Oral cancer (OC) is an aggressive malignant neoplasm originating from the oral cavity and its associated anatomical structures, including the gingiva, oropharynx, tongue, lips, and palate [
1–
3]. Global epidemiological data from 2022 indicates that oral cancer accounted for approximately 390,000 newly diagnosed cases and 200,000 mortality-related outcomes worldwide [
4]. Oral cancer exhibits notable geographic disparities, with escalating incidence rates observed in Southeast Asia and specific Chinese regions including Hainan and Taiwan provinces [
5]. Contemporary chemotherapeutic regimens, though clinically indispensable, show suboptimal response rates (~50%), highlighting critical gaps in targeted molecular interventions [
6].
Ferroptosis, a newly characterized form of regulated cell death driven by iron-dependent accumulation of lipid peroxides, has emerged as a promising therapeutic strategy [
7], particularly for refractory oral cancers [
8]. Moreover, induction of ferroptosis has also shown potential in reversing resistance to chemotherapy [
9] and radiotherapy, thereby maximizing clinical therapeutic efficacy [
10].
The cystine/glutamate antitransporter (xCT)-glutathione/glutathione peroxidase 4 (GPX4) axis serves as the central regulatory machinery governing cellular defense against ferroptosis, with coordinated cystine uptake and lipid peroxide reduction constituting its core mechanistic framework. The xCT inhibitor erastin significantly induced ferroptosis in HSC-2 and SAS oral squamous cell carcinoma (OSCC) cell lines, particularly in those exhibiting an epithelial-mesenchymal transition (EMT)-driven mesenchymal phenotype, which appears more susceptible to ferroptotic cell death. Similarly, RSL3, a GPX4 inhibitor, enhances the efficacy of radiation therapy in OSCC cells in mouse xenograft models, providing a mechanistic rationale for ferroptosis-targeted therapies in aggressive OSCC subtypes [
8,
11].
In our previous work, a more selective ferroptosis inducer compound
26a targeting GPX4 was discovered, and
26a demonstrated excellent anti-proliferative activity on a variety of tumor cell lines [
12,
13]. However,
26a exhibits poor water solubility and exhibits certain toxic effects. Its chloroacetyl moiety, acting as an electrophilic warhead, is the primary cause of its unfavorable pharmacokinetic and pharmacodynamic profiles. ML210 is regarded as a classic GPX4 inhibitor, and combining its piperazine moiety with novel electrophilic fragments offers significant potential for addressing the issues [
14].
In this study, structure-based molecular hybridization was used to strategically integrate the privileged benzothiophene acetamide scaffold of
26a with the piperazine pharmacophore from ML210. Building upon this molecular framework, we systematically explored structure-activity relationships (SAR) through strategic incorporation of diverse electrophilic warheads (
Fig. 1). Optimized compound XW19 demonstrates potent ferroptosis-inducing activity with remarkable selectivity (selectivity index (SI) = 1200) and enhanced aqueous solubility. This novel agent exhibits strong growth suppression against multiple oral cancer cell lines, and thermal shift assays validate its direct engagement with the GPX4 protein target. Notably,
in vivo evaluations reveal significant tumor growth inhibition in oral cancer xenograft models coupled with an excellent safety profile, positioning XW19 as a promising therapeutic candidate for oral cancer treatment.
The synthetic routes for targeted compounds were strategically developed through multicomponent reaction-driven divergent synthesis (
Scheme 1). The synthesis of XW1–XW5 begins with the Passerini reaction, generating the key benzothiophene acetamide intermediate
1. Following sequential hydrolysis, chlorination, nucleophilic substitution, and acylation transformations, the intermediate was elaborated into the final A-series compounds. In parallel, XW6–XW22 was synthesized through a four-component Ugi protocol, yielding chloroacetamide intermediates
5a–
5e. These intermediates underwent Boc deprotection, followed by nucleophilic cyclization to establish the core scaffold, and were subsequently subjected to late-stage acylation to afford the target compounds XW6–XW22.
Our initial investigation focused on two molecular hybrid compounds (XW1 and XW2) containing chloracetyl and 5-methyl-4-nitroisoxazole-3-carbonyl as distinct electrophilic warheads, respectively. As shown in
Table 1, cytotoxicity profiling revealed that compound
26a exhibited moderate antiproliferative activity against CAL27, SCC-25, and WSU-HN30 oral cancer cells (half maximal inhibitory concentration (IC
50) = 2.85–13.68 µmol/L), but showed minimal effect on SAS cells (IC
50 > 20 µmol/L). In contrast, ML210 demonstrated potent cytotoxicity in fibrosarcoma HT1080 cells with high ferroptosis selectivity (confirmed by inhibitor rescue assays), yet was inactive against all four oral cancer cell lines (IC
50 > 20 µmol/L). XW1 exhibited neither significant cytotoxicity nor ferroptosis selectivity, while XW2 demonstrated promising oral cancer inhibitory activity. This marked contrast in biological outcomes highlights the critical influence of electrophilic warhead selection on pharmacological profiles within this structural class. Guided by these findings, we subsequently designed compounds XW3–XW5 through the replacement of the electrophilic warheads. Furthermore, to investigate the impact of heterocyclic modifications on bioactivity, we implemented an isosteric replacement strategy by substituting the CH
2 group in the piperazine ring with a carbonyl group (CO), leading to the synthesis of analogs XW6–XW10. Lactam ring-containing analogs exhibited enhanced anti-oral cancer activity. For instance, compound XW8 demonstrated superior potency with IC
50 values of 1.61 µmol/L (Cal27), 3.09 µmol/L (SCC25), 0.91 µmol/L (WSU-HN30), and 4.78 µmol/L (SAS), representing an improvement compared to the parent compound XW3 across all tested cell lines, underscoring the pharmacophoric advantage of the lactam. This SAR-driven discovery prompted systematic exploration of electrophilic warheads (XW11–XW19) [
8,
15–
20] to evaluate their ability to induce ferroptosis and inhibit oral cancer cell proliferation. The results revealed that electrophilic fragments such as 5-formylthiophene-3-carbonyl, 5-formylfuran-3-carbonyl, and 5-nitrofuran-2-carbonyl conferred certain ferroptosis-inducing activity and anti-oral cancer effects. In contrast, fragments such as 5-nitrothiophene-2-carbonyl (compounds XW14–XW18) resulted in a complete loss of activity. These findings suggest that GPX4 inhibitors exhibit a degree of conservativity regarding the electrophilic moiety. Among the evaluated warheads, 2-ethynylthiazole-4-carboxamide (XW19) emerged as the most favorable, conferring marked ferroptosis selectivity (SI = 1201) and potent antiproliferative activity against oral cancer cell lines (IC
50 = 0.71–2.39 µmol/L). Finally, the effects of distinct amides on biological activity were investigated. Compounds bearing benzyl (XW19), tert‑butyl (XW20), 4-fluorophenethyl (XW21), or cyclohexyl (XW22) substituted amides all retained good potency, and the benzyl-containing compound XW19 exhibited enhanced inhibitory activity against oral cancer cells with ferroptosis selectivity. Consequently, XW19 was selected as the lead compound for further investigation.
To validate the protein targets of the compounds, extracellular GPX4 inhibitory activity was assessed (Fig. S1 in Supporting information). The results indicated that both XW19 and
26a exhibited significant inhibitory activity against GPX4. In contrast, ML210 did not show any extracellular GPX4 inhibition, as it is a prodrug that requires intracellular biotransformation to bind GPX4 [
21]. Following compound treatment, a decrease in intracellular GPXs activity was also observed. These results collectively demonstrate that XW19 possesses remarkable GPX4 inhibitory activity. To elucidate the molecular recognition mode of XW19 with GPX4, molecular docking studies were performed. As depicted in
Fig. 2A, the alkyne group on the thiazole ring forms a critical covalent bond with CYS-46, consistent with previous reports. Additionally, a network of
π-π interactions involving the thiazole, benzyl ring, and benzothiophene moieties with residue TRP-136 stabilizes the compound conformation. Notably, hydrogen bonds form between the oxygen atoms of the two amide bonds and residue LYS-48. Collectively, these interactions underpin the strong binding affinity of compound XW19 for GPX4, accounting for its potent GPX4 inhibitory activity and consequent antitumor effects. To further validate the target engagement, the cellular thermal shift assay (CETSA) [
22] was performed in Cal27 and WSU-HN30 cell lines. As shown in
Figs. 2B–E, incubation with XW19 increased the thermal stability of GPX4, resulting in thermal shifts (Δ
Tm) of 3.40 and 2.51 ℃, respectively. These findings demonstrate that XW19 effectively binds to GPX4 in a cellular context. Furthermore, treatment with XW19 resulted in both dose- (
Figs. 2F–I) and time- (
Figs. 2J–M) dependent depletion of cellular GPX4 protein levels. However, XW19 treatment did not affect GPX4 mRNA expression (Fig. S2A in Supporting information), suggesting that compound binding may facilitate GPX4 protein degradation. Consistently, MG132, a proteasome inhibitor, blocked XW19-induced GPX4 degradation, whereas leupeptin had no effect (Figs. S2B–E in Supporting information). Collectively, these results indicate that XW19 binds to GPX4 and promotes its proteasome-mediated degradation.
The ferroptosis selectivity of XW19 was evaluated in oral cancer CAL27 cells. As shown in
Figs. 3A and B, the cytotoxicity of XW19 was significantly reversed by the ferroptosis inhibitor ferrostatin-1 (Fer-1), whereas the apoptosis inhibitor Z-VAD-FMK, the necroptosis inhibitor necrostatin-1 (Nec-1), and the autophagy inhibitor 3-MA did not attenuate XW19-induced cytotoxicity (
Fig. 3C, Figs. S3A and B in supporting information). Moreover, XW19 exhibited lower cytotoxicity toward normal cells compared with
26a, with IC
50 values of 11.55 ± 0.48 µmol/L in mouse fibroblasts L929 and 13.65 ± 1.05 µmol/L in human oral keratinocytes HOK. In contrast,
26a displayed IC
50 values of 7.00 ± 0.48 and 8.21 ± 0.36 µmol/L in L929 and HOK cells, respectively (Figs. S3C and D in Supporting information). These results demonstrate that XW19 selectively induces ferroptosis in oral cancer cells. Consistently, both flow cytometry and confocal microscopy analyses (
Figs. 3D and E) [
23] demonstrated that XW19 significantly increased the levels of oxidative stress, which were reversed by Fer-1. Importantly, the elevation of oxidative stress and lipid peroxidation exhibited a concentration-dependent pattern in Cal27 and WSU-HN30 cell lines following XW19 treatment (
Figs. 3F and G,
Figs. 4A and B). Additionally, XW19 increased the labile iron pool (LIP) in both cell lines (
Figs. 4C and D), thereby amplifies lipid peroxidation and establishing a self-reinforcing cycle that drives ferroptotic cell death. Morphological evidence of ferroptosis was further confirmed by the presence of characteristic mitochondrial alterations, including shrinkage and loss of cristae, observed in XW19-treated cells (
Fig. 4E).
Given the potent ferroptosis-inducing and anti-oral cancer activity of XW19, we further evaluated its aqueous solubility and metabolic stability. The results showed that XW19 exhibits improved solubility (3.6 µg/mL) and a plasma half-life of 119.6 min. Its
in vivo efficacy was assessed using xenograft models. All animal experiments were approved by the Ethics Committee of Hainan University (approval No HPIACUC2024150). As shown in
Fig. 5, both compound
26a (30 mg/kg) and XW19 moderately inhibited tumor growth in Cal27-derived xenograft model, each achieving a tumor growth inhibition (TGI) rate of 59%, with significantly reduced tumor weights compared to control group (
P < 0.001). However, treatment with
26a resulted in >10% body weight loss, indicative of systemic toxicity. A comprehensive safety assessment (
Fig. 6) further revealed hepatocellular injury in the
26a-treated group, as evidenced by increased organ coefficients (liver: +16%; spleen: +146%) and altered serum biochemical parameters (alanine aminotransferase (ALT): +91.8%; alkaline phosphatase (ALP): −22%). In contrast, XW19 did not induce significant hepatocellular injury and may exert a regulatory effect of hepatic metabolism without reaching a toxic threshold, indicating an overall favorable safety profile.
In summary, we developed a series of hybrid molecules by integrating the privileged scaffold of 26a and ML210. Following systematic structural optimization, the lead compound XW19 was identified. XW19 exhibits remarkable ferroptosis selectivity (SI = 1200) and potent inhibitory activity against oral cancer cell lines with IC50 values ranging from 0.71 µmol/L to 2.39 µmol/L in Cal27, SCC25, WSU-HN30, and SAS cells. Molecular docking and CETSA in both Cal27 and WSU-HN30 cells confirmed its strong binding affinity for GPX4. Furthermore, XW19 significantly induced lipid peroxide accumulation in oral cancer cells, an effect that was reversed by ferroptosis inhibitors. Treatment with XW19 also increased intracellular ferrous iron (Fe2+) levels and induced characteristic ferroptotic mitochondrial morphology. Crucially, in a Cal27 ectopic xenograft model, XW19 demonstrated significant tumor growth inhibition (TGI = 59%) and exhibited an improved safety profile compared to 26a. This study may provide ideas for the application of ferroptosis inducers in the treatment of oral cancer.
PDF
Full
102
55
关闭全屏
No. 86 Xueyuan South Road, Haidian District, Beijing
010-62662699(Operation and Technical Support)
qkjq@cast.org.cn
