Evaluation of various RNA-seq approaches for identification of gene outrons in the flatworm Opisthorchis felineus.

D E Maslov,N I Ershov,N P Bondar

doi:10.18699/vj20.688

Abstract

The parasitic flatworm Opisthorchis felineus is one of the causative agents of opisthorchiasis in humans. Recently, we assembled the O. felineus genome, but the correct genome annotation by means of standard methods was hampered by the presence of spliced leader trans-splicing (SLTS). As a result of SLTS, the original 5’-end (outron) of the transcripts is replaced by a short spliced leader sequence donated from a specialized SL RNA. SLTS is involved in the RNA processing of more than half of O. felineus genes, making it hard to determine the structure of outrons and bona fide transcription start sites of the corresponding genes and operons, being based solely on mRNA-seq data. In the current study, we tested various experimental approaches for identifying the sequences of outrons in O. felineus using massive parallel sequencing. Two of them were developed by us for targeted sequencing of already processed branched outrons. One was based on sequence-specific reverse transcription from the SL intron toward the 5’-end of the Y-branched outron. The other used outron hybridization with an immobilized single-stranded DNA probe complementary to the SL intron. Additionally, two approaches to the sequencing of rRNA-depleted total RNA were used, allowing the identification of a wider range of transcripts compared to mRNAseq. One is based on the enzymatic elimination of overrepresented cDNAs, the other utilizes exonucleolytic degradation of uncapped RNA by Terminator enzyme. By using the outron-targeting methods, we were not able to obtain the enrichment of RNA preparations by processed outrons, which is most likely indicative of a rapid turnover of these trans-splicing intermediate products. Of the two rRNA depletion methods, a method based on the enzymatic normalization of cDNA (Zymo-Seq RiboFree) showed high efficiency. Compared to mRNA-seq, it provides an approximately twofold increase in the fraction of reads originating from outrons and introns. The results suggest that unprocessed nascent transcripts are the main source of outron sequences in the RNA pool of O. felineus.

Highlights

Opisthorchis felineus is a representative of parasitic flat worms (Trematoda: Opisthorchiidae), which has a complex life cycle with two intermediate hosts and the mammalian definitive host, including humans (Beer, 2005)
To identify O. felineus outrons, we evaluated two candidate approaches based on targeted enrich ment for Y-outrons, in one case, by sequence-specific reverse transcription primed from the spliced genes (SL) intron towards the 5′-end of the outron, in the other – by hybridization of Y-outrons with an immobilized single-stranded DNA probe complementary to the SL-intron
Description of the approaches At the moment, we have not been able to find any reference in the literature to experimental methods for the direct massive identification of Y-outrons in the transcriptome

Summary

Introduction

Opisthorchis felineus is a representative of parasitic flat worms (Trematoda: Opisthorchiidae), which has a complex life cycle with two intermediate hosts and the mammalian definitive host, including humans (Beer, 2005). To describe gene outrons, researchers combine information on transcrip tion initiation sites obtained by such methods as GRO-seq, GRO-cap, ChIP-seq to RNA polymerase II, with data on SLTS sites detected by mRNA-seq (Chen et al, 2013; Kruesi et al, 2013). In this way, the region of the genome from the start of transcription to the 3′-site of the SLTS is accepted as an outron. On non-model organisms, including trematodes, RNA-seq “SL Trapping” methods have been successfully used for direct massive detection of trans-spliced mRNAs, based on selec tion for the universal 5′-sequence of processed transcripts, the SL exon (Nilsson et al, 2010; Boroni et al, 2018). The proportion of sequences containing outrons in these libraries should certainly be higher than in standard polyA-mRNA-seq libraries

Materials and methods

Results and discussion

Conclusion