Free circular introns with an unusual branchpoint in neuronal projections

Abstract

9The polarized structure of axons and dendrites in neuronal cells depends in part on RNA 10 localization. Previous studies have looked at which polyadenylated RNAs are enriched in neuronal 11 projections or at synapses, but less is known about the distribution of non-adenylated RNAs. By 12 physically dissecting projections from cell bodies of primary rat hippocampal neurons and 13 sequencing total RNA, we found an unexpected set of free circular introns with a non-canonical 14 branchpoint enriched in neuronal projections. These introns appear to be tailless lariats that 15 escape debranching. They lack ribosome occupancy, sequence conservation, and known 16 localization signals, and their function, if any, is not known. Nonetheless, their enrichment in 17 projections has important implications for our understanding of the mechanisms by which RNAs 18 reach distal compartments of asymmetric cells. 19 20 42The interplay between intron retention and neuronal RNA localization has been studied in several 43 individual cases (Chen et al., 2008; Bell et al., 2010; Buckley et al., 2011; Khaladkar et al., 2013; 44 Ortiz et al., 2017; Sharangdhar et al., 2017). In this work, our aim was to systematically identify 45 localized RNAs in primary rat hippocampal neurons by sequencing total RNA (rRNA depleted) as 46 opposed to polyadenylated (polyA+) RNA, with a particular focus on the repertoire of projection-47 localized introns (both retained and excised). Our analyses identify hundreds of transcripts with 48 retained introns. Unexpectedly, we also found a set of free circular introns localized to distal 49 neuronal projections. 50 Results 51 Experimental Design and Validation 52 To physically separate cellular projections from cell bodies, we cultured dissociated primary rat 53 hippocampal cells on membranes with 1 µm diameter pores (Poon et al., 2006). These cultures 54 are a mixture of neuronal and glial cells; we add a DNA replication inhibitor to block cell division 55 and prevent dividing glia from overgrowing post-mitotic neurons. We refer to the projections 56 as "neuro-glial" projections because both neuronal (Map2-immunopositive) and non-neuronal 57 (Gfap/Vimentin-immunopositive) projections extend through the pores and continue growing on 58 the underside of the membrane, whereas cell bodies and nuclei are restricted to the top surface 59 (Figure 1A and Figure 1-Figure Supplement 1). Lysates prepared by scraping the underside are 60 highly enriched for projections ("projection" samples), while lysates prepared from the top surface 61 comprise whole cells with nuclei and projections ("whole cell" samples). 62 To capture both adenylated (polyA+) and non-adenylated (polyA-) long RNAs in our lysates, 63 we prepared rRNA-depleted total RNAseq libraries (mean insert size~200 nt) from five biological 64 replicates (10 samples total). The RNAseq libraries were subjected to paired-end sequencing 65 (100-125 nt reads) on the Illumina platform to obtain 30-80 million mate pairs per sample. 66 We generated addit...