Novel Approach for the Search for Chemical Scaffolds with Dual

molecules

Article Novel Approach for the Search for Chemical Scaffolds with Dual Activity with Acetylcholinesterase and the α7 Nicotinic Acetylcholine Receptor—A Perspective for the Treatment of Neurodegenerative Disorders

Natalia M. Kowal 1,2,3, Dinesh C. Indurthi 2,3 , Philip K. Ahring 2,3, Mary Chebib 2,3, Elin S. Olafsdottir 1 and Thomas Balle 2,3,*

1 Faculty of Pharmaceutical Sciences, School of Health Sciences, University of Iceland, 107 Reykjavik, Iceland; [email protected] (N.M.K.); [email protected] (E.S.O.) 2 Sydney School of Pharmacy, Faculty of Medicine and Health, The University of Sydney, Sydney, NSW 2006, Australia; [email protected] (D.C.I.); [email protected] (P.K.A.); [email protected] (M.C.) 3 Brain and Mind Centre, The University of Sydney, Camperdown, NSW 2050, Australia * Correspondence: [email protected]; Tel.: +61-2-9036-7035

Academic Editor: Simone Brogi Received: 7 December 2018; Accepted: 23 January 2019; Published: 27 January 2019

Abstract: Neurodegenerative disorders, including Alzheimer’s disease, belong to the group of the most difficult and challenging conditions with very limited treatment options. Attempts to find new drugs in most cases fail at the clinical stage. New tactics to develop better drug candidates to manage these diseases are urgently needed. It is evident that better understanding of the neurodegeneration process is required and targeting multiple receptors may be essential. Herein, we present a novel approach, searching for dual active compounds interacting with acetylcholinesterase (AChE) and the α7 nicotinic acetylcholine receptor (nAChR) using computational chemistry methods including homology modelling and high throughput virtual screening. Activities of identified hits were evaluated at the two targets using the colorimetric method of Ellman and two-electrode voltage-clamp electrophysiology, respectively. Out of 87,250 compounds from a ZINC database of natural products and their derivatives, we identified two compounds, 8 and 9, with dual activity and balanced IC50 values of 10 and 5 µM at AChE, and 34 and 14 µM at α7 nAChR, respectively. This is the first report presenting successful use of virtual screening in finding compounds with dual mode of action inhibiting both the AChE enzyme and the α7 nAChR and shows that computational methods can be a valuable tool in the early lead discovery process.

Keywords: virtual screening; multi modal; dual mode of action; nAChR; nicotinic acetylcholine receptor; AChE; acetylcholinesterase; docking; lead identiﬁcation; neurodegenerative disorders

1. Introduction Alzheimer’s disease (AD) is the most common form of dementia affecting, in particular, the elderly population. So far, the disease is not well understood, and despite huge research efforts by academia and the pharmaceutical industry to understand the origins of the disease, the identiﬁcation of new drugs to treat or alleviate symptoms of memory impairment have proven difﬁcult. The drugs that are currently on the market are mainly acetylcholinesterase (AChE) inhibitors including galantamine (1), donepezil (2), and rivastigmine (3), and the NMDA receptor antagonist, memantine (4) (Figure1). These drugs do not slow progression of the disease [1], but they help to alleviate some of the symptoms. To further enhance their effect, it has been suggested that combining AChE inhibitors

Molecules 2019, 24, 446; doi:10.3390/molecules24030446 www.mdpi.com/journal/molecules Molecules 2019, 24, 446 2 of 17 with other “pro-cognitive receptor” drug targets such as NMDA receptors could be a better strategy to alleviate cognitive impairment in dementia. Thus, combining treatments such as AChE inhibitors with memantine could improve outcomes [2–4] as reported in clinical trials involving patients with moderate-to-severeMolecules 2019, 24, x AD FOR [PEER5,6]. REVIEW 8 of 18

FigureFigure 1. Structures 1. Structures of currently of currently approved approved Alzheimer’s Alzheimer’s disease disease (AD) (AD) drugs drugs galantamine galantamine (1), donepezil (1), donepezil (2), rivastigmine(2), rivastigmine (3) and (3) and memantine memantine (4 ).(4).

Instead2. Results of a poly-pharmaceutical approach where two or more drugs are combined, an attractive alternativeIn this study would we aimed be to tohave identify one comp compoundounds possessing simultaneously activity targetingat both the multiple α7 nAChRs receptors and [7]. Recently,AChE. it wasWe suggestedchose galantamine that a multi-targetas our lead compou approachnd and might performed be a viable virtual tactic screening to combat towards complex neurodegenerativehomology models diseases of the α like7 nAChR AD [8 and,9]. an However, X-ray structure identification of an AChE. of leadCompounds molecules that forscored development well of multi-modalin both screens drugs were represents subsequently a challenge. purchased and evaluated by two-electrode voltage-clamp Computationalelectrophysiology techniques in vitro at including the human virtual α7 nAChR screening expressed (VS) arein Xenopus attractive oocytes technologies and against for faster identificationElectrophorus of drug electricus leads AChE [8,9 using] and the may Ellman’s also prove colorimetric to be assay. powerful tools for finding multi-modal drugs,2.1. if sufficientVirtual Screening structural information of the desired drug targets is available. In the past decade, the number of characterized crystal structures has increased significantly along with developments in We focused the search on natural products and natural product derivatives, which are a valuable docking software and hardware, so it is now possible to filter millions of compounds within a short source of active compounds against both targets. We screened a database consisting of 87,250 natural timeframe.products Further, and natural virtual product screening derivatives against from one the target ZINC proved database to merged be more with efficient an in-house for identification database of drug leadscontaining versus 250 conventional lycopodium alkaloids. high throughput The compound screenings were virtually (HTS) with screened, success first rates at two of homology 1–40% versus 0.01–0.14%,models respectivelyof the α7 nAChR, [10]. and However, then the top dual hits or were multi-target docked to the screenings active site areof a notco-crystal common structure and their successof ratesthe AChE differ enzyme [11,12 bound]. with galantamine. Galantamine was identified amongst the top scoring Forhits the in all treatment screens. After of AD, post-docking an interesting filtering, multi-modal 78 compounds drug were profile left out, would of which be a asub-set compound of 13 that simultaneouslycompounds acts(Table as 1), an i.e., AChE galantamine inhibitor analogues and a (6 positive, 7) and structurally allosteric modulatorunrelated compounds (PAM) at (8– one of the pro-cognitive18), were selected nicotinic and purchased acetylcholine for biological receptors assessment. (nAChR), especially the α7 or α4β2 subtype. This combination2.2. Assessment would of Activity lead with to enhancedAChE and α synaptic7 nAChRs acetylcholine (ACh) levels along with a subtype specific potentiation of nAChRs, which would be a more refined way to modulate the cholinergic The selected compounds and galantamine were initially evaluated for activity with AChE and system, compared to increasing the dose of an AChE inhibitor alone. In particular, the α7 subtype of the α7 nAChR using a single concentration of a compound. For assessment of AChE inhibitory the nAChRactivity, is we an used interesting a colorimetric target microplate for AD therapy. assay ba Specificsed on the activation colour change of this during receptor reaction is knownof ACh to be pro-cognitivewith Ellman’s [13–17 reagent.]. All compounds were evaluated at 100 µM concentration and all selected Incompounds, the search except for new compound compounds 13 that thatwas simultaneouslyinactive, showed inhibition target AChE ranging and from the 7α to7 nAChR,95.4%. we embarkedCompounds on a lead 6 and discovery 7, sanguinine project and norgalantamine, using high throughput which are virtualgalantamine screening analogues, (HTVS). showed As 96.4 a model compoundand 96.5% we selectedinhibition galantamine of AChE activity that at is100 marketed µM, respectively. as an AD Two drug additional and has compounds been described showed in the literatureinhibition as a druggreater with than the 70%. desired These dual compounds, mode of 8 action,and 9, i.e.,which inhibitor are structurally of the AChE unrelated and to a PAM galantamine, at 100 µM inhibited AChE by 78.5 and 88.1%, respectively. Galantamine (10 µM), the at the α7 nAChRs [18,19]. Further, as co-crystal structures with galantamine are available of both known reference compound, inhibited AChE activity by 91.5%. At a similar concentration (10 µM),

Molecules 2019, 24, 446 3 of 17

AChE [20] and an acetylcholine binding protein (AChBP) [21], a commonly used structural surrogate for nAChRs [22], it seemed to be the obvious choice for a structure guided lead discovery project. We developed a protocol for high throughput virtual screening to identify multi-modal drug leads utilizing α7 homology models and an AChE X-ray structure and selected compounds that ranked high in both screens for in vitro evaluation at α7 nAChRs and of the AChE. In the current paper we present the virtual screening results conﬁrming that virtual screening is an attractive technique that can successfully be used to identify compounds simultaneously targeting the AChE and the α7 nAChR.

2. Results In this study we aimed to identify compounds possessing activity at both the α7 nAChRs and AChE. We chose galantamine as our lead compound and performed virtual screening towards homology models of the α7 nAChR and an X-ray structure of an AChE. Compounds that scored well in both screens were subsequently purchased and evaluated by two-electrode voltage-clamp electrophysiology in vitro at the human α7 nAChR expressed in Xenopus oocytes and against Electrophorus electricus AChE using the Ellman’s colorimetric assay.

2.1. Virtual Screening We focused the search on natural products and natural product derivatives, which are a valuable source of active compounds against both targets. We screened a database consisting of 87,250 natural products and natural product derivatives from the ZINC database merged with an in-house database containing 250 lycopodium alkaloids. The compounds were virtually screened, first at two homology models of the α7 nAChR, and then the top hits were docked to the active site of a co-crystal structure of the AChE enzyme bound with galantamine. Galantamine was identified amongst the top scoring hits in all screens. After post-docking filtering, 78 compounds were left out, of which a sub-set of 13 compounds (Table1), i.e., galantamine analogues ( 6, 7) and structurally unrelated compounds (8–18), were selected and purchased for biological assessment.

2.2. Assessment of Activity with AChE and α7 nAChRs The selected compounds and galantamine were initially evaluated for activity with AChE and the α7 nAChR using a single concentration of a compound. For assessment of AChE inhibitory activity, we used a colorimetric microplate assay based on the colour change during reaction of ACh with Ellman’s reagent. All compounds were evaluated at 100 µM concentration and all selected compounds, except compound 13 that was inactive, showed inhibition ranging from 7 to 95.4%. Compounds 6 and 7, sanguinine and norgalantamine, which are galantamine analogues, showed 96.4 and 96.5% inhibition of AChE activity at 100 µM, respectively. Two additional compounds showed inhibition greater than 70%. These compounds, 8 and 9, which are structurally unrelated to galantamine, at 100 µM inhibited AChE by 78.5 and 88.1%, respectively. Galantamine (10 µM), the known reference compound, inhibited AChE activity by 91.5%. At a similar concentration (10 µM), compounds 6 and 7 showed 95.5 and 88.6% AChE inhibition, respectively, reaching a plateau at 10 µM. All compounds were subsequently evaluated at the human α7 nAChR expressed in Xenopus oocytes using two-electrode voltage-clamp electrophysiology. The oocytes were pre-incubated with the test compound and subsequently the test compound was co-applied with an EC20 concentration of ACh, an experimental design adopted to facilitate identification of PAMs. We first confirmed the ability to identify PAMs by testing NS1738, a well-established α7 PAM [16]. As evident from Figure2, robust potentiation (440% at 31.6 µM) of (30 µM) ACh-evoked currents was observed. Unexpectedly, galantamine did not show any PAM activity at concentrations ranging from 10 nM to 100 µM, instead, inhibition of the (30 µM) ACh-evoked response was observed. At the highest concentration, galantamine inhibited ACh by 67.3%. These results triggered an in depth evaluation of galantamine effects at the α7 and the α4β2 nAChRs. The outcome of this study was published recently [23], with the conclusion that galantamine is not a PAM of the investigated nAChRs. Out of Molecules 2019, 24, 446 4 of 17 the 13 tested compounds in the present study, all except compound 18 inhibited the α7 nAChR and compounds 9–13 showed more than 90% inhibition at 100 µM. Molecules 2019, 24, x FOR PEER REVIEW 8 of 18

Figure 2.FigureRepresentative 2. Representative current current traces traces for NS1738, NS1738, galantamine, galantamine, and selected and selected compounds compounds from α7 from α7 nAChRsnAChRs expressed expressed in inXenopus Xenopus oocytes.oocytes. Cells Cells were were subjected subjected to two-electrode to two-electrode voltage-clamp voltage-clamp electrophysiology experiments; the oocyte membrane potential was clamped at −60 mV. All electrophysiology experiments; the oocyte membrane potential was clamped at −60 mV. All experiments experiments involved a pre-incubation protocol that consisted of 25 s application of the test solution involved(or a a pre-incubation saline solution for protocol the reference that consistedtrace) followed of 25 by s 20 application s co-application of the with test 30 solutionµM ACh. (orThe a saline solutionrepresentative for the reference traces were trace) baseline followed subtracted, by 20 and s co-application the bars above each with trace 30 µrepresentM ACh. the The application representative traces wereperiods, baseline and concentrations subtracted, of and the thetest barssolutions above appear each above trace the represent bars. The themajority application of the washing periods, and concentrationsperiods (3 of min) the testbetween solutions each trace appear are omitted. above the bars. The majority of the washing periods (3 min) between each trace are omitted. Molecules 2019, 24, 446 5 of 17

For compounds exceeding 70% and 90% inhibition at 100 µM at AChE and the α7 nAChR, respectively, IC50 values were determined based on full concentration response relationships (Table1, Figures2 and3A). To verify the AChE assay, we first tested physostigmine ( 5) and galantamine inhibition and found IC50 values of 0.78 and 0.68 µM, respectively, which are in agreement with the previously published results [24–27]. Two galantamine analogues, 6 and 7, were ca 3-fold more potent than galantamine itself at the AChE with IC50 values of 0.28 and 0.23 µM, respectively. Compounds 8 and 9 were ca 13- and 6-fold less potent compared to galantamine, with IC50 values of 10.6 and 5.0 µM, respectively. All compounds for which IC50 values were determined at the α7 nAChR (8–13) were stronger inhibitors than galantamine, with compounds 10–12 displaying single digit micromolar potencies corresponding to 6- to 14-fold higher potency. Compound 13 inhibited α7 nAChRs with an IC Molecules 2019, 24, x FOR PEER REVIEW 8 of 18 50 value of 13.1 µM. Compounds 8 and 9 were less potent with IC50 values of 34.3 and 14.5 µM, respectively.

Figure 3. Inhibition of ACh-evoked currents from the human α7 nAChR expressed in Xenopus oocytes Figure 3. Inhibition of ACh-evoked currents from the human α7 nAChR expressed in Xenopus oocytes of galantamine and selected VS hits. (A) Concentration-response relationships measured by whole-cell, of galantamine and selected VS hits. (A) Concentration-response relationships measured by whole- voltage-clamp experiments (holding potential −60 mV). Inward nicotinic currents were recorded after cell, voltage-clamp experiments (holding potential −60 mV). Inward nicotinic currents were recorded µ 25 s incubationafter 25 s incubation with a test with or Gala test solution or Gal solution followed followed by 20 by s 20 co-application s co-application with with30 30µMM ACh. ACh. Peak Peak currentcurrent amplitudes amplitudes were measured were measured and normalizedand normalized with with respect respect to theto the amplitude amplitude of of current current elicited elicited by 30 µM ACh,by 30 µMn = ACh, 3. (B )n Potential-inhibition= 3. (B) Potential-inhibition dependence dependence was was determined determined by by conductingconducting experiments experiments at two holdingat two holding potentials potentials−50 and−50 and−100 −100 mV. mV. Oocytes Oocytes were were pre-incubated pre-incubated with with the the test test solution solution (25 (25 s) s) and theand test the solution test solution with 30 withµM 30 ACh µM ACh was co-applied.was co-applied. Gal Gal was was tested tested at at 30 30µ M,µM, compounds compounds 88 andand 9 at 10 µM. Peakat 10 µM. current Peak amplitudes current amplitudes were normalized were normalized with with respect respect to the to the amplitude amplitude of of current current elicited elicited by 30 µM ACh,by 30 nµM= 3–4ACh, cells. n = 3–4 cells.

Of the 13 screened compounds, two compounds, 8 and 9, met the criteria for being “dual mode of action”, corresponding to a virtual screening hit rate of 15% across two targets. Since galantamine at high concentrations acts as an open channel pore blocker [23], we investigated if the inhibitory activity of the dual mode of action compounds was dependent on membrane potential. We designed a simple experiment where application of a single concentration of a drug was applied at two different membrane holding potentials, −100 and −50 mV. At a lower, more negative potential, the driving force for ion flow is higher and causes a stronger inhibition of ACh induced currents for positively charged channel blockers. Currents elicited in the presence of antagonists (compounds binding elsewhere on the receptor) are independent of the membrane potential. In this assay, galantamine as well as compounds 8 and 9 showed ~30% inhibition of (30 µM) ACh-induced currents when tested at −50 mV (note that galantamine was tested at 30 µM and compounds 8 and 9 due to their higher potency, at 10 µM). Inhibition increased for galantamine and compound 9 when tested at −100 mV and was different for each compound: 42% for galantamine and 25% for compound 9. Inhibition decreased 13% for compound 8 (Figure3B). These results indicate that compound 8 is binding in the ACh pocket; however, compound 9 might work as a pore blocker or possibly has a mixed mechanism of action.

Molecules 2019, 24, 446 6 of 17

TableTableTable 1. Structures 1. Structures1. Structures and and corresponding and corresponding corresponding docking docking docking scores scores scores and and in and vitroin vitroin activitiesvitroactivities activities of ofselected selected of selected hits hits and hits and reference and reference reference compounds. compounds. compounds. All All compounds compoundsAll compounds exc exceptept exc references referencesept references galantamine galantamine galantamine and and Table Table1. Structures 1. Structures and corresponding and corresponding docking docking scores scores and in and vitro in activities vitro activities of selected of selected hits and hits reference and reference compounds. compounds. All compounds All compounds except exc referencesept references galantamine galantamine and and TableTable 1. Structures 1. Structures and corresponding and corresponding docking docking scores scores and in and vitro in activitiesvitro activities of selected of selected hits and hits reference and reference compounds. compounds. All compounds All compounds except exc referencesept references galantamine galantamine and and physostigmineandphysostigmine physostigmine were initiallywere were initially initially evaluated evaluated evaluated at 100 atµM. at 100 100 For µM.µ M.comp For For oundscomp compoundsounds exceeding exceeding exceeding 70% inhibition 70% inhibition at the atatAChE thethe AChEAChE and 90% andand at 90% 90% the at atα the7 the nAChR,α α77 nAChR, nAChR, IC50 valuesvalues IC IC505050values valuesvalues werewere were determined.determined.werewere determined. determined.determined. physostigminephysostigmine were initiallywere initially evaluated evaluated at 100 atµM. 100 For µM. comp For oundscompounds exceeding exceeding 70% inhibition 70% inhibition at the atAChE the AChE and 90% and at 90% the atα7 the nAChR, α7 nAChR, IC50 values IC50 values were determined.were determined. physostigminephysostigmine were initiallywere initially evaluated evaluated at 100 atµM. 100 For µM. comp For oundscompounds exceeding exceeding 70% inhibition 70% inhibition at the atAChE the AChE and 90% and at 90% the atα7 the nAChR, α7 nAChR, IC50 values IC50 values were determined.were determined. AChEAChE α7 nAChRα α7 nAChR AChEAChE α α7 nAChRα7 nAChR AChEAChE AChE α7 nAChR7α nAChR7 nAChR AChEAChEAChE 7α nAChR7 nAChRα7 nAChR AChEAChE a a a a α7 nAChRα7 nAChRa a aa AChEAChE aa a a α7 nAChRαa7 nAChRa aa G-ScoreG-ScoreG-Score a a G-ScoreG-ScoreG-Score a a G-ScoreG-ScoreG-Score a a G-ScoreG-ScoreG-Score a a G-ScoreG-Score a a b G-ScoreG-Score a ab G-ScoreG-Score a b a G-ScoreG-Score ab a G-ScoreG-Score b bb G-ScoreG-Score b bb G-ScoreG-Score b bb G-ScoreG-Score b bb NoNo No Structure Structure % Inhibition% Inhibition% Inhibition IC50 b IC5050 % Inhibition% Inhibition% Inhibition b NoNo No StructureStructure %% InhibitionInhibition% Inhibition b % Inhibition% Inhibition% Inhibition b No No StructureStructure % Inhibition% Inhibition b IC50 IC50 % Inhibition% Inhibition b No No StructureStructure % Inhibition% Inhibition b % Inhibition% Inhibition b ICb 50(µM)b 50 IC (µM)b b IC (µM) b b IC (µM) b b No No StructureStructure % Inhibition%(µM) Inhibition (µM)(µM) IC50 IC % ICInhibition50% (µM)(µM) IC50Inhibition5050 (µM)(µM) No No StructureStructure % ICInhibition5050% (µM)(µM) ICInhibition5050 (µM)(µM) % IC50Inhibition50% (µM)(µM) ICInhibition5050 (µM)(µM) (µM) (µM) IC50 (µM)IC50 (µM) IC50 (µM)IC50 (µM) IC50 (µM)IC50 (µM) (pIC50 ± SEM) (pIC50 50 ±50 SEM) (pIC5050± SEM)50 (pIC50 ±50 SEM)50 (pIC(µM)50(pIC (pIC ±± SEM)SEM) (µM)50 50 ±± SEM)SEM) (pICIC50(pIC (pIC (µM)±± ICSEM)SEM)5050 (µM)±± SEM)SEM) (pICIC50(pIC (pIC (µM)±± ICSEM)SEM)5050 (µM)±± SEM)SEM) (pICIC50(pIC (pIC (µM)±± ICSEM)SEM)5050 (µM)±± SEM)SEM) (pIC50(pIC ± SEM)50 ± SEM) (pIC50(pIC ± SEM)50 ± SEM) (pIC50(pIC ± SEM)50 ± SEM) (pIC50(pIC ± SEM)50 ± SEM) (pIC50(pIC ± SEM)50 ± SEM) (pIC50(pIC ± SEM)50 ± SEM) (pIC50(pIC ± SEM)50 ± SEM) (pIC50(pIC ± SEM)50 ± SEM) - - - - −−11.28− 11.28 − −14.33− 14.33 ------−11.28− 11.28 14.33−14.33− 14.33 - cc - cc c - - −11.28− 11.28 −14.33− 14.33 91.5 ± 91.50.3% 91.5± 0.3%IC50± == 0.3%c ICIC5050 == - - -- 38.338.3 ±± 38.34.2% ± 4.2% 95.295.2± 2.0%± 95.22.0% ± 2.0% 5 5 5 91.5 ± 91.50.3% ± c 0.3%IC50 = IC50 = - - 1212 12 38.3 ± 38.34.2% ± 4.2% 95.2 ± 95.22.0% ± 2.0% 5 5 c 50 c 50 12 12 5 5 91.5 ± 91.50.3%0.78 ± IC0.3%IC0.7850 = IC 0.78 = -- - -- 12 12 38.3 ±-- 38.3 4.2% ± -- 4.2% IC95.250IC=50 ± 8.3== 95.22.0% ICIC8.38.35050 ± == 2.0% 8.38.3 5 5 0.78 0.78 - - 12 12 - - IC50 = IC8.350 = 8.3 PhysostigminePhysostigmine (6.130.78 ±(6.13 0.02)(6.13 0.78 ± 0.02)± 0.02) ------(5.08(5.08IC±50 0.04) ±=(5.08 0.04)IC8.350 ±= 0.04)8.3 PhysostigminePhysostigmine (6.13(6.13 ±±(6.13 0.02)0.02) ± 0.02) ------(5.08(5.08 ±±(5.08 0.04)0.04) ± 0.04) PhysostigminePhysostigmine (6.13 ±(6.13 0.02) ± 0.02) - - - - (5.08 ±(5.08 0.04) ± 0.04)

−11.03− 11.03 −15.10− 15.10 −10.30− 10.30 −13.39− 13.39 −11.03− 11.03−11.03 −15.10−15.10− 15.10 −−10.30− 10.30 −13.39−13.39− 13.39 −11.03− 11.03d dd d −15.10− 15.10 −10.30− 10.30 −13.39− e13.39ee ee 91.5 ± 91.50.1% ± 0.1% d 67.3 ± 67.30.5% ± 0.5% 0% 0% 76.3 ± 76.30.5% ± 0.5% e 1 1 91.5 ± 91.50.1%91.5 ± d ± 0.1%0.1% 67.367.3 ± 67.30.5%± 0.5%± 0.5% 13 13 0% 0% 76.376.3± 0.5%± 76.30.5% ± e 0.5% 1 1 1 91.5 ± 91.50.1% ± d 0.1% d 67.3 ± 67.30.5% ± 0.5% 1313 13 0% 0% 76.3 ± 76.30.5% ± e 0.5% e 1 IC50 == IC0.68IC0.68IC5050 50 == 0.680.68= 0.68 IC50IC == 50IC54.8IC54.8=5050 54.8 == 54.854.8 13 -- -- IC50IC=50 13.1== IC13.1IC13.15050 == 13.113.1 1 1 IC50 = IC0.6850 = 0.68 IC50 = IC54.850 = 54.8 13 13 - - IC50 = IC13.150 = 13.1 (6.14IC50 =±(6.14(6.14 IC0.680.01)(6.1450 =±± 0.680.01)±0.01)0.01) (4.26IC50(4.26 =±(4.26(4.26 IC54.80.04)±50 0.04)=±± 54.80.04)0.04) --- -- (4.88(4.88IC±50 0.02)=±(4.88(4.88 IC13.10.02)50 =±± 13.10.02)0.02) (6.14 ±(6.14 0.01) ± 0.01) (4.26 ±(4.26 0.04) ± 0.04) - - (4.88 ±(4.88 0.02) ± 0.02) (6.14 ±(6.14 0.01) ± 0.01) (4.26 ±(4.26 0.04) ± 0.04) NefopamNefopam - - (4.88 ±(4.88 0.02) ± 0.02) NefopamNefopam NefopamNefopam −11.51− 11.51 −15.10− 15.10 −12.07− 12.07 −13.87− 13.87 −11.51− 11.51−11.51 −15.10−15.10− 15.10 −−12.07− 12.07 −13.87−13.87− 13.87 96.4−11.51 ± 96.40.3%− 11.51 ± 0.3% 45.4−15.10 ± 45.42.3%− 15.10 ± 2.3% 6.5− 12.07± 1.7%6.5− 12.07± 1.7% 87.4−13.87 ± 87.41.4%− 13.87 ± 1.4% 96.4 ± 96.40.3% ± 0.3%± 45.4 ± 45.42.3%± ± 2.3% 6.5 ±± 1.7%6.5 ± 1.7% 87.4± ± 87.41.4% ± 1.4% 6 6 96.4 ± 96.40.3%96.4 ± 0.3%0.3% 45.445.4 ± 45.42.3% 2.3%± 2.3% 14 14 6.56.5 ± 1.7%1.7%6.5 ± 1.7% 87.487.4 1.4%± 87.41.4% ± 1.4% 6 6 6 IC50 == IC0.28IC0.285050 == 0.280.28 - -- 1414 14 ------6 6 IC50 = 0.28ICIC50 50= 0.28= 0.28 - - - 14 14 - - - - - (6.55IC50 =±(6.55(6.55 IC0.280.02)50 =±± 0.280.02)0.02) ------(6.55 ±(6.55 0.02)(6.55 ± ±0.02)0.02) ------(6.55 ±(6.55 0.02) ± 0.02) ------−10.28− 10.28 −13.05− 13.05 −11.88− 11.88 −13.03− 13.03 −10.28− 10.28 −13.05− 13.05 −11.88− 11.88 −13.03− 13.03 96.5−10.28 ± 96.50.1%− 10.28 −± 10.280.1% 64.8−13.05 ±− 64.80.7%13.05− 13.05 ± 0.7% 30.3−−11.88 ± 30.34.7%− 11.88 ± 4.7% −75.813.03−13.03 ± 75.80.4%− 13.03 ± 0.4% 96.5 ± 96.50.1% ± 0.1% 64.8 ± 64.80.7% ± 0.7% 30.3 ± 30.34.7% ± 4.7% 75.8 ± 75.80.4% ± 0.4% 7 7 96.5 ± 96.50.1% ± 0.1% 64.8 ± 64.80.7% ± 0.7% 15 15 30.3 ± 30.34.7% ± 4.7% 75.8 ± 75.80.4% ± 0.4% 7 7 IC50 == IC0.23IC0.2396.55050 == ±0.230.230.1% 64.8- ± 0.7%-- 15 15 30.3 ±- 4.7%-- 75.8 ± 0.4%- -- 7 7 7 IC50 = IC0.2350 = 0.23 - - 1515 15 - - - - (6.13IC50 =±(6.13(6.13 IC0.230.02)IC50 50=±± 0.230.02)0.02)= 0.23 ------(6.13 ±(6.13 0.02) ± 0.02) ------(6.13 ±(6.13 0.02)(6.13 ± 0.02)± 0.02) ------

−13.99− 13.99 −13.03− 13.03 −10.64− 10.64 −13.11− 13.11 −13.99− 13.99 −13.03− 13.03 −10.64− 10.64 −13.11− 13.11 78.5−13.99 ± 78.50.6%− 13.99 ± 0.6% 81.8−13.03 ± 81.81.3%− 13.03 ± 1.3% 56.7−10.64 ± 56.74.2%− 10.64 ± 4.2% 79.0−13.11 ± 79.01.6%− 13.11 ± 1.6% 8 8 78.5 ± 78.50.6% ± 0.6% 81.8 ± 81.81.3% ± 1.3% 16 16 56.7 ± 56.74.2% ± 4.2% 79.0 ± 79.01.6% ± 1.6% 8 8 78.5IC50 ±== 78.5 0.6%IC10.6IC10.65050 ± == 0.6%10.610.6 81.8IC50 ±== 81.8 1.3%IC34.3IC34.35050 ± == 1.3%34.334.3 16 16 56.7 ±- 56.7 4.2% ± -- 4.2% 79.0 ±- 79.0 1.6% ± -- 1.6% 8 8 IC50 = IC10.650 = 10.6 IC50 = IC34.350 = 34.3 16 16 - - - - (4.96IC50 =±(4.96 IC10.60.03)50 =± 10.60.03) (4.46IC50 =±(4.46 IC34.30.05)50 =± 34.30.05) - - - - (4.96(4.96 ±±(4.96 0.03)0.03) ± 0.03) (4.46(4.46 ±±(4.46 0.05)0.05) ± 0.05) ------(4.96 ±(4.96 0.03) ± 0.03) (4.46 ±(4.46 0.05) ± 0.05) - - - -

Table Table1. Structures 1. Structures and corresponding and corresponding docking docking scores scores and in and vitro in activities vitro activities of selected of selected hits and hits reference and reference compounds. compounds. All compounds All compounds except exc referencesept references galantamine galantamine and and

physostigminephysostigmine were initiallywere initially evaluated evaluated at 100 atµM. 100 For µM. comp For oundscompounds exceeding exceeding 70% inhibition 70% inhibition at the atAChE the AChE and 90% and at 90% the atα7 the nAChR, α7 nAChR, IC50 values IC50 values were determined.were determined.

AChEAChE α7 nAChRα7 nAChR AChEAChE α7 nAChRα7 nAChR G-ScoreG-Score a a G-ScoreG-Score a a G-ScoreG-Score a a G-ScoreG-Score a a No No StructureStructure % Inhibition% Inhibition b IC50 b IC50 % Inhibition% Inhibition b b No No StructureStructure % Inhibition% Inhibition b b % Inhibition% Inhibition b b (µM) (µM) IC50 (µM)IC50 (µM) IC50 (µM)IC50 (µM) IC50 (µM)IC50 (µM) (pIC50(pIC ± SEM)50 ± SEM) (pIC50(pIC ± SEM)50 ± SEM) (pIC50(pIC ± SEM)50 ± SEM) (pIC50(pIC ± SEM)50 ± SEM)

- - - - −11.28− 11.28 −14.33− 14.33 91.5 ± 91.50.3% ± c 0.3%IC50 = c IC50 = - - 38.3 ± 38.34.2% ± 4.2% 95.2 ± 95.22.0% ± 2.0% 5 5 12 12 0.78 0.78 - - - - IC50 = IC8.350 = 8.3 PhysostigminePhysostigmine (6.13 ±(6.13 0.02) ± 0.02) - - - - (5.08 ±(5.08 0.04) ± 0.04)

−11.03− 11.03 −15.10− 15.10 −10.30− 10.30 −13.39− 13.39 91.5 ± 91.50.1% ± d 0.1% d 67.3 ± 67.30.5% ± 0.5% 0% 0% 76.3 ± 76.30.5% ± e 0.5% e 1 1 13 13 IC50 = IC0.6850 = 0.68 IC50 = IC54.850 = 54.8 - - IC50 = IC13.150 = 13.1 (6.14 ±(6.14 0.01) ± 0.01) (4.26 ±(4.26 0.04) ± 0.04) - - (4.88 ±(4.88 0.02) ± 0.02) NefopamNefopam Molecules 2019, 24, 446 7 of 17 −11.51− 11.51 −15.10− 15.10 −12.07− 12.07 −13.87− 13.87 96.4 ± 96.40.3% ± 0.3% 45.4 ± 45.42.3% ± 2.3% 6.5 ± 1.7%6.5 ± 1.7% 87.4 ± 87.41.4% ± 1.4% 6 6 14 14 IC50 = 0.28IC50 = 0.28 - - Table 1. Cont. - - - - (6.55 ±(6.55 0.02) ± 0.02) ------

AChE α7 nAChR AChE α7 nAChR −10.28−G-Score 10.28 a −13.05G-Score− 13.05a G-Score−11.88− a11.88 G-Score−13.03a − 13.03 No Structure 96.5 ± %96.50.1% Inhibition ± 0.1% b 64.8% Inhibition± 64.80.7% ± 0.7%b No Structure %30.3 Inhibition ± 30.34.7% ± b 4.7% % Inhibition75.8 ± 75.80.4%b ± 0.4% 7 7 15 15 IC50 = IC0.23IC50 =50 0.23(µM) IC- 50 (µ-M) IC50 -(µ M) - IC50 (µ-M) - (6.13 ±(pIC(6.13 0.02)50 ± 0.02)± SEM) (pIC-50 ± SEM)- (pIC50 ±- SEM)- (pIC50 ± SEM)- -

−13.99− 13.99−13.99 −13.03−13.03− 13.03 −−10.64− 10.64 −13.11−13.11− 13.11 78.5 ± 78.50.6%78.5 ± ±0.6%0.6% 81.881.8 ± 81.81.3%± 1.3%± 1.3% 56.756.7± ± 56.74.2% ± 4.2% 79.079.0± 1.6%± 79.01.6% ± 1.6% 8 8 8 1616 16 IC50 = IC10.6IC50 50 = 10.6= 10.6 IC50IC = 50IC34.3=50 34.3 = 34.3 ------MoleculesMolecules 2019, 24 2019, x FOR, 24, xPEER FOR REVIEWPEER REVIEW (4.96(4.96 ± ±0.03)0.03) (4.46(4.46± 0.05)± 0.05) - - - -8 of 188 of 18 MoleculesMolecules 2019, 24 2019, x FOR, 24, xPEER FOR REVIEWPEER REVIEW (4.96 ± 0.03) (4.46 ± 0.05) - - 8 of 18 8 of 18 Molecules 2019, 24, x FOR PEER REVIEW 8 of 18

−10.71− 10.71−10.71 −13.11−13.11− 13.11 −−12.92− 12.92 −14.94−14.94− 14.94 −10.71− 10.71 −13.11− 13.11 −12.92− 12.92 −14.94− 14.94 88.1 ± 88.10.8%88.1 ± 0.8%± 0.8% 93.293.2 ± 93.20.3%±−13.11 0.3%± 0.3% 23.923.9± ± 23.95.7%−12.92 ± 5.7% 84.884.8± 0.7%± 84.80.7%−14.94 ± 0.7% 9 9 9 88.1 ± 88.10.8% ± 0.8% 93.2 ± 93.20.3% ± 0.3% 1717 17 23.9 ± 23.95.7% ± 5.7% 84.8 ± 84.80.7% ± 0.7% 9 9 IC50 = IC5.04IC50 50 = 5.04= 5.04 IC50IC = 50IC14.5=50 14.5= 14.5 17 17 ------IC50 = IC5.0450 = 5.04 IC50 = IC14.550 = 14.5 - - - - (5.29 ±(5.29 IC0.03)(5.2950 =± 0.03)5.04± 0.03) (4.84(4.84 ±(4.84 IC0.04)±50 0.04)=± 0.04)14.5 ------(5.29 ±(5.29 0.03) ± 0.03) (4.84 ±(4.84 0.04) ± 0.04) - - - -

−10.35− 10.35 −13.27− 13.27 −11.23− 11.23 −15.05− 15.05 −10.35− 10.35−10.35 −13.27−13.27− 13.27 −−11.2311.23− 11.23 − −15.05− 15.05 32.7 ± 32.74.2%−10.35 ± 4.2% 97.2 ± 97.20.5%−13.27 ± 0.5% 18.0 ± 18.08.9%−11.23 ± 8.9% 2.515.05 ± 2.5%2.5− 15.05± 2.5% 10 10 32.7 ± 32.74.2%32.7 ± 4.2%± 4.2% 97.297.2 ± 97.20.5%± 0.5%± 0.5% 18 18 18.018.0 ±± 18.08.9%8.9% ± 8.9% 2.52.5± 2.5%± 2.5%2.5 ± 2.5% 10 10 10 - - IC50 = IC3.950 = 3.9 1818 18 - - - - 10 - - IC50IC = 3.9= 3.9 18 - - - - 50IC5050 == 3.93.9 - - - - - (5.44(5.44 ±(5.44 0.05)± 0.05)± 0.05) ------(5.44 ±(5.44 0.05) ± 0.05) - - - - CyamemazineCyamemazine CyamemazineCyamemazine

−11.29− 11.29 −14.60− 14.60 −11.29− 11.29 −14.60− 14.60 14.0 ± 14.09.3%−11.29 −± 11.299.3% 98.0 ±− 98.00.1%14.60−14.60 ± 0.1% 11 11 14.0 ± 14.09.3%14.0 ± 9.3%± 9.3% 98.098.0 ± 98.00.1%± 0.1%± 0.1% 11 11 11 - - IC50 = IC7.250 = 7.2 - - IC50 = IC7.250 = 7.2 - - - (5.14IC ±(5.14 500.05)IC=50 ± 7.2= 0.05) 7.2 - - - (5.14(5.14 ±(5.14 0.05)± 0.05)± 0.05) FlazaloneFlazalone FlazaloneFlazalone a a Flazaloneb b c c d d e e a Glide GlideG-score G-score (kcal/mol); (kcal/mol); b % inhibition % inhibition at 100 at µM; 100 - µM;indicates - indicates “value “value not determined”; not determined”; c % inhibition % inhibition at 5.12 at µM; 5.12 d µM;% of AChE % of AChE inhibition inhibition at 10 µM;at 10 e µM;inhibition inhibition at 33 µM.at 33 µM. Glideaa GlideG-score G-score (kcal/mol); (kcal/mol); % inhibition b % inhibition at 100 atµM; 100 - µM;indicates - indicates “value “value not determined”; not determined”; % inhibition cc % inhibition at 5.12 at µM; 5.12 µM;% of dAChE % of AChE inhibition inhibition at 10 µM; at 10 µM;inhibition ee inhibition at 33 µM. at 33 µM. Glidea Glide G-score G-score (kcal/mol); (kcal/mol); b%% inhibition inhibition at 100100µ µM;M; - indicates- indicates “value “value not determined”;not determined”;c % inhibition % inhibition at 5.12 µ atM; 5.12d % µM; of AChE % inhibitionof AChE atinhibition 10 µM; e inhibitionat 10 µM; at 33inhibitionµM. at 33 µM.

Molecules 2019, 24, 446 8 of 17

3. Discussion In the present study we explored the use of HTVS as a tool for identification of lead compounds targeting two biological targets, AChE and the α7 nAChRs as inhibitors and PAMs, respectively. A dual-acting compound working at these two targets would be highly interesting in relation to developing drugs for the treatment of dementias where both targets independently have been shown to effectively enhance neurotransmission and improve cognitive functions [28]. HTVS is a valuable method; however, it is not devoid of challenges, and hit rates vary significantly depending on the target and the quality of the crystal structure used [12,29]. Requiring a hit to have activity at two independent targets is even more challenging, and there are only a few such examples in the literature. For identification of drug leads against AD, an interesting approach was taken by Domínguez et al., where compounds with neuroprotective activity previously observed in mice were investigated against two enzymes: AChE and γ-secretase [29]. In another approach, Lepailleur et al. searched an in-house library of presumed GPCR compounds and looked for additional effects on histamine H3 and serotonin 5-HT4 receptors [30]. These dual-target screening projects proved to be successful with hit-rates of 12–33%. We searched for compounds targeting both AChE and the α7 nAChR. These are two structurally and functionally very different biological targets, an enzyme and a ligand gated ion channel. However, through evolution, their binding pockets have become optimised to efficiently recognise the same neurotransmitter, ACh. Both targets contain a characteristic aromatic cassette that efficiently recognises and binds cations, and identification of dual-action compounds towards these two particular targets may, therefore, not be as difficult as could be anticipated. The screening approach presented here proves that finding structures with dual activity at the AChE and the α7 nAChR is indeed possible. Most of the compounds identified had some activity at both targets and two met the cut-off criteria of 70% inhibition at AChE and 90% inhibition at the nAChR to merit full evaluation. Interestingly, there appeared to be no correlation between docking scores and measured IC50 values. Compounds 8 and 9 were identified as dual-target compounds with a balanced potency at AChE/α7 of 10/34 and 5/14 µM, respectively. Interestingly, the two compounds are not galantamine-like structures, and they differ from each other, but both contain a large aromatic portion and a basic tertiary amine. Their docked poses (Figure4A,B), which can be superimposed in a way that overlays the protonated nitrogens and aromatic nitrogen heterocycles (Figure4C), create interactions with Y195 and Q117. This superimposition may define a pharmacophore for compounds targeting both AChE and nAChRs. Of the remaining compounds, an additional two, galantamine analogues 6 and 7, had high potency at the AChE, meaning that 4/13 or 30% of the tested compounds were potent AChE inhibitors. At the α7 nAChR, the individual target HTVS hit rate was higher, 6/13 or 46%, which is not surprising as the VS was biased towards nAChRs in the first place in that only compounds from the top scoring list for nAChR were virtually screened against the AChE. In addition to the two dual action inhibitors 8 and 9, four other compounds (10–13) were identified as potent α7 nAChR antagonists. Their nicotinic activity has not been previously described. Compounds 8, 9, and 12 are new ligands without any previously known biological activities. Compounds 10 and 13 are clinically used drugs. Compound 10 is a well-known antipsychotic drug, cyamemazine, which has anxiolytic properties and is used in the treatment of schizophrenia. It is known to bind to dopamine, serotonin, muscarinic, and histamine receptors [31] but activity at nAChRs has not been reported. The α7 nAChR is considered a target for the treatment of schizophrenia. However, only agonists or PAMs are expected to produce cognition enhancement [32,33]. Compound 13, nefopam, is another clinically used drug that appeared as an α7 antagonist in our screening. It is a centrally-acting, non-opioid analgesic used for the relief of moderate to severe pain. Its mechanism of action is thought to be by inhibition of serotonin, dopamine, and noradrenaline reuptake, but is not well understood [34]. Again, the α7 nAChR is considered as a target in non-opioid pain treatment with interest in compounds positively affecting the receptor [35]. Compound 11, flazalone, is known Molecules 2019, 24, 446 9 of 17

Molecules 2019, 24, x FOR PEER REVIEW 8 of 18 for its anti-inﬂammatory properties, but has no clinical use [36]. For the above clinically used drugs, thedesired antagonism clinical of α effect7 nAChRs of the is drugs, unlikely as in to beall linkedcases, activation to the desired would clinical be required; effect of however, the drugs, blockade as in all cases,of α7 activationnAChRs may would be belinked required; to side however, effects. blockadeDurrieu et of al.α7 [34] nAChRs studied may side be linkedeffects to of side nefopam effects. (13) Durrieuand reported et al. [34] studiedoccurrence side of effects serious, of nefopam unexpected (13) andneuropsychiatric reported occurrence side effects of serious, like unexpectedconfusion and neuropsychiatrichallucinations. side These effects were like confusioncorrelated andto high hallucinations. doses of Thesethe drug, were which correlated is consistent to high doses with of its themicromolar drug, which inhibition is consistent of α with7 nAChRs its micromolar described inhibition here. of α7 nAChRs described here.

FigureFigure 4. Binding 4. Binding mode mode of the of twothe two dual-active dual-active compounds compounds in the in αthe7 nAChRα7 nAChR homology homology models models and and AChEAChE X-ray X-ray structure. structure. Compound Compound8 in the8 inα the7 nAChR α7 nAChR (A) and (A) AChEand AChE (D) binding (D) binding pockets. pockets. Compound Compound9 in the9 inα7 the nAChR α7 nAChR (B) and ( AChEB) and ( EAChE) binding (E) pockets.binding Bothpockets. compounds Both compounds superimposed superimposed in the α7 nAChR in the α7 (C) pocketnAChR showing(C) pocket common showing pharmacophores. common pharmacophores. Compound Compound8 binds to the 8 bindsα7 nAChR to the( Aα7) throughnAChR (A) threethrough hydrogen three bonds hydrogen (yellow bonds dotted (yellow lines) dotted to Q57, lines) Q117, to andQ57, W149; Q117, compoundand W149; 9compoundexhibits similar 9 exhibits interactionssimilar interactions with the same with amino the same acids amino Q57, acids Q117, Q5 and7, Q117, W149. and In theW149. AChE In the pocket, AChE compound pocket, compound8 (D) π binds8 ( throughD) binds hydrogen through hydrogen bonds to Y124,bonds Q202, to Y124, S203, Q202, and S203, H447 and and H447 a cation- and ainteraction cation-π interaction to W86 (blue to W86 π dotted(blue line). dotted Compound line). Compound9 (E) creates 9 hydrogen(E) creates bonds hydrogen to Y72, bonds Y337, to and Y72, R296 Y337, and and a cation- R296 andinteraction a cation-π F 8 9 to Y341.interaction ( ) Structure to Y341. of ( compoundsF) Structure ofand compounds. 8 and 9. 4. Materials and Methods 4. Materials and Methods 4.1. Materials 4.1. Materials Plant origin galantamine hydrobromide analytical standard was purchased from PhytoLab GmbH & Co., KGPlant (Vestenbergsgreuth, origin galantamine Germany). hydrobromide Screened analytical compounds standard were purchased was purchased from Ambinter from PhytoLab (c/o Greenpharma,GmbH & Co. Orl éans,KG France).(Vestenbergsgreuth, Restriction enzymes Germany). were Screened from New compounds England Bio were Labs Inc.purchased (Ipswich, from MA,Ambinter USA), and (c/o DNA Greenpharma, and RNA purification Orléans, France). kits were Restriction from QIAGEN enzymes N.V. were (Venlo, from The New Netherlands). England Bio TheLabs mMessage Inc. (Ipswich, mMachine MA, USA), T7 transcription and DNA and kit RNA was from purification ThermoFisher kits were Scientific from QIAGEN (Waltham, N.V. MA,(Venlo, USA).The Acetylcholinesterase Netherlands). The mMessage from Electrophorus mMachine electricus T7 transcription, acetylcholine kit was chloride, from ThermoFisher acetylthiocholine Scientific iodide,(Waltham, 5,5-dithiobis-(2-nitro-benzoic MA, USA). Acetylcholinesterase acid), bovine from serum Electrophorus albumin, kanamycin, electricus theophylline,, acetylcholine tricaine, chloride, collagenase,acetylthiocholine HEPES, Trizma,iodide, salts,5,5-dithiobis-(2-nitro-benzoic and other chemicals not acid), mentioned bovine specifically serum albumin, were purchased kanamycin, fromtheophylline, Sigma-Aldrich tricaine, Co., LLC collagenase, (St. Louis, MI,HEPES, USA) Trizma, and were salts, of analytical and other grade. chemicals not mentioned specifically were purchased from Sigma-Aldrich Co. LLC (St. Louis, MI, USA) and were of analytical grade.

4.2. Homology Modeling and Preparation of Protein Structures

Molecules 2019, 24, 446 10 of 17

4.2. Homology Modeling and Preparation of Protein Structures The sequence of the human α7 nAChR was downloaded from the UniProt database (entry P36544, The UniProt Consortium, 2011, www.uniprot.org) and the signal peptide (residues 1–22) and the transmembrane domain (residues 231–490) were deleted. Template structures Aplysia californica AChBP with galantamine bound (PDB ID: 2PH9 [21]) and the mus musculus α1 nAChR extracellular subunit (PDB ID: 2QC1 [37]) were downloaded from the Protein Data Bank (www.rcsb.org,[38]). After removal of unwanted ligands and heteroatoms, the sequences were initially aligned in T-Coffee [39], and the alignment was subsequently edited according to the following considerations: both templates were used in areas where the two templates were structurally similar, otherwise, 2QC1 was the template of choice, except for regions where this template was distorted due to introduced mutations and bound antagonists. A detailed alignment highlighting the choice of template for specific regions is shown in Figure5. Homology models with galantamine included in the interfacial binding sites were constructed with MODELLER v.9.9 (Sali-Lab, San-Francisco, CA, USA) [40]. One hundred models were generated and sorted according to DOPE scores [41]. After visual inspection, the model with the most favourable DOPE score was selected and modified as follows: two water molecules (HOH 314 and 315) and a third bridging water molecule (HOH 83) were copied across from the 2PH9 and 2QC1 templates, respectively. Subsequently, the model was energy minimized in Macro Model (Schrödinger Release 2012-1: Macro Model, Schrödinger, LLC, New York, NY, USA, 2017) using the OPLS_2005 force field [42] and the GB/SA water solvation model, after preparation of the structure using the Protein Preparation workflow in Schrodinger’s MAESTRO where hydrogen atoms were added and oriented in an energetically favourable manner followed by protonation/deprotonation of acidic/basic residues and a constrained energy minimization [43]. The water molecules were deemed important to include to block part of the classical agonist pharmacophore to preclude prototypical nicotine-like agonists from the top of the scoring list, which were not of interest in this study. Finally, given that the experimentally determined binding mode of galantamine in the X-ray structure is ambiguous [21], a series of 20 docking runs using the Schrodinger Induced Fit Docking Protocol [44,45] were performed in which protein side chains were sampled and backbone atoms were allowed to adapt in vicinity (10 Å) of the docked ligand. After confirming that no outliers in the Ramachandran plot were present in the vicinity of the binding sites, two top ranked models that differed only by the rotameric state of Q57 were selected as targets for docking. In these structures, galantamine forms cation-π interactions with residues on the principle side of the interface (Y93, Y188, Y195, and W149) and a hydrogen bond to Y188. On the complementary side, galantamine is anchored via a π-π interaction with W55 and hydrogen bonds to a water molecule, Q117, and when in the rotameric state facing the binding pocket, also to Q57 (Figure6). For VS in the AChE enzyme, an X-ray structure of human recombinant AChE co-crystalized with galantamine (PDB ID: 4EY6 [20]) was downloaded from the PDB database and prepared for virtual screening as follows: water molecules were removed except HOH 860, which bridges from galantamine to the receptor (S203 and G122). The X-ray structure was subsequently prepared using the Protein Preparation workflow [43]. Molecules 2019, 24, x FOR PEER REVIEW 8 of 18 Molecules 2019, 24, 446 11 of 17 Molecules 2019, 24, x FOR PEER REVIEW 8 of 18

Figure 5. Sequence alignment. Sequence of the human α7 nAChR and templates, AChBP co- crystalized with galantamine (PDB ID: 2PH9), and rat α1 subunit (PDB ID: 2QC1) were aligned using FigureFigure 5.5.Sequence Sequence alignment. alignment. Sequence Sequence of the of humanthe humanα7 nAChR α7 nAChR and templates, and templates, AChBP co-crystalized AChBP co- T-Coffee and used to build the homology model. Residues from both templates used in homology withcrystalized galantamine with galantamine (PDB ID: 2PH9), (PDB and ID: rat 2PH9),α1 subunit and rat (PDB α1 subunit ID: 2QC1) (PDB were ID: aligned2QC1) were using aligned T-Coffee using and modelling are shown in bold. Conserved residues are marked with grey boxes. The sequence usedT-Coffee to build and theused homology to build the model. homology Residues model. from Re bothsidues templates from both used templa in homologytes used modelling in homology are identities counting only the parts of the templates that were used for modelling constitute 30 and 39%, shownmodelling in bold. are Conservedshown in residuesbold. Conserved are marked residues with grey are boxes. marked The sequencewith grey identities boxes. The counting sequence only respectively, for 2PH9 and 2QC1. For the human α7 nAChR, the sequence was used without the signal theidentities parts of counting the templates only the that parts were of the used templates for modelling that were constitute used for 30 modelling and 39%, respectively,constitute 30 forand 2PH9 39%, peptide. andrespectively, 2QC1. For for the 2PH9 human and α2QC1.7 nAChR, For the the human sequence α7 nAChR, was used the without sequence the was signal used peptide. without the signal peptide.

FigureFigure 6.6. InteractionsInteractions ofof galantaminegalantamine (orange)(orange) inin thethe bindingbinding pocketpocket ofof homologyhomology modelsmodels ofof thethe αα77 nAChR.nAChR. InteractingInteracting residues residues are are shown shown as as purple purple sticks. sticks. Q57 Q57 in in alternative alternative rotameric rotameric state state is shownis shown in Figure 6. Interactions of galantamine (orange) in the binding pocket of homology models of the α7 magenta.in magenta. Galantamine Galantamine creates creates three three hydrogen hydrogen bonds bonds with with Q57, Q57, Q117, Q117, and Y188 and (yellowY188 (yellow dotted dotted lines) nAChR. Interacting residues are shown as purple sticks. Q57 in alternative rotameric state is shown alonglines) withalong several with several aromatic aromatic contacts contacts with the with principal the principal receptors: receptors: Y93, W149, Y93, W149, Y188, andY188, Y195 and (blueY195 in magenta. Galantamine creates three hydrogen bonds with Q57, Q117, and Y188 (yellow dotted dotted(blue dotted lines) andlines) to and W55 to on W55 the on complementary the complementary side (grey side dotted(grey dotted line). line). lines) along with several aromatic contacts with the principal receptors: Y93, W149, Y188, and Y195 4.3. Preparation(blue dotted of lines) Databases and to W55 on the complementary side (grey dotted line). 4.3. Preparation of Databases A subset of the ZINC database (ZINC version 12, subset Znp98) containing 210,000 natural 4.3. PreparationA subset ofof Databasesthe ZINC database (ZINC version 12, subset Znp98) containing 210,000 natural products and natural product derivatives was downloaded and merged with an in-house database products and natural product derivatives was downloaded and merged with an in-house database containingA subset lycopodium of the ZINC alkaloid database structures (ZINC (~250version compounds, 12, subset oneZnp98) low containing energy conformation 210,000 natural per containing lycopodium alkaloid structures (~250 compounds, one low energy conformation per compound).products and Allnatural structures product were derivatives protonated was according downloaded to pHand 7.4. merged After with calculation an in-house of molecular database compound). All structures were protonated according to pH 7.4. After calculation of molecular propertiescontaining with lycopodium QikProp (Schrödingeralkaloid structures Release (~250 2018-4: compounds, QikProp, New one York, low NY,energy USA) conformation [46], compounds per properties with QikProp (Schrödinger Release 2018-4: QikProp, New York, NY, USA) [46], compound). All structures were protonated according to pH 7.4. After calculation of molecular properties with QikProp (Schrödinger Release 2018-4: QikProp, New York, NY, USA) [46],

Molecules 2019, 24, x FOR PEER REVIEW 8 of 18 compounds that were not compliant with Lipinski’s rule of five [47], contained reactive or toxic groups, or contained more than one formal charge were discarded. Post filtering, 87,250 compounds remained for high throughput virtual screening.

Molecules4.4. HTVS2019 and, 24 ,Hits 446 Selection 12 of 17 HTVS was performed on the two α7 nAChR homology models following the Schrödinger VS thatprotocol were (Figure not compliant 7). Briefly, with thre Lipinski’se levels rule of ofscreening five [47], were contained performed reactive (HTVS-, or toxic Standard-, groups, or and contained Extra morePrecision than docking) one formal with charge GLIDE were (version discarded. 5.9, Schrödinger, Post filtering, New 87,250 York, compoundsNY, USA, 2013) remained [48,49]. for Default high throughputsettings were virtual used, screening.except the number of compounds that were carried forward between stages was 20% instead of the standard 10%. From the last stage of the nAChR screening, the top 50% of 4.4.compounds HTVS and from Hits each Selection homology model (1714 compounds per model) were kept and subsequently filtered, applying a 5 kcal/mol conformational energy penalty cut-off and leaving 1018 and 1031 HTVS was performed on the two α7 nAChR homology models following the Schrödinger VS compounds from the two models, respectively. Then, 1607 unique structures were retrieved from the protocol (Figure7). Briefly, three levels of screening were performed (HTVS-, Standard-, and Extra original database and docked to the AChE crystal structure using Extra-Precision (XP) GLIDE Precision docking) with GLIDE (version 5.9, Schrödinger, New York, NY, USA, 2013) [48,49]. Default docking and filtered by applying the 5 kcal/mol conformational energy penalty cut-off. settings were used, except the number of compounds that were carried forward between stages was 20% The ranges of docking scores were broad, therefore, hits with docking scores worse than −10 instead of the standard 10%. From the last stage of the nAChR screening, the top 50% of compounds kcal/mol for AChE and −13 kcal/mol for the nAChR were eliminated. Further, a 400 g/mol molecular from each homology model (1714 compounds per model) were kept and subsequently filtered, applying weight cut-off filter was applied, leading to 78 compounds that were clustered in Canvas [50] based a 5 kcal/mol conformational energy penalty cut-off and leaving 1018 and 1031 compounds from the on the Tanimoto score using Daylight’s fingerprints into eight clusters. After visual inspection, 13 two models, respectively. Then, 1607 unique structures were retrieved from the original database compounds, at least one from each cluster, were selected and purchased. Purity of purchased and docked to the AChE crystal structure using Extra-Precision (XP) GLIDE docking and filtered by compounds was >90% as stated by the supplier. The chemical structure and docking scores for the applying the 5 kcal/mol conformational energy penalty cut-off. selected compounds and reference compounds are shown in Table 1.

Figure 7. Screening workflow. Virtual screening was performed at two α7 nAChR homology models Figure 7. Screening workflow. Virtual screening was performed at two α7 nAChR homology models having different volumes of the binding pocket. Top scoring hits from both models were subsequently having different volumes of the binding pocket. Top scoring hits from both models were subsequently screened at the X-ray structure of AChE (PDB ID: 4EY6). From hits scoring high at both targets, screened at the X-ray structure of AChE (PDB ID: 4EY6). From hits scoring high at both targets, 13 13 compounds were selected and tested in vitro. HTVS = high throughput virtual screening. compounds were selected and tested in vitro. HTVS = high throughput virtual screening. The ranges of docking scores were broad, therefore, hits with docking scores worse than 4.5. Validation of Activity at nAChRs −10 kcal/mol for AChE and −13 kcal/mol for the nAChR were eliminated. Further, a 400 g/mol molecular weight cut-off filter was applied, leading to 78 compounds that were clustered in Canvas [50] 4.5.1. Molecular Biology based on the Tanimoto score using Daylight’s fingerprints into eight clusters. After visual inspection, 13 compounds,Human α7 atnACh least onereceptor from subunits each cluster, were were cloned selected and andinserted purchased. into expression Purity of purchasedvectors as compoundsdescribed previously was >90% [16]. as statedPlasmid by cDNAs the supplier. were linearized The chemical using structure a downstream and docking Not I restriction scores for site the selectedand purified. compounds cRNA andwas reference prepared compounds and capped are from shown the in linearized Table1. cDNA using the mMessage mMachine T7 transcription kit according to the manufacturer’s protocol. Purified cRNA was 4.5.aliquoted Validation and ofstored Activity at a at concentration nAChRs of 0.5 µg·µL−1 at −80 °C until further use.

4.5.1. Molecular Biology 4.5.2. Expression of nAChR in Xenopus laevis Oocytes Human α7 nACh receptor subunits were cloned and inserted into expression vectors as described Xenopus laevis oocytes were obtained as described previously [51], briefly, ovary lobes were previously [16]. Plasmid cDNAs were linearized using a downstream Not I restriction site and removed by surgical incision, sliced into small pieces, and defolliculated by collagenase treatment. puriﬁed. cRNA was prepared and capped from the linearized cDNA using the mMessage mMachine T7 transcription kit according to the manufacturer’s protocol. Puriﬁed cRNA was aliquoted and stored at a concentration of 0.5 µg·µL−1 at −80 ◦C until further use. Molecules 2019, 24, 446 13 of 17

4.5.2. Expression of nAChR in Xenopus laevis Oocytes Xenopus laevis oocytes were obtained as described previously [51], brieﬂy, ovary lobes were removed by surgical incision, sliced into small pieces, and defolliculated by collagenase treatment. The protocol for this speciﬁc study was approved by the Animal Ethics Committee of the University of Sydney (Protocol number: 2013/5915). Stage V and VI oocytes were injected with a total of ~25 ng of cRNA encoding human α7 nACh receptor with RIC3 (in 5:1 ratio), a protein enhancing the expression of the receptor. Injected oocytes were incubated for 2–5 days at 18 ◦C in a saline solution (96 mM NaCl, 2 mM KCl, 1 mM MgCl2, 1.8 mM CaCl2, 5 mM HEPES (hemisodium, pH 7.4)) supplemented with 2.5 mM sodium pyruvate, 0.5 mM theophylline, and 50 µM kanamycin.

4.5.3. Oocyte Electrophysiology ACh and galantamine were initially dissolved in milliQ water as 10 mM stock solutions. Screened compounds were dissolved as a 10 mM stock solution in DMSO. The maximal DMSO concentration in the final dilution did not exceed 1%. This DMSO concentration did not evoke any current from the receptors. Compound dilutions were prepared in a saline solution on the day of the experiment. Electrophysiological recordings from Xenopus laevis oocytes were performed using the two-electrode voltage-clamp technique as described previously [23,51]. Briefly, oocytes were placed in a custom-built recording chamber and continuously perfused with a saline solution. The saline solution contained 115 mM NaCl, 2.5 mM KCl, 1.8 mM BaCl2, 10 mM HEPES, and was adjusted to pH 7.4 with NaOH. Pipettes were backfilled with 3 M KCl, and open-pipette resistances ranged from 0.3–1.5 MΩ when submerged in the saline solution. Oocytes were voltage clamped at a holding potential of −60 mV unless otherwise stated using an Axon Geneclamp 500B amplifier (Molecular Devices, San Jose, CA, USA). Rapid solution exchange in the oocyte vicinity (order of a few seconds) was ensured by application through a 1.5 mm diameter capillary tube placed approximately 2 mm from the oocyte as described previously [51]. Solution flow rate through the capillary was 2.0 mL·min−1. Experiments were performed as follows: nAChR currents were initially evoked with three AChcontrol applications (~EC20, 30 µM), a maximum efficacious concentration of ACh (EC100, 3 mM), followed by three additional AChcontrol applications. Thereafter, test compounds in increasing concentration were applied following a pre-incubation protocol. This involved pre-incubation with the test compound for ~25 s followed by co-application of the test compound with AChcontrol for 20 s and a wash period of at least 2 min before the next application. Initially, the effect of application of 100 µM of a test compound was evaluated, and for the most potent compounds (inhibition at 100 µM > 90%), full concentration response relationships were determined using six concentrations of test compounds from 0.4–100 µM. A detailed description of the protocol was published recently [23]. For experiments evaluating the magnitude of the response as a function of membrane potential, experiments were conducted at two holding potentials: −50 and −100 mV. These experiments also involved a pre-incubation protocol where, first, oocytes were pre-incubated with the test solution (25 s), and secondly, the test solution was applied together with 30 µM ACh (20 s). Galantamine was tested at 30 µM, all other compounds at 10 µM. Concentrations were selected based on the potency of compounds. Peak current amplitudes were normalized with respect to the amplitude of current elicited by 30 µM ACh. All experiments were conducted in triplicate using oocytes from at least two batches from different frogs.

4.6. Validation of Activity at AChE In vitro AChE inhibitory activity was studied using the colorimetric method of Ellman [52] using the AChE enzyme from Electrophorus electricus. To a 96-well microplate, test solutions were applied along with 0.2 mg/mL bovine serum albumin, 1 mM 5,5-dithiobis-(2-nitro-benzoic acid), and 0.05 mM acetylthiocholine iodide. The reaction was initiated by addition of 0.20 units/mL AChE enzyme and followed by colorimetric detection performed at 405 nm. Experiments were conducted in triplicate. Molecules 2019, 24, 446 14 of 17

All compounds were dissolved in methanol (maximum 2% methanol at assay conditions that did not affect the enzyme activity) and screened at 100 µM with physostigmine (5) as a positive control. For compounds exhibiting more than 70% inhibition of ACh degradation, IC50 values were determined using six concentrations within the 20–90% inhibition range.

4.7. Data Analysis Electrophysiological data were analysed using pClamp 10.2 (Molecular Devices, San Jose, CA, USA). During analysis, traces were baseline subtracted and responses to individual applications quantified as peak-current amplitudes. All fitting and statistical calculations were performed using GraphPad Prism 7 (GraphPad, San Diego, CA, USA). A monophasic Hill equation was used in all the non-linear regression calculations. For evaluation of the inhibitory activity, the percentage of remaining peak-current amplitudes relative to that of the AChcontrol application was calculated. Data were then fitted with the slope set to 1 and the remaining current amplitude at infinitely high compound concentrations set to 0. AChE inhibition data were analysed using GraphPad Prism 7. Absorption readings from the AChE inhibition assay were plotted versus time and linear regression was performed. From the obtained slopes, percentages of inhibition were calculated, and IC50 values were determined from non-linear regression calculations.

5. Conclusions We have shown that HTVS can successfully be used to identify compounds with activity at both AChE and the α7 nAChR. Hit rates obtained at both targets signiﬁcantly exceeded those generally reported in the literature. The choice of galantamine as a lead compound has biased the screening towards antagonists, since galantamine during the course of the project was shown to be an antagonist rather than a PAM. However, ﬁnding compounds interacting with two separate targets presents a success in itself and an achievement from a virtual screening technology perspective. The current project can serve as an encouragement for further, more detailed study of the required interactions and more sophisticated designs of HTVS projects providing dual active structures positively affecting the α7 nAChR and inhibiting AChE.

Author Contributions: N.M.K., T.B., and P.K.A. designed the studies. N.M.K. and D.C.I. performed the experiments. N.M.K. and T.B. drafted the manuscript, and all authors contributed to the final version of the manuscript. E.S.O., M.C., and T.B. acquired funding and performed supervision of the experiments. All authors approved the final version of the manuscript. Funding: This research was supported by the Australian National Health and Medical Research Council [NHMRC grant number APP1069417], The Australian Research Council [ARC grant number LP140100781], a doctoral grant and financial support from the University of Iceland, a grant from the Bergthóru and Thorsteins Scheving Thorsteinsson Fund, and the Icelandic Research Fund [grant number: 152604051]. Conflicts of Interest: The authors declare no conflict of interest.

References

1. Reiman, E.M.; Quiroz, Y.T.; Fleisher, A.S.; Chen, K.; Velez-Pardo, C.; Jimenez-Del-Rio, M.; Fagan, A.M.; Shah, A.R.; Alvarez, S.; Arbelaez, A.; et al. Brain imaging and ﬂuid biomarker analysis in young adults at genetic risk for autosomal dominant Alzheimer’s disease in the presenilin 1 E280A kindred: A case-control study. Lancet Neurol. 2012, 11, 1048–1056. [CrossRef] 2. Grossberg, G.T.; Edwards, K.R.; Zhao, Q. Rationale for combination therapy with galantamine and memantine in Alzheimer’s disease. J. Clin. Pharm. 2006, 46, 17s–26s. [CrossRef] 3. Zhao, X.; Marszalec, W.; Toth, P.T.; Huang, J.; Yeh, J.Z.; Narahashi, T. In vitro galantamine-memantine co-application: Mechanism of beneﬁcial action. Neuropharmacology 2006, 51, 1181–1191. [CrossRef] 4. Lopes, J.P.; Tarozzo, G.; Reggiani, A.; Piomelli, D.; Cavalli, A. Galantamine potentiates the neuroprotective effect of memantine against NMDA-induced excitotoxicity. Brain Behav. 2013, 3, 67–74. [CrossRef] Molecules 2019, 24, 446 15 of 17

5. Atri, A.; Molinuevo, J.L.; Lemming, O.; Wirth, Y.; Pulte, I.; Wilkinson, D. Memantine in patients with Alzheimer’s disease receiving donepezil: New analyses of efficacy and safety for combination therapy. Alzheimer’s Res. 2013, 5, 6. [CrossRef] 6. Grossberg, G.T.; Manes, F.; Allegri, R.F.; Gutierrez-Robledo, L.M.; Gloger, S.; Xie, L.; Jia, X.D.; Pejovic, V.; Miller, M.L.; Perhach, J.L.; et al. The safety, tolerability, and efficacy of once-daily memantine (28 mg): A multinational, randomized, double-blind, placebo-controlled trial in patients with moderate-to-severe Alzheimer’s disease taking cholinesterase inhibitors. CNS Drugs 2013, 27, 469–478. [CrossRef][PubMed] 7. Reddy, A.S.; Zhang, S. Polypharmacology: Drug discovery for the future. Exp. Rev. Clin. Pharm. 2013, 6. [CrossRef] 8. Hughes, R.E.; Nikolic, K.; Ramsay, R.R. One for all? Hitting multiple Alzheimer’s Disease targets with one drug. Front. Neurosci. 2016, 10, 177. [CrossRef][PubMed] 9. Nikolic, K.; Mavridis, L.; Djikic, T.; Vucicevic, J.; Agbaba, D.; Yelekci, K.; Mitchell, J.B.O. Drug design for cns diseases: polypharmacological profiling of compounds using cheminformatic, 3D-QSAR and Virtual screening methodologies. Front. Neurosci. 2016, 10.[CrossRef] 10. Zhu, T.; Cao, S.; Su, P.C.; Patel, R.; Shah, D.; Chokshi, H.B.; Szukala, R.; Johnson, M.E.; Hevener, K.E. Hit identification and optimization in virtual screening: Practical recommendations based on a critical literature analysis. J. Med. Chem. 2013, 56, 6560–6572. [CrossRef][PubMed] 11. Moser, D.; Wisniewska, J.M.; Hahn, S.; Achenbach, J.; Buscató, E.L.; Klingler, F.-M.; Hofmann, B.; Steinhilber, D.; Proschak, E. Dual-target virtual screening by pharmacophore elucidation and molecular shape filtering. ACS Med. Chem. Lett. 2012, 3, 155–158. [CrossRef] 12. Zhou, S.; Li, Y.; Hou, T. Feasibility of using molecular docking-based virtual screening for searching dual target kinase inhibitors. J. Chem. Inf. Model. 2013, 53, 982–996. [CrossRef] 13. Sadigh-Eteghad, S.; Talebi, M.; Mahmoudi, J.; Babri, S.; Shanehbandi, D. Selective activation of alpha7 nicotinic acetylcholine receptor by PHA-543613 improves Abeta25-35-mediated cognitive deficits in mice. Neuroscience 2015, 298, 81–93. [CrossRef] 14. Gotti, C.; Zoli, M.; Clementi, F. Brain nicotinic acetylcholine receptors: Native subtypes and their relevance. Trends Pharm. Sci. 2006, 27, 482–491. [CrossRef] 15. Rubboli, F.; Court, J.A.; Sala, C.; Morris, C.; Chini, B.; Perry, E.; Clementi, F. Distribution of Nicotinic Receptors in the Human Hippocampus and Thalamus. Eur. J. Neurosci. 1994, 6, 1596–1604. [CrossRef] 16. Timmermann, D.B.; Gronlien, J.H.; Kohlhaas, K.L.; Nielsen, E.O.; Dam, E.; Jorgensen, T.D.; Ahring, P.K.; Peters, D.; Holst, D.; Christensen, J.K.; et al. An allosteric modulator of the α7 nicotinic acetylcholine receptor possessing cognition-enhancing properties in vivo. J. Pharm. Exper. 2007, 323, 294–307. [CrossRef] 17. Prickaerts, J.; van Goethem, N.P.; Chesworth, R.; Shapiro, G.; Boess, F.G.; Methfessel, C.; Reneerkens, O.A.H.; Flood, D.G.; Hilt, D.; Gawryl, M.; et al. EVP-6124, a novel and selective α7 nicotinic acetylcholine receptor partial agonist, improves memory performance by potentiating the acetylcholine response of α7 nicotinic acetylcholine receptors. Neuropharmacology 2012, 62, 1099–1110. [CrossRef] 18. Samochocki, M.; Hoffle, A.; Fehrenbacher, A.; Jostock, R.; Ludwig, J.; Christner, C.; Radina, M.; Zerlin, M.; Ullmer, C.; Pereira, E.F.; et al. Galantamine is an allosterically potentiating ligand of neuronal nicotinic but not of muscarinic acetylcholine receptors. J. Pharm. Exper. 2003, 305, 1024–1036. [CrossRef] 19. Texido, L.; Ros, E.; Martin-Satue, M.; Lopez, S.; Aleu, J.; Marsal, J.; Solsona, C. Effect of galantamine on the human alpha7 neuronal nicotinic acetylcholine receptor, the Torpedo nicotinic acetylcholine receptor and spontaneous cholinergic synaptic activity. Br. J. Pharm. 2005, 145, 672–678. [CrossRef] 20. Cheung, J.; Rudolph, M.J.; Burshteyn, F.; Cassidy, M.S.; Gary, E.N.; Love, J.; Franklin, M.C.; Height, J.J. Structures of human acetylcholinesterase in complex with pharmacologically important ligands. J. Med. Chem. 2012, 55, 10282–10286. [CrossRef] 21. Hansen, S.B.; Taylor, P. Galanthamine and non-competitive inhibitor binding to ACh-binding protein: Evidence for a binding site on non-alpha-subunit interfaces of heteromeric neuronal nicotinic receptors. J. Mol. Biol. 2007, 369, 895–901. [CrossRef] 22. Shahsavar, A.; Gajhede, M.; Kastrup, J.S.; Balle, T. Structural studies of nicotinic acetylcholine receptors: Using acetylcholine-binding protein as a structural surrogate. Basic Clin. Pharm. Toxicol. 2016, 118, 399–407. [CrossRef] Molecules 2019, 24, 446 16 of 17

23. Kowal, N.M.; Ahring, P.K.; Liao, V.W.Y.; Indurti, D.C.; Harvey, B.S.; O’Connor, S.M.; Chebib, M.; Olafsdottir, E.S.; Balle, T. Galantamine is not a positive allosteric modulator of human α4β2 or α7 nicotinic acetylcholine receptors. Br. J. Pharm. 2018, 175, 2911–2925. [CrossRef] 24. Han, S.Y.; Sweeney, J.E.; Bachman, E.S.; Schweiger, E.J.; Forloni, G.; Coyle, J.T.; Davis, B.M.; Joullié, M.M. Chemical and pharmacological characterization of galanthamine, an acetylcholinesterase inhibitor, and its derivatives. A potential application in Alzheimer’s disease? Eur. J. Med. Chem. 1992, 27, 673–687. [CrossRef] 25. Bores, G.M.; Huger, F.P.; Petko, W.; Mutlib, A.E.; Camacho, F.; Rush, D.K.; Selk, D.E.; Wolf, V.; Kosley, R.W., Jr.; Davis, L.; et al. Pharmacological evaluation of novel Alzheimer’s disease therapeutics: Acetylcholinesterase inhibitors related to galanthamine. J. Pharm. Exper. 1996, 277, 728–738. 26. Triggle, D.J.; Mitchell, J.M.; Filler, R. The pharmacology of physostigmine. Cns Drug Rev. 2006, 4, 87–136. [CrossRef] 27. Jia, P.; Sheng, R.; Zhang, J.; Fang, L.; He, Q.; Yang, B.; Hu, Y. Design, synthesis and evaluation of galanthamine derivatives as acetylcholinesterase inhibitors. Eur. J. Med. Chem. 2009, 44, 772–784. [CrossRef] 28. Weed, M.R.; Polino, J.; Signor, L.; Bookbinder, M.; Keavy, D.; Benitex, Y.; Morgan, D.G.; King, D.; Macor, J.E.; Zaczek, R.; et al. Nicotinic alpha 7 receptor agonists EVP-6124 and BMS-933043, attenuate scopolamine-induced deficits in visuo-spatial paired associates learning. PLoS ONE 2017, 12, e0187609. [CrossRef] 29. Dominguez, J.L.; Fernandez-Nieto, F.; Castro, M.; Catto, M.; Paleo, M.R.; Porto, S.; Sardina, F.J.; Brea, J.M.; Carotti, A.; Villaverde, M.C.; et al. Computer-aided structure-based design of multitarget leads for Alzheimer’s disease. J. Chem. Inf. Model. 2015, 55, 135–148. [CrossRef] 30. Lepailleur, A.; Freret, T.; Lemaitre, S.; Boulouard, M.; Dauphin, F.; Hinschberger, A.; Dulin, F.; Lesnard, A.; Bureau, R.; Rault, S. Dual histamine H3R/serotonin 5-HT4R ligands with antiamnesic properties: Pharmacophore-based virtual screening and polypharmacology. J. Chem. Inf. Model. 2014, 54, 1773–1784. [CrossRef] 31. Hameg, A.; Bayle, F.; Nuss, P.; Dupuis, P.; Garay, R.P.; Dib, M. Affinity of cyamemazine, an anxiolytic antipsychotic drug, for human recombinant dopamine vs. serotonin receptor subtypes. Biochem. Pharm. 2003, 65, 435–440. [CrossRef] 32. Martin, L.F.; Freedman, R. Schizophrenia and the alpha7 nicotinic acetylcholine receptor. Int. Rev. Neurobiol. 2007, 78, 225–246. 33. Young, J.W.; Geyer, M.A. Evaluating the role of the alpha-7 nicotinic acetylcholine receptor in the pathophysiology and treatment of schizophrenia. Biochem. Pharm. 2013, 86, 1122–1132. [CrossRef] 34. Durrieu, G.; Olivier, P.; Bagheri, H.; Montastruc, J.L. Overview of adverse reactions to nefopam: An analysis of the French Pharmacovigilance database. Fundam. Clin. Pharmacol. 2007, 21, 555–558. [CrossRef] 35. Dineley, K.T.; Pandya, A.A.; Yakel, J.L. Nicotinic ACh receptors as therapeutic targets in CNS disorders. Trends Pharmacol. Sci. 2015, 36, 96–108. [CrossRef] 36. Levy, L.; McClure, D. The pharmacology of flazalone: A new class of anti-inflammatory agent. J. Pharm. Exper. 1976, 198, 473–480. 37. Dellisanti, C.D.; Yao, Y.; Stroud, J.C.; Wang, Z.Z.; Chen, L. Crystal structure of the extracellular domain of nAChR alpha1 bound to alpha-bungarotoxin at 1.94 A resolution. Nat. Neurosci. 2007, 10, 953–962. [CrossRef] 38. Berman, H.M.; Westbrook, J.; Feng, Z.; Gilliland, G.; Bhat, T.N.; Weissig, H.; Shindyalov, I.N.; Bourne, P.E. The Protein Data Bank. Nucleic Acids Res. 2000, 28, 235–242. [CrossRef] 39. Notredame, C.; Higgins, D.G.; Heringa, J. T-Coffee: A novel method for fast and accurate multiple sequence alignment. J. Mol. Biol. 2000, 302, 205–217. [CrossRef] 40. Sali, A.; Blundell, T.L. Comparative protein modelling by satisfaction of spatial restraints. J. Mol. Biol. 1993, 234, 779–815. [CrossRef] 41. Shen, M.-Y.; Sali, A. Statistical potential for assessment and prediction of protein structures. Protein Sci. Publ. Protein Soc. 2006, 15, 2507–2524. [CrossRef] 42. Jorgensen, W.L.; Maxwell, D.S.; Tirado-Rives, J. Development and Testing of the OPLS All-Atom Force Field on Conformational Energetics and Properties of Organic Liquids. J. Am. Chem. Soc. 1996, 118, 11225–11236. [CrossRef] 43. Sastry, G.M.; Adzhigirey, M.; Day, T.; Annabhimoju, R.; Sherman, W. Protein and ligand preparation: Parameters, protocols, and influence on virtual screening enrichments. J. Comput. Aid. Mol. Des. 2013, 27, 221–234. [CrossRef] Molecules 2019, 24, 446 17 of 17

44. Sherman, W.; Beard, H.S.; Farid, R. Use of an induced fit receptor structure in virtual screening. Chem. Biol. Drug Des. 2006, 67, 83–84. [CrossRef] 45. Sherman, W.; Day, T.; Jacobson, M.P.; Friesner, R.A.; Farid, R. Novel procedure for modeling ligand/receptor induced fit effects. J. Med. Chem. 2006, 49, 534–553. [CrossRef] 46. Jorgensen, W.L.; Duffy, E.M. Prediction of drug solubility from structure. Adv. Drug Deliv. Rev. 2002, 54, 355–366. [CrossRef] 47. Lipinski, C.A.; Lombardo, F.; Dominy, B.W.; Feeney, P.J. Experimental and computational approaches to estimate solubility and permeability in drug discovery and development settings. Adv. Drug Deliv. Rev. 1997, 23, 3–25. [CrossRef] 48. Friesner, R.A.; Banks, J.L.; Murphy, R.B.; Halgren, T.A.; Klicic, J.J.; Mainz, D.T.; Repasky, M.P.; Knoll, E.H.; Shelley, M.; Perry, J.K.; et al. Glide: A new approach for rapid, accurate docking and scoring. 1. Method and assessment of docking accuracy. J. Med. Chem. 2004, 47, 1739–1749. [CrossRef] 49. Halgren, T.A.; Murphy, R.B.; Friesner, R.A.; Beard, H.S.; Frye, L.L.; Pollard, W.T.; Banks, J.L. Glide: A new approach for rapid, accurate docking and scoring. 2. Enrichment factors in database screening. J. Med. Chem. 2004, 47, 1750–1759. [CrossRef] 50. Sastry, M.; Lowrie, J.F.; Dixon, S.L.; Sherman, W. Large-Scale Systematic Analysis of 2D Fingerprint Methods and Parameters to Improve Virtual Screening Enrichments. J. Chem. Inf. Model. 2010, 50, 771–784. [CrossRef] 51. Mirza, N.R.; Larsen, J.S.; Mathiasen, C.; Jacobsen, T.A.; Munro, G.; Erichsen, H.K.; Nielsen, A.N.; Troelsen, K.B.; Nielsen, E.O.; Ahring, P.K. NS11394 [30-[5-(1-hydroxy-1-methyl-ethyl)-benzoimidazol- 1-yl]-biphenyl-2-carbonitrile], a unique subtype-selective GABAA receptor positive allosteric modulator: In vitro actions, pharmacokinetic properties and in vivo anxiolytic efficacy. J. Pharm. Exper. 2008, 327, 954–968. [CrossRef] 52. Ellman, G.L.; Courtney, K.D.; Andres, V., Jr.; Feather-Stone, R.M. A new and rapid colorimetric determination of acetylcholinesterase activity. Biochem. Pharmacol. 1961, 7, 88–95. [CrossRef]

Sample Availability: Samples of the compounds are not available from the authors.

© 2019 by the authors. Licensee MDPI, Basel, Switzerland. This article is an open access article distributed under the terms and conditions of the Creative Commons Attribution (CC BY) license (http://creativecommons.org/licenses/by/4.0/).