Skip to main content
Dryad logo

Light availability and light demand of plants shape the arbuscular fungal communities in their roots


Neuenkamp, Lena et al. (2021), Light availability and light demand of plants shape the arbuscular fungal communities in their roots, Dryad, Dataset,


Plants involved in the arbuscular mycorrhizal (AM) symbiosis trade photosynthetically derived carbon for fungal-provided soil nutrients. However, little is known about how plant light demand and ambient light conditions influence root-associating AM fungal communities.

We conducted a manipulative field experiment to test whether plants’ shade tolerance influences their root AM fungal communities in open and shaded grassland sites. We found similar light-dependent shifts shifts in AM fungal community structure for experimental bait plant roots and the surrounding soil. Yet, deviation from the sorrounding soil towards lower AM fungal beta-diversity in the roots of shade-intolerant plants in shade suggested preferential carbon allocation to specific AM fungi in conditions where plant-assimilated carbon available to fungi was limited. We conclude that favourable environmental conditions widen the plant biotic niche, as demonstrated here with optimal light availability reducing plants' selectivity for specific AM fungi, and promote compatibility with a larger number of AM fungal taxa.

The deposited dataset contains the raw fungal sequence table for each sample and information on the characteristics of the sample (light treatment, sample type, sampling point in time etc).


Lines from the method section of the related paper (accepted in Ecology Letters)

Experimental site: 

We conducted a field experiment in a species rich wooded meadow situated on the west coast of Estonia (Laelatu, 58.5835001 N and 23.5677568 E) and covering an area of 153 ha. The meadow has been under continuous, extensive management for centuries; notably mowing for hay and cutting shrubs and trees for fuel. The wooded meadow is a mosaic of open, well illuminated sites and shaded sites containing shrubs (Corylus avellana L.) and some trees such as Quercus robur L. and Betula pendula Roth (Semchenko & Zobel 2007). Shaded sites are characterized by lower plant diversity and cover of herbaceous species compared with open sites.

Sampling design: 

We chose 40 perennial plant species native to the wooded meadow and with contrasting plant shade tolerance as experimental species. Plant species shade-tolerance was estimated based on their Ellenberg indicator value for light (Ellenberg et al. 1992); species with a light indicator value ≥ 5 were categorised as shade-intolerant, and species with lower light indicator values were categorised as shade-tolerant.

We grouped the seeds into 10 unique seed mixtures each containing seeds of two shade-intolerant and two shade-tolerant species, with some exceptions due to differences in seed availability. In October 2011, we established the experiment at four sites (10 x 10 m) within the continuously managed part of Laelatu wooded meadow: two sites in open and two sites in shaded parts. We created a grid of 1 m2 plots in each of the four experimental sites, and each seed mixture was assigned to five randomly selected plots in a site. In each of the plots, 50 seeds per seed mixture were sown into a metal ring of 15cm diameter (experimental unit) positioned in the centre of the plot. Altogether, we created 200 experimental units (10 seed mixtures x 5 replicates x 4 sites).

Data collection:

One year after seeding we conducted the first sampling campaign (October 2012), collecting 3-5 seedlings of all bait plant species that germinated within the first year (10 out of 40 species; Table 1) from each experimental unit. Seedlings from each experimental unit were cleaned from soil and dried for 24 h at 50 ˚C; the roots of each plant species were separated from the shoots and stored dry for further molecular analysis of root colonizing AM fungi. We repeated the same procedure in July 2013 for the bait plant species that had not germinated in the first year (7 out of 40 species; Table 1). For the second census, we sampled all the previously sampled plant species – i.e., all those that had germinated in October 2012 or July 2013 – again in June 2014. While the interval between censues varied, we interpret sampling census as an indicator of AM fungal community maturity, with young, potentially dynamic composition in the first census, and mature, potentially stable composition in the second census. The final set of species inlcuded into the analyses (and those present in the uploaded dataset) comprised eight shade-tolerant and nine shade-intolerant species. 

During selection of experimental plant species, we took care to limit phylogenetic differences between shade-tolerant and -intolerant species, thus froming pairs of two species from the same genus or familiy with contrasting shade tolerance (where possible). This balance remained evident also in the final set of data.

After the final sampling campaign (October 2014), we collected 10 cm3 soil cores from five random locations per site (n = 20 , hereafter `background samples´) to characterise the AM fungal communities in plant roots (across all plant species present in the community) and in soil surrounding the experimental units. Background samples provided local context to AM fungal community shifts observed in bait root samples. From background samples, we separated plant roots from soil, dried them for 24 h at 50 ˚C and stored them dry for further molecular analysis. To homogenize root composition within samples, we crushed the roots in each sample with liquid nitrogen and thoroughly mixed them as described by Garcia de Leon et al. (2016). We collected 10g of soil from each background sample, dried it with silica gel and stored it airtight at room temperature for further molecular analysis.

Molecular Analyses and Bioinformatics:

AM fungal communities in bait plant roots were analysed for each bait plant species and experimental unit separately. We extracted DNA from 70 mg (dry weight) of roots from bait plants (1-3 individuals per experimental unit) and mixed root samples using the Power Soil® DNA Isolation Kit (MO BIO Laboratories, Inc., Carlsbad, CA, USA) following Saks et al. (2014). For soil samples, we extracted DNA from 5 g of dried soil using the Power Max Soil® DNA Isolation Kit (MO BIO Laboratories, Inc., Carlsbad, CA, USA) according to Gazol et al. (2016). We amplified Glomeromycotina SSU rRNA gene sequences with the primers NS31 and AML2 (Simon et al. 1992; Lee et al. 2008) and identified them using 454-sequencing on a Genome Sequencer FLX System with Titanium Series reagents (Roche Applied Science, Mannheim, Germany) at GATC Biotech (Konstanz, Germany) as described in Davison et al. (2012) and Öpik et al. (2013).

We assigned taxonomic information to quality-filtered reads using the MaarjAM database (status May 2020) that classifies the central part of published Glomeromycotina SSU rRNA gene sequences into phylogenetically delimited sequence clusters – virtual taxa (VT, cf. Öpik et al. 2010, 2014). Taxonomic identification and quality filtering followed the same approach as described in Neuenkamp et al. 2018 (see Appendix S3 for a more detailed description). The final data matrix consisted of 198 samples, 265 375 reads and 142 VT for AM fungi in bait plants root samples, and 36 samples, 56 755 reads and 86 VT for AM fungi in background samples (mixed roots, soil) (Appendix S4). We excluded singletons (VT represented by a single read in the whole dataset; 22 VT in total) from the dataset. For bait plants root samples, the minimum number of reads per sample was 9 and the median number of reads per sample was 1055; for background samples, the respective numbers were 195 and 1969. A set of representative sequence reads was deposited in the EMBL nucleotide collection (accession numbers: LR890754-LR897751).  

Usage Notes

The dataset of fungal sequences is uploaded and contains a "Glossary" sheet where terms if necessary are explained. 

The dataset is organised so that each row is a sample and then sample information is provided in colums. 

Read counts on arbuscular mycorrhizal (AM) fungi are presented as quality-filerted but raw reads. So no rarefaction has been applied to them, but is encouraged before analysis.

Reads were assigned to virtual fungal taxa (VT) based on their similarity to the taxonomy established in the MaarjAM database (see methods above; 


University of Tartu, Award: PLTOM20903

Estonian Research Council, Award: SLTOM20001T

European Regional Development Fund, Award: Centre of Excellence EcolChange

University of Tartu, Award: PLTOM20903

Estonian Research Council, Award: SLTOM20001T