The docking results show that the HIF-1α receptor has the best binding energy when it interacts with compound 1 (− 8.49 kcal/mol) followed by compound 3 (− 8.43 kcal/mol) and compound 2 (− 7.80 kcal/mol). The desolvation energies are in agreement with the binding energies. Compounds 1 and 3 are so close in terms of stability.
Further, Fig. 4 reveals that both compounds 1 and 2 docked to residues located in the HIF-1α active site with four hydrogen bonds. Compound 2 interacts via its OH and O=C functional groups with active site amino acids HIS279, ARG238, THR196, and ASP201 of the receptor HIF-1α, while compound 1 interacts with active site amino acids ARG238 with its OH group and SER184 with its O=C group. Compound 4 docked to the residues located in the HIF-1α active site with two hydrogen bonds (THR196, SER184).
Surprisingly, no H-bonding interaction occurs in the complex that involves ligand 3 despite its large binding energy (− 8.43 kcal/mol), very close to that of the most stable complex. Such a result was also observed by other previous studies (Tunga et al. 2020; Kasende et al. 2017). However, it is noteworthy to mention that the complex between ligand 3 and the receptor HIF-1α (3KCX) is mainly stabilized by π-π interactions (stacked and T-shaped) and van der Walls interactions that contain a large quantity of dispersion energy. The stabilization of other complexes is strengthened by the π–π, π-cation, π-alkyl, π-sigma, and vdW interactions as well (Mpiana et al. 2020; Matondo et al. 2021).
Overall, modeling of contact shows lipophilic and hydrogen-bonding interactions involving typical hydrophobic and polar amino acid residues which are crucial for the inhibitory interaction of HIF-1α (Vidya et al. 2019). In particular, ASP201 and ARG238 residues play a significant function in the inhibitory interaction between compounds 1/2 and HIF-1α. These key residues are located in the binding site of the HIF-1α protein.
In terms of ligand efficiency, a compound to be selected as a hit should possess a threshold value of 0.3 (Hopkins et al. 2004). All investigated compounds have ligand efficiency greater than the threshold value, except compound 1 (− 0.30 kcal/mol). Nevertheless, there do appear to be situations in which a compound with an inhibitory effect has a LE value lower than the threshold value (Hopkins et al. 2004; Alamri 2020).
Thermodynamically, compounds 1, 2, and 3 play a significant role in the inhibition of HIF-1α regarding the interaction energy values of the formed complexes. Recent findings revealed that hypoxia, a condition closely associated with the tumor environment due to the fast growth of cancerous cells, is caused by the lack of adequate blood supply. This condition causes the activation of hypoxia-inducible factor (HIF), a protein transcription factor involved in carcinogenesis, and tumor growth through the up-regulation of genes involved in angiogenesis.
Indeed, in hypoxia conditions, HIF-1 interacts with the positive regulatory sequence or enhancer called hypoxia response element (HER) of the vascular endothelial growth factor (VEGF) gene in order to facilitate the creation of new blood vessels at the tumor level and to answer the energy demand efficiently and therefore of oxygen bound to the anarchic cells proliferation (Gothié and Pouysségur 2002; Pezzuto and Carico 2018).
In relation to the second receptor, 1E3G, the results in Table 1 show that the score values vary between − 10.04 and + 14.38 kcal/mole. The 2D representation of interactions of the 4 ligands with the 1E3G receptor complexes is depicted in Fig. 5. Among the top binding ligands to 1E3G, two potential inhibitors are identified, ligands 1 and 3 with binding energies of − 10.08 and − 9.61 kcal/mole, respectively. These two compounds are also those identified as potential inhibitors for the receptor 3KCX, and have ligand efficiency values lower than the threshold value. For both 3KCX and 1E3G receptors, the weakest complex is obtained with ligand 4.
Prostate cancer is the second most common malignancy among men all over the world. Its therapy is based on the use of drugs acting as antagonists of hormone receptors against the prostate tissue (Anshika et al. 2017).
Overall, cancer treatment can be performed in various ways, such as surgery, chemotherapy, radiotherapy, immunotherapies, gene therapy, or protein therapy (Siegel et al. 2019). However, on the one hand, the cost of treatment is too high that in low- and middle-income countries, people are unable to support their medical care, and on other hand, most anticancer drugs have several side effects in addition to the drug resistance problem. Currently, the use of plant-based bio-active extracts and phytochemicals is gaining recognition as these have structural diversity, negligible side effects, and bio-availability, as well as exhibit multiple target activities (Ashfaq et al. 2013).
The search for phytochemicals having significant androgen receptor inhibition activity could help the development of new drugs from African plant biodiversity. The results showed that compounds 1, 2 and 3 have significant activity in inhibiting the receptors. The present study therefore reveals that these two phytochemicals could be promising candidates for further evaluation in the prostate cancer prevention or management. In fact, the interaction of these secondary metabolites with androgen receptor could inhibit signaling pathways implicated in the development and progression of prostate cancer especially MAPK and PI3K pathways (Lonergan and Tindall 2011). Further post-docking calculations, namely molecular dynamics simulations (Hollingsworth and Dror 2018) and or MM-PBSA/MM-GBSA (Genheden and Ryde 2015) of four ligand–receptor complexes obtained, should be carried out in near future for confirmation of anti-cancer activities of compounds. However, the three compounds studied in this work belong to the quinone-methide family and several studies have highlighted the anticancer activities of these compounds or even of compounds which include a quinone-methide moity (Dufrasne et al. 2011; Hernandes et al. 2020).
According to threshold energy, a best anti-cancer activity is obtained if the binding energy is above − 6.60 kcal/mole for 3KCX or − 11.99 kcal/mole for 1E3G. Taking into consideration this threshold, the two plant species are more efficient against brain cancer than prostate cancer. The target amino-acids implicated in hydrogen bonds are given in Table 2.
Based only on the H-bonding interaction, this table shows that ligands 1 and 2 have a good pharmacophore profile to target both the 3KCX and 1E3G proteins. More importantly, it can be seen that all amino acids residues involving in the hydrogen bonding interaction in ligands 1 and 2 in the receptors sites are present in the co-crystalized binding site (PDB ligand).
Inspection of Table 3 shows that all ligands meet every single criterion of Lipinski’s rule of five and thus fully obey the rule. Indeed, Lipinski's rules predict that the absorption of a compound is low when (1) the molecular weight is greater than 500; (2) the number of donor hydrogen bonds is greater than 5; (3) the number of acceptor hydrogen bonds is greater than 10; (4) the partition coefficient is greater than 5 (Lipinski 2000). The bio-availability score reveals that all four compounds virtually tested are orally bioavailable in rats (Martin 2005) and can be absorbed in the intestine with respect to the TPS value ≤ 140 Å2 (Clark 1999; Verber et al. 2002). However, only compound 3 is capable according to Hitchcock (2008) criteria of crossing the blood–brain barrier. As a result, they possess drug-like properties, of which only compound 3 fulfills the conditions for serving as a lead-like compound (Scheneider 2002). The partition coefficient, a physico-chemical parameter used to measure the tendency of a molecule to dissolve in membranes, which is correlated with its tendency to dissolve in an organic solvent, is inversely proportional to solubility. Thus, a Log P value that is too high will suggest that the molecule may be poorly soluble in an aqueous environment. The logarithm of partitioning coefficient between n-octanol and water phases ranges for 95% of existing drugs: − 2 to 6 (Ntie-Kang 2013). All the four compounds have a Log P comprising in this interval. These two physicochemical parameters (P and S) are related to the absorption of drugs in the organism. In the present study, the values of Log S (solubility) are between − 6.54 and − 3.78. Compound 2 has the best solubility (Log S ˃ − 4) while the other compounds are poorly soluble (Bergstrom et al. 2003).
It can also be noted that Log Kp values are within normal limits (between − 8 and − 1) indicating that these compounds are likely to be distributed in the organism for bio-transformation (Yadav et al. 2018).
Table 4 shows that compounds 1 and 2 are highly soluble and have the best absorption rate in the intestine. Only compounds 1 and 2 can serve as substrate for glycoprotein P, but all are inhibitors of this transmembrane protein. It is also noted that apart from compound 3, all other compounds (1, 2 and 4) have a VDss < 0.45 (low). Only compound 3 (LogBB ˃ 0.3) can cross the blood–brain barrier, while the others can be moderately distributed in the brain (Log BB < − 1). Compounds 3 and 4 can enter the CNS (Log PS ˃ − 2). None of these four compounds is metabolized by cytochrome P450 2D6 or its inhibitor. Compound 1 is an inhibitor of CYP3A4 while compounds 3 and 4 are its substrates. In addition, ligand 3 is an inhibitor of CYP1A2, CYP2C19 and, CYP2C9. The inhibition of CYP2C9 by compound 3 means that there is a possibility of interaction between the plant and NSAIDs/antiarrhythmics (amiodarone) and anticoagulants (acenocoumarol) because the latter are bio-transformed by CYP2C9. Cytochrome P450 2C19 is involved in the bioactivation of pro-drugs into bio-active metabolites. Its inhibition by compound 3 would lead to a decrease in the activity of the pro-drugs. This is notably the case with clopidogrel, a platelet anti-aggregate used in acute coronary syndrome (Desmeules 2010). Of the above, only compound 4 has a good metabolic profile (P-gp inhibitor and CYP3A4 substrate), while the risk of interaction between compounds 1, 2 and 3 is possible in individuals taking D. quercina herbal tea alone or when combined with drugs that are substrates for these enzymes. Indeed, the inhibition of P-gp and CYP3A4 by compound 1, for example, will lead to the accumulation of compound 3. A drug interaction is also possible between the four compounds when the phyto-drug is formulated from two plants (D. quercina and S. leptoclada).
Biochemically, various transmembrane proteins, such as P-glycoprotein (Pgp), CD243, or Multi-drug Resistance protein (MDR1), part of the ATB-Binding cassette (ABC) super family of transporter proteins, play an important role in the bioavailability of drugs. Indeed, several drug chemistry compounds can inhibit or induce ATPase P-gp activity, thus altering the kinetics of drug substrates for this transporter (Kale et al. 2012). Once distributed, these chemical compounds undergo biotransformation in the liver using P450 cytochromes (phase I) and other enzymes such as glutathione s-transferases (GST), catechol o-methyl transferase (COMT), or glucuro-nyltransferases (phase II: conjugations) that facilitate their elimination from the body (ADME process). In humans, the expression of detoxification genes (MDR1, Cytochrome P450) and therefore the response to drug treatment vary from one individual to another (Fattinger and Meier-Abt 2003). This inter-individual variation constitutes what is called "genetic polymorphism." Genetic polymorphism is a non-pathological variation in a nucleotide sequence between individuals due to a genetic mutation (Ameziane et al. 2006). This is the case in particular of the C3435 T mutation located in exon 26 of the MDR1 gene, whose CC genotype (high prevalence rate in negroids) is characterized by overexpression of P-gp, significant excretion (efflux) of substrates, and a low plasma level of substrates compared to the CT and TT genotypes (Caucasian populations). It is well established that P-gp substrates are also those of CYP3A4.
Indeed, an induction of CYP3A4 leads to an increase in P-gp expression as the genes encoding these two are located on chromosome 7 (P-gp: 7q21.1 and CYP3A4: 7q22.1) and would be regulated by the same mechanism (Feaz 2016).
In addition, it should be noted that P-gp limits the absorption of xenobiotic compounds from the gastrointestinal tract by promoting elimination in urine and bile but also participates in a protective barrier role for the CNS and the fetus (blood–brain and placental barrier).
After absorption and diffusion, five iso-enzymes (alternative forms) of cytochrome P450 (CYP1A2, CYP3A4, CYP2C9, CYP2C19, and CYP2D6) are involved in the metabolism (Phase I) of 50–90% of the drugs (Desmeules 2010). Clinically, their inhibition can decrease the elimination of the drug and increase its bioavailability, resulting in an overdose that can be fatal for the patient. A drug interaction occurs when a given chemical compound inhibits the enzymatic activity of cytochrome P450, leading to the accumulation of its drug-substrate in the body. Table 4 also shows that compound 4 has a low clearance value, indicating that its concentration in the body is low. This means that its rate of elimination from the body is very high. This elimination is essentially metabolic. Only compound 3 can be eliminated via the organic cation transporter OCT2. Regarding the toxicological profile, only compound 1 is potentially mutagenic and hepatotoxic. However, none of the four compounds is cardio-toxic.