The dynamics of oxygen concentration, pH value, and organic acids in the rhizosphere of Juncus spp

lable at ScienceDirect

Soil Biology & Biochemistry 43 (2011) 1186e1197

Contents lists avai

Soil Biology & Biochemistry

journal homepage: www.elsevier .com/locate/soi lbio

The dynamics of oxygen concentration, pH value, and organic acidsin the rhizosphere of Juncus spp.

Stephan Blossfeld a,*, Dirk Gansert b, Björn Thiele c, Arnd J. Kuhn a, Rainer Lösch d

a IBG-2: Plant Sciences, Forschungszentrum Jülich, D-52425 Jülich, GermanybDept. of Ecology & Ecosystem Research, University Göttingen, Untere Karspüle 2, D-37073 Göttingen, Germanyc IBG-2: Plant Sciences/BioSpec, Forschungszentrum Jülich, D-52425 Jülich, GermanydNebensteingasse 1, D-63739 Aschaffenburg, Germany

a r t i c l e i n f o

Article history:Received 3 February 2010Received in revised form20 December 2010Accepted 10 February 2011Available online 2 March 2011

Keywords:Planar optodesNon-invasive imagingRootesoil interfaceMinimal-invasive samplingMicrosuction capillariesWaterloggingHypoxiaRadial oxygen loss

* Corresponding author. Tel.: þ49 2461 61 4894; faE-mail address: [email protected] (S. Blossf

0038-0717/$ e see front matter � 2011 Elsevier Ltd.doi:10.1016/j.soilbio.2011.02.007

a b s t r a c t

A novel type of planar optodes for simultaneous optical analysis of pH and oxygen dynamics in therhizosphere is introduced. The combination of the optical, non-invasive measurement of these param-eters with sterile sampling of rhizosphere solution across and along growing roots by use of a novel typeof rhizobox provides a methodical step forward in the investigation of the physicochemical dynamics ofthe rhizosphere and its underlying matter fluxes between roots and soil. In this study, this rhizobox wasused to investigate the effect of oxygen releasing roots of three Juncus species on the amount anddistribution of organic acids in reductive, oxygen-deficient soils of different pH (pH 3.9epH 5.9).Pronounced diurnal variations of oxygen concentration and pH along the roots, particularly along theelongation zone were observed. Long-term records over more than eight weeks revealed considerablespatial and temporal patterns of oxygen over a range of almost 200 mmol O2 L�1 and pH dynamics of �1.4pH units in the rhizosphere. A strong effect of oxidative acidification due to oxygen release by the plantroots was clearly visible for Juncus effusus, whereas the roots of Juncus articulatus alkalinized therhizosphere. In contrast, roots of Juncus inflexus induced no effects on rhizospheric pH. Only fourdifferent organic acids (oxalate, acetate, formate and lactate) were detectable in all soil solutions.Maximal concentration of all organic acids occurred at pH 3.9, whereas the lowest concentration of eachorganic acid was found at pH 5.9. Hence, considering the pH-dependence of the redox potential, the acidsoil provided increased reductive conditions leading to slower anaerobic degradation of organic acids toCO2 or methane (CH4). The concentration of organic acids decreased by up to 58% within a distance ofonly 4 mm from the bulk soil to the root surface, i.e. reciprocal to the pronounced O2-gradient. Thedecreasing presence of organic acids toward the oxygen releasing roots is possibly due to a change in thecomposition of the microbial community from anaerobic to aerobic conditions. The present studyhighlights the dynamic interplay between O2 concentration, pH and organic acids as key parameters ofthe physicochemical environment of the rhizosphere, particularly for wetland plants growing in oxygen-deficient waterlogged soils.

� 2011 Elsevier Ltd. All rights reserved.

1. Introduction

In the rhizosphere, especially in those of wetland plants, impor-tant physicochemical parameters of biogeochemical processes areoxygen concentration ([O2]), pH value and redox potential (Eh). It iswell-known, that plant roots can actively alter the rhizospheric [O2],pH and Eh. For example, the roots of well adaptedwetland plants canrelease substantial amounts of oxygen into the rhizosphere without

x: þ49 2461 61 2492.eld).

All rights reserved.

suffering from hypoxia in the roots (Armstrong et al., 1992; Chabbiet al., 2000; Visser et al., 2000; Mainiero and Kazda, 2005). Oxida-tive processes in the roots of wetland plants are maintained due tothe formation of an aerenchyma and by a variety of plant internalventilation processes that supply the root tissues with atmosphericoxygen (Armstrong et al., 1996; Sorrell and Brix, 2003). On the otherhand, root-derived pH changes of the rhizosphere can also be quiteconsiderable. Rhizospheric pH changes ofmore than one pH unit areknown for aerated and also for hypoxic rhizospheres (Hinsingeret al., 2003; Blossfeld and Gansert, 2007; Bravin et al., 2009; Devauet al., 2009; Blossfeld et al., 2010). These pH changes are ofteninducedbyadeficiencyof nutrients (e.g. phosphate or iron), or by the

AA

501

05

581

56 501

53

004

300

°07

°53

081

03

35ØC

C

BD

E

051

08

08

-wercsseloh

IIIIII

Fig. 1. Front view of the rhizobox (distances and dimensions in mm). Twometal bodies(A) and two oblique metal bars (B) created equal volumes of three test-compartments(IeIII). The rhizobox was sealed by conventional rubber cord (C), and each test-compartment was equipped with a water inflow/outflow port (D). Four holes (E,d ¼ 53 mm) in the backside of the rhizobox granted access to the soil by insertion ofthe raster access ports (the drawing is not to scale).

S. Blossfeld et al. / Soil Biology & Biochemistry 43 (2011) 1186e1197 1187

type of nitrogen acquisition, i.e. nitrate versus ammonium nutrition(Lin, 1979; Marschner et al., 1987, 1991; Hinsinger, 2001; Hinsingeret al., 2005; Devau et al., 2009).

The plant-induced changes of these physicochemical parame-ters will of course dramatically affect the microbial community ofthe soil. For example, the oxygenation of a hypoxic wetland soil byplant roots will raise the redox potential toward oxidative condi-tions. This enables the oxidation of e.g. Fe(II) to Fe(III) in therhizosphere by chemical reactions, which in turn favors oxidizingbacteria like Acidithiobacillus ferroxidans or those that metabolizeorganic acids to CO2 and CH4 (Küsel et al., 2003; Weiss et al., 2003;Neubauer et al., 2007). On the contrary, depending on the amountof oxygen released into the hypoxic soil, anaerobic production oforganic acids and methane is hampered or even ceases (Laanbroek,2010). Hence, wetland plants alter the structure and abundance ofthe microbial community in the rhizosphere along with essentialbiogeochemical processes like acidogenesis, acetogenesis ormethanogenesis, including the production and consumption oforganic acids (Jones et al., 2003; Lu et al., 2006, 2007). However, theprocesses of oxygen release or pH modification by plant roots arenot constant over time and do not occur all along the entire rootsystem (Marschner and Römheld, 1983; Jaillard et al., 1996; Chabbiet al., 2000; Blossfeld and Gansert, 2007). As a result, the growingplant roots create a dynamic and heterogeneous rhizosphere withmicro niches that can show extreme differences in pH or [O2] overa short distance. However, a detailed non-invasive and simulta-neous mapping of the dynamic interplay of rhizospheric [O2] andpH is a major challenge for rhizosphere research. Most of theexisting rhizosphere mapping methods and techniques provideonly qualitative and static information for one of the two param-eters (Marschner and Römheld, 1983; Chabbi et al., 2000). Othersprovide quantitative and dynamic information, but they needartificial agar substrates (Jaillard et al., 1996). Since few years a newtechnology, called planar optodes, is available, which allowsa quantitative mapping of rhizosphere [O2] or pH dynamics innatural soils (Jensen et al., 2005; Frederiksen and Glud, 2006;Blossfeld and Gansert, 2007; Blossfeld et al., 2010). However, upto now there was no technique available, which allows simulta-neous mapping of rhizospheric [O2] and pH dynamics.

In this study, we examined the spatial and temporal dynamics ofoxygen concentration, pH value and the abundance of organic acidsin the rhizosphere of three wetland plant species of the genusJuncus, simultaneously. The genus Juncus is spread worldwide andoffers a variety of differently adapted species. In this particularstudy we investigated Juncus effusus, Juncus inflexus and Juncusarticulatus, because these species are common in Central Europeand they exhibit strong differences in their aerenchyma structure.The oxygen and pH measurements were done by application ofa new, non-invasive O2epH hybrid optode imaging technology(Gansert et al., 2006; Gansert and Blossfeld, 2008). Furthermore,we present a new type of temperature-regulated rhizobox for free-choice root growth in differentially treated soil compartments (i.e.a modified split root system) equipped with microsuction rasteraccess ports in order to identify the effect of roots on the amountand distribution of organic acids in reductive soils of different pH(pH 3.9epH 5.9).

This combination of a non-invasive imaging technology anda minimal-invasive sampling method allows the study of theinteraction of root structure and physiology with soil processes. Inparticular, the present study provides evidence of species specificand considerable spatialetemporal dynamics of oxygen release andpH changes by the roots of Juncus spp. that create conditionscompletely different from the bulk soil, along with contrastingzones of production and consumption of organic acids that char-acterize the microenvironment of the particular rhizospheres.

2. Materials and methods

2.1. Plant species

The investigated plant species J. effusus L., J. inflexus L., andJ. articulatus L. are common wetland plants in Central Europe, andthus, were selected as the living test organisms in this study. Onemajor anatomical difference between these species is the structureof the aerenchyma. The aerenchyma of J. effusus is distinguished bya spongy type pith, whereas the aerenchyma spaces of J. inflexusand J. articulatus are separated by several walls into caverns ofdifferent size. The average volume of such caverns in the culms ofJ. inflexus is 1.9 mm3, whereas that of J. articulatus is 44.1 mm3. Theplants were grown in pots with commercially available peatsubstrate (Stender AG, Schermbeck, Germany, soil type E400)under waterlogged conditions for at least two weeks before start ofthe experiments, and waterlogged conditions were maintainedthroughout all investigations.

2.2. Novel compartmented rhizobox for free-choice root ingrowth

The temperature-controlled and compartmented rhizobox(Fig. 1) was made of anodized aluminum plates (thickness ¼ 8 mm)with an acrylic glass front plate (thickness ¼ 3 mm Rhöm GmbH,Darmstadt, Germany). The inner dimensions of the box were406 � 300 � 30 mm in width, height and depth, respectively. Twotrapezoid metal bodies (Fig. 1 “A”, surface area 115.5 cm2, thickness30 mm) and two oblique metal bars (Fig. 1 “B”, width ¼ 30 mm,length ¼ 150 mm, thickness ¼ 8 mm) formed three test-compart-ments (IeIII) of equal volume (V ¼ 550 cm3), to be used forinvestigations in the rhizosphere of growing roots under differen-tially manipulated soil conditions. A covering soil layer, the parentsoil compartment (V ¼ 450 cm3), in equidistance from the test-compartments provided equal conditions for root growth in thebeginning of the experiment. The trapezoid metal bodies increasedthe surface for thermal conductivity, and thus, considerablyimproved the thermal homogeneity in the rhizobox. Each test-compartment was equipped with a raster access port (Fig. 2 ‘RAP’),and a water inflow/outflow port, to be used for individual regula-tion of the water level (Fig. 1 “D”). Two fluid coolers with internal

°54

O

Q

R

P

LW

oR

N

SoS

PF

F

G

G

PAR

OH

FO

Fig. 2. Side view of the rhizobox. Raster access ports (RAP) were placed through circular ring adapters (F, di ¼ 53 mm) close to the roots (Ro) at the front plate (FP). Water level (WL)was constantly 3 cm above soil surface (SoS). Tubing connectors (O) connected steel capillaries (N) to a vacuum sampling chamber (P) for sampling of soil solution. Vacuum wasgenerated by a vacuum pump (Q) and controlled by a valve (R). The two cooling devices (G) were connected to a water bath. The O2epH hybrid optode (HO) was scanned fromoutside via a movable optical fiber (OF). The drawing is not to scale (distances and dimensions in mm).

S. Blossfeld et al. / Soil Biology & Biochemistry 43 (2011) 1186e11971188

fin structure (FLKU 140, Fischer Elektronik GmbH, Lüdenscheid,Germany) were fixed from outside to the back of the rhizobox,below and above the raster access ports, covering 53.3% surfacearea of the back plate of the rhizobox (Fig. 2 “G”). A thermalconductance paste (Conrad Electronic, Hirschau, Germany)increased the thermal conductivity between the fluid coolers andthe rhizobox. The fluid coolers were connected in series to a waterbath and flowed through with deionized water at 20 �C. For indi-vidual tilting the rhizobox was mounted on a tiltable mount(Blossfeld and Gansert, 2007).

2.2.1. Raster access portsThe raster access ports (RAP) were constructed for sterile

sampling of small quantities of both bulk soil and rhizospheresolution in the test and parent compartments. The RAP could beinserted into the rhizobox at arbitrary depths (0e30 mm) throughthe sealing adapter (Fig. 2 “F”) of the holes in the back plate. TheRAP contained raster plates with a 1 mm raster of 34 � 34 holes(d ¼ 0.7 mm) for taking up steel capillaries which serve as micro-suction cups (see below). The rotation and moving of the rasteraccess ports allowed an optimized positioning of the micro capil-lary raster in close contact with roots (0.5 mm distance) in order toderive transects of samples across and along roots (for additionalconstruction information see Supplementary material).

2.2.2. Vacuum sampling of soil solutionFor minimal-invasive sampling of soil solution, a series of

stainless steel capillaries (Fig. 2 “N”, do ¼ 0.6 mm, di ¼ 0.4 mm,length ¼ 30 cm, SWS Edelstahl GmbH, Emmingen, Germany) wereinserted in the raster plates penetrating a sealing film but stoppedin the multi-step holes of the raster plates 0.5 mm in front ofa sterile filter membrane (for additional information seeSupplementary material). For prevention of blockage of the capil-laries by intrusion of sealing film a steel wire (d ¼ 0.4 mm) wasinserted into each capillary prior to its insertion into the rasterplates, which was removed after positioning of the capillary. Due to

the elasticity of the sealing film it affixed to the steel capillariesmaintaining its function as a watertight sealant.

Each steel capillary was connected to a vacuum samplingchamber (Göttlein et al., 1996) via 10 mm Ø tubing connectors(Fig. 2 “O”, di ¼ 0.5 mm, Pharmed, Saint-Gobain PerformancePlastics Verneret, Charny, France). The sampling chamber wasconstructed with 100 ports available (Fig. 2 “P”). A vacuum of300 hPa below ambient pressure (¼700 hPa absolute) was appliedto this chamber by a vacuum pump (Fig. 2 “Q”, Laboport N 86 KT.18,KNF Neuberger GmbH, Freiburg, Germany). The vacuum wasregulated by a manual valve (Fig. 2 “R”) that was connectedbetween the pump and the vacuum sampling chamber. Thesampled soil and rhizosphere solution (250e500 ml) was collectedin standard 2 ml test tubes, positioned in the sampling chamberunderneath of each of the 100 ports. Sampling time was short(30e45 min). After sampling of the solution, the test tubes wereimmediately closed and stored in a freezer (�20 �C) until they wereused for capillary electrophoresis (CE) analysis.

2.3. Non-invasive measurement of O2 and pH by hybrid optodes

2.3.1. Measuring principle and calibrationThe optical measurement of O2 concentration and pH value by

planar O2epH hybrid optodes is based on the measurement of thefluorescence decay times of O2-, and pH-sensitive indicator dyes(Huber et al., 2001; Klimant et al., 2001; Gansert et al., 2006;Blossfeld and Gansert, 2007; Gansert and Blossfeld, 2008). Oxy-genepH hybrid optodes represent a one layeretwo particle systemof analyte sensitive indicator dyes. In brief, the pH indicator dye iscovalently attached to ion permeable polymer particles, whereasthe oxygen indicator dye is electrostatically bound to the oxygenpermeable polymer nanoparticles. These pH and oxygen sensitiveparticles are dispersed in an ion and oxygen permeable polymerlayer (Gansert et al., 2006). A multi-frequency approach has beenapplied for the O2 and pH measurement (Borisov et al., 2006;Gansert et al., 2006), i.e. both the indicator dyes have overlapping

S. Blossfeld et al. / Soil Biology & Biochemistry 43 (2011) 1186e1197 1189

absorption and emission spectra. Thereby, simultaneous excitationof both indicator dyes is possible, and one emission filter is chosento pass the emission light from both indicator dyes. For separationof the O2- and pH-specific signals, measurements are done at twodifferent modulation frequencies, and mathematical models areused for data separation. Simultaneous excitation of both indicatorsis performed by a blue 470 nm LED, and emission light of theindicators is filtered through an OG 530 long pass filter.

Calibration of the O2epH hybrid optode was done by a combi-nation of the calibration procedures described elsewhere for the pHand O2 optodes (Blossfeld and Gansert, 2007). (for additional cali-bration protocols see Supplementary material.)

2.3.2. Experimental setupAs described elsewhere (Blossfeld and Gansert, 2007; Blossfeld

et al., 2010) a pH-1-Mini system (PreSens GmbH, Regensburg)was connected with a stepper motor device for non-invasivemeasurement of O2 concentration ([O2]) and pH from outside,through the acrylic glass front plate of the rhizobox. In each rhi-zobox four O2epH hybrid optodes (35 mm � 40 mm; PreSensGmbH) were fixed with conventional silicon grease to the innerside of the acrylic glass plate, each opposite to a raster access port.The scanned surface of an O2epH hybrid optode was generally setto 60% of the total sensor size, in order to maintain a constantrepetition period of 60 min. Across this scanned surface 105sampling points were sampled per hour, by using a stepper motorstep size of 3 mm. The data from these axial and radial gradients ofpH and [O2] along the roots and across the rhizospheres werevisualized by color contour plots (SigmaPlot 11, Systat Software Inc.,San Jose, CA, USA).

2.4. Analysis of organic acids by capillary electrophoresis (CE)

Capillary electrophoresis with a salicylate electrolyte(Bazzanella et al., 1997) was applied for identification and quanti-fication of organic acids sampled from the bulk soil and from therhizosphere. The instrumental parameters are listed in theSupplementary materials.

Problems arose from measurement of acetate. Presence oforganic acids in laboratory air, as impurities in chemicals, onglassware and on human skin is well-known (Peldszus, 2007).Thus, the accumulation of organic acids was tested in blanksamples, revealing that the concentration of acetate increased frombelow detection limit up to 87.0 mM during performance of theanalysis sequences. Hence, samples derived from the bulk soil andthe rhizosphere were corrected for the drift in acetate concentra-tion. Calibration lines for the organic acids investigated were linearin the range between 5 and 20 mM with excellent correlationcoefficients (r ¼ 0.9965e0.9991). The limits of detection (LOD, 3s)were about 0.5 mM.

2.5. Variation of soil pH in the test-compartments of the rhizobox

In order to investigate the effect of wetland plant roots on theabundance of organic acids in submerged soils of different pH, thetest-compartments IeIII of the rhizobox were filled witha commercially available peat substrate (E400, A600, E910, StenderAG, Schermbeck, Germany) with pH values of 5.5, 3.9, and 5.9,respectively. The parent soil compartment was filled with the samesubstrate as was used in the test-compartment I (E400, pH 5.5).Prior to filling of the rhizobox, the RAPs were positioned 1 mm offthe front plate, and were pushed forward by 0.5 mm after fillingwas completed. In each compartment three temperature sensors(Pt 1000) were installed at three depths (top, center and bottom)and temperature was recorded at 15 min intervals. To maintain

equal nutrient conditions in all test-compartments the rhizoboxwas watered with Knop standard nutrient solution (Ca(NO3)21000 mg L�1, MgSO4 250 mg L�1, KH2PO4 250 mg L�1, KNO3250 mg L�1, FeSO4 20 mg L�1) during the experiment. To simulatewetland conditions, the rhizobox was kept waterlogged by use ofthe nutrient solution. Light conditions in the laboratory were keptconstant at 700 mmol photons m�2 s�1 during a 14/10 h dayenightcycle (HIT 400 W, Norka GmbH & Co. KG, Hamburg, Germany).These light conditions were proven to be sufficient for maintainingphotosynthesis for all three plant species (Blossfeld, 2008).The mean air temperature varied between 21.8 � 0.6 �C at nightand 23.8 � 1.6 �C during daytime. By use of the fluid coolerssoil temperature at all depths was nearly constant at 19.9 �Cthroughout the experiment, and showed negligible diurnal varia-tion (19.8e19.9 �C).

Plants of J. effusus, J. inflexus and J. articulatus were grown inseparate rhizoboxes for three weeks before the first sampling ofrhizosphere and soil solution in the parent soil compartmentoccurred (25 samples per species). Twoweeks later, a second seriesof samples were derived from test-compartment II (50 samples perspecies). Up to that time no roots were grown across the rasterplates in the other two test-compartments. However, samples fromthe bulk soil of the test-compartment I and III were taken duringthis sampling period. During themeasurements the steel capillarieswere placed in 1 mm raster intervals for evaluation of the spatialresolution of this new sampling technique.

3. Results

None of the investigated plants grew into the substrate adjustedat pH 5.9 (test-compartment III) and, furthermore, J. articulatus alsoavoided the acidic substrate of pH 3.9 (test-compartment II). Bothfindings indicate a small pH optimum of J. articulatus, and thepreference of the rush species for growth in moderately acidic soilconditions.

3.1. Qualification of the O2epH hybrid optode

The continuous use of the planar O2epH hybrid optode for morethan eight weeks during each experiment revealed a high stabilityof the O2- and pH-sensitive indicator dyes (lumiophores) merged inthe hybrid optode. However, the indicator dyes showed a slightdrift of the phase angle F so that a recalibrationwas necessary aftereach experiment. The drift of F ranged from 0.42� to 3.57�,depending on pH and oxygen concentration (for additional infor-mation see Supplementary material). To compensate this drift,a linearized adaptation model of the calibration coefficients duringthe experiment was developed. The preliminary calibration coef-ficients were gradually changed over time by adding or subtractinga coefficient-specific increment until they equaled the results of therecalibration after the experiments. This resulted in different cali-bration coefficients per day which were used for the iterationmodels.

3.2. Variation of pH and O2 concentration in the rhizosphere

The use of the non-invasive, and thus, disturbance-free O2epHhybrid optode revealed, that the roots of all three investigated plantspecies strongly oxidized the rhizosphere (Figs. 3e6, Table 1).While the root tips approached, the oxygen concentration in thesoil raised from strong oxygen-deficient conditions in the rangebetween 11.4 and 38.3 mmol O2 L�1, i.e. between 4 and 13% of airsaturation, respectively, to strong oxidative conditions of up to198 mmol O2 L�1, corresponding to an air saturation of about 70% inthe case of J. effusus (Table 1). The highest rhizospheric oxygen

Fig. 3. Visualization of a series of pH and O2 profiles across a growing root of Juncus effusus, rooting in E400-soil. The figure illustrates snapshots at different times during the timeseries from day 3 of the experiment (D3) to day 4 (D4). Illumination started at 0800 h and ended at 2200 h each day. The colors indicate different pH and O2 values. The black linesindicate the position of the root.

S. Blossfeld et al. / Soil Biology & Biochemistry 43 (2011) 1186e11971190

concentration for J. inflexus and J. articulatus root tips were 184.7and 101.3 mmol O2 L�1, respectively, corresponding to an air satu-ration of about 66% and 36% (Table 1). In general, O2 release fromlateral roots was higher than from main roots, but species-specificdifferences were observed. For example, lateral roots of J. articulatusdid not release oxygen, while those of J. effusus and J. inflexusshowed high O2 concentrations in the rhizosphere of up to nearly200 mmol O2 L�1. By contrast, the O2 concentration in the bulk soilaveraged at 25 mmol O2 L�1 (Table 1).

J. effusus and J. inflexus showed an oxygen release pattern quitedifferent from that of J. articulatus. For J. effusus and J. inflexus,oxygen release was restricted to the elongation zone of the roots,whereas the roots of J. articulatus released oxygen along the entireroot surface (Figs. 3e6). Fig. 3 shows a series of repetitivemeasurements of the same section of a hybrid optode (4.08 cm2,25.5% of total sensor area). This series illustrates the growth ofa root of J. effusus and the related changes of the pH and O2concentration. Within 10 h (Day 3 of experiment 1040 he2040 h),the root grew 8.2 mm across the optode, corresponding to anaverage growth rate of 0.82mmh�1. Due to the root growth and therestriction of radial oxygen loss (ROL) to the elongation zone, theoxygenation of this particular section of the submerged soil wasonly temporary. After 10 h of root growth and O2 release (Fig. 3, D32040 h), the maximal oxygenation of this soil section was reached(120 mmol O2 L�1). During the following hours, oxygen releaseceased from the basal parts of the roots toward the direction of themoving root tip (Fig. 3, D4 0500 heD4 0900 h). As a result, the

oxidative zone around the root tips moved in a cone-shaped formthrough the strongly hypoxic soil.

The same pattern of temporary oxygenation of the soil wasobserved for J. inflexus (Fig. 4). However, the duration of theoxygenation was increased due to a slower root growth rate ofJ. inflexus (6.5mmh�1) by about 50% (Table 1). In contrast to the rootsof J. effusus and J. inflexus, the roots of J. articulatus continuouslyreleased oxygen into the soil. Hence, the oxygen concentration in therhizospherewaspermanently higher than thatof thebulk soil (Fig. 5).

Concomitantly with the formation of lateral roots, the roots ofJ. inflexus and J. effusus released large amounts of oxygen into the soil.This second phase of oxygen release was evenmore prominent thanthe oxygenation by the root tips because the oxygenation by lateralroots was permanent. Moreover, the oxygenation by lateral rootsshowed a pronounced diurnal rhythm. Around the lateral rootsdistinctly higher oxygenconcentrationsweremeasured compared tothe oxygenation by the root tips (Table 1). Fig. 6 exemplarily illus-trates the formation of the lateral roots of J. inflexus, accompanied bya strong oxygen release into the soil.Within about 3 h after the onsetof illumination the O2 concentration increased by 30 mmol L�1,corresponding to an O2 release rate of 2.8 nmol L�1 s�1.

In contrast to the findings in alkaline soil substrate (Blossfeld andGansert, 2007), no continuous acidification along the roots growingin these acidic soil substrates was observed for any of the threeinvestigated plant species. Nevertheless, a temporary rhizosphericacidificationwas detected for the roots of J. effusus (Fig. 3, D4 0500 h).This rhizospheric acidification was coupled to the rhizospheric

Fig. 4. Visualization of a series of pH and O2 profiles across a growing root of Juncus inflexus, rooting in E400-soil. The figure illustrates snapshots at different times during the timeseries from day 17 of the experiment (D17) to day 18 (D18). Illumination started at 0800 h and ended at 2200 h each day. The colors indicate different pH and O2 values. The blacklines indicate the position of the root.

S. Blossfeld et al. / Soil Biology & Biochemistry 43 (2011) 1186e1197 1191

oxygenation. During the transit of the elongation zone, the averagerhizospheric pH value dropped e.g. from pH 5.9 to pH 5.5 and roseback quickly to the initial value afterward (Fig. 3). Due to this shortterm reaction, the rhizosphere pH was not stable over time. Incontrast to J. effusus, the average rhizospheric pH value of J. inflexusroots was not lowered by the oxygen release along the elongationzone (pH 5.3) compared to the situation shortly before the approachof the root tip (pH 5.3, Fig. 4). The roots of J. articulatus did not acidify,but alkalinized the rhizosphere (pH¼ 5.6) compared to the bulk soil(pH¼ 5.5), although the rhizospheric oxygen concentrationwasfive-fold higher than the bulk soil oxygen concentration (Table 1, Fig. 5).

3.3. Organic acids in the bulk soil and in the rhizosphere

3.3.1. Organic acids in the bulk soilDespite considerable amounts of inorganic anions (e.g.

carbonate, nitrate, sulfate, chloride, etc.) as matrix compounds inthe bulk soil solution, the organic acids oxalate, formate, acetate,

and lactate, could clearly be separated and identified by CE analysis.Other common organic acids that often occur in soils e.g. citratewere not detectable.

The distribution of organic acids in the bulk soil (Table 2) high-lighted a strong variation of quality and quantity of organic acids inthe different soil substrates. Even in the same soil substrate a vari-ation of the concentration of a given organic acid could be observed.For example, the concentration of lactate varied from 14.6 mM to116.5 mM in the test-compartment ‘II’, depending on the plantspecies growing in the respective rhizobox (i.e. J. effusus, J. inflexus orJ. articulatus). Formate was detectable in all test-compartments ofthe J. inflexus rhizoboxes and in test-compartment ‘II’ of theJ. articulatus rhizoboxes, but not in any test-compartment of therhizoboxes with J. effusus. However, the total amount of organicacids in the different test-compartments and rhizoboxes was ina comparable range (concentrations 75.9 mMe91.2 mM), except forthe test-compartments ’II’. These latter ones showed strongervariations of the total amount of organic acids, ranging from92.2 mM

Fig. 5. Visualization of a series of pH and O2 profiles across a growing root of Juncus articulatus, rooting in E400-soil. The figure illustrates snapshots at different times during thetime series from day 7 of the experiment (D7) to day 9 (D9). Illumination started at 0800 h and ended at 2200 h each day. The colors indicate different pH and O2 values. The blacklines indicate the position of the root.

S. Blossfeld et al. / Soil Biology & Biochemistry 43 (2011) 1186e11971192

to 208.2 mM. This is basically attributable to large changes in theformate concentration (<LOD to 45.2 mM) and lactate concentration(14.6 mMe116.5 mM)within these low-pH test-compartments of thedifferent rhizoboxes.

3.3.2. Organic acids in the rhizosphereAs in the bulk soil, oxalate, formate, lactate and acetate were

detected also in the rhizosphere of the investigated rush species.Other organic acids commonly found in the rhizosphere of plants,such as citrate or malate, were not detected.

Presence of roots of all three rush species influenced thedistribution of the organic acids. In these cases, a strong gradient ofdecreasing total organic acid concentration was found from thebulk soil toward the root surface. For example, a sampling transectof 1 mm-intervals across three roots of J. inflexus growing in theparent soil compartment showed that the highest concentration oflactate occurred 2 mm off the root (96.2 and 77.3 mM at positions 1and 5 respectively, Fig. 7), while at the root surface lactate wasreduced to 28.1 mMand 17.7 mM, respectively (Fig. 7, positions 3 and9). A similar decrease effects of the roots on the presence of formateconcentration was also observed around root 2 that crossed thesampled area, where formate dropped from 39.5 to 20.7 mM fol-lowed by an increase to 26.4 mMat a distance of 2 mm from the rootsurface (Fig. 7, positions 7e11). However, root 1 showed a lesspronounced radial profile of formate concentration. Similar effectsof plant roots on the organic acid concentrations were also

observed for J. effusus and J. articulatus (Table 3). Lactate was theprevailing acid in the rhizosphere of J. inflexus in the soil substratewith pH 5.5 (32.5 mM, test-compartment ‘P’/‘I’), while for J. effususand J. articulatus in this soil the concentrations were only 19.6 mMand 17.5 mM, respectively. This difference was even morepronounced in the acidic soil substrate (pH 3.9, test-compartment‘II’), where lactate and formate were present in considerableamounts in the rhizosphere of J. inflexus, while the concentration ofthese acids remained below the limit of detection in the rhizo-sphere of J. effusus. Regarding acetate, the reverse pattern wasfound with a strongly reduced acetate concentration in the rhizo-sphere of J. inflexus (19 mM).

4. Discussion

4.1. Non-invasive optical O2 and pH measurements

The O2epH hybrid optode technology allows without anydisturbance a simultaneous quantification and visualization of theroot-mediated heterogeneity of O2 and pH dynamics of biologicaland physicochemical processes in the rhizosphere. For the inter-pretation of the driving forces of the O2 and pH dynamics, the ratiobetween the diffusion and the convection rate of molecular oxygen(O2) and protons (Hþ) in the submerged soil has to be determined(Walter et al., 2009). A guide for interpretation is the so-calledPéclet number P (Kim and Silk, 1999).

Fig. 6. Visualization of a series of pH and O2 profiles across the lateral roots of Juncus inflexus L, rooting in A600-soil. The figure illustrates snapshots at different times during thetime series from day 40 of the experiment (D40) to day 41 (D41). Illumination started at 0800 h and ended at 2200 h each day. The colors indicate different pH and O2 values. Theblack lines indicate the position of the root.

S. Blossfeld et al. / Soil Biology & Biochemistry 43 (2011) 1186e1197 1193

P ¼ V L D�1; (1)

where V is the root growth rate (mm h�1), L is the root diameter(mm) and D is the diffusion coefficient of O2 (7.2 mm2 h�1,Greenwood,1967) and protons (33.84mm2 h�1, Kim and Silk,1999),respectively, in waterlogged soil. A Péclet number much lowerthan 1 implies that diffusion of O2 and protons acts to produce

a rhizosphere that is wide relative to the root radius. When P ismuch larger than 1, the convection of O2 and protons by thegrowing roots dominates, so that the rhizosphere is narrow. In thisstudy, the Péclet numbers for molecular oxygen were calculated as0.065, 0.056 and 0.085 for J. effusus, J. inflexus and J. articulatus, andfor protons 0.015, 0.013 and 0.019, respectively. These Pécletnumbers indicate that in all experiments diffusion will produce

Table 1Effects of roots of Juncus effusus, J. inflexus and J. articulatus on rhizosphere and bulk soil pH and O2 concentration [O2]. Maximal oxidation time means the duration ofoxygenation of a single spot in the soil by passage of a growing root tip. (‘P’/‘I’/‘II’ indicate the different soil compartments).

Soil Numberof roots

Max [O2] mainroot, mmol O2 L�1

Max [O2] lateralroots, mmol O2 L�1

Mean [O2] in bulksoil, mmol O2 L�1

Mean pHrhizosphere

Mean pHbulk soil

Max oxidationtime, h

Juncus effusus (did not grow in compartment ‘III’)‘II’ 5 174.1 198.1 32.6 � 15.1 4.52 � 0.67 5.11 � 0.16 31‘P’/‘I’ 16 159.4 167.3 22.5 � 10.7 5.49 � 0.47 5.46 � 0.18 28

Juncus inflexus (did not grow in compartment ‘III’)‘II’ 8 164.5 184.7 38.3 � 12.9 4.75 � 0.27 4.93 � 0.26 49‘P’/‘I’ 27 170.3 179.7 20.6 � 14.5 5.31 � 0.16 5.41 � 0.17 40

Juncus articulatus (did not grow in compartments ‘II’ and ‘III’)‘P’/‘I’ 9 101.3 n.d. 11.4 � 4.5 5.63 � 0.06 5.44 � 0.09 Permanent

S. Blossfeld et al. / Soil Biology & Biochemistry 43 (2011) 1186e11971194

a wide acidified and oxygenized rhizosphere around the roots as itwas measured indeed and is shown in Figs. 3e5.

The non-invasive studies demonstrate that the dynamics ofrhizospheric O2 concentration and pH changes are species specific.For example the results clearly visualize that the roots of twoinvestigated rush species (J. inflexus and J. effusus) release largeamounts of oxygen into the submerged soil mainly along the rootapex and the lateral roots. The advancing root tips of these twospecies in fact move a cone-shaped oxic zone through the soil,while the basal parts of the roots and the bulk soil remain underhypoxic to anoxic conditions. The lateral roots on the other handform awide diurnally fluctuating oxic zone. Even during nighttimesthe rhizosphere is partly aerobic due to either diffusion or venti-lation processes like for example Venturi-convection that areindependently operative from photosynthesis (Armstrong et al.,1992, 1996; Soukup et al., 2007). In the case of J. articulatus onthe other hand, oxygen is continuously released into the soil alongthe complete surface of the roots including the basal parts, whereasno lateral roots are formed. These different patterns of oxygenrelease from the main and lateral roots over time indicate anotherspecies-specific factor that determines oxygen dynamics in therooted soil, governed by species-specific peculiarities. This meansthat the long-termmosaic-like pattern of oxygen and consequently,the pHeEh relationship as well as the spatial differences in micro-bial communities in the soil depend on root-specific processes.

Moreover, our results clearly link spatialetemporal dynamics ofoxygen release with another key parameter of the biogeochemistryof the waterlogged soils, the pH value. The considerable input ofoxygen from the roots into the soil, e.g. up to 200 mmol O2 L�1 maystoichiometrical cause an acidification of the affected soil area, dueto the oxygenation of reduced ions, especially Fe2þ (Eqn. (2); Begget al., 1994; Kirk and Bajita, 1995).

4Fe2þ þ O2 þ 10H2O54FeðOHÞ3þ8Hþ (2)

However, from the rhizosphere pH measurements we foundthat high oxygen concentrations there did not always correspond to

Table 2Distribution of organic acids in the bulk soil of the parent soil compartment ‘P’ (i.e. the samthree investigated species. Limit of detection (LOD) was reached at 0.5 mM (‘P’/’I’/ ‘II’ ind

Organicacid

Mean concentration of organic acids (mM), n ¼ 10

J. inflexus J. effusus

‘P’/‘I’ ‘II’ ‘III’ ‘P’/‘I’ ‘II

Oxalate 10.4 � 7.1 36.9 � 9.4 9.0 � 6.7 23.3 � 5.0 36Formate 25.6 � 7.7 45.2 � 13.7 27.1 � 11.4 <LOD <

Lactate 50.0 � 26.3 36.8 � 24.3 47.1 � 18.0 21.8 � 12.5 14Acetate <LOD 42.4 � 26.0 8.0 � 5.2 31.5 � 7.7 41Sum 86.0 161.3 91.2 76.6 92

a strong local oxidative acidification although all older parts of theroots of all species showed a brownish color (i.e. most likely due toiron oxides). While with J. effusus indeed a slight oxidative acidifi-cation of the strongly oxidized rhizosphere occurred, the rhizo-sphere pH of J. inflexus was not affected and J. articulatus evenrevealed an alkalinization of the oxidized rhizosphere. Due to therestriction of oxygen release to the root tips and lateral roots theoxidative acidification along the roots of J. effusus is not permanent(Visser et al., 2000; Van der Welle et al., 2007). Moreover, thecortical barrier also restricts the uptake and release of solutes(Soukup et al., 2007), which could be another reason for theabsence of root-derived pH changes along the proximal rootsections of J. effusus. In contrast to the roots of J. effusus, the roots ofJ. inflexus and J. articulatus show a stronger buffering effect (foradditional information see Supplementary material). Interestingly,although the pattern of the root-derived radial oxygen loss (ROL) issimilar and the range of ROL differs only slightly (maximalDO2 ¼ 13.4 mmol O2 L�1), the impact of the ROL of J. inflexus on thebulk soil is stronger than the impact of the ROL of J. effusus. This ismainly attributable to the slower root growth rates, and thereforelonger oxidation times of J. inflexus roots, causing a higher oxygencontent in the bulk soil coupled with a stronger acidification. This isremarkable, because the ROL of J. articulatus never ceases after theapproach of the root tip. This also causes a continuous increase ofthe bulk soil oxygen content. However, this is accompanied by anonly weak acidification of the bulk soil.

These results point out that the roots of the investigated plantspecies buffer the oxidative acidification differently during phasesof oxygen release, and that this buffering effect is even strongerthan in the bulk soil. Due to these species-specific patterns ofrhizosphere oxygenation and the resulting heterogeneity ofoxidative and reductive micro-areas along the roots, a similar smallscaled heterogeneity of the pH must also be taken into account(Blossfeld and Gansert, 2007).

The cause for the small scaled heterogeneity of the pH foundwith the investigated rush species is the diversity of biotic and

e substrate as in test-compartment ‘I’) and the test-compartments ‘II’ and ‘III’ of theicate the different soil compartments).

J. articulatus

’ ‘III’ ‘P’/‘I’ ‘II’ ‘III’

.4 � 19.3 29.6 � 6.5 24.1 � 10.9 15.5 � 9.1 21.1 � 4.0LOD <LOD <LOD 6.3 � 3.6 <LOD.6 � 4.4 31.7 � 19.0 45.1 � 24.7 116.5 � 54.4 37.8 � 21.5.2 � 20.2 27.3 � 15.2 20.2 � 6.0 69.9 � 35.3 17.0 � 5.1.2 88.6 89.4 208.2 75.9

Fig. 7. Distribution pattern of organic acids along two linear transect across two roots of Juncus inflexus grown in the parent soil compartment at pH 5.5. The abscissa indicatessampling positions in mm. The concentrations of the organic acids were stacked up per each sampling point. The numbers in each part of the columns indicate the singleconcentrations of the different organic acids per sampling point. The positions of both root surfaces (root 1 and 2) are indicated by arrows pointing at the corresponding column.

S. Blossfeld et al. / Soil Biology & Biochemistry 43 (2011) 1186e1197 1195

abiotic reactions in the soil. For example, the consumption ofoxygen by the microbiota can quickly reduce the availability ofoxygen, and therefore, reduce the amount of protons generated byabiotic oxidation processes. Some studies report that up to 50% ofroot-released oxygen can be consumed by microorganisms(Howeler and Bouldin, 1971; Kirk and Solivas, 1994). On the otherhand, protons can be consumed as well by several processes in thesoil. One important process is the uptake of carbon dioxide byaerenchymatous roots (Begg et al., 1994; Greenway et al., 2006).Under anaerobic conditions the fermentative production of carbondioxide strongly increases the partial pressure of CO2 compared tothe atmosphereeplant gradient (>5 kPa; Greenway et al., 2006).This steep concentration gradient is a driving force for CO2 uptakeby plant roots and CO2 transport via the aerenchyma toward thecarbon assimilating shoot organs and the atmosphere (Higuchiet al., 1984; Brix, 1990; Constable et al., 1992). CO2 dissolved inthe rhizosphere solution is in equilibrium with HCO3

� or CO32�,

depending on the pH value (Eqn. (3)). Within the relevant pH rangeof this study (pH 4epH 6), dissociation of dissolved CO2 equilibratesbetween HCO3

� and CO2aq. Hence, during a continuous uptake of

CO2 by aerenchymatous roots the balance between HCO3� and

CO2aq (H2CO3, is not stable and indicated in Eqn. (3) as H2CO3

*) isshifted toward [CO2

aq], consuming one proton per CO2 taken up.Furthermore, in the course of water uptake into the plants, CO2 canbe taken up as HCO3

� or CO32� as well.

Table 3Distribution of organic acids in the rhizosphere of the selected rush species growingin the parent soil compartment ‘P’ (i.e. the same soil substrate as used in test-compartment ‘I’), and the test-compartments ‘II’ and ‘III’ of the three investigatedspecies. Limit of detection (LOD) was reached at 0.5 mM (‘P’/‘I’/‘II’ indicate thedifferent soil compartments).

Organicacid

Mean concentration of organic acids (mM), n ¼ 10

J. inflexus J. effusus J. articulatus

‘P’/‘I’ ‘II’ ‘P’/‘I’ ‘II’ ‘P’/‘I’

Oxalate 9.7 � 3.5 26.6 � 10.6 22.4 � 2.0 29.5 � 5.3 23.2 � 6.2Formate 17.9 � 3.6 26.5 � 7.8 <LOD <LOD <LODLactate 32.5 � 12.2 22.8 � 16.2 19.6 � 5.2 <LOD 17.5 � 0.8Acetate <LOD 19.0 � 15.2 9.9 � 5.4 51.7 � 12.2 19.2 � 9.4Sum 40.1 94.9 51.9 81.2 59.9

COaq2 þ H2O5H2CO

*35HCO�

3 þHþ5CO2�3 þ 2Hþ (3)

As stated elsewhere (Colmer, 2003; Greenway et al., 2006;Hinsinger et al., 2009), there is still a lack of data providingunequivocal information about CO2 effects on the pH of therhizosphere.

Another possible factor that might hamper the effect of soilacidification could be excretion of bases or uptake of protons by theroots. It is well-known that during the uptake of nutrients likenitrate (NO3

�) e other anions (e.g. HCO3�, OH�, organic anions) are

excreted (or protons are taken up) by the plant roots to maintaincationeanion balance (Marschner et al., 1991; Marschner, 1995;Hinsinger et al., 2003). These processes can cause an alkaliniza-tion of the rhizosphere and affect the pH of the bulk soil as well(Blossfeld et al., 2010). However, under wetland conditions nitrateuptake is only possible, if nitrifying bacteria accompany the aerobiczones of the rhizosphere in the generally anaerobic wetland soils.However, nitrification generates two protons per NH4

þ molecule(Eqn. (4); Scheffer and Schachtschabel, 2002), and only one OH�

will be excreted by the roots during the uptake of one moleculeNO3

�.

NHþ4 þ 2O25NO�

3 þ H2Oþ 2Hþ (4)

Thus, the net balance should still cause an acidification, ifnitrification and nitrate uptake processes would be predominant inthe rhizosphere. In the present study nitrate was continuouslysupplied by watering with the standard Knop nutrient solution.Therefore NO3

� uptake was possible without the presence ofnitrifying bacteria.

4.2. Organic acids in the bulk soil and in the rhizosphere

With concern to organic acids one major result of this study isthat under strong as compared withmoderate acidic soil conditions(pH 3.9 versus 5.5) their total quantities are higher by up to a factorof three. The high amount of organic acids indicates reducedmicrobial degradation of organic acids to CO2 or CH4 due to the lackof inorganic electron acceptors (Fe(III), nitrate, sulfate, etc.) atanaerobic conditions (Yao et al., 1999; Küsel et al., 2003; Lu et al.,

S. Blossfeld et al. / Soil Biology & Biochemistry 43 (2011) 1186e11971196

2006). It is well-known, that especially the amount of nitrate andsulfate is quickly reduced within the first days after the onset ofwaterlogging (Kirk, 2004). Sulfate reducers are in competition withFe(III) reducers for acetate, but the lack of black spots on the rootsor in the bulk soil indicates that sulfate reduction played onlya minor role in our study. Furthermore, according to the EhepHstability field of iron compounds (Scheffer and Schachtschabel,2002; Kirk, 2004), iron occurs in its reduced form (i.e. Fe(II)) atstrong acidic and anaerobic conditions. The lack of Fe(III) as anelectron acceptor appears to be a limiting factor in anaerobicmicrobial metabolism, that co-determines the catabolism oforganic acids as intermediates in the anaerobic degradation oforganic matter toward CH4 and CO2 (Dannenberg and Conrad,1999; Hori et al., 2007; Koelbener et al., 2010; Laanbroek, 2010).Although some fermentation processes might be still operative atlow pH (Wust et al., 2009), the dissimilatory activity and growth ofFe(III)-reducing bacteria, such as Geobacter spp. and Anaeromyx-obacter spp. (Petrie et al., 2003; Lu et al., 2006; Hori et al., 2007),which are able to oxidize acetate (Hori et al., 2007), will be reducedat low pH. Therefore, the pH-dependent shift in composition andconcentration of organic acids in the anoxic bulk soil pinpoints thesignificance of pH and redox potential on the microbial processdynamics of anaerobic organic matter degradation with respect toCH4 and CO2 emission from submerged soils.

A pronounced gradient of decreasing concentration of organicacids was found from the bulk soil toward the roots. Since the rhi-zobox was continuously flooded throughout the experiment, thebulk soil was hypoxic to strictly anoxic in all compartments,providing the appropriate conditions for anaerobic microbialproduction of organic acids. By release of large amounts of oxygeninto the rhizosphere, thus raising the oxygen level from anoxia tooxic conditions of up to almost 200 mmol O2 L�1, the roots of the rushspecies effect directly on the production and consumption of organicacids (Laanbroek, 2010). Conditions change from anoxic to oxic lifeconditions on a micro-scale. Hence, in the outer rhizosphereperiphery anaerobic bacteria can produce organic acids like lactate(e.g. Lactobacillus) and acetate (e.g. Acetobacterium). Proximal to theroot surfaces the activity of the anaerobic microflora declines, whilethe activity of aerobic microflora increases, mirrored by lowconcentrations of the organic acids. But not all roots andnot even thesame root release similar amounts of oxygen at the same time, thusa spatially and temporally heterogeneous pattern of oxygen releaseand acid turnover is common for all three investigated plant species.

The intrusion of oxygen from the roots into the anaerobicenvironment will raise the redox potential and thus, favor theoxidation of reduced iron. Also these dynamics can be mediated bymicrobial biota. For example, Neubauer et al. (2007) proved that Fe(II) is oxidized to Fe(III) by iron oxidizing bacteria at the root surfaceof J. effusus. Enhanced iron oxidation can led to a coating of theroots with iron plaques as often observed on wetland plant roots(Chabbi et al., 2001; Küsel et al., 2003; Neubauer et al., 2007).Furthermore, the improved availability of oxidized iron in theproximate environment of the roots of wetland plants will supportthe metabolic activity and the growth of Fe(III)-reducing bacteria,and therefore, facilitate a rapid microbial-mediated cycling of ironin the rhizosphere (Küsel et al., 2003;Weiss et al., 2003). Hence, theactivity of Fe(III)-reducing bacteria, able to oxidize organic acids asmentioned above, will also contribute to the decrease of organicacids like lactate and acetate in an oxygenized rhizosphere. Infreshwater sediments rooted by J. effusus Fe(III) reduction accoun-ted for 65% of total carbon metabolism in the rhizospherecompared to 22% methanogenesis (Roden andWetzel, 1996). Theseauthors stated that Fe(III) oxide reduction could mediate a consid-erable amount of organic carbon oxidation and significantlysuppress CH4 production in freshwater wetlands.

5. Conclusions

The technical approaches presented in this study allow newinsights into the plantesoil interactions, especially with regard tothe dynamics of root-derived changes of the pH and oxygen micro-pattern in the rhizosphere. These approaches provide new resultsfor further understanding of the dynamic interplay between rootactivities and the rhizosphere microbial community in soils undernearly natural conditions. For example, the present study revealthat the roots of the investigated Juncus species (J. effusus, J. inflexusand J. articulatus) form a dynamic transition zone from oxic tohypoxic conditions, highly variable over space and time. This inturn determines the distribution and movement of the microbialcommunity, both aerobes and anaerobes, indicated by changingzones of production and consumption of organic acids in therhizosphere. Deduced from these first results, it becomes evidentthat the use of non-invasive and automated quantitative imagingmethods in combinationwith minimal-invasive sampling tools willpave the way for an advanced understanding of the influence ofplant roots on the network of the rhizosphere biota and the rela-tions between physicochemical soil conditions and biotic activitieslike the production and consumption of greenhouse gases.

Acknowledgments

Wewould like to express our thanks to the precisionmechanicalengineers W. Seidel, A. Lanzinger and M. Laug for their support inconstructing the rhizobox, and the raster access ports. We alsothank the PreSens GmbH for supporting uswith planar optodes andequipment. Additionally, we appreciate the very constructive workof the chief editor J. Schimel and two anonymous reviewers.

Appendix. Supplementary material

Supplementary material related to this article can be found atdoi:10.1016/j.soilbio.2011.02.007.

References

Armstrong, J., Armstrong, W., Beckett, P.M., 1992. Phragmites australis: venturi- andhumidity-induced pressure flows enhance rhizome aeration and rhizosphereoxidation. New Phytologist 120, 197e207.

Armstrong, J., Armstrong, W., Beckett, P.M., Halder, J.E., Lythe, S., Holt, R., Sinclair, A.,1996. Pathways of aeration and the mechanisms and beneficial effects ofhumidity- and Venturi-induced convections in Phragmites australis (Cav.) Trin.ex Steud. Aquatic Botany 54, 177e197.

Bazzanella, A., Lochmann, H., Mainka, A., Bächmann, K., 1997. Determination ofinorganic anions, carboxylic acids and amino acids in plant matrices by capillaryzone electrophoresis. Chromatographia 45, 59e62.

Begg, C.B.M., Kirk, G.J.D., MacKenzie, A.F., Neue, H.-U., 1994. Root-induced ironoxidation and pH changes in the lowland rice rhizosphere. New Phytologist 128,469e477.

Blossfeld, S., Gansert, D., 2007. A novel non-invasive optical method for quantitativevisualization of pH dynamics in the rhizosphere of plants. Plant, Cell andEnvironment 30, 176e186.

Blossfeld, S., Perriguey, J., Sterckeman, T., Morel, J.-L., Lösch, R., 2010. RhizospherepH dynamics in trace-metal-contaminated soils, monitored with planar pHoptodes. Plant and Soil 330, 173e184.

Blossfeld, S., 2008. Plant growth dynamics in relation to soil moisture, oxygenconcentration and pH-value. In: Mathematisch Naturwissenschaftliche Fakul-tät. Heinrich-Heine Universität, Düsseldorf, p. 144.

Borisov, S.M., Neurauter, G., Schroeder, C., Klimant, I., Wolfbeis, O.S., 2006. Modifieddual lifetime referencing method for simultaneous optical determination andsensing of two analytes. Applied Spectroscopy 60, 1167e1173.

Bravin, M.P., Tentscher, T., Rose, J., Hinsinger, P., 2009. Rhizosphere pH gradientcontrols copper availability in a strongly acidic soil. Environmental Science &Technology 43, 5686e5691.

Brix, H., 1990. Uptake and photosynthetic utilization of sediment-derived carbon byPhragmites australis (Cav.) Trin. ex Steudel. Aquatic Botany 38, 377e389.

Chabbi, A., McKee, K.L., Mendelssohn, I.A., 2000. Fate of oxygen losses from Typhadomingensis (Typhaceae) and Cladium jamaicense (Cyperaceae) and conse-quences for root metabolism. American Journal of Botany 87, 1081e1090.

S. Blossfeld et al. / Soil Biology & Biochemistry 43 (2011) 1186e1197 1197

Chabbi, A., Hines, M.E., Rumpel, C., 2001. The role of organic carbon excretion bybulbous rush roots and its turnover and utilization by bacteria under ironplaques in extremely acid sediments. Environmental and Experimental Botany46, 237e245.

Colmer, T.D., 2003. Long-distance transport of gases in plants: a perspective oninternal aeration and radial oxygen loss from roots. Plant, Cell and Environment26, 17e36.

Constable, J.V.H., Grace, J.B., Longstreth, D.J., 1992. High carbon dioxide concentra-tions in aerenchyma of Typha latifolia. American Journal of Botany 79, 415e418.

Dannenberg, S., Conrad, R., 1999. Effect of rice plants on methane production andrhizospheric metabolism in paddy soil. Biogeochemistry 45, 53e71.

Devau, N., Cadre, E.L., Hinsinger, P., Jaillard, B., Gérard, F., 2009. Soil pH controls theenvironmental availability of phosphorus: experimental and mechanisticmodelling approaches. Applied Geochemistry 24, 2163e2174.

Frederiksen, S.M., Glud, R.N., 2006. Oxygen dynamics in the rhizosphere of Zosteramarina: a two-dimensional planar optode study. Limnology and Oceanography51, 1072e1083.

Gansert, D., Blossfeld, S., 2008. The application of novel optical sensors (Optodes) inexperimental plant ecology. Progress in Botany, 333e358.

Gansert, D., Stangelmayr, A., Krause, C., Borisov, A.Y., Wolfbeis, O.S., Arnold, A.,Müller, A., 2006. Hybrid optodes (HYBOP). In: Popp, J., Strehle, M. (Eds.), Bio-photonics: Visions for a Better Health Care. Wiley VCH, Weinheim, pp. 477e518.

Göttlein, A., Hell, U., Blasek, R., 1996. A system for microscale tensiometry andlysimetry. Geoderma 69, 147e156.

Greenway, H., Armstrong, W., Colmer, T.D., 2006. Conditions leading to high CO2(>5 kPa) in waterlogged-flooded soils and possible effects on root growth andmetabolism. Annals of Botany 98, 9e32.

Greenwood, D.J., 1967. Studies on the transport of oxygen through the stems androots of vegetable seedlings. New Phytologist 66, 337e347.

Higuchi, T., Yoda, K., Tensho, K., 1984. Further evidence for gaseous CO2 transport inrelation to root uptake of CO2 in rice plant. Soil Science and Plant Nutrition 30,125e136.

Hinsinger, P., Plassard, C., Tang, C., Jaillard, B., 2003. Origins of root-mediated pHchanges in the rhizosphere and their responses to environmental constraints:a review. Plant and Soil 248, 43e59.

Hinsinger, P., Gobran, G.R., Gregory, P.J., Wenzel, W.W., 2005. Rhizosphere geometryand heterogeneity arising from root-mediated physical and chemical processes.New Phytologist 168, 293e303.

Hinsinger, P., Bengough, A., Vetterlein, D., Young, I., 2009. Rhizosphere: biophysics,biogeochemistry and ecological relevance. Plant and Soil 321, 117e152.

Hinsinger, P., 2001. Bioavailability of soil inorganic P in the rhizosphere as affectedby root-induced chemical changes: a review. Plant and Soil 237, 173e195.

Hori, T., Noll, M., Igarashi, Y., Friedrich, M.W., Conrad, R., 2007. Identification ofacetate-assimilating microorganisms under methanogenic conditions in anoxicrice field soil by comparative stable isotope probing of RNA. Applied andEnvironmental Microbiology 73, 101e109.

Howeler, R.H., Bouldin, D.R., 1971. The diffusion and consumption of oxygen insubmerged soils. Soil Science Society of America Journal 35, 202e208.

Huber, C., Klimant, I., Krause, C., Wolfbeis, O.S., 2001. Dual lifetime referencing asapplied to a chloride optical sensor. Analytical Chemistry 73, 2097e2103.

Jaillard, B., Ruiz, L., Arvieu, J.-C., 1996. pH mapping in transparent gel using colorindicator video densitometry. Plant and Soil 183, 85e95.

Jensen, S.I., Kühl, M., Glud, R.N., Jorgensen, L.B., Priemé, A., 2005. Oxic microzonesand radial oxygen loss from roots of Zostera marina. Marine Ecology ProgressSeries 293, 49e58.

Jones, D.L., Dennis, P.G., Owen, A.G., van Hees, P.A.W., 2003. Organic acid behavior insoils e misconceptions and knowledge gaps. Plant and Soil 248, 31e41.

Kim, T.K., Silk, W.K., 1999. A mathematical model for pH patterns in the rhizo-spheres of growth zones. Plant, Cell and Environment 22, 1527e1538.

Kirk, G.J.D., Bajita, J.B., 1995. Root-induced iron oxidation, pH changes and zincsolubilization in the rhizosphere of lowland rice. New Phytologist 131, 129e137.

Kirk, G.J.D., Solivas, J.L., 1994. Coupled diffusion and oxidation of ferrous iron insoils: III. Further development of the model and experimental testing. EuropeanJournal of Soil Science 45, 369e378.

Kirk, G.J.D., 2004. The Biogeochemistry of Submerged Soils. John Wiley & Sons,Chichester. 304.

Klimant, I., Huber, C., Liebsch, G., Neurauter, G., Stangelmayr, A., Wolfbeis, O.S., 2001.Dual lifetime referencing (DLR) e a new scheme for converting fluorescenceintensity into a frequency domain or timeedomain information. In: Valeur, B.,Brochon, J.C. (Eds.), New Trends in Fluorescence Spectroscopy: Application toChemical and Life Sciences. Springer, Berlin, pp. 257e275.

Koelbener, A., Ström, L., Edwards, P., Olde Venterink, H., 2010. Plant species frommesotrophic wetlands cause relatively high methane emissions from peat soil.Plant and Soil 326, 147e158.

Küsel, K., Chabbi, A., Trinkwalter, T., 2003. Microbial processes associated with rootsof bulbous rush coated with iron plaques. Microbial Ecology 46, 302e311.

Laanbroek, H.J., 2010. Methane emission from natural wetlands: interplay betweenemergent macrophytes and soil microbial processes. A mini-review. Annals ofBotany (London) 105, 141e153.

Lin, W., 1979. Potassium and phosphate uptake in corn roots: further evidence foran electrogenic Hþ/Kþ exchanger and an OH�/Pi antiporter. Plant Physiology 63,952e955.

Lu, Y., Rosencrantz, D., Liesack, W., Conrad, R., 2006. Structure and activity ofbacterial community inhabiting rice roots and the rhizosphere. EnvironmentalMicrobiology 8, 1351e1360.

Lu, Y., Abraham, W.-R., Conrad, R., 2007. Spatial variation of active microbiota in therice rhizosphere revealed by in situ stable isotope probing of phospholipid fattyacids. Environmental Microbiology 9, 474e481.

Mainiero, R., Kazda, M., 2005. Effects of Carex rostrata on soil oxygen in relation tosoil moisture. Plant and Soil 270, 311e320.

Marschner, H., Römheld, V., 1983. In vivo measurement of root induced pH changesat the soil-root interface: effect of plant species and nitrogen source. Zeitschriftfür Pflanzenphysiologie 111, 241e251.

Marschner, H., Römheld, V., Kissel, M., 1987. Localization of phytosiderophorerelease and of iron uptake along intact barley roots. Physiologia Plantarum 71,157e162.

Marschner, H., Häussling, M., George, E., 1991. Ammonium and nitrate uptake ratesand rhizosphere pH in non-mycorrhizal roots of Norway spruce [Picea abies (L.)Karst.]. Trees e Structure and Function 5, 14e21.

Marschner, H., 1995. Mineral Nutrition of Higher Plants. Academic Press, London,UK, 889 pp.

Neubauer, S.C., Toledo-Durán, G.E., Emerson, D., Megonigal, J.P., 2007. Returning totheir roots: iron-oxidizing bacteria enhance short-term plaque formation in thewetland-plant rhizosphere. Geomicrobiology Journal 24, 65e73.

Peldszus, S., 2007. Organic acids. In: Nollet, L. (Ed.), Handbook of Water Analysis.CRC Press, Boca Raton, pp. 393e408.

Petrie, L., North, N.N., Dollhopf, S.L., Balkwill, D.L., Kostka, J.E., 2003. Enumerationand characterization of iron(III)-reducing microbial communities from acidicsubsurface sediments contaminated with uranium(VI). Applied and Environ-mental Microbiology 69, 7467e7479.

Roden, E.E., Wetzel, R.G., 1996. Organic carbon oxidation and suppression ofmethane production by microbial Fe(III) oxide reduction in vegetated andunvegetated freshwater wetland sediments. Limnology and Oceanography 41,1733e1748.

Scheffer, F., Schachtschabel, P., 2002. Lehrbuch der Bodenkunde. Spektrum,Heidelberg.

Sorrell, B.K., Brix, H., 2003. Effects of water vapour pressure deficit and stomatalconductance on photosynthesis, internal pressurization and convective flow inthree emergent wetland plants. Plant and Soil 253, 71e79.

Soukup, A., Armstrong, W., Schreiber, L., Franke, R., Votrubová, O., 2007. Apoplasticbarriers to radial oxygen loss and solute penetration: a chemical and functionalcomparison of the exodermis of two wetland species, Phragmites australis andGlyceria maxima. New Phytologist 173, 264e278.

Van der Welle, M.E.W., Niggebrugge, K., Lamers, L.P.M., Roelofs, J.G.M., 2007.Differential responses of the freshwater wetland species Juncus effusus L. andCaltha palustris L. to iron supply in sulfidic environments. EnvironmentalPollution 147, 222e230.

Visser, E.J.W., Colmer, T.D., Blom, C.W.P.M., Voesenek, L.A.C.J., 2000. Changes ingrowth, porosity, and radial oxygen loss from adventitious roots of selectedmono- and dicotyledonous wetland species with contrasting types of aeren-chyma. Plant, Cell and Environment 23, 1237e1245.

Walter, A., Silk, W.K., Schurr, U., 2009. Environmental effects on spatial andtemporal patterns of leaf and root growth. Annual Review of Plant Biology 60.

Weiss, J.V., Emerson, D., Backer, S.M., Megonigal, J.P., 2003. Enumeration of Fe(II)-oxidizing and Fe(III)-reducing bacteria in the root zone of wetland plants:implications for a rhizosphere iron cycle. Biogeochemistry 64, 77e96.

Wust, P.K., Horn, M.A., Drake, H.L., 2009. Trophic links between fermenters andmethanogens in a moderately acidic fen soil. Environmental Microbiology 11,1395e1409.

Yao, H., Conrad, R., Wassmann, R., Neue, H.U., 1999. Effect of soil characteristics onsequential reduction and methane production in sixteen rice paddy soils fromChina, the Philippines, and Italy. Biogeochemistry 47, 269e295.