SSSAJ Journal of Natural Resources and Life Sciences Education
HOME HELP FEEDBACK SUBSCRIPTIONS ARCHIVE SEARCH TABLE OF CONTENTS
QUICK SEARCH:   [advanced]


     

Published online 28 June 2005
Published in Soil Sci Soc Am J 69:1238-1247 (2005)
DOI: 10.2136/sssaj2004.0289
© 2005 Soil Science Society of America
677 S. Segoe Rd., Madison, WI 53711 USA
This Article
Right arrow Abstract Freely available
Right arrow Figures Only
Right arrow Full Text (PDF) Free
Right arrow Alert me when this article is cited
Right arrow Alert me if a correction is posted
Services
Right arrow Similar articles in this journal
Right arrow Similar articles in ISI Web of Science
Right arrow Alert me to new issues of the journal
Right arrow Download to citation manager
Right arrow reprints & permissions
Citing Articles
Right arrow Citing Articles via HighWire
Right arrow Citing Articles via ISI Web of Science (9)
Right arrow Citing Articles via Google Scholar
Google Scholar
Right arrow Articles by McMahon, S. K.
Right arrow Articles by Myrold, D. D.
Right arrow Search for Related Content
PubMed
Right arrow Articles by McMahon, S. K.
Right arrow Articles by Myrold, D. D.
Agricola
Right arrow Articles by McMahon, S. K.
Right arrow Articles by Myrold, D. D.
Related Collections
Right arrow Soil Microbiology
Right arrow Isotopes
Right arrow Biogeochemical Processes

Soil Biology & Biochemistry

Dynamics of Microbial Communities during Decomposition of Carbon-13 Labeled Ryegrass Fractions in Soil

Shawna K. McMahona,c, Mark A. Williamsa,d, Peter J. Bottomleya,b and David D. Myrolda,*

a Dep. of Crop and Soil Science, Oregon State Univ., Agric. Life Sci. Bldg. 3017, Corvallis, OR 97331-7306
b Dep. of Microbiology, 220 Nash Hall, Oregon State Univ., Corvallis, OR 97331-3804
c Biological Sciences, Ecology, Evolution & Marine Biology, Univ. of California, Santa Barbara, CA 93106
d Dep. of Crop and Soil Sciences, Univ. of Georgia, 3121 Miller Plant Sciences Bldg., Athens, GA 30602

* Corresponding author (david.myrold{at}oregonstate.edu)


   ABSTRACT
TOP
 ABSTRACT
INTRODUCTION
 MATERIALS AND METHODS
 RESULTS
 DISCUSSION
 CONCLUSIONS
 REFERENCES
 
The soluble fraction of ryegrass [Lolium perenne L. ssp. multiflorum (Lam.) Husnot.] straw comprises a major component of residue C and its presence or absence should influence the succession of decomposer communities. Changes in phospholipid fatty acid (PLFA) profiles are indicative of shifts in microbial community structure. We used 13C-labeled ryegrass to track substrate-derived C into microbial lipids during decomposition in a microcosm-based study. Treatments were unleached straw, leached straw, and leachate, plus an unamended control. Destructive sampling took place after 0.6, 1.6, 15, 18, 50, and 80 d of incubation. Phospholipid fatty acids were extracted from bulk soil and isolated straw (detritusphere), and analyzed by gas chromatography combustion isotope ratio mass spectroscopy (GC-C-IRMS). Distinct temporal shifts occurred in PLFAs in bulk soil samples; cy19:0 was an indicator for late succession communities in unleached straw and leachate treatments, and 18:2{omega}6,9 characterized late samples in the leached straw treatment. The temporal shift was affected by the presence of the soluble fraction of straw. Microscale spatial effects were clear. Bulk soil and detritusphere communities were different and more 13C was detected in the 16:0 and 18:2{omega}6,9 PLFAs of the detritusphere than bulk soil. Carbon-13 was slowest to appear in PLFAs of leached straw samples, illustrating the importance of the soluble fraction in promoting growth of the biomass in bulk soil and in detritusphere. The 18:2{omega}6,9 PLFA, a fungal biomarker, was the most highly labeled in all treatments. Carbon-13 PLFA analysis provides greater insight into microbial community structure and functioning, facilitating more concrete conclusions regarding the role of functional groups of organisms.

Abbreviations: FAME, fatty acid methyl ester • GC-C-IRMS, gas chromatography-combustion-isotope ratio mass spectrometry • IRMS, isotope ratio mass spectrometer • ISA, indicator species analysis • MRPP, multi-response permutation procedure • PCA, principal components analysis • PLFA, phospholipid fatty acid


   INTRODUCTION
TOP
 ABSTRACT
 INTRODUCTION
MATERIALS AND METHODS
 RESULTS
 DISCUSSION
 CONCLUSIONS
 REFERENCES
 
DECOMPOSITION OF PLANT residues is a major ecosystem process that is important in recycling nutrients, maintaining organic matter, and fueling food webs in soils. Consequently, litter decomposition has been studied extensively (e.g., Swift et al., 1979; Cadisch and Giller, 1997; Berg and McClaugherty, 2003). Regulation of decomposition is a function of the intrinsic quality of the substrate (the chemical constituents of the residue), extrinsic environmental factors (such as temperature, moisture, etc.), and the composition of the decomposing community. Although the effects of microclimatic variables and substrate quality on decomposition are well understood and are the primary controllers used in models of decomposition (Smith et al., 1998), less is known about the interactions of substrate characteristics and microbial decomposers during residue decomposition (Schimel, 1995).

Residue decomposition is often mathematically described by dividing the residue C into two, or more, compartments that decompose at faster and slower rates (e.g., van Veen et al., 1984; Saviozzi et al., 1997). Although these compartments do not neatly correspond to chemically defined pools of C, many studies have shown that the fast or labile pool is comprised primarily of soluble C compounds (Reinertsen et al., 1984; Cogle et al., 1989; Collins et al., 1990; Marstorp, 1996; Trinsoutrot et al., 2000). The slow pools are made up of structural, polymeric C compounds, such as hemicelluloses, cellulose, and lignin (Wessén and Berg, 1986; Saviozzi et al., 1997). Thus, as decomposition proceeds, the chemical composition of the residue changes (Horwath and Elliott, 1996). It is tempting to postulate that the utilization of these C pools is associated with different groups of microorganisms, resulting in an orderly microbial succession that is linked to the changes in residue chemistry during decomposition. The real situation is more complex, of course, and each succession is unique (Frankland, 1998).

Plant litter is already colonized with microorganisms before its senescence (Tester, 1988), but bacterial and fungal biomass increase as decomposition proceeds (e.g., Wessén and Berg, 1986; Parmelee et al., 1989; Henriksen and Breland, 1999; Malosso et al., 2004). Many studies have shown an increase in the relative amounts of fungi vs. bacteria during decomposition (Neely et al., 1991; Beare et al., 1992; Lundquist et al., 1999; Henriksen and Breland, 2002), but some have observed fairly constant or even declining fungal/bacterial ratio (Broder and Wagner, 1988; Lundquist et al., 1999). The different patterns in fungal and bacterial biomass that have been observed may be due in part to differences in the methods used in these studies. The ratio of culturable copiotrophic to oligotrophic bacteria was found to increase during the first 3 wk of decomposition and then decline (Hu et al., 1999) and the types of fungi isolated changed during litter decomposition (Frankland, 1998). The newer methods of community level physiological profiles, PLFA analysis, denaturing gradient gel electrophoresis, and RNA fingerprinting have also found microbial community composition to shift as plant materials decompose (Sharma et al., 1998; Thirup et al., 2001, 2003; Nakamura et al., 2003; Aneja et al., 2004). How tightly these shifts in microbial communities are linked to the availability of different substrates in the decomposing litter is not clear.

The activity of microbial communities has been linked to the degradation of 13C-labeled substrates by using GC-C-IRMS, or compound-specific IRMS (Boschker and Middelburg, 2002; Zhang, 2002). At first, this approach was focused on the use of individual C compounds, such as acetate, glucose, methane, and toluene (e.g., Boschker et al., 1998; Arao, 1999; Hanson et al., 1999; Bull et al., 2000), but more recently it has been applied to more complex substrates and systems. Phillips et al. (2002) added 13C-labeled cellobiose and N-acetylglucosamine to temperate forest soils and found that 13C was not incorporated into all PLFAs; nine PLFAs contained 80% of the excess 13C recovered. Based on ratios between fungal and bacterial lipids, they were also able to determine that compared with bacterial populations, fungi consumed proportionally greater quantities of cellobiose than N-acetylglucosamine. N-acetylglucosamine was used for biosynthesis more than cellobiose, as more 13C was recovered in PLFAs. More recently, Waldrop and Firestone (2004) compared the utilization of 13C-labeled starch, xylose, vanillin, and pine needles in two soils. They found all PLFAs to be labeled after 9 d of incubation, with PLFAs associated with gram-negative bacteria most highly labeled. These studies highlight the usefulness of 13C as a tracer of substrate C into active portions of the microbial community.

The objective of this experiment was to explore the interaction between the composition of the decomposing community and the constituents of ryegrass straw during decomposition in soil. In particular, we were interested in determining: (1) how microbial community structure, as indicated by PLFA composition, changes through time and if predictable trends in microbial community structure could be discerned; (2) if the presence or absence of the soluble component of ryegrass straw would result in the development of different microbial communities; (3) whether different lipids, representing different functional groups of organisms, are enriched differently in 13C between treatments and over time; and (4) if microbial community composition varied between bulk soil and decomposing straw (detritusphere).


   MATERIALS AND METHODS
TOP
 ABSTRACT
 INTRODUCTION
 MATERIALS AND METHODS
RESULTS
 DISCUSSION
 CONCLUSIONS
 REFERENCES
 
Soil and Amendment Preparation
In fall 2001, annual ryegrass was planted in the field at the Hyslop Research Farm located north of Corvallis, OR (44°38' N, 123°12' W). Plants were labeled with 13C using a pulse–chase technique weekly between early April and mid-May 2002. A Plexiglas labeling chamber (60 x 60 x 75 cm) was placed over the plants and the soil-chamber interface sealed with wet mud. To generate 13CO2, 5 mL of 4 M HCl was added to 150 mg of 99 atom% NaH13CO3 in a sealed 120-mL serum bottle. The gas was collected in two 60-mL syringes and injected into the headspace of the labeling chamber. The 120-mL injection raised the CO2 in the chamber by about 400 µL L–1. To ensure complete transfer of 13CO2 to the chamber, air was injected into the bottle, removed, and injected into the chamber three times. Carbon dioxide concentration in the chamber headspace was monitored using a LI-COR 6200 CO2 analyzer (LI-COR Inc. Lincoln, NE). When CO2 was drawn down to 150 µL L–1, the 13CO2 injection procedure was repeated a total of four times. The 13CO2 was chased with six 120-mL injections of unlabeled CO2. The headspace concentration was allowed to drop to about 150 µL L–1 between chase injections. Ryegrass was allowed to mature in the field and was harvested in July 2002.

Soil (0–20 cm) was collected from a location adjacent to the plots in July 2002. The soil is classified as a Woodburn silty loam (Fine-silty, mixed, superactive, mesic Aquultic Argixerolls). The climate is characterized by cool wet winters and dry warm summers with 108.5 cm of annual precipitation. The surface horizon is about 25 cm deep and 1.3% organic C. The soil was nearly air dry and hard at the time of sampling. In the lab, clods were broken and soil was passed through a 2-mm sieve to remove large organic debris. Following sieving, the air-dry soil was stored in buckets at 4°C until needed.

To prepare leachate, fresh straw from each of the four plots was cut into 1- to 2-cm lengths. Cut straw (18 g) was placed with 300 mL of cold deionized water in 1-L canning jars and shaken horizontally at 270 rpm at 5°C for 24 h (Christensen, 1985). Leachates were decanted through 250-µm mesh, centrifuged at 11325 x g for 10 min, and filtered through Whatman #2 filter paper to remove fine particulate matter. All leachates were pooled and stored at 4°C for 6 d. Leached straw was composited and dried on brown paper in a forced-air oven at 60°C for 48 h.

Soil, unleached and leached straw, and leachate were analyzed by IRMS (PDZ Europa Ltd., Crewe, Cheshire, England) to determine total C content and 13C abundance relative to Pee Dee belemnite (PDB). The {delta}13C values were: 129 ± 4{per thousand} for unleached straw, 130 ± 9{per thousand} for leached straw, and 121 ± 1{per thousand} for leachate; the {delta}13C signature of soil was –26.5 ± 0.1{per thousand}.

Experimental Design
Microcosms were prepared in 1-L canning jars with air-dried soil (150 g oven dry weight), a C amendment, and deionized water to bring the soil to about field capacity (0.29 g water g–1soil). The treatments consisted three C amendments—2.25 g of unleached straw (917 mg C), 1.69 g of leached straw (675 mg C), 29.0 mL of leachate (133 mg C)—and an unamended control. The amount of soluble C added in leachate approximated the amount of soluble C added in the unleached straw.

To prepare straw treatments, residue was mixed with about 120 g of soil and transferred to the jar, which contained 24.0 mL of water. This method resulted in more uniform wetting of the fine-textured soil. The remaining soil was poured on top to ensure burial of all plant residue. An additional 7.5 mL of water was pipetted evenly over the soil surface. Leachate treatments were prepared by dispensing 29.0 mL of leachate into the bottom of the jar, adding all the soil and pipetting an additional 2.5 mL of water on the soil surface. The control without added C was prepared similarly to the leachate treatment. Twenty-four replicates of each treatment were prepared. Jars were covered with eight layers of cheesecloth to allow adequate ventilation but slow water loss. Microcosms were placed in an incubator at 25°C using a blocked design with each of four shelves representing a block. Replicates were arranged on the shelves using a Latin rectangle design.

Four randomly selected replicates (one from each block) of each treatment were destructively sampled at six times throughout the 80-d incubation. Sampling took place after 0.6, 1.6, 15, 18, 50, and 80 d. Soil was homogenized in each jar before sampling. Straw was removed from soil to form a detritusphere (straw and tightly adhering soil) sample and a sample of bulk (straw-free) soil was collected for immediate PLFA analysis.

Phospholipid Fatty Acid Analysis
We isolated PLFAs using a single-phase extraction method (White and Ringelberg, 1998) as modified by Butler et al. (2003) with the following adjustments. Detritusphere samples (0.5–1.5 g oven dry weight) were extracted using 10 mL of methanol, 5 mL of chloroform, and 4 mL of phosphate buffer. The centrifugation step was repeated using 20 mL of methanol and 10 mL of chloroform with soil; detritusphere samples received 7 mL of methanol and 3.5 mL of chloroform. Thirty milliliters of 3 M NaCl solution was added to filtrate from soil samples to induce phase separation; 1.0 g of Na2SO4 crystals was also added to remove trapped water mixed in the chloroform phase. Detritusphere samples received 10 mL of 3 M NaCl and 0.5 g of Na2SO4. Samples were placed in a warm water bath (<35°C) during all N2 drying steps. Bulk lipids were dried in serum bottles, which were closed with a septum and crimp-top, flushed with N2 and evacuated twice, overpressurized with N2 gas, and frozen at –80°C until separation of PLFAs. Lipids were separated by solid-phase extraction using silicic acid columns and PLFAs stored under N2 at –80°C until methylation. Fatty acid methyl esters (FAMEs) produced during mild alkaline methanolysis were extracted using 2 mL of 4:1 hexane/chloroform, 200 µL of 1 M acetic acid, and 2 mL of deionized water. Phospholipids partitioned into the top layer, which was transferred to another tube. An additional 2 mL of 4:1 hexane/chloroform was added to the first tube, allowed to sit for 5 min, and also transferred to the second tube. Fatty acid methyl esters were transferred to 200-µL pulled glass inserts in 2-mL Agilent vials (Agilent Inc., Palo Alto, CA) using three 50-µL aliquots of chloroform. Chloroform was gently evaporated under N2 gas. Fatty acid methyl esters were resuspended in 40 µL of chloroform, closed with a septum screw lid, and stored at –20°C until analysis.

Gas Chromatography-Combustion Isotope Ratio Mass Spectroscopy
Fatty acid methyl esters were separated by capillary chromatography using an Agilent 6890 (Agilent Inc., Palo Alto, CA) capillary gas chromatograph equipped with a 30-m Hewlett-Packard Innowax 2 column (Hewlett Packard, Palo Alto, CA; 0.25 mm i.d., 0.25 µm film) connected to a Europa ORCHID on-line combustion interface attached to a Europa 20-20 isotope ratio mass spectrometer (PDZ Europa Ltd., Crewe, Cheshire, England). The starting temperature was 120°C, and was ramped up to 250°C in 5°C per min increments. The run was terminated after the system had been at 250°C for 5 min. Fatty acid methyl esters move off the column based on boiling points and pass through the combustion unit where they are converted to CO2 before analysis by the IRMS.

Chromatograms were processed using GC Post Processor v.2.5 (PDZ Europa Ltd., Crewe, Cheshire, England). Peaks were identified based on comparison with a known lipid profile generated by GC–MS analysis of PLFAs from similar soil (Butler et al., 2003). Twenty peaks were identified and analyzed on a mole percentage basis based on relative areas of the peaks in a chromatogram.

The {delta}13C values were determined based on CO2 pulse standards ({delta}13C = –49.9{per thousand}) at the beginning of each run. Fatty acid methyl ester isotopic signatures were corrected for the methanol-derived C ({delta}13C = –44.6{per thousand}) during methylation. We used a mixing model to calculate the amount of substrate-derived C incorporated into specific PLFAs using their {delta}13C values, and the {delta}13C values of the soil and substrate.

Nomenclature
Phospholipid fatty acids were named using standard nomenclature and assigned to taxonomic groups based on reviews and recent literature (Stahl and Klug, 1996; Zelles, 1999; Myers et al., 2001; Ruess et al., 2002; DeForest et al., 2004; Waldrop et al., 2004). Terminally branched saturated PLFAs (i14:0, a15:0, i15:0, i16:0, a17:0, i17:0) were considered markers for gram-positive bacteria and mid-chain branched saturated PLFAs (10Me16:0, 10Me17:0) were associated with actinomycetes. Gram-negative bacteria were associated with some monounsaturated PLFAs (16:1{omega}7, 17:1, 18:1{omega}7) and cyclopropyl saturated PLFAs (cy17:0, cy19:0). Short or odd-chain saturated PLFAs (14:0, 15:0, 17:0) were considered non-specific bacterial markers. We considered 18:2{omega}6,9 and 18:1{omega}9 as markers for fungi. Common saturated PLFAs (16:0, 18:0) that are present in all organisms were not assigned to a taxonomic group. Ratios of fungal/bacterial PLFAs were calculated from the mole percentage of the designated fungal and bacterial PLFAs (Bailey et al., 2002; Waldrop et al., 2004).

Data Analysis
All 20 peaks were not present in every sample. Incomplete lipid profiles were caused by low total lipids measured, presumably because of variations in extraction efficiency. When this occurred, we excluded the sample from further analysis. Our general criterion for exclusion was when more than four peaks were missing relative to the number present in other replicates. This criterion was based on empirical observations that the absence of more than four peaks resulted in that sample being an outlier in an ordination analysis. Twenty-four of 144 samples were either lost during extraction or excluded based on this criterion.

All multivariate data analysis was conducted using PC-ORD v. 4.0 (McCune and Mefford, 1999). Phospholipid fatty acid profiles were analyzed using principal components analysis (PCA) on a correlation matrix of mole percentage values. Data were determined to meet linearity and normality assumptions through examination of summary statistics of each PLFA. A multi-response permutation procedure (MRPP) was used to test the hypotheses of no difference between detritusphere and bulk soil at all times and no difference between times for a given treatment. The MRPP is a nonparametric method for testing group differences that is not constrained by distributional assumptions (McCune and Grace, 2002). The MRPP provides a measure of effect size (A), which describes with-in group homogeneity; perfect homogeneity produces A = 1. Significance of A is tested using a randomization test. Euclidean distance measure was used for MRPP to be compatible with PCA.

When significant treatment or time differences were identified by MRPP, groups were chosen from PCA ordination plots for indicator species analysis (ISA). A perfect indicator, or PLFA in this case, is always present in the group for which it is an indicator and occurs in no other group (McCune and Grace, 2002). Statistical significance of an indicator value was tested by a Monte Carlo method with 1000 randomizations of samples among groups. The PLFAs were considered indicators if they were significant with p = 0.001, except in leached straw through time, for which p = 0.002 was used.

Analysis of variance was used to test for time and treatment effects on substrate-derived C and fungal/bacterial ratios using SAS 8.2 (SAS Institute, Cary, NC).


   RESULTS
TOP
 ABSTRACT
 INTRODUCTION
 MATERIALS AND METHODS
 RESULTS
DISCUSSION
 CONCLUSIONS
 REFERENCES
 
The dominant trend observed when all treatments were analyzed together using PCA was that detritusphere and bulk soil PLFA profiles were different (Fig. 1) . This trend accounted for 47% of variation in the data. These groups were strongly significant according to MRPP (p ≤ 0.0001).



View larger version (26K):
[in this window]
[in a new window]
  		Fig. 1. Principal components ordination of all treatments through all times. Percentage of variance explained is indicated on each axis. Symbols indicate treatments. Ellipses enclose bulk soil and detritusphere samples, which were significantly different from each other based on multi-response permutation procedure (MRPP) analysis.

 
A number of significant indicators were identified by ISA (p ≤ 0.001) for distinguishing bulk soil and detritusphere samples. For example, 16:0 and 18:2{omega}6,9 were strong indicators of detritusphere samples; in contrast, i15:0, a15:0, i16:0, 16:1{omega}7, i17:0, a17:0, 17:1, cy17:0, and 10Me17:0 were dominant in bulk soil community profiles.

Individual Treatments through Time
Bulk Soil
No distinct groups based on sampling dates were observed in the unamended control, but MRPP results indicated that groups in the amended treatments were strongly significant (p < 0.001; Fig. 2) . In the unleached straw treatment, two distinct groups consisting of samples from 0.6, 1.6, and 15 d, and from 18, 50, and 80 d were produced by PCA (Fig. 2b). These groups were separated primarily by PC 1, which accounted for 51% of variation in the data. The leached straw and leachate treatments formed two distinct groups when analyzed using PCA: 0.6 and 1.6 d, and 15, 18, 50, and 80 d (Fig. 2c and 2d). These groups were also separated by PC 1, which summarized 44% of the variation in the leached straw and 65% in the leachate treatments.



View larger version (25K):
[in this window]
[in a new window]
  		Fig. 2. Principal components ordination of bulk soil treatments through time. Percentage of variance explained is indicated on each axis. Symbols indicate sampling date. The four plots show: (a) control, (b) leachate, (c) leached straw, and (d) unleached straw treatments. Ellipses enclose samples from early (dashed line) and late (solid line) samplings, with indicator PLFAs for each group listed. No significant temporal trends were observed in the control but significant temporal trends were observed for all amended treatments based on multi-response permutation procedure (MRPP) analysis (p < 0.001).

 
Results of ISA showed that a15:0 was a strong indicator of microbial communities at early samplings (p ≤ 0.002) in all amended treatments. 10Me16:0 was also a significant indicator of early samples in the leached straw treatment. In contrast, longer and more complex PLFAs were indicative of samples taken later in the incubation. In unleached straw and leachate treatments, 16:1{omega}7, 18:0, and cy19:0 were associated with late samples; in contrast, 18:2{omega}6,9 was the only significant indicator of late samples in the leached straw treatment. Because there was no temporal pattern in control samples, no significant indicators could be identified.

Detritusphere
Despite MRPP indicating statistically significant groups among sampling times (p < 0.0001), temporal trends in detritusphere samples were not as obvious as bulk soil samples (Fig. 3) . In unleached straw detritusphere, two groups consisting of 0.6, 18, and 50 d, and 1.6, 15, and 80 d could be distinguished. The PCA of the leached straw detritusphere structure was dominated by the difference between 50 d and all other sample times (data not shown).



View larger version (20K):
[in this window]
[in a new window]
  		Fig. 3. Principal components ordination of unleached straw detritusphere through time. Percentage of variance explained is indicated on each axis. Symbols indicate sampling date and dotted arrows show the temporal shifts among sampling dates, which were significant based on multi-response permutation procedure (MRPP) analysis (p < 0.0001). The bold arrow shows the correlation of the fungal/bacterial ratio with respect to the data, with higher ratios associated positively with PC 1 and negatively with PC 2.

 
Fungal/Bacterial Phospholipid Fatty Acid Ratios
Table 1 summarizes fungal/bacterial PLFA ratios. Ratios for bulk soil treatments were not significantly different from each other at any time, although ratios for leached straw samples were slightly higher than other bulk soil samples at most times. Detritusphere ratios were much more variable and often significantly higher than bulk soil ratios by two- to ten-fold. Fungal/bacterial ratios for the leached straw detritusphere and unleached straw detritusphere samples occasionally varied among sampling times but no significant trends were observed. In bulk soils, no temporal trends in fungal/bacterial ratios were observed for control and leachate samples but there were significant positive correlations (p < 0.05) of increasing fungal/bacterial ratios with time for leached and unleached straw treatments.


View this table:
[in this window]
[in a new window]
  		Table 1. Mean ratio of fungal/bacterial PLFAs, calculated on a mole percentage basis for all treatments through time. There was a significant treatment-by-time interaction. Treatment effects are shown in this table.{dagger}

 
Incorporation of Carbon-13 into Selected Phospholipid Fatty Acids
All PLFAs in the unamended control had similar {delta}13C values, which were slightly depleted relative to the {delta}13C of native soil organic matter, whereas all PLFAs from amended soils and detritusphere samples were significantly enriched with 13C (data not shown). To simplify sample comparisons we calculated the percentage of C derived from the added substrate. A time-by-treatment interaction was significant only in 16:0 of bulk soil; in all other cases the main effects of treatment and time were considered.

The amount of substrate-derived C varied among PLFAs, treatments, and changed with time (Fig. 4) . Detritusphere samples contained significantly more substrate C than bulk soil samples for 16:0 but not 18:2{omega}6,9 (Fig. 4a, and 4b). About 10% more substrate-derived C was found in 18:2{omega}6,9 than 16:0 at each time in the detritusphere samples, but there was no significant difference in substrate-derived C between the unleached and leach straw detrituspheres. The amount of substrate-derived C in detritusphere PLFAs increased by 20 to 30% from 0.6 to 80 d of incubation.



View larger version (31K):
[in this window]
[in a new window]
  		Fig. 4. Percentage of substrate-derived C of selected PLFAs through time. Symbols indicate treatments and are means with error bars representing ± SE. The six plots show: (a) 16:0, (b) 18:2{omega}6,9, (c) i15:0, (d) a15:0, (e) 10Me16:0, and (f) cy19:0.

 
Substrate-derived C in PLFAs extracted from bulk soil varied by PLFA, treatment, and time. In general, 18:2{omega}6,9 and a15:0 were the most highly labeled PLFAs, and i15:0, 10Me16:0, and cy19:0 contained the least substrate-derived C in bulk soil samples (Fig. 4). Except for cy19:0, which showed no significant effect of treatment, there was a consistent, and often significant, trend for substrate-derived C in PLFAs to decrease from the unleached straw to leachate to leached straw treatments. This trend was particularly evident at the first two sampling times. Four temporal patterns in substrate-derived C were observed among the PLFAs in bulk soil. An increase in substrate-derived C with time, sometimes leveling off at later times, was observed in all treatments for 18:2{omega}6,9 and cy19:0, and in 16:0 of the leached straw treatment (Fig. 4a, 4b, and 4f). The amount of substrate-derived C increased to a maximum and then declined for i15:0 in all treatments, and for 16:0 in the unleached straw and leachate treatments (Fig. 4b and 4c). The opposite trend, with substrate-derived C reaching a minimum and then increasing was seen for a15:0 in all treatments (Fig. 4d). No temporal trends were observed for 10Me16:0 in any treatment (Fig. 4e). It is interesting to note that often a lag in the temporal pattern of substrate-derived C was observed with the leached straw treatment compared with the unleached straw and leachate treatments.


   DISCUSSION
TOP
 ABSTRACT
 INTRODUCTION
 MATERIALS AND METHODS
 RESULTS
 DISCUSSION
CONCLUSIONS
 REFERENCES
 
Microbial Communities of Detritus and Bulk Soil
Despite clear temporal trends in bulk soil and detritusphere PLFAs when treatments were considered separately (Fig. 2 and 3), analysis of all samples together revealed one clear pattern: Detritusphere and bulk soil PLFA profiles formed two, distinct groups (Fig. 1). This grouping may have been caused by different microbial communities developing in the two microhabitats or may be an artifact of residual plant and microbial PLFAs remaining from the original straw.

Determining the influence of residual PLFAs is problematic, in part because there is no definitive way to differentiate between plant and microbial PLFAs associated with plant material. Plant PLFAs are generally less diverse than those of the microbial community, consisting of straight-chain PLFAs at least 16 C long with even numbers of C. For example, 16:0 and 16:1 made up more than half of the plant-associated PLFAs of grass and rice straws (Klamer and Bååth, 1998; Kimura et al., 2001; Nakamura et al., 2003). But some PLFAs found in plants are also found in microorganisms, including 18:2{omega}6,9, which is often used in microbial studies as a fungal marker. Thus, a single measurement of PLFAs from plant tissue cannot differentiate between plant and microbial PLFAs.

Temporal shifts in the mole percentage of PLFA are likely caused by changes in microbial communities during decomposition of plant material (Nakamura et al., 2003) and are relatively insensitive to residual PLFAs, although the degradation of plant PLFAs could also result in shifts in the mole percentage of PLFA. If extraction efficiencies are known, changes in absolute amounts of PLFAs, particularly increases in PLFAs, would be a more definitive indicator of shifts in microbial communities and the residual PLFAs from the first sampling time might even be subtracted as a way to isolate PLFAs of the colonizing microorganisms.

In this study we were not able to calculate absolute amounts of PLFAs because of highly variable extraction efficiencies; however, the fact that substrate-derived C of detritusphere PLFAs increased during the incubation (Fig. 4a and 4b) suggests that changes in the mole percentage of PLFA of detritusphere samples were due to microbial production of PLFAs rather than decomposition of plant PLFAs. Furthermore, it seems unlikely that any residual PLFAs remained by the end of the incubation because at 80 d about 25% of the substrate C had been respired (S.K. McMahon, unpublished data, 2003), which means that about half of the straw had decomposed, assuming a growth efficiency of 0.5. Consequently, given that the detritusphere samples from different times form a cluster distinct from the cluster of bulk soil samples (Fig. 1), we think it unlikely that residual PLFAs were the dominant influence on either the temporal shifts in the mole percentage of PLFA that we observed in the detritusphere or in discriminating between detritusphere and bulk soil microbial communities.

Microbial communities in bulk soil were characterized by bacterial PLFA markers, whereas those of the detritusphere were associated with the 18:2{omega}6,9 fungal marker. This is also reflected by fungal/bacterial PLFA ratios greater than 1.0 in detritusphere samples and low ratios between 0.17 and 0.33 in bulk soil (Table 1). These ratios fall within the wide range reported in the literature, with values as low as 0.01 (Bardgett and McAlister, 1999; Bååth and Anderson, 2003) and as high as 5.0 (Bailey et al., 2002). The greater relative amounts of fungi associated with the detritus is consistent with their role in degrading complex, polymeric C compounds (Swift et al., 1979; Neely et al., 1991; Schimel, 1995; Henriksen and Breland, 2002), and is further supported by the greater amount of substrate-derived C found in the fungal vs. bacterial markers (Fig. 4).

However, despite higher proportions of fungi in detritusphere compared with bulk soil communities, 18:2{omega}6,9 was often not significantly more enriched in 13C in detritusphere than in bulk soil. Because fungi form long hyphae, they are able to access and translocate C and nutrients over larger spatial scales than bacteria (Wessén and Berg, 1986; Frey et al., 2003; Klein and Paschke, 2004). Therefore, fungi found in the bulk soil may represent hyphae growing out from detritus or soil-based hyphae colonizing labeled straw and translocating C back into the soil. This is supported by the difference in the amount of substrate-derived C in 18:2{omega}6,9 of the leached straw compared with leachate and unleached straw treatments (Fig. 4): The fungal marker was highly labeled in treatments containing highly mobile, soluble C at 0.6 d but not until 15 d in the leached straw treatment.

Temporal Trends in Community Composition
We expected that the C amendments would stimulate microbial activity, resulting in growth and potentially changes in the composition of the microbial community. These temporal trends might represent a succession of microorganisms responsible for the sequential utilization of C substrates during decomposition.

Detritusphere
The clearest pattern observed for detritusphere samples was an increase in fungal colonization with time. During the first 1.6 d of the study, the 18:2{omega}6,9 fungal marker averaged 21 ± 1 mol% for the unleached straw detritusphere and 24 ± 1 mol% for the leached straw detritusphere; this marker averaged 32 ± 2 mol% for the last four sampling dates of each treatment. This is similar to increases observed by Nakamura et al. (2003) for rice straw decomposing in the field, and may represent colonization and growth of fungi capable of degrading the hemicellulose and cellulose left in the detritus once the readily degradable soluble C has been used (Reinertsen et al., 1984; Horwath and Elliott, 1996; Saviozzi et al., 1997). Bacterial PLFA markers were much more varied, however.

In unleached straw detritusphere, the fraction of bacterial PLFAs oscillated through time, being relatively high at 0.6, 18, and 50 d (Fig. 3). The high proportion of bacteria initially present may reflect the residual community on the straw (Tester, 1988). The flush of bacterial-associated lipids seen at 18 and 50 d may be in response to nutrients liberated by fungal activity early in the incubation.

Leached straw detritusphere PLFA profiles followed a different trajectory through time, with bacterial PLFAs being high only at 50 d. The lower proportion of bacterial PLFAs initially (Table 1) may have been caused by removal of surface bacteria during the leaching process and the absence of soluble C may have further slowed bacterial colonization. Thus, the delay in the development of a bacterial community seems reasonable, although the reason for its decline between 50 and 80 d is unclear. Nevertheless, a similar trend was seen in the unleached straw detritusphere, and may indicate a general decline in bacteria during later stages of decomposition.

The different microbial community dynamics observed in the leached and unleached straw detritusphere leads to interesting questions about the roles of bacterial and fungal communities during decomposition: Do these different groups of microorganisms work independently or cooperatively? If the latter, do they work in series or in consort?

Bulk Soil
There was no detectable shift in the PLFAs of control soil communities, which suggests that changes in other treatments are attributable to added C. During the early stage of decomposition, microbial communities in bulk soil from all amended treatments had similar PLFA profiles, being primarily characterized by higher mole percentage of 14-C and 15-C PLFAs, which are characteristic of bacteria (Myers et al., 2001). Bossio et al. (1998) also saw increases in these short branched-chain fatty acids in plots receiving large straw inputs. Because all three amended treatments had similar indicator PLFAs during this early stage of decomposition (Fig. 2), it does not appear that the type of C amendment had a major influence on structuring the microbial community at this time.

As decomposition continued, a distinct shift occurred in PLFA composition of the bulk soil communities in amended soils toward longer and more complex PLFAs (Fig. 2). This shift occurred between 1.6 and 15 d for leached straw and leachate, and between 15 and 18 d for unleached straw. Regardless of the exact timing of the shift, unleached straw and leachate communities were distinguished by increasing the mole pecentage of PLFAs associated with gram-negative bacteria (e.g., 16:1{omega}7 and cy19:0) as decomposition progressed (Fig. 2). This may reflect an increase in bacteria in response to the soluble C added in both of these treatments, and is consistent with the interpretations of others that bacteria are the primary users of monomeric C compounds (e.g., Paul and Clark, 1996; Myers et al., 2001).

The increase in cy19:0, which was also observed in the leached straw treatment, might also be a response to stress conditions (Bossio and Scow, 1998). In the case of the unleached straw and leachate treatments, cy19:0 might have been produced once the bacterial community ran out of the easily degradable, soluble C. With soluble C removed from leached straw, the bacterial community may have become more stressed as a result of competition with a growing fungal community.

Both unleached and leached straw treatments showed increasing ratios of fungal/bacterial PLFAs with time (Table 1), although the 18:2{omega}6,9 fungal marker was a significant indicator for only the leached straw treatment (Fig. 2). An increase in fungi as decomposition progresses is consistent with their commonly regarded role as the primary decomposers of more recalcitrant C (Schimel, 1995; Myers et al., 2001; Henriksen and Breland, 2002).

Microbial Community Processing of Substrate Carbon
Phospholipid fatty acids measured in this experiment were not enriched equally with 13C. Substrate quality influenced both the rate at which 13C was incorporated into PLFAs and the order in which PLFAs were labeled. Although most PLFAs in the detritusphere or bulk soil were already labeled at 0.6 d, the response was often greater for treatments containing soluble C (Fig. 4). The exception was cy19:0 of the leached straw. Such an increase in substrate-derived C can only be attributed to degradation and assimilation of labeled substrate degradation, because PLFAs closely reflect the signature of substrates used during synthesis (Abraham et al., 1998). These results are consistent with Hanson et al. (1999), who found 13C in all PLFAs after adding 13C-labeled glucose to soil. Generally, the more highly labeled the substrate, the more highly labeled the PLFAs; unleached straw and leachate were more enriched than leached straw until later time points when most of the leachate had been consumed.

Although both i15:0 and a15:0 are generally associated with gram-positive bacteria, a15:0 became labeled more quickly and to a larger extent (Fig. 4c and 4d) in bulk soil than i15:0. Several explanations are possible for this phenomenon. First, gram-positive bacteria are a highly diverse group of organisms containing such genera as aerobic Bacillus and Arthrobacter, and anaerobic Clostridium (Paul and Clark, 1996). Thus, organisms included in this group may have very different life strategies and, although they may all produce i15:0 and a15:0, they may do so in radically different proportions (Haack et al., 1994). Enrichment of a15:0 but not i15:0 may indicate that a subset of the gram-positive bacteria with higher proportions of a15:0 in their cell membranes were active during early decomposition. Second, it is possible that the same types of organisms produced both i15:0 and a15:0 but at different times during their lifecycle or under various growth conditions (Haack et al., 1994; Petersen et al., 2004). In this case, the high percentage of substrate-derived C seen in a15:0 could merely indicate that a15:0 is produced preferentially during periods of rapid growth, as Haack et al. (1994) found for Arthrobacter and Bacillus in pure culture.

10Me16:0, which is considered to be an indicator of actinomycetes, was fairly uniformly labeled in leachate throughout the experiment and increased slightly in unleached straw and leached straw through time (Fig. 4e). Actinomycetes are a diverse group of gram-positive bacteria capable of producing depolymerization enzymes for recalcitrant compounds, such as chitin, hemicellulose, and starches (Paul and Clark, 1996). Thus, they play an important role in the decomposition of straw resulting in more 13C incorporation into PLFAs than with leachate alone. Additional work with more discriminating methods would be required to determine if different types of actinomycetes respond to unleached straw and leached straw.

Of the lipids analyzed for 13C, cy19:0 was the most variable among treatments over time (Fig. 4f). Cyclopropyl fatty acids have been shown to increase in organisms under physiological stress (Bossio and Scow, 1998), although cy17:0 and cy19:0 have also been attributed to gram-negative bacteria (O'Leary and Wilkinson, 1988) and anaerobic bacteria (Zelles, 1997). Because the microcosms were maintained at or below water saturation significant anaerobic activity is unlikely. However, the state of rapid microbial growth and activity early in the incubation may have resulted in more gram-negative bacteria and/or nutrient shortages that induced production of cy19:0.

Effect of Soluble Carbon on Microbial Community Composition
We were surprised that there were no significant differences in the microbial communities among the four bulk soil treatments based on the mole percentage of PLFA (Fig. 1), but it is worth noting that PLFAs provide only broad groupings of microorganisms and are unable to differentiate among phylotypes of fungi or bacteria at finer taxonomic levels. Coupling PLFA analysis with other molecular approaches, such as LH-PCR or T-RFLP (e.g., Marsh, 1999; Ritchie et al., 2000; Dickie et al., 2002), could provide additional insights into microbial community dynamics in response to C amendments.

Nevertheless, the temporal shifts in PLFAs observed in each of the C-amended bulk soils suggest some dynamics in microbial communities during decomposition (Fig. 2). These dynamics appeared to be less influenced by soluble C, which was incorporated quickly into all PLFAs (Fig. 4), than by the presence of the more recalcitrant straw residue. At later stages of decomposition, there was a shift toward a greater relative abundance of fungal PLFAs in the straw-amended treatments (Table 1). Treatments containing straw residue often showed more substrate-derived C in PLFAs at the end of the incubation than the leachate treatment (Fig. 4), which suggests that the recalcitrant C had a longer impact on microbial communities than the soluble C did.


   CONCLUSIONS
TOP
 ABSTRACT
 INTRODUCTION
 MATERIALS AND METHODS
 RESULTS
 DISCUSSION
 CONCLUSIONS
REFERENCES
 
Phospholipid fatty acid methodology was adequate for identifying temporal trends in microbial community composition during decomposition of ryegrass straw. Contrary to our expectations, microbial succession appeared to be driven more by the presence of recalcitrant materials, rather than the presence of soluble substrates. Combining 13C and PLFA analysis provided new insights into microbial community dynamics associated with residue decomposition. Information was gained through the use of 13C about the spatial scale at which decomposition occurs: PLFAs extracted from detritusphere communities were much more highly labeled than those from bulk soil communities. These communities also proved to be compositionally different. Labeling of PLFAs in bulk soil community with 13C also raises interesting questions about C transfer from residue and the role of water soluble components in stimulating bulk soil versus residue-associated communities.


   ACKNOWLEDGMENTS
 
We thank Rockie Yarwood, Stephanie Boyle, Alicia Lyman-Holt, Justin Brant, Stacie Kageyama, and Naoyuki Ochiai for assistance with experimental set-up, maintenance, and sampling. The comments by three anonymous reviewers and the associate editor were most helpful. Shawna McMahon was funded by a PGS-A fellowship from the Natural Science and Engineering Research Council of Canada. This material is based on work supported by the National Science Foundation under Grant No. 0075777.

Received for publication August 28, 2004.


   REFERENCES
TOP
 ABSTRACT
 INTRODUCTION
 MATERIALS AND METHODS
 RESULTS
 DISCUSSION
 CONCLUSIONS
 REFERENCES
 




This article has been cited by other articles:


Home page
 Soil Sci.Home page
H. Minoshima, L. E. Jackson, T. R. Cavagnaro, and H. Ferris
Short-Term Fates of Carbon-13-Depleted Cowpea Shoots in No-Till and Standard Tillage Soils
Soil Sci. Soc. Am. J., October 29, 2007; 71(6): 1859 - 1866.
[Abstract] [Full Text] [PDF]

Home page
 Soil Sci.Home page
H. W. Kreuzer-Martin
Stable Isotope Probing: Linking Functional Activity to Specific Members of Microbial Communities
Soil Sci. Soc. Am. J., March 12, 2007; 71(2): 611 - 619.
[Abstract] [Full Text] [PDF]

Home page
 Soil Sci.Home page
E. A. Paul, S. J. Morris, R. T. Conant, and A. F. Plante
Does the Acid Hydrolysis-Incubation Method Measure Meaningful Soil Organic Carbon Pools?
Soil Sci. Soc. Am. J., April 19, 2006; 70(3): 1023 - 1035.
[Abstract] [Full Text] [PDF]

This Article
Right arrow Abstract Freely available
Right arrow Fig