Synthesis of new trans-ferulic acid derivatives as potential anticancer agents and VEGFR-2 inhibitors

Abstract

VEGFR-2 signaling is the primary driver of angiogenesis, a process essential for tumor development and metastasis. Although trans-ferulic acid (TFA), a naturally occurring polyphenol, has demonstrated anticancer and antiangiogenic potential, its low solubility and rapid metabolism limit its therapeutic application. To overcome these drawbacks, seven novel TFA derivatives with chemically masked hydroxyl (–OH) groups were designed and synthesized, aiming to improve their metabolic stability and pharmacokinetic properties. This study evaluates their safety in normal WI-38 fibroblasts and anticancer efficacy in HepG2, Hep3B, and Huh7 hepatocellular carcinoma (HCC) cells. Compound 4e emerged as the lead candidate, demonstrating exceptional cytotoxicity against HCC cells (HepG2: IC₅₀ = 1.8 μg mL⁻¹; Huh7: IC₅₀ = 6.7 μg mL⁻¹, and Hep3B: IC₅₀ = 7.1 μg mL⁻¹) with 23-fold greater potency than TFA and 2-fold superiority to doxorubicin while maintaining minimal toxicity in WI-38 fibroblasts. Further mechanistic studies revealed that 4e significantly modulates key cancer-associated biomarkers in HepG2 lysates, including the downregulation of AFP (α-fetoprotein), BCL-2, MCL-1, and γ-carboxyprothrombin, accompanied by the upregulation of pro-apoptotic caspase-3 and the tumor suppressor P53. The compound also exhibited good inhibitory activity against VEGFR-2, with its binding interaction further supported by molecular docking studies. These findings suggest that compound 4e is a promising anticancer candidate worthy of further therapeutic development research.

By using –OH blocking, seven TFA derivatives were created. Compound 4e regulated AFP, BCL-2, caspase-3, and P53, inhibited VEGFR-2 in docking, and exhibited the maximum cytotoxicity against HepG2, Hep3B, and Huh7 with a safe profile.

Introduction

Trans-Ferulic acid (TFA) is a naturally occurring polyphenolic compound belonging to the hydroxycinnamic acid family.¹ It is widely found in plant-based foods such as whole grains, fruits, vegetables, and medicinal herbs.² As a potent antioxidant, TFA plays a crucial role in neutralizing free radicals, protecting cellular structures from oxidative damage, and modulating various biological processes.³ Due to its structural similarity to other bioactive phenolics, it has garnered attention for its anti-inflammatory,⁴ antimicrobial,⁵ and anticancer^6,7 properties. Research has shown that TFA can influence multiple molecular pathways, making it a promising candidate for cancer therapy.⁸ In cancer research, TFA has demonstrated its ability to target key signaling pathways involved in tumor progression.⁹ It modulates oncogenic and tumor-suppressive molecules, such as p53, BCL-2, and caspases, which regulate cell survival, proliferation, and apoptosis.¹⁰ TFA inhibits cell growth and metastasis by suppressing angiogenesis, invasion, and the epithelial–mesenchymal transition (EMT).¹¹ Additionally, its ability to reduce oxidative stress and inflammation contributes to the suppression of cancer cell survival, making it an attractive compound for therapeutic interventions in various cancers, particularly hepatocellular carcinoma (HCC).

One of TFA's most significant anticancer effects is its role in inducing apoptosis, the programmed cell death essential for eliminating malignant cells. TFA promotes apoptosis through intrinsic (mitochondrial) and extrinsic (death receptor) pathways, enhancing caspase-3 activation and downregulating anti-apoptotic proteins such as BCL-2 and MCL-1.⁸ By restoring the apoptotic balance in cancer cells, TFA effectively reduces tumor progression and improves the efficacy of conventional cancer therapies. With its multifaceted biological effects, TFA holds great promise as a natural anticancer agent, providing a foundation for future research in oncology and therapeutic drug development.¹²

Furthermore, the design and synthesis of TFA-based inhibitors have gained significant attention as a strategy to block key cancer-promoting enzymes and signaling pathways. By modifying TFA's core structure, researchers can develop potent inhibitors targeting kinases (e.g., VEGFR-2),¹¹ histone deacetylases (HDACs),¹³ and matrix metalloproteinases (MMPs),¹⁴ which are critical for cancer cell proliferation, metastasis, and survival. Additionally, TFA derivatives have shown promise in inhibiting cyclooxygenase-2 (COX-2) and NF-κB, which play a major role in inflammation-driven cancers.^15,16 Cancer growth, metastasis, and treatment resistance are all significantly influenced by vascular endothelial growth factor receptor-2 (VEGFR-2), a crucial regulator of tumor angiogenesis. Sorafenib, sunitinib, regorafenib, and lenvatinib are among the small-molecule VEGFR-2 inhibitors that have received clinical approval for the treatment of different types of cancer. Long-term use of these drugs is often linked to serious drawbacks, including dose-limiting toxicities, off-target effects, and the quick development of drug resistance, despite their clinical advantages. Furthermore, poor cardiovascular and gastrointestinal events are frequently caused by inadequate selectivity for VEGFR-2, which jeopardizes patient compliance and therapy results. These difficulties point to a therapeutic need that has not yet been satisfied: the creation of new VEGFR-2 inhibitors with better pharmacokinetic characteristics, lower toxicity, and increased selectivity. In this regard, developing structurally optimized, naturally derived VEGFR-2 inhibitors is a viable way to address the drawbacks of existing treatments and produce anti-angiogenic cancer therapeutics that are safer and more effective.

The rational design of TFA-based small molecules enhances their anticancer activity and provides a foundation for developing targeted therapies with reduced toxicity. The clinical application of TFA is limited due to its rapid metabolic degradation, which is attributed to the presence of hydroxyl groups, thereby restricting its bioavailability and therapeutic potential. Therefore, masking the hydroxyl (–OH) groups of TFA could be a promising strategy to develop new TFA-based derivatives with potentially improved pharmacokinetic and metabolic properties.

Results and discussion

Chemistry

The trans-ferulic acid derivatives (4a–e, 6, and 12) were synthesized from trans-ferulic acid (TFA) (1) and different organobromide compounds (3a–e, 5, and 11) (Schemes 1–3). Initially, different 2-(bromoacetyl)phenyl derivatives (3a–e) were prepared from their respective acetophenones and cupric bromide in a mixture of chloroform and ethyl acetate in a 5 : 1 ratio. The reaction mixture was refluxed for 3 hours (ref. 17–19) (Scheme 1). Using anhydrous potassium carbonate as the base, compound 1 then underwent an S_N2 reaction in acetone with 2-(bromoacetyl)phenyl derivatives (3a–e) to produce the target compounds 4a–e (Scheme 1).

Scheme 1. Synthesis of trans-ferulic acid derivatives (4a–e): reagents and conditions: (i) cupric bromide, chloroform, ethyl acetate reflux 3 h; (ii) anhydrous acetone, K₂CO₃, stirring at room temperature, 12 h.

Scheme 2. Synthesis of trans-ferulic acid derivative (6): reagents and conditions: (i) anhydrous acetone, K₂CO₃, stirring at room temperature, 12 h.

Scheme 3. Synthesis of trans-ferulic acid derivative (12): reagents and conditions: (i) dry toluene, reflux 4 h; (ii) H₂SO₄, stirring 2 h (0–5 °C); (iii) NBS, CCl₄, reflux 3 h; (iv) anhydrous DMF, K₂CO₃, stirring at room temperature, 9 h.

Furthermore, target compound 6 (Scheme 2) was created by an S_N2 reaction between TFA (1) and benzyl bromide (5) in acetone, with anhydrous potassium carbonate serving as the base.

Scheme 3, which includes a crucial benzofuran ring cyclization step, shows an appropriate synthesis pathway for ethyl 3-(bromomethyl)benzofuran-2-carboxylate (11). The process starts with the nucleophile sodium phenoxide (7) and ethyl α-aceto-α-chloroacetate (8) forming an O–C₂ bond in the developing furan ring by an S_N2 reaction. After that, the intermediate α-phenoxy-β-ketoester (9) is dehydrated using sulfuric acid in mild circumstances,²⁰ forming the C₃–C₄ link and completing the synthesis of the benzo[b]furan scaffold (10). N-Bromosuccinimide is then used to brominate the methyl group at position 3 of compound 10, producing the 3-(bromomethyl)benzofuran derivative (11). Lastly, as illustrated in Scheme 3, compound 11 reacts with TFA (1) in acetone via an S_N2 reaction with anhydrous potassium carbonate acting as a base to produce the desired benzofuranyl ester (12).

It is noteworthy from the previously shown Schemes 1–3 that although trans-ferulic acid (1) contains two reactive functional groups (a carboxylic acid and a phenolic hydroxyl), the reaction behavior varies depending on the electrophile used. Both benzyl bromide (5) and phenacyl bromides (3a–e) were found to react with trans-ferulic acid at room temperature in the presence of K₂CO₃, leading to substitution at both the carboxylic acid and phenolic hydroxyl sites. In contrast, ethyl 3-(bromomethyl)benzofuran-2-carboxylate (11) reacts selectively with the carboxylic acid group under the same conditions, without affecting the phenolic hydroxyl group. This selectivity can be attributed to the steric hindrance and reduced electrophilicity of the bromomethyl group within the benzofuran framework, which makes nucleophilic substitution at this position less favorable compared to simpler alkyl bromides. The preservation of the phenolic hydroxyl functionality is clearly supported by the ¹H NMR spectrum, showing a distinct signal for the OH proton at δ 9.65 ppm (compound 12), confirming that this group remains free and unreacted.

The elemental and spectral data (¹H NMR, ¹³C NMR, DEPT-135, and HRMS) supported the target compounds 4a–e, 6, and 12 structures. The spectra are in the supplementary file. Furthermore, the ¹H NMR spectra of compounds 4a–e and 6 confirmed the disappearance of two hydroxyl groups of trans-ferulic acid 1,²¹ which appeared at 12.13 ppm (of COOH group) and 9.54 ppm (of phenolic moiety) (as well as in compound 12), and the appearance of two singlet peaks in the range 5.14–5.67 regarding two OCH₂ groups. On the other hand, ¹³C NMR spectra of target compounds were characterized by the appearance of two peaks in the range 65.9–71.0 regarding two OCH₂ groups and two singlet peaks in the range 191.6–194.6 regarding two ketonic groups.

Due to significant ¹⁹F–¹³C spin–spin interaction, the para-fluorinated compound showed more recognizable peaks.^22,23 The compound, 4b, had coupling values of ¹J_CF = 252.8 Hz, ²J_CF = 22.0 Hz, and ³J_CF = 13.2 Hz. The molecular ion peaks were in good agreement with the molecular formula within the permissible range (±0.4), in addition to the elemental analysis results.

Biological evaluations

Antiproliferative activities on different cancerous cell lines

The MTT assay was used to evaluate the cytotoxic activity of the newly synthesized TFA derivatives (4a–e, 6, and 12) against the human hepatocellular carcinoma (HCC) cell lines HepG2, Hep3B, and Huh7 (ref. 24 and 25) and non-HCC cell lines such as MCF7 (breast cancer) and A549 (lung cancer). To find the IC₅₀ values (the concentration needed to cause 50% cell death) and the percentage of cell death at a fixed concentration of 100 μg mL⁻¹ after 48 hours of exposure, experiments were carried out in triplicate. Doxorubicin was used as the positive control due to its well-established and broad-spectrum cytotoxic activity against hepatocellular carcinoma (HCC) cells and its widespread use as a reference anticancer drug in in vitro cytotoxicity assays. DMSO served as the vehicle control (negative control) (Table 1). Several TFA compounds exhibited moderate to high growth-inhibitory effects on these cell lines, as indicated by the data. Regarding HCC, HepG2 cells (IC₅₀ range: 1.8–72.0 μg mL⁻¹) were more sensitive to the compounds than Huh7 cells (IC₅₀ range: 6.7–81.9 μg mL⁻¹) and Hep3B cells (IC₅₀ range: 5.7–76.3 μg mL⁻¹).

Table 1. The in vitro cytotoxic activity of trans-ferulic acid derivatives on cancerous (HepG2, Hep3B, Huh7, MCF7 and A549) cell lines and in vitro safety assay towards human normal (WI-38) cell lines using MTT assay.

Compound	Cytotoxicity (IC50 μg mL−1)
	HepG2	HuH7	Hep3B	MCF7	A549	WI38
4a	19.8 ± 1.1	29.8 ± 1.6	36.1 ± 1.4	44.2 ± 1.8	26.3 ± 1.0	73.0 ± 1.8
4b	53.3 ± 0.8	68.8 ± 0.8	70.6 ± 1.0	32.7 ± 1.5	40.0 ± 1.2	11.9 ± 0.7
4c	72.0 ± 0.4	81.9 ± 1.2	76.3 ± 1.2	55.9 ± 1.6	49.3 ± 1.7	8.4 ± 0.6
4d	7.6 ± 0.3	27.6 ± 1.4	16.9 ± 0.3	13.3 ± 0.6	16.9 ± 0.2	32.9 ± 0.8
4e	1.8 ± 0.1	6.7 ± 0.3	7.1 ± 0.4	9.7 ± 0.1	7.3 ± 0.2	31.8 ± 0.8
6	45.2 ± 0.8	59.8 ± 0.8	52.0 ± 0.7	67.3 ± 1.7	83.4 ± 0.6	17.6 ± 0.6
12	4.3 ± 0.2	7.2 ± 0.4	5.7 ± 0.3	11.6 ± 0.5	8.4 ± 0.3	77.1 ± 1.3
TFA	40.7 ± 1.7	56.6 ± 2.9	23.6 ± 1.2	75.4 ± 1.9	>100	77.2 ± 2.8
DOX	3.4 ± 0.1	9.0 ± 0.5	1.00 ± 0.02	3.1 ± 0.2	4.2 ± 0.1	23.7 ± 0.9

Concerning the antiproliferative activity against HepG2 cells, the parent compound TFA, containing free carboxylic acid and phenolic OH groups, exhibited moderate activity with an IC₅₀ value of 40.7 ± 1.7 μg mL⁻¹. Introduction of double O-alkylation with benzyl bromide (compound 6) slightly decreased the activity (IC₅₀ = 45.2 ± 0.8 μg mL⁻¹). In contrast, incorporation of two carbonyl groups (compound 4a) significantly enhanced activity (IC₅₀ = 19.8 ± 1.1 μg mL⁻¹), highlighting the positive contribution of carbonyl functionalities. Further modifications at the para position revealed notable structure–activity relationships. Substitution with a methoxy group (compound 4e) dramatically increased potency, achieving an IC₅₀ of 1.8 ± 0.1 μg mL⁻¹, representing a 23-fold improvement over TFA and surpassing doxorubicin (IC₅₀ = 3.4 ± 0.1 μg mL⁻¹). Conversely, para-chloro substitution (compound 4c) markedly decreased activity (IC₅₀ = 72.0 ± 0.4 μg mL⁻¹), while replacement with fluorine (4b) moderately improved activity (IC₅₀ = 53.3 ± 0.8 μg mL⁻¹), and bromine substitution (4d) substantially enhanced it (IC₅₀ = 7.6 ± 0.3 μg mL⁻¹). Mono O-alkylation of the carboxylic acid group of TFA with a bromomethyl benzofuran moiety (compound 12) also resulted in increased antiproliferative effect (IC₅₀ = 4.3 ± 0.2 μg mL⁻¹), indicating that selective mono-substitution can be beneficial. Overall, these findings underscore the critical influence of carbonyl introduction, para-substituent nature, and selective O-alkylation on enhancing HepG2 inhibitory activity (Fig. 1).

Fig. 1. Potential inhibitory activity of target compounds 4a–e, 6, and 12 against the HepG2 cancer cell line. The red arrow denotes reduced activity, whereas the green arrow denotes enhanced activity.

A comparable cytotoxicity trend to that observed in HepG2 cells was also evident in Huh7 and Hep3B cell lines. Compounds 4a, 4d, 4e, and 12 emerged as the most active derivatives, exhibiting markedly higher potency than the parent trans-ferulic acid (TFA) (IC₅₀ = 56.6 μg mL⁻¹ and 23.6 μg mL⁻¹ in Huh7 and Hep3B cells, respectively). Among these, compound 4e showed the strongest cytotoxic effect (IC₅₀ = 6.7 μg mL⁻¹ in Huh7 and 7.1 μg mL⁻¹ in Hep3B), followed by compound 12 (IC₅₀ = 7.2 μg mL⁻¹ and 5.7 μg mL⁻¹, respectively). In contrast, compounds 4d and 4a displayed moderate activity with IC₅₀ values of 27.6 μg mL⁻¹ and 16.9 μg mL⁻¹ for 4d, and 29.8 μg mL⁻¹ and 36.1 μg mL⁻¹ for 4a against Huh7 and Hep3B cells, respectively.

To further evaluate the cytotoxic potential of the tested compounds beyond hepatic-derived cancer models, their effects were examined on non-hepatic cancer cell lines, specifically MCF-7 (breast cancer) and A549 (lung cancer). As expected, the overall sensitivity of these non-hepatic cancer cells was lower compared to HepG2 cells, indicating a degree of selectivity of the compounds toward hepatic-derived cancer cells. Despite this reduced sensitivity, compounds 4e and 12 exhibited the most pronounced cytotoxic effects among the series, with IC₅₀ values of 9.7 μg mL⁻¹ (MCF-7) and 7.3 μg mL⁻¹ (A549) for 4e, and 11.6 μg mL⁻¹ (MCF-7) and 8.4 μg mL⁻¹ (A549) for 12. These findings suggest that, although less potent against non-hepatic lines, these two compounds retain significant anticancer activity across different tumor types.

This activity pattern suggests a clear structure–activity relationship among the tested HCC cell lines. Furthermore, the findings revealed that the structural modifications performed for TFA markedly enhanced its anticancer potency, particularly against HepG2 cells, with compound 4e emerging as the most promising molecule demonstrating 23- and 2-fold enhanced activity compared to TFA and doxorubicin, respectively.

This emphasizes its potential as a promising candidate for HCC therapy and motivates us to conduct further mechanistic studies to investigate the possible mode of action.

In vitro cytotoxic assay towards human normal WI-38 cell lines

To evaluate the potential cytotoxicity toward normal cells, the prepared compounds (4a–e, 6, and 12) were examined using the non-cancerous WI-38 cell line. The safety evaluation of TFA derivatives (4a–e, 6, 12) on normal WI-38 fibroblasts revealed critical insights into their therapeutic potential. Key derivatives 4b, 4d, 4e, and 12 demonstrated negligible cytotoxicity against non-cancerous cells (Table 1), with compound 4e exhibiting particularly low toxicity. These results suggest that strategic chemical modifications (e.g., hydroxyl group blocking) successfully improved biosafety parameters. The >10-fold differential in cytotoxicity between HepG2 cells and WI-38 fibroblasts for lead compound 4e indicates a robust therapeutic window, which validates the design rationale of modifying TFA's phenolic structure to reduce non-specific cellular toxicity while enhancing anticancer potency. The preserved viability of WI-38 cells at therapeutic concentrations supports further preclinical evaluation of these derivatives as targeted therapies for HCC.

Effect of trans-ferulic acid derivatives on cancer hallmark markers assay

Tumor markers are crucial for cancer diagnosis, prognosis, and monitoring therapy. A well-known biomarker that aids in early detection and disease monitoring is alpha-fetoprotein (AFP), which increases in hepatocellular carcinoma (HCC) and certain germ cell tumors.^26,27 The anti-apoptotic protein BCL-2 (B-cell lymphoma 2) is frequently overexpressed in various malignancies, including solid tumors and lymphomas, which enables tumor cells to evade programmed cell death. On the other hand, cancer typically results in downregulated levels of caspase-3, a crucial pro-apoptotic enzyme, which inhibits apoptosis and allows uncontrolled cell division. Leukemia, lymphoma, and other cancers are often associated with increased MCL-1 (myeloid cell leukemia-1), another anti-apoptotic protein that promotes cell survival. The “guardian of the genome”, the p53 tumor suppressor gene, is frequently altered or inactivated in various malignancies, resulting in the loss of its ability to control cell cycle arrest and induce cell death. Aggressive malignancies frequently overexpress urokinase plasminogen activator (uPA), a protease involved in the breakdown of the extracellular matrix and metastasis, which promotes tumor invasion.

Last but not least, a crucial biomarker in hepatocellular carcinoma is γ-carboxyprothrombin (DCP), also known as des-γ-carboxy prothrombin, which increases as a result of impaired liver function leading to defective vitamin K-dependent carboxylation. Knowledge of such tumor markers enhances diagnostic and therapeutic approaches, offering important insights into cancer biology. As presented in Table 1, the top four trans-ferulic acid derivatives (4b, 4d, 4e, and 12) were evaluated for their effects on various cancer hallmark markers in HepG2 cell lysates (Table 2).

Table 2. Effects of treatment with trans-ferulic acid derivatives on the levels of cancer hallmark markers in cell lysate from HepG2 cells treated with the selected compounds at their IC₅₀.

Codes	Parameter
	AFP (ng mL−1)	BCL-2 (ng mL−1)	Caspase-3 (ng mL−1)	MCL-1 (pg mL−1)	P53 (pg mL−1)	Urokinase (ng mL−1)	DCP (ng mL−1)
4a	1756	10.6	257	646	547	2302	0.96
	(±8)	(±0.4)	(±9)	(±13)	(±11)	(±36)	(±0.06)
4d	1880	6.3	236	400	434	1753	0.68
	(±16)	(±0.15)	(±17)	(±13)	(±22)	(±18)	(±0.07)
4e	921	3.3	469	206	1018	952	0.17
	(±15)	(±0.2)	(±11)	(±10)	(±19)	(±29)	(±0.01)
12	998	3.53	413	229	722	1002	0.61
	(±17)	(±0.4)	(±15)	(±8)	(±11)	(±19)	(±0.03)
DMSO a	3545	15.0	131	943	246	3847	1.64
	(±24)	(±0.5)	(±15)	(±29)	(±14)	(±34)	(±0.10)
DOX b	612	2.69	599	153	1282	709	0.25
	(±12)	(±0.45)	(±17)	(±9)	(±16)	(±27)	(±0.04)
TFA	1459	4.67	282	345	556	1585	0.29
	(±20)	(±0.4)	(±16)	(±9)	(±13)	(±34)	(±0.02)

^aDifference from non-treated HepG2 cells (negative control).

^bDifference from doxorubicin-treated HepG2 cells (positive control).

AFP results

The addition of doxorubicin, a trans-ferulic acid derivative, 4e, separately into the medium of HepG2, significantly decreased AFP activity compared to the negative control group and TFA, such as DOX (Fig. 2).

Fig. 2. AFP levels in HepG2 cells treated with the promising compounds 4a, 4d, 4e, and 12. Data are shown as mean ± SD (n = 3) and compared to positive and negative controls, as indicated.

BCL-2 results

Anti-apoptotic repression via B-cell lymphoma-2 (BCL-2) antigen level restoration, the addition of doxorubicin, compound 4e, separately into the medium of HepG2, significantly decreased BCL-2 activity in comparison to the negative control group and TFA such as DOX (Fig. 3).

Fig. 3. BCL-2 levels in HepG2 cells treated with the promising compounds 4a, 4d, 4e, and 12. Data are shown as mean ± SD (n = 3) and compared to positive and negative controls, as indicated.

Caspase-3 results

Addition of doxorubicin, TFA derivative 4e, separately into the medium of HepG2, significantly increases caspase-3 activity in comparison to the negative control group and significantly increases caspase-3 level in comparison to TFA, such as DOX (Fig. 4).

Fig. 4. Caspase-3 levels in HepG2 cells treated with the promising compounds 4a, 4d, 4e, and 12. Data are shown as mean ± SD (n = 3) and compared to positive and negative controls, as indicated.

MCL-1 results

The addition of doxorubicin, trans-ferulic acid derivative, 4e, separately into the medium of HepG2 significantly decreased MCL-1 activity in comparison to the negative control group and TFA, such as DOX (Fig. 5).

Fig. 5. MCL-1 levels in HepG2 cells treated with the promising compounds 4a, 4d, 4e, and 12. Data are shown as mean ± SD (n = 3) and compared to positive and negative controls, as indicated.

P53 results

The addition of doxorubicin, TFA analogue 4e, separately into the medium of HepG2 significantly increases P53 activity compared to negative control group and TFA such as DOX (Fig. 6).

Fig. 6. P53 levels in HepG2 cells treated with the promising compounds 4a, 4d, 4e, and 12. Data are shown as mean ± SD (n = 3) and compared to positive and negative controls, as indicated.

Urokinase results

Anti-invasion testing via testing the protease urokinase plasminogen activator (uPA), addition of doxorubicin, compound 4e, separately into the medium of HepG2 significantly decreased urokinase activity in comparison to negative and TFA such as DOX (Fig. 7).

Fig. 7. Urokinase levels in HepG2 cells treated with the promising compounds 4a, 4d, 4e, and 12. Data are shown as mean ± SD (n = 3) and compared to positive and negative controls, as indicated.

γ-Carboxyprothrombin results

The addition of doxorubicin, compound 4e, separately into the medium of HepG2 significantly decreased γ-carboxyprothrombin (DCP) activity compared to the negative control and TFA, such as DOX (Fig. 8).

Fig. 8. γ-Carboxyprothrombin levels in HepG2 cells treated with the promising compounds 4a, 4d, 4e, and 12. Data are shown as mean ± SD (n = 3) and compared to positive and negative controls, as indicated.

VEGFR-2 kinase inhibition

To determine its possible molecular target, compound 4e's inhibitory effect against VEGFR-2 was assessed. With an IC₅₀ value of 0.618 μM, compound 4e exhibited significant VEGFR-2 inhibition, as indicated in Table 3, which is similar to that of sorafenib (IC₅₀ = 0.560 μM).

Table 3. IC₅₀ value for the most potent cytotoxic compound's inhibition of VEGFR-2 kinase (4e).

Compound	IC50 ± SDa (μM)
	VEGFR-2
4e	0.618 ± 0.04
Sorafenib	0.560 ± 0.01

^aThe values are the average of three separate tests. A sigmoidal non-linear regression analysis of % inhibition across five distinct concentrations of each chemical was used to calculate the IC₅₀ values.

Assessment of the impact of derivative 4e on VEGFR-2 total expression

Compound 4e's IC₅₀ concentration (1.8 μg mL⁻¹) was used to assess its impact on VEGFR-2 expression in the HepG2 cancer cell line, whereas sorafenib was used as a positive control. In comparison to untreated control cells, the data showed that sorafenib reduced VEGFR-2 expression by 25.0%, whereas compound 4e reduced it by 17.5%. These results suggest that compound 4e has the ability to downregulate VEGFR-2 levels (Fig. 9).

Fig. 9. Diagram showing the effect of compound 4e on VEGFR-2 concentrations in treated HepG2 cells compared to sorafenib.

Assessment of the effect of compound 4e on VEGFR-2 phosphorylation

At a dose of 1.8 μg mL⁻¹, compound 4e's capacity to prevent VEGFR-2 phosphorylation was examined in HepG2 cells. In comparison to untreated control cells, compound 4e treatment led to a 69.2% inhibition of VEGFR-2 phosphorylation, whereas sorafenib caused a 73.0% inhibition. All of these findings demonstrate that compound 4e efficiently lowers overall VEGFR-2 and phosphorylated VEGFR-2 (p-VEGFR-2) in HepG2 cells (Fig. 10).

Fig. 10. Bar graph illustration for the impact of compound 4e on VEGFR-2 phosphorylation in HepG2 cells in comparison to sorafenib.

Assessment of the effect of compound 4e on the PI3K/Akt/mTOR pathway

The progression of cancer, including angiogenesis, migration, metastasis, and cell proliferation, is significantly influenced by the PI3K/Akt/mTOR signaling pathway. The impact on phosphorylation within the PI3K/Akt/mTOR pathway was assessed in HepG2 cells at 1.8 μg mL⁻¹ in order to investigate the molecular mechanism of apoptosis induction by compound 4e and identify pertinent pathway targets. In comparison to untreated controls, compound 4e dramatically reduced PI3K phosphorylation by 77.6%, whilst sorafenib accomplished a 79.2% inhibition. Furthermore, in contrast to sorafenib, which inhibited p-AKT and p-mTOR by 75.4% and 72.1%, respectively, compound 4e decreased phosphorylation of AKT and mTOR by 74.9% and 72.5% (Fig. 11). These findings show that compound 4e significantly reduces the phosphorylation of important PI3K/Akt/mTOR pathway proteins.

Fig. 11. The effect of compound 4e on phosphorylation of the PI3K/Akt/mTOR pathway in HepG2 cells compared to sorafenib.

To evaluate the kinase selectivity of the target compounds, compound 4e was screened at 10 μM against a panel of 5 kinases, and the percentage inhibition for each enzyme was measured (Table 4). The results revealed that compound 4e did not exhibit notable inhibition toward any of the tested kinases. Overall, these findings indicate that 4e possesses favorable selectivity for VEGFR-2 kinase.

Table 4. Percent activity values of compound 4e against a small panel of five kinases at 10 μM.

Kinase	BTK	FLT-3	EGFR	JAK3	Aurora B
Kinase activity (%)	85	93	68	79	99

Computational studies

In silico ADMET profile of compound 4e

The ADMETlab 3.0 platform was utilized to predict the absorption, distribution, metabolism, excretion, and toxicity (ADMET) properties of the developed compounds, thereby evaluating their pharmacokinetic behavior and drug-likeness. One of the most important variables affecting a compound's efficacy in vivo is its metabolic stability. Human liver microsomal (HLM) stability, in particular, indicates a compound's vulnerability to cytochrome P450-mediated hepatic metabolism. While a high HLM stability implies resistance to enzymatic breakdown and a potentially longer half-life, a low HLM stability denotes rapid metabolic degradation, resulting in poor bioavailability.

Accordingly, the HLM stability model predicted that trans-ferulic acid would be red, indicating a high metabolic liability. Free carboxylic (–COOH) and phenolic hydroxyl (–OH) groups, which are vulnerable to enzymatic oxidation and conjugation processes, are the cause of this. On the other hand, compound 4e yielded a green color in the HLM stability prediction, indicating that both the carboxyl and phenolic hydroxyl functions were chemically masked. This suggests increased metabolic stability, which is likely a consequence of the decreased accessibility of metabolic enzymes to these reactive sites, thereby improving the compound's overall pharmacokinetic profile.

Molecular docking of compound 4e against VEGFR-2 TK

The binding mode of compound 4e exhibited a binding energy of −9.57 kcal mol⁻¹ against VEGFR-2 TK. Three hydrophobic π–π and π–sigma interactions were observed with Phe918, Leu840, and Phe1047. Additionally, sixteen π–alkyl interactions were conducted with Phe918, Cys919, Ala866, Leu1035, Val916, Val848, Cys1045, Leu889, Val899, Ile892, and Val898 amino acids side chain. Moreover, compound 4e formed a hydrogen bond through the interaction of Asp1046 with the oxygen atom in the ether-like group, with a distance of 2.34 Å (Fig. 12). Furthermore, the co-crystalized ligand complexed with VEGFR-2 TK (sorafenib) exhibited an affinity score of −10.14 kcal mol⁻¹ (Table 5) with an RMSD value equal to 0.15 Å, which indicates the docking process was valid (Fig. S26), sorafenib formed thirteen hydrophobic π–alkyl and π–π interactions with Phe918, Leu840, Phe1047, Val916, Val848, Lys868, Leu889, Ile1044, Leu1019, Cys1045, Ala866, and Cys919, additionally, the interaction was supported by five hydrogen bonds with essential pharmacophoric amino acids side chain as Cys919, Asp1046, and Glu885 with distances of 2.04, 2.07, 1.82, 2.34 and 1.74 Å (Fig. 13).

Fig. 12. 3D figure of the proposed binding mode of compound 4e against VEGFR-2 TK. The tested compound is colored (turquoise), and the amino acid side chain is colored (yellow).

Table 5. Molecular docking analysis of the tested compound 4e against VEGFR-2 TK.

Codes	RMSD value (Å)	Docking (affinity) score (kcal mol−1)
4e	1.18	−9.57
Sorafenib	0.15	−10.14

Fig. 13. The proposed binding mode of the co-crystalized ligand (sorafenib) is complexed with VEGFR-2 TK. Sorafenib is colored (brown), and the amino acid side chain is colored (yellow).

Conclusions

This study presents the rational design and biological evaluation of hydroxyl-masked trans-ferulic acid (TFA) derivatives aimed at overcoming the intrinsic metabolic limitations of the parent compound. Among the synthesized series, compound 4e emerged as the most promising candidate, exhibiting potent cytotoxic activity against hepatocellular carcinoma (HCC) cell lines while maintaining a favorable safety profile toward normal WI-38 cells. Molecular docking analysis revealed that 4e effectively occupies the VEGFR-2 ATP-binding site through stable and specific interactions, supporting its role as a VEGFR-2 inhibitor. Notably, in contrast to sorafenib, a clinically approved multi-kinase inhibitor associated with off-target toxicity and resistance development, compound 4e represents a structurally simpler, natural product-derived scaffold with potentially improved selectivity toward VEGFR-2. The masking of hydroxyl groups in 4e is expected to enhance metabolic stability, which may contribute to improved pharmacokinetic behavior compared to the parent TFA and existing VEGFR-2 inhibitors. Collectively, these features highlight compound 4e as a promising lead structure for further optimization as a safer and more selective anti-angiogenic agent. Future studies will focus on in vivo pharmacokinetic and toxicity evaluations to further validate its therapeutic potential.

Experimental

Chemistry

Unless otherwise noted, all materials were purchased from commercial vendors and utilized without additional purification. Thin-layer chromatography (TLC) on FLUKA silica gel aluminum plates (0.2 mm thick) with a fluorescent indicator at 254 nm was used to monitor the reaction and evaluate the purity of the final product. Melting points were recorded uncorrected using an SMP10 device (Stuart, Japan). A PerkinElmer-1430 spectrometer (PerkinElmer, USA) was used to gather FTIR spectra. Using DMSO-d₆ as the solvent, ¹H NMR, ¹³C NMR, and DEPT-135 spectra were captured on Bruker spectrometers (500 MHz for ¹H and 125 MHz for ¹³C; Bruker, Japan). The PerkinElmer 2400 analyzer (PerkinElmer, USA) was used to do elemental analysis. Samples were burned in a sealed combustion tube with pure oxygen to analyze the carbon and hydrogen. A thermal conductivity detector was used to examine the mixed gases at atmospheric pressure after the resultant CO₂ and H₂O had been treated using particular reagents. Bruker MicroTOF was used to obtain high-resolution mass spectra (HRMS). The SI contains spectral data and the interpretations that go with it.

General procedure for the synthesis of 2-(bromoacetyl)phenyl derivatives (3a–e)

In a rounded-bottom flask, different acetophenones (10 mmol) and cupric bromide (2.46 g, 11 mmol) were added to a mixture of chloroform : ethyl acetate 25 : 5 mL. The reaction mixture was refluxed for 3 hours. The progress of the reaction was monitored by TLC (n-hexane : ethyl acetate (1 : 2)). The resulting reaction mixture was filtrated, washed with ethanol, and finally dried under vacuum to give derivatives 3a–e.¹⁹

The general synthesis of target analogues (4a–e and 6)

Using anhydrous acetone as the solvent, trans-ferulic acid (1) (0.19 g, 1.2 mmol) was reacted with a mixture of 2-(bromoacetyl)phenyl derivatives (3a–e) or benzyl bromide (5) (1.0 mmol) in the presence of potassium carbonate (0.21 g, 1.5 mmol) as a base. For twelve hours, the reaction mixture was agitated at room temperature. Compounds 4a–e and 6 were obtained by recrystallizing the residue from 95% ethanol after it had been triturated with 20 mL of cold water and agitated for another half hour.

2-Oxo-2-phenylethyl (E)-3-(3-methoxy-4-(2-oxo-2-phenylethoxy)phenyl)acrylate (4a)

Yield (93%) as a white powder, with Mp: 101–102 °C.²⁸¹H NMR (500 MHz, DMSO-d₆) δ (ppm): 8.05–8.00 (m, 4H), 7.73–7.69 (m, 3H), 7.59 (t, J = 7.6 Hz, 4H), 7.46 (d, J = 2.1 Hz, 1H), 7.25 (dd, J = 8.4, 2.0 Hz, 1H), 6.95 (d, J = 8.4 Hz, 1H), 6.73 (d, J = 16.1 Hz, 1H), 5.67 (s, 2H, OCH₂), 5.61 (s, 2H, OCH₂), 3.87 (s, 3H, OCH₃). ¹³C NMR (125 MHz, DMSO-d₆) δ (ppm): 194.6, 193.4, 166.4, 150.2, 149.5, 145.9, 134.8, 134.4, 134.4, 134.3, 129.4, 129.3, 128.3, 128.2, 127.7, 123.2, 115.5, 113.5, 111.5, 71.0, 66.9, 56.2. ¹³C NMR-DEPT (125 MHz, DMSO-d₆) at 135°, the CH and CH₃ groups were detected as positive (upward) signals and the CH₂ groups as negative (downward) signals. The following chemical shifts were noted at δ: 145.9 (Arom. ↑), 134.4 (Arom. ↑), 134.3 (Arom. ↑), 129.4 (Arom. ↑), 129.3 (Arom. ↑), 128.3 (Arom. ↑), 128.2 (Arom. ↑), 123.2 (Arom. ↑), 115.5 (Arom. ↑), 113.5 (Arom. ↑), 111.5 (Arom. ↑), 71.0 (OCH₂ ↓), 66.9 (OCH₂ ↓), 56.2 (OCH₃ ↑). HRMS (ESI): m/z: [M + Na]⁺ calcd. 453.1309 and found 453.1320. Anal. calcd. for C₂₆H₂₂O₆: C, 72.55; H, 5.15. Found: C, 72.39; H, 5.12%.

2-(4-Fluorophenyl)-2-oxoethyl (E)-3-(4-(2-(4-fluorophenyl)-2-oxoethoxy)-3-methoxyphenyl)acrylate (4b)

Yield (88%) as a white powder, with Mp: 105–107 °C. ¹H NMR (500 MHz, DMSO-d₆) δ (ppm): 8.14–8.08 (m, 4H), 7.67 (d, J = 16.0 Hz, 1H), 7.46–7.40 (m, 5H), 7.24 (dd, J = 8.3, 2.0 Hz, 1H), 6.96 (d, J = 8.4 Hz, 1H), 6.73 (d, J = 15.9 Hz, 1H), 5.65 (s, 2H, OCH₂), 5.60 (s, 2H, OCH₂), 3.87 (s, 3H, OCH₃). ¹³C NMR (125 MHz, DMSO-d₆) δ (ppm): 193.3, 192.1, 166.8, 166.8, 166.4, 164.8 (d, ¹J_CF = 252.8 Hz), 164.8, 150.1, 149.5, 145.9, 131.5, 131.5, 131.5, 131.4, 131.4, 131.3, 131.2, 131.2 (d, ³J_CF = 13.2 Hz), 127.7, 123.2, 116.6, 116.4, 116.4, 116.2 (d, ²J_CF = 22.0 Hz), 115.5, 113.5, 111.5, 70.9, 66.8, 56.2. ¹³C NMR-DEPT (125 MHz, DMSO-d₆) at 135°, the CH and CH₃ groups were detected as positive (upward) signals and the CH₂ groups as negative (downward) signals. The following chemical shifts were noted at δ: 145.9 (Arom. ↑), 131.5 (Arom. ↑), 131.4 (Arom. ↑), 131.4 (Arom. ↑), 131.3 (Arom. ↑), 123.2 (Arom. ↑), 116.6 (Arom. ↑), 116.4 (Arom. ↑), 116.4 (Arom. ↑), 116.2 (Arom. ↑), 115.5 (Arom. ↑), 113.5 (Arom. ↑), 111.5 (Arom. ↑), 70.9 (OCH₂ ↓), 66.8 (OCH₂ ↓), 56.2 (OCH₃ ↑). HRMS (ESI): m/z: [M + Na]⁺ calcd. 489.1120 and found 489.1129. Anal. calcd. for C₂₆H₂₀F₂O₆: C, 66.95; H, 4.32. Found: C, 67.12; H, 4.29%.

2-(4-Chlorophenyl)-2-oxoethyl (E)-3-(4-(2-(4-chlorophenyl)-2-oxoethoxy)-3-methoxyphenyl)acrylate (4c)

Yield (85%) as a white powder, with Mp: 117–119 °C. ¹H NMR (500 MHz, DMSO-d₆) δ (ppm): 8.06–8.01 (m, 4H), 7.67–7.65 (m, 5H), 7.46 (d, J = 2.0 Hz, 1H), 7.24 (dd, J = 8.5, 2.0 Hz, 1H), 6.97 (d, J = 8.4 Hz, 1H), 6.73 (d, J = 15.9 Hz, 1H), 5.65 (s, 2H, OCH₂), 5.60 (s, 2H, OCH₂), 3.87 (s, 3H, OCH₃). HRMS (ESI): m/z: [M + Na]⁺ calcd. 521.0529 and found 521.0551 and [M + Na + 2]⁺ calcd. 523.0506 and found 523.0530. Anal. calcd. for C₂₆H₂₀Cl₂O₆: C, 62.54; H, 4.04. Found: C, 62.73; H, 4.02%.

2-(4-Bromophenyl)-2-oxoethyl (E)-3-(4-(2-(4-bromophenyl)-2-oxoethoxy)-3-methoxyphenyl)acrylate (4d)

Yield (82%) as a white powder, with Mp: 128–129 °C. ¹H NMR (500 MHz, DMSO-d₆) δ (ppm): 7.98–7.92 (m, 4H), 7.81 (s, 2H), 7.79 (s, 2H), 7.67 (d, J = 15.9 Hz, 1H), 7.46 (d, J = 2.0 Hz, 1H), 7.24 (dd, J = 8.5, 2.0 Hz, 1H), 6.96 (d, J = 8.5 Hz, 1H), 6.72 (d, J = 16.0 Hz, 1H), 5.64 (s, 2H, OCH₂), 5.59 (s, 2H, OCH₂), 3.87 (s, 3H, OCH₃). ¹³C NMR (125 MHz, DMSO-d₆) δ (ppm): 194.0, 192.8, 166.4, 150.1, 149.5, 146.0, 133.7, 133.4, 132.4, 132.3, 130.4, 130.2, 128.5, 128.3, 127.8, 123.2, 115.5, 113.6, 111.5, 71.0, 66.8, 56.2. ¹³C NMR-DEPT (125 MHz, DMSO-d₆) at 135°, the CH and CH₃ groups were detected as positive (upward) signals and the CH₂ groups as negative (downward) signals. The following chemical shifts were noted at δ: 146.0 (Arom. ↑), 132.4 (Arom. ↑), 132.3 (Arom. ↑), 130.4 (Arom. ↑), 130.2 (Arom. ↑), 123.2 (Arom. ↑), 115.4 (Arom. ↑), 113.6 (Arom. ↑), 111.5 (Arom. ↑), 71.0 (OCH₂ ↓), 66.8 (OCH₂ ↓), 56.2 (OCH₃ ↑). HRMS (ESI): m/z: [M + Na]⁺ calcd. 608.9519 and found 608.9540 and [M + Na + 2]⁺ calcd. 610.9501 and found 610.9511. Anal. calcd. for C₂₆H₂₀Br₂O₆: C, 53.09; H, 3.43. Found: C, 52.91; H, 3.46%.

2-(4-Methoxyphenyl)-2-oxoethyl (E)-3-(3-methoxy-4-(2-(4-methoxyphenyl)-2-oxoethoxy)phenyl)acrylate (4e)

Yield (92%) as a white powder, with Mp: 113–114 °C. ¹H NMR (500 MHz, DMSO-d₆) δ (ppm): 8.03–7.97 (m, 4H), 7.66 (d, J = 15.9 Hz, 1H), 7.45 (d, J = 2.0 Hz, 1H), 7.23 (dd, J = 8.5, 2.0 Hz, 1H), 7.11–7.08 (m, 4H), 6.91 (d, J = 8.4 Hz, 1H), 6.71 (d, J = 16.0 Hz, 1H), 5.58 (s, 2H, OCH₂), 5.54 (s, 2H, OCH₂), 3.87 (s, 9H, 3OCH₃). ¹³C NMR (125 MHz, DMSO-d₆) δ (ppm): 192.9, 191.6, 166.4, 164.1, 164.0, 150.2, 149.5, 145.8, 130.7, 130.6, 127.7, 127.3, 123.2, 115.6, 114.6, 114.5, 113.4, 111.5, 70.7, 66.5, 56.2, 56.1. ¹³C NMR-DEPT (125 MHz, DMSO-d₆) at 135°, the CH and CH₃ groups were detected as positive (upward) signals and the CH₂ groups as negative (downward) signals. The following chemical shifts were noted at δ: 145.8 (Arom. ↑), 130.7 (Arom. ↑), 130.6 (Arom. ↑), 123.2 (Arom. ↑), 115.6 (Arom. ↑), 114.6 (Arom. ↑), 114.5 (Arom. ↑), 113.4 (Arom. ↑), 111.5 (Arom. ↑), 70.7 (OCH₂ ↓), 66.5 (OCH₂ ↓), 56.2 (OCH₃ ↑), 56.0 (2OCH₃ ↑). HRMS (ESI): m/z: [M + Na]⁺ calcd. 513.1520 and found 513.1492. Anal. calcd. for C₂₈H₂₆O₈: C, 68.56; H, 5.34. Found: C, 68.38; H, 5.35%.

Benzyl (E)-3-(4-(benzyloxy)-3-methoxyphenyl)acrylate 6

Yield (87%) as a white powder, with Mp: 95–97 °C.²⁹¹H NMR (500 MHz, DMSO-d₆) δ (ppm): 7.64 (d, J = 15.9 Hz, 1H), 7.47–7.31 (m, 11H), 7.24 (dd, J = 8.4, 2.0 Hz, 1H), 7.07 (d, J = 8.4 Hz, 1H), 6.64 (d, J = 15.9 Hz, 1H), 5.22 (s, 2H, OCH₂), 5.14 (s, 2H, OCH₂), 3.82 (s, 3H, OCH₃). ¹³C NMR (125 MHz, DMSO-d₆) δ (ppm): 166.8, 150.4, 149.7, 145.4, 137.2, 136.8, 128.9, 128.9, 128.5, 128.5, 128.4, 128.3, 127.6, 123.3, 115.9, 113.5, 111.1, 70.2, 65.9, 56.1. ¹³C NMR-DEPT (125 MHz, DMSO-d₆) at 135°, the CH and CH₃ groups were detected as positive (upward) signals and the CH₂ groups as negative (downward) signals. The following chemical shifts were noted at δ: 145.4 (Arom. ↑), 128.9 (Arom. ↑), 128.9 (Arom. ↑), 128.5 (Arom. ↑), 128.5 (Arom. ↑), 128.4 (Arom. ↑), 128.3 (Arom. ↑), 123.3 (Arom. ↑), 115.9 (Arom. ↑), 113.5 (Arom. ↑), 111.1 (Arom. ↑), 70.2 (OCH₂ ↓), 65.9 (OCH₂ ↓), 56.1 (OCH₃ ↑). HRMS (ESI): m/z: [M + Na]⁺ calcd. 397.1410 and found 397.1424. Anal. calcd. for C₂₄H₂₂O₄: C, 76.99; H, 5.92. Found: C, 77.18; H, 5.89%.

Preparation of ethyl 3-methylbenzofuran-2-carboxylate (10)

Sulfuric acid was added to the ester 9 (0.44 g, 2 mmol) in an ice-cooled flask that was kept between 0 and 5 °C for two hours. After that, the reaction mixture was gradually added to distilled water while being constantly stirred. To separate the benzofuran derivative 10, the resultant aqueous layer was divided into three sections (3 × 12 mL) and extracted using dichloromethane (DCM). After a water wash and a 10% aqueous sodium bicarbonate solution wash, the mixed organic layers were dried on anhydrous magnesium sulfate. The crude product was recrystallized from methanol after the solvent was removed under low pressure, yielding 0.36 g (88%) of compound 10 as white crystals. The product's authenticity was confirmed when its melting point was measured at 49–51 °C, which confirms its structure by matching with the reported one 48–50 °C.³⁰

Preparation of ethyl 3-(bromomethyl)benzofuran-2-carboxylate (11)

For three hours, compound 10 (0.41 g, 2 mmol) and N-bromosuccinimide (NBS, 0.36 g, 2 mmol) reacted in refluxing carbon tetrachloride. The reaction was carried out in a fume hood with adequate ventilation. After everything was finished, the solvent was extracted using distillation, and the residue that was left behind was dissolved in heated methanol to create a saturated solution. In accordance with the reported value of 86–87 °C,³⁰ the bromo derivative 11 crystallized upon cooling, producing 0.50 g (89%) of product with a melting point of 87–89 °C.

Preparation of ethyl (E)-3-(((3-(4-hydroxy-3-methoxyphenyl)acryloyl)oxy)methyl)-benzofuran-2-carboxylate 12

In the presence of potassium carbonate (0.28 g, 2 mmol) as a base, trans-ferulic acid 1 (0.29 g, 1.5 mmol) was added to ester 11 (0.28 g, 1 mmol). The reaction mixture was agitated for nine hours at room temperature while DMF was utilized as a solvent. The residue was then triturated in 20 mL of cold water, agitated for 30 minutes, and then crystallized from 95% ethanol to create compound 12. The yield (86%) was a white powder with a Mp of 165–167 °C. DMF was then evaporated. ¹H NMR (500 MHz, DMSO-d₆) δ (ppm): 9.65 (s, 1H), 7.93 (dd, J = 8.0, 2.3 Hz, 1H), 7.77 (dd, J = 8.5, 2.4 Hz, 1H), 7.63–7.56 (m, 2H), 7.42 (td, J = 7.7, 2.4 Hz, 1H), 7.34 (q, J = 1.8 Hz, 1H), 7.13 (dt, J = 8.2, 1.8 Hz, 1H), 6.79 (dt, J = 8.2, 1.7 Hz, 1H), 6.56 (dt, J = 15.9, 1.7 Hz, 1H), 5.74–5.70 (m, 2H), 4.43–4.38 (m, 2H), 3.82–3.77 (m, 3H), 1.36 (tt, J = 7.2, 1.7 Hz, 3H). ¹³C NMR (125 MHz, DMSO-d₆) δ (ppm): 166.9, 159.4, 154.2, 150.0, 148.4, 146.3, 141.9, 128.8, 127.1, 126.9, 124.6, 124.0, 123.9, 122.7, 115.9, 114.1, 112.7, 111.7, 61.9, 56.6, 56.1, 14.5. ¹³C NMR-DEPT (125 MHz, DMSO-d₆) at 135°, the CH and CH₃ groups were detected as positive (upward) signals and the CH₂ groups as negative (downward) signals. The following chemical shifts were noted at δ: 146.3 (Arom. ↑), 128.8 (Arom. ↑), 124.6 (Arom. ↑), 123.9 (Arom. ↑), 122.7 (Arom. ↑), 115.9 (Arom. ↑), 114.1 (Arom. ↑), 112.7 (Arom. ↑), 111.7 (Arom. ↑), 61.9 (OCH₂CH₃ ↓), 56.6 (OCH₂ ↓), 56.1 (OCH₃ ↑), 14.5 (OCH₂CH₃ ↑). Anal. calcd. for C₂₂H₂₀O₇: C, 66.66; H, 5.09. Found: C, 66.87; H, 5.06%.

Biochemical reagents, chemicals and solvents

All of the reagents and compounds utilized in this investigation were analytical grade. The main source of the reagents was Sigma-Aldrich Chemical Co. (St. Louis, MO, USA). The solvent used for trans-ferulic acid derivatives was dimethyl sulfoxide (DMSO), which was modified based on the molecular weights of the compounds. As a reference cytotoxic agent (positive control), doxorubicin was employed. Biomed Laboratories in Egypt provided further analytical-grade reagents. Amygdalin (Amy) was dissolved in DMEM medium, and sorafenib (Sor) stock solutions were made fresh in DMSO. VACSERA (Dokki-Giza, Egypt) provided the human hepatocellular carcinoma (HepG2) and normal human lung fibroblast (WI-38) cell lines.

Biochemical/molecular assay kits

Abcam (Boston, MA, USA) provided the propidium iodide (PI) staining flow cytometry assay kit and the MTT assay kit.³¹

Cell culture and treatment

Human cell lines HepG2, Hep3B, Huh7, and normal WI-38 were cultivated in DMEM medium (Lonza, BioWhittaker®, USA, CAS No. 12-614) supplemented with 10% fetal bovine serum (FBS) (Sigma, USA, CAS No. 1943609-65-1), 100 μg mL⁻¹ penicillin, and 100 μg mL⁻¹ streptomycin (Lonza, BioWhittaker®, USA, Cat. No. DE17-602E). The cells were then incubated in a humidified chamber with 5% CO₂. The cells were subcultured at a 1 : 6 ratio when they had achieved 80–90% confluency. When cultures achieved 40–50% confluency, treatments were started using the corresponding drugs' predefined half-maximal inhibitory concentration (IC₅₀/2) for 48 hours. Every cell culture experiment was carried out in triplicate, and statistical analysis was done using the data from three separate experiments.²⁴

Cell viability assay

The MTT colorimetric test kit was used to assess the impact of trans-ferulic acid derivatives on the viability of HepG2, Hep3B, Huh7, and WI-38 cells (Sigma-Aldrich, USA, CAS-No. 298-93-1). To put it briefly, cells were seeded in 96-well plates at a density of 5 × 10³ cells per well, and they were left to adhere for the entire night at 37 °C. To ensure that the final DMSO concentration did not surpass 0.1% (v/v), a stock solution of each chemical was produced in DMSO at a concentration of 10 mM and diluted with growth medium to the necessary working concentrations. After then, the cells were exposed to different doses of trans-ferulic acid derivatives for 48 hours. Following treatment, each well received 100 μL of MTT working solution, and the plates were incubated for 4 hours at 37 °C in the dark. Following the cautious removal of the medium, 100 μL of DMSO was applied to each well in order to dissolve the formazan crystals. The microplate reader (BIO-RAD PR4100, USA) was used to measure the optical density (OD) at 570 nm. The percentage of cell viability was expressed in relation to untreated control cells, which were thought to be 100% viable. Using nonlinear regression analysis in GraphPad Prism, IC₅₀ values were computed from the dose–response curves. Every experiment was carried out in triplicate (n = 3), and the mean ± standard deviation (SD) was used to express the results.^24,32

Enzyme-linked immunosorbent assay (ELISA)

Quantikine human AFP (alpha-fetoprotein) ELISA (cat# E-EL-H0070, Elbasciense, Houston, Texas, USA), human caspase-3 (Catalog # KHO1091, Invitrogen Corporation, Camarillo, CA), human abnormal prothrombin (APT)/des-γ-carboxyprothrombin (DCP) ELISA kit (Cat. No. MBS2602615), Zymed® BCL-2 ELISA kit (Cat. No. 99-0042), human MCL1 (induced myeloid leukemia cell differentiation protein Mcl-1) ELISA kit (Wuhan Fine Biotech Co., Ltd., Wuhan, China, Catalog No. EH2099), human p53 ELISA (Product Number CS0070, Sigma-Aldrich, Missouri, USA), and Urokinase-type Plasminogen Activator Human ELISA kit (Cat. No. DEIA1630, Creative Diagnostics, NU, USA). Error bars show the standard error of the mean, and the data are displayed as the average of all biological replicates.

Statistical analysis

GraphPad Prism software was used to analyze the data. The mean ± SD is used to report the results, and Tukey's post hoc test was used to evaluate significant differences between groups after one-way analysis of variance (ANOVA). Pearson's correlation coefficient (r) was used to assess correlations between variables. At p < 0.05, statistical significance was taken into account.

VEGFR2 kinase inhibitory assay

The inhibiting impact of compound 4e on VEGFR2 kinase (Catalog #40325) was evaluated. Four serial concentrations were made in accordance with the manufacturer's instructions after it was dissolved in 0.1% DMSO.^33–38

VEGFR-2 (total and phosphorylated) and PI3K/Akt/mTOR pathway

Total VEGFR-2: the impact of compound 4e on total VEGFR-2 expression was evaluated in HepG2 cells at its IC₅₀ (1.8 μg mL⁻¹) using a cell-based ELISA. Absorbance at 450 nm was measured and concentrations were calculated using a standard curve. Phosphorylated VEGFR-2 (p-VEGFR-2): the effect on VEGFR-2 phosphorylation at Tyr951 was assessed using a colorimetric cell-based ELISA, with quantification via absorbance at 450 nm. PI3K/Akt/mTOR pathway: PI3K phosphorylation: compound 4e inhibited PI3K phosphorylation in HepG2 cells at IC₅₀ using a phospho-specific ELISA. Biotinylated detection antibodies and streptavidin-HRP were used, with absorbance measured at 450 nm. AKT phosphorylation: inhibition of AKT phosphorylation at Ser473 was measured using a DuoSet® IC ELISA, distinguishing phosphorylated from total AKT. mTOR phosphorylation: the effect on mTOR phosphorylation at Ser2448 was evaluated using a PathScan® sandwich ELISA, where phosphorylated mTOR was detected by specific antibodies and quantified via HRP/TMB colorimetric readout.

Computational studies

In silico ADMET profile of compound 4e

Virtual labs can be used to evaluate the physicochemical properties, drug-likeness, and pharmacokinetic profiles of chemical compounds. ADMETlab 3.0 is the most comprehensive platform of these tools, providing comprehensive predictions regarding drug-likeness, ADMET characteristics, and potential toxicity issues.^39,40

Molecular docking analysis

To assess compound 4e's possible binding affinity against VEGFR-2 tyrosine kinase (TK), molecular docking was used. The Protein Data Bank provided the VEGFR-2 TK structure (PDB ID: 4ASD).^41,42 Water molecules and other extraneous elements were first eliminated from the protein complex. After fixing crystallographic flaws and missing valence atoms, the protein structure was energy-minimized and stored as PDBQT files. Chem-Bio create Ultra 16.0 was used to create the compound 4e's 2D structure, which was then saved as an SDF file and transformed into a 3D structure. Energy minimization and protonation were carried out and stored in PDBQTP files. Using Autodock Vina 1.5.7 software, docking simulations were performed using a stiff docking technique in which the ligand was flexible⁴³ and the receptor stayed stationary. Twenty different stances could be produced by each chemical. Discovery Studio 2024 was used to create both 2D and 3D visualizations, and the docking scores (affinity energy) for the best-fitting postures with the target protein were noted.⁴⁴

Conflicts of interest

There are no conflicts to declare.

Data availability

All data generated or analyzed during this study are included in this manuscript and its supplementary information (SI) file.

Supplementary information is available. See DOI: https://doi.org/10.1039/d5md00980d.