Molecular prospecting for cryptic species of the Hypholoma fasciculare complex: toward the effective and practical delimitation of cryptic macrofungal species

Hirotoshi Sato,Noriaki Murakami,Ryoma Ohta

doi:10.1038/s41598-020-70166-z

Hirotoshi Sato, Noriaki Murakami + Show 1 more

Open Access

https://doi.org/10.1038/s41598-020-70166-z

Copy DOI

Abstract

Many macrofungal cryptic species remain unidentified. A possible solution is to increase the number of loci analyzed and use rigorous statistics for macrofungal species delimitation. To validate this assumption, cryptic species of the Hypholoma fasciculare complex, a group of common wood-decomposing fungi, were attempted to be delineated. Massively parallel sequencing of mitochondrial ribosomal RNA (mt_rRNA), nuclear ribosomal internal transcribed spacer (ITS) region, and 24 single-copy genes were performed for 96 specimens collected in Japan. Then, the species boundaries were inferred using comparative gene genealogies (mt_rRNA vs. ITS), Bayesian Poisson tree process (bPTP) model for the phylogeny of concatenated nuclear sequences, and analysis of molecular variance (AMOVA) for single nucleotide polymorphisms. In both the mt_rRNA and ITS phylogenies, the H. fasciculare complex was not divided into well-supported clades. Nevertheless, based on the bPTP, two mitochondrial haplotypes were inferred to represent distinct species (H. fasciculare and H. subviride). The results of AMOVA also indicated that the differentiation of nuclear loci can be explained mostly by differences between haplotype. These results suggest that it is necessary to increase the number of target loci to 20 or more and use both phylogeny-based and population genetics-based statistics for the accurate delimitation of macrofungal species.

Highlights

Single molecular marker is not necessarily sufficient for accurate species d elimitation[12,13]
Over the last three decades, comparative gene genealogies have been used to delimitate fungal species, including pathogenic fungi[6,15,16,17,18], lichenized fungi[19,20,21], and macrofungi[7,22,23]. This method clearly differs from single-locus phylogeny because distinct linkage disequilibrium among different loci indicates the restriction of gene flow, which is an important indication of reproductive isolation among sympatrically distributed fungal t axa[7,17]
The Poisson tree processes (PTP) model and its Bayesian implementation are similar to the GYMC model[29], and are often integrated with the evolutionary placement algorithm (EPA), in which short reads are placed into a given reference tree obtained from full-length sequences to determine the evolutionary origin of r eads[30]

Summary

Introduction

Single molecular marker is not necessarily sufficient for accurate species d elimitation[12,13]. Analysis of population genetics using single nucleotide polymorphisms (SNPs) is a powerful method of detecting reproductively isolated species This method has some advantages over phylogeny-based species delimitation by testing distinct linkage disequilibrium among different loci. A fundamental question addressed in the present study is how increasing the number of loci analyzed (i.e., using 20 or more loci) and using two different statistical frameworks (i.e., both phylogeny-based and population genetics-based species delimitation) could improve the accuracy of the delimitation of closely related cryptic macrofungal species. To explore this question, species boundaries were examined in the species complex of Hypholoma fasciculare (Strophariaceae, Agaricales). The empirical study of the H. fasciculare complex should show the importance of increasing number of loci analyzed and using rigorous statistical frameworks for the accurate detection of cryptic macrofungal species

Methods

Results

Discussion

Conclusion