6. CONCLUSION
Rapidly accumulating evidences of the role of RNA modifications in cell differentiation and development and their dysregulation in cancer have emerged in the last years (Barbieri and Kouzarides, 2020; Rosselló-Tortella et al., 2020). Although still in its infancy, the potential of pharmacological targeting of RNA modifiers for reverting aberrant epitranscriptomes strongly energized the research of chemical biology and drug discovery in this field (Boriack-Sjodin et al., 2018). Below, we highlight some of the key limitations and challenges that will need to be addressed in the near future to improve the drug discovery in the RNA scenario (Figure 2 ).
It should be noticed that the field does not start from the zero \soutpoint, since the structures of key enzymes (writers, erasers) and reader proteins are currently well-characterized. This is a good starting point for guided and precise design of small compounds targeting these structures based on computational methods. To date a sizeable number of epitranscriptomic inhibitors have been reported, but in actuality not all small molecule inhibitors have demonstrated acceptable target potency or enzyme selectivity. Thus, there is an unmet need to develop selective and more effective small-molecule inhibitors of RNA modifiers for therapeutic applications. Apart from rational chemical design approaches, we still need to learn more about the biochemistry of RNA modifiers, their ligand binding pockets and the downstream pathways.
A very important matter to consider is our current limited vision of the mode of action of RNA modifiers. In other words, we are establishing associations between two observations (i.e., an altered m6A level and a tumour-suppressive effect) but, to date, we cannot determine causality between both observations. Most of the studies trying to establish phenotypic connections applied approaches that involve genetic manipulation of the full RNA-related gene. These approaches do not allow to solve whether the resulting phenotype is due to “druggable” biochemical mechanisms (e.g., enzyme catalysis or ligand binding) or other non-easy actionable targets linked to protein interactions (e.g., scaffolding, protein–protein interactions or chaperone mechanisms) (Boriack-Sjodin et al., 2018). Or even less, whether the RNA modifications result in changes in secondary structures or protein-protein interactions is still uncertain.
More sophisticated biological approaches apart from cell-based assays are essential to identify the functional effects of targeting the epitranscriptome. On the basis of what we have learned from epigenetic- based therapy, it is expected that RNA modifications work in multimeric protein complexes. Consequently, the translation of the results obtained in in vitro cell based assays do not replicate the physiological conditions from in vivo models. Animal models carrying RNA modification defects (e.g., inducible knockout or mutant mice for RNA-modifying enzymes) are a move in the right direction to unravel the physiological function of RNA modifications. Furthermore, as previously described (Cheng et al., 2018; Huang et al., 2019a), most of the actual RNA modifying enzymes shared catalytic sites and cofactors with DNA and histone epigenetic machinery, and thus, drug-based intervention on RNA modifications can induce unforeseen effects on other regulation systems including epigenetics.
Although we are seeing improvements, the high number of possible RNA modifications and the complexity of molecular pathways involved are still an impediment to evaluate the biological consequences of small molecules intervention targeting the epitranscriptome (Jia et al., 2011; Wang et al., 2014). An aspect of intense debate is whether the ubiquitous nature of RNA modifications could increase the toxicity of associated drugs as it would be extremely difficult to prevent a pleiotropic effect. This knowledge, undoubtedly, also draws from the technical side. RNA modification detection methods have shown a tremendous improvement, particularly since the development of valid methods for studying the epitranscriptome at wide scale using NGS (Linder et al., 2015). Improvements to avoid methods dependent on antibody recognition or site-specific cleavage linked to radiolabelling, like MAZTER-seq (Garcia-Campos et al., 2019) or m6A-REF-seq (Zhang et al., 2019), held promises to allow a better quantification and precision of the epitranscriptome in specific contexts. Technological advances for the identification and quantification of low-abundance RNA modifications, establishment of internal standards as controls or development of bioinformatic tools to generate, analyse, and standardize protocols should be also a priority to ensure the reliability of the data (Morena et al., 2018).
In overcoming these (and other) chemical, biological and technical barriers, we will have a better and clearer view of the epitranscriptome map, its contribution to signalling pathways and its role in human health and disease. With this comprehensive overview of “epitranscriptome Science”, the development of innovative therapeutic intervention of RNA modifications will be an exciting reality.