Although THCA is a relatively common and less lethal cancer in the head and neck, certain types of THCA exhibit early systemic metastasis, rapid infiltration of surrounding tissues, and high malignancy [47]. The pathogenesis and clinical manifestations of THCA are more associated with genetic alterations [48]. Therefore, the search for reliable and clinically applicable novel biomarkers is crucial for the diagnosis and treatment of THCA. Previous studies have shown that ferroptosis is involved in the development of cancer, and the discovery of ferroptosis, a novel type of cell death, offers new possibilities for cancer development and treatment [49]. Our research used ferroptosis-related genes to uncover biomarkers that can diagnose or treat THCA and used them as targets to find compounds that can modulate their expression as potential therapeutic agents.
In this study, we first searched for DEGs in the GSE27155 and GSE3678 datasets from the GEO database, compared them with the validated ferroptosis-related genes from the FerrDb database, and identified three ferroptosis-related genes, MUC1, GDF15 and ALOX5. Then, we validated them in the GEPIA and TCGA databases, and the the expression of MUC1, ALOX5, and GDF15 were higher in THCA tissues than in normal tissues and correlated with the pathological stage of THCA (p < 0.05). In addition, we investigated the expression of ALOX5, GDF15 and MUC1 in pancarcinoma and found that ALOX5 and GDF15 were equally highly expressed in GBM and OV, while MUC1 was not differentially expressed in THCA. Therefore, MUC1 was excluded from further studies. Combined with the large differences between GBM and OV and THCA in terms of site, cytologic features, clinical manifestations, and laboratory tests, we can still conclude that ALOX5 and GDF15 can be specific biomarkers for THCA [50, 51].
ALOX5, an iron-containing dioxygenase, is highly expressed in tumors along with its metabolite 5-hydroxyeicosatetraenoic acid (5-HETE), which promotes tumour cell growth [52–54]. Previous studies have shown that ALOX5 can regulate cell death through lipid peroxidation and inflammation, and its overexpression increases the sensitivity of cells to ferroptosis [18], while also increasing invasiveness by relying on 5-HETE to induce matrix metalloproteinase-9 [55]. Other studies have now confirmed that ALOX5 can be used as a biomarker or potential therapeutic target in various cancers, including lung cancer [56], colon adenocarcinoma [57], pancreatic cancer [58], chronic myeloid leukaemia [59] and renal clear cell carcinoma [60]. GDF15, a member of the transforming growth factor- β (TGF- β) superfamily, has received much attention for its potential role in cancer. It can both inhibit and promote cancer development [61, 62]. Previous studies have shown that GDF15, which is overexpressed in tissue injury, inflammation and other pathological conditions and affects cellular energy metabolism [63], plays an anticancer role in the early stage of malignant tumors, and promotes tumorigenesis in the late stages of cancer [64–66]. Prostate cancer, cervical cancer, and colorectal cancer have all been associated with GDF15 overexpression [67, 68]. GDF15 may serve as a diagnostic biomarker and prognostic index for the occurrence and development of cancers [69], such as gastric cancer [70] and liver cancer [71]. In addition, we found that GDF15 plays a key role in ferroptosis [72], and alters tumour progression in papillary THCA through activation of STAT3 [73]. In the present study, we confirmed by IHC and qRT‒PCR that ALOX5 and GDF15 were overexpressed in THCA tissues and that knockdown of ALOX5 and GDF15 protein expression inhibited THCA cell viability.
Based on the AUC values in the ROC curves, ALOX5 and GDF15 have high diagnostic value in THCA, which indicates that ALOX5 and GDF15 have high sensitivity and specificity in istinguishing patients with thyroid tumors from healthy individuals. Therefore, we believe that ALOX5 and GDF15 can be used as diagnostic indicators for thyroid cancer.
It has been reported that ALOX5 can promote tumour cell growth through the PI3K/AKT pathway and MAPK pathway [54, 74], while GDF15 can act through the PI3/Akt pathway and TGFβ/SMAD pathway to promote tumour growth [75, 76]. Therefore, in the functional study of ALOX5 and GDF15, we used the GO and KEGG databases and found by enrichment analysis that the main functions of the ALOX5 and GDF15 genes were related to lipid metabolism, the cell signalling pathway, arachidonic acid metabolism, the MAPK signalling pathway and the IL-17 signalling pathway by enrichment analysis. The MAPK signalling pathway is involved in regulating cell growth, differentiation, the inflammatory response and many other cellular physiological and pathological processes and is impacted by cancer [77]. Interleukin-17 (IL-17), a proinflammatory cytokine, is involved in the secretory process of immune cells to induce inflammatory responses and immune disease pathology [78]. These data suggest that ALOX5 and GDF15, which are differentially expressed in THCA, are potential drug therapeutic targets.
Immune cells in the TME have tumour-promoting or antagonistic functions, which are increasingly being studied [79]. Mounting evidence shows that the single-cell analysis of immune cells around and within the tumour of THCA plays a crucial role in the diagnosis and treatment of THCA [80]. We found an association with immune pathways in our functional analysis of ALOX5 and GDF15, so we briefly explored the relationship between ALOX5 and GDF15 and immune infiltration using the TIMER database. The results showed that the expression of ALOX5 and GDF15 was significantly correlated with the infiltration of six types of immune cells, including B cells, CD8+ T cells, CD4+ T cells, macrophages, neutrophils and dendritic cells. Considering the relationship between immune cell infiltration and tumour progression, was propose that ALOX5 and GDF15 could be used not only as diagnostic markers but also to reflect the immune status of cells.
We used the CTD database to find compounds that could target ALOX5 and GDF15, and since overexpression of these two genes in THCA tissues affects tumour development, we identified four compounds that could reduce ALOX5 and GDF15, namely, BDE47, MWCNT, indomethacin and (+)-JQ1.
BDE47 is a polybrominated diphenyl ether (PBDE) homologue. High levels of exposure can promote cell damage and increase the expression of inflammatory proteins [81]. Previous studies have shown that BDE47 can downregulate several protective pathways involved in cell homeostasis, cell growth and survival, including the eIF2, PAK, PDGF and EGF signalling pathways, which affect cell growth, proliferation, migration and apoptosis. At present, research on BDE47 is mainly based on its endocrine-disrupting activity and neurobehavioural toxicity [82, 83]. Because of its structural similarity to triiodothyronine (T3) and thyroid hormone (T4), the toxicity of BDE47 is related to its ability to disrupt the dynamic homeostasis and function of thyroid hormone (TH) in animals[45]. BDE-47 damages normal thyroid tissue by triggering endoplasmic reticulum stress, apoptosis, and autophagy and has been shown to damage the hypothalamic-pituitary-thyroid axis [84, 85]. Therefore, we did not consider it as a potential compound for THCA treatment. MWCNTs are nanomaterials that are toxic to the lung, reproduction, and development. Animal studies have shown that exposure to carbon nanotubes can lead to persistent inflammation of the lungs, fibrosis, lung cancer and even genetic damage to the lungs [44]. Due to their low biocompatibility and toxicity, carbon nanotubes have thus far been used in medicine primarily for their unique physicochemical properties, which make them a valuable drug delivery system [86, 87]. Therefore, neither MWCNTs nor BDE47 can be used directly as a targeted compound. Indomethacin is a class of nonsteroidal compounds that effectively inhibit the cyclooxygenase 1 and 2 pathways and is widely used in cancer and viral infection research [88, 89]. A large number of studies have reported that indomethacin inhibits the growth of THCA cells and affects the proliferation of thyroid follicular cells in rats, so we did not investigate it as a therapy here [90–94].
JQ1 is a small molecular bromination domain inhibitor, that can destroy the bromination domain-histone acetylation interaction, resulting in a decrease in BRD4 binding. BRD4 is an epigenetic regulator and transcriptional cofactor. Its abnormal expression leads to an imbalance in the expression of many genes, affecting the function of related genes [95, 96]. Previous studies have found that JQ1 can effectively promote cancer cell death, induce miR-4516 upregulation [97], and decrease the expression of Mcm5 mRNA in anaplastic thyroid carcinoma (ATC) cell lines and thus reduce cell proliferation [35]. We conducted in vitro studies to assess the effect of JQ1 on ferroptosis. JQ1 induced ferroptosis by inhibiting BRD4. Other studies have shown that JQ1 downregulates the expression of histone lysine methyltransferase G9a and upregulates the expression of histone deacetylase SIRT1 [98], thus initiating ferritin phagocytosis or changes in the ferroptosis genes Gpx4, SLC7A11 and SLC3A2 and inducing ferroptosis [99]. Additionally, JQ1 can activate ATF2, and ATF2 via the JNK1/2 pathway and induce ferroptosis in breast cancer cells. JQ1 combined with ATF2 inhibitors might be a novel cancer treatment option [100]. It has been reported that combining an apoptosis inducer that targets the B-cell lymphoma-2 (Bcl-2) protein family with JQ1 can significantly enhance the inhibitory effect on GBM cells with high levels of ALOX5 and GDF15, and this finding suggests that combination therapy with apoptosis and ferroptosis inducers may improve treatment outcomes for GBM [101]. At the gene level we verified that JQ1 can reduce the expression of ALOX5 and GDF15, and at the cellular level we also verified that JQ1 can inhibit the proliferation of THCA cells. Therefore, JQ1 can serve as a potential compound for the treatment of THCA.
In summary, ALOX5 and GDF15 have high diagnostic value and a significant correlation with immune infiltrating cells, suggesting that ALOX5 and GDF15 could be relevant biomarkers for cancer diagnosis and treatment and potential targets for tumour immunotherapy. JQ1 can target THCA cells by regulating ALOX5 and GDF15, providing THCA with new potential therapeutic compounds. Our study was limited by the lack of in vivo animal experiments to validate specific mechanisms, as all data analysed were from online databases and only a few basic experiments were performed.