Introduction

Cancer is the world’s most prevalent cause of morbidity and death. In terms of mortality, cancer has become the second most prevalent cause of illness globally, after cardiovascular disorders1. Lung, prostate, and breast cancer (BC) are the three cancer types that are widespread in various geographical areas. Chemotherapy is still the first-line treatment for cancer, despite the many therapeutic options that have been proposed. A modeling study published in “The Lancet Oncology” predicts a roughly 53% rise in the number of patients requiring chemotherapy, from 9.8 million in 2018 to 15 million in 20402. These circumstances make it evident that the need for chemotherapeutic medications is rising, so an important goal should be the development of novel scaffolds with a main focus on cancer3,4. Two common chemotherapeutic medications used to treat these conditions are tamoxifen and cisplatin. However, after being employed in therapy, these medications show significant toxicity.

BC has become a prevalent root of cancer-related death among women globally5. According to the American Cancer Society’s most recent data, approximately one in eight (12%) American women may experience invasive BC at a specific stage in their life6. Based on its molecular features, BC is divided into three main types: triple-negative BC (ER−, PR−, and HER2−), hormone-related BC (progesterone receptor (PR+) or estrogen receptor (ER+), and human epidermal growth factor receptor 2 (HER2+)- related BC7. The therapy method is determined by the type of BC. Hormone therapy is used to treat hormone-based BC. It acts by preventing the synthesis or function of hormones that stimulate the development of cancerous cells. Tamoxifen, a selective estrogen receptor modulator (SERM) which prevents the impacts of hormone on BC cells, as well as aromatase inhibitors, a type of medications which prevent estrogen’ production by blocking the aromatase enzyme, are two examples of common hormone therapy for BC8,9. The cell cycle is regulated primarily by HER2 inhibitors, which prevent the function of the HER2 protein, which is upregulated in certain types of BC, and CDK4/6 inhibitors, which impede the function of the cyclin-dependent kinases 4 and 610,11. The treatments for hormonal BC are chemotherapy, surgery, radiation therapy, and targeted medications that disrupt the signaling routes that foster the growth of tumor cells12. Treatment for TNBC, which contributes up to 10–15% of BC cases, is more difficult because there are few modes for specific therapy as TNBC cells do not overexpress HER2/neu, progesterone, or estrogen13.

Schiff bases have a vital role as ligands and are among the most commonly used organic compounds14,15,16. Schiff bases with the CO − NH − N = CH unit have long been the centre of interest due to their intriguing properties and potential medical applications. They can be categorized as a subclass of imines, which contain the imine (-C = N-) group in their structure that occurs through the reaction of an aldehyde or ketone with a primary amine under suitable conditions17,18. In this instance, the compounds demonstrated an enhanced ability to donate a pair of electrons through their azomethine group. Schiff bases exhibit a range of biological activities. In particular, they exhibit antidiabetic19, antifungal20, anticancer4,21,22, anti-inflammatory23, and antioxidant activities24. In addition, it has been suggested that the azomethine linkage might be responsible for their biological activities. The hydrazones are stable at neutral pH but are rapidly destroyed in the acidic environment around the tumor, and hence designing a pH-sensitive drug delivery system is an advantage. These characteristics have allowed hydrazones to be linked to many drugs for many applications25. Recently, many compounds containing this moiety have been reported, indicating that the use of this pharmacophore can result in high potential activities22,26.

The indole scaffold, which has a pyrrole structure parallel to benzene, is a principal structural nuclei in drug discovery because of its unique ability to mimic peptide structure and interact with enzymes27,28,29,30,31. Additionally, it has been reported by previous researchers that the activity of indole analogs with N substitution, such as phenyl, benzyl and sulphonyl substitution, has enhanced markedly32,33. The indole moiety is an excellent building block for manufacturing anticancer medications34. Recently, bis-indole alkaloids consisting of two indole moieties bound to a spacer through their third position were found to exhibit a wide spectrum of potent biological activities, including antifungal, antitumor, antiviral, antimicrobial, anti-inflammatory, and cytotoxic activities. The bis-indole alkaloids can bear either an acyclic chain or a six-membered carbocyclic or a five-membered heterocyclic ring or a six-membered heterocyclic ring between two indole rings35. The bis-indole drugs, such as 3,5-bis(3′-indolyl) triazinones36, 2,5-bis(indolyl)−1,3,4-oxadiazoles37, and 1,3,4-oxadiazole-linked bis-indole35 derivatives demonstrate strong inhibitory effects against a variety of tumor cell lines, including lung cancer, breast cancer, cervical cancer, colon cancer, and oral cancer.

Recent studies have highlighted sulfonyl-containing heterocyclic compounds as promising scaffolds for anticancer drug development due to their diverse biological activities, structural versatility, and favorable pharmacokinetic properties. Sulfonamide and sulfonate moieties, when integrated into heterocyclic backbones, have demonstrated enhanced cytotoxicity and selectivity against various cancer cell lines. For instance, benzothiazole-carbohydrazide sulfonate hybrids have shown potent anticancer effects, with compound 6i inducing the generation of reactive oxygen species (ROS), DNA fragmentation, and G₂ phase cell cycle arrest in MCF-7 cells, while exhibiting low toxicity to normal cells38. Acyl sulfonamide spirodienone (compound 4a) showed potent antiproliferative activity by inducing apoptosis and cell cycle arrest in MDA-MB-231 cancer cells, demonstrated low toxicity in mice, inhibited tumor growth in vivo, and acted as a potential MMP2 inhibitor, making it a promising lead for anticancer drug development39. Sulfonamide-pyrimidine-pyrazole hybrids (9, 19, MTT assay) were highly active against MDA-MB-231 breast cancer cells40.

Molecular hybridization is a commonly utilized approach for drug design and development that aims at the direct or indirect combination of two or more pharmacophores to synthesize a unique molecule. In this strategy, the primary focus is on enhancing the biological profile while minimizing the side effects. Based on these considerations, we aimed to synthesize a new series of N-substituted indole-based hydrazone derivatives 5(a-r) by coupling N-substituted indole with substituted hydrazides. Indole-based hydrazones have gained a lot of attention because of their diverse biological activities that render them anticancer agents41. Some previously reported compounds that exhibited anti-cancer activity are given in Fig. 1.

Fig. 1
figure 1

Previously reported hydrazones as anti BC agents.

Full size image

Result and discussion

Chemistry

Scheme 1 details the synthetic route of the newly synthesized compounds 5(a-r). N-tosyl indole-3-carbaldehyde based hydrazones were synthesized in two steps. In the first step, N-tosyl indole-3-carbaldehyde (3) was synthesized from commercially available indole-3-carbaldehyde (1). N tosylation of indole-3-carbaldehyde was carried out by refluxing indole-3-carbaldehyde with p-toluene sulphonyl chloride at 90–95 °C for 1 h in triethylamine solvent.

In the second step, equimolar concentration of N-tosyl indole-3-carbaldehyde (3) and substituted hydrazide 4(a-r) was refluxed in MeOH to synthesize hydrazone, and CH3COOH was used as a catalyst. The obtained product was filtered, washed, and dried to acquire the desired yield of hydrazones 5(a-r) in Scheme 1.

Different characterization techniques, such as FTIR, 1H NMR, 13C NMR spectroscopy, and mass spectrometry, were performed to confirm the structures of new compounds 5(a-r). The FTIR data showed that the NH bands in CONH were absorbed in the region of 3500–3200 cm−1. And aromatic absorption was seen in the region of 600–900 cm−1.

A singlet was observed in 1H NMR at 11.04–12.17 ppm, which was ascribed to the carbohydrazide’s NH group. An additional peak was seen in compounds 5(k-r), which indicates the presence of tautomers. And methyl protons of the tosyl group appeared at 2.29–2.33 ppm region. The remaining aromatic protons all appeared in the 7–8 ppm range.

For compounds 5(k-r), peaks in the 13C NMR spectroscopy data showed an increase in carbon atoms in contrast to the targeted synthesized compounds. Two peaks were seen in the C = O region at 160–210 ppm, indicating the presence of tautomerism.

In the HRMS spectra, molecular ion peaks denoted as [M + H]+, precisely corresponded to the molecular weight of the newly synthesized compounds.

Scheme 1
scheme 1

Synthetic route of N-tosyl indole-based hydrazones.

Full size image

Biological activity

A different concentration (6.5µM, 12.5µM, 25µM and 50µM) of the synthetic compounds 5(a-r) was utilized to examine against the MDA-MB-231 to determine the suppression of cancer cells proliferation. Simultaneously, MCF-10 A cell lines were maintained as a control in the study. The MTT [3-(4,5-dimethylthiazol-2-yl)−2,5-diphenyltetrazolium bromide] assay was utilized to evaluate the decline in viability of cancerous cells caused by cytotoxic drugs. The IC50 values, percent viability, and inhibition of compounds 5(a-r) for MDA-MB-231 are demonstrated in Table 1. To calculate the IC50 values and evaluate the dose-response, IBM SPSS Statistics 26 software was used. MTT assay results disclosed that all new compounds demonstrated potent inhibitory potential against MDA-MB-231 breast cancer (BC) cells. Compound 5p exhibited greater potency against the MDA-MB-231 TNBC cell line (IC50 = 12.2 ± 0.4µM) among all.

The non-malignant MCF-10 A cells were introduced to synthesized compounds at various concentrations (6.5µM, 12.5µM, 25µM, and 50µM) in the same way as the malignant cells to ascertain whether the compounds’ cytotoxicity was selective for malignant cells compared to non-malignant cells. Following the MTT assay, the findings of percent viability and inhibition for MCF-10 A cell lines by compounds 5(a-r) are shown in Table 1. The outcomes demonstrated that these cells were less prone to the compounds’ actions, especially to compound 5p which displayed greater cell death in BC cells.

Data from current research demonstrated that the aggressive TNBC MDA-MB-231 cells performed more effectively against most of the compounds and displayed more cytotoxic effects. The reduced cytotoxicity was seen when non-malignant MCF-10 A cells were introduced to compounds 5(a-r) indicating that these new compounds will provide a promising treatment for BC patients.

Table 1 Percent viability and IC50 values of new compounds on MDA-MB-231 and MCF-10 A.
Full size table

Structure-Activity relationship

SAR is mainly influenced by different substitutions on the hydrazide’ R group. We employed a variety of aromatic rings in the current research, including pyridyl, naphthyl, furan, phenyl, and benzyl at varying positions, as well as different substituents attached to aromatic rings.

The trend of monocyclic, bicyclic, and heterocyclic is as follows: 5p (naphthyl, IC50 value = 18.1 ± 0.2 µM) > 5q (3-hydroxy-2-naphthoic, IC50 value = 16.4 ± 0.2 µM) > 5 g (nicotinic, IC50 value = 18.1 ± 0.2 µM) > 5e (isonicotinic, IC50 value = 23.4 ± 0.6 µM) > 5d (phenyl, IC50 value = 24.1 ± 0.4 µM) > 5 h (furan, IC50 value = 29.6 ± 0.2 µM). Correlating the inhibitory potency of different rings linked to the hydrazide, compound 5 g with the nicotinic group exhibited promising activity, having an IC50 = 18.1 ± 0.2 µM and is the third potent compound among all. Compound 5d with a phenyl group showed significant activity, having an IC50 = 24.1 ± 0.4 µM. Compound 5 h with furan group demonstrated moderate inhibitory potency, having an IC50 = 29.6 ± 0.2 µM. Compound 5p, which has naphthyl substitution linked to the hydrazide, is the most intriguing candidate in the complete pool of compounds exhibiting an IC50 value of 12.2 ± 0.4 µM, leading to compound 5q with 3-hydroxy naphthoic substitution attached at the ortho position of the hydrazide demonstrated outstanding inhibitory potency with an IC50 = 16.4 ± 0.2 µM and ranks as series’ second most potent compound. The subsequent observation demonstrated that the inhibitory potency is significantly influenced by the size of ring, as bicyclic ring compounds are more potent than others in the series. Compound 5e, containing an iso-nicotinic ring showed lower inhibition potential as compared to compound 5 g with a nicotinic ring. Compound 5c with 4-methoxyphenyl substitution is the second least potent compound of the series.

Compound 5r with a 2-methylphenyl group demonstrated good inhibition, possessing an IC50 value of 27.2 ± 0.6 µM.

Comparing the inhibition potential of halogen-containing compounds, the ranking order of halogen substitutions is as follows; 5i (4-(trifluoromethyl)phenyl, IC50 value = 19.2 ± 0.4 µM) > 5n (2-(trifluoromethyl)benzyl, IC50 value = 20.1 ± 0.4 µM) > 5f (3-fluoro-4-methylphenyl, IC50 value = 21.3 ± 0.2 µM) > 5j (4-bromophenyl, IC50 value = 22.4 ± 0.2 µM) > 5b (3-chlorophenyl, IC50 value = 25.1 ± 0.4 µM) > 5 m (2,4-difluorobenzyl, IC50 value = 26.4 ± 0.2 µM) > 5k (diclofenac, IC50 value = 27.8 ± 0.1 µM) > 5 L (2-bromophenyl, IC50 value = 29.4 ± 0.6 µM) > 5o (3,5-difluorobenzyl, IC50 value = 32.1 ± 0.6 µM). Remarkably, Compounds 5i and 5n, which had trifluoromethyl groups at 4 and 2 positions of phenyl and benzyl ring, respectively, exhibited marvellous inhibitory potential, possessing an IC50 = 19.2 ± 0.4 µM and IC50 = 20.1 ± 0.4 µM, respectively. The similar behavior was shown by compounds 5j and 5 L containing 4-bromophenyl and 2-bromophenyl groups, respectively. Compound 5j with the 4-bromophenyl group showed significantly higher inhibition potential as compared to compound 5 L with the 2-bromophenyl group. It demonstrated the significance of para-substitution, which is advantageous for suppressing breast cancer (BC) cells. 5 m with 2,4-diflourobenzyl group demonstrated significant inhibition possessing an IC50 value of 26.4 ± 0.2 µM in comparison with compound 5o with 3,5-diflourobenzyl group showed much lower potency, possessing an IC50 value of 32.1 ± 0.6 µM. Compound 5b with 3-chlorophenyl substitution exhibited moderate potency with an IC50 = 25.1 ± 0.4 µM.

Compound 5k demonstrated another decline in inhibitory potential when chloro substitutions were introduced at positions 2 and 6 of the phenyl ring. Comparing compound 5 m with difluoro substituents to compound 5k with dichloro substituents shows a sharp drop in the inhibition potential, indicating that high electronegativity promotes the suppression of BC cells. Compound 5f, which has a 3-flouro-4-methylphenyl group, demonstrated an impressive rise in inhibition activity with an IC50 = 21.3 ± 0.2 µM.

The inhibitory potential is considerably reduced in compound 5a when the nitro ring is substituted at the phenyl ring, and the least potent compound among all, with an IC50 = 38.6 ± 0.2 µM. This might be because of the steric hindrance when compounds interact with the cancer cells. Indole is a pharmacologically active entity against BC, and its activity is enhanced by further tosylation at its N-position. SAR of newly synthesized compounds 5(a-r) is illustrated in Fig. 2.

Fig. 2
figure 2

Pictorial representation of SAR.

Full size image

Molecular docking analysis

In the docking studies, we accessed the binding affinity of various synthetic derivatives that were effective in EGFR T790M/L858R mutants using experimental crystal structure 3w2s. results represented in Table 1S reflect a range of binding energies, attached in the supporting information, suggesting that compound 5q demonstrated the highest interaction with a docking score of −11.533 kcal/mol, indicating robust interaction with target protein. Binding energy suggests the efficacy of all compounds as inhibitors. The docking results highlight that hydrophobic interactions dominate, with crucial residues like LEU718A, VAL726A, and ALA743A frequently involved across multiple ligands that are important for sustaining the ligand’ structural integrity within the binding pocket. Moreover, hydrogen bonds (H-bonds) provide specificity and enhanced binding affinity like 5q with LYS745A at distances of 2.11 and 2.67 Å, contributing significantly to its high binding affinity. This specificity is crucial for the selectivity of the inhibitors against the mutant EGFR over the wild-type, potentially reducing off-target effects. The presence of π-stacking and π-cation interactions, like in 5j and 5 L, displays a significant role in orientation of the ligands in the binding site of the receptors. Supported by interaction, we can say that compounds with functional groups that compounds which form additional H-bonds or enhance hydrophobic interactions show the lowest binding energy. Compounds with polar substituents such as hydroxyl groups (in 5q, Fig. 3), fluorine atoms, or trifluoromethyl groups (in 5i and 5n) impacted the electron distribution and hydrophobic character influencing binding affinity. Bulky groups (e.g., CF3 in 5i) restrict or enhance the compound’s ability to fit or adapt to the binding site topology. It means Modifications that improve fit within the binding pocket without causing steric clashes to enhance efficacy. The variation in the side chains impacts not only the binding affinity but also the ligand’s orientation and interactions within the active site.

Fig. 3
figure 3

2D and 3D contacts of protein 3W2S with 5q.

Full size image

Binding free energy calculation

The approach to calculating binding free energies of protein-ligand complexes is the Molecular Mechanics combined with the Generalized Born Surface Area (MMGBSA), which was calculated using the Python script thermalmmgbsa.py using the last 50 frames of simulation trajectories with each step sampling size. MMGBSA (kcal/mol) were estimated by adding various energy modules like covalent, coulombic, Vander Waal, lipophilic solvation, etc., which were collectively considered.

The following equation employed to determine ΔGbind is:

$$\:{\varDelta\:G}_{bind}={\varDelta\:G}_{MM}+{\varDelta\:G}_{Solv}-{\varDelta\:G}_{SA}$$

Whereas:

  • ΔGbind indicates the binding free energy,

  • ΔGMM indicates the difference between the ligand-protein complexes’ free energies and the total energies of protein and ligand in isolated form,

  • ΔGSolv indicates the difference in the GSA solvation energies of the ligand-receptor complex and the sum of the solvation energies of the receptor and the ligand in the unbound state,

The difference between the ligand and protein surface area energies is represented by ΔGSA.

Molecular dynamics simulation

Molecular dynamics (MD) simulation and studies were carried out to ascertain the stability28,29,30,31,32 and confluence of 3W2S with 5q and W2R represented as 3W2S_5q and 3W2S_W2R respectively. Figure 4A represents a Root Mean Square Deviation (RMSD) analysis over a simulation period of 100 ns. Both systems exhibit a rapid increase in RMSD, indicating initial equilibration as the system adjust to simulation period. The complex 3W2S_5q (Red line) maintains a higher average RMSD 2.23 Å than 3W2S_W2R (Blue line) with average 2.15 Å, suggesting more conformational flexibility from initial structure. 3W2S_W2R exhibits slightly lower fluctuations, indicating greater structural stability. Overall, both systems achieve stability as RMSD values remain consistently below 3 Å. Figure 4B represents the Root Mean Square Fluctuation (RMSF) analysis over a simulation period. Both systems exhibit similar fluctuation profiles, residues show fluctuations below 2 Å, suggesting stability. A pronounced peak observed around residues 0, 162–164, 272 in 3W2S_5q complex (Red line) with an average value of RMSF 1.13 Å whereas in complex 3W2S_W2R (Blue line) exhibits RMSF more than 3 Å at residues 0–1, 306 with an average value 1.08 Å. Overall both systems exhibit fluctuations relatively low RMSF values suggesting rigidity and stability of complexes. Figure 4C depicts the Radius of Gyration (Rg) analyzed to investigate the compactness and structural stability of protein. The complex 3W2S_5q (Red line) shows significant fluctuations in the Rg values ranges between 19.86 Å and 25.49 Å, particularly after 20ns, indicating substantial structural stability or conformational flexibility. The complex 3W2S_W2R (Blue line) exhibits more consistent Rg Values with fluctuations limited to a narrow range of 20.18 Å − 22.03 Å, suggesting more stable structural conformation over time. Figure 4D illustrates the number of H-bonds formed in complexes. H-bonds exhibit essential role in retaining the stability and structural integrity of complex. The complex 3W2S_W2R (Blue line) maintains a consistently higher number of H-bonds during the simulation, fluctuating between 2 and 6. This suggests strong and stable intermolecular or intramolecular interactions. In contrast, the complex 3W2S_5q (Red line) exhibits a lower and more inconsistent number of H-bonds, fluctuating between 0 and 5, indicating weaker and less stable interactions. Figure 5 depicts the Solvent Accessible Surface Area (SASA) analysis, highlighting the conformational changes and stability of 5q and W2R when bound to protein. The lower SASA in the receptor bound systems suggests that ligand binding leads to a significant reduction in solvent exposed areas. The difference in SASA between the unbound and bound systems suggests that ligand binding induces structural changes, resulting in more compact receptor conformation and reduced solvent attainability.

Fig. 4
figure 4

(A) RMSD of 3W2S_5q (Red line) and 3W2S_W2R (Blue line) (B) RMSF of 3W2S_5q (Red line) and 3W2S_W2R (Blue line) (C) Rg of Cα backbone of 3W2S_5q (Red line) and 3W2S_W2R (Blue line) (D) H-bonds formation in 3W2S_5q (Red line) and 3W2S_W2R (Blue line).

Full size image
Fig. 5
figure 5

MD simulation study of 1000 Frame work of (A) SASA of 3W2S_5q (B) SASA of 3W2S_W2R.

Full size image

Protein ligand interactions between the complexes were monitored by simulation studies over the course of 100 ns summarised by stacked coloured bars, normalized over the course of trajectories which are categorizes as water bridges, hydrophobic interactions, Ionic interactions and H-bonds. If protein ligand complex has multiple contacts for particular type of interactions, then value remains ≥ 1. Figure 6A and B represents various bar graphs of various types of interaction fractions against residues present over the period of 100 ns. It reveals that high interaction fractions for protein-ligand complexes 3W2S_5q and 3W2S_W2R were made by hydrogen bonding, ionic interaction and water bridges.

Figure 7A illustrates the percentage of protein-ligand interactions for the 3W2S_5q complex. ARG748 (36%), LYS745 (94%), GLU746 (32%) and ASP855 (122%) interacts via forming water bridges. PHE723 (15%) and PHE856 (60%) interacts through hydrophobic interactions. ARG836 (12%) and LYS745 (51%) interacts through ionic interactions. GLY857 (74%), PHE856 (33%), and ASP855 (40%) interacts via hydrogen bonding. Figure 7B illustrates the percentage of protein-ligand interactions for the 3W2S_W2R complex. ASP855 (114%), LYS745 (174%), CYS775 (22%) and THR854 (166%) interacts via forming water bridges. PHE856 (21%) interact through hydrophobic interactions. GLY857 (84%), PHE856 (134%), LYS745 (38%), and MET793 (100%) interacts via hydrogen bonding.

Fig. 6
figure 6

Bar graph of interaction of protein and ligand (A) 3W2S_5q (B) 3W2S_W2R.

Full size image
Fig. 7
figure 7

Percent contacts of protein and ligand (A) 3W2S_5q (B) 3W2S_W2R.

Full size image

Molecular mechanics generalized born surface area (MM-GBSA) calculations

For complexes 3W2S_5q and 3W2S_W2R, the ΔGbind incorporating further contributing energy in the form of MM-GBSA was computed utilizing MD simulation trajectory. The findings in Table 2 indicate the binding free energies for complexes for complexes 3W2S_5q and 3W2S_W2R. The binding free energy ΔGbind of 3W2S_5q is −63.41 Kcal/mol, while 3W2S_W2R exhibits a significantly more favourable binding energy ΔGbind of −96.17 Kcal/mol. The significantly lower ΔGbind of 3W2S_W2R relative to 3W2S_5q suggests that the former complex is thermodynamically more stable. The much stronger coulombic and hydrogen bonding contribution in 3W2S_W2R indicate the formation of robust electrostatic interactions and a well-established hydrogen bonding network. The greater ΔGbindLipo for 3W2S_W2R reflects a highly favourable hydrophobic interactions and stronger Vander Waals highlights steric complementarily and optimized packing which stabilizes the binding as compared to 3W2S_5q. Despites a higher desolvation penalty for 3W2S_W2R, the favourable coulombic, lipophilic and Vander Waals contributions compensate for energy, resulting in lower ΔGbind. The slightly more favourable ΔGbindPacking in 3W2S_5q suggests that it has steric interactions as compared to 3W2S_W2R.

Table 2 ΔGbind components for the 3W2S_5q and 3W2S_W2R determined by MM-GBSA.
Full size table

QSAR analysis

“PIC50=5.84 + 0.96 * ringC_O_3Bc + 0.167 * sp2C_notringC_3B + −0.061 * fsp3Csp2C3B”----------------------QSAR Model 1.

(ringC_O_3Bc: total of ring carbon atom’ partial charges within three bonds from the oxygen atoms; sp2C_notringC_3B: presence of non-ring carbon atom within three bonds from the sp2 hybridized carbon atoms; fsp3Csp2C3B: frequency of presence of sp2 hybridized carbon atom at exact three bonds from the sp3 hybridized carbon atoms).

The QSAR plots for the correlation between experimental and observed activities are also shown in Fig. 8. Other necessary plots are given in the supporting information.

Fig. 8
figure 8

Experimental Vs predicted activities for model-1 obtained with QSAR equations.

Full size image

QSAR model description and validation

The descriptor ringC_O_3Bc computes the total of ring carbon atoms’ partial charges located inside three bonds of oxygen atoms, with a positive coefficient (+ 0.96), indicating that elevated values augment pIC50. This indicates that oxygen-adjacent ring carbons enhance biological activity, perhaps owing to their capacity for electrical interactions or hydrogen bonding with the biological target. sp2C_notringC_3B quantifies the prevalence of non-ring sp2-hybridized carbon atoms within three bonds of other sp2 carbons, exhibiting a positive coefficient (+ 0.167), indicating that planar or conjugated systems augment receptor binding via enhanced molecular interactions. Conversely, fsp3Csp2C3B, denoting the presence of sp2-hybridized carbons precisely three bonds distant from sp3-hybridized carbons, exhibits a negative coefficient (−0.061), indicating that these spatial configurations diminish biological activity, likely attributable to steric hindrance or diminished structural compatibility with the target. This model highlights the interaction of electrical effects (via partial charges), molecular planarity (by sp² hybridization), and steric factors in influencing biological potency. The substantial influence of ringC_O_3Bc underscores the significance of electronegative areas in binding interactions, while the negligible positive contribution of sp2C_notringC_3B reinforces the advantageous function of conjugated systems. In contrast, the adverse impact of fsp3Csp2C3B underscores the need to reduce steric hindrances in molecular design. The validation of Model 1 QSAR was as per the Table 3. This QSAR model offers essential insights for optimizing molecular characteristics to improve biological activity, illustrating the efficacy of descriptor-based methodologies in rational drug design.

Table 3 QSAR model validation parameters.
Full size table

ADME study 5(a-r)

The absorption, distribution, metabolism, and excretion (ADME) characteristics of the synthesized derivatives 5(a-r) demonstrate a consistent drug-like physicochemical profile that aligns systematically with Lipinski’s rule-of-five (Ro5) criteria, though selective deviations are noted, indicating potential avenues for optimization. The molecular weight (MW) range of 407.44–591.51 g/mol indicates that the majority of compounds fall within the Ro5-compliant domain (16/18 ≤ 500 Da). Notably, compounds 5k and 5q slightly surpass the threshold, implying that structural elaboration at these specific loci significantly contributes to the overall molecular bulk. The quantification of hydrogen-bond donors (HBD) and acceptors (HBA) is rigorously maintained within specific limits, with values ranging from 1 to 2 for HBD and 4 to 7 for HBA. This adherence to established criteria, specifically the Ro5 thresholds (HBD ≤ 5, HBA ≤ 10), underscores the likelihood of enhanced membrane permeability. Lipophilicity, characterized by consensus cLogP values ranging from 3.25 to 6.13, with a mean of approximately 4.37, predominantly resides within the optimal lipophilic efficiency window. Nevertheless, compound 5k surpasses the Ro5 threshold (cLogP > 5), suggesting a potential liability concerning suboptimal aqueous solubility and metabolic instability. The topological polar surface area (TPSA) ranges from 88.91 to 134.73 Ų, while the number of rotatable bonds (RB) is between 6 and 9, both of which are below the established thresholds set by Veber (TPSA < 140 Ų; RB < 10). This observation highlights the theoretical capacity for sufficient oral bioavailability. An analysis based on Lipinski’s rule of five revealed that seven compounds exhibited full compliance with no violations, ten compounds presented a single violation—predominantly influenced by logP values—and one compound (5k) was identified with two violations, thereby categorizing it as the least favorable candidate in terms of developability. Significantly, the anticipated gastrointestinal absorption was categorized as High for 13 derivatives and Low for five (5a, 5i, 5k, 5n, 5q), aligning with the interaction of increased TPSA and heightened lipophilicity in these less advantageous instances. None of the molecules demonstrated the anticipated ability to penetrate the blood–brain barrier, a characteristic that is beneficial in preventing central nervous system-related off-target liabilities for applications outside of neurotherapeutics. Furthermore, all derivatives were forecasted to be non-substrates of P-glycoprotein, thereby reducing the potential risks associated with efflux-mediated loss of oral bioavailability. The anticipated aqueous solubility exhibited a range from “Moderately soluble” for 11 compounds to “Poorly soluble” for 7 compounds, illustrating the intrinsic lipophilicity–solubility trade-off associated with the scaffold. The skin permeation coefficients (log Kp ranging from − 6.4 to − 4.8 cm/s) consistently demonstrated a low level of dermal permeability. The analysis of bioavailability scores revealed a value of 0.55 for the majority of derivatives, while a score of 0.17 was noted for the 5k compound. This data further substantiates the enhanced oral drug-likeness exhibited by most analogues. In conjunction with these data, the SwissADME BOILED-Egg predictive model (WLOGP vs. TPSA) revealed that all compounds were situated within the “white” region, signifying a high likelihood of passive gastrointestinal absorption (HIA). Furthermore, none of the compounds entered the “yellow yolk” region, thereby affirming the predicted lack of permeability across the blood–brain barrier (see Fig. 9).

Fig. 9
figure 9

BOILED-Egg model illustrating WLOGP vs. TPSA for candidate molecules. Most compounds fall within the acceptable HIA/BBB regions, except Molecule 1 which lies outside the range.

Full size image

Significantly, a particular compound (Molecule 1) was observed at the boundary of the absorption ellipse, indicating its marginal interaction between topological polar surface area (TPSA) and lipophilicity. In contrast, the majority of compounds were closely clustered within the designated absorption zone. All molecules were categorized as P-glycoprotein non-substrates (PGP–), thereby supporting the hypothesis of unobstructed intestinal absorption devoid of efflux-related liabilities. This comprehensive analysis of physicochemical properties and graphical modeling substantiates the hypothesis that compounds within the series 5(a-r) fall within the parameters of oral drug-likeness. The optimization strategies should primarily focus on adjusting lipophilicity and molecular size for outliers like 5k, while preserving the advantageous balance of hydrogen-bonding and topological polar surface area (TPSA) that facilitates absorption (See supplementary material for ADME parameters).

Conclusion

In conclusion, a new series of N-tosyl indole-3-carbaldehyde based hydrazones 5(a-r) was assessed for their anti-cancer activity against the MDA-MB-231 breast cancer (BC) cell line. All the new compounds showed strong anti BC activity on the MDA-MB-231. Compound 5p with a naphthyl group exhibited promising potential against the MDA-MB-231 BC cell line (IC50 = 12.2 ± 0.4 µM). SAR research revealed that employing the bicyclic groups improved the effectiveness of inhibiting cancer cell types. Provided that the recently investigated hydrazone derivatives may serve as a viable therapeutic candidate for BC, the findings of this research motivate us to carry out further structural modification and anticancer activity screening in the future.

Experimental

General

All of the starting materials, including indole, used to synthesize indole-based hydrazones, were purchased from Sigma Aldrich. EtOH, Et3N, MeOH, CH3COOH, petroleum ether, and EtOAc were utilized as original and purchased from Merck. The progress and completion of the reaction were monitored by aluminum-backed silica gel plates. A Bruker Ascend 600 MHz NMR spectrometer was employed to obtain 1H and 13C NMR spectra in DMSO-d6 at 25 °C (600 MHz for 1H and 151 MHz for 13C). To illustrate signal multiplicity, NMR results were displayed as ppm for chemical shifts and Hertz (Hz) for coupling constants (J). The mass spectrum was recorded on QTOF HRMS 6530 With 1260 HPLC.

General procedure for synthesis of N-tosyl indole-3-carbaldehyde (3)

N-tosyl indole-3-carbaldehyde was obtained by refluxing indole-3-carbaldehyde (1.5 g, 10.3mmol) and p-toluene sulphonyl chloride (2.99 g, 15.7mmol) in triethylamine (25 ml). The resultant mixture was refluxed at 90–95 °C for 1 h while constantly being stirred. The progression of reaction was checked by TLC. After that, the resulting residue was cooled and then poured into cold H2O that had ice fragments in it. Subsequently, the product was filtered, washed, and then left to dry.

General procedure for synthesis of N-tosyl indole-3-carbaldehyde based hydrazones 5(a-r)

15 ml of methanol and 0.33mmol (0.1 g) of N-tosyl indole-3-carbaldehyde were added in a round bottom flask. Further, the reaction mixture was stirred until all amount of the N-tosyl indole-3-carbaldehyde had been dissolved. 0.33mmol of substituted hydrazide was added in the resulting mixture and 3–4 drops of CH3COOH were used as a catalyst. The resultant mixture was then refluxed for about 4 to 6 h, and precipitates were formed. EtOAc and petroleum ether (1:3) were used to examine the progression of the reaction on TLC. Then, the resultant mixture was cooled to 25 °C. The product was filtered, and methanol was used to wash off the impurities. The yield was determined after drying and weighing the product. The N-tosyl indole-3-carbaldehyde based hydrazones 5(a-r) were purified by crystallization.

(E)−4-Nitro-N‘-[(1-tosyl-1H-indol-3-yl)methylene]benzohydrazide (5a)

Yield = 95, m.p = 256–258 °C, off-white solid, IR(KBr)cm−1; 3118, 2310, 1681, 1618, 1447, 1362 δ H (600 MHz, DMSO-d6) 11.04 (1 H, s), 8.96 (1 H, s), 8.51 (1 H, s), 8.39 (3 H, dd, J = 8.4 Hz, 6.6 Hz), 8.17 (2 H, d, J = 8.5 Hz), 7.99 (1 H, d, J = 8.3 Hz), 7.94 (2 H, d, J = 8.1 Hz), 7.49–7.45 (1 H, m), 7.41 (3 H, dd, J = 8.0 Hz, 5.7 Hz), 2.32 (3 H, s); 13C NMR (151 MHz, DMSO) δ 164.75, 156.45, 149.98, 146.56, 138.38, 135.29, 134.08, 133.10, 130.95, 129.52, 127.47, 127.43, 126.43, 124.92, 124.31,123.65, 118.14, 113.71, 21.52. QTOF HRMS (m/z): [M + H]+, calcd: 463.1076, found: 463.1076.

(E)−3-Chloro-N’-[(1-tosyl-1H-indol-3-yl)methylene]benzohydrazide (5b)

%Yield = 60, m.p = 201–203 °C, white solid, IR(KBr)cm−1; 3056, 2324, 1687, 1607, 700 δ H (600 MHz, DMSO-d6) 12.03 (1 H, s), 8.62 (1 H, s), 8.46–8.39 (2 H, m), 8.01–7.87 (5 H, m), 7.68 (1 H, dd, J = 7.9 Hz, 2.2 Hz), 7.58 (1 H, t, J = 7.9 Hz), 7.42 (4 H, m), 2.32 (3 H, s); 13C NMR (151 MHz, DMSO) δ 162.06, 146.38, 143.15, 135.96, 135.25, 134.23, 133.74, 132.01, 130.99, 130.88, 130.70, 127.81, 127.44, 127.35, 126.97, 126.32, 124.76, 123.81, 118.59, 113.59, 21.52. QTOF HRMS (m/z): [M + H]+, calcd: 452.0835, found: 452.2971.

(E)−4-Methoxy-N’-[(1-tosyl-1 H-indol-3-yl)methylene]benzohydrazide (5c)

%Yield = 90, m.p = 202–204 °C, off-white solid, IR(KBr)cm−1: 3056, 2324, 1687, 1607, 1256, 1168, δ H (600 MHz, DMSO-d6) 11.83 (1 H, s), 8.60 (1 H, s), 8.43 (1 H, d, J = 7.8 Hz), 8.35 (1 H, s), 8.01–7.88 (5 H, m), 7.41 (4 H, m), 7.08 (2 H, d, J = 8.3 Hz), 3.84 (3 H, s), 2.32 (3 H, s); 13C NMR (151 MHz, DMSO) δ 162.88, 162.48, 146.34, 141.94, 135.27, 134.25, 130.86, 130.15, 130.01, 127.54, 127.33, 126.27, 125.96, 124.69, 123.86, 118.87, 114.19, 113.58, 55.91, 21.52. QTOF HRMS (m/z): [M + H]+, calcd: 448.1331, found: 448.2462.

(E)-N’-[(1-Tosyl-1H-indol-3-yl)methylene]benzohydrazide (5d)42

%Yield = 70, m.p = 219–221 °C, off-white solid, IR(KBr)cm−1: 3120, 2323, 1652, 1601, δ H (400 MHz, DMSO-d6) 11.93 (1 H, s), 8.61 (1 H, s), 8.43 (1 H, d, J = 7.7 Hz), 8.36 (1 H, s), 8.00-7.88 (5 H, m), 7.62–7.50 (3 H, m), 7.39 (4 H, m), 2.31 (3 H, s); 13C NMR (101 MHz, DMSO) δ 163.48, 146.35, 142.58, 135.25, 134.23, 133.95, 132.17, 130.86, 130.41, 128.93, 128.10, 127.49, 127.33, 126.29, 124.72, 123.83, 118.73, 113.57, 21.51. QTOF HRMS (m/z): [M + H]+, calcd: 418.1225, found: 418.2466.

E)-N‘-[(1-Tosyl-1H-indol-3-yl)methylene]isonicotinohydrazide (5e)

%Yield = 60, m.p = 206–208 °C, off-white solid, IR(KBr)cm−1: 3101, 2356, 1733, 1651, 1603, δ H (600 MHz, DMSO-d6) 12.17 (1 H, s), 8.84–8.77 (2 H, m), 8.63 (1 H, s), 8.47–8.39 (2 H, m), 7.95 (3 H, m), 7.87–7.83 (2 H, m), 7.48–7.36 (4 H, m), 2.32 (3 H, s); 13C NMR (151 MHz, DMSO) δ 161.98, 150.80, 150.03, 146.41, 143.85, 141.02, 135.24, 134.21, 131.03, 130.89, 127.37, 126.35, 124.81, 123.76, 122.04, 118.42, 113.61, 21.53. QTOF HRMS (m/z): [M + H]+, calcd: 419.1177, found: 419.2231.

(E)−3-Fluoro-4-methyl-N‘[(1-tosyl-1H-indol-3-yl)methylene]benzohydrazide (5f)

%Yield = 93, m.p = 232–234 °C, white solid, IR(KBr)cm−1: 3351, 2348, 1678, 1549, 700, δ H (600 MHz, DMSO-d6) 11.94 (1 H, s), 8.61 (1 H, s), 8.47–8.34 (2 H, m), 7.95 (3 H, m), 7.71 (2 H, d, J = 9.1 Hz), 7.42 (5 H, m), 2.32 (6 H, s); 13C NMR (151 MHz, DMSO) δ 162.05, 161.62, 146.37, 142.83, 135.25, 134.23, 133.47, 132.24, 130.88, 130.56, 128.87, 127.47, 127.35, 126.30, 124.74, 123.82, 118.66, 114.59, 114.43, 113.59, 21.52, 14.73. QTOF HRMS (m/z): [M + H]+, calcd: 450.1287, found: 450.2259.

(E)-N’-[(1-Tosyl-1H-indol-3-yl)methylene]nicotinohydrazide (5 g)

%Yield = 72, m.p = 179–181 °C, off-white solid, IR(KBr)cm−1: 3123, 2356, 1620, δ H (600 MHz, DMSO-d6) 12.11 (1 H, s), 9.09–8.95 (1 H, d, J = 2.7 Hz), 8.60–8.5 (1 H, m), 8.42–8.26 (3 H, m), 8.28 (1 H, dt, J = 7.9 Hz, 2.0 Hz), 8.02–7.88 (3 H, m), 7.61–7.56 (1 H, m), 7.49–7.36 (4 H, m), 2.32 (3 H, s); 13C NMR (151 MHz, DMSO) δ 162.06, 152.72, 149.09, 146.40, 143.25, 135.94, 134.22, 130.96, 130.89, 129.73, 127.46, 127.44, 127.42, 126.34, 124.79, 124.08, 123.78, 118.52, 113.60, 21.53. QTOF HRMS (m/z): [M + H]+, calcd: 419.1177, found: 419.2590.

(E)-N’-[(1-Tosyl-1H-indol-3-yl)methylene]furan-2-carbohydrazide (5 h)

%Yield = 94, m.p = 169–171 °C, light brown solid, IR(KBr)cm−1: 3125, 2357, 1658, 1604, 1232, 1170, δ H (600 MHz, DMSO-d6) 11.95 (1 H, s), 8.62 (1 H, s), 8.39 (1 H, d, J = 7.8 Hz), 8.36 (1 H, s), 7.99–7.89 (4 H, m), 7.46–7.35 (4 H, m), 7.31 (1 H, d, J = 3.5 Hz), 6.71 (1 H, dd, J = 3.5 Hz, 1.8 Hz), 2.31 (3 H, s); 13C NMR (151 MHz, DMSO) δ 154.60, 147.18, 146.36, 146.27, 142.70, 135.24, 134.23, 130.86, 130.42, 127.43, 127.35, 126.30, 124.74, 123.77, 118.65, 115.31, 113.59, 112.53, 21.51. QTOF HRMS (m/z): [M + H]+, calcd: 408.1018, found: 408.2236.

(E)-N’-[(1-Tosyl-1H-indol-3-yl)methylene]−4-(trifluoromethyl)benzohydrazide (5i)

%Yield = 90, m.p = 244–246 °C, white solid, IR(KBr)cm−1: 3120, 2356, 1653, 1609, 710, δ H (600 MHz, DMSO-d6) 12.14 (1 H, s), 8.64 (1 H, s), 8.43 (1 H, s), 8.14 (2 H, d, J = 8.1 Hz), 7.97 (1 H, d, J = 8.3 Hz), 7.93 (4 H, d, J = 8.4 Hz), 7.88 (1 H, d, J = 8.1 Hz), 7.47–7.36 (4 H, m), 2.32 (3 H, s); 13C NMR (151 MHz, DMSO) δ 162.33, 146.38, 143.45, 137.79, 135.26, 134.24, 132.07, 131.86, 130.88, 129.81, 127.43, 127.36, 126.33, 125.99, 125.92, 124.78, 123.80, 118.54, 113.6, 21.52. QTOF HRMS (m/z): [M + H]+, calcd: 486.1099, found: 486.2564.

(E)−4-Bromo-N’-[(1-tosyl-1 H-indol-3-yl)methylene]benzohydrazide (5j)

%Yield = 70, m.p = 222–224 °C, light brown solid, IR(KBr)cm−1: 3169, 2355, 1649, 1601, 665 δ H (600 MHz, DMSO-d6) 12.01 (1 H, s), 8.61 (1 H, s), 8.41 (2 H, m), 7.97 (1 H, d, J = 8.3 Hz), 7.93 (2 H, d, J = 8.1 Hz), 7.89 (2 H, d, J = 8.4 Hz), 7.80–7.74 (2 H, m), 7.48–7.36 (4 H, m), 2.32 (3 H, s); 13C NMR (151 MHz, DMSO) δ 162.52, 146.37, 142.95, 135.25, 134.24, 133.01, 131.99, 130.88, 130.63, 130.22, 127.45, 127.35, 126.31, 125.96, 124.75, 123.81, 118.63, 113.59, 21.53. QTOF HRMS (m/z): [M + H]+, calcd: 496.0330, found: 496.1237.

(E)−2-((2,6-Dichlorophenyl)amino)-N’-[(1-tosyl-1H-indol-3-yl)methylene]benzohydrazide (5k)

%Yield = 95, m.p = 270–272 °C, off-white solid, IR(KBr)cm−1: 3452, 1657, 1274, 1141, 974, 757, 569, δ H (600 MHz, DMSO-d6) 11.86 (1 H, s), 11.70 (1 H, s), 8.44–8.34 (1 H, m), 8.32–8.26 (1 H, m), 8.04–7.88 (3 H, m), 7.68 (1 H, s), 7.53 (1 H, d, J = 8.1 Hz), 7.47 (1 H, d, J = 8.1 Hz), 7.45–7.38 (3 H, m), 7.35 (1 H, td, J = 7.6 Hz, 2.5 Hz), 7.26 (1 H, m), 7.16 (1 H, dt, J = 23.0 Hz, 8.1 Hz), 7.11–7.01 (1 H, m), 6.81 (1 H, t, J = 7.4 Hz), 6.31 (1 H, m), 4.18 (1 H, s), 3.73 (1 H, s), 2.32 (3 H, d, J = 5.1 Hz); 13C NMR (151 MHz, DMSO) δ 168.02, 146.39, 146.36, 143.60, 142.47, 139.49, 137.57, 135.28, 134.20, 131.22, 130.88, 130.38, 129.69, 127.95, 127.74, 127.38, 126.27, 125.88, 125.29, 124.89, 124.73, 123.57, 118.31, 113.57, 35.96, 21.52. QTOF HRMS (m/z): [M + H]+, calcd: 591.1024, found: 591.3970.

(E)−2-Bromo-N’-[(1-tosyl-1H-indol-3-yl)methylene]benzohydrazide (5 L)

%Yield = 70, m.p = 211–213 °C, off-white solid, IR(KBr)cm−1: 3068, 2178, 1657, 1593, 670, δ H (600 MHz, DMSO-d6) 12.05 (1 H, s), 8.45 (1 H, s), 8.43–8.36 (1 H, m), 8.23 (2 H, m), 7.95 (2 H, m), Hz), 7.73 (2 H, t, J = 8.5 Hz), 7.57 (1 H, dd, J = 7.5 Hz, 1.8 Hz), 7.50 (2 H, q, J = 7.5 Hz), 7.28 (1 H, t, J = 7.6 Hz), 7.00 (1 H, d, J = 7.9 Hz), 6.88 (1 H, t, J = 7.6 Hz), 2.32 (3 H, s); 13C NMR (151 MHz, DMSO) δ 169.96, 146.37, 142.91, 138.71, 137.97, 135.25, 134.20, 133.27, 131.94, 130.87, 130.30, 128.74, 128.22, 127.38, 127.27, 126.34, 124.79, 119.96, 118.44, 113.61, 21.53. QTOF HRMS (m/z): [M + H]+, calcd: 496.0330, found: 498.3337.

(E)−2-(2,4-Difluorophenyl)-N’-[(1-tosyl-1H-indol-3-yl)methylene]acetohydrazide (5 m)

%Yield = 90, m.p = 231–233 °C, white solid, IR(KBr)cm−1: 3072, 2355, 1653, 1607, 651, δ H (600 MHz, DMSO-d6) 11.70 (1 H, s), 8.42–8.12 (3 H, m), 8.03–7.84 (3 H, m), 7.52–7.29 (5 H, m), 7.21 (1 H, td, J = 9.7 Hz, 2.6 Hz), 7.05 (1 H, m), 4.07 (1 H, s), 3.63 (1 H, s), 2.32 (3 H, s); 13C NMR (151 MHz, DMSO) δ 165.79, 162.53, 160.91, 160.42, 146.35, 141.49, 135.29, 134.22, 133.54, 133.48, 130.86, 130.52, 127.42, 127.31, 126.26, 124.86, 123.66, 119.63, 118.53, 113.57, 34.08, 21.51. QTOF HRMS (m/z): [M + H]+, calcd: 468.1193, found: 450.2145.

(E)-N’-[(1-Tosyl-1H-indol-3-yl)methylene]−2-(2-(trifluoromethyl)phenyl)acetohydrazide (5n)

%Yield = 94, m.p = 237–239 °C, white solid, IR(KBr)cm−1: 3079, 2355, 1666, 1599, 713, δ H (600 MHz, DMSO-d6) 11.91 (1 H, s), 8.33–8.14 (3 H, m), 7.99–7.89 (3 H, m), 7.72 (1 H, d, J = 7.9 Hz), 7.68–7.60 (1 H, m), 7.56–7.45 (2 H, m), 7.41 (3 H, dd, J = 11.2 Hz, 8.2 Hz), 7.32 (1 H, q, J = 8.1 Hz), 4.28 (1 H, s), 3.83 (1 H, s), 2.32 (3 H, s); 13C NMR (151 MHz, DMSO) δ 165.89, 146.35, 138.49, 135.29, 134.50, 134.19, 133.72, 132.72, 132.66, 130.86, 130.54, 128.07, 127.80, 127.70, 127.44, 126.27, 126.07, 124.74, 123.66, 118.57, 113.57, 36.55, 21.51. QTOF HRMS (m/z): [M + H]+, calcd: 500.1255, found: 500.3520.

(E)−2-(3,5-Difluorophenyl)-N’-[(1-tosyl-1H-indol-3-yl)methylene]acetohydrazide (5o)

%Yield = 90, m.p = 245–247 °C, white solid, IR(KBr)cm−1: 3081, 2356, 1688, 1593, 714, δ H (600 MHz, DMSO-d6) 11.70 (1 H, s), 8.32–8.18 (3 H, m), 7.99–7.88 (3 H, m), 7.45–7.31 (4 H, m), 7.17–7.01 (3 H, m), 4.10 (1 H, s), 3.63 (1 H, s), 2.32 (3 H, s); 13C NMR (151 MHz, DMSO) δ 165.06, 163.42, 161.80, 146.37, 141.79, 140.62, 138.73, 135.26, 134.21, 130.87, 130.56, 127.39, 127.31, 126.27, 124.84, 123.64, 118.47, 113.57, 38.99, 21.52. QTOF HRMS (m/z): [M + H]+, calcd: 468.1193, found: 450.2908.

(E)-N’-[(1-Tosyl-1H-indol-3-yl)methylene]−1-naphthohydrazide (5p)

%Yield = 90, m.p = 208–210 °C, off-white solid, IR(KBr)cm−1: 3130, 2353, 1650, 1608, δ H (600 MHz, DMSO-d6) 12.11 (1 H, s), 8.59–8.44 (1 H, m), 8.37 (1 H, s), 8.22 (1 H, m), 8.15–7.89 (4 H, m), 7.86–7.71 (2 H, m), 7.60 (3 H, m), 7.46 (3 H, m), 7.34 (1 H, d, J = 8.0 Hz), 2.33 (3 H, s); 13C NMR (151 MHz, DMSO) δ 165.04, 146.37, 142.63, 138.33, 135.29, 134.98, 134.10, 133.64, 133.27, 130.91, 130.47, 129.75, 128.83, 127.54, 127.36, 126.91, 126.56, 125.83, 125.49, 124.79, 123.85, 122.88, 118.65, 113.62, 21.5. QTOF HRMS (m/z): [M + H]+, calcd: 468.1381, found: 468.2694.

(E)−3-Hydroxy-N’-[(1-tosyl-1H-indol-3-yl)methylene]−2-naphthohydrazide (5q)

%Yield = 97, m.p = 265–267 °C, off-white solid, IR(KBr)cm−1: 3120, 2348, 1642, 1553, 757, δ H (600 MHz, DMSO-d6) 12.10 (1 H, s), 8.64 (1 H, s), 8.40 (1 H, s), 8.03–7.75 (7 H, m), 7.57–7.31 (8 H, m), 2.32 (3 H, s); 13C NMR (151 MHz, DMSO) δ 164.35, 146.42, 143.42, 136.33, 135.26, 134.21, 130.90, 130.81, 130.75, 130.53, 129.13, 128.70, 127.43, 127.37, 127.21, 126.37, 126.34, 124.82, 124.27, 123.78, 120.74, 118.52, 113.62, 111.07, 21.53. QTOF HRMS (m/z): [M + H]+, calcd: 484.1331, found: 484.1331.

(E)−2-Methyl-N’-[(1-tosyl-1H-indol-3-yl)methylene]benzohydrazide (5r)

%Yield = 95, m.p = 194–196 °C, off-white solid, IR(KBr)cm−1: 3134, 2355, 1647, 1632, δ H (600 MHz, DMSO-d6) 11.88 (1 H, s), 8.48 (1 H, s), 8.45–8.41 (1 H, m), 8.34 (1 H, s), 7.97 (1 H, d, J = 8.3 Hz), 7.94–7.90 (1 H, m), 7.86 (1 H, dd, J = 8.3 Hz, 4.2 Hz), 7.48–7.25 (8 H, m), 2.39 (2 H, s), 2.32 (2 H, s), 2.29 (1 H, s), 2.22 (1 H, s); 13C NMR (151 MHz, DMSO) δ 165.54, 146.33, 142.23, 138.27, 135.85, 134.35, 134.17, 131.06, 130.85, 130.50, 129.11, 127.95, 127.34, 127.08, 126.30, 125.78, 124.74, 123.82, 118.51, 113.59, 21.52, 19.81. QTOF HRMS (m/z): [M + H]+, calcd: 432.1381, found: 432.2604.

In vitro cytotoxicity assay of synthetic compounds

An invasive breast cancer (BC) cell line MDA-MB-231 was utilized to perform an MTT (yellow tetrazolium salt, 3-(4, 5-dimethylthizol-2-yl)−2, 5-diphenyl tetrazolium bromide) assay in order to determine the in vitro cytotoxic effects of synthetic compounds. MCF-10 A cell lines were used as a control in the research43,44. Cells were grown in DMEM supplemented with 1% antibiotics (100 U/ml penicillin) and 10% FBS. After seeding in a 96-well plate at a density of 1.0 × 104 cells/well, the cells were then incubated with 5% CO2 at 37 °C for 24 h. The medium was disposed of, and both cell lines were exposed to varying concentrations (6.5µM, 12.5µM, 25µM and 50µM) of synthesized triazole derivatives45. After 48 h of incubation46, 20µL of MTT solution (5 mg/mL) was poured into every well then incubated for further 4 hours. After discarding the medium, DMSO was used to dissolve the formazan precipitate. A microplate reader at 570 nm was utilized to determine the mixture’s absorbance. Each experiment was performed in triplicates and the cytotoxic effects were demonstrated as % cell viability in comparison to untreated control cells47.

$$\:\begin{array}{c}\%\:Viability\:=\frac{Absorbance\:of\:sample}{Absorbance\:of\:control}\times\:100\end{array}$$
(1)

Computational methods

Molecular Docking simulations

Synthesized compounds were designed and optimized using the software MarvinSketch (ChemAxon, Version 22.13). Hydrogens were added and cleaned up each structure in 2D and 3D, both dimensional perspectives. Several possible molecule conformers were generated and chose the lowest energy configuration for additional analysis. DockPrep module in Chimera version 1.17.1 (build 42449) was used to process our Mol2 files of the 3D structures by choosing default parameters like protonation states with AM1-BCC during conjugate gradient optimization. The RCSB Protein Data Bank47 offered us the crystal structure of Pyrrolo[3,2-d]pyrimidine-based inhibitor bound with EGFR T790M/L858R variant through PDB ID: 3W2S46. The selected protein was validated by examining protein resolution and wwPDB scores, observing missing residues in binding sites on PDBsum48, and looking at Ramachandran plot data. Same process was used to optimized the raw PDB file using the DockPrep module on Chimera by removing unwanted residues and adding hydrogen atoms to prepare the structure for the AMBER force field and charge adjustment and saved refined structure to PDB file format. AutoDockTools version 1.5.6, developed by The Scripps Research Institute, was used to convert our protein framework to the PDBQT format. The molecular docking analysis was conducted utilizing AutoDockTools 1.5.649, Chimera50, and Maestro51 were used for grid Generation and validation. The grid was developed at the binding site where W2R was positioned as the co-crystalized ligand and was designed using 20 × 20 × 20, orienting in x, y, and z directions, respectively, with a grid point spacing of 0.375 Å and the center dimensions (x = 5.07, y = 1.11, z = 10.35) respectively. Docking simulations used AutoDock Vina52,53 Version 1.2.5 through a Windows operating system. Procedure for docking were ran in triplicate using different grid sizes to verify reliability and repeatability. Software operated based on preset conditions that handled CPU speed, grid size, search intensity, mode quantity, and energy limits. Post-Docking Analysis: Python scripts created from AutoDockTools features used to process the docking results which enabled us to split and form protein and ligand connections. Visualization of protein-ligand complexes were studied through Discovery Studio54, PLIP55, and MAESTRO visualization programs.

GA-MLR QSAR study

The dataset was utilized for developing GA-MLR (genetic algorithm multiple linear regression) models with the widely used software QSARINS version 2.2.2, and it was validated both internally and externally. This methodological approach is similar to earlier publication22,56,57.

Molecular dynamics simulation (MDs)

MD simulations were carried out on the dock complex for 3W2S with 5q and W2R represented as 3W2S_5q and 3W2S_W2R, respectively, utilizing the Desmond 2020.1 from Schrödinger, LLC. The OPLS-2005 force field and explicit solvent model with the TIP3P water molecules were utilized in this system in period boundary salvation box with dimensions 10 Å x 10 Å x 10 Å. NaCl solutions were put into the system to activate the physiological environment, and Na+ ions were added to neutralize the 0.15 M charge58,59. To retrain over the protein-ligand complexes, the equilibration of the system was first carried out for 10 ns utilizing an NVT ensemble. Following the initial step, an NPT ensemble was utilized to perform a brief run of minimization and equilibration for 12 ns. The Nose-Hoover chain coupling strategy was utilized to configure an NPT ensemble with variable temperatures, a 1.0 ps relaxation time, and 1 bar pressure was retained throughout all simulations58,59. A 2 fs time scale was employed. The Martyna-Tuckerman–Klein chain coupling strategy barostat method was utilized for controlling pressure with 2 ps relaxation time. The particle mesh Ewald approach was used to determine long-range electrostatic contacts, and the radius for coulomb contacts was set at 9Å. The final production run was carried out for 100 ns. The RMSD, RMSF, Rg, and salt bridges, numerous H-bonds, and SASA are computed to evaluate MD simulations’ stability58,59.