Design, Synthesis and Biological Evaluation of New Antioxidant and Neuroprotective Multitarget Directed Ligands Able to Block Calcium Channels

We report herein the design, synthesis and biological evaluation of new antioxidant and neuroprotective multitarget directed ligands (MTDLs) able to block Ca2+ channels. New dialkyl 2,6-dimethyl-4-(4-(prop-2-yn-1-yloxy)phenyl)-1,4-dihydropyridine-3,5-dicarboxylate MTDLs 3a–t, resulting from the juxtaposition of nimodipine, a Ca2+ channel antagonist, and rasagiline, a known MAO inhibitor, have been obtained from appropriate and commercially available precursors using a Hantzsch reaction. Pertinent biological analysis has prompted us to identify the MTDL 3,5-dimethyl-2,6–dimethyl–4-[4-(prop–2–yn–1-yloxy)phenyl]-1,4-dihydro- pyridine- 3,5-dicarboxylate (3a), as an attractive antioxidant (1.75 TE), Ca2+ channel antagonist (46.95% at 10 μM), showing significant neuroprotection (38%) against H2O2 at 10 μM, being considered thus a hit-compound for further investigation in our search for anti-Alzheimer’s disease agents.


Introduction
Alzheimer's disease (AD) is a neurodegenerative pathology characterized by a highly interconnected biological processes leading to neuronal death, accumulation and aggregation of abnormal extracellular deposits of beta-amyloid peptide (Aβ) and neurofibrillary tangles, composed of hyperphosphorylated tau protein [1], and low level of neurotransmitter acetylcholine. Therefore, new strategies, based on the multitarget directed ligand (MTDL) approach [2,3], have been developed for the design of new drugs able to bind simultaneously at diverse enzymatic systems or receptors involved in the progress of AD [4][5][6][7][8]. Accordingly, and following this paradigm a number of MTDLs has been described by many research groups [9][10][11]. Our contributions in this area have used multicomponent

Synthesis
The synthesis of the new MTDL 3a-t has been carried via a one-pot Hantzsch reaction of aldehydes 1a-e/2a-e with ethyl or methyl acetoacetate and ammonium carbonate in EtOH/water (Scheme 1). Aldehydes 1a-e were prepared from the appropriate substituted 4-hydroxybenzaldehydes and propargyl bromide, under typical Williamson reaction conditions (Scheme 1). Aldehydes 2a-e were synthesized by a Mitsunobu reaction under the conditions described by Mertens [38], from but-3-yn-1-ol and 3-substituted 4-hydroxybenzaldehydes, in the presence of Ph 3 P and diisopropyl azodicarboxylate (DIAD), in THF, at room temperature (rt) (Scheme 1). All new compounds showed excellent analytical and spectroscopic data, in good agreement with the expected values (see Material and Methods, and Supplementary Material).
To verify, the effectiveness of our design, compounds 3a-t were submitted to Ca +2 channel blockade, antioxidant and MAO inhibition evaluation assays, followed by the neuroprotection analysis of selected compounds.

Ca +2 Channel Blockade
The Ca +2 channel blockade capacity of compounds 3a-t, and nimodipine as standard, at 10 μM concentration, has been carried out following the usual methodology [39]. As shown in Table 1, the observed % values ranged from 20.2 (3k) to 47.0 (3a). The most potent DHP corresponded, in decreasing order, to 3a (47.0 ± 6.6%), 3h (42.8 ± 14.0%), and 3j (39.0 ± 4.8%), comparing thus very favorably witH-Nimodipine (52.8 ± 5.5%). From the point of view of the structure activity relationship (SAR), compounds bearing n = 1 length as linker showed better results than those bearing n = 2 linkers. In fact, only one adduct with n = 1 presented an inhibition value under 25% (3f, 22.9 ± 6.7%), the rest surpassed 30% Ca 2+ influx blockage, which is 2/3-fold the value of nimodipine (52.8%). Concerning the influence of R 1 over the blockade activity, no conclusions can be drawn. However, it is worth mentioning that compounds 3a and 3h with R 1 = H and Cl, respectively, are the most active compounds, as they presented values of inhibition that almost doubled the rest (47% and 43%). To verify, the effectiveness of our design, compounds 3a-t were submitted to Ca 2+ channel blockade, antioxidant and MAO inhibition evaluation assays, followed by the neuroprotection analysis of selected compounds.

Ca 2+ Channel Blockade
The Ca 2+ channel blockade capacity of compounds 3a-t, and nimodipine as standard, at 10 µM concentration, has been carried out following the usual methodology [39]. As shown in Table 1, the observed % values ranged from 20.2 (3k) to 47.0 (3a). The most potent DHP corresponded, in decreasing order, to 3a (47.0 ± 6.6%), 3h (42.8 ± 14.0%), and 3j (39.0 ± 4.8%), comparing thus very favorably witH-Nimodipine (52.8 ± 5.5%). From the point of view of the structure activity relationship (SAR), compounds bearing n = 1 length as linker showed better results than those bearing n = 2 linkers. In fact, only one adduct with n = 1 presented an inhibition value under 25% (3f, 22.9 ± 6.7%), the rest surpassed 30% Ca 2+ influx blockage, which is 2/3-fold the value of nimodipine (52.8%). Concerning the influence of R 1 over the blockade activity, no conclusions can be drawn. However, it is worth mentioning that compounds 3a and 3h with R 1 = H and Cl, respectively, are the most active compounds, as they presented values of inhibition that almost doubled the rest (47% and 43%).

Antioxidant Assay
The antioxidant activity of compounds 3a-t, compared to melatonin, used as positive control, showing an ORAC value of 2.45 [12], was determined by the ORAC-FL method [40]. The antioxidant activities are expressed as Trolox equivalents (TE) units. As shown in Table 1, the values for the antioxidant capacity range from 0.52 (3e) to 2.78 (3n). Three compounds, 3l (2.45 TE), 3n (2.78 TE) and 3p (2.78 TE), showed antioxidant activities equal or higher than melatonin (2.45 TE). Concerning the SAR, for the same R 1 substituent, compounds with a linker length of n = 2 showed better ORAC values than those with n = 1, except for the pairs 3g, 3q and 3h, 3r. For the same linker length the best results for n = 1 were obtained for compounds bearing R 1 = H, whereas for compounds with n = 2, the best results corresponded to molecules bearing R 1 = OMe or R 1 = OEt.

hMAOs Inhibition
The effect of the compounds 3a-t on the activity of both human MAO (hMAO) isoforms was evaluated by measuring the production of 4-hydroxyquinoline (4-HQ, λmax = 316 nm) from kynuramine, using microsomal recombinant hMAO isoforms. Unexpectedly and unfortunately these compounds showed a very low inhibition.
Based on the previously described biological results, the three most balanced compounds (3a, 3h and 3j) against calcium channel blockade and antioxidant activity were evaluated for their capacity to protect human neuronal cells (SH-SY5Y cell line) from cell death.

Neuroprotective Activity
Several in vitro approaches have been performed to mimic human neuronal features, based on neuronal-like cells such as the neuroblastoma line SH-SY5Y, a human cell line that divides quickly and has the ability to differentiate in post-mitotiC-Neurons, thus it is considered a convenient and popular model to study neuroprotective activity for PD and AD [41]. For this purpose, cytotoxicity was induced by mitochondrial respiratory chain blockers oligomycin rotenone (O/R) and by H 2 O 2 , a well-known toxic responsible for the generation of ROS. Prior to the neuroprotective assay, the effect of the compounds on the cell viability was evaluated at 1 and 10 µM, showing no cytotoxicity against SH-SY5Y cells. As shown in Table 2, compounds 3a and 3j showed a modest neuroprotective effect against O/R. However, and very interestingly, the two compounds showed an interesting effect against H 2 O 2 , particularly at 10 µM where they showed a percentage of neuroprotection equal to 38 and 39 for 3a and 3j, respectively.

General Information
All reagents were purchased from Sigma Aldrich (Saint-Quentin Fallavier France) or TCI (Zwijndrecht, Belgium). 1 H-and 13 C-NMR spectra were recorded on a Bruker (Wissembourg France) spectrometer, operating at 400 and 100 MHz, respectively, in solution in dimethylsulfoxide (DMSO-d 6 ) at rt. Chemical shift values are given in δ (ppm) relatively to TMS as internal reference. Coupling constants are given in Hz. The following abbreviations were used: s= singlet, d= doublet, t= triplet, q= quartet, m= multiplet. Elemental analyses were obtained by a Carlo Erba EA 1108 analyzer and the analytical results were within ± 0.2% of the theoretical values for all compounds. High resolution mass spectra were obtained at Centre Commun de Spectrométrie de Masse, Lyon, France on a Bruker micrOTOF-Q II spectrometer (Bruker Daltonics, Champs sur Marne France) in positive ESI-TOF (electrospray ionization-time of flight).

Synthesis of Propargylic Aldehydes 1a-e
A suspension of the corresponding 4-hydroxybenzaldehyde (1 equiv) and K 2 CO 3 (1.3 equiv) in acetone (1.6 mmol/mL) was stirred at reflux for 30 min. The mixture was cooled to rt and propargyl bromide (1.6 equiv) was added dropwise. The resulting suspension was stirred at reflux for 2 h 30 min. After that time, the solvent was removed under pressure conditions. The residue was dissolved in water and extracted with ethyl acetate three times. Organic layers were joined and dried over Na 2 SO 4 . Activated carbon was added to the solution and the mixture is stirred over 15 min at 40 • C. The crude was finally filtered over Celite ® , the filtrate was evaporated and the obtained residue was recrystallized from EtOAct/hexane (1:2 v/v) to afford the desired products in yields ranging from 34% to 98%.

Synthesis of Propargylic Aldehydes 2a-e
A solution of 4-hydroxybenzaldehyde (1 equiv), triphenylphosphine (2 equiv) and 3-butyn-1-ol (1.5 equiv) in THF (0.8 mmol/mL) was cooled to 0 • C. DIAD (1.5 equiv) is then added dropwise and the resulting mixture is stirred overnight at rt. The solvent was evaporated, the residue solubilized in ethyl acetate and washed 3 times with 1M NaOH solution and brine. The organic layers were dried over Na 2 SO 4 , filtered and evaporated under pressure conditions. The residue was triturated with ethyl ether and filtered. The filtrate was purified by flash chromatography with hexane/EtOAc (9:1 v/v) to afford the desired products in yields ranging from 45% to 97%.

General procedure of compounds 3a-t
3-Substitued-4-alkynyloxy-benzaldehydes 1a-e and 2a-e (1 equiv) and the corresponding acetoacetate (3.5 equiv) were dissolved in a mixture of EtOH (4 mmol/mL) and the same volume of H 2 O. The resulting mixture is stirred and heated at 75 • C for 1 h. Next, ammonium carbonate (2.5 equiv) was added to the mixture, and the reaction stirred and heated at 75 • C overnight. The desired product precipitated once the crude reaction reached rt, or was triturated with diethyl ether. The solid was then filtered and washed with diethyl ether again to finally afford compounds 3a-t in yields ranging from 12% to 76%.

Oxygen Radical Absorbance Capacity Assay
The antioxidant activity of hybrids 3a-tI was carried out by the ORAC-FL using fluorescein as a luorescent probe. Briefly, fluorescein and antioxidant were incubated in a black 96-well microplate Nunc, Thermo Scientific, 67403 Illkirch France) for 15 min at 37 °C. 2,2′-Azobis(amidinopropane) ihydrochloride was then added quickly using the built-in injector of a Varioskan Flash plate reader

Calcium Channel Blockade
Human neuroblastoma cell line SH-SY5Y (CRL-2266) obtained from the American Type Culture Collection (Manassas, VA, USA) was maintained at 37 • C in humidified atmosphere of 5% CO2/95% air. Mixture of Dulbecco's Modified Eagle Medium and Nutrient Mixture F-12 (1:1) containing 10% of fetal bovine serum, was used as a culture medium (Thermo Fisher Scientific). The cells were seeded out in 96-well dark-walled plates at a density of 1 × 10 5 cells per well. The cells were between second and eighth passage number at the time of the experiment. After 24 h, the cells were loaded with FLIPR Calcium 6 indicator (Molecular Devices, San Jose, CA, USA) for 2 h at 37 • C, according to the manufacturers protocol. Compounds of interest were dissolved in appropriate amount of DMSO, in order to prepare stock solutions at 10 mM concentration. They were then subsequently diluted to a final concentration of 10 µM in Hanks Balanced Salt Solution (HBSS, Thermo Fisher Scientific, Waltham, MA, USA) buffered with HEPES (Sigma-Aldrich) at pH = 7.4 and as such used for treatment of indicator loaded cells (10 min at 37 • C). Fluorescence of indicator loaded cells was measured with Synergy H1 (Biotek Instruments, Winooski, VT, USA) multilabel plate reader at excitation and emission wavelengths of 485 and 525 nm, respectively. The baseline fluorescence was recorded for 5 sec. Then, the cells were stimulated with KCl/CaCl 2 solution (in HBSS, final concentration of KCl and CaCl 2 was 90 mM and 5 mM, respectively) and the fluorescence was recorded for further 30 s. Dimethyl sulfoxide 1% solution in HBSS was used as a vehicle control. Nimodipine (10 µM) was used as a reference inhibitor. Compounds were assessed at the same final concentration as nimodipine in triplicates in three independent experiments. Fluorescent intensity values were normalized to the baseline. Outlier detection by Grubbs' test was performed and outlying values were excluded from the further analysis.

Oxygen Radical Absorbance Capacity Assay
The antioxidant activity of hybrids 3a-tI was carried out by the ORAC-FL using fluorescein as a fluorescent probe. Briefly, fluorescein and antioxidant were incubated in a black 96-well microplate (Nunc, Thermo Scientific, 67403 Illkirch France) for 15 min at 37 • C. 2,2 -Azobis(amidinopropane) dihydrochloride was then added quickly using the built-in injector of a Varioskan Flash plate reader (Thermo Scientific). The fluorescence was measured at 485 nm (excitation wavelength) and 535 nm (emission wavelength) each min for 1h. All the reactions were made in triplicate and at least three different assays were performed for each sample.

hMAOs Inhibition Screening
The effect of the test compounds on the activity of both hMAO isoforms was evaluated by measuring the production of 4-hydroxyquinoline (4-HQ, λ max = 316 nm) from kynuramine, using microsomal

Conclusions
Compounds 3a-t have been successfully synthesized in modest to high yields by Hantzsch multicomponent reactions, and their biological evaluation, as potential MAO inhibitors, Ca 2+ channel blockers, antioxidant and neuroprotective agents has been assessed. Unfortunately, our molecules displayed very low MAO inhibition power. However, concerning Ca 2+ channel blockade results, it is worth mentioning that compounds 3a and 3h showed Ca 2+ influx blockage values of 47%, and 42.8%, respectively, very close to the standard reference nimodipine (52.8%) at 10 µM. The most active compounds were those with n = 1 linker length, with R 1 = H (3a) and R 1 = Cl (3h). This suggests that the C3 position at the aryl core is involved in the Ca 2+ channel blockade, and that this finding is a good strategy for further pharmacomodulation studies. The antioxidant activity allowed us to identify molecules 3l (2.45 TE), 3n (2.78 TE) and 3p (2.78 TE) as interesting antioxidants with ORAC values equal or higher than melatonin (2.45TE). Moreover, and very interestingly, the most potent Ca 2+ channel agonists, 3a and 3h, showed interesting ORAC values equal to 1.75TE and 1.35TE. The neuroprotective activity results support the interest of compounds 3a and 3h, particularly for their neuroprotective activity against H 2 O 2 in SHSY5Y cells. To sum up, the biological analyses have prompted us to identify the multifunctional compound 3,5-dimethyl-2,6-dimethyl-4-[4-(prop-2-yn-1-yloxy)phenyl]-1,4-dihydropyridine-3,5-dicarboxylate