FTO – A Common Genetic Basis for Obesity and Cancer

1The morbidity of obesity and cancer is increasing year by year in most countries around the world and represents a threat to human health. Obesity causes changes in the body’s physiological and hormonal environments that promote many diseases, including diabetes and cardiovascular diseases. Obesity has been proven to increase the risks of at least 13 different types of cancers, such as esophageal adenocarcinoma, colon cancer, endometrial cancer, postmenopausal breast cancer, kidney cancer, and hematopoietic cancers. Of all Americans diagnosed with cancer in 2014, the overweight and obese population account for 40%. Furthermore, another prospective study of large samples of Americans confirmed that 14% of cancer deaths in males and 20% of females are due to them being overweight or having obesity. The biological mechanism of obesity and cancer are complex, including obesity-related hormones, growth factors, multiple signaling pathways, and chronic inflammation. In recent years, FTO SNPs have been firmly associated with increased body mass index (BMI) and higher risks of various types of cancers in people of multiple races, and the role of FTO SNPs in the development of obesity and cancer has been gradually revealed. This review details this role and the molecular mechanisms of FTO in obesity and cancer, as well as its potential clinical applications as a therapeutic target.

2In 1999, FTO was first cloned by exon trapping analysis in Fused toes (Ft) mutation mice. Initially, FTO was expected to be associated with programmed cell death because scientists observed that heterozygous mice with Ft mutation developed syndactyly in the forelimb part and thymus hyperplasia. In 2007, the GWAS study identified FTO as an obesity sensitivity gene, and multiple SNPs in the intron 1 region were strongly associated with BMI, body fat rate, waist circumference, hip circumference, and energy intake. As a result, the gene was named as the fat mass and obesity-associated (FTO) gene and has received extensive attention.

3According to current genomics research, the FTO gene only exists in vertebrates and a few kinds of marine algae with highly conserved nucleotide and amino acid sequences. The human FTO gene is located on chromosome 16q12.2, encoding a 2-oxoglutarate (2-OG) Fe(II)-dependent AlkB family dioxygenase, with a total length of 410.50 kb including 9 exons and 8 introns. About 3.4 kb upstream of FTO gene was Merkel’s diverticulum syndrome-associated gene (RPGRIP1L), and its downstream was close to Iroquois gene family (including IRX3, IRX5, IRX6) (Supplementary Figure). FTO is extensively expressed in adipose tissues and the skeletal muscles of human tissues, with the highest expression in the hypothalamus in the region that controls energy balance, namely the arcuate nucleus, which indicates that it may play a critical role in regulating appetite and energy metabolism.

4In 2007, Thomas et al. revealed that the FTO gene encodes Fe(II)/2-OG dependent demethylase, which is the ninth AlkB family protein found in mammals (also called ALKBH9). They also used purified FTO protein from recombinant mice or humans that can catalyze the demethylation of 3-methylthymine(3-meT) and 3-methyluracil (3-meU) with the help of Fe(II)/2-OG. Later, He et al. found that N6-methyladenosine (m6A) in nuclear RNA was a main substrate of the FTO. Therefore, the FTO was identified as the first RNA demethylase, thus initiating a wave of research on epigenetic modifications of RNA. Since then, the complex and diverse functions of FTO proteins have been gradually revealed. FTO can bind to multiple types of RNAs, including mRNA, snRNA, and tRNA, and can demethylate m6A and N6,2′-O-dimethyladenosine (m6Am) in mRNA, m6A in U6RNA, m6Am in snRNAs, and N1-methyladenosine (m1A) in tRNA (; Figure 1A). However, m6A is the most favorable nucleobase substrate of FTO.

5M6A, methylation modification on the sixth nitrogen atom of adenine is the most common mRNA methylation enriched in the 3′-untranslated region (3′-UTRs), between the stop codon and the start codon. M6A modifications were subjected to reversible and dynamic regulations including writers (METTL3, METTL14, and WTAP), erasers (FTO and ALKBH5), and readers (YTH domain family and IGF2BPs) (;;;;,,;;; Figure 1B). Based on this extensive existence and dynamic regulation, m6A plays an important role in post-transcriptional regulatory processes such as RNA splicing, nuclear production, degradation, and translation.

6Since FTO has been identified as the first obesity-related gene. By conducting GWAS analysis, researchers have found that FTO SNPs are associated with obesity and higher risks of various cancers in multiracial populations (Supplementary Figure).

7The connection between FTO SNPs and BMI was first found in European people with diabetes. The classic BMI-related FTO SNPs were rs9939609 (T/A), and compared with those who did not carry the risk allele, 16% of adults who carried the homozygous risk allele gained nearly 3 kg in weight, and the risk of obesity increased by 1.67 times. Serial GWAS studies on obesity-related traits in people of European descent have confirmed the important role of the FTO locus, and many other FTO SNPs in the intron 1 region have been reported, such as rs9930506, rs1421085, rs8050136, rs1121980, and so forth. The obesity-associated FTO SNPs in East Asian populations are comparable to that of people of European descent. The risk allele A of FTO SNP rs9939609 was closely related to obesity and BMI in Chinese, ethnic Chinese, Malaysian, Singaporean, East, and South Asian people. A large-scale meta-analysis targeting GWAS analysis of East Asian populations found that FTO SNPs rs17817449 have the most significant correlation with BMI in people of Chinese descent. In recent years, global studies have associated the rs9939609 variant with higher obesity risks in other populations (including Brazilian people, early adolescence in China, and adults in Shiraz, Iran). It is also associated with increased BMI and waist circumference (in Brazilian youths), adipose tissue distribution (in Italian people), and increased metabolic syndrome susceptibility (in Chinese populations). In terms of the potential mechanisms between mutations in FTO and increased risks of obesity, studies have proven the role of FTO in the influence of food intake. People carrying FTO risk alleles and are inclined to higher energy intake foods like fat or proteins, reduced satiety, resulting in overeating, and many even lose control when eating. Another large-scale meta-analysis showed that the homozygous FTO risk allele was associated with a 27% lower risk of obesity in physically active adults.

8To date, it has been studied that variants of FTO rs9939609, rs8050136, rs1477196, rs6499640, rs1121980, rs17817449, rs11075995, rs8047395, and rs7206790 have an association with a higher risk of cancers. The most typical FTO SNP rs9939609 was associated with lung cancer, renal cancer, breast cancer, prostate cancer, pancreatic cancer, endometrial cancer. Multiple SNPs in the intron 1 region of the FTO (including rs9939609, rs1477196, rs7206790, rs8047395) have been correlated with the risk of breast cancer, with rs1477196 strongly associated. Interestingly, Da et al. observed that the interaction of FTO and MC4R polymorphisms showed a strong association with breast cancer: there was a 4.59-fold increased risks for women who have the allele combination C/T/C (FTO rs1121980/FTO rs9939609/MC4R rs17782313). In addition, in 2013, Iles et al. found an association between FTO rs16953002 and rs12596638 and melanoma susceptibility. However, these two SNPs are located in intron 8 of the FTO gene rather than intron 1 (the BMI-related region). This suggests that the association between the FTO variant and a wide range of diseases may play a role beyond BMI.

9FTO proteins are widely involved in both adipogenesis and tumorigenesis by m6A-dependent demethylase activity which influences several mRNA processing events (Table 1).

10FTO proteins are involved in the development of obesity by affecting the m6A level of hormones related to eating or molecules related to adipogenesis (Figure 2A). In 2013, Efthimia et al. found that FTO over-expression limited the m6A modification of ghrelin mRNA in cell models and increased ghrelin mRNA and peptide levels concomitantly. This article provided insights into how FTO predisposes to stimulated energy intake and obesity in humans. Simultaneously, substantial evidence has proved that FTO participates in the process of adipogenesis. Zhao et al. found that FTO regulates the exonic splicing of the adipogenic regulator RUNX1T1 by influencing the level of m6A around the splice site, thereby modulating cell differentiation. Moreover, FTO affects adipogenesis by regulating the process of mitotic clonal expansion (MCE), which is a prerequisite for adipocyte differentiation that occurs within 48 h of adipogenic stimulation. The overexpression of FTO can induce MCE and regulate the differentiation of preadipocytes by influencing the expression of m6A-dependent transcription factors. Furthermore, Wu et al. found that FTO regulated adipogenesis by dominating cell cycle proteins such as CCNA2 and CDK2 by m6A-YTHDF2 dependent pathway, revealing a new mechanism about anti-obesity and anti-adipogenesis activity of Epigallocatechin gallate (EGCG).

11FTO is highly expressed in many cancer tissues, which plays the role of an oncogene in an m6A-dependent way and participates in the regulation of the malignant phenotype of cancer cells (Figure 2A). In 2017, Li et al. found that FTO enhanced leukemia oncogene-mediated cell transformation and leukemogenesis and inhibited all-trans-retinoic acid (ATRA)-induced AML cell differentiation by affecting the expression of targets mRNAs such as ASB2 and RARA, through reducing the level of m6A. Moreover, the elevated level of m6A was found to promote the growth, self-renewal, and tumorigenesis of Glioblastoma Stem Cells (GSC) as well as prolong the lifespan of GSC-grafted mice. In 2018, Niu et al. showed that FTO can promote breast cancer cell malignant phenotype through epigenetically demethylated m6A in BNIP3 mRNA 3′UTR. Similarly, the overexpression level of FTO was also found in gastric cancer, advanced non-small cell lung cancer, and hepatocellular carcinoma, which can regulate cell proliferation and/or migration/invasion through targeting demethylation for the m6A of Wnt/PI3K-Akt, USP7, MZF1, or PKM2, respectively.

12FTO may also have an impact on the therapeutic response of cancer (Figure 2A). In 2018, Zhou et al. found that FTO enhanced the chemo-radiotherapy resistance of cervical squamous cell carcinoma both in vitro and in vivo through influencing the expression of ß-catenin by reducing m6A levels. The overexpression of FTO in leukemia cells can promote its expression by catalyzing the demethylation of cell proliferation-related genes such as m6A of MERTL and BCL-2 mRNA and affecting the generation of resistance phenotypes in the treatment with tyrosine kinase inhibitors (TKI). Recently, He et al. found that the down-regulation of FTO made melanoma cells sensitive to interferon-gamma (IFN-γ) and anti-PD-1. FTO played a crucial role by promoting melanoma tumorigenesis and anti-PD-1 resistance.

13The mammalian target protein rapamycin (mTOR) is an atypical serine/threonine kinase, which is the core component of regulating mRNA translation and can promote cell growth according to environmental signals. mTOR binds to a variety of chaperone proteins to form two different kinase complexes, i.e., mTOR complex 1 (mTORC1) and mTORC2.

14In 2013, in vivo experiments demonstrated that FTO played a role in the coupling of amino acid level and mTORC1 signaling pathway. FTO deficient cells showed reduced activation of the mTORC1 pathway, decreased mRNA translation rate, and increased autophagy. mTORC1 is a negative regulator of autophagy, which is a major cellular digestion process. In response to nutrition and environmental stress, autophagy plays a critical role in the occurrence and progression of obesity and cancer. Furthermore, mTOR activates the Warburg effect by inducing PKM2 and other glycolytic enzymes under normoxic conditions. In summary, mTORC1 can regulate the development of obesity and cancer through autophagy or direct activation of downstream signaling pathways.

15Mutations in the FTO gene raise blood levels of leptin, a known mediator or growth factor between obesity and colon cancer, which activates a variety of pathways associated with colon cancer. In addition, leptin has been suggested as an intermediate link between obesity and breast or prostate cancer. Intriguingly, mTOR is also one of the signal mediators of obesity related factors, such as leptin, adiponectin, and inflammatory cytokines, through the Akt/PI3K or AMPK pathways. This seems to coincide with the FTO-mTOR pathway, discussed in the previous paragraph. In summary, FTO can directly or indirectly target mTOR, thus regulating the occurrence and progress of obesity and cancer in a broad manner (Figure 2B).

16With the gradual disclosure of the important functions of FTO as mRNA demethylase in many diseases, the crystal structure of FTO has been resolved since 2010, and the development and applications of its specific inhibitors have attracted extensive attention. FTO inhibitors that have been shown to have anti-obesity or anti-cancer effects in vitro or in vivo are summarized in Table 2.

17In 2012, Yang et al. reported natural product Rhein as an inhibitor of human FTO demethylase, which can competitively bind to the FTO catalytic site. Before this study, Rhein was thought to prevent or even reverse weight gain and obesity caused by high-fat diets. The catechin EGCG, another natural compound rich in green tea, was found to play anti-obesity and anti-adipogenesis roles through the FTO-m6A-YTHDF2 axis. Recently, Peng et al. identified Entacapone as a potential FTO inhibitor, which has the effect of reducing weight and lowering blood glucose. It was initially approved as an adjunctive therapy combined with levodopa and carbidopa for the treatment of Parkinson’s disease. Entacapone had an effect on gluconeogenesis and adipogenesis in the liver of mice by acting on an FTO-FOXO1 regulatory axis.

18As for cancers, there are more studies on the application of FTO inhibitors, especially in the treatment of leukemia, glioblastoma, and breast cancer. Compared with single drug therapy, Rhein combined with nilotinib is a more effective treatment for leukemia in mice. Recently, Yang et al. identified meclofenamic acid (MA) as a highly selective inhibitor of FTO, which can compete with FTO binding for the m6A-containing nucleic acids. The inhibitor FB23 was designed and synthesized from the chemical scaffold of MA, which exhibited a more potent inhibition for FTO demethylation in vitro. Its bioisostere FB23-2 can inhibit the leukemogenesis in cells and in the patient-derived xenografted (PDX) mouse model. Su et al. found that R-2HG (oncometabolite produced by mutant isocitrate dehydrogenase 1/2 (IDH1/2) enzymes), compounds CS1 and CS2 were also targeted inhibitors of FTO. By inhibiting its demethylation function, they affected related signaling pathways (such as the MYC pathway) and played an active role in inhibiting the proliferation of AML cells in vivo and in vitro. Compounds CS1 and CS2 extended the overall survival of transplanted mice with primary MLL-AF16 cells and made AML cells sensitive to other curative drugs, such as decitabine, a tyrosine kinase inhibitor, and IDH2mut inhibitor. Comparing the anti-leukemic activities of the four FTO inhibitors, CS1 and CS2 showed higher activity in inhibiting cell viability, and their IC50 values were 10–30 times lower than FB23-2 or MO-I-500. MO-I-500 is another FTO inhibitor, which selectively inhibits the demethylation of FTO and increases the m6A levels in cells. In addition, among the above-mentioned FTO inhibitors, Rhein and MO-I-500 have been reported to significantly inhibit the growth ability of breast cancer cells in vivo and in vitro. MA2 (the ethyl ester form of MA) and R-2HG had inhibitory effects on glioma. Compared with MA, MA2 has a better cell penetration, significantly increased m6A methylation in cells, suppresses glioblastoma progression, and prolongs the lifespan of GSC-grafted animals. Interestingly, the previous anti-obesity EGCG, Entacapone, also had an inhibitory effect on cancer. For example, EGCG had an inhibitory effect on lung cancer, breast cancer, colon cancer, metastatic pancreatic cancer, and prostate cancer or had a sensitivity to chemotherapy. It was noted that it can be used as an adjuvant drug in cancer treatments. The combination of Entacapone and EGCG can synergistically enhance the growth inhibitor of lung cancer cell lines.

19In summary, more FTO inhibitors are displaying positive therapeutic effects in animal disease models, and represent promising therapeutic targets for obesity and cancer (Figure 2A).

20Single nucleotide polymorphism sites are the main form of human genome DNA sequence variation and can regulate gene expression. From FTO polymorphisms which have a risk for obesity and cancer to the specific mechanisms regulating these diseases through nucleic acid demethylation of FTO proteins, FTO SNPs seem to regulate the expression level of FTO and affect its enzymatic function, playing an important role in obesity and cancer. Some studies support this hypothesis. For example, In 2008, a study of a Mexican population revealed that in obese patients, the FTO risk allele was significantly correlated with high FTO expression. Subsequently, Tea et al. and Efthimia et al. revealed that FTO mRNA caused by the risk allele was more abundant than non-risk alleles at least in blood cells. Unfortunately, the mechanism for the correlation between FTO SNPs and obesity or cancer has been elusive. So far, there are no studies that provide indisputable evidence for these associations.

21Moreover, some studies have suggested that FTO SNPs may be associated with obesity by regulating the expression of adjacent genes (; Figure 3A). The Leibel group found that the rs8050136 of the intron 1 region of FTO overlapped with the binding site of transcription factor Cut Like Homeobox 1 (CUX1). This SNP nucleotide type can affect the transcriptional activation of FTO and retinitis pigmentosa GTPase regulator interacting protein 1 like (RPGRIP1L) by CUX1 P110. For individuals with the obesity risk allele at rs8050136, the expression of RPGRIP1L and FTO in the hypothalamus were decreased due to the low binding affinity of CUX1 P110 to DNA, RPGRIP1L can affect the location of leptin receptors and leptin signaling in neurons and lead to increased food intake and obesity. In addition, Jowett et al. found that in combination with gene variation and expression data from the human cohort, the A allele of rs8050136 was positively associated with the expression level of RBL2, and an increase in RBL2 level might help to restrict the clonal expansion of A population of precursor adipose cells during development. Moreover, Smemo et al. found that these sites also contained an enhancer sequence that can bind to the promoter of IRX3. Using expression quantitative trait loci (eQTL), they found that obesity-related SNPs such as rs9930506 were correlated with the expression of IRX3 in human brain samples. Mice lacking IRX3 lost 25–30% of their body weight through increasing basal metabolic rate and browning of white adipose tissues. The study of Claussnitzer et al. also supported the regulatory relationships between FTO SNPs and IRX3 expression. They suggested that changes in the rs1421085 risk allele led to a double expression of IRX3 and IRX5 through disruption of the conservative motif of the ARID5B repressor in the early stage of adipocyte differentiation. In this case, brown fat cells transform automatically into white fat cells and lower down the mitochondrial thermogenesis by five times. This study explained the correlation between FTO SNPs and obesity by using the effect of the autonomous transformation of fat cells on thermogenesis. These studies provided a plausible mechanism for the correlation between SNP variation of FTO intron 1 and obesity (Figure 3B).

22Adjacent genes (RPGRIP1L, RBL2, IRX3, IRX5) regulated by FTO SNPs are also involved in the occurrence and progression of cancer in various ways (Figure 3B). For example, studies have shown that RPGRIP1L was one of the tumor suppressor genes of human hepatocellular carcinoma. RBL2, a member of the retinoblastoma (Rb) family, was inactivated by cell cycle kinases and was the basis of various cancer types. The emerging role of RBL2 in aging and apoptosis also appeared to play an active part in tumor inhibition. IRX3 was highly expressed in approximately 30% of patients with AML, and high expression of IRX3 alone can perpetuate hematopoietic stem cells and progenitor cells (HSPC) in bone marrow cultures and induce lymphoid leukemia in vivo. In addition, IRX3 and IRX5 have been reported to participate in the transition from intestinal adenoma to colorectal cancer by negatively regulating the Dpp/TGF-ß pathway. IRX5 alone has also been reported to be an oncogenic gene in hepatocellular carcinoma, colorectal cancer, prostate cancer, and non-small cell lung cancer by regulating cancer cell cycle and apoptosis. Although there is no direct evidence that FTO SNPs are associated with cancer by affecting the expression of adjacent genes, we cannot rule it out as a possibility.

23Based on these findings, the correlation between FTO SNPs and obesity and cancer may be due to the regulation of FTO enzyme activity or expression of adjacent genes. However, more convincing and systematic research studies are needed to decipher the causal mechanism between FTO non-coding variants and obesity or cancer. A healthy lifestyle such as proper diet and moderate exercise is recommended to minimize the negative effects of obesity susceptibility genes before we can fully understand the underlying mechanisms.

24Although the specific mechanisms for FTO polymorphism and high risk of obesity and cancer are elusive, the correlation is definite. As for FTO, it can participate in the disease progression of obesity and cancer in m6A-dependent post-transcriptional regulation, or by targeting mTOR. More importantly, some drugs have been shown to inhibit obesity, and some cancers such as leukemia, glioblastoma, and breast cancer by targeting FTO. In particular, FB23, R-2HG, and CS1/CS2 have shown anti-leukemia effects through in vivo experiments, and MA2 can also inhibit the activities of glioblastoma cells in vivo. This evidence suggests FTO as the common genetic basis of obesity and cancer and a potential target for obesity and some cancers.

25WY oversaw and guided the process of writing this manuscript. NL wrote and edited the manuscript. YL, YZ, SP, XN, HX, and JL put forward suggestions for the manuscript. All authors read and approved the final manuscript.

26The authors declare that the research was conducted in the absence of any commercial or financial relationships that could be construed as a potential conflict of interest.