Synthesis and biological assessment of diversely substituted furo[2,3-b]quinolin-4-amine and pyrrolo[2,3-b]quinolin-4-amine derivatives, as novel tacrine analogues

at SciVerse ScienceDirect

European Journal of Medicinal Chemistry 46 (2011) 6119e6130

Contents lists available

European Journal of Medicinal Chemistry

journal homepage: http: / /www.elsevier .com/locate/ejmech

Laboratory note

Synthesis and biological assessment of diversely substitutedfuro[2,3-b]quinolin-4-amine and pyrrolo[2,3-b]quinolin-4-aminederivatives, as novel tacrine analoguesq

Carla Martins a, M. Carmo Carreiras a,*, Rafael León b,c, Cristóbal de los Ríos d,e, Manuela Bartolini f,Vincenza Andrisano f, Isabel Iriepa d, Ignacio Moraleda d, Enrique Gálvez d, Manuela García c,g,Javier Egea c, Abdelouhaid Samadi b, Mourad Chioua b, José Marco-Contelles b,**

aResearch Institute for Medicines and Pharmaceutical Sciences (iMed.UL), Faculty of Pharmacy, University of Lisbon, Av. Prof. Gama Pinto, 1649-003 Lisbon, Portugalb Laboratorio de Radicales Libres y Química Computacional (IQOG, CSIC), C/Juan de la Cierva 3, 28006 Madrid, Spainc Instituto Teófilo Hernando and Departamento de Farmacología y Terapéutica, Facultad de Medicina, Universidad Autónoma de Madrid, C/Arzobispo Morcillo 4, 28029 Madrid, SpaindDepartamento de Química Orgánica, Facultad de Farmacia, Universidad de Alcalá, Ctra. Barcelona, Km. 33.6, 28817 Alcalá de Henares, Spaine Instituto de Investigación Sanitaria, Hospital Universitario de la Princesa, C/Diego de León, 62, 28006 Madrid, SpainfDepartment of Pharmaceutical Sciences, Alma Mater Studiorum, University of Bologna, Via Belmeloro 6, 40126 Bologna, ItalygHospital La Paz Health Research Institute-IdiPAZ, Madrid, Spain

a r t i c l e i n f o

Article history:Received 21 March 2011Received in revised form20 September 2011Accepted 21 September 2011Available online 29 September 2011

Keywords:AChE/BuChE inhibitorsNeuroprotectionAlzheimer’s disease

q This manuscript is dedicated to Prof. Christian Brubirthday.* Corresponding author. Tel.: þ351 96 7070270; fax** Corresponding author. Fax: þ34 91 5644853.

E-mail addresses: [email protected], [email protected] (J. Marco-Contelles).

0223-5234/$ e see front matter � 2011 Elsevier Masdoi:10.1016/j.ejmech.2011.09.038

a b s t r a c t

The synthesis and pharmacological analyses of a number of furo[2,3-b]quinolin-4-amine, and pyrrolo[2,3-b]quinolin-4-amine derivatives are reported. Thus, we synthesized diversely substituted tacrineanalogues 1e11 and 12e16 by Friedländer-type reaction of readily available o-amino(furano/pyrrolo)nitriles with suitable and selected cycloalkanones. The biological evaluation of furanotacrines 1e11 andpyrrolotacrine 13 showed that these are good, in the micromolar range, and highly selective inhibitors ofBuChE. In the furanotacrine group, the most interesting inhibitor was 2-(p-tolyl)-5,6,7,8-tetrahydrofuro[2,3-b]quinolin-4-amine (3) [IC50 (eqBuChE)¼ 2.9� 0.4 mM; IC50 (hBuChE)¼ 119� 15 mM]. Conversely,pyrrolotacrines 12 and 14 proved moderately equipotent for both cholinesterases, being 1,2-diphenyl-5,6,7,8-tetrahydro-1H-pyrrolo[2,3-b]quinolin-4-amine (12) the most potent for the inhibition of bothenzymes [IC50 (EeAChE)¼ 0.61� 0.04 mM; IC50 (eqBuChE)¼ 0.074� 0.009 mM]. Moreover, pyrrolotacrine12, at concentrations as low as 300 nM can afford significant neuroprotective effects against Ab-inducedtoxicity. Docking studies show that compounds 3 and 12 bind in the middle of the AChE active site gorge,but are buried deeper inside BuChE active site gorge, as a consequence of larger BuChE gorge void. Allthese data suggest that these new tacrine analogues could be used for the potential treatment of Alz-heimer’s disease.

� 2011 Elsevier Masson SAS. All rights reserved.

1. Introduction

Alzheimer’s disease (AD) is a multifactorial neurodegenerativedisorder with several target proteins contributing to its aetiology[1e3]. Hence, drug discovery in AD is gradually moving from thedevelopment of agents able to modulate the biological function ofa single target to the multi-target-directed-ligands (MTDLs), which

neau on occasion of his 60th

: þ351 21 7946470.

@gmail.com (M.C. Carreiras),

son SAS. All rights reserved.

are, in principle, effective in treating complex diseases because oftheir ability to interact with multiple targets supposed to beresponsible for the pathogenesis [1e11]. This approach normallyinvolves the use of synthetic hybrid compounds, rationallydesigned to incorporate into a single framework at least twodifferent pharmacophoric moieties [12]. The use of multifunctionaldrugs should be advantageous over the classical polypharmacy,since it reduces the risk of drugedrug interactions and simplifiespharmacokinetic and pharmacodynamic studies, manufacture,formulation, clinical trial designs, and also the therapeutic regi-mens, thus resulting in a great patient compliance [2,7,12]. Sinceacetylcholinesterase (AChE) binds through its peripheral anionicsite (PAS) to the amyloid b (Ab) non-amyloidogenic form and acts asa pathological chaperone inducing a conformational transition

Scheme 1. Synthesis of 2-aryl-5,6,7,8-tetrahydrofuro[2,3-b]quinolin-4-amines (1, 3, 5),2-aryl-6,7,8,9-tetrahydro-5H-cyclohepta[b]furo[3,2-e]pyridin-4-amines (2, 4, 6), 4-amino-2-(p-tolyl)-5,6,7,8-tetrahydrofuro[2,3-b]quinoline-3-carbonitrile (7), 4-amino-2-aryl-5,6,7,8-tetrahydrofuro[2,3-b]quinoline-3-carbonitriles (8, 10), and 4-amino-2-aryl-6,7,8,9-tetrahydro-5H-cyclohepta[b]furo[3,2-e]pyridine-3-carbonitriles (9, 11).

Scheme 2. Synthesis of 2-aryl-1-phenyl-5,6,7,8-tetrahydro-1H-pyrrolo[2,3-b]quinolin-4-amines (12, 14, 16), and 2-aryl-1-phenyl-1,5,6,7,8,9-hexahydrocyclohepta[b]pyrrolo[3,2-e]pyridin-4-amines (13, 15).

C. Martins et al. / European Journal of Medicinal Chemistry 46 (2011) 6119e61306120

to the amyloidogenic form [13,14] most of the MTDLs are dual-binding AChE inhibitors endowed with additional properties.Thus, they may simultaneously alleviate cognitive deficits andbehave as disease-modifying agents by inhibiting Ab peptideaggregation through binding to both catalytic and peripheral sitesof the enzyme [12].

Besides the pathologic hallmarks of the disease, i.e., neurofi-brillary tangles and neuritic plaques, AD is characterized chemicallyby a consistent deficit in cholinergic transmission, particularlyaffecting cholinergic neurons in the basal forebrain. Therefore, thefirst clinical relevant approach has been the use of AChE inhibitors

Chart 1. General structure and heterocyclic ring systems for the new tacrine analo

(AChEI) [15]. Due to its pharmacological nature, cholinesteraseinhibitory therapy may be considered as a simple symptomaticshort-term intervention. However, data emerging from long-termshows that the effects of AChEI are not necessarily confined toAChE inhibition; they additionally improve the cognitive symptomsof AD through regulation of the processing and secretion of APP[16e27]. Stabilization of the cognitive status of patients treatedwith AChEI may be related to the interactions of these drugs withthe amyloid cascade.

Due to the overwhelming evidence regarding the neuro-protective effects of AChEI, most AD clinicians and researchersunanimously agree that there is no single treatment or therapeuticapproach that will alleviate this condition, and that a multiple-functional approach or multipotent drugs will be required [1e12].Moreover, it is highly conceivable that MTDLs may represent thefuture treatment for AD and other similarly complex multifactorialdiseases [28].

Some years ago, we embarked on a long-term research projectaimed at the synthesis of a series of multipotent compoundsdesigned to target both AChE and neuronal Ca2þ modulation,showing neuroprotection and antioxidant properties. As a result,we have synthesized and evaluated a number of hybrid compounds[29], where the benzene ring in tacrine has been replaced bya heterocyclic ring. Other tacrine analogues have also been reportedfrom other laboratories [30e34]. Kirsch and co-workers syn-thesised a few analogues where the benzene ring in tacrine wasreplaced by a thiazole ring [30], a 4-azaisoindole ring [31], a sele-nophene ring [32], and a thiophene ring [33]. In all these analoguesthe cyclohexane-fused ring was maintained, contracted or enlarged[30e33]. Moreover, in the 4-azaisoindole derivatives [31] it wasreplaced by functionalized cyclohexane rings, a nitrogen-containingsix-membered heterocyclic ring, a pyran and a thiopyran ring.Tzeng and collaborators [34] reported the synthesis of certainalkoxy, aryloxy, alkylamino or arylamino substituted benzofuro[2,3-b]quinoline derivatives, all of them are tetracyclic analogues oftacrine.

With these precedents in mind, and with the purpose ofdeveloping AChE and butyrylcholinesterase (BuChE) inhibitorsendowed with neuroprotective properties, we report here thesynthesis and pharmacological activities of fused five-memberedaromatic ring tacrine analogues, such as the furanotacrines 1e11(Scheme 1) and the pyrrolotacrines 12e16 (Scheme 2), bearing thefuro[2,3-b]quinolin-4-amine (I) or the pyrrolo[2,3-b]quinolin-4-amine (II) (Chart 1) heterocyclic ring systems.

2. Chemistry

2.1. Furo[2,3-b]quinolin-4-amine derivatives

The synthesis of the new furanotacrines 1e11 has been ach-ieved, as shown in Scheme 1, starting from the readily available,previously synthesized in our laboratory, 2-amino-5-arylfuran-3-carbonitriles 17e19 [34] or 2-amino-5-arylfuran-3,4-dicarboni-triles 20e22 [34], according to typical Friedländer reaction (FR)

gues: furo[2,3-b]quionlin-4-amine (I) and pyrrolo[2,3-b]quinolin-4-amine (II).

Table 1Inhibition of EeAChE/hrAChE and eqBuChE/hBuChE by 2-aryl-3a,5,6,7,8,9a-hexahydrofuro[2,3-b]quinolin-4-amine (1e6), 4-amino-2-aryl-5,6,7,8-tetrahydrofuro[2,3-b]quin-oline-3-carbonitrile (10), and 2-aryl-1-phenyl-5,6,7,8-tetrahydro-1H-pyrrolo[2,3-b]quinolin-4-amine (12e14) derivatives.a

Compound Structure IC50 IC50 Selectivity IC50 IC50

EeAChE (mM) eqBuChe (mM) BuChE/AChE hrAChE (mM) hBuChE (mM)

1 >100 3.6� 0.7 >100 3.78� 0.11

2 >100 6� 1

12 0.61� 0.04 0.074� 0.009 0.12 45.5� 3.1 8.57� 0.40

13b >30 1.1� 0.2

3 (Ar: 40-MeC6H4) > 100 2.9� 0.4 >100 119� 15

4b (Ar: 40-MeC6H4) > 30 13� 1

5 (Ar: 40-ClC6H4) >100 23� 2 >100 26.1� 1.0

6b (Ar: 40-ClC6H4) >100 43� 3

10b (Ar: 40-ClC6H4) >100 32.4� 0.6 >100 >100

14b (Ar: 40-ClC6H4) 24� 2 15� 2 0.62

Tacrine 0.027� 0.002 0.0052� 0.0002 0.193

a IC50 values are the mean� SEM of quadruplicates of at least three independent experiments.b Measured in presence of bovine serum albumin to improve solubility.

C. Martins et al. / European Journal of Medicinal Chemistry 46 (2011) 6119e6130 6121

-4 -2 0 2 4 6 8 10

0

10

20

30

40

50

60

Control

100nM

300nM

600nM

1uM

2uM

V-1

(m

in

Δ AU

-1

)

[AThCh]-1

(mM-1

)

Fig. 1. Steady-state inhibition of AChE hydrolysis of acetylthiocholine (ATCh) bycompound 12. LineweavereBurk reciprocal plots of initial velocity and substrateconcentrations (0e0.8 mM) are presented. Lines were derived from a weighted least-squares analysis of data.

0

20

40

60

80

100

*

Compound 12

303Basal

***

303

Tacrine

[ ] μM

###

Ce

ll V

iab

ilit

y (

%)

0

20

40

60

80

100

###

*****

**

Aβ25-35

10μM

[Compound 12] μM0.3 1 3Basal

Ce

ll V

iab

ilit

y (

%)

A

B

Fig. 2. (A) Cell viability, measured as reduction of MTT, in SH-SY5Y cells exposed for48 h with 3 or 30 mM of compound 12 or tacrine alone. Data correspond to the meanand S.E.M. of 4 different experiments performed in triplicate. *p< 0.05, ***p< 0.001 incomparison to basal condition in the absence of drug and ###p< 0.001 comparing30 mM compound 12 with 30 mM tacrine. (B) Protective effect of increasing concen-trations of compound 12 in cells exposed to the cytotoxic fragment of Ab25e35.Compound 12 was incubated with the cells 24 h before and during the 24 h exposureto 10 mM Ab25e35. Data correspond to the mean and S.E.M. 4 Different experimentsperformed in triplicate. ###p< 0.001, comparing with basal conditions. *p< 0.05,**p< 0.01 and ***p< 0.001 comparing to Ab25e35 e induced toxicity in the absence ofdrug.

C. Martins et al. / European Journal of Medicinal Chemistry 46 (2011) 6119e61306122

protocols [35]. Under the usual conditions [29], using cyclohex-anone or cycloheptanone, we isolated the expected targetmolecules in good yields. The structures of these compounds arein very good agreement with their analytical and spectroscopicdata (see Experimental part).

2.2. Pyrrolo[2,3-b]quinolin-4-amine derivatives

The synthesis of the new pyrrolotacrines 12e16 has been ach-ieved, as shown in Scheme 2, starting from the readily available,previously synthesized in our laboratory 2-amino-5-aryl-1-phenyl-1H-pyrrole-3-carbonitriles 23e25 [35], according to typical FRprotocols [36]. Under the usual conditions [29], using cyclohexa-none or cycloheptanone, the expected target molecules were iso-lated in good yields. The structure of these compounds has beenestablished and assigned by their analytical and spectroscopic data(see Experimental part).

3. Results and discussion

3.1. Pharmacology

3.1.1. AChE/BuChE inhibitory activity of furo[2,3-b]quinolin-4-amine and pyrrolo[2,3-b]quinolin-4-amine derivatives

Furanotacrines 1e11 and pyrrolotacrines 12e16 have been eval-uated as inhibitors of AChE from Electrophorus electricus (EeAChE)and BuChE from horse serum (eqBuChE) according to the methodbased on Ellman [37]. Next, some selected inhibitors (1, 3, 5 and 10)were also evaluated as inhibitors of human recombinant AChE(hrAChE), and human serum BuChE (hBuChE), according to themethod based on Ellman [37].

Unfortunately, compounds 7 and 16 were not soluble, even intheir hydrochloride form. Molecules 8, 9, 11 and 15 were soluble inDMSO, but proved to be inactive on both cholinesterases. The activemolecules gave the IC50 values shown in Table 1. From these data,and for the furanotacrines we conclude that: (a) these compoundsare from potent to moderate, in the micromolar range, selectiveinhibitors of eqBuChE regarding to EeAChE; (b) the most potentinhibitor was 2-(p-tolyl)-5,6,7,8-tetrahydrofuro[2,3-b]quinolin-4-amine (3) [IC50 (eqBuChE)¼ 2.9� 0.4 mM]; (c) for compoundsbearing the same substituent in the aryl ring at C2, the most potentinhibitors were those bearing a fused cyclohexane ring, instead ofa fused cycloheptane ring [compare 1 with 2 (Ar¼ C6H5); 3 with4 (Ar¼ 40-MeC6H4); 5 with 6 (Ar¼ 40-ClC6H4)]; (d) for thecompounds bearing the same cycloalkane fused ring system, andregarding the type of the substituent at C40 in the aromatic ring, theinhibitory potency followed the order: H� 40-Me[ Cl [compare 1with 3 and 5 (cyclohexane)]; or H[ 40-Me[ Cl [compare 2 with4 and 6 (cycloheptane)]; (e) the incorporation of a nitrile (CN)group at C3 resulted in a total loss of BuChE inhibitory activity(compare compound 6 with 11, or see molecules 8, and 9), in mostof the cases, only CN-substituted compound 10 afforded amoderateeqBuChE inhibitory activity (IC50¼ 32.4 mM); (f) on going fromeqBuChE to hBuChE, the inhibitory potency remained similar(compounds 1 and 5) or was significantly decreased (compound 3),the most potent being 2-phenyl-5,6,7,8-tetrahydrofuro[2,3-b]qui-nolin-4-amine (1) [IC50 (hBuChE)¼ 3.78� 0.11 mM], or was totallylost as in compound 10.

From the data in Table 1, and for the pyrrolotacrines, the samestructureeactivity relationships (SARs) could also be determined,regarding the effect of the size of the fused cycloalkane ring, andthe type of the substituent at the C40 in the aromatic substituent.The most significant difference with the furanotacrines concerningthe inhibition of AChE/BuChE was that in this group of molecules(12e16), two of them, 12 and 14, also inhibited EeAChE. Thus,

inhibitor 14 shows an equipotent profile on both cholinesterases,whereas compound 12 is much more potent for the inhibition ofeqBuChE. Very interestingly, compared with 12, pyrrolotacrine 13,differing just by a cycloheptane instead of cyclohexane ring, is

Fig. 3. Binding mode of compound 12 at the EeAChE gorge predicted by the docking simulation. (a) The compound is represented as stick model and illustrated in green. Relevantresidues are shown as sticks and coloured by atoms. Hydrogen bond is represented as pink dashed lines. (b) Molecular docking model of 12eEeAChE complex: 12 (green) (Forinterpretation of the references to color in this figure legend, the reader is referred to the web version of this article.).

C. Martins et al. / European Journal of Medicinal Chemistry 46 (2011) 6119e6130 6123

almost inactive. Pyrrolotacrine 12 was also assayed for the inhibi-tion of hrAChE and hBuChE (Table 1). The results show that thiscompound inhibited both enzymes, but 5-fold more selectively forhBuChE.

Overall, 1,2-diphenyl-5,6,7,8-tetrahydro-1H-pyrrolo[2,3-b]qui-nolin-4-amine (12) was the most potent for the inhibition of bothenzymes [IC50 (EeAChE)¼ 0.61�0.04 mM; IC50 (eqBuChE)¼0.074� 0.009 mM].

3.1.2. Kinetic analysis of AChE inhibition by compound 12To gain further insight into the mechanism of action on AChE of

this family of compounds, a kinetic study was carried out with themost promising compound of the series, 12, using EeAChE. Graph-ical analysis of the reciprocal LineweavereBurk plots (Fig. 1)showed both increased slopes (decreased Vmax) and intercepts(higher Km) at increasing concentration of the inhibitor. Thispattern indicates a mixed-type inhibition. Replots of the slopeversus concentration of compound 12 gave an estimate of theinhibition constant, Ki, of 621 nM.

3.1.3. Measurement of cell viability with MTTThe effects of compound 12 on cell viability were analysed, first

at the concentration of 3 mM and after, at a concentration 10 timeshigher, 30 mM, since this drug is a tacrine analogue. For compar-ative purposes we also used tacrine in the experiments. Cellviability measured as MTT reduction showed that exposure of SH-

Fig. 4. Binding mode of 3 (violet sticks) at the EeAChE gorge. (a) Relevant residues are show(b) Molecular docking model of 3eEeAChE complex (For interpretation of the references to

SY5Y cells during 48 hwith 3 mMof compound 12 or tacrine did notcause reduction of viability. However, at 30 mM, compound 12reduced viability by 28% and tacrine by 50%; this cytotoxic effectwas significantly higher for tacrine in comparison to compound 12(p< 0.001) (Fig. 2A).

3.1.4. Neuroprotection against Ab-induced cytotoxicityIn another set of experiments we evaluated the potential

protective effect of compound 12 on Ab-induced toxicity. Exposureof SH-SY5Y cells for 24 h to 10 mM of Ab25e35 reduced cell viabilityto 36.6� 4.5%. Cells that were pre-treated 24 h before withcompound 12 and during the 24 h exposure to the Ab25e35 toxinshowed lower reductions of cell viability, i.e. 11.4� 2.2%, 19� 5.7%and 23.7�4.3% at 0.3, 1 and 3 mM, respectively. Therefore, theseresults indicate that compound 12, at concentrations as low as300 nM, can afford significant neuroprotective effects against Ab-induced toxicity (Fig. 2B).

3.1.5. Molecular docking studiesMolecular docking studies with pyrrolotacrine 12 and furanota-

crine 3 were carried out in the active site of EeAChE and eqBuChE.The 3D structures of both enzymes were employed since ligandsbind to enzyme species differently despite the high homologysequence among species.

The technique blind docking [38,39] was adopted since it mightdetect possible binding sites and ligand conformations by

n as sticks and coloured by atoms. Hydrogen bond is represented as pink dashed lines.color in this figure legend, the reader is referred to the web version of this article.).

Fig. 5. Complex of compounds 12 (green) and 3 (violet) and eqBuChE homology build3D-model (For interpretation of the references to color in this figure legend, the readeris referred to the web version of this article.).

C. Martins et al. / European Journal of Medicinal Chemistry 46 (2011) 6119e61306124

scanning the entire surface of a protein target so that a locationwith the highest binding affinity on the protein may be found. Theprogram AutoDock Vina [40] was carried out to perform the blinddocking using a single catalytic sub-unit of EeAChE (PDB: 1C2B).This program was shown to be more accurate compared toprevious versions of AutoDock employed in blind docking studies[38,39].

Comparative crystallographic analysis of the PAS and thecatalytic ligandeAChE complexes revealed distinct features of thecrystalline PAS and the active site gorge, which may reflectconformational states of AChE associated with PAS and catalyticligands in solution. Although the overall structures are virtuallyidentical, superimposition revealed significant mobility for theside chain residues of Trp286, located at the entry of the gorge(PAS), Tyr124 and Tyr337 in the constricted gorge region and Tyr72in the gorge path. These residues delineate the shape of the gorgeentry and its lining. In particular, the phenol moieties of Tyr124and Tyr337, located midway, swing in and out with large devia-tions of its hydroxyl group. This motion is sufficient by itself to

Fig. 6. Binding modes of pyrrolotacrine 12 and furanotacrine 3 on eqBuChE homology buiposition of 12 (green stick) and 3 (violet stick) with choline (red stick), butyrate (blue stick)in this figure legend, the reader is referred to the web version of this article.).

enlarge the gorge and enhance access for the ligand to the activecentre.

To account for side chain flexibility during docking, flexibletorsions in the ligandwere assigned, and the acyclic dihedral angleswere allowed to rotate freely. In addition, receptor flexibility wasintroduced by keeping flexible Trp286, Tyr124, Tyr337 and Tyr72receptor residues mentioned earlier under the docking simulation,in which the pose with the lowest docking energy was selected asthe best solution.

Fig. 3 illustrates the most energetic favourable binding mode of12 at the active site of EeAChE. In the complex, TYR341 was foundto allow a pep stacking interaction with the pyrrolotacrinemoietyof the ligand. The aromatic nitrogen at the tetrahydroquinolinering is hydrogen bonded to the hydroxyl group of TYR124 and thephenyl group at one-position is involved in hydrophobic contactswith the central hydrophobic region (TRP86 and ASP74). In thisorientation, the phenyl ring at the two-position is buried insidethe protein gorge and it is stacked between TYR337 and TRP86, aswell as it is involved in hydrophobic contacts with the amino acidresidue of the catalytic triad, HIS447. Besides, the cyclohexanering is involved in hydrophobic contacts with TYR341 andTRP286, the latter residue located in the PAS. These resultssuggest that this compound could favourably interact with thecatalytic site and the PAS of AChE, therefore, docking results are inagreement with the experimental kinetic data, indicatinga mixed-type inhibition.

However, a similar behaviour was not observed for furanota-crine 3, the best pose accommodates the ligand within thecentral region of the active site gorge, although no interactionswith the catalytic triad and the PAS residues were found (Fig. 4).Conversely to pyrrolotacrine 12, the cyclohexane unit is buriedinside the protein gorge and does not make favourable contacts.The aromatic nitrogen at the tetrahydroquinoline ring is hydrogenbonded to the hydroxyl group of TYR337 and the p-methylphenylring is involved in hydrophobic contacts with TYR341 and PHE338.

A close inspection of the 12eAChE and 3eAChE complexessuggests that critical rearrangements of the ligand are thoseinduced by the phenyl ring at the one-position. In summary, thegeometric and energetic features of the 12eenzyme complexcontribute to the higher affinity of pyrrolotacrine 12 and make ita more potent inhibitor than 3.

With respect to BuChE, and in the absence of X-ray structure ofeqBuChE, a homology model was used to rationalize experimentaldata. The automated homology-modelling SWISS-MODEL programperformed the modelling of the 3-D structure [41e43]. A putativethree-dimensional structure of eqBuChE was created based on the

ld 3D-model. (a) Top view of the accessible surface of the active site gorge. (b) Super-and the best pose for tacrine (yellow stick) (For interpretation of the references to color

C. Martins et al. / European Journal of Medicinal Chemistry 46 (2011) 6119e6130 6125

crystal structure of hBuChE (pdb: 2pm8). These two enzymesexhibited 89% sequence identity.

In order to simulate the binding of compounds 3 and 12 toeqBuChE, the reported position of butanoic acid (PDB ID: 1P0I) andcholine (PDB ID: 1P0M) in hBuChE were used to mimic the inter-mediate product-bound form during the docking studies due to thelack of native BuChE-substrate structure. The non-availability ofco-crystallized structures for eqBuChE with tacrine prompted us toapply the same molecular docking protocol that was adopted forcompounds 3 and 12, in order to explore its proteineligandinteractions and compare with the positions of butanoic acid andcholine. All docking experiments were performed as blind dock-ings following the same computational protocol adopted inEeAChE.

The best-ranked docking solutions revealed that BuChE couldeffectively accommodate pyrrolotacrine 12 and furanotacrine 3inside the active site gorge (Fig. 5). The 12-BuChE complex isstabilized mainly by the hydrophobic interactions with theamino acid residues of the catalytic active site region of theenzyme.

Fig. 6 shows the position of 12 and tacrine at the active site ofeqBuChE, and choline and butyrate at the active site of hBuChE. Thephenyl ring at one-position is in the region where tacrine andcholine are bound and the cyclohexane moiety occupies the sameregion that butanoic acid does (Fig. 5). In the 3eBuChE complex theinappropriate position of the ligand compared to the positionoccupied by tacrine, choline and compound 12 accounts fora higher affinity of 12 towards eqBuChE andmakes it a more potentinhibitor than compound 3.

The key to increasing the inhibitory potency against AChE andBuChE seems to be the phenyl ring at one-position of pyrrolotacrine12 rather than changing the heteroatom of the five-memberedrings.

4. Conclusions

The synthesis andbiological evaluationof furanotacrines1e11, andpyrrolotacrines 12e14 were reported. These tacrine analogues showa selective affinity in the micromolar range towards BuChE. In thefurotacrine group, the most attractive inhibitor was 2-(p-tolyl)-5,6,7,8-tetrahydrofuro[2,3-b]quinolin-4-amine (3) [IC50 (eqBuChE)¼2.9� 0.4 mM; IC50 (hBuChE)¼ 119�15 mM].However, pyrrolotacrines12 and 14 proved to be moderately equipotent for both cholinester-ases. Pyrrolotacrine 1,2-diphenyl-5,6,7,8-tetrahydro-1H-pyrrolo[2,3-b]quinolin-4-amine (12) was the most potent [IC50 (EeAChE)¼0.61�0.04 mM; IC50 (eqBuChE)¼ 0.074� 0.009 mM] though itdoesn’t reach affinities to BuChE comparable to tacrine or physo-stigmine [46]. However, as a BuChEI it shows higher activity thanphenserine [47] and rivastigmine [3]. Cell viability measured as MTTreduction showed that exposure of SH-SY5Y cells during 48 h with3 mM of compound 12 or tacrine did not cause reduction of viability.Nevertheless, at 30 mM, compound 12 reduced cell viability by 28%while tacrine diminished by 50%. Thus, this cytotoxic effect wassignificantly higher for tacrine in comparison to compound 12. Pyr-rolotacrine 12 also showed that at concentrations as low as 300 nM,can afford significant neuroprotective effects against Ab-inducedtoxicity. The molecular modelling studies provided valuable infor-mation on how this new molecular scaffold represented by thecompounds containing both pyridine and pyrrole or furan ringsinteracts with AChE and BuChE. The presence of a phenyl ring at theone-positionof the pyrrole unit is beneficial for bothAChE andBuChEinhibition. These results suggest pyrrolotacrine 12 may representa new lead capable of structural improvement for potential treatmentof AD.

5. Experimental part

5.1. Chemistry

Melting points were determined on a Koffler apparatus and areuncorrected. 1H NMR and 13C NMR spectra were recorded in CDCl3or DMSO-d6 at 300, 400 or 500 MHz, and at 75.4, 100.5 or125.7 MHz, respectively, using solvent peaks [CDCl3: 7.27 (D), 77.2(C) ppm; D2O: 4.60 ppm and DMSO-d6: 2.49 (D), 40 (C) ppm] asinternal reference. The assignment of chemical shifts is based onstandard NMR experiments (1H, 13C-DEPT, 1H, 1H-COSY, gHSQC,gHMBC). NMR signals with (*) can be interchanged. Mass spectrawere recorded on a GC/MS spectrometer with an API-ES ionizationsource. Elemental analyses were performed on a LECO CHNS-932apparatus. TLC was performed on silica F254 and detection by UVlight at 254 nm or by charring with either ninhydrin, anisaldehyde,or phosphomolybdic-H2SO4 dyeing reagents. Anhydrous solventswere used in all experiments. Column chromatography was per-formed on silica gel 60 (230mesh). Reactions under MW irradiation(250W) were performed in a CEM Discover system�, equippedwith electromagnetic sample stirrer, an infrared temperaturedetector and a pressure sensor. The reaction was performed in30 mL glass tube equipped with septa.

5.2. General method for the Friedländer reaction

To a suspension of AlCl3 (2.2 equiv) in 1,2-dichloroethane (8 mLsolvent/mmol of aminonitrile), was added cyclohexanone(2.2 equiv) at rt under nitrogen atmosphere. After ca. 5 min, thecorresponding furo or pyrrolo aminonitrile (1 equiv) was addedand the reaction mixture was refluxed (1e8 d). After the reactionwas complete, a NaOH or NaHCO3 10% aqueous solution was addeddropwise until pH 9. Next, the solvents were evaporated to dryness.The solid residue was purified by column chromatography, usinghexaneeethyl acetate or hexaneedichloromethane mixtures aseluent, to afford the corresponding furan or pyrrolotacrines.

5.2.1. 2-Phenyl-5,6,7,8-tetrahydrofuro[2,3-b]quinolin-4-amine (1)Following the general method for the Friedländer reaction, to

a solution of AlCl3 (599 mg, 4.5 mmol, 1.7 equiv) in 1,2-dichloroethane (30 mL) was added 2-amino-5-phenylfuran-3-carbonitrile (17) [35] (503 mg, 2.73 mmol) under argon atmo-sphere. After a few minutes, cyclohexanone (425 mL, 4.1 mmol,1.5 equiv) was added and the reaction mixture was refluxed for72 h, yielding furanotacrine 1 (618 mg, 86%) as light brown crystals:mp 207e210 �C; IR (KBr) n 3470, 3326, 3215, 3078, 2935, 2909,2860, 1633, 1601, 1469, 1336, 1107 cm�1; 1H NMR (CDCl3, 400 MHz)d 7.81 (d, J¼ 7.2 Hz, 2H, H20), 7.41 (t, J¼ 7.3 Hz, 2H, H30), 7.30 (t,J¼ 7.3 Hz,1H, H40), 6.84 (s,1H, H3), 4.36 (br s, 2H, NH2), 2.93 (m, 2H,H8), 2.51 (m, 2H, H5), 1.89 (m, 4H, H6, H7); 13C NMR (CDCl3,100.5 MHz) d 161.1 (C9a), 152.8 (C8a), 151.6 (C2), 145.1 (C4), 130.3(C10), 128.7 (C30), 128.1 (C40), 124.4 (C20), 110.5 (C4a), 105.5 (C3a),96.6 (C3), 33.1 (C8), 23.1 (C5), 22.8 (C7)*, 22.7 (C6)*; MS (EIþ) m/z:264 [Mþ] (100), 263 (36), 236 (27). Anal. Calcd. for C17H16N2O: C,77.25; H, 6.10; N, 10.60. Found: C, 77.05; H, 6.15, N, 10.46.

5.2.2. 2-Phenyl-6,7,8,9-tetrahydro-5H-cyclohepta[b]furo[3,2-e]pyridin-4-amine (2)

Following the general method for the Friedländer reaction, toa solution of AlCl3 (694 mg, 5.21 mmol, 1.9 equiv) in 1,2-dichloroethane (30 mL) was added 2-amino-5-phenylfuran-3-carbonitrile (17) [35] (500 mg, 2.71 mmol) under argon atmo-sphere. After a few minutes, cycloheptanone (577 mL, 4.89 mmol,1.8 equiv) was added and the reaction mixture was refluxed for24 h, yielding furanotacrine 2 (659 mg, 87%): mp 216e218 �C; IR

C. Martins et al. / European Journal of Medicinal Chemistry 46 (2011) 6119e61306126

(KBr) n 3433, 3323, 3250, 3103, 3067, 3032, 2949, 2912, 2846, 1642,1599, 1565, 1471, 1340, 1267, 1186, 1114 cm�1; 1H NMR (DMSO-d6,400 MHz) d 7.72 (d, J¼ 7.2 Hz, 2H, H20), 7.47 (t, J¼ 7.3 Hz, 2H, H30),7.38 (s, 1H, H3), 7.34 (t, J¼ 7.3 Hz, 1H, H40), 6.31 (br s, 2H, NH2), 2.86(m, 2H, H9), 2.69 (m, 2H, H5), 1.78 (m, 2H, H7), 1.58 (m, 2H, H8), 1.51(m, 2H, H6); 13C NMR (DMSO-d6, 100.4 MHz) d 160.2 (C10a), 159.1(C9a),149.3 (C2),146.5 (C4),130.2 (C10),129.1 (C30),128.0 (C40),123.6(C20),114.9 (C4a),105.2 (C3a), 99.8 (C3), 38.5 (C9), 31.9 (C7), 27.4 (C6),26.5 (C8), 24.6 (C5); MS (EIþ) m/z (%): 278 [M]þ (100), 277 (33), 263(17), 250 (23), 249 (39), 237 (12), 224 (12). Anal. Calcd. forC18H18N2O:C, 77.67; H, 6.52; N, 10.06. Found: C, 77.54; H, 6.35; N, 9.92.

5.2.3. 2-(p-Tolyl)-5,6,7,8-tetrahydrofuro[2,3-b]quinolin-4-amine(3)

Following the general method for the Friedländer reaction, toa solution of AlCl3 (277 mg, 2.08 mmol, 1.7 equiv) in 1,2-dichloroethane (33 mL) was added 2-amino-5-(p-tolyl)furan-3-carbonitrile (18) [35] (202 mg, 1.02 mmol) under argon atmo-sphere. After a few minutes, cyclohexanone (218 mL, 2.10 mmol,1.8 equiv) was added and the reaction mixture was refluxed for24 h, yielding furanotacrine 3 (220 mg, 78%): mp 256e258 �C; IR(KBr) n 3481, 3333, 3221, 3068, 3020, 2943, 2861, 1632, 1599, 1503,1468, 1445, 1336, 1112 cm�1; 1H NMR (DMSO-d6, 300 MHz) d 7.63(d, J¼ 6.0 Hz, 2H, H20), 7.29 (d, J¼ 6.0 Hz, 2H, H30), 7.27 (s, 1H, H3),6.01 (br s, 2H, NH2), 2.74 (m, 2H, H8), 2.50 (m, 2H, H5), 2.35 (s, 3H,CH3), 1.79 (m, 4H, H6, H7); 13C NMR (DMSO-d6, 75.4 MHz) d 161.7(C9a), 152.6 (C8a), 150.4 (C2), 147.7 (C4), 138.3 (C40), 130.4 (C30),128.5 (C10), 124.6 (C20), 110.5 (C4a), 105.5 (C3a), 99.4 (C3), 33.7 (C8),23.8 (C5), 23.5 (C7)*, 23.3 (C6)*, 21.6 (CH3); MS (EIþ) m/z (%): 278[M]þ$ (100), 277 (32), 250 (30). Anal. Calcd. for C18H18N2O: C, 77.67;H, 6.52; N, 10.06. Found: C, 77.54; H, 6.47; N, 9.97.

5.2.4. 2-(p-Tolyl)-6,7,8,9-tetrahydro-5H-cyclohepta[b]furo[3,2-e]pyridin-4-amine (4)

Following the general method for the Friedländer reaction, toa solution of AlCl3 (271 mg, 2.04 mmol, 2.0 equiv) in 1,2-dichloroethane (36 mL) was added 2-amino-5-(p-tolyl)furan-3-carbonitrile (18) [35] (200 mg, 1.01 mmol) under argon atmo-sphere. After a few minutes, cycloheptanone (240 mL, 2.04 mmol,2.0 equiv) was added and the reaction mixture was refluxed for24 h, yielding furanotacrine 4 (238 mg, 81%): mp 272e274 �C; IR(KBr) n 3460, 3337, 3228, 3022, 2953, 2923, 2847, 1638, 1601, 1503,1471, 1338 cm�1; 1H NMR (DMSO-d6, 400 MHz) d 7.57 (d, J¼ 8.0 Hz,2H, H20), 7.26 (s, 1H, H3), 7.25 (d, J¼ 8.0 Hz, 2H, H30), 6.23 (br s, 2H,NH2), 2.81 (m, 2H, H9), 2.65 (m, 2H, H5), 2.29 (s, 3H, CH3), 1.75 (m,2H, H7), 1.54 (m, 2H, H8), 1.47 (m, 2H, H6); 13C NMR (DMSO-d6,100.5 MHz) d 160.1 (C10a), 158.8 (C9a), 149.5 (C2), 146.3 (C4), 137.4(C40), 129.6 (C30), 127.5 (C10), 123.6 (C20), 114.8 (C4a), 105.2 (C3a),98.9 (C3), 38.4 (C9), 31.8 (C7), 27.3 (C6), 26.5 (C8), 24.6 (C5), 20.8(CH3); MS (EIþ) m/z (%): 292 [M]þ� (100), 291 (22), 264 (10), 263(22). Anal. Calcd. for C19H20N2O: C, 78.05; H, 6.89; N, 9.58. Found: C,77.78; H, 7.10; N, 9.58. The hydrochloride of furanotacrine 4 wasobtained by treating it with hydrogen chloride gas generated by thereaction of sulphuric acid with sodium chloride crystals: Anal.Calcd. for C18H19ClN2O$HCl: C, 68.67, H, 6.08; N, 8.90; Cl, 11.26.Found: 68.70; H, 6.20; N, 8.45; Cl, 10.84.

5.2.5. 2-(4-Chlorophenyl)-5,6,7,8-tetrahydrofuro[2,3-b]quinolin-4-amine (5)

Following the general procedure for the Friedländer reaction, toa solution of AlCl3 (693 mg, 5.20 mmol, 2.1 equiv) in 1,2-dichloroethane (35 mL) was added 2-amino-5-(40-chlorophenyl)furan-3-carbonitrile (19) [35] (537 mg, 2.46 mmol) under argonatmosphere. After a fewminutes, cyclohexanone (510 mL, 4.92 mmol,2.0 equiv) was added and the reaction mixture was refluxed for

48 h, yielding furanotacrine 5 (510 mg, 69%) as dark yellow crystals:mp 263e265 �C; IR (KBr) n 3451, 3328, 3221, 3091, 2932, 2855,1633,1601, 1555, 1486, 1466, 1339, 1088 cm�1; 1H NMR (DMSO-d6,400 MHz) d 7.72 (d, J¼ 8.4 Hz, 2H, H20), 7.53 (d, J¼ 8.4 Hz, 2H, H30),7.39 (s, 1H, H3), 6.34 (br s, 2H, NH2), 2.72 (m, 2H, H8), 2.43 (m, 2H,H5), 1.75 (m, 4H, H6, H7); 13C NMR (DMSO-d6, 100.5 MHz) d 160.9(C9a), 152.2 (C8a), 147.9 (C2)*, 147.3 (C4)*, 132.3 (C40), 129.2 (C30),129.1 (C10), 125.3 (C20), 109.6 (C4a), 104.3 (C3a), 100.3 (C3), 32.9 (C8),23.1 (C5), 22.7 (C6)**, 22.5 (C7)**; MS (EIþ)m/z (%): 300 [Mþ 2]þ (57),298 [M]þ (100), 283 (10), 270 (37). Anal. Calcd. for C17H15ClN2O: C,68.34; H, 5.06; N, 9.38. Found: C, 68.21; H, 5.13; N, 9.02.

5.2.6. 2-(4-Chlorophenyl)-6,7,8,9-tetrahydro-5H-cyclohepta[b]furo[3,2-e]pyridin-4-amine (6)

Following the general procedure for the Friedländer reaction, toa solution of AlCl3 (635 mg, 4.76 mmol, 2.1 equiv) in 1,2-dichloroethane (30 mL) was added 2-amino-5-(40-chlorophenyl)furan-3-carbonitrile (19) [35] (504 mg, 2.31 mmol) under argonatmosphere. After a few minutes, cycloheptanone (550 mL,4.66 mmol, 2.0 equiv) was added and the reaction mixture wasrefluxed for 48 h, yielding furanotacrine 6 (563 mg, 78%): mp271e273 �C; IR (KBr) n 3469, 3339, 3250, 3221, 2949, 2921, 2843,1637, 1594, 1576, 1482, 1471, 1335, 1091 cm�1; 1H NMR (DMSO-d6,400 MHz) d 7.71 (d, J¼ 8.4 Hz, 2H, H20), 7.53 (d, J¼ 8.4 Hz, 2H, H30),7.40 (s, 1H, H3), 6.35 (br s, 2H, NH2), 2.85 (m, 2H, H9), 2.69 (m, 2H,H5), 1.78 (m, 2H, H7), 1.58 (m, 2H, H8), 1.50 (m, 2H, H6); 13C NMR(DMSO-d6, 100.5 MHz) d 160.9 (C10a), 160.0 (C9a), 148.7 (C2), 147.2(C4), 132.9 (C40), 129.8 (C30), 129.7 (C10), 125.9 (C20), 115.6 (C4a),105.8 (C3a), 101.2 (C3), 39.1 (C9), 32.5 (C7), 28.0 (C6), 27.1 (C8), 25.2(C5); MS (EIþ) m/z (%): 314 [Mþ 2]þ (51), 312 [M]þ (100), 311 (26),297 (15), 285 (15), 284 (20), 283 (31). Anal. Calcd. for C18H17ClN2O:C, 69.12; H, 5.48; N, 8.96. Found: C, 69.31; H, 5.68; N, 8.56.

5.2.7. 4-Amino-2-(p-tolyl)-5,6,7,8-tetrahydrofuro[2,3-b]quinoline-3-carbonitrile (7)

Following the general procedure for the Friedländer reaction, toa solution of AlCl3 (714 mg, 5.35 mmol, 4.1 equiv) in 1,2-dichloroethane (35 mL) was added 2-amino-5-(p-tolyl)furan-3,4-dicarbonitrile (20) [35] (290 mg, 1.30 mmol) under argon atmo-sphere. After a few minutes, cyclohexanone (516 mL, 5.27 mmol,4.1 equiv) was added and the reaction mixture was refluxed for48 h, yielding furanotacrine 7 (333 mg, 84%): mp 268e270 �C; IR(KBr) n 3472, 3322, 3203, 3032, 2936, 2855, 2217, 1646, 1599, 1504,1467,1361,1328,1141 cm�1; 1H NMR (DMSO-d6, 500 MHz) d 7.92 (d,J¼ 8.2 Hz, 2H, H20), 7.43 (d, J¼ 8.2 Hz, 2H, H30), 5.77 (br s, 2H, NH2),2.76 (m, 2H, H8), 2.51 (m, 2H, H5), 2.41 (s, 3H, CH3), 1.81 (m, 2H,H7)*, 1.80 (m, 2H, H6)*; 13C NMR (DMSO-d6, 125.7 MHz) d 159.8(C9a), 158.0 (C2), 154.9 (C8a), 147.7 (C4), 142.1 (C40), 130.8 (C30),126.5 (C20),125.6 (C10),116.0 (CN),113.0 (C4a),101.9 (C3a), 84.8 (C3),33.6 (C8), 23.7 (C5), 23.1 (C7)**, 22.9 (C6)**, 21.9 (CH3); MS (EIþ)m/z(%): 303 [M]þ� (100), 302 (34), 288 (10), 275 (32). Anal. Calcd. forC19H17N3O: C, 75.23; H, 5.65; N, 13.85. Found: C, 74.99; H, 5.46; N,13.76. The hydrochloride of furanotacrine 7 was obtained bytreating it with hydrogen chloride gas generated by the reaction ofsulphuric acid with sodium chloride crystals: Anal. Calcd. forC19H18ClN3O$HCl: C, 67.15, H, 5.34; N,12.37; Cl, 10.43. Found: 66.91;H, 5.52; N, 12.55; Cl, 10.84.

5.2.8. 4-Amino-2-(4-methoxyphenyl)-5,6,7,8-tetrahydrofuro[2,3-b]quinoline-3-carbonitrile (8)

Following the general procedure for the Friedländer reaction, toa solution of AlCl3 (458 mg, 3.43 mmol, 4.0 equiv) in 1,2-dichloroethane (38 mL) was added 2-amino-5-(4-methoxyphenyl)furan-3,4-dicarbonitrile (21) [35] (200 mg, 0.85 mmol) under argonatmosphere. After a few minutes, cyclohexanone (356 mL,

C. Martins et al. / European Journal of Medicinal Chemistry 46 (2011) 6119e6130 6127

3.44 mmol, 4.0 equiv) was added and the reaction mixture wasrefluxed for 24 h, yielding furanotacrine 8 (241 mg, 89%): mp267e269 �C; IR (KBr) n 3477, 3338, 3077, 2923, 2851, 2213, 1634,1604, 1504, 1468, 1261, 1176, 1108 cm�1; 1H NMR (DMSO-d6,300 MHz) d 8.04 (d, J¼ 8.9 Hz, 2H, H20), 7.18 (d, J¼ 8.9 Hz, 2H, H30),5.72 (br s, 2H, NH2), 3.91 (s, 3H, OCH3), 2.77 (m, 2H, H8), 2.50 (m, 2H,H5), 1.80 (m, 4H, H6, H7); 13C NMR (DMSO-d6, 75.4 MHz) d 162.4(C40), 154.6 (C8a), 147.5 (C4), 132.2 (C2), 128.4 (C20), 120.8 (C10), 116.6(CN),116.2 (C3a)*,115.9 (C30)*,113.0 (C4a), 83.3 (C3), 56.4 (OCH3), 33.6(C8), 23.7 (C5), 23.1 (C7)**, 22.9 (C6)**; MS (EIþ) m/z (%): 319 [M]þ�

(100), 318 (23), 304 (16), 291 (17). Anal. Calcd. for C19H17N3O2: C,71.46; H, 5.37; N, 13.16. Found: C, 71.58; H, 5.49; N, 13.26.

5.2.9. 4-Amino-2-(4-methoxyphenyl)-6,7,8,9-tetrahydro-5H-cyclohepta[b]furo[3,2-e]pyridine-3-carbonitrile (9)

Following the general procedure for the Friedländer reaction, toa solution of AlCl3 (1.15 g, 8.69 mmol, 4.2 equiv) in 1,2-dichloroethane (50 mL) was added 2-amino-5-(4-methoxyphenyl)furan-3,4-dicarbonitrile (21) [35] (490 mg, 2.05 mmol) under argonatmosphere. After a few minutes, cycloheptanone (1 mL,8.48 mmol, 4.1 equiv) was added and the reaction mixture wasrefluxed for 24 h, yielding furanotacrine 9 (521 mg, 76%): mp249e251 �C; IR (KBr) n 3478, 3339, 3222, 2928, 2848, 2219, 1642,1603, 1507, 1476, 1300, 1262, 1177, 1111 cm�1; 1H NMR (DMSO-d6,500 MHz) d 7.98 (d, J¼ 8.7 Hz, 2H, H20), 7.18 (d, J¼ 8.7 Hz, 2H, H30),5.79 (br s, 2H, NH2), 3.87 (s, 3H, OCH3), 2.91 (m, 2H, H9), 2.75 (m,2H, H5), 1.81 (m, 2H, H7), 1.62 (m, 2H, H8), 1.56 (m, 2H, H6); 13CNMR (DMSO-d6, 125.7 MHz) d 162.4 (C40), 161.6 (C9a), 159.2 (C10a),158.4 (C2), 146.9 (C4), 128.4 (C20), 120.8 (C10), 118.3 (C4a), 116.3(CN), 115.9 (C30), 102.9 (C3a), 83.9 (C3), 56.4 (OCH3), 39.1 (C9), 32.2(C7), 27.7 (C6), 26.9 (C8), 25.5 (C5); MS (EIþ) m/z (%): 333 [M]þ

(100), 332 (18), 318 (12), 305 (10), 304 (19). Anal. Calcd. forC20H19N3O2: C, 72.05; H, 5.74; N, 12.60. Found: C, 72.27; H, 6.01; N,12.58.

5.2.10. 4-Amino-2-(4-chlorophenyl)-5,6,7,8-tetrahydrofuro[2,3-b]quinoline-3-carbonitrile (10)

Following the general procedure for the Friedländer reaction, toa solution of AlCl3 (556 mg, 4.17 mmol, 3.3 equiv) in 1,2-dichloroethane (35 mL) was added 2-amino-5-(4-chlorophenyl)furan-3,4-dicarbonitrile (22) [35] (307 mg, 1.26 mmol) under argonatmosphere. After a few minutes, cyclohexanone (392 mL,3.78 mmol, 3.0 equiv) was added and the reaction mixture wasrefluxed for 5 d, yielding furanotacrine 10 (306 mg, 75%): mp240e242 �C; IR (KBr) n 3344, 2943, 2845, 2219, 1634, 1605, 1463,1102 cm�1; 1H NMR (CDCl3, 300 MHz) d 8.06 (d, J¼ 8.5 Hz, 2H, H20),7.48 (d, J¼ 8.5 Hz, 2H, H30), 4.84 (br s, 2H, NH2), 2.91 (m, 2H, H8),2.51 (m, 2H, H5), 1.92 (m, 4H, H6, H7); 13C NMR (CDCl3, 75.4 MHz)d 159.7 (C9a), 158.9 (C2), 156.9 (C8a), 146.3 (C4), 137.4 (C40), 129.9(C30), 127.6 (C20), 126.6 (C10), 115.9 (CN), 112.5 (C4a), 103.5 (C3a),84.6 (C3), 33.6 (C8), 23.3 (C5), 22.9 (C7)*, 22.8 (C6)*; MS (EIþ) m/z(%): 325 [Mþ 2]þ (37), 324 (35), 323 [M]þ (100), 297 (10), 295 (29),139 (10). Anal. Calcd. for C18H14ClN3O: C, 66.77; H, 4.36; N 12.98.Found. C, 67.08; H, 4.35; N, 12.54.

5.2.11. 4-Amino-2-(4-chlorophenyl)-6,7,8,9-tetrahydro-5H-cyclohepta[b]furo[3,2-e]pyridine-3-carbonitrile (11)

Following the general procedure for the Friedländer reaction, toa solution of AlCl3 (1.14 g, 8.59 mmol, 4.0 equiv) in 1,2-dichloroethane (20 mL) was added 2-amino-5-(4-chlorophenyl)furan-3,4-dicarbonitrile (22) [35] (518 mg, 2.13 mmol) under argonatmosphere. After a few minutes, cycloheptanone (500 mL,4.24 mmol, 2.0 equiv) was added and the reaction mixture wasrefluxed. After 48 h, AlCl3 (579 mg, 4.34 mmol, 2.0 equiv), in dry1,2-dichloroethane (20 mL), cycloheptanone (500 mL, 4.24 mmol,

2.0 equiv) were added and the reaction mixture was refluxed for3 d yielding furanotacrine 11 (521 mg, 72%): mp 265e267 �C; IR(KBr) n 3479, 3398, 3332, 3217, 2919, 2851, 2219, 1644, 1600, 1487,1474, 1407 cm�1; 1H NMR (DMSO-d6, 500 MHz) d 8.02 (d,J¼ 10.0 Hz, 2H, H20), 7.69 (d, J¼ 10.0 Hz, 2H, H30), 5.88 (br s, 2H,NH2), 2.92 (m, 2H, H9), 2.76 (m, 2H, H5), 1.82 (m, 2H, H7), 1.62 (m,2H, H8), 1.56 (m, 2H, H6); 13C NMR (DMSO-d6, 125.7 MHz) d 162.4(C9a), 159.5 (C10a), 156.6 (C2), 147.2 (C4), 136.5 (C40), 130.4 (C30),128.2 (C20), 127.1 (C10), 118.4 (C4a),115.6 (CN),102.9 (C3a), 86.4 (C3),39.2 (C9), 32.2 (C7), 27.7 (C6), 26.9 (C8), 25.3 (C5); MS (EIþ)m/z (%):339 [Mþ 2]þ (34), 338 (30), 337 [M]þ (100), 336 (26), 322 (16), 310(15), 309 (23), 308 (35), 283 (11), 139 (7). Anal. Calcd. forC19H16ClN3O: C, 67.56; H, 4.77; N, 12.44; Cl, 10.50. Found: C, 67.87;H, 5.01; N, 12.28; Cl, 10.88.

5.2.12. 1,2-Diphenyl-5,6,7,8-tetrahydro-1H-pyrrolo[2,3-b]quinolin-4-amine (12)

Following the general procedure for the Friedländer reaction, toa solution of AlCl3 (700 mg, 5.25 mmol, 3.0 equiv) in 1,2-dichloroethane (50 mL) was added 2-amino-1,5-diphenyl-1H-pyrrole-3-carbonitrile (23) [35] (450 mg, 1.74 mmol) under argonatmosphere. After a few minutes, cyclohexanone (540 mL,5.21 mmol, 3.0 equiv) was added and the reaction mixture wasrefluxed for 24 h yielding pyrrolotacrine 12 (405 mg, 69%): mp308e310 �C; IR (KBr) n 3467, 3360, 3061, 2930, 2852, 1633, 1595,1565, 1534, 1499, 1467, 1451, 1440, 1340 cm�1; 1H NMR (DMSO-d6,400 MHz) d 7.37 (t, J¼ 7.8 Hz, 2H, H300), 7.28 (d, J¼ 7.2 Hz, 2H, H20),7.27 (m, 1H, H400), 7.23 (d, J¼ 7.8 Hz, 2H, H200), 7.19 (t, J¼ 7.2 Hz, 2H,H30), 7.16 (d, J¼ 7.2 Hz,1H, H40), 6.89 (s,1H, H3), 6.01 (br s, 2H, NH2),2.62 (m, 2H, H8), 2.48 (m, 2H, H5), 1.73 (m, 4H, H6, H7); 13C NMR(DMSO-d6, 100.5 MHz) d 151.5 (C8a), 148.9 (C9a), 145.6 (C4), 137.7(C100), 134.9 (C2), 132.7 (C10), 128.6 (C300), 128.3 (C30)*, 128.2 (C200)*,127.8 (C20), 126.7 (C400), 126.5 (C40), 106.8 (C4a), 106.2 (C3a), 100.1(C3), 33.3 (C8), 23.1 (C5), 23.0 (C7), 22.8 (C6); MS (EIþ) m/z (%): 339[M]þ (75), 338 (100). Anal. Calcd. for C23H21N3: C, 81.38; H, 6.24; N,12.38. Found: C, 81.07; H, 6.51; N, 12.09.

5.2.13. 1,2-Diphenyl-1,5,6,7,8,9-hexahydrocyclohepta[b]pyrrolo[3,2-e]pyridin-4-amine (13)

Following the general procedure for the Friedländer reaction, toa solution of AlCl3 (730 mg, 5.47 mmol, 3.1 equiv) in 1,2-dichloroethane (36 mL) was added 2-amino-1,5-diphenyl-1H-pyrrole-3-carbonitrile (23) [35] (450 mg, 1.74 mmol) under argonatmosphere. After a few minutes, cycloheptanone (644 mL,5.46 mmol, 3.1 equiv) was added and the reaction mixture wasrefluxed for 48 h yielding pyrrolotacrine 13 (561 mg, 91%): mp205e206 �C; IR (KBr) n 3432, 3337, 3046, 2909, 2847, 1628, 1594,1566, 1532, 1499, 1472, 1452, 1417, 1339, 1203 cm�1; 1H NMR(DMSO-d6, 400 MHz) d 7.37 (t, J¼ 7.4 Hz, 2H, H300), 7.28 (d,J¼ 7.2 Hz, 2H, H20), 7.26 (m, 1 H, H400), 7.23 (d, J¼ 7.4 Hz, 2H, H200),7.19 (t, J¼ 7.2 Hz, 2H, H30), 7.14 (d, J¼ 7.2 Hz, 1H, H40), 6.88 (s, 1H,H3), 6.02 (br s, 2H, NH2), 2.76 (m, 2H, H9), 2.68 (m, 2H, H5), 1.75 (m,2H, H7), 1.51 (m, 4H, H6, H8); 13C NMR (DMSO-d6, 75.4 MHz)d 159.3 (C9a), 149.0 (C10a), 145.6 (C4), 138.3 (C100), 135.6 (C2), 133.4(C10), 129.2 (C300)*, 129.0 (C200)*, 128.9 (C30), 128.5 (C20), 127.4 (C400),127.2 (C40), 112.9 (C4a), 107.5 (C3a), 101.1 (C3), 39.3 (C9), 32.7 (C7),28.6 (C8), 27.6 (C6), 25.5 (C5); MS (EIþ) m/z (%): 353 [M]þ (88), 352(100), 324 (12). Anal. Calcd. for C24H23N3: C, 81.55; H, 6.56; N, 11.89.Found: C, 81.36; H, 6.58; N, 11.75.

5.2.14. 2-(4-Chlorophenyl)-1-phenyl-5,6,7,8-tetrahydro-1H-pyrrolo[2,3-b]quinolin-4-amine (14)

Following the general procedure for the Friedländer reaction, toa solution of AlCl3 (449 mg, 3.06 mmol, 1.8 equiv) in 1,2-dichloroethane (45 mL) was added 2-amino-5-(4-chlorophenyl)-

C. Martins et al. / European Journal of Medicinal Chemistry 46 (2011) 6119e61306128

1-phenyl-1H-pyrrole-3-carbonitrile (24) [35] (500 mg, 1.70 mmol)under argon atmosphere. After a few minutes, cyclohexanone(320 mL, 3.06 mmol, 1.8 equiv) was added and the reaction mixturewas refluxed. After 48 h, AlCl3 (209 mg, 1.57 mmol, 0.9 equiv), indry 1,2-dichloroethane (5 mL), and cyclohexanone (200 mL,1.95 mmol, 1.1 equiv) were added and the reaction refluxed for 6 dyielding pyrrolotacrine 14 (516 mg, 81%): mp 293e295 �C; IR (KBr)n: 3470, 3366, 3072, 2937, 2847, 1631, 1594, 1564, 1530, 1499, 1467,1440, 1339 cm�1; 1H NMR (CDCl3, 400 MHz) d 7.37 (t, J¼ 7.8 Hz, 2H,H300), 7.28 (d, J¼ 7.8 Hz, 3H, H200, H400), 7.19 (d, J¼ 6.6 Hz, 2H, H20),7.13 (d, J¼ 6.6 Hz, 2H, H30), 6.53 (s, 1H, H3), 4.31 (br s, 2H, NH2), 2.87(m, 2H, H8), 2.58 (m, 2H, H5), 1.87 (m, 4H, H6, H7); 13C NMR (CDCl3,100.4 MHz) d 153.2 (C8a), 149.0 (C9a), 143.7 (C4), 137.2 (C100), 136.0(C10), 132.9 (C40), 131.4 (C2), 129.7 (C30), 128.7 (C300), 128.4 (C200),128.3 (C20), 126.7 (C400), 108.2 (C4a), 107.0 (C3a), 98.0 (C3), 33.8 (C8),29.6 (C5), 23.2 (C7)*, 23.1 (C6)*; MS (EIþ)m/z (%): 375 [Mþ 2]þ (27),374 (49), 373 [M]þ (79), 372 (100). Anal. Calcd. for C23H20ClN3: C,73.89; H, 5.39; N, 11.24. Found: C, 73.92; H, 5.87; N, 10.87.

5.2.15. 2-(4-Chlorophenyl)-1-phenyl-1,5,6,7,8,9-hexahydrocyclohepta[b]pyrrolo[3,2-e]pyridin-4-amine (15)

Following the general procedure for the Friedländer reaction, toa solution of AlCl3 (548 mg, 4.11 mmol, 2.0 equiv) in 1,2-dichloroethane (30 mL) was added 2-amino-5-(4-chlorophenyl)-1-phenyl-1H-pyrrole-3-carbonitrile (24) [35] (593 mg, 2.02 mmol)under argon atmosphere. After a few minutes, cycloheptanone(476 mL, 4.04 mmol, 2.0 equiv) was added and the reaction mixturewas refluxed for 48 h, yielding pyrrolotacrine 15 (517 mg, 73%): mp274e276 �C; IR (KBr) n: 3430, 2918, 2848, 1630, 1593, 1566, 1530,1451, 1334 cm�1; 1H NMR (DMSO-d6, 500 MHz) d 7.41 (t, J¼ 5.0 Hz,2H, H300), 7.32 (t, J¼ 5.0 Hz, 2H, H20, H400), 7.24 (d, J¼ 5.0 Hz, 2H,H200), 7.18 (d, J¼ 5.0 Hz, 2H, H30), 6.94 (s, 1H, H3), 5.88 (br s, 2H,NH2), 2.81 (m, 2H, H9), 2.72 (m, 2H, H5), 1.78 (m, 2H, H7), 1.56 (m,4H, H6, H8); 13C NMR (DMSO-d6, 125 MHz) d 159.8 (C9a), 149.1(C10a), 145.8 (C4), 138.4 (C100), 134.8 (C10), 132.6 (C40)*, 132.4 (C2)*,130.3 (C30), 129.4 (C300), 129.1 (C20)**, 129.1 (C200)**, 127.5 (C400), 113.3(C4a), 107.9 (C3a), 101.7 (C3), 39.7 (C9), 32.6 (C7), 28.6 (C8), 27.6(C6), 25.7 (C5); MS (EIþ) m/z (%): 389 [Mþ 2]þ (31), 388 (50), 387[M]þ (91), 386 (100), 358 (15). Anal. Calcd. for C24H22ClN3: C, 74.31;H, 5.72; N, 10.83; Cl, 9.14. Found: C, 74.12; H, 5.93; N, 11.03; Cl, 9.44.

5.2.16. 1-Phenyl-2-(p-tolyl)-5,6,7,8-tetrahydro-1H-pyrrolo[2,3-b]quinolin-4-amine (16)

Following the general procedure for the Friedländer reaction, toa solution of AlCl3 (306 mg, 2.30 mmol, 2.1 equiv) in 1,2-dichloroethane (20 mL) was added 2-amino-1-phenyl-5-(p-tolyl)-1H-pyrrole-3-carbonitrile (25) (302 mg, 1.10 mmol) under argonatmosphere. After a few minutes, cyclohexanone (231 mL,2.23 mmol, 2.1 equiv) was added and the reaction mixture wasrefluxed. After 72 h, AlCl3 (228 mg,1.71 mmol, 1.5 equiv), in dry 1,2-dichloroethane (10 mL), and cyclohexanone (171 mL, 1.5 mmol)were added and the reaction was refluxed for 4 d yielding pyrro-lotacrine 16 [35] (285 mg, 73%): mp 305e306 �C; IR (KBr) n 3469,3364, 3067, 3014, 2940, 2849, 1632, 1595, 1563, 1539, 1500, 1467,1441, 1338 cm�1; 1H NMR (DMSO-d6, 300 MHz) d 7.38 (t, J¼ 9.0 Hz,2H, H300), 7.31 (d, J¼ 9.0 Hz, 1H, H400), 7.22 (d, J¼ 9.0 Hz, 2H, H200),7.07 (br s, 4H, H20, H30), 6.84 (s, 1H, H3), 5.75 (br s, 2H, NH2), 2.66(m, 2H, H8), 2.50 (m, 2H, H5), 2.26 (s, 3H, CH3), 1.78 (m, 4H, H6, H7);13C NMR (DMSO-d6, 75.4 MHz) d 152.2 (C8a), 149.8 (C9a), 146.2(C4), 138.8 (C100), 136.9 (C40), 136.3 (C2), 130.9 (C10), 129.6 (C300),129.3 (C200), 129.2 (C30), 128.7 (C20), 127.2 (C400), 107.9 (C4a), 107.3(C3a), 100.3 (C3), 34.1 (C8), 23.9 (C5), 23.8 (C7), 23.7 (C6), 21.4(CH3); MS (EIþ) m/z (%): 353 [M]þ (100), 352 (98), 176 (11). Anal.Calcd. for C24H23N3: C, 81.55; H, 6.56; N, 11.89. Found: C, 81.28; H,6.67; N, 11.59. The hydrochloride of pyrrolotacrine 16 was obtained

by treating it with hydrogen chloride gas generated by the reactionof sulphuric acid with sodium chloride crystals. Anal. Calcd. forC24H24ClN3O$HCl: C, 73.93, H, 6.20; N, 10.78; Cl, 9.09. Found: 73.85;H, 6.48; N, 10.49; Cl, 9.21.

5.3. Pharmacology

5.3.1. Evaluation of EeAChE and eqBuChE inhibitory activitiesTo assess the inhibitory activity of both cholinesterases, the

spectrophotometric method of Ellman [37] was followed, usingpurified AChE from E. electricus (Type V-S, Sigma), or BuChE fromhorse serum (lyophilized powder) (SigmaeAldrich, Madrid, Spain).The reaction took place in a final volume of 3 mL of a phosphate-buffered solution (0.1 M) at pH 8, containing 0.035 U of AChE or0.05 U of BuChE and 0.35 mM of 5,50-dithio-bis(2-nitrobenzoicacid) (DTNB, SigmaeAldrich, Madrid, Spain). Inhibition curveswere made by pre-incubating this mixture with at least nineconcentrations of each compound for 10 min. A sample with nocompound was always present to determine the 100% of enzymeactivity. After this pre-incubation period, acetylthiocholine iodide(0.35 mM) or butyrylthiocholine iodide (0.5 mM) (SigmaeAldrich,Madrid, Spain) were added, allowing 15 min of incubation, wherethe DTNB produces the yellow anion 5-thio-2-nitrobenzoic acidalong with the enzymatic degradation of acetylthiocholine iodideor butyrylthiocholine iodide. Changes in absorbance were detectedat 405 nm in a spectrophotometric plate reader (FluoStar OPTIMA,BMG Labtech). Compounds inhibiting AChE or BuChE activitywould reduce the colour generation, thus IC50 values were calcu-lated as the concentration of compound that produces 50% of theAChE activity inhibition. Data are expressed as means� SEM of atleast three different experiments in quadruplicate.

5.3.2. Evaluation of hAChE and hBuChE inhibitory activitiesThe evaluation of the inhibitory activity against human

recombinant AChE (Sigma, Italy) and BuChE from human serum(Sigma, Italy) was carried out with 0.02 U/mL of enzyme, 550 mM ofsubstrate (acetylthiocholine and butyrylthiocholine, respectively),340 mM 5,50-dithio-bis(2-nitrobenzoic acid) and 0.1 M potassiumphosphate buffer pH 8 as assay medium, in a final reaction volumeof 1 mL. The method of Ellman [37] was followed to evaluateactivity and calculate inhibition.

5.3.3. Kinetic analysis of the AChE inhibition by compound 12To obtain estimates of the mechanism of action of compound 12,

reciprocal plots of 1/V versus 1/[S] were constructed at differentconcentrations of the substrate acetylthiocholine (0.1e0.8 mM) byusing Ellman’s method [37]. Experiments were performed ina transparent 48-well plate containing each well 400 mL of theDTNB 0.875 mM solution in PBS 0.1 M, 1 mL of DMSO (control) orinhibitor solution to give the desired final concentration. Finalvolume (1 mL) was reached by adding PBS (pH 8). Reaction wasinitiated by adding 14 mL of AChE at 30 �C to give a final concen-tration of 0.05 U/mL. Progress curves were monitored at 412 nmover 1.33 min in a fluorescence plate reader Fluostar Optima (BMG-technologies, Germany). Progress curves were characterized bya linear steady-state turnover of the substrate and values of a linearregression were fitted according to LineweavereBurk replots usingOrigin software. The plots were assessed by aweighted least squareanalysis. Determination of Michaelis constant for the substrateATCh was done at 6 different concentrations (0e0.8 mM) to givea value of KM¼ (0.4� 0.1) mM, and Vmax¼ (0.41�0.07) min�1.Slopes of the reciprocal plots were then plotted against theconcentration of 12 (range 0e2 mM) to evaluate Ki data. Dataanalysis was performed with Origin Pro 8.0 software (OriginLab Corp.).

C. Martins et al. / European Journal of Medicinal Chemistry 46 (2011) 6119e6130 6129

5.3.4. Cell culture of the human neuroblastoma cell line SH-SY5YSH-SY5Y cells were maintained in a 1:1 mixture of F-12 nutrient

mixture (Ham12) (SigmaeAldrich, Spain) and Eagle’s minimumessential medium (EMEM) supplemented with 15 nonessentialamino acids, 1 mM sodium pyruvate, 10% heat-inactivated foetalbovine serum (FBS), 100 U/mL penicillin, and 100 mg/mL strepto-mycin (reagents from Invitrogen, Spain). Cultures were seeded intoflasks containing supplemented medium and maintained at 37 �Cin a humidified atmosphere of 5% CO2 and 95% air. For assays, SH-SY5Y cells were sub-cultured in 48-well plates at a seedingdensity of 1�105 cells per well. Cells were treated with the testedcompound before confluence in F12/EMEM with 1% FBS. All thecells used in this study were used at a low passage number (<13).

5.3.5. Incubation of drugsTo compare the effects of compound 12 with tacrine on cell

viability, SH-SY5Y cells were incubated for 48 hwith 3 and 30 mMofcompound 12, and thereafter, cell viability was measured as MTTreduction.

To assess the neuroprotective effect of compound 12 againstAb25e35 e induced toxicity, SH-SY5Y cells were pre-incubated withcompound 12 at the concentrations of 0.3, 1, and 3 mM for 24 h.Then, cells were co-incubated for another 24 h period withcompound 12 in the presence of 10 mM Ab25e35. At the end of theexperiment, cell viability was assessed using the MTT method.

5.3.6. Measurement of cell viability with MTTCell viability, virtually the mitochondrial activity of living cells,

was measured by quantitative colorimetric assay with MTT (3-[4,5-dimethylthiazol-2-yl]-2,5-diphenyltetrazolium bromide, Sigma-eAldrich, Spain), as described previously. MTT was added to allwells (final concentration 0.5 mg/mL) and allowed to incubate inthe dark at 37 �C for 2 h. The tetrazolium ring of MTTcan be cleavedby mitochondrial reductases in order to produce a precipitatedformazan derivative. After this 2 h period, the formazan producedwas dissolved by adding 200 mL of DMSO, resulting in a colouredcompound whose optical density was measured in an ELISA readerat 540 nm. All MTT assays were performed in triplicate. Absorbancevalues obtained in control cells untreated with the toxic wereconsidered as 100% viability.

5.4. Molecular docking

Compounds 3 and 12 were assembled within Discovery Studio,version 2.1, software package, using standard bond lengths andbond angles without protonation at the pyridine ring. With theCHARMm force field [44] and partial atomic charges, the moleculargeometry of these compounds was energy-minimized using theadopted-based NewtoneRaphson algorithm. Structures wereconsidered fully optimized when the energy changes betweeninteractions were less than 0.01 kcal/mol [45].

5.4.1. Molecular docking into EeAChEThe coordinates of EeAChE (PDB ID: 1C2B) were obtained from

the Protein Data Bank (PDB). For docking studies, initial proteinwasprepared by removing all water molecules, heteroatoms, any co-crystallized solvent and the ligand. It is well known that PDB filesoften have poor or missing assignments of explicit hydrogens, andthe PDB file format cannot accommodate bond order information.Therefore, proper bonds, bond orders, hybridization and chargeswere assigned using proteinmodel tool in Discovery Studio, version2.1, software package. CHARMm force field was applied using thereceptoreligand interactions tool in Discovery Studio, version 2.1,software package.

Docking calculations were performed with the program Auto-Dock Vina [40]. AUTODOCKTOOLS (ADT) (version 1.5.4) was used toadd hydrogens and partial charges for proteins and ligands usingGasteiger charges. Flexible torsions in the ligands were assignedwith the AUTOTORS module, and the acyclic dihedral angles wereallowed to rotate freely.

Because Vina uses rectangular boxes for the binding site, the boxcentre was defined and the docking box was displayed using ADT.For EeAChE (PDB: 1C2B) the docking procedure was applied to thewhole protein target, without imposing the binding site (“blinddocking”). A grid box of 60� 60� 72 with grid points separated 1�A,was positioned at the middle of the protein (x¼ 21.5911;y¼ 87.752; z¼ 23.591). Default parameters were used exceptnum_modes, which was set to 40.

5.4.2. Molecular docking of compounds 3 and 12 into eqBuChEThe eqBuChEmodel has been retrieved from The SWISS-MODEL

Repository. This is a database of annotated three-dimensionalcomparative protein structure models generated by the fullyautomated homology-modelling pipeline SWISS-MODEL. A puta-tive three-dimensional structure of eqBuChE has been createdbased on the crystal structure of hBuChE (pdb: 2pm8). These twoenzymes exhibited 89% sequence identity. Proper bonds, bondorders, hybridization and charges were assigned using proteinmodel tool in Discovery Studio, version 2.1, software package.CHARMm force field was applied using the receptoreligand inter-actions tool in Discovery Studio, version 2.1, software package.

Docking calculations were performed following the sameprotocol described before for EeAChE. All dockings were performedas blind dockings where a cube of 75�A with grid points separated1�A was positioned at the middle of the protein (x¼ 29.885;y¼�54.992; z¼ 58.141). Default parameters were used exceptnum_modes, which was set to 40.

The lowest docking-energy conformation was considered as themost stable orientation. Finally, the docking results generated weredirectly loaded into Discovery Studio, version 2.1.

Acknowledgements

Carla Martins thanks Fundação para a Ciência e Tecnologia doMinistério da Ciência, Tecnologia e Ensino Superior of Portugal forher Ph D grant SFRH/BD/17577/2004. JMC thanks MICINN(SAF2006-08764-C02-01; SAF2009-07271), Comunidad de Madrid(S/SAL-0275-2006), and CSIC-GRICES (2007PT-13) for financialsupport.

MG and JE thank the Spanish Ministry of Science and InnovationRef. SAF2009-12150 and the Spanish Ministry of Health (Institutode Salud Carlos III) RETICS-RD06/0026 to MGL. CM thanks COSTAction D34 for her STSM, which made possible this collaborativework.

References

[1] M.L. Bolognesi, A. Cavalli, L. Valgimigli, M. Bartolini, V. Andrisano,M. Recanatini, C. Melchiorre, J. Med. Chem. 50 (2007) 6446e6449.

[2] M. Rosini, E. Simoni, M. Bartolini, A. Cavalli, L. Ceccarini, N. Pascu,D.W. McClymont, A. Tarozzi, M.L. Bolognesi, A. Minarini, V. Tumiatti,V. Andrisano, I.R. Mellor, C. Melchiorre, J. Med. Chem. 51 (2008) 4381e4384.

[3] S. Rizzo, C. Rivière, L. Piazzi, A. Bisi, S. Gobbi, M. Bartolini, V. Andrisano,F. Morroni, A. Tarozzi, J.P. Monti, A. Rampa, J. Med. Chem. 51 (2008)2883e2886.

[4] R. Morphy, Z. Rankovic, J. Med. Chem. 48 (2005) 6523e6543.[5] M.B.H. Youdim, J.J. Buccafusco, Trends Pharmacol. Sci. 26 (2005) 27e35.[6] L. Piazzi, A. Cavalli, F. Colizzi, F. Belluti, M. Bartolini, F. Mancini, M. Recanatini,

V. Andrisano, A. Rampa, Bioorg. Med. Chem. Lett. 18 (2008) 423e426.[7] A. Cavalli, M.L. Bolognesi, A. Minarini, M. Rosini, V. Tumiatti, M. Recanatini,

C. Melchiorre, J. Med. Chem. 51 (2008) 347e372.

C. Martins et al. / European Journal of Medicinal Chemistry 46 (2011) 6119e61306130

[8] L. Fang, D. Appenroth, M. Decker, M. Kiehntopf, A. Lupp, S. Peng, C. Fleck,Y. Zhang, J. Lehmann, J. Med. Chem. 51 (2008) 7666e7669.

[9] M. Decker, Mini-Rev. Med. Chem. 7 (2007) 221e229.[10] M.L. Bolognesi, R. Banzi, M. Bartolini, A. Cavalli, A. Tarozzi, V. Andrisano,

A. Minarini, M. Rosini, V. Tumiatti, C. Bergamini, R. Fato, G. Lenaz, P. Hrelia,A. Cattaneo, M. Recanatini, C. Melchiorre, J. Med. Chem. 50 (2007) 4882e4897.

[11] A. Cavalli, M.L. Bolognesi, S. Capsoni, V. Andrisano, M. Bartolini, E. Margotti,A. Cattaneo, M. Recanatini, C. Melchiorre, Angew. Chem., Int. Ed. 46 (2007)3689e3692.

[12] D. Muñoz-Torrero, Curr. Med. Chem. 15 (2008) 2433e2455.[13] N.C. Inestrosa, A. Alvarez, F. Calderón, Mol. Psychiatry 1 (1996) 359e361.[14] M. Bartolini, C. Bertucci, V. Cavrini, V. Andrisano, Biochem. Pharmacol. 65

(2003) 407e416.[15] M. Racchi, E. Porrello, C. Lanni, S.C. Lenzken, M. Mazzucchelli, S. Govoni, Aging

Clin. Exp. Res. 18 (2006) 149e152.[16] E. Giacobini, F. Mori, C.C. Lai, Ann. N.Y. Acad. Sci. 777 (1996) 393e398.[17] E. Giacobini, Ann. N.Y. Acad. Sci. 920 (2000) 321e327.[18] E. Giacobini, CNS Drugs 15 (2001) 85e91.[19] M. Pakaski, P. Kasa, Curr. Drug Targets CNS Neurol. Disord. 2 (2003) 163e171.[20] M. Racchi, M. Mazzucchelli, S.C. Lenzken, E. Porrello, C. Lanni, S. Govoni, Chem.

Biol. Interact. 157e158 (2005) 335e338.[21] F. Mori, L. Chen-Ching, F. Fusi, E. Giacobini, Neuroreport 6 (1995) 633e636.[22] E. Giacobini, Neurochem. Res. 25 (2000) 1185e1190.[23] M. Zimmermann, B. Borroni, F. Cattabeni, A. Padovani, M. Di Luca, Neurobiol.

Dis. 19 (2005) 237e242.[24] H.C. Liu, C.W. Chi, S.Y. Ko, H.C. Wang, C.J. Hong, K.N. Lin, P.N. Wang, T.Y. Liu,

Dement. Geriatr. Cogn. Disord. 19 (2005) 345e348.[25] M. Pakaski, J. Kálmán, Neurochem. Int. 53 (2008) 103e111.[26] R. Cacabelos, Neuropsychiatr. Dis. Treat 3 (2007) 303e333.[27] A. Castro, A. Martinez, Curr. Pharm. Des. 12 (2006) 4377e4387.[28] M.L. Bolognesi, M. Rosini, V. Andrisano, M. Bartolini, A. Minarini, V. Tumiatti,

C. Melchiorre, Curr. Pharm. Des. 15 (2009) 601e613.[29] J. Marco-Contelles, R. León, C. de Los Ríos, A. Samadi, M. Bartolini,

V. Andrisano, O. Huertas, X. Barril, F.J. Luque, M.I. Rodríguez-Franco, B. López,

M.G. López, A.G. García, M.C. Carreiras, M. Villarroya, J. Med. Chem. 52 (2009)2724e2732 and references cited therein.

[30] (“Thiazolotacrines”) D. Thomae, E. Perspicace, S. Hesse, G. Kirsch, P. Seck,Tetrahedron 64 (2008) 9309e9314.

[31] ("Pyrrolotacrines”) H. Bekolo, G. Kirsch, Can. J. Chem. 85 (2007) 1e6.[32] (“Selenophenetacrines”) D. Thomae, G. Kirsch, P. Seck, Synthesis (2008)

1600e1606.[33] (“Thiophenetacrines”) D. Thomae, G. Kirsch, P. Seck, Synthesis (2007)

1027e1032.[34] (“Benzofuranotacrines”) C.-L. Yang, C.-H. Tseng, Y.-L. Chen, C.-M. Lu, C.-L. Kao,

M.-H. Wu, C.-C. Tzeng, Eur. J. Med. Chem. 45 (2010) 602e607.[35] C. Martins, A. Eleuterio, M.C. Carreiras, A. Samadi, E. Soriano, R. León,

J.M. Marco-Contelles, Mol. Divers. 13 (2009) 459e468.[36] J. Marco-Contelles, E.M. Pérez-Mayoral, A. Samadi, M.C. Carreiras, E. Soriano,

Chem. Rev. 109 (2009) 2652e2671.[37] G.L. Ellman, K.D. Courtney, B.J. Andres, R.M. Featherstone, Biochem. Pharma-

col. 7 (1961) 88e90.[38] C. Hetenyi, D. Van der Spoel, Protein Sci. 20 (2011) 880e893.[39] C. Hetenyi, D. Van der Spoel, FEBS Lett. 580 (2006) 1447e1450.[40] O. Trott, A.J. Olson, J. Comput. Chem. 31 (2010) 455e461.[41] K. Arnold, L. Bordoli, J. Kopp, T. Schwede, Bioinformatics 22 (2006)

195e201.[42] F. Kiefer, K. Arnold, M. Künzli, L. Bordoli, T. Schwede, Nucleic Acids Res. 37

(2009) D387eD392.[43] M.C. Peitsch, Bio/Technology 13 (1995) 658e660.[44] B.R. Brooks, R.E. Bruccoleri, B.D. Olafson, D.J. States, S. Swaminathan,

M. Karplus, M. Charmm, J. Comput. Chem. 4 (1983) 187e217.[45] A. Morreale, F. Maseras, I. Iriepa, E. Galvez, J. Mol. Graph. Model. 21 (2002)

111e118.[46] M.L. Bolognesi, V. Andrisano, M. Bartolini, A. Cavalli, A. Minarini,

M. Recanatini, M. Rosini, V. Tumiati, C. Melchiorre, J. Med. Chem. 60 (2005)465e473.

[47] Q.-S. Yu, H.W. Holloway, T. Utsuki, A. Brossi, N.H. Greig, J. Med. Chem. 42(1999) 1855e1861.