ARTICLE IN PRESS

0038-0717/$ - se

doi:10.1016/j.so

�CorrespondE-mail addr

Soil Biology & Biochemistry 40 (2008) 1496–1505

www.elsevier.com/locate/soilbio

Stability and composition of soil organic matter controlrespiration and soil enzyme activities

Peter Leinwebera,�, Gerald Jandla, Christel Bauma, Kai-Uwe Eckhardta, Ellen Kandelerb

aInstitute for Land Use, University Rostock, D-18051 Rostock, GermanybInstitute for Soil Science and Land Evaluation (310), Universitat Hohenheim, D-70593 Stuttgart, Germany

Received 1 August 2007; received in revised form 21 December 2007; accepted 8 January 2008

Available online 07 February 2008

Abstract

Relationships between soil organic matter (SOM) molecular composition, thermal stability and decomposability by soil enzymes

and microbes are largely unknown. We incubated soils from unfertilized and NPK-fertilized neighboring field plots of a long-term rye

(Secale cereale) monoculture experiment and investigated relationships between changes in the molecular-chemical composition of SOM,

the CO2 flux and the activities of enzymes. Pyrolysis-field ionization mass spectrometry (Py-FIMS) showed larger ion intensities in the

NPK-fertilized than in the unfertilized soil at start of the incubation, only small changes in composition and thermal stability in the

unfertilized soil, and a preferential reduction in thermally stable components as well as general shifts towards lower pyrolysis temperature

after three weeks of incubation in the NPK-treatment. We found evidence that thermally labile and stable proportions of various

compound classes were differently susceptible to decomposition, depending on the fertilization history of the soil. Irrespective of

fertilization treatment, peaks in xylanase activity after 7 days of incubation followed by decreasing values were reflected by the ratio of

xylan (m/z 114) to xylose (m/z 132) marker signals in the Py-FI mass spectra. Thus, the study proved that (1) SOM composition was

changed due to long-term rye cropping without and with NPK-fertilization, (2) the modified SOM composition affected the

decomposability and microbial parameters under optimized conditions and (3) the thermal properties of individual compound classes

derived from Py-FI mass spectra can be sensitive predictors of microbial decomposition.

r 2008 Elsevier Ltd. All rights reserved.

Keywords: Soil organic matter; Pyrolysis mass spectrometry; Microbial biomass; Long-term experiment; Fertilization

1. Introduction

Total content and the chemical composition of soilorganic matter (SOM) depend on the input from vegetationresidues, anthropogenic inputs such as compost, manure(Leinweber and Reuter, 1992; Schulten and Leinweber,1991), sewage (Leinweber et al., 1996) and air-borne dusts(Rumpel, 1999; Schmidt et al., 1996), and the soil manage-ment such as agricultural tillage, grazing and forest cutting(Magdoff and Weil, 2004). Whereas these inputs tend toincrease SOM contents, tillage and aeration favor organicmatter decomposition, and thus tend to decrease SOMcontents. However, stabilization reactions prevent parts ofthe SOM from microbial attack and maintain SOM levels

e front matter r 2008 Elsevier Ltd. All rights reserved.

ilbio.2008.01.003

ing author.

ess: peter.leinweber@uni-rostock.de (P. Leinweber).

in soil. Since the SOM is linked directly to the atmosphericCO2 and, thus, to the climate change (Schlesinger andAndrews, 2000; Lal, 2004), great research efforts arecurrently undertaken to elucidate mechanisms of SOMstabilization (e.g. Kogel-Knabner et al., 2008; Lutzowet al., 2008; Marschner et al., 2008).The stability of organic matter in soils is defined as

resistance to microbial decomposition and can be studiedby aerobic incubation experiments. This approach waswidely applied to study the decomposability of organicmatter from primary sources (Klimanek, 1990; Jensenet al., 2005), SOM from whole soil samples and physicalsoil fractions (Christensen, 1987; Gregorich et al., 1989;Leinweber, 1995; Schulten and Leinweber, 1999), anddissolved organic matter which enters soils (Kalbitz et al.,2003). The decomposability of crop residues in soil dependson the chemical composition such as lignin contents or

ARTICLE IN PRESSP. Leinweber et al. / Soil Biology & Biochemistry 40 (2008) 1496–1505 1497

lignin to plant nitrogen (N) ratio (Klimanek, 1990;Vanlauwe et al., 1996). Holocellulose carbon (C) andneutral detergent soluble plant N were found to be betterpredictors of C and N mineralization than lignin-relatedparameters (Jensen et al., 2005). Nevertheless, up to nowthe relationships between the chemical composition ofSOM, its decomposability and the component fluxesreleased are not fully understood (Ryan and Law, 2005).Pyrolysis-field ionization mass spectrometry (Py-FIMS)provided evidence for relationships between the chemicalcomposition and stability of dissolved organic matter andits decomposition in an incubation experiment (Kalbitzet al., 2003). Long-term monoculture cropping and thefertilization modified the chemical composition of SOM(Schmidt et al., 2000), but the consequences for stabilityand turnover are unknown.

Since microbial decomposition of SOM is mediated bysoil enzymes, detailed studies of these enzymes are centralin understanding the decomposition of organic matter(Sinsabaugh et al., 1994; Paul, 2007). For example, croprotations, fertilization and tillage influenced the activity ofvarious enzymes involved in the SOM turnover (Kandeleret al., 1999a,b,c; Vepsalainen et al., 2004). A few studiestried to link the activity of enzymes involved in C cyclingwith the chemical composition of the vegetation litter orsoils under long-term monoculture (Magid et al., 1997;Luxhøi et al., 2002). In an early state of decomposition,X-glucosidase activity was a very good predictor of releaseof low molecular weight organic compounds from the litterinto the adjacent soil (Poll et al., 2006). In the later phase ofdecomposition, chemical and physical properties of litterfrom Miscanthus controlled its stability. Whereas brownmaterial retained its rigid structure at sub-cellular levelover long time, green material was quickly invaded andtransformed into an amorphous tissue architecture bymicroorganisms (Luxhøi et al., 2002).

Although causal relationships between the chemicalcomposition of litter and SOM as important substratesfor microbial growth on the one hand and activities ofenzymes involved in the turnover of these substrates on theother appear highly probable and plausible, in-depthunderstanding of these relationships is still lacking.Methodological reasons may be the insensitivity ofanalytical methods for probing the chemical compositionof SOM, the indirect approaches to enzyme activitydeterminations and interference of decomposition reac-tions with neoformation of metabolites in one sample.

Therefore, the objectives of the present study were:

(1)

to investigate how long-term monoculture of rye underzero and NPK-fertilization altered the chemical com-position of SOM in a Haplic Phaeozem developed in asandy loess;

(2)

to study the decomposability of whole SOM and SOMfractions, differing in chemical composition and ther-mal stability, during short-term aerobic incubationunder optimal conditions; and

(3)

to ascertain if and how the molecular-chemicalcomposition of SOM and the activities of soil enzymesinvolved in the C-, N- and P-cycling reflected thedifferent decomposability of SOM in fertilized andunfertilized soils.

2. Materials and methods

2.1. Field experiment and soil samples

Field moist soil samples were taken in fall 2000 fromplots of the ‘Eternal Rye Cultivation’ experiment at Halle,Saxony-Anhalt, Germany. This field experiment on aHaplic Phaeozem was established in 1878 (Merbachet al., 2000). The plots sampled were uniformly croppedwith rye (Secale cereale L.). We sampled the Ap horizon(0–20 cm depth) of the treatments ‘unfertilized’ (U, withoutany fertilization since 1878) and mineral fertilization(NPK). Mineral N fertilization was increased during theexperimental period from 40 kg N (ha yr)�1 (1878–1990) to60 kg N (ha yr)�1 (since 1991). The P (24 kg (ha yr)�1) andK (75 kg (ha yr)�1) applications remained constant for thewhole experimental period.For a general chemical characterization sub-samples

were air-dried and sieved o2mm. The soils containedabout 10% clay, 18% silt and 72% sand (sandy loam). Thecontents of organic C and total N were 9.3 g kg�1 (U) and11.5 g kg�1 (NPK), and 0.7 g kg�1 (U) and 0.9 g kg�1

(NPK), respectively. The pH value was lower in theunfertilized (pH: 5.3) than in the fertilized soil (pH: 6.0).No lime was applied at these plots. For the aerobicincubation experiment and the subsequent analyses ofincubated samples (see below), the soil material was storedfrozen (�20 1C) and thawed in a fridge immediately priorto the experiment.

2.2. Aerobic incubation experiment

A soil weight of 200 g from each treatment, moistened to60% of water holding capacity, was incubated at 25 1Cunder aerobic conditions. The incubation experimentsstarted with 10 (treatment U) and 16 (treatment NPK)replicates for the soil samples collected from the rye plotsto enable destructive sampling after certain incubationperiods. Samples from the unfertilized soil received a Pfertilization equivalent to the concentration of DL-P in thetreatment NPK. This was done to prevent P limitationbecause the decomposition of organic C and N compoundswas to be studied. Prior to the incubated samples ambientair was sucked through an entrance filter filled with 400mLof 10% potassium hydroxide solution to remove CO2. TheCO2-free gas purged the CO2 formed during incubationand pumped subsequently to the analyzer. The deliveryrate of the pump was 200L h�1. The CO2 was determinedwith the infrared CO2 analyzer BERLINA (MSA-AUER,Berlin, Germany). In the treatment U each three replicates

ARTICLE IN PRESSP. Leinweber et al. / Soil Biology & Biochemistry 40 (2008) 1496–15051498

of the samples were removed from the incubation experi-ment after 7, 21 and 63 days. Since the data evaluationshowed the most intensive transformations in the initialperiod, in treatment NPK replicate samples for micro-biological and chemical analyses were removed after 2, 7,21 and 35 days. Sub-samples were immediately frozen orfreeze-dried for later determination of enzyme activitiesand Py-FIMS, respectively.

2.3. Determination of enzyme activities

Microbial biomass C and N were determined by thechloroform/K2SO4 direct extraction procedure of Vanceet al. (1987). The C and N concentrations in the extractswere measured with a DIMATOC 2000 with DIMA Nmodule (DIMATEC, Essen, Germany). The correctionfactor for carbon was kc ¼ 0.33 (Smith and Paul, 1990) andfor nitrogen kn ¼ 0.38 (Voroney and Paul, 1984). Theenzyme activities were determined in three replicatesamples of each treatment and sampling date. The activityof alkaline phosphomonoesterase (ALP, EC 3.1.3.1) wasexpressed as mmol p-nitrophenol released from addedp-nitrophenyl phosphate within 1 h at 37 1C using themethod of Tabatabai (1994). The activity of urease wasexpressed as mmol NH4-N released from added urea perhour within a total incubation period of 2 h at 37 1C usingthe method of Kandeler and Gerber (1988). The activitiesof invertase, and xylanase were expressed as mmol glucoseequivalent released from added saccharose or xylan within1 h at 50 1C during an incubation period of 3 or 24 h,respectively, using the method of Schinner and von Mersi(1990).

2.4. Py-FIMS

About 0.5mg of dry sample were thermally degraded inthe ion-source of a modified high-performance FinniganMAT 731 mass spectrometer. The samples were heated inhigh vacuum (10�4 Pa) from 110 to 700 1C in temperaturesteps of 10K per magnetic scan. Between two scans theemitter was flash-heated to 1500 1C. During 18min of totalregistration time, about 60 magnetic scans were recordedfor the mass range m/z 15–900. These were combined toobtain one thermogram of total ion intensity (TII) and anaveraged Py-FI mass spectrum. For each of the singlescans, the absolute and relative ion intensities of ten classesof chemical compounds in the organic matter werecalculated by summation of the ion intensities of 8–39indicator signals (Schulten and Leinweber, 1999) to obtainthermograms of their volatilization. We integrated thesethermograms over the whole temperature range to discusschanges in the abundance of compound classes, and overcertain temperature ranges to discuss changes in theabundance of differently stable pools within these com-pound classes. All Py-FIMS data were normalized per mgsample. This procedure was done for each of two to threereplicate measurements per soil sample and results were

averaged for statistical analyses. For detailed descriptionsof the Py-FIMS methodology and of statistical evaluationsof sample weight and residue, volatilized matter and TII;see Schulten (1996) and Sorge et al. (1993).

2.5. Statistical data evaluation

Standard errors were calculated for the microbiologicalparameters and the ion intensities, intensity ratios andabundance of compound classes from Py-FIMS. Compar-ison of respiration, microbial biomass and enzymeactivities in treatments U and NPK was done by ANOVAwith fertilization as main effect (Statistica, 6th ed.).

3. Results

3.1. Thermograms and Py-FI mass spectra

The thermograms of TII (Fig. 1, inserts) clearly showed(1) larger ion intensities in the NPK-fertilized than in theunfertilized soil at start of the incubation and (2) apreferential reduction in thermally stable components anda shift towards lower pyrolysis temperature at day 21 in theNPK-treatment. Furthermore, at start of the incubationthe soil from treatment NPK compared to U was relativelyenriched in phenols/lignin monomers (+2.6% TII) andalkylaromatics (+1.4% TII). In the Py-FI mass spectrathis was reflected, e.g. by the m/z signals 184, 194, 196, 208and 210 being more intensive in Fig. 1c than in Fig. 1a.Incubation for 21 days led to decreased proportions ofcarbohydrates (e.g. m/z 96), phenols/lignin monomers(e.g. m/z 124) and lignin dimers (e.g. m/z 260, 296, 310)in treatment U. Generally, the changes in chemicalcomposition during incubation were small in this treat-ment. More obvious was the change in spectral patternin the NPK-treatment. Preferential losses in phenols/lignin monomers (e.g. m/z 194, 196, 210), lignin dimers(e.g. m/z 260, 296, 310), lipids (e.g. m/z 230), alkylaro-matics (e.g. m/z 184) and N-containing compounds. On theother hand, carbohydrates (e.g. m/z 82, 84, 96) andpeptides (e.g. m/z 70) were relatively enriched uponincubation of the NPK-treated soil (see Fig. 1d andcompare with Fig. 1c).

3.2. CO2-C flux, microbial biomass and enzyme activities

during incubation

Generally, almost all soil microbial properties weresignificantly affected by the treatment without and withNPK-fertilization (Table 1). The cumulative CO2-C fluxeswere significantly larger in treatment NPK than in Uduring the whole incubation period (Table 1). At days 7and 21 the factors of increase due to NPK-fertilizationwere 3.9. Initially, the microbial biomass C and N werelarger in treatment U than in NPK. The NPK-fertilized soilhad a steep increase in microbial biomass C within the firsttwo days of incubation. In the treatment U the microbial

ARTICLE IN PRESS

2 0.062 0.06

96

1 200 400 600Temperature in °C

Inte

nsity96

110

252

220206

82 266280

230

1 200 400 600Temperature in °C

Inte

nsity

110

124

252

220206

82

246

266280

50 100 150 200 250 300 350 400 450 500

1.0

1.5 0.06

Inte

nsity

82110

96

208

252 2661.0

1.5 0.06

96

110

230194

210

266184

252

280

50 100 150 200 250 300 350 400 450 500m / z m / z

50 100 150 200 250 300 350 400 450 50050 100 150 200 250 300 350 400 450 500m / z m / z

Rel

ativ

e ab

unda

nce

in %

TII

0.5

200 400 600Temperature in °C

I

70280

294

196

0.5

200 400 600Temperature in °C

Inte

nsity

110

84

280

294

Fig. 1. Summed and averaged pyrolysis-field ionization mass spectra and thermograms of total ion intensity (TII, inserts) of soils from the rye

monoculture, unfertilized at: (a) day 0 and (b) day 21, and fertilized with NPK (c) day 0 and (d) day 21.

Table 1

Soil microbial and biochemical properties (means) during a 9-week (U) and 5-week (NPK) aerobic incubation of unfertilized (treatment U) and fertilized

(treatment NPK) soil from a Phaeozem Ap horizon under rye (n ¼ 3)

Property Treatment Duration of incubation (days)

0 2 7 21 35 63

Cumulative CO2-C flux (mg g�1) U 0 n.d. 154 (4) a 375 (13) a n.d. 669 (13)

NPK 0 255 (8) 610 (10) b 1481 (53) b 2758 (74) n.d.

Biomass C (mg g�1) U 211 (2) b n.d. 266 (3) b 213 (1) a n.d. 192 (6)

NPK 107 (2) a 375 (6) 221 (4) a 304 (4) b 178 (9) n.d.

Biomass N (mg g�1) U 53 (3) b n.d. 36 (6) a 32 (3) a n.d. 40 (7)

NPK 17 (1) a 35 (2) 36 (9) a 35 (2) a 45 (3) n.d.

Activity (mmol product gsoil�1 h�1)

Invertase U 2.4 (0.09) a n.d. 2.2 (0.01) a 2.2 (0.02) a n.d. 2.3 (0.04)

NPK 3.5 (0.01) b 4.0 (0.11) 4.4 (0.05) b 4.3 (0.03) b 4.4 (0.18) n.d.

Urease U 0.3 (0.01) a n.d. 0.2 (0.01) a 0.1 (0.03) a n.d. 0.1 (0.02)

NPK 0.5 (0.01) b 0.5 (0.01) 0.4 (0.12) b 0.3 (0.03) b 0.2 (0.02) n.d.

Alkaline phosphatase U 0.4 (0.01) a n.d. 0.4 (0.01) b 0.4 (0.01) b n.d. 0.3 (0.00)

NPK 0.5 (0.01) b 0.4 (0.01) 0.3 (0.01) a 0.3 (0.01) a 0.4 (0.05) n.d.

Standard errors are given in parentheses.

n.d., not determined. Different letters for the same property indicate significant differences between treatment U and NPK (Po0.05) at this sampling date.

P. Leinweber et al. / Soil Biology & Biochemistry 40 (2008) 1496–1505 1499

ARTICLE IN PRESSP. Leinweber et al. / Soil Biology & Biochemistry 40 (2008) 1496–15051500

biomass C also reached a maximum at the first samplingdate after the start of incubation (day 7) and decreased inthe following. After 21 days of incubation it wassignificantly smaller than in treatment NPK (Table 1).Microbial biomass N showed no clear trend in U butincreases in NPK. Fertilization caused increased activitiesof all investigated enzymes in the soil samples during thewhole incubation period (Table 1). With exception ofthe alkaline phosphatase the activities of the enzymescontinued to be higher in NPK than in U treatment duringthe whole incubation period. After 7 and 21 days ofincubation the alkaline phosphatase activity was lower inNPK than in U.

3.3. Explanation of CO2-C flux and enzyme activities by

abundance and stability of SOM compounds

To find out which compounds contributed most to theCO2-flux, we plotted the absolute ion intensities(countsmg�1) of 10 important compound classes (Schultenand Leinweber, 1999) versus the time of incubation.Furthermore, considering relationships between microbialdecomposition and thermal stability as revealed by Py-FIMS (Kalbitz et al., 2003), we separated the compoundclasses into thermally labile (volatilizedo400 to 500 1Cpyrolysis temperature) and stable (volatilized4400 to500 1C pyrolysis temperature) proportions. These valuesas well were plotted versus the time of incubation. First, themost intensive changes in the abundance of almost allcompound classes and their labile and stable proportionswere observed in the initial phase of the incubation.Second, there were scarcely any trends in compound classesand their thermally labile and stable proportions consistent

3000CO2-C flux UCO2-C flux NPK

25002

2000

(µg

g -1

)

1500

O2-

C fl

ux (

500

1000

CO

0

Days of incuba0 10 20 30

Fig. 2. CO2-C flux during the incubation experiment and ion intensities of th

from 460 to 700 1C, fertilization treatments U and NPK. Vertical bars indicat

for the two fertilization treatments. Fig. 2 shows a slightenrichment in stable lignin dimers nearly parallel to thecumulative CO2-C flux in treatment U. By contrast, intreatment NPK the thermally stable lignin dimers firstdecreased dramatically and then remained at a relativelyconstant level from day 2 on, whereas the cumulativeCO2-C flux increased permanently during the incubation.The stable and labile proportions of the investigated

N-containing compounds, which comprise heterocycles,nitriles and other N functions but not peptides (Schultenand Leinweber, 1999), showed distinctively reverse trendsin the NPK-treatment (Fig. 3). This was observed duringthe whole incubation period but with the strongest changesoccurring until day 7. By contrast, in the unfertilizedsoil both, labile and stable N-containing compoundsremained at a rather constant level during the incubation.For the period 7th–21st day the order for thermallydifferently stable N-containing compounds was stableU4stable NPKElabile NPK4labile U. After 35 daysthe labile proportions of N-containing compounds in theNPK-treatment became more prominent than the stableproportions.In general the stable and labile proportions of the other

compound classes remained at a rather constant level in thetreatment U, whereas nearly all compound classes in theNPK-treatment showed a decrease of stable proportionsand an increase in labile proportions (not shown). Enzymeactivities were correlated with the molecular-chemical com-position of SOM determined by Py-FIMS. The temporalvariations in xylanase activity were reflected by the ratio ofxylan (m/z 114) to xylose (m/z 132) marker signals (Fig. 4).Xylanase activity and the signal ratio m/z 114 : m/z 132 inboth treatments increased and reached a peak after 7 days

0.04Lignin dimers ULignin dimers NPK

0.03

g

6m

g-1

)un

ts x

10

6

0.02

ensi

ty (c

ou

0.01

Ion

inte

0 000.00

tion40 50 60 70

e thermally stable lignin dimer indicator signals in the temperature range

e standard errors of replicate analyses.

ARTICLE IN PRESSP. Leinweber et al. / Soil Biology & Biochemistry 40 (2008) 1496–1505 1501

before they fell down to the initial level. This agreementresulted in a significant positive correlation betweenxylanase activity and the signal ratio m/z 114 : m/z 132.There were no significant correlations between the activitiesof other enzymes (urease, phosphomonoesterase, invertase)and the mass spectral data.

4. Discussion

The TII thermograms and the Py-FI mass spectra inFig. 1 showed effects of (1) the different soil management

6NPK 110 to 410 °CNPK 420 to 700 °C

5 U 110 to 410 °CU 420 to 700 °C

4

3

1

2

0

N-c

onta

inin

g co

mpo

unds

in

Py-

FIM

S (%

TII)

0 10 20 30 40 50 60 70

Days of incubation

Fig. 3. Ion intensities of marker signals for N-containing compounds in

characteristic temperature intervals from 110 to 410 1C and from 420 to

700 1C of the Py-FI mass spectra during the aerobic incubation

experiment; fertilization treatments U and NPK. Vertical bars indicate

standard errors of replicate analyses.

0.7

0.8

0.6

0.5

0.4

0.3

0.2

0.0

0.1Xyl

anas

e ac

tivity

(�m

ol p

rodu

ct g

soil-1

h-1

)

0 10 20 30

Days of incu

Fig. 4. Xylan (m/z 114) to xylose (m/z 132) ratio calculated from the indicato

aerobic incubation experiment; fertilization treatments U and NPK. Vertical b

without and with NPK-fertilization and (2) the short-termaerobic incubation on the thermal stability and molecular-chemical composition of SOM. The relative enrichment ofsample NPK in phenols/lignin monomers and alkylaro-matics confirmed Schmidt et al. (2000), who also analyzedthis pair of treatments sampled at another date withPy-FIMS. Incubation altered the chemical composition ofSOM, but extent and direction of these alterations differedamong soil treatments. The small alterations in U can beexplained by stable equilibrium between a small input ofprimary organic matter from rye residues and its decom-position at a low level, developed in the soil due to ryemonoculture for more than 120 years. Thus, only a smalladditional decomposition can be expected if the growthconditions for microorganisms are somewhat optimized bysoil moisture at 60% water holding capacity, 25 1Ctemperature and moderate O2-supply (Jason et al., 2000).The decomposition of lignin dimers in U and NPK asindicated by the summed Py-FI mass spectra in Fig. 1corresponded to incubation studies with dissolved organicmatter (Kalbitz et al., 2003). For the NPK-treatment thelosses in lignin dimers derived from the summed spectrumover the whole temperature range and incubation period(Fig. 1) was confirmed by the pronounced decrease inthermally stable lignin dimers in the initial incubation(Fig. 2). The difference in carbohydrate depletion in U andenrichment in NPK is explained by the fact thatcarbohydrates can be decomposed but at the same timesynthesized in the microbial biomass (Derrien et al., 2007).Aerobic decomposition of small amounts of originallyabundant and metabolized carbohydrates, resulting insuccessively decreasing carbohydrate contents, may be the

1.2Xylanase U

Xylanase NPK1.0

Xylan / xylose ratio U

Xylan / xylose ratio NPK

0.6

0.8

0.4

0.2

0.0

Xyl

an /

xylo

se ra

tio o

f ion

inte

nsiti

es in

Py-

FIM

S

40 50 60 70

bation

r signals of Py-FIMS and the xylanase activity in soil samples during the

ars indicate standard errors of replicate analyses.

ARTICLE IN PRESSP. Leinweber et al. / Soil Biology & Biochemistry 40 (2008) 1496–15051502

reason for carbohydrate depletion in U. By contrast, alarger pool of carbohydrates and other decomposablecompounds, consequently resulting in microbial metaboli-zation at higher quantitative level, probably explain therelative carbohydrate enrichment during incubation of theNPK samples. This hypothesis is supported by greatdifferences in the organic matter inputs to soils U andNPK each year. According to Kramer (2004) the plantresidues introduced to soils about 0.048 kg organicCm�2 yr�1 in treatment U and 0.061 kg organic Cm�2 yr�1

in treatment NPK. Furthermore, mass spectrometricanalyses have shown that straw from the unfertilized anda manured plot of this experiment differed in the chemicalcomposition, and it was suggested that these differencesmay affect its microbial decomposability and turnover insoils (Leinweber and Schulten, 1993). In summary, SOM intreatment NPK resulted from the long-term input of moreplant residues of different quality, probably higher nutrientconcentration, and thus, better decomposability, comparedto SOM in treatment U.

The cumulative CO2-C fluxes in the unfertilized soilagreed with comparable data of other field experimentsby Klimanek (1980). However, Klimanek (1980) foundsmaller differences between NPK fertilized and unfertilizedsoils than observed in the present study. On the other handboth, proportions of CO2-C fluxes and factors of increasedue to fertilization, agreed with incubation studies oforganic-mineral particle-size fractions from unfertilizedand manured plots of a loess Chernozem (Leinweber,1995). The CO2-C flux from the unfertilized soil after long-term monoculture was lower than from an arable soilwith less organic C (6.2 g kg�1 compared to 9.3 g kg�1 inTable 1) investigated by Wiedow et al. (2007).

The C/N ratio of the microbial biomass in the incubatedsoils ranged from 4 to 7 in the treatments U and from 4 to11 in the treatment NPK (Table 1). The highest C/N ratiosof the microbial biomass in treatment NPK after two daysmight be caused by promotion of fungal r-selected species.This can be an aeration effect during soil preparation of theincubation experiment in combination with a highernutrient content caused by the NPK-fertilization. Forinstance, Esser (1986) observed high mycelial biomassproduction of Mucoraceae, a r-selected taxon alreadywithin 2 days.

The activities of some enzymes were low compared tosimilar soils under short-term monoculture (Drissner et al.,2007) or crop rotation (Sonnleitner and Schinner, 2003;Meysner et al., 2006). Low soil respiration rates andenzyme activities following long-term monoculture werealso observed by Pascual et al. (2001). In this line, alkalinephosphatase activities in the present study were lower thanin arable Chernozems (Sonnleitner and Schinner, 2003). Ina podzolic soil under wheat monoculture urease activitydecreased with increasing duration of this type of cultiva-tion (Meysner et al., 2006). This supports our observationthat the urease activity was low in the investigated soilform long-term monoculture (Table 1). On the other hand,

the xylanase activity was in the same range in theunfertilized soils and even higher in the fertilized soils thanin the arable Chernozems investigated by Sonnleitner andSchinner (2003). The strongly increased enzyme activities inNPK compared to the U-treatment confirms correspond-ing results of Hoffmann et al. (2002) after NPK-fertiliza-tion of commercially cropped arable soils, of Graham andHaynes (2005) in sugar cane and of Nayak et al. (2007) inrice monoculture.In previous incubation studies organic matter stability

and decomposability often was derived (1) indirectlyby comparisons of widely differing soil samples fromfertilization experiments (e.g. Klimanek, 1990; Klimanekand Korschens, 1982), and (2) correlations between thechemical composition of organic matter added to soil andthe resulting CO2-C release (e.g. Vanlauwe et al., 1996;Jensen et al., 2005). To the best of our knowledge, this isthe first study in which the CO2-C flux from incubated soilwas compared with changes in the chemical compositionof SOM at molecular level. The disagreement betweenrelatively constant contents of stable lignin dimers intreatments U and NPK (period day 2–35) and the gradientsin cumulative CO2-C flux (Fig. 2) indicates that lignindimers, thermally volatilized at 4460 1C, were scarcelydecomposed in this incubation experiment, if the firstunusually high value for stable lignin dimers is ignored.Generally, lignin constituents are relatively stable in thesoil (Rabinovich et al., 2004). By contrast to the presentresults, Wiedow et al. (2007) found a significant decrease inlignin dimers during decomposition of wheat straw addedto soil already after 28 days of incubation. This contra-diction might be explained by the different sources of thelignin dimers either from wheat straw (Wiedow et al., 2007)or from SOM.The changes in N-containing compounds during the

incubation differed fundamentally between the treatments.The relatively constant levels of thermally labile andthermally stabile N-containing compounds in treatmentU (Fig. 3) agreed with the low CO2-C flux, microbialbiomass C and N, and urease activity in this treatment. Inthe NPK-treatment the initial strong decrease of thermallystabile N-containing compounds along with the increase inthermally labile N-containing compounds is explained bythe combination of SOM decomposition with growth ofmicrobial biomass. The latter is proved by the increase inmicrobial biomass N from 17 to 35 mg g�1 (Table 1).Comparison of Figs. 2 and 3 indicates that thermally stableproportions of lignin dimers and N-containing compoundshad not the same meaning for stability against microbialdecomposition. Kuzyakov et al. (2006) concluded fromd13C and d15N isotopic investigations combined withthermogravimetry that the thermal stability of SOM wasnot very strongly related to the biological availability ofSOM. This weak relationship may be explained by our datathat thermally labile and stable proportions of variouscompound classes showed different changes during incuba-tion (Figs. 2 and 3). This also confirms a recent study by

ARTICLE IN PRESSP. Leinweber et al. / Soil Biology & Biochemistry 40 (2008) 1496–1505 1503

Rovira et al. (2008) who showed that litter decompositionwas reflected by general shifts of differential thermalgravimetry (DTG) curves towards lower temperature,and changes in the quantitative relationship betweenthermally labile and stable proportions of the litter.

For the first time we derived positive relationshipsbetween signal ratios of xylan to xylose from Py-FI massspectra and the xylanase activity. Thus obtained verysimilar curves during decomposition (Fig. 4) and positivecorrelation (not shown) indicated decreasing xylanaseactivity with decreasing xylan to xylose signal ratio. Thiscan be interpreted as a progressive decomposition of thehemicellulose xylan to the metabolite xylose. Therefore,this relationship confirms the signal assignment in thePy-FI mass spectra on the one hand and the chemicalbackground of the indirectly determined xylanase activityon the other.

5. Conclusions

(1)

The described experimental approach involving aerobicincubation with samples from long-term differentlyfertilized soils under rye monoculture, molecular-chemical plus thermal (Py-FIMS) and biochemical(enzyme activities) characterization of samples duringdecomposition was well suited to give first hints whichcompound classes and which fractions of thesecompound classes were potentially mineralizable andwhich enzymes were involved in the decomposition.

(2)

Since within one class of chemical compound thermallydifferently stable fractions showed contradictingtrends in quantity and correlation with the CO2 fluxunder incubation, a pure chemical substrate character-ization appears insufficient to explain its microbialdecomposability. The thermal stability of compoundsseems to be a universal predictor for the decomposi-tion, perhaps because energy required for bondcleavage in Py-FIMS reflects the activation energyin the enzymatic driven biochemical decompositionreactions.

(3)

It appeared that the most intensive transformationsalready occurred in the early incubation period (days 0to 2 or 7, respectively). Therefore, forthcoming studieswill involve more intensive sampling in the initialdecomposition periods and analyses at higher temporalresolution. Furthermore, the molecular-chemical back-ground for the wide range of indirectly determinedenzyme activities is a challenge for broad applicationsof field-ionization and other soft-ionization massspectrometry techniques in the future.

Acknowledgments

This work was financially supported by the DFGPriority Program, ‘‘Soils as source and sink for CO2—mechanisms and regulation of organic matter stabilization

in soil’’ (Project LE 903/3). The authors are grateful toR. NuXler and R. Beese (Institute for Land Use, Universityof Rostock) for technical support with the incubationstudies and the Py-FIMS, respectively. We thank the staffof the Faculty of Agriculture (Martin-Luther-University,Halle-Wittenberg, Germany) for the support in soilsampling and providing data from the long-term experi-ments. We also thank two anonymous reviewers for veryhelpful comments and suggestions, which considerablyimproved the manuscript.

References

Christensen, B.T., 1987. Decomposability of organic matter in particle-

size fractions from field soils with straw incorporation. Soil Biology &

Biochemistry 19, 429–435.

Derrien, D., Marol, C., Balesdent, J., 2007. Microbial biosyntheses of

individual neutral sugars among sets of substrates and soils. Geoderma

139, 190–198.

Drissner, D., Blum, H., Tscherko, D., Kandeler, E., 2007. Nine years of

enriched CO2 changes the function and structural diversity of soil

microorganisms in a grassland. European Journal of Soil Science 58,

260–269.

Esser, K., 1986. Kryptogamen. Springer, Berlin, Heidelberg, New York,

Tokyo.

Graham, M.H., Haynes, R.J., 2005. Organic matter accumulation and

fertilizer-induced acidification interact to affect soil microbial and

enzyme activity on a long-term sugarcane management experiment.

Biology and Fertility of Soils 41, 249–256.

Gregorich, G.E., Kachanowski, R.G., Voroney, R.P., 1989. Carbon

mineralization in soil size fractions after various amounts of aggregate

disruption. Journal of Soil Science 40, 649–659.

Hoffmann, S., Csitari, G., Hegedus, L., 2002. The humus content and soil

biological properties as a function of organic and mineral fertilization.

Archives of Agronomy and Soil Science 48, 141–146.

Jason, S.L., Daniels, B.L., Eberiel, D.T., Farrell, R.E., 2000. Polymer

mineralization in soils: effects of cold storage on microbial populations

and biodegradation potential. Journal of Polymers and the Environ-

ment 8, 81–89.

Jensen, L.S., Tapio, S., Palmason, F., Tor, A.B., Trond, M.H., Stenberg,

B., Pedersen, A., Lundstrom, C., Esala, M., 2005. Influence of

biochemical quality on C and N mineralisation from a broad variety of

plant materials in soil. Plant and Soil 273, 307–326.

Kalbitz, K., Schwesig, D., Schmerwitz, J., Kaiser, K., Haumeier, L.,

Glaser, B., Ellerbrock, R., Leinweber, P., 2003. Changes in properties

of soil-derived dissolved organic matter induced by biodegradation.

Soil Biology & Biochemistry 35, 1129–1142.

Kandeler, E., Gerber, H., 1988. Short-term assay of soil urease activity

using determination of ammonium. Biology and Fertility of Soils 6,

68–72.

Kandeler, E., Stemmer, M., Klimanek, E.M., 1999a. Response of soil

microbial biomass, urease and xylanase within particle size fractions to

long-term soil management. Soil Biology & Biochemistry 31, 261–273.

Kandeler, E., Tscherko, D., Spiegel, H., 1999b. Long-term monitoring of

microbial biomass, N-mineralisation and enzyme activities of a

Chernozem under different tillage management. Biology and Fertility

of Soils 28, 343–351.

Kandeler, E., Palli, S., Stemmer, M., Gerzabek, M.H., 1999c. Tillage

changes microbial biomass and enzyme activities in particle-size

fractions of a Haplic Chernozem. Soil Biology & Biochemistry 31,

1253–1264.

Klimanek, E.-M., 1980. Mineralisierungsleistung unterschiedlicher Boden

in Abhangigkeit von der Dungung. Archiv fur Acker—und Pflanzen-

bau und Bodenkunde 24, 225–232.

ARTICLE IN PRESSP. Leinweber et al. / Soil Biology & Biochemistry 40 (2008) 1496–15051504

Klimanek, E.-M., 1990. Umsetzungsverhalten von Ernte—und Wurzel-

ruckstanden im Boden. Archiv fur Acker—und Pflanzenbau und

Bodenkunde 34, 559–567.

Klimanek, E.-M., Korschens, M., 1982. Die Mineralisierungsleistung

unterschiedlicher Boden und ihre Beziehung zum Gehalt an umsetz-

barem Humus. Archiv fur Acker—und Pflanzenbau und Bodenkunde

26, 289–294.

Kogel-Knabner, I., Guggenberger, G., Kleber, M., Kandeler, E., Kalbitz,

K., Scheu, S., Eusterhues, K., Leinweber, P., 2008. Organo-mineral

associations in temperate soils: integrating biology, mineralogy and

organic matter chemistry. Journal of Plant Nutrition and Soil Science

171, 61–82.

Kramer, C., 2004. Umsatz und Stabilisierung von organischem Kohlenst-

off in Boden. Thesis, University of Jena.

Kuzyakov, Y., Mitusov, A., Schneckenberger, K., 2006. Effect of C3–C4

vegetation change on d13C and d15N values of soil organic

matter fractions separated by thermal stability. Plant and Soil 283,

229–238.

Lal, R., 2004. Soil carbon sequestration impacts on global climate change

and food security. Science 304, 1623–1627.

Leinweber, P., 1995. Organische Substanzen in PartikelgroXenfraktionen:

Zusammensetzung, Dynamik und EinfluX auf Bodeneigenschaften.

In: Vechtaer Studien zur Angewandten Geographie und Regionalwis-

senschaft, vol. 15. Vechtaer Druckerei und Verlag, Vechta, p. 148.

Leinweber, P., Reuter, G., 1992. The influence of different organic

fertilization practices on concentrations of organic carbon and

total nitrogen in particle-size fractions during 34 years of a soil

formation experiment in loamy marl. Biology and Fertility of Soils 13,

119–124.

Leinweber, P., Schulten, H.-R., 1993. Dynamics of soil organic matter

studied by pyrolysis-field ionization mass spectrometry. Journal of

Analytical and Applied Pyrolysis 25, 123–136.

Leinweber, P., Blumenstein, O., Schulten, H.-R., 1996. Organic matter

composition in sewage farm soils: investigations by 13C�NMR and

pyrolysis-field ionization mass spectrometry. European Journal of Soil

Science 47, 71–80.

Lutzow, M.v., Kogel-Knabner, I., Ludwig, B., Matzner, E., Flessa, H.,

Ekschmitt, K., Guggenberger, G., Marschner, B., Kalbitz, K., 2008.

Stabilization mechanisms of organic matter in four temperate soils:

development and application of a conceptional model. Journal of Plant

Nutrition and Soil Science 171, 111–124.

Luxhøi, J., Magid, J., Tscherko, D., Kandeler, E., 2002. Dynamics

of invertase, xylanase and coupled quality indices of decomposing

green and brown plant residues. Soil Biology & Biochemistry 34,

501–508.

Magdoff, F., Weil, R.R., 2004. Strategies for managing organic matter. In:

Magdoff, F., Weil, R.R. (Eds.), Soil Organic Matter in Sustainable

Agriculture. CRC Press, Boca Raton, FL, pp. 44–65.

Magid, J., Jensen, L.S., Mueller, T., Nielsen, N.E., 1997. Size-density

fractionation for in situ measurements of rape straw decomposition—

an alternative to litterbag approach? Soil Biology & Biochemistry 29,

1125–1133.

Marschner, B., Brodowski, S., Dreves, A., Gleixner, G., Grootes, P.M.,

Hamer, U., Heim, A., Jandl, G., Ji, R., Kaiser, K., Kalbitz, K.,

Kramer, C., Leinweber, P., Rethemeyer, J., Schmidt, M.W.I.,

Schwark, L., Wiesenberg, G.L.B., 2008. How relevant is recalcitrance

for the stabilization of organic matter in soils? Journal of Plant

Nutrition and Soil Science 171, 91–110.

Merbach, W., Garz, J., Schliephake, W., Stumpe, H., Schmidt, L., 2000.

The long-term fertilization trials in Halle—introduction and overview.

Journal of Plant Nutrition and Soil Science 163, 627–636.

Meysner, T., Szajdak, L., Kus, J., 2006. Impact of farming systems on the

content of biologically active substances and the forms of nitrogen in

soils. Agronomy Research 4, 531–542.

Nayak, D.R., Babu, J.Y., Adhya, T.K., 2007. Long-term application of

compost influences microbial biomass and enzyme activities in a

tropical Aeric Endoaquept planted with rice under flooded condition.

Soil Biology & Biochemistry 39, 1897–1906.

Pascual, J.A., Ros, M., Hernandez, T., Garcia, C., 2001. Effect of long-

term monoculture on microbiological and biochemical properties in

semiarid soils. Communications in Soil Science and Plant Analysis 32,

537–552.

Paul, E.A., 2007. Soil Microbiology, Biochemistry and Soil Ecology, third

ed. Elsevier, San Diego.

Poll, C., Ingwersen, J., Stemmer, M., Gerzabek, M.H., Kandeler, E., 2006.

Mechanisms of solute transport influence small-scale abundance and

function of soil microorganisms at the soil–litter interface. European

Journal of Soil Science 57, 583–595.

Rabinovich, M.L., Bolobova, A.V., Vasilchenko, L.G., 2004. Fungal

decomposition of natural aromatic structures and xenobiotics: a

review. Applied Biochemical Microbiology 40, 1–17.

Rovira, P., Kurz-Besson, C., Couteaux, M.-M., Vallejo, V.R., 2008.

Changes in litter properties during decomposition: a study by

differential thermogravimetry and scanning calorimetry. Soil Biology

& Biochemistry 40, 172–185.

Rumpel, C., 1999. Differenzierung und Charakterisierung pedogener und

geogener organischer Substanz in forstlich kultivierten Kippboden.

Cottbuser Schriften zu Bodenschutz und Rekultivierung, Band 5,

Brandenburgische Technische Universitat Cottbus.

Ryan, M.G., Law, B.E., 2005. Interpreting, measuring, and modeling soil

respiration. Biogeochemistry 73, 3–27.

Schinner, F., von Mersi, W., 1990. Xylanase-, CM-cellulase and invertase

activity in soil: an improved method. Soil Biology & Biochemistry 22,

511–515.

Schlesinger, W.H., Andrews, J.A., 2000. Soil respiration and the global

carbon cycle. Biogeochemistry 48, 7–20.

Schmidt, L., Warnstorff, K., Dorfel, H., Leinweber, P., Lange, H.,

Merbach, W., 2000. The influence of fertilization and rotation on soil

organic matter and plant yields in the long-term Eternal Rye trial in

Halle (Saale), Germany. Journal of Plant Nutrition and Soil Science

163, 639–648.

Schmidt, M.W.I., Knicker, H., Hatcher, P.G., Kogel-Knabner, I.,

1996. Impact of brown coal dust on a soil and its size fractions—

chemical and spectroscopic studies. Organic Geochemistry 25,

29–39.

Schulten, H.-R., 1996. Direct pyrolysis-mass spectrometry of soils: a novel

tool in agriculture, ecology, forestry, and soil science. In: Yamasaki, S.,

Boutton, T.W. (Eds.), Mass Spectrometry of Soils. Marcel Dekker,

New York, pp. 373–436.

Schulten, H.-R., Leinweber, P., 1991. Influence of long-term fertilization

with farmyard manure on soil organic matter: characteristics of

particle-size fractions. Biology and Fertility of Soils 12, 81–88.

Schulten, H.-R., Leinweber, P., 1999. Thermal stability and composition

of mineral-bound organic matter in density fractions of soil. European

Journal of Soil Science 50, 237–248.

Sinsabaugh, R.L., Moorhead, D.L., Linkins, A.E., 1994. The enzymatic

basis of plant litter decomposition: emergence of an ecological process.

Applied Soil Ecology 1, 97–111.

Smith, J.L., Paul, E.A., 1990. The significance of soil biomass estimations.

In: Bollag, J.M., Stolzky, G. (Eds.), Soil Biochemistry. Marcel Dekker,

New York, pp. 357–396.

Sonnleitner, R., Schinner, F., 2003. Microbiological and enzymatic

properties of soil deposits originating from wind erosion. Journal of

Plant Nutrition and Soil Science 166, 484–489.

Sorge, C., Muller, R., Leinweber, P., Schulten, H.-R., 1993. Pyrolysis-field

ionization mass spectrometry of whole soils, soil particle-size fractions

and humic substances: statistical evaluation of sample weight, residue,

volatilized matter and total ion intensity. Fresenius Journal of

Analytical Chemistry 346, 697–703.

Tabatabai, A., 1994. Soil enzymes. In: Weaver, R.W., Angle, R.W., Angle,

J.S., Bottomley, P.S. (Eds.), Methods of Soil Analysis, Part 2.

Microbiological and Biochemical Properties. SSSA, Madison, WI,

pp. 775–833.

Vance, E.D., Brookes, P.C., Jenkinson, D.S., 1987. An extraction method

for measuring soil microbial biomass C. Soil Biology & Biochemistry

19, 703–707.

ARTICLE IN PRESSP. Leinweber et al. / Soil Biology & Biochemistry 40 (2008) 1496–1505 1505

Vanlauwe, B., Nwoke, O.C., Sanginga, N., Merckx, R., 1996. Impact of

residue quality on the C and N mineralization of leaf and root residues

of three agroforestry species. Plant Soil 183, 221–231.

Vepsalainen, M., Erkomaa, K., Kukkonen, S., Vestberg, M., Wallenius, K.,

Niemi, R.M., 2004. The impact of crop plant cultivation and peat

amendment on soil microbial activity and structure. Plant Soil 264, 273–286.

Voroney, R.P., Paul, E.A., 1984. Determination of kc and kn in situ for

calibration of the chloroform fumigation–incubation method. Soil

Biology amp;amp; Biochemistry 16, 9–14.

Wiedow, D., Baum, C., Leinweber, P., 2007. Inoculation with Tricho-

derma saturnisporum accelerates wheat straw decomposition on soil.

Archives of Agronomy and Soil Science 53, 1–12.