Soil microbial diversity across Europe

Factors driving microbial community composition and diversity are well established but the relationship with microbial functioning is poorly understood, especially at large scales. We analysed microbial biodiversity metrics and distribution of potential functional groups along a gradient of increasing land-use perturbation, detecting over 79,000 bacterial and 25,000 fungal OTUs in 715 sites across 24 European countries. We found the lowest bacterial and fungal diversity in less-disturbed environments (woodlands) compared to grasslands and highly-disturbed environments (croplands). Highly-disturbed environments contain significantly more bacterial chemoheterotrophs, harbour a higher proportion of fungal plant pathogens and saprotrophs, and have less beneficial fungal plant symbionts compared to woodlands and extensively-managed grasslands. Spatial patterns of microbial communities and predicted functions are best explained when interactions among the major determinants (vegetation cover, climate, soil properties) are considered. We propose guidelines for environmental policy actions and argue that taxonomical and functional diversity should be considered simultaneously for monitoring purposes.

Data

We analyse DNA sequences of two groups of microorganisms (bacteria and fungi) from 715 soil samples collected from 23 countries of the European Union and the United Kingdom (EU + UK) as part of LUCAS2018. We assess continental-scale effects of vegetation cover (and associated land-use), soil properties, climate and their two-way interactions on soil microbial communities and potential functional groups (i.e. inferred based on taxonomy). Soil sampling is performed within the framework of the soil module of the Land Use/Cover Area frame Survey (LUCAS). Sampling locations cover both semi-natural and highly-managed environments, including six vegetation cover types subjected to an increasing land-use perturbation gradient: from coniferous and broadleaved woodlands, to extensively- and intensively-managed grasslands, and permanent and non-permanent croplands. Vegetation cover is combined with a broad set of 9 soil physico-chemical properties and 6 climatic variables. We hypothesise that (i) land-use perturbation affects soil microbial diversity and community structure and potential functional groups, as observed for the above-ground biodiversity6; (ii) bacterial and fungal guilds and potential functional groups are shaped by different forces30,31; (iii) interactions between vegetation cover, climate and soil properties are more important and informative than single-effects in driving the assembly of bacterial and fungal soil communities and potential functional groups.

Sampling points distribution coloured by vegetation cover type across biogeographical regions. The number of sites is indicated between brackets. b Vegetation cover types ordered along a gradient of increasing land-use perturbation.

Methods

Soil samples were analysed for their bacterial and fungal biodiversity through a DNA metabarcoding approach. Primer sets for barcode amplification were 515F (GTGYCAGCMGCCGCGGTAA) and 926R (GGCCGYCAATTYMTTTRAGTTT) for the bacterial 16S region and ITS9mun (GTACACACCGCCCGTCG) and ITS4ngsUni (CGCCTSCSCTTANTDATATGC) for the fungal ITS region. Sequencing was performed by Illumina MiSeq platform with 2 × 300 paired-end mode for bacterial data and PacBio Sequel II platform for fungal data. Protocols for DNA extraction and amplification are described hereafter.

DNA was extracted with the Qiagen DNeasy PowerSoil HTP 96 Kit Q12955-4. Three 0.2 g aliquots per sample were extracted. The three subsamples were pooled after extraction. A negative control and positive control were used during extraction to locate any external contamination and cross-contamination. Quality check and quantification of DNA were performed with Qubit™ 1X dsDNA HS Assay Kit using Qubit 3 fluorometer (Invitrogen).

Land-use perturbation effects

The entire dataset included 79,593 bacterial zero-radius operational taxonomic units (zOTUs) and 25,962 fungal OTUs. Bacterial observed zOTU richness and Shannon diversity index were lowest in woodlands and significantly higher in croplands and grasslands. Similarly, fungal richness and diversity were lower in woodlands compared to grasslands and croplands. Vegetation cover had a significant impact on bacterial and fungal community structure (β-diversity). The highest difference in community structure was found between microbial communities in croplands and woodlands, as supported by pairwise multiple comparison that displayed the highest determination coefficients. The distribution of inferred bacterial and fungal functional groups also differed among vegetation cover types. Bacterial chemoheterotrophs (12,786 zOTUs, i.e. 16.1% of the total number of bacterial zOTUs) dominated in croplands and intensive grasslands

Data available: The dataset includes bacterial 16S data and fungal ITS raw DNA sequences for 885 samples collected as part of LUCAS 2018 Soil survey (Biodiversity module).

Reference: Labouyrie, M., Ballabio, C., Romero, F., Panagos, P., Jones, A.,  Schmid, M., Mikryukov, V., Dulya, O., Tedersoo, L., Bahram, M., Lugato, E., van der Heijden, M.G.A., Orgiazzi, A. 2023. Patterns in soil microbial diversity across Europe. Nature Communications. 14, Article number: 3311,  doi: 10.1038/s41467-023-37937-4