Wax worm saliva and the enzymes therein are the key to polyethylene degradation by Galleria mellonella
Description
Plastic degradation by biological systems with re-utilization of the by-products can be the future solution to the global threat of plastic waste accumulation. We report that the saliva of Galleria mellonella larvae (wax worms) is capable of oxidizing and depolymerizing polyethylene (PE), one of the most produced and sturdy polyolefin-derived plastics. This effect is achieved after a few hours’ exposure at room temperature and physiological conditions (neutral pH). The wax worm saliva can indeed overcome the bottleneck step in PE biodegradation, that is the initial oxidation step. Within the saliva, we identified two enzymes that can reproduce the same effect. This is the first report of enzymes with this capability, opening up the way to new ground-breaking solutions for plastic waste management through bio-recycling/up-cycling. This data contains the gene models generated for the reference G. mellonella genome ASM258982v1 in GTF and GFF3 formats
Files
Steps to reproduce
A specific pipeline was developed to combine the information from RNA-seq (Bertocchini, NCBI: PRJNA822887) with ab initio predictors in order to obtain the most accurate annotation. Briefly, the RNA-seq data was mapped on the reference genome using STAR (version 2.5 0c) in local mode and used to perform a reference guided transcriptome assembly with Trinity (v2.11.0). The obtained transcripts and the mapping files were used as input for the Braker2 pipeline to combine AUGUSTUS ab initio annotation with the transcriptome assembly to obtain the annotation in GFF format, together with transcript and protein sequences. Proteins were used as input for the PANNZER2 pipeline to obtain descriptions , Gene Ontology and KEGG annotations. About 32000 genes could be annotated in the Galleria genome. The corresponding proteins were analyzed to assess their completeness performing a BLASTP alignment against the UniRef90 database and calculating the percentage of alignment. A similarity search against the UniRef90 (Nov2018) database showed that about 50% of the predicted proteins covered 100% of the corresponding hits (i.e. full length) and that about 80% of the predicted proteins covered at least 50% of the corresponding hits.