Biotechnology Letters

, Volume 31, Issue 9, pp 1321–1326 | Cite as

Bio-hydrogen production from cellulose by sequential co-culture of cellulosic hydrogen bacteria of Enterococcus gallinarum G1 and Ethanoigenens harbinense B49

  • Aijie Wang
  • Lingfang Gao
  • Nanqi Ren
  • Jifei Xu
  • Chong Liu
Original Research Paper


Microbial conversion of lignocellulose to hydrogen is a fascinating way to provide a renewable energy source. A mesophilic bacterium strain G1 that had high cellulose degradation and hydrogen production activity (2.38 mmol H2 g−1 cellulose) was isolated from rumen fluid and identified as the Enterococcus gallinarum. Hydrogen production from cellulose by using sequential co-cultures of a cellulosic-hydrolysis bacterium G1 and Ethanoigenens harbinense B49 was investigated. With an initial Avicel concentration of 5 g l−l, the sequential co-culture with G1 and strain Ethanoigenens harbinense B49 produced H2 yield approximately 2.97 mmol H2 g−1 cellulose for the co-culture system.


Hydrogen Isolation lignocellulose Sequential co-culture 


Cellulosic materials mainly consist of hexose and pentose sugars with potential use for the production of fuel (Kuhad and Singh 1993; Chandel et al. 2007; Herrera, 2004). Recently, hydrogen production from cellulose has received more and more attention. In nature, cellulosic materials are degraded by mixed microbial consortia consisting of cellulose-hydrolytic, fermentative and other microorganisms depending on the substrate. For instance, rumen fluid cultures comprising cellulolytic and non-cellulolytic bacteria are ideal for cellulose degradation (Odom and Wall 1983; Lewis et al. 1988). However, most high efficient hydrogen-producing organisms are non (or low)-cellulose-degrading fermentative bacteria. A combination of high-active cellulose-hydrolyzing bacteria and hydrogen-producing bacteria could result in synergistic hydrogen production. Lay (2001) investigated the potential of producing H2 from microcrystalline cellulose under mesophilic digestion condition with heat shocked sludge as inoculum. Lo et al. (2008) determined that mixed culture of cellulosic-hydrolysis sludge and Clostridium pasteurianum produced H2 1.09 mmol H2/g cellulose from 10 g l−1 carboxymethyl cellulose.

In this study, we isolated a mesophilic bacterium from rumen fluid and identified it as Enterococcus gallinarum G1. The growth and hydrogen production from cellulose by strain G1 and its companion bacterium, a hexose-hydrogen producing Ethanoigenens harbinense B49 were investigated. We also investigated the synergistic effect of cellulose degradation and hydrogen production that occurred during sequential co-culture of these two strains.

Materials and methods

Medium and isolation of bacteria

Rumen fluid obtained from a slaughter-house in Harbin, China was used as the source for microbial isolation. The fluid was filtered through four-layers of gauze and stored in a vial purged with N2 gas prior to use. The rumen fluid (15 ml) was then inoculated into anaerobic flasks (Bellco Glass, USA) with sterilized modified 1191 Clostridium thermocellum medium (ATCC medium 1190) to enrich cellulose-degrading microorganisms under N2 atmosphere. The medium used 5 g l−1 Avicel as carbon source. The Avicel had cellulose content at 97.2% (v/v, dry basis) and a water solubility of 0.1% (w/v). The mixtures were incubated at 37°C for 100 h. Suspension samples collected from the incubated tubes were plated on agar plates with the same medium mentioned above for the isolation of cellulose-degrading and hydrogen-production bacteria.

DNA extraction and sequencing

The identification by 16S rDNA gene sequence analysis was carried out as follows. The genomic DNA was extracted from cells with the standard method (Sambrook and Russell 2001) and the 16S rDNA gene was amplified by PCR as described (Wang et al. 2008).The double-stranded PCR products were sequenced. The 16S rDNA sequence was aligned and identified best matches of the genes against existing DNA sequences in GenBank database using the BLAST program.

Co-culture test

Co-culture tests were performed using the same medium as that in the cultivation tests. Briefly, medium (150 ml) was mixed with 15 ml inoculum (0.016 g dry cells) and was kept at 37°C for 100 h. The isolated strain that had the highest hydrogen production potential among those isolates from Avicel suspensions was tested further. Strain Ethanoigenens harbinense B49 (AF481148 in NCBI), a H2-producing, fermentative bacteria isolated (Ren et al.2007) was utilized as a partner for co-culture with the isolated strain G1 in fermentation tests. Strain B49 does not hydrolyze cellulose, but produced hydrogen from glucose with a maximum hydrogen production rate of 25 mmol H2 h−1 g−1 dry cells and a hydrogen yield of 1,810 ml l−1 medium (Wang et al. 2008).

In the co-culture tests, G1 was cultured individually as either control or sequential co-culture with B49 at the same culture volumes at a total biomass quantity of 0.0184 g dry cells, namely B49 was inoculated into the same medium after 20 h that of G1 at 37°C, considering nearly 20 h gap of doubling time between G1 and B49.

Three replicates of culture tubes were used at each experimental sampling point. The time zero samples were collected immediately after inoculation and used as controls.

Analytical methods

The cell dry weight was measured based on centrifuging the filtered culture broths without cellulose (3,000 g, 10 min), washing twice with distilled water and drying at 105°C until constant. The gas composition was measured as described by Wang et al. (2008). The Avicel concentration in the medium was determined by phenol–H2SO4 method after removal of cell mass as described by Minato et al. (1962).

Results and discussions

Isolation and characterization of the cellulosic bacterial strains

Four strains G1, G2, G4, and G5, had high hydrogen yields from the enrichment of rumen fluid were screened and purified using anaerobic pressure technology at 37°C under anaerobic conditions (Table 1). Among these isolates, a bacterial strain G1 with the highest cellulose degradation ratio and hydrogen production activity was obtained. Several round and transparent circle-shaped white colonies were formed using agar roll tubes containing Avicel (Fig. 1a). The bacteria G1 grown in liquid medium or taken from colonies in agar roll tubes usually occurred as pair or chain-link spherical shape with 0.6–2.0 × 0.6–2.5 μm in diametric (Fig. 1b). The partial genome of 1,496 bp obtained from 16S rDNA amplification of strain G1 was isolated and aligned using NCBI Blast. This strain had 99% sequence similarity with Enterococcus gallinarum. Therefore, it is named as E. gallinarum strain G1. Physiological tests demonstrated that the optimal temperature and pH for strain G1 to degrade Avicel were 37°C and 6.5, respectively (data not shown). Doubling time of G1 was 13.6 h.
Table 1

Hydrogen yields, cellulose degradation ratio and metabolites from Avicel with different strains

Substrate 5 g l−1


CDR(%) (v/v)

Final pH

YH2 (ml L−1)

Metabolites (mg l−1)

































CDR, Cellulose degradation ratio

Fig. 1

The SEM image of Enterococcus gallinarum G1 (a). Cellulosic hydrolyzed circle caused by Enterococcus gallinarum G1 on Hungate tube (b)

Cellulose degradation and hydrogen production of G1 started after the 5 h lag phase. The entire fermentation process was completed in 55–60 h. The corresponding hydrogen yield, maximum hydrogen production rate and cellulose hydrolysis ratio reached 107.5 ml l−1 medium, 1.16 mmol H2 h−1 g−1 dry cell, and 42.6% (v/v), respectively. Hence, strain G1 has a good capability to hydrolyze Avicel and then convert it into hydrogen.

Growth and corresponding changes in pH during sequential co-culture

The efficiency of fermentative conversion of Avicel with cellulosic strain G1 and hydrogen-producing strain B49 into H2 was examined. Time courses of G1 mono-cultures and sequential co-cultures of G1 and B49 were shown in Fig. 2. In the mono-culture of G1, as cellulose was degraded gradually, the cell dry weight increased to 0.9 g l−1 while slightly decreased at the end of the fermentation process, corresponding to the increase in pH. It was estimated that autolysis reaction of cells due to the unfavorable growth condition contributes to this. Figure 2 also revealed that more cells (1.2 g l−1) were produced without cell autolysis phenomena when G1 and B49 were co-cultured. This indicated that there were more cell divisions in the co-culture since B49 improved cellulose hydrolysis and hydrogen production of co-culture by removing the reducing saccharides produced by G1. In the absence of B49, the final pH of the culture media from G1 monocultures was slightly higher at pH 6.0. The addition of B49 caused a decrease in the pH of the media to pH 3.5 toward the end of the culture period, meaning more organic acids were produced when G1 was co-cultured with B49.
Fig. 2

Time course of cell growth and corresponding pH for mono-culture of G1 and co-culture of G1 + B49 at 37°C and pH 6.5. CDW Cell dry weight

H2 production in sequential co-culture

Effects of co-culture of G1 and B49 on cumulative H2 yields were shown in Fig. 3. When G1 was cultured alone, specific hydrogen yield of 2.38 mmol H2 g−1 cellulose was observed after 55 h. However, hydrogen production was initiated following a 7 h lag phase after addition of B49 when the two strains were sequential co-cultured (Fig. 3). The entire fermentation process was completed in 50 h, while the co-culture fermentation process was only 25 h, generating a better hydrogen yield, hydrogen production rate, and a cellulose hydrolysis ratio of 182 ml l−1 medium, 1.68 mmol H2 h−1 g−1 dry cell, and 53.8% (v/v), respectively. The specific hydrogen production rate by strains G1 + B49 increase by 30.9% (v/v) when compared with the mono-culture test of G1. Hence, sequential co-culture with the two strains significantly improved cellulose hydrolysis and subsequent hydrogen production yield and shorten the hydrogen production peak period from 55 to 60 h (mono-culture of G1) to 45–50 h (co-culture of G1 + B49).
Fig. 3

Evolved hydrogen test for mono-culture G1 and co-culture G1 + B49 at 37°C and pH 6.5. Total dry biomass 0.015 g for G1 and 0.0184 g for G1 and B49. QH2, mmol H2 g−1 Avicel

From the above results, it was clear that G1 and B49 formed stable synergistic co-cultures system. When G1 was co-cultured with B49, B49 utilized partial reducing sugars as a carbon source to produce H2, organic acids and ethanol. The cellulases produced by the strain G1 continued to decompose cellulose as B49 removed the reduced saccharides quickly.


The metabolites of end liquid products of the mono-cultures of G1 and the sequential co-cultures of the two strains G1 + B49 were shown (Fig. 4a). The end liquid products of mono-culture test with G1 were primarily acetate, propionate and butyrate of 971.2, 623.5, and 450.9 mg l−1, respectively. When both G1 and B49 were sequential co-cultured in the same medium, more end liquid products were formed, the end liquid products were acetate, propionate and butyrate, and ethanol. This difference was attributed to the fact that strain B49 degrades mono-saccharides via an ethanol-type fermentation pathway (Ren et al.2007). After 50 h incubation, the ethanol, acetate, propionate, and butyrate were 255.5, 1515.4, 976.52, and 638.05 mg l−1. Reduced saccharides were peaked at 75 mg l−1 after 10 h of cultivation and kept fluctuation with the increase in cellulose hydrolysis ratio when G1 was cultured alone (Fig. 4b). When G1 was co-cultured with B49, reduced saccharides decreased to an undetectable level.
Fig. 4

Mono and co-culture tests with strains G1 and B49 of VFA (a) contents of reduced saccharides (b). Experimental conditions corresponding to those in Fig. 4

Conclusions and discussion

In this research, a mesophilic anaerobic bacterium named as Enterococcus gallinarum G1 was isolated to effectively decompose cellulose and produce hydrogen, At 37°C and pH 6.5, the mono-culture tests with G1 revealed that cellulose hydrolysis and hydrogen production occurred with a 5 h time lag. The cellulose was then promptly hydrolyzed and hydrogen was produced in the following 60 h. The corresponding hydrogen yield, maximum hydrogen production rate, and cellulose hydrolysis ratio reached 107.5 ml l−1 medium, 1.16 mmol H2 h−1 g−1 dry cell, and 42.6% (v/v), respectively. Equivalently, the hydrogen yield was 2.38 mmol H2 g−1 cellulose. The end liquid products were primarily acetate, propionate and butyrate.

Bio-hydrogen was directly produced from cellulose efficiently using sequential co-culture of G1 and B49. The hydrogen yield, maximum hydrogen production rate, and cellulose hydrolysis ratio of 182 ml l−1 medium, 1.68 mmol H2 h−1 g−1 dry cell, and 53.8% (v/v), respectively, were achieved. The equivalent hydrogen yield reached 2.97 mmol H2 g−1 cellulose. The corresponding end liquid products were acetate, ethanol, propionate and butyrate. In addition, the efficiency of G1 + B49 co-culture cellulosic H2 production system is comparable to that reported in the other studies (Table 2). The results indicated that co-culture of cellulosic degradation strain G1 and saccharides-based hydrogen-producing strains B49 presents a potential route in converting renewable biomass such as cellulose into hydrogen energy.
Table 2

Comparison of H2 production performance using cellulosic material as substrate under mesphilic condition


Culture type


Temperature (°C)

H2 yield (mmol H2 g−1 substrate)


X9 + B49


MC (10 g l−1)



Wang et al. (2007)


Stream-exploded corn stover (15 g l−1)


Wang et al. (2008)

Anaerobic digested sludge


MC (12.5 g l−1)



Lay (2001)

Sludge + Clostridium pasteurianum


CMC (10 g l−1)



Lo et al. (2008)

G1 + B49


Avicel (5 g l−1)



This study

MC Microcrystalline cellulose, CMC Carboxymethyl cellulose

Rumen microbes and its co-culture companion for cellulose degradation and hydrogen production activity have been demonstrated in this study. However, further study on the optimization of the co-culture system is required, such as the different cellulase systems in rumen microorganism. Understanding optimization parameters and cellulase systems can potentially lead to significant enhancement of hydrogen production from cellulosic biomass.



This project was supported by Natural Science Foundation of China (NSFC 50678049, 50878062, and by Program for New Century Excellent Talents in University (NECT-2005)).


  1. Chandel AK, Kapoor RK, Singh A, Kuhad RC (2007) Detoxification of sugarcane bagasse hydrolysate improves ethanol production by Candida shehatae NCIM 3501. Bioresour Technol 98:1947–1950PubMedCrossRefGoogle Scholar
  2. Herrera S (2004) Industrial biotechnology: a chance at redemption. Nat Biotechnol 22:671–675PubMedCrossRefGoogle Scholar
  3. Kuhad RC, Singh A (1993) Lignocellulose biotechnology current and future prospects. Crit Rev Biotechnol 13:151–172CrossRefGoogle Scholar
  4. Lay JJ (2001) Biohydrogen generation by mesophilic anaerobic fermentation of microcrystalline cellulose. Biotechnol Bioeng 74:281–287CrossRefGoogle Scholar
  5. Lewis SM, Montgomery L, Garleb KA, Berger LL, Fahey GC Jr (1988) Effects of alkaline hydrogen peroxide treatment on in vitro degradation of cellulosic substrates by mixed ruminal microorganisms and Bacteroides succinogenes S85. Appl Environ Microbiol 54:1163–1169PubMedGoogle Scholar
  6. Lo Y-C, Bai M-D, Chen W-M, Chang J-S (2008) Cellulosic hydrogen production with a sequencing bacterial hydrolysis and dark fermentation strategy. Bioresour Technol doi:10.1016/j.biortech.2008.03.004
  7. Mackie RI, Bryant MP (1994) Acetogenesis and the rumen: syntrophic relationships. In: Drake HL (ed) Acetogenesis. Chapman and Hall, Inc, New York, NY, pp 331–364Google Scholar
  8. Minato H, Endo A, Kouriyama H, Uemura T (1962) On the ecological mechanism of the rumen fermentation. Influence of the addition of cellulose preparations upon the rumen microbial activities. Nippon Nōgeikagaku Kaishi 36:101–106Google Scholar
  9. Odom JM, Wall JD (1983) Photoreduction of H2 from cellulose by an anaerobic bacterial coculture. Appl Environ Microbiol 45:1300–1395PubMedGoogle Scholar
  10. Ren N, Xing D, Rittman BE, Zhao L, Xie T, Zhao X (2007) Microbial community structure of ethanol type fermentation in bio-hydrogen production. Environ Microbiol 9:1112–1115PubMedCrossRefGoogle Scholar
  11. Sambrook J, Russell DW (2001) Molecular cloning: a laboratory manual, 3rd edn. Cold Spring Harbor Laboratory Press, New YorkGoogle Scholar
  12. Wang AJ, Ren NQ, Shi YQ, Lee DJ (2007) Bioaugmented hydrogen production from microcrystalline cellulose using co-culture—Clostridium acetobutylicum X9 and Ethanoigenens harbineses B49. Int J Hydrogen Energy 33(2):912–917CrossRefGoogle Scholar
  13. Wang AJ, Ren NQ, Gao L, Lee DJ (2008) Bioaugmented hydrogen production from carboxymethyl cellulose and partially delignified corn stalks using isolated cultures. Int J Hydrogen Energy 33(19):5250–5255CrossRefGoogle Scholar

Copyright information

© Springer Science+Business Media B.V. 2009

Authors and Affiliations

  • Aijie Wang
    • 1
  • Lingfang Gao
    • 2
  • Nanqi Ren
    • 1
  • Jifei Xu
    • 2
  • Chong Liu
    • 2
  1. 1.State Key Laboratory of Urban Water Resource and Environment (SKLUWRE, HIT)HarbinPeople’s Republic of China
  2. 2.School of Municipal and Environmental EngineeringHarbin Institute of TechnologyHarbinPeople’s Republic of China

Personalised recommendations