| HOME | HELP | FEEDBACK | SUBSCRIPTIONS | ARCHIVE | SEARCH | TABLE OF CONTENTS |
|
|
|
|||||||
|
|||||||||
a Laboratoire d'Ecologie, UMR 7625, Ecole Normale Supérieure, 46 rue d'Ulm, F-75 230 Paris cedex 05, France
b Laboratoire de Biogéochimie Isotopique, Université Pierre et Marie Curie, Case 120, 4 Place Jussieu, F-75 252 Paris cedex 05, France
* Corresponding author (fontaine{at}biologie.ens.fr).
 |
ABSTRACT |
|---|
 TOP ABSTRACT INTRODUCTIONMATERIALS AND METHODSRESULTSDISCUSSIONREFERENCES |
|---|
Abbreviations: FOM, fresh organic matter • PE, priming effect • SOM, soil organic matter
|
INTRODUCTION |
|---|
|
TOPABSTRACTINTRODUCTIONMATERIALS AND METHODSRESULTSDISCUSSIONREFERENCES |
|---|
The possible extra decomposition of native SOM decomposition by fresh organic matter (FOM) input to soil has been subject to controversy since Löhnis (1926). Although Löhnis suggested that the PE played a central role in the long-term SOM dynamics, it was not until the 1940s that the PE could be quantified thanks to isotopic facilities (Broadbent, 1947; Broadbent and Bartholomew, 1948; Bingeman et al., 1953; Broadbent and Nakashima, 1974). The labeling of either FOM or SOM allowed for mineralization of these two C sources to be separated. The PE was quantified by comparing the native soil C released as CO2 in FOM amended soils to the C in control soils. During the 1980s, however, the validity of these studies was questioned. Dalenberg and Jager (1980)(1989) demonstrated that the input of labeled glucose to soil induced a large and immediate release of unlabeled CO2 originating from an increased microbial turnover and a substitution in microbial C. Priming effect was therefore an artifact since the increase in unlabeled CO2 resulted from a decrease in unlabeled microbial C that was replaced by labeled C, and not from a decrease in native soil C pool. Since this work, it is generally believed that the PE in soils is only apparent and that the rate of SOM mineralization is not controlled by FOM input. This idea was further called into question in the 1990s due to the fumigation technique that allows researchers to measure the microbial biomass, that is, to control the dynamics of C within the microbial biomass (Voroney and Paul, 1984; Sparling and West, 1988; Bremer and van Kessel, 1990). By comparing the native microbial C in ryegrass (Lolium L.) amended soil to that in control soil, Wu et al. (1993) suggested that the PE induced by ryegrass input was real despite several technical problems with the monitoring of microbial biomass and the possible heterogeneous labeling of C ryegrass. To date, there is no experiment in controlled conditions showing an indisputable PE and quantifying the impact of extra soil C output on the final balance of C.
The mechanisms leading to the PE remain elusive. Priming effect is often attributed to an increased microbial activity or enzyme production of the whole microbial community following the addition of energy to soil (Broadbent and Norman, 1947; Norman, 1947; Kuzyakov et al., 2000). Priming effect is also attributed to an increased microbial biomass induced by the input of FOM to soil, the newly formed microbes being able to decompose the SOM especially when the FOM is exhausted (Tate, 1987; Pascual et al., 1998). Wu et al. (1993), however, observed that a ryegrass input to soil induced PE whereas glucose had no effect. This result was surprising since glucose highly stimulated microbial growth and activity thus suggesting that neither increased microbial activity nor increased microbial biomass were involved in the PE.
In a recent review on the PE (Fontaine et al., 2003) we proposed two potential mechanisms leading to the PE. The starting point is the dramatic change in the structure of the microbial community after the supply of FOM (Griffiths et al., 1998), that is, the enhancement of the activity and growth of previously starving microbial populations, that become available to specifically use this new substrate, (Winogradzky, 1924; Lemoigne et al., 1951; Holding, 1960; Behera and Wagner, 1974; Kshattriya et al., 1991). There is thus a strong relationship between the type of FOM, the type of active microbes and the type of enzymes involved in FOM degradation (Garett, 1951; Swift et al., 1979). This specificity of the organic matter decomposing activities could be the reason why the rate of SOM decomposition can remain unchanged after the input of FOM. The PE, however, could occur in two ways. First, extracellular enzymes that are produced to decompose FOM by FOM specialized microbes may be partly efficient in degrading SOM (Mechanism 1). This mechanism obviously depends on biochemical similarities between FOM and SOM. The higher the chemical diversity of FOM, the higher the diversity of the produced enzymes will be and the probability of occurrence of the PE. Wu et al. (1993) suggested this mechanism to explain why a ryegrass input led to a PE when glucose had no effect. Second, depending on the competition with FOM specialized microbes, part of the FOM may be absorbed by SOM decomposing microbes (Mechanism 2). This absorption increases the populations of SOM decomposing microbe and hence the decomposition rate of SOM. In this second case, the intensity of the PE will depend on the intensity of the competition for energy acquisition between FOM specialized microbes and SOM decomposing microbes.
To test whether the input of FOM to soil does or does not accelerate SOM decomposition, we incubated soil with cellulose (the major chemical form of C entering soils). We used 13C-labeled cellulose to separate soil C and cellulose C decompositions. Usually, the PE is quantified by comparing the native soil C released as CO2 in FOM amended soil to that in control soil. However, C substitution may occur in the microbial biomass, which results in an artifact PE as shown by Dalenberg and Jager (1980)(1989). To control this artifact, we monitored microbial biomass and used it in the calculation of the intensity of PE. We also tested the mechanisms really involved in PE by two approaches. First, we performed a cellulase amendment to estimate the potential contribution of Mechanism 1 to the PE. Second, we compared the dynamic of microbial biomass with the decomposition rates of soil C and cellulose to validate the competition hypothesis (Mechanism 2). At the end of the incubation, we determined the effect of cellulose addition on soil C content by calculating the balance between the amount of 13C (C originated from cellulose) remaining in soil as organic form to the amount of 12C (C originated from soil C) lost through primed respiration.
|
MATERIALS AND METHODS |
|---|
|
TOPABSTRACTINTRODUCTIONMATERIALS AND METHODSRESULTSDISCUSSIONREFERENCES |
|---|
13C of 2000
.
Incubation Conditions
Surface mineral soil (0–10 cm) was collected from a site of grass savannah located in the Lamto Ecological Station in Côte d'Ivoire (6°13' N, 5°20' W). The sample was air-dried in the shade, homogenized, sieved and stored before the experiment. The soil was a sandy silt soil (USDA: Ultisol) with the following characteristics: 6 pH KCl; 10.5 g kg–1 total C and 0.63 g kg–1 total N; 16.7 C/N ratio. Most of the plant residues were removed by flotation in a second step before soil was dried a final time.
Two sets of incubations were performed: soil by itself (control) and soil with added cellulose. Sufficient soil samples were prepared for allowing five destructive samples in triplicate per treatment. The experimental units consisted of 20-g samples of dried soil placed into a 120-mL flask that was sealed with a septum. Distilled water was added to soils to perform water potential of 31.6 kPa. Moistened soils were incubated at 28°C for 3 d. Following the preincubation 13C-labeled cellulose was uniformly added to the soil at rate of 495 mg C kg–1.
At each sampling date, gas samples were taken in the headspace of a subset of flasks and were injected into a chromatograph coupled to a mass spectrometer for CO2 and 13C measurements. After measuring CO2, all the flasks were flushed with reconstituted air (19% O2, 81% N2). The air was led through a bottle of distilled water to moisten it and prevent the drying of soils. The CO2 concentration in the flask headspace never exceeded 3% by volume. Mineral N, soil microbial biomass, and soil C contents were determined by destructive sampling at Days 3, 13, 21, 44, and 70.
The microbial biomass was determined by the fumigation-extraction technique (Sparling and West, 1988). For each harvested flask, 5 g of soil was extracted with 20 mL of 30 mM K2SO4 and shacked for 1 h. A subsequent 5-g sample was fumigated for 2 d with ethanol-free chloroform in a glass desiccator. Chloroform was removed from the soil by ventilation, and the soils were immediately extracted with 20 mL of 30 mM K2SO4. The K2SO4 extracts were filtered (0.45 µm) and then lyophilized. The recovered crystals were stored until analysis of C content and 13C. The microbial biomass was calculated from B = 1/k x E where: E = soluble microbial C and k = extraction yield. E is calculated using E = (organic C extracted by 30 mM K2SO4 from fumigated soil) minus (organic C extracted by 30 mM K2SO4 from nonfumigated soil).
Extraction yield of microbial biomass (k = 16%) was estimated in a separate experiment by the in situ labeling procedure (Voroney and Paul, 1984; Bremer and van Kessel, 1990). Aliquots of soil used in this experiment were amended with 13C-labeled glucose, KNO–3 and H3PO2–4 at 500, 50, and 3 mg of C, N and P per kilogram of soil, respectively. After 3 d of incubations, soluble microbial C (E) was determined as described above. The extraction yield was defined as the proportion of 13C in the microbial soluble C (E) relative to the total amount of 13C remaining in the soil at Day 3.
The C content and 13C of soil and K2SO4 crystals were measured by an elemental analyzer coupled to a mass spectrometer (double inlet mass spectrometer).
Enzyme Amendment
A solution of lyophilized cellulase powder (from Trichoderma ressei) mixed with cooled water (5°C) was used as amending cellulase (50 units mL–1). The pH of the incubation medium was adjusted to 5 by using a 0.1 M acetate buffer. Toluene was used to inhibit the microbial uptake of soluble sugars (Skuji
, 1976). For the incubation medium, 2 g of dried soil, 1.2 mL of toluene, 1.2 mL of enzyme solution, and 10.8 mL of buffer were reacted at 28°C for 3 d in a slowly shaken test tube (Tateno, 1987). To check the potential activities of the amended cellulase, a subset of soil suspension was added to 1.5% carboxymethyl cellulose (CMC) solution. To measure the effect of cellulase on soil C decomposition, the soil solution was centrifuged (560 x g [2500 rpm], 10 min) and the glucose accumulated in the supernatant was determined by a glucose oxidase method, using the Glucose B-test (Wako). We used two controls: a control soil (soil + acetate buffer + toluene – cellulase) and a control enzyme (cellulase + acetate buffer + toluene – soil) because we observed that commercialized cellulase powder may contain glucose. The extra release of glucose due to the cellulase amendment was calculated by subtracting the glucose of controls to that of cellulase amended soils. This cellulase effect could not directly compared with the PE of cellulose because the two experiments were conducted under different conditions. However, because the solution state incubation will presumably maximize the opportunity of contact between enzyme and substrate, the extra release of glucose due to the cellulase amendment represented the potential contribution of the cellulase to the PE of cellulose (Mechanism 1).
Calculations and Statistics
Calculation of Priming Effect Intensity
Carbon-13-labeled cellulose was added to the soil so that the C metabolized as CO2 and microbial biomass can be divided into two fractions, that originating from cellulose decomposition (13C) and that originating from the decomposition of native soil C (12C). Thus, a comparison of the amount of 12C metabolized as CO2 and microbial biomass in cellulose amended soil and in control soil allowed estimation of the magnitude of the PE. The calculation of the PE was made in three steps:
If it was observed that MC 
0, then the excess of mineralized soil C was real, that is, it resulted from an acceleration of SOM decomposition by cellulose input. If MC < 0, then the excess of mineralized soil C was at least partly apparent, that is, it resulted partly from microbial C substitution.
Effect on Total Soil Organic Matter
The analysis of the total C and 13C data allowed us to calculate the amount of 13C that remained in the soil as an organic form (Microbial biomass 13C + Nonliving 13C). Thus, the effect of cellulose addition on total soil C was calculated as the balance between the amount of 13C remaining in the soil as an organic form (Total 13C) and the amount of 12C lost due to the excess of mineralized soil C:
 |
CT is the change in total soil C due to cellulose addition. If CT is positive, then the cellulose input to the soil increased the total soil C. If CT is negative, then the cellulose input to soil decreased the total soil C.
Effect on Nonliving Soil Carbon (= Soil Humus)
Because microbial biomass C (Microbial C) is labile, we also examined the effect of cellulose addition on humus C (Humus C = Total soil C – Microbial C). On one hand, the supply of 13C cellulose may increase humus C through microbial turnover. This positive effect on humus C may be quantified as:
 |
(this calculation was made when the cellulose was completely decomposed)
Alternatively, the supply of cellulose may decrease humus C by accelerating the decomposition of humus 12C due to the PE. Finally, the effect of cellulose on humus C (12C + 13C) was calculated as the balance between the two effects of cellulose addition:
 |
Statistics
Analyses of variance of microbial biomass were performed by using the ANOVA Procedure of SAS (SAS Institute, Inc., 1999). Analysis of variance of CO2 was performed with the ANOVA Repeated Measure Procedure because CO2 measurements were conducted on the same flasks throughout the incubation. The effect of cellulose addition on total soil C was analyzed by comparing the Total 13C to the Excess with the ANOVA Procedure. If Total 13C was significantly different from Excess, then cellulose addition induced significant change in total soil C.
|
RESULTS |
|---|
|
TOPABSTRACTINTRODUCTIONMATERIALS AND METHODSRESULTSDISCUSSIONREFERENCES |
|---|
|
|
|
The higher rate of SOM decomposition started between Day 3 and 13 during the phase of general stimulation of the activity and growth of microbes induced by the decomposition of cellulose (Fig. 4) . However, the parallel between the stimulation of microbes and the higher SOM decomposition stopped here. Indeed, the increased activity and growth of microbes ceased by Day 21 likely due to cellulose exhaustion. On the contrary, the SOM decomposition rate remained high throughout the incubation. It is also remarkable that the SOM decomposition rate remained roughly constant despite the 77% decrease of labeled biomass. Finally, the higher rate of SOM decomposition resulted in 140 mg C kg–1 of excess mineralized soil C, representing four times the amount of glucose C released due to enzyme amendment (Mechanism 1).
|
|
The analysis of the C content and 13C of soil indicated that 110 mg 13C kg–1 remained in soil as organic form at the end of the incubation, that is, only 22% of the initial amount of cellulose C. The excess of mineralized soil 12C (140 mg C kg–1) was significantly higher than the amount of 13C remaining in soil as an organic form. Therefore, the input of cellulose input to soil decreased the total soil C content by 30 mg C kg–1 (P < 0.001).
The microbial 13C amounted for 50 mg C kg–1 of the 110 mg 13C kg–1 remaining in soil as organic form. This indicated that only 60 mg C kg–1 of the applied cellulose actually entered in soil humus. Given the 234 mg 12C kg–1 leaving soil humus due to the PE, the soil humus stock was depleted by 174 mg C kg–1 after cellulose addition.
|
DISCUSSION |
|---|
|
TOPABSTRACTINTRODUCTIONMATERIALS AND METHODSRESULTSDISCUSSIONREFERENCES |
|---|
We have proposed that the release of extracellular enzymes to decompose FOM by FOM specialized microbes is a possible origin of the PE (Mechanism 1). In our experiment, we estimated the potential contribution of this mechanism by applying cellulase to soil. The effect on SOM mineralization, however, was very low as observed by Tateno (1987). This indicates that the potential contribution of Mechanism 1 to PE is very low and that SOM mineralization required the production of specific enzymes. Therefore, the PE of cellulose resulted from the stimulation of microbes that were able to provide SOM specific enzymes.
If the PE resulted only from the transient stimulation of microbes that were already present, then one might expect a decrease in SOM mineralization after cellulose exhaustion. In our experiment, the PE was maintained until the end of the incubation though most of the supplied cellulose was exhausted by Day 13. This strongly suggests that the cellulose supply increased the populations of SOM feeding microbes, which were able to continue to survive on SOM after the cellulose was exhausted.
The large increase in labeled biomass within the two first weeks was paralleled by an increase in SOM mineralization rate. However, the decrease in labeled biomass from Day 13 to 70 was not followed by a decrease in SOM decomposition rate. This indicated that most newly formed biomass did not contribute to the PE and only decomposed the cellulose. Thus, the cellulose induced the growth of cellulose specialized microbes that then died or became dormant as the cellulose was exhausted because they were unable to use SOM.
Our results provide evidence that the PE resulted from the stimulation of SOM decomposing microbes by the supply of FOM. They also show that SOM decomposing microbes competed with FOM specialized microbes. These indicate that the PE depends on microbial competition: depending on the competition with FOM specialized microbes, part of the FOM may be absorbed by SOM decomposing microbes (Mechanism 2). This interpretation could explain why some C compounds like glucose have no effect on SOM decomposition. Indeed, although these compounds might be metabolized by SOM feeding microbes, they may also escape to microbes that decompose SOM depending on the dominance of specialized microbes. Because they have slow growth rate SOM feeding microbes only dominate in the later stages of plant litter decomposition when easily assimilable compounds have already been exhausted (Zvyagintsev, 1994; Paul and Clark, 1989). These populations are therefore unable to benefit from the supply of easily assimilable compounds, which could explain why the rate of SOM decomposition remains constant after glucose input. Further investigations should focus on the relationship between the intensity of microbial competition and the magnitude of PE.
Similar to previous research, the increase in CO2 release due to PE was low compared with the amount of CO2 released by cellulose decomposition (e.g., Dalenberg and Jager, 1989). This is why the impact of PE on C and nutrients cycles in soils is often neglected (Dalenberg and Jager, 1989; Wu et al., 1993). Yet, in our experiment, the primed respiration persisted many weeks after the complete decomposition of cellulose, which makes the PE an important determinant of the final C balance. This result suggests that models and short-term incubations tend to overestimate the ability of FOM input to increase the SOM pool, and may explain some of the discrepancies between the data from theoretical and lab-models and those from field experiments (Melillo et al., 1982; Parton et al., 1996; Gill et al., 2002). Clearly, the interactions between fresh litter and ancient SOM controlled by microbial processes may decrease the long-term ability of ecosystems to sequester C.
|
ACKNOWLEDGMENTS |
|---|
Received for publication November 27, 2002.
|
REFERENCES |
|---|
|
TOPABSTRACTINTRODUCTIONMATERIALS AND METHODSRESULTSDISCUSSIONREFERENCES |
|---|
, J. 1976. Extracellular enzymes in soil. CRC Crit. Rev. Microbiol. 4:383–421.[Medline]This article has been cited by other articles:
|
|
 |
|
  A. E. Jimenez Esquilin, M. E. Stromberger, and W. D. Shepperd Soil Scarification and Wildfire Interactions and Effects on Microbial Communities and Carbon Soil Sci. Soc. Am. J., January 11, 2008; 72(1): 111 - 118. [Abstract] [Full Text] [PDF]
|
|
|
|
|
|
H. Arai, N. Tokuchi, and K. Koba Possible Mechanisms Leading to a Delay in Carbon Stock Recovery after Land Use Change Soil Sci. Soc. Am. J., August 27, 2007; 71(5): 1636 - 1638. [Abstract] [Full Text] [PDF]
|
|
| ||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||
| HOME | HELP | FEEDBACK | SUBSCRIPTIONS | ARCHIVE | SEARCH | TABLE OF CONTENTS |
| The SCI Journals | Agronomy Journal | Crop Science | |||
| Vadose Zone Journal | Journal of Plant Registrations | ||||
| Journal of Natural Resources and Life Sciences Education |
Journal of Environmental Quality |
||||
