Taxonomic assignment of arbuscular mycorrhizal fungi in an 18S metagenomic dataset: a case study with saltcedar (Tamarix aphylla)

• Published in: Mycorrhiza Vol. 30; no. 2-3; pp. 243 - 255
• Main Authors: Stefani, Franck, Bencherif, Karima, Sabourin, Stéphanie, Hadj-Sahraoui, Anissa Lounès, Banchini, Claudia, Séguin, Sylvie, Dalpé, Yolande
• Format: Journal Article
• Language: English
• Published: Berlin/Heidelberg Springer Berlin Heidelberg 01.05.2020; Springer Nature B.V; Springer Verlag
• Subjects: Agriculture; Arbuscular mycorrhizas; Biomedical and Life Sciences; Chemical Sciences; Ecology; Forestry; Fungi; Life Sciences; Metagenomics; Microbiology; Mycorrhizae - genetics; Original Article; Plant Sciences; Symbiosis; Tamaricaceae - genetics; Metagenomics; Arbuscular mycorrhizal fungi; Taxonomic assignment
• ISSN: 0940-6360; 1432-1890
• DOI: 10.1007/s00572-020-00946-y

Many studies describing communities of arbuscular mycorrhizal fungi (AMF, Glomeromycota) based on high-throughput sequencing target the V4 variable region of the 18S ribosomal gene. However, an accurate taxonomic assignment of these short 18S sequences is challenging. Here we describe a simple approach based on a phylogenetic analysis using a backbone of reference sequences with taxonomic names updated in MycoBank to improve the taxonomic assignment of amplicon sequence variants (ASVs). As a case study, paired-end sequencing (2 × 250 bp) was carried out to describe the community of AMF associated with Tamarix aphylla from Algerian steppe ecosystems. AMF from root and soil samples were targeted with a nested PCR, using the AMF-discriminating primers pair AML1/AML2 for the first amplification and a new primer pair for the second. The proportion of the sequences assigned to Glomeromycota was 85.9%, representing a total of 87 ASVs. Seven well-defined genera (Claroideoglomus, Dominikia, Funneliformis, Innospora, Microkamienskia, Rhizophagus, Septoglomus) and seven phylogenetic divergent clades of Glomus (27% of the ASVs) were identified with the proposed approach. This taxonomic assignment was in sharp contrast with querying the MaarjAM or GenBank databases. Out-of-date taxonomy led the MaarjAM database to attribute 85% of the ASVs to the genus Glomus and the GenBank database to assign 18% of the ASVs to unclassified taxa. We recommend using the simple workflow presented in this study so that up-to-date taxonomic information is accurately assigned to AMF communities analyzed by high-throughput sequencing.