Investigation of the barrier to the rotation of carbamate and amide C-N bonds in antidepressant (6aR*,11bS*)-7-[carbobenzyloxy-l-alanyl]-2-[(4-methylphenyl)sulfonyl]-1,2,3,4,6,6a,7,11b,2,12a(S)-decahydropyrazino[2′,1′:6,1]pyrido[3,4-

Article abstract of DOI:10.1016/j.tet.2007.03.171

Two concurrent exchanges arising due to the restricted rotation around the carbamate C-N bond and amide C-N bond were observed in (6aR^*,11bS^*)-7-[carbobenzyloxy-l-alanyl]-2-[(4-methylphenyl)sulfonyl]-1,2,3,4,6,6a,7,11b,2,12a(S)-decah

Full text of DOI:10.1016/j.tet.2007.03.171

Tetrahedron 63 (2007) 5236–5243
Investigation of the barrier to the rotation of carbamate and amide
C–N bonds in antidepressant (6aR*,11bS*)-7-[carbobenzyloxy-L-
alanyl]-2-[(4-methylphenyl)sulfonyl]-1,2,3,4,6,6a,7,11b,2,12a(S)-
decahydropyrazino[20,10:6,1]pyrido[3,4-b]indole by
dynamic NMR and molecular mechanics*
a,
Aparna Anand,a Abhijeet Deb Roy,a Ruchika Chakrabarty,b Anil K. Saxenab, and Raja Roy
*
*
aDivision of SAIF, Central Drug Research Institute, Lucknow 226 001, UP, India
bMedicinal and Process Chemistry Division, Central Drug Research Institute, Lucknow 226 001, UP, India
Received 1 December 2006; revised 14 March 2007; accepted 29 March 2007
Available online 4 April 2007
Abstract—Two concurrent exchanges arising due to the restricted rotation around the carbamate C–N bond and amide C–N bond were
observed in (6aR*,11bS*)-7-[carbobenzyloxy-L-alanyl]-2-[(4-methylphenyl)sulfonyl]-1,2,3,4,6,6a,7,11b,2,12a(S)-decahydropyrazino-
[20,10:6,1]pyrido[3,4-b]indole by NMR spectroscopic experiments. A total of four low energy conformers were evaluated in the
molecule, out of those, two were observed because of the restricted rotation of the amide C–N bond in CDCl3 and two were observed due
to restricted carbamate C–N bond rotation in DMSO-d6 and (CD3)2CO. The barrier to the rotation (DGz) around carbamate C–N bond
and amide C–N bond was determined using dynamic NMR calculations. Molecular mechanics calculations also provided evidence for the
presence of four low energy conformers for the compound due to restricted amide rotation and carbamate C–N bond rotation, with the value
of barriers (DGz) between them of the order of 15.0 kcal/mol, which is in agreement with the dynamic NMR results. Since the molecule has
shown potent antidepressant activity, it is proposed that these dynamic properties could influence the activity profile of these classes of
molecules.
Ó 2007 Elsevier Ltd. All rights reserved.
1. Introduction
DGz of the amide C–N bond was considerably increased
upon changing the polarity of solvents.4 Moreover, the low-
ering of the rotational barrier of carbamate C–N bond has
also been reported,5 which was attributed to the resonance
effect exerted by an N-substituted aryl ring. This evidence
indicated that the barrier to rotation of the carbamate and
amide C–N bonds might depend upon some of the other
analogous factors, which still remained unexplored. Al-
though much of the work carried out so far specifically deals
with the study of solvent effects on DGz of a molecule con-
taining either the carbamate functionality or amide function-
ality, there has been no report in the literature regarding the
effect on DGz of either of the functional groups when both of
them existed in the same system.
In recent years, the kinetics and thermodynamics of the
stereodynamic processes occurring due to restricted intra-
molecular rotation have been extensively explored using dy-
namic NMR studies.1 The phenomenon of hindered rotation
about carbamate2 and amide3 C–N bonds has received much
attention as rotational changes play significant role in con-
formational stereochemistry of a compound, which could
influence the activity profile. From the previously reported
experimental data, it has been observed that carbamate
C–N bond and amide C–N bond have nearly equal barriers
to the rotation (DGz).4 Previously Lectka and Cox demon-
strated4 the effect of various solvents on DGz of the carba-
mate and amide C–N bonds, where the rotational barrier of
the carbamate C–N bond was found to have least sensitivity
toward the solvent polarity. Contrary to that of carbamates,
In our ongoing process of synthesizing derivative of CNS
active nucleus decahydropyrazino[20,10:6,1]pyrido[3,4-b]-
indole, we synthesized a potential antidepressant molecule
(6aR*,11bS*)-7-[carbobenzyloxy-L-alanyl]-2-[(4-methyl-
phenyl)sulfonyl]-1,2,3,4,6,6a,7,11b,12,12a(S)-decahydropyr-
azino[20,10:6,1]pyrido[3,4-b]indole (4). During the process of
characterization of this molecule the 1H NMR spectrum
*
CDRI Communication No. 6970.
Keywords: Dynamic NMR; Carbamate; Rotational barrier; Decahydropyr-
azino[20,10:6,1]pyrido[3,4-b]indole.
* Corresponding authors. E-mail addresses: anilsak@hotmail.com;
rajaroy_cdri@yahoo.com
0040–4020/$ - see front matter Ó 2007 Elsevier Ltd. All rights reserved.
doi:10.1016/j.tet.2007.03.171

A. Anand et al. / Tetrahedron 63 (2007) 5236–5243
5237
revealed few broad resonances at room temperature, indicat-
ing the effect of barrier to the rotation of the carbamate and
or amide C–N bond, which led us to the current study. In
the present communication, we herein describe the investiga-
tion of the barrier to the rotation of carbamate and amide C–N
bonds when both are present in a single system. Further calcu-
lations of the activation parameters involved in this dynamic
exchange process were also determined in two different sol-
vents by two-dimensional exchange spectroscopy and molec-
ular mechanics.
various one- and two-dimensional NMR experiments. The
unequivocal assignments of 4 were performed by the com-
bined use of 1H, 13C, DEPT, COSY, edited HSQC, HMBC,
and NOESY NMR spectra recorded in three solvents, in
CDCl3 at 248 K and in (CD3)2CO and DMSO-d6 solution
at 298 K.
2.2. Exchange phenomenon in CDCl3 solution
1
The H NMR spectrum of 4 recorded at 298 K in CDCl3
solution showed pronounced line broadening of all the reso-
nances, revealing the expected dynamic exchange process in
the system. Hindered rotation around the amide C–N bond
was indicated by the signal of the H-8 proton, which
2. Results and discussion
2.1. Synthesis
1
appeared distinctly as a broad resonance at 7.8 ppm in H
NMR spectrum recorded in CDCl3 solution at 298 K, and
further emerged as an ortho-coupled doublet resonating at
8.0 ppm on decreasing the temperature to 248 K (Fig. 1a).
The HSQC spectrum recorded at 248 K allowed observation
of cross peaks owing to two rotamers and the exact ratio of
population of rotamers was found to be 4:1 by taking the
integral values of the partially resolved signals.
We synthesized a new analogue of CNS active drug cent-
butindole,6 (6aR*,11bS*)-7-[carbobenzyloxy-L-alanyl]-2-[(4-
methylphenyl)sulfonyl]-1,2,3,4,6,6a,7,11b,12,12a(S)-deca-
hydropyrazino[20,10:6,1]pyrido[3,4-b]indole (4), which has
both carbamate and amide functional groups. The selection
of this molecule was based upon the unique semi-rigid
conformation of the pyrazino[20,10:6,1]pyrido[3,4-b]indole
nucleus (1) that does not acquire any dynamic behavior
and also possesses CNS depressant,7 antihistaminic,8 hypo-
tensive,6 phosphodiesterase inhibitory,9 and neoplasm inhib-
itory activities.10
Moreover, the 2D exchange spectrum recorded at 248 K
showed fully resolved exchange cross peaks for H-8 proton
(Fig. 1b) further reinstating the hindered rotation of amide
C–N bond of the carbobenzyloxy-L-alanyl side chain in the
molecule. The rotation around the carbamate C–N bond
was found beyond the NMR time-scale in CDCl3 solution
as no exchange cross peak for the NH proton was observed
at any temperature. Armed with these observations it was
stated that in CDCl3 solution at 298 K, the presence of two
populations was clearly visible but can only be determined
within NMR time-scale by restricting the rotation around
amide C–N bond at a low temperature of 248 K. The magni-
tude of the downfield shift of the major H-8 proton and
upfield shift of minor H-8 proton also indicated the confor-
mation of the molecule as reported earlier by Nagarajan
et al. in case of N-acylindolines.12 Although the exact con-
formation of both the rotamers was not deduced at this point
of time, it was presumed that at lower temperature (248 K),
the CO group is preferentially oriented toward the phenyl
ring (rotamer A) while at room temperature it was in its
reverse orientation (rotamer B). These presumptions were
further supported by the results obtained by theoretical stud-
ies using Maestro 7.0.113 suite of programs with MMFF’s
force field.
The synthetic methodology commenced with the synthesis
of 2-[(4-methylphenyl)sulfonyl]-1,2,3,4,6,7,12,12a(S)-octa-
hydropyrazino[20,10:6,1]pyrido[3,4-b] indole (2) from the
tosylation of the previously reported 1,2,3,4,6,7,12,12a(S)-
octahydropyrazino[20,10:6,1]pyrido[3,4-b]indole molecule6
(1) using tosyl chloride in pyridine. This was followed by
the reduction of indole double bond using boranedimethyl-
sulfide complex11 in the presence of TFA at 0 ꢀC. The
reaction was completed in 3 h providing 2-[(4-methylphenyl)-
sulfonyl]-1,2,3,4,6,6a,7,11b,12,12a(S)-decahydropyrazino-
[20,10:6,1]pyrido[3,4-b]indole (3) with a high degree of
purity and yield. Finally, compound 4 was prepared by the
amidation of 3, treating it with carbobenzyloxy-L-alanine
in the presence of coupling agent DIC in dichloromethane.
The overall yield of compound 4 achieved after three steps
of reaction was 47% (Scheme 1).
Once the synthesis of the final compound (4) was achieved,
the complete structure elucidation was carried out using
NH
O2
p-TosCl, C5H5N,
N
S
(CH3)2SBH3, TFA,
N
30 °C, 25 min
(ether) 0 °C-30 °C, 3 h
N
N
H
N
H
(2)
(1)
6"
7"
3"
8''
5"
2"
O2S
O2S
N
1
4"
N
11b
DIC, dry DCM,
carbobenzyloxy-L-alanine,
0 °C-30 °C, 12 h
12
3
11
H
10
H
11a
4
N
N
9
6'
5'
6 O
8'
7a
H6a
N
8
N
4'
H
7'
O12'
9'
10'
H
NH
(3)
1'
3'
O
2'
11'
13'
(4)
Scheme 1. Strategy for the synthesis of 4.

5238
A. Anand et al. / Tetrahedron 63 (2007) 5236–5243
Figure 1. (a) 1H NMR spectrum of 4 in CDCl3 solution as a function of temperature. (b) Expanded region of NOESY spectrum showing exchange cross peak for
H-8 in CDCl3 solution at 248 K.
2.3. Exchange phenomenon in (CD3)2CO and DMSO-d6
solutions
carbamate C–N bond in (CD3)2CO and DMSO-d6, that is
within the NMR time-scale whereas the rotation of the amide
C–N bond was not observed in these solvents.
Since the rotation around the carbamate C–N bond was
not observed in CDCl3 solution at any temperature, the
exchange phenomenon was further re-examined in polar
solvents, viz., DMSO-d6 and (CD3)2CO. Contrary to CDCl3
solution, 2D exchange spectra recorded in (CD3)2CO and
DMSO-d6 solutions revealed the exchange cross peak for
the NH proton (Fig. 2). No exchange cross peak for the H-8
proton was observed in either solvent in the temperature
range of 278–318 K. These observations straightforwardly
suggested the existence of slow hindered rotation of the
2.4. Kinetics of exchange process
To establish the kinetic order of both the exchange pro-
cesses, the concentration dependence of the NOESY spectra
has been studied. No variation was found in the intensity of
the cross peak for the NH proton in DMSO-d6 and the H-8
proton in CDCl3 solution upon diluting the initial solution
concentration to its half value at 298 K taking the mixing
time of tm¼200 ms. Under these conditions the exchange
Figure 2. Expanded region of NOESY spectrum showing exchange cross peak for NH in (a) (CD3)2CO and (b) DMSO-d6 solutions.

A. Anand et al. / Tetrahedron 63 (2007) 5236–5243
5239
was about 13% complete in DMSO-d6 and 24% in CDCl3
solution. Since any substantial change in the rate constants
would immediately be reflected in the cross peak intensities,
it was therefore concluded that exchange of NH proton and
H-8 proton in their respective solvents obeyed first order
kinetics.13
ID is the intensity of diagonal peak and IC is the intensity of
cross peak in the NOESY spectrum.
The activation parameters involved in both exchange pro-
cesses were determined using the Eyring equation15 (Eq.
2) and are presented in Tables 3 and 4 and the Eyring plots
are given in Figure 3.
The Eyring analysis of 4 was performed to estimate the bar-
rier to rotation (DGz) around carbamate and amide C–N
bonds in their respective solvents. The rates of exchange
process were calculated by performing phase sensitive 2D-
NOESY using mixing times of 200, 300, 400 ms in CDCl3,
(CD3)2CO, and DMSO-d6 solutions at five different tem-
peratures. These temperature ranges varied for all three
solvents, as temperatures at which good separation of
exchange cross peaks of the respective signals was observed
were selected for the analysis. For instance, in the case of
(CD3)2CO the temperature range was 298–278 K. The
values of the rate constants have been obtained by Eq. 1,14
and presented in Tables 1 and 2.
À
Á
À
Á
Â
À
ÁÃ
ln k=T ¼ ꢂ DHz=RT þ 23:76 þ DSz=R
ð2Þ
Similar to the previously documented analysis,4 DGz of
our complex system did not show remarkable variation
with change of solvents. Despite having two concurrent
hindered rotations existing in the same system, the barrier
to rotation of carbamate and amide C–N bonds was found
to be z16 kcal/mol, which is also supported by the
theoretically calculated value. Therefore, it was concluded
that the rotational barrier of carbamate C–N bond
and amide C–N bond in a particular solvent is indepen-
dent of other exchange processes existing in the same
system.
Kz1=½tMðID=IC þ 1Þꢁ
ð1Þ
Table 1. Rate constants determined using Eq. 1 in (CD3)2CO and DMSO-d6
for DGz of carbamate bond
Table 3. Activation parametersa of carbamate C–N bond obtained using
Eq. 2 in (CD3)2CO and DMSO-d6
(CD3)2CO
T/K
Solvent
DHzb
DSzc
DGzb
278
283
288
293
298
k1 (sꢂ1
)
0.21ꢃ0.02 0.25ꢃ0.03 0.29ꢃ0.05 0.32ꢃ0.06 0.39ꢃ0.07
(CD3)2CO
DMSO-d6
2.89ꢃ0.98
4.27ꢃ1.01
ꢂ47.82ꢃ3.11
ꢂ45.16ꢃ3.68
15.93ꢃ0.17
16.60ꢃ0.06
k
(sꢂ1
)
2.36ꢃ0.23 2.65ꢃ0.44 2.95ꢃ0.44 3.08ꢃ0.55 3.48ꢃ0.68
ꢂ1
a
0 ꢀC.
DMSO-d6
T/K
b
Values in kcal/mol.
Values in cal/mol.
298
303
308
313
318
c
k1 (sꢂ1
)
0.29ꢃ0.04 0.33ꢃ0.05 0.38ꢃ0.07 0.44ꢃ0.12 0.49ꢃ0.13
k
(sꢂ1
)
1.28ꢃ0.30 1.53ꢃ0.40 1.79ꢃ0.47 2.02ꢃ0.60 2.43ꢃ0.75
ꢂ1
Table 4. Activation parametersa of amide C–N bond obtained using Eq. 2 in
CDCl3
Table 2. Rate constants determined using Eq. 1 in CDCl3 solution for DGz of
amide bond
Solvent
CDCl3
DHzb
DSzc
DGzb
6.42ꢃ0.07
ꢂ34.75ꢃ1.04
15.86ꢃ0.26
CDCl3
T/K
238
243
248
253
258
a
0 ꢀC.
Values in kcal/mol.
Values in cal/mol.
k1 (sꢂ1
)
0.22ꢃ0.02 0.31ꢃ0.04 0.41ꢃ0.04 0.54ꢃ0.04 0.72ꢃ0.07
b
k
(sꢂ1
)
0.11ꢃ0.01 0.15ꢃ0.01 0.20ꢃ0.01 0.29ꢃ0.01 0.34ꢃ0.01
ꢂ1
c
Figure 3. Eyring plots of the rate constants (a) in (CD3)2CO and DMSO-d6, (b) in CDCl3 solution with 200 ms mixing time determined by fitting the rate
constants to the experimental data at five different temperatures.

5240
A. Anand et al. / Tetrahedron 63 (2007) 5236–5243
Figure 4. Conformational profile for 4 obtained for amide C–N bond rota-
tion at the molecular mechanics (MMFF) level of theory.
Figure 6. Conformational profile for 4 obtained for carbamate C–N bond
rotation at the molecular mechanics (MMFF) level of theory.
2.5. Theoretical studies
conformation (total energy 315.4 kJ/mol) shows that the
structure adopts a conformation where the carbonyl group
is facing the phenyl ring (rotamer A). In order to visualize
the effect of rotation along C–N bond on the overall energy
(potential energy) of the molecule, a conformational search
(Fig. 4) was carried out employing macromodel module,
dihedral drive with an increment of 5ꢀ rotation over 360ꢀ.
Inspection of the results reveals that the dip in Figure 4
corresponds to rotamer B (349.6 kJ/mol) having undergone a
rotation of w160ꢀ from the initial structure. The carbonyl
group in rotamer B lies almost opposite to the phenyl
ring as predicted earlier by NMR study. The above results
clearly indicate the appearance of two NMR signals for
H-8 proton (Fig. 5). The DGz values for 4, obtained from
Figure 4, show that rotamers A and B can be inter-converted
What is immediately striking from our observations is that
free rotation along the C–N bond can cause a difference in
the magnetic environment around the H-8 phenyl core and
hence two signals in the NMR spectrum for H-8 proton can
be observed. To gain insight into the problem of how the
amide group at indole-N may be impacting the variation
in the magnetic field of H-8 proton via changes in the
conformation and also to obtain the exact conformation
of these two rotamers along with the confirmation of the
barrier to rotation of the four rotamers, conformational
dynamics of 4 was studied. The minimum energy confor-
mation of 4 was obtained using Maestro 7.0.113 suite of
programs with MMFF’s force field. The energy minimized
Figure 5. Exact conformation of the two rotamers A and B obtained in CDCl3 due to amide C–N bond rotation.

A. Anand et al. / Tetrahedron 63 (2007) 5236–5243
5241
Figure 7. Exact conformation of the two rotamers C and D obtained in DMSO-d6 due to carbamate C–N bond rotation.
by overcoming energy barriers of 14.46 kcal/mol (60.46 kJ/
mol).
3. Conclusions
In summary, our investigations demonstrated the complex
dynamic behavior of (6aR*,11bS*)-7-[carbobenzyloxy-
L-alanyl]-2-[(4-methylphenyl)sulfonyl]-1,2,3,4,6,6a,7,11b,12,
12a(S)-decahydropyrazino[20,10:6,1]pyrido[3,4-b]indole (4).
The molecule was specifically synthesized for the first time
to examine the mutual effect of the hindered rotation around
carbamate and amide C–N bonds and it demonstrated the
presence of four rotamers of the molecule. It was proposed
using concentration dependent NOESY experiment that
both the exchange processes obey first order kinetics. The
barrier to rotation of both amide and carbamate bonds of 4
was found to be around 15–16 kcal/mol by both dynamic
NMR calculations and theoretical calculations and typically
insensitive to the coexisting bond rotations. Our study sig-
nifies a new dimension to the ongoing research in this field
to explore such type of complex systems bearing two or
more functional groups. Moreover, the molecule has been
found to be active as antidepressant, it is therefore believed
that the effect of this dynamic behavior using similar types of
substitutions may be one of the deciding factors in influenc-
ing the activity profile of these classes of compounds.
Similar conformational dynamics study of the molecule 4
was carried out for the identification of the two rotamers,
which originated due to restricted rotation along the carba-
mate C–N bond. Since the restricted rotation of the carba-
mate C–N bond was observed in the case of DMSO-d6 and
(CD3)2CO and not in CDCl3, the exact conformation for
the molecule 4 in DMSO-d6 was calculated using MMFF’s
force field. The molecular modeling was carried out using
water as a solvent and keeping the dielectric constant of
DMSO at 298 K. The energy minimized conformation (total
energy 364.79 kJ/mol) shows that the structure adopts a
conformation where the carbonyl group of the carbamate
is parallel to the NH (rotamer C). Examination of results
again reveals that the dip in Figure 6 corresponds to
rotamer D (383.53 kJ/mol) having incorporated rotation of
w170ꢀ from the initial structure. The carbonyl group in
rotamer B lies almost opposite to the NH group in this
case. The DGz values for 4, obtained in DMSO-d6 from
Figure 6, indicated that rotamers C and D can be inter-
converted by overcoming energy barriers of 16.41 kcal/
mol (68.63 kJ/mol) (Fig. 7).
2.6. Biological activity
4. Experimental
In vitro receptor binding studies for dopaminergic (D2) and
seratonergic (5-HT2A) were carried out using rat brain cau-
date region and cortex membrane, respectively. The radioli-
gand and reference compound used for D2 receptor binding
studies were [3H] Spiperone and Haloperidol (Ki¼0.43 nM),
respectively, and for 5-HT2A receptor binding studies the
radioligand and reference compound used were [3H]
Ketanserin and Ketanserin (Ki¼0.38 nM), respectively.
Compound 4 has shown a very good affinity for 5-HT2A re-
ceptors (Ki¼0.52 nM) with no affinity for the D2 receptor.
The selectivity of this compound toward seratonergic recep-
tors may be converted into its use as an antidepressant. The
compound has also shown 16% antihistaminic (H1) activity
(at 1 mg/mL) when tested on Guinea pig ileum (GPI) for
blockade of histamine-induced contractions. The standard
drug used in this case was Ceterizine with 34% histamine
blockade (at 1 mg/mL).
4.1. General considerations
All the solvents used for the reactions and purifications were
commercially available and used after distillation. All the
products were characterized by 1H, 13C, DEPT90, DEPT135,
two-dimensional Correlation Spectroscopy (COSY), Hetero-
nuclear Single Quantum Coherence (HSQC), and Hetero-
nuclear Multiple Bond Correlation spectroscopy (HMBC).
The NMR spectra were recorded using a BRUKER Avance
DRX 300 MHz FT NMR spectrometer equipped with
a 5 mm multinuclear inverse probehead with z-shielded gra-
dient. Experiments were recorded in DMSO-d6, (CD3)2CO,
and in CDCl3 in the temperature range from 298 K to 248 K.
Chemical shifts are given on the d scale and are referenced to
TMS at 0.00 ppm for the proton and 0.00 ppm for the car-
bon. In the 1D measurement (1H and 13C) 32K data points

5242
A. Anand et al. / Tetrahedron 63 (2007) 5236–5243
were used for the FID. The pulse programs of the following
2D experiments were taken from the Bruker software library
and the parameters were as follows.
106.7 (C-11b), 110.9 (C-8), 118.1 (C-11), 119.8 (C-10),
121.9 (C-9), 127.1 (C-11a), 128.1 (C-30, 70), 129.4 (C-40,
70), 130.3 (C-20), 132.4 (C-6a), 136.2 (C-7a), 144.1 (C-50).
Mass (FAB) m/z 382 (M++1). Anal. Calcd for C21H23N3O2S:
C, 66.12; H, 6.08; N, 11.01. Found C, 66.05; H, 6.15; N,
11.02.
300/75 MHz gradient HSQC spectra: relaxation delay
d1¼2 s; evolution delay d2¼3.44 ms; 90ꢀ pulse, 6.85 ms for
1H, 10 ms for 13C, hard pulses at ꢂ3.0 dB and 60 ms for
13C GARP decoupling with gradient ratio GPZ1:GPZ2:
GPZ3¼50:30:40.1; 1024 data points in t2; spectral width
9.0 ppm in F2 and 160 ppm in F1; number of scans 32;
256 experiments in t1; linear prediction to 512; zero filling
up to 1K and apodization with sine-bell in both dimensions
prior to double Fourier transformation.
4.1.3. (6aR*,11bS*)-2-[(4-Methylphenyl)sulfonyl]-1,2,
3,4,6,6a,7,11b,12,12a(S)-decahydropyrazino[20,10:6,1]-
pyrido[3,4-b]indole (3). A solution of borane dimethyl
sulfide complex (10 M, 0.5 mL) was added dropwise to a
stirred solution of 2 (0.95 g, 2.5 mmol) in trifluoroacetic
acid (7.3 mL) at 0 ꢀC in an atmosphere of nitrogen for
10 min. The reaction mixture was stirred for another 3 h at
30 ꢀC. Workup was done by adding water (0.5 mL) to the re-
action mixture, concentrating, and basifying with ammonia
solution. The precipitated product was filtered off and crys-
tallized from ethanol/water to give white crystals of 3
(0.65 g, 68%), mp 110 ꢀC. IR (KBr): ~n 3367, 2922, 2834,
1596, 1350, 1165, 1005, 760, 742, 658, 550 cmꢂ1. FABMS:
m/z 384 [MH+]. 1H NMR (300 MHz, CDCl3) 298 K, d¼1.07
(m, 1H, H-12ax), 1.68 (m, 1H, H-12eq), 1.96 (t, J¼10.4 Hz,
1H, H-1ax), 2.11 (m, 1H, H-12a), 2.34 (m, 1H, H-4ax), 2.38
(s, 3H, H-80), 2.43 (m, 1H, H-6ax), 2.53 (dt, J¼2.3 Hz and
11.5 Hz, 1H, H-3eq), 2.83 (br d, J¼11.3 Hz, 1H, H-4eq),
2.99 (m, 1H, H-11b), 3.11 (d, J¼13.5 Hz, 1H, H-6eq),
3.50 (m, 1H, H-1eq), 3.63 (dd, J¼1.8 Hz and 10.5 Hz, 1H,
H-3ax), 3.81 (br d, J¼6.1 Hz, 1H, 6a), 6.64 (d, J¼7.8 Hz,
1H, H-8), 6.71 (dt, J¼0.5 Hz and 7.5 Hz, 1H, H-10), 6.97
(dt, J¼1.0 Hz and 7.6 Hz,1H, H-9), 7.08 (d, J¼7.2 Hz, 1H,
H-11), 7.26 (d, J¼8.1 Hz, 2H, H-30 and H-70), 7.57 (d,
J¼8.1 Hz, 2H, H-40 and H-60); 13C NMR (75 MHz,
CDCl3) 298 K, d¼21.6 (C-130), 33.8 (C-12), 38.6 (C-11b),
45.9 (C-3), 51.3 (C-1), 54.2 (C-4), 56.2 (C-6), 57.1
(C-12a), 58.9 (C-6a), 111.0 (C-8), 119.5 (C-10), 123.6
(C-11), 127.8 (C-9), 128.1 (C-30, 70), 129.7 (C-40, 70),
131.9 (C-20), 134.5 (11b), 143.9 (C-50), 150.1 (C-7a). Mass
(FAB) m/z 384 (M++1). Anal. Calcd for C21H25N3O2S: C,
65.77; H, 6.57; N, 10.96. Found C, 65.75; H, 6.59; N, 11.02.
300/75 MHz gradient HMBC spectra: relaxation delay
d1¼2 s; delay of the low-pass J-filter d2¼3.44 ms; delay
for evolution of long-range coupling d6¼71 ms with gradi-
ent ratio same as HSQC; 2048 data points in t2; spectral
width 11.0 ppm in F2 and 240 ppm in F1; number of scans
52; 256 experiments in t1; linear prediction to 512; zero fill-
ing up to 2K and apodization with 90ꢀ shifted square sine-
bell in F1 dimension and sine-bell in F2 dimension prior
to double Fourier transformation.
4.1.1. General procedure for the synthesis of 1,2,3,
4,6,7,12,12a(S)-octahydropyrazino[20,10:6,1]pyrido[3,4-
b]indole (1). Compound 1 was prepared by reported pro-
cedure taking (S)-(ꢂ)-tryptophan as starting material.6 1H
NMR (300 MHz, DMSO-d6) 298 K, d¼3.07 (m, 3H, H-
12a, H-12ax, H-4ax), 3.27 (t(o), 1H, H-1ax), 3.37 (m, 1H,
H-12eq), 3.50 (t(o), 1H, H-3eq), 3.61 (t(o), 2H, H-3ax, H-
4eq), 3.76 (d, J¼11.8 Hz, 1H, H-1eq), 4.06 (d, J¼15.2 Hz,
1H, H-6ax), 4.35 (t(o), 1H, NH-2), 4.56 (d, J¼14.9 Hz,
1H, H-6eq), 7.70 (t(o), 1H, H-10), 7.78 (t(o), 1H, H-9),
8.05 (d(o), J¼8.0 Hz, 1H, H-11), 8.10 (d(o), J¼7.7 Hz,
1H, H-8), 11.52 (s, 1H, NH-7); 13C NMR (75 MHz,
DMSO-d6) 298 K, d¼25.6 (C-12), 45.7 (C-3), 52.1 (C-6),
52.4 (C-1), 55.9 (C-4), 57.9 (C-12a), 105.5 (C-11b), 110.5
(C-11), 117.3 (C-8), 118.3 (C-10), 120.3 (C-9), 126.6 (C-
11a), 131.9 (C-6a), 135.8 (C-7a). Mass (FAB) m/z 228
(M++1). Anal. Calcd for C14H17N3: C, 73.98; H, 7.54; N,
18.49. Found C, 74.05; H, 7.56; N, 18.59.
4.1.4. (6aR*,11bS*)-7-[Carbobenzyloxy-L-alanyl]-2-[(4-
methylphenyl)sulfonyl]-1,2,3,4,6,6a,7,11b,12,12a(S)-deca-
hydropyrazino[20,10:6,1]pyrido[3,4-b]indole (4). DIC
(0.2 mL, 1.3 mmol) was added to a stirred solution of carbo-
benzyloxy-L-alanine (0.3 g, 1.3 mmol) in dry DCM (15 mL),
stirring was continued for another 20 min and this reac-
tion mixture was cooled to 0 ꢀC. Compound 3 (0.5 g,
1.3 mmol) was added to the above mixture in portions at
0 ꢀC during 15 min. The reaction mixture was stirred for
another 12 h at 30 ꢀC. The reaction mixture was filtered
and the filtrate was concentrated under vacuum, water was
added to the residue, and extracted with ethyl acetate
(3ꢄ30 mL). The combined ethyl acetate layer was washed
with water, dried over sodium sulfate, and concentrated to
one third of the volume. The separated diisopropyl urea
was filtered off and the filtrate was finally concentrated.
The residue obtained was subjected to crystallization using
methanol yielding white needle-like crystals of 4 (0.52 g,
67.5%); mp 190ꢀ C. IR (KBr): ~n 3383, 2829, 2363, 1712,
4.1.2. 2-[(4-Methylphenyl)sulfonyl]-1,2,3,4,6,7,12,12a(S)-
octahydropyrazino[20,10:6,1]pyrido[3,4-b]indole (2). A
solution of 1 (1.4 g, 6.0 mmol) and 4-toluenesulfonyl chlo-
ride (1.4 g, 7.2 mmol) in dry pyridine 15 mL was stirred at
30 ꢀC for 25 min. The solid separated was filtered off,
washed with water (6ꢄ50 mL), and crystallized from
C2H5OH/H2O to give light yellow crystals of 2 (1.53 g,
65%) mp 178 ꢀC. IR (KBr): ~n 3361, 2904, 2833, 2364,
1592, 1450, 1360, 1164, 742, 658, 547 cmꢂ1. FABMS: m/z
1
382 [M+H+]. H NMR (300 MHz, CDCl3) 298 K, d¼2.36
(m, 1H, H-4ax), 2.44 (s, 3H, H-80), 2.55 (m, 1H, H-12ax),
2.64 (m, 2H, H-3eq, H-6ax), 2.75 (s, 1H, H-12a), 2.83 (m,
1H, H-12eq), 3.04 (m, 1H, H-6eq), 3.54 (d, J¼14.3 Hz,
1H, H-1ax), 3.64 (br s, 1H, H-3ax), 3.78 (d, J¼10.9 Hz,
1H, H-1eq), 3.88 (d, J¼14.3 Hz, 1H, H-4eq), 7.10 (m, 2H,
H-9, H-10), 7.28 (d, J¼7.7 Hz, 1H, H-8), 7.34 (d,
J¼8.1 Hz, 2H, H-40, H-60), 7.41 (d, J¼7.5 Hz, 1H, H-11),
7.68 (d (o), J¼8.1 Hz, 2H, H-30, H-70); 13C NMR
(75 MHz, CDCl3) 298 K, d¼21.8 (C-80), 25.5 (C-12), 46.3
(C-3), 51.6 (C-1), 51.8 (C-4), 53.8 (C-6), 56.3 (C-12a),
1658, 1597, 1362, 1238, 1162, 1114, 1024, 757 cmꢂ1
.
1
FABMS: m/z 589 [M+H+]. H NMR (300 MHz, CDCl3)
298 K, d¼1.26 (br s, 2H, H-12), 1.38, (d, J¼6.3 Hz, 3H,
H-130), 1.89 (t, J¼10.2 Hz, 1H, H-1ax), 2.10 (m, 1H,

A. Anand et al. / Tetrahedron 63 (2007) 5236–5243
5243
H-3eq), 2.41 (s, 3H, H-800), 2.54 (m, 1H, H-12a), 2.63 (m,
Acknowledgements
3H, H-6, H-4ax), 3.04 (s, 1H, H-4eq), 3.35 (br d, 2H, H-
1eq, H-3ax), 3.51 (s, 1H, H-11b), 4.44 (br s, 1H, H-6a),
4.68 (br s, 1H, H-20), 5.09 (q, J¼9.2 Hz, 2H, H-60), 5.65
(br s, 1H, NH), 6.94 (t, J¼6.9 Hz, 1H, H-10), 7.04 (m, 2H,
H-11, H-9), 7.23 (d, J¼7.6 Hz, 2H, H-400, 600), 7.33 (s, 5H,
H-80, H-90, H-100, H-110, H-120), 7.48 (d, J¼7.4 Hz, 2H,
H-300, H-700), 7.79 (br s, 1H, H-8); 13C NMR (75 MHz,
CDCl3) 298 K, d¼19.8 (C-130), 21.7 (C-80), 29.9 (C-12),
37.3 (C-11b), 44.6 (C-3), 49.5 (C-20), 49.7 (C-1), 52.8 (C-4),
53.2 (C-6), 56.5 (C-12a), 60.0 (C-6a), 67.1 (C-60), 116.8
(C-8), 123.2 (C-11), 124.1 (C-10), 127.6 (C-9), 127.8 (C-300,
700), 128.1 (C-100), 128.3 (C-80, 120), 128.7 (C-90, 110),
129.7 (C-400, 600), 132.7 (C-200), 134.7 (C-11a), 136.5 (C-70),
142.7 (C-7a), 143.7 (C-500), 155.9 (C-40), 171.8 (C-10). Mass
(FAB) m/z 589 (M++1). Anal. Calcd for C32H36N4O5S: C,
65.28; H, 6.16; N, 9.52. Found C, 65.25; H, 6.19; N, 9.55.
A.A., A.D.R., and R.C. are thankful to CSIR New Delhi for
providing fellowship. The authors are extremely grateful to
Dr. Rajesh K. Grover of The Scripps Research Institute, La
Jolla, California for his valuable suggestions and theoretical
MMFF calculations.
Supplementary data
Supplementary data associated with this article can be found
in the online version, at doi:10.1016/j.tet.2007.03.171.
References and notes
1. Stewart, W. E.; Siddall, T. H., III. Chem. Rev. 1970, 70, 517–
551.
4.2. Standard protocols used for determining biological
activities
2. (a) Rablen, P. R. J. Org. Chem. 2000, 65, 7930–7937; (b) Suarez-
Castillo, O. R.; Contreras-Martinez, Y. M. A.; Beiza-Granados,
L.; Melendez-Rodriguez, M.; Villagomez-Ibarra, J. R.; Torres-
Valencia, J. M.; Morales-Rios, M. S.; Joseph-Nathan, P.
Tetrahedron 2005, 61, 8809–8820; (c) Kam, T. S.; Yoganathan,
K.; Li, H.-Y. Tetrahedron Lett. 1996, 37, 8811–8814.
3. Scherer, G.; Kramer, M. L.; Schutkowski, M.; Reimer, U.;
Fischer, G. J. Am. Chem. Soc. 1998, 120, 5568–5574.
4. Lectka, T.; Cox, C. J. Org. Chem. 1998, 63, 2426–2427.
5. Deetz, M. J.; Forbes, C. C.; Jonas, M.; Malerich, J. P.; Smith,
B. D.; Wiest, O. J. Org. Chem. 2002, 67, 3949–3952.
6. Saxena, A. K.; Jain, P. C.; Anand, N.; Dua, P. R. J. Med. Chem.
1973, 16, 560–564.
7. (a) Kumar, N.; Dhoan, M. K.; Agarwal, S. K.; Saxena, A. K.;
Jain, P. C.; Anand, N. Eur. J. Med. Chem. 1982, 17, 312–316;
(b) Kumar, N.; Dhoan, M. K.; Agarwal, S. K.; Saxena, A. K.;
Jain, P. C.; Anand, N. Indian J. Chem. 1980, 19b, 1087–1088.
8. (a) Saxena, A. K.; Dhoan, M. K.; Ram, S.; Saxena, M.; Jain,
P. C.; Patnaik, G. K.; Anand, N. Indian J. Chem. 1983, 22b,
1224–1232; (b) Saxena, M.; Agarwal, S. K.; Patnaik, G. K.;
Saxena, A. K. J. Med. Chem. 1990, 33, 2970–2976.
9. (a) Jiang, W.; Alford, V. C.; Qiu, Y.; Bhattacharjee, S.; Mathew,
T. J.; Haynes-Jhonson, D.; Kraft, P. J.; Lundeen, S. G.; Sui, Z.
Bioorg. Med. Chem. 2004, 12, 1505–1515; (b) Daugan, A.;
Grondin, P.; Ruault, C.; Le Monnier de Gouville, A.-C.;
Coste, H.; Kirilovsky, J.; Hyafil, F.; Labaudiniere, R. J. Med.
Chem. 2003, 46, 4525–4532; (c) Daugan, A.; Grondin, P.;
Ruault, C.; Le Monnier de Gouville, A.-C.; Coste, H.;
Kirilovsky, J.; Hyafil, F.; Labaudiniere, R. J. Med. Chem.
2003, 46, 4533–4542.
10. Pamukcu, R.; Larry, A. P. U.S. Patent 6,046,199, April 4, 2000;
Chem. Abstr. 2000, 132, 257762.
11. Rao, J.; Chakrabarty, R.; Roy, R.; Mishra, A.; Saxena, A. K.
Arkivoc 2005, ii, 20–28.
12. (a) Nagrajan, K.; Nair, M. D.; Pillai, P. M. Tetrahedron 1967,
23, 1683–1690; (b) Anderson, J. E.; Lehn, J. M. Tetrahedron
1968, 24, 123–125.
13. Grover, R. K.; Maurya, R.; Roy, R. Tetrahedron 2004, 60,
2005–2010.
14. Ernst, R. R.; Bodenhausen, G.; Wokaun, A. Principle of
Magnetic Resonance in One and Two Dimensions; Clerendon:
Oxford, 1987; p 499.
15. G€unther, H. NMR Spectroscopy Basic Principles, Concepts,
and Applications in Chemistry; Wiley: New York, NY, 1995;
p 342.
4.2.1. Dopamine (D2) receptor binding. Brains were iso-
lated by decapitation of Sprague Dawley rats of either sex
with 200–250 g body weight. Caudate regions of the brain
were dissected out at 4 ꢀC. Immediately tissues were homo-
genized in 50 mM Tris buffer of pH 7.7. The homogenates
were centrifuged at 50,000ꢄg at 4 ꢀC. Pellets obtained
were washed twice in the above buffer followed by resus-
pending in Tris-salt buffer. Tissue suspension was incubated
at 37 ꢀC for 10 min followed by centrifugation at 50,000ꢄg
at 4 ꢀC. Final tissue pellets were resuspended in Tris-salt
buffer, aliquoted, and stored at ꢂ20 ꢀC until binding assays.
Ligand binding assays were carried out by incubating
300 mg membrane protein suspension with various concen-
trations of test drug at 37 ꢀC for 20 min in 1 mL incubation
mixture containing 50 mM ketanserin, 0.067 nM [3H]-
spiperone. After completion of the incubation, the reaction
was terminated by filtering the incubation mixture using
Whatman GF/C filters under vacuum. Dried filters were
collected and placed in scintillation vials to which was added
5 mL of scintillation cocktail. Non-specific binding was
determined using similar conditions in the presence of
100 nM haloperidol. Scintillation counts were determined
and data were analyzed by non-linear regression analysis
to determine Ki and IC50 of test drug using Graph pad Prism
software.
4.2.2. Serotonin 5-HT2A binding. Potency of test com-
pounds to bind to serotonin 5-HT2A receptors was deter-
mined using cortex membranes prepared by isolating brain
from Sprague Dawley rats of either sex of 200–250 g
body weight. Cortex regions were dissected out and homo-
genized in 50 mM Tris buffer, pH 7.7. Pellets were washed
twice and resuspended in the same buffer. Competitive bind-
ing assays were carried out by incubating 300 mg protein
with 0.39 nM [3H]-ketanserin at 37 ꢀC for 20 min in the
presence of various concentrations of test drug to a final
volume of 1 mL. Incubation mixtures were filtered onto
Whatman GF/C filters. Filters were collected followed by
addition of scintillation cocktail. Scintillation counts were
obtained and data were analyzed by non-linear regression
analysis to determine Ki and IC50 using Graph pad Prism
software.