3,4-Dihydro-1,3,5-triazin-2(1H)-ones as the First Dual BACE-1/GSK-3β Fragment Hits against Alzheimer's Disease

Article abstract of DOI:10.1021/acschemneuro.5b00121

One of the main obstacles toward the discovery of effective anti-Alzheimer drugs is the multifactorial nature of its etiopathology. Therefore, the use of multitarget-directed ligands has emerged as particularly suitable. Such ligands, able to modulate different neurodegenerative pathways, for example, amyloid and tau cascades, as well as cognitive and neurogenic functions, are fostered to come. In this respect, we report herein on the first class of BACE-1/GSK-3β dual inhibitors based on a 3,4-dihydro-1,3,5-triazin-2(1H)-one skeleton, whose hit compound 1 showed interesting properties in a preliminary investigation. Notably, compound 2, endowed with well-balanced potencies against the two isolated enzymes (IC₅₀ of 16 and 7 μM against BACE-1 and GSK-3β, respectively), displayed effective neuroprotective and neurogenic activities and no neurotoxicity in cell-based assays. It also showed good brain permeability in a pharmacokinetic assessment in mice. Overall, triazinone derivatives, thanks to the simultaneous modulation of multiple points of the diseased network, might emerge as suitable candidates to be tested in in vivo Alzheimer's disease models.

Full text of DOI:10.1021/acschemneuro.5b00121

Page 1 of 70
ACS Chemical Neuroscience
1
2
3
4
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
ACS Paragon Plus Environment

ACS Chemical Neuroscience
Page 2 of 70
1
2
3
4
5
6
7
8
9
1
1
1
1
1
1
1
1
1
1
2
2
2
2
2
2
2
2
2
2
3
3
3
3
3
3
3
3
3
3
4
4
4
4
4
4
4
4
4
4
5
5
5
5
5
5
5
5
5
5
6
3
,4ꢀDihydroꢀ1,3,5ꢀtriazinꢀ2(1H)ꢀones as the first
dual BACEꢀ1/GSKꢀ3β fragment hits against
Alzheimer’s disease
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
‡
,#
‡,×
‡
‡
‡
Federica Prati, Angela De Simone, Andrea Armirotti, Maria Summa, Daniela Pizzirani,
‡
‡
⊥
×
Rita Scarpelli, Sine Mandrup Bertozzi, Daniel I. Perez, Vincenza Andrisano, Ana Perezꢀ
§
#
#
∞
ǁ
Castillo, Barbara Monti, Francesca Massenzio, Letizia Polito, Marco Racchi, Piera
ǁǁ
‡
⊥
*,‡,#
*,#
Sabatino, Giovanni Bottegoni, Ana Martinez, Andrea Cavalli,
and Maria L. Bolognesi
‡
Department of Drug Discovery and Development, Istituto Italiano di Tecnologia, via Morego
30, 16163 Genova, Italy
#
Department of Pharmacy and Biotechonology, University of Bologna, via Belmeloro 6/Selmi 3,
40126 Bologna, Italy
×
Department for Life Quality Studies, University of Bologna, Corso D’Augusto 237, 47921,
Rimini, Italy
⊥Centro de Investigaciones Biologicas, CIBꢀCSIC, Ramiro de Maetzu 9, 28040 Madrid, Spain.
§
Instituto de Investigaciones Biomédicas, CSICꢀUAM, Arturo Duperier, 4, 28029 Madrid, Spain
and Centro Investigación Biomédica en Red sobre Enfermedades Neurodegenerativas
(CIBERNED)
∞
Fondazione Golgi Cenci, Corso San Martino 10, 20081 Abbiategrasso, Italy
ACS Paragon Plus Environment
1

Page 3 of 70
ACS Chemical Neuroscience
1
2
3
4
5
6
7
8
9
ǁ
Department of Drug SciencesꢀPharmacology, University of Pavia, viale Taramelli 12, 27100
Pavia, Italy
ǁǁ
Department of Chemistry “Giacomo Ciamician”, University of Bologna, via Selmi 2, 40126
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
Bologna, Italy
ABSTRACT
One of the main obstacles toward the discovery of effective antiꢀAlzheimer drugs is the
multifactorial nature of its etiopathology. Therefore, the use of multitargetꢀdirected ligands has
emerged as particularly suitable. Such ligands, able to modulate different neurodegenerative
pathways, e.g. amyloid and tau cascades, as well as cognitive and neurogenic functions, are
fostered to come. In this respect, we report herein on the first class of BACEꢀ1/GSKꢀ3β dual
inhibitors based on a 3,4ꢀdihydroꢀ1,3,5ꢀtriazinꢀ2(1H)ꢀone skeleton, whose hit compound 1
showed interesting properties in a preliminary investigation. Notably, compound 2, endowed
with wellꢀbalanced potencies against the two isolated enzymes (IC of 16 and 7 ꢁM against
5
0
BACEꢀ1 and GSKꢀ3β, respectively) displayed effective neuroprotective and neurogenic
activities and no neurotoxicity in cellꢀbased assays. It also showed good brain permeability in a
pharmacokinetic assessment in mice.
Overall, triazinone derivatives, thanks to the simultaneous modulation of multiple points of the
diseased network might emerge as suitable candidates to be tested in in vivo AD models.
KEYWORDS Alzheimer’s disease, 6ꢀaminoꢀ3,4ꢀdihydroꢀ1,3,5ꢀtriazinꢀ2(1H)ꢀone, drug design,
multitargetꢀdirected ligands, multitarget drug discovery, βꢀsecretase, glycogenꢀsynthase kinaseꢀ
3β.
ACS Paragon Plus Environment
2

ACS Chemical Neuroscience
Page 4 of 70
1
2
3
4
5
6
7
8
9
1
1
1
1
1
1
1
1
1
1
2
2
2
2
2
2
2
2
2
2
3
3
3
3
3
3
3
3
3
3
4
4
4
4
4
4
4
4
4
4
5
5
5
5
5
5
5
5
5
5
6
INTRODUCTION
Alzheimer’s disease (AD) is the most common form of ageꢀrelated dementia, and the one with
the strongest societal impact for what concerns incidence, prevalence, mortality rate, and cost of
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
1
care. Against this backdrop, government and industry have increased their support for drug
discovery and development. However, despite the past and ongoing massive investments, the
available treatments have only moderate palliative effects and a truly diseaseꢀmodifying drug has
yet to come. The cause for the incredible high attrition rate for AD drug discovery has been
attributed to several factors, including the fact that the AD pathogenesis is not yet fully
2
, 3
understood.
Nevertheless, what is increasingly recognized is that AD is a multifactorial
4
syndrome, characterized by massive deposits of amyloidꢀβ (Aβ) peptide, neurofibrillary tangles
(NTF) of the hyperꢀphosphorylated τ protein (Pꢀτ), inflammatory mediators, and reactive oxygen
5
species (ROS), leading to neuronal death via a complex array of interꢀrelated pathways. On this
basis, only therapeutic tools with a similar complexity and ability to hamper the multiple
6,
7
components of the diseased network might turn effective.
For this reason,
polypharmacological strategies are envisaged as specifically suitable to contrast the complex
nature of AD. In this respect, in 2008 we proposed multitargetꢀdirected ligands (MTDLs),
namely small organic molecules able to hit multiple targets responsible for the underlying
8
neurodegeneration, as promising therapeutic options. Since then, by appreciating challenges and
opportunities, we and others have continued to refine and evaluate multitarget concepts, trying to
9
ꢀ17
develop ever better MTDLs against AD.
A pivotal aspect for the success of a MTDL drug discovery project is the initial selection of two
or more suitable target proteins to start with. In particular, proper targets must not only be
ACS Paragon Plus Environment
3

Page 5 of 70
ACS Chemical Neuroscience
1
2
3
4
5
6
7
8
9
validated for AD, but also belong to different neurodegenerative pathways and/or be involved in
1
8
cognitive and neurogenic functions, thus leading to potential additive or synergistic effects. On
this basis, βꢀsecretase (BACEꢀ1) and glycogenꢀsynthase kinaseꢀ3β (GSKꢀ3β) enzymes have
emerged as ideal candidates for such multitarget approach. Notably, BACEꢀ1 and GSKꢀ3β
belong to the two main pathways of AD that are the amyloid and tau cascade, respectively. Thus,
their activities are deeply involved in AD pathogenesis and progression: BACEꢀ1, the aspartylꢀ
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
19
protease which catalyzes the cleavage of the amyloid precursor protein (APP), and GSKꢀ3β,
2
0
the major kinase responsible for τ hyperꢀphosphorylation, are implicated in the formation of Aβ
21, 22
plaques and NFTs, the two main AD pathological hallmarks.
In addition, GSKꢀ3β has been
proposed as the possible link between Aβ and τ, and it also modulates inflammatory response,
axonal transport and microtubule dynamics impairment, apoptosis, cell cycle deregulation, and
23
adult hyppocampal neurogenesis impairment. Therefore, the simultaneous modulation of both
BACEꢀ1 and GSKꢀ3β, by intervening at crucial points in the neurotoxic pathways might
represent a breakthrough for the treatment of AD.
On these premises, we have preliminary reported on triazinones as the first class of BACEꢀ1 and
2
4
GSKꢀ3β dualꢀinhibitors in the search for innovative diseaseꢀmodifiers against AD.
In particular, we identified the fluorinated derivative 1 as promising hit fragment with balanced
2
4
low micromolar activities against the BACEꢀ1/GSKꢀ3β enzymes (Figure 1). In addition, 1
showed an interesting cellular activity profile in terms of neuroprotection, immunomodulation
and neurogenesis, with no sign of toxicity. It also displayed good brain exposure, a fundamental
property for central nervous system (CNS)ꢀdirected drugs. We also demonstrated that the two
2
4
enatiomers of 2 displayed a similar enzymatic profile, with no enantioselectivity effect. On this
basis and in the pursuit of more effective molecules, we have performed different chemical
ACS Paragon Plus Environment
4

ACS Chemical Neuroscience
Page 6 of 70
1
2
3
4
5
6
7
8
9
1
1
1
1
1
1
1
1
1
1
2
2
2
2
2
2
2
2
2
2
3
3
3
3
3
3
3
3
3
3
4
4
4
4
4
4
4
4
4
4
5
5
5
5
5
5
5
5
5
5
6
modifications of the triazinone core, providing compounds 3ꢀ34 (Figure 2). Therefore, in this
paper, we delineate the general structureꢀactivity relationships (SAR) of 2ꢀ34 against BACEꢀ1
and GSKꢀ3β and outline the pharmacological and pharmacokinetic profile of most promising
derivatives.
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
RESULTS AND DISCUSSION
Design. In multitarget drug discovery, fragmentꢀbased strategies have been reported to play a
2
5
pivotal role. Indeed, small fragments, which could be grown by stepꢀwise addition of functional
groups, are good starting points for the development of MTDLs. This is based on the assumption
that the lower the complexity of a molecule, the higher its probability to interact with multiple
2
6
biological targets. With these concepts in mind, we exploited a fragmentꢀbased approach to
2
4
design the dual BACEꢀ1/GSKꢀ3β inhibitors 1 and 2. In particular, we aimed to combine in a
single scaffold the pharmacophoric features responsible for binding to BACEꢀ1 and GSKꢀ3β,
such as a guanidino motif and a cyclic amide group, respectively (Figure 1). The guanidino
moiety, common to several BACEꢀ1 inhibitors, such as acylguanidines and aminoimidazoles
2
7, 28
may bind to the catalytic aspartic dyad of BACEꢀ1.
Whereas, the amino and carbonyl
functionalities of the cyclic amide group may act as Hꢀbond donor and acceptor, respectively,
thus forming Hꢀbond interactions with the backbone of GSKꢀ3β hinge region. This cyclic amide
function, present in numerous ATPꢀcompetitive inhibitors of GSKꢀ3β, that is indirubines,
maleimides, and paullones, among others, seems to play a key role in the kinase binding,
29
providing a specific Hꢀbond network.
As a result, the 6ꢀaminoꢀ4ꢀphenyl triazinone scaffold was identified by means of molecular
modelling studies as potential starting point with the aforementioned structural features (Figure
1). On this basis, a preliminary SAR exploration of the triazinone core with respect to the
ACS Paragon Plus Environment
5

Page 7 of 70
ACS Chemical Neuroscience
1
2
3
4
5
6
7
8
9
substituent on the paraꢀposition of the phenyl ring (F, CF , Br, CH , N(CH ) ), led to the 4ꢀ
3
3
3 2
2
4
fluoro derivative 1. With 1 in hand, herein we have expanded the aromatic substitution pattern,
and a new set of twenty five 6ꢀamino triazinones (3ꢀ27, Figure 2) was developed. In particular,
various electronꢀwithdrawing and electronꢀdonating groups at the ortho (o), meta (m) and para
(p) positions of the phenyl ring, as well as the replacement with other heteroaromatic nuclei were
investigated. The substituents were selected for both SAR and aqueous solubility purposes.
Notably, to cover this latter aspect, compounds 18ꢀ21, bearing a polar group on the aromatic
ring, such as diethylaminoꢀ, morpholinoꢀ, piperidinoꢀ, and 4ꢀmethylꢀpiperazinoꢀmethyl
functions, were synthesized. These solubilizing moieties were carefully selected among those
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
0
structural elements more frequently employed for the design or optimization of a CNS drug.
Afterwards, we planned the isosteric replacement of the 3,4ꢀdihydroꢀtriazinone carbonyl oxygen
with a sulphur atom, to access a subset of 3,4ꢀdihydroꢀtriazinthiones 28ꢀ31 (Figure 2).
In addition, on the basis of the wellꢀbalanced activities of the 6ꢀethylamino derivative 2 against
both targets, the introduction of different Nꢀalkyl and Nꢀaryl groups on the exocyclic amino
group at position C6 was also envisaged, providing derivatives 32ꢀ34 (Figure 2).
24
Considering the lack of stereorecognition by the target enzymes of enantiomers (ꢀ)ꢀ2 and (+)ꢀ2,
all compounds were synthesized and tested as racemates.
Chemistry. Compounds 3ꢀ27 were synthesized following the twoꢀstep synthetic route optimized
2
4
for the synthesis of 1 (Scheme 1). First, the acidꢀcatalyzed hydration of 1ꢀcyanoguanidine 35
provided guanylurea sulfate 36 in excellent yield. Subsequently, 36 was coupled to the aromatic
3
1
aldehydes of interest (37ꢀ61) through a condensation reaction, affording the target compounds
3
ꢀ27 in poor to moderate yields (10ꢀ80%).
ACS Paragon Plus Environment
6

ACS Chemical Neuroscience
Page 8 of 70
1
2
3
4
5
6
7
8
9
1
1
1
1
1
1
1
1
1
1
2
2
2
2
2
2
2
2
2
2
3
3
3
3
3
3
3
3
3
3
4
4
4
4
4
4
4
4
4
4
5
5
5
5
5
5
5
5
5
5
6
All the aromatic aldehydes used in the cyclization were purchased from commercial vendors,
with the exception of 4ꢀ(aminomethyl)benzaldehydes 52ꢀ55. These were prepared following a
3
2
reported three stepꢀsynthetic protocol, slightly modified (Scheme 2). The commercially
available methyl 4ꢀ(bromomethyl)benzoate (62) underwent nucleophilic substitution with the
suitable amines 63ꢀ66 to provide methyl 4ꢀ(aminomethyl)benzoates 67ꢀ70. Subsequent reduction
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
with LiAlH provided the corresponding alcohols 71ꢀ74, which were then oxidized under Swern
4
conditions to afford aldehydes 52ꢀ55 in good yield. Notably, this synthetic pathway resulted
more efficient with respect to the attempted singleꢀstep reduction of 67 to 52 by lithium
bis(diethylamino)aluminum hydride (not shown).
6ꢀAminoꢀ4ꢀarylꢀ3,4ꢀdihydroꢀ1,3,5ꢀtriazinthiones 28ꢀ31 were synthesized by treatment of the
commercially available 2ꢀiminoꢀ4ꢀthiobiuret 75 with benzaldehydes 76, 42, 77, and 49,
according to the aforementioned cyclization procedure (Scheme 3).
6ꢀalkyl/arylaminoꢀ4ꢀarylꢀ3,4ꢀdihydroꢀ1,3,5ꢀtriazinꢀ2(1H)ꢀones 32ꢀ34 were prepared following
the twoꢀstep synthetic route described for 3ꢀ27. However, in this case an additional step was
necessary to provide the starting alkylcyanoguanidines 78ꢀ80 (Scheme 4). To this end, the
commercially available sodium dicyanamide 81 was treated with the amines of interest 82ꢀ84 to
afford the corresponding alkylcyanoguanidines 78ꢀ80. These were then converted to
alkylguanylurea derivatives 85ꢀ87 by the developed acidꢀcatalyzed hydration and then coupled
with the aldehyde of interest 49 and 77 in concentrated H SO providing 32ꢀ34.
2
4,
Xꢀray structure of 2. Concerning the possibility of different tautomeric forms, that could
generate different Hꢀbonding patterns, single crystal Xꢀray investigations on 2 were performed.
They unambiguously showed that two independent molecules (A and B) are present in the
ACS Paragon Plus Environment
7

Page 9 of 70
ACS Chemical Neuroscience
1
2
3
4
5
6
7
8
9
asymmetric unit, plus a water molecule (crystallization medium) strictly connected to both of
them (see Supporting Information and Figure S1).
The possibility of different conformers and tautomeric forms lands support to the high molecular
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
25, 33
versatility of the triazinone fragment and to its potential as a multitarget ligand.
Biological evaluation. We have envisaged MTDLs as the most powerful options for the
treatment of AD, a multifactorial syndrome that currently available singleꢀtarget drugs cannot
cure. The MTDLs developed herein were rationally designed to hit two main ADꢀrelated targets,
that is BACEꢀ1 and GSKꢀ3β enzymes. Thus, by intervening at two crucial points of the amyloid
and tau network, they may display a truly diseaseꢀmodifying effect against AD. To this end, very
3
4ꢀ36
recently Aβ and tau antiꢀaggregating agents
and dual inhibitors of the tau kinase Dyrk 1A
3
7
and of Aβ aggregation have been reported.
To disclose the proposed dualꢀtarget profile for the newly synthesized compounds, a number of
assays were performed, including BACEꢀ1 and GSKꢀ3β inhibition in biochemical assays, as well
as the cellular neuroprotective and neurogenic effects mediated by the inhibition of the two
enzymes. Furthermore, brain permeation was preliminary assessed in vitro in parallel artificial
membrane permeability assay (PAMPAꢀBBB), and in vivo.
Enzymatic profile. As a primary screening, the ability of compounds 3ꢀ34 to inhibit both
BACEꢀ1 and GSKꢀ3β activities was investigated in comparison to inhibitor IV and SB415286 as
reference compounds for BACEꢀ1 and GSKꢀ3β, respectively, as well as the parent compounds 1
and 2. The antiꢀβꢀsecretase potential of 3ꢀ34, was determined in a biochemical assay based on
the cleavage of a peptide substrate mimicking the human APP sequence with the Swedish
mutation (methoxycoumarinꢀSerꢀGluꢀValꢀAsnꢀLeuꢀAspꢀAlaꢀGluꢀPheꢀLysꢀdinitrophenyl), using
38
the fluorescence resonance energy transfer (FRET) methodology. GSKꢀ3β biochemical
ACS Paragon Plus Environment
8

ACS Chemical Neuroscience
Page 10 of 70
1
2
3
4
5
6
7
8
9
1
1
1
1
1
1
1
1
1
1
2
2
2
2
2
2
2
2
2
2
3
3
3
3
3
3
3
3
3
3
4
4
4
4
4
4
4
4
4
4
5
5
5
5
5
5
5
5
5
5
6
inhibition was assessed using the KinaseꢀGlo luminescent assay, which quantifies the decrease in
3
9
ATP levels following the kinase reaction. IC values on the two enzymes are reported in Table
50
1, and were determined by using the linear regression parameters.
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
Considering the 6ꢀaminoꢀ4ꢀarylꢀtriazinone series (3ꢀ27), all tested compounds modulated GSKꢀ
3β activity in the doubleꢀdigit micromolar range (10.14 ꢁM ≤ IC50 ≤ 57.65 ꢁM), with the
exception of 4ꢀ((diethylamino)methyl)phenyl (18) and 4ꢀ((Nꢀmethylpiperazino)methyl)phenyl
(
21) derivatives, which resulted less effective GSKꢀ3β inhibitors (IC > 100 ꢁM). These results
50
indicate that no significant improvement in GSKꢀ3β inhibitory activity was achieved with this
new series of compounds, with derivatives 8 (2ꢀF), 15 (2ꢀCH ) and 23 (3,4ꢀdiCl) being
3
equipotent to hits 1 and 2.
With regards to BACEꢀ1, the 6ꢀaminoꢀ4ꢀ(oꢀ, mꢀ, pꢀhalogen)ꢀsubstituted triazinones 6ꢀ14, the 3,4ꢀ
dichloroꢀ and 3,5ꢀdifluoroꢀsubstituted derivatives 23 and 24, and the 3ꢀpyridinyl analogue 26
showed interesting IC50 values, ranging from 10.18 to 84.72 ꢁM. Notably, the 3ꢀFꢀphenyl
derivative 9 resulted the most active βꢀsecretase inhibitor, with an IC of 10.18 ± 1.02 µM,
5
0
which is even slightly improved with respect to the 4ꢀFꢀphenyl derivative 2.
Generally, triazinthiones 28ꢀ30 were active in the micromolar range against both enzymes (13.78
ꢁ
M ≤ IC ≤ 39.00 ꢁM for GSKꢀ3β; 43.15 ꢁM ≤ IC ≤ 60.19 ꢁM for BACEꢀ1) and showed
50 50
similar or higher activities compared to the corresponding triazinones. Therefore, the
oxygen/sulfur isosteric replacement at position 2 of the triazinone core only marginally affected
activity.
Nꢀalkyl/Nꢀaryl derivatives 32ꢀ34 displayed low micromolar activities against GSKꢀ3β (4.34 ꢁM
≤
IC ≤ 40.71 ꢁM). Remarkably, 33 and 34 together with the parent compound 2 showed IC
50 50
values in the singleꢀdigit micromolar range, with the propyl derivative 33 as the most potent
ACS Paragon Plus Environment
9

Page 11 of 70
ACS Chemical Neuroscience
1
2
3
4
5
6
7
8
9
GSKꢀ3β inhibitor of the whole series (IC = 4.34 ꢁM). As for BACEꢀ1, 32 and 33 turned out to
5
0
be moderate inhibitors, with IC50 ranging from 36 to 50 ꢁM, whereas 34 did not show any
significant effect up to a concentration of 100 µM. It can be derived that alkylation and arylation
of the 6ꢀamino group of the triazinone core slightly improve the antiꢀGSKꢀ3β potency, probably
due to the generation of weak hydrophobic interactions within the binding pocket of the enzyme.
Conversely, the introduction of the same substituents, especially the aryl moiety of 34,
negatively affects BACEꢀ1 inhibitory activity. This is likely due to the fact that these substituents
generate steric and electronic effects (strongly influencing the amino group basicity) that might
hamper BACEꢀ1 aspartyl dyad recognition.
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
Overall, the collected SAR results revealed that, despite the number of synthesized congeners,
the parent compounds 1 and 2 are still the most promising from a multitarget standpoint, with a
24
balanced enzymatic profile and good ligand efficiency (LE) metrics against the two targets. In
particular, LE metrics, that is measures of the in vitro biological activity corrected for the
physicochemical property ‘load’ of a molecule, account for the effective use of the molecule’s
40
structural features in binding to the target. In this respect, LE metrics has recently found
increasing application in fragmentꢀbased drug discovery as useful tool to normalize the low
binding affinities (1 mM to 10 ꢁM) of fragment hits and to identify and prioritize the best
4
0
starting points for a subsequent fragmentꢀtoꢀlead campaign. Therefore, despite the moderate
activity, we have proposed that 1 and 2, due to their structural simplicity, low molecular weight,
and favorable LE values, might be suitable binder.
On this basis, 2 was pursued for further cellular studies. Moreover, derivatives 9 and 33, being
the most active BACEꢀ1 and GSKꢀ3β inhibitors of the series, respectively, were also selected for
additional investigations.
ACS Paragon Plus Environment
10

ACS Chemical Neuroscience
Page 12 of 70
1
2
3
4
5
6
7
8
9
1
1
1
1
1
1
1
1
1
1
2
2
2
2
2
2
2
2
2
2
3
3
3
3
3
3
3
3
3
3
4
4
4
4
4
4
4
4
4
4
5
5
5
5
5
5
5
5
5
5
6
Aβ(1ꢀ40) secretion. The effect of 9 and 33 on the secretion of Aβ(1ꢀ40) was examined in
2
4
24
cellular context, in comparison with parent compounds 1 and 2 . The antiꢀβꢀsecretase activity
of the tested compounds in neuroglioma cell line overexpressing the hAPP gene harboring the
KM670/671NL (Swedish) mutation (H4ꢀAPP ) is reported in Figure 3a. H4ꢀAPP cells were
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
sw
sw
treated with 1, 2, 9, and 33 for 24 h at a concentration corresponding to their respective IC50
values on BACEꢀ1. Similarly to what observed for 1 and 2, in the case of 9 enzymatic activity
was mirrored by a moderate reduction of cellular Aβ levels, though with much less effectiveness
with respect to the reference inhibitor (inhibitor IV). As regards to 33, which is the poorest
BACEꢀ1 inhibitor among the selected ones, the observed activity is negligible. As a positive
feature, Figure 3b shows that 1, 2, 9, and 33 did not affect H4ꢀAPP cell viability at the tested
sw
concentrations, as monitored through the colorimetric tetrazolium salt (MTT) assay. So, a low
neurotoxicity for the triazinone chemotype may be anticipated.
Neuroprotection. The effects of 2, 9, and 33 on neuroprotection were also assessed, in
2
4
comparison with 1. In fact, neuroprotection, preserving neuronal structure and functionality
against toxic insults, and thus reducing neuronal loss and degeneration, is a crucial property for
new ADꢀmodifying drugs.
Various proꢀinflammatory stimuli might enhance GSKꢀ3β activity, resulting in increased level of
glia activation and in a peculiar cytokines pattern, ultimately leading to the neuronal cell death in
4
1
AD. Therefore, we explored the potential neuroprotective activity of the selected compounds in
primary cultures of astrocytes and microglia, by evaluating nitrite production.
2
4
As reported for 1, primary cultured glial cells were first incubated with compounds 2, 9, and 33
(10 ꢁM) for 1 h, and then cultured for further 24 h with lipopolysaccharide (LPS) (10 ꢁg/mL), a
potent cytotoxic inducer of inflammation and of a cascade of intracellular events involved in
ACS Paragon Plus Environment
11

Page 13 of 70
ACS Chemical Neuroscience
1
2
3
4
5
6
7
8
9
neuronal death. In LPSꢀtreated cells, we observed an important induction of nitrite production,
which was significantly reduced by treatment with 2 and 33 (Figure 4). Their observed cellular
activities nicely match their inhibitory potencies on GSKꢀ3β isolated enzyme. Indeed, 2 and 33,
displaying enzymatic potency in the singleꢀdigit micromolar range, turned out to be potent
neuroprotective agents at 10 ꢁM concentration. To note, 2 and 33 were mostly effective on
astrocytes, where they significantly drop nitrite production to levels lower than the basal ones.
On the other hand, 9, with an IC of 32.41 ꢁM against GSKꢀ3β, showed modest activity on
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
5
0
microglia, but no effect on astrocytes, where it even caused an increase in nitrite generation.
Importantly, similar results were also obtained by coꢀtreatment of primary rat glial cells with
LPS and compounds 2 and 33 at different concentrations (0, 5, 10, 20 and 50 µM) (Figure S2).
Motivated by these interesting findings, we investigated the observed neuroprotective activities
of 2 and 33 more in depth. In particular, we evaluated the ability of our compounds to modulate
the glial phenotypic switch from the proꢀinflammatory M1 to the antiꢀinflammatory M2 type.
Importantly, the transformation from the neuroprotective M2 to the cytotoxic M1 glial cells is
42
now considered a crucial step in the progression of neurodegenerative diseases, including AD.
Therefore, compounds that prompt the switch from the M1 to M2 form have been proposed as
capable to attenuate neuroinflammation and boost neuronal protection and recovery. In details,
this M1/M2 phenotypic classification is based on a specific pattern of proꢀ or antiꢀinflammatory
cytokines and receptors, whose release and expression is regulated, among others, by GSKꢀ3β
43
activity. In this respect, GSKꢀ3β activation has been reported to foster and maintain the proꢀ
inflammatory state. On this basis, we evaluated the ability of 2 and 33 to modulate the expression
level of the inducible nitric oxide synthase (iNOS) as M1 marker, and the triggering receptor
expressed on myeloid cells 2 (TREM2) as M2 marker on glial cells. iNOS, an inducible enzyme
ACS Paragon Plus Environment
12

ACS Chemical Neuroscience
Page 14 of 70
1
2
3
4
5
6
7
8
9
1
1
1
1
1
1
1
1
1
1
2
2
2
2
2
2
2
2
2
2
3
3
3
3
3
3
3
3
3
3
4
4
4
4
4
4
4
4
4
4
5
5
5
5
5
5
5
5
5
5
6
with a prevailingly glial localization, is expressed upon proꢀinflammatory stimulation and it is
responsible for the elevated NO concentrations associated with the neurodegenerative
4
4
pathology, whereas TREM2 stimulates phagocytosis for Aβ clearance and suppresses cytokine
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
45
release, thus reducing inflammation.
We were pleased to verify that when cultures of primary rat glial cells were stimulated with LPS,
we observed the expected iNOS induction, which was reduced by a 24 h coꢀtreatment with 2 and
33 in a doseꢀdependent manner and to a greater extent than what reported for 1 (Figures 5aꢀd).
Furthermore, along the same line, in microglia cells treated with LPS we observed a reduction of
TREM2 expression, which was restored by 24 h coꢀtreatment with 2 and 33 (Figures 5d and e).
Altogether these results indicate that 2 and 33 are highly promising antiꢀinflammatory and
neuroprotective agents, able to decrease the neurotoxic microglial activation, while not affecting
the neuroprotective one. Moreover, 2 and 33 did not display any toxicity in glial and neuronal
cells up to 50 ꢁM (Figure S3).
Neurogenesis. Neurogenesis is a crucial property for new ADꢀmodifying drugs, since it confers
the potential to increase endogenous regeneration as a repair mechanism in the damaged brain,
and reduce neuronal loss and degeneration. In this respect, considering that GSKꢀ3β inhibition
4
6, 47
has been reported to regulate and increase neurogenesis,
we verified whether addition of 2,
9
, and 33 to neurosphere (NS) cultures of primary rat neural stem cells could regulate cell
differentiation toward a neuronal phenotype. The NSs were cultivated in the absence or presence
of compounds 2, 9, and 33 (10 ꢁM) during a week. After that, they were first incubated with
antiꢀβꢀtubulin and antiꢀmicrotubule associated protein 2 (MAPꢀ2) antibodies, and then treated
with the corresponding labelled secondary antibodies to reveal the immature and mature
neuronal markers βꢀtubulin (green label) and MAPꢀ2 (red label), respectively. Encouragingly,
ACS Paragon Plus Environment
13

Page 15 of 70
ACS Chemical Neuroscience
1
2
3
4
5
6
7
8
9
evident neurogenic effects were detected after treatment of NSs with 2 and 33 (Figure 6).
Particularly, when compared to control and even to parent compound 1, the number of βꢀtubulin
positive cells was considerately augmented in cultures treated with 2, whereas, 33 significantly
amplified the number of MAPꢀ2 positive cells. Moreover, 2 and 33 showed a dense crown of
either βꢀtubulin or MAPꢀ2ꢀpositive cells around the NS core and additional positive cells outside
of it, suggesting a migration out of the NS.
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
Notably, these findings underlined the ability of 2 and 33 to differentiate neural stem cells to
immature and mature neurons, respectively. Furthermore, our results also demonstrate that 2 and
33 can induce migration of cells out of the NSs. These outcomes are particularly relevant in a
clinical setting, since the identification of small molecules that not only promote neural stem cell
differentiation as endogenous regenerative and repair mechanisms, but also affect their migration
capacity might have an important regulatory role in hippocampal migratory events in a brain
4
7
injury context.
Bloodꢀbrainꢀbarrier (BBB) penetration. The BBB penetration has been considered as a major
bottleneck in CNS drug development and as an important factor limiting the future growth of
4
8
neurotherapeutics. Therefore, to reduce attrition rate, the brain permeability of new CNSꢀ
directed molecules needs to be evaluated in the early drug discovery stage. Considering that the
majority of CNS drugs enter the brain by transcellular passive diffusion, PAMPAꢀBBB has been
proposed as a useful high throughput technique to predict passive permeability through
49
biological membranes. Hence, prediction of brain penetration for compounds 1ꢀ34 was
evaluated in PAMPAꢀBBB, using a brain lipid porcine membrane. First, an assay validation
study was carried out by testing ten commercial drugs and comparing the obtained permeability
(Pe) values with the corresponding literature data. A good linear correlation between the two
ACS Paragon Plus Environment
14

ACS Chemical Neuroscience
Page 16 of 70
1
2
3
4
5
6
7
8
9
1
1
1
1
1
1
1
1
1
1
2
2
2
2
2
2
2
2
2
2
3
3
3
3
3
3
3
3
3
3
4
4
4
4
4
4
4
4
4
4
5
5
5
5
5
5
5
5
5
5
6
2
data set was obtained: Yexperimental = 1.1999Xliterature ꢀ 0.8494 (R = 0.974) (Figure S4). According
to this equation and following the pattern established in the literature for BBB permeation
5
0
prediction, we could classify compounds as able to enter the brain when they present Pe > 3.95
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
ꢀ6
ꢀ1
×
10 cm s . Unfortunately, all our compounds, but 22 and 34, showed low Pe and were
predicted not to penetrate the CNS (Table S1). However, considering that 1 showed an
2
4
interesting in vivo pharmacokinetic profile, and that triazinones are small and polar molecules,
we reasoned that the PAMPAꢀBBB model is not the right one for the current series. Indeed, it is
highly conceivable that they do not cross BBB by passive diffusion, but rather might exploit
51
membrane specific transporters to enter the brain, i.e. the guanidine compound trasporters.
Accordingly, we generated mouse pharmacokinetic data for compound 2. Notably, 2 showed
good BBB penetration, confirming that triazinones may rely on different mechanisms rather than
passive diffusion to enter the brain.
−
1
In details, 15 min after intraperitoneal dosing of 2 (10 mg kg ), we measured a maximal plasma
−
1
concentration of 695 ng mL , accompanied with an halfꢀlife for the elimination phase of 163
−1
min. 2 showed a volume of distribution of 7.2 L kg and disappeared from the systemic
−
1
circulation with a clearance of 306 mL min kg . Importantly, 2 reached a maximum
concentration in 1 mL of brain homogenate of 1.50 ng/mgprotein, 30 min after administration
(
Table 2). On this basis we could approximately estimate that 2 reached total cerebral levels of
1.34 µM, which, although not ensuring in vivo target engagement, represent a promising starting
point towards an efficacious hit optimization campaign (details of the calculation are reported in
the Experimental Section). Indeed, effective in vivo concentrations could be reached by lowering
IC values on both enzymes just around 10 folds or even less, considering that a peritoneal
5
0
−1
administration of 10 mg kg dose is not very high, and it can be increased.
ACS Paragon Plus Environment
15

Page 17 of 70
ACS Chemical Neuroscience
1
2
3
4
5
6
7
8
9
CONCLUSIONS
Herein, we described a new series of derivatives belonging to the first class of dual inhibitors of
BACEꢀ1/GSKꢀ3β enzymes, two of the most validated targets in AD drug discovery. The
remarkable neuroprotective and neuroregenerative profile shown by 2 is undoubtedly the most
promising achievement of the current investigation, and support the idea that balanced, although
relatively low inhibitory potencies are key molecular features for such dual inhibitors.
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
This is reinforced by the suggestion that a partial inhibition of BACEꢀ1 and GSKꢀ3β might
provide clinical benefits with limited side effects. Heterozygous BACEꢀ1 knockout APP
transgenic mice with an only 15% reduction in Aβ cerebral level showed a significant reduction
5
2
in brain amyloid burden at old age. As for GSKꢀ3β, considering that it is generally upꢀregulated
in neurodegenerative conditions, a similar mild inhibition would be enough to produce an
5
3
important therapeutic effect. Particularly, a smooth inhibition of GSKꢀ3β would normalize its
activity in the diseased tissues, without significantly affecting the healthy ones, where
54
compensatory mechanisms will likely balance the deficit.
More importantly, a basic concept of polypharmacology is that where connections exist between
two targets within a network, dual inhibitors with only moderate activities produce superior in
vivo effects compared to higherꢀaffinity singleꢀtargeted compounds, and with potential minor
3
3, 55
toxicity.
Indeed, a crossꢀtalk between the BACEꢀ1 and GSKꢀ3β has been clearly demonstrated. BACEꢀ1
fosters Aβ generation, which exerts its neurotoxic effects in a variety of ways. Particularly, Aβ
5
6
has been reported to induce GSKꢀ3β activation, thus increasing τꢀphosphorylation and NFT
toxicity, neuroinflammation, apoptosis, cell cycle deregulation, and impairment of adult
ACS Paragon Plus Environment
16

ACS Chemical Neuroscience
Page 18 of 70
1
2
3
4
5
6
7
8
9
1
1
1
1
1
1
1
1
1
1
2
2
2
2
2
2
2
2
2
2
3
3
3
3
3
3
3
3
3
3
4
4
4
4
4
4
4
4
4
4
5
5
5
5
5
5
5
5
5
5
6
hippocampal neurogenesis. What’s the more, GSKꢀ3β not only responds to Aβ peptide, but it
5
7ꢀ59
also regulates its accumulation, by modulating αꢀ, βꢀ and γꢀsecretase enzymatic activities,
thus establishing a vicious feedꢀforward loop.
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
In addition, the conformational flexibility of 2 highlighted by the crystal structure elucidation
expands the multitarget potential of this fragment.
If these concepts are arguable in a multitarget drug discovery perspective, we should remark that
BACE inhibitors currently in Phase III for AD, that is AZD3293 or LY3314814 and MK 8931,
display an inhibitory potency in the (sub)ꢀnanomolar range. Thus, we are confident that
triazinone derivatives might serve as the basis for developing suitable candidates to be tested in
in vivo model of AD, following a further round of multitarget activities and pharmacokinetic
properties optimization.
METHODS
Chemistry. All the commercial available reagents and solvents were used as purchased from
SigmaꢀAldrich, Fluka (Italy) and Alfa Aesar (Germany) without further purification.
CEM Discover SP focused microwave reactor was used for microwaveꢀmediated reactions.
Column chromatography purifications were performed under “flash conditions” using Sigmaꢀ
Aldrich silica gel grade 9385, 60 Å, 230−400 mesh. Thin layer chromatography (TLC)
separations were performed on 0.20 mm silica gel 60 F254 plates (Merck, Germany), which
were visualized by exposure to ultraviolet light (254 and 366 nm), and potassium permanganate
stain. Reactions involving generation or consumption of amine were visualized by using
bromocresol green spray (0.04% in EtOH made blue by NaOH) following heating of the plate.
ACS Paragon Plus Environment
17

Page 19 of 70
ACS Chemical Neuroscience
1
2
3
4
5
6
7
8
9
Compounds were named following IUPAC rules as applied by ChemBioDraw Ultra (version
1
3.0).
The purifications by preparative high performance liquid chromatographyꢀmass spectrometry
HPLC/MS) were run on a Waters Autopurification system consisting of a 3100 single
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
(
quadropole mass spectrometer (SQDꢀMS) equipped with an electrospray ionization (ESI)
interface and a 2998 photodiode array detector (PDA). The HPLC system included a 2747
sample manager, 2545 binary gradient module, system fluidic organizer and 515 HPLC pump.
TM
The separations were performed on a XBridge prep C18 OBD column (100 × 19 mm ID,
TM
particle size 5 µm) with a XBridge prep C18 (10 × 19 mm ID, particle size 5µm) guard
cartridge, using 10 mM NH OAc in H O at pH 5 adjusted with AcOH (A) and 10 mM NH OAc
4
2
4
in MeCNꢀH O (95:5) at pH 5 (B) as mobile phase. A linear gradient was applied starting at 0% B
2
(initial hold for 0.5 min) to 40% B in 7 min. From 40% to 100% B in 0.1 min and hold at 100%
for 2.4 min. The PDA range was 210ꢀ400 nm. ESI in positive mode was used in the mass scan
range 100ꢀ500 Da.
Nuclear magnetic resonance (NMR) experiments were run on Varian VXR 200 and 400 and
1
13
Bruker Avance III 400 (200 and 400 MHz for H; 50 and 100 MHz for C). Spectra were
1
acquired at 300 K, using DMSOꢀd , CD OD, D O and CDCl as solvents. Chemical shifts for H
6
3
2
3
1
3
and C spectra were recorded in parts per million (ppm) using the residual nonꢀdeuterated
solvent as the internal standard (I.S). Data are reported as follows: chemical shift (ppm),
multiplicity (indicated as: s, singlet; br s, broad singlet; exch, exchangeable proton with D O; d,
2
doublet; t, triplet; q, quartet; m, multiplet and combinations thereof), coupling constants (J) in
Hertz (Hz) and integrated intensity.
ACS Paragon Plus Environment
18

ACS Chemical Neuroscience
Page 20 of 70
1
2
3
4
5
6
7
8
9
1
1
1
1
1
1
1
1
1
1
2
2
2
2
2
2
2
2
2
2
3
3
3
3
3
3
3
3
3
3
4
4
4
4
4
4
4
4
4
4
5
5
5
5
5
5
5
5
5
5
6
Ultra performance liquid chromatographyꢀmass (UPLCꢀMS) analyses were run on a Waters
ACQUITY UPLCꢀMS system consisting of SQDꢀMS equipped with an ESI interface and a PDA
detector. PDA range was 210ꢀ400 nm. Analyses were performed on an ACQUITY UPLC HSS
T3 C18 column (50 mm × 2.1 mm ID, particle size 1.8 ꢁm) with a VanGuard HSS T3 C18
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
precolumn (5 mm × 2.1 mm ID, particle size 1.8 ꢁm). Mobile phase was 10 mM NH OAc in
4
H O at pH 5 adjusted with AcOH (A) and 10 mM NH OAc in MeCNꢀH O (95:5) at pH 5 (B).
2
4
2
ESI in positive and negative mode was applied, in the mass scan range 100ꢀ500 Da.
Melting points were determined in glass capillary tubes on a Gallenkamp melting point apparatus
and are uncorrected.
All the finals compounds showed ≥95% purity by NMR and UPLCꢀMS (UV at 215nm) analysis.
General procedure (A) for the synthesis of methyl 4ꢀ((amino)methyl)benzoates (67ꢀ70). To
an iceꢀcold solution of methyl 4ꢀ(bromomethyl)benzoate 62 (1.0 eq.) in THF (0.7 M), the proper
amine (63ꢀ66) (4.0 eq.) was added. The reaction mixture was heated at 100 °C under microwave
irradiation for 20 min, affording a white precipitate, which was filtered off. The filtrate was
concentrated under vacuum, and the resulting residue was taken up with 30 mL of 2N aqueous
HCl solution. The aqueous phase was washed with Et O (30 mL × 3), made basic with Na CO ,
2
2
3
and extracted with EtOAc (30 mL × 3). The organic layers were collected, dried over anhydrous
Na SO , filtered, and concentrated under vacuum. The so obtained title compound (67ꢀ70) was
2
4
used in the next step without further purification.
Methyl 4ꢀ((diethylamino)methyl)benzoate (67). The title compound was synthesized according
to general procedure A using diethylamine 63 (2.4 g, 35.00 mmol). 67 was obtained as a yellow
ACS Paragon Plus Environment
19

Page 21 of 70
ACS Chemical Neuroscience
1
2
3
4
5
6
7
8
9
1
oil: 1.73 g (90%). HꢀNMR (CDCl , 400 MHz) δ 1.03 (t, J = 7.2 Hz, 6H), 2.51 (q, J = 7.2 Hz,
3
1
3
4
H), 3.60 (s, 2H), 3.89 (s, 3H), 7.41 (d, J = 8.4 Hz, 2H), 7.97 (d, J = 8.4 Hz, 2H). CꢀNMR
(
CDCl , 100 MHz) δ 11.8, 46.9, 51.8, 57.4, 128.4, 128.5, 129.4, 145.8, 167.0
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
Methyl 4ꢀ(morpholinomethyl)benzoate (68). The title compound was synthesized according to
general procedure A using morpholine 64 (3.0 g, 35.00 mmol). 68 was obtained as a yellow oil:
1
1
.80 g (90%). HꢀNMR (CDCl , 400 MHz) δ 2.24ꢀ2.25 (m, 4H), 3.33 (s, 2H), 3.50ꢀ3.52 (m, 4H),
3
13
3.71 (s, 3H), 7.22 (d, J = 8.4 Hz, 2H), 7.81 (d, J = 8.4 Hz, 2H). CꢀNMR (CDCl , 100MHz,) δ
3
52.0, 53.7, 63.0, 70.0, 128.9, 129.1, 129.6, 143.4, 167.0.
Methyl 4ꢀ(piperidinꢀ1ꢀylmethyl)benzoate (69). The title compound was synthesized according
to general procedure A using piperidine 65 (2.98 g, 35.00 mmol). 69 was obtained as a yellow
1
oil: 1.89 g (94%). HꢀNMR (CDCl , 400 MHz) δ 1.37ꢀ1.42 (m, 2H), 1.50ꢀ1.55 (m, 4H), 2.31ꢀ
3
1
3
2
.33 (m, 4H), 3.46 (s, 2H), 3.86 (s, 3H), 7.35 (d, J = 8.4 Hz, 2H), 7.93 (d, J = 8.4 Hz, 2H). Cꢀ
NMR (CDCl , 100 MHz) δ 25.3, 25.4, 50.2, 52.5, 60.5, 125.6, 128.3, 131.5, 140.4, 169.3.
3
Methyl 4ꢀ((4ꢀmethylpiperazinꢀ1ꢀyl)methyl)benzoate (70). The title compound was synthesized
according to general procedure A using Nꢀmethylpiperazine 66 (3.50 g, 35.00 mmol). 70 was
1
obtained as a yellow oil: 2.10 g (97%). HꢀNMR (CDCl , 200 MHz) δ 2.21 (s, 3H), 2.38ꢀ2.42
3
13
(
m, 8H), 3.50 (s, 2H), 3.84 (s, 3H), 7.34 (d, J = 8.4 Hz, 2H), 7.94 (d, J = 8.4 Hz, 2H). CꢀNMR
(
CDCl , 100 MHz,): 46.0, 51.9, 53.2, 55.0, 62.5, 128.8, 128.9, 129.5, 143.8, 167.0.
3
ACS Paragon Plus Environment
20

ACS Chemical Neuroscience
Page 22 of 70
1
2
3
4
5
6
7
8
9
1
1
1
1
1
1
1
1
1
1
2
2
2
2
2
2
2
2
2
2
3
3
3
3
3
3
3
3
3
3
4
4
4
4
4
4
4
4
4
4
5
5
5
5
5
5
5
5
5
5
6
General procedure (B) for the synthesis of methyl 4ꢀ((amino)methyl)benzyl alcohols (71ꢀ
7
4). To an iceꢀcold suspension of LiAlH (2.0 eq.) in anhydrous THF (2.3 M), a solution of
4
methyl 4ꢀ((amino)methyl)benzoate (67ꢀ70) (1.0 eq.) in the same solvent (1.6 M) was added
dropwise. After stirring at rt for 2 h, the reaction mixture was cooled down in an iceꢀbath, and
quenched by slow addition of cold water (4 mL). The resulting mixture was stirred at 0 °C for 30
min, added to 1N aqueous NaOH solution (4 mL) and stirred at 0 °C for additional 10 min. This
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
suspension was diluted with Et O and filtered through a Celite cake. The filtrate was
2
concentrated in vacuo, and the resulting residue was taken up with EtOAc (20 mL) and washed
with H O (20 mL). The organic phase was dried over anhydrous Na SO , filtered and
2
2
4
concentrated under vacuum to afford the title compound (71ꢀ74). Further purification by flash
chromatography was performed when required.
(
(
(
4ꢀ((Diethylamino)methyl)phenyl)methanol (71). The title compound was synthesized from 67
1.10 g, 5.00 mmol) according to general procedure B. 71 was obtained as a yellow oil: 0.84 g
1
88%). HꢀNMR (CDCl , 400 MHz) δ 0.99ꢀ1.03 (m, 6H), 2.45ꢀ2.50 (m, 4H), 3.52 (s, 2H), 3.90
3
1
3
(br s, exch, 1H), 4.55 (s, 2H), 7.24ꢀ7.26 (m, 4H). CꢀNMR (CDCl , 100 MHz) δ 11.4, 46.4,
3
56.0, 64.5, 126.8, 129.2, 138.0, 140.0.
(4ꢀ(Morpholinomethyl)phenyl)methanol (72). The title compound was synthesized from 68
(1.80 g, 7.65 mmol) according to general procedure B. Elution with DCM/MeOH/33% aqueous
1
NH solution (9:1:0.1) afforded 72 as light pink solid: 1.40 g (88%). HꢀNMR (CDCl , 400
3
3
MHz) δ 2.30ꢀ2.32 (m, 4H), 3.38 (s, 2H), 3.55ꢀ3.57 (m, 4H), 4.00 (br s, exch, 1H), 4.51 (s, 2H),
ACS Paragon Plus Environment
21

Page 23 of 70
ACS Chemical Neuroscience
1
2
3
4
5
6
7
8
9
13
7
.18ꢀ7.22 (m, 4H). CꢀNMR (CDCl , 100 MHz) δ 53.4, 63.0, 64.3, 66.7, 126.9, 129.4, 136.3,
3
140.4.
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
(
4ꢀ(Piperidinꢀ1ꢀylmethyl)phenyl)methanol (73). The title compound was synthesized from 69
(1.89 g, 8.10 mmol) according to general procedure B. 73 was obtained as a yellow oil: 1.50 g
1
(90%). HꢀNMR (CDCl , 400 MHz) δ 1.34ꢀ1.37 (m, 2H), 1.49ꢀ1.52 (m, 4H), 2.27ꢀ2.31 (m, 4H),
3
1
3
3
.38 (s, 2H), 4.10 (br s, exch, 1H), 4.55 (s, 2H), 7.18ꢀ7.22 (m, 4H). CꢀNMR (CDCl , 100 MHz)
3
δ 25.3, 25.4, 50.2, 64.5, 66.5, 126.8, 129.2, 138.0, 140.0.
(4ꢀ((4ꢀMethylpiperazinꢀ1ꢀyl)methyl)phenyl)methanol (74). The title compound was
synthesized from 70 (2.10 g, 8.45 mmol) according to general procedure B. Elution with
DCM/MeOH/33% aqueous NH solution (9:1:0.2) afforded 74 as a white solid: 1.44 g (77%).
3
1
HꢀNMR (CDCl , 400 MHz) δ 2.11 (s, 3H), 2.28ꢀ2.32 (m, 8H), 3.38 (s, 2H), 4.52 (s, 2H), 4.80
3
1
3
(br s, exch, 1H), 7.18ꢀ7.23 (m, 4H). CꢀNMR (CDCl , 100 MHz,): 46.0, 53.2, 55.0, 65.2, 66.8,
3
1
26.9, 128.3, 137.8, 140.2.
General procedure (C) for the synthesis of benzaldehydes 52ꢀ55.
To a ꢀ78 °C solution of oxalyl chloride (1.2 eq.) in DCM (0.5 M) anhydrous DMSO (4.0 eq.) in
DCM (0.5 M) was slowly added. After stirring for 15 min, the reaction mixture was added to a
solution of the proper alcohol (71ꢀ74) (1.0 eq.) in DCM (0.4 M) and allowed to stir at ꢀ78 °C for
1
h. Et N (5.0 eq.) was subsequently added and the resulting mixture was stirred at ꢀ78 °C for
3
additional 20 min and then warmed up to rt. The reaction medium was diluted with H O (30
2
mL), and the immiscible phases were separated over a separatory funnel. The organic layer was
ACS Paragon Plus Environment
22

ACS Chemical Neuroscience
Page 24 of 70
1
2
3
4
5
6
7
8
9
1
1
1
1
1
1
1
1
1
1
2
2
2
2
2
2
2
2
2
2
3
3
3
3
3
3
3
3
3
3
4
4
4
4
4
4
4
4
4
4
5
5
5
5
5
5
5
5
5
5
6
washed with brine (30 mL), dried over anhydrous Na SO filtered and concentrated under
2
4,
vacuum to afford the title compound (52ꢀ55). Further purification by flash chromatography was
performed when required.
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
4ꢀ((Diethylamino)methyl)benzaldehyde (52). The title compound was synthesized from 71
(0.83 g, 4.20 mmol) according to general procedure C. 52 was obtained as a yellow oil: 0.48 g
1
(
60%). HꢀNMR (CDCl , 200 MHz) δ 0.91 (t, J = 7.0 Hz, 6H), 2.43 (q, J = 6.6 Hz, 4H), 3.54 (s,
3
1
3
2H), 7.41 (d, J = 7.8 Hz, 2H), 7.65 (d, J = 7.8, 2H), 9.80 (s, 1H). CꢀNMR (CDCl 100 MHz) δ
3
,
11.5, 47.0, 57.3, 129.4, 129.8, 135.5, 148.1, 192.0.
4ꢀ(Morpholinomethyl)benzaldehyde (53). The title compound was synthesized from 72 (1.40
g, 6.75 mmol) according to general procedure C. Elution with DCM/MeOH/33% aqueous NH3
1
solution (9.5:0.5:0.03) afforded 5 as a yellow solid: 0.81 g (59%). HꢀNMR (CDCl , 400 MHz) δ
3
2.36ꢀ2.38 (m, 4H), 3.49 (s, 2H), 3.61ꢀ3.64 (m, 4H), 7.43 (d, J = 8.0 Hz, 2H), 7.75 (d, J = 8.0 Hz,
13
2
H), 9.90 (s, 1H). CꢀNMR (CDCl , 100 MHz) δ 53.6, 62.9, 66.8, 129.4, 129.7, 135.5, 145.3,
3
191.8.
4ꢀ(Piperidinꢀ1ꢀylmethyl)benzaldehyde (54). The title compound was synthesized from 73 (1.18
g, 5.74 mmol) according to general procedure C. Elution with DCM/MeOH/33% aqueous NH3
1
solution (9.5:0.5:0.03) afforded 54 as a yellow oil: 0.92 g (79%). HꢀNMR (CDCl , 400 MHz) δ
3
1
7
1
.28ꢀ1.32 (m, 2H), 1.44ꢀ1.48 (m, 4H), 2.23ꢀ2.27 (m, 4H), 3.40 (s, 2H), 7.38 (d, J = 7.2 Hz, 2H),
1
3
.69 (d, J = 7.2 Hz, 2H), 9.86 (s, 1H). CꢀNMR (CDCl , 100 MHz) δ 23.8, 25.4, 53.8, 62.3,
3
29.0, 129.3, 135.0, 146.0, 192.5.
ACS Paragon Plus Environment
23

Page 25 of 70
ACS Chemical Neuroscience
1
2
3
4
5
6
7
8
9
4
ꢀ((4ꢀMethylpiperazinꢀ1ꢀyl)methyl)benzaldehyde (55). The title compound was synthesized
from 74 (1.44 g, 6.53 mmol) according to general procedure C. Elution with DCM/MeOH/33%
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
1
aqueous NH solution (9:1:0.1) afforded 55 as a yellow oil: 1.25 g (88%). HꢀNMR (CDCl , 400
3
3
MHz) δ 2.20 (s, 3H), 2.26ꢀ2.56 (m, 8H), 3.49 (s, 2H), 7.42 (d, J = 7.6 Hz, 2H), 7.74 (d, J = 7.6
1
3
Hz, 2H), 9.90 (s, 1H). CꢀNMR (CDCl , 100 MHz) δ 45.9, 53.1, 55.0, 62.5, 129.3, 129.6, 135.4,
3
145.7, 191.8.
General procedure (F) for the synthesis of NꢀcyanoꢀN'ꢀalkyl/arylguanidines (78ꢀ80). To a
suspension of sodium dicyanamide (81) (1.2 eq.) in 1ꢀbutanol (1.2 M), the proper amine
hydrochloride (82ꢀ84) (1.0 eq.) was added. The resulting mixture was heated at reflux for 6ꢀ8 h,
affording a white precipitate, which was filtered off. The filtrate was concentrated under vacuum
to yield crude NꢀcyanoꢀN'ꢀalkyl/arylguanidine (78ꢀ80). Further purification methods were
employed when required.
NꢀCyanoꢀN'ꢀethylguanidine (78). The title compound was obtained according to general
procedure F using ethylamine hydrochloride 82 (2.29 g, 27.07 mmol). Elution with
1
DCM/MeOH/33% aqueous NH solution (9:1:0.1) afforded 78 as a yellow oil: 2.5 g (73%). Hꢀ
3
13
NMR (D O, 400 MHz) δ 1.19 (t, J = 6.8 Hz, 3H), 3.23 (q, J = 6.8 Hz, 2H). CꢀNMR (D O, 100
2
2
MHz) δ 13.3, 36.5, 120.5, 161.1.
NꢀCyanoꢀN'ꢀpropylguanidine (79). The title compound was obtained according to general
procedure F using propylamine hydrochloride 83 (2.30 g, 24.53 mmol). Elution with
ACS Paragon Plus Environment
24

ACS Chemical Neuroscience
Page 26 of 70
1
2
3
4
5
6
7
8
9
1
1
1
1
1
1
1
1
1
1
2
2
2
2
2
2
2
2
2
2
3
3
3
3
3
3
3
3
3
3
4
4
4
4
4
4
4
4
4
4
5
5
5
5
5
5
5
5
5
5
6
1
DCM/MeOH/33% aqueous NH solution (9:1:0.1) afforded 79 as a waxy solid: 1.9 g (63%). Hꢀ
3
NMR (CD OD, 400 MHz) δ 0.90 (t, J = 7.2 Hz, 3H), 1.49ꢀ1.54 (m, 2H), 3.09 (t, J = 6.8 Hz, 2H).
3
1
3
CꢀNMR (CD OD, 100 MHz) δ 10.1, 22.2, 42.8, 118.8, 161.7.
3
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
NꢀCyanoꢀN'ꢀphenylguanidine (80). The title compound was obtained according to general
procedure F using aniline hydrochloride 84 (0.65 g, 5.05 mmol). The crude material was
triturated with H O, affording 80 as a white solid, which was used in the next step without
2
1
further purification: 0.80 g (quantitative yield). HꢀNMR (DMSOꢀd , 400 MHz) δ 6.98 (br s,
6
13
exch, 2H), 7.05ꢀ7.09 (m, 1H), 7.28ꢀ7.35 (m, 4H), 9.03 (br s, exch, 1H). CꢀNMR (DMSOꢀd ,
6
100 MHz) δ 117.6, 121.7, 124.2, 129.2, 138.4, 159.9.
NꢀGuanylurea sulfate hydrate (36). To a solution of cyanoguanidine (35) (1.0 g, 11.9 mmol) in
water (8 mL) 70% aqueous H SO (3.30 g, 23.8 mmol.) was added. The resulting mixture was
2
4
stirred at rt for 15 min and heated at reflux for 45 min. The reaction mixture was then cooled to
rt, affording the precipitation of a white solid, which was collected by filtration. The residue was
triturated with Et O, and used in the next step without further purification. 36 was obtained as a
2
1
white solid: 2.20 g (93%). Mp: 199ꢀ200°C (dec.) HꢀNMR (400 MHz; DMSOꢀd ) δ 6.92 (br s,
6
1
3
exch, 2H), 8.23 (br s, exch, 4H), 11.38 (br s, exch, 1H). CꢀNMR (100 MHz; DMSOꢀd ) δ
6
+
155.4, 156.1. MS (ESI) m/z 103 [M + H] .
General procedure (G) for the synthesis of N'ꢀalkyl/arylguanylureas (85ꢀ87). To a solution
of cyanoguanidine (78ꢀ80) (1.0 eq.) in water (1.2 M) 70% aqueous sulfuric acid (2.0 eq.) was
added. The resulting mixture was stirred at rt for 15 min and heated at reflux for 1 h. The
ACS Paragon Plus Environment
25

Page 27 of 70
ACS Chemical Neuroscience
1
2
3
4
5
6
7
8
9
reaction mixture was then cooled to rt and basified with Na CO . The aqueous phase was
2
3
evaporated in vacuo, and the crude material was taken up with MeOH. The residue was filtered
off and the organic solvent was concentrated under vacuum, to afford the title compound (85ꢀ
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
8
7), which was used in the next step without further purification.
(N'ꢀEthylcarbamimidoyl)urea (85). The title compound was obtained according to general
procedure G using 78 (2.50 g, 20.46 mmol). 85 was obtained as a gummy solid: 3.90 g
1
(quantitative yield). HꢀNMR (D O, 400 MHz) δ 0.48 (t, J = 6.8 Hz, 3H), 2.58 (q, J = 6.8 Hz,
2
13
2
H). CꢀNMR (D O, 100 MHz) δ 12.1, 35.8, 152.8, 155.2.
2
(N'ꢀPropylcarbamimidoyl)urea (86). The title compound was obtained according to general
procedure G using 79 (1.80 g, 14.26 mmol). 86 was obtained as a gummy solid: 2.40 g
1
(quantitative yield). HꢀNMR (D O, 400 MHz) δ 0.49ꢀ0.54 (m, 3H), 1.14ꢀ1.25 (m, 2H), 2.71 (t, J
2
1
3
=
6.8 Hz, 1H), 2.83 (t, J = 6.8 Hz, 1H). CꢀNMR (D O, 100 MHz) δ 10.1, 21.1, 42.6, 155.7,
2
1
56.3.
(N'ꢀPhenylcarbamimidoyl)urea (87). The title compound was obtained according to general
1
procedure G using 80 (0.80 g, 4.90 mmol). 87 was obtained as a white solid: 0.72 g (82%). Hꢀ
NMR (CDCl , 400 MHz) δ 6.10 (br s, exch, 2H), 7.01ꢀ7.06 (m, 3H), 7.11(br s, exch, 3H), 7.21ꢀ
3
13
7
.25 (m, 2H). CꢀNMR (CDCl , 100 MHz) δ 124.0, 125.0, 129.5, 140.0, 155.6, 163.6.
3
General procedure (H) for the synthesis of 6ꢀaminoꢀ4ꢀarylꢀ3,4ꢀdihydroꢀ1,3,5ꢀtriazinꢀ2(1H)ꢀ
ones (3ꢀ27), 6ꢀaminoꢀ4arylꢀ3,4ꢀdihydroꢀ1,3,5ꢀtriazineꢀ2(1H)ꢀthiones (28ꢀ31) and 6ꢀNꢀ
ACS Paragon Plus Environment
26

ACS Chemical Neuroscience
Page 28 of 70
1
2
3
4
5
6
7
8
9
1
1
1
1
1
1
1
1
1
1
2
2
2
2
2
2
2
2
2
2
3
3
3
3
3
3
3
3
3
3
4
4
4
4
4
4
4
4
4
4
5
5
5
5
5
5
5
5
5
5
6
alkyl/arylꢀ4ꢀarylꢀ3,4ꢀdihydroꢀ1,3,5ꢀtriazinꢀ2(1H)ꢀones (32ꢀ34). To a solution of Nꢀguanylurea
sulfate hydrate (36), or 2ꢀiminoꢀ4ꢀthiobiuret (75), or N'ꢀalkyl/arylguanylurea (85ꢀ87) (1.0 eq.) in
concentrated H SO (5.5 M) the proper benzaldehyde (37ꢀ61, 76, and 77) (1.2 eq.) was added.
2
4
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
After stirring at rt for 72 h, the reaction mixture was diluted with a small amount of cold H O.
2
The solution was then made basic with Na CO , affording a precipitate which was collected by
2
3
filtration. Whereas no precipitation was observed, the aqueous phase was concentrated in vacuo.
The crude material was either purified by trituration with organic solvents or by chromatographic
techniques affording the title compound (3ꢀ34).
6ꢀAminoꢀ4ꢀ(2ꢀnitrophenyl)ꢀ3,4ꢀdihydroꢀ1,3,5ꢀtriazinꢀ2(1H)ꢀone (3). The title compound was
obtained according to general procedure H using 2ꢀnitrobenzaldehyde (37) (0.91 g, 6.00 mmol)
and 36 (0.80 g, 5.00 mmol). The crude material was triturated with a mixture of Et O/DCM and
2
1
washed with H O to afford 3 as a brown solid: 0.19 g (16%). Mp: 196ꢀ197 °C. HꢀNMR
2
(
DMSOꢀd , 400 MHz) δ 5.62 (br s, exch, 2H), 6.16 (s, 1H), 7.54ꢀ7.59 (m, 3H), 7.74 (t, J = 8.0
6
13
Hz, 1H), 7.91 (d, J =8.0 Hz, 1H), 8.78 (br s, exch, 1H). CꢀNMR (DMSOꢀd , 100 MHz) δ 64.1,
6
+
124.0, 126.9, 128.3, 132.6, 137.5, 147.4, 149.9, 153.3. MS (ESI) m/z 236 [M + H] .
6ꢀAminoꢀ4ꢀ(3ꢀnitrophenyl)ꢀ3,4ꢀdihydroꢀ1,3,5ꢀtriazinꢀ2(1H)ꢀone (4). The title compound was
obtained according to general procedure H using 3ꢀnitrobenzaldehyde (38) (0.91 g, 6.00 mmol)
and 36 (0.80 g, 5.00 mmol). The crude material was triturated with a mixture of Et O/DCM and
2
1
washed with H O to afford 4 as a white solid: 0.63 g (53%). Mp: 213ꢀ214 °C. HꢀNMR (DMSOꢀ
2
d6, 400 MHz) δ 5.47 (br s, exch, 2H), 5.73 (br s, 1H), 7.66 (t, J = 8.0 Hz, 1H), 7.73 (br s, exch,
13
1
H), 7.80 (d, J = 8.0 Hz, 1H), 8.15 (d, J = 8.0 Hz, 1H), 8.19 (s, 1H), 8.86 (br s, exch, 1H). Cꢀ
ACS Paragon Plus Environment
27

Page 29 of 70
ACS Chemical Neuroscience
1
2
3
4
5
6
7
8
9
NMR (DMSOꢀd , 100 MHz) δ 67.1, 120.6, 122.2, 129.7, 132.9, 147.8, 148.1, 150.2, 153.6. MS
6
+
ꢀ
(
ESI) m/z 236 [M + H] , MS (ESI) m/z 234 [M ꢀ H] .
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
ꢀAminoꢀ4ꢀ(4ꢀnitrophenyl)ꢀ3,4ꢀdihydroꢀ1,3,5ꢀtriazinꢀ2(1H)ꢀone (5). The title compound was
obtained according to general procedure H using 4ꢀnitrobenzaldehyde (39) (0.23 g, 1.50 mmol)
and 36 (0.20 g, 1.25 mmol). The crude material was triturated with a mixture of Et O/DCM and
2
1
washed with H O to afford 5 as a white solid: 53 mg (22%). Mp: 253ꢀ254 °C. HꢀNMR (DMSOꢀ
2
d6, 400 MHz) δ 5.78 (br s, exch, 2H), 5.92 (s, 1H), 7.65 (d, J= 8.0 Hz, 2H), 7.67 (br s, exch, 1H),
13
8
.28 (d, J =8.0 Hz, 2H), 8.54 (br s, exch, 1H). CꢀNMR (DMSOꢀd , 100 MHz) δ 64.0, 123.8,
6
+
ꢀ
1
27.6, 147.5, 148.9, 150.7, 152.3. MS (ESI) m/z 236 [M + H] , MS (ESI) m/z 234 [M ꢀ H] .
6ꢀAminoꢀ4ꢀ(2ꢀbromophenyl)ꢀ3,4ꢀdihydroꢀ1,3,5ꢀtriazinꢀ2(1H)ꢀone (6). The title compound
was obtained according to general procedure H using 2ꢀbromobenzaldehyde (40) (0.70 mL, 6.00
mmol) and 36 (0.80 g, 5.00 mmol). The crude material was triturated with a mixture of
Et O/DCM and washed with H O to afford 6 as a white solid: 0.73 g (54%). Mp: 212ꢀ213 °C.
2
2
1
HꢀNMR (DMSOꢀd , 400 MHz) δ 5.68 (br s, exch, 2H), 5.88 (s, 1H), 7.22ꢀ7.26 (m, 1H), 7.41 (d,
6
1
3
J = 8.0 Hz, 2H), 7.46 (br s, exch, 1H), 7.58 (d, J = 8.0 Hz, 1H), 8.62 (br s, exch, 1H). CꢀNMR
DMSOꢀd , 100 MHz) δ 67.6, 120.6, 127.6, 129.3, 132.4, 134.6, 141.6, 149.4, 153.0.
(
6
6ꢀAminoꢀ4ꢀ(3ꢀbromophenyl)ꢀ3,4ꢀdihydroꢀ1,3,5ꢀtriazinꢀ2(1H)ꢀone (7). The title compound
was obtained according to general procedure H using 3ꢀbromobenzaldehyde (41) (0.56 mL, 4.80
mmol) and 36 (0.60 g, 4.00 mmol). The crude material was triturated with a mixture of
Et O/DCM and washed with H O to afford 7 as a white solid: 0.40 g (38%). Mp: 196ꢀ197 °C.
2
2
ACS Paragon Plus Environment
28

ACS Chemical Neuroscience
Page 30 of 70
1
2
3
4
5
6
7
8
9
1
1
1
1
1
1
1
1
1
1
2
2
2
2
2
2
2
2
2
2
3
3
3
3
3
3
3
3
3
3
4
4
4
4
4
4
4
4
4
4
5
5
5
5
5
5
5
5
5
5
6
1
HꢀNMR (DMSOꢀd , 400 MHz) δ 5.64 (s, 1H), 5.70 (br s, exch, 2H), 7.34ꢀ7.36 (m, 2H), 7.48 (br
6
1
3
s, exch, 1H), 7.50ꢀ7.53 (m, 2H), 7.88 (br s, exch, 1H). CꢀNMR (DMSOꢀd , 100 MHz) δ 66.6,
6
1
21.8, 125.7, 129.2, 131.2, 131.3, 145.9, 152.0, 153.6.
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
6ꢀAminoꢀ4ꢀ(2ꢀfluorophenyl)ꢀ3,4ꢀdihydroꢀ1,3,5ꢀtriazinꢀ2(1H)ꢀone (8). The title compound was
obtained according to general procedure H using 2ꢀfluorobenzaldehyde (42) (0.50 mL, 4.80
mmol) and 36 (0.64 g, 4.00 mmol). The crude material was triturated with a mixture of
Et O/DCM and washed with H O to afford 8 as a white solid: 0.52 g (62%). Mp: 152ꢀ154 °C.
2
2
1
HꢀNMR (DMSOꢀd , 400 MHz) δ 5.39 (br s, exch, 2H), 5.83 (s, 1H), 7.12ꢀ7.21 (m, 2H), 7.30ꢀ
6
13
7
.37 (m, 2H), 7.48 (br s, exch, 1H), 8.80 (br s, exch, 1H). CꢀNMR (DMSOꢀd , 100 MHz) δ
6
64.1, 125.1(d, J = 6.0 Hz), 125.7 (d, J = 29.0 Hz), 128.4, 129.0, 132.9, 150.1, 152.6, 159.1 (d, J
+
=
234.0 Hz). MS (ESI) m/z 209 [M + H] .
6
ꢀAminoꢀ4ꢀ(3ꢀfluorophenyl)ꢀ3,4ꢀdihydroꢀ1,3,5ꢀtriazinꢀ2(1H)ꢀone (9). The title compound was
obtained according to general procedure H using 3ꢀfluorobenzaldehyde (43) (0.64 mL, 6.00
mmol) and 36 (0.80 g, 5.00 mmol). The crude material was triturated with a mixture of
1
Et O/DCM and washed with H O to afford 9 as a white solid: 0.74 g (71%). Mp: 182ꢀ183°C. Hꢀ
2
2
NMR (DMSOꢀd , 400 MHz) δ 5.47 (br s, exch, 2H), 5.57 (s, 1H), 7.08ꢀ7.12 (m, 2H), 7.17ꢀ7.19
6
13
(m, 1H), 7.36ꢀ7.42 (m, 1H), 7.56 (br s, exch, 1H), 8,85 (br s, exch, 1H). CꢀNMR (DMSOꢀd ,
6
1
00 MHz) δ 67.7, 112.6, 113.2, 121.6, 129.8, 147.9, 150.4, 152.3, 161.0 (d, J = 246.0 Hz). MS
+
(ESI) m/z 209 [M + H] .
ACS Paragon Plus Environment
29