Long-read sequencing-based analyses of the adult Drosophila brain transcriptome in physiological and pathological settings

Optimal brain function requires that neurons carry out extensive post-transcriptional RNA processing to produce a vast diversity of transcripts. Accurate reconstruction and quantification of highly processed RNA using standard RNA sequencing approaches is challenging due to their short read lengths. Long-read direct RNA sequencing can resolve multiple variations within RNA isoforms by capturing full-length transcripts spanning multiple exon-exon junctions, repetitive regions (e.g. retrotransposons), and intronic structures. Here we produce an isoform-level map of N6-methyladenosine (m6A) RNA modifications using Oxford Nanopore Technologies (ONT) long-read sequencing of native RNA strands extracted from heads of Drosophila melanogaster aged to day 10 of adulthood. In addition to identifying 930 transcripts that are not present in the reference transcriptome, we find that almost half of the total detected isoforms have polyadenylated tails in excess of 104 nucleotides and that over 59% of transcripts possessed detectable m6A-modified bases. RNA modifications are present in RNA transcribed from transposable elements, which are important drivers of genetic diversity and relevant to human neurodegenerative diseases, including Alzheimer’s disease and related tauopathies. Applying nanopore direct RNA sequencing to a Drosophila model of tauopathy with known transposable element activation and various types of errors in RNA handling reveals exceptionally diverse RNA processing events in regions that are considered difficult to characterize with traditional short-read sequencing. Taken together, we have uncovered complex transcript structures in adult Drosophila head in a physiological setting and in the context of tauopathy, laying the groundwork for future studies to characterize the diverse tau transcriptome in brain tissue from patients with Alzheimer’s disease and related tauopathies.