Bioresource Technology 157 (2014) 22–27

Contents lists available at ScienceDirect

Bioresource Technology

journal homepage: www.elsevier .com/locate /bior tech

Bacterial community structure in maximum volatile fatty acidsproduction from alginate in acidogenesis

http://dx.doi.org/10.1016/j.biortech.2014.01.0720960-8524/� 2014 Elsevier Ltd. All rights reserved.

⇑ Corresponding authors. Tel.: +82 51 629 6436; fax: +82 51 629 6429 (H.C. Woo).Tel.: +82 51 629 7646; fax: +82 51 629 6429 (M. Song).

E-mail addresses: woohc@pknu.ac.kr (H.C. Woo), songmk@pknu.ac.kr (M. Song).

Jiyun Seon a,b, Taeho Lee b, Seong Chan Lee c, Hong Duc Pham c, Hee Chul Woo c,⇑, Minkyung Song a,⇑a The Institute of Cleaner Production, Pukyong National University, 365 Sinseon-ro, Nam-gu, Busan 608-739, Republic of Koreab Department of Civil and Environmental Engineering, Pusan National University, 63 Busandaehak-ro, Geumjeong-gu, Busan 609-735, Republic of Koreac Department of Chemical Engineering, Pukyong National University, 365 Sinseon-ro, Nam-gu, Busan 608-739, Republic of Korea

h i g h l i g h t s

� The bacterial community structure was monitored in acidogenesis from alginate.� Bacteroides-related organism might be contributed to hydrolysis of alginate.� Clostridium spp. corresponded to alginate-fermenting were mainly detected.� Bacterial community shifts corresponded to VFA profiles was statistically verified.

a r t i c l e i n f o

Article history:Received 12 December 2013Received in revised form 15 January 2014Accepted 19 January 2014Available online 25 January 2014

Keywords:AlginateAcidogenesisBacterial communityVolatile fatty acids

a b s t r a c t

Alginate as biomass feedstock for bioconversion into volatile fatty acids (VFAs) is limited primarily by thelow solubility in water or little utilization as microbial substrate and yet unknown about the microbialcommunity structure for acidogenesis. The bacterial community structure was demonstrated thereflected changes in VFAs profiles in the maximized acidogenic process from alginate. Bacteroides- andClostridium-related microorganisms were suggested to be mainly responsible for the hydrolysis of algi-nate and VFAs production, respectively. And the bacterial community shifted corresponded to VFAs pro-ducing was statistically demonstrated. A number of features discussed in this research can stimulatefurther interests on bioconversion of alginate into anaerobic biofuels production.

� 2014 Elsevier Ltd. All rights reserved.

1. Introduction

Marine algae (including macroalgae and microalgae) are receiv-ing increasing attention as attractive renewable resources for bio-fuels production with many advantages over biomass from thecrop-based or cellulosic matters (Wei et al., 2013). In particular,macroalgae (i.e., seaweed) has the growth rate with high produc-tivity, greater potential for carbon dioxide fixation, and abundantcontent of carbohydrates which are easily converted to biofuels(Pham et al., 2013). Among the macroalgae, massive brown algaeare primarily composed of polysaccharides such as alginate, lamin-aran, fucoidan, mannitol, and trace cellulose (Chang et al., 2010).The main polysaccharide, alginate, accounts for up to 40% dry wt.in brown algae as a principal material of the cell wall (Dragetet al., 2005; Jung et al., 2013) and has a chemical structure that

consists of two uronic acids (i.e., manuronic acid and gulonic acid)each containing a carboxyl group (Yang et al., 2011). A high yield oforganic acids can be theoretically available by simple decomposi-tion of alginate and maintaining its carboxyl group structures (Aidaet al., 2012). However fermentative production of organic acidsfrom alginate is still challenge due to its low solubility in wateror little utilization as microbial substrate (Pham et al., 2013).

Volatile fatty acids (VFAs) including short-chain fatty acids fromC2 to C6 (i.e., acetate, propionate, and butyrate, etc.) are valuablechemical compounds which have diverse industrial applicationssuch as feedstock resources for biofuels (i.e., biohydrogen, biome-thane, and mixed alcohols, etc.) (Singhania et al., 2013). Recently,a study of VFAs production from alginate has been firstly reported(Pham et al., 2013). Although the study has been showed a poten-tial of VFAs production in anaerobic alginate fermentation, infor-mation is still lacking on the behavior of microbial communitiesfor better understanding of the process.

In this research, we performed to operate acidogenic processusing alginate and especially focused on the changes in bacterial

J. Seon et al. / Bioresource Technology 157 (2014) 22–27 23

communities as functions of performance and operating conditionsfor insight understanding of the process. Therefore, the aim of thisresearch was to investigate bacterial community structure, in rela-tion to changes in VFAs profiles, throughout an acidogenic processfrom alginate.

2. Methods

2.1. Microorganism

Anaerobically digested sludge was obtained from a municipalwastewater treatment plant in Busan, Korea. Acid pre-treatment(2 N HCl) was applied at 35 �C for 24 h in waterbath in order to en-hance the activity of acidogenic bacteria (Lee et al., 2009). Contin-uous acidogenic reactor as an inoculum system was operated at35 �C and 150 rpm in a 3 L LiFlus GX bioreactor (BioTron Inc., Pu-chon, Korea) with a working volume of 2 L and equipped with tem-perature and pH controller. Acidogenic microorganism wascultivated in feed medium with 20.0 g/L of glucose as a sole carbonsource and nutrient solution containing: 5.2 g/L NH4HCO3; 0.12 g/LK2HPO4; 0.1 g/L MgCl2�6H2O; 0.015 g/L MnSO4�6H2O; 0.025 g/LFeSO4�7H2O; 0.005 g/L CuSO4�5H2O; 0.0005 g/L CoCl2�5H2O; and6.7 g/L NaHCO3. The pH value was maintained at 5.5 with buffersolution of 5 N NaOH or 5 N HCl to enhance acidogenic bacteriaactivity and suppress methanogenic bacteria activity. The opera-tion of acidogenic reactor was continued in a fill and draw once aday and hydraulic retention time (HRT) was kept at 1 day. The aci-dogenic biomass was taken before feeding and the concentrationsof TVFAs and alcohols in the system were remained at 12–14 g/L.

2.2. Experimental set-up

Sodium alginate (80–120 mPa s, Wako Pure Chemical IndustriesLtd., Osaka, Japan) was dissolved in distilled water and autoclaved(121 �C for 15 min) then used as a microbial growth substrate.Alginate was the sole carbon source in the medium, which alsocontained NH4HCO3, 2.0 g/L; KH2PO4, 1.0 g/L; MgSO4�7H2O,0.01 g/L; NaCl, 0.001 g/L; Na2MoO4�2H2O, 0.001 g/L; CaCl2�2H2O,0.001 g/L; MnSO4�7H2O, 0.0015 g/L; and FeCl2�4H2O, 0.00388 g/Las nutrient additives. The initial pH was adjusted as required using5 N NaOH or 5 N HCl. The effluent including acidogenic biomasswas added to distilled water with volume ratio of 1:1 and centri-fuged at 1000 rpm for 3 min. This procedure was repeated threetimes for removal of residue in the effluent and then was used asseed culture (equivalent to 10% of working volume), resulting ina volatile suspended solids (VSS) concentration of approximately7500 mg/L as cell concentration. The acidogenic process usingthe alginate was operated at 35 �C and 120 rpm. Chloroform(CHCl3; 100 lM) was used as a methanogenic inhibitor from bothH2/CO2 and acetate. It also inhibited acetate consumption by sul-fate reducers (Hu and Chen, 2007; Pham et al., 2013).

The effects of changing alginate concentration (4–9 g/L) and pH(6.0–10.0) on the production of VFAs in acidogenic batch processwere evaluated used a 2 � 2 (Conc. � pH; two levels each) orthog-onal central composite cube (CCC) design. The optimum conditionwas estimated by response surface methodology (RSM) usingMinitab software (version 15.1.1.0, Minitab Inc., State College,Pennsylvania, USA). Changes in VFA profiles and in acidogenic bac-terial communities were investigated under the estimated opti-mum condition as validated points.

2.3. DNA extraction

Raw samples were collected from the reactor for optimum con-dition. Samples were centrifuged at 10,000�g for 1 min. The super-

natant was removed and resuspended in 1 mL of distilled water.Total genomic DNA in the suspension was immediately extractedusing PowerSoil™ DNA kit (Mo Bio Labs, Carlsbad, USA). PurifiedDNA was eluted with 100 lL of Tri-HCl buffer (pH 8.0) and storedat �80 �C for further analyses. DNA extraction of was performed induplicate.

2.4. PCR–DGGE analysis

The extracted DNA was amplified using a Mastercycler gradientautomated thermal cycler (Bio-Rad Laboratories Inc., Hercules,USA) with primer set consisting of a GC-clamp eub 341F (50-CGCCCG CCG CGC GCG GCG GGC GGG GCG GGG GCA CGG GGG GCCTAC GGG AGG CAG CAG-30) and eub 518R (50-ATT ACC GCG GCTGCT GG-30) for V3–V5 region of bacterial 16S rRNA gene at the fol-lowing program: an initial denaturation at 95 �C for 2 min; 30 cy-cles of denaturation (30 s at 95 �C), annealing (40 s at 55 �C),extension (30 s at 72 �C); and a final extension at 72 �C for10 min. The reactions were carried out in a 25 lL volume contain-ing 1 lL of template DNA, 0.25 lL of the forward and reverse prim-ers (10 pmol), 2.5 lL of 10� Taq buffer, 10 lL of 10 mM dNTP, and0.125 lL of DNA polymerase (Solgent Co., Seoul, Korea). Polymer-ase chain reaction (PCR) products were checked electrophoreticallyand were purified using a PCR purification kit (Bioneer, Alameda,USA).

The denaturing gel gradient electrophoresis (DGGE) was thenperformed with a DCode™ Universal Mutation Detection System(Bio-Rad Laboratories Inc., Hercules, CA) using a 6% polyacrylamidegel with denaturing gradient ranging from 20% to 55% denaturantin 0.5� TAE buffer. The electrophoresis was run at a constant volt-age of 150 V at 60 �C for 7 h. After the electrophoresis, the gel wasstained in ethidium bromide solution for 30 min, rinsed for 30 minin water and visualized by UV transillumination. For identificationof DGGE bands, each band was eluted into 40 lL of deionized anddistilled water and then mixture was incubated overnight at 4 �C toextract the DNA from DGGE bands. And then solution was used asthe template in the reamplification reaction using same primerwithout GC-lamp, the specific primers, eub 341F (50-TAC GGGAGG CAG CAG-30) and eub 518R (50-ATT ACC GCG GCT GCT GG-30).

PCR fragments were purified using a purification kit (SolgentCo., Seoul, Korea). Sequencing was performed using an ABI3730XL capillary DNA sequencer (Applied Biosystems, Foster City,USA) and an ABI Prism Bigdye terminator cycle sequencing readyreaction kit (version 3.1, Applied Biosystems, Foster City, USA).The obtained sequences were finally compared with the related se-quences in the Genbank DNA database using BLAST program in theNational Center for Biotechnology Information (NCBI). Sequenceswere aligned with the closest 16S rRNA gene found in GenbankDatabase with Clustal X2 software (version 2.1, www.clustal.org)and the phylogenetic tree was constructed using the neighbor-joining method with MEGA 5.2 software package (version 5.2,www.megasoftware.net). Bootstrap analysis with 500 replicateswas performed to estimate the confidence of the tree topologies.Statistical analyses were performed using the Statistical Packagefor Social Sciences, fingerprinting II informatix software (version3.0 for windows, Bio-Rad Laboratories Inc., Hercules, USA) to ana-lyze Pearson correlation coefficient. Also, principle componentanalysis (PCA) was performed, to analyze the relationships amongthe bacterial sequence structures derived from DGGE gels, withSPSS Statistics software (version 14.0 for windows, SPSS Inc., Chi-cago, USA).

2.5. Analytical methods

The liquid sample was measured for pH values, and supernatantextracted from centrifugation at 2500 rpm for 10 min was filtered

24 J. Seon et al. / Bioresource Technology 157 (2014) 22–27

through a 0.2 lm membrane filter for further analyses. A high per-formance liquid chromatography (HPLC) (HPLC Ultimate 3000,Dionex, Sunnyvale, USA) equipped with Refractive Index detectorusing Aminex HPX-87H column was used to quantify VFAs (C2–C6) with alcohols (ethanol, butanol, and propanol). Every analysiswas performed at 65 �C under isocratic condition with 2.5 mMH2SO4 as mobile phase. Total organic carbon (TOC) was quantifiedusing a TOC analyzer (TOC-VCPH, Shimadzu, Kyoto, Japan). Theyield of TVFAs (g carbon in TVFAs/g carbon in alginate) was calcu-lated as the carbon amount of TVFAs produced divided by the sol-uble carbon amount of substrate feeding. Total suspended solids(TSS) and VSS measurement of inoculum sample were conductedin accordance with the procedures listed in standard methods(APHA, 1995).

3. Results and discussion

3.1. Process performance

To find out an optimum condition for the maximum TVFAsyield, the effect of alginate concentration and initial pH were eval-uated with RSM. The TVFAs yield of 37.1 ± 2.7% was determined at6.2 g/L of alginate and the initial pH of 7.6. The slight alkali condi-tion of initial pH significantly affected the yield of TVFAs at the 1%a-level. The adequacy of the fit of model was verified by comparingthe maximum model outputs of the TVFAs yield with experimentalvalidation at an optimum condition. The TVFAs yield of validatedexperiment of 39.4 ± 3.5 % was similar with the estimated opti-mum value.

Fig. 1 shows the production profiles of TVFAs and individualVFAs in the validated experiment. Acetate, butyrate, propionate,valerate, and caproate were detected as the bioconversion metab-olites, whereas alcohols (i.e., ethanol, butanol, and propanol) werenot detected during the operation. Acetate (i.e., 74.5 ± 2.1%) andbutyrate (i.e., 18.1 ± 2.7%) were the major components of TVFAsmixture. The production of acetate and propionate (i.e., average50% for each) has been reported as the major acidogenic productsin cellulose (lignocellulosic biomass) which has similar chemicalstructure with alginate (Nakashima et al., 2011). This suggests thatthe irregular chemical sequences of alginate (i.e., varying composi-tion of M- and G- residues with no regular repeating unit) may nothave similar acidogenic pathway comparing with cellulose. There-fore, the study of microbial community structure corresponded toVFA profiles in the acidogenic process from alginate could bestrongly induced.

Fig. 1. Profiles of (a) volatile fatty acids (VFAs) production in acidogenic processunder validated optimum condition (6.2 g alginate/L and pH 7.6); acetate (d),propionate (s), butyrate (.), valerate (4), caproate (j), and total VFAs (h).

3.2. Bacterial community structure

Bacterial community profiles in the optimized acidogenic pro-cess were investigated by PCR-DGGE and phylogenetic analysis.Eleven band sequences (bands 1–11) of interest were retrievedfrom the DGGE gels (Fig. 2) and sequenced for phylogenetic analy-sis in Table 1. Neighbor-joining trees illustrating the phylogeneticaffiliations of the recovered 16S rRNA gene sequences were repre-sented in Fig. 3. All bacterial sequences were closely related toknown species of the three phyla, Firmicutes (bands 3–8), Actino-bacteria (bands 9–11), and Bacteroidetes (bands 1–2).

Band 2 was related to Bacteroides coprophilus (88% similarity)which is strict anaerobe isolated from human faces with pH 7.2and 37 �C for the growth optimum (Hayashi et al., 2007). GenusBacteroides are able to ferment complex carbohydrate such as algi-nate with cell bound enzymes (Hayashi et al., 2007; Salyers et al.,1978). Indeed, the B. coprophilus-like band was detected with grad-ually increasing intensity during the operation and might havecontributed to the hydrolysis of alginate in the process (Fig. 2).Bands 9 and 11 showed 99% similarity to strains of Bifidobacteriumthermacidophilium isolated from an anaerobic digester treatingbean-curd wastewater is a carbohydrates-hydrolyzing bacteriumto produce acetate and lactate at low pH (4.0) and moderately ther-mophilic condition (maximum 49.5 �C) (Dong et al., 2000; Zhuet al., 2003). Indeed, the strong and stable intensity of band 11was detected during the overall incubation (Fig. 2), albeit the oper-ating condition (i.e., pH 7.60 and 35 �C) was not suitable for itsgrowing. B. thermacidophilum spp. is able to grow through rangefrom 30 to 49.5 �C (Dong et al., 2000; Zhu et al., 2003). The bacteriacan resist to heat treatment with unique survival structure(Duangmanee et al., 2007). Also, the enhanced survival and stabil-ity of B. thermacidophilium loaded in alginate microparticles as amicroencapsulation, which is widely utilized to protect microor-ganisms from environmental degradation (Cui et al., 2000), mightbe occurred during the incubation in this research.

Bands 3 and 8 were closely affiliated with the strictly anaerobicgenus Clostridium with 94–98% similarities and these bands werelikely derived from Clostridium disporicum and Clostridium quiniistrains. C. disporicum and C. quinii utilize carbohydrates (mainlyglucose and sugars) to mainly produce acetate, butyrate,and hydrogen at slight alkali pH (around 7.4) and mesophilic

Fig. 2. Denaturing gel gradient electrophoresis (DGGE) profiles of bacterial 16SrRNA gene fragments at the optimum acidogenic process using alginate. Lanes arelabeled with sampling time, and for the overall operation period in parentheses, indays.

Table 1Phylogenetic affiliation of bacterial 16S rRNA gene sequences from the DGGE bands.

Band Closest sequence Accession no. % Similarity

1 Alkalibacter saccharofermentans NR_042834 892 Bacteroides coprophilus NR_041313 883 Clostridium disporicum NR_026491 98

Clostridium quinii NR_026167 984 Megaspaera paucivorans NR_043657 97

Megasphaera sueciensis NR_043656 965 Anaeroglobus geminatus NR_028812 966 Lactobacillus acidophilus NR_075045 1007 Dialister invisus sp. NR_025680 948 Clostridium disporicum NR_026491 94

Clostridium quinii NR_026167 949 Bifidobacterium thermacidophilum subsp. porcinum NR_102973 99

Bifidobacterium thermacidophilum subsp. thermacidophilum NR_037049 9910 Corynebacterium kroppenstedtii strain DSM 44385 NR_074408 96

Corynebacterium kroppenstedtii strain CCUG 35717 NR_026380 9611 Bifidobacterium thermacidophilum subsp. porcinum NR_102973 99

Bifidobacterium boum NR_025672 99

Fig. 3. Neighbor-joining trees illustrating the phylogenetic identities of the 16S rRNA gene sequences from bacterial DGGE bands.

J. Seon et al. / Bioresource Technology 157 (2014) 22–27 25

temperature condition (Svensson et al., 1992). This is supportive ofthe more pronounced production of acetate and butyrate produc-tions in the process. Indeed, bands 3 and 8 were detected with highintensity at 5–6 days of operation (i.e., the maximum acetate andbutyrate yields points). Usually, Clostridium species are highlyabundant and very frequently observed in anaerobic fermentativeprocesses producing hydrogen and VFAs from various carbohy-

drates (e.g., glucose, sucrose, starch, and cellulose) (Svenssonet al., 1992). In particular, hydrogen productions by Clostridiumspecies (i.e., Clostridium thermocellum and Clostridium butyricum)have been intensively reported from terrestrial biomass includingstarch or cellulose (Cheng et al., 2008; Levin et al., 2006; Wangand Wan, 2009), while almost no hydrogen production (belowthe detection level) was observed in this research. This indicates

Fig. 4. Principal component analysis (PCA) plot illustrating shifts in bacterialcommunity structure over the operation. Points are labeled with sampling time indays and arrows indicate the time-course community shifts in the process.

26 J. Seon et al. / Bioresource Technology 157 (2014) 22–27

that the bacterial enzymatic degradation of alginate into hydrogenpresents a challenge rather than into VFAs.

Bands 4 and 5, detected gradually weak intensity, were closelyrelated to Megasphaera spp. (96–97% similarity) and Anaeroglobusgeminatus (96% similarity), respectively. Megasphaera paucivoransand Megasphaera sueciensis isolate from the brewery environmentare strictly anaerobes fermenting pyruvate and gluconate intobutyrate and iso-valerate (Juvonen and Suihko, 2006) in acidicpH (4.3–4.9). A. geminatus was found in gastrointestinal tract ormouth, fermenting galactose and mannose into acetate, propio-nate, and butyrate in the pH of 6.5–8.0 (Carlier et al., 2002). Band6 was closely affiliated to Lactobacillus acidophilius (100% similar-ity) is a homofermentative species, fermenting sugars into lactateand growing readily at rather low pH values (below pH 5.0) (Sand-ers and Klaenhammer, 2001). Band 7 showed 94% similarity to Dia-lister invisus spp. isolated from the human oral cavity, producingacetate and propionate from glucose (Dowens et al., 2003). Band10 was related to Corynebacterium kroppenstedtii, a saccharolyticbacterium producing acetate, butyrate, and ethanol at pH 7.0 and37 �C (Collins et al., 1998).

PCA was employed to visualize the changes in bacterial commu-nity structure in the process. The considerable variance(PC1 = 25.13% and PC2 = 20.57%) and relatively closed distancesof the samples grouped indicate that the bacterial community fer-menting carbohydrates into organic acids showed some similarityfor each other. On the PCA plot, the bacterial community structureshifted clockwise with operation time (Fig. 4) and any bacterialcommunity shift from one cluster to another cluster with opera-tion time corresponded to VFAs (mainly acetate) production inthe process. The first cluster shift occurred between 0 and 2 days,the point at which the alginate started to ferment (Fig. 1). It wasprobably affected by the hydrolytic bacteria (i.e., B. coprophilus)utilizing alginate. The second shift between 3 and 7 days corre-sponded to the increasing acidogenic products. This suggested thatthe main bacterial community would change from the alginate-hydrolyzing bacteria to the VFAs-producing bacteria (i.e., mainlyClostridium spp.). The third shift was observed between 8 and14 days, the point at which VFAs started to be slight consumedand stable accumulated. Because the process was designed as aninhibition environment for methanogens and sulfate reducers thedramatically decreased accumulation of VFAs was not detected(Fig. 1). Our results demonstrated that the bacterial communityshifts reflected the changes in VFAs profiles in the optimized acido-genic process from alginate. This implied that the diagnosis andprediction of the alginate acidification might be possible by moni-

toring bacterial community shifts corresponded to VFA profilesthroughout the operation.

4. Conclusion

The bacterial community structure reflected changes in VFAsprofile in the maximized acidogenic process from alginate wasfirstly investigated in this research. A Bacteroides-related organismwas suggested to be mainly responsible for the hydrolysis of algi-nate. Additionally, Clostridium-related organisms were assumed tobe involved in fermenting alginate into VFAs, in particular acetateand butyrate. The shift of bacterial community, indicating hydroly-sis into acidogenesis of alginate, was statistically demonstratedduring the overall operation. A number of features discussed in thisresearch may induce future interests on bioconversion of alginateinto anaerobic biofuels production.

Acknowledgement

This work was financially supported by the Korea Fisheries Re-source Agency of Ministry of Oceans and Fisheries (No.20131039449).

References

Aida, T.M., Yamagata, T., Abe, C., Kawanami, H., Watanabe, M., Smith, R.L., 2012.Production of organic acids from alginate in high temperature water. J.Supercrit. Fluids 65, 39–44.

APHA, 1995. Standard Methods for Examination of Water and Wastewater.American Public Health Association, Washington, DC.

Carlier, J.P., Marchandin, H., Estelle, J.B., Lorin, V., Christine, H., Christian, C., Helene,J.P., 2002. Anaeroglobus geminatus gen. nov., sp. nov., a novel member of thefamily Veillonellaceae. Int. J. Syst. Evol. Microbiol. 52 (3), 983–986.

Chang, H.N., Kim, N.J., Kang, J., Jeong, C.M., 2010. Biomass-derived volatile fatty acidplatform for fuels and chemicals. Biotechnol. Bioprocess Eng. 15 (1), 1–10.

Cheng, C.H., Hung, C.H., Lee, K.S., Liau, P.Y., Liang, C.M., Yang, L.H., Lin, P.J., Lin, C.-Y.,2008. Microbial community structure of a starch-feeding fermentativehydrogen production reactor operated under different incubation conditions.Int. J. Hydrogen Energy 33 (19), 5242–5249.

Collins, M.D., Falsen, E., Akervall, E., Sooden, B., Alvarez, A., 1998. Corynebacteriumkroppenstedtii sp. nov., a novel Corynebacterium that does not contain mycolicacids. Int. J. Syst. Bacteriol. 48, 1449–1454.

Cui, J.H., Goh, J.S., Kim, P.H., Choi, S.H., Lee, B.J., 2000. Survival and stability ofBifidobacteria loaded in alginate poly-L-lysine microparticle. Int. J. Pharm. 210,51–59.

Dong, X., Xin, Y., Jian, W., Ling, D., 2000. Bifidobacterium thermacidophilum sp. nov.,isolated from an aerobic digester. Int. J. Syst. Evol. Microbiol. 50 (1), 119–125.

Dowens, J., Munson, M., Wade, W.G., 2003. Dialister invisus sp. nov., isolated fromthe human oral cavity. Int. J. Syst. Evol. Microbiol. 53 (6), 1937–1940.

Draget, K.I., Smidsrod, O., Skjak-Braek, G., 2005. Alginates from algae, first ed. In:Steinbuchel, A., Rhee, S.K. (Eds.), Polysaccharides and Polyamides in the FoodIndustry: Properties, Production, and Patents, vol. 1. Wiley-Blackwell, pp. 1–30.

Duangmanee, T., Padmasiri, S.I., Simmons, J.J., Raskin, L., Sung, S., 2007. Hydrogenproduction by anaerobic microbial communities exposed to repeated heattreatments. Water Environ. Res. 79, 975–983.

Hayashi, H., Shibata, K., Bakir, M.A., Sakamoto, M., Tomita, S., Benno, Y., 2007.Bacteroides coprophilus sp. nov., isolated from human faeces. Int. J. Syst. Evol.Microbiol. 57 (6), 1323–1326.

Hu, B., Chen, S., 2007. Pretreatment of methanogenic granules for immobilizedhydrogen fermentation. Int. J. Hydrogen Energy 32 (15), 3266–3273.

Jung, K.A., Lim, S.R., Kim, Y., Park, J.M., 2013. Potentials of macroalgae as feedstocksfor biorefinery. Bioresour. Technol. 135, 182–190.

Juvonen, R., Suihko, M.L., 2006. Megasphaera paucivorans sp. nov., Megasphaerasueciensis sp. nov. and Pectinatus haikarae sp. nov., isolated from brewerysamples, and emended description of the genus Pectinatus. Int. J. Syst. Evol.Microbiol. 56 (4), 695–702.

Lee, M.J., Song, J.H., Hwang, S.J., 2009. Effects of acid pre-treatment on bio-hydrogenproduction and microbial communities during dark fermentation. Bioresour.Technol. 100, 1491–1493.

Levin, D.B., Islam, R., Cicek, N., Sparling, R., 2006. Hydrogen production byClostridium thermocellum 27405 from cellulosic biomass substrates. Int. J.Hydrogen Energy 31 (11), 1496–1503.

Nakashima, K., Yamaguchi, K., Taniguchi, N., Arai, S., Yamada, R., Katahira, S., Ishida,N., Takahashi, H., Ogino, C., Kondo, A., 2011. Direct bioethanol production fromcellulose by the combination of cellulase-displaying yeast and ionic liquidpretreatment. Green Chem. 13 (10), 2948–2953.

J. Seon et al. / Bioresource Technology 157 (2014) 22–27 27

Pham, H.D., Seon, J., Lee, S.C., Song, M., Woo, H.C., 2013. Maximization of volatilefatty acids production from alginate in acidogenesis. Bioresour. Technol. 148,601–604.

Salyers, A.A., Palmer, J.K., Wikins, T.D., 1978. Degradation of polysaccharides byintestinal bacterial enzymes. Am. J. Clin. Nutr. 31, 128–130.

Sanders, M.E., Klaenhammer, T.R., 2001. Invited review:the scientific basis ofLactobacillus acidophilus NCFM functionally as probiotic. J. Dairy Sci. 84, 319–331.

Singhania, R.R., Patel, A.K., Christophe, G., Fontanille, P., Larroche, C., 2013. Biologicalupgrading of volatile fatty acids, key intermediates for the valorization of biowastethrough dark anaerobic fermentation. Bioresour. Technol. 145, 166–174.

Svensson, B.H., Dubourguier, H.C., Prensier, G., Zehnder, A.J.B., 1992. Clostridiumquinii sp. nov., a new saccharolytic anaerobic bacterium isolated from granularsludge. Arch. Microbiol. 157, 97–103.

Wang, J., Wan, W., 2009. Factors influenting fermentative hydrogen production: areview. Int. J. Hydrogen Energy 34 (2), 799–811.

Wei, N., Quarterman, J., Jin, Y.S., 2013. Marine macroalgae: an untapped resource forproducing fuels and chemicals. Trends Biotechnol. 31 (2), 70–77.

Yang, J.S., Xie, Y.J., He, W., 2011. Research progress on chemical modification ofalginate: a review. Carbohydr. Polym. 84 (1), 33–39.

Zhu, L., Li, W., Dong, X., 2003. Species identification of genus Bifidobacterium basedon partial HSP60 gene sequences and proposal of Bifidobacteriumthermacidophilum subsp. porcinum subsp. nov. Int. J. Syst. Evol. Microbiol. 53,1619–1623.