Synthesis and Biological Assessment of Racemic Benzochromenopyrimidinimines as Antioxidant, Cholinesterase, and Aβ1?42Aggregation Inhibitors for Alzheimer's Disease Therapy

Article abstract of DOI:10.1002/cmdc.201500539

Given the complex nature of Alzheimer′s disease (AD), compounds that are able to simultaneously address two or more AD-associated targets show greater promise for development into drugs for AD therapy. Herein we report an efficient two-step synthesis and biological evaluation of new racemic benzochromene derivatives as antioxidants, inhibitors of cholinesterase and β-amyloid (Aβ_1?42) aggregation. Based on the results of the primary screening, we identified 15-(3-methoxyphenyl)-9,11,12,15-tetrahydro-10H,14H-benzo[5,6]chromeno[2,3-d]pyrido[1,2-a]pyrimidin-14-imine (3 e) and 16-(3-methoxyphenyl)-9,10,11,12,13,16-hexahydro-15H-benzo[5′,6′]chromeno[2′,3′:4,5]pyrimido[1,2-a]azepin-15-imine (3 f) as new potential multitarget-directed ligands for AD therapy. Further in-depth biological analysis showed that compound 3 f is a good human acetylcholinesterase inhibitor [IC₅₀=(0.36±0.02) μm], has strong antioxidant activity (3.61 μmol Trolox equivalents), and moderate Aβ_1?42antiaggregating power (40.3 %).

Full text of DOI:10.1002/cmdc.201500539

DOI: 10.1002/cmdc.201500539
Full Papers
Synthesis and Biological Assessment of Racemic
Benzochromenopyrimidinimines as Antioxidant,
Cholinesterase, and Ab1À42 Aggregation Inhibitors for
Alzheimer’s Disease Therapy
Youssef Dgachi,[a] Lhassane Ismaili,*[b] Damijan Knez,[c] Mohamed Benchekroun,[b]
HØl›ne Martin,[d] Natalia Szałaj,[e] Sarah Wehle,[f] Oscar M. Bautista-Aguilera,[b]
Vincent Luzet,[b] Alexandre Bonnet,[d] Barbara Malawska,[e] Stanislav Gobec,[c]
Mourad Chioua,[g] Michael Decker,[f] Fakher Chabchoub,*[a] and JosØ Marco-Contelles*[g]
Given the complex nature of Alzheimer’s disease (AD), com-
pounds that are able to simultaneously address two or more
AD-associated targets show greater promise for development
into drugs for AD therapy. Herein we report an efficient two-
step synthesis and biological evaluation of new racemic benzo-
chromene derivatives as antioxidants, inhibitors of cholinester-
ase and b-amyloid (Ab1À42) aggregation. Based on the results
of the primary screening, we identified 15-(3-methoxyphenyl)-
9,11,12,15-tetrahydro-10H,14H-benzo[5,6]chromeno[2,3-
d]pyrido[1,2-a]pyrimidin-14-imine (3e) and 16-(3-methoxy-
phenyl)-9,10,11,12,13,16-hexahydro-15H-benzo[5’,6’]chrome-
no[2’,3’:4,5]pyrimido[1,2-a]azepin-15-imine (3 f) as new poten-
tial multitarget-directed ligands for AD therapy. Further in-
depth biological analysis showed that compound 3 f is a good
human acetylcholinesterase inhibitor [IC50 =(0.36Æ0.02) mm],
has strong antioxidant activity (3.61 mmol Trolox equivalents),
and moderate Ab1À42 antiaggregating power (40.3%).
Introduction
Alzheimer’s disease (AD) is a severe form of dementia and the
most common cause of memory loss and cognitive deficiency
in elderly people, with more than 35 million cases documented
worldwide.[1] The incidence of AD will increase in the coming
years owing to the high age of the population and life expect-
ancy.[2] Several supramolecular lesions have been identified in
AD patients such as amyloid plaques consisting of aggregated
b-amyloid (Ab) peptide and intracellular neurofibrillary tangles,
constituted of hyperphosphorylated tau protein, along with
a substantial loss of synapses and cholinergic neurons located
in the nucleus basalis of Meynert.[3,4] The latter finding results
in low levels of the neurotransmitter acetylcholine in the syn-
aptic cleft, which is closely related to a decline in cognition
and related memory functions. Oxidative stress is also thought
to play a fundamental role in the onset and progression of AD.
Several studies have revealed that the accumulation of reactive
oxygen species released by senile plaques[5] corrupts the mo-
lecular content of neurons (e.g., the phospholipid membrane,
mitochondria, and cytoplasmic proteins),[6] which lead to cell
death and apoptosis.
Currently, the approved pharmacotherapy of AD is only pal-
liative and is based on targeting one pathological event at
a time. Namely, there is one N-methyl-d-aspartate receptor an-
tagonist, memantine, and three acetylcholinesterase inhibitors
(AChEIs): galantamine, rivastigmine, and donepezil.[7,8] Research
in the field of AChEIs has also been encouraged by the poten-
tial role of acetylcholinesterase (AChE) in accelerating the as-
sembly of Ab into insoluble amyloid fibrils in the brain[9]
[a] Y. Dgachi, Prof. F. Chabchoub
[e] Dr. N. Szałaj, Prof. B. Malawska
Laboratory of Applied Chemistry: Heterocycles, Lipids and Polymers
Faculty of Sciences of Sfax, University of Sfax, BP 802, 3000 Sfax (Tunisia)
E-mail: fakher.chabchoub@yahoo.fr
Department of Physicochemical Drug Analysis, Jagiellonian University
Medical College, Medyczna 9 Street, 30-688 Krakow (Poland)
[f] S. Wehle, Prof. Dr. M. Decker
[b] Dr. L. Ismaili, Dr. M. Benchekroun, Dr. O. M. Bautista-Aguilera, V. Luzet
Neurosciences IntØgratives et Cliniques EA 481, Laboratoire de Chimie Orga-
nique et ThØrapeutique, UFR SMP, UniversitØ de Franche-ComtØ, UniversitØ
Bourgogne Franche-ComtØ, 19 rue Ambroise ParØ, 25000 BesanÅon (France)
E-mail: lhassane.ismaili@univ-fcomte.fr
Pharmazeutische und Medizinische Chemie, Institut für Pharmazie und
Lebensmittelchemie, Julius-Maximilians-Universität Würzburg, Am Hubland,
97074 Würzburg (Germany)
[g] Dr. M. Chioua, Prof. J. Marco-Contelles
Laboratory of Medicinal Chemistry (IQOG, CSIC)
C/Juan de la Cierva 3, 28006 Madrid (Spain)
E-mail: iqoc21@iqog.csic.es
[c] D. Knez, Prof. S. Gobec
Faculty of Pharmacy
ˇ
ˇ
University of Ljubljana, Askerceva 7, 1000 Ljubljana (Slovenia)
Supporting Information for this article is available on the WWW under
http://dx.doi.org/10.1002/cmdc.201500539.
[d] Dr. H. Martin, A. Bonnet
Laboratory of Cell Toxicology, EA 4267, University of Franche-ComtØ
19 rue Ambroise ParØ, 25030 BesanÅon (France)
This article is part of a Special Issue on Polypharmacology and
Multitarget Drugs. To view the complete issue, visit:
http://onlinelibrary.wiley.com/doi/10.1002/cmdc.v11.12/issuetoc.
ChemMedChem 2016, 11, 1318 – 1327
1318
ꢀ 2016 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim

Full Papers
through the interaction with the peripheral anionic site of
AChE.[10] Alternatively, and owing to the multifactorial nature of
AD, the design of multitarget-directed ligands (MTDLs) able to
simultaneously interact with different enzymatic systems or re-
ceptors involved in the progress and development of AD rep-
resents one of the most promising approaches for the devel-
opment of new drugs for AD therapy.[11–17]
ties. As a result, we identified benzochromenopyrimidinimine
3 f (Figure 1) as a potential new multitarget-directed ligand for
AD therapy displaying sub-micromolar human AChE inhibition,
moderate Ab1À42 aggregation inhibitory activity, and strong an-
tioxidant properties.
Following this paradigm, one of us previously developed Results and Discussion
a number of quinazolinimine-based compounds (compound
type I, Figure 1)[18–20] by identifying novel tricyclic quinazolini-
Chemistry
mines as strong butyrylcholinesterase inhibitors (BuChEIs) rela-
tive to galantamine and rivastigmine.[18] Homo-bivalent quina-
zolinimines were also disclosed as novel nanomolar cholines-
terase (ChE) inhibitors with marked and tunable selectivity
toward butyrylcholinesterase (BuChE).[19] In the same vein, they
also reported a series of quinazolinimine–lipoic acid hybrids
that showed remarkable selectivity toward BuChE inhibition as
well as moderate antioxidant features.[20] Although these deriv-
atives demonstrated an attractive polypharmacologic profile,
they remain largely unexplored, and to the best of our knowl-
edge, scarce attention has been devoted to the quinazolini-
mine core to build new disease-modifying agents for AD thera-
py.[21]
The synthesis of compounds 3a–f and 4a–f was performed by
a straightforward two-step synthetic procedure with good
overall yields, as depicted in Scheme 1. The one-pot multicom-
ponent reaction starting from malononitrile, aromatic alde-
hydes, naphthol, and piperidine gave 3-amino-1-aryl-1H-ben-
zo[f]chromene-2-carbonitriles
1 and 2-amino-4-aryl-4H-ben-
zo[h]chromene-3-carbonitrile 2. Interestingly, this step was very
fast (reaction time 1 to 2 min), solvent free, and afforded inter-
mediates 1a/1b and 2a/2b in high yields (ranging from 90 to
95%) (Scheme 1). Next, condensation of compounds 1a/1b (or
2a/2b) with the appropriate lactams in the presence of phos-
phorus oxytrichloride in bromobenzene under microwave irra-
diation (MWI) for 15 min gave expected benzochromenopyri-
midinimines 3a–f and 4a–f in high yields (83–94%)
(Scheme 1).
Continuing with our interest in multicomponent reactions as
a very convenient method for introducing molecular diversi-
ty,[16,17,22,23] we designed new racemic benzochromenopyrimidi-
nimines IIA and IIB, in which a fused benzochromene motif
was added instead of the fused benzene ring on quinazolines I
to afford new MTDLs (Figure 1). Thus, in this paper we report
the highly convergent and simple two-step synthesis of 12
new benzochromenopyrimidinimines, as well as the inhibition
and kinetic analyses of AChE/BuChE enzymes, their Ab1À42 ag-
gregation inhibitory, and antioxidant (ORAC-FL method) activi-
All new compounds showed analytical and spectroscopic
data in good agreement with their structures and with the
data reported in the literature for similar compounds (see the
Experimental Section).
Biological evaluation
The in vitro evaluation of compounds 3a–f and 4a–f started
with analysis of their antioxidant activity and continued with
the inhibition of ChEs and evaluation of the Ab1À42 aggregation
inhibitory activity.
Antioxidant evaluation of compounds 3a–f and 4a–f
The antioxidant power of compounds 3a–f and 4a–f was eval-
uated by the oxygen radical absorbance capacity (ORAC)
method[24,25] by using fluorescein (FL) as the fluorescent probe,
2,2’-azobis(2-amidinopropane) dihydrochloride (AAPH) as the
peroxyl radical source, Trolox as a standard, and ferulic acid as
a reference molecule (ORAC value 3.7).[26] The antioxidant ca-
pacity of compounds 3a–f and 4a–f was determined by de-
tecting their time-dependent ability to compete with FL to
capture the peroxyl radical.
As shown in Table 1, all new derivatives displayed good abili-
ty to reduce a peroxyl radical with Trolox equivalents ranging
from 2.3 (for 4c) to 4.0 (for 4b), which are better than the
modest ORAC value (0.2) found for compound I (n=6).[20] This
is possibly due to the weak NÀH bond present in the (C)=NH
motif, which is absent in compounds of type I[20] (Figure 1). De-
rivatives 3d–f and 4b have the strongest antioxidant proper-
ties showing Trolox equivalents similar to that shown by ferulic
acid.
Figure 1. Structures of previously investigated derivatives I,[20] new benzo-
chromenopyrimidines II, and MTDLs 3e and 3 f described in this work.
ChemMedChem 2016, 11, 1318 – 1327
www.chemmedchem.org
1319
ꢀ 2016 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim

Full Papers
Scheme 1. Synthesis of benzochromenopyrimidinimines 3a–f and 4a–f.
Table 1. Inhibition (IC50) of EeAChE and eqBuChE and antioxidant (ORAC-FL method) activities for 3a–f, 4a–f, and reference compounds.
Compd
IC50 [nm][a]
Selectivity[b]
ORAC[c]
EeAChE
eqBuChE
3a
3b
3c
3d
3e
3 f
4a
4b
4c
4d
4e
4 f
67.8Æ4.3
9.9Æ0.5
34.0Æ3.9
6.0Æ0.1
3.2Æ0.2
5.3Æ0.2
379.5Æ12.4
93.5Æ6.9
96.1Æ2.8
44.9Æ1.9
12.1Æ0.8
52.6Æ0.4
640[18]
473.4Æ11.2
1163.0Æ49.8
455.0Æ7.9
85.6Æ3.9
7
118
13
14
15
23
11
6
3.3Æ0.1
2.7Æ0.2
3.2Æ0.3
3.9Æ0.2
3.4Æ0.2
3.6Æ0.2
3.3Æ0.0
4.0Æ0.1
2.6Æ0.1
3.1Æ0.1
2.3Æ0.1
2.5Æ0.1
n.d.[d]
46.7Æ4.8
123.7Æ5.3
4180Æ290
546.7Æ17.8
298.6Æ5.7
3
2865Æ80
64
26
2
13
0.1
–
317.0Æ6.8
89.6Æ2.1
8400[18]
12[20]
galantamine
I (n=6)
ferulic acid
130[20]
0.1[20]
n.d.[d]
n.d.[d]
3.7Æ0.1[26]
[a] Inhibition curves were obtained by nonlinear regression; Ee: Electrophorus electricus (electric eel), eq: equine. Values are the meanÆSEM of at least
three different experiments performed in quadruplicate. [b] Ratio IC50(EeAChE)/IC50(eqBuChE). [c] Data are expressed as (mmol Trolox equivalents)(mmol test
compound)À1 and are shown as meanÆSD. [d] Not determined.
Evaluation of AChE/BuChE inhibition
EeAChE inhibitors with IC50 values in the nanomolar range, but
For the preliminary experiments, Electrophorus electricus AChE
(EeAChE) and equine BuChE (eqBuChE) were used, following
Ellman’s protocol for the determination of inhibitory poten-
cy.[27] As shown in Table 1, compounds 3a–f and 4a–f were
they were less potent against eqBuChE, with selectivities
toward EeAChE ranging from 118 (for 3a) to 2 (for 4 f). Notably,
this selectivity is reversed with respect to that of quinazolini-
mine I (n=6),[20] a potent and selective eqBuChE inhibitor. The
ChemMedChem 2016, 11, 1318 – 1327
www.chemmedchem.org
1320
ꢀ 2016 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim

Full Papers
most potent EeAChE and eqBuChE inhibitor was compound
3e, which displayed an IC50 value of (3.2Æ0.2) nm for EeAChE;
it is 40- and 200-fold more potent than quinazolimine I (n=6)
(Figure 1) and galantamine, respectively. Compound 3e also in-
hibited eqBuChE [IC50 =(46.7Æ4.8) nm], and it is a fourfold less
potent inhibitor than compound I (n=6) (Figure 1). Potent an-
tioxidant benzochromenopyrimidinimine 3 f also strongly in-
hibited EeAChE [IC50 =(5.3Æ0.2) nm].
more active than galantamine; it displayed inverse selectivity
and is thus more selective toward hAChE than human BuChE
(hBuChE). Regarding the inhibition of hBuChE, these benzo-
chromenopyrimidinimines showed sub-micromolar inhibition,
with IC50 ranging from 0.302 mm for 3e to 0.907 mm for 3 f;
these compounds are 61- and 20-fold more active than galant-
amine, respectively. Among the tested compounds, inhibitor
3e was the most potent BuChEI [IC50 =(0.302Æ0.011) mm].
Next, the kinetic analysis and mechanism of hAChE inhibi-
tion by compounds 3b and 3e were investigated by obtaining
Lineweaver–Burk double reciprocal plots. Analysis of these
plots (Figure 2) showed that Km appears unaltered and Vmax is
decreased with increasing inhibitor concentrations. This pat-
tern indicated that inhibitors 3b and 3e were characterized by
a linear noncompetitive inhibition mechanism, a result that
could also be applied to benzochromenopyrimidinimines 3d
and 3 f. Replots of the slope versus concentration of com-
pounds 3b and 3e provide an estimate of the inhibition con-
stants (Ki of 162 and 144 nm, respectively).
In the same way, the mechanism of hBuChE inhibition by
compounds 3b and 3d–f was examined. The Lineweaver–Burk
reciprocal plots for benzochromenopyrimidinimine 3b
(Figure 3) showed increasing slopes and very similar intercepts
on the y axis (higher Km and equal Vmax values) upon increasing
the inhibitor concentration. This suggests a competitive-type
inhibition. A Ki value of 73.7 nm was estimated from the slopes
of double reciprocal plots versus concentration of compound
3b. For 3d–f, Lineweaver–Burk plots displayed increasing
slopes and increasing intercepts with higher inhibitor concen-
trations. The intersection of the lines above the x axis
(Figure 2) indicates that 3d–f are able to interact with both
the free and substrate-bound enzyme and therefore act as
mixed-type inhibitors of hBuChE. Note that benzochromeno-
pyrimidinimines 3b and 3e, differing only by a methoxy group
on the phenyl ring, presented the same inhibition mechanism
on hAChE and two different mechanisms on BuChE.
Concerning the structure–activity relationships (SARs), we
first conclude that the IC50 values for EeAChE and eqBuChE in-
hibition were much lower if the benzene ring attached to the
pyran core had an additional methoxy group. Next, com-
pounds of type IIA, that is, 3b–f (Figure 1), are more potent
EeAChE inhibitors than analogues 4a–f of type IIB (Figure 1),
with the exception of compound 4e. Finally, regarding the size
of the saturated carbocyclic ring attached to the pyrimidini-
mine moiety, for the same substituent in the aromatic ring at-
tached at the stereogenic center (C14 for n=1, C15 for n=2,
C16 for n=2), the most potent AChEIs were compounds bear-
ing a six-membered ring (i.e., 3b, 3e, 4b, and 4e), whereas for
eqBuChE inhibition no such clear SAR was observed.
On the basis of these results, we selected most-balanced
compounds 3b and 3d–f in terms of antioxidant properties
and EeAChE/eqBuChE inhibition for an in-depth study in
human cholinesterases and Ab1À42 aggregation inhibitory
assays.
As shown in Table 2 and regarding human AChE (hAChE) in-
hibition, we found similar results to those obtained for EeAChE
and similar SAR trends, but significant lower inhibition power.
In any case, these four benzochromenopyrimidinimines are
AChEIs, showing IC50 values ranging from 0.058 mm for 3e to
0.355 mm for 3 f, and selectivity toward hAChE ranging from
2.30 to 5.55. The most potent hAChE inhibitor was compound
3e with an IC50 value of 58 nm. Compound 3e was 24-fold
Table 2. Inhibition (IC50) of hAChE and hBuChE and Ab1À42 aggregation
inhibitory activity of 3b, 3d–f, and reference compounds.
Inhibition of Ab1À42 aggregation inhibitory activity by
benzochromenopyrimidinimines 3b and 3d–f
Continuing with the exploration of the biological activity of
compounds 3b and 3d–f, evaluation of their Ab1À42 aggrega-
tion inhibition power was undertaken by using the thioflavin T
(ThT)-based fluorescence assay.[29] As it is well known that com-
pounds with an absorbance spectrum that overlaps the excita-
tion and/or emission wavelengths of ThT can interfere with the
ThT assay results causing false positives, we first registered the
absorbance spectra of compounds 3b, 3e, and 3 f (see the
Supporting Information), and no absorbances were detected
with the wavelengths used for the test. Consequently, we pro-
ceeded with Ab1À42 aggregation inhibition analysis. As shown
in Table 2, benzochromenopyrimidinimine 3d was inactive,
compounds 3b and 3e presented very weak activity, and mol-
ecule 3 f was the most significant Ab1À42 aggregation inhibitor
(40.3%) at 10 mm, in the same range as resveratrol[30] at the
same concentration (Table 2).
Compd
IC50 [mm][a]
hAChE hBuChE
Selectivity[b] Inhibition [%][c]
3b
3d
3e
3 f
0.155Æ0.004 0.357Æ0.009
0.151Æ0.003 0.779Æ0.048
0.058Æ0.002 0.302Æ0.011
0.355Æ0.015 0.907Æ0.040
2.3
5.2
13.3Æ3.6
n.a.[d]
5.2
2.6
11.4
14.1Æ6.7
40.3Æ9.4
n.d.[e]
galantamine 1.60Æ0.14[28] 18.30Æ0.14[28]
resveratrol
n.d.[e]
n.d.[e]
40.1Æ5.8
[a] Values are the meanÆSEM of at least three different experiments per-
formed in quadruplicate. [b] Ratio IC50(hAChE)/IC50(hBuChE). [c] Percent in-
hibition at 10 mm compound and 1.5 mm Ab1À42; values are the meanÆ
SD. [d] Not active, defined as percent inhibition <10%. [e] Not deter-
mined.
ChemMedChem 2016, 11, 1318 – 1327
www.chemmedchem.org
1321
ꢀ 2016 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim

Full Papers
Figure 2. Steady-state inhibition of hAChE hydrolysis of acetylthiocholine (ATC) by compounds 3b and 3e. Lineweaver–Burk reciprocal plots of initial velocity
dependence on various inhibitor and substrate concentrations (0.067–0.50 mm) are presented. Lines were derived from weighted least-squares analysis of the
data.
Figure 3. Steady-state inhibition of hBuChE hydrolysis of butyrylthiocholine (BTC) by benzochromenopyrimidinimines 3b and 3d–f. Lineweaver–Burk recipro-
cal plots of initial velocity dependence on various inhibitor and substrate concentrations (0.05–0.50 mm) are presented. Lines were derived from weighted
least-squares analysis of the data.
Molecular modeling
hAChE) and an additional hydrogen bond 3.0 Š in length be-
The reported biological data refer to racemic mixtures, and the
observed final result is a combination of the effects of both
enantiomers in an extent that clearly cannot be predicted.
Nevertheless, to obtain more information about the mode of
action, docking studies were performed on the S and R enan-
tiomers of compound 3e, the most potent hAChEI.
Docking studies were performed on the crystal structure
4EY5[31] of hAChE with the program Gold, which was applied in
previous and related docking studies and thus was also used
herein to identify a representative binding mode.[32,33]
tween the protonated imine and the Asp74 residue (Figure 4).
These interactions were not found for the R enantiomer of 3e.
Conclusions
In conclusion, we synthesized and evaluated 12 new racemic
benzochromenopyrimidinimine derivatives as promising multi-
functional compounds, selective acetylcholinesterase inhibitors,
and good antioxidants. Compared to quinazoline I (n=6)
(Figure 1), a reference compound for this work,[20] the new
benzochromenopyrimidinimines showed reversed selectivity
against cholinesterases (ChEs) and improved antioxidant activi-
ty. In particular, 15-(3-methoxyphenyl)-9,11,12,15-tetrahydro-
10H,14H-benzo[5,6]chromeno[2,3-d]pyrido[1,2-a]pyrimidin-14-
imine (3e), although showing low b-amyloid (Ab1À42) aggrega-
tion inhibitory activity, exhibited nanomolar noncompetitive in-
hibition of human acetylcholinesterase and high antioxidant
The predicted docking solution of benzochromenopyrimidi-
nimine 3e is valid for the chosen crystal structure and its
amino acid arrangement within the binding gorge. Within this
crystal structure, a binding mode representing a noncompeti-
tive inhibition mode was found with the inhibitor binding in
the middle of the hAChE binding gorge. The S enantiomer of
3e seems to be preferred for binding over the R enantiomer
because of stabilization by p–p interactions of the benzochro-
meno unit to the Trp86 residue (the choline binding site of
power.
More
interestingly,
16-(3-methoxyphenyl)-
9,10,11,12,13,16-hexahydro-15H-benzo[5’,6’]chromeno[2’,3’:4,5]-
ChemMedChem 2016, 11, 1318 – 1327
www.chemmedchem.org
1322
ꢀ 2016 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim

Full Papers
Chemistry
General procedure for the synthesis and spectral data of com-
pounds 1a/1b and 2a/2b: A round-bottomed flask was charged
with equimolar (0.01 mmol) amounts of a- or b-naphthol, aromatic
aldehyde, malonitrile, and piperidine. An exothermic reaction was
observed. After 1 min, ethanol (5 mL) was added, and the product
was isolated by filtration, with yields ranging between 90 and
95%.
General procedure for the synthesis and spectral data of benzo-
chromenopyrimidines derivatives 3a–f and 4a–f: POCl3 (0.14 mL,
0.23 g, 1.5 equiv) was added dropwise to a mixture of 3-amino-1-
aryl-1H-benzo[f]chromene-2-carbonitriles (1) {or 2-amino-4-aryl-4H-
benzo[h]chro-mene-3-carbonitriles (2) (1 mmol)} and the appropri-
ate lactam (1.5 equiv) in bromobenzene (20 mL). The mixture was
stirred under microwave irradiation for 15 min. The mixture was
thereafter concentrated under reduced pressure to remove approx-
imately 80% of the solvent. Then, water (10 mL) was added, and
the mixture was treated with an aqueous solution of sodium hy-
droxide (20%) to pH 11–12. The mixture was extracted with CH2Cl2
(3”25 mL), washed with water (20 mL), and dried (MgSO4). The sol-
vent was evaporated under reduced pressure, and the solid ob-
tained was washed with ether to afford corresponding derivative 3
or 4.
Figure 4. Predicted docking solution for the S enantiomer of 3e (light
green) in hAChE showing binding in the middle of the binding gorge. Stabi-
lization might occur through p–p interactions to Trp86 and a hydrogen
bond to Asp74. The catalytic site is shown in orange, the choline binding
site in turquoise, the peripheral anionic site in green, and other binding site
residues in gray; hAChE numbering is used. Distances are specified in Š as
italic numbers.
14-Phenyl-9,10,11,14-tetrahydro-13H-benzo[5,6]chromeno[2,3-
d]pyrrolo[1,2-a]pyrimidin-13-imine (3a): Product 3a was obtained
according to the general procedure for the preparation of benzo-
pyrimido[1,2-a]azepin-15-imine (3 f) demonstrated moderate
ChE inhibitory activities and Ab1À42 antiaggregation properties
as well as antioxidant capacity. To sum up, the benzochrome-
nopyrimidinimines reported in this paper constitute an addi-
tional exploration of the quinazolinimine scaffold[20] and
prompt further investigations with this type of motif to enrich
the multitarget-directed ligands pipeline for the potential
treatment of Alzheimer’s disease.
chromenopyrimidines as
a brown solid (325 mg, 89%); mp:
1
>2608C; H NMR (300 MHz, CDCl3): d=8.15 (d, J=8.40 Hz, 1H, H-
1), 7.97–7.78 (m, 2H, H-4, H-5), 7.75–7.06 (m, 9H, N-H, H-2, H-3, H-
6, H-6’, H-2’, H-3’, H-4’, H-5’), 5.93 (s, 1H, H-14), 4.02 (m, 2H, H-11),
3.05 (m, 2H, H-9), 2.16 ppm (m, 2H, H-10); 13C NMR (75 MHz,
CDCl3): d=162.8 (C, C-13), 156.4 (C, C-7a), 147.5 (C, C-8a), 143.9 (C,
C-6a), 131.5 (C, C-1’), 130.8 (C, C-14b), 129.4 (C, C-4a), 128.6 (CH, C-
4), 128.5 (CH, C-5), 128.3 (CH, C-2’, C-6’), 127.8 (CH, C-3’, C-5’), 126.8
(CH, C-2), 126.6 (CH, C-4’), 124.9 (CH, C-3), 123.6 (CH, C-1), 123.4
(CH, C-6), 117.5 (C, C-14a), 97.3 (C, C-13a), 47.0 (CH, C-11), 36.5 (CH,
C-14), 32.1 (CH, C-9), 18.7 ppm (CH, C-10); IR: n˜max =3330,
1593 cmÀ1; HRMS (ESI-TOF): m/z: calcd for C24H20N3O: 366.1590
[M+H]+; found: 366.1601.
Experimental Section
General methods
Melting points were determined with an electrothermal apparatus.
Progress of the reactions was monitored by TLC by using silica gel
covered aluminum sheets (60 F254, Merck). IR spectra were record-
ed with a PerkinElmer PARAGON FTIR spectrometer covering the
15-Phenyl-9,11,12,15-tetrahydro-10H,14H-benzo[5,6]chrome-
no[2,3-d]pyrido[1,2-a]pyrimidin-14-imine (3b): Product 3b was
obtained according to the general procedure for the preparation
of benzochromenopyrimidines as a brown solid (352 mg, 93%);
mp: >2608C; 1H NMR (300 MHz, [D6]DMSO): d=8.30 (d, J=
8.70 Hz, 1H, H-1), 8.27–7.55 (m, 2H, H-4, H-5), 7.52–7.07 (m, 9H, N-
H, H-2, H-3, H-6, H-6’, H-2’, H-3’, H-4’, H-5’), 5.87 (s, 1H, H-15), 3.68
(m, 2H, H-12), 2.74 (m, 2H, H-11), 1.83 ppm (m, 4H, H-9, H-10);
13C NMR (75 MHz, [D6]DMSO): d=158.6 (C, C-14), 154.8 (C, C-7a),
147.3 (C, C-8a), 144.2 (C, C-6a), 131.2 (C, C-1’), 130.7 (C, C-15b),
130.5 (C, C-4a), 130.4 (CH, C-4), 129.1 (CH, C-5), 128.4 (CH, C-2’, C-
6’), 128.2 (CH, C-3’, C-5’), 128 (CH, C-2), 127.3 (CH, C-4’), 126.8 (CH,
C-3), 126.5 (CH, C-1), 124.7 (CH, C-6), 117.4 (C, C-15a), 96.7 (C, C-
14a), 43.7 (CH, C-12), 35.5 (CH, C-15), 31.2 (CH, C-9), 21.5 (CH, C-11),
18.3 ppm (CH, C-10); IR: n˜max =3331, 1591 cmÀ1; HRMS (ESI-TOF):
m/z: calcd for C25H22N3O: 380.1750 [M+H]+; found: 380.1757.
1
field 400–4000 cmÀ1. H NMR and 13C NMR were recorded in CDCl3
or [D6]DMSO with a Bruker spectrometer (1H NMR at 300 MHz,
13C NMR at 75 MHz). Chemical shifts are reported in parts per mil-
lion (ppm) by using tetramethylsilane (SiMe4) as an internal refer-
ence. The multiplicities of the signals are indicated by the follow-
ing abbreviations: s, singlet; d, doublet; t, triplet; q, quadruplet;
and m, multiplet. Coupling constants are expressed in Hz. High-res-
olution mass spectra were obtained at the Centre Commun de
SpectromØtrie de Masse, Lyon, France, with a Bruker micrOTOF-Q II
spectrometer (Bruker Daltonics) in positive ESI-TOF mode. The mi-
crowave-assisted reactions were performed in a synthetic micro-
wave: Monowave 300 apparatus with a maximum power of 300 W.
All tested compounds were found to be ꢀ95% pure by HPLC anal-
ysis by using a Hitachi Lachrom Elite series instrument. 3-Amino-1-
phenyl-1H-benzo[f]chromene-2-carbonitrile (1a),[34] 3-amino-1-(3-
methoxyphenyl)-1H-benzo[f]chromene-2-carbonitrile (1b),[35] 2-
amino-4-phenyl-4H-benzo[h]chromene-3-carbonitrile (2a),[36] and 2-
amino-4-(3-methoxyphenyl)-4H-benzo[h]chromene-3-carbonitrile
(2b)[37] were previously described.
16-Phenyl-9,10,11,12,13,16-hexahydro-15H-benzo[5’,6’]chrome-
no[2’,3’:4,5]pyrimido[1,2-a]azepin-15-imine (3c): Product 3c was
obtained according to the general procedure for the preparation
of benzochromenopyrimidines as a brown solid (354 mg, 90%);
mp: >2608C; 1H NMR (300 MHz, CDCl3): d=8.07 (d, J=8.10 Hz,
1H, H-1), 7.94–7.76 (m, 2H, H-4, H-5), 7.75–7.10 (m, 9H, N-H, H-2,
ChemMedChem 2016, 11, 1318 – 1327
www.chemmedchem.org
1323
ꢀ 2016 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim

Full Papers
H-3, H-6, H-6’, H-2’, H-3’, H-4’, H-5’), 5.48 (s, 1H, H-16), 3.72 (m, 2H,
H-13), 3.45 (m, 2H, H-12), 2.77 (m, 2H, H-11), 2.17 ppm (m, 4H, H-9,
H-10); 13C NMR (75 MHz, CDCl3): d=163.3 (C, C-15), 146.9 (C, C-7a),
130.8 (C, C-8a), 130.3 (C, C-6a), 131.2 (C, C-1’), 130.4 (C, C-16b),
129.1 (C, C-4a), 128.6 (CH, C-4), 128.4 (CH, C-5), 128.1 (CH, C-2’, C-
6’), 127.5 (CH, C-3’, C-5’), 126.9 (CH, C-2), 126.7 (CH, C-4’), 124.9
(CH, C-3), 123.7 (CH, C-1), 122.6 (CH, C-6), 117.1 (C, C-16a), 96.9 (C,
C-15a), 44.9 (CH, C-13), 36.7 (CH, C-16), 29.1 (CH, C-12), 29.0 (CH, C-
7-Phenyl-7,10,11,12-tetrahydro-8H-benzo[7,8]chromeno[2,3-
d]pyrrolo[1,2-a]pyrimidin-8-imine (4a): Product 4a was obtained
according to the general procedure for the preparation of benzo-
chromenopyrimidines as
a brown solid (325 mg, 89%); mp:
1
>2608C; H NMR (300 MHz, CDCl3): d=8.46 (d, J=8.40 Hz, 1H, H-
1), 8.41–7.58 (m, 2H, H-4, H-5), 7.55–7.16 (m, 9H, N-H, H-2, H-3, H-
6, H-6’, H-2’, H-3’, H-4’, H-5’), 5.28 (s, 1H, H-7), 3.14 (m, 2H, H-10),
2.19 (m, 2H, H-12), 1.25 ppm (m, 2H, H-11); 13C NMR (75 MHz,
CDCl3): d=163.2 (C, C-8), 143.4 (C, C-13a), 142.8 (C, C-12a), 142.7
(C, C-14a), 131.5 (C, C-1’), 130.0 (C, C-4a), 129.0 (CH, C-4), 128.5 (CH,
C-2), 128.4 (CH, C-3’, C-5’), 127.5 (CH, C-2’, C-6’), 126.7 (CH, C-4’),
126.5 (C, C-14b), 126.1 (CH, C-3), 124.7 (CH, C-6), 121.5 (CH, C-1),
118.2 (CH, C-5), 117.9 (C, C-6a), 96.0 (C, C-7a), 45.6 (CH, C-10), 40.2
(CH, C-7), 32.5 (CH, C-12), 19.0 ppm (CH, C-11); IR: n˜max =3332,
1597 cmÀ1; HRMS (ESI-TOF): m/z: calcd for C24H20N3O: 366.1600
[M+H]+; found: 366.1601.
9), 26.2 (C, C-11), 24.1 ppm (C, C-10); IR: n˜max =3334, 1591 cmÀ1
;
HRMS (ESI-TOF): m/z: calcd for C26H24N3O: 394.1903 [M+H]+;
found: 394.1914.
14-(3-Methoxyphenyl)-9,10,11,14-tetrahydro-13H-benzo[5,6]-
chromeno[2,3-d]pyrrolo[1,2-a]pyrimidin-13-imine (3d): Product
3d was obtained according to the general procedure for the prep-
aration of benzochromenopyrimidines as a brown solid (332 mg,
84%); mp: >2608C; 1H NMR (300 MHz, CDCl3): d=8.15 (d, J=
8.40 Hz, 1H, H-1), 7.80–7.76 (m, 2H, H-4, H-5), 7.50–6.64 (m, 9H, N-
H, H-2, H-3, H-6, H-6’, H-2’, H-3’, H-4’, H-5’), 5.69 (s, 1H, H-14), 3.71
(s, 3H, OCH3), 3.03 (m, 2H, H-11), 2.21 ppm (m, 4H, H-9, H-10);
13C NMR (75 MHz, CDCl3): d=162.7 (C, C-3’), 159.8 (C, C-13), 150.0
(C, C-7a), 156.7 (C, C-8a), 147.5 (C, C-6a), 144.6 (C, C-1’), 131.4 (C, C-
14b), 130.9 (C, C-14a), 129.4 (CH, C-5’), 128.5 (CH, C-5), 127.0 (CH,
C4), 128.2 (CH, C-2), 124.2 (CH, C-3), 123.3 (CH, C-1), 120.8 (CH, C-
6’), 117.5 (C, CH, C-6, C-4a), 112.4 (CH, C-4’), 97.2 (C, C-13a), 55.1
(OCH3), 48.3 (CH, C-11), 37.3 (CH, C-14), 32.2 (CH, C-9), 18.8 ppm
(CH, C-10); IR: n˜max =3325, 1599 cmÀ1; HRMS (ESI-TOF): m/z: calcd
for C25H22N3O2: 396.1691 [M+H]+; found: 396.1707.
7-Phenyl-10,11,12,13-tetrahydro-7H,8H-benzo[7,8]chromeno[2,3-
d]pyrido[1,2-a]pyrimidin-8-imine (4b): Product 4b was obtained
according to the general procedure for the preparation of benzo-
chromenopyrimidines as
a brown solid (349 mg, 92%); mp:
1
>2608C; H NMR (300 MHz, CDCl3): d=8.46 (d, J=8.10 Hz, 1H, H-
1), 8.20–7.57 (m, 2H, H-4, H-5), 7.55–7.17 (m, 9H, N-H, H-2, H-3, H-
6, H-6’, H-2’, H-3’, H-4’, H-5’), 5.27 (s, 1H, H-7), 3.87 (m, 2H, H-10),
2.66 (m, 2H, H-13), 1.89 ppm (m, 4H, H-11, H-12); 13C NMR
(75 MHz, CDCl3): d=158.3 (C, C-8), 143.1 (C, C-14a), 142.9 (C, C-
13a), 142.6 (C, C-15a), 131.6 (C, C-1’), 130.5 (C, C-4a), 129.1 (CH, C-
4), 128.5 (CH, C-2), 127.5 (CH, C-3’, C-5’), 127.0 (CH, C-2’, C-6’), 126.2
(CH, C-4’), 125.9 (C, C-15b), 125.7 (CH, C-3), 124.0 (CH, C-6), 121.2
(CH, C-1), 118.4 (CH, C-5), 117.5 (C, C-6a), 95.6 (C, C-7a), 45.2 (CH, C-
10), 40.2 (CH, C-7), 31.6 (CH, C-13), 21.5 (CH, C-11), 18.2 ppm (C, C-
12); IR: n˜max =3330, 1591 cmÀ1; HRMS (ESI-TOF): m/z: calcd for
C25H22N3O: 380.1754 [M+H]+; found: 380.1757.
15-(3-Methoxyphenyl)-9,11,12,15-tetrahydro-10H,14H-ben-
zo[5,6]chromeno[2,3-d]pyrido[1,2-a]pyrimidin-14-imine
(3e):
Product 3e was obtained according to the general procedure for
the preparation of benzochromenopyrimidines as a brown solid
(377 mg, 92%); mp: >2608C; 1H NMR (300 MHz, CDCl3): d=8.11
(d, J=8.40 Hz, 1H, H-1), 7.78–7.74 (m, 2H, H-4, H-5), 7.49–6.62 (m,
9H, N-H, H-2, H-3, H-6, H-6’, H-2’, H-3’, H-4’, H-5’), 5.54 (s, 1H, H-
15), 3.67 (s, 3H, OCH3), 3.88 (m, 2H, H-12), 2.86 (m, 2H, H-11), 1.92
(m, 2H, H-9), 1.83 ppm (m, 2H, H-10); 13C NMR (75 MHz, CDCl3): d=
159.3 (C, C-3’), 158.8 (C, C-14), 157.9 (C, C-7a), 154.9 (C, C-8a), 147.0
(C, C-6a), 144.1 (C, C-1’), 131.0 (C, C-15b), 130.8 (CH, C-15a), 129.5
(CH, C-5’), 128.9 (CH, C-5), 128.1 (CH, C4), 126.4 (CH, C-2), 124.2
(CH, C-3), 122.7 (CH, C-1), 120.4 (CH, C-6’), 117.2 (C, CH, C-6, C-4a),
114.2 (CH, C-2’), 114.4 (CH, C-4’), 96.6 (C, C-14a), 54.5 (OCH3), 44.17
(CH, C-15), 37.4 (CH, C-12), 31.4 (CH, C-9), 21.7 (CH, C-11), 18.4 ppm
(CH, C-10); IR: n˜max =3329, 1594 cmÀ1; HRMS (ESI-TOF): m/z: calcd
for C26H24N3O2: 410.1850 [M+H]+; found: 410.1863.
7-Phenyl-7,10,11,12,13,14-hexahydro-8H-benzo[7’,8’]chrome-
no[2’,3’:4,5]pyrimido[1,2-a]azepin-8-imine (4c): Product 4c was
obtained according to the general procedure for the preparation
of benzochromenopyrimidines as a brown solid (354 mg, 90%);
mp: >2608C; 1H NMR (300 MHz, CDCl3): d=8.49 (d, J=7.80 Hz,
1H, H-1), 7.76–7.57 (m, 2H, H-4, H-5), 7.54–7.13 (m, 9H, N-H, H-2,
H-3, H-6, H-6’, H-2’, H-3’, H-4’, H-5’), 5.03 (s, 1H, H-7), 3.01 (m, 2H,
H-10), 2.80–1.29 ppm (m, 8H, H-14, H-11, H-12, H-13); 13C NMR
(75 MHz, CDCl3): d=164.1 (C, C-8), 159.2 (C, C-15a), 156.3 (C, C-
14a), 143.9 (C, C-16a), 143.5 (C, C-1’), 133.1 (C, C-4a), 130.0 (CH, C-
4), 129.1 (CH, C-2), 128.8 (CH, C-3’, C-5’), 127.4 (CH, C-2’, C-6’), 126.8
(CH, C-4’), 126.6 (C, C-16b), 126.1 (CH, C-3), 125.7 (CH, C-6), 122.5
(CH, C-1), 121.9 (CH, C-5), 117.8 (C, C-6a), 96.0 (C, C-7a), 43.0 (CH, C-
10), 40.4 (CH, C-7), 37.2 (CH, C-14), 27.3 (CH, C-11), 26.8 (CH, C-13),
24.7 ppm (CH, C-12); IR: n˜max =3333, 1595 cmÀ1; HRMS (ESI-TOF):
m/z: calcd for C26H24N3O: 394.1905 [M+H]+; found: 394.1914.
16-(3-Methoxyphenyl)-9,10,11,12,13,16-hexahydro-15H-ben-
zo[5’,6’]chromeno[2’,3’:4,5]pyrimido[1,2-a]azepin-15-imine (3 f):
Product 3 f was obtained according to the general procedure for
the preparation of benzochromenopyrimidines as a brown solid
(364 mg, 86%); mp: >2608C; 1H NMR (300 MHz, CDCl3): d=8.13
(d, J=8.40 Hz, 1H, H-1), 7.80–7.51 (m, 2H, H-4, H-5), 7.50–6.63 (m,
9H, N-H, H-2, H-3, H-6, H-6’, H-2’, H-3’, H-4’, H-5’), 5.52 (s, 1H, H-
16), 3.71 (s, 3H, OCH3), 2.95 (m, 2H, H-13, H-12), 1.79 ppm (m, 6H,
H-9, H-10, H-11); 13C NMR (75 MHz, CDCl3): d=163.8 (C, C-3’), 159.8
(C, C-15), 159.1 (C, C-7a), 155.6 (C, C-8a), 147.4 (C, C-6a), 144.5 (C,
C-1’), 131.5 (C, C-16b), 130.9 (C, C-16a), 130.0 (CH, C-5’), 129.1 (CH,
C-5), 128.6 (CH, C-4), 126.9 (CH, C-2), 124.7 (CH, C-3), 123.2 (CH, C-
1), 120.9 (CH, C-6’), 117.5 (C, CH, C-6, C-4a), 114.8 (CH, C-2’), 111.9
(CH, C-4’), 97.4 (C, C-15a), 55.0 (OCH3), 44.7 (CH, C-13), 38.2 (CH, C-
16), 37.3 (CH, C-12), 29.5 (CH, C-9), 26.7 (CH, C-11), 24.6 ppm (CH,
C-10); IR: n˜max =3328, 1593 cmÀ1; HRMS (ESI-TOF): m/z: calcd for
C27H26N3O2: 424.2014 [M+H]+; found: 424.2020.
7-(3-Methoxyphenyl)-7,10,11,12-tetrahydro-8H-benzo[7,8]chro-
meno[2,3-d]pyrrolo[1,2-a]pyrimidin-8-imine (4d): Product 4d
was obtained according to the general procedure for the prepara-
tion of benzochromenopyrimidines as a brown solid (352 mg,
89%); mp: >2608C; 1H NMR (300 MHz, CDCl3): d=8.46 (d, J=
8.10 Hz, 1H, H-1), 7.77–7.57 (m, 2H, H-4, H-5), 7.54–6.72 (m, 9H, N-
H, H-2, H-3, H-6, H-6’, H-2’, H-3’, H-4’, H-5’), 5.08 (s, 1H, H-7), 3.73 (s,
3H, OCH3), 4.08 (m, 2H, H-10), 3.08 (m, 2H, H-12), 2.22 ppm (m,
2H, H-11); 13C NMR (75 MHz, CDCl3): d=163.1 (C, C-3’), 160.1 (C, C-
8), 158.3 (C, C-13a), 157.1 (C, C-14a), 145.6 (C, C-1’), 143.5 (C, C-4a),
133.6 (CH, C-5’), 130.1 (CH, C-4), 130.0 (C, C-14b), 127.4 (CH, C-2),
126.8 (CH, C-6), 126.0 (CH, C-3), 124.2 (CH, C-1), 121.7 (C, C-6a),
120.4 (CH, C-6’), 117.6 (CH, C-5), 114.2 (CH, C-2’), 112.3 (CH, C-4’),
95.6 (C, C-7a), 55.1 (OCH3), 48.1 (CH, C-10), 41.0 (CH, C-7), 32.3 (CH,
ChemMedChem 2016, 11, 1318 – 1327
www.chemmedchem.org
1324
ꢀ 2016 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim

Full Papers
C-12), 18.8 ppm (CH, C-11); IR: n˜max =3335, 1597 cmÀ1; HRMS (ESI-
TOF): m/z: calcd for C25H22N3O2: 396.1701 [M+H]+; found:
396.1707.
calculated according to Equation (1):
fi
ð1Þ
AUC ¼ 1 þ sumð
Þ
f0
7-(3-Methoxyphenyl)-10,11,12,13-tetrahydro-7H,8H-benzo[7,8]-
chromeno[2,3-d]pyrido[1,2-a]pyrimidin-8-imine (4e): Product 4e
was obtained according to the general procedure for the prepara-
tion of benzochromenopyrimidines as a brown solid (389 mg,
95%); mp: >2608C; 1H NMR (300 MHz, CDCl3): d=8.47 (d, J=
8.10 Hz, 1H, H-1), 7.76–7.56 (m, 2H, H-4, H-5), 7.53–6.71 (m, 9H, N-
H, H-2, H-3, H-6, H-6’, H-2’, H-3’, H-4’, H-5’), 5.06 (s, 1H, H-7), 3.72 (s,
3H, OCH3), 3.87 (m, 2H, H-10), 2.92 (m, 2H, H-13), 1.99 ppm (m,
4H, H-11, H-12); 13C NMR (75 MHz, CDCl3): d=160.1 (C, C-3’), 159.4
(C, C-8), 158.8 (C, C-14a), 156.3 (C, C-15a), 145.5 (C, C-1’), 143.5 (C,
C-4a), 133.1 (CH, C-5’), 131.5 (CH, C-4), 130.0 (C, C-15b), 127.4 (CH,
C-2), 126.6 (CH, C-6), 126.1 (CH, C-3), 124.1 (CH, C-1), 121.8 (C, C-
6a), 120.4 (CH, C-6’), 117.6 (CH, C-5), 114.2 (CH, C-2’), 112.3 (CH, C-
4’), 95.7 (C, C-7a), 55.1 (OCH3), 44.7 (CH, C-10), 41.3 (CH, C-7), 32.0
(CH, C-13), 22.1 (CH, C-12), 18.9 ppm (CH, C-11); IR: n˜max =3332,
1595 cmÀ1; HRMS (ESI-TOF): m/z: calcd for C26H24N3O2: 410.1854
[M+H]+; found: 410.1863.
in which f0 is the initial fluorescence reading at 0 min and fi is the
fluorescence value at time i.
The net AUC corresponding to a sample was calculated according
to Equation (2):
Net AUC ¼ AUCantioxidantÀAUCblank
ð2Þ
Regression equations were calculated by plotting the net AUC
against the antioxidant concentration. The ORAC value was ob-
tained by dividing the slope of the latter curve by the slope of the
Trolox curve obtained in the same assay. Final ORAC values were
expressed as (mmol Trolox equivalent)(mmol benzochromenopyri-
midinimines)À1. Data are expressed as meanÆSD.
Biochemical methods
7-(3-Methoxyphenyl)-7,10,11,12,13,14-hexahydro-8H-ben-
zo[7’,8’]chromeno[2’,3’:4,5]pyrimido[1,2-a]azepin-8-imine
Inhibition of EeAChE and eqBuChE: Inhibitory activity on ChEs was
performed following the spectrophotometric method of Ellman[27]
by using purified AChE from Electrophorus electricus (Type V-S,
Sigma–Aldrich) or BuChE from horse serum (lyophilized powder,
Sigma–Aldrich). Enzymes were first dissolved in 0.1 m phosphate
buffer (pH 8.0) and then aliquoted in small vials. Compounds stock
solutions in DMSO (10 mm) were diluted with DMSO to prepare
nine serial dilutions of each compound. The reaction was per-
formed in a final volume of 3 mL of a 0.1m phosphate-buffered so-
lution at pH 8.0, containing 5,5’-dithiobis-2-nitrobenzoic acid
(DTNB, 2625 mL, 0.35 mm, final concentration), EeAChE (29 mL,
0.035 UmLÀ1 final concentration) or eqBuChE (60 mL, 0.05 UmLÀ1
final concentration), tested compound (3 mL, different concentra-
tions), and 1%w/v bovine albumin serum phosphate-buffered
(pH 8.0) solution (BSA, 40 mL). The inhibition with respect to control
without compound was determined by pre-incubating this mixture
at room temperature (RT) with nine concentrations of each com-
pound for 10 min. After this period, acetylthiocholine iodide
(105 mL, 0.35 mm, final concentration) or butyrylthiocholine iodide
(150 mL, 0.5 mm final concentration) was added and incubated for
15 min at RT. Then, the absorbances were measured at l=412 nm
in a spectrophotometer plate reader (iEMS Reader MF, Labsystems).
IC50 values were calculated graphically as the concentration of
compound that decreases 50% of the enzymatic activity by using
a nonlinear regression method (four parameter logistic method).
Data are expressed as meanÆSEM of at least three different ex-
periments performed in quadruplicate.
(4 f):
Product 4 f was obtained according to the general procedure for
the preparation of benzochromenopyrimidines as a brown solid
(394 mg, 93%); mp: >2608C; 1H NMR (300 MHz, CDCl3): d=8.49
(d, J=8.10 Hz, 1H, H-1), 8.46–7.73 (m, 2H, H-4, H-5), 7.56–6.73 (m,
9H, N-H, H-2, H-3, H-6, H-6’, H-2’, H-3’, H-4’, H-5’), 4.95 (s, 1H, H-7),
3.73 (s, 3H, OCH3), 3.01 (m, 2H, H-10), 1.77 ppm (m, 8H, H-14, H-11,
H-12, H-13); 13C NMR (75 MHz, CDCl3): d=164.1 (C, C-3’), 160.2 (C,
C-8), 159.4 (C, C-15a), 156.0 (C, C-16a), 145.6 (C, C-1’), 143.5 (C, C-
4a), 133.5(CH, C-5’), 131.2 (CH, C-4), 130.0 (C, C-16b), 127.4 (CH, C-
2), 126.8 (CH, C-6), 126.2 (CH, C-3), 124.0 (CH, C-1), 121.9 (C, C-6a),
120.6 (CH, C-6’), 117.6 (CH, C-5), 114.0 (CH, C-2’), 112.4 (CH, C-4’),
95.8 (C, C-7a), 55.1 (OCH3), 44.1 (CH, C-10), 41.8 (CH, C-7), 37.2 (CH,
C-14), 29.6 (CH, C-11), 26.8 (CH, C-13), 24.8 ppm (CH, C-12); IR:
n˜max =3330, 1594 cmÀ1; HRMS (ESI-TOF): m/z: calcd for C27H26N3O2:
424.2015 [M+H]+; found: 424.2020.
Oxygen radical absorbance capacity assay: The radical scaveng-
ing capacity of the new compounds was determined by the ORAC-
FL method, for which fluorescein was used as a fluorescent
probe[24,25] by using a Varioskan Flash plate reader with built-in in-
jectors (Thermo Scientific). (Æ)-6-Hydroxy-2,5,7,8-tetramethylchro-
mane-2-carboxylic acid (Trolox), fluorescein (FL), and 2,2’-azobis(a-
midinopropane) dihydrochloride (AAPH) were purchased from
Sigma–Aldrich. The reaction was performed at 378C in 75 mm
phosphate buffer (pH 7.4). The final volume of the reaction mixture
was 200 mL. Trolox and tested compounds were dissolved in DMSO
(10 mm) and further diluted in 75 mm phosphate buffer (pH 7.4).
The final concentrations were 1–8 mm for Trolox standard and 0.1–
1 mm for the tested compounds. In a black 96-well microplate
(Nunc), antioxidant (20 mL) and fluorescein (FL, 120 mL, final con-
centration of 70 nm) were incubated for 15 min at 378C. Then,
AAPH (60 mL, final concentration of 40 mm) solution was added
rapidly by using the built-in injector. The fluorescence was mea-
sured every minute for a 60 min period (excitation wavelength
485 nm and emission wavelength 535 nm). For each assay a blank
composed of FL solution (120 mL), AAPH (60 mL), and phosphate
buffer (pH 7.4, 20 mL) was used. All the experiments were made in
triplicate and at least three different assays were performed for
each sample. Antioxidant curves (fluorescence versus time) were
first normalized to the curve of the blank (without antioxidant),
and then, the area under the fluorescence decay curve (AUC) was
In vitro inhibition of hAChE: To assess the inhibitory activity of the
target compounds toward hAChE, we followed Ellman’s assay[27] by
using human recombinant AChE (Sigma–Aldrich). hAChE (500 U)
was dissolved in 1 mL of a gelatin solution (1% in water) and dilut-
ed with demineralized water to give a stock solution of 5 UmLÀ1
.
The hAChE solution was further diluted before use to give a final
concentration of 3.125 UmLÀ1. The 12.5 mm 5,5’-dithiobis(2-nitro-
benzoic acid) (DTNB, Ellman’s reagent) solution containing
0.15%w/v sodium carbonate and the 18.75 mm acetylthiocholine
(ATC) iodide solution were prepared in demineralized water. All
assays were performed in 0.1m phosphate buffer, pH 8.0. The test
compounds or water in a blank sample (25 mL) were incubated
with the enzyme (20 mL) for 5 min at 378C in buffer (765 mL) prior
to start the reaction. Then, DTNB (20 mL) and ATC (20 mL) were
added. After 5 min, changes in absorbance were measured at l=
ChemMedChem 2016, 11, 1318 – 1327
www.chemmedchem.org
1325
ꢀ 2016 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim

Full Papers
412 nm in EnSpire Multimode Microplate Reader (PerkinElmer). The
percentage of inhibition of the enzyme was calculated by compari-
son with a blank sample (100% enzyme activity). Calculation of
IC50 values was performed with GraphPad Prism 5. Each concentra-
tion was measured in triplicate. Data is expressed as meanÆSEM.
bottom þ ðtopÀbottomÞ
f1 þ 10½ðlog IC50ÀXÞ Â Hill slopeŠg
ð4Þ
Y ¼
Kinetic characterization of hBuChE inhibition: The kinetic studies
were performed by using the method of Ellman as described
above.[27] The stock solution of butyrylthiocholine was prepared in
0.1m phosphate buffer (1.5 mm) and was diluted before use. Final
concentrations of butyrylthiocholine in the wells were 500, 300,
200, 100, and 50 mm. The reactions were started by adding the
substrate to the enzyme and inhibitor (of various concentrations)
that had been pre-incubated for 5 min to allow complete equilibra-
tion of the enzyme–inhibitor complexes. The initial velocity (v0) in
the absence and presence of the test compound was calculated
from the slope of the linear trend obtained, with each measure-
ment performed in triplicate. Lineweaver–Burk plots were calculat-
ed by using linear regression in GraphPad Prism 6.
Kinetic characterization of hAChE inhibition: To estimate the type of
inhibition of hAChE, we performed the same experimental protocol
as that reported for hAChE inhibition. Different concentrations of
the substrate ATC (0.067–0.5 mm) were used to create Lineweaver–
Burk plots by plotting the inverse initial velocity (1/V) as a function
of the inverse of the substrate concentration (1/[S]). The stock solu-
tion of ATC (0.5 mm in a well) was prepared in water and diluted
before use to obtain 0.4, 0.3, 0.2, 0.1, and 0.067 mm concentrations
of substrate. The double reciprocal plots were assessed by
a weighted least-square analysis that assumed the variance of V to
be a constant percentage of V for the entire data set. Each experi-
ment was performed in triplicate. To confirm the mode of inhibi-
tion, Cornish–Bowden plots were obtained by plotting S/V (sub-
strate/velocity ratio) versus the inhibitor concentration.[27,38] Data
analysis was performed with GraphPad Prism 5. Calculation of the
inhibitor constant (Ki) value was performed by replotting slopes of
lines from the Lineweaver–Burk plot versus the inhibitor concentra-
tion, and Ki was determined as the intercept on the negative
x axis.[39]
Ab1À42 aggregation inhibitory activity: Thioflavin T (ThT) fluorimetric
assay was performed to investigate the effect of the test com-
pounds on the aggregation of Ab1À42
[29]
.
Recombinant human
1,1,1,3,3,3-hexafluoro-2-propanol (HFIP)-pretreated Ab1À42 peptide
(lot number 2387442, Merck Millipore, Darmstadt, Germany) was
dissolved in DMSO. Prior to the incubations, the Ab1À42 peptide
stock solution was diluted in 150 mm 4-(2-hydroxyethyl)piperazine-
1-ethanesulfonic acid (HEPES) buffer (pH 7.4) containing 150 mm
NaCl to give a concentration of 7.5 mm. Then, 20 mL Ab1À42 was
mixed with 10 mL of the test compounds (100 mm stock in HEPES;
10 mm final concentration), added to the corresponding wells in
black-walled 96-well plates, and diluted with 70 mL ThT solution
(14.3 mm stock solution in HEPES; 10 mm final concentration) to
a final volume of 100 mL (1.5 mm final Ab1À42 concentration). Each
sample was prepared in quadruplicate, and the DMSO was always
at 3%. To quantify amyloid fibril formation, the ThT fluorescence
was measured through the top of the plate every 300 s at an exci-
tation wavelength of 440 nm and emission wavelength of 490 nm,
with the medium continuously shaking between measurements by
using a 96-well microplate reader (Synergy H4, BioTek Instruments,
Inc., USA). The ThT emission of the Ab1À42 began to rise after 4 h,
reached a plateau after 20 h, and remained almost unchanged for
an additional 28 h of incubation. The fluorescence intensities at
the plateau in the absence and presence of the test compounds
were averaged, and the average fluorescence of the corresponding
wells at t=0 h was subtracted. The Ab1À42 aggregation inhibitory
potencies are expressed as the percentage inhibition [Eq. (5)]:
Inhibition of human BuChE: The inhibitory potencies of compounds
against hBuChE was determined by using the method of Ellman.[27]
5,5’-Dithiobis(2-nitrobenzoic acid) (Ellman’s reagent; DTNB) and bu-
tyrylthiocholine iodide were purchased from Sigma–Aldrich (Stein-
heim, Germany). Recombinant hBuChE at the stock concentration
of 4.6 mgmLÀ1 in 10 mm 2-(N-morpholino)ethanesulfonic acid
buffer (pH 6.5) was kindly donated by Florian Nachon (IBS, Greno-
ble). The enzyme solution was prepared by dilution of the concen-
trated stocks in phosphate-buffered solution (0.1m, pH 8.0). The re-
actions were performed in a final volume of 300 mL 0.1m phos-
phate-buffered solution, pH 8.0, containing 333 mm DTNB, 500 mm
butyrylthiocholine, and 1 nm hBChE. The reaction was started by
adding the substrate at RT. The final content of DMSO was always
1%. The formation of the yellow 5-thio-2-nitrobenzoate anion as
a result of the reaction of DTNB with the thiocholines was moni-
tored for 1 min as the change in absorbance at l=412 nm by
using a 96-well microplate reader (Synergy H4, BioTek Instruments,
Inc., USA). To determine the blank value (b), phosphate-buffered
solution replaced the enzyme solution. The initial velocity (v0) was
calculated from the slope of the linear trend obtained, with each
measurement performed in triplicate. For the IC50 measurements,
seven different concentrations of each compound dissolved in
DMSO were used to obtain enzyme activities of between 5 and
90%. The reactions were started by adding the substrate to the
enzyme and inhibitor that had been pre-incubated for 5 min to
allow complete equilibration of the enzyme–inhibitor complexes.
The initial velocities in the presence of the test compounds (vi)
were calculated. The inhibitory potencies are expressed as the re-
sidual activities (RAs) [Eq. 3]. The IC50 values were obtained by plot-
ting the residual enzyme activities against the applied inhibitor
concentrations, with the experimental data fitted to Equation (4),
in which X is the logarithm of the inhibitor concentration and Y is
the residual activity. For the fitting procedure, the Gnuplot soft-
ware and an in-house python script were used. Tacrine was used
as positive control.
1ÀFi
ð5Þ
% Inhibition ¼ ð
Þ Â 100 %
F0
in which Fi is the increase in fluorescence of Ab1À42 treated with
the test compounds, and F0 is the increase in fluorescence of
Ab1À42 alone.
Docking protocols: Docking studies were conducted with the pro-
gram Gold v.5.2.2[32] by using the crystal structure PDB ID: 4EY5 of
hAChE (resolution: 2.30 Š).[31] Protein and ligand preparation was
performed as described in a previous study.[33] The binding site
was defined as stated by F. H. Darras et al.[40]
Among the four tested scoring functions in Gold, GoldScore[41,42]
with the numbers of operation set to 1, Mio performed best in re-
docking studies by using 1EVE,[43] 2CKM, and 4EY7.[31] For 1EVE, the
lowest RMSD pose (0.51 Š) was found on the highest rank relative
to the other scoring functions. For 2CKM, GoldScore showed the
lowest RMSD to the original pose (0.63 Š), and in 4EY7 the top
pose and the pose with lowest RMSD to the top pose (0.39 Š)
ðviÀbÞ
ð3Þ
RA ¼
ðvoÀbÞ
ChemMedChem 2016, 11, 1318 – 1327
www.chemmedchem.org
1326
ꢀ 2016 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim

Full Papers
si, M. I. Rodríguez-Franco, M. Decker, J. Marco-Contelles, L. Ismaili,
ChemMedChem 2015, 10, 523–539.
[18] M. Decker, F. Krauth, J. Lehmann, Bioorg. Med. Chem. 2006, 14, 1966–
1977.
[19] M. Decker, J. Med. Chem. 2006, 49, 5411–5413.
[20] M. Decker, B. Kraus, J. Heilmann, Bioorg. Med. Chem. 2008, 16, 4252–
4261.
were in the biggest cluster. GoldScore also gave consistent results
in previous AChE docking studies.[33]
Ligand structures were built in both enantiomeric forms and pro-
tonated at the iminium nitrogen atom, as in the study of Chen
et al.[44] and based on pKa calculations with MoKa.[45]
Docking of 3e was performed by using the same parameters as
those used for redocking. The pose with highest score of the big-
gest cluster was chosen as representative docking solution. Clus-
tering was performed with the complete linkage method applying
a cluster cut-off of 1.5 Š. For further optimization of this pose, re-
scoring with MiniMuDS in the DSX package was performed.[46,47]
Scores showed a preference for the S enantiomers, which might be
due to hydrogen bonding of the imine function to the Asp74 resi-
due, which cannot be formed by the R enantiomer of 3e.
[21] V. E. Gregor (Warner-Lambert Co., Philadelphia, PA, USA), Quinazoline
Derivatives as Acetylcholinesterase Inhibitors, US Pat. No. US5486512
(A), 1996.
[22] J. E. Biggs-Houck, A. Younai, J. T. Shaw, Curr. Opin. Chem. Biol. 2010, 14,
371–382.
[23] I. Tomassoli, L. Ismaili, M. Pudlo, C. de Los Ríos, E. Soriano, I. Colmena, L.
Gandía, L. Rivas, A. Samadi, J. Marco-Contelles, B. Refouvelet, Eur. J.
Med. Chem. 2011, 46, 1–10.
[24] B. Ou, M. Hampsch-Woodill, R. L. Prior, J. Agric. Food Chem. 2001, 49,
4619–4626.
[25] A. Dµvalos, C. Gómez-CordovØs, B. BartolomØ, J. Agric. Food Chem.
2004, 52, 48–54.
[26] L. Fang, B. Kraus, J. Lehmann, J. Heilmann, Y. Zhang, M. Decker, Bioorg.
Med. Chem. Lett. 2008, 18, 2905–2909.
[27] G. L. Ellman, K. D. Courtney, V. Andres, Jr., R. M. Featherstone, Biochem.
Pharmacol. 1961, 7, 88–95.
Distances were measured in PyMOL[48] and geometric centers were
used for distance measurements if aromatic ring systems were in-
volved. The figures of the docking poses were made with
PyMOL.[48]
[28] Y.-H. Wang, Q.-L. Wan, C.-D. Gu, H.-R. Luo, C.-L. Long, Chem. Cent. J.
2012, 6, 96.
[29] H. LeVine 3rd, Protein Sci. 1993, 2, 404–410.
Acknowledgements
[30] F. Mao, J. Yan, J. Li, X. Jia, H. Miao, Y. Sun, L. Huang, X. Li, Org. Biomol.
Chem. 2014, 12, 5936.
Appreciation is expressed to Professor Christoph A. Sotriffer (Uni-
versity of Würzburg, Germany) for helpful discussions with S.W.
about the docking data. The award of a PhD scholarship to S.W.
by the German Academic National Foundation (Studienstiftung
des deutschen Volkes) is gratefully acknowledged.
[31] J. Cheung, M. J. Rudolph, F. Burshteyn, M. S. Cassidy, E. N. Gary, J. Love,
M. C. Franklin, J. J. Height, J. Med. Chem. 2012, 55, 10282–10286.
[32] GOLD Suite ver. 5.2, CCDC Software: www.ccdc.cam.ac.uk/solutions/csd-
discovery/components/gold/; G. Jones, P. Willett, R. C. Glen, A. R. Leach,
R. Taylor, J. Mol. Biol. 1997, 267, 727–748.
[33] F. H. Darras, S. Wehle, G. Huang, C. A. Sotriffer, M. Decker, Bioorg. Med.
Chem. 2014, 22, 4867–4881.
Keywords: Alzheimer’s disease · antioxidants · inhibitors ·
molecular modeling · multitarget-directed ligands
[34] M. G. Dekamin, M. Eslami, Green Chem. 2014, 16, 4914–4921.
[35] E. Maalej, F. Chabchoub, A. Samadi, C. de Los Ríos, A. Perona, A. Mor-
reale, J. Marco-Contelles, Bioorg. Med. Chem. Lett. 2011, 21, 2384–2388.
[36] E. Maalej, F. Chabchoub, M. J. Oset-Gasque, M. Esquivias-PØrez, M. P.
Gonzµlez, L. Monjas, C. PØrez, C. de Los Ríos, M. I. Rodríguez-Franco, I.
Iriepa, I. Moraleda, M. Chioua, A. Romero, J. Marco-Contelles, A. Samadi,
Eur. J. Med. Chem. 2012, 54, 750–763.
[1] H. W. Querfurth, F. M. LaFerla, N. Engl. J. Med. 2010, 362, 329–344.
[2] M. J. Prince, F. Wu, Y. Guo, L. M. Gutierrez Robledo, M. O’Donnell, R. Sul-
livan, S. Yusuf, Lancet 2015, 385, 549–562.
[3] P. Davies, A. J. Maloney, Lancet 1976, 308, 1403.
[4] S. M. de La Monte, Ann. Neurol. 1989, 25, 450–459.
[5] J.-C. Desport, P. Couratier, Nutr. Clin. Metab. 2002, 16, 253–259.
[6] Y. Christen, Am. J. Clin. Nutr. 2000, 71, 621s–629s.
[7] V. N. Talesa, Mech. Ageing Dev. 2001, 122, 1961–1969.
[8] P. B. Watkins, H. J. Zimmerman, M. J. Knapp, S. I. Gracon, K. W. Lewis,
JAMA J. Am. Med. Assoc. 1994, 271, 992–998.
[9] B. S. P. Marcin Sadowski, B. S. P. Thomas Wisniewski, Curr. Pharm. Des.
2007, 13, 1943–1954.
[10] M. Bartolini, C. Bertucci, V. Cavrini, V. Andrisano, Biochem. Pharmacol.
2003, 65, 407–416.
[11] R. Morphy, Z. Rankovic, Curr. Pharm. Des. 2009, 15, 587–600.
[12] A. Agis-Torres, M. Sollhuber, M. Fernandez, J. M. Sanchez-Montero, Curr.
Neuropharmacol. 2014, 12, 2–36.
[13] A. Cavalli, M. L. Bolognesi, A. Minarini, M. Rosini, V. Tumiatti, M. Recana-
tini, C. Melchiorre, J. Med. Chem. 2008, 51, 347–372.
[14] R. León, A. G. Garcia, J. Marco-Contelles, Med. Res. Rev. 2013, 33, 139–
189.
[15] A. Rampa, F. Belluti, S. Gobbi, A. Bisi, Curr. Top. Med. Chem. 2011, 11,
2716–2730.
[37] Y. Zhen, H. Lin, S. Wang, M. Tao, RSC Adv. 2014, 4, 26122–26128.
[38] A. Cornish-Bowden, Biochem. J. 1974, 137, 143–144.
[39] R. B. Silverman, The Organic Chemistry of Enzyme-Catalyzed Reactions,
Academic Press, San Diego, 2000.
[40] F. H. Darras, S. Pockes, G. Huang, S. Wehle, A. Strasser, H.-J. Wittmann,
M. Nimczick, C. A. Sotriffer, M. Decker, ACS Chem. Neurosci. 2014, 5,
225–242.
[41] G. Jones, P. Willett, R. C. Glen, A. R. Leach, R. Taylor, J. Mol. Biol. 1997,
267, 727–748.
[42] G. Jones, P. Willett, R. C. Glen, J. Mol. Biol. 1995, 245, 43–53.
[43] G. Kryger, I. Silman, J. L. Sussman, Structure 1999, 7, 297–307.
[44] X. Chen, I. G. Tikhonova, M. Decker, Bioorg. Med. Chem. 2011, 19, 1222–
1235.
[45] F. Milletti, L. Storchi, G. Sforna, G. Cruciani, J. Chem. Inf. Model. 2007, 47,
2172–2181.
[46] A. Spitzmüller, H. F. G. Velec, G. Klebe, J. Chem. Inf. Model. 2011, 51,
1423–1430.
[47] G. Neudert, G. Klebe, J. Chem. Inf. Model. 2011, 51, 2731–2745.
[48] The PyMOL Molecular Graphics System, ver. 1.7.4.3, Schrçdinger LLC,
New York, NY (USA), 2015: www.pymol.org.
[16] M. Benchekroun, L. Ismaili, M. Pudlo, V. Luzet, T. Gharbi, B. Refouvelet, J.
Marco-Contelles, Future Med. Chem. 2015, 7, 15–21.
[17] M. Benchekroun, M. Bartolini, J. Egea, A. Romero, E. Soriano, M. Pudlo,
V. Luzet, V. Andrisano, M.-L. Jimeno, M. G. López, S. Wehle, T. Gharbi, B.
Refouvelet, L. de AndrØs, C. Herrera-Arozamena, B. Monti, M. L. Bologne-
Received: November 18, 2015
Published online on January 25, 2016
ChemMedChem 2016, 11, 1318 – 1327
www.chemmedchem.org
1327
ꢀ 2016 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim