Do different growth rates of trees cause distinct habitat qualities for saproxylic assemblages?

In production forests, a common silvicultural objective is enhancing tree growth rates. The growth rate influences both mechanical and biochemical properties of wood, which may have an impact on dead wood inhabiting (i.e. saproxylic) species. In this study, we tested for the first time whether tree growth rates affect dead-wood associated assemblages in general and the occurrence of red-listed species in particular. We sampled saproxylic beetles (eclector traps) and fungi (DNA metabarcoding of wood samples) in dead trunks of Norway spruce (Picea abies), which had different growth rates within the same hemiboreal forests in Sweden. A high proportion of fungi showed a positive association to increasing tree growth. This resulted in higher fungal richness in fast-grown trees both at the trunk scale and across multiple studied trunks. Such patterns were not observed for saproxylic beetles. However, a set of species (both beetles and fungi) preferred slow-grown wood. Moreover, the total number of red-listed species was highest in slow-grown trunks. We conclude that dead wood from slow-grown trees hosts relatively fewer saproxylic species, but a part of these may be vulnerable to production forestry. It implies that slow-grown trees should be a target in nature conservation. However, where slow-grown trees are absent, for instance in forests managed for a high biomass production, increasing the volumes of dead wood from fast-grown trees may support many species. Supplementary Information The online version contains supplementary material available at 10.1007/s00442-021-05061-z.


The protocol of molecular analyses and bioinformatics for sampling fungi from wood
Dried sawdust samples of 100 mg were homogenised using stainless steel beads (∅ 3.2 mm) in a Retsch MM400 homogenizer (Retsch GmbH, Haan, Germany). DNA was extracted with DNeasy PowerSoil DNA Isolation Kit (Qiagen GmbH, Hilden, Germany) following manufacturer's instructions. We selected rDNA ITS2 region for fungal species identification.
Primers gITS7ngs and ITS4ngsUni (Tedersoo and Lindahl 2016) were tagged with 12-base identifier barcodes as described in Tedersoo et al. (2014). The PCR mixture comprised 1 µl DNA, 1 µl of tagged gITS7ngs primer (20 µM), 0.5 µl of tagged ITS4ngsUni primer (20 µM), 5 µl 5x HOT FIREPol Blend Master Mix (Solis Biodyne, Tartu, Estonia) and 17.5 µl mQ water. PCR was carried out in duplicate replicates in the following thermocycling conditions: an initial 15 min at 95 °C, followed by 30 cycles of 95 °C for 30 s, 52 °C for 30 s, 72 °C for 1 min, and a final cycle of 10 min at 72 °C. PCR products from replicate samples were pooled and their relative quantity was estimated on 1% agarose gel. DNA samples with low or no visible bands were re-amplified using 35 cycles and 2 µl of DNA template. Both negative (mQ water) and positive controls were included in PCR and sequencing runs. PCR products were pooled at approximately equimolar ratios as determined by gel band strength, with the exception of negative control from which 5 µl were added. The library was purified by FavorPrep™ Gel/PCR Purification Kit (Favorgen Biotech Corp., Vienna, Austria), following the manufacturer's instructions. Thereafter, the library was subjected to ligation of Illumina adaptors using the TruSeq DNA PCR-free Library Prep kit (Illumina Inc., San Diego, CA, USA) and Illumina MiSeq sequencing using 2 x 250 bp paired-end mode at the Institute of Genomics (University of Tartu, Estonia).
Illumina sequencing provided 10,190,731 raw reads that were processed using PipeCraft 1.0 platform (Anslan et al., 2017). Paired-end reads were merged and quality trimmed using vsearch v1.11.1 (Rognes et al., 2016;trimming options: maxee=1, maxee_rate=999, truncqual=5). The resulting 8,145,716 sequences were re-assigned to samples based on the identifier barcodes using mothur v1.36.1 (Schloss et al., 2009;demultiplexing options: bdiffs=1, pdiffs=2). A total of 7,333,412 sequences were subjected to de novo as well as reference database (UCHIME v7.1; Nilsson et al. 2015) based chimera filtering using vsearch. To extract the full-length ITS2 subregion the sequences were processed with ITSx 1.0.11 (Bengtsson-Palme et al., 2013) to remove flanking gene fragments. Full-length ITS2 reads were assigned to operational taxonomic units (OTUs) by clustering with USEARCH v8.1.1861 (Edgar 2010) using UPARSE OTU algorithm (Edgar 2013) at 97% similarity threshold. The most abundant sequence of each cluster was selected as a representative for BLASTn sequence similarity search (Camacho et al. 2009) against both INSDC (International Nucleotide Sequence Databases Collaboration) and UNITE 8.0 (UNITE Community 2019). BLASTn e-values < e −50 were considered reliable to assign OTUs to kingdoms, whereas OTUs with e-values > e −20 were treated as unknown taxa. E-values between e −50 and e −20 were checked manually against the ten best matches for accurate assignment. The OTUs with e>-20 were removed, as well as all sequences representing nonfungi, and all singletons (OTUs present only as a single sequence read per sample). We relied on 98%, 90%, 85%, 80%, and 75% sequence identity as a criterion for assigning OTUs to species, genus, family, order or class level, respectively (Tedersoo et al., 2014). Each fungal genus, family or order was assigned to functional categories based on FUNGuild (Nguyen et al., 2016).

Appendix 4
Community-level summary of the HMSC models assessing environmental factors for functional guilds among frequent fungal OTUs and beetle species in spruce trunks. Mean diff. shows the difference in species/OTU richness between the levels of categorical variables or between the minimum and maximum of continuous variables (summed posterior means of each species response to the variable). For decay stage (two levels: early/late), early stage was used as reference; for trunk type (two levels: standing/fallen), fallen trunk was the reference. P[effect>0] shows the probability of an effect (proportion of posterior distributions of the difference above zero).

Cambium consumers Fungivores Wood consumers
Mean diff.