- Email: [email protected]
Microscale distribution and function of soil microorganisms in the interface between rhizosphere and detritusphere
Soil Biology & Biochemistry 49 (2012) 174e183
Contents lists available at SciVerse ScienceDirect
Soil Biology & Biochemistry journal homepage: www.elsevier.com/locate/soilbio
Microscale distribution and function of soil microorganisms in the interface between rhizosphere and detritusphere Petra Marschner a, *, Sven Marhan b, Ellen Kandeler b a b
School of Agriculture, Food and Wine, Faculty of Sciences, The Waite Research Institute, University of Adelaide, Adelaide, SA, Australia Soil Science and Land Evaluation, Soil Biology Section, University of Hohenheim, Stuttgart, Germany
a r t i c l e i n f o
a b s t r a c t
Article history: Received 7 August 2011 Received in revised form 20 January 2012 Accepted 22 January 2012 Available online 8 March 2012
The rhizosphere and the detritusphere are hot spots of microbial activity, but little is known about the interface between rhizosphere and detritusphere. We used a three-compartment pot design to study microbial community structure and enzyme activity in this interface. All three compartments were filled with soil from a long-term field trial. The two outer compartments were planted with maize (root compartment) or amended with mature wheat shoot residues from a free air CO2 enrichment experiment (residue compartment) and were separated by a 50 mm mesh from the inner compartment. Soil, residues and maize differed in 13C signature (d13C soil 26.5&, maize roots 14.1& and wheat residues 44.1&) which allowed tracking of root- and residue-derived C into microbial phospholipid fatty acids (PLFA). The abundance of bacterial and fungal PLFAs showed clear gradients with highest abundance in the first 1e2 mm of the root and residue compartment, and generally higher values in the vicinity of the residue compartment. The d13C of the PLFAs indicated that soil microorganisms incorporated more carbon from the residues than from the rhizodeposits and that the microbial use of wheat residue carbon was restricted to 1 mm from the residue compartment. Carbon incorporation into soil microorganisms in the interface was accompanied by strong microbial N immobilisation evident from the depletion of inorganic N in the rhizosphere and detritusphere. Extracellular enzyme activities involved in the degradation of organic C, N and P compounds (b-glucosidase, xylosidase, acid phosphatase and leucin peptidase) did not show distinct gradients in rhizosphere or detritusphere. Our microscale study showed that rhizosphere and detritusphere differentially influenced microbial C cycling and that the zone of influence depended on the parameter assessed. These results are highly relevant for defining the size of different microbial hot spots and understanding microbial ecology in soils. Ó 2012 Elsevier Ltd. All rights reserved.
Keywords: 13 C C flow Enzymes Maize Microbial community composition Roots Wheat
1. Introduction In soils, the rhizosphere and the detritusphere are two hot spots of microbial activity due to the presence of easily available compounds from either roots or plant residues. Microbial density and enzyme activity are high close to the roots or residues and decrease with increasing distance, forming distinct gradients in millimetre scales. Moreover, microbial community structure changes with distance from roots or residues (e.g. Kandeler et al., 2001; Poll et al., 2006, 2008, 2010). Roots modify the rhizosphere by taking up nutrients and releasing various substances such as sugars, organic acid anions and amino acids which are easily available nutrient sources for
* Corresponding author. E-mail address: [email protected] (P. Marschner). 0038-0717/$ e see front matter Ó 2012 Elsevier Ltd. All rights reserved. doi:10.1016/j.soilbio.2012.01.033
microorganisms. Hence, compared to the bulk soil, the rhizosphere of active roots is characterized by depletion of nutrients such as P and a high abundance and activity of microorganisms (Foster, 1986; Joner et al., 1995; Chen et al., 2002; Wang et al., 2005) as well as by high activity of enzymes released by roots and microorganisms (Tarafdar and Jungk, 1987; Joner et al., 1995; Badalucco et al., 1996). Abundance and activity of microorganisms and enzyme activity decrease with distance from the root surface creating distinct gradients within a few millimetres (Tarafdar and Jungk, 1987; Kandeler et al., 2001). As the rhizosphere, the detritusphere is characterized by high concentrations of easily available compounds, particularly in the early stages of residue decomposition when water-soluble compounds are released (Bastian et al., 2009; Poll et al., 2010). After depletion of the water-soluble compounds, more complex compounds requiring specialized enzymes are decomposed more slowly (Theuerl and Buscot, 2010). Similarly to the rhizosphere
P. Marschner et al. / Soil Biology & Biochemistry 49 (2012) 174e183
there is a gradient of nutrient availability as well as of enzyme activity (Gaillard et al., 2003), microbial density and community composition with distance from the residues (Poll et al., 2006; Nicolardot et al., 2007; Bastian et al., 2009). Moreover, fungal community composition and abundance change over time as compounds within the residues are successively depleted (Poll et al., 2010). As a result of diffusion and mass flow as well as translocation via fungal hyphae and soil animals, residue C can be detected up to 4 mm from the residues with a greater distance in moist compared to dry soil (Gaillard et al., 2003; Poll et al., 2008). Depending on the C/N or C/P ratio of the residues, there may be net mineralization or net immobilisation in the microbial biomass in the residues, but also within the detritusphere several millimetres away from the residues (Moritsuka et al., 2004; Ha et al., 2007). Thus, chemical, physical and biological properties of the rhizosphere and detritusphere have been studied extensively, but separately. However, in the field, roots usually grow in the vicinity of decomposing plant residues where rhizosphere and detritusphere may meet resulting an interface that is influenced by both rhizosphere and detritusphere properties. The conditions in the interface may change over time as release of C from decomposing residues and roots change. Thus the relative dominance of rhizosphere or detritusphere properties in the interface may also change. Microbial communities characteristic for rhizosphere and detritusphere may compete and create new communities that are specific for this interface. We designed a three-compartment pot system with living roots on one side and soil with residues on the other, each separated by a fine mesh from the 5 mm-wide middle compartment by a 50 mm mesh which was filled with soil only. In this system the three dimensions of the rhizosphere and detritusphere were reduced to two dimensions to allow accurate smallscale sampling at different distance from roots and residues. The slicing of the middle compartment into 1 mm layers provided information about extent and possible interactions of rhizosphere and detritusphere. In each layer, enzyme activity and microbial community composition by PLFA were measured. The aims of this study were to (i) compare the changes induced by roots and residues and the extent of their zone of influence, (ii) assess if there is an overlap of their zones of influence when roots and residues are separated by 5 mm, and (iii) track the carbon flow from roots and residues into the microbial biomass by using soil, plants and residues with different 13C signatures and 13C-PLFA analyses. 2. Materials and methods 2.1. Experimental set up The experiment was conducted in a three-compartment pot system (diameter 100 mm, height 190 mm) with two equally-sized outer compartments, separated by a 50 mm polyamide mesh (SEFAR Nitex PA 6.6) from a 5 mm-wide middle compartment (Fig. 1). The outer compartments were filled with 664 g and the middle compartment with 99 g dry soil equivalent, sieved to 2 mm. The middle compartment consisted of three PVC frames with the outer two frames with the mesh and the middle frame defining the width of the compartment (5 mm). The three frames were held together by plastic binders threaded through holes in the frames. After filling the middle compartment with the defined moist soil weight, and assembling the ‘sandwich’, the frames were placed into the pots and the outer compartments immediately filled with moist soil. In order to track C from plants and residues into the microbial biomass, we used soil, plants and residues with differential d13C values. The soil was an agricultural topsoil (Chernozem; Corg 1.5%;
175
Fig. 1. Design of pots with two outer compartments and a middle compartment (5 mm wide) representing the interface between the two outer compartments.
Ntotal 0.13%; pH (CaCl2) 7.0; inorganic N 26 mg kg1; PCAL 93 mg kg1; KCAL 137 mg kg1; soil texture: sand 12%, silt 66%, clay 22%, water-holding capacity 381 g H2O kg1) from a long-term field trial in Bad Lauchstädt (Germany) with crop rotations including only C3 plants. The soil was sieved to 2 mm mesh size and frozen two times (20 C) for defaunation before storage at 4 C. The initial soil water content was 51% water-holding capacity and the bulk density was 1.2 g cm3. The mature wheat residues (C/N 74, water-soluble C and N 23 and 0.6 g kg1, respectively) were obtained from a free air CO2 enrichment (FACE) experiment conducted at the University of Hohenheim (Erbs and Fangmeier, 2005). The shoot residues were cut into 2 cm length and mixed thoroughly into the soil of one outer compartments at a rate of 7.5 g kg1 dry soil. Pre-germinated maize (Zea mays L., cv Amadeo) was planted in the other outer compartment. The d13C values were: wheat residues 44.1&, soil 26.5& and maize roots 14.1&. There were four treatments which differed in configuration of the outer compartments: (1) unamended soil-unamended soil (soilesoil), (2) soil with wheat residues-unamended soil (residue-soil), (3) maize-unamended soil (maize-soil), (4) maize-soil with wheat residues (maize-residue). The pots were sealed and placed in a water bath in order to maintain the soil temperature at 20 C. The water bath was situated in a glasshouse with ambient light and temperature (summer). Soil moisture was maintained by weight daily. Despite watering daily to weight, the soil in the lower parts of the pots with plants dried out because of strong water uptake by the roots, particularly in the second growth phase from 14 to 23 DAP.
176
P. Marschner et al. / Soil Biology & Biochemistry 49 (2012) 174e183
Between 14 days after planting (DAP) and 23 DAP, some fine roots grew through the mesh or gaps between the frames of the middle compartment thus the middle compartment was not entirely root-free. 2.2. Sampling Four replicates of each treatment were harvested on 14 DAP and 23 DAP. The first sampling date was chosen to allow sufficient root growth to occur for the development of a rhizosphere at the mesh of the middle compartment. The second sampling was selected to have a denser root mat at the mesh and greater decomposition of the wheat straw. Soil from the outer compartment was collected after removing roots or residue particles. The middle compartment was frozen at 20 C and sliced into 1 mm vertical slices using a kitchen knife. The layers 0e1, 1e2, 2e3, 3e4 and 4e5 mm refer to the distance in mm from the outer root compartment if plants were present (maize-soil and maize-residue) or the residue compartment in the residue-soil treatment. In the soilesoil treatment the side of the first slice was selected randomly. To obtain layers with the correct distance from the mesh, the middle compartment was placed horizontally on a flat surface and the top frame with the mesh and the middle frame were removed leaving the soil rectangle resting on the mesh of the bottom frame. Then PVC frames of a certain thickness were placed around the soil rectangle. For the first slice, the frame thickness was 4 mm (thus obtaining the layer 0e1 mm from the top mesh). After removing the soil above this frame, the frame was replaced by one of 3 mm thickness and the process of removing the soil repeated. This was repeated with frames of 2 and 1 mm thickness. Any soil remaining after removal of the 1 mm frame was defined as from the 4e5 mm layer. Care was taken to minimize the mixing of soil from different layers during slicing, but some mixing may have occurred, particularly between layers 3e4 and 4e5 mm. The soil from each layer was mixed and then stored at 20 C until analyses. Maize shoot and root dry weight were determined after drying at 50 C for 48 h. Soil water content of each sample was estimated by drying samples at 105 C for 24 h.
i16:0, and i17:0 are predominantly found in Gram-positive, and cy17:0 and cy19:0 in Gram-negative bacteria. The PLFAs 18:1u9 and 18:2 u 6,9 c or t were used as marker for fungal biomass (Kandeler, 2007). The peaks of the fungal fatty acids 18:2u6t, 18:1u9 and 18:3u3 could not be separated. In the following, the sum of these three fatty acids will be referred to as 18:1u9. Isotopic ratios of the PLFAs were determined using a GC (HP 6890, Agilent Inc., USA) coupled via a combustion III Interface (Thermo Finnigan, Waltham, USA) to a Delta Plus XP mass spectrometer (Thermo Finnigan, Waltham, USA). Carbon dioxide of known isotopic composition was injected at the beginning and at the end of each run, which was finished by a re-oxidation of the combustion interface with O2 for 10 s. Isotopic ratios of SOC and PLFAs were reported relative to the Pee Dee Belemnite (PDB) standard:
d13C (&) ¼ [(Rsam/Rstd) 1] 1000 where Rsam is the 13C/12C ratio of the sample and Rstd is the C/12C ratio of the standard (PDB). During the methylation of the PLFAs, a methyl group with a d13C different from that of the fatty acid is added. For the correction of the d13C of the PLFAs, the equation according to Abraham et al. (1998) was used.
13
2.3.3. Microbial biomass C and N Microbial biomass carbon (Cmic) and nitrogen (Nmic) were assessed using the fumigationeextraction method according to Vance et al. (1987) only for the soil from the outer compartments because the amount of soil in many of the layers from the middle compartment was not sufficient. 2.3.4. Inorganic N Inorganic N was extracted from 2 g moist soil from the different layers and the outer compartment with 20 mL 2M KCl (30 min at 250 rpm on a horizontal shaker) and the extracts centrifuged at 4070 g for 30 min. Ammonium and nitrate in the supernatant were determined with an autoanalyzer (Bran & Luebbe, Germany). þ Inorganic N was calculated as the sum of NO 3 and NH4 .
2.3. Analyses 2.3.1. Enzyme activity The activities of the following enzymes were measured as described in Marx et al. (2005): b-D-glucosidase (EC 3.2.1.21), xylosidase (EC 3.2.1.37), N-acetyl-glucosaminidase (EC 3.2.1.52), phosphatase (EC 3.1.3.1) and leucine peptidase (EC 3.4.11.1). Substrates containing the fluorescent compounds 4-methylumbelliferone (4-MUF) and 7-amino-4-methyl coumarin (7-AMC), standards and buffer (MES-buffer 2-[N-orpholino]ethanesulphonic acid, Trizma buffer (mixture of a-Tris-(hydroxymethyl)-methylamin and Tris(hydroxymethyl)-aminomethane hydrochloride)) were obtained from SigmaeAldrich (St. Louis, USA). 2.3.2. Phospholipid fatty acid extraction and analysis Phospholipid fatty acids were extracted from 4 g of soil using the procedure described by Frostegard et al. (1993). Methylnondecanoate (19:0) was used as internal standard. The fatty acid methyl esters were identified by chromatographic retention time comparison with a standard mixture composed of 37 different PLFAs that ranged from C11 to C24 (SigmaeAldrich, St Louis). To ensure correct identification of PLFAs (chain length and saturation), a random selection of samples were analysed by GCMS. The sum of the following PLFAs was used to estimate bacterial biomass: i15:0, a15:0, i16:0, 16:1o7, i17:0, cy17:0, 18:1u7, cy19:0 (Frostegard et al., 1993; Zelles et al., 1995). The PLFAs i15:0, a15:0,
2.3.5. Shoot N Total shoot N was determined after grinding with a CeN analyser (Carlo Erba, Germany). 2.4. Statistical analyses Significant differences in microbial parameters between harvests, treatments and layers were determined by ANOVA followed by Tukey test. For non-metric multidimensional scaling (MDS), PLFA data was log (x þ 1) transformed. The PLFA data were log (x þ 1) transformed to balance the contributions of fatty acids by down-weighing the dominant fatty acids and increasing the weighting of rare fatty acids (Clarke and Warwick, 2001). Transformations and MDS were carried out using Primer E software (Primer-E Ltd, Plymouth Marine Laboratory, Plymouth, UK). Unlike principle component analysis plots, MDS plots do not have axis units or labels. The 2D stress indicates how well the plot represents the variability in the data with a 2D stress <0.2 considered to represent a good reflection of the resemblance matrix (Clarke and Warwick, 2001). Significant differences in microbial community structure between treatments were determined by PERMANOVA (Primer-E Ltd, Plymouth Marine Laboratory, Plymouth, UK). The DistLM procedure in Primer was used to determine the environmental variables (soil moisture, inorganic N concentration, enzyme activities) and fatty acids which
P. Marschner et al. / Soil Biology & Biochemistry 49 (2012) 174e183
explain the observed patterns. Significant differences refer to P 0.05.
A
177
80
(i) soil-soil
(ii) residue-soil
60 14 DAP 23 DAP
3. Results 40
3.1. Plant growth Inorganic N (mg kg-1)
20
Plant shoot and root dry matter increased 2.5 fold from 14 DAP to 23 DAP and did not differ significantly between the treatments (Table 1). The shoot N concentration at 23 DAP was 13e18 g kg1 which is low to adequate (Marschner, 2012). On 14 DAP, maize had formed a few roots close to the mesh of the middle compartment; by the second harvest, 23 DAP, a dense root mat had formed in the lower half of the pot. A few fine roots had grown though the mesh or the gaps in the frame extending several mm into the middle compartment; but since their diameter and their density was very low (5e10/pot), the effect is considered to be negligible. Soil water content at sampling varied between 38 and 51% water-holding capacity, with lower values in treatments with plants, suggesting that although the water content of the pots was maintained by weight, the water did not penetrate sufficiently into the lower part of the pots to compensate for water uptake by the roots.
0 soil
0-1
1-2
2-3
3-4
4-5
soil
resid
0-1
1-2
2-3
3-4
4-5
soil
80
(iv) maize-residue
(iii) maize-soil 60
40
20
0 maize 0-1
B 25
1-2
2-3
3-4
4-5
soil
maize 0-1
1-2
2-3
3-4
4-5
resid
(ii) residue-soil
(i) soil-soil
20 15 10
3.2. Inorganic N
Sum PLFA (ng g-1)
5
The inorganic N concentrations were low, ranging between 0 and 60 mg g1 (Fig. 2A). They were highest in the soilesoil treatment, where there were no differences in N concentration between layers. Compared to the soilesoil treatment (Fig. 2A (i)), residues and roots decreased inorganic N concentrations in the middle compartment as well as in the outer compartments.
0 soil
0-1
1-2
2-3
3-4
4-5
soil
resid
0-1
1-2
2-3
3-4
4-5
soil
25
(iii) maize-soil
(iv) maize-residue
20 15
3.3. Microbial biomass C and N and enzyme activity
10
Microbial biomass C (Cmic) and N (Nmic) were only determined in the outer compartments. At both sampling times, compared to the outer compartment with soil only, Cmic was about 50% higher in the outer compartment with maize and more than twice as high in the outer compartment with residues (Table 2). However, there was no difference in Cmic between the outer compartments with maize and residues. At 14 DAP, Nmic in the outer compartments with residues or maize was about three times higher than in outer compartment with soil only (Table 2). At 23 DAP, Nmic in the outer compartment with residues was more than 2-fold higher than in the outer compartment with soil only or with maize. There were no significant differences in Cmic/Nmic ratio among treatments. The activity of all enzymes was highly variable, but tended to be higher in the outer compartments with maize or residues compared to that with soil only (Table 2) and in layer 0e1 mm;
Table 1 Shoot, root and total dry weight of maize grown in pots with unamended soil (maize-soil) or soil amended with wheat residues (maize-residue) in the other outer compartment at 14 and 23 DAP (n ¼ 4). Values in the same column followed by different letters are significantly different (P 0.05). Shoot g pot1
Root g pot1
Total g pot1
14 DAP
Maize-soil Maize-residues
0.77 a 0.72 a
0.34 a 0.38 a
1.10 a 1.10 a
23 DAP
Maize-soil Maize-residues
1.71 b 1.56 b
0.97 b 1.07 b
2.68 b 2.63 b
5 0 maize 0-1
1-2
2-3
3-4
4-5
soil
maize 0-1
1-2
2-3
3-4
4-5
resid
Distance from mesh (mm) Fig. 2. Inorganic N (A) and sum of PLFAs (B) in the outer compartments and different layers in the middle compartment of the treatments (i) soilesoil, (ii) residue-soil, (iii) maize-soil and (iv) maize-residue 14 and 23 DAP (n ¼ 4). Vertical lines indicate standard error.
there was no consistent gradient in enzyme activity in the middle compartment (data not shown). 3.4. Microbial community composition The sum of bacterial and fungal fatty acids, the bacteria to fungi (B/F) ratio and the sum of all PLFA varied between layers in the soilesoil treatment, but there was no consistent trend (Figs. 2 and 3). The sum of bacterial and fungal fatty acids (Fig. 3A, B) and the sum of all PLFA (Fig. 2B) were increased in the vicinity of the residues and the maize roots with generally higher concentrations in the vicinity of the residues. The zone of higher concentrations extended 2e3 mm from the outer compartments, with the zone being wider for the residues than for the maize roots. Although roots and residues increased the concentrations of bacterial and fungal fatty acids, the increase was greater for the fungal fatty acids, thus the B/F ratio was reduced in the vicinity of residues and maize roots (Fig. 3C). In the maize-residue treatment ((iv) in Fig. 3), there was no overlap of gradients for the bacterial fatty acids; however the gradients overlapped for the fungal fatty acids.
178
P. Marschner et al. / Soil Biology & Biochemistry 49 (2012) 174e183
Table 2 Microbial biomass C, N C/N ratio and activities of b-D-glucosidase, xylosidase, N-acetyl-glucosaminidase, phosphatase and leucine peptidase in the outer compartments without plants or residues, with residues or with plants at 14 and 23 DAP (n ¼ 4). Values in the same row followed by different letters are significantly different (P 0.05). Outer compartment
Cmic (mg g1) Nmic (mg g1) Cmic/Nmic
14 DAP
23 DAP
Soil only
Residues
Maize
Soil only
Residues
Maize
23.8 a 2.2 a 10.8 a
61.6 c 8.7 c 7.1 a
41.8 ab 4.9 b 8.5 a
26.9 a 3.4 ab 7.8 a
59.9 bc 8.3 c 7.2 a
45.8 bc 5.1 b 8.9 a
122 a 14 a 18 a 48 a 147 b
157 abc 19 a 31 ab 107 c 178 c
204 c 14 a 29 a 66 ab 149 ab
125 ab 12 a 25 a 38 a 116 a
174 bc 43 a 50 b 69 ab 136 ab
176 c 22 a 36 ab 90 bc 150 bc
Enzyme activity (nmol g dw h1)
b-D-glucosidase Xylosidase N-acetyl-glucosaminidase Phosphatase Leucine peptidase
PERMANOVA of all PLFA data showed that the main effects (treatment, layer and sampling time) as well as all interactions were significant, with the interactions sampling time layer and treatment sampling time layer as well as the main effect layer being the most important factors. When the treatments were analysed separately (Fig. 4), PERMANOVA showed that the interaction layer sampling time was more important than the main effects. In the soilesoil treatment (Fig. 4A), there were no significant differences in microbial community composition between the layers 1e5 mm at 14 DAP; but at 23 DAP, the community composition of outermost layers (0e2 mm and 4e5 mm) differed significantly from that in the layers 2e4 mm. There was no consistent gradient or change in microbial community composition with distance from the outer compartments. At 14 DAP in the residue-soil treatment (Fig. 4B), the microbial community composition of the outer compartment with residues and the layers 0e1, 1e2 and 2e3 mm differed significantly from each other and the community composition of layers 3e5 mm and the outer compartment with soil only. At 23 DAP, the two outer compartments and the layers 0e1 and 4e5 mm had distinct communities and the layer 1e2 mm differed from that of the layer 3e4 mm. The microbial community composition of the layers 0e1 and 1e2 mm in the maize-soil treatment (Fig. 4C), was similar, but differed from that of the other layers. The community composition of the remaining layers were distinct at 14 DAP, whereas at 23 DAP, the communities of layers 2e3 and 3e4 mm were similar. At both sampling times, the community composition of the outer compartment with maize and layers 0e2 mm was different, and distinct from the other layers as well as from the outer compartment with soil only. In the maize-residue treatment (Fig. 4D), at 14 DAP, the community composition of the layers 1e4 mm was similar, whereas at 23 DAP, the microbial community composition of each layer was distinct. At 23 DAP, the microbial communities of the outer compartment with maize and layers 0e2 mm differed from all others and also the communities in the outer compartment with residues and the layers 3e5 mm were distinct from all other layers. Comparison of the community composition of the different layers in the maize-residue treatment with the respective layers in the residue-soil and the maize-soil treatment showed that roots and residues maintained their distinct communities in the interface up to 2 mm from the mesh (data not shown). However, the community composition of the middle layer maize-residue treatment (3 mm from the roots and the residues) did not differ from the same layer in both the residue-soil and the maize-soil treatments, suggesting that the community is a mixture of detritusphere and rhizosphere microbes.
The concentrations of the following fatty acids were consistently increased in the vicinity of the residues and of the roots, a15:0, 16:0 (Gram-positive bacteria), 17:0 (bacteria), 18:2u6 and 18:1u9 (fungi), with higher concentrations in the vicinity of the residues (data not shown). 3.5. d13C of PLFA Of the 22 detected PLFAs, only 9 (i15:0, a15:0, 15:1, 16:0, 16:1, 16:1u7, cy17:0/17:1/i17:0, C18:0 and 18:1u9) showed gradients in d13C values (examples shown in Fig. 5). In all cases, the d13C values in 1 mm distance from the outer compartment with residues were more strongly depleted than those in greater distance. For some fatty acids, the depletion zone extended up to 2e3 mm: i15:0, 16:0, i16:0, 18:0 and 18:1u9 in the residue-soil treatment and i15:0, a15:0, cy17:0 and 18:1u9 in the maize-residue treatment. The degree of depletion varied between fatty acids, 15:1, 16:0, 16:1, i16:0, cy17:0 and 18:1u9 were more strongly depleted than the others. The fatty acids 16:0 and i16:0 are indicators of Grampositive bacteria, cy17:0 of Gram-negative bacteria and 18:1u9 is a fungal signature fatty acid. Thus, no particular group of microorganisms was more strongly depleted than others. The extent of the depletion zone varied slightly between the two sampling times, but there was no consistent effect of sampling time. The d13C values in the vicinity of the root compartment were only slightly enriched compared to the outer compartment with only soil (data not shown). Based on the two-compartment mixing model, the contribution of maize root C to PLFA C in the middle section was very small, ranging from 0 to 7.2% (data not shown). For most PLFAs, maize root C contribution was zero, only for the PLFAs i15:0, a15:0, 15:1, 16:1u7 and 18:1u9 a slight contribution (<2%) in the 0e1 mm layer was found. The contribution of residue C to PLFA C was generally higher than that of the maize roots and could be detected up to 5 mm from the residue interface. For some PLFAs there was a clear gradient, with a high percentage contribution of residue C in the 0e1 mm layer which decreased with increasing distance from the residue compartment (Table 3). In the 0e1 mm layer, the percentage contribution ranged between 5 and 20%, with lowest values in 15:0 and highest values in 16:1. At a distance from the residue interface greater than 3 mm, the percentage contribution was usually <5%. 4. Discussion The experiment showed microscale gradients and C fluxes in the detritusphere and rhizosphere. This experimental design allows direct comparison of rhizosphere and detritusphere effects and assessing whether their zones of influence overlap. In general, roots
P. Marschner et al. / Soil Biology & Biochemistry 49 (2012) 174e183
A
different d13C, we showed that residue C is incorporated more strongly than root C into the microbial biomass up to 3 mm from the residues/roots.
30
(ii) residue-soil
(i) soil-soil
25
179
20
4.1. Extent of rhizosphere and detritusphere
Bacterial fatty acids -1 (ng g )
15 10 5 0 soil
0-1
1-2
2-3
3-4
4-5
soil
resid 0-1
1-2
2-3
3-4
4-5
soil
30
(iv) maize-residue
(iii) maize-soil
25 20 15 10 5 0
B
6
14 DAP 23 DAP
4 3
Fungal fatty acids -1 (ng g )
(ii) residue-soil
(i) soil-soil
5
2 1 0 soil
0-1
1-2
2-3
3-4
4-5
soil
resid 0-1
1-2
2-3
3-4
4-5
soil
6
(iv) maize-residue
(iii) maize-soil
5 4 3 2 1 0
C
16
(i) soil-soil
14
(ii) residue-soil
12 10 8 6
B/F ratio
4 2 0 soil
0-1
1-2
2-3
3-4
4-5
soil
resid 0-1
1-2
2-3
3-4
4-5
soil
16
(iii) maize-soil
14
(iv) maize-residue
12 10 8 6 4 2 0 maize 0-1
1-2
2-3
3-4
4-5
soil
maize 0-1
1-2
2-3
3-4
4-5 resid
Distance from mesh (mm) Fig. 3. Sum of bacterial (A), fungal signature fatty acids (B) and bacteria/fungi ratio (C) in the outer compartments and different layers in the middle compartment of the treatments (i) soilesoil, (ii) residue-soil, (iii) maize-soil and (iv) maize-residue 14 and 23 DAP (n ¼ 4). Vertical lines indicate standard error.
and residues had similar effects on the measured parameters. For some properties (fungal abundance, B/F ratio, N depletion), their gradients overlapped, whereas for most of the measured parameters, the two zones of influence remained separate despite the close vicinity of root and residues. Using plants, residues and soil with
In agreement with previous studies, we found that roots and residues stimulate enzyme activity, microbial abundance and change microbial community composition, e.g. (Tarafdar and Jungk, 1987; Kandeler et al., 2001; Gaillard et al., 2003; Poll et al., 2006, 2008). However, the zone of influence varied with the parameter assessed, being narrower for enzyme activity and wider for changes in microbial community composition and inorganic N concentration. In the present study, increased enzyme activity was confined to the outer compartment and the first mm from the mesh. The narrow zone of increased enzyme activity is in agreement with Kandeler et al. (2001) who reported high activity of invertase, xylanase, acid phosphatase and protease only in <1 mm distance from the root surface in a soil of similar texture. On the other hand, gradients in enzyme activity have been found up to 3 mm (Poll et al., 2006) in a clay-rich soil and 4 mm (Gaillard et al., 1999) from the residue interface in a sandy soil. Thus, the width of the zone of increased activity does not seem to be related to soil type, but could be due to differences in root exudation and residue decomposition among the studies. Root and residue C was detected in the PLFAs up to 1 and 3 mm from the root and residue compartment, respectively, suggesting that the narrow zone of increased enzyme activity in the detritusphere can not be explained by lack of C flow from the residues. Previous studies have shown that microbial community composition changes with distance from the root surface (Kandeler et al., 2001) and the residue-soil interface (Nicolardot et al., 2007; Poll et al., 2010). In agreement with these studies, microbial community composition in the present study differed between layers and even the community in the layers up to 3 mm from the mesh often differed from that of the outer compartments (Fig. 4) with a similar extent in the vicinity of the roots and the residues. Indeed, in the maize-residue treatment, roots and residues maintained their distinct communities up to 2 mm from the mesh, whereas the community in the middle layer (3 mm from both residues and roots) was a mixture of the two communities. In most soils, carbon availability limits the growth of microorganisms (De Nobili et al., 2001; Hoyle et al., 2008). Hence, the availability of C is probably the main driver of differential microbial biomass and community composition in the various layers. Indeed, a model including carbon availability, respiration and microbial biomass C was able to accurately predict changes in microbial biomass and community composition in the detritusphere (Ingwersen et al., 2008). Differences in soil moisture between rhizosphere and detritusphere alone are unlikely to explain the differential community composition in the vicinity of maize roots and residues, because when comparing the percentage explanation of factors such as N availability, enzyme activity and soil moisture for patterns in microbial community composition, soil moisture had little importance, explaining not more than 10% of the variability. In agreement with other studies (Moritsuka et al., 2004; Appuhn and Joergensen, 2006; Poll et al., 2008), microbial biomass (sum of PLFAs, concentration of bacterial and fungal fatty acids) was increased up to 4 mm from the residues and maize roots (Figs. 2 and 3). The increased microbial growth can also be explained by the supply of relatively easily decomposable C (compared to native soil organic matter) from maize roots and residues. This is supported by the greater microbial biomass C in the outer compartments with residues or maize roots compared to the control soil (Table 2). In the
180
P. Marschner et al. / Soil Biology & Biochemistry 49 (2012) 174e183
14 DAP
2 44 3
S S S 22 5 45 5 3 3 5 1
23 DAP S
1
3 3
1
soil-soil
5
1
32 S
3
S
S S
S
S
4 2 4 2
R RR
2
11
5
R
5
R
3
33 3
M
1
2 2 1 22 1 M 44 4 4 5 SS 5 5 S 5
M
2D Stress: 0.04
M 2 1 34 2 34 4 2 2 3 3 4
C
R
R
2D Stress: 0.05 1
1
1 1
5
5
S
5
4 44 4 32 23 2 3
S
R
5 5
residue-soil
5 5
2D Stress: 0.03
11
5
1
2D Stress: 0.1
1
5 44 2
S
2 S 22
1 1 1
3
A
B
2
44
1
S
3
4
3
33 3
S
4
4 2
M M MM
1 1
5 S
5 5
S
5
S
maize-soil
S
2D Stress: 0.07
S
2D Stress: 0.06 4
33 3 3
2
11 1
R R R
5
24 2 4 4 2
R
5 5 5
D maize-residue
M M M
M
2D Stress: 0.06
2 33 2 1 2 1 1 32 1 3
4 44 M M M M
5 5 55 R
R RR
2D Stress: 0.03
Fig. 4. Non-metric multidimensional scaling plots of microbial communities in the outer compartments and different layers in the middle compartment of the treatments soilesoil (A), residue-soil (B), C: maize-soil (C) and maize-residue (D) 14 and 23 DAP (n ¼ 4). Outer compartments are indicated by S, R and M for soil only, residues and maize, respectively. Shaded areas encircle communities that are not significantly different.
vicinity of the residues, the abundance of fungi was more strongly increased than in the rhizosphere (Fig. 3B). This may be due to the low concentration of residues in the outer compartment. With their hyphae, fungi would be better able to grow towards the residue particles and transport C and other nutrients into to the adjacent layers than bacteria (Poll et al., 2006). The zone of influence of roots and residues was widest for inorganic N depletion. Compared to the outer compartment with soil only and the soilesoil treatment, inorganic N was depleted in the vicinity of residues and roots (Fig. 2A). This can be explained by immobilisation of N in the microbial biomass and, in the
rhizosphere, uptake of N by the roots. The high C/N ratio of the wheat residues (C/N 74) would have resulted in net N immobilisation by the microbial biomass. The depletion of inorganic N occurred up to 5 mm from the residues and the roots in the residuesoil and maize-soil treatments, which is further than the increase in microbial biomass (up to 3 mm). This may be explained by diffusion of N (particularly nitrate) to the microorganisms. In the treatments with maize, mass flow induced by water uptake by the roots would have further resulted in movement of N towards the root surface. Due to the differential extent of the zone of influence of roots and residues for the assessed parameters, the two zones
P. Marschner et al. / Soil Biology & Biochemistry 49 (2012) 174e183
A
i15:0
-22
-22 -24
-24
-26
-26 -28
-28
C
-32
13
-30
14 DAP 23 DAP
-30
-34
-32 -34
(i) residue-soil resid
0-1
1-2
2-3
3-4
4-5
resid 0-1
soil
-20
-20
-22
-22
-24
-24
-26
-26
-28
-28
-30
-30
-34
(ii) maize-residue
-36 2-3
3-4
4-5 resid
cy17:0
3-4
4-5
soil
(ii) maize-residue maize 0-1
D
-22
-24
-24
-26
-26
-28
-28
-30
-30
-32
-32 -34
(i) residue-soil 1-2
2-3
3-4
4-5
-20
-22
-22
-24
-24
-26
-26
-28
-28
-30
-30
-32
-32 -34
(ii) maize-residue 1-2
2-3
3-4
3-4
4-5 resid
(i) residue-soil resid 0-1
soil
-20
maize 0-1
2-3
-36
-36 resid 0-1
1-2
18:1ω9
-20
-22
-34
13C
1-2
-20
-36
2-3
-36 maize 0-1
-34
1-2
-32
-32
C
(i) residue-soil
-36
-36
-34
15:1
B -20
-20
181
1-2
2-3
3-4
4-5
soil
(ii) maize-residue
-36 4-5 resid
Distance from mesh (mm)
maize 0-1
1-2
2-3
3-4
4-5 resid
Distance from mesh (mm)
Fig. 5. d13C concentrations in selected fatty acids (A: a15:0, B: 15:1, C: cy17:0, D: 18:2u6t/18:1u9/18:3u3) in different layers in the middle compartment of the treatments (i) residuesoil and (ii) maize-residue 14 and 23 DAP (n ¼ 4). Dotted line indicates d13C of the fatty acid in outer compartment with soil only. Vertical lines indicate standard error.
overlapped only for inorganic N depletion, abundance of fungal fatty acids, B/F ratio and microbial community composition. Inorganic N is mobile in soils and can move by diffusion and mass flow. Unlike bacteria, fungi can exploit a large soil volume by transporting carbon from the source to the tips of hyphae (Frey et al., 2003). Bacteria and also enzymes are less mobile which may explain why their zone of influence did not overlap. 4.2. Differential extent of rhizosphere and detritusphere Root exudates include a wide range of easily decomposable compounds (Neumann, 2007) which are considered to be more readily available than C in mature wheat straw, therefore we had expected that the root would exert a greater effect than the residues. However, there was no difference in extent of the zone for most parameters and the reverse was true for C flow, where more
residue C was incorporated than root C and the zone of incorporation of residue C was greater (Fig. 5, Table 3). Residue and maize root C in the PLFAs may have been taken up by the microbes directly as it diffused into the middle compartment. However, secondary uptake of C from dying microorganisms into actively growing microbes can not be ruled out, particularly at 23 DAP when turnover of biomass is likely to have occurred. The maize plants grew well and should have therefore released substantial amounts of exudates into the rhizosphere. The relatively limited incorporation of maize C into the microbial biomass may be explained by three factors: lower water content in the vicinity of the roots, N availability and priming effect. The soil water content was lower in outer compartment with maize (38% water-holding capacity) compared to 47e50% waterholding capacity in outer compartments with residues or soil only. Despite daily watering to weight, the soil in the lower part of the
182
P. Marschner et al. / Soil Biology & Biochemistry 49 (2012) 174e183
Table 3 Contribution of residue 13C to selected PLFAs in different layers in the middle compartment of the treatments residue-soil and maize-residue 14 and 23 DAP (n ¼ 4 standard error of the means). DAP
0e1 mm
1e2 mm
2e3 mm
14
Residue-soil Maize-residue
i15:0 4.71.0 1.70.2
3e4 mm
4e5 mm
3.60.7 1.80.4
3.40.4 3.40.7
2.20.5 4.51.4
0.90.4 7.31.7
23
Residue-soil Maize-residue
6.70.4 0.00.7
3.40.3 1.60.2
3.40.5 0.60.8
1.10.6 7.61.3
4.21.2 8.73.4
14
Residue-soil Maize-residue
cy17:0 1.21.6 0.00.5
0.01.4 0.00.7
0.01.0 0.00.8
0.00.6 0.03.7
0.01.6 8.44.7
23
Residue-soil Maize-residue
12.22.7 0.00.7
4.30.9 2.01.4
5.31.0 3.61.0
1.30.4 10.54.4
3.21.1 12.06.3
14
Residue-soil Maize-residue
18:1u9 10.92.4 0.00.2
3.40.7 0.00.7
1.84.2 1.00.2
0.00.6 2.02.4
0.00.7 10.33.4
28
Residue-soil Maize-residue
12.70.9 0.00.3
5.20.8 3.70.7
6.50.9 4.70.4
3.60.3 8.62.5
2.20.5 9.55.0
compartments with plants dried out due to water uptake by the roots. This was also the case, to a lesser extent, for the soil in the middle compartment. Diffusion of soluble compounds occurs in water films/water-filled pores. Poll et al. (2008) showed that diffusion of C is restricted by low water content. Thus, exudates released by the roots may not have entered the middle compartment and moved within it because of low diffusion rates. Since the soil in the residue compartment and also in the adjacent layers in the middle compartment remained moister, diffusion would have been greater than in the root compartment (Poll et al., 2008). The strong effect of the residues was surprising because of the low density and the large particle size of the residues. Moreover, even at 23 DAP, the wheat residue particles appeared to be intact and only slightly discoloured. In the vicinity of the residues, the abundance of fungi was more strongly increased than that of bacteria (Fig. 2). This may be due to the low concentration of residues in the outer compartment. With their hyphae, fungi would be better able to grow towards the residue particles and transport C and other nutrients into to the adjacent layers than bacteria (Poll et al., 2006). However, the differential water content can not fully explain the limited diffusion of root C compared to residue C as the extent of the zone of influence for other parameters such as microbial community composition and abundance of bacterial PLFA was similar in the vicinity of roots and residues. Nitrogen availability may have limited the ability of microorganisms in the vicinity of the roots to utilise C from exudates. Inorganic N concentrations were decreased to the same extent in the rhizosphere and in the detritusphere (Table 2); but whereas only immobilisation of N into the microbial biomass can explain the decrease in the detritusphere, N would have been taken up by plants and microbes in the rhizosphere. This argument is supported by the lower microbial biomass N in the root than in the residue compartment, although this difference was not statistically significant. Root exudates may have induced a priming effect (Kuzyakov, 2002), i.e. increased decomposition of native organic matter to a greater extent than compounds from the poorly degradable wheat residues. Hence, microorganism would have utilised root and soil organic matter-derived C for synthesis of PLFAs. It should be noted, that the incorporation of root C and residue C into the microbial biomass may change over time; whereas C flux from the roots may increase as the plants become bigger, C release from residues will decrease when only poorly decomposable compounds remain. In order to understand the processes in the
interface between roots and residues, it would be important to quantify the C flux from roots and residues. The gradients in the interface will however not only depend on the sinks, but also on water content and movement and will differ according to type of process or nutrient. The differential d13C and percentage residue C incorporation (Fig. 5, Table 3) showed that not all PLFAs that were more abundant in the vicinity of the residues incorporated residue C to the same extent. The PLFAs 16:0 (Gram-positive), 16:1 and cy17:0 (Gramnegative) had a higher percentage of residue C than the other fatty acids. Thus, residue C was incorporated by both Gram-positive and Gram-negative bacteria. However, the incorporation of residue C appears to be confined to some PLFAs of Gram-positive and Gramnegative bacteria. This may be due to preferential uptake by certain bacterial species or preferential utilisation of residue C for synthesis of certain PLFAs. Although fungi were also stimulated be the presence of residues, even more so than bacteria, they showed a relatively small incorporation of residue C into their PLFAs. This discrepancy may be due to the greater C use efficiency of fungi compared to bacteria (more biomass per unit C taken up) (Bailey et al., 2002) or the greater capacity of fungi to decompose native soil organic matter because they release enzymes which hydrolyse recalcitrant compounds in the organic matter (Killham, 1994; Theuerl and Buscot, 2010). 5. Conclusion The results of this study show the existence of millimetre scale gradients in abundance of bacterial and fungal fatty acids, differential microbial community composition in the interface between rhizosphere and detritusphere as well as carbon flow from residues into soil microorganisms. Despite the small distance between roots and residues (5 mm), the two interfaces did not overlap from most parameters assessed. This suggests that unless roots and residues are separated by less than 2e3 mm, their zones of influence remain distinct. The surprising finding that residue C was incorporated into the microbial biomass to a greater extent than root C would have to be verified with an improved experimental design which prevents drying of the soil in the bottom part of the pots. Acknowledgements We thank Sabine Rudolph and Heike Haslwimmer for expert technical assistance. We would also like to thank the
P. Marschner et al. / Soil Biology & Biochemistry 49 (2012) 174e183
two anonymous reviewers for their insightful and constructive comments which helped us to improve the manuscript considerably. This study was supported by a grant from the AustralianGerman Go8-DAAD funding scheme. References Abraham, W.-R., Hesse, C., Pelz, O., 1998. Ratios of carbon isotopes in microbial lipids as an indicator of substrate usage. Applied and Environmental Microbiology 64, 4202e4209. Appuhn, A., Joergensen, R.G., 2006. Microbial colonisation of roots as a function of plant species. Soil Biology & Biochemistry 38, 1040e1051. Badalucco, L., Kiukman, P.J., Nannipieri, P., 1996. Protease and deaminase acivities in wheat rhizosphere and their relation to bacterial and protozoan populations. Biology and Fertility of Soils 23, 99e104. Bailey, V.L., Smith, J.L., Bolton, H., 2002. Fungal-to-bacterial ratios in soils investigated for enhanced C sequestration. Soil Biology & Biochemistry 34, 997e1007. Bastian, F., Bouziri, L., Nicolardot, B., Ranjard, L., 2009. Impact of wheat straw decomposition on successional patterns of soil microbial community structure. Soil Biology & Biochemistry 41, 262e275. Chen, C.R., Condron, L.M., Davis, M.R., Sherlock, R.R., 2002. Phosphorus dynamics in the rhizosphere of perennial ryegrass (Lolium perenne L.) and radiata pine (Pinus radiata D. Don.). Soil Biology & Biochemistry 34, 487e499. Clarke, K.R., Warwick, R.M., 2001. Change in Marine Communities: An Approach to Statistical Analysis and Interpretation, second ed. Primer-E Ltd, Plymouth, UK. De Nobili, M., Contin, M., Mondini, C., Brookes, P.C., 2001. Soil microbial biomass is triggered into activity by trace amounts of substrate. Soil Biology & Biochemistry 33, 1163e1170. Erbs, M., Fangmeier, A., 2005. A chamberless field exposure system for ozone enrichment of short vegetation. Environmental Pollution 133, 91e102. Foster, R.C., 1986. The ultrastructure of the rhizoplane and rhizosphere. Annual Review of Phytopathology 24, 211e234. Frey, S.D., Six, J., Elliott, E.T., 2003. Reciprocal transfer of carbon and nitrogen by decomposer fungi at the soil-litter interface. Soil Biology & Biochemistry 35, 1001e1004. Frostegard, A., Tunlid, A., Bååth, E., 1993. Phospholipid fatty acid composition, biomass, and activity of microbial communities from two soil types experimentally exposed to differing heavy metals. Applied and Environmental Microbiology 59, 3605e3617. Gaillard, V., Chenu, C., Recous, S., 2003. Carbon mineralisation in soil adjacent to plant residues of contrasting biochemical quality. Soil Biology & Biochemistry 35, 93e99. Gaillard, V., Chenu, C., Recous, S., Richard, G., 1999. Carbon, nitrogen and microbial gradients induced by plant residues decomposing in soil. European Journal of Soil Science 50, 567e578. Ha, K.V., Marschner, P., Buenemann, E.K., Smernik, R.J., 2007. Chemical changes and phosphorus release during decomposition of pea residues in soil. Soil Biology & Biochemistry 39, 2696e2699. Hoyle, F.C., Murphy, D.V., Brookes, P.C., 2008. Microbial response to the addition of glucose in low-fertility soils. Biology and Fertility of Soils 44, 571e579.
183
Ingwersen, J., Poll, C., Streck, T., Kandeler, E., 2008. Micro-scale modelling of carbon turnover driven by microbial succession at a biogeochemical interface. Soil Biology & Biochemistry 40, 864e878. Joner, E.J., Magid, J., Gahoonia, T.S., Jakobsen, I., 1995. P depletion and activity of phosphatases in the rhizosphere of mycorrhizal and non-mycorrhizal cucumber (Cucumis sativus L.). Soil Biology & Biochemistry 27, 1145e1151. Kandeler, E., 2007. Physiological and biochemical methods for studying soil biota and their function. In: Paul, E.A. (Ed.), Soil Microbiology, Ecology and Biochemistry. Academic Press, pp. 53e84. Kandeler, E., Marschner, P., Tscherko, D., Gahoonia, T.S., Nielsen, N.E., 2001. Microbial community composition and functional diversity in the rhizosphere of maize. Plant and Soil 238, 301e312. Killham, K., 1994. Soil Ecology. Cambridge University Press, Cambridge. Kuzyakov, Y., 2002. Review: factors affecting rhizosphere priming effects. Journal of Plant Nutrition and Soil Science 165, 382e396. Marschner, P. (Ed.), 2012. Marschner’s Mineral Nutrition of Higher Plants. Elsevier. Marx, M.C., Kandeler, E., Wood, M., Wermbter, N., Jarvis, S.C., 2005. Exploring the enzymatic landscape: distribution and kinetics of hydrolytic enzymes in soil particle-size fractions. Soil Biology & Biochemistry 37, 35e48. Moritsuka, N., Yanai, J., Mori, K., Kosaki, T., 2004. Biotic and abiotic processes of nitrogen immobilisation in the soil-residue interface. Soil Biology & Biochemistry 36, 1141e1148. Neumann, G., 2007. Root exudates and nutrient cycling. In: Marschner, P., Rengel, Z. (Eds.), Nutrient Cycling in Terrestrial Ecosystems. Springer, pp. 123e157. Nicolardot, B., Bouziri, L., Bastian, F., Ranjard, L., 2007. A microcosm experiment to evaluate the influence of location and quality of plant residues on residue decomposition and genetic structure of soil microbial communities. Soil Biology & Biochemistry 39, 1631e1644. Poll, C., Brune, T., Begerow, D., Kandeler, E., 2010. Small-scale diversity and succession of fungi in the detritusphere of rye residues. Microbial Ecology 59, 130e140. Poll, C., Ingwersen, J., Stemmer, M., Gerzabek, M.H., Kandeler, E., 2006. Mechanisms of solute transport affect small-scale abundance and function of soil microorganisms in the detritusphere. European Journal of Soil Science 57, 583e595. Poll, C., Marhan, S., Ingwersen, J., Kandeler, E., 2008. Dynamics of litter carbon turnover and microbial abundance in a rye detritusphere. Soil Biology & Biochemistry 40, 1306e1321. Tarafdar, J.C., Jungk, A., 1987. Phosphatase activity in the rhizosphere and its relation to the depletion of soil organic phosphorus. Biology and Fertility of Soils 3, 199e204. Theuerl, S., Buscot, F., 2010. Laccases: toward disentangling their diversity and functions in relation to soil organic matter cycling. Biology and Fertility of Soils 46, 215e225. Vance, C.P., Brookes, P.C., Jenkinson, D.S., 1987. An extraction method for measuring soil microbial biomass C. Soil Biology & Biochemistry 19, 703e707. Wang, Z.Y., Kelly, J.M., Kovar, J.L., 2005. Depletion of macro-nutrients from rhizosphere soil solution by juvenile corn, cottonwood, and switchgrass plants. Plant and Soil 270, 213e221. Zelles, L., Rackwitz, R., Bai, Q.Y., Beck, T., Beese, F., 1995. Discrimination of microbial diversity by fatty acid profiles of phospholipids and lipopolysaccharides in differently cultivated soils. Plant and Soil 170, 115e122.
Contents lists available at SciVerse ScienceDirect
Soil Biology & Biochemistry journal homepage: www.elsevier.com/locate/soilbio
Microscale distribution and function of soil microorganisms in the interface between rhizosphere and detritusphere Petra Marschner a, *, Sven Marhan b, Ellen Kandeler b a b
School of Agriculture, Food and Wine, Faculty of Sciences, The Waite Research Institute, University of Adelaide, Adelaide, SA, Australia Soil Science and Land Evaluation, Soil Biology Section, University of Hohenheim, Stuttgart, Germany
a r t i c l e i n f o
a b s t r a c t
Article history: Received 7 August 2011 Received in revised form 20 January 2012 Accepted 22 January 2012 Available online 8 March 2012
The rhizosphere and the detritusphere are hot spots of microbial activity, but little is known about the interface between rhizosphere and detritusphere. We used a three-compartment pot design to study microbial community structure and enzyme activity in this interface. All three compartments were filled with soil from a long-term field trial. The two outer compartments were planted with maize (root compartment) or amended with mature wheat shoot residues from a free air CO2 enrichment experiment (residue compartment) and were separated by a 50 mm mesh from the inner compartment. Soil, residues and maize differed in 13C signature (d13C soil 26.5&, maize roots 14.1& and wheat residues 44.1&) which allowed tracking of root- and residue-derived C into microbial phospholipid fatty acids (PLFA). The abundance of bacterial and fungal PLFAs showed clear gradients with highest abundance in the first 1e2 mm of the root and residue compartment, and generally higher values in the vicinity of the residue compartment. The d13C of the PLFAs indicated that soil microorganisms incorporated more carbon from the residues than from the rhizodeposits and that the microbial use of wheat residue carbon was restricted to 1 mm from the residue compartment. Carbon incorporation into soil microorganisms in the interface was accompanied by strong microbial N immobilisation evident from the depletion of inorganic N in the rhizosphere and detritusphere. Extracellular enzyme activities involved in the degradation of organic C, N and P compounds (b-glucosidase, xylosidase, acid phosphatase and leucin peptidase) did not show distinct gradients in rhizosphere or detritusphere. Our microscale study showed that rhizosphere and detritusphere differentially influenced microbial C cycling and that the zone of influence depended on the parameter assessed. These results are highly relevant for defining the size of different microbial hot spots and understanding microbial ecology in soils. Ó 2012 Elsevier Ltd. All rights reserved.
Keywords: 13 C C flow Enzymes Maize Microbial community composition Roots Wheat
1. Introduction In soils, the rhizosphere and the detritusphere are two hot spots of microbial activity due to the presence of easily available compounds from either roots or plant residues. Microbial density and enzyme activity are high close to the roots or residues and decrease with increasing distance, forming distinct gradients in millimetre scales. Moreover, microbial community structure changes with distance from roots or residues (e.g. Kandeler et al., 2001; Poll et al., 2006, 2008, 2010). Roots modify the rhizosphere by taking up nutrients and releasing various substances such as sugars, organic acid anions and amino acids which are easily available nutrient sources for
* Corresponding author. E-mail address: [email protected] (P. Marschner). 0038-0717/$ e see front matter Ó 2012 Elsevier Ltd. All rights reserved. doi:10.1016/j.soilbio.2012.01.033
microorganisms. Hence, compared to the bulk soil, the rhizosphere of active roots is characterized by depletion of nutrients such as P and a high abundance and activity of microorganisms (Foster, 1986; Joner et al., 1995; Chen et al., 2002; Wang et al., 2005) as well as by high activity of enzymes released by roots and microorganisms (Tarafdar and Jungk, 1987; Joner et al., 1995; Badalucco et al., 1996). Abundance and activity of microorganisms and enzyme activity decrease with distance from the root surface creating distinct gradients within a few millimetres (Tarafdar and Jungk, 1987; Kandeler et al., 2001). As the rhizosphere, the detritusphere is characterized by high concentrations of easily available compounds, particularly in the early stages of residue decomposition when water-soluble compounds are released (Bastian et al., 2009; Poll et al., 2010). After depletion of the water-soluble compounds, more complex compounds requiring specialized enzymes are decomposed more slowly (Theuerl and Buscot, 2010). Similarly to the rhizosphere
P. Marschner et al. / Soil Biology & Biochemistry 49 (2012) 174e183
there is a gradient of nutrient availability as well as of enzyme activity (Gaillard et al., 2003), microbial density and community composition with distance from the residues (Poll et al., 2006; Nicolardot et al., 2007; Bastian et al., 2009). Moreover, fungal community composition and abundance change over time as compounds within the residues are successively depleted (Poll et al., 2010). As a result of diffusion and mass flow as well as translocation via fungal hyphae and soil animals, residue C can be detected up to 4 mm from the residues with a greater distance in moist compared to dry soil (Gaillard et al., 2003; Poll et al., 2008). Depending on the C/N or C/P ratio of the residues, there may be net mineralization or net immobilisation in the microbial biomass in the residues, but also within the detritusphere several millimetres away from the residues (Moritsuka et al., 2004; Ha et al., 2007). Thus, chemical, physical and biological properties of the rhizosphere and detritusphere have been studied extensively, but separately. However, in the field, roots usually grow in the vicinity of decomposing plant residues where rhizosphere and detritusphere may meet resulting an interface that is influenced by both rhizosphere and detritusphere properties. The conditions in the interface may change over time as release of C from decomposing residues and roots change. Thus the relative dominance of rhizosphere or detritusphere properties in the interface may also change. Microbial communities characteristic for rhizosphere and detritusphere may compete and create new communities that are specific for this interface. We designed a three-compartment pot system with living roots on one side and soil with residues on the other, each separated by a fine mesh from the 5 mm-wide middle compartment by a 50 mm mesh which was filled with soil only. In this system the three dimensions of the rhizosphere and detritusphere were reduced to two dimensions to allow accurate smallscale sampling at different distance from roots and residues. The slicing of the middle compartment into 1 mm layers provided information about extent and possible interactions of rhizosphere and detritusphere. In each layer, enzyme activity and microbial community composition by PLFA were measured. The aims of this study were to (i) compare the changes induced by roots and residues and the extent of their zone of influence, (ii) assess if there is an overlap of their zones of influence when roots and residues are separated by 5 mm, and (iii) track the carbon flow from roots and residues into the microbial biomass by using soil, plants and residues with different 13C signatures and 13C-PLFA analyses. 2. Materials and methods 2.1. Experimental set up The experiment was conducted in a three-compartment pot system (diameter 100 mm, height 190 mm) with two equally-sized outer compartments, separated by a 50 mm polyamide mesh (SEFAR Nitex PA 6.6) from a 5 mm-wide middle compartment (Fig. 1). The outer compartments were filled with 664 g and the middle compartment with 99 g dry soil equivalent, sieved to 2 mm. The middle compartment consisted of three PVC frames with the outer two frames with the mesh and the middle frame defining the width of the compartment (5 mm). The three frames were held together by plastic binders threaded through holes in the frames. After filling the middle compartment with the defined moist soil weight, and assembling the ‘sandwich’, the frames were placed into the pots and the outer compartments immediately filled with moist soil. In order to track C from plants and residues into the microbial biomass, we used soil, plants and residues with differential d13C values. The soil was an agricultural topsoil (Chernozem; Corg 1.5%;
175
Fig. 1. Design of pots with two outer compartments and a middle compartment (5 mm wide) representing the interface between the two outer compartments.
Ntotal 0.13%; pH (CaCl2) 7.0; inorganic N 26 mg kg1; PCAL 93 mg kg1; KCAL 137 mg kg1; soil texture: sand 12%, silt 66%, clay 22%, water-holding capacity 381 g H2O kg1) from a long-term field trial in Bad Lauchstädt (Germany) with crop rotations including only C3 plants. The soil was sieved to 2 mm mesh size and frozen two times (20 C) for defaunation before storage at 4 C. The initial soil water content was 51% water-holding capacity and the bulk density was 1.2 g cm3. The mature wheat residues (C/N 74, water-soluble C and N 23 and 0.6 g kg1, respectively) were obtained from a free air CO2 enrichment (FACE) experiment conducted at the University of Hohenheim (Erbs and Fangmeier, 2005). The shoot residues were cut into 2 cm length and mixed thoroughly into the soil of one outer compartments at a rate of 7.5 g kg1 dry soil. Pre-germinated maize (Zea mays L., cv Amadeo) was planted in the other outer compartment. The d13C values were: wheat residues 44.1&, soil 26.5& and maize roots 14.1&. There were four treatments which differed in configuration of the outer compartments: (1) unamended soil-unamended soil (soilesoil), (2) soil with wheat residues-unamended soil (residue-soil), (3) maize-unamended soil (maize-soil), (4) maize-soil with wheat residues (maize-residue). The pots were sealed and placed in a water bath in order to maintain the soil temperature at 20 C. The water bath was situated in a glasshouse with ambient light and temperature (summer). Soil moisture was maintained by weight daily. Despite watering daily to weight, the soil in the lower parts of the pots with plants dried out because of strong water uptake by the roots, particularly in the second growth phase from 14 to 23 DAP.
176
P. Marschner et al. / Soil Biology & Biochemistry 49 (2012) 174e183
Between 14 days after planting (DAP) and 23 DAP, some fine roots grew through the mesh or gaps between the frames of the middle compartment thus the middle compartment was not entirely root-free. 2.2. Sampling Four replicates of each treatment were harvested on 14 DAP and 23 DAP. The first sampling date was chosen to allow sufficient root growth to occur for the development of a rhizosphere at the mesh of the middle compartment. The second sampling was selected to have a denser root mat at the mesh and greater decomposition of the wheat straw. Soil from the outer compartment was collected after removing roots or residue particles. The middle compartment was frozen at 20 C and sliced into 1 mm vertical slices using a kitchen knife. The layers 0e1, 1e2, 2e3, 3e4 and 4e5 mm refer to the distance in mm from the outer root compartment if plants were present (maize-soil and maize-residue) or the residue compartment in the residue-soil treatment. In the soilesoil treatment the side of the first slice was selected randomly. To obtain layers with the correct distance from the mesh, the middle compartment was placed horizontally on a flat surface and the top frame with the mesh and the middle frame were removed leaving the soil rectangle resting on the mesh of the bottom frame. Then PVC frames of a certain thickness were placed around the soil rectangle. For the first slice, the frame thickness was 4 mm (thus obtaining the layer 0e1 mm from the top mesh). After removing the soil above this frame, the frame was replaced by one of 3 mm thickness and the process of removing the soil repeated. This was repeated with frames of 2 and 1 mm thickness. Any soil remaining after removal of the 1 mm frame was defined as from the 4e5 mm layer. Care was taken to minimize the mixing of soil from different layers during slicing, but some mixing may have occurred, particularly between layers 3e4 and 4e5 mm. The soil from each layer was mixed and then stored at 20 C until analyses. Maize shoot and root dry weight were determined after drying at 50 C for 48 h. Soil water content of each sample was estimated by drying samples at 105 C for 24 h.
i16:0, and i17:0 are predominantly found in Gram-positive, and cy17:0 and cy19:0 in Gram-negative bacteria. The PLFAs 18:1u9 and 18:2 u 6,9 c or t were used as marker for fungal biomass (Kandeler, 2007). The peaks of the fungal fatty acids 18:2u6t, 18:1u9 and 18:3u3 could not be separated. In the following, the sum of these three fatty acids will be referred to as 18:1u9. Isotopic ratios of the PLFAs were determined using a GC (HP 6890, Agilent Inc., USA) coupled via a combustion III Interface (Thermo Finnigan, Waltham, USA) to a Delta Plus XP mass spectrometer (Thermo Finnigan, Waltham, USA). Carbon dioxide of known isotopic composition was injected at the beginning and at the end of each run, which was finished by a re-oxidation of the combustion interface with O2 for 10 s. Isotopic ratios of SOC and PLFAs were reported relative to the Pee Dee Belemnite (PDB) standard:
d13C (&) ¼ [(Rsam/Rstd) 1] 1000 where Rsam is the 13C/12C ratio of the sample and Rstd is the C/12C ratio of the standard (PDB). During the methylation of the PLFAs, a methyl group with a d13C different from that of the fatty acid is added. For the correction of the d13C of the PLFAs, the equation according to Abraham et al. (1998) was used.
13
2.3.3. Microbial biomass C and N Microbial biomass carbon (Cmic) and nitrogen (Nmic) were assessed using the fumigationeextraction method according to Vance et al. (1987) only for the soil from the outer compartments because the amount of soil in many of the layers from the middle compartment was not sufficient. 2.3.4. Inorganic N Inorganic N was extracted from 2 g moist soil from the different layers and the outer compartment with 20 mL 2M KCl (30 min at 250 rpm on a horizontal shaker) and the extracts centrifuged at 4070 g for 30 min. Ammonium and nitrate in the supernatant were determined with an autoanalyzer (Bran & Luebbe, Germany). þ Inorganic N was calculated as the sum of NO 3 and NH4 .
2.3. Analyses 2.3.1. Enzyme activity The activities of the following enzymes were measured as described in Marx et al. (2005): b-D-glucosidase (EC 3.2.1.21), xylosidase (EC 3.2.1.37), N-acetyl-glucosaminidase (EC 3.2.1.52), phosphatase (EC 3.1.3.1) and leucine peptidase (EC 3.4.11.1). Substrates containing the fluorescent compounds 4-methylumbelliferone (4-MUF) and 7-amino-4-methyl coumarin (7-AMC), standards and buffer (MES-buffer 2-[N-orpholino]ethanesulphonic acid, Trizma buffer (mixture of a-Tris-(hydroxymethyl)-methylamin and Tris(hydroxymethyl)-aminomethane hydrochloride)) were obtained from SigmaeAldrich (St. Louis, USA). 2.3.2. Phospholipid fatty acid extraction and analysis Phospholipid fatty acids were extracted from 4 g of soil using the procedure described by Frostegard et al. (1993). Methylnondecanoate (19:0) was used as internal standard. The fatty acid methyl esters were identified by chromatographic retention time comparison with a standard mixture composed of 37 different PLFAs that ranged from C11 to C24 (SigmaeAldrich, St Louis). To ensure correct identification of PLFAs (chain length and saturation), a random selection of samples were analysed by GCMS. The sum of the following PLFAs was used to estimate bacterial biomass: i15:0, a15:0, i16:0, 16:1o7, i17:0, cy17:0, 18:1u7, cy19:0 (Frostegard et al., 1993; Zelles et al., 1995). The PLFAs i15:0, a15:0,
2.3.5. Shoot N Total shoot N was determined after grinding with a CeN analyser (Carlo Erba, Germany). 2.4. Statistical analyses Significant differences in microbial parameters between harvests, treatments and layers were determined by ANOVA followed by Tukey test. For non-metric multidimensional scaling (MDS), PLFA data was log (x þ 1) transformed. The PLFA data were log (x þ 1) transformed to balance the contributions of fatty acids by down-weighing the dominant fatty acids and increasing the weighting of rare fatty acids (Clarke and Warwick, 2001). Transformations and MDS were carried out using Primer E software (Primer-E Ltd, Plymouth Marine Laboratory, Plymouth, UK). Unlike principle component analysis plots, MDS plots do not have axis units or labels. The 2D stress indicates how well the plot represents the variability in the data with a 2D stress <0.2 considered to represent a good reflection of the resemblance matrix (Clarke and Warwick, 2001). Significant differences in microbial community structure between treatments were determined by PERMANOVA (Primer-E Ltd, Plymouth Marine Laboratory, Plymouth, UK). The DistLM procedure in Primer was used to determine the environmental variables (soil moisture, inorganic N concentration, enzyme activities) and fatty acids which
P. Marschner et al. / Soil Biology & Biochemistry 49 (2012) 174e183
explain the observed patterns. Significant differences refer to P 0.05.
A
177
80
(i) soil-soil
(ii) residue-soil
60 14 DAP 23 DAP
3. Results 40
3.1. Plant growth Inorganic N (mg kg-1)
20
Plant shoot and root dry matter increased 2.5 fold from 14 DAP to 23 DAP and did not differ significantly between the treatments (Table 1). The shoot N concentration at 23 DAP was 13e18 g kg1 which is low to adequate (Marschner, 2012). On 14 DAP, maize had formed a few roots close to the mesh of the middle compartment; by the second harvest, 23 DAP, a dense root mat had formed in the lower half of the pot. A few fine roots had grown though the mesh or the gaps in the frame extending several mm into the middle compartment; but since their diameter and their density was very low (5e10/pot), the effect is considered to be negligible. Soil water content at sampling varied between 38 and 51% water-holding capacity, with lower values in treatments with plants, suggesting that although the water content of the pots was maintained by weight, the water did not penetrate sufficiently into the lower part of the pots to compensate for water uptake by the roots.
0 soil
0-1
1-2
2-3
3-4
4-5
soil
resid
0-1
1-2
2-3
3-4
4-5
soil
80
(iv) maize-residue
(iii) maize-soil 60
40
20
0 maize 0-1
B 25
1-2
2-3
3-4
4-5
soil
maize 0-1
1-2
2-3
3-4
4-5
resid
(ii) residue-soil
(i) soil-soil
20 15 10
3.2. Inorganic N
Sum PLFA (ng g-1)
5
The inorganic N concentrations were low, ranging between 0 and 60 mg g1 (Fig. 2A). They were highest in the soilesoil treatment, where there were no differences in N concentration between layers. Compared to the soilesoil treatment (Fig. 2A (i)), residues and roots decreased inorganic N concentrations in the middle compartment as well as in the outer compartments.
0 soil
0-1
1-2
2-3
3-4
4-5
soil
resid
0-1
1-2
2-3
3-4
4-5
soil
25
(iii) maize-soil
(iv) maize-residue
20 15
3.3. Microbial biomass C and N and enzyme activity
10
Microbial biomass C (Cmic) and N (Nmic) were only determined in the outer compartments. At both sampling times, compared to the outer compartment with soil only, Cmic was about 50% higher in the outer compartment with maize and more than twice as high in the outer compartment with residues (Table 2). However, there was no difference in Cmic between the outer compartments with maize and residues. At 14 DAP, Nmic in the outer compartments with residues or maize was about three times higher than in outer compartment with soil only (Table 2). At 23 DAP, Nmic in the outer compartment with residues was more than 2-fold higher than in the outer compartment with soil only or with maize. There were no significant differences in Cmic/Nmic ratio among treatments. The activity of all enzymes was highly variable, but tended to be higher in the outer compartments with maize or residues compared to that with soil only (Table 2) and in layer 0e1 mm;
Table 1 Shoot, root and total dry weight of maize grown in pots with unamended soil (maize-soil) or soil amended with wheat residues (maize-residue) in the other outer compartment at 14 and 23 DAP (n ¼ 4). Values in the same column followed by different letters are significantly different (P 0.05). Shoot g pot1
Root g pot1
Total g pot1
14 DAP
Maize-soil Maize-residues
0.77 a 0.72 a
0.34 a 0.38 a
1.10 a 1.10 a
23 DAP
Maize-soil Maize-residues
1.71 b 1.56 b
0.97 b 1.07 b
2.68 b 2.63 b
5 0 maize 0-1
1-2
2-3
3-4
4-5
soil
maize 0-1
1-2
2-3
3-4
4-5
resid
Distance from mesh (mm) Fig. 2. Inorganic N (A) and sum of PLFAs (B) in the outer compartments and different layers in the middle compartment of the treatments (i) soilesoil, (ii) residue-soil, (iii) maize-soil and (iv) maize-residue 14 and 23 DAP (n ¼ 4). Vertical lines indicate standard error.
there was no consistent gradient in enzyme activity in the middle compartment (data not shown). 3.4. Microbial community composition The sum of bacterial and fungal fatty acids, the bacteria to fungi (B/F) ratio and the sum of all PLFA varied between layers in the soilesoil treatment, but there was no consistent trend (Figs. 2 and 3). The sum of bacterial and fungal fatty acids (Fig. 3A, B) and the sum of all PLFA (Fig. 2B) were increased in the vicinity of the residues and the maize roots with generally higher concentrations in the vicinity of the residues. The zone of higher concentrations extended 2e3 mm from the outer compartments, with the zone being wider for the residues than for the maize roots. Although roots and residues increased the concentrations of bacterial and fungal fatty acids, the increase was greater for the fungal fatty acids, thus the B/F ratio was reduced in the vicinity of residues and maize roots (Fig. 3C). In the maize-residue treatment ((iv) in Fig. 3), there was no overlap of gradients for the bacterial fatty acids; however the gradients overlapped for the fungal fatty acids.
178
P. Marschner et al. / Soil Biology & Biochemistry 49 (2012) 174e183
Table 2 Microbial biomass C, N C/N ratio and activities of b-D-glucosidase, xylosidase, N-acetyl-glucosaminidase, phosphatase and leucine peptidase in the outer compartments without plants or residues, with residues or with plants at 14 and 23 DAP (n ¼ 4). Values in the same row followed by different letters are significantly different (P 0.05). Outer compartment
Cmic (mg g1) Nmic (mg g1) Cmic/Nmic
14 DAP
23 DAP
Soil only
Residues
Maize
Soil only
Residues
Maize
23.8 a 2.2 a 10.8 a
61.6 c 8.7 c 7.1 a
41.8 ab 4.9 b 8.5 a
26.9 a 3.4 ab 7.8 a
59.9 bc 8.3 c 7.2 a
45.8 bc 5.1 b 8.9 a
122 a 14 a 18 a 48 a 147 b
157 abc 19 a 31 ab 107 c 178 c
204 c 14 a 29 a 66 ab 149 ab
125 ab 12 a 25 a 38 a 116 a
174 bc 43 a 50 b 69 ab 136 ab
176 c 22 a 36 ab 90 bc 150 bc
Enzyme activity (nmol g dw h1)
b-D-glucosidase Xylosidase N-acetyl-glucosaminidase Phosphatase Leucine peptidase
PERMANOVA of all PLFA data showed that the main effects (treatment, layer and sampling time) as well as all interactions were significant, with the interactions sampling time layer and treatment sampling time layer as well as the main effect layer being the most important factors. When the treatments were analysed separately (Fig. 4), PERMANOVA showed that the interaction layer sampling time was more important than the main effects. In the soilesoil treatment (Fig. 4A), there were no significant differences in microbial community composition between the layers 1e5 mm at 14 DAP; but at 23 DAP, the community composition of outermost layers (0e2 mm and 4e5 mm) differed significantly from that in the layers 2e4 mm. There was no consistent gradient or change in microbial community composition with distance from the outer compartments. At 14 DAP in the residue-soil treatment (Fig. 4B), the microbial community composition of the outer compartment with residues and the layers 0e1, 1e2 and 2e3 mm differed significantly from each other and the community composition of layers 3e5 mm and the outer compartment with soil only. At 23 DAP, the two outer compartments and the layers 0e1 and 4e5 mm had distinct communities and the layer 1e2 mm differed from that of the layer 3e4 mm. The microbial community composition of the layers 0e1 and 1e2 mm in the maize-soil treatment (Fig. 4C), was similar, but differed from that of the other layers. The community composition of the remaining layers were distinct at 14 DAP, whereas at 23 DAP, the communities of layers 2e3 and 3e4 mm were similar. At both sampling times, the community composition of the outer compartment with maize and layers 0e2 mm was different, and distinct from the other layers as well as from the outer compartment with soil only. In the maize-residue treatment (Fig. 4D), at 14 DAP, the community composition of the layers 1e4 mm was similar, whereas at 23 DAP, the microbial community composition of each layer was distinct. At 23 DAP, the microbial communities of the outer compartment with maize and layers 0e2 mm differed from all others and also the communities in the outer compartment with residues and the layers 3e5 mm were distinct from all other layers. Comparison of the community composition of the different layers in the maize-residue treatment with the respective layers in the residue-soil and the maize-soil treatment showed that roots and residues maintained their distinct communities in the interface up to 2 mm from the mesh (data not shown). However, the community composition of the middle layer maize-residue treatment (3 mm from the roots and the residues) did not differ from the same layer in both the residue-soil and the maize-soil treatments, suggesting that the community is a mixture of detritusphere and rhizosphere microbes.
The concentrations of the following fatty acids were consistently increased in the vicinity of the residues and of the roots, a15:0, 16:0 (Gram-positive bacteria), 17:0 (bacteria), 18:2u6 and 18:1u9 (fungi), with higher concentrations in the vicinity of the residues (data not shown). 3.5. d13C of PLFA Of the 22 detected PLFAs, only 9 (i15:0, a15:0, 15:1, 16:0, 16:1, 16:1u7, cy17:0/17:1/i17:0, C18:0 and 18:1u9) showed gradients in d13C values (examples shown in Fig. 5). In all cases, the d13C values in 1 mm distance from the outer compartment with residues were more strongly depleted than those in greater distance. For some fatty acids, the depletion zone extended up to 2e3 mm: i15:0, 16:0, i16:0, 18:0 and 18:1u9 in the residue-soil treatment and i15:0, a15:0, cy17:0 and 18:1u9 in the maize-residue treatment. The degree of depletion varied between fatty acids, 15:1, 16:0, 16:1, i16:0, cy17:0 and 18:1u9 were more strongly depleted than the others. The fatty acids 16:0 and i16:0 are indicators of Grampositive bacteria, cy17:0 of Gram-negative bacteria and 18:1u9 is a fungal signature fatty acid. Thus, no particular group of microorganisms was more strongly depleted than others. The extent of the depletion zone varied slightly between the two sampling times, but there was no consistent effect of sampling time. The d13C values in the vicinity of the root compartment were only slightly enriched compared to the outer compartment with only soil (data not shown). Based on the two-compartment mixing model, the contribution of maize root C to PLFA C in the middle section was very small, ranging from 0 to 7.2% (data not shown). For most PLFAs, maize root C contribution was zero, only for the PLFAs i15:0, a15:0, 15:1, 16:1u7 and 18:1u9 a slight contribution (<2%) in the 0e1 mm layer was found. The contribution of residue C to PLFA C was generally higher than that of the maize roots and could be detected up to 5 mm from the residue interface. For some PLFAs there was a clear gradient, with a high percentage contribution of residue C in the 0e1 mm layer which decreased with increasing distance from the residue compartment (Table 3). In the 0e1 mm layer, the percentage contribution ranged between 5 and 20%, with lowest values in 15:0 and highest values in 16:1. At a distance from the residue interface greater than 3 mm, the percentage contribution was usually <5%. 4. Discussion The experiment showed microscale gradients and C fluxes in the detritusphere and rhizosphere. This experimental design allows direct comparison of rhizosphere and detritusphere effects and assessing whether their zones of influence overlap. In general, roots
P. Marschner et al. / Soil Biology & Biochemistry 49 (2012) 174e183
A
different d13C, we showed that residue C is incorporated more strongly than root C into the microbial biomass up to 3 mm from the residues/roots.
30
(ii) residue-soil
(i) soil-soil
25
179
20
4.1. Extent of rhizosphere and detritusphere
Bacterial fatty acids -1 (ng g )
15 10 5 0 soil
0-1
1-2
2-3
3-4
4-5
soil
resid 0-1
1-2
2-3
3-4
4-5
soil
30
(iv) maize-residue
(iii) maize-soil
25 20 15 10 5 0
B
6
14 DAP 23 DAP
4 3
Fungal fatty acids -1 (ng g )
(ii) residue-soil
(i) soil-soil
5
2 1 0 soil
0-1
1-2
2-3
3-4
4-5
soil
resid 0-1
1-2
2-3
3-4
4-5
soil
6
(iv) maize-residue
(iii) maize-soil
5 4 3 2 1 0
C
16
(i) soil-soil
14
(ii) residue-soil
12 10 8 6
B/F ratio
4 2 0 soil
0-1
1-2
2-3
3-4
4-5
soil
resid 0-1
1-2
2-3
3-4
4-5
soil
16
(iii) maize-soil
14
(iv) maize-residue
12 10 8 6 4 2 0 maize 0-1
1-2
2-3
3-4
4-5
soil
maize 0-1
1-2
2-3
3-4
4-5 resid
Distance from mesh (mm) Fig. 3. Sum of bacterial (A), fungal signature fatty acids (B) and bacteria/fungi ratio (C) in the outer compartments and different layers in the middle compartment of the treatments (i) soilesoil, (ii) residue-soil, (iii) maize-soil and (iv) maize-residue 14 and 23 DAP (n ¼ 4). Vertical lines indicate standard error.
and residues had similar effects on the measured parameters. For some properties (fungal abundance, B/F ratio, N depletion), their gradients overlapped, whereas for most of the measured parameters, the two zones of influence remained separate despite the close vicinity of root and residues. Using plants, residues and soil with
In agreement with previous studies, we found that roots and residues stimulate enzyme activity, microbial abundance and change microbial community composition, e.g. (Tarafdar and Jungk, 1987; Kandeler et al., 2001; Gaillard et al., 2003; Poll et al., 2006, 2008). However, the zone of influence varied with the parameter assessed, being narrower for enzyme activity and wider for changes in microbial community composition and inorganic N concentration. In the present study, increased enzyme activity was confined to the outer compartment and the first mm from the mesh. The narrow zone of increased enzyme activity is in agreement with Kandeler et al. (2001) who reported high activity of invertase, xylanase, acid phosphatase and protease only in <1 mm distance from the root surface in a soil of similar texture. On the other hand, gradients in enzyme activity have been found up to 3 mm (Poll et al., 2006) in a clay-rich soil and 4 mm (Gaillard et al., 1999) from the residue interface in a sandy soil. Thus, the width of the zone of increased activity does not seem to be related to soil type, but could be due to differences in root exudation and residue decomposition among the studies. Root and residue C was detected in the PLFAs up to 1 and 3 mm from the root and residue compartment, respectively, suggesting that the narrow zone of increased enzyme activity in the detritusphere can not be explained by lack of C flow from the residues. Previous studies have shown that microbial community composition changes with distance from the root surface (Kandeler et al., 2001) and the residue-soil interface (Nicolardot et al., 2007; Poll et al., 2010). In agreement with these studies, microbial community composition in the present study differed between layers and even the community in the layers up to 3 mm from the mesh often differed from that of the outer compartments (Fig. 4) with a similar extent in the vicinity of the roots and the residues. Indeed, in the maize-residue treatment, roots and residues maintained their distinct communities up to 2 mm from the mesh, whereas the community in the middle layer (3 mm from both residues and roots) was a mixture of the two communities. In most soils, carbon availability limits the growth of microorganisms (De Nobili et al., 2001; Hoyle et al., 2008). Hence, the availability of C is probably the main driver of differential microbial biomass and community composition in the various layers. Indeed, a model including carbon availability, respiration and microbial biomass C was able to accurately predict changes in microbial biomass and community composition in the detritusphere (Ingwersen et al., 2008). Differences in soil moisture between rhizosphere and detritusphere alone are unlikely to explain the differential community composition in the vicinity of maize roots and residues, because when comparing the percentage explanation of factors such as N availability, enzyme activity and soil moisture for patterns in microbial community composition, soil moisture had little importance, explaining not more than 10% of the variability. In agreement with other studies (Moritsuka et al., 2004; Appuhn and Joergensen, 2006; Poll et al., 2008), microbial biomass (sum of PLFAs, concentration of bacterial and fungal fatty acids) was increased up to 4 mm from the residues and maize roots (Figs. 2 and 3). The increased microbial growth can also be explained by the supply of relatively easily decomposable C (compared to native soil organic matter) from maize roots and residues. This is supported by the greater microbial biomass C in the outer compartments with residues or maize roots compared to the control soil (Table 2). In the
180
P. Marschner et al. / Soil Biology & Biochemistry 49 (2012) 174e183
14 DAP
2 44 3
S S S 22 5 45 5 3 3 5 1
23 DAP S
1
3 3
1
soil-soil
5
1
32 S
3
S
S S
S
S
4 2 4 2
R RR
2
11
5
R
5
R
3
33 3
M
1
2 2 1 22 1 M 44 4 4 5 SS 5 5 S 5
M
2D Stress: 0.04
M 2 1 34 2 34 4 2 2 3 3 4
C
R
R
2D Stress: 0.05 1
1
1 1
5
5
S
5
4 44 4 32 23 2 3
S
R
5 5
residue-soil
5 5
2D Stress: 0.03
11
5
1
2D Stress: 0.1
1
5 44 2
S
2 S 22
1 1 1
3
A
B
2
44
1
S
3
4
3
33 3
S
4
4 2
M M MM
1 1
5 S
5 5
S
5
S
maize-soil
S
2D Stress: 0.07
S
2D Stress: 0.06 4
33 3 3
2
11 1
R R R
5
24 2 4 4 2
R
5 5 5
D maize-residue
M M M
M
2D Stress: 0.06
2 33 2 1 2 1 1 32 1 3
4 44 M M M M
5 5 55 R
R RR
2D Stress: 0.03
Fig. 4. Non-metric multidimensional scaling plots of microbial communities in the outer compartments and different layers in the middle compartment of the treatments soilesoil (A), residue-soil (B), C: maize-soil (C) and maize-residue (D) 14 and 23 DAP (n ¼ 4). Outer compartments are indicated by S, R and M for soil only, residues and maize, respectively. Shaded areas encircle communities that are not significantly different.
vicinity of the residues, the abundance of fungi was more strongly increased than in the rhizosphere (Fig. 3B). This may be due to the low concentration of residues in the outer compartment. With their hyphae, fungi would be better able to grow towards the residue particles and transport C and other nutrients into to the adjacent layers than bacteria (Poll et al., 2006). The zone of influence of roots and residues was widest for inorganic N depletion. Compared to the outer compartment with soil only and the soilesoil treatment, inorganic N was depleted in the vicinity of residues and roots (Fig. 2A). This can be explained by immobilisation of N in the microbial biomass and, in the
rhizosphere, uptake of N by the roots. The high C/N ratio of the wheat residues (C/N 74) would have resulted in net N immobilisation by the microbial biomass. The depletion of inorganic N occurred up to 5 mm from the residues and the roots in the residuesoil and maize-soil treatments, which is further than the increase in microbial biomass (up to 3 mm). This may be explained by diffusion of N (particularly nitrate) to the microorganisms. In the treatments with maize, mass flow induced by water uptake by the roots would have further resulted in movement of N towards the root surface. Due to the differential extent of the zone of influence of roots and residues for the assessed parameters, the two zones
P. Marschner et al. / Soil Biology & Biochemistry 49 (2012) 174e183
A
i15:0
-22
-22 -24
-24
-26
-26 -28
-28
C
-32
13
-30
14 DAP 23 DAP
-30
-34
-32 -34
(i) residue-soil resid
0-1
1-2
2-3
3-4
4-5
resid 0-1
soil
-20
-20
-22
-22
-24
-24
-26
-26
-28
-28
-30
-30
-34
(ii) maize-residue
-36 2-3
3-4
4-5 resid
cy17:0
3-4
4-5
soil
(ii) maize-residue maize 0-1
D
-22
-24
-24
-26
-26
-28
-28
-30
-30
-32
-32 -34
(i) residue-soil 1-2
2-3
3-4
4-5
-20
-22
-22
-24
-24
-26
-26
-28
-28
-30
-30
-32
-32 -34
(ii) maize-residue 1-2
2-3
3-4
3-4
4-5 resid
(i) residue-soil resid 0-1
soil
-20
maize 0-1
2-3
-36
-36 resid 0-1
1-2
18:1ω9
-20
-22
-34
13C
1-2
-20
-36
2-3
-36 maize 0-1
-34
1-2
-32
-32
C
(i) residue-soil
-36
-36
-34
15:1
B -20
-20
181
1-2
2-3
3-4
4-5
soil
(ii) maize-residue
-36 4-5 resid
Distance from mesh (mm)
maize 0-1
1-2
2-3
3-4
4-5 resid
Distance from mesh (mm)
Fig. 5. d13C concentrations in selected fatty acids (A: a15:0, B: 15:1, C: cy17:0, D: 18:2u6t/18:1u9/18:3u3) in different layers in the middle compartment of the treatments (i) residuesoil and (ii) maize-residue 14 and 23 DAP (n ¼ 4). Dotted line indicates d13C of the fatty acid in outer compartment with soil only. Vertical lines indicate standard error.
overlapped only for inorganic N depletion, abundance of fungal fatty acids, B/F ratio and microbial community composition. Inorganic N is mobile in soils and can move by diffusion and mass flow. Unlike bacteria, fungi can exploit a large soil volume by transporting carbon from the source to the tips of hyphae (Frey et al., 2003). Bacteria and also enzymes are less mobile which may explain why their zone of influence did not overlap. 4.2. Differential extent of rhizosphere and detritusphere Root exudates include a wide range of easily decomposable compounds (Neumann, 2007) which are considered to be more readily available than C in mature wheat straw, therefore we had expected that the root would exert a greater effect than the residues. However, there was no difference in extent of the zone for most parameters and the reverse was true for C flow, where more
residue C was incorporated than root C and the zone of incorporation of residue C was greater (Fig. 5, Table 3). Residue and maize root C in the PLFAs may have been taken up by the microbes directly as it diffused into the middle compartment. However, secondary uptake of C from dying microorganisms into actively growing microbes can not be ruled out, particularly at 23 DAP when turnover of biomass is likely to have occurred. The maize plants grew well and should have therefore released substantial amounts of exudates into the rhizosphere. The relatively limited incorporation of maize C into the microbial biomass may be explained by three factors: lower water content in the vicinity of the roots, N availability and priming effect. The soil water content was lower in outer compartment with maize (38% water-holding capacity) compared to 47e50% waterholding capacity in outer compartments with residues or soil only. Despite daily watering to weight, the soil in the lower part of the
182
P. Marschner et al. / Soil Biology & Biochemistry 49 (2012) 174e183
Table 3 Contribution of residue 13C to selected PLFAs in different layers in the middle compartment of the treatments residue-soil and maize-residue 14 and 23 DAP (n ¼ 4 standard error of the means). DAP
0e1 mm
1e2 mm
2e3 mm
14
Residue-soil Maize-residue
i15:0 4.71.0 1.70.2
3e4 mm
4e5 mm
3.60.7 1.80.4
3.40.4 3.40.7
2.20.5 4.51.4
0.90.4 7.31.7
23
Residue-soil Maize-residue
6.70.4 0.00.7
3.40.3 1.60.2
3.40.5 0.60.8
1.10.6 7.61.3
4.21.2 8.73.4
14
Residue-soil Maize-residue
cy17:0 1.21.6 0.00.5
0.01.4 0.00.7
0.01.0 0.00.8
0.00.6 0.03.7
0.01.6 8.44.7
23
Residue-soil Maize-residue
12.22.7 0.00.7
4.30.9 2.01.4
5.31.0 3.61.0
1.30.4 10.54.4
3.21.1 12.06.3
14
Residue-soil Maize-residue
18:1u9 10.92.4 0.00.2
3.40.7 0.00.7
1.84.2 1.00.2
0.00.6 2.02.4
0.00.7 10.33.4
28
Residue-soil Maize-residue
12.70.9 0.00.3
5.20.8 3.70.7
6.50.9 4.70.4
3.60.3 8.62.5
2.20.5 9.55.0
compartments with plants dried out due to water uptake by the roots. This was also the case, to a lesser extent, for the soil in the middle compartment. Diffusion of soluble compounds occurs in water films/water-filled pores. Poll et al. (2008) showed that diffusion of C is restricted by low water content. Thus, exudates released by the roots may not have entered the middle compartment and moved within it because of low diffusion rates. Since the soil in the residue compartment and also in the adjacent layers in the middle compartment remained moister, diffusion would have been greater than in the root compartment (Poll et al., 2008). The strong effect of the residues was surprising because of the low density and the large particle size of the residues. Moreover, even at 23 DAP, the wheat residue particles appeared to be intact and only slightly discoloured. In the vicinity of the residues, the abundance of fungi was more strongly increased than that of bacteria (Fig. 2). This may be due to the low concentration of residues in the outer compartment. With their hyphae, fungi would be better able to grow towards the residue particles and transport C and other nutrients into to the adjacent layers than bacteria (Poll et al., 2006). However, the differential water content can not fully explain the limited diffusion of root C compared to residue C as the extent of the zone of influence for other parameters such as microbial community composition and abundance of bacterial PLFA was similar in the vicinity of roots and residues. Nitrogen availability may have limited the ability of microorganisms in the vicinity of the roots to utilise C from exudates. Inorganic N concentrations were decreased to the same extent in the rhizosphere and in the detritusphere (Table 2); but whereas only immobilisation of N into the microbial biomass can explain the decrease in the detritusphere, N would have been taken up by plants and microbes in the rhizosphere. This argument is supported by the lower microbial biomass N in the root than in the residue compartment, although this difference was not statistically significant. Root exudates may have induced a priming effect (Kuzyakov, 2002), i.e. increased decomposition of native organic matter to a greater extent than compounds from the poorly degradable wheat residues. Hence, microorganism would have utilised root and soil organic matter-derived C for synthesis of PLFAs. It should be noted, that the incorporation of root C and residue C into the microbial biomass may change over time; whereas C flux from the roots may increase as the plants become bigger, C release from residues will decrease when only poorly decomposable compounds remain. In order to understand the processes in the
interface between roots and residues, it would be important to quantify the C flux from roots and residues. The gradients in the interface will however not only depend on the sinks, but also on water content and movement and will differ according to type of process or nutrient. The differential d13C and percentage residue C incorporation (Fig. 5, Table 3) showed that not all PLFAs that were more abundant in the vicinity of the residues incorporated residue C to the same extent. The PLFAs 16:0 (Gram-positive), 16:1 and cy17:0 (Gramnegative) had a higher percentage of residue C than the other fatty acids. Thus, residue C was incorporated by both Gram-positive and Gram-negative bacteria. However, the incorporation of residue C appears to be confined to some PLFAs of Gram-positive and Gramnegative bacteria. This may be due to preferential uptake by certain bacterial species or preferential utilisation of residue C for synthesis of certain PLFAs. Although fungi were also stimulated be the presence of residues, even more so than bacteria, they showed a relatively small incorporation of residue C into their PLFAs. This discrepancy may be due to the greater C use efficiency of fungi compared to bacteria (more biomass per unit C taken up) (Bailey et al., 2002) or the greater capacity of fungi to decompose native soil organic matter because they release enzymes which hydrolyse recalcitrant compounds in the organic matter (Killham, 1994; Theuerl and Buscot, 2010). 5. Conclusion The results of this study show the existence of millimetre scale gradients in abundance of bacterial and fungal fatty acids, differential microbial community composition in the interface between rhizosphere and detritusphere as well as carbon flow from residues into soil microorganisms. Despite the small distance between roots and residues (5 mm), the two interfaces did not overlap from most parameters assessed. This suggests that unless roots and residues are separated by less than 2e3 mm, their zones of influence remain distinct. The surprising finding that residue C was incorporated into the microbial biomass to a greater extent than root C would have to be verified with an improved experimental design which prevents drying of the soil in the bottom part of the pots. Acknowledgements We thank Sabine Rudolph and Heike Haslwimmer for expert technical assistance. We would also like to thank the
P. Marschner et al. / Soil Biology & Biochemistry 49 (2012) 174e183
two anonymous reviewers for their insightful and constructive comments which helped us to improve the manuscript considerably. This study was supported by a grant from the AustralianGerman Go8-DAAD funding scheme. References Abraham, W.-R., Hesse, C., Pelz, O., 1998. Ratios of carbon isotopes in microbial lipids as an indicator of substrate usage. Applied and Environmental Microbiology 64, 4202e4209. Appuhn, A., Joergensen, R.G., 2006. Microbial colonisation of roots as a function of plant species. Soil Biology & Biochemistry 38, 1040e1051. Badalucco, L., Kiukman, P.J., Nannipieri, P., 1996. Protease and deaminase acivities in wheat rhizosphere and their relation to bacterial and protozoan populations. Biology and Fertility of Soils 23, 99e104. Bailey, V.L., Smith, J.L., Bolton, H., 2002. Fungal-to-bacterial ratios in soils investigated for enhanced C sequestration. Soil Biology & Biochemistry 34, 997e1007. Bastian, F., Bouziri, L., Nicolardot, B., Ranjard, L., 2009. Impact of wheat straw decomposition on successional patterns of soil microbial community structure. Soil Biology & Biochemistry 41, 262e275. Chen, C.R., Condron, L.M., Davis, M.R., Sherlock, R.R., 2002. Phosphorus dynamics in the rhizosphere of perennial ryegrass (Lolium perenne L.) and radiata pine (Pinus radiata D. Don.). Soil Biology & Biochemistry 34, 487e499. Clarke, K.R., Warwick, R.M., 2001. Change in Marine Communities: An Approach to Statistical Analysis and Interpretation, second ed. Primer-E Ltd, Plymouth, UK. De Nobili, M., Contin, M., Mondini, C., Brookes, P.C., 2001. Soil microbial biomass is triggered into activity by trace amounts of substrate. Soil Biology & Biochemistry 33, 1163e1170. Erbs, M., Fangmeier, A., 2005. A chamberless field exposure system for ozone enrichment of short vegetation. Environmental Pollution 133, 91e102. Foster, R.C., 1986. The ultrastructure of the rhizoplane and rhizosphere. Annual Review of Phytopathology 24, 211e234. Frey, S.D., Six, J., Elliott, E.T., 2003. Reciprocal transfer of carbon and nitrogen by decomposer fungi at the soil-litter interface. Soil Biology & Biochemistry 35, 1001e1004. Frostegard, A., Tunlid, A., Bååth, E., 1993. Phospholipid fatty acid composition, biomass, and activity of microbial communities from two soil types experimentally exposed to differing heavy metals. Applied and Environmental Microbiology 59, 3605e3617. Gaillard, V., Chenu, C., Recous, S., 2003. Carbon mineralisation in soil adjacent to plant residues of contrasting biochemical quality. Soil Biology & Biochemistry 35, 93e99. Gaillard, V., Chenu, C., Recous, S., Richard, G., 1999. Carbon, nitrogen and microbial gradients induced by plant residues decomposing in soil. European Journal of Soil Science 50, 567e578. Ha, K.V., Marschner, P., Buenemann, E.K., Smernik, R.J., 2007. Chemical changes and phosphorus release during decomposition of pea residues in soil. Soil Biology & Biochemistry 39, 2696e2699. Hoyle, F.C., Murphy, D.V., Brookes, P.C., 2008. Microbial response to the addition of glucose in low-fertility soils. Biology and Fertility of Soils 44, 571e579.
183
Ingwersen, J., Poll, C., Streck, T., Kandeler, E., 2008. Micro-scale modelling of carbon turnover driven by microbial succession at a biogeochemical interface. Soil Biology & Biochemistry 40, 864e878. Joner, E.J., Magid, J., Gahoonia, T.S., Jakobsen, I., 1995. P depletion and activity of phosphatases in the rhizosphere of mycorrhizal and non-mycorrhizal cucumber (Cucumis sativus L.). Soil Biology & Biochemistry 27, 1145e1151. Kandeler, E., 2007. Physiological and biochemical methods for studying soil biota and their function. In: Paul, E.A. (Ed.), Soil Microbiology, Ecology and Biochemistry. Academic Press, pp. 53e84. Kandeler, E., Marschner, P., Tscherko, D., Gahoonia, T.S., Nielsen, N.E., 2001. Microbial community composition and functional diversity in the rhizosphere of maize. Plant and Soil 238, 301e312. Killham, K., 1994. Soil Ecology. Cambridge University Press, Cambridge. Kuzyakov, Y., 2002. Review: factors affecting rhizosphere priming effects. Journal of Plant Nutrition and Soil Science 165, 382e396. Marschner, P. (Ed.), 2012. Marschner’s Mineral Nutrition of Higher Plants. Elsevier. Marx, M.C., Kandeler, E., Wood, M., Wermbter, N., Jarvis, S.C., 2005. Exploring the enzymatic landscape: distribution and kinetics of hydrolytic enzymes in soil particle-size fractions. Soil Biology & Biochemistry 37, 35e48. Moritsuka, N., Yanai, J., Mori, K., Kosaki, T., 2004. Biotic and abiotic processes of nitrogen immobilisation in the soil-residue interface. Soil Biology & Biochemistry 36, 1141e1148. Neumann, G., 2007. Root exudates and nutrient cycling. In: Marschner, P., Rengel, Z. (Eds.), Nutrient Cycling in Terrestrial Ecosystems. Springer, pp. 123e157. Nicolardot, B., Bouziri, L., Bastian, F., Ranjard, L., 2007. A microcosm experiment to evaluate the influence of location and quality of plant residues on residue decomposition and genetic structure of soil microbial communities. Soil Biology & Biochemistry 39, 1631e1644. Poll, C., Brune, T., Begerow, D., Kandeler, E., 2010. Small-scale diversity and succession of fungi in the detritusphere of rye residues. Microbial Ecology 59, 130e140. Poll, C., Ingwersen, J., Stemmer, M., Gerzabek, M.H., Kandeler, E., 2006. Mechanisms of solute transport affect small-scale abundance and function of soil microorganisms in the detritusphere. European Journal of Soil Science 57, 583e595. Poll, C., Marhan, S., Ingwersen, J., Kandeler, E., 2008. Dynamics of litter carbon turnover and microbial abundance in a rye detritusphere. Soil Biology & Biochemistry 40, 1306e1321. Tarafdar, J.C., Jungk, A., 1987. Phosphatase activity in the rhizosphere and its relation to the depletion of soil organic phosphorus. Biology and Fertility of Soils 3, 199e204. Theuerl, S., Buscot, F., 2010. Laccases: toward disentangling their diversity and functions in relation to soil organic matter cycling. Biology and Fertility of Soils 46, 215e225. Vance, C.P., Brookes, P.C., Jenkinson, D.S., 1987. An extraction method for measuring soil microbial biomass C. Soil Biology & Biochemistry 19, 703e707. Wang, Z.Y., Kelly, J.M., Kovar, J.L., 2005. Depletion of macro-nutrients from rhizosphere soil solution by juvenile corn, cottonwood, and switchgrass plants. Plant and Soil 270, 213e221. Zelles, L., Rackwitz, R., Bai, Q.Y., Beck, T., Beese, F., 1995. Discrimination of microbial diversity by fatty acid profiles of phospholipids and lipopolysaccharides in differently cultivated soils. Plant and Soil 170, 115e122.