Metabarcoding for parallel identification of species, sex and diet of obligate scavengers: an application to globally-threatened Gyps vultures

After suffering a massive decline (~99%) in numbers caused by feeding on livestock carcasses containing the nephrotoxic drug diclofenac, critically endangered Gyps vultures now persist in low numbers in the Indian subcontinent, mostly concentrated within or near National Parks. This spatial association might be attributed to availability of wild ungulate carcasses free from toxic veterinary drugs. Hence, quantification of vulture diets is critical to test this hypothesis. We describe a validated “field-to-benchtop-to-desktop” metabarcoding workflow for assessing the species- and sex-specific diet of these obligate scavengers from non-invasively collected faecal samples. Seven metabarcodes were designed to simultaneously identify the vulture species, sex and diet species.  The amplicons generated from multiplex PCRs were indexed and sequenced on an Illumina Miseq platform. We included controls and three replicates per sample to establish a series of non-arbitrary thresholds to filter the sequence data and eliminate cross-contamination, PCR/sequencing errors and false positives. Using this strategy enabled identification of species and sex for all samples. Diet-specific sequences could be identified with high taxonomic resolution for 97% of samples. Out of the seven metabarcodes, just four (one for species identification, one for sexing and two for diet) were sufficient to meet the objectives. This preliminary analysis suggests that domestic livestock is the most frequently consumed diet item across samples from inside and outside protected habitats. Our method provides a rapid and reliable tool for describing large-scale variation in consumption of domestic versus wild species in the diet of these scavengers, paving the way for a better understanding of the role protected areas play in persistence and recovery of the remaining Gyps vultures in the wild. 

Methods
We collected fecal samples from a captive individual (Himalayan Griffon, Gyps himalayensis) held at the vulture conservation breeding centre at Pinjore, Haryana, (March 2018) and from wild vultures at sites located in Pench, Kanha, Sariska, Ranthambore and Mukundra Hills Tiger Reserves in India (June 2018-December 2018). At each field site, we located resident Gyps vulture (White-rumped and Long-billed) roosting and nesting sites and collected faecal droppings from substrates (soil, leaves, rocks) directly under the roost/nest. We swabbed the faecal dropping using a plastic applicatory rayon swab dipped in Longmire lysis buffer (Longmire et al. 1997). The tip of the swab was broken and stored in a 2 ml vial (with 1 ml buffer) at room temperature for up to 30 days, after which it was transported to the laboratory and stored at -20°C until DNA extraction. Using the extracted DNA from these samples, we prepared library using seven metabarcodes (two for vulture species identification, two for sexing vultures and three for diet identification). We sequenced the library on the Miseq platform using 2x150 bp paired-end sequencing.  The resulting sequence data was processed using the OBITOOLS bioinformatics pipeline and subsequently filterered using low frequency noise filters based on the experimnental controls included.