Small molecules that hit colchicine binding site could exert their efficient biological effects by inhibiting tubulin assembly and suppressing microtubule formation (Lu et al., 2012), numerous modifications of the colchicine chemical structure have been thus made to develop new anti-cancer candidate drug molecules (Gracheva et al., 2020). However, there were few reports about the hybridization of colchicine with other bioactive natural molecules. In this current study, we designed a new type of C-7-modified colchicinoid which was a hybrid of colchicine and magnolol with simple amino acid as the linker.

As shown in Scheme 1, the synthesis of CMH was commenced with the S_N2 reaction of magnolol with ethyl bromoacetate, and compound 3 was isolated in 78% yield. The ester was hydrolyzed with NaOH to generate the acid 4 which was used directly without further purification. In addition, a three-step sequence was employed to remove the -Ac group at C(7) of colchicine (Lagnoux et al., 2005; Yasobu et al., 2011). Furthermore, the amino group was then acylated with compound 4 to give the novel hybrid CMH.

We then evaluated the antitumor activity of colchicine, CMH and cisplatin (positive control) in Lewis lung carcinoma (LLC) cells. As shown in Figure 2, the cell viability of LLC cells was greatly reduced after 24 h treatment with three compounds. Specifically, the half-maximal inhibitory concentration (IC₅₀) of colchicine, CMH and cisplatin against LLC cells was 0.06, 0.26 and 26.05 μM, respectively. It was evident that our novel hybrid CMH inhibited proliferation of LLC cells 100 times more potently than cisplatin did. In consistent with the cell viability results, images from confocal microscopy showed that compared to the control group, a lower density and rounder shape were observed in LLC cells incubated with different concentrations of CMH (Figure 3). Moreover, CMH remarkably decreased the number of FDA-stained viable cells (Figure 3). These results indicated that the proliferation of LLC cells was inhibited to a larger extent by CMH treatment.

Next, we further used normal lung epithelial BEAS-2B cells to evaluate the toxicity of CMH and its parent molecule colchicine. As shown in Figure 4, CMH did not show any toxicity until its concentration reached 1 μM. Specifically, 24 h exposure of BEAS-2B cells to CMH at 1 μM decreased cell viability from (100.00 ± 1.48) % to (72.59 ± 3.17) %. In comparison, colchicine began to induce toxicity at the concentration of 0.1 μM, colchicine at 0.1 μM decreased cell viability to (73.27 ± 3.89) %. These results indicated the cytotoxicity of CMH might be 10-fold lower than that of colchicine in normal human lung cells.

Glycogen synthase kinase-3β (GSK3β) is a serine-threonine kinase that is responsible for promoting cancer cell survival, growth and proliferation (Domoto et al., 2020). It has been well documented that aberrant GSK3β activity is firmly associated with multiple tumor-related diseases and that GSK3β is generally accepted as a potential anti-tumor target (Bai et al., 2022; Fan et al., 2022; You et al., 2022). In light of this, we extended our effort to test the possibility that CMH may provide antiproliferative capacity through the direct inhibition of GSK3β enzyme activity. As shown in Figure 5, CMH at 100 μM decreased enzyme activity to approximately 40% of control, while failed to inhibit GSK3β at concentrations below 100 μM (data not shown). Since CMH showed toxicity in normal lung epithelial BEAS-2B cells when its concentration exceeded 1 μM, we speculate that GSK3β may not be the critical mechanism underlying CMH-mediated antitumor efficacy and some other potential targets are expected to be explored.

There is extensive evidence that the phosphorylation of mitogen-activated protein kinases (MAPKs), extracellular signal-regulated kinase (ERK) subtype in particular, is closely associated with the growth and proliferation of tumor cells in cellular and animal experimental paradigms (Yang et al., 2022; Yu et al., 2022). Specifically, the increase of phospho-ERK level was usually observed in cancer cell lines such as LLC cells, particularly those treated with pro-tumorogenic compounds (Stoyanov et al., 2012), bio-molecules that could down-regulate ERK phosphorylation may thereby offer effective anti-tumor efficacy (Shi et al., 2022; Yuan et al., 2022).

In our cell system, the protein level of phospho-ERK (p-ERK) was evaluated using Western blot and the results in Figure 6 showed that CMH at 0.25 μM and 1 μM declined this phosphorylated protein to (0.61 ± 0.11) and (0.57 ± 0.13), respectively, compared to the control group (1.00 ± 0.20). This phenomenon was consistent with earlier findings where this phosphorylated protein level in LLC cells greatly declined in the presence of anti-tumorogenic chemicals (Kim et al., 2007; Xie et al., 2018) and indicated that the inhibition of p-ERK may be a critical mechanism that underlied the anti-tumor effects of CMH.

Microtubules, which maintain the shape of the cell through the dynamic assembly of tubulin heterodimers, are generally accepted as an attractive target for the development of anti-cancer drugs (Andres et al., 2022). Specifically, colchicine binding site agents bind to colchicine binding domain and prevent the polymerization of tubulin proteins, thereby destabilize microtubules and provide antitumor potential (Wang et al., 2016; Sueth-Santiago et al., 2017). We then tested the possibility that our novel hybrid CMH could occupy colchicine binding site. Results from docking studies of compounds with tubulin protein (PDB entry: 1SA0) showed that both colchicine and CMH were well fitted in the colchicine binding site of tubulin. Specifically, colchicine molecule formed a hydrogen bond to Asn258 of tubulin protein with an estimated binding free energy of -5.93 kcal/mol, an observation consistent with an earlier study (colchicine docking score: −5.5 kcal/mol) (Dwivedi et al., 2022). In contrast, the best-docked conformation of CMH in the tubulin showed that the methoxy and the phenolic hydroxyl group of this ligand from the colchicine fragment and B fragment interacted with Asn258 and Thr353 through several hydrogen bonds with an estimated binding free energy of -8.04 kcal/mol (Figure 7). These results suggested that our novel CMH could be served as an effective colchicine binding site inhibitor.

In this docking system, CMH showed better docking score than the standard inhibitor colchicine, on the other hand, CMH (IC₅₀ = 0.26 μM) displayed anti-proliferative potential 4 times less potently than colchicine (IC₅₀ = 0.06 μM, Figure 2). This discrepancy could be explained by the existence of some other possible targets, such as taxane, vinca, laulimalide binding domains of tubulin, that CMH or colchicine may hit. Such interesting topics will be further investigated in our future projects.

Finally, the antitumor ability of CMH was verified in C57BL/6 mice xenograft model. As shown in Figure 8, CMH exhibited promising antitumor efficacy, with a tumor growth inhibition (TGI) of 79.37% and 85.44% at the dosages of 0.5 mg/kg and 0.75 mg/kg, respectively. No mortality was observed in the group treated with CMH. For a reference, cisplatin (positive control) at 1.0 mg/kg inhibited tumor growth by 79.22% and colchicine at 0.5 mg/kg inhibited tumor growth by 78.81%. However, colchicine treatment was accompanied with a high risk of mortality. It was obvious that 7 out of 10 mice (colchicine group, 0.75 mg/kg) and one out of 9 mice (colchicine group, 0.5 mg/kg) were dead during 10 days of drug treatment (Table 1). These findings taken together suggested that CMH inhibited tumor growth 2 times more potently than cisplatin, and that CMH displayed antitumor capacity with a lower mortality and an efficacy comparable or even superior to colchicine.