Design, synthesis and evaluation of benzimidazole-NsCOXis hybrids for the treatment of Alzheimer’s disease

Usage of an acetylcholinesterase (AChE) inhibitor and a non-selective COX inhibitor (NsCOXi) have been documented to exhibit signicantly protective and recuperative effects in AD patients. Therefore, it is hypothesised that a compound capable of exhibiting both AChE inhibition and anti-inammatory activities can be potential pluripotent drug candidate for de-accelerating the progression of AD, as well as for providing relief from the associated inammatory pathological indications. The present study involves the coupling of ibuprofen (IB) or naproxen (NP) to varied disubstituted amines (AChE inhibitor pharmacophore) through benzimidazole to develop two series of compounds i.e. IB01-IB05 and NP01-NP05 as pluripotent anti-AD compounds. All target compounds are evaluated for in vitro AChE inhibitory and COX inhibitory activities. Compounds IB01-IB05 are found more potent as compared to NP01-NP05. Compound IB04 being the most active in in vitro evaluation is selected for in vivo evaluation of memory restoration using scopolamine-induced amnesia model in mice. It signicantly reverses the scopolamine-induced changes (i.e., escape latency time, mean time spent in target quadrant, brain AChE activity and oxidative stress) in a dose-dependent manner. IB04 at higher dose i.e. 8 mg/kg is signicantly effective in lowering AD manifestation in comparison to donepezil. The ndings indicate that Benzimidazole-NsCOXi derivatives having pyrrolidine moiety may prove a useful template for the development of new chemical moieties against AD with multiple potencies. the reaction mixture was poured, while it was still hot into excess of water. The solution was then neutralized by sodium carbonate. The precipitated ester was ltered and dried under vacuum. The crude product was recr + ystallised from boiling water. Yield and melting point were noted.


Introduction
Alzheimer's disease (AD) is a progressive neurodegenerative disorder characterized by progressive cognitive decline and memory loss. US-FDA has approved three cholinesterase inhibitors (donepezil, rivastigmine and galantamine) for the treatment of AD. Memantine is another drug approved by FDA to treat the cognitive symptoms (memory loss, confusion, and problems with thinking and reasoning) of AD, which acts through regulation of glutamate activity [1,2]. These drugs, though modestly improve memory and cognitive function in AD patients, are unable to prevent the progression of the disease. This may be attributed to the pathological complexity of AD. Over the time, various strategies such as complex drug therapy or drug cocktails (independent dosage of the drugs) and combined drugs ( xed combination of two or more drugs in one dosage form) have evolved for the treatment of AD [3]. Compelling evidences from multiple epidemiology studies have revealed that long-term dosing with non-steroidal antiin ammatory drugs (NSAIDs) dramatically reduces AD risk in the elderly, including delayed disease onset, reduced symptomatic severity and slowed cognitive decline [4][5][6]. NSAIDs are predicted to dampen the neuro-in ammatory response and impact AD progression via several mechanisms, when administered along with drugs that inhibit beta-amyloid oligomerization [7]. The simultaneous administration of an acetylcholinesterase (AChE) inhibitor and a NSAID has been documented to exhibit signi cantly protective and recuperative effects in AD patients [8]. Clinical studies have proven that NSAIDs belonging to the category of non-selective COX inhibitors such as ibuprofen and naproxen protect against cognitive decline and reduce the levels of β-amyloid better than other NSAIDs ( urbiprofen, fenoprofen, indomethacin, meclofenamic acid, sulindac and fenbufen). A recent molecular docking study has suggested that (i) carboxylic group in a NSAID is not an essential requirement for inhibition of Aβ brils, and (ii) -O-; -NH-; -CO-and -CONH-linker groups in NSAIDs afford maximum binding a nity [9]. Based on these reports, it is hypothesized that combining the AChE inhibitory and anti-in ammatory potentials of a non-selective COX inhibitors (NsCOXi) in a single molecule may be an attractive strategy for both deaccelerating the progression of AD and providing relief from the associated in ammatory pathological indications. Therefore, the present study is designed to hybridise NsCOXi with appropriate pharmacophore for AChE inhibition to develop pluripotent molecules for the treatment of AD. A disubstituted amine in the form of an aliphatic chain or piperidine ring is common pharmacophore identi ed from most of the studies reported for AChE inhibition. Naproxen and ibuprofen are the most commonly used NsCOXi. Benzimidazole is a versatile heteronucleus, which is exploited as an anchor to develop numerous biologically active molecules [10]. Therefore, the compounds have been designed by coupling naproxen or ibuprofen with some di-substituted amines through benzimidazole nucleus (Fig. 1), and are evaluated for in vitro AChE inhibition, COX inhibition and in vivo behavioural study.

Chemistry
The target compounds have been synthesized by following a synthetic scheme comprising 4 steps as depicted in Scheme 1. In the rst step, carboxylic acid group of 3,4-diaminobenzoic acid (1) is esteri ed for its protection by re uxing 1 with ethanol in the presence of dry HCl gas to produce ethyl 3,4-diamino benzoate (2). Esteri cation of 1 to 2 is con rmed through 1 H-NMR spectral analysis which reveals that the broad singlet due to -COOH in 1 disappears, and a 2-proton quintet at d 4.27 and a 3-proton triplet at d 1.32 appear due to ethyl group. In the second step, the esteric intermediate (2) is re uxed with ibuprofen (3a) or naproxen (3b) in the presence of orthophosphoric acid (OPA) to yield benzimidazole-NsCOXi conjugate intermediates (4a/4b). Formation of these conjugates is con rmed by IR and 1 H-NMR spectral analyses. Mainly, a double band at 3442 and 3300 cm -1 due to -NH 2 in IR spectrum of 2 is replaced by a single broad band due to -NH-in IR spectrum of the conjugates; and a band appears at 1672.1 cm -1 due to C=N stretching, which indicate that conversion of -NH 2 group in 2 to -NH-group involves the formation of benzimidazole nucleus. 1 H-NMR spectra of 4a and 4b show the signals due to three protons of benzimidazole nucleus along with the proton signals due to respective NsCOXi i.e., ibuprofen and naproxen. For instance, a 2-proton quartet and a 3-proton doublet due to esteric ethyl group are noted at d 4.62-3.34 and d 1.98-1.26, respectively whereas three benzimidazole protons are noted in the range of d 7.78-6.98 in the spectra of both 4a and 4b. In spectrum of 4a, ibuprofen-speci c signals include two doublets at d 7.15 and d 7.00 due to four protons of the para-disubstituted phenyl ring, and a set of signals in the range d 3.04-1.06 due to isobutyl group, whereas the signals characteristic to naproxen include a 3 proton singlet at d 3.87 due to methoxyl group and aromatic protons at d 7.75-7.58. Deesteri cation of 4a/4b by 5 M NaOH in THF results in free carboxylic acid intermediates 5a/5b, which is con rmed by the disappearance of protons of ethyl group in their 1 H-NMR spectra.
Finally, the intermediate 5a/5b are coupled to varied disubsitituted amines by stirring in the presence of dicyclohexycarbodiimde (DCC) and dichloromethane (DCM) to obtain the target compounds IB 01-05/NP 01-05. IR spectrum of each target compound is found to lack the C=O stretching band corresponding to the free -COOH group, which shows that -COOH group in intermediates 5a/5b is couple to the disubstituted amine. Further, Amide I and Amide II bands at about 1680 and 1575 cm -1 , respectively, suggest an amide linkage in the compound due to coupling of -NH-with -COOH. The 1 H-NMR spectra of all target compounds have shown signals similarly as in the respective intermediates (5a/5b).
Additionally, protons of the substituted amine moieties are detected at the expected chemical shift positions in the region of d 0-5. The 13 C-NMR spectra of all target compound show a maximally down eld signal at d 169.8-166.9 due to carbonyl carbon, along with signals due to other carbons as expected. Finally, the high resolution mass spectral data (+ESI) of all target compounds show the parent ion peak as [M+H] + at m/z values corresponding to their theoretical masses con rming the formation of the compounds.

In vitro AChE inhibitory activity
AChE inhibitory activity of target compounds (IB01-IB05 and NP01-NP05) was determined by using Ellman method [11] with slight modi cations using donepezil as stranded drug. IC 50 values of all target compounds for AChE inhibition are summarized in Table 1. The data reveal that almost all compounds have signi cant AChE inhibitory activity. In the IB series, a heterocyclic moiety (morpholine, pyrrolidine and piperizine) appended to benzimidazole rings is found to incur signi cantly higher activity (IC 50 0.34 -0.55 µM) than a dialkylamino moiety (IC 50 0.84 and 1.01 µM). Contrastingly in NP series, the dimethylamino substituted analog (NP01) is found to be maximally potent (IC 50 0.65 µM).
Statistical analysis of the data by one way ANOVA followed by multiple comparison (Tukey's) test revealed that ibuprofen-derived compounds (IB01-IB05) are more potent than the naproxen-derived ones (NP01-NP05). Compound IB04 is the most potent among all the synthesized compounds from both the series, which is however less active than donepezil.

In vitro COX inhibitory activity
The target compounds were tested for their inhibitory activity against both COX-1 and COX-2 enzymes as well as their selectivity index (SI = IC 50 COX-1/IC 50 COX-2) using ibuprofen and naproxen as reference drugs. Amongst the ibuprofen-derived compounds, IB04 is found to be the most potent for both COX-1 and COX-2 with a COX-1/COX-2 selectivity index of 0.27. The compound IB02 is statistically equipotent to IB-04 towards COX-1 and has maximum selectivity for COX-1 (Selectivity index 0.17) among ibuprofen derived molecules. Amongst the naproxen-derived molecules, NP05 and NP04 are most potent for COX-1 and COX-2, respectively. The methyl and ethyl substituted compounds (NP01-NP02) are less active as compared to closed ring amine derivatives (NP03-NP05) for both COX-1 and COX-2. Interestingly, all target compounds are found to have selectivity towards COX-signi cantly increased as compared ibuprofen and naproxen. Further, the ibuprofen-derived compounds (IB01-IB05) are more selective towards COX-1 (SI 0.16 -0.37) as compared to the naproxen-derived compounds (NP01-NP05) (SI 0.36-0.49).
In vivo biological evaluation IB-04 is found to be the most potent inhibitor of AChE (IC 50 0.34 μM/ml) as well as COX-1 (IC 50 10.96 μM/ml) with good selectivity towards COX-1 (SI 0.27). Therefore, it was selected for evaluation of learning and memory restoration through scopolamine-induced amnesia model in mice at three different doses (i.e., 2, 4 and 8 mg/kg) using Morris Water Maze. The learning and memory were assessed in terms of Escape Latency Time (Fig. 2) and Mean Time Spent in Target Quadrant (Fig. 3).

Escape Latency Time (ELT)
All the animals were subjected to trials in morris water maze for 5 days, and assessed for ELT for rst for 4 days. Normal control group animals showed downward trend in ELT exposure during acquisition trial of day 4 that re ects normal learning abilities. However, scopolamine-treated mice showed a signi cant rise in ELT on day 4 in comparison to the control group mice indicating an impairment of acquisition (learning). Administration of IB04 (at 2, 4 and 8mg/kg; p.o) and donepezil (2 mg/kg, i.p) in scopolaminetreated mice resulted in fall in ELT on day indicating reversal in scopolamine-induced impaired learning and memory. Treatment with IB04 showed improvement in ELT in a dose-dependent manner with response at 8 mg/kg being comparable to donepezil.

Mean Time Spent in Target Quadrant (TSTQ)
Normal control mice when subjected to retrieval test on day 5 spent signi cantly (p<0.05) more time in the target quadrant (Q4) in search for the missing platform as compared to the time spent in other quadrants (Q1, Q2 and Q3), re ecting normal memory capacity. On the other hand, scopolamine-treated mice showed signi cant decrease in time spent in Q4 quadrant, which depicts memory impairment. However, administration of donepezil and IB04 in scopolamine-treated mice showed increase in day 5 TSTQ in a dose-dependent manner, which indicates normal cognitive function (Fig. 3). These ndings of in vivo evaluations disclosed that IB04 exhibits signi cant actions against AD manifestations in a dosedependent fashion. Effect of IB04 on scopolamine-induced changes in brain AChE activity Brain AChE activity was found to increase signi cantly in scopolamine-treated mice as compared to that in normal control mice. Administration of donepezil and IB04 prevented this scopolamine-induced rise in brain AChE activity in a dose-dependent manner (Fig. 4).
Effect of IB04 on scopolamine-induced changes in brain oxidative stress levels TBARS levels were enhanced and reduced glutathione (GSH) levels were decreased signi cantly (p<0.05) in scopolamine-treated mice in comparison to normal control mice, indicating increased oxidative stress in scopolamine-treated mice. Treatment of scopolamine-treated mice with donepezil and IB04 showed a signi cantly decrease in TBARS levels and increase in brain GSH level, which depicts a decrease in oxidative stress in a dose-dependent manner (Fig. 5). These observations suggested that IB04 is effective in reducing oxidative stress in dose-dependent fashion.

Conclusion
Though numerous multifunctional strategies for the treatment of AD are reported in literature, yet inhibition of both in ammatory pathways and AChE activity or level needs to be explored. Hence, the present study involves the coupling of pharmacophore from anti-in ammatory drugs (ibuprofen and naproxen) and from varied AChE inhibitors to design and synthesize benzimidazole-based two series of compounds i.e., IB01-05 and NP01-05. All compounds were primarily evaluated for in vitro AChE and COX inhibitory activities. Compounds IB01-IB05 are found to be more potent in vitro as compared as compounds NP01-NP05. Among all, compound IB04 is found to be the most potent, and hence, it is evaluated for its memory restoration activity using scopolamine-induced amnesia model in mice. It signi cantly reversed the scopolamine-induced changes (i.e., escape latency time, mean time spent in target quadrant, brain AChE activity and oxidative stress) in a dose-dependent manner. IB04 at 8 mg/kg is found to be signi cantly effective in lowering AD manifestation in comparison to donepezil. The ndings of this study indicate that the benzimidazole-NsCOXi derivatives having pyrrolidine moiety may provide a useful template for the development of new chemical moieties effective against AD.

Experimental
Chemistry All reactions were monitored by TLC using pre-coated aluminum plates. The purity of the compounds was ascertained con rmed by the developed TLC plate visualised in UV chamber at short as well as long Synthesis of BZ-Ibuprofen/Naproxen conjugate Ethyl-3,4-diaminobenzoate (2.5 mM, 0.45 g) was re uxed with 3.75 mM of ibuprofen or naproxen in the presence of OPA in a 50 ml round bottom ask. Completion of the reaction was ascertained through the disappearance of spot of reactant in TLC. After completion of the reaction, the reaction mixture was poured into excess of cold water and basi ed with aqueous ammonia. The resultant product was ltered, washed with cold water and recrystallized from hot ethanol-water mixture to obtain ne crystals, which were dried in vacuum desiccator.  General procedure Synthesis of target compound (IB01 -05 and NP01-NP05 series) Dicyclohexyl carbodiimide (DCC; 0.1 mM, 0.4 g) was added to a solution of the BZ-NsCOXi conjugates (0.1 mM) in dicholoromethane (10 mL). The content were stirred for 30 min followed by addition of disubsituted amine (0.3 mM) and a solution of DMAP in dichloromethane (10 mL). The reaction mixture was stirred at 0°C for rst 2 h, followed by stirring at room temperature overnight. The precipited dicyclohexyurea (DHU) was ltered and the solvent was removed in vacuum. Ethyl acetate (10 mL) was added to the residue in order to separate the product from residual DHU. The ethyl acetate solution was itered, washed with 10 % aqueous solution of sodium bicarbonate followed by distilled water and then dried with magnesium sulphate (anhydrous). The solvent was recovered in vacuum to obtain the crude product. ve consecutive days and the memory was assessed in terms of: (i) escape latency time (ELT), that is, the time taken by the animal to locate the hidden platform in the target quadrant (Q4) for the rst 4 days of training, and (ii) time spent in target quadrant (Q4) on fth day of trial, that is, the day of retrieval.

Experimental Protocol
Seven groups, each group comprising ve mice, were employed in this study.
Group I (normal control) Normal mice, without any treatment, were subjected to trials on the water maze for 5 days to note escape latency time (ELT) for rst 4 days (an index of learning) and time spent in target quadrant (TSTQ) on 5th day of trial (an index of retrieval).
Group II (scopolamine control) Scopolamine (0.4 mg/kg i.p.) was administered to each mouse, 30 min prior to each trial, for rst 4 days of trials. In scopolamine-treated mice, the ELT and TSTQ were noted as described in group I.
Group III (vehicle control) 10 % DMSO (1 ml/kg) was administered in scopolamine-treated mice (30 min before scopolamine administration) to each mouse before subjecting to rst four trials and rest of the procedure was same as described in group I Group IV (Donepezil in scopolamine control) Donepezil (5 mg/kg) was administered in scopolamine-treated mice (30 min before scopolamine administration) to each mouse before subjecting to rst four trials and rest of the procedure was same as described in group I.
Group V-VII (Compound IB-04 (2, 4 and 8 mg/kg) in scopolamine control) The test compound with the most potent AChE, i.e. compound IB04 was administered by oral route through canula in different doses (2, 4 and 8 mg/kg) in scopolamine-treated mice (30 min before scopolamine administration) to each mouse before subjecting to rst four trials and rest of the procedure was same as described in group I.

Biochemical estimations
At the end of the protocol, animals were euthanized by cervical dislocation and brains were removed carefully. Different parts of brain i.e. cortex and hippocampus were separated. Isolated parts were homogenized in ice cold phosphate buffer of pH 7.4. The homogenate was centrifuged at 14,500 rpm for 15 min at 4 C. The clear supernatant was used for estimation of thiobarbituricacid reactive substance (TBARS), reduced glutathione (GSH), and brain AChE activity by the methods reported by Ohkawa et al [15] Beutler, et al [16], and Ellman method [11], respectively.

Statistical analysis
Data was expressed as mean ± SD of the obtained data. The statistical analysis of the data was performed using analysis of variance (ANOVA) followed by multiple comparison test. The Morris water maze data (Day 1 ELT, Day 4 ELT, and Day 5 TSTQ data) was analysed using two-way ANOVA followed by Bonferroni post hoc testwhereas biochemical estimations were statistically analyzed by one-way ANOVA followed by Tukey's test. A value of P < 0.05 was considered statistically signi cant.

CONFLICT OF INTEREST
There is no con ict of interest among authors.  Effect of IB04 on escape latency time at day 1 and day 4 using water maze test for memory evaluation. Data is presented as Mean ± S.D. and analyzed by two way ANOVA followed by Tukey's multiple range test; a p< 0.05 vs day 1 ELT in control; bp< 0.05 vs day 4 ELT in normal; cp< 0.05 vs day 4 ELT in scopolamine, dp< 0.05 vs day 4 ELT in donepezil + scopolamine.  Effect of IB04 on brain AChE activity. Values are expressed as Mean±S.D. and analyzed by one way ANOVA followed by Tukey's multiple range test; a(p < 0.05) in comparison to normal; b(p < 0.05) in comparison to scopolamine; c(p < 0.05) in comparison to donepezil + scopolamine. Effect of IB04 on TBARs level. Data are presented as mean ± standard deviation; analyzed by one way ANOVA. a(p < 0.05) in comparison to Normal; b(p < 0.05) in comparison to scopolamine; c(p < 0.05) in comparison to donepezil + scopolamine.