Synthesis and conformational studies on 3-o-tolylhydantoins by NMR and molecular modeling: Dipole-π attractions in peptides and proteins

Article abstract of DOI:10.1246/bcsj.74.1917

5-Pentafluorobenzyl-3-o-tolylhydantoin (2) was synthesized for reducing this π-electron density, and NMR studies have shown that we have succeeded in changing the conformation of 2 into an extended system instead of the folded conformation for the non-fluorinated compound, 5-benzyl-3-o-tolylhydantoin (1). Molecular modeling has also confirmed the extended structure for 2. An approach has been found for possibly modulating the physiological activity of peptides containing aromatic amino acids.

Full text of DOI:10.1246/bcsj.74.1917

c. Jpn.
© 2001 The Chemical Society of Japan
Bull. Chem. Soc. Jpn., 74, 1917–1925 (2001) 1917
e Chemical
ciety of Ja-
n
e Chemical
ciety of Ja-
n
Synthesis and Conformational Studies on 3-o-Tolylhydantoins by NMR
and Molecular Modeling: Dipole-π Attractions in Peptides and Proteins
Conformational Studies on 3-o-Tolylhydantoins
S. H. Park et al.
Sang Hyun Park* and Ajay K. Bose†
01
17
25
SJA8
09-2673
089
Metabolic and Biomolecular Engineering National Research Laboratory, Department of Chemical Engineering and
BioProcess Engineering Research Center, Korea Advanced Institute of Science and Technology,
373-1 Kusong-dong, Yusong-gu, Taejon 305-701, Korea,
†Department of Chemistry and Chemical Biology, Stevens Institute of Technology, Hoboken, New Jersey 07030, USA
(Received March 14, 2001)
01
01
.1
5-Pentafluorobenzyl-3-o-tolylhydantoin (2) was synthesized for reducing this π-electron density, and NMR studies
have shown that we have succeeded in changing the conformation of 2 into an extended system instead of the folded con-
formation for the non-fluorinated compound, 5-benzyl-3-o-tolylhydantoin (1). Molecular modeling has also confirmed
the extended structure for 2. An approach has been found for possibly modulating the physiological activity of peptides
containing aromatic amino acids.
When the conformation of molecules is studied, one usually
considers steric repulsion resulting in a less crowded confor-
mation. For example, in the case of butane, the anti-conformer
is favored over the eclipsed conformer due to steric and tor-
sional strains. But when studying the conformation of small
peptides and other polar compounds derived from aromatic
amino acids, one should consider non-bonded attraction,
which can result in a more crowded conformation. The inter-
action between a dipole and the dipole induced by it in a π-
electron cloud constitutes a force of attraction.1 Such attrac-
tion at the intramolecular level could influence the conforma-
tion of peptides and proteins. Recently, several reports de-
scribing cation-π attractions in peptides and proteins have ap-
peared in the literature.2 This type of non-bonded attraction
could be an important factor governing the biological charac-
teristics of these molecules. The conformation of these mole-
cules and indications of the non-bonded attraction between the
aromatic ring and polar group in another part of the molecule
were therefore studied here with the help of nuclear magnetic
resonance (NMR) and molecular modeling.
Over the past several decades, NMR spectroscopy has be-
come firmly established as the most powerful method for
structure analysis in solution in the field of biochemistry as
well as organic chemistry. The parameters normally obtained
from a study of the NMR spectra are the chemical shift,3 nu-
clear spin-spin coupling4 via electrons and nuclear relaxation
time5 (temperature dependent).
Monte Carlo (MC) and molecular dynamics are the primary
methods used for the free-energy simulation of molecular sys-
tems. The application of molecular dynamics to molecules
that have multiple conformations separated by energy barriers
of more than 3 kcal/mol is problematic because of slow rates
of convergence.6 Monte Carlo sampling does not efficiently
sample the bottom of the potential energy wells, thereby caus-
ing slow rates of convergence. Guarnieri and Still introduced a
hybrid simulation method termed MC-SD,6 which mixes Mon-
te Carlo (MC) and stochastic dynamics (SD) sampling. The
new method generates a canonical ensemble by altering MC
and SD steps and combines the local exploration strengths of
dynamics with the barrier-crossing ability of large-step Monte
Carlo.6 MC-SD simulations converge faster than either MC or
SD alone and generate ensembles which are equivalent to
those created by MC or SD alone.6
Results and Discussion
In order to obtain detailed information about the dipole-π at-
traction in aromatic amino acid derivatives, ( )-5-substituted
3-o-tolylhydantoins 1–3 and 3-o-tolylhydantoin (4) were pre-
pared in our laboratory from aromatic and aliphatic amino ac-
ids because of their easy preparation and conformation deter-
mination from NMR spectra. Conformational studies on the
hydantoins were conducted by both 1H and 13C NMR (nuclear
magnetic resonance) and molecular modeling. Conformation-
al information was derived from several aspects of NMR stud-
ies of some hydantoins. The aspects studied were proton
chemical shifts of the o-methyl group, the coupling constants
between the benzylic protons and C-5 proton from A2X pat-
tern, 13C NMR, and the temperature dependence of the o-meth-
yl group in 3-o-tolylhydantoins.
Proton Chemical Shifts of o-Methyl and o-Proton in 3-o-
Tolylhydantoins. In view of the well-known restricted rota-
tion of amide derivatives, it appeared to be of interest to study
3-o-tolylhydantoins. The structures of rotamers A and B re-
sulting from restricted rotation about the N-aryl bond are
shown in Fig. 1. The chemical shifts of the o-methyl group in
3-o-tolylhydantoins derived from aromatic and aliphatic amino
acids are listed in Table 1. The methyl resonance of the o-tolyl
group gives two peaks for compounds 1, 2, and 3 due to the re-
stricted rotation about the N-aryl single bond.
The separation of the two peaks in compound 1 is larger

1918 Bull. Chem. Soc. Jpn., 74, No. 10 (2001)
Conformational Studies on 3-o-Tolylhydantoins
Table 1. 1H NMR Chemical Shifts of o-Methyl and o-Proton in 3-o-Tolylhydantoins
Chemical shift/ppm
Population ratio
Compd
R
Solvent
a)
a)
b)
b)
CH3d
CH3
Hd
Hu
CH3d(Hu):CH3u(Hd)
1
C6H5CH2
CDCl3
DMSO-d6
DMSO-d6
CDCl3
2.20
2.06
2.10
2.16
1.76
1.34
1.93
2.13
7.12 6.63
7.10 6.23
7.16 7.10
7.10 7.06
1:1
4:6
4:6
2
3
4
C6F5CH2
isobutyl
H
CDCl3
2.23c)
a) Subscript d and u stand for the downfield and upfield o-methyl resonances, respectively.
b) Subscript d and u stand for the downfield and upfield o-proton resonances, respectively.
c) The value indicates no separation for the proton resonance.
Fig. 1. The folded conformation of 5-benzyl-3-o-tolylhydan-
toin (1).
Fig. 3. The extended conformation of 5-pentafluorobenzyl-
3-o-tolylhydantoin (2).
yl (ICH3u) must be equal to the ratio of the upfield o-proton (IHu)
and the downfield o-proton (IHd), (ICH3d/ICH3u = IHu/IHd). When
the C-5 position in the 3-o-tolylhydantoin ring in 1 is substitut-
ed by the pentafluorobenzyl group, the separation of the two
peaks of the o-methyl in compound 2 is smaller than that in 1.
This is an indication of a conformation extended away from
the hydantoin ring in compound 2 (Fig. 3). This change is due
to a reduced π-dipole interaction in 2. The lower NMR peak of
the o-methyl group shows up around the normal position of an
aryl o-methyl. The other peak of the methyl is at 0.17 ppm up-
field from the normal peak. One can easily understand that the
upper peak is not diamagnetically shielded by the C-5 pen-
tafluorobenzyl group. The methyl group in o-tolyl is just
chemically nonequivalent when the N-aryl bond rotates by 180
degrees. Therefore, there are two kinds of o-methyl peaks in
compound 2. The conformer population of forms A and B in
Fig. 3 is also easily determined by the integration of the down-
field and upfield o-methyl and o-proton in C-3 o-tolyl group as
shown before. 1H NMR spectrum of 2 is given in Fig. 4.
When the C-5 position in the 3-o-tolylhydantoin ring in 1 is
substituted by an isobutyl group, two slightly separated methyl
peaks are observed. The separation of the two peaks of the o-
methyl in compound 3 was much smaller than that in 1. This is
an indication of a conformation extended away from the hy-
dantoin ring in compound 3 (Fig. 5).
Fig. 2. 1H NMR spectrum of 5-benzyl-3-o-tolylhydantoin
(1).
than that in 2 and 3. This is an indication of a folded confor-
mation in compound 1. The lower NMR peak of the o-methyl
group shows up around the normal position of an aryl o-meth-
yl. The other peak of the methyl group is 0.44 ppm upfield
from the normal peak. Obviously, the upper peak is diamag-
netically shielded by the C-5 benzyl group (Fig. 1). On the
other hand, the methyl group in the o-tolyl is not shielded
when the N-aryl bond rotates through 180 degrees (Fig. 1).
Therefore, there are two kinds of o-methyl peaks in compound
1. The conformer population of forms A and B in Fig. 1 is eas-
ily determined by integration of the downfield and upfield o-
methyl and o-proton in C-3 o-tolyl group. The 1H NMR spec-
trum of 1 is given in Fig. 2. It is obvious that the integration
ratio of the downfield o-methyl (ICH3d) and the upfield o-meth-
There is no π-dipole interaction in compound 3. The lower
NMR peak of the o-methyl group shows up around the normal
position of an aryl o-methyl. The other peak of the methyl is at
0.03 ppm upfield from the normal peak. It would appear that
the upper and lower peaks are just chemically nonequivalent

S. H. Park et al.
Bull. Chem. Soc. Jpn., 74, No. 10 (2001) 1919
Fig. 7. The conformation of 3-o-tolylhydantoin (3).
Fig. 4. 1H NMR spectrum of 5-pentafluorobenzyl-3-o-tolyl-
hydantoin (2).
Fig. 8. 1H NMR spectrum of 3-o-tolylhydatoin (4).
Fig. 5. The conformation of 5-isobutyl-3-o-tolylhydantoin
(3).
counts for a reduced non-bonded attraction and no non-bonded
attraction, respectively. But in 1, there is clear evidence for a
shielding of the methyl group due to the aromatic ring at C-5,
resulting from the non-bonded attraction. The downfield me-
thyl signal can be ascribed to the rotamer A in Fig. 1, while the
upfield signal is due to rotamer B in Fig. 1.
The Coupling Constants of the Benzylic Protons and the
C-5 Proton from A2X Patterns in 3-o-Tolylhydantoins.
The two benzylic protons (Ha, Hb) couple with the C-5 proton
(Hx) in 1 and 2 to a coupling pattern, depending upon the sol-
vent. In general, the ABX pattern is obtained in a CDCl3 solu-
tion while a DMSO-d6 solution gives an A2X pattern. The cou-
pling constants of 1 and 2 are listed in Table 2. The rotational
isomerism of hydantoin derivatives in solution is usually repre-
sented in terms of the equilibrium mixture of the three stag-
gered rotamers7,8 illustrated in Fig. 9. Since the internal rota-
tion is sufficiently rapid, the observed coupling constants are
weighted averages over those corresponding to the individual
rotamers (Eq. 1):
Fig. 6. 1H NMR spectrum of 5-isobutyl-3-o-tolylhydantoin
(3).
when the N-aryl bond rotates by 180 degrees (Fig. 5). There-
fore, there are two kinds of o-methyl peaks in compound 3.
The conformer population of forms A and B in Fig. 5 is also
easily determined by the integration of the downfield and up-
field o-methyl and o-proton in C-3 o-tolyl group, as shown be-
fore. The 1H NMR spectrum of 3 is given in Fig. 6. When the
C-5 position in the 3-o-tolylhydantoin ring in 1 is not substitut-
ed, no two separated methyl peaks are observed for compound
4.
Table 2. The Coupling Constants between Benzylic Protons
and C-5 Proton from A2X Pattern in 3-o-Tolylhydantoins
As expected from a consideration of symmetry, the 1H NMR
spectrum of 4 shows a single peak for the o-methyl in tolyl
group (Figs. 7 and 8).
In summary, 4 shows a single peak for the methyl protons of
o-tolyl group due to symmetry of the hydantoin ring. In 2 and
3, two slightly separated methyl signals are observed. This ac-
Population ratio
Compd
R
Solvent
Javg/Hz
Extended:Folded
1
2
C6H5
C6F5
DMSO-d6
DMSO-d6
4.5
7.0
3:7
8:2

1920 Bull. Chem. Soc. Jpn., 74, No. 10 (2001)
Conformational Studies on 3-o-Tolylhydantoins
the C–C fragment12 (Eq. 4):
Jav = 18.0 − 0.80ΣEi
= 18.0 − 0.80 × 14.56 = 6.35 (Hz).
(Eq. 4)
Unfortunately the elctronegativity of the hydantoin ring has
not yet been reported. However, in a simple comparison of the
vicinal coupling constants of the C-2 methyl protons and the
C-5 methyl protons between DL-alanine (1) and 5-methylhy-
dantoin (2), there is not much difference (Fig. 10).
The calculated average coupling constant from Eq. 4 for DL-
alanine is 6.7 Hz and for phenylalanine is 6.25 Hz. The ob-
served and estimated coupling constants of 3-ethyl-5-pheneth-
ylhydatoin are 6.9 Hz and 6.6 Hz, respectively. Therefore, it
can be assumed that the values of Jg and Jt for the hydantoin
derivative are very close to that for an α-amino acid (Jg =
2.60, Jt = 13.56). Since it is not possible to distinguish which
is the Ha and Hb in the benzyl protons without a labeling study,
the population of rotamer ꢀ and ꢁ in Fig. 9 is not distinguish-
able. The total population of rotamer ꢀ and ꢁ is independent of
the assignment of Ha and Hb. The average vicinal coupling
Fig. 9. Conformers of 3-o-tolylhydantoins.
Ⅰ
Ⅱ
Ⅲ
Jax = X J ax + X J ax + X J
,
Ⅰ
Ⅱ
Ⅲ
ax
Ⅰ
Ⅱ
Ⅲ
Ⅲ
Jbx = X J bx + X J bx + X J
,
bx
Ⅰ
Ⅱ
Ⅲ
Ⅰ
Ⅱ
Jab = X J ab + X J ab + X J
,
ab
Ⅰ
Ⅱ
Ⅲ
X + X + X = 1.
(Eq. 1)
Ⅰ
Ⅱ
Ⅲ
1
constant (Jav) can be measured directly from the H NMR
spectra. The average coupling constant was derived from Eq.
3:
where Jax, Jbx, and Jab are accurately calculated experimental
vicinal and geminal coupling constants, and J ax means the
coupling constant between nuclei a and x in the rotamer ꢀ. X ,
Ⅰ
Ⅰ
X , and X are also defined as the rotamer populations for
Ⅱ
Ⅲ
Jav = (Jax + Jbx)/2 = {2 − (X + X )Jg + (X + X )Jt}/2
Ⅰ
Ⅱ
Ⅰ
Ⅱ
forms ꢀ, ꢁ, and ꢂ, respectively. The dependence on coupling
constants on the dihedral angle in accordance with the Karplus
equation9 (Eq. 2) is well documented:
= {(1 + X )Jg + (1 − X )Jt}/2.
Ⅲ
Ⅲ
The rotamer populations of 1 and 2 are listed in Table 2.
13C NMR Spectra of 3-o-Tolylhydantoins. The 13C
NMR spectra are considerably more sensitive to the stere-
ochemistry of the molecule. The naturally abundant 13C NMR
spectra of o-tolylhydantoins were measured in CDCl3 and
DMSO-d6. The chemical shifts are listed in Table 3. The
Jav = K1 cos2 ϕ − 0.28 for 0° < ϕ < 90°,
Jav = K2 cos2 ϕ − 0.28 for 90° < ϕ < 180°. (Eq. 2)
Ⅰ, Ⅱ, Ⅲ
J
ax in Eq. 1 may be replaced by Jg and Jt for vicinal pro-
ton gauch and trans to one another. Then Eq. 1 becomes Eq. 3:
Jax = (1 − X ) Jg + X Jt,
Ⅱ
Ⅱ
Jbx = (1 − X ) Jg + X Jt.
(Eq. 3)
Ⅰ
Ⅰ
If Jg and Jt were known, the populations of the three rotam-
ers could be evaluated from the Eq. 3. According to the values
of K1 and K2 of the Karplus equation by Pachler.10 Jg and Jt
calculated to be 2.60 and 13.56 for α-amino acid, respectively.
The average coupling constant, Jav, can also be calculated from
the electronegativity values11 for the atom directly attached to
Fig. 10. DL-Alanine (1) and 5-methylhydantoin (2).
Table 3. 13C NMR Chemical Shifts of 3-o-Tolylhydantoins
Chemical shifts/ppm
Compd
R
CH3
C-5
CO(2)
CO(4)
156.6
156.0
156.7
157.8
1
2
3
4
C6H5CH2
C6F5CH2
isobutyl
H
17.2, 17.9
17.1, 17.9
17.8, 18.0
17.7
58.5, 58.7
56.6, 56.8
41.1, 41.7
46.7
172.1
172.2
173.3
170.5

S. H. Park et al.
Bull. Chem. Soc. Jpn., 74, No. 10 (2001) 1921
chemical shifts of aromatic carbon from the spectra are diffi-
cult to assign due to the complex pattern, but the signals of the
o-methyl carbon, benzylic carbon, and hydantoin ring carbons
are relatively easy to identify. Compounds 1, 2, and 3 lose
symmetry. The 13C NMR spectra of o-methyl carbon show
two distinguishable peaks due in part to the asymmetric center
at C-5. Compound 4 has symmetry, and the 13C NMR spec-
trum of o-methyl carbon displays a single peak. This 13C
NMR result strongly supports the existence of conformational
isomers. Since compounds 1, 2, and 3 are asymmetric at the
C-5, the 13C NMR spectra for C-5 also give two peaks. The
carbonyl resonance is split and appears downfield with C-2 at a
relatively higher field than C-4. This agrees with the carbonyls
in 3-hydroxyethyl-5-dimethylhydantoin13 and in substituted
uracils.14 The down-field carbonyl peaks are moved downfield
2–3 ppm by substitution at the 5-position. Similar effects in
alkyl substitution on acetone are also reported by Jackman and
Kelly.15 The difference in the chemical shifts between two res-
onances of the o-methyl increases considerably when the sub-
stituent at C-5 is a benzyl group, indicating a substantial con-
tribution from the folded conformation of the hydantoin. The
split for o-methyl peaks in compound 1 is largest (0.73 ppm) in
Table 3, and supports the folded conformation between the 5-
benzyl and the hydantoin ring.
perature increases in DMSO-d6. The downfield o-methyl peak
moves further upfield, while the upfield methyl peak moves
further downfield (Table 4).
Molecular Modeling Studies on 3-o-Tolylhydantoins.
We were interested in predicting the relative populations of
folded and extended conformations of the hydantoin system 1
and 2 described earlier. This can be done efficiently using
modern computational techniques. A technique was needed
which would generate a canonical ensemble for the benzyl and
pentafluorobenzyl hydantoin system in a high dielectric sol-
vent, since the NMR experiment was performed in a DMSO-d6
solution. To this end, a hybrid Monte Carlo (MC)–stochastic
dynamics (SD) calculation6 was performed using a continuum
water solvent mode16 implemented in Macro Model17 version
5.0. The simulation method MC–SD was chosen because it
combines the local exploration strengths of dynamics and the
barrier-crossing ability of large-step Monte Carlo.6 The simu-
lation was done in water (dielectric constant = 78.7), as being
the most representative dielectric available which is similar to
DMSO (dielectric constant = 45). After a period of 100 ps of
stochastic dynamics equilibrium, long simulations (5000 ps)
were carried out at 300 K. The simulations were to be con-
verged by the obtained constant temperature and average en-
thalpies. All three ratable bonds were permitted to vary during
the simulations. During the simulations, 1000 snapshots were
taken at equal intervals. From this sampling, the populations
of the folded versus extended forms were obtained, as well as
the populations of the o-tolyl groups in the two conformations-
above and below the plane of the hydantoin ring. The torsions
T1 and T2 defined by H23–C1–C16–H32 (Fig. 11) and H6–
C1–C16–H31 (Fig. 11) were used to quantitate to the groups
1
Temperature Dependence of H NMR Spectra of 3-o-
Tolylhydantoins. The chemical shifts and the coupling con-
stants of hydantoin derivatives change with temperature.1
Temperature studies on 5-benzyl-3-o-tolylhydantoin (1) in
DMSO-d6 solution show that the o-methyl and o-proton coa-
lesce at 150 °C (Table 4). The o-tolyl group can rotate freely
above this temperature. The downfield o-methyl peak moves
further upfield, while the upfield methyl peak moves further
downfield. As the temperature increases, the 5-benzyl group
also starts to rotate and the diamagnetically shielded peak
moves rapidly downfield. The coupling constant of the benzyl-
ic protons with the adjacent C-5 proton and the chemical shift
of o-methyl are changed when the temperature decreases in ac-
etone-d6.1 The downfield o-methyl peak is independent of the
temperature. In contrast, the upfield methyl peak moves fur-
ther upfield, and at the same time the coupling constant (Javg
)
decreases, indicating the stability of the folded conformation at
low temperature. Temperature studies on 5-pentafluorobenzyl-
3-o-tolylhydantoin (2) in DMSO-d6 solution show that the o-
methyl and o-proton coalesce at 120 °C (Table 4). The o-tolyl
group can rotate freely above this temperature. The coupling
constant of the benzylic protons with the adjacent C-5 proton
and the chemical shift of o-methyl are changed when the tem-
Table 4. Coalescence Temperatures for 3-o-Tolylhydantoins
Compd
R
Coalescence temp./°C
Fig. 11. The torsions T1 and T2 defined by H23–C1–C16–
H32 and H6–C1–C16–H31.
1
2
C6H5CH2
C6H5CH2
150
120

1922 Bull. Chem. Soc. Jpn., 74, No. 10 (2001)
Conformational Studies on 3-o-Tolylhydantoins
Table 5. Torsion Angle Constraints to Determine Populations of States ꢀ, ꢁ, and ꢂ
Torsion (T)
Atoms defining T
State ꢀ
State ꢁ
State ꢂ
T1
T2
H23–C1–C16–H32
H6–C1–C16–H31
−60 60
180 60
180 60
60 60
60 60
−60 60
Table 6. Populations of Each Conformational State Based on the 1000 Snapshot Sampling for
the Folded versus Extended Form
Compound 1
Compound 1
Compound 2
Compound 2
State
# of conformers
% of conformers
# of conformers
% of conformers
ꢀ
ꢁ
ꢂ
293
105
601
29.3
10.5
60.1
557
409
32
55.7
40.9
3.2
Table 7. Populations of Each Conformational State Based on the 1000 Snapshot Sampling of the
o-Tolyl Group (Torsion: C7–C4–C9–C14)
B (90 45°)
B (90 45°)
A (−90 45°)
A (−90 45°)
Compound
# of conformers
% of conformers
# of conformers
% of conformers
1
2
373
423
37.3
42.3
616
568
61.6
56.8
previously defined as conformer ꢀ, ꢁ and ꢂ described earlier
(Fig. 11).
Conclusions
The torsion defined by C7–C4–C9–C14 (Fig. 11) was used
to determine the rotamer states of the o-tolyl groups. The re-
sults are summarized in the Tables 5–7 below.
The folded form is defined as the conformers where the ben-
zyl group is over the hydantoin ring, state ꢂ. The extended
form is defined by states ꢀ and ꢁ (Figs. 12 and 13). For com-
pound 1, the population ratio of the extended form to the fold-
ed form was calculated to be 4:6. For compound 2, the popu-
lation ratio of extended form to folded form was found to be
97:3.
It is necessary to take into consideration the dipole-π attrac-
tion as well as the steric repulsion in studying the conforma-
tion of peptides and enzyme-substrate binding. Basic hydroly-
sis of 5-benzyl-3-o-tolylhydantoin (1) gives phenylalanine.18
Similarly, pentafluorophenylalanine could be obtained by the
hydrolysis of 5-pentafluorobenzyl-3-o-tolylhydantoin (2).
Phenylalanine is often a constituent of peptides and other polar
compounds derived from aromatic amino acids. The biologi-
cal characteristics of these molecules could be changed by sub-
stituting pentafluorophenylalanine or less fluorinated phenyla-
lanine for phenylalanine. Thus, an approach has been found
for possibly modulating the physiological activity of peptides
containing aromatic amino acids.
The two main states relative to the hydantoin ring were de-
fined as A, o-methyl group on the same side of the hydantoin
ring as the benzyl substituent (torsion C7–C4–C9–C14; −90
45°) and B, o-methyl group on the opposite side as the ben-
zyl substituent (torsion C7–C4–C9–C14; 90 45°). For com-
pound 1, the population ratio of B to A is 37.3%:61.6%. For
compound 2, the population ratio of B to A is 42.3%:56.8%.
Experimental
1H and 13C NMR Studies on 3-o-Tolylhydantoins. One-Di-
mensional 1H and 13C NMR. 1H NMR spectra were taken us-
ing a 30° pulse angle by either a 500 MHz Varian Unity spectrom-
eter equipped with a 3-mm triple-resonance gradiant probe (Nalo-
1
This result corresponds to the result of a H NMR study (see
Table 1).

S. H. Park et al.
Bull. Chem. Soc. Jpn., 74, No. 10 (2001) 1923
Fig. 12. Supereimposed conformers of state ꢀ, ꢁ, and ꢂ of 1
from the MC/SD Simulation.
Fig. 13. Supereimposed conformers of state ꢀ, ꢁ, and ꢂ of 2
from the MC/SD Simulation.
rac) or a 300 MHz Varian Unity spectrometer equipped with a 5-
mm four-nuclei (1H, 13C, 19F, and 31P) probe. Each spectrum was
obtained by a Sun “Spark” Data Station using the in-house
MNMR software. Each sample was dissolved in either CDCl3 or
DMSO-d6.
fined by C7–C4–C9–C14 to determine the rotamer states of the o-
tolyl groups.
General Procedure for the Synthesis of 3-o-Tolylhydantoins
(1–4). The melting points were determined on a Mel-Temp (50/
60 cycles, 110–120 volts, 250 watts) apparatus and are uncorrect-
ed. Infrared spectra were recorded on either a Perkin–Elmer 1310
or a Perkin-Elmer 1760 FTIR spectrophotometer. NMR spectra
were recorded on a Bruker 200-MHz FTNMR spectrometer, a 500
MHz Varian Unity spectrometer equipped with a 3-mm triple-res-
onance gradiant probe (Nalorac) or a 300 MHz Varian Unity spec-
trometer equipped with a 5-mm four-nuclei (1H, 13C, 19F, and 31P)
probe. 1H NMR and 13C NMR chemical shifts are reported in δ
value vs TMS and CHCl3, respectively. Mass spectra were ob-
tained on a Scientific Research Instruments Biospect mass spec-
trometer. Flash column chromatography was performed on silica-
gel 60 (230–400 mesh, Merck). All chromatographic separations
were monitored by TLC analyses, performed using glass plates
precoated with 0.25-mm 230–400-mesh silica gel impregnated
with a fluorescent indicator (254 nm).
Preparation of 5-Isobutyl-3-o-tolylhydantoin (3). DL-Leu-
cine (1.00 g, 7.6 mmol) was dissolved in 15 mL of water contain-
ing 0.51 g (9.1 mmol) of potassium hydroxide at 0–5 °C. o-Tolyl
isocyanate (1.21 g, 9.1 mmol) was added dropwise into the amino
acid solution over a period of 5 min. The reaction mixture was
warmed to 60–70 °C and kept at that temperature for 1 h. The
mixture was filtered and the filtrate was acidified with conc. hy-
drochloric acid in order to precipitate the hydantoic acid, which
was then separated by filtration. A solution of 4 mL of conc. hy-
Temperature Study. 1H NMR spectra at various tempera-
tures were recorded by a 300 MHz Varian Unity spectrometer
equipped with a 5-mm two-nuclei (1H and 19F) probe. The cali-
bration of the probe temperature was made immediately after the
sample was analyzed at the temperature using an ethylene glycol
calibration sample. Each sample was dissolved in DMSO-d6.
Molecular Modeling Studies on 3-o-Tolylhydantoins.
MC–SD simulations were applied to 3-o-tolylhydantoins to deter-
mine their conformational populations. A hybrid Monte Carlo
(MC)–stochastic dynamics (SD) calculation was performed using
a continuum water solvent mode16 implemented in Macro Model17
version 5.0. The simulation was conducted in water (dielectric
constant = 78.7), as being the most representative dielectric avail-
able which is similar to DMSO (dielectric constant = 45). After a
period of 100 ps of stochastic dynamics equilibrium, long simula-
tions (5000 ps) were carried out at 300 K. The simulations were
to be converged by the obtained constant temperature and average
enthalpies. All three rotatable bonds were permitted to vary dur-
ing the simulations. During the simulations, 1000 snapshots were
taken at equal intervals. From this sampling, the populations of
the folded versus extended forms were obtained, as well as the
populations of the o-tolyl group being up, versus down. The tor-
sions were defined by H23–C1–C16–H32 and H6–C1–C16–H31
to quantitate to the groups previously defined as conformer ꢀ, ꢁ,
and ꢂ described earlier (Figs. 9 and 11). The torsions were de-

1924 Bull. Chem. Soc. Jpn., 74, No. 10 (2001)
Conformational Studies on 3-o-Tolylhydantoins
drochloric acid and 4 mL of water was added to the hydantoic acid
derivative and refluxed until the precipitate of the hydantoic acid
dissolved in the acidic solution (1–2 h), effecting cyclization to the
hydantoin. After cooling the solution, the hydantoin precipitated
out and was filtered. Recrystallization from absolute ethanol gave
0.40 g (23% yield) of the title compound as a white solid: mp
123–125 °C (Ref. 1, 182–183 °C); IR (CHCl3) 1710 cm−1
claisen head was attached to the flask and residual methanol was
distilled under a slow stream of carbon dioxide until the tempera-
ture of the head reached 110 °C. Then, a 5-plate bubble-cap col-
umn was attached and distillation continued until the head temper-
ature reached 152 °C. The solution was then cooled to room tem-
perature and stirred under carbon dioxide for 1–3 h.
Preparation of 5-Benzyl-3-o-tolylhydantoin (1) (adapted
(–NHCO–); CIMS (CH4, 160 °C) m/z 233 [M+H]+; H NMR
from that of Finkbeiner19).
2 M MMC in DMF (2 mL) was
1
(300 MHz, CDCl3) δ 0.90–0.95 (m, 6H, 2 CH3), 1.53–1.61 (m,
1H, CH), 1.76–1.83 (m, 2 H, CH2), 2.13 and 2.16 (2s, total 3H, o-
Me), 4.12–4.18 (m, 1H, C5 H), 6.44 (s, 1H, NH), 7.05 and 7.10
(2d, J = 7.5 Hz, total 1H, o-H), 7.10–7.29 (m, 3H, Ar); 13C NMR
(200 MHz, CDCl3) δ 17.6 17.8, 21.7, 21.8, 23.0, 23.1, 25.0, 25.2,
25.3, 41.1, 41.7, 56.1, 56.2, 126.8, 126.9, 128.4, 128.6, 129.4,
130.4, 130.5, 131.1, 131.2, 136.2, 136.4, 156.7, 173.3.
placed into a 50-mL round-bottom flask equipped to maintain a
nitrogen atmosphere. The solution was stirred with a magnetic
stirring bar for 0.5 h under carbon dioxide and then heated to 80
°C. The carbon dioxide was replaced with nitrogen and the nitro-
gen atmosphere was maintained throughout the reaction. To the
reaction mixture was added 4 (380 mg, 2 mmol), and the solution
was heated at 80 °C and kept at that temperature for 1 to 1.5 h.
Then, benzyl bromide (376 mg, 2.2 mmol) was added and the re-
action mixture was allowed to be stirred for 5 h at 110 °C. It was
then cooled and poured onto a mixture of 2 g of ice and 0.5 mL of
conc. hydrochloric acid. A gummy solid precipitated, which be-
came hard upon standing. The mixture was allowed to stand over-
night to complete precipitation. Once precipitation was complete,
the solid was filtered. The crude product was then dried. Column
chromatography on silica gel (hexane/ethyl acetate: 7/3) gave 3.60
g (64% yield) of the title compound as a white solid: mp 134–135
°C (Ref. 20, 134–135 °C); IR (CHCl3) 1700 cm−1 (–NHCO–); 1H
NMR (300 MHz, CDCl3) δ 1.76 and 2.20 (2s, total 3H, o-Me),
3.04–3.30 (m, 2H, benzylic), 4.44–4.48 (m, 1H, CH), 6.18 (s, 1H,
NH), 6.63 and 7.12 (2 dd, J = 7.5 Hz, total 1H, o-H), 7.23–7.37
(m, 8H, Arm); 13C NMR (300 MHz, CDCl3) δ 17.2 and 17.9
(CH3), 37.8 and 38.0 (PhCH2), 58.5 and 58.7 (C5), 127.0, 127.7,
127.8, 128.3, 128.5, 129.0, 129.7, 129.8, 130.0, 130.1, 131.2,
131.3, 134.7, 136.4, 156.6, 172.0, 172.1; CIMS (CH4, 160 °C) m/z
281 [M+H]+; Anal. Calcd for C17H16N2O2: C, 72.89; H, 5.62; N,
10.05%. Found: C, 72.92; H, 5.62; N, 10.09%.
Preparation of o-Tolylhydantoic Acid.
o-Tolylhydantoic
acid was prepared by a modification of the method used by
Finkbeiner.19 Glycine (1.5 g, 20 mmol) was dissolved in 8 mL of
potassium hydroxide solution (68 g of potassium hydroxide in 400
mL of water). The solution was stirred on a magnetic stirrer while
o-tolyl isocyanate (2.9 mL, 22 mmol) was added dropwise over a
period of 20 min (during the addition of the o-tolyl isocyanate, the
mixture first became cloudy and then a solid precipitate of 1,3-di-
o-tolylurea separated). The reaction mixture was kept at room
temperature overnight. The solid urea was removed by centrifug-
ing (5000 rpm, 15 min) and a cloudy supernatant layer was fil-
tered. The precipitate was washed three times with a small
amount of the potassium hydroxide stock solution. The combined
filtrate and washes were acidified with conc. hydrochloric acid
when a white precipitate was formed. After cooling in an ice bath
for 1–2 h to complete precipitation, the precipitated o-tolylhydan-
toic acid was filtered and dried. The product was purified by re-
crystallization from hot methanol to give 3.00 g (73% yield) of the
title compound as a white solid: mp 200–201 °C; IR (CHCl3) 1710
cm−1 (–NHCO–); CIMS (CH4, 160 °C) m/z 209 [M+H]+.
Preparation of 5-Pentafluorobenzyl-3-o-tolylhydantoin (2)
Preparation of 3-o-Tolylhydantoin (4). 3-o-Tolylhydantoin
was prepared by a modification of the method used by
Finkbeiner.19 The o-tolylhydantoic acid (3.00 g, 14.4 mmol) ob-
tained previously was added to a solution of 6 mL of conc. hydro-
chloric acid and 1.5 mL of water. Upon heating to reflux, the mix-
ture became a clear solution. After refluxing for 2 h, the reaction
mixture was allowed to cool to room temperature and stand over-
night to complete precipitation. The precipitated 3-o-tolylhydan-
toin was filtered and dried. The product was purified by recrystal-
lization from absolute methanol to give 2.53 g (92% yield) of the
title compound as a white solid: mp 152–153 °C (Ref. 10, 153
°C); IR (CHCl3) 1710 cm−1 (–NHCO–); CIMS (CH4, 160 °C) m/z
191 [M+H]+; 1H NMR (200 MHz, CDCl3) δ 2.23 (s, 3H, o-Me),
4.07 (s, 2H, CH2), 7.20 (s, 1H, NH), 7.05–7.40 (m, 4H, Arm).
Preparation of Magnesium Methyl Carbonate (MMC).
MMC in DMF–A 2 M solution of MMC in N,N-dimethylform-
amide (DMF) was prepared as described by Finkbeiner and
Stiles.21 A stock solution of magnesium methyl carbonate in N,N-
dimethylformamide was prepared as follows and used within one
month. Over a period of 3 h, magnesium turnings (6 g) were add-
ed to dry methanol (100 mL) in a 250-mL round-bottom flask
equipped with a condenser, a stirrer and a gas inlet tube. After the
magnesium had reacted completely, methanol was removed under
reduced pressure at a bath temperature of 50 °C. N,N-Dimethyl-
formamide (commercial untreated) was added to give a total vol-
ume of 125 mL and carbon dioxide was passed to the stirred solu-
tion as rapidly as it was absorbed. After the solid had dissolved, a
(adapted from that of Finkbeiner19).
Following the procedure
described above for the preparation of 1 using benzyl bromide, α-
bromo-2,3,4,5,6-pentafluorotoluene (1.73 g, 6.6 mmol) was used.
Column chromatography on silica gel (hexane/ethyl acetate: 7/3)
gave 1.10 g (50% yield) of the title compound as a white solid: mp
189–191 °C (EtOAc); IR (CHCl3) 1685 cm−1 (–NHCO–); 1H
NMR (500 MHz, DMSO-d6) δ 1.93 and 2.10 (2s, total 3H, o-Me),
3.13 and 3.16, 3.20 and 3.23 (2 dd, J = 7.0 Hz, total 1H, benzyl-
ic), 4.52 and 4.53, 4.58 and 4.59 (2 dd, J = 7.0 Hz, total 1H, CH),
7.10 and 7.16 (2d, J = 7.5 Hz, total 1H, o-H), 7.25–7.34 (m, 3 H,
Arm), 8.53 (s, a H, NH); 13C NMR (300 MHz, DMSO-d6) δ 17.1
and 17.9 (o-CH3), 25.5, 45.2, 56.0 and 56.2 (C5), 127.2, 129.3,
129.6, 131.1, 156.0, 172.2; CIMS (CH4, 170 °C) m/z 372
[M+2]+, 371 [M+H]+, 189 [M−181]+; Anal. Calcd for
C17H16N2O2: C, 55.14; H, 2.99; N, 7.56%. Found: C, 55.06; H,
2.82; N, 7.60%.
We would like to thank Dr. Hideji Fujiwara, Monsanto
Company, St. Louis, MO for NMR spectra and Dr. Frank P.
Hollinger, Shering–Plough Research Institute, Kenilworth, NJ
for molecular modeling.
References
1
A. K. Bose, M. S. Manhas, R. F. Tavares, R. F. Van Der
Veen, and H. Fujiwara, Heterocycles, 7, 1227 (1977).

S. H. Park et al.
Bull. Chem. Soc. Jpn., 74, No. 10 (2001) 1925
2
a) D. S. Dougherty, Science, 271, 163 (1996). b) J. L.
(1956).
Sussman, I. Silman, M. Harel, D. M. Quinn, and H. K. Nair, J.
Am. Chem. Soc., 118, 2340 (1996).
13 L. F. Johnson and W. C. Janknowski, “Carbon-13 NMR
Spectra,” A Wiley-Interscience Publication, New York (1972).
14 G. C. Levy and G. L. Nelson, “Carbon-13 Nuclear Magnet-
ic Resonance for Organic Chemists,” John Wiley and sons Inc.,
New York (1972).
3
W. G. Protor and F. C. Yu, Phys. Rev., 77, 717 (1950).
4
(1951).
5
H. S. Gutowsky and D. W. MaCall, Phys. Rev., 82, 748
E. R. Andrew, “Nuclear Magnetic Resonance,” Cambridge
15 L. M. Jackman and D. P. Kelly, J. Chem. Soc. (B), 102
(1970).
University Press, London (1955), Chap. 2.
6
(1994).
7
F. Guarnieri and W. C. Still, J. Comput. Chem., 15, 1302
16 a) W. Hasel, T. F. Hendrickson, and W. C. Still, Tetrahe-
dron Comput. Method, 1, 103 (1988). b) W. C. Still, A.
Tempezyk, R. C. Hawley, and T. Hendrickson, J. Am. Chem. Soc.,
112, 6127 (1990). c) W. C. Still and F. P. Hollinger, U.S. Patent 5
420 805, May 30, 1995.
a) J. R. Cavanough, J. Am. Chem. Soc., 89, 1558 (1967). b)
J. R. Cavanough, J. Am. Chem. Soc., 90, 4533 (1968). c) J. R.
Cavanough, J. Am. Chem. Soc., 92, 1488 (1970).
8
a) A. K. Bose and R. F. Tavares at the 156th American
17 F. Mohamadi, N. G. J. Richards, W. C. Guida, R. Liskamp,
M. Lipton, C. Caufield, G. Chang, T. Hendrickson, and W. C. Still,
J. Comput. Chem., 11, 440 (1990).
18 R. F. Tavares, Ph.D. Dissertation, Stevens Institute of Tech-
nology (1968).
Chemical Society National Meeting, Atlantic City, N. J., PHYS-
152, (1968). b) R. F Tavares, Ph.D. Dissertation, Stevens Institute
of Technology (1968).
9
M. Karplus, J. Chem. Phys., 30, 11 (1959).
10 K. G. R. Pachler, Spectro. Chem. Acta, 19, 2085 (1963); R.
B. Martin and R. Mathur, J. Am. Chem. Soc., 87, 1065 (1965).
11 Group electronegativities of benzyl group, φCH2, and hy-
dantoin ring are estimated to 2.4 and 5.5, respectively.
19 H. Finkbeiner, J. Org. Chem., 30, 3414 (1965).
20 H. Fujiwara, Ph.D. Dissertation, Stevens Institute of Tech-
nology (1974).
21 H. Finkbeiner and M. Stiles, J. Am. Chem. Soc., 85, 614
(1963).
12 R. E. Glick and J. Bothner-By, Chem. Phys., 25, 362,