Phylogenetic analysis of the distribution of deadly amatoxins among the little brown mushrooms of the genus Galerina

Abstract

Some but not all of the species of ’little brown mushrooms’ in the genus Galerina contain deadly amatoxins at concentrations equaling those in the death cap, Amanita phalloides. However, Galerina’s ~300 species are notoriously difficult to identify by morphology, and the identity of toxin-containing specimens has not been verified with DNA barcode sequencing. This left open the question of which Galerina species contain toxins and which do not. We selected specimens for toxin analysis using a preliminary phylogeny of the fungal DNA barcode region, the ribosomal internal transcribed spacer (ITS) region. Using liquid chromatography/mass spectrometry, we analyzed amatoxins from 70 samples of Galerina and close relatives, collected in western British Columbia, Canada. To put the presence of toxins into a phylogenetic context, we included the 70 samples in maximum likelihood analyses of 438 taxa, using ITS, RNA polymerase II second largest subunit gene (RPB2), and nuclear large subunit ribosomal RNA (LSU) gene sequences. We sequenced barcode DNA from types where possible to aid with applications of names. We detected amatoxins only in the 24 samples of the G. marginata s.l. complex in the Naucoriopsis clade. We delimited 56 putative Galerina species using Automatic Barcode Gap Detection software. Phylogenetic analysis showed moderate to strong support for Galerina infrageneric clades Naucoriopsis, Galerina, Tubariopsis, and Sideroides. Mycenopsis appeared paraphyletic and included Gymnopilus. Amatoxins were not detected in 46 samples from Galerina clades outside of Naucoriopsis or from outgroups. Our data show significant quantities of toxin in all mushrooms tested from the G. marginata s.l. complex. DNA barcoding revealed consistent accuracy in morphology-based identification of specimens to G. marginata s.l. complex. Prompt and careful morphological identification of ingested G. marginata s.l. has the potential to improve patient outcomes by leading to fast and appropriate treatment.