The specific work carried out to achieve the objectives of the project is summarized below:
Work Package 1: Sample collection
A field expedition was carried out in which a total of 28 replicated samples of the species Lineus longissimus, Lineus lacteus, Cerebratulus marginatus, Paranemertes peregrina and Amphiporus formidabilis were collected. Samples of the hoplonemerteans Antarctonemertes valida and Antarctonemertes riesgoae were already available. These samples were preserved in RNAlater for RNA sequencing, flash frozen for proteomics and ST, and in glutaraldehyde and ethanol for TEM and SEM respectively.
Work package 2: Proteogenomics
For all species, two tissues (i.e. proboscis and posterior end of body) from three individuals per species (biological replicates) have been sequenced to generate RNAseq data for Differential Gene Expression (DGE) analyses. The fellow generated cDNA libraries for each replicate, that were sequenced with Illumina and assembled with Trinity v2.4.0 into a reference transcriptome. The individual libraries were mapped to the reference transcriptome to calculate expression levels and perform, DGE analyses with edgeR. In addition, mass spectrometry-based proteomic data has been generated for L. longissimus and A. valida, producing a catalogue of ca. 100 candidate toxins validated through proteotranscriptomics.
Work package 3: Spatial Omics
The heteronemerteans L. longissimus and C. marginatus, and the hoplonemertean A. valida have been analyzed following a spatial omics approach, including ST and spatial proteomics. MALDI-IMS color maps have been created for tissue sections including the proboscis of L. longissimus and A. valida, revealing the spatial distribution of venom peptides. The fellow did a research stay with Dr Stefania Giacomello at SciLifeLab (Stockholm, Sweden) where she optimized ST for ribbon worms. She analysed 3 replicates of L. longissimus and C. marginatus obtaining 3.5 and 6 million UMIs per replicate, respectively. This data was used to resolve mRNA spatial distribution and identify putative novel ribbon worm toxins, including a novel paralytic predatory toxin.
Work package 4: Morphology
Samples of L. longissimus proboscis have been imaged with TEM to investigate the structure of the venom secreting system. A complete proboscis, and anterior and posterior sections of a proboscis placed in ethanol to induce venom discharge, were fixed with glutaraldehyde and osmium tetroxide, dehydrated with acetone and embedded in resin for ultrathin sectioning and TEM imaging at MNCN-CSIC. Samples of L. longissimus, C. marginatus, and A. valida have also been processed.
Work package 5: Evolution
Homologs of the putative toxins identified have been searched across Nemertea, and molecular evolution analyses have been completed. The putative toxins identified in A. valida were used to search nemertean transcriptomes, identifying 32 orthogroups with sequences form other nemertean species. Of the 32 orthologous toxin clusters, 9 are common to all species analyzed including representatives of Palaeonemertea, Pilidiophora and Hoplonemertea, 2 are shared by Pilidiophora and Hoplonemertea, and 21 are lineage-specific, including only hoplonemertean sequences. This suggests that venom evolution is divergent across lineages, with a high proportion of lineage-specific toxins tuned to particular predatory habits.
