Long-Term Drought and Warming Alter Soil Bacterial and Fungal Communities in an Upland Heathland

Abstract

The response of soil microbial communities to a changing climate will impact global biogeochemical cycles, potentially leading to positive and negative feedbacks. However, our understanding of how soil microbial communities respond to climate change and the implications of these changes for future soil function is limited. Here, we assess the response of soil bacterial and fungal communities to long-term experimental climate change in a heathland organo-mineral soil. We analysed microbial communities using Illumina sequencing of the 16S rRNA gene and ITS2 region at two depths, from plots undergoing 4 and 18 years of in situ summer drought or warming. We also assessed the colonisation of Calluna vulgaris roots by ericoid and dark septate endophytic (DSE) fungi using microscopy after 16 years of climate treatment. We found significant changes in both the bacterial and fungal communities in response to drought and warming, likely mediated by changes in soil pH and electrical conductivity. Changes in the microbial communities were more pronounced after a longer period of climate manipulation. Additionally, the subsoil communities of the long-term warmed plots became similar to the topsoil. Ericoid mycorrhizal colonisation decreased with depth while DSEs increased; however, these trends with depth were removed by warming. We largely ascribe the observed changes in microbial communities to shifts in plant cover and subsequent feedback on soil physicochemical properties, especially pH. Our results demonstrate the importance of considering changes in soil microbial responses to climate change across different soil depths and after extended periods of time.

Altered activities of extracellular soil enzymes by the interacting global environmental changes

“…One of the most frequent phenomena of indirect driver induction is the reduction of soil water availability caused by warming (Steinauer et al, 2015), so that it is not always easy to discern between indirect or combined effects. Temperature and soil moisture can affect enzymatic activities through changes in microbial composition (de Vries et al, 2018; Ochoa‐Hueso et al, 2018; Seaton et al, 2021), amount and quality of the organic matter (Guoju et al, 2012; Siebert et al, 2019), and/or pH (Jiao et al, 2016; Slessarev et al, 2016), while in turn pH can affect EEAs through changes of quality and quantity of organic matter and alter the soil capacity to store and supply nutrients (Slessarev et al, 2016). N deposition can cause, as seen above, soil acidification (Wang et al, 2013), and together with nutrient availability can cause changes in the structure of the microbial population (Jiang et al, 2009).…”

Evaluation of metagenomic, 16S rRNA gene and ultra-plexed PCR-based sequencing approaches for profiling antimicrobial resistance gene and bacterial taxonomic composition of polymicrobial samples

“…AmpliSeq was conducted using the community-designed Illumina for AmpliSeq AMR Research Panel and the Library PLUS kit with manufacturer guidelines; libraries were pooled and sequenced on the Illumina MiniSeq (150 bp PE reads) with replicate libraries prepared for the mock sample. 16S rRNA gene sequencing utilised 515F–806R primers for library preparation and libraries were sequenced on the Illumina MiSeq (250 bp PE reads) as previously described 19 .…”

High-resolution characterisation of short-term temporal variability in the taxonomic and resistome composition of wastewater influent

“…Taxonomic processing of 16S rRNA gene sequence data was conducted using DADA2 v1.16 as previously described (Seaton et al, 2021). Briefly, Illumina demultiplexed 16S rRNA gene sequences underwent DADA2 workflows to quality filter, merge, denoise and assign taxonomy using SILVA v138.1 as the training database.…”

“…All samples (n=75) underwent 16S rRNA gene sequencing using 515F-806R primers on the MiSeq (Illumina, San Diego, USA) as previously described (Seaton et al, 2021) generating 250bp paired-end reads. Two grab samples (2/72; timepoints 4:00 and 14:00) were excluded from analyses due to low sequencing quality.…”