Inhibiting Acetylcholinesterase to Activate Pleiotropic Prodrugs with Therapeutic Interest in Alzheimer’s Disease

Alzheimer’s disease (AD) is a multifactorial neurodegenerative disease which is still poorly understood. The drugs currently used against AD, mainly acetylcholinesterase inhibitors (AChEI), are considered clinically insufficient and are responsible for deleterious side effects. AChE is, however, currently receiving renewed interest through the discovery of a chaperone role played in the pathogenesis of AD. But AChE could also serve as an activating protein for pleiotropic prodrugs. Indeed, inhibiting central AChE with brain-penetrating designed carbamates which are able to covalently bind to the enzyme and to concomitantly liberate active metabolites in the brain could constitute a clinically more efficient approach which, additionally, is less likely to cause peripheral side effects. We aim in this article to pave the road of this new avenue with an in vitro and in vivo study of pleiotropic prodrugs targeting both the 5-HT4 receptor and AChE, in order to display a neuroprotective activity associated with a sustained restoration of the cholinergic neurotransmission and without the usual peripheral side effects associated with classic AChEI. This plural activity could bring to AD patients effective, relatively safe, symptomatic and disease-modifying therapeutic benefits.


Introduction
Acetylcholinesterase (AChE) is the main target of the currently marketed drugs against Alzheimer's Disease (AD). Inhibiting AChE and the physiological degradation of acetylcholine (ACh), these drugs aim at restoring the cholinergic neurotransmission impaired by the neurodegeneration which induces the cognitive troubles associated with AD. These medicines are controversial today, mainly due to their loss of efficacy over time. AChE, however, could benefit from renewed interest as a target for novel anti-AD drugs for two reasons: The first concerns the discovery of a new chaperone role played by AChE which would be able to interact, thanks to its peripheral anionic site (PAS), with the β-amyloid (Aβ) peptide and to form neurotoxic aggregates with the latter [1]. Inhibiting this interaction, if possible in a concomitant manner with the catalytic inhibition of AChE, would yield a dual therapeutic benefit against AD.
The second reason is inherent to the explanation which can be given for the loss of efficiency of its inhibitors. The latter can be imputed to the evolution of the disease and to the consequent neurodegeneration which leads to a decrease in the production of AChE, whose inhibition becomes less The second reason is inherent to the explanation which can be given for the loss of efficiency of its inhibitors. The latter can be imputed to the evolution of the disease and to the consequent neurodegeneration which leads to a decrease in the production of AChE, whose inhibition becomes less and less useful over time. However, preserving the functional status of the cholinergic neurons could make AChE an efficient target in AD treatment once again. This explains why a lot of clinical trials assessing potential of novel neuroprotective agents are often done in association with donepezil, the most used AChE inhibitors (AChEI), in order to display a synergistic effect [2]. Alternatively, another approach was recently proposed. It consists of the design of Multi-Target Directed Ligands (MTDL) which are able to both inhibit AChE and to express a neuroprotective effect through an interaction with another target involved in the pathogenesis of AD [3]. Our group recently described, within this framework, the in vitro and in vivo properties of donecopride, the first MTDL able to both inhibit AChE and to activate 5-HT4 receptors (5-HT4R) with potent antiamnesic effects in animal models of AD [4][5].
These considerations have led to a revival of interest in AChE as an efficient therapeutic target. Furthermore, this enzyme can be used to reduce the peripheral side effects displayed by novel anti-AD drugs. Indeed, the latter can be specifically liberated at the central level through hydrolysis by AChE of brain-penetrating prodrugs. AChE, being inhibited during this hydrolysis, can be considered in this case both as a therapeutic target and as an activating enzyme (Figure 1). A unique example of such a pleiotropic anti-AD prodrug is currently reported in literature. Ladostigil (1) is a carbamate compound which is able to covalently bind to AChE and to liberate a hydroxy derivative of rasagiline, displaying Monoamine Oxidase (MAO)-B inhibitory activity ( Figure 2) [6][7][8][9].
Concerning the present work, we wanted to develop an innovative approach through the design of novel pleiotropic prodrugs which were likely to covalently inhibit AChE and then to liberate active metabolites related to donecopride (2) while keeping its potent 5-HT4R agonist and Dual Binding Site (DBS) AChE inhibitory activities. These compounds will be designed based on the model of rivastigmine (3) with which they are structurally related. Rivastigmine is a carbamate compound which covalently binds and temporarily inhibits AChE. Rivastigmine, which is the only marketed pseudo-irreversible AChEI, consequently restores cholinergic neurotransmission, then alleviating the cognitive symptoms in AD patients. However, the phenolic derivative, released after the carbamoylation of AChE by rivastigmine, is devoid of any therapeutic activity.
The novel phenolic derivatives (4)(5)(6) of donecopride which we propose herein were synthesized and evaluated as potential 5-HT4R ligands and DBS AChEI. They were carbamoylated into carbamates (7)(8)(9). The capacity of the latter to inhibit AChE in a covalent manner was investigated. In order to avoid possible peripheral side effects, their lack of other activities was verified. The ability of a selected carbamate to liberate the corresponding phenolic derivative upon AChE-dependent hydrolysis has been studied, as well as its capacity to easily cross the blood brain barrier (BBB). A unique example of such a pleiotropic anti-AD prodrug is currently reported in literature. Ladostigil (1) is a carbamate compound which is able to covalently bind to AChE and to liberate a hydroxy derivative of rasagiline, displaying Monoamine Oxidase (MAO)-B inhibitory activity ( Figure 2) [6][7][8][9].
Concerning the present work, we wanted to develop an innovative approach through the design of novel pleiotropic prodrugs which were likely to covalently inhibit AChE and then to liberate active metabolites related to donecopride (2) while keeping its potent 5-HT 4 R agonist and Dual Binding Site (DBS) AChE inhibitory activities. These compounds will be designed based on the model of rivastigmine (3) with which they are structurally related. Rivastigmine is a carbamate compound which covalently binds and temporarily inhibits AChE. Rivastigmine, which is the only marketed pseudo-irreversible AChEI, consequently restores cholinergic neurotransmission, then alleviating the cognitive symptoms in AD patients. However, the phenolic derivative, released after the carbamoylation of AChE by rivastigmine, is devoid of any therapeutic activity.
The novel phenolic derivatives (4-6) of donecopride which we propose herein were synthesized and evaluated as potential 5-HT 4 R ligands and DBS AChEI. They were carbamoylated into carbamates (7)(8)(9). The capacity of the latter to inhibit AChE in a covalent manner was investigated. In order to avoid possible peripheral side effects, their lack of other activities was verified. The ability of a selected carbamate to liberate the corresponding phenolic derivative upon AChE-dependent hydrolysis has been studied, as well as its capacity to easily cross the blood brain barrier (BBB). Finally, in vivo tests have been performed in mice in order to potentially highlight, for this pleiotropic prodrug, a procognitive effect. Finally, in vivo tests have been performed in mice in order to potentially highlight, for this pleiotropic prodrug, a procognitive effect.  (3), phenolic derivatives (4-6) of Donecopride and targeted pleiotropic prodrugs (7-9).

Chemistry
The targeted phenolic derivatives of donecopride (4)(5)(6) and their carbamates (7-9) were obtained starting from 4-amino-5-chloro-2-methoxybenzoic acid (10) in three series: ketone, ester and amide. In the ketone series, the β-ketoester (11) was synthesized through the carbonyldiimidazole (CDI) activation of the carboxylic acid group of 10 followed by treatment with the potassium salt of ethylmalonate (Scheme 1). The methylpiperidine moiety was then installed through nucleophilic substitution using N-Boc 4-(iodomethyl)piperidine carboxylate at room temperature to avoid the risk of a double substitution, immediately followed by a saponification−decarboxylation sequence with hydroalcoholic potassium hydroxide, which gave 12 [5]. In the ester series, the analog compound 13 was obtained by esterification of 10, with N-Boc 4-hydroxymethylpiperidine, while in the amide series, the analog compound 14 was obtained through a peptidic coupling involving 10 and N-Boc 4aminomethylpiperidine.   (3), phenolic derivatives (4-6) of Donecopride and targeted pleiotropic prodrugs (7-9).

Chemistry
The targeted phenolic derivatives of donecopride (4)(5)(6) and their carbamates (7-9) were obtained starting from 4-amino-5-chloro-2-methoxybenzoic acid (10) in three series: ketone, ester and amide. In the ketone series, the β-ketoester (11) was synthesized through the carbonyldiimidazole (CDI) activation of the carboxylic acid group of 10 followed by treatment with the potassium salt of ethylmalonate (Scheme 1). The methylpiperidine moiety was then installed through nucleophilic substitution using N-Boc 4-(iodomethyl)piperidine carboxylate at room temperature to avoid the risk of a double substitution, immediately followed by a saponification−decarboxylation sequence with hydroalcoholic potassium hydroxide, which gave 12 [5]. In the ester series, the analog compound 13 was obtained by esterification of 10, with N-Boc 4-hydroxymethylpiperidine, while in the amide series, the analog compound 14 was obtained through a peptidic coupling involving 10 and N-Boc 4-aminomethylpiperidine. Finally, in vivo tests have been performed in mice in order to potentially highlight, for this pleiotropic prodrug, a procognitive effect.  (3), phenolic derivatives (4-6) of Donecopride and targeted pleiotropic prodrugs (7-9).

Chemistry
The targeted phenolic derivatives of donecopride (4)(5)(6) and their carbamates (7-9) were obtained starting from 4-amino-5-chloro-2-methoxybenzoic acid (10) in three series: ketone, ester and amide. In the ketone series, the β-ketoester (11) was synthesized through the carbonyldiimidazole (CDI) activation of the carboxylic acid group of 10 followed by treatment with the potassium salt of ethylmalonate (Scheme 1). The methylpiperidine moiety was then installed through nucleophilic substitution using N-Boc 4-(iodomethyl)piperidine carboxylate at room temperature to avoid the risk of a double substitution, immediately followed by a saponification−decarboxylation sequence with hydroalcoholic potassium hydroxide, which gave 12 [5]. In the ester series, the analog compound 13 was obtained by esterification of 10, with N-Boc 4-hydroxymethylpiperidine, while in the amide series, the analog compound 14 was obtained through a peptidic coupling involving 10 and N-Boc 4aminomethylpiperidine.  Having been selected for in vivo studies, the fumaric acid salts of phenol 4 and its carbamate 7, (19,20) were synthesized using fumaric acid in iPrOH (Scheme 4). In parallel, the bromobenzyl carbamate (15) was synthesized through the following sequence (Scheme 2). 3-Hydroxybenzaldehyde (16) was carbamoylated using ethylmethylcarbamic chloride to yield the carbamate 17 [10]. The aldehyde group of the latter was reduced by NaBH4, and the alcohol derivative 18 was brominated by PBr3 to give 15. The targeted compounds were finally obtained through the TFA-N-deprotection of 12-14, immediately followed by another nucleophilic substitution with 3-iodomethylphenol [11] to yield phenols 4-6 or with compound 15 to form carbamates 7-9 (Scheme 3). Having been selected for in vivo studies, the fumaric acid salts of phenol 4 and its carbamate 7, (19,20) were synthesized using fumaric acid in iPrOH (Scheme 4).

In Silico Results
During in silico studies, we evaluated whether carbamate derivative 7 could access the (h)AChE catalytic triad, specially the Ser203 residue playing a crucial role during carbamoylation. According to the proposed mechanism of rivastagmine carbamoylation [12], the carbamate-to-serine grouping approach starts via a hydrogen bond between the oxygen atom of the carbamate carbonyl group and

In Silico Results
During in silico studies, we evaluated whether carbamate derivative 7 could access the (h)AChE catalytic triad, specially the Ser203 residue playing a crucial role during carbamoylation. According to the proposed mechanism of rivastagmine carbamoylation [12], the carbamate-to-serine grouping approach starts via a hydrogen bond between the oxygen atom of the carbamate carbonyl group and the hydroxy group of Ser203. Therefore, during the docking study of compound 7 into the (h)AChE active site (PDB code: 4EY7 [13]), performed using the Gold software (v5.7.2, Cambridge Crystallographic Data Center, CCDC), a hydrogen bond constraint between carbamate carbonyl and hydroxy group of Ser203 was applied. In parallel, the Ser203 side chain was kept flexible during the docking. The generated poses of compound 7 with scores between 85.44 and 100.64 could be divided in two major clusters which differ principally in the position of the methoxy phenyl moiety and phenyl one. In cluster 1 (see Figure 3B), the carbonyl group established a hydrogen bond with Ser293, and in cluster 2, a π-stacking between compound 7 phenol ring and the Tyr 337 one was observed, as well as a possible interaction through a hydrogen bond between the NH 2 group of compound 7 and C=O of the Trp286 backbone (see Figure 3C).
Compound 7 docks in the AChE active site with the carbamate group in proximity of Ser203, as imposed by the constraint, but it is positioned much higher in the AChE binding site in two clusters compared to the donepezil and/or the donecopride (Figure 3) [5][6][7][8][9][10][11][12][13]. Consequently, according to our molecular modeling results, compound 7 loses the characteristic interactions of the donecopride ligands family, i.e.,: i) interaction between NH + of piperidine ring through the water molecule lying between the hydroxy groups of two tyrosines (Tyr341 and Tyr337); ii) interaction via C=O group with the backbone NH of Phe295; and iii) the methoxyphenyl moiety is placed a little high but stacking with Trp286 remains possible. On the other hand, the C=O group of compound 7 establishes a new interaction through a hydrogen bond with the hydroxy group of Ser293 in cluster 1. So, the 'donecopride' part attached to carbamate of compound 7 brings few interactions in addition to carbamate group ones, and the interaction with the AChE binding site is realized mainly thanks to the carbamate group. the hydroxy group of Ser203. Therefore, during the docking study of compound 7 into the (h)AChE active site (PDB code: 4EY7 [13]), performed using the Gold software (v5.7.2, Cambridge Crystallographic Data Center, CCDC), a hydrogen bond constraint between carbamate carbonyl and hydroxy group of Ser203 was applied. In parallel, the Ser203 side chain was kept flexible during the docking. The generated poses of compound 7 with scores between 85.44 and 100.64 could be divided in two major clusters which differ principally in the position of the methoxy phenyl moiety and phenyl one. In cluster 1 (see Figure 3B), the carbonyl group established a hydrogen bond with Ser293, and in cluster 2, a π-stacking between compound 7 phenol ring and the Tyr 337 one was observed, as well as a possible interaction through a hydrogen bond between the NH2 group of compound 7 and C=O of the Trp286 backbone (see Figure 3C). Compound 7 docks in the AChE active site with the carbamate group in proximity of Ser203, as imposed by the constraint, but it is positioned much higher in the AChE binding site in two clusters compared to the donepezil and/or the donecopride ( Figure 3) [5][6][7][8][9][10][11][12][13]. Consequently, according to our molecular modeling results, compound 7 loses the characteristic interactions of the donecopride ligands family, i.e.: i) interaction between NH + of piperidine ring through the water molecule lying between the hydroxy groups of two tyrosines (Tyr341 and Tyr337); ii) interaction via C=O group with the backbone NH of Phe295; and iii) the methoxyphenyl moiety is placed a little high but stacking with Trp286 remains possible. On the other hand, the C=O group of compound 7 establishes a new interaction through a hydrogen bond with the hydroxy group of Ser293 in cluster 1. So, the 'donecopride' part attached to carbamate of compound 7 brings few interactions in addition to carbamate group ones, and the interaction with the AChE binding site is realized mainly thanks to the carbamate group.

AChE Inhibition and 5-HT4R Binding.
All the synthesized phenolic derivatives and their carbamates were evaluated as potential inhibitors of human AChE, using the Ellman assay [14], as well as potential ligands for human 5-HT4R using a radioligand displacement assay [15]. In these tests, Donepezil and Rivastigmine (3) were used as reference AChEI, and RS67333 as a reference 5-HT4R agonist [5]. The results are depicted in Table  1.
All the carbamates appeared able to inhibit AChE with IC50 values in the same range as rivastigmine. The keto derivative 7 exhibited the best activity with an IC50 value of 4.15 µM which was recovered with its fumaric salt, 20. The phenolic derivatives in the ester (5) and amide (6) series

AChE Inhibition and 5-HT 4 R Binding
All the synthesized phenolic derivatives and their carbamates were evaluated as potential inhibitors of human AChE, using the Ellman assay [14], as well as potential ligands for human 5-HT 4 R using a radioligand displacement assay [15]. In these tests, Donepezil and Rivastigmine (3) were used as reference AChEI, and RS67333 as a reference 5-HT 4 R agonist [5]. The results are depicted in Table 1. All the carbamates appeared able to inhibit AChE with IC 50 values in the same range as rivastigmine. The keto derivative 7 exhibited the best activity with an IC 50 value of 4.15 µM which was recovered with its fumaric salt, 20. The phenolic derivatives in the ester (5) and amide (6) series appeared to be devoid of any AChE inhibitory activities, unlike their analog 4 in keto series displaying a noticeable AChEI activity which turned to be even better for its fumaric salt 19 (respectively, IC 50 value of 148 and 72 nM). Concerning 5-HT 4 R affinity, carbamates 7-9 and 20 appeared to be devoid of such activity, while their phenolic analogs 4-6 and 19 appeared to be potent ligands with Ki values in the same range as RS67333 (5 nM), or even decreased for the ester derivative 5 (0.6 nM). appeared to be devoid of any AChE inhibitory activitie,s, unlike their analog 4 in keto series displaying a noticeable AChEI activity which turned to be even better for its fumaric salt 19 (respectively, IC50 value of 148 and 72 nM). Concerning 5-HT4R affinity, carbamates 7-9 and 20 appeared to be devoid of such activity, while their phenolic analogs 4-6 and 19 appeared to be potent ligands with Ki values in the same range as RS67333 (5 nM), or even decreased for the ester derivative 5 (0.6 nM).

Pharmacological Profile Results
The pharmacological profile of the selected phenolic derivative 19 was first established towards (h)5-HT4R. It acts as a partial agonist in a similar manner as RS67333 ( Figure 4 and Table 2).

Pharmacological Profile Results
The pharmacological profile of the selected phenolic derivative 19 was first established towards (h)5-HT 4 R. It acts as a partial agonist in a similar manner as RS67333 ( Figure 4 and Table 2).
On the other hand, the mechanism of AChE inhibition for compounds 19 and 20 was evaluated by means of a kinetic study, the results of which are reported in typical Lineweaver-Burk plots. Phenol 19 acts as a non-competitive inhibitor, accounting for a possible interaction of 19 with the PAS of the enzyme ( Figure 5A). At the same time, the carbamate 20 showed a mixed-type inhibition, illustrated by a typical Lineweaver-Burk plot similar to those obtained with the covalent AChEI, rivastigmine ( Figure 5B) [18].
Molecules 2019, 24, x 6 of 20 appeared to be devoid of any AChE inhibitory activitie,s, unlike their analog 4 in keto series displaying a noticeable AChEI activity which turned to be even better for its fumaric salt 19 (respectively, IC50 value of 148 and 72 nM). Concerning 5-HT4R affinity, carbamates 7-9 and 20 appeared to be devoid of such activity, while their phenolic analogs 4-6 and 19 appeared to be potent ligands with Ki values in the same range as RS67333 (5 nM), or even decreased for the ester derivative 5 (0.6 nM).

Pharmacological Profile Results
The pharmacological profile of the selected phenolic derivative 19 was first established towards (h)5-HT4R. It acts as a partial agonist in a similar manner as RS67333 ( Figure 4 and Table 2).     On the other hand, the mechanism of AChE inhibition for compounds 19 and 20 was evaluated by means of a kinetic study, the results of which are reported in typical Lineweaver-Burk plots. Phenol 19 acts as a non-competitive inhibitor, accounting for a possible interaction of 19 with the PAS of the enzyme ( Figure 5A). At the same time, the carbamate 20 showed a mixed-type inhibition, illustrated by a typical Lineweaver-Burk plot similar to those obtained with the covalent AChEI, rivastigmine ( Figure 5B) [18].
A B Figure 5. Lineweaver-Burk plots of inhibition kinetics show that 19 acts as a non-competitive AChE inhibitor (A) and 20 as a mixed-type AChE inhibitor (B).

Brain Penetration
The ability of carbamate 20 to cross the BBB was assessed using a parallel artificial membrane permeability assay experiment. The compound was classified among the compounds having good brain penetration with logPe = −4.39 (Pe represents the PAMPA effective permeability coefficient). The test was performed using a good (corticosterone) and weak brain-penetrating reference (theophylline), respectively (Table 3).

AChE-Dependent Decarbamoylation
The in vitro decarbamoylation of 20 in the presence of AChE was established according to a new analytical method developed by Alix et al [19]. A large excess of Electrophorus electricus (eel)AChE was added to a solution of 20 in a phosphate-buffered saline (PBS) buffer (pH 7.4) at 25 °C. After 24 h of incubation, the solution was extracted by ethyl acetate, evaporated and solubilized in acetonitrile, and then analyzed by UPLC and LCMS. Owing to its high homology with the active-site sequence of

Brain Penetration
The ability of carbamate 20 to cross the BBB was assessed using a parallel artificial membrane permeability assay experiment. The compound was classified among the compounds having good brain penetration with logPe = −4.39 (Pe represents the PAMPA effective permeability coefficient). The test was performed using a good (corticosterone) and weak brain-penetrating reference (theophylline), respectively (Table 3).

AChE-Dependent Decarbamoylation
The in vitro decarbamoylation of 20 in the presence of AChE was established according to a new analytical method developed by Alix et al. [19]. A large excess of Electrophorus electricus (eel)AChE was added to a solution of 20 in a phosphate-buffered saline (PBS) buffer (pH 7.4) at 25 • C. After 24 h of incubation, the solution was extracted by ethyl acetate, evaporated and solubilized in acetonitrile, and then analyzed by UPLC and LCMS. Owing to its high homology with the active-site sequence of the human AChE and its commercial availability, (ee)AChE is commonly used for in vitro assays and was chosen for this study.
According to ULPC analysis and on the basis of the retention times of 4 and 20 without (ee)AChE ( Figure 6A), the first result showed a decarbamoylation of 20 and release of 4 ( Figure 6B). The control experiment in a PBS buffer without (ee)AChE, under the same reaction conditions, was also performed to ensure that in the absence of (ee)AChE, 20 was not chemically converted into 4 ( Figure 6C). Rivastigmine was used as a positive control. These results were confirmed by mass analysis.
the human AChE and its commercial availability, (ee)AChE is commonly used for in vitro assays and was chosen for this study.
According to ULPC analysis and on the basis of the retention times of 4 and 20 without (ee)AChE ( Figure 6A), the first result showed a decarbamoylation of 20 and release of 4 ( Figure 6B). The control experiment in a PBS buffer without (ee)AChE, under the same reaction conditions, was also performed to ensure that in the absence of (ee)AChE, 20 was not chemically converted into 4 ( Figure  6C). Rivastigmine was used as a positive control. These results were confirmed by mass analysis.

In Vivo Results.
Some in vivo investigations have been performed on the selected carbamate 20 as a potential pleiotropic prodrug.

Pharmacological Screening
None of the dose tested (1, 10 and 100 mg/kg) for compound 20 showed serious side effects (Table 3). Indeed, even though only for the highest tested dose, tremors and following hypoactivity phase were noticed, all symptoms were absent after 24 h, suggesting a LD50 that is significantly higher than 100 mg/kg.

In Vivo Results
Some in vivo investigations have been performed on the selected carbamate 20 as a potential pleiotropic prodrug.

Pharmacological Screening
None of the dose tested (1, 10 and 100 mg/kg) for compound 20 showed serious side effects (Table 4). Indeed, even though only for the highest tested dose, tremors and following hypoactivity phase were noticed, all symptoms were absent after 24 h, suggesting a LD 50 that is significantly higher than 100 mg/kg.

Spontaneous Alternation Deficit
We then decided to evaluate the in vivo efficacy of compound 20 in cognition models. Whatever the pharmacological agent used to induce a cognitive deficit (scopolamine -SCOP or dizocilpine, MK801), ANOVA of percentage of spontaneous alternation revealed a group effect (respectively, F(2,23) = 7.281; p = 0.0036 and F(2,25) = 4.778; p = 0.0175) (Figures 8 and 9). In both cases, the control group displayed an alternation percentage significantly higher than animals receiving either SCOP or MK801 plus NaCl, demonstrating a pharmacologically-induced cognitive deficit (SNK, p < 0.001). Further, while the control group showed higher spontaneous alternation than animals receiving SCOP + compound 20 (SNK, p < 0.05), no statistical difference was observed compared to animals receiving MK801 + compound 20 (SNK, p = 0.0621). Additionally, a univariate t-test revealed that in the two conditions (SCOP and MK801), only the control group and animals having received compound 20 displayed an alternation percentage which was significantly different from the value of the chance level (i.e. 50%). Hence, those results point to the ability of compound 20 to offset-at least partially-working memory deficits, whether they result from cholinergic or glutamatergic neurotransmission alteration. Interestingly, these anti-amnestic effects are effective at a relatively low dose with respect to the estimated LD50.

Spontaneous Alternation Deficit
We then decided to evaluate the in vivo efficacy of compound 20 in cognition models. Whatever the pharmacological agent used to induce a cognitive deficit (scopolamine -SCOP or dizocilpine, MK801), ANOVA of percentage of spontaneous alternation revealed a group effect (respectively, F(2,23) = 7.281; p = 0.0036 and F(2,25) = 4.778; p = 0.0175) (Figures 8 and 9). In both cases, the control group displayed an alternation percentage significantly higher than animals receiving either SCOP or MK801 plus NaCl, demonstrating a pharmacologically-induced cognitive deficit (SNK, p < 0.001). Further, while the control group showed higher spontaneous alternation than animals receiving SCOP + compound 20 (SNK, p < 0.05), no statistical difference was observed compared to animals receiving MK801 + compound 20 (SNK, p = 0.0621). Additionally, a univariate t-test revealed that in the two conditions (SCOP and MK801), only the control group and animals having received compound 20 displayed an alternation percentage which was significantly different from the value of the chance level (i.e., 50%). Hence, those results point to the ability of compound 20 to offset-at least partially-working memory deficits, whether they result from cholinergic or glutamatergic neurotransmission alteration. Interestingly, these anti-amnestic effects are effective at a relatively low dose with respect to the estimated LD 50 .

Discussion
An in vitro evaluation of the three phenolic derivatives of donecopride 4-6 led us to select 4 in the keto series on the basis of its AChE inhibitory activity (IC50 = 149 nM); such activity appeared to be absent in 5 and 6. The 5-HT4R affinity of 4 (Ki = 5.1 nM) appeared to be slightly less potent than its ester analogue (Ki = 0.6 nM), but was preferred due to its dual activity. The fumaric acid salt of 4, 19, which maintains the dual activity, further expresses a partial 5-HT4R agonist pharmacological profile and a non-competitive-type AChE inhibitory profile. Consequently, it would act as a DBS AChEI. The carbamate 7, as well as its fumaric acid salt 20, derived from 4, are devoid of any activity towards 5-HT4R in a similar manner as 8 and 9. This lack of activity was sought in order to reduce the occurrence of possible peripheral side effects. Compounds 7 and 20 further inhibited AChE with a

Discussion
An in vitro evaluation of the three phenolic derivatives of donecopride 4-6 led us to select 4 in the keto series on the basis of its AChE inhibitory activity (IC50 = 149 nM); such activity appeared to be absent in 5 and 6. The 5-HT4R affinity of 4 (Ki = 5.1 nM) appeared to be slightly less potent than its ester analogue (Ki = 0.6 nM), but was preferred due to its dual activity. The fumaric acid salt of 4, 19, which maintains the dual activity, further expresses a partial 5-HT4R agonist pharmacological profile and a non-competitive-type AChE inhibitory profile. Consequently, it would act as a DBS AChEI. The carbamate 7, as well as its fumaric acid salt 20, derived from 4, are devoid of any activity towards 5-HT4R in a similar manner as 8 and 9. This lack of activity was sought in order to reduce the occurrence of possible peripheral side effects. Compounds 7 and 20 further inhibited AChE with a

Discussion
An in vitro evaluation of the three phenolic derivatives of donecopride 4-6 led us to select 4 in the keto series on the basis of its AChE inhibitory activity (IC 50 = 149 nM); such activity appeared to be absent in 5 and 6. The 5-HT 4 R affinity of 4 (Ki = 5.1 nM) appeared to be slightly less potent than its ester analogue (Ki = 0.6 nM), but was preferred due to its dual activity. The fumaric acid salt of 4, 19, which maintains the dual activity, further expresses a partial 5-HT 4 R agonist pharmacological profile and a non-competitive-type AChE inhibitory profile. Consequently, it would act as a DBS AChEI. The carbamate 7, as well as its fumaric acid salt 20, derived from 4, are devoid of any activity towards 5-HT 4 R in a similar manner as 8 and 9. This lack of activity was sought in order to reduce the occurrence of possible peripheral side effects. Compounds 7 and 20 further inhibited AChE with a similar intensity as rivastigmine (IC 50 = 4.1 µM). The kinetic study of 20, furthermore, established its covalent binding to the enzyme, and an analytic study demonstrated its concomitant AChE-dependent hydrolysis and the release of the corresponding phenolic derivative 4. Finally, the capacity of 19 to easily cross the BBB was attested using the PAMPA model.
On the basis of these predictive pharmacodynamic and pharmacokinetic profiles, 20 was selected for two in vivo studies aiming at demonstrating its potential antiamnesic effect in two different models of spontaneous alternation deficit in mice using either scopolamine or MK801 to induce cognitive impairments.
These results support our expected mechanism of action and afford preliminary proof of concept of an in vivo release of a phenolic derivative (from carbamate precursor). Hence, having irreversibly blocked acetylcholinesterase during its release from carbamate, the phenolic derivative would then be capable of modulating our serotoninergic receptors of interest (5-HT 4 receptors). Further studies will be needed to validate this, but some data from the literature reinforce this hypothesis. In particular, some studies in rodents have shown the ability of rivastigmine (which has a pharmacodynamic profile close to that of our compound 20) to counterbalance scopolamine-induced memory performance deficit in various tests [20,21]. Interestingly, this same rivastigmine, used under similar conditions, proved to be ineffective in improving memory deficits induced by MK 801 [22]. Thus, the superiority in the MK801 model of our compound relative to rivastigmine may be related to the release of the phenolic derivative having affinity and significant activity towards 5-HT 4 receptors.

General Methods
All chemical reagents and solvents were purchased from commercial sources and used without further purification. Melting points were determined on a STUART SMP50 melting point apparatus. 1 H, 13 C and 19 F NMR spectra were recorded on a BRUKER AVANCE III 400MHz with chemical shifts expressed in parts per million (in chloroform-d, methanol-d 4 or DMSO-d 6 ) downfield from TMS as an internal standard and coupling in Hertz. IR spectra were recorded on a Perkin-Elmer BX FT-IR apparatus using KBr pellets. High resolution mass spectra (HRMS) were obtained by electrospray on a BrukermaXis. The purities of all tested compounds were analyzed by LC−MS, with the purity all being higher than 95%. Analyses were performed using a Waters Alliance 2695 as separating module (column XBridge C18 2.5 µM/4.6 × 50 mM) using the following gradients: A (95%)/B (5%) to A (5%)/B (95%) in 4.00 min. This ratio was hold during 1.50 min before returning to initial conditions in 0.50 min. Initial conditions were then maintained for 2.00 min (A = H 2 O, B = CH 3 CN, each containing HCOOH: 0.1%). MS were obtained on a SQ detector by positive ESI.

In Silico Study
The initial model of compound 7 was built from donecopride X-ray structure [5], and its protonation state at pH 7.4 was predicted using standard tools of the ChemAxon Package (http: //www.chemaxon.com/). The majority microspecies protonated on piperidine nitrogen at this pH was used for docking studies.
The crystallographic coordinates of human acetylcholinesterase used in this study were obtained from X-ray structure of the donepezil/AChE complex (PDB ID 4EY7, a structure refined to 2.35 Å with an R factor of 17.7%). Structures of human acetylcholinesterase in complex with pharmacologically important ligands [13]. The AChE amino acid protonation state was checked before the docking study using the ProPKA software and the proposed protonation for Glu202 was applied.
The docking of the compound into the AChE was carried out with the GOLD program (v5.7.2, Cambridge Crystallographic Data Center, CCDC) using the default parameters [23,24]. This program applies a genetic algorithm to explore conformational spaces and ligand binding modes. To evaluate the proposed ligand positions, the ChemPLP fitness function was used. The binding site in the AChE model was defined as a 7 Å sphere from the co-crystallized donepezil ligand, and a water molecule interacting with a protonated piperidine ring of donepezil was conserved during the docking (residue number 931). The Ser203 side chain was kept flexible and a hydrogen bond constraint between Ser203 side chain and carbonyl oxygen was applied during the docking.

In Vitro Tests of AChE Inhibitory Activity
The inhibitory capacity of compounds on AChE biological activity was evaluated through the use of the spectrometric method of Ellman [14]. Acetylthiocholine iodide and 5,5-dithiobis-(2-nitrobenzoic) acid (DTNB) were purchased from Sigma Aldrich. AChE from human erythrocytes (buffered aqueous solution, ≥500 units/mg protein (BCA), Sigma Aldrich, St. Louis, MO, USA) was diluted in 20 mM HEPES buffer pH 8, 0.1% Triton X-100 such as to have enzyme solution with 0.25 unit/mL enzyme activity. In the procedure, 100 µL of 0.3 mM DTNB dissolved in a phosphate buffer with pH 7.4 were added into the 96-well plates followed by 50 µL of the test compound solution and 50 µL of enzyme (0.05 U final). After 5 min of preincubation at 25 • C, the reaction was then initiated by the injection of 50 µL of 10 mM acetylthiocholine iodide solution. The hydrolysis of acetylthiocholine was monitored by the formation of yellow 5-thio-2-nitrobenzoate anion as the result of the reaction of DTNB with thiocholine, released by the enzymatic hydrolysis of acetylthiocholine, at a wavelength of 412 nm using a 96-well microplate plate reader (BioTek, Synergy 2, Winooski, VT, USA). Test compounds were dissolved in analytical grade DMSO. Donepezil (DPZ) and Rivastigmine were used as a reference standard. The rate of absorbance increase at 412 nm was followed every minute for 10 min. Assays were performed with a blank containing all components except acetylthiocholine, in order to account for non-enzymatic reactions. The reaction slopes were compared and the percent inhibition due to the presence of test compounds was calculated by the following expression: 100 − (v i /v 0 × 100), where v i is the rate calculated in the presence of inhibitor and v 0 is the enzyme activity. The first screening of AChE activity was carried out at a 10 −6 concentration of compounds under study. For the compounds with significant inhibition (≥50%), IC 50 values were determined graphically by plotting the % inhibition versus the logarithm of six inhibitor concentrations in the assay solution using the GraphPad Prism 6 software.

Kinetic Study for AChE Inhibition
To try to clarify the mechanism of action of 19 and 20, reciprocal plots of 1/velocity vs. 1/[substrate] were constructed at different concentrations of the substrate acetylthiocholine iodide (0.01-1 mM) by using the spectrometric method by Ellman et al. [14]. Four concentrations of 19 were selected for the studies: 10 −6 , 10 −7 , 10 −8 and 10 −9 M for the kinetic analysis of AChE inhibition. Four concentrations of 20 were selected for the studies: 10 −5 , 10 −6 , 10 −7 and 10 −8 M for the kinetic analysis of AChE inhibition. The plots were assessed by a weighted least squares analysis that assumed the variance of velocity (ν) to be a constant percentage of ν for the entire dataset. Slopes of these reciprocal plots were then plotted against the concentration of 19 and 20 in a weighted analysis.

Pharmacological Characterization of Drugs on Human 5-HT 4 R
For competition studies, 2.5 µg of proteins (5-HT 4(b) membrane preparations, HTS110M, Eurofins. Eurofins' 5-HT 4(b) membrane preparations are crude membrane preparations made from their proprietary stable recombinant cell lines to ensure high-level of GPCR surface expression.) were incubated in duplicate at 25 • C for 60 min in the absence or the presence of 10 −6 or 10 −8 M of each drug (9 was used as a reference standard) and 0.2 nM [ 3 H]-GR113808 (NET 1152, Perkin Elmer) in 25 mM Tris buffer (pH 7.4, 25 • C). At the end of the incubation, homogenates were filtered through Whatman Spatial working memory. The antiamnesic activity of the tested compounds was evaluated by a reversal of deficit on spontaneous alternation behavior in the Y maze test [27]. Deficits were pharmacologically induced either by scopolamine (SCOP, 0.5 mg/kg) or dizocilpine (MK801, 0.1 mg/kg). The Y maze made of grey plastic consisted of three equally-spaced arms (21-cm long, 7-cm wide with walls 15-cm high). The mouse was placed at the end of one of the arms and allowed to move freely through the maze during a 5 min session, while the sequence of arm entries was recorded by an observer. An arm entry was scored when all four feet crossed into the arm. An alternation was defined as entries into all three arms on a consecutive occasion. The number of possible alternations is thus the total number of arm entries minus two; the percentage of alternation was calculated as (actual alternation/possible alternation) 100. Compound 20 was tested at 1, 3 and 10 mg/kg. Pharmacological treatments. Amphetamine (+)-α-Methylphenethylamine hemisulfate, chlorpromazine hydrochloride, MK801 hydrogen maleate and scopolamine hydrobromide were purchased from Sigma (France). All pharmacological compounds were dissolved in NaCl 0.9%, used as vehicle. Besides, all were administered IP 30 min before tests, except for scopolamine and MK801, which were subcutaneously administered 20 min before spontaneous alternation test.
Statistical analysis. Results were expressed as mean ± SD and were analyzed by one-way analysis of variance (ANOVA), with Statview ® software (Abacus Concepts, Berkeley, CA, USA). In case of significance, a SNK (Student-Newman-Keuls) post hoc test was realized. Additionally, for the spontaneous alternation test, the percentage of alternation was compared to a theoretical 50% value (random alternation) by a univariate t-test. Differences were considered as statistically significant if the p value was strictly under 0.05.

Conclusions
We succeeded in designing and synthesizing a pleiotropic prodrug which benefits from a carbamate structure 20 to easily cross the BBB and covalently bind to central AChE. Upon hydrolysis by the enzyme, 20 releases a phenol compound 4 which, contrarily to 20, acts both as a DBS AChEI and as a 5-HT 4 R partial agonist. With this unique in vitro profile, 20 displayed, in vivo at a low dose of 1 mg/kg, antiamnesic effects towards cognitive deficits, resulting from either cholinergic or glutamatergic neurotransmission impairments in mice. These results seem to support conjecture that this pleiotropic prodrug is both peripherally safe and shows great efficiency in AD pathogenesis and symptoms.