Multiscale computational investigation of β-himachalene oxidation: Density functional theory-based reactivity, chemo- and stereoselectivity, molecular docking, and ADME predictions

3.1. Analysis of the global molecular descriptors of β-himachalene and potassium permanganate

To better understand the electronic interactions between β-himachalene and potassium permanganate (KMnO₄), global molecular descriptors were calculated using DFT and have been presented in Table 1. These indices, such as chemical hardness, chemical potential, global electrophilicity index, and nucleophilicity power, allow for the characterization of a molecule’s tendency to accept or donate electrons. The comparative analysis of these parameters provides valuable information on the nature of the electron transfer involved in this oxidation reaction and contributes to the interpretation of the chemoselectivity observed experimentally [7].

The fact that the chemical hardness (η = 3.89 eV) of potassium permanganate is moderate implies that its electronic density is rather easily modifiable. This rather small value suggests that KMnO₄ is a ‘soft’ species, subject to very strong reactivity and, as is well known, a powerful oxidizing agent. By contrast, β-himachalene is considerably harder (η = 6.55 eV), bearing a lower reactivity in general, consistent with its nature as a stable polycyclic hydrocarbon. The chemical potential (μ), a measure of the reactivity of a species toward electron acceptance or donation, is much more negative for β-himachalene (-2.44 eV) compared to permanganate (-1.01 eV). This indicates that β-himachalene is more electron capturing. The global electrophilicity index (ω) determines the extent to which a species can accept electrons. β-Himachalene has a greater value compared to KMnO₄ (ω = 0.13 eV), which translates to a relatively stronger propensity for electron donor interactions. The nucleophilicity (N) also estimates the capability of a system to donate electrons. (4) KMnO₄ has a relatively high value (N = 6.39), which might appear as a paradox, but is actually due to its high global reactivity. In contrast, β-himachalene possesses a lower nucleophilicity (N = 3.63), which means a weaker but focused ability to embed in the electron transfer process. In conclusion, this study underlines an electron transfer electronic complementarity between KMnO₄ and β-himachalene, oxidant being an electron donor, and β-himachalene, despite its relative hardness and stability, can act as an electrophile in the oxidation process.

3.2. Prediction of oxidation sites of β-himachalene by KMnO₄ based on the local electrophilicity function

In as much as reactivity this reactivity of β-himachalene is concerned, it is important to know what are the sites in this molecule which undergo chemical modification, especially in the presence of a KMnO₄ equivalent; the local electrophilicity function of Parr [19], coated with theoretical framework, enables us to map reactivity with high precision, revealing the sites in the molecule being very sensitive to nucleophilic attacks. This section provides an analysis of these electrophilic functionalities, with the purpose of predicting the sites that would be more susceptible to oxidation by KMnO₄, and can aid in interpreting the chemoselectivity found experimentally. Figure 1 shows the values of the local electrophilic functions calculated for the major reactive sites of β-himachalene.

The mapping function of the local electrophilic properties of Parr for the β-himachalene molecule has been presented in Figure 1, to look for the potential oxidation centers of the potassium permanganate (KMnO₄). The latter, being a strong oxidizing agent, exhibits a marked preference for attacking unsaturated systems, such as carbon–carbon double bonds, as well as electrophilically activated centers. The examination of the local analysis of the P⁺ values reveals which areas of the molecule are most susceptible to a nucleophilic attack by MnO₄⁻ of the carbon atoms examined, C₇ has the largest P⁺ value of 0.28 and is thus the most electrophilic. This is followed by atoms C₂ (0.25) and C₆ (0.24), both of which also show a potential reactivity. Carbon C₃ (0.17) seems to be less reactive, although it can also be involved with strong or persistent oxidative situations. The fact that carbons C₆ and C₇ are highly electrophilic suggests that coupling of a double bond between these carbons speeds up the directed oxidative attack by MnO₄⁻, resulting in the formation of diols. A gentle oxidation (dilute MnO₄⁻ at room temp) would have the diols forming on this double bond. Finally, the analysis enables explicit prediction of the favored reactor centers of β-himachalene in the oxidation by KMnO₄, and this may rationalize the experimentally observed chemoselectivity [7].

Close

Figure 1.

3D visualization map of the local electrophilic Parr functions (P⁺) of β-himachalene.

Export to PPT

3.3. Theoretical analysis of the energy profile and of the chemo- and stereoselectivity

To rationalize the preferential formation of product A1 in the oxidation of β-himachalene, a theoretical analysis of the reaction energy profiles was performed (Scheme 1). This method permits to compare of the diverse reaction pathways by looking at the activation barriers (ΔG^‡) and the free energies of the corresponding products [20]. The goal here is to separate mechanisms in kinetic versus thermodynamic control and then test the theory to see if it matches observed chemo- chemo-stereoselectivity. The thermodynamic quantities of the four processes have been listed in Table 2, and the corresponding energy profiles have been shown in Figure 2.

Close

Scheme 1.

Overall scheme of the four mechanistic pathways leading to the hydroxylated products A1-F to A4-F resulting from the oxidation of β-himachalene by KMnO₄ via the transition states TS-A1 to TS-A4.

Export to PPT

Close

Figure 2.

Energy profile of the reaction leading to A1-A4 and their transition states (TS-A1 to TS-A4), expressed in ΔG (kcal/mol)

Export to PPT

A corresponding reaction energy profile in further Gibbs free energy expression, ΔG in kcal/mol, has been presented in Figure 2, where four reactions, which yield respective three complexes, A1, A2, A3, and A4, each one passing via the transition state denoted again as TS-A1 to TS-A4. It allows for the assessment of the kinetic feasibility via the activation barriers, ΔG^‡, as well as the thermodynamic stability of the resulted products.

Kinetically, the lowest activation barrier is for pathway A1 (TS-A1=24.27 kcal/mol) followed by pathway A3 (TS-A3=24.89 kcal/mol). Both A2 (34.56 kcal/mol) and, especially, A4 (110.25 kcal/mol) are highly disfavored, especially for the latter, whose barrier is too high under standard conditions. That TS-A1 is the product with the lowest relative energy suggests that the product is the fastest to form, and hence, by kinetic favorability.

According to the thermodynamic analysis, however, the most stable among the products is A3 (ΔG = –21.50 kcal/mol), followed by A1 (–18.91 kcal/mol), A4 (–13.54 kcal/mol), and A2 (–12.99 kcal/mol). In equilibrium, or at high temperatures, A3 would be anticipated to be the predominant product. However, the experimental results indicate that product A1 is indeed the product that is formed. This observation shows that the reaction operates via kinetic control: the product that appears must quickly be formed, regardless of its ultimate stability. Furthermore, that the kinetically favored product is reasonably stable (–18.91 kcal/mol) supports its dominance even under slightly high energy conditions. This behavior highlights a marked chemoselectivity of the reaction, perfectly predicted by the analysis of the energy diagram. The selectivity in favor of product A1 is therefore not random but results from a compromise between energy accessibility (low activation barrier) and sufficient stability of the formed product [7].

3.4. Bioinformatics prediction of potential biological activities of the hydroxylated derivative of β-himachalene

As a further step in the theoretical analysis of the reactivity of β-himachalene, an in silico investigation of the potential biological properties of the alcohol derivative resulting from an oxidation process was performed. The aim of this strategy is to find a set of the most likely pharmacological activities by means of a bioinformatics screening relying on activity probability scores (Pa). Only tasks where Pa > 0.7 to interpret were elicited to maintain a high degree of reliability [21]. Such an approach can be used to predict the drugs for repurposing and lay the foundation for experimental research. In silico prediction from the alcoholic derivative of β-himachalene of prediction of the biological activities (Pa > 0.7) has been summarized in Table 3.

Indeed, the systematic evaluation of the predictive biological activities in Table 3 for Beta-himachalene-alcohol in a bioinformatics-derived bioactivity profiling approach (Pa > 0.7) reflects a wide and interesting pharmacological profile. Generally, Pa > 0.7 is considered the significant cut-off, indicating that the molecule is likely to have in vivo activity towards the predicted target. For the 15 activities, a relatively high fraction are related to hepatic metabolism enzymes of cytochromes P450 (CYP) family, in particular that of CYP2J (Pa = 0.855), CYP2C12 (Pa = 0.853), CYP2C8 (Pa = 0.753), CYP2B6 (Pa = 0.726), CYP2H (Pa = 0.741), CYP3A4 (Pa = 0.713), and CYP3A (Pa = 0.701). This classification indicates that the oxidized derivative of β-himachalene might act as a candidate substrate for several major CYP isoforms. The compound might be quickly metabolized in the liver, and its pharmacokinetic profile (absorption, distribution, biotransformation, elimination) could rely on the individuals’ enzymatic activity. This property makes the compound a potential risk in drug-drug interaction studies, but also a good candidate as a probe or specific inhibitor in individual enzymatic systems.

In parallel with these metabolic activities, several pharmacodynamic activities for therapeutic purposes are predictively associated with the derivative. Notably, there is an anti-arthritic activity with a high probability (Pa = 0.802), which suggests a potential systemic anti-inflammatory effect. In addition, there are interesting dermatological properties, such as antieczematic activity (Pa = 0.766) and antiseborrheic activity (Pa = 0.701), which could be exploited in the treatment of chronic dermatoses with an inflammatory or hormonal component. An antidyskinetic activity (Pa = 0.702) is also predicted, paving the way for a neuromodulatory activity yet to be explored.

Two individual enzymatic functions also deserve special mention since their roles are directly implicated in clearly characterized pathological processes. On the other hand, the compound is expected to decrease the expression of the metalloproteinase MMP9 (Pa = 0.774), targeted by fluoro-3-(-2-[-8′-hydroxy-5′-quinoline]-ethenyl) benzoic acid, an enzyme with essential functions in the degradation of the extracellular matrix, tumor progression, and chronic inflammation. This feature indicates an anti-invasive or anti-metastatic role of the compound. On the contrary, the alcoholic derivative is predicted to have a high probability of being an inhibitor of testosterone 17β-dehydrogenase [NADP⁺] (Pa = 0.771), a crucial enzyme involved in the metabolism of androgens. This inhibition when compete can have an anti-androgenic action which is interesting in the clinic, in case of hirsutism, hormonal acne or some sensitive cancers (prostate, ovary). Lastly, inhibition of phosphatidylcholine-retinol O-acyltransferase (Pa = 0.722) implicated in vitamin A metabolism hints modulation of cell differentiation pathways, epithelium development or even vision, which needs further elucidation. The prediction of activity as a redox enzyme inhibitor (Pa = 0.706) also suggests a possible redox enzyme as the target, thereby the compound probably would have anti-inflammatory or antioxidant effect. In conclusion, this β-himachalene alcoholic derivative became, according to the in silico modeling, a multitarget profile molecule, with pharmacokinetic (active: substrate to many CYP450) and pharmacological (anti-inflammatory, dermatological, hormonal, antitumor) actions. This profile entirely supports the further molecular docking investigations on the most promising targets (CYP3A4, MMP9, 17β-HSD), to verify the stability of the ligand-receptor interaction and to select the most interesting complexes to be tested in experimental trials.

3.5. Target identification analysis and binding affinity determination with molecular docking

Molecular docking is a structure-based modeling method that can be used to predict the preferred orientation of a ligand that binds to a biological target, generally a protein [22]. This can be used to approximate binding affinity, describe stabilizing non-covalent interactions, and describe steric and electronic complementarity between molecular entities. In this study, we assess the ability of β-himachalene alcoholic derivative, which is the product of oxidation, to stably interact with different enzymatic targets to support the bioinformatics screening results (Pa > 0.7). Based on these predictions and the pharmacological implications of the activities they were associated with, six proteins were logically selected. These targets play roles in important biological functions such as hepatic metabolism (cytochrome P450 family), inflammation, hormone regulation, and cell differentiation [23]. This modeling step represents a crucial structural validation in aid of the proposed mechanistic hypotheses and for guiding future experimental work. Known reference ligands were used for each enzymatic target to guarantee the reliability of the docking protocol. These are midazolam and erythromycin for CYP3A4, paclitaxel for CYP2C8, terfenadine or astemizole for CYP2J2, ilomastat (GM6001) for MMP9, dihydrotestosterone or estradiol for 17β-hydroxysteroid dehydrogenase (NADP⁺), and retinol for LRAT [24-27]. These are widely reported molecules characterized in the literature to be used as a structure reference to ensure the docking approach and to compare the affinity of the tested compound with the native or pharmacologically active ligands. The match between the targets, the reference drugs, and their pharmacological rationales have also been shown in Table 4.

3.5.1. Interaction with CYP3A4: alcoholic derivative compared with midazolam

As for the selected targets, since CYP3A4 plays an important role in hepatic metabolism, the molecular docking was performed to study the interaction between CYP3A4 and the alcoholic derivative of β-himachalene (Pa = 0.713). Midazolam, a CYP3A4 substrate, was selected as a reference compound for validation [28]. Comparison of ligand-protein interactions of the compounds (E) with the alcoholic derivative of β-himachalene and Midazolam in the active site of CYP3A4 has been depicted in Figure 3. The purpose of this study is to assess the stability of anchoring of the derivative and compare its reactivity with that of the natural ligand.

Close

Figure 3.

Matching the ligand-protein interactions of the alcoholic derivative of β-himachalene and Midazolam in the active site of CYP3A4.

Export to PPT

The above Figure 3 shows molecular docking results of the alcoholic derivative of β-himachalene with reference ligand Miazolam in the active site cavity of major metabolic target CYP3A4. A binding energy of –6.5 kcal/mol is shown by the configuration alcoholic derivative–CYP3A4 complex on the left, which suggests a stable and thermodynamically favorable binding. This interaction is stabilized by a number of hydrophobic contacts, including contacts with L366, F367, P474, and L475 in the form of alkyl or π-alkyl interactions. In addition, a common hydrogen bond is formed with a residue GLU412, and a favorable electrostatic interaction is created with ARG403. It is worth mentioning, however, the presence of a detrimental donor-donor interaction with this same residue (ARG403), which would slightly modulate the overall stability of the complex. Nevertheless, the 3D structure reveals a good fit of the ligand in the enzyme cavity. In contrast, the Midazolam-CYP3A4 complex (right), as a reference, shows a stronger binding affinity (–7.4 kcal/mol), in accordance with its classification as a reference substrate of CYP3A4; in compare to the alcoholic derivative, Midazolam does not make any classic HB-bond but tight and well oriented hydrophobic interactions with residues PHE419, LYS424, ILE427, PRO429, and PRO434. These contact interactions provide strong stabilization of the complex due to π–π stacking and π–alkyl, conditioning Midazolam in a flatter and more compact conformation, deeply positioned in the hydrophobic core of the enzyme.

In comparison, while Midazolam has a higher affinity, the alcoholic analog has a competitive binding energy with Midazolam computed value, which is confirmed by the presence of the specific hydrogen bonds that the reference ligand does not have. This implies that the derivative might work as a substrate, and even a partial enzyme inhibitor of CYP3A4. Therefore, the latter findings validate the predictions of the bioinformatics screening (Pa=0.713 for CYP3A4), and structurally support the active metabolization of the derivative by this enzymatic pathway. Such an attitude perfectly justifies the pursuit of a more detailed analysis of the ADME profile as well as of the interaction of the compound with other CYP isoforms.

3.5.2. Molecular docking of CYP2C8; structural validation and comparison with paclitaxel

Of the cytochromes P450 enzymes, CYP2C8 is involved in the metabolism of many xenobiotics, such as anticancer agents, including paclitaxel, to which it is the reference substrate. Since the alcoholic derivative of β-himachalene showed a very high predictive capacity for this target (Pa = 0.753), its docking into the active site of CYP2C8 was also performed. The importance of the anchoring was confirmed by comparing with the reference molecule, paclitaxel. Comparison of the molecular interactions of the alcoholic analog of β-himachalene and paclitaxel in the active site of CYP2C8 (PDB ID: 2NNJ) has been given in Figure 4 [29]. With this comparison, we can report on the stability of the complex formed and on the nature of the interactions established, and predict whether the derivative analyzed will act as a substrate or modulator of the enzyme.

Close

Figure 4.

Comparison of molecular interactions between the alcoholic derivative of β-himachalene and paclitaxel in the active site of CYP2C8 (PDB ID: 2NNJ).

Export to PPT

The alcoholic derivative of β-himachalene (left) shows an affinity energy of -6.4 kcal/mol, which confers a stable and favorable interaction in the catalytic site; its anchoring is guaranteed by several hydrophobic contacts of alkyl and π–alkyl types, notedly, with the LEU128, ARG125, LYS121, TRP120, and LYS432 residues. These interactions reflect a nice fit of the ligand’s lipophilic skeleton into a nonpolar pocket of the protein. However, there is no classical hydrogen bond, which may contribute to the moderate selectivity of the complex and thus, the slightly lower K d value measured in comparison to the reference compound.

Paclitaxel, used here as a reference (right), exhibits a small higher binding energy (–6.6 kcal/mol). Such greater stability should stem from the combination of several specific contacts which the catechol forms, namely a π–cation bond with LYS270, a π–sulfur bond with CYS266, and several C–H•••O hydrogen bonds with one of the cation binding site aromatic residues (PHE126). These specific interactions, together with a longer and polarized conformation, should contribute to more specific anchoring within the enzymatic pocket. Paclitaxel also interacts with ARG125, which corresponds to a residue common to the alcoholic derivative, and suggests the existence of a common recognition site in the enzyme. In conclusion, while the binding affinity of paclitaxel is slightly superior, the oxidized derivative of β-himachalene is a strong and believable interacting candidate with CYP2C8, which supports its potential to be used as a substrate or a competitive inhibitor. These data validate the computational predictions (Pa = 0.753) and indicate a potential metabolic route for the compound through this enzymatic pathway, which should be followed by further pharmacokinetic and enzymatic studies.

3.5.3. Docking to the binding sites of CYP2J2: comparison of interactions of the alcoholic derivative of β-himachalene and astemizole

CYP2J2 is one of the putative enzymatic targets and holds a high probability of interaction score (Pa = 0.855) for the alcoholic derivative of β-himachalene. To confirm this prediction, a comparative molecular docking study was performed between the tested derivative and the known inhibitor of the isoform, Astemizole. The molecular interactions of the alcoholic derivative of β-himachalene and Astemizole in the active site of CYP2J2 were compared, as shown in Figure 5. This assessment will determine the stability of the complex and the metabolic potential toward this target.

Close

Figure 5.

Comparison of molecular interactions between the alcoholic derivative of β-himachalene and astemizole in the active site of CYP2J2.

Export to PPT

CYP2J2 (cytochrome P450 2J2) is a significant isoform in xenobiotic metabolism and endogenous biotransformation, especially regarding the cardiovascular and hepatic systems. As this (Pa = 0.855) is the highest value of predictive capacity achieved in the in silico screen of the alcoholic derivative of β-himachalene, a comparative-docking-based molecular modeling was performed. The intention is to describe the interaction modes of the compound with CYP2J2, and to juxtapose them with those of a standard drug, astemizole, as a reference ligand of this enzyme. The results have been shown in Figure 5 in the 3D representation and 2D plots of interactions for the two ligands.

The alcohol derivative (on the left-hand side of the plot) has an attraction strength of -6.3 kcal/mol, which is stable and likely that have biological relevance. The ligand forms a classical hydrogen bond between its terminal hydroxyl group and the ASP B:458 of the catalytic pocket. This contact is flanked by a π-alkyl interaction with ARG B:457, and an alkyl contact with TYR A:204, providing moderate but well-directed anchoring for the whole complex. The molecular surface presents a partial insertion into a mixed hydrophobic/hydrophilic cavity, with regions of electronic complementarity evident via the donor (purple) and acceptor (green) sites placed on the surface map. By contrast, the reference ligand Astemizole (right) displays a markedly superior binding affinity (–7.7 kcal/mol), as expected for a very affine substrate and well-reported inhibitor of CYP2J2 metabolism. The linear and planar form of astemizole outline of TDNA glycosylase reveals that it can utilize the whole cavity to establish many stacked π–π interactions with the residues PHE A:283, PHE A:325, and PHE A:350. These cation−π interactions are facilitated by hydrophobic contacts with PRO A:287, LEU A:291, AND LEU A:321, and a halogen bond (fluor) with GLN A:234, which stabilizes the steric and electronic anchoring. The distribution of the aromatic rings within the cavity allows for optimal filling at the catalytic site. Consequently, even if less favorable an interaction network for the derivative of β-himachalene than for Astemizole, the derivative of β-himachalene’s capacity to establish a crucial hydrogen bond and to remain energetically sizeable points out that this compound could be considered as a potential substrate or inhibitor of CYP2J2. The consistency of the docking data (–6.3 kcal/mol), the interaction mapping, and the screening prediction (Pa > 0.85), largely support this suggestion.

3.5.4. MMP9 targeted molecular docking: A comparison of β-himachalene alcoholic derivative and Ilomastat interactions

MMP9, an enzyme that plays a central role in inflammatory and tumor processes, is one of the most important targets from a pharmacological point of view. From the in silico screening (Pa = 0.774), we carried out a comparative molecular docking study of the alcoholic derivative of β-Himachalene with ilomastat, the standard inhibitor. Figure 6 shows the comparison of interactions of the alcoholic derivative of β-himachalene with ilomastat in the active site of MMP9. This analysis permits to evaluate stability of formed complexes and the significance of the derivative as a potential inhibitor of MMP9.

Close

Figure 6.

Comparison of molecular interactions between the alcoholic derivative of β-himachalene and ilomastat in the active site of MMP9 (PDB ID: 1GKC).

Export to PPT

The etched enzyme, MMP9, is the most significant mediator of extracellular matrix degradation [30], chronic inflammation, and tumor growth. To evaluate the relevance of the predicted interaction (Pa = 0.774), we performed molecular docking modeling with the alcoholic β-himachalene derivative and the target enzyme MMP9 in comparison with ilomastat, as the standard inhibitor of this enzyme. Both ligands show almost the same affinity (–5.8 kcal/mol) with a similar anchoring potential in the active pocket. The alkyloxy derivative predominantly forms hydrophobic contacts (alkyl and π–alkyl) with VAL A:167 and ILE A:198, suggesting a stable embedding in a non-polar environment. On the other hand, its nonspecific interaction is highly restricted by the lack of hydrogen bonds. On the other hand, ilomastat presents a more detailed network of specific interactions, including hydrogen bonds with GLN A:169 and HIS A:203 and one C–H•••O bond with ALA A:164 and several hydrophobic contacts with VAL A:167 and ARG A:162. Such diversity allows the ligand to gain in steric and electronic stability, explaining its previously reported inhibitory behavior. The overlapping affinities toward these two compounds, with different interaction profiles, indicate the potential use for the alcohol of β-himachalene as a non-specific MMP9 inhibitor. These findings corroborate those of in silico screening and warrant experimental investigations of the molecule’s anti-inflammatory or anti-proteolytic effects.

3.5.5. Molecular docking focused on 17β-hydroxysteroid dehydrogenase (NADP⁺): alcoholic vs. estradiol derivative of β-himachalene interactions

17β-Hydroxysteroid dehydrogenase (NADP⁺) is a rate-limiting enzyme involved in the steroid hormones metabolism which has a direct relation to the regulation of the androgens and estrogen. According to in silico production (Pa = 0.771), a molecular docking analysis was performed to compare the docking between alcoholic ß-himachalene with that of Estradiol, which is the natural substrate of this enzyme. The comparison of molecular interactions between β-himachalene alcoholic derivative and Estradiol in the active site of 17β-hydrogenase (NADP⁺) have been shown in Figure 7. This analogy is an attempt to illustrate what the derivative may be doing as a hormonal modulator.

Close

Figure 7.

Comparison of molecular interactions between the alcoholic derivative of β-himachalene and estradiol in the active site of 17β-hydroxysteroid dehydrogenase (PDB ID: 1JTV).

Export to PPT

HSD17B is the key enzyme for the metabolism of sex steroid/steroid hormones [31]. It catalyzes the transfer of the hydroxyl group of estradiol to NAD to form a beta-keto group. It is a key player in endocrine equilibration and participates in numerous pathophysiologic functions, such as hormone-dependent cancers. We undertook a comparative molecular docking modeling of the alcoholic derivative of β-himachalene and Estradiol, to study their interaction in the active site of the enzyme (PDB ID: 1JTV) [32]. The representations of interactions in 2D and 3D have been shown in Figure 7. The alcohol derivative is found to bind by –7.1 kcal/mol (thermodynamic anchoring). It forms a directional hydrogen bond to residue ASN A:90, parallelly oriented to the active site. Also, various alkyl-type hydrophobic contacts are found to be associated with VAL A:188, LYS A:159, and ILE A:14 to fix the 3D structure of the anchoring. The conformation of the derivative has sufficient steric and electronic complementarity to the active site pocket, but it lacks the contacts observed with the native ligand. For its part, estradiol, with a slightly higher affinity (–7.5 kcal/mol), confirms its role as a high-affinity natural substrate. It interacts through two classic hydrogen bonds with the residues SER A:12 and SER A:142, involved in the stabilization of steroids in the active site. An unfavorable donor-donor interaction with VAL A:143 is also noted, but it does not seem to compromise the overall efficiency of the complex. Its rigid and planar structure allows for optimal insertion, with a central symmetry axis aligned with the enzyme’s catalytic axis. Moreover, its ability to form a network of stabilizing polar interactions explains its superiority in terms of binding energy. Comparison of the two ligands shows that despite different structures and interaction numbers, the alcoholic derivative of β-himachalene has a very competitive profile of binding, which could become the modulating or alternative substrate of 17β-HSD. The above hypothesis is further substantiated by the close similarity of binding affinity to estradiol and specific interaction with a hydrogen bond acceptor, as well as favorable hydrophobic complementarity. The present results add the pharmacological value of the derivative in the hormonal modulators area, suggesting that it is worthwhile experimental research, especially in vitro, on possible modulation of the androgen/estrogen ratio.

3.6. Prediction of ADMET and oral bioavailability of alcohol derivative of β-himachalene

To further improve the pharmacokinetic profile for the studied compounds, an in silico ADME assessment was carried out. This procedure permits prediction of oral bioavailability and the blood-brain barrier crossing, two relevant parameters during the development of drug candidates [33]. To do this, we apply two synergistic ways: (i) using radar charts to show a multi-dimensional bio-physicochemical profile of each molecule, and (ii) applying the BOILED-egg drawing (Brain Or IntestinaL EstimateD permeation) to allow us to visualize the predictive ability of the intestinal absorption and brain permeation from the topological polar surface area (TPSA) and lipophilicity (WLOGP). The relative bio-physicochemical profiles (radar diagrams) and permeability prediction (BOILED-Egg model) regarding the six compounds under investigation, the alcoholic derivative of β-himachalene (M1) and five bio-referents (M2–M6), have been depicted in Figure 8.

Close

Figure 8.

Comparative representation of the bio-physicochemical profiles (radar charts) and permeability prediction (BOILED-Egg model) for the six compounds studied: the alcoholic derivative of β-himachalene (M1) and five reference ligands (M2–M6).

Export to PPT

To finalize the in silico pharmacokinetic evaluation of the analyzed molecules, a dual analysis was performed, combining the previously established multi-parameter biophysicochemical profiles (represented as radar charts) with the predictive BOILED-Egg model (Brain Or IntestinaL EstimateD permeation method) for a set of six compounds: the alcoholic derivative of β-himachalene (M1) and five reference ligands (midazolam, paclitaxel, astemizole, ilomastat, and estradiol). The radar plots showed the simultaneous comparison of the six dimensions (lipophilicity, size, polarity, unsaturation, insolubility, and flexibility) and highlighted that the M1 derivative has a very well-balanced profile of moderate lipophilicity, low polarity, good solubility, small size, and intermediate flexibility, all these according to the passage through intestinal membranes by passive diffusion. By contrast, midazolam (M2) is more lipophilic and less polar and stiffer, conferring a BBB crossing property, while paclitaxel (M3) is the largest, the most polar and the most rigid, which impairs its biodisponibility without vectorization. Astemizole (M4) has high lipophilicity and higher flexibility; good absorption can be expected, but with a distribution to the brain, and ilomastat (M5) shows high polarity and poor solubility and is not flexible compared to the other.

Estradiol (M6): This steroid incorporates low molecular weight, low polarity, and high lipophilicity, general features of steroids, which facilitate a fast penetration through biological membranes. These findings are supported by the BOILED-Egg diagram that combines TPSA with lipophilicity (WLOGP) and predicts intestinal permeability (white zone) and brain penetration (yellow zone). The M1 derivative lies in the white area (strong oral absorption efficacy); however, outside of the yellow area (no CNS permeability), a profile of interest for use at peripheral sites without a narcoreceptor effect. Midazolam (M2) and estradiol (M6) are in the yellow zone, and therefore they might act centrally; astemizole (M4) is exactly on the border between both zones, and consequently a mixed behavior can be inferred. Ilomastat (M5) and paclitaxel (M3) are outliers of the two predictive zones due to a lack of a suitable formulation for their poor oral absorption. In summary, taken across approaches, this study underpins that the alcohol derivative of β-himachalene is both orally absorbed, predicted to possess optimal physicochemical properties, and not to diffuse into the CNS, thus establishing it as a preferred compound for non-neuroactive peripheral indications, such as anti-inflammatory, metabolic, or cardiovascular. Those findings support the utility for targeted in vivo and pharmacokinetic investigations of this compound.