hERG stereoselective modulation by mexiletine‐derived ureas: Molecular docking study, synthesis, and biological evaluation

Long QT syndrome (LQTS) is a disorder of cardiac electrophysiology resulting in life‐threatening arrhythmias; nowadays, only a few drugs are available for the management of LQTS. Focusing our attention on LQT2, one of the most common subtypes of LQTS caused by mutations in the human ether‐à‐go‐go‐related gene (hERG), in the present work, the stereoselectivity of the recently discovered mexiletine‐derived urea 8 was investigated on the hERG potassium channel. According to preliminary in silico predictions, in vitro studies revealed a stereoselective behavior, with the meso form showing the greatest hERG opening activity. In addition, functional studies on guinea pig isolated left atria, aorta, and ileum demonstrated that 8 does not present any cardiac or intestinal liability in our ex vivo studies. Due to its overall profile, (R,S)‐8 paves the way for the design and development of a new series of compounds potentially useful in the treatment of both congenital and drug‐induced forms of LQTS.

ileum demonstrated that 8 does not present any cardiac or intestinal liability in our ex vivo studies.Due to its overall profile, (R,S)-8 paves the way for the design and development of a new series of compounds potentially useful in the treatment of both congenital and drug-induced forms of LQTS.

| INTRODUCTION
Urea represents a privileged structure in drug design and development, providing potent drug-target interactions by forming multiple stable hydrogen bonds with proteins.][11] On the other hand, some urea-based small molecules have also been proposed as human ether-à-go-go-related gene (hERG) potassium channel activators (NS1643 and NS3623) which accelerate myocardial repolarization by increasing the outward potassium current during the ventricular action potential. [12,13]Therefore, they are potentially useful for the treatment of both congenital and druginduced forms of long QT syndrome (LQTS), [12] a cardiovascular disorder characterized by abnormal cardiac repolarization, leading to a prolonged QT interval and T-wave irregularities on the surface electrocardiogram (ECG), and responsible for a concomitant risk of Torsade de Pointes, a well-known potentially life-threatening arrhythmia. [14,15]Starting from the consideration that mexiletine-a wellknown voltage-gated sodium channel blocking agent clinically useful as class Ib antiarrhythmic and antimyotonic agent [16][17][18][19][20] -can reverse the action potential prolongation in patients with LQTS, [21][22][23][24][25][26] and in analogy with the urea-based hERG potassium channel activators, we recently described an asymmetric mexiletine urea [(R,R)-1,3-bis [1-(2,6-dimethylphenoxy)propan-2-yl]urea, MC450, Figure 1], which shared some structure and electrophysiological features with the "Type 2" activators of the hERG potassium channel. [27] mainly acts on the inactivation mechanism of the channel unlike "Type 1" activators that slow the rate of channel deactivation.
The mechanism of action of MC450 is similar to that described for the two previously reported ureas NS-1643 and NS-3623, thus corroborating the role of the urea scaffold as a useful chemotype in the quest for hERG openers. [28]arting from the observation that the hERG channel is involved in cardiac disorders and very few and even old studies have been devoted to the evaluation of the cardiac activity of ureabased compounds so far, [29][30][31][32][33][34][35] we aimed to explore the cardiovascular effect of MC450, together with its SS enantiomer and the RS (meso) form to gain insight on a possible stereoselective behavior.These compounds were prepared in high optical purity by an alternative and more efficient stereospecific route than what was previously reported for the RR enantiomer. [27]Their respective chemical and optical purity were checked through a chiral high performance liquid chromatography (HPLC) method and crystal structures were determined by the X-ray crystallography.

| Chemistry
Despite our previously reported accidental synthesis of (R,R)-8, [27] the desired optically active urea has been firstly synthesized by a convergent synthesis reported in Scheme 1.

| Crystal structure determination
Being known the dependence of the E/Z conformation equilibrium on the pattern of N-substitution in ureas, [38,39] the crystal structures of (S,S)-8 and (R,S)-8 were determined by X-ray diffraction powder data to confirm the Z,Z conformation to support molecular docking simulation study.A real-space method based on the simulated annealing algorithm (SA) was used and the crystal cell and symmetry were determined by an automatic analysis of the diffraction profile (Supporting Information: Table S1).
The crystal structures of the individual molecules and their crystal packing are shown in Figure 2, while crystal data and refinement parameters are reported in Supporting Information: Table S1.
The analysis of crystal packing (Figure 2c,d For such interactions, the H-O distances range from 1.9 Å to 2.1 Å and the NH-O angles range from 148°to 158°, as expected for moderate hydrogen bonds. [40](ii) Molecules in the crystal are stacked along the shorter unit cell axis (c).Such stacking could be ascribed to π-π interactions among the aromatic moieties of the molecules in the crystal, although the observed ring-to-ring distances (4.5-6.5 Å) are larger than those expected for such interactions. [41]The aromatic moieties involved in π-π interactions show a parallel-displaced geometry, as a result of the attractive interaction between one of the methyl groups bound to phenyl and the phenyl group of a symmetry-related molecule.
Despite the relative position of the methyl and phenyl group being similar for both the molecules under investigation, (S,S)-8 shows shorter ring-to-ring distances than (R,S)-8, suggesting a stronger stacking interaction in this case.

| Molecular docking
LQT2, one of the most frequent subtypes of congenital LQTS, is associated with mutations in genes encoding hERG and is related to the loss of function of the channel. [42]In 2017, we reported on the discovery of MC450 (Figure 1), a new mexiletine-derived opener of the hERG K + channel able to shorten action potential duration, thus having therapeutic potential for the treatment of LQT2.Although this compound could exist in three different stereoisomers, only the isolated RR enantiomer [(R,R)-8] was at first evaluated as a hERG opener.Based on the interesting obtained results, in-depth studies on the SS enantiomer [(S,S)-8] and the meso form [(R,S)-8] would be advisable as well.Therefore, a preliminary in silico investigation on a possible stereoselective behavior has been performed on the recently published cryo-EM structure of the human ERG (hERG) channel (pdb code: 5VA1).Notably, ligand-induced fit effects were included during the simulations, as proved to be of utmost importance for obtaining reliable docking data on the available hERG protein structures. [43]gure 3 shows the obtained top-scored docking poses.Remarkably, (R,R)-8 and (R,S)-8 share a very similar binding mode where the urea moiety is crucial to engage a dual well-oriented H-bond with the sidechain of S631(D).In addition, molecular recognition seems to be the result of favorable hydrophobic interactions/contacts with N633(D), N588(C), and G628 (D).A different binding mode is instead returned by (S,S)-8 which is predicted to engage an H-bond interaction with the backbone of G628 (B) (via its urea moiety) and hydrophobic interactions with G628 (D) and N588 (B).The picture that emerged from this analysis indicates that (R,R)-8 and (R,S)-8 fit better than (S,S)-8 in the hERG binding site.This seems to be the result of a better orientation of the urea scaffold allowing the establishment of a well-oriented dual H-bond interaction with the protein cavity.Interestingly, the computed binding free energies are in agreement with the performed visual inspection, with (R,R)-8 and (R,S)-8 outperforming (S,S)-8 in terms of molecular mechanics/ generalized Born surface area (molecular mechanics/generalized Born surface area [MM-GBSA]) score (60.3 and 58.9 kcal/mol vs. 45.7 kcal/mol).These data, taken as a whole, support the hypothesis, which is consistent with the available literature, [44,45] whereby the hERG molecular recognition might be stereoselective, and underline the importance of S631 and G628 in the interaction between hERG and urea-based activators.Note that this hypothesis is in full agreement with the experimental literature displaying that the hERG double mutant G628C/S631C is insensitive to (R,R)-8. [27]I G U R E 2 Crystal structures of (S,S)-8 (a, c) and (R,S)-8 (b, d).(a, b) Asymmetric unit with color legend: carbon (light gray), hydrogen (white), oxygen (red), nitrogen (blue).(c, d) Packing in the crystal cell, where all hydrogen atoms have been removed for clarity.The unit cell of both molecules is shown in perspective view along the c-axis.

| hERG binding activity
The newly synthesized (S,S)-8 and (R,S)-8 were tested on hERG wildtype channels.In particular, the effect of each compound was evaluated on the activation (Figures 4a and 5a,b), inactivation (Figures 4b,c and 5c,d) and deactivation (Figures 4d and 5e channel activation (V 0.5 ), a double pulses protocol was employed.
The hERG activation currents were determined by a holding potential of -80 mV, where the activation gate is closed and the channel is in its stable nonconducting state.Then, to allow activation to reach as close as possible to the steady state, the cells were stepped by pulses between -60 and +80 mV for 7 s.Then, currents were repolarized to more hyperpolarized potentials at -60 mV (for 3 s) to generate deactivating tail currents whose amplitude is related to the proportion of channel recovering from inactivation.The recorded currents (examples in Figure 4a) were measured at the end of the 7 s pulse (steady-state currents), were normalized to the maximal evoked current, and graphed as a function of the voltage used to build the steady-state activation I/V plot (Figure 5).The hERG steady-state current increases progressively and then slowly decreases to more positive voltage values, indicating that the voltage-dependent inactivation behavior characteristic of hERG current (control [CTL]) is maintained after the perfusion with the SS or RS isomers.The S,S isomer decreases the outward currents, and (R,S)-8 induces a right shift in the steady-state curve without distorting particularly the bell-shaped curve.To investigate the possible effect of (R,S)-8 on the pharmacologically induced LQTS, the inhibitor E4031 (50 nM) was perfused.By comparing the I/V plot of steady-state currents of (R,S)-8 versus (R,S)-8+E4031 (Figure 5a), a statistical difference from -20 to +80 mV (p < 0.05) was revealed.To study the voltage dependence of activation, we analyzed the peak tail current amplitude that is elicited by a single potential (-60 mV) and reflects the proportion of channels activated during the previous depolarizing pulse (Figure 4a).The peak currents were normalized to the maximal evoked current and graphed as a function of the voltage used; the activation curve was fitted with a sigmoidal Boltzmann function (Figure 5b).The peak activation current of (R,S)-8, differently from (S,S)-8, shows statistically significant effect in comparison with the control conditions CTL: V 0.5 = -12.19± 2.30, k = 8.30 ± 1.45 versus (S,S)-8: V 0.5 = -12.45± 3.24, k = 9.59 ± 2.39; CTL: V 0.5 = -12.34± 1.00 mV, k = 7.62 ± 0.59 versus (R,S)-8: V 0.5 = -24.09± 2.05, k = 8.32 ± 1.32 (p < 0.05).
-Inactivation.A three steps protocol was used to investigate the steady-state inactivation (Figures 4b and 5c): after the application of a holding potential of -80 mV, the cells were depolarized at +60 mV for 200 ms to ensure the full inactivation of the hERG channel, then repolarized from -100 mV to +20 mV for 10 ms and repolarized at +60 mV for 200 ms.The currents evoked at this last pulse were measured and normalized to the maximum current and fitted with the sigmoidal Boltzmann function (Figure 5c) (CTL: V 0.5 = -48.37 ± 0.6, k = 17.20 ± 0.68 vs (S,S)-8: V 0.5 = -41.28± 1.03, k = 13.27 ± 1.00; CTL: V 0.5 = -50.96± 0.6, k = 14.4 ± 0.59 vs (R,S)-8: V 0.5 = -37.06± 0.75, k = 17.11 ± 0.81).This analysis was carried out to compare the perturbating effect of SS and RS on hERG current versus the control conditions.There is a statistically significant difference for the treatment with (S,S)-8 (p < 0.05) from -80 mV, for CTL versus (R,S)-8 (p < 0.05) from -60 to 0 mV (p < 0.05).To investigate more in deep the inactivation phase, we have studied the onset rate of inactivation by using a three-pulses voltage clamp protocol shown in Figure 4c.Depolarizing pulse at +60 mV for 200 ms was employed to inactivate the channel, then a hyperpolarizing pulse at -100 mV for 10 ms was used to allow inactivation recovery, and finally, pulses between +60 and -20 mV (300 ms) elicited the outward inactivating currents (Figure 4c).Summing up these findings, (RS)-8 shows a negative shift in the inactivation as previously reported for the RR enantiomer, and the SS enantiomer was the less potent stereoisomer, according to the in silico predictions.The different binding modes returned by (S,S)-8 docking simulations are probably responsible for the impaired ability to activate the channel while allowing it to act with the same mechanism of action as the other two isomers.Overall, we can state that the urea of mexiletine (8) retains the hERG opening activity of the parent compound and that the stereoselective behavior observed in our study confirms the stereoselectivity of binding reported in the literature for the hERG channel. [44]tably, both the tested stereoisomers proved to have mixed mechanisms by combining properties of the different types of the hERG potassium channel activators.Considering that Type 2 activators have been reported as drugs with a risk of repolarization overcorrection that could itself be pro-arrhythmic, [46,47] unlike Type 1 activators, [48,49] the observed mixed mechanism suggests a safer cardiac profile for the most potent RS isomer compared with both the previously reported RR enantiomer and the other Type 2 activators reported in the literature.

| Functional studies on cardiac parameters and smooth muscle function
To evaluate possible undesired effects, our compounds were tested in functional studies to evaluate their cardiac and vascular activities.
First of all, the three urea stereoisomers were in vitro tested for their antiarrhythmic activity on guinea pig isolated left atria driven at 1 Hz.The results were reported in Table 1 along with data for mexiletine taken as a reference compound.None of the stereoisomers increased the threshold of ac-arrhythmia, with the intrinsic activities being at the most 11%: such a mild effect was observed that the EC 50 values were not determined.Looking at the flip side, being known that hERG agonists are associated with a substantial risk of proarrhythmia, [13,46,47] no arrhythmogenic effect was observed, this result highlighting the inability of our compounds to alter cardiac rhythm as a side effect.
Both negative inotropic and chronotropic activities of the three stereoisomers under evaluation were then determined, obtaining dose-response curves, and the results were reported in Table 2 along with the data for mexiletine.Unlike the latter, which showed quite similar negative inotropic and chronotropic effects, compounds (R,R)-, (S,S)-, and (R,S)-8 showed an interesting selective negative inotropic effect, also on the spontaneously beating right atrium, being devoid of the chronotropic one.Their chronotropic intrinsic activity does not exceed 33%, thus confirming the previously discussed lack of the arrhythmogenic effect.It is noteworthy that the negative inotropic effect on the spontaneously beating right atrium was not recorded for mexiletine since this insignificant inotropic effect mainly results from the chronotropic effect.Furthermore, the selectivity for cardiac inotropism allows us to speculate about the possible preferential binding of our compounds with the Cav1.2 subtype, mainly involved in cardiomyocyte contractility, rather than Cav1.3, responsible for chronotropy. [51]It is noteworthy that the most interesting hERG opener (R,S)-8 displayed an inotropic EC 50 value higher than that of the RR and SS isomers and even one order of magnitude higher than that of mexiletine on the left atrium thus possibly acting as the safest stereoisomer in terms of cardiac profile.All compounds were also tested on K + -depolarized (80 mM KCl) guinea pig aortic strips (Table 3) and none of them displayed significant vascular effect, with the intrinsic vasorelaxant activity percentage on the aorta being lower than 30%, thus confirming a safe cardiovascular profile.
Finally, being known the undesired spasmolytic effect of Ca 2+ channel antagonists, the activity on K + -depolarized (80 mM KCl) guinea pig ileum smooth muscle was evaluated (Table 3), with the meso form [(R,S)-8] being devoid of possible intestinal side effects.
Unexpectedly, the results obtained so far pointed to possible offtarget pharmacological activities for the RR isomer.It was the most potent negative inotrope stereoisomer with an EC 50 value of 8.5 nM on the left atria.(R,R)-8 was 3.9-and 48-fold more potent than (S,S)-8 and (R,S)-8, respectively, and about fivefold more potent than mexiletine.A greater inotropic potency of (R,R)-8 on right atria in the spontaneous activity was also established, with its EC 50 value being 0.10 μM in comparison to 0.38 and 0.62 μM of (S,S)-8 and (R,S)-8, respectively.Furthermore, since no effect on the aorta was observed, it displayed a remarkable selectivity for cardiac functional parameters compared with vascular ones.Furthermore, when tested on the ileum smooth muscle, it displayed an IC 50 value very close to the EC 50 value determined on the left atrium, thus leading to some considerations: (1) a possible L-type calcium channel-mediated activity could be responsible for the observed inotropic activity, (2)   the RR isomer could also be suggested as a possible spasmolytic agent, [52] (3) more in-depth studies aimed at the dissociation of the two observed pharmacological activities are required and will hereafter be carried out.
Figure 6 illustrates the above considerations by comparing the potencies of mexiletine and its urea stereoisomers.The figure clearly shows the negative inotropic effect observed for all compounds, the selectivity of (R,S)-8 with respect to mexiletine, and the highest potency of (S,S)-8 both on the isolated guinea-pig left atrium and on ileum.

| Cytotoxicity
The cytotoxic effect was assessed by the 3-(4,5-dimethylthiazol-2yl)-2,5-diphenyltetrazolium bromide (MTT) test (Figure 7) on the human neuroblastoma (SH-SY5Y) cell line which was used to predict possible neural toxicity of our compounds being known that mexiletine clinical use is often associated with CNS toxicity. [16,19]xiletine was found to be nontoxic, showing an IC 50 value >100 μM, which is consistent with its current use as a drug.The IC 50 b Calculated from log concentration-response curves (Probit analysis according to Litchfield and Wilcoxon [50] with n = 6-8).When the maximum effect was <50%, the EC 50 values were not calculated.c At 10 -4 M.
T A B L E 2 Effects of the tested compounds on cardiac parameters.M concentration gave the maximum effect for most compounds.b Calculated from log concentration-response curves (Probit analysis by Litchfield and Wilcoxon [50] with n = 6-7).When the maximum effect was <50%, the EC values obtained for (R,R)-8, (S,S)-8, and (R,S)-8 were 72 ± 4 μM, >100 μM, and 52 ± 2 μM, respectively.Notably, the cytotoxicity of (R,R)-8 was observed at concentrations remarkably higher than those responsible for the negative inotropic effect (3-5 orders of magnitude), thus highlighting a safe profile for this compound.
Concerning the RS isomer, the cytotoxic IC 50 value was only twofold higher than the concentration used to investigate its hERG agonism.
However, further electrophysiological investigations on (R,S)-8 as an hERG opener at concentrations lower than those so far explored could be carried out to look for a wider gap between active and possibly toxic concentrations.

|
In summary, in the present work, we report on the stereoselective behavior of the mexiletine-derived symmetric and asymmetric urea 8 as an hERG opener, with its cardiovascular profile having also been determined.Electrophysiological investigations in HEK cells expressing hERG channels demonstrated that the meso form (R,S)-8 is the most potent activator of the hERG potassium channel among the three possible urea stereoisomers.The induced increase in the repolarizing potassium current could be helpful for the treatment of the LQTS since it could induce an action potential shortening.In addition, functional studies on guinea pig isolated left atria and aorta demonstrated its safe cardiovascular profile, unlike Type 2 hERG activators reported in the literature.It was not able to alter the cardiac rhythm, showing no proarrhythmic effect and being devoid of the chronotropic one, without even weakening the force of the cardiac contraction.The (R,S)-8 inotropic EC 50 value is one order of magnitude higher than that of mexiletine.Furthermore, it showed neither vasorelaxant activity on guinea pig aortic strips nor spasmolytic effect on guinea pig ileum smooth muscle, all these results together confirming a safe cardiovascular and intestinal profile.Due to its overall profile, (R,S)-8 may be considered a good starting point for the development of a novel congeneric series of hERG openers to treat both congenital and drug-induced forms of LQTS.The possible absence of unintended cardiovascular and nonvascular effects paves the way for the study of these mexiletine derivatives in in vivo models of LQTS.  a Percent inhibition of calcium-induced contraction on K + -depolarized (80 mM) guinea pig nonvascular (ileum) and vascular (aorta) smooth muscle at 5 × 10 -5 M. The 5 × 10 -5 M concentration gave the maximum effect for most compounds.

| Statistical analysis
Data were analyzed using Student's t test and are presented as mean ± S.E.M.Since the drugs were added in a cumulative manner, the difference between the control and the experimental values at each concentration was tested for a p value < 0.05.The potency of drugs defined as EC 50 and IC 50 was evaluated from log concentrationresponse curves (Probit analysis using Litchfield and Wilcoxon) [50] in the appropriate pharmacological preparations.reported by Rullo et al. [57] Absorbance values at 570 nm were measured using a multilabel plate counter Victor3 V (PerkinElmer), with dimethyl sulfoxide medium as the blank solution.Cells in DMEM alone represented negative control.Triplicate cultures were set up for each concentration of the tested compounds and each experiment was repeated three times.For each compound, IC 50 value was determined ± SD.Crystal structure determination was performed by EXPO software, [58] a package capable of carrying out the following steps: (a) determination of unit-cell parameters and identification of space group; (b) structure solution by direct methods and/or real-space approach; (d) structure model refinement by the Rietveld method. [59]e first low-angle well-defined peaks in the experimental diffraction pattern were selected and actively used for indexing via N-TREOR09 [60] and DICVOL04 [61] programs embedded in EXPO.
The space group determination was determined on the evaluation of the systematic absences.
The structures were solved with a real-space method based on the simulated annealing algorithm implemented in EXPO.The starting models were assembled using the sketching facilities of ACD/ ChemSketch [62] and the geometry optimization was achieved by the program MOPAC2016. [63]The simulated annealing algorithm was run 100 times under Linux workstation in default mode and in parallel calculation over 20 CPUs.The solutions with the lowest cost function value were selected.The criterion to accept the solutions was also based on the soundness of crystal packing.The solutions obtained by the direct-space method were also confirmed by direct methods.
) of I Kr flowing through the hERG channel expressed in human embryonic kidney (HEK) cells.-Activation.To study the activation and assess if the newly synthesized (S,S)-8 and (R,S)-8 influence the bell-shaped I/V relationship of hERG and avoid undermining the half point for the F I G U R E 3 Top-scored docking poses of (a) (R,R)-8, (b) (R,S)-8, and (c) (S,S)-8 within the human ether-à-go-go-related gene (hERG) binding site.Ligands and important residues are rendered as sticks, whereas the protein is represented as a surface.H-bonds are represented by dotted black lines.For the sake of clarity, only polar hydrogen atoms are shown.F I G U R E 4 Voltage clamp recordings of human ether-à-go-go-related gene (hERG) currents in HEK cells.The currents were elicited in HEK cells by the protocol indicated on the left of the figure (a−d).On the right, the recorded currents are reported in the control condition (CTL, in black) and after the perfusion of 25 μM of the compound of interest (S,S)-8 or (R,S)-8 in red.The arrows indicate the zero-current level.F I G U R E 5 Electrophysiological investigations in HEK cells expressing human ether-à-go-go-related gene (hERG) channels.Black squares indicate the value of current in basal condition (control = CTL) and the red dots indicate the values of current after the perfusion of 25 μM of the indicated compound (a−c).The I/V plots represent the steady-state activation currents (a), peak activation currents (b), and the steady-state inactivation currents (c).The gray triangles indicated the effect of (R,S)-8 on the steady-state current under the block of E4031 (50 nM).The histograms shown in (d) and (e) indicate the values of time constant in basal conditions (control = CTL, black columns) and after the perfusion of 25 μM of the indicated compound (red columns), for the inactivation and deactivation phase, respectively.In all panels is reported the number of cells for the single investigation and the standard error bar.
Inactivation currents were fitted by a single exponential function to obtain the time constants (tau).The mean tau values are represented by the histograms in Figure 5d.No statistically significant differences were observed in the slowed inactivation kinetics.-Deactivation.To determine the rates of deactivation of SS and RS on the hERG channel we used a two-step voltage protocol to elicit tail currents that were fitted by a biexponential function.Starting from a holding potential of −80 mV, the deactivation kinetics (Figures 4d and 5e) were examined in the voltage range from -100 to -20 mV after a depolarizing step at +60 mV to activate the hERG channel.During the second pulse phase, a growth of current is visible as the hERG channel recovers quickly from inactivation and then follows a current decay as the channel deactivates.The time course of the deactivating currents was fitted with a double exponential function to obtain the fast and slow time constants (tau) at −70 mV pulse.There is a statistically significant difference for CTL versus RS tau slow (p < 0.05).

T A B L E 1 a
Antiarrhythmic activity of tested compounds.CompdMax % increase of threshold of ac-arrhythmia after pretreatment with compounds a (M ± SEM) Increase of the threshold of ac-arrhythmia: increase in the current strength of 50 Hz alternating current required to produce arrhythmia in guinea pig left atria driven at 1 Hz in the presence of each tested compound at 5 × 10 -5 M. The 5 × 10 -5 M concentration gave the maximum effect for most compounds.

a
Decrease in developed tension on isolated guinea pig left atrium at 10 -5 M, expressed as percent changes from the control (n = 5-6).The 10 -5

5 M 5 M
developed tension on guinea pig spontaneously beating isolated right atrium at 10 -5 M, expressed as percent changes from the control (n = 7-8).The 10 -concentration gave the maximum effect for most compounds.d Decrease in atrial rate on guinea pig spontaneously beating isolated right atrium at 5 × 10 -5M, expressed as percent changes from the control (n = 7-8).The 10 -concentration gave the maximum effect for most compounds.Pretreatment heart rate ranged from 165 to 190 beats/min.
methods Chemicals were purchased from Sigma-Aldrich or Lancaster in the highest quality commercially available.Solvents were RP grade unless otherwise indicated.Yields refer to purified products and were not optimized.The structures of the compounds were confirmed by routine spectrometric analyses.Only spectra for compounds not previously described are given.Melting points were determined on a Gallenkamp melting point apparatus in open glass capillary tubes and are uncorrected. 1H NMR and 13 C NMR spectra were recorded on either a Varian VX Mercury spectrometer operating at 300 and 75 MHz for 1 H and 13 C, respectively, or an Agilent 500 MHz T A B L E 3 Activity of tested compounds on K + -depolarized guinea pig smooth muscle.

d
At 5 × 10 -8 M. F I G U R E 6 Each bar represents the -log EC 50 obtained from 6 to 8 independent experiments.When bar errors are not shown, they are covered by the corresponding points.Ino, inotropy; Chro, chronotropy; Arry, antiarrhythmic activity; Ao, aorta; Il, ileum.operating at 500 and 125 MHz for 1 H and 13 C, respectively, using CDCl 3 as solvent, unless otherwise indicated.Chemical shifts are reported in ppm relative to the residual nondeuterated solvent resonance: CDCl 3 , δ 7.26 ( 1 H NMR) and δ 77.3 ( 13 C NMR).J values are given in Hertz.Enantiomeric excess (ee) values were determined by chiral HPLC with an Agilent chromatograph (model 1100), equipped with a diode array detector, on a Daicel Chiralpak IA column.Electron ionization mass spectroscopy spectra were recorded on a Hewlett-Packard 6890-5973 MSD gas chromatograph/mass spectrometer at low resolution.The molecular ion is given as [M] + .Elemental analyses were performed on a Eurovector Euro 3000 analyzer, and the data for C, H, and N were within ±0.4 of theoretical values.Optical rotations were measured on a PerkinElmer Mod 341 spectropolarimeter; concentrations are expressed in g 100 mL -1 , and the cell length was 1 dm; thus, [α] 20 D values are given in units of 10 -1 deg cm 2 g -1 .Chromatographic separations were performed on silica gel columns (Kieselgel 60, 0.040-0.063mm, Merck).Thin layer chromatography (TLC) analyses were performed on precoated silica gel on aluminum sheets (Kieselgel 60 F254, Merck).TLC plates were visualized under UV light.The purity of the final compounds was determined by elemental analysis.

4. 3 . 2 |
Cell viability on neuroblastoma SH-SY5YImmortalized human neuroblastoma SH-SY5Y cell line was cultured in DMEM containing 4.5 g/L D-glucose and non-essential amino acids, supplemented with 10% heat-inactivated FBS, 1% L-glutamine 200 mM, 100 U/mL penicillin and 100 μg/mL streptomycin at 37°C in an atmosphere of 5% CO 2 .At confluence 80%-90%, the cells were harvested and seeded at a density of 4 × 10 4 cells/well in a 96-well plate for 24 h, to approximately 70%-80% confluency.After this time, the culture medium was replaced by fresh medium containing the compounds (R,R)-8, (S,S)-8, and (R,S)-8 used in the range of concentration from 0 to 100 μM, respectively.The cells were treated for 24 h at 37°C in 5% CO 2 with the indicated concentrations of the compounds and after the cell viability was tested by MTT[3-(4,5-  dimethylthiazol-2-yl)-2,5-diphenyl tetrazolium bromide] assay as

4. 4 |
Crystallographic analysis 4.4.1 | General X-ray diffraction powder data were collected at room temperature by a Rigaku RINT2500 diffractometer operating in transmission mode at Cu K α radiation (λ = 1.540560Å).Samples were put on a glass capillary and spun during data collection.Data were collected in the 2θ range from 5°to 80°.