Higher-Affinity Agonists of 5-HT1AR Discovered through Tuning the Binding-Site Flexibility

Article abstract of DOI:10.1021/acs.jcim.5b00164

Discovery of high-affinity and high-selectivity agonists of 5-HT_1AR has become very attractive due to their potential therapeutic effects on multiple 5-HT_1AR-related psychological and neurological problems. On the basis of our previously designed lead compound FW01 (K_i = 51.9 nM, denoted as 9a in the present study), we performed large-scale molecular dynamics simulations and molecular docking operations on 5-HT_1AR-9a binding. We found the flip-packing events for the headgroup of 9a, and we also found that its tail group could bind flexibly at the agonist-binding site of 5-HT_1AR. By finely tuning the flip-packing phenomenon of the 9a headgroup and tuning the binding flexibility of 9a tail group, we virtually designed a series of new 9a derivatives through molecular docking operations and first-principles calculations and predicted that these newly designed 9a derivatives should be higher-affinity agonists of 5-HT_1AR. The computational predictions on the new 9a derivatives have been confirmed by our wet-experimental studies as chemical synthesis, binding affinity assays, and agonistic-function assays. The consistency between our computational design and wet-experimental measurements has led to our discovery of higher-affinity agonists of 5-HT_1AR, with ～50-fold increase in receptor-binding affinity and ～25-fold improvements in agonistic function. In addition, our newly designed 5-HT_1AR agonists showed very high selectivity of 5-HT_1AR over subtype 5-HT_2AR and also over three subtypes of dopamine receptors (D₁, D₂, and D₃). (Graph Presented).

Full text of DOI:10.1021/acs.jcim.5b00164

Subscriber access provided by NEW YORK UNIV
Article
Higher-Affinity Agonists of 5-HT1AR Discovered
through Tuning the Binding-Site Flexibility
Peng Lian, Linlang Li, Chuanrong Geng, Xuechu Zhen, and Wei Fu
J. Chem. Inf. Model., Just Accepted Manuscript • DOI: 10.1021/acs.jcim.5b00164 • Publication Date (Web): 01 Jul 2015
Downloaded from http://pubs.acs.org on July 5, 2015
Just Accepted
“Just Accepted” manuscripts have been peer-reviewed and accepted for publication. They are posted
online prior to technical editing, formatting for publication and author proofing. The American Chemical
Society provides “Just Accepted” as a free service to the research community to expedite the
dissemination of scientific material as soon as possible after acceptance. “Just Accepted” manuscripts
appear in full in PDF format accompanied by an HTML abstract. “Just Accepted” manuscripts have been
fully peer reviewed, but should not be considered the official version of record. They are accessible to all
readers and citable by the Digital Object Identifier (DOI®). “Just Accepted” is an optional service offered
to authors. Therefore, the “Just Accepted” Web site may not include all articles that will be published
in the journal. After a manuscript is technically edited and formatted, it will be removed from the “Just
Accepted” Web site and published as an ASAP article. Note that technical editing may introduce minor
changes to the manuscript text and/or graphics which could affect content, and all legal disclaimers
and ethical guidelines that apply to the journal pertain. ACS cannot be held responsible for errors
or consequences arising from the use of information contained in these “Just Accepted” manuscripts.
Journal of Chemical Information and Modeling is published by the American Chemical
Society. 1155 Sixteenth Street N.W., Washington, DC 20036
Published by American Chemical Society. Copyright © American Chemical Society.
However, no copyright claim is made to original U.S. Government works, or works
produced by employees of any Commonwealth realm Crown government in the course
of their duties.

Page 1 of 36
Journal of Chemical Information and Modeling
1
2
3
4
5
HigherꢀAffinity Agonists of 5ꢀHT1AR Discovered through
6
7
8
9
Tuning the BindingꢀSite Flexibility
Peng Liana, LinLang Lib, Chuanrong Genga, Xuechu Zhenb*, Wei Fua*
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
29
30
31
32
33
34
35
36
37
38
39
40
41
42
43
44
45
46
47
48
49
50
51
52
53
54
55
56
57
58
59
60
a Department of Medicinal Chemistry & Key Laboratory of Smart Drug Delivery, Ministry of Education,
School of Pharmacy, Fudan University, Shanghai 201203, China
b Jiangsu Key laboratory for Translational Research for Neuropsychiatric-diseases, Department of
Pharmacology, College of Pharmaceutical Sciences, Soochow University, Suzhou 215123, China
Wei Fu, Prof./Ph.D.
Department of Medicinal Chemistry & Key Laboratory of Smart Drug Delivery, Ministry of Education
& PLA, School of Pharmacy, Fudan University
826 Zhangheng Road, Shanghai 201203, P.R. China
Fax: +86ꢀ21ꢀ51980010, Phone: +86ꢀ21ꢀ51980010
Eꢀmail: wfu@fudan.edu.cn
Xuechu Zhen Prof./Ph.D.
Department of Pharmacology, College of pharmaceutical Sciences
Soochow University
Suzhou 215123, P. R. China.
Email: zhenxuechu@suda.edu.cn
ACS Paragon Plus Environment

Journal of Chemical Information and Modeling
Page 2 of 36
1
2
3
4
ABSTRACT
5
6
7
8
Discovery of highꢀaffinity and highꢀselectivity agonists of 5ꢀHT1AR has become very
attractive due to their potential therapeutic effects on multiple 5ꢀHT1ARꢀrelated
psychological and neurological problems. Based on our previously designed lead
compound FW01 (Ki = 51.9 nM, denoted as 9a in the present study), we performed
largeꢀscale molecular dynamics simulations and molecular docking operations on
5ꢀHT1ARꢀ9a binding. We found out the flipꢀpacking events for the head group of 9a, and
we also found that its tail group could bind flexibly at the agonistꢀbinding site of 5ꢀHT1AR.
By finely tuning the flipꢀpacking phenomenon of 9a head group, and tuning the binding
flexibility of 9a tail group, we virtually designed a series of new 9a derivatives through
molecular docking operations and first principles calculations, and predicted that these
newly designed 9a derivatives should be higherꢀaffinity agonists of 5ꢀHT1AR. The
computational predictions on the new 9a derivatives have been confirmed by our
wetꢀexperimental studies as chemical synthesis, binding affinity assays and
agonisticꢀfunction assays. The consistency between our computational design and
wetꢀexperimental measurements has lead to our discovery of pretty higherꢀaffinity agonists
of 5ꢀHT1AR, with ~50ꢀfold increase of receptorꢀbinding affinity and ~25ꢀfold
improvements of agonistic function. In addition, our newly designed 5ꢀHT1AR agonists
showed very high selectivity of 5ꢀHT1AR over subtype 5ꢀHT2AR, and also over three
subtypes of dopamine receptors (D1, D2 and D3).
9
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
29
30
31
32
33
34
35
36
37
38
39
40
41
42
43
44
45
46
47
48
49
50
51
52
53
54
55
56
57
58
59
60
ACS Paragon Plus Environment

Page 3 of 36
Journal of Chemical Information and Modeling
1
2
3
4
INTRODUCTION
5
6
7
8
Serotonin (5ꢀhydroxytryptamine, 5ꢀHT), as a specific endogenous neurotransmitter,
plays an important role in a plethora of human physiological and pathological processes in
the central nervous system (CNS), such as mood, sleep, cognition, memory and chronic
pain alleviation.1 5ꢀHT exerts its effects via 5ꢀHT receptors, which can be classified into
seven subfamilies, 5ꢀHT1 to 5ꢀHT7.2ꢀ7 Among them, 5ꢀHT1AR has drawn particular
9
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
29
30
31
32
33
34
35
36
37
38
39
40
41
42
43
44
45
46
47
48
49
50
51
52
53
54
55
56
57
58
59
60
attention because of its important therapeutic potentials for anxiety8ꢀ11, depression3, 12, 13
,
cognition impairment11, 13 and other psychiatric syndrome.3, 14 The 5ꢀHT1AR agonists have
also been found as effective treatments for neurological disorders (e.g. Parkinson’s
disease)2, 14, 15 and for extrapyramidal side effects (EPS).10, 16ꢀ18 Recent findings showed
that the 5ꢀHT1AR agonists could be the potential treatments for ischemic stroke.4, 9, 19
Medicines like buspirone, gepirone, tandospirone and vilazodone that act as 5ꢀHT1AR
1, 2, 20
partial agonists have been used for years as anxiolytics or antidepressants.
The
reported new 5ꢀHT1AR ligands, Fꢀ15599, which showed potent antidepressant activity, and
F13640 as a potential treatment of chronic pain, have entered clinical trials.21ꢀ23 Giving the
large needs of 5ꢀHT1AR agonists as potential treatments for various health problems, it is
very interesting to design novel chemical agents with higher binding affinities and better
pharmacological profiles.
In our previous studies,24, 25 we have put great efforts to design novel 5ꢀHT1AR
agonists through the combination of computational approaches (e.g. dynamic
pharmacophoreꢀbased virtual screening) and experimental tests (e.g. [35S]GTPγS binding
assay). We obtained a series of lead compounds as 5ꢀHT1AR agonists, which have novel
scaffolds and high binding affinities. The previous best compound, FW01 (denoted as 9a
in Scheme 1), showed a pretty high affinity for 5ꢀHT1ARꢀbinding (i.e. the Ki value of 51.9
nM, and EC50 value of 160.80 nM). Our previous results24 of docking compound 9a into
the agonistꢀbinding site of 5ꢀHT1AR, revealed that 9a adopted an extended conformation
and was hydrogen bonded with residue D3.32 (Figure S1 in Supporting Information). The
ACS Paragon Plus Environment

Journal of Chemical Information and Modeling
Page 4 of 36
1
2
3
4
5
6
7
8
9
head group (i.e. indole ring) of 9a was located at subꢀpocket 1 (denoted as SP1 in Figure S1,
and here after) around residues S5.42, F6.51 and F6.52 of 5ꢀHT1AR, and the tail group
packed with residues Y2.64, F3.28 and W7.40 at subꢀpocket 2 (denoted as SP2 in Figure
S1, and here after) of 5ꢀHT1AR. Such mode of 5ꢀHT1ARꢀ9a binding is generally consistent
with the observed binding modes of other GPCRꢀligand complexes and receptorꢀactivation
as reported in the Xꢀray crystal structures.2, 26ꢀ28 Structurally, the head group of 9a is
connected to the central piperazine group through rotatable bonds, and the similar
intraꢀmolecular connection for the tail group of 9a (Scheme 1). At the SP1 of 5ꢀHT1AR, the
head group of 9a could be hydrogenꢀbonded with the side chain of residue S5.42 through
either its –NH or –F atom. Based on these observed features, we wonder that how flexible
of the head group of 9a could be when packing with the surrounding residues of 5ꢀHT1AR.
Similarly, we also wonder whether or not the packing between the tail group of 9a and SP2
of 5ꢀHT1AR can be tuned to the best possibility. Tuning of these local interꢀmolecular
interactions, if further explored, can very possibly provide new clues of structural
modifications of 9a, which will definitely help us design higherꢀaffinity agonists of
5ꢀHT1AR.
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
29
30
31
32
33
34
35
36
37
38
39
40
41
42
43
44
45
46
47
48
49
50
51
52
53
54
55
56
57
58
59
60
Scheme 1
In the present study, we tested the binding flexibility of compound 9a at the
agonistꢀbinding site of 5ꢀHT1AR by largeꢀscale molecular dynamics (MD) simulations and
conformational analysis on the 5ꢀHT1ARꢀ9a binding structure. Based on the newly
observed flipꢀpacking phenomena between structural parts (i.e. head group and tail group)
of 9a and the subꢀpockets (SP1 and SP2) of 5ꢀHT1AR, we designed a set of new derivatives
of 9a by substituting its head group and tail group, and by predicting their binding free
energies with 5ꢀHT1AR. These newly designed 9a derivatives were then synthesized, and
tested for binding and functional assays. Through the tuning of 5ꢀHT1ARꢀligand binding
flexibility, and our wetꢀexperimental studies, we discovered a new series of higherꢀaffinity
ACS Paragon Plus Environment

Page 5 of 36
Journal of Chemical Information and Modeling
1
2
3
4
5
6
7
8
agonists of 5ꢀHT1AR, with ~50ꢀfold increase of binding affinity (the lowest Ki value of
1.01±0.1 nM) and ~25ꢀfold improvement of receptorꢀagonizing function (the smallest
EC50 value is 6.53 nM).
9
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
29
30
31
32
33
34
35
36
37
38
39
40
41
42
43
44
45
46
47
48
49
50
51
52
53
54
55
56
57
58
59
60
MATERIALS AND METHODS
Molecular Dynamics Simulations. In order to test how flexible of compound 9a at
the agonistꢀbinding site of 5ꢀHT1AR, we performed MD simulations on the 5ꢀHT1ARꢀ9a
binding structure at the environment of POPC lipid bilayer and the surrounding water
molecules, which mimics the physiological condition of the receptor. The binding structure
of 5ꢀHT1ARꢀ9a complex was derived directly from our previous studies on the modeling of
5ꢀHT1AR and its binding with FW01, and the simulation system was set up in a similar way
as that described in our previous study.29 In brief, the 5ꢀHT1ARꢀ9a complex was inserted
into a POPC lipid bilayer by using the inflated GRO method.30 Meanwhile, the area per
lipid was maintained to ~75 Å. Then, the system was solvated into a SPC water box.31 Na+
and Clꢀ ions were randomly placed to neutralize and maintain the system ionic
concentration of 154 mmol/L. In total, the system contains 37610 atoms, including 9a,
5ꢀHT1AR, 121 POPC lipids, 9442 SPC water molecules, 26 Na+ and 38 Clꢀ ions. The MD
simulations were performed using GROMACS 4.5.1 package.32 GROMOS96 53A6 force
field in combination with Berger lipid parameter for the POPC molecules33 was used for
the whole simulation. For the force field parameters of 9a, the topology was prepared with
PRODRG28 and the CHelpG partial charge was calculated using Gaussian 0934 with the
DFT/B3LYP/6ꢀ311g** basis set.
After the initial system was set up, 1000 steps of steepest decent and 200 steps of
conjugate gradient energy minimization were performed to remove the energetically
unfavorable contacts. After that, the whole system was gradually heated from 0 to 310 K
using velocity rescaling, and equilibrated for 1.0 ns in the NPT ensemble with the protein
and ligand restrained. NoseꢀHoover35, 36 and ParrinelloꢀRahman method37, 38 were applied
ACS Paragon Plus Environment

Journal of Chemical Information and Modeling
Page 6 of 36
1
2
3
4
5
6
7
8
9
to maintain the system’s temperature at 310 K and the pressure of 1 bar, respectively. Then,
the equilibration was continued for another 500 ps without any restrains. After that, the
system was used for further unrestrained 200 ns MD simulations. During the simulation
NPT ensemble, periodic boundary conditions were applied. LINCS algorithm was adopted
to constrain all bonds.39 Particalꢀmesh Ewald (PME) algorithm40 with the grid size of about
1 Å was applied to deal with the longꢀrange electrostatic interactions. The van der Waals
interactions were treated by using a cutoff of 12 Å. All the simulations were performed
with a time step of 2 fs and the coordinates were saved every 2 ps for later analysis. Similar
strategy has been successfully applied in the studies of other biomacromolecule
systems.41ꢀ44
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
29
30
31
32
33
34
35
36
37
38
39
40
41
42
43
44
45
46
47
48
49
50
51
52
53
54
55
56
57
58
59
60
Based on the trajectory of MD simulations on 5ꢀHT1ARꢀ9a binding structure, we
tracked the hydrogen bonding distance between the side chain of S5.42 (Oxygen atom at
–OH group) and the nitrogen atom of –NH group or –F atom of 9a.
Docking 9a Derivatives into 5ꢀHT1AR AgonistꢀBinding Site. In order to see how
compatible of the substitutes of the head group and the tail group of compound 9a with
both subꢀpockets (SP1 and SP2), we performed molecular docking operations to dock all
the 9a derivatives into the agonistꢀbinding site of 5ꢀHT1AR. The Autodock4.245 and
AutoDockTools version 1.5.646 (ADT) were applied for molecular docking. The docking
files were prepared for both ligands and the receptors through the tool ADT. For the
ligands, the Gasteiger charges were assigned and the nonꢀpolar hydrogen atoms were
merged into the attached carbon atoms. The rigid root of each ligand was defined
automatically. The backbones and the amide bonds were allowed to rotate. The total
number of the rotatable bonds of our ligands was counted about 14. For the receptors, the
Gasteiger charges were added and the nonꢀpolar hydrogen atoms were merged. Two
typical conformations from the conformational analysis were selected as the target receptor
(see Supporting Information for the coordinates of these conformations). The grid box was
centered in the orthosteric agonistꢀbinding site of 5ꢀHT1AR. The box size was set large
ACS Paragon Plus Environment

Page 7 of 36
Journal of Chemical Information and Modeling
1
2
3
4
5
6
enough to cover all the residues involved in agonist binding. The spacing between grid
points was 0.375 angstroms.
7
8
9
The LGA (Lamarckian Genetic Algorithm) search algorithm was adopted to search for
the best conformers. During the docking processes, the total energy evaluations were
increased to 25,000,000, which is 10 fold of the default value to achieve a more thorough
conformation search for all 9a derivatives with unusually large number of rotatable bonds.
Gene mutation and the crossing over rate were set to 0.02 and 0.80, respectively. For each
cycle, top two individuals were allowed to survive to the next generation. 100 instead of 10
(default value) LGA runs were launched for every compound. All the generated 100
conformers were ranked according to the calculated binding free energies (∆Gbind), and
also classified using the cluster analysis function of ADT with a threshold of
rootꢀmeanꢀsquare deviation (RMSD) value of 2.0 Å. Those conformations with both the
distance between the oxygen atom of the side chain of S5.42 and the nitrogen atom of –NH
group or –F atom of 9a derivatives, and the distance between the oxygen atom of the side
chain of D3.32 and the proton atom of the piperazine group of 9a derivatives less than 3.5
Å were selected as initial candidates, and were counted for the percentage over the total
number of selected candidates.
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
29
30
31
32
33
34
35
36
37
38
39
40
41
42
43
44
45
46
47
48
49
50
51
52
53
54
55
56
57
58
59
60
First Principles Calculations. As mentioned above, the head group of 9a has two
options (ꢀNH or –F) to hydrogen bonding with the –OH group at S5.42 side chain of
5ꢀHT1AR. In order to evaluate the contribution of this hydrogen bonding from different
substituents at the head group of 9a derivatives, we performed first principles calculations
by quantum mechanics (QM) method implemented inside the Gaussian 09 package.47 For
this purpose, we used an Fꢀsubstituted indole molecule to represent the head group of 9a,
and a methanol molecule to mimic the S5.42 side chain of 5ꢀHT1AR. Such simplification
can still reasonably represent the actual situation of this hydrogen bonding interaction, and
are within the acceptable molecular size for usual QM calculations. The first principles
calculations were also performed for different head groups of 9a derivatives. The possible
ACS Paragon Plus Environment

Journal of Chemical Information and Modeling
Page 8 of 36
1
2
3
4
5
6
7
8
9
hydrogen binding options for each head group of each 9a derivative and the methanol
molecule were explored by these procedures. Firstly, the hydrogen bonding options were
searched using the mixed torsional/lowꢀmode sampling method providing by Schrodinger
Suite 201348 with OPLS2005 force field. 100,000 options for each pair of molecules were
visited during the stochastic search. Then, the conformers were fully optimized using DFT
method with Jaguar in the same Suite 2013. The B3LYP hybrid exchangeꢀcorrelation
functional and the 6ꢀ31+G* basis set were adopted for the optimization. For each option of
hydrogen bonding, the dihedral angle between the plane of the head group and the
Cαꢀhydroxyl plane of the methanol was further scanned with an interval of 15° at the same
QM level as described before. The obtained conformations were then optimized by using
Gaussian09 package49 with the DFT method of B3LYP/6ꢀ311+G** in gas phase. The
single point energy and the formation enthalpy for each pair of hydrogen bonding were
calculated using larger basis set (i.e. augꢀccꢀpVTZ set) with polarizable continuum model
(PCM) in aqueous solution. After this calculation, the population analyses of the molecular
orbitals were carried out, and the charges for each pair of hydrogen bonding were fitted to
the electrostatic potential using the CHelpG scheme.
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
29
30
31
32
33
34
35
36
37
38
39
40
41
42
43
44
45
46
47
48
49
50
51
52
53
54
55
56
57
58
59
60
Chemical Synthesis. Each of the virtually designed 9a derivatives was synthesized
according to the route as shown in Scheme 1. The indolepropanol moiety 2 was prepared
from the corresponding phenylhydrazine derivative via Fischer indole synthesis. The
generated alcohol intermediate was then converted to the bromo derivative 3 with carbon
tetrabromide and triphenylphosphine50. The 5ꢀcyanoindole propanol, 2f, was prepared
from the 5ꢀbromo compound 2e by the treatment with copper (I) cyanide in DMF. 3e was
obtained from 3d via the demethylation reaction. Starting from the aniline, 4, the
intermediate 8 was synthesized. Firstly, compound 4 was treated with ketone derivative in
the reductive amination condition to achieve a good yield of the secondary
Nꢀalkylarylamine, 5. Subsequently, the amine was acylated with acyl chloride to obtain
compounds 6 and followed by alkylation with NꢀBocꢀpiperazine to yield derivative 7, and
ACS Paragon Plus Environment

Page 9 of 36
Journal of Chemical Information and Modeling
1
2
3
4
5
6
7
8
9
then, the deprotection of the NꢀBocꢀpiperazine moiety of compound 7 lead to the formation
of intermediate 8. With the combination of alkylindole moiety 3 and Nꢀsubstituted
piperazine 8, each of the 9a derivatives was synthesized in the presence of potassium
carbonate and acetonitrile. In addition, compounds 9a_1 and 9b_1 were obtained via
methylation of 9a and 9b, respectively. Structures of all the 9a derivatives were verified by
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
29
30
31
32
33
34
35
36
37
38
39
40
41
42
43
44
45
46
47
48
49
50
51
52
53
54
55
56
57
58
59
60
13
1H and C NMR spectra. The detailed spectral data are provided at the Experimental
Section of the Supporting Information.
Binding Assays. All the synthesized 9a derivatives were subjected to competitive
binding assays for the human dopamine (D1, D2 and D3) and serotonin (5ꢀHT1A, 5ꢀHT2A)
receptors, in order to validate our virtual design based on the MD simulations and the first
principles calculations, and also to wetꢀexperimentally test how selective of these
synthesized 9a derivatives for 5ꢀHT1AR over the 5ꢀHT2AR and dopamine receptors. The
binding assays were performed by using membrane preparation obtained from stable
transfected HEK293 cells as described in our previous studies.39, 51 For each 9a derivative,
the initial screening was conducted with 10 ꢁM compound to test its inhibition on the
binding of a tritiated radioligand to each of these receptors. A compound that inhibited
binding by more than 85% was further assayed for IC50. The Ki value was calculated based
on this relation: Ki = IC50 / (1 + C/Kd). [3H]ꢀSCH23390, [3H]ꢀSpiperone, [3H]ꢀ8ꢀOHꢀDPAT
and [3H]ꢀKetanserin were used as standard radioligands for D1, D2/D3, 5ꢀHT1A, and 5ꢀHT2A
receptors, respectively. The binding reaction was conducted in duplicated tubes in 200 ꢁL
binding buffer containing 50 mM Tris, 4 mM MgCl2 (pH 7.4) at 30 °C for 50 min with
increasing concentrations (within the range of 1 nM ~ 100 ꢁM) of the testing compound in
the presence of 0.7 nM [3H]SCH23390, [3H]Spiperone, [3H]ꢀ8ꢀOHꢀDPAT or
[3H]ꢀKetanserin. The reaction was started by addition of membranes (15 ꢁg/tube), stopped
by rapid filtration through a Whatman GF/B glass fiber filter and subsequently washed
with cold buffer. Nonspecific binding was determined by parallel incubations with 10 ꢁM
SCH23390 for D1R, 10 ꢁM Spiperone for D2R/D3R, 10 ꢁM 5ꢀHT for 5ꢀHT1AR, 10 ꢁM
ACS Paragon Plus Environment

Journal of Chemical Information and Modeling
Page 10 of 36
1
2
3
4
5
6
7
8
butaclmol for 5ꢀHT2AR. Scintillation cocktail was added, and the radioactivity was
determined in a MicroBeta liquid scintillation counter. The IC50 and Ki values were
calculated by nonlinear regression using a sigmoidal function.
9
[35S]GTPγS Binding Assays. To test that our synthesized 9a derivatives are able to
activate the 5ꢀHT1AR, which in turn to bind the intracellular G protein for signal
transduction over the membrane, we performed standard [35S]GTPγS binding assay for each
of the 9a derivatives. The [35S]GTPγS binding assay was performed as described in our
previous studies.29, 39, 52 Briefly, the [35S]GTPγS binding reaction was conducted at 30 °C for
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
29
30
31
32
33
34
35
36
37
38
39
40
41
42
43
44
45
46
47
48
49
50
51
52
53
54
55
56
57
58
59
60
30 min with 30 ꢁg of membrane protein in a final volume of 200 ꢁL with various
concentration of each of the tested compound. The natural substrate of 5ꢀHT1AR, serotonin,
was used as a positive control. The binding buffer contains 50 mM Tris (pH 7.4), 5 mM
MgCl2, 1 mM ethylenediaminetetraacetic acid (EDTA), 100 mM NaCl, 1 mM
DLꢀdithiothreitol (DTT), and 40
ꢁ
M guanosine triphosphate (GDP). The [35S]GTPγS
binding reaction was initiated by adding of [35S]GTPγS (final concentration of 0.1 nM).
Nonꢀspecific binding was measured in the presence of 100 ꢁM 5′ꢀguanylimidodiphosphate
(Gpp(NH)p). The [35S]GTPγS binding reaction was terminated by the addition of 1 mL of
iceꢀcold washing buffer, and was rapidly filtered with GF/C glass fiber filters (Whatman),
and then washed for three times. Radioactivity was determined by liquid scintillation
counting.
RESULTS AND DISCUSSION
Flipꢀpacking of 9a Head Group at SP1 Subsite in 5ꢀHT1AR. Depicted in Figure 1 is
the dynamic orientation of lead compound 9a at the agonist binding site of 5ꢀHT1AR, which
were tracked through the 200 ns MD simulations. Figure 1A and 1B showed the two
typical options of hydrogen bonding between the head group of 9a and the –OH group of
S5.42 side chain at the subꢀpocket SP1 of 5ꢀHT1AR. These are, the –NH at the head group
of 9a is hydrogen bonded with –OH group at S5.42 side chain through the option of
ACS Paragon Plus Environment

Page 11 of 36
Journal of Chemical Information and Modeling
1
2
3
4
5
6
7
8
9
ꢀO…HꢀNꢀ (denoted as HBS5.42ꢀ1, and here after), and the –F atom at the head group of 9a is
hydrogen bonded with the –OH group at S5.42 side chain (Figure 1B) in the option of
–OꢀH…F (denoted as HBS5.42ꢀ5, and here after). In order to see whether these two options
of hydrogen bonding are dynamically exchangeable, we tracked the distances during the
MD simulations for these two hydrogenꢀbonding interactions. As shown in Figure 1C (red
curve), the HBS5ꢀ42ꢀ1 option of hydrogen bonding lasted for the first ~50 ns of MD
simulations, as the tracked O(S5.42)ꢀN(9a) distance fluctuated flatly around 3Å. Such
hydrogen bonding interaction was flipped gradually, and was replaced by the HBS5.42ꢀ5
option of hydrogen bonding (blue curve in Figure 1C), which lasted till ~ 150 ns along the
MD trajectory. The second flip, i.e. from HBS5.42ꢀ5 to HBS5.42ꢀ1, then happened at once, and
lasted till the end of MD simulations. Figure 1D depicts the superimposed conformations
of 9a at the binding site of 5ꢀHT1AR, which were derived from the first 150 ns of MD
trajectory with the interval of 2 ns, These dynamic properties of 5ꢀHT1ARꢀ9a binding
structure demonstrated that the head group of 9a flipped between two options of hydrogen
bonding interaction with S5.42 residue, while located always inside SP1 of 5ꢀHT1AR.
Using the OPLS2005 force field implemented in Schrodinger Suite 201318, the minimum
energy path (MEP) for the rotation of the head group of 9a was searched without the
presence of 5ꢀHT1AR, and we found that the energy barrier for the flip of 9a head group is
only ~1.0 kcal/mol (Figure S2 in Supporting Information). This indicates that the head
group of 9a is readily to rotate, supporting the observed flipꢀpacking events along the MD
simulations on 5ꢀHT1ARꢀ9a binding structure (Figure 1).
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
29
30
31
32
33
34
35
36
37
38
39
40
41
42
43
44
45
46
47
48
49
50
51
52
53
54
55
56
57
58
59
60
Figure 1.
Four possible Orientations of 9a for 5ꢀHT1ARꢀ9a Binding. Since the head group of
9a can flip inside the SP1 of the receptor (Figure 1A, Figure S1), we wanted to know how
many possible typical orientations of 9a at the agonistꢀbinding site of 5ꢀHT1AR, and their
contributions to the overall 5ꢀHT1ARꢀ9a binding. According to our results of molecular
ACS Paragon Plus Environment

Journal of Chemical Information and Modeling
Page 12 of 36
1
2
3
4
5
6
7
8
9
docking, there are 4 possible orientations of 9a at agonistꢀbinding site, which contribute
significantly to the overall receptor binding (Figure 2). Besides the observed two options of
hydrogen bonding with S5.42 residue by the head group of 9a, we also found that the tail
group (Scheme 1) of 9a can also pack in two typical options with the surround residues at
SP2 of 5ꢀHT1AR, i.e. the phenyl group of 9a packs mainly around F3.28 residue (as labeled
in Figure 1A), denoted as phenylꢀup pose; or the phenyl group of 9a mainly packs around
W7.40 (as labeled in Figure 1A) of 5ꢀHT1AR, denoted as phenylꢀdown pose. Through
cluster analysis on the selected conformational candidates of 9a from molecular docking,
we found that the HBS5.42ꢀ1 option of hydrogen bonding occupied 74.4% of all the
conformations of 9a at the binding site of 5ꢀHT1A, while 25.6% for the HBS5.42ꢀ5 option.
The phenylꢀdown pose for the tail group of 9a takes 59.0% among all the binding
conformations of 9a at the SP2 of the binding site of 5ꢀHT1AR, and 41.0% for the
phenylꢀup option of the 9a tail group. As observed, the orientation of 9a with HBS5.42ꢀ5 and
phenylꢀdown pose at the binding site (Figure 2A) takes only 12.8% over the population of
9a binding conformations, and has the calculated binding free energy (∆Gbind) of ꢀ10.55
kcal/mol. The orientation of 9a with HBS5.42ꢀ5 and phenylꢀup pose has also 12.8% of the
whole population, and has the ∆Gbind value of ꢀ10.45 kcal/mol (Figure 2B). The 9a
orientation with HBS5.42ꢀ1 and phenylꢀdown pose (Figure 2C) at the binding site of
5ꢀHT1AR takes 46.2%, with the ∆Gbind value of ꢀ11.0 kcal/mol, and the ∆Gbind value of
ꢀ10.87 kcal/mol for the HBS5.42ꢀ1 plus phenylꢀup pose (Figure 2D). The total binding free
energy contributed from these four possible orientations of 9a at the agonistꢀbinding site of
5ꢀHT1AR is ꢀ10.83 kcal/mol.
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
29
30
31
32
33
34
35
36
37
38
39
40
41
42
43
44
45
46
47
48
49
50
51
52
53
54
55
56
57
58
59
60
Figure 2.
Testing the Observed FlipꢀPacking of 9a Head Group. In order to test that the
observed flipꢀpacking of 9a head group at the SP1 site of 5ꢀHT1A, and the importance of the
hydrogen bonding between the 9a head group and S5.42 residue to the affinity of
ACS Paragon Plus Environment

Page 13 of 36
Journal of Chemical Information and Modeling
1
2
3
4
5
6
7
8
9
5ꢀHT1Aꢀ9a binding, we virtually designed three 9a analogs. These are, 9b with the
HBS5.42ꢀ5 blocked (i.e. removed –F atom) (Scheme 1, Table 1); 9a_1 with HBS5.42ꢀ1 blocked
(i.e. added methyl substituent to the nitrogen atom at the head group); and 9b_1 (no
hydrogen bonding with S5.42 residue). Using the same protocols of molecular docking as
described above, we found that 9b takes the HBS5.42ꢀ1 option up to 87.9%, that is, 7.3:1 ratio
for the HBS5.42ꢀ1 over the nonꢀHBS5.42ꢀ1, and the calculated value of ∆Gbind for 9b is ꢀ10.29
kcal/mol. The calculated ∆Gbind values for 9a_1 (the ratio of HBS5.42ꢀ5 over nonꢀHBS5.42ꢀ5 as
1.7:1), and 9b_1 (the head group flips equally toward S5.42 residue of 5ꢀHT1A) are ꢀ10.09
kcal/mol, and ꢀ9.02 kcal/mol, respectively. Based on these calculated values of ∆Gbind, we
predicted that the order of receptorꢀbinding affinity for these four compounds should be
9a > 9b > 9a_1 > 9b_1. We then submitted these four compounds to chemical synthesis (as
shown in Scheme 1) and binding assays (as described above). The wetꢀexperimental
measurements are listed in Table 1. Interestingly, the measured binding affinities of these
four compounds followed same decreasing order as 9a > 9b > 9a_1 > 9b_1, i.e. the
increasing Ki (nM) value or the percentage for the displacement of radioligand at 10 ꢀM,
The consistency between our computational predictions and wetꢀexperimental results
strongly suggests that the hydrogen bonding between 9a and S5.42 residue of 5ꢀHT1A is
very important for agonistꢀbinding, and the HBS5.42ꢀ1 option of hydrogen bonding
contributes more to the agonist affinity than that from the HBS5.42ꢀ5 option.
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
29
30
31
32
33
34
35
36
37
38
39
40
41
42
43
44
45
46
47
48
49
50
51
52
53
54
55
56
57
58
59
60
Table 1
Testing the Flexibility of 9a Tail Group at SP2 Site. As found in the above 9a
analogs (9b, 9a_1, 9b_1), the tail group of 9a can take the pose as both phenylꢀup and
phenylꢀdown at the SP2 subsite of 5ꢀHT1AR (Figure 2), we wanted to test how flexible of
the tail group. The similar MEP for the rotation of the tail group of 9a was also searched
without the presence of 5ꢀHT1AR, and we found that the energy barrier for the rotation of
9a tail group is as small as ~1.3 kcal/mol (Figure S3 in Supporting Information). We first
ACS Paragon Plus Environment

Journal of Chemical Information and Modeling
Page 14 of 36
1
2
3
4
5
6
7
8
9
changed the size of the hexacycle substituent at the tail group, i.e. the pentaꢀcycle
substituent for 9f, heptacycle substituent for 9g, and 9h by adding a methylꢀsubstituent at
the hexacycle group at the tail of 9a (Scheme 1). Results of docking these three molecules
(9f, 9g and 9h) into the agonistꢀbinding site of 5ꢀHT1AR showed that, these three analogs
can still have the options of phenylꢀup and phenylꢀdown for the tail at SP2 site (Figure S4
in Supporting Information). However, results of binding assays (Table 2) showed that
these three analogs have decreased affinities with 5ꢀHT1AR, suggesting that the SP2 subsite
in the receptor does not accept largerꢀsize alkylꢀcycle substituents. For this reason, we
designed more 9a derivatives by changing the hexacycle substituent with smaller
alkylꢀchain substituents, namely 9j, 9k, 9l, 9m and 9n (Scheme 1). The results of binding
assays on these 9a derivatives showed that the 9m, with the pentyl substituent at the tail
group, has better affinity for 5ꢀHT1ARꢀbinding, i.e. Ki = 3.5 ± 0.1 nM (Table 2), which
corresponded to the ratio of phenylꢀup/phenylꢀdown of the tail group to be 1.10:1 from
molecular docking operations.
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
29
30
31
32
33
34
35
36
37
38
39
40
41
42
43
44
45
46
47
48
49
50
51
52
53
54
55
56
57
58
59
60
Table 2
Rational Design of HigherꢀAffinity 9a Derivatives. Encouraged by the above
preliminary testing results on the binding flexibility of 9a analogs at the agonistꢀbinding
site of 5ꢀHT1AR, we extended our efforts to design higher affinity agonists by tuning the
observed flipꢀpacking of 9a head group and the phenylꢀup/phenylꢀdown poses of 9a tail
group. For this purpose, we designed three more 9a derivatives, namely 9c with –Cl
substituent at the original Fꢀatom position, 9d with a –CN substituent and 9e with –OH
substituent (Scheme 1), i.e. playing the HBS5.42ꢀ5 pattern of hydrogen bonding at the head
group of 9a. We performed the first principles calculations on these three more derivatives,
and 9a and 9b. The results of QM calculations as shown in Figure 3 indicate that all these
9a derivatives have at least one option of hydrogen bonding, i.e. HBS5.42ꢀ1 and/or HBS5.42ꢀ5
.
As the substituent at the original –F atom of the head group changed from –H, to –F, to –Cl,
ACS Paragon Plus Environment

Page 15 of 36
Journal of Chemical Information and Modeling
1
2
3
4
5
6
7
8
9
to –CN, and to –OH, the chance of forming HBS5.42ꢀ5 became higher (Figure 3A), and
therefore, the total contribution from hydrogen bonding to the affinity should increase.
Based on the results of these QM calculations and the ratio of HBS5.42ꢀ5/HBS5.42ꢀ1 from
docking operations, we would predict that these 9a derivatives have the order of
5ꢀHT1ARꢀaffinity as 9b < 9a ~9c < 9d < 9e. The results of binding assays on these
synthesized 9a derivatives are shown in Figure 3B and also listed in Table 3. The measured
Ki values for these 9a derivatives confirmed the predicted order of affinity from QM
calculations and molecular docking operations. Specifically, compound 9e showed much
higher affinity of 5ꢀHT1ARꢀbinding, i.e. the Ki value is as low as 1.1 ± 0.1 nM, about
50ꢀfold increase in receptorꢀbinding affinity when compared with that of 9a.
Figure 3
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
29
30
31
32
33
34
35
36
37
38
39
40
41
42
43
44
45
46
47
48
49
50
51
52
53
54
55
56
57
58
59
60
Table 3
Figure 4 depicts the four possible orientations of compound 9e at the agonistꢀbinding
site of 5ꢀHT1AR, which were obtained by the same protocols as that applied on lead
compound 9a (Figure 2). These results suggest that the way of our rational design is
reasonable.
Figure 4
Further tuning on the 9a tail group has led to the design of 9i. By adding a –OꢀCH3
substituent to the paraꢀposition of hexacycle part at the tail of 9a, we synthesized
compound 9i (as shown in Scheme 1), and submitted it to the biding affinity assay. It turned
out that 9i has higher affinity of 5ꢀHT1ARꢀbinding than that of 9a. As listed in Table 3 for 9i,
its Ki = 11.0 ± 2 nM. Structurally, 9i is a racemic mixture, that is, it has the cisꢀconformer
(9i_1) and the transꢀconformer (9i_2). In a similar way, we docked both these two
compounds into the agonistꢀbinding site of 5ꢀHT1AR, and found that the oxygen atom of
–OꢀCH3 at the tail group of 9i_2 can form one additional hydrogen bond with W7.40 side
chain of 5ꢀHT1AR (as shown in Figure 5C). Based on these docking results, we would
ACS Paragon Plus Environment

Journal of Chemical Information and Modeling
Page 16 of 36
1
2
3
4
5
6
7
8
9
predict that the 9i_2 should have higher affinity with 5ꢀHT1AR than that of 9i itself. After
both 9i_1 and 9i_2 were synthesized (rout as shown in Scheme 1), we again submitted
these two compounds 9i_1 and 9i_2 to receptorꢀbinding assays. The results of
wetꢀexperimental tests are listed in Table 3, and that compound 9i_2 showed the Ki = 1.6 ±
1 nM, pretty higher affinity for 5ꢀHT1ARꢀbinding than that of 9i.
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
29
30
31
32
33
34
35
36
37
38
39
40
41
42
43
44
45
46
47
48
49
50
51
52
53
54
55
56
57
58
59
60
Figure 5
Allꢀinꢀone Tuning for the Design of Higherꢀaffinity Agonists. As described above,
the rationally designed 9e and 9i_2 became the nMꢀlevel agonists of 5ꢀHT1AR. This is
because these two higherꢀaffinity agonists were designed either through taking advantage
of the observed flipꢀpacking phenomenon of the head group of 9a at the SP1 subsite of
5ꢀHT1AR, or through tuning the flexibility of the tail of 9a at the SP2 subsite of the receptor.
If we combined these two merits together, we wanted to see whether or not the allꢀinꢀone
tuning of the binding flexibility at the agonistꢀbinding site of 5ꢀHT1AR could help us make
new active compounds. Followed the same protocols as described above, i.e. through a
combination of molecular docking and wetꢀexperimental tests. We discovered new
compounds 9o and 9p (rout as shown in Scheme 1), which have both the best
hydrogenꢀbonding options at the head group, and the best flexibility of the tail group
(phenylꢀup/phenylꢀdown poses). As listed in Table 3, both 9o and 9p showed nMꢀlevel
affinities toward the 5ꢀHT1ARꢀbinding, i.e. Ki values are 1.01±0.1 nM, and 1.05±0.3 nM,
respectively.
In addition, we also checked the 5ꢀHT1ARꢀactivation abilities of our newly designed
9a derivatives by performing the standard [35S]GTPγS binding assays. Table 4 listed the
results of wetꢀexperimental tests for some of our newly designed 9a derivatives. As shown
in Table 4, the most active compounds 9e, 9i_2 and 9o, have the EC50 values of 24.42 nM,
11.95 nM and 15.81 nM, respectively. The muchꢀlowered EC50 values of these compounds
confirm again that our rationale to design higherꢀaffinity agonists of 5ꢀHT1AR is
ACS Paragon Plus Environment

Page 17 of 36
Journal of Chemical Information and Modeling
1
2
3
4
successful.
5
6
7
8
Meanwhile, we also performed wetꢀexperimental measurements for the affinities of
our newly designed 9a derivatives with subtype 5ꢀHT2AR, and also for the dopamine
receptors (D1, D2 and D3), in order to test their selectivity about 5ꢀHT1AR. As listed in
Table 3, the 5ꢀHT2AR/5ꢀHT1AR selectivity for our newly designed 9a derivatives became
higher as their binding affinities with 5ꢀHT1AR became higher. For example, the
5ꢀHT2AR/5ꢀHT1AR selectivity for 9a is ~4ꢀfold, and it goes up to ~33ꢀfold for 9i, to
~56ꢀfold for 9i_2, to ~148ꢀfold for 9e, and even to ~333ꢀfold for 9p. Additionally, all of
our newly designed 9a derivatives are also highly selective over dopamine receptors
(Table 3). These data of selectivity suggest that the observed binding flexibility for 9a
derivatives at the subsites SP1 and SP2 of 5ꢀHT1AR is unique, and our tuning approaches
can be generally applied in the design of highꢀaffinity agonists targeting other subtype
5ꢀHT receptors.
9
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
29
30
31
32
33
34
35
36
37
38
39
40
41
42
43
44
45
46
47
48
49
50
51
52
53
54
55
56
57
58
59
60
CONCLUSION
In summary, we observed a dynamic flipꢀpacking phenomenon for the head group of
lead compound 9a through largeꢀscale molecular dynamics simulations, and we also
observed four possible orientations of 9a at the agonistꢀbinding site of 5ꢀHT1AR by
molecular docking and binding free energy calculations. We found that the tail group of
lead compound 9a can also take different poses at the subsite of 5ꢀHT1AR, namely the
phenylꢀup and phenylꢀdown. Taking advantage of the flipꢀpacking, we virtually designed
new 9a derivatives with different substituents at the head group through first principles
calculations and molecular docking operations. These computational designs were
validated by our wetꢀexperimental measurements (i.e. chemical synthesis, binding assays
and functional assays), and have led to the discovery of nMꢀlevel agonists (e.g. the new
compound 9e). By tuning the binding flexibility of the tail group of lead compound 9a, we
also discovered nMꢀlevel 5ꢀHT1AR agonists with the same head group of 9a, e.g. the new
ACS Paragon Plus Environment

Journal of Chemical Information and Modeling
Page 18 of 36
1
2
3
4
5
6
7
8
9
compound 9i_2. The allꢀinꢀone tuning on lead compound 9a have resulted the discovery of
higherꢀaffinity agonists (9o and 9p) of 5ꢀHT1AR, with ~50ꢀfold increase of binding affinity,
significantly improved agonistic function (EC50), and also very high 5ꢀHT1AR selectivity.
The combinatory approach based on both the computational predicting and
wetꢀexperimental validation has been proved working perfectly in our design of
higherꢀaffinity agonists for 5ꢀHT1AR, and thus, this approach can also be generally applied
to design highly active ligands of other subtype 5ꢀHT receptors.
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
29
30
31
32
33
34
35
36
37
38
39
40
41
42
43
44
45
46
47
48
49
50
51
52
53
54
55
56
57
58
59
60
ACKNOLEDGMENTS
This work was supported by the National Natural Science Foundation of China (No.
81172919, 81130023, 81373382, 31400624), National Basic Research Plan (973) of the
Ministry of Science and Technology of China (2011CB5C4403). Support from the Priority
Academic Program Development of Jiangsu Higher Education Institutes (PAPD) and a
grant obtained from the Jiangsu Science and Technology commission (BM2013003) are
appreciated.
ASSOCIATED CONTENT
Supporting Information Available: Four supporting figures, procedures for chemical
synthesis, detailed spectral data, coordinates for the modeled 5ꢀHT1AR and the possible
orientations for 9e. This material is available free of charge via the Internet at
http://pubs.acs.org.
REFERENCES
1. Franchini, S.; Prandi, A.; Sorbi, C.; Tait, A.; Baraldi, A.; Angeli, P.; Buccioni, M.; Cilia, A.;
Poggesi, E.; Fossa, P.; Brasili, L., Discovery of A New Series of 5ꢀHT1A Receptor Agonists.
Bioorg. Med. Chem. Lett. 2010, 20, 2017ꢀ2020.
2. Valhondo, M.; Marco, I.; MartínꢀFontecha, M.; VázquezꢀVilla, H.; Ramos, J. A.; Berkels, R.;
ACS Paragon Plus Environment

Page 19 of 36
Journal of Chemical Information and Modeling
1
2
3
4
5
6
7
8
9
Lauterbach, T.; Benhamú, B.; LópezꢀRodríguez, M. L., New Serotonin 5ꢀHT1A Receptor Agonists
Endowed with Antinociceptive Activity in Vivo. J. Med. Chem. 2013, 56, 7851ꢀ7861.
3. Enza Lacivita, P. D. P., Paola De Giorgio, Nicola A Colabufo, Francesco Berardi, Roberto
Perrone & Marcello Leopoldo, The Therapeutic Potential of 5ꢀHT1A Receptors: A Patent Review.
Expert Opin. Ther. Pat. 2012, 22, 887ꢀ902.
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
29
30
31
32
33
34
35
36
37
38
39
40
41
42
43
44
45
46
47
48
49
50
51
52
53
54
55
56
57
58
59
60
4. Liu, Z.; Zhang, H.; Ye, N.; Zhang, J.; Wu, Q.; Sun, P.; Li, L.; Zhen, X.; Zhang, A., Synthesis of
Dihydrofuroaporphine Derivatives: Identification of a Potent and Selective Serotonin 5ꢀHT 1A
Receptor Agonist. J. Med. Chem. 2010, 53, 1319ꢀ1328.
5. Wrobel, M. Z.; Chodkowski, A.; Herold, F.; Gomolka, A.; Kleps, J.; Mazurek, A. P.; Plucinski,
F.; Mazurek, A.; Nowak, G.; Siwek, A.; Stachowicz, K.; Slawinska, A.; Wolak, M.; Szewczyk, B.;
Satala, G.; Bojarski, A. J.; Turlo, J., Synthesis and Biological Evaluation of Novel
Pyrrolidineꢀ2,5ꢀdione Derivatives as Potential Antidepressant Agents. Part 1. Eur. J. Med. Chem.
2013, 63, 484ꢀ500.
6. Mueller, C. P.; Carey, R. J.; Huston, J. P.; Silva, M. A. D. S., Serotonin and Psychostimulant
Addiction: Focus on 5ꢀHT1Aꢀreceptors. Prog. Neurobiol. 2007, 81, 133ꢀ178.
7. Lanoir, J.; Hilaire, G.; Seif, I., Reduced Density of Functional 5ꢀHT1A Receptors in the Brain,
Medulla and Spinal Cord of Monoamine OxidaseꢀA Knockout Mouse Neonates. J. Comp. Neurol.
2006, 495, 607ꢀ623.
8. Jason Hannon, D. H., Serotonin Receptors and Systems: Endless Diversity. Acta Biol. (Szeged)
2002, 42, 1ꢀ12.
9. Weber, K. C.; Salum, L. B.; Honorio, K. M.; Andricopulo, A. D.; da Silva, A. B.,
Pharmacophoreꢀbased 3D QSAR Studies on a Series of High Affinity 5ꢀHT1A Receptor Ligands.
Eur. J. Med. Chem. 2010, 45, 1508ꢀ1514.
10. Dilly, S.; Graulich, A.; Liegeois, J. F., Molecular Modeling Study of 4ꢀphenylpiperazine and
4ꢀphenylꢀ1,2,3,6ꢀtetrahydropyridine Derivatives: A New Step Towards the Design of Highꢀaffinity
5ꢀHT1A Ligands. Bioorg. Med. Chem. Lett. 2010, 20, 1118ꢀ1123.
11. Hendry, N.; Christie, I.; Rabiner, E. A.; Laruelle, M.; Watson, J., In vitro Assessment of the
Agonist Properties of the Novel 5ꢀHT1AReceptor Ligand, CUMIꢀ101 (MMP), in Rat Brain Tissue.
Nucl. Med. Biol. 2011, 38, 273ꢀ277.
12. Kuipers, W.; Kruse, C. G.; van Wijngaarden, I.; Standaar, P. J.; Tulp, M. T. M.; Veldman, N.;
Spek, A. L.; Ijzerman, A. P., 5ꢀHT1Aꢀ versus D2ꢀReceptor Selectivity of Flesinoxan and
Analogous Nꢀ4ꢀSubstituted Nꢀ1ꢀArylpiperazines. J. Med. Chem. 1997, 40, 300ꢀ312.
13. Vicente, M. A.; Zangrossi, H., Jr.; dos Santos, L.; de Macedo, C. E.; Andrade, T. G.,
Involvement of Median Raphe Nucleus 5ꢀHT1A Receptors in the Regulation of Generalized
Anxietyꢀrelated Defensive Behaviours in Rats. Neurosci. Lett. 2008, 445, 204ꢀ208.
14. Albert, P. R.; Lemonde, S., 5ꢀHT1A Receptors, Gene Repression, and Depression: Guilt by
Association. Neuroscientist 2004, 10, 575ꢀ593.
15. Slowinski, T.; Stefanowicz, J.; Dawidowski, M.; Kleps, J.; Czuczwar, S.; AndresꢀMach, M.;
Luszczki, J. J.; Nowak, G.; Stachowicz, K.; Szewczyk, B.; Slawinska, A.; Mazurek, A. P.; Mazurek,
A.; Plucinski, F.; Wolska, I.; Herold, F., Synthesis and Biological Investigation of Potential
Atypical Antipsychotics with a Tropane Core. Part 1. Eur. J. Med. Chem. 2011, 46, 4474ꢀ4488.
ACS Paragon Plus Environment

Journal of Chemical Information and Modeling
Page 20 of 36
1
2
3
4
5
16. Yuan, Q.; Joiner, W. J.; Sehgal, A., A Sleepꢀpromoting Role for the Drosophila Serotonin
Receptor 1A. Curr. Biol. 2006, 16, 1051ꢀ1062.
6
7
8
9
17. Mo, J.; Zhang, H.; Yu, L.ꢀP.; Sun, P.ꢀH.; Jin, G.ꢀZ.; Zhen, X., lꢀStepholidine Reduced
lꢀDOPAꢀinduced Dyskinesia in 6ꢀOHDAꢀlesioned Rat Model of Parkinson's Disease. Neurobiol.
Aging 2010, 31, 926ꢀ936.
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
29
30
31
32
33
34
35
36
37
38
39
40
41
42
43
44
45
46
47
48
49
50
51
52
53
54
55
56
57
58
59
60
18. Zhang, H.; Ye, N.; Zhou, S.; Guo, L.; Zheng, L.; Liu, Z.; Gao, B.; Zhen, X.; Zhang, A.,
Identification of Nꢀpropylnoraporphinꢀ11ꢀyl 5ꢀ(1,2ꢀdithiolanꢀ3ꢀyl)pentanoate as a New
AntiꢀParkinson's Agent Possessing a Dopamine D2 and Serotonin 5ꢀHT1A Dualꢀagonist Profile. J.
Med. Chem. 2011, 54, 4324ꢀ4338.
19. Zhao, R.; Lu, W.; Fang, X.; Guo, L.; Yang, Z.; Ye, N.; Zhao, J.; Liu, Z.; Jia, J.; Zheng, L.; Zhao,
B.; Zhang, A.; Zhen, X., (6aR)ꢀ11ꢀaminoꢀNꢀpropylꢀnoraporphine, a New Dopamine D2 and
Serotonin 5ꢀHT1A Dual Agonist, Elicits Potent Antiparkinsonian Action and Attenuates
Levodopaꢀinduced Dyskinesia in a 6ꢀOHDAꢀlesioned Rat Model of Parkinson's Disease.
Pharmacol., Biochem. Behav. 2014, 124, 204ꢀ210.
20. Altieri, S. C.; GarciaꢀGarcia, A. L.; Leonardo, E. D.; Andrews, A. M., Rethinking 5ꢀHT1A
Receptors: Emerging Modes of Inhibitory Feedback of Relevance to Emotionꢀrelated Behavior.
ACS Chem. Neurosci. 2013, 4, 72ꢀ83.
21. Fox, S. H.; Chuang, R.; Brotchie, J. M., Serotonin and Parkinson's Disease: On Movement,
Mood, and Madness. Mov. Disord. 2009, 24, 1255ꢀ1266.
22. Celada, P.; Bortolozzi, A.; Artigas, F., Serotonin 5ꢀHT1A Receptors as Targets for Agents to
Treat Psychiatric Disorders: Rationale and Current Status of Research. CNS Drugs 2013, 27,
703ꢀ716.
23. Murata, Y.; Yanagihara, Y.; Mori, M.; Mine, K.; Enjoji, M., Chronic Treatment with
Tandospirone, a Serotonin 1A Receptor Partial Agonist, Inhibits Psychosocial Stressꢀinduced
Changes in Hippocampal Neurogenesis and Behavior. J. Affective Disord. 2015, 180, 1ꢀ9.
24. McKean, J.; Watts, H.; Mokszycki, R., Breakthrough Seizures after Starting Vilazodone for
Depression. Pharmacotherapy 2015, 35, E6ꢀE8.
25. Gommoll, C.; Durgam, S.; Mathews, M.; Forero, G.; Nunez, R.; Tang, X.; Thase, M. E., A
Doubleꢀblind, Randomized, Placeboꢀcontrolled, Fixedꢀdose Phase III Study of Vilazodone in
Patients with Generalized Anxiety Disorder. Depression Anxiety 2015, 32, 451ꢀ459.
26. Colpaert, F. C.; Tarayre, J. P.; Koek, W.; Pauwels, P. J.; Bardin, L.; Xu, X. J.;
WiesenfeldꢀHallin, Z.; Cosi, C.; CarillaꢀDurand, E.; Assie, M. B.; Vacher, B., Largeꢀamplitude
5ꢀHT1A Receptor Activation: A New Mechanism of Profound, Central Analgesia.
Neuropharmacology 2002, 43, 945ꢀ958.
27. Guo, Y.; Zhang, H.; Chen, X.; Cai, W.; Cheng, J.; Yang, Y.; Jin, G.; Zhen, X., Evaluation of the
Antipsychotic Effect of Biꢀacetylated lꢀstepholidine (lꢀSPDꢀA), a Novel Dopamine and Serotonin
Receptor Dual Ligand. Schizophr. Res. 2009, 115, 41ꢀ49.
28. Xu, L.; Zhou, S.; Yu, K.; Gao, B.; Jiang, H.; Zhen, X.; Fu, W., Molecular Modeling of the 3D
Structure of 5ꢀHT(1A)R: Discovery of Novel 5ꢀHT(1A)R Agonists via Dynamic
Pharmacophoreꢀbased Virtual Screening. J. Chem. Inf. Model. 2013, 53, 3202ꢀ3211.
29. Lian, P.; Xu, L.; Geng, C.; Qian, Y.; Li, W.; Zhen, X.; Fu, W., A Computational Perspective on
ACS Paragon Plus Environment

Page 21 of 36
Journal of Chemical Information and Modeling
1
2
3
4
5
6
7
8
Drug Discovery and Signal Transduction Mechanism of Dopamine and Serotonin Receptors in the
Treatment of Schizophrenia. Curr. Pharm. Biotechnol. 2014, 15, 916ꢀ926.
30. Bang, I.; Choi, H.ꢀJ., Structural Features of β2 Adrenergic Receptor: Crystal Structures and
Beyond. Mol. Cells 2015, 38, 105ꢀ111.
31. Lebon, G.; Warne, T.; Edwards, P. C.; Bennett, K.; Langmead, C. J.; Leslie, A. G. W.; Tate, C.
G., Agonistꢀbound Adenosine A(2A) Receptor Structures Reveal Common Features of GPCR
Activation. Nature 2011, 474, 521ꢀ525.
9
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
29
30
31
32
33
34
35
36
37
38
39
40
41
42
43
44
45
46
47
48
49
50
51
52
53
54
55
56
57
58
59
60
32. Zhang, J.; Zhang, K.; Gao, Z.ꢀG.; Paoletta, S.; Zhang, D.; Han, G. W.; Li, T.; Ma, L.; Zhang, W.;
Mueller, C. E.; Yang, H.; Jiang, H.; Cherezov, V.; Katritch, V.; Jacobson, K. A.; Stevens, R. C.; Wu,
B.; Zhao, Q., Agonistꢀbound Structure of the Human P2Y(12) Receptor. Nature 2014, 509,
119ꢀ122.
33. Xu, F.; Wu, H. X.; Katritch, V.; Han, G. W.; Jacobson, K. A.; Gao, Z. G.; Cherezov, V.; Stevens,
R. C., Structure of an AgonistꢀBound Human A(2A) Adenosine Receptor. Science 2011, 332,
322ꢀ327.
34. Kandt, C.; Ash, W. L.; Peter Tieleman, D., Setting up and Running Molecular Dynamics
Simulations of Membrane Proteins. Methods 2007, 41, 475ꢀ488.
35. Berendsen, H. J. C.; Postma, J. P. M.; van Gunsteren, W. F.; Hermans, J. Interaction Models for
Water in Relation to Protein Hydration. In Intermol. Forces, Pullman, B., Ed.; D. Reidel Publishing
Company: 1981, pp 331ꢀ342.
36. Hess, B.; Kutzner, C.; van der Spoel, D.; Lindahl, E., GROMACS 4: Algorithms for Highly
Efficient, LoadꢀBalanced, and Scalable Molecular Simulation. J. Chem. Theory Comput. 2008, 4,
435ꢀ447.
37. Berger, O.; Edholm, O.; Jähnig, F., Molecular Dynamics Simulations of a Fluid Bilayer of
Dipalmitoylphosphatidylcholine at Full Hydration, Constant Pressure, and Constant Temperature.
Biophys. J. 1997, 72, 2002ꢀ2013.
38. Schuttelkopf, A. W.; van Aalten, D. M., PRODRG: A Tool for Highꢀthroughput
Crystallography of Proteinꢀligand Complexes. Acta Crystallogr., Sect. D: Biol. Crystallogr. 2004,
60, 1355ꢀ1363.
39. M.J.Frisch; G.W.Trucks; H.B.Schlegel; G.E.Scuseria; M.A.Robb; J.R.Cheeseman; G.Scalmani;
V.Barone; B.Mennucci; G.A.Petersson; H.Nakatsuji; M.Caricato; X.Li; H.P.Hratchian;
A.F.Izmaylov; J.Bloino; G.Zheng; J.L.Sonnenberg; M.Hada; M.Ehara; K.Toyota; R.Fukuda;
J.Hasegawa; M.Ishida; T.Nakajima; Y.Honda; O.Kitao; H.Nakai; T.Vreven; J.A.Montgomery; Jr.;
J.E.Peralta; F.Ogliaro; M.Bearpark; J.J.Heyd; E.Brothers; K.N.Kudin; V.N.Staroverov; T.Keith;
R.Kobayashi; J.Normand; K.Raghavachari; A.Rendell; J.C.Burant; S.S.Iyengar; J.Tomasi;
M.Cossi; N.Rega; J.M.Millam; M.Klene; J.E.Knox; J.B.Cross; V.Bakken; C.Adamo; J.Jaramillo;
R.Gomperts; R.E.Stratmann; O.Yazyev; A.J.Austin; R.Cammi; C.Pomelli; J.W.Ochterski;
R.L.Martin; K.Morokuma; V.G.Zakrzewski; G.A.Voth; P.Salvador; J.J.Dannenberg; S.Dapprich;
A.D.Daniels; O.Farkas; J.B.Foresman; J.V.Ortiz; J.Cioslowski; D.J.Fox Gaussian 09, Gaussian,
Inc.: Wallingford CT, 2010.
40. Hoover, W., Canonical Dynamics: Equilibrium Phaseꢀspace Distributions. Phys. Rev. A 1985,
31, 1695ꢀ1697.
ACS Paragon Plus Environment

Journal of Chemical Information and Modeling
Page 22 of 36
1
2
3
4
5
41. Nosé, S., A Molecular Dynamics Method for Simulations in the Canonical Ensemble. Mol.
Phys. 1984, 52, 255ꢀ268.
6
7
8
42. Nosé, S.; Klein, M. L., Constant Pressure Molecular Dynamics for Molecular Systems. Mol.
Phys. 1983, 50, 1055ꢀ1076.
43. Parrinello, M.; Rahman, A., Polymorphic Transitions in Single Crystals: A New Molecular
Dynamics Method. J. Appl. Phys. 1981, 52, 7182ꢀ7190.
9
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
29
30
31
32
33
34
35
36
37
38
39
40
41
42
43
44
45
46
47
48
49
50
51
52
53
54
55
56
57
58
59
60
44. Hess, B.; Bekker, H.; Berendsen, H. J. C.; Fraaije, J. G. E. M., LINCS: A Linear Constraint
Solver for Molecular Simulations. J. Comput. Chem. 1997, 18, 1463ꢀ1472.
45. Darden, T.; York, D.; Pedersen, L., Particle Mesh Ewald: An N . log(N) Method for Ewald
Sums in Large Systems. Chem. Phys. 1993, 98, 10089ꢀ10092.
46. Lian, P.; Angela Liu, L.; Shi, Y.; Bu, Y.; Wei, D., TetheredꢀHopping Model for ProteinꢀDNA
Binding and Unbinding Based on Sox2ꢀOct1ꢀHoxb1 Ternary Complex Simulations. Biophys. J.
2010, 98, 1285ꢀ1293.
47. Lian, P.; Li, J.; Wang, D. Q.; Wei, D. Q., CarꢀParrinello Molecular Dynamics/Molecular
Mechanics (CPMD/MM) Simulation Study of Coupling and Uncoupling Mechanisms of
Cytochrome P450cam. J. Phys. Chem. B 2013, 117, 7849ꢀ7856.
48. Lian, P.; Wei, D.ꢀQ.; Wang, J.ꢀF.; Chou, K.ꢀC., An Allosteric Mechanism Inferred from
Molecular Dynamics Simulations on Phospholamban Pentamer in Lipid Membranes. PLoS One
2011, 6, e18587.
49. Lian, P.; Guo, H.ꢀB.; Smith, J. C.; Wei, D.ꢀQ.; Guo, H., Catalytic Mechanism and Origin of
High Activity of Cellulase TmCel12A at High Temperature: A Quantum Mechanical/Molecular
Mechanical Study. Cellulose 2013, 1ꢀ13.
50. Morris, G. M.; Huey, R.; Lindstrom, W.; Sanner, M. F.; Belew, R. K.; Goodsell, D. S.; Olson, A.
J., AutoDock4 and AutoDockTools4: Automated Docking with Selective Receptor Flexibility. J.
Comput. Chem. 2009, 30, 2785ꢀ2791.
51. Sanner, M. F., Python: A programming language for software integration and development. J.
Mol. Graphics Modell. 1999, 17, 57ꢀ61.
52. Schrödinger Release 2013-3, Schrödinger, LLC, New York, NY: 2013.
ACS Paragon Plus Environment

Page 23 of 36
Journal of Chemical Information and Modeling
1
2
3
4
Tables
5
6
7
8
9
Table 1. The wetꢀexperimentally measured 5ꢀHT1A binding affinities (Ki, nM, or the
percentage for the displacement of the readioligand at 10 ꢁM) for the designed 9a
derivatives based on the observed flipꢀpacking of the head group (Scheme 1) of 9a from
MD simulations and molecular docking. Also listed the calculated binding free energies
(∆Gbind, kcal/mol)
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
29
30
31
32
33
34
35
36
37
38
39
40
41
42
43
44
45
46
47
48
49
50
51
52
53
54
55
56
57
58
59
60
Compound
HB options
9a
9b
9a_1
9b_1
2
1
1
0
F
F
Head group
N
H
N
H
N
N
HBS5.42ꢀ1
HBS5.42ꢀ5
/
*
2.9:1
7.3:1
1:1.7
1:1.1
ꢀ9.02
∆Gbind**
Ki
ꢀ10.83
51±16
ꢀ10.29
314±7
ꢀ10.09
542±3
82.29%
* The ratio of hydrogen bonding HBS5.42ꢀ1/HBS5.42ꢀ5 counted from results of molecular
docking operations.
** Each calculated free energy value was the weighted average based on its population.
ACS Paragon Plus Environment

Journal of Chemical Information and Modeling
Page 24 of 36
1
2
3
4
5
6
7
8
9
Table 2. Wetꢀexperimentally measured 5ꢀHT1A binding affinities (Ki, nM) for the designed
9a derivatives based on the observed flipꢀpacking of the tail group (Scheme 1) of 9a from
MD simulations and molecular docking. Also listed the calculated binding free energies
(∆Gbind, kcal/mol)
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
29
30
31
32
33
34
35
36
37
38
39
40
41
42
43
44
45
46
47
48
49
50
51
52
53
54
55
56
57
58
59
60
Carbon
Phenyl_down/
Phenyl_up*
Compound
R1 group
∆Gbind**
Ki
atoms in R1
1.32:1
ꢀ10.33
ꢀ10.83
ꢀ10.55
ꢀ10.25
42±5
51±16
192±11
88±13
9f
9a
9g
9h
5
6
1.44:1
1.85:1
1:2.96
7
7
0
1
H
0.24:1
0.45:1
ꢀ9.64
ꢀ9.89
103±3
51±10
9j
CH3
9k
41±4
3.5±0.1
41±5
3
5
6
0.59:1
1.10:1
1.00:1
ꢀ10.23
ꢀ10.31
ꢀ10.18
9l
9m
9n
* The ratio of phenyl position Phenyl_down/Phenyl_up counted from the results of
molecular docking operations.
** Each calculated free energy value was the weighted average based on its population.
ACS Paragon Plus Environment

Page 25 of 36
Journal of Chemical Information and Modeling
1
2
3
4
5
6
7
Table 3. Wetꢀexperimentally measured binding affinities (Ki, nM, or the percentage for the
displacement of radioligand at 10 ꢁM) for all the newly designed 9a derivatives with
dopamine receptors (D1, D2 and D3) and serotonin receptors (5ꢀHT1A and 5ꢀHT2A).
8
9
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
29
30
31
32
33
34
35
36
37
38
39
40
41
42
43
44
45
46
47
48
49
50
51
52
53
54
55
56
57
58
59
60
Ki ± SEM (nM)a
Compd
R1
R2
R3
D1
D2
D3
5ꢀHT1A
5ꢀHT2A
206.7±7
FW01
(9a)
F
H
34.1%
31.7%
2161±2
51±16
H
F
H
CH3
CH3
H
26.6%
ND
34.7%
ND
73.2%
ND
314±7
542±3
82.29%
51±7
200±16
520.1±11
76.68%
1936±15
1365±42
163±9
9b
9a_1
9b_1
9c
H
ND
ND
ND
Cl
CN
OH
60.35%
9.5%
5.13%
75.76%
43.2%
37.85%
666±33
77.7%
83.99%
H
9.2±1
9d
H
1.1±0.1
9e
F
H
42.3%
19.8%
795±1.3
42±5
155±5
9f
F
F
F
H
H
H
25.3%
34.0%
27.28%
25.2%
24.4%
76.44%
1107±40
409 ±53
83.84%
192±11
88±13
11±2
373±22
163±13
363±94
9g
9h
9i
H
H
F
H
13.1%
22.0%
72.6%
127±2
122±22
9i_1
Cis
O
F
F
H
H
6.0%
54.5%
9.13%
1375±3
52.99%
1.6±1
89.6±12
75.30%
9i_2
9j
H
18.08%
103±3
F
F
F
H
H
H
1.63%
1.48%
43.7%
9.49%
21.8%
46.9%
57.21%
73.3%
51±10
41±4
169±5
230±23
60±15
CH3
9k
9l
255 ±34
3.5±0.1
9m
F
H
H
40.53%
5.13%
74.94%
37.85%
874±7
41±5
971±11
158±9
9n
9o
OH
83.99%
1.01±0.1
ACS Paragon Plus Environment

Journal of Chemical Information and Modeling
Page 26 of 36
1
2
3
4
5
6
7
OH
H
10.08%
1.61%
70.04%
1.05±0.3
350±16
9p
a
Binding data are the mean values of five to six individual experiments with 5ꢀHT1A receptors each
done in triplicate.
8
9
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
29
30
31
32
33
34
35
36
37
38
39
40
41
42
43
44
45
46
47
48
49
50
51
52
53
54
55
56
57
58
59
60
ACS Paragon Plus Environment

Page 27 of 36
Journal of Chemical Information and Modeling
1
2
3
4
Table 4. Results of [35S]GTPγS assays (EC50, nM) for the newly designed 9a derivatives
5
6
7
8
5ꢀHT1AR agonistic activity
Compound
EC50（nM）
106.80
43.63
Emax%
9a
9c
133.14±4.03
124.16±33.47
105.23±22.07
110.42±7.85
86.75±3.47
100.77±25.06
116.83±2.96
105.52±3.39
80.32±1.07
100
9
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
29
30
31
32
33
34
35
36
37
38
39
40
41
42
43
44
45
46
47
48
49
50
51
52
53
54
55
56
57
58
59
60
9d
6.53
9e
24.42
9f
287.11
46.73
9h
9i_2
9m
9o
11.95
29.67
15.81
5ꢀHT
1.7016
ACS Paragon Plus Environment

Journal of Chemical Information and Modeling
Page 28 of 36
1
2
3
4
Schemes
5
6
7
8
9
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
29
30
31
32
33
34
35
36
37
38
39
40
41
42
43
44
45
46
47
48
49
50
51
52
53
54
55
56
57
58
59
60
Scheme 1 Synthetic route for the newly designed 9a derivatives
ACS Paragon Plus Environment

Page 29 of 36
Journal of Chemical Information and Modeling
1
2
3
4
5
6
7
8
9
Regents and conditions: (A) dihydropyran, DMAC/4%H2SO4, reflux; (B) CuCN, DMF,
130°C, reflux; (C) CBr4, PPh3, DCM; (D) BBr3, DCM, ꢀ78°C; (E) Ketones, Zn,
AcOH/H2O, reflux; (F) Chloroacetyl chloride, TEA, DCM; (G) Bocꢀpiperazine, K2CO3,
CH3CN,reflux; (H) TFA, DCM (I) K2CO3, CH3CN, reflux, 85oC; (J) NaH, CH3I, DMF, 65
oC, reflux.
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
29
30
31
32
33
34
35
36
37
38
39
40
41
42
43
44
45
46
47
48
49
50
51
52
53
54
55
56
57
58
59
60
ACS Paragon Plus Environment