Background

Lung cancer is still the leading and deadly type of all cancers worldwide. Lung cancer was estimated to account for about 25% of the 7 million people that died as a result of cancer-related issues/mortality every year in the world [1,2,3]. The lung is classified traditionally into small cell lung cancer (SCLC) and non-small cell lung cancer (NSCLC) [4]. NSCLC is the lethal/deadly class of lung cancer with nearly 1.5 million reported cases and less than 20% survival rate [5].

Receptor kinases (RKs) belong to the ErbB family, located on cell membranes of living organisms, and played a significant role in the management of the neoplasms’/malignant tumors’ physiological cycle [6]. EGFR is a member of RKs which was recognized to be the most significant target for the management of neoplasms/malignant tumors, which plays a vital role in the control of cancer cell growth, proliferation, and differentiation [7, 8].

Most of the EGFR inhibitors/NSCLC therapeutic agents share a common structure of quinazoline and acrylamide. Many EGFR inhibitors/NSCLC therapeutic agents have been designed and developed starting from the first, the second, and up to the third generations. The first-generation EGFR inhibitors/NSCLC therapeutic agents were developed to treat patients with EGFR mutation caused by the L858R mutation [9]. The first-generation EGFR inhibitors were reported to have a remarkable therapeutic effect in the early stage of clinical treatment; in the first year of treatment with these drugs, more than 50% of the patients developed resistance to these drugs by the T790M mutation [10]. As such, many second-generation EGFR inhibitors/NSCLC therapeutic agents such as afatinib and docatininb were designed and developed to manage the resistance caused by the T790M mutation. The stated objective was not achieved due to serious side effects caused by the so-called second-generation EGFR inhibitors/NSCLC therapeutic agents such as diarrhea and skin rashes. Many third-generation EGFR inhibitors/NSCLC therapeutic agents such as AZD9291, CO1686, and WZ4002 were designed and synthesized to manage the developed resistance caused by the EGFRT790M/L858R mutations [11].

This work targets the computational design of new EGFRWT inhibitors with improved inhibitory activities, as well as docking investigation and pharmacokinetic property prediction of these new EGFRWT inhibitors.

Method

Compound selection

From our previous study, virtual screening via molecular docking was performed which identified molecule 18 (Fig. 1) with the most promising affinity of − 8.6 kcal/mol toward the receptor and a pIC50 of 7.5639 as the template hit compound [12, 13].

Fig. 1
figure 1

The 2D structure of the selected hit compound

Full size image

Design, optimum conformation search, and preparations of designed EGFRWT inhibitors

According to the results of molecular docking virtual screening and pharmacokinetic study investigated in our previous research on some EGFRWT inhibitors, molecule 18 with a binding affinity of − 8.6 kcal/mol and good pharmacokinetic properties was retained as a template to be modified for designing new EGFRWT inhibitors. Ten new EGFRWT inhibitors were designed by the addition of halogens, halo substituted phenyl ring, phenyl methanone, and amino rings on the meta and para positions of the phenyl ring on the sulfinyl group of the template compound.

The Chemdraw software developed by Cambridge University was then utilized to draw the two-dimensional (2D) structures of these newly designed EGFRWT inhibitors (Table 1) [14, 15]. The transformation of two-dimensional (2D) to three-dimensional (3D) structures of these newly designed EGFRWT inhibitors was carried out by direct importation of the structures onto the interface of Spartan. The search for the optimum conformation of all the newly designed EGFRWT inhibitors was performed using Merck molecular force field (MMFF) with density functional theory (DFT) at Becke’s three-parameter hybrid function utilizing LYP correlation functional using 6-311G* basis set. Then, the stable conformations of the anilinopyrimidine analogs were then saved in a file format that is going to be used in the next investigation (Protein Data Bank file format) [12, 16].

Table 1 The two-dimensional (2D) structures of the newly designed NSCLC drugs
Full size table

Retrieval and preparation of EGFR kinase receptor and molecular docking execution

The crystal structure of EGFR kinase domain T790M mutation in complex with AEE788 with pdb entry code 4ZAU was successfully retrieved from the RCSB Protein Data Bank database, prepared and used in this study [17].

Vina of the Pyrex-virtual screening tool was utilize for the docking execution of these newly designed EGFRWT inhibitors with the active site of EGFR kinase receptor [18]. The UCSF chimera software was then used to re-couple the docked newly designed EGFRWT inhibitors and the EGFR kinase receptor and saved in pdb format for the investigation of the respective amino acids the newly designed EGFRWT inhibitors interacted with in the active site of the EGFR kinase receptor. Discovery studio was used for the investigation of the respective amino acids the newly designed EGFRWT inhibitors interacted with in the active site of the EGFR kinase receptor [19].

Pharmacokinetic properties prediction of newly designed EGFRWT inhibitors

SWISSADME (http://www.swissadme.ch/index.php), an online web tool, was utilized in predicting the drug-likeness of these newly designed EGFRWT inhibitors, while the ADMET properties of these newly designed EGFRWT inhibitors under investigation were predicted using the pkCSM, also an online web server (http://structure.bioc.cam.ac.uk/pkcsm) which uses graph-based signatures to generate predictive models of central ADMET properties for drug discovery. Many rules were developed to guide the choice of compounds in the early phases of drug discovery. Among the famous rules applied for the selection of compounds based on the drug-likeness properties in the early phase of drug discovery was Lipinski’s rule of five (RO5). The studied compounds would be evaluated for their drug-likeness properties based on the RO5 criteria [20,21,22].

Results

Activity modeling of newly designed EGFRWT inhibitors

The result of the activity modeling is shown in Table 2.

$$ {\mathbf{pIC}}_{\mathbf{50}}=0.951235508\ast \mathbf{ATSC}\mathbf{1}\mathbf{p}+1.336909853\ast \mathbf{GATS}\mathbf{1}\mathbf{s}-0.811346737\ast \mathbf{GATS}\mathbf{8}\mathbf{s}-3.524533882\ast \mathbf{SpMin}\mathbf{8}\_\mathbf{Bhm}+0.727035851\ast \mathbf{RDF}\mathbf{6}\mathbf{5e}+2.794753386\ast \mathbf{P}\mathbf{1}\mathbf{e}+7.074425161. $$
(1)
$$ {{\mathrm{R}}^2}_{\mathrm{trng}}=0.9459,{{\mathrm{R}}^2}_{\mathrm{adj}}=0.9311,{{\mathrm{Q}}^2}_{\mathrm{cv}}=0.8947,{{\mathrm{R}}^2}_{\mathrm{test}}=0.7008\ \mathrm{and}\ \mathrm{LOF}=0.1195 $$
Table 2 The predicted pIC50 of the newly designed series of newly designed EGFRWT inhibitors
Full size table

Molecular docking investigation of newly designed EGFRWT inhibitors

The results of the molecular docking are presented in Table 3 and Figs. 2, 3, 4, and 5.

Table 3 Types of interactions of the newly designed EGFRWT inhibitors
Full size table
Fig. 2
figure 2

2D and 3-dimensional structure of the SFD10 receptor

Full size image
Fig. 3
figure 3

2D and 3-dimensional structure of the SCD8 receptor

Full size image
Fig. 4
figure 4

2D and 3-dimensional structure of the SCD9 receptor

Full size image
Fig. 5
figure 5

2D and 3-dimensional structure of the SCD4 receptor

Full size image

Pharmacokinetic properties of newly designed EGFRWT inhibitors

The results of the pharmacokinetic properties are presented in Tables 4 and 5.

Table 4 The drug-likeness of the newly designed EGFRWT inhibitors
Full size table
Table 5 The ADMET properties of the newly designed EGFRWT inhibitors
Full size table

Discussion

Activity modeling of newly designed EGFRWT inhibitors

The predictive power of the model developed in our previous study (Eq. 1) was further confirmed in this study by modeling/predicting the pIC50 of these newly designed EGFRWT inhibitors (Table 2). The model was not just chosen but because of its significance with R2trng = 0.9459, R2adj = 0.9311, Q2cv = 0.8947, R2test = 0.7008, and LOF = 0.1195 [23]. From the table, it can be inferred that the predicted pIC50 of these newly designed EGFRWT inhibitors (range from 7.746966 to 11.09261) were better than that of the hit compound with pIC50 of 7.563837 and gefitinib the positive control used in this study with pIC50 of 5.879426. This further reaffirmed the high predicting performance of the model used in modeling the pIC50 of these newly designed EGFRWT inhibitors.

Molecular docking of newly designed EGFRWT inhibitors

The ligand-binding interaction between these newly designed EGFRWT inhibitors and the EGFR tyrosine kinase receptor as shown in Table 3 was investigated and elucidated using the molecular docking protocol. Based on the molecular docking results, the binding affinities of these newly designed EGFRWT inhibitors were found to be between − 8.8 and − 9.5 kcal/mol. The designed compound SFD10 has the highest binding affinity of − 9.5 kcal/mol followed by compound SFD8 (with a binding affinity of − 9.3 kcal/mol), then by compound SFD9 and 4 (each with a binding affinity of − 9.3 kcal/mol).

SFD10 (− 9.5 kcal/mol) formed conventional and carbon-hydrogen bonds with THR790 (2.48 Å) and PRO794 (3.31 Å) amino acids backbone of the receptor. Besides hydrogen bonds, it formed halogen with ALA743, ILE744, and LEU788 amino acids, electrostatics bond with LYS745 amino acid, and hydrophobic bond with LYS745, LEU792, LYS728, PRO794, VAL726, and LEU718 amino acid residues of the receptor.

The second best designed compound SFD8 (− 9.3 kcal/mol) formed a conventional hydrogen bond with PRO794 (2.61178 Å) and LYS745 (2.23017 Å) amino acid residues. It also formed a carbon-hydrogen bond with PHE795 (3.35205 Å) and TYR801 (3.6175 Å). On the other hand, it also formed a halogen bond with LYS745 amino acids. In addition to the halogen bond, it formed a hydrophobic bond with PHE795, GLY796, LEU844 (2), LEU718, VAL726, ALA743 (2), LYS745, and HIS805 amino acid residues of the target protein.

Only one conventional carbon-hydrogen bond interaction was observed between the designed compound SFD9 (− 9.2 kcal/mol) and the binding site of the target protein with PRO794 (2.43) amino acid residue. Also, two carbon-hydrogen bond interactions were observed between the designed compound SFD9 and the binding site of the target protein with GLN791 (3.52) and PHE795 (3.42) amino acid residues, respectively. The following amino acid residues PHE795, GLY796, ALA743, MET793, LEU788, LEU718, LEU844, VAL726, and LYS745 were also observed between the designed compound SFD9 and the binding site of the target protein employing hydrophobic interaction.

SFD4 among the designed compounds with higher affinity (− 9.2 kcal/mol) toward the target interacted in the binding site of the target receptor with PRO794 (2.46) and THR790 (2.32) residues employing conventional carbon-hydrogen bond. Two carbon-hydrogen bond interactions were observed between the designed compound SFD4 and the binding site of the target protein with GLN791 (3.33) and MET793 (3.79) amino acid residues, respectively. The binding site of the receptor was seen to interact with SFD4 employing hydrophobic interaction with PHE795, GLY796, MET793, LEU844, MET766, LEU718, LEU844, and VAL726. Not only that, ALA743 and LEU788 amino acid residues were observed to interact with SFD4 employing a halogen bond. LYS745 amino acid in the binding site of the receptor interacted with SFD4 employing electrostatic interaction.

The following amino acid residues, PRO794, LEU718, ALA743, VAL726, and LYS745 are common to the best designed compounds which might be the reason why they have a higher binding affinity. On comparing the designed compounds with the template and the control gefitinib, the designed compounds possessed better binding energy than the template and the control gefitinib. Furthermore, the 3D and 2D structures of the best four discussed designed compounds are shown in Figs. 2, 3, 4 and 5.

Pharmacokinetic properties of newly designed EGFRWT inhibitors

The drug-likeness properties of the designed compounds were also predicted following Lipinski’s rule of five (Table 4). All the newly designed compounds were found to have one violation for Lipinski’s rule of five (WM > 500). The number of hydrogen bond donors and acceptors for all was less than 5 and 10, respectively. The TPSA and the WLOGP values were less than 140 Å and 5, respectively. The synthetic accessibility scores of these newly designed compounds on the scale were in the easy portion (< 5). It means that there is high tendency these newly designed compounds can be easily synthesize in the laboratory. On that basis, the newly designed compounds predicted to be drug-like compounds, orally bioavailable, and active [21, 22].

The predicted ADMET properties of these newly designed compounds are represented in Table 5. The intestinal absorption values for these newly designed compounds were all above 90% but less than 100. Their intestinal absorption values have passed the threshold value of 30%, which clearly shows that these newly designed compounds have high human intestinal absorption properties. The BBB permeability (log BB) values of all newly designed series F compounds were all < − 1, which implies that all these newly designed compounds are poorly distributed through the brain. The CNS permeability (Log PS) values for all were > − 2 which are considered to penetrate the central nervous system. Moreover, they were found to be both substrate and inhibitors of CYP3A4, thereby affirming their metabolic properties. Furthermore, the total clearance for a drug molecule in the body for these newly designed compounds was within the accepted value. All the newly designed compounds were found to be non-toxic. Based on these predicted parameters, the newly designed compounds are said to have high absorption value, low toxicity level, and good permeability across the cell membrane. In general, all these newly designed NSCLC drugs were predicted to have good pharmacokinetic profiles [21, 22].

Conclusion

In the end, the modeled activities of these newly designed EGFRWT inhibitors were seen to be higher than that of the template compound and the control used in this research as well as their binding affinities toward their target (EGFR kinase enzyme). They were seen to possess drug-like properties by non-violating more than 1 of Lipinski’s rule of five, the filtering criterion used in this work, meaning they were predicted to be orally bioavailable. More so, they were also seen to have good ADMET properties, and none of them was found to be toxic. Furthermore, based on their synthetic accessibility, they can be easily synthesized in the laboratory. Therefore, this study recommends that these newly designed EGFRWT inhibitors should be synthesized.