"Methylene Bridge" to 5-HT3 Receptor Antagonists: Conformationally Constrained Phenylguanidines

Article abstract of DOI:10.1021/acschemneuro.8b00431

Arylguanidines, depending upon their aromatic substitution pattern, display varying actions at 5-HT₃ receptors (e.g., partial agonist, agonist, superagonist). Here, we demonstrate that conformational constraint of these agents as dihydroquinazo

Full text of DOI:10.1021/acschemneuro.8b00431

Subscriber access provided by UNIV OF NEWCASTLE
Article
“Methylene Bridge” to 5-HT3 Receptor Antagonists:
Conformationally-Constrained Phenylguanidines"
Ahmed S Abdelkhalek, Genevieve S Alley, Osama I Alwassil, Shailesh Khatri, Philip
D Mosier, Heather L Nyce, Michael M White, Marvin K Schulte, and Malgorzata Dukat
ACS Chem. Neurosci., Just Accepted Manuscript • DOI: 10.1021/acschemneuro.8b00431 • Publication Date (Web): 30 Oct 2018
Downloaded from http://pubs.acs.org on November 4, 2018
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 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 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.
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 44
ACS Chemical Neuroscience
1
2
3
4
5
6
7
8
“Methylene Bridge” to 5-HT3 Receptor Antagonists: Conformationally-
Constrained Phenylguanidines
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
Ahmed S. Abdelkhalek,†,ǂ Genevieve S. Alley,† Osama I. Alwassil,†,€ Shailesh Khatri,‡,‖ Philip D.
Mosier,†,⊥ Heather L. Nyce,§ Michael M. White,§ Marvin K. Schulte,‡ and Małgorzata Dukat*,†
†Department of Medicinal Chemistry, School of Pharmacy, Virginia Commonwealth University,
Richmond, VA 23298, USA
‡Department of Pharmaceutical Sciences, Philadelphia College of Pharmacy, University of the
Sciences, Philadelphia, PA 19104, USA
§Department of Biochemistry and Molecular Biology, College of Medicine, Drexel University,
Philadelphia, PA 19102, USA
*Corresponding author. Tel.: 8048285256; Fax: 8048287625.
E-mail address: mdukat@vcu.edu (M. Dukat)
Address: Department of Medicinal Chemistry, School of Pharmacy, Virginia Commonwealth
University, PO Box 980540, Richmond, VA 23298-0540, USA
ACS Chem Neurosci _revised_Oct29_2018
Current Address: ǂDepartment of Medicinal Chemistry, Faculty of Pharmacy, Zagazig University,
Zagazig P.C. 44519, Egypt; €Department of Pharmaceutical Sciences, College of Pharmacy,
1
ACS Paragon Plus Environment

ACS Chemical Neuroscience
Page 2 of 44
1
2
3
4
5
6
7
8
9
‖
King Saud bin Abdulaziz University for Health Sciences, Riyadh 11426, Saudi Arabia; Department
of Cellular and Integrative Physiology, UT Health San Antonio, San Antonio, TX 78229, USA;
⊥Department of BioMolecular Sciences, School of Pharmacy, University of Mississippi,
University, MS 38677, USA
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
2
ACS Paragon Plus Environment

Page 3 of 44
ACS Chemical Neuroscience
1
2
3
4
ABSTRACT
5
6
7
8
9
Arylguanidines, depending upon their aromatic substitution pattern, display varying actions at 5-
HT3 receptors (e.g., partial agonist, agonist, superagonist). Here, we demonstrate that
conformational constraint of these agents as dihydroquinazolines (such as A6CDQ; 1) results in
their conversion to 5-HT3 receptor antagonists. We examined the SAR of 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
Replacement/removal of any of the guanidinium nitrogen atoms of 1 resulted in decreased
affinity. All three nitrogen atoms of 1 are necessary for optimal binding affinity at 5-HT3
receptors. Introduction of substituents as small as an N2-methyl group abolishes affinity. The
results are consistent with homology modeling/docking studies and binding data from site-
directed mutagenesis studies. Introducing a “methylene bridge” to the arylguanidine structure,
regardless of its functional activity, results in a 5-HT3 receptor antagonist.
KEYWORDS: Dihydroquinazolines, functional activities, binding affinities, site-directed
mutagenesis, SAR, 3D-graphic models
3
ACS Paragon Plus Environment

ACS Chemical Neuroscience
Page 4 of 44
1
2
3
4
INTRODUCTION
5
6
7
8
Despite their wide distribution both in the peripheral and central nervous systems, the role of
ligand-gated ion channel 5-HT3 receptors as therapeutic targets appears to be limited and
underexplored. Currently, only the antiemetic properties of 5-HT3 receptor antagonists have
earned them the status of clinically used agents in the chemotherapeutic treatment of cancer.1
However, recent literature reports might refresh the outlook on the possible therapeutic role of 5-
HT3 receptor ligands and open new avenues for further investigations. In addition to their
postulated involvement in cognition, pain, depression, anxiety, alcohol abuse, schizophrenia, and
inflammation,1, 2 5-HT3 receptors have been shown to play a role in neuroprotection.3 The latter
might be due to a modulatory role of 5-HT3 receptors in neurotransmission and/or signaling
pathways of other endogenous ligands (i.e., acetylcholine, dopamine, GABA and glutamate)4
implicating the 5-HT3 receptor as a beneficial target in the treatment of Alzheimer’s disease,
multiple sclerosis, and stroke.3 Recently, antagonists of 5-HT3 receptors were also indicated as a
potential pharmacological therapy for the treatment of opioid-induced pruritus.5
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
Our laboratory recently identified a novel class of 5-HT3 receptor antagonists.6 An example of
this class is 2-amino-6-chloro-3,4-dihydroquinazoline (A6CDQ; 1). Little is known about the
structure-activity relationships (SAR) of 1.
N2
NH2
N1
N3
The aim of the studies reported here was several-fold: 1) to determine the
necessity of the N1, N2 and N3 nitrogen atoms of A6CDQ for binding at h5-
HT3A receptors; 2) to examine the role of primary amine (i.e., N2) substituents
of A6CDQ on binding at h5-HT3A receptors; 3) to investigate the structural
N
NH
7
5
6
Cl
A6CDQ 1
( )
4
ACS Paragon Plus Environment

Page 5 of 44
ACS Chemical Neuroscience
1
2
3
4
5
6
features of A6CDQ underlying its 5-HT3 receptor antagonist action; and 4) to understand how
A6CDQ and its analogs interact/bind with 5-HT3A receptors at the atomic level.
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
CHEMISTRY
To understand the role of the nitrogen atoms N1, N2, and N3 and their contribution to the
binding affinity at h5-HT3A receptors of the parent 1 we synthesized analogs 2-5 by removing
the N2 atom (i.e., 2) or replacing the N1 and/or N3 atoms by sp3 hybridized carbon atoms (i.e., 3-
5). The parent compound 1 was resynthesized according to a previously reported procedure by
us.6 Des-amino quinazoline analog 2 was obtained in a two-step reaction (Scheme 1). First, the
nitrile group of commercially available 2-amino-5-chlorobenzonitrile (19) was reduced with
7
LiAlH4 to the corresponding diamine 20, followed by cyclization with HCO2H in the presence
of a catalytic amount of H2SO4 to afford the free base of 2 in 38% yield.
Scheme 1. aSynthetic route for preparation of compounds 2 and 7-11.
5
ACS Paragon Plus Environment

ACS Chemical Neuroscience
Page 6 of 44
1
2
3
4
5
NH2
NH2 NH2
ii
. HCl
NH
N
CN
i
6
7
8
9
Cl
Cl
Cl
19
2
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
20
iii
CH3
. HI
NHR
. HI
SCH3
HN
. HI
N
NH
N
NH
N
NH
v
vi
Cl
Cl
Cl
9
10
11
; R = CH3CH2CH2
; R = CH(CH3)2
; R = C6H5CH2
8
21
iv
H3C
CH3
NH
. HCl
N
N
Cl
7
aReagents and conditions: (i) LiAlH4/THF, reflux, 2 h; (ii) (a) HCO2H, H2SO4, reflux, 18 h; (b)
HCl/EtOAc, room temperature; (iii) (a) CS2, KOH, room temperature, 2 h, reflux, 18 h; (b) CH3I,
EtOH, reflux, 2 h; (iv) (a) (CH3)2NH (40% wt in H2O), EtOH, reflux, 18 h; (b) HCl/EtOH, room
temperature, overnight; (v) CH3CH2NH2 (70% wt in H2O), EtOH, reflux, overnight; (vi) RNH2,
MeCN, reflux, 4-8 h.
The des-N3 quinoline analog 3 was obtained from known 6-chloro-3,4-dihydroquinoline-2(1H)-
thione8 via an amination reaction9 with ammonium hydroxide and HgCl2 as Lewis acid in a
THF/DMF (5:1) mixture.
6
ACS Paragon Plus Environment

Page 7 of 44
ACS Chemical Neuroscience
1
2
3
4
5
6
7
8
9
The des-N1 dihydroisoquinoline, analog 4 was obtained in four steps (Scheme 2). Initially, an
acid-catalyzed cyclization reaction of commercially available p-chlorophenylacetonitrile (22)
was carried out with pyrophosphoric acid and paraformaldehyde at 200 °C (oil bath) to afford 7-
chloro-1,2,3,4-tetrahydroisoquinolin-3-one (23) in 72% yield.10 Similar to a literature
procedure,11 thionation of 23 with Lawesson’s reagent in anhydrous THF gave 7-chloro-1,2,3,4-
tetrahydroisoquinolin-3-thione (24). Compound 24 was methylated with iodomethane to give the
corresponding 7-chloro-3-(methylthio)-1,4-dihydroisoquinoline hydroiodide (25). Amination of
25 was achieved by treatment with ammonia to give 4.
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 2. aSynthetic route for compound 4.
. HI
S
NH2
N
O
SCH3
N
. HI
CN
NH
NH
i
ii
iii
iv
Cl
Cl
Cl
Cl
Cl
24
4
25
23
22
aReagents and conditions: (i) Pyrophosphoric acid, paraformaldehyde, 200 °C, 2 h; (ii)
Lawesson’s reagent, THF, reflux, 18 h; (iii) CH3I, MeCN, reflux, 2 h; (iv) (a) NH3, butanol, 200
°C, 1 h.
Compound 5 was obtained in one step by reductive amination of 6-chloro-β-tetralone with
NaCNBH3 and NH4OAc.
7
ACS Paragon Plus Environment

ACS Chemical Neuroscience
Page 8 of 44
1
2
3
4
The N-methyl analog of 1, secondary amine 6, was prepared in a similar manner to 1, except
5
6
N,S-dimethylthioisourea hydroiodide was used instead of S-methylisothiourea sulfate (Scheme
7
8
9
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
Scheme 3. aSynthetic pathway for preparation of compounds 6 and 18.
R'
R'
. HCl
NH
NH
NH
NH
O
N
N
HN
O
SCH3
NH2
i
ii
O
O
+
R'
R
N
R
R
27
30
; R' = CH3
; R' = H
6
18
26
29
; R = 6-Cl, R' = CH3
; R = 5,7-diCl, R' = H
; R = 6-Cl
; R = 5,7-diCl
28
31
; R = 6-Cl, R' = CH3
; R = 5,7-diCl, R' = H
.
a Reagents and conditions: (i) Na2CO3, aq. MeCN (80%), reflux, 24 h; (ii) (a) BH3 THF, reflux,
6 h; (b) HCl/EtOH, room temperature, 1 h.
The tertiary amine 7 and secondary amines 8-11 were synthesized by amination of 21 with
dimethylamine (aq. 40%) and the corresponding primary amines, respectively (Scheme 1). The
unsubstituted dihydroquinazoline analog 1312 and a positional isomer of 1, 5-Cl
dihydroquinazoline 1613 were synthesized according to a literature procedure except that HCl
salts were prepared instead of the reported HI and HBr salts, respectively. Analog 18, the 5,7-
diCl analog of 13, was prepared by a general procedure (Scheme 3) for the preparation of
dihydroquinazolines utilized by us in the preparation of 1.6
All target compounds were prepared as water soluble salts and their structures confirmed by 1H
NMR, and C, H, N analysis.
8
ACS Paragon Plus Environment

Page 9 of 44
ACS Chemical Neuroscience
1
2
3
4
RESULTS AND DISCUSSION
5
6
7
8
9
Radioligand Binding. Figure 1 shows the h5-HT3A receptor affinity of A6CDQ (1) and the
analogs where one (i.e., 2-4) or two (i.e., 5) of the nitrogen atoms have been removed or replaced
by a carbon atom. All three nitrogen atoms of 1 contribute to binding and removal of any one of
them (i.e., 2-4) results in a 5- to 10-fold reduction in affinity (Figure 1). Removal of both ring
nitrogen atoms or, more accurately, replacement of the ring nitrogen atoms N1 and N3 by sp3-
hybridized carbon atoms, resulted in loss of affinity (i.e., 5; Ki > 10,000 nM).
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
Retaining all three nitrogen atoms, N-alkylated analogs of 1 were examined next. The N-
alkylated analogs 6-11 were screened in a primary binding assay and none displayed >26%
inhibition at a concentration of 10,000 nM (Table 1). As shown in Table 1, amine substitution
was not tolerated. Even the simplest N-alkyl analog, 6, lacked affinity (i.e., Ki > 10,000 nM).
The des-Cl analog of 1 (Ki = 209 nM), dihydroquinazoline 13, resulted in a substantial decrease
in affinity (Ki > 10,000 nM) indicating that the presence of a 6-Cl group is essential for 1 to bind
at h5-HT3A receptors. Changing the position of the 6-Cl group of 1 to a 5-Cl group (16, Ki = 718
nM; pKi = 6.14 ± 0.07) or 7-Cl group (15, Ki = 1,975 nM)6 resulted in diminished affinity
compared to the parent 1. A hybrid molecule of 15 and 16, 5,7-diCl dihydroquinazoline 18 also
binds with >2-fold lower affinity (Ki = 589 nM; pKi = 6.2 ± 0.10) than 1. It is apparent that both,
the position and, even more so, the presence of the chloro group are essential for optimal affinity
of 1 at h-5HT3A receptors.
9
ACS Paragon Plus Environment

ACS Chemical Neuroscience
Page 10 of 44
1
2
3
4
5
6
7
8
9
Functional Activity. Recently, we reported that 5-HT3A receptor functional activity in the
arylguanidinium series ligands depends on its substitution pattern.14 The guanidinium side chain
in the arylguanidinium series can exist in a number of conformations (i.e., as rotamers) in which
the guanidinium group is approximately perpendicular to the plane of the aryl group. We tested
one conformational extreme where the guanidinium side chain is constrained via a methylene
bridge to the phenyl ring to form dihydroquinazolines, resulting in near co-planarity of the aryl
and guanidinium moieties, a conformation not energetically feasible for arylguanidines.
Introduction of the methylene bridge to 1214 resulted in a change of function from a partial
agonist to an antagonist as seen with A6CDQ (1)6 (Figure 2). The observed change in function
could be due to the introduction of the methylene bridge, the position of the chloro substituent
(i.e., 1 and 12 might be binding somewhat differently), or a combination of these. To examine
the effect of the position of the chloro group we prepared a positional isomer of A6CDQ (1),
A7CDQ (15). Dihydroquinazoline 15 can be also visualized as a conformationally constrained
analog of the agonist 3-chlorophenylguanidine14 (mCPG; 14). There are two possible ways in
which 14 (Figure 2) can be restricted to form a dihydroquinazoline ring. Thus, A5CDQ (16) with
a 5-chloro substituent was synthesized as well. Although diminished, both A7CDQ (15) and
A5CDQ (16) retained affinity at 5-HT3 receptors and both behaved as antagonists (Figures 2 and
3). Thus, constraint of arylguanidine 1414 to dihydroquinazolines 15 and 16 also resulted in a
change of function from an agonist to an antagonist (Figure 2).
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
Up to this point we were able to show that both a partial agonist and an agonist arylguanidine,
when constrained via a methylene bridge to a quinazoline ring system, yielded functionally
different 5-HT3 ligands, antagonists. Next, we examined the effect of constraint on the
superagonist arylguanidine 1714 and found that introduction of the methylene bridge also
10
ACS Paragon Plus Environment

Page 11 of 44
ACS Chemical Neuroscience
1
2
3
4
5
6
7
8
9
converts superagonist 17 to antagonist 18 (Figures 2 and 3). Finally, to test the functional effect
of the presence of the chloro group, we prepared des-chloro dihydroquinazoline 13 and found
that, although binding with very low affinity (Ki > 10, 000 nM), it retained 5-HT3 receptor
antagonist action (Figures 2 and 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
3-D Molecular Modeling. Prediction of favorable binding modes for the constrained
dihydroquinazoline compounds was achieved using the same modeling protocol that was used
for the unconstrained arylguanidines (see Methods section). The m5-HT3A crystal structure15
was also used for docking studies involving the dihydroquinazoline antagonists, as functional
analysis showed that for this structure, “the channel conformation likely corresponds to that of an
inhibited, non-conducting channel” compared to the recently-solved cryo-EM m5-HT3A
structure, in which “the ligand binding domain adopts a conformation reminiscent of the
unliganded state”.16 A consistent binding mode was identified for the dihydroquinazolines,
which bears both similarities and significant differences compared to the consistent binding
mode previously identified14 for the arylguanidine agonists (Figure 4).
In both the arylguanidine and dihydroquinazoline series, the guanidinium group is predicted to
engage the 5-HT3A receptor in a similar fashion: The N1 and N2 nitrogen atoms form hydrogen
bonds with the side chain carboxylate group of E236, the N2 nitrogen forms a second hydrogen
bond with the T181 side chain hydroxyl group, and the N3 nitrogen atom forms a hydrogen bond
with the backbone carbonyl oxygen atom of S182. Thus, each N–H moiety in the guanidinium
group is engaged in a hydrogen bond. This binding mode is therefore consistent with the current
finding that dihydroquinazoline analogs produced by either 1) removal or replacement of any of
11
ACS Paragon Plus Environment

ACS Chemical Neuroscience
Page 12 of 44
1
2
3
4
5
6
the three nitrogen atoms or 2) N-alkylation, exhibit substantial or complete loss of affinity and
very weak activity.
7
8
9
The way in which the aryl portion of the ligands is predicted to interact with the receptor,
however, differs significantly between the arylguanidine and dihydroquinazoline series. By
bridging the guanidinium group back to the aryl ring, the guanidinium and aryl groups in the
dihydroquinazoline compounds are locked in a near co-planar arrangement. This prevents them
from simultaneously forming the aforementioned hydrogen bond interactions with E236, T181
and S182, and forming a thermodynamically favorable aromatic “sandwich” with residues W183
and W90 in the “aromatic box” region of the 5-HT3A orthosteric binding site. We previously
proposed, based on combined experimental and computational results for arylguanidines, that
such simultaneous interactions involving both parts of the ligand are important for the agonist
activity of the arylguanidines. The dihydroquinazoline compounds are not able to bury their
aromatic moiety in the aromatic box (specifically, between W183 and W90), but instead the
aromatic portion forms edge-to-face aromatic interactions only with W90 and is partially
exposed to the solvent.
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 potentially important factor that may influence the compounds’ ability to interact with E236,
and thus their measured binding affinity, is the basicity of the nitrogen-containing functional
group, as measured by the pKa value of the conjugate protonated acid. Of primary importance is
the major microspecies that exist in the environment of the receptor binding site. The cationic
conjugate acid form, with an additional proton available for H-bonding, is able to form a much
more thermodynamically stable (i.e. favorable) interaction with the E236 carboxylate than the
neutral free base. Although the basic functional groups of the reported compounds, particularly
guanidines and amidines, tend to be inherently very strongly basic (pKa ≥ 12), their inclusion in
12
ACS Paragon Plus Environment

Page 13 of 44
ACS Chemical Neuroscience
1
2
3
4
5
6
7
8
9
certain environments, such as conjugation with an aromatic ring, can dramatically lower their
pKa. For example, 3,4-dihydroquinazoline (pKa = 9.19)17 is nearly four log units lower than that
of 1,4,5,6-tetrahydropyrimidine (pKa = 13.0).18 Thus, although it is possible that a substantial
fraction of the low-affinity neutral form may exist for the compounds reported here, we expect
that the protonated, cationic form will usually predominate at physiological pH = 7.4.
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
The modeling results are also generally consistent with our mutagenesis results (Tables 2 and 3).
Site-directed Mutagenesis Studies. Dihydroquinazoline antagonists 1 and 16 were examined,
and the results were compared to their arylguanidine counterparts 12 (a partial agonist)14 and 14
(a full agonist),14 respectively. Two mutations in loop D were tested, W90F and R92A. The
W90F mutation showed reduced affinity for both arylguanidines by about 10-fold, likely due to
non-optimal interaction with the ligand via formation of pi-interactions with F compared to W in
the putative aromatic “sandwich” (Table 2). The same mutation showed either no difference or a
modest increase in affinity (~7-fold) for dihydroquinazolines 1 and 16, respectively, perhaps due
to a reduction of unfavorable steric interactions in the wild-type receptor, where the aryl portion
of the molecule is located in a crowded location enclosed not only by W90, but also by R92,
Y153, and F226 in a small area near the opening of the “aromatic box” (Figures 4 and S2). The
R92A mutant showed only a small reduction in affinity (Table 2) for all four ligands, which is
consistent with the prediction that, individually, R92 is not expected to have a large influence on
the binding of the ligands due to its location at the edge of the orthosteric binding site. Similarly,
the N128A loop A mutant showed negligible changes in binding affinity for the four ligands
compared to the wild-type receptor (Table 2). Although N128 forms a small part of the
13
ACS Paragon Plus Environment

ACS Chemical Neuroscience
Page 14 of 44
1
2
3
4
5
6
orthosteric binding site, its contribution to the binding of arylguanidines and dihydroquinazolines
is not expected to be great.
7
8
9
In loop C, the conservative F226Y mutation resulted in modestly reduced affinity (5- to 10-fold)
for the arylguanidines and essentially no change in affinity for the dihydroquinazolines (Table 3).
The Y226 side chain hydroxyl group is predicted to be partially solvent exposed (see Figure S2
for the putative location of F/Y226), and because the residue at position 226 on loop C may also
serve as a “lid” controlling access to the orthosteric binding site aromatic box, the presence of
the water-soluble side chain hydroxyl group may cause the lid to exist more often in an open,
solvated conformation. This effect would presumably be less deleterious to the binding affinity
of the dihydroquinazolines than to the arylguanidines because the aryl portion of the
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
dihydroquinazolines is predicted to be near the opening of the aromatic box, where it may
occupy the same space as the lid (including F226) in its closed conformation (see Figure S1).
Preventing complete and effective closure of loop C also provides a plausible mechanism by
which the dihydroquinazolines exert their antagonistic effect. Loop C closure has been predicted
to be related to the degree of agonist activity at the 5-HT3 and other pentameric Cys-loop ligand-
gated ion channels,19 and also appears to be correlated with the ability of agonists to form an
energetically favorable inter-subunit bridge between regions of the primary (including loop C)
and complementary subunits comprising the orthosteric binding site.20 Finally, two mutations at
loop C position 236 (E236A and E236N) were evaluated. All ligands, with the exception of the
weak partial agonist 12, showed substantial reductions in binding affinity to these mutants (10-
to 40-fold), highlighting the importance of this residue in hydrogen bond/salt bridge formation
with the ligands. Compound 12 was not substantially affected by the mutations, which may be
14
ACS Paragon Plus Environment

Page 15 of 44
ACS Chemical Neuroscience
1
2
3
4
5
6
related to the fact that 12 did not bind with high affinity even at the wild-type receptor (pKi =
5.86; Ki = 1,400 nM), suggesting a degree of non-specific binding (Table 3).
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
Conclusions
We have shown that A6CDQ (1) binds with higher affinity than its deconstructed analogs 2-5. It
would appear that all nitrogen atoms of A6CDQ (1) contribute to an optimal receptor interaction
indicating that an intact guanidine moiety is important for optimal binding affinity at h5-HT3A
receptors. Substituents at the terminal N2 nitrogen atom (even a single methyl group) are not
tolerated by h5-HT3A receptors; a primary amine seems optimal for binding at h5-HT3A
receptors. Ring closure (“methylene bridge” formation) of the guanidinium moiety is a
dominating factor in the function of dihydroquinazolines. Furthermore, the antagonist action of
dihydroquinazolines at 5-HT3A receptors is substitution-pattern and substituent-presence
independent.
Our previously-proposed hypothesis that explained the wide range of functional activity seen in
the arylguanidine series (i.e. antagonist, partial agonist, full agonist, and superagonist) posited
that “the arylguanidines that are able to simultaneously engage the primary and complementary
subunits, thus keeping them in close proximity, have greater agonist character whereas those that
are deficient in this ability are antagonists”.14 The inability of the dihydroquinazolines to
completely occupy the orthosteric binding pocket likely hinders their ability to keep the C loop
of the principal subunit in close proximity to the D, E, and F loops of the complementary
subunit, so that the ring-constrained dihydroquinazoline analogs become antagonists, regardless
of the functional activity of their unconstrained arylguanidine counterparts (see Figure 4 and
compare panel A to panel B, C to D, and E to F).
15
ACS Paragon Plus Environment

ACS Chemical Neuroscience
Page 16 of 44
1
2
3
4
5
6
7
8
9
Experimental Section
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
Synthesis. Melting points (°C) were taken in glass capillary tubes on a Mel-Temp or Thomas
Hoover apparatus and are uncorrected.1H NMR were recorded with a Bruker ARX 400 MHz or
Bruker AVANCE III 400 MHz spectrometer, and peak positions are given in parts per million
(ppm) downfield from tetramethylsilane as internal standard. IR spectra were determined using a
Thermo Nicolet iS10 FT-IR. MS were obtained using a Waters Acquity TQD (tandem
quadrupole) spectrometer utilizing electrospray ionization in positive ion mode. Microanalyses
were performed by Atlantic Microlab Inc. (Norcross, GA) for the indicated elements, and results
are within 0.4% of calculated values. Chromatographic separations were performed on silica gel
columns (silica gel 62, 60−200 mesh, Sigma-Aldrich). Flash chromatography was performed on
a CombiFlash Companion/TS (Teledyne Isco Inc., Lincoln, NE) apparatus.
Reactions were monitored by thin-layer chromatography (TLC) on silica gel GHLF plates (250
μm, 2.5 × 10 cm2; Analtech Inc., Newark, DE). Compound 1 HCl was previously synthesized by
our laboratory.6 Quinazoline analogs 2-8 HCl and 9-11 HI were prepared adapting literature
procedures for preparation of similar compounds.
6-Chloro-3,4-dihydroquinazoline Hydrochloride (2)
A catalytic amount of conc H2SO4 (10 drops) was added to a solution of 20 (0.25 g, 1.59 mmol)
in formic acid (5 mL), and the reaction mixture was heated at reflux overnight. The reaction
mixture was cooled to room temperature, quenched by careful addition of 6N NaOH (to pH 11),
16
ACS Paragon Plus Environment

Page 17 of 44
ACS Chemical Neuroscience
1
2
3
4
5
6
7
8
9
and extracted with CHCl3 (3 x 10 mL). The combined organic portion was dried (MgSO4) and
evaporated to dryness under reduced pressure to yield 0.10 g of the free base of 2 as a white
solid: mp 152-156 oC; 1H NMR (DMSO-d6) δ 4.51 (s, 1H, CH2), 6.78 (d, J = 8.40 Hz, 1H, ArH),
7.01 (d, J = 2.00 Hz, 1H, ArH), 7.13 (m, 2H, ArH, NH=CH-NH). The hydrochloride salt was
prepared by addition of a saturated solution of gaseous HCl in EtOAc to yield a white solid that
was recrystallized from EtOH to give 0.10 g (83%) of 2 as a white solid: mp > 300 oC; 1H NMR
(DMSO-d6) δ 4.75 (s, 2H, CH2), 7.17 (d, J = 8.80 Hz, 1H, ArH), 7.32 (d, J = 2.00 Hz, 1H, ArH),
7.39 (dd, J = 2.40, 8.40 Hz,1H, ArH), 8.38 (s, 1H, NH=CH-NH), 10.75 (s, 1H, NH, D2O ex),
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
.
12.35 (s, 1H, NH+, D2O ex); Anal. Calcd for (C8H7ClN2 HCl) C, H, N.
2-Amino-6-chloro-3,4-dihydroquinoline Hydrochloride (3)
A solution of 6-chloro-3,4-dihydroquinoline-2(1H)-thione8 (0.13 g, 0.66 mmol) in anhydrous
THF (5 mL) was added to a solution of NH4OH (5 mL) and HgCl2 (0.18 g, 0.66 mmol) in
anhydrous DMF (1 mL) and the reaction mixture was stirred at room temperature overnight. The
reaction mixture was diluted with CH2Cl2 (15 mL), filtered, and acidified with 6N HCl (to pH
1). The aqueous portion was basified with 1N NaOH (to pH 9) and extracted with EtOAc (3 x
10 mL). The combined organic portion was evaporated under reduced pressure to yield 0.08 g of
the free base of 3 as a pale yellow solid: mp 115-118 oC; 1H NMR (CDCl3) δ 2.39- 2.43 (m, 2H,
CH2), 2.73-2.76 (m, 2H,CH2), 4.22 (br s, 1H, NH), 6.86 (d, J = 8.36 Hz, 1H, ArH), 6.99 (d, J =
2.20 Hz, 1H, ArH), 7.06- 7.08 (dd, J = 2.32, 8.32 Hz, 1H, ArH). The addition of a saturated
solution of gaseous HCl in EtOAc (10 mL) yielded a white solid which upon recrystallization
from EtOH/Et2O gave 0.05 g (55%) of 3 as a white solid: mp 222-225 oC; 1H NMR (DMSO-d6)
δ 2.86-2.94 (m, 4H, 2-CH2), 7.15 (d, J = 8.48 Hz, 1H, ArH), 7.32 (dd, J = 2.40, 8.48 Hz, 1H,
17
ACS Paragon Plus Environment

ACS Chemical Neuroscience
Page 18 of 44
1
2
3
4
ArH), 7.38 (d, J = 2.24 Hz, 1H, ArH), 9.07 (s, 1H, NH, D2O ex), 9.92 (s, 1H, NH, D2O ex),
5
6
12.36 (s, 1H, NH+, D2O ex). Anal. Calcd for (C9H9ClN2·HCl) C, H, N.
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
2-Amino-7-chloro-1,4-dihydroisoquinoline Hydroiodide (4)
A solution of 25 (0.10 g, 0.29 mmol) in n-butanol (5 mL) at 0 oC (ice-bath) was charged with
gaseous NH3. The ice bath was removed and the reaction mixture was stirred at 100 oC for 2 h.
The solution was allowed to cool, concentrated under reduced pressure, the resultant yellow oil
was digested with petroleum ether, and the residue was collected by filtration and dried (MgSO4)
to yield 0.05 g (55%) of 4 as a yellow solid: mp 246-250 oC; 1H NMR (DMSO-d6) δ 3.87 (s, 2H,
CH2), 4.49 (s, 2H, CH2), 7.36-7.42 (m, 2H, ArH), 7.50 (s, 1H, ArH), 8.63 (s, 1H, NH, D2O ex),
9.05 (s, 1H, NH, D2O ex), 9.72 (s, 1H, NH+, D2O ex); HRMS m/z (ESI+) found (M)+ 181.0525,
+
C9H10ClN2 (requires (M)+ 181.0527); Anal. Calcd for (C9H9ClN2·HI.0.1C4H9OH) C, H, N.
2-Amino-6-chlorotetralin Hydrochloride (5)
Sodium cyanoborohydride (0.08 g, 1.33 mmol) was added to a solution of 6-chloro-β-tetralone
(0.20 g, 1.11 mmol) and NH4OAc (0.85 g, 11.07 mmol) in MeOH (15 mL) at room temperature.
The resulting solution was stirred for 37 h at room temperature. The reaction mixture was
acidified with 10% HCl (to pH 2), concentrated under reduced pressure, and then extracted with
CH2Cl2 (2 x 75 mL). The aqueous portion was basified with 6 N NaOH (to pH 10) and
extracted with CH2Cl2 (3 x 75 mL). The combined organic portion was dried (Na2SO4) and
evaporated under reduced pressure to yield 0.11 g of the free base of 5 as a greenish-grey oil.
The hydrochloride salt was prepared by addition of saturated solution of gaseous HCl in absolute
EtOH (10 mL). The resulting solid was collected by filtration and recrystallized from EtOH/Et2O
18
ACS Paragon Plus Environment

Page 19 of 44
ACS Chemical Neuroscience
1
2
3
4
5
6
7
8
9
to yield 0.05 g (38%) of 5 as a pink solid: mp >250 oC; 1H NMR (DMSO-d6) δ 1.69-1.79 (m, 1H,
CH), 2.09-2.13 (m, 1H, CH), 2.74-2.91 (m, 3H, CH), 3.04-3.10 (m, 1H, CH), 3.41-3.45 (m, 1H,
CH), 7.15-7.20 (m, 3H, ArH), 8.29 (br s, 3H, NH3+, D2O ex); Anal. Calcd for
(C10H12ClN·HCl·0.1H2O) C, H, N.
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
2-(N-Methylamino)-6-chloro-3,4-dihydroquinazoline Hydrochloride (6)
.
A complex of BH3 THF (1M, 7 mL) was added in a dropwise manner to a stirred solution of 32
(0.35 g, 1.66 mmol) in anhydrous THF (6 mL) under an N2 atmosphere at 0 oC (ice-bath). The
stirred reaction mixture was heated at reflux overnight, cooled to room temperature, and
quenched by addition of 6N HCl (to pH ~1). The mixture was basified with 6N NaOH (to pH
~13) and extracted with hot CHCl3 (3 x 25 mL). The combined organic portion was dried
(MgSO4) and evaporated under reduced pressure to afford 0.09 g, of the free base of 6 as a white
solid: mp 152-155 oC; 1H NMR (DMSO-d6) δ 2.88 (s, 3H, CH3), 4.45 (s, 2H, CH2), 6.81 (d, J =
8.4 Hz, 1H, ArH), 7.09 (d, J = 2.44 Hz, 1H, ArH), 7.16 (dd, J = 2.52, 8.40 Hz, 1H, ArH). A
solution of the crude product (0.08 g, 0.41 mmol) in EtOAc (10 mL) was allowed to stir at 0 oC
(ice-bath). A saturated solution of HCl gas in absolute EtOAc (10 mL) was added, and the
reaction mixture was stirred at room temperature for 1 h. The solvent was evaporated to yield a
white solid which upon recrystallization from EtOH/Et2O gave 0.06 g (38%) of 6 as a white
solid: mp 230-233 oC; 1H NMR (DMSO-d6) δ 2.90 (d, J = 4.8 Hz, 3H, CH3), 4.49 (s, 2H, CH2),
7.12 (s, 1H,ArH), 7.31(d, J = 2.3 Hz,1H, ArH), 7.34 (s, 1H, ArH), 8.11 (br s, 1H, NH, D2O ex),
8.70 (br s, 1H, NH, D2O ex), 10.87 (br s, 1H, NH+, D2O ex). Anal. Calcd for
(C9H10ClN3·HCl·0.35 H2O) C, H, N.
19
ACS Paragon Plus Environment

ACS Chemical Neuroscience
Page 20 of 44
1
2
3
4
2-(N,N-Dimethylamino)-6-chloro-3,4-dihydroquinazoline Hydrochloride (7)
5
6
7
8
An aqueous solution of dimethylamine (40% w/w; 60 mL) was added to a solution of 21 (1.00 g,
4.70 mmol) in absolute EtOH (10 mL) and the stirred reaction mixture was heated at reflux
overnight. The dimethylamine hydrochloride was collected by filtration and the filtrate was
allowed to cool to room temperature for 1 h. The precipitate was collected by filtration and dried
to yield 0.18 g of a yellow solid: mp 168-170 oC; 1H NMR (DMSO-d6) δ 2.91(s, 6H, 2 x CH3),
4.27 (s, 2H, CH2), 6.64 (d, J = 8.32, 1H, ArH), 6.96 (m, 2H, ArH). A saturated solution of
gaseous HCl in EtOH (5 mL) was added to form a hydrochloride salt as a white solid that was
collected by filtration and recrystalized from EtOH/Et2O to yield 0.13 g (69%) of 7 as a white
solid: mp 280-282 oC; 1H NMR (DMSO-d6)  3.14 (s, 6H, 2-N(CH3)2), 4.46 (s, 2H, CH2), 7.36
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
(m, 3H, ArH), 8.67 (s, 1H, NH), 10.61 (s, 1H, NH+); Anal. Calcd for (C10H12ClN3HCl) C, H, N.
6-Chloro-2-(N-ethylamino)-3,4-dihydroquinazoline Hydroiodide (8)
An aqueous solution of ethylamine (70% w/w; 0.50 mL) was added to a solution of 21 (1.00 g,
4.70 mmol) in absolute EtOH (10 mL) and the stirred reaction mixture was heated at reflux
overnight. The solvent was evaporated under reduced pressure to yield a residue that was
recrystallized from EtOAc to afford 0.32 g (79%) of 8 as a white solid: mp 189-192 oC; 1H NMR
(DMSO-d6) δ 1.16 (t, J = 7.16 Hz, 3H, CH3), 3.26-3.31 (m, 2H, CH2), 4.49 (s, 2H, CH2), 7.08 (d,
J = 8.36 Hz, 1H, ArH), 7.33-7.36 (m, 2H, ArH), 7.94 (s, 1H, NH), 8.40 (s, 1H, NH), 10.28 (s,
.
1H, NH+); Anal. Calcd for (C11H12ClN3 HI) C, H, N.
General procedure for synthesis of compounds (9-11)
20
ACS Paragon Plus Environment

Page 21 of 44
ACS Chemical Neuroscience
1
2
3
4
5
6
7
8
9
In a sealed tube, the appropriate amine (0.34 mmol) was added to a solution of 21 (0.31 mmol) in
anhydrous MeCN, and the reaction mixture was heated at 80 oC overnight, and allowed to cool to
room temperature. The solvent was evaporated to dryness under reduced pressure to yield a
crude solid followed by recrystallization from EtOH/Et2O.
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
2-(N-Propylamino)-6-chloro-3,4-dihydroquinazoline Hydroiodide (9)
Yield: 58%; white solid: mp 176-178 oC; 1H NMR (DMSO-d6) δ 0.91 (t, J = 7.2 Hz, CH3), 1.51-
1.60 (m, 2H, CH2), 3.20 (t, J = 7.2 Hz, CH2), 4.48 (s, 2H, CH2), 7.08 (d, J = 8 Hz, 1H, ArH),
7.32-7.35 (m, 2H, ArH), 7.93 (s, 1H, NH), 8.40 (s, 1H, NH), 10.24 (s, 1H, NH+); Anal. Calcd for
.
(C11H14ClN3 HI) C, H, N.
2-(N-isoPropylamino)-6-chloro-3,4-dihydroquinazoline Hydroiodide (10)
Yield: 48%; white solid: mp 140-142 oC; 1H NMR (DMSO-d6) δ 1.21 (t, J = 6.4 Hz, 2-CH3),
3.82-3.87 (m, 1H, CH), 4.47 (s, 2H, CH2), 7.09 (d, J = 8.4 Hz, 1H, ArH), 7.32-7.39 (m, 2H,
ArH), 7.91 (s, 1H, NH), 8.30 (s, 1H, NH), 10.11 (s, 1H, NH+); Anal. Calcd for
.
(C11H14ClN3 HI.0.1H2O) C, H, N.
2-(N-Benzylamino)-6-chloro-3,4-dihydroquinazoline Hydroiodide (11)
Yield: 62%; white solid: mp 230-233 oC; 1H NMR (DMSO-d6) δ 4.50 (s, 2H, CH2), 4.57 (s, 2H,
CH2), 7.09 (d, J = 8 Hz, 1H, ArH), 7.30-7.45 (m, 7H, ArH), 8.39 (s, 1H, NH), 8.55 (s, 1H, NH),
.
10.47 (s, 1H, NH+); Anal. Calcd for (C15H14ClN3 HI) C, H, N.
2-Amino-3,4-dihydroquinazoline Hydrochloride (13)
21
ACS Paragon Plus Environment

ACS Chemical Neuroscience
Page 22 of 44
1
2
3
4
5
6
7
8
9
The hydrochloride salt of 13, was prepared by addition of a saturated solution of gaseous HCl in
absolute EtOH to its free base12 and isolated as a white solid that upon recrystallization from
EtOH yielded 74% of 13 as an off-white solid: mp 158-160 oC; 1H NMR (DMSO-d6) δ 4.49 (s,
2H, CH2), 6.96-6.98 (d, 1H, ArH), 7.07-7.10 (t, 1H, ArH), 7.17-7.19 (d, 1H, ArH), 7.24-7.27 (t,
1H, ArH), 7.63 (br s, 2H, NH2, D2O ex), 8.51 (br s, 1H, NH, D2O ex), 10.83 (br s, 1H, NH+, D2O
ex); IR (solid, cm-1): 3261, 3088, 2963, 2870, 1671, 1626; Anal. Calcd for (C8H9N3·HCl·0.25
EtOH·0.25H2O) C, H, N.
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
2-Amino-5-chloro-3,4-dihydroquinazoline Hydrochloride (16)
The free base of 1613 was dissolved in absolute EtOH (2 mL) and converted to the hydrochloride
salt by the addition of 1 N HCl/MeOH. Precipitation was forced by adding Et2O; the precipitate
was collected by filtration and recrystallized from EtOH to yield 45% of the desired target as
white crystals: mp 238-239 oC; 1H NMR (DMSO-d6) δ 4.5 (s, 2H, CH2), 6.93 (dd, J = 7.98 Hz, J
= 0.74 Hz, 1H, ArH), 7.18 (dd, J = 8.12 Hz, J = 0.92 Hz, 1H, ArH), 7.28 (t, J = 8.06 Hz, 1H, ArH);
IR (Diamond) cm-1: 1618, 3192 (NH), 3285 (NH2); Anal Calcd for (C8H8ClN3HCl) C, H, N.
2-Amino-5,7-dichloro-3,4-dihydroquinazoline Hydrochloride (18)
A solution of BH3·THF complex (2.9 mL, 1.0 M) was added in a dropwise manner to 2-amino-
5,7-dichloroquinazolin-4(3H)-one (31) under a N2 atmosphere. The green reaction mixture was
heated at reflux for 1 h. A solution of 6 N HCl (1 mL) was added in a dropwise manner at 0 oC
(ice-bath) to hydrolyze the borate complex and excess reagent. The dark-blue suspension was
basified with 6 N NaOH (1.5 mL). The reaction mixture was concentrated under reduced pressure
and the resulting residue was extracted with hot CHCl3 (2 x 15 mL). The crude product
22
ACS Paragon Plus Environment

Page 23 of 44
ACS Chemical Neuroscience
1
2
3
4
5
6
7
8
9
precipitated when the combined organic extracts cooled to room temperature. The combined
organic extracts were filtered. The filtrate was concentrated and the product was recrystallized
from CHCl3. The combined solid was dried in an Abderhalden over toluene heated at reflux for 8
h and yielded 0.09 g (59%) of the free base of 18 as an off-white solid: mp 207-208 oC; 1H NMR
(DMSO-d6): δ 4.34 (s, 2H, CH2), 5.87 (br s, 2H, NH2, D2O ex), 6.41 (br s, 1H, NH, D2O ex), 6.48-
6.49 (d, 1H, ArH), 6.78-6.79 (d, 1H, ArH); IR (solid, cm-1): 3466, 3309, 3162, 1648, 1607.
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
The hydrochloride salt was formed by addition of a saturated solution of HCl in EtOH (5 mL). The
resulting off-white solid was recrystallized from abs EtOH to yield 0.04 g (68%) of 18 as an off-
o
1
white solid: mp 272-273 C; H NMR (DMSO-d6) δ 4.48 (s, 2H, CH2), 7.03-7.04 (d, 1H, ArH),
7.36-7.37 (d, 1H, ArH), 7.85 (br s, 2H, NH2, D2O ex), 8.69 (br s, 1H, NH, D2O ex), 11.22 (br s,
1H, NH+, D2O ex); IR (solid, cm-1): 3251, 3026, 2972, 2915, 2837, 1662, 1618; Anal Calcd for
(C8H7Cl2N3·HCl·0.25 EtOH·0.25 H2O) C, H, N.
2-Amino-5-chlorobenzylamine (20)
A solution of 2-amino-5-chlrobenzonitrile (19) (3.00 g, 19.76 mmol) in anhydrous THF (20 mL)
was added in a dropwise manner to a suspension of LiAlH4 (1.27 g, 33 mmol) in anhydrous THF
(50 mL) under an N2 atmosphere and cooled to 0 oC (ice-bath). The stirred reaction mixture was
heated at reflux for 2 h, cooled to 0 oC (ice-bath), and quenched by the careful addition of H2O (2
mL), 5% NaOH (2 mL), and H2O (6 mL). The white precipitate was collected by filtration and
the filtrate was extracted with EtOAc (3 x 25 mL). The combined organic portion was dried
(MgSO4) and evaporated under reduced pressure to give 2.90 g (94%) of a crude product as a
yellow solid: mp 97-99 oC; 1H NMR (DMSO-d6) δ 1.70 (br s, 2H, NH2), 3.57 (s, 2H, CH2), 5.20
23
ACS Paragon Plus Environment

ACS Chemical Neuroscience
Page 24 of 44
1
2
3
4
(br s, 2H, NH2), 6.59 (d, J = 8.44 Hz, 1H, ArH), 6.94 (dd, J = 2.56, 8.40 Hz, 1H, ArH), 7.08 (d, J
5
6
= 2.52 Hz, 1H, ArH).
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
6- Chloro-2-(methylthio)-3,4-dihydroquinazoline Hydroiodide (21)
A solution of CS2 (2 mL, 31.61 mmol) and KOH (1.18 g, 21.03 mmol) was added to a solution
of 20 (3.30 g, 21.07 mmol) in EtOH (25mL) and the reaction mixture was stirred at room
temperature for 2 h, then heated at reflux for 16 h. The precipitate was collected by filtration and
dried to yield 3.60 g of a yellow solid: mp > 250 oC; 1H NMR (DMSO-d6) δ 4.36 (s, 2H, CH2),
6.93 (dd, J = 2.12, 7.20 Hz, 1H, ArH), 7.21 (s, 1H, ArH), 7.24 (d, J = 2.32 Hz, 1H, ArH), 8.67 (s,
1H, NH), 10.47 (s, 1H, NH). A solution of MeI (3.70 mL, 60.40 mmol) in absolute EtOH was
added to a solution of the yellow solid (3.00 g, 15.10 mmol) in absolute EtOH (50 mL) and the
reaction mixture was heated at reflux for 2 h. The solid was collected by filtration and
recrystallized from EtOH to give 4.00 g (78%) of 21 as a pale-yellow solid: mp 236-240 oC; 1H
NMR (DMSO-d6) δ 2.74 (s, 3H, CH3), 4.72 (s, 2H, CH2), 7.08 (d, J = 8.12 Hz, 1H, ArH), 7.39-
7.42 (m, 2H, ArH); Anal. Calcd for (C9H9ClN2S.HI) C, H, N.
7-Chloro-1,2,3,4-tetrahydroisoquinolin-3-one (23)10
A mixture of 4-chlorophenylacetamide (1.5 g, 10 mmol), paraformaldehyde (0.33 g, 15 mmol)
and pyrophosphoric acid (20 g, 112 mmol) was heated at 200 oC for 1 h, cooled to room
temperature and quenched by the careful addition of saturated aqueous solution of NaHCO3 and
extracted with CHCl3 (3 x 25 mL). The combined organic portion was washed with H2O, dried
(MgSO4), and evaporated to dryness under reduced pressure to yield 1.3 g (72%) of 23 as a
yellow-colored powder: mp 166-170 oC (lit.10 mp 209-211 oC); 1H NMR (DMSO-d6) δ 3.42 (s,
24
ACS Paragon Plus Environment

Page 25 of 44
ACS Chemical Neuroscience
1
2
3
4
2H, CH2), 4.32 (s, 2H, CH2), 7.22 (d, J = 8 Hz, 1H, ArH ), 7.28 (dd, J = 2, 8 Hz, 1H, ArH), 7.39
5
6
(s, 1H, ArH), 8.06 (s, 1H, NH).
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
7-Chloro-1,2,3,4-tetrahydroisoquinolin-3-thione (24)
A solution of Lawesson's reagent (1.3 g, 3.3 mmol) in anhydrous THF (10 mL) was added in a
dropwise manner to a stirred solution of 2310 (0.6 g, 3.3 mmol) in anhydrous THF (10 mL) at
room temperature. The reaction mixture was then heated at reflux overnight, cooled to room
temperature and the solvent was evaporated under reduced pressure to yield a crude product
which was purified by column chromatography (silica gel; CHCl3/MeOH; 100:0 to 80:20) to
afford 0.3 g (47%) of 24 as a yellow solid: mp 256-260 oC; 1H NMR (DMSO-d6) δ 3.92 (s, 2H,
CH2), 4.41 (s, 2H, CH2), 7.27 (d, J = 8 Hz, 1H, ArH ), 7.34 (dd, J = 2, 8 Hz, 1H, ArH), 7.45 (s,
1H, ArH), 10.65 (br s, 1H, NH).
7-Chloro-3-(methylthio)-1,4-dihydroisoquinoline Hydroiodide (25)
Methyl iodide (0.15 mL, 3.00 mmol) was added to a solution of 24 (0.15 g, 0.76 mmol) in
MeCN (20 mL) and the reaction mixture was heated at reflux for 2 h, cooled to room
temperature and the residue was collected by filtration and dried to yield 0.17 g (66%) of 25 as a
pale yellow solid: mp 216-220 oC; 1H NMR (DMSO-d6) δ 2.70 (s, 3H, CH3), 4.16 (m, 2 H, CH2),
4.81 (s, 2H, CH2), 7.34 (d, J = 8.4 Hz, 1H, ArH), 7.43 (d, J = 7.6 Hz, 1H, ArH), 7.54 (s, 1H,
ArH).
6-Chloro-2-(N-methylamino)-3,4-dihydroquinazolin-4-(3H)-one (28)
25
ACS Paragon Plus Environment

ACS Chemical Neuroscience
Page 26 of 44
1
2
3
4
5
6
7
8
9
N,S-dimethylthioisourea hydroiodide (1.17 g, 5.06 mmol) and Na2CO3 (0.58 g, 5.51 mmol) were
added to a stirred solution of isatoic anhydride (1.00 g, 5.06 mmol) in aq. MeCN (25 mL, 80%)
and the resulting yellow solution was heated at reflux for 24 h. The reaction mixture was allowed
to cool to room temperature for 1.5 h. The resulting precipitate was collected by filtration and
washed successively with aq. MeCN (3 x 25 mL; 80%) and H2O (75 mL). The solid was dried
under reduced pressure and recrystallized from anhydrous THF to give 0.90 g (47%) of 28 as a
white solid: mp >300 oC; 1H NMR (DMSO-d6) δ 2.85 (d, J = 4.68 Hz, 3H, CH3), 6.28 (br s, 1H,
NH), 7.28 (d, J = 8.76 Hz, 1H, ArH), 7.57 (dd, J = 2.64, 6.12 Hz, 1H, ArH), 7.80 (d, J = 2.56 Hz,
1H, ArH), 11.20 (br s, 1H, NH).
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
2-Amino-5,7-dichloroquinazolin-4(3H)-one (31). S-Methylthioisourea sulfate (0.30 g, 1.07
mmol) and Na2CO3 (0.11 g, 1.07 mmol) were added to a solution of 4,6-dichloroisatoic
anhydride (0.25 g, 1.07 mmol) (29)21 dissolved in aq MeCN (11 mL, 80%) and heated at reflux
for 23 h. The reaction mixture was allowed to cool to room temperature over a period of 30 min,
H2O (20 mL) was added, and the reaction mixture was allowed to stir for 2 h. The reaction
mixture was filtered and the precipitate was washed with aq 80% MeCN (3 x 15 mL). The solid
was dried in an Abderhalden over toluene heated at reflux for 24 h to yield 0.17 g (70%) of 31 as
a beige solid: mp >300 oC; 1H NMR (DMSO-d6) δ 6.64 (br s, 2H, NH2, D2O ex), 7.13-7.15 (m,
2H, ArH), 11.10 (br s, 1H, NH, D2O ex); IR (solid, cm-1): 3385, 2925, 2257, 2130, 1652.
Radioligand Binding Assay. Assays were carried out with h5-HT3 receptors expressed in
HEK293 cells using [3H]GR65630 as radioligand and zacopride as reference compound.22 In
brief: target compounds were subjected to a primary binding assay in which compounds were
26
ACS Paragon Plus Environment

Page 27 of 44
ACS Chemical Neuroscience
1
2
3
4
5
6
7
8
9
examined at a single concentration (10 μM) in quadruplicate in 96-well plates; compounds that
showed a minimum of 50% inhibition at h5-HT3A receptors were evaluated in secondary
radioligand binding assays to determine equilibrium binding affinity at 5-HT3 receptors. In the
secondary binding assay, selected compounds were tested at 11 concentrations (0.1, 0.3, 1, 3, 10,
30, 100, 300 nM, 1, 3, 10 μM) in triplicate. Ki values were obtained from the following equation:
Ki = IC50/(1+ [radioligand]/Kd).
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
Electrophysiology. Compounds 13, 16 and 18 were evaluated functionally using two-electrode
voltage clamp of Xenopus oocytes expressing murine 5-HT3A receptors using previously
published procedures.6, 14
Oocytes were surgically excised, then defolliculated by shaking for 90 min in 1.5% collagenase
dissolved in Ca2+-free OR-II medium (82.5 mM NaCl, 2.5 mM KCl, 1 mM MgCl2 and 5 mM
HEPES, pH 7.4). cRNA was synthesized in vitro using an appropriate cDNA and the
mMESSAGE MACHINE mRNA synthesis kit (Ambion). Each oocyte was injected with 50 nL
of mRNA (0.3 μg/μL concentration) and incubated at 19 °C for at least 24 h prior to
electrophysiological recording. An oocyte was placed in the oocyte recording chamber and
perfused with ND-96 buffer (96 mM NaCl, 2 mM KCl, 1.8 mM CaCl2, 1mM MgCl2, 2 mM
phosphate, pH 7.4) at a flow rate of 20 mL/min. Two electrodes (1-2 MΩ filled with 3 M KCl)
corresponding to the current and voltage electrodes were inserted and the holding potential was
clamped at −60 mV using a Warner instruments amplifier. Ligands were prepared in ND-96
buffer then injected into the chamber. Current responses were recorded using a Axon Instruments
A/D board and PClamp software. A dose response curve was determined using pooled peak
amplitudes of individual responses and analyzed using nonlinear curve fitting and GraphPad
27
ACS Paragon Plus Environment

ACS Chemical Neuroscience
Page 28 of 44
1
2
3
4
5
6
7
8
9
Prism software. All experiments were repeated a minimum of 4 times using oocytes from at least
2 different frogs. Compounds were initially applied alone, in the absence of 5-HT to determine
if they were capable of activating the receptors (agonist action). Compounds that do not activate
the receptors were further evaluated for their ability to inhibit 5-HT induced responses
(antagonist action). For determination of antagonist Ki values, the IC50 was first determined by
inhibition of 5-HT induced responses by increasing concentrations of the antagonist (compound
13, 16 and 18). The Ki was calculated from the IC50 value and 5-HT concentration using the
Cheng−Prusoff relationship. A concentration of 5-HT equal to the EC50 and the EC90 for 5-HT
stimulation of the receptor was used for inhibition and agonist experiments, respectively.
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
Site-Directed Mutagenesis and Ligand Binding. The generation of mutants and analysis of
ligand binding to the homomeric murine 5-HT3A(b) receptor was carried as previously
described.14 Briefly, membranes from transfected cells were incubated for 2 h at 37 C in a total
volume of 0.5 mL of 154 mM NaCl, 20 mM Tris-HCl, pH 7.4 containing the appropriate
concentrations of the competing unlabeled ligand (e.g., mCPG; 14) and radioligand ([3H]
granisetron; PerkinElmer, 85 Ci/mmol). Binding was terminated by rapid vacuum filtration onto
GF/B filters pretreated with 50 mM Tris-HCl, pH 7.4, 0.2% polyethyleneimine and the filters
were washed with 10 mL of cold 50 mM Tris-HCl, pH 7.4 per sample. Nonspecific binding was
defined as that binding not displaced by 10 M mCPBG (meta-chlorophenylbiguanide). IC50
values for the various compounds were determined by fitting the data to Eq. 1 using a
Levenberg–Marquardt algorithm in a commercially available software package (Igor Pro,
WaveMetrics, Oswego, OR):
28
ACS Paragon Plus Environment

Page 29 of 44
ACS Chemical Neuroscience
1
2
3
4
5
(Eq. 1)
)n)1
[I]
  (1 (
IC50
6
7
8
9
where  is the fractional amount of [3H]granisetron bound in the presence of the antagonist at
concentration [I] compared to that in the absence of antagonist, IC50 is the concentration of
antagonist at which =0.5, and n is the apparent Hill coefficient. All experiments were carried
out with a [3H]granisetron concentration equal to its experimentally-determined dissociation
constant for the particular receptor (WT: 2.2 nM; W90F: 11 nM; R92A: 7.5 nM; N128A:1.7 nM;
F226Y: 1.6 nM).
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
Molecular Modeling. Modeling studies were performed as described previously.14 and are
briefly described here. The ligands were sketched using SYBYL-X 2.1.1 (Certara USA, Inc.,
Princeton, NJ) and energy-minimized using the Tripos Force Field (TFF) with Gasteiger–Hückel
charges and a distance-dependent dielectric constant of 4.0 D/Å to an energy gradient cut-off of
0.05 kcal(mol×Å)-1.
Models representing the extracellular ligand binding domains (LBDs) of two adjacent subunits of
the homopentameric mouse 5-HT3A receptor were generated using the A (primary) and E
(complementary) subunits from the m5-HT3A receptor crystal structure (PDB ID: 4PIR).15 The
orthosteric ligand binding site, consisting of loops A–G, was modeled at the interface of subunits
A and E. Residues preceding T35 and following P247 were not included in the models. To model
the flexibility of the C loop, a population of 17 models of the LBD were generated with varying
degrees of C loop openness. Although the receptor modeling was performed using the mouse 5-
29
ACS Paragon Plus Environment