Pharmacophore modeling and conformational analysis in the gas phase and in aqueous solution of regioisomeric melatonin analogs. A theoretical and experimental study

Article abstract of DOI:10.1016/j.molstruc.2016.12.040

In this work, 2-(N-acylaminoalkyl)indoles 1a–1d, that incorporate a pMeOBn group at the 3-position of the indole ring were virtual screened as potential melatoninergic ligands by analog-based design study using pharmacophore modeling. Pharmacophore models for melatoninergic agonist and antagonist activity were developed in order to identify the molecular constraints that define the geometric relationship among chemical features in each model. The best hypothesis consisted of six features for agonists and eight features for antagonists. The models suggest that the agonists and antagonists can share the same 3D arrangement for the six common pharmacophoric elements identified: two hydrogen bond acceptors (HBA), one hydrogen bond donor (HBD), one hydrophobic area (H), and two aromatic rings (AR). The extra hydrofobic interaction might be used as criterion for identified the pharmacological antagonist profile. Based on the pharmacophore fit, it was found that structures 1c and 1d show a good structural overlay that meets the requirements for the antagonistic pharmacophore hypothesis. Molecular modeling studies using the PCM solvation model predicted that the most stable conformers of 1a–1d match the antagonist pharmacophore hypothesis in contrast to those in the gas phase. Structures 1a–1c were synthesized only but the activities were not tested.

Full text of DOI:10.1016/j.molstruc.2016.12.040

Accepted Manuscript
Pharmacophore modeling and conformational analysis in the gas phase and in
aqueous solution of regioisomeric melatonin analogs. A theoretical and experimental
study
Humberto Mendoza-Figueroa, Gelacio Martínez-Gudiño, Jorge E. Villanueva-Luna,
Joel J. Trujillo-Serrato, Martha S. Morales-Ríos
PII:
S0022-2860(16)31353-9
10.1016/j.molstruc.2016.12.040
MOLSTR 23238
DOI:
Reference:
To appear in: Journal of Molecular Structure
Received Date: 12 August 2016
Revised Date: 8 December 2016
Accepted Date: 14 December 2016
Please cite this article as: H. Mendoza-Figueroa, G. Martínez-Gudiño, J.E. Villanueva-Luna, J.J. Trujillo-
Serrato, M.S. Morales-Ríos, Pharmacophore modeling and conformational analysis in the gas phase
and in aqueous solution of regioisomeric melatonin analogs. A theoretical and experimental study,
Journal of Molecular Structure (2017), doi: 10.1016/j.molstruc.2016.12.040.
This is a PDF file of an unedited manuscript that has been accepted for publication. As a service to
our customers we are providing this early version of the manuscript. The manuscript will undergo
copyediting, typesetting, and review of the resulting proof before it is published in its final form. Please
note that during the production process errors may be discovered which could affect the content, and all
legal disclaimers that apply to the journal pertain.

ACCEPTED MANUSCRIPT

ACCEPTED MANUSCRIPT
Pharmacophore modeling and conformational analysis in the gas phase
and in aqueous solution of regioisomeric melatonin analogs. A theoretical
and experimental study
Humberto Mendoza-Figueroaa, Gelacio Martínez-Gudiñoa, Jorge E. Villanueva-Lunaa, Joel
J. Trujillo-Serratob, Martha S. Morales-Ríosa,*
a
b
Departamento de Química and Programa de Posgrado en Farmacología, Centro de
Investigación y de Estudios Avanzados del Instituto Politécnico Nacional, Apartado 14-
740, Mexico City, 07000 Mexico.
Abstract
In this work, 2-(N-acylaminoalkyl)indoles 1a-1d, that incorporate a pMeOBn group at the
3-position of the indole ring were virtual screened as potential melatoninergic ligands by
analog-based design study using pharmacophore modeling. Pharmacophore models for
melatoninergic agonist and antagonist activity were developed in order to identify the
molecular constraints that define the geometric relationship among chemical features in
each model. The best hypothesis consisted of six features for agonists and eight features for
antagonists. The models suggest that the agonists and antagonists can share the same 3D
arrangement for the six common pharmacophoric elements identified: two hydrogen bond
acceptors (HBA), one hydrogen bond donor (HBD), one hydrophobic area (H), and two
aromatic rings (AR). The extra hydrofobic interaction might be used as criterion for
identified the pharmacological antagonist profile. Based on the pharmacophore fit, it was
found that structures 1c and 1d show a good structural overlay that meets the requirements
for the antagonistic pharmacophore hypothesis. Molecular modeling studies using the PCM
solvation model predicted that the most stable conformers of 1a-1d match the antagonist
pharmacophore hypothesis in contrast to those in the gas phase. Structures 1a-1c were
synthesized only but the activities were not tested.

ACCEPTED MANUSCRIPT
Keywords: 2-(N-acylaminoalkyl)indoles, synthesis, virtual screening, pharmacophore
modeling, gas-phase, continuum solvent
*Corresponding author. E-mail address: smorales@cinvestav.mx (M. S. Morales-Ríos)
1. Introduction
Melatonin (5-methoxy-N-acetyltryptamine), hormone primarily secreted by the
pineal gland, has wide functions in vertebrate organisms, with effects almost on all tissues
and organs [1-3]. Melatonin production and secretion are regulated by sunlight sensed by
the eyes and control circadian rhythms. This hormone has a complex mechanism of action.
It can cross the cell plasma membrane and exert its actions in all cells of the body. Most of
the regulatory functions exerted by this hormone are mediated by two high-affinity G-
protein-coupled receptors, named MT1 and MT2 [4], which have been recognized as
promising targets in the treatment of a number of human diseases and disorders. Melatonin
modulates neurogenesis, synaptic functions, neuronal cytoskeleton and gene expression [5].
Additionally, melatonin is a powerful endogenous antioxidant involved in the prevention of
the oxidative stress both through its direct free radical scavenging effect and by increasing
antioxidant activity [6-10]. Therefore, a lot of experimental and theoretical work has been
devoted to understanding the molecular mechanisms underlying such unique properties of
melatonin and of alternate molecules capable of mimicking its effects [11].
Previous studies on a great number of structurally different MLT receptor ligands, which
range from simple indole derivatives and their bioisosteres to phenylalkyl amides and
constrained melatoninergic agents [12], have demonstrated that suitably spaced alkoxyaryl
core and the amido side chain seem to be critical in determining the binding affinity and
biological activity of these melatoninergic ligands [13-17]. In this line, structure activity
relationship studies of regioisomeric melatonin-based analogues in which the MLT side
chain was shifted from the C-3 to the either the N-1 or to the C-2 indole position, allowed
to the development of indole MLT agonist and antagonist compounds, provided they keep
an appropriate spatial relationship of its pharmacophoric groups (i.e. the aromatic moiety,
the methoxy substituent and the amido group) [18].

ACCEPTED MANUSCRIPT
We now report a virtual screened of a series of 2-(N-acylaminoalkyl)indole
derivatives 1a-1d bearing a methoxyaryl group separated by an indole ring spacer from the
secondary amide, as potential melatoninergic ligands. Their structure design was based on
pharmacophore mapping study using available agonists and antagonists for melatonin
receptors. Because compounds 1a-1d have hydrogen bond donor and acceptor sites, their
conformational stability in the gas phase and in continuum solvent was also analyzed. The
most stable conformers in continuum solvent were spatial superposed with the
pharmacophore models and served to validate the obtained results. The synthesis of
compounds 1a-1c has been performed. The structure, atom numbering scheme, and
dihedral angle definitions of studied compounds 1a-1d are displayed in Fig. 1.
Fig. 1. (a) Structures of compounds 1a-1d and numbering. The indole nitrogen atom is
designated as N1 and the amide nitrogen as N2. Minimal structural requirements for the
biological activity of melatoninergic ligands are shown in bold. (b) Dihedral angle
definitions for 1a and 1b: τ1 = ∠(C2C3C8C1'); τ2 = ∠(C3C2C9N2); τ3 = ∠(C2C9N2C=O). (c)
Dihedral angle definitions for 1c and 1d: τ1 = ∠(C2C3C8C1'); τ2 = ∠(C3C2C9C10); τ3 =
∠(C2C9C10N2); τ4 = ∠(C9C10N2C=O).
2. Results and discussion
2.1. Pharmacophore modeling

ACCEPTED MANUSCRIPT
Pharmacophore mapping methodology based on 3D alignments of the
pharmacophore features of already known active molecules acting as agonists and
antagonists for melatonin receptors was explored in order to predict whether
indolylbenzylamides 1a-1d meet the requirements of the pharmacophore hypotheses and
therefore could function as potential melatoninergic ligands. In addition, the goal was to
estimate to what extent the pharmacophore maps could be valuable in dissecting important
features for the modeling of new melatoninergic ligands with agonist or antagonist profiles.
Pharmacophore features were identified based on representative training sets of 8 agonists
and 8 antagonists. The selection of these molecules was supported on their structural
diversity, affinity and potency (Supporting Information). Screens were carried out using the
LigandScout 3.12 [19] program for automatic generation of the hypotheses, which were
scored by setting root mean square deviation (RMSD) below 1.0, and using default
parameters.
The generated agonist and antagonist models were analyzed using several criteria:
(i) the number of pharmacophoric elements, (ii) the RMSD fit of corresponding
pharmacophoric elements, (iii) the degree of overlap of molecular volumes, and (iv) the
distances between pharmacophore features. The agonist top hypothesis was composed of
six essential features for activity: two hydrogen bond acceptors (HBA); one hydrogen bond
donor (HBD); one hydrophobic area (H); and two aromatic ring features (AR). In Fig. 2,
the pharmacophoric elements allow a certain spherical tolerance surrounding the ideal
position of a particular feature in 3D space. The results were consistent with a previously
developed melatonin receptor ligands model [20-24].
a
b
Fig. 2. (a) 6-Points pharmacophore model, key features include HBA (red), HBD (green),
H (yellow), and AR (blue), distance relation in Å. (b) Superposition of the 8 representative
agonist melatoninergic compounds on the 6-points pharmacophore model.

ACCEPTED MANUSCRIPT
The antagonist pharmacophore hypothesis is shown in Fig. 3. In this case, the best
quality option was composed of eight key features, six of which are in common with the
agonist model above described. The extra hydrophobic interaction identified as a ring
aromatic feature, might contribute to conferring antagonist activity. These results suggest
that agonists and antagonists can share the same arrangement for their mutual
pharmacophoric features in agreement with previously prediction [25].
a
b
Fig. 3. (a) 8-Points pharmacophore model, key features include HBA (red), HBD (green),
H (yellow), and AR (blue), distance relation in Å. (b) Superposition of the 8 representative
antagonist melatoninergic compounds on the 8-points pharmacophore model.
The 3D-pharmacophore models for melatoninergic agonist and antagonist activity
were then used for the prediction of melatoninergic activity of indolylbenzylamides 1a-1d.
The test set molecules were flexibly aligned with the chemical features of the models and
the inspection of the resulting structural superpositions showed the best-fit with the
antagonist pharmacophore hypothesis, (Supporting Information) largely because the
hydrophobic benzyl center has no comparable constituent in the agonist model. As shown
in Fig. 4, the antagonist model identified the following six mutual pharmacophoric
elements in 1a-1d: one HBA (the amide oxygen); one HBD (the amide hydrogen); two H
(the hydrophobic part of the indole ring and the benzyl group); and two AR (the benzene
part of the indole ring and the benzyl group). The results suggest that the hydrophobic
benzyl center led to compounds with antagonistic properties. This hypothesis obviously
needs to be supported by further experimental data. In addition, compounds 1c and 1d
showed superior structural overlay with the antagonist model compared to 1a and 1b in the

ACCEPTED MANUSCRIPT
series. Shortening the acetyl side chain by a methylene could prevent the proper alignment
of 1a and 1b.
b
a
Fig. 4. Alignment of indolylbenzylamides 1a-1d with the chemical features of (a) agonist
model (b) antagonist model.
2.2. Conformational Analysis
Analysis of the conformational stability of 1a-1d structures is important especially
when these molecules can be involved in attractive intramolecular N–H···O or NC–H···O
H-bond interactions. One of the main questions has always been whether intramolecular H-
bonds, which are supposed to exist in some gas-phase conformations, are maintained after
aqueous solvation. Therefore, taking into account that the aqueous phase results better
simulate the biological environment, the structural parameters of the lower energy 3D
structures of 1a-1d in the gas phase and aqueous solution were comparatively studied. All
computational optimizations were carried out by using the Gaussian 03 or 09 program
package [26,27]. The characterization of the minimum energy conformers in gas phase (Cg)
or aqueous solution (Cs) of 1a-1d (Gibbs energies among conformers, expressed in relative
terms, the respective mole fractions given by the Boltzmann populations, and relevant
dihedral angles) are presented in Table 1. The absolute values for the lowest Gibbs energy
conformer of every compound are given in Table 2, while the corresponding distances (d,
D) and angular parameters (θ, φ) used in the description of H-bonds are given in Table 3.
The schematic drawing of the most stable conformers of 1a-1d in the gas phase are given in
Fig. 5, while those in the presence of the implicit water solvent are given in Fig. 6.

ACCEPTED MANUSCRIPT
2.2.1. Gas Phase
Analysis of the distances (d, D) and angular parameters (θ, φ) (Table 3) suggests
that the conformational stability of 1a-1d in the gas phase is affected by attractive
intramolecular N–H···O or NC–H···O H-bond interactions. In the most stable two pair of
specular Cga1/Cga1′ and Cga2/Cga2′ conformations of 1a, carrying 91% of the total
population, the arrangements make possible the formation of an intramolecular N–H···O
hydrogen bond between the indole sp2-NH proton and the O atom of the carbonyl amide
group, closing a seven-membered ring, as assigned for the relatively short N–H···O < 2.2 Å
distances (Table 3). As a result, the amidomethyl side chain is fairly rigid. Therefore, the
main conformational freedom involves the rotation of the aromatic methoxy group which
was close to 0° or 180° owing to the resonance with the benzyl group (Fig. 5). While for
the homologue 1c, six most stable conformers were identified. The Cgc1 conformation and
the pair of specular Cgc2/Cgc2′ conformations, contributing over 90% of the total
population, are stabilized by an eight-membered intramolecular H-bond, as assigned for the
relatively short N–H···O < 2.1 Å distances (Table 3). The aforementioned hypothesized
cyclic H-bond formation delivers significant stability to the particular structure in the gas
phase. Concerning the minimum energy conformers found for the N-methyl derivatives 1b
(Cgb1 and Cgb2) and 1d (Cgd1/Cgd1′ specular pair) with highest abundance (a Boltzman
population density greater than 85%), show that they exhibited typical directional features
between the N-methyl group and the oxygen atom of the carbonyl amide group in
concordance with the formation of a carbon-oxygen N–CH···O hydrogen bonding. H-bond
(d) and carbon-oxygen distances (D) in the minimum energy conformations of 1b and 1d
are within the typical van der Waals distances: dC–H···O = 2.7 Å; D = 3.7 Å, frequently used
as cutoffs for C–H···O hydrogen bond identification [28-31] (Table 3).

ACCEPTED MANUSCRIPT
Fig. 5. Geometries for the minimum energy conformers in the gas phase found for 1a (Cga),
1b (Cgb), 1c, (Cgc) and 1d (Cgd). The stability increases from right to left in each row.
Specular conformations, with the same absolute values of dihedral angles (τ) but opposite
signs, were found for the minimum energy conformers of 1a (Cga1/Cga1′; Cga2/Cga2′), 1b
(Cgb3/Cgb3′), 1c (Cgc2/Cgc2′), and 1d (Cgd1/Cgd1′). Intramolecular H-bonding is indicated
as a red dashed line.
2.2.2. Aqueous Solution
Analysis of the dihedral angles (Table 1) as well as the corresponding lengths (d,
D), and angular preferences (θ, φ) of H-bond interactions (Table 3) provides evidence that
the structural parameters of the conformers of 1a in vacuo, namely Cga1/Cga1′and
Cga2/Cga2′, (Fig. 5) or in solution Csa1 and Csa2 (Fig. 6) are not much different. The
exception are the dihedral angles characterizing the orientation of the benzyl group. The
relatively short N–H···O < 2.3 Å distance (Table 3) in conformers Csa1 and Csa2, carrying
100% of the total population, suggested the presence of intramolecular H-bond. Therefore,
the intramolecular N–H···O H-bond supposed to exist in gas-phase remains after aqueous
solvation. In contrast, as can be seen in Tables 1 and 3, the solvent environment modify

ACCEPTED MANUSCRIPT
severely the geometries of the conformers of 1b-1d altering the dihedral angles involved in
the intramolecular H-bond interaction predicted in the corresponding gas phase structures.
As a result of these geometric changes, the stabilizing intramolecular H-bond force
important in the gas-phase is breaking in water to enable the formation of intermolecular H-
1
bonds with solvent molecules. It is of further interest to observe that in the H NMR
spectra, recorded in CDCl3, δN1H for 1a at 8.90 occurs 0.38 ppm high frequency from
δN1H of 1c at 8.52 (Table 4), consistent with the assumption that an amount of
intramolecular H-bonding is present in 1a [32]. The change of solvent from CDCl3 to
DMSO-d6 moves the N1H indole and N2H amide resonance proton signals at higher-
frequency (Table 4), indicating the formation of intermolecular H-bonding between the NH
groups and solvent DMSO.
Fig. 6. Geometries for the minimum energy conformers in the aqueous solution found for
1a (Csa), 1b (Csb), 1c, (Csc) and 1d (Csd). The stability increases from right to left in each
row.

ACCEPTED MANUSCRIPT
Table 1
Relative Gibbs energies in the gas phase (Ggas,rel) and in aqueous solution (Gsoln,rel), molar fractions (χ, %), and selected dihedral angles (τ,
°) for the minimum energy conformers of 1a-1d.a
gas phase
aqueous solution
b
c
d
d
d
d
b
c
d
d
d
d
Ggas,rel
χgas
τ1
τ2
τ3
τ4
Gsoln,rel
χsoln
τ1
τ2
τ3
τ4
1a
1a
Csa1
Csa2
-
-
1b
Cga1
Cga1′
Cga2
Cga2′
1b
0.00 28.36
0.00 28.36 -55.3 108.5
0.16 21.64 68.5 -108.4 -77.0
0.16 21.64 -68.6 108.4 77.0
55.2 -108.5 -77.2
-
-
-
-
0.00 74.57 -111.2 115.4
0.64 25.43 -82.3 112.7
79.6
78.8
-
-
-
-
-
-
77.2
-
-
-
-
-
-
-
-
Cgb1
Cgb2
Cgb3
Cgb3′
1c
0.00 46.68
0.08 40.59
68.5 -92.0 -110.4
73.3 -90.9 -109.1
-
-
-
-
Csb1
Csb2
-
-
1c
0.00 58.35
0.20 41.65
-78.7 105.1 164.4
-81.2 105.6 159.4
-
-
-
-
1.18
1.18
6.36 -110.3 -104.8 -120.6
6.36 110.3 104.8 120.6
-
-
-
-
-
-
-
-
-
-
Cgc1
Cgc2
Cgc2′
Cgc3
Cgc4
Cgc5
1d
0.00 42.59 -75.6 158.7 -60.1
0.35 23.47 80.9 -119.2 -124.2
0.35 23.47 -80.8 119.2 124.1
107.1
74.8
-74.7
-61.7
Csc1
Csc2
Csc3
Csc4
Csc5
Csc6
1d
0.00 42.17
0.32 18.68
0.44 15.26
0.59 11.85
0.62 11.26
0.64 10.89
112.3
-82.8
92.1
95.5
61.8
64.7
88.9
89.0
-84.8 120.0 -61.6 178.3
-90.2 125.3 -62.6 -91.8
-85.2 121.0 -61.6 137.1
-86.5 119.9 -62.1 -176.8
1.17
1.69
1.79
5.95
2.45 -102.2
2.07
86.5 -75.9 -45.3
99.2 -71.3 -127.5
91.4 -112.7
64.2
158.5
Cgd1
Cgd1′
Cgd2
Cgd3
Cgd4
Cgd5
0.00 43.00
80.5 -97.5 -171.8
78.4
-78.7
Csd1
Csd2
Csd3
Csd4
-
0.00 36.67 -112.0
85.0
-89.1 113.5 -60.6 173.7
-85.5 88.1 63.1 89.4
-93.0 112.0 -63.2 -174.3
61.4
88.6
0.01 42.46 -80.6
97.2 171.6
81.0 -112.8
4.15 -89.2 73.8
3.61 -84.6 112.2 -47.0
0.27 23.25
0.31 21.73
0.41 18.35
1.35
1.39
1.47
1.71
4.39
46.9 -115.0
50.5
76.7
114.2
-
-
-
-
-
-
-
-
-
-
-
-
2.39 -96.3 106.7 -65.9 -154.3
-
a All energy values are in kcal/mol. Vales relative the most stable conformer. Boltzmann populations relative to the lowest Gibbs
energy conformer. d Dihedral definitions for 1a-1d see Fig. 1. The absolute Gibbs energy values for the lowest energy conformers are
given in Table 2.
b
c

ACCEPTED MANUSCRIPT
Table 2
Absolute values (hartrees) of Ggas and Gsol for the conformers of 1a-1d presenting the
lowest Gibbs energy (Cga1-Cgd1 and Csa1-Csd1).
compound
Ggas
Gsoln
-996.112861 -989.872149
-1035.420662 -1028.896868
-1035.429171 -1028.910397
-1074.738495 -1067.935904
1a
1b
1c
1d

ACCEPTED MANUSCRIPT
Table 3
Distances (d and D in Å) and angular parameters (θ and φ in °) for the minimum energy
conformers of 1a-1d.a
gas phase
aqueous solution
d
D
θ
φ
d
D
θ
φ
1a
1a
Cga1 2.17 3.32 132.51 109.69
Cga1′ 2.17 3.32 132.50 109.68
Cga2 2.18 3.33 131.72 110.22
Cga2′ 2.18 3.33 131.70 110.21
1b
Csa1 2.27 3.30 132.05 105.84
Csa2 2.29 3.34 130.91 106.64
-
-
-
-
-
-
-
-
-
-
1b
b
b
Cgb1 2.36 3.31 164.01 103.12
Cgb2 2.35 3.79 164.10 104.90
Csb1 3.68 4.25
Csb2 3.54 4.18
-
-
-
-
-
b
b
b
b
Cgb3 2.47 3.79
Cgb3′ 2.47 3.48
1c
-
-
-
-
b
b
-
1c
b
b
b
b
b
b
b
b
b
b
b
b
Cgc1 2.01 3.31 150.69 105.23
Cgc2 2.02 3.79 152.22 133.43
Cgc2′ 2.02 3.79 152.15 133.40
Csc1 4.05 4.02
Csc2 4.03 4.21
Csc3 5.00 4.71
Csc4 5.71 4.89
Csc5 4.03 4.21
Csc6 5.12 4.77
1d
Csd1 3.19 3.74
Csd2 4.69 5.07
Csd3 3.25 3.73
Csd4 5.12 5.28
-
-
-
-
-
-
-
-
-
-
-
-
b
b
b
b
b
b
Cgc3 2.95 3.48
Cgc4 6.18 5.46
Cgc5 5.10 5.62
1d
-
-
-
-
-
-
b
b
b
b
b
b
b
b
Cgd1 2.62 3.71 177.23 126.56
Cgd1′ 2.62 3.71 178.97 126.68
-
-
-
-
-
-
-
-
b
b
b
b
b
b
b
b
Cgd2 2.64 3.69
Cgd3 2.58 3.49
Cgd3′ 2.58 3.48
Cgd4 5.57 5.99
-
-
-
-
-
-
-
-
-
-
-
-
-
-
-
-
-
-
a According to geometric criteria, the hydrogen bonds in the generic molecule X
−
H···O (X
= C, N) are determined by the following parameters: d = 1.5
−3.2 Å, D = 2.5−4.0 Å, θ =
90 180°, φ = 100 260° [28-31].
−
−
b Electron lone pair direction of the acceptor carbonyl oxygen is deviated too severely from
the hydrogen donor atom.

ACCEPTED MANUSCRIPT
Table 4
Selected 1H spectral data (δ in ppm) for 1a-1c in CDCl3 and DMSO-d6 solutions.a
compond
solvent
N1-H
8.90
N1-Me N2-H Meamide
CDCl3
-
5.75
8.25
5.07
8.20
5.40
7.94
1.93
1.85
1.82
1.82
1.77
1.76
1a
DMSO-d6
CDCl3
10.69
-
-
3.70
3.69
-
1b
1c
DMSO-d6
CDCl3
-
8.52
10.80
DMSO-d6
-
a The indole nitrogen atom is designated as N1 and the amide nitrogen as N2.
2.2.3. Mapping of the solvated minimum energy conformers of 1a-1d
The benefit of complementing pharmacophore mapping results with the information
obtained about the predicted conformational preference for 1a-1d in aqueous solution is to
determine how well statistically significant conformers characterize the antagonist
pharmacophore model. In order to check whether a chemical feature is present in a given
conformer all sets of conformers of 1a-1d were aligned with the agonist and antagonist
hypotheses by a rigid fit procedure. An inspection of results showed that, in general, the
two most stable conformers of 1a-1d, namely Csa1, Csa2, Csb1, Csb2, Csc1, Csc2, and Csd1,
together with Csd3, mapped reasonably well the antagonist hypothesis. Except for Csa2,
Csc1 and Csd1, the pharmacophore maps on all these conformers on two hydrogen-bond
acceptor sites, one hydrogen-bond donor site and two aromatic hydrophobic regions,
whereas these conformers either do not map at all or map fewer of the features of the
agonist model (Fig. 7). The fit scores are presented in Table S4 in the Supplementary.

ACCEPTED MANUSCRIPT
b
a
Fig. 7. Alignment of the lower energy conformers of 1a-1d with the chemical features of
(a) agonist model (b) antagonist model.
2.3. Synthesis of 2-(N-acylaminoalkyl)indole derivatives
In sharp contrast with the number of 3-(N-acylaminoethyl)indoles identified as
melatoninergic ligands, those examples related with their regioisomeres in which the 2-
aminoethyl appendage is attached at the C2 indole core are very scarce [33-35]. The
synthesis of 2-(N-acylaminoalkyl)indoles 1a-1d was started with the study of the
electrophilic benzylation of electron deficient indole 2 (Scheme 1) [36]. Thus, the
benzylation of 2 with pMeOBnCl conducted under various conditions, from 10 to 70 °C
and for 45 min up to 90 h reaction time, showed that the best result was observed when the
reaction was performed with 2.25 equiv of pMeOBnCl in glacial acetic acid at 60 °C for
1.5 h. Under these conditions the pMeOBn group was introduced at the C3 and C5-position
of the indole nucleus to afford the mono- and di-benzylated compounds 3a and 3b in 2:1
ratio and in a global yield of 62%. In every experiment, we found that 2 was never fully
consumed and in no case was N-substitution observed [37]. It is worthy to mention that the
ease of displacement of the halide depended strongly on the electrophilicity; under the
conditions analyzed no reaction occurred with BnCl. Single crystals of 3a and 3b were
obtained, the X-ray data showed that the carbonyl group and the hydrogen atom at N-1 are
syn oriented with respect to one another, and the dihedral angle between the carbonyl group
and the N(1)-C(2) bond is about 0° in both compounds favoring conjugative interactions
(Fig. 8).

ACCEPTED MANUSCRIPT
R2
R1
CO2Me
RCl
CO2Me
N
H
N
H
2
1
2
3a
3b
: R = pMeOBn, R = H
1
2
: R = R = pMeOBn
Scheme 1. Preparation of 3-substituted- and 3,5-disubstituted 2-carbometoxyindoles 3a and
3b. Reagents and conditions: (a) p-MeOBnCl, glacial AcOH, room temperature, 1.5 h.
a
b
Fig. 8. X-ray structures of 3a (a) and 3b (b); ellipsoids drawn at the 30% probability level.
The multistep transformation of 3a to 1a-1c (Scheme 2) was conducted following a
slightly modified literature procedure [38]. Briefly, ester 3a was saponified and the
carboxilic acid 4 was converted to the carboxamide 5 by reaction with oxalyl chloride and
then with ammonium hydroxide in 88% combined yield. Reduction of amide 5 with LiAlH4
afforded the corresponding unstable amine 6. The 2-(N-acylaminomethyl)indole derivative
1a was prepared by N-acylation of 6 with acetic anhydride in 62% global yield from 5.
Treatment of 1a with KOH in DMSO, followed by MeI provided the N1,N2-dimethylated
product 7. Selective N1-methylation of 1a was carried out with MeI in the presence of a
weaker base, Cs2CO3, to give 1b in 62% isolated yield. Alternatively, the carboxilic acid 4
was converted to the N,N-diethylcarboxamide 8 by reaction with dicyclohexylcarbodiimide
(DCC) and then with diethylamine, albeit in only in 20% yield. Reduction of amide 8 with
LiAlH4 gave the N,N-diethylmethanamine 9 in 49% yield. Reaction of the ammonium salt
of 9 with potassium cyanide provided the nitrile 10 in 36% yield. In solution 10 was

ACCEPTED MANUSCRIPT
unstable, decomposition occurs noticeably for 1 day, whereas upon standing neat at room
temperature 10 crystallized as reddish needles. Structures of compounds 8 and 10 from
single crystal X-ray diffraction studies are shown in Fig. 9. Treatment of 10 with MeI using
Cs2CO3 as a base furnished the N-methylated product 11a in mixture with the C-
monomethylated and C,N-dimethylated products 11b and 11c in 5:2:4 ratio, respectively,
and in a global yield of 55%. Finally, the 2-(N-acylaminoethyl)indole derivative 1c was
prepared by hydrogenation of 10 over Raney nickel and concomitant N-acylation with
acetic anhydride in 65% yield of the crude product. It was observed that compound 1c was
spontaneously oxidized in air. Indeed, mass spectrometric and NMR analyses revealed that
upon storage, neat 1c degraded in the presence of oxygen, allowing cleanly an entry to the
2,2-disubstituted indoline-3-one derivative 12 (Scheme 3). Under these circumstances, the
N-methyl derivative 1d could not be obtained from unstable 1c.

ACCEPTED MANUSCRIPT
Scheme 2. Synthesis of 2-(N-acylaminoalkyl)indoles 1a-1d. Reagents and conditions: (a)
20% aq NaOH, MeOH, THF, reflux, 6 h; (b) oxalyl chloride, CH2Cl2, DMF, rt, 4.5 h; (c)
NH4OH, rt, 18 h; (d) LiAlH4, THF, reflux, 6 h; (e) (Ac)2O, NEt3, THF, rt, 6 h; (f) Cs2CO3,
DMF, MeI, rt, 24 h; (g) KOH, DMSO, MeI, reflux, 4 h; (h) DMPA, DCC, CHCl3, Et2NH,
reflux, 24 h; (i) MeI, MeOH, rt, 2 h, then 40% aq KCN, reflux, 1.5 h (j) H2, Raney-Ni W2,
Ac2O, 3 atm, rt, 5 h.

ACCEPTED MANUSCRIPT
a
b
Fig. 9. X-ray structures of 8 (a) and 10 (b); ellipsoids drawn at the 30% probability level.
Scheme 3. Formation of 2,2-disubstituted indoline-3-one derivative 12.
3. Conclusions
Pharmacophoric models for melatoninergic agonist and antagonist ligands were
generated and used these models for the virtual screening of 1a-1d compounds. Analysis of
the resulting structural alignment showed the best-fit with the antagonist eight-point
pharmacophore model, largely because the hydrophobic benzyl center has no comparable
constituent in the agonist model. Compounds 1c and 1d exhibited superior structural
overlay with the antagonist model over its lower homologues 1a and 1b. In addition,
theoretical conformational analysis in the gas phase at the B3LYP/6-31G+(d,p) level
hypothesized intramolecular H-bonding as essential factor in determining stable
conformations for 1a-1d. These results diverge with those in the aqueous phase where

ACCEPTED MANUSCRIPT
conformational changes do not permit such H-bond. In addition, statistically significant
conformers calculated at the same level using PCM solvation model matched the antagonist
pharmacophore mapping. Although this initial study suggests that indoles 1a-1d bearing an
amidoalkyl side chain at C-2 and a p-methoxybenzyl substructure at C-3 are expected to
show activity as antagonist melatoninergic ligands, their relative stability, which decreases
in the order 1b > 1a > 1c, discourages biological studies.
4. Experimental section
4.1 Computational Part
Automated pharmacophoric studies were carried out using the LigandScout 3.12
program available from Inte:Ligand GmbH (http://www.inteligand.com/). The three-
dimensional structure of the molecules 1a-1d was built using ChemBioOffice 12.0 [39]. All
computational optimizations were carried out by using the Gaussian 03 or 09 program
package [26,27]. A Monte Carlo random conformational search with the MMFF94 force-
field minimization [40] was applied for every compound to give 14-16 conformers in the
first 3 kcal/mol. The calculations were carried out using the Spartan 04 molecular modeling
software program [41,42]. Geometry optimizations for gas-phase conformers were
performed at the DFT level by applying the B3LYP functional [43] and 6-31G+(d,p) basis
set [44]. A frequency analysis was carried out in each case to ascertain that a real minimum
had been obtained (i.e., no imaginary vibrational frequencies). To simulate physiological
conditions, the conformational search was performed in the presence of an aqueous
environment using the free energy perturbation MMFFaq method implemented in Monte
Carlo simulations. The calculations were carried out using the Spartan 14 software [45].
Geometry optimizations were performed at the DFT level by applying the B3LYP
functional and 6-31G+(d,p) basis set using Tomasi’s polarizable continuum model (PCM)
[46-48] and the dielectric constant of water at 298.15 K (ε = 78.4).

ACCEPTED MANUSCRIPT
4.2 Chemistry
All reagent-grade chemicals were purchased from Sigma-Aldrich Co. and were used
as received. Melting points were determined on an electrothermal instrument and are
uncorrected. IR spectra were obtained in chloroform. Low-resolution mass spectra were
conducted on a Varian CP 3800 GC spectrometer equipped with a selective mass Varian
Saturn 2000 detector. MS analyses were obtained in the electron impact (EI) mode at an
ionizing voltage of 70 eV. High-resolution mass spectra (HRMS) were recorded on an
Agilent LCTOF spectrometer at the UCR Mass Spectrometry Facility, University of
California, Riverside, CA and on a Waters Synapt G2 HDMS spectrometer at the Central
Analytical Laboratory, Department of Chemistry and Biochemistry, University of Colorado
at Boulder. 1H and 13C NMR spectra were recorded on Varian Mercury spectrometers (300
1
MHz for H and 75 MHz for 13C) using CDCl3 as a solvent unless indicated otherwise.
Chemical shifts are given in ppm (δ) and are referenced to TMS as internal standard. J
values are quoted in Hertz, with the normal abbreviations (s, singlet; d, doublet; t, triplet; q,
quartet; br, broad; m, multiplet). All structural assignments were supported by gHMBC,
gHSQC, and NOESY. Analytical thin-layer chromatography (TLC) was performed on
silica gel 60 F254 coated aluminum sheets (0.25 mm thickness) with a fluorescent
indicator. Visualization was accomplished with UV light (254 nm). Flash chromatography
was performed on silica gel 60 (230-400 mesh).
4.2.1. Methyl 3-(4-methoxybenzyl)-1H-indole-2-carboxylate (3a) and methyl 3,5-bis(4-
methoxybenzyl)-1H-indole-2-carboxylate (3b). A glacial acetic acid suspension (16.5 M, 48
mL) of commercial available methyl 1H-indole-2-carboxylate (2) (2.98 g, 17.01 mmoL)
was stirred at 60 °C until solution, then brought up to room temperature and p-MeOBnCl
(2.25 mL, 16.5 mM) was added. After stirring at room temperature for 1.5 h, the reaction
was diluted with water until cloudy (15 mL). The aqueous layer was extracted with EtOAc
(3 x 30 mL), and the combined organic layers were washed with saturated aqueous
NaHCO3 and brine, dried over Na2SO4, and concentrated in vacuo to give a mixture of
1
mono- and di-benzylated compounds 3a and 3b in 2:1 ratio, as determined by H NMR.
The mixture was separated by flash chromatography (hexane/EtOAc 4:1) to afford

ACCEPTED MANUSCRIPT
successively 3a (1.94 g, 7.0 mmol, 41% yield) and 3b (0.99 g, 2.38 mmol, 21%).
Recrystallization of 3a or 3b from hexane/EtOAc gave colorless crystals suitable for X-ray
crystallography. For 3a: Mp: 158-159 °C. Rf = 0.60 (hexane/EtOAc 7:3); 1H NMR (CDCl3)
δ: 8.75 (br s, 1H, NH), 7.61 (dm, J = 8.2 Hz, 1H, H-4), 7.37 (dm, J = 8.2 Hz, 1H, H-7),
7.31 (ddd, J = 8.3, 6.7, 1.2 Hz, 1H, H-6), 7.19 (dm, J = 8.8 Hz, 2H, Ho), 7.10 (ddd, J = 8.2,
6.7, 1.2 Hz, 1H, H-5), 6.77 (dm, J = 8.8 Hz, 2H, Hm), 4.44 (s, 2H, CH2) 3.94 (s, 3H,
CO2Me), 3.75 (s, 3H, OMe); 13C NMR (CDCl3) δ: 162.8 (C=O), 157.7 (Cp), 136.1 (C-7a),
133.1 (Ci), 129.3 (Co), 127.9 (C-3a), 125.6 (C-6), 123.3 (C-3), 123.2 (C-2), 121.2 (C-4),
120.2 (C-5), 113.7 (Cm), 111.7 (C-7), 55.1 (OMe), 51.7 (CO2Me), 29.6 (CH2); IR (CHCl3,
cm-1) νmax = 3460, 3344, 1697; GC-MS (EI, 70 eV): m/z (%) = 295 (M+, 87), 264 (100);
HRMS: m/z calcd for C18H17NO3 (M + H) 296.1282, found 296.1282. CCDC-1469876.
For 3b: Mp: 164-166 °C Rf = 0.45 (hexane/EtOAc 7:3); 1H NMR (CDCl3) δ: 8.75 (br s, 1H,
NH), 7.39 (m, 1H, H-4), 7.25 (dd, J = 8.5, 0.7 Hz, 1H, H-7), 7.16 (dm, J = 8.8 Hz, 2H, Ho),
7.12 (d, J = 1.8 Hz, 1H, H-6), 7.08 (dm, J = 8.8 Hz, 2H, Ho'), 6.81 (dm, J = 8.8 Hz, 2H,
Hm'), 6.76 (dm, J = 8.8 Hz, 2H, Hm'), 4.39 (s, 2H, CH2), 3.96 (s, 2H, CH2'), 3.91 (s, 3H,
13
CO2Me), 3.77 (s, 3H, OMe'), 3.75 (s, 3H, OMe); C NMR (CDCl3) δ: 162.7 (CO), 157.9
(Cp'), 157.7 (Cp), 134.7 (C-7a), 133.8 (Ci'), 133.5 (C-5), 133.1 (Ci), 129.8 (Co'), 129.3
(Co), 128.1. (C-3a), 127.3 (C-6), 123.3 (C-3), 123.2 (C-2), 120.6 (C-4), 113.8 (Cm'), 113.7
(Cm), 111.8 (C-7), 55.2 (OMe'), 55.2(OMe), 51.7 (CO2Me), 41.1 (CH2'), 29.6 (CH2); IR
(CHCl3, cm-1) νmax = 3455, 3351, 1695; GC-MS (EI, 70 eV): m/z (%) = 295 (56), 281 (19),
264(100), 221(30); HRMS: m/z calcd for C26H25NO4 (M + Na) 438.1681, found 438.1673.
CCDC-1469877.
4.2.2. 3-(4-Methoxybenzyl)-1H-indole-2-carboxylic acid (4). To a stirred and cooled (0 °C)
suspension of 3a (0.3 g, 1.02 mmol) in MeOH (6 mL) was added 20% aqueous NaOH (1.2
mL). The ice bath was removed, and the reaction was heated at reflux for 6 h. The reaction
mixture was cooled (0 °C) and 1 M HCl was added with stirring to adjust the pH to about
3-4. The solids were filtered off, washed with water and dried at 50 °C overnight to give 4
as a white solid (0.27 g, 0.97 mmol, 95%). Mp: 242-243 °C. Rf = 0.18 (hexane/EtOAc 7:3).
1H NMR (AcOD-d4) δ: 7.61 (br d, J = 8.2 Hz, 1H, H-4), 7.44 (br d, J = 8.2 Hz, 1H, H-7),
7.28 (ddd, J = 8.2, 7.0, 1.2 Hz , H-6), 7.19 (dm, J = 8.8 Hz, 2H, Ho), 7.05 (ddd, J = 8.2, 7.0,

ACCEPTED MANUSCRIPT
1.2 Hz, 1H, H-5), 6.77 (dm, J = 8.8 Hz, 2H, Hm), 4.46 (s, 2H, CH2) 3.71 (s, 3H, OMe); 13C
NMR (AcOD-d4) δ: 167.8 (CO), 158.8 (Cp), 138.3 (C-7a), 134.2 (Ci), 130.2 (Co), 128.7
(C-3a), 126.8 (C-6), 125.9 (C-3), 123.4 (C-2), 122.1 (C-4), 121.5 (C-5), 114.6 (Cm), 113.2
(C-7), 55.4 (OMe), 30.3 (CH2); IR (CHCl3, cm-1) νmax = 3080, 1712; GC-MS (EI, 70 eV):
m/z (%) = 121 (100). HRMS: m/z calcd for C17H15NO3 (M - H) 280.0979, found 280.0987.
4.2.3. 3-(4-Methoxybenzyl)-1H-indole-2-carboxamide (5). To a suspension of 4 (273 mg,
0.97 mmol) in CH2Cl2 (6 mL) and DMF (0.02 mL) was added dropwise oxalyl chloride
(0.2 mL, 2.35 mmol) and the mixture was allowed to stir for 4.5 h at room temperature. The
reaction mixture was cooled (0 °C) and concentrated aqueous NH4OH (1.05 mL) was
added by portions. The ice bath was removed, and the reaction was stirred at room
temperature for 18 h. The solids were filtered off, washed with water, dissolved in EtOAc,
dried over Na2SO4, and concentrated in vacuo. The resulting yellowish solid was purified
by crystallization (CH2Cl2/hexane) to give 5 (250 mg, 0.89 mmol, 93%) as colorless
needles. Mp: 155-156 °C (CH2Cl2/hexane) (lit. [49] 150-152 °C). Rf = 0.37 (hexane/EtOAc
3:7); 1H NMR (CDCl3) δ: 9.55 (br s, 1H, NH), 7.65 (dm, J = 8.1, Hz, 1H, H-4), 7.45 (dm, J
= 8.2 Hz, 1H, H-7), 7.32 (ddd, J = 8.2, 6.9, 1.2 Hz, 1H, H-6), 7.15 (ddd, J = 8.1, 6.9, 1.0
Hz, 1H, H-5), 7.11 (dm, J = 8.8 Hz, 2H, Ho), 6.82 (dm, J = 8.8 Hz, 2H, Hm), 5.87 (br s,
2H, NH2), 4.36 (br s, 2H, CH2), 3.75 (s, 3H, OMe); 13C NMR (CDCl3) δ: 163.5 (CO), 158.5
(Cp), 135.4 (C-7a), 131.0 (Ci), 129.0 (Co), 128.9 (C-3a), 127.2 (C-2), 125.2 (C-6), 120.4
(C-5), 120.3 (C-4), 116.2 (C-3), 114.5 (Cm), 111.8 (C-7), 55.3 (OMe), 29.3 (CH2); IR
(CHCl3, cm-1) νmax = 3493, 3445, 3384, 1659; GC-MS (EI, 70 eV): m/z (%) = 280 (M+, 8),
263 (100), 249 (18), 220 (26). HRMS m/z calcd for C17H16N2O2 (M + H) 281.1285, found
281.1291.
4.2.4. (3-(4-Methoxybenzyl)-1H-indol-2-yl)methanamine (6). In a dry flask, under Ar, a
solution of 5 (424 mg, 1.51 mmol) in anhydrous THF (10 mL) was added dropwise to a
suspension of LiAlH4 (424 mg, 11.2 mmol) in anhydrous THF (10 mL). The mixture was
stirred at reflux during 20 h. The reaction was cooled (0 °C), quenched with water (8 mL)
and filtered over Celite. Following concentration, the residue was partitioned between
CH2Cl2 (30 mL) and water (15 mL). The aqueous layer was extracted with CH2Cl2 (2 x 15

ACCEPTED MANUSCRIPT
mL), and the combined organic layers were washed with brine, dried over Na2SO4, and
concentrated in vacuo. The reddish oil residue was purified by flash chromatography
(CH2Cl2/MeOH 9:1) to give 6 (321 mg, 1.21 mmol, 80%) as an unstable yellowish oil,
experiencing decomposition even after short NMR acquisition times. Rf = 0.16
(CH2Cl2/MeOH 9:1); 1H NMR (CDCl3) δ: 8.63 (br s, 1H, NH), 7.42 (br d, J = 7.6 Hz, 1H,
H-4), 7.32 (br d, J = 7.6 Hz, 1H, H-7), 7.14 (br dd, J = 7.0, 7.0 Hz, 1H, H-6), 7.10 (dm, J =
8.8 Hz, 2H, Ho), 7.04 (br dd, J = 7.6, 7.6 Hz, 1H, H-5), 6.76 (dm, J = 8.8 Hz, 2H, Hm),
4.01 (s, 2H, CH2Ph), 3.98 (s, 2H, CH2N) 3.74 (s, 3H, OMe), 1.94 (br s, 2H, NH2); 13C
NMR (CDCl3) δ: 157.7 (Cp), 135.3 (C-7a), 134.7 (C-2), 133.6 (Ci), 129.0 (Co), 128.6 (C-
3a), 121.6 (C-6), 119.2 (C-5), 118.7 (C-4), 113.7 (Cm), 110.7 (C-7 and C-3), 55.2 (OMe),
37.1 (CH2N), 28.9 (CH2Ph); IR (CHCl3, cm-1) νmax = 3456, 3363; GC-MS (EI, 70 eV): m/z
(%) = 266 (M+, 1), 248 (100), 233 (24), 218 (36). HRMS: m/z calcd for C17H18N2O (M +
Na) 289.1317, found 289.1317.
4.2.5. N-((3-(4-Methoxybenzyl)-1H-indol-2-yl)methyl)acetamide (1a). To a solution of
amine 6 (0.152g, 0.57 mmol) in a mixture of THF (3 mL) and triethylamine (0.1 mL, 0.8
mmol) was added acetic anhydride (0.075mL, 0.8 mmol). The mixture was stirred at room
temperature for 6 h until disappearance of starting material (TLC). The resulting solution
was concentrated in vacuo and the residue was taken up in EtOAc (25 mL). The organic
layer was washed with saturated aqueous NaHCO3 and brine, dried over Na2SO4, and
concentrated in vacuo. The oil residue was purified by flash chromatography
(hexane/EtOAc 3:2) to give 1a (135 mg, 0.44 mmol, 77%) as an unstable yellowish oil. Rf
1
= 0.22 (37 (hexane/EtOAc 3:7); H NMR (CDCl3) δ: 8.90 (br s, 1H, NH), 7.46 (br d, J =
7.6 Hz, 1H, H-4), 7.31 (br d, J = 8.2 Hz, 1H, H-7), 7.16 (ddd, J = 7.7, 7.7,1.2 Hz, 1H, H-6),
7.10 (dm, J = 8.8 Hz, 2H, Ho), 7.05 (ddd J = 7.7, 7.7, 1.2 Hz, 1H, H-5), 6.80 (dm, J = 8.8
Hz, 2H, Hm), 5.75 (br t, J = 5.9 Hz, 1H, NHCO), 4.39 (d, J = 5.9 Hz, 2H, CH2N), 4.05 (br
13
s, 2H, CH2Ph), 3.75 (s, 3H, OMe), 1.93 (s, 3H, COMe); C NMR (CDCl3) δ: 171.5 (CO),
157.8 (Cp), 135.4 (C-7a), 133.7 (Ci), 132.7 (C-2), 129.1 (Co), 127.8 (C-3a), 122.1 (C-6),
119.3 (C-5), 118.9 (C-4), 113.8 (Cm), 111.7 (C-3), 111.0 (C-7), 55.2 (OMe), 35.1
(CH2NH), 28.9 (CH2Ph), 23.0 (COMe); 1H NMR (DMSO-d6) δ: 10.69 (br s, 1H, NH), 8.25
(br t, J = 5.3 Hz, 1H, NHCO), 7.30 (br d, J = 7.6 Hz, 1H, H-4), 7.28 (br d, J = 8.2 Hz, 1H,

ACCEPTED MANUSCRIPT
H-7), 7.12 (dm, J = 8.8 Hz, 2H, Ho), 6.98 (ddd, J = 8.2, 7.0, 1.2 Hz, 1H, H-6), 6.85 (ddd J
= 7.6, 7.0, 1.2 Hz, 1H, H-5), 6.76 (dm, J = 8.8 Hz, 2H, Hm), 4.41 (d, J = 5.8 Hz, 2H,
13
CH2N), 3.94 (s, 2H, CH2Ph), 3.68 (s, 3H, OMe), 1.85 (s, 3H, COMe); C NMR (DMSO-
d6) δ: 169.2 (CO), 157.2 (Cp), 135.4 (C-7a), 133.7 (Ci), 132.7 (C-2), 129.1 (Co), 127.6 (C-
3a), 120.7 (C-6), 118.4 (C-5), 118.3 (C-4), 113.5 (Cm), 111.0 (C-7), 110.7 (C-3), 54.9
(OMe), 34.1 (CH2N), 28.5 (CH2Ph), 22.5 (COMe); IR (CHCl3, cm-1) νmax = 3441, 1692,
1665; GC-MS (EI, 70 eV): m/z (%) = 308 (M+, 25), 248 (72), 237 (100), 219 (23).
4.2.6. N-((3-(4-methoxybenzyl)-1-methyl-1H-indol-2-yl)methyl)acetamide (1b). To a stirred
suspension of amide 1a (120 mg, 0.37 mmol) and Cs2CO3 (192 mg, 0.58 mmol) in DMF (2
mL) was added MeI (0.1 mL, 1.8 mmol). After stirring at room temperature for 72 h, the
reaction was quenched with water (20 mL). The aqueous layer was extracted with EtOAc
(3 x 30 mL), and the combined organic layers were washed with brine (20 mL), dried over
Na2SO4, and concentrated in vacuo. The oil residue was purified by flash chromatography
(hexane/EtOAc 1:1 to give 1b (60 mg, 0.19 mmol, 62%) as colorless solid. Mp: 170-171
1
°C. Rf = 0.22 (hexane/EtOAc 3:7); H NMR (CDCl3) δ: 7.54 (br d, J = 7.6 Hz, 1H, H-4),
7.28 (br d, J = 8.2 Hz, 1H, H-7), 7.22 (ddd, J = 8.2, 7.0, 1.2 Hz, 1H, H-6), 7.11 (dm, J = 8.8
Hz, 2H, Ho), 7.10 (ddd J = 7.6, 7.0, 1.2 Hz, 1H, H-5), 6.80 (dm, J = 8.8 Hz, 2H, Hm), 5.07
(br d, J = 5.3 Hz, 1H, NH), 4.57 (d, J = 5.3 Hz, 2H, CH2N), 4.08 (br s, 2H, CH2Ph), 3.75 (s,
3H, OMe), 3.70 (s, 3H, NMe), 1.82(s, 3H, COMe); 13C NMR (CDCl3) δ: 169.3 (CO), 157.8
(Cp), 136.8 (C-7a), 134.0 (Ci), 132.5 (C-2), 129.1 (Co), 127.4 (C-3a), 122.1 (C-6), 119.3
(C-5), 118.9 (C-4), 113.9 (Cm), 113.5 (C-3), 109.2 (C-7), 55.2 (OMe), 33.2 (CH2N), 29.6
1
(NMe), 29.1 (CH2Ph), 23.0 (COMe); H NMR (DMSO-d6) δ: 8.20 (br t, J = 5.1 Hz, 1H,
NHCO), 7.37 (dm, J = 8.3 Hz, 1H, H-7), 7.35 (dm, J = 7.6 Hz, 1H, H-4), 7.18 (dm, J = 8.7
Hz, 2H, Ho), 7.09 (ddd, J = 8.2, 7.0, 1.3 Hz, 1H, H-6), 6.92 (ddd, J = 8.0, 7.0, 1.0 Hz, 1H,
H-5), 6.77 (dm, J = 8.8 Hz, 2H, Hm), 4.48 (d, J = 5.1 Hz, CH2N), 4.02 (br s, 2H, CH2Ph),
3.69 (s, 3H, NMe) 3.66 (s, 3H, OMe) 1.82 (s, 3H, COMe); 13C NMR (DMSO-d6) δ: 168.4
(CO), 157.2 (Cp), 136.6 (C-7a), 133.7 (Ci), 133.6 (C-2), 129.1 (Co), 126.8 (C-3a), 121.2
(C-6), 118.7 (C-4), 118.6 (C-5), 113.5 (Cm), 112.1 (C-3), 109.3 (C-7), 54.9 (OMe), 32.2
(CH2N), 29.6 (NMe), 28.8 (CH2Ph), 22.4 (COMe); IR (CHCl3, cm-1) νmax = 3434, 1666;

ACCEPTED MANUSCRIPT
GC-MS (EI, 70 eV): m/z (%) = 322 (M+, 5), 262 (55), 251 (100), 233 (18). HRMS: m/z
calcd for C20H22N2O2 (M + Na) 345.1579, found 345.1580.
4.2.7. N-((3-(4-Methoxybenzyl)-1-methyl-1H-indol-2-yl)methyl)-N-methylacetamide (7). To
a solution of amide 1a (120 mg, 0.37 mmol) in DMSO (2 mL) was added KOH (90 mg, 1.6
mmol). The mixture was stirred at room temperature for 1 h, and MeI (0.1 mL, 1.8 mmol)
was added. After reacting for further for 4 h in the same conditions, the reaction was
quenched with water (20 mL). The aqueous layer was extracted with EtOAc (3 x 30 mL),
and the combined organic layers were washed with brine (20 mL), dried over Na2SO4, and
concentrated in vacuo. The oil residue was purified by flash chromatography
(hexane/EtOAc 1:1) to give 7 (60 mg, 0.18 mmol, 46%) as yellowish solid. Mp: 89-90 °C.
Rf = 0.22 (hexane/EtOAc 3:7); 1H NMR (CDCl3) δ: 7.46 (br d, J = 7.6 Hz, 1H, H-4), 7.30
(br d, J = 8.2 Hz, 1H, H-7), 7.24 (ddd, J = 8.2, 6.4, 1.2 Hz, 1H, H-6), 7.09 (dm, J = 8.8 Hz,
2H, Ho), 7.07 (ddd J = 7.6, 6.4, 1.2 Hz, 1H, H-5), 6.75 (dm, J =8.8 Hz, 2H, Hm), 4.83 (br s,
2H, CH2N), 4.12 (br s, 2H, CH2Ph), 3.75 (s, 3H, OMe), 3.66 (s, 3H, NMe), 2.62 (s, 3H,
NMeCO), 2.08 (s, 3H, COMe); 13C NMR (CDCl3) δ: 170.4 (CO), 157.7 (Cp), 137.2 (C-7a),
133.3 (Ci), 131.4 (C-2), 129.0 (Co), 127.5 (C-3a), 121.9 (C-6), 119.2 (C-5), 119.1 (C-4),
114.2 (C-3), 113.7 (Cm), 109.0 (C-7), 55.2 (OMe), 39.0 (CH2N), 33.7 (NMeCO), 29.6
(NMe), 29.1 (CH2Ph), 21.9 (COMe); 1H NMR (DMSO-d6) δ: 7.35 (br d, overlapped, 2H,
H-4, H-7), 7.11 (ddd, J = 8.2, 6.5,1.2 Hz, 1H, H-6), 7.08 (dm, J = 8.8 Hz, 2H, Ho), 6.94 (dd
J = 7.5, 6.6 Hz, 1H, H-5), 6.76 (dm, J =8.8 Hz, 2H, Hm), 4.78 (br s, 2H, CH2N), 4.04 (br s,
2H, CH2Ph), 3.65 (s, 3H, OMe), 3.60 (s, 3H, NMe), 2.62 (s, 3H, NMeCO), 2.01(s, 3H,
13
COMe); C NMR (DMSO-d6) δ: 169.7 (CO), 157.2 (Cp), 136.8 (C-7a), 133.5 (Ci), 131.9
(C-2), 129.0 (Co), 126.9 (C-3a), 121.4 (C-6), 118.8 (C-4), 118.7 (C-5), 113.6 (C-3), 113.5
(Cm), 109.4 (C-7), 54.9 (OMe), 38.2 (CH2N), 33.5 (NMeCO), 30.0 (NMe), 28.4 (CH2Ph),
21.7 (COMe); IR (CHCl3, cm-1) νmax = 1629; GC-MS (EI, 70 eV): m/z (%) = 336 (M+, 0.3),
263 (95), 251 (100), 233 (25). HRMS: m/z calcd for C21H24N2O2 (M + Na) 359.1736, found
359.1737.
4.2.8. N,N-Diethyl-3-(4-methoxybenzyl)-1H-indole-2-carboxamide (
8). To a solution of 4
(1.0 g, 3.6 mmol) in CHCl3 (60 mL) was added successively 4-(dimethylamino)pyridine

ACCEPTED MANUSCRIPT
(DMAP) (549 mg, 4.5 mmol) and dicyclohexylcarbodiimide (DCC) (618 mg, 3.0 mmol) at
room temperature, and Et2NH (300 µL) was then added dropwise. After stirring at reflux
for 24 h, the reaction was quenched with brine (30 mL). The organic layer was separated,
and the aqueous layer was extracted with CHCl3 (2 x 10 mL). The combined organic
extracts were dried over Na2SO4, and concentrated in vacuo. The residue was suspended in
cool THF (30 mL), filtered, and the filtrate concentrated in vacuo. The oil residue was
purified by flash chromatography (hexane/EtOAc 9:1 to 1:1 gradient) to give 8 (243 mg,
1
0.72 mmol, 20%) as colorless solid. Mp: 174-175 °C; Rf = 0.53 (hexane/EtOAc 1:1); H
NMR (CDCl3) δ: 8.30 (br s, 1H, NH), 7.40 (dm, J = 8.1 Hz, 1H, H-4), 7.34 (dm, J = 8.1
Hz, 1H, H-7), 7.19 (ddd, J = 8.2, 7.0, 1.3 Hz, 1H, H-6), 7.15 (dm, J = 8.9 Hz, 2H, Ho), 7.03
(ddd, J = 8.1, 7.0, 1.2 Hz, 1H, H-5), 6.77 (dm, J = 8.8 Hz, 2H, Hm), 4.09 (s, 2H, CH2Ph),
13
3.75 (s, 3H, OMe), 3.45 (q, J = 7.1 Hz, 4H, 2 NCH2), 1.12 (t, J = 7.1 Hz, 6H, 2 Me); C
NMR (CDCl3) δ: 165.2 (C=O), 157.8 (Cp), 135.8 (C-7a), 132.4 (Ci), 129.3 (Co), 128.9 (C-
2), 127.5 (C-3a), 123.2 (C-6), 120.2 (C-4), 119.8 (C-5), 114.2 (C-3), 113.7 (Cm), 111.5 (C-
7), 55.2 (OMe), 41.0 (NCH2), 29.9 (CH2Ph), 13.6 (Me); IR (CHCl3, cm-1) νmax = 3458,
3262, 1611; GC-MS (EI, 70 eV): m/z (%) = 336 (M+, 3), 263 (100). HRMS: m/z calcd for
C21H24N2O2 (M + H) 337.1911, found 337.1926. CCDC-1469878.
4.2.9. N-Ethyl-N-((3-(4-methoxybenzyl)-1H-indol-2-yl)methyl)ethanamine (9). In a dry
flask, under Ar, a solution of 8 (300 mg, 0.9 mmol) in anhydrous THF (6 mL) was added
dropwise to a suspension of LiAlH4 (84 mg, 2.2 mmol) in anhydrous THF (2 mL), and the
mixture was stirred at reflux during 4 h. After cooled (0 °C), the reaction was quenched
with water (8 mL) and filtered over Celite. Following concentration, the residue was
partitioned between CH2Cl2 (30 mL) and water (15 mL). The aqueous layer was extracted
with EtOAc (2 x 15 mL), and the combined organic layers were washed with brine, dried
over Na2SO4, and concentrated in vacuo. The oil residue was purified by crystallization
(hexane/Et2O) to give 9 (140 mg, 0.44 mmol, 49%) as pale yellow crystals. Mp: 113-116
°C; Rf = 0.47 (hexane/EtOAc 1:1); 1H NMR (CDCl3) δ: 8.54 (br s, 1H, NH), 7.39 (dm, J =
7.9 Hz, 1H, H-4), 7.30 (dm, J = 7.9 Hz, 1H, H-7), 7.12 (ddd, J = 8.1, 7.0, 1.3 Hz, 1H, H-6),
7.10 (dm, J = 8.8 Hz, 2H, Ho), 7.00 (ddd, J = 7.8, 7.0, 1.2 Hz, H-5), 6.76 (dm, J = 8.7 Hz,
2H, Hm), 4.03 (s, 2H, CH2Ph), 3.73(s, 3H, OMe), 3.67 (s, 2H, CH2N), 2.51 (q, J = 7.2 Hz,

ACCEPTED MANUSCRIPT
4H, 2 CH2CH3), 1.00 (t, J = 7.2 Hz, 6H, 2 CH2CH3); 13C NMR (CDCl3) δ: 157.6 (Cp),
135.3 (C-7a), 133.9 (C-2), 133.7 (Ci), 129.1 (Co), 128.9 (C3a), 121.2 (C-6), 118.9 (C-5),
118.6 (C-4), 113.6 (Cm), 111.3 (C-3), 110.5 (C-7), 55.2 (OMe), 48.9 (CH2N), 47.0
(CH2CH3), 28.9 (CH2Ph), 11.7 (CH2CH3); IR (CHCl3, cm-1) νmax = 3447, 3349; GC-MS
(EI, 70 eV): m/z (%) = 323 (M+, 1), 249 (100), 219 (42).
4.2.10. 2-(3-(4-Methoxybenzyl)-1H-indol-2-yl)acetonitrile (10). To a stirred and cooled (0
°C) solution of 9 (300 mg, 0.93 mmol) in MeOH (1.1 mL) was added MeI (145 µL, 2.3
mmol). The reaction was stirred at room temperature for 2 h, then 40% aqueous KCN (450
µL, 2.9 mmol) was added, and the mixture was refluxed for 1.5 h. The mixture was
extracted with EtOAc (3 x 10 mL), and the combined organic layers were washed with
brine, dried over Na2SO4, and concentrated in vacuo. The residue was purified by flash
chromatography (hexane/EtOAc 7:3) to give 10 as an oil, which upon standing neat at room
temperature crystallized as reddish needles (93 mg, 0.33 mmol, 36%). Mp: 120-122 °C; Rf
= 0.38 (hexane/EtOAc 7:3); 1H NMR (CDCl3) δ: 8.17 (br s, 1H, NH), 7.45 (dm, J = 7.6 Hz,
1H, H-4), 7.33 (dm, J = 8.2 Hz, 1H, H-7), 7.20 (ddd, J = 8.2, 7.0, 1.2 Hz, 1H, H-6), 7.09
(dm, J = 8.8 Hz, 2H, Ho), 7.09 (overlapped, 1H, H-5); 6.79 (dm, J = 8.8 Hz, 2H, Hm), 4.01
(s, 2H, CH2Ph), 3.74 (s, 3H, OMe), 3.70 (s, 2H, CH2CN); 13C NMR (CDCl3) δ: 158.0 (Cp),
135.7 (C-7a), 132.0 (Ci), 129.1 (Co), 128.2 (C-3a), 122.8 (C-6), 122.1 (C-2), 120.1 (C-5),
119.1 (C-4), 116.4 (CN), 114.0 (Cm), 113.6 (C-3), 110.9 (C-7), 55.2 (OMe), 29.0 (CH2Ph),
15.6 (CH2CN); IR (CHCl3, cm-1) νmax = 3456, 3408, 2258; GC-MS (EI, 70 eV): m/z (%) =
276 (M+, 100), 237 (15); HRMS: m/z calcd for C18H16N2O (M - H) 275.1179, found
275.1182. CCDC-1469879.
4.2.11. 2-(3-(4-Methoxybenzyl)-1-methyl-1H-indol-2-yl)acetonitrile
(11a), 2-(3-(4-
methoxybenzyl)-1H-indol-2-yl)propanenitrile (11b) and 2-(3-(4-methoxybenzyl)-1-methyl-
1H-indol-2-yl)propanenitrile (11c). To a stirred suspension of nitrile 10 (120 mg, 0.44
mmol) and Cs2CO3 (225 mg, 1.4 mmol) in DMF (3 mL) was added MeI (120 µL, 1.9
mmol). After stirring at room temperature for 24 h, the reaction was quenched with water
(20 mL). The aqueous layer was extracted with EtOAc (3 x 15 mL), and the combined
organic layers were washed with brine (10 mL), dried over Na2SO4, and concentrated in

ACCEPTED MANUSCRIPT
vacuo to give a mixture of mono- and dimethylated compounds 11a, 11b, and 11c in ca.
5:2:4 ratio, respectively as determined by CG/MS. The mixture was separated by flash
chromatography using hexane/CH2Cl2/EtOAc solvent mixture. Elution with 50:48:2 solvent
mixture ratio gave successively 11c (27 mg, 0.09 mmol, 20% yield), 11a (30 mg, 0.1
mmol, 25% yield), and 11b (12 mg, 0.04 mmol, 10% yield) as yellow oils. Further elution
with 50:40:10 solvent mixture ratio yielded the starting material 10 (18 mg). For 11a: Rf =
0.30 (hexane/CH2Cl2/EtOAc 50:48:2); 1H NMR (CDCl3) δ: 7.47 (dm, J = 7.6 Hz, 1H, H-4),
7.32 (dm, J = 8.2 Hz, 1H, H-7), 7.26 (overlapped, 1H, H-6), 7.11 (dm, J = 8.8 Hz, 2H, Ho),
7.09 (overlapped, 1H, H-5); 6.80 (dm, J = 8.8 Hz, 2H, Hm), 4.08 (s, 2H, CH2Ph), 3.81 (s,
3H, NMe), 3.79 (s, 2H, CH2CN), 3.76 (s, 3H, OMe); 13C NMR (CDCl3) δ: 158.0 (Cp),
137.2 (C-7a), 132.3 (Ci), 129.1 (Co), 127.3 (C-3a), 124.2 (C-2), 122.6 (C-6), 119.7 (C-5),
119.3 (C-4), 115.9 (CN), 114.0 (Cm), 113.4 (C-3), 109.1 (C-7), 55.2 (OMe), 30.0 (NMe),
29.3 (CH2Ph), 14.0 (CH2CN); IR (CHCl3, cm-1) νmax = 2255, 1827; GC-MS (EI, 70 eV):
m/z (%) = 290 (M+, 100), 250 (9), 182 (43). HRMS: m/z calcd for C19H18N2O (M + Na)
1
313.1317, found 313.1320. For 11b: Rf = 0.22 (hexane/CH2Cl2/EtOAc 50:48:2); H NMR
(CDCl3) δ: 8.12 (br s, 1H, NH), 7.43 (dm, J = 7.9 Hz, 1H, H-4), 7.37 (dm, J = 8.1 Hz, 1H,
H-7), 7.22 (ddd, J = 8.2, 7.0, 1.2 Hz, 1H, H-6), 7.09 (dm, J = 8.8 Hz, 2H, Ho), 7.09
(overlapped, 1H, H-5); 6.80 (dm, J = 8.8 Hz, 2H, Hm), 4.12 (q, J = 7.3 Hz, 1H, CHCN),
4.05 (AB, J = 16.4 Hz, 2H, CH2Ph), 3.77 (s, 3H, OMe), 1.60 (d, J = 7.3 Hz, 3H, CHMe);
13C NMR (CDCl3) δ: 158.0 (Cp), 135.6 (C-7a), 132.1 (Ci), 129.1 (Co), 128.8 (C-2), 128.3
(C-3a), 122.8 (C-6), 120.2 (CN), 120.1 (C-5), 119.3 (C-4), 113.9 (Cm), 112.3 (C-3), 111.0
(C-7), 55.2 (OMe), 28.9 (CH2Ph), 23.5 (CHCN), 19.9 (CHMe); IR (CHCl3, cm-1) νmax
=
3450, 2245, 1727; GC-MS (EI, 70 eV): m/z (%) = 290 (M+, 100), 236 (68). HRMS: m/z
calcd for C19H18N2O (M + Na) 313.1317, found 313.1319. For 11c: Rf = 0.43
(hexane/CH2Cl2/EtOAc 50:48:2); 1H NMR (CDCl3) δ: 7.48 (dm, J = 7.9 Hz, 1H, H-4), 7.32
(dm, J = 8.1 Hz, 1H, H-7), 7.26 (ddd, J = 8.2, 7.0, 1.2 Hz, 1H, H-6), 7.09 (overlapped, 1H,
H-5);7.07 (dm, J = 8.8 Hz, 2H, Ho), 6.78 (dm, J = 8.8 Hz, 2H, Hm), 4.28 (q, J = 7.3 Hz,
1H, CHCN), 4.11 (AB, J = 16.9 Hz, 2H, CH2Ph), 3.89 (s, 3H, NMe), 3.76 (s, 3H, OMe),
1.54 (d, J = 7.3 Hz, 3H, CHMe); 13C NMR (CDCl3) δ: 157.9 (Cp), 137.4 (C-7a), 132.7 (Ci),
130.0 (C-2), 129.0 (Co), 127.3 (C-3a), 122.5 (C-6), 120.0 (CN), 119.8 (C-5), 119.3 (C-4),
113.9 (Cm), 112.3 (C-3), 109.1 (C-7), 55.2 (OMe), 30.7 (NMe), 29.1 (CH2Ph), 22.1