYP/6-31G(d,p)
and B3LYP/TZVP stability trends are quite homo-
geneous between them, although the TZVP basis set
reduces the energy differences among S4U2, S4U3,
and S4U4. The 6-311++G(d,p) set gives the same
results than the 6-31+G(d,p) [12] one but enhances
the energy difference between S4U3 and S4U4
tautomers. The MP2/6-31G(d) data of Lamsabhi
Our data concerning the stability order in so-
lution cannot be compared with other previous
studies because in the work of Kryachko and
Nguyen [12] the interaction with the water mole-
cules regards only the most stable tautomer of 2-
48
VOL. 82, NO. 1

TAUTOMERIC EQUILIBRIA OF 2- AND 4-THIOURACIL
TABLE IV
Gas phase total (in a.u.) and relative energies (1E in kcal/mol) at 0 K, for the five tautomers of 4-thiouracil,
obtained at various levels of theory.
S4U1
S4U2
S4U3
S4U4
S4U5
B3LYP/6-31G(d,p)
Total energy
1E
−737.698662
−737.682390
−737.678865
−737.679389
−737.678510
0.0
10.21
12.42
12.09
12.64
B3LYP/6-311++G(d,p)
Total energy
1E
−737.817260
−737.799995
−737.799424
−737.798581
−737.798171
0.0
10.85
11.31
11.81
11.98
B3LYP/TZVP
Total energy
1E
−737.803622
−737.787195
−737.786380
−737.786504
−737.785573
0.0
10.31
10.82
10.74
11.32
B3LYP/6-31+G(d,p)a
1E
1E
1E
1E
1E
0.0
0.0
0.0
0.0
0.0
0.0
10.30
11.43
11.41
11.5
12.00
12.1
—
MP2/6-31G(d)b
11.2
12.8
MP2/6-31G(d,p)c
16.59
10.26
11.89
12.72
9.88
MP4(SDQ)/6-311G(2d,2p)//MP2/6-31G(d,p)c
10.39 15.27
—
MP2/6-311++G(2d,2p)//MP2/6-31G(d,p)c
9.13
14.19
—
SCF/6-31G(d,p)+MBPTd
1E
10.08
11.80
9.10
—
a
From Ref. [12].
b
c
d
From Ref. [30].
From Ref. [28].
From Ref. [2].
et al. [30] propose a stability order with S4U1 >
S4U2 ≈ S4U4 > S4U5 > S4U3. This result differs
with respect to all B3LYP calculations, essentially
for the relative positions of the latter two tautomers,
and is in disagreement with those coming from the
previous MP2/6-31G(d,p) optimization and from
MPn (n = 2, 4)/6-311G(2d,2p)//MP2/6-31G(d,p)
single-point computations [28], in the energy differ-
ence estimation. As in the case, of 2-thiouracil and
because MP2 computations seem to be more influ-
enced by the basis set effect, we can hypothesize that
the B3LYP/6-311++G(d,p) order of stability is quite
reliable.
in Table V. The situation appears to be analogous
to that of 2-thiouracil. Again, the structural parame-
ters are well reproduced, except the N3C4 distance,
which is considerably longer than the experimen-
tal one [44]. The same arguments, as in the case
of 2-thiouracil, can be advanced for explaining the
disagreement. On the other hand the vibrational
frequency for the stretching relative to the N3C4
bond is very close to experimental value (1357 vs.
1344–1350 cm−1, Ref. [14]), although this distance
is abundantly overestimated. All other frequencies,
available upon request to authors, are in good agree-
ment with the comparison data [12, 14].
The B3LYP/6-311++G(d,p) geometrical parame-
In Table VI are collected the relative stabili-
ties (1E) and the free hydration energies (1G) of
ters of the S4U1 most stable tautomer are reported
INTERNATIONAL JOURNAL OF QUANTUM CHEMISTRY
49

MARINO ET AL.
TABLE V
the 4-thiouracil tautomers in solution. Dipole mo-
ment trends are practically the same in all the three
computations, so only B3LYP/6-311++G(d,p) val-
ues will be given along the discussion.
Optimized geometry of the most stable tautomer
(S4U1) of 4-thiouracil at B3LYP/6-311++G(d,p) level
in vacuum and in solvent.a
The results show that the S4U1 (µ = 4.944 D)
species is the most stable one also in sol-
vent. B3LYP/6-31G(d,p) and B3LYP/TZVP full-
optimized computations indicate the same stability
order for the five tautomers. The only difference in
the B3LYP/6-311++G(d,p) order regards the S4U4
(µ = 1.784 D) and S4U5 (µ = 2.281 D) forms, with
the first favored over the second by 0.82 kcal/mol.
Both these tautomers are characterized by two in-
tramolecular hydrogen bonds and, as in the case of
2-thiouracil, are those less solvated with 1G values
sensibly lower with respect to that of S4U1. Because
the dipole moment of S4U5 is higher than that of
S4U4, the B3LYP/6-311++G(d,p) result seems to be
much more reasonable than our other findings.
The S4U2 (µ = 4.138 D) tautomer keeps the
same stability order as in the gas phase at all lev-
els of theory, while the S4U3 (µ = 5.299 D) sys-
tem gains a position in both B3LYP/6-31G(d,p) and
B3LYP/TZVP computations. The 1G value for S4U3
Parameter
Gas phase
Water
Expb
N1C2
C2N3
N3C4
C4C5
C5C6
C6N1
C2O8
C4S7
1.391
1.387
1.391
1.441
1.351
1.374
1.211
1.659
1.381
1.385
1.382
1.431
1.356
1.365
1.219
1.675
1.396
1.373
1.349
1.445
1.353
1.372
1.214
1.656
N1C2N3
C2N3C4
N3C4C5
C4C5C6
C5C6N1
C6N1C2
N1C2O8
N3C4S7
112.9
127.9
114.1
120.1
121.4
123.3
123.1
120.9
113.8
127.2
114.8
119.7
121.5
123.0
123.4
120.5
114.5
127.7
115.3
119.0
122.2
121.1
123.4
119.2
a
Distances are in Å and angles in degrees.
From Ref. [44].
b
TABLE VI
Relative stabilities (1E in kcal/mol) and free hydration energies (1G in kcal/mol) for the five tautomers of
4-thiouracil in water, obtained by the PCM model.
Method
1E(S4U1-S4U2)
1E(S4U1-S4U3)
1E(S4U1-S4U4)
1E(S4U1-S4U5)
B3LYP/6-31G(d,p)
9.86
10.41
11.38
12.02
15.65
13.60
13.53
18.13
18.92
17.89
18.20
20.28
21.09
17.07
19.25
—
B3LYP/6-311++G(d,p)
B3LYP/TZVP
MP4(SDQ)/6-311G(2d,2p)//
MP2/6-31G(d,p) (SCI-PCM)a
MP4(SDQ)/6-311G(2d,2p)//
MP2/6-31G(d,p) (AM1-SM2)a
MP2/6-311++G(2d,2p)//
MP2/6-31G(d,p) (SCI-PCM)a
MP2/6-311++G(2d,2p)//
MP2/6-31G(d,p) (SCI-PCM)a
13.11
10.76
11.84
13.16
17.05
12.08
18.58
17.44
15.74
—
—
—
1G(S4U1-S4U2)
1G(S4U1-S4U3)
1G(S4U1-S4U4)
1G(S4U1-S4U5)
B3LYP/6-31G(d,p)
−0.49
−0.12
−1.96
2.72
0.65
−0.19
2.72
−2.11
6.99
6.24
5.20
5.86
6.36
6.22
6.08
—
B3LYP/6-311++G(d,p)
B3LYP/TZVP
MP4(SDQ)/6-311G(2d,2p)//
MP2/6-31G(d,p) (AM1-SM2)a
MP4(SDQ)/6-311G(2d,2p)//
MP2/6-31G(d,p) (SCI-PCM)a
1.63
2.86
7.56
—
a
From Ref. [28].
50
VOL. 82, NO. 1

TAUTOMERIC EQUILIBRIA OF 2- AND 4-THIOURACIL
is the most negative accordingly to the dipole mo-
ment trend.
2. The S4U1 species is more stable than S2U1
in both phases, and all other tautomers of 4-
thiouracil appear to be better hydrated than
those of 2-thiouracil.
3. The order of stability of tautomers depends
on the basis set, especially in the presence of
degenerate systems. The convergence of the
data is obtained in correspondence with the
6-311++G(d,p) basis set from which we think
to derive out best proposal for the stability
trends.
4. The solvent effects modify partially the stabil-
ity order of tautomers with respect to that in
gas phase in the sense that only the enol forms
are interested in the phenomenon. The energy
differences in water of the U2, U3, U4, and U5
tautomers of both 2- and 4-thiouracils with re-
spect to the U1 one become greater than those
in gas phase, underlying a major difficulty to
the tautomerism in this medium. The results
do not depend on the very small variation
of the geometrical structures but are substan-
tially in line with dipole moment trends and
with the pre-existent energetic situation in the
gas phase.
The geometry optimization leaves the structural
parameters of tautomers practically unchanged
with respect to those obtained in vacuum. Fur-
thermore the contributions due to nonelectrostatic
interactions are much smaller than the 1G values
and very similar for all tautomers [2.12 (S4U1), 1.80
(S4U2), 1.72 (S4U3), 1.54 (S4U4), and 1.52 (S4U5)
kcal/mol]; thus, they cannot decide the reversal of
the stability trend. For these reasons, the stability
order in water of the first three tautomers should be
essentially conditioned by the energy differences in
gas phase.
The B3LYP/6-311++G(d,p) geometrical parame-
ters of the most stable S4U1 tautomer in solution are
reported in Table V. Differences of less than 0.02 Å
for bond lengths and 1.0◦ for valence angles are
found in going from gas to condensed phase.
B3LYP results are in agreement with all the previ-
ous theoretical data reported in Table VI. However,
it should be emphasized that the MPn (n = 2, 4)
values concerning the S4U3 system are referred to
a tautomer in which the N3C4O7H torsion (see
Fig. 1) is 180◦ [28]. This means that the stability of
S4U3 species, in the MPn computations, is estimated
without the hydrogen bond contribution. Because
normally the intramolecular hydrogen bonds play
a significant role in the stabilization of a molecule, it
is possible that taking into account this interaction,
the order of stability between S4U3 and S4U2 can be
influenced.
5. The geometrical parameters and the harmonic
vibrational frequencies are well reproduced.
ACKNOWLEDGMENTS
This work was supported by grants from the
MURST (Progetto Nazionale Sistemi a Grandi Inter-
fasi) and Università della Calabria.
Conclusions
References
In this work we have studied the tautomeric
equilibria of the 2- and 4-thiouracil derivatives both
in gas phase and in solution employing the hybrid
B3LYP functional and different basis sets. On the ba-
sis of the obtained results the following conclusions
can be drawn:
1. Hall, R. J.; Burton, N. A.; Hillier, I. H.; Young, P. E. Chem
Phys Lett 1994, 220, 129.
2. Les, A.; Adamowicz, L. J Am Chem Soc 1990, 112, 1504.
3. Beak, P. Acc Chem Res 1977, 10, 186, and references therein.
4. Kwiatkowski, J. S.; Pullmann, B. Adv Heterocycl Chem
1975, 18, 199.
5. Liguori, A.; Marino, T.; Napoli, A.; Sindona, G.; Turbante, D.
J Mass Spectr Rapid Commun Mass Spectrom 1995, 3, 212.
1. All computations indicate the di-keto (U1) tau-
tomer as the most stable and perhaps the
unique form existing in the gas phase and
in water for both 2- and 4-thiouracil sys-
tems. This observation suggests that muta-
tions should not depend on the presence of un-
usual enol forms but on other circumstances
and in particular on the different stability
of the hydrogen bonds and the structural
changes introduced by the sulfur atom.
6. Scanlan, M. J.; Hillier, I. H. Chem Phys Lett 1983, 98, 545.
7. Sponer, J.; Leszczynski, J.; Hobza, P. J Phys Chem A 1997,
101, 9489.
8. Chojnacki, H.; Sokalski, W. A. J Theor Biol 1975, 54, 167.
9. Aida, M.; Nagata, C.; Ohmine, I.; Morokuma, K. J Theor Biol
1982, 99, 599.
10. Lipinski, J. J Mol Struct (Theochem) 1989, 201, 87.
11. Geller, M.; Jaworski, A.; Pohorille, A. Int J Quantum Chem
1979, 15, 369.
INTERNATIONAL JOURNAL OF QUANTUM CHEMISTRY
51

MARINO ET AL.
12. Kryachko, E. S.; Nguyen, M. T. Adv Quantum Chem 2000,
private communications.
30. Lamsabhi, M.; Alcami, M.; Mo, O.; Bouab, W.; Essefar, M.;
Abboud, J. L. M.; Yanez, M. J Phys Chem A 2000, 104, 5122.
13. Rostkowska, H.; Szczepaniak, K.; Nowak, M. J.; Leszczyn-
ski, J.; Kubulat, K.; Person, W. B. J Am Chem Soc 1990, 112,
2147, and references therein.
31. Kryachko, E. S.; Nguyen, M. T.; Zeegers-Huyskens, T. J Phys
Chem 2000, private communications.
32. Frisch, M. J.; Trucks, G. W.; Schlegel, H. B.; Scuseria, G. E.;
Robb, M. A.; Cheeseman, J. R.; Zakrzewski, V. G.; Mont-
gomery, J. A.; Stratmann, R. E., Jr.; Burant, J. C.; Dapprich, S.;
Millam, J. M.; Daniels, A. D.; Kudin, K. N.; Strain, M. C.;
Farkas, O.; Tomasi, J.; Barone, V.; Cossi, M.; Cammi, R.;
Mennucci, B.; Pomelli, C.; Adamo, C.; Clifford, S.; Ochter-
ski, J.; Petersson, G. A.; Ayala, P. Y.; Cui, Q.; Morokuma,
K.; Malick, D. K.; Rabuck, A. D.; Raghavachari, K.; Fores-
man, J. B.; Cioslowski, J.; Ortiz, J. V.; Baboul, A. G.; Stefanov,
B. B.; Liu, G.; Liashenko, A.; Piskorz, P.; Komaromi, I.; Gom-
perts, R.; Martin, R. L.; Fox, D. J.; Keith, T.; Al-Laham,
M. A.; Peng, C. Y.; Nanayakkara, A.; Gonzalez, C.; Challa-
combe, M.; Gill, P. M. W.; Johnson, B.; Chen, W.; Wong, M.
W.; Andres, J. L.; Gonzalez, C.; Head-Gordon, M.; Replogle,
E. S.; Pople, J. A. Gaussian: Pittsburgh, 1998.
14. Saenger, W. In Principles of Nucleic Acid Structure;
Springer: New York, 1984; Chapter 7.
15. Napolitano, A.; Palumbo, A.; d’Ischia, M.; Prota, G. J Med
Chem 1996, 39, 5192.
16. Stewart, G. M.; Tiekink, E. R. T.; Buntine, M. A. J Phys
Chem A 1997, 101, 5368.
17. Wang, Z.; Rana, T. M. Biochemistry 1996, 35, 6491.
18. van Mourik, T.; Price, S. L.; Clary, D. C. J Phys Chem A 1999,
103, 1611.
19. Orozco, M.; Hernandez, B.; Luque, F. J. J Phys Chem B 1998,
102, 5228.
20. Shugar, D.; Szczepaniak, K. Int J Quantum Chem 1981, 20,
573.
33. Becke, A. D. J Chem Phys 1993, 98, 5648.
21. Ruterjans, H.; Kaun, E.; Hull, W. E.; Limbach, H. H. Nucleic
Acids Res 1982, 10, 7027.
34. Lee, C.; Yang, W.; Parr, R. G. Phys Rev B 1988, 37, 785.
35. Godbout, N.; Salahub, D. R.; Andzelm, J.; Wimmer, E. Can J
Chem 1992, 70, 560.
22. Brown, R. D.; Godfrey, P. D.; McNaughton, D.; Pierlot, A. P.
J Am Chem Soc 1988, 110, 2329.
23. Rostkowska, H.; Barski, A.; Szczepaniak, K.; Szczejniak, M.;
Person, W. B. J Mol Struct 1988, 176, 137, and references
therein.
36. Petersson, G. A.; Bennet, A.; Tensfeldt, T. G.; Al-Laham,
M. A.; Shirley, W. A.; Mantzaris, J. J Chem Phys 1988, 89,
2193, and references therein.
24. Scanlan, M. J.; Hillier, I. H. J Am Chem Soc 1984, 106,
3737.
37. Clark, T.; Chandrasekhar, J.; Spitznagel, G. W.; Schleyer,
P. v. R. J Comput Chem 1983, 4, 294, and references therein.
25. Les, A.; Adamowicz, L. J Chem Phys 1989, 93, 7078.
38. Barone, V.; Cossi, M.; Tomasi, J. J Comput Chem 1998, 19,
404.
26. Mastryukov, V. S.; Fan, K.; Boggs, I. E. J Mol Struct 1995, 346,
173.
39. Barone, V.; Cossi, M. J Phys Chem A 1998, 102, 1995.
27. Marino, T.; Russo, N.; Toscano, M. Int J Quantum Chem
1997, 62, 489.
40. Miertus, S.; Tomasi, J. Chem Phys 1982, 65, 239, and refer-
ences therein.
28. Rubin, Y. V.; Morozov, Y.; Venkateswarlu, D.; Leszczynski, J.
J Phys Chem A 1998, 102, 2194.
41. Tomasi, J.; Persico, M. Chem Rev 1994, 94, 2027.
42. Hawkinson, S. W. Acta Crystallogr B 1977, 33, 80.
43. Orozco, M.; Alhambra, C.; Barril, X. J Mol Model 1996, 2, 1.
44. Lesyng, B.; Saenger, W. Naturforsch 1981, 36C, 956.
29. Adamo, C.; Barone, V. In Recent Advances in Density Func-
tional Methods, Part II; Chong, D. P., Ed.; World Scientific:
Singapore, 1997; pp. 115–164.
52
VOL. 82, NO. 1