Citation

Song Jian, Wang Shenghui, Liu Qiuge, Wang Xiao, Yuan Shuo, Liu Hongmin, Zhang Saiyang. N-Benzyl arylamide derivatives as novel and potent tubulin polymerization inhibitors against gastric cancers: Design, structure–activity relationships and biological evaluations[J]. Chin. Chem. Lett., 2025, 36(2): 109678  

Jian Song^a,1, Shenghui Wang^a,1, Qiuge Liu^a, Xiao Wang^a, Shuo Yuan^d, Hongmin Liu^b,*, Saiyang Zhang^a,c,*     

Microtubules, as the crucial component of the tumor cell cytoskeleton, play pivotal roles in essential biological functions such as mitosis, cell signaling, and intracellular transport [1,2]. Consequently, microtubules have emerged as significant and highly efficacious targets for cancer treatment [3,4]. By interfering with microtubule dynamics, microtubule-targeting agents (MTAs) could effectively induce cycle arrest and apoptosis in tumor cells, thus exhibiting potent anti-tumor effects. Some of MTAs, such as paclitaxel and vinblastine along with their analogues, have gained U.S. Food and Drug Administration (FDA) approval for frontline cancer treatment [5]. Nevertheless, these drugs still encounter certain limitations in clinical application due to concerns regarding drug toxicity and resistance [4,6].

Currently, seven binding sites on microtubules have been identified, including the paclitaxel site, vinca site, laulimalide site, pironetin site, maytansine site, colchicine site and the seventh site [7-9]. Among them, the MTAs targeting the colchicine binding site (colchicine binding site inhibitors or CBSIs) have garnered significant attention due to their favorable attributes compared to other MTAs, such as simplistic structures, enhanced water solubility, wide therapeutic index, and diminished multi-drug resistance [10,11]. To date, some CBSIs, such as, OXI4503 [12], AVE8062 [13], BNC-105P [14] and ABT-751 [15], have been or are being investigated in clinical trials for cancer therapy (Fig. 1A).

Figure 1

	Download: JPG larger image
Fig. 1. (A) Representative CBSIs that have been approved for clinical trials. (B) CA-4 and representative CA-4 analogues as CBSIs. (C) Design of N-benzyl arylamide derivatives as CBSIs in this work.

The natural product combretastatin A-4 (CA-4, 1), derived from the South African bushwillow tree Combretum caffrum, demonstrates potent inhibitory activity against a variety of tumor cells [16]. CA-4 exerts its effects by targeting the colchicine binding site, leading to disrupt cell mitosis and induce cell apoptosis [17]. Despite the significant in vitro anti-tumor activity of CA-4, its lower bioavailability and less effective activity in vivo, further limited its clinical applications [18]. As a result, numerous research groups have conducted extensive structural modifications on CA-4 [17,19-21], with some derivatives such as OXI4503 and AVE8062, successfully entering clinical trials. In the researches on structural modifications of CA-4, the cis-ethylene bridge in the structure of CA-4 was usually modified, such as replacing it with a longer or shorter connecting chain or directly modifying the double bond [22-26]. For example, compounds isoCA-4 (2) and phenstain (3) were also exhibited excellent antitumor properties [27]. In addition, by replacing the cis-ethylene bridge with a rigid cyclic system consisting of three to six-membered rings, flipping could be prevented, thereby increasing stability and maintaining biological activity [28-30]. This approach also enables the synthesis of compounds exhibiting excellent antitumor activity, such as cyclopropyl analog 4 [29], β-lactam 5 [31] and 1,2,5-selenadiazol 6 (Fig. 1B) [30].

Many groups have incorporated the cis-double bond of CA-4 into a rigid β-lactam ring to design and synthesize β-lactam-based CBSIs [32-36], which exhibit excellent inhibitory activity against tubulin polymerization with potent anti-tumor activities. However, β-lactam compounds, especially β-lactam antibiotics, with chiral centers, may undergo hydrolysis or molecular rearrangement, leading to loss of efficacy and further limiting clinical applications [37,38]. In addition, as a commonly found structural fragment in FDA-approved small molecule compounds, the amide scaffold is frequently in the design of anticancer agents [39-41]. Hence, we attempted to obtain the amide scaffold unit by opening up the β-lactam ring to circumvent potential disadvantages associated with β-lactams (Fig. 1C). By employing the ring-opening approach, we here try to transform the β-lactam scaffold-based inhibitors derived from CA-4 into an N-benzyl arylamide scaffold, which was then used to derive novel N-benzyl arylamide derivatives. Meanwhile, in the design of N-benzyl arylamide derivatives, the 3, 4, 5-trimethoxyphenyl moiety was retained, which is considered as an essential active moiety to act on colchicine binding site [20]. We found that novel N-benzyl arylamide derivative 13n (also known as MY-1388), whose mechanism is the inhibition of tubulin polymerization by targeting the colchicine binding site, demonstrated excellent in vivo and in vitro activities.

The synthetic procedures of compounds 13a–ac, 14a–k and 16a–g were illustrated in Scheme S1 (Supporting information). Under standard aldimine condensation reaction conditions, substituted aromatic aldehydes 7a–i were reacted with 3, 4, 5-trimethoxyaniline (8) to afford schiff bases 9a–i, which were reduced by sodium borohydride to generate the intermediates 10a–i. Substitution reactions of compound 10a with commercially-available acyl chlorides (11a–z and 12a–k) gave the final products 13a–m, 13o–y, 13aa–ab and 14a–k. The amino compounds 13m, 13y and 13ac were prepared by the reduction reactions of nitros 13n, 13z and 13ab. Under the substitution reaction conditions, compounds 10b–i reacted with 4-nitrobenzoyl chloride (12m) to get intermediates 15a–g. Subsequently, intermediates 15a–g were reduced by Fe to obtain the final products 16a–g.

Given these compounds were designed as anticancer agents, the in vitro antiproliferative activities of 13a–ac, 14a–k and 16a–g were first explored via using the MTT assay MGC-803 (human gastric cancer cell line), KYSE450 (human esophageal cancer cell line), and HCT-116 (human colorectal cancer cell line) after a 48-h incubation period with two well-known CBSIs, colchicine and CA-4, as positive control drugs.

As shown in Table S1 (Supporting information), N-benzyl arylamides 13a to 13z exhibited excellent antiproliferative activities against these three cancer cells in vitro, with half maximal inhibitory concentration (IC₅₀) values all below 0.5 µmol/L. Among them, 13a, 13b, 13c, 13g, 13h, 13i, 13n (also known as MY-1388) and 13w showed the best activity within the range of IC₅₀ values from 0.09 µmol/L to 0.047 µmol/L. Particularly, 13n (MY-1388) with an amino-substituted group on right-side phenyl ring demonstrated the strongest activity, displaying excellent IC₅₀ values of 0.009 µmol/L for MGC-803 cells, 0.018 µmol/L for HCT-116 cells and 0.014 µmol/L for KYSE450 cells respectively, surpassing colchicine. Importantly, it exhibited comparable CA-4-like activity (IC₅₀ = 0.008 µmol/L) in inhibiting MGC-803 cells.

Further elucidating the structure-activity relationship, it was observed that substituents on right-side phenyl ring exert an influence on the in vitro antiproliferative activities. 13a demonstrated notable antiproliferative activity (MGC-803: IC₅₀ = 0.027 µmol/L) when devoid of substituents on the right-side phenyl ring. When electron-donating groups such as methyl, methoxy, and amino were introduced as para-substituents on the right-side phenyl ring, 13b, 13c and 13n (MY-1388) exhibited enhanced antiproliferative potency against MGC-803 cells. Conversely, when these substituents were positioned at the meta-position of right-side phenyl ring (13e and 13f), their activity decreased compared to 13a, 13b and 13c. In comparison to 13a, compounds containing electron-donating groups like tert-butyl and dimethylamino as para-substituents on the right-side phenyl ring (13d and 13l) displayed reduced activities against MGC-803 cells. Compared to 13a, the activity of the compound weakens as the electron-withdrawing effect strengthens when the para-substituents the right-side phenyl ring are F, Cl, Br, CF₃, CN or NO₂. Among them, 13g, 13h and 13i demonstrated superior activities compared to compound 13a. Similarly, the electron-withdrawing groups on the right-side phenyl ring as meta-substituents showed weaker activity than those as para-substituents (13g vs. 13p; 13h vs. 13q; 13i vs. 13r). To further investigate the structure-activity relationship, 13s–ac were designed and synthesized to explore the impact of alkyl chain length (n) on activities (Table S1), compared to 13a–r, the activity of the decreases 13s–ac with the same substituent as the length (n) of the alkyl chain increases (13a vs. 13s vs. 13aa, 13b vs. 13t, 13c vs. 13u, 13g vs. 13v, 13h vs. 13w, 13i vs. 13x, 13m vs. 13y vs. 13ab, 13n vs. 13z vs. 13ac).

Subsequently, the right-side phenyl ring was substituted with heterocyclic groups such as thiophene, furan and pyridine (14a–k, Table S2 in Supporting information). Concurrently, the impact of alkyl chain length (n) on compound activity was also investigated. Most of the 14a–k exhibited potent antiproliferative activities against these three cancer cells, with IC₅₀ values ranging from 0.011 µmol/L to 0.37 µmol/L. In addition, 14a, 14d and 14e featuring unsubstituted thiophene, furan and pyridine moieties demonstrated exceptional activity, surpassing 14b, 14c, 14f and 14g containing substituted thiophene and pyridine moieties. Among them, 14a and 14d exhibited slightly lower activity than 13n (MY-1388), with IC₅₀ values ranging from 0.011 µmol/L to 0.020 µmol/L. Similarly, an increase in alkyl chain length (n) results in a decrease in activities, as evidenced by the comparison of 14a vs. 14h vs. 14j, 14d vs. 14k, 14e vs. 14i.

Next, we continued to change the substituents on left-side phenyl ring to synthesize 16a–g (Table S3 in Supporting information). However, the results revealed that modifications of the left-side phenyl ring led to a significant reduction in antiproliferative activities. Compared to 16g, the incorporation of a hydroxy group (MY-1388) at the meta-position of the left-side phenyl ring could significantly enhance the antiproliferative activities. The introduction of electron-donating group (methoxy) or electron-withdrawing groups (F, Cl, Br) on the meta-position significantly diminished the antiproliferative activities, compared to 16g. Compared with 13n (MY-1388), 16f, which retains only the meta-hydroxy substituent, exhibited a significantly reduced antiproliferative activity. On the other hand, 16g, which only retains a methoxy substituent at the para-position, maintained its antiproliferative activity at nanomolar levels. These results indicated that the presence of a para-methoxy group on the left-side phenyl ring is crucial for maintaining antiproliferative activity, and the meta-hydroxy substitution significantly enhanced its potency.

Considering that MY-1388 exhibited excellent proliferative inhibition activity against MGC-803, HCT-116 and KYSE450 cells, its antiproliferative potency on other 15 cancer cells, gastric normal cells GES-1, and normal vascular endothelial cells HUVEC were also further evaluated using MTT assay (Table S4 in Supporting information). MY-1388 also exhibited a broad spectrum of antiproliferative activity against other 12 cancer cells with IC₅₀ values ranging from 0.008 µmol/L to 0.048 µmol/L with high selective inhibitory activities over normal cells GES-1 (IC₅₀ = 0.76 µmol/L) and HUVEC (IC₅₀ = 48.47 µmol/L). Next, we also conducted further investigations to evaluate the potential of MY-1388 overcome drug resistance. To accomplish this, three resistant cancer cell lines were selected: paclitaxel-resistant A549/Taxol cells, as well as multidrug-resistant MCF-7/ADR and SGC-7901/ADR cells. The results presented in Table S5 (Supporting information) demonstrate that MY-1388 exhibited more potent antiproliferative activities compared to those of paclitaxel (IC₅₀ = 0.045 vs. 0.23 µmol/L for A549/Taxol; 0.0227 vs. 5.23 µmol/L for MCF-7/ADR; 0.0188 vs. 6.27 µmol/L for SGC-7901/ADR).

These compounds were designed as a specific tubulin polymerization inhibitor by targeting the colchicine binding site of β-tubulin. Therefore, the effects of MY-1388 on tubulin polymerization were further evaluated. As shown in Figs. 2A and B, MY-1388 could concentration-dependently inhibit tubulin polymerization under cell-free condition with an IC₅₀ value of 0.62 µmol/L, which was better than that of colchicine (IC₅₀ = 6.7 µmol/L) [42] and CA-4 (IC₅₀ = 1.98 µmol/L) [43], and could directly interacted with the colchicine binding site of β-tubulin (Fig. 2C). To predict the binding modes of MY-1388 with tubulin, molecular docking studies were further explored (PDB: 5LYJ) by MOE.2019.01 software. The docking results indicated that MY-1388 could well bind into the colchicine binding site of β-tubulin (Fig. 2E), and it shared similar binding modes with CA-4 (Fig. 2D). The 4-methoxy group could form a hydrogen bond with Cys241 at a distance of 2.4 Å. At the same time, the 3-hydroxy-4-methoxyphenyl group could engage in extensive hydrophobic interactions with surrounding amino acid residues, including Leu242, Val238, Ala354 and Ile318. Compared CA-4, the 4-aminophenyl group formed an additional hydrogen bond with Asn101’s backbone at a distance of 2.3 Å. The binding between MY-1388 and colchicine binding site of β-tubulin is maintained through hydrogen bonding and hydrophobic interactions, resulting in its potent tubulin polymerization inhibitory activity. In addition, in MY-1388 treated MGC-803 and SGC-7901 cells, multiple nuclei obtained from cell replication could not be drawn to complete mitosis, and the phenomenon of multiple nuclei appeared (Figs. 2F and G), which was similar to colchicine, indicating that MY-1388 exerted a potent impact on the tubulin assembly and disrupting microtubule networks.

Figure 2

Download: JPG larger image

Fig. 2. Compound MY-1388 bound to colchicine binding site to inhibit tubulin polymerization. (A) Cell-free tubulin polymerization detection, the fluorescence intensity indicates the level of tubulin polymerization. (B) The activity percentage of tubulin treated with compound MY-1388. (C) N,N′-Ethylene-bis(iodoacetamide) (EBI) competition assay. Immunoblotting analysis for β-tubulin adduct protein and glyceraldehyde-3-phosphate dehydrogenase (GAPDH) protein from MGC-801 and SGC-7901 cells. (D, E) Molecular docking models between compound MY-1388 with tubulin (PDB: 5LYJ). The binding mode comparison between compound MY-1388 with CA-4 (D). The proposed binding mode of compound MY-1388 in the colchicine binding site of β-tubulin (E). The hydrogen bonds were shown in yellow dotted lines. (F, G) Effects of compound MY-1388 on microtubule networks. Microtubules were visualized with an anti-β-tubulin antibody (green), and the cell nucleus was visualized with 4′,6-diamidino-2-phenylindole (DAPI, blue). Scale bar: 25 µm.

Inhibition of tubulin polymerization could result in mitotic arrest of cancer cells, thus inhibiting cell colony formatting ability, and inducing cell cycle arrest and cell apoptosis. Therefore, the effects of MY-1388 on cell colony formatting ability, cell cycle arrest and apoptosis were further explored. After treatment with MY-1388, cell colony formatting ability was concentration-dependently inhibited (Figs. 3A–D), and the morphological changes in MGC-803 and SGC-7901 cells were observed (Figs. S1A–C in Supporting information). In addition, MY-1388 also displayed potent effects on inducing G2/M phase cell cycle arrest (Figs. 3E–H and Figs. S2A–D in Supporting information) and promoting apoptosis (Figs. 3I–L, and Figs. S2E–H in Supporting information) by modulating the expression levels of proteins involved in cell cycle and apoptosis.

Figure 3

Download: JPG larger image

Fig. 3. Effects of compound MY-1388 on cell colony formatting ability, cell cycle and apoptosis. Cells were treated for 48 h. (A–D) Images of formatted cell colonies. (E, F) Cell cycle distribution of treated cells detected by flow cytometry. (G, H) Levels of cell cycle related proteins. (I, J) Cell apoptosis ratio of treated cells detected by flow cytometry. (K, L) Level of cell apoptosis related proteins. Quantitative data were represented as the mean ± standard deviation (SD) (n = 3). P  <  0.05, **P  <  0.01, ***P  <  0.001 vs. the vehicle control.

In order to facilitate the development of candidate drugs, it is imperative to achieve a harmonious equilibrium between biological activity and physicochemical properties. MY-1388 showed acceptable druglike properties according to the Lipinski’s rule of five (Table S6 in Supporting information). In addition, MY-1388 also exhibited good liver microsomal stability in human (T_1/2 = 37.5 min) and rat liver (T_1/2 = 133 min) microsomes (Table S7 in Supporting information). After 60 min incubations, while 32.0% and 73.9% of MY-1388 still remained in human and rat liver microsomes, respectively. However, according to previous research, only 26.4% of CA-4 remained in rat liver microsomes [44]. Given MY-1388 exhibited improved stability compared to CA-4 in the human liver microsome. A pharmacokinetic study in vivo was further performed for MY-1388 in rats (n = 3) after intravenous injection (10 mg/kg). The results of Table S8 (Supporting information) indicated MY-1388 was absorbed quickly (T_max = 0.0833 h) and exhibited a moderate half-life (T_1/2 = 0.938 h).

In order to investigate the in vivo inhibitory effects of MY-1388 on gastric cancer cells, a xenograft model bearing MGC-803 cells was established. CA-4 was used as a positive control and the MY-1388 was administered daily by intraperitoneal injection. Animal welfare and experimental procedures have been reviewed and approved by the Animal Ethics Committee of Zhengzhou University, Zhengzhou, China. As shown in Fig. 4, compared to the negative control group, the CA-4 group (30 mg kg⁻¹ day⁻¹), MY-1388 group (15 mg kg⁻¹ day⁻¹), and MY-1388 group (30 mg kg⁻¹ day⁻¹) exhibited significantly reduced tumor volume and weight (Figs. 4A–C), and also exhibited a good safety profile (Fig. S3 in Supporting information). The tumor growth inhibition rates for the CA-4 group (30 mg kg⁻¹ day⁻¹), MY-1388 group (10 mg kg⁻¹ day⁻¹), and MY-1388 group (30 mg kg⁻¹ day⁻¹) were 53.2%, 69.7%, and 76.4%, respectively, indicating that MY-1388 demonstrated remarkable in vivo anti-tumor effects surpassing those of CA-4. In addition, the protein changes in tumor tissues were the same as in vitro, and MY-1388 led to up-regulation of cleaved-poly ADP-ribose polymerase (cleaved-PARP), p-Histone H3, and down-regulation of Ki67, cyclin B1 (Figs. 4D–G).

Figure 4

Download: JPG larger image

Fig. 4. Antitumor activity of compound MY-1388 in vivo. (A) Tumor volume during the treatment. (B) Tumor weight of each group. (C) Images of tumor tissues. (D, F) Apoptosis and proliferation-related proteins levels detected by Western blot assay. Scale bar: 10 mm. (E, G) Protein levels detected by immunohistochemistry. Scale bar: 500 µm. Quantitative data were represented as the mean ± SD (n = 3). P  <  0.05, **P  <  0.01, ***P  <  0.001 vs. the vehicle control.

In summary, we designed and synthesized a series of novel N-benzylaryl amide derivatives as CSBIs with potential anticancer by ring-opening strategy. Among them, 13n (MY-1388) exhibited excellent antiproliferative potency on fifteen human cancer cell lines with IC₅₀ values ranging from 8 nmol/L to 48 nmol/L. Further studies confirmed that 13n (MY-1388) could target the colchicine-binding site of β-tubulin, potently inhibited tubulin polymerization (IC₅₀ = 0.62 µmol/L), significantly inhibited cell colony formation, and effectively induced G2/M phase cell cycle and cell apoptosis. Additionally, 13n (MY-1388) shows enhanced and adequate liver microsomal stability in human (T_1/2 = 37.5 min) and rat liver (T_1/2 = 133 min) microsomes. Importantly, 13n (MY-1388) demonstrated potent antitumor effects (TGI rate was 76.4% at a dosage of 30 mg/kg) with a favorable safety profile in an MGC-803 xenograft model, surpassing the antitumor efficacy of CA-4 (30 mg kg⁻¹ day⁻¹). The aforementioned results demonstrate that 13n (MY-1388) exhibits significant potential as a tubulin polymerization inhibitor, warranting further exploration for its development in cancer therapy.

The authors declare that they have no known competing financial interests or personal relationships that could have appeared to influence the work reported in this paper.

This work was supported by the National Natural Science Foundation of China (Nos. 82273782 and U2004123) and Training Program for Young Key Teachers of Colleges and Universities in Henan Province (No. 2023GGJS008).

Supplementary material associated with this article can be found, in the online version, at doi:10.1016/j.cclet.2024.109678.