TED is a chronic inflammatory autoimmune disease with a complex etiology. Recent studies have demonstrated that both m6A modification and autophagy are involved in the process of TED [7, 10]. Increased evidence has shown that m6A plays a critical role in regulating autophagy in various diseases via directly modifying the expression of ATGs or affecting the autophagy-related signaling mechanisms [11–14, 23–25]. However, there is limited knowledge about the interactive effects of m6A modification and autophagy in TED. The aim of our research was to investigate t the effects of m6A modification and autophagy interactivity in TED.
In this study, we comprehensively analyzed the expression profiles of m6A and autophagy regulators between TED patients and normal samples. We first identified dysregulation of five m6A regulators and 44 ARGs in TED patients compared to healthy controls. Each of the five m6A regulators positively correlated with most dysregulated ARGs. After overlapping with the validated target genes of five m6A regulators from the RM2target online database and combined with correlation analysis, two dysregulated ARGs, BNIP3 and CTSB, were selected to verify in our samples. Finally, FTO and BNIP3 were considered biomarkers for TED.
FTO, fat mass and obesity-associated protein, is a well-known RNA demethylase involved in regulating m6A levels. FTO has been shown to play a role in various physiological and pathological processes, including autophagy. Several studies have investigated and confirmed the relationship between FTO and autophagy. In adipogenesis, suppression of FTO could directly downregulate the expression of ATG7 and ATG5 in an m6A-YTHDF2-dependent manner, inhibiting autophagosome formation and subsequently preventing autophagy and adipogenesis [14]. Another study found that the downregulation of FTO appears to trigger autophagy activation via stabilizing BECN1 mRNA in liver fibrosis [26]. A previous study demonstrated that several m6A regulators were upregulated in surgically excised EOMs from TED patients compared to the control subjects. Among them, the expression level of FTO was lower in the TED group; however, no statistical difference was obtained [7]. In our study, FTO was decreased in the anterior orbit tissues of patients with TED. The consistent result was also observed in our validated surgically excised orbit samples.
BNIP3, also known as BCL2/adenovirus E1B 19kDa interacting protein 3, is involved in cellular processes such as apoptosis and oxidative stress. Recent studies have suggested that BNIP3 is involved in the development of various ocular diseases, including age-related macular degeneration (AMD) [27, 28], retinal detachment [29], and glaucoma [30]. Negatively correlation between FTO and BNIP3 has been confirmed in breast cancer, as high FTO expression decreases the expression of BNIP3 via an m6A-YTHDF2-independent mechanism, thus promoting tumor growth retardation and metastasis [31]. Another study showed that BNIP3 is involved in the regulation of m6A levels in the context of neuronal apoptosis [32]. The study found that BNIP3 interacts with the m6A writer METTL3 and the m6A reader YTHDF2, which are involved in the addition and recognition of m6A modifications, respectively. This interaction was found to play a role in the regulation of neuronal apoptosis, suggesting that BNIP3 may play a role in the regulation of m6A-mediated RNA metabolism in the nervous system. In the present study, BNIP3 was invested downregulated in TED patients from GSE58331 and our surgically excised orbit samples, indicating autophagy might play a suppressive role in the process of TED. In addition, BNIP3 is positively correlated with FTO in our study, however, which was opposite to the results in breast cancer [31]. These results indicate that the influence of FTO on autophagy is complex and dynamic, and their role in TED pathogenesis needs to be further determined and verified on mechanism experiments and more clinical samples.
Next, we elucidate the m6A- and autophagy-related expression patterns in TED to gain further insight into their interactions. We discovered that both dysregulated m6A regulators and ARGs have well and similar discrimination ability to distinguish the 27 TED samples into two clusters. This finding highlights the existence of m6A-autophagy interaction in the process of TED. Further, two distinct groups were identified based on the expression levels of 2,751 overlapped DEGs from m6A and autophagy patterns. In addition, 507 genes were identified as cluster-specific DEGs in TED patients compared to healthy controls among the 2,751 genes. Combining the SVM-RFE and LASSO model results, five hub genes (F13A1, FGR, CFH, FCGR3B, and MXRA5) were selected and constructed the Nomogram Model. Calibration curves revealed that the predictivity of the nomogram model was accurate. The DCA result demonstrated that the nomogram model had an obvious net benefit in predicting the occurrence of TED.
Adipose tissue remodeling is an important aspect of the pathogenesis of TED. In TED, the adipose tissue in orbit undergoes a process of expansion and fibrosis, which can lead to compression of the optic nerve and other orbital structures. This expansion of adipose tissue is thought to be driven by the autoimmune response that characterizes the disease, which leads to the accumulation of immune cells and cytokines in the adipose tissue [1, 2, 33, 34]. FTO has been extensively studied in relation to obesity and metabolic disorders, which has been shown to play a role in the regulation of adipocyte differentiation and function via an m6A-dependent manner [14, 35]. BNIP3 has been shown to play a role in the regulation of adipocyte remodeling, particularly through its involvement in autophagy [36, 37]. Therefore, it is reasonable to the downregulation of FTO and BNIP3 in hypertrophic orbital adipose tissue of TED patients in our study, which is consistent with their previous inhibition effects in adipocyte differentiation and adipocyte remodeling.
F13A1, also known as coagulation factor XIII A chain, is a cellular process involved in the degradation and recycling of damaged cellular components. Studies have shown that F13A1 was significantly altered, with higher expression in adipose tissue, linking to increased weight, pro-inflammatory, cell stress, and tissue remodeling pathways [38, 39]. FGR is a gene that encodes the FGR protein, which is a member of the Src family of non-receptor protein tyrosine kinases. A recent study has shown that FGR expression is upregulated in macrophages of obese adipose tissue, and bone marrow FGR deletion can inhibit adipose inflammation and metabolic dysfunction caused by M1 polarization of adipocyte macrophages in diet-induced obesity [40]. The gene MXRA5 encodes the matrix remodeling-associated protein 5, which has been linked to several biological processes, including extracellular matrix remodeling, cell adhesion, and immune responses. MXRA5 has been found to be upregulated in the adipose tissue of obese individuals and to be associated with inflammation and insulin resistance [41]. For CFH and FCGR3B, the exact roles in adipose tissue biology or TED are not well understood. However, studies have confirmed that they may play critical roles in the regulation of inflammation and metabolic disorders [42, 43]. Additionally, these genes were linked to various ocular diseases, including AMD [42, 44], uveitis [45], and Sjogren's syndrome [46].
Although our prognostic model performed well in predicting the TED and rigorous evidence supported the interaction between m6A and autophagy in this disease, our study had some limitations. These results were based on and only verified in small clinic samples; we need verification in a large number of TED patients in the future. Additionally, the experiment in vitro or in vivo needed to perform to clarify the details of the interplay mechanism between FTO and BNIP3 in the process of TED.