biomethanation i

81 Advances in Biochemical Engineering / Biotech nology Series Editor: T. Scheper

Edi tor ia l Board:

W. Babe l - H. W. B lanch . I. E n d o . S.-O. Enfors A. Fiechter • M. H o a r e • B. M a t t i a s s o n • H. S a h m K. ScMigerl • G. S t e p h a n o p o u l o s • 13. yon S tockar G. T. Tsao . J. Vi l ladsen • C. W a n d r e y • J. J. Z h o n g

Springer Berlin Heidelberg New York Hong Kong London Milan Paris Tokyo

Biomethanat ion I

Volume Editor: Birgitte K. Ahring

With contributions by B. K. Ahring, I. Angelidaki, E. Conway de Macario, H. N. Gavala, J. Hofman-Bang, A. ]. L. Macario, S. J.W.H. Oude Elferink, L. Raskin, A. ]. M. Stams, P. Westermann, D. Zheng

~ Springer

Advances in Biochemical Engineering/Biotechnology reviews actual trends in modern biotechnology. Its aim is to cover all aspects of this interdisciplinary technology where knowledge, methods and expertise are required for chemistry, biochemistry, microbiology, genetics, chemical engineering and computer science. Special volumes are dedi- cated to selected topics which focus on new biotechnological products and new processes for their synthesis and purification. They give the state-of-the-art of a topic in a comprehensive way thus being a valuable source for the next 3-5 years. It also discusses new discoveries and applications.

In general, special volumes are edited by well known guest editors. The series editor and publisher will however always be pleased to receive suggestions and supplementary information. Manuscripts are accepted in English.

In references Advances in Biochemical Engineering/Biotechnology is abbreviated as Adv Biochem Engin/Biotechnol as a journal. Visit the ABE home page at http://link.springer.de/series/abe/ http://link.springer-ny.com/series/abel

ISSN 0724-6145 ISBN 3-540-44322-3 Springer-Verlag Berlin Heidelberg New York

Library of Congress Catalog Card Number 72-152360

This work is subject to copyright. All rights are reserved, whether the whole or part of the material is concerned, specifically the rights of translation, reprinting, reuse of illustrations, recitation, broadcasting, reproduction on microfilm or in any other way, and storage in data banks. Duplication of this publication or parts thereof is permitted only under the provisions of the German Copyright Law of September 9, 1965, in its current version, and permission for use must always be obtained from Springer-Verlag. Violations are liable for prosecution under the German Copyright Law.

Springer-Vedag Berlin Heidelberg New York a member of BertelsmannSpringer Science+Business Media GmbH

http://www.springer.de

© Springer-Verlag Berlin Heidelberg 2003 Printed in Germany

The use of general descriptive names, registered names, trademarks, etc. in this publication does not imply, even in the absence of a specific statement, that such names are exempt from the relevant protective laws and regulations and therefore free for general use.

Typesetting: Fotosatz-Service K6hler GmbH,Wtirzburg Cover: K/inkelLopka GmbH, Heidelberg/design & production GmbH, Heidelberg

Printed on acid-free paper 02/3020mh - 5 4 3 2 1 0

Series Editor Professor Dr. T. Scheper Institute of Technical Chemistry University of Hannover Callinstrafle 3 30167 Hannover, Germany E-mail: [email protected]

Volume Editor Professor Birgitte K. Ahring Biocentrum The Technical University of Denmark DTU, Block 227 2800 Lyngby Denmark E-mail: [email protected]

Editorial Board Prof. Dr. W. Babel Section of Environmental Microbiology Leipzig-Halle GmbH Permoserstrafle 15 04318 Leipzig, Germany E-maih [email protected]

Prof. Dr. I. Endo Faculty of Agricukure Dept. of Bioproductive Science Laboratory of Applied Microbiology Utsunomiya University Mine-cho 350, Utsunomiya-shi Tochigi 321-8505, Japan E-maih [email protected]

Prof. Dr. A. Fiechter Institute of Biotechnology Eidgen6ssische Technische Hochschule ETH-H6nggerberg 8093 Ztirich, Switzerland E-maih [email protected]

Prof. Dr. H.W. Blanch Department of Chemical Engineering University of California Berkely, CA 94720-9989, USA E-maih [email protected]

Prof. Dr. S.-O. Enfors Department of Biochemistry and Biotechnology Royal Institute of Technology Teknikringen 34 100 44 Stockholm, Sweden E-mail: [email protected]

Prof. Dr. M. Hoare Department of Biochemical Engineering University College London Torrington Place London, WC1E 7JE, UK E-maih [email protected]

VI Editorial Board

Prof. Dr. B. Mattiasson Department of Biotechnology Chemical Center, Lund University P.O. Box 124, 221 00 Lund, Sweden E-mail: [email protected]

Prof. Dr. K. Schfigerl Institute of Technical Chemistry University of Hannover Callinstrat~e 3 30167 Hannover, Germany E-maih [email protected]

Prof. Dr. U. von Stockar Laboratoire de G~nie Chimique et Biologique (LGCB) D~partment de Chimie Swiss Federal Institute of Technology Lausanne 1015 Lausanne, Switzerland E-mail'. [email protected]

Prof. Dr. J. Villadsen Center for Process of Biotechnology Technical University of Denmark Building 223 2800 Lyngby, Denmark E-mail: john. [email protected]

Prof. Dr. J.-J. Zhong State Key Laboratory of Bioreactor Engineering East China University of Science and Technology 130 Meflong Road Shanghai 200237, China E-mail.'[email protected]

Prof. Dr. H. Sahm Institute of Biotechnolgy Forschungszentrum Jfilich GmbH 52425 Jfilich, Germany E-mail: [email protected]

Prof. Dr. G. Stephanopoulos Department of Chemical Engineering Massachusetts Institute of Technology Cambridge, MA 02139-4307, USA E-maih [email protected]

Prof. Dr. G. T. Tsao Director Lab. of Renewable Resources Eng. A.A. Potter Eng. Center Purdue University West Lafayette, IN 47907, USA E-maih [email protected]

Prof. Dr. C. Wandrey Institute of Biotechnology Forschungszentrum Jfilich GmbH 52425 Jfilich, Germany E-mail: c. [email protected]

Advances in Biochemical Engineering/Biotechnology also Available Electronically

For all customers with a standing order for Advances in Biochemical Engineer- ing/Biotechnology we offer the electronic form via SpringerLink free of charge. Please contact your librarian who can receive a password for free access to the full articles. By registration at:

http://www.springer.de/series/abe/reg_form.htm

If you do not have a standard order you can nevertheless browse through the table of contents of the volumes and the abstracts of each article at:

http://link.springer.de/series/abe/ http://link.springer_ny.com/series/abe/

There you will find also information about the

- Editorial Board - Aims and Scope - Instructions for Authors

Attention all Users of the "Springer Handbook of Enzymes"

Information on this handbook can be found on the internet at http:llwww.springer.de/enzymesl

A complete list of all enzyme entries either as an alphabetical Name Index or as the EC-Number Index is available at the above mentioned URL. You can down- load and print them free of charge.

A complete list of all synonyms (more than 25,000 entries) used for the enyzmes is available in print form (ISBN 3-540-41830-X).

Save 15 % We recommend a standing order for the series to ensure you automatically receive all volumes and all supplements and save 15 % on the list price.

Contents

Perspectives for Anaerobic Digestion B. K. Ahring . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .

Metabolic Interactions Between Methanogenic Consortia and Anaerobic Respiring Bacteria A. J. M. Stams, S. J. W. H. Oude Elferink, P. Westermann . . . . . . . . . . .

Kinetics and Modeling of Anaerobic Digestion Process H. N. Gavala, I. Angelidaki, B. K. Ahring . . . . . . . . . . . . . . . . . . . .

Molecular Biology of Stress Genes in Methanogens: Potential for Bioreactor Technology E. Conway de Macario, A. J. L. Macario . . . . . . . . . . . . . . . . . . . .

Molecular Ecology of Anaerobic Reactor Systems J. Hofman-Bang, D. Zheng, P. Westermann, B. K. Ahring, L. Raskin . . . . .

Author Index Volumes 51-81 . . . . . . . . . . . . . . . . . . . . . . . . .

Subject Index . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .

31

57

95

151

205

217

Contents of Volume 82

Biomethanation II Volume Editor: Birgitte K. Ahring

Applications of the Anaerobic Digestion Process I. Angelidaki, L. Ellegaard, B. K. Ahring

Anaerobic Granular Sludge and Biof'dm Reactors I.V. Skiadas, H. N. Gavala, J. E. Schmidt, B. K. Ahring

Potential for Anaerobic Conversion of Xenobiotics A. S. Mogensen, J. Dolfing, E Haagensen, B. K. Ahring

Monitoring and Control of Anaerobic Reactors P. F. Pind, I. Angelidaki, B. K. Ahring, K. Stamatelatou, G. Lyberatos

Preface

In November 1776, Alessandro Volta performed his classic experiment disturb-ing the sediment of a shallow lake, collecting the gas and demonstrating that thisgas was flammable. The science of Biomethanation was born and, ever since, sci-entists and engineers have worked at understanding this complex anaerobic bio-logical process and harvesting the valuable methane gas produced during anaer-obic decomposition. Two lines of exploitation have developed mainly during thelast century: the use of anaerobic digestion for stabilization of sewage sludge,and biogas production from animal manure and/or household waste. Lately, theemphasis has been on the hygienic benefit of anaerobic treatment and its effecton pathogens or other infectious elements. The importance of producing a safeeffluent suitable for recirculation to agricultural land has become a task just asimportant as producing the maximum yield of biogas from a given type ofwaste. Therefore, anaerobic digestion at elevated temperatures has become themain area of interest and has been growing during the last few years.

Anaerobic digestion demands the concerted action of many groups ofmicrobes each performing their special role in the overall degradation process.Both Bacteria and Archaea are involved in the anaerobic process while theimportance, if any, of eukaryotic microorganisms outside the rumen environ-ment is still unknown. The basic understanding of the dynamics of the complexmicroflora was elucidated during the latter part of the last century where theconcept of inter-species hydrogen transfer was introduced and tested. The isola-tion of syntrophic bacteria specialized in oxidation of intermediates such asvolatile fatty acids gave strength to the theories. Lately the use of molecular tech-niques has provided tools for studying the microflora during the biomethana-tion process in situ. However, until now the main focus has been on probing thedynamic changes of specific groups of microorganisms in anaerobic bioreactorsand less emphasis has been devoted to evaluating the specific activities of thedifferent groups of microbes during biomethanation. In the future we can expectthat the molecular techniques will be developed to allow more dynamic studiesof the action of specific microbes in the over-all process. From the present studies we know that many unknown microbes are found in anaerobic bio-reactors. Especially within the domain of Archaea, there are whole phyla ofmicrobes such as the Crenarchaeota, which make up significant fractions ofmicrobes in a reactor but without cultured representatives. Improving the tech-niques for the isolation of presently unculturable microbes is a major task forthe future.

Preface

Anaerobic digestion of waste has been implemented throughout the world fortreatment of wastewater, manure and solid waste and most countries have sci-entists, engineers and companies engaged in various aspects of this technology.Even though the implementation of anaerobic digestion has moved out of theexperimental phase, there is still plenty of room for improvements. The basicunderstanding of the granulation process, the basis for the immobilization ofanaerobic microbes to each other without support material in UASB reactors, isstill lacking. Like any other bioprocess, anaerobic digestion needs further con-trol and regulation for optimization. However, until now suitable sensors fordirect evaluation of the biological process have been lacking and anaerobicbioreactors have generally been controlled by indirect measurements of biogasor methane production along with measurements of pH and temperature. Thenewly development of an on-line monitoring system for volatile fatty acids couldbe a major step in the right direction and the use of infra-red monitoring sys-tems could bring the price down to a reasonable level. A better performance oflarge-scale anaerobic bioreactor systems for treatment of complex mixtures ofwaste can be expected to be based on on-line monitoring of the process in thefuture along with controlling software for qualified management of these plants.

Besides treatment of waste, anaerobic digestion possesses a major potentialfor adding value to other biomass converting processes such as gasification,bioethanol or hydrogen from ligno-cellulosic materials.Conversion of ligno-cel-lulosic biomass will often leave a large fraction of the raw material untouchedwhich will be a burden for the over-all economy of the process and will demandfurther treatment.Anaerobic digestion will on the other hand be capable of con-verting the residues from the primary conversion into valuable methane, whichwill decrease the cost and the environmental burden of the primary production.

Biomethanation is an area in which both basic and applied research isinvolved. Major new developments will demand that both disciplines worktogether closely and take advantage of each other’s field of competence. The twovolumes on Biomethanation within the series of Advances in Biochemical Engi-neering and Biotechnology have been constructed with this basic idea in mindand, therefore, both angles have been combined to give a full picture of the area.The first volume is devoted to giving an overview of the more fundamentalaspects of anaerobic digestion while the second volume concentrates on somemajor applications and the potential of using anaerobic processes. The two vol-umes will therefore be of value for both scientists and practitioners within thefield of environmental microbiology, anaerobic biotechnology, and environ-mental engineering. The general nature of most of the chapters along with theunique combination of new basic knowledge and practical experiences should,in addition, make the books valuable for teaching purposes.

The volume editor is indebted to all the authors for their excellent contribu-tions and their devotion and cooperation in preparing these two volumes onBiomethanation.

Lyngby, January 2003 Birgitte K. Ahring

X Preface

Perspectives for Anaerobic Digestion

Birgitte K. Ahring

University of California, Los Angeles (UCLA), School of Engineering and Applied Science,Civil and Environmental Engineering Dept., 5732 Boelter Hall, Box 951593, Los Angeles,California 90095-1593, USA

Present address: Biocentrum, The Technical University of Denmark, DTU, Block 227,2800 Lyngby, Denmark. E-mail: [email protected]

The modern society generates large amounts of waste that represent a tremendous threat tothe environment and human and animal health. To prevent and control this, a range of differ-ent waste treatment and disposal methods are used. The choice of method must always bebased on maximum safety, minimum environmental impact and, as far as possible, on val-orization of the waste and final recycling of the end products. One of the main trends oftoday’s waste management policies is to reduce the stream of waste going to landfills and torecycle the organic material and the plant nutrients back to the soil.Anaerobic digestion (AD)is one way of achieving this goal and it will, furthermore, reduce energy consumption or mayeven be net energy producing.This chapter aims at provide a basic understanding of the worldin which anaerobic digestion is operating today. The newest process developments as well asfuture perspectives will be discussed.

Keywords. Anaerobic digestion, Carbon-flow, Microbiology, Antimization, Gas yild, Effluentquality

1 Introduction . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 2

2 Microbiology of Anaerobic Digestion . . . . . . . . . . . . . . . . 3

2.1 General Scheme . . . . . . . . . . . . . . . . . . . . . . . . . . . . 32.2 Syntrophic Acetate Conversion . . . . . . . . . . . . . . . . . . . . 52.3 Microbiology of Thermophilic Digestion . . . . . . . . . . . . . . 72.4 Establishing a Stable Microflora in Thermophilic Reactors . . . . 8

3 Anaerobic Digestion Plants . . . . . . . . . . . . . . . . . . . . . 12

4 Anaerobic Digestion as a Way to Add Extra Value . . . . . . . . . 14

5 Optimization of Anaerobic Digestion . . . . . . . . . . . . . . . . 15

5.1 Increasing the Digestibility of the Waste . . . . . . . . . . . . . . 155.2 Optimization of Reactor Configuration . . . . . . . . . . . . . . . 175.3 Optimizing Process Control and Stability . . . . . . . . . . . . . . 195.4 Improving the Microbial Process and its Efficiency . . . . . . . . . 22

CHAPTER 6

Advances in Biochemical Engineering/Biotechnology, Vol. 81Series Editor: T. Scheper© Springer-Verlag Berlin Heidelberg 2003

6 Optimization of Effluent Quality . . . . . . . . . . . . . . . . . . 23

6.1 Inactivation of Pathogens and Other Biological Hazards . . . . . . 236.2 Control of Chemical Pollutants . . . . . . . . . . . . . . . . . . . 25

7 Conclusions . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 26

8 References . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 27

1Introduction

The modern society generates large amounts of waste that represent a tremen-dous threat to the environment and human and animal health. To prevent andcontrol this, a range of different waste treatment and disposal methods is used.The choice of method must always be based on maximum safety, minimumenvironmental impact and, as far as possible, on valorization of the waste andfinal recycling of the end products. One of the main trends of today’s waste man-agement policies is to reduce the stream of waste going to landfills and to recy-cle the organic material and the plant nutrients back to soil.Waste is increasing-ly becoming a problem and secure recirculation is gaining more and more atten-tion. Anaerobic digestion (AD) is one way of achieving this goal and it will, fur-thermore, reduce energy consumption, or may even be energy producing, whichis of major importance to the global environment.Anaerobic digestion has beenimplemented for years as a means for the stabilization of sewage sludge; how-ever, during the past years anaerobic digestion technologies have been expand-ed to emphasize treatment and energy recovery from many other types ofwastes including animal wastes, source-sorted household wastes, organic indus-trial wastes and industrial wastewater. Compared to incineration, anaerobicdigestion creates more energy during the treatment of wastes, which normallyhave high water content. During incineration the nutrients are lost. Followingthe increasing interest in implementation of anaerobic digestion, optimizationof this process is becoming increasingly more important. Despite the increasedefforts spent on waste reduction, the amounts of waste are increasing through-out the world. This has led to ideas for a total removal of waste through injectioninto the deep underground (below 2 km) into old oil wells far below any thegroundwater level [1]. The recovery of methane will, however, be of importancefor the feasibility and economy of this technique and methane development atthese high temperatures, pressures and salinity is now under investigation.

This chapter focuses on the perspectives for optimization of anaerobic diges-tion after a brief introduction to the microbiology of anaerobic digestion. Opti-mization is a double-sided task: it involves both an increase of the biogas yield,which again implies an increased removal of the organic material in the waste,as well as ensuring an effluent with a sufficiently high quality to allow for recy-cling of the material as a fertilizer. A number of areas for improving the biogasyield will be discussed such as, for example, increasing the digestibility of the

2 B. K. Ahring

waste, optimizing process control, and improving the microbial process. Withrespect to effluent quality, emphasis will be on inactivation of pathogens andcontrol of chemical pollutants.

2Microbiology of Anaerobic Digestion

2.1General Scheme

A major value of anaerobic digestion is linked to the production of biogas(methane and carbon dioxide) formed as the end product during degradation oforganic material without oxygen. This energy is renewable and CO2 neutral andcan be used for production of electricity and heat. Many different consortia ofmicroorganisms with different roles in the overall process scheme are neededfor the AD process, which occurs naturally in anaerobic ecosystems such as sed-iments, paddy fields, water-logged soils and in the rumen [2].

Three major groups of microorganisms have been identified with differentfunctions in the overall degradation process [3] (Fig. 1):

1. The hydrolyzing and fermenting microorganisms are responsible for the ini-tial attack on polymers and monomers found in the waste material and pro-duce mainly acetate and hydrogen, but also varying amounts of volatile fattyacids (VFA) such as propionate and butyrate as well as some alcohols.

Perspectives for Anaerobic Digestion 3

Fig. 1. The anaerobic degradation process

2. The obligate hydrogen-producing acetogenic bacteria convert propionateand butyrate into acetate and hydrogen.

3. Two groups of methanogenic Archaea produce methane from acetate or hydrogen, respectively.

The major part of the carbon flow in a well-operating anaerobic reactor occursbetween the fermentative microorganisms and the methanogens. Only between20 and 30 % of the carbon is transformed into intermediary products beforethese are metabolized to methane and carbon dioxide (Fig. 2) [4].

A balanced anaerobic digestion process demands that the products from thefirst two groups of microbes responsible for hydrolyzing and fermenting thematerial to hydrogen and acetate, simultaneously are used by the third group ofmicrobes for the production of methane and carbon dioxide. The first group ofmicroorganisms can survive without the presence of methanogens but will,under these conditions, form an increased amount of reduced products such asVFA (Fig. 3).

The second group does, however, rely on the activity of the methanogens forremoving hydrogen to make their metabolism thermodynamically possible astheir reactions are endergonic under standard conditions and only occur whenhydrogen is kept below a certain concentration. The relationship between theVFA-degrading bacteria and the hydrogen-utilizing methanogens is defined assyntrophic due to the dependent nature of this relationship and the process is

4 B. K. Ahring

Fig. 2. Carbon flow in anaerobic environments with active methanogens

Fig. 3. Carbon flow in anaerobic environments without active methanogens

called interspecies hydrogen transfer (Fig. 4) [3]. The lower the hydrogen con-centration the better are the thermodynamics of the VFA degradation. The dis-tance between the VFA degrader and the hydrogen utilizer does, therefore, affectthe concentration of hydrogen in the liquid phase, which again affects the ther-modynamics of the process. Therefore, the conversion is improved in granulesand flocks compared to a situation where the microbes are distributed freely ina liquid solution [5]. The two partners have to share a very small amount ofenergy and the conditions for ensuring energy for both microbes is very strictand can only be met within a narrow range of hydrogen concentrations [6].

2.2Syntrophic Acetate Conversion

Syntrophic relationships have also been found to be of importance for conver-sion of acetate when the acetate-degrading methanogens are inhibited by highconcentrations of ammonia [7, 8] or sulfite (unpublished). Under these condi-tions the acetate-utilizing methanogens are inhibited and other groups ofmicrobes replace them to obtain energy from the oxidation of acetate to hydro-gen and carbon dioxide (Fig. 5).

Due to thermodynamic constrains this reaction proceeds much better atincreased temperatures and is the way of acetate transformation when the tem-perature is higher than 60°C, close to the upper temperature limit of ther-mophilic acetate-utilizing methanogens [9, 10]. In accordance with this, the pop-ulation of Methanosarcina species disappeared more or less instantaneouslyfrom a biogas reactor operated on manure, when the temperature was increasedfrom 55 to 65°C [11]. Concurrently, the acetate concentration first increased and


Fig. 4. Interspecies hydrogen transfer

then stabilized at a level somewhat higher than that found at 60°C [12]. Thiscoincided with a significant increase in the population of hydrogen-utilizingmethanogens [11] indicating that this group had become dominant in the over-all conversion. Both syntrophic acetate oxidation and methanogenesis fromacetate can be simultaneously active in a reactor system as indicated by severalisotope studies often showing that less than 95 % of the methane produced fromacetate is derived from the methyl group. Isotope experiments with biomassfrom thermophilic reactors have further shown that the concentration of acetateaffects the competition between the two processes. When the concentration ofacetate is low, syntrophic acetate conversion is the major process for acetatetransformation [13, 14]. However, when the concentration of acetate is above thethreshold level [15] for the specific population of acetate-utilizing methanogensin the reactor, these will be the major group active in the system. These findingsfurther explain why the numbers of hydrogen-utilizing methanogens are high inthermophilic granules, which have exclusively been fed with acetate for a longperiod [16]. Furthermore, the numbers of acetate-utilizing methanogens arehighest close to the surface of the granules, where the concentration of acetate ishighest, while the populations of hydrogen-utilizing methanogens increasedtowards the center of the granules [16].

The first microbe found to perform acetate oxidation was a thermophilic bac-terium belonging to the group of homo-acetogenic bacteria capable of reversingthe acetate-forming reaction from hydrogen and carbon dioxide [17]. This bac-terium used a very limited range of substrates all related to its homo-acetogenicnature [17]. Over time more microbes have been identified as being capable of

6 B. K. Ahring

Fig. 5. Modified anaerobic degradation process with syntrophic acetate conversion

carrying out this reaction. Some of these microbes have been found to use alarge variety of substrates [18] and, furthermore, to be normal members of thepopulations of fermentative microbes in thermophilic reactors. This indicatesthat, at least in thermophilic reactors, syntrophilic acetate oxidation could beperformed by a variety of the fermentative bacteria in the reactor when no other substrates are available. This needs, however, further verification.

2.3Microbiology of Thermophilic Digestion

Microbes thriving at high temperatures have been known for years [19]. Thereaction rate of many chemical reactions will double by an increase of 10°Caccording to the Arrhenius equation. The same is, however, not always the casefor microbial reactions where the temperature response is specific for the par-ticular microbe. Different groups of microbes have been identified where theones of interest for anaerobic digestion are mesophilic strains with an optimumbetween 30 and 40°C, and thermophilic strains with an optimum between 50and 60 °C [20]. The mixed microflora found in an anaerobic bioreactor general-ly shows an increasing rate from a temperature of 20 to 60 °C and the theoreti-cal temperature gap between mesophilic and thermophilic strains is not appar-ent when viewing the process as a whole [21]. Anaerobic digestion at a temper-ature below 20°C, or at a temperature above 60 °C, generally shows a lowermethane yield than within these limits. However, anaerobic digestion has beenshown to be possible even at extreme thermophilic conditions of 70 °C and more[28 – 30]. Experiments with high temperature digestion of manure showed thatmajor changes occurred in the microbial populations of the anaerobic reactorwhen the temperature was increased from 55 to 65 °C [12]. Besides a significantincrease in the population of Archaea compared to Bacteria, also the popula-tions of methanogens underwent large changes over time. The population ofhydrogen-utilizing methanogens did, for example, change from a major popula-tion belonging to the genus Methanobacterium to another belonging toMethanococcus over a 3-month period [11]. Such results clearly demonstratethat reactors operated at extreme conditions can take months before a stablemicroflora has established. With this in mind it is difficult to guess if themethane yield actually will be lower after an extended period of many months.

Within the normal temperature range the general carbon flow of ther-mophilic reactors was found to be very similar to that of mesophilic reactors[31]. A slightly higher amount of the carbon was channeled directly into acetateand a slightly smaller amount of carbon was turned over via the pool of VFA [4].Many extreme thermophilic Bacteria or Archaea have been found to producemainly acetate and hydrogen as their end products [32]. Therefore, less butyrateand propionate can be expected at these high temperatures. Different maximumtemperatures were found for the different microbial groups in a thermophilicanaerobic reactor treating manure [33]. For instance, among the methanogens,the acetate-utilizing methanogens have a much lower temperature maximum(ca. 62 °C) compared to the hydrogen-utilizing methanogens (ca. 75 °C) [33].However, the actual temperature of the reactor affects the specific populations


which are active in the reactor. Therefore, a higher temperature optimum andmaximum is found for the main metabolic groups in extreme thermophilic reac-tors compared to thermophilic reactors [28]. Methane production was found inmicrobial mat samples taken from a slightly alkaline hot spring at 80 °C [34].This demonstrates that methanogenesis is possible even at this very high tem-perature.

2.4Establishing a Stable Microflora in Thermophilic Reactors

Waste such as sewage sludge, manure or household waste contains many differ-ent populations of anaerobic or facultative anaerobic microorganisms. Most ofthese microbes are mesophilic and only a very small number of true ther-mophiles is present. The number of microbes in raw sewage sludge utilizing sub-strates such as acetate or cellulose at 60 °C is extremely low (ca. 100 per g) [35].The numbers are somewhat higher at 55 °C but still much lower than the num-bers at 37 °C [35]. These facts clearly show the problems of establishing stablereactors at higher temperatures. Where the microflora of mesophilic reactorscan be established directly based on the raw material fed to the reactor, themicroflora of the thermophilic reactor has to be propagated from small minor-ity populations found in the raw materials [36]. Many thermophilic full-scalereactors have failed through history, especially within the area of sewage sludgetreatment. The reason is basically a lack of understanding of the principles forestablishing a stable thermophilic microflora in the reactor. The same alsoapplies to the literature, which is full of experiments with unstable thermophiliclaboratory reactors often performing poorly compared to mesophilic reactors.When reviewing the literature describing these experiments, Wiegant [37] con-cluded that process stability is lower in thermophilic reactors and that ther-mophilic reactors generally have higher concentrations of volatile fatty acids inthe effluent compared to mesophilic reactors. During recent years where morethermophilic reactors have been implemented, it has been shown that this con-clusion it not correct and that stable thermophilic reactors with a balanced thermophilic microflora perform just as well as stable mesophilic reactors [33,38, 39].

The key to obtain a balanced thermophilic microflora is to give optimalgrowth conditions to the small numbers of thermophilic populations found inthe raw material during start-up of the bioreactor [36]. If sufficient thermophilicseed material is available, it is possible to carry out a rapid start-up of a ther-mophilic reactor [33]. The seed material should be evaluated before use withrespect to the destruction of volatile solids in the reactor from which the seed isobtained as well as the concentration of VFA. If possible, it will be beneficial toperform a methanogenic activity testing of the seed material to establish thepotential of this seed for transforming extra loads of methanogenic substrates(acetate and hydrogen) [40, 41]. After addition of the seed to an empty reactor itshould be allowed to equilibrate for 1 day before feeding is initiated at thedesired thermophilic temperature. A slow and graduate change of the tempera-ture only prolongs the start-up phase and does not select for true thermophiles

8 B. K. Ahring


Fig.

6.St

art-

up u

sing

ther

mop

hilic

dig

este

d se

ed m

ater

ial

10 B. K. Ahring

Fig. 7. Unbalanced growth of methanogens (dots) and fermentative bacteria (rods)

possessing an optimum growth rate at thermophilic temperatures. Probing ofmethanogens in bioreactors has clearly demonstrated that mesophilic meth-anogens are present in thermophilic reactors and vice versa [42]. However, themajor populations are those with an optimum temperature close to the reactortemperature [43]. After equilibration, feeding should then be initiated corre-sponding to a hydraulic retention time approx. 25 % higher than the retentiontime of the reactor where the seed came from. Normally the reactor is only part-ly full at this stage and, therefore, the reactor is operated in a fed-batch modeduring this period of time. During start-up, the VFA concentration should bemonitored on a daily basis. If the VFA concentrations continue to decrease afterapprox. 3 days of feeding or remain at a stable low level, the hydraulic retentiontime can be lowered. By repeating this pattern and, at the same time, keeping atight eye on the concentration of VFA – especially the isoacids [44, 45] – it is pos-sible to reach the desired final retention time in approx. 1 month. A schematicdrawing of the expectable feeding pattern and the expectable response in VFA isshown in Fig. 6. For sewage sludge it is possible to obtain a stable process with ahigh reduction of volatile solids at a hydraulic retention time as low as 6 days atthermophilic conditions.

If no seed is available, it is even more important to plan the start-up in a con-trolled condition. It is important to avoid over-loading and build-up of VFA.Therefore, a start-up material containing only small amounts of organic mater-ial should be chosen. Mesophilic digested waste material has a much lowerorganic content than raw waste material and has at least as many thermophilicmicroorganisms as found in this material [46]. Immediately upon an increase ofthe temperature to the thermophilic region, these thermophilic microbes willstart growing. As the acid-producing microbes grow much faster than the

24 h

24 h

td = 24 h

td = 8 h


Fig.

8.St

art-

up u

sing

mes

ophi

lic d

iges

ted

seed

mat

eria

l.A

rrow

s in

dica

te fe

edin

g of

the

reac

tor

12 B. K. Ahring

methanogens, the first reaction is often an increase in the concentration of VFA[47] (Fig. 7).

Due to the limited amount of undigested material present, this increase isonly relatively small and does not affect the over-all digestion process. Immedi-ately after a decreasing trend is seen in the volatile fatty acids corresponding togrowth of the population of VFA-degrading and methanogenic microbes, it isappropriate to start feeding.A portion corresponding to approximately the dou-ble of the desired hydraulic retention time is appropriate. Depending on theresponse in VFA concentration, this trend can be continued every day unless theVFA starts to increase. After 3 to 5 days of continuous feeding it is time to lowerthe hydraulic retention time again and this pattern can be continued until thereactor has reached the desired hydraulic retention time. This is normallyreached within a period of 2 months. A schematic drawing of the expectablefeeding pattern and the expectable response in VFA is shown in Fig. 8.

The time needed for performing the start-up with a small amount of ther-mophilic seed material can further be reduced by addition of mesophilic-digest-ed material in addition to the daily feeding with raw waste material. This wasused with success for start-up of the new thermophilic sewage sludge digester at Western Lake Superior Sanitary District in Duluth, Minnesota during thesummer of 2001 [48].

3Anaerobic Digestion Plants

A large number of different AD-technologies and AD plants are found todaythroughout the world. The largest number of AD plants in the modern societytreats primary and secondary sludge (biosolid) in municipal wastewater treat-ment plants. These plants basically stabilize the waste material and the biogasproduced is often of minor importance. For some of the large wastewater plants,the biogas produced is used for electricity production and the idea of improvingthe biogas yield is attracting increased interest [49, 50]. A large number of sin-gle household biogas units have further been implemented in developing coun-tries such as China, India and Africa [51, 52]. These units will normally providegas for cooking and lighting in the households.

Another major field for anaerobic digestion is the industrial wastewater from,especially, food processing industries where the wastewater is heavily pollutedwith easily degradable organic carbon [53]. Treatment of municipal wastewaterhas further been implemented in developing countries such as India especiallywhere the average temperature is rather constant [54]. Anaerobic treatment isbasically a way to reduce BOD while the nutrients such as nitrogen and phosphorare left untouched [50]. Recent studies have, however, shown that nitrogen can bedenitrified in a chemoautotrophic anaerobic process using nitrite as an electronacceptor [55, 56]. A better way to implement anaerobic treatment of municipalwastewater would be to recover all nutrients and heavy metals from the waste-water after the anaerobic treatment using membrane technology. In this way, thebenefits of anaerobic process with respect to space, speed, low sludge productionand cost can be fully exploited and the valuable nutrients can be reused.

A large-scale biogas facility for treating manure from several farms in combi-nation with other organic wastes such as food wastes and source-sorted house-hold wastes – the so-called co-digestion – was launched in Denmark at the endof the 1980s [57, 58]. Addition of even small amounts of organic industrialwastes increases the gas production significantly (Fig. 9). Especially fatty or oilywastes have a much higher gas potential than manure and a much higher con-centration of organic material (higher dry matter content) – but also wastes richin carbohydrates and proteins will improve the gas yield per unit of reactor vol-ume. Digestion of sewage sludge or manure yields from about 1 – 2 cubic meterbiogas per cubic meter reactor volume per day while the reactors will producebetween 4 and 10 cubic meter biogas per cubic meter reactor volume with addi-tion of ca. 20 % fatty waste.

Today around 22 large-scale AD plants have been built in Denmark mainly inthe regions with high manure production and all of these plants are co-digest-ing many types of raw materials [33, 59]. The idea of large scale centralized ADplants treating mixtures of waste have spread throughout Europe and to the restof the world especially during the last ten years [60]. Besides common biogasplants, the numbers of farm biogas plants for large pig farms have steadilyincreased in many European countries [61]. Many of these plants further sup-plement the manure with raw materials with a higher gas potential. Recently, alarge number of biogas projects are on their way in USA [62] and especially inCalifornia due to the energy crisis starting at the end of year 2000, which again


Fig. 9. Addition of 5 % fish oil will double the biogas production

has resulted in higher prices on electricity. Biogas plants in USA are often muchlarger compared to the so-called “large-scale” biogas plants in Denmark andtreat manure from diaries or feeding lots having up to hundred thousand cowsor from major pig or chicken production. The AD technology is, therefore, suit-able both for small and large-scale applications. The economics depend uponthe scale, and larger plants will in general have a better economy than smallerplants. Raw sewage sludge has a very low dry matter content and, therefore, thepotential for treating waste in these plants is only used to a low degree.Additionof food waste, restaurant waste or organic industrial waste could be a good wayto make use of this potential. Several concepts are based on treatment of mix-tures of sewage sludge in combination with household waste such as the FinishWaasa process [63]. Very good results have been obtained in Grindsted, Den-mark with co-digestion of source-sorted food waste together with sewagesludge. The food waste was collected in paper bags and only a small amount wasremoved before the digestion process [65].

4Anaerobic Digestion as a Way to Add Extra Value

Production of biofuels from biomass such as bioethanol or gasification of bio-mass only makes use of a fraction of the biomass. The same is true for many oth-er biomass-based productions of non-food products. The biomass fraction leftis, however, often a good substrate for methane production (Fig. 10). In this waybiogas production can add approximately 30 % more value to the production ofbioethanol from biomass such as wet straw or corn stovers [66].

The AD process will further purify the process water allowing for recircula-tion within the system, which will further decrease the cost of ethanol produc-tion. In the future it is expected that more valuable products than methane willbe sought from waste. However, these niche productions will as a rule only use apart of the waste and methane production from the final residues can add further value to the production and will decrease the pollution load of the endproducts before their final disposal.

14 B. K. Ahring

Fig. 10. Simultaneous production of bioethanol and biogas

5Optimization of Anaerobic Digestion

The economy of a biogas plant is directly linked to the amount of biogas pro-duced per unit of raw material treated in the plant. Some costs are fixed such asthe cost of transporting the material to the AD plant and back again to the enduser or the end destination, while others are variable such as construction costs.Lowering the water content of the raw material and running the process withhigher dry matter content can significantly decrease the cost of treatment.This is of major importance when the raw material has to be transported to a centralized biogas facility and in this case it is often beneficial to separate the manure into a solid fraction and a liquid fraction, which is left behind at the farm. The liquid fraction can be used as a nitrogen-rich fertilizer at the form.

The potential for increasing the biogas yield of manure or sewage sludge islarger as only approximately half of the organic material is converted in this typeof material. This is, however, not the case for most organic industrial wastes orsource-sorted household waste, which have been found to be more easilydegradable, and approximately 80 % of the organic material is converted to bio-gas [67, 68]. Manure is, however, the major raw material available for a large-scale use of AD technology in most of the world and a large-scale implementa-tion of AD will have to be based on this raw material. AD will further improvethe quality of manure by making a more stable material with fewer pathogensand less odor. For wastewater sludge the interest in increasing the conversion ofthe organic material is further linked to the reduction in the final amount ofbiosolid, which has to be disposed after the treatment. Suitable end-use ofdigested sewage sludge or biosolids is becoming an increasing problem for manycommunities throughout the world. Some major ways to improve the gas yieldin AD plants will be by (Fig. 11):

1. Increasing the digestibility of the waste,2. Optimizing the reactor configuration,3. Optimizing process control and stability, and4. Improving the microbial process and its efficiency

5.1Increasing the Digestibility of the Waste

Several methods have been discovered to increase the digestibility of manure orsewage sludge ranging from mechanical, chemical to biological methods such asenzyme treatment. Chemical treatment with bases or acids or treatment withmixtures of enzymes have generally been found to increase the accessibility tomicrobial conversion into biogas – but these processes have all been found to betoo expensive for practical implementation [69]. Decreasing the particle sizewas found to increase the gas production from manure and the increase in gasproduction exceeded by far the extra costs of implementing a macerating unitwith several knives [70].


A handful of wastewater plants in the world have further implemented theuse of thermal hydrolysis where concentrated sludge is treated by a combinationof high temperature (133 – 180 °C) and pressure (3 –10 bar) with the aim ofimproving the digestibility of the sludge (The Cambi Process) [71]. However, theinfluence of this process on gas production per unit sludge treated is still not ful-ly documented, and the amount of sludge ending as carbon dioxide due to thetreatment is unknown. The wet oxidation process using alkaline conditions andoxygen in addition to high temperature and pressure has been found to be supe-rior for breaking the lignin associated to hemicellulose and cellulose as ligno-celluloses [72]. The products of the lignin-oxidation (carboxylic acids and phe-nolics) are further found to have highly convertible to methane and carbondioxide (approx. 80 % COD removal) [73]. The pure cellulose and hemicellulosefound in the hydrolysate is expected to give a methane yield corresponding tothe methane potential of the mannouronic sugars. Due to the hydrolytic capa-bility of microbes in the AD process, it is expected that enzyme addition is notneeded for conversion of hydrolysates produced by wet oxidation. However, thisstill needs to be verified along with the optimal way of implementing wet oxida-tion as part of the AD process for materials such as manure containing a highfraction of lignocellulosic material. Furthermore, the economics of this extrastep needs to be evaluated.

Another way to enhance the digestibility of the raw material is by Pulse PowerTM technology developed by Scientific Utilization Inc. in Decatur, AL [74].This equipment incorporates rapid-pulse high-power electric technology ori-

16 B. K. Ahring

Fig. 11. Major goals for anaerobic digestion of today

Sewage

Fertilizer

Animal wastes

Source-sortedhousehold wastes

Food-processingwastes

Crops

Anaerobicdigestion

A. Optimization ofbiogas production• Increasing the

digestibility• Optimizing reactor

configuration• Optimizing process

control• Improving microbial

process

B. Optimization ofeffluent quality

• Inactivation ofpathogens

• Control of chemicalpollutants

Biogenicwastes

More gas

ginally developed for antimissile laser and particle-beam devices, which pro-duces disruptive shock waves in the raw material. The shocks are expected tobreak large molecules into shorter fractions and have been claimed to enhancedestruction of volatile solids by 50 to 100 %. However, a study of the efficiency ofthis method in a full-scale system showed no improvements.

5.2Optimization of Reactor Configuration

The AD process can be conducted in a single-step or multi-step process [47].Continuous processes are generally most favorable when treating large amountsof waste and thermophilic temperatures have the largest potential due to higherreaction rates which corresponds to smaller reactor volumes. Separation of thesolid phase from the liquid phase of manure or sewage sludge is a technical solu-tion which has been well documented during the last years and which can beimplemented both before and after the anaerobic reactor [75 – 77]. Separationwill further allow for optimal design of the process so that the liquid fraction canbe left at the farm, or treated locally in small compact plants, while the solid fac-tion can be transported to a centralized plant for treatment. Especially pigmanure contains very high concentrations of phosphorus and, therefore, largeland areas are needed to use the manure as a fertilizer afterwards. If the solidfraction is removed from the manure, the farm is able to use the liquid fractionon a much smaller land area and pipes can be used for the spreading. Afterdigestion the phosphorus-rich solid fraction is an excellent fertilizer [78]. If theseparation is carried out on fresh manure approximately 70 % of the gas poten-tial will remain in the solids [70].

Recently, we demonstrated that the conversion of organic material in manurecould be increased along with an increase in the over-all biogas yield by using atwo-phase system combining a short hydrolysis step performed at 70 °C fol-lowed by a methane-producing step at 55 °C, both done in continuous stirredtank reactors (CSTRs) (Fig. 12). The performance was compared to a single-phase process in a CSTR reactor with the same over-all retention time, and thefirst estimation showed that the extra gas produced was sufficient to justify theimplementation of an additional reactor and the need for extra heating energy(unpublished).

The possibility of using an immobilized reactor system after a short hydrolyt-ic step during a two-phase conversion of waste such as manure, sewage sludgeand household waste does possess a potential, which needs more attention(Fig. 11).A number of systems such as the up-flow anaerobic sludge blanket reac-tor are in use throughout the world for treatment of industrial wastewaters. How-ever, only limited experience has been obtained from full-scale use of this reactorfor the treatment of solids [79, 80]. Having retention times in the range of hours,the potential is apparent, as much smaller reactors are needed to treat the sameamount of waste. Furthermore, the immobilized system has lower constructionand running costs, no stirring system will be needed. The sludge produced in thesystem can be recirculated back to the hydrolysis step decreasing the final sludgeproduction from the system,and again the phosphorus is concentrated in the sol-


18 B. K. Ahring

Fig.

12.

Diff

eren

t rea

ctor

con

figu

rati

ons

for

anae

robi

c di

gest

ion

ofw

aste

id fertilizer while the liquid fertilizer is rich in nitrogen. The over-all economy is,therefore, improved although a separation is needed between the hydrolysis reac-tor and the immobilized reactor to ensure that the amount of suspended solids inthe influent to the immobilized system is acceptable.

5.3Optimizing Process Control and Stability

Process stability is important for the operation and economy of any AD plant.Imbalance often affects the methanogens in the anaerobic process and leads toa VFA accumulation [44]. It is important to note that some inhibitory com-pounds equally affect all the major groups in the anaerobic digestion process.This is the case for long-chain fatty acids [81, 82] or for phthalate esters such asDEHP [83]. No VFA accumulation was observed when reactors were inhibitedwith DEHP. Inhibitory compounds in waste are, however, generally eitherammonia or sulfide, which are found in high concentrations in some types ofwaste [84–87]. Furthermore, high concentrations of proteins in the incomingwaste can lead to the development of inhibitory concentrations of ammonia andsulfide. For both of these compounds, the toxic effect is dependent on pH andtemperature – the higher the temperature and pH, the higher the toxic effect[88]. Due to the high ammonia concentration, thermophilic digestion of swinemanure has been found to be difficult [89]. Adaptation to an inhibitory com-pound is, however, possible over time and the anaerobic process can work withstable performance but with a lower gas yield as long as the concentration of thetoxic compound is kept relatively constant. Process stability is, however, lost ifthe concentration of the inhibitor is fluctuating as seen in the large-scale biogasplants when treating many types of wastes in different ratios. In immobilizedanaerobic systems, the biomass has generally been found capable of withstand-ing much higher concentrations of inhibitory compounds [90]. This is probablydue to concentration gradients in the biofilm creating niches where themicrobes are protected from toxic concentrations of the inhibitory compounds.The use of a two-phase digestion system is, therefore, expected to show superi-or performance by compared to a one-phase system for waste containing highconcentrations of inhibitors. Increasing the biomass concentration in a biogasreactor by recirculation of the biomass was found to increase the gas yield dur-ing anaerobic digestion of swine waste [91, 92]. In accordance with this, theinhibitory effect of swine manure can be counter-acted by addition of otherwastes such as lipid-containing wastes, which result in a higher biomass con-centration in the reactor besides a dilution of the manure.

Process problems in AD plants are generally difficult to detect before theprocess is severely affected and the gas production decreases. In general theplant operator has very little information about the condition of the process andno instruments inform him when the process is becoming unstable (Fig. 13).

As a result, the plant is often operated with a very low organic loading to pre-vent problems from occurring. A good sensor would, however, make it possibleto optimize the operation and to ensure maximum use of the reactor space with-out having any process failures (Fig. 14).


20 B. K. Ahring

Fig.

13.

Ope

rati

on o

ftod

ay’s

biog

as p

lant

Supp

liers

Stor

age

Tank

Stor

age

Tank

Gas

Brav

e

Opt

imum

Cau

tiou

s

Tim

e

Ope

rato

r


Fig.

14.

Futu

re o

pera

tion

Supp

liers

Stor

age

Tank

Stor

age

Tank

Ope

rato

r

PLC

Reg

istr

atio

nof

dataG

as

Opt

imum

Tim

e

Recently, a sensor has been developed that can measure VFA on-line in bio-gas reactors [93]. This development allows a continuous monitoring of theanaerobic process and with the development of logic control systems it shouldbe possible to improve the economy of AD plants through a more stable andoptimized operation in the future. A number of kinetic models have been developed for anaerobic waste reactors but so far no control algorithm has been developed and tested based on VFA data in addition to the normal dataavailable at the AD plant (flow rates, amount of raw materials, gas production,temperature etc.).A combined sensor and control system can be expected in thefuture.

5.4Improving the Microbial Process and its Efficiency

Improving bioprocesses by implementation of microbial populations withimproved degradation abilities (bioaugmentation) has been known for years.However, only few studies have been done on bioaugmentation and the resultsare inconsistent. To obtain a clear picture of the potential to use specificmicrobes for improvement of the process, it is necessary to follow the fate of themicrobes added to the reactor system over time. Only microbes with the abili-ty to thrive and proliferate in the reactor will be of importance in a longer-termprospective. Molecular techniques are available today for studies of popula-tions in reactor systems and using such techniques we demonstrated that a spe-cific cellulolytic bacterium, present in manure inhabited the reactor [94]. Thesame technique could be used to test for specific added microbes. Compared tocontrols without any pretreatment a more than 20 % increase in the methaneyield was found by incubating separated fibers from cow manure with specificextreme thermophilic xylanolytic microbes for 2 days before the material wasresuspended in the liquid and digested [69]. This finding seems promising inthe context of the two-phase system described above and deserves furtherexamination.

Isolation and characterization of the acetate-utilizing methanogens fromthermophilic manure plants in Denmark showed important differences betweenthe different isolates of Methanosarcina species with respect to temperatureoptimum and growth rates [95]. The strain derived from the best performingthermophilic biogas plant was the acetate-utilizing methanogen with the high-est growth rate and highest temperature optimum.When using this methanogento seed reactors where the organic loading was increased by a sudden additionof lipids to the feed of manure, the seeded reactor was found to be superior toovercome the changes compared to the unseeded reactor, which was inhibitedseverely and accumulated VFA. No major accumulation of VFA was found in theseeded reactor compared to the unseeded reactor, and biomass from this reac-tor had a much higher specific methanogenic activity on acetate than for hydro-gen and formate, which was almost the same in both reactors. The fact that seed-ing had an effect even after several retention times indicates that the addedmethanogen grew in the reactor as further demonstrated using a probe specificto this strain (unpublished). These findings have practical implications and

22 B. K. Ahring

show that better performance can be obtained when lipid-containing waste isintroduced into a biogas reactor operated on manure if the reactor is seededwith a robust acetate-utiling methanogenic strain with a higher growth ratethan the native strain in the system.

6Optimization of Effluent Quality

Besides production of biogas the AD plant produces a residue or effluent with apotential market value. Until now, no applicable standards for these productshave been available and recycling of AD-residues has generally been poorly reg-ulated in most countries. The main issues related to quality management whenrecycling AD-residues is

1. to break the chain of disease transmission by inactivation of pathogens andother biological hazards and

2. to control chemical pollutants (organic and inorganic).

Inactivation of pathogens is increased with increasing temperatures and, there-fore, thermophilic digestion has a much high sanitary effect than mesophilicdigestion.

6.1Inactivation of Pathogens and Other Biological Hazards

Sewage sludge and segregated household wastes are both high-risk wastes thatcan be heavily contaminated with pathogens [96]. Several reviews on pathogensin livestock waste, factors influencing microbial movement and methods forinactivating pathogens have been published [97 – 99] New regulations withrespect to concentrations of pathogens and organic pollutants could potential-ly be threatening to land-disposal of digested material. The regulations madeboth by the US EPA and the European Union demand specific treatmentprocesses before the use of sewage sludge on agricultural land [1, 100]. How-ever, for unrestricted use of digested sewage sludge a further reduction ofpathogens will be required. Several studies found anaerobic digestion to besuperior to aerobic digestion in reducing the density level of pathogens [101].Conventional mesophilic digestion was found to be insufficient for meeting thenew requirements for unrestricted land-use [101, 102]. An increased elimina-tion of pathogens can be achieved by using different treatment processesincluding composting of sewage sludge at high temperatures, exposing thematerial to radiation or specific temperatures for a defined period. Thermo-philic digestion is still not included as a mean for producing a clean effluent.Biosolid A demands that the number of coliforms is less than 1000 per g drysolid, the number of Salmonella is less than 3 per 4 g dry solid, enteric virus andhelmic ova should not be detectable. These restrictions are based on logarith-mic reduction experiments performed in buffer with the respective microbes.However, experience obtained under anaerobic digestion showed that otherparameters than time and temperature are of importance for reduction of


pathogens. The anaerobic environment seems to have an additional effect,which is not accounted for in the biosolid classification [102]. It was demon-strated that the high level of VFA [103], ammonia and sulfide [104] and alkalinepH enhance inactivation of pathogens [105]. Several reports on thermophilicAD of sewage sludge showed that thermophilic digestion is more efficient inreducing the pathogens and pathogen indicators than mesophilic digestion[102, 106–108]. Further improvement of the process by extension of onedigester into a series of several reactors operating at 55 °C [110], application of a two-stage process with an acidogenic reactor operating at 55 °C and 60 °Ccoupled to a 37 °C methanogenic reactor [111], or combinations of thermo-philic pretreatment or posttreatment at 62 °C with conventional mesophilicdigestion [109] all showed that it is possible to make a stabilized sludge fulfill-ing the requirements for Biosolid A. Recent experiments done at TerminalIsland Treatment Plant in Los Angeles clearly showed that Class A biosolid canbe produced by switching the process from mesophilic digestion at around35 °C to thermophilic digestion temperatures at 55 °C [112]. However, for unrestricted use of the effluent a required holding period of one day is neededat 55 °C, which is difficult to meet in a conventional sewage sludge treatmentsystem. By testing the effluent quality and demonstrating that the solid meetsall requirements it is possible to obtain a Class A biosolid classification. Thetesting program, however, has to be repeated on a regular basis to maintain theclassification.

Pathogens are further of major interest when manure from several farms istreated in centralized large-scale biogas plants. When the Danish Action Pro-gram for Large Scale Biogas Plants was implemented more than 10 years ago,hygienic aspects were central as a consequence of transperation of manure fromseveral farms. Therefore, a veterinary program was initiated and this led toimplementation of a number of control functions, which have gained majorinterest and respect throughout the world [96, 113]. In this program the fecalStreptococci (enterococci) were found to be excellent indicator organismsinstead of coliforms (the FS method) during digestion at temperatures up to60 °C. These microbes are present in manure or other materials of intestinal ori-gin as the coliforms, but in contrast to coliforms, they are much more resistantto high temperatures and the anaerobic environments and, therefore, most path-ogenic bacteria, viruses and parasitic eggs will be inactivated long before thesemicrobes. An FS log10 reduction of around 4 and 5 was needed to give an accept-able effluent quality, which basically implies that the AD process is operated atthermophilic conditions or that a high-temperature step is added to a mesophilicreactor [96].

Pathogenic viruses have been identified in sewage sludge, segregated house-hold waste and manure [96]. The absence of enteric viruses showed no correla-tion with porcine parvovirus in a previous study of thermophilic anaerobicdigestion of manure [102]. This indicates that this virus could be a poor indica-tor for human pathogenic viruses. The effect of the AD process on viruses fur-ther depends on the way the virus is found the environment. For instance, it wasfound that a virus in tissue was less sensitive than a free-living virus [114]. Muchmore work is, however, needed to understand the fate of a virus during anaero-

24 B. K. Ahring

bic digestion including the possibility to remove a resistant virus by pretreat-ment such as thermal hydrolysis or wet oxidation.

Besides the potential for pathogens, material of animal origin such as wastefrom slaughterhouses and milled bones from cows and sheep can contain infec-tious elements resulting in transmittable spongiform encephalopathy (TSE). Toreduce the risk of spreading of these diseases the European Commission hasvery recently defined special methods for slaughtering of animals to ensure thatall risk material is removed from the feed-chain [115]. This regulation defineshow the different types of wastes (animal by-products) not used for human con-sumption have to be treated. All waste of animal origin is divided in three cate-gories with different demands. The new regulation demands an approval of allplants treating animal waste including quality control by the plant and society.The banning of meat and bone meal as fodder for animals intended for humanconsumption following the increased number of European cases of bovinespongiform encephalopathy (BSE) has led to investigations of possible and safedisposal methods of the meal. During the discussion of disposal methods,anaerobic digestion followed by utilization of the fertilizer value by spreadingthe digested sludge on arable land was suggested. The idea was, however,abandoned, and instead, a large part of the Danish meat and bone meal is utilized in cement production and is thereby lost from cycling of nutrients.BSE and the variant of Creutzfeldt-Jakob disease (vCJD) are generally consid-ered to be transmitted by the ingestion of proteinaceous agents (prions), whichaccumulate in the brain and spinal cord of infected animals and humans.The disease-causing protein (PrPSc) is an abnormal isomer of a host-encodedprotein (PrP) that has the ability to change the conformation of normal PrP to PrPSc. The infectious PrPSc is, however, considered to be extremely resis-tant to enzymatic degradation, heat, and chemical treatment. Proteases are ineffective in inactivating PrPSc, and bioassays have shown that proteinremained infectious after autoclaving at temperatures up to 138 °C for 60 min[116]. Among the different chemical inactivation methods tested, alkaline treat-ments have so far shown most promise, although they are not completely effec-tive. Complete inactivation might, however, be achieved by combination ofmethods. Based upon one study, in which scrapie-infected hamster brainhomogenate remained infectious after 3 years incubation in soil [117], weassume that only a minor reduction of prions will occur during the AD processand that sufficient pretreatment will be necessary to eliminate prions before theanaerobic reactor.

6.2Control of Chemical Pollutants

Among the chemical pollutants, heavy metals are mainly problematic in wastesof industrial origin and are found in high concentrations in some organic wasteand in sewage sludge from wastewater treatment plants with certain industrialinfluents. Through source reduction and elimination of specific types of wastes,it is generally possible to meet the standards regarding heavy metals for use ofresidues produced by AD.


Agricultural waste can contain persistent organic contaminants such as pes-ticides, antibiotics, and other medicine residues. Industrial wastes, sewagesludge, and household wastes can contain aromatics, aliphatic and halogenatedhydrocarbons, organochlorine pesticides, PCBs, PAHs, phthalates, linear alkylbenzenesulfonates (LAS), nonylphenol and nonylphenol ethoxylates. During ADmost of the water-soluble organic contaminants are degraded to variousdegrees. However, hydrophobic compounds such as high molecular phthalates,PAH, and LAS are tightly bound to the particulate phase and are partly unavail-able for biological conversion [118]. The potential to remove organic pollutantsby pretreatment of sewage sludge by wet oxidation was studied very recently.Unfortunately, these results showed that the conditions suitable for keeping abiogas potential in the waste material resulted in production of high amounts of organic pollutants with a smaller molecular weight than the initial pollutantsindustries (unpublished). Effectively, a complete decontamination demandsincubation at very high temperatures (more than 250 °C) and pressure, whichimplies that the final gas potential is marginal and that the costs are very high.When comparing full-scale mesophilic and thermophilic AD-reactors operatedon the same sewage sludge, it was found that the thermophilic process deliveredan effluent with significant lower concentrations of organic pollutants than theeffluent from the mesophilic reactor [119]. A higher bioavailability due to ahigher solubility of the hydrophobic elements could explain the differencesobserved. Recent experiments indicated that extreme thermophilic processesimprove this reaction further but this needs to be further investigated before anyconclusions can be drawn.

7Conclusions

Aaerobic digestion is an important way of handling waste in society. While theemphasis previously was focused on stabilization of sewage sludge, emphasistoday is focusing on creation of an effluent, which safely can be used as a fertil-izer on farmland. Production of biogas is furthermore gaining more attention,especially for treatment of manure from large-scale animal production. In thispicture other types of organic waste such as wastes from food processing orfrom households will be interesting as a mean of boosting the gas productionand, thereby, the economy of the AD plant. Anaerobic digestion can further addvalue during use of waste or other biomasses for the production of chemicalsand energy and this synergy is expected to be further exploited in the future.

Anaerobic digestion is a mature technology today. However, as demonstratedin this chapter, there is plenty of room for optimization and improvements. Thestandardized CSTR reactor has its limitations and implementation of more effi-cient reactor types such as the immobilized reactor systems has a major poten-tial for treatment of solid waste. Process control on the current AD plant is stillrelying on in- and output data and no information is available on-line for check-ing the state of the process and its performance. The microbiology in AD plantsis normally regarded as a big black box and very few attempts have been madeto control the actual microflora in bioreactors treating waste. Recent research

26 B. K. Ahring

has demonstrated that AD plants within close distance of each other can possessdifferent microfloras with different characteristics. Some microbial strains willadd superior characteristics to the reactor system and this has major implica-tions for the future of AD plants. Most waste is only partly degraded in the ADplant. Improving the digestibility of waste by using physical or chemical pre-treatment methods, which will make the waste more accessible for anaerobicdegradation, is another area with major perspectives.

One of the most promising areas for the future is the use of extreme ther-mophilic digestion within the AD plant. The high temperature process will allowfor better hydrolysis of the solids, for better sanitation and for better removal ofxenobiotics during the treatment process.

Acknowledgement. I would like to thank Zuzana Mladenovska, Hinnerk Hartmann, ThomasIshøy and Peter Westermann for valuable input to this chapter.

8References

1. Iranpour R, Oh S, Kim H, Shao YJ, Hagekhalil A, Schafer P, Stenstrom MK, Ahring BK(2002) Wat Environ Federation in press

2. Zinder SH (1993) Physiological ecology of methanogens. In: Ferry JG (ed), Methanogen-esis. Ecology, physiology, biochemistry & genetics. Chapman & Hall, New York, p 128

3. Schink B (1992) Syntrophism among prokaryotes. In: Balows A, Trüper HG, Dworkin M,Harder W, Schleifer K-H (eds), The prokaryotes. Springer Verlag, Berlin Heidelberg New York p 276

4. Mackie RI, Bryant MP (1981) Appl Environ Microbiol 41:13635. Thiele JH, Zeikus JG (1988) Interactions between hydrogen- and formate-producing bac-

teria and methanogens during anaerobic digestion. In: Erickson LE, Fung DY (eds),Handbook on anaerobic fermentation. Marcel Dekker, New York, p 537

6. Westermann P (1996) World J Microbiol Biotechnol 12:4977. Schnürer A, Zellner G, Svensson BH (1999) FEMS Microbiol Ecol 29: 2498. Schnürer A, Houwen FP, Svensson BH (1994) Arch Microbiol 162:709. Zinder SH, Koch M (1984) Arch Microbiol 138:263

10. Lee MJ, Zinder SH (1988) Appl Environ Microbiol 54:145711. Ahring BK, Ibrahim A, Mladenovska Z (2002) FEMS Microbiology Ecology submitted12. Ahring BK, Ibrahim AA, Mladenovska Z (2001) Wat Res 35:244613. Petersen SP, Ahring BK (1991) FEMS Microbiol Ecol 86:14914. Ahring BK (1995) Antonie van Leeuwenhoek 67:9115. Westermann P, Ahring BK, Mah RA (1989) Appl Environ Microbiol 55:51416. Ahring BK, Schmidt JE, Winther-Nielsen M, Macario AJL, Conway de Macario E (1993)

Appl Environ Microbiol 59:253817. Lee MJ, Zinder SH (1988) Appl Environ Microbiol 54:12418. Schnürer A, Schink B, Svensson BH (1996) Int J Syst Bacteriol 46:114519. Wiegel J (1990) FEMS Microbiol Rev 75:15520. Lettinga G (1995) Antonie van Leeuwenhoek 67:321. van Lier JB, ten Brummeler E, Lettinga G (1993) J Ferment Bioeng 76:14022. Kettunen RH, Rintala JA (1997) Appl Microbiol Biotechnol 48:57023. Kotsyurbenko OR, Nozhevnikova AN, Kalyuzhnyi SV, Zavarzin GA (1993) Microbiology

62:46224. Zeikus JG, Winfrey MR (1976) Appl Environ Microbiol 31:9925. Westermann P (1994) FEMS Microbiol Ecol 13:29526. Varel VH, Hashimoto AG, Chen YR (1980) Appl Environ Microbiol 40:217


27. Uemura S, Tseng I-C, Harada H (1995) Environ Technol 16:98728. Rintala J, Lepistö S, Ahring B (1993) Appl Environ Microbiol 59:174229. Lepistö R, Rintala J (1996) Biores Technol 56:22130. van Lier JB (1996) Antonie van Leeuwenhoek 69:131. Zinder SH (1990) FEMS Microbiol Rev 75:12532. Wiegel J (1992) The obligatory anaerobic thermophilic bacteria. In: Kristjansson JK (ed),

Thermophilic bacteria. Boca Raton, CRC Press, p 10533. Ahring BK (1994) Wat Sci Tech 30:24134. Ahring BK (1992) Proceedings from International conference Thermophiles.Science and

Technology. Reykjavik, Iceland, p 13035. Chen M (1983) Appl Environ Microbiol 45:127136. Ahring BK, Mladenovska Z, Iranpour R, Westermann P (2002) Wat Sci Tech 45:29337. Wiegant WM (1986) PhD Thesis, Landbouwhogeschool Wageningen, Netherlands38. Dinsdale RM, Hawkes FR, Hawkes DL (1997) Wat Res 31:16339. Mackie RI, Bryant MP (1995) Appl Microbiol Biotechnol 43:34640. Sørensen AH, Ahring BK (1993) Appl Microbiol Biotechnol 40:42741. Nopharatana A, Clarke WP, Pullammanappallil PC, Silvey P, Chynoweth DP (1998) Biores

Technol 64:16942. Zheng D, Raskin L (2000) Microb Ecol 39:24643. van Lier JB, Grolle KCF, Stams AJM, Conway de Macario E, Lettinga G (1992) Appl Micro-

biol Biotechnol 37:13044. Ahring BK, Sandberg M, Angelidaki I (1995) Appl Microbiol Biotechnol 43:55945. Angelidaki I, Ahring BK (1995) Antonie van Leeuwenhoek 68:28546. Griffin ME, McMahon KD, Mackie RI, Raskin L (1998) Biotechnol Bioeng 57:34247. Pohland FG, Ghosh S (1971) Environ Lett 1:25548. Krugel S, Hamel K, Ahring BK (2002) WEF’s 16th Annual Residuals and Biosolids

Management Conference. Austin, Texas, USA, March 3–6, 200249. Verstraete W, de Beer D, Pena M, Lettinga G, Lens P (1996) World J Microbiol Biotechnol

12:22150. Verstraete W, Vandevivere P (1999) Critical Reviews Environ Sci Technol 28:15151. Qureshi MA, Kharbanda VP (1983) J Scient Ind Res 42:59752. Day DL, Chen TH, Anderson JC, Steinberg MP (1990) Biomass 21:8353. Driessen W, Yspeert P (1999) Wat Sci Tech 40:22154. Seghezzo L, Zeeman G, van Lier JB, Hamelers HVM, Lettinga G (1998) Biores Technol

65:17555. Jetten MSM, Wagner M, Fuerst J, van Loosdrecht M, Kuenen G, Strous M (2001) Current

Opinion in Biotechnology 12:28356. Van Loosdrecht MCM, Jetten MSM (1998) Wat Sci Tech 38:157. Ahring BK, Angelidaki I, Johansen K (1992) Wat Sci Tech 25:31158. Tafdrup, S (1992) Proceedings from Seventh International Symposium on Anaerobic

digestion, 23–27 January, 1994, Cape Town, South Africa, p 46059. Mæng H, Lund H, Hvelplund F (1999) Appl Energy 64:19560. Dagnall S (1995) Biores Technol 52:27561. Hammond G (1993) Biorecovery 2:14162. Lusk P (1999) BioCycle 40:5263. Rintala JA, Järvinen KT (1996) Waste Management Research 14:16364. Rosenwinkel K-H, Meyer H (1999) Wat Sci Tech 40:10165. Hartmann H, Møller HB,Ahring BK (2002) Proceedings from VII Latin American Work-

shop and Symposium on Anaerobic Digestion. Yucatán, México, October 200266. Clausen A (2001) PhD Thesis, Technical University of Denmark, Lyngby, Denmark67. Hartmann H, Angelidaki I, Ahring BK (2001) Proceedings from 9th World Congress of

Anaerobic Digestion, Antwerpen – Belgium, Sept 2–6, 2001, p 30168. Scherer PA, Vollmer G-R, Fakhouri T, Martensen S (2000) Wat Sci Tech 41:8369. Angelidaki I, Ahring BK (2000) Wat Sci Tech 41:18970. Hartmann H, Angelidaki I, Ahring BK (2000) Wat Sci Tech 41:145

28 B. K. Ahring

71. Kepp U, Machenbach I, Weisz N, Solheim OE (2000) Wat Sci Tech 42:8972. Ahring BK, Jensen K, Nielsen P, Bjerre AB, Schmidt AS (1996) Biores Technol 58:10773. Torry-Smith M, Sommer P, Ahring BK (2003) Biotechnol – Bio Wat Res submitted74. Greene H (1995) Anaerobic digester process enhancement by pulse power treatment.

Scientific Utilization, Inc., Decatur, AL75. Møller HB, Sommer SG, Ahring BK (2002) Biores Technol submitted76. Kalyuzhnyi S, Fedorovich V, Nozhevnikova A (1998) Biores Technol 65:22177. Kalyuzhnyi S, Sklyar V, Fedorovich V, Kovalev A, Nozhevnikova A, Klapwijk A (1999) Wat

Sci Tech 40:22378. Worley JW, Das KC (2000) Appl Eng Agri 16:55579. Aoki N, Kawase M (1991) Wat Sci Tech 23:114780. Kübler H, Schertler C (1994) Wat Sci Tech 30:36781. Angelidaki I, Ahring BK (1992) Appl Microbiol Biotechnol 37:80882. Koster IW, Cramer A (1987) Appl Environ Microbiol 53:40383. Alatriste-Mondragon F, Iranpour R, Ahring BK (2002) Wat Res in press84. Sandberg M, Ahring BK (1992) Appl Microbiol Biotechnol 36:80085. Omil F, Mendez R, Lema JM (1996) Water SA 22:17386. Athanassopoulos N, Kouinis J, Papadimitriou A, Koutinas AA (1989) Biological Wastes

30:5387. Daoming S, Forster CF (1994) Environ Technol 15:28788. Koster IW, Koomen E (1988) Appl Microbiol Biotechnol 28:50089. Hansen KH, Angelidaki I, Ahring BK (1998) Wat Res 32:590. Speece RE. (1996) Anaerobic biotechnology for industrial wastewaters. Archae Press,

Nashville, Tennessee, USA91. Hansen KH, Angelidaki I, Ahring BK (1999) Wat Res 33:180592. Boopathy R (1998) Biores Technol 64:193. Pind PF, Angelidaki I, Ahring BK (2002) Biotechnol Bioengineering submitted94. Mladenovska Z, Ishøy T, Mandiralioglu A, Westermann P, Ahring BK (2001) Proceedings

from International conference Anaerobic Digestion 2001, 2–6 September 2001, Antwer-pen, Belgium, p 183–188

95. Mladenovska Z, Ahring BK (2000) FEMS Microbiol Ecol 31:22596. Bendixen HJ (1999) IEA Bioenergy Workshop. Hygienic and environmental aspects of

anaerobic digestion: Legislation and experience in Europe, Stuttgart 29–31 March 1999,p 27

97. Mawdsley JL, Bardgett RD, Merry RJ, Pain BF, Theodorou MK (1995) Appl Soil Ecol 2:198. Pell AN (1997) J Dairy Sci 80:267399. Turner C, Burton CH (1997) Biores Technol 61:9

100. Aitken MD, Mullennix RW (1992) Wat Environ Res 64:915101. Ponugoti PR, Dahab MF, Surampalli R (1997) Wat Environ Res 69:1195102. Lund B, Jensen VF, Have P, Ahring B (1996) Antonie van Leeuwenhoek 69:25103. Kunte DP, Yeole TY, Chiplonkar SA, Ranade DR (1998) J Appl Microbiol 84:138104. Arridge H, Oragui JI, Pearson HW, Mara DD, Silva SA (1995) Wat Sci Tech 31:249105. Carrington EG, Pike EB, Auty D, Morris R (1991) Wat Sci Tech 24:377106. Watanabe H, Kitamura T, Ochi S, Ozaki M (1997) Wat Sci Tech 36:25107. Nielsen B, Petersen G (2000) Wat Sci Tech 42:65108. Duarte EA, Mendes B, Oliveira JS (1992) Wat Sci Tech 26:2169109. Cheunbarn TP, Krishna R (2000) J Environ Engineering 126:796110. Krugel S, Nemeth L, Peddie C (1998) Wat Sci Tech 38:409111. Huyard A, Ferran B, Audic J-M (2000) Wat Sci Tech 42:41112. Iranpour R, Shao YJ, Stenstrom M, Ahring BK (2002) Wat Environ Res in press113. Bendixen HJ (1995) Wat Sci Tech 30:171114. Ahring BK, Lund B, Jungersen G, Have P, Frøkjær Jensen V (1995) Modelstudier

vedrørende overlevelse af virus i gyllebaseret biomasse under udrådning i laboratorie-skala biogasanlæg. Smitstofreduktion i biomasse. Danish Veterinary Service, Frederiks-berg. Vol II: Rep. no. 10


115. Regulation of the European Parliament and of the Council laying down health rules con-cerning animal by-products not intended for human consumption, 2001

116. Taylor DM (1998) Journal of Food Safety 18:265117. Brown P, Gajdusek DC (1991) Lancet 337:269118. Ejlertsson J, Alnervik M, Jonsson S, Svensson BH (1997) Environ Sci Tech 31:2761119. Alatriste-Mondragon F, Ahring BK (2002) Wat Res submitted

Received: March 2002

30 B. K. Ahring: Perspectives for Anaerobic Digestion

Metabolic Interactions Between Methanogenic Consortia and Anaerobic Respiring Bacteria

A.J.M. Stams 1 · S. J.W.H. Oude Elferink 2 · P. Westermann 3

1 Wageningen University and Research Centre, Laboratory of Microbiology,Hesselink van Suchtelenweg 4, 6703 CT Wageningen, The Netherlands.E-mail: [email protected]

2 ID TNO Animal Nutrition, Edelhertweg 15, P.O. Box 65, 8200 AB, The Netherlands.E-mail: [email protected]

3 Department of Environmental Microbiology and Biotechnology, The Technical Universityof Denmark, Building 227, 2800 Lyngby, Denmark. E-mail: [email protected]

Most types of anaerobic respiration are able to outcompete methanogenic consortia for com-mon substrates if the respective electron acceptors are present in sufficient amounts. Further-more, several products or intermediate compounds formed by anaerobic respiring bacteriaare toxic to methanogenic consortia. Despite the potentially adverse effects, only few inorgan-ic electron acceptors potentially utilizable for anaerobic respiration have been investigatedwith respect to negative interactions in anaerobic digesters. In this chapter we review com-petitive and inhibitory interactions between anaerobic respiring populations and methano-genic consortia in bioreactors. Due to the few studies in anaerobic digesters, many of our dis-cussions are based upon studies of defined cultures or natural ecosystems.

Keywords. Competition, Sulfate reduction, Denitrification, Acetogenesis, Inhibition

1 Introduction . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 32

2 Metabolic Interactions in Methanogenic Bioreactors . . . . . . . . 32

2.1 Competitive Interactions . . . . . . . . . . . . . . . . . . . . . . . . 322.1.1 Kinetic Competition . . . . . . . . . . . . . . . . . . . . . . . . . . 352.1.2 Thermodynamic Competition . . . . . . . . . . . . . . . . . . . . . 362.2 Inhibitory Interactions . . . . . . . . . . . . . . . . . . . . . . . . . 36

3 Competition . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 37

3.1 Competition in the Presence of Oxygen . . . . . . . . . . . . . . . . 373.2 Competition Between Nitrogen Reducers and Methanogenic

Consortia . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 383.3 Competition Between Manganese and Iron Reducers and

Methanogenic Consortia . . . . . . . . . . . . . . . . . . . . . . . . 393.4 Competition Between Sulfate-Reducing and Acetogenic Bacteria

and Methanogenic Consortia . . . . . . . . . . . . . . . . . . . . . 403.4.1 Competition for Hydrogen . . . . . . . . . . . . . . . . . . . . . . . 413.4.2 Competition for Acetate . . . . . . . . . . . . . . . . . . . . . . . . 433.4.3 Competition for Methanol . . . . . . . . . . . . . . . . . . . . . . . 45

CHAPTER 6


3.4.4 Competition for Organic Acids and Ethanol . . . . . . . . . . . . . 463.4.5 Competition for Sulfate . . . . . . . . . . . . . . . . . . . . . . . . 483.5 Competition Between Sulfate-Reducers and Acetogens

in the Absence of Sulfate . . . . . . . . . . . . . . . . . . . . . . . . 49

4 Inhibition . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 50

5 Conclusion . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 52

6 References . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 53

1Introduction

Very few environments exist in which only one population of microorganismsthrives or where populations of microorganisms do not affect each other eitherpositively or negatively. As discussed in Chap. 1, anaerobic ecosystems suchas methanogenic bioreactors are characteristic by their complex food-chainsand the close symbiotic relationship between the different links in the chain,and are often exemplified as classical symbiotic ecosystems in which organismsconsume the products of the preceding link in the chain, rather than consum-ing each other. The symbiosis between hydrogen-producing and hydrogen-consuming microorganisms is confined to a narrow range of hydrogen partialpressures outside which the reactions become thermodynamically unfavorablefor one or the other part of the relationship. This can be caused by overloadingwith easily degradable compounds or by unintentional influence of inhibitorycompounds. Compounds inhibiting methane production in a digester mightexert their action either by direct inhibition of microbes in the anaerobic degra-dation chain or by stimulating microorganisms present in the digester to com-pete with methanogens or preceding links leading to reduced methane produc-tion and other unfavorable effects such as corrosion [1]. In this chapter wewill discuss various types of direct and indirect competitive interactions be-tween methanogenic consortia and anaerobic respiring bacteria in anaerobicbioreactors.

2Metabolic Interactions in Methanogenic Bioreactors

2.1Competitive Interactions

Competition between two or more populations of microorganisms is a negativerelationship in which the different populations often are adversely affected withrespect to their survival and growth. Also competition is considered the mostimportant interaction among organisms, and is one of the major responsiblecauses of the selection pressure leading to the evolution of species.

32 A.J.M. Stams et al.

The competitive interactions among anaerobic microorganisms can be roughlydivided into kinetic competition and thermodynamic competition (Fig. 1).Kinetic competition refers to the determination of competitive capabilities bykinetic measurements of microbial growth,although the underlying mechanismfor the observed effects might be thermodynamic. Thermodynamic competi-tion means that one organism is capable of growing at and maintaining a sub-strate concentration below the minimum concentration for uptake (thresholdconcentration) of other organisms due to a higher energy yield in the conver-sion of the compound.

In anaerobic fermentation of organic compounds, numerous pathways andcombinations of pathways are used leading to different energy yields. However,since anaerobic fermentation is internally optimized in the cells to gain a maxi-mum energy yield and an optimal redox balance [2, 3] the energetic outcome isoften the same. This has the consequence that fermentative competitive interac-tions are mainly of kinetic character. Most of the studies which have examinedcompetition between anaerobic fermenting bacteria have focused on gastro-intestinal systems [4] and very little is known on this type of competitive inter-action in anaerobic digestion processes. Apart from interactions between fer-menting sulfate-reducing bacteria and acetogenic bacteria, we will not discussthis topic in this chapter.

Metabolic Interactions Between Methanogenic Consortia and Anaerobic Respiring Bacteria 33

Fig. 1. Model of kinetic and thermodynamic competition among sulfate-reducing bacteria andmethanogenic Archaea

In contrast to aerobic conditions where most heterotrophic microorganismsutilize oxygen as a terminal electron acceptor and in most cases follow the samemetabolic pathway ending in complete mineralization of the organic com-pounds into CO2 and H2O, the biochemical diversity of anaerobic microbialcommunities is huge. A large number of electron acceptors can be used by dif-ferent anaerobic organisms in anaerobic respiration processes (Table 1). Themost important inorganic electron acceptors are Mn4+,Fe3+,NO3

– ,SO42– and CO2 .

The respiration processes where these acceptors are used are normally separat-ed either in space or time. This is due to a different energy outcome of theprocesses according to the Gibbs equation: DG0¢ = –n · F · DE0¢ in which DG0¢ is theGibbs free energy at pH = 7; n is the number of electrons transferred in the oxi-dation-reduction reaction; F is Faraday’s constant (96.490 kJ/V) and DE0¢ is theredox potential (E0¢) of the electron-accepting reaction minus the redox poten-tial of the electron-donating reaction. From this equation it is obvious that thelarger the difference is between the redox potentials of the half-reactions, thelarger is the amount of energy available to the organism performing the reac-tion. The consequence is a hierarchy, which often resembles the order seen inTable 1.

In most environments, some of the respiration processes do not occur, or onlyoccur to a minor extent, due to the lack or exhaustion of available electronacceptors. The energy available to a respiring organism is not only dependentupon the difference in redox potential between electron donor and acceptor.Also concentrations of the reactants and temperatures deviating from standardconditions affect the energy outcome according to the Nernst equation DG =DG0 + RT · ln [B]/[A] in which DG0 is the change in Gibbs free energy understandard conditions, R is the gas constant, T is temperature and [B] and [A] arethe concentrations of the two components of the reaction A ¤ B. According tothe respiration hierarchy, sulfate reduction excludes methanogenic utilizationof common substrates, which is verified in high-sulfate environments such asmarine sediments [5]. However in, e.g., freshwater sediments, the two processescan coexist or even be dominated by methanogenesis due to equilibrium dis-placements caused by low sulfate concentrations making sulfate reduction ther-modynamically less favorable than methane production [6].


Table 1. The respiration hierarchy

Acceptor Product E0¢ (V)

Oxygen O2 Water H2O +0.82Manganic ion Mn4+ Manganous ion Mn2+ +0,80Ferric ion Fe3+ Ferrous ion Fe2+ +0.77Nitrate NO3

– Nitrogen N2 +0.76Selenate SeO4

2– Selenite SeO32– +0.48

Arsenate AsO43– Arsenite AsO3

3– +0.14Sulfate SO4

2– Sulfide HS– –0.22Carbon dioxide CO2 Methane CH4 –0.24Carbon dioxide CO2 Acetate CH3COO– –0.29

2.1.1Kinetic Competition

This is the classical competitive interaction, the theory of which has been estab-lished in studies of defined cultures in chemostats [7, 8]. According to kine-tically-based competition models, the outcome of interactions between twomicroorganisms competing for the same growth-limiting substrate can be pre-dicted from the relationship between substrate concentration and the specificgrowth rate (µ) according to the Monod equation: µ = µmax ¥ S/Ks + S. Twotypical relationships can be observed in studies of competitive interactions(Fig. 2a, b).

In Fig. 2a, organism I will grow faster than organism II at any substrate con-centration, while the outcome in Fig. 2b is dependent upon the substrate con-centration. The pattern seen in Fig. 2a is typical of organisms utilizing differentelectron acceptors with different energy yields for the oxidation of a commonsubstrate, since the energy yield is higher for the electron acceptor with thehighest redox potential at all electron donor concentrations. The pattern seen inFig. 2b is typical for organisms utilizing the same metabolism but having differ-ent ecological strategies. In natural ecosystems, such as sediments, the concen-tration of nutrients needed to support growth is often very low. Among theorganisms using the same type of metabolism under these conditions, type II inFig. 2b having a high substrate affinity (low Ks) and a relatively low maximalgrowth rate (µmax) will normally dominate. This group is assigned to “K selec-


Fig. 2. Growth rate as a function of substrate concentration in two different scenarios (a andb). a represents two organisms with different energy metabolism, I having the highest energyyield. b represents two organisms with the same energy metabolism, but with different eco-logical strategies. I is assigned to “r” selection while II is assigned to “K” selection

Gro

wth

rat

e

Substrate concentration

a

b

tion” which refers to organisms that can most effectively utilize the resourcesavailable [9]. In gastrointestinal environments and anaerobic bioreactors, oppor-tunistic types of organisms (type I) will normally dominate, since type II has alonger doubling time than the retention time of the system. This group isassigned to “r selection” referring to a high potential r value (rate of populationgrowth/individual) [9].

2.1.2Thermodynamic Competition

In natural environments, the substrate concentration for most organisms isnormally well below Ks . For all organisms, there is a specific minimum concen-tration of substrate necessary to gain conservable energy. This minimal “quan-tum” of energy, which can be conserved, corresponds to the energy needed fortranslocation of 1 proton. The phosphorylation of ATP to ADP has a DG¢ of+49 kJ/mol corresponding to 60–70 kJ/mol when compensating for energy con-servation efficiency [10]. Since 3 protons are needed in the phosphorylation ofADP to ATP, we can assume that the smallest amount of energy which can beconserved is 1/3 of the phosphorylation energy, corresponding to a minimumDG¢ of –20 kJ/mol. Inserting this value and DG¢0 for different respiration process-es in the Nernst equation, the substrate concentration yielding the minimumamount of energy (the threshold concentration) can be calculated for each processunder the prevailing conditions of the specific ecosystem. Several authors haveshown that organisms utilizing electron acceptors with higher redox potentialscan maintain electron donor concentrations below the threshold for uptakeof organisms utilizing electron acceptors with lower redox potentials [11–13].Other studies have shown that significant differences in threshold values forcommon substrates also can be found among species utilizing the same type ofmetabolism [14].

2.2Inhibitory Interactions

Several compounds, which serve as electron donors to respiring bacteria, mightinhibit members of the methanogenic consortia. Also some products fromanaerobic respiration might affect the activity of these consortia. The modes ofaction can be indirect by increasing the redox potential to levels that interferewith the biochemistry of the anaerobic microorganisms, or direct by chemicalreaction with proteins or other cell constituents.

It has been assumed that many anaerobic microorganisms have specificdemands for low redox potentials in their environment to make their energymetabolism thermodynamically possible [15]. This conception has since beenmoderated and several reports have shown that the parameters controllinggrowth of most anaerobes is the oxygen concentration and only to a lesserdegree the redox potential of the environment. This has been demonstrated instudies of fermentative rumen bacteria, but also in studies of microorganismsconsidered extremely sensitive to aerobic conditions [16]. Fetzer and Conrad


[17] have, for instance, demonstrated that methane production in axenic cul-tures of Methanosarcina barkeri proceeded at normal rates in oxygen-freemedia where the redox potential was elevated to +420 mV.

The direct inhibition of methanogenic consortia by electron acceptors ismediated by several mechanisms. Oxygen is toxic to all obligatory anaerobicmicroorganisms. Many anaerobes are rich in flavine enzymes, and may also con-tain quinones and iron-sulfur proteins, which can react spontaneously with oxy-gen to yield hydrogen peroxide, superoxide and hydroxyl radicals. Since mostanaerobes lack peroxidase, catalase and superoxide dismutases, which destroythe reactive oxygen species,damage of essential cell components can occur uponoxygen exposure. Superoxide dismutase has, however, been demonstrated insome anaerobic microorganisms. Kirby et al. [18] have, for instance, character-ized a superoxide dismutase from the obligatory anaerobe Methanobacteriumbryantii. Other electron acceptors, such as oxidized nitrogen and sulfur species,have also been shown inhibitory to anaerobic microorganisms.

Although the metabolism of these electron acceptors is competitive to anaer-obes utilizing electron acceptors with a more negative redox potential, the reduc-tion of the inhibitory compounds might lead to the production of less inhibitorycompounds and, hence, relieve the inhibition. In some cases, however, the prod-ucts of anaerobic respiration are more toxic than the parent compounds. Thiswill be discussed in details in the next chapters.

3Competition

3.1Competition in the Presence of Oxygen

Although oxygen is the naturally occurring electron acceptor yielding the high-est amount of energy leading to effective outcompetition of anaerobic microor-ganisms, oxygen respiration and anaerobic metabolism are mutually exclusiveprocesses mainly due to the toxicity of oxygen, which can be observed in allaerobic environments. Most facultatively aerobic microorganisms capable ofanaerobic respiration suppress these processes in favor of oxygen respirationwhen oxygen is present. Only environments in which rapid changes betweenoxic and anoxic conditions occur, such as alternating sludge treatment basins,favor constitutively anaerobic respiring bacteria [19]. In true oxic environments,anaerobic processes are normally only occurring in organic-rich micro- andmacro-niches, where oxygen is depleted at a higher rate than it diffuses into theniche.

Oxygen is normally excluded in anaerobic digestion processes,and only smallamounts might enter the reactors together with, e.g., strongly aerated substrates[20]. Due to the low solubility of oxygen, this does normally not pose a problemto the anaerobic microorganisms in the digester and is rapidly scavenged byfacultative bacteria. Kato et al. [21] demonstrated a high oxygen tolerance ofmethanogens in granular sludge due to mainly oxygen consumption by faculta-tively anaerobic bacteria metabolizing easily degradable substrates.


3.2Competition Between Nitrogen Reducers and Methanogenic Consortia

From an immediate evaluation of redox potentials of methanogenesis and nitratereduction, it is obvious that nitrate reducers should outcompete methanogensdue to the much higher energy yield of nitrate respiration. This has been veri-fied in a few natural environments [22]. Under most circumstances, however, theeffects of nitrogen oxides to anaerobic digestion are ambiguous and very com-plex, and to our knowledge no certain verification of competition in anaerobicdigesters in which inhibition has been excluded has been published so far.

Denitrification and methanogenesis are performed by microbial populationseach requiring their distinct environmental conditions. Most true denitrifiersare facultatively anaerobic bacteria utilizing either oxygen respiration or deni-trification as sole energy source. If none of these metabolisms are possible dueto the lack of appropriate electron acceptors, the bacteria will probably notthrive in the digester. Instead, fermentative bacteria reduce oxidized nitrogenspecies for dissimilatory electron dissipation. The product is either nitrite orammonia, and only the reduction of nitrate to nitrite is energy yielding. The fur-ther reduction to ammonia is considered non-energy yielding and hence with-out competitive value. Several authors have shown that high carbon to nitrogenratios which are normally found in anaerobic digesters favor dissimilatorynitrate reduction to ammonia [23], while others [24] found that a high COD/NO3

–

did not favor dissimilatory reduction of nitrate to ammonia. The nature of thecarbon source has also been shown to influence whether nitrate is reduced toammonia or dinitrogen [25]. When glucose or glycerol was added as carbonsource, 50% of the nitrate was reduced to ammonium, while 100% was denitri-fied completely in the presence of acetate or lactate.

Several authors have demonstrated that denitrification and methanogenesiscan proceed in the same reactor as long as the two processes are spatially sepa-rated. Hendriksen and Ahring [26] found that denitrification took place in thebottom of an upflow anaerobic sludge blanket reactor utilizing all availablenitrate. Methanogenensis occurred in the uppermost part of the reactor, whichwas depleted from nitrogen oxides. In a mixed culture system of denitrifyingand methanogenic sludge in a digester enriched with methanol, Chen andLin [27] observed no competitive interactions between the two communities.Methanogenesis was, however, inhibited as long as nitrate or nitrite was presentin the reactor. Percheron et al. [24] studied methanogenesis and nitrate reduc-tion in an anaerobic digester fed with sulfate-rich wastewater. Sulfate reductionwas inhibited by the presence of nitrate while methanogenesis proceeded untilthe onset of denitrification and production of nitrite after which it also wasinhibited. Sulfide served as electron donor for some of the denitrifying bacteria.When sulfide was precipitated by ferrous iron, only dissimilatory nitrate reduc-tion occurred with no nitrite production. This led to a stimulation of metha-nogenesis compared to a control digester, probably due to extensive acetateproduction by the dissimilatory nitrate reducers. No specific competitive inter-actions between methanogens and denitrifiers were verified in this study either.Clarens et al. [28] studied the effects of nitrogen oxides and denitrification on a


pure culture of an aceticlastic methanogen (Methanosarcina mazei) and mixedcultures of M. mazei and a denitrifying bacterium (Pseudomonas stutzeri). Itwas demonstrated that the observed cessation of methanogenesis upon nitrateaddition was a consequence of inhibition by denitrification products (NO2

– ,N2O) rather than competition by the denitrifying bacterium. The authors foundthat 50 mM NO3

– inhibited methanogenesis from acetate by 65% while 0.18 mMNO2

– and 0.32 mM N2O were almost completely inhibitory.In a study of the effects of nitrogen oxides on methanogenesis and other

metabolic activities in an anoxic rice-field soil, Klüber and Conrad [22] tried toresolve inhibition and competition among nitrogen-respiring bacteria andmethanogens. The addition of nitrate, nitrite, nitrous oxide and nitric oxide allresulted in an immediate arrest of methanogenesis until the nitrogen oxideswere consumed. Methanogenesis then resumed at a similar or lower rate. Noneof the nitrogen oxides affected acetate concentrations negatively while nitrate,nitrite and nitrous oxide additions temporarily reduced hydrogen partial pres-sures to low exergonic or even endergonic values for methanogenesis. Sincemore than 70% of the methane produced was derived from acetate, Klüber andConrad’s results indicate that toxicity rather than competition is responsiblefor the inhibition observed. When nitrate, nitrite or nitrous oxide were added,sulfate or/and ferric iron concentrations increased, probably as respirationproducts of sulfide and ferrous iron oxidation coupled to denitrification. Themaintenance of low hydrogen partial pressures might, therefore, be due to theactivity of iron or sulfate reducing bacteria rather than denitrifying bacteria.

The mechanisms of nitrogen oxide inhibition of methanogenesis in anaero-bic digesters can be considered far from solved, and is probably a complexmechanism composed of toxicity, competition, and indirect stimulation of otherrespiring bacteria by oxidation of reduced electron acceptors such as ferrousiron and sulfide.

3.3 Competition Between Manganese and Iron Reducers and Methanogenic Consortia

Both manganic ions [Mn(IV)] and ferric ions [Fe(III)] can act as potent electronacceptors in anaerobic respiratory processes carried out by a variety of microor-ganisms coupled to the oxidation of organic and inorganic compounds. Mn(IV)and Fe(III) can also be reduced in non-enzymatic chemical reactions underanaerobic conditions, and much of the effort in earlier studies of respiration ofthe two compounds was devoted to the separation of non-biological from bio-logical reductions [29]. The isolation of numerous bacteria capable of Fe(III)and Mn(IV) reduction and properly designed experiments with environmentalsamples have, however, unambiguously verified this type of bacterial respira-tion. Several authors have shown that methanogenesis and other terminalanaerobic processes can be outcompeted by ferric- and manganic-reducing bac-teria due to their maintenance of acetate concentrations and H2 partial pressuresbelow the threshold of methanogenic Archaea and sulfate-reducing bacteria[12]. Although respiration with Mn(IV) or Fe(III) is thermodynamically morefavorable than sulfate reduction or methanogenesis, several authors have shown


that Mn(IV) and Fe(III) respiration is less efficient with crystalline than withamorphous forms of the two electron acceptors [30, 31]. Lovley and Phillips [12]showed that sulfate reduction and methanogenesis are only inhibited by Fe(III)-reducing bacteria when Fe(III) is in an amorphous form. In a study of anaerobicrespiration processes in flooded soils, Peters and Conrad [32] found that Mn(IV),Fe(III), and sulfate reduction proceeded simultaneously, possibly due to thecrystalline structures of the Mn and Fe minerals in the soils.

Since oxidation of short-chain fatty acids and H2 are the main electron-donating processes of both Fe(III) and Mn(IV) reduction, one could expect thatthese two electron acceptors could play a significant role in anaerobic digestionwhen present in high concentrations. Besides direct competitive interactions,Mn(IV) has been shown to act as an electron acceptor in the oxidation of ele-mental sulfur (S0) to sulfate catalyzed by sulfate-reducing bacteria [33]. Thiscould lead to the stimulation of sulfate reduction upon exhaustion of Mn(IV).

Very few investigations have, however, been carried out regarding the effectsof Fe(III) and Mn(IV) on anaerobic digestion. One major reason for this couldbe the very low contents of iron and manganese normally found in wastewater.In average wastewater with a BOD of 290 g O2/m3, the typical iron and manganeseconcentrations have been estimated to 3.5 mg/g BOD and 0.35 mg/g BOD,respectively [34]. In a study of the effect of ferric chloride addition to anaerobicsludge digesters to precipitate struvite (MgNH4PO4 · 6 H2O), Mamais et al. [35]added FeCl3 at doses ranging from 0 to 20.5 mM Fe/L. A slight increase in gasproduction was observed upon FeCl3 addition, but no other effects were found.

Further investigations are needed with respect to these two electron accep-tors to clarify their actual and potential effects on different anaerobic digestionprocesses.

3.4Competition Between Sulfate-Reducing and Acetogenic Bacteria and Methanogenic Consortia

In environments where sulfate is present, sulfate-reducing bacteria will competewith methanogenic consortia for common substrates. Direct competition willoccur for substrates like hydrogen, acetate and methanol. Compared with metha-nogens, sulfate-reducing bacteria are much more versatile than methanogens.Compounds like propionate and butyrate, which require syntrophic consortia inmethanogenic environments, are degraded directly by single species of sulfate-reducing bacteria. The physiology of sulfate-reducing bacteria has been reviewedbefore by Widdel [36],Widdel and Hansen [37] and Colleran et al. [38], while thephysiology of methanogenic consortia was reviewed by Stams [3], Schink [39]and Verstraete et al. [40]. Some key reactions in anaerobic environments are list-ed in Table 2.

Kinetic properties of sulfate-reducers, methanogens, and acetogens can beused to predict the outcome of the competition for these common substrates [6,41–44]. For bacteria growing in suspension, Monod kinetic parameters such asthe half-saturation constant (Ks) and the specific growth rate (µmax) can be used.When bacterial growth is negligible, as is often the case in reactors with a dense


biomass concentration, Michaelis-Menten kinetics may be used to predict whichtype of organism has the most appropriate enzyme systems to degrade sub-strates. Therefore, both the Vmax/Km and the µmax/Ks ratio gives an indication ofthe outcome of competition at low substrate concentrations [42].

3.4.1Competition for Hydrogen

In anaerobic environments methanogens, homoacetogens and sulfate-reducerswill compete for hydrogen. Thermodynamically, homoacetogenesis is less favor-able than methanogenesis and sulfate reduction. Homoacetogens are very poorhydrogen-utilizing organisms [13]. When grown on organic substrates likeethanol and lactate in the presence of hydrogenotrophic methanogens, theyeven produce hydrogen. In the absence of methanogens 1.5 acetate is producedper lactate or ethanol that is degraded. However, in the presence of methanogensonly 1 acetate per lactate or ethanol is produced, while reducing equivalents aredisposed of as hydrogen.


Table 2. Acetogenic and methanogenic reactions, and sulfate-reducing reactions involved inthe degradation of organic matter in methanogenic bioreactors, and sulfate-reducing biore-actors, respectively

Reaction DG0¢ a

[kJ/reaction]

Syntrophic Acetogenic reactionsPropionate– + 3 H2O Æ Acetate– + HCO3

– + H+ + 3 H2 +76.1Butyrate– + 2 H2O Æ 2 Acetate– + H+ + 2 H2 +48.3Lactate– + 2 H2O Æ Acetate– + HCO3

– + H+ + 2 H2 –4.2Ethanol + H2O Æ Acetate– + H+ + 2 H2 +9.6Methanol + 2 H2O Æ HCO3

– + H+ + 3 H2 +23.5

Methanogenic reactions4 H2 + HCO3

– + H+ Æ CH4 + 3 H2O –135.6Acetate– + H2O Æ CH4 + HCO3

– –31.0Methanol Æ 3/4 CH4 + 1/4 HCO3

– + 1/4 H+ + 1/4 H2O –78.2

Sulfate-reducing reactions4 H2 + SO4

2– + H+ Æ HS– + 4 H2O –151.9Acetate– + SO4

2– Æ 2 HCO3– + HS– –47.6

Propionate– + 3/4 SO42– Æ Acetate– + HCO3

– + 3/4 HS– + 1/4 H+ –37.7Butyrate– + 1/2 SO4

2– Æ 2 Acetate– + 1/2 HS– + 1/2 H+ –27.8Lactate– + 1/2 SO4

2– Æ Acetate– + HCO3– + 1/2 HS– + 1/2 H+ –80.0

Ethanol + 1/2 SO42– Æ Acetate– + 1/2 HS– + 1/2 H+ + H2O –66.4

Methanol + 3/4 SO42– + 1/4 H+ Æ HCO3

– + 3/4 HS– –90.4

Homoacetogenic reactionsLactate– Æ 11/2 Acetate– + 1/2 H+ –56.6Ethanol + HCO3

– Æ 11/2 Acetate– + H2O + 1/2 H+ –42.6Methanol + 1/2 HCO3

– Æ 3/4 Acetate– + H2O –55.04 H2 + 2 HCO3

– + H+ Æ Acetate– + 4 H2O –104.6

a DG0¢-values are taken from Thauer et al. (1977) [2].

Studies with sediments and sludge from bioreactors have indicated that at anexcess of sulfate hydrogen is mainly consumed by sulfate reducers [6, 45–49]. Inreactors with immobilized biomass the activity of hydrogenotrophic methano-gens is completely suppressed within a few weeks when sulfate is added [50]. Ashydrogenotrophic methanogens are still present in high numbers in such reac-tors, this effect cannot simply be explained by Michaelis-Menten or Monodkinetic data (Table 3). In methanogenic environments the hydrogen partial pres-sure is low. However, by addition of sulfate the hydrogen partial pressure mayeven become lower. The hydrogen partial pressure becomes so low that thermo-dynamically hydrogenotrophic methanogenesis is not possible any more (Fig. 1).In freshwater sediments a threshold hydrogen concentration of 1.1 Pa has beenmeasured; this value was lowered to 0.2 Pa by the addition of sulfate [6].

An additional effect of the addition of sulfate is that hydrogen formationbecomes less important. In the absence of sulfate, hydrogen has to be formed byacetogenic bacteria in the oxidation of compounds like lactate, alcohols, propi-onate and butyrate. However, in the presence of sulfate, all these compoundscan be oxidized directly by sulfate-reducers without the intermediate formation


Table 3. Selected growth kinetic data of hydrogenotrophic sulfate-reducing bacteria andmethanogens. For references see Oude Elferink [81] and Oude Elferink et al. [149]

Microorganism Ks µmax Yield a Km Vmax (µM) (1/day) (g/mol H2) (µM) (µmol/min · g)

Sulfate reducersDesulfovibrio

desulfuricans b 1.6–4.3 1.9 1.8–4.0 88vulgaris b 0.7–5.5 0.6–3.1 1.3–4.0 30

Desulfovibrio G11 2.4–4.2 1.2–1.6 1.4–2.0 1.1 65Desulfobacter hydrogenophilus 1.0Desulfobacterium autotrophicum 0.7–1.1Desulfobulbus propionicus b 0.2–1.7Desufomicrobium escambium 1.4

MethanogensMethanobacterium

bryantii 0.3–1.9 0.6formicicum b 1.2–3.1 0.9 2ivanovii 0.8–1.7 1.1 14

Methanobrevibacterarboriphilus b 0.7–3.4 0.6–1.3 6.6smithii 4.1

Methanococcus vannielii 4.1Methanospirillum hungatei 5.8–7.3 1.2–1.8 0.3–0.5 5.0 70

strain BD 2.4–2.8Methanosarcina

barkeri b 1.4–1.8 1.6–2.2 13 110mazei 1.4–1.7

a The yield is given in gram cell dry weight per mol.b Several strains.

of hydrogen. However, this explanation cannot be the only one because fer-mentative glucose- and amino acid-degrading bacteria will always form somehydrogen.

Methanogens,which grow on H2/CO2 ,are autotrophic [51].Among the hydro-gen-utilizing sulfate-reducing bacteria both autotrophic and heterotrophicspecies have been isolated [37]. The classical Desulfovibrio species requireacetate and carbon dioxide or another organic carbon source for growth where-as, e.g., Desulfobacterium sp. can use CO2 as the sole source of carbon [37, 52, 53].An interesting observation has been made by Brysch et al. [54]. Enrichments inmedia with H2 and sulfate as energy substrates and carbon dioxide as the solecarbon substrate resulted in stable cultures of Desulfovibrio and Acetobacteri-um, in a cell ratio of about 20 to 1. The Desulfovibrio species required acetate forgrowth, which was provided by the homoacetogenic Acetobacterium species.Sulfate-reducing bacteria have a higher affinity for hydrogen than homoaceto-gens, but apparently the sulfate-reducers are dependent on the homoacetogensfor synthesis of their carbon source acetate. It can be speculated that under theseconditions the kinetic properties of homoacetogens determine the kineticproperties of the sulfate-reducers. In that case, methanogens would win thecompetition for hydrogen from the sulfate-reducers even at an excess of sulfate.Unfortunately, an experiment which could demonstrate this has never beenperformed. Van Houten et al. [55, 56] started up bioreactors at high hydrogenpartial pressures with solely bicarbonate as carbon source. This led to the coex-istence of sulfate-reducers and homoacetogens.

3.4.2Competition for Acetate

It has been shown that in marine and freshwater sediments acetate is mainlyconsumed by sulfate-reducers when sufficient sulfate is present [45, 46, 49, 57].However, for anaerobic digesters it is less clear how acetate is degraded. A com-plete conversion of acetate by methanogens, even at an excess of sulfate, hasbeen reported [46–48, 50, 58–61]. However, in some studies a predominance ofacetate-degrading sulfate-reducers was found [62–64]. Some factors which mayaffect the competition between sulfate-reducers and methanogens are discussedbelow.

The work of Schönheit et al. [43] has indicated that the predominance ofDesulfobacter postgatei in marine sediments could be explained by its higheraffinity for acetate than Methanosarcina barkeri. The Km values were 0.2 and3.0 mM, respectively (Table 4). However, in bioreactors Methanosarcina sp. areonly present in high numbers when the reactors are operated at a high acetateconcentration or operated at a low pH [65]. Generally, Methanosaeta (formerMethanothrix, [66]) sp. are the most important aceticlastic methanogens inanaerobic bioreactors [65, 67–69]. Also in freshwater sediments Methanosaetaseems to be the most numerous acetoclastic methanogen [70]. Methanosaeta sp.have a higher affinity for acetate than Methanosarcina sp.; their Ks is about0.4 mM [71]. In addition, D. postgatei and other Desulfobacter species are typi-cal marine bacteria, which have not yet been isolated in freshwater media [72].


The aceticlastic sulfate-reducers that prefer freshwater conditions, such asDesulfoarculus baarsii [73], Desulfobacterium catecholicum [74], and Desulfo-coccus biacutus [75], show very poor growth with acetate. Only Desulfobacteri-um strain AcKo and Desulfotomaculum acetoxidans show good growth withacetate under mesophilic conditions (see Table 4). Unfortunately no Ks or Km

values are available for these bacteria.Two abundant acetate-degrading sulfate-reducers, Desulforhabdus amnigenus

and Desulfobacca acetoxidans, were isolated from sulfate-reducing bioreactors[76, 77]. The Michaelis-Menten parameters for D. amnigenus (KM = 0.2–1 mM,Vmax = 21–35 µmol · min–1 · g protein–1) and D. acetoxidans (KM = 0.1–1 mM,Vmax = 29–57 µmol · min–1 · g protein–1) were in the same range as or slightlybetter than those of most Methanosaeta species (KM = 0.4–1.2 mM, Vmax =32–170 µmol · min–1 · g protein–1). This was also the case for the specific growthrate and the threshold value for acetate, which were 0.14–0.20 day–1 and <15 µMfor D. amnigenus and 0.31–041 day–1 and <15 µM for D. acetoxidans. Reportedvalues for Methanosaeta species are 0.08–0.69 day–1 and 7–69 µM, respectively.Putting all kinetic information together, it seems that the growth kinetic prop-erties of acetate-degrading sulfate-reducers are only slightly better than those ofMethanosaeta.

When the growth kinetic properties of the sulfate-reducers are only slightlybetter than those of the methanogens it can be expected that the initial relativecell numbers affect the outcome of competition experiments. This is in particu-lar the case for methanogenic sludge from bioreactors where a major part of themicrobial biomass may consist of Methanosaeta. When methanogenic bioreac-


Table 4. Selected growth kinetic data of acetotrophic sulfate-reducing bacteria and methano-genic bacteria. For references see Oude Elferink [81] and Oude Elferink et al. [149]

Microorganism Ks µmax Yield a Km Vmax (µM) (1/day) (g/mol ac.) (mM) (µmol/min · g)

Sulfate reducersDesulfobacter

curvatus 0.79hydrogenophilus 0.92latus 0.79postgatei b 0.72–1.11 4.3–4.8 0.07–0.23 53

Desulfotomaculum acetoxidans 0.65–1.39 5.6Desulforhabdus amnigenus 0.14–0.20 0.6 28Desulfobacca acetoxidans 0.31–0.41 0.6 43

MethanogensMethanosarcina barkeri b 5.0 0.46–0.69 1.6–3.4 3.0mazei b 0.49–0.53 1.9Methanosaetasoehngenii b 0.5 0.08–0.29 1.1–1.4 0.39–0.7 38concilii 0.21–0.69 1.1–1.2 0.84–1.2 16

a The yield is given in gram cell dry weight per mol.b Several strains.

tors are fed with sulfate, the few initial acetate-degrading sulfate-reducers haveto compete with huge numbers of aceticlastic Methanosaeta species. In UASBreactors the sludge age can be as high as 0.5–1 year [78]. Visser et al. [79] havesimulated the competition between sulfate-reducing bacteria and methanogensusing a biomass retention time in the reactor of 0.02 day–1, a maximum specificgrowth rate of 0.055 and 0.07 day–1 for the methanogen and sulfate-reduc-ing bacterium, respectively, a Ks value for acetate of 0.08 and 0.4 mM acetate,respectively, and different initial ratios of bacteria. Starting with a ratio ofmethanogens/sulfate reducers of 104, it will take already one year before thenumbers of acetate-degrading sulfate-reducing bacteria and acetate-degradingmethanogens are equal. Nevertheless, long-term UASB reactor experiments ofVisser [65] showed that sulfate-reducers are able to outcompete methanogensfor acetate, even if the seed sludge initially only contains low numbers of aceti-clastic sulfate-reducers. In his acetate- and sulfate-fed UASB reactor it took50 days before acetate degradation via sulfate reduction was observed, andanother 50 days to increase it to 10%. The shift from 50 to 90% of acetate degra-dation via sulfate reduction took approximately 400 days.

Methanosaeta can only grow on acetate, whereas Methanosarcina can use afew other substrates besides acetate, like hydrogen, methanol and methylatedamines [71, 79]. Aceticlastic Desulfobacter sp. also use a limited range of sub-strates; solely hydrogen, acetate and ethanol provide good growth [72]. Desul-fobacca acetoxidans is also a true specialist. It only showed growth on acetate[76]. However, Desulfotomaculum acetoxidans and Desulforhabdus amnigenususe a wide range of the common substrates for sulfate-reducers for growth [77,80]. It is not clear to which extent these bacteria can grow mixotrophically.During growth on, e.g., butyrate or ethanol acetate is even excreted [80, 81].However, if low concentrations of acetate and other substrates are used at thesame time the outcome of the competition between Methanosaeta and these sul-fate-reducers will be affected. Gottschal and Thingstad [82] described a modelin which it is shown that during competition on mixtures of substrates in con-tinuous cultures not only the specific growth rate determines the outcome of acompetition, but also the yield on the different substrates.

3.4.3Competition for Methanol

Methanol is an excellent substrate for mesophilic methanogens and homoaceto-gens. Methanosarcina species, Acetobacterium woodii, Eubacterium limosum andButyribacterium methylotrophicum show very fast growth on methanol [83–88](Table 5). The homoacetogens require externally supplied bicarbonate forgrowth, while the methanogens do not. Remarkably, only a very few mesophilicspecies of sulfate-reducing bacteria can grow on methanol [89–91]. The maxi-mum specific growth rates of these sulfate-reducers are much lower than thoseof the methanogens and homoacetogens. This suggests that sulfate-reducers arepoor competitors for methanol.

The competition between methanogens and homoacetogens in bioreactorshas been studied by Florencio [92] it appears that the Ks value of methanogens


for methanol is 0.25 mM, while that of the homoacetogens is much higher(16 mM). This indicates that at a low methanol concentration methanol is main-ly used by methanogens. Only at a high methanol concentration, and addition-ally a high bicarbonate concentration, was a substantial part of the methanolconsumed by homoacetogens.

During growth on methanol methanogens and homoacetogens producesome hydrogen. The amount of hydrogen which is produced is affected by thepresence of sulfate-reducers. This results in the coexistence of methanol-utiliz-ing and hydrogen-utilizing anaerobes [84, 93–95]. Thus, it seems that in mixedcommunities growing on methanol there is an indirect competition betweenmethanogens and sulfate-reducers as well.

We have studied methanol conversion in mesophilic and thermophilic sulfate-reducing bioreactors at high sulfate concentrations.At low temperature methano-genesis became the dominant process, indicating that methanol is mainlyconsumed by methanogens (Weijma, unpublished results). However, at a hightemperature (65 °C) sulfate reduction became the dominant process [96]. Somethermophilic Desulfotomaculum species show excellent growth with methanol.

3.4.4Competition for Organic Acids and Ethanol

In anaerobic environments with high sulfate concentrations, sulfate-reducingbacteria compete with acetogenic bacteria for substrates like lactate, ethanol,propionate and butyrate. Little is known about this competition.

The fate of ethanol and lactate in anaerobic environments is not completelyclear.A few methanogens are able to oxidize ethanol and other alcohols [97, 98].In the presence of sulfate they can be oxidized by, e.g., Desulfovibrio species.However, lactate and ethanol (+CO2) can also be fermented by bacteria in a pro-


Table 5. Specific growth rates and growth yields (g dry weight · mol–1) of methanol-utilizinganaerobic bacteria. For references see Florencio [92], and Nanninga and Gottschal [90]

Microorganism µmax (1/day) Yield (g/mol methanol)

MethanogensMethanosarcina barkeristrain MS 2.35 3.5strain 227 1.85 3.8Methanosarcina mazei 3.24Methanosarcina acetivorans 3.20

HomoacetogensAcetobacterium woodii 5.3–8.2Eubacterium limosum 2.38 7.1Butyribacterium methylotrophicum 1.85 8.2

Sulfate reducersDesulfovibrio carbinolicum 0.22

pionic acid or homoacetogenic fermentation. In addition, lactate (+acetate) andethanol (+acetate) can be fermented in a butyric acid fermentation by Clostrid-ium kluyveri. Chemostat experiments have indicated that at low concentrationslactate and probably also ethanol are mainly consumed by sulfate-reducers.Desulfomicrobium outcompeted Veillonella and Acetobacterium at low acetateconcentration. However, it appeared that the Veillonella sp. had a much higherspecific growth rate than the sulfate-reducer, 0.30 and 0.17 h–1, respectively.Interestingly, sulfate-reducers are also able to ferment lactate and ethanol. Lac-tate and ethanol can be oxidized to acetate and hydrogen, provided that thehydrogen partial pressure is kept low by methanogens [99], while Desulfobulbusspecies are able to ferment lactate and ethanol in a propionic acid fermentation[37, 100, 101].

For wastewater with an excess of sulfate it is to be expected that sulfate-reduc-ing bacteria become predominant over syntrophic fatty acid-degrading consor-tia, because of their better growth kinetic properties (Table 6). It is obvious thatat high sulfate concentrations, sulfate-reducing bacteria grow much faster thanthe syntrophic consortia.Almost no Ks and Km values for propionate and butyrate


Table 6. Specific growth rates (1/day) of sulfate-reducing bacteria and of acetogenic bacteriain co-cultures with hydrogenotrophic methanogens/sulfate reducers, growing on butyrate orpropionate. For references see Oude Elferink [81] and Oude Elferink et al. [149]

Sulfate-reducing Syntrophicculture co-culture

– sulfate + sulfate

Butyrate-degrading strainsDesulfoarculus baarsii 0.4Desulfobacterium autotrophicum 0.67–1.11 27Desulfococcus multivorans 0.17–0.23Desulfotomaculum acetoxidans 1.11Desulfotomaculum strain Gro111 1.2–1.3Syntrophomonas sapovorans 0.6Syntrophomonas wolfei 0.2 0.3Syntrophospora (Clostridium) bryantii 0.25sporeforming strain FMS2 0.31sporeforming strain FSS7 0.34non-sporeforming strain FM4 0.24non-sporeforming strain B1 0.1

Propionate-degrading strainsDesulfobulbus elongatus 1.39Desulfobulbus propionicus a 0.89–2.64Desulfococcus multivorans 0.17–0.23Syntrophobacter fumaroxidans 0.02 0.15–0.17Syntrophobacter pfennigii 0.07 0.07Syntrophobacter wolinii 0.06 0.02–0.10 0.18–0.21culture PT 0.1culture PW 0.23 0.14

a Several strains.

degradation have been reported. Therefore, a comparison of the growth ofsyntrophic cultures and sulfate-reducers at low substrate concentrations is notpossible. The existence of two subpopulations of propionate-oxidizers in metha-nogenic sludge was reported [102], a fast-growing one with a µmax of 1.2 day–1

and a Ks of 4.5 mM, and a slow-growing one with a higher affinity (µmax of0.13 day–1 and a Ks of 0.15 mM).

Several researchers investigated the competition for propionate and butyratebetween sulfate-reducers and acetogens in anaerobic reactors and in sedimentslurries. In most cases syntrophic consortia are easily outcompeted by sulfate-reducers [48, 50, 60, 103]. However, in some of these studies no distinction can be made between a direct oxidation of propionate and butyrate by sulfate-reducers and an indirect conversion whereby the fatty acids are oxidized toacetate and hydrogen by the acetogenic bacteria followed by hydrogen con-version via sulfate reduction. In this respect it is important to note that sulfate-reducers keep the hydrogen partial pressure lower than methanogens, and that propionate- and butyrate-degrading acetogens grow much faster in co-culture with hydrogen-consuming sulfate-reducers than with hydrogen-con-suming methanogens [104, 105]. Therefore, the reported critical role of sulfate-reducers in mediating propionate and butyrate degradation [48, 50, 106, 108]may be that of a hydrogen-consumer or that of a direct propionate or butyrate-oxidizer.

Findings of Harmsen [108] and Raskin et al. [109] seem to support the directpropionate oxidation by sulfate-reducers. The population dynamics of propi-onate-oxidizing bacteria in two UASB reactors, one fed with propionate andsulfate and the other with only propionate were studied. In the first reactor thenumber of Desulfobulbus sp. increased rapidly, and in the second reactor thenumber of syntrophic propionate oxidizers increased. It seems unlikely thatDesulfobulbus acted as a hydrogen scavenger in the first reactor, although Desul-fobulbus sp. are able to use H2 as well as propionate, because no syntrophic pro-pionate-oxidizers were enriched in this reactor, and all Desulfobulbus cells werelocalized on the outside of the granule, not intertwined with other bacteria.

Remarkably, Syntrophobacter species are also able to grow on propionate andsulfate [110–113]. The importance of the sulfate-dependent growth of these bac-teria is not fully understood

3.4.5Competition for Sulfate

At low sulfate concentrations the growth of the sulfate-reducing bacteria will besulfate-limited.Also under conditions of high sulfate concentrations, sulfate lim-itation may occur due to mass transfer limitation of sulfate into the biofilm.Nielsen [114] reported that sulfate limitation could already occur in a biofilm ofa few hundred µm thick when the sulfate concentration in the bulk solution wasbelow 0.5 mM.

Under sulfate-limiting conditions aceticlastic sulfate-reducers will have tocompete with other sulfate-reducers for the available sulfate. Laanbroek et al.[115] experimented with three bacterial strains, Desulfobacter postgatei, Desul-


fobulbus propionicus and Desulfomicrobium baculatum in sulfate-limitedchemostats. They found that D. baculatum was the most successful competitorfor limiting amounts of sulfate, followed by D. propionicus and then by D. post-gatei. The Km for sulfate of D. postgatei is 200 µM [116], a value which is muchhigher than the reported Ks and Km values for several Desulfovibrio strains(5–77 µM) [117–119]. The affinities for sulfate of Desulfobacter strain AcKo,Desulfotomaculum acetoxidans, Desulforhabdus amnigenus and Desulfobaccaacetoxidans are not known. However, if these species have a higher Ks value thanother sulfate-reducers, one might speculate that limiting amounts of sulfatewould result in an oxidation of compounds like hydrogen, formate and butyrateby sulfate-reducing bacteria, while acetate is used by the aceticlastic metha-nogens.

Competition for sulfate between sulfate-reducing bacteria could explain theresults obtained in studies with sulfate-limited reactors, where acetate seemed tobe the least favored substrate for sulfate reduction, compared to propionate,butyrate and hydrogen [50, 107, 120]

When hydrogen-utilizing sulfate-reducers have the highest affinity for sulfatethis would indicate that under sulfate-limiting conditions fatty acids are oxi-dized in syntrophy with hydrogen-utilizing sulfate-reducers and not directly byDesulfobulbus species.

3.5Competition Between Sulfate-Reducers and Acetogens in the Absence of Sulfate

The role of sulfate-reducing bacteria in the anaerobic digestion in the absence ofsulfate has hardly been investigated. Yet, recent studies showed that sulfate-reducing bacteria can be present in methanogenic sludge to upto 15% of thetotal biomass [109]. It is known that several types of sulfate-reducing bacteriahave fermentative or syntrophic capacities. Widdel and Hansen [37] gave anoverview of the fermentative and syntrophic growth of sulfate-reducing bacte-ria. Growth of sulfate-reducers in the absence of sulfate could explain the fastresponse of methanogenic ecosystems to the addition of sulfate.Some substrateswhich can be fermented by sulfate-reducers are pyruvate, lactate, ethanol,fumarate and malate, fructose, serine, choline, acetoin and S-1,2-propanedioland propanol + acetate. Sulfate-reducers can also grow as acetogens in theabsence of sulfate. Desulfovibrio sp. oxidize ethanol or lactate to acetate when co-cultured with methanogens [99, 121–124]. It has been reported that Desul-fovibrio sp. were the main lactate- and ethanol-degrading bacteria in a reactortreating whey in the absence of sulfate [125, 126]. However, others reportedthat only in the presence of sulfate were Desulfovibrio sp. the dominant lactatedegraders, while in the absence of sulfate lactate was fermented according tothe usual fermentation pattern of Propionibacterium [48]. Syntrophic formatedegradation has been reported for Desulfovibrio vulgaris in association withMethanobacterium bryantii [127], and a Desulfovibrio-like organism couldsyntrophically degrade alcohols like 1,3-butanediol, 1,4-butanediol, 1-butanoland 1-propanol in the presence of 10 mM acetate and Methanospirillum hun-gatei [128].


The role of sulfate-reducing bacteria in propionate degradation becomesmore intricate by the work of Wu et al. [129, 130]. They were the first to reportthat the syntrophic conversion of propionate was mainly performed by sulfate-reducing bacteria, and they were able to isolate such an organism. This indicatesthat in the absence of sulfate certain propionate-degrading sulfate-reducing bac-teria are able to oxidize propionate in syntrophic association with H2-consum-ing anaerobes, while in the presence of sulfate they couple propionate oxidationto sulfate reduction. This represents a considerable ecological advantage of thistype of sulfate-reducing bacteria over obligate syntrophic propionate-degradersin ecosystems where sulfate is continuously or intermittently available.

Interestingly, as mentioned before, several Syntrophobacter species, includingS. wolinii [111], S. pfennigii [112], S. fumaroxidans [110, 131], strain HP1.1 [113],were shown to grow on propionate with sulfate. For S. wolinii this finding wasvery remarkable because S. wolinii grows as an acetogen in the presence ofDesulfovibrio G11 [104]. Phylogenetic research, based on 16S rRNA sequences,showed that Syntrophobacter species belong to the Gram-negative sulfate-reduc-ers [108, 132].

Thus far, growth of sulfate-reducers on butyrate in the absence of sulfate butin the presence of methanogens was not yet demonstrated. However, Desul-fovibrio sp. were detected in a fixed-bed reactor fed with butyrate without sul-fate [133, 134].

4Inhibition

As discussed earlier in this chapter, several substrates and products of anaerobicrespiration might have inhibitory effects on the methanogenic consortia inanaerobic digesters.

Much of the decrease in methane production caused by intermediate nitro-gen oxides of the denitrification process (NO2

–, NO and N2O) is due to toxicity ofthese compounds rather than competition and unfavorable redox conditions.The inhibition mechanism of nitrate and its denitrification products is stilllargely unknown. The reduction of oxidized nitrogen species for dissimilatoryelectron dissipation by fermentative bacteria yields ammonia which numerousauthors have demonstrated to be toxic to methanogenic consortia. Ammoniais mainly toxic in its un-ionized form (NH3) while the ammonium ion (NH4

+)is much less toxic, and toxicity is therefore dependent upon pH and tempera-ture of the reactor. Fig. 3 shows the effect of temperature and pH on the per-centage of total ammonium (NH4

+ + NH3) which appears as NH3 . It is obviousthat increasing temperature and pH leads to increased NH3 concentrations in areactor.

If the sludge fed to the reactor simultaneously contains high amounts of pro-teinaceous material or/and pig manure, large amounts of ammonia are releasedfrom the fermentation of amino acids and other nitrogen-rich compounds[135]. Ammonia has been shown to mainly affect acetate-utilizing methano-genic Archaea, and to a lesser degree, hydrogen-utilizing methanogens and syn-trophic bacteria [136]. A decrease in pH and an increase in the concentration of



Fig.

3.T

he e

ffec

t of

pH a

nd t

empe

ratu

re o

n th

e di

ssoc

iati

on o

fH

2S a

nd N

H3.D

H o

fth

e tw

o re

acti

ons

(H2S

´ H

S–+

H+,a

nd

NH

3+

H+

´ N

H4+) w

ere

calc

ulat

ed fr

om e

ntha

lpie

s off

orm

atio

n [1

50].

Equi

libri

um c

onst

ants

at t

he th

ree

tem

pera

ture

s wer

e th

enca

lcul

ated

from

Van

’t H

oof’s

equ

atio

n (l

nK

2/K

1=

DH

/R(1

/T1–

1/T

2).F

inal

ly,t

he u

ndis

soci

ated

frac

tion

s at

diff

eren

t pH

val

ues

(f)

wer

e ca

lcul

ated

from

the

equa

tion

f=

100

¥ (1

+K

/10–p

H)–1

.(--

-)25

°C;(

–––)

37

°C;(

––––

) 60

°C

volatile fatty acids observed in ammonia-inhibited reactors, however, pointtowards an inhibition of all terminal microorganisms of the anaerobic degrada-tion chain [137]. In two studies on the effects of high ammonia concentrations(7 g NH4

+ – N/L) on methanogenesis from acetate, Blomgren et al. [138] andSchnürer et al. [139] demonstrated that aceticlastic methanogenesis was dis-placed in favor of syntrophic acetate oxidation in enriched and defined culturesgrowing with acetate as the only substrate. When the anaerobic processes areinhibited by ammonia, the decrease in pH will counteract the effect of ammoniadue to a decrease in the free ammonia concentration.

Since anaerobic reactors used in different ammonia toxicity studies haveoften been operated at different pH values, it is difficult to generalize about theinhibitory concentration as different concentrations of NH3 ammonia are pre-sent. In most reactor studies, however, inhibitory concentrations are in the range1.7–5 g total ammonia-N/L, corresponding to 0.4–1 g NH3-ammonia/L [135,140, 141]. Several authors have also shown that the biogas process can be adapt-ed to ammonia concentrations above 4 g total ammonia/L without any reductionof the methane yield [135, 142, 143].

Sulfide produced by sulfate-reducing bacteria and by fermentation of sulfur-containing amino acids has been shown to be inhibitory to the biogas process byseveral authors [144, 145]. Similar to ammonia, it is generally assumed that theneutral undissociated sulfide is the agent of toxicity since it is only membranepermeable in this form [146]. The pH is therefore also an important determinantof the toxicity, but contrary to ammonia, low pH values and low temperaturesfavor the undissociated sulfide (Fig. 3) . Much of the published literature on sul-fide toxicity does not take pH into consideration, which makes general conclu-sions about toxicity levels difficult. Since sulfide readily reacts with most metalsto form insoluble metal sulfides, the toxicity of sulfide is also related to metalconcentrations in the sludge. However, several authors have found that sulfideinhibits the biogas process at concentrations around 50 mg/L [144, 147]. Sulfideand ammonia have been shown to inhibit methanogenesis in thermophilicanaerobic digesters synergistically. A sulfide concentration of only 23 mg/L ledto an approximately 40% decrease of the methane production in a digestertreating material with a high ammonium concentration [140]. From Fig. 3 itis obvious that optimal conditions for maintaining a low concentration of un-dissociated H2S and NH3 are occurring at lower pH values for thermophilicdigesters than for mesophilic digesters.

5Conclusion

Compared to many other anaerobic environments,anaerobic digesters receivingmunicipal sludge or animal wastes are generally sparsely exposed to inorganicelectron acceptors. Of the large amounts of easily degradable carbon only a tinyfraction is consumed by respiring bacteria. In digesters receiving industrialwastes, however, significant amounts of electron acceptors stimulating anaero-bic respiration other than methanogenesis might occur and pose a problem asdescribed in this chapter. In most natural environments, inorganic electron


acceptors and the corresponding respiration types are confined to distinct zonesin a stratified system. A similar zonation can be established in sludge blanketreactors such as UASB reactors. In stirred reactors, however, the maintenance ofgradients outside particles is difficult and probably only sporadically occurring;thus, in principle, several types of anaerobic respiration might proceed simulta-neously given sufficient amounts of organic electron donors. The interactivepattern of electron acceptors, intermediate products and respiring microorgan-isms is therefore very complex under these conditions and only partly under-stood as discussed in this chapter. The role of electron acceptors such as ferriciron and manganese has only been very sparsely studied in anaerobic digesters.Also the role of newly described respiration systems such as humic acid respira-tion [148] awaits thorough investigation in anaerobic digesters.

6References

1. Hamilton WA (1985) Annu Rev Microbiol 39:1952. Thauer RK, Jungermann K, Decker K (1977) Bacteriol Rev 41:1003. Stams AJM (1994) Antonie van Leeuwenhoek 66:2714. Coleman ME, Dreesen DW, Wiegert RG (1996) Appl Environ Microbiol 62:36325. Abram JW, Nedwell DB (1978) Appl Environ Microbiol 117:896. Lovley DR, Dwyer DF, Klug MJ (1982) Appl Environ Microbiol 43:13737. Kuenen JG, Harder W (1982) In Burns RG, Slater JH (eds), Experimental Microbial

Ecology. Blackwell, London, p 3428. Veldkamp H, Jannasch HW (1972) J Appl Chem Biotechnol 22:1059. MacArthur RH, Wilson EO (1967) The theory of island biogeography. Princeton NJ,

Princeton University Press10. Schink B, Friedrich M (1994) FEMS Microbiol Rev 15:8511. Lovley DR (1985) Appl Environ Microbiol 49:153012. Lovley DR, Phillips EJP (1987) Appl Environ Microbiol 53:263613. Cord-Ruwisch R, Seitz H-J, Conrad R (1988) Arch Microbiol 149:35014. Westermann P, Ahring BK, Mah RA (1989) Appl Environ Microbiol 55:51415. Oremland R S (1988) In: Zehnder AJB (ed), Biology of anaerobic microorganisms.Wiley,

New York, p 64116. Marounek M, Wallace RJ (1984) J Gen Microbiol 130:22317. Fetzer S, Conrad R (1993) Arch Microbiol 160:10818. Kirby T W, Lancaster JR Jr, FridovichI (1981) Arch Biochem Biophys 210:14019. Frette L, Gejlsbjerg B, Westermann P (1997) FEMS Microbiol Ecol 24:36320. O’Keefe D M, Brigmon R L, Chynoweth DP (2000) Biores Technol 71:21721. Kato M T, Field J A, Lettinga G (1993) Biotechnol Bioeng 42:136022. Klüber H D, Conrad R (1998) FEMS Microbiol Ecol 25:30123. Akunna J C, Binzeau C, Moletta R (1992) Environ Technol 13:82524. Percheron G, Bernet N, Moletta R (1999) FEMS Microbiol Ecol 29:34125. Akunna J C, Binzeau C, Moletta R (1993) Water Res 27:130326. Hendriksen HV, Ahring BK (1996) Water Res 30:145127. Chen KC, Lin YF (1993) Water Res 27:174928. Clarens M, Bernet N, Delgenés J-P, Moletta R (1998) FEMS Microbiol Ecol 25:27129. Lovely DR (1997) FEMS Microbiol Rev 20:30530. Lovley DR (1991) Microbiol Rev 55:25931. Wahid PA, Kamalam NV (1993) Biol Fert Soils 15:14432. Peters V, Conrad R (1996) Soil Biol Biochem 3:37133. Lovley DR, Phillips EJ (1994) Appl Environ Microbiol 60:2394


34. Henze M (1992) Water Sci Technol 25:135. Mamais D, Pitt PA, Cheng YW, Loiacono J, Jenkins D (1994) Wat Environ Res 66:91236. Widdel F (1988) In: Zehnder AJB (ed), Biology of anaerobic microorganisms. Wiley, New

York, p 46937. Widdel F, Hansen TA (1992) In: Balows A, Trüper, Dworkin M, Harder W, Schleifer K-H,

(eds), The Prokaryotes, 2nd edn. Springer, Berlin, Heidelberg, New York, p 58338. Colleran E, Finnegan S, Lens P (1995) Antonie van Leeuwenhoek 67:2939. Schink B (1997) Microbiol Biol Rev 61:26240. Verstraete W, de Beer D, Pena M, Lettinga G, Lens P (1996) World J Microbiol Biotechnol

12:22141. Kristjansson JK, Schönheit P, Thauer RK (1982) Arch Microbiol 131:27842. Robinson JA, Tiedje JM (1984) Arch Microbiol 137:2643. Schönheit P, Kristjansson JK, Thauer RK (1982) Arch Microbiol 132:28544. Lupton FS, Zeikus JG (1984) Curr Microbiol 11:745. Banat IM, Lindström EB, Nedwell DB, Balba MT (1981) Appl Environ Microbiol 42:98546. Isa Z, Grusenmeyer S, Verstraete W (1986) Appl Environ Microbiol 51:57247. Isa Z, Grusenmeyer S, Verstraete W (1986) Appl Environ Microbiol 51:58048. Qatibi AI, Bories A, Garcia JL (1990) Anthonie van Leeuwenhoek 58:24149. Winfrey MR, Zeikus JG (1977) Appl Environ Microbiol 33:27550. Visser A, Gao Y, Lettinga G (1993) Water Res 27:54151. Whitman WB, Bowen TL, Boone DR (1992) In: Balows A, Trüper, Dworkin M, Harder W,

Schleifer K-H (eds), The Prokaryotes, 2nd edn. Springer, Berlin, Heidelberg, New York,p 719

52. Badziong W, Thauer RK, Zeikus JG (1978) Arch Microbiol 116:4153. Widdel F, Bak F (1992) In: Balows A, Trüper, Dworkin M, Harder W, Schleifer K-H (eds),

The Prokaryotes, 2nd edn. Springer, Berlin, Heidelberg, New York, p 335254. Brysch K, Schneider C, Fuchs G, Widdel F (1987) Arch Microbiol 148:26455. Van Houten RT, Hulshoff Pol LW, Lettinga G (1994) Biotechnol Bioengin 44:58656. Van Houten RT, Oude Elferink SJWH, van Hamel SE, Hulshoff Pol LW, Lettinga G (1995)

Bioresource Technol 54:7357. Smith RL, Klug MJ (1981) Appl Environ Microbiol 42:11658. Ueki K, Ueki A, Simogoh Y (1988) J Gen Appl Microbiol 34:42559. Ueki A, Ueki K, Oguma A, Ohtsuki C (1989) J Gen Appl Microbiol 35:51560. Visser A,Beeksma I,van der Zee F,Stams AJM,Lettinga G (1993b) Appl Microbiol Biotech-

nol 40:54961. Yoda M, Kitagawa M, Miyaji Y (1987) Water Res 21:154762. Gupta A, Flora JRV, Gupta M, Sayles GD, Suidan MT (1994) Water Res 28:78163. Rinzema A (1988) Anaerobic treatment of wastewater with high concentrations of lipids

or sulphate. PhD Thesis, Wageningen Agricultural University, Wageningen64. Visser A (1995) The anaerobic treatment of sulfate containing wastewater. PhD Thesis,

Wageningen Agricultural University, Wageningen65. Grotenhuis JTC (1992) Structure and stability of methanogenic granular sludge. PhD

Thesis, Wageningen Agricultural University, Wageningen66. Patel GB (1984) Can J Microbiol 30:138367. MacLoed FA, Guiot SR, Costerton JW (1990) Appl Environ Microbiol 56:159868. Morvai L, Mihàhz P, Czakó L (1992) Water Sci Tech 25:11369. Nishio N, Silveira RG, Hamato K, Nagai S (1993) J Ferment Bioeng 75:30970. Scholten JCM (1999) The influence of sulfate and nitrate on the methane formation by

methanogenic archaea in freshwater sediments. PhD thesis, Wageningen AgriculturalUniversity, Wageningen

71. Jetten MSM, Stams AJM, Zehnder AJB (1992) FEMS Microbiol Rev 88:18172. Widdel F (1987) Arch Microbiol 148:28673. Widdel F (1980) Anaerober Abbau von Fettsäuren und Benzoesäure durch neue isolierte

Arten Sulfate-reduzierender Bacterien. PhD Thesis, University of Göttingen, Germany74. Szewzyk R, Pfennig N (1987) Arch Microbiol 147:162


75. Platen H, Temmes A, Schink B (1990) Arch Microbiol 154:35576. Oude Elferink SJWH, Akkermans-van Vliet WM, Bogte JJ, Stams AJM (1999) Int J Syst

Bacteriol 49:34577. Oude Elferink SJWH, Maas RN, Harmsen HJM, Stams AJM (1995) Arch Microbiol

164:11978. Hulshoff Pol LW (1989) The phenomenon of granulation of anaerobic sludge. PhD

Thesis, Wageningen Agricultural University, Wageningen79. Boone DR, Whitman WB (1988) Int J Syst Bacteriol 38:21280. Widdel F, Pfennig N (1981) Arch Microbiol 129:39581. Oude Elferink SJWH, Luppens SBI, Marcelis CLM, Stams AJM (1998) Appl Environ

Microbiol 64:230182. Gottschal JC, Thingstad TF (1982) Biotechnol Bioeng 24:140383. Genthner BRS, Davis CL, Bryant MP (1981) Appl Environ Microbiol 42:1284. Genthner BRS, Bryant MP (1987) Appl Environ Microbiol 53:47185. Bache R, Pfennig N (1981) Arch Microbiol 130: 25586. Hutten TJ, Bongaerts HCM, van der Drift C,Vogels GD (1980) Antonie van Leeuwenhoek

46:7687. Lynd LH, Zeikus JG (1983) J Bacteriol 153:142588. Maestrojuán GM, Boone DR (1991) Int J Syst Bacteriol 41:26789. Nanninga HJ, Gottschal JC (1986) FEMS Microbiol Ecol 38:12590. Nanninga HJ, Gottschal JC (1987) Appl Environ Microbiol 53:80291. Braun M, Stolp H (1985) Arch Microbiol 142:7792. Florencio L (1994) The fate of methanol in anaerobic bioreactors. PhD thesis, Wagenin-

gen Agricultural University93. Cord-Ruwisch R, Ollivier B (1986) Arch Microbiol 144:16394. Heijthuijsen JHFG, Hansen TA (1986) FEMS Microbiol Lett 38:5795. Phelps TJ, Conrad R, Zeikus JG (1985) Appl Environ Microbiol 50:58996. Weijma J, Stams AJM, Hulshoff Pol LW, Lettinga G (2000) Biotech Bioeng 67:35497. Widdel F, Rouviere PE, Wolfe RS (1988) Arch Microbiol 150:47798. Zellner G, Bleicher K, Braun E, Kneifel H, Tindall BJ, Conway de Macario E, Winter J

(1989) Arch Microbiol 151:1–999. Bryant MP, Campbell LL, Reddy CA, Crabill MR (1977) Appl Environ Microbiol 33:1162

100. Laanbroek HJ, Abee T, Voogd IL (1982) Arch Microbiol 133:178101. Tasaki M, Kamagata Y, Nakamura K, Mikami E (1992) J Ferment Bioeng 73:329102. Heyes RH, Hall RJ (1983) Appl Environ Microbiol 46:710103. Scholten JCM, Stams AJM (1995) Antonie van Leeuwenhoek 68:309104. Boone DR, Bryant MP (1980) Appl Environ Microbiol 40:626105. McInerney MJ, Bryant MP, Hespell RB, Costerton JW (1981) Appl Environ Microbiol

41:1029106. Harada H, Uemura S, Monomoi K (1994) Water Res 28:355107. Mizuno O, Li YY, Noike T (1994) Water Sci Tech 30:45108. Harmsen HJM (1996) Detection, phylogeny and population dynamics of syntrophic

propionate-oxidizing bacteria in anaerobic granular sludge. PhD Thesis, WageningenAgricultural University, Wageningen

109. Raskin L, Rittmann BE, Stahl DA (1996) Appl Environ Microbiol 62:3847110. Kuijk van BLM, Stams AJM (1995) Antonie van Leeuwenhoek 68:293111. Wallrabenstein C, Hauschild E, Schink B (1994) FEMS Microbiol Lett 123:249112. Wallrabenstein C, Hauschild E, Schink B (1995) Arch Microbiol 164:346113. Zellner G, Busmann A, Rainey FA, Diekmann H (1996) Syst Appl Microbiol 19:414114. Nielsen PH (1987) Appl Environ Microbiol 53:27115. Laanbroek HJ, Geerligs HJ, Sijtsma L,Veldkamp H (1984) Appl Environ Microbiol 47:329116. Ingvorsen K, Zehnder AJB, Jørgensen BB (1984) Appl Environ Microbiol 47:403117. Ingvorsen K, Jørgensen BB (1984) Arch Microbiol 139:61118. Nethe-Jaenchen R, Thauer RK (1984) Arch Microbiol 137:236119. Okabe S, Nielsen PH, Characklis WG (1992) Biotechnol Bioeng 40:725


120. Uberoi V, Bhattacharya SK (1995) Water Eviron Res 67:330121. McInerney MJ, Bryant MP (1981) Appl Environ Microbiol 41:346122. Tatton MJ, Archer DB, Powell GE, Parker ML (1989) Appl Environ Microbiol 55:440123. Traore AS, Fardeau M-L, Hatchikian CE, Le Gall J, Belaich J-P (1983) Appl Environ Micro-

biol 46:1152124. Yadav VK, Archer DB (1988) Lett Appl Microbiol 7:165125. Chartrain M, Zeikus JG (1986) Appl Environ Microbiol 51:188126. Zellner G, Vogel P, Kneifel H, Winter J (1987) Appl Microbiol Biotechnol 27:306127. Guyot J-P, Brauman A (1986) Appl Environ Microbiol 52:1436128. Tanaka K (1992) J Ferment Bioeng 73:362129. Wu W-M, Hickey RF, Zeikus JG (1991) Appl Environ Microbiol 57:3438130. Wu W-M, Jain M, Conway de Macario E, Thiele JH, Zeikus JG (1992) Appl Microbiol

Biotechnol 38:282131. Harmsen HJM, van Kuijk BLM, Plugge CM, Akkermans ADL, de Vos WM, Stams AJM

(1998) Int J Syst Bacteriol 48:1383132. Harmsen HJM, Wullings B, Akkermans ADL, Ludwig W, Stams AJM (1993) Arch Micro-

biol 160:238133. Zellner G, Geveke M, Conway de Macario E, Diekmann H (1991) Appl Microbiol Biotech-

nol 36:404134. Zellner G, Geveke M, Diekmann H (1991) Biotechnol Lett 13:687135. Angelidaki I, Ahring BK (1993) Appl Microbiol Biotechnol 38:560136. Heinrichs DM, Poggi-Varaldo HM, Oleszkiewicz JA (1990) J Environ Engineering

116:698137. Poggi-Varaldo HM., Rodriguez-Vazquez R, Fernandez-Villagomez G, Esparza-Garcia F

(1997) Appl Microbiol Biotechnol 47:284138. Blomgren A, Hansen A, Svensson B (1990) In: Belaich JP, Bruschi M, Garcia J-L (eds),

Microbiology and biochemistry of strict anaerobes involved in interspecies hydrogentransfer. Plenum Press, New York, p 225

139. Schnürer A, Houwen FP, Svensson BH (1994) Arch Microbiol 162:70140. Hansen KH, Angelidaki I, Ahring BK (1999) Wat Res 33:1805141. Borja R, Sánchez E, Weiland P (1996) Process Biochem 31:477142. Hashimoto AG (1986) Agricul Wastes 17:241143. van Velsen AFM (1979) Wat Res 13:995144. Karhadkar PP, Audic J-M, Faup GM, Khanna P (1987) Wat Res 21:1061145. Hilton BL, Oleszkiewicz JA (1988) J Environ Eng 114:1377146. O’Flaherty V, Mahony T, O’Kenndey R, Colleran E (1998) Process Biochem 33:555147. Parkin GF, Speece RE, Yang CHJ, Kocher WM (1983) J Wat Pol Contr Fed 55:44148. Lovley DR, Coates JD, Blunt-Harris EL, Phillips EJP, Woodward JC (1996) Nature 382:445149. Oude Elferink SJWH, Visser A, Hulshoff Pol LW, Stams AJM (1994) FEMS Microbiol Rev

15:119150. Stumm W, Morgan JL (1981) Aquatic chemistry. Wiley, New York

Received: January 2002

56 A.J.M. Stams et al.: Metabolic Interactions Between Methanogenic Consortia and …

Kinetics and Modeling of Anaerobic Digestion Process

Hariklia N. Gavala 1 · Irini Angelidaki 2 · Birgitte K. Ahring 1

1 The Environmental Microbiology and Biotechnology Group (EMB), Biocentrum-DTU,bldg 227, The Technical University of Denmark, 2800 Lyngby, Denmark.E-mail: [email protected]

2 Environment and Resources DTU, Bldg 115, The Technical University of Denmark,2800 Lyngby, Denmark

Anaerobic digestion modeling started in the early 1970s when the need for design and effi-cient operation of anaerobic systems became evident.At that time not only was the knowledgeabout the complex process of anaerobic digestion inadequate but also there were computa-tional limitations. Thus, the first models were very simple and consisted of a limited numberof equations. During the past thirty years much research has been conducted on the peculiar-ities of the process and on the factors that influence it on the one hand while an enormousprogress took place in computer science on the other. The combination of both parametersresulted in the development of more and more concise and complex models. In this chapterthe most important models found in the literature are described starting from the simplestand oldest to the more recent and complex ones.

Keywords. Anaerobic digestion, Kinetics, Modeling

1 Introduction . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 58

2 Kinetics of Anaerobic Digestion . . . . . . . . . . . . . . . . . . . . . 60

2.1 Microbial Growth Kinetics . . . . . . . . . . . . . . . . . . . . . . . . 602.2 Hydrolysis of Biopolymers . . . . . . . . . . . . . . . . . . . . . . . . 622.3 Acidogenesis . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 662.4 Acetogenesis . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 682.5 Methanogenesis . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 68

3 Modeling of Anaerobic Digestion . . . . . . . . . . . . . . . . . . . . 68

3.1 Models Using Un-Ionized VFA Inhibition as the Primary Key Parameter . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 68

3.2 Models Using Total VFA Inhibition as the Primary Key Parameter . . 733.3 Models Considering the Different Composition of Wastewater . . . . 753.4 Models Using H2 as the Primary Key Parameter . . . . . . . . . . . . 773.5 Models Using NH3 as the Primary Key Parameter . . . . . . . . . . . 843.6 Recent Developments on Anaerobic Digestion Modeling:

Anaerobic Modeling Task Group Work Presentation . . . . . . . . . . 89

4 Conclusions . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 90

5 References . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 91

CHAPTER 6


Abbreviations

ATP adenosine 5-triphosphateCOD chemical oxygen demandCSTR continuous stirred-tank reactorFBR fluidized sand-bed reactorLCFA long-chain fatty acidsNADH reduced form of nicotinamide adenine dinucleotideNAD+ oxidized form of nicotinamide adenine dinucleotideVFA volatile fatty acids

1Introduction

Anaerobic digestion is one of the main processes used for sludge stabilization.Furthermore, anaerobic digestion is widely used for the treatment of manure,industrial wastewaters and the organic fraction of municipal solid waste. Themicrobiology of anaerobic digestion is complicated, since it involves severalbacterial groups, each performing a separate task of the overall degradationprocess. So far, up to nine steps have been identified during the anaerobic con-version of organic matter. However, one can distinguish four main steps andthree major bacterial groups (Fig. 1): the hydrolytic-fermentative bacteria thathydrolyze and convert the organic compounds to volatile fatty acids with thesimultaneous production of hydrogen (H2) and carbon dioxide (CO2), theacetogenic bacteria that convert the above-mentioned acids to acetic acid andfinally the methanogenic bacteria that produce methane, either from acetate orfrom H2 and CO2 .

Anaerobic digestion has the advantages of producing small amounts ofsludge, requiring less nutrients and energy than an aerobic treatment processwhereas the generated biogas can be used as an energy source. Unfortunately,anaerobic systems can be unstable and this instability is usually caused by feedoverload or by the presence of an inhibitor or even by inadequate temperaturecontrol. This is certainly a factor that limits the applicability of anaerobic diges-tion. Therefore, appropriate mathematical models need to be developed in orderto overcome this problem and also to design and operate efficiently anaerobicsystems.

Several models have been developed during the last 35 years. In the first stud-ies on anaerobic process modeling, special attention was paid to the descriptionof the final stage of the anaerobic digestion, methanogenesis, which was consid-ered also as the most important step of the overall process. These models werevery simple and consisted of a limited number of equations. More complicatedmodels describing two or even more bacterial groups and also including inhibi-tion kinetics, pH calculations and gas-phase dynamics came later. Also muchattention has been paid to the modeling of the anaerobic degradation of “syn-thetic substrates”such as glucose. On the other hand, and despite the difficulties,many successful attempts exist on the modeling of the anaerobic degradation of

58 H.N. Gavala et al.

real and complex wastewater. Nowadays, a large number of models can be foundin the literature each of them having its own potential and worth. No unifiedmodeling framework for the anaerobic digestion process exists so far. However,an international anaerobic modeling task group was established in Japan in1997. This group has now formulated a common platform for the establishmentof an anaerobic model.

In the following sections, kinetics and existing models on anaerobic sus-pended growth systems are discussed. At first, general microbial growth kinet-ics is presented and a discussion on hydrolytic kinetics during anaerobic diges-tion process follows. Only some representative kinetics on acidogenesis, aceto-genesis and methanogenesis are included in this chapter since some excellentreviews on this subject have already been published [1–3]. Recently, a review ofsome of the important models for the anaerobic digestion has been published aswell [4].

In this chapter and in order to facilitate the study of the numerous modelsfound in the literature, a classification has been attempted according to the pri-mary key parameter used. Thus five major categories are distinguished: modelsconsidering (a) the non-ionized VFA inhibition, b) the total VFA inhibition,c) the different composition of the wastewaters, d) the H2 as regulator of thevolatile fatty acids production and e) the un-ionized ammonia inhibition. Dueto the complexity of the existing models it could be that some of the modelsdescribed and especially the recent ones use more than one key parameter. As a

Kinetics and Modeling of Anaerobic Digestion Process 59

Fig. 1. Bioconversion of organic matter to methane during the anaerobic digestion process

rule of thumb the first two categories are mostly referring to older models thatuse only one key parameter and consider the organic content of a wastewater asa whole whereas the third category refers to models that gave much attention tothe different composition of wastewaters. On the other hand, most of the mod-els described in fourth and fifth category are more complex and consider morethan one key parameter. Finally, a description of the common platform for ananaerobic model that has been established so far by the international anaerobicmodeling task group is given.

2Kinetics of Anaerobic Digestion

2.1Microbial Growth Kinetics

Cell growth generally involves a respiratory (electron transport phosphoryla-tion, [5]) or a fermentative (substrate-level phosphorylation, [5]) conversion ofthe substrate to products (catabolism) which releases energy in the form ofadenosine 5-triphosphate (ATP). The energy obtained from the catabolic reac-tions is used for both the synthesis of new cells and the maintenance of old ones(anabolism).

Catabolism Substrate Æ Microbial products + EnergyAnabolism Substrate + Energy Æ MicroorganismsMetabolism Substrate Æ Microbial products + Microorganisms

In general, the metabolism of the microorganisms is coupled with the produc-tion of ATP. The ratio of the ATP mass produced per substrate mass consumedis defined as the ATP yield factor, YATP [6]. Accordingly, the biomass yield factorand the product yield factor are expressed as follows:

Biomass yield factor:

DXYX/S = 6 (1)

DS

Product yield factor:

DPYP/S = 6 (2)

DS

where X, S, P symbolize the amounts of the biomass, the substrate and the prod-uct, respectively.

Anaerobic degradation gives low biomass yield factors compared to aerobic;this is due to the low energy (ATP) yield of anaerobic metabolism. In particular,the anaerobic biomass yield factor usually lies between 0.05–0.2 g of biomassproduced per g of substrate consumed whereas the aerobic one could be as highas 0.5 g of biomass produced per g of substrate consumed. The yield factorscould be determined either experimentally or theoretically from the stoichiom-etry of the biochemical reactions.


The bacterial growth is often described by a series of mathematical expres-sions according to the following Eq. (3)

rX = µ (S, X) · X (3)

where rX is the bacterial growth rate and µ symbolizes the specific growth rate ofthe microorganisms.

Monod [7] suggested the following Eq. (4) for the specific growth rate:

µmax · Sµ = 03 (4)

KS + S

where µmax is the maximum specific growth rate achievable when S � KS and KSsymbolizes the saturation constant, meaning the value of the limiting nutrient(substrate) concentration at which the specific growth rate is half its maximumvalue.

By combining the Eqs. (3) and (4) the microbial growth rate is as follows:

µmax · SrX = 03 · X (5)

KS + S

whereas the substrate consumption rate, rS , follows Eq. (6).

1 µmax · SrS = 7 · 03 · X (6)

YX/S KS + S

However, Monod’s equation is incapable of predicting the decrease of the bio-mass concentration that is due to the endogenous respiration and the cell lysis.McCarty [8] developed the following modified Monod equation taking into con-sideration the endogenous respiration and the cell lysis [Eq. (7)]:

µmax · SrX = 03 · X – b · X (7)

KS + S

where b is the specific decay rate (or decay coefficient).In general, the decay coefficient lies around 5% of the maximum specific

growth rate. Nevertheless, the methanogens have a relatively low decay coeffi-cient (almost 1% of their maximum specific growth rate) and thus it can beignored during modeling and simulations of the anaerobic process.

Other expressions of the microbial growth rate as a function of the substrateconcentration are presented in Table 1.

The above expressions are incapable of describing the bacterial growth whenan inhibitory factor is present. In anaerobic digestion many factors could inhib-it the whole process and especially the methanogenesis step. Intermediate prod-ucts such as volatile fatty acids and even compounds that are used as substratecould be inhibitory at high concentration. Also hydrogen sulfide (H2S), ammo-nia (NH3), chlorinated hydrocarbons, aromatic compounds, fatty acids andheavy metals among other compounds are either inhibitory or toxic – depend-ing on their concentration [13]. The most common inhibition types used inanaerobic models are expressed according to Eqs. (8) and (9) and are those of


Haldane [14] first used by Andrews [15] and the non-competitive inhibitiontype, first introduced by Ierusalimsky [16], respectively.

1µ = µmax · 001 (8)

KS I5 + 5 + 1S KI

S KIµ = µmax · 01 · 0 (9)KS + S KI + I

where KI is the inhibition constant and I symbolizes the concentration of theinhibitor.

Haldane inhibition has been used by several researchers for describing theinhibition caused by either the un-ionized volatile fatty acids (butyrate, propi-onate, acetate) or the total volatile acids concentration. Other investigators haveused the non-competitive inhibition type in order to describe the inhibitioncaused by either the volatile fatty acids or other toxic substances, e.g., ammonia.Dinopolou et al. [17] studied the inhibition of acidogenesis by volatile fatty acidsand they concluded that the non-competitive type describes it better. Möscheand Jördening [18] studied the inhibition of acetate and propionate degradationby propionate (substrate inhibition) and the inhibition of propionate degrada-tion by acetate (product inhibition). They concluded that the substrate inhibi-tion is best described by the Haldane equation whereas the non-competitivetype of inhibition describes the product inhibition best.

2.2Hydrolysis of Biopolymers

Hydrolysis means both the solubilization of insoluble particulate matter and thebiological decomposition of organic polymers to monomers or dimers, whichcan pass the cell membrane. Hydrolysis of organic polymers is usually carriedout by extracellular enzymes (hydrolases) and it may or may not be the rate-lim-


Table 1. Expressions of the microbial growth rate as a function of the substrate concentration

Equation Specific growth rate, µ

S n

Moser [9] µmax · 02S n + KS

um · SContois [10] 04B · x + S

µmax · SGrau et al. [11] 02S0

µmax · SChen and Hashimoto [12] 0208K · S0 + (1 – K) · S

iting step of their bioconversion under anaerobic conditions. However, solubi-lization is not necessarily an enzymatic process catalyzed by biologically pro-duced enzymes but could take place due to physicochemical reactions as well. Itis very difficult to describe the whole process by reliable kinetics since hydroly-sis of a complex, insoluble substrate depends on many different parameters suchas the particle size, pH, production of enzymes, diffusion and adsorption ofenzymes to particles.

Hydrolysis of organic polymers is often described by a first-order kineticmodel [Eq. (10)] as the enzymatic activity is not directly coupled to the bacteri-al growth [1, 2]. Nevertheless, it has been reported that a first-order functionmay be most appropriate for complex, heterogeneous substrates, while otherhydrolysis functions may be more appropriate for single homogeneous sub-strates [19]. However, McCarty and Mosey [20] claimed that hydrolysis couldbe considered as a microbial process and, besides first-order kinetics, they sug-gested a “pseudo-Monod” equation with a high saturation constant (5000–10000 mg/L).

rS = Kh · S (10)

where Kh is the hydrolytic constant.A really interesting approach is described in the study of Vavilin et al. [21]

where hydrolysis of complex organic matter is considered as a two-phaseprocess. The first phase is a bacterial colonization, during which the hydrolyticbacteria cover the surface of the solids and its rate depends on the contact areaavailable. Hydrolytic enzymes degrade the solid surface at a constant depth perunit of time during the second phase.According to the aforementioned study thehydrolysis rate is given by Eq. (11).

rS = Kh · SF1/3 · S2/3 (11)

where SF is the concentration of influent biodegradable organic matter. Thehydrolytic constant Kh of the two-phase model is a function of the ratio betweenthe characteristic sizes of bacteria and particles hydrolyzed according toEq. (12).

ÇB dKh = 6rmS · 5 · 4 (12)ÇS dS

where rmS is the maximum specific hydrolysis rate,ÇB and ÇS are the bacterial andparticles densities, respectively, d denotes the depth of the bacterial layer and dSis the current diameter of particles. This approach may be reduced, in some cas-es, to the Contois kinetics [10] (Table 1) as it predicts exponential growth of thehydrolytic biomass at a high solids-to-biomass ratio and first-order kinetics at alow solids-to-biomass ratio.

Information about the hydrolysis of the undissolved part of different waste-waters is presented in Table 2. In all studies first-order hydrolysis kinetics wereassumed. However, in the study of Miron et al. [22] on hydrolysis of the compo-nents of primary sludge, it was reported that none of them followed first-orderhydrolysis. They concluded that hydrolysis still remains the less defined step inthe anaerobic digestion process. Furthermore, Schober et al. [23] concluded that


zero-order kinetics describes better the hydrolysis in the acidogenic reactor dur-ing the two-stage anaerobic digestion of municipal solid organic wastes. On theother hand,Vavilin et al. [24] reported that Contois kinetics (Table 1) are prefer-able to the traditional first-order kinetics when considering the optimal designof a two-stage anaerobic digestion system.

A wide range of hydrolysis rate constants concerning the hydrolysis of carbo-hydrates, proteins and lipids has been reported assuming first-order hydrolysis.Some representative values coming from different studies are presented in thefollowing paragraphs. However, one should take into account that the substratehydrolysis rate depends very much on the origin and the previous acclimationof the anaerobic culture [30–32]. The dependency of the hydrolytic constant onthe previous acclimation of the anaerobic culture coming from the study ofGavala et al. [31] is presented in Table 3.

Many studies on the hydrolysis of carbohydrates under anaerobic conditionshave been made while much attention was paid to the hydrolysis of cellulose inthe rumen [33–36]. The study of O’Rourke [28] on the hydrolysis of cellulose ina continuous, lab-scale reactor gives interesting information on the factors thatinfluence the hydrolytic constant assuming first-order hydrolysis. The results ofthe aforementioned study are presented in Table 4.

Proteins are hydrolyzed by extracellular enzymes, the proteases, intopolypeptides and amino acids. It has been reported in the past that the proteinhydrolysis is a slower process than the carbohydrate hydrolysis [37]. However,


Table 2. Kinetic constant (d–1) values for the hydrolysis of the undissolved part of differentwastewaters

Substrate kh (d–1) Temperature (°C) Reference

Mixture of primary and 0.077 25 calculations from [1]secondary sludge 0.150 35 experimental

results from [25]Primary sludge (from a domestic 0.007–0.990 35–60 [26]wastewater treatment plant)Algae biomass 0.11–0.032 20 [27]Primary sludge (from a domestic 0.4–1.2 35 [28]wastewater treatment plant)Secondary sludge 0.168–0.6 35 [29]

Table 3. Dependence of the hydrolytic constant on the previous acclimation of the anaerobicculture [31]. The substrate was a mixture of piggery, olive mill and dairy wastewater

Inoculum Hydrolysis constant (d–1)

Undissolved proteins Undissolved carbohydrates

Digested piggery wastewater 0.68 0.28Digested olive-mill wastewater 0.35 0.19Digested dairy wastewater 0.24 0.13

the hydrolysis rate depends very much on the solubility of the protein, the pHand the origin of the anaerobic culture. Some representative values of thehydrolytic constant, Kh , (assuming first-order hydrolysis) for the hydrolysis ofcasein, gelatin and corn-protein under anaerobic conditions, are presented inTable 5.

The term “lipids” includes a heterogeneous group of biomolecules which aresoluble in organic solvents of low polarity and not in water. The first step oflipids biodegradation under anaerobic conditions is their hydrolysis by a groupof esterases, the lipases. For example, hydrolysis of one molecule of a phospho-glyceride results in one molecule each of glycerin and phosphoric acid and twomolecules of fatty acids. In the literature, not much information exists on theanaerobic biodegradation of specific lipids. On the contrary, many studies existon the hydrolysis of lipids when considering them as a homogeneous part of theorganic load of a wastewater. The study of O’Rourke [28] on the hydrolysis of thelipid part of the primary sludge in a continuous, lab-scale reactor gives interest-ing information on the factors that influence the hydrolytic constant assumingfirst-order hydrolysis. The results are presented in Table 6.

Christ et al. [40] studied the hydrolysis of carbohydrate, protein and lipidfractions in different organic waste. The ranges of the hydrolysis constant valuesare presented in Table 7. For comparison purposes, the results of the literaturestudy of Gujer and Zehnder [1] on the first-order hydrolysis of complex bio-molecules are presented in Table 7 as well.


Table 4. Hydrolytic constants (d–1) for cellulose hydrolysis as a function of temperature andsolids retention time [1, 28]

Temperature (°C) Solids retention time (days)

5 10 15 30 60

35 1.95 1.21 0.62 0.38 0.2125 0.29 0.27 0.27 0.34 0.1620 0.09 0.14 0.13 0.14 0.1015 – 0.05 0.03 0.10 0.08

Table 5. Representative values of the hydrolytic constant, Kh , for different proteins hydrolysis

Substrate Kh (d–1) Reference

Casein 0.35 [38]Gelatin 0.60 [38]Corn-protein 0.04 [39]

2.3Acidogenesis

During the acidogenesis step the dissolved organic matter is biodegraded main-ly to volatile fatty acids and alcohols by a heterogeneous microbial population.The dominant species in anaerobic digesters is the bacteria while small popula-tions of protozoa, fungi and yeasts have been reported as well [41]. It is mainlythe obligatory and facultative anaerobic bacteria that carry out fermentativeconversion of the substrate to products. Much attention was paid to the acido-genesis of carbohydrates during the last decades and in almost all cases Monodkinetics was assumed. On the contrary, limited data exist on the kinetics ofanaerobic degradation of amino acids despite the fact that the pathways of theiranaerobic biodegradation and their corresponding products have been exten-sively studied. In Table 8 representative kinetics concerning the anaerobicbiodegradation of glucose, cellulose and starch are reported. Many studies alsoexist on the production stoichiometry of the various products coming from car-bohydrates and/or proteins metabolism. Most important factors that influencethe production stoichiometry are the interspecies hydrogen transfer [42–45], thepH [46], the dilution rate [46, 47] and the previous acclimation of the anaerobicculture [32].


Table 6. Hydrolytic constants (d–1) for the hydrolysis of the lipid part of the primary sludge asa function of temperature and solids retention time [1, 28]

Temperature (°C) Solids retention time (days)

5 10 15 30 60

35 0.01 0.17 0.11 0.06 0.0425 0 0.01 0.09 0.07 0.0320 0 0 0.02 0.05 0.0315 – 0 0 0 0

Table 7. Kinetic constants (d–1) for carbohydrate, protein and lipid hydrolysis

Substrate Reference

[40] [1]

Carbohydrates 0.025–0.2 –Cellulose – 0.04–0.13Proteins 0.015–0.075 0.02–0.03Lipids 0.005–0.010 0.08–1.7


Tabl

e8.

Rep

rese

ntat

ive

valu

es o

fkin

etic

con

stan

ts c

once

rnin

g th

e ac

idog

enes

is o

fcar

bohy

drat

es

Cul

ture

acc

limat

ed in

Subs

trat

eµ m

ax (h

–1)

YX

/S(m

gVSS

/mgC

OD

)K

S(g

/L)

Dou

blin

g ti

me

Tem

pera

ture

R

efer

ence

(h)

(°C

)

Prim

ary

slud

gegl

ucos

e0.

30.

150.

42.

336

.5[4

8]ce

llulo

se0.

071

0.16

0.03

689.

8

Synt

heti

c su

bstr

ate

gluc

ose

1.25

0.27

0.02

2537

[49]

com

ing

from

agr

i-cu

ltura

l pro

duct

s

dext

rose

gluc

ose

1.25

0.16

20.

0225

0.5

37[5

0]

Prim

ary

slud

gegl

ucos

e2.

760.

0706

35[5

1]st

arch

1.56

0.59

1gl

ucos

e0.

323

0.49

437

[52]

vari

ous

0.3–

1.25

0.14

–0.1

724

–672

37[2

]ca

rboh

ydra

tes

2.4Acetogenesis

In general, two different types of acetogenic mechanisms can be distinguished:(a) acetogenic hydrogenations and (b) acetogenic dehydrogenations.Acetogenichydrogenations include the production of acetate as a sole end product eitherfrom fermentation of hexoses or from CO2 and H2 . Usually the step of acetoge-nesis in anaerobic digestion refers to acetogenic dehydrogenations and specifi-cally to the anaerobic oxidation of long and short (volatile) chain fatty acids.Obligate proton-reducing or obligate hydrogen-producing bacteria carry outanaerobic oxidation of fatty acids. They are inhibited by even low hydrogen par-tial pressures and consequently they can survive only in syntrophic associationwith microorganisms that consume hydrogen such as the acetoclastic methano-gens. Many studies have been performed so far on the anaerobic oxidation oflong- and short-chain fatty acids. Table 9 includes representative kinetics con-cerning the anaerobic biodegradation of some long-chain fatty acids whilekinetics concerning the bioconversion of propionate and butyrate to acetate arereported in Table 10.

2.5Methanogenesis

A very limited number of organic compounds are used as carbon and energysources supporting growth of methanogenic bacteria. So far, CO2 , CO, formicand acetic acid, methanol, methylamines and dimethyl sulfide have been identi-fied as substrates for methanogenesis. Almost the 65–70% of the methane pro-duced in the anaerobic digesters comes from acetate. On the other handmethanogenesis from CO2 and H2 has a significant role as well by keeping a lowhydrogen pressure and thus supporting the growth of bacteria which carry outanaerobic oxidation of long- and short-chain fatty acids. Methanogenic bacteriaare extremely sensitive to temperature, loading rate and pH fluctuations andthey are inhibited by a number of compounds as has already been reported.Many studies exist so far on the isolation and kinetic characterization of specif-ic methanogenic bacteria utilizing acetate and/or hydrogen. For the purposes ofthis chapter only representative kinetics of methanogenesis from mixed culturesare presented in Table 11.

3Modeling of Anaerobic Digestion

3.1Models Using Un-Ionized VFA Inhibition as the Primary Key Parameter

The first model that takes into consideration the inhibition of methanogenesiscaused by the volatile fatty acids (VFA) is that of Graef and Andrews [63]. Thisstudy considers only one bacterial population, the acetoclastic methanogens.Assuming that all VFA can be represented and expressed in acetic acid units,



Tabl

e9.

Rep

rese

ntat

ive

valu

es o

fkin

etic

con

stan

ts c

once

rnin

g th

e an

aero

bic

oxid

atio

n of

long

cha

in fa

tty

acid

s [2

]

Tem

pera

ture

K

Sµ m

ax

Yb

Ref

eren

ce(°

C)

(mgC

OD

/l)(d

–1)

(mgV

SS/m

gCO

D)

(d–1

)

Prod

ucts

com

ing

from

20

4620

0.13

90.

040.

015

[28]

hydr

olys

is o

flip

ids

2537

200.

171

0.04

0.01

5fo

und

in p

rim

ary

slud

ge35

2000

0.25

20.

040.

015

Satu

rate

d lo

ng-c

hain

fatt

y ac

ids

37(m

ean

valu

e)[5

3]m

yris

tic

(C14

)10

50.

105

0.11

0.

01pa

lmit

ic (C

16)

143

0.11

00.

110.

01st

eari

c (C

18)

417

0.08

50.

110.

01

Uns

atur

ated

long

-cha

in fa

tty

acid

s37

[53]

olei

c (C

18)

3180

0.44

0.11

0.01

linol

eic

(C18

)18

160.

550.

110.

01


Tabl

e10

.R

epre

sent

ativ

e va

lues

ofk

inet

ic c

onst

ants

con

cern

ing

the

bioc

onve

rsio

n of

prop

iona

te a

nd b

utyr

ate

to a

ceta

te

Subs

trat

eK

inet

ics

for

Tem

pera

ture

K

Sµ m

ax

Yb

Ref

eren

ce(°

C)

( mgC

OD

/l)(d

–1)

( mgV

SS/m

gCO

D)

(d–1

)

Prop

iona

tepr

opio

nate

2511

440.

051

0.04

[54]

Prop

iona

tepr

opio

nate

3578

0.04

20.

01Bu

tyra

tebu

tyra

te35

130.

047

0.02

7G

luco

sebu

tyra

te37

298

0.86

––

[55]

Prop

iona

tepr

opio

nate

35

170.

13–

–[5

6](H

RT:1

4.5d

)Pr

opio

nate

prop

iona

te

3549

91.

2–

–(H

RT:8

.2d)

Ace

tate

:pro

pion

ate:

[57]

Buty

rate

= 2

:1:1

mix

ed a

cids

3516

60.

414

0.03

00.

099

Buty

rate

buty

rate

6012

0.45

0.01

9–

[58]

Prop

iona

tepr

opio

nate

3311

––

–[5

9]

Tabl

e11

.R

epre

sent

ativ

e va

lues

ofk

inet

ic c

onst

ants

con

cern

ing

met

hano

gene

sis

from

ace

tate

and

hyd

roge

n in

ana

erob

ic m

ixed

cul

ture

s

Cul

ture

acc

limat

ed in

Met

hano

gene

sis

from

Tem

pera

ture

Y

X/S

KS

µ max

Ref

eren

ce(°

C)

(mgV

SS/m

gCO

D)

(mgC

OD

/l)(d

–1)

Mun

icip

al w

aste

wat

erac

etat

e35

[60]

Mun

icip

al w

aste

wat

erac

etat

e25

0.05

093

00.

25[5

4]M

unic

ipal

was

tew

ater

acet

ate

300.

054

356

0.27

5[5

4]M

unic

ipal

was

tew

ater

acet

ate

350.

041

165

0.35

7[5

4]M

unic

ipal

was

tew

ater

acet

ate

36.5

0.28

50.

49[4

8]M

unic

ipal

was

tew

ater

acet

ate

3320

[59]

Glu

cose

acet

ate

3719

8[6

1]A

ceta

teac

etat

e37

49[6

1]M

unic

ipal

was

tew

ater

hydr

ogen

300.

07–0

.109

11–6

9 m

gCO

D/l/

h[6

2]


Graef and Andrews developed the following stoichiometry [Eq. (13)] for theirconversion to methane.

CH3COOH + 0.032NH3 Æ 0.032C5H7NO2 + 0.92CO2 + 0.92CH4 + 0.096H2O(13)

where the empirical formula C5H7NO2 corresponds to the composition of metha-nogenic bacteria.

The growth of biomass and the substrate consumption is assumed to followinhibition kinetics according to Eq. (8). Both the substrate and the inhibitor arethe un-ionized VFA (AcH) expressed as acetic acid. The concentration of the un-ionized form is calculated according to acetate dissociation reaction:

AcH ¤ Ac– + H+ (14)

The model includes a total ion balance and takes into consideration three phas-es, gas, liquid and biological (solid) phase. Methane is considered to be waterinsoluble, whereas the carbon dioxide produced is partly dissolved and partlyescapes to the gas phase. This model has been used to simulate digester start-upand response to organic overload and was able to predict digester failure dueto a temporary accumulation of VFA, which lowers the pH and subsequentlyincreases the concentration of un-ionized VFA. The introduction of a first-orderequation describing the rate of microorganisms’ death in the model [Eq. (15)]gives us the possibility to predict digester failure due to toxic substances.

rK = KT · TX (15)

where rK is the rate of organisms death due to the toxic substance, KT is the tox-icity rate constant and TX is the concentration of the toxic compound.

Hill and Barth [64] developed a model describing the animal waste digestion(Fig. 2). Their model considers two microbial groups, the acid-formers andthe acetoclastic methane-formers and it is using un-ionized VFA (expressed asacetic acid) inhibition of both microbial groups. Furthermore, it considers ahydrolytic step and ammonia (un-ionized) inhibition of methane-formers.This double inhibition has been incorporated into the growth kinetics of themethanogens according to the Eq. (16).

µmaxµ = 0000 (16)KS AcH NH31 + 8 + 8 + 8AcH KI, 1 KI, 2

Kleinstreuer and Poweigha [65] and Marsili-Libelli and Nardini [66] publishedsimulation studies of a digester receiving soluble and insoluble organic com-pounds, respectively. Their models are based on un-ionized VFA inhibition ofmethanogenesis and the basic steps considered are presented in Fig. 3.

Moletta et al. [67] developed a model for the anaerobic digestion of glucose(Fig. 4). It considers two steps: glucose biodegradation to acetate and methano-genesis from acetate and it is based on un-ionized VFA inhibition of both acido-genesis and methanogenesis. The model simulated satisfactorily the methaneproduction during batch experiments with pea bleaching wastewater and with asynthetic medium consisting of sucrose and acetic acid.


Fig. 2. Block diagram of the Hill and Barth [64] mathematical model

Fig. 3. Block diagram of the Kleinstreuer and Poweigha [65] (a) and the Marsili-Libelli andNardini [66] (b) mathematical models

Fig. 4. Flow chart of the Moletta et al. [67] model

A model developed by Smith et al. [68] considers three steps (Fig. 5) assum-ing that the insoluble organic material used as feedstock (biomass) consists oftwo components, a rapidly and a slowly biodegradable one. In the first step, thetwo insoluble biomass components are converted to soluble intermediates thatserve as substrate for volatile fatty acid-producing bacteria during the secondstep. Finally in the third step, the methanogenic bacteria convert volatile fattyacids to methane and carbon dioxide. Smith et al. used two inhibition types: a)the un-ionized volatile fatty acids inhibition of the methanogenesis step and b)the total volatile fatty acids inhibition of the acidogenesis step. However, themodel has not been experimentally verified.

Finally, Märkl [69] published a study concerning the quantitative analysis ofthe modern biogas reactor systems. This study was based on the model of Graefand Andrews [63] and especially on its physicochemical assumptions.

3.2Models Using Total VFA Inhibition as the Primary Key Parameter

The first model considering total VFA inhibition was that of Hill [70]. This mod-el was developed in order to simulate methanogenesis from animal wastes. Themodel considers five bacterial groups and four steps (Fig. 6). During the firststep complex organic material enters the digester and is converted by extracel-lular enzymes to soluble, biodegradable organic matter.A set of “biodegradabil-ity constants” from another study [71] was used in this hydrolysis stage. Duringthe second step (acidogenesis), the soluble organic matter is biodegraded main-ly to butyrate, propionate and acetate. In the third step (acetogenesis) acetate isproduced from butyrate and propionate whereas the fourth step (methanogen-esis) refers to the methane production from acetate and hydrogen. All fivebacterial groups catalyzing the three last steps are assumed to be inhibited bytotal VFA concentration. This inhibition is expressed both in the growth rateaccording to Eq. (8) and microbial decay rate according to Eq. (17), which wasdeveloped by Hill et al. [72].

bmaxb = 04 (17)Kb1 + 8VFA

where b is the specific decay rate [Eq. (7)], bmax is the maximum specific decayrate and VFA is the concentration of total VFA.


Fig. 5. Flow chart of the Smith et al. [68] model

Comparison of steady-state predictions with experimental data from six-teen pilot and full-scale biogas plants digesting animal wastes validated themodel of Hill (1982). The same model was used for the design of param-eters and operating characteristics of swine and poultry [73], beef cattle[74] and dairy cattle manure [75] anaerobic digestion systems. Additionally,Durand et al. [76] used the Hill’s (1982) model for predicting the steady stateand dynamic performance of swine manure digesters and a satisfactoryagreement between the experimental results and the theoretical predictionswas noted.

The model of Kalyuzhnyi [77] is the last one of a series of models [78–82]dealing with anaerobic digestion of glucose (Fig. 7). It consists of five steps andconsiders five bacterial groups: the acidogens that ferment glucose to butyrate,propionate, acetate and ethanol at an experimentally defined stoichiometrywith the simultaneous production of hydrogen and carbon dioxide, the obligateproton reducers that convert butyrate and propionate to acetate, the ethanoldegrading acetogens, the acetoclastic methanogens and finally the hydrogeno-trophic methanogens. All steps are pH-dependent according to the pH function[Eq. (18)] of Angelidaki et al. [83]. Hydrogen inhibition of the acidogenic andboth acetogenic steps, acetate inhibition of the butyrate-degrading step andethanol and butyrate inhibition of both methanogenic steps is taken into con-


Fig. 6. Flow chart of the Hill [70] model

sideration. Model validation has been made using batch kinetic experimentswith glucose.

1 + 2 · 100.5(pKl–pKh)

F (pH) = 00002 (18)1 + 10(pH – pKh ) + 10(pKl – pH)

where pKl and pKh denote the lower and upper pH values where the microbialgrowth rates are approximately 50% of the uninhibited rate.

3.3Models Considering the Different Composition of Wastewater

The models described so far do not take the different compositions of waste-water into account. Specifically, models dealing with the anaerobic digestion ofcomplex wastewaters, such as animal slurries, considered the organic content asa whole, which was hydrolyzed and degraded with an experimentally measuredrate. This assumption on the one hand simplifies the models and makes theiruse easier, but on the other hand limits their applicability since such models areuseful only for the anaerobic digestion of the specific wastewater they are devel-oped for.

The first model suggesting that the complex biodegradable particulate part ofa wastewater is hydrolyzed to protein, carbohydrates and lipids and subsequentlyto amino acids, simple sugars and fatty acids respectively, is that of Bryers [84].However, due to the lack of information, the particulate matter, proteins, carbo-hydrates and lipids are considered as a whole, while the amino acids and simple


Fig. 7. Flow chart of the Kalyuzhnyi [77] model

sugars are lumped together (Fig. 8). Consequently, in the first step hydrolysistakes place resulting in amino acids, sugars and fatty acids while in the secondand third steps intermediates such as propionate, butyrate and acetic acid arederived from amino acids/sugars and fatty acids acidogenesis. The fourth stepcorresponds to the acetogenesis while the fifth and sixth steps are the methano-genesis from acetate and hydrogen, respectively. The model predicted fairly wellthe observations in two different experimental systems that treated biomassparticulates anaerobically after some parameter optimization concerning initialbacterial compositions.

In 1996 a mathematical model for the codigestion of piggery, olive-mill anddairy wastewaters in continuous stirred tank reactors (CSTR) was developed byGavala et al. [85]. It was assumed that wastewaters consist mainly of carbohy-drates and proteins (undissolved and dissolved) and other dissolved organicmatter (fatty acids and lipids are the major constituents of the latter). The mod-el considers four steps and three bacterial groups, the acidogens, the acetogensand the acetolytic methanogens (Fig. 9). The model was based on batch kineticexperiments and is capable of predicting adequately the COD and fatty acidsdependence on the operating conditions when an anaerobic culture acclimatedto a specific wastewater starts being fed a mixture of other agroindustrialwastewaters [86]. Thus, the model can be useful for predicting the short-termresponse of a digester subjected to feed changes and thus avoiding system fail-


Fig. 8. Flow chart of the Bryers [84] model

ure, as well as for maximizing the co-digestion process efficiency. The changesof anaerobic culture’s biological characteristics during the acclimation processto each one of the three aforementioned wastewaters were determined as well[31]. It was found that a dairy acclimated culture was characterized by thehigher total biomass percentage in acidogens, whereas methanogen and aceto-gens concentrations were higher in the piggery and olive-mill acclimated cul-tures, respectively. On the other hand, different bacterial species of acidogenshad predominated in each culture. This suggests that the composition of thewastewater in different organic compounds (carbohydrates, proteins, lipids etc.)should definitely be taken into account when designing anaerobic digestionprocesses.

In the study of Jeyaseelan [87] a model for anaerobic digestion of municipalwastewater was proposed considering its composition in carbohydrates, pro-teins, lipids and other organics. However, this study was a theoretical one sinceno experimental part is involved and the kinetic parameters used were collect-ed from the available literature.

Some other models consider the different composition of wastewater as well[88–92]. However, these models use different key parameters than the onesalready reported and thus they are presented in the next two sections.

3.4Models Using H2 as the Primary Key Parameter

The first model using H2 as the key parameter that regulates the production offatty acids from glucose is that of Mosey [93]. The main topic of this study is theinvestigation and the expression of the effect that the dissolved hydrogen con-centration has on the regulation of the redox potential inside the bacterial celland subsequently on the produced mixture of volatile fatty acids.

The model considers four bacterial groups: a) the fast-growing (minimumdoubling times around 30 minutes) acid-forming bacteria that ferment glucoseto a mixture of butyrate, propionate and acetate, b) the slow-growing (minimumdoubling times of 1.5–4 days) acetogenic bacteria that convert the butyrate andpropionate to acetate,c) the slow-growing (minimum doubling times of 2–3 days)acetoclastic methanogenic bacteria that produce methane from acetate and d)


Fig. 9. Flow chart of the Gavala et al. [85] model

the relatively fast-growing (minimum doubling times around 6 hours) hydro-gen-utilizing methanogens that produce methane from carbon dioxide.

The key assumption made by Mosey is that the formation of butyrate, propi-onate and acetate from glucose is regulated by the availability of hydrogen.The main hypothesis of his study is that the relative availability of the reduced(NADH) and oxidized form (NAD+) of the carrier molecule nicotinamide ade-nine dinucleotide controls both the overall rate of the conversion and the com-position of the mixture of acids formed. The ratio [NADH]/[NAD] is a functionof the hydrogen partial pressure in the gas phase and it is expressed accordingto Eq. (19).

[NADH] [NADH]04 = 1500 · PH2 or 04 = 1.5 · 10–3 · H (19)[NAD] [NAD]

where PH2 is the partial pressure of hydrogen in the gas phase and H is the con-centration of hydrogen (ppm by volume) in the digester gas. Eq. (19) appliesunder the following restrictions: a) the bacteria maintain a constant internal pHvalue of 7.0 regardless of the variations in the pH value of their growth mediumand b) gaseous hydrogen diffuses both freely and rapidly into and out of the bac-terial cells. Consequently, the partial pressure of hydrogen inside the bacteria isthe same as the partial pressure of hydrogen in the digester gas and the redoxpotential of the bacteria is the same as the potential of the growth medium.

The model considers hydrogen-regulated production of volatile fatty acidsfrom glucose via the Embden-Meyerhof-Parnas metabolic pathway and alsohydrogen regulated acetogenesis from butyrate and propionate. It includes yieldequations for all the four bacterial groups that are involved in the overall processof glucose fermentation. The yield equations are based on the production of ATPat each step and on the assumption that one mole of ATP provides sufficientenergy for the formation of about 10 g of biomass [94].

The metabolic pathways inside acid-forming bacteria are shown in Fig. 10.Steps and reaction rates are shown in Tables 12, 13 and 14.

Mosey’s model combines in a very fine way biochemical and microbiologicalconsiderations. It is the first model that takes into consideration the varied stoi-chiometry of the produced fatty acids during the acidogenesis step of glucose.According to this model and during stress situations that usually result in tohigh H2 concentrations, acidogenesis is diverted to the production of propionate(and butyrate) which then are degraded slowly.This is consistent with the exper-imentally observed persistency of propionate concentrations during stress situ-ations. Nevertheless, the Mosey model has not been experimentally verified.

Thereafter, many models were based on the assumption that H2 is the keyparameter that regulates the production of fatty acids from glucose. The studyof Pullammanapallil et al. [95] is mainly based on Mosey’s model and addition-ally it includes a description of the gas phase and acetoclastic inhibition byundissociated fatty acids.

Costello et al. [96] developed a model taking into consideration that lacticacid may also be an important intermediate in the anaerobic degradation of glu-cose. Their model consists of five bacterial groups (Fig. 11) and the main differ-ences between this model and the model of Mosey [93] is that the glucose is con-



verted to a mixture of butyric, acetic and lactic acids with the subsequent degra-dation of the lactate to propionic and acetic acids. These two steps are inhibitedand regulated by hydrogen through the redox reactions of NAD. The conversionof butyrate and propionate to acetate is subjected to hydrogen inhibition aswell. pH inhibition functions were applied to each group of bacteria whileproduct inhibition was taken into consideration for the acid-forming, lacticacid and acetogenic bacteria by using non-competitive and competitive inhi-bition terms. In the study of Costello et al. [97] an attempt at experimental veri-fication of the aforementioned model was made by using independent setsof experimental data. In some cases the model was able to provide a reasonablecomparison between the theoretical predictions and the experimental results;however, in most cases an under-estimation of the propionic and butyric acidconcentrations as well as an over-estimation of the total biogas flow rate wasobserved.

Keller et al. [98] and Romli et al. [99] used the model of Costello et al. [98] forthe prediction and simulation of experimental results coming from a two-stagehigh-rate anaerobic wastewater treatment system fed with diluted molasses.This system consisted of a continuous stirred tank reactor (CSTR) as the acidi-

Fig. 10. Schematic biochemical pathways of the Mosey [93] model. The model of Pullam-manapallil et al. [95] is based on this scheme as well


Tabl

e12

.St

oich

iom

etri

c re

acti

ons,

rate

and

cel

ls y

ield

equ

atio

ns fo

r th

e ac

idog

enes

is s

tep

duri

ng g

luco

se fe

rmen

tati

on

Stoi

chio

met

ric

reac

tion

sR

ate

and

cells

yie

ld e

quat

ions

d[g

luc]

RG

k G· X

G· [

gluc

]0

3=

046

whe

reas

RG

= 0

04

dt[N

AD

H]

(K

G+

[glu

c])

1 +

04

[NA

D+]

d[b

ut]

R

GC

12H

12O

6Æ

CH

3CH

2CH

2CO

OH

+ 2

CO

2+

2H

2+

2AT

P0

3=

046

00

05

dt[N

AD

H]

2[N

AD

+]

�1 +

04�· �1

+ 0

4�

[NA

D+]

[NA

DH

]

d[p

rop]

2R

GC

12H

12O

6+

2H

2Æ

2CH

3CH

2CO

OH

+ 2

H2O

+ 2

ATP

022

= 0

46

00

05

dt[N

AD

H]

[NA

D+]

�1 +

04�·

�1 +

04�

[NA

D+]

[NA

DH

]

d[a

cet]

2R

G2R

GC

12H

12O

6Æ

H2O

Æ2C

H3C

OO

H +

2C

O2

+ 4

H2

+ 4

ATP

021

=0

46

61–

046

00

02

dt[N

AD

H]

2[N

AD

H]

2

[NA

D+]

�1 +

04�

�1 +

04�· �1

+ 0

4�

[NA

D+]

[NA

D+]

[NA

DH

]

dXG

d[a

cet]

d

[pro

p]

[but

] 7

= Y

¢ acet· 7

6+

Y¢ pro

p· 7

7+

Y¢ bu

t· 7

2– X

G· b

dt

dt

dt

dt

Whe

reR

G:

unre

gula

ted

rate

ofu

ptak

e of

gluc

ose

(mm

oles

/L/d

).[p

rop]

:co

ncen

trat

ion

ofpr

opio

nic

acid

(mM

).k G

:m

axim

um r

ate

cons

tant

(mm

oles

/mg/

d).

[but

]:co

ncen

trat

ion

ofbu

tyri

c ac

id (m

M).

KG

:M

icha

elis

-typ

e co

nsta

nt (m

M).

Y¢ ace

t:bi

omas

s yi

eld

coef

ficie

nt (2

0m

g/m

M a

ceta

te).

XG

:co

ncen

trat

ion

ofgl

ucos

e fe

rmen

ters

(m

g/L)

.Y

¢ prop:

biom

ass

yiel

d co

effic

ient

(10

mg/

mM

pro

pion

ate)

.[g

luc]

:co

ncen

trat

ion

ofgl

ucos

e (m

M).

Y¢ bu

t:bi

omas

s yi

eld

coef

ficie

nt (2

0m

g/m

M b

utyr

ate)

.[a

cet]

:co

ncen

trat

ion

ofac

etic

aci

d (m

M).

b:de

cay

coef

ficie

nt (0

.2/d

ay).


Tabl

e13

.St

oich

iom

etri

c re

acti

ons,

rate

and

cel

ls y

ield

equ

atio

ns fo

r th

e ac

etog

enes

is s

tep

duri

ng g

luco

se fe

rmen

tati

on

Stoi

chio

met

ric

reac

tion

sR

ate

and

cells

yie

ld e

quat

ions

d[b

ut]

RB

k B· X

B· [

but]

CH

3CH

2CH

2CO

OH

+ 2

H2O

Æ 2

CH

3CO

OH

+ 2

H2

+ 2

ATP

02=

029

whe

reas

RB

= 0

228

dt[N

AD

H]

(K

B+

[but

])1

+ 0

41

[NA

D+]

dXB

d[b

ut]

7=

YB

· 02–

XB

· bdt

dt

d[p

rop]

RP

k P· X

P· [

prop

]C

H3C

H2C

OO

H +

2H

2O Æ

CH

3CO

OH

+ C

O2

+ 3

H2

+ 1

ATP

77

= 0

0w

here

as R

P=

00

5dt

[NA

DH

]

(K

P+

[pro

p])

1 +

76

1[N

AD

+]

dXP

d[p

rop]

7=

YP

· 0

4–

XP

· bdt

dt

Whe

reR

B,R

P:

unre

gula

ted

rate

ofu

ptak

e of

buty

rate

and

pro

pion

ate

(mm

oles

/L/d

).k B

,kP:

max

imum

rat

e co

nsta

nt fo

r bu

tyra

te a

nd p

ropi

onat

e (m

mol

es/m

g/d)

.K

B,K

P:

Mic

hael

is-t

ype

cons

tant

for

buty

rate

and

pro

pion

ate

(mM

).X

B,X

P:

conc

entr

atio

n of

buty

rate

and

pro

pion

ate

utili

zers

(mg/

L).

YB:

biom

ass

yiel

d co

effic

ient

(20

mg/

mM

but

yrat

e).

YP:

biom

ass

yiel

d co

effic

ient

(10

mg/

mM

pro

pion

ate)

.b:

deca

y co

effic

ient

(0.2

/day

).


Table 14. Stoichiometric reactions, rate and cells yield equations for the methanogenesis stepduring glucose fermentation

Stoichiometric reactions Rate and cells yield equations

d [acet] kA · XA · [acet]CH3COOH Æ CH4 + CO2 03 = 004dt (KA + [acet])

dXA d [acet]7 = YA · 76 – XA · bdt dt

d [H2] kH · XH · [H]4H2 + CO2 + CH4 + 2H2O 01 = 704dt KH + [H]

dXH d [H2]8 = YH · 74 – XH · b

dt dt

Where kA , kH: maximum rate constant for acetate and hydrogen (mmoles/mg/d).KA: Michaelis-type constant for acetate (mM).KH : Michaelis-type constant for hydrogen expressed as partial pressure

of hydrogen in the digester gas (atm).XA , XH: concentration of acetate and hydrogen utilizers (mg/L).YA: biomass yield coefficient (2.5 mg/mM acetate).YH: biomass yield coefficient (2.5 mg/mM hydrogen).b: decay coefficient (0.2/day).[H2]: concentration of hydrogen gas in the solution (mM).[H]: partial pressure of hydrogen in the gas (atm).

fication reactor and a fluidized sand-bed reactor (FBR) as the methanogenicreactor. They stated – although not giving details – that a physico-chemical reac-tion system was included in the model to calculate the pH at any time given theconcentration of the acidic and basic species in the reactor. The model was pret-ty well capable to predict the pH, the alkali consumption rate at the acidificationreactor, the gas generation and composition and the effluent concentration inorganic acids when hydraulic step changes at various organic loading rates weretaking place. On the contrary, theoretical predictions were not in good agree-ment with experimental results when a recycle stream from the second to thefirst reactor was introduced [98].Also the system responses in combination withthe model predictions during pH changes and shock loads are described inRomli et al. [100, 101].

In the study of Ruzicka [102] an extension of Mosey model is proposed tak-ing into account the fact that in some cases formation of higher acids thanbutyrate takes place during saccharide fermentation. Other modifications alsoare suggested in Ruzicka [103] and are based on hydrogen transport across thecell membrane as well as on reaction phenomena in the cell involving hydrogenand NAD. This interesting approach, mentioned by the author as a “nonequilib-rium concept”, could explain the contradictory experimental observations how,on the one hand, small amounts of hydrogen could completely inhibit glucosecleavage and how glucose uptake could proceed even in a medium fully saturat-


ed with hydrogen, on the other. Unfortunately, no experimental validation of thismodel exists so far.

The model of Batstone et al. [92] was developed in order to simulate theanaerobic degradation of complex wastewaters such as the slaughterhouse efflu-ent. It is a complex model consisting of nine generic biological groups and threeenzymatic groups (Fig. 12). In this study the catabolic reactions are as follows:the lipids,particulate proteins and carbohydrates are hydrolyzed by exo-enzymesto long-chain fatty acids and simple sugars, soluble proteins and carbohydrates,respectively. Acetic acid is coming from the further biodegradation of long-chain fatty acids via b-oxidation whereas acidogenesis of proteins results inacetate, propionate, butyrate and valerate. The pathway of protein degradation isassumed to be via coupled Stickland reactions. The soluble carbohydrate degra-dation follows H2 regulation of VFA formation as suggested by Mosey [93] andthe pathways are the same as used by Costello et al. [96], thus resulting in acetate,butyrate and lactate. Subsequently, the lactate is biodegraded to propionic andacetic acids. All the aforementioned volatile fatty acids are biodegraded toacetate and finally the acetoclastic and the H2-utilizing methanogens produce

Fig. 11. Schematic biochemical pathways of the Costello et al. [96] model. The studies of Kelleret al. and Romli et al. [98, 99, 101] are based on this scheme as well

methane. The degradation rates of the substrates are subjected to hydrogen andpH inhibition. The model includes physico-chemical reactions determining thepH of the liquid phase and the gas-liquid transfer of carbon dioxide. Concen-trations of saline and free organic components as well as ammonia and carbondioxide are calculated using acid-base equilibrium relationships. The model wasvalidated by performing experiments with a two-stage, high-rate anaerobictreatment plant that treated wastewater from a pig slaughterhouse [104].

3.5Models Using NH3 as the Primary Key Parameter

Ammonia is considered as a major inhibitor of the methanogenesis processespecially when animal wastes are being digested. The inhibition is mainlycaused by the un-ionized form of ammonia and thus it is very much dependenton the pH value. Detailed description of the ammonia chemistry, diffusion andrelease from liquid manure can be found in the review of Ni [105].The first mod-el using un-ionized ammonia inhibition is that of Hill and Barth [64] alreadydescribed earlier in this chapter. During the last decade many models have beendeveloped including ammonia inhibition as one of the key parameters and mostof them focus on the anaerobic digestion of animal wastes. Kiely et al. [106]developed a model for anaerobic digestion of piggery slurry, primary sewage


Fig. 12. Schematic biochemical pathways of the Batstone et al. [92] model

sludge and the organic fraction of municipal solid waste. The model was basedon those of Hill and Barth [64] and Moletta [67] (Fig. 4). It consisted of twostages (hydrolysis/acidogenesis producing acetate and methanogenesis) andconsiders un-ionized VFA inhibition of both steps and un-ionized ammoniainhibition for the methanogenesis step [Eq. (16)].The model was validated usingexperimental results from lab-scale continuous stirred-tank reactors anddespite its simplicity the theoretical results fitted satisfactorily the experimen-tal ones.

Digesters fed with substrate with a high ammonia content, such as manure,exhibit a self-regulation of the pH and self-resistance on un-ionized ammoniatoxicity that can be described as follows: when free ammonia concentrationexceeds the toxicity threshold, inhibition of methanogenesis occurs which leadsto an accumulation of VFA with a subsequent decrease of pH and thus reductionof the free ammonia concentration. This mechanism tends to stabilize theprocess at a certain volatile fatty acids concentration and pH level. The firstmodel that mentioned this mechanism is that of Angelidaki et al. [83]. It is acomplex model (Fig. 13) consisting of four microbial groups: the glucose fer-menting acidogens, the propionate degrading acetogens, the butyrate degrading


Fig. 13. Flow chart of the Angelidaki et al. [83] model including pH and ammonia regulationscheme

acetogens and the acetoclastic methanogens. The primary substrates are solubleand insoluble carbohydrates whereas a fraction (almost 32%) of the total ammo-nia is bound to the insoluble fraction and is released during hydrolysis. Thestoichiometry of the microbial reactions is based on the study of Hill [70] withminor modifications to some coefficients. Equilibrium relationships for ammo-nia, carbon dioxide and pH as well as gas phase dynamics and temperatureeffects are included. Total VFA inhibition of hydrolysis, total acetate inhibition ofacetogenesis and free ammonia inhibition of methanogenesis is assumed in themodel. The type of inhibition used is the non-competitive one [Eq. (9)]. For thelast two steps, acetogenesis and methanogenesis, the effect of the pH on themicrobial growth rate was described by a Michaelis pH function, normalized togive a value of 1.0 as center value [Eq. (18)]. The model was validated with ther-mophilic (55 °C) laboratory scale anaerobic digestion experiments with cattlemanure where the feed concentration of ammonia was increased from 2.5 to6.0 g-N/l. Experimental results and theoretical simulations showed on the onehand that the process was inhibited but stabilized at a lower level of methaneproduction on the other.

The above-mentioned model was extended in order to be able to simulate theanaerobic treatment of cattle manure with olive oil mill effluent [107]. Two morebacterial groups were added to the model: the lipolytic and the long-chain fattyacids-degrading bacteria (Fig. 14). Subsequently, the model was extended oncemore for simulating the anaerobic bioconversion of complex substrates to bio-


Fig. 14. Extension of the Angelidaki et al. [83] model in order to include lipid and proteinanaerobic biodegradation [91, 107]

gas [91]. Hydrolysis of undissolved proteins and two more bacterial groups (theamino-acid- and valerate-degrading acidogens and acetogens, respectively) arealso considered (Fig. 14). Specifically, the substrate in this model is defined by itsorganic and inorganic composition. The organic part consists of carbohydrates,proteins, lipids and volatile fatty acids. The inorganic part includes ammonia-N,phosphate-P, carbonate-C, hydrogen sulfide, anions and cations (i.e., Ca2+, Mg2+,and K+). The latter plays an important role in determining the pH and bufferbalance of the process. Product inhibition of VFA degradation to acetate wasfurther included along with inhibition of all the bacterial groups by the long-chain fatty acids. The model was validated with thermophilic (55 °C) experi-ments in (a) laboratory-scale CSTRs receiving a mixture of cattle manure andglycerol trioleate or gelatin and (b) full-scale reactors receiving a mixture ofcattle manure, bentonite-bound oil and a proteinaceous wastewater comingfrom bone extraction.

Another model that considers the ammonia inhibition of acetate conversionto methane is that of Siegrist et al. [89, 90]. This model was developed in orderto describe the dynamic behavior of the mesophilic anaerobic digestion ofsewage sludge according to the reaction scheme shown in Fig. 15. Gujer andZehnder [1] first developed this reaction scheme. The model consists of fivemicrobial groups: the acidogens that ferment the amino acids and sugars to


Fig. 15. Reaction scheme for anaerobic digestion of domestic sewage sludge [1, 89, 90]

butyrate, propionate and acetate, the microorganisms that perform the b-oxida-tion of fatty acids to hydrogen and acetate, the acetogens that produce acetatefrom butyrate and propionate and finally the acetogenic and hydrogen-utilizingmethanogens. The model includes equilibrium relationships for ammonia, car-bon dioxide and pH and gas phase dynamics as well. The propionate and acetatedegradation and hydrogen consumption are subjected to pH dependence.Addi-tionally, propionate and fatty acid degradation have non-competitive inhibitionterms for increased hydrogen pressure and acetate concentration whereas foracetate conversion a non-competitive inhibition term for free ammonia isincluded. The model is verified with load variation experiments in laboratoryand full-scale digesters.

Vavilin et al. [88] developed a model named “methane” in order to describethe anaerobic digestion of complex organic matter. According to Fig. 16, themodel considers hydrolysis of the biodegradable organic matter to dissolvedorganic substrate by hydrolytic enzymes released by acidogenic bacteria.Acetate- and propionate-producing bacteria with the simultaneous release ofhydrogen, ammonia and carbon dioxide consume the dissolved organic sub-strate, which is a mixture of carbohydrates, proteins and lipids. Propionate isbiodegraded to acetate, which is used for the production of methane along with


Fig. 16. Schematic biochemical pathways of the Vavilin et al. [88] model

the hydrogen. Simultaneously, the sulfate-reducing bacteria, resulting in therelease of hydrogen sulfide, use the propionic and acetic acids. The inhibitioneffect of non-ionized molecules of ammonia, sulfide and propionate was takeninto consideration as well.

The innovation brought by this model is the consideration of two moregroups of bacteria than the ones already reported in this chapter: the bacteriathat perform sulfate reduction coupled with the utilization of acetate and propi-onate, respectively. Also, the aforementioned model takes into account that themetabolism of the carbohydrates, proteins and lipids is regulated by the hydro-gen partial pressure and thus the latter controls the relative production of pro-pionate and acetate. This is accomplished by the consideration of two differentgroups of acidogens: the propionate- and the acetate-producers.

The model “methane” was used successfully for the simulation of the resultsof batch experiments on anaerobic digestion of food industry wastewater. In1996, the same scientists developed a theory describing the hydrolysis as a two-phase process. The model “methane” was used to compare the results of fourtypes of hydrolytic kinetics during anaerobic digestion of swine waste, sewagesludge, cattle manure and cellulose. Experimental results from different studieswere used for this purpose [21].

3.6Recent Developments on Anaerobic Digestion Modeling:Anaerobic Modeling Task Group Work Presentation

An international anaerobic modeling task group was established in Japan in1997 in order to formulate a common platform for the establishment of ananaerobic model (http://www.awmc.uq.edu.au/admtg/tg). This task group hasnow formulated an anaerobic model and the key assumptions of this genericmodel are presented below. Conversion processes taken into account are biolog-ical and physico-chemical ones and the carbon flow chart is presented in Fig. 17.Acidogenesis, acetogenesis and methanogenesis are the three main biologicalsteps whereas the degradation of complex particulate matter is considered as acombination of an extracellular, partly non-biological disintegration step and aextracellular biological hydrolysis step. All extracellular steps follow first-orderkinetics [Eq. (10)] and substrate uptake and biomass growth and decay proceedaccording to Eqs. (6) and (7), respectively. The stoichiometric yields of volatilefatty acids during the acidogenesis step of carbohydrates are considered to beconstant with no hydrogen regulation function suggested so far. Additionally,Stickland pathways are considered for the estimation of products yield fromamino acids. pH influence on all biological processes is included according tothe empirical function proposed by Angelidaki et al. [Eq. (18)]. Hydrogen inhi-bition of acetogenesis from long-chain fatty acids (LCFA), propionate, butyrateand valerate and for hydrogenotrophic methanogenesis is proposed along withnon-competitive free ammonia inhibition of acetolytic and hydrogenotrophicmethanogenesis. Physico-chemical processes such as liquid-liquid transforma-tions (ion association/dissociation) and gas-liquid transfers of CO2, CH4 and H2

are included in the model since they are considered very important for anaero-


bic systems. The IWA Anaerobic Digestion Model No 1 was presented during the9th World Congress on Anaerobic Digestion [108]. However, discussions aboutthe final version of the model are still under way.

4Conclusions

A systematic classification of the mathematical models describing the anaerobicdigestion process in suspended growth systems was made in this chapter. Thiseffort focused on the overview of the most important models found in the liter-ature so that the reader can acquire the knowledge of how the mathematicalmodeling of the anaerobic digestion developed throughout the years. It is a well-known fact that anaerobic digestion is a complex process affected and regulatedby many factors. Also, anaerobic digestion is applied to the treatment of a wide


Fig. 17. Flow chart of the generic anaerobic model as suggested by the international anaerobicmodeling task group

range of wastes/wastewater with significant differences in their composition.In order to have appropriate mathematical models that combine simplicity and accuracy and are applicable to almost every type of waste/wastewater, anintegrated approach should focus on the a) formulation of a common platform,which will take into account all the important parameters and factors in-fluencing the process, b) classification of the wastes/wastewater based on theirdifferent composition in organic and inorganic compounds and c) formulationof a model for each category selecting each time the key elements from thecommon platform, e.g., NH3 effect should be taken into account in case of waste-water with high concentration of NH3 or proteins; on the other hand the intro-duction of the NH3 effect into a mathematical model concerning anaerobicdigestion of dairy wastewater will only add complexity with no practical advan-tage. In this way, it will be possible to fully exploit the information and knowl-edge gained through decades of scientific research on the area of anaerobicdigestion process.

5References

1. Gujer W, Zehnder AJB (1983) Water Science and Technology 15:1272. Pavlostathis SG, Giraldo-Gomez E (1991) Water Science and Technology 24:353. Pavlostathis SG, Giraldo-Gomez E (1991) Critical Reviews in Environmental Control

21:4114. Lyberatos G, Skiadas IV (1999) Global Nest: the International Journal 1:635. Gottschalk G (1986) Nutrition of bacteria. In: Bacterial metabolism. Springer, Berlin,

Heidelberg, New York6. Gottschalk G (1986) Bacterial fermentations. In: Bacterial metabolism. Springer, Berlin,

Heidelberg, New York7. Monod J (1949) Ann Rev Microbiol 3:3718. McCarty PL (1966) Kinetics of waste assimilation in anaerobic treatment. In: Devel-

opments in Industrial Microbiology.American Institute of Biological Sciences,Washing-ton DC

9. Moser H (1958) Carnegie Institute Washington Publ. No. 614:10. Contois DE (1959) Journal of General Microbiology 21:4011. Grau P, Dohanyos M, Chubota J (1975) Water Res 9:63712. Chen YR, Hashimoto AG (1978) Biotechnology And Bioengineering Symp 8:26913. Bitton G (1994) Anaerobic digestion of wastewater and sludge. In: Wastewater micro-

biology. Wiley-Liss, New York14. Haldane JBS (1930) Enzymes. Longmans, London15. Andrews JF (1969) J Sanit Engng Div Am Soc Civ Engrs SA1:9516. Ierusalimsky ND (1967) Bottle-necks in metabolism as growth rate controlling factors.

In: Powell EO, Evans CGT, Strange RE, Tempest DW (eds), Microbial Physiology and Con-tinuous Culture, 3rd International Symposium. Her Majesty’s Stationery Office, London

17. Dinopoulou G, Sterritt RM, Lester JN (1988) Biotechnol Bioeng 31:96918. Mosche M, Jordening HJ (1999) Water Res 33:254519. Eastman JA, Ferguson JF (1981) J WPCF 53:35220. McCarty PL, Mosey FE (1991) Water Science and Technology 24:1721. Vavilin VA, Rytov SV, Lokshina LYa (1996) Bioresource Technol 56:22922. Miron Y, Zeeman G, van Lier JB, Lettinga G (2000) Water Res 34:170523. Schober G, Schaefer J, Schmid-Staiger U, Troesch W (1999) Water Res 33:85424. Vavilin VA, Rytov SV, Lokshina LYa, Rintala JA, Lyberatos G (2001) Water Res 35:424725. Pfeffer JT (1968) J Water Pollution Control Federation 40:1933


26. Pfeffer JT (1974) BioTechnol and Bioengineering 26:77127. Foree EG, McCarty PL (1969) Proc 24th Ind Waste Conf, Purdue University 1328. O’Rourke JT (1968) PhD thesis, Stanford University, California29. Ghosh S (1981) BioTechnol And Bioengineering Symp 11:30130. Doyle O, O’Malley J, Clausen E, Gaddy J (1983) Kinetic improvements in the production

of methane from cellulosic residues. In: Energy from biomass and wastes. 7:54631. Gavala HN, Skiadas IV, Lyberatos G (1999) Water Science and Technol 40:33932. Gavala HN, Lyberatos G (2001) BioTechnol and Bioengineering 74:18133. Colberg PJ (1988) Anaerobic microbial degradation of cellulose, lignin, oligolignols and

monoaromatic lignin derivatives. In: Zehnder AJB (ed), Biology of Anaerobic Microor-ganisms. Wiley, New York

34. Stack RJ, Cotta MA (1986) Applied and Environmental Microbiology 52:20935. Pavlostathis SG, Miller TL, Wolin MJ (1988) Applied and Environmental Microbiology

54:256636. Pavlostathis SG, Miller TL, Wolin MJ (1988) Applied and Environmental Microbiology

54:266037. Heukelekian H (1958) Basic principles of sludge digestion. In: McCabe J, Eckenfelder

WW (eds), Biological Treatment of Sewage and Industrial Wastes. Reinhold, New York38. Nagase M, Matsuo T (1982) BioTechnol and Bioengineering 24:222739. Greco RL, Coto JM, Dentel SK, Gosset JM (1983) Technical report. Environmental Engi-

neering department, Cornell University, Ithaca, New York40. Christ O,Wilderer PA,Angerhofer R, Faulstich M (2000) Water Science and Technol 41:6141. Toerien DF, Hattingh WHJ (1969) Water Res 3:38542. Iannotti EL, Kafkewitz D, Wolin MJ, Bryant MP (1973) J Bacteriol 114:123143. Miller TL, Wolin MJ (1973) J Bacteriol 116:83644. Thauer RK, Jungermann K, Decker K (1977) Bacteriol Rev 41:10045. Cohen A, van Gemert JM, Zoetemeyer RJ, Breure AM (1984) Process Biochemistry

(December) 22846. Zoetemeyer RJ, van den Heuvel JC, Cohen A (1982) Water Res 16:30347. Kisaalita WS, Lo KV, Pinder KL (1989) Biotechnol and Bioengineering 34:123548. Ghosh S, Klass DL (1978) Process Biochemistry 13:1549. Pohland FG, Ghosh S (1971) Biotechnol And Bioengineering Symp 2:8550. Ghosh S, Pohland FG (1974) J WPCF 46:74851. Noike T, Endo G, Chang J-E, Yaguchi J-I, Matsumoto J (1985) BioTechnol and Bioengi-

neering XXVII:148252. Huang CJ (1983) The effect of dilution rate on the kinetics of anaerobic acidogenesis. Pro-

ceedings of the thirteenth Annual Biochemical Engineering Symposium, Reilly PJ (ed)53. Novak JT, Carlson DA (1970) J WPCF 42:193354. Lawrence AW, McCarty PL (1969) J WPCF 41:R155. Massey ML, Pohland FG (1978) J WPCF 50:220556. Heyes RH, Hall RJ (1983) Applied and Environmental Microbiology 43:71057. Lin C-Y, Sato K, Noike T, Matsumoto J (1986) Water Res 20:38558. Ahring BK, Westermann P (1987) Applied and Environmental Microbiology 53:42959. Kaspar HF, Wuhrmann K (1978) Applied and Environmental Microbiology 36:160. Smith PH, Mah RA (1966) Applied Microbiology 14:36861. Aguilar A, Casas C, Lema JM (1995) Water Res 29:50562. Robinson JA, Tiedje JM (1982) Applied and Environmental Microbiology 44:137463. Graef SP, Andrews JF (1973) AIChe symposium series. Water 70:10164. Hill DT, Barth CL (1977) J WPCF October:212965. Kleinstreuer C, Poweigha T (1982) BioTechnol and Bioengineering XXIV:194166. Marsili-Libelli S, Nardini M (1985) Environ Technol Lett 6:60267. Moletta R, Verrie D, Albagnac G (1986) Water Res 20:42768. Smith PH, Bordeaux FM, Goto M, Shiralipour A,Wilkie A,Andrews JF, Ide S, Barnett MW

(1988) Biological production of methane from biomass. In: Smith WH, Frank JR (eds),Methane from Biomass. A Treatment Approach. Elsevier, London


69. Markl H (1999) Modeling of biogas reactors. In: Rehm HJ, Reed G (eds), Biotechnology.Wiley-VCH, Weinheim

70. Hill DT (1982) Transactions of the ASAE 25:137471. Hill DT (1983) Agricultural wastes 5:172. Hill DT, Tollner EW, Holmberg RD (1983) Agricultural wastes 5:10573. Hill DT (1983) Agricultural wastes 5:15374. Hill DT (1983) Agricultural wastes 5:20575. Hill DT (1983) Agricultural wastes 5:21976. Durand JH, Iannotti EL, Fischer JR, Miles JB (1988) Biological wastes 24:177. Kalyuzhnyi SV (1997) Bioresource Technol 59:24978. Kalyuzhnyi SV, Davlyatshina MA (1997) Bioresource Technol 59:7379. Kalyuzhnyi SV, Davlyatshina MA, Varfolomeev SD (1994) Applied Biochemistry And

Microbiology 30:16280. Kalyuzhnyi SV, Davlyatshina MA, Varfolomeev SD (1994) Applied Biochemistry And

Microbiology 30:2081. Kalyuzhnyi SV, Gachok VP, Davlyatshina MA, Varfolomeyev SD (1993) Applied Bio-

chemistry And BioTechnol 39:60182. Kalyuzhnyi SV, Gachok VP, Sklyar VI,Varfolomeev SD (1991) Applied Biochemistry And

BioTechnol 28–29:18383. Angelidaki I, Ellegaard L, Ahring BK (1993) BioTechnol and Bioengineering 42:15984. Bryers JD (1985) BioTechnol and Bioengineering XXVII:63885. Gavala HN, Skiadas IV, Bozinis NA, Lyberatos G (1996) Water Science and Technol 34:6786. Lyberatos G, Gavala HN, Stamatelatou A (1997) Nonlinear Analysis 30:234187. Jeyaseelan S (1997) Water Science and Technol 35:18588. Vavilin VA, Vasiliev VB, Ponomarev AV, Rytov SV (1994) Bioresource Technol 48:189. Siegrist H, Renggli D, Gujer W (1995) Mathematical modelling of anaerobic mesophilic

processes in a digester. In: International meeting on anaerobic processes for bioenergyand environment. Colle – Colle, Copenhagen, Denmark

90. Siegrist H, Renggli D, Gujer W (1993) Water Science and Technol 27:2591. Angelidaki I, Ellegaard L, Ahring BK (1999) BioTechnol and Bioengineering 63:36392. Batstone DJ, Keller J, Newell RB, Newland M (2000) Bioresource Technol 75:6793. Mosey FE (1983) Water Science and Technol 15:20994. Bauchop T, Elsden SR (1960) J of General Microbiology 23:45795. Pullammanapallil P, Owens JM, Svoronos SA, Lyberatos G, Chynoweth DP (1991) AIChe

Annual Meeting 4396. Costello DJ, Greenfield PF, Lee PL (1991) Water Res 25:84797. Costello DJ, Greenfield PF, Lee PL (1991) Water Res 25:85998. Keller J, Romli M, Lee PL, Greenfield PF (1993) Water Science and Technol 28:19799. Romli M, Keller J, Lee PL, Greenfield PF (1994) Advances in Bioprocess Engineering 379

100. Romli M, Keller J, Lee PL, Greenfield PF (1994) Water Science and Technol 30:35101. Romli M, Keller J, Lee PJ, Greenfield PF (1995) Process Safety and Environmental Protec-

tion 73:151102. Ruzicka M (1996) Water Res 30:2440103. Ruzicka M (1996) Water Res 30:2447104. Batstone DJ, Keller J, Newell RB, Newland M (2000) Bioresource Technol 75:75105. Ni J (1999) J of Agricultural Engineering Research 72:1106. Kiely G, Tayfur G, Dolan C, Tanji K (1997) Water Res 31:534107. Angelidaki I, Ellegaard L, Ahring BK (1997) Water Science and Technol 36:263108. Batstone DJ, Keller J, Angelidaki RI, Kalyuzhny SV, Pavlostathis SG, Rozzi A, Sanders

WTM, Siegrist H,Vavilin VA (2001) 9th World Congress on Anaerobic Digestion,Antwer-pen, September 2–6 Proceedings of the Workshop on ADM 1

Received: April 2002


Molecular Biology of Stress Genes in Methanogens:Potential for Bioreactor Technology

Everly Conway de Macario · Alberto J. L. Macario

Wadsworth Center, Division of Molecular Medicine, New York State Department of Health;and Department of Biomedical Sciences, School of Public Health, The University at Albany(SUNY), Albany, New York 12201-0509, USA. E-mail: [email protected]

Many agents of physical, chemical, or biological nature, have the potential for causing cellstress. These agents are called stressors and their effects on cells are due to protein denatura-tion. Cells, microbes, for instance, perform their physiological functions and survive stressonly if they have their proteins in the necessary concentrations and shapes. To be functional aprotein shape must conform to a specific three-dimensional arrangement, named the nativeconfiguration. When a stressor (e.g., temperature elevation or heat shock, decrease in pH,hypersalinity, heavy metals) hits a microbe, it causes proteins to lose their native configura-tion, which is to say that stressors cause protein denaturation. The cell mounts an anti-stressresponse: house-keeping genes are down-regulated and stress genes are activated. Among thelatter are the genes that produce the Hsp70(DnaK), Hsp60, and small heat protein (sHsp) fam-ilies of stress proteins. Hsp70(DnaK) is part of the molecular chaperone machine togetherwith Hsp40(DnaJ) and GrpE, and Hsp60 is a component of the chaperonin complex. Both thechaperone machine and the chaperonins play a crucial role in assisting microbial proteins toreach their native, functional configuration and to regain it when it is partially lost due tostress. Proteins that are denatured beyond repair are degraded by proteases so they do not accumulate and become a burden to the cell. All Archaea studied to date possess chaperonins but only some methanogens have the chaperone machine. A recent genome sur-vey indicates that Archaea do not harbor well conserved equivalents of the co-chaperonestrigger factor, Hip, Hop, BAG-1, and NAC, although the data suggest that Archaea have proteinsrelated to Hop and to the NAC alpha subunit whose functions remain to be elucidated. Otheranti-stress means involve osmolytes, ion traffic, and formation of multicellular structures. Allcellular anti-stress mechanisms depend on genes whose products are directly involved incounteracting the effects of stressors, or are regulators. The latter proteins monitor and mod-ulate gene activity. Biomethanation depends on the concerted action of at least three groupsof microbes, the methanogens being one of them. Their anti-stress mechanisms are brieflydiscussed in this Chapter from the standpoint of their role in biomethanation with emphasison their potential for optimizing bioreactor performance. Bioreactors usually contain stres-sors that come with the influent, or are produced during the digestion process. If the stressorsreach levels above those that can be dealt with by the anti-stress mechanisms of the microbesin the bioreactor, the microbes will die or at least cease to function. The bioreactor will mal-function and crash. Manipulation of genes involved in the anti-stress response, particularlythose pertinent to the synthesis and regulation of the Hsp70(DnaK) and Hsp60 molecularmachines, is a promising avenue for improving the capacity of microbes to withstand stress,and thus to continue biomethanation even when the bioreactor is loaded with harsh waste.The engineering of methanogenic consortia with stress-resistant microbes, made on demandfor efficient bioprocessing of stressor-containing effluents and wastes, is a tangible possibilityfor the near future. This promising biotechnological development will soon become a realitydue to the advances in the study of the stress response and anti-stress mechanisms at the molecular and genetic levels.

CHAPTER 6


Keywords. Stress, Stress genes, Methanogens, Anti-stress mechanisms, Multicellular struc-tures

1 Introduction . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 98

1.1 Terminology, Stress, Heat-Shock Proteins, Chaperones,and Anti-Stress Mechanisms . . . . . . . . . . . . . . . . . . . . . . 98

1.2 Objectives . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 991.3 Scope . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 1001.4 Life on Earth, Evolution, and Stressors . . . . . . . . . . . . . . . . 1001.5 Methanogens . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 101

2 Stress Genes . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 102

2.1 Definition . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 1022.2 Classification . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 1022.3 Evolution . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 1022.4 Strategies and Methods for Study and Utilization of Stress Genes . . 103

3 Molecular Chaperones . . . . . . . . . . . . . . . . . . . . . . . . . 103

3.1 Definition and Functions . . . . . . . . . . . . . . . . . . . . . . . 1033.2 Interactions . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 1043.3 Families . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 1043.4 Mechanisms and Manipulation . . . . . . . . . . . . . . . . . . . . 105

4 Stress Response . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 105

4.1 Definition . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 1054.2 Characteristics . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 1054.3 Stressors . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 106

5 Stress Genes and Molecular Chaperones in Archaea . . . . . . . . . 106

5.1 Overview . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 1065.2 Evolution . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 1075.3 Structure . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 1095.4 Expression and Regulation . . . . . . . . . . . . . . . . . . . . . . . 109

6 The Hsp70(DnaK) Chaperone Machine in Methanogens . . . . . . 109

6.1 Components . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 1096.2 Expression . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 1106.3 Stressor-Response Relationships . . . . . . . . . . . . . . . . . . . 1146.4 Other Stressors Pertinent to Methanogenic Bioreactors . . . . . . . 1146.5 Factors that Modify the Stress Response . . . . . . . . . . . . . . . 1166.6 Other Methanogens . . . . . . . . . . . . . . . . . . . . . . . . . . 1196.7 Co-Chaperones . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 123

96 E. Conway de Macario · A. J. L. Macario

7 The Hsp60 (Chaperonin) System in Methanogens . . . . . . . . . . 123

7.1 Examples . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 1237.2 Structure and Potential for Bioreactor Technology . . . . . . . . . . 124

8 Other Stress or Stress-Related Molecules, Genes and Proteins,and Anti-Stress Mechanisms in Methanogens . . . . . . . . . . . . 125

8.1 Examples . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 1258.2 Osmolytes . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 1258.3 TrkA . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 1258.4 Prefoldin or GimC . . . . . . . . . . . . . . . . . . . . . . . . . . . 1278.5 Small Heat-Shock Proteins (sHsp) . . . . . . . . . . . . . . . . . . . 1288.6 PPIase . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 1298.7 Proteases . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 1298.8 Putative Stress Genes and Proteins Found in Fully Sequenced

Genomes . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 130

9 Other Manifestations of the Stress Response . . . . . . . . . . . . . 130

9.1 Introduction . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 1309.2 Thermoprotectants . . . . . . . . . . . . . . . . . . . . . . . . . . . 1319.3 Multicellular Structures . . . . . . . . . . . . . . . . . . . . . . . . 131

10 Perspectives and Applications . . . . . . . . . . . . . . . . . . . . . 138

10.1 Introduction . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 13810.2 Diversity of Methanogens . . . . . . . . . . . . . . . . . . . . . . . 13810.3 Dynamics of Methanogenic Subpopulations in Bioreactors . . . . . 13810.4 Diversity of Stressors . . . . . . . . . . . . . . . . . . . . . . . . . . 14110.5 Diversity of Response . . . . . . . . . . . . . . . . . . . . . . . . . 14110.6 Diversity of Methanogens: A Source of Useful Microbes? . . . . . . 14210.7 Cooperation Between Molecules and Between Cells . . . . . . . . . 14510.8 Proteases as Builders . . . . . . . . . . . . . . . . . . . . . . . . . . 14510.9 Intrinsic Stress Resistance . . . . . . . . . . . . . . . . . . . . . . . 146

11 Conclusion and Perspectives . . . . . . . . . . . . . . . . . . . . . 146

12 References . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 147

Molecular Biology of Stress Genes in Methanogens: Potential for Bioreactor Technology 97

1Introduction

1.1Terminology, Stress, Heat-Shock Proteins, Chaperones, and Anti-Stress Mechanisms

Terms will be defined in the text when pertinent. A few are introduced here toabate the sense of awe some uninitiated readers might experience when theyencounter specialized words for the first time. This ought to make the Chaptermore “user friendly.”

Stress is a word applied in many fields of intellectual endeavor and does notneed a lengthy explanation. It will be used in this Chapter to indicate an alteredstatus of the cell caused by an agent (stressor) of a physical, chemical, or biolog-ical nature [1]. This status is the stress response, which manifests itself by anincrease in the products of a series of genes (stress genes). These produce thestress proteins, also called for historical reasons, heat-shock proteins, abbreviat-ed Hsp [2]. While the genes’ names are written in italics, those of their productsare in Roman characters.

A gene, when active, is transcribed to messenger RNA (mRNA), which in turnis translated in the ribosome into a linear series of amino acids to form apolypeptide or protein. Thus, when one measures intracellular levels of themRNA from a given gene, an estimate of the degree of activity of that gene isobtained. mRNA levels may increase in a cell also via other mechanisms, butthese will be explained in the text.

A protein is synthesized as a string of amino acids but to become functional-ly competent it has to fold into a three-dimensional configuration, i.e., it has toaccommodate itself to what is called the native configuration for that particularprotein. There are many occasions in which this functional configuration is par-tially or totally lost, a process named protein denaturation. Denatured proteinsmust be renatured, or eliminated, lest they cause serious problems to the cell.The major effect of stressors is protein denaturation. Consequently, the stressresponse is made of events unchained by protein denaturation, and many ofthese events aim at counteracting it.

Some Hsp are molecular chaperones [2]. They assist other polypeptides toacquire their native configurations, or to regain them if they have been partial-ly denatured. Proteins that are denatured beyond repair are eliminated by pro-teases, i.e., enzymes that digest the damaged molecules [3–5].

There are several groups of Hsp, but the two most studied are those that con-stitute the molecular chaperone machine and the chaperonin system [6–11].Both will be explained in the text.


NOTE. This review was completed in January 2000. Since then other genomes have beensequenced and new publications have appeared. Most relevant is a genome survey in search ofarchaeal co-chaperones, a discussion of which was added to this article, after completion ofthe original manuscript. Other updates may be found in Frontiers of Bioscience, Vol. 5 and 6,Special Issue, Anti-stress mechanisms in Archaea, at:http://www.bioscience.org/current/special/macario.htm; Genome Res. 12:532–542, 2002; andCrit. Rev. Biochem. Mol. Biol., December 2002

Stress genes may be active in the absence of recognizable stress. This physio-logical activity is called basal or constitutive to distinguish it from that inducedby stressors. The two gene activities, constitutive and stress-induced, are impor-tant to the cell. They have to be properly regulated, so the cell can perform all itsfunctions under physiological conditions, and can react speedily and efficientlyin the face of stress, and survive. It follows that the gene-regulatory mechanismsare as important to the cell as the stress genes themselves. Gene regulation ismediated by proteins called transcription and regulatory factors that are theproduct of other genes. Some of these factors interact with critical DNAsequences named cis-acting sites or elements, usually located near the genesthey regulate.

There are other components of the stress response beyond stress genes andchaperones. A few will be discussed in this Chapter, such as simple compoundsthat protect the cell against stressors and are called thermoprotectants, and mul-ticellular structures [12]. The latter structures allow cells to survive adverse con-ditions as they are shielded inside a large body surrounded by extracellularmaterial.

It should be easy to envisage from the above general considerations howimportant anti-stress mechanisms are for biomethanation. The microbial cellsinvolved in the bioconversion of wastes in bioreactors are constantly exposed tostressors. If the cells are not prepared to withstand these attacks by stressorsthey will die, or at least cease to function.

There is a very promising future for the use of stress genes and proteins in theoptimization of biomethanation technology.A rational application of these anti-stress means must be based on adequate knowledge of the mechanisms involvedin transcription and regulation of stress genes. The aim of this Chapter is to pre-sent in a simplified manner part of what is known about the stress response in agroup of microbial cells that are key to biomethanation.

1.2Objectives

This Chapter, as stated above, will introduce the topic defined in the title to non-specialists in molecular biology or stress. Data will be presented in Tables withonly a brief explanation in the text, in which selected references will be quotedfor consultation by those interested in more details. Moreover, recent reviews onthe stress genes and proteins will be quoted extensively because they containpractically all the information available at the end of 1999 [8, 12, 13]. In addition,the few pertinent publications that have appeared since then will be discussed.

The information on stress genes and proteins will be discussed mainly toindicate how it might be pertinent to methanogenesis and, more specifically, tothe designing, monitoring, improving, and controlling of anaerobic methan-ogenic bioreactors (digestors).


1.3Scope

An attempt will be made to cover all aspects of the stress response, and the stressgenes and proteins that might be applicable to bioreactor technology, focusingon the methanogens. However, the emphasis will lie on the systems that are bet-ter known, namely the molecular chaperone machine and the chaperonins. Inaddition, a rather detailed discussion of multicellular structures that play a crit-ical role in resistance to stressors will be presented.

Likewise, we will focus on two organisms, Methanosarcina mazeii andMethanosarcina thermophila, whose stress response and molecular chaperone-machine genes are the best studied among methanogens [12, 14]. They are keyfor methanogenesis in bioreactors and in many other ecosystems of biotechno-logical relevance [15–19]. M. mazeii has an optimal temperature for growth(OTG) of 37°C and is a key element for mesophilic digestion, while the other,M. thermophila, has an OTG of 50°C and plays a central role in thermophilicbioreactors.

This Chapter is based on an extensive search of written and electronic litera-ture, and on consultations with colleagues, but only a minimal list of essentialreferences will be quoted and work done in our laboratories will be primarilysurveyed.As stated above, there are comprehensive reviews available that may beused by the reader to advantage, and that allow us to simplify this Chapter, andto direct its focus on matters that are, or might be, relevant to methanogenesisand bioreactor technology.

1.4Life on Earth, Evolution, and Stressors

Life on Earth today resulted from the evolution of organisms that survivedthrough the changes in temperature, pH, oxygen levels, etc. that the planet expe-rienced since the most primitive life form appeared. Since these environmentalchanges, and many others that must have occurred during the millennia, are cellstressors, one is inclined to think that stress genes have played a decisive role indetermining which organisms survived. It would seem reasonable to think thatlife forms endowed with the best anti-stressor means (among which stress genesand proteins are paramount) would be those that withstood the stressful condi-tions when they appeared more or less suddenly, until the conditions eitherchanged back to a more benign character or the organism adapted itself to thenew situation. In the latter case, the stressful conditions were no longer such.They became normal, physiological conditions for the adapted organism. Thismechanism may be one of the reasons why we see today such a variety of habi-tats that are optimal for their aboriginal organisms [20], but may be deadly toforeigners. This is a crucial concept while defining stressors. An agent, forinstance a certain level of temperature, may be a potent stressor for a givenorganism while it may be optimal for the growth and physiology of another [1].The same applies to acidity, alkalinity, salinity, etc. It is with these ideas in mindthat one may now look at the list of common cell stressors displayed in Table 1.


1.5Methanogens

All living cells have been classified into three main lines of evolutionary descentbased mostly on comparative analyses of the sequences of the small subunit ofribosomal RNA (rRNA) [21–24].The lines or phylogenetic domains are: Archaea(formerly archaebacteria), Bacteria (eubacteria), and Eucarya (eukaryotes).Methanogens belong to the Archaea, and consequently they are prokaryoteswith many characteristics that distinguish them from the other prokaryotes, thebacteria.

Methanogens are very important organisms for a number of reasons. Themost pertinent to this Chapter is that they are key elements in the bioconversionof organic matter with the generation of methane gas, i.e., biomethanation [17,19, 25]. They are widespread, diverse, and capable of metabolizing a range of dif-ferent substrates. Therefore, they are useful in waste treatment under a varietyof conditions, and have the potential for application in the bioconversion of arange of materials in many different locations in our planet.A major objective ofthis Chapter is to indicate possibilities for improving methanogens, and to sug-gest means to do it, so they become very resistant to stressors and thus able tofunction efficiently even in very inhospitable environments.


Table 1. Cell stressors a

Type Name description b

Physical Heat; several types of irradiation, including ultraviolet light; pressure;sound

Oxygen H2O2; oxygen-derived free radicals or reactive oxygen species (ROS);anaerobiosis to aerobiosis shift; hypoxia-anoxia

pH Alkalinity; acidity; pH shiftOsmotic Changes in the concentration of salt, sugars, other osmolytes

(hyper- or hypoosmotic shock)Nutritional Starvation: multiple; specific (carbon, glucose, nitrogen, phosphate,

nitrate)Antibiotics Puromycin; tetracycline; nalidix acidAlcohols Ethanol; methanol; butanol; propanol; octanolMetals Cadmium; copper; chromium; zinc; tin; aluminum; mercury; lead; nickelInsecticides, Lindane; diazinon; paraquat; thiram; tributyltin

pesticidesMechanical Compression; shearingOther Benzene and derivatives; phenol and derivatives; mutagens; ammonia;

arsenite; arsenate; amino acid analogues; nicotine; anesthetics;desiccation

a Reproduced modified from reference [1] with permission from the copyright owner.b These agents cause stress in cells from the three phylogenetic domains, Bacteria, Archaea

(both prokaryotes), and Eucarya (eukaryotes).

2Stress Genes

2.1Definition

In a broad sense, stress genes are those that work during and after a stressor hitsa cell differently as compared with the pre-stress situation [1].Stress genes beginto be transcribed, or increase their transcriptional activity, in response to thestressor’s impact on the cell. In contrast to other cellular genes, the stress genesare activated upon stress. Non-stress genes are down-regulated or shut off bystress.

In agreement with a broad definition, stress genes form a large group encom-passing those that produce the molecular chaperones and chaperonins (themost representative of the group) [2] and many others [12, 26]. Among the lat-ter are those that participate in the formation of multicellular structures, ofwhich we know very little despite their obvious importance in cell survival andfunction.

2.2Classification

The stress genes and their proteins (Hsp) are classified according to the molec-ular mass of the latter as shown in Table 2. The Hsp70 and Hsp60 families are thebest known in Archaea, including methanogens [8, 12, 27, 31, 32]. There is alsoinformation on the components of the molecular chaperone machine other thanHsp70(DnaK), namely the Hsp40(DnaJ) and GrpE proteins, particularly in M.mazeii and M. thermophila, as we shall discuss later in this Chapter. While thesegene/protein families are very important and their study should continue if onewants to develop means to improve methanogens for biotechnologic purposes,another important group shown in Table 2 is “Other”. Under this heading, there


Table 2. Classification of Hsp into families according to their molecular mass a

Family Name Mass (kDa) Found in methanogens(synonyms within parentheses) b

Heavy (High M.W.; Hsp100) 100 or higher No c

Hsp90 81–99 NoHsp70 (Chaperones) 65–80 YesHsp60 (Chaperonins) 55–64 YesHsp40 (DnaJ) 35–54 YesSmall Hsp (sHsp) <35 YesOther (proteases, etc.) Various Yes

a Reproduced from reference [2] with permission from the copyright owner.b For additional information on the various families see references [7, 10, 11, 27–36].c Not yet investigated, or investigated but not yet found, or found but incompletely character-

ized (see also references [8, 37]).

are several gene/proteins that are surely crucial for cell survival in the face ofstress but that are not yet well understood [12]. This group offers a great poten-tial for research and as a source of genes and molecules to engineer more resis-tant methanogens as we shall see in other sections of this Chapter.

2.3Evolution

The evolutionary history of stress genes and proteins is extremely interestingfrom both the theoretical and practical standpoints.The topic has been reviewedin detail [8, 12, 32, 38–40], and we shall not deal with it here again. We encour-age the reader to consult the references quoted, and others cited in them, tobecome acquainted with the evolution of these important genes, and to under-stand why they are useful for constructing evolutionary trees, and how they canbe used for such a purpose.

What are the practical dividends of understanding stress gene/protein evolu-tion? It provides a comprehensive picture of their presence and variations inmany organisms, and thus their functions and adaptations in a variety of cir-cumstances. Thus, it helps us to understand the functions of the stress proteinsas a whole, and those of their specialized domains, which in turn is instrumen-tal to designing strategies for engineering more efficient genes and moleculesthat will “fortify” methanogens pertinent to bioreactors.

2.4Strategies and Methods for the Study and Utilization of Stress Genes

Because of the diversity of methanogens mentioned above, including the capac-ity to grow in a wide variety of ecosystems, the methods for studying them arevery varied [21, 25, 41, 42]. It follows that the methods for manipulating thestress genes and proteins from methanogens are also diverse and may be quitecomplex.

A wide array of species possibly also exists, with many families, but this typeof diversity has not yet been fully assessed because only a limited number of iso-lates have been thoroughly characterized.We shall come back to this topic in thecourse of the Chapter to show the array of possibilities that are available for dis-covering useful methanogens and for obtaining useful genes and proteins.

3Molecular Chaperones

3.1Definition and Functions

A molecular chaperone is a protein that assists others to fold correctly, refold ifpartially denatured (unfolded) by stressors, and translocate to the cell’s locale(e.g., the periplasm in bacteria, or the mitochondria and the endoplasmic retic-ulum in eukaryotes) where they reside and function [6, 9, 43]. They are also


implicated in the degradation of intracellular proteins that are damaged orabnormal to the extent that they might aggregate and precipitate, and thus causeserious problems to the cell [3–5].

In addition to the functions listed above, molecular chaperones have otheractivities, the list of which grows rapidly as more research is done on theseimportant molecules. For example, they participate in the regulation of theirown synthesis [44–46] and in the disaggregation and refolding of partially dena-tured polypeptides [47–49], and some of them interact with nucleic acids [50].Furthermore, it has been suggested that molecules other than proteins (e.g.,lipids) can play a role as chaperones in protein folding [51].

In this Chapter, the molecular chaperones that are the product of stress geneswill be treated in more detail. However, others that are less well studied will alsobe mentioned to stimulate the curiosity of the reader and, hopefully, to promoteresearch aiming at clarifying their functional role. This should provide a widerbasis for biotechnologic developments than that furnished by the analysis of theclassical molecular chaperones exclusively.

3.2Interactions

Molecular chaperones usually exercise their functions by interacting with othermolecules that are chaperones themselves, or that act as co-chaperones or co-factors, forming complex multimolecular assemblies [6, 8, 9, 35, 43, 52].Examples are the molecular chaperone machine formed by the chaperonesHsp70(DnaK), Hsp40(DnaJ), and GrpE; and the thermosome or chaperonincomplex, as we shall see later in this Chapter. Also, the molecular chaperonemachine may interact with chaperonins in the process of de novo polypeptidefolding. Thus, in the chaperoning process and in the regulation of their own syn-thesis, chaperones do not act alone but in association with teammates. It followsthat if one aims at, for example, controlling the levels of stress proteins in a cellto make it more resistant to stressors over periods of time longer than usual, onemust understand their self-regulatory circuits and the array of other moleculesinvolved in this mechanism.

3.3Families

As mentioned above, the Hsp, many of which are molecular chaperones, are clas-sified into families according to their molecular mass, see Table 2. It is importantto bear in mind that while many molecular chaperones are stress proteins in thesense that their levels are increased by stress, others are not. Also, the reverse istrue, namely, many stress proteins are not molecular chaperones. Their levelsincrease in response to stress, or they are more active as a consequence of stress,but they do not assist other proteins to fold, re-fold, or mobilize; they have other functions.


3.4Mechanisms and Manipulation

It is well known that stress proteins, including those that are molecular chaper-ones, increase in quantity inside the cell in response to stressors. This increase isone of the landmarks of the stress response [1]. There are several mechanismsby which the concentration of a given protein may be increased in a cell, forexample, increases in the rate of transcription of the gene that encodes the pro-tein, life-time of the mRNA, efficiency of mRNA translation, and life-time of theprotein itself. Hence, there are at least four points to consider when designingstrategies to increase the cell’s resistance to stressors. One may focus on tran-scription initiation and elongation and try to improve these processes to makethem faster and more efficient. Or one may try to do the same with the processof translation at the level of the ribosome. In addition, one may aim at decreas-ing the degradation rate of the mRNA, or the protein, and thus lengthen their lifespans. Obviously, a combination of these approaches seems the most promisingstrategy, however complicated it might be.

4Stress Response

4.1Definition

Now that we have introduced the stressors, stress genes and proteins, and themolecular chaperones, it is timely to define the stress response. This is a complexseries of events unchained by a stressor upon hitting a cell [1]. The stress genes,proteins, and molecular chaperones are the main players in these events. Theyincrease in concentration and/or activity to counteract the effects of the stres-sor.

The central, most damaging effect of stressors is on cellular proteins thatbecome denatured, i.e., they lose their native configuration [1, 12]. Hence, thestress response aims at avoiding protein denaturation, renaturing (refolding)partially damaged molecules, and degrading those damaged beyond repair.

4.2Characteristics

A stress response manifests itself in several ways [1, 12]:

(i) Most, if not all, house-keeping genes are down-regulated or shut off;(ii) The stress genes are activated;(iii) Many cellular proteins become denatured in various degrees;(iv) The cell may change its motility and move more or less than before

stress;(v) The cell membrane and wall may change to become more resistant to the

stressor, more or less permeable to certain compounds, ions, etc.;


(vi) The cell may aggregate with others via synthesis of an intercellular con-nective material and build multicellular structures of various degrees ofcomplexity; and

(vii) Other phenomena may occur, ranging from alterations in the intracellularosmolytes through changes in the electrolytes and other components.

All the above events, or some of them, may occur in a cell in response to a givenstressor, but down-regulation of housekeeping genes and activation of stressgenes are by definition essential landmarks. These two main components areevidenced by a change in the levels of the proteins that these genes encode: theproducts of the housekeeping genes diminish, even become undetectable,whereas the products of the stress genes increase. Thus, in monitoring microbesin a bioreactor to check their functional status and potential for recovery ifstressed, one may measure the levels of representatives of these two groups ofproteins, or their respective mRNAs. This in fact ought to be a straightforward,relatively simple approach to the monitoring and controlling of bioreactors toprevent, or at least anticipate, failure with disruption of a waste-treatment oper-ation.

4.3Stressors

Anything endowed with the capacity to elicit a stress response is a stressor [1].The most common or best studied are listed in Table 1. It is clear from the listthat stressors are ubiquitous. They can be found in air, soil, water, all kinds offoods and nutrients, and in a great variety of natural or manufactured materials.Many of them reach the microbes in methanogenic bioreactors with the influentor are produced in the bioreactor itself [17, 53–56]. It is therefore not surprisingthat bioreactor malfunction is a rather frequent occurrence, even when themechanical and most chemical conditions seem to be adequate. It follows thatmonitoring bioreactors ought to include testing the degree of stress of themicrobes by measuring manifestations of the stress response.

5Stress Genes and Molecular Chaperones in Archaea

5.1Overview

This topic has been reviewed recently [12]. It will, therefore, not be necessary to treat it again here in any detail, except in what pertains directly to biometha-nation.

The study of stress genes in Archaea began very recently by comparison withtheir counterparts in organisms of the other two domains,Bacteria and Eucarya.The earliest studies on the stress response in Archaea may be traced back to thelate 1980s, but it was only in 1991 that an archaeal stress gene, hsp70(dnaK), wascloned and sequenced for the first time [57]. The same year, a chaperonin gene


was also cloned and sequenced [7]. Since then, a few others have been se-quenced, some as part of genome sequencing projects [12]. In parallel, func-tional studies in vitro have been carried out to elucidate how and when thesegenes are expressed and how they respond to stress. Nevertheless, very little isknown about these genes in comparison with those from bacteria and eukary-otes. A lot of work remains to be done before one can think of using archaealgenes, or their products, for improving methanogens pertinent to bioreactortechnology.

5.2Evolution

The evolution of archaeal stress genes, hsp70(dnaK) in particular, has attractedthe attention of a number of investigators. Considerable information exists onthe evolution of hsp70(dnaK), which has revealed very interesting features [8, 32,39, 40]. For example, the gene is absent from several archaeal species, includingsome methanogens, Table 3, seemingly in contradiction of the generally ack-nowledged fact that this gene is very important, if not essential, for life, espe-cially for surviving stress. In fact, the discovery that the gene has a discontinu-ous distribution among Archaea may be considered one of the major contribu-tions of researchers working with Archaea to the fields of evolution, stress


Table 3. Occurrence, or lack thereof, of the hsp70(dnaK) gene among Archaea and represen-tatives of thermophilic and hyperthermophilic bacteria a

Organism OTG b hsp70 Genome size Demonstrated (°C) (dnaK) (Mb) by:

ARCHAEAMethanosarcina

mazei S-6 37 Yes 2.8 S, N, W, seq.cmazei JC3 37 Yes n.d.d Nmazei LYC 37 Yes n.d. Nsp. JVC 37 Yes n.d. Nacetivorans C2A 37 Yes 2.7 Nbarkeri 37 Yes 2.7 Sthermophila TM-1 50 Yes 2.7 S, N, seq.

Methanospirillum hungateii 37 No n.d. S

Methanobacterium 65 Yes 1.7 seq.thermoautotrophicum DH

Methanococcusvoltae 37 No n.d. S, Wvannielii 37 No n.d. S, Pjannaschii 85 No 1.7 S, seq.

Methanothermus fervidus 85 No n.d. S, P

Methanopyrus kandleri 100 No n.d. S, P

Haloarcula marismortui 45 Yes n.d. seq.

response, and molecular chaperones in general. Several of the critical questionsposed by this finding have been discussed recently [8, 12]. The finding has had asignificant impact on our views regarding the evolutionary conservation of thegene, its role in cell physiology and survival, and its substitute in those speciesthat do not have it (but still need to maintain an optimal set of functional pro-teins via folding and refolding under “normal” conditions and in the face ofstress).

Also very interesting is the evolution of the gene encoding the Hsp60 archaealchaperonin, of which one, two, or three representatives may be found depending


Table 3 (continued)

Organism OTG b hsp70 Genome size Demonstrated (°C) (dnaK) (Mb) by:

Halobacterium cutirubrum 45 Yes n.d. seq.halobium 45 Yes n.d. S, P

Thermoplasma acidophilum 55 Yes 1.7 seq., P

Sulfolobus solfataricus 70 No 3.1 S, P

Sulfolobus sp. 70 No n.d. S

Archaeoglobus fulgidus 83 No 2.2 seq., P

Desulfurococcus mobilis 85 No n.d. S, P

Thermococcus tenax 88 No n.d. S, P

Pyrococcus furiosus 100 No 2.0 seq.horikoshii 100 No 1.7 seq.woesei 100 No n.d. S, Pabyssi 100 No 1.8 seq.

Pyrobaculum aerophilum 100 No 2.2 seq.

Aeropyrum pernix K1 100 No 1.7 seq.

BACTERIA

Thermus thermophilus 70 Yes n.d. seq.

Thermomicrobium roseum 70 Yes n.d. seq.

Thermotoga maritima 80 Yes n.d. seq.

Aquifexaeolicus 83 Yes n.d. seq.pyrophilus 83 Yes n.d. seq.

a Reproduced from reference 12 with permission from the copyright owner.b OTG, optimal temperature for growth.c S, N, and W, Southern, Northern, and Western blotting, respectively; P, PCR; seq., sequenc-

ing of gene or genome.d n.d., not determined.e R. Weiss, personal communication.f S.A. Fitz-Gibbon, personal communication.


on the species [31, 38]. The proteins encoded in these chaperonin genes havereceived different names: chaperonin subunit 1 and 2, a and b (and g in casethere were three in the same organism), or TF55 and TF56. Comparative analy-ses of amino acid sequences using computer programs that reveal phylogeneticrelationships suggested that multiple subunit species arose several times, inde-pendently, during archaeal evolution.

5.3Structure

The analyses of sequences mentioned above have revealed other features ofinterest. For example, the variation in the number of chaperonin subunitsdepending on the species is intriguing. It may have important consequences onthe structural details of their assembly when they build up the chaperonin com-plex (for details on this complex, including three-dimensional reconstructions,see [12]). These structural details may in turn play a decisive role on the way thecomplex functions as a chaperoning machine. A complex formed by a singletype of subunit may work differently from complexes formed by two, or three,types of subunits. This remains to be determined and points to another area inwhich research with archaeal chaperonins will most likely have reverberationson biomethanation technology. No doubt, the engineering of chaperonin com-plexes with the proper structure will help cells to make functionally competentproteins (such as the enzymes needed for methanogenesis) even under stress.

5.4Expression and Regulation

Very little is known about the expression of the genes that produce the compo-nents of the chaperone machine, the chaperonin complex, and other stress pro-teins and molecular chaperones in Archaea. The topic has been recentlyreviewed [12], and the reader is encouraged to consult this article in order tounderstand its relevance to the art and science of fortifying cells, to make themresistant to stress. Some detailed information is available for the methanogensM. mazeii and M. thermophila, which will be discussed in subsequent Sectionsof this Chapter.

6The Hsp70(DnaK) Chaperone Machine in Methanogens

6.1Components

As mentioned above, some methanogens lack the hsp70(dnaK) gene, and thegenes for the other components of the machine, Hsp40(DnaJ), and GrpE, Table 3.The latter two genes seem to always accompany hsp70(dnaK) [8]. If hsp70(dnaK)is missing, hsp40(dnaJ) and grpE are also absent in the genome. This parallelismhas been demonstrated whenever enough sequence data have been available.

Thus, one may conclude that the product of hsp70(dnaK) cannot operate with-out the products of the other two. Needless to say, this conclusion from sequenc-ing data alone must translate into functional inferences, and impact on the wayone plans to engineer microbes to make them stronger in the face of stress.Manipulation of only hsp70(dnaK) would not suffice. The other two genesshould also be included, so the three of them would be expressed in a coordi-nated fashion for their products to act in unison, namely, to form a balancedchaperone machine, functionally efficient.

The hsp70(dnaK), hsp40(dnaJ), and GrpE genes from methanogens that havebeen cloned and sequenced are listed in Tables 4, 5, and 6, respectively. The firsthsp70(dnaK) loci to be fully sequenced in methanogens (and in the wholedomain Archaea) are depicted in Fig. 1.

6.2Expression

Expression of the Hsp70(DnaK) molecular chaperone machine genes inresponse to the stressor heat has been studied in M. mazeii S-6, Fig. 2, and to alesser extent in M. thermophila TM-1 [14, 59–64].

Data in Fig. 2 demonstrate that the genes in M. mazeii S-6 respond to heatshock and, thus, that they are in this regard similar to counterparts in the organ-


Fig. 1. hsp70(dnaK)-locus genes of the methanogens for which sequences are available thatinclude genes up- and downstream of hsp70(dnaK). The genes are represented by rectangularboxes from the 5¢ to the 3¢ end (left to right) with their names above their respective boxes inthe locus on top (dnaK and dnaJ are used instead of hsp70(dnaK) and hsp40(dnaJ) for sim-plicity). The figures within the boxes indicate the number of amino acids encoded. The linesjoining the boxes represent the intergenic regions with their lengths, in base pairs, shownunderneath. The sequences of M. thermophila TM-1 grpE and trkA are still incomplete (whatis available would encode 53 and 401 amino acids, respectively). Accession numbers and other details are provided in Tables 3–6. Reproduced from references [12] and [14] with per-mission from the copyright owners


Tabl

e4.

hsp7

0(dn

aK)g

enes

in m

etha

noge

nsa

Org

anis

mA

cces

sion

Ba

se p

airs

/am

ino

Prom

oter

/ter

min

ator

/O

ther

str

uctu

res

Expr

essi

onnu

mbe

rac

ids

enco

ded

RBS

b

Met

hano

bact

eriu

m

AE0

0089

417

91/5

96n.

r./n.

r./ga

ggtg

(-8)

cD

owns

trea

m r

epea

ts;

n.r.

ther

moa

utot

roph

icum

DHst

emlo

ops

Met

hano

sarc

ina

maz

ei S

-6X

6026

518

60/6

19aa

actt

taat

taa

(-79

)/

Up-

and

dow

nstr

eam

H

eat-

shoc

k in

vert

ed r

epea

ts/

repe

ats;

stem

loop

sin

duci

ble

agga

tata

a (-

5)

Met

hano

sarc

ina

Y17

862

1833

/610

aact

ttta

tcta

(-60

)/Pa

lindr

ome;

Hea

t-sh

ock

ther

mop

hila

TM

-1tc

tttt

tt (+

38)/

di

stin

ctiv

e fe

atur

es

indu

cibl

eag

tgag

gata

aa (

-7)

upst

ream

aD

ata

extr

acte

d fr

om r

efer

ence

[12]

wit

h pe

rmis

sion

from

the

copy

righ

t ow

ner.

bR

BS,r

ibos

ome-

bind

ing

site

.c

n.r.,

not r

epor

ted;

(–) a

nd (+

) ref

er to

pos

itio

n of

cent

er o

fseq

uenc

e up

stre

am fr

om th

e tr

ansl

atio

n st

art c

odon

or

dow

nstr

eam

from

the

tran

slat

ion

stop

cod

on,r

espe

ctiv

ely.


Tabl

e5.

hsp4

0(dn

aJ)g

enes

in m

etha

noge

nsa

Org

anis

mA

cces

sion

Ba

se p

airs

/am

ino

Prom

oter

/ter

min

ator

/O

ther

str

uctu

res

Expr

essi

onnu

mbe

rac

ids

enco

ded

RBS

b

Met

hano

bact

eriu

m

AE0

0089

411

31/3

76ac

attt

tttt

att (

-63)

Up-

and

dow

nstr

eam

n.

r.th

erm

oaut

otro

phic

ucm

DH/n

.r./ a

ggtg

(-9)

cre

peat

s;st

emlo

ops

Met

hano

sarc

ina

X

6026

511

70/3

89aa

acct

gttc

aca

Up-

and

dow

nstr

eam

H

eat-

shoc

k m

azei

S-6

(-10

0)/t

-ric

h re

gion

/re

peat

s;st

emlo

ops

indu

cibl

eaa

cagg

gaat

ctg

(-8)

Met

hano

sarc

ina

AJ0

1015

211

67/3

88aa

acct

gcac

t (-5

5)/t

cttt

tt

Inve

rted

rep

eat

n.r.

ther

mop

hila

TM

-1(+

30),

t-ri

ch r

egio

n/up

stre

am;t

-ric

h at

gaca

ggga

a (-

11)

regi

on d

owns

trea

m

aD

ata

extr

acte

d fr

om r

efer

ence

12

wit

h pe

rmis

sion

from

the

copy

righ

t ow

ner.

bR

BS,r

ibos

ome

bind

ing

site

.c

n.r.,

not r

epor

ted;

(–) a

nd (+

) ref

er to

pos

itio

n of

cent

er o

fseq

uenc

e up

stre

am fr

om th

e tr

ansl

atio

n st

art c

odon

or

dow

nstr

eam

from

the

tran

slat

ion

stop

cod

on,r

espe

ctiv

ely.

Tabl

e6.

grpE

gene

s in

met

hano

gens

a

Org

anis

mA

cces

sion

Ba

se p

airs

/am

ino

Prom

oter

/ter

min

ator

/O

ther

str

uctu

res

Expr

essi

onnu

mbe

rac

ids

enco

ded

RBS

b

Met

hano

bact

eriu

m th

erm

o-A

E000

894

525/

174

aaaf

fttt

atat

a (-

87)/

Ups

trea

m r

epea

ts,

n.r.

auto

trop

hicu

m D

Hn.

r/ag

gtg

(-7)

cst

emlo

ops

Met

hano

sarc

ina

maz

ei S

-6X

7435

363

0/20

9aa

ctat

ttat

aga

(-69

)/

Up-

and

dow

nstr

eam

H

eat-

shoc

k in

vert

ed r

epea

t re

peat

s;st

emlo

ops

indu

cibl

e(+

76)/

atgg

g (-

11)

aD

ata

extr

acte

d fr

om r

efer

ence

12

wit

h pe

rmis

sion

from

the

copy

righ

t ow

ner.

bR

BS,r

ibos

ome

bind

ing

site

.c

n.r.,

not r

epor

ted;

(–)

and

(+)

refe

r to

pos

itio

n of

cent

er o

fseq

uenc

e up

stre

am fr

om th

e tr

ansl

atio

n st

art c

odon

or

dow

nstr

eam

from

the

tran

slat

ion

stop

cod

on,r

espe

ctiv

ely.


isms of the other two phylogenetic domains. However, the M. mazeii genesresemble eucaryal homologues in as much as the message is monocistronic, incontrast to the bacterial counterparts. The latter are transcribed polycistroni-cally into an mRNA molecule at least as long as the sum of the lengths of thethree genes [12, 65]. In contrast, the mRNAs from the M. mazeii genes are dis-tinct from one another, and their individual lengths are about the same as thoseof the respective genes.

Fig. 2. Northern blots with M. mazeii S-6 total RNA (10 µg/lane) showing an increase in thetranscripts of hsp70(dnaK) (A), hsp40 (dnaJ) (B), and grpE (C), and a decrease in the tran-script of orf16 (D), in response to heat shock. (E) dot blot showing a decrease in the transcriptof orf11-trkA in response to heat shock. Hybridizations were done in all cases with radiola-beled probes specific for the respective genes. In A, B, and D, I the gel is stained with ethidiumbromide showing the ribosomal RNAs, 23S and 16S, while II is the corresponding Northernblot. Lanes A: total RNA from M. mazeii S-6 cells maintained at the optimal growth tempera-ture of 37°C, i.e., non-heat-shocked cells. Lanes B–C, or B–D: total RNA from cells heat-shocked at 45°C for increasing time periods, from 15 to 60 min. The sizes of the transcripts inA–D are indicated in kilobases (kb). Transcripts were detected for all the genes in non-heat-shocked cells. Heat shock caused an increase in the transcripts of hsp70(dnaK), hsp40(dnaJ),and grpE. The reverse occurred for orf16, and orf11-trkA. The latter two genes overlap, and arecotranscribed, whereas the other genes are transcribed monocystronically. Reproduced fromreferences [59, 60, 62, 64] with permission from the copyright owners

The other two genes in the M. mazeii S-6 hsp70(dnaK0 locus, orf16 and orf11-trkA, Fig. 1, respond differently [62, 64, 66]. Their transcripts decrease instead ofincreasing, after heat shock, Fig. 2.

6.3Stressor-Response Relationships

In order to be able to develop methanogens with improved stress resistance –that can be used for bioconversion of wastes in harsh, changing environments –one must first understand the range of conditions within which the stress genesare able to respond, and also, one must learn how the cell responds at theextremes of that range.

Data in Fig. 3 demonstrate that the three genes of the chaperone machine triad in M. mazeii S-6 (OTG 37°C) respond very well at temperatures between 45 and 60°C, even though cell viability is considerably diminished already at55°C [63].

The conclusion from the data is that while an increase in temperature abovecertain limits kills many M. mazeii S-6 cells, the heat-shock response of thechaperone machine triad is still in operation, at least with respect to increasingthe genes’ transcripts. This observation suggests that in a seriously damagedmicrobial population, with many of its members dying, there are still many cellscapable of mounting a sizable stress response. The data also suggest that byimproving the cells’ ability to mount a stress response, using genetic engineer-ing procedures for example, one will increase the size of the surviving, func-tional subpopulation, and thus insure continuity in the bioconversion processand avoid bioreactor failure.

It must be mentioned, however, that it remains to be established whetherdamaged cells still capable of mounting a stress response as measured by anincrease in the stress genes’ transcripts, are also capable of proceeding with the entire pathway of protein biogenesis and produce functional stress pro-teins. This is a very promising area of research pertinent to biomethanationtechnology.

6.4Other Stressors Pertinent to Methanogenic Bioreactors

The list of cell stressors that can affect methanogens is long, as one may inferfrom the sample listed in Table 1. Examples of stressors pertinent to industrialand other effluents and environments are heavy metals and sound.

Cadmium (Cd++) and sound do elicit a stress response in M. mazeii S-6, asshown by the data in Figs. 4 and 5, respectively [58]. It is likely that othermethanogens, and other microbes pertinent to methanogenesis in bioreactors,are also susceptible to be stressed by Cd++ and sound. It follows thatmethanogens and the other components of methanogenic consortia ought to beable to withstand stress caused by heavy metals and sound to a degree above andbeyond that of cells in other, less stressful environments.



Fig. 3. Response of the M mazeii S-6 genes grpE, hsp70(dnaK), and hsp40(dnaJ) to heat shockat various temperatures demonstrated by slot-blotting. The levels of mRNA for grpE,hsp70(dnaK), and hsp40(dnaJ) (top three panels) are represented by vertical bars expressed inthe OD. X mm units given by the densitometer. The respective slot blots (10 µg/slot of totalRNA) are shown at the foot of the bars, while the heat-shock temperatures are indicated in thehorizontal axis at the bottom of the figure (°C). Hybridization was done with the respectivelabeled probes. The culture density is shown in the bottom panel. The OD660 was determinedat time 0 (open bars) and at 30 min (hatched bars), in cultures maintained at 37°C or heat-shocked during this 30-min period at the temperatures indicated at the foot of the bars. Repro-duced from reference [63] with permission from the copyright owner

Another compound very pertinent to methanogenic bioreactors is ammonia.Bioconversion of wastes from humans and other animals with feces and urine isinhibited when the ammonia present in these wastes reaches certain levels [53,54]. Data in Fig. 6 demonstrate that ammonia induces a stress response inM. mazeii S-6 as evidenced by the increase in the levels of the mRNAs from themolecular chaperone machine genes [67].

Interestingly,ammonia also induces a response by the adjacent gene trkA.Theresponse of this gene seems to be tightly regulated in view of its strict dosedependency. Overall, the results show that concentrations of ammonia over 20-fold higher than that which is adequate for physiological growth of M. mazeiiS-6 cause stress in this methanogen (see also Sects. 6.6 and 8.3).

6.5Factors that Modify the Stress Response

It is well established that a number of cellular activities and physiological func-tions are associated with, or are dependent on, the cell cycle and the growth


Fig. 4. Response of the M. mazeii S-6 genes grpE, hsp40(dnaJ), and hsp70(dnaK) to the stres-sors cadmium (Cd++) and heat. The bars represent levels of mRNA determined by slot-blot-ting with probes for the grpE, hsp40(dnaJ), and hsp70(dnaK) genes. The total RNAs were fromcells grown at 37°C (i.e., the optimal temperature for growth for M. mazeii S-6) in mediumwithout Cd++ (a), and in medium with 5 or 27 mM CdCl2 (b and c, respectively); and from cellsgrown in medium without Cd++ but heat-shocked at 45°C for 30 min (d). Note that the mRNAsfrom the three genes increased after heat shock by comparison with the levels before heatshock (constitutive or basal levels; compare a vs. d). Likewise, the presence of Cd++ in themedium also induced an increase in the three mRNAs. This effect was more marked with 27than with 5 mM CdCl2; compare a vs. b and c; and b vs. c. Reproduced from reference [58] withpermission from the copyright owner

a b c d

grpEdnaJ dnaK

% a

rea

ofpe

ak

phase. Thus, growth phase affects many cellular properties. Among these is thestress response. In M. mazeii S-6, both the basal (constitutive) and heat-shockinduced levels of the mRNAs from the molecular chaperone machine genes areaffected by growth phase, as illustrated by the data in Fig. 7, which pertain to thehsp70(dnaK) gene [63]. The highest levels of this gene’s mRNA were induced byheat stress in cells in early stationary phase, as compared to the levels induced incells in the exponential and late-stationary phases. The highest basal levels ofmRNA were observed in cells in late stationary phase. This, together with thediminished response to heat stress, suggests that cells in late stationary phase arestressed. The degree of stress gene activity in late-stationary phase in theabsence of an added stressor is higher than in the other two phases. Concomi-tantly, in late stationary phase, the capacity to respond to an added stressor isimpaired.


Fig. 5. Response of the M. mazeii S-6 hsp70(dnaK) gene to the stressors heat and sound. Thelevels of hsp70(dnaK) mRNA were determined by slot blotting (A) and were measured by den-sitometry (B). RNAs were from cells cultured at the optimal temperature for growth, i.e., 37°C(slot-blots 1 and 3, counting from the top down); or from cells heat-shocked at 45°C for 30 min(slot blot 2); or from cells maintained at 37°C and exposed to sound (90 decibels) for 15, 30,60, and 120 min (slot blots 4, 5, 6, and 7, respectively). Notice the increase of hsp70(dnaK)mRNA after heat shock (compare peaks 1 vs. 2), and after sound stress (compare peaks 3 vs.4–7). Reproduced from reference [58] with permission from the copyright owner


Fig. 6. Effect of ammonia on grpE, hsp70(dnaK), hsp40(dnaJ) and orf11-trkA mRNA levels inM. mazeii S-6. Total RNA was extracted from single cells cultivated in medium with the stan-dard concentration of NH4Cl (i.e., 1 g/L; lanes A), or from cells incubated for 30, 60, or 180 minin medium containing either 10 (lanes B, C, and D, respectively) or 25 (lanes E, F, and G, respec-tively) g/L of NH4Cl, and electrophoresed (10 µg/lane) in a denaturing gel. The upper portionof each panel represents the gel stained with ethidium bromide showing the 23S and 16SrRNAs, whereas the bottom portion displays the respective Northern blot hybridized withprobes for grpE (top left panel), hsp70(dnaK) (top right), hsp40(dnaJ) (bottom left) and orf11-trkA (bottom right). The sizes of the rRNAs, and those of the hybridization bands in kilobases(kb) are indicated to the right. Reproduced from [63] with permission from the copyrightowner

The observations describe above are critical for developing strategies to mon-itor the performance of microbes in bioreactors, and to control and improve themicrobial populations so that they are always capable of responding to stressors,and to proceed with methanogenesis, despite changes in growth rates and envi-ronmental factors.

6.6Other Methanogens

The studies referred to above were done using M. mazeii S-6, which is a keyorganism in mesophilic methanogenic ecosystems, including bioreactors [19,68–73]. There are, in addition, data pertaining to M. thermophila TM-1 (OTG50°C), which is important for methanogenesis in many thermophilic ecologicniches and bioreactors [15, 17, 18, 71, 74, 75].

The effect of heat-shock of various durations on the chaperone machinegenes of M. thermophila TM-1 was determined. Illustrative data for


Fig. 7. Response of hsp70(dnaK) to heat shock in M. mazeii S-6 cells at different growth phas-es. Top panel. Slot blots of total RNA (10 µg/slot) from cells in exponential (1), and early (2)and late (3) stationary phases, before (37°C) and after (45°C) a heat shock at 45°C for 30 min,hybridized with a probe for hsp70(dnaK). Bottom panel. Densitometric readings (OD. X mm)of the slot blots shown in the top panel, before and after heat shock (open and hatched bars,respectively). Reproduced from reference [63] with permission from the copyright owner

dnaK

37°C

45°C

1 2 3

hsp70(dnaK) and hsp40(dnaJ) are shown in Fig. 8, top panel [14]. Cells wereheat-shocked at 60 or 65°C for 15, 30, or 60 min. Gene transcripts were minimalbefore stress, but after a 60-min heat shock at 60°C they increased. A similartrend was already evident after a 15-min heat shock at 65°C, and the levels oftranscripts were still higher after longer heat shocks by comparison with cellheat-shocked at 60°C. The mRNA from hsp40(dnaJ) increased little under all theconditions tested, suggesting that this gene is regulated differently as comparedwith hsp70(dnaK).

It is clear from the results in Fig. 8 that 65°C is a critical temperature forM. thermophila TM-1. The cells suffer a stress and mount a stress response afteronly 15 min of exposure at 65°C. It may be concluded that assessing the levels ofhsp70(dnaK) transcripts will provide a sensitive indicator of TM-1-cell stress ina bioreactor with temperatures above 60°C. By the same token, the data alsoindicate that the bioreactor temperature must be maintained below 60°C toinsure the well being of M. thermophila TM-1 (and the same is probably true forother methanogens that have OTGs similar to TM-1).

The response of M. thermophila TM-1 to ammonia stress was also studied[14]. An increase in the mRNA from hsp70(dnaK) was induced by all threeammonia doses tested: 5, 10, and 25 g/L in the series of experiments shown inFig. 8, bottom panel. Interestingly, the effect of 5 g/L correlated with exposurelength: no mRNA increase after a 10-min exposure, clear increase after 30 and60 min, and clear but slightly lower increase after 120 min.

A dose-dependent response to ammonia stress was also observed for the trkAgene, Fig. 8, bottom panel, lower section.A clear increase in this gene’s transcriptwas produced by the 5 g/L-10 min dose. Longer exposure times caused tran-script increases that were less and less marked as the times increased. A differ-ent pattern was observed for the 10 g/L dose. The trkA mRNA augmented pro-gressively as the incubation time with ammonia increased from 10 to 60 min, buta 120-min exposure caused about the same effect as 60 min. The highest dosetested, 25 g/L, caused the same increase in the trkA mRNA levels at all the incu-bation times tested with ammonia, except for the 30-min exposure, which pro-duced a slight but evident higher increase.


Fig. 8. Top panel. Response of M. thermophila TM-1 to heat shock. Northern blotting of totalRNA (10 µg/lane) from TM-1 hybridized with a probe for hsp70(dnaK), top and middle sec-tions, or for hsp40(dnaJ), bottom section. Lanes 1 and 5 contained RNA from cells maintainedat 50°C. The other lanes contained RNA from cells heat-shocked for 15 min (lanes 2 and 6),30 min (lanes 3 and 7), or 60 min (lanes 4 and 8), at 60°C (top section) or at 65°C (middle andbottom sections). The left half of each section displays the gel stained with ethidium bromideto show the 16 and 23 rRNAs, while the right half shows the respective Northern blot with thesize of the hybridization bands in kilobases (kb). Bottom panel. Response of M. thermophilaTM-1 to increasing concentrations of ammonia assessed by slot blotting to determine levelsof hsp70(dnaK) and trkA transcripts after incubation with the stressor for various time peri-ods. Total RNA was extracted from exponentially-growing cells and 10 µg of RNA/slot wasused. The blots were hybridized with biotin-labeled probes for hsp70(dnaK) and trkA (top andbottom sections, respectively). In each section the slot blots are at the bottom, and the verticalbars above the slot blots represent the respective densitometric readings. Reproduced fromreference [14] with permission from the copyright owner

These data, together with those obtained with M. mazeii S-6 and described inthe previous Sect. 6.4, indicate that the stress response to the stressor ammoniais dose dependent, and that it must be tightly regulated. The levels of ammoniain the medium in which the cells grow, and the length of time during which thecells are exposed to this compound, play critical roles in determining the degreeof the stress response to ammonia (and probably have distinctive effects on theresponse to other stressors that may act simultaneously in real-life situations).

The implications for bioreactor management and technology of the data dis-cussed above are manifold. For example, the level of stress caused by ammoniacan be monitored by assessing the levels of mRNAs from one or more stressgenes, including trkA. Also, it is clear from the data that ammonia can be a pow-erful cell stressor, and that its effects are more pronounced as the dose and expo-sure times increase. Lastly, a time of exposure to elevated levels of ammonia asshort as 2 h will cause severe cell stress.

The effects of relatively long heat shocks on both Methanosarcina species areillustrated by data in Fig. 9 [63]. The results show that, contrary to bacteria [12],M. mazeii S-6 and M. thermophila TM-1 (and most likely many othermethanogens) can withstand heat shocks longer than 15–30 min without adecrease in the stress response as measured by the levels of the hsp70(dnaK)-locus gene products. Even after a 3-h heat shock, the M. mazeii S-6 hsp70(dnaK)mRNA is quite increased.


Fig. 9. Response of the hsp70(dnaK) gene from M. mazeii S-6 and M. thermophila TM-1 toheat shocks of various durations. Left panel. Northern blots of total RNA (10 µg/lane) extract-ed from S-6 cells before heat shock (lane 0, in both sections), or after a heat shock at 45°C forthe length of time indicated in the horizontal axis, in minutes (min) or hours (h). Hybridiza-tion was done with a probe for hsp70(dnaK). The size of the hybridization bands in kilobases(kb) is indicated to the right. Right panel. Total RNA (20 µm/lane) from M. thermophila TM-1was run in a denaturing gel and stained with ethidium bromide to show the 23S and 16SrRNAs (top section). The respective Northern blot obtained with a probe for hsp70(dnaK) isshown below (bottom section). Lanes from left to right are: RNA from cells maintained at 50°C(lane 0, non-heat-shocked) and RNAs from cells heat-shocked at 60°C for 1, 2, or 3 h (lanes 1,2, and 3, respectively). The size of the hybridization bands in kilobases (kb) is shown to theright. Reproduced from reference [63] with permission from the copyright owner

The capacity of methanogens to respond efficiently to temperature elevationsabove their optima, even if they are exposed to these temperatures for hours, is auseful and encouraging feature. It shows that the need to correct the temperatureof a bioreactor before it causes irreversible cell damage is not of extreme urgency.Likewise, it also shows that any measure that would improve the cell’s stressresponse, even slightly, will add time to the period during which corrections tothe bioreactor conditions can be made. This time gain should greatly facilitatebioreactor operation, and reduce the frequency of total, irreversible failures.

6.7Co-Chaperones

In view of the discontinuous distribution of the chaperone machine amongArchaea and the occurrence of chaperonins in these organisms (see follow-ing Section), it is pertinent to ask whether Archaea have the co-chaperones –also named chaperone co-factors – known to coexist, and interact with themachine and the chaperonins in bacteria and eukaryotes. Examples of co-chaperones are the bacterial trigger factor (TF), and the eucaryal Hop, Hip,BAG-1, and NAC.

A recent survey of five fully sequenced archaeal genomes, including one withthe machine and four lacking it, but all containing chaperonins, showed absenceof conservation of the genes encoding the co-chaperones [75a]. There were nogenes readily identifiable by common genome-searching methods as being thehomologues of the five co-chaperones listed above. However, two families ofmolecules were identified that might be related to Hop and to one of the sub-units of NAC. These results, which open the road to a more detailed analysis ofthe chaperoning mechanisms in Archaea as compared to those of bacteria andeukaryotes, are available in the Internet at:http://www.bioscience.org/2001/v6/d/macario/fulltext.htm

7 The Hsp60 (Chaperonin) System in Methanogens

7.1Examples

Extensive descriptions and discussions of the archaeal chaperonin system areavailable in the literature [7, 10, 12, 76, 77]. Therefore, only aspects directly orindirectly pertinent to methanogens will be touched in this chapter.

Examples of chaperonins in methanogens are listed in Table 7. The functionsof these genes’ products have not been elucidated in any detail either in vivo orin vitro. However, extrapolating from what is known from studies in otherarchaeal, non-methanogenic species, and in eukaryotes and bacteria (referencesin [12]), it may be said that the chaperonins of methanogens are likely to play acritical role in de novo protein biogenesis. They may also play a role during thestress response, and in the cell recovery after stress, but these roles althoughprobable have not yet been demonstrated.


7.2Structure and Potential for Bioreactor Technology

The chaperonins form multimeric complexes of comparatively very large size(thousands of kDa) with a spheroidal or cylindrical shape, and with a centralcavity that serves as a protected chamber inside which polypeptide folding isthought to occur [12].

The implications for biotechnology and bioreactors are significant. The phys-iological performance of methanogens is tied to a protein balance within thenormal range. Protein balance here is understood as the entire set of proteins ina cell, which is composed of many functionally distinct subsets. Each subsetmust be maintained within the normal range of number of molecules (concen-tration), and each molecule must be kept with its structural and conformationalintegrity (native configuration). Most of these properties are maintained by theconcerted action of the molecular chaperone machine, the chaperonins, andother molecules including proteases. It follows that research ought to be direct-ed towards the elucidation of how the chaperonin system in methanogens withdifferent OTGs, and pertinent to anaerobic digestion of wastes (e.g., M. mazeiiS-6 and M. thermophila TM-1, and others) assembles itself and functions undernormal circumstances, and under stress. Regulation of the chaperonin genesought to be clarified in order to manipulate them in a way that will maintain pro-tein biogenesis during stress to ensure continuous biomethanation.


Table 7. A sample of hsp60 (chaperonin) genes in methanogens a

Organism Chaperonin Base Promoter Terminator RBS b

(accession number) pairs

Methanobacterium Chaperonin 1617 n.r. c n.r n.rthermo- (a-subunit)autotrophicum (mt0794)

DH Chaperonin (mt0218) 1659 n.r. n.r. n.r.

Methanococcus Chaperonin 1626 n.r. n.r. n.r..jannaschii (U67542)

Methanococcus MTTS (AB015435) 1632 tttatata t-rich region n.r.thermo- (–75) d (+20)lithotrophicus

Methanopyrus Thermosome 1635 tttaaata c-rich region aggtgat kandleri (Z50745) (–60), (+18)

atgc (–42)

a Data extracted from reference 12, with permission from the copyright owner.b RBS, ribosome binding site.c n.r., not reported.d (–) and (+) refer to position of center of sequence upstream from the translation start codon

or downstream from the translation stop codon, respectively.

8Other Stress or Stress-Related Molecules, Genes and Proteins,and Anti-Stress Mechanisms in Methanogens

8.1Examples

A number of molecules different from those discussed in the preceding Sectionsappear, or increase in concentration, in response to stressors [12]. Illustrativeexamples of molecules that have been studied are listed in Table 8. These mole-cules are integral parts of the stress response as the molecular chaperones andchaperonins. Similar genes, and other pertinent examples have been identifiedin the genomes of methanogenic species that have been fully sequenced, asshown in Table 9. Also, some methanogenic species have a cell envelope withextraordinary stability [42, 78].

8.2Osmolytes

Some of the molecules listed in Table 8 (e.g., inositol compounds) are not pro-teins and participate in maintaining the internal osmotic pressure. Compoundsof this sort are called osmolytes and they come into action when the osmolarityof the medium surrounding the cell increases or decreases (osmotic shock) [13,79–81]. These compounds are of paramount importance for cells that inhabithypersaline environments, or that suddenly encounter such environments, forexample when the influent of a bioreactor contains unusual concentration ofsalts. In addition, the enzymes that participate in the synthesis import of osmo-lytes are also stress proteins in as much as they must be active during stress, andmust produce the osmolytes to protect the cells from osmotic stress.

8.3TrkA

TrkA has been identified in M. mazeii S-6, M. thermophila TM-1, and in othermethanogens whose genomes have been fully sequenced [14, 62, 66]. It is a pro-tein member of the Trk K+ transport system in Escherichia coli and other bacte-ria [82, 83]. By analogy, the archaeal homologue is assumed to be also involvedin the transport of this cation.

TrkA is involved in maintaining the K+ balance of the cell [82, 83]. The inter-nal K+ balance of methanogens in anaerobic bioreactors may be affected by fac-tors in the immediate environment. Ammonia is one of these factors, known toinhibit methanogenesis [55, 56]. Ammonia may reach inhibitory levels when abioreactor is fed with protein-rich wastes or swine manure, for example [53].The unionized ammonia causes a pH increase. An intracellular pH increase willresult in a K+ efflux coupled to a H+ influx in order to counteract the pH increase[55, 84]. TrkA is involved in this counter transport. As described in a previousSection, the trkA of both M. mazeii S-6 and M. thermophila TM-1 respond to


ammonia stress as shown by an increase of its mRNA. Since it may be assumedthat this gene’s product, TrkA, is involved in maintaining a physiological level ofintracellular K+ also in methanogens, and since this cation is essential for themolecular chaperone machine to function [refs. in 62, 66, 67], one may hypo-thesize that trkA is a stress gene. It probably plays a major role in cell physiologyand survival in bioreactors. More research on this interesting topic with many


Table 8. Other examples of stress, or stress-related, genes and proteins that have been foundand studied in methanogens a

Gene/protein Protein mass Organism Presumed Inducer(kDa) function

Crx protein trio 40.8, 42.3, Methanobacterium Copper or Copperand 42.9 bryantii general resistance

Betaine transporter n.r. b Methanosarcina Maintains internal Osmoticthermophila TM-1 ionic balance stress

Inositol n.r. Methanococcus Maintains internal Osmotic compounds igneus ionic balance stress

TrkA 44.1 Methanosarcina Maintains internal Ammoniamazei S-6 K+ balance

Prefoldin or 14–23 c Methanococcus Protein folding n.rjannaschii;

GimC Methanobacterium n.r. n.r.thermo-autotrophicum

Small heat-shock n.r. Methanococcus RNA stabilization, n.r.protein (sHsp) jannaschii thermotolerance

ClpB n.r. Methanosarcina Affects growth and n.r.acetivorans survival at hightem-

peratures, involvedin proteolysis

PPIase (peptidyl 19.4–31 d Methanococcus Accelerates rate n.r.prolyl cis-trans 16 or 42 d thermo- limiting step in isomerase) lithotrophicus, protein folding

Proteasome 24 and 22 e Methanosarcina Protein n.r.hermophila TM-1, degradation

25.8 and Thermoplasma 22.3 e acidophilum

a Data extracted from reference 12 with permission from the copyright owner. See alsoTable 9.

b n.r., not reported.c Six subunits within the indicated size range in eukaryotes but only two subunits in Archaea.d Depends on the method used.e a- and b-subunit, respectively.

potential applications for the monitoring and control of bioreactors should bedone. This research will lay the foundations for using the trkA gene to fortifycells and make them more resistant to the ammonia and other stressors, whichmay also provoke imbalances of intracellular electrolytes.

8.4Prefoldin or GimC

Another multimeric complex named prefoldin or GimC seems to be involved inprotein folding in eukaryotes and in Archaea, including methanogens [85–87].Six subunits have been identified in eukaryotes, but only two have been found inArchaea.

The role of this complex in the stress response is unclear. The genes codingfor the subunits do not seem to be activated by stressors. Nonetheless, we willdiscuss prefoldin in this Chapter because of its probable participation in proteinfolding in vivo. Very little is known in this regard but current research will soonadd to our knowledge of this “chaperone machine” and may unveil functions


Table 9. Stress related gene/protein-homologues identified in sequenced genomes frommethanogens a

Organism Gene/protein ID b

Methanococcus Heat-shock protein X MJ1682jannaschii Heat-shock protein 31 MJ0285

DNA repair protein 45 MJ0869DNA repair protein RAD51 MJ0254DNA repair protein RAD2 MJ1444PPIase MJ0278PPIase MJ0825Proteasome a-subunit MJ0591Proteasome b-subunit MJ1237Survival protein MJ0559

Methanobacterium Heat-shock protein X MTH569thermoautotrophicum Heat-shock related protein X MTH1817DH Heat-shock protein class I MTH859

DNA repair protein rad2 MTH1633DNA repair protein rad51 MTH1693DNA repair protein radA MTH541DNA repair protein rad32 MTH1383PPIase MTH1125PPIase B MTH1338Proteasome, a-subunit MTH686Proteasome, b-subunit MTH1202Survival protein (SurE) MTH1435

a Excluding the Hsp70(DnaK) chaperone machine and the Hsp60 (chaperonin) family.Reproduced from reference 12 with permission from the copyright owner.

b ID, identification number in genome project.

that are essential for survival during stress, and/or for recovery after stress. Boththese functions are important for methanogens in bioreactors.

The representative from Methanobacterium thermoautotrophicum namedMtGimC has been studied in some detail [86]. It is a complex of 87 kDa made oftwo different subunits, a and b. Preliminary studies have indicated that the com-plex is a hexamer consisting of two a and four b subunits. The a subunit is theequivalent of the eukaryotic subunits Gim2 and Gim5, while b is the homologueof the eukaryotic Gim 1, 3, 4, and 6 subunits.

A preliminary in vitro search for possible chaperone functions of MtGimCshowed that it:

(i) Formed a complex with unfolded actin, and bound this substrate with rel-atively low affinity;

(ii) Suppressed aggregation of unfolded hen lysozyme (14 kDa);(iii) Prevented aggregation of chemically unfolded bovine mitochondrial rho-

danese (30 kDa) and glucose dehydrogenase (39 kDa);(iv) Formed complexes with non-native dihydrofolate reductase (DHFR;

23 kDa) and firefly luciferase (62 kDa);(v) Stabilized non-native actin for at least 15 min, which allowed transfer of

actin to TRiC (the eukaryotic chaperonin complex) for folding in the pres-ence of ATP; and

(vi) Prevented aggregation of unfolded rhodanese (as mentioned above) andallowed its folding by the bacterial chaperonin GroEL.

While the above series of observations demonstrate a certain degree of partici-pation of MtGimC in preventing the aggregation of partially denatured poly-peptides, and in assisting folding by way of interaction with TRiC in the case of actin or GroEL for rhodanese, how much the results reflect in vivo, physio-logically meaningful situations remains to be seen. Further analysis in vitro, andnew studies in vivo should be done to elucidate the functions of GimC inmethanogens, its mechanism of action, preferred substrates, and activity (orlack thereof) during stress. One may anticipate that important information isgoing to emerge from these analyses, which will be very useful in understandingthe intracellular situation of stressed methanogens and in thinking of ways forcoping with it so the cell will be able not only to survive, but also to continue thebioconversion pathway unabated.

8.5Small Heat-Shock Proteins (sHsp)

The sHsp are currently the focus of active investigation in organisms of the three phylogenetic domains, including the methanogens [12]. An sHsp fromMethanococcus jannnaschii has been purified and crystallized [33]. Like otherstress proteins, it forms a large multimeric complex. It protects other proteinsfrom heat denaturation and prevents aggregation of partially denaturedpolypeptides in vitro [34]. More research is necessary to determine the role ofthis, and other, sHsp in vivo. Such research should provide the basis for design-ing strategies to use sHsp and/or their genes for improving the mechanisms


against protein denaturation in methanogens, and thus their resistance to stres-sors. In this regard, it is noteworthy that sHsp form large complexes, as the chap-eronins for example do. These complexes are seemingly essential for the chap-erones in general to exercise their function of assisting other proteins to fold andrefold. One of the aims of research in this area, with direct implications for bio-methanation technology, should be the elucidation of how the multimeric struc-tures form, what stressors do tend to damage these structures, and what keepsthem from being disrupted by stressors. Obviously, information on these areaswill help in identifying the most damaging stressors, and in developing meansto avoid their accumulation in a bioreactor, and tools, genetic and otherwise, tostrengthen the stability of the multimeric complexes.

8.6PPIase

PPIase (for peptidyl prolyl cis-trans isomerase) is an enzyme that is found inmany organisms of the three phylogenetic domains [12]. It mediates peptidyl-prolyl isomerization, an important step in protein folding. There are variousforms of the enzyme, similar to each other, that have been grouped into threefamilies: cyclophilins (Cyp), FK506-binding proteins (FKBPs), and parvulins[29, 30, 88–90]. A PPIase from Methanococcus thermolithotrophicum thatbelongs to the FKBP family has been characterized in some detail [88]. Thegenes encoding other PPIases have been found in the genomes of othermethanogens, as seen in Table 9.

The observations described above demonstrate that methanogens possess acomplex battery of tools, including PPIases, to generate and maintain a balancedset of proteins within the ranges of concentrations and configurations requiredfor growth and survival. Study of PPIases will help in understanding proteinfolding in methanogens pertinent to bioreactors, and will pave the way to devis-ing means for protecting the folding machinery from damage due to stressors.

8.7Proteases

Proteases constitute a large group of enzymes, some of which should be consid-ered under the umbrella of stress.We will not discuss them here in any detail butrefer to reviews available in the literature [3–5, 91, 92]. Suffice it to say that pro-teases are involved in the degradation of abnormal proteins lest they interfered,or might interfere, with the trafficking of normal proteins and other functionsinside the cell. Abnormal protein in this context means molecules that are par-tially or completely unfolded due to stress or to some structural alteration(mutation, or post-synthetic modification that went wrong). These abnormalmolecules tend to misfold, aggregate, and build up precipitates. If these are toolarge, they will be an obstacle to the physiological movement of molecules insidethe cells,and cause a disturbance in many functions.Hence,abnormal moleculesmust be refolded into a correct configuration or, if this is impossible, they mustbe eliminated. Molecular chaperones participate in folding, refolding, dissolvingaggregates, and degradation. For the latter purpose, some molecular chaperones


present the abnormal polypeptide to the protease for digestion. When all thepreventive and corrective measures aiming at keeping the proteins in the correctconcentrations and configurations fail, or are overwhelmed owing to an excessof substrate for the chaperoning systems, proteases are called into action. Theseenzymes degrade the abnormal polypeptides and, in doing so, they not only ridthe cell of aggregates, but they also generate building blocks (i.e., amino acids)for the synthesis of new protein molecules.

Proteases are surely also involved in the construction of multicellular struc-tures (see Section below), a process that requires the action of many moleculesin addition to proteases. The formation of multicellular structures requires alsothe migration of these molecules towards the cell’s outside, as we shall discuss ina subsequent Section of the Chapter.

Interestingly, proteases tend to form multimeric complexes. One example isthe proteasome, which is a large multimolecular machine similar to the chaper-onin complex [91–95].

8.8Putative Stress Genes and Proteins Found in Fully Sequenced Genomes

The availability of full genome sequences has opened the doors to computer-assisted searches for stress genes, or candidate (putative) stress genes, whichencode proteins likely to play a role in the stress response but which have not yetbeen isolated and tested in the laboratory. A sample of these genes/proteinsfound in two genomes from methanogens (Methanococcus jannaschii andMethanobacterium thermoautotrophicum) that have been sequenced is dis-played in Table 9. Excluded from the list are the genes for the members of theHsp70(DnaK) chaperone machine and those for the Hsp60 (chaperonin) sys-tem, both groups already discussed in previous Sections (see Tables 4–7). It isimportant to re-emphasize that the chaperone machine genes are not present inthe genome of M. jannaschii, as discussed previously (see Table 3), whereas thechaperonin genes do occur in this methanogen and in M. thermoautotrophicum.

The functions of the genes/proteins listed in Table 9 remain to be determined.This is a challenging task for the near future made attractive because of theavailability of the clones that contain the genes, and the promise of informationuseful to devise strategies and tools for improving methanogens so that they willdevelop increased resistance to stressors.

9Other Manifestations of the Stress Response

9.1Introduction

In addition to the components of the stress response described in the precedingSections of this Chapter, which have been identified in methanogens and otherArchaea, there are other pertinent molecules, anatomical structures, and eventsthat must be discussed [12]. These are either induced or are associated in some


meaningful way with the stress response, the molecular chaperoning process,the development of stress resistance also called thermotolerance or stress toler-ance [26], or the recovery of cells after stress. What follows is a brief account ofseveral of these stress-related molecules, structures, and phenomena that areimportant for the survival and functioning of methanogens, and that havepotential for the devising of means to improve bioreactor performance despitechanging environmental conditions.

9.2Thermoprotectants

Cells produce compounds that somehow improve their thermotolerance. Someof these are sugars and simple molecules such as di-myo-inositol phosphate(DIP) and cyclic diphosphoglycerate (cDPG). They have been demonstrated, forinstance, in the hyperthermophilic methanogens Methanopyrus kandleri (OTG100°C) and Methanothermus fervidus (OTG 85°C). A more detailed discussionon the functions and possible mechanism of action may be found in recent arti-cles with pertinent bibliography [13, 81, 96]. See also Table 8.

9.3Multicellular Structures

A few methanogenic species have the ability to build multicellular structures,either by themselves (single-species structure) or in association with one ormore different species (multispecies structure) [17, 25, 42, 97, 98]. These struc-tures may be formed in response to stressors and confer more resistance to themby comparison with the isolated cells growing as independent, free units.

Morphologically, the multicellular structures appear as flat sheets of one orvery few cell-diameters in thickness, or as globular masses vaguely spheroidal inshape with diameters equivalent to many (e.g., 10–20) cell diameters. The cellsare kept together by an intercellular connective material, whose components arenot yet fully elucidated, and that ought to be considered elements of the stressresponse as a working hypothesis for future research (see below).

Examples of single-species multicelluar structures are produced by M. mazeiiS-6, which can be flat (named lamina) or globular (named packet), as illustratedin Fig. 10 [97, 98]. The packet morphotype is considerably more resistant tomechanical, chemical, and physical stressors, and to antibiotics, than the single-cell morphotype (AJLM and ECdeM, unpublished data). For instance, inductionof a heat-shock response measurable by an increase in the mRNAs from the mol-ecular chaperone machine genes (as shown in Figs. 2 and 3, for example)requires higher temperatures and longer exposure times in packets than in sin-gle cells. Illustrative data for the grpE gene are presented in Fig. 11 [60]. The sin-gle-cell morphotype showed a more pronounced response than that of the pack-ets to a heat-shock at 45°C for 30 or 60 min. In fact, the packets showed aresponse only after a heat shock of 60 min.

Other multicellular structures directly pertinent to anaerobic methanogenicbioreactors are the globular multispecies consortium termed granule [17], and


the biofilm [72, 99, 100]. They are usually composed of a variety of methanogensand bacteria interlaced in a food web. Histological thin sections of a granulefrom a thermophilic bioreactor are shown in Fig. 12, where the spheroidal shape may be inferred from the visible segment of the outer profile [101].Methanosarcinal packets can be seen in panel A, while panel B shows laminarstructures formed by M. thermophila TM-1 as demonstrated by the antigenicfingerprinting method. The methanogens form colonies of various shapes andsizes that are lodged in the supporting scaffolding provided by the intercellularconnective material as represented in the model shown in Fig. 13 [101].

Little is known about the mechanism of granule formation (granulogenesis)at the molecular and genetic levels, or about the biochemistry and synthesis ofthe components of the intercellular connective material. Also, the functions ofthis material, beyond the obvious mechanical support for cells, are largelyunknown. These functions are surely more complex than just providing a scaf-fold for the growth of cellular colonies. They must also include insulation, trans-port of nutrients and catabolites in opposite directions, concentration ofmicronutrients, passive barrier or active defense against agents of various kinds(chemical, physical, and biological such as antibiotics), and others that futureresearch ought to discover. Granules have an inner communication networkmade of a small tubes [101], as shown in Fig. 14. These tubes can conceivably bethe route for nutrients to reach the cells inside the granule, and the way of escapefor catabolites and other cellular products away from the cells.

There is some information about the composition of the intercellular con-nective material in Methanosarcina packets [102, 103], but beyond that there isnot much that would allow the developing of means to manipulate this material


Fig. 10. Multicellular structures formed by M. mazeii S-6. Packets (A) and lamina (B) are dis-played along with the single-cell morphotype (C), for comparison (see references [97, 98]).The diameter of the single cells is 1–3 µm, and the magnification factor is the same for thethree panels. The photographs were taken with phase contrast optics of wet samples from livecultures between glass slide and cover slip. Reproduced from reference [12] with permissionfrom the copyright owner

at the molecular and genetic levels for biotechnologic purposes. This is a veryimportant area for investigation since efficient methanogenesis in bioreactorsdepends on the presence of a stable population of microbes retained in positionwithin the granule, and inside the bioreactor, in the appropriate spatial relation-ship with one another. This three-dimensional distribution of different speciesis key to the metabolic interactions between them, as required by the food webleading to waste bioconversion with generation of methane.


Fig. 11. Heat resistance of a multicellular structure formed by a methanogen as comparedwith its own single-cell phenotype. Primer-extension mapping of the transcription initiationsite for M. mazeii S-6 grpE.A radiolabeled oligonucleotide primer complementary to bases 57through 77 within the grpE coding region was used with 10 µg of total RNA from single cells(lanes 1 to 3) or packets (lanes 4 to 6) per test. Single cells and packets were grown at 37°C(lanes 1 and 4) or heat shocked at 45°C for 30 (lanes 2 and 5) or 60 (lanes 3 and 6) min. Theprimer-extended products were electrophoresed in a 6% acrylamide sequencing gel in paral-lel with the products of a sequencing reaction that was done with the same primer and thedideoxychain-termination method (lanes G, A, T, and C). These lanes show the complemen-tary (anti-sense) strand sequence. The coding (sense) strand sequence and the initiation site(asterisk) are shown on the left. Reproduced from reference [60] with permission from thecopyright owner

Most likely, granules are among other things a mechanism to protect the cellsfrom stressors, as may be inferred from the data mentioned above, obtained with methanosarcinal packets. Also, because of their large size and weight ascompared with individual cells, the granules will not be washed away by the cir-culating bioreactor contents, and thus they will maintain a steady functionalprofile.


Fig. 12. Spheroidal multicellular structure (granular consortium, or granule) formed bymethanogens associated with bacteria in a thermophilic (50°C), anaerobic, methanogenicbioreactor, as seen in a thin histological section. (A) Cross section of the granule showing thecortex and medulla (see reference [101]) and a large island of methanosarcina packets(arrows). Hematoxylin-eosin (magnification ¥800). (B) Another section of the same granulein which the presence of Methanosarcina thermophila TM-1 (optimal temperature for growth,50°C) is demonstrated with a antibody probe for TM-1 by immunofluorescence. Themethanosarcina cells are arranged mostly in laminae (see Fig. 10; magnification ¥4000).Reproduced from reference [12] with permission from the copyright owner

The structure of the granule is complex (see Figs. 12–14) [75, 101, 104],including well defined zones, such as the cortex and the medulla, and subzonesthat probably represent functionally specialized areas. In addition, there are the small tubes (Fig. 14), which provide still another proof that a granule is acomplex anatomic structure with a complicated physiology, seemingly wellequipped to withstand stressors. It is then important to realize that understand-ing how a granule forms and maintains its integrity as a functional unit in anenvironment as full of stressors as the bioreactors influents is essential for thedeveloping of means to monitor and control biomethanation, and to correct itwhen the bioreactors malfunction. The same type of considerations apply to thebiofilm [70, 72, 73, 99, 100, 105], an example of which is presented in Fig. 15.

Fortification of cells to withstand stressors should, therefore, also includeimprovements in their granule- or biofilm-formation ability. In this regard, allthe molecules that form the intercellular connective material and the enzymesthat synthesize as well as those that translocate them to the cell’s outside shouldbe considered components of the stress response.As such, they should be targetsfor investigations aiming at developing means to improve granulogenesis, andbiofilm formation, and, thereby methanogenesis.


Fig. 13. Computer-assisted three-dimensional representation of a granular consortium likethe one shown in the preceding figure, depicting the aggregates or bundles formed bymethanogens. Methanobacterium thermoautotrophicum, surface (SC) and inner (IC) colonies;Methanosarcina thermophila packets (P) and laminae (L); Methanosaeta (Methanothrix) rodsin bundles of more or less intertwined filaments (Mx); Methanobrevibacter arboriphilus (Ma),clouds that appear in cross-section as lawns of variable density; and Methanobrevibactersmithii (Ms), thin clouds that look like sparse lawns in cross-section. The filaments formed bythe Methanosaeta rods are shown for the sake of clarity in only two areas but they are moregeneralized. Reproduced from reference [101] with permission from the copyright owner


Fig. 14. Superficial, histological thin section of a granule like the one shown in Fig. 12, pass-ing through the cortex. Visible are circular openings that are cross-sections of the tubes thatcrisscross the granule (possibly communicating different zones of it between themselves andwith the immediate surroundings of the granule [101]). Hematoxylin-eosin (magnification¥800). Reproduced from reference [12] with permission from the copyright owner


Fig. 15. Example of a microbial consortium in the form of biofilm made of methanogens andassociated, syntrophic bacteria visualized by scanning electron microscopy (SEM). Thebiofilm was attached to the substratum (curler-type polypropylene) in a fixed-bed anaerobicmethanogenic bioreactor processing synthetic waste water containing acetate,propionate,andbutyrate at 35°C. The samples were collected from the top (A and B), middle (C and D) andbottom (E and F) of the bioreactor 57 days after its inoculation with sludge from anotherdigestor treating municipal sewage. Discernible are cells that were identified as related to M.mazeii S-6 (single cells, 1), Methanosaeta (Methanothrix) soehngenii (2), Methanospirillumhungatei (3), and Desulfovibrio sp. (4). M. mazeii occurred as single cells (best visible in B) and as laminae (see Fig. 10). The exopolymer of the intercellular connective material in thelaminae appeared as filaments, as illustrated in D. Scale bars (in µm) are shown at the right-bottom corner of each panel. Reproduced from reference [73] with permission from the copy-right owner

10Perspectives and Applications

10.1Introduction

The study of the stress response, stress genes and proteins, other components ofthe stress response, and molecules and phenomena pertinent to resistanceagainst stressors and recovery after stress is essential to deal with stressors,counteract them, and avoid or abate their effects. Stressors of many kinds reachbioreactors with the influent, or are produced inside the bioreactors. It is there-fore of paramount importance to develop means to deal with the problemscaused by stressors. As repeatedly stated in this Chapter, before preventive andcorrective means can be developed, information from basic and appliedresearch is needed. This research should focus on several topics, some of whichwill be dealt with in the following portions of this Chapter.

10.2Diversity of Methanogens

In the preceding Sections we have referred to methanogens in bioreactors andfocused chiefly on M. mazeii S-6 and M. thermophila TM-1. These twoMethanosarcina species are key for methanogenic bioconversion in meso- andthermophilic environments, respectively. However, other methanogens alsooccur in bioreactors, Fig. 13. In fact, there is considerable diversity ofmethanogenic species, strains, and immunotypes in bioreactors as demonstrat-ed as early as 1988, Fig. 16 [69]. An important conclusion drawn from these andsubsequent findings is that the study of the stress response, stress genes and pro-teins, and anti-stress mechanisms should be extended to other methanogens, inaddition to methanosarcinas (see also Sect. 10.6).

10.3Dynamics of Methanogenic Subpopulations in Bioreactors

Qualitative and quantitative analyses using immunologic and other comple-mentary methods have revealed that the population of methanogenic organismsin bioreactors (and several other ecosystems) is composed of subpopulations,each of these representing a different species [15, 18, 69–71, 74, 75, 100]. Sub-populations have also been identified within a single species. Time-course stud-ies have demonstrated that methanogenic subpopulations change in distribu-tion and in size (number of organisms in each subpopulation), during bioreac-tor operation [70, 75]. An illustrative study is displayed in Fig. 17. A few speciesof methanogens identified in a bioreactor fed with sulfite evaporator condensatewere followed over a period of 14 months [70]. Some stressful manipulationswere done to the bioreactor during this period. Methanogenic subpopulationswere assessed by qualitative and quantitative methods at different time points.The results showed that:


(i) The subpopulations differed in size at the beginning, thus adding an extradimension (quantitative) to the diversity already evident from the varietyof species present;

(ii) The subpopulations closely related to Methanobrevibacter smithii ALI andM. mazeii S-6 were the most abundant at the beginning, while the subpop-ulations closely related to Methanobacterium formicicum MF, Methanobre-vibacter arboriphilus AZ, and M. arboriphilus DC, were the smallest;


Fig. 16. Diversity of methanogens in bioreactors. Methanogens identified in a series of 14 dif-ferent bioreactors (DIGESTOR A-N) with antibody probes and the antigenic fingerprintingmethod using indirect immunofluorescence and the quantitative slide immunoenzymaticassay, SIA. The variety of methanogens occurring in these bioreactors as a group and in eachone of them is evident from the total number (14) of species identified and the range of speciesfound in the individual bioreactors (from one, bioreactor K, up to 8, bioreactor B). In most cas-es the methanogens found were not identical to the reference species, neither were they of thesame immunotype within each species identified.Abbreviations are: Mx.,Methanothrix; Msp.,Methanospirillum; Mbr., Methanobrevibacter; Mc., Methanococcus; Mb., Methanobacterium;Ms., Methanosarcina. Reproduced from reference [69] with permission from the copyrightowner

(iii) Some subpopulations increased after seven months (e.g., MF, DC, ALI, andMethanobrevibacter smithii PS), while others remained the same (AZ) ordecreased (Methanosarcina barkeri W, and S-6);

(iv) Seven months after stressful manipulations of the bioreactor performedwithin a short interval, which had caused a decrease in all methanogens(data not shown), practically all the subpopulations had recovered;

(v) The two Methanosarcina species recovered to some extent but only in theform of packets, namely the phenotype most resistant to stressors;

(vi) The presence of Methanosarcina packets at the end of the observation peri-od was even more striking if one considers that at the beginning there werevirtually only single cells, and suggests that stressors along the way select-ed against single cells and perhaps induced them to form multicellularstructures.


Fig. 17. Diversity and dynamics over time (months) of methanogenic subpopulations inbioreactors subjected to manipulations known to cause cell stress (e.g., change in pH, nutri-ents’ availability, and configuration of functional space). Samples A and D (abscissa) were tak-en from different levels inside the chamber at the beginning, when the bioreactor reached sta-ble conditions, i.e., steady flow of substrate and yield of biogas. Seven months later, sample Fwas obtained at a time in which modifications in pH and substrate chemical oxygen demand(COD) were introduced. Immediately thereafter, configuration changes were also made, andseven months later, samples H and J were collected, from different levels. Methanogens wereidentified and quantified in each sample as follows: Methanobacterium formicicum MF (a);Methanobacterium arboriphilus AZ (b) and DC (c); Methanobrevibacter smithii ALI (d) andPS (e); a rod related to Methanosarcina thermophila TM-1 (f); Methanosarcina barkeri W (g);and Methanosarcina mazeii S-6 (h). Each bar represents the number (arithmetic mean ±range; n = 2) of organisms per species identified – that in most cases were not identical to thereference organism. In g and h, open and closed bars represent the packets and single-cellmorphotype, respectively, of M. mazeii S-6 (see Fig. 10). The wavy lines at the top of the barsfor samples H and J in panel f indicate abundance beyond the quantifiable by the methodused. Reproduced from reference [70] with permission from the copyright owner

Efforts to improve bioreactor operation and yield ought to take into account thediversity of methanogens involved and their time-course dynamics in terms ofquantity, described above. Those species more productive of methane will haveto be targeted first for improving their anti-stress machinery using genetic engi-neering procedures, or stress-gene inducers that are not harmful, such as drugsthat mimic physiologic stress-gene inducers. These drugs could be added to theinfluent at doses predetermined to induce stress genes without secondary,unwanted effects on the cells.

10.4Diversity of Stressors

A major goal of future research aiming at improving bioreactor technologyshould be the identification of stressors that might affect methanogens and oth-er pertinent microbes. A list of representative stressors for all kinds of cells isdisplayed in Table 1, but only a minority of them have actually been tested withmethanogens, as shown in Table 10. These stressors are relevant to bioreactortechnology because they are found in relatively high levels in the effluents frommany factories, homes, farms, and other man-made sources that require anaer-obic bioconversion in bioreactors.

10.5Diversity of Response

Another important task for the near future will be that of characterizing in detailthe response to the different stressors that are relevant to methanogenic biotech-nology. It is well established that a series of components of the stress responseare the same for any stressor. These are the basic or common components, whichare those discussed in this Chapter for the most part. It is very likely, however,that the stress response has, in addition to the basic components, other elements


Table 10. Examples of stressors, other than heat, tested with methanogens a

Stressor Organism

Hyperosmolarity Methanococcus igneus, Methanococcus thermolitholitrophicus,Methanosarcina thermophila TM-1, Methanosarcina mazei S-6

Pressure Methanococcus thermolithotrophicus, Methanococcus jannaschii

Ethanol Methanococcus voltae

Copper Methanobacterium bryantii

Cadmium Methanosarcina mazei S-6

H2O2 Methanococcus voltae

Ammonia Methanosarcina mazei S-6, Methanosarcina thermophila TM-1

Sound Methanosarcina mazei S-6

a Data extracted from reference [12] with permission from the copyright owner.

that are specific for each stressor or family of similar stressors. It will beextremely interesting and useful to identify at least some of the stress-responsecomponents that are specific for each of the stressors most relevant to bio-methanation technology. They could be genes/proteins, signal transducers,membrane sensors or receptors, gene-activating and gene-repressing factors,molecules for signaling the formation of multicellular structures, etc. Also, themechanism of action of some of these molecules may differ depending on thestressor. A case in point would be a gene activator that would induce a stressgene by one mechanism (e.g., using a heat-shock cis-acting element) if the stres-sor is heat, and by another if the stressor is a heavy metal (e.g., by interactingwith a metal element in the DNA instead of binding to a heat-shock element).One can also hypothesize that the response to stress by ammonia implicatesDNA elements and transcription and regulatory factors that are different fromthose used in the response to a heat shock, at least for the activation of the trkAgene.

10.6Diversity of Methanogenes: A Source of Useful Microbes?

A rational approach to the improvement of bioreactor technology includes themanipulation of relevant genes to construct better methanogens, more resistantto stressors and also more efficient bioconverters. This must be based on knowl-edge provided by basic and applied research on the molecular biology and bio-chemistry of the various components of the stress response, as outlinedthroughout this Chapter up to this point.A second avenue towards assembling avery efficient and resistant microbial population inside a bioreactor is the searchfor “good” microbes in natural ecosystems. If one or more are found with thecharacteristics required, they could be used for bioconversion in bioreactors, oras a source of useful genes.

The diversity of methanogens we have mentioned several times before prob-ably reflects their universality [106]. They can be found in a wide variety of eco-logic niches. An idea of the ubiquity of methanogens is provided by data inTable 11 [AJLM and ECdeM, unpublished data]. In it, we have listed the sourcesof methanogenic isolates recorded as tested and identified immunologically inour laboratories between 1981 and 1986, and the number of isolates from eachsource. The variety of sources is evident and encompasses ecologic niches withvery different characteristics. We have also identified methanogens in otherecosystems, different from those mentioned in Table 11, such as the Antarcticcontinent [107], deep subterranean aquifers [108], and temperate marine waters[109], just to mention a few. In each ecosystem explored, whenever it was possi-ble to study at least a few isolates, species diversity was evident, and withinspecies a diversity of immunotypes was usually discovered. For example,46 methanogens isolated from human feces were all identified as Methanobre-vibacter smithii, but they were distributed into at least seven groups with dis-tinctive antigenic mosaics demonstrable with a panel of six monoclonal anti-bodies [110]. These findings show that the diversity of methanogens, even with-in a single species, is quite remarkable, and that with the appropriate tools


(e.g., a panel of calibrated antibody probes) this diversity can be demonstratedfairly easily.

Another source of methanogens for possible use in biotechnology is the sea.In one study of the water column of the Chesapeake Bay, we demonstrated sev-eral species, Fig. 18. Some were related more or less closely to the referenceorganisms available in culture collections but others were not [109]. The deeperthe water layer, the more abundant were the methanogens less similar to theknown species.

The diversity of methanogens demonstrated by the antigenic fingerprintingmethod is phenotypic. It might not reflect to the last detail structural diversityat the genome level. Nevertheless, phenotypic diversity does suggest genomicdifferences, particularly functional ones. It shows that even if the gene contentsof different genomes are very similar, their functional patterns are not. Genesactive in one phenotype may be inactive in another.Thus, there is diversity in thepattern of regulatory mechanisms. What are the regulatory genes involved indetermining the phenotypes? Which are the most useful phenotypes? An impor-tant endeavor in the near future should be the identification of useful pheno-types, and of the genes involved in producing them. It will then be possible tosearch for the “good”microbes, stress resistant and efficient for biomethanation,or to make them by means of genetic engineering procedures.

It is evident from the data described above, more recently confirmed by oth-ers with different methods [111], that microbial diversity is probably enormous,


Table 11. Methanogenic isolates from various ecosystems identified by antigenic fingerprint-ing during the period 1981–1986

Isolate Ecosystem Total Antigenically identifiableSource studiedCountry Yes No

USA human feces 67 67 0USA dental plaque 14 12 2USA animal feces 16 11 5USA rumen herbivores 11 9 2USA cockroach digestive tract 2 2 0The Netherlands marine ciliate 1 0 1Germany; USA marine sediments 14 12 2Germany hot spring 1 1 0Germany swamp 1 1 0USA peat lands 5 1 4Germany; France soil 4 3 1USA fresh-water sediments 5 5 0United Kingdom landfills 12 11 1Japan; USA; France waste-water sludge 5 5 0Germany; USA; bioreactors (digestors) 39 33 6

FranceCanada; Germany; undetermined 21 17 4

Japan; New Zealand;USA

TOTAL 218 190 (87%) 28 (13%)

and that what we have so far uncovered is only a very minimal portion of it. Wehave seen only the tip of the iceberg, as it were. Hence, there is hope that a searchfor methanogens in nature will yield abundant dividends in terms of speciesuseful for methanogenic biotechnology, endowed with the necessary resistanceto the stressors that usually threaten bioreactor stability and efficiency. Thissearch for naturally “good” organisms can be complemented with genetic engi-neering to make them optimal not only to withstand stress but also to proceedthrough the methanogenic pathway with speed and efficiency.


Fig. 18. Diversity of methanogens in an aquatic ecosystem. Methanogens were isolated fromthe water column of Chesapeake Bay (USA) and characterized by antigenic fingerprinting andother methods. Samples for isolating the microbes were collected from three different layersof the water column (abscissa). Twelve, eight, and thirteen isolates from the upper, middle(pycnocline), and lower layer, respectively, were characterized and identified. Each isolate isrepresented by a square with the number inside indicating the species-strain most closelyrelated, as follows: 3, Methanosarcina barkeri MS; 6, Methanosarcina barkeri R1M3; 18,Methanosarcina mazeii S-6; 19, Methanosarcina barkeri W; 20, Methanosarcina thermophilaTM-1; and 0 (zero), unrelated to the known reference methanogens. The striped bars indicatethe percentage of methanogenic isolates that were unrelated to the reference organisms,namely they were novel methanogens, not yet available in the pure-culture collections. Therelative abundance of isolates related to the Methanosarcinae is noteworthy, as is the increasedabundance of novel methanogens with increasing water-column depth. Data extracted fromreference [109] with permission from the copyright owner

10.7Cooperation Between Molecules and Between Cells

Stress proteins like the molecular chaperones function as members of a team ormolecular machine with several interconnected and interacting parts. Membersof a machine, for example the Hsp70(DnaK)-Hsp40(DnaJ)-GrpE molecularchaperone machine, interact with each other, and also with other molecules andmachines [6, 9, 43, 49, 52]. Chaperonins and sHsp also assemble into large mul-timeric complexes [7, 11, 33, 76, 77]. It is obvious that natural selection hasfavored these complexes, and one must infer that they are functionally betterthan the sum of the separate activities of their single components. Alternatively,one might think that multimerism is a requirement for the functioning of cer-tain types of molecules as multicellular communities (tissues, organs, and theirprimitive prokaryotic counterparts) would be for cells.

A parallelism to molecular multimerism seems to occur with the tendency toform multicomponent (communities, tissues, organs) structures by cells. Theseforms of association for function seem to be far more effective under physio-logic circumstances and in the face of stress than solitary molecules or cells.Molecular machines, tissues and organs, and microbial consortia, all appear tobe landmarks of evolutionary success. The main conclusion one may draw fromthese observations is that strategies for optimizing bioreactor technology oughtto include the development of means that enhance the formation of multicom-ponent machines at the molecular and cellular levels.

10.8Proteases as Builders

Proteases are essentially destructive in as much as they degrade molecules intosmaller parts [4, 5, 91]. However, this process may be essential in some instancesfor building complex multicellular structures. Enzymatic digestion of moleculesby proteases frees the space occupied by the initial, larger whole,when the small-er parts have been used up or removed (e.g., washed away with fluids in circu-lation, or engulfed by cells). The freed space can then be occupied by anothercomponent of the complex, this time a more appropriate one for that particularlocation. Alternatively, the voided space may remain empty of solids and thusbecome a vesicle or tube for storage or circulation, respectively. It is likely thatproteases take an active role in the formation of the tubes that crisscross thegranular microbial consortia discussed earlier in this Chapter, and shown in Fig.14. There can be little doubt that this internal circulatory system is essential forthe survival of cells inside the structure, or at least for distribution of nutrientswithin it. It also is a convenient way for removal of catabolites and for delivery ofproducts from one cell (or colony) to another (see Fig. 13), or to the outside(e.g., methane).

The observations discussed above suggest that proteases should also be thetarget of basic and applied research that would provide the basis for engineer-ing more efficient microbial consortia. For example, a consortium should have acirculation network commensurate with its size, and the needs and metabolic


activities of all its cellular constituents, regardless of the location of these con-stituents in the whole structure.

10.9Intrinsic Stress Resistance

The molecules of organisms that have high or very high OTG, or that grow underhigh hydrostatic pressure, are able to function under these extreme conditions(as compared to those that are optimal for humans, for instance). The moleculesare endowed with intrinsic stress resistance; the mechanisms implicated in this resistance are only now beginning to be examined [13, 112, 113]. This is aninteresting point for investigation, potentially useful for methanogenic bio-technology.

11Conclusion and Perspectives

Stress proteins, molecular chaperones, formation of functional multimericstructures by molecules and cells, thermoprotectants, ion transporters, and oth-er anti-stress mechanisms ultimately depend on the presence of genes properlyregulated, capable of responding to the attack of stressors. Hence, elucidation of the gene regulatory mechanisms is an important step towards optimizing biomethanation. Tools to study gene regulation and to manipulate genes inmethanogens are being developed [114–117]. Likewise, means to manipulatemicrobial cells, including methanogens and syntrophs, using antibodies andrelated techniques are available [118]. The perspectives for rapid progress are,therefore, promising. If the study of stress genes, proteins, and other anti-stressmechanisms,and the identification of novel microbial species continue, it will bepossible in the near future to optimize the biological component of bioreactors.Furthermore, it will be possible to monitor bioreactor function to anticipate fail-ure, and to repair it in case of malfunction, by removing unwanted microbesand/or introducing those pre-selected or pre-engineered (genetically) to meetthe requirements for stress resistance and optimal biomethanation. It has beendemonstrated that it is possible to introduce a microbial species in a granularconsortium to add to it a lacking metabolic ability [119]. The consortium wasthus endowed with the capacity to bioconvert a substrate that could not bemetabolized prior to the microbial graft. This procedure is easy to perform andhas a promising future as a means to build consortia with stress-resistantmicrobes tailored to bioconvert specific substrates. Biofilms and granules con-structed on demand with stress-resistant microbes, which are also efficient forbiomethanation of specific substrates (e.g., a certain type of waste), should be areality in the first decade of the third millennium.

Acknowledgement. Work in the authors’ laboratories has been supported over the years bygrants from NYSERDA, DOE, and NSF.We thank our collaborators of yesterday and today, toonumerous to mention by name (many appear in the list of references), for their help. We alsothank the members of the Photo-Art unit of this Center for their excellent assistance withgraphs and photographs.


12References

1. Macario AJL, Conway de Macario E (2000) Heat-shock response, overview. In: Fink G(ed), The encyclopedia of stress. Academic Press, San Diego, California, USA,Vol 2, p 350

2. Macario AJL, Conway de Macario E (2000) Chaperone proteins. In: Fink G (ed), TheEncyclopedia of Stress. Academic Press, San Diego, California, USA, Vol 1, p 429

3. Gottesman S, Squires C, Pichersky E, Carrington M, Hobbs M, Mattick JS, Dalrymple B,Kuramitsu H, Shiroza T, Foster T, Clark WP, Ross B, Squires CL, Maurizi MR (1990) ProcNatl Acad Sci USA 87:3513

4. Herman C, D’Ari R (1998) Curr Op Microbiol 1:2045. Wickner S, Maurizi MR, Gottesman S (1999) Science 286:18886. Hartl FU, Martin J (1995) Curr Op Struct Biol 5:927. Klumpp M, Baumeister W (1998) FEBS Lett 430:738. Macario AJL, Conway de Macario E (1999) Genetics 152:12779. Netzer WJ, Hartl FU (1998) Trends Biochem Sci 23:68

10. Ranson NA, White HE, Saibil HR (1998) Biochem J 333:23311. Willison KR (1999) Composition and function of the eukaryotic cytosolic chaperonin-

containing TCP-1. In: Bukau B (ed), Molecular chaperones and folding catalysis. Har-wood Academic Publishers, Sydney, Australia, p 555

12. Macario AJL, Lange M, Ahring BK, Conway de Macario E (1999) Microbiol Mol Biol Rev63:923

13. Holden J, Adams MWW, Baross JA (2000). Heat-shock response in hyperthermophilicmicroorganisms. In: Bell CR, Brylinsky M, Johnson-Green P (eds), Microbial biosystems:New frontiers. Atlantic Canada Society for Microbial Ecology, Acadia University,Wolfville, Nova Scotia, Canada, p 663

14. Hofman-Bang JP, Lange M, Conway de Macario E, Macario AJL, Ahring BK (1999) Gene218:387

15. Ahring BK, Schmidt JE, Winther-Nielsen M, Macario AJL, Conway de Macario E (1993)Appl Environ Microbiol 59:2538

16. Clarens M, Cairó JJ, Paris, JM, Macario AJL, Conway de Macario E (1993) Curr Microbiol26:167

17. Schmidt JE, Ahring BK (1996) Biotechnol Bioeng 49:22918. Schmidt JE, Macario AJL, Ahring BK, Conway de Macario E (1992) Appl Environ Micro-

biol 58:86219. Zinder, S (1993) Physiological ecology of methanogens. In: Ferry JG (ed), Methanogene-

sis. Chapman and Hall, New York, New York, USA, p 12820. Madigan MT, Martinko JM, Parker J (2000). Brock Biology of microorganisms, 9th edn.

Prentice-Hall Inc, Upper Saddle River, New Jersey, USA21. Balch WE, Fox GE, Magrum LJ, Woese CR, Wolfe RS (1979) Microbiol Rev 43:26022. Woese CR (1987) Microbiol Rev 51:22123. Woese CR (1998) Curr Biol 8:R78124. Woese CR, Kandler O, Wheelis ML (1990) Proc Natl Acad Sci USA 87:457625. Boone DR, Whitman W B, Rouviere P (1993) Diversity and taxonomy of methano-

gens. In: Ferry JG (ed), Methanogenesis. Chapman and Hall, New York, New York, USA,p 35

26. Macario AJL, Conway de Macario E (2000) Heat resistance. In: Fink G (ed), The encyclo-pedia of stress. Academic Press, San Diego, California, USA, Vol 2, p 338

27. Bustard K, Gupta RS (1997) J Mol Evol 45:19328. Csermely P, Schnaider T, Sôti C, Prohászka Z, Nardai G (1998) Pharmacol Ther 79:12929. Fisher G, Tradler T, Zarnt Z (1998) FEBS Lett 426:1730. Galat A (1999) Arch Biochem Biophys 371:14931. Gupta RS (1995) Mol Microbiol 15:132. Gupta RS, Singh B (1994) Curr Biol 4:110433. Kim KK, Kim R, Kim S-H (1998) Nature 394:595


34. Kim R, Kim KK, Yokota H, Kim S-H (1998) Proc Natl Acad Sci USA 95:912935. Lee GJ, Vierling E (2000) Plant Physiol 122:18936. Oh HJ, Easton D, Murawski M, Kaneko Y, Subjeck JR (1999) J Biol Chem 274:1571237. Conway de Macario E, Macario AJL (1994) Trends Biotechnol 12:51238. Archibald JM, Logsdon Jr JM, Doolittle WF (1999) Curr Biol 9:105339. Gribaldo S, Lumia V, Creti R, Conway de Macario E, Sanangelantoni A, Cammarano P

(1999) J Bacteriol 181:43440. Gupta RS (1998) Microbiol Mol Biol Rev 62:143541. Robb FT, Place AR, Sowers KR, Schreier HJ, DasSarma S, Fleischmann EM (eds), (1995)

Archaea. A laboratory manual. Cold Spring Harbor Laboratory Press, Plainview, NewYork, USA

42. Sprott GD, Beveridge TJ (1993) Microscopy. In: Ferry JG (ed), Methanogenesis. Chapmanand Hall, New York, New York, USA, p 81

43. Pfanner N (1999) Curr Biol 9:R72044. Arsene F, Tomoyasu T, Mogk A, Schirra C, Schulze-Specking A, Bukau B (1999) J Bacteri-

ol 181:355245. Blaszczak A, Georgopoulos C, Liberek K (1999) Mol Microbiol 31:15746. Morimoto RI (1998) Genes Dev 12:378847. Glover JR, Lindquist S (1998) Cell 94:7348. Weber-Ban EU, Reid BG, Miranker AD, Horwich AL (1999) Nature 401:9049. Zolkiewski M (1999) J Biol Chem 274:2808350. Korber P, Zander T, Herschlag D, Bardwell JCA (1999) J Biol Chem 274:24951. Bogdanov M, Dowhan W (1999) J Biol Chem 274:3682752. Wang Y-Y, Chen X, Oh H-J, Repasky E, Kazim L, Subjeck J (2000) FEBS Lett 465:9853. Angelidaki I, Ahring BK (1994) Wat Res 28:72754. Angelidaki I, Petersen SP, Ahring BK (1990) Appl Microbiol Biotechnol 33:46955. Koster IW, Koomen E (1988) Appl Microbiol Biotechnol 28:50056. Koster IW, Lettinga G (1988) Biological Wastes 25:5157. Macario AJL, Dugan CB, Conway de Macario E (1991) Gene 108:13358. Macario AJL, Conway de Macario E (1997) Stress 1:12359. Clarens M, Macario AJL, Conway de Macario E (1995) J Mol Biol 250:19160. Conway de Macario E, Clarens M, Macario AJL (1995) J Bacteriol 177:54461. Conway de Macario E, Dugan CB, Macario AJL (1994) J Mol Biol 240:9562. Conway de Macario E, Macario AJL (1995) J Bacteriol 177:607763. Lange M, Macario AJL, Ahring BK, Conway de Macario E (1997) Curr Microbiol 35:

11664. Macario AJL, Simon VH, Conway de Macario E (1995) Biochim Biophys Acta 1264:17365. Hanawa T, Kai M, Kamiya S, Yamamoto T (2000) Cell Stress Chap 5:2166. Macario AJL, Dugan CB, Conway de Macario E, (1993) Biochim Biophys Acta 1216:

49567. Lange M, Macario AJL, Ahring BK, Conway de Macario E (1997) FEMS Microbiol Lett

152:37968. Kobayashi HA, Conway de Macario E, Williams RS, Macario AJL (1988) Appl Environ

Microbiol 54:69369. Macario AJL, Conway de Macario E (1988) Appl Environ Microbiol 54:7970. Macario AJL, Conway de Macario E, Ney U, Schoberth AM, Sahm H (1989) Appl Environ

Microbiol 55:199671. Macario AJL, Earle JFK, Chynoweth DP, Conway de Macario E (1989) Syst Appl Microbi-

ol 12:21672. Zellner G, Geveke M, Conway de Macario E, Diekmann H (1991) Appl Microbiol Biotech-

nol 36:40473. Zellner G, Macario AJL, Conway de Macario E (1996) Appl Microbiol Biotechnol 46:

44374. Koornneef E, Macario AJL, Grotenhuis JTC, Conway de Macario E (1990) FEMS Micro-

biol Ecol 73:225


75. Visser FA, van Lier JB, Macario AJL, Conway de Macario E (1991) Appl Environ Microbiol57:1728

75a. Macario AJL, Conway de Macario E (2001) Frontiers in Bioscience 6:262–283http://www.bioscience.org/2001/v6/d/macario/fulltext.htm

76. Ditzel L, Löwe J, Stock D, Stetter KO, Huber H, Huber R, Steinbacher S (1998) Cell 93:125

77. Nitsch M, Walz J, Typke D, Klumpp M, Essen LO, Baumeister W (1998) Nature Struc Biol5:855

78. Beveridge TJ, Stewart M, Doyle RJ, Sprott GC (1985) J Bacteriol 162:72879. Da Costa MS, Santos H, Galinski EA (1998) Adv Biochem Eng Biotechnol 61:11880. Hensel R, Koenig H (1988) FEMS Microbiol Lett 49:7581. Martin DD, Ciulla RA, Roberts MF (1999) Appl Environ Microbiol 65:181582. Nakamura T, Yamamuro N, Stumpe S, Unemoto T, Bakker EP (1998) Microbiology

144:228183. Schlösser A, Hamann A, Bossemeyer D, Schneider E, Bakker EP (1993) Mol Microbiol

9:53384. Sprott G, Patel G (1986) Syst Appl Microbiol 7:35885. Geissler S, Siegers K, Schiebel E (1998) EMBO J 17:95286. Leroux M, Faendrich M, Klunker D, Siegers K, Lupas AN, Brown JR, Schiebel E, Dobson

CM, Hartl FU (1999) EMBO J 18:673087. Vainberg IE, Lewis SA, Roemmelaere H, Ampe C,Vandekerckhove J, Klein HL, Cowan NJ

(1998) Cell 93:86388. Furutani M, Iida T, Yamano S, Kamino K, Maruyama T (1998) J Bacteriol 180:38889. Herdegen T, Fisher G, Gold BG (2000) Trends Pharmacol Sci (TiPS) 21:390. Iida T, Furutani M, Nishida F, Maruyama T (1998) Gene 222:24991. Baumeister W, Walz J, Zuehl F, Seemueller E (1998) Cell 92:36792. DeMartino GN, Slaughter CA (1999) J Biol Chem 274:2212393. Maupin-Furlow JA, Aldrich HC, Ferry JG (1998) J Bacteriol 180:148094. Maupin-Furlow JA, Ferry JG (1995) J Biol Chem 270:2861795. Zwickel P, Ng D, Woo KM, Klenk H-P, Goldberg AL (1999) J Biol Chem 274:2600896. Shen B, Hohmann S, Jensen RG, Bohnert HJ (1999) Plant Physiol 121:4597. Mayerhofer LE, Macario AJL, Conway de Macario E (1992) J Bacteriol 174:300998. Yao R, Macario AJL, Conway de Macario E (1992) J Bacteriol 174:468399. Zellner G, Feuerhake E, Joerdening H-J, Macario AJL, Conway de Macario E (1995) Appl

Microbiol Biotechnol 43:566100. Zellner G, Macario AJL, Conway de Macario E (1997) FEMS Microbiol Ecol 22:295101. Macario AJL, Visser FA, van Lier JB, Conway de Macario E (1991) J Gen Microbiol

137:2179102. Garberi JC, Macario AJL, Conway de Macario E (1985) J Bacteriol 16:1103. Koenig H (1988) Can J Microbiol 34:395104. Howgrave-Graham AR, Macario AJL, Wallis FM (1997) J Appl Microbiol 83:587105. Ney U, Macario AJL, Conway de Macario E, Aivasidis A, Schoberth SM, Sahm H (1990)

Appl Environ Microbiol 56:2389106. Conway de Macario E, Macario AJL (1997) FEMS Microbiol Rev 20:59107. Franzmann PD, Liu Y, Balkwill DL, Aldrich HC, Conway de Macario E, Boone DR (1997)

Intl J Syst Bacteriol 47:1068108. Kotelnikova S, Macario AJL, Pedersen K (1998) Intl J Syst Bacteriol 48:357109. Sieburth JMcN, Johnson PW, Macario AJL, Conway de Macario E (1993) Mar Ecol Prog

Ser 95:81110. Conway de Macario E, Macario AJL, Pastini A (1985) Arch Microbiol 142:311111. Sekiguchi Y, Kamagata Y, Syutsubo K, Ohashi A, Harada H, Nakamura K (1998) Microbi-

ology 144:2655112. Haney PJ, Badger JH, Buldak GL, Reich CI, Woese CR, Olsen GJ (1999) Proc Natl Acad Sci

USA 96:3578113. Scandurra R, Consalvi V, Chiaraluce R, Politi L, Engel PC (1998) Biochimie 80:933


114. Conway de Macario E, Guerrini M, Dugan CB, Macario AJL (1996) J Mol Biol 262:12115. Metcalf WM, Zhang JK,Apolinario E, Sowers KR,Wolfe RS (1997) Proc Natl Acad Sci USA

94:2626116. Metcalf WW, Zhang JK, Wolfe RS (1998) Appl Environ Microbiol 64:768117. Tumbula DL, Whitman WB (1999) Mol Microbiol 33:1118. Macario AJL, Conway de Macario E (1993) Manipulation and mapping of microbes with

antibodies. In: Guerrero R, Pedros-Alio C (eds), Trends in microbial ecology. SpanishSociety for Microbiology, Barcelona, Spain, p 505

119. Ahring BK, Christiansen N, Mathrani I, Hendriksen HV, Macario AJL, Conway deMacario E (1992) Appl Environ Microbiol 58:3677

Received: December 2001

150 E. Conway de Macario · A. J. L. Macario: Molecular Biology of Stress Genes in Methanogens

Molecular Ecology of Anaerobic Reactor Systems

J. Hofman-Bang 1 · D. Zheng 2 · P. Westermann 1 · B. K. Ahring 1 · L. Raskin 3

1 Environmental Microbiology and Biotechnology, Biocentrum DTU, The Technical University of Denmark, Building 227, 2800 Lyngby, Denmark.E-mail: [email protected]; E-mail: [email protected];E-mail: [email protected]

2 Alpha Therapeutic Corporation, Los Angeles, CA 90032, USA.E-mail: [email protected]

3 Department of Civil and Environmental Engineering, University of Illinois at Urbana-Champaign, Urbana, Illinois 61801, USA. E-mail: [email protected]

Anaerobic reactor systems are essential for the treatment of solid and liquid wastes and con-stitute a core facility in many waste treatment plants.Although much is known about the basicmetabolism in different types of anaerobic reactors, little is known about the microbes respon-sible for these processes. Only a few percent of Bacteria and Archaea have so far been isolated,and almost nothing is known about the dynamics and interactions between these and othermicroorganisms. This lack of knowledge is most clearly exemplified by the sometimes un-predictable and unexplainable failures and malfunctions of anaerobic digesters occasionallyexperienced, leading to sub-optimal methane production and wastewater treatment.

Using a variety of molecular techniques, we are able to determine which microorganismsare active, where they are active, and when they are active, but we still need to determine whyand what they are doing. As genetic manipulations of anaerobes have been shown in only afew species permitting in-situ gene expression studies, the only way to elucidate the functionof different microbes is to correlate the metabolic capabilities of isolated microbes in pure cul-ture to the abundance of each microbe in anaerobic reactor systems by rRNA probing.

This chapter focuses on various molecular techniques employed and problems encoun-tered when elucidating the microbial ecology of anaerobic reactor systems. Methods such asquantitative dot blot/fluorescence in-situ probing using various specific nucleic acid probesare discussed and exemplified by studies of anaerobic granular sludge, biofilm and digestersystems.

Keywords. rRNA, rDNA, PCR, Biofilm, UASB, Granular sludge

1 Introduction . . . . . . . . . . . . . . . . . . . . . . . . . . . . 153

2 Nucleic Acid-Based Analysis of Anaerobic Bioreactors . . . . . . 154

2.1 Background . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 1542.2 Retrieving Nucleic Acid Sequences . . . . . . . . . . . . . . . . 1562.2.1 Nucleic Acid Isolation . . . . . . . . . . . . . . . . . . . . . . . 1572.2.2 PCR Reaction . . . . . . . . . . . . . . . . . . . . . . . . . . . . 1582.2.3 Cloning . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 1592.2.4 rDNA Sequences . . . . . . . . . . . . . . . . . . . . . . . . . . 1592.2.5 Community Fingerprints . . . . . . . . . . . . . . . . . . . . . . 1602.2.6 Quantification Based on Sequence Retrieval . . . . . . . . . . . 1612.2.7 Quantitative PCR . . . . . . . . . . . . . . . . . . . . . . . . . . 1622.2.8 Summary . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 162

CHAPTER 6


2.3 Oligonucleotide Probes . . . . . . . . . . . . . . . . . . . . . . . 1632.3.1 Probe Design . . . . . . . . . . . . . . . . . . . . . . . . . . . . 1642.3.1.1 Probe Specificity . . . . . . . . . . . . . . . . . . . . . . . . . . 1642.3.1.2 Target Accessibility . . . . . . . . . . . . . . . . . . . . . . . . . 1652.3.2 Quantitative Slot (Dot) Blot Hybridization . . . . . . . . . . . . 1652.3.2.1 Hybridization Stringency . . . . . . . . . . . . . . . . . . . . . 1662.3.2.2 Quantification . . . . . . . . . . . . . . . . . . . . . . . . . . . . 1662.3.2.2.1 Interpreting the Quantification Results . . . . . . . . . . . . . . 1662.3.2.2.2 Sensitivity . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 1672.3.2.2.3 Variation . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 1672.3.2.3 Factors that May Interfere with Quantification . . . . . . . . . . 1672.3.2.3.1 Membrane Saturation . . . . . . . . . . . . . . . . . . . . . . . 1672.3.2.3.2 Target Accessibility . . . . . . . . . . . . . . . . . . . . . . . . . 1672.3.2.3.3 Co-Extracted Substances . . . . . . . . . . . . . . . . . . . . . . 1682.3.2.3.4 In Vitro Transcribed rRNA . . . . . . . . . . . . . . . . . . . . . 1682.3.3 Reverse Genome Sample Probing . . . . . . . . . . . . . . . . . 1692.3.4 Whole Cell or in Situ Hybridization . . . . . . . . . . . . . . . . 1692.3.4.1 Hybridization Stringency . . . . . . . . . . . . . . . . . . . . . 1702.3.4.2 Cell Fixation . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 1712.3.4.3 Signal Enhancement . . . . . . . . . . . . . . . . . . . . . . . . 1712.3.4.3.1 Indirect Assays . . . . . . . . . . . . . . . . . . . . . . . . . . . 1722.3.4.3.2 Enzyme-Labeled Oligonucleotides . . . . . . . . . . . . . . . . . 1722.3.4.3.3 Multi-Probe and Multi-Labeling . . . . . . . . . . . . . . . . . . 1732.3.4.3.4 Amplification of the Target Sequence . . . . . . . . . . . . . . . 1732.3.5 Solution-Based Hybridizations (Molecular Beacons) . . . . . . . 1732.4 FISH and Reporter Systems . . . . . . . . . . . . . . . . . . . . 1752.5 FISH and Antibody Probes . . . . . . . . . . . . . . . . . . . . . 1762.6 FISH and Microautoradiography . . . . . . . . . . . . . . . . . 1782.7 Peptide Nucleic Acid Probes . . . . . . . . . . . . . . . . . . . . 179

3 rRNA-Based Analyses of Anaerobic Reactors . . . . . . . . . . . 180

3.1 Biofilm Reactors . . . . . . . . . . . . . . . . . . . . . . . . . . 1803.1.1 Biofilm Formation . . . . . . . . . . . . . . . . . . . . . . . . . 1803.1.2 Biofilm Composition and Dynamics . . . . . . . . . . . . . . . . 1893.2 Granular Sludge Reactors . . . . . . . . . . . . . . . . . . . . . 1903.2.1 Granular Sludge . . . . . . . . . . . . . . . . . . . . . . . . . . . 1903.2.2 Microbial Composition of Granules . . . . . . . . . . . . . . . . 1913.2.3 Structure of Granular Sludge . . . . . . . . . . . . . . . . . . . . 1933.2.4 The Granulation Process . . . . . . . . . . . . . . . . . . . . . . 1943.3 Continuously Stirred Tank Reactors (CSTR) . . . . . . . . . . . 1943.3.1 Microbial Composition in CSTRs . . . . . . . . . . . . . . . . . 1943.3.2 Microbial Dynamics in CSTRs . . . . . . . . . . . . . . . . . . . 196

4 Concluding Remarks . . . . . . . . . . . . . . . . . . . . . . . . 197

5 References . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 198

152 J. Hofman-Bang et al.

1Introduction

Most anaerobic microbial processes are characterized by close association ofnumerous functional groups of microorganisms. The understanding of anaero-bic processes has improved greatly during recent decades with advances madein microbial physiology, biochemistry, ecology, kinetics, and mathematicalmodeling. These contributions have led to an expansion of anaerobic processesby introducing better designs and operational controls. However, the under-standing of anaerobic processes is far from complete. Understanding the micro-bial ecology in anaerobic reactor systems requires

(1) identification and classification of microorganisms,(2) quantification of microbial abundance, and(3) quantification and identification of activity.

Morphology and other microbial traits have previously been used for identifi-cation and quantification of microbial populations. Grotenhuis et al. [1] micro-scopically counted cell numbers of methanogens and identified aceticlasticmethanogens based on morphology, and hydrogenotrophic methanogens byvisualizing autofluorescence at 420 nm. Morphology and ultrastructure havealso been used extensively in scanning or transmission electron microscopystudies to show the location of certain microorganisms in anaerobic granules [2, 3]. Information gained from morphology-based techniques is, however,ambiguous and limited since most microorganisms are small in size, and simplein morphology and ultrastructure.

In the absence of special morphological features or autofluorescence, physio-logical and biochemical traits have been used for identification. Furthermore,enrichments on defined substrates have been helpful to identify prevalentspecies in anaerobic granules [4], and Most Probable Number (MPN) estimateshave been used frequently for quantification of different trophic groups ofanaerobic microorganisms [1, 4]. These methods are, however, cultivation-dependent and therefore limited by the ability of microorganisms to grow underlaboratory conditions. It is well known that only a very small fraction of themicroorganisms in nature is culturable by present cultivation techniques,because of unrecognized nutrient and growth conditions, or the interruption ofintrinsic interdependencies such as syntrophic interactions [5]. Amann et al.estimated that the culturability ranges from 0.001% in seawater to 15% in acti-vated sludge [6]. Culturing may be especially difficult for anaerobes due to theirlow growth rates and fastidious nutritional and environmental requirements.

Recently, more direct methods have been developed for identification, quan-tification, and localization of microorganisms in environmental samples.Immunology techniques utilize monoclonal or polyclonal species-specific anti-bodies to detect and even quantify the abundance of cultivable microorganismsin environmental samples [7–9]. In combination with electron microscopy, anti-bodies have been used to localize microorganisms in sections of anaerobic gran-ules [1]. The major disadvantages of immunotechnology are the need for axeniccultures or defined co-cultures to produce the specific antibodies, and the high

Molecular Ecology of Anaerobic Reactor Systems 153

specificity that limits the detection to the species or subspecies level [10–12]. Inaddition, cross-reactions often might cause a problem [13, 14]. Adsorption tocross-reacting cells can broaden the specificity of an antibody [15], but thisapproach is limited by the number of species that are used for the specificity test.

Molecular phylogeny, which employs nucleic acid sequences to document thehistory of evolution, has provided a new basis for the direct identification andquantification of microorganisms [16]. Nucleic acid-based methods allowmicrobial community characterization without cultivation. So far, ribosomalRNA (rRNA) and ribosomal DNA (rDNA) have been the most commonly usedtarget nucleic acids in microbial ecology studies. This chapter focuses mainly on the use of rRNA- and rDNA-based methods for the study of anaerobic reactor systems. In addition, some other molecular approaches are discussedbriefly.

We first present the fundamentals and principles of different nucleic acid-based techniques to study anaerobic reactor systems. The second part of thechapter reviews literature in which rRNA- and rDNA-based techniques havebeen applied to studies of anaerobic bioreactors.

2Nucleic Acid-Based Analysis of Anaerobic Bioreactors

This section provides an overview of the most widely used or potentially appli-cable rRNA- and rDNA-based methods, but also presents studies in which func-tional diversity has been investigated by analyses of expressed messenger RNA.When available, we have used examples from anaerobic bioreactor work.

2.1Background

Studying microbial ecology requires identification of microorganisms, basedupon a comprehensive classification system that ideally should reflect the evo-lutionary relatedness of organisms [5].As pointed out by numerous authors, tra-ditional classification systems based on phenotypic characteristics (morpholo-gy, physiology, and structure of cell components) offer little information on evo-lutionary relatedness and require cultivation for identification [5, 17].

In the mid-1960s, Zuckerkandl and Pauling pointed out that molecularsequences could document evolutionary history [18]. Due to the pioneeringwork of especially Carl Woese, the rRNAs have become the most commonly usedmolecules for phylogenetic analyses. rRNA or the corresponding rDNA are par-ticularly suitable as evolutionary chronometers [19–21] since

(1) they are key elements of the cells and are functionally and evolutionarilyhomologous for all organisms;

(2) they are very conserved in overall structure;(3) their regions of different conservation levels allow phylogenetic analysis

and design of probes and primers;


(4) they are very abundant in most cells (103 to 105 copies) [6], and are easilyrecovered and detected;

(5) the small subunit (SSU) rRNA (16S and 16S-like rRNA) and the large rRNAof the large subunit (LSU) of the ribosome (23S rRNA and 23S-like rRNA)are sufficiently long for statistically significant comparisons; and

(6) their genes have so far not been shown to be transferable among organisms.

Using 16S rRNA comparative sequence analysis, Woese and colleagues devel-oped the first universal phylogenetic tree, which reflects the evolutionary relat-edness of all organisms, grouping them into three domains: Eucarya, Archaea,and Bacteria [22–24].

16S rRNA sequences are most commonly used for molecular ecology investi-gations since a huge number is available through the Ribosomal Database Pro-ject II (16,277 aligned and 30,322 unaligned 16S rRNA sequences in June, 2000)[25]. However, 23S rRNA-based analyses should become more common whenmore sequence data become available, since 16S rRNA comparisons sometimesfail to resolve very closely related species [20, 21, 26]. Internal transcribed spac-er (ITS) regions that separate rRNA genes may provide additional informationfor resolving very close phylogenetic relationships [27]. A complication relatedto the use of rRNA as a target for the quantification of population abundance isthe limited information currently available on the number of rRNA operons pre-sent on microbial genomes. Information on the level of gene redundancy presentin the rRNA operons has recently been catalogued in the Ribosomal RNA Oper-on Copy Number Database (rrndb) [28]. Besides the ribosomal RNA genes, oth-er gene sequences have been used for phylogenetic analyses, including genes forthe elongation factor Tu, and F1F0ATPase b-subunit [29].

A phylogenetic analysis allows the identification of a microorganism basedonly on a molecular sequence, eliminating the need for cultivation. In otherwords, a sequence can be retrieved from an environmental sample, sequenced,and compared to known sequences for identification of the correspondingorganism [19]. If the retrieved sequence is new, characteristics associated withhousekeeping functions of the cells (e.g., characteristics of ribosomes, DNAreplication machinery, biosynthetic pathways and their regulation mechanisms)can be inferred from closely related species [5, 17]. The metabolic diversity ofcells, however, is more variable than reflected by the housekeeping functions,due to events such as lateral transfer of metabolic genes and symbiotic fusions[17]. Thus, caution has to be taken when metabolic characteristics of a newlyidentified microorganism are inferred from characteristics of close phylogenet-ic relatives.

Based on “signature” sequences of specific groups of microorganisms, probescan be designed and used to identify and quantify these microorganisms incomplex microbial ecosystems. The strategies based on rRNA sequences analy-sis for characterizing a microbial community are summarized in Fig. 1. It shouldbe noted that many of these strategies are also applicable to other genes.


2.2Retrieving Nucleic Acid Sequences

Methods for retrieving nucleic acid sequences from environmental samples aremainly used to detect and identify microorganisms, although some quantitativemethods are being developed, especially for populations present in low num-bers. The sequences obtained should ideally represent the diversity present inthe sample, but all methods introduce at least some bias and may not identify allpopulations present as discussed below.

The process starts with the extraction of nucleic acid (DNA or RNA) from asample.Extracted DNA can be randomly digested and cloned (shotgun cloning).Subsequently, the clones are screened for rRNA genes using dot/colony blothybridization. More commonly, however, the rRNA genes in DNA extracts arespecifically amplified by PCR, or rRNA genes are produced by reverse tran-scription from rRNA in RNA extracts followed by PCR (i.e., RT-PCR). Next, PCRor RT-PCR products are cloned or separated by gel electrophoresis (denaturinggradient gel electrophoresis [DGGE], temperature gradient gel electrophoresis[TGGE], terminal restriction fragment length polymorphism [T-RFLP]). TherDNA clone library or the DNA bands from the electrophoresis gel can besequenced and the obtained sequences are deposited in sequence databases.


Fig. 1. Strategies based on rRNA sequence analysis for characterization of microbial commu-nities without cultivation (arrows indicate the interconnected use of methods, experimentalmaterials, and information in the study of microbial ecosystems. RT-PCR: reverse transcrip-tion to produce DNA from RNA, followed by PCR. DGGE: denaturing gradient gel elec-trophoresis. RFLP: restriction fragment length polymorphism. Modified from [5, 6]

Subsequently, phylogenetic trees showing the diversity of the correspondingenvironmental sample can be constructed by comparative sequence analysis.Details of these methods and their applications can be found in a number ofreviews [5, 6, 17, 19, 20, 26, 27, 30, 31]. Brief descriptions of the methods and ofsome factors that affect the overall results are discussed below.

2.2.1Nucleic Acid Isolation

Ideally, a sample should be representative and free from bias, especially whenquantification is the objective. It should also contain sufficient biomass for sub-sequent analyses. In addition, the presence of materials that interfere with nucle-ic acid recovery or manipulation (e.g., humic substances) should be avoided orsuch compounds should be removed from the sample if possible. If the sampleis collected for RNA extraction, nuclease activity should be reduced as much aspossible [32]. Measures such as quick freezing, storing at –80°C, avoiding thaw-ing the sample before extraction, and preventing the introduction of foreignnucleases should be practiced.

Several methods have been published in the recent years for extraction ofintact RNA from environmental samples, such as manure [33], sediment, soil,and water samples [34], sediment and microbial mat samples [35], and rumenfluid [39]. Recovering RNA or DNA quantitatively from all cells in a complexcommunity without bias can be difficult. In general, mechanical lysis methodshave shown less bias than enzymatic lysis methods, leading to the recovery ofintact high molecular weight nucleic acids [36].

Ibrahinm et al. reported on a rapid method for extracting high purity rRNAfrom manure [33]. The bead-beating-based method involves citrate buffered,low-pH phenol and chloroform extractions.This method effectively disrupts thecell wall of cells that are difficult to break such as Gram-positive bacteria andmethanogens. Citrate has been shown to strongly inhibit RNases [37]. Humicacids were removed from the samples by repeated washes with low-pH citrateprior to cell disruption.

Moran and coworkers used a low-pH, hot-phenol extraction method and sub-sequent gel filtration with Sephadex G-75 spin columns for sediment, soil, andwater samples [34]. Since they used lysozyme to open the cells, this method mayintroduce a bias, since lysozyme is not equally effective for all types of cells.

Alm and Stahl compared different lysis solutions and subsequent vortexing inlow-pH phenol and chloroform to extract RNA from sediments and microbialmats [38]. Use of a guanidine thiocyanate/b-mercaptoethanol lysis solutionresulted in an efficient recovery of intact, high purity rRNA. When they in-creased the ratio of lysis buffer to sample volume from a 1 :1 to 5 :1, an order ofmagnitude more rRNA was extracted. Adding the sample to the lysis bufferwhile mixing, instead of the opposite also proved to be important. This extrac-tion method left the final extract with considerable amounts of organic conta-mination.

Raskin and coworkers used a low-pH, hot phenol, bead-beating extrac-tion method to isolate RNA from rumen fluid [39]. They demonstrated that


the total amount of RNA recovered per gram of rumen fluid neither changedsignificantly with the duration of the beating, nor with the amount of beadsused. However, the amount of RNA recovered from Gram-positive Rumino-coccus cells increased significantly when the beating period was extended. In the same study, the total amount of RNA recovered in replicate extractionsshowed very high variability. The efficiency of RNA recovery was also investi-gated [39]. The study demonstrated proportional recovery of RNA for a specifictarget population. The study also demonstrated loss of RNA in the phenol phase and during precipitation/rinse/resuspension steps of the extractionprocess.

Yu and Mohr developed a fast method (one hour) to simultaneously extractDNA and RNA [36]. Bead beating combined with 2 M ammonium acetate pre-cipitation of proteins in the presence of DEPC (diethyl pyrocarbonate) resultedin intact rRNA and non-sheared DNA. No phenol or chloroform was used. Byadding RNase or 200 mM NaOH, either DNA or RNA was extracted. The nucleicacids were of a sufficient quality to perform PCR and RT-PCR. This method hasbeen adapted to extract intact DNA or rRNA from municipal solid waste andcow manure without any washing prior to extraction in our laboratory. Theresuspended nucleic acids obtained after extraction are colorless, indicating thathumic substances and other impurities are removed effectively during theextraction procedure.

Humic substances coextracted with nucleic acids may interfere with subse-quent enzymatic reactions (such as PCR) [27]. They can also interfere withmembrane hybridizations (see below). A number of methods has been devel-oped for removing the humic substances. These include polyvinylpolypyrroli-done (PVPP) adsorption, gel purification, and dilution [27, 34, 35]. Furthermore,DNA is often present in RNA extracted from environmental samples [38, 40].The influence of DNA on membrane hybridization is discussed below. Whennecessary, DNase can be used to remove DNA, although concerns of partialdegradation of RNA due to impurities in commercial DNase should be consid-ered [40].

2.2.2PCR Reaction

The polymerase chain reaction (PCR) can be used to amplify DNA sequencesfrom environmental samples. The PCR products can be analyzed by techniquessuch as DGGE (denaturation gradient gel electrophoresis), TGGE (temperaturegradient gel electrophoresis), T-RFLP (terminal restriction fragment lengthpolymorphism), or SSCP (single stranded conformation polymorphism), whichhave the potential to separate the PCR products originating from different DNAsequences representing populations in the original samples. The PCR productscan also be cloned and subsequently sequenced to allow identification of popu-lations. For details about PCR, the reader is referred to a review by Steffan andAtlas and “The PCR application manual, 2nd ed. [41, 42]. Since the amount ofDNA produced by PCR ideally increases exponentially during the amplification,errors occurring early in the process will result in biased results [6].


Factors that cause bias of PCR are:

– “Universal” primers or other specific primers are designed based onsequence information available in databases (obtained from cultured organ-isms and clones) [5]. However, primers targeting multiple groups of organ-isms may not amplify all target genes since the primer sites are not complete-ly conserved.

– Bias can be caused by an inappropriate annealing stringency, which results inamplification of genes that are not intended to be amplified [27].

– There is some evidence that PCR does not amplify all rRNA sequences in thesample to the same extent (preferential amplification) [6, 27].

– Contaminating sequences from chemicals and enzymes can be erroneouslyincluded in the analysis.

– Chimeric sequences are often produced [27] due to the presence of partialfragments of rDNA in DNA extracts, partially reverse transcribed DNA whenperforming RT-PCR, or premature PCR products acting as primers in a sub-sequent PCR cycle [6].

2.2.3Cloning

Cloning can produce large amounts of DNA segments originally isolated fromenvironmental samples. The DNA fragments can be produced after digestionwith restriction enzymes of the DNA extracted from a sample (i.e., shotguncloning), or after PCR or RT-PCR (if RNA is the template). Compared to cloningafter PCR, shotgun cloning introduces less bias and produces clones of multiplegenes at the same time [5]. Cloning after PCR is rapid and convenient, but canbe biased [5, 27]. The bias can be introduced during the PCR step as discussedabove or during cloning. For instance, the use of rare-cutting restrictionenzymes during cloning might also cut amplified rDNA [6]. In addition, it is pos-sible that different rRNA gene fragments are cloned with different efficiencies.

2.2.4rDNA Sequences

Cloned DNA fragments can be sequenced to study the phylogenetic diversity ofthe microbial community from which the sample was originally obtained. Theresulting sequences can be compared to sequences in databases to identify theclosest phylogenetic relatives.

There are a number of problems that need to be taken into considerationwhen using this technique. First, when the retrieved sequences exhibit high similarity to sequences available in databases (98% to 99% identity), it is difficult to rule out PCR errors [27]. Secondly, it is difficult to convert the rRNAsimilarity to the nomenclature level of species or genus [6]. In general, morethan 97% 16S rRNA identity indicates that two sequences belong to the samespecies, which typically corresponds to DNA:DNA hybridization values above70%. The definition of a similar threshold for genera is not as clear, but 16S


rRNA differences greater than 5–7% may be used to support a new genus [43].Finally, the heterogeneity of 16S rRNA genes increases the complexity. It hasbeen observed that some species express different 16S rRNA genes at differentgrowth stages, or multiple 16S rRNA genes are expressed at the same time [27].For example, sequence heterogeneity was found in Paenibacillus polymyxa [45],strains of the Mycoplasma mycoides cluster [44], and in Clostridium para-doxum [45a].

2.2.5Community Fingerprints

Several fingerprinting techniques, such as DGGE, TGGE, RFLP, T-RFLP, andSSCP, have been developed to screen clone libraries, to estimate the level ofdiversity in environmental samples, to follow changes in community structure(e.g., trace one or more populations over time) and to compare diversity andcommunity characteristics in various samples. These techniques usually involvegel electrophoresis that can separate different DNA segments of a communityrDNA library.

DGGE (denaturing gradient gel electrophoresis) separates DNA fragments ofequal length (obtained after PCR of DNA extracted from an environmental sam-ple) into distinct bands on a chemical denaturing gradient polyacrylamide gel.PCR amplification of the 16S rRNA gene utilizing conserved primers targetingeither the V3 or the V8 +V9 variable regions is normally used to produce a300–500 bp fragment. Larger fragments are typically not used as the DGGE tech-nique cannot resolve these into distinct bands [46]. One of the primers used hasa GC-clamp consisting of a 30 nucleotide GC-rich 5¢ end, which maintains thetwo denaturated single stranded DNA fragments together in the denaturing gel.As the double stranded DNA migrates through the gel experiencing increasing-ly higher denaturant concentrations, the double stranded DNA separates intotwo single strands at a specific point and the migration stops due to the largervolume of the denaturated molecule kept together by the GC clamp. The DGGEtechnique has been used to characterize the microbial diversity in different envi-ronments such as activated sludge [47], sediments [46], lake water [8], hotsprings [48], soils [49, 50], biofilm [51]. DGGE has been used to monitor changesin complex communities [48, 52–55] and to identify microorganisms present inwall painting [56].

The banding pattern reveals the community components at best semi-quan-titatively due to the possible bias caused by PCR and difficulties to quantify theamount of DNA associated with a band. An advantage of the technique is that itcan resolve the microbial diversity of up to 15 different species by optimizing thedenaturing gradient concentration in the gel. By using narrow gradients, rDNAsthat differ in only one bp can be separated in DGGE [46]. A drawback of the technique is that the reproducibility is not optimal; one DNA fragment may generate more than one band on the gel and a DNA sample analyzed on two different gels may not generate the same band pattern [57]. In addition, it is possible that a band in a DGGE gel may contain different sequences with similar denaturation characteristics.


Alternatively, the obtained rDNA can be digested with restriction enzymesand analyzed on an agarose gel. This technique is called restriction fragmentlength polymorphism (RFLP) [58] or amplified ribosomal DNA restrictionanalysis (ARDRA) [59]. The banding pattern obtained has been used to identi-fy different genotypes of microorganisms [60] and to monitor populationchanges in environmental samples [61, 62]. The RFLP patterns have also beenused to deduce the phylogeny of axenic cultures of microorganisms [63–67].Compared to cloning, DGGE and RFLP are faster and less laborious, and bandsof interest can be cut out, extracted from the gels, and cloned and sequenceddirectly.

SSCP (single-strand conformation polymorphism) is based on the separationof the double stranded DNA PCR product by NaOH prior to non-denaturingpolyacrylamide gel electrophoresis. The single stranded DNA forms secondarystructures (analogous to the cloverleaf structure of tRNAs). Scheinert andcoworkers employed this technique to differentiate between 15 Mycoplasmaspecies and to analyze a mixed sample of six different species based upon analy-sis of the spacer region between the 16S rRNA and the 23S rRNA gene [68]. Theadvantage of the technique is that even point mutations can be detected as achange of conformation in the secondary structure of the single stranded DNA.In the study of the Mycoplasma species, the variable size of the PCR product ofthe spacer region (280–1300 bp) further discriminated the populations while the16S rDNA DGGE technique might not have been able to do so since the gel bandpattern would have been too compressed.A critical parameter in the SSCP tech-nique is to control the temperature in the gel tank to reduce smearing. As theavailable sequence data of the rRNA spacer region is limited compared to the16S rRNA sequence databases, the SSCP technique is mainly used to differenti-ate between different cloned PCR products.

2.2.6Quantification Based on Sequence Retrieval

Theoretically, the abundance of a population can be inferred from the frequen-cy of a particular sequence appearing in the sequence collection obtained froman environmental sample. For instance, the microbial community structure ofan anaerobic fluidized-bed reactor (treating wine distillation wastewater) wascharacterized by PCR and sequencing [69]. The PCR was conducted using threepairs of primers specific for the three domains. The authors obtained 460 and96 clones from Bacteria and Archaea, respectively. Of these 556 clones, 76% wereBacteria, 10% corresponded to Methanobacterium formicicum, 4% representedMethanosarcina frisius, 8% were Methanosarcina barkeri, and 2% representedother Archaea. Within the bacterial domain, there were 6% high G + C Gram-positives, 4 % Planctomyces, 33% low G + C Gram-positives, 4% Spirochaetes,12% delta Proteobacteria, 2% gamma Proteobacteria, 1% beta Proteobacteria,2% alpha Proteobacteria, 26% Cytophaga-Flexibacter-Bacteroides, and 7%green non-sulfur bacteria. As discussed by the authors, this method has manyshortcomings when used as a quantitative tool. First, the primers were designedbased on previously isolated cultures, thus it is not free from the well-known cul-


tivation-derived limitations. Secondly, bias can be introduced during PCR andnucleic acid extraction as previously discussed.

2.2.7Quantitative PCR

Recently, a number of quantitative PCR methods have been developed that havethe potential of detecting low-level populations in environmental samples.

One of these methods is competitive PCR, in which an internal standard isadded to the sample. The sample and the internal standard are amplified usingthe same pair of primers. The corresponding assumptions are: The gene of inter-est and the internal standard are equally accessible to primers after denatura-tion; both templates have the same efficiency to hybridize to the primer and tobe extended by the polymerase; substrate exhaustion affects the extension ofboth templates equally [70]. However, in a competitive PCR experiment [70], theauthors observed a bias that was strongly dependent on the number of replica-tion cycles. They demonstrated that reannealing of genes progressively inhibit-ed the formation of template-primer hybrids.

Taqman PCR is another quantitative method exploited for detection of low-level populations. This method takes advantage of the 5¢ to 3¢ exonuclease activ-ity of the Taq DNA polymerase [71]. A probe targeting one strand of the PCRtemplate is labeled at the 5¢ end, and its 3¢ end is phosphorylated to preventextension. During PCR, the Taq polymerase extends the ordinary primer alongthe template strand. When it meets the probe that binds to the template strand,it cleaves the 5¢ terminal nucleotide and produces mono- or oligonucleotides,which are shorter than the original probe. In the first Taqman study [71], theprobe was labeled with 32P. Autoradiography after TLC (thin layer chromatogra-phy) was needed to detect the hydrolyzed probes. The original Taqman methodwas modified [72] to allow rapid analysis by labeling the probe with a fluores-cent dye at the 5¢ end and with a quencher at the seventh nucleotide from the 5¢end. The dye fluoresces when the probe is cleaved between the dye and thequencher (Fig. 2). Therefore, there is no need for post-amplification separation.

Subsequently, Taqman PCR was demonstrated to be quantitative since theintensity of fluorescence was proportional to the amount of PCR product, andunder appropriate conditions, to the initial number of the templates [73]. How-ever, it is necessary to assume that the efficiency of the PCR to amplify DNAfrom an environmental sample is similar to the efficiency of the PCR used forconstructing the standard curve, before this method can be used to quantifypopulations in environmental samples. Since this assumption may not always bevalid, bias can occur.

2.2.8Summary

Despite the potential biases of the various methods discussed above, retrieving16S rRNA sequences directly from environmental samples allows the investiga-tion of microbial communities without cultivation. The use of these techniques


has revealed that the microbial diversity is much greater than was anticipatedbased on cultivation studies.

2.3Oligonucleotide Probes

The first step of oligonucleotide probe hybridizations consists of probe designas illustrated in Fig. 1. The probes are used in various types of hybridizations todetect, quantify, and localize the target sequences or cells in a sample, in a nucle-ic acid extract, or in a clone library. For more information on environmentalapplication of nucleic acid hybridization, several reviews are available [10, 39,74, 75].

This section focuses on probes rationally designed, using the phylogeneticframework provided through comparative analysis of sequences available indatabases [76]. In particular, the large collection of 16S rRNA sequences makesit possible to design a nested set of 16S rRNA-targeted oligonucleotide probeswith different levels of specificity. By comparing aligned 16S rRNA sequences,unique regions can be found that are shared only by the target population(s).

Empirically designed probes have traditionally been generated from agenomic recombinant library or simply are the total genomic DNA obtainedfrom a target organism [76]. These probes have not been used much in micro-bial ecology research in the last decade because of limitations with nesting andquantification. An obvious advantage of oligonucleotide probes is that targetsdiffering in a single nucleotide can be discriminated under appropriate experi-mental conditions. A few applications of empirically designed probes have beenpublished. DeLong et al. [12] studied the correlation between growth rates ofEscherichia coli, the average ribosome contents, and the fluorescence conferredby hybridization probes. They observed that with decreasing growth rates the


Fig. 2. Taqman PCR. Modified from [72]

hybridization signal quickly approached the limit of detection of epifluores-cence microscopy or flow cytometry. Using oligonucleotides carrying multiplelabels both in the hybridizing probe and in a non-complementary tail did notsignificantly increase the sensitivity [77]. A possible way to identify cells withlow metabolic activity, i.e., low amount of ribosomes, is by applying polynu-cleotide probes carrying multiple fluorescent reporter molecules [78]. In thisstudy a polynucleotide probe (ca. 200 to 300 nucleotides in length) was generat-ed by transcription of a cloned probe sequence from the 23S rRNA gene fromPseudomonas stutzeri. The probe was selected to target the variable domain IIIregion in the 23S rRNA molecule. Whole-cell hybridization proved that thepolynucleotide probe was superior to the oligonucleotide probes for in situdetection of cells with low cellular rRNA contents. Also, the larger probes coulddifferentiate between two closely related organisms Pseudomonas stutzeri andPseudomonas diminuta.

Heuer and coworkers utilized digoxigenin-labeled probes targeting the 16SrRNA molecule to fingerprint the microbial community in rhizosphere samplestaken from potato plants [79]. Briefly, PCR amplified 16S rDNA genes (the V6region) were separated by temperature gradient gel electrophoresis (TGGE),cloned, sequenced and used to probe dot blotted 16S rDNA amplified from bac-terial isolates. To optimize the specificity of the probes, flanking conservedregions were removed. One truncated probe hybridized to three different bacte-rial isolates all with different electrophoretic mobility in the TGGE. Subsequentsequencing of these isolates revealed an identical V6 region but different V7 andV8 regions explaining the above observations. Also, the polynucleotide probeswere shown to discriminate between targets differing only by two nucleotides.

The advantage of polynucleotide probes compared to oligonucleotide probesis their higher specificity. Even though the specificity of oligonucleotide probesis sufficient for many applications, more specific probes are sometimes neededto discriminate between two closely related organisms. Several groups of organ-isms have been identified which share almost identical 16S rRNA sequences butamong which DNA:DNA hybridization values are lower than 70% [80].

2.3.1Probe Design

2.3.1.1Probe Specificity

The specificity of a newly designed probe has to be tested before it can be usedwith confidence. The specificity can be checked using rRNA databases such asthe CHECK_PROBE software provided in the Ribosomal Database project(http://www.cme.mwu.edu/RDP/html/analyses.html) [25] and the Oligonu-cleotide Probe Database (OPD) [81]. Alternatively, the BLAST network serviceavailable from the National Center for Biotechnology Information athttp://www.ncbi.nlm.nih.gov can be used. Due to the limited collection of rRNAsequences compared to the total estimated prokaryotic diversity, there is a pos-sibility that some yet undiscovered sequences are targeted by probes designed


for other organisms [27]. As pointed out by Ward et al., “probes should beregarded as tools subject to refinement” [27]. Probe nesting is another way tocheck the specificity of a probe [39]. Theoretically, the sum of quantificationresults from an environmental sample at one taxonomic level (e.g., family)should equal the quantification result of the taxa at one higher level (e.g., order).If this is observed,all the probes used are probably specific. If the population sizeobtained for, e.g., an order, however, is larger than the sum of populations rec-ognized at the family level, an unknown family not targeted by the family probesmight exist [39].

2.3.1.2Target Accessibility

This needs to be considered when probes are used for in situ or whole cellhybridization, since the higher order structure of the target rRNAs in this typeof hybridization is intact and since rRNAs are associated with ribosomal pro-teins [6]. Fixation can denature the higher order structure. However, since theinfluence of fixation is hardly predictable, there is no easy estimation of accessi-bility to the target [6]. Some of the 16S rRNA and 23S rRNA sites that have beensuccessfully targeted were summarized by Amann and coworkers [6, 82].A moresystematic study of target accessibility was reported by Frischer and co-workers[83]. Five probes each consisting of 12 nucleotides were designed to target the515, 786, 1063, 1341, and 1369 sites of Escherichia coli 16S rRNA. Hybridizationsignals of all the five probes were equal in hybridization to cell blotted mem-brane, but different in whole-cell hybridization. Only probe 1341 gave a goodsignal in whole-cell hybridization. Probe 515, which targeted a ribosomal pro-tein-binding site, showed moderate signals, but the inhibition by the proteinsseemed to be outcompeted by a longer probe. The study showed that probe 786,which targeted a loop site, gave a moderate signal for unknown reasons. Target-ing the self-complementary sites did not seem to be a problem since probe 1369,which targeted a less self-complementary site, gave a lower signal thanprobe 1341, which targeted a higher self-complementary site. Another study also showed that targeting highly structured sites with probes consisting of30 nucleotides resulted in good signals [84]. More recently, Fuchs et al. con-vincingly conducted a systematic study on the accessibility of 16S rRNA targetsites in E. coli by probing with more than 200 probes along the 16S rRNA mole-cule showing regions with high accessibility and other regions with low accessi-bility [85]. Fuchs et al. also showed improved accessibility to otherwise lowaccessibility regions by applying unlabeled helper oligonucleotides binding nextto the labeled probe’s target site [86].

2.3.2Quantitative Slot (Dot) Blot Hybridization

In general, the quantitative slot (dot) blot hybridization involves the applicationof rRNAs extracted from environmental samples on membranes together with adilution series of RNA from an axenic culture (reference RNA). The membranes


are then prehybridized, hybridized with the probes, and washed. Usually mem-brane hybridizations are conducted at low stringency (high salt concentrationsand low temperatures). The washing step is then used to remove all excess non-specific binding probe. The signals from the environmental samples are quanti-fied by comparison with the reference rRNA. The abundance of a specific popu-lation is expressed as a percentage of the total rRNA determined by a universalprobe targeting all rRNA. Alternatively, results can be reported in terms of µg ofrRNA per sample volume or weight.

2.3.2.1Hybridization Stringency

The washing conditions are critical in order to distinguish between targetsequences and sequences with one or more mismatches. The specificity is usu-ally controlled by temperature. The optimum washing temperature (Tw) is rec-ommended to be equal to, or slightly higher than, the dissociation temperature(Td) which is the temperature at which 50% of the duplexes remain intact dur-ing a specified washing period [87]. Another closely related parameter is themelting temperature (Tm) which is defined as the equilibrium temperaturewhere half of the duplexes are dissociated [76, 87]. Thus, Tm is defined for equi-librium conditions and is time-independent, whereas the Td value is time-dependent. There are a number of empirical equations that can be used to esti-mate the Tm and Td of a duplex [76, 87]. These equations provide guidelines forprobe design, especially when certain Td values are needed (e.g., the design of anumber of probes for simultaneous in situ hybridization experiments) [39].However, since the Td is a function of numerous factors [76, 87] such as duplexstructure and length, sequence, nucleotide content, number and type of mis-matches, terminal unpaired bases, as well as the hybridization and washing con-ditions, the experimental determination of Td is highly recommended.

2.3.2.2Quantification

2.3.2.2.1Interpreting the Quantification Results

When interpreting the quantification results obtained from membrane hybrid-ization experiments, it is important to consider that rRNA abundance does notequal cell abundance, since the number of rRNA molecules in each cell lies with-in a very broad range (103 to 105) [6]. However, since the cellular rRNA contentof a cell often is correlated to its growth rate [12], the abundance of the rRNA isan indication of the metabolic activity. Since one genotype can be related to sev-eral phenotypes, derivation of specific physiological activities from rRNA abun-dance should be carried out with caution.Although the absolute abundance of apopulation in terms of mass of rRNA per weight or volume of a sample is desir-able, this type of quantification result should also be carried out with cautiondue the high variability of RNA recovery in the extraction process as previouslydiscussed.


2.3.2.2.2Sensitivity

The sensitivity of membrane hybridizations can be as low as 0.1% when an iso-tope-labeled probe is used [6, 27]. Based on observations in our laboratory, thissensitivity can be obtained by proper loading (highest loading without saturat-ing the membrane, see [40]) and high radioactivity of the probes. This has theconsequence that populations having a rRNA abundance below 0.1% of the totalrRNA in a sample cannot be detected by membrane hybridization. However,low-abundance populations can be detected with the aid of PCR.

2.3.2.2.3Variation

By comparing different types of commercially available membranes, [32] it wasshown that the detectability and local variation differed significantly from onetype of membrane to another. The local variability of a membrane (Type I)ranged from 10 to 50%, and was believed to be the primary cause for variationin quantification results. Several other types of variability during membranehybridizations were tested [39]. It was demonstrated that variability introducedby denaturation/dilution (Type II) and prehybridization/hybridization (TypeIII) were not statistically significant. The washing step (Type IV), however, couldintroduce significant variation. Therefore, it was recommended to apply eachsample in triplicate to compensate for the Type I variation. Samples and refer-ence rRNA series should also be washed in the same tube or beaker to avoidType IV variation.

2.3.2.3Factors that May Interfere with Quantification

2.3.2.3.1Membrane Saturation

Membrane saturation can be one of the many reasons that a non-linearhybridization response is observed [39, 40]. The saturation of Marga Chargemembranes (MSI, Westborough, MA) was determined to be around 150 ngnucleic acid/slot using 32P-labeled E. coli rRNA [40].

2.3.2.3.2Target Accessibility

Accessibility of probes to the 16S rRNA sequence can be different from site tosite, even in membrane hybridizations [32]. This study showed that thehybridization signal increased as the denaturation conditions increased (interms of temperature, time, and the concentration of glutaraldehyde) for the target site 628 ( E. coli 16S rRNA numbering). However, for site 1392, higher lev-els of denaturation resulted in lower signals. It was hypothesized that a higher


level of denaturation might result in loss of signal because site 1392 has rela-tively low levels of secondary structure [32]. In another study, McMahon et al.[88] also showed that the effect of denaturation conditions on membranehybridization signals was target site-dependent. The difference in accessibilityalong the 16S rRNA could therefore cause bias in quantification if differentprobes are used.

2.3.2.3.3Co-Extracted Substances

The presence of as little as 5.6 µg humic substances in 10 ng RNA can lowerhybridization signals [38].Although DNA does not contribute much to non-spe-cific binding during membrane hybridization [38], high concentrations of DNA(higher than 10 ng in 10 ng RNA) can reduce hybridization signals [40]. Since atotal of 20 ng nucleic acids (DNA and RNA) is much lower than the saturationlimit of the membrane [40], the inhibition is caused by mechanisms other thansaturation. However, since hybridization signals are reduced for both specificand universal probes, results expressed in terms of percentage should notchange significantly due to the presence of inhibitory compounds, but detect-ability is reduced.

2.3.2.3.4In Vitro Transcribed rRNA

An advantage of the oligonucleotide probe method to previously used antibodymethods is that a probe can be designed and synthesized without the avail-ability of an axenic culture. However, there is still a requirement for pure culture rRNA for probe characterizations, for specificity studies, and for standards during quantitative membrane hybridizations. In vitro transcribedRNA or rcDNA (obtained by reverse transcription) was suggested as a substi-tute for native RNA by Ward et al. [27] and this method has been used in somestudies [89–92]. However, the behavior of the in vitro transcribed rRNA andrcDNA was not compared to native rRNA in those studies. Polz and Cavanaugh[93] and McMahon et al. [88] both agreed that in vitro transcribed rRNA can be used to determine the Td values of the native rRNA, although there is some disagreement between the two studies. Polz and Cavanaugh found that transcribed rRNA resulted in 2 to 3°C higher Td values compared to native rRNA using probes S-D-Bact-0338-a-A-18 for Bacteria and S-S-V.ang-0219-a-A-20 for Vibrio anguillarum. McMahon et al. found, however, that transcribed rRNAs have the same Td values as the native rRNA for probes S-*-Synb-0222-a-A-19, S-S-S.fum-0464-a-A-19, S-F-Synm-0700-a-A-23, and S-*-Univ-1390-A-a-18. Both studies demonstrated bias when the transcribed rRNAs were used for quantification by membrane hybridization, even though Polz and Cavanaugh claimed that the bias was not statistically significant due to the high variability of the hybridization signals. This will restrict the usage of in vitro transcribed RNA for absolute quantification of microbial population abundance.


2.3.3Reverse Sample Genome Probing

The reverse sample genome probing technique was developed by Voordouw andcoworkers in the mid-1990s [95]. Instead of using a probe targeting a genus or afamily, genomic DNAs from reference strains are immobilized on a membrane.DNA extracted from an environmental sample is then isotope-labeled by nicktranslation and used as probes. This method is well suited to study diversity ofmicrobial groups like sulfate-reducing bacteria, syntrophs or methanogens.rRNA probing detects how many known and unknown species containing theprobe sequence are present in a sample while the reverse sample genome prob-ing detects specific species or strains. To carry out the same task by rRNA prob-ing, many hybridizations would need to be performed using different probes.The DNA probe is much more specific because it contains thousands of genes.Aprerequisite, however, is a proper selection of the reference strains to minimizecross-hybridization to closely related strains. Each DNA reference spot onlydetects other genomes present if they share enough homology. This techniquehas been shown to discriminate down to the species level [96].

Voordouw and coworkers have used the technique to study diversity in oil-containing environments [96, 97]. Twenty-six sulfate-reducing bacteria (16belonging to the genus Desulfovibrio) plus other reference species were used toprobe oil field samples to detect environmental nitrate- and sulfate-reducingbacteria. Hybridization signals were shown to specific Desulfovibrio species butnot to others. Thereby the authors were able to identify and to quantify the rel-ative amount of the different species present in a single hybridization event. Aproblem with this technique is that the DNA extracted from the samples mayoriginate from dead cells. Also, the generation time of specific populations hasto be considered to ensure that environmental changes are reflected in the bac-terial populations. rRNA probing may detect changes in population sizes andactivities that occur within days, while the reverse sample genome probing cando the same on a week scale, but in higher detail.

2.3.4Whole Cell or in Situ Hybridization

Amann et al. define whole cell hybridization as hybridization performed withmorphologically intact cells, and in situ hybridization as whole cell hybridiza-tion targeting cells in their natural habit [6, 82]. The term “fluorescence in situhybridization (FISH)” is used for both whole cell and in situ hybridizations. Theperspectives of whole cell and in situ hybridizations are discussed below.

1) FISH can show the three-dimensional spatial distribution and morphology ofuncultured cells. For instance, Harmsen et al. [90, 98] used fluorescentlylabeled probes to reveal the internal structure of anaerobic granules. Thistechnology was also used to show the dense aggregation of Paracocci in a den-itrifying biofilm [99], as well as locations of Nitrosomonas and Nitrobacter ina nitrifying biofilm [100]. With careful design of probes and their fluorescent


labeling, the distribution of at least seven different types of microorganismscan be shown on one slide at the same time [101].

2) Following FISH labeling, target cells can be counted and concentrations can be expressed in terms of cell numbers. Flow cytometry in combinationwith FISH can also be used to count a large number of cells in a short time[10].

3) Since the number of ribosomes in each cell is assumed proportional to itsgrowth rate, quantification of the signal from each cell may be used to inferits growth rate.This approach was used to detect the in situ growth rate of sul-fate-reducing bacteria in a biofilm [102]. The results should, however, beinterpreted with caution, since it is not known if the relation between theribosome number and the growth rate is similar for cells under different con-ditions [6]. The sensitivity of FISH can be very high, 1 in 106 cells (comparedto 0.1% in membrane hybridization and 1 in 103 for cloning).

4) Probe specificity can be controlled even for very complicated communities byusing multiple probes with different color labels that target the same organ-ism at different sites in their 16S or 23S rRNA sequences.

2.3.4.1Hybridization Stringency

In contrast to membrane hybridizations, FISH is usually conducted at a relative-ly high stringency, i.e., a stringency that can differentiate target from non-targetcells. The wash step is merely used to remove excess probes. Although tempera-ture could be used as the parameter for controlling stringency [103], salt or for-mamide concentrations are more often used [99, 104]. This is more convenientsince only one temperature and hence, one oven is needed for hybridizationreactions at different stringencies.

The optimal stringency for FISH is determined in a similar way as a Td study.Target cells (perfect match) as well as non-target cells with a few mismatchesshould be used for stringency tests. Hybridizations are conducted at a numberof stringencies. The average hybridization intensities obtained for a large num-ber of cells (e.g., 100 to 200 cells) are plotted versus the stringencies in terms offormamide or salt concentration. The optimal hybridization stringency is deter-mined as the point where the signals from non-target cells are low while thosefrom target cells are still strong (see [104] and [99] for examples). Using thesame hybridization and wash solutions employed for membrane hybridization,Amann and coworkers carried out a Td study with whole cells using a 32P-labeledprobe [105]. It was shown that there was no significant difference in Td valuesbetween the two methods. Empirical equations are available for the conversionof stringencies expressed in terms of temperature, salt concentration, and for-mamide concentration [76]. These, again, can be used to predict the necessarystringency for whole cell hybridization, but are not a substitute for experimen-tal examinations.


2.3.4.2Cell Fixation

The first step of FISH is fixation, which permeates the cell wall and cell mem-brane to allow the penetration of probes. Insufficient fixation can be a cause oflow signal [6]. It is recommended to check the permeability of the cell wallsbefore further studies involving hybridization with well-labeled universal ordomain-specific probes are carried out [6].

Fixatives can be ethanol or methanol, aldehydes, enzymes, or heat [6]. In gen-eral, paraformaldehyde (PFA) and 50% ethanol offer good results for bothGram-positive and Gram-negative cells for fluorescently labeled oligonucleotideprobes [106]. PFA is often preferred for the fixation of Gram-negative cells, sincePFA might cross-link the thick Gram-positive cell wall to such an extent that theprobe cannot pass through the wall. The 50% ethanol fixation is less used forGram-negative cells, since the stability of ethanol-fixed Gram-negative cells islow. Heating or ethanol-formalin (v/v = 9:1) fixation also have proven to be suit-able for Gram-positive cells [107]. Some of the Gram-positive cells may need lyt-ic enzymes, hydrophobic solvents (toluene or diethyl ether), or acids for properfixation [82]. PFA solution (4%) has proven to be a good fixative for most of theArchaea tested in a study by Burggraf et al. [108]. However, 4% PFA was too mildin some cases due to the rigid cell wall of Methanopyrus kandleri, Methanother-mus fervidus, Methanobacterium thermoautotrophicum, and Halococcus mor-rhuae [108]. Sørensen et al. [109] found that 4% PFA fixation of Methanosarcinamazeii resulted in disruption of the cells. They demonstrated satisfying fixationresults by washing the M. mazeii cells in saline-formaldehyde (1.6% formalde-hyde and 0.85% NaCl).

Fixation time also plays an important role. In a study by de los Reyes et al.[104], one minute fixation in 4% PFA is optimal for mycolic-acid-containingGordona amarae, Rhodococcus rhodochrous, and Mycobacterium semegmatis.Longer fixation caused excessive cross-linking of the proteins in the cell wallpreventing the access of probes.

A number of fixation methods, including PFA fixation, ethanol/formaldehydefixation, solvent extraction using chloroform/methanol, acid methanolysis, andacid hydrolysis, were evaluated for mycolic-acid-containing actinomycetes andsome other Gram-positive and Gram-negative cells [110]. It was demonstratedthat the optimum fixation was species-dependent. The wide variety of cell walltypes complicated a proper fixation of all cells in a complex community by a single treatment method. It was suggested that different fixation methodsshould be used corresponding to the cell wall properties of the populations to be detected [106].

2.3.4.3Signal Enhancement

Besides poor fixation, low signals can be the result of non-complementarity be-tween the probe and the target, ineffective probe labeling, non-optimal hybridiza-tion conditions, low ribosome numbers, or low accessibility of the target site.


Several methods can be used to enhance the signal from cells having a low ribosome content, and thus increase the sensitivity of detection [see below].It should be noted that none of these methods can improve the signal quanti-tatively.

The sensitivity of detection can alternatively be improved by using instru-ments such as CCCD (cooled charged-coupled device) cameras, which candetect very low levels of emitted light, and confocal laser scanning microscopy(CLSM), which can exclude out-of-focus fluorescence.

2.3.4.3.1Indirect Assays

In this technique the oligonucleotide probe is labeled with digoxigenin. Afterhybridization, the digoxigenin is detected by a binding protein labeled with afluorescent dye or an enzyme. The signal increase by fluorescently labeled bind-ing proteins is limited, since the molar fluorescent dye/protein ratio cannotexceed 3. Enzyme-labeled binding protein has the advantage that hybridizationis detected by the precipitation of suitable substrates, eliminating interferencefrom background fluorescence or autofluorescence. The major problem withindirect assays is the limited permeability of the cell wall to relatively large flu-orescent dyes or enzyme-labeled binding proteins. Usually, enzymatic or morerigid fixation methods are needed, even for Gram-negative cells [111, 112].

2.3.4.3.2Enzyme-Labeled Oligonucleotides

Enzyme-labeled oligonucleotides are formed by covalent linking of the oligonu-cleotide probe to an enzyme. After hybridization, the probe is detected by theprecipitate formed from suitable substrates. Similar to the indirect assay, auto-fluorescence and background fluorescence do not interfere with this method.Animprovement compared to indirect assays is the lack of problems with non-spe-cific binding of the binding proteins. A study by Urdea et al. (1988) showed thatenzyme-labeled oligonucleotide probes had lower detection limits than fluores-cently labeled probes and that their sensitivity was comparable to 32P-labeledprobes [113]. Since the probe is smaller compared to binding proteins of indi-rect assays, this method can be used for most of the Gram-negative cells andsome Gram-positive, ethanol-fixed cells. Lysozyme treatment of Gram-negativecells is also suitable for this method [114], while SDS might be used for treat-ment of Archaea [114]. In a study of the bacterial community in the gut of anearthworm Fischer et al. (1995) used fluorescence-, peroxidase- (enzyme-labeled), and digoxigenin- (indirect assay) labeled probes [115]. The authorsshowed that the peroxidase- and digoxigenin-labeled probes were limited by therequirement of enzymatic fixation, diffuse images of stained cells, and interfer-ence with DAPI.


2.3.4.3.3Multi-Probe and Multi-Labeling

In general, the multi-probe method using several monolabeled fluorescentprobes that target the same cell can increase the hybridization signal [103, 105].However, limitation of specific target sites in a cell restrict the signal increase ofthis method. Alternatively, a probe can be labeled with several fluorescent dyes.But for unknown reasons, this method does not result in a significant improve-ment of signal intensity [6].

2.3.4.3.4Amplification of the Target Sequence

Amplification of the target sequence before detecting has been shown effectivein detection of weak hybridization signals [27]. In situ PCR with a fluorescentprimer was used to localize the single prokaryotic cells in a complex communi-ty [116]. Non-specific amplification, however, might be a potential problemassociated with this method.

2.3.5Solution-Based Hybridizations (Molecular Beacons)

When performing membrane hybridization or FISH, it is necessary to removeexcess probe from the hybridization mixture before the detection step. Thisrequires immobilization of cells or nucleic acids on solid surfaces (slides ormembranes), which may lower the sensitivity of hybridization due to non-spe-cific binding of probes to the surface [117]. It also prohibits real-time monitor-ing of nucleic acid synthesis and location of specific nucleic acids in living cells[117]. Furthermore, these hybridization methods are labor intensive and cannotbe automated. The use of solution hybridization techniques, in contrast, offersadvantages such as fast kinetics and the suitability for automatic analysis [74].Molecular beacon techniques eliminate the need of removing excess probe afterhybridization, and thus provide the feasibility for quick and automatedhybridization.

A molecular beacon is a probe that contains a stem-and-loop structure(Fig. 3) [117]. The loop part consists of a sequence that is complementary to thetarget sequence, while the stem part consists of two short sequences (arms)located at each end of the probe and complementary to each other. One of thearms is end-labeled with a fluorescent dye, while the end of the other arm con-tains a quencher. The quencher is selected so that its absorption spectrum over-laps the emission band of the fluorophore.When the molecular beacon is closed,the fluorescent dye and the quencher are held closely together by the stem. As aresult, no fluorescence is emitted. When the molecular beacon hybridizes to atarget, or when the temperature is higher than the Tm of the stem, the fluorescentdye is spatially removed from the quencher and the fluorophore emits lightupon exitation (Fig. 3).


Tyagi and Kramer [117] showed that the hybridization was very fast and thatthe reaction rate increased as a function of the beacon concentration, the targetconcentration, the salt concentration, and the temperature. The molecular bea-con technique was found to be very specific when oligonucleotides were used asa target. Almost no hybridization was observed to targets with one mismatch orone deletion. Molecular beacons, therefore, can be used for detection of livingcells or for real-time monitoring of PCR reactions. In addition, it is possible todetect a number of different populations in the same reaction tube if molecularbeacons are labeled with different types of fluorescent dyes. In a comparativestudy of membrane hybridizations with oligonucleotide probes and solutionhybridizations with molecular beacons, Schonfield et al. [118] obtained similarresults for the detection of 16S rRNA. However, several problems were encoun-tered by the authors. Firstly, the specificity of molecular beacons was not as highas reported by Tyagi and Kramer [117]. Only an 80% signal decrease wasobserved when one mismatch occurred in the target sequence. Secondly, denat-uration had to be carried out carefully. Chemical denaturants may not be applicable, since they denature the molecular beacons if they are not removedfrom the solution before hybridization. In addition, denaturants cause highbackground fluorescence interfering with the detection of molecular beacons.Schonfield et al. denatured the RNA by heating (95°C for 5 min) followed byovernight hybridization at 39°C. Thirdly, the post-hybridization mixture need-ed to be centrifuged to remove particles that otherwise interfered with thedetection of the molecular beacons. Nevertheless, this study showed that themolecular beacon technique reduced the total time needed for analysis of a sample from 3–4 days to 12 hours, and it was possible to use the technique fordetecting 16S rRNA in environmental samples.


Fig. 3. Concept of the molecular beacon technique. Modified from [117, 118]

2.4FISH and Reporter Systems

Direct hybridization methods (e.g., FISH) are only possible when the abundanceof target nucleic acids is sufficiently high. Detecting genomic DNA or mRNAusing in situ hybridizations is therefore difficult or impossible since the amountof target is too low. The amount of target can be increased by in situ RT-PCRapplied to whole, permeabilized cells. By this technique, it is possible to addressthe questions regarding gene expression under different growth conditions andthe influence of metabolites and chemicals on the different pathways at the singlecell level. This approach is very laborious when compared to direct studies withwell known model systems like Echerichia coli, Saccharomyces cerevisiae, andBacillus subtilis, but at the current state of technology it is the only possibility.

To our knowledge the only report describing gene expression in anaerobic reac-tor systems on the mRNA level is by Lange et al. [119]. This paper describes theexpression of the heat shock gene dnaK in Methanosarcina mazei S-6 under dif-ferent stress situations in situ. Paraformaldehyde-fixed cells were permeabilizedby lysozyme and heat treatment, and the cellular DNA was removed by DNAsetreatment.By means of a semi-nested RT-PCR protocol,DIG-labeled primers wereused to amplify the dnaK gene product. Detection of the dnaK reporter moleculewas based on the HPPN/Fast Red detection system and the binding of anti-digoxigenin-AP conjugate. Key parameters in this technique are the permeabi-lization of the cells and the number of cycles in the RT-PCR step. Analyses of dif-ferent ecosystems by this technique, therefore, have to be carefully and individual-ly optimized as especially Gram-positive bacteria and many archaeal species havea cell wall that is difficult to permeabilize.As opposed to traditional mRNA analy-sis techniques, e.g., Northern blotting, the in situ RT-PCR technique may revealheterogeneous gene expression in microbial populations [120], providing a moredetailed picture of the physiological state of populations.

A few reporter systems have been developed in organisms present in anaero-bic reactor systems. In the methanogenic archaeon Methanococcus maripaludis,genetic manipulations are now possible, as a naturally occurring plasmid inMethanococcus has been modified to include the genes encoding the puromycinresistance marker and the reporter gene lacZ encoding a b-galactosidase [105,121, 122]. A uidA b-glucuronidase reporter gene has been used in Methanococ-cus voltae [123]. By fusing the nifH promotor region to the lacZ coding region,mutational analysis showed the presence of a regulatory palindromic sequencerepressing the nifH gene expression.

In the genus Methanosarcina, a plasmid from Methanosarcina acetivoranswas reported to be able to replicate in 9 of 11 Methanosarcina strains with hightransformation efficiency [124]. Lange and Ahring have utilized the plasmidfound by Metcalf to construct a reporter system based on the heat shock pro-motors of dnaK and grpE in Methanosarcina mazei S-6 and the lacZ markergene. The resistance marker was the puromycin cassette. This system showed anice correlation between the amount of stress and the activity of b-galactosi-dase. However, after several transfers of the transformants, the presence of thepromotor-lacZ cassette could no longer be confirmed due to the size of the plas-


mid (12 kb), but the strains retained their resistance towards the original level ofpuromycin [125]. These findings should encourage new investigations to con-struct smaller plasmids to allow routine genetic manipulations in the genusMethanosarcina. Also, other resistance markers are needed as the puromycinmarker is one of a few shown to work in methanogens [126–128].

In the extremely halophilic archaeon Haloferax alicantei, a b-galactosidaseprotein was purified and the sequence of the N-terminal part of the protein wasdetermined which facilitated the cloning of the gene [108]. b-Galactosidasesfrom other organisms do not function in the halophiles due to salt concentra-tions above 4 M NaCl. Since plasmids already have been found in the halophilicArchaea, in vivo studies of gene expression in this group of organisms is nowpossible [127].

Future gene candidates to be used as reporter genes may be the luxA and theluxB gene system; these genes have been expressed in the anaerobe Clostridiumperfringens under the transcriptional control of the a-toxin promotor regionunder anoxic conditions [129]. However, this gene system has only limited rele-vance in anaerobic prokaryotes, as the lux protein complex needs oxygen to pro-duce light. The green fluorescent protein (GFP) also requires oxygen to matureinto a functional fluorescent protein [130]. Errampalli et al. [131] have recentlyreviewed the applications of the GFP protein as a molecular marker in environ-mental studies, so we refer to this paper for further reading. The GFP may bevaluable as a reporter gene in anaerobic laboratory systems if organisms trans-formed with the GFP gene can be detected by fluorescence when subsequentlyexposed to oxygen [130]. It will, therefore, not be possible to detect the GFPreporter system in situ, but ex situ as a tool to monitor gene expression and tocharacterize promotor regions. Further work has to be done on the applicabili-ty of the GFP in anoxic and anaerobic systems.

An unusual blue protein Ambineela, has been isolated from the archaeonAcidianus ambivalens [132]. Spectrum analyses of this protein show two emis-sion bands around 395 nm and 625 nm. The protein does not require any tran-sition metals in order to display this blue color. Ambineela might be a possiblereporter gene candidate in archaeal species for flow cytometric measurementsor absorbance measurements in axenic cultures.

2.5FISH and Antibody Probes

Molecular studies of anaerobic reactor systems started using antibodies raisedagainst different microbial groups and labeled with fluorochromes [133–139].The technique was used in phylogenetic studies to construct 2D pictures of thespatial distribution of different groups of microorganisms in granular sludgeand flocs. Several drawbacks are, however, associated with the production anduse of antibodies in microbial ecology studies [10]. In order to raise the desiredantibodies, axenic cultures are required to induce antibody production by theanimals implying that only known organisms can be targeted. The antibodiesare less specific than oligonucleotide probes and antibodies raised against oneorganism may target different epitopes on the cell leading to putative cross-reac-


tivity to other non-target cells. Furthermore, probing with antibodies does notdiscriminate between dead and living cells.

An advantage of antibody probing is that cells having different metabolicactivities are equally visualized. The difference in metabolic activity, on the oth-er hand, is one of the problems of oligonucleotide probing when targeting the16S rRNA molecule. The cells of a population do not contain an equal amount ofribosomes [27, 119, 120] and as the oligonucleotides normally are quite small,only a certain amount of fluorochrome can be targeted to each cell. If the ribo-some content in the cell, therefore, is low, the fluorescent signal cannot be dis-tinguished from the background fluorescence.

To circumvent this problem, antibody probing can be combined with fluores-cent oligonucleotide probes targeting the 16S rRNA molecule. The antibodyprobe and the oligonucleotide probe can be labeled with different colors and byapplying different filters when examining the samples in the microscope, onlycells labeled by both probes are visible.

Raskin and coworkers have successfully applied both antibody and oligonu-cleotide probes in situ for the characterization of Gordona species in activatedsludge and anaerobic digesters [140, 141]. Gordona species are slow growing fil-amentous bacteria commonly found in activated sludge foam causing opera-tional problems in wastewater treatment plants. Quantification of Gordonaspecies in activated sludge samples and in anaerobic digester samples by fluo-rescent antibodies showed that Gordona species biomass accounted for 10–28%of the volatile suspended solids in a full-scale activated sludge system and for8–19% in anaerobic digester sludge. Quantification using traditional Gramstaining and filament counting for the same activated sludge sample resulted insignificantly lower values of 2–10% [142].

Analyzing seven full-scale wastewater treatment plants with antibody probesor FISH targeting Gordona species showed that the antibody probing techniqueled to estimates of a higher number of Gordona species than when FISH wasused [141]. Simultaneous staining with antibody probes and FISH showed thatsome branched filaments were clearly stained by the antibodies, but gave a lowsignal when using the FISH probe. The ribosome content of the cells, therefore,indicated that individual cells differed in metabolic activity. These findingspoint to the difficulties when estimating the number of species and the meta-bolic state of the species in an environmental sample. Quantifying the numberof Gordona species by membrane hybridization targeting 16S rRNA and FISHtargeting 16S rRNA, expressed as biomass by measuring cell length and calcu-lating the cell volume, gave different results. As observed in other studies [143],relative cell numbers, therefore, cannot be reliably extrapolated from rRNAabundance levels.

Although both immunostaining and FISH might yield non-specific signalsfrom non-target microorganisms, the possibility of misidentifying the samenon-target microorganisms by two independent methods should be low. Thesimultaneous application of a polyclonal antibody serum and an oligonu-cleotide probe targeting 16S rRNA also offers the advantage of detecting slowgrowing or metabolically less active microorganisms while maintaining highphylogenetic specificity in complex environments [140].


2.6FISH and Microautoradiography

To identify and to evaluate the metabolic activity of a single cell simultaneously,microautoradiography (MAR) has successfully been used in combination withFISH. MAR is used to study the microscale distribution of radiolabeled com-pounds. The radiolabeled compound is taken up by either active or passivetransport across the cell membrane. After appropriate treatment, the radiola-beled cell sample is placed in contact with a layer of radiosensitive emulsion.Theemulsion is then developed after days or weeks of exposure to the radioisotopeby means of standard photographic procedures. Subsequent microscopy of thecell sample will show cells covered with silver grains if the radiolabeled com-pound has been taken up or metabolized. Cells on the developed slide can thenbe investigated further by FISH to correlate the phylogenetic affiliation andmetabolic activity.

Nielsen et al. and Ouverney et al. have combined FISH and MAR protocols to correlate cell identity and function in activated sludge reactors [144–147].The uptake of 14C- or 3H-labeled substrates (acetate, glucose, ethanol, glycine,leucine, and oleic acid) in bulking sludge was investigated in seven industrial ormunicipal activated sludge treatment plants. The authors concluded that thetaxonomic variability assessed using FISH with probes specific for type 021 N,Thiothrix and Leucothrix is high since only some of the filaments morphologi-cally identified as type 021 N hybridized with the 021 N probe. Moreover, no fil-aments took up all the tested substrates, and type 021 N from the various treat-ment plants varied in their uptake capabilities. The study demonstrates thatstrain differences with regard to substrate utilization are likely to occur amongbacteria within the same genus.

Some parameters to obtain good results with the MAR technique are:

– The choice of radioactive tracer: image resolution will decrease if high ener-gy isotopes are used.

– The selection of incubation conditions: amount of biomass and isotope, ratioof “hot” to “cold” substrate, presence of electron acceptors and incubationtime.

– The sample handling and fixation: the sample must be washed thoroughly toremove excess isotopes.

– Other staining procedures: various cell stains such as Gram and Neisser stain-ing for identification of filamentous microorganisms and fluorescent non-specific stains such as Acridine orange or DAPI targeting DNA may be com-bined with MAR. The dye CTC may be used to discriminate between viableand dead cells.

The limitation of the MAR technique is low resolution unless cryosectioning ofthe sample is conducted and 3D distribution of non-fluorescent silver grains inthe emulsion is accurately analyzed by confocal laser scanning microscopy.


2.7Peptide Nucleic Acid Probes

Peptide nucleic acids (PNA) are artificial oligoamides reported for the first timeby Nielsen et al. [148]. PNA are capable of forming not only double-strandedduplexes, but also triple-stranded complexes with poly- or oligonucleotides.Due to their neutral backbone, PNA can hybridize to nucleic acids in the absenceof the counterions needed to stabilize nucleic acid duplexes. Thus, PNA probesexhibit superior hybridization characteristics compared to DNA probes underconditions where double-stranded DNA is electrostatically unstable, e.g., at lowsalt conditions. PNA probes have been used for a wide range of applications andthe advantages of PNA over DNA are numerous: rapid hybridization, hybridiza-tion independent of the salt concentration, resistant to nuclease and proteaseattack, more specific and shorter probes can be used for higher sensitivity [149].

Prescott and Fricker [149] have used PNA oligonucleotide probes for in situdetection of Eccherichia coli in water. They targeted the PNA biotinylated probestowards the V1 region of the E. coli 16S rRNA molecule and the specific detec-tion was carried out in less than 3 hours. The specificity of the PNA probeagainst E. coli was confirmed by comparative dot-blot analyses using the generaKlebsiella, Enterobacter, and Citrobacter.

Work done by Perry-O’Keefe et al. [150] has shown the superior hybridizationconditions under which PNA probes can detect DNA target sequences comparedto DNA probes. Investigations of parameters like hybridization rate and salt concentrations showed that PNA probes hybridized much faster under a widerange of salt concentrations. Because the PNA molecule is not charged, pre-gelhybridization can be performed with subsequent transfer to nylon membraneand fast detection.

Another new method of detection of nucleic acids is the application of PNAprobes together with the BIAcore biosensor system. The principle behind thissystem is the detection of PNA hybridizing to DNA through signaling to a sur-face plasmon resonance unit. Specific hybridization occurs in 10 min within aflow stream at room temperature [151].

This fast and reproducible technique is promising for the detection and quan-tification of rRNA extracted from environmental samples, as it is much fasterthan the traditional techniques employed today. The problem with the sec-ondary structure of the rRNA may be overcome by an addition of formamidewhich will inhibit double strand and hydrogen bond formation, thereby reduc-ing the secondary structure of the rRNA but still allowing the PNA probe to bindto its target sequence.


3rRNA-Based Analyses of Anaerobic Reactors

3.1Biofilm Reactors

3.1.1Biofilm Formation

Anaerobic biofilm development was monitored by Araujo et al. [152]. Theyinvestigated biofilm formation with respect to the methanogens and the prop-erties of biofilm formation on hydrophilic and hydrophobic surfaces. FISH andconfocal laser microscopy were used to quantify the microbial composition andto elucidate the spatial distribution of microbes in the biofilm. Table 1 outlinesthe probes used in this study.

In this experiment, axenic cultures of Methanobacterium formicicum, Me-thanosaeta concilli and Methanosarcina barkeri were grown together for amonth before introduction into a sterile, anaerobic chemostat connected to amodified Robbins device (MRD) [153]. The system was operated at 30°C ± 3°Cand the biofilm was grown on either polypropylene or glass discs inside theMRD. The carbon sources were methanol, formate and acetate. Samples waretaken on days 0, 2, 7, and 9. Even though the setup was supposed only to analyzefor methanogens, contaminants were observed when samples were taken. Dur-ing the experiment, the chemostat microbial community changed from 100%Archaea at day 0 to contain 40% Bacteria within 2 days.After 7 days the Bacteriaconstituted 80% of the total microbial community. The number of Methano-sarcina barkeri and Methanosaeta concilii decreased from about 15% day 0 toclose to 0% day 9. The number of Methanobacterium formicicum cells wasreduced from 80% to 10% within 9 days. The findings were not reflected in the composition of the biofilm growing on the discs. After 7 days, the bacterialpopulation had only increased to about 15%. Methanobacterium formicicumpredominated in all the biofilms ranging from 40% to 80% after 9 days.Methanosaeta concilii cells were found at low numbers up to 2.4% whileMethanosarcina barkeri was not detected. No difference in biofilm colonizationwas observed between the polypropylene and the glass discs.

In a second experiment crushed granular sludge was used as seed in thechemostat and the biofilm in the MRD was grown on polypropylene discs at33°C ± 2°C. Initially, the inoculum in the chemostat consisted of 13% sulfate-reducing bacteria (SRB) and 93% Archaea mostly Methanobacterium species.Methanosarcina species accounted for 5%.After eleven days the SRB populationhad decreased to 1.7% although the bacterial population still constituted 13%.Since no sulfate was present, the SRB population probably functioned as protonreducers in syntrophic association with hydrogen and formate-consumingmethanogens. From day zero to day eleven Methanobacteriaceae increased from74% to 85% while Methanosarcinales decreased from 11% to 4%. Microscopicobservations of the biofilm by confocal laser microscopy showed areas where no growth occurred, and areas up to 9 µm where growth occurred. This was



Tabl

e1.

Vari

ous

16S

rRN

A p

robe

s us

ed in

env

iron

men

tal s

tudi

es

Rea

ctor

type

TC

arbo

n so

urce

sPr

obe

tech

niqu

ePr

obes

use

dTa

rget

org

anis

ms

Ref

eren

ce

Fixe

d-be

d M

Glu

cose

FISH

S-D

-Bac

t-03

38-a

-R-1

8V

irtu

ally

all

bact

eria

Am

ann

et a

l.19

99an

aero

bic

S-*-

Srb-

0385

-a-R

-18

All

sulfa

te-r

educ

ing

bact

eria

bior

eact

orS-

Ss-P

t1–6

47-a

-R-1

9U

nkno

wn

Des

ulfu

rom

onas

sp.

S-Ss

-Pt2

–647

-a-R

-19

Unk

now

n D

esul

fovi

brio

vul

gari

ssp.

Fixe

d-be

d M

Lact

ate

FISH

S-Ss

-Pt2

–647

-a-R

-19

Unk

now

n D

esul

fovi

brio

vul

gari

ssp.

Kan

e et

al.

1993

anae

robi

c D

ot b

lot

bior

eact

or

Mod

ified

M

Met

hano

l,FI

SH,C

LSM

S-D

-Bac

t-03

38-a

-R-1

8V

irtu

ally

all

bact

eria

Ara

ujo

et a

l.20

00R

obbi

ns

form

ate,

acet

ate

devi

ce r

eact

orS-

D-A

rch-

0915

-a-R

-20

Vir

tual

ly a

ll A

rcha

eaA

ceta

te,b

utyr

ate,

FISH

,CLS

MS-

*-Sr

b-03

85-a

-R-1

8A

ll su

lfate

-red

ucin

g ba

cter

iafo

rmat

e +

Etha

nol a

nd

S-F-

Mco

c-11

09-a

-R-2

0M

etha

noco

cace

aem

etha

nol

S-F-

Mba

c-11

74-a

-R-2

2M

etha

noba

cter

iace

aeS-

O-M

mic

-120

0-a-

R-21

Met

hano

mic

robi

ales

S-O

-Msa

rc-8

60-a

-R-2

1M

etha

nosa

rcin

ales

Fixe

d-be

d M

Glu

cose

FISH

S-*-

Uni

v-13

92-a

-R-1

5V

irtu

ally

all

orga

nism

sR

aski

n et

al.

1995

anae

robi

c D

ot b

lot

S-D

-Bac

t-03

38-a

-R-1

8V

irtu

ally

all

Bact

eria

Ras

kin

et a

l.19

96bi

orea

ctor

S-D

-Arc

h-09

15-a

-R-2

0V

irtu

ally

all

Arc

haea

S-D

-Arc

h-34

4-a-

R-20

Vir

tual

ly a

ll A

rcha

eaS-

D-E

uca-

0502

-a-R

-16

Vir

tual

ly a

ll Eu

cary

aS-

F-M

coc-

1109

-a-R

-20

Met

hano

coca

ceae

S-F-

Mba

c-03

10-a

-R-2

2M

etha

noba

cter

iace

aeS-

O-M

mic

-120

0-a-

R-21

Met

hano

mic

robi

ales


Tabl

e1

(con

tinu

ed)

Rea

ctor

type

TC

arbo

n so

urce

sPr

obe

tech

niqu

ePr

obes

use

dTa

rget

org

anis

ms

Ref

eren

ce

S-O

-Msa

rc-8

60-a

-R-2

1M

etha

nosa

rcin

ales

S-F-

Msa

r-08

21-a

-R-2

4M

etha

nosa

rcin

aS-

F-M

sae-

0825

-a-R

-23

Met

hano

saet

acea

eS-

F-D

sv-0

687-

a-R-

16D

esul

fovi

brio

nace

aeS-

G-D

sbm

-022

1-a-

R-20

Des

ulfo

bact

eriu

mS-

G-D

sb-0

129-

a-R-

18D

esul

foba

cter

S-*-

Dsc

oc-0

814-

a-R-

18D

esul

foco

ccus

+D

esul

fosa

rcin

a+

Des

ulfo

botu

lus

S-G

-Dsb

b-06

60-a

-R-2

0D

esul

fobu

lbus

Ana

erob

ic

MG

luco

seD

ot b

lot

S-U

-Uni

v-13

92-a

-R-1

5V

irtu

ally

all

orga

nism

sR

aski

n et

al.

1994

chem

osta

t

Solid

was

te

M,T

Mun

icip

al w

aste

FISH

S-D

-Bac

t-03

38-a

-R-1

8V

irtu

ally

all

bact

eria

dige

ster

Sew

age

slud

ge

MSe

wag

e sl

udge

S-D

-Arc

h-09

15-a

-R-2

0V

irtu

ally

all

Arc

haea

dige

ster

S-D

-Euc

a-05

16-a

-R-1

6V

irtu

ally

all

Euca

rya

S-F-

Mco

c-11

09-a

-R-2

0M

etha

noco

cale

sS-

F-M

bac-

0310

-a-R

-22

Met

hano

bact

eria

les

S-O

-Mm

ic-1

200-

a-R-

21M

etha

nom

icro

bial

esS-

O-M

sarc

-860

-a-R

-21

Met

hano

sarc

inal

esS-

F-M

sar-

821-

a-R-

24M

etha

nosa

rcin

acea

eS-

G-M

sar-

0821

-a-R

-21

Met

hano

sarc

ina

sp.

S-F-

Msa

e-08

25-a

-R-2

3M

etha

nosa

eta

One

-pha

se,

M,T

Sew

age

slud

geD

ot b

lot

S-*-

Uni

v-13

90-a

-R-1

8V

irtu

ally

all

orga

nism

sR

aski

n et

al.

1995

two-

phas

e


Sew

age

slud

ge

S-D

-Bac

t-03

38-a

-R-1

8V

irtu

ally

all

bact

eria

dige

ster

sS-

D-A

rch-

0915

-a-R

-20

Vir

tual

ly a

ll A

rcha

eaS-

D-A

rch-

344-

a-R-

20V

irtu

ally

all

Arc

haea

S-D

-Euc

a-05

02-a

-R-1

6V

irtu

ally

all

Euca

rya

S-F-

Mco

c-11

09-a

-R-2

0M

etha

noco

cace

aeS-

F-M

bac-

0310

-a-R

-22

Met

hano

bact

eria

ceae

S-O

-Mm

ic-1

200-

a-R-

21M

etha

nom

icro

bial

esS-

O-M

sarc

-860

-a-R

-21

Met

hano

sarc

inal

esS-

F-M

sar1

414-

a-R-

21M

etha

nosa

rcin

acea

eS-

G-M

sar-

0821

-a-R

-21

Met

hano

sarc

ina

sp.

S-F-

Msa

e-08

25-a

-R-2

3M

etha

nosa

etac

eae

S-G

-Dsb

-080

4-a-

R-18

Des

ulfo

bact

ergr

oup

S-F-

Dsv

-068

7-a-

R-16

Des

ulfo

vibr

iona

ceae

S-G

-Dsb

m-0

221-

a-R-

20D

esul

foba

cter

ium

spp.

S-G

-Dsb

-012

9-a-

R-18

Des

ulfo

bact

ersp

p.S-

*-D

scoc

-081

4-a-

R-18

Des

ulfo

cocc

usm

ulti

vora

nsD

esul

fosa

rcin

ava

riab

ilis

Des

ulfo

botu

luss

apov

oran

sS-

G-D

sbb-

0660

-a-R

-20

Des

ulfo

bulb

ussp

p.

CST

R d

iges

ter:

MC

ow m

anur

eD

ot b

lot

S-F-

Synm

-070

0-a-

R-23

Stra

in F

SS7

Han

sen

et a

l.19

99Bi

ogas

pla

nt+

lipid

sSt

rain

FM

S2Sy

ntro

phos

pora

bry

anti

iSy

ntro

phom

onas

sapo

vora

nsSy

ntro

phom

onas

wol

feis

ubsp

.wol

fei

Synt

roph

omon

as w

olfe

isub

sp.L

YB

S-G

-Syn

m-0

126-

a-R-

19Sy

ntro

phom

onas

sapo

vora

nsSy

ntro

phom

onas

wol

feis

ubsp

.wol

fei

Synt

roph

omon

as w

olfe

isub

sp.L

YB

S-S-

S-w

ol-0

180-

a-R-

21Sy

ntro

phom

onas

wol

feis

ubsp

.wol

fei

Synt

roph

omon

as w

olfe

isub

sp.L

YB

S-S-

S.sa

p-01

81-a

-R-2

0Sy

ntro

phom

onas

sapo

vora

nsS-

S-S.

bry-

0181

-a-R

-21

Synt

roph

ospo

ra b

ryan

tii


Tabl

e1

(con

tinu

ed)

Rea

ctor

type

TC

arbo

n so

urce

sPr

obe

tech

niqu

ePr

obes

use

dTa

rget

org

anis

ms

Ref

eren

ce

CST

R r

eact

orM

Glu

cose

Dot

blo

tS-

G-M

sar-

0821

-a-R

-24

Met

hano

sarc

ina

Fern

ande

z et

al.

S-G

-Msa

e-03

81-a

-R-2

2M

etha

nosa

eta

2000

S-O

-Mm

ic-1

200-

a-R-

21M

etha

nom

icro

bial

esS-

F-M

bac-

0310

-a-R

-22

Met

hano

bact

eria

les

Tem

p.ph

ase

TSe

wag

e sl

udge

Dot

blo

tS-

G-D

tm-0

229-

a-R-

18D

esul

foto

mac

ulum

Hri

stov

a et

al.

dige

ster

2000

Mun

icip

al s

olid

w

aste

CST

R d

iges

ter

TM

unic

ipal

S-

*-D

tm(a

)-02

29a-

R-18

Sele

cted

Des

ulfo

tom

acul

umsp

.w

aste

wat

erPa

per

mill

S-

*-D

tm(b

e)-0

152-

a-R-

20Se

lect

ed D

esul

foto

mac

ulum

sp.

was

tew

ater

S-*-

Dtm

(c)-

0428

-a-R

-19

Sele

cted

Des

ulfo

tom

acul

umsp

.S-

*-D

tm(c

d)-0

216-

a-R-

19Se

lect

ed D

esul

foto

mac

ulum

sp.

S-*-

Dtm

(bcd

)-23

0-a-

R-18

Sele

cted

Des

ulfo

tom

acul

umsp

.

Lab-

scal

e M

,TM

unic

ipal

D

ot b

lot

S-U

-Uni

v-13

90-a

-R-1

8V

irtu

ally

all

orga

nism

sZh

eng

and

Ras

kin,

dige

ster

was

te20

00

Tem

p.ph

ased

T

Act

ivat

ed s

ludg

eS-

D-A

rch-

0915

-a-R

-20

Vir

tual

ly a

ll A

rcha

eadi

gest

erS-

F-M

coc-

1109

-a-R

-20

Met

hano

coca

ceae

S-F-

Mba

c-03

10-a

-R-2

2M

etha

noba

cter

iace

aeS-

O-M

mic

-120

0-a-

R-21

Met

hano

mic

robi

ales

S-O

-Msa

rc-8

60-a

-R-2

1M

etha

nosa

rcin

ales

S-F-

Msa

r141

4-a-

R-21

Met

hano

sarc

inac

eae

S-G

-Msa

r-08

21-a

-R-2

1M

etha

nosa

rcin

asp

.S-

F-M

sae-

0825

-a-R

-23

Met

hano

saet

acea

eS-

G-M

sae-

0733

-a-R

-22

Met

hano

saet

a


S-G

-Msa

e-07

81-a

-R-2

2M

etha

nosa

eta

S-G

-Msa

e-03

32-a

-R-2

2M

etha

nosa

eta

S-G

-Msa

e-03

22-p

-R-2

2M

etha

nosa

eta

com

peti

tive

pro

beS-

S-M

.con

-038

1-a-

R-22

Met

hano

saet

aco

ncili

i

UA

SBM

Etha

nol,

Dot

blo

tS-

D-A

rch-

0915

-a-R

-20

Vir

tual

ly a

ll A

rcha

eaZh

eng

et a

l.20

00w

aste

wat

erG

luco

seFI

SHS-

O-M

mic

-120

0-a-

R-21

Met

hano

mic

robi

ales

Glu

cose

+S-

F-M

bac-

0310

-a-R

-22

Met

hano

bact

eria

ceae

prop

iona

teS-

F-M

coc-

1109

-a-R

-20

Met

hano

coca

ceae

S-O

-Msa

rc-8

60-a

-R-2

1M

etha

nosa

rcin

ales

S-G

-Msa

r-08

21-a

-R-2

4M

etha

nosa

rcin

aS-

G-M

sae-

0332

-a-R

-22

Met

hano

saet

aS-

S-M

.con

-038

1-a-

R-22

Met

hano

saet

aco

ncili

iS-

S-M

.the-

0396

-a-R

-22

Met

hano

saet

ath

erm

ophi

la

UA

SBM

Sulfa

te-r

ich

mill

Dot

blo

tS-

D-B

act-

0338

-a-R

-18

Vir

tual

ly a

ll ba

cter

iaEl

feri

nk e

t al.

1998

pape

r w

aste

wat

erS-

D-A

rch-

0915

-a-R

-20

Vir

tual

ly a

ll A

rcha

eaS-

F-M

bac-

1174

-a-R

-22

Met

hano

bact

eria

ceae

S-F-

Mco

c-11

09-a

-R-2

0M

etha

noco

cace

aeS-

O-M

mic

-120

0-a-

R-21

Met

hano

mic

robi

ales

S-G

-Msa

r-08

21-a

-R-2

1M

etha

nosa

rcin

aS-

F-M

sae-

0825

-a-R

-23

Met

hano

saet

aS-

*-Sr

b-03

85-a

-R-1

8G

ram

-neg

ativ

e su

lfate

-red

ucin

g ba

cter

iaS-

F-D

sv-0

687-

a-R-

16D

esul

fovi

brio

nace

aeS-

G-D

sbb-

0660

-a-R

-20

Des

ulfo

bulb

usS-

G-D

sbm

-022

1-a-

R-20

Des

ulfo

bact

eriu

mS-

G-D

sb-0

804-

a-R-

18D

esul

foba

cter

S-S-

D.a

mn-

0454

-a-R

-20

Des

ulfo

habd

us a

mni

genu

sS-

*-D

scoc

-080

4-a-

R-18

Des

ulfo

cocc

us,D

esul

foba

cter

ium

Des

ulfo

sarc

ina,

Des

ulfo

bact

er,

Des

ulfo

botu

lus


Tabl

e1

(con

tinu

ed)

Rea

ctor

type

TC

arbo

n so

urce

sPr

obe

tech

niqu

ePr

obes

use

dTa

rget

org

anis

ms

Ref

eren

ce

S-*-

Sbac

-022

2-a-

R-19

Synt

roph

obac

ter

fum

arox

idan

sSy

ntro

phob

acte

r pf

enni

gii

Des

ulfo

rhab

dus a

mni

genu

sS-

S-S.

pfe-

0460

-a-R

-21

Synt

roph

obac

ter

fpen

nigi

iS-

S-S.

wol

-022

3-a-

R-19

Synt

roph

obac

ter

wol

inii

S-Ss

-SY

N7–

0177

-a-R

-23

Synt

roph

ic p

ropi

onat

e-ox

idiz

er S

YN

7S-

F-M

bac-

1174

-a-R

-22

Met

hano

bact

eria

les

S-F-

Mco

c-11

09-a

-R-2

0M

etha

noco

cale

sS-

O-M

mic

-120

0-a-

R-21

Met

hano

mic

robi

ales

S-G

-Msa

r-08

21-a

-R-2

1M

etha

nosa

rcin

aS-

F-M

sae-

0825

-a-R

-23

Met

hano

saet

a

UA

SBM

Prop

iona

teD

ot b

lot

S-*-

Uni

v-13

90-a

-R-1

8V

irtu

ally

all

orga

nism

sH

arm

sen

et a

l.+

sulfa

te19

96a

FISH

S-D

-Bac

t-03

38-a

-R-1

8V

irtu

ally

all

bact

eria

S-D

-Arc

h-09

15-a

-R-2

0V

irtu

ally

all

Arc

haea

S-S-

SYN

7–01

77-a

-R-2

3Sy

ntro

phic

pro

pion

ate-

oxid

izer

SY

N7

S-S-

S.w

ol-0

223-

a-R-

19Sy

ntro

phob

acte

r w

olin

iiS-

S-M

POB-

0222

-a-R

-19

MPO

BS-

S-S.

pfe-

0460

-a-R

-21

Synt

roph

obac

ter

fpen

nigi

iS-

F-D

sv-0

687-

a-R-

16D

esul

fovi

brio

S-G

-Dsb

b-06

60-a

-R-2

0D

esul

fobu

lbus

S-F-

Msa

e-08

25-a

-R-2

3M

etha

nosa

eta

S-O

-Mm

ic-1

200-

a-R

A-2

1M

etha

nom

icro

bial

esS-

F-M

bac-

0310

-a-R

-22

Met

hano

bact

eria

ceae


UA

SBM

Sucr

ose+

sulfa

teFI

SHS-

D-B

act-

0338

-a-R

-18

Vir

tual

ly a

ll ba

cter

iaH

arm

sen

et a

l.V

FAs+

sulfa

teS-

D-A

rch-

0915

-a-R

-20

Vir

tual

ly a

ll A

rcha

ea19

96b

S-S-

MPO

B-02

22-a

-R-1

9M

POB

S-S-

Spfe

-046

0-a-

R-21

Synt

roph

obac

ter

fpen

nigi

ialia

s K

OPR

OP

S-G

-Msa

e-08

25-a

-R-2

3M

etha

nosa

eta

S-O

-Mm

ic-1

200-

a-R-

21M

etha

nom

icro

bial

esS-

F-M

bac-

0310

-a-R

-22

Met

hano

bact

eria

ceae

UA

BSM

Pape

r m

ill

Dot

blo

tS-

D-B

act-

0338

-a-R

-18

Vir

tual

ly a

ll ba

cter

iaEl

feri

nk e

t al.

1997

was

tew

ater

S-*-

Sbac

-022

2-a-

R-19

Synt

roph

obac

ter

fum

arox

idan

sSy

ntro

phob

acte

r pf

enni

gii

Des

ulfo

rhab

dus a

mni

genu

sS-

S-D

.am

n-04

54-a

-R-2

0D

esul

foha

bdus

am

nige

nus

UA

SBM

,TA

lcoh

ol d

isti

llery

FISH

S-D

-Arc

h-09

15-a

-R-2

0V

irtu

ally

all

Arc

haea

Taga

wa

et a

l.20

00Su

cros

e+V

FAs

S-G

-Msa

e-07

57-a

-R-1

8M

etha

nosa

eta

Dai

ry m

ilkS-

F-M

bac-

1174

-a-R

-22

Met

hano

bact

eria

ceae

Food

bre

wer

yB

ean

jam

Pota

to w

aste

Mun

icip

al s

ewag

e

UA

SBT

Sucr

ose+

VFA

s FI

SHS-

D-B

act-

0338

-a-R

-18

Vir

tual

ly a

ll ba

cter

iaIm

achi

et a

l.20

00+

Yeas

t ext

ract

Dot

blo

tS-

D-A

rch-

0915

-a-R

-20

Vir

tual

ly a

ll A

rcha

eaS-

F-M

bac-

1174

-a-R

-22

Met

hano

bact

eria

ceae

S-Ss

-TG

P-06

90-a

-R-2

0St

rain

SI

UA

SBM

,TSu

cros

e,V

FAs

PCR

Seki

guch

i et a

l.19

98


Tabl

e1

(con

tinu

ed)

Rea

ctor

type

TC

arbo

n so

urce

sPr

obe

tech

niqu

ePr

obes

use

dTa

rget

org

anis

ms

Ref

eren

ce

UA

SBM

,TSu

cros

e,V

FAs

CLS

MS-

D-B

act-

0338

-a-R

-18

Vir

tual

ly a

ll ba

cter

iaSe

kigu

chi e

t al.

FISH

S-D

-Arc

h-09

15-a

-R-2

0V

irtu

ally

all

Arc

haea

1999

S-F-

Mba

c-11

74-a

-R-2

2M

etha

noba

cter

iale

sS-

O-M

mic

-120

0-a-

R-21

Met

hano

mic

robi

ales

S-F-

Msa

r-14

14-a

-R-2

1M

etha

nosa

rcin

acea

eS-

F-M

sae-

0825

-a-R

-23

Met

hano

saet

acea

eS-

G-D

sbb-

0660

-a-R

-20

Des

ulfo

bulb

usS-

*-M

UG

28–0

701-

a-R-

20rD

NA

clo

ne M

UG

28S-

*-G

NSB

-063

3-a-

R-20

rDN

A c

lone

s in

gre

en n

on-s

ulfu

r ba

cter

ia

UA

SBM

Prop

iona

te,

Dot

blo

tS-

D-B

act-

0338

-a-R

-18

Vir

tual

ly a

ll ba

cter

iaH

ofm

an-B

ang

etal

.bu

tyra

teS-

D-A

rch-

0915

-a-R

-20

Vir

tual

ly a

ll A

rcha

ea20

01S-

*-Sr

b-03

85-a

-R-1

8G

ram

-neg

ativ

e su

lfate

-red

ucin

g ba

cter

iaS-

*-R

6B-7

–098

0-a-

R-18

Clo

stri

dium

-lik

e ba

cter

iaS-

*-R

1B-1

6–09

80-a

-R-1

8C

lost

ridi

um-l

ike

bact

eria

explained by the syntrophic growth of the SRB population in conjunction withthe methanogens. The high number of Methanobacteriales found may beexplained by the competitive advantage over Methanococcales and Methanosar-cinales, when hydrogen and formate kinetics are compared.

Amann et al. [155] investigated the role of sulfate-reducing bacteria (SRB) inthe establishment and development of a biofilm. This was done by comparativesequence analysis and FISH in biofilm targeting selected SRB populations.Biofilms were grown on coverslips and reached a thickness of 10 µm. Two cloneswere identified as delta-Proteobacteria. One was closely related to Desul-furomonas acetoxidans. The other was closely related to Desulfovibrio vulgaris.Specific probes constructed to target these two SRB did not hybridize to any cellsin the biofilm. If specific probes constructed on the basis of the two cloned 16SrDNA genes were used, single cells could be identified. If the PCR primer S-*-Srb-0385-a-R-18, used to construct the clone library, however, was used as afluorescent probe, only one of the populations could not be identified. The gen-eral observation was that the biofilm formation was patchy and that the identi-fied cells were gathered in colonies. The authors also concluded that the differ-ent results obtained with the FISH probes demonstrate that a better insight inthe SRB diversity is needed to construct more specific probes.

3.1.2Biofilm Composition and Dynamics

Raskin et al. [154] studied the competition and coexistence of sulfate-reducingbacteria (SRB) and methanogens in anaerobic biofilms. Four biofilm reactors,MA, MB, SA, SB, were operated on glucose as sole carbon source with (SA, SB) orwithout sulfate (MA, MB). After eleven months of operation, sulfate was addedto MB and omitted from SB. MA and SA were maintained as control reactors.Probing of samples taken throughout the experiment using the probes shown inTable 1 was carried out.

Reactor MA contained 25% methanogens. Specific probing showed thatMethanobacteriales accounted for 12% and Methanosarcinales, Methanomicro-biales, Methanococcales each occurred in the range of 4 –5 %. Prior to sulfateaddition, SRB comprised a significant fraction of the community in themethanogenic reactors. Desulfovibrio and Desulfobacterium genera were pre-sent in high amounts (16% and 2.8%, respectively). As previously mentioned,these findings may be explained by the capability of several SRB to grow syn-trophically on lactate, ethanol, propionate, and pyruvate.

After the addition of sulfate to the MB reactor, sulfate reduction started after6 hours, reaching steady state levels within a few days. Desulfovibrio and Desul-fobacterium sharply increased to 26% and 7.7%, respectively. These levels de-creased subsequently to 20% and 3.5%, followed by a slow increase to 35% and4.5%, respectively. Desulfobulbus and Desulfobacterium spp. increased shortlyafter the addition of sulfate, but later fell to the levels observed before sulfateaddition. Desulfosarcina,Desulfococcus, and Desulfobotulus spp.all fell below thelevels observed before the sulfate addition (from 2% to 0.5%). 100 days after thesulfate addition, the acetate levels increased in the reactor for about 100 days.


This change in acetate was not accompanied by detectable population changes,thus emphasizing the need for a careful evaluation of the concept “steady-state”in anaerobic reactors. Even though the methane production decreased to unde-tectable levels after the addition of sulfate, the methanogenic populations stillconstituted about 8%. Obviously, the ribosome content was maintained formonths in the methanogens despite the inactivity of their primary metabolicpathways.

After the sulfate was removed from the inflow of reactor SB, methanogenicpopulations slowly increased to steady-state levels dominated by Methanobacte-riales (18%). The relative abundance of other methanogens remained fairly con-stant. Following sulfate removal, Desulfovibrio and Desulfobacterium popula-tions decreased to levels comparable to the control reactor (MA). These obser-vations show that hydrogen is the key substrate when the metabolism shifts fromsulfidogenesis to methanogenesis.After several hundred days of operation with-out sulfate, the microbial community structure and function of the SB reactorwas similar to that of the methanogenic control reactor MA.

Similar to observations from other sulfate-rich environments, the SRB wereshown to outcompete methanogens in high-sulfate biofilms. Also, the presenceof Desulfovibrio and Desulfobacterium spp. only varied a factor 2 as a functionof the presence or absence of sulfate. Moreover, Desulfosarcina, Desulfococcus,and Desulfobotulus spp. turned out to be better adapted to the biofilms withoutsulfate. The content of ribosomes in methanogens only slowly decreased uponsulfate addition although methane could not be detected. Wagner et al. [143]have made similar observations when probing denitrifying bacteria in waste-water treatment plants. It therefore seems reasonable to correlate microbialresponses to environmental perturbations to increased ribosome content, whileit is more problematic to correlate decreasing microbial activity to decreasingribosome content.

3.2Granular Sludge Reactors

3.2.1Granular Sludge

Granular sludge consists of conglomerates of anaerobic microorganisms, whichare still visible as granules after settling and is considered a major form forimmobilization of microorganisms in anaerobic wastewater treatment systems[156]. Similar to biofilms, granular sludge provides minimized mass transferlimits, optimal micro-environment, and protection for microorganisms such asmethanogens and syntrophic bacteria. Granules typically form in upflow anaer-obic sludge blanket (UASB) reactors, although they also might be found in oth-er anaerobic systems, such as expanded granular sludge blanket (ESGB) reactors[157, 158], upflow sludge bed filters (UBF) [2, 159], compartmentalized UASBreactors [160, 161], anaerobic migrating blanket reactors (AMBR) [162], anaer-obic baffled reactors (ABR) [163], and anaerobic sequencing batch reactors(ASBR) [164]. Granular sludge has been used successfully to treat wastewaters


having varying COD contents. The loading can be as high as 20 g COD/L [165],or as low as 1 g COD/L, such as for domestic wastewater [166]. Anaerobic gran-ular sludge reactors can be operated at mesophilic, thermophilic, and even psy-chrophilic conditions. After acclimation, granular sludge can treat wastewatercontaining refractory, toxic, or xenobiotic compounds [167–174]. Even thoughUASB reactors are usually operated at neutral pH, granules can adapt to high pH(8.1–8.5) [175] or low pH (6.0) [176]. UASB reactors have also been used to treatsulfate rich wastewater (SO4

2–/COD >1) [177].To further optimize the process, it is of great interest to understand the micro-

bial species composition, and structure of granular sludge, as well as the granu-lation processes.

3.2.2Microbial Composition of Granules

Zheng and Raskin [178] used membrane hybridization with radioactivelylabeled oligonucleotide probes to quantify methanogens in granules from twolab-scale UASB reactors. The two reactors were fed synthetic wastewater(COD = 4 g/L) containing glucose or glucose/propionate as the only energysource. In both reactors, the granular sludge contained around 40% archaeal 16SrRNA of which aceticlastic methanogens, Methanosarcinales were dominating.Quantification by a species-specific probe revealed that Methanosaeta conciliiwas the predominant species among the Methanosarcinales. This finding wasconsistent with the observation that Methanosaeta-like filaments often are dom-inating in granules.Among the hydrogenotrophic methanogens, Methanobacte-riaceae were dominant in the glucose-fed reactor (about 5%). In the glucose/propionate-fed reactor, Methanobacteriaceae and Methanomicrobiales eachmade up around 5% of the microbial populations.

Zheng and Raskin also analyzed granules sampled at different heights from afull-scale UASB reactor treating wastewater from a corn wet milling plant [178].The archaeal 16S rRNA constituted between 30 to 50% of total 16S rRNA. Thehighest numbers of Archaea were found in granules sampled from the top of thereactor. Methanosarcinales and Methanobacteriaceae were the two dominantmethanogenic groups. Methanomicrobiales constituted less than 5% whileMethanococcaceae were almost absent. Within the Methanosarcinales, whichconstituted between 7.4 and 27.9%, only 2.2 to 2.8% were Methanosaeta concilii,and less than 1% were Methanosaeta thermophila and Methanosarcina species.The authors concluded that genera such as Methanolobus, Methanococcoides,Methanohalophilus, Methanohalobium, and Methanosalsus within Methano-sarcinales may be present in high numbers in these granules.

The population changes of propionate-oxidizing bacteria after granule for-mation in potato-processing industry wastewater adapted to different substrateswas investigated by Harmsen et al. [90]. Very low amounts of propionate-utiliz-ing, sulfate-reducing bacteria Desulfobulbus spp. (around 2% of total 16S rRNA)and the syntrophic propionate degrader strain SYN7 (less than 1% of total 16SrRNA) were found in the inoculum. After adaptation, the Desulfobulbus spp.increased up to 35% in the granules fed propionate and sulfate, while SYN7


increased to around 10% in the granules fed only propionate. The universalprobe used in this study underestimated the amounts of bacterial 16S rRNA,which may have introduced significant biases [179]. Sekiguchi et al. [180] ana-lyzed the microbial diversity of two types of methanogenic granular sludge,sampled from a mesophilic (35°C) and a thermophilic (55°C) UASB reactortreating synthetic wastewater containing sucrose, propionate, and acetate. Clonelibraries of 16S rDNA were constructed using a prokaryote-specific primer set,followed by partial sequencing of the cloned rDNAs. It was found that 19% of theclones from the mesophilic granules and 22% from the thermophilic granuleswere closely related to methanogens, while the rest were Bacteria.A major groupof bacterial clones from the mesophilic granules showed homology to the deltasubclass of the Proteobacteria (27%) harboring syntrophic bacteria and sulfate-reducing bacteria. The bacterial clones from the thermophilic granules, how-ever, were mainly Thermodesulfovibrio spp. (19%), green non-sulfur bacteria(18%) and low G+C Gram-positive bacteria (18%). The authors also found thatthe microbial diversity of the thermophilic granules was lower compared to themesophilic granules.

The population changes in granular sludge in a sucrose-fed five-compart-ment AMBR system were monitored when staging was established [181]. Usingprobes for methanogens and Archaea, the authors demonstrated that the high-est amounts of methanogenic 16S rRNA was found in the middle compartment,where 42% of the total 16S rRNA belonged to Archaea. Methanosaeta account-ed for 32% of the 16S rRNA, Methanobacteriaceae for 8%, and Methanomicro-biales for 2%. Throughout the process, Methanosaeta spp. were always the pre-dominant aceticlastic methanogens. Even though acetate concentrations in theside compartments were as high as 600 mg/L, Methanosarcina spp. alwaysamounted to less than 1%. With respect to hydrogenotrophic methanogens,Methanobacteriaceae occurred in highest amounts followed by Methanomicro-biales. Methanococcaceae were almost absent. Interestingly, syntrophic bacteriasuch as Syntrophomonas and Syntrophobacter, and sulfate-reducing bacteriaDesulfobulbus occurred in similar amounts in the different compartments evenafter the staging was established.

Microbial diversity of syntrophic bacteria in granular sludge in UASB reac-tors was studied by Hofman-Bang et al. [182]. Microbial enrichment was con-ducted in a UASB reactor at 32°C fed with butyrate and propionate. After threemonths of stable operation, the granular sludge was divided between three newUASB reactors fed with butyrate and propionate. One reactor was kept as a con-trol reactor, a second reactor was fed 10 mM sulfate, and in a third reactor thetemperature was increased by 10°C. After three months of operation, the granu-lar sludge from each reactor was again divided into two new UASB reactors fedwith either butyrate or propionate and operated for six weeks. Community fin-gerprinting (DGGE) from the six reactors showed a low number of bands indi-cating a bacterial enrichment consisting of 3–5 different species. Bacterial clonelibraries were constructed from each reactor and 15–30 clones from each librarywere fully sequenced. Phylogenetic analysis indicated that bacterial clones were85.9%–99.7% homologous to members of the Gram-negative d-Proteobacteriaand the Gram-positive Syntrophomonas cluster.


3.2.3Structure of Granular Sludge

The microbial structure of granular sludge has been studied using fluorescence insitu hybridization (FISH) with rRNA-targeted oligonucleotide probes. Granulescollected from a full-size UASB reactor treating potato-processing wastewaterwere shown to have a layered structure [90]. The thick outer layer of the gran-ules was dominated by Bacteria and contained very few methanogens. Desulfob-ulbus spp. were present in this layer. The inner layer of the granules had twotypes of microcolonies arranged in concentric circles. One type only containedmethanogens and the other type consisted of Bacteria tangled with filamentousmethanogens. Two lab-scale reactors were inoculated with the granules, one wasfed propionate (methanogenic) and the other was supplied with propionate andsulfate (sulfidogenic). After 8–12 weeks of adaptation, the granules in themethanogenic reactor had lost the thick outer bacterial layer. Hybridizationwith bacterial probes revealed two types of microcolonies yielding strong andweak signals, respectively. Microcolonies yielding strong signals were found tocontain syntrophic propionate oxidizers and methanogens. Granules that adapt-ed in the sulfidogenic reactor had established a new outer layer dominated byDesulfobulbus.

Harmsen et al. observed three layers in granules that were originally obtainedfrom a system treating sugar beet wastewater and then adapted to sucrose for sixmonths [90]. The external layer contained mainly Bacteria. The middle layerconsisted of syntrophic microcolonies containing propionate-degraders andMethanobrevibacter spp. (detected by the Methanobacteriaceae-specific probeand morphologically similar to Methanobrevibacter). In this layer, transmissionelectron microscopy and FISH revealed microcolonies of Methanosaeta spp.located next to the syntrophic microcolonies. The core contained inorganicmaterial with large cavities and some methanogens were detected by anArchaea-specific probe. In the same study, the authors showed that granulesfrom a system fed a sugar beet wastewater that were adapted to a mixture of VFA(butyrate :propionate :acetate = 42:32:24), created similar structures as thesucrose-adapted granules, except for an additional thick layer between theexternal and the middle layer rich in Methanosaeta spp. and Methanosarcinaspp. microcolonies. The authors explained this extra layer by the high acetateconcentration in the feed. High concentrations of acetate are inhibitory to syn-trophic propionate degradation due to the unfavorable thermodynamic condi-tions. Methanosaeta spp. and Methanosarcina spp. present in the thick layercould remove the acetate before it reached the syntrophic microcolonies. Sincethe feed contained only butyrate, propionate, and acetate, the authors concludedthat bacteria in the external layer were mainly butyrate degraders.

Granules from a mesophilic and a thermophilic lab-scale UASB reactor wereanalyzed by Sekiguchi et al. [183]. Granules from both reactors had an outer lay-er dominated by Bacteria and an inner layer dominated by Archaea and anunstable center. Methanosaeta spp. were the predominant Archaea in both typesof granules. In granules from the mesophilic UASB reactor, some Methanobac-teriaceae cells were observed together with Bacteria. Some Methanomicrobiales


cells were also detected that were spread in the granules. In granules from thethermophilic reactor, some Methanobacteriaceae and Methanosarcina cellswere detected. The authors also tried to locate Bacteria that were dominating inthe clone libraries from their previously mentioned study [180]. In mesophilicgranules, Desulfobulbus cells were found in the outer layer of the granules. Cellsclosely related to Syntrophobacter were shown to form microcolonies togetherwith Methanobacteriaceae cells in the mesophilic granules. A probe targetinggreen non-sulfur bacteria revealed filamentous cells on the surface of the ther-mophilic granules.

3.2.4The Granulation Process

The potentials for Methanosaeta concilii and propionate-degrading syntrophicconsortia to serve as nuclei for granulation were tested by monitoring the gran-ulation processes from non-granular sludge in two laboratory-scale upflowanaerobic sludge blanket reactors treating synthetic wastewater [184]. The influ-ent of one reactor contained glucose as the only energy sources, while the influ-ent of the other reactor contained glucose and propionate. The two reactors werestarted following the recommended procedures and the effluent acetate concen-trations were maintained below 200 mg/L. Quantitative membrane hybridiza-tions and FISH were used to monitor the changes of microbial communities andto investigate the cell aggregate structures during the granulation processes.Methanosaeta concilii demonstrated good settling ability and a large populationdeveloped in the microbial community. The increase in population size was cor-related with the significant increase in cell aggregate sizes at the early stage ofgranulation. Methanosaeta concilii cells were found to serve as backbones in thesmall cell aggregates having other archaeal and bacterial cells attached to them.They remained dominant in larger cell aggregates and in the mature granules.These findings support the hypothesis that Methanosaeta serves as nuclei forgranulation. Syntrophic propionate-oxidizing bacteria, on the other hand,exhibited poor settling capability and were easily washed out from the system.Their contribution to granulation, therefore, is probably minimal.

3.3Continuously Stirred Tank Reactors (CSTR)

3.3.1Microbial Composition in CSTRs

Zheng and Raskin reevaluated the probes available for detection of Methano-saeta species in mesophilic and thermophilic anaerobic digesters [178]. Gen-erally Methanosaeta species are detected with the S-F-Msae-0825-a-A-23 probe.However, some of the Methanosaeta spp. 16S rRNA sequences have a deletion in the target site of this probe at position 838 (based on E. coli numbering). Tocircumvent the problems with this probe, new probes targeting the genusMethanosaeta according to Table 1 were constructed and tested on anaerobic


digesters. Also, specific probes for Methanosaeta concilii and Methanosaetathermophila were constructed. Three probes targeting the genus Methanosaetawere tested against reference strains with or without mismatches. The probes S-G-Msae-0733-a-A-22 and S-G-Msae-0781-a-A-22 were not able to discriminatebetween target and non-target sites since the optimal wash temperatures weretoo close to each other. A third probe, S-G-Msae-0332-a-A-22 was designed andshown to be specific to the genus Methanosaeta.

To evaluate the probes, environmental samples from various mesophilic andthermophilic digesters were analyzed for the presence of Methanosaeta spp.In general, the two new probes detected a higher level of Methanosaeta than the old S-F-Msae-0825-a-A-23 probe. This could indicate putative deletions in target sites of some Methanosaeta species. Moreover, the amount of Me-thanosaeta spp. in the digesters decreased when acetate concentrations in-creased, and increased when acetate concentrations decreased. Methanosarcinaincreased with increasing acetate concentrations which is in agreement withprevious findings[185].

Hansen et al. [186] quantified the syntrophic fatty acid-b-oxidizing bacteriain a mesophilic biogas digester treating cow and swine manure. Syntrophic fat-ty acid b-oxidizing bacteria were found in the Gram-positive Syntrophomon-adaceae cluster. This family currently contains three genera, Syntrophomonas,Syntrophospora, and Thermosyntropha as well as two lost strains FSM2 and FSS7[187–190]. Probing of samples from the digester showed that members ofMethanomicrobiales were the most abundant hydrogenotrophic methanogensconstituting 10% of the prokaryotes. Syntrophomonadaceae were estimated tocomprise 0.2–1% and probing for the different syntrophic genera showed thatonly members of the genus Syntrophomonas degrading butyrate were present.Methanosarcina spp. were the only methanogens present apart from the Methanomicrobiales. This is consistent with previous studies demonstratingthat Methanosarcina spp. are the main acetate utilizers in Danish biogas plants[191]. The relatively low numbers of butyrate degrading bacteria imply thatthese have high metabolic rates corresponding to the low energy yield of VFAoxidation.

Raskin et al. [192] quantified the abundance of sulfate reducing bacteria(SRB) and methanogens in twenty-one single-phase and two-phase full-scaleanaerobic sewage sludge digesters by oligonucleotide probing of 16S rRNA. Itwas determined that methanogens in well-functioning mesophilic, single-phasedigesters accounted for 8–12% of which Methanosarcinales and Methanomicro-biales constituted the majority. Methanobacteriales and Methanococcales onlyplayed a minor role.

The SRB were present in significant amounts. Desulfovibrio and Desulfobul-bus spp. dominated the community while Desulfobacter and Desulfobacteriumwere less abundant. Desulfosarcina, Desulfococcus and Desulfobotulus spp. werenot detected. Even though sulfate was present in small amounts in the sludge, therelative high level of SRB found in most of the digesters indicates that SRB stillcan compete with the methanogens for available electrons and/or with proton-reducing syntrophs for fermentation products such as propionate, butyrate, lac-tate, or ethanol [192].


Hristova et al. [194] constructed genus- and subgenus-specific probes forGram-positive SRB. Thermophilic anaerobic digesters were probed to evaluatethe probes and to quantify the abundance of Desulfotomaculum spp. Desulfo-tomaculum spp. accounted for about 2% of the total bacterial population at thestart up of a thermophilic CSTR digester inoculated with mesophilic digestersludge and cow manure.After 31 days of operation the amount had decreased to0.3%. Probing against Desulfovibrio, however, indicated that this genus account-ed for 1.7%. This may be explained by the Desulfovibrio being more competitivethen Desulfotomaculum under the low sulfate concentrations found in the reactor.

Probing another thermophilic digester operated for a longer period showedDesulfotomaculum at a stable level of 2%. However, the sulfate concentrationwas two times higher than in the start-up reactor, and the long operation timemay have stabilized the digester allowing Desulfotomaculum to gain foothold.

Godon et al. [195] have determined the microbial diversity in an anaerobicdigester treating wine distillation waste by extensive sequencing of clonelibraries of 16S rDNA. 579 clones were partially sequenced and analyzed byoperational taxonomic unit (OTU) phylogenetic analysis; 146 OTUs were foundof which 133 belonged to Bacteria, 6 to Archaea and 7 to Eucarya. The bacterialclones were distributed among at least eight of the major groups of the Bacteriadomain. Despite the large bacterial diversity, the 20 most frequent bacterialOTUs represented 50% of the total clones.

3.3.2Microbial Dynamics in CSTRs

Ahring and coworkers investigated the archaeal population dynamics in adigester treating cow manure during a temperature shift from 55°C to 65°C[196]. Thermophilic anaerobic digestion of complex organic wastes is usuallycarried out at 50–55°C. Previous studies demonstrated that an increase in theoperational temperature to 65°C caused a disturbance of the biogas processreflected in an increase in VFA concentrations [191]. This indicates that the pop-ulations participating in the terminal part of the biogas process are much moresensitive to a temperature increase than the fermentative populations responsi-ble for VFA production. Probing of samples from the digester showed that thearchaeal population increased significantly as a consequence of the temperatureincrease and that the bacterial population decreased correspondingly. Analysesof the archaeal population showed that the Crenarchaeota increased dramati-cally at the cost of the Euryarchaeota after the temperature increase. Meth-anosaetaceae were never detected, Methanosarcinaceae disappeared and Meth-anococcus increased to a relatively high amount after the temperature change.These findings indicate that our knowledge about the diversity in thermophilicreactors is limited and that much more work is needed to elucidate the popula-tion dynamics at elevated temperatures.

Fernandez et al. [67] conducted glucose perturbation studies on CSTRdigesters under mesophilic conditions to evaluate the relationship between sta-bility and community structure. Two digesters were studied; one (HS) was inoc-


ulated with fluid from a bioreactor operating for 200 days on glucose and theother (LS) was inoculated with fluid from a bioreactor operating for 60 days onglucose. The effect of shock loading the digesters with glucose was monitored bymorphological analysis, 16S rRNA probing, restriction analysis of amplifiedrDNA, and partial 16S rDNA sequencing. The HS reactor was characterized bygood replicability, a high proportion of spirochete-like and short thin rod mor-photypes, a dominance of spirochete-related 16S rDNA genes, and a high per-centage of Methanosarcina-related 16S rRNA. The LS reactor was characterizedby higher morphotype diversity dominated by cocci, a predominance of Strep-tococcus-related and deeply branching spirochete-related 16S rDNA genes, anda high percentage of Methanosaeta-related 16S rRNA.

In the HS reactor, the glucose shock caused a dramatic shift in the relativeabundance of fermentative bacteria, resulting in a temporary displacement ofspirochete-related ribotypes by Eubacterium-related ribotypes followed by areturn to the pre-shock community structure. The LS reactor was less affected,and the Streptococcus-related organisms still dominated after the glucose shock,although changes in the relative abundance of some of the members weredetected by morphotype analysis.

4Concluding Remarks

During the last decade, environmental microbiology has changed markedly as aconsequence of the exploitation of molecular biology methods for answeringquestions such as which organisms are active and where and when do they showactivity. The number of isolated microbes has increased markedly and basedupon molecular analyses, researchers are able to improve experimental designsand to isolate new bacteria. Also, as our knowledge of microbial metabolism isincreasing, we are coming closer to be able to answer questions related to whatmicrobes are doing in situ.

Molecular microbiology has increased our field of vision and our resolutioncapability. Now, traces of DNA from one gene are often sufficient to assign thebacterium from which the DNA originated in a phylogenetic context. Wholegenomes are being sequenced using the same effort needed for sequencing of asingle gene a few decades ago. Sequence information can now be used to linkmetabolism, phylogeny, and ecology. Sequence databases are growing exponen-tially in size and bioinformatics, therefore, will inevitably play an increasing rolein microbial ecology. From an evolutionary point of view, massive amounts ofDNA sequences and powerful computers have made it possible to conduct phy-logenetic analysis on these data and to broaden our view of how life on Earthevolved. In the foreseeable future, genome sequence analysis will to a higherdegree link 16S rRNA/23S rRNA gene sequences to metabolic functions of wholebacterial families.

To a large degree, the different techniques developed for microbial diversityscreening discussed in this chapter answer the questions of who is present.To a limited extent, we are able answer questions on who is where since homogeneous cell accessibility and probe specificity still are two key problems


in in situ probing. Furthermore, we still need to develop methods to deter-mine activity levels in situ (i.e., who is doing what in which location at which point in time?). For instance, interactions and interdependence amongdifferent microbial populations are probably much more pronounced than generally assumed. The metabolic expression pattern of a single microbial cell probably differs from the neighboring cell yielding a multi dimensionalpatchwork. Several major constraints, therefore, have to be overcome, and several new concepts have to be adapted and resolved to fully exploit the use of molecular methods in understanding complex ecosystems such as anaerobicreactors.

5References

1. Grotenhuis J-TC, Smit M, Plugge CM, Yuansheng X, van Lammeren A-AM, Stams A-JM,Zehnder A-JB (1991) Appl Environ Microbiol 57:1942

2. MacLeod FA, Guiot SR, Costerton JW (1990) Appl Environ Microbiol 56:15983. Morgan JW, Evison LM, Forster CF (1991) J Chem Tech Biotech 50:2114. Wu WM, Jain MK, Conway de Macario E, Thiele JH, Ziekus JG (1992) Appl Microbiol

Biotech 38:2825. Pace NR (1996) ASM News 62:4636. Amann RI, Ludwig W, Schleifer KH (1995) Microbiol Rev 59:1437. Macario AJL, Visser FA, van Lier JB, Conway de Macario E (1991) J Gen Microbiol

137:21798. Ovreas L, Forney L, Daae FL, Torsvik V (1997) Appl Environ Microbiol 63:33679. Schmidt JE, Ahring BK (1991) Appl Microbiol Biotech 35:681

10. Amann R, Lemmer H, Wagner M (1998) FEMS Microbiol Ecol 25:20511. Abu A, Radstrom P (1998) Appl Environ Microbiol 64:374812. Delong EF, Wickham GS, Pace NR (1989) Science 243:136013. Archer DB (1984) Appl Environ Microbiol 48:79714. Kemp HA, Archer DB, Morgan MRA (1988) Appl Environ Microbiol 54:100315. Sorensen AH, Ahring B (1997) Appl Environ Microbiol 63:200116. Olsen GJ, Woese C (1993) Fase B Journal 7:11317. Pace NR (1997) Science 276:73418. Zuckerkandl E, Pauling L (1965) J Theor Biol 8:35719. Olsen GJ, Lane DJ, Giovannoni S, Pace NR, Stahl DA (1986) Ann Rev Microbiol 40:33720. Pace NR, Stahl DA, Lane DJ, Olsen GL (1986) Adv Microb Ecol 9:121. Stahl DA, Flescher B, Mansfield HR, Montgommery L (1988) Appl Environ Microbiol

54:107922. Woese C, Fox GE (1977) Proc Natl Acad Sci USA74:508823. Woese C (1987) Microbiol Rev 51:22124. Woese C, Kandler O, Wheelis ML (1990) Proc Natl Acad Sci USA 87:457625. Maidak BL, Cole JR, Lilburn TG, Parker CT, Saxman PR, Stredwick JM, Garrity GM, Li B,

Olsen GJ, Pramanik S, Schmidt TM, Tiedje J (2000) NAR 28:17326. Stahl DA (1986) Biotechnology 4:62327. Ward DM, Bateson MM, Weller R, Ruff-Roberts AL (1992) Adv Micro Ecol 12:21928. Klappenbach JA, Saxman PR, Cole JR, Schmidt TM (2001) Nucl Acids Res 29:18129. Ludwig W, Schleifer K-H (1999) ASM News 65:75230. Swofford DL, Olsen GJ, Waddell PJ, Hillis DM (1996) In: Hillis DM, Moritz C (eds),

Molecular systematics. p 407, Sinauer Associates, Sunderland, MA, USA31. van Peer Y (1996) In: Durbin R, Eddy S, Krogh A (eds), Biological sequence analysis,

vol 24, p 3381, Cambridge, Mitchison, USA


32. Raskin L, Capman WC, Kane MD, Rittmann BE, Stahl DA (1996) Appl Environ Microbiol62:300

33. Ibrahim A, Ahring BK (1999) Biotechniques 27:113234. Moran MA, Torsvik VL, Torsvik T, Hodson RE (1993) Appl Environ Microbiol 59:91535. Frischer ME, Danforth JM, Healy MN, Saunders FM (2000) Appl Environ Microbiol

66:303736. Yu Z, Mohr WW (2001) Can J Microbiol 45:26937. Libonati M, Sorrentino S (1992) Mol Cell Biochem 117:13938. Alm EW, Stahl DA (2000) J Microbiol Meth 40:15339. Raskin L, Capman WC, Sharp R, Poulsen LK, Stahl DA (1997) In: Mackie RI, White BA,

Isaacson RE (eds) Ecology and Physiology of Gastrointestinal, vol 2. p 243 Chapman andHall, New York

40. Alm EW, Zheng DD, Raskin L (2000) Appl Environ Microbiol 66:454741. Steffan RJ, Atlas RM (1991) Ann Rev Microbiol 45:13742. Roche Molecular Biochemicals, PCR Application Manual, 2nd edition43. Madigan MT, Martinko JM, Parker J (1997) Biology of microorganisms. Prentice-Hall,

Upper Saddle River, NJ, USA44. Nubel U, Engelen B, Felske A, Snaidr J, Wieshuber A, Aaann RI, Ludwig W, Backhaus H

(1996) J Bact 178:563645. Petterson B, Leitner T, Ronaghi M, Bolske G, Uhlen M, Johansson KE (1996) J Bact

178:413145a. Rainey FA,Ward-Rainey NL, Janssen PH, Hippe H, Stackebrandt E (1996) Microbiology

142:208746. Muyzer G, De-Wall ED, Uitterlinden AG (1993) Appl Environ Microbiol 59:69547. Curtis TP, Craine NG (1998) Water Sci Technol 37:7148. Santegoeds CM, Nold SC, Ward DM (1996) Appl Environ Microbiol 62:392249. Stephen JR, Kowalchuk GA, Bruns MV, McCaig AE, Phillips CJ, Embley TM, Prosser JI

(1998) Appl Environ Microbiol 64:295850. Jensen S, Ovreas L, Daae FL, Torsvik V (1998) FEMS Microbiol Ecol 26:1751. Santegoeds CM, Ferdelman TG, Muyzer G, de Beer D (1998) Appl Environ Microbiol

64:373152. Ferris MJ, Ward DM (1997) Appl Environ Microbiol 63:137553. Ferris MJ, Muyzer G, Ward DM (1996) Appl Environ Microbiol 62:34054. Murray AE, Hollbaugh JT, Orrego C (1996) Appl Environ Microbiol 62:267655. Teske A, Wawer C, Muyzer G, Ramsing NB (1996) Appl Environ Microbiol 62:140556. Rollele S, Muyzer G, Wawer G, Wanner G, Lubitz W (1996) Appl Environ Microbiol

62:205957. Muyzer G, Smalla K (1998) Antonie van Leeuwenhoek 73:12758. Dunbar J, Ticknor LO, Kuske CR (2001) Appl Environ Microbiol 67:19059. Massol-Deya A, Weller R, Rios-Hernandez L, Zhou JZ, Hickey RF, Tiedje JM (1997) Appl

Environ Microbiol 63:27060. Ingianni A, Petruzzalli S, Morandotti G, Pompei R (1997) FEMS Immun Med Microbiol

18:6161. Vachee A, Vincent P, Struijk CB, Mossel DA, Leclerc H (1997) Syst Appl Microbiol

20:49262. Smit E, Leeflang P, Wemars K (1997) FEMS Microbiol Ecol 23:24963. Heyndrichx M, Lebbe L, Vancanneyt M, Kersters K, de Vos P, Logan NA, Forsyth G, Nazli

S, Ali N, Berkeley RCW (1996) Int J Syst Bact 47:80864. Lee IM, Bartoszyk IM, Gundersen-Rindal DE (1997) Appl Environ Microbiol 63:263165. Urakawa H, Kita-Tsukamoto K, Ohwada K (1997) FEMS Microbiol Lett152:12566. van Damme P, Heyndrichx M, de Rosse I, Lammens C, de Vos P, Kersters K (1997) Int J

Syst Bact 47:80267. Fernandez AS, Hashsham SA, Dollhopf SL, Raskin L, Glagoleva O, Dazzo FB, Hickey RF,

Criddle CS, Tiedje JM (2000) Appl Environ Microbiol 66:405868. Scheinert P, Krausse R, Ullmann U, Soller R, Krupp G (1996) J Microbiol Meth 26:103


69. Godon JJ, Zumstein E, Dabert P, Habouzit F, Moletta R (1997) Water Sci Technol 36:49

70. Suzuki MT, Giovannoni SJ (1996) Appl Environ Microbiol 62:62571. Holland PM, Abramson RD, Watson R, Gelfand DH (1991) Proc Natl Acad Sci USA

88:727672. Lee LG, Connell CR, Block W (1993) Nucl Acid Res 21:376173. Bassler HA,Flood SJA,Livak KJ,Marmaro J,Knorr R,Batt CA (1995) Appl Environ Micro-

biol 61:372474. Sayler GS, Layton AC (1990) Ann Rev Microbiol 44:62575. Amann R, Ludwig W (2000) FEMS Microbiol Rev 24:55576. Stahl DA, Amann RI (1991) In: Stackebrandt E, Goodfellow M (eds), Nucleic acid tech-

niques in bacterial systematics. p 205, John Wiley & Sons Inc, New York77. Wallner G, Amann R, Beisker W (1993) Cytometry 14:13678. Trebesius K,Amann R, Ludwig W, Muhlegger K, Schleifer KH (1994) Appl Environ Micro-

biol 60:322879. Heuer H, Hartung K, Wieland G, Kramer I, Smalla K (1999) Appl Environ Microbiol

65:104580. Stackebrandt E, Goebel BM (1994) Int J Syst Bacteriol 44:84681. Alm EW, Oerther DB, Larsen N, Stahl DA, Raskin L (1996) Appl Environ Microbiol

62:355782. Amann RI (1995) Mol Ecol 4:54383. Frischer ME, Floriani PJ, Nierzwicki-Bauer SA (1996) Can J Microbiol 42:106184. Jurtshuk RJ, Blick M, Bresser J, Fox GE (1992) Appl Environ Microbiol 58:257185. Fuchs BM, Wallner G, Beisker W, Schwippl I, Ludwig W, Amann R (1998) Appl Environ

Microbiol 64:497386. Fuchs BM, Glockner FO, Wulf J, Amann R (2000) Appl Environ Microbiol 66:360387. Tijssen P (1993) In: Vliet PCVD (ed), Laboratory techniques in biochemistry and mole-

cular biology, vol 24. p 613, Elsevier Science Publishers, Amsterdam88. McMahon KD, Stahl DA, Raskin L (1998) Mol Ecol 36:36289. Delong EF, Wu K, Prezelin BB, Jovine RVM (1994) Nature 371:69590. Harmsen H-JM,Akkermans A-DL, Stams A-JM, De-Vos WM (1996) Appl Environ Micro-

biol 62:216391. Massana R, Murray AE, Preston CM, Delong EF (1997) Appl Environ Microbiol

63:5092. Macgregor BJ, Moser DP, Alm EW, Nealson KH, Stahl DA (1997) Appl Environ Microbiol

63:117893. Polz MF, Cavanaugh CM (1997) Appl Environ Microbiol 63:102894. Giovannoni SJ, Delong EF, Olsen GJ, Pace NR (1988) J Bact 170:72095. Voordouw G, Shen Y, Harrington CS, Telang AJ, Jack TR, Westlake DWS (1993) Appl

Environ Microbiol 59:412196. Telang AJ, Ebert S, Foght JM, Westlake DS, Jenneman GE, Gevertz D, Voordouw G (1997)

Appl Env Microbiol 63:178597. Shen Y, Stehmeier LG, Voordouw G (1998) Appl Environ Microbiol 64:63798. Harmsen H-JM, Kengen H-MP, Akkermans A-DL, Stams A-JM, De-Vos WM (1996) Appl

Environ Microbiol 62:165699. Neef A, Zaglauer A, Meier H, Amann R, Lemmer H, Schleifer K-H (1996) Appl Environ

Microbiol 62:4329100. Schramm A, Larsen LH, Revsbech NP, Ramsing NB, Amann R, Schleifer K-H (1996) Appl

Environ Microbiol 62:4641101. Amann R, Snaidr J, Wagner M, Ludwig W, Schleifer KH (1996) J Bact 178:3496102. Poulsen LK, Ballard G, Stahl DA (1993) Appl Environ Microbiol 59:1354103. Amann RI, Krumholz LR, Stahl DA (1990) J Bact 172:762104. de los Reyes FL, Ritter W, Raskin L (1997) Appl Environ Microbiol 63:1107105. Amann RI, Binder BJ, Olson RJ, Chisholm SW, Devereux R, Stahl DA (1990) Appl Environ

Microbiol 56:1919


106. Manz W, Wagner M, Amann R, Schleifer KH (1994) Wat Res 28:1715107. Amann RI (1995) Mol Ecol Manual p3.3.6108. Burggraf S, Mayer T, Amann R, Schadhauser S, Woese CR, Stetter KO (1994) Appl Envi-

ron Microbiol 60:3112109. Sorensen AH, Torsvik VL, Torsvik T, Poulsen LK, Ahring BK (1997) Appl Environ Micro-

biol 63:3043110. Macnaughton SJ, O’Donnell AG, Embley TM (1994) Microbiology 140:2859111. Hahn D, Amann R, Zeuer J (1993) Appl Environ Microbiol 59:1709112. Zarda B, Amann R, Wallner G, Schleifer KH (1991) J Gen Microbiol 137:2823113. Urdea MS, Warner BD, Runnign JA, Stempien CM, Horn T (1988) NAR 16:1988114. Amann RI, Zarda B, Stahl DA, Schleifer K-H (1992) Appl Environ Microbiol 58:3007115. Fischer K, Hahn D, Amann R, Daniel O, Zeuer J (1995) Can J Microbiol 41:666116. Hodson RW, Dustman WA, Garg RP, Moran MA (1995) Appl Environ Microbiol 61:4074117. Tyagi S, Kramer FR (1996) Nature Biotech 14:303118. Schonfield P, Pell AN, Krause DO (1997) Appl Environ Microbiol 63:1143119. Lange M, Tolker-Nielsen T, Molin S, Ahring BK (2000) Appl Environ Microbiol 66:1796120. Tolker-Nielsen T, Holmstrom K, Boe L, Molin S (1998) Mol Microbiol 27:1099121. Tumbula DL, Whitman WB (1999) Mol Microbiol 33:1122. Cohen-Kupiec R, Blank C, Leigh JA (1997) PNAS 94:1316123. Beneke S, Bestgen H, Klein A (1995) Mol Gen 248:225124. Metcalf WW, Zhang JK,Apolinario E, Sowers KR,Wolfe RS (1997) Proc Natl Acad Sci USA

94:2626125. Lange M, Ahring B (2002) in preparation126. Metcalf WW (1999) Meth Microbiol 29:277127. Holmes M, Pfeifer F, Dyallsmith M (1994) Gene 146:117128. Fukuzaki S, Nishio N (1997) J Ferment Bioeng 84:382129. Phillips-Jones MK (1993) FEMS Microbiol Lett 106:265130. Hansen MC, Palmer RJ, Udsen C, White DC, Molin S (2001) Microbiol 147:1383131. Errampalli D, Leung K, Cassidy MB, Kostrzynska M, Blears M, Lee H, Trevors JT (1999) J

Microbiol Meth 35:187132. Gomes CM, Teixeira M (1998) Biochem Biophys Res Commun 249:23133. Howgrave-Graham AR, Macario AL, Wallis FM (1997) J Appl Microbiol 83:587134. Macario AL, Conway de Macario E (1993) Biomass Bioenergy 5:203135. Schmidt JE, Ahring BK (1996) Biotech Bioeng 49:229136. Zellner G, Feuerhake E, Jordening HJ, Macario AL, Conway de Macario E (1995) Appl

Microbiol Biotech 43:566137. Harper SR, Pohland FG (1997) Water Sci Technol 36:33138. Warikoo V, McInerney MJ, Robinson JA, Suflita JM (1996) Appl Environ Microbiol 62:26139. Wu WM, Jain MK, Hickey RF, Zeikus JG (1996) Biotech Bioeng 52:404140. Oerther DB, de los Reyes F, Hernandez M, Raskin L (1999) FEMS Microbiol Ecol 29:129141. de los Reyes F, Oerther DB, Hernandez M, Raskin L (1998) Water Sci Technol 37:485142. Hernandez M, Jenkins D, Beaman BL (1994) Water Sci Technol 29:249143. Wagner M, Amann R, Lemmer H, Schleifer KH (1993) Appl Environ Microbiol 59:1520144. Nielsen PH, Andreasen K, Lee N, Wagner M (1999) Water Sci Technol 39:1145. Nielsen PH,Andreasen K,Wagner M, Blackall LL, Lemmer H, Seviour RJ (1998) Water Sci

Technol 37:423146. Lee N, Nielsen PH, Andreasen KH, Juretschko S, Nielsen JL, Schleifer KH, Wagner M

(1999) Appl Environ Microbiol 65:1289147. Ouverney CC, Fuhrman JA (1999) Appl Env Microbiol 65:1746148. Nielsen PE, Egholm M, Berg RH, Buchardt O (1991) Science 254:1497149. Prescott AM, Fricker CR (1999) Mol Cell Probe 13:261150. Perry-O’Keefe H, Yao XW, Coull JM, Fuchs M, Egholm M (1996) Proc Natl Acad Sci USA

93:14670151. Sawata S, Kai E, Ikebukuro K, Iida T, Honda T, Karube I (1999) Biosensors & Bioelectron

14:397


152. Araujo JC, Brucha G, Campos JR, Vazoller RF (2000) Water Sci Technol 41:69153. Ruseska I, Robbins J, Osterton JW (1982) Oil Gas J 8:253154. Raskin L, Rittmann BE, Stahl DA (1996) Appl Environ Microbiol 62:3847155. Amann RI, Stromley J, Devereux R, Key R, Stahl DA (1992) Appl Environ Microbiol

58:614156. Dubourguier HC,Prensier G,Albagnac G (1988) In: Lettinga G,Zehnder A-JB,Grotenhuis

J-TC, Pol L-WH (eds), GASMAT-workshop. p 18, Puduc Wageningen, Lunteren, Nether-lands

157. Lettinga G (1996) Water Sci Technol 33:85158. van Lier JB, Rebac S, Lens P, van Bijnen F, Elferink SH, Stams AM, Lettinga G(1997) Water

Sci Technol 36:317159. Fang HHP, Kwong TS (1995) Environ Tech 16:13160. van Lier JB, Boersma F, Debets MMWH, Lettinga G (1994) Water Sci Technol 30:251161. van Lier JB, Sanz-Martin JL, Lettinga G (1996) Water Res 30:199162. Angenent LT, Dague RR (1996) In: The 10th forum for applied biotechnology. Brugge,

Belgium163. Barber WP, Stuckey DC (1998) Environ Tech 19:489164. Banik GC, Ellis TG, Dague RR (1997) In: The 8th international conference on an aerobic

digestion. Sendai, Japan 2:548165. Lettinga G, van Velsen AF, Hobma SW, de Zeeuw W, Klapwijk A (1980) Biotech Bioeng

22:699166. Lettinga G (1995) Antonie van Leeuwenhoek 67:3167. Parkash, Gupta SK (1997) In: The 8th international conference of anaerobic digestion.

Sendai, Japan 2:500168. Cheng SS, Ho CY, Wu JH (1997) In: The 8th international conference on anaerobic diges-

tion. Sendai, Japan 2:445169. Razo-Flores E, Luijten M, Donlon B, Lettinga G, Field J (1997) In: The 8th international

conference on anaerobic digestion. Sendai, Japan 2:429170. Demirer GN, Speece RE (1997) In: The 8th international conference of anaerobic diges-

tion. Sendai, Japan 2:316171. He YL, Geng XL, Yang SH (1995) Bioresource Tech 51:213172. Donlon BA, Razo-Flores E, Lettinga G, Field JA (1996) Biotech Bioeng 51:439173. Ahring BK, Christiansen N, Mathrani IM, Hendriksen HV, Macario AJL, Conway de

Macario E (1992) Appl Environ Microbiol 58:3677174. Christiansen N, Ahring BK (1996) Int J Syst Bact 46:442175. Sandberg M, Ahring BK (1992) Appl Microbiol Biotech 36:800176. Ten Brummeler E, Hulshoff Pol LW, Dolfing J, Lettinga G, Zehnder AJB (1985) Appl

Environ Microbiol 49:1472177. Verstraete W, Debeer D, Pena M, Lettinga G, Lens PN (1996) World J Microbiol Biotech

12:221178. Zheng D, Raskin L (2000) Mol Ecol 39:246179. Zheng D, Alm EW, Stahl DA, Raskin L (1996) Appl Environ Microbiol 62:4504180. Sekiguchi Y, Kamagata Y, Syutsubo K, Ohashi A, Harada H, Nakamura K (1998) Micro-

biology-UK 144:2655181. Angenent LT, Zheng D, Sung S, Raskin L (2000) Water Sci Technol 41:35182. Hofman-Bang J, Ahring B (2002) Appl Environ Microbiol, submitted183. Sekiguchi Y, Kamagata Y, Nakamura K, Ohashi A, Harada H (1999) Appl Environ Micro-

biol 65:1280184. Zheng D (1999) PhD thesis. University of Illinios at Urbana-Champaign, USA185. Zinder SH (1993) In: Ferry JG (ed), Methanogenesis: ecology, physiology, biochemistry

and genetics. p 128, Chapman and Hall, New York186. Hansen KH, Ahring BK, Raskin L (1999) Appl Environ Microbiol 65:4767187. Zellner G, Macario A-JL, Conway de Macario E (1997) FEMS Microbiol Ecol 22:295188. Zhao H, Yang D, Woese CR, Bryant MP (1993) Int J Syst Bact 43:278189. Svetlitshnyi V, Rainey F, Wiegel J (1996) Int J Syst Bact 46:1131


190. Lorowitz WH, Zhao H, Bryant MP (1989) Int J Syst Bacteriol 39:122191. Ahring BK (1995) Antonie van Leeuwenhoek 67:91192. Raskin L, Zheng D, Griffin ME, Stroot PG, Misra P (1995) Antonie van Leeuwenhoek

68:297193. Wu WM, Hickey R, Zeikus JG (1991) Appl Environ Microbiol 57:3438194. Hristova K, Mau M, Zheng D, Raskin L (2000) Environ Microbiol 2:143195. Godon JJ, Zumstein E, Dabert P, Habouzit F, Moletta R (1997) Appl Environ Microbiol

63:2802196. Ahring B, Ibrahim A, Mladenovska Z (2001) Water Res 35:2446

Received: March 2002


Author Index Volumes 51-81 A u t h o r I n d e x Volumes 1 - 5 0 see Volume 50

Ackermann, I.-U. see Babel, W.: Vol. 71, p. 125 Adam, W.., Lazarus, M., Saha-Mi~ller, C. R., Weichhold, 0., Hoch, U., Hi, ring, 19., Schreier, (.Y.:

Biotransformations with Peroxidases. Vol. 63, p. 73 Ahring, B. K.: Perspectives for Anaerobic Digestion. Vol. 81, p. 1 Ahring, B. K. see Gavala, H. N.: Vol. 81, p. 57 Ahring, B. K. see Hofman-Bang, l.: Vol. 81, p. 151 Akhtar, M., Blanchette, R. A., Kirk, T. K.: Fungal Delignification and Biochemical Pulping of

Wood. Vol. 57, p. 159 Allan, J. V., Roberts, S. M., Williamson, N. M.: Polyamino Acids as Man-Made Catalysts. Vol. 63,

p. 125 Allington, R. W. see Xie, S.: Vol. 76, p. 87 AI-Rubeai, M.: Apoptosis and Cell Culture Technology. Vol. 59, p. 225 AI-Rubeai, M. see Singh, R. P.: Vol. 62, p. 167 Alsberg, B. K. see Shaw, A. D.: Vol. 66, p. 83 Angelidaki, L see Gavala, H. N.: Vol. 81, p. 57 Antranikian, G. see Ladenstein, R.: Vol. 61, p. 37 Antranikian, G. see Miiller, R.: Vol. 61, p. 155 Archelas, A. see Orru, R. V. A.: Vol. 63, p. 145 Argyropoulos, D. S.: Lignin. Vol. 57, p. 127 Arnold, E H., Moore, ]. C.: Optimizing Industrial Enzymes by Directed Evolution. Vol. 58, p. 1 Autuori, E, Farrace, M. G., Oliverio, S., Piredda, L., Piacentini, G.: "Tissie" Transglutaminase

and Apoptosis. Vol. 62, p. 129 Azerad, R.: Microbial Models for Drug Metabolism. Vol. 63, p. 169

Babel, W.., Ackermann, ].-U., Breuer, U.: Physiology, Regulation and Limits of the Synthesis of Poly(3HB).Vol. 71, p. 125

Baj15ai, P., Bajpai, P. 1(.: Realities and Trends in Emzymatic Prebleaching of Kraft Pulp. Vol. 56, p. 1

Bajpai, P., Bajpai, P. K.: Reduction of Organochlorine Compounds in Bleach Plant Effluents. Vol. 57, p. 213

Bajpai, P. K. see Bajpai, P.: Vol. 56, p. 1 Bajpai, P. K. see Bajpai, P.: Vol. 57, p. 213 Banks, M. K., Schwab, P., Liu, B., gulakow, P. A., Smith, J. &, Kim, R.: The Effect of Plants on the De-

gradation and Toxicity of Petroleum Contaminants in Soil: A Field Assessment. Vol. 78, p. 75 Barut, M. see Strancar, A.: Vol. 76, p. 49 Bdrzana, E.: Gas Phase Biosensors. Vol. 53, p. 1 Bathe, B. see Pfefferle, W.: Vol. 79, p. 59 Bazin, M. J. see Markov, S.A.: Vol. 52, p. 59 Bellgardt, K.-H.: Process Models for Production of 13-Lactam Antibiotics. Vol. 60, p. 153 Beppu, T.: Development of Applied Microbiology to Modern Biotechnology in Japan. Vol. 69,

p.41 Berovic, M. see Mitchell, D.A.: VoL 68, p. 61

206 Author Index Volumes 51-81

Beyeler, W., DaPra, E., Schneider, K.: Automation of Industrial Bioprocesses. Vol. 70, p. 139 Beyer, M. see Seidel, G.: Vol. 66, p. 115 Bhatia, P. K., Mukhopadhyay, A.: Protein Glycosylation: Implications for in vivo Functions and

Thereapeutic Applications. Vol. 64, p. 155 Bisaria, V.S. see Ghose, T.K.: Vol. 69, p. 87 Blanchette R. A. see Akhtar, M.: Vol. 57, p. 159 Bocker, H., Knorre, W..A.: Antibiotica Research in lena from Penicillin and Nourseothricin to

Interferon. Vol. 70, p. 35 de Bont, ].A.M. see van der Werf, M. l-: Vol. 55, p. 147 van den Boom, D. see lurinke, C.: Vol. 77, p. 57 Brainard, A. P. see Ho, N. W. Y.: Vol. 65, p. 163 Brazma, A., Sarkans, U., Robinson, A., Vilo, ]., Vingron, M., Hoheisel, ]., Fellenberg, K.: Micro-

array Data Representation, Annotation and Storage. Vol. 77, p. 113 Breuer, U. see Babel, W.: Vol. 71, p. 125 Broadhurst, D. see Shaw, A. D.: Vol. 66, p. 83 Bruckheimer, E. M., Cho, S. H., Sarkiss, M., Herrmann, ]., McDonell, T. ].: The Bcl-2 Gene

Family and Apoptosis. Vol 62, p. 75 Briiggemann, O.: Molecularly Imprinted Materials - Receptors More Durable than Nature Can

Provide. Vol. 76, p. 127 Bruggink, A., Straathof, A. I. l., van der Wielen, L. A. M.: A 'Fine' Chemical Industry for Life

Science Products: Green Solutions to Chemical Challenges. Vol. 80, p. 69 Buchert, ]. see Suurn~kki, A.: Vol. 57, p. 261 Bungay, H. R. see Miihlemann, H. M.: Vol. 65, p. 193 Bungay, H. R., Isermann, H. P.: Computer Applications in Bioprocessin. Vol. 70, p. 109 Biissow, K. see Eickhoff, H.: Vol. 77, p. 103 Byun, S. Y. see Choi, I. W.: Vol. 72, p. 63

Cabral, ]. M. S. see Fernandes, P.: Vol. 80, p. 115 Cantor, C. R. see Jurinke, C.: Vol. 77, p. 57 Cao, N.J. see Gong, C. S.: Vol. 65, p. 207 Cao, N. ]. see Tsao, G. T.: Vol. 65, p. 243 Carnell, A. ].: Stereoinversions Using Microbial Redox-Reactions. Vol. 63, p. 57 Cen, P., Xia, L.: Production of Cellulase by Solid-State Fermentation. Vol. 65, p. 69 Chang, H. N. see Lee, S. Y.: Vol. 52, p. 27 Cheetham, P.S.I.: Combining the Technical Push and the Business Pull for Natural

Flavours.Vol. 55, p. 1 Chen, Z. see Ho, N. W. Y.: Vol. 65, p. 163 Cho, S. H. see Bruckheimer, E. M.: Vol. 62, p. 75 Cho, G.H. see Choi, J.W.: Vo172, p. 63 Choi, I. see Lee, S.Y.: Vol. 71, p. 183 Choi, ]. W.., Cho, G.H., Byun, S.Y., Kim, D.-I.: Integrated Bioprocessing for Plant Cultures.

Vol. 72, p. 63 Christensen, B., Nielsen, ].: Metabolic Network Analysis - A Powerful Tool in Metabolic

Engineering. Vol. 66, p. 209 Christians, E C. see McGall, G.H.: Vol. 77, p. 21 Chui, G. see Drmanac, R.: Vol. 77, p. 75 Ciaramella, 34. see van der Oost, J.: Vol. 61, p. 87 Contreras, B. see Sablon, E.: Vol. 68, p. 21 Conway de Macario, E., Macario, A. ]. L.: Molecular Biology of Stress Genes in Methanogens:

Potential for Bioreactor Technology. Vol. 81, p. 95 Cordero Otero, R.R. see Hahn-H~igerdal, B.: Vol. 73, p. 53 Cornet, ].-E, Dussap, C. G., Gros, ].-B.: Kinetics and Energetics of Photosynthetic Micro-

Organisms in Photobioreactors. Vol. 59, p. 153 da Costa, M. S., Santos, H., Galinski, E.A.: An Overview of the Role and Diversity of

Compatible Solutes in Bacteria and Archaea. Vol. 61, p. 117

Author Index Volumes 51- 81 207

Cotter, T. G. see McKenna, S. L.: Vol. 62, p. 1 Croteau, R. see McCaskill, D.: Vol. 55, p. 107

Danielsson, B. see Xie, B.: Vol. 64, p. 1 DaPra, E. see Beyeler, W.: Vol. 70, p. 139 Darzynkiewicz, Z., Traganos, E: Measurement of Apoptosis. Vol. 62, p. 33 Dave),, 14. M. see Shaw, A. D.: Vol. 66, p. 83 Dean, ]. E D., LaFayette, P. R., Eriksson, K.-E. L., Merkle, S. A.: Forest Tree Biotechnolgy. Vol. 57, p. 1 Debabov, V.. G.: The Threonine Story. Vol. 79, p. 113 Demain, A.L., Fang, A.: The Natural Functions of Secondary Metabolites. Vol. 69, p. 1 Diaz, R. see Drmanac, R.: Vol. 77, p. 75 Dochain, D., Perrier, M.: Dynamical Modelling, Analysis, Monitoring and Control Design for

Nonlinear Bioprocesses. Vol. 56, p. 147 Drmanac, R., Drmanac, S., Chui, G., Diaz, R., Hou, A., ]in, H., ]in, P., Kwon, S., Lacy, S., Moeur,

B., Shafto, ]., Swanson, D., Ukrainczyk, T., Xu, C., Little, D.: Sequencing by Hybridization (SBH): Advantages, Achievements, and Opportunities. Vol. 77, p. 75

Drmanac, S. see Drmanac, R.: Vol. 77, p. 75 Du, ]. see Gong, C. S: Vol. 65, p. 207 Du, ]. see Tsao, G. T.: Vol. 65, p. 243 Dueser, M. see Raghavarao, K. S. M. S.: Vol. 68, p. 139 Dussap, C. G. see Cornet J.-F.: Vol. 59, p. 153 Dutta, N. N. see Ghosh, A. C.: Vol. 56, p. 111 Dutta, N. N. see Sahoo, G. C.: Vol. 75, p. 209 Dynesen, J. see McIntyre, M.: Vol. 73, p. 103

Eggeling, L., Sahm, 14., de Graafi A. A.: Quantifying and Directing Metabolite Flux: Application to Amino Acid Overproduction. Vol. 54, p. 1

Eggeling, L. see de Graaf, A.A.: Vol. 73, p. 9 Eggink, G., see Kessler, B.: Vol. 71, p. 159 Eggink, G., see van der Walle, G. J. M.: Vol. 71, p. 263 Ehrlich, H. L. see Rusin, P.: Vol. 52, p. 1 Eickhoff, H., Konthur, Z., Lueking, A., Lehrach, H., Walter, G., Nordhoff, E., Nyarsik, L., Bfissow,

K.: Protein Array Technology: The Tool to Bridge Genomics and Proteomics. Vol. 77, p. 103

Elias, C. B., ]oshi, J. B.: Role of Hydrodynamic Shear on Activity and Structure of Proteins. Vol. 59, p. 47

Elling, L.: Glycobiotechnology: Enzymes for the Synthesis of Nucleotide Sugars. Vol. 58, p. 89

Eriksson, K.-E. L. see Kuhad, R. C.: Vol. 57, p. 45 Eriksson, K.-E. L. see Dean, J. F. D.: Vol. 57, p. 1

Faber, K. see Orru, R. V. A.: Vol. 63, p. 145 Fang, A. see Demain, A.L.: Vol. 69, p. 1 Farrace, M. G. see Autuori, F.: Vol. 62, p. 129 Farrell, R. L., Hata, K., Wall, M. B.: Solving Pitch Problems in Pulp and Paper Processes. Vol.

57, p. 197 Fellenberg, K. see Brazma, A.: Vol. 77, p. 113 Fernandes, P., Prazeres, 1). M. E, Cabral, J. M. S.: Membrane-Assisted Extractive Biocon-

versions. Vol. 80, p. 115 Ferro, A., Gefell, M., Kjelgren, R., Lipson, D. S., Zollinger, N., Jackson, S.: Maintaining Hydraulic

Control Using Deep Rooted Tree Systems. Vol. 78, p. 125 Fiechter, A.: Biotechnology in Switzerland and a Glance at Germany. Vol. 69, p. 175 Fiechter, A. see Ochsner, U. A.: Vol. 53, p. 89 Flechas, E W., Latady, M.: Regulatory Evaluation and Acceptance Issues for Phytotechnology

Projects. Vol. 78, p. 171

208 Author Index Volumes 51- 81

Foody, B. see Tolan, J. S.: Vol. 65, p. 41 Frdchet, J. M.J. see Xie, S.: Vol. 76, p. 87 Freitag, R., H6rvath, C.: Chromatography in the Downstream Processing of Biotechnological

Products. Vol. 53, p. 17 Furstoss, R. see Orru, R. V. A.: Vol. 63, p. 145

Galinski, E.A. see da Costa, M.S.: Vol. 61, p. 117 Gardonyi, M. see Hahn-Higerdal, B.: Vol. 73, p. 53 Gatfield, L L.: Biotechnological Production of Flavour-Active Lactones. Vol. 55, p. 221 Gavala, H. N., Angelidaki, L, Ahring, B. K.: Kinetics and Modeling of Anaerobic Digestion

Process. Vol. 81, p. 57 GefeU, M. see Ferro, A.: Vol. 78, p. 125 Gerneiner, P. see Stefuca, V.: Vol. 64, p. 69 Gerlach, S. R. see Schfigerl, K.: Vol. 60, p. 195 Ghose, T.K., Bisaria, ES.: Development of Biotechnology in India. Vol. 69, p. 71 Ghosh, A. C., Mathur, R. K., Dutta, N. N.: Extraction and Purification of Cephalosporin

Antibiotics. Vol. 56, p. 111 Ghosh, P.. see Singh, A.: Vol. 51, p. 47 Gilbert, R. ]. see Shaw, A. D.: Vol. 66, p. 83 Gill, R.T. see Stephanopoulos, G.: Vol. 73, p. 1 Gomes, J., Menawat, A. S.: Fed-Batch Bioproduction of Spectinomycin. Vol. 59, p. 1 Gong, C. S., Cao, N.J., Du, J., Tsao, G. T.: Ethanol Production from Renewable Resources.

Vol. 65, p. 207 Gong, C. S. see Tsao, G. T.: Vol. 65, p. 243 Goodacre, R. see Shaw, A. D.: Vol. 66, p. 83 de Graaf,, A. A., Eggeling, L., Sahm, H.: Metabolic Engineering for L-Lysine Production by

Corynebacterium glutarnicurn. Vol. 73, p. 9 de Graaf, A. A. see Eggeling, L.: Vol. 54, p. 1 de Graaf, A. A. see Weuster-Botz, D.: Vol. 54, p. 75 de Graaf,, A. A. see Wiechert, W.: Vol. 54, p. 109 Grabley, S., Thiericke, R.: Bioactive Agents from Natural Sources: Trends in Discovery and

Application. Vol. 64, p. 101 Griengl, H. see Johnson, D. V.: Vol. 63, p. 31 Gros, ].-B. see Larroche, C.: Vol. 55, p. 179 Gros, J.-B. see Cornet, J. E: Vol. 59, p. 153 Guenette M. see Tolan, J. S.: Vol. 57, p. 289 Gutman, A. L., Shapira, M.: Synthetic Applications of Enzymatic Reactions in Organic

Solvents. Vol. 52, p. 87

Hahn-Hiigerdal, B., Wahlbom, C.E, Gdrdonyi, M., van Zyl, W..H., Cordero Otero, R.R., J6nsson, L.J.: Metabolic Engineering of Saccharomyces cerevisiae for Xylose Utilization. Vol. 73, p. 53

Haigh, J.R. see Linden, J.C.: Vol. 72, p. 27 Hall, D. O. see Markov, S. A.: Vol. 52, p. 59 Hall, P. see Mosier, N. S.: Vol. 65, p. 23 Harnmar, F.: History of Modern Genetics in Germany. Vol. 75, p. 1 Hannenhalli, S., Hubbell, E., Lipshutz, R., Pevzner, P. A.: Combinatorial Algorithms for Design

of DNA Arrays. Vol. 77, p. 1 Haralampidis, D., Trojanowska, M., Osbourn, A. E.: Biosynthesis of Triterpenoid Saponins in

Plants. Vol. 75, p. 31 Hiiring, D. see Adam, E.: Vol. 63, p. 73 Harvey, N. L., Kumar, S.: The Role of Caspases in Apoptosis. Vol. 62, p. 107 Hasegawa, S., Shimizu, K.: Noninferior Periodic Operation of Bioreactor Systems. Vol. 51,

p. 91 Hata, K. see Farrell, R. L.: Vol. 57, p. 197

Author Index Volumes 51- 81 209

van der Heijden, R. see Memelink, I.: VoL 72, p. 103 Hein, S. see Steinbiichel, A.: Vol. 71, p. 81 Hembach, T. see Ochsner, U. A.: Vol. 53, p. 89 Henzler, H.-1.: Particle Stress in Bioreactor. Vol. 67, p. 35 Herrmann, ]. see Bruckheimer, E. M.: Vol. 62, p. 75 Hill, D. C., Wrigley, S. K., Nisbet, L. ].: Novel Screen Methodologies for Identification of New

Microbial Metabolites with Pharmacological Activity. Vol. 59, p. 73 Hiroto, M. see Inada, Y.: Vol. 52, p. 129 Ho, N. W. Y., Chen, Z., Brainard, A. lq Sedlak, M.: Successful Design and Development of

Genetically Engineering Saccharomyces Yeasts for Effective Cofermentation of Glucose and Xylose from Cellulosic Biomass to Fuel Ethanol. Vol. 65, p. 163

Hoch, U. see Adam, W.: Vol. 63, p. 73 Hofman-Bang, ]., Zheng, D., Westermann, P., Ahring, B. K., Raskin, L.: Molecular Ecology of

Anaerobic Reactor Systems. Vol. 81, p. 151 Hoheisel, 1. see Brazma, A.: Vol. 77, p. 113 Hol16,1., Kralovdnsky, U.P.: Biotechnology in Hungary. Vol. 69, p. 151 Honda, H., Liu, C., Kobayashi, T.: Large-Scale Plant Micropropagation. Vol. 72, p. 157 H6rvath, C. see Freitag, R.: Vol. 53, p. 17 Hou, A. see Drmanac, R.: Vol. 77, p. 75 Hubbell, E. see Hannenhalli, S.: Vol. 77, p. 1 Huebner, S. see Mueller, U.: Vol. 79, p. 137 Hummel, W.: New Alcohol Dehydrogenases for the Synthesis of Chiral Compounds.Vol. 58,p. 145

Ikeda, M.: Amino Acid Production Processes. Vol. 79, p. 1 Imamoglu, S.: Simulated Moving Bed Chromatography (SMB) for Application in Bio-

separation. Vol. 76, p. 211 Inada, Y.., Matsushima, A., Hiroto, M., Nishimura, H., Kodera, Y.: Chemical Modifications of

Proteins with Polyethylen Glycols. Vol. 52, p. 129 Irwin, D. C. see Wilson, D. B.: Vol. 65, p. 1 lsermann, H.P. see Bungay, H. R.: Vol. 70, p. 109 Iyer, P. see Lee, Y. Y.: Vol. 65, p. 93

Jackson, S. see Ferro, A.: Vol. 78, p. 125 ]ames, E., Lee, I. M.: The Production of Foreign Proteins from Genetically Modified Plant

Cells. Vol. 72, p. 127 ]effries, T. W.., Shi, N.-Q.: Genetic Engineering for Improved Xylose Fementation by Yeasts.

Vol. 65, p. 117 ]endrossek, D.: Microbial Degradation of Polyesters. Vol. 71, p. 293 ]enne, M. see Schmalzriedt, S.: Vol. 80, p. 19 ]in, H. see Drmanac, R.: Vol. 77, p. 75 ]in, P. see Drmanac, R.: Vol. 77, p. 75 Johnson, D. V., Griengl, H.: Biocatalytic Applications of Hydroxynitrile. Vol. 63, p. 31 Johnson, E. A., Schroeder, W.. A.: Microbial Carotenoids. Vol. 53, p. 119 ]ohnsurd, S.C.: Biotechnolgy for Solving Slime Problems in the Pulp and Paper Industry.

Vol. 57, p. 311 ]Onsson, L. ]. see Hahn-H~igerdal, B.: Vol. 73, p. 53 ]oshi, ]. B. see Elias, C. B.: Vol. 59, p. 47 ]urinke, C., van den Boom, D., Cantor, C. R., K~ster, H.: The Use of MassARRAY Technology for

High Throughput Genotyping. Vol. 77, p. 57

Kaderbhai, N. see Shaw, A. D.: Vol. 66, p. 83 Karanth, N. G. see Krishna, S. H.: Vol. 75, p. 119 Karthikeyan, R., Kulakow, P. A.: Soil Plant Microbe Interactions in Phytoremediation. Vol. 78,

p. 51 Kataoka, M. see Shimizu, S.: Vol. 58, p. 45


Kataoka, M. see Shimizu, S.: Vol. 63, p. 109 Katzen, R., Tsao, G. T.: A View of the History of Biochemical Engineering. Vol. 70, p. 77 Kawai, 17.: Breakdown of Plastics and Polymers by Microorganisms. Vol. 52, p. 151 Kell, D. B. see Shaw, A. D.: Vol. 66, p. 83 Kessler, B., Weusthuis, R., Witholt, B., Eggink, G.: Production of Microbial Polyesters: Fer-

mentation and Downstream Processes. Vol. 71, p. 159 Khosla, C. see McDaniel, R.: Vol. 73, p. 31 Kieran, P.M., Malone, D.M., MacLoughlin, P.E: Effects of Hydrodynamic and Interfacial

Forces on Plant Cell Suspension Systems. Vol. 67, p. 139 Kijne, ]. 144. see Memelink, J.: Vol. 72, p. 103 Kirn, D.-L see Choi, J.W.: Vol. 72, p. 63 Kim, R. see Banks, M. K.: Vol. 78, p. 75 Kim, Y.B., Lenz, R. W.: Polyesters from Microorganisms. Vol. 71, p. 51 Kimura, E.: Metabolic Engineering of Glutamate Production. Vol. 79, p. 37 King, R.: Mathematical Modelling of the Morphology of Streptomyces Species. Vol. 60, p. 95 Kino-oka, M., Nagatome, H., Taya, M.: Characterization and Application of Plant Hairy Roots

Endowed with Photosynthetic Functions. Vol. 72, p. 183 Kirk, T. K. see Akhtar, M.: Vol. 57, p. 159 Kjelgren, R. see Ferro, A.: Vol. 78, p. 125 Knorre, W.A. see Bocker, H.: Vol. 70, p. 35 Kobayashi, M. see Shimizu, S.: Vol. 58, p. 45 Kobayashi, S., Uyama, H.: In vitro Biosynthesis of Polyesters. Vol. 71, p. 241 Kobayashi, T. see Honda, H.: Vol. 72, p. 157 Kodera, E see Inada, Y.: Vol. 52, p. 129 Kolattukudy, P. E.: Polyesters in Higher Plants. Vol. 71, p. 1 Ki~nig, A. see Riedel, K: Vol. 75, p. 81 de Koning, G.]. M. see van der Walle, G. A. M.: Vol. 71, p. 263 Konthur, Z. see Eickhoff, H.: Vol. 77, p. 103 Kossen, N. W.. E: The Morphology of Filamentous Fungi. Vol. 70, p. 1 K6ster, H. see ]urinke, C.: Vol. 77, p. 57 Krabben, P., Nielsen, ].: Modeling the Mycelium Morphology of Penicilium Species in Sub-

merged Cultures. Vol. 60, p. 125 Kralovdnszky, U.P. see Hol16, ].: Vol. 69, p. 151 Kriimer, R.: Analysis and Modeling of Substrate Uptake and Product Release by Procaryotic

and Eucaryotik Cells. Vol. 54, p. 31 Kretzmer, G.: Influence of Stress on Adherent Cells. Vol. 67, p. 123 Krieger, N. see Mitchell, D.A.: Vol. 68, p. 61 Krishna, S. H., Srinivas, N. D., Raghavarao, K. S. M. S., Karanth, N. G.: Reverse Micellar

Extraction for Downstream Processeing of Proteins/Enzymes. Vol. 75, p. 119 Kuhad, R. C., Singh, A., Eriksson, K.-E. L.: Microorganisms and Enzymes Involved in the

Degradation of Plant Cell Walls. Vol. 57, p. 45 Kuhad, R. Ch. see Singh, A.: Vol. 51, p. 47 Kulakow, P. A. see Karthikeyan, R.: Vol. 78, p. 51 KuIakow, P. A. see Banks, M. K.: Vol. 78, p. 75 Kumagai, H.: Microbial Production of Amino Acids in Japan. Vol. 69, p. 71 Kurnar, S. see Harvey, N. L.: Vol. 62, p. 107 Kunze, G. see Riedel, K.: Vol. 75, p. 81 Kwon, S. see Drmanac, R.: Vol. 77, p. 75

Lacy, S. see Drmanac, R.: Vol. 77, p. 75 Ladenstein, R., Antranikian, G.: Proteins from Hyperthermophiles: Stability and Enzamatic

Catalysis Close to the Boiling Point of Water. Vol. 61, p. 37 Ladisch, C. M. see Mosier, N. S.: Vol. 65, p. 23 Ladisch, M. R. see Mosier, N. S.: Vol. 65, p. 23 LaFayette, P. R. see Dean, J. F. D.: Vol. 57, p. 1

Author Index Volumes 51-81 211

Lammers, E, Scheper, T.: Thermal Biosensors in Biotechnology. Vol. 64, p. 35 Larroche, C., Gros, ].-B.: Special Transformation Processes Using Fungal Spares and

Immobilized Cells. Vol. 55, p. 179 Latady, M. see Flechas, F. W.: Vol. 78, p. 171 Lazarus, M. see Adam, W.: Vol. 63, p. 73 Leak, D. ]. see van der Weft, M. l.: Vol. 55, p. 147 Lee, ].M. see James, E.: Vol. 72, p. 127 Lee, S. Y., Chang, H. N.: Production of Poly(hydroxyalkanoic Acid). Vol. 52, p. 27 Lee, S. Y., Choi, ].: Production of Microbial Polyester by Fermentation of Recombinant

Microorganisms. Vol. 71, p. 183 Lee, Y. Y., Iyer, P., Torget, R. W.: Dilute-Acid Hydrolysis of Lignocellulosic Biomass. Vol. 65, p. 93 Lehrach, H. see Eickhoff, H.: Vol. 77, p. 103 Lenz, R. W. see Kim, Y. B.: Vol. 71, p. 51 Licari, P. see McDaniel, R.: Vol. 73, p. 31 Lievense, L. C., van't Riet, K.: Convective Drying of Bacteria II. Factors Influencing Survival.

Vol. 51, p. 71 Linden, J.C., Haigh, J.R., Mirjalili, N., Phisaphalong, M.: Gas Concentration Effects on

Secondary Metabolite Production by Plant Cell Cultures. Vol. 72, p. 27 Lipshutz, R. see Hannenhalli, S.: Vol. 77, p. 1 Lipson, D. S. see Ferro, A.: Vol. 78, p. 125 Little, D. see Drmanac, R.: Vol. 77, p. 75 Liu, B. see Banks, M. K.: Vol. 78, p. 75 Liu, C. see Honda, H.: Vol. 72, p. 157 Lueking, A. see Eickhoff, H.: Vol. 77, p. 103

MacLoughlin, P.F. see Kieran, P. M.: Vol. 67, p. 139 Macario, A. ]. L. see Conway de Macario, E.: Vol. 81, p. 95 Malone, D.M. see Kieran, P. M.: Vol. 67, p. 139 Malone),, S. see Miiller, R.: Vol. 61, p. 155 Mandenius, C.-F.: Electronic Noses for Bioreactor Monitoring. Vol. 66, p. 65 Markov, S. A., Bazin, M. ]., Hall,/9, O.: The Potential of Using Cyanobacteria in Photobio-

reactors for Hydrogen Production. Vol. 52, p. 59 Marteinsson, V. T. see Prieur, D.: Vol. 61, p. 23 Marx, A. see Pfefferle, W.: Vol. 79, p. 59 Mathur, R. K. see Ghosh, A. C.: Vol. 56, p. 111 Matsushima, A. see Inada, Y.: Vol. 52, p. 129 Mauch, K. see Schmalzriedt, S.: Vol. 80, p. 19 McCaskill, D., Croteau, R.: Prospects for the Bioengineering of Isoprenoid Biosynthesis.

Vol. 55, p. 107 McDaniel, R., Licari, P., Khosla, C.: Process Development and Metabolic Engineering for the

Overproduction of Natural and Unnatural Polyketides. VoL 73, p. 31 McDonell, T.. ]. see Bruckheimer, E. M.: Vol. 62, p. 75 McGall, G.H., Christians, F.C.: High-Density GeneChip Oligonucleotide Probe Arrays. Vol. 77,

p.21 McGovern, A. see Shaw, A. D.: Vol. 66, p. 83 McGowan, A. ]. see McKenna, S. L.: Vol. 62, p. 1 Mclntyre, M., Mfiller, C., Dynesen, ]., Nielsen, 1.: Metabolic Engineering of the Aspergillus. Vol.

73, p. 103 Mclntyre, T.: Phytoremediation of Heavy Metals from Soils. Vol. 78, p. 97 McKenna, S. L., McGowan, A. ]., Cotter, T. G.: Molecular Mechanisms of Programmed Cell

Death. Vol. 62, p. 1 McLoughlin, A. 1.: Controlled Release of Immobilized Cells as a Strategy to Regulate

Ecological Competence of Inocula. Vol. 51, p. 1 Memelink, ]., Kijne, ]. W., van der Heijden, R., Verpoorte, R.: Genetic Modification of Plant

Secondary Metabolite Pathways Using Transcriptional Regulators. Vol. 72, p. 103


Menachem, S. B. see Argyropoulos, D. S. : Vol. 57, p. 127 Menawat, A. S. see Gomes J.: Vol. 59, p. 1 Menge, M. see Mukerjee, J.: Vol. 68, p. 1 Merkle, S. A. see Dean, J. F. D.: Vol. 57, p. 1 Mirjalili, N. see Linden, J.C.: Vol. 72, p. 27 Mitchell, D.A., Berovic, M., Krieger, N.: Biochemical Engineering Aspects of Solid State Bio-

processing. Vol. 68, p. 61 MiJckel, B. see Pfefferle, W.: Vol. 79, p. 59 Moeur, B. see Drmanac, R.: Vol. 77, p. 75 Moore, J. C. see Arnold, F. H.: Vol. 58, p. 1 Moracci, M. see van der Oost, J.: Vol. 61, p. 87 Mosier, N.S., Hail, P., Ladisch, C.M., Ladisch, M.R.: Reaction Kinetics, Molecular Action, and

Mechanisms of Cellulolytic Proteins. Vol. 65, p. 23 Mithlemann, H.M., Bungay, H.R.: Research Perspectives for Bioconversion of Scrap Paper.

Vol. 65, p. 193 Mukherjee, ]., Menge, M.: Progress and Prospects of Ergot Alkaloid Research. Vol. 68, p. 1 Mukhopadhyay, A.: Inclusion Bodies and Purification of Proteins in Biologically Active

Forms. Vol. 56, p. 61 Mukhopadhyay, A. see Bhatia, P.K.: Vol. 64, p. 155 Mueller, U., Huebner, S.: Economic Aspects of Amino Acids Production. Vol. 79, p. 137 Mfiller, C. see Mclntyre, M.: Vol. 73, p. 103 Miiller, R., Antranikian, G., Maloney, S., Sharp, R.: Thermophilic Degradation of Environ-

mental Pollutants. Vol. 61, p. 155

Nagatome, H. see Kino-oka, M.: Vol. 72, p. 183 Nag),, E.: Three-Phase Oxygen Absorption and its Effect on Fermentation. Vol. 75, p. 51 Necina, R. see Strancar, A.: Vol. 76, p. 49 Nielsen, ]. see Christensen, B.: Vol. 66, p. 209 Nielsen, ]. see Krabben, P.: Vol. 60, p. 125 Nielsen, ]. see McIntyre, M.: Vol. 73, p. 103 Nisbet, L.]. see Hill, D.C.: Vol. 59, p. 73 Nishimura, H. see Inada, Y.: Vol. 52, p. 123 Nordhoff, E. see Eickhoff, H.: Vol. 77, p. 103 Nyarsik, L. see Eickhoff, H.: Vol. 77, p. 103

Ochsner, U.A., Hembach, T., Fiechter, A.: Produktion of Rhamnolipid Biosurfactants. Vol. 53, p. 89

O'Connor, R.: Survival Factors and Apoptosis: Vol. 62, p. 137 Ogawa, ]. see Shimizu, S.: Vol. 58, p. 45 Ohta, H.: Biocatalytic Asymmetric Decarboxylation. Vol. 63, p. 1 Oliverio, S. see Autuori, F.: Vol. 62, p. 129 van tier Oost, ]., Ciaramella, M., Moracci, M., Pisani, EM., Rossi, M., de Vos, W..M.: Molecular

Biology of Hyperthermophilic Archaea. Vol. 61, p. 87 Orlich, B., Schomiicker, R.: Enzyme Catalysis in Reverse Micelles. Vol. 75, p. 185 Orru, R. V.A., Archelas, A., Furstoss, R., Faber, K.: Epoxide Hydrolases and Their Synthetic

Applications. Vol. 63, p. 145 Osbourn, A. E. see Haralampidis, D.: Vol. 75, p. 31 Oude Elferink, S. I. W. H. see Stares, A. J. M.: Vol. 81, p. 31

Paul, G.C., Thomas, C.R.: Characterisation of Mycelial Morphology Using Image Analysis. Vol. 60, p. 1

Pettier, M. see Dochain, D.: Vol. 56, p. 147 Pevzner, P. A. see Hannenhalli, S." Vol. 77, p. 1 Pfefferle, W.., Mi~ckel, B., Bathe, B., Marx, A.: Biotechnological Manufacture of Lysine. Vol. 79,

p. 59


Phisaphalong, M. see Linden, J. C.: Vol. 72, p. 27 Piacentini, G. see Autuori, E: Vol. 62, p. 129 Piredda, L. see Autuori, E: Vol. 62, p. 129 Pisani, EM. see van der Oost, J.: Vol. 61, p. 87 Podgornik, A. see Strancar, A.: Vol. 76, p. 49 Podgornik, A., Tennikova, T.B.: Chromatographic Reactors Based on Biological Activity. Vol.

76, p. 165 Pohl, M.: Protein Design on Pyruvate Decarbox-ylase (PDC) by Site-Directed Mntagenesis.

Vol. 58, p. 15 Poirier, I(.: Production of Polyesters in Transgenic Plants. Vol. 71, p. 209 Pons, M.-N., Vivier, H.: Beyond Filamentous Species. Vol. 60, p. 61 Pons, M.-N., Vivier, H.: Biomass Quantification by Image Analysis. Vol. 66, p. 133 Prazeres, D. M. E see Fernandes, P.: Vol. 80, p. 115 Prieur, D., Marteinsson, V. T.: Prokaryotes Living Under Elevated Hydrostatic Pressure. Vol. 61,

p. 23 Prior, A. see Wolfgang, J.: Vol. 76, p. 233 Pulz, 0., Scheibenbogen, K.: Photobioreactors: Design and Performance with Respect to Light

Energy Input. Vol. 59, p. 123

Raghavarao, K. S. M. S., Dueser, M., Todd, R: Multistage Magnetic and Electrophoretic Extraction of Ceils, Particles and Macromolecules. Vol. 68, p. 139

Raghavarao, K. S. M. S. see Krishna, S. H.: Vol. 75, p. 119 Ramanathan, K. see Xie, B.: Vol. 64, p. 1 Raskin, L. see Hofman-Bang, J.: Vol. 81, p. 151 Reuss, M. see Schmalzriedt, S.: Vol. 80, p. 19 Riedel, K., Kunze, G., K6nig, A.: Microbial Sensor on a Respiratory Basis for Wastewater

Monitoring. Vol. 75, p. 81 van't Riet, K. see Lievense, L. C.: Vol. 51, p. 71 Roberts, S. M. see Allan, I. V.: Vol. 63, p. 125 Robinson, A. see Brazma, A.: Vol. 77, p. 113 Rock, S. A.: Vegetative Covers for Waste Containment. Vol. 78, p. 157 Roehr, M.: History of Biotechnology in Austria. Vol. 69, p. 125 Rogers, R L., Shin, H. S., Wang, B.: Biotransformation for L-Ephedrine Production. Vol. 56, p. 33 Rossi, M. see van der Oost, J.: Vol. 61, p. 87 Rowland, ]. ]. see Shaw, A. D.: Vol. 66, p. 83 Roychoudhury, P. K., Srivastava, A., Sahai, V.: Extractive Bioconversion of Lactic Acid. Vol. 53,

p. 61 Rusin, R, Ehrlich, H. L.: Developments in Microbial Leaching - Mechanisms of Manganese

Solubilization. Vol. 52, p. 1 Russell, N.J.: Molecular Adaptations in Psychrophilic Bacteria: Potential for Biotechnological

Applications. Vol. 61, p. 1

Sablon, E., Contreras, B., Vandamme, E.: Antimicrobial Peptides of Lactic Acid Bacteria: Mode of Action, Genetics and Biosynthesis. Vol. 68, p. 21

Sahai, V. see Singh, A.: Vol. 51, p. 47 Sahai, V. see Roychoudhury, P. K.: Vol. 53, p. 61 Saha-Mi~ller, C. R. see Adam, W.: Vol. 63, p. 73 Sahm, H. see Eggeling, L.: Vol. 54, p. 1 Sahm, H. see de Graaf, A.A.: Vol. 73, p. 9 Sahoo, G. C., Dutta, N. N.: Perspectives in Liquid Membrane Extraction of Cephalosporin

Antibiotics: Vol. 75, p. 209 Saleemuddin, M.: Bioaffinity Based Immobilization of Enzymes. Vol. 64, p. 203 Santos, H. see da Costa, M.S.: Vol. 61, p. 117 Sarkans, U. see Brazma, A.: Vol. 77, p. 113 Sarkiss, M. see Bruckheimer, E. M.: Vol. 62, p. 75


Sauer, U.: Evolutionary Engineering of Industrially Important Microbial Phenotypes. Vol. 73, p. 129

Scheibenbogen, K. see Pulz, O.: Vol. 59, p. 123 Scheper, T.. see Lammers, F.: Vol. 64, p. 35 Schmalzriedt, S., ]enne, M., Mauch, K., Reuss, M.: Integration of Physiology and Fluid

Dynamics. Vol. 80, p. 19 Schneider, K. see Beyeler, W.: Vol. 70, p. 139 Schomiicker, R. see Orlich, B.: Vol. 75, p. 185 Schreier, R: Enzymes and Flavour Biotechnology. Vol. 55, p. 51 Schreier, P. see Adam, W.: Vol. 63, p. 73 Schroeder, W. A. see Johnson, E. A.: Vol. 53, p. 119 Schftgerl, K., Gerlach, S. R., Siedenberg, D.: Influence of the Process Parameters on the

Morphology and Enzyme Production of Aspergilli. Vol. 60, p. 195 Schiigerl, K. see Seidel, G.: Vol. 66, p. 115 Schiigerl, K.: Recovery of Proteins and Microorganisms from Cultivation Media by Foam

Flotation. Vol. 68, p. 191 Schiigerl, K.: Development of Bioreaction Engineering. Vol. 70, p. 41 Schumann, W..: Function and Regulation of Temperature-Inducible Bacterial Proteins on the

Cellular Metabolism. Vol. 67, p. 1 Schuster, K. C.: Monitoring the Physiological Status in Bioprocesses on the Cellular Level.

Vol. 66, p. 185 Schwab, P. see Banks, M. K.: Vol. 78, p. 75 Scouroumounis, G. K. see Winterhalter, P.: Vol. 55, p. 73 Scragg, A. 14.: The Production of Aromas by Plant Cell Cukures. Vol. 55, p. 239 Sedlak, M. see Ho, N. W. Y.: Vol. 65, p. 163 Seidel, G., Tollnick, C., Beyer, M., Schi~gerl, K.: On-line and Off-line Monitoring of the

Production of Cephalosporin C by Acremonium Chrysogenum. Vol. 66, p. 1 I5 Shafto, ]. see Drmanac, R.: Vol. 77, p. 75 Shamlou, P. A. see Yim, S. S.: Vol. 67, p. 83 Shapira, M. see Gutman, A. L.: Vol. 52, p. 87 Sharp, R. see Mtiller, R.: Vol. 61, p. 155 Shaw, A. D., Winson, M. K., Woodward, A. M., McGovern, A., Davey, H. M., Kaderbhai, N.,

Broadhurst, D., Gilbert, R. ]., Taylor, ]., Timmins, E. M., Alsberg, B. K., Rowland, ]. ]., Goodacre, R., Kell, 1). B.: Rapid Analysis of High-Dimensional Bioprocesses Using Multivariate Spectroscopies and Advanced Chemometrics. Vol. 66, p. 83

Shi, N.-Q. see leffries, T. W.: Vol. 65, p. 117 Shimizu, K. see Hasegawa, S.: Vol. 51, p. 91 Shimizu, S., Ogawa, ]., Kataoka, M., Kobayashi, M.: Screening of Novel Microbial for the

Enzymes Production of Biologically and Chemically Useful Compounds. Vol. 58, p. 45 Shimizu, S., Kataoka, M.: Production of Chiral C3- and C4-Units by Microbial Enzymes.

Vol. 63, p. 109 Shin, H. S. see Rogers, P. L.: Vol. 56, p. 33 Siedenberg, D. see Schiigerl, K.: Vol. 60, p. 195 Singh, A., Kuhad, R. Ch., Sahai, V., Ghosh, P.: Evaluation of Biomass. Vol. 51, p. 47 Singh, A. see Kuhad, R. C.: Vol. 57, p. 45 Singh, R. P., AI-Rubeai, M.: Apoptosis and Bioprocess Technology. Vol. 62, p. 167 Smith, ]. S. see Banks, M. K.: Vol. 78, p. 75 Sohail, M., Southern, E. M.: Oligonucleotide Scanning Arrays: Application to High-Through-

put Screening for Effective Antisense Reagents and the Study of Nucleic Acid Inter- actions. Vol. 77, p. 43

Sonnleitner, B.: New Concepts for Quantitative Bioprocess Research and Development. Vol. 54, p. 155

Sonnleitner, B.: Instrumentation of Biotechnological Processes. Vol. 66, p. 1 Southern, E. M. see Sohail, M.: Vol. 77, p. 43 Srinivas, N. D. see Krishna, S. H.: Vol. 75, p. 119


Srivastava, A. see Roychoudhury, P. K.: Vol. 53, p. 61 Stafford, D.E., Yanagimachi, K.S., Stephanopoulos, G.: Metabolic Engineering of Indene

Bioconversion in Rhodococcus sp. Vol. 73, p. 85 Stares, A. J. M., Oude Elferink, S. J. W. H., Westermann, P.: Metabolic Interactions Between

Methanogenic Consortia and Anaerobic Respiring Bacteria. Vol. 81, p. 31 Stark, D., yon Stockar, U.: In Situ Product Removal (ISPR) in Whole Cell Biotechnology

During the Last Twenty Years. Vol. 80, p. 149 Stefuca, V., Gemeiner, P.: Investigation of Catalytic Properties of Immobilized Enzymes and

Cells by Flow Microcalorimetry. Vol. 64, p. 69 Steinbiichel, A., Hein, S.: Biochemical and Molecular Basis of Microbial Synthesis of Poly-

hydroxyalkanoates in Microorganisms. Vol. 71, p. 81 Stephanopoulos, G., Gill, R.T.: After a Decade of Progress, an Expanded Role for Metabolic

Engineering. Vol. 73, p. 1 Stephanopoulos, G. see Stafford, D. E.: Vol. 73, p. 85 yon Stockar, U., van der Wielen, L. A. M.: Back to Basics: Thermodynamics in Biochemical

Engineering. Vol. 80, p. 1 yon Stockar, U. see Stark, D.: Vol. 80, p. 149 Straathof,, A. J. ]. see Brnggink, A.: Vol. 80, p. 69 Strancar, A., Podgornik, A., Barut, M., Necina, R.: Short Monolithic Columns as Stationary

Phases for Biochromatography. Vol. 76, p. 49 Suurniikki, A., Tenkanen, M., Buchert, J., Viikari, L.: Hemicellulases in the Bleaching of

Chemical Pulp. Vol. 57, p. 261 Svec, F.: Capillary Electrochromatography: a Rapidly Emerging Separation Method. Vol. 76,

p. 1 Svec, F.. see Xie, S.: Vol. 76, p. 87 Swanson, D. see Drmanac, R.: Vol. 77, p. 75

Taya, M. see Kino-oka, M.: Vol. 72, p. 183 Taylor, ]. see Shaw, A. D.: Vol. 66, p. 83 Tenkanen, M. see Suurn~cki, A.: Vol. 57, p. 261 Tennikova, T. B. see Podgornik, A.: Vol. 76, p. 165 Thiericke, R. see Grabely, S.: Vol. 64, p. 101 Thomas, C. R. see Paul, G. C.: Vol. 60, p. 1 ThOmmes, ].: Fluidized Bed Adsorption as a Primary Recovery Step in Protein Purification.

Vol. 58, p. 185 Timmens, E. M. see Shaw, A. D.: Vol. 66, p. 83 Todd, P. see Raghavarao, K.S.M.S.: Vol. 68, p. 139 Tolan, ]. S., Guenette, M.: Using Enzymes in Pulp Bleaching: Mill Applications.Vol. 57,

p. 289 Tolan, ]. S., Foody, B.: Cellulase from Submerged Fermentation. Vol. 65, p. 41 ToUnick, C. see Seidel, G.: Vol. 66, p. 115 Torget, R. W.. see Lee, Y. Y.: Vol. 65, p. 93 Traganos, F. see Darzynkiewicz, Z.: Vol. 62, p. 33 Trojanowska, M. see Haralampidis, D.: Vol. 75, p. 31 Tsao, D. T.: Overview of Phytotechnologies. Vol. 78, p. 1 Tsao, G. T., Cao, N. ]., Du, ]., Gong, C. S.: Production of Multifunctional Organic Acids from

Renewable Resources. Vol. 65, p. 243 Tsao, G. T. see Gong, C. S.: Vol. 65, p. 207 Tsao, G. T. see Katzen, R.: Vol. 70, p. 77

Ukrainczyk, T. see Drmanac, R.: Vol. 77, p. 75 Uyama, H. see Kobayashi, S.: Vol. 71, p. 241

Vandamme, E. see Sablon, E.: Vol. 68, p. 21 Verpoorte, R. see Memelink, J.: Vol. 72, p. 103


Viikari, L. see Suurn~dci, A.: Vol. 57, p. 261 Vilo, ]. see Brazma, A.: Vol. 77, p. 113 Vingron, M. see Brazma, A.: Vol. 77, p. 113 Vivier, H. see Pons, M.-N.: Vol. 60, p. 61 Vivier, H. see Pons, M.-N.: Vol. 66, p. 133 de Vos, W..M. see van der Oost, J.: Vol. 61, p. 87

Wahlbom, C.E see Hahn-H~igerdal, B.: Vol. 73, p. 53 Wall, M. B. see Farrell, R. L.: Vol. 57, p. 197 van der Walle, G.A.M., de Koning, G.].M., Weusthuis, R.A., Eggink, G.: Properties, Modi-

fications and Applications of Biopolyester. Vol. 71, p. 263 Walter, G. see Eickhoff, H.: Vol. 77, p. 103 Wang, B. see Rogers, P. L.: Vol. 56, p. 33 Weichold, O. see Adam, W.: Vol. 63, p. 73 van der Werf, M. ]., de Bont, ]. A. M. Leak, D. ].: Opportunities in Microbial Biotransformation

of Monoterpenes. Vol. 55, p. 147 Westermann, P. see Hofman-Bang, l-: Vol. 81, p. 151 Westerrnann, P. see Stams, A. J. M.: Vol. 81, p. 31 Weuster-Botz, D., de Graaf, A. A.: Reaction Engineering Methods to Study Intracellular

Metabolite Concentrations. Vol. 54, p. 75 Weusthuis, R. see Kessler, B.: Vol. 71, p. 159 Weusthuis, R.A. see van der Walle, G.J.M.: Vol. 71, p. 263 Wiechert, W.., de Graaf, A. A.: In Vivo Stationary Flux Analysis by 13C-Labeling Experiments.

Vol. 54, p. 109 van der Wielen, L. A. M. see Bruggink, A.: Vol. 80, p. 69 van der Wielen, L. A. M. see yon Stockar, U.: Vol. 80, p. 1 Wiesmann, U.: Biological Nitrogen Removal from Wastewater. Vol. 51, p. 113 Williamson, N. M. see Allan, J. V.: Vol. 63, p. 125 Wilson, D. B., Irwin, D. C.: Genetics and Properties of Cellulases. Vol. 65, p. 1 Winson, M. K. see Shaw, A. D.: Vol. 66, p. 83 Winterhalter, P., Skouroumounis, G. K.: Glycoconjugated Aroma Compounds: Occurence, Role

and Biotechnological Transformation. Vol. 55, p. 73 Witholt, B. see Kessler, B.: Vol. 71, p. 159 Wolfgang, ]., Prior, A.: Continuous Annular Chromatography. Vol. 76, p. 233 Woodley, ]. M.: Advances in Enzyme Technology - UK Contributions. Vol. 70, p. 93 Woodward, A. M. see Shaw, A. D.: Vol. 66, p. 83 Wrigley, S. K. see Hill, D. C.: Vol. 59, p. 73

Xia, L. see Cen, P.: Vol. 65, p. 69 Xie, B., Ramanathan, K., Danielsson, B.: Principles of Enzyme Thermistor Systems: Applica-

tions to Biomedical and Other Measurements. Vol. 64, p. 1 Xie, S., Allington, R. W, Fr~chet, ]. M. ]., Svec, E: Porous Polymer Monoliths: An Alternative to

Classical Beads. Vol. 76, p. 87 Xu, C. see Drmanac, R.: Vol. 77, p. 75

Yanagimachi, K. S. see Stafford, D.E.: Vol. 73, p. 85 Yim, S. S., Shamlou, P.A.: The Engineering Effects of Fluids Flow and Freely Suspended Bio-

logical Macro-Materials and Macromolecules. Vol. 67, p. 83

Zheng, D. see Hofman-Bang, J.: Vol. 81, p. 151 Zhong, ].-].: Biochemical Engineering of the Production of Plant-Specific Secondary

Metabolites by Cell Suspension Cultures. Vol. 72, p. 1 Zollinger, N. see Ferro, A.: Vol. 78, p. 125 van Zyl, W. H. see Hahn-H~igerdal, B.: Vol. 73, p. 53

Subject Index

Acetate 3,4,7 Acetate competition 43 Acetate conversion, syntrophic 5 Acetate oxidation 6 Acetogenesis 68 Acetogenic bacteria 40 - - , hydrogen-producing 4 Acidodenesis 66 Adenosin 5-triphosphate (ATP) 60 A M B R 192 Ammonia 84, 116, 118, 121, 122, 125,

141 Anabolism 60 Anaerobic baffled reactors (ABR)

190 Anaerobic reactor systems,

molecular ecology 151 Anaerobic sequence batch reactors

(ASBR) 190 Anti-stress mechanisms 98, 146 Antibodies 153, 176 Antigenic fingerprinting 143 Aquatic ecosystem 144 Archaea 101, 106, 123 -, methanogenic 4 ARDRA 161 ATP 60 - yield factor 60

Bacterial growth 61 BAG-1 123 Biases 192 Bioethanol 14 Biofilm 132, 137, 146, 180, 189 Biogas yield 2, 12, 15, 17 Biomass yield factor 60 Biomethanation technology 99,109,114,

124 Biopolymers, hydrolysis 62 Bioreactor malfunction 106, 114 Bioreactor technology 124, 143 Butyrate 3, 4 - to acetate 70

Cadmium 114, 116 Cambi Process 16 Carbohydrates, acidogenesis 67 -, hydrolysis 64 Carbon flow 1 Catabolism 60 Cell fixation 171 Cell growth 60 Cell stressors 100, 101 Chaperones 98 Chaperonin complex 109, 124, 145 Chaperonin subunits 109 Chaperonins 102, 104, 123, 124 Chimeric sequences 159 Clone libraries 160 Cloning 159 Clostridium perfringens 176 Co-chaperones 123 Co-digestion 13 Competition 31 - , kinetic 35 - , thermodynamic 36 Cooperation between cells 145 - - molecules 145

Decay rate/coefficient 61 DEHP 19 Dehydrogenation, acetogenic 68 Denitrification 38 Desulfotomaculum spp. 196 DGGE 156, 158,160 Digester failure 71 Digestion, anaerobic 1, 57 -, anaerobic, kinetics/modeling 57 -, co- 13 -, glucose 71, 74 -, thermophilic 7 Digestors 139 Digoxigenin 172 Digoxigenin-labeled probes 164 Dissociation temperature 166 Diversity of stressors 141 Diversity, phylogenetic 159

218 Subject Index

DNA, ribosomal (rDNA) 154 DNA:DNA hybidization 159 Dot blot 113

Effluent quality 1, 23 Esterases 65 Ethanol 14 Expanded granular sludge blanket (ESGB)

190

Fatty acid-fl-oxidizing bacteria, syntrophic 195

Fatty acids, long-chain, anaerobic oxidation 69

- -, volatile (VFA) 3, 4, 68 FISH 168, 193 Fixation 165 Fluorescent in situ hybridization 168

Gel electrophoresis 160 Gene expression 109, 110, 113, 115, 175 Genetic engineering 114, 141,145 Genus 160 GimC 127, 128 Glucose, digestion 71, 74 Glucose fermentation, acidogenesis 80 Gordona 177 Granular consortium 131,134, 135, 145,

146 Granules 133-136 Green fluorescent protein (GFP) 176 Growth kinetics 60 Growth phase 116, 119 Growth rate 61 grpE 109, 110, 112, 116

Ha/dane inhibition 62 Haloferax alicantei 176 Heat shock 119-122, 131 Heat-shockproteins (Hsp) 98, 127 Heavy metals 25 Helper oligonucleotides 165 Hip 123 Histological sections 134, 136 Hop 123 Hsp families 102, 104 hsp-dnaKlocus 110, 115 hsp40(dna]) in methanogens 110, 112, 119,

121 Hsp60chaperonins 108, 123, 124 hsp70(dnaK) in methanogens 106-109,

111,117, 121 Humic substances 157, 168 Hybidization, in situ 169 -,FISH 168, 169, 193, 194

-,membrane 166, 194 -, slot (dot) blot 165 - stringency 170 -, whole-cell 164, 165, 169 Hydrogen 3, 4, 7 - competition 41 -, model for anaerobic digestion - transfer, interspecies 5 Hydrogenation, acetogenic 68 Hydrolases 62 Hydrolysis of biopolymers 62 Hydrolytic constant 63

Immunology techniques 153 Immunotechnology 153 Immunotypes 138, 142 Inhibition, non-competitive 62 Intercellular connective material

132, 135 Internal transcribed spacers 155 Intrinsic stress resistance 146 Iron reduction 39 IST regions 155

K balance 125 K selection 35 Kinetic constants 64-66

77

106,131,

Lamina 131,132, 135, 137 Lipases 65 Lipid biodegradation 65

Manganese reduction 39 Manure, AD technology 15 Manure, high temperature digestion 7 MAR 178 Melting temperature 166 Membrane hybridization 166 Membrane saturation 167 Metabolism 60 Methane 2, 4 Methanobacterium 7 Methanobacterium formicicum 180 Methanobacterium thermoautotrophicum

127, 128, 130, 135 Methanobervibacter 135, 137-140, 142 Methanococcus 7 Methanococcusjannaschii 127, 128,130 Methanogenesis 68 -, acetate/hydrogen 70 -, animal wastes 73, 84 -, inhibition 50 Methanogenic bioreactors 99, 106, 133 Methanogenic biotechnology 144 Methanogenic ecosystems 142, 143

Subject Index 219

Methanogens 4, 7, 100, 101,107, 123, 124, 132, 189, 195

-, acetate-utilizing 5 -, diversity 103, 138-142, 144 -, dynamics in bioreactors 138-143 Methanol 45 Methanosaeta (Methanothrix) 135, 137 Methanosaeta concilli 180, 191,194, 195 Methanosaeta thermophila 195 Methanosarcina 5,22, 176 Methanosarcina barkeri 180 Methanosarcina mazeii 100, 109-122, 131,

132, 137, 139, 140, 144 Methanosarcina thermophila 100, 109, 110,

119, 121,132, 135, 140, 144 Microautoradiography (MAR) 178 Microbial consortia 114, 137, 146 Microbial growth kinetics 60 Microorganisms, fer menting/hydrolyzing

3 Molecular beacon 173 Molecular chaperone machine 102, 104,

109, 114 Molecular chaperones 98, 102-105 Monitoringbioreactors 106, 119 Multi-probe method 173 Multicellular communities 145 Multicellular structures 100, 102, 130-132,

140

NAC 123 Native configuration 98, 105 Nitrogen oxide inhibition 38 Northern blots 113, 118-122

Oligonucleotide probes 163 Osmolytes 125 Oxygen toxicity 37

Packets 131-135, 140 Pathogens 23 PCR 156, 158 -, bias 159 - , i n s i t u 173 -, quantitative 162 -, Taqman 162 Peptide nucleic acids 179 Phosphorylation 60 Phthalate esters 19 Phylogenetic analyses 154 Phylogenetic domains 101,113 Phylogenetic trees 155,157 Phylogeny, molecular 154 PNA 179 Polynucleotide probes 164

Polypeptides, refolding of denatured 104 PPIase 126, 129 Prefoldin 126, 127 Primer extension mapping 133 Probe nesting 165 Probe specificity 164 Product yield factor 60 Propionate 3, 4 - to acetate 70 Propionate-oxidizing bacteria, syntrophic

194 Proteases 64, 129, 130, 145 Proteasome 126, 130 Protein degradation 129, 145 Protein denaturation 98, 105 Proteins, hydrolysis 64 Pulse Power technology 16

r selection 36 Refolding 105, 129 Reporter systems 175 Reverse sample genome probing 169 RFLP 161 RibosomalDatabase Project II 155 RNA, extraction of intact 157 - , ribosomal (rRNA) 154 Robbins device, modified (MRD) 180 rRNA, in vitro transcribed 168 - operons 155 RT-PCR 156, 158, 159, 160 -, in situ 175

Salmonella 23 Saturation constant 63 Shotgun cloning 156, 159 Signal enhancement 172 Slot (dot) blot hybridization 165 Slot blots 117, 119-121 Sludge, granular 180, 190 Sludge stabilization 58 Small heat-shock proteins 126, 128 Solution hybridization technique 173 Sound, response 114, 117 Sound stress 117, 141 Species 159 SSCP 158, 160, 161 Stress 98 Stress genes 102-106, 130

, and proteins 99 , evolution 103, 107

Stress proteins 104, 105 Stress resistance 114, 146 Stress response 105, 106, 114, 121,123,

135 Stressors 100, 106, 138, 141

220 Subject Index

Stringency 166 Substrate consumption rate 61 Sulfate reduction 40 Sulfate-reducing bacteria (SRB)

195 Sulfide inhibition 52 Syntrophic bacteria 190, 192 Syntrophy 4, 189

189, 192,

T-RFLP 156, 160 Taq DNA polymerase 162 Taqman PCR 162 Target accessibility 165, 167 TGGE 156, 158, 160 Thermophilic bioreactors 8, 132, 134 Thermoprotectants 131 Thermosome 104 Thermotolerance 131

Transcription initiation Trigger factor 123 trkA 121,122 TrkA 125, 126

105

Universal phlogenetic tree 155 Universal probe 192 Upflow anaerobic sludge blanket (UASB)

reactors 190, 191

VFA 3, 4, 68 VFA inhibition 59 - -, total 73 - -, un-ionized 68 Volatile fatty acids (VFA)

Wastewater 12 -, composition 75

3, 4, 68

Printing: Druckhaus Berlin-Mitte Binding: Buchbinderei Stein & Lehmann, Bedin

biomethanation i

Documents