Microb Ecol DOI 10.1007/s00248-014-0396-3
Fungal and Bacterial Community Succession Differs for Three Wood Types during Decay in a Forest Soil Lynn Prewitt & Youngmin Kang & Madhavi L. Kakumanu & Mark Williams
Received: 6 August 2012 / Accepted: 11 February 2014 # Springer Science+Business Media New York 2014
Abstract Wood decomposition by soil microorganisms is vital to carbon and nutrient cycles of forested ecosystems. Different wood types decompose at different rates; however, it is not known if there are differences in microbial community succession associated with the decay of different wood types. In this study, the microbial community associated with the decay of pine (decay-susceptible wood), western red cedar (decay resistant) and ACQ-treated pine (Ammoniacal Copper Quaternary, preservative-treated pine for decay resistance) in forest soil was characterized using DNA sequencing, phospholipid fatty acid (PLFA) analysis, and microbial activity over a 26-month period. Bray–Curtis ordination using an internal transcribed spacer (ITS) sequence and PLFA data indicated that fungal communities changed during succession and that wood type altered the pattern of succession. Nondecay fungi decreased over the 26 months of succession; however, by 18 months of decay, there was a major shift in the fungal communities. By this time, Trametes elegans
L. Prewitt Department of Forest Products, Forest and Wildlife Research Center, College of Forest Resources, Mississippi State University, P.O. Box 9820, Starkville, MS 39762, USA M. L. Kakumanu Horticulture, Rhizosphere-Soil Microbial Ecology and Biogeochemistry Lab, Virginia Polytechnic and State University, 311 Latham Hall, Blacksburg, VA 24061, USA Y. Kang The Basic Herbal Medicine Research Group, Herbal Medicine Research Division, Korea Institute of Oriental Medicine (KIOM), Daejeon 305-811, Republic of Korea M. Williams (*) Horticulture, Rhizosphere-Soil Microbial Ecology and Biogeochemistry Lab, Virginia Polytechnic Institute and State University, 311 Latham Hall, Blacksburg, VA 24061, USA e-mail: [email protected]
dominated cedar and Phlebia radiata dominated pine and ACQ-treated pine. The description of PLFA associated with ACQ-treated pine resembled cedar more than pine; however, both PLFA and ITS descriptions indicated that fungal communities associated with ACQ-treated pine were less dynamic, perhaps a result of the inhibition by the ACQ preservative, compared with pine and cedar. Overall, fungal community composition and succession were associated with wood type. Further research into the differences in community composition will help to discern their functional importance to wood decay.
Introduction Microbial decomposition of wood plays a key role in regulating forest carbon and nutrient cycles [1–3]. The activity of these microorganisms is dependent upon an array of environmental factors that include water availability and temperature. Wood chemistry is also an important predictor of decomposition, with rates varying across a broad range of wood types. For example, rates are influenced by wood density and the content of soluble wood extractives, cellulose, and lignin [4, 5]. These differences are likely to influence the dominant microbial community associated with wood decomposition. Different types of microbes filling different ecological niches could result in feedbacks that influence the rate and types of biochemical processes of decay [2, 3]. While a general model of microbial succession during the decomposition of wood has been described for decades, details about the specific types of microbes associated with the process remain unpredictable. Fungi are considered dominant members during the decomposition of wood; however, bacteria are often the initial colonizers , feeding on available sugars and increasing the permeability of wood [7–9]. During this time, the so-called “nondecay” fungi, such as certain molds and sapstain fungi,
L. Prewitt et al.
also utilize freely available nonstructural wood substrates such as sugars . The true wood-decay fungi (soft, brown, and white-rot fungi) then cause loss in wood strength and generally appear during the mid to late stages of wood decay [11, 12]. This general model of fungal community succession during the decomposition of wood has also been shown to predict broad shifts from soft-rot, brown-rot, and then to white-rot fungi [13–15] and is thought to be related to the stage of decomposition and the availability of substrate [16, 17]. While ultimately limited by the immigration and occurrence of available fungal taxa, changes have been linked to the ability of fungi to compete and dominate substrate use during each successional stage. Hence, successional change in fungal communities during wood decomposition can vary partially due to resource competition [18, 19], wood chemistry, and other factors related to microbial inhibition. The details of microbial community change, however, are less well known. Fungal taxa differ in the capacity and efficiency of wood catabolism, and so, an understanding of the natural dynamics of fungal community succession during the decomposition of wood will help describe some of these functional changes [19, 20]. The structural framework of wood consists of cellulose, hemicelluloses, and lignin and comprises approximately 95 % of the wood’s total composition . Wood extractives (organic soluble materials) represent a smaller percentage of the wood’s mass (~5 %) but might have a significant effect on wood-decay fungi and activity. In particular, cedrol and thujaplicins, the extractives found in cedars and that are known to have antifungal properties provide cedar with a highly durable wood compared with most other softwoods such as pine . Pines, which contain different extractives, are generally considered more susceptible to microbial decay than cedars and junipers . Though the extractives have been shown to modify fungal activity, questions remain about the changes that occur to microbial communities during the decay and decomposition of wood . To better describe the succession of microorganisms associated with the decomposition of wood in forest ecosystems, with the objectives to observe (a) the community membership of decomposing wood and (b) fungal community change associated with three different wood types with different susceptibilities to degradation over >2 years of decomposition. Fungal gene expression related to lignin degradation was reported to be very different among the three woods [24, 25]. The microbial community structure during decomposition of the three wood types pine, (a relatively decaysusceptible wood), cedar (decay-resistant wood), and ACQtreated pine (ammonium copper quaternary, chemically treated to be decay resistant) was characterized. The hypothesis was that fungal communities would follow a pattern of succession from early colonizing nondecay fungi to wood-decay fungi during the decomposition of wood. It was also hypothesized that membership and structure of fungal communities would differ during decomposition based on wood type. The
term “decomposition” is used to broadly describe catabolism, whether it originates from structural or nonstructural wood.
Materials and Methods Preparation of Wood Stakes for Soil Incubation Pine (Pinus taeda L.) and western red cedar (Thuja plicata.Donn ex D.Don ) boards (5.1 cm×10.2 cm×180 cm and 2.5 cm× 10.2 cm×180 cm, respectively) used for this study were purchased from Lowes’s Home Improvement Center, Starkville, Mississippi. One set of pine stakes were later treated with a wood preservative ACQ to 0.15 pcf by the full cell method . Each board was cut into strips measuring 14 mm× 14 mm×115 mm (T × R × L) and numbered for identification. Samples were wrapped in Saran™ plastic wrap to equilibrate for 7 days, air-dried for 1 week, and equilibrated to approximately 12 % moisture content (MC). Afterward, the samples were soaked for 7 days with daily water changing. The stakes were then air-dried for several days until they reached 60 % MC as determined by weight loss, based on fresh weight . The wood stakes were placed in plastic containers (250 mm× 365 mm×220 mm) filled with sieved silty clay soil collected from the top 7.6 cm of undisturbed forested soil at Dorman Lake, Oktibbeha County, Mississippi. Eight circular holes (5mm diameter) were made in the bottom of each container for drainage. A screen (250 mm×365 mm) was placed at the bottom of each container followed by gravel (20 mm deep) and the Dorman soil (100 mm deep). The MC of soil in each container was adjusted to 90 % of its field capacity and monitored weekly. Six unsterilized stakes each of pine, cedar, and ACQ-treated pine were inserted 8 cm deep into the soil in replicate containers for each sampling time for a total of 84 stakes per wood type. The containers were placed in a greenhouse at 25 °C with a relative humidity of 30–50 % from November to March and outside from April to October (experiment was conducted between December 2007 through April 2010). Two stakes per container were covered with nylon stocking material and used to monitor 40–80 % MC on the wood samples . Stakes were weighed every week, and water was added to the soil if needed. Random sampling of the wood was conducted on day 0 and bimonthly over 26 months.
Modulus of Elasticity Dynamic modulus of elasticity (MOE) provided a metric of wood decay and was measured for each sampled stake on a bimonthly schedule. The average percentage of MOE change was calculated using the formula [(initial MOE − current MOE)/initial MOE] × 100 % . The percent MOE loss of the wood was used as an
Fungal and Bacterial Community Succession on Three Wood Types
indication of decay: The higher the % MOE loss, the more decayed the wood. Sample Collection Three of the six randomly harvested wood stakes showing the most decay (based on MOE results) were selected at 0, 4, 10, 18, and 26 months and then individually cut into 16 equal sections. Four sections were combined and ground for fungal genomic DNA and phospholipid fatty acid (PLFA) extraction, while four unground sections were used each for bacterial genomic DNA extraction and CO2 respiration. The remaining samples were immediately frozen in liquid nitrogen and stored at −70 °C. Extraction of Fungal and Bacterial Genomic DNA from Wood The fungal genomic DNA was extracted from 50-mg grounded wood in CTAB (1,000 μl, 2 % w/v hexadecyltrimethylammonium bromide, 100 mM Tris, 20 nM Na2EDTA, and 1.4 mM NaCl. The resulting mixture was processed according to the MachereyNagel Nucleospin Plant DNA extraction kit protocol (Easton, PA, USA) as previously described . Bacterial genomic DNA was extracted by placing four unground wood sections in 5 mL of nutrient broth (OIDCO, Becton Dickinson) overnight at 28 °C with shaking. Following overnight incubation, the cell cultures were transferred in 1-mL aliquots to 1.5-mL microcentrifuge tubes and centrifuged for cell separation. The liquid portion was removed from each sample, and 10 μL of RNase A was added to each tube and incubated for 2 h at 65ºC for cell lysis, mixing every 15– 20 min by inverting the tubes. The mixture was transferred to a Nucleospin® spin column and centrifuged for 5 min at 11,000×g to filter the lysate. The filtrate was mixed with 850 μL of binding buffer and passed through a second spin column containing a silica membrane for 1 min at 11,000×g, binding the genomic DNA. The silica membrane was washed and dried by centrifugation at 13,000×g for 2 min. The DNA was eluted from the silica membrane by adding 50 μL of 65 °C elution buffer, then incubated at 25 °C for 5 min, and centrifuged at 8,000×g for 1 min to collect the eluted DNA. The quality and quantity of the extracted fungal and bacterial genomic DNAs were determined by UV absorbance at 260 and 280 nm using the NanoDrop spectrophotometer ND1000 (NanoDrop Technologies, Inc.). Extracted genomic DNAs were stored at −70 °C. Amplification of ITS and 16s rRNA Genes Bacterial and fungal DNA associated with each wood type were amplified using 16s ribosomal RNA (rRNA) gene and the internal transcribed spacer (ITS) region, respectively.
Amplification of each gene was conducted with the following thermocycler settings: initial denaturation at 94 °C for 2 min, followed by 35 cycles at 94 °C for 30 s, annealing at 60 °C for 30 s, extension at 72 °C for 30 s, and a final extension at 72 °C for 10 min. PCR products were visualized by agarose gel electrophoresis stained with ethidium bromide. Primers used for amplification were 5′-CTTGGTCATTTAGAGGAAGT AA-3′ (ITS-F) and 5′-TCCTCCGCTTATTGATATGC-3′ (ITS-R) for general fungi ITS and 5′-AGACTCGATCCTGG CTCAG-3′ (16s-F) and 5′-GGTTACCTTGTTACGACTT-3′ (16s-R) for general bacterial 16s rRNA gene . Cloning and Sequencing of Amplified DNA Products for Taxa Identification Amplified PCR products from decaying stakes were transformed into Escherichia coli plasmids using the TOPOcloning kit for sequencing (K4575-40 Invitrogen, Co., Carlsbad, CA, USA). The plasmids of positively transformed E. coli were isolated and extracted using the PureLink™ Quick Plasmid Miniprep Kit (K2100-11, Invitrogen, Carlsbad, CA, USA). Plasmids were analyzed for inserts by restriction digest using EcoRI, gel electrophoresis, and prepared for sequencing using the Dye Terminator Cycle Sequencing with Quick Start Kit (608120, Beckman Coulter Co, Brea, CA, USA). Automated sequencing was performed using a Beckman CEQ 8000 DNA Analysis System. Sequences were edited by EditSeq™ (DNASTAR Inc.). Phospholipid Fatty Acid (PLFA) Extraction Total lipids were extracted from 2 g of ground wood at 0, 4, 10, 18, and 26 months of wood aging using a modified Bligh and Dyer method . The phospholipid fraction was recovered and converted to fatty acid methyl esters for analysis . Fatty acid methyl esters were separated, quantified, and detected by an Agilent 6890 series gas chromatograph (Santa Clara, CA, USA) equipped with a flame ionization detector, an Ultra-2 column (19091B-102; 0.2 mm by 25 m), controlled by computerized ChemStation and Sherlock software. Ultrahigh-purity H2 was the carrier gas at a column head pressure of 20 kPa, septum purge of 5 mL min−1, a split ratio of 40:1, injection temperature of 300 °C, and an injection volume of 2 μL. The oven temperature increased from 170 to 288 °C at 28 °C min−1, and the analysis time of each sample was 6 min. Peak identification was carried out by the Microbial Identification System (MIDI, Inc., Newark DE, USA) following calibration with a standard mixture of 17 fatty acid methyl esters (1,300 A calibration mix). The PLFA markers used to determine the fungal population were 18:2ω6c and 18:1w9c and for the bacterial population were i15:0, a15:0, 15:0, i16:0, 16:1ω7c, i17:0, a17:0, cy17:0, 17:0, and 18:1ω7c and cy19:0.
L. Prewitt et al.
Data Analyses The statistical analysis of MOE was performed by two-way analysis of variance (ANOVA) and Tukey's test (α=0.05) for randomized complete block design (RCBD) using SAS program (SAS 9.1, SAS Institute Inc., Cary, NC, USA). Multivariate analysis of the PLFA data was conducted using PC-ORD (version 4.2) software (Gleneden Beach, OR, USA). The dominant fatty acids were relativized and analyzed by nonmetric multidimensional scaling (NMS) using Sorenson distance, as previously described . NMS is a nonparametric method that provides graphical ordination of the experimental data . Fungal sequences were separately aligned using Clustal W and analyzed by Mallard Software to check for chimeras and anomalies. Sequences were grouped by the computer program DOTUR  at 97 % evolutionary distance (D=0.03) to generate operational taxonomic units (OTUs). The relative abundance of the OTU was then analyzed by NMS using Bray–Curtis ordination. Sequences with the closest match (>98 %) were used for identification of bacterial and fungal species. Analysis of variance with repeated measures was conducted to analyze for differences in respiration and PLFA abundances. The multiresponse permutation procedure (MRPP), a nonparametric test, was used to assess differences in fungal community structure across wood type and incubation.
Fig. 1 Wood decay as determined by loss in % modulus of elasticity (MOE) on pine (denoted by x), cedar (circle), and ACQ-treated pine (square) over 26 months of decay. MOE losses were significantly greater for pine than cedar and ACQ pine (P0.90). MOE was significantly different from 0 at 6 months for pine and at 8 months for both cedar and ACQ-treated pine. Decay was thus measurable at 6 and 8 months, respectively
cedar–ACQ-pine MOE data, with polynomials showing a strong fit to the data (R2 >0.90; P