New DNA metabarcoding workflow detects terrestrial gastropods from ingested DNA

An invasive predator is quietly reshaping soil communities across France, and until now scientists had no reliable molecular way to see what it eats. A newly published study in Metabarcoding and Metagenomics describes a complete DNA metabarcoding workflow for detecting terrestrial gastropods from ingested DNA, developed and validated specifically for the exotic flatworm Obama nungara, a South American land planarian now considered a potential threat to soil biodiversity across much of Europe. The work delivers a purpose-built primer pair, a rigorously curated reference database, and a calibrated set of filtering parameters that together allow researchers to identify which slugs and snails a predator has consumed, down to species and even genetic lineage level.

Obama nungara is no ordinary invader. Citizen science reports have documented its spread across several European countries, and it is now considered widespread in France, covering roughly 75 percent of the territory. Laboratory experiments have shown that it feeds on earthworms, slugs, snails, and even other flatworms. While earlier research had used DNA metabarcoding to analyze earthworm prey in its digestive contents, terrestrial gastropods, which are widely consumed by the flatworm, had never been subjected to similar molecular analysis. The new study fills that gap with a targeted approach designed to overcome the technical challenges that plague dietary metabarcoding when applied to new taxa.

DNA metabarcoding of predator gut contents faces three well-known obstacles: the specificity of the primers used to amplify target DNA, amplification bias that can overrepresent or underrepresent certain species, and contamination issues that can produce false positives. The research team addressed each of these systematically. They first built a reference database of mitochondrial 16S rRNA gene sequences, combining 96 newly obtained sequences from gastropods collected at French sites invaded by O. nungara with 244 sequences retrieved from NCBI, bringing the total to 340 sequences from 58 species and lineages, roughly 13 percent of the French terrestrial gastropod fauna. The 16S gene was chosen over the more commonly used COI because preliminary tests revealed no suitable regions in COI for designing exclusive gastropod primers, and because COI primer bias is known to cause underrepresentation or non-detection of some taxa even within closely related groups.

Six new primers, ranging from 14 to 20 base pairs, were designed by visually inspecting a 16S alignment that included gastropod, earthworm, and O. nungara sequences. These were complemented by three primers from the literature, resulting in eight candidate primer pairs targeting short DNA fragments of 98 to 217 base pairs. In silico PCR tests against 18,210 mitochondrial reference sequences for Eukaryota and 26,578 bacterial 16S sequences showed that none of the pairs amplified bacterial DNA, with amplification restricted to eukaryotic taxa. Laboratory cross-tests on DNA extracts from three gastropod species, three earthworm species, and eleven O. nungara samples with different feeding histories revealed that the primer pair VRF-VRR amplified only the targeted gastropod taxa and O. nungara samples fed with gastropods, with no co-amplification of earthworm DNA or flatworm DNA from empty or earthworm-fed stomachs.

The VRF-VRR primer pair targets a short 16S fragment of about 162 base pairs, a length well suited to degraded digestive DNA. PCR assays on 49 DNA samples from 38 gastropod species and lineages showed successful amplification in 35 cases. Only Pomatias elegans failed to amplify, and two lineages of Theba pisana showed unspecific results. Phylogenetic reconstruction based on the short fragment revealed that 56 of the 58 species and lineages were monophyletic, grouping into distinct clades or singletons. This is particularly relevant for taxonomically complex genera such as Deroceras, where closely related species like D. invadens and D. panormitanum could be clearly differentiated, and for the slug genus Arion, where two previously undescribed lineages of A. hortensis and five lineages of A. subfuscus were distinguished. Even recently differentiated subspecies of Arion ater could be separated using the short metabarcoding region.

To calibrate the workflow, the researchers deployed an extensive battery of metabarcoding controls. Three mock communities composed of DNA from five gastropod species at equal concentration, and three more composed of 27 species at graduated concentrations, were used to assess contamination levels and set filtering thresholds for Operational Taxonomic Units. Six controlled-feeding controls involved O. nungara individuals fed for three weeks with known gastropod prey, either a single species or a different species each week. Three negative controls from independent PCR amplifications without added DNA checked for contamination during PCR. Finally, three field-collected O. nungara samples were replicated three times each to validate the parameters on real digestive contents.

High-throughput sequencing produced 714,602 paired-end reads across the 40 samples. In the equal-concentration mock communities, all five expected gastropod species were recovered in two of three samples, with more than 99 percent of sequences from these species on average. In the graduated mock communities, 24 of the 27 expected species and lineages were amplified and discriminated. The controlled-feeding controls confirmed that all given prey could be detected after three weeks of experimentation, although the first prey given in the mixed-feeding controls, Hygromia cinctella, was not detected in any sample, consistent with its poor in silico binding score of 254. These results allowed the researchers to set a 0.4 percent OTU filtering threshold, which removed unassigned and artefactual sequences while preserving all correctly identified gastropod detections.

Applying the validated workflow to 16 field samples of O. nungara digestive contents from four sites across mainland France revealed significant dietary diversity. Between one and six prey species were detected per individual, and between five and eight species were recorded per site. Thirty-two of the 43 detections in gut contents, or 74 percent, corresponded to species also observed during field sampling. Deroceras invadens and Arion hortensis L1 were detected in nearly all sites. Notably, at the Cleguer site in Brittany, two species not found during field inventories, Oxyloma elegans and Lauria cylindracea, were detected in gut contents, suggesting that molecular analysis can capture diversity that field surveys miss, particularly for micro-snails that are underrepresented in macrofauna inventories.

The study’s findings suggest that O. nungara exhibits a broad feeding spectrum, though the disparity between field and gut contents at some sites hints at possible selective feeding influenced by prey accessibility or active preference. The researchers deliberately chose a qualitative presence-absence framework over quantitative sequence counts, recognizing that relative read abundance in digestive contents is heavily influenced by time since consumption, species-specific amplification bias, and DNA degradation. The mock community results confirmed this: Cornu aspersum L1 was systematically overrepresented, while Hygromia cinctella and Testacella haliotidea were consistently underrepresented relative to their known DNA proportions.

Beyond O. nungara, the primers and assay developed here have considerable potential for broader application. In silico analyses suggest the VRF-VRR primer pair could theoretically amplify DNA from at least 178 terrestrial gastropod species found in France across 37 families, and 607 species found in Europe across 47 families. The assay could be applied to study the diets of other terrestrial predators, including malacophagous beetles, snail-eating snakes, rodents, and omnivorous birds, or to detect gastropods in environmental DNA samples from soil. This is particularly relevant given the growing presence of other exotic predatory flatworms in regions with high gastropod endemism, such as Platydemus manokwari in the Caribbean and Bipalium vagum in Jamaica. As invasive predators continue to expand globally, tools like this validated metabarcoding workflow will be essential for understanding their ecological impact and informing conservation strategies for native soil communities.

Subject of Research: DNA metabarcoding

Article Title: A DNA metabarcoding workflow for the detection of terrestrial gastropods from ingested DNA

Article References: Ventura, M., Mazuras, N., Noël, S., Dupont, L., & Roy, V. (2025). A DNA metabarcoding workflow for the detection of terrestrial gastropods from ingested DNA. Metabarcoding and Metagenomics, 9, Article e161359. https://doi.org/10.3897/mbmg.9.161359

Image Credits: AI Generated

DOI: 10.3897/mbmg.9.161359

Keywords: DNA metabarcoding, terrestrial gastropods, invasive species, Obama nungara, diet analysis, 16S rRNA, conservation, metabarcoding, workflow, detection, terrestrial, gastropods