Rapid bacterial and fungal successional dynamics in first year after Chaparral wildfire

Abstract

The rise in wildfire frequency in the western United States has increased interest in secondary succession. However, despite the role of soil microbial communities in plant regeneration and establishment, microbial secondary succession is poorly understood owing to a lack of measurements immediately post-fire and at high temporal resolution. To fill this knowledge gap, we collected soils at 2 and 3 weeks and 1, 2, 3, 4, 6, 9, and 12 months after a chaparral wildfire in Southern California. We assessed bacterial and fungal biomass with qPCR of 16S and 18S and richness and composition with Illumina MiSeq sequencing of the 16S and ITS2 amplicons. We found that fire severely reduced bacterial biomass by 47% and richness by 46%, but the impacts were stronger for fungi, with biomass decreasing by 86% and richness by 68%. These declines persisted for the entire post-fire year, but bacterial biomass and richness oscillated in response to precipitation, whereas fungal biomass and richness did not. Fungi and bacteria experienced rapid succession, with 5-6 compositional turnover periods. As with plants, fast-growing surviving microbes drove successional dynamics. For bacteria, succession was driven by the phyla Firmicutes and Proteobacteria, with the Proteobacteria Massilia dominating all successional time points, and the Firmicutes (Domibacillus and Paenibacillus) dominating early- to mid-successional stages (1-4.5 months), while the Proteobacteria Noviherbaspirillum dominated late successional stages (4.5-1 year). For fungi, succession was driven by the phyla Ascomycota, but ectomycorrhizal basidiomycetes, and the heat-resistant yeast, Geminibasidium were present in the early successional stages (1 month). However, pyrophilous filamentous Ascomycetes Pyronema, Penicillium, and Aspergillus, dominated all post-fire time points. While wildfires vastly decrease bacterial and fungal biomass and richness, similar to plants, pyrophilous bacteria and fungi increase in abundance and experience rapid succession and compositional turnover in the first post-fire year, with potential implications for post-fire chaparral regeneration

Not All Is Lost: Resilience of Microbiome Samples to Freezer Failures and Long-term Storage

“…We selected 20 soil samples from which DNA had previously been extracted, sequenced, and analyzed ("original" samples) to test against soil samples that had been stored in the -80ºC freezer that thawed ("thawed" samples) vs resequencing extracted DNA that had been stored in an uncompromised -20ºC freezer for roughly 2 years ("frozen" samples). We selected 20 previously analyzed soil samples that had been collected from burned vs unburned plots at 25, 34, 67, 95, and 286 days after the 2018 Southern California Holy Fire, selecting 2 burned and 2 unburned samples for each of the 5 timepoints (21).…”

“…In October 2021, we extracted DNA from the 20 soil samples that had experienced the short thaw, using the same methods as the "original" and "frozen" samples with Qiagen DNeasy PowerSoil Kits following the manufacturer's protocol, as was done in the original study (21). We then used the same methods as had been used for the "original" samples to PCR amplify and sequence with Illumina MiSeq the 16S and ITS2 regions of the "thawed" and "frozen" DNA samples (21).…”

“…In October 2021, we extracted DNA from the 20 soil samples that had experienced the short thaw, using the same methods as the "original" and "frozen" samples with Qiagen DNeasy PowerSoil Kits following the manufacturer's protocol, as was done in the original study (21). We then used the same methods as had been used for the "original" samples to PCR amplify and sequence with Illumina MiSeq the 16S and ITS2 regions of the "thawed" and "frozen" DNA samples (21). We performed bioinformatics in QIIME2 ( 22) on all 60 samples ("original" vs "frozen" vs "thawed") and performed downstream analyses on bacterial ASV tables rarefied to 6,493 sequences/sample and fungal (5,786 sequences/sample) as previously published (21).…”

“…We then used the same methods as had been used for the "original" samples to PCR amplify and sequence with Illumina MiSeq the 16S and ITS2 regions of the "thawed" and "frozen" DNA samples (21). We performed bioinformatics in QIIME2 ( 22) on all 60 samples ("original" vs "frozen" vs "thawed") and performed downstream analyses on bacterial ASV tables rarefied to 6,493 sequences/sample and fungal (5,786 sequences/sample) as previously published (21).…”

“…We tested how the major taxonomic groups assessed in burned vs unburned plots varied in "original" vs "frozen" vs "thawed" with relative abundance plots created in ggplot2 (26) as had been done on the "original" study (21).…”