Multi-target-directed ligands for Alzheimer's disease: Discovery of chromone-based monoamine oxidase/cholinesterase inhibitors

Article abstract of DOI:10.1016/j.ejmech.2018.07.056

There has been a substantial research effort to design multi-target ligands for the treatment of Alzheimer's disease (AD), an approach that is moved by the knowledge that AD is a complex and multifactorial disease affecting many linked to pathological pathways. Accordingly, we have devoted efforts to develop multi-target ligands based on the chromone scaffold. As a result, a small library of chromone derivatives was synthesized and screened towards human cholinesterases and monoamine oxidases. Compounds 2-(dimethylamino)ethyl (E)-3-(4-oxo-2-(p-methylphenlcarbamoyl)-4H-chromen-6-yl)acrylate (9a) and 2-(dimethylamino)ethyl (E)-3-(4-oxo-3-(phenylcarbamoyl)-4H-chromen-6-yl)acrylate (23a) were identified as the most promising multi-target inhibitors of the series. Compound 9a acted as a potent, selective and bifunctional AChEI (IC₅₀ = 0.21 μM, K_i = 0.19 μM) and displayed dual hMAO inhibitory activity (hMAO-A IC₅₀ = 0.94 μM, K_i = 0.057 μM and hMAO-B IC₅₀ = 3.81 μM, K_i = 0.48 μM). Compound 23a acted as a selective IMAO-B (IC₅₀ = 0.63 μM, K_i = 0.34 μM) while still displaying hChE inhibitory and bifunctional activity in the low micromolar range. Overall, these two compounds stand out as reversible multi-target inhibitors with favourable permeability, toxicological and drug-like profiles, thus being valid candidates for subsequent optimization and pre-clinical studies.

Full text of DOI:10.1016/j.ejmech.2018.07.056

Accepted Manuscript
Multi-target-directed ligands for Alzheimer's disease: Discovery of chromone-based
monoamine oxidase/cholinesterase inhibitors
Joana Reis, Fernando Cagide, Martín Estrada Valencia, José Teixeira, Donatella
Bagetta, Concepción Pérez, Eugenio Uriarte, Paulo J. Oliveira, Francesco Ortuso,
Stefano Alcaro, María Isabel Rodríguez-Franco, Fernanda Borges
PII:
S0223-5234(18)30621-4
10.1016/j.ejmech.2018.07.056
EJMECH 10591
DOI:
Reference:
To appear in: European Journal of Medicinal Chemistry
Received Date: 23 February 2018
Revised Date: 18 July 2018
Accepted Date: 23 July 2018
Please cite this article as: J. Reis, F. Cagide, Martí.Estrada. Valencia, José. Teixeira, D. Bagetta,
Concepció. Pérez, E. Uriarte, P.J. Oliveira, F. Ortuso, S. Alcaro, Marí.Isabel. Rodríguez-Franco, F.
Borges, Multi-target-directed ligands for Alzheimer's disease: Discovery of chromone-based monoamine
oxidase/cholinesterase inhibitors, European Journal of Medicinal Chemistry (2018), doi: 10.1016/
j.ejmech.2018.07.056.
This is a PDF file of an unedited manuscript that has been accepted for publication. As a service to
our customers we are providing this early version of the manuscript. The manuscript will undergo
copyediting, typesetting, and review of the resulting proof before it is published in its final form. Please
note that during the production process errors may be discovered which could affect the content, and all
legal disclaimers that apply to the journal pertain.

ACCEPTED MANUSCRIPT
Multi-target-directed ligands for Alzheimer´s disease: discovery of chromone-
based monoamine oxidase/cholinesterase inhibitors
Joana Reis[a], Fernando Cagide[a], Martín Estrada Valencia[b], José Teixeira[a,c]
,
Donatella Bagetta[d], Concepción Pérez[b], Eugenio Uriarte[e,f], Paulo J. Oliveira[c],
Francesco Ortuso[d], Stefano Alcaro[d], María Isabel Rodríguez-Franco[b], Fernanda
Borges[a]
[a] CIQUP/Department of Chemistry and Biochemistry, Faculty of Sciences, University
of Porto, 4169-007 Porto, Portugal
[b] Instituto de Química Médica, (IQM-CSIC), C/Juan de la Cierva 3, 28006 Madrid,
Spain
[c] CNC – Center for Neuroscience and Cell Biology, UC-Biotech Building, Biocant
Park, University of Coimbra, Cantanhede 3060-197, Portugal
[d] Department of “Scienze della Salute”, University “Magna Græcia” of Catanzaro,
Campus “Salvatore Venuta”, Viale Europa, 88100 Catanzaro, Italy
[e] Department of Organic Chemistry, Faculty of Pharmacy, Universidad of Santiago de
Compostela, 15782 Santiago de Compostela, Spain
[f] Applied Chemical Science Institute, Universidad Autonoma de Chile, 7500912
Santiago de Chile, Chile
1

ACCEPTED MANUSCRIPT
ABSTRACT
There has been a substantial research effort to design multi-target ligands for the
treatment of Alzheimer disease (AD), an approach that is moved by the increase
knowledge that AD is a complex and multifactorial disease affecting many interlinked
pathological pathways. Accordingly, we have devoted efforts to develop multi-target
ligands based on the chromone scaffold. As a result, a small library of chromone
derivatives was synthesized and screened towards human cholinesterases and
monoamine oxidases. Compounds 2-(dimethylamino)ethyl (E)-3-(4-oxo-2-(p-
methylphenlcarbamoyl)-4H-chromen-6-yl)acrylate (9a) and 2-(dimethylamino)ethyl
(E)-3-(4-oxo-3-(phenylcarbamoyl)-4H-chromen-6-yl)acrylate (23a) were identified as
the most promising multi-target inhibitors of the series. Compound 9a acted as a potent,
selective and bifunctional AChEI (IC50 = 0.21 µM, Ki = 0.19 µM) and displayed dual
hMAO inhibitory activity (hMAO-A IC50 = 0.94 µM, Ki = 0.057 µM and hMAO-B IC50
= 3.81 µM, Ki = 0.48 µM). Compound 23a acted as a selective IMAO-B (IC50 = 0.63
µM, Ki = 0.34 µM) while still displaying hChE inhibitory and bifunctional activity in
the low micromolar range. Overall, these two compounds stand out as reversible multi-
target inhibitors with favourable permeability, toxicological and drug-like profiles, thus
being valid candidates for subsequent optimization and pre-clinical studies.
Keywords: Alzheimer’s disease, MAO-B inhibitors, AChE inhibitors, New Chemical
Entities, Chromone, Multi-target-directed ligands
2

ACCEPTED MANUSCRIPT
1. Introduction
Alzheimer’s disease (AD) is a chronic neurodegenerative disease (ND) with a high
social impact characterized by a progressive decline in cognitive function and dementia.
Major pathological hallmarks include extensive neuronal loss, formation of intracellular
neurofibrillary tangles (NFT) and extracellular deposition of β-amyloid peptides (Aβ)
[1].
Currently approved drugs for the clinical management of AD are only for symptomatic
control and relief, thus improving the patients’ quality of life. Therapeutic approaches
are based on single target drugs that interact with specific targets within the
neurodegeneration cascade, cognitive decline and/or motor control [2,3]. However,
considering the multifactorial etiology of AD new therapeutic solutions supported on a
multi-target-directed ligand (MTDL) approach has been an increased point of focus [4].
In fact, several biological targets are considered when dealing with ND’s; enzymatic
targets such as cholinesterases (ChEs), monoamine oxidases (MAOs) or β-secretase-1
(BACE-1, EC 3.4.23.46), are currently the subject of the development of several new
chemical entitities (NCE) [5].
Acetylcholinesterase (AChE, EC 3.1.1.7), the main target in the palliative therapy of
AD, is present in both central and peripheral nervous system and in muscular motor
plaques, and is responsible for the enzymatic cleavage of neurotransmitter acetylcholine
(ACh). Another ChE, butyrylcholinesterase (BuChE, EC 3.1.1.8), present in the brain
peripheral tissues and in the serum, is up-regulated in advanced AD and may play a role
in the pathophysiology of the disease [6]. Currently approved drugs for AD are the ChE
inhibitors rivastigmine, galantamine, and donepezil (Figure 1A) [7], which are not able
to modify disease progression [8].
3

ACCEPTED MANUSCRIPT
Monoamine oxidases (MAOs; amine-oxygen oxidoreductase, EC 1.4.3.4) are
intracellular flavine-containing enzymes that play a major role in the in vivo inactivation
of biogenic amines, which are important mediators of neuronal signaling in the central
and peripheral nervous system [9]. MAO-A and MAO-B are two isoforms of MAOs
expressed in mammals, which can be distinguished based on their substrate preference
and their interaction with specific inhibitors. Clorgyline for MAO-A and (R)-(−)-
deprenyl, rasagiline and safinamide for MAO-B (Figure 1B) are well-known inhibitors
used in depression and Parkinson therapy [9,10]. MAO-B expression in neuronal tissues
increases 4-fold in the elderly, resulting in increased dopamine deactivation and higher
production of hydrogen peroxide, which play a relevant role in the aetiology of ND
[11]. On the other hand, emerging evidence has shown that MAO-A mediates growth-
factor withdraw [12,13], protection against mitochondrial toxins [14] and toxin-induced
apoptosis of neuronal cells [15]. As a result, it was hypothesized that MAO-A inhibition
can protect against neuronal death and that both isoforms play a role in ND [16].
Figure 1. Drugs approved as ChE (A) and MAO (B) inhibitors.
As part of our continued endeavor to develop NCE for the therapy of neurodegenerative
diseases, we have focused our investigation on the identification of multi-
4

ACCEPTED MANUSCRIPT
target/multifunctional ligands based on the chromone scaffold. Several studies reported
by our group have shown that chromone (4H-benzopyran-4-one) [17] is a privileged
structure and a relevant framework for the development of novel MAO-B inhibitors
(IMAO-B) [18-22]. Additionally, the effect of the type of substituents located at C3-
and/or C6-positions of the benzopyrone ring on the potency and selectivity have been
underlined [18-23]. Previous lead optimization studies showed that the amide linker,
located at position 3 of the chromone scaffold, productively contributes to the
stabilization of the complex between the inhibitor and MAO-B. Among such
derivatives, the most potent MAO-B inhibitors were N-(3′-chlorophenyl)-4-oxo-4H-
chromene-3-carboxamide (IC50 = 403 pM) and N-(3′,4′-dimethylphenyl)-4-oxo-4H-
chromene-3-carboxamide (IC50 = 669 pM) [22]. Based on the data acquired so far, a
new chromone-based library was designed, synthesized and screened towards MAO and
ChE. Briefly, the chromone scaffold was functionalized at position C2- or C3 with a
phenylcarboxamide moiety and at position C6 with an acrylate moiety containing a
tertiary amine function (a fragment present in ChE inhibitors) (Figure 2). Additionally,
the putative metabolites formed upon hydrolysis of the acrylate side chain were also
synthesized and their in vitro biological activity evaluated.
Figure 2. Rational design strategy followed for the development of MTDL based on the chomone
scaffold.
In addition, the evaluation of blood-brain barrier (BBB) permeability, drug-like
properties, kinetics and mechanism of enzymatic inhibition and cytotoxicity profile of
5

ACCEPTED MANUSCRIPT
the most promising derivatives was performed. Finally, to understand the enzyme(s)-
inhibitor(s) interactions, molecular modeling studies were performed using models
based in the crystal structures of the targets.
2. Results and discussion
2.1. Chemistry
2.1.1. Synthesis of chromone 2- and 3-phenylcarboxamide derivatives
The general synthetic strategy for the development of chromone 2- and 3-
phenylcarboxamide derivatives is depicted in Schemes 1 and 2. The first series (Scheme
1, compounds 2-13) encompasses 2-phenylcarboxamide-based chromones synthesized
from the commercially available 6-bromo-4-oxo-4H-chromene-2-carboxylic acid (1).
The amide chromone derivatives 2-7 were obtained by a condensation reaction using
compound 1 as starting material and the appropriate phenylamine derivatives. Final
compounds 8-13 were obtained by a microwave-assisted Pd(II)-catalyzed Heck cross
coupling reaction using 2-(dimethylamino)ethyl acrylate (8a-13a) or 2-
(diethylamino)ethyl acrylate (8b-13b) as reagents.
R2
H2N
R1
R2
R2
O
O
R
N
O
O
O
COOH
O
O
O
O
POCl3
N
R1
R
O
N
R1
H
H
Br
R
O
Br
N
R
DCM
DMF
NaHCO3
TBAB
O
Pd(OAc)2
1
2-7
8-13
2
3
4
5
6
7
R1 = R2 = H
8a, 8b
9a, 9b
R1 = R2 = H
R1 = CH3; R2 = H
a, R = CH3
b, R = CH2CH3
R1 = CH3; R2 = H
R1 = OCH3; R2 = H
R1 = Cl; R2 = H
R1 = H; R2 = Cl
R1 = R2 = CH3
10a, 10b R1 = OCH3; R2 = H
11a, 11b R1 = Cl; R2 = H
12a, 12b R1 = H; R2 = Cl
13a, 13b R1 = R2 = CH3
Scheme 1. Synthetic strategy followed for the synthesis of compounds 2-13.
Chromone 3-phenylcarboxamides (Scheme 2, compounds 17-28) were obtained from
bromo-4-oxo-4H-chromene-3-carboxylic acid (16) using the same synthetic strategy as
6

ACCEPTED MANUSCRIPT
described above. The carboxylic acid 16 used as starting material was prepared from 5’-
bromo-2’-hydroxyacetophenone (14). Briefly, a POCl3-induced cyclization of the
acetophenone was performed to attain 6-bromo-4-oxo-4H-chromene-3-carbaldehyde
(15) that was subsequently oxidized, in the presence of sulfamic acid and sodium
chlorite, to the carboxylic acid 16.
Scheme 2. Synthetic strategy followed for the synthesis of compounds 15-28.
Chromones 29-34 and 35-40 were obtained from chromone 2-phenylcarboxamides
(compounds 8-13) and chromone 3-phenylcarboxamides (compounds 23-28),
respectively, by acidic hydrolysis) (Scheme 3).
7

ACCEPTED MANUSCRIPT
Scheme 3. Synthetic strategy followed for the synthesis of compounds 29-40.
The overall yields were moderate-to-high except for the Heck cross-coupling reaction
(compounds 8-13 and 23-28), which resulted in lower yields mainly due to problems of
the purification steps, mainly involving the presence of the starting material and the
removal of the excess of the catalyst
2.2. Biological evaluation
The screening of the chromone library toward hMAOs and hChEs was performed
following previously described methods (see Experimental Section) [22,24].
The results of the in vitro evaluation of inhibitory potency (IC50 values) and selectivity
expressed as SI toward hMAOs and hChEs of the chromones under study and reference
drug inhibitors (clorgyline for MAO-A; (R)-(−)-deprenyl, rasagiline, safinamide for
MAO-B; donepezil and tacrine for ChEs) are shown in Tables 1 and 2.
2.2.1. Evaluation of hMAO/hChE inhibitory activities of chromone 2-
phenylcarboxamide derivatives
Compounds 8a, 9a, 10a and 13b (Table 1) presented MAO-A and MAO-B inhibitory
activity in the low micromolar and submicromolar range. Remarkably, this is the first
8

ACCEPTED MANUSCRIPT
report of chromone 2-phenylcarboxamides as monoamine oxidase inhibitors. No
significant activity was observed for the remaining C2-substituted chromone
derivatives. On the other hand, all tested chromone-2-carboxamides (compounds 8-13)
displayed low submicromolar and micromolar activity toward AChE, and generally no
significant activity toward BuChE. It was also observed that the more active compounds
toward AChE were the ones bearing a dimethyl-substituted tertiary amine (8a-13a).
Conversely, diethylamine derivatives 8b-13b were active against AChE in the
micromolar range and, with the exception of 13b, were inactive toward MAOs.
Table 1. hMAOs and hChEs inhibitory activity and selectivity of chromone 2-
phenylcarboxamide (8-13) and chromone 3-phenylcarboxamide derivatives (23-28).
IC50 (µM) ± S.D.a
hMAO-B hMAO-A
IC50 (µM) ± S.D.a
Compound
R
R1
R2
SI
SI
hAChE
hBuChE
CH3
CH3
CH3
CH3
CH3
H
CH3
OCH3
Cl
H
CH3
H
CH3
OCH3
Cl
H
CH3
H
H
H
2.28±0.01
3.81±0.01
*
> 4.4 b
0.74±0.07
0.21±0.07
0.91±0.29
1.22±0.41
5.92±0.41
0.21±0.08
3.47±0.38
9.87±0.18
1.14±0.40
0.86±0.19
4.45±1.09
2.47±0.42
1.57±0.21
2.1
8a
9a
0.94±0.03
4.91±0.01
0.25
*
*
*
*
*
*
*
*
*
*
*
> 47.6 b
> 11.0 b
> 8.2 b
> 1.7 b
> 47.6 b
> 2.9 b
> 1.0 b
> 8.8 b
> 11.6 b
> 2.2 b
> 4.0 b
*
*
*
*
*
*
*
*
*
> 0.49 b
10a
11ª
12ª
13a
8b
H
*
*
*
*
*
*
*
*
*
-
-
-
-
-
-
-
-
Cl
CH3
H
H
H
H
Cl
CH3
CH3
CH2CH3
CH2CH3
CH2CH3
CH2CH3
CH2CH3
CH2CH3
9b
10b
11b
12b
13b
2.54±0.01
> 3.9 b
CH3
CH3
CH3
CH3
CH3
H
CH3
OCH3
Cl
H
CH3
H
H
H
H
0.63±0.01
2.37±0.01
*
0.72±0.01
2.21±0.01
3.65±0.01
*
*
*
*
*
*
*
*
> 15.9 b
3.69±0.24
0.97±0.12
3.54±0.19
9.32±0.21
4.32±0.29
7.42±0.31
1.2
23a
24a
25a
26a
27a
28a
23b
> 4.2 b
7.6
-
*
*
*
*
*
> 2.8 b
H
> 13.9 b
> 1.1 b
Cl
CH3
H
> 4.5 b
*
*
*
-
-
-
CH3
CH2CH3
-
-
9

ACCEPTED MANUSCRIPT
CH2CH3
CH2CH3
CH2CH3
CH2CH3
CH2CH3
CH3
OCH3
Cl
H
CH3
H
H
H
Cl
CH3
*
*
*
*
*
*
*
*
-
-
-
8.99±0.64
5.43±0.71
7.68±0.53
8.54±1.02
*
*
*
*
> 1.1 b
24b
25b
26b
27b
28b
> 1.8 b
> 1.3 b
1.25±0.01
> 8.0 b
> 1.2 b
*
*
*
-
0.017±0.002
0.050±0.002
0.023±0.002
63.4±0.1
-
68.7±0.4
53.0±0.7
**
4041
1060
-
-
-
-
-
-
-
-
-
-
-
(R)-(−)-Deprenyl
Rasagiline
> 4347 c
Safinamide
Clorgyline
Donepezil
0.0045±0.0003 0.00007
-
-
-
-
-
0.010±0.002
0.23±0.07
2.50±0.07
0.040±0.002
250
0.17
-
Tacrine
a
Data are expressed as mean ± S.D. from three diferent experiments;. * Inactive at 10 µM. ** Inactive at 100 µM (highest concentration
tested). b Values obtained under the assumption that the corresponding IC50 against MAO-A or MAO-B is the highest concentration tested (10
µM).
c Values obtained under the assumption that the corresponding IC50 against MAO-A or MAO-B is the highest concentration tested (100 µM).
SI: hMAO-B selectivity index = IC50 (hMAO-A) /IC50 (hMAO-B). SI: hAChE selectivity index = IC50 (hBuChE) /IC50 (hAChE)
.
Compound 8a (Table 1), with no substituents on the exocyclic phenyl ring and bearing
two methyl groups in the tertiary amine moiety, resulted in a selective IMAO-B (IC50 =
2.28 ± 0.01 µM, MAO-B/ MAO-A SI > 4.4). Additionally, it displayed a relevant
inhibitory activity toward AChE (IC50 = 0.74 ± 0.07 µM), and BuChE activity (IC50 =
1.57 ± 0.21 µM). Compound 9a (Table 1), bearing a methyl group in para position of
the chromone exocyclic phenyl ring, and two methyl groups linked to the tertiary amine,
showed MAO-B (IC50 = 3.81 ± 0.01 µM), MAO-A (IC50 = 0.94 ± 0.03 µM) and AChE
(IC50 = 0.21 ± 0.07 µM, AChE/ BuChE SI > 47.6) inhibition. These results show a
decrease of MAO-B inhibitory activity when compared with the chromone analogs
previously studied [18-20,22].
2.2.2. Evaluation of hMAO/hChE inhibitory activities chromone-3-
phenylcarboxamide derivatives
Chromone 3-carboxamide derivatives (Table 2) generally showed MAO-B selective
inhibition in the micromolar and submicromolar range, with exception of compounds
25a, 23b-26b and 28b. In this series, MAO-B inhibitory activity was noticeably
affected by the spacial volume of the substituent located on the tertiary amine and/or in
10

ACCEPTED MANUSCRIPT
the exocyclic aromatic ring. With exception of 27b, dimethyl-N-substituted derivatives
(23b-28b) were inactive toward MAOs. Regarding AChE, and with exception of 27a-
23b and 28b, selective AChE inhibitory activity in the low micromolar range was
observed. Interestingly, 23a (IC50 = 4.32 ± 0.29 µM, Table 2) and 24a (IC50 = 7.42 ±
0.31 µM, Table 2) were able to inhibit BuChE. Compound 23a (Table 2), bearing no
substituents on the exocyclic ring, showed noteworthy MAO-B (IC50 = 0.63 ± 0.01 µM,
SI > 15.9), AChE and BuChE (IC50 = 3.69 ± 0.24 µM and IC50 = 4.32 ± 0.29 µM,
respectively, SI =1.2) inhibitory activities. Compound 24a (Table 2), bearing a p-methyl
group, displayed MAO-B selective inhibition (IC50 = 2.37 ± 0.01 µM) and a higher
selectivity index toward AChE over BuChE (SI = 7.6), with IC50 = 0.97 ± 0.12 µM and
7.42 ± 0.31 µM, respectively.
Table 2. hMAO and hChE inhibitory activity and selectivity of chromone 2-phenyl and
3-phenylcarboxamide carboxylic acid derivatives (29-40)
IC50 (µM) ± S.D.a
IC50 (µM) ± S.D.a
Compound
R1
R2
hMAO-B
hMAO-A
hAChE
hBuChE
29
30
31
32
33
34
H
CH3
OCH3
Cl
H
H
*
*
*
*
*
*
*
*
*
*
*
*
5.99±0.61
*
*
*
*
*
*
5.10±0.42
H
7.43±0.51
H
*
*
*
H
Cl
CH3
CH3
35
36
37
38
H
CH3
OCH3
Cl
H
H
H
H
4.66±0.11
*
*
*
*
9.25±0.72
*
*
*
*
*
*
*
*
8.70±0.50
*
11

ACCEPTED MANUSCRIPT
39
40
H
Cl
*
*
*
*
*
*
CH3
CH3
4.52±0.46
*
a Data are expressed as mean ± S.D. from three diferent experiments; * Inactive at 10 µM.
2.2.3. Evaluation of hMAO/hChE inhibitory activities of chromone 2-phenyl and 3-
phenylcarboxamide carboxylic acid derivatives
As an additional study, we evaluated the inhibition of MAOs and ChEs by the putative
metabolites 29-40 resulted from the hydrolysis of the acrylated-substituted chromones
(Table 2). Results showed that the presence of a C6-carboxylic acid moiety resulted in a
significant decrease of MAO and ChE inhibitory activity. Specifically, the hydrolyzed
chromone 2-phenylcarboxamide derivatives (29-34) displayed no activity toward both
MAO isoforms. Chromone 3-phenylcarboxamide 35 (IC50 = 6.66 ± 0.11 µM) and 40
(IC50 = 4.52 ± 0.46 µM) retained MAO-B inhibitory activity in the micromolar range,
albeit with a significant or moderate decrease, respectively, when compared to the
corresponding precursors 23a (IC50 = 0.63 ± 0.01 µM) and 28a (IC50 = 3.65 ± 0.01 µM).
Compounds 32, 33, 38 and 39 (Table 3) displayed no inhibitory activity toward AChE,
while chromone derivatives 29 (IC50 = 5.99 ± 0.61 µM), 30 (IC50 = 5.10 ± 0.42 µM), 31
(IC50 = 7.43 ± 0.51 µM), 35 (IC50 = 9.25 ± 0.72 µM) and 37 (IC50 = 8.70 ± 0.50 µM)
maintained moderate AChE inhibitory activity when compared to their corresponding
precursors. No inhibitory activity was observed for all the compounds in BuChE.
Overall, the data point toward the importance of the acrylate moiety for ChEs inhibition.
2.3. Molecular modeling studies
Molecular modeling studies were performed to investigate the activity and selectivity
of the chromone derivatives toward hChEs and hMAOs. The interaction of chromone
derivatives with the targets was estimated by the Induced Fit Docking (IFD) approach
(Experimental Section). According to the experimental inhibition data, IFD results
12

ACCEPTED MANUSCRIPT
indicated that inhibitors’ catalytic site recognition was better for hAChE than hBuChE.
(Table S2). Excluding 11b in hBuChE and 23b in hAChE, the top poses of our
inhibitors showed the positively charged ammonium net toward the inner side of the
binding sites. On the other hand, the favourite binding mode of 29-40 was characterized
by the orientation of their carboxylic group outside of the active site (Figure 3, S1-S70).
Figure 3. IFD top poses of 9a and 29 into the hAChE active site. Ligands are depicted in green carbon
polytube; most relevant target residues are reported in gray carbon wireframe.
The reason of these binding modes was addressed to electrostatic interactions, as
hAChE has an overall net negative charge. There are numerous studies reporting the
electrostatic mechanism that enhances the rate of cationic ligands binding down to the
catalytic cleft [25]. In fact, for both hChEs, there is a prevalence of negatively charged
amino acids (Table S1), despite their poor exposure in the active site gorge. In addition,
the x-ray structures of hChEs showed that catalytic residues lay in a cleft mostly
13

ACCEPTED MANUSCRIPT
surrounded by aromatic groups, promoting the attraction of ACh or exogenous ligands
with positively charged ammonia groups. In fact, the cation-π interaction between the
charged nitrogen atom of our ligands and the side chains of Trp86 and Trp82, in hAChE
and hBuChE respectively, was evident in our theoretical complexes. Moreover, the
ligand’s ammonium group was also involved in electrostatic interactions to Glu202 in
the case of hAChE, and Glu197 in the case of hBuChE. The same inhibitor moiety
established a hydrogen bond to the backbone of His447 and Hys438 in hAChE and
hBuChE, respectively. The analysis of the ligand(s)-target(s) interaction addressed the
hAChE selectivity to stacking interactions between the ligand and Tyr124, Trp286,
Phe297, Tyr341, Tyr337 and Phe338 residues. In fact, Tyr124, Phe297, Trp286, and
Tyr337 are replaced in hBuChE by non-aromatic residues such as Gln119, Val288,
Ala277 and Ala328, respectively. The benzopyrone and amide moieties also stabilized
the hAChE complexes by hydrogen bonds to Ser293, Phe295 and Arg296 backbones.
Regarding hAChE selectivity, docking analysis suggested a key role of both chromone
and phenylcarboxamide moieties, regardless of the position on the pyrone ring.
Concerning docking results on MAO enzymes, it was evident that hMAO-A recognition
was strongly prevented by steric hindrance. Indeed, IFD studies showed that the
hMAO-A conformational fitting was unable to find an energetically favoured
accommodation of the most part of the phenylcarboxamide derivatives into the target
cleft. As a matter of fact, among our most promising hMAOs inhibitors, 10a, 27a, 35,
and 40 were the only derivatives fitting the binding pocket of hMAO-A (Figures S71-
S106). A different outcome was observed in the case of hMAO-B. Compounds 8a
(Figure 4), 9a, 13b, 23a and 24a reported good predicted affinity, with the orientation
of the phenylcarboxamide moiety toward the FAD cofactor. On the other hand,
compound 28b was located outside of the target active site (Figures S107-S142). Such a
14

ACCEPTED MANUSCRIPT
binding mode was addressed to productive electrostatic interactions between the
ligands’ ammonium group and Glu84, and to stacking contacts of the
phenylcarboxamide moiety to Tyr398 and Tyr435 (Figure 4).
Figure 4. IFD top poses of 8a and 35 into the hMAO-B active site. Ligands are depicted in green carbon
polytube, most relevant target residues are reported in gray carbon wireframe, FAD cofactor is in CPK
colored spacefill.
2.4. Selection of multi-target chromone-based compounds
The results obtained in the biological screening, corroborated by molecular docking
studies, identified chromones 8a and 9a (chromone 2-phenylcarboxamide derivatives)
and 23a and 24a (chromone 3-phenylcarboxamide derivatives) as the most promising
MAO/ChE inhibitors (Table 3). In general, an unsubstituted exocyclic phenyl ring
combined with the presence of a dimethyl acrylate moiety, were beneficial for the multi-
15

ACCEPTED MANUSCRIPT
target inhibitory activity. Moreover, the presence of a para-methyl group substituent in
the phenyl exocyclic was well tolerated.
As a result, a more detailed analysis by molecular modeling studies was performed for
the most promising inhibitors. The top poses of compounds 8a, 9a, 23a, and 24a were
almost equivalent, demonstrating a very low influence of the methyl group located at
the phenyl moiety para positon. Regarding the C2-substituted derivatives, the best
affinity binding modes of compounds 8a and 9a were characterized by hydrogen bonds
to Tyr326 and Cys172, mediated by the sp2 oxygen atom of the benzopyrone and amide
groups, respectively. As for the C3-substituted analogs, the top poses of compounds 23a
and 24a showed the benzopyrone sp2 oxygen atom establishing hydrogen bonds to
Tyr326 and Gln206, with the chromone C3 substitution resulting in the loss of one
hydrogen bond.
Table 3. MAO/ChE inhibitory activities and in vitro BBB permeability data of the
selected chromones
hMAO-B
hMAO-A
hAChEa
IC50(µM)
hBuChE
PAMPA-BBBb
Pe (10-6 cm s-1)
Compound
IC50(µM)a
IC50(µM)a
IC50(µM) a
Chromone 2-
phenylcarboxamides
3.6±0.1
(CNS +/-)
2.28 ± 0.01
*
0.74 ± 0.07
5.99±0.61
0.21 ± 0.07
5.10±0.42
1.57 ± 0.21
8a
29
9a
30
O
O
N
H
0.9±0.1
(CNS -)
*
*
*
*
*
HO
O
O
5.2±0.2
(CNS +)
3.81 ± 0.01
*
0.94 ± 0.03
*
1.5±0.1
(CNS -)
Chromone-3-
phenylcarboxamides
6.4±0.2
(CNS +)
0.63 ± 0.01
4.66 ± 0.11
2.37 ± 0.01
*
*
*
3.69 ± 0.24
9.25±0.72
0.97 ± 0.12
4.32 ± 0.29
*
23a
35
1.2±0.1
(CNS -)
9.7±0.4
(CNS +)
7.42 ± 0.31
24a
16

ACCEPTED MANUSCRIPT
1.6±0.2
(CNS -)
*
*
*
*
36
a Data are expressed as mean ± S.D. from three diferent experiments; bPredictive CNS -: CNS+: compounds with high BBB permeation.
CNS±: compounds with uncertain BBB permeation.CNS−: compounds of low BBB permeation* Inactive at 10 µM.
Regarding compound 35, docking simulations didn’t indicate a remarkable isoform
selectivity regarding affinity or geometries (Figures S101 and S137), since it assumed a
similar carboxylic group orientation for both hMAOs. The carboxylic acid was involved
in electrostatic interactions to Lys305 in hMAO-A, and to Lys209 and Lys296 in
hMAO-B (Figure 4). Accordingly, the selectivity of our ligands toward hMAOs was
mainly addressed to steric hindrance, since the hMAO-B active site is larger than
hMAO-A.
2.4.1. Evaluation of in vitro blood-brain barrier (BBB) permeability and drug-like
properties
The BBB permeability of the selected chromones and their corresponding putative
metabolites (Table 3) was evaluated using the in vitro PAMPA-BBB assay described by
Di et al. [26], and partially modified allowing the evaluation of molecules with limited
water-solubility [27,28]. According to the outlines previously established [26],
compounds with an effective permeability value (Pe) exceeding 4·10-6 cm s-1 can cross
the BBB whereas those displaying Pe less than 2·10-6 cm s-1 cannot reach the CNS.
Compounds 9a, 23a, and 24a (Table 3) showed permeability values above 4·10-6 cm s-1
in this in vitro BBB model and thus, it is expected that they could diffuse across the
BBB and interact with their biological targets. Compound 8a, displaying a Pe equal to
3.6·10-6 cm s-1, still demonstrated a promising permeability profile. Finally, it was
predicted that the putative metabolites were not able to cross the membrane (29, 30, 35,
36, Table 3).
17

ACCEPTED MANUSCRIPT
This data is also consistent with the drug-like properties of the most promising
chromone derivatives (8a, 9a, 23a, 24a) that were subsequently estimated (Table 4).
The calculated physicochemical parameters included: molecular weight (MW), number
of heavy atoms (N), partition coefficient (clog P), topological polar surface area (tPSA
in Å2), number of hydrogen acceptors (HBA), number of hydrogen donors (HBD) and
number of rotatable bonds (nrotb) (Table 4). The same parameters were calculated for
the reference inhibitors selegiline, safinamide, clorgyline, donepezil and tacrine (Table
5). The most relevant parameters, clogP and tPSA values, were within the optimal range
for oral bioavailability and BBB permeability (log P 2−4, MW < 400 and tPSA 50−90
Å2), respectively [29,30]. The estimated properties were in line with general CNS
requirements for drug-likeness, namely those related to the rule-of-5 (Table 4) [29].
Table 4. Evaluation of drug-like properties of the selected chromone-based compounds
Compound
8a
MWa
406.44
420.46
406.44
420.46
302.35
187.29
272.18
365.52
198.27
<450
Na cLogPa tPSA (Å2)a HBAa HBDa nrotba
30
31
30
31
22
14
17
27
15
___
3.54
3.99
3.23
3.68
2.91
2.64
3.74
4.76
3.05
<5
88.85
88.85
88.85
88.85
64.36
3.24
7
7
1
1
8
8
9a
7
1
8
23a
7
1
8
24a
Safinamide
(R)-(−)-Deprenyl
Clorgyline
Donepezil
Tacrine
4
3
7
1
0
4
12.47
21.71
38.91
<60-90
2
0
6
3
0
6
2
2
0
CNS+ drugs[31-35]
<7
<3
<8
MW: molecular weight; N: number of heavy atoms; clog P: calculated logarithm of the octanol-water partition
coefficient; HBA: number of hydrogen acceptors; HBD: number of hydrogen donors; nrotb: number of rotatable
bonds; a Properties calculated using Cheminformatics software [http://www.molinspiration.com].
18

ACCEPTED MANUSCRIPT
2.4.2. Evaluation of hMAOs inhibition mechanism
Kinetic experiments were performed for the chromone analogs comprising the most
promising multi-target profile. Therefore, the inhibition mechanism of compounds 9a
and 23a was investigated. The measurement of the initial rates of the MAO-catalyzed
oxidation of p-tyramine at five different substrate concentrations was performed in the
absence or presence of the inhibitors (Figure 5). Graphical analyses of the reciprocal
Lineweaver−Burk plots allowed the determination of Michaelis−Menten reaction
kinetic parameters (Michaelis constant, Km and maximum velocity, Vmax). Concerning
compound 9a, the kinetic studies were performed in both MAO-A and MAO-B.
Accordingly, it was observed for both isoforms that the Vmax remained unchanged while
Km increased. The Lineweaver−Burk plots for different concentrations of 9a (Figure 5A,
MAO-B and Figure 5B, MAO-A) were linear and intersected at the y-axis, indicating
that this compound acted as a competitive IMAO. The same behaviour was observed for
compound 23a, with the Lineweaver-Burk plots displaying a series of converging lines
on the same point of the y-axis (1/[S]), thus profiling a competitive MAO-B inhibition
mechanism (Figure 5C).
A
B
Dixon plot
30
Dixon plot
3000
2000
1000
0
500
400
300
200
100
20
10
0
1000 nM
2000 nM
20
10
500 nM
1000 nM
Ki = 478 nM
Ki = 57 nM
-1000
1000
2000
3000
[I] (nM)
-500
500
[I] (nM)
1000
1500
-1000
-100
0.005
0.010
0.005
0.010
1/[p-tyramine] (µM-1
)
1/[p-tyramine] (µM-1)
-10
-10
C
Dixon plot
3000
20
10
0
2000
1000
500 nM
Ki
= 337 nM
1000 nM
-1000
-500
500
[I] (nM)
1000
1500
-1000
0.005
0.010
1/[p-tyramine] ( µM-1
)
-10
19

ACCEPTED MANUSCRIPT
Figure 5. Kinetic studies on the inhibition mechanism of hMAO-B and hMAO-A by chromone 9a (A and
B, respectively), and chromone 23a (C). The effect of the inhibitors on the enzyme was determined from
the double reciprocal plot of 1/rate (1/V) versus 1/substrate concentration in presence of varying
concentrations of the inhibitors. The Ki values were calculated by the intersection of the curves obtained
by plotting 1/V versus the inhibitor concentration for each substrate concentration (Dixon plots insets on
the top left).
From the Dixon plots, obtained from the replots of the slopes of the Lineweaver–Burk
plots vs inhibitor concentrations (upper left corner), the hMAO-B inhibition binding
affinities, determined as inhibition constants (Ki), were calculated. Chromone 9a
displayed Ki values of 478 nM (Figure 5A) and 57 nM (Figure 5B) for hMAO-B and
hMAO-A, respectively. Chromone 23a exhibited the same behaviour, displaying a Ki
value of 337 nM for hMAO-B (Figure 5C). The estimated Ki values were smaller than
the respective IC50 values, as expected of reversible and competitive inhibitors [36,37].
Furthermore, the reversibility of hMAO-B inhibition for chromones 9a and 23a was
assessed by time-dependent inhibition studies (Figure 6). The behaviour of irreversible
((R)-(−)-deprenyl and rasagiline) and reversible (safinamide) inhibitors was also
evaluated under the same experimental conditions (Figure 6A). Additionally, the
reversibility of hMAO-A inhibition by chromone 9a and the inhibitor clorgyline was
also evaluated (Figure 6B). hMAO-B and hMAO-A activities (% of control) were
measured during a 60 min incubation with the enzyme inhibitors (Figure 6).
A
B
50
Clorgyline
(R)-(-)-Deprenyl
60
40
20
40
30
20
10
0
Compound 9a
Rasagiline
Safinamide
Compound 23a
Compound 9a
0
20
40
60
0
20
40
60
20
t (min)
t (min)

ACCEPTED MANUSCRIPT
Figure 6. Time-dependent inhibition of recombinant human MAO-B by the drugs (R)-(−)-deprenyl (50
nM), safinamide (40 nM), rasagiline (200 nM) and chromones 9a (500 nM) and 23a (250 nM) (Figure
6A), and time-dependent inhibition of recombinant human MAO-A of clorgyline (12.5 nM) and
chromone 9a (5.0 µM) (Figure 6B). The remaining activity was expressed as % of activity. Data are the
mean ± S.D. of three different experiments.
The analysis of time-dependent enzyme inhibition studies performed with rasagiline and
(R)-(−)-deprenyl (Figure 6A) showed that the enzyme residual activity decayed
continuously after the first 15 minutes of incubation, which is consistent with
irreversible enzymatic inhibition. In the case of the reversible inhibitor safinamide
(Figure 6A), an enhancement on enzymatic activity was observed along the analysis
time. The same behaviour was observed for chromone 9a with both MAO isoforms, and
for chromone 23a with hMAO-B, in a reversible hMAO-A and hMAO-B inhibition
profile.
2.4.3. Evaluation of hAChE inhibition mechanism
The most potent hAChE inhibitor (9a, IC50 = 0.21 µM) was selected for studying its
inhibition mechanism. The initial velocity of the enzymatic inhibition was measured at
four concentrations of the substrate acetylthiocholine (ATCh), in absence and presence
of three increasing concentrations of inhibitor.
21

ACCEPTED MANUSCRIPT
Figure 7. Kinetic study on the inhibition mechanism of hAChE inhibition by chromone 9a. (A) Overlaid
Lineweaver-Burk reciprocal plots of hAChE initial velocity at increasing substrate concentration (ATCh,
0.2 – 1.6 mM) in the absence of inhibitor and in the presence of 9a (0 – 0.20 µM) are shown. Lines were
derived from a least-squares analysis of the data points. (B) Replot of slopes vs. inhibitor concentration
for calculating Ki as the intersection on the x-axis.
The Lineweaver - Burk plots at different concentrations of compound 9a (Figure 7A)
resulted in an increase of inhibitor concentration as intercepts at y- and x-axes also
increased, pointing out a mixture of competitive and non-competitive mechanisms in
the enzymatic inhibition. These results suggested the simultaneous interaction of
compound 9a with both catalytic anionic site (CAS) and peripheral anionic site (PAS)
of hAChE, underlining compound 9a as a AChE bifunctional inhibitor. Reploting of
slopes vs. inhibitor concentration yielded a straight line (Dixon plot) that was fitted by
least-squares analysis and its intersection on the negative x-axis provided an inhibition
constant around Ki = 190 nM (Figure 7B).
2.4.4. Evaluation β-secretase-1 inhibitory activity of chromone 2-phenyl and 3-
phenylcarboxamides
To extend our studies on another relevant AD biological target, the most promising
chromone-based derivatives, namely compounds 8a, 9a, 23a, 24a, and 35, were also
22

ACCEPTED MANUSCRIPT
screened toward human BACE-1, a key enzyme which has been identified as being
responsible for β-amyloid formation [38]. However, none of the compounds displayed
relevant potency (IC50 > 10 µM).
2.4.5. Evaluation of the cytotoxicity in differentiated human neuroblastoma (SH-
SY5Y) and human hepatocarcinoma (HepG2) cells
The cytotoxicity profile of the most promising compounds was evaluated in
differentiated human neuroblastoma (SH-SY5Y) and human hepatocarcinoma (HepG2)
cells after a 48h incubation period in a concentration range of 0.1 to 50 µM. Both cell
lines are often used in the preclinical safety assessment of drug candidates, being the
SH-SY5Y cells of particular interest concerning the development of drug candidates for
CNS [39]. Cytotoxicity was evaluated by measuring cellular metabolic activity, using
the resazurin reduction assay (Figure 8), and total cell protein, which is directly
proportional to cell number, using the SRB assay (Figure 9).
The cytotoxicity profile of compounds 9a and 23a (Figure 8A and 8B, respectively) as
well as of compound 35 (Figure 8C), a putative metabolite of compound 23a, followed
the same trend in both neuroblastoma and hepatocarcinoma cells, although cytotoxic
effects were more noticeable in neuroblastoma cells. For compounds 23a and 35 no
significant decrease on cellular viability was noticed in SH-SY5Y and HepG2 cells for
all tested concentrations. Compound 23a, as well as its putative metabolite (compound
35), presented a wider safety profile and a promising safety margin, thus clearing it for
further preclinical studies.
23

ACCEPTED MANUSCRIPT
C
A
B
10000
8000
6000
4000
2000
0
15000
10000
8000
6000
4000
2000
0
*
*
10000
****
****
5000
0
M
M
M
M
M
M
M
M
M
M
M
M
µ
µ
µ
µ
µ
µ
µ
M
M
M
M
M
M
µ
µ
µ
µ
µ
µ
µ
µ
µ
µ
µ
1
2
6
2
5
0
5
.
1
2
6
2
5
0
5
1
2
.
1
.
0
2
6
2
5
0
5
0
1
2
0
1
2
SH-SY5Y cells
SH-SY5Y cells
SH-SY5Y cells
B
A
C
15000
10000
5000
0
15000
10000
5000
0
15000
10000
5000
0
*
M
M
M
M
M
µ
M
M
M
M
M
M
M
µ
µ
µ
µ
µ
M
M
M
M
M
M
µ
µ
µ
µ
µ
µ
µ
µ
µ
µ
µ
µ
2
6
2
5
0
5
1
2
6
2
5
0
5
1
.
.
1
2
1
2
6
2
5
0
5
1
2
.
0
0
1
2
0
HepG2 cells
HepG2 cells
HepG2 cells
Figure 8- Cytotoxicity profile of compounds 9a (8A), 23a (8B) and 35 (8C) measured by changes in
cellular metabolic activity on human neuroblastoma SH-SY5Y and human hepatocarcinoma HepG2 cells
after a 48 h treatment with different concentrations (0.1 to 50 µM). Untreated cells were used as control.
Results are expressed as mean % of untreated controls ± SEM. (n = 4). * P < 0.05 and **** P < 0.0005.
Compound 9a, a chromone analog bearing a methyl group on para position of the
exocyclic ring, exhibited significant cytotoxic effects towards SH-SY5Y cells at the
highest tested concentrations (25 and 50 µM). However, compound 9a presented an
acceptable safety window considering the nanomolar and submicromolar range
enzymatic inhibitory activities. No significant cytotoxicity was observed for any of the
compounds on HepG2 cells, demonstrating low risk drug-induced hepatotoxicity.
The cytotoxicity profile of compounds 9a, 23a (Figure 9A and 9B, respectively) and 35
(Figure 9C) was also evaluated by measuring changes in cell density.
24

ACCEPTED MANUSCRIPT
C
A
B
1.0
0.8
0.6
0.4
0.2
0.0
1.0
1.0
0.5
0.0
0.8
*
*
0.6
*
0.4
****
0.2
0.0
M
M
M
M
M
M
µ
M
M
M
M
M
M
µ
M
M
M
M
M
M
µ
µ
µ
µ
µ
µ
µ
µ
µ
µ
µ
µ
µ
µ
µ
µ
1
2
6
2
5
0
5
1
2
6
2
5
0
5
.
1
2
6
2
5
0
5
.
1
2
.
1
2
0
1
2
0
0
SH-SY5Y cells
SH-SY5Y cells
SH-SY5Y cells
B
C
A
2.0
2.0
2.0
1.5
1.0
0.5
0.0
1.5
1.0
0.5
0.0
1.5
1.0
0.5
0.0
M
M
M
M
M
M
µ
M
M
M
M
M
M
M
M
M
M
M
M
µ
µ
µ
µ
µ
µ
µ
µ
µ
µ
µ
µ
µ
µ
µ
µ
µ
1
2
6
2
5
0
5
1
2
6
2
5
0
5
1
2
6
2
5
0
5
.
.
.
1
2
1
2
1
2
0
0
0
HepG2 cells
HepG2 cells
HepG2 cells
Figure 9 – Cytotoxicity profile of compounds 9a (9A), 23a (9B) and 35 (9C) measured by changes in cell
mass density of human neuroblastoma SH-SY5Y and human hepatocarcinoma HepG2 cells after a 48 h
treatment with different concentrations (0.1 to 50 µM).. Untreated cells were used as control. Results are
expressed as mean % of untreated controls ± SEM. (n = 4). * P < 0.05 and **** P < 0.0005.
The same cytotoxicity pattern was observed for the SRB assay. Compounds 23a and 35
displayed no significant cytotoxic effects, for all tested concentrations, on SH-SY5Y
and HepG2 cell lines. On the other hand, compound 9a induced a significant decrease of
SH-SY5Y cell mass (44 %, p <0.0005) at the highest concentration tested (50 µM)
,
when compared to the control group. Again, this cytotoxic effect is less significant
considering that compound 9a displayed hMAOs/hAChE inhibition at much lower
concentrations. No significant cytotoxicity effect was observed on HepG2 cells.
2.4.6. Pan Assay INterference compoundS (PAINS) evaluation. Artificial activity of
synthetized compounds was theoretically investigated and compared with the biological
data. Compounds 8-52 were analysed by means of ZINC PAINS Pattern Identifier [40]
25

ACCEPTED MANUSCRIPT
and FAF-Drug4 [41] and none were highlighted as PAINS or aggregator. Taking into
account theoretical results and the biological data on hMAOs and hChEs our molecules
cannot be considered as false positive inhibitors. Finally, the inhibitory activity of the
most promising inhibitors has been demonstrated as a reversible process making us
confident that we are not in presence of artificial activities.
3. Conclusions
Along this project we have successfully developed BBB permeable chromone-based
multi-target compounds targeting human hChEs and hMAOs. Overall, compounds 9a
and 23a were identified as the most promising chromone derivatives of the series.
Compound 9a showed the advantage of acting as a potent and selective AChEI (Ki =
190 nM), able to interact with both CAS and PAS, in a bifunctional inhibition profile.
Furthermore, it possessed dual hMAO inhibitory activity (Ki = 478 nM toward hMAO-
B and Ki = 57 nM toward hMAO-A). On the other hand, compound 23a acted as a
selective IMAO-B (Ki = 337 nM) while still displaying hChE inhibitory activity in the
low micromolar range. Compound 23a demonstrated a wider safety window, with no
significant decrease on cellular viability in SH-SY5Y and HepG2 cells for all tested
concentrations. Furthermore, it should be stressed that its putative metabolite
(compound 35) was non-cytotoxic and retained hMAO-B and hAChE inhibitory
activities in the low micromolar range. Due to their different biological profiles
chromones 9a and 23a can be categorized as promising lead compounds for the
subsequent optimization and pre-clinical stages.
4. Experimental section
4.1. Chemistry. Reagents and general conditions
26

ACCEPTED MANUSCRIPT
6-Bromo-4-oxo-4H-chromene-2-carboxylic acid, 5’-bromo-2’-hydroxyacetophenone,
tetra-n-butylammonium bromide (TBAB), palladium (II) acetate, sulfamic acid, 2-
(dimethylamino)ethyl
acrylate,
2-(diethylamino)ethyl
acrylate,
N,N-
diisopropylethylamine (DIPEA), dimethylformamide anhydrous, phosphorus
oxychloride (POCl3) and phenylamine derivatives were purchased from Sigma-Aldrich.
All other reagents and solvents were pro analysis grade and were acquired from Carlo
Erba Reagents and Scharlab, and were used without additional purification. Thin-layer
chromatography (TLC) was carried out on precoated silica gel 60 F254 (Merck) with
layer thickness of 0.2 mm. For analytical control the following systems were used:
acetone, dichloromethane/methanol, dichloromethane in several proportions. The spots
were visualized under UV detection (254 and 366 nm). Flash chromatography was
performed using silica gel 60 0.2-0.5 or 0.040-0.063 mm (Carlo Erba Reagents) or
cellulose (Cellulose Powder MN 2100 Macherey-Nagel). Solutions were decolorized
with activated charcoal, when necessary. Solvents were evaporated in a Buchi
Rotavapor.
For the PAMPA-BBB assays, all reference drugs, phosphate buffer saline solution at pH
7.4 (PBS) and dodecane were purchased from Sigma-Aldrich and Across. Millex filter
units (PVDF membrane, diameter 25 mm, pore size 0.45 µm) were acquired from
Millipore. The porcine brain lipid (PBL) was obtained from Avanti Polar Lipids
(Alabaster). The donor microplate was a 96-well filter plate (PVDF membrane, pore
size 0.45 µm) and the acceptor microplate was an indented 96-well plate, both from
Millipore. The acceptor 96-well microplate was filled with 180 µL of PBS:ethanol
(70:30) and the filter surface of the donor microplate was impregnated with 4 µL of
porcine brain lipid (PBL) in dodecane (20 mg mL-1).
27

ACCEPTED MANUSCRIPT
Reagents used in cell culture, including nonessential amino acids, heat inactivated fetal
bovine serum (FBS), 0.05 % trypsin/1 mM EDTA, antibiotic (10000 U/mL penicillin,
10000 µg/mL streptomycin) and Hank’s balanced salt solution (HBSS) were purchased
from Gibco Laboratories. Dulbecco’s modified Eagle’s medium (DMEM) with 4.5 g/L
glucose were obtained from Sigma. Dimethylsulfoxide (DMSO-d6), absolute ethanol
and acetic acid were obtained from Merck. All reagents were of analytical grade or of
the highest grade available.
4.2. Apparatus. NMR data were acquired at room temperature on a Brüker AMX 400
spectrometer operating at 400 MHz for 1H and 101 MHz for 13C, or on a Bruker Avance
1
III HD 600 MHz operating at 600 MHz for for H and 151 MHz for 13C. Chemical
shifts were expressed in δ (ppm) values relative to tetramethylsilane (TMS) as internal
reference; coupling constants (J) were given in Hz. Assignments were also made from
DEPT (Distortionless Enhancement by Polarization Transfer). Mass spectra (MS) were
carried out on a Bruker Microtof (ESI) apparatus; the data were reported as m/z (% of
relative intensity of the most important fragments). Microwave-assisted synthesis was
performed in a Biotage Initiator Microwave Synthesizer.
The purity of the final products (>97% purity) was verified by high-performance liquid
chromatography (HPLC) equipped with a UV detector. Chromatograms were obtained
in an HPLC/DAD system, a Waters 2690 Alliance HPLC system (Milford, MA, USA),
equipped with a commercially prepacked Waters Spherisorb RP-18 (5 µm) ODS2
column, with 4.6 mm internal diameter and 250 mm length, and a Waters 996 PDA
detector at the maximum wavelength of 254 nm. The mobile phase consisted of a
methanol/water or acetonitrile/water (gradient mode, room temperature) at a flow rate of
1 mL/min. The chromatographic data was processed in a Compaq computer, fitted with
CSW 1.7 software (DataApex, Prague, Czech Republic).
28

ACCEPTED MANUSCRIPT
4.3. Synthesis of chromone 2-phenylcarboxamides (compounds 2-7) and chromone
3-phenylcarboxamides (compounds 17-22)
4.3.1. General procedure for amidation reaction. To a solution of 6-bromo-4-oxo-
4H-chromene-2-carboxylic acid (1, 2.6 mmol) or 6-bromo-4-oxo-4H-chromene-3-
carboxylic acid (16, 2.6 mmol) in DMF (4 mL) POCl3 (2.6 mmol) was added. The
mixture was stirred at room temperature for 30 min for the in situ formation of the acyl
chloride. Then, the aromatic amine with the desired aromatic pattern was added [21].
After 1-5 h, the mixture was diluted with dichloromethane (20 mL), washed with H2O
(2 x 10 mL) and with saturated NaHCO3 solution (2 x 10 mL). The organic phase was
dried, filtered and concentrated under reduced pressure. The crude product was purified
by flash chromatography and/or crystallization.
6-Bromo-4-oxo-N-phenyl-4H-chromene-2-carboxamide (2)
1
Yield: 65%. Purified by flash column chromatography with CH2Cl2/MeOH (9:1). H
NMR (400 MHz, DMSO-d6) δ: 10.75 (1H, s, CONH), 8.15 (1H, d, J = 2.3 Hz, H(5)),
8.11 (1H, dd, J = 8.9, 2.3 Hz, H(7)), 7.84 (1H, d, J = 8.9 Hz, H(8)), 7.79 (2H, dd, J =
8.7, 1.2 Hz, H(2’) H(6’)), 7.46 - 7.41 (2H, m, H(3’) H(5’)), 7.24 - 7.19 (1H, m, H(4’)),
13
7.02 (1H, s, H(3)). C NMR (101 MHz, DMSO-d6) δ: 177.5 (C4), 158.8 (C2), 157.3
(CONH), 155.5 (C8a), 139.1 (C7), 138.8 (C1’), 130.2 (C3’ C5’), 128.4 (C5), 126.6
(C4’), 126.4 (C3), 123.1 (C6), 122.5 (C2’ C6’), 119.9 (C4a), 112.5 (C8).
6-Bromo-4-oxo-N-(4-methylphenyl)-4H-chromene-2-carboxamide (3)
1
Yield: 62%. Purified by flash column chromatography with CH2Cl2/MeOH (9:1). H
NMR (400 MHz, DMSO-d6) δ: 10.68 (1H, s, CONH), 8.15 (1H, d, J = 2.4 Hz, H(5)),
8.09 (1H, dd, J = 8.9, 2.4 Hz, H(7)), 7.83 (1H, d, J = 8.9 Hz, H(8)), 7.68 (2H, d, J = 8.4
Hz, H(3’) H(5’), 7.23 (2H, d, J = 8.4 Hz, H(2’) H(6’)), 7.00 (1H, s, H(3)), 2.31 (3H, s,
29