Thiazole‐Derived Dual EGFR/CDK‐2 Inhibitors: Rational Design, Synthesis, In Vitro Anticancer Evaluation, Mechanistic Profiling, and Computational Binding Analysis

ABSTRACT

New series of thiazole analogues were designed, synthesized, and tested for their potential anticancer activity. All the synthesized compounds were biologically tested against HCT‐116 and MCF‐7 cell lines. Compounds 14d, 14h, and 14i were determined to be the most active members in this series against HCT‐116 cancer cell lines with a considerable safety profile. The three lead compounds were further investigated for their inhibitory activities against EGFR and CDK‐2. Compound 14i exhibited the most potent inhibition, with IC₅₀ values of 0.056 and 0.215 µM against EGFR and CDK‐2, respectively, suggesting a possible association between kinase inhibition and the observed cellular activity. Compounds 14d, 14h, and 14i induced early apoptosis (14.40%–18.09%) and G₂/M arrest (51.55%–66.15% population) in HCT‐116 cells, with minimal necrosis. Additionally, the screened compounds upregulated pro‐apoptotic factors (Bax, cytochrome c, and cleaved caspase‐3), while downregulating pro‐survival/anti‐apoptotic markers (p‐AKT1, Bcl‐2) and the angiogenic factor VEGF, offering preliminary insight into the potential mechanism of action. EGFR and CDK‐2 were among the most prominent genes identified in the network pharmacology analysis of the tested compounds. Molecular docking and molecular dynamics simulations suggested favorable binding modes of compound 14i within the active sites of EGFR and CDK‐2, with stable interaction patterns observed during the simulation period. Furthermore, compounds 14d, 14h, and 14i showed encouraging in silico ADMET and drug‐likeness profiles. Overall, these findings highlight the thiazole series as promising lead compounds with potential dual inhibitory activity, warranting further optimization and mechanistic validation.

A new series of thiazole derivatives was rationally designed and synthesized. Compounds 14d, 14h, and 14i exhibited potent cytotoxicity in HCT‐116 cells, with 14i showing strong dual EGFR/CDK‐2 inhibition (IC₅₀ = 0.056 and 0.215 µM). These leads induced apoptosis and G₂/M arrest, supported by docking and dynamics simulations for 14i.

1. Introduction

Cancer ranks among the leading causes of morbidity and mortality worldwide, garnering extensive research attention for decades [1]. It is a complex disease characterized by the uncontrolled growth and spread of abnormal cells [2]. Cancer is currently the second leading cause of death worldwide, accounting for one in six fatalities. The International Agency for Research on Cancer (IARC) estimates that there were 10 million cancer‐related deaths and 19.3 million new cases of the disease in 2020 [3]. Conventional chemotherapy remains a cornerstone of cancer treatment [4]. Anticancer agent research dates back to the early 20th century, yet the pursuit of safe, efficacious, and innovative anticancer therapies remains a major research priority [5, 6, 7]. Accordingly, the development of medications targeting specific targets has been a major focus of anticancer drug research, however, resistance to single‐target agents remains a major clinical challenge in targeted cancer therapy. Consequently, designing a single molecule that can interact with multiple targets at once is becoming a greater emphasis on drug research. This could boost efficacy and lower the possibility of drug resistance mutations [8, 9, 10].

Protein kinases are a large family of enzymes that speed up the phosphorylation of proteins. This process is one of the key mechanisms that control many biological processes, including growth, differentiation, cell cycle, apoptosis, motility, and proliferation. Consequently, protein kinases have received considerable attention as promising therapeutic targets for cancer treatment [11, 12, 13]. Most kinase inhibitors are ATP‐competitive heterocycles that directly compete with ATP for the kinase ATP‐binding site. Figure 1 illustrates the five primary domains that make up the ATP‐binding site of protein kinases: (a) the adenine‐binding pocket, which fits the heterocyclic ring; (b) the hydrophobic pocket I; (c) the hydrophobic pocket II; (d) the hydrophilic ribose domain; and (e) the phosphate pocket [14, 15]. EGFR, a member of the tyrosine kinase family, is a transmembrane receptor that plays a crucial role in various cell signaling pathways, including proliferation, apoptosis, angiogenesis, and metastatic migration. EGFR overexpression is a primary driver of the development and progression of multiple carcinomas, including pancreatic, breast, and lung cancers; consequently, substantial therapeutic emphasis has been placed on inhibiting EGFR signaling [16, 17, 18]. Meanwhile, cyclin‐dependent kinase 2 (CDK‐2) is a key mediator of cell cycle regulation and is engaged in several biological processes such as intracellular transport, protein degradation, signal transmission, DNA and RNA metabolism, and translation. Therefore, targeting the cell‐cycle system by suppressing cyclin‐dependent kinases (CDKs) has been regarded as a vital treatment option for cancer [19, 20, 21].

Figure 1.

The architecture of the ATP‐binding pocket of protein kinases: maintained hinge area, adenine binding region, ribose and triphosphate pockets, and both hydrophobic subpockets (I and II) employed for inhibitor selectivity.

This study aims to develop new thiazole‐based compounds that function as dual‐target inhibitors, specifically inhibiting the EGFR and CDK‐2 proteins, which play vital roles in cancer development. The targeted compounds function by blocking both kinases to stop abnormal molecular pathways that lead to tumor development and sustained growth and survival because single‐target drugs face limitations due to rapid resistance development via compensatory pathways or genetic mutations. The desired analogues are designed to act as potential apoptosis inducers for cancer cells through their control of essential apoptotic regulators because they suppress anti‐apoptotic proteins which include vascular endothelial growth factor (VEGF) and serine/threonine kinase AKT‐1 and B‐cell lymphoma 2 (Bcl‐2). At the same time, they boost pro‐apoptotic proteins which include Bcl‐2‐associated X protein (Bax), cytochrome c (Cyto‐c), and caspase‐3. The approach combines kinase inhibition and apoptosis induction, using a polypharmacological approach supported by network pharmacology studies to show that EGFR and CDK‐2 function as vital components in cancer development networks, thereby enhancing therapeutic efficacy, improving selectivity, and reducing toxic effects from non‐targeted interactions in colorectal and breast cancer models.

Simultaneous inhibition of EGFR and CDK‐2 is considered as solid polypharmacological strategy. While EGFR is an upstream receptor tyrosine kinase, it tends to switch on several key mitogenic and pro survival signaling routes, like RAS/RAF/MEK/ERK and PI3K/AKT. At the same time, CDK‐2 acts as a crucial cell cycle regulator, and it really helps steer the G1 to S transition, then it basically pushes S phase progression forward. If we hit both targets together, the idea is that cancer cells lose their ability to keep dividing more effectively, and you can get a more forceful cell cycle arrest and apoptosis. Also, it may lower the chances of resistance showing up, since that often happens when only one pathway is blocked, and then compensatory signaling takes over. A few recent investigations have shown the treatment value of dual EGFR/CDK inhibitors, where synergistic anticancer effects appear in kinase‐addicted tumors [10, 22, 23, 24, 25].

The thiazole heterocycle serves as a privileged scaffold in drug discovery, enabling researchers to develop drugs that exhibit multiple biological effects. Such core structure produces derivatives that demonstrate effective drug properties across various medical fields which include antimicrobial, antitubercular, antihypertensive, analgesic, anti‐inflammatory, anticancer, antioxidant, neuroprotective, antidiabetic, and antiviral effects [26, 27]. To achieve optimal rational design, the key pharmacophoric features shared by the FDA‐approved kinase inhibitor dasatinib and literature‐reported anticancer compounds I–IV—essential for effective ATP‐competitive inhibition of the ATP‐binding sites of EGFR and CDK‐2—were incorporated. These include: terminal hydrophobic substituents to occupy hydrophobic region I; a central thiazole ring capable of forming critical hydrogen bonds with the adenine‐binding region (Met769 and Gln767 in EGFR; Asp145 in CDK‐2); terminal hydrophobic or moderately polar head groups to engage hydrophobic region II and the ribose‐binding domain; and a hydrophilic linker oriented toward the solvent‐exposed front pocket. These structural elements were systematically integrated during the design and synthesis of four novel series (A–D) of thiazole‐containing compounds to evaluate their potential as dual EGFR/CDK‐2 inhibitors. We retained the structural fragments that are critical for tight, ATP‐competitive binding to the kinase domains of EGFR and CDK‐2, while making thoughtful changes to boost potency, selectivity, and overall drug‐like behavior.

In Series A (Figure 2), we utilized the same thiazole scaffold and amide linkers that formed the basis of chemotype I [28], thereby retaining the same critical hinge region interactions that most high‐quality kinase inhibitors require for binding. In addition, we systematically varied each of the peripheral heterocycles and terminal substituents based on their effect on binding both inside of hydrophobic pockets, as well as on solvent exposed regions. Ultimately, our goal was to characterize how those changes impact activity against EGFR and CDK‐2 by applying classic SAR techniques. Series B was designed similarly to Series A; however, we intentionally removed the oxazole ring from chemotype II [29] and replaced it with a thiazole ring. This is an example of how bioisosteric replacements can be leveraged for improved binding affinity; thiazoles have been shown to demonstrate superior hydrogen‐bond accepting capacity due to the presence of sulfur atoms, as well as generally having improved metabolic stability and enhanced π‐stacking. Thus, we believed that removing the oxazole ring and adding the thiazole ring would allow us to maintain the overall shape of the pharmacophore while enhancing the opportunity for each compound to bind to the respective target proteins.

Figure 2.

Rational design approach for new EGFR/CDK2 inhibitors that combined pharmacophoric components of FDA‐approved dasatinib with literature anticancer compounds I–IV via spacer modification, bioisosteric replacement, and systematic terminal substitution.

In Series C, we substituted the oxadiazole ring present in chemotype III [30] with a thiazole to keep the central heterocyclic scaffold that effectively anchors to the adenine‐binding region of kinases. We also made the linker a di‐amide to hopefully provide additional hydrogen bonding, slightly increase polarity, and improve solvation in the front pocket—all things that tend to lead to improved cellular penetration, reduced efflux transporter‐mediated elimination of the compounds, and more potent overall inhibition of kinases. In Series D, we used the existing thiazole core from chemotype IV [31], but modified the amide linker and substituted terminal groups. These changes were intended to improve the steric and electronic fit of the compounds into the hydrophobic pockets (I, II) and the ribose moiety. Overall, this rational for dual targeting of EGFR and CDK‐2 is based on their complementary roles in cancer progression. EGFR regulates key signaling pathways associated with cell proliferation, while CDK‐2 is critically involved in cell cycle transition. Simultaneous modulation of these pathways has been proposed an approach to enhance anticancer efficacy and potentially overcome resistance mechanisms associated with single‐target therapies. Accordingly, the designed thiazole compounds were hypothesized to exhibit dual inhibitory potential against these kinases.

2. Results and Discussion

2.1. Chemistry

Scheme 1 illustrates the synthetic pathways to our main intermediates (primary and secondary amines 2, 5a–c, and 8a–c) under the previously described conditions [32, 33, 34, 35, 36]. 1‐[(4‐Chlorophenyl)sulfonyl]piperazine 2 was obtained in 84% yield by selectively monosulfonylating of piperazine (6.0 equiv.) with 4‐chlorobenzenesulfonyl chloride (1.0 equiv.) 1 at 0°C in dichloromethane. Treatment of 1‐fluoro‐4‐nitrobenzene 3 with morpholine, 1‐acetylpiperazine, or 1‐methylpiperazine (2.0 equiv.) under S_NAr conditions furnished the 4‐amino‐substituted nitrobenzenes 4a–c in 78%–98% yield after precipitation in acetonitrile at 60°C. Using zinc powder and ammonium chloride in methanol/water, the nitro group was reduced under mild conditions, producing the corresponding anilines 5a–c with 52%–57% yield. Furthermore, 4‐substituted anilines 6a–c were acylated with chloroacetyl chloride (1.1 equiv.) using triethylamine in dichloromethane to obtain N‐(4‐substituted phenyl)−2‐chloroacetamides 7a–c in high yields (86%–89%). The primary α‐aminoacetamides 8a–c were obtained in moderate to good yields (52%–77%) through direct S_N2 displacement of the chloro group in the corresponding 2‐chloroacetamides 7a–c, by heating with excess 25% aqueous ammonia in ethanol at 70°C for 12 h.

Scheme 1.

Synthesis of 1‐((4‐chlorophenyl)sulfonyl)piperazine 2, 4‐substituted anilines 5a‐c, and 2‐amino‐N‐(4‐substituted phenyl)acetamides 8a–c. Reagents and conditions: (a) piperazine (6.0 equiv.), DCM, 0°C, 30 min, 84%; (b) morpholine, 1‐acetylpiperazine or 1‐methylpiperazine, acetonitrile, 60°C, overnight, 78%–98%; (c) Zn, NH₄Cl, methanol/H₂O, rt, overnight, 52%–57%; (d) chloroacetyl chloride, Et₃N, DCM, rt, overnight, 86%–98%; (e) ammonium hydroxide, ethanol, 70°C, 12 h, 52%–77%.

Scheme 2 demonstrates the synthetic pathway of the proposed target compounds 14a–m. The benzoic acid derivatives 9a–e were converted to the corresponding primary amides 10a–e in 20%–60% yield via mixed anhydride formation followed by ammonolysis. The acids (1.0 equiv.) were activated with ethyl chloroformate (1.5 equiv.) and triethylamine in THF at 0°C, warmed to room temperature, and then treated with excess 25% aqueous ammonia overnight [37]. These benzamides 10a–e (1.0 equiv.) were then reacted with phosphorous pentasulfide (1.2 equiv.) in tetrahydrofuran at 60°C to give the corresponding benzothioamides 11a–e with a yield of 71%–79%. The benzothioamide intermediates 11a–e underwent Hantzsch‐type cyclization with ethyl bromopyruvate (1.2 equiv.) in ethanol at 90°C, providing the ethyl 2‐arylthiazole‐4‐carboxylate derivatives 12a–e as brown oils in 71%–90% yield [38]. Hydrolysis of the thiazole esters 12a–e was achieved using aqueous NaOH solution in methanol or dioxane, furnishing the desired carboxylic acids 13a–e in 20%–78% yield [39, 40]. The acids 13a–e were subjected to amide coupling with a range of primary and secondary amines (1.0 equiv.) employing EDC (1.5 equiv.) and DMAP (0.2 equiv.) as coupling agents in dichloromethane or dimethylformamide, furnishing the targeted N‐substituted‐2‐arylthiazole‐4‐carboxamides 14a–m in 16%–48% yields.

Comp. No.	14a	14b	14c	14d	14e	14f	14g
R1
R2	H	H	H	H	H	H	H

Comp. No.	14h	14i	14j	14k	14l	14m
R1
R2	H	H	OCH3	Br	Cl	F

Scheme 2.

Synthesis of N‐(substituted)−2‐phenylthiazole‐4‐carboxamides 14a–m. Reagents and conditions: (a) (i) ethyl chloroformate, Et₃N, THF, 0°C, 30 min; (ii) ammonium hydroxide, rt, overnight, 20%–60%; (b) P₂S₅, THF, 60°C, 3 h, 71%–79%; (c) ethyl bromopyruvate, EtOH, 90°C, 6 h, 71%–90%; (d) NaOH, H₂O, MeOH, rt, overnight, 20%–60% for compounds 13a–c or NaOH, H₂O, dioxane, 50°C, 24 h, 70%–78% for compounds 13d–e; (e) 2, 5a–c, 8a–c, phenylpiperazine or morpholine, EDC, DMAP, DCM (14a–j) or DMF (14k–m), rt, overnight, 16%–48%.

The target compound 18 was prepared following the reactions sequence shown in Scheme 3. The synthesis embarked on cyclization of phenacyl bromide 15 (1.0 equiv.) and ethyl 2‐amino‐2‐thioxoacetate (1.5 equiv.) in ethanol at 90°C, which cleanly furnished ethyl 4‐phenylthiazole‐2‐carboxylate 16 in 95% yield [41]. Subsequent hydrolysis of ethyl 4‐phenylthiazole‐2‐carboxylate 16 with aqueous sodium hydroxide solution furnished 4‐phenylthiazole‐2‐carboxylic acid 17 in 49% yield [39]. Amide coupling between acid 17 (1.0 equiv.) and 2‐amino‐N‐phenylacetamide 8a (1.0 equiv.) was achieved using EDC (1.5 equiv.) and DMAP (0.2 equiv.) in dimethylformamide at room temperature, affording the target product 18 in 37% yield.

Scheme 3.

Synthesis of N‐(2‐oxo‐2‐(phenylamino)ethyl)−4‐phenylthiazole‐2‐carboxamide 18. Reagents and conditions: (a) ethyl 2‐amino‐2‐thioxoacetate, EtOH, 90°C, 12 h, 95%; (b) NaOH, H₂O, MeOH, rt, overnight, 49%; (c) 2‐amino‐N‐phenylacetamide 8a, EDC, DMAP, DMF, rt, overnight, 37%.

The designed compounds 22, 26, and 30a–b were obtained through multiple steps shown in Scheme 4. We performed N‐acylation on aminothiazoles 19, 23, and 27a–b using benzoyl chloride at room temperature in tetrahydrofuran with triethylamine and DMAP serving as base and catalyst, respectively which produced N‐thiazolyl benzamides 20, 24, and 28a–b with 69%–84% yield [42]. These benzamides were then hydrolyzed under basic conditions with aqueous sodium hydroxide to give the carboxylic acids 21, 25, and 29a–b in 63%–69% yield [39]. Finally, EDC/DMAP‐mediated amide coupling of these acids (1.0 equiv.) with 2‐amino‐N‐phenylacetamide 8a (1.0 equiv.) in DMF at room temperature delivered the target products 22, 26, and 30a–b in yields ranging from 22% to 44%. The structures of all newly synthesized target compounds were confirmed by melting point determination, mass spectrometry (MS), and detailed ¹H and ¹³C NMR spectroscopy.

Scheme 4.

Synthesis of 4/5‐substituted (thiazol‐2‐yl)benzamides 22, 26, and 30a–b. Reagents and conditions: (a) benzoyl chloride, DMAP, Et₃N, THF, rt, overnight, 69%–84%; (b) NaOH, H₂O, MeOH, rt, overnight, 63%–69%; (c) 2‐amino‐N‐phenylacetamide 8a, EDC, DMAP, DMF, rt, overnight, 22%–44%.

2.2. Biological Evaluation

2.2.1. In Vitro Cytotoxicity Evaluation Against HCT‐116 and MCF‐7 Cancer Cell Lines

All newly synthesized compounds (14a–m, 18, 22, 26, and 30a–b) were initially screened for their antiproliferative activity against two human cancer cell lines—MCF‐7 (breast carcinoma) and HCT‐116 (colorectal carcinoma) as representative models 3‐[4,5‐dimethylthiazol‐2‐yl]‐2,5 diphenyl tetrazolium bromide (MTT) assay [43, 44] at a fixed concentration of 10 µM. HCT‐116 and MCF‐7 are good cell lines to use for initial tests of anticancer drugs targeting EGFR and CDK‐2 because they are well characterized. HCT‐116 cells are documented to overexpress EGFR, are highly dependent on EGFR signaling pathways, and have undergone changes in the regulation of the cell cycle that result in increased sensitivity to CDK‐2 inhibition [45]. Similarly, MCF‐7 cells express both EGFR and CDK‐2, frequently serve as a screening system for compounds that inhibit kinase‐mediated growth of breast cancer cells, and are frequently used to test the effects of drugs on cell cycle progression [46, 47, 48]. Thus, these two cell lines provide a logical yet complementary system for evaluating the potential of the synthesized thiazole derivatives to inhibit both EGFR and CDK‐2.

The growth inhibition percentage (GI %) was used as a measure of cytotoxicity for the synthesized compounds. The outcome of the experiment indicated that compounds 14d, 14h, and 14i showed evidence of promising inhibition percentages against the HCT‐116 cell line, with inhibition percentages ranging from 33.29% up to 35.09% (Figure 3). Compounds 14f and 14m also demonstrated acceptable inhibition, with values of 25.64% and 25.12%, respectively. Additionally, compounds 14c, 14g, 14l, 26, and 30a had adequate levels of GI%, 19.55% against both MCF‐7 and HCT‐116. The compounds that had the highest levels of activity during the initial screening experiments, 14d, 14h, and 14i, were used for the following studies to identify their mechanism of action. The IC₅₀ values (µM) of compounds 14d, 14h, and 14i against HCT‐116 cells were compared to doxorubicin, the positive control (Table 1).

Figure 3.

In vitro percent growth inhibition of the newly synthetized compounds 14a–m, 18, 22, 26, and 30a–b against HCT‐116 and MCF‐7 cancer cell lines.

Table 1.

MTT assay showing the IC₅₀ (µM) of compounds 14d, 14h, and 14i versus doxorubicin in HCT‐116 and WI‐38 cell lines (mean ± SEM; n = 3).

Comp. no.	Measured IC50 (µM)		SI
	WI‐38	HCT‐116
14d	> 100	36.00 ± 1.5	> 2.7
14h	> 100	26.60 ± 2.1	> 3.8
14i	> 100	35.49 ± 2.8	> 2.8
Doxorubicin	77.20 ± 0.77	4.208 ± 0.99	18.34

Note: The safety index (SI) was calculated as the IC₅₀ of the compound (WI‐38)/IC₅₀ of the compound (cancer cell line). Sigmoidal curve for the compounds on HCT‐116 cell line was added in Supporting Information: S2.

Abbreviations: MTT, 3‐[4,5‐dimethylthiazol‐2‐yl]‐2,5 diphenyl tetrazolium bromide; SEM, standard error of mean.

2.2.2. Selectivity Profiling: Cytotoxicity Against Normal Cells

Based on our initial cytotoxicity screening, compounds 14d, 14h, and 14i stood out as the most active and were therefore carried forward for more detailed evaluation. Their IC₅₀ values against HCT‐116 colorectal carcinoma cells were determined and are listed in Table 1 (ranging from 26.60 to 36.00 µM). To gauge their therapeutic potential and safety margin, we also tested these three compounds against the WI‐38 normal human lung fibroblast cell line, using doxorubicin as the reference standard.

Compounds 14d, 14h, and 14i were identified from the initial screening for cytotoxicity as the three most effective against HCT‐116 cells. They were selected for subsequent testing with respect to their active concentrations (IC₅₀ = 26.60–36.00 μM) on HCT‐116 colorectal cancer cells (listed in Table 1). In addition, we were eager to determine how well these three compounds compared in viability to each other by considering their effects on WI‐38 (normal human lung fibroblast) cells, with doxorubicin being the reference standard for comparison. The results will provide additional information regarding the potential therapeutic benefit and safety margin of each compound. All three compounds showed very low cytotoxicity toward WI‐38 cells, with IC₅₀ values > 100 µM and selectivity indices (SI = IC₅₀ (WI‐38)/IC₅₀ (HCT‐116)) of > 2.7, > 3.8, and > 2.8, respectively. These results strongly suggest that 14d, 14h, and 14i possess promising cancer‐cell‐selective cytotoxicity.

2.2.3. Enzyme Inhibition Profiling: Potency Against EGFR and CDK‐2

The three lead compounds (14d, 14h, and 14i) that stood out in the initial MTT screening were advanced to biochemical assays to assess their ability to inhibit EGFR and CDK‐2. Erlotinib and seliciclib were used as the respective reference inhibitors. Looking at the EGFR data in Table 2, 14i turned out to be the most potent analog in the series with an IC₅₀ of 0.056 µM—only about five times less potent than erlotinib (IC₅₀ = 0.012 µM). Compounds 14d and 14h were moderately active, giving IC₅₀ values of 0.10 µM and 0.11 µM, respectively. These sub‐micromolar values show that the series really can engage EGFR effectively, with 14i clearly among the better examples. On the CDK‐2 side, 14i and 14 h again delivered solid results, with IC₅₀ values of 0.215 µM and 0.252 µM—roughly ten‐fold weaker than seliciclib (IC₅₀ = 0.019 µM). Compound 14d still managed acceptable inhibition with an IC₅₀ value of 1.011 µM. Collectively, these results suggest that our synthesized thiazole chemotype has a valid potential to inhibit both kinases.

Table 2.

In vitro EGFR and CDK‐2 inhibitory assay of compounds 14d, 14h, and 14i with erlotinib and seliciclib employed as reference chemotherapeutic drugs.

Comp. no.	Measured IC50 (µM)*
	EGFR	CDK‐2
14d	0.108 ± 0.007	1.011 ± 0.13
14h	0.113 ± 0.009	0.252 ± 0.04
14i	0.056 ± 0.003	0.215 ± 0.07
Erlotinib	0.012 ± 0.001	—
Seliciclib	—	0.019 ± 0.001

^*The inhibitory activity of the compounds was expressed as IC₅₀ (µM) values. Values are expressed as mean ± SD of three independent trials. Dose–response curves were added in Supporting Information S2.

The correlation between the enzymatic results and cellular data for HCT‐116 cells is very optimistic. The best dual‐inhibitor compound, 14i, also displayed good cellular growth inhibition (GI%) and promising enzymatic IC₅₀ values, which strengthens the notion that the antiproliferative effect arises from inhibiting both of the target kinases together. The enzyme activities of 14d and 14h were somewhat lower than those of 14i and corresponded with lower GI% in the cellular data. However, the cellular potencies of both compounds were significantly lower than their sub‐micromolar IC₅₀ values, which is consistent with a typical finding in the early development of kinase inhibitors. Most likely, the disparity is due to low permeability of the compounds through cell membranes (due to their polar amide nature), potential efflux by transporters (like P‐gp and BCRP), degree of intracellular metabolism or requiring higher sustained target occupancy in whole‐cell assays compared with isolated enzyme assays. Collectively, these types of issues are frequently encountered when transitioning from biochemical to cellular activity, and they also provide us with multiple chances to enhance the efficacy of our compounds.

The enzymatic inhibition profiles of EGFR and CDK‐2 have moderate consistency with the cytotoxicity for the HCT‐116 cell line; however, the cellular potency of 14i (IC₅₀ = 0.056 µM for EGFR and 0.215 µM for CDK‐2) is also associated with the best correlation to the sub‐micromolar biochemical IC₅₀ values of all three compounds (range of 26.60–36.00 µM). Therefore, this indicates that dual inhibition of the two kinases is critical for driving the observed cytotoxicity. Conversely, compounds 14d (4‐morpholinophenyl) and 14 h (1‐(4‐phenylpiperazine‐1‐yl) had only moderate EGFR activity and 14h similar but somewhat lower CDK‐2 activity, indicating that they had lower overall engagement in the cellular environment compared to 14i.

The comparative activity of 14i (4‐morpholino) and 14h (1‐(4‐phenylpiperazin‐1‐yl)) versus 14h (1‐(4‐phenylpiperazine‐1‐yl)) with respect to their relative activity for inhibition of catalytic activity is an important observation regarding how sensitive this series is to minor differences in the terminal carbonyl attached substituent groups. When 4‐morpholino is attached directly, as in 14i, it may achieve a favorable balance between polarity and lipophilicity, thus allowing improved access to solvent‐exposed sites on the target ATP (in both the front pocket and ribose) with minimal hydrophilicity. Alternatively, the addition of the 1‐(4‐phenylpiperazine‐1‐yl) substituent in 14h provides complementary nitrogen content and conformational flexibility, probably enhancing binding affinity for this compound. This is further supported by the fact that both 14i and 14h display better than 14d because the morpholine in 14d is positioned at an increased distance from the core as a result of being linked via a phenyl group. The fact that the common thiazole (core) and amide (linker) bands in all compounds remains unchanged indicates that the nitrogen‐containing moiety is positioned closer to the core and directly contributes to binding. While these polar terminal substituents enhance the biochemical activity of both 14i and 14h, they may also detract from the cellular activity of either compound due to increasing hydrophilicity overall, increasing the number of H‐bond donors and/or acceptors, and thus potentially promoting efflux from cells; all of these factors may contribute to reducing passive cellular permeation and/or loss of cellular retention.

It seems quite realistic to bridge the gap between biochemical potency and cellular potency by optimizing the lead compounds based on the current understanding of structure–activity relationship (SAR). It may be beneficial to optimize the terminal substituent groups of the lead compounds by decreasing their polarity slightly; for example, by shortening the distance between the phenyl ring and the nitrogen‐containing group of a compound like 14d, or by trying various smaller cyclic amines/bioisosteres in place of the cyclic amine linker (i.e., morpholine or piperazine), but keeping the required interactions intact. In addition, adding small lipophilic substituents to other areas of the lead compound would increase the logP and permeability of the lead compounds without affecting the binding geometry. Potentially, modifying the linkers' length or flexibility could enhance the affinity of the lead compounds for their targets (i.e., the dual‐target binding profile) and increase the likelihood that their cellular potencies are more therapeutically relevant.

Although biochemical assays proved that compound 14i is a highly potent inhibitor of both EGFR and CDK‐2 (IC₅₀ = 0.056 µM for EGFR and IC₅₀ = 0.215 µM for CDK‐2), we did not perform direct evaluation of target cell engagement, such as Western Blot analysis with antibodies to detect phosphorylation level of EGFR or CDK‐2 substrates in this study; therefore, we will pursue this type of experiment in order to confirm the on‐target mechanism when we exclusively optimize compound 14i moving forward. Taking these results into consideration and supporting that both EGFR and CDK‐2 are valid drug targets that can be used for this thiazole chemotype development and demonstrate the validity of the approach. These results—along with cell line toxicity (discussed above)—provide either a rationale and/or roadmap for the next round of analog(s) from compound 14i, that is, focusing mostly on the physicochemical characteristics of the next compound to improve membrane permeability and intracellular exposure (while still maintaining the currently established binding modes) sets up for a strong foundation to complete the lead optimization process.

2.2.4. Cell Cycle Disruption by Lead Compounds: G₂/M Checkpoint Arrest in HCT‐116 Colorectal Carcinoma

Anticancer agents use cell cycle arrest as their primary mechanism to stop tumor growth because it prevents cancer cells from completing their division and replication process, leading to uncontrolled growth [49, 50]. We studied how compounds 14d, 14h, and 14i inhibit cancer growth by assessing their effects on HCT‐116 colorectal carcinoma cell cycle progression using flow cytometry analysis. The results (Figures 4 and 5) showed that treatment caused a substantial rise in cells in the G₂/M phase. The G₂/M population in untreated control cells reached 43.89%. Treatment with 14d, 14h, and 14i increased in this fraction which reached 66.15% for 14d and 51.55% and 51.59% for the other two compounds. The G₂/M arrest demonstrates that thiazole derivatives prevent cells from transitioning to G₂/M phase which halts cell division and causes antiproliferative effects.

Figure 4.

Percentage distribution of HCT‐116 cell subpopulation in different cell cycle phases for control and cells treated with 14d, 14h, 14i, and doxorubicin.

Figure 5.

Flow cytometry of control HCT‐116 cells, doxorubicin‐, 14d‐, 14h‐, and 14i‐treated HCT‐116 cells, respectively.

The EGFR and CDK‐2 biochemical inhibition patterns match the cell cycle disruption, which produces moderate cellular toxicity as shown by the MTT assay (IC₅₀ values in the 26.60–36.00 µM range for HCT‐116 cells). The G₁/S transition and S phase progression depend on CDK‐2 which serves as a crucial regulatory mechanism; however, its inhibition through EGFR blockade triggers compensatory mechanisms or checkpoint activation, leading to G₂/M arrest. EGFR signaling functions through its downstream pathways to control cyclin B1/CDK‐1 activity and mitotic entry which affects G₂/M checkpoints and other cell cycle checkpoints. Compound 14i (IC₅₀ 0.056 µM for EGFR and 0.215 µM for CDK‐2) shows strong dual inhibition which causes significant G₂/M accumulation because it disrupts both EGFR‐driven growth signaling and CDK‐2 cell cycle progression synchronously. The G₂/M cell‐cycle arrest and early apoptosis found with the use of compound 14d, 14h, 14i correspond as would be expected for a strong inhibition of both EGFR and CDK‐2, providing a more substantial disruption to cancer cell survival/division than could be achieved through inhibiting one of the targets alone; therefore supporting the rationale for dual‐targeting even though details of the pathway may be somewhat noisy.

The moderate cellular potency relative to the sub‐micromolar enzymatic IC₅₀ values likely reflects limitations in cellular permeability, intracellular distribution, or the need for prolonged target engagement in the whole‐cell context. The tested compounds induce G₂/M arrest, providing supportive evidence that their inhibitory activity against kinases leads to the antiproliferative effects seen in cytotoxicity testing. The study results propose the dual targeting approach of the series, while G₂/M checkpoint disruption may serve as the primary mechanism by which they fight cancer. We will use this to optimize the treatment for greater cellular effectiveness while keeping the beneficial treatment effects.

2.2.5. Apoptosis Induction by Lead Thiazole Derivatives in HCT‐116 Cells

Inducing apoptosis in cancer cells serves as the most effective method for developing new anticancer drugs, enabling patients' treatment through selective malignant cell destruction while protecting healthy cells. We tested 14d, 14h, and 14i in HCT‐116 colorectal carcinoma cells to determine whether our lead compounds' antiproliferative effects were mediated by apoptosis. Treatment with 14d, 14h, and 14i at their IC₅₀ concentrations for 24 h resulted in a significant increase of early apoptotic cells, as evidenced by Annexin V‐positive and PI‐negative cells in the lower‐right quadrant. As shown in (Figures 6 and 7), the tested compounds 14d, 14h, and 14i induced a total apoptosis of 18.60%, 19.77% and 15.97%, respectively compared to 0.23% in the negative control, while Doxorubicin induced total apoptosis by 17.08%. Regarding necrosis, the tested compounds 14d, 14h, and 14i decreased necrotic cell death of 2.02%, 2.40%, and 3.11%, respectively compared to 6.53% in the negative control.

Figure 6.

Percentage distribution of HCT‐116 cancer cell subpopulation across the different apoptotic phases following treatment with 14d, 14h, 14i, and doxorubicin.

Figure 7.

Annexin V/PI staining revealing early apoptosis induction by doxorubicin and lead compounds 14d, 14h, and 14i in HCT‐116 cells (Quartiles UL as necrosis, UR as late apoptosis, LL as live cells and LR as early apoptosis).

The results of apoptosis testing suggested that the observed cytotoxicity may be associated with the induction of apoptotic pathways, potentially linked to kinase modulation. The most effective dual inhibitor 14i, with an IC₅₀ of 0.056 µM for EGFR and 0.215 µM for CDK‐2, produced a significant early apoptosis rate because of its superior enzymatic potency and the low cellular IC₅₀ value in the MTT test. The two compounds 14d and 14h demonstrated EGFR inhibition at moderate levels while 14h showed slightly reduced CDK‐2 activity which resulted in high early apoptosis rates. The tested three compounds may lead to apoptosis through EGFR and CDK‐2 inhibition because their blocking activity creates downstream apoptotic signals which disrupt survival pathways that depend on EGFR‐mediated PI3K/AKT and cell cycle control through CDK‐2‐dependent G₁/S transition.

The process of apoptosis induction works together with the earlier observed G₂/M arrest which showed a G₂/M population increase from 43.88% in controls to 66.15% with 14d, 51.55% with 14h, and 51.59% with 14i. The G₂/M checkpoint becomes active when CDK‐2 and EGFR signaling pathways are blocked because these pathways lead to cyclin B1/CDK‐1 complex formation and subsequent failure of cells to enter mitosis which results in cell death when the G₂/M arrest continues without resolution. The absence of significant necrosis further indicates that these thiazole derivatives act primarily through targeted, programmed cell death rather than non‐specific cytotoxicity which makes it a desirable quality for future drug development.

2.2.6. Gene Expression Profiling: Modulation of Apoptotic and Survival Pathways in HCT‐116 Cells

Enzymatic inhibition of EGFR and CDK‐2 is linked to the observed cellular cytotoxicity and subsequent executional apoptosis, according to the quantitative real‐time polymerase chain reaction (RT‐qPCR) analysis conducted in HCT‐116 cells treated at the relevant IC₅₀ concentrations. The top three potent compounds 14d, 14h, and 14i, had significant impacts on the apoptotic key regulatory genes that were completely consistent with their dual‐target mechanism (Figure 8). Bax is a pro‐apoptotic gene whose upregulation stimulates mitochondrial permeabilization, leading to the release of pro‐apoptotic factors such as Cyto‐c, which triggers the activation of Caspase‐3, ultimately leading to cell death [51, 52]. Bax was overexpressed in the cells treated with compounds 14d, 14h, and 14i, while compound 14d induced the expression of Caspase‐3 in the treated cells. In addition, there was a significant increase in the expression of Cyto‐c by compounds 14d, 14h, and 14i by 19.6, 1.4 and 2.4 folds, respectively compared to doxorubicin.

Figure 8.

Relative gene expression level in HCT‐116 cell lines treated with doxorubicin, 14d, 14h, and 14i.

In addition, there are some factors that promote cell growth and counteract the apoptotic process, such as (VEGF, AKT‐1, and Bcl‐2). Treatment with compounds 14d, 14h, and 14i led to a significant downregulation of VEGF expression—an angiogenic growth factor critical for tumor vascularization [53]—with fold reductions of 1.9, 1.1, and 2.5, respectively, relative to doxorubicin‐treated cells. We also observed downregulation of AKT‐1, an oncogene that drives cancer progression. Treatment with compounds 14d, 14h, and 14i reduced AKT‐1 expression by 1.3‐, 2.0‐, and 5.8‐fold, respectively, compared to doxorubicin. Furthermore, only 14h and 14i significantly decreased the expression of anti‐apoptotic gene Bcl‐2, with fold reductions of 1.2 and 1.7, respectively, relative to doxorubicin.

The gene expression alterations that we observed provide preliminary insight into the possible molecular mechanisms underlying the observed biological activity. The strongest gene modulation by 14i stems from its superior enzymatic potency which enables the compound to achieve its low cellular IC₅₀ in the MTT assay and high G₂/M arrest at 51.59% of the population, and high early apoptosis induction at 14.40% Both compounds 14d and 14h exhibited lower levels of gene expression than other compounds studied because the inhibition of their corresponding kinases was less effective than 14i, which resulted in less pronounced effects in terms of the cell cycle and/or apoptosis. In addition, Annexin V/PI assays performed with these compounds revealed minimal necrosis. In contrast, the effect of the thiazole series on both the growth/survival factors tested (AKT‐1, bcl‐2 and VEGF) and the apoptotic effectors (Bax, cytochrome‐c, and caspase‐3) was supportive of a mechanism through which each compound acted via selective, pre‐programmed cell death rather than through nonselective, toxic effects.

The process begins with kinase inhibition, which triggers a sequence of events that starts with EGFR/CDK‐2 blockade and results in the shutdown of survival and angiogenic pathways through decreased AKT‐1, VEGF, and Bcl‐2 levels. The G₂/M checkpoint is activated as the mechanism proceeds to the stage where cells begin early apoptosis, marked by elevated levels of Bax and Cyto‐c and Caspase‐3. Targeting multiple sequence mechanisms shows scientific validity because it supports the potential of the series to function as selective anticancer agents, with 14i being the most promising lead for future optimization. The combined findings from enzyme assays, cellular cytotoxicity tests, cell‐cycle arrest, apoptosis induction, and gene expression analysis support that compounds 14d, 14h, and 14i function as dual‐target EGFR/CDK‐2 apoptotic inducers while inducing selective intrinsic apoptosis in colorectal cancer cells. This thiazole‐containing series continues to be a preferred substrate for new colorectal cancer treatments.

Gene expression analysis presented here gives a sort of preliminary insight into the potential molecular mechanisms that could explain the apoptotic effects that we observed. While the results show fairly consistent trends in how pro‐apoptotic genes are being modulated (Bax, cytochrome c, caspase‐3) and how anti‐apoptotic or more survival related genes (VEGF, AKT‐1, and Bcl‐2) are also shifting, we should still say that this analysis is limited in scope and really it should be read as supportive, not definitive.

2.3. Molecular Modeling Simulations

2.3.1. Network Pharmacology Screening

2.3.1.1. Network Pharmacology Profiling of Lead Compounds 14d, 14h, and 14i to Identify Molecular Pathways Underlying Anti‐Colon Cancer Activity

To elucidate the molecular mechanisms underlying the anti‐colorectal cancer activity of the most potent thiazole derivatives (14d, 14h, and 14i), we employed network pharmacology to predict and prioritize the key signaling pathways and target networks involved. The DISGENET, Gene Cards, and OMIM databases were used to identify genes linked to colon cancer as well as the anticipated molecular targets of compounds 14d, 14h, and 14i. Twenty‐six overlapping genes were revealed by Venn diagram analysis, indicating shared molecular targets of genes linked to colon cancer and targets associated with thiazole derivatives, via which our tested compounds may have an anticancer effect (Figure 9A).

Figure 9.

Network pharmacology analysis of colon cancer and compounds 14d, 14h, and 14i. (A) Venn diagram of genes overlapped with colon cancer and those predicted for compounds 14d, 14h, and 14i. (B) Protein–protein interactions network of the predicted genes. (C) Interaction network of core hub proteins.

2.3.1.2. Protein–Protein Interaction (PPI) Network of Common Targets Between Colon Cancer and Compounds 14d, 14h, and 14i

To investigate the functional associations between these overlapping targets, a PPI network was constructed. Figure 9B,C revealed that the PPI network is highly interconnected, showing strong functional linkages between the shared targets. Several proteins, including EGFR, AKT‐1, CDK‐1, CDK‐2, CCNB1, CCNA2, AURKA, AURKB, BRAF, BRCA1, CDC20, CDC45, and ATM, emerged as hup nodes with high connectivity. Functional analysis of PPI network demonstrated that these genes are associated with key oncogenic processes such as cell cycle regulation, apoptosis, DNA damage and proliferation. Overall, these results revealed that these thiazole derivatives may exhibit their anticancer activity through multi‐target interactions.

2.3.1.3. Enrichment Analysis based on Gene Ontology (GO) and Kyoto Encyclopedia of Genes and Genomes (KEGG) databases

The top 10 pathways were selected based on Log10 (P) values following KEGG and GO enrichment analyses, as shown in Figure 10. Molecular functions (MF), biological processes (BP), and cellular components (CC) were included in the GO enrichment analysis for overlapping targets. According to MF (Figure 10A), there was a predominant enrichment in protein serine/threonine kinase activity, histone kinase activity and MAP kinase‐related activity, highlighting the vital role of kinase‐related signaling. Cellular components analysis (CC) (Figure 10B) revealed that the targets were mainly associated with cyclin‐dependent kinase complexes, germ cell nucleus, pronucleus, spindle centrosome, kinetochores, and DNA repair complex. BP enrichment (Figure 10C) exhibited significant involvement in mitotic segregation, spindle assembly, spindle signaling, chromatid segregation, mitotic metaphase/anaphase transition, and mitotic sister chromatid separation, suggesting the role of these targets in mitotic and apoptosis processes. KEGG pathway analysis showed a predominant enrichment in cancer‐related pathways such as colorectal, bladder, prostate, melanoma and glioma, as well as signaling pathways associated with cell cycle, apoptosis and EGFR inhibitor resistance (Figure 10D).

Figure 10.

Enrichment analysis results including GO categories: (A) BP, (B) CC, (C) MF. (D) KEGG pathway analysis. BP, biological processes; CC, cellular components; KEGG, Kyoto Encyclopedia of Genes and Genomes; MF, molecular functions.

2.3.2. Molecular Docking Studies

The molecular operating environment (MOE) 2024.06 was used to carry out and visualize the docking process of the most active compound 14i as it showed the most inhibitory effect against EGFR and CDK‐2 enzymes with IC₅₀ 0.056 and 0.215 μM, respectively. These simulations fully explained the observed SARs, provided preliminary insights into hinge‐region occupation, hydrophobic‐pocket filling, and hydrogen‐bonding interactions, and may guide future lead‐optimization techniques. To verify the docking process, the co‐crystallized ligand was obtained from the reference crystal structure and then redocked using the standard docking approach into the target protein's binding pocket. The redocked ligand was superimposed onto the initial co‐crystallized posture to calculate the root mean square deviation (RMSD). The RMSD values remained consistently below 2 Å throughout the redocking and docking methodology validation. These low RMSD values suggest little variation in atomic locations, supporting the strategy for additional virtual screening or ligand‐optimization research (Supporting Information: Figures S4 and S5).

2.3.2.1. Molecular Docking Into the EGFR Active Site

In order to predict the binding affinity of the most potent synthesized compound in this series, molecular docking was performed for compound 14i (IC₅₀ = 0.056 µM) in the active site of the EGFR enzyme (code ID 1M17) [54] using erlotinib (IC₅₀ = 0.012 μM) as a reference EGFR inhibitor. The docking scores and the binding interactions of compound 14i and erlotinib are listed in Supporting Information: Table S1 and Figures 11 and 12. Erlotinib interacted with the EGFR enzyme through a range of hydrophobic and hydrophilic interactions. Among the most common are hydrogen bonds involving Gln767 and Met769 via the quinazoline moiety. It was also bound by strong hydrophobic interactions with Leu820, Gly772, Leu694, and Val702, in addition to an arene‐H interaction with Leu820. On the other hand, compound 14i displayed a comparable binding mode and a range of interactions with the EGFR active site compared to erlotinib. Compound 14i formed two hydrogen bonds with Gln767 and Met769 via the sulfur atom of the thiazole ring moiety, in addition to the hydrophobic interactions with Leu820, Gly772, Leu694, and Val702. Compound 14i also formed an arene‐H interaction with Pro770. The calculated binding energy for 14i (−6.26 Kcal/mol) was comparable to that of erlotinib (− 8.27 Kcal/mol). Finally, molecular docking of compound 14i in the active site of EGFR offered plausible interactions with EGFR amino acid residues, comparable to those observed for erlotinib, which correlates with its inhibitory activity against EGFR enzyme.

Figure 11.

2D binding interactions of erlotinib and compound 14i with key residues of EGFR.

Figure 12.

3D visualization of the binding interactions of erlotinib and compound 14i with the amino acid residues of EGFR.

2.3.2.2. Molecular Docking Into the CDK‐2 Active Site

Molecular docking of compound 14i (IC₅₀ = 0.215 μM) in the CDK‐2 enzyme's active site (code ID 2A4L) [55] was done using seliciclib (IC₅₀ = 0.019 μM) as a reference CDK‐2 inhibitor in order to determine the binding affinity of compound 14i. Supporting Information: Table S2 summarizes the docking scores and binding interactions of compound 14i and seliciclib. The CDK‐2 enzyme and seliciclib interacted via a variety of hydrophobic and hydrophilic interactions. Seliciclib formed hydrogen bonds with Leu83, Glu81, and Gln131. Seliciclib engaged in hydrophobic interactions with Ile10, Leu134, Gln131, and Val18, in addition to an arene‐H interaction with Ile10, as shown in Figure 13 and 14. Compound 14i revealed a similar binding pattern and several interactions with the CDK‐2 active site. Compound 14i formed two hydrogen bonds with Leu83 and Asp145 by the oxygen atom of the morpholine moiety and the sulfur atom of the thiazole moiety, respectively. Compound 14i also formed stabilizing hydrophobic contacts with Ile10, Leu134, Gln131, and Val18 in the EGFR active site, as shown in Figure 13 and 14. The computed binding free energy of 14i (−6.11 Kcal/mol) was comparable to seliciclib (−7.83 Kcal/mol). In conclusion, the suggested notable interactions of compound 14i with CDK‐2 amino acid residues that are similar to those observed for seliciclib may explain its inhibitory activity against CDK‐2. The molecular docking studies suggest favorable binding modes of compound 14i within the active sites of EGFR and CDK‐2. These computational results are consistent with the observed biochemical inhibitory activities and provide structural rationale for the structure–activity trends, but should be interpreted as hypothetical binding poses that require experimental validation.

Figure 13.

Comparative 2D binding mode of seliciclib and compound 14i, illustrating the key residues of CDK‐2.

Figure 14.

3D visualization for the binding interactions of seliciclib and compound 14i with CDK‐2 binding pocket amino acid residues.

2.3.3. Molecular Dynamics (MD) Simulations

Using the EGFR crystal structure (PDB 1M17), which represents an active conformation of the kinase domain bound to the inhibitor erlotinib, the results of MD simulations were obtained over a period of 100 ns. The choice of crystal structure was by design to include well‐defined regions (ATP binding site, hinge region, P‐loop, and activation loop) having excellent electron density for critical residues. So, it is fairly suitable for examining type‐I ATP competitive inhibitors like our thiazole‐derived compounds. Even though there are newer EGFR structures with higher resolution in the Protein Data Bank, PDB 1M17 still stays among the most commonly used and validated models in the literature for docking and MD investigations of EGFR tyrosine kinase inhibitors, enabling straightforward comparison with data that has been reported earlier. The most effective compound 14i was tested alongside the co‐crystallized ligand Erlotinib (Co).

2.3.3.1. RMSD Analysis

A 100 ns molecular dynamics simulation was used to evaluate the structural stability of the two investigated complexes (14i–1M17 and Co–1M17) by monitoring the RMSD trajectories of the protein Cα atoms and co‐crystallized ligand, as shown in Figure 15. The protein RMSD (blue line) for the first complex (14i–1M17) increased quickly during the initial equilibration period (0–10 ns), then stabilized at 4.5–4.8 Å, indicating a stable conformational assembly. The ligand RMSD (red line) fluctuates more during the initial phase, indicating conformational adjustment inside the binding pocket, before stabilizing at 6.5–7.5 Å for the rest of the simulation, demonstrating persistent binding despite modest flexibility (Figure 15A). For Co–1M17 complex, the protein RMSD (blue line) gradually increased and reached equilibrium after 20–25 ns. The RMSD values remained low (2.8–3.2 Å), indicating enhanced structural rigidity. The ligand RMSD (red line) followed a similar equilibration pattern, with initial deviations before stabilizing with decreased fluctuations (Figure 15B). Overall, the RMSD analysis recommended the dynamic stability of both complexes (14i–1M17 and Co–1M17) throughout the molecular dynamics simulation period. The results indicate that the binding poses of compound 14i remain relatively stable over the 100 ns simulation period. These findings support the plausibility of the docking poses but do not constitute direct experimental evidence of binding.

Figure 15.

The root mean square deviation (RMSD) (Å) plots for ligand–protein complexes in the EGFR active site (PDB: 1M17) simulated over 100 ns (RMSD vs. time): (A) 14i–1M17 complex; (B) Co–1M17 complex.

2.3.3.2. Root Mean Square Fluctuation (RMSF)

RMSF analyses were used to assess the residue‐level flexibility of the two protein–ligand complexes throughout the 100‐ns period. Alpha helices and beta strands are examples of secondary structural parts that fluctuate less than loop areas because they are less flexible than the unstructured region of the protein. As shown in Figure 16, red and blue background colors were used to identify beta strands and alpha helices, respectively. Green vertical bars indicated areas that interact with the ligand. According to RMSF profile of the 14i–1M17 complex (Figure 16A), most residues remains stable, with fluctuations less than 2.4 Å. While residues within the binding site have low RMSF values, indicating structural rigidity. The higher flexibility is primarily limited to the terminal sections and a few loop regions. The distribution of B‐factors and RMSF trends corresponded well with one another, demonstrating stable dynamic behavior of the protein. In particular, the vast majority of residues in the Co–1M17 complex exhibited low RMSF values (< 2.4 Å), indicating that they were generally stabilized structurally. Notably, the residues located near the ends (terminal) of the proteins and/or in loop regions were far more susceptible to variation, while the residues at the interface of the protein that participated in binding and/or incorporation of ligand exhibited substantial structural stability through the duration of the simulations. The consistency in the RMSF patterns with the B‐factor profile provides additional support for the stabilizing influences of the overall dynamic behavior observed in the two complexes (Figure 16B). Ultimately, the results suggest that the relative structural integrity of the two complexes is maintained over the course of the simulations, with most of the flexibility being limited to the terminal and loop regions of the proteins, while the critical functional residues involved in ligand incorporation remained relatively stable, allowing for stable protein‐ligand interactions.

Figure 16.

The root mean square fluctuation (RMSF) (Å) plots for ligand–protein complexes in the EGFR active site (PDB: 1M17) against residue index: (A) 14i–1M17 complex; (B) Co–1M17 complex.

2.3.3.3. Histogram and Contact Map Analyses

The interaction histograms with respect to the residues during the 100 ns MD simulation indicated that amino acids could provide multiple binding contributions to stabilize the ligand within the two different complexes, that is, the 14i–1M17 and Co–1M17. For 14i–1M17 (Figure 17A), there is a direct hydrogen bond interaction between the ligand and key residues (Lys721, Cys751, Arg817, and Thr830), suggesting a critical role of these residues in the recognition of the ligand and in anchoring the ligand within the active site of the protein. The primary amino acids that contribute to hydrophobic interactions are Leu694, Phe699, Val702, Ala719, and Leu768, thus creating a hydrophobic environment to stabilize the ligand conformation. In addition, interactions via water molecules between Ser696; Lys721; Cys731; and Asp776 help to maintain the general stability of the complex. In the case of the Co–1M17 complex, the primary hydrogen bond interactions are associated with Leu694, Ser696, Met769, Arg817, Thr830, and Asp831, thereby indicating an important role in anchoring the ligand to the protein. Hydrophobic interactions between Phe699, Val702, Ala719, Leu775, and Leu820 also enhance the stability of the ligand within the binding pocket, while water‐bridged interactions (notably, Ser696, Lys721, and Arg817) further strengthen the protein–ligand interaction (Figure 17B). Thus, the interaction profiles for both complexes suggest stable and well‐defined ligand binding.

Figure 17.

Histograms demonstrating the protein–ligand interaction fractions in the EGFR active site (PDB: 1M17): (A) 14i–1M17 complex; (B) Co–1M17 complex. Interaction fraction = 0.1, suggesting a 10% persistence of protein–ligand contact.

The heat contact maps in Figure 18 represent how ligands interact with protein residues (14i–1M17 and Co–1M17) while undergoing MD for 100 ns. For both complexes, the blue line in the top portion of the page illustrates that the overall number of interactions did not fluctuate throughout the MD, suggesting that ligands remain continuously bound to the same protein binding sites. The durability of the interaction between individual protein residues across the duration of the MD is represented in the bottom section of the heatmap. There were residues (Ser696, Val702, Lys721, and Arg817) in the 14i–1M17 complex, which had sustained interaction profiles over the entire 100‐ns MD trajectory, meaning they contribute significantly to the stability of the complex. Other residues exhibited fluctuating amounts of interaction, suggesting the residues in the binding pocket are more flexible. The second complex (Co–1M17) exhibited significantly greater density and persistence of interactions than the first complex (14i–1M17) between Leu694, Phe699, Val702, Met769, and Arg817, suggesting stronger and more stable ligand binding. Therefore, the heatmap analyses support stable binding for both complexes.

Figure 18.

The heat contact map visualization describing the interaction dynamics between the ligand and protein residues in the EGFR active site (PDB: 1M17) over 100 ns. (A) 14i–1M17 complex. (B) Co–1M17 complex. The heat intensity (color) reflects the number of contacts: lighter colors represent fewer interactions, while darker shades (orange to red) signify more frequent and stable interactions.

2.3.4. ADMET prediction

Lipinski's rule of five was applied to evaluate the drug‐likeness and potential oral bioavailability of the most active compounds (14d, 14h, and 14i) by assessing their key physicochemical properties. Using Swiss ADME prediction website [56], compounds 14d, 14h, and 14i were assessed to see if they comply with Lipinski rule of five. For a drug to be orally active, the rule permits just one violation of the criteria mentioned in Table 3. All compounds obey Lipinski's rule of five and had zero violations, making them suitable for use as oral drugs.

Table 3.

Calculated parameters of Lipinski's rule of five for the most active compounds 14d, 14h, and 14i.

Comp. no.	Parameters
	Log P	TPSA	MW	nHBA	nHBD	nRB	nVs
14d	2.6	70.67	274.34	3	0	3	0
14h	3.29	64.68	349.45	2	0	4	0
14i	2.97	82.70	365.45	3	1	5	0

Abbreviations: Log P, calculated lipophilicity; MW, molecular weight; nHBA, number of hydrogen bond acceptors; nHBD, number of hydrogen bond donors; nRB, number of rotatable bonds; nVs, number of violations from Lipinski's rule of five; TPSA, total polar surface area.

Admetsar website [57] was used to calculate ADMET properties of the most potent compounds 14d, 14h, and 14i. The ADMET data were summarized in Table 4. All the compounds showed high human intestinal absorption (HIA > 90%), so they can be easily absorbed from GIT. Additionally, the tested compounds had acceptable water solubility as their log S values ranged from −4.77 to −3.05 which guarantees rapid disintegration in digestive fluids and pharmaceutical formulations. In addition, the tested compounds showed low predicted affinity for CYP2D6, indicating a reduced likelihood of drug–drug interactions. Carcinogenicity testing predicted that none of the compounds exhibit carcinogenic potential. Given their clean and favorable in silico ADMET profile, these promising leads (14d, 14h, and 14i) warrant progression to in vivo pharmacokinetic studies in mice or rats to confirm adequate oral bioavailability and acceptable half‐life for further preclinical development. Moreover, these compounds would be most appropriate for efficacy investigations in long‐term administration regimens due to the lack of CYP2D6 liability and potential non‐carcinogenicity.

Table 4.

Predicted ADMET data for the most active compounds 14d, 14h, and 14i.

Comp. no.	Log S	HIA%	CYP2D6 inhibition	Carcinogenicity
14d	−3.05	High	Noninhibitor	Noncarcinogenic
14h	−4.77	High	Noninhibitor	Noncarcinogenic
14i	−4.48	High	Noninhibitor	Noncarcinogenic

2.3.4.1. SAR

The in vitro cytotoxicity data against HCT‐116 colorectal carcinoma cells revealed preliminary structure–activity trends across the synthesized thiazole series. In Series A and B, the R¹ position in compounds 14d–14i showed a major impact on potency when different heteroaryl and aryl groups were applied. The morpholine‐substituted analogues (14d and 14i) showed the highest active potential followed by compounds that contained the phenylpiperazine 14h and (4‐methylpiperazin‐1‐yl)aniline 14f structural elements. The introduction of bulkier or more electron‐withdrawing substituents, such as ((4‐chlorophenyl)sulfonyl)piperazine 14g or (4‐acetylpiperazin‐1‐yl)aniline 14e, resulted in compounds that displayed moderate to decreased therapeutic potential. Those findings show that small, moderately polar heterocycles at R¹ are optimal because they provide good binding to the solvent‐exposed front pocket and the ribose‐binding region of the ATP site. Bulky or strongly electron‐withdrawing groups create steric clashes which lead to negative electronic effects that decrease binding affinity.

The para‐position of the phenyl ring in Series C compounds 14a–14c and 14j–14m showed that two different para‐position substitutions preferred halogen elements with smaller atomic sizes. Potency followed the order F > Cl > H > OCH₃ > Br, with fluorine (14 m) providing the greatest enhancement. Halogenation at R³‐position brought about better activity results, with chlorine showing more potency than bromine and hydrogen. This study shows that small electron‐withdrawing groups function correctly because they fit into hydrophobic pocket I or II without causing steric repulsion. The methoxy group, together with bigger bromine atoms, causes disruption of the best interactions due to its their electron‐donating power. Also, this study involved a reciprocal exchange of the substituents: the phenyl group and amide functionality on the thiazole core were swapped between the 4‐ and 2‐positions (4‐phenyl/2‐amide in 18 → 2‐phenyl/4‐amide in 14a), enabled more precise spatial positioning conferring markedly improved potency.

The potency of Series D dimidamide‐linked derivatives 22, 26, 30a, and 30b varied when we lengthened the linker and changed the amide substituent. The most active analogue was 30a (bearing an oxo‐2‐acetamido‐N‐phenylacetamide group), followed by 30b (2‐acetamido‐N‐phenylacetamide), while the simplest analogue 22 was the least active. The study results show that longer and more functional amide chains at R⁴ position produce better activity results because they enable more hydrogen‐bonding and hydrophobic interactions to occur in the solvent‐exposed area. The study revealed that substituting the 4‐position 22 for the 5‐position 26 improved potency because the 5‐position enabled more precise spatial alignment with the kinase binding pocket.

Overall, the SAR analysis (Figure 19) revealed relatively modest variations in activity across the compound series. This behavior suggests a degree of structural tolerance within the scaffold, which may reflect flexible binding interactions at the target sites. It should be noted that SAR conclusions are based on single‐dose screening data and therefore represent preliminary observations. Full dose–response studies and IC₅₀ determination for the entire series are currently underway as part of ongoing lead optimization efforts.

Figure 19.

Structure–activity relationships of thiazole analogues: Key substituent effects on HCT‐116 cytotoxicity.

3. Conclusion

In summary, new thiazole analogues were synthesized and investigated for their potential anticancer properties. MTT anticancer assay recommended that compounds 14d, 14h, and 14i are considered to be the most active members in this series against HCT‐116 cancer cell lines with growth inhibition percentage (GI %), ranging from 33.29% to 35.09% and minimal cytotoxic effect against normal fibroblast WI‐38 cells. Additionally, compound 14i exhibited the most potent EGFR and CDK‐2 inhibitory activity with IC₅₀ values 0.056 and 0.215 µM, respectively. Cell cycle analysis and apoptosis assays supported that compounds 14d, 14h, and 14i induced early‐stage apoptosis and arrested the cell cycle at G₂M phase. Gene expression analysis suggested that treatment with compounds 14d, 14h, and 14i significantly upregulated the pro‐apoptotic factors Bax and cytochrome c, with caspase‐3 showing notable upregulation particularly in cells treated with 14d. On the other hand, VEGF, AKT‐1 and Bcl‐2 factors which promote cell growth and counteract the apoptotic process, were downregulated in the cells after the treatment with compounds 14d, 14h, and 14i, explaining the potential apoptotic mechanism of the tested compounds. Molecular docking study was conducted to support the obtained results with compound 14i showing a significant binding affinity toward the active site of EGFR and CDK‐2 enzymes. Furthermore, molecular dynamic simulations have been conducted to evaluate the stability of ligand‐protein complexes. Also, the top‐three active compounds 14d, 14h, and 14i obeyed Lipinski's rule of five. The combined outcomes from biochemical studies, cellular research, molecular investigations and computational analysis suggested that compounds 14d and 14h, and especially 14i, function as potential dual EGFR/CDK‐2 inhibitors. Those findings provide preliminary evidence for advancing this thiazole series into direct functional validation of cellular target engagement, in vivo pharmacokinetic and efficacy studies, offering significant potential for the development of novel dual‐targeted anticancer agents.

4. Experimental

4.1. Chemistry

4.1.1. General remarks

All reagents, solvents, and chemicals were purchased from Sigma‐Aldrich (U.S.A.) or Alfa Aesar and used as received without further purification. Melting points (°C) were determined using Stuart melting point apparatus SMP30. The ¹H NMR and ¹³C NMR spectra were recorded using Bruker Avance III HD FT‐high resolution NMR spectrometer at Faculty of Pharmacy, Mansoura University, ¹H and ¹³C NMR spectra were attained at 400/100 MHz, respectively in either CDCl₃ or DMSO‐d _6. All chemical shifts were expressed in δ ppm and referenced to TMS. Mass spectroscopy was carried out on a direct inlet part to the mass analyzer in Thermo‐ Scientific GCMS type ISQ at the Regional Center for Mycology and Biotechnology (RCMB), Al‐Azhar University, Nasr City, Cairo. Reactions progresses were monitored using TLC on silica gel plates 60F245 E. Merk, DCM/methanol 9:1 or hexane/ethyl acetate 7:3 or 3:7 were used as eluting system and the eluted compounds were visualized using a UV lamp set at (366–245 nm). HRMS data were recorded on LC/Q‐TOF, 6530 (Agilent Technologies, Santa Clara, CA, USA). HPLC system was used to determine purities for each final compound and found to be ≥ 95%. HPLC system used an LC column (4.6 mm × 150 mm, 5 μM) which was kept at a temperature of 40°C. The applied isocratic program involved a flow rate of 1 mL/min with 50% methanol/50% buffer as a mobile phase. The eluted peaks were monitored using UV absorbance at 280 nm. Supporting Information: S2 contain detailed procedures for the synthesis of intermediates 2, 4a–c, 5a–c, 7a–c, 8a–c, 10a–e, 11a–e, 12a–e, 13a–e, 16–17, 20–21, 24–25, 28a–b, and 29a–b.

4.1.2. General procedure for synthesis N‐(substituted)−2‐phenylthiazole‐4‐carboxamides (14a–m)

To a stirred solution of the corresponding carboxylic acids 13a–e (1.0 equiv.) and the appropriate amine (1.0 equiv.) in dichloromethane (for 14a–j) or dimethylformamide (for 14k–m), EDC·HCl (1.5 equiv.) and DMAP (0.2 equiv.) were added at room temperature. The reaction mixture was stirred overnight at room temperature. Upon completion (monitored by TLC), the pH was acidified using 1 N HCl solution. The mixture was then extracted with dichloromethane (3 × 20 mL), and the combined organic layers were washed with brine, dried over anhydrous sodium sulfate, filtered, and concentrated under reduced pressure. The crude residue was purified by trituration with methanol followed by hexane to afford the target compounds 14a–m in good purity and moderate to good yields.

N‐[2‐Oxo‐2‐(phenylamino)ethyl]−2‐phenylthiazole‐4‐carboxamide (14a): Off‐white powder, yield (0.1 g, 30%), MP 203°C; ¹H NMR (400 MHz, DMSO‐d ₆) δ 10.12 (s, 1H, Ar NH CO‐), 8.78 (t, J = 6.0 Hz, 1H, thiazole‐CO NH ‐), 8.37 (s, 1H, CH ‐thiazole), 8.09 (dd, J = 6.7, 2.8 Hz, 2H, ArH), 7.61 (d, J = 8.1 Hz, 2H, ArH), 7.55 (dt, J = 6.9, 3.4 Hz, 3H, ArH), 7.33 (t, J = 7.8 Hz, 2H, ArH), 7.07 (t, J = 7.4 Hz, 1H, ArH), 4.15 (d, J = 6.0 Hz, 2H, CH₂); ¹³C NMR (100 MHz, DMSO‐d ₆) δ 167.4 (ArNH CO ‐), 167.3 (C ² ‐thiazole), 160.7 (thiazole‐ CO ‐), 150.3 (Ar), 138.9 (Ar), 132.4 (Ar), 130.8 (Ar), 129.3 (Ar), 128.8 (Ar), 126.4 (Ar), 124.4 (Ar), 123.3 (Ar), 119.1 (Ar), 42.8 (CH₂); MS m/z (%) for C₁₈H₁₅N₃O₂S: 337.67 (11.44, M⁺), 69.49 (100.00).

N‐{2‐[(4‐Bromophenyl)amino]−2‐oxoethyl}−2‐phenylthiazole‐4‐carboxamide (14b): White powder, yield (0.2 g, 48%), MP 238°C; ¹H NMR (400 MHz, DMSO‐d ₆) δ 10.27 (s, 1H, Ar NH CO‐), 8.80 (t, J = 6.0 Hz, 1H, thiazole‐CO NH ‐), 8.35 (s, 1H, NH, CH ‐thiazole), 8.07 (dd, J = 6.9, 2.8 Hz, 2H, ArH), 7.65–7.47 (m, 7H, ArH), 4.13 (d, J = 5.9 Hz, 2H, CH₂); ¹³C NMR (100 MHz, DMSO‐d ₆) δ 167.7 (ArNH CO ‐), 167.3 (C ² ‐thiazole), 160.8 (thiazole‐ CO ‐), 150.3 (Ar), 138.3 (Ar), 132.4 (Ar), 131.6 (Ar), 130.9 (Ar), 129.3 (Ar), 126.5 (Ar), 124.4 (Ar), 121.1 (Ar), 114.9 (Ar), 42.9 (CH₂); MS m/z (%) for C₁₈H₁₄BrN₃O₂S: 417.20 (21.15, M⁺ +2), 415.13 (28.14, M⁺), 269.97 (100.00).

N‐{2‐[(4‐Chlorophenyl)amino]−2‐oxoethyl}−2‐phenylthiazole‐4‐carboxamide (14c): White powder, yield (0.15 g, 40%), MP 237°C; ¹H NMR (400 MHz, DMSO‐d ₆) δ 10.30 (s, 1H, Ar NH CO‐), 8.81 (t, J = 6.0 Hz, 1H, thiazole‐CO NH ‐), 8.35 (s, 1H, CH ‐thiazole), 8.10–8.04 (m, 2H, ArH), 7.65 (d, J = 8.5 Hz, 2H, ArH), 7.55 (d, J = 5.6 Hz, 3H, ArH), 7.37 (d, J = 8.4 Hz, 2H, ArH), 4.13 (d, J = 5.9 Hz, 2H, CH₂); ¹³C NMR (100 MHz, DMSO‐d ₆) δ 167.7 (ArNH CO ‐), 167.3 (C ² ‐thiazole), 160.8 (thiazole‐ CO ‐), 150.3 (Ar), 137.9 (Ar), 132.5 (Ar), 130.9 (Ar), 129.4 (Ar), 128.7 (Ar), 126.9 (Ar), 126.5 (Ar), 124.5 (Ar), 120.7 (Ar), 42.9 (CH₂); MS m/z (%) for C₁₈H₁₄ClN₃O₂S: 373.7 (12.46, M⁺ +2), 371.81 (35.46, M⁺), 93.35 (100.00).

N‐(4‐Morpholinophenyl)−2‐phenylthiazole‐4‐carboxamide (14d): Off‐white powder, yield (0.07 g, 23%), MP 166°C; ¹H NMR (400 MHz, DMSO‐d ₆) δ 10.20 (s, 1H, thiazole‐CO NH ‐), 8.46 (s, 1H, CH ‐thiazole), 8.19 – 8.12 (m, 2H, ArH), 7.81 (d, J = 8.4 Hz, 2H, ArH), 7.56 (d, J = 5.5 Hz, 3H, ArH), 7.23 (d, J = 8.5 Hz, 2H, ArH), 3.85 (t, J = 4.7 Hz, 4H, 2CH₂O of morpholine), 3.24 (t, J = 4.7 Hz, 4H, 2CH₂N of morpholine); ¹³C NMR (100 MHz, DMSO‐d ₆) δ 167.4 (C ² ‐thiazole), 158.8 (thiazole‐ CO ‐), 150.5 (Ar), 132.4 (Ar), 130.9 (Ar), 129.3 (Ar), 126.7 (Ar), 125.1 (Ar), 121.6 (Ar), 117.2 (Ar), 65.4 (2CH₂O of morpholine), 50.4 (2CH₂O of morpholine); MS m/z (%) for C₂₀H₁₉N₃O₂S: 365.01 (14.38, M⁺), 73.85 (100.00); HRMS (ESI‐TOF) m/z: [M]^‐ Calcd for C₂₀H₁₉N₃O₂S 365.1198; Found 365.1172.

N‐[4‐(4‐Acetylpiperazin‐1‐yl)phenyl]−2‐phenylthiazole‐4‐carboxamide (14e): Gray powder, yield (0.08 g, 20%), MP 227°C; ¹H NMR (400 MHz, DMSO‐d ₆) δ 10.26 (s, 1H, thiazole‐CO NH ‐), 8.47 (s, 1H, CH ‐thiazole), 8.15 (dd, J = 6.6, 2.9 Hz, 2H, ArH), 7.85 (d, J = 8.5 Hz, 2H, ArH), 7.63 – 7.59 (m, 3H, ArH), 7.41 – 7.26 (m, 2H, ArH), 3.73 (t, J = 5.0 Hz, 4H, 2CH₂ of piperazine), 3.33 (t, J = 5.0 Hz, 2H, CH₂ of piperazine), 3.27 (d, J = 5.0 Hz, 2H, CH₂ of piperazine), 2.07 (s, 3H, CH₃); ¹³C NMR (100 MHz, DMSO‐d ₆) δ 168.4 (CH₃ CO ), 167.4 (C ² ‐thiazole), 158.9 (thiazole‐ CO ‐), 150.5 (Ar), 132.4 (Ar), 130.9 (Ar), 129.3 (Ar), 126.7 (Ar), 126.5 (Ar), 125.3 (Ar), 121.6 (Ar), 53.5 (2CH₂ of piperazine), 44.6 (2CH₂ of piperazine), 21.2 ( CH ₃ CO).; MS m/z (%) for C₂₂H₂₂N₄O₂S: 406.99 (19.98, M⁺), 312.56 (100.00).

N‐[4‐(4‐Methylpiperazin‐1‐yl)phenyl]−2‐phenylthiazole‐4‐carboxamide (14f): Off‐white powder, yield (0.06 g, 16%), MP 130°C; ¹H NMR (400 MHz, DMSO‐d ₆) δ 10.07 (s, 1H, thiazole‐CO NH ‐), 8.42 (s, 1H, CH ‐thiazole), 8.15 (s, 2H, ArH), 7.70 (d, J = 8.4 Hz, 2H, ArH), 7.55 (s, 3H, ArH), 6.95 (d, J = 8.6 Hz, 2H, ArH), 3.11 (s, 4H, 2CH₂ of piperazine), 2.45 (s, 4H, 2CH₂ of piperazine), 2.22 (s, 3H, CH₃); ¹³C NMR (100 MHz, DMSO‐d ₆) δ 167.3 (C ² ‐thiazole), 158.6 (thiazole‐ CO ‐), 150.8 (Ar), 147.8 (Ar), 132.5 (Ar), 130.9 (Ar), 130.1 (Ar), 129.3 (Ar), 126.7 (Ar), 124.7 (Ar), 121.6 (Ar), 115.5 (Ar), 54.7 (2CH₂ of piperazine), 48.4 (2CH₂ of piperazine), 45.8 (CH₃); MS m/z (%) for C₂₁H₂₂N₄OS: 378.7 (22.1, M⁺), 73.9 (100.00).

{4‐[(4‐Chlorophenyl)sulfonyl]piperazin‐1‐yl}(2‐phenylthiazol‐4‐yl)methanone (14g): White powder, Yield (0.18 g, 32%), MP 184°C; ¹H NMR (400 MHz, DMSO‐d ₆) δ 8.15 (s, 1H, CH ‐thiazole), 8.01–7.87 (m, 2H, ArH), 7.75 (q, J = 8.4 Hz, 4H, ArH), 7.55 – 7.49 (m, 3H, ArH), 3.90 (s, 2H, CH₂ of piperazine), 3.75 (s, 2H, CH₂ of piperazine), 3.03 (t, J = 5.8 Hz, 4H, 2CH₂ of piperazine); ¹³C NMR (100 MHz, DMSO‐d ₆) δ 166.7 (C ² ‐thiazole), 161.9 (thiazole‐ CO ‐), 150.0 (Ar), 138.4 (Ar), 133.9 (Ar), 132.4 (Ar), 130.7 (Ar), 129.7 (Ar), 129.5 (Ar), 129.3 (Ar), 128.3 (Ar), 126.4 (Ar), 125.2 (Ar), 46.6 (4CH₂ of piperazine); MS m/z (%) for C₂₀H₁₈ClN₃O₃S₂: 449.86 (12.64, M⁺ +2), 447.00 (44.21, M⁺), 248.83 (100.00).

(4‐Phenylpiperazin‐1‐yl)(2‐phenylthiazol‐4‐yl)methanone (14h): Off‐white powder, yield (0.07 g, 20%), MP 135°C; ¹H NMR (400 MHz, DMSO‐d ₆) δ 8.20 (s, 1H, CH ‐thiazole), 7.98 (d, J = 6.4 Hz, 2H, ArH), 7.53 (br s, 3H, ArH), 7.22 (d, J = 7.9 Hz, 2H, ArH), 6.97 (d, J = 8.1 Hz, 2H, ArH), 6.81 (t, J = 7.3 Hz, 1H, ArH), 3.92 (s, 2H, CH₂ of piperazine), 3.81 (s, 2H, CH₂ of piperazine), 3.21 (s, 4H, 2CH₂ of piperazine); ¹³C NMR (100 MHz, DMSO‐d ₆) δ 166.7 (C ² ‐thiazole), 161.9 (thiazole‐ CO ‐), 150.8 (Ar), 150.6 (Ar), 132.5 (Ar), 130.7 (Ar), 129.4 (Ar), 129.0 (Ar), 126.4 (Ar), 124.7 (Ar), 119.4 (Ar), 115.9 (Ar), 49.0 (CH₂ of piperazine), 48.4 (CH₂ of piperazine), 46.5 (CH₂ of piperazine), 42.0 (CH₂ of piperazine); MS m/z (%) for C₂₀H₁₉N₃OS: 349.47 (48.19, M⁺), 306.47 (100.00).

Morpholino(2‐phenylthiazol‐4‐yl)methanone (14i): Off‐white powder, yield (0.12 g, 44%), MP 100°C; ¹H NMR (400 MHz, CDCl₃) δ 7.77 (s, 1H, CH ‐thiazole), 7.75–7.71 (m, 2H, ArH), 7.26 (d, J = 3.0 Hz, 2H, ArH), 7.06 (s, 1H, ArH), 3.89 (s, 1H, CH₂O of morpholine), 3.60 (d, J = 12.6 Hz, 7H, CH₂N of morpholine); ¹³C NMR (100 MHz, CDCl₃) δ 167.5 (C ² ‐thiazole), 162.7 (thiazole‐ CO ‐), 151.2 (Ar), 133.2 (Ar), 130.7 (Ar), 129.3 (Ar), 129.2 (Ar), 126.8 (Ar), 126.7 (Ar), 125.2 (Ar), 67.2 (2 CH₂O of morpholine), 48.1 (CH₂N of morpholine), 43.4 (CH₂N of morpholine); MS m/z (%) for C₁₄H₁₄N₂O₂S: 274.99 (36.48, M⁺), 84.11(100.00); HRMS (ESI‐TOF) m/z: [M ‐ H]^‐ Calcd for C₁₄H₁₄N₂O₂S 273.0703; Found 273.07188.

2‐(4‐Methoxyphenyl)‐N‐[2‐oxo‐2‐(phenylamino)ethyl]thiazole‐4‐carboxamide (14j): White powder, yield (0.06 g, 19%), MP 211°C; ¹H NMR (400 MHz, DMSO‐d ₆) δ 10.08 (s, 1H, Ar NH CO‐), 8.71 (t, J = 6.0 Hz, 1H, thiazole‐CO NH ‐), 8.24 (s, 1H, CH ‐thiazole), 7.99 (d, J = 8.3 Hz, 2H, ArH), 7.61 (d, J = 8.0 Hz, 2H, ArH), 7.31 (t, J = 7.8 Hz, 2H, ArH), 7.15–7.01 (m, 3H, ArH), 4.14 (d, J = 6.0 Hz, 2H, CH₂), 3.84 (s, 3H, CH₃); ¹³C NMR (100 MHz, DMSO‐d ₆) δ 167.4 (ArNH CO ‐), 167.2 (C ² ‐thiazole), 161.3 (Ar C ‐OCH₃), 160.7 (thiazole‐ CO ‐), 150.0 (Ar), 138.8 (Ar), 128.7 (Ar), 128.0 (Ar), 125.2 (Ar), 123.2 (Ar), 123.2 (Ar), 119.2 (Ar), 114.6 (Ar), 55.4 (O CH ₃ ), 42.8 (CH₂); MS m/z (%) for C₁₉H₁₇N₃O₃S: 366.94 (33.87, M⁺), 281.34 (100.00).

2‐(4‐Bromophenyl)‐N‐[2‐oxo‐2‐(phenylamino)ethyl]thiazole‐4‐carboxamide (14k): Off‐white powder, yield (0.16 g, 36%), MP 196°C; ¹H NMR (400 MHz, DMSO‐d ₆) δ 10.10 (s, 1H, Ar NH CO‐), 8.80 (t, J = 6.0 Hz, 1H, thiazole‐CO NH ‐), 8.37 (s, 1H, CH ‐thiazole), 8.02 (d, J = 8.5 Hz, 2H, ArH), 7.76 (d, J = 8.5 Hz, 2H, ArH), 7.60 (d, J = 8.0 Hz, 2H, ArH), 7.31 (t, J = 7.8 Hz, 2H, ArH), 7.05 (t, J = 7.4 Hz, 1H, ArH), 4.13 (d, J = 5.9 Hz, 2H, CH₂); ¹³C NMR (100 MHz, DMSO‐d ₆) δ 167.5 (ArNH CO ‐), 166.1 (C ² ‐thiazole), 160.7 (thiazole‐ CO ‐), 150.4 (Ar), 138.9 (Ar), 132.4 (Ar), 131.7 (Ar), 128.9 (Ar), 128.4 (Ar), 124.9 (Ar), 124.3 (Ar), 123.4 (Ar), 119.2 (Ar), 42.9 (CH₂); MS m/z (%) for C₁₈H₁₄BrN₃O₂S: 418.54 (24.81, M⁺ +2), 416.22 (25.11, M⁺), 262.55 (100.00).

2‐(4‐Chlorophenyl)‐N‐[2‐oxo‐2‐(phenylamino)ethyl]thiazole‐4‐carboxamide (14l): Off‐white powder, yield (0.11 g, 24%), MP 193°C; ¹H NMR (400 MHz, DMSO‐d ₆) δ 10.11 (s, 1H, Ar NH CO‐), 8.80 (t, J = 6.0 Hz, 1H, thiazole‐CO NH ‐), 8.38 (s, 1H, CH ‐thiazole), 8.09 (d, J = 8.1 Hz, 2H, ArH), 7.62 (t, J = 8.6 Hz, 4H, ArH), 7.32 (t, J = 7.7 Hz, 2H, ArH), 7.05 (t, J = 7.4 Hz, 1H, ArH), 4.13 (d, J = 5.9 Hz, 2H, CH₂); ¹³C NMR (100 MHz, DMSO‐d ₆) δ 167.6 (ArNH CO ‐), 166.0 (C ² ‐thiazole), 160.8 (thiazole‐ CO ‐), 150.5 (Ar), 139.0 (Ar), 135.5 (Ar), 131.4 (Ar), 129.5 (Ar), 128.9 (Ar), 128.3 (Ar), 125.0 (Ar), 123.4 (Ar), 119.3 (Ar), 43.0 (CH₂); MS m/z (%) for C₁₈H₁₄ClN₃O₂S: 373.40 (13.44, M⁺ +2), 371.71 (27.63, M⁺), 316.00 (100.00).

2‐(4‐Fluorophenyl)‐N‐[2‐oxo‐2‐(phenylamino)ethyl]thiazole‐4‐carboxamide (14m): Off‐white powder, yield (0.08 g, 17%), MP 204°C; ¹H NMR (400 MHz, DMSO‐d ₆) δ 10.07 (s, 1H, Ar NH CO‐), 8.76 (t, J = 6.0 Hz, 1H, thiazole‐CO NH ‐), 8.31 (s, 1H, CH ‐thiazole), 8.14 – 8.06 (m, 2H, ArH), 7.58 (d, J = 8.0 Hz, 2H, ArH), 7.38 (t, J = 8.6 Hz, 2H, ArH), 7.29 (t, J = 7.8 Hz, 2H, ArH), 7.02 (t, J = 7.4 Hz, 1H, ArH), 4.10 (d, J = 5.9 Hz, 2H, CH₂); ¹³C NMR (100 MHz, DMSO‐d ₆) δ 167.5 (ArNH CO ‐), 166.1 (C ² ‐thiazole), 164.8 (ArC‐F), 162.3 (ArC‐F), 160.7 (thiazole‐ CO ‐), 150.3 (Ar), 138.9 (Ar), 129.2 (Ar), 129.1 (Ar), 128.9 (Ar), 128.8 (Ar), 124.5 (Ar), 123.3 (Ar), 119.2 (Ar), 116.5 (Ar), 116.3 (Ar), 42.9 (CH₂); MS m/z (%) for C₁₈H₁₄FN₃O₂S: 355.14 (32.30, M⁺), 55.06 (100.00).

4.1.3. Synthesis of N‐[2‐oxo‐2‐(phenylamino)ethyl]−4‐phenylthiazole‐2‐carboxamide (18)

To a mixture of carboxylic acid 17 (0.33 g, 1.6 mmol, 1.0 equiv.) and 2‐amino‐N‐ phenylacetamide 8a (0.24 g, 1.6 mmol, 1.0 equiv.) in dimethylformamide (5 mL), EDC.HCl (0.38 g, 2.4 mmol, 1.5 equiv.) and DMAP (0.04 g, 0.32 mmol, 0.2 equiv.) were added. After that, the reaction was stirred at room temperature overnight. After the reaction was completed, the reaction mixture was poured into water, then the formed precipitate was filtrated, washed with methanol, and dried to afford compound 18. Off‐white powder, yield (0.2 g, 37%), MP 208°C; ¹H NMR (400 MHz, DMSO) δ 10.14 (s, 1H, Ar NH CO‐), 9.12 (t, J = 6.1 Hz, 1H, thiazole‐CO NH ‐), 8.45 (s, 1H, CH ‐thiazole), 8.10 (d, J = 7.5 Hz, 2H, ArH), 7.61 (d, J = 8.0 Hz, 2H, ArH), 7.50 (t, J = 7.6 Hz, 2H, ArH), 7.40 (t, J = 7.4 Hz, 1H, ArH), 7.32 (t, J = 7.8 Hz, 2H, ArH), 7.06 (t, J = 7.4 Hz, 1H, ArH), 4.15 (d, J = 6.0 Hz, 2H, CH₂); ¹³C NMR (100 MHz, DMSO‐d ₆) δ 167.1 (ArNH CO ‐), 163.0 (C ² ‐thiazole), 159.5 (thiazole‐ CO ‐), 155.3 (Ar C ‐thiazole), 138.9 (Ar), 133.5 (Ar), 128.9 (Ar), 128.8 (Ar), 128.7 (Ar), 126.3 (Ar), 123.4 (Ar), 119.7 (Ar), 119.2 (Ar), 43.0 (CH₂); MS m/z (%) for C₁₈H₁₅N₃O₂S: 337.61 (31.93, M⁺), 313.98 (100.00).

4.1.4. Synthesis of 2‐benzamido‐N‐(substituted)thiazole‐(4/5)‐carboxamides (22, 26, and 30a‐b)

To a mixture of carboxylic acids 21, 25, and 29a–b (1.0 equiv.) and 2‐amino‐N‐phenylacetamide 8a (1.0 equiv.) in dimethylformamide, EDC.HCl (1.5 equiv.) and DMAP (0.2 equiv.) were added, then the reaction was stirred at room temperature overnight. After the reaction was completed, it was poured into water, then the formed precipitate was filtrated, dried and washed with hexane to afford compounds 22, 26, and 30a‐b in high purity.

2‐Benzamido‐N‐[2‐oxo‐2‐(phenylamino)ethyl]thiazole‐4‐carboxamide (22): White powder, yield (0.2 g, 44%), MP 284°C; ¹H NMR (400 MHz, DMSO‐d ₆) δ 12.88 (s, 1H, thiazole NH CO‐), 10.20 (s, 1H, Ar NH CO‐), 8.12 – 8.09 (m, 3H, 2 ArH and thiazole‐CO NH ), 7.93 (s, 1H, ‐ CH ‐thiazole), 7.68 – 7.53 (m, 5H, ArH), 7.32 (t, J = 7.7 Hz, 2H, ArH), 7.06 (t, J = 7.3 Hz, 1H, ArH), 4.18 (d, J = 5.4 Hz, 2H, CH₂); ¹³C NMR (100 MHz, DMSO‐d ₆) δ 167.3 (ArNH CO ‐), 165.6 (thiazoleNH CO ‐), 160.8 (C ² ‐thiazole), 158.6 (thiazole‐ CO ‐), 144.2 (Ar), 138.8 (Ar), 132.9 (Ar), 131.8 (Ar), 128.9 (Ar), 128.7 (Ar), 128.2 (Ar), 123.4 (Ar), 119.1 (Ar), 118.2 (Ar), 42.8 (CH₂); MS m/z (%) for C₁₉H₁₆N₄O₃S: 380.00 (25.77, M⁺), 112.39 (100.00).

2‐Benzamido‐N‐[2‐oxo‐2‐(phenylamino)ethyl]thiazole‐5‐carboxamide (26): Pink powder, yield (0.1 g, 22%), MP > 300°C; ¹H NMR (400 MHz, DMSO‐d ₆) δ 12.88 (s, 1H, thiazole NH CO‐), 10.09 (s, 1H, Ar NH CO‐), 8.92 (t, J = 5.9 Hz, 1H, thiazole‐CO NH ‐), 8.24 (s, 1H, CH ‐thiazole), 8.11 (d, J = 7.6 Hz, 2H, ArH), 7.68 – 7.53 (m, 5H, ArH), 7.31 (t, J = 7.7 Hz, 2H, ArH), 7.05 (t, J = 7.4 Hz, 1H, ArH), 4.06 (d, J = 5.8 Hz, 2H, CH₂); ¹³C NMR (100 MHz, DMSO‐d ₆) δ 167.7 (ArNH CO ‐), 165.5 (thiazoleNH CO ‐), 161.7 (thiazoleNH CO ‐), 161.3 (C ² ‐thiazole), 139.8 (Ar), 138.9 (Ar), 132.9 (Ar), 131.8 (Ar), 128.8 (Ar), 128.7 (Ar), 128.3 (Ar), 127.4 (Ar), 123.3 (Ar), 119.2 (Ar), 43.0 (CH₂); MS m/z (%) for C₁₉H₁₆N₄O₃S: 380.50 (25.85, M⁺), 66.08 (100.00).

N‐[4‐(2‐Oxo‐2‐{[2‐oxo‐2‐(phenylamino)ethyl]amino}acetyl)thiazol‐2‐yl]benzamide (30a): Yellow powder, yield (0.07 g, 24%), MP 218°C; ¹H NMR (400 MHz, DMSO‐d ₆) δ 13.11 (s, 1H, thiazole NH CO‐), 10.15 (s, 1H, Ar NH CO‐), 9.22 (t, J = 6.1 Hz, 1H, thiazole‐COCO NH ‐), 8.75 (s, 1H, CH ‐thiazole), 8.14 (d, J = 7.7 Hz, 2H, ArH), 7.67 – 7.50 (m, 5H, ArH), 7.32 (t, J = 7.8 Hz, 2H, ArH), 7.06 (t, J = 7.4 Hz, 1H, ArH), 4.08 (d, J = 6.0 Hz, 2H, CH₂); ¹³C NMR (100 MHz, DMSO‐d ₆) δ 180.9 (thiazole‐ CO CONH‐), 166.8 (ArNH CO ‐), 165.8 (thiazoleNH CO ‐), 163.6 (C ² ‐thiazole), 159.0 (thiazole‐CO CO NH‐), 145.2 (Ar), 138.8 (Ar), 133.0 (Ar), 131.5 (Ar), 130.1 (Ar), 128.8 (Ar), 128.7 (Ar), 128.6 (Ar), 128.3 (Ar), 123.5 (Ar), 119.3 (Ar), 42.5 (CH₂); MS m/z (%) for C₂₀H₁₆N₄O₄S: 408.46 (26.33, M⁺), 93.08 (100.00).

N‐[4‐(2‐Oxo‐2‐{[2‐oxo‐2‐(phenylamino)ethyl]amino}ethyl)thiazol‐2‐yl]benzamide (30b): Yellow powder, yield (0.12 g, 40%), MP 255°C; ¹H NMR (400 MHz, DMSO‐d ₆) δ 12.68 (s, 1H, thiazole NH CO‐), 10.02 (s, 1H, Ar NH CO‐), 8.34 (t, J = 5.8 Hz, 1H, thiazole‐CH₂CO NH ‐), 8.13 – 8.06 (m, 2H, ArH), 7.67–7.49 (m, 5H, ArH), 7.30 (t, J = 7.8 Hz, 2H, ArH), 7.08–7.00 (m, 2H, ArH), 3.94 (d, J = 5.7 Hz, 2H, ‐NH CH ₂ CO‐), 3.64 (s, 2H, thiazole‐ CH ₂ CONH‐); ¹³C NMR (100 MHz, DMSO‐d ₆) δ 169.5 (thiazole‐CH₂ CO NH‐), 167.8 (ArNH CO ‐), 165.0 (thiazoleNH CO ‐), 161.4 (C ² ‐thiazole), 158.2 (thiazole‐CO CO NH‐), 145.5 (Ar), 138.9 (Ar), 132.6 (Ar), 132.1 (Ar), 128.8 (Ar), 128.6 (Ar), 128.1 (Ar), 123.3 (Ar), 119.2 (Ar), 110.4 (C ⁵ ‐thiazole), 42.9 (‐NH CH ₂ CO‐), 38.2 (thiazole‐ CH ₂ CONH‐); MS m/z (%) for C₂₀H₁₈N₄O₃S: 394.60 (18.52, M⁺), 93.88 (100.00).

4.2. Biological Screening

4.2.1. Cell Culture and Anti‐proliferative Activity Using MTT Assay

The Centre of Scientific Excellence “Helwan Structural Biology Research, (HSBR)” received the cancer cell lines, which included the colon cancer cell line (HCT‐116), the breast cancer cell line (MCF‐7), and the fibroblast lung cells (WI‐38) [58, 59]. These cell lines were employed with a low passage number of +10 and maintained contamination‐free cultures, which are necessary for cell growth. The initial source of cells was obtained from VACSERA, an Egyptian business that produces vaccinations. Each cell line was cultivated at 37°C with 5% CO₂ in its appropriate medium with 10% fetal bovine serum (FBS) and 1% penicillin/streptomycin. Culture medium was refreshed every 3–4 days. After the cultures reached 85%–90% confluency, they were passaged using the proper methods for downstream applications using a 0.25% trypsin/EDTA solution [60]. The cytotoxic and antiproliferative properties of the synthesized compounds 14d, 14h, and 14i were assessed in vitro against HCT‐116 and WI‐38, in comparison to the standard drug doxorubicin to determine IC₅₀ values and safety index (SI).

4.2.2. EGFR and CDK‐2 Kinase Inhibition Assay

Compounds 14d, 14h, and 14i were investigated for their EGFR and CDK‐2 kinase inhibition assay, using erlotinib and seliciclib as standard chemotherapeutic drugs. The EGFR (catalog #40321) and CDK‐2 (catalog #79599) kinase kits from BPS Bioscience, San Diego, were used at 1 mg/mL concentration, according to the manufacturer's instructions [61].

4.2.3. Flow Cytometry Analysis

The cell cycle and apoptosis were performed in accordance with standard procedures [59]. In summary, the HCT‐116 cancer cell line was cultured for 24 h at a density of 1 × 10⁶ cells per 25 cm² flask. The synthesized compounds 14d, 14h, and 14i were used in these experiments for in vitro IC₅₀ values evaluation after 24‐treatment to achieve the best possible detection of apoptosis while maintaining cell viability, using doxorubicin as a positive control at its IC₅₀ concentration. Cells were concentrated at 1500 × g, suspended in 50 μg/mL propidium iodide (PI) staining solution and 20 g/mL RNaseA, and incubated for 30 min at room temperature for the final analysis to detect cells with a sub‐G₁ DNA content. The flow cytometric analysis was measured using a Cytoflex flow cytometer (Beckman Coulter, USA) and CytExpert software (version 2.4.0.28).

4.2.4. Gene Expression Analysis Using RT‐qPCR

As previously reported, RT‐qPCR was used to assess the gene expression analysis for crucial regulatory genes [58]. The Favor‐PrepTM Blood/Cultured cell total RNA purification mini kit (Favorgen Biotech Corp., Ping‐Tung, Taiwan) was used to extract total RNA from HCT‐116 cancer cells after they were collected and treated with a concentration matching the IC₅₀ of compounds 14d, 14h, and 14i for 24 h. Using the Revert Aid First‐Strand cDNA Synthesis Kit (Thermo Scientific, Waltham, MA, USA), the purified RNA was then reverse transcribed into the first‐strand cDNA. Gene expression analysis was performed to evaluate key regulatory genes (VEGF, AKT‐1, Bcl‐2, Bax, Cyto‐c and Caspase‐3) using HERAPLUS SYBR® Green qPCR Kit (Willowfort, Nottingham, UK). Using β‐actin as the reference gene, the 2^{ˆ −ΔΔCT} method was used to perform differential gene expression [62].

4.3. Molecular Modeling Simulations

4.3.1. Molecular Docking Studies

The molecular operating environment (MOE) 2024.06 [63] was used to carry out the docking process of the lowest energetic conformer of the most potent synthesized compound 14i into the active site of EGFR enzyme (code ID 1M17) and CDK‐2 enzyme (code ID 2A4L). They were downloaded from the RCSB protein data bank and then, they were prepared for molecular docking by automatically adding hydrogen atoms, adding the missed bonds, and fixing potential.

4.3.2. MD Simulations

The ligand‐protein interactions have been investigated using Desmond's MD software, and the stability of the MD simulations was continually evaluated by tracking the RMSDs of the ligand and protein atom sites. For ligands under investigation, the AMBER programs used the AMBER forcefield ff99 [63] to carry out a number of activities, including minimization, counterion addition, solvation, equilibration, and periodic box molecular dynamics simulations in explicit water (TIP4P). The 6‐31 G basis set was used to improve the geometries of the investigated ligands at the B3LYP density functional theory (DFT) level. Functional blends of B3LYP A computationally very accurate description of molecular systems can be obtained by combining Becke's three‐parameter exchange functional with the Lee‐Yang‐Parr correlation functional. The protein–ligand–water system was maintained in a flexible state throughout the simulations. The simulations were conducted ten times independently, changing the initial conditions at random over 10 ns at a time step of 1 fs.

MD simulations were carried out using Desmond software [64]. The temperature was raised from 0 to 300 K using an NPT system. To achieve a total increase of 100 ps, the equilibration phase was completed in 1000 steps. The production phase lasted 100 ns, with a time shift of 100 ps with temperature and pressure remained at 300 K and 1.01325 atm, respectively. The Nose‐Hoover system was used for both complexes' temperature at 300 K and pressure at 1 bar in all NPT ensemble simulations. To analyze the ligands, a 100‐ns simulation with a 1 ps relaxation time lag was run. The settings used during the runs were consistent with the OPLS_2005 force field that was integrated into it. Long‐range electrostatic interactions were assessed using the particle mesh Ewald method, which has a Coulomb cutoff radius of 9.0 Å. The Martyna–Tuckerman–Klein chain coupling method with a 2.0 ps coupling factor was used to regulate pressure.

Funding

The authors have nothing to report.

Consent

All authors contributed directly to the execution and completion of this study. All authors approve the release of this manuscript under a Creative Commons Attribution License.

Conflicts of Interest

The authors declare no conflicts of interest.

Supporting information

Supporting File 1

Supporting File 2

Data Availability Statement

The data that supports the findings of this study are available in the supporting material of this article.