1 Introduction

Cancer is defined as a condition in which the body is unable to regulate the spread of aberrant cell division and growth to neighboring tissues and other organs [1]. All over the world, it continues to be one of the major causes of death and morbidity despite decades of scientific and clinical research and trials of promising new medicines [2]. Therefore, cancer therapy is a significant challenge for medicinal chemists who want to create secure and efficient tailored chemotherapeutic drugs [3]. Apoptosis is a critical cellular process that eliminates damaged or potentially harmful cells in response to various oncogenic stresses, such as uncontrolled proliferation or DNA damage [4]. Through this process, apoptosis acts as a crucial safeguard against the development of cancer by removing cells that are at risk of transformation. Additionally, apoptosis is known to suppress tumorigenesis through various mechanisms, including the clearance of oncogenic proteins and the modulation of cellular signaling pathways [5]. The epidermal growth factor receptor (EGFR) is a particularly important factor that was linked to cancer cells’ apoptosis [6, 7]. It has been taken as a prime target for abnormal signaling-mediated cancer [8]. Additionally, EGFR has critical contributions in the development and progression of various carcinoma types [9,10,11]. Overexpression of EGFR promotes cell proliferation, differentiation, and survival [12]. A high level of EGFR expression has been associated with a lower survival rate across a variety of cancer types, and this expression serves as a powerful prognostic indicator [13].

Ligand and structure-based drug design are forefront of major anticancer drug design targeting EGFR [14, 15]. In addition to the discovery and repurposing of more potent drugs, the study of structure–activity relationships can also enhance the drug-likeness, pharmacokinetics, and pharmacodynamics of new drugs [16]. By using software employing a variety of techniques, computational chemistry can be used to study the interactions between potential drugs and biomolecules. Computational chemistry has been widely used in the pharmaceutical industry [17,18,19]. A range of computational chemistry applications has been developed over the years, including molecular and drug design [20, 21], docking simulations [22, 23], ligand-based approaches such as ADMET [24, 25], DFT [26], structure similarity [27, 28], and pharmacophore assessment [29].

As part of our team’s continuing efforts to search for compounds with potential anticancer activity targeting EGFR protein, including various derivatives of xanthines [30,31,32], thieno[2,3-d]pyrimidine [33,34,35], and theobromine [36,37,38], this paper describes a novel lead anticancer compound (T-1-NCA) that was synthesized for the first time and demonstrated promising in silico and in vitro anticancer properties.

1.1 Rational

To counteract aberrant signal transduction through EGFR, the design of EGFR inhibitors using natural molecules as scaffolds has been potentiated [8]. Studies have demonstrated that some of the bioactive compounds in medicinal herbs overcome drug resistance to EGFR-TKIs and potentiate the therapeutic effects of EGFR-TKIs. These compounds include polyphenols, saponins, terpenoids, alkaloids, quinones, resins, and nucleosides [39]. Resveratrol I belongs to a class of polyphenolic compounds called stilbenes [40]. Resveratrol decreases CYP1A1 and ABCG2 expression and impairs gefitinib elimination from cells, increasing intracellular gefitinib concentration, which helps to overcome gefitinib resistance [41]. The accumulated gefitinib triggers apoptosis, autophagy, and senescence in gefitinib-resistant NSCLC cells [41]. Curcumin II potentiates the therapeutic effects of gefitinib. Curcumin induces apoptosis in TKI-resistant NSCLC cells by decreasing EGFR phosphorylation and increasing EGFR degradation, thus inhibiting cancer growth [42]. More importantly, the combination treatment of curcumin and EGFR-TKI markedly inhibits NSCLC growth by decreasing the expression of EGFR, c-MET, and cyclin D1 [43]. Shikonin III is a naphthoquinone compound that exhibits cytotoxicity, increases ROS production, and induces apoptosis in NSCLC cells. In addition, shikonin suppresses EGFR phosphorylation and increases EGFR proteasomal degradation in NSCLC [44]. Gambogenic acid IV is a polyprenylated xanthone that induces apoptosis in NSCLC cells by inhibiting the JAK/STAT3 signaling pathways. Furthermore, gambogenic acid abrogated the resistance to erlotinib in NSCLC, as demonstrated in a xenograft mouse model and patient-derived xenograft model by inhibiting c-Met activity and decreasing EGFR phosphorylation [45]. Cordycepin V induces cell cycle arrest at the G0/G1 phase and apoptosis in lung cancer cells by inhibiting the phosphorylation of EGFR, Akt, and ERK1/2 [46] and by interacting with and activating AMP-activated protein kinase [47, 48]. Cordycepin is also capable of inhibiting NSCLC cell cycle progression. NSCLC cells containing EGFR mutations are more sensitive to cordycepin treatment than those without EGFR mutations [49].

In this work, a new EGFR-based semisynthetic compound was designed and synthesized. Such compound has the same features of EGFR inhibitors [50, 51] (Fig. 1). The xanthine moiety was used as a heteroaromatic system to occupy the adenine binding pocket of the EGFR binding site. The acetamide moiety was used as a linker. The cyclohexyl moiety was used as a terminal hydrophobic head to occupy the hydrophobic region I. The two methyl groups at the 3- and 7-positions of xanthine moiety were used and proposed to occupy the hydrophobic region II.

Fig. 1
figure 1

Rational of design

Full size image

2 Results and discussion

2.1 DFT

2.1.1 Geometry optimization and Mulliken charge

The geometry optimization of T-1-NCA was adopted at B3LYP/6-31G +  + (d, p) utilizing the Gaussian 09 program. The important bond length (C14-N2) and bond angles at the two ends of the formed bond are given in the optimized structure, Fig. 2a. The ground energy or total energy (TE) was found to be -29,498.4 eV for the current chemical system of 44 atoms and 170 electrons, indicating a stable structure. The magnitude of the dipole moment was calculated to be 7.795 Debye, suggesting a polarizable structure with high reactivity.

Fig. 2
figure 2

The fully optimized chemical structure (a) and the Mulliken charge of T-1-NCA at B3LYB/6-311G +  + (d,p) level

Full size image

The color scale of the Mulliken charge is represented in Fig. 2b, which shows the polarizability and distribution of charges along the chemical structure and hence predicts the charge transfer possibility in the designed molecule. The most electronegative and electropositive atoms are highlighted in Fig. 2b.

2.1.2 Frontier molecular orbital (FMO) analysis

HOMO and LUMO (FMO) orbitals can explain quantitative information regarding the affinity of T-1-NCA to bind with the target. According to Fig. 3, both HOMO and LUMO density lobes are mainly distributed over the structure except the cyclohexane ring. The HOMO energy (EHOMO), LUMO energy (ELUMO), and the energy gap between HOMO and LUMO, Egap are calculated and represented in Fig. 3. The Egap is a major stability metric that describes a molecule's reactivity and stability. Structures with a narrow Egap are more reactive and polar because electrons are readily offered to an acceptor. Theoretical Egap indicates high inhibition reactivity and feasible charge transfer [52]. The ionization potential (IP) and electron affinity (EA) amounts, Table 1, with the small Egap, suggesting strong binding with the target due to feasible electron donation from the prepared compound to the target molecule.

Fig. 3
figure 3

FMO levels of T-1-NCA by using DFT calculations at B3LYB/6-311G +  + (d,p) level

Full size image
Table 1 The calculated global reactivity indices and energetic parameters for the selected compound
Full size table
2.1.2.1 Chemical reactivity descriptors and total density of state (TDOS)

Koopman's theorem was utilized to compute the electronic properties through various global reactivity parameters, including “ionization potential (IP), electron affinity (EA), chemical potential (μ), electrophilicity (ω), chemical hardness (η), maximal charge acceptance (Nmax), chemical electronegativity (χ), and global softness (δ)” as follows:

IP = –EHOMO.

EA = –ELUMO.

µ = (IP + EA)/2.

η = (IP−EA).

χ = −µ.

ω = µ2/(2 η).

σ = 1/η.

∆Nmax = − (μ/η).

∆E = −ω.

Egap = ELUMO–EHOMO.

The computed parameters listed in Table 1 show high (σ); softness, low (ω), and (η); electrophilicity and hardness, which indicate a high chemical reactivity. The magnitude of (μ), (ω), and (η) determines the efficiency of the compound to gain an extra charge ∆Nmax from an adjacent chemical system (target) [52]. In addition, the μ and χ control HOMO, LUMO, and MEP which are primarily assigned to a compound's reactivity and biological activity, as electronegativity (χ) is the negative of chemical potential (− μ) [53]. Hence, the easiness of gaining extra charge and the small value of Egap were the reason behind the high inhibition reactivity of the prepared drug.

In equilibrium, "the number of occupied states per unit volume" of a compound can be calculated by multiplying the probability function by the density of states and the results can be used to study various characteristics of the molecule [54]. The overall total density of states has been calculated and is depicted in Fig. 4. The TDOS spectrum indicated the highest density was found in the unoccupied orbitals.

Fig. 4
figure 4

FMO spectrum and TDOS of T-1-NCA obtained at B3LYP/6–311 +  + G(d,p)

Full size image

2.1.3 Electrostatic potential maps (ESP)

Electrostatic attractions are one of the main variables that influence molecule–protein binding. The computational analysis of these attraction forces helps to determine the energy of the protein–drug complex. Electrostatic forces with steric force are the main driving forces in biomolecules.

Depending on the availability of electron density, the ESP surface has many colored spots. Atoms with a partial negative charge (highly electronegative) glow red and can act as hydrogen-bonding acceptors, while atoms with a low electronegativity (poor in electrons) and having a partial positive charge display blue and act as hydrogen-bonding donors. Neutral atoms (with zero charge) display green to yellow color and create p- and other kinds of staking attractions. These colored domains aid in anticipating their ability to participate in chemical bonding and understanding their reaction mechanism. T-1-NCA displayed blue and red patches, suggesting the probability of hydrogen bond formation with the target. The red patches at the carbonyl groups can create hydrogen bonds with polar amino acids, while the blue zones of hydrogen atoms in the methyl group of the imidazole ring could act as hydrogen donors. The high electron clouds with a green color primarily at cyclohexane moiety favor the π-staking bonding with aromatic amino acid residues (Fig. 5).

Fig. 5
figure 5

ESP; maps of T-1-NCA at the 6-311G +  + (d,p) level

Full size image

2.2 Molecular docking against wild and mutant EGFR

The molecular modeling tool was able to foresee interactions between molecules and their biological targets. Therefore, the binding mode of T-1-NCA was examined against EGFRWT (wild form) and EGFRT790M (mutant form) using MOE. As references, the co-crystallized ligands erlotinib and TAK-285 were utilized for wild and mutant types, respectively.

Before we go to how T-1-NCA binds wild and mutant types, a validation step was performed to verify the docking process Figs. 6 and 7.

Fig. 6
figure 6

Validation step of erlotinib

Full size image
Fig. 7
figure 7

Validation step of TAK-285

Full size image

With an affinity value of −21.44 kcal/mol, erlotinib binds the active site of EGFRWT. In details, the quinazoline moiety occupied the adenine pocket of EGFRWT forming one hydrogen bond with Met769. In addition, it was incorporated with many hydrophobic interactions with Leu820, Ala719, Leu694, Val702, and Thr830. The terminal ethynylphenyl moiety was oriented into the hydrophobic pocket I forming several hydrophobic interactions with Val702, Thr830, Lys721, Asp831, and Thr766. Additionally, the hydrophobic region II was occupied by the two 2-methoxyethoxy groups forming one hydrogen bonds with Lys692. Furthermore, it formed hydrophobic interactions with Gly695, Leu694, Gly772, Phe771, and Pro770. These findings were matched with the reported results [55] (Fig. 8).

Fig. 8
figure 8

Binding mode of erlotinib against the active site of EGFRWT. A 3D binding mode of erlotinib, B 2D binding mode of erlotinib

Full size image

A comparable affinity value was obtained by T-1-NCA (−18.71 kcal/mol). It interacts with the EGFRWT active site similarly to erlotinib. The purine moiety formed one hydrogen bond with Met769 in the adenine pocket. In addition, it formed many hydrophobic interactions with Lue694, Leu820, Val702, Thr830, and Ala719. On the other side, the cyclohexyl group was oriented into the hydrophobic pocket I forming several hydrophobic interactions with Lys721, Leu764, Leu820, Thr830, Lys721, Asp831, Ala719, and Thr766. Furthermore, the two methyl group formed a close contact with Gly695, Leu694, Val702, Gly772, and Phe771 (Fig. 9).

Fig. 9
figure 9

A Mapping surface MS, B 3D, and C 2D binding pattern of T-1-NCA with the active site of EGFRWT

Full size image

To support the aforementioned findings of the wild type, the docking outcomes of the mutant type of EGFR (EGFRT790M) were examined. The binding pattern TAK-285 to the EGFRT790M active site revealed an important hydrogen bond with Met793 [56] as presented in Fig. 10.

Fig. 10
figure 10

TAK-285 docked into EGFRT790M active site

Full size image

As displayed in Fig. 11, T-1-NCA was stacked onto the EGFRT790M catalytic site in a way like to that of TAK-285. The purine arm was fixed in the adenine pocket to form one hydrogen bond with Met793 and four pi-pi bonds with Leu844, Leu718, and Cys797. Additionally, the cyclohexyl moiety interacted with the hydrophobic pocket forming three hydrophobic bonds with Lys745, Met790, and Val726.

Fig. 11
figure 11

A 3D and B 2D interaction pattern of T-1-NCA with EGFRT790M

Full size image

2.3 Molecular dynamics (MD) simulations

According to the analyses performed on the production run, the protein shows a stable behavior, while the T-1-NCA shows large fluctuations which result from the unbinding and rebinding to the active site. The RMSD plot of the protein (Fig. 12a (blue curve)) shows a stable average during the whole trajectory with a value of 1.79 Å. On the other hand, the RMSD values of the complex (green curve) show two states. For the first 20 ns, the RMSD of the complex has a stable average of, approximately, 2 Å. After that, it shows a slightly large average of around 2.65 Å except for the spike at 29 ns, reaching 4.3 Å, which corresponds to the transient unbinding of the T-1-NCA. On the other hand, the RMSD of the T-1-NCA shows very large values with the maximum value reaching 27 Å corresponding to the unbinding of the compound. After rebinding, the RMSD shows a slightly large fluctuation before stabilizing at an average of 13.3 Å for the last 20 ns. The RoG (Fig. 12c) and SASA (Fig. 12d) show a similar trend. The first 80 ns show an average of 19.42 Å and 15,088 Å2, respectively. On the other hand, the last 20 ns show a decreasing trend due to the motion of the loop Glu841:Lys851 as this loop has a large RMSF value as can be seen in Fig. 12f. H-bonds (Fig. 12e) show a stable fluctuation with an average of 61 bonds. The fluctuation of the amino acids depicted in the RMSF plot (Fig. 12f) shows very low fluctuation (less than 2 Å) except for the Glu841:Lys851 loop, the Asp892:Ile899 loop, and the free C-terminal reaching 5, 2.7, and 8 Å, respectively. During the simulation time, the ligand transiently dissociated from the protein before rebinding again for the rest of the MD duration (Fig. 12g), with an average of 14.4 Å. This confirms that the large RMSD values are due to the free movement of the ligand inside the binding pocket.

Fig. 12
figure 12

The RMSD values from the trajectory. A the RMSD for the EGFR protein and EGFR & T-1-NCA complex. EGFR is the blue curve, and complex is the green curve, B the RMSD for the T-1-NCA compound, C the radius of gyration of the EGFR protein, D the solvent-accessible surface area (SASA) of the EGFR protein, E the change in the number of H-bonds during the trajectory, F the RMSF values of the amino acids, G the center of mass distance between the T-1-NCA compound and the EGFR protein, H the different energetic components of MM-GBSA and their values. Bars represent the standard deviations, I the binding free energy decomposition of the EGFR-T-1-NCA complex

Full size image

The binding free energy analysis using MM-GBSA (Fig. 12i) shows the different components that contribute to the binding. T-1-NCA shows a total binding with an average value of −11.63 kcal/Mol. The largest favorable contribution is the van der Waals energy with an average value of −22.73 kcal/Mol followed by the electrostatic interaction energy with an average value of −9.65 kcal/Mol. Moreover, we did a decomposition analysis (Fig. 12i) to know which amino acids within 1 nm of T-1-NCA have a contribution to the interaction with a value less than −0.5 kcal/mol. Leu694 (−1.01 kcal/Mol), Gly695 (−0.86 kcal/Mol), Ser696 (−1.29 kcal/Mol), Gly697 (−0.73 kcal/Mol), Val702 (−0.81 kcal/Mol), and Arg817 (−0.64 kcal/Mol) are the amino acids that have a contribution with a value better (less) than −0.5 kcal/Mol. In addition, one amino acid Asp776 shows a positive contribution to the binding with an average value of + 0.48 kcal/Mol.

The trajectory was clustered to get a representative frame for each cluster produced. The number of clusters was selected automatically using the elbow method, and this produced five clusters. For each cluster representative, the PLIP webserver was utilized to know the number and types of interactions between T-1-NCA and the EGFR protein. Table 2 shows the number and types of interactions obtained from the PLIP webserver. There are two types of interactions detected with nearly the same numbers (5 hydrophobic interactions and 4 H-bonds). Val702 and Ser696 are the most common amino acids forming hydrophobic interactions and H-bonds, respectively. In addition to producing the interaction types and numbers from PLIP, it also generates a.pse file to see the 3D conformation of the ligand and its interaction with the protein (Fig. 13).

Table 2 The number and types of interactions detected from the PLIP webserver
Full size table
Fig. 13
figure 13

Different interactions produced from PLIP for each cluster representative. H-bond: blue solid line, hydrophobic interaction: dashed gray line, amino acids: blue sticks representation, and T-1-NCA: orange sticks representation

Full size image

2.4 ADMET profiling study

In medicinal chemistry, the field of predictive ADMET (absorption, distribution, metabolism, excretion, and toxicity) has grown in importance since Lipinski's rule of five was introduced [57]. Nowadays, it consists of a wide range of techniques including high-throughput assay development, machine learning, data mining, visualization, and structure-based modeling [58]. During the early stages of drug development, it is urgent to evaluate the ADMET properties of the new compounds to avoid dragging them off the market later [59]. Herein, the ADMET properties for T-1-NCA were computationally predicted using Discovery Studio® based on its reference molecule, erlotinib. The ADMET results of T-1-NCA versus erlotinib (Fig. 14) demonstrated an acceptable degree of drug-likeness because it was predicted to have a very low ability to cross the blood–brain barrier, BBB, and to be non-hepatotoxic and non-inhibitor of the cytochrome P-450, CYP2D6. T-1-NCA also had good aqueous solubility and medium intestinal absorption levels, as shown in Table 3.

Fig. 14
figure 14

Computational prediction of ADMET parameters for T-1-NCA and erlotinib

Full size image
Table 3 ADMET parameters for T-1-NCA and erlotinib
Full size table

2.5 In silico toxicity studies

The use of in silico methods has proven to be crucial in drug development because they reduce the need for in vitro and in vivo experiments and reduce the associated time delay [60]. As part of in silico toxicity prediction, the software compares the basic chemical structural descriptors of the studied molecule/s with those of thousands of molecules that have been described as safe or toxic [61], using the structure–activity relationship (SAR)-predictive toxicity method. Through Discovery Studio software, toxicity models have been created, which are used to estimate toxicity parameters: FDA Rodent Carcinogenicity in Mouse-male (FDA-C-MM), Ames Mutagenicity (A-M), a mouse tumor cell potency (TD50-R), a rat maximum tolerable dose, a rat oral LD50 (R-O-LD50), a rat chronic LOAEL (R–C-LOAEL). Based on the computed models, the designed theobromine derivative T-1-NCA revealed general safety levels, as shown in Table 4.

Table 4 In silico toxicity studies of T-1-NCA and erlotinib
Full size table

2.6 Chemistry

In the current study, theobromine 1 was treated with alcoholic KOH while being continuously stirred to produce potassium salt 2 [51]. In order to generate the corresponding target product T-1-NCA (5), the formed salt 2 was then reacted with 2-chloro-N-cyclohexylacetamide 4 in DMF (Scheme 1).

Scheme 1
scheme 1

Synthetic pathway of T-1-NCA (5)

Full size image

Compound 5 was obtained as off-white crystals (yield, 84%); m. p. = 257–259 °C. The molecular formula C15H21N5O3 was deduced based on the basis of the elemental analysis and EI-MS. Besides, the 1H NMR spectrum in (DMSO-d6) (Table 5) displayed signals for amidic proton at δH 8.05 ppm and singlet signal for the proton of theobromine moiety at δH 7.97 ppm. On the other hand, the aliphatic region contains two protons that appeared as a singlet signal at δH 4.41 ppm corresponding to the methylene group, the α-amino-methine proton at 3.52 (m), and two methyl signals, at δH 3.89 (s, 3H) and 3.42 (s, 3H). Finally, the 1H NMR spectrum displayed the multiple protons of the cyclohexyl ring at δH (1.69–1.03) ppm. The 13C NMR spectrum afforded 13 carbon signals that were assigned as two methyls, one methylene, six sp3 methines, one sp2 methine, two quaternary sp2, and three amide carbonyl carbons. All the above-mentioned data confirmed the validity of the proposed structure.

Table 5 NMR spectroscopic data for 5 in DMSO-d6, δ ppm)
Full size table
figure a

2.7 Biology

2.7.1 In vitro EGFR inhibition

The potentiality of T-1-NCA against EGFRWT protein was investigated in vitro to examine the performed design as well as the great computational results. Compare the strongly inhibited EGFR protein with an IC50 value of 24.25 nM (Table 6). These results were in agreement with those acquired in silico, demonstrating the strong suppression potential of the compound.

Table 6 IC50 values of T-1-NCA and erlotinib against EGFR kinase
Full size table

2.7.2 Cytotoxicity

Considering the notably inhibitory potentials (in silico and in vitro) of T-1-NCA against EGFRWT, it is expected to exert promising anticancer effects. Therefore, the cytotoxicity of T-1-NCA was examined in vitro against the lung carcinoma epithelial A549 and the human colon cancer HCT-116 malignant cell lines using erlotinib as a reference drug (Table 7). Interestingly, T-1-NCA exhibited potent anticancer against the aforementioned cancer cells with IC50 values of 40.20 and 34.05 µM, respectively, compared to erlotinib, which had IC50 values of 17.32 and 17.13 µM, respectively.

Table 7 In vitro anti-proliferative of T-1-NCA against HCT-116 and A549 cell lines
Full size table

To verify the safety profile of T-1-NCA and investigate its specificity, experiments were conducted using a Vero cell line. Results indicated that T-1-NCA exhibited a significant IC50 value of 132.4 μM and remarkably high selectivity index values (SI) of 3.89 and 3.29 against the two cancer cell lines.

2.7.3 Flow cytometric analysis and apoptosis assay

Apoptosis (programmed cell death) is a regular sequential process of cell death that maintains a homeostatic equilibrium between the rate of cell production and cell death. A misalignment of this balancing function, on the other hand, can contribute to aberrant cell development/proliferation, autoimmune diseases, and so on. Apoptosis is therefore believed to be critical from the development of an embryo through the growth of an organism, contributing to tissue regeneration and the elimination of inflammatory cells [62]. Since the discovery of DNA breakage in thymocytes following exposure to glucocorticoids in the 1980s, the induction of apoptosis became a logical and viable therapeutic strategy [5, 63].

To confirm the apoptotic properties of T-1-NCA, the apoptosis percentage in the A549 cells was checked by Annexin V and PI double stains [64, 65]. The results revealed that T-1-NCA induced an increase of the apoptotic cells percentage in the early stage of apoptosis (from 0.71 to 4.39%), late stage of apoptosis (from 0.13 to 21.02%), as well as the total stage (from 3.06 to 33.52%). Furthermore, the necrosis percentage was promoted to be 8.11, compared to 2.22% in the control cells (Fig. 15 and Table 8). Briefly, T-1-NCA successfully arrested the HCT-116 cell cycle at the G2/M phase resulting in cytotoxic potentialities that may be related to apoptosis.

Fig. 15
figure 15

Flow cytometric charts of apoptosis in A549 cells exposed to T-1-NCA

Full size image
Table 8 Effect of T-1-NCA on stages of the cell death process in HCT-116 cells
Full size table

2.7.4 Apoptotic proteins assay

The effects of compound T-1-NCA on HCT-116 were investigated in this study, with a focus on the expression levels of BAX, Bcl-2, caspase-3, and caspase-9. T-1-NCA was administered at a concentration of 34.05 µM, which is the IC50 value. As shown in Table 9,  T-1-NCA had a significant impact on the expression levels of these proteins compared to the control group. The results showed that the proapoptotic protein BAX was increased by 3.8 times when compared to the control. BAX is a protein that promotes cell death or apoptosis. The anti-apoptotic protein Bcl-2 was also significantly reduced by 5.8-fold compared to the control. Bcl-2 is a protein that inhibits apoptosis and promotes cell survival. Therefore, the increase in BAX and decrease in Bcl-2 indicate that T-1-NCA is promoting apoptosis in HCT-116 cells.

Table 9 T-1-NCA’s activities on apoptotic proteins in HCT-116 cell lines
Full size table

The study also found that T-1-NCA significantly increased the levels of caspase-3 and caspase-9. Caspases are a family of enzymes that play a critical role in apoptosis. Caspase-3 is an initiator caspase, which means it is activated at the start of the apoptotic pathway, while caspase-9 is an executioner caspase, which means it is activated downstream in the pathway. The increase in caspase-3 levels by 4.6-fold and caspase-9 levels by threefold compared to the control indicates that T-1-NCA is an apoptosis promotor. These findings provide a better understanding of the mechanism of action of T-1-NCA and its potential as an anti-cancer agent (Fig. 16).

Fig. 16
figure 16

T-1-NCA’s activities on apoptotic proteins in HCT-116 cell lines

Full size image

3 Experimental

3.1 Docking studies

It was operated for T-1-NCA against EGFRWT (wild form) and EGFRT790M (mutant form) by MOE2014 software [66]. The supplementary section provides additional illustrations and details.

3.2 M D simulations

It was operated for EGFR-T-1-NCA complex by the CHARMM-GUI web server [67] and GROMACS 2021. The supplementary section provides additional illustrations and details. The supplementary section provides additional illustrations and details [68, 69].

3.3 MM-GBSA

It was operated for EGFR-T-1-NCA complex by the Gmx_MMPBSA package [70, 71]. The supplementary section provides additional illustrations and details.

3.4 ED analysis

Principal component analysis (PCA) was employed for EGFR-T-1-NCA complex to investigate the dynamic motion of alpha carbons located in the amino acid sequence spanning from Glu826 to Leu1161 [72]. Supplementary section provides additional illustrations and details.

3.5 Bi-dimensional assays

To compare frames within the reduced subspace, we merged, aligned, created a new C matrix, and plotted the projections [73]. The supplementary section provides additional illustrations and details.

3.6 DFT

It was operated for T-1-NCA by Gaussian 09 and GaussSum3.0 programs. The supplementary section provides additional illustrations and details.

3.7 ADMET studies

It was operated for T-1-NCA by Discovery Studio 4.0 [74]. The supplementary section provides additional illustrations and details.

3.8 Toxicity studies

It was operated for T-1-NCA by Discovery Studio 4.0 [75]. The supplementary section provides additional illustrations and details.

3.9 In vitro EGFR inhibition

It was operated for T-1-NCA by Human EGFR ELISA kit. The supplementary materials show a comprehensive explanation.

3.9.1 In vitro antiproliferative activity

It was operated for T-1-NCA by MTT procedure [76, 77]. The supplementary materials show a comprehensive explanation.

3.9.2 Safety assay

It was operated for T-1-NCA by the calculation of selectivity index after MTT assay utilizing Vero cell lines [78]. The supplementary section provides additional illustrations and details.

3.9.3 Flow cytometry of apoptosis

It was operated for T-1-NCA flow cytometry analysis technique. The supplementary section provides additional illustrations and details.

3.9.4 Apoptotic proteins assay

It was operated by RT-PCR using the kit (Qiagen RNA extraction/BioRad SYBR green PCR MMX). The supplementary section provides additional illustrations and details.

4 Conclusion

In conclusion, this study has shown that the newly designed semisynthetic N-cyclohexyl acetamide derivative, T-1-NCA, has potent anti-EGFR apoptotic anti-proliferative activities in vitro. The computational investigations, including molecular docking, molecular dynamics simulations, and PLIP experiments, suggested that T-1-NCA can effectively bind and inhibit the EGFR protein. The in vitro assays demonstrated that T-1-NCA had a significantly strong IC50 value for EGFRWT inhibition. Moreover, T-1-NCA demonstrated significant and safe anti-proliferative effects against A549 and HCT-116 cell lines compared to erlotinib, indicating its potential as a therapeutic agent for EGFR-dependent cancers. The apoptotic effects of T-1-NCA were also confirmed through the significant increase in BAX, Casp3, and Casp9 levels and the significant decrease in Bcl-2 levels. These results suggest that T-1-NCA is a promising candidate for further development as an anti-cancer drug targeting EGFR.