DNA quantification of basidiomycetous fungi during storage of logging residues

Experimental setup and sample taking

The detailed description of the experimental setup and sampling is given in Filbakk et al. (2011). Briefly, this study was carried out with residues originating from five different harvesting sites, all located close to Braskereidfoss (60°62′N/12°02′E), Norway. Three stands were predominantly Norway spruce (Picea abies (L.) Karst.) and two were dominated by Scots pine (Pinus sylvestris L.), all 70–100 years old. The experimental setup is illustrated in Fig. 1. Stand 2 was harvested in the autumn 2007 and logging residues were pre-stored; left lying on the clear-cut site until spring 2008 when piles were constructed. Stands 1 and 3 were harvested in spring 2008, and Stands 4 and 5 were harvested in autumn 2008. After each harvesting, the residues were stored in two types of piles, either in pyramid-like piles consisting of bundled residue material or loose piles consisting of unbundled material (Fig. 1). To protect the piles from precipitation, each pile was covered at the top with 2 mm thick residue wrapping paper produced by UPM Kymmene. At each sampling point the entire bundle was chipped in an industrial grinder (Peterson 4700B), resulting in about 1,000 kg of chipped material. From this source five replicates (1–2 kg each) were provided for further analysis. The chips were then removed before the next sample was chipped and sampled. Samples from piles were taken before storage (Start), then in spring 2008, autumn 2008, spring 2009, and summer 2009 (Fig. 1). All samples were analyzed for moisture content by the oven drying method at 103 °C (CEN/TS-14774-1, 2004), and for calorific value (CEN/TS-14918, 2005). To get a rough estimate of the dry matter loss in bundles, each bundle was weighed before placing into piles and at then before chipping by using a scale with accuracy of one kg. The total dry matter loss in bundled piles was calculated as the difference between initial and final mass of each bundle (Filbakk et al., 2011).

DNA extraction

From each chipped sample, about 0.4 kg was randomly taken, dried at 50 °C for 24 h, and grinded first coarsely (Retsch Mühle grinder, 5 mm mesh; Retsch Gmbh, Haan, Germany) and then finely (IKA Werke MF 10 basic grinder, 0.5 mm mesh; IKA^®-Werke Gmbh & Co., Staufen, Germany). The same sample batch was used for analyses by Filbakk et al. (2011). About 80 mg of finely grinded material was milled to powder-consistency in liquid nitrogen using a Retsch mixer mill (MM 300, Retsch Gmbh, Haan, Germany). Aliquots of 20 mg were prepared from powdered material for each treatment and total DNA was extracted using a DNeasy Plant Mini Kit (Qiagen, Hilden, Germany). The protocol provided by the manufacturer was followed. To account for the variation in DNA extractability in the environmental samples and to normalize for this variation, 5 ng of an external reference DNA, pGEM plasmid (pGEM-3Z Vector; Promega, Madison, Wisconsin, USA), was added to each sample upon start of each DNA extraction (Hietala et al., 2009). The extracted DNA was eluted in 50 µl of buffer AE and stored at −80 °C until processed by qPCR.

Primer selection

The primers were based on prior findings by Vilgalys & Hester (1990) and Fierer et al. (2005) with minor modification to improve the number of basidiomycetes amplified. As a forward primer the 5.8sr TCGATGAAGAACGCAGCG primer was used (Fierer et al., 2005; Vilgalys & Hester, 1990) and as a reverse primer we selected a 2 nucleotide truncated ITS4-X primer CAG GAG ACT TGT ACA CGG TCC as it amplifies a larger set of basidiomycetous species. Primers for the internal pGEM control for extractability were selected as previously described by Coyne et al. (2005) and Pilgård, Alfredsen & Hietala (2010).

To test the primer specificity for basidiomycetes, pure cultures of organisms representing the target DNA (basidiomycetes) were used as positive controls (Armillaria borealis 2005-713/2, Fomitopsis pinicola 1946-755/2, Coniophora puteana 1982-97/3 and Schizophyllum commune 1956-1236/1), and as negative controls non-target DNA (aseptic spruce seedling roots, Trichoderma sp. 1959-1919/471, Penicillium sp. 2007-161/14/3 and Cladosporium cladosporoides 1967-149/11), all fungi obtained from the Norwegian Forest and Landscape Institute’s fungal culture collection (http://www.skogforsk.no/skogpatologi/database/searchform.cfm) were grown in petri dishes on cellophane over malt agar medium and incubated at 25 °C. To verify the specificity and sensitivity of the primers also on degraded wood samples, known dual mixtures of basidiomycetous DNA and host tree (Serpula-degraded and Trametes-degraded pine wood) together with internal control pGEM plasmid DNA were used. The presence of amplified qPCR product in the expected size range was verified by agarose gel runs. Negative controls consisted of both water and purified Scots pine DNA.

qPCR conditions

The qPCR detection of basidiomycetous DNA (DNA_bas) was performed using SYBR Green PCR Mastermix (Applied Biosystems, Foster City, California, USA) and the reference pGEM was quantified with TaqMan Universal PCR Master Mix (Applied Biosystems #4304437; Applied Biosystems, Foster City, California, USA).

The internal standard pGEM for calibrating DNA extractability was quantified as described by Coyne et al. (2005). The pGEM standard curve was prepared from serial diluted samples containing 0.066–0.000066 ng of pGEM DNA giving a standard curve of y = 2.7371 − 0.2642x (x is the Cq value and y the logarithmic amount of pGEM DNA present in the sample).

For quantification of DNA_bas a standard curve was prepared from DNA isolated from four basidiomycetous fungi (A. borealis, F. pinicola, C. puteana and S. commune), using serial diluted samples of 0.03–0.0003 ng of DNA giving a standard curve of y = 3.0922 − 0.176x (x is the Cq value obtained by qPCR and y the logarithmic amount of DNA_bas present in the sample).

For qPCR detection and quantification of DNA_bas the primer concentrations of 60 nM and 3 µl of extracted DNA solution from each sample were used for each reaction. The qPCR cycling was carried out using cycle parameters of 95 °C for 10 min to activate the polymerase, followed by 40 amplification cycles of 95 °C for 15 s and 60 °C for 1 min with signal thresholds set automatically.

The qPCR was performed with ABI PRISM 7700 (Applied Biosystems Foster City, California, USA). After amplification the data were analyzed and plotted (fluorescence vs. cycle number) using the Sequence Detection System, version 1.7a, Software Package (Applied Biosystems Foster City, California, USA). The extent of amplification was calculated as a mean Cq value of 2 technical replicates for each sample. All PCR reactions were performed in singleplex conditions under standard PCR cycling parameters. Undiluted, 10×, 100×, 1,000× and 10,000× diluted experimental sample concentrations were tested to investigate the presence of compounds inhibitory (impurities) to PCR amplification and to ensure that the amount of template fell in the linear range of the standard curve. The 100× dilution showed no indications of inhibition and will thus be presented. Five biological and two technical replicates were used for each sample.

Statistical analysis

We analyzed the amounts of DNA_bas in logging residues from five forest stands as dependent parameter against the following independent categorical parameters: tree species, harvesting times, storage, within bundle placement, storage time. Precipitation, moisture, bundling time, mean temperature, dry matter loss and calorific value were continuous parameters in the model (Tables 1 and 2). The values for DNA_bas were normally distributed. To test whether the parameters had significant effects on the amount of DNA_bas, we used analysis of variance (ANOVA) for all categorical parameters and linear regression for all continuous parameters. In addition, we tested interaction among wood moisture and storage method. First we tested whether pre-storage (leaving the logging residues on the harvesting site during the winter) had a significant effect on the DNA_bas between Stands 1 and 2. Because it did not, we included all five stands in the same model. Means in categorical parameters were compared by Tukey–Kramer HSD test at P < 0.05). In all cases, a null hypothesis was rejected at the 5% level of significance. All statistics were done using JMP (SAS Institute Inc., Cary, NC, USA).

qPCR assay

The qPCR primers used in our assay were specific for tested DNA_bas, while the non-target DNA used as negative controls (water, host tree DNA, bacterial and non-basidiomycetous fungi) was not amplified under the conditions used. Likewise, the primers used for the calibration of DNA extractability between samples amplified only the pGEM plasmid used as an internal control in these studies. The qPCR can detect down to one average size fungal genome (i.e., one fungal cell) corresponding to the lowest point on our standard curve (∼0.0003 ng).

Our qPCR assay was able to quantify DNA_bas in logging residues. It detected DNA_bas in the range of 0.01–0.69 ng DNA_bas/mg in chipped logging residue (Fig. 2). That the most elevated amounts of DNA_bas detected was after the highest precipitation event i.e., conditions promoting basidiomycetous growth (see below, ‘Variation in DNA_bas over time’ and ‘Moisture’) is an indicator that the sampling was representative and that the selected primers were efficient. This same general pattern was found in all stands: the lowest DNA_bas value was detected at the start of the storage period, followed by increase in the wetter period in spring 2009, followed by a decrease during the dryer period at the end of the storage, suggesting that our detected values may reflect the environmental conditions impact on basidiomycetous growth. Our DNA_bas values were considerably higher than those found by Pilgård et al. (2011) in Scots pine heartwood, preservative treated and furfurylated stakes in a soil contact (EN 252, 1989), when using qPCR. However, the substrates and exposure situation in these two studies were very different; Scots pine heartwood has some natural resistance to fungal decay and preservative treated or modified wood are commercial treatments used to hinder the basidiomycetous deterioration. In this study, the forest residues were not treated with wood protection systems in order to prolong their service life and thus were likely exposed to higher colonization potential by fungi, which may explain higher DNA_bas in samples tested in our study.

Although the qPCR assays are highly sensitive with high resolution and correlation to mass loss at the early stages of decay, they may lack accuracy in advanced decay stages due to substrate depletion (Eikenes et al., 2005). We did not have substrate depletion in our material, since the level of decay was at all times only in its initial stages of colonization and thus substrate availability was plentiful for further basidiomycetous growth when the environmental conditions were favorable.

The accuracy of the qPCR based quantification of fungal biomass depends on extraction efficiency and purity of extracted DNA. The mean and modus concentrations of extracted DNA in our samples were even 95 and 83 ng/µl, respectively. The ratio of absorbance at 260 nm and 280 nm is used to assess the purity of DNA. The mean and modus value for 260/280 ratio in our extracted DNA measured by NanoDrop (NanoDrop Technologies, Inc, Wilmington, Delaware, USA) was 1.8, which is accepted as “good quality” DNA. The absorbance ratio 260/230 is used as a secondary measure of nucleic acid purity and expected values are commonly in range 2.0–2.2. Mean and modus values in our samples were 1.4, suggesting that contaminants absorbing at 230 nm, such as carbohydrates or phenols, may have been present. Impact of contaminants was avoided by diluting samples, and dilution 100X ensured optimal accuracy.

In spite of the increasing use of qPCR for fungal biomass estimation, it is known that there is variation in ITS copy numbers (per ng DNA) among fungal species (Baldrian et al., 2013; Song et al., 2014), which will impact on the biomass estimation. We used a DNA from four different basidiomycetes to make our standard curves to partly accommodate for such ITS differences, but the obtained values must be considered as estimates of DNA_bas. When Baldrian et al. (2013) compared fungal biomass content in litter and soil by using three different methods (ergosterol, phospholipid fatty acid and qPCR), they demonstrated that the obtained estimates led to large differences in relative amount of litter and soil fungal biomass. They estimated 28, 7 and 2 times larger fungal biomass in litter than soil by ergosterol, phospholipid fatty acid and qPCR, respectively, indicating that qPCR was the one least likely to overestimate the fungal biomass in litter.

DNA_bas and storage conditions

The statistical model we used included data from all stands and explained 68% of the variability in the data (Table 2, n = 133, R² = 0, 68). Of all the parameters we tested, only harvesting time, storage time, precipitation, temperature, wood moisture content, tree species and storage method significantly affected the amount of DNA_bas detected in the logging residues (Table 2). In addition, the wood moisture*storage method contributed significantly. As fungi colonize the material, fungal biomass, species diversity and material decomposition change. Therefore in our study, each time the sample was taken, offers a mere snapshot of the situation at the given time.

Variation in DNA_bas over time

In the model, the storage time affected the DNA_bas amount significantly (p < 0.0001). Generally, the amount of DNA_bas decreased at the end of the storage (Figs. 2 and 3). In all stands and pile types we detected the lowest DNA_bas values at the start, slightly rising during the storage and decreasing at the end of the study (Fig. 2). The transient DNA_bas increase was significant for the combinations illustrated in Figs. 2B, 2C, 2E and 2F.

The highest DNA_bas content was detected in bundles after 270 days, in spring 2009, in a Norway spruce stand (n = 5, mean 0.39 ng DNA, Fig. 2B) and in a Scots pine stand (n = 5, mean 0.51 ng DNA, Fig. 2F). Both peaks coincided with the highest precipitation (9 mm) measured during the entire study period, almost 5-fold higher than the mean precipitation.

The general decline in amount of DNA_bas with time may be due to transpiration drying, as transpiration continues from foliage or open wood surfaces after harvesting (Andersson et al., 2002). Indeed, such a decline in moisture has already previously been documented in this material by Filbakk et al. (2011).

Material from Stands 4 and 5, with the lowest DNA_bas values, was harvested and piled in autumn 2008 and had the shortest storage period. It is likely that the conditions were not conducive for fungal development during most of the storage period (winter/spring), since basidiomycetous fungi have restricted growth at low temperatures.

Practical implications

Both drying (including fall off needles and twigs) and microbial activity will result in dry matter loss in logging residues. In Filbakk et al. (2011) about 20% total dry matter loss was measured for bundled residues. However, this dry matter loss was most probably caused by handling together with the process of transpiration drying, causing the dry foliage and small twigs to fall off. Because the error in exact measuring of the dry matter loss in bundles was large, the mass loss caused by fungi alone could not be specified. The needle and twig fall off is considered beneficial because (1) its presence lowers the final combustion efficiency and (2) the nutrients contained in this material remain at the site as opposed to nutrient depletion by its complete removal.

We show that moisture is a key factor for basidiomycetous growth in piles. During storage time of maximum 460 days the moisture content in logging residues gradually decreased together with DNA_bas content. Hence, our results indicate that that storage of logging residues for maximum 460 days under these conditions give low levels of decay. Because we found higher values of DNA_bas in loose residues compared to bundles, the practice of bundling does not seem to promote more basidiomycetous growth than leaving the residues in loose piles. Neither did pre-storing of residues during winter seem to make a difference to DNA_bas content.

Our data suggest that basidiomycetous growth may be temporarily stimulated when the material has the optimal water content, such as in periods of increased precipitation, This supports the practice of covering the piles during storage.