In silico and in vitro studies of gemcitabine derivatives as anticancer agents

Ali Jabbar Radhi; Mahmood M. Fahad; Ihsan Alrubaie; Nusrat Shafiq; Kinza Afzal

doi:10.3897/pharmacia.72.e153719

Abstract

Cancer remains one of the leading causes of death globally and is expected to increase in incidence. The present investigation aimed to evaluate the anticancer efficacy of gemcitabine derivatives on MCF7, A549, and PC3 cell lines. Compounds Com. 10 and Com. 16 were the most potent, with IC₅₀ values of 9.45, 6.93, and 12.09 µM (Com. 10) and 12.23, 6.60, and 21.57 µM (Com. 16) for the three cell lines, showing a better response compared to the reference drug gemcitabine. Com. 9 and Com. 14 were also moderately active, preferentially on A549 and PC3. Com. 16 showed strong binding with the active residues as determined by molecular docking. Com. 16 bound to key cancer targets Vav1 (6NEW), ERK2 (4ZXT), and CYP3A4 (7LXL), with docking scores of −191.72, −187.66, and −221.08 kcal/mol, respectively. These interactions were electrostatic, hydrophobic, and π-type, similar to those induced by gemcitabine. Several other compounds also demonstrated good docking scores (−160 to −180 kcal/mol), and eight protein–ligand complexes emerged as leads. Moreover, ADME and physicochemical analyses of the top six active compounds indicated good drug-likeness and a favorable PK profile compared to gemcitabine.

Keywords

MCF7, PC3, A549, gemcitabine, molecular docking, anticancer activity

Introduction

Cancer ranks third as a cause of death worldwide, resulting in an estimated 10 million deaths in 2020 (Sung et al. 2021). Before the COVID-19 pandemic, it was the WHO’s second leading cause of death. In developed countries, cancer remains a considerable public health problem because of the persistent resistance to therapies and the process of metastasis, which prevents effective management and therapeutic success (Sung et al. 2021; Wu et al. 2024). The failure of current treatment, partly ascribed to resistance phenomena but also to inefficient treatment against metastases, is a major contributing factor to this high mortality rate. Nucleoside analogs are a major component of the chemotherapy regimens used to treat cancer today. These compounds are designed to resemble purine and pyrimidine nucleosides found in nature. Among these nucleoside analogs is gemcitabine. Gemcitabine is effective against a variety of solid tumors, including malignancies of the pancreas, non-small cell lung, prostate, breast, and ovaries (Carmichael et al. 1996; Hoang et al. 2003; Ozols 2005; Yardley 2005). Tumors often develop resistance over time, which is one of the main challenges in cancer treatment. The mechanism of action of gemcitabine is closely associated with this resistance (Scheme 1). Different transporters, including hENT1, are involved in the absorption of gemcitabine into cells. Reduced expression of hENT1 leads to a reduction in the activity of gemcitabine by obstructing its uptake (Rauchwerger et al. 2000). To become active, gemcitabine must undergo phosphorylation several times inside the cell, with deoxycytidine kinase (dCK) performing the initial phosphorylation.

Scheme 1.

Chemical structure of gemcitabine.

There is a correlation between low gemcitabine cytotoxicity and low dCK levels (Farrell et al. 2009). Moreover, cytidine deaminase (CDA) rapidly deaminates gemcitabine, resulting in a brief plasma half-life. Thus, novel therapeutic approaches may arise from methods that promote high metabolic bioevasion and improved transport by chemical alteration (Itoi et al. 2007). Gemcitabine’s structure has been modified over time by introducing functional groups at the 4-(N)-position, which corresponds to the pyrimidine ring’s amino group, or at the 5’-(O) position, which is where the hydroxyl group is attached to the furan ring (Vandana and Sahoo 2010). There are many benefits to this technique, including increased water solubility, biocompatibility, longer half-lives for produced gemcitabine derivatives, and, most importantly, enzyme protection. PEG–gemcitabine is one well-known example. It is made by PEGylation at the 4-(N)-position, where PEG polymer chains are attached to the medication via covalent or non-covalent bonds (Immordino et al. 2004; Pasut and Veronese 2009). In 1998, Eli Lilly received a patent for creating lipophilic encapsulated gemcitabine. Using long-chain saturated lipids C18 and C20, the amino group at position four was covalently linked in this procedure. Compared to unaltered gemcitabine, the resultant lipophilic compounds showed increased cytotoxicity (Tsume et al. 2014a, b). In an attempt to improve gemcitabine’s metabolic stability, Tokunaga Y and associates synthesized a number of prodrugs. By joining the 4-(N)-position with derivatives of stearoyl, valeroyl, and lauroyl and then encasing them in liposomes, these prodrugs enhanced lipophilicity (Tsume et al. 2014a, b). Gemcitabine’s 4-(N)-amino group and PEG’s COOH were implicated in the drug linkage, whereas folic acid was connected to the PEG amino group via its carboxylic function. Because folic acid receptors are overexpressed in a number of cancer types, including lung, breast, kidney, and ovarian cancers (Iwakiri et al. 2008; Kalli et al. 2008; Yang et al. 2010), and because these receptors may be sparsely distributed in healthy human tissues, folic acid was selected as the targeted agent. Folate on the surface of numerous nanocarriers, such as gold nanoparticles, liposomes, or magnetic nanoparticles, has demonstrated effective targeting for the detection of cancer cells and the release of anticancer medications (Shmeeda et al. 2010; Yang et al. 2010; Kaaki et al. 2012). To increase their stability, prodrugs of gemcitabine were made with both D- and L-amino acids. When compared to gemcitabine alone, both prodrug variations demonstrated greater efficacy against pancreatic cancer cells (Bender et al. 2009). Furthermore, gemcitabine dipeptide prodrugs showed improved antiproliferative action against pancreatic cancer cells (Koolen et al. 2011). Gemcitabine, a deoxycytidine analog, exhibits high antitumor activity against various cancer cells. IC₅₀ values vary in the literature depending on the cell type and experimental protocol. For instance, gemcitabine has an IC₅₀ of 8–20 µM in MCF7 (breast cancer) cells (Chou 2003; Giovannetti 2006), 10–25 µM in A549 (lung cancer) cells (Nakano et al. 2007; Ewald et al. 2008), and 15–30 µM in PC3 (prostate cancer) cells (Nakano et al. 2007; Ewald et al. 2008; Khan 2012). Such values are consistent with the wide spectrum of pharmacological activity of gemcitabine, which validates its continued use as a reference compound for anticancer activity. This study used gemcitabine derivatives modified at position 2 of the furan ring by a pentacyclic ring, such as 1,2,3-triazoles and tetrazole derivatives (Salman et al. 2019; Mohammed et al. 2020; Naser et al. 2020; Ali Jabbar Radhi et al. 2021). This research examined the anticancer activities of gemcitabine derivatives against MCF7, A549, and PC3 cell lines and compared them to that of the reference drug gemcitabine. Gemcitabine, a deoxycytidine analogue antimetabolite, is a widely used agent for several types of solid tumors, such as breast (Ferrazzi and Stievano 2006), lung (Cappuzzo 2009), and prostatic cancer (Thakkar et al. 2006). In this study, the reference drug compound was used for in vitro testing against the selected cancer cell lines (MCF7, A549, and PC3), as well as for comparative in silico molecular docking studies against the selected targets: Vav1 (PDB ID: 6NEW), ERK2 (PDB ID: 4ZXT), and CYP3A4 (PDB ID: 7LXL). While binding data for Vav1 are sparse, gemcitabine has been reported to interfere with cytoskeleton dynamics and to inhibit downstream signaling of Vav1-controlled signals (Rho/Rac) that modulate cell motility and proliferation (Aslan 2019). In docking predictions, all Vav1-binding compounds possessed stable hydrogen bonds and hydrophobic contacts with Vav1 and could be modulators of Vav1 activity. The inhibition of the MAPK/ERK pathway, including decreased phosphorylation of ERK2 by gemcitabine, is widely known (Zhu et al. 2023). Docking simulations also validated its binding to critical ERK2 residues, suggesting obstruction of ATP binding or kinase activity. Gemcitabine docked in the active site of CYP3A4 (a drug metabolic enzyme) in positions that interacted with this protein in a manner consistent with its known metabolic clearance in hepatic and tumor tissues (Torres Robles et al. 2025). The docking results confirm the binding of gemcitabine to each of the three targets and consequently its status as a mechanistically related reference compound. This is the first work to use site-2 substituted gemcitabine derivatives and study their efficacy against cancer cells. Gemcitabine derivatives were studied in silico to better understand the potential for drug-receptor interactions.

In vitro anticancer activity

The cytotoxic activity of gemcitabine (obtained from Sigma) and its derivatives (Com. 1–Com. 16), which were synthesized in our previous studies as reported in the literature (Ali Jabbar Radhi et al. 2021; Mohammed et al. 2020; Naser et al. 2020; Salman et al. 2019) (Scheme 2), was examined using the well-known MTT protocol (Mahshid Ghasemi et al. 2023) against selected human cancer cell lines: A549 (human lung cancer), PC3 (human prostate cancer), and MCF7 (human breast cancer). Gemcitabine served as the positive reference control. Spectrophotometry was used; the examination determines when the metabolically active cells used in this study reduce tetrazolium yellow (MTT) to purple formazan. At a density of 5000 cells per well, the cell lines were seeded onto 96-well plates. The growth medium was then replaced with a mixture of 10% fetal calf serum (FCS), 2 mM L-glutamine, 100 IU mL^–1 penicillin, and 100 mg mL^–1 streptomycin. For twenty-four hours, the cells were cultured in 5% CO₂ at 37 °C. Next, gemcitabine in DMSO solvent was applied to each well of the plate, along with a variety of concentrations of the studied drugs, and the plate was left to incubate for 48 hours. Following that, each well received 10 µL of a freshly prepared MTT reagent solution, and the plate that contained a sample with human cells was incubated in a CO₂ incubator for four hours at 37 °C. The human cells used in this study were dissolved in ethanol, and their optical density was measured at 570 nm after the purple precipitate was formed. Six wells were used in the experiment, and the test sample concentration was tested in tandem with the control DMSO concentration. The percentage of all the samples used in this work, through measured proliferation inhibition, was determined using the below formula:

Scheme 2.

Chemical structure of the gemcitabine derivatives (Com. 1–Com.16).

% Cell inhibition = 100 – [(At – Ab) / (Ac – Ab)] × 100

Where, At = absorbance of test compound, Ab = absorbance of blank, and Ac = absorbance of control.

The gemcitabine is symbolized in blue, and its functional groups are emphasized in red. These colors are for illustrative purposes only.

ADME predictions

Physicochemical property calculation was performed using ChemBio3D in the ChemBioOffice Ultra 18.0 software package (PerkinElmer, USA, https://www.perkinelmer.com). In silico absorption, distribution, metabolism, excretion, and toxicity (ADMET) predictions were performed using two online programs: SwissADME and ADMETlab 2.0. The pharmacokinetic/drug-like features profile screener SwissADME (http://www.swissadme.ch) offers basic pharmacokinetic and drug-like profiles, while ADMETlab 2.0 uses machine learning algorithms, including random forest (RF), support vector machine (SVM), and k-nearest neighbor (k-NN), trained on curated and validated chemical datasets. These models were developed based on structure–activity relationship (QSAR) analysis to predict important descriptors in five pharmacokinetic and toxicological domains, with classification (e.g., toxic/non-toxic) and regression (e.g., logP, half-life) capabilities. In this QSAR, molecular structure in SMILES format was given as input, and molecular- and physicochemical descriptor-based predictions were made (Xiong et al. 2021).

In silico molecular docking studies

Target selection rationale

Three protein-based targets one for each cancer cell line (MCF7, A549, and PC3) were chosen for molecular docking studies in the present work. The docking accuracy was verified by re-docking the native ligands of 4ZXT and 7LXL, with the resulting RMSD at 1.83 Å and 1.17 Å, respectively. Overfitting was estimated by re-docking one of the ligands (Com. 16) already tested, and an RMSD of 1.24 Å was obtained for 6NEW, which had no native ligand, thus confirming the validation of our docking methodology. The criteria for selection were based on relevance to cancer biology, structural availability in the Protein Data Bank (PDB), and the importance of each target in critical oncogenic signaling pathways. The nature of each target is described in Table 1. 6NEW is a synthetic compressed form of Vav1 from MCF7 Her2 cons GEF to Rho/Rac GTPases, involved in the regulation of the actin cytoskeleton, cell survival, and cell proliferation. Overexpression of Vav1 is associated with several types of cancer, including breast cancer (Liu et al. 2007; Zhang et al. 2014). 4ZXT is an ERK2 MAPK signal transduction kinase that mediates cell growth, survival, and differentiation. The ERK/MAPK pathway is frequently altered in non-small cell lung cancer (A549), and its inhibition has been a target in the development of targeted drugs (Dhillon et al. 2007). 7LXL, cytochrome P450 3A4 (CYP3A4), a member of the cytochrome P450 superfamily, is essential for the metabolism of many drugs. CYP3A4 overexpression or increased function has also been observed in prostate cancer and is associated with drug resistance to chemotherapy (Mann 2008; Varani et al. 2003). These proteins were chosen not only because they are established cancer-related proteins but also because they possess 3D crystallographic structures stored in the RCSB PDB, which made more reliable docking and interaction studies feasible. The tested gemcitabine derivatives (Com. 1–Com. 16) were chosen and developed to attack dominant proteins regulating major signaling pathways in relation to breast (MCF7), lung (A549), and prostate (PC3) cancers.

Table 1.

Download as

CSV

XLSX

Structural and functional profiling of tumor suppressor proteins: a mosaic of PDB sum contents and cancer-associated context.

Target protein (PDB ID)	Proteome source	Protein Function	Cancer relevance	Justification for selection	Resolution	Organism
6NEW (Vav1)	Human (MCF7)	Rho/Rac GTPase exchange factor for guanine nucleotide.	Modulator of cell survival and migration; involved in breast cancer	Reservoirs of oncogenic signaling and cytoskeletal regulation	2.50 Å	Homo sapiens
4ZXT (ERK2 Complex)	Human (A549)	MAPK/ERK signaling kinase	Overexpressed in lung cancer; promotes growth and survival	Targeted by MAPK inhibitors; implications for NSCLC treatment	2.00 Å	Homo sapiens
7LXL (CYP3A4)	Human (PC3)	Cytochrome P450 responsible for metabolism of drugs	Related to drug resistance of prostate cancer	Affects gemcitabine bioactivation and elimination	2.75 Å	Homo sapiens

Cellular functions and biological roles of selected targets

Vav1 (PDB ID: 6NEW): Vav1 belongs to the Vav family of guanine nucleotide exchange factors (GEFs), which selectively activate Rho family GTPases, including Rac1 and Cdc42. These small GTPases are involved in various cellular functions, including cytoskeletal remodeling, transcription, cell proliferation, and survival. Vav1 is an important hematopoietic cell-mediated transducer, but its expression is dysregulated in several solid tumors, such as breast cancer. Its oncogenic function correlates with the induction of cell migration, invasion, and metastasis via the activation of actin cytoskeleton reorganization and regulation of downstream signaling pathways, including NF-κB and MAPK. High levels of Vav1 are associated with poor prognosis for breast cancer patients and hence become attractive targets for therapy (Zhang et al. 2014). Extracellular Signal-Regulated Kinase 2 (ERK2, PDB ID: 4ZXT): ERK2 is a mitogen-activated protein kinase (MAPK) serine/threonine kinase. It serves as a key effector, relaying extracellular growth signals from receptor tyrosine kinases to the nucleus to regulate cell proliferation, differentiation, survival, and apoptosis. The activity of ERK2 is highly controlled under physiological conditions, but many cancers, including non-small cell lung cancer (NSCLC), activate ERK2 constitutively following mutation or upregulation of upstream regulators. Such sustained signaling leads to a loss of normal control of cell division and resistance to apoptosis. Inhibition of ERK2 has been successful in abrogating such oncogenic signals (Dhillon et al. 2007). 3A4 (cytochrome P450 3A4, PDB ID: 7LXL): CYP3A4 is the most important enzyme from the cytochrome P450 superfamily and participates in the oxidative metabolism of a broad spectrum of endogenous compounds and xenobiotics, including a large number of chemotherapeutic drugs. In prostate cancer, CYP3A4 affects the metabolism and bioavailability of anticancer compounds such as gemcitabine. Changes in CYP3A4 function may lead to chemoresistance by altering the activation or clearance of specific drugs and may therefore impact the effects of therapy. Furthermore, CYP3A4 is involved in the metabolism of steroid hormones, which play a crucial role in the progression of prostate cancer. Knowledge of CYP3A4 is important for establishing correct drug dosing and understanding resistance mechanisms (Mann 2008; Varani et al. 2003).

Ligand preparation

Primarily, chemical structures were carefully illustrated using ChemDraw (version 19.0.1.8, PerkinElmer) and subsequently brought into Chem3D Ultra (version 19.0.1.8, PerkinElmer). Optimization was conducted using MM2 and MMFF9 force fields, safeguarding the refinement of three-dimensional geometries for precise structural analysis, and saved in Mol2 format.

Protein preparation

The 3D protein structures of target proteins corresponding to the three cancer cell lines (MCF7, PC3, and A549) were retrieved from the RCSB PDB Protein Data Bank (www.rcsb.org) and prepared within Discovery Studio 2021 in accordance with conventional crystallographic principles. This preparation consisted of loop building, completion of missing atoms in unresolved residues, and removal of water and heteroatoms. The processed protein structures were subsequently loaded into Molegro Virtual Docker (MVD) 6.0 for docking (Thomsen and Christensen 2006). For apo structures such as 6NEW, binding sites were determined using MVD’s cavity detection algorithm in a 15 Å radius centered on the largest cavity. All protein remainder atoms were considered rigid, except for the selected binding site remainder atoms: GLU60, LYS58, TYR23, and ARG64 (6NEW); LYS54 and GLU71 (4ZXT); and LEU216, GLY481, and PHE304 (7LXL); these were assigned to be flexible. Ligands (gemcitabine and Com. 1–Com. 16) were prepared with ChemDraw, transformed to 3D using Chem3D, and energy-minimized applying MMFF94. The ligands were treated as completely flexible. Docking was carried out using the MolDock scoring function (PLP) with a maximum of 1500 iterations and a population size of 50. The best pose with the lowest MolDock score was chosen and then analyzed by Discovery Studio to investigate the key binding interactions. The detailed characteristics of the proteins are summarized in Table 1.

Protein–ligand docking

For in silico studies, Molegro Virtual Docker (MVD) was used for docking simulations, followed by post-docking modeling and visualization in Discovery Studio. First, the prepared protein was imported into MVD, then the desired ligand was introduced. Docking parameters were kept at default. The docking exercise was directed to generate a bioactive binding pose at active sites. The lead pose was chosen based on the MolDock score, representing the binding potential between the protein and ligand.

Afterward, a 3D image of the docking results was analyzed. Field interactions were examined to explain hydrogen bonding in the workspace. The docked elements were transferred to Discovery Studio for ligand interaction and hydrophobicity assessment. The hydrophobicity of a protein–ligand complex was computed on a scale of 3 to −3, with positive values indicating an increase and negative values indicating a decrease in hydrophobicity. These hydrophobic values were used to quantify the energy contributions of hydrophobic interactions within the protein–ligand complex. A two-dimensional image was created to detect interactions between the ligand and its locale. The image was annotated with hydrogen bonds, covalent bonds, π, and π–alkyl interactions. However, unfavorable steric bumps, donor/acceptor clashes, and charge repulsion were purposely disregarded. Additionally, amino acid labels with three-letter codes were added for clarity.

Results and discussion

The gemcitabine compounds (Com. 1–Com. 16) produced in earlier studies (Ali Jabbar Radhi et al. 2021; Mohammed et al. 2020; Naser et al. 2020; Salman et al. 2019) were also examined for their anticancer activity using the MTT assay against three different cancer cell lines: MCF7, PC3, and A549. Gemcitabine (Re.) served as a non-selective positive control. The results are summarized in Table 2.

Table 2.

Download as

CSV

XLSX

Anticancer (IC₅₀, µM) activities of prepared compounds (Com. 1–Com. 16).

References	WI-38	PC3	A549	MCF7	Sr. no.
(Salman et al. 2019)	-	51.93 ± 1.78	86.12 ± 2.17	106.4 ± 1.14	Com.1
(Salman et al. 2019)	-	124.4 ± 1.24	128.5 ± 1.67	130.1 ± 1.01	Com.2
(Salman et al. 2019)	-	102.9 ± 1.09	45.32 ± 1.17	103.5 ± 1.08	Com.3
(Mohammed et al. 2020)	-	113.8 ± 2.45	114.9 ± 0.89	45.48 ± 0.54	Com.4
(Mohammed et al. 2020)	-	71.06 ± 1.68	190.3 ± 2.07	269.1 ± 0.91	Com.5
(Mohammed et al. 2020)	-	69.39 ± 1.49	135.7 ± 1.27	146.7 ± 1.51	Com.6
(Mohammed et al. 2020)	-	44.56 ± 1.65	43.96 ± 0.57	135.2 ± 2.04	Com.7
(Mohammed et al. 2020)	-	53.11 ± 1.42	70.72 ± 1.78	628.6 ± 1.35	Com.8
(Ali et al. 2021)	38.78 ± 0.79	39.39 ± 0.86	29.42 ± 1.68	91.12 ± 1.98	Com.9
(Ali et al. 2021)	27.14 ± 1.08	21.57 ± 2.34	6.599 ± 1.08	12.23 ± 1.38	Com.10
(Naser et al. 2020)	-	45.95 ± 2.17	106.2 ± 0.97	81.58 ± 0.87	Com.11
(Naser et al. 2020)	-	85.15 ± 1.08	89 ± 2.52	113.7 ± 1.09	Com.12
(Naser et al. 2020)	-	89.71 ± 1.37	52.37 ± 1.38	114.2 ± 1.68	Com.13
(Naser et al. 2020)	74.25 ± 1.12	36.29 ± 1.54	58.94 ± 1.82	88.4 ± 1.47	Com.14
(Naser et al. 2020)	-	40.81 ± 1.33	41.05 ± 1.68	70.34 ± 0.98	Com.15
(Naser et al. 2020)	27.18 ± 1.27	12.09 ± 1.03	6.931 ± 0.76	9.454 ± 1.78	Com.16
	51.18 ± 1.05	24.09 ± 2.04	19.35 ± 0.99	15.34 ± 1.07	Re.

In a series of gemcitabine derivatives (Com. 1–Com. 16) used as anticancer agents, good to moderate activity (IC₅₀ = 6.599–45.48 µM) or weak activity (IC₅₀ > 50 µM) was observed toward MCF7, A549, and PC3 cell lines. The extent of cell viability after treatment with the gemcitabine derivatives used in this study was visualized through microscopic images obtained using the MTT method (Figs 1–3). We must focus on the color and shape changes that indicate cellular metabolic activity. The MTT test measures cellular metabolic activity as an indicator of cell viability, proliferation, and toxicity. Metabolically active cells reduce MTT to formazan, which can be observed as a color change (Mahshid Ghasemi et al. 2023). In images (Fig. 1a, b), the cells appear elongated and spindle-shaped, and the density seems relatively low, which indicates an early growth stage or low seeding density. That is, there is minimal rounding of the cells, which suggests that the majority of cells are viable. As for (Fig. 1c, d), the cells are more rounded and less elongated, indicating possible exposure to stress or a reduced ability to survive. The density is also higher, which may indicate a more advanced growth stage or a response to treatment. The density of the formazan product from the MTT assay was measured using a spectrophotometer. Higher absorbance values correspond to higher cell viability and metabolic activity, from which the results shown in Table 2 were obtained.

Figure 1.

Morphological changes in PC3 cell lines as viewed under the inverted microscope after post-treatment with 100 µM: a. Treated cells with Com. 1; b. Treated cells with Com. 8; c. Treated cells with Com. 10; d. Treated cells with Com. 16.

Figure 2.

Morphological changes in MCF7 cell lines as viewed under the inverted microscope after post-treatment with 100 µM: a. Treated cells with Com. 10; b. Treated cells with Com. 12; c. Treated cells with Com. 14; d. Treated cells with Com. 16.

Figure 3.

Morphological changes in A549 cell lines as viewed under the inverted microscope after post-treatment with 100 µM: a. Treated cells with Com. 3; b. Treated cells with Com. 7; c. Treated cells with Com. 11; d. Treated cells with Com. 15.

Compounds Com. 10 and Com. 16 showed significant anticancer activity against MCF7 cell lines (IC₅₀ = 12.23 and 9.454 µM, respectively) when compared to gemcitabine (IC₅₀ = 15.34 µM). On the other hand, compound Com. 4 had moderate anticancer activity against MCF7 cell lines (IC₅₀ = 45.48 µM), and other gemcitabine derivatives had weak anticancer activity (IC₅₀ > 50.00 µM), as shown in Fig. 4.

Figure 4.

IC₅₀ values of anticancer activity against the MCF7 cell lines (Re., Com. 10, and Com. 16.

Compounds (Com9, Com10, and Com16) showed good anticancer activity against the human lung cancer A549 cell lines with IC₅₀ values at (29.42, 6.599, and 6.931 µM), respectively, in comparison with the control drug gemcitabine (IC₅₀ = 19.35 µM), but compounds (Com.3, Com.7, and Com.15) exhibited moderate anticancer activity against A549 cell lines (IC₅₀ = 45.32, 43.96, and 41.05 µM, respectively). The remaining drugs have poor anticancer action against the A549 cell lines, with IC₅₀ values greater than 50 µM. (Fig. 5).

Figure 5.

IC₅₀ values of anticancer activity against the A549 cell lines (Re., Com. 9, Com. 10, and Com. 16).

All gemcitabine derivatives were also tested for anticancer activity against the human prostate cancer PC3 cell lines. Compounds Com. 10, Com. 14, and Com. 16 showed good anticancer activity against PC3 cell lines, with IC₅₀ values of 21.57, 36.29, and 12.09 µM, respectively, compared with the control drug gemcitabine (IC₅₀ = 24.09 µM). Some gemcitabine compounds (Com. 7, Com. 9, Com. 11, and Com. 15) exhibited moderate anticancer activity against PC3 cell lines (IC₅₀ = 44.56, 39.39, 45.95, and 40.81 µM, respectively). The other gemcitabine derivatives showed weak anticancer activity against PC3 cell lines, with IC₅₀ values greater than 50.00 µM (Fig. 6).

Figure 6.

IC₅₀ values of anticancer activity against the PC3 cell lines (Re., Com. 10, Com. 14, and Com. 16).

On the other hand, some compounds exhibited good anticancer activity against only one cell line, such as compound Com. 9 against A549 and compound Com. 14 against PC3 cell lines, whereas compounds Com. 10 and Com. 16 showed good anticancer activity across all cell lines employed in this investigation.

Analyzing the cytotoxicity of a normal cell line in vitro

Here, gemcitabine was taken as a reference, by which the in vitro cytotoxicity of the top potent compounds Com. 9, Com. 10, Com. 14, and Com. 16 was tested against the human lung normal cell line WI-38. The synthetic compounds Com. 9, Com. 10, Com. 14, and Com. 16 had IC₅₀ values of 38.78, 24.14, 74.25, and 27.18 µM in non-cancerous WI-38 cells, respectively (Table 2). These values are within the same range as their IC₅₀ values in cancer cell lines (6.60–39.39 µM), suggesting little to no selectivity. Although IC₅₀ values are in the micromolar range both in normal and cancer cell lines (which indicate some degree of safety margin), the relatively low potency raises caution toward therapeutic development. Additional optimization is required to enhance the selectivity window.

Cytotoxicity and selectivity indices in normal cell lines

The selectivity index (SI) is an important endpoint in evaluating how well a compound is able to selectively kill cancer cells relative to normal cells (Equation 1), and it serves as a quantitative indicator of the therapeutic window for these compounds (Zhou et al. 2020; Krzywik et al. 2020).

$SI Formula: SI = \frac{{IC}_{50} (no cancer cell)}{{IC}_{50} (cancer cell)}$

Cytotoxicity and SI determination of the tested gemcitabine derivatives showed that Com. 16 possessed the most attractive profile in terms of potent anticancer activity against MCF7, A549, and PC3 (IC₅₀: 9.45, 6.93, 12.09 µM) with high selectivity against cancer cells over normal cells (SI: 2.88, 3.92, and 2.25, respectively) and thus could be considered the worthiest candidate for future investigation. Com. 10 also exerted significant cytotoxicity, especially for MCF7, PC3, and A549, and showed moderate SI values (2.22, 1.26, and 4.11), indicating possible cancer selectivity. Com. 14 showed some selectivity, predominantly for PC3 (SI = 2.05), but it was less potent. In contrast, Com. 9 presented low SI values for all lines (0.43–1.32), signifying poor discrimination between cancerous and non-cancerous cells, which would limit its use as a therapy. These results also emphasize the need for normal cell line toxicity testing to determine compounds with a favorable therapeutic index early in drug development. All results are summarized in Table 3.

Table 3.

Download as

CSV

XLSX

Selectivity indices (SI) of tested compounds vs. gemcitabine.

Com.	PC3 SI (Prostate)	A549 SI (Lung)	MCF7 SI (Breast)
Com.9	0.98 ± 0.02	1.32 ± 0.07	0.43 ± 0.01
Com.10	1.26 ± 0.08	4.11 ± 0.67	2.22 ± 0.20
Com.14	2.05 ± 0.05	1.26 ± 0.04	0.84 ± 0.01
Com.16	2.25 ± 0.10	3.92 ± 0.43	2.88 ± 0.54
Re.	2.12 ± 0.09	2.64 ± 0.14	3.34 ± 0.23

Molecular docking analysis

The threshold for docking scores was established by comparing the docking scores of the test compounds (Com. 1–Com. 16) with the reference drug gemcitabine. Compounds with more negative scores than gemcitabine were considered to have better binding affinities. Favorable interactions with the major active site residues were also visualized by binding mode inspection (Ventura and Serrano 2004). The computational technique MVD was used for 16 compounds against three different cancer cell lines MCF7, A549, and PC3 with PDB IDs: 6NEW, 4ZXT, and 7LXL. The main objective was to determine the minimum potential energy. MVD operates by locating cavities on the protein surface, generating poses, and selecting the best pose based on the MolDock score. Using Molegro Virtual Docker, five possible binding sites were identified in the Vav1 crystal structure (PDB ID: 6NEW). The MVD workspace was then exported to Discovery Studio, providing a more visually intuitive representation of interactions between protein and ligand, elucidating their type, category, and bond distances. Additionally, all Vav1-binding compounds formed stable hydrogen bonds and hydrophobic contacts with essential residues in the Vav1 binding pocket, including LYS58, GLU60, and ARG64, potentially implicating them as modulators of Vav1 (interaction details in Table 4). The computational observations are consistent with previous studies that have shown small-molecule inhibition of the Vav1-regulated Rac/Rho signaling pathway (Zhu et al. 2023). For ERK2, gemcitabine has a demonstrated inhibitory effect on the MAPK/ERK pathway, including reduction in ERK2 phosphorylation, in various types of cancer cells (Torres Robles et al. 2025). For the analysis of docking results, the threshold was set to −160 to −180 (Tables 5–7). Only eight protein–ligand complexes emerged as leads based on their scores. Investigating the MCF7 cell line, protein 6NEW gave the highest MolDock score of −163.385 with Com. 10; the rest were below the threshold. Remarkably, Com. 10 exhibited a total of 14 interactions, including hydrogen bonds and carbon–hydrogen bonds with amino acids LYS58, LYS58, GLU53, GLU56, GLU57, PHE133, LYS58, GLY61, GLU60, TYR23, and GLU137. Furthermore, one intriguing halogen contact of fluorine was found with HIS132. The halogen interactions underscore the importance of short-range electrostatics, reflecting the polar nature of halogens and their bond formation capacity with another donor system (Fig. 7). Com. 16 exhibited a total of 19 interactions, including hydrogen bonds, carbon–hydrogen bonds, π–sigma, π–sulfur, π–lone pair, π–π T-shaped, and π–alkyl interactions with amino acids LYS58, ARG64, LYS143, LYS143, GLU63, GLU63, LYS139, ARG64, GLU60, GLU60, GLU136, TYR23, ARG64, PHE133, ARG64, LYS67, LEU140, LEU140, and LYS139. Furthermore, one intriguing sulfur contact of fluorine was also found with TYR23. The molecular docking scores and interactions of ligands Com. 10 and Com. 16 with MCF7 (PDB ID: 6NEW) are listed in Table 4.

Table 4.

Download as

CSV

XLSX

The molecular docking scores and interaction of ligands Com.10 and Com.16 with MCF7.

Structure	Amino acid	Distance	Category	Interaction type	Mol Dock Score	H-B
Structure	Amino acid	(A)	Category	Interaction type	Mol Dock Score	H-B
	LYS58	2.03951	H-B	Conv-H-B	-163.385	-11.628
	LYS58	2.48834	H-B	Conv-H-B
	GLU53	2.4754	H-B	Conv-H-B
	GLU56	2.52664	H-B	Conv-H-B
	GLU57	2.19955	H-B	Conv-H-B
	PHE133	1.75813	H-B	Conv-H-B
	LYS58	2.97071	H-B	Conv-H-B
	GLY61	2.29203	H-B	Carbon H-B
	GLU60	2.62038	H-B	Carbon H-B
	TYR23	2.12039	H-B	Carbon H-B
	GLU137	2.94131	H-B	Carbon H-B
Com. 10	HIS132	3.31023	Hal.	F
	GLU57	3.26092	H-B	π-Donor H-B
	GLU57	2.87295	Other	π-Lone Pair
	LYS58	2.36096	H-B	Conv-H-B	-191.721	-13.002
	ARG64	2.37134	H-B	Conv-H-B
	LYS143	2.76044	H-B	Conv-H-B
	LYS143	2.61637	H-B	Conv-H-B
	GLU63	1.7465	H-B	Conv-H-B
	GLU63	1.95269	H-B	Conv-H-B
	LYS139	1.98333	H-B	Conv-H-B
	ARG64	2.70225	H-B	Carbon H-B
	GLU60	2.14635	H-B	Carbon H-B
	GLU60	2.21068	Hydroph.	π-Sigma
	GLU136	2.37183	Hydroph.	π-Sigma
	TYR23	5.54103	Other	π-Sulfur
	ARG64	2.32158	Other	π-Lone Pair
	PHE133	5.49409	Hydroph.	π- π T-shaped
Com. 16	ARG64	4.82842	Hydroph.	π-Alkyl
	LYS67	5.37618	Hydroph.	π-Alkyl
	LEU140	4.5043	Hydroph.	π-Alkyl
	LEU140	3.761	Hydroph.	π-Alkyl
	LYS139	5.43276	Hydroph.	π-Alkyl

Table 5.

Download as

CSV

XLSX

The molecular docking scores and interaction of ligands Com.10 and Com.16 with A549.

Structure	Amino acid	Distance	Category	Interaction type	Mol Dock Score	H-B
Structure	Amino acid	(A)	Category	Interaction type	Mol Dock Score	H-B
	LYS114	1.81547	H-B	Conv-H-B	-170.099	-10.1334
	SER153	2.86114	H-B	Conv-H-B
	MET108	2.1401	H-B	Conv-H-B
	THR110	1.84786	H-B	Conv-H-B
	ASN154	2.06793	H-B	Conv-H-B
	GLU33	1.76549	H-B	Conv-H-B
	ALA35	2.96513	H-B	Carbon H-B
	LYS114	3.07598	H-B	Carbon H-B
	GLY34	2.15675	H-B	Carbon H-B
Com. 10	LYS114	4.09629	Hydroph.	Alkyl
	ALA35	4.60344	Hydroph.	π-Alkyl
	LYS54	1.88794	H-B-Hal.	Conv-H-B-F	-187.661	-14.2654
	ARG67	2.40994	H-B	Conv-H-B
	LYS151	2.2787	H-B	Conv-H-B
	LYS151	2.51907	H-B	Conv-H-B
	LYS151	2.75789	H-B	Conv-H-B
	GLY37	2.70501	H-B	Conv-H-B
	MET38	3.03441	H-B	Conv-H-B
	GLU71	2.10816	H-B	Conv-H-B
	GLU71	2.44728	H-B	Conv-H-B
	GLY34	2.37648	H-B	Conv-H-B
	GLU71	3.42735	H-B- Hal.	Carbon H-B-F
	TYR36	2.95815	Hydroph.	π-Donor H-B
	LYS54	3.15883	Hydroph.	π-Donor H-B
	TYR113	4.47795	Hydroph.	π-π T-shaped
Com. 16	ALA35	5.38219	Hydroph.	Alkyl
	ALA35	3.62479	Hydroph.	Alkyl
	ALA35	4.94274	Hydroph.	Alkyl
	LYS54	4.17284	Hydroph.	Alkyl
	ILE56	4.11395	Hydroph.	Alkyl

Table 6.

Download as

CSV

XLSX

The molecular docking scores and interaction of ligands Com.1- Com.16 with PC3.

Structure	Amino acid	Distance	Category	Interaction type	Mol Dock Score	H-B
Structure	Amino acid	(A)	Category	Interaction type	Mol Dock Score	H-B
	LEU216	2.93771	H-B-Hal.	Conv-H-B-F	-189.399	-6.25661
	LEU216	2.09129	H-B	Conv-H-B
	ASP217	2.64494	H-B-Hal.	Conv-H-B-F
	GLY481	2.03213	H-B	Conv-H-B
	PHE304	2.84701	H-B	Conv-H-B
	ILE369	2.79252	H-B	Conv-H-B
	LEU210	2.59839	H-B	Carbon H-B
	GLY481	2.1127	H-B	Carbon H-B
	GLY481	1.66432	H-B	Carbon H-B
	ASP217	3.57624	Hal.	F
	PHE220	4.93396	Hydroph.	π-π Stacked
	PHE220	4.28801	Hydroph.	π-πStacked
Com. 10	PHE241	4.21739	Hydroph.	π-Alkyl
	PHE304	4.37551	Hydroph.	π-Alkyl
	LEU482	4.13431	Hydroph.	π-Alkyl
	PRO242	5.0414	Hydroph.	π-Alkyl
	LEU216	2.35059	H-B	Conv-H-B	-221.078	-7.86421
	GLU308	2.07382	H-B	Conv-H-B
	THR309	2.17966	H-B	Conv-H-B
	GLY109	1.68235	H-B	Conv-H-B
	ILE369	2.06542	H-B	Conv-H-B
	ILE369	2.34983	H-B	Conv-H-B
	PHE304	1.80493	H-B	Conv-H-B
	PHE304	2.72547	H-B	Conv-H-B
	PHE108	2.3283	H-B-Hal.	Carbon H-B-F
	PHE108	2.54019	H-B	Carbon H-B
	THR224	2.84776	H-B	Carbon H-B
	GLY481	2.64411	H-B	Carbon H-B
	ILE223	2.81007	H-B	Carbon H-B
	PHE304	1.73239	H-B	Carbon H-B
	PRO107	2.26226	Hal.	F
	GLU308	3.46764	Hal.	F
	PHE57	5.65399	Other	π-Sulfur
Com. 16	ILE223	4.45239	Hydroph.	π-Alkyl
	ARG106	4.31909	Hydroph.	π-Alkyl
	ARG107	4.5699	Hydroph.	π-Alkyl
	LEU482	4.68019	Hydroph.	π-Alkyl
	LEU48	4.63407	Hydroph.	π-Alkyl

Table 7.

Download as

CSV

XLSX

Physicochemical properties of tested gemcitabine derivatives.

St. NO	M. W.	H-B-A	H-B-D	TPSA	ILOGP	H-A	R-B	Bio-S	GI	BBB
Com.7	468.44	9	3	150.54	1.19	34	6	0.55	Low	No
Com.9	834.7	21	6	366.09	1.69	59	12	0.17	Low	No
Com.10	882.75	21	6	366.09	1.24	63	12	0.17	Low	No
Com.14	546.49	13	4	234.53	1.77	38	7	0.17	Low	No
Com.15	804.72	18	6	308.4	2.73	58	9	0.17	Low	No
Com.16	840.73	20	6	350.92	2.1	59	10	0.17	Low	No

Figure 7.

Visualization by Discovery Studio. a. Three-dimensional view of Com. 10 with its hydrophobic site map; b. Two-dimensional view of Com. 10 docked with Vav1 (PDB ID: 6NEW) from the MCF7 breast cancer cell line and type of interaction.

Inspecting cell line A549, the protein 4ZXT among all compounds, Com. 7, Com. 9, Com. 10, and Com. 16 stood out significantly, displaying MolDock scores of −163.136, −160.115, −170.009, and −187.661, respectively, suggesting their potential. Firstly, Com. 7 displays 19 contact points with LYS114, MET108, THR110, GLU33, GLY34, GLU33, ILE31, ALA35, ALA35, TYR36, LYS54, and LYS114. They were engaged in hydrogen bonding. In contrast, LEU156, VAL39, LYS54, ALA35, LYS54, LYS54, and ILE56 are hydrophobic contacts (Fig. 8). Secondly, Com. 9 has 17 interactions; LYS54, LYS151, THR68, LYS55, GLY37, LYS55, ASP111, LYS54, GLU71, and ASP111 are involved in hydrogen bonding. GLU33 and GLU71 are bonded with halogen; the rest of ALA35, ALA35, LYS54, and ILE56 are hydrophobic in nature (Fig. 9). On the other hand, Com. 10 has only 11 interactions and possesses the high MolDock score of −170.009 because all the interactions are hydrogen bonded and two are hydrophobic, displaying its stability (Fig. 10). Lastly, Com. 16 has 19 interactions LYS54, ARG67, LYS151, LYS151, LYS151, GLY37, MET38, GLU71, GLU71, GLY34, GLU71, TYR36, LYS54, TYR113, ALA35, ALA35, ALA35, LYS54, and ILE56 and possesses the highest MolDock score of −187.661 because all the interactions are hydrogen bonded and two are hydrophobic, displaying its stability. The molecular docking scores and interactions of ligands Com. 10 and Com. 16 with MCF7 (PDB ID: 4ZXT) are listed in Table 5.

Figure 8.

Visualization by Discovery Studio; a. Three-dimensional view of Com. 7 with its hydrophobic site map; b. Two-dimensional view of Com. 7 docked with ERK2 kinase (PDB ID: 4ZXT) from the A549 lung cancer cell line and type of interaction.

Figure 9.

Visualization by Discovery Studio; a. Three-dimensional view of Com. 9 with its hydrophobic site map; b. Two-dimensional view of Com. 9 docked with ERK2 kinase (PDB ID: 4ZXT) from the A549 lung cancer cell line and type of interaction.

Figure 10.

Visualization by Discovery Studio; a. Three-dimensional view of Com. 10 with its hydrophobic site map; b. Two-dimensional view of Com. 10 docked with ERK2 kinase (PDB ID: 4ZXT) from the A549 lung cancer cell line and type of interaction.

Examination of cell line PC3, the protein 7LXL, displays notable interactions with Com. 2, Com. 8, Com. 16, and Com. 17 with MolDock scores of −161.155, −170.469, −167.632, −163.718, and −221.078, respectively. Initially, Com. 1 showed five contact points with ALA370, ASP76, ASP76, PHE57, and ILE223. The first three are hydrogen bonds, and the last two are hydrophobic in nature (Fig. 11). Subsequently, ARG440, ILE118, SER119, and ASN441 are hydrogen bonded, and ALA305, ILE443, ALA305, ALA448, ILE369, CYS442, and ALA448 are hydrophobic as π–alkyl type interactions (Fig. 12). Com. 14 has GLU308, ILE369, LEU216, PHE304, GLY481, PHE304, and GLU308 amino acids involved in hydrogen bonding. PHE220, LEU482, LEU482, LEU482, and LEU216 residues are hydrophobic, being π–alkyl and π–π T-shaped type interactions (Fig. 13). Additionally, ARG372, GLU374, VAL240, LYS209, LEU210, LEU210, PRO242, VAL240, LYS209, ASP76, and LYS209 are hydrogen-bonded interactions for Com. 15. ASP76 and LYS209 are halogen contact points, making this complex more stable. PHE220, LEU482, LEU210, PHE57, PHE304, and LEU216 are hydrophobic-type interactions (Fig. 14). Finally, LEU216, GLU308, THR309, GLY109, ILE369, ILE369, PHE304, and PHE304 amino acids are involved in hydrogen bonding, but PHE108, PRO107, and GLU308 are fluorine contact points, making this complex more stable. PHE108, THR224, GLY481, and ILE223 are hydrogen-bonded interactions. Furthermore, one intriguing sulfur contact of fluorine is also found with PHE57. Amino acids ILE223, ARG106, ARG107, LEU482, and LEU48 residues are hydrophobic, being of the π–alkyl type. The molecular docking scores and interactions of ligands Com. 10 and Com. 16 with MCF7 (PDB ID: 7LXL) are listed in Table 6.

Figure 11.

Visualization by Discovery Studio; a. Three-dimensional view of Com. 1 with its hydrophobic site map; b. Two-dimensional view of Com. 1 docked with CYP3A4 enzyme (PDB ID: 7LXL) from the PC3 prostate cancer cell line and type of interaction.

Figure 12.

Visualization by Discovery Studio; a. Three-dimensional view of Com. 7 with its hydrophobic site map; b. Two-dimensional view of Com. 7 docked with CYP3A4 enzyme (PDB ID: 7LXL) from the PC3 prostate cancer cell line and type of interaction.

Figure 13.

Visualization by Discovery Studio; a. Three-dimensional view of Com. 14 with its hydrophobic site map; b. Two-dimensional view of Com. 14 docked with CYP3A4 enzyme (PDB ID: 7LXL) from the PC3 prostate cancer cell line and type of interaction.

Figure 14.

Visualization by Discovery Studio; a. Three-dimensional view of Com. 15 with its hydrophobic site map; b. Two-dimensional view of Com. 15 docked with CYP3A4 enzyme (PDB ID: 7LXL) from the PC3 prostate cancer cell line and type of interaction.

Docking simulations indicated that gemcitabine had stable binding to the main ERK2 residues (LYS54, GLU71, ASP111) in the ATP binding pocket (RMSD 1.83 Å), which could imply the inhibition of kinase activity (Yoshimi Nakajima et al. 2012; Subhani and Jamil 2015). This is further demonstrated by the fact that ERK2 phosphorylation was suppressed in NSCLC cells in vitro. Gemcitabine also interacted closely with LEU216 and PHE304 of the active site of CYP3A4 as predicted, and the drug is a known CYP3A4 metabolic clearance substrate, showing its interaction and CYP3A4 metabolism (Subhani and Jamil 2015).

ADME prediction studies

During the drug discovery process, it is imperative to investigate the variables of absorption, distribution, metabolism, and excretion (ADME) to minimize the risk of pharmacokinetics-related clinical failure. With their molecular weights more than 500 (MW > 500), except for Com. 7 (MW = 468.44), IlogP values ranging from 1.19 to 2.73 (IlogP < 5), number of hydrogen bond acceptors (H-B-As) ranging from 9 to 21 (HBA ≥ 10), and number of hydrogen bond donors (HBDs) ranging from 3 to 6 (HBDs ≥ 5) (Lipinski et al. 2001), ADME prediction using SwissADME software (http://www.swissadme.ch) revealed that not all compounds follow Lipinski’s Rule of Five. In addition, according to SwissADME predictions, the compounds were found to be blood–brain barrier (BBB)-impermeable and poorly absorbed from the gastrointestinal tract (GIT). ADME predictions for some gemcitabine derivatives are shown in Table 7. The most cytotoxic compounds in this study, Com. 10 and Com. 16, were analyzed for drug-likeness applying Lipinski’s Rule of Five, which predicts the oral bioavailability of a compound based on specific physicochemical features. While Com. 10 (882.75 g/mol) and Com. 16 (840.73 g/mol) are above the traditional Lipinski MW cutoff (<500 g/mol), their prodrug strategies (e.g., PEGylation, triazole conjugates, etc.) may enable alternative administration strategies (e.g., IV or nanoparticle delivery). Moreover, these compounds did not contravene Lipinski’s rules based on hydrogen bond donors (≤5), hydrogen bond acceptors (≤10), or logP (≤5), which suggests a good balance between hydrophilicity and lipophilicity. This agreement indicates that Com. 10 and Com. 16 exhibit promising drug-like characteristics, which increase their chances of serving as lead molecules for subsequent optimization. Slight variations in the P-field for other derivatives require optimization toward better pharmacokinetics.

Conclusion

In summary, sixteen gemcitabine derivatives, including three heterocyclic moieties, are 1,2,3-triazoline, five compounds of 1,2,3-triazole, and eight tetrazole compounds. All these compounds were prepared from previous studies using gemcitabine as a starting material. In this study, we evaluated them for their anticancer activities. In addition, compounds Com. 10 and Com. 16 showed potent anticancer activity against all tested cell lines. The compound Com. 16 exhibited potent anticancer activity with IC₅₀ = 12.23, 6.599, and 21.57 µM against all MCF7, A549, and PC3 compared with the reference drug gemcitabine. Whereas, Com. 10 exhibited potent anticancer activity with IC₅₀ = 9.454, 6.931, and 12.09 µM against all MCF7, A549, and PC3 compared with the reference drug gemcitabine. Moreover, other compounds, Com. 9 and Com. 14, also exhibited potent anticancer activity against some tested cell lines; Com. 9 showed potent anticancer activity with IC₅₀ = 29.42 µM against the A549 cell line. On the other hand, Com. 9 and Com. 14 showed potent anticancer activity with IC₅₀ = 39.39 and 36.29 µM against the PC3 cell line, respectively. Molecular docking showed that the most suitable targets for anticancer efficacy are dihydrofolate reductase protein from cell lines, vascular endothelial growth factor receptor, and histone deacetylase. Compound Com. 16, the most effective anticancer, showed good interactions against 6NEW, 4ZXT, and 4LXT with affinities of −191.721, −187.661, and −221.078 kcal/mol, respectively. Especially, this compound showed electrostatic carbon–hydrogen bonds with halogen and hydrophobic interactions, hydrophobic being π–sigma, π–sulfur, π–lone pair, π–π T-shaped, and π–alkyl type, that resemble the co-crystallization ligand and reference drugs. Other compounds showed good docking results; the threshold is set to be −160 to −180. Only eight protein–ligand complexes emerged as lead based on their score. The ADME profile was evaluated for the six most active compounds in comparison to gemcitabine as a reference drug.

Acknowledgement

The authors are greatly thankful to the Al-Kafeel Center for Medical and Pharmaceutical Sciences staff for carrying out the anticancer activity tests and for the assistance provided in completing this study.

Additional information

Conflict of interest

The authors have declared that no competing interests exist.

Ethical statements

The authors declared that no clinical trials were used in the present study.

The authors declared that no experiments on humans or human tissues were performed for the present study.

The authors declared that no informed consent was obtained from the humans, donors or donors’ representatives participating in the study.

The authors declared that no experiments on animals were performed for the present study.

The authors declared that no commercially available immortalised human and animal cell lines were used in the present study.

Use of AI

No use of AI was reported.

Funding

This work was supported by AlKafeel Center for Medical and Pharmaceutical Sciences.

Author contributions

All authors have contributed equally.

Author ORCIDs

Ali Jabbar Radhi https://orcid.org/0000-0002-7578-1716

Data availability

All of the data that support the findings of this study are available in the main text or Supplementary Information.

References

Aslan JE (2019) Platelet Rho GTPase regulation in physiology and disease. Platelets 30: 17–22. https://doi.org/10.1080/09537104.2018.1475632

Azmal M, Paul JK, Prima FS, Talukder OF, Ghosh A (2024) An in silico molecular docking and simulation study to identify potential anticancer phytochemicals targeting the RAS signaling pathway. PLOS ONE 19: e0310637. https://doi.org/10.1371/journal.pone.0310637

Bender DM, Bao J, Dantzig AH, Diseroad WD, Law KL, Magnus NA, Peterson JA, Perkins EJ, Pu YJ, Reutzel-Edens SM, Remick DM, Starling JJ, Stephenson GA, Vaid RK, Zhang D, McCarthy JR (2009) Synthesis, crystallization, and biological evaluation of an orally active prodrug of gemcitabine. Journal of Medicinal Chemistry 52: 6958–6961. https://doi.org/10.1021/jm901181h

Ali FN, Ammar KM, Dheyaa H M, Ehab KO, Ali JR (2020) Synthesis of new gemcitabine derivatives linked tetrazole ring as antibacterial activity. International Journal of Psychosocial Rehabilitation 20: 1806–1813. https://doi.org/10.37200/IJPR/V24I5/PR201852

Cappuzzo F (2009) Gemcitabine for the treatment of advanced nonsmall cell lung cancer. OncoTargets and Therapy 2: 209. https://doi.org/10.2147/OTT.S4645

Carmichael J, Fink U, Russell R, Spittle M, Harris A, Spiessi G, Blatter J (1996) Phase II study of gemcitabine in patients with advanced pancreatic cancer. British Journal of Cancer 73: 101–105. https://doi.org/10.1038/bjc.1996.18

Dhillon AS, Hagan S, Rath O, Kolch W (2007) MAP kinase signalling pathways in cancer. Oncogene 26: 3279–3290. https://doi.org/10.1038/sj.onc.1210421

Ewald B, Sampath D, Plunkett W (2008) Nucleoside analogs: molecular mechanisms signaling cell death. Oncogene 27: 6522–6537. https://doi.org/10.1038/onc.2008.316

Farrell JJ, Elsaleh H, Garcia M, Lai R, Ammar A, Regine WF, Abrams R, Benson AB, Macdonald J, Cass CE, Dicker AP, Mackey JR (2009) Human equilibrative nucleoside transporter 1 levels predict response to gemcitabine in patients with pancreatic cancer. Gastroenterology 136: 187–195. https://doi.org/10.1053/j.gastro.2008.09.067

Ferrazzi E, Stievano L (2006) Gemcitabine: monochemotherapy of breast cancer. Annals of Oncology 17: v169–v172. https://doi.org/10.1093/annonc/mdj975

Ghasemi M, Liang S, Luu QM, Kempson I (2023) The MTT Assay: A Method for Error Minimization and Interpretation in Measuring Cytotoxicity and Estimating Cell Viability. In: Friedrich O, Gilbert DF (Eds) Cell Viability Assays. Methods in Molecular Biology, vol 2644. Humana, New York, NY. https://doi.org/10.1007/978-1-0716-3052-5_2

Hoang T, Kim K, Jaslowski A, Koch P, Beatty P, McGovern J, Quisumbing M, Shapiro G, Witte R, Schiller JH (2003) Phase II study of second-line gemcitabine in sensitive or refractory small cell lung cancer. Lung Cancer 42: 97–102. https://doi.org/10.1016/S0169-5002(03)00273-3

Immordino ML, Brusa P, Rocco F, Arpicco S, Ceruti M, Cattel L (2004) Preparation, characterization, cytotoxicity and pharmacokinetics of liposomes containing lipophilic gemcitabine prodrugs. Journal of Controlled Release 100: 331–346. https://doi.org/10.1016/j.jconrel.2004.09.001

Itoi T, Sofuni A, Fukushima N, Itokawa F, Tsuchiya T, Kurihara T, Moriyasu F, Tsuchida A, Kasuya K (2007) Ribonucleotide reductase subunit M2 mRNA expression in pretreatment biopsies obtained from unresectable pancreatic carcinomas. Journal of Gastroenterology 42: 389–394. https://doi.org/10.1007/s00535-007-2017-0

Iwakiri S, Sonobe M, Nagai S, Hirata T, Wada H, Miyahara R (2008) Expression status of folate receptor α is significantly correlated with prognosis in non-small-cell lung cancers. Annals of Surgical Oncology 15: 889–899. https://doi.org/10.1245/s10434-007-9755-3

Kaaki K, Hervé-Aubert K, Chiper M, Shkilnyy A, Soucé M, Benoit R, Paillard A, Dubois P, Saboungi M-L, Chourpa I (2012) Magnetic nanocarriers of doxorubicin coated with poly(ethylene glycol) and folic acid: relation between coating structure, surface properties, colloidal stability, and cancer cell targeting. Langmuir 28: 1496–1505. https://doi.org/10.1021/la2037845

Kalli KR, Oberg AL, Keeney GL, Christianson TJH, Low PS, Knutson KL, Hartmann LC (2008) Folate receptor alpha as a tumor target in epithelial ovarian cancer. Gynecologic Oncology 108: 619–626. https://doi.org/10.1016/j.ygyno.2007.11.020

Koolen SLW, Witteveen PO, Jansen RS, Langenberg MHG, Kronemeijer RH, Nol A, Garcia-Ribas I, Callies S, Benhadji KA, Slapak CA, Beijnen JH, Voest EE, Schellens JHM (2011) Phase i study of oral gemcitabine prodrug (LY2334737) alone and in combination with erlotinib in patients with advanced solid tumors. Clinical Cancer Research 17: 6071–6082. https://doi.org/10.1158/1078-0432.CCR-11-0353

Krzywik J, Mozga W, Aminpour M, Janczak J, Maj E, Wietrzyk J, Tuszyński JA, Huczyński A (2020) Synthesis, antiproliferative activity and molecular docking studies of novel doubly modified colchicine amides and sulfonamides as anticancer agents. Molecules 25: 1789. https://doi.org/10.3390/molecules25081789

Leskelä S, Honrado E, Montero-Conde C, Landa I, Cascón A, Letón R, Talavera P, Cózar JM, Concha A, Robledo M, Rodríguez-Antona C (2007) Cytochrome P450 3A5 is highly expressed in normal prostate cells but absent in prostate cancer. Endocrine-Related Cancer 14: 645–654. https://doi.org/10.1677/ERC-07-0078

Lipinski CA, Lombardo F, Dominy BW, Feeney PJ (2001) 46 Advanced Drug Delivery Reviews Experimental and computational approaches to estimate solubility and permeability in drug discovery and development q settings. www.elsevier.com/locate/drugdeliv.

Liu Y, Tarsounas M, O’Regan P, West SC (2007) Role of RAD51C and XRCC3 in Genetic Recombination and DNA Repair. Journal of Biological Chemistry 282: 1973–1979. https://doi.org/10.1074/jbc.M609066200

Mohammed MJ, Oleiwi ZK, Al-Hilo MAH, Mubarak AKH, Obaid EK, Radhi AJ (2020) Synthesis and study biological activity of gemcitabine linked heterocyclic hybrids. Research Journal of Pharmacy and Technology 13: 3257. https://doi.org/10.5958/0974-360X.2020.00577.6

Nakano Y, Tanno S, Koizumi K, Nishikawa T, Nakamura K, Minoguchi M, Izawa T, Mizukami Y, Okumura T, Kohgo Y (2007) Gemcitabine chemoresistance and molecular markers associated with gemcitabine transport and metabolism in human pancreatic cancer cells. British Journal of Cancer 96: 457–463. https://doi.org/10.1038/sj.bjc.6603559

Naser AF, Madlool AK, Mohsin DH, Obaid EK, Radhi AJ (2020) Synthesis of new gemcitabine derivatives linked tetrazole ring as antibacterial activity. International Journal of Psychosocial Rehabilitation 20: 1806–1813. https://doi.org/10.37200/IJPR/V24I5/PR201852

Ozols R (2005) Gemcitabine and carboplatin in second-line ovarian cancer. Seminars in Oncology 32: 4–8. https://doi.org/10.1053/j.seminoncol.2005.06.023

Pace E, Melis M, Siena L, Bucchieri F, Vignola AM, Profita M, Gjomarkaj M, Bonsignore G (2000) Effects of gemcitabine on cell proliferation and apoptosis in non-small-cell lung cancer (NSCLC) cell lines. Cancer Chemotherapy and Pharmacology 46: 467–476. https://doi.org/10.1007/s002800000183

Pasut G, Veronese FM (2009) PEG conjugates in clinical development or use as anticancer agents: An overview. Advanced Drug Delivery Reviews 61: 1177–1188. https://doi.org/10.1016/j.addr.2009.02.010

Pasut G, Veronese FM (2012) State of the art in PEGylation: The great versatility achieved after forty years of research. Journal of Controlled Release 161: 461–472. https://doi.org/10.1016/j.jconrel.2011.10.037

Radhi AW, AL-Yasiry ANT, Radhi AJ (2021) Synthesis of new tetrazole derivatives as potential antibacterial agents. International Journal of Pharmaceutical Research 13: 2931–2935. https://doi.org/10.31838/ijpr/2021.13.01.330

Robinson BW, Ostruszka L, Im MM, Shewach DS (2004) Promising combination therapies with gemcitabine. Seminars in Oncology 31: 2–12. https://doi.org/10.1053/j.seminoncol.2004.03.021

Salman FW, Twayej AJ, Shaheed HA, Radhi AJ (2019) New Gemcitabine Derivatives as Potent in vitro α-Glucosidase Inhibitors. Nano Biomedicine and Engineering 11(1): 84–90. https://doi.org/10.5101/nbe.v11i1.p84-90

Seyedi S, Teo R, Foster L, Saha D, Mina L, Northfelt D, Anderson KS, Shibata D, Gatenby R, Cisneros L, Troan B, Anderson ARA, Maley CC (2023) Testing adaptive therapy protocols using gemcitabine and capecitabine on a mouse model of endocrine-resistant breast cancer. https://doi.org/10.1101/2023.09.18.558136

Shmeeda H, Amitay Y, Gorin J, Tzemach D, Mak L, Ogorka J, Kumar S, Zhang JA, Gabizon A (2010) Delivery of zoledronic acid encapsulated in folate-targeted liposome results in potent in vitro cytotoxic activity on tumor cells. Journal of Controlled Release 146: 76–83. https://doi.org/10.1016/j.jconrel.2010.04.028

Subhani S, Jamil K (2015) Molecular docking of chemotherapeutic agents to CYP3A4 in non-small cell lung cancer. Biomedicine & Pharmacotherapy 73: 65–74. https://doi.org/10.1016/j.biopha.2015.05.018

Sung H, Ferlay J, Siegel RL, Laversanne M, Soerjomataram I, Jemal A, Bray F (2021) Global cancer statistics 2020: GLOBOCAN Estimates of incidence and mortality worldwide for 36 cancers in 185 countries. CA: A Cancer Journal for Clinicians 71: 209–249. https://doi.org/10.3322/caac.21660

Thakkar S, Hutson T, Garcia J, Rothaermal J, Bart M, Dreicer R (2006) A phase II trial of gemcitabine and docetaxel in hormone-refractory metastatic prostate cancer. Journal of Clinical Oncology 24: 14501–14501. https://doi.org/10.1200/jco.2006.24.18_suppl.14501

Thomsen R, Christensen MH (2006) MolDock: A new technique for high-accuracy molecular docking. Journal of Medicinal Chemistry 49: 3315–3321. https://doi.org/10.1021/jm051197e

Torres Robles J, Stiegler AL, Boggon TJ, Turk BE (2025) Cancer hotspot mutations rewire ERK2 specificity by selective exclusion of docking interactions. Journal of Biological Chemistry 301: 108348. https://doi.org/10.1016/j.jbc.2025.108348

Tsume Y, Borras Bermejo B, Amidon G (2014a) The dipeptide monoester prodrugs of floxuridine and gemcitabine—feasibility of orally administrable nucleoside analogs. Pharmaceuticals 7: 169–191. https://doi.org/10.3390/ph7020169

Tsume Y, Incecayir T, Song X, Hilfinger JM, Amidon GL (2014b) The development of orally administrable gemcitabine prodrugs with d-enantiomer amino acids: Enhanced membrane permeability and enzymatic stability. European Journal of Pharmaceutics and Biopharmaceutics 86: 514–523. https://doi.org/10.1016/j.ejpb.2013.12.009

Vandana M, Sahoo SK (2010) Long circulation and cytotoxicity of PEGylated gemcitabine and its potential for the treatment of pancreatic cancer. Biomaterials 31: 9340–9356. https://doi.org/10.1016/j.biomaterials.2010.08.010

Varani K, Gessi S, Merighi S, Iannotta V, Cattabriga E, Pancaldi C, Cadossi R, Borea PA (2003) Alteration of A3 adenosine receptors in human neutrophils and low frequency electromagnetic fields. Biochemical Pharmacology 66: 1897–1906. https://doi.org/10.1016/S0006-2952(03)00454-4

Ventura S, Serrano L (2004) Designing proteins from the inside out. Proteins: Structure, Function, and Bioinformatics 56: 1–10. https://doi.org/10.1002/prot.20142

Wu Z, Xia F, Lin R (2024) Global burden of cancer and associated risk factors in 204 countries and territories, 1980–2021: a systematic analysis for the GBD 2021. Journal of Hematology & Oncology 17: 119. https://doi.org/10.1186/s13045-024-01640-8

Xiong G, Wu Z, Yi J, Fu L, Yang Z, Hsieh C, Yin M, Zeng X, Wu C, Lu A, Chen X, Hou T, Cao D (2021) ADMETlab 2.0: an integrated online platform for accurate and comprehensive predictions of ADMET properties. Nucleic Acids Research 49: W5–W14. https://doi.org/10.1093/nar/gkab255

Yang S-J, Lin F-H, Tsai K-C, Wei M-F, Tsai H-M, Wong J-M, Shieh M-J (2010) Folic Acid-Conjugated Chitosan Nanoparticles Enhanced Protoporphyrin IX Accumulation in Colorectal Cancer Cells. Bioconjugate Chemistry 21: 679–689. https://doi.org/10.1021/bc9004798

Yardley D (2005) Gemcitabine Plus Paclitaxel in Breast Cancer. Seminars in Oncology 32: 14–21. https://doi.org/10.1053/j.seminoncol.2005.06.025

Zhang J, Wan L, Dai X, Sun Y, Wei W (2014) Functional characterization of Anaphase Promoting Complex/Cyclosome (APC/C) E3 ubiquitin ligases in tumorigenesis. Biochimica et Biophysica Acta (BBA) - Reviews on Cancer 1845: 277–293. https://doi.org/10.1016/j.bbcan.2014.02.001

Zhu C, Hu H, Ma Y, Xiong S, Zhu D (2023) Vav1‐dependent Rac1 activation mediates hypoxia‐induced gemcitabine resistance in pancreatic ductal adenocarcinoma cells through upregulation of HIF‐1α expression. Cell Biology International 47: 1835–1842. https://doi.org/10.1002/cbin.12074

Supplementary material

Supplementary material 1

Supplamantary data

Ali Jabbar Radhi, Mahmood M. Fahad, Ihsan Alrubaie, Nusrat Shafiq, Kinza Afzal

Data type: docx

This dataset is made available under the Open Database License (http://opendatacommons.org/licenses/odbl/1.0). The Open Database License (ODbL) is a license agreement intended to allow users to freely share, modify, and use this Dataset while maintaining this same freedom for others, provided that the original source and author(s) are credited.

Download file (480.33 kb)

﻿Abstract

Keywords

﻿Introduction

﻿In vitro anticancer activity

﻿ADME predictions

﻿In silico molecular docking studies

﻿Target selection rationale

﻿Cellular functions and biological roles of selected targets

﻿Ligand preparation

﻿Protein preparation

﻿Protein–ligand docking

﻿Results and discussion

﻿Analyzing the cytotoxicity of a normal cell line in vitro

﻿Cytotoxicity and selectivity indices in normal cell lines

﻿Molecular docking analysis

﻿ADME prediction studies

﻿Conclusion

﻿Acknowledgement

﻿Additional information

﻿References

Supplementary material

Abstract

Introduction

In vitro anticancer activity

ADME predictions

In silico molecular docking studies

Target selection rationale

Cellular functions and biological roles of selected targets

Ligand preparation

Protein preparation

Protein–ligand docking

Results and discussion

Analyzing the cytotoxicity of a normal cell line in vitro

Cytotoxicity and selectivity indices in normal cell lines

Molecular docking analysis

ADME prediction studies

Conclusion

Acknowledgement

Additional information

References