Focused design of polypharmacophoric neuroprotective compounds: Conjugates of γ-carbolines with carbazole derivatives and tetrahydrocarbazole

Kinetic measurements

The results summarized in Table 1 demonstrate that all γ-carboline conjugates with carbazoles (3a–o) and tetrahydrocarbazole (5a–f) selectively inhibit BChE, while the activity towards AChE and CaE is low. Most of these compounds are more potent BChE inhibitors than the basic pharmacophores – 2,8-dimethyltetrahydro-γ-carboline, carbazole, and Dimebon. All conjugates with unsubstituted carbazoles (3a–f) and tetrahydrocarbazole (5a–f) show high inhibitory activity towards BChE with the IC₅₀ values of about 1–6 μM, the replacement of carbazole by tetrahydrocarbazole having no effect on the anti-BChE activity.

For the conjugates of unsubstituted carbazoles (3a–f) (R₂=H), a change in the structures of R and R₁ groups in the γ-carboline moiety has virtually no effect on the anti-BChE activities: all of compounds (3a–f) are potent inhibitors of the enzyme. Meanwhile, in the case of γ-carboline conjugates with dichloro- or dibromo-substituted carbazoles, (R₂=Cl: 3g–k) and (R₂=Br: 3l–o), the degree of BChE inhibition depends on the structure of R and R₁. Among compounds with dichloro-substituted carbazoles, the highest activity was found for 3g (R=H, R₁=CH₃) and 3j (R=F, R₁=CH₃) characterized by IC₅₀=3.48±0.69 and IC₅₀=1.93±0.03, respectively. Out of the dibromo-substituted carbazoles, the activity was maximum for 3l (R=H, R₁=CH₃), 3n (R=СН₃, R₁=C₂H₅), and 3o (R=F, R₁=CH₃) with IC₅₀ of 2.69±0.25, 3.16±2.1, and 1.18±0.06, respectively.

Thus, the lead compounds in this series (Table 1) are all conjugates with unsubstituted carbazoles 3a–f and tetrahydrocarbazole 5a–f and dichloro derivatives 3g (R=H, R₁=CH₃) and 3j (R=F, R₁=CH₃) and dibromo derivatives 3l (R=H, R₁=CH₃), 3n (R=СН₃, R₁=C₂H₅), and 3o (R=F, R₁=CH₃). One more conclusion that can be drawn from the data of Table 1 is that the conjugates formed by γ-carboline derivatives with R=F and R₁=CH₃, that is, 3e (R₂=H), 3j (R₂=Cl), and 3o (R₂=Br), show high anti-BChE activity, irrespective of the structure of the carbazole moiety.

For the series of active compounds 3a, 3c, 3k, 3n, 5a, 5c, and 5e, the kinetics and mechanism of BChE inhibition were studied in detail. The kinetic data were analyzed using the Lineweaver–Burk double reciprocal plot (Fig. 2). The resulting values of BChE inhibition constants (K_i is the competitive component and αK_i is non-competitive component) are summarized in Table 2. As shown in Table 2, all the γ-carboline conjugates with carbazoles and tetrahydrocarbazole are mixed-type BChE inhibitors.

Fig. 2:

Steady-state inhibition of BChE by compounds 3c (a) and 5e (b). Lineweaver-Burk reciprocal plots of initial velocity and substrate concentrations in the presence of inhibitors 3c, 5e (three concentrations) and without the inhibitors are presented. The plots a and b show mixed-type inhibition.

The mechanism of inhibitory action of γ-carboline conjugates with carbazoles and tetrahydrocarbazole is illustrated in relation to two of the active compounds 3с and 5е. The graphical analysis of kinetic data on the BChE inhibition by compounds 3с and 5е in the Lineweaver–Burk plot is presented in Fig. 2.

The tested compounds are combinations of various pharmacophores: 3с contains the γ-carboline and carbazole moieties, while 5e comprises the γ-carboline and tetrahydrocarbazole units. As observed in Figs. 2a and b, binding of either 3c or 5e to BChE induces changes in both V_max and K_m, which is typical of mixed inhibition. Variation of the pharmacophore combination does not affect the mechanism of BChE inhibition by these compounds. The inhibition constants were determined: for compound 3с, K_i=2.55±0.18 μM (competitive component) and αK_i=12.5±0.7 μM (non-competitive component), and for compound 5е, K_i=3.89±0.15 μM (competitive component) and αK_i=10.2±1.1 μM (non-competitive component).

Thus, unlike the reference compounds, 2,8-dimethyltetrahydro-γ-carboline and carbazole, their conjugates are potent and selective BChE inhibitors efficient in the micromolar range. The esterase profile of the main reference compound, Dimebon, also changes. The BChE inhibition is reversible and corresponds to the mixed type.

Molecular modeling

According to the molecular docking of conjugates into BChE, they were found to bind both to PAS in the middle of the gorge and at the gorge bottom where they interact with the active site amino acid residues (Fig. 3), which accounts for the mixed type of BChE inhibition observed for these compounds.

Fig. 3:

Docked poses of compound 5a at the gorge bottom (the carbon atoms of 5a are shown in pink) and the same compound in the middle of BChE gorge (the carbon atoms are shown in blue). The yellow dashed line indicates the hydrogen bond with a catalytic serine residue, the orange dashed line shows the π-cationic interaction, and the blue one stands for the salt bridge with the PAS amino acid residues. Presented poses correspond to the most populated clusters after 256 AutoDock runs (see more parameters in the Methods section).

The positions in the PAS are energetically preferred, but at the same time, they correspond to less specific interactions of the carbazole and tetrahydrocarbazole moieties. The replacement of carbazole in the conjugate molecule by tetrahydrocarbazole, for example, for compounds 3a/5a does not affect the anti-BChE activity (Fig. 4a).

Fig. 4:

(a) Docked positions of conjugates with the carbazole 3a (carbon atoms are shown in orange) and tetrahydrocarbazole moieties 5a (carbon atoms are shown in pink) with identical substituents in the γ-carboline moiety in the BChE active site. (b) Docked positions of tetrahydrocarbazole 5b (carbon atoms are shown in white) and 5e (carbon atoms are shown in red) conjugates with various substituents in the γ-carboline moiety in the BChE active site. Presented poses correspond to the most populated clusters after 256 AutoDock runs (see more parameters in the Methods section).

Similarly, in some cases, the substituent in the γ-carboline moiety located outside the contact area with active site amino acids (Fig. 4b) may also have no effect on the position of molecule in the active site and, hence, on the inhibitory activity, as indicated by comparison of the conjugates formed by tetrahydrocarbazoles 5b and 5e.

Meanwhile, comparison of the positions of compounds with different substituents in γ-carboline for the series of dichloro- and dibromocarbazoles demonstrated that, although the radii and properties of these two halogens differ little, in some cases, this can dictate the preferred position of the dihalogenated conjugate in the BChE active site and, hence, affect the inhibitory efficiency to some extent. Indeed, the positions are very similar for compounds 3g, 3l, and 3n, whereas the position of less active compound 3i is different (Fig. 5), although the difference is not very large.

Fig. 5:

Docked positions of dichloro- and dibromocarbazole conjugates with various substituents in the γ-carboline moiety, compounds 3l (carbon atoms are shown in green), 3n (carbon atoms are shown in yellow), and 3g (carbon atoms are shown in pink), and compound 3i (carbon atoms are shown in turquoise), which has a somewhat lower binding affinity. The yellow dashed lines show hydrogen bonds with BChE amino acid side chains. Presented poses correspond to the most populated clusters after 256 AutoDock runs (see more parameters in the Methods section).

As regarding the weak interaction of the γ-carboline–carbazole and γ-carboline–tetrahydrocarbazole conjugates with AChE, it is known that the major difficulty involved in binding of compounds with bulky groups to AChE is to overcome the so-called bottleneck, a narrow part of the gorge formed by the Tyr341, Tyr124, and Tyr337 side chains [78]. Furthermore, compounds without a fixed positive charge (quaternary nitrogen atom), like the conjugates in question, can form non-specific contacts in the AChE peripheral anionic site and on the surface, thus partly inhibiting the enzyme [79]; this accounts for the low inhibitory activity of the studied conjugates towards AChE.

The low inhibitory activity of γ-carboline conjugates with carbazoles and tetrahydrocarbazole towards AChE indicates that these compounds would not induce undesirable adverse effects inherent in AChE inhibitors. The lack of inhibitory activity towards CaE in these compounds implies the absence of drug-drug interactions typical of CaE inhibitors.

In vitro AChE, BChE and CaE inhibition

Acetylcholinesterase (AChE, EC 3.1.1.7, from human erythrocyte), butyrylcholinesterase (BChE, EC 3.1.1.8, from equine serum), carboxylesterase (CaE, EC 3.1.1.1, from porcine liver), acetylthiocholine iodide (ATCh), butylthiocholine iodide (BTCh), 5,5′-dithiobis-(2-nitrobenzoic acid) (DTNB), and 4-nitrophenol acetate (4-NPA), were purchased from Sigma-Aldrich (Germany).

AChE and BChE activities were measured by the method of Ellman and coworkers as described earlier [82]. The assay solution consisted of 0.1 M K/Na phosphate buffer pH 7.5, 25°C with the addition of 0.33 mM DTNB, 0.02 unit/mL of AChE or BChE and 1 mM of substrate (ATCh or BTCh, respectively). Assays were carried out with a blank containing all components except ATCh and BTCh.

The activity of CaE was determined spectrophotometrically by the release of 4-nitrophenol at 405 nm [83]. The assay solution consisted of 0.1 M K/Na phosphate buffer pH 8.0, 25°C, with the addition of 1 mM 4-nitrophenyl acetate and 0.02 unit/mL of CaE. Assays were carried out with a blank containing all components except CaE.

The tested compounds were dissolved in DMSO; the incubation mixture contained 2% of the solvent. Eight different concentrations of the test compounds in the range of 10⁻¹¹–10⁻⁴ M were selected in order to obtain inhibition of AChE and BChE activity between 20% and 80%. The test compounds were added to the assay solution and preincubated with the enzymes for 5 min at 25^oC followed by the addition of substrate. A parallel control was made for the assay solution with no inhibitor. Measurements were performed in a BioRad Benchmark Plus microplate spectrophotometer (France). Each experiment was performed in triplicate. The results were expressed as the mean±SEM. The reaction rates in the presence or absence of inhibitor were compared, and the percentage of residual enzyme activity due to the presence of test compounds was calculated. The IC₅₀ (the concentration of inhibitor required to decrease the enzyme activity by 50%) values were determined graphically from inhibition curves (log inhibitor concentration vs percent residual enzyme activity) using the Origin 6.1 software.

Kinetic analysis of BChE inhibition. Determination of steady-state inhibition constants

To elucidate the inhibition mechanisms for the most active compounds, the BChE residual activity was determined in the presence of three increasing concentrations of the test compounds and six decreasing concentrations of the substrates. The test compounds were preincubated with the enzymes at 25°C for 5 min, followed by the addition of the substrates. Parallel controls were made for an assay of the rate of hydrolysis of the same concentrations of substrates in the solutions with no inhibitor. The kinetic parameters of substrate hydrolysis were determined. Measurements were performed in a BioRad Benchmark Plus microplate spectrophotometer (France). Each experiment was performed in triplicate. Results were fitted into Lineweaver-Burk double reciprocal kinetic plots of 1/V versus 1/[S] and values of inhibition constants K_i (competitive component) and αK_i (noncompetitive component) were calculated using the program Origin 6.1.

ABTS radical cation scavenging activity assay

Radical scavenging activity of the compounds was assessed using an ABTS˙⁺ radical decolorization assay. The procedure followed the method [80] with some minor modifications.

ABTS (2,2′-azino-bis-{3-ethylbenzothiazoline-6-sulfonic acid} diammonium salt) was purchased from TCI (Tokyo, Japan), potassium persulfate (di-potassium peroxodisulfate), Trolox^® (6-hydroxy-2,5,7,8-tetramethychroman-2-carboxylic acid (Sigma-Aldrich), and ascorbic acid were obtained from Sigma-Aldrich Chemical Co. (St. Louis, MO, USA). Ethanol was HPLC grade. Aqueous solutions were prepared using deionized water.

Trolox and ascorbic acid were used as the antioxidant standards. Five millimolarity solutions of standards were prepared in DMSO for use as stocks. Fresh working solutions of the standards of known concentrations (1–100 μM) were prepared on the day of experiments and used for calibration and as positive controls for ABTS˙⁺ scavenging activity.

ABTS was dissolved in deionized water to a 7 mM concentration. The solution of ABTS radical cation (ABTS˙⁺) was produced by mixing 7 mM ABTS stock solution with 2.45 mM potassium persulfate aqueous solution in equal quantities and allowing them to react for 12–16 h at room temperature in the dark. At the time of activity, ABTS˙⁺ solution was diluted with ethanol to adjust an absorbance value of about 0.80±0.02 at 734 nm. Fresh working ABTS˙⁺ solution was prepared for each assay.

The radical scavenging capacity of the compounds was analyzed by mixing 10 μL of compound with 240 μL of ABTS˙⁺ working solution. The reduction in absorbance was measured spectrophotometrically at 734 nm after 1 h of mixing the solutions using the microplate UV/VIS spectrophotometer BioRad xMark (Japan). EtOH blanks were run in each assay. Values were obtained from three replicates of each sample and three independent experiments.

The antioxidant activity is reported as Trolox equivalent antioxidant capacity (TEAC values) by comparing (A₀–A_test) of the test antioxidant with (A₀–A_Trolox) of the Trolox standard at concentration of 20 μM after reaction time of 1 h, where A₀ is the absorbance of a control lacking any radical scavenger and A_test/Trolox is the absorbance of the remaining ABTS˙⁺ in the presence of the test compound or Trolox:

T E A C = ( A 0 − A T r o l o x ) / ( A 0 − A t e s t ) .

For the most active compounds, we also determined the IC₅₀ values (compound concentration required for 50% reduction of ABTS radical). The IC₅₀ values were calculated using Origin 6.1 for Windows (OriginLab, Northampton, MA, USA).

Rat liver mitochondria isolation

Rat liver mitochondria were isolated by conventional differential centrifugation from the livers of adult Wistar strain rats fasted overnight. The isolation was carried out in an ice-cold isolation buffer (IB), pH 7.6 [81]. The mitochondrial protein concentration was determined using a biuret procedure with bovine serum albumin as the standard [84].

Mitochondrial potential and permeability transition

Safranine O (10 μM) was used as a membrane potential probe [85]. Fluorescence intensity at 580 nm (excitation at 520 nm) was measured with Victor3 multi-well fluorescence plate reader (Perkin Elmer, Germany). Mitochondrial protein concentration was 0.2 mg/mL. The medium for measurements contained 75 mM sucrose, 225 mM mannitol, 10 mM K-HEPES (pH 7.4), 0.02 mM EGTA, 1 mM KH₂PO₄. After a 5-min incubation, 5 mM glutame/malate or the 5 mM succinate and 0.5 μM rotenone were added to produce the mitochondrial potential. Then the compounds (30 μM) or the same volume of vehicle (DMSO) were injected to mitochondrial suspension. After 15–20 min, 12.5 μM CaCl₂ was added to each probe to induce the depolarization of mitochondria.

Сa²⁺-induced mitochondrial swelling was used to study the mitochondrial permeability transition induction. The mitochondrial swelling was determined by monitoring the absorbance at 620 nm using a Victor3 multi-well fluorescence plate reader (Perkin Elmer, Germany). The rat liver mitochondria (0.2 mg/mL) were incubated in a buffer containing 75 mM sucrose, 225 mM mannitol, 10 mM K-HEPES (pH 7.4), 0.02 mM EGTA, 1 mM KH₂PO₄, 5 mM succinate, and 0, 5 ϻM rotenone for energized conditions. The mitochondrial suspensions (0.2 mg/mL) were infused with the 25 μM CaCl₂ solution.

Tubulin polymerization

The assembly of tubulin into microtubules was carried out using pure tubulin from HTS-Tubulin polymerization assay kit (Cytoskeleton, Inc., Denver, USA). Tubulin polymerization assay was based on an adaptation of the original method of Shelanski et al. [86], who demonstrated that light is scattered by microtubules to an extent that is proportional to the concentration of microtubule polymer. The standard polymerization reaction contained 100 μL of a 4 mg/mL tubulin solution in 80 mM PIPES (pH 6.9), 0.5 mM EGTA, 2 mM MgCl₂, and 1 mM GTP. Polymerization was monitored by observing the change in the absorbance on a Victor microplate reader (Perkin Elmer, Finland) at λ=355 nm. The parameter V_max=(dA₃₅₅/dt)^max normalized to control probes (%) was used to compare the action of the compounds. Electron microscopic monitoring was carried out on A Carl Zeiss Libra 120 Electron Microscope (Carl Zeiss Meditec AG, Jena, Germany) at 120 kV using negative contrasting [87].

Statistical analyses

Each experiment was performed in triplicate. Average results are presented as the mean±S.D.

Protein and ligand structure preparation

To determine the protonation state of the piperidine nitrogen atom of γ-carboline moiety and other ionizable groups of the compounds, Marvin 14.9.1.0 (ChemAxon, http://www.chemaxon.com) was used to estimate pK_a values. Since they were found to be close to 7.4, protonated and neutral forms were used for molecular docking of all molecules.

Geometries of the ligands were quantum-mechanically (QM) optimized in Gamess-US package [88] using DFT method B3LYP and basis 6-31G*. Partial atomic charges were taken from QM results according to Mulliken scheme [89]. These optimized geometries and partial charges were used for molecular docking.

The X-ray structure of human BChE (PDB ID 1P0I [90]) was used. Previously the importance of saturation of BChE gorge with water molecules was demonstrated [91]. Protein structure was prepared, saturated with water molecules, and optimized using QM/MM method as reported previously [91], [92].

Molecular docking

Molecular docking with a Lamarckian Genetic Algorithm [93] was performed with Autodock 4.2.6 software [94]. Grid box for docking included the whole BChE active site and the gorge with dimensions 15 Å×20.25 Å×18 Å with grid spacing 0.375 Å. The main of selected Lamarckian Genetic Algorithm parameters were 256 runs, 25×10⁶ evaluations, 27×10⁴ generations and population size 300. Docking positions with the lowest binding energies were used for analysis. Structural images were prepared with Accelrys Discovery Studio Visualiser 4.0 (http://www.accelrys.com), PyMOL (Schrödinger, LLC).