Inhibition of methanogens increases photo-dependent nitrogenase activities in anoxic paddy soil amended with rice straw

Authors

  • Naoki Harada,

    Corresponding author
    1. Department of Applied Biological Chemistry, Graduate School of Agricultural and Life Sciences, The University of Tokyo, 1-1-1 Yayoi, Bunkyo-ku, Tokyo 113-8657, Japan
    Search for more papers by this author
  • Masaya Nishiyama,

    1. Department of Applied Biological Chemistry, Graduate School of Agricultural and Life Sciences, The University of Tokyo, 1-1-1 Yayoi, Bunkyo-ku, Tokyo 113-8657, Japan
    Search for more papers by this author
  • Satoshi Matsumoto

    1. Department of Applied Biological Chemistry, Graduate School of Agricultural and Life Sciences, The University of Tokyo, 1-1-1 Yayoi, Bunkyo-ku, Tokyo 113-8657, Japan
    Search for more papers by this author

*Corresponding author. Tel.: +81 (3) 5841-5176; Fax: +81 (3) 5841-8042; E-mail: aa87032@mail.ecc.u-tokyo.ac.jp

Abstract

The interaction between phototrophic dinitrogen fixers and methanogens was examined in soil slurries amended with rice straw using 2-bromoethanesulfonic acid (BES), a specific methanogenic inhibitor. Slurries incubated in light increased phototrophic nitrogenase activity (acetylene reducing activity), and showed growth of phototrophic purple bacteria and reduction of CH4 emission, indicating outcompetition of purple bacteria with methanogens in photic zones. Adding BES effectively inhibited methane production and markedly increased phototrophic acetylene reducing activity accompanied with acetate accumulation, but did not affect populations of purple bacteria in the slurries. More acetate accumulated in the inhibited slurries incubated in dark. We suggest that increased availability of organic substrates for purple bacteria after stopping methanogenic consumption by BES caused the increased phototrophic acetylene reducing activity. These results indicate that, after purple bacteria grow enough, performance of their N2 fixation may be limited by substrate availability, which methanogenesis may profoundly influence.

1. Introduction

Straw application into lowland rice paddies is a traditional agricultural practice, which has been primarily used to add nutrients and organic matter to the soil. Long-term experiments indicated that repeated application of straw increases nutrient elements in soil, such as carbon, phosphate, potassium, silica, and nitrogen [1], thus resulting in the advantage of increased plant growth and grain yield [1–4]. In particular, increasing effects on N fertility have been noted, because N is often a limiting nutrient element in crop production systems and several studies have clarified that straw application promotes biological N2 fixation in rice paddies [5–8].

N2-fixing microorganisms in rice paddies are generally classified into the following two groups: heterotrophs in the rice rhizosphere or bulk soil that depend on external carbon for energy, and phototrophs in floodwater and the soil surface that obtain their energy from photosynthesis. Many studies have shown the improving effects of straw application on biological N2 fixation in laboratory experiments [9–12]. Reddy and Patrick [13] published the earliest study that compared the straw effects on phototrophs with heterotrophs. They incubated waterlogged soil with straw in light and dark, and found that the incubation in light resulted in a 3–4 times higher nitrogenase activity. In a field experiment, Matsuguchi and Yoo [14] measured nitrogenase activity during cultivation after straw application and showed that the photo-dependent nitrogenase activity in the surface layer was much higher than the photo-independent nitrogenase activity in the whole profile of the soil. These findings indicate the importance of straw as a substrate or habitat for phototrophs.

Despite frequent reports on N2 fixers in paddy soils, no attempts have been made to study the interaction between phototrophic N2 fixers and other microorganisms, except for a series of experiments on nitrogenase activity in mixed cultures [15–19]. The possibility of interactions between phototrophic N2 fixers and methanogens in rice fields amended with rice straw is interesting, as straw application greatly increases emission of methane, a powerful greenhouse gas [20–25]. Recent studies on vertical profiles of methanogenic populations or CH4-producing activity indicated that the numbers of methanogens are constant in plough layers of soil regardless of the depth [26] and that they occasionally initiate CH4 production even at surface layers [27,28]. These findings indicate the possibility that methanogens exist near phototrophs. This study examined the interaction between phototrophic N2 fixers, particularly phototrophic purple bacteria (PB), and methanogens, assuming that CH4 production is able to start in the surface of rice paddies. To study the influence of methanogens on biological N2 fixation, 2-bromoethanesulfonic acid (BES), a specific methanogenic inhibitor [29,30], was used.

2. Materials and methods

2.1 Soils and rice straw

An alluvial paddy soil was sampled from an unflooded rice field of the Saitama Prefecture Agriculture and Forestry Research Center (Kumagaya, Japan) in April 1998 and was stored in dark and cool (4°C) conditions. Before the experiments, the soil was passed through a 2 mm mesh stainless steel sieve. Properties of the soil were: total C: 1.60% (w/w), total N: 0.18% (w/w), pH (H2O): 5.6, pH (KCl): 4.4.

Rice straw was collected in the fall of 1997 and was used after it was finely ground and passed through a 0.5 mm mesh stainless steel sieve. Total C and total N were 37.6% (w/w) and 0.70% (w/w), respectively.

2.2 Preparation and incubation of soil slurries

Soil slurries were prepared in 68-ml clear serum bottles to measure soil nitrogenase activities (five replications), to measure the pH (three replications), to enumerate PB (three replications), and to assay low molecular mass fatty acids (three replications). Each bottle received fresh soil equivalent to 10 g dry weight (dw) soil and 0.3 g ground rice straw, and distilled water was added to give a total water content of 20 ml. Each bottle was closed with a butyl rubber septum and an aluminum cap. A 0.70 mm internal diameter (i.d.) hollow needle was inserted through the septum to keep the inside at atmospheric pressure during the experiment. The bottles were statically incubated at 30°C in a 12-h light and dark cycle (light incubation) or in 24-h dark (dark incubation). The light source was 75-W tungsten lamps (ca. 1000 lux).

The procedure to prepare and incubate soil slurries to measure CH4 emissions (three replications) was the same as for measuring the nitrogenase activities with the following two exceptions: the soil slurries were prepared in 120-ml clear serum bottles, not 68-ml, and the top needles were not used.

To evaluate the influence of methanogens on biological N2 fixation, some bottles received BES (sodium salt, Sigma, St. Louis, USA) before starting the incubation. Stock solutions of BES were prepared with distilled water and were added to soil slurries to obtain a final concentration of 0.0005, 0.05 or 5 mM in soil solution. In another experiment, BES was added after CH4 production was steady. After 4 weeks incubation in light, aliquots (4 ml) of surface water were replaced with the same aliquots of 25 mM BES solution (5 mM final concentration) and the incubation was continued.

2.3 pH measurement

An incubated bottle was shaken vigorously for 1 min and the soil suspension was moved into a beaker. A combination pH electrode (GST-5311C, Toa Electronics, Tokyo, Japan) was inserted into the soil suspension, and the soil pH was measured using a pH meter (HM-40S, Toa Electronics, Tokyo, Japan).

2.4 Enumeration of PB

PB in the slurries were enumerated by the most probable number (MPN) method with the medium described by Hoshino and Satoh [31]. As carbon sources, 1.0 g of sodium DL-malate and 1.7 g of sodium acetate trihydrate were added to 1 l of the medium. The medium is usually used for growing phototrophic purple non-sulfur bacteria, but a few species of phototrophic purple sulfur bacteria, which can grow photoheterotrophically using organic substrates in the absence of sulfide or elemental sulfur [33], may also be enumerated. The total slurry in an incubated bottle was transferred to a 250-ml plastic bottle with 80 ml of sterile distilled water and was shaken vigorously for 10 min at room temperature (defined as the first dilution solution). Five sterile clear 10-ml screw-cap glass tubes for each dilution solution were prepared. Each tube filled with 1 ml of dilution solution and 9 ml of autoclaved medium (121°C, 20 min) was capped tightly and incubated at 30°C under tungsten lamps (ca. 1000 lux). After 3 weeks incubation, the tubes that had turned purple, red or brown were noted as positive and numbers of viable cells were determined.

2.5 CH4 emission

CH4 emitted from the slurries was measured using gas chromatography (GC). An incubated bottle was shaken for 1 min vigorously. A gas sample (0.1 ml) taken from the headspace of each bottle with a gas-tight syringe was injected into the gas chromatograph. To avoid excessive pressure build-up, a headspace gas pressured by emitted gases was transferred to another bottle every 2 weeks. CH4 in the collected gases was also measured using GC and the total amount of CH4 emitted during the incubation was calculated. Section 2.7 describes the detailed analytic conditions.

2.6 Determination of soil nitrogenase activities

Soil nitrogenase activity was estimated by acetylene (C2H2) reducing activity (ARA). The method of Matsuguchi et al. [32] for waterlogged soils was used with modifications. Incubated bottles were gradually evacuated using a gas exchanger (IP-8, Sanshin, Yokohama, Japan) equipped with a vacuum pump to 105 Pa of the suction port pressure. After being kept at 105 Pa for 1 min, the pressure was gradually returned to atmospheric pressure by filling with 10% C2H2 gas (Ar base) and was maintained for 1 min. This procedure for gas exchange was repeated four times. Thereafter, the bottles were incubated for 3 h at 30°C under fluorescent lamps (ca. 10 000 lux) to determine total ARA or in dark to determine photo-independent ARA. Before measuring the produced ethylene (C2H4), the bottles were vigorously shaken for 1 min to diffuse gases remaining in the soil to the headspace. A gas sample (0.5 ml) from the headspace of each bottle was injected into the gas chromatograph. Section 2.7 describes the detailed analytic conditions.

2.7 Gas analysis

Both CH4 and C2H4 were analyzed using a Shimadzu GC14B gas chromatograph (Kyoto, Japan) equipped with a Porapack R column (2 m×3 mm i.d. stainless steel tubing, 80/100 mesh packing, Waters, Milford, USA) and a flame ionization detector. The flow rates of the carrier gas (high-purity He), H2 and air were 60, 50 and 500 ml min−1, respectively. The injector and detector temperatures were 100°C, and the oven temperature was 80°C.

2.8 Assay of low molecular mass fatty acids

Concentrations of formate, acetate, and propionate were measured using reverse phase high performance liquid chromatography. The total slurry in an incubated bottle was transferred to a 250-ml plastic bottle with 80 ml of distilled water and was shaken vigorously for 10 min at room temperature. The suspension was then centrifuged at 3000 rpm for 10 min and the supernatant was frozen at −20°C in a tight bottle. During the analysis, the extract sample was filtered through a 0.20 μm pore size cellulose acetate membrane (DISMIC-13CP, Toyo Roshi, Tokyo, Japan) and each aliquot (50 μl) of the filtrate was injected into a Shimadzu LC6A high performance liquid chromatograph (Kyoto, Japan) equipped with a UV detector set at 210 nm. The separation was made using an Inertsil ODS-3 column (5 μm, 250 mm×4.6 mm i.d., GL Sciences, Tokyo, Japan) thermostated at 40°C and 0.1 M ammonium dihydrogen phosphate solution adjusted to pH 2.5 with phosphoric acid (flow rate 1.0 ml min−1). The detection limits of the system for formate, acetate, and propionate corresponded to 0.3, 0.4, and 0.6 μmol g−1 dw soil, respectively.

2.9 Statistical analysis

Due to non-equivalence of variances, non-parametric methods were used for the following statistical analysis. To find the effects of the treatments on CH4 emission and ARA, the Friedman test with the treatment as a factor and time as a block was used. As a BES effect on CH4 emission depending on the concentration was foreseen, the Jonkheere test with the treatments as a factor was used to compare the total amounts during the incubation. For the other statistical analyses, two-way ANOVA with the treatment as a factor and time as a block was used.

3. Results

3.1 General observation and enumeration of PB

Growth of reddish microorganisms was visible in all bottles incubated in light. They appeared within 2 weeks and remained until the end of the incubation. The area of the red coloration was limited to a small part of the soil surface when the slurries were incubated without BES, whereas it spread, and even the surface water was pink to red, when incubated with the inhibitor. The slurries incubated in dark never showed such coloration.

PB were enumerated to less than 1×102 MPN g−1 dw soil in the initial slurries. The incubation in light increased the number of PB to 1.4×107 MPN g−1 dw soil after 1 week and to over 108 MPN g−1 dw soil after 2 weeks or later (Fig. 1). The numbers were significantly higher than those of the slurries incubated in dark (two-way ANOVA, P<0.001), in which PB ranged from 104 to 106 MPN g−1 dw soil. When the slurries modified by 5 mM BES were incubated in light, the changes of the MPN of PB were similar to those in the unmodified slurries (two-way ANOVA, P>0.05).

Figure 1.

Changes in populations of phototrophic PB in the slurries. The MPN method was used to enumerate PB in the slurries; mean±1 S.D. after logarithmic transformation, n=3.

3.2 Changes in pH

The initial pH of the slurries was 5.6 and gradually increased to 6.5. However, this increase leveled off in the second week to show only negligible changes between 6.3 and 6.6. Neither the light exposure nor the inhibitor treatment made any serious impacts on the time course.

3.3 CH4 emission

Fig. 2 shows the CH4 emission from the slurries during the incubation. Without BES, only a small amount of CH4 was detected in the first week and extensive CH4 production was recorded from the second week. CH4 production continued almost linearly to the end of the incubation. The amount of CH4 emitted in light was slightly but significantly smaller than in the dark (Friedman test, P<0.01). CH4 emissions during the 8 weeks amounted to 264 μmol g−1 dw soil from the slurries incubated in light and 316 μmol g−1 dw soil in dark. Adding BES at the sample preparation inhibited the CH4 production in relation to the concentration (Jonkheere test, P<0.001). The inhibitory effects were maintained for the first 3 weeks and thereafter the release of CH4 from the slurries increased somewhat. CH4 emissions from the slurries incubated with 0.0005, 0.05, and 5 mM BES solution were 197, 94.6, and 19.7 μmol g−1 dw soil, respectively, during the 8 weeks.

Figure 2.

Changes in CH4 emissions from the slurries. Each point shows the amount of CH4 emitted from the slurries during the incubation, measured using GC; mean±1 S.D., n=3.

3.4 Nitrogenase activity

The light exposure significantly increased the total ARA of the slurries (Friedman test, P<0.001, Fig. 3A). The effects appeared in the second week when the highest value was 23.2 nmol C2H4 h−1 g−1 dw soil. Afterwards, the total ARA gradually decreased and the effects were no longer observed from the sixth week. The photo-independent ARA remained unaffected by light exposure throughout the experiments (Friedman test, P>0.05, Fig. 3B). Adding BES at the sample preparation changed the nitrogenase activity of the slurries markedly. When the slurries were incubated in light, total ARA was significantly stimulated (Friedman test, P<0.001) and the amount depended on the concentration of BES (Fig. 4A). Total ARA increased to 424 nmol C2H4 h−1 g−1 dw soil in the fourth week when the samples were incubated with 5 mM BES, and it was approximately 18 times higher than the peak without the inhibitor. Photo-independent ARA was not affected by adding the inhibitor throughout the experiment (Friedman test, P>0.05, Fig. 4B). The slurries incubated in dark showed no changes in ARA, even when they received 5 mM BES (data not shown).

Figure 3.

Changes in (A) total and (B) photo-independent ARA of the slurries incubated without BES. To measure total ARA, the assay was carried out under fluorescent lamps (ca. 10 000 lux) after replacement of the headspace gas with 10% acetylene gas. To measure photo-independent ARA, the assay was carried out in dark after the replacement of the headspace gas; mean±1 S.D., n=5.

Figure 4.

Changes in (A) total and (B) photo-independent ARA of the slurries incubated in light with 0.0005, 0.05, and 5 mM BES at the sample preparation. To measure total ARA, the assay was carried out under fluorescent lamps (ca. 10 000 lux) after replacement of the headspace gas with 10% acetylene gas. To measure photo-independent ARA, the assay was carried out in dark after the replacement of the headspace gas; mean±1 S.D., n=5.

3.5 Changes in low molecular mass fatty acids

Acetate was the major low molecular mass fatty acid in the incubated slurries (Fig. 5A) and propionate was the next (Fig. 5B). Formate was not detected in any samples. When the slurries were incubated without BES, changes in the acetate were not significantly different regardless of the light exposure throughout the incubation (two-way ANOVA, P>0.05, Fig. 5A). The concentrations of the acetate were around 10 μmol g−1 dw soil in the first week and then dropped rapidly. From the third week to the end of the examination, the concentrations of acetate were almost under the detection limit. When the slurries were incubated in light with the inhibitor, another acetate accumulation was recorded after a temporary decrease in the second week. The latter peak of acetate accumulation was relatively broad and low, and continued to the fifth week. When the slurries were incubated in dark with BES, a marked accumulation of acetate was recorded throughout the experiment. Propionate was found only in the slurries incubated in light without BES and the inhibited slurries incubated in dark (Fig. 5B). The concentration of propionate in the former slurries was 1 μmol g−1 dw soil in the first week and it disappeared thereafter. In the latter slurries, propionate was detected throughout the incubation and was inclined to increase. The final concentration after the 8-week incubation was 5 μmol g−1 dw soil, around one tenth of the acetate concentration.

Figure 5.

Changes in concentrations of (A) acetate and (B) propionate. Low molecular mass fatty acids in the slurries were extracted with distilled water and were assayed using reverse phase high performance liquid chromatography. Propionate was detected only in the slurries plotted in the graph; mean±1 S.D., n=3.

3.6 Methanogenic inhibition after CH4 production was steady

Inhibition of CH4 production after it was steady also increased nitrogenase activity of the slurries (Fig. 6A). The ARA reached a peak of 49.3 nmol C2H4 h−1 g−1 dw soil 1 week after adding BES and then decreased gradually. The stimulatory effects on the ARA were lower compared with adding 5 mM BES to the fresh soil (Fig. 4A), although the final BES concentrations were the same. The treatment resulted in some accumulation of acetate (Fig. 6B). Formate and propionate were not detected. The concentrations of acetate were less than 1 μmol g−1 dw soil even at the maximum. The accumulation of acetate was smaller than when adding 5 mM BES to the fresh soil (Fig. 5A).

Figure 6.

Effects of adding BES after CH4 production was steady on (A) total ARA and (B) acetate concentrations of the slurries. Arrows indicate the timing of adding BES and figures in parentheses on the horizontal line indicate time in weeks after adding BES. Broken lines show controls incubated without BES during the incubation; mean±1 S.D., A: n=5, B: n=3.

4. Discussion

Reddish microorganisms were observed in all samples incubated in light. The coloration was associated with growth of PB enumerated to over 108 MPN g−1 dw soil in the slurries incubated in light (Fig. 1). This finding is consistent with earlier studies: growth of Rhodopseudomonas sp. was promoted on rice straw top-dressed onto a submerged soil [14] and populations of phototrophic purple non-sulfur bacteria increased markedly in rice soil improved with straw [8]. In this study, the pH values of the slurries remained weakly acid to neutral, which is the optimum for many representatives of PB [33]. In the experiments of Matsuguchi and Yoo [14], Rhodopseudomonas sp. was succeeded by cyanobacteria. Haque et al. [34] described a similar succession of microflora during the incubation of rice straw with soil extraction. However, the red coloration remained during the incubation and succession of microflora was not observed in this study, probably because the slurries remained anoxic to the end of the incubation, which is unsuitable for cyanobacteria which grow aerobically.

CH4 production started after a short lag phase and remained constant to the end of the experiments (Fig. 2). These results are consistent with earlier laboratory-scale experiments on the effects of straw application on CH4 production in rice paddy soils [35–37]. The lag generally means that time is necessary to complete a sequential reduction of electron acceptors and provision of methanogenic substrates by multimicrobial activities [38–41]. It is inferred that sequential reduction was critical to start CH4 production in this study, as acetate accumulated in the first week of the incubation (Fig. 5A). The fact that the amounts of CH4 emitted from the slurries incubated in light were significantly smaller than those in dark is very interesting. The degree of inhibition by light exposure was calculated as 10–20%. The difference in the final concentrations of CH4 due to the light condition was mainly caused by the difference in the emission rates from the first to the third week and from the sixth to the eighth week. These results indicate that PB outcompete methanogens in photic zones and suggest that sunlight has a certain role in reducing CH4 production through activating PB. The magnitude of this effect in actual rice fields needs to be investigated later.

We measured total and photo-independent ARA for each treatment. As heterotrophic N2 fixers contribute to both total and photo-independent ARA, whereas PB contribute mainly to the total ARA [33], the photo-independent ARA measured in this study can be regarded as heterotrophic ARA and the difference between total and photo-independent ARA may represent nitrogenase activity by PB. Therefore, the result that the light exposure increased the total ARA but not the photo-independent ARA (Fig. 3) supports the importance of PB in straw-applied conditions, as Reddy and Patrick pointed out [13]. On the other hand, the time courses of the phototrophic ARA (Figs. 3A and 4A) were not in line with the changes in the populations of PB (Fig. 1). This fact clearly indicates restrictions other than the population of PB on their nitrogenase activity measured in this study.

Adding BES effectively suppressed methanogenic activity (Fig. 2) and greatly increased the phototrophic ARA (Fig. 4A). To our knowledge, such an increase in ARA by suppressing methanogens has not been reported so far. Why was the phototrophic ARA stimulated by adding BES? Growth of Rhodopseudomonas sp. is not affected by 25 mM BES in a pure culture [30]. In this study, no significant difference in the populations of PB between the inhibited and control slurries was found (Fig. 1). In addition, concentrations of BES were maintained during the experiments (data not shown), and adding BES after CH4 production was steady was less effective in increasing nitrogenase activity than adding BES to the fresh soil, although the final BES concentrations were the same (Figs. 4A and 6A). Together these findings indicate little possibility that BES acts directly on PB. Instead, noteworthy in this study are the accumulations of low molecular mass fatty acids, mainly acetate (Fig. 5). While little acetate was detected in the slurries incubated without BES, except in the first week, marked accumulation of acetate was observed on incubation in dark with BES. Acetate is easily produced by decomposing rice straw in paddy fields [37,42–45], and methanogens in rice fields produce CH4 mainly from the fermenting of acetate to CH4 and CO2[35,46–48]. Many species of PB grow photoheterotrophically using acetate under anaerobic conditions and in light [33]. Therefore, the accumulation of acetate can be explained by reduced acetate consumption due to suppression of methanogens and PB and constant acetate production by straw-decomposing microorganisms. In the slurries incubated in light with BES, only a little acetate accumulation was found, which is evidence for consumption of acetate by PB. These differences in acetate concentration among the treatments allow us to hypothesize that phototrophic ARA increased by BES was due to increased availability of acetate after stopping methanogenic consumption. Accepting that the supply of energy from photosynthesis regulates the expression of nitrogenase activity in PB [49,50], photosynthesis increased by acetate may fulfill the high energy demand of the nitrogen fixation process of PB. From the above hypothesis, the relatively low ARA in the slurries that received BES after CH4 production was steady can be regarded as a result of less substrate (Fig. 6). Besides acetate, propionate accumulated in the BES-added slurries when incubated in dark (Fig. 5B). In an Italian paddy soil, accumulations of other fermentation products of rice straw such as butyrate, isobutyrate, and caproate were found in the absence of methanogenic activity [37,45]. Even though concentrations of these intermediates for methanogenesis are inferred to be low [37,45], they as well as acetate may contribute to increasing nitrogenase activity, as PB can utilize diverse organic substrates for photoassimilation [33]. However, H2, which also accumulates when methanogenesis is inhibited [37,45], is assumed to be less important in this competition, as photoreduction of H2 and CO2 by PB initiates only after the disappearance of external organic substrates [51], which decomposition of rice straw may continuously provide in the slurries.

Consequently, our results show a competitive relationship between PB and methanogens in surface soils of rice fields amended with rice straw. The facts that light exposure had a suppressive effect on CH4 production (Fig. 2) may indicate outcompetition of PB with methanogens in photic zones. As population sizes of PB were almost the same in the soil slurries regardless of the addition of BES (Fig. 1), the growth of PB is not affected by the presence of methanogenesis. However, after PB grow enough, their nitrogenase activity may be limited by substrate concentrations as a result of supply and consumption by other microorganisms. As decreasing methanogenic activity due to increasing BES concentration led to increasing concentrations of low molecular mass fatty acids and a stimulated ARA (Figs. 4A, 5, and 6), methanogenesis may profoundly influence the resultant availability of substrates necessary for PB to perform N2 fixation.

Acknowledgements

We thank Dr. Hidenori Wada for his valuable comments on this work. We thank the members of the Saitama Prefecture Agriculture and Forestry Research Center for providing us with soils and rice straw.

Ancillary