FEMS Microbiology Ecology 31 (2000) 87^94

www.fems-microbiology.org

Pattern of non-methanogenic and methanogenic degradation of cellulose in anoxic rice ¢eld soil Amnat Chidthaisong, Ralf Conrad * Max-Planck-Institut fu«r terrestrische Mikrobiologie, Karl-von-Frisch-Strasse, D-35043 Marburg, Germany Received 30 March 1999 ; received in revised form 6 October 1999; accepted 9 October 1999

Abstract Rice field soils turn anoxic upon flooding. The complete mineralization of organic matter, e.g. cellulose, to gaseous products is then accomplished by the sequential reduction of nitrate, ferric iron, sulfate and finally by methanogenesis. Therefore, the anaerobic turnover of [U-14 C]cellulose was investigated in fresh, non-methanogenic and in preincubated, methanogenic slurries of Italian rice field soil. In anoxic soil slurries freshly prepared from air-dried soil [U-14 C]cellulose was converted to 14 CO2 and 14 CH4 in a ratio of 3:1. In methanogenic soil slurries, on the other hand, which had been preincubated for 45 days under anaerobic conditions, [U-14 C]cellulose was converted to 14 CO2 and 14 CH4 in the ratio of 1:1. The turnover times (7^14 days) of cellulose degradation were not significantly different (P s 0.05) in fresh and methanogenic soil. Chloroform addition abolished CH4 production, but only slightly (30%) inhibited cellulose degradation in both fresh and methanogenic soil. Under both soil conditions, [14 C]acetate was the only labeled intermediate detected. A maximum of 24% of the applied radioactivity was transiently accumulated as [14 C]acetate in both fresh and methanogenic soil slurries. However, when methanogenesis was inhibited by chloroform, 46% and 66% of the applied radioactivity were recovered as [14 C]acetate in fresh and methanogenic soil, respectively. Only non-radioactive propionate accumulated during the incubation with [U-14 C]cellulose, especially in the presence of chloroform, indicating that propionate was produced from substrates other than cellulose. ß 2000 Federation of European Microbiological Societies. Published by Elsevier Science B.V. All rights reserved. Keywords : Methane; Acetate; Propionate ; Turnover ; Paddy soil; Chloroform ; Radioactive cellulose

1. Introduction Cellulose is a major constituent of plant material and its degradation in natural ecosystems is exclusively biological. In oxic environments cellulose is oxidized to CO2 . In anoxic environments, on the other hand, the end product of cellulose degradation is either only CO2 or CH4 +CO2 , depending on the active microbial populations. When electron acceptors other than CO2 are absent, the end products of cellulose degradation are CH4 and CO2 in a ratio of 1:1 according to C6 H12 O6 C3CH4 +3CO2 . Accordingly, Vogels [1] estimated that about 5^10% of the worldwide production of cellulose is eventually converted into CH4 thus contributing to the global methane cycle. In soil, about 5^20% of soil organic matter consists of carbohy-

* Corresponding author. Tel. : +49 (6421) 178 801 ; Fax: +49 (6421) 178 809; E-mail : [email protected]

drates, 8^14% of which is cellulose [2]. This content is low compared to the cellulose content (70%) of plant biomass, indicating that cellulose is rapidly degraded in soil [2]. The main source of cellulose in rice ¢eld soil is rice straw which consists of 20^50% cellulose on a dry weight basis [3,4]. After rice harvest straw has traditionally been incorporated into the soil, to enhance soil fertility and rice productivity [5,6]. Besides enhancing crop productivity, however, straw incorporation also causes a large increase in microbial biomass and in production and emission of CH4 [7^11]. Since cellulose is a major constituent of rice straw, its decomposition probably serves as one of the most important carbon and energy sources for the production of CH4 which is known as one of the important greenhouse gases [12,13]. Knowing the ecology of cellulose degradation in rice ¢eld soil, therefore, would lead to an understanding of the e¡ect of rice straw decomposition on CH4 production. Complete decomposition of macropolymers such as cellulose to gaseous products in anoxic environments requires the co-operation of several microbial populations [14].

0168-6496 / 00 / $20.00 ß 2000 Federation of European Microbiological Societies. Published by Elsevier Science B.V. All rights reserved. PII: S 0 1 6 8 - 6 4 9 6 ( 9 9 ) 0 0 0 8 4 - 7

FEMSEC 1090 27-12-99

88

A. Chidthaisong, R. Conrad / FEMS Microbiology Ecology 31 (2000) 87^94

Generally, hydrolytic microorganisms are responsible for conversion of macropolymers to mono- and oligomers in the ¢rst step. The products of this hydrolysis, e.g. cellobiose and glucose, are then fermented by cellulolytic and associated saccharolytic organisms, yielding various fatty acids, alcohols, H2 and CO2 . The next step involving the conversion of fatty acids larger than acetate to H2 , CO2 and acetate is carried out by the obligate proton-reducing bacteria [15,16]. This step is thermodynamically unfavorable unless the partial pressure of H2 is maintained at very low levels. Under natural conditions, the low partial pressure of H2 can be achieved by the presence of H2 -utilizing microorganisms such as sulfate-reducing or methane-producing microorganisms. Thus, the ¢nal step in organic carbon mineralization involves the conversion of acetate, H2 and CO2 either to CH4 in methanogenic environments or to only CO2 in the environments where electron acceptors such as sulfate or ferric iron and the corresponding bacterial populations are present. Although a general scheme of organic carbon mineralization in anaerobic environments can be outlined as described above, detailed information on each step, especially the break-down of macropolymers such as cellulose are still lacking. Most of the studies of anaerobic cellulose degradation have concentrated on the functional behavior of microbial communities in rumen, digestors and de¢ned mixed cultures [17^21]. Anoxic soils, on the other hand, have received only little attention [22,23]. Anoxic soils are basically di¡erent from environments such as rumen as methanogenesis is not the exclusive ¢nal step in cellulose degradation. In anoxic soil, reduction of nitrate, ferric iron and sulfate plays an additional role, especially during the early phase of £ooding of the rice ¢elds when nitrate, ferric iron and sulfate have still not completely been reduced [24^26]. Therefore, we investigated the degradation of [14 C]cellulose in rice ¢eld soil slurries immediately after onset of anoxic conditions before the start of methanogenesis, and later on when methanogenesis was the only ¢nal mineralization step. The purpose of our study was to understand the fermentation pattern of cellulose degradation, both the initial phase and the later phase of anoxia, and to balance the £ow of carbon and electrons during the degradation of cellulose in both phases. For this purpose, we used a model system in which the processes are mimicked that normally take place in the beginning of the rice growing season when the drained, ploughed and harrowed rice ¢elds are £ooded, i.e. we prepared anoxic soil slurries from dry Italian rice ¢eld soil. 2. Materials and methods 2.1. Soil sample and slurry incubation Rice ¢eld soil was taken in 1993 from the experimental

¢eld of the Italian Rice Research Institute in Vercelli. Detailed ¢eld descriptions and soil characteristics have already been provided [27]. The air-dried soil was mechanically crushed and sieved ( 6 0.5 mm) before use. Soil slurry was prepared by adding 28 ml of distilled and sterilized water to 28 g dry soil in a sterile 120 ml serum bottle, giving a ¢nal volume of the soil slurry of 32.7 ml. The bottles were then closed with sterile black rubber stoppers and crimped with aluminum caps. The gaseous head space was exchanged to N2 . All treatments were done in duplicate. The soil slurries were incubated at 30³C without shaking to avoid damage of the methanogenic community [28]. At given time intervals, gas samples (1 ml) were taken from the head space after vigorously shaking the bottles by hand and then analyzed for H2 , CO2 and CH4 . Liquid samples (1 ml) were taken and centrifuged at 14 000Ug for 5 min. The supernatant was membrane-¢ltered (0.2 Wm; polytetra£uoroethylene; Sartorius, Go«ttingen, Germany) and stored frozen (320³C) until analysis. Concentrations are given in Wmol g31 dry soil; 1 Wmol g31 is equivalent to 1 mM in the aqueous phase. 2.2. Tracer experiments Amorphous, [U-14 C]cellulose ((C6 H10 O5 )n ; 925 MBq g ) from Nicotiana tabacum L. was purchased from American Radiolabeled Chemical Inc. The amount of radioactivity added per bottle of soil slurry was 1.2^2.5 Mdpm (60 Bq = 1 dpm). Two di¡erent experiments were conducted. The ¢rst experiment was designed to study anaerobic cellulose degradation under the conditions when methanogenesis was still inactive and instead, endogenous nitrate, sulfate and iron(III) were being reduced. According to previous studies, endogenous electron acceptors (nitrate, ferric iron and sulfate) are sequentially reduced during the ¢rst 2 weeks of anoxic soil incubation [29,30]. To study the degradation of cellulose during the phase of reduction of these endogenous electron acceptors, [U-14 C]cellulose was added to freshly prepared soil slurries, immediately after the gaseous head space was exchanged with N2 . For convenience, the term `fresh soil' will be used to refer to this experiment. The second experiment was designed to study cellulose degradation during the steady state phase of methanogenesis (`methanogenic soil'). For this purpose, the soil slurries were preincubated for 45 days until all the endogenous electron acceptors had been reduced and CH4 was produced under steady state conditions. Absence of endogenous electron acceptors, establishment of the stable concentrations of H2 and acetate, and a linear increase in CH4 production were considered as the criteria for steady state methanogenesis. [U-14 C]cellulose was then added to the soil slurries and formation of degradation products was followed. Degradation of [U-14 C]cellulose was also measured in the presence of chloroform (100 WM) applied as methanogenic inhibitor. 31

FEMSEC 1090 27-12-99

A. Chidthaisong, R. Conrad / FEMS Microbiology Ecology 31 (2000) 87^94

2.3. Analyses and calculations CH4 , CO2 , and H2 concentrations were measured by gas chromatography as previously described [31]. Radioactive CH4 and CO2 were determined by a gas chromatograph with a radioactivity detector [32]. Organic acids and their radioactivities were determined by high pressure liquid chromatography with the outlet of the detector being connected to a radioactive scintillation monitor [33]. Radioactive peaks of cellobiose and glucose were not completely separated and therefore are shown as a combined `cellobiose/glucose' peak. Soil incubation was terminated by addition of 1 ml of 7 N sulfuric acid to each soil bottle. The ratio of CO2 concentration and its radioactivity between before and after acidi¢cation was used to calculate the total CO2 pool and its radioactivity. The use of this ratio was based on the assumption that the CO2 in the gas and aqueous phases were in equilibrium throughout the experiment. Typically, slurry acidi¢cation resulted in the release of 14 CO2 from dissolved radioactive bicarbonate and carbonate to the gas phase, but had no signi¢cant e¡ect on the recovery of 14 CH4 . When used below, the term 14 CO2 is assumed to represent total 14 CO2 (CO2 +bicarbonate+carbonate). Since the endogenous pool concentration of cellulose was not determined, the actual turnover rate of cellulose could not be determined. Instead, the transformation rate constant (k) of cellulose was calculated from the temporal accumulation of total 14 CO2 and 14 CH4 , based on the assumption that the maximum radioactivity recovered as gaseous products corresponded to the radioactivity available to the soil microbes [34,35]. A k-value was calculated from the slope of a semi-logarithmic plot of {13[(14 CO2 +14 CH4 )/(max 14 CO2 +max 14 CH4 )]} against the incubation time, with (14 CO2 +14 CH4 ) = radioactivity at a given time point; and (max 14 CO2 +max 14 CH4 ) = maximum radioactivity in the gas phase at the

89

end of incubation when [U-14 C]cellulose was exhausted. In the presence of chloroform, k was calculated from the accumulation of 14 CO2 +[14 C]acetate because of the absence of CH4 production. 3. Results 3.1. Decomposition of [U-14 C]cellulose in fresh soil The sequential reduction process in Italian rice ¢eld soil has been well investigated, e.g. in a parallel study using the same soil [36]. Nitrate was consumed within less than one day. Reduction of ferric iron also started immediately but usually lasted until about day 5^7. Sulfate reduction started at about day 5 and continued until about day 10^15. Thus, reduction of endogenous electron acceptors in this Italian rice soil was completed within 10^15 days after the beginning of anoxic incubation. In the present study, acetate accumulated with a lag phase of about 3 days and reached a maximum concentration of 2 mM at about day 10 (Fig. 1A). Propionate accumulation started after a lag phase of about 10 days. In the presence of chloroform, however, the accumulation patterns were di¡erent. Acetate accumulation started immediately without lag phase, and steadily continued for the entire incubation period of 40 days (Fig. 1A). Propionate also started to increase earlier in the presence than in the absence of chloroform. Before day 7, the accumulation of acetate and propionate in the presence of chloroform was not due to inhibition of methanogenesis, since linear CH4 production did not begin before day 7. After day 7 the rate of CH4 production was about 675 nmol g31 day31 . No CH4 production was observed when chloroform was added (Fig. 1B). Production of CO2 , on the other hand, increased without lag phase at a rate of 1056 and 975 nmol g31 day31 in the absence and the

Fig. 1. Accumulation of (A) acetate and propionate, and (B) CO2 and CH4 in anoxically incubated fresh rice ¢eld soil in the presence and absence of chloroform; mean þ S.D. of duplicates.

FEMSEC 1090 27-12-99

90

A. Chidthaisong, R. Conrad / FEMS Microbiology Ecology 31 (2000) 87^94

Fig. 2. Conversion of [U-14 C]cellulose to radioactive (A) cellobiose/glucose and acetate, and radioactive (B) CO2 and CH4 in anoxically incubated fresh rice ¢eld soil in the presence and absence of chloroform; mean þ S.D. of duplicates.

presence of chloroform, respectively (Fig. 1B). Thus, chloroform addition had no clear e¡ect on CO2 production when it was added to the freshly prepared soil slurry. When [U14 -C]cellulose was added to the fresh soil at day 0, radioactive cellobiose/glucose (Fig. 2A) and 14 CO2 were formed immediately (Fig. 2B). Radioactive cellobiose/glucose subsequently disappeared within a few days of soil incubation (Fig. 2A). After a lag phase of about 5 days, [14 C]acetate started to accumulate, reached a maximum at day 7^13 and then decreased again (Fig. 2A). The decrease of [14 C]acetate was paralleled by an increase of 14 CH4 (Fig. 2B). The time course of [14 C]acetate accumulation was similar to that of the acetate accumulation shown in Fig. 1A. At the maximum, 24% of the added radioactivity was recovered as [14 C]acetate if no chloroform was added. With chloroform addition, [14 C]acetate accumulated steadily until a maximum recovery of 46% of the initially added radioactivity in the [U-14 C]cellulose was reached (Table 1). No radioactive propionate was detectable throughout the experimental period, indicating that cellulose was not an important substrate for propionate production. Radioactive CH4 was detected after about day 7, at approximately the same time as when unlabeled CH4 be-

came detectable (Fig. 2B). Radioactive CO2 , on the other hand, was detected immediately after addition of [U-14 C]cellulose. At the end of incubation, 20% of the added radioactivity was converted to 14 CH4 (Table 1). Regardless of the presence or absence of chloroform, approximately 50^56% of radioactive cellulose was converted to 14 CO2 . Thus, the ratio of 14 CH4 /14 CO2 produced from [U-14 C]cellulose at the end of the experiment was approximately 1:3. The total recovery of radioactivity (radioactive CO2 +CH4 +acetate) at the end of incubation was 76% and 96% in the absence and the presence of chloroform, respectively (Table 1). A semi-logarithmic plot of the [U-14 C]cellulose converted to radioactive CO2 +CH4 or acetate showed slightly lower transformation rate constants in the presence than in the absence of chloroform (Fig. 3; Table 1), equivalent to cellulose turnover times of 7^9 days. 3.2. Decomposition of [U-14 C]cellulose in methanogenic soil Some characteristics of soil conditions after 45 days of preincubation were: H2 partial pressure varied between

Table 1 Transformation rate constant (k) and recovery of [U-14 C]cellulose during 40 days of anoxic incubation of fresh and methanogenic rice ¢eld soil in the presence and absence of chloroform as inhibitor of methanogenesis; mean þ S.D. of n = 2; means were statistically not di¡erent at P = 0.05 Soil sample

Fresh soil Control +CHCl3 Methanogenic soil Control +CHCl3

k (day31 )

Maximum recovery (%) of initially added radioactivity in Acetate

CO2

CH4

Total

0.148 þ 0.024 0.110 þ 0.003

24a 46

56 50

20 0

76 96

0.103 þ 0.017 0.073 þ 0.005

24a 66

38 28

38 0

76 94

a 14

[ C]acetate accumulated only transiently and disappeared before maximum accumulation of

FEMSEC 1090 27-12-99

14

CO2 and

14

CH4 was obtained.

A. Chidthaisong, R. Conrad / FEMS Microbiology Ecology 31 (2000) 87^94

2 and 5 Pa (data not shown), acetate concentrations varied between 130 and 170 WM, and propionate was below the detection limit (Fig. 4). Rates of CH4 production were 354 nmol g31 day31 . In the presence of chloroform, however, CH4 production was completely inhibited (data not shown), acetate accumulated steadily right from the beginning and propionate started to accumulate after about 5 days (Fig. 4). Rates of CO2 production were 580 and 407 nmol g31 day31 in the absence and the presence of chloroform, respectively. When the [U-14 C]cellulose was added to this methanogenic soil, radioactive cellobiose/glucose, [14 C]acetate, 14 CO2 and 14 CH4 were produced immediately (Fig. 5). Compared to the fresh soil, however, [14 C]acetate decreased more rapidly and disappeared within 3^5 days after [U-14 C]cellulose addition (Fig. 5A). In the presence of chloroform, on the other hand, [14 C]acetate continued to accumulate until the end of incubation (40 days), when about 66% of the initially added [U-14 C]cellulose were recovered as [14 C]acetate (Table 1). In the methanogenic soil slurry, both 14 CH4 and 14 CO2 were produced simultaneously at a ratio of about 1:1 (Fig. 5B). At the end of the experiment, 37.7 and 38.2% of the added radioactivity were recovered as 14 CH4 and 14 CO2 , respectively (Table 1). Addition of chloroform completely inhibited production of the 14 CH4 and partially inhibited the production of 14 CO2 (by 30% compared to control). At the end of the experiment, the recovery of the radioactivity as

91

Fig. 4. Accumulation of acetate and propionate in methanogenic rice ¢eld soil in the presence and absence of chloroform; mean þ S.D. of duplicates.

14

CO2 +14 CH4 +[14 C]acetate was 76% and 94% in the absence and the presence of chloroform, respectively. The cellulose transformation rate constant calculated from the accumulation of 14 CO2 plus 14 CH4 or [14 C]acetate were 0.103 day31 (turnover time = 10 days) and 0.073 day31 (14 days) without and with chloroform, respectively (Fig. 3; Table 1). Thus, addition of chloroform inhibited the rate of cellulose transformation approximately 30% compared to the control. 4. Discussion

Fig. 3. Semi-logarithmic plot of the transformation of [U-14 C]cellulose to radioactive products in fresh and in methanogenic rice ¢eld soil; mean þ S.D. of duplicates ; the radioactive products formed at time t (14 Ct ) were normalized against the maximum amount of radioactive products formed (14 Cmax ), being 14 CO2 +14 CH4 in the absence of chloroform and 14 CO2 +[14 C]acetate in the presence of chloroform.

Methanogenic decomposition of cellulose produces CO2 and CH4 at a ratio of 1:1. However, when methanogenesis is not yet developed, cellulose degradation can couple with reduction of nitrate, ferric iron and sulfate. In this case, cellulose degradation serves as source of electrons and hence CO2 instead of CH4 is produced. Thermodynamic theory predicts that electron acceptors with a higher redox potential are reduced ¢rst, i.e. nitrate before ferric iron and before sulfate [24]. Hence, linear CH4 production will not start until reduction of nitrate, ferric iron and sulfate are complete [25,26]. The actual ratio between CO2 and CH4 produced from cellulose degradation, therefore, will depend on how much nitrate, ferric iron and sulfate are present in the soil. Our results show that [U-14 C]cellulose was exclusively oxidized to 14 CO2 , as long as endogenous electron acceptors were available. However, after 10 days, when endogenous electron donors became limiting (see [36]), [U-14 C]cellulose was also converted to 14 CH4 .

FEMSEC 1090 27-12-99

92

A. Chidthaisong, R. Conrad / FEMS Microbiology Ecology 31 (2000) 87^94

Fig. 5. Conversion of [U-14 C]cellulose to radioactive (A) cellobiose/glucose and acetate, and radioactive (B) CO2 and CH4 in methanogenic rice ¢eld soil in the presence and absence of chloroform; mean þ S.D. of duplicates.

Addition of [U-14 C]cellulose to freshly prepared soil slurry allowed us to follow the degradation products right from the beginning of anoxic incubation. Radioactive compounds accumulated in a similar manner as non-radioactive CO2 , acetate, and CH4 accumulated in the absence of added [U-14 C]cellulose, indicating that endogenous cellulose was degraded similarly as the added [U-14 C]cellulose. Degradation of [U-14 C]cellulose started right after the soil was made anoxic. However, [14 C]acetate started to accumulate later, possibly indicating that [14 C]acetate produced may have been immediately oxidized to 14 CO2 by coupling to reduction of nitrate, ferric iron and/or sulfate. [14 C]acetate accumulated transiently when these endogenous electron acceptors became limiting and 14 CH4 production started. Eventually, the accumulated [14 C]acetate decreased and was replaced by production of only 14 CH4 and 14 CO2 . This result is consistent with similar patterns of product formation when adding [U-14 C]glucose, demonstrating that the concentration of acetate in rice ¢eld soil is the result of acetate production by fermentation of carbohydrates that is more or less balanced by acetate consumption by either the reduction of nitrate, ferric iron and/or sulfate (in the early phase of submergence) or methanogenesis (in the late phase of submergence) [37]. Transient accumulation of acetate is generally observed within 2 weeks of soil £ooding [25,36, 38,39]. The transformation rate constants of [U-14 C]cellulose were similar when either fresh or methanogenic soil was used (Table 1). The recovery of radioactivity was also similar under both soil conditions. However, the experiment with the fresh soil contained both the phase during which endogenous electron acceptors were reduced and the phase when CH4 was produced. Therefore, it is not possible to clearly di¡erentiate the e¡ect of the late methanogenic versus the early reduction phase on cellulose

turnover. Nevertheless, it seems that cellulose degradation progressed similarly. This was also reported for freshwater lake sediment, where organic matter decomposition was accompanied by reduction of ferric iron or methanogenesis [40]. These authors found that addition of ferric iron to the sediment did not signi¢cantly alter the amount of organic matter mineralized, although the production of CH4 was greatly inhibited. Thus, reducing equivalents not used for CH4 production were instead used for ferric iron reduction. Our results can be explained in a similar way. In the fresh soil, the recovery of [U-14 C]cellulose as 14 CO2 and 14 CH4 was 20% and 56%, respectively. In the methanogenic soil, recovery was 38% for both gases. The discrepancy between the amount of CH4 produced under these two soil conditions was attributed to the fact that the reducing equivalents generated from cellulose degradation were diverted towards CH4 production under the methanogenic soil condition. Namely, about 20% more CO2 was produced in the fresh soil and this can be compensated for by a corresponding increase in labeled CH4 production of about 20% (Table 1). Thus, more CH4 production from cellulose in the methanogenic soil than in fresh soil was a result of a diversion of electron £ow from cellulose decomposition towards CH4 production, but not a result of more rapid degradation of cellulose. Furthermore, when methanogenesis was inhibited by chloroform, reducing equivalents were disposed as acetate. With chloroform, 46% of [U-14 C]cellulose was converted to [14 C]acetate in fresh soil, compared to 66% in methanogenic soil. Since as much CO2 as CH4 is produced during aceticlastic methanogenesis, the [14 C]acetate accumulated in the presence of chloroform would in the absence of chloroform be equivalent to 23% and 33% recovery of 14 CH4 in fresh and methanogenic soil, respectively. This agrees reasonably well with the amount of 14 CH4 produced under these two soil conditions (Table 1).

FEMSEC 1090 27-12-99

A. Chidthaisong, R. Conrad / FEMS Microbiology Ecology 31 (2000) 87^94

We were unable to detect [14 C]propionate among the degradation products of [U-14 C]cellulose. A previous study [33] detected small amounts ( 6 10% of initially added radioactivity) of propionate during degradation of position-labeled glucose. However, Chidthaisong et al. [36,37] were also unable to detect [14 C]propionate when [U-14 C]glucose was added to anoxic rice ¢eld soil. We assume that the discrepancy is due to di¡erent soil characteristics, since the experiments were conducted with Italian rice ¢eld soil from di¡erent years. In the present study, we detected no radioactive propionate from [U-14 C]cellulose, although non-radioactive propionate accumulated transiently. Hence, degradation of compounds other than cellulose must have served as substrate for propionate production. We found that accumulation of propionate was especially observed when acetate accumulated to a high concentration, regardless of the presence of chloroform (Figs. 1, 4). Accumulation of propionate may thus be explained by thermodynamically unfavorable conditions of propionate degradation when methanogenesis was inhibited (compare [39]). The soil organic compounds from which propionate was produced are unknown. Xylan or pectin may be possible substrates. Propionate was found to accumulate when xylan but not when cellulose was added to a paddy soil from the Camargue [23]. Similarly, Chin et al. [41] observed stimulation of propionate production after addition of xylan or pectin, but not cellulose to methanogenic Italian rice ¢eld soil. A moderate propionate production from glucose and xylose fermentation was only observed with bacteria isolated on xylan or pectin, but not cellulose. A cellulose-degrading isolate produced only marginal amounts of propionate, even when coupled to methanogenesis by co-culture with a H2 -consuming methanogen [41]. In conclusion, it appears that only the pattern of product formation but not the rate of cellulose degradation is in£uenced by the presence or absence of endogenous electron acceptors (nitrate, ferric iron, sulfate) and that the CO2 produced during the early phase of anoxia is replaced by CH4 during the later methanogenic phase in £ooded rice ¢eld soil. Furthermore, degradation of organic substrates other than cellulose seems to a¡ect the pattern of fermentation products, e.g. propionate. Acknowledgements A.C. was supported by a fellowship of the Alexandervon-Humboldt foundation. We also thank the Fonds der Chemischen Industrie for ¢nancial support.

References [1] Vogels, G.D. (1979) The global cycle of methane. Antonie van Leeuwenhoek 45, 347^352.

93

[2] Ljungdahl, L.G. and Eriksson, K.-E. (1985) Ecology of microbial cellulose degradation. Adv. Microb. Ecol. 8, 237^299. [3] Vlasenko, E.Y., Ding, H., Labavitch, J.M. and Shoemaker, S.P. (1997) Enzymatic hydrolysis of pretreated rice straw. Bioresour. Technol. 59, 109^119. [4] Watanabe, A., Katoh, K. and Kimura, M. (1993) E¡ect of rice straw application on CH4 emission from paddy ¢elds. 2. Contribution of organic constituents in rice straw. Soil Sci. Plant Nutr. 39, 707^ 712. [5] Inoko, A. (1984) Compost as a source of plant nutrients. In: Organic Matter and Rice (IRRI, Ed.), pp. 137^145. International Rice Research Institute, Los Banos, The Philippines. [6] Ponnumperuma, F.N. (1984) Straw as a source of nutrients for wetland rice. In: Organic Matter and Rice (IRRI, Ed.), pp. 117^136. International Rice Research Institute, Los Banos, The Philippines. [7] Chidthaisong, A. and Watanabe, I. (1997) Methane formation and emission from £ooded rice soil incorporated with 13 C-labeled rice straw. Soil Biol. Biochem. 29, 1173^1181. [8] Kimura, M., Miura, Y., Watanabe, A., Katoh, T. and Haraguchi, H. (1991) Methane emission from paddy ¢eld. 1. E¡ect of fertilization, growth stage and midsummer drainage pot experiment. Environ. Sci. 4, 265^271. [9] Shibahara, F. and Inubushi, K. (1997) E¡ect of organic matter application on microbial biomass and available nutrients in various types of paddy soils. Soil Sci. Plant Nutr. 43, 191^203. [10] Sass, R.L., Fisher, F.M., Harcombe, P.A. and Turner, F.T. (1991) Mitigation of methane emissions from rice ¢elds: Possible adverse e¡ects of incorporated rice straw. Glob. Biogeochem. Cycles 5, 275^287. [11] Yagi, K. and Minami, K. (1990) E¡ect of organic matter application on methane emission from some Japanese paddy ¢elds. Soil Sci. Plant Nutr. 36, 599^610. [12] Seiler, W. (1984) Contribution of biological processes to the global budget of CH4 in the atmosphere. In: Current Perspectives in Microbial Ecology (Klug, M.J. and Reddy, C.A., Eds.), pp. 468^477. American Society for Microbiology, Washington, DC. [13] Khalil, M.A.K. and Rasmussen, R.A. (1983) Sources, sinks, and seasonal cycles of atmospheric methane. J. Geophys. Res. 88, 5131^ 5144. [14] Zehnder, A.J.B. (1978) Ecology of methane formation. In: Water Pollution Microbiology, Vol. 2 (Mitchell, R., Ed.), pp. 349^376. Wiley, New York. [15] Schink, B. (1997) Energetics of syntrophic cooperation in methanogenic degradation. Microb. Mol. Biol. Rev. 61, 262. [16] Dol¢ng, J. (1988) Acetogenesis. In: Biology of Anaerobic Microorganisms (Zehnder, A.J.B., Ed.), pp. 417^468. Wiley, New York. [17] Ljungdahl, L.G. and Eriksson, K.-E. (1985) Ecology of microbial cellulose degradation. Adv. Microb. Ecol. 8, 237^299. [18] Colberg, P.J. (1988) Anaerobic microbial degradation of cellulose, lignin, oligolignols, and nomoaromatic lignin derivatives. In: Biology of Anaerobic Microorganisms (Zehnder, A.J.B., Ed.), pp. 333^372. Wiley, New York. [19] Malburg, L.M., Lee, J.M.T. and Forsberg, C.W. (1992) Degradation of cellulose and hemicelluloses by rumen microorganisms. In: Microbial Degradation of Natural Products (Winkelmann, G., Ed.), pp. 127^159. VCH, Weinheim. [20] Beguin, P. and Aubert, J.P. (1994) The biological degradation of cellulose. FEMS Microbiol. Rev. 13, 25^58. [21] Leschine, S.B. (1995) Cellulose degradation in anaerobic environments. Annu. Rev. Microbiol. 49, 399^426. [22] Saito, M., Wada, H. and Takai, Y. (1990) Development of a microbial community on cellulose buried in waterlogged soil. Biol. Fertil. Soils 9, 301^305. [23] Joulian, C., Ollivier, B., Neue, H.U. and Roger, P.A. (1996) Microbiological aspects of methane emission by a rice¢eld soil from the Camargue (France). 1. Methanogenesis and related micro£ora. Eur. J. Soil Biol. 32, 61^70.

FEMSEC 1090 27-12-99

94

A. Chidthaisong, R. Conrad / FEMS Microbiology Ecology 31 (2000) 87^94

[24] Ponnamperuma, F.N. (1972) The chemistry of submerged soils. Adv. Agron. 24, 29^96. [25] Yao, H. and Conrad, R. (1999) Thermodynamics of methane production in di¡erent rice paddy soils from China, the Philippines, and Italy. Soil Biol. Biochem. 31, 463^473. [26] Yao, H., Conrad, R., Wassmann, R. and Neue, H.U. (1999) E¡ect of soil characteristics on sequential reduction and methane production in sixteen rice paddy soils from China, the Philippines, and Italy. Biogeochemistry 47, 269^295. [27] Holzapfel-Pschorn, A., Conrad, R. and Seiler, W. (1986) E¡ects of vegetation on the emission of methane from submerged paddy soil. Plant Soil 92, 223^233. [28] Dannenberg, S., Wudler, J. and Conrad, R. (1997) Agitation of anoxic paddy soil slurries a¡ects the performance of the methanogenic microbial community. FEMS Microbiol. Ecol. 22, 257^263. [29] Klu«ber, H.D. and Conrad, R. (1998) E¡ects of nitrate, nitrite, NO and N2 O on methanogenesis and other redox processes in anoxic rice ¢eld soil. FEMS Microbiol. Ecol. 25, 301^318. [30] Roy, R., Klu«ber, H.D. and Conrad, R. (1997) Early initiation of methane production in anoxic rice soil despite the presence of oxidants. FEMS Microbiol. Ecol. 24, 311^320. [31] Conrad, R., Schu«tz, H. and Babbel, M. (1987) Temperature limitation of hydrogen turnover and methanogenesis in anoxic paddy soil. FEMS Microbiol. Ecol. 45, 281^289. [32] Conrad, R., Mayer, H.P. and Wu«st, M. (1989) Temporal change of gas metabolism by hydrogen-syntrophic methanogenic bacterial associations in anoxic paddy soil. FEMS Microbiol. Ecol. 62, 265^274. [33] Krumbo«ck, M. and Conrad, R. (1991) Metabolism of position-la-

[34]

[35]

[36]

[37]

[38]

[39]

[40]

[41]

belled glucose in anoxic methanogenic paddy soil and lake sediment. FEMS Microbiol. Ecol. 85, 247^256. Phelps, T.J. (1991) Similarity between biotransformation rates and turnover rates of organic matter biodegradation in anaerobic environments. J. Microbiol. Methods 13, 243^254. Schu«tz, H., Seiler, W. and Conrad, R. (1989) Processes involved in formation and emission of methane in rice paddies. Biogeochemistry 7, 33^53. Chidthaisong, A. and Conrad, R. (1999) Turnover of glucose and acetate coupled to reduction of nitrate, ferric iron and sulfate and to methanogenesis in anoxic rice ¢eld soil. FEMS Microbiol. Ecol. 31, 73^86. Chidthaisong, A., Rosenstock, B. and Conrad, R. (1999) Measurement of monosaccharides and conversion of glucose to acetate in anoxic rice ¢eld soil. Appl. Environ. Microbiol. 65, 2350^2355. Chin, K.J. and Conrad, R. (1995) Intermediary metabolism in methanogenic paddy soil and the in£uence of temperature. FEMS Microbiol. Ecol. 18, 85^102. Krylova, N.I., Janssen, P.H. and Conrad, R. (1997) Turnover of propionate in methanogenic paddy soil. FEMS Microbiol. Ecol. 23, 107^117. Lovley, D.R. and Phillips, E.J.P. (1986) Organic matter mineralization with reduction of ferric iron in anaerobic sediments. Appl. Environ. Microbiol. 51, 683^689. Chin, K.J., Rainey, F.A., Janssen, P.H. and Conrad, R. (1998) Methanogenic degradation of polysaccharides and the characterization of polysaccharolytic clostridia from anoxic rice ¢eld soil. Syst. Appl. Microbiol. 21, 185^200.

FEMSEC 1090 27-12-99