Microbial community structure and dynamics during anaerobic digestion of various agricultural waste materials

Download Microbial community structure and dynamics during anaerobic digestion of various agricultural waste materials

Post on 09-Dec-2016

214 views

Category:

Documents

1 download

Embed Size (px)

TRANSCRIPT

<ul><li><p>BIOENERGYAND BIOFUELS</p><p>Microbial community structure and dynamicsduring anaerobic digestion of various agriculturalwaste materials</p><p>Ayrat M. Ziganshin &amp; Jan Liebetrau &amp; Jrgen Prter &amp;Sabine Kleinsteuber</p><p>Received: 1 October 2012 /Revised: 18 March 2013 /Accepted: 20 March 2013 /Published online: 28 April 2013# Springer-Verlag Berlin Heidelberg 2013</p><p>Abstract The influence of the feedstock type on the microbialcommunities involved in anaerobic digestion was investigatedin laboratory-scale biogas reactors fed with different agricul-tural waste materials. Community composition and dynamicsover 2 months of reactors operation were investigated byamplicon sequencing and profiling terminal restriction frag-ment length polymorphisms of 16S rRNA genes. Major bac-terial taxa belonged to the Clostridia and Bacteroidetes,whereas the archaeal community was dominated bymethanogenic archaea of the orders Methanomicrobiales andMethanosarcinales. Correlation analysis revealed that thecommunity composition was mainly influenced by the feed-stock type with the exception of a temperature shift from 38 to55 C which caused the most pronounced community shifts.Bacterial communities involved in the anaerobic digestion ofconventional substrates such as maize silage combined withcattle manure were relatively stable and similar to each other.In contrast, special waste materials such as chicken manure orJatropha press cake were digested by very distinct and lessdiverse communities, indicating partial ammonia inhibition or</p><p>the influence of other inhibiting factors. Anaerobic digestion ofchicken manure relied on syntrophic acetate oxidation as thedominant acetate-consuming process due to the inhibition ofaceticlastic methanogenesis. Jatropha as substrate led to theenrichment of fiber-degrading specialists belonging to thegenera Actinomyces and Fibrobacter.</p><p>Keywords Biogas . Co-digestion . 16S rRNA genes .</p><p>T-RFLP . Pyrosequencing</p><p>Introduction</p><p>Accumulation of agricultural, municipal, and industrialwastes results in contamination of the environment. Oneof the most effective and sustainable methods to reduceharmful effects of these wastes on the environment istheir treatment under anaerobic conditions. Anaerobicdigestion destructs the significant portion of organiccontent of waste products and leads to production of arenewable energy source, biogas. While distinct energycrops are used for commercial production of bioethanoland fatty acid methyl esters (biodiesel), biogas produc-tion does not depend on specific energy crops but canalso utilize residual biomass and various organic wastes.Such biowaste treatment reduces the negative effects onthe environment and contributes further to the economicbenefit of this approach. In many cases, digestates canbe used as high-quality organic fertilizers in agriculture(Weiland 2010).</p><p>Due to low carbon content, the anaerobic digestion ofmanure results in relatively low biogas yields. This makescommercial production of biogas from manure economicallyunprofitable. Co-digestion of such biowastes with other bio-mass substrates is a very attractive solution for improving thefermentation process, as it results in better distribution of</p><p>Electronic supplementary material The online version of this article(doi:10.1007/s00253-013-4867-0) contains supplementary material,which is available to authorized users.</p><p>A. M. ZiganshinDepartment of Microbiology, Kazan (Volga Region) FederalUniversity, Kazan 420008, Russia</p><p>J. Liebetrau : J. PrterDepartment of Biochemical Conversion, DeutschesBiomasseforschungszentrum (DBFZ), Torgauer Str. 116,04347 Leipzig, Germany</p><p>S. Kleinsteuber (*)Department of Environmental Microbiology, Helmholtz Centrefor Environmental Research (UFZ), Permoserstr. 15,04318 Leipzig, Germanye-mail: sabine.kleinsteuber@ufz.de</p><p>Appl Microbiol Biotechnol (2013) 97:51615174DOI 10.1007/s00253-013-4867-0</p></li><li><p>nutrients and trace elements in bioreactors, supporting micro-bial activity and providing potential for higher methane yield(Holm-Nielsen et al. 2009; El-Mashad and Zhang 2010).However, the extensive use of maize as energy crop, ascurrently practiced in Germany, has been come under criticismfor its negative effects on agro-ecosystems. To develop moresustainable bioenergy systems, the exploitation of agriculturalwaste material and by-products such as straw or other ligno-cellulosic feedstock not competing with food production is anoption. The cascade usage of biomass for the production ofvarious fuels such as biodiesel, bioethanol, and biogas or othervaluable products improves the ecobalance of energy crops.For instance, stillage from bioethanol production or press cakefrom oil seeds used for biodiesel production still containsconsiderable percentages of organic carbon which can beconverted to methane in anaerobic digestion. However,the composition of such residual biomass is often chal-lenging in anaerobic digestion processes due to imbal-anced C/N ratios, high fiber content, or the presence ofcompounds inhibiting microbial activity. For instance,protein-rich biomass such as stillage or press cake canlead to ammonia or sulfide inhibition, whereas ligno-cellulosic biomass is recalcitrant to anaerobic hydroly-sis. Also animal excrements such as poultry dung aremore problematic than conventional cattle manure dueto the high nitrogen content leading to ammonia inhi-bition. Therefore, the development of adapted conver-sion technologies requires a detailed understanding ofthe microbial processes in anaerobic digestion ofnonconventional biomass.</p><p>Anaerobic degradation of biomass with biogas productionoccurs by distinct bacterial and archaeal consortia. Microbialactivity and composition of the biogas depend on the substratetype. The first three phases of anaerobic digestionhydroly-sis, acidogenesis, and acetogenesisoccur by distinct bacte-rial consortia, while the fourth step of methanogenesis iscarried out by specialized groups of methanogenic archaea.During the last years, various molecular methods have beendescribed to analyze the community structure of differentfunctional groups of microbes. PCR primers were developedtargeting 16S rRNA genes as a phylogenetic marker as well asmetabolic key genes reflecting specific functional groups. Theapplication of these molecular approaches disclosed that mostmicroorganisms in natural habitats as well as in biogas re-actors are still uncultivated. Their isolation and cultivationwould provide a deeper understanding of the ecology andfunctions of these microbial consortia (Narihiro andSekiguchi 2007). Phylogenetic and metabolic marker genesfor studying microbial populations in anaerobic digesters arewidely used, most frequently targeting methanogenic archaea(Lee et al. 2009; OReilly et al. 2009; Steinberg and Regan2009). The reason that most community analyses in anaerobicdigesters target methanogenic archaea is that they are less</p><p>diverse than bacteria and represent the bottleneck of the wholeprocess under conditions of process failure, due to a lack offunctional redundancy. In contrast to other anaerobic digestionprocesses, the rate-limiting step in the digestion of lignocellu-losic biomass is the hydrolysis due to the poor bioavailabilityof organic carbon within plant fibers under anaerobic condi-tions. Moreover, process failures leading to an inhibition ofmethanogenesis also affect acetogenic bacteria which rely ona syntrophic process with hydrogenotrophic and aceticlasticmethanogens (Demirel and Scherer 2008). Such disturbanceson the level of acetogenesis are reflected by the accumulationof volatile fatty acids in a malfunctioning reactor, a parameterwhich is also used for process control (Gerardi 2003). For bothreasonsthe limitation of hydrolysis in the digestion of lig-nocellulosic biomass and the inhibition of acetogenesis underspecific process conditionscommunity analyses should alsoconsider the bacterial consortia. However, data on qualitativeand quantitative bacterial community shifts depending on thesubstrate type and process parameters are still meager.Therefore, investigation of both bacterial and archaealcommunity dynamics during the digestion of agriculturalby-products in combination with process monitoringshould provide insights into microbial community be-havior and interactions as a prerequisite to improve theanaerobic digestion process. Temporal changes in com-position of microbial consortia in the environment canbe effectively analyzed using molecular techniques suchas terminal restriction fragment length polymorphism (T-RFLP) analysis of phylogenetic and metabolic markergenes (Abdo et al. 2006).</p><p>In this study, the impact of substrate type on the efficien-cy of anaerobic digestion and concomitant dynamics of thebacterial and archaeal communities were investigated in 11laboratory-scale continuous stirred biogas reactors. The fol-lowing agricultural waste materials served as feedstock:chicken manure combined with cattle manure, cattle manurealone or in combination with maize straw or distillers grains,and Jatropha press cake. For comparison, cattle manurecombined with maize silage was used as a conventionalfeedstock. Jatropha sp. is a widely used energy plant forcommercial production of biodiesel. After extraction of oilfrom Jatropha nascent, residual biomass was tested here as apossible substrate for biogas generation. Bacteria participat-ing in the anaerobic digestion of Jatropha residues were notyet reported in the literature. The composition and dynamicsof microbial communities during bioreactors operation wereinvestigated by molecular methods targeting 16S rRNAgenes. T-RFLP fingerprinting was performed on 16SrRNA amplicons retrieved at three different sampling times.The correlation of community composition with substratesand process parameters was analyzed by multivariatestatistics to identify the key factors shaping the bacterialand archaeal community structure.</p><p>5162 Appl Microbiol Biotechnol (2013) 97:51615174</p></li><li><p>Materials and methods</p><p>Lab-scale biogas reactors and running conditions</p><p>Table 1 shows the biogas reactors running conditions andfeedstock composition. The study was started after all re-actors had been running under stable conditions for at leastthe threefold hydraulic retention time (HRT) to ensure sta-tionary conditions. The reactors were operated at mesophilictemperatures (3740 C) with one exception: reactors R 4.5and R 4.6 were shifted to thermophilic conditions (55 C)between the second and the third sampling points. All re-actors were fed every day, and the digestates were taken outevery day as well. Biogas volume and composition as wellas pH values were measured every day, while acid capacity,volatile fatty acids (VFA), and ammonium concentrationswere determined twice per week. Samples for microbialcommunity analyses were taken at three distinct times: 29September 2009 (day 1), 2 November 2009 (day 35), and 30November 2009 (day 63).</p><p>Measurement of process parameters and analytical techniques</p><p>Biogas production was measured by using milligascountersMGC-1 and drum-type gas meters TG 05 (Ritter, Germany),whereas biogas composition was analyzed with an infra-redlandfill gas analyzer GA 94 (Ansyco, Germany). For am-monium analysis, Nesslers reagent (Merck, Germany) wasadded to the liquid phase of the reactor effluents, and sam-ples were then assayed with a DR/2000 spectrophotometerat 425 nm (Hach Company, USA). Acid capacity of theeffluents was determined by titration with 0.0250.1 MH2SO4 in a pH range of 4.5 and 3.5 using a TitrationExcellence T90 titrator (Mettler-Toledo, Switzerland). VFAwere analyzed on a 5890 series II gas chromatograph(Hewlett Packard, USA) equipped with an Agilent HP-FFAP column (30 m0.32 mm0.25 m) as described byZiganshin et al. (2011). The samples were analyzed afteradding of 0.5 mL of 85 % H3PO4 and 3.0 mL of the aqueouseffluents into 10 mL vials. The vials were then closedtightly, and the gaseous phase was injected into the GC.Acetate, propionate, butyrate, isobutyrate as well as othersodium salts of organic acids were used as standards(Sigma-Aldrich, Germany). All chemicals used in this workwere of analytical or higher grade.</p><p>Sequencing and T-RFLP fingerprinting of 16S rRNA genes</p><p>From the digester effluent of each reactor, 15 mL sampleswere withdrawn and instantly used for DNA extraction.Samples were sedimented by centrifugation at 20,000gfor 10 min. DNA was then extracted and purified from0.5 g of sediment using a FastDNA SPIN Kit for soil (MP</p><p>Biomedicals, Germany) and quantified with a NanoDropND-1000 UVvis spectrophotometer (ThermoFisherScientific, Germany).</p><p>Bacterial 16S rRNA gene fragments were PCR-amplifiedwith the primers Bac27F (5-GAG TTT GAT CMT GGCTCA G-3) and Bac519R (5-GWA TTA CCG CGG CKGCTG-3) using the Phire Hot Start II DNA Polymerase(Thermo Scientific). After 25 cycles, additional ten cycleswere performed applying 454 fusion primers tagged withmultiplex identifier sequences. Amplicons were purifiedfrom an agarose gel using the MinElute Gel Extraction Kit(Qiagen) and quality-checked on an Agilent 2100Bioanalyzer. Fluorometric quantitation and preparation ofthe amplicon library were performed as described in the GSJunior Amplicon Library Preparation Method Manual(Roche). Amplicons from 13 samples (one of each reactorat the second sampling time and additionally the third sam-pling times from reactors R 4.5 and R 4.6) were pooled andapplied for emulsion PCR using the Lib-L emPCR Kit(Roche). Pyrosequencing of the library was run on a GSJunior picotiter plate according to the manufacturers rec-ommendations. Analysis of raw data and sorting of themultiplex identifiers was done using the GS Junior software.The sequences were further processed using the RDPpyrosequencing pipeline (http://pyro.cme.msu.edu/). TheRDP Classifier was used for the taxonomic assignment(http://rdp.cme.msu.edu).</p><p>Archaeal 16S rRNA gene fragments were PCR-amplifiedand cloned as previously described (Ziganshin et al. 2011)with the exception that the forward primer UniArc21F (5-TTC YGK TTG ATC CYG SCR G-3) was used for theamplification of archaeal 16S rRNA genes. Recombinantclones were picked up and screened for the suitable insertsize in PCR reactions using the vector-specific primersM13uni(21) and M13rev(29). The clone libraries wereanalyzed for restriction fragment length polymorphisms ap-plying the restriction endonuclease HaeIII (New EnglandBiolabs, Germany). Restriction patterns were clusteredusing the Phoretix 1D software (Nonlinear Dynamics,UK). Representative clones from each cluster were partiallysequenced as described by Ziganshin et al. (2011). Thesequences were compared to the NCBI database using thenucleotide Basic Local Alignment Search Tool (BLAST)program (http://www.ncbi.nlm.nih.gov/BLAST) and taxo-nomically assigned according to the RDP Classifier. Datawere checked for chimeric sequences using Bellerophon(http://comp-bio.anu.edu.au/bellerophon/bellerophon.pl).The partial archaeal 16S rRNA gene sequences were depos-ited in the GenBank datab...</p></li></ul>