The potential inhibitory activity of the compounds MI_a, MI_b, MI_c, MI_d, MI_e, MII_a, MII_b, MII_c, MII_d and MII_e on the evaluated Cytochrome P450 enzyme isoforms was measured using the corresponding specific fluorimetric screening kits CYP1A2 Inhibitor Screening Kit (Fluorometric) (Abcam) with 3-cyano-7-hydroxycoumarin (3-CHC) applied as substrate and α-naphthoflavone used as an inhibitor. The CYP2D6 Inhibitor Screening Kit (Fluorometric) (Abcam) with 3-[2-(N,N-diethyl-N-methylammonium)ethyl]-7-methoxy-4-methylcoumarin applied as a substrate and quinidine as an inhibitor, and CYP3A4 Inhibitor Screening Kit (Fluorometric) (Abcam) with resorufin applied as a substrate and ketoconazole as an inhibitor. The fluorescence for the different isoforms was measured at Ex/Em = 406/468 nm for CYP1A2; Ex/Em = 535/587 nm for CYP3A4 and Ex/Em = 390/468 nm for CYP2D6, respectively.

The CYP450 superfamily of enzymes are involved in the metabolic pathways of various pharmacologically active compounds. The assessment of CYP450-mediated interactions is important because of the involvement in undesirable drug-drug interactions. Therefore, an initial evaluation toward the CYP450 activity for each active molecule is required in the process of drug development (Manikandan et al. 2018). The examined structures were initially synthesized by Tzankova et al. 2023. The series MI_a-e and MII_a-e were studied for possible inhibitory effects on some isoforms of Cytochrome P450 – CYP1A2; CYP2D6, which revealed genetic polymorphism; and CYP3A4, which metabolized over 60 % of the drugs on the market.

The examined compounds (at concentration 1 µM) did not reveal any statistically significant inhibitory effects on the different isoforms (Figs 2–4). The risk for possible drug interactions between these compounds and drugs, substrates of the exam isoforms, does not exist. Inhibitory effects exerted only the classical inhibitors: α-Naphthoflavone, a classical CYP1A2 inhibitor, at concentration 1 µM, inhibited CYP1A2 enzyme activity with 50 %, compared to the control (pure CYP1A2) (Fig. 2).

Quinidine, a classical CYP2D6 inhibitor, at concentration 1 µM, inhibited CYP2D6 enzyme activity with 55%, compared to the control (pure CYP2D6) (Fig. 3).

Ketoconazole, a classical CYP3A4 inhibitor, at concentration 1 µM, inhibited CYP3A4 enzyme activity with 45%, compared to the control (pure CYP3A4) (Fig. 4).

The in vitro results demonstrated that none of the applied pyrrole-based compounds (MI_a-e and MII_a-e) are capable of inhibiting the CYP1A2, CYP2D6 and CYP3A4 isoforms. The results could be related to the unavailability of the MAO-B inhibitors to be placed in the active site of the CYP450 isoforms. It could be hypothesized that the presented compounds are metabolized by non-specific arylamidase. The latter theory is in relation with the similarity of the title compounds with the antidepressant Isocarboxazid. Isocarboxazid (Marplan) is a non-selective, irreversible monoamine oxidase (MAO) inhibitor classified as a hydrazide antidepressant. An arylamidase purified from guinea pig liver microsomes is known to be responsible for the hydrolysis of Isocarboxazid (Moroi and Sato 1975), therefore, the hydrolysis of the hydrazide group to a hydrazine and an acid could be plausible for the title structures. Similar metabolism after hydrolysis was discussed for another hydrazide-containing compound – Ipronizide (Nelson et al. 1978). The study was followed by the development of an in silico approach to examine the active conformations and the scores of the complexes ligand-CYP450.

Figure 2. 

Effects of two series MI_a-e and MII_a-e, as well as α-Naphthoflavone (at concentration 1 µM), on CYP1A2 enzyme activity. *** P < 0.001 vs control (pure CYP1A2).

Figure 3. 

Effects of two series MI_a-e and MII_a-e, as well as Quinidine (at concentration 1 µM), on CYP2D6 enzyme activity. *** P < 0.001 vs control (pure CYP2D6).

Figure 4. 

Effects of two series MI_a-e and MII_a-e, as well as Ketoconazole (at concentration 1 µM), on CYP3A4 enzyme activity. *** P < 0.001 vs control (pure CYP3A4).

The in silico calculations could provide data about the additional optimization of the novel compounds (Iliev et al. 2023), thus molecular docking in the active sites of the CYP450 isoforms was conducted. The conducted dockings simulations evaluated the correlation between the experimental and the theoretical results.

The validation of the proteins were done through redocking procedures provided in our recent study (Mateev et al. 2022). After the validation of the crystallographic structures, all 10 ligands were docked into the three CYP isoforms to observe the potential correlation between the in vitro and in silico results. The data from the docking simulations is provided in Table 1. The co-crystallized ligands were docked into their native enzymes for reference.

From the docking calculations, it was found that only MI_b could be situated in the active site of CYP1A2. However, the compounds showed docking score of -2.66 kcal/mol and returned no MM/GBSA data, not IFD. Moreover, the native CYP1A2 inhibitor – alpha-naphthoflavone, demonstrated significantly lower binding energies, which contributes to more stable complex.

The simulations in the active site of CYP2D6 also correlated well with the in vitro data. Only compound MI_d returned a score of -3.13 after applying the XP docking of Glide. However, no active poses were observed after using IFD and MM/GBSA modes. Moreover, quinidine showed score of -9.89 kca/mol which is not comparable to the pyrrole-based compound.

Interestingly, the results from the docking simulations in CYP3A4 demonstrated that all ligands can be situated in the active site of CYP3A4 and good XP scores were recorded. Subsequently, IFD docking led to good results as well. However, after rescoring the enzyme-ligand complexes, it was found that only MI_c, MII_e, MI_a and MII_b returned scores of -13.82, -6.35, -5.39 and -3.37 kcal/mol, respectively. The scores were not comparable to the native CYP3A4 inhibitor ketoconazole which demonstrated MM/GBSA score of -117.64. Therefore, it could be concluded that none of the ligands poses significant CYP3A4 inhibition capacity. However, MI_c and MII_e returned low binding scores, which do not correlated with the in vitro experiments where all compounds showed zero inhibition capacity. Therefore, the next step was to analyse and visualize the active conformation of both pyrrole-based compounds in 2D and 3D paneles (Fig. 5).

Based on the visualized stabilization forces, the hem group and Phe304 were the major drug-binding residues in CYP3A4. The Hem group formed a stable hydrogen bond with thep-bromophenyl moiety of MI_c. However, in the case of MII_e, only hydrophobic interactions were found. Therefore, MI_c formed more stable complex with the active site of CYP3A4 and the lower binding energy was observed. Several weak hydrophobic interactions were formed between the two ligands and the active amino residues Ile120, Leu210, Leu211, Phe241, Ala305, Ile369, Leu482 and Leu483. After the XP docking, the active pose of MI_c demonstrated that the carbonyl group from the ester group was in close proximity to the hem structure. In contrast, the more hardware demanding, and more precise IFD displayed different orientation in the active site of CYP3A4. During the latter simulations, the p-bromophenyl moiety in the pyrrole ring was facing the Hem moiety and a stable hydrogen bond was formed. Thus, the implementation of IFD for virtual simulations in various CYPs isoforms is essential considering the enhanced reliability of the acquired results, which was also underlined in recent studies (Angelov et al. 2022; Mateev et al. 2022).

Figure 5. 

Visualized major intermolecular interactions of the pyrrole-based compounds MI_c (A, B) and MII_e (C, D) in the active site CYP3A4 (PDB: 2V0M) after employing IFD and MM/GBSA recalculations. The interactions are provided in both 3D and 2D forms. The enzyme is depicted in grey, the active amino acids were shown and ligands are presented as green sticks.