Evaluating DNA Barcoding for Species Identification and Discovery in European Gracillariid Moths

Carlos Lopez-Vaamonde,Kyung Min Lee,Stanislav Gomboc,Erik J Van Nieukerken,Jean-François Landry,Camiel Doorenweerd,Zdenek Laštůvka,Paolo Triberti,H Charles J Godfray,Christian Wieser,Ales Laštůvka,David C Lees,Marko Mutanen,Peter Huemer,Antoine Guiguet,Andreas H Segerer,Rodolphe Rougerie,Natalia Kirichenko,Alain Cama

doi:10.3389/fevo.2021.626752

Abstract

Gracillariidae is the most species-rich leaf-mining moth family with over 2,000 described species worldwide. In Europe, there are 263 valid named species recognized, many of which are difficult to identify using morphology only. Here we explore the use of DNA barcodes as a tool for identification and species discovery in European gracillariids. We present a barcode library including 6,791 COI sequences representing 242 of the 263 (92%) resident species. Our results indicate high congruence between morphology and barcodes with 91.3% (221/242) of European species forming monophyletic clades that can be identified accurately using barcodes alone. The remaining 8.7% represent cases of non-monophyly making their identification uncertain using barcodes. Species discrimination based on the Barcode Index Number system (BIN) was successful for 93% of species with 7% of species sharing BINs. We discovered as many as 21 undescribed candidate species, of which six were confirmed from an integrative approach; the other 15 require additional material and study to confirm preliminary evidence. Most of these new candidate species are found in mountainous regions of Mediterranean countries, the South-Eastern Alps and the Balkans, with nine candidate species found only on islands. In addition, 13 species were classified as deep conspecific lineages, comprising a total of 27 BINs with no intraspecific morphological differences found, and no known ecological differentiation. Double-digest restriction-site associated DNA sequencing (ddRAD) analysis showed strong mitonuclear discrepancy in four out of five species studied. This discordance is not explained by Wolbachia-mediated genetic sweeps. Finally, 26 species were classified as “unassessed species splits” containing 71 BINs and some involving geographical isolation or ecological specialization that will require further study to test whether they represent new cryptic species.

Highlights

Alarming projections of biodiversity loss indicate that we have entered the world’s sixth mass extinction (Raven and Miller, 2020)
We looked at whether there was any correspondence between Barcode Index Numbers (BINs) and RADseq clades in two species initially classified as Unassessed species splits” (USS) [Phyllocnistis saligna (Zeller) and Phyllonorycter salictella (Zeller)] and three as Deep conspecific lineages” (DCL) [Phyllonorycter roboris (Zeller), Ph. maestingella (Müller) and Ph. salicicolella (Sircom)]
We generated a total of 6,791 COI sequences representing 242 of the 263 species (92%) of Gracillariidae known to occur in Europe (Supplementary Table S2 and Supplementary Figure S4) with an average of 25.8 specimens per species [minimum one specimen (in 12 (5%) of the 242 named species), maximum 1,932 specimens] (Supplementary Table S3)

Summary

Introduction

Alarming projections of biodiversity loss indicate that we have entered the world’s sixth mass extinction (Raven and Miller, 2020). DNA barcoding, a tool for species identification based on the use of a single standard DNA marker (a fragment of the COI mtDNA gene; Hebert et al, 2003), has contributed to the discovery of many new species, including among them the most closely studied and best documented groups of organisms (Hrbek et al, 2014). Some of those new species are cryptic, rare and of conservation value; assessing the level of cryptic diversity and identifying their biodiversity hotspots should be a conservation priority (Bickford et al, 2007)

Objectives

Methods

Results

Discussion

Conclusion