Thesis title: Structural and Mechanistic Studies of Nucleic Acid Demethylase FTO
Supervisors: Prof Wanjin Hong and Assoc Prof Yong-Gui Gao
FTO gene was the first gene that provided the strongest direct link to obesity uncovered by GWAS. Functional FTO is crucial for normal growth and development. Later on, FTO was identified as an m6A demethylase. M6A modification was known to be a prevalent mark across all kingdoms of organisms. In mammals, m6A occurred within a consensus sequence and was initially believed to be a permanent post-transcription modification in RNA. The ability of FTO to reverse m6A methylation prompted revisitation on the role of m6A in RNA transcripts. The status of m6A in RNA transcripts are further dynamically regulated by methyltransferases and m6A-binding proteins. This affects RNA splicing, translation and degradation, representing epigenetic regulation at the RNA level. FTO and ALKBH5 belonging to the AlkB 2OG oxygenases family are the only m6A demethylases known so far. They are implicated in unrelated diseases such as obesity for FTO and spermatogenesis for ALKBH5, revealing very different biological roles. It is not clear how FTO is selectively expressed over ALKBH5 and how FTO regulate the RNA methylation status in vivo. Thus, the mechanistic and structural studies of FTO will aid in the elucidation of FTO roles and functions.
We developed and reported the first potent, selective and cell-active inhibitor for FTO through a combined approach of X-ray crystallography, DSF and biochemical assays. Human AlkB oxygenases have conserved 2OG-binding site but not substrate-binding site. This is supported by DSF which revealed m3T substrate only preferentially binds to FTO. Encouraged by this, we envisioned that selectivity and potency could be achieved by exploiting the substrate-binding site. The inhibitor we have developed has at least 30-fold selectivity over other human AlkB oxygenases. Particularly, the inhibitor demonstrated a 100-fold preference for FTO (IC50 = 0.81 μM, Tm shift = 11.2°C) over ALKBH5 (IC50 = 108.1 mM; Tm = 2.7°C). The co-crystal structure of FTO complexed to inhibitor solved at atomic resolution revealed the basis of selectivity. The position of the pyridyl nitrogen of the inhibitor and Glu234FTO is crucial for the selectivity. This is important as we can potentially, selectively perturb FTO in cells. We further validated that the cell-activity of the inhibitor. Potentially, this inhibitor can serve as a functional probe for mechanistic studies of FTO in cells.
Next, we asked what determined the substrate specificity of FTO. Through in vitro demethylation studies, we concluded that FTO does not need m6A-containing consensus sequence for activity. Various secondary structural conformations are adopted by mRNA. Although m6A do not prevent duplex formation, the methylation was found to destabilise Watson-Crick base pairing by 0.5-1.7 kcal/mol depending on sequential context. We purposefully designed various m6A-containing RNA oligonucleotides that are highly similar in sequences that were able to form duplexes or hairpins in vitro. We showed that FTO and ALKBH5 are able to selectively demethylate m6A-containing hairpins. Demethylation caused a conformational change of the corresponding RNA oligonucleotides, adopting a duplex conformation. With this validation, it is fair to ask whether FTO performs its biological role by affecting secondary structures of m6A-containing mRNA.
For more information on Wanjin HONG’s lab, please click here.
For more information on Yong-Gui GAO’s lab, please click here.