Identification of novel HER2 ınhibitors: potential therapeutics for breast cancer

Abstract

Human epidermal growth factor receptor-2 (HER2) is a tyrosine kinase receptor involved in cell growth and differentiation. Targeting HER2 is a critical strategy in HER2-positive breast cancer treatment. Despite advancements in HER2-targeted therapies, drug resistance and side effects remain significant challenges. Therefore, identifying novel HER2 inhibitors with the potential to overcome resistance mechanisms while maintaining favorable drug-like properties is essential. Identifying novel HER2 inhibitors with high binding affinity and favorable drug-like properties is essential for overcoming these limitations. This study employed molecular docking and molecular dynamics simulations to evaluate the binding potential of plant-derived and synthetic compounds against HER2. The most promising candidates were further analyzed using ADMET profiling and binding free energy calculations to assess their drug-likeness and binding free energy. Among the tested compounds, axitinib, prunetin, and silymarin demonstrated strong HER2-binding affinities comparable to established inhibitors such as TAK-285 and lapatinib. Molecular dynamics simulations revealed that prunetin formed the most stable HER2-ligand complex, while axitinib exhibited the lowest binding free energy, indicating a strong interaction potential. ADMET analysis confirmed axitinib and prunetin as favorable drug candidates, whereas silymarin exhibited lower intestinal absorption. In conclusion, axitinib and prunetin emerged as promising HER2 inhibitors that may offer therapeutic advantages by addressing both drug resistance and toxicity concerns in HER2-positive breast cancer treatment. Prunetin, with its lower toxicity and higher stability, presents a safer therapeutic option, whereas axitinib offers high binding affinity. These findings suggest that these compounds could help overcome resistance and side effects associated with current HER2-targeted therapies.

Introductıon

Human epidermal growth factor receptor-2 (HER2) is a member of the transmembrane epidermal growth factor family of tyrosine kinase receptors involved in signaling pathways regulating cell growth and differentiation [1]. HER2 is overexpressed in approximately 15–30% of breast cancers and this overexpression is associated with increased risk of disease recurrence and poor prognosis. Therefore, targeting HER2 has become a critical strategy in the treatment of HER2-positive breast cancer, leading to the development of various HER2-targeted therapies such as trastuzumab, pertuzumab, and lapatinib [2].

Lapatinib and neratinib are examples of HER2 tyrosine kinase inhibitors that bind to the ATP-binding site of the receptor, thereby preventing ATP from binding. The absence of ATP binding inhibits phosphorylation, which in turn blocks the downstream signaling pathways responsible for cell proliferation. As a result, the growth of cancerous cells is suppressed [3]. However, there remains a need for novel alternatives to these ATP-mimetic inhibitors to overcome resistance and improve therapeutic outcomes.

One of the biggest difficulties in treating breast cancer is the emergence of drug resistance in HER2-targeted therapy [4]. Several studies have identified specific molecular mechanisms contributing to resistance against HER2 tyrosine kinase inhibitors such as lapatinib [5]. These mechanisms can be listed as: the upregulation of compensatory pathways like PI3 K/Akt and ER signaling, overexpression of alternative receptors such as AXL and HER3, and genetic alterations like PIK3 CA mutations or PTEN loss. These adaptive changes allow cancer cells to bypass HER2 inhibition and continue proliferating despite targeted therapy, highlighting the importance of combination strategies and the development of next-generation inhibitors [5]. Resistance mechanisms include changes in the HER2 receptor, activation of alternative signaling pathways, and changes in cell survival mechanisms [6, 7]. Therefore, the need to discover new HER2 inhibitors that can overcome these resistance mechanisms and develop treatment options continues.

Another critical problem in HER2-targeted therapies is treatment-related side effects that negatively affect patients'quality of life [8]. These side effects reduce adherence to treatment and can threaten the overall health of patients. Therefore, developing alternative HER2 inhibitors that are less toxic and better tolerated is crucial to overcome the limitations of current treatment options.

Molecular docking is a computational technique that predicts the binding affinity of the stable complex formed by two molecules, such as a ligand and receptor protein, by modeling the bound structure. This technique is widely used in drug discovery to screen and optimize potential drug candidates by predicting their binding affinity to target proteins [9]. Recent advances in computational power and algorithm development have significantly improved the accuracy and efficiency of molecular docking studies, making this method very useful for the early stages of drug discovery [10]. Molecular dynamics (MD) simulations, on the other hand, simulate the physical movements of atoms and molecules over time, offering dynamic insights into molecular interactions and conformational changes. This technique complements molecular docking by validating binding stability and exploring the flexibility of protein–ligand complexes under physiological conditions [11].

Natural products have historically been a rich source of therapeutic agents and many existing drugs are derived from natural compounds [12]. In particular, plant-derived compounds have shown significant potential as anticancer agents due to their diverse chemical structures and biologically active properties [13, 14]. The investigation of HER2 inhibitors through molecular docking and molecular dynamics simulations aims to discover and validate natural and chemical compounds as potential therapeutic agents against HER2-positive cancers [15]. HER2 is an important target in the treatment of various cancers, including breast, gastric, and lung cancers, due to its role in cell proliferation and survival [16]. Traditional HER2 inhibitors include monoclonal antibodies, tyrosine kinase inhibitors, and antibody–drug conjugates, each designed to inhibit the oncogenic signaling pathways of HER2 [17]. Molecular computational techniques facilitate the discovery of new inhibitors by allowing researchers to predict and analyze interactions between small molecules and HER2 at the molecular level [18, 19].

The search for HER2 inhibitors represents a promising frontier in cancer treatment by combining traditional knowledge with modern computational techniques. Although the efficacy and safety of these natural and chemical compounds require further clinical and laboratory validation, their potential to provide novel and less toxic treatment options is significant. The integration of metabolomic and genomic data with molecular docking and dynamic simulations, together with advances in non-invasive diagnostic techniques, suggests that a new era of personalized and effective cancer treatment strategies has begun.

In this study, we focused on determining the potential of plant-derived and chemical compounds to be evaluated as HER2 inhibitors using molecular docking and molecular dynamics simulation techniques. We aimed to identify novel HER2 inhibitors with high binding affinity and favorable drug-like properties. By searching for new and effective drug molecules that can be used in HER2-targeted therapies, we tried to contribute to the discovery of natural compounds that can overcome drug resistance and side effects and thus improve cancer treatment. The findings of this study are expected to guide in vitro and in vivo research for the development of new drugs targeting HER2.

Methods

Protein structures

HER2 (PDB ID:3PP0) 3D crystal structure was downloaded from the protein databank (PDB) [20]. Water molecules and native ligand (03Q) structure were removed from the protein structure using Discovery Studio Visualizer [21]. Protein structures were prepared for molecular docking using Autodock Tools: Polar hydrogen bonds were added and Gasteiger and Kollman charges were calculated [22].

Ligand library

HER2 plant inhibitors were obtained from the ChemFaces database (http://chemfaces.com/). Natural ligand (03Q), silymarin, and axitinib (according to literature search) were added to this library. In addition, lapatinib and TAK-285 anticancer drugs acting on HER2 were used as ligands in molecular docking studies for comparison with phytochemical ligands. The 3D structures of the ligands were retrieved from the PubChem database (https://pubchem.ncbi.nlm.nih.gov/). Ligand structures were prepared for molecular docking using Autodock Tools.

Molecular docking

Autodock Vina [23, 24] was used for molecular docking of ligands to the target protein. The binding site coordinates (x = 17.098971, y = 16.548676, z = 26.600647) and grid size (x = 15, y = 15, z = 15) were determined using Discovery Studio Visualizer. It was determined as the binding site of the natural ligand in the crystal protein structure. Taking this binding site as a reference, other plant-derived ligands were bound to the same site. The grid size was also determined with reference to the natural ligand. After docking, the interactions between receptor and ligand molecules were examined using the Discovery Studio Visualizer program.

We obtained the target structure from the PDB database (PDB ID: 3PP0). The structure was determined using X-ray crystallography. Within the structure, the ligand named 03Q is bound to the ATP-binding pocket of the receptor. Our aim is to inhibit the target by preventing phosphorylation, which we plan to achieve by allowing a molecule to bind to the ATP-binding site, thereby blocking ATP binding. For this reason, the region where the 03Q ligand is bound was selected as the reference site during the docking procedure. While defining the docking grid box, its size was adjusted to include the residues located in the ATP-binding site and to encompass the reference ligand.

Drug similarity and ADMET prediction

Drug similarity was assessed using SWISSADME (http://swissadme.ch/index.php). All ligands were filtered according to Lipinski, Ghose, Veber, Egan, and Muegge rules and the ligands that met all rules were selected (Table 1).

Table 1.

Filters used for drug similarity

Lipinski [25]	Ghose [26]	Veber [27]	Egan [28]	Muegge [29]
MW ≤ 500	160 ≤ MW ≤ 480	Rotatable bonds ≤ 10	WLOGP ≤ 5.88	200 ≤ MW ≤ 600
MLOGP ≤ 4.15	− 0,4 ≤ WLOGP ≤ 5.6	TPSA ≤ 140	TPSA ≤ 1316	− 2 ≤ XLOGP ≤ 5
N or O ≤ 10	− 40 ≤ MR ≤ 130			TPSA ≤ 150
NH or OH ≤ 5	20 ≤ atoms ≤ 70			Num. rings ≤ 7 Num. carbon > 4 Num. heteroatoms > 1 Num. rotatable bonds ≤ 15
				H-bond acc. ≤ 10 H-bond don. ≤ 5

Molecular dynamics simulations

The molecular dynamics simulations were performed for the best predicted complex structures of HER2 targets with the candidate drugs, using GROMACS software [30] and CHARMM36 force field. Initially, the ligand and protein structures were prepared separately and combined in the final step. After generating the protein's.gro file via GROMACS, the ligands’ topology files were prepared by adding hydrogens using Avogadro software [31, 32], using CGenFF [33–35] and creating the final.gro files in GROMACS. The.gro files of the protein and ligands were then combined, while including the ligands'topology files in the topol.top files. A cubic water box with dimensions ensuring a minimum distance of 1.0 nm between the complex and the box boundaries was generated using GROMACS with the -d flag. The complex was then solvated with TIP3P water molecules. Sodium (Na +) and chloride (Cl-) ions were added to neutralize the system. After energy minimization and equilibration under 310 K temperature and 1 atm pressure, a molecular dynamics simulation of 250 ns was performed. These procedures were applied to three drug ligand molecules: Axitinib, Silymarin, and Prunetin and three parallel simulations were run for each ligand.

After the simulations were completed, the gmx rms, gmx rmsf, gmx gyrate, and gmx hbond commands were used to obtain the RMSD, RMSF, radius of gyration, and the number of hydrogen bonds formed between the ligands and the protein throughout the simulation, respectively. For the simulation analyses, averages of the parallel runs were considered.

Binding free energy analysis

The molecular mechanics Poisson–Boltzmann surface area (MM-PBSA) method was used to calculate the free binding energy, and per-residue decomposition analysis (PRED) was performed. The gmx_mmpbsa package of GROMACS was utilized for this method [36]. For each protein–ligand complex, energy calculations were repeated three times, and the averages of the runs were taken. MM-PBSA calculations were conducted for 2500 frames obtained from 250 ns molecular dynamics simulations, covering the entire trajectory. The formula used to calculate the free binding energy is shown below:

$$\mathrm{\Delta }{\text{G}}_{\text{bind}}=\mathrm{\Delta }{\text{G}}_{\text{gas}}+\mathrm{\Delta }{\text{G}}_{\text{sol}}$$

$$\mathrm{\Delta }{\text{G}}_{\text{gas}}={\text{E}}_{\text{gas}}-\text{T}\mathrm{\Delta }\text{S}$$

$${\text{E}}_{\text{gas}}={\text{E}}_{\text{int}}+{\text{E}}_{\text{vdw}}+{\text{E}}_{\text{ele}}$$

$${\text{G}}_{\text{sol}}={\text{G}}_{\text{GB}}+{\text{G}}_{\text{SA}}$$

$${\text{G}}_{\text{SA}}=\gamma \text{SASA}$$

In the formula, the binding energy of the complex (∆G_bind) is defined as the difference between the free energies of the receptor and the ligand. Gas phase energy (∆G_gas) includes gas phase potential energy (E_gas) and entropic contributions (T∆S). E_gas consists of internal bond energies (E_int), van der Waals energy (E_vdw), and electrostatic energy (E_ele). Solvation energy (G_sol) is composed of the polar solvent contribution (G_GB) and the surface area energy (G_SA), which represents the effect of the apolar solvent. SASA refers to the solvent-accessible surface area [37].

Results

Molecular docking studies

Molecular docking is a crucial method to investigate interactions between target proteins and ligands. Binding energy (kcal/mol) data allow the binding affinities of various ligands to their respective target proteins to be analyzed and compared. A lower binding energy indicates a greater affinity of the ligand for the receptor. The ligand exhibiting the highest affinity could be selected as a potential drug for further research.

Molecular docking studies were validated against the reference native ligand (03Q) present in the 3D crystal structure of HER2 (PDB ID:3PP0). The RMSD of the 03Q ligand in the crystal structure and the re-docked 03Q ligand was calculated as 1.0065 Å < 1.5 Å, which means a desirable result. The binding energies of 03Q (reference), TAK-285, and lipatinib anticancer drugs acting on HER2 were determined as − 11.9 kcal/mol, − 11.9 kcal/mol, and − 10.7 kcal/mol, respectively (Table 2). Axitinib (− 11.1 kcal/mol), considered in the literature to be a HER2 inhibitor, exhibited a lower binding energy compared to lipatinib. Silymarin showed a binding energy similar to that of lipatinib. Likewise, prunetin (− 10.1 kcal/mol), obtained from the Chemfaces database, demonstrated a binding energy comparable to both lipatinib and silymarin (Table 2).

Table 2.

Summary of estimated binding affinity (kcal/mol) of docked ligands against HER2 and interacting residues in the binding sites

Ligands	Binding energy (kcal/mol)	Target residues interacting with the ligand
		Hydrogen bond	Hydrophobic Interactions
03Q (Reference)	− 11.9	MET801, ASP863	LEU726, VAL734, ALA751, LYS753, MET774, LEU785, PHE864
TAK-285 (Anticancer drug)	− 11.9	MET801, SER728	LEU726, VAL734, ALA751 LYS753, LEU785, MET774, LEU796, LEU852, PHE864
Axitinib	− 11.1	CYS805	VAL734, ALA751, MET774, LEU785, LEU796, THR798, LEU852, PHE864
Lapatinib (Anticancer drug)	− 10.7	MET801	LEU726, VAL734, ALA751, LYS753, MET774, LEU785, LEU796, PHE864
Silymarin	− 10.2	LYS753, ASP863	LEU726, VAL734, LEU796, LEU800, CYS805, LEU852
Prunetin	− 10.1	SER783, ARG784, THR798, THR862	VAL734, ALA751, LYS753, ALA771, MET774, LEU785, LEU796, LEU852, PHE864

Ligands with strong binding affinity attach to a binding site similar to the reference ligand, as indicated by the red-marked amino acids in Table 2. The selected ligands have the potential to be HER2 inhibitors by interacting with amino acids in the HER2 active site such as TAK-285 and lapatinib anticancer drugs. It was also revealed by two-dimensional (2D) and three-dimensional (3D) visualization that axitinib, silymarin, and prunetin ligands bind to the HER2 binding site with a high affinity similar to the reference ligands (Fig. 1).

Fig. 1.

Interaction analysis of binding poses of ligands in the HER2 binding domain represented in 2D and 3D. The interactions were visualized by Discovery Studio

Drug similarity properties

The ligands with high binding affinity to HER2 were determined to comply with the Lipinski, Ghose, Veber, Egan, and Muegge rules by using the SwissADME (http://swissadme.ch/index.php) web tool. Axitinib and prunetin, which were investigated as potential inhibitors of HER2 by molecular docking, fulfill all drug-like property rules. Silymarin, on the other hand, fulfills Lipinski's five rules, but violates one rule each in the Ghose, Veber, Egan, and Muegge rules (Table 3).

Table 3.

Investigation of Drug Similarity Properties

Compound	Lipinski (Yes/No, Violation)	Ghose (Yes/No, Violation)	Veber (Yes/No, Violation)	Egan (Yes/No, Violation)	Muegge (Yes/No, Violation)
Axitinib	Yes	Yes	Yes	Yes	Yes
Silymarin	Yes	No, 1 violation: MW > 480	No; 1 violation: TPSA > 140	No; 1 violation: TPSA > 131.6	No; 1 violation: TPSA > 150
Prunetin	Yes	Yes	Yes	Yes	Yes

MW Molecular Weight, TPSA Topological Polar Surface Area

Investigation of ADMET properties

Intestinal absorption and colon adenocarcinoma-2 cell (Caco2) permeability are two important parameters determining drug absorption. If the percentage absorbed intestinally is less than 30%, the substance is considered poorly absorbed. Caco2 permeability logPapp > 0.9 Caco2 permeability is considered high [38]. Axitinib and prunetin have acceptable intestinal absorption values, whereas silymarin has low Caco2 permeability (Table 4). VDss is a measure of the distribution of the molecule of interest in tissues, and if logVDss is greater than 0.45, the compounds are well distributed in tissues, whereas logVDss smaller than − 0.15 means they are poorly distributed [38]. Accordingly, the distribution of axitinib in tissues was low, while the distribution of silymarin and prunetin in tissues was moderate. Blood–brain barrier (BBB) permeability is considered to be poor when LogBB values are less than −1 [38]. BBB permeability of silymarin is found to be poor (Table 4). Cytochrome P450 s play an important role in the metabolism of many drugs. Cytochrome P450 inhibitors can significantly alter the pharmacokinetics of these drugs. None of the compounds studied are CYP2D6 substrates, but all compounds except silymarin are CYP3 A4 substrates (Table 4). Total clearance of prunetin is higher than axitinib and silymarin. None of the compounds, except axitinib, have AMES and hepatotoxicity (Table 4).

Table 4.

ADMET features

	Parameters	Axitinib	Silymarin	Prunetin
Absorption	Caco2 permeability (log Papp in 10–6 cm/s) logPapp > 0.9 Caco2 permeability is considered high	0.913	0.435	1.023
	Intestinal absorption (human) (%) less than 30%, the substance is considered poorly absorbed	92.321	61.861	95.535
	Skin Permeability (log Kp)	− 2.743	− 2.735	− 2.745
Distribution	VDss (human)(log L/kg) logVDss > 0.45 means images are well distributed in the tissues, logVDss < −0.15 means poor distribution	− 0.18	0.369	− 0.066
	BBB permeability (log BB) LogBB values less than − 1.0 indicate poor blood–brain barrier permeability	− 0.18	− 1.21	− 0.32
	CNS permeability (log PS)	− 2.105	− 3.639	− 2.188
Metabolism	CYP2D6 substrate (Yes/No)	No	No	No
	CYP3 A4 substrate (Yes/No)	Yes	No	Yes
	CYP2D6 inhibitior (Yes/No)	No	No	No
	CYP3 A4 inhibitior (Yes/No)	Yes	No	Yes
Excretion	Total Clearance (log ml/min/kg)	0.116	− 0.103	0.272
	Renal OCT2 substrate	No	No	No
Toxicity	AMES toxicity (Yes/No)	Yes	No	No
	Hepatotoxicity (Yes/No)	Yes	No	No

Molecular dynamics simulations and binding free energy analysis

According to the results of molecular dynamics simulations of axitinib, silymarin, and prunetin molecules with their target HER2 protein, all three complexes were observed to stabilize over time. The prunetin complex exhibited greater structural stability compared to the other two molecules, while the silymarin complex demonstrated higher structural flexibility (Fig. 2A). When the RMSD values of only the ligands were analyzed, prunetin showed a good fit to the protein’s binding site and displayed lower mobility in this region (Fig. 2B). This indicates that prunetin may achieve stronger or more specific binding (Figs. 2A and B). In contrast, axitinib exhibited higher RMSF values for the residues at the binding site, compared to the other molecules (Fig. 2C). This finding suggests that axitinib forms a less stable structure in the binding region. The structures taken at 175 th nanoseconds of simulations, when all three complex structures have approximately the same RMSD value, are shown in Fig. 3 and it can be observed that the ligands bind to similar regions on the HER2 protein. In the radius of gyration graph, which is a measure of compactness, silymarin was found to form a less compact complex than the other ligands (Fig. 2D). This suggests that silymarin may lead to reduced binding stability and undesirable conformational flexibility in the protein. Conversely, prunetin exhibited the most compact structure, indicating stronger molecule-protein interactions and a more stable, organized complex.

Fig. 2.

A RMSD values of protein–ligand complex structures, B RMSD values of ligands, C RMSF of protein–ligand complex structures, D Radius of gyration values of protein–ligand complex structures, E Number of hydrogen bonds between ligands and protein

Fig. 3.

Structures of the protein–ligand complexes from molecular dynamics simulations at the 175 th nanosecond

When the number of hydrogen bonds formed by the ligand molecules with the protein during simulation was analysed, silymarin was found to have the highest average number of hydrogen bonds (Fig. 2E). This may imply that silymarin establishes stronger electrostatic interactions with the protein. However, silymarin molecule has a higher total number of atoms compared to the other two molecules and the increase in number of hydrogen bonds may be misleading. Therefore, we examined the 2D interactions of the ligands with the target structures corresponding to the nanoseconds where all three molecules exhibited the highest number of hydrogen bonds during their individual simulations (Fig. 4 and Table 5). The common residues of the reference ligand, interacting with HER2 molecule are MET801, ASP863, LEU726, VAL734, ALA751, LYS753, MET774, LEU785, and PHE864 (Table 2) and drugs are analyzed in terms of tehir interaction with the same residues on the target (Table 5, residues indicated in red). Despite prunetin being the smallest ligand, it targets a higher number of common residues with the reference ligand than the larger drug silymarin.

Fig. 4.

The 2D interactions of the three molecules with the protein at the time when they exhibited the highest number of hydrogen bonds during the simulation

Table 5.

The interacting target residues with the three ligand molecules for frames where they exhibited the highest number of hydrogen bonds during the simulation

Ligands	Interacting Target Residues
	Hydrogen Bond	Hydrophobic Interactions
Axitinib	ALA751, LEU796	LEU726, VAL734, LYS753, MET774, CYS805, LEU852, PHE864
Silymarin	THR798, LEU800, MET801, CYS805, ASP863	GLY727, ALA751, LYS753, LEU852
Prunetin	ALA751, THR798, MET801, ASP863	LYS753, LEU785, LEU796, LEU852

MM-PBSA and per-residue decomposition (PRED) analyses provide insights into the energetic contributions of individual residues to ligand binding, highlighting key interaction regions at the molecular level. The binding free energies of axitinib, silymarin, and prunetin are − 36.17 kcal/mol, − 16.78 kcal/mol, and − 24.9 kcal/mol, respectively (Table 6 and Fig. 5A). When the binding free energies of axitinib, prunetin, and silymarin ligands are compared, axitinib exhibits the lowest binding energy, indicating the strongest interaction with the protein (Fig. 5A). Prunetin and silymarin, on the other hand, display higher binding energies, suggesting weaker interactions with the protein.

Table 6.

Binding free energies of Axitinib, Silymarin, and Prunetin

Energy Components (kcal/mol)	Axitinib	Silymarin	Prunetin
∆Gbind	− 36.17	− 16.78	− 24.9

Fig. 5.

A Binding free energies of Axitinib, Silymarin and Prunetin, B Per-residue decomposition analysis of protein–ligand complexes

Per-residue decomposition analysis for each ligand reveals the contributions of specific residues to ligand binding energy and emphasizes critical interaction points, such that LYS753 weakens silymarin’s binding by contributing positive energy (Fig. 5B). In contrast, target protein’s LEU852, VAL734, and LEU726 residues provide negative energy contributions, enhancing the stability of silymarin’s interaction. LEU796 and ALA751 are key residues that enhance axitinib’s binding, through hydrophobic and electrostatic interactions. LEU796, THR798, and VAL734 are primary residues that strengthen Prunetin binding. However, LYS753 contributes positive energy, weakening prunetin’s binding and resulting in a relatively less favorable interaction (Fig. 5B).

Reevaluating the ADMET analyses in light of the binding dynamics results, axitinib and prunetin appear to be more suitable as drug candidates in terms of binding stability, although silymarin demonstrates low toxicity and high safety. Thus, when all the results are evaluated together, axitinib and prunetin emerge as the most prominent molecules targeting HER2.

Dıscussıon

Plant-derived drugs have been used therapeutically for a very long time. The development of phytochemicals for the prevention and treatment of difficult-to-treat diseases such as cancer is possible thanks to advanced screening methods for bioactive components of plants. By adopting state-of-the-art technology and innovative research techniques, the discovery of plant-derived and synthetic bioactive chemicals that can be used to treat and prevent malignancies without major adverse effects is critical for innovative cancer therapies [39]. In this study, plant-derived and synthetic compounds were comparatively evaluated using molecular docking and molecular dynamics simulation methods to identify new potential HER2-targeted inhibitors, and compounds thought to be HER2 inhibitors were identified. These compounds were filtered according to Lipinski, Ghose, Veber, Egan, and Muegge drug eligibility criteria and included in molecular docking and molecular dynamics simulation studies. The study results revealed that Axitinib, Silymarin, and Prunetin showed high affinity for the HER2 binding site, while Axitinib and Prunetin were particularly promising in terms of binding energy values and interacting amino acid residues. This study was particularly motivated by the need to identify HER2 inhibitors that not only offer high binding affinity but also have the potential to overcome the issue of drug resistance and minimize side effects associated with current treatments. This dual objective is critical given that drug resistance and toxicity continue to limit the long-term efficacy of HER2-targeted therapies.

The search for HER2 inhibitors constitutes an important part of combating HER2-positive cancers by utilizing natural bioactive compounds found in various plants. Despite significant advances in conventional therapies, including monoclonal antibodies and small-molecule tyrosine kinase inhibitors, challenges such as side effects and resistance to treatment persist [40]. Therefore, there is growing interest in the potential of phytochemicals to offer additional and less toxic alternatives or complements to existing therapies [41]. Due to the important role of HER2 in the regulation of cell proliferation and differentiation, its overexpression is associated with aggressive forms of breast cancer and other malignancies [2]. The anti-cancer activity of plant-derived compounds can be explained by the fact that they block HER2-specific signaling pathways and thus prevent the growth and spread of cancer cells. These effects of phytochemicals make them potential therapeutic agents against HER2-positive breast cancer and other cancers.

Therapeutic targeting of HER2 using plant-derived and synthetic compounds depends on their ability to inhibit the HER2 signaling pathway, which is crucial in tumor growth and survival. Several phytochemicals show anticancer effects through the inhibition of key metabolic pathways, including inhibition of HER2 [42]. Recent studies highlight the anticancer properties of extracts from various medicinal plants, which show the ability to reduce HER2 expression levels and induce apoptosis in HER2-positive breast cancer cells [43, 44]. It is stated that the efficacy of these plant-derived compounds can be increased when used in combination with conventional HER2-targeted therapies. For example, clinical studies have shown that the combination of lapatinib, a tyrosine kinase inhibitor, and monoclonal antibody gives more effective results in breast cancer treatment [45]. This supports the rationale that novel compounds, such as those identified in this study, could be used either as standalone or adjunctive agents to overcome drug resistance and reduce side effects. The current situation shows that the potential application area of plant-derived HER2 inhibitors is expanding and provides a reason for continuing potential inhibitor research for HER2.

The results of this study provide valuable insights into the potential of some natural compounds as HER2 inhibitors, specifically in the context of overcoming resistance to traditional HER2-targeted therapies. HER2 overexpression, which is implicated in approximately 15–30% of breast cancers, has been a critical target in cancer treatment due to its association with increased recurrence risk and poor prognosis [46]. Despite the clinical success of HER2-targeted drugs such as trastuzumab, pertuzumab, and lapatinib, resistance remains a significant hurdle. Our molecular docking and molecular dynamics simulation studies aimed to identify novel HER2 inhibitors, which could offer alternative therapeutic options with potentially lower toxicity. Among the compounds tested, axitinib, silymarin, and prunetin demonstrated promising binding affinities comparable to established anticancer drugs like TAK-285 and lapatinib. Axitinib, in particular, showed a binding energy of − 11.1 kcal/mol, which is comparable to TAK-285 (− 11.9 kcal/mol) and significantly lower than lapatinib (− 10.7 kcal/mol), indicating strong potential as a HER2 inhibitor. This supports the idea that these compounds can effectively inhibit HER2.

Molecular dynamics simulation analyses showed that the candidate compounds formed stable complexes with the HER2 protein. The high binding affinity of the molecules evaluated in our study to HER2 allows these compounds to be evaluated as potential therapeutic agents that may be alternatives to existing treatments. RMSD analyses revealed that Prunetin showed lower mobility at the binding site compared to the other two compounds and formed a relatively more stable complex. Axitinib has higher RMSF values, indicating that it shows more structural flexibility at the binding site. On the other hand, prunetin was found to form the most compact complex structure, indicating stronger molecule-protein interactions and a more stable binding structure. The binding free energy analysis (MM-PBSA) calculations showed that axitinib had a value of − 36.17 kcal/mol, silymarin −16.78 kcal/mol, and prunetin − 24.9 kcal/mol. The fact that axitinib has the lowest free energy indicates that it forms the strongest interaction with the HER2 protein and can be considered a potential HER2 inhibitor.

Furthermore, the ADMET analysis revealed that axitinib and prunetin possess favorable drug-like properties, complying with multiple drug similarity rules and demonstrating acceptable absorption, distribution, metabolism, and excretion profiles. Notably, silymarin, despite its high binding affinity, exhibited lower intestinal absorption and poor blood–brain barrier permeability, which could limit its clinical applicability. However, the absence of hepatotoxicity and AMES toxicity in silymarin and prunetin suggests a safer toxicity profile compared to axitinib with these toxicities.

Our findings emphasize the role of novel compounds in the development of new and less toxic therapeutic agents. The integration of molecular computational techniques with ADMET analysis provides a comprehensive approach to identifying and optimizing these compounds for clinical use. Our study suggests that axitinib, prunetin, and silymarin emerge as strong candidates for further investigation. The results of the study show that especially axitinib and prunetin can be evaluated as stronger drug candidates targeting HER2. Prunetin's more stable complex formation and lower toxicity favor it, but Axitinib's higher binding energy supports its efficacy as a HER2 inhibitor. Thus, these compounds represent a promising avenue for overcoming resistance mechanisms while minimizing side effects. Future in vitro and in vivo validation studies are required for evaluation of the biological effects of these compounds. In this context, the results of this study will guide future research and contribute to the development of more effective and specific treatment options for HER2-positive breast cancer patients.

Conclusıon

HER2-positive breast cancer is one of the aggressive malignancies with difficulties in treatment. Current HER2-targeted therapies have problems such as drug resistance and side effects. It is of critical importance to discover new HER2 inhibitors that can overcome these problems and enable the development of treatment options. In this context, the natural and synthetic compounds evaluated in our study have the potential to offer new and more effective therapeutic options in breast cancer treatment.

In conclusion, this study provides important findings for the discovery of potential new inhibitors for the treatment of HER2-positive breast cancer. Axitinib and Prunetin stand out as promising candidates as HER2 inhibitors with high binding affinity and stable complexes with HER2 protein. In particular, Prunetin's low toxicity profile and stable binding properties make this compound a safer drug candidate.

Lımıtatıon

The findings of the current study are based on computational analyses. Experimental validation is required to confirm the pharmacological effects of the molecules in the study.

Author contributions

C.Y. study design, data analysis of docking/dynamic studies, wrote the main manuscript text, editing S.E.A. study design, data analysis, editing S.E.A. dynamic and energy binding analysis, editing H.O. docking analysis E.B. dynamic and energy binding analysis N.C. data analysis, editing All authors reviewed the manuscript.

Funding

No funding was utilized in this study.

Data availability

The datasets generated during and/or analysed during the current study are available from the corresponding author on reasonable request.

Declarations

Ethical approval and consent to particpate

No ethics committee approval is required for this study.

Consent for publication

Not applicable.

Competing interests

The authors declare no competing interests.