Effective cellulose degradation by a mixed-culture system composed of a cellulolytic Clostridium and aerobic non-cellulolytic bacteria

Abstract

A stable cellulose-degrading microflora enriched from composting materials has been analyzed in our laboratory. Cellulose-degrading efficiency of an anaerobic cellulolytic isolate, Clostridium straminisolvens CSK1, was remarkably lower than that of the original microflora. We successfully constructed bacterial communities with effective cellulose degradation by mixing C. straminisolvens CSK1 with aerobic non-cellulolytic bacteria isolated from the original microflora. Comparison of the cellulose degradation processes of the pure culture of C. straminisolvens CSK1 and the mixed-culture indicated that non-cellulolytic bacteria essentially contribute to cellulose degradation by supplying anaerobic environment, consuming metabolites, which otherwise deteriorate the cellulolytic activity, and by neutralizing pH.

Introduction

In nature, many species of microorganisms coexist by interacting with each other. Many species of microorganisms are most effective only when they are present in association with other groups of organisms. For example, this cooperative activity is seen in interspecies hydrogen transfer within methanogenic systems [1], and in the degradation of xenobiotics such as halogenated organic compounds [2].

Likewise, in nature, lignocellulose is degraded with the cooperation of many species of microorganisms. Lignocellulose is one of the earth's most abundant, renewable resources and its degradation and utilization by microorganisms is considered very important [3]. However, utilizing isolated microorganisms and pure enzymes to process natural lignocellulose without pretreatment and/or sterilization is difficult. On the other hand, it has been reported that a mixed-culture of a cellulolytic microorganism with a non-cellulolytic microorganism (e.g., hydrogen consumer [4,5], carbohydrate consumer [6–9]) is ideal for degrading cellulose.

In our laboratory, a stable microflora (designated as “original microflora”) capable of effectively degrading various cellulosic materials (e.g., filter paper, cotton and rice straw) under aerobic static conditions was constructed by succession of enrichment cultures as reported by Haruta et al. [10]. The original microflora degraded natural cellulosic materials without sterilization. Furthermore, the original microflora had a high stability; the cellulose-degrading efficiency and the composition of bacteria did not change after more than 20 subcultures. Denaturing gradient gel electrophoresis (DGGE) and 16S rRNA gene sequence analyses indicated the coexistence of aerobic and anaerobic bacteria in the original microflora.

An anaerobic, cellulose-degrading bacterium, Clostridium straminisolvens CSK1 was successfully isolated from the original microflora [11]. Although C. straminisolvens CSK1 had cellulose-degrading capability under anaerobic conditions, it did not grow under the conditions used for the original microflora (i.e., aerobic static conditions). Anaerobic cellulolytic Clostridia and aerobic bacteria are often simultaneously detected at various sites where cellulose degradation occurs, such as rice paddy soils [12]. As the coexistence of anaerobic and aerobic bacteria is assumed to be important for effective cellulose degradation, comprehension of the mechanism of effective cellulose degradation by the original microflora is valuable for understanding the process of cellulose degradation in nature.

Although the original microflora is a good model for studying cellulose degradation by cooperation of microorganisms in terms of the effectiveness and stability, there is an experimental pitfall; not all members of the microflora can be completely defined. Hence, the reconstruction of a stable community with isolated bacteria is the first step in clarifying the role of each bacterium and their interactions.

In this paper, we describe the reconstruction of effective cellulose degradation under aerobic conditions by mixed-culture systems consisting of C. straminisolvens CSK1 and aerobic bacteria isolated from the original microflora. The mixed-culture systems were analyzed for degradation efficiency, physicochemical conditions, concentration of metabolites, and relative abundance of each member during cellulose degradation processes. Furthermore, the functional and structural stability of the mixed-culture system was evaluated.

Materials and methods

Maintenance of the original microflora

The original microflora was constructed by successive enrichment cultures using composting materials as sources of microorganisms [10]. The original microflora was cultivated in PCS medium containing (l⁻¹): yeast extract, 1 g; peptone, 5 g; CaCO₃, 2 g; NaCl, 5 g; rice straw, 10 g (pH 8.0), under static conditions with a loose cap at 50 °C. The culture solution was used as isolation sources.

Isolation of bacteria from the original microflora

Isolation was performed using Tryptic Soy agar (3% [w/v] Tryptic soy broth, Difco, containing 1.5% [w/v] agar) at 50 °C under air. Several kinds of colonies were transferred onto the same medium. After more than three successive subcultures, the isolates were examined for purity, anaerobic growth and cellulose-utilizing capability, and phylogenetic analyses were performed.

Anaerobic growth capability of aerobic isolates was determined using VLG agar medium containing (l⁻¹): yeast extract, 5 g; peptone, 10 g; meat extract, 2 g; glucose, 10 g; NaCl, 5 g; l-cysteine–HCl–H₂O 0.5 g; agar, 1.5 g (pH 7.0), under anaerobic conditions using an AnaeroPack® pouch bag with oxygen absorber, AnaeroPack® (Mitsubishi Gas Chemical, Tokyo) at 50 °C. Cellulose utilization capability of the aerobic isolates under aerobic and anaerobic conditions were checked using PCS-FP medium (PCS medium with 1% [w/v] of ADVANTEC No.5 filter paper [Advantec, Tokyo] instead of rice straw), and β-1,4 endoglucanase activity was checked by the carboxymetyl cellulose agar method described by Hankin and Anagnostakis [13]. The aerobic bacteria were maintained with PCS medium without carbohydrate.

Phylogenetic analysis

The 16S rRNA gene sequences of isolated bacteria were determined by direct sequencing of the purified PCR-amplified 16S rDNA fragment. Bacterial cells used for DNA extraction were cultivated in PCS medium with no carbohydrate for approximately 16 h with shaking. Genomic DNA was extracted by the benzyl chloride method [14] and was used as the PCR template. PCR was performed with universal bacterial primers complementary to conserved regions of the 5′ and 3′ ends of the 16S rRNA gene, 27F (forward) 5′-AGAGTT TGATCCTGGCTCAG-3′ (positions 8–27 in Escherichia coli numbering) and 1512R (reverse) 5′-ACGGCTACCTTGTTACGACT-3′ (positions 1512–1493 in E. coli numbering) [15]. PCR was performed using AmpliTaqGold® (Applied Biosystems). After initial denaturation for 10 min at 95 °C, target DNA was amplified in 30 cycles. Each cycle consisted of denaturation for 30 s at 93 °C, annealing for 30 s at 55 °C and extension for 2 min at 72 °C. The final extension was 5 min at 72 °C. The PCR products were purified with QIAquick® PCR purification kit (Qiagen) according to the manufactures' instruction. The purified 16S rDNA was sequenced directly using the ABI PRISM® BigDye® Terminator Cycle Sequencing Ready Reaction Kit (Applied Biosystems) and an ABI PRISM model 377 genetic analyzer (Applied Biosystems). The obtained 16S rDNA sequences of isolated bacteria were compared with those from the DDBJ nucleotide sequence database, using the program BLAST.

Maintenance of an anaerobic bacterium

Clostridium straminisolvens CSK1 (IAM 15070T = DSM 16021T) was isolated from the original microflora [11] and used for mixed-culture experiments. Unless stated otherwise, stringent anaerobic procedures [16] were used for cultivation of C. straminisolvens CSK1. An anaerobic medium containing a reducing agent (0.25 g l⁻¹ of glutathione [reduced]) was pre-reduced by cooling after sterilization inside an anaerobic chamber operated with anaerobic gas mixture (10% [v/v] H₂; 90% [v/v] N₂). Only fully reduced media, as indicated by the colorless state of resazurin in the media, were used for cultivation. C. straminisolvens CSK1 was maintained in 4 ml of medium 122 [11,17] containing 0.5% (w/v) ball-milled cellulose (Avicel®; Merck, Darmstadt, Germany) [11] as a substrate in 10 ml vial with a butyl rubber cap under N₂ 100% atmosphere without shaking at 50 °C.

Cultivation of mixed-culture systems

Aerobic isolates used for mixed-culture experiments were cultivated in PCS medium with no carbohydrate for approximately 20 h without shaking. C. straminisolvens CSK1 used for mixed-culture experiments was cultivated in medium 122 with ball-milled cellulose for 4–5 days.

PCS-FP medium in test tubes (16 by 100 mm for 4.0 ml scale or 40 by 130 mm for 70.0 ml scale) was used for mixed-culture experiments. Forty microliters of each pre-culture solution was inoculated into 4 ml of PCS-FP medium in various combinations (1st generations). These were incubated under static condition with a loose cap at 50 °C. Filter paper degradation capability of each mixed-culture was checked after 4–8 days cultivation. When the filter paper in the medium started to degrade, 40 or 700 μl of culture solution was transferred to 4.0 or 70.0 ml of the same medium (2nd generation).

Analyses of the processes during filter paper degradation

For the analyses of filter paper degradation, metabolites and bacterial community structures, test tubes filled with 70.0 ml of PCS medium or medium 122 with 1% (w/v) filter paper were used. For anaerobic cultures, the media were supplemented with 0.25 g l⁻¹ of glutathione (reduced) and cultivated in a pouch bag with oxygen absorber. Unless stated otherwise, the 3rd generation of mixed-cultures was used for analyses. For each bacterial combination and culture condition, two identical cultures were subjected to all analyses. All analyses were performed in more than duplicate for each sample. Residual filter paper was gravimetrically determined using a method described by Taillisz et al. [18]. Uninoculated medium was used as a control. The oxidative-reductive potential (ORP value) in the culture solution was determined using ACT pH meter D24 with ORP electrode 9300-10D (Horiba, Kyoto, Japan). The soluble cellooligosaccharides concentration in the culture solution was determined colorimetrically by an anthrone method [19]. The acetate concentration in the culture solution was determined enzymatically by F-kit (Boehringer–Mannheim, Mannheim, Germany). The cell number in the culture solution (only the planktonic population, not containing adherent cells) was determined by direct counting of 4′,6-diamidino-2-phenylindole (DAPI)-stained cells under a fluorescence microscope, using the method described by Haruta et al. [20].

Quantitative PCR

From the 16S rDNA sequences of the isolates (C. straminisolvens CSK1, Pseudoxanthomonas sp. M1-3, Brevibacillus sp. M1–5 and Bordetella sp. M1–6), specific primers were designed using the program PRIMROSE [21] according to the sequence database of ribosomal database project II (RDP II). The designed primers were checked with the PROBE_MATCH tool of the RDP II and are listed in Table 1.

The genomic DNAs extracted from the mixed-culture samples were used as templates for the quantitative PCR. For the standard curves, the genomic DNAs extracted from each isolate cultivated in pure culture were used as templates. The concentration of DNA was determined spectrophotometrically.

The LightCycler (Roche Diagnostics) was used with LightCycler-FastStart DNA Master SYBR Green I (Roche Diagnostics). Each reaction mixture was prepared as follows: LightCycler-FastStart DNA Master SYBR Green I, 2 μl; 25 mM MgCl₂, 2.4 μl; forward and reverse specific primers (10 pmol μl⁻¹), 1 μl each; and PCR grade distilled water to give 18 μl. Finally, 2 μl of DNA solution was added.

Real-time PCR was started with an initial denaturation at 95 °C for 10 min. Subsequently, target DNA was amplified in 45 cycles. Each cycle consisted of denaturation for 5 s at 95 °C, annealing for 6 s at 68 °C (for C. straminisolvens CSK1 and Brevibacillus sp. M1–5) or 10 s at 61 °C (for Pseudoxanthomonas sp. M1–3 and Bordetella sp. M1–6) and extension for 20 s at 72 °C. Fluorescence was detected at the end of extension reaction. The temperature transition rate was 20 °C s⁻¹ except for annealing temperature to extension temperature, where it was 2 °C s⁻¹. The specificity of the amplified PCR product was assessed by performing a melting curve analysis, which consisted of denaturation at 95 °C, annealing for 15 s at 70 °C and gradual denaturation with 0.1 °C s⁻¹ of the temperature transition rate until 95 °C with continuous detection of fluorescence. Two trials were conducted to analyze each sample.

Results

Isolation of aerobic bacteria from the original microflora

Eight strains of aerobic bacteria were isolated from the culture solution of the original microflora. From the 16S rRNA gene sequences analyses, they were clustered into five groups (Table 2). 16S rDNA sequences of these isolates were deposited in DDBJ database under the Accession nos. AB039328–AB039331 and AB039334–AB039336. One strain from each group (Bacillus sp. M1-1, Pseudoxanthomonas sp. M1-3, Virgibacillus sp. M1-4, Brevibacillus sp. M1–5 and Bordetella sp. M1–6) was used for subsequent experiments. Two isolated strains (Bacillus sp. M1–1 and Pseudoxanthomonas sp. M1–3) were able to grow anaerobically whereas three strains (Virgibacillus sp. M1-4, Brevibacillus sp. M1–5 and Bordetella sp. M1–6) did not. None of the aerobic isolates had cellulose-utilizing capability or β-1,4 endoglucanase activity under both aerobic and anaerobic conditions.

Cellulose degradation by cellulolytic C. straminisolvens CSK1 together with non-cellulolytic aerobic isolates

In a mixed-culture consisting of C. straminisolvens CSK1 and all of the aerobic isolates, cellulose degradation occurred under aerobic static conditions, while it did not in the pure culture of C. straminisolvens CSK1 or a mixed-culture of the aerobic isolates only (Fig. 1). The growth of C. straminisolvens CSK1 in the mixed-culture with aerobic bacteria was confirmed by specific PCR. The results indicate that the aerobic isolates enable C. straminisolvens CSK1 to grow and degrade cellulose under the aerobic static conditions.

To clarify which aerobic isolates have potential for enabling C. straminisolvens CSK1 to grow and degrade cellulose in mixed-culture systems, filter paper degradation capability of two-strains-mixed-culture consisting of C. straminisolvens CSK1 and each aerobic isolate were investigated (Fig. 2). The mixed-culture of C. straminisolvens CSK1 with Pseudoxanthomonas sp. M1-3, Brevibacillus sp. M1–5 or Bordetella sp. M1–6 degraded filter paper, while the other mixed-culture did not.

Furthermore, a four-strains-mixed-culture consisting of C. straminisolvens CSK1, Pseudoxanthomonas sp. M1-3, Brevibacillus sp. M1–5 and Bordetella sp. M1–6 (namely CSK + M356) had a distinguished filter paper degradation capability (8.30 mg ml⁻¹, for 8 days). The degradation efficiency was almost the same as that of the original microflora (7.77 mg ml⁻¹, for 8 days). These three aerobic isolates seemed to have positive effects on cellulose degradation of the mixed-culture systems. However, under anaerobic condition, the filter paper degradation efficiency by CSK + M356 was remarkably low (1.93 mg ml⁻¹, for 8 days). This result indicates that metabolism of aerobic isolates under aerobic condition is crucial for effective degradation besides supplying anaerobic environment.

On the other hand, a mixed-culture of C. straminisolvens CSK1 with all aerobic isolates had a definitely lower filter paper degradation capability (6.64 mg ml⁻¹, for 8 days). A mixed-culture of CSK + M356 together with Bacillus sp. M1–1 (6.63 mg ml⁻¹, for 8 days) or Virgibacillus sp. M1–4 (6.77 mg ml⁻¹, for 8 days) had a relatively lower filter paper degradation rate, indicating that these aerobic bacteria (Bacillus sp. M1–1 and Virgibacillus sp. M1–4) do not have positive effects for cellulose degradation in the mixed-culture systems.

Comparison between filter paper degradation processes of pure culture of C. straminisolvens CSK1 and of the mixed-culture (CSK + M356)

To clarify how the aerobic isolates assist C. straminisolvens CSK1 in degrading cellulose so effectively, the filter paper degradation processes of C. straminisolvens CSK1 pure culture and that of the CSK + M356 were compared. The filter paper degradation and the pH profile of strain CSK1 pure culture cultivated in PCS-FP medium under anaerobic conditions are shown in Fig. 3. The amount of filter paper degradation was remarkably lower than that of the pure culture cultivated in medium 122 (Fig. 4(a)). Hence, medium 122 was used for analyses of the filter paper degradation processes of C. straminisolvens CSK1 pure culture.

The transitions of filter paper degradation amounts of the pure- and mixed-culture are shown in Fig. 4(a). Although there is little difference between the pure- and mixed-culture in the filter paper degradation amounts in the initial phase (e.g., day 2 to 3, 3.76 and 3.44 mg ml⁻¹ day⁻¹, respectively), there is an obvious difference in the later phase (e.g., day 4 to 8, 1.87 mg ml⁻¹ [4 days]⁻¹ and 3.26 mg ml⁻¹ [4 days]⁻¹, respectively). After 10 days cultivation, 6.50 mg ml⁻¹ of filter paper was degraded by C. straminisolvens CSK1 pure culture and 8.91 mg ml⁻¹ by the CSK + M356.

The ORP value in the culture solution of CSK + M356 decreased to under −300 mV within 1 day of incubation, and was retained under −300 mV throughout the cultivation, indicating that aerobic bacteria were consuming oxygen, and supplying an anaerobic environment. The transitions of pH values in culture solutions of the pure- and mixed-culture were measured (Fig. 4(b)). Until day 3, at which vigorous filter paper degradation was observed, the pH value of each culture solution dropped to below 6. In the latter phases, however, the pH value of the mixed-culture system returned to neutral (around 7.2) while that of the pure-culture was constant at around 5.6.

The soluble cellooligosaccharides concentration in each culture solution was determined (Fig. 4(c)). In the initial phase (until day 4), the increase of the soluble cellooligosaccharides concentrations appeared to correspond to the cellulose degradation amounts in each system. In the latter phase, however, the concentration in the mixed-culture system was obviously lower than that in the pure-culture system while the cellulose degradation amount was higher.

The acetate concentration in each culture solution was determined (Fig. 4(d)). The concentration in the pure-culture system increased sharply to 1127 mg l⁻¹ during days 2–4, and then gradually increased to 1467 mg l⁻¹ by day 10. On the other hand, the concentration in the mixed-culture system increased to 674 mg l⁻¹ during days 1–3, and then gradually decreased to 474 mg l⁻¹ by day 10.

Transition of bacterial composition of the mixed-culture

The bacterial cell numbers in culture solution of CSK + M356 (3rd generation) at days 1, 2, 4 and 6 were determined by direct counting with DAPI staining. The cell numbers vigorously increased (more than 10 times) during day 1 (1.2 ± 0.3 × 10⁸ cells ml⁻¹) to day 2 (1.9 ± 0.4 × 10⁹ cells ml⁻¹). Thereafter the cell numbers scarcely changed up to day 6 (1.8 ± 0.3 × 10⁹ cells ml⁻¹ at day 4 and 9.2 ± 2.5 × 10⁸ cells ml⁻¹ at day 6).

The transition of relative abundance of each bacterium during the filter paper degradation by CSK + M356 was determined by real-time PCR (Fig. 5(a)). C. straminisolvens CSK1 was indicated to exist in low abundance (less than 1% DNA mass) at day 1, when filter paper degradation scarcely occurred, and then it increased to around 20% with accompanying filter paper degradation by day 2. Pseudoxanthomonas sp. M1-3, which is a facultative anaerobe, was present in high abundance (more than 73%) throughout the period. Brevibacillus sp. M1-5, which is considered a strict aerobe, was indicated to exist more than 10% at day 1, and then it decreased (around 1–7% during days 2–6). Bordetella sp. M1-6, which is also considered a strict aerobe, was indicated to exist in low abundance (less than 1%) at day 1, and then it increased (around 1–8% during days 2–6).

Functional and structural stability of the mixed-culture

The filter paper degradation capability and the transition of pH value throughout the cultivation of CSK + M356 in 12th and 41st generation were examined. The filter paper degradation amounts at days 4 and 8 of 12th generation were 5.02 and 8.70 mg ml⁻¹, respectively. In addition, those of 41st generation were 5.18 and 8.42 mg ml⁻¹. The transitions of pH value of 12th and 41st generation showed almost the same pattern as that of 3rd generation (data not shown).

In contrast, as for the structural stability, the mixed-culture was unstable. To clarify the stability of structure (the bacterial composition) of the mixed-culture CSK + M356, the bacterial communities of 12th and 37th generation were analyzed by real-time PCR (Figs. 5(b) and (c)). In the 37th generation, one of the mixed-culture members (Brevibacillus sp. M1–5) was not detected (<0.01%). It is obvious that the relative abundance of Brevibacillus sp. M1–5 gradually decreased with successive subcultures (3rd–12th generation).

Discussion

Eight strains (clustered into five groups) of aerobic bacteria were isolated from the original microflora. Facultative anaerobes, strain M1-2, -3, and -7 were clustered into γ-proteobacteria, Pseudoxanthomonas sp. Identical strain sequences were detected by denaturing gradient gel electrophoresis (DGGE) in the original microflora throughout the cultivation [10]. Pseudoxanthomonas taiwanensis, the closest relative of these strains, is classified as a thermophilic facultative anaerobe that was originally isolated from a hot spring [22]. Homologous sequences have been detected in water from a thermophilic wastewater treatment bioreactor operated at 55 °C [23]. A strict aerobe, strain M1–5 was closely related to Brevibacillus agri, which is gram-positive, strictly aerobic soil bacterium [24]. A strict aerobe, strain M1–6 was clustered into β-proteobacteria, genus Bordetella. Most species within the genus Bordetella are reported as strict aerobes and some strains are moderately thermophilic growing at 50 °C or above [25]. Strain M1–5 and -6 had an identical sequence for a DGGE band sequence derived from the original microflora [10]. Strain M1–1 and -8, and M1–4 were closely related to gram-positive bacteria, Bacillus licheniformis and Virgibacillus pantothenticus, respectively. The DGGE bands corresponding to these strains had not been detected from the original microflora.

It was revealed that some species with various oxygen requisiteness and susceptibility coexist steadily in the original microflora. This feature would be caused by the culture conditions, i.e., aerobic static conditions. While there would be a plentiful supply of oxygen from air in the upper phase (especially the liquid-air interface) of the culture system, the culture solution in the lower phase would be an anaerobic, reductive condition. Indeed, the ORP value of the culture solution during the cellulose degradation by CSK + M356 was under −300 mV, and that by the original microflora was reported as under −400 mV [10].

Although C. straminisolvens CSK1 did not grow under the conditions used for the original microflora, a mixed-culture with aerobic non-cellulolytic bacteria isolated from the original microflora enabled it to grow and degrade cellulose under the conditions (Fig. 1). These results suggest that coexistence of an anaerobic cellulolytic bacterium and aerobic non-cellulolytic bacteria is crucial for cellulose degradation by the original microflora. However, not all bacteria had positive effects on cellulose degradation in the mixed-culture systems, specifically, Bacillus sp. M1–1 and Virgibacillus sp. M1-4.

The four-strains-mixed-culture consisting of C. straminisolvens CSK1 and the three strains of aerobic isolates, CSK + M356, had high cellulose degradation capability compared to that of the original microflora. There are some reports concerning improvement of cellulose degradation efficiency by pairing a cellulolytic Clostridium with a non-cellulolytic microorganism that scavenge metabolites from cellulose (e.g., Clostridium thermocellum with Methanobacterium thermoautotrophicus [formerly Methanobacterium thermoautotrophicum] [4], C. thermocellum with Thermoanaerobacter themohydrosulfuricus [formerly Clostridium themohydrosulfuricum] [6]). All of these systems are, however, two-strains-mixed-culture, all microbial species involved is anaerobic, and the systems are cultivated under anaerobic conditions. Veal and Lynch [8] have described an enhancement of cellulose degradation with coculture consisting of the aerobic cellulolytic fungus Tricoderma harzianum and the anaerobic bacterium Clostridium butyricum. On the other hand, CSK + M356 consists of cellulolytic Clostridium and 3 strains of aerobic bacteria. This is a novel finding for effective cellulose degradation by microorganisms. Furthermore, anaerobic cellulolytic Clostridia and aerobic bacteria are often simultaneously detected at various sites where cellulose degradation occurs, such as rice paddy soils [12]. Our bacterial community is a good model for understanding the relationships between cellulolytic Clostridia and aerobic non-cellulolytic bacteria.

Comparison of the cellulose degradation processes of C. straminisolvens CSK1 pure culture and CSK + M356 indicated that the factors mentioned below are essential for effective cellulose degradation in the mixed-culture system. First, the aerobic bacteria supply anaerobic environment, which is an essential condition for growth of C. straminisolvens CSK1. The aerobic bacteria would consume oxygen by utilizing substrates contained in yeast extract and peptone, such as peptides and amino acids. There are other factors involved in effective degradation of cellulose, considering CSK + M356 degraded a smaller amount of filter paper under anaerobic conditions (Fig. 2). Second, the aerobic isolates scavenge metabolites derived from cellulose, which otherwise deteriorate cellulolytic activity. Water-soluble cellooligosaccharides, especially cellobiose, are known to repress cellulose degradation by cellulolytic Clostridia [26]. The addition of the aerobic bacteria reduced the concentration of cellooligosaccharides in the culture solution (Fig. 3(c)). Third, the aerobic isolates neutralize the pH of the culture solution. It had been shown that the optimum initial pH for growth and cellulose degradation of C. straminisolvens CSK1 was 7.5, and little cellulose degradation occurred under pH 6.0 [11]. During cellulose degradation by CSK + M356, although the pH value dropped to below 6, it returned to and remained around 7 (Fig. 3(b)). Although it is not accurately clear how the aerobic bacteria neutralize the pH value, acetic acid consumption (Fig. 3(d)) would be one of the factors.

In addition to the three functions mentioned above, another possible function for strengthening the cellulose degradation in the mixed-cultures might be stimulation of growth of other species by excretion of low-molecular weight compounds. In fact, addition of boiled supernatant from the culture solution of Pseudoxanthomonas sp. M1–3 reinforced the growth and cellulose degradation of C. straminisolvens CSK1 pure culture (data not shown).

As expected, C. straminisolvens CSK1 vigorously grew in CSK + M356 cultivation during day 1 (0.18%) to day 2 (23%) in an anaerobic culture solution, and filter paper degradation corresponded with the growth. C. straminisolvens CSK1 was not the dominant bacterium present (23% at day 2 and 11% at day 4) during the period of vigorous cellulose degradation. The relative abundance was nearly the same as in the original microflora (13% at day 3 and 3% at day 7). Taking into consideration that the abundance of cellulolytic bacteria in bovine rumen was less than 10% as calculated by competitive PCR [27], the primary degrader of cellulose in microbial communities might generally not be the most abundant species.

The bacterium with highest abundance in CSK + M356 throughout the cultivation was Pseudoxanthomonas sp. M1–3 (73–84%). This bacterium existed in high abundance at day 1 before the onset of cellulose degradation, indicating that it would grow without substrates derived from cellulose degradation such as soluble carbohydrates. Pseudoxanthomonas sp. M1–3 is considered to utilize peptides or amino acids contained in the liquid medium under aerobic condition. Using this metabolic pathway, Pseudoxanthomonas sp. M1–3 would consume oxygen and produce ammonia, which would cause the ORP value of the culture solution to drop and the pH value to increase. Moreover, Pseudoxanthomonas sp. M1–3 was also the dominant bacterium through the latter phase of cultivation when the culture solution was anaerobic. Furthermore, Pseudoxanthomonas sp. M1–3 was detected as the dominant bacterium not only from the upper phase of culture solution (considered to be aerobic condition) but also from lower phase (data not shown). These results indicate that Pseudoxanthomonas sp. M1–3 simultaneously lives in either anaerobic or aerobic environments. It is very fascinating that the bacterium selects its metabolism by responding to the environment in a test tube.

The function of the community, CSK + M356, as well as the original microflora, was stable in subculture as the cellulose degradation efficiency and pH transition pattern did not change after successive subcultures. However, the structure of the mixed-culture was not stable; one of the members (Brevibacillus sp. M1–5) was washed-out through the successive subcultures (Fig. 5). On the other hand, in the original microflora, Brevibacillus sp. M1–5 survived after more than 40 subcultures [10]. Although the reason for this result is not yet clear, one possibility is that the mixed-culture lacks a particular bacterium existing in the original microflora, which has positive effects on the growth of Brevibacillus sp. M1-5. Although there are some reports about the stability of microbial communities [28,29], the mechanism(s) have not been clarified yet. In the ecological field, computer modelling indicated that stability of ecosystems is derived from various interactions among many kinds of species [30]. Likewise, the functional and structural stability of the original microflora would be derived from various relationships amongst many species of bacteria. To clarify the role of each bacterium and their interactions relating to the stability of the community, construction of a stable mixed-culture system of CSK + M356 with appropriate modifications would be required.

In nature, many microorganisms coexist in an environment. There are various types of microbial interactions known, such as antagonism, competition and symbiosis. Throughout the long history of microbiology, most researchers were interested in the “negative interactions” amongst microorganisms such as antibiotic production and the scramble for substrates. In contrast, microflora used in this study is a community that was selected from natural environment by the succession of subcultures, whose members would have chiefly positive interactions with each other. We consider this microbial community to be a good model for studying such positive interactions that universally exist in nature but have not been clarified yet.

Acknowledgements

This work was partially supported by a Grant-in-Aid (no. 14206011) for scientific research from the Ministry of Education, Science and Culture of Japan.

References