1 s2.0 S092913930400109X Main
1 s2.0 S092913930400109X Main
1 s2.0 S092913930400109X Main
www.elsevier.com/locate/apsoil
Abstract
Annual plant species differ in their rhizosphere microbial community composition. However, rhizosphere communities are
often investigated under controlled conditions, and it is unclear if perennial plants growing in the field also have rhizosphere
communities that are specific to a particular plant species. The aim of our study was to determine the bacterial community
composition of three species of Banksia (B. attenuata R. Brown, B. ilicifolia R. Brown and B. menziesii R. Brown) growing in
close proximity in a native woodland in Western Australia and to relate community structure to function. All three species are
small trees that produce cluster roots in the field following winter rains. Cluster roots and rhizosphere soil were sampled in early
spring (August 2001) and again four weeks later (September 2001). Many new cluster roots were formed in the period between
the August and the September sampling. Rhizosphere soil pH, percent soil moisture and C and N content did not differ
significantly among species or sampling times. However, the bacterial community composition on the cluster roots and in the
rhizosphere soil, studied by denaturing gradient gel electrophoresis (DGGE), differed among the three species, with cluster root
age class (young or mature to senescing) and also between sampling times. These changes in community composition were
accompanied by changes in the activity of some of the enzymes studied. The activities of b-glucosidase and protease increased
over time. The three species differed in asparaginase activity, but not in the activity of acid and alkaline phosphatase in the
rhizosphere. These results suggest a relationship between the changes in composition and function of bacterial communities.
# 2004 Elsevier B.V. All rights reserved.
Keywords: Bacterial community composition; Banksia; Cluster roots; DGGE; Enzyme activity; Rhizosphere
* Corresponding author. Tel.: +61 8 8303 7379; fax: +61 8 8303 6511.
E-mail address: petra.marschner@adelaide.edu.au (P. Marschner).
0929-1393/$ – see front matter # 2004 Elsevier B.V. All rights reserved.
doi:10.1016/j.apsoil.2004.09.001
192 P. Marschner et al. / Applied Soil Ecology 28 (2005) 191–201
single species often have little effect on a given (0–5 cm) were taken from beneath the Banksias at four
function (Nannipieri et al., 2002; Avrahami et al., different locations within an area of approximately
2003; Miethling et al., 2003). However, there are 1 ha. At each location, the three species grew within
indications that changes in overall microbial commu- 2–3 m distance from each other. Sampling locations
nity composition can be correlated with changes in differed between the two sampling dates. In order
certain functions both under controlled conditions to ensure that the sampled roots belonged to the
(Kandeler et al., 2002; Avrahami et al., 2003) and in correct species, the samples were taken within 50 cm
the field (Marschner et al., 2003). of the stem of each species and the lateral root
The objective of this study was to determine traced back to the parent tree. Root mats with adhering
bacterial rhizosphere community composition and soil were carefully extracted, put in plastic bags,
enzyme activities in three Banksia species growing placed on ice and processed within 3 h of collection.
close to each other in a natural Banksia woodland Size, color and location along the cluster roots were
north of Perth, Western Australia. Banksia is an used to differentiate between young (small, light
interesting model for understanding interactions brown, close to root tip of the laterals) and mature
among rhizosphere microorganisms and plant adapta- cluster roots (larger, dark brown, further away from
tions to low nutrient soils. We examined three co- the root tip).
occurring species of Banksia for differences in Lightly adhering soil was removed by shaking the
bacterial rhizosphere community structure in relation roots gently and then two young cluster roots or one
to the activity of several enzymes involved in C, N and mature cluster root were placed in microcentrifuge
P cycling. Similarity among species would indicate tubes and freeze-dried for determination of the
that there is a generalized effect of Banksia on nutrient bacterial community composition. The lightly adher-
cycling processes while differences would suggest that ing soil from old and young cluster roots was mixed
each Banksia species has a distinct effect on soil (hereafter referred to as rhizosphere soil) and sieved to
microorganisms and function in the rhizosphere. 2 mm. Approximately 500 mg of rhizosphere soil
were placed in microcentrifuge tubes and freeze-dried
for determination of the bacterial community compo-
2. Materials and methods sition. The rhizosphere soil for the chemical analyses
and enzyme activity measurements was stored at 4 8C
The sampling site was a native woodland at until further analysis. Bulking of the rhizosphere soil
Wanneroo, approximately 20 km north of Perth from both cluster root ages was necessary because the
(318450 S, 1158480 E). Soils are nutrient-poor siliceous amount of soil from each cluster root age would have
sands that dry out to great depth during summer, with been insufficient for analyses of all properties
nutrients concentrated in the litter layer and the top 0– examined in this study.
5 cm of soil (Lamont, 1993; Pate et al., 1998). The At the second sampling only, soil was also
three Banksia species studied (B. attenuata R. Brown, collected from four open areas between trees adjacent
B. ilicifolia R. Brown and B. menziesii R. Brown) are to the locations where the Banksia cluster roots were
small trees of 3–7 m height with hard sclerophyllous sampled. This soil was free of vegetation and litter and
leaves. Cluster roots and rhizosphere soil were is referred to hereafter as bare soil. Samples of bare
sampled in mid-August 2001 (early spring) and again soil contained some fine roots, but no cluster roots.
4 weeks later in September 2001. The August
sampling was at the beginning of the seasonal growth 2.1. Chemical properties of the rhizosphere
period (Grierson and Adams, 2000), with only a few and bare soil
new cluster roots initiated; the majority of cluster roots
were from the previous year (dark-colored cluster Soil moisture content was measured after drying
roots and no longer active). At the second sampling in at 105 8C for 5 h and given on a gravimetric basis.
September, numerous new cluster roots had been Soil pH was determined in a water:soil ratio of 1:1.
formed at the soil–litter interface by all three Banksia Total C and N were determined using a Carlo Erba
species. Samples of roots and soils in the top soil layer NA 1500 Elemental Analyzer (Milan, Italy).
194 P. Marschner et al. / Applied Soil Ecology 28 (2005) 191–201
To determine water-soluble C, 20 g oven dry and to increase the pH. The nitrophenol concentration
weight equivalents of fresh soil samples were shaken was determined photometrically at 400 nm.
with 80 ml of ultrapure water for 1 h and then Enzyme activities were compared by one-way
centrifuged for 15 min at 3000 rpm. The supernatant ANOVA for the two sampling dates separately and by
was first filtered through a Whatman #42 filter to two-way ANOVA for both sampling dates together
remove larger particles followed by filtration through using the least significant difference to compare the
a 0.45 mm-pore membrane (Millipore, USA). Total mean values (GenStat 5th edition, Rothamsted
carbon in the extract was measured using a TOC Experimental Station, 2000).
analyzer (Shimadzu, 5000a).
2.3. Assessment of the bacterial community
2.2. Enzyme activities in the rhizosphere and composition in the rhizosphere soil and on
bare soil cluster roots
Protease activity was measured according to Ladd The method described in Marschner et al. (2001b)
and Butler (1972). Briefly, 5 mL Tris buffer (pH 8.1) was used for bacterial community composition
and 5 mL caseinate solution (2% (w/v)) were added to analysis by polymerase chain reaction-denaturing
1 g of moist soil. The samples were shaken for 2 h gradient gel electrophoresis (PCR-DGGE). Briefly,
at 50 8C. After incubation, 5 mL trichloric acid (15% 200 mM phosphate buffer and 10% (w/v) SDS were
(v/v)) were added and centrifuged for 10 min at added to the freeze-dried root or rhizosphere soil
12,000 rpm. After alkalinization of the supernatant, samples followed by homogenization in a bead beater
tryrosine was determined photometrically at 700 nm (Fast-Prep, Model FP120, Qbiogene) at 5.5 m s1 for
using the Folin-Ciocalteu reagent (33%). 30 s. After proteins were removed with a protein
L-Asparaginase activity was measured according precipitation solution (PPS1, Qbiogene), the DNA
to Frankenberger and Tabatabai (1991). Nine was bound to a silica matrix (Binding matrix1,
milliliters of Tris buffer (pH 10) and 1 mL of L- Qbiogene), washed twice with an ethanol–salt solution
asparagine solution (0.5 M) were added to 5 g of (SEWS1, Qbiogene) and then desorbed into sterile
moist soil. Samples were incubated at 37 8C for 2 h. water. The DNA samples were stored at –20 8C for
After incubation, 40 mL KCl (2.5 M)–Ag2SO4 further analysis. The DNA extract was diluted 10-fold
(100 mg L1) solution were added. Ammonium con- to reduce the concentration of compounds that may
centration was determined by steam distillation. inhibit the amplification.
The method by Alef and Nannipieri (1995) was A highly variable section of the 16S rDNA was
used for measurement of b-glucosidase activity. One g amplified by using the primer set F984 and R1378
moist, sieved soil was incubated with 4 mL modified (Heuer et al., 1997). A GC clamp (Sheffield et al.,
universal buffer (pH 6.0) and 1 mL p-nitrophenyl-b-D- 1989) was attached to primer F984 to prevent
glucoside (25 mM) solution. After incubation at 37 8C complete separation of the strands in the DGGE
for 1 h, 1 mL CaCl2 solution (0.5 M) and 4 mL Tris gel. For some samples, namely young cluster roots
buffer (0.1 M, pH 12) were added. The nitrophenol sampled in September, amplification of root DNA
concentration was determined photometrically at extracts was not successful, probably due to high
400 nm. concentrations of inhibiting substances such as
Phosphatase activities were determined according phenolics. For samples that could be amplified, the
to Tabatabai and Bremner (1969). Alkaline phospha- resulting fragments were separated by DGGE. To
tase activity was measured at the second sampling in obtain an adequate number of replicates, in some cases
September only. One g of moist fresh soil was DNA extracts from more than one cluster root of a
incubated with 4 mL modified universal buffer (pH 6.5 Banksia species at a given location was amplified. The
for acid phosphatase, pH 11 for alkaline phosphatase) community-specific band patterns were digitized,
and 1 mL p-nitrophenyl phosphate (15 mM) for 1 h at normalized and analyzed by multivariate analysis
37 8C. After incubation, 1 mL CaCl2 (0.5 M) and (for details see Marschner et al., 2001b) (CANOCO
4 mL NaOH (0.5 M) were added to stop the reaction 4.0, Microcomputer Power, Ithaca, USA). We used
P. Marschner et al. / Applied Soil Ecology 28 (2005) 191–201 195
principal component analyses (PCA) to assess the C content in the rhizosphere soil ranged between 9
general differences between band patterns. Redun- and 13 mg L1 and did not differ among species or
dancy discriminate analyses (RDA) provides more sampling time. Total C and N contents of the
information because not only banding patterns of rhizosphere soil were similar among the three species
different samples are compared but RDA also takes and between sampling times, and ranged between 10
environmental factors such as plant species, time, soil and 17 g kg1 C and 0.4 and 0.7 g kg1 N, giving a
moisture and activity of enzymes related to this C/N ratio of 22–28.
sample into account. The Monte Carlo Permutation After initiation of new cluster roots in September,
test calculates the proportion of variation explained b-glucosidase and protease activities were signifi-
and the significance of a given environmental factor, cantly greater than in August, whereas the activities of
and thus its importance to community composition. acid phosphatase and asparaginase activity did not
Even though these are only mathematical correlations, change over time. When both sampling dates were
they can, in most cases, be interpreted in a biological analyzed together, the three species significantly
meaningful way (Jongman et al., 1995). differed in asparaginase activity. At the August
sampling, there was no significant difference in
enzyme activities of the rhizosphere soils among
3. Results the three Banksia species (Table 1). In September,
asparaginase activity differed significantly between
In August 2001 (the first sampling), only few young the species, being more than three times greater in
cluster roots were found, with the majority being old, B. menziesii than in B. attenuata. Compared to the
dark-colored cluster roots. In the 4 weeks between the rhizosphere soil, acid phosphatase, b-glucosidase and
first and the second sampling, many new cluster roots protease activities were about three times lower in the
were formed by all three Banksia species directly at bare soil.
the soil–litter interface. Thirty different bands could be distinguished in the
Moisture content of the rhizosphere soil of all three DGGE gels, with each sample containing between 5 and
species ranged between 6.2 and 7.5% (w/w) and was 14 bands. When the DGGE banding patterns of the
similar between sampling times. The moisture content bacterial communities from both sampling dates were
of the bare, vegetation-free soil was only determined analyzed together, all environmental variables used in
for the September sampling and was significantly the redundancy discriminate analysis (RDA) had a
lower (2.8% (w/w)). The pH of the rhizosphere soil at significant effect on the bacterial community composi-
the second harvest ranged between pH 4.6 and pH 5.1 tion. Plant species (explaining 8% of the total variation
and did not differ significantly among species, or within the banding patterns), location (root or rhizo-
between rhizosphere and the bare soil. Water-soluble sphere soil) (5%) and root age (6%) had the strongest
Table 1
Activities of asparaginase b-glucosidase, acid and alkaline phosphatase and protease in the rhizosphere soil of Banksia attenuata, B. ilicifolia
and B. menziesii in August and September (means of four replicates standard error)
Asparaginase b-Glucosidase Acid phosphatase Alkaline phosphatasea Protease
(mg N g1 soil) (mg nitrophenol g1 soil) (mg nitrophenol g1 soil) (mg nitrophenol g1 soil) (mg tyrosine g1 soil)
August
B. attenuata 81 166 24 3677 448 – 732 262
B. ilicifolia 71 250 56 3669 828 – 627 163
B. menziesii 62 163 24 2994 475 – 619 139
September
B. attenuata 3 1 323 82 3774 1107 506 112 1044 225
B. ilicifolia 6 2 342 60 5089 1267 577 152 1401 356
B. menziesii 10 1 370 143 4258 1447 426 83 1225 564
Bare soil 5 2 96 4 1429 128 411 65 213 65
a
Alkaline phosphatase activity was only determined in September.
196 P. Marschner et al. / Applied Soil Ecology 28 (2005) 191–201
Table 2
Results of the Monte Carlo Permutation test in percent of variation explained and significance of environmental factors (P value) for the bacterial
community composition on the roots of B. attenuata, B. ilicifolia and B. menziesii in August and September determined by denaturing gradient
gel electrophoresis (DGGE)
Both dates August September Average (%)
% P % P % P
** ** **
Banksia species 8 11 14 11
** a a
Sampling time 3 3
** * **
Rhizosphere soil vs. root 5 5 12 7
** ** a
Root age 6 6 6
** ** **
Soil moisture 4 11 11 9
** a **
pH 3 17 10
** ** **
Asparaginase 3 11 12 9
** ** **
Protease 3 10 10 8
** a **
Alkaline phosphatase 3 16 6
** ** **
Acid phosphatase 4 11 10 8
** ** **
Glucosidase 4 8 11 8
Data sets of the two sampling dates were compared for the two dates together or separately. The average percent of variation explained represents
the average of the comparisons; ns indicates non-significant at P 0.10.
*
P 0.05.
**
P 0.01.
a
Correlation could not be calculated.
effect (Table 2). The community composition on the activity (explaining 4% of the total variation). The RDA
cluster roots and in the rhizosphere soil also changed plot shows that the community composition of B.
with time in all three species. The bacterial community attenuata differed clearly from that of B. ilicifolia and B.
composition was correlated with enzyme activities, menziesii; in all three species the communities changed
especially with b-glucosidase and acid phosphatase from August to September (Fig. 1). The bacterial
Fig. 1. Ordination plot of bacterial communities in the rhizosphere soil and the cluster roots (young or mature) of B. attenuata, B. ilicifolia and B.
menziesii in August and September generated by redundancy discriminate analysis (RDA) of 16S rDNA denaturing gradient gel electrophoresis
(DGGE) profiles. The values on the axes refer to the % of the total variance explained by the axis. Communities are represented as symbols.
Communities (symbols) close to each other have similar compositions, whereas communities far apart differ strongly in composition.
P. Marschner et al. / Applied Soil Ecology 28 (2005) 191–201 197
communities of the young cluster roots of B. ilicifolia only one DNA extract of young cluster roots from
and B. menziesii differed from those of the old cluster B. menziesii could be amplified. Amplification of
roots in the August sampling. the DNA extracts of the young cluster roots of
Although many young cluster roots were found in B. attenuata was not successful, but the bacterial
September, none of their DNA extracts could be communities of the old cluster roots and those of the
amplified. Amplification of DNA extracts of mature rhizosphere soil differed from each other.
cluster roots of B. attenuata and B. ilicifolia from the In September, the bacterial community composi-
September sampling were also not successful; hence, tion was affected by pH (explaining 17% of the total
only the rhizosphere soil samples of these two species variation), Banksia species (14%), location (12%) and
were included in the analysis. soil moisture (11%) (Table 2). Community composi-
In order to obtain a clearer picture of the relation- tion was correlated with the activity of all enzymes
ship between bacterial community composition and measured, most strongly with that of alkaline
environment, the two sampling dates were analyzed phosphatase (explaining 16% of the total variation).
separately. In August, the bacterial community The effect of root age could not be examined, because
composition was affected by Banksia species and DNA extracts from young cluster roots sampled in
soil moisture (each explaining 11% of the total September could not be amplified. The PCA plot of
variation of the banding patterns) and to a lesser extent the September sampling shows that the bacterial
by root age (6%) and location (5%) (Table 2). The communities of B. menziesii differ from those of
bacterial community composition was correlated with B. ilicifolia and B. attenuata, and the bare soil com-
the activities of all enzymes. The PCA plot shows that, munities were different from those in the rhizosphere
in August, the community composition of B. ilicifolia soil (Fig. 2).
differed from that of B. menziesii whereas that of To assess the overall importance of the measured
B. attentuata showed similarities with B. ilicifolia and environmental factors, the mean percent explanation
B. menziesii (Fig. 2). In B. ilicifolia, the communities was calculated (Table 2). Bacterial community
of the young cluster roots and those in the rhizosphere composition was most strongly affected by Banksia
soil differed from those of the old cluster roots. In species (explaining 11% of the total variation), soil pH
B. menziesii the community of the young cluster (10%) and moisture (9%) and was correlated with the
root differed from those of the old roots and the activity of all enzymes measured to a similar extent
rhizosphere soil. However, it should be noted that (6–9%).
Fig. 2. Ordination plots of bacterial communities in the rhizosphere soil and the cluster roots (young or mature) of B. attenuata, B. ilicifolia and
B. menziesii in August (A) and September (B) separately generated by principal component analysis (PCA) of 16S rDNA denaturing gradient gel
electrophoresis (DGGE) profiles. For explanation see Fig. 1.
198 P. Marschner et al. / Applied Soil Ecology 28 (2005) 191–201
rhizosphere soil than in the vegetation-free soil. This longer periods (weeks), both nitrification rates and the
stimulation of microbial activity in the rhizosphere is community structure of ammonia-oxidizing bacteria
in agreement with earlier studies (Tarafdar and Jungk, changed (Avrahami et al., 2003). It should be noted
1987; Kandeler et al., 2002). that due to the high temporal and spatial heterogeneity
Changes in bacterial community composition were of ecosystems (Ettema and Wardle, 2002), the concept
related to differences in the activity of some of the of functional redundancy may only hold on a larger
enzymes studied. The differences in bacterial com- scale (cm to m), whereas in a particular niche (for
munity composition among the Banksia species were example in the rhizosphere) the presence of a certain
reflected in significant differences in asparaginase species may be crucial for a certain function.
activity. The change in community composition over In conclusion, this study demonstrated that the
time was associated with an increase in activities of b- rhizospheres of Banksia species growing under field
glucosidase and protease from August to September. conditions in Western Australia differed in bacterial
At the September sampling, the vegetation-free bare community composition and that the bacterial com-
soil had lower acid phosphatase, glucosidase and munity changed rapidly with the onset of new root
protease activity than the rhizosphere soil (Table 1) growth after winter rains. Shifts in bacterial commu-
and differed in bacterial community composition nity composition were associated with changes in
from the rhizosphere soil (Fig. 2). A link between activity of enzymes involved in N and C cycling. Thus
community composition and function is also indicated microbial community composition and enzyme
by the strong correlation between bacterial commu- activity are highly variable in time and space. In
nity composition and enzyme activity (Table 2). the rhizosphere, the main driving factors for changes
Recently, Kandeler et al. (2002) found that decreasing in bacterial community composition and activity are
activities of acid and alkaline phosphatase and likely to be root exudate amount and composition.
invertase with increasing distance from the root Consequently, each plant species may modify the
surface were associated with changes in bacterial conditions in the rhizosphere in order to maximize
community composition. nutrient acquisition from organic matter by promoting
However not all enzyme activities showed con- particular functional groups of microorganisms.
sistent links with bacterial community composition in
the present study. For example, acid phosphatase
activity did not change over time and did not differ
between Banksia species. This may be due to fact that Acknowledgements
acid phosphatase is also released by plant roots and
fungi and the activities measured here cannot be solely This study was supported by the Australian
attributed to the bacterial community (Grierson and Research Council (IREX X00001664). We thank
Adams, 2000). Alternatively, decreases in abundance Dr. Nui Milton (University of Western Australia) for
and/or activity of a certain species can be compensated the analysis of total C and N and water-soluble C.
for by increases in abundance and/or activity of
another species with a similar function, i.e. there is
functional redundancy (Zak et al., 1994; Nannipieri et
al., 2002). Miethling et al. (2003) also demonstrated References
that microbial community composition and function
Alef, K., Nannipieri, P., 1995. b-Glucosidase activity. In: Alef, K.,
(substrate utilisation pattern) are not necessarily Nannipieri, P. (Eds.), Methods in Applied Soil Microbiology
correlated. Changes in metabolic processes may occur and Biochemistry. Academic Press, London, pp. 350–352.
without concomitant changes in microbial community Aulakh, M.S., Wassmann, R., Bueno, C., Kreuzwieser, J., Rennen-
structure. For example, nitrification rates may increase berg, H., 2001. Characterization of root exudates at different
without short-term changes (days) in the community growth stages of ten rice (Oryza sativa L.) cultivars. Plant Biol.
3, 139–148.
structure of ammonia-oxidizing bacteria, indicating Avrahami, S., Liesack, W., Conrad, R., 2003. Effects of temperature
that only the activity of the species already present and fertilizer on activity and community structure of soil
increased (Avrahami et al., 2003). However, over ammonia oxidizers. Environ. Microbiol. 5, 691–705.
200 P. Marschner et al. / Applied Soil Ecology 28 (2005) 191–201
Baudoin, E., Benizri, E., Guckert, A., 2003. Impact of artificial root bedingungen und auf Standorten in Baden-Würtemberg. Kali Br.
exudates on the bacterial community composition in bulk soil 20, 431–439.
and maize rhizosphere. Soil Biol. Biochem. 35, 1183–1192. Heuer, H., Krsek, M., Baker, P., Smalla, K., Wellington, E.M.H.,
Beadle, N.C., 1968. Some aspects of the ecology and physiology of 1997. Analysis of actinomycete communities by specific ampli-
Australian xeromorphic plants. Aust. J. Sci. 30, 348–355. fication of genes encoding 16S rRNA and gel electrophoretic
Dinkelaker, B., Hengeler, C., Marschner, H., 1995. Distribution and separation in denaturing gradients. Appl. Environ. Microbiol.
function of proteoid roots and other root clusters. Bot. Acta 108, 63, 3233–3244.
183–200. Ibekwe, A.M., Kennedy, A.C., 1998. Fatty acid methyl ester
Ettema, C.H., Wardle, D.A., 2002. Spatial soil ecology. Trends Ecol. (FAME) profiles as a tool to investigate community composition
Evol. 17, 177–183. of two agricultural soils. Plant Soil 206, 151–161.
Frankenberger, W.T., Tabatabai, M.A., 1991. Factors affecting Jeffrey, D.W., 1967. Phosphate nutrition of Australian heath plants.
L-asparaginase activity in soils. Biol. Fertil. Soils 11, 1–5. I. The importance of proteoid roots in Banksia (Proteaceae).
Gardner, W.K., Parbery, D.G., Barber, D.A., 1982. The acquisition Aust. J. Bot. 15, 403–411.
of phosphorus by Lupinus albus L. I. Some characteristics of the Jongman, R.H., ter Braak, C.J.F., Van Tongeren, O.F.R. (Eds.),
soil/root interface. Plant Soil 68, 19–32. 1995. Data Analysis in Community and Landscape Ecology.
Gardner, W.K., Barber, D.A., Parbery, D.G., 1983. The acquisition Cambridge University Press, Cambridge, UK.
of phosphorus by Lupinus albus L. III. The probable mechanism Jones, D.L., Darrah, P.R., 1994. Role of root derived organic acids in
by which phosphorus movement in the soil/root interface is the mobilization of nutrients from the rhizosphere. Plant Soil
enhanced. Plant Soil 70, 107–124. 166, 247–257.
Gerke, J., Römer, W., Jungk, A., 1994. The excretion of citric and Kandeler, E., Marschner, P., Tscherko, D., Gahoonia, T.S.,
malic acid by proteoid roots of Lupinus albus L., effects on soil Nielsen, N.E., 2002. Microbial community composition and
solution concentrations of phosphate, iron, and aluminum in the functional diversity in the rhizosphere of maize. Plant Soil
proteoid rhizosphere in samples of an Oxisol and a Luvisol. Z. 238, 301–312.
Pflanzenern. Bdkde. 155, 339–343. Klein, D.A., Frederick, B.A., Biodini, M., Trlica, M.J., 1988.
Germida, J.J., Sicillano, S.D., Defreitas, J.R., Seib, A.M., 1998. Rhizosphere microorganisms effects on soluble amino acids,
Diversity of root-associated bacteria associated with field-grown sugars and organic acids in the root zone of Agropyron cristatum
canola (Brassica napus L.) and wheat (Triticum aestivum L.). A. smithii and Bouteloua gracilis. Plant Soil 110, 19–25.
FEMS Microb. Ecol. 26, 43–50. Kourtev, P.S., Ehrenfeld, J.G., Häggblom, M.M., 2003. Experimen-
Gilbert, G.A., Knight, J.D., Vance, C.P., Allan, D.L., 1999. Acid tal analysis of the effect of exotic and native plant species on the
phosphatase activity in phosphorus-deficient white lupin roots. structure and function of soil microbial communities. Soil Biol.
Plant Cell Environ. 22, 801–810. Biochem. 35, 895–905.
Grayston, S.J., Vaughan, D., Jones, D., 1996. Rhizosphere carbon Ladd, J.N., Butler, J.H.A., 1972. Short-term assays of soil proteo-
flow in trees in comparison with annual plants: the importance of lytic enzyme activities using proteins and dipeptide derivatives
root exudation and its impact on microbial activity and nutrient as substrates. Soil Biol. Biochem. 4, 19–30.
availability. Appl. Soil Ecol. 5, 29–56. Lamont, B.B., 1973. Factors affecting the distribution of proteoid
Grierson, P.F., 1992. Organic acids in the rhizosphere of Banksia roots within the root system of two Hakea species. Aust. J. Bot.
integrifolia L.f. Plant Soil 144, 259–265. 21, 165–187.
Grierson, P.F., Adams, M.A., 2000. Plant species affect acid phos- Lamont, B., McComb, A.J., 1974. Soil microorganisms and the
phatase, ergosterol and microbial P in a jarrah (Eucalyptus formation of proteoid roots. Aust. J. Bot. 22, 681–688.
marginata Donn ex Sm) forest in south-western Australia. Soil Lamont, B.B., 1993. Why are hairy roots clusters so abundant in the
Biol. Biochem. 32, 1817–1827. most nutrient-impoverished soils of Australia? In: Barrow, N.J.
Grierson, P.F., Attiwill, P.M., 1989. Chemical characteristics of the (Ed.), Plant Nutrition—From Genetic Engineering to Field
proteoid root mat of Banksia integrifolia L.f. Aust. J. Bot. 37, Practice. Kluwer Academinc Publishers, Dordrecht, The Nether-
137–143. lands, pp. 309–312.
Griffiths, B.S., Bonkowski, M., Dobson, G., Caul, S., 1999. Changes Lamont, B.B., 2003. Structure, ecology and physiology of root
in soil microbial community structure in the presence of micro- clusters—a review. Plant Soil 248, 1–19.
bial-feeding nematodes and protozoa. Pedobiology 43, 297–304. Liljeroth, E., Bååth, E., Mathiasson, I., Lundborg, T., 1990. Root
Groves, R.H. 1964. Experimental studies on heath vegetation. PhD exudation and rhizoplane bacterial abundance of barley
Thesis. University of Melbourne. (Hordeum vulgare L.) in relation to nitrogen fertilization and
Groves, R.H., Keraitis, K., 1976. Survival and growth of seedlings of root growth. Plant Soil 127, 81–89.
three sclerophyll species at high levels of phosphorus and Malajczuk, N., Bowen, G.D., 1974. Proteoid roots are microbially
nitrogen. Aust. J. Bot. 24, 681–690. induced. Nature 251, 316–317.
Grundon, N.J., 1972. Mineral nutrition of some Queensland heath Marschner, P., Yang, C.-H., Lieberei, R., Crowley, D.E., 2001a. Soil
plants. J. Ecol. 60, 171–181. and plant specific effects on bacterial community composition in
Häussling, M., Leisen, H., Marschner, H., 1990. Gradienten von the rhizosphere. Soil Biol. Biochem. 33, 1437–1445.
pH-Werten und Nährstoffaufnahmerated bei Langwurzeln von Marschner, P., Crowley, D.E., Lieberei, R., 2001b. Arbuscular
Fichten (Picea abies (L.) Karst.) unter kontrollierten Umwelt- mycorrhizal infection changes the bacterial 16S rDNA commu-
P. Marschner et al. / Applied Soil Ecology 28 (2005) 191–201 201
nity composition in the rhizosphere of maize. Mycorrhiza 11, Roelofs, R.F.R., Rengel, Z., Cawthray, G.R., Dixon, K.W., Lambers,
297–302. H., 2001. Exudation of carboxylates in Australian Proteaceae:
Marschner, P., Neumann, G., Kania, A., Weisskopf, L., Lieberei, R., chemical composition. Plant Cell Environ. 24, 891–903.
2002. Spatial and temporal dynamics of bacterial community Rosenfield, C.L., Reed, D.W., Kent, M.W., 1991. Dependency of
composition in the rhizosphere of cluster roots of white lupin iron reduction on development of an unique root morphology in
(Lupinus albus L.). Plant Soil 246, 167–174. Ficus benjamina L. Plant Physiol. 95, 1120–1124.
Marschner, P., Kandeler, E., Marschner, B., 2003. Structure and Rovira, A.D., 1959. Root excretions in relation to the rhizosphere
function of the soil microbial community in a long-term ferti- effect. IV. Influence of plant species, age of plant, light,
lizer experiment. Soil Biol. Biochem. 35, 453–461. temperature, and calcium nutrition on exudation. Plant Soil
Marschner, P., Crowley, D.E., Yang, C.H., 2004. Development of 11, 53–64.
specific rhizosphere bacterial communities in relation to plant Rumberger, A., Marschner, P., 2003. Phenylethylisothiocyanate
species, nutrition and soil type. Plant Soil 261, 199–208. liberated by living canola roots affects the microbial community
Merbach, W., Mirus, E., Knof, G., Remus, R., Ruppel, S., Russow, composition in the rhizosphere. Soil Biol. Biochem. 35, 445–
R., Gransee, A., Schulze, J., 1999. Release of carbon and 452.
nitrogen compounds by plant roots and their possible ecological Shane, M.W., De Vos, M., De Roock, S., Cawthray, G.R., Lambers,
importance. J. Plant Nutr. Soil Sci. 162, 373–383. H., 2003. Effects of external phosphorus supply on internal
Miethling, R., Wieland, G., Backhaus, H., Tebbe, C.C., 2000. phosphorus concentration and the initiation, growth and exuda-
Variation of microbial rhizosphere communities in response tion of cluster roots in Hakea prostata R. Br. Plant Soil 248,
to crop species, soil origin, and inoculation with Sinorhizobium 209–219.
meliloti L. Microb. Ecol. 40, 43–56. Sheffield, V.C., Cox, D.R., Lerman, L.S., Myers, R.M., 1989.
Miethling, R., Ahrends, K., Tebbe, C.C., 2003. Structural differ- Attachment of a 40-base-pair G + C-rich sequence (GC-clamp)
ences in the rhizosphere communities of legumes are not equally to genomic DNA fragments by the polymerase chain reaction
reflected in community-level physiological profiles. Soil Biol. results in improved detection of single-base changes. Proc. Natl.
Biochem. 35, 1405–1410. Acad. Sci. U.S.A. 86, 232–236.
Nannipieri, P., Kandeler, E., Ruggiero, P., 2002. Enzyme activities Tabatabai, M.A., Bremner, J.M., 1969. Assay of urease activity in
and microbiological and biochemical processes in soil. In: soils. Soil Biol. Biochem. 4, 479–487.
Burns, R., Dick, R. (Eds.), Enzymes in the Environment: Activ- Taranto, M.T., Adams, M.A., Polglase, P.J., 2000. Sequential frac-
ity, Ecology, and Application. Marcel Dekker, New York, USA, tionation and characterization (31P NMR) of phosphorus-
pp. 1–34. amended soils in Banksia integrifolia (L.f.) woodland and
Neumann, G., Massonneau, A., Martinoia, E., Römheld, V., 1999. adjacent pasture. Soil Biol. Biochem. 32, 169–177.
Physiological adaptations to phosphorus deficiency during pro- Tarafdar, J.C., Jungk, A., 1987. Phosphatase activity in the rhizo-
teoid root development in white lupin. Planta 208, 373–382. sphere and its relation to depletion of soil organic phosphorus.
Neumann, G., Massonneau, A., Langlade, N., Dinkelaker, B., Biol. Fertil. Soils 3, 199–204.
Hengeler, C., Römheld, V., Martinoia, E., 2000. Physiological Waldrop, M.P., Balser, T.C., Firestone, M.K., 2000. Linking micro-
aspects of cluster root function and development in phosphorus- bial community composition to function in a tropical soil. Soil
deficient white lupin (Lupinus albus L.). Ann. Bot. 85, 909–919. Biol. Biochem. 32, 1837–1846.
Pate, J.S., Jeschke, W.D., Dawson, T.E., Raphael, C., Hartung, W., Westover, K.M., Kennedy, A.C., Kelley, S.E., 1997. Patterns of
Bowen, B.J., 1998. Growth and seasonal utilization of water and rhizosphere microbial community composition associated with
nutrients by Banksia prionoites. Aust. J. Bot. 46, 511–532. co-occurring plant species. J. Ecol. 85, 863–873.
Purnell, H.M., 1960. Studies of the family Proteaceae. I. Anatomy Zak, J.C., Willig, M.R., Moorhead, D.L., Wildman, H.G., 1994.
and morphology of the rots of some Victorian species. Aust. J. Functional diversity of microbial communities: a quantitative
Bot. 8, 38–50. approach. Soil Biol. Biochem. 26, 1101–1108.