Novel Tacrine-Scutellarin Hybrids as Multipotent Anti-Alzheimer’s Agents: Design, Synthesis and Biological Evaluation

A novel series of 6-chlorotacrine-scutellarin hybrids was designed, synthesized and the biological activity as potential anti-Alzheimer’s agents was assessed. Their inhibitory activity towards human acetylcholinesterase (hAChE) and human butyrylcholinesterase (hBChE), antioxidant activity, ability to cross the blood-brain barrier (BBB) and hepatotoxic profile were evaluated in vitro. Among these compounds, hybrid K1383, bearing two methylene tether between two basic scaffolds, was found to be very potent hAChE inhibitor (IC50 = 1.63 nM). Unfortunately, none of the hybrids displayed any antioxidant activity (EC50 ≥ 500 μM). Preliminary data also suggests a comparable hepatotoxic profile with 6-Cl-THA (established on a HepG2 cell line). Kinetic studies performed on hAChE with the most active compound in the study, K1383, pointed out to a mixed, non-competitive enzyme inhibition. These findings were further corroborated by docking studies.


Introduction
Alzheimer's disease (AD) is a progressive neurodegenerative brain disorder characterized by gradual degeneration and loss of neurons [1]. According to the World Alzheimer Report 2015, approximately 46.8 million people suffer from AD and other forms of dementia worldwide. This number is expected to increase due to the rapid growth of the aging population, reaching 74.7 million by 2030 and 131.5 million by 2050 [2]. To date the pathogenesis of this neurodegenerative disease has not been fully elucidated. However, several factors have been linked to play an important role in AD pathogenesis, including loss of cholinergic transmission, excessive accumulation of Aβ aggregates extracellularly in the brain, accumulation of tau protein intracellularly, oxidative stress with free radicals formation, biometal dyshomeostasis, glutamate excitotoxicity, inflammatory processes as well as disturbed mitochondrial transport to synaptic terminals [3,4].
Current AD therapeutic approaches rely on the administration of cholinesterase inhibitors (ChEIs) (donepezil, galantamine, and rivastigmine) and a non-competitive N-methyl-D-aspartate receptor antagonist (memantine), which contribute to an improvement of cognitive functions, but do not prevent from progressive neurodegeneration [5,6].
The multifactorial character of AD requires new therapeutics to successfully affect multiple AD-relevant pathways responsible for the pathogenesis of the disease [7]. The so-called multi-target directed ligands (MTDLs) strategy has been postulated as a promising one [8][9][10][11][12][13][14][15][16]. Accordingly, the most widely accepted theory for drug discovery in AD by medicinal chemists is tied to the modification of acetylcholinesterase inhibitors (AChEIs). The possible benefit of MTDLs are reduced incidence of possible drug-drug interactions, a higher efficacy of these drugs and concomitantly a simple dosing mode which may enhance patient compliance [17]. However, today it remains unclear which pathological aspects of AD are best to target to resolve the multiple nature of the disease.
6-Chlorotacrine (6-Cl-THA) is a potent acetylcholinesterase inhibitor (AChEI) widely used as a basic moiety to be the introduced into novel MTDLs. It displays toxicity comparable to that of tacrine (THA), [18] but higher AChE inhibitory activity. Several 6-Cl-THA-based hybrids, such as 6-Cl-THA-melatonin hybrids [19,20], 6-Cl-THA-1,4-naphthoquinone hybrids [21], 6-Cl-THA-caffeic acid heterodimers [22], 6-Cl-THA-Trolox heterodimers [23], hybrids combining 6-Cl-THA with caffeic, ferulic and lipoic acid [24] and 6-Cl-THA-ferulic acid heterodimers [25], were synthesized and evaluated in vitro, revealing interesting biological profiles. receptor antagonist (memantine), which contribute to an improvement of cognitive functions, but do not prevent from progressive neurodegeneration [5,6]. The multifactorial character of AD requires new therapeutics to successfully affect multiple ADrelevant pathways responsible for the pathogenesis of the disease [7]. The so-called multi-target directed ligands (MTDLs) strategy has been postulated as a promising one [8][9][10][11][12][13][14][15][16]. Accordingly, the most widely accepted theory for drug discovery in AD by medicinal chemists is tied to the modification of acetylcholinesterase inhibitors (AChEIs). The possible benefit of MTDLs are reduced incidence of possible drug-drug interactions, a higher efficacy of these drugs and concomitantly a simple dosing mode which may enhance patient compliance [17]. However, today it remains unclear which pathological aspects of AD are best to target to resolve the multiple nature of the disease.
Based upon the aforementioned findings, we report the design, synthesis and biological evaluation of a novel series of 6-Cl-THA-scutellarin hybrids K1383-K1389 (Figure 1), including their AChE and BChE inhibitory activity, the antioxidant activity, molecular modeling studies, ability to cross BBB, as well as the evaluation of their hepatotoxicity on HepG2 cell line.
The synthon 8 and the 1,ω-diaminotacrines 9-15 were utilized for formation of the desired 6-Cl-THA-scutellarin hybrids K1383-K1389 in a Buchwald-Hartwig cross-coupling reaction under catalysis of Pd 2 (dba) 3 + BINAP + NaOtBu in dry toluene [47]. The final compounds were obtained as yellow powders in yields ranging from 15-88%. Structural determination and signal assignments were accomplished by the application of the usual combination of 1 H-and 13 C-NMR spectroscopy. Final compounds were also characterized by high-resolution mass spectrometry (HRMS) and melting points.

Cholinesterase Inhibition
The in vitro inhibitory activities of the seven novel 6-Cl-THA-scutellarin hybrids K1383-K1389 were determined on the models of recombinant hAChE (E.C. 3.1.1.7) and plasmatic hBChE (E.C. 3.1.1.8) using spectroscopic Ellman's method [48,49] and compared with 6-Cl-THA as reference (Table 1). All new heterodimers were found to be potent inhibitors of both cholinesterases. In general, these hybrids exerted high affinity towards hAChE with IC50 values ranging from 1.63 nM for the most active hybrid K1383 to the least potent K1387 31.0 nM. Structure-activity relationship (SAR) related to hAChE revealed that derivatives bearing shorter linkers slightly exceeded the activity of those with longer chains. Especially derivatives K1383, K1384 and K1385 displayed single nanomolar digit hAChE inhibition, surpassing the inhibition capability of 6-Cl-THA by one order of magnitude. Regarding hBChE inhibition, all the hybrids were less potent inhibitors of this enzyme, with IC50 values lying in the micromolar to submicromolar region. The highest inhibition power for hBChE was demonstrated by hybrid K1384 (IC50 = 174 nM). It is well established that the activity of AChE decreases during the course of AD, while that of BChE increases. Several reports indicate the ability of BChE to compensate the lack of AChE thus continuing in regulation of cholinergic transmission. Therefore, the agents with affinity to both cholinesterases could play a more profound role in the AD therapy [50]. The selectivity index (SI), reflecting the relative preference for hAChE over hBChE, was

Cholinesterase Inhibition
The in vitro inhibitory activities of the seven novel 6-Cl-THA-scutellarin hybrids K1383-K1389 were determined on the models of recombinant hAChE (E.C. 3.1.1.7) and plasmatic hBChE (E.C. 3.1.1.8) using spectroscopic Ellman's method [48,49] and compared with 6-Cl-THA as reference (Table 1). All new heterodimers were found to be potent inhibitors of both cholinesterases. In general, these hybrids exerted high affinity towards hAChE with IC 50 values ranging from 1.63 nM for the most active hybrid K1383 to the least potent K1387 31.0 nM. Structure-activity relationship (SAR) related to hAChE revealed that derivatives bearing shorter linkers slightly exceeded the activity of those with longer chains. Especially derivatives K1383, K1384 and K1385 displayed single nanomolar digit hAChE inhibition, surpassing the inhibition capability of 6-Cl-THA by one order of magnitude. Regarding hBChE inhibition, all the hybrids were less potent inhibitors of this enzyme, with IC 50 values lying in the micromolar to submicromolar region. The highest inhibition power for hBChE was demonstrated by hybrid K1384 (IC 50 = 174 nM). It is well established that the activity of AChE decreases during the course of AD, while that of BChE increases. Several reports indicate the ability of BChE to compensate the lack of AChE thus continuing in regulation of cholinergic transmission. Therefore, the agents with affinity to both cholinesterases could play a more profound role in the AD therapy [50]. The selectivity index (SI), reflecting the relative preference for hAChE over hBChE, was calculated presuming high selectivity of novel 6-Cl-THA-scutellarin family for hAChE. From this point of view, K1385, showing the highest SI for hAChE (SI = 1687) which is comparable to that for donepezil (SI value for donepezil = 1100) [51], can be highlighted. Previously prepared compound 1 [31] showed sub-micromolar and micromolar value inhibitory activity toward rat AChE and rat BChE, respectively. Hybrid 2 [42] exhibited sub-micromolar inhibitory activity toward EeAChE and hAChE and micromolar eqBChE inhibitory activity. Although these compounds were tested on different type of enzyme than novel 6-Cl-THA-scutellarin hybrids, they displayed in general better inhibitory potency against hAChE.

Kinetic Analysis
To determine the interaction of K1383 with hAChE, a kinetic study was carried out. Enzyme rate curves were obtained at several substrate concentrations of the inhibitor. The type of inhibition was determined from the nonlinear regression analysis, which confirmed mixed, non-competitive mode of AChE inhibition (p < 0.05), consistent with graphical representation of corresponding Lineweaver-Burk plot in Figure 2, where all of the lines intersect on the one point of the x-axis.  [51], can be highlighted. Previously prepared compound 1 [31] showed sub-micromolar and micromolar value inhibitory activity toward rat AChE and rat BChE, respectively. Hybrid 2 [42] exhibited sub-micromolar inhibitory activity toward EeAChE and hAChE and micromolar eqBChE inhibitory activity. Although these compounds were tested on different type of enzyme than novel 6-Cl-THA-scutellarin hybrids, they displayed in general better inhibitory potency against hAChE.

Kinetic Analysis
To determine the interaction of K1383 with hAChE, a kinetic study was carried out. Enzyme rate curves were obtained at several substrate concentrations of the inhibitor. The type of inhibition was determined from the nonlinear regression analysis, which confirmed mixed, non-competitive mode of AChE inhibition (p < 0.05), consistent with graphical representation of corresponding Lineweaver-Burk plot in Figure 2, where all of the lines intersect on the one point of the x-axis. Inhibitors bind reversibly to both free AChE and AChE-substrate complex with equal affinity (K i = K i ) thus affecting the binding of substrate in the active site of enzyme and interacting with its allosteric peripheral anionic site (PAS). Interaction in this site causes conformational changes of the enzyme, including its active site. K m remained unchanged, while V max was reduced at higher concentration of K1383. The corresponding K i value of 2.75 ± 0.32 nM was calculated from measured data of the tested inhibitor.

Antioxidant Activity
To determine the antioxidant activity of novel 6-Cl-THA-scutellarin hybrids, we used the 1,1-diphenyl-2-picrylhydrazyl (DPPH) assay [52] with Trolox as positive control and 6-Cl-THA, 1 and 2 as standards. As shown in Table 2, the results suggest that all selected hybrids exhibited weak antioxidant activity (EC 50 higher than 10 −4 M) compared to Trolox, 1 and 2. The free phenolic group of 1 is presumably able to easily undergo keto-enol tautomerism which has previously been postulated to be responsible for radical scavenging [53]. Within the family of 6-Cl-THA-scutellarin hybrids, all the phenolic hydroxyls of scutellarin moiety are hindered by methyl containing methoxy groups and thus unable to exert such an effect under in vitro conditions. However, the antioxidant activity of K1383-K1389 in vivo may occur due to metabolism of the compound whereby demethylation of one or more methoxy groups can take place. This is known to be the case of 7-methoxytacrine and its major metabolite found in vivo, 7-hydroxytacrine [54].
16.20 ± 0.42 b a Results are expressed as µM of Trolox/µM of tested compound as the mean of at least three experiments determined by the Oxygen Radicals Absorbance Capacity by Fluorescence (ORAC-FL) [55] from reference [31] for compound 1 and from reference [42] for compound 2; b Results are expressed as the mean of at least three experiments.

Hepatotoxicity
In consequence of the hepatotoxicity of tacrine and some of its derivatives a 3-(4,5-dimethylthiazol-2-yl)-2,5-diphenyltetrazolium bromide (MTT) assay [56] was carried out to determine the potential hepatotoxicity of novel 6-Cl-THA-scutellarin hybrids K1383-K1389 [57]. Cytotoxic effects were assayed on human liver carcinoma (HepG2) cell lines and the results are displayed in Table 3. After 24 h of incubation, only two novel derivatives, namely K1388 and K1389, bearing seven and eight methylene linkers, showed slightly suppressed cytotoxicity compared to the reference compound 6-Cl-THA. These results should however, be taken with some precaution due to the limited solubility and unknown pharmacological profile of K1383-K1389 in vivo (ADME properties) [23].

In Vitro Blood-Brain Barrier Permeation Assay
The ability of novel derivatives from 6-Cl-THA-scutellarin subset to permeate the BBB was predicted using a parallel artificial membrane permeation assay of the BBB (PAMPA-BBB) described by Di et al. [58,59]. All the data are gathered in Table 4. Based on the obtained results, three of the tested hybrids (compounds K1384, K1388, K1389) displayed the potential to cross the BBB, whereas the rest of the series demonstrated uncertain CNS permeation in the passive diffusion model. 3.93 ± 1.00 CNS± K1387 3.31 ± 0.38 CNS± K1388 4.92 ± 0.02 CNS+ K1389 8.10 ± 3.10 CNS+ 6-Cl-THA 5.00 ± 0.45 CNS+ a Values are expressed as the mean ± SD of three independent experiments; b Prediction of BBB penetration by PAMPA-BBB assay; CNS+ (high BBB permeation predicted, P e × 10 −6 cm/s > 4.0), CNS± (BBB permeation uncertain, P e × 10 −6 cm/s from 4.0 to 2.0), CNS− (low BBB permeation predicted, P e × 10 −6 cm/s < 2.0).

Molecular Docking
Initially, it should be noted that both cholinesterases share high degrees of homology (50-60%) [60]. However, such homology is not enough to provide same degree of ligand affinity to the active site of both AChE and BChE. The BChE active site is much bulkier compared to that of AChE. This results from some the absence of aromatic residues within the active gorge of BChE and also observable structural differences near the catalytic anionic site (CAS) of BChE [61]. In addition, broadly cited PAS containing several aromatic residues and located at the rim of the cavity gorge is highly conserved and specific for AChE, whereas in PAS in the BChE active site the number of such residues is greatly reduced [62]. This is also the reason for the different inhibition efficacies of the compounds under the study, i.e., the 6-Cl-THA-scutellarin family.
Having all these prerequisites in mind, we have conducted molecular modeling studies with K1383, K1384 and K1385 in order to explore their high affinity towards hAChE or hBChE and high hAChE selectivity profile observed mainly in case of K1385. The selection of hAChE and hBChE for the computational studies is rationally in accordance with in vitro studies where the enzymes of human origin were also exploited. Firstly, we have taken hAChE (PDB ID: 4EY7) and hBChE (PDB ID: 4BDS) from the Protein Data Bank as source receptors for embedding the ligands and, ultimately, explore their binding modes via flexible ligand docking protocol [63,64].
Top-scored docking poses of K1383, K1384 and K1385 in the hAChE active site are illustrated in Figure 3. Focused on K1383 (−13.3 kcal/mol; Figure 3A,B), the ligand revealed a dual binding site pattern of inhibition with the scutellarin moiety protruding out of the cavity gorge and THA moiety stacked in the CAS of the enzyme. Such an orientation of THA scaffold is highly consistent with other recently described THA dual binders like tacrine-tianeptine or 7-methoxytacrine-adamantylamine heterodimers [65][66][67]. On contrary, it has to be stated that opposite orientation, i.e., THA lodging within PAS region, can be also found in the literature like in 7-methoxytacrine-p-anisidine, tacrine-Trolox or 7-methoxytacrine-donepezil like compounds [68][69][70]. In case of K1383, the 6-Cl-THA moiety is facing Trp86 (4.1 Å) in a π-π/cation-π parallel manner. The secondary amino group of 6-Cl-THA is further stabilized by a hydrogen bond to Gly329 (1.4 Å). Glu202 (2.5 Å) is the only amino acid residue from the catalytic triad being involved in ligand enzyme interactions, forming a hydrogen bond with a protonated nitrogen from the 6-Cl-THA template. The high enzyme affinity of K1383 is caused also by proper orientation of the chlorine atom to Tyr124. Scutellarin from K1383 is sandwiched between Trp286 (4.5 Å), a key aromatic residue of PAS, and Tyr341 (4.0 Å) by slightly deflected parallel π-π stacking. Moreover, some other hydrogen bond interactions to scutellarin can be also found with Arg525. The scutellarin-attached phenyl ring provided a π-π interaction in the mid-gorge of hAChE to Phe338. Many other hydrophobic interactions contribute to K1383-hAChE complex constriction (e.g., with Tyr124, Tyr337 or Phe297). On the other hand, Tyr337 seems to form unfavorable hydrogen bonds to the secondary amino group of 6-Cl-THA.
Very close spatial orientation to that observed between K1383 and hAChE can be found within the hAChE-K1384 complex (−13.6 kcal/mol; Figure 3C,D). More in detail, the 6-Cl-THA scaffold is proximally lodged while the scutellarin moiety is situated distally at the cavity entrance. The THA core is stacked by π-π/cation-π interactions against Trp86 (3.9 Å). Interestingly, more hydrogen contacts can be found compared to K1383-hAChE, like between the secondary amino group of 6-Cl-THA moiety and Tyr337 (2.4 Å) and Tyr428 (2.5 Å). However, Tyr341 is involved in an unfavorable hydrogen bond to the amino group of the THA moiety. Additionally, the charged nitrogen from THA supports ligand anchoring via hydrogen bonding to Glu202 (2.3 Å). Other catalytic triad residues are not involved in enzyme-ligand interactions. Contrary to the hAChE-K1383 complex, Tyr341 (3.8 Å) provided a parallel-interaction to the scutellarin-attached phenyl ring. This is presumably caused by the one methylene group longer tether and a more favorable ligand arrangement in the gorge. The scutellarin moiety faces directly to Trp286 (3.9 Å) forming a parallel π-π interaction. In general, slightly suppressed potency can be attributed to the missing interaction of the scutellarin moiety with Tyr341. Overall binding poses of K1383 and K1384 within hAChE active site are in good agreement with the docking results obtained for 2-TcAChE complex [71].
Opposite positioning of K1385 (−12.8 kcal/mol; Figure 3E,F) to K1383 and K1384 was proposed in the hAChE active site. Accordingly, the 6-Cl-THA unit is oriented by parallel π-π/cation-π stacking to Tyr72 (4.0 Å) and disordered π-π bonding to Trp286 towards the outside of the cavity, suggesting its anchoring in the PAS. Indeed, this binding pose is possible since it has been previously reported that THA itself can be a CAS as well as a PAS binder [72]. Notwithstanding the PAS orientation of 6-Cl-THA, its chlorine atom also exerted alkyl-π bonding to Tyr124 pointing out to the flexibility of this aromatic residue. This observation is in line with the findings where Tyr133, and Tyr337 in conjunction with Tyr124 regulates the flow of the substrate into the catalytic site from the entrance thus presuming high-flexibility of these residues [73]. The scutellarin moiety in K1385-hAChE complex resides centrally in CAS. Herein, scutellarin revealed a specific T-shaped π-π bonding with Phe338 and Tyr33. Trp86 is also implicated in π-π interactions to scutellarin and additionally forms a hydrogen bond with one of the methoxy groups. Ser203 is the only residue from the catalytic triad being involved in the ligand settling by van der Waals bonding.
The missing cation-π interaction together with opposite K1385 anchoring in the hAChE active site are plausibly the culprits of the slightly reduced affinity to the enzyme. This is somewhat counterbalanced by not displaying unfavorable donor-donor interaction with Tyr337 like in case of K1383-and K1384-hAChE complexes, thus the overall discrepancy in enzyme affinity is marginal. Focused on the K1383-hBChE complex (−12.8 kcal/mol, Figure 4A,B), we perceived that the 6-Cl-THA heterocycle is situated in the CAS of hBChE facing in a parallel manner towards Trp82 (4.1 Å) via π-π/cation-π. The phenyl ring of scutellarin moiety is accommodated by typical T-shaped π-π interactions with Tyr332 (3.7 Å) and simultaneously with Phe329 (4.0 Å). Two methoxy groups from scutellarin moiety form triangular bonding to Gln119 (2.7 Å and 2.6 Å). Regarding the catalytic triad, His438 contacts the cyclohexyl moiety of tacrine core via an alkyl-π interaction while Glu197 with Ser198 were tightly bound by a non-specific van der Waals hydrophobic interaction. Contrary to hAChE, several important π-π interactions to scutellarin were omitted due to "the more-aliphatic nature" of hBChE PAS. Focused on the K1383-hBChE complex (−12.8 kcal/mol, Figure 4A,B), we perceived that the 6-Cl-THA heterocycle is situated in the CAS of hBChE facing in a parallel manner towards Trp82 (4.1 Å) via π-π/cation-π. The phenyl ring of scutellarin moiety is accommodated by typical T-shaped π-π interactions with Tyr332 (3.7 Å) and simultaneously with Phe329 (4.0 Å). Two methoxy groups from scutellarin moiety form triangular bonding to Gln119 (2.7 Å and 2.6 Å). Regarding the catalytic triad, His438 contacts the cyclohexyl moiety of tacrine core via an alkyl-π interaction while Glu197 with Ser198 were tightly bound by a non-specific van der Waals hydrophobic interaction. Contrary to hAChE, several important π-π interactions to scutellarin were omitted due to "the more-aliphatic nature" of hBChE PAS. Opposite ligand settling can be observed for K1384-hBChE (−11.1 kcal/mol; Figure 4C,D) complex. In this case, the scutellarin moiety is preferentially oriented toward the enzyme cavity providing π-π stacking to Trp82 (4.0 Å). More importantly, all the catalytic triad residues are involved in ligand harboring; these are conventional hydrogen bonds of His438 to a ketone and π-sigma contact to a methoxy group, van der Waals interaction with Ser198 and carbon-hydrogen bonding with Glu197. 6-Cl-THA is further stabilized by its chlorine atom via a π-alkyl connection to Phe329. Interestingly, the latter amino acid residue is also involved in a π-π interactions with the scutellarin unit. Generally, such an opposite orientation of K1384 may sufficiently explain the high affinity to hBChE, leading us to conclude that scutellarin itself is a more potent CAS binder than 6-Cl-THA. We note that our observation is in line with the data found from the in vitro tests.
Taking into account the most hAChE selective ligand under study, K1385, and its anchoring to hBChE (−11.1 kcal/mol; Figure 4E,F), several conclusions justifying its low hBChE inhibitory activity Opposite ligand settling can be observed for K1384-hBChE (−11.1 kcal/mol; Figure 4C,D) complex. In this case, the scutellarin moiety is preferentially oriented toward the enzyme cavity providing π-π stacking to Trp82 (4.0 Å). More importantly, all the catalytic triad residues are involved in ligand harboring; these are conventional hydrogen bonds of His438 to a ketone and π-sigma contact to a methoxy group, van der Waals interaction with Ser198 and carbon-hydrogen bonding with Glu197. 6-Cl-THA is further stabilized by its chlorine atom via a π-alkyl connection to Phe329. Interestingly, the latter amino acid residue is also involved in a π-π interactions with the scutellarin unit. Generally, such an opposite orientation of K1384 may sufficiently explain the high affinity to hBChE, leading us to conclude that scutellarin itself is a more potent CAS binder than 6-Cl-THA. We note that our observation is in line with the data found from the in vitro tests.
Taking into account the most hAChE selective ligand under study, K1385, and its anchoring to hBChE (−11.1 kcal/mol; Figure 4E,F), several conclusions justifying its low hBChE inhibitory activity can be drawn. Firstly, the scutellarin moiety is accommodated into CAS like in K1384-hBChE, however, in slightly diverse manner. We found parallel π-π stacking with Trp82 (3.9 Å) and also with Phe329. In this case, Phe329 does not connect to the chlorine atom from the THA core. On the contrary, chlorine forms a π-alkyl contact with His438 from the catalytic triad. Other members of catalytic triad (Ser198, Glu197) similarly provide ligand bonding to scutellarin like in case of K1384-hBChE. More importantly, the four carbon chain between the two moieties is presumably too long to efficiently contact CAS and PAS concomitantly, thus causing the 6-Cl-THA moiety to jut out of from the active site.
Our computational studies confirmed that the relationship between hBChE and derivatives from the 6-Cl-THA-scutellarin subset involves a lower affinity to the hBChE active site. This finding, supported by in vitro data, can be rationalized by the primary structure of both cholinesterases where six of these residues in hBChE are changed to smaller ones, including Tyr 72 to Asn 68, Tyr 124 to Gln 119, Trp 286 to Ala 277, Phe 295 to Leu286, Phe 297 to Val 288, and Tyr 337 to Ala 328 [74]. These smaller residues significantly enlarge the entrance radius, which makes it possible for bulkier substrates to enter the active site.

General Information
All reagents were of reagent grade quality obtained from Sigma-Aldrich (Prague, Czech Republic). Solvents for synthesis were obtained from Penta Chemicals Co. (Chrudim, Czech Republic). All experiments were carried out under nitrogen atmospheres. Thin layer chromatography (TLC) was performed on aluminium sheets precoated with silica gel 60 F254 (Merck, Prague, Czech Republic) and then visualized by UV 254 nm and 366 nm. Column chromatography was performed at normal pressure on silica gel 100 (particle size 0.063-0.200 mm, 70-230 mesh ASTM, Fluka, Prague, Czech Republic). Mass spectra were recorded using a combination of high performance liquid chromatography (HPLC) and mass spectrometry (MS). The Dionex Ultimate 3000 LC-MS analytical system was connected with an Orbitrap Q Exactive Plus hybrid spectrometer (Thermo Fisher Scientific, Bremen, Germany). 1 H-NMR and 13 C-NMR spectra were recorded with a Varian Mercury VX BB 300 (operating at 300 MHz for 1 H and 75 MHz for 13 C) or on a Varian S500 spectrometer operating at 500 MHz for 1 H and 126 MHz for 13 C (Varian Inc. Palo Alto, CA, USA). Chemical shifts are reported in parts per million (ppm, δ) relative to TMS. The assignment of chemical shifts is based in standard NMR experiments ( 1 H, 13

Inhibition of Human AChE and BChE
The AChE and BChE inhibitory activities of the tested compounds were determined using a modified Ellman's method [48,49] and were expressed as IC 50 values, i.e., the concentration of compound required for 50% reduction in cholinesterase activity. Human erythrocyte hemolyzate, prepared from blood samples, was applied as a source of AChE (EC 3.1.1.7). The blood samples were collected from healthy volunteers from the vein into a disposable syringe containing 3.8% sodium citrate solution (the ratio/citrate was 1:10 w/w). . Assayed solutions of target compounds (10 µL, 10 −3 -10 −9 M) were preincubated with enzyme for 5 min. The reaction was initiated by addition of 20 µL of substrate. The activity was determined by measuring of the increase in absorbance at 412 nm at 37 • C in 2 min intervals using a Synergy 2 multimode microplate reader (BioTek Instruments, Winooski, VT, USA). Each concentration was assayed in triplicate. Percentage of inhibition (I) was calculated from the measured data as follows Equation (1): ∆A i indicates absorbance change provided by cholinesterase exposed to tested compound. ∆A 0 indicates absorbance change caused by intact cholinesterase, where phosphate bufffer was applied in the same was as the anticholinesterase compound. Software Microsoft Excel (Redmond, WA, USA) and GraphPad Prism version 5.02 for Windows (GraphPad Software, San Diego, CA, USA) (www.graphpad.com) were used for the statistical data evaluation. The IC 50 values are expressed as a mean ± SEM.

Kinetic Study of AChE Inhibition
The kinetic study of AChE inhibition was performed using Ellman's method [48]. Following concentrations of substrate were used for the measurements: 0.156, 0.313, 0.625 and 1.250 mM. The type of inhibition and corresponding inhibition constant (K i and K i ) were determined by the nonlinear regression analysis of the substrate velocity curves at different concentrations of tested compounds. Results for each type model of inhibition (competitive, uncompetitive, mixed and mixed-non-competitive) were compared with sum-of-squares F-test. Lineweaver-Burk (double-reciprocal) plot, used for visualization of the enzyme-inhibitor interactions, was calculated using weighted least-squares regression. All calculations were performed using the GraphPad Prism software.

Evaluation of Antioxidant Activity
Diphenyl-1-picrylhydrazyl stable free radical assay (DPPH) [52] is a simple method to determine antioxidant activity and is expressed as EC 50 , that is the concentration of compound that causes a 50% decrease in the DPPH activity. DPPH, methanol, and Trolox (as reference standard) were purchased from Sigma-Aldrich. Polystyrene Nunc 96-well microplates with flat bottom shape (Thermo Fisher Scientific) were used for measuring purposes. All of the assays were carried out in methanol. DPPH solution was prepared at 0.2 mM concentration. The assay medium (200 µL) consisted of 100 µL of DPPH solution and 100 µL of tested compound (10 −3 -10 −6 M). The reaction time constituted 30 min. The antioxidant activity was determined by measuring the increase in absorbance at 517 nm at laboratory temperature using Synergy 2 multimode microplate reader [75]. Each concentration was tested in triplicate. The software GraphPad Prism version 5 for Windows was used for statistical data evaluation.

Determination of in vitro Blood-Brain Barrier Permeation
The parallel artificial membrane permeation assay (PAMPA) has been used based on reported protocol [58,59]. The filter membrane of the donor plate was coated with PBL (Polar Brain Lipid, Avanti, Alabaster, AL, USA) in dodecane (4 µL of 20 mg/mL PBL in dodecane) and the acceptor well was filled with 300 µL of PBS pH 7.4 buffer (V D ). Tested compounds were dissolved first in DMSO and then diluted with PBS pH 7.4 to reach the final concentration 100 µM in the donor well. The concentration of DMSO did not exceed 0.5% (v/v) in the donor solution. 300 µL of the donor solution was added to the donor wells (V A ) and the donor filter plate was carefully put on the acceptor plate so that coated membrane was "in touch" with both donor solution and acceptor buffer. Test compound diffused from the donor well through the lipid membrane (Area = 0.28 cm 2 ) to the acceptor well. The concentration of the drug in both donor and the acceptor wells was assessed after 3, 4, 5 and 6 h of incubation in quadruplicate using the Synergy HT UV plate reader at the maximum absorption wavelength of each compound. Concentration of the compounds was calculated from the standard curve and expressed as the permeability (Pe) according to Equation (2) [76,77]: 3.9.5. Determination of Hepatotoxicity on HepG2 Cells Hepatotoxicity of tested compounds was evaluated using the HepG2 cell line originally from human liver hepatocellular carcinoma (ATCC, Manassas, VA, USA). These cells were plated at density 17 × 10 3 per well in a 96-well plate in Dulbecco's Modified Eagle's Medium (DMEM; Gibco, Langley, OK, USA) with 10% FBS (Gibco) and were left to attach overnight. The incubation was performed under condition 37 • C, 5% CO 2 and 80-95% air humidity. Stock solutons of tested compounds were prepared in DMSO (Sigma Aldrich, Prague, Czech Republic) and then diluted in DMEM medium. The final concentration of DMSO was less than 0.25%. The cell viability was detected using (3-4,5-dimethylthiazol-2-yl)-2,5-diphenyltetrazolium bromide (MTT; Sigma Aldrich) assay after 24 h of incubation with compounds [56]. Subsequently, the medium was aspirated and 100 µL MTT solution (0.5 mg/mL) in serum-free DMEM medium was added to each well. Cells were then incubated for 1 h. The medium was then discarded and purple crystals of MTT formazan were dissolved in 100 µL DMSO under shaking. The absorbance was measured with a microplate reader (BioTek) at a minimum wavelength of 570 nm. The IC 50 values were calculated using four parametric nonlinear regression with statistic software GraphPad Prism. Data were obtained from three independent experiments performed in triplicates.

Molecular Modeling Studies
Two structures of hAChE and hBChE were obtained from the RCSB Protein Data Bank-PDB ID: 4EY7 (crystal structure of hAChE) and 4BDS (crystal structure of hBChE) [63,64]. All receptor structures were prepared by DockPrep function of UCSF Chimera (version 1.4) and converted to pdbqt-files by AutodockTools (v. 1.5.6) [78,79]. Flexible residues selection was based on previous experience (hAChE) or spherical region around the binding cavity [80][81][82][83]. Three-dimensional structures of ligands were built by Open Babel (v. 2.3.1), minimized by Avogadro (v 1.1.0) and converted to pdbqt-file format by AutodockTools [84]. The docking calculations were done by Autodock Vina (v. 1.1.2) with the exhaustiveness of 8 [85]. Calculation was repeated ten times for each ligand and receptor and the best-scored result was selected for manual inspection. The visualization of enzyme-ligand interactions was prepared using 1.

Conclusions
While the exact cause(s) of Alzheimer's disease remain unknown, new therapies aimed at a number of biological targets have been identified to design new therapies. In this context, we report the design, synthesis and pharmacological profiles of some novel multi-functional anti-AD agents K1383-K1389 containing 6-Cl-THA and scutellarin fragments. These derivatives were proven to be potent ChE inhibitors exerting selective AChE dual binding site character. Several of novel hybrids K1383-K1385 were better hAChE inhibitors in relation to a reference standard, 6-Cl-THA. The most prominent hybrid K1383 was found to be a very potent hAChE inhibitor (IC 50 = 1.63 nM). The AChE kinetic analysis revealed a mixed, non-competitive mode of AChE inhibition. Although designed as potential antioxidants, the free scavenging ability was lost (EC 50 higher than 10 −4 M) by the hybridization. A probable reason is the presence of methoxy groups in the scutellarin moiety instead of free phenolic groups which could presumably confer antioxidant potential. An important issue that needs to be addressed in the development of novel tacrine-based MTDLs is their potential hepatotoxicity. Our preliminary data verified on HepG2 cell line suggested a comparable hepatotoxic profile of K1383-K1389 as the standard 6-Cl-THA. Moreover, the PAMPA assay predicted a probability of some hybrids, namely K1384, K1388, K1389, to cross the BBB while the remainder demonstrated rather uncertain CNS permeation via a passive diffusion model. This work presents a novel tacrine-based family as potential anti-AD agents, but further in vitro and in vivo testing are needed to fully established their real perspectives.
Author Contributions: Katarina Spilovska and Jan Korabecny carried out the synthesis, spectra analysis, interpretation and wrote the manuscript. Jan Korabecny and Tomas Kucera performed molecular modeling study. Vendula Sepsova performed the cholinesterase inhibition. Daniel Jun performed the kinetic study. Martina Hrabinova performed the antioxidant analysis. Lubica Muckova and Petr Jost performed the hepatotoxicity analysis. Jana Janockova and Ondrej Soukup performed the PAMPA assay. Rafael Dolezal performed HRMS analysis. Eva Mezeiova performed spectra analysis. Daniel Kaping and Kamil Kuca supervised the study. All authors contributed to and have approved the final manuscript.

Conflicts of Interest:
The authors declare no conflict of interest.