molecules

Review Memantine Derivatives as Multitarget Agents in Alzheimer’s Disease

Giambattista Marotta , Filippo Basagni , Michela Rosini and Anna Minarini * Department of Pharmacy and Biotechnology, Alma Mater Studiorum-University of Bologna, Via Belmeloro 6, 40126 Bologna, Italy; [email protected] (G.M.); fi[email protected] (F.B.); [email protected] (M.R.) * Correspondence: [email protected]



Academic Editors: Maria Novella Romanelli and Silvia Dei
 Received: 10 July 2020; Accepted: 1 September 2020; Published: 2 September 2020

Abstract: Memantine (3,5-dimethyladamantan-1-amine) is an orally active, noncompetitive N-methyl-D-aspartate receptor (NMDAR) antagonist approved for treatment of moderate-to-severe Alzheimer’s disease (AD), a neurodegenerative condition characterized by a progressive cognitive decline. Unfortunately, memantine as well as the other class of drugs licensed for AD treatment acting as acetylcholinesterase inhibitors (AChEIs), provide only symptomatic relief. Thus, the urgent need in AD drug development is for disease-modifying therapies that may require approaching targets from more than one path at once or multiple targets simultaneously. Indeed, increasing evidence suggests that the modulation of a single neurotransmitter system represents a reductive approach to face the complexity of AD. Memantine is viewed as a privileged NMDAR-directed structure, and therefore, represents the driving motif in the design of a variety of multi-target directed ligands (MTDLs). In this review, we present selected examples of small molecules recently designed as MTDLs to contrast AD, by combining in a single entity the amantadine core of memantine with the pharmacophoric features of known neuroprotectants, such as antioxidant agents, AChEIs and Aβ-aggregation inhibitors.

Keywords: memantine; Alzheimer’s disease; multi target directed ligand; hybrid structures

1. Introduction Alzheimer’s disease (AD) is a chronic neurodegenerative condition that slowly destroys nerve cells, leading to a progressive cognitive decline. As for major chronic diseases, a pluralism of causative features is thought to be implicated, resulting in complex networks of perturbation. The mechanistic understanding of synaptic damage might offer a valuable key for developing future therapeutic strategies to tackle AD progression. In this context, excitotoxicity is emerging as a pivotal event. Caused by excessive or prolonged glutamate exposure, this pathological condition alters neurons to death. N-methyl-D-aspartate (NMDA) receptors (NMDAR) are ionotropic glutamate receptors primarily involved in synaptic plasticity underlying learning and memory. However, they are also prime actors of excitotoxic damage occurring during chronic neurodegenerative injuries [1]. Increasing consensus indicates that functional properties of NMDAR are governed by their localization, with synaptic NMDAR (sNMDAR) contributing to cell plasticity and neurotrophic processes, while extrasynaptic NMDAR (eNMDAR) triggering apoptotic signalling pathways [2]. This effect accounts for the difficulties encountered in the design of safe and effective cures acting on NMDARs, and contemporary suggests that a selective targeting of the eNMDARs represents a promising approach to treat neurodegenerative conditions [3]. Memantine is an uncompetitive NMDAR antagonist approved to treat moderate-severe AD patients. It acts as an open-channel blocker with a relatively rapid off-rate from the channel [4].

Molecules 2020, 25, 4005; doi:10.3390/molecules25174005 www.mdpi.com/journal/molecules Molecules 2020, 25, x FOR PEER REVIEW 2 of 20

Memantine is an uncompetitive NMDAR antagonist approved to treat moderate-severe AD Molecules 2020, 25, 4005 2 of 20 patients. It acts as an open-channel blocker with a relatively rapid off-rate from the channel [4]. Due to these peculiar kinetics, memantine mainly enters the channel in conditions of excessive and Dueprolonged to these glutamate peculiar kinetics,exposure, memantine preferentially mainly acting enters on theextrasynaptic/tonically channel in conditions-activated of excessive NMDAR and prolongedover synaptic/phasically glutamate exposure,-activated preferentially NMDAR [5,6]. acting This on peculiar extrasynaptic profile,/tonically-activated which allows memantine NMDAR to overcontrast synaptic excitotoxicity/phasically-activated while preservi NMDARng glutamatergic [5,6]. This synaptic peculiar functioning, profile, which possibly allows accounts memantine for the to contrastclinical tolerability excitotoxicity of the while drug preserving [7]. Unfortunately, glutamatergic however, synaptic memantine functioning, offers only possibly palliative accounts benefits for theto patients.clinical tolerability Possibly, ofthe the explanation drug [7]. Unfortunately, for this issue however, might be memantine the complex off ers neurotransm only palliativeitters benefitsdysfunction to patients. characterizing Possibly, AD, the which explanation suggests for the this modulation issue might of be a thesingle complex neurotransmitter neurotransmitters system dysfunctionto be inadequate characterizing to face the AD, disease. which Increasingsuggests the evidence modulation indicates of a singlea mutual neurotransmitter feedback connection system tobetween be inadequate the glutamatergic to face the and disease. the cholinergic Increasing systems, evidence which indicates both apossess mutual strong feedback implications connection in betweencognitive the functions glutamatergic [8]. Furthermore, and the cholinergic the failure systems, of calcium which homeostatic both possess balance strong deriving implications from inNMDAR cognitive overactivation functions [8 ]. as Furthermore, well as metal the dyshomeostasis failure of calcium (Fe, homeostatic Zn, Cu) [9,10] balance orchestrate derivings several from NMDARpathologic overactivation features, including as well oxidative as metal damage dyshomeostasis [11,12], neuroinflammation (Fe, Zn, Cu) [9,10 and] orchestrates protein misfolding several pathologicand aggregation features, [13,14] including, all contributingoxidative damage to the [11 neurotoxicity,12], neuroinflammation mediated by and excitotoxic protein misfolding damage. andIndeed, aggregation the presence [13,14 of], neuritic all contributing plaques toconstituted the neurotoxicity mainly by mediated insoluble by Aβ excitotoxic-peptide damage.(Aβ) in the Indeed, brain theparenchyma, presence ofis one neuritic of the plaques hallmarks constituted required for mainly AD diagnosis. by insoluble Aβ-peptide (Aβ) in the brain parenchyma,On this basis, is one research of the hallmarks efforts have required recently for focused AD diagnosis. on the identification of therapeutic tools that couldOn target this basis, glutamatergic research e hyperactivationfforts have recently and focused excitotoxicity on the-related identification mechani ofsms, therapeutic triggering tools a thatsynergistic could target response glutamatergic [15]. Thus, hyperactivationMulti Target Directed and excitotoxicity-related Ligands (MTDLs), i.e. mechanisms,, single molecules triggering acting a synergisticon different response targets simultaneously, [15]. Thus, Multi are Target emerging Directed at Ligandsa rapid pace (MTDLs), as a valuable i.e., single opportunity molecules actingto restore on dithefferent complex targets interplay simultaneously, among are multi emergingple NMDAR at a rapid-mediated pace as aalterations valuable opportunity [16,17]. In to this restore context, the complexmemantine interplay is viewed among as multiple a privileged NMDAR-mediated NMDAR-directed alterations structure, [16,17 and]. In therefore this context,, represents memantine the isdriving viewed motif as a privilegedin the design NMDAR-directed of a variety of structure, multifunctional and therefore, compounds. represents In the the search driving for motif a disease in the- designmodifyin of ag varietydrug for of AD, multifunctional the MTDL approach compounds. has been In the applied search by for different a disease-modifying research groups drug to for obtain AD, thehybrid MTDL compounds approach hasable been to hit applied simultaneously by different different research pathways groups to obtainimplicated hybrid in compounds AD [18]. In able this toreview, hit simultaneously we present selected different examples pathways of small implicated molecules in AD designed [18]. In by this mergi review,ng in we a presentsingle entity selected the examplesamantadine of smallcore moleculesof memantine designed with bythe merging chemical in features a single entityof known the amantadine neuroprotectants core of memantineperforming withthrough the chemicaldifferent featuresmechanism of known of action, neuroprotectants such as antioxidant performing and anti through-aggregating different activity mechanism or AChE of action,and monoaminoxidase such as antioxidant inhibition and anti-aggregating (Figure 1). activity or AChE and monoaminoxidase inhibition (Figure1).

Figure 1. Conjugation strategies exploiting memantine’s NMDAR antagonism to tackle AD.

Molecules 2020, 25, 4005 3 of 20

2. Memantine and Cholinesterase Inhibitor Hybrids To date, only two classes of drugs are licensed for the treatment of AD: AChEIs for mild to-moderate disease and memantine for moderate and severe disease. AChEIs prevent cholinesterase enzymes (ChEs) from degrading acetylcholine (ACh), in order to increase the synaptic concentration and the duration of action of the neurotransmitter [19]. Among the various theories that try to explain the characteristic etiopathological cascade of AD, one of the most investigated is the cholinergic hypothesis. Indeed, cholinergic loss is one of the driving impairments since the early symptoms of the disease, unavoidably leading to the characteristic cognitive decline [20]. Actually, the standard-of-care treatment for AD is the combination of cholinesterase’s inhibitors, with the aim to potentiate the cholinergic transmission, and memantine [16]. AChEIs currently used in AD therapy, such as Rivastigmine, Donepezil and Galantamine, are able to stabilize cognitive decline for 3 to 6 months, but do not halt or slow down the progression of the disease [19,21]. Also, some Donepezil-based derivatives prevent the formation of amyloid plaques, since a connection has been demonstrated between AChE activity and plaque formation [22]. Indeed, Aβ peptide interacts with the peripheric anionic site (PAS) of human AChE (hAChE), that operates as a nucleation site enhancing Aβ aggregation [23]. Moreover, throughout AD progression there is a reduction of AChE levels and an increase of those of a nonspecific cholinesterase, butyrylcholinesterase (BChE); thus, targeting the latter could result in higher therapeutic efficacy [20]. Recently, some novel compounds bearing a carbamate moiety have proved to selectively bind to BChE and to act as neuroprotectors in vivo, pointing out a potential advantage for a disease-modifying therapy [24]. Following the polypharmacological approach and the clinical relevance of both memantine and AChEIs, several MTDLs have been developed by joining memantine to pharmacophores, inhibiting ChE with the aim to synergistically and more effectively tackle the AD pathological cascade. Herein there are reported a series of these hybrids divided by their ChEI core.

2.1. Tacrine-Adamantanes Hybrids Tacrine(THA) is a non-competitive reversible AChEI active in the central nervous system (CNS) and slightly selective for BChE [25]. It was the first AChEI drug approved by FDA to combat AD, but it was withdrawn from the market due to its hepatoxicity [26]. Its derivatives 7-methoxytacrine (7-MEOTA) and 6-chlorotacrine (6-Cl-THA) are potent anti-ChE compounds with lower side effects [27,28]. An interesting strategy to obtain a multimodal profile towards ChE and NMDAR is to combine tacrine scaffold with the adamantane core through tethers of different length. In this view, hybrids of 7-MEOTA and amantadine, linked by methylene-thiourea or methylene-urea tethers, were designed and synthesized. The new compounds showed micromolar to submicromolar activities towards hAChE and human BChE (hBChE), NMDA receptor and Aβ1-40 aggregation (Figure2)[29,30]. These hybrids act as competitive inhibitors of hAChE and hBChE with thioureas displaying higher selectivity for the latter; this may be beneficial, since it has been demonstrated that the levels of this enzyme increase in the course of AD (Figure2)[ 20]. They have lower IC50 values for both enzymes than reference 7-MEOTA (10.50 µM for hAChE and 21 µM for hBChE), with the hybrids bearing a five-carbon linker that are the most potent in both series, suggesting that this is the optimal distance for target interaction [29,30]. Furthermore, the urea derivative 2 showed the highest inhibition of Aβ1-40 aggregation, while the thiourea 1 inhibits NMDAR with a similar IC50 value as memantine (1.80 µM vs. 1 µM) and the same efficacy as open channel blocker, but with slower kinetics, which may suggest why it preferentially binds to eNMDAR [31]. In fact, selectivity for eNMDARs is highly desirable due to their involvement in glutamate-mediated excitotoxicity and it is achieved through use-dependent inhibitors with a slow onset of action [32]. In addition to these activities, most of the compounds tested displayed weak-to-moderate activity towards BACE-1 enzyme, which, alongside inhibitory activity on Aβ1-40 aggregation previously showed, proves their potential antiamyloid properties. Molecules 2020, 25, 4005 4 of 20 Molecules 2020, 25, x FOR PEER REVIEW 4 of 20

Compound n X AChE IC50 (μM) BChE IC50 (μM) Aβ IC50 (μM) NMDAR IC50 (μM) 1 5 S 0.47 0.11 66.5 1.80 2 8 O 2.03 18.82 0.038 nd

Figure 2. Drug design and biological activities of 7 7-MEOTA-amantadine-MEOTA-amantadine hybridshybrids 1 and 22..

MolecularThese hybrids modelling act as competitivestudies performed inhibitors on ofcompounds hAChE and1 and hBChE2 showed with thioureas how the displaying 7-MEOTA moietyhigher selectivity binds to the for catalytic the latter anionic; this may site be (CAS) beneficial, while thesince amantadine it has beencore demonstrated binds to the that PAS the of levels both cholinesterases.of this enzyme increase In hAChE, in the the course 7-MEOTA of AD sca (Fiffoldgure creates 2) [20].π- πTheyinteractions have lower within IC50 the values cation- forπ bothsite, whileenzymes the methoxythan reference group stabilizes7-MEOTA the (10.50 complex μM throughfor hAChE hydrogen and 21 bonding. μM for ThehBChE), alkyl with chain the contributes hybrids tobearing stabilize a five the-carbon molecule-enzyme linker that are complex the most through potent in the both interaction series, suggesting of phenyl that groups this ofis the amino optimal acid residuesdistance located for target in the interaction middle of [29,30] the gorge,. Furthermore, whereas the the amantadine urea derivative moiety 2 establishes showed theπ-aliphatic highest interactionsinhibition of and Aβ1 several-40 aggregation, Van der Waalswhile interactionsthe thiourea at 1 theinhibits edge NMDAR of the gorge with in a the similar PAS. InIC the50 valuehBChE, as thememantine interactions (1.80 are μ similar,M vs. 1 with μM) the and nitrogen the same atom efficacy of the three-fused as open channel ring structure blocker, and but the with secondary slower aminokinetics, group which that may create suggest further why hydrogen it preferentially bonding. binds Based to oneNMDAR the results [31] obtained. In fact, fromselectivity docking for studies,eNMDARs assumptions is highly desirable were made due also to totheir understand involvement the abilityin glutamate of these-mediated hybrids toexcitotoxicity counteract Aandβ1-40 it aggregation;is achieved through it was supposed use-dependent that the inhibitors three-ring structurewith a slow of 7-MEOTA onset of action establishes [32]. πIn-π additioninteractions to these with theactivities, hydrophobic most of regions the compou of Aβ, whilends tested the adamantane displayed weakcage is-to involved-moderate in non-aromaticactivity towards interactions. BACE-1 Mostenzyme, urea which, derivatives alongside studied inhibitory here are activity superior on than Aβ1 their-40 aggregation thiourea counterparts previously showed, in inhibiting proves amyloid their fibrillization,potential antiamyloid probably properties. due to the urea moiety that creates stronger hydrogen bonds. FollowingMolecular modellinga similar strategy, studies performed Kaniakova, on M. compounds and coworkers 1 and designed2 showed and how synthesized the 7-MEOTA the 6-Cl-THA-memantinemoiety binds to the catalytic hybrid anionic3 (Figure site3 )[(CAS)33]. while The esamethylene the amantadine linker core wasbinds chosen to the based PAS of on both the optimalcholinesterases. spacer previouslyIn hAChE, the found 7-MEOTA in a series scaffold of galantamine-memantine creates π-π interactions within hybrids, the which cation will-π site, be discussedwhile the later. methoxy group stabilizes the complex through hydrogen bonding. The alkyl chain contributesCompound to stabilize3 demonstrated the molecule to act-enzyme in a multimodal complex throu fashion,gh the with interaction a good activity of phenyl towards groups AChE of andamino NMDAR, acid residues as well located as a higherin the middle neuroprotection of the gorge, ability whereas than memantinethe amantadine after moiety microinjection establishes of NMDAπ-aliphatic into interactions the dorsal ratand hippocampus. several Van der In Waals particular, interactions3 turned at out the to edge be moreof the potent gorge thanin the parent PAS. compoundIn the hBChE, 6-Cl-THA the interactions in inhibiting are AChEsimilar, (IC with50 = the9.41 nitrogen nM vs. 20 atom nM), of but the less three potent-fused than ring memantine structure inand blocking the secondary NMDAR amino (IC50 group= 1.80 thatµM vs.create 0.79 furtherµM). Likehydrogen memantine, bonding. it targets Based the on GluN1the results/N2B obtained subunit, actingfrom docking as an open studies, channel assumptions blocker. Due were to made its promising also to understand pharmacological the ability properties, of these its hybrids ability to crosscounteract the blood-brain Aβ1-40 aggregation barrier (BBB); it was was supposed predicted that using the three the PAMPA-ring structure assayand of 7 its-MEOTA cytotoxic establishes effect in π-π interactions with the hydrophobic regions of Aβ, while the adamantane cage is involved in non- aromatic interactions. Most urea derivatives studied here are superior than their thiourea

Molecules 2020, 25, x FOR PEER REVIEW 5 of 20 counterparts in inhibiting amyloid fibrillization, probably due to the urea moiety that creates stronger hydrogenMolecules 2020 bonds., 25, 4005 5 of 20 Following a similar strategy, Kaniakova, M. and coworkers designed and synthesized the 6-Cl- THA-memantine hybrid 3 (Figure 3) [33]. The esamethylene linker was chosen based on the optimal CHO-K1 mammalian cells was assessed. Results suggest that it has a high probability of passing the spacer previously found in a series of galantamine-memantine hybrids, which will be discussed later. BBB (Pe = 9.20 10 6 cm s 1), but is more cytotoxic than 6-Cl-THA (IC = 5.93 µM vs. 70.7 µM). × − − 50

Figure 3. Drug design and biological activities of 6-chlorotacrine-memantine hybrid 3. Figure 3. Drug design and biological activities of 6-chlorotacrine-memantine hybrid 3. Beside adamantylamine scaffold, its benzofused analogs have been extensively studied as Compound 3 demonstrated to act in a multimodal fashion, with a good activity towards AChE NMDAR antagonists, with fluorobenzohomoadamantanamine being more potent than memantine and andamantadine NMDAR, [34 as]. Due well to as itsa promisinghigher neuroprotection activity, Perez-Areales ability than and memantine coworkers after combined microinjection this scaffold of NMDA into the dorsal rat hippocampus. In particular, 3 turned out to be more potent than parent with 6-Cl-THA moiety through different linker lengths and in different linkage position to evaluate the compound 6-Cl-THA in inhibiting AChE (IC50 = 9.41 nM vs. 20 nM), but less potent than memantine potential pharmacological value of hybrids formed in this way (Figure4)[35]. in blocking NMDAR (IC50 = 1.80 μM vs. 0.79 μM). Like memantine, it targets the GluN1/N2B subunit, The resulting hybrids are better inhibitors of hAChE than 6-Cl-THA (IC50 = 0.3–2.5 nM vs. acting as an open channel blocker. Due to its promising pharmacological properties, its ability to cross 14.5 nM). In particular, with a primary amino group attached to the adamantane moiety, the activity is the blood-brain barrier (BBB) was predicted using the PAMPA assay and its cytotoxic effect in CHO- augmented, with compound 5 being the most potent; the importance of the unsubstituted amino group K1 mammalian cells was assessed. Results suggest that it has a high probability of passing the BBB is prominent in the inhibition of hBChE, for which hybrid 6 has the lowest IC50 value. However, only (Pe = 9.20 × 10−6 cm s−1), but is more cytotoxic than 6-Cl-THA (IC50 = 5.93 μM vs. 70.7 μM). few of the tested hybrids turned out to be more potent than 6-Cl-THA (0.5 µM) in inhibiting hBChE. Beside adamantylamine scaffold, its benzofused analogs have been extensively studied as Compounds 4 and 5 resulted as the most potent against NMDARs. Unlike in memantine, NMDAR antagonists, with fluorobenzohomoadamantanamine being more potent than memantine where the methylation of the amino group leads to a reduced activity, here the monomethylation and amantadine [34]. Due to its promising activity, Perez-Areales and coworkers combined this of the bridgehead amino group in the polycyclic scaffold is tolerated and both compounds are scaffold with 6-Cl-THA moiety through different linker lengths and in different linkage position to about 2-fold more potent than benzohomoadamantanamine (IC50 = 0.9 and 1.2 µM vs. 1.9 µM). evaluate the potential pharmacological value of hybrids formed in this way (Figure 4) [35]. Particularly, in nitrogen-connected series, the hybrid with the longer linker has the best activity, while in the benzo-attached series, the opposite pattern is observed. Apart from the discussed targets, the compounds showed negligible inhibitory efficacy on BACE-1 enzyme and on Aβ42 and tau aggregation. Finally, PAMPA assay for brain permeation predictions was carried out with the results showing a more efficient BBB permeation for derivatives bearing a substituted amino group in the benzofused core (4), probably due to their higher lipophilicity. These studies show that hybrid molecules bearing an aminoadamantane moiety bound to THA or one of its derivatives exert their action in various pathways involved in AD in low-micromolar or nanomolar ranges [35]. In many cases, the spacer plays an important role in modulating the activity: the proper spacers’ length and nature is fundamental to the optimal interaction of pharmacophores to their respective binding sites.

Molecules 2020, 25, 4005 6 of 20 Molecules 2020, 25, x FOR PEER REVIEW 6 of 20

Compound n AChE IC50 (nM) BChE IC50 (μM) NMDAR IC50 (μM) 4 - 1.4 0.2 0.9 5 3 0.3 0.5 1.2 6 4 2.5 0.02 3.4

FigureFigure 4. 4. DrugDrug design design and and biological biological activities of 6-chlorotacrine-benzohomoadamantane6-chlorotacrine-benzohomoadamantane hybrids hybrids4 4––66. .

2.2. Galantamine-MemantineThe resulting hybrids are Hybrids better inhibitors of hAChE than 6-Cl-THA (IC50 = 0.3–2.5 nM vs. 14.5 nM). In particular,Galantamine with is a an primary alkaloid-derived amino group AChEI attached approved to the adamantane since 2001 moiety for the, the treatment activity of is mild-to-moderateaugmented, with formscompound of AD. 5 Furthermore,being the most galantamine potent; the associates importance AChEI of ethefficacy unsubstituted to potentiation amino of NMDARgroup is prominent transmission in mediatedthe inhibition by protein of hBChE, kinase for C whi (PKC)ch hybrid and to 6 the has ability the lowest of allosterically IC50 value. modulated However, αonly-subunit few of of the nicotinic tested hybrids receptors turned [36,37 out]. Particularly,to be more potent through than the 6-Cl modulation-THA (0.5 ofμM)α7-nAChR in inhibiting on presynaptichBChE. glutamatergic neurons, it can enhance the beneficial effect of NMDAR’s synaptic activation, potentiatingCompounds at the 4same and 5 time resulted memantine’s as the most neuroprotective potent against eNMDARsffect [38].. TheseUnlike premises in memantine, were further where assessedthe methylation in an in vitroof thevalidation amino group assay whereleads inactiveto a reduced doseof activity, memantine here and the galantamine monomethylation produced of fullthe neuroprotectivebridgehead amino effi groupcacy, confirming in the polycyclic promising scaff experimentalold is tolerated effi andcacy both of their compounds combination are therapyabout 2- [fold39]. Inmore this potent context, than after benzohomoadamantanamine a computationally-driven SAR (IC50 campaign, = 0.9 and a1.2 series μM vs of. galantamine-memantine1.9 μM). Particularly, in conjugates,nitrogen-connected connected series through, the hybrid the nitrogen with the atom longer of both,linker were has the developed, best activity, demonstrating while in the to benzo exert- bothattach NMDARed series, antagonism the opposite and pattern AChE is inhibitionobserved. withApart high from and the well-balanceddiscussed targets, potencies the compounds [40]. It is importantshowed negligible to note thatinhibitory for AChE efficacy inhibition, on BACE it- is1 fundamentalenzyme and on to A identifyβ42 and thetau properaggregation. linker Finally, length betweenPAMPA theassay two for moieties, brain permeation allowing interaction predictions both was to AChE’s carried CAS out and with PAS. the In results fact, 4-methylene showing a linkermore ofefficient7 prevented BBB permeation it from properly for derivatives interacting bearing with thea substituted PAS (identified amino with group Trp286), in the thereby benzofused reducing core its(4), a probablyffinity when due compared to their higher to derivatives lipophilicity. with These longer studies alkyl chainsshow that (e.g., hybrid8, Figure molecules5). Concurrently, bearing an foraminoadamantane NMDAR antagonism, moiety alkyl bound substituents to THA on or memantine’s one of its derivatives nitrogen are exert allowed. their In action terms in of various AChEI activity,pathways di ffinvolvederent lengths in AD of in the low alkyl-micromolar linker or N-methyl or nanomolar substitutions ranges [35] on memantine. In many cases, moiety the are spacer well tolerated,plays an whereas important spacer role with in modulating heteroatoms the and activity: switching the amine proper function spacers’ of memantine length and into nature amide is greatlyfundame affntalected to AChEIthe optimal profile. interaction Regarding of pharmacophores NMDAR, only alkyl to their or alkylamino respective binding linkers weresites. allowed, whereas all other modifications were detrimental for NMDAR affinity. 2.2. Galantamine-Memantine Hybrids Galantamine is an alkaloid-derived AChEI approved since 2001 for the treatment of mild-to- moderate forms of AD. Furthermore, galantamine associates AChEI efficacy to potentiation of NMDAR transmission mediated by protein kinase C (PKC) and to the ability of allosterically

Molecules 2020, 25, x FOR PEER REVIEW 7 of 20

modulated α-subunit of nicotinic receptors [36,37]. Particularly, through the modulation of α7- nAChR on presynaptic glutamatergic neurons, it can enhance the beneficial effect of NMDAR’s synaptic activation, potentiating at the same time memantine’s neuroprotective effect [38]. These premises were further assessed in an in vitro validation assay where inactive dose of memantine and galantamine produced full neuroprotective efficacy, confirming promising experimental efficacy of their combination therapy [39]. In this context, after a computationally-driven SAR campaign, a series of galantamine-memantine conjugates, connected through the nitrogen atom of both, were developed, demonstrating to exert both NMDAR antagonism and AChE inhibition with high and well-balanced potencies [40]. It is important to note that for AChE inhibition, it is fundamental to identify the proper linker length between the two moieties, allowing interaction both to AChE’s CAS and PAS. In fact, 4-methylene linker of 7 prevented it from properly interacting with the PAS (identified with Trp286), thereby reducing its affinity when compared to derivatives with longer alkyl chains (e.g., 8, Figure 5). Concurrently, for NMDAR antagonism, alkyl substituents on memantine’s nitrogen are allowed. In terms of AChEI activity, different lengths of the alkyl linker or N-methyl substitutions on memantine moiety are well tolerated, whereas spacer with heteroatoms and switching amine function of memantine into amide greatly affected AChEI profile. Regarding NMDAR, only alkyl or alkylamino linkers were allowed, whereas all other modifications were Moleculesdetrimental2020, 25for, 4005 NMDAR affinity. 7 of 20

Compound n AChEI IC50 (nM) NMDAR Ki (μM) 7 2 695.9 2.32 8 3 1.16 4.6

FigureFigure 5.5. Drug design and biological activitiesactivities ofof memantine-galantaminememantine-galantamine hybridshybrids 77 andand 88..

DerivativeDerivative 88 stoodstood outout asas thethe bestbest ofof thethe seriesseries regardingregarding thethe AChEIAChEI/NMDAR/NMDAR profileprofile asas wellwell asas eefficientfficient NR2BNR2B binderbinder (NMDAR(NMDAR withwith 2B2B subunit,subunit, peculiarpeculiar ofof eNMDARseNMDARs involvedinvolved inin glutamatergicglutamatergic overexcitation)overexcitation) andand validvalid neuroprotectiveneuroprotective agent agent after after insult insult by by NMDA NMDA in in SHSY-5Y SHSY-5Y cells cells (IC (IC5050= = 0.280.28 nMnM,, FigureFigure5 5))[ [40]40].. CompoundCompound 7 waswas selectedselected for furtherfurther inin vivovivo biologicalbiological evaluationevaluation as aa resultresult ofof bestbest compromisecompromise between between well-balanced well-balanced pharmacological pharmacological potencies potencies and and physico-chemical physico-chemical properties properties (Figure(Figure5 5))[ [41]41].. DueDue toto aa suboptimalsuboptimal pharmacokineticpharmacokinetic profile,profile, itit waswas testedtested intracerebroventricularlyintracerebroventricularly inin ADAD mousemouse model through osmotic osmotic minipump. minipump. After After 77 days days tr treatmenteatment at at 7.5 7.5 µg/day,µg/day, compound compound 7 7waswas able able to tocompletely completely revert revert the the induced induced neurotoxicity neurotoxicity in vivoin vivo in termsin terms of behavioral of behavioral readouts readouts (e.g., (e.g.,neuroprotective neuroprotective effect e forffect short for short-term-term memory) memory) and biochemical and biochemical analysis analysis (e.g., (e.g.,strong strong reduction reduction of all of all neurodegenerative-related, apoptotic and synaptic plasticity biomarker) [41]. More recently, preliminary in vivo studies were conducted with compound 7 by transdermal iontophoresis delivery, looking to maintain the same pharmacological effects with a more patient-friendly administration. For now, compound 7 has proved to be delivered consistently to the brain through iontophoresis but further preclinical studies and pharmacokinetic-pharmacodynamic optimization are needed to prove its biological effects with this new innovative delivery system [42].

2.3. Aminoadamantane-Carbazole/Tetrahydrocarbazole Hybrids It has been reported that some carbazole derivatives are potential AD-modifying compounds, i.e., N-alkyl carbazoles are able to favor the formation of soluble form of Aβ, through different mechanisms [43] and are also able to block apoptosis, thus exerting a neuroprotective effect [44]. Moreover, recent studies suggested that some carbazole-based compounds are valuable anti-AD drug candidates, since they are able to inhibit both AChE and BChE activity [45] as well as AChE- and self-induced Aβ aggregation [46]. These findings support the rationale to develop novel hybrids with a multimodal profile by adding to biological properties of the carbazole core the neuroprotection derived by the memantine nucleus. Hybrids of aminoadamantanes with carbazole and tetrahydrocarbazole (Figure6) were synthesized and tested as inhibitors of hAChE and equine serum BChE, as antagonists of NMDARs and as stimulators Molecules 2020, 25, x FOR PEER REVIEW 8 of 20 neurodegenerative-related, apoptotic and synaptic plasticity biomarker) [41]. More recently, preliminary in vivo studies were conducted with compound 7 by transdermal iontophoresis delivery, looking to maintain the same pharmacological effects with a more patient-friendly administration. For now, compound 7 has proved to be delivered consistently to the brain through iontophoresis but further preclinical studies and pharmacokinetic-pharmacodynamic optimization are needed to prove its biological effects with this new innovative delivery system [42].

2.3. Aminoadamantane-Carbazole/Tetrahydrocarbazole Hybrids It has been reported that some carbazole derivatives are potential AD-modifying compounds, i.e., N-alkyl carbazoles are able to favor the formation of soluble form of Aβ, through different mechanisms [43] and are also able to block apoptosis, thus exerting a neuroprotective effect [44]. Moreover, recent studies suggested that some carbazole-based compounds are valuable anti-AD drug candidates, since they are able to inhibit both AChE and BChE activity [45] as well as AChE- and self-induced Aβ aggregation [46]. These findings support the rationale to develop novel hybrids with a multimodal profile by adding to biological properties of the carbazole core the neuroprotection Moleculesderived 2020by ,the25, 4005memantine nucleus. 8 of 20 Hybrids of aminoadamantanes with carbazole and tetrahydrocarbazole (Figure 6) were synthesized and tested as inhibitors of hAChE and equine serum BChE, as antagonists of NMDARs ofand microtubule as stimulators assembly of microtubule [47]. In AD, assembly microtubule [47] structure. In AD, disturbance microtubule and structure tubulin disturbance polymerization and suppressiontubulin polymerization are signs of suppression the incoming are neurodegeneration signs of the incoming that neurodegeneration lead to failures in the that axonal lead to transport; failures therefore,in the axonal it is necessarytransport; totherefore develop, compoundsit is necessary that to are develop able to compounds stabilize microtubule that are able structure to stabilize and to enhancemicrotubule tubulin structure assembly and [to47 enhance,48]. tubulin assembly [47,48].

Compound Core R1 R2 R3 BChE IC50 (μM) NMDAR IC50 (μM) 9 Carbazole - - - 15.9 8.2 10 Tetrahydrocarbazole CH3 CH3 H 9.17 8.1 11 Tetrahydrocarbazole F H CH3 5.43 10.4

Figure 6. Drug design and biological activities of carbazole/tetrahydrocarbazole-aminoadamantane Figure 6. Drug design and biological activities of carbazole/tetrahydrocarbazole-aminoadamantane hybrids 9–11. hybrids 9–11. All the hybrids act as weak inhibitors of hAChE but are more potent and more selective for All the hybrids act as weak inhibitors of hAChE but are more potent and more selective for BChE BChE and their inhibition activity is generally higher than reference carbazole (IC50 = 5–20 µM vs. >and20 theirµM). inhibition The substitution activity ofis generally a carbazole higher moiety than with reference a tetrahydrocarbazole carbazole (IC50 = 5 fragment–20 µM vs. leads >20 toµM). an enhanced inhibition for BChE, while minimal differences between memantine and amantadine moiety have been observed. Overall, the highest inhibition activity for BChE is showed by compound 11 with a fluorinated tetrahydrocarbazole core. At the same time these compounds were tested as antagonists of NMDA receptors, and their ability to bind to the intrachannel (or MK-801 binding) site, present in all NMDARs, and/or to the allosteric (or ifenprodil-binding) site of NR2B-containing NMDARs, was assessed. The NR2B subunit is of particular interest because its blockade is beneficial for neuroprotection and Aβ-induced neuronal disruption [49]. In the carbazole series, compounds show different activities towards the allosteric site, and this feature depends primarily on the substitution patterns of the carbazole moiety: Indeed, compound 9 with no substituents is the most active. However, in the tetrahydrocarbazole series, this trend is not observed, in fact all the compounds are differently active towards both binding sites, except for compound 10, which is selective towards the allosteric site of the NR2B subunit and also exhibits the most balanced multimodal profile. These heterodimers have substantially lower IC50 values than memantine and amantadine. In both series, neither memantine moiety nor amantadine one seems important to stimulate tubulin polymerization, while the insertion of a chlorine or a bromine in the carbazole causes a reduction of this Molecules 2020, 25, 4005 9 of 20 stimulating effect. Furthermore, the introduction of a fluorine in the tetrahydrocarbazole core and the unsubstitution of the carbazole one considerably enhances this effect, hence the most active compounds are the fluorinated tetrahydrocarbazole-memantine conjugate and the carbazole-memantine derivative. The fluorine-containing tetrahydrocarbazole hybrids reveal a high neuroprotection profile: Indeed, they are able to reduce the Ca2+-induced swelling of mitochondria and thus inhibit the mitochondria permeability transition (MPT), which is a key process in cell death, either in physiological conditions or in neurotoxicity models. The same compounds were also evaluated and tested as hydrochlorides with compound 10 that showed the lowest IC50 value for the ifenprodil site [50]. Along with this, their effects were studied on the membrane potential of mitochondria, on the Ca2+ ion-caused depolarization and swelling of mitochondria, both indicating MPT. These novel compounds depolarize the membrane, indicating that they are toxic; however, exceptions can be made for compounds containing a fluorine as R substituent, which causes just a slight depolarization potential and shows a dose-dependent-inhibiting effect on Ca2+ ion-induced mitochondrial swelling, thus increasing mitochondrial stability and being assessed as potential neuroprotectors [50]. Following evaluation of the activity and selectivity of these compounds, molecular docking studies were performed, in order to shed light on the binding poses. It was reported that memantine and amantadine lay in the cavity located at the active site gorge of BChE, and their positions were very stable, while the ones of the linker and of the carbazole moiety changed. These compounds have a chiral centre, but the binding poses of the two enantiomers are very similar, with the hydroxyl groups that establish hydrogen bonds with different amino acid residues depending on the R or S isomer. Taking into account the position of the entire ligand, it was highlighted that it does not vary when amantadine is replaced by memantine, and a similar behavior is observed with the switch from carbazole to tetrahydrocarbazole; however, when a halogen substituent is inserted, the carbazole core undergoes a flip to reduce the steric strain [47]. These studies demonstrate that also carbazoles linked to memantine are able to maintain important NMDAR-blocking activities while adding interesting neuroprotection effects through tackling several pathological pathways involved in AD.

3. Memantine-Antioxidant Hybrids to Explore Neuroinflammation In light of the above described multifaced nature of AD, the multitarget approach could help to elucidate insights in the complex pathogenetic network triggering pathologic conditions. Beside glutamate-driven excitotoxicity, oxidative stress, misfolded proteins and neuroinflammation are only some of the actors which, playing in liaison, define AD-impaired neuroenvironment. Particularly, prolonged eNMDAR activation leads to increasing cytoplasmic Ca2+ concentration and the onset of oxidative damage, which can in turn pull the trigger of aberrant misfolded protein cascade (e.g., amyloid plaques formation) [11]. Amyloid plaques are one of the classical trademarks of AD, characterized by Aβ42 deposition, which originates from the sequential cleavage of APP by β- and γ-secretases rather than the non-amyloidogenic and physiologic α-secretase cleavage. In concert, Aβ deposition triggered by excitotoxicity-leading oxidative stress can exacerbate oxidative damage in an etiopathogenetic loop, which involves microglia activation and the following neuroinflammatory conditions [14,51]. Based on these premises, linking memantine with bioactive payload able to tackle Aβ/ROS/ neuroinflammation could be a promising pathway to deepen knowledge in NMDAR-mediated neurotoxic events in AD.

3.1. Memantine-Ferulic Acid Hybrids Recently, Rosini and coworkers developed a series of hybrids between memantine and ferulic acid (FA), envisioned by the concurrent ability to reduce Aβ-induced neurotoxicity and oxidative stress of this latter [52,53]. The best derivatives of the series are those directly linked at the carboxylic function of FA and nitrogen atom of memantine through alkyl chain of different lengths, which were initially screened for their NMDAR antagonism efficacy. Derivative 12 came out as the most promising Molecules 2020, 25, x FOR PEER REVIEW 10 of 20 amyloid plaques formation) [11]. Amyloid plaques are one of the classical trademarks of AD, characterized by Aβ42 deposition, which originates from the sequential cleavage of APP by β- and γ- secretases rather than the non-amyloidogenic and physiologic α-secretase cleavage. In concert, Aβ deposition triggered by excitotoxicity-leading oxidative stress can exacerbate oxidative damage in an etiopathogenetic loop, which involves microglia activation and the following neuroinflammatory conditions [14,51]. Based on these premises, linking memantine with bioactive payload able to tackle Aβ/ROS/neuroinflammation could be a promising pathway to deepen knowledge in NMDAR- mediated neurotoxic events in AD.

3.1. Memantine-Ferulic Acid Hybrids Recently, Rosini and coworkers developed a series of hybrids between memantine and ferulic acid (FA), envisioned by the concurrent ability to reduce Aβ-induced neurotoxicity and oxidative stress of this latter [52,53]. The best derivatives of the series are those directly linked at the carboxylic functionMolecules 2020 of FA, 25, and 4005 nitrogen atom of memantine through alkyl chain of different lengths, which10 were of 20 initially screened for their NMDAR antagonism efficacy. Derivative 12 came out as the most promising NMDAR-blocking agent, with potency only 3-fold less than memantine (IC50 = 6.9 µM vs. NMDAR-blocking agent, with potency only 3-fold less than memantine (IC50 = 6.9 µM vs. 2.3 µM, 2.3 µM, Figure 7), sharing with the parent compound the same open channel blocking mechanism Figure7), sharing with the parent compound the same open channel blocking mechanism and the and the binding site midway through the channel pore. Compound 12 exerted interesting antioxidant binding site midway through the channel pore. Compound 12 exerted interesting antioxidant efficacy, efficacy, both directly as radical scavenger and indirectly through activation of Nrf2-ARE pathway. both directly as radical scavenger and indirectly through activation of Nrf2-ARE pathway. Interestingly, Interestingly, differently from memantine and FA alone, it was also able to stimulate the non- differently from memantine and FA alone, it was also able to stimulate the non-amyloidogenic amyloidogenic pathway and consequently limiting Aβ production at 10 μM concentration [53]. This pathway and consequently limiting Aβ production at 10 µM concentration [53]. This new balanced new balanced memantine-hybrid, acting at a different level of AD pathological cascade, could be a memantine-hybrid, acting at a different level of AD pathological cascade, could be a starting point starting point for further evaluation in more complex AD models, to deepen our findings in NMDAR- for further evaluation in more complex AD models, to deepen our findings in NMDAR-guided guided oxidative damage. oxidative damage.

Figure 7. Drug design and biological activities of ferulic acid-memantine hybrid 12.

3.2. Memantine-GlutathioneFigure 7. Drug design/Lipoic and Acid biological Hybrids activities of ferulic acid-memantine hybrid 12.

3.2. MemantineThe possibility-Glutathione/Lipoic of connecting Acid antioxidant Hybrids agents with the memantine core has also been explored [53,54] using radical scavengers such as glutathione (GSH) and (R)-α-Lipoic Acid (LA). The possibility of connecting antioxidant agents with the memantine core has also been explored They are endogenous molecules capable of removing free radicals that accumulate in cells, thus [53,54] using radical scavengers such as glutathione (GSH) and (R)-α-Lipoic Acid (LA). They are maintaining ROS balance at physiological levels and preventing oxidative stress and cell damage [55,56]. endogenous molecules capable of removing free radicals that accumulate in cells, thus maintaining Furthermore, LA has been shown to mitigate Aβ aggregation and to protect from Aβ-mediated citotoxicity in vitro [57]. In light of these findings, Sozio, P. and coworkers developed conjugates of memantine linked to GSH and LA through an amide bond and tested them as NMDAR antagonists, Aβ aggregation inhibitors and as radical scavengers, pointing out compound 13 as the best derivative (Figure8)[58]. The antioxidant properties of compound 13 were assessed in GL15 cell line upon the addition of hydrogen peroxide (H2O2) and superoxide radical anion (O2−); compared to LA, it showed a similar antioxidant activity towards H2O2, while a slightly lower one was shown towards O2−. In addition, compound 13 displayed Aβ antiaggregating properties (41% at 10 µM). In order to investigate the effects of these conjugates on NMDAR, the authors used an in vitro protocol evaluating the release of [3H]Noradrenaline ([3H]NA) modulated by presynaptic NMDAR from rat hippocampal synaptosomes. Compound 13 turned out to be inactive as NMDAR ligand. This is consistent with previous SAR studies on memantine, which reported a loss of NMDAR activity upon conversion of the free adamantane’s amine group into amide group [40]. Molecules 2020, 25, x FOR PEER REVIEW 11 of 20

ROS balance at physiological levels and preventing oxidative stress and cell damage [55,56]. Furthermore, LA has been shown to mitigate Aβ aggregation and to protect from Aβ-mediated citotoxicity in vitro [57]. In light of these findings, Sozio, P. and coworkers developed conjugates of memantine linked to GSH and LA through an amide bond and tested them as NMDAR antagonists, MoleculesAβ aggr2020egation, 25, 4005 inhibitors and as radical scavengers, pointing out compound 13 as the best derivative11 of 20 (Figure 8) [58].

Figure 8. Drug design and biological activities of lipoic acid-memantine hybrid 13. Figure 8. Drug design and biological activities of lipoic acid-memantine hybrid 13. 3.3. Amantadine-Propargylamine Hybrids The antioxidant properties of compound 13 were assessed in GL15 cell line upon the addition of Monoamine oxidase B (MAO-B) with monoamine oxidase A (MAO-A) belong to flavoenzyme hydrogen peroxide (H2O2) and superoxide radical anion (O2−); compared to LA, it showed a similar family bound to outer mitochondrial membrane. They exert a pivotal role in neurotransmission, antioxidant activity towards H2O2, while a slightly lower one was shown towards O2−. In addition, regulatingcompound neurotransmitter 13 displayed Aβ antiaggregating metabolism and properties cellular redox (41% balance. at 10 µM). Particularly, MAO-B has been gainingIn order particular to investigate interest in the recent effects years of these due to conjugates its overexpression on NMDAR, in neuroinflammatory the authors used an condition in vitro andprotocol the consequential evaluating the ROS release overproduction of [3H]Noradrenaline and neurotransmitters ([3H]NA) modulated impairment by presynaptic [59,60]. Accordingly, NMDAR developmentfrom rat hippocampal of MAO-B syna inhibitorsptosomes. seems Compound to be an interesting13 turned out strategy to be to inactive deepen as our NMDAR knowledge ligand. on diseasesThis is consistent characterized with by previous neuroinflammatory SAR studies conditions, on memantine like AD, which and Parkinson reported disease a loss of(PD) NMDAR [59]. activityIn 2014,upon Malanconversion and coworkersof the free adamantane’s developed a seriesamine of group polycyclic into amide propargylamine group [40]. or acetylene derivatives with the aim to synergistically limit apoptotic processes and excitotoxicity, due to aberrant 2+ intracellular3.3. Amantadine Ca -Propargylamineconcentration andHybrids oxidative damage deriving from increasing MAO-B activity in AD. Polycyclic cores, such as memantine and amantadine, are established Ca2+ channel blockers, acting on Monoamine oxidase B (MAO-B) with monoamine oxidase A (MAO-A) belong to flavoenzyme NMDAR, while bigger polycyclic nucleus exert their effect, modulating L-type calcium channels [61]. family bound to outer mitochondrial membrane. They exert a pivotal role in neurotransmission, In parallel, propargylamine moiety is the responsible fragment of MAO-B irreversible inhibition in regulating neurotransmitter metabolism and cellular redox balance. Particularly, MAO-B has been compounds like selegiline, a drug used for the treatment of PD. Amantadine-propargylamine hybrids gaining particular interest in recent years due to its overexpression in neuroinflammatory condition 14 and 15, bearing respectively one or two propargyl functions directly connected to the nitrogen atom, and the consequential ROS overproduction and neurotransmitters impairment [59,60]. Accordingly, demonstrated anti-apoptotic activity similar to selegiline (Figure9)[ 62]. Compound 14 exhibited the development of MAO-B inhibitors seems to be an interesting strategy to deepen our knowledge on best L-type voltage-gated calcium channels (VGCC) inhibitory activity (45% at 100 µM), whereas with diseases characterized by neuroinflammatory conditions, like AD and Parkinson disease (PD) [59]. compound 15 there was a drop of activity (18% at 100 µM), demonstrating the importance of N-benzyl In 2014, Malan and coworkers developed a series of polycyclic propargylamine or acetylene substitution for target recognition. Both compounds maintained mild NMDAR inhibitory activity derivatives with the aim to synergistically limit apoptotic processes and excitotoxicity, due to (52% for 15 and 59% for 14 at 100 µM) thanks to the presence of an amantadine core. Unfortunately, aberrant intracellular Ca2+ concentration and oxidative damage deriving from increasing MAO-B none of them exerted acceptable MAO-B inhibitory potency at 300 µM. These derivatives confirm the activity in AD. Polycyclic cores, such as memantine and amantadine, are established Ca2+ channel ability of memantine/amantadine scaffold to partially maintain NMDAR inhibitory activity, also in blockers, acting on NMDAR, while bigger polycyclic nucleus exert their effect, modulating L-type N-substituted derivatives (e.g., as previously seen in ferulic and galantamine hybrids), whereas it does not seem a good starting scaffold to reach satisfactory MAO-B inhibition.

Molecules 2020, 25, x FOR PEER REVIEW 12 of 20

calcium channels [61]. In parallel, propargylamine moiety is the responsible fragment of MAO-B irreversible inhibition in compounds like selegiline, a drug used for the treatment of PD. Amantadine-propargylamine hybrids 14 and 15, bearing respectively one or two propargyl functions directly connected to the nitrogen atom, demonstrated anti-apoptotic activity similar to selegiline (Figure 9) [62]. Compound 14 exhibited the best L-type voltage-gated calcium channels (VGCC) inhibitory activity (45% at 100 μM), whereas with compound 15 there was a drop of activity (18% at 100 μM), demonstrating the importance of N-benzyl substitution for target recognition. Both compounds maintained mild NMDAR inhibitory activity (52% for 15 and 59% for 14 at 100 μM) thanks to the presence of an amantadine core. Unfortunately, none of them exerted acceptable MAO- B inhibitory potency at 300 μM. These derivatives confirm the ability of memantine/amantadine scaffold to partially maintain NMDAR inhibitory activity, also in N-substituted derivatives (e.g., as previously seen in ferulic and galantamine hybrids), whereas it does not seem a good starting scaffold Moleculesto reach2020 satisfactory, 25, 4005 MAO-B inhibition. 12 of 20

Apoptosis NMDAR MAO-B VGCC Inhibition Compound Induced at 10 μM Inhibition at 100 Inhibition at 300 at 100 μM (%) (%) μM (%) μM (%) 14 70 45 59 20 15 68 18 52 14

FigureFigure 9.9. DrugDrug designdesign andand biologicalbiological activitiesactivities ofof propargyl-amantadinepropargyl-amantadine hybridshybrids14 14and and15 15..

4.4. MiscellaneousMiscellaneous MemantineMemantine DerivativesDerivatives InIn thisthis section,section, wewe reviewreview studiesstudies reportedreported inin thethe recentrecent chemicalchemical andand biologicalbiological literatureliterature onon thethe miscellaneousmiscellaneous memantine memantine derivatives derivatives for for whichwhich ADAD neuroprotectiveneuroprotective activity activity derived derived byby mergingmerging twotwo pharmacophorespharmacophores is is claimed, claimed, but but which which lie lie outside outside the the hybrid hybrid series series described described in thein the other other paragraphs paragraphs of thisof this review. review. In this In regard,this regard, a lot ofa lot eff ortsof efforts have been have made been overmade the over years the to years enhance to enhance memantine memantine potency, synthesizingpotency, synthesizing double-acting double prodrug-acting orprodrug targeting or targeting the second the region second of region the same of the receptor same toreceptor increase to itsincrease efficacy its [6 , efficacy63]. Otherwise, [6,63]. Otherwise, merging two mergi well-knownng two well NMDAR-known antagonists NMDAR has antagonists been exploited has been as aexploited strategy to as obtain a strategy more to effi obtaincient channel more efficient blockers channel with potential blockers further with clinical potential applications further clinical [64]. Furthermore,applications [64] linking. Furthermore, memantine linking to a purinergic memantine receptor to a purinergic blocker has receptor been pursued blocker as has a new been approach pursued toas investigatea new appr otheroach pathwaysto investigate involved other inpathways AD [65]. involved in AD [65].

4.1. Memantine-Polyamine Conjugates

Kumamoto, T. and coworkers developed a series of memantine-polyamine conjugates showing very potent NMDAR antagonism, higher than memantine [64]. The rationale is based on the well-known potent blocking NMDAR activity of the polyamine backbone, that the same authors have already conjugated with polycyclic scaffolds, developing compounds endowed with proved strong reversible and voltage-dependent NMDA blocking activity [66–68]. These derivatives were further explored and optimized, merging polyamine moiety with memantine scaffold (Figure 10). The triamine 16 and guanidino-diamine 17 demonstrated the highest inhibitory activities against GluN1/N2A and GluN1/N2B NMDAR (Figure 10). Particularly, compound 17 exhibited IC50 3-fold lower against both receptors when compared with parent compound memantine (IC50 = 379 nM and 433 nM vs. 1376 nM and 2099 nM, respectively) [64]. Molecules 2020, 25, x FOR PEER REVIEW 13 of 20

4.1. Memantine-Polyamine Conjugates Kumamoto, T. and coworkers developed a series of memantine-polyamine conjugates showing very potent NMDAR antagonism, higher than memantine [64]. The rationale is based on the well- known potent blocking NMDAR activity of the polyamine backbone, that the same authors have already conjugated with polycyclic scaffolds, developing compounds endowed with proved strong reversible and voltage-dependent NMDA blocking activity [66–68]. These derivatives were further explored and optimized, merging polyamine moiety with memantine scaffold (Figure 10). The triamine 16 and guanidino-diamine 17 demonstrated the highest inhibitory activities against GluN1/N2A and GluN1/N2B NMDAR (Figure 10). Particularly, compound 17 exhibited IC50 3-fold lower against both receptors when compared with parent compound memantine (IC50 = 379 nM and Molecules 2020, 25, 4005 13 of 20 433 nM vs. 1376 nM and 2099 nM, respectively) [64].

Compounds Glu1/GluN2A IC50 (nM) Glu1/GluN2B IC50 (nM) 16 656 587 17 379 433

FigureFigure 10. 10. DrugDrug design design and and biological biological activities activities of of polyamine polyamine-memantine-memantine hybrids hybrids 1616 andand 1717. .

4.2.4.2. H H2S2S-Releasing-Releasing Memantine Memantine Prodrug Prodrug InIn recent recent years, years, the the possibility possibility of of link linkinging moieties moieties releasing releasing endogenous endogenous gaseous gaseous signalling signalling molemolecules,cules, i.e. i.e.,, gasotransmitters, gasotransmitters, with with a a well well-known-known pharmacophore pharmacophore,, has has assumed assumed particular particular interest interest inin multitarget multitarget drug drug design design [69,70] [69,70].. Gasotransmitters, Gasotransmitters, such such as as carbon carbon monoxide monoxide (CO), (CO), nitric nitric oxide oxide (NO)(NO) and and hydrogen hydrogen sulfyde sulfyde (H (H2S),2 S),are areendogenously endogenously produced produced by enzymes by enzymes and play and important play important roles asroles antioxidants, as antioxidants, neuromodulators neuromodulators and and antinflammatory antinflammatory agents agents [71] [71. ]. Particularly, Particularly, H H2S2 S combines combines neuroprotection,neuroprotection, antinflammatory antinflammatory and and antiapoptotic antiapoptotic activities activities to to NMDAR NMDAR-modulation-modulation ability ability both both directly,directly, with with sulfhydration sulfhydration of of crit criticalical cysteine cysteine residues, residues, and and indirectly, indirectly, through through regulation regulation of of intracellularintracellular calcium calcium concentration concentration [72 [72,73],73]. On. On this basis, this basis, the interesting the interesting biological biological outputs of outputs compound of compound18, an H2S-releasing 18, an H2S memantine-releasing memantine derivative byderivative Sestito and by Sestito coworkers, and coworkers can open a, rangecan open of possibilities a range of possibilitiesfor new MTDL for campaignnew MTDL [63 campaign]. Compound [63]18. Compoundbears an isothiocyanate 18 bears an isothiocyanate moiety as H2S-donating moiety as moiety, H2S- donatingreplacing moiety the characteristic, replacing the free characteristic amino group free of amino memantine group (Figure of memantine 11). First (Figure of all, 11) compound. First of all,18 compoundshowed to 18 maintain showed memantine’sto maintain memantine’s NMDAR inhibition NMDAR (K inhibitioni = 458 vs. (Ki 329 = 458 nM) vs. only329 nM) in the only presence in the presenceof 4 mM ofL-cysteine, 4 mM L which-cysteine can, mediatewhich can H2 S mediate release and H2S memantine’s release and unveiling. memantine’s Amperometric unveiling. Amperometricanalysis on derivative analysis18 ondemonstrated derivative 18 demonstrated a prolonged and a prolonged persisting and cysteine-mediated persisting cysteine H2S-mediated releasing. Furthermore, compound 18 was able to protect cells from oxidative stress and LPS-TNFα insult, reducing ROS production and stimulating cell proliferation in neuroinflammation in an in vitro model at 10 µM (Figure 11). Antiamyloidogenic properties of memantine were confirmed with compound 18, which demonstrated to reduce Aβ42 self-induced aggregation and to reduce Aβ-induced damage in rat microglia cells to roughly the same extent as its parent compound [63]. Molecules 2020, 25, x FOR PEER REVIEW 14 of 20

H2S releasing. Furthermore, compound 18 was able to protect cells from oxidative stress and LPS- TNFα insult, reducing ROS production and stimulating cell proliferation in neuroinflammation in an in vitro model at 10 μM (Figure 11). Antiamyloidogenic properties of memantine were confirmed Moleculeswith 2020compound, 25, 4005 18, which demonstrated to reduce Aβ42 self-induced aggregation and to reduce14 of 20 Aβ- induced damage in rat microglia cells to roughly the same extent as its parent compound [63].

Figure 11. Drug design and biological activities of memantine prodrug 18. Figure 11. Drug design and biological activities of memantine prodrug 18. 4.3. Dual P2X7-NMDA Receptor Antagonists 4.3. Dual P2X7-NMDA Receptor Antagonists P2X7 receptor (P2X7R) is an ATP-sensitive ion channel, present in the CNS (astrocytes, oligodendrocytesP2X7 receptor and (P microglia),2X7R) is an where ATP it-sensitive mediates ion K + channel,efflux andpresent Ca 2 in+/ Na the+ CNSinflux (astrocytes, [74,75]. Inoligodendrocytes physiological conditions, and microglia), it exerts where a neuroprotective it mediates e Kff+ect, efflux i.e., and the activation Ca2+/Na+ influx of the [74,75] cAMP. In responsephysiological element-binding conditions, (CREB),it exerts a which neuroprotective inhibits the effect, transcription i.e., the activation of genes of involvedthe cAMP inresponse the microglialelement- inflammatorybinding (CREB), response which [76 inhibits] and the the stimulation transcription of ofα -secretase, genes involved which in is responsiblethe microglial forinflammatory the non-amyloidogenic response [76] cleavage and theof the stimulation amyloid precursor of α-secretase, protein which (APP), is withresponsible the formation for the of non- solubleamyloidogenic and non-toxic cleavage sAPP αof[ 77the]. amyloid precursor protein (APP), with the formation of soluble and nonIn- thetoxic presence sAPPα of[77] high. ATP levels, like in AD, microglial P2X7Rs are upregulated, giving rise to an inflammatoryIn the cascade,presence where of high IL-1 ATPβ, INF- levels,γ, TNF- like αin, andAD, other microglial cytokines P2X are7Rs released, are upregulated and this, contributes giving rise to to thean progression inflammatory of the cascade, disease where [74,78 , IL79-].1β Furthermore,, INF-γ, TNF P2X7Rs-α, and release other glutamate cytokines [ are80], releasedwhich activates, and this NMDARscontributes [81]. to A the research progression conducted of the on disease AD mouse [74,78,79] model. Furthermore, reported that theP2Xin7Rs vivo releaseP2X7R glutamate inhibition [80], reducedwhich theactivates levels NMDARs of Aβ, thus [81] showing. A research a neuroprotective conducted on AD effect. mouse However, model reported the effects that of the P2X7R in vivo inhibitionP2X7R in inhibition pathological reduced conditions the levels are still of Aβ, poorly thus understood showing a and neuroprotective need further investigation effect. However [78]., the effectsBased of on P2X these7R inhibition premises, in newpathological P2X7R-NMDAR conditions antagonists are still poorly were understood developed and based need on further an amantadineinvestigation or memantine[78]. moiety bound to a N’-arylcarbohydrazide core, which is correlated to strongBased P2X7R on antagonism these premises, (Figure new 12)[ 65 P2].X7R-NMDAR antagonists were developed based on an amantadineCompound or19 memantinebearing a o moiety-chloro bound phenyl to ring a N’ showed-arylcarbohydrazide the highest inhibitory core, which activity is correlated towards to P2X7R,strong but P2 lowerX7R antagonism blocker activity (Figure towards 12) [65] NMDAR. compared to amantadine. Hence, following the encouraging P2X7 inhibition, structural modifications were made to improve NMDAR antagonism by replacing the amantadine core of compound 19 with a memantine moiety. However, the resulting hybrids were not able to significantly inhibit NMDARs and also showed higher IC50 values for P2X7R than compound 19. The study demonstrates that these molecules could potentially slow down neuroinflammation by preventing the P2X7R-induced inflammatory cascade. Even if they effectively bind to the P2X7R, they did not show an improved activity for NMDAR compared to memantine. Molecules 2020, 25, 4005 15 of 20 Molecules 2020, 25, x FOR PEER REVIEW 15 of 20

Figure 12. Drug design and biological activities of N’-arylcarbohydrazide-aminoadamantane hybrid 19.

5. ConclusionsFigure 12. Drug and design Future and Perspectivesbiological activities of N’-arylcarbohydrazide-aminoadamantane hybrid 19. CompoundAlthough AD 19 currentbearing pharmacotherapy a o-chloro phenyl is ring founded showed on cholinergic the highest and inhibitory glutamatergic activity hypotheses, towards Pthe2X7 drugsR, but licensed lower blocker for AD activity treatment towards acting NMDAR as AChEIs compared or as NMDAR to amant antagonistadine. Hence, (memantine) following provide the encouragingonly symptomatic P2X7 inhibition, and temporary structural support. modifications In recent years,were made the research to improve of new NMDAR therapeutic antagonism agents byhas replacing been mainly the amantadine centred on thecore amyloidogenic of compound 19 hypothesis with a memantine and based moiety. on the However, multifactorial the resulting nature of hybridsAD. The were assumption not able to that significantly intractable inhibit neurodegenerative NMDARs and diseases also showed like ADhigher are IC caused50 values by for more P2X7 thanR thanone compound interacting 19 mechanism. has driven many research groups in designing and synthesising small moleculesThe study by merging demonstrates two or morethat these distinct molecules pharmacophores could potentially with different slow biologicaldown neuroinflammation properties. In this bycontext, preventing memantine the P2X7 representsR-induced a inflammatory privileged structure cascade. in Even the design if they ofeffectively a variety bind of multifunctional to the P2X7R, theycompounds did not show to be an developed improved as activity new AD for therapeutic NMDAR compared agents. Into thismemantine. review, we have reported the more recent studies relative to the molecular hybridization of memantine or its amantadine nucleus 5. Conclusion and Future Perspectives with the chemical features of AChEI (tacrine and galantamine), on the basis of the close interplay betweenAlthough the glutamatergic AD current and pharmacotherapy the cholinergic systems is founded in disease on process, cholinergic and withand MAO glutamatergic inhibitor hypotheses,(propargylamine) the drugs and neuroprotectantlicensed for AD (ferulic treatment and a lipoiccting asacid, AChEIs and carbazole) or as NMDAR scaffolds antagonist with the (memantine)aim to halt the provide etiopathogenetic only symptomatic loop of and neuroinflammation temporary support. involving In recent glutamatergic years, the research excitotoxicity, of new therapeuticoxidative stress agents and has protein been aggregation. mainly centred The on biological the amyloidogenic profile of the hypothesis new molecules, and based in some on cases the multifactorialvery promising, nature highlights of AD. theThe greatassumption challenge that for intractable medicinal neurodegenerative chemists in finding diseases disease-modifying like AD are causedagents by with more balanced than one activities interacting towards mechanism the different has driven targets many involved research in AD. groups Several in issues designing remain and to synthesisingbe addressed small to translate molecules AD multitarget by merging approaches two or intomore e ff distinctective drug pharmacophores candidates against with AD, differen mainlyt biologicala better comprehension properties. In this of disease context, processes memantine and represents a higher predictive a privileged validity structure of animal in the models. design of a variety of multifunctional compounds to be developed as new AD therapeutic agents. In this review, Author Contributions: G.M. collected articles, organized and wrote the review; F.B. collected articles and wecontributed have reported in writing the more the review; recent M.R. studies contributed relative into writingthe molecular and editing hybridization the draft; A.M. of memantine conceived, or wrote its amantadineand supervised nucleus the preparation with the chemical of the review. features All authorsof AChEI have (tacrine read and and agreed galantamine), to the published on the versionbasis of of thethe close manuscript. interplay between the glutamatergic and the cholinergic systems in disease process, and withFunding: MAOThis inhibitor research (propargylamine) was funded by Italianand neuroprotectant Ministry of University (ferulic andand Researchlipoic acid, (MIUR), and carbazole) PRIN 2017 scaffolds(2017MT3993_007). with the aim to halt the etiopathogenetic loop of neuroinflammation involving glutamatergicConflicts of Interest: excitotoxicity,The authors oxidative declare no stress conflict and of protein interest. aggregation. The biological profile of the new molecules, in some cases very promising, highlights the great challenge for medicinal chemists in finding disease-modifying agents with balanced activities towards the different targets involved in AD. Several issues remain to be addressed to translate AD multitarget approaches into effective

Molecules 2020, 25, 4005 16 of 20

References

1. Parsons, M.P.; Raymond, L.A. Extrasynaptic NMDA receptor involvement in central nervous system disorders. Neuron 2014, 82, 279–293. [CrossRef] 2. Léveillé, F.; El Gaamouch, F.; Gouix, E.; Lecocq, M.; Lobner, D.; Nicole, O.; Buisson, A. Neuronal viability is controlled by a functional relation between synaptic and extrasynaptic NMDA receptors. FASEB J. 2008, 22, 4258–4271. [CrossRef] 3. Folch, J.; Busquets, O.; Ettcheto, M.; Sánchez-López, E.; Castro-Torres, R.D.; Verdaguer, E.; Garcia, M.L.; Olloquequi, J.; Casadesús, G.; Beas-Zarate, C.; et al. Memantine for the Treatment of Dementia: A Review on its Current and Future Applications. J. Alzheimer’s Dis. 2018, 62, 1223–1240. [CrossRef] 4. Xia, P.; Chen, H.S.; Zhang, D.; Lipton, S.A. Memantine preferentially blocks extrasynaptic over synaptic NMDA receptor currents in hippocampal autapses. J. Neurosci. 2010, 30, 11246–11250. [CrossRef][PubMed] 5. Lipton, S.A. Pathologically activated therapeutics for neuroprotection. Nat. Rev. Neurosci. 2007, 8, 803–808. [CrossRef][PubMed] 6. Takahashi, H.; Xia, P.; Cui, J.; Talantova, M.; Bodhinathan, K.; Li, W.; Saleem, S.; Holland, E.A.; Tong, G.; Piña-Crespo, J.; et al. Pharmacologically targeted NMDA receptor antagonism by NitroMemantine for cerebrovascular disease. Sci. Rep. 2015, 5, 14781. [CrossRef] 7. Lipton, S.A. Paradigm shift in neuroprotection by NMDA receptor blockade: Memantine and beyond. Nat. Rev. Drug Discov. 2006, 5, 160–170. [CrossRef] 8. Parsons, C.G.; Danysz, W.; Dekundy, A.; Pulte, I. Memantine and cholinesterase inhibitors: Complementary mechanisms in the treatment of Alzheimer’s disease. Neurotox. Res. 2013, 24, 358–369. [CrossRef] 9. Barnham, K.J.; Bush, A.I. Biological metals and metal-targeting compounds in major neurodegenerative diseases. Chem. Soc. Rev. 2014, 43, 6727–6749. [CrossRef] 10. Santos, M.A.; Chand, K.; Chaves, S. Recent progress in multifunctional metal chelators as potential drugs for Alzheimer’s disease. Coord. Chem. Rev. 2016, 327–328, 287–303. [CrossRef] 11. Lipton, S. NMDA receptor activity regulates transcription of antioxidant pathways. Nat. Neurosci. 2008, 11, 381–382. [CrossRef] 12. Nakamura, T.; Lipton, S.A. Preventing Ca2+-mediated nitrosative stress in neurodegenerative diseases: Possible pharmacological strategies. Cell Calcium 2010, 47, 190–197. [CrossRef][PubMed] 13. Kodis, E.J.; Choi, S.; Swanson, E.; Ferreira, G.; Bloom, G.S. N-methyl-D-aspartate receptor-mediated calcium influx connects amyloid-β oligomers to ectopic neuronal cell cycle reentry in Alzheimer’s disease. Alzheimer’s Dement. 2018, 14, 1302–1312. [CrossRef] 14. Tu, S.; Okamoto, S.; Lipton, S.A.; Xu, H. Oligomeric Aβ-induced synaptic dysfunction in Alzheimer’s disease. Mol. Neurodegener. 2014, 9, 48. [CrossRef] 15. Banks, W.A. A Spectrum of Topics for 2019: Advances in Neuroinflammation, Oxidative Stress, Obesity, Diabetes Mellitus, Cardiovascular Disease, Autism, Exosomes, and Central Nervous System Diseases. Curr. Pharm. Des. 2020, 26, 1–5. [CrossRef][PubMed] 16. Rosini, M.; Simoni, E.; Minarini, A.; Melchiorre, C. Multi-target design strategies in the context of Alzheimer’s disease: Acetylcholinesterase inhibition and NMDA receptor antagonism as the driving forces. Neurochem. Res. 2014, 39, 1914–1923. [CrossRef][PubMed] 17. Rosini, M.; Simoni, E.; Caporaso, R.; Minarini, A. Multitarget strategies in Alzheimer’s disease: Benefits and challenges on the road to therapeutics. Future Med. Chem. 2016, 8, 697–711. [CrossRef][PubMed] 18. Rosini, M. Polypharmacology: The rise of multitarget drugs over combination therapies. Future Med. Chem. 2014, 6, 485–487. [CrossRef] 19. Colovi´c,M.B.; Krsti´c,D.Z.; Lazarevi´c-Pašti,T.D.; Bondži´c,A.M.; Vasi´c,V.M. Acetylcholinesterase inhibitors: Pharmacology and toxicology. Curr. Neuropharmacol. 2013, 11, 315–335. [CrossRef] 20. Li, Q.; Yang, H.; Chen, Y.; Sun, H. Recent progress in the identification of selective butyrylcholinesterase inhibitors for Alzheimer’s disease. Eur. J. Med. Chem. 2017, 132, 294–309. [CrossRef] 21. Korabecny, J.; Spilovska, K.; Mezeiova, E.; Benek, O.; Juza, R.; Kaping, D.; Soukup, O. A Systematic Review on Donepezil-based Derivatives as Potential Cholinesterase Inhibitors for Alzheimers Disease. Curr. Med. Chem. 2019, 26, 5625–5648. [CrossRef][PubMed] 22. Li, Q.; He, S.; Chen, Y.; Feng, F.; Qu, W.; Sun, H. Donepezil-based multi-functional cholinesterase inhibitors for treatment of Alzheimer’s disease. Eur. J. Med. Chem. 2018, 158, 463–477. [CrossRef] Molecules 2020, 25, 4005 17 of 20

23. Lushchekina, S.V.; Kots, E.D.; Novichkova, D.A.; Petrov, K.A.; Masson, P. Role of Acetylcholinesterase in beta-amyloid Aggregation Studied by Accelerated Molecular Dynamics. BioNanoScience 2017, 7, 396–402. [CrossRef] 24. Hoffmann, M.; Stiller, C.; Endres, E.; Scheiner, M.; Gunesch, S.; Sotriffer, C.; Maurice, T.; Decker, M. Highly Selective Butyrylcholinesterase Inhibitors with Tunable Duration of Action by Chemical Modification of Transferable Carbamate Units Exhibit Pronounced Neuroprotective Effect in an Alzheimer’s Disease Mouse Model. J. Med. Chem. 2019, 62, 9116–9140. [CrossRef][PubMed] 25. Horak, M.; Holubova, K.; Nepovimova, E.; Krusek, J.; Kaniakova, M.; Korabecny, J.; Vyklicky, L.; Kuca, K.; Stuchlik, A.; Ricny, J.; et al. The pharmacology of tacrine at N-methyl-d-aspartate receptors. Prog. Neuropsychopharmacol. Biol. Psychiatry 2017, 75, 54–62. [CrossRef][PubMed] 26. Romero, A.; Cacabelos, R.; Oset-Gasque, M.J.; Samadi, A.; Marco-Contelles, J. Novel tacrine-related drugs as potential candidates for the treatment of Alzheimer’s Disease. Bioorg. Med. Chem. Lett. 2013, 23, 1916–1922. [CrossRef] 27. Gancar, M.; Ho, K.; Mohid, S.A.; Thai, N.Q.; Bednarikova, Z.; Nguyen, H.L.; Bhunia, A.; Nepovimova, E.; Li, M.S.; Gazova, Z. 7-Methoxytacrine and 2-Aminobenzothiazole Heterodimers: Structure-Mechanism Relationship of Amyloid Inhibitors Based on Rational Design. ACS Chem. Neurosci. 2020, 11, 715–729. [CrossRef] 28. Misik, J.; Nepovimova, E.; Pejchal, J.; Kassa, J.; Korabecny, J.; Soukup, O. Cholinesterase Inhibitor 6-Chlorotacrine—In Vivo Toxicological Profile and Behavioural Effects. Curr. Alzheimer Res. 2018, 15, 552–560. [CrossRef] 29. Spilovska, K.; Korabecny, J.; Kral, J.; Horova, A.; Musilek, K.; Soukup, O.; Drtinova, L.; Gazova, Z.; Siposova, K.; Kuca, K. 7-Methoxytacrine-adamantylamine heterodimers as cholinesterase inhibitors in Alzheimer’s disease treatment–synthesis, biological evaluation and molecular modeling studies. Molecules 2013, 18, 2397–2418. [CrossRef] 30. Spilovska, K.; Korabecny, J.; Horova, A.; Musilek, K.; Nepovimova, E.; Drtinova, L.; Gazova, Z.; Siposova, K.; Dolezal, R.; Jun, D.; et al. Design, synthesis and in vitro testing of 7-methoxytacrine-amantadine analogues: A novel cholinesterase inhibitors for the treatment of Alzheimer’s disease. Med. Chem. Res. 2015, 24, 2645–2655. [CrossRef] 31. Gazova, Z.; Soukup, O.; Sepsova, V.; Siposova, K.; Drtinova, L.; Jost, P.; Spilovska, K.; Korabecny, J.; Nepovimova, E.; Fedunova, D.; et al. Multi-target-directed therapeutic potential of 7-methoxytacrine-adamantylamine heterodimers in the Alzheimer’s disease treatment. Biochim. Biophys. Acta Mol. Basis Dis. 2017, 1863, 607–619. [CrossRef] 32. Vyklicky, V.; Smejkalova, T.; Krausova, B.; Balik, A.; Korinek, M.; Borovska, J.; Horak, M.; Chvojkova, M.; Kleteckova, L.; Vales, K.; et al. Preferential Inhibition of Tonically over Phasically Activated NMDA Receptors by Pregnane Derivatives. J. Neurosci. 2016, 36, 2161–2175. [CrossRef] 33. Kaniakova, M.; Nepovimova, E.; Kleteckova, L.; Skrenkova, K.; Holubova, K.; Chrienova, Z.; Hepnarova, V.; Kucera, T.; Kobrlova, T.; Vales, K.; et al. Combination of Memantine and 6-Chlorotacrine as Novel Multi-Target Compound against Alzheimer’s Disease. Curr. Alzheimer Res. 2019, 16, 821–833. [CrossRef][PubMed] 34. Valverde, E.; Sureda, F.X.; Vázquez, S. Novel benzopolycyclic amines with NMDA receptor antagonist activity. Bioorg. Med. Chem. 2014, 22, 2678–2683. [CrossRef] 35. Pérez-Areales, F.J.; Turcu, A.L.; Barniol-Xicota, M.; Pont, C.; Pivetta, D.; Espargaró, A.; Bartolini, M.; De Simone, A.; Andrisano, V.; Pérez, B.; et al. A novel class of multitarget anti-Alzheimer benzohomoadamantane-chlorotacrine hybrids modulating cholinesterases and glutamate NMDA receptors. Eur. J. Med. Chem. 2019, 180, 613–626. [CrossRef] 36. Santos, M.D.; Alkondon, M.; Pereira, E.F.; Aracava, Y.; Eisenberg, H.M.; Maelicke, A.; Albuquerque, E.X. The nicotinic allosteric potentiating ligand galantamine facilitates synaptic transmission in the mammalian central nervous system. Mol. Pharmacol. 2002, 61, 1222–1234. [CrossRef] 37. Moriguchi, S.; Marszalec, W.; Zhao, X.; Yeh, J.Z.; Narahashi, T. Mechanism of action of galantamine on N-methyl-D-aspartate receptors in rat cortical neurons. J. Pharmacol. Exp. Ther. 2004, 310, 933–942. [CrossRef] 38. McGehee, D.S.; Heath, M.J.; Gelber, S.; Devay, P.; Role, L.W. Nicotine enhancement of fast excitatory synaptic transmission in CNS by presynaptic receptors. Science 1995, 269, 1692–1696. [CrossRef] 39. 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 2020, 25, 4005 18 of 20

40. Simoni, E.; Daniele, S.; Bottegoni, G.; Pizzirani, D.; Trincavelli, M.L.; Goldoni, L.; Tarozzo, G.; Reggiani, A.; Martini, C.; Piomelli, D.; et al. Combining galantamine and memantine in multitargeted, new chemical entities potentially useful in Alzheimer’s disease. J. Med. Chem. 2012, 55, 9708–9721. [CrossRef] 41. Reggiani, A.M.; Simoni, E.; Caporaso, R.; Meunier, J.; Keller, E.; Maurice, T.; Minarini, A.; Rosini, M.; Cavalli, A. In Vivo Characterization of ARN14140, a Memantine/Galantamine-Based Multi-Target Compound for Alzheimer’s Disease. Sci. Rep. 2016, 6, 33172. [CrossRef][PubMed] 42. Singhal, M.; Merino, V.; Rosini, M.; Cavalli, A.; Kalia, Y.N. Controlled Iontophoretic Delivery in vitro and in vivo of ARN14140—A Multitarget Compound for Alzheimer’s Disease. Mol. Pharm. 2019, 16, 3460–3468. [CrossRef][PubMed] 43. Saturnino, C.; Iacopetta, D.; Sinicropi, M.S.; Rosano, C.; Caruso, A.; Caporale, A.; Marra, N.; Marengo, B.; Pronzato, M.A.; Parisi, O.I.; et al. N-alkyl carbazole derivatives as new tools for Alzheimer’s disease: Preliminary studies. Molecules 2014, 19, 9307–9317. [CrossRef][PubMed] 44. Pieper, A.A.; McKnight, S.L.; Ready, J.M. P7C3 and an unbiased approach to drug discovery for neurodegenerative diseases. Chem. Soc. Rev. 2014, 43, 6716–6726. [CrossRef] 45. Choubdar, N.; Golshani, M.; Jalili-Baleh, L.; Nadri, H.; Küçükkilinç, T.T.; Ayazgök, B.; Moradi, A.; Moghadam, F.H.; Abdolahi, Z.; Ameri, A.; et al. New classes of carbazoles as potential multi-functional anti-Alzheimer’s agents. Bioorg. Chem. 2019, 91, 103164. [CrossRef] 46. Ghobadian, R.; Nadri, H.; Moradi, A.; Bukhari, S.N.A.; Mahdavi, M.; Asadi, M.; Akbarzadeh, T.; Khaleghzadeh-Ahangar, H.; Sharifzadeh, M.; Amini, M. Design, synthesis, and biological evaluation of selective and potent Carbazole-based butyrylcholinesterase inhibitors. Bioorg. Med. Chem. 2018, 26, 4952–4962. [CrossRef] 47. Bachurin, S.O.; Shevtsova, E.F.; Makhaeva, G.F.; Grigoriev, V.V.; Boltneva, N.P.; Kovaleva, N.V.; Lushchekina, S.V.; Shevtsov, P.N.; Neganova, M.E.; Redkozubova, O.M.; et al. Novel conjugates of aminoadamantanes with carbazole derivatives as potential multitarget agents for AD treatment. Sci. Rep. 2017, 7, 45627. [CrossRef] 48. Ballatore, C.; Brunden, K.R.; Huryn, D.M.; Trojanowski, J.Q.; Lee, V.M.; Smith, A.B. Microtubule stabilizing agents as potential treatment for Alzheimer’s disease and related neurodegenerative tauopathies. J. Med. Chem. 2012, 55, 8979–8996. [CrossRef] 49. Zhang, Y.; Li, P.; Feng, J.; Wu, M. Dysfunction of NMDA receptors in Alzheimer’s disease. Neurol. Sci. 2016, 37, 1039–1047. [CrossRef] 50. Sokolov, V.B.; Aksinenko, A.Y.; Goreva, T.V.; Epishina, T.A.; Grigor’ev, V.V.; Gabrel’yan, A.V.; Vinogradova, D.V.; Neganova, M.E.; Shevtsova, E.F.; Bachurin, S.O. Molecular design of multitarget neuroprotectors 3. Synthesis and bioactivity of tetrahydrocarbazole-aminoadamantane conjugates. Russ. Chem. Bull. 2016, 65, 1354–1359. [CrossRef] 51. Von Bernhardi, R.; Eugenín-von Bernhardi, L.; Eugenín, J. Microglial cell dysregulation in brain aging and neurodegeneration. Front. Aging Neurosci. 2015, 7, 124. [CrossRef][PubMed] 52. Benchekroun, M.; Romero, A.; Egea, J.; León, R.; Michalska, P.; Buendía, I.; Jimeno, M.L.; Jun, D.; Janockova, J.; Sepsova, V.; et al. The Antioxidant Additive Approach for Alzheimer’s Disease Therapy: New Ferulic (Lipoic) Acid Plus Melatonin Modified Tacrines as Cholinesterases Inhibitors, Direct Antioxidants, and Nuclear Factor (Erythroid-Derived 2)-Like 2 Activators. J. Med. Chem. 2016, 59, 9967–9973. [CrossRef][PubMed] 53. Rosini, M.; Simoni, E.; Caporaso, R.; Basagni, F.; Catanzaro, M.; Abu, I.F.; Fagiani, F.; Fusco, F.; Masuzzo, S.; Albani, D.; et al. Merging memantine and ferulic acid to probe connections between NMDA receptors, oxidative stress and amyloid-β peptide in Alzheimer’s disease. Eur. J. Med. Chem. 2019, 180, 111–120. [CrossRef][PubMed] 54. Jalili-Baleh, L.; Forootanfar, H.; Küçükkılınç, T.T.; Nadri, H.; Abdolahi, Z.; Ameri, A.; Jafari, M.; Ayazgok, B.; Baeeri, M.; Rahimifard, M.; et al. Design, synthesis and evaluation of novel multi-target-directed ligands for treatment of Alzheimer’s disease based on coumarin and lipoic acid scaffolds. Eur. J. Med. Chem. 2018, 152, 600–614. [CrossRef] 55. Bansal, A.; Simon, M.C. Glutathione metabolism in cancer progression and treatment resistance. J. Cell Biol. 2018, 217, 2291–2298. [CrossRef] 56. Tibullo, D.; Li Volti, G.; Giallongo, C.; Grasso, S.; Tomassoni, D.; Anfuso, C.D.; Lupo, G.; Amenta, F.; Avola, R.; Bramanti, V. Biochemical and clinical relevance of alpha lipoic acid: Antioxidant and anti-inflammatory activity, molecular pathways and therapeutic potential. Inflamm. Res. 2017, 66, 947–959. [CrossRef] Molecules 2020, 25, 4005 19 of 20

57. Ono, K.; Hirohata, M.; Yamada, M. Alpha-lipoic acid exhibits anti-amyloidogenicity for beta-amyloid fibrils in vitro. Biochem. Biophys. Res. Commun. 2006, 341, 1046–1052. [CrossRef] 58. Sozio, P.; Cerasa, L.S.; Laserra, S.; Cacciatore, I.; Cornacchia, C.; Di Filippo, E.S.; Fulle, S.; Fontana, A.; Di Crescenzo, A.; Grilli, M.; et al. Memantine-sulfur containing antioxidant conjugates as potential prodrugs to improve the treatment of Alzheimer’s disease. Eur. J. Pharm. Sci. 2013, 49, 187–198. [CrossRef] 59. Edmonson, D.E.; Binda, C. Monoamine Oxidases. In Membrane Protein Complexes: Structure and Function; Harris, J.R., Boekema, E.J., Eds.; Springer Nature: Singapore, 2018; pp. 117–140. 60. Schedin-Weiss, S.; Inoue, M.; Hromadkova, L.; Teranishi, Y.; Yamamoto, N.G.; Wiehager, B.; Bogdanovic, N.; Winblad, B.; Sandebring-Matton, A.; Frykman, S.; et al. Monoamine oxidase B is elevated in Alzheimer disease neurons, is associated with γ-secretase and regulates neuronal amyloid β-peptide levels. Alzheimers Res. Ther. 2017, 9, 57. [CrossRef] 61. Joubert, J.; Geldenhuys, W.J.; Van der Schyf, C.J.; Oliver, D.W.; Kruger, H.G.; Govender, T.; Malan, S.F. Polycyclic cage structures as lipophilic scaffolds for neuroactive drugs. ChemMedChem 2012, 7, 375–384. [CrossRef] 62. Zindo, F.T.; Barber, Q.R.; Joubert, J.; Bergh, J.J.; Petzer, J.P.; Malan, S.F. Polycyclic propargylamine and acetylene derivatives as multifunctional neuroprotective agents. Eur. J. Med. Chem. 2014, 80, 122–134. [CrossRef][PubMed] 63. Sestito, S.; Daniele, S.; Pietrobono, D.; Citi, V.; Bellusci, L.; Chiellini, G.; Calderone, V.; Martini, C.; Rapposelli, S. Memantine prodrug as a new agent for Alzheimer’s Disease. Sci. Rep. 2019, 9, 4612. [CrossRef][PubMed] 64. Kumamoto, T.; Nakajima, M.; Uga, R.; Ihayazaka, N.; Kashihara, H.; Katakawa, K.; Ishikawa, T.; Saiki, R.; Nishimura, K.; Igarashi, K. Design, synthesis, and evaluation of polyamine-memantine hybrids as NMDA channel blockers. Bioorg. Med. Chem. 2018, 26, 603–608. [CrossRef][PubMed] 65. Karoutzou, O.; Kwak, S.H.; Lee, S.D.; Martínez-Falguera, D.; Sureda, F.X.; Vázquez, S.; Kim, Y.C.; Barniol-Xicota, M. Towards a Novel Class of Multitarget-Directed Ligands: Dual P2X7-NMDA Receptor Antagonists. Molecules 2018, 23, 230. [CrossRef] 66. Jin, L.; Sugiyama, H.; Takigawa, M.; Katagiri, D.; Tomitori, H.; Nishimura, K.; Kaur, N.; Phanstiel, O.; Kitajima, M.; Takayama, H.; et al. Comparative studies of anthraquinone- and anthracene-tetraamines as blockers of N-methyl-D-aspartate receptors. J. Pharmacol. Exp. Ther. 2007, 320, 47–55. [CrossRef] 67. Takayama, H.; Yaegashi, Y.; Kitajima, M.; Han, X.; Nishimura, K.; Okuyama, S.; Igarashi, K. Design, synthesis, and biological evaluation of tricyclic heterocycle-tetraamine conjugates as potent NMDA channel blockers. Bioorg. Med. Chem. Lett. 2007, 17, 4729–4732. [CrossRef] 68. Saiki, R.; Yoshizawa, Y.; Minarini, A.; Milelli, A.; Marchetti, C.; Tumiatti, V.; Toida, T.; Kashiwagi, K.; Igarashi, K. In vitro and in vivo evaluation of polymethylene tetraamine derivatives as NMDA receptor channel blockers. Bioorg. Med. Chem. Lett. 2013, 23, 3901–3904. [CrossRef] 69. Sestito, S.; Nesi, G.; Pi, R.; Macchia, M.; Rapposelli, S. Hydrogen Sulfide: A Worthwhile Tool in the Design of New Multitarget Drugs. Front. Chem. 2017, 5, 72. [CrossRef] 70. Basagni, F.; Lanni, C.; Minarini, A.; Rosini, M. Lights and shadows of electrophile signaling: Focus on the Nrf2-Keap1 pathway. Future Med. Chem. 2019, 11, 707–721. [CrossRef] 71. Shefa, U.; Yeo, S.G.; Kim, M.S.; Song, I.O.; Jung, J.; Jeong, N.Y.; Huh, Y. Role of Gasotransmitters in Oxidative Stresses, Neuroinflammation, and Neuronal Repair. Biomed. Res. Int. 2017, 2017, 1689341. [CrossRef] 72. Kimura, H. Physiological role of hydrogen sulfide and polysulfide in the central nervous system. Neurochem. Int. 2013, 63, 492–497. [CrossRef] 73. Nagai, Y.; Tsugane, M.; Oka, J.; Kimura, H. Hydrogen sulfide induces calcium waves in astrocytes. FASEB J. 2004, 18, 557–559. [CrossRef][PubMed] 74. Cie´slak,M.; Wojtczak, A. Role of purinergic receptors in the Alzheimer’s disease. Purinergic Signal. 2018, 14, 331–344. [CrossRef][PubMed] 75. Deussing, J.M.; Arzt, E. P2X7 Receptor: A Potential Therapeutic Target for Depression? Trends Mol. Med. 2018, 24, 736–747. [CrossRef][PubMed] 76. Kopp, R.; Krautloher, A.; Ramírez-Fernández, A.; Nicke, A. P2X7 Interactions and Signaling—Making Head or Tail of It. Front. Mol. Neurosci. 2019, 12, 183. [CrossRef] 77. Erb, L.; Woods, L.T.; Khalafalla, M.G.; Weisman, G.A. Purinergic signaling in Alzheimer’s disease. Brain Res. Bull. 2019, 151, 25–37. [CrossRef] Molecules 2020, 25, 4005 20 of 20

78. Martin, E.; Amar, M.; Dalle, C.; Youssef, I.; Boucher, C.; Le Duigou, C.; Brückner, M.; Prigent, A.; Sazdovitch, V.; Halle, A.; et al. New role of P2X7 receptor in an Alzheimer’s disease mouse model. Mol. Psychiatry 2019, 24, 108–125. [CrossRef] 79. Illes, P.; Rubini, P.; Ulrich, H.; Zhao, Y.; Tang, Y. Regulation of Microglial Functions by Purinergic Mechanisms in the Healthy and Diseased CNS. Cells 2020, 9, 1108. [CrossRef] 80. Burnstock, G. Purinergic signalling and disorders of the central nervous system. Nat. Rev. Drug Discov. 2008, 7, 575–590. [CrossRef] 81. Gao, P.; Ding, X.; Khan, T.M.; Rong, W.; Franke, H.; Illes, P. P2X7 receptor-sensitivity of astrocytes and neurons in the substantia gelatinosa of organotypic spinal cord slices of the mouse depends on the length of the culture period. Neuroscience 2017, 349, 195–207. [CrossRef]

© 2020 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/).