Abstract
Background/Aim: Demethylase fat mass and obesity-related protein (FTO), which belongs to the AlkB homologous (ABH) family, is associated with various neurological diseases, cancer, and obesity. This protein, which contains many structurally and functionally different regions, contains a COOH-terminal domain whose function, unlike other ABH members, is not fully understood. This study aimed to investigate the effects of the exonic V493F mutation in this region of FTO on the soluble proteome.
Materials and Methods: SH-SY5Y cells stably over-expressing wild-type (WT-FTO) or mutant FTO (V493F-FTO) proteins under the control of the Tet promoter were created and used. Comparative proteomic analysis using two-dimensional gel electrophoresis (2DE) identified over 500 protein spots, with 10 showing significant (≥2-fold) differential expression. These proteins were identified by MALDI-TOF/TOF mass spectrometry and validated by western blotting.
Results: WT-FTO over-expression primarily affected proteins related to DNA replication and repair, including PCNA, whereas V493F-FTO over-expression altered the expression of stress response and endoplasmic reticulum-associated degradation (ERAD) pathway proteins, such as HSPA4, ARHGDIA, and VCP. Although the mutation did not alter the nuclear localization or predicted 3D structure of FTO, it distinctly modulated pathways associated with protein homeostasis and cellular stress.
Conclusion: FTO participates in the regulation of the cellular stress response and the ubiquitin-dependent ERAD pathway, functions potentially independent of its demethylase activity. Importantly, dysregulation of these pathways has been implicated in cancer initiation, progression, and therapeutic resistance. Therefore, our findings provide new insights into how FTO mutations might influence oncogenic processes, highlighting FTO as a potential biomarker and therapeutic target in cancer biology.
Introduction
The AlkB homolog (ABH) protein family has nine homologs (ALKBH1-8 and FTO) in humans (1). These proteins are involved in several biological processes, including RNA metabolism and demethylation, DNA repair, and fatty acid metabolism. They exhibit Fe2+ and α-ketoglutarate-dependent demethylase activity on various substrates, including nucleic acids (DNA and RNA), and proteins (2). FTO protein is known to demethylate mRNAs and non-coding RNAs (such as tRNA and snRNA), N6-methyladenosine (m6A), and N6,2′-O-dimethyladenosine (m6Am) (3). These modifications affect stabilization and translation of mRNA and non-coding RNAs (4, 5). The FTO protein has two distinct domains: an N-terminal domain (32-326 residues, NTD) and a C-terminal domain (327-498 residues, CTD). The FTO protein is distinguished from other ALKBH members by its strong nucleotide selectivity in demethylase activity. NTD functions as the catalytic domain for the demethylation function. The CTD adopts a α-helix fold that is structurally well-conserved among species. However, sequence analysis revealed no significant homology with any known CTD-containing structures, suggesting that this domain represents a novel fold (6). CTD may play a role in maintaining the conformation of NTD and hence in regulating the catalytic activity of the FTO (6, 7). Also, the non-homologous region at the C-terminus together with the catalytic domain form an “L” shaped enzyme structure. This structure of FTO is thought to be related to the binding of tRNA or snRNA as well as mRNA (8, 9). The presence of a CTD that is not found in other ALKBH members may also help FTO to interact with potential protein partners in cells to demethylate different RNA substrates (9).
In a study by Meyre et al., the coding domain of FTO was sequenced from extremely obese and lean individuals, and the enzymatic activity of selected nonsynonymous variants was measured. Screening tests revealed the V493F variant in four obese individuals, and this variant had no significant effect on the enzyme’s in vitro enzymatic activity (10). Variation in this region, which is highly conserved across species, may affect other possible unknown functions of FTO, independent of demethylase activity. Therefore, in this study, we evaluated the effect of the exonic V493F mutation of FTO on the soluble proteome. Our results showed that over-expression of mutant V493F-FTO protein caused changes in the expression levels of several proteins involved in the ubiquitin-dependent ERAD pathway. The effect was more pronounced on stress-sensitive proteins.
Materials and Methods
Generation of plasmid constructs and stable cell lines expressing WT or mutant V493F FTO proteins. Full-length cDNA of human WT FTO was cloned into pcDNA4/TO (Life Tech, Carlsbad, CA, USA) by reverse transcription as detailed in our previous publication (11). The pcDNA4/TO clone for the mutant V493F-FTO was generated using Site-Directed Mutagenesis (Agilent Tech, Santa Clara, CA, USA) using the primers GCCGTTTGACCTCACAGACATCTT TTCAGAACTCAGAGGTCAGCTTC and GAAGCTGACC TCTGAGTTCTGAAAAGATGTCTGTGAGGTCAAACGGC for sense and antisense FTO sequences, respectively. The procedure for establishing stable SH-SY5Y cell lines expressing wild-type (WT) or mutant (V493F) FTO proteins under tetracycline-responsive (Tet-R+) regulation was described in detail in our previous studies (11, 12). SH-SY5Y cells were grown in EMEM supplemented with 10% (Vol/Vol) tetracycline-reduced fetal bovine serum, 100 U/mL penicillin-streptomycin, and 2 mM l-glutamine at 37°C in a humidified 5% CO2 atmosphere. Three selected clones expressing either WT-FTO or mutant FTO proteins were evaluated. These clones were labeled as clones #3, #5, and #12 for the WT and clones #3, #4, and #7 for the mutant. For the localization study, the mutant FTO gene was amplified with forward and reverse primers containing XhoI and XbaI restriction enzyme cutting sites, respectively, and cloned into the GFP-harboring pcDNA4/TO plasmid. This construct was also sequenced, and the in-frame FTO sequence was verified (Eurofins, Ebersberg, Germany).
Fluorescence microscopy. SH-SY5Y-WT-FTO- and SH-SY5Y-V493F-FTO-expressing cells were cultured under standard culture conditions (11). Culture plates contained glass coverslips that allowed fluorescence imaging. After 24 h of FTO expression, the cells were fixed with formaldehyde and stained for FTO with an anti-FTO antibody (Santa Cruz, Dallas, TX, USA). The nuclei of the cells were stained with DAPI (Lifetech). Cover slips were mounted in Mowiol (4-88, Sigma-Aldrich, St. Louis, MO, USA) before the analysis (13). Imaging was performed using an inverted microscope (Olympus CKX41, Shinjuku, Tokyo, Japan) using appropriate filter sets.
Protein extraction and assembly of the pooled sample. Protein extraction from cells, protein concentration measurements using the Bradford assay, and preparation of protein pools for each group were performed as detailed in our previous study (13). Protein concentrations and profiles were further verified using SDS-PAGE.
Two-dimensional gel electrophoresis (2DE). Cell-free extracts prepared from cells that expressed or do not express WT-FTO or V493F-FTO proteins were used for comparative analysis of protein profiles by 2DE. Three sets of gels were created for each sample. Briefly, 750 μg of protein was passively rehydrated for 16 h at 22°C onto immobilized 17 cm nonlinear pH 3-10 strips (BioRad, Hercules, CA, USA), focused, and equilibrated as previously described (14). The strips were then run on 12% SDS-PAGE gels to separate the proteins in the second dimension. After separation, the gels were fixed and stained with colloidal Coomassie blue G250 (Bio-Rad).
Image analysis and spot cutting. For the comparative analysis of the protein spots, PD Quest Advance 2D analysis software (BioRad) was used. The outer edges of the images were trimmed using the cropping tool. Protein spots were matched using an automated spot matching tool and then manually edited. The quantity of each spot was normalized using the linear regression model. Total spot counts and volumes inside the normalized region were calculated by automated analysis. The differentially regulated protein spots were cut from the respective gels using an ExQuest automated spot cutter (Bio-Rad).
In-gel tryptic digestion and protein identification. These experiments were performed as stated in our previous work (13). The MALDI-TOF/TOF data were analyzed against the MASCOT database using Protein Pilot, a streamlined software (ABSCIEX, Marlborough, MA, USA). Only significant hits (p<0.005) from the MASCOT probability analysis were allowed.
Western blotting. Western blot (WB) analysis was performed as described by Kanli et al. (13) GAPDH was used as an internal control of each protein sample to ensure equal protein loading. FTO mouse monoclonal antibody (Clone C-3, sc-271713, Santa Cruz), VCP rabbit monoclonal antibody (#2649, Cell Signaling, Boston, MA, USA) and GAPDH mouse monoclonal antibody (sc-365062, Santa Cruz) were used as primary antibodies. HRP labeled secondary antibody (Bio-Rad) was used as secondary antibody.
Bioinformatics analysis. Protein-protein interaction networks were created by STRING using the UniProt accession number of identified proteins and selecting the organism as Homo sapiens (15). Hits with 1e-04 false discovery rates (FDR) were considered significant. The results obtained as bitmap images were reconstructed using Adobe Illustrator Version 6.
Homology modelling. The three-dimensional structure of the WT- and V493F-FTO proteins was also predicted using the homology modeling program SWISS-MODEL web server.
Results
SH-SY5Y cell lines expressing WT-FTO and mutant V493F-FTO under Tet induction used in this study were produced using recombinant plasmids (pcDNA4/TO-FTO-WT and pcDNA4/TO-FTO-V493F) as we previously reported (11, 12). Experiments were performed using three stable cell lines selected from each group. In our earlier investigation, we demonstrated using immunofluorescence microscopy that the WT-FTO protein was localized to the nucleus in SH-SY5Y cells (13). Similarly, immunofluorescence was used to determine the location of mutant V493F-FTO proteins in SH-SY5Y cells. Prior to imaging, the mutant protein was expressed in SH-SY5Y cells grown on coverslips for 16 h. The Mutant FTO protein was also located in the nucleus (Figure 1).
Immunofluorescence microscopy analysis of cells expressing the mutant V493F FTO-GFP fusion protein showing localization of the protein to the nucleus. Anti-FTO monoclonal primary antibody (green) and Texas Red conjugated anti-mouse secondary antibody were used at a ratio of 1:100. Cell nuclei were stained with DAPI (blue) at a ratio of 1:20,000.
2DE experiments were performed with cell-free extracts prepared from WT and mutant FTO-expressing cells (Figure 2). A total of 500 (±20) protein spots per gel were detected and compared according to two-fold regulation criteria. Representative images of the triplicate 2D gels and the differentially-regulated protein spots are shown in Figure 2A for SH-SY5Y cells expressing WT FTO and Figure 2B for SH-SY5Y cells expressing V493F. There were changes in four and six protein spots in WT-FTO and mutant-FTO expressing cells, respectively. The protein spots, cut from the preparative gel using an automatic cutting tool, were identified using MALDI-TOF/TOF analysis following in-gel tryptic digestion (Table I). The identified proteins were peroxiredoxin-2 (PRDX2), proliferation cell nuclear antigen (PCNA), 14-3-3 protein zeta/delta (YWHAZ, 14-3-3) and phosphoglycerate mutase 1 (PGAM1) for WT-FTO-expressing cells and heat shock 70 kDa protein 4 (HSPA4), DDB1- and CUL4-associated factor 7 (DCAF7), ran-specific GTPase-activating protein (RANBP1), actin-cytoplasmic 1 (ACTB), Rho GDP-dissociation inhibitor 1 (ARHGDIA), and transitional endoplasmic reticulum ATPase (TERA, Valosin-containing protein VCP) for mutant FTO-expressing cells. The expresion ratios of these proteins are presented in Table I. These ratios were also verified by taking close-up images of the corresponding spot, as shown in Figure 3A and B. WB analysis was used to confirm differential protein expression. WB experiments showed that FTO protein levels were significantly higher in induced samples of cells expressing WT-FTO and V493F-FTO under tetracycline control, and expression levels were similar between both WT-FTO and V493F-FTO (Figure 4A). VCP protein was selected from the differentially expressed proteins in the V493F-FTO group in 2D experiments for further examination. VCP protein levels were decreased in induced cells compared to uninduced cells (Figure 4B). Therefore, the results of the WB analyses confirmed the findings of the mass spectrometry experiments, demonstrating that the observed changes represent actual changes at the proteome level.
Images of 2-DE gels prepared with protein extracts from WT-FTO (A) and V493F-FTO (B) samples. The larger 2D gel below shows the locations of the protein spots that were cut and identified using MALDI-TOF/TOF analysis. The term “Uninduced” refers to cells that do not over-express the WT-/mutant-FTO protein, while “Tet-induced” refers to cells that over-express the WT-/mutant-FTO protein by adding tetracycline to the cell culture. An unstained protein marker was used in the experiments (#26610, Thermo Scientific Waltham, MA, USA).
The identified proteins’ names and MALDI scores, and the regulation ratios and trends of the identified proteins.
Close-up images of proteins identified using MALDI-TOF/TOF in WT-FTO (A) or V493F-FTO (B) expressing and non-expressing SH-SY5Y cells.
Representative western blot images and relative expression levels of (A) FTO-WT and FTO-V493F proteins, (B) VCP protein (*p<0.05; ns: not significant).
Discussion
The FTO protein, a member of the ALKBH family, causes posttranscriptional modifications on mRNA, tRNA, and snRNA species in cells through its demethylation activity (4, 5). Multiple physiological functions are attributed to the FTO, as it has effects on the translation and stability of target RNAs (13, 16-19). While NTD provides the catalytic activity of the protein, it has been suggested that CTD may regulate the catalytic activity of the protein by maintaining the conformation of the NTD (6, 7). Meyre et al. (2010) showed that the V493F mutation located in the C-terminal domain and only detected in obese people does not affect the in vitro enzymatic activity of the FTO. They noted that although this variant appears to have comparable enzymatic activity with WT-FTO, caution should be taken when its physiological function is considered (10). This change may impact the possible unknown functions of FTO independent of its demethylase activity. Recent studies have clarified that FTO’s functional specificity depends not only on its catalytic domain but also on its structural conformation and partner interactions. Particularly, the C-terminal domain (CTD) plays an important role in determining FTO’s interaction landscape and substrate targeting (20, 21). The discovery that FTO can be recruited to specific RNAs by zinc finger proteins, such as ZBTB48, demonstrates that its activity is modulated through protein-protein interactions rather than sequence alone (21). Thus, mutations in the CTD, such as V493F, although not altering enzymatic demethylase activity, may disrupt partner binding or subcellular targeting, ultimately leading to altered proteomic regulation. The lack of sequence similarity of the CTD across known structures might limit predictable partner binding, underlining the novelty of CTD fold and its potential role in mediating such mutations’ effects. Therefore, studies on the C-terminal domain will contribute to elucidating the function of this region.
To determine whether the V493F mutation affects the intracellular localization of the FTO protein, we examined the localization of the protein using immunofluorescence microscopy and observed that mutant FTO is localized to the nucleus. We and others have demonstrated that WT-FTO is also localized to the nucleus, suggesting that FTO is not likely to be targeted to the plasma membrane, extracellular space, or any other organelles (1, 22, 23). In our previous studies, we have also shown that another mutant (R316Q) of FTO is also localized to the nucleus in different cell lines (13, 24). However, Gulati et al. using a live-cell imaging technique, discovered that FTO shuttles between the nucleus and cytoplasm, most likely by interacting with the exportin 2 protein, and can be found in two cellular compartments (25). Our findings indicated that nuclear localization of FTO was unaffected by the V493F mutation. Whether this mutation affects the shuttling of the FTO protein to the cytoplasm can be elucidated using live-cell imaging techniques.
There were no major differences in the proteomes of cells expressing WT-FTO and mutant FTO. However, we observed changes in the levels of some proteins that guided us to elucidate changes in associated pathways. In cells over-expressing WT-FTO protein, PRDX2 was up-regulated, whereas PCNA, YWAHAZ (14-3-3), and PGAM1 were down-regulated. STRING analysis using stringent search conditions yielded a metabolic network with four nodes (Figure 5A). One of the nodes associated with PCNA directed our attention to cell cycle associated events like transcription-coupled nucleotide-excision repair (FDR: 9.79e-05), nucleotide-excision repair (FDR: 5.26e-07), nuclear DNA replication (FDR: 3.14e-07), and mismatch repair (FDR: 1.21e-07). However, the down-regulated protein PGAM1 was correlated with a respiratory burst (FDR: 1.58e-02), whereas the other down-regulated proteins, PRDX2 and YWHAZ, were related to the response to stress (FDR: 5e-03). Interestingly, PRDX2 was also associated with respiratory burst, regulation of transferase activity (FDR: 1.78e-02), and cellular homeostasis (FDR: 1.17e-02). Changes in YWHAZ protein levels (14-3-3) were associated with the cell cycle (FDR: 1.7e-03) and PI3K-Akt signaling pathways (FDR: 1.67e-02). Overall, over-expression of WT-FTO affected mainly cell cycle-associated DNA replication and repair processes.
STRING analysis of proteins identified in WT-FTO (A) and V493F-FTO (B) interactions. This analysis was conducted with high confidence and without any restriction on the maximum number of interactions in the first and second shells. Three-dimensional structure of the WT-FTO protein (C) and the mutant V493F-FTO protein (D). The red region indicates position 493 in the CTD (http://swissmodel.expasy.org). The structure modelling was performed using Swiss-Model tool based on WT-FTO sequence, >sp|Q9C0B1|FTO_HUMAN Alpha-ketoglutarate-dependent dioxygenase. The mutant FTO contains Valine (V) to Phenylalanine (F) substitution at position 493.
Cells over-expressing the mutant V493F-FTO protein up-regulated HSPA4, DCAF7, RANBP1, and ACTB proteins, whereas they down-regulated ARHGDIA and VCP (TERA) proteins. STRING analysis of these proteins predicted a metabolic network with six nodes (Figure 5B). The VCP protein was associated with several biological processes, including retrograde protein transport from the endoplasmic reticulum (ER) to the cytosol, the ubiquitin-dependent ERAD pathway, cellular response to and protein quality control for misfolded or incompletely synthesized proteins (FDR: 1.81e-05). In addition, both VCP and HSPA4 proteins were associated with topologically incorrect protein (FDR: 2.00e-06) and unfolded protein (FDR: 1.81e-05) responses. In addition, ACTB, ARHGDIA, and RANBP1 proteins were associated with the regulation of catalytic activity displaying pathways (FDR: 4.42e-05). VCP and ACTB proteins were also associated with DNA repair, which is also an associated function of FTO (FDR: 4.01e-2). ARHGDIA and RANBP1 proteins were also associated with positive regulation of hydrolase activity in biological processes (FDR: 1.2e-04). In addition, ARHGDIA was associated with ER unfolded protein response (FDR: 2.95e-02). According to STRING analysis using Kyoto Encyclopedia of Genes and Genomes (KEGG), ARHGDIA was also associated with protein processing in the ER (FDR: 8.88e-05). DCAF7 was associated with protein modification via small protein conjugation or dissociation (FDR: 3.88e-02). Overall, STRING analysis showed that cellular over-expression of WT-FTO caused changes in the regulation of proteins associated with DNA repair and carbon metabolism. Over-expression of the mutant V493F-FTO mainly affects the levels of proteins responsive to cellular stress. This over-expression causes minimal changes in the regulation of proteins involved in the ubiquitin-dependent ERAD pathway and the regulation of cellular responses to misfolded proteins.
We have previously demonstrated that over-expression of either WT- or mutant R316-FTO proteins results in alterations in the regulation of proteins, particularly those involved in energy metabolism and cellular stress (13, 24). The 4-fold over-expression of the HSPA4 protein in the V493F mutant FTO-expressing cultures supported our previous studies. The 3-fold increase in the expression of the ACTB level observed in our study may also be evidence of changes in actin dynamics. Our interest in this work stemmed from the fact that the expression of mutant V493F-FTO also changed the expression levels of ARHGDIA and VCP proteins, which are associated with the ubiquitin-dependent ERAD pathway and protein quality control, may have important implications for cancer biology. These proteins are also involved in the control of misfolded proteins, and cellular response to these events. The expression levels of both proteins decreased 3-fold in mutant protein-expressing cells. The small G proteins of the Ras family are functional proteins with two conformational states (GDP- or GTP-bound), each interacting with different cellular partners. Rho family GTPases, such as Cdc42, Rac1, and RhoA, serve as “molecular switches” that control a variety of cellular processes, such as organelle development and cytoskeletal dynamics (26-28). GTPase activator proteins and guanine nucleotide exchange factors ensure cycling between the GTP- or GDP-bound states of Rho family GTPases. Rho guanine nucleotide dissociation inhibitors (RhoGDIs), which are negative regulators, control their location and activity by blocking the GDP/GTP exchange (29). RhoGDIs can also remove Rho proteins from membranes. Therefore, RhoGDIs modulate the activation of RhoGTPases in response to specific signals and allow them to rapidly migrate to different membrane structures in the cell. Thus, RhoGDIs regulate various functions, including cell division, migration, morphology, vesicular trafficking, and gene expression (30, 31). RhoGDI1, also known as ARHGDIA, is one of the three known members of the most widely expressed RhoGDIs (32). The cytosolic protein ARHGDIA controls both the Rho/Rac membrane association/dissociation cycle and the GDP/GTP cycle (29). It keeps Rho proteins in an inactive cytosolic pool, thereby regulating their stability and preventing their degradation. Studies have shown that without ARHGDIA, the cytosolic pool of Rho GTPases cannot maintain its stability and undergoes rapid degradation through a process that depends on proteasome activity (33). The 3-fold down-regulation of ARGHDIA expression may directly or indirectly affect the cytosolic pool of Rho GTPases, although no major changes were observed after over-expression of the mutant V493F-FTO. Thus, a wide variety of cellular events may be affected by changes in RhoGTPase levels. Changes in the expression levels of two members of the RhoGDI family (RhoGDI1-ARHGDIA and RhoGDI2) have been linked to multiple cancer types, but their nature varies depending on the tumor type (31). Rho protein activation abnormalities have been linked to cancer, infectious and cognitive problems, and cardiovascular disease (34). ARHGDIA expression was increased in colorectal and ovarian cancers and associated with increased invasion and resistance to chemotherapy, whereas it was decreased in brain cancers and hepatocellular carcinoma (31). Therefore, the results of this study, which we performed on neuroblastoma SH-SY5Y cells, cannot be generalized to all tumor types.
VCP (TERA or p97) is also differentially regulated in mutant-expressing cells. VCP is an ATPase-associated protein that plays a role in various cellular activities. Among the metabolic processes, maintaining the quality of nuclear proteins through a process known as nuclear protein quality control can be considered (35). It is also necessary for the division and reassembly of Golgi stacks during mitosis (36). In addition, it plays roles in transferring membranes from the ER to the Golgi (37). Studies have demonstrated that VCP can move incorrectly folded proteins from the ER lumen to the proteasome for degradation (38, 39). Subsequent studies have also revealed the role of this protein in DNA double-strand break (DSB) repair (40, 41). Overall, VCP is a multifunctional protein that maintains homeostasis by regulating protein turnover during ER-associated protein degradation, chromatin-associated degradation, mitochondrial-associated degradation, autophagy, and endosomal trafficking (42-44). Mutations or dysfunctions in VCP have been directly and indirectly associated with multiple neurodegenerative disorders. Because of its critical involvement in protein homeostasis, it is over-expressed in various types of cancer, including human melanoma, breast carcinoma, and cervical cancer cells (45-47). In addition, VCP may play a role in various cardiovascular diseases, including heart failure, myocardial infarction, and diabetic cardiomyopathy (35). Recent work supports the idea that VCP’s regulation of proteostasis is tightly connected to stress signaling, and disruptions in this axis can lead to oncogenic or degenerative outcomes (48). In our study, V493F-FTO over-expression caused a marked down-regulation of VCP, indicating that this mutation may interfere with protein quality control and degradation pathways. Dysregulation of these proteins has previously been linked to tumorigenesis, cancer progression, and therapeutic resistance, suggesting that the V493F-FTO mutation might influence oncogenic processes through modulation of stress response and proteostasis pathways. These findings extend the role of FTO beyond RNA demethylation and propose its involvement in cancer-related cellular mechanisms. To date, there is no report presenting an interaction between VCP and FTO. This is the first study to provide the first indications on the interaction between VCP and FTO. However, further studies are needed to elucidate the details of this interaction to understand the involvement of FTO in VCP-associated metabolic events.
The three-dimensional (3D) structures of WT- and V493F-FTO proteins were predicted using the Swiss model server. This analysis showed no significant structural differences between the WT-FTO (Figure 5C) and V493F-FTO proteins (Figure 5D). A change from valine to phenylalanine at position 493 of the WT-FTO protein does not appear to alter the 3D structure of the protein or affect its catalytic activity, as previously reported by Meyre et al. (10). Both amino acids have hydrophobic side chains, whereas phenylalanine has the additional characteristic of an aromatic ring. Interestingly, this change, which did not change the catalytic activity of the protein, did lead to a change in the expression of certain proteins involved in the ubiquitin-dependent ERAD pathway, such as VCP and ARGHDIA, after over-expression of the mutant V493F-FTO protein. Several works in recent years have reinforced FTO’s role in stress responses, cellular repair, autophagy, and disease pathology. FTO is over-expressed in patients with nonalcoholic steatohepatitis with hepatic lipotoxicity, and FTO knockdown using siRNA protects cells from mitochondrial dysfunction by reducing ER stress (49). In another study, it was found that an increase in FTO expression caused mild ER stress, whereas tunicamycin-induced severe ER stress inhibited osteogenic differentiation by suppressing FTO expression (50). Zhang et al. found that the expression levels of some m6A regulators (METTL3, METTL4, KIAA1429, FTO, and YTHDF2) were increased in patients with heart failure and preserved ejection fraction. Gene ontology analysis of m6A peaks obtained by MeRIP sequencing showed that biological processes such as protein folding (such as Hspa1a, Hspa1b, and HSPH1) and the ubiquitin-dependent ERAD pathway (Serpinh1) are affected (51). In another recent study, it was shown that NR3C1-mediated FTO over-expression increases autophagy by reducing m6A modifications of autophagy-related genes (Atg12, Atg5, Atg16l2, Atg9a), and inhibitor-mediated FTO suppression reverses this process (52). In ischemia–reperfusion injury, FTO over-expression activated autophagy via the Ambra1/ULK1 axis and reduced oxidative stress, indicating that FTO may promote survival pathways under injury conditions (53). Similarly, in mesenchymal stem cells, FTO demethylates SMAD3 mRNA to promote chondrogenic differentiation, highlighting its ability to affect differentiation and stress-related signaling (54). Our proteomic results for the V493F mutant show increased regulation of proteins responding to misfolded proteins and ER stress (e.g., HSPA4, VCP, ARHGDIA), suggesting that this mutation may tilt FTO’s function more toward proteostasis and stress response rather than classical cell-cycle or DNA repair roles observed with WT. These observations align with the broader trend in literature that FTO is not only a regulator of mRNA methylation per se, but also a modulator of cellular stress, protein folding, and quality control pathways. These studies indicate that the FTO protein can affect the expression levels of proteins targeted not only to nuclear and cytoplasmic structures but also to organelles and membranous structures, indicating that FTO is a multifunctional protein. Our results suggest additional potential roles for FTO. This study provided indications for undiscovered roles of FTO. To shed more light on these roles, additional in vivo and in vitro studies are needed.
Conclusion
This was the first study to investigate the effects of over-expressing the mutant V493F-FTO protein on the SH-SY5Y proteome. Our findings revealed that over-expression of the mutant protein significantly affected the expression of proteins responding to cellular stress, whereas minimal changes were observed in the regulation of proteins involved in the ubiquitin-dependent ERAD pathway and cellular response to misfolded proteins. These changes implied that FTO may be associated with the ubiquitin-dependent ERAD pathway and that this involvement may be independent of its well-known demethylase activity. Taken together, these findings not only provide the first evidence for a potential link between FTO and the ERAD pathway but also suggest that FTO mutations may contribute to oncogenic mechanisms by altering protein homeostasis and cellular stress responses. Further studies are warranted to explore the role of FTO variants in cancer development and progression.
Acknowledgements
This research was supported by TUBITAK under Grant number 113S965 and Kocaeli University under Grant number TAA-2025-4290. The author would like to express her sincere gratitude to Prof. Dr. Murat Kasap, Prof. Dr. Gurler Akpinar, and the members of the proteomics laboratory for their guidance regarding the use of the laboratory infrastructure and for providing general technical assistance throughout the study. She also gratefully acknowledges Prof. Dr. Mehmet Dogan Gulkac for his continuous motivation and encouragement.
Footnotes
Conflicts of Interest
The Author declares that there are no conflicts of interest regarding the publication of this article.
Artificial Intelligence (AI) Disclosure
During the preparation of this manuscript, a large language model (ChatGPT 4.0, OpenAI) was used solely for language editing and stylistic improvements in select paragraphs. No sections involving the generation, analysis, or interpretation of research data were produced by generative AI. All scientific content was created and verified by the authors. Furthermore, no figures or visual data were generated or modified using generative AI or machine learning–based image enhancement tools.
- Received September 3, 2025.
- Revision received October 7, 2025.
- Accepted October 9, 2025.
- Copyright © 2026 The Author(s). Published by the International Institute of Anticancer Research.
This article is an open access article distributed under the terms and conditions of the Creative Commons Attribution (CC BY-NC-ND) 4.0 international license (https://creativecommons.org/licenses/by-nc-nd/4.0).












