SINEUP lncRNAs: from molecular mechanism to therapeutic application.

The post-genomic era has brought to light a previously unknown world of transcripts with the discovery of non-coding RNAs (ncRNAs). Indeed, it became evident that only as few as 1-2% of mammalian transcriptome consists of protein-coding mRNAs. Among several families of ncRNAs, long non-coding RNAs (lncRNAs) are under intense scrutiny for their heterogenicity of forms and molecular activities. A new class of antisense lncRNAs, known as SINEUPs, were previously identified for their ability to specifically enhance the translation of their target sense mRNA. LncRNAs and mRNA were transcribed from a sense/antisense pair locus with an head-to-head divergent configuration. SINEUPs activity relies on the combination of two domains: an overlapping region, or binding domain (BD), that confers specificity, and an embedded inverted SINEB2 element, or effector domain (ED), enhancing target mRNA translation. This new class of transcripts embodies the model of lncRNAs as flexible and versatile modular scaffolds enabling interactions between RNA, DNA and proteins. Furthermore, it represents a promising new RNA therapeutics platform to increase endogenous expression of a protein of interest within a physiological range. In this work, I provided new insights on the molecular mechanism of SINEUP activity, focusing on the role of N6-methyladenosine (m6A) modification, and on a Proof-Of-Concept therapeutic application of SINEUPs to rescue haploinsufficient OPA1 gene expression in Dominant Optic Atrophy (DOA). m6A is the most common RNA modification found in mRNAs and ncRNAs, where it is post-transcriptionally installed in the cell nucleus and can exert regulatory functions in many cellular processes such as nuclear export and translation. Here, I observed that both the natural SINEUP AS Uchl1, acting in rodent cells, and the synthetic shorter miniSINEUP-DJ1, acting in human cells, are m6A-modified. Results indicate METTL3 enzyme as the main responsible for SINEUP RNA modification. I then applied Nanopore direct RNA sequencing to map m6A-modified residues and a reverse transcription assay for validation. I monitored SINEUP activity upon METTL3 knock-down and in the presence of mutations on sites of m6A deposition. Interfering with a proper m6A modification led to a dominant negative effect of SINEUPs RNA on endogenous DJ1 protein levels in both experimental conditions. Applying ribosome fractionation analysis in conditions of inhibition of proper m6A deposition, I observed an enrichment of the target DJ1 mRNA associated to 40S and 60S ribosome fractions and a concomitant depletion from polysomes. These results provide a mechanistic model for its dominant negative effect on endogenous DJ1 protein. These data also suggest the presence of an m6A-dependent step in the molecular mechanism of SINEUP activity at the ribosome and contribute to a better understanding of the role of RNA modifications in the regulation of lncRNAs function. From a therapeutic point of view, SINEUPs are proposed as a new platform for the treatment of i. haploinsufficient diseases, where the lack of a functional allele prevents healthy phenotype formation; ii. complex multifactorial diseases, where increasing a compensatory pathway could preserve or restore physiological activities. Here, I applied SINEUP technology to increase endogenous levels of OPA1 protein to treat DOA, the most common inherited optic neuropathy caused in 75% of cases by heterozygous mutations in OPA1 gene. DOA is an early-onset autosomal dominant haploinsufficient disorder, with a prevalence ranging from 1:12000 to 1:50000 births and characterized by degeneration of the retinal ganglion cells that leads to optic nerve atrophy and blindness. OPA1 is a ubiquitously expressed dynamin-related GTPase protein with crucial functions in mitochondrial homeostasis, that localizes in the Inner Mitochondrial Membrane (IMM), reaching highest expression levels in brain, retina and heart. By in vitro screening, I identified OPA1-specific miniSINEUPs able to increase selectively both human and murine OPA1 proteins in a range sufficient to restore neuronal cell functions. Currently, a major limitation to the development of SINEUPs as a RNA drug is represented by their length, that should be reduced to less than 60 nts to allow cost-effective manufacturing and efficient in vivo delivery. Recently, encouraging data have proved that the incorporation of chemically modified ribonucleotides restores IVT SINEUP RNA activity, making an important progress for its development as a drug. Here, I successfully designed and tested shorter SINEUP RNA variants that allowed us to reduce their size from 250 nts down to 50 nts. Indeed, by transfecting 2’OMeA modified ASO-SINEUP-OPA1, I was able to upregulate endogenous OPA1 protein translation of around 1.8 fold, as achieved with standard plasmid-driven expression of the same nanoSINEUP-OPA1 RNA. Most importantly, I applied previously selected mini- and nanoSINEUP to prove the functional rescue of DOA patients’ fibroblasts defects in mitochondrial morphology and activity. In summary, I was able to identify OPA1-specific SINEUPs promoting the recovery of disease-associated defects in patient-derived cellular model of DOA and I optimized SINEUP technology for its development as RNA therapeutic molecule for the treatment of haploinsufficient diseases.