14 Aug 2001 17:41 AR AR135K-19.tex AR135K-19.SGM ARv2(2001/05/10) P1: GSR

Annu. Rev. Microbiol. 2001. 55:485–529Copyright c© 2001 by Annual Reviews. All rights reserved

AMMONIA-OXIDIZING BACTERIA: A Model forMolecular Microbial Ecology

George A. Kowalchuk1 and John R. Stephen21Netherlands Institute of Ecology, Centre for Terrestrial Ecology, Boterhoeksestraat 48,P.O. Box 40, 6666 ZG, Heteren, The Netherlands; e-mail: gkowal@cto.nioo.knaw.nl2Horticulture Research International, Crop and Weed Science Department, CV35 9EFWellesbourne, United Kingdom; e-mail: john.stephen@hri.ac.uk

Key Words nitrification, ammonia oxidation,Nitrosospira, Nitrosomonas,β-subclass Proteobacteria

■ Abstract The eutrophication of many ecosystems in recent decades has led to anincreased interest in the ecology of nitrogen transformation. Chemolitho-autotrophicammonia-oxidizing bacteria are responsible for the rate-limiting step of nitrificationin a wide variety of environments, making them important in the global cycling ofnitrogen. These organisms are unique in their ability to use the conversion of ammoniato nitrite as their sole energy source. Because of the importance of this functionalgroup of bacteria, understanding of their ecology and physiology has become a subjectof intense research over recent years. The monophyletic nature of these bacteria interrestrial environments has facilitated molecular biological approaches in studyingtheir ecology, and progress in this field has been rapid. The ammonia-oxidizing bacteriaof the β-subclass Proteobacteria have become somewhat of a model system withinmolecular microbial ecology, and this chapter reviews recent progress in our knowledgeof their distribution, diversity, and ecology.

CONTENTS

INTRODUCTION . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 486Chemolitho-Autotrophic Ammonia Oxidation. . . . . . . . . . . . . . . . . . . . . . . . . . . . . 486Role of Ammonia-Oxidizing Bacteria in Nitrogen Cycling. . . . . . . . . . . . . . . . . . . 488Economic Importance and Utilizationof Ammonia-Oxidizing Bacteria. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 489

Problems Associated with Nitrification andAmmonia-Oxidizing Bacteria. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 490

Isolation of Ammonia-Oxidizing Bacteria from the Environment. . . . . . . . . . . . . . 491DIVERSITY AND DISTRIBUTION OF AMMONIA-OXIDIZINGBACTERIA BASED UPON 16S rDNA ANALYSES . . . . . . . . . . . . . . . . . . . . . . . . 492Phylogeny of Culture Collection. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 492

0066-4227/01/1001-0485$14.00 485

14 Aug 2001 17:41 AR AR135K-19.tex AR135K-19.SGM ARv2(2001/05/10) P1: GSR

486 KOWALCHUK ¥ STEPHEN

Development of Phylogenetically Based Primers and Probes. . . . . . . . . . . . . . . . . 493The Characterization of Ammonia-Oxidizing Bacteriain the Environment Based upon 16S Analyses:What Have We Learned So Far?. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 497

USE OF NON-16S PCR TECHNIQUESFOR STRAIN IDENTIFICATION . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 505The amoAGene as a Marker for Ammonia-Oxidizing Bacteria. . . . . . . . . . . . . . . 505Recovery ofamoA-Like Sequences and Comparisonwith 16S-Based Results. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 506

Use of Other Markers Within the rRNA Operon. . . . . . . . . . . . . . . . . . . . . . . . . . . . 507TRACKING AMMONIA-OXIDIZINGBACTERIA POPULATIONS . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 507Development of Quantitative Techniques TargetingamoAand 16S rDNA . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 507

Comparison of Enrichments with Direct Extraction Results. . . . . . . . . . . . . . . . . . 508Following the Course of Ammonia-OxidizingBacteria in Laboratory Experiments. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 509

FUTURE PERSPECTIVES. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 509Future Advances in Molecular Techniques forβ-SubclassAmmonia-Oxidizing Bacteria. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 509

Techniques for Other Nitrifiers. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 513Modeling and Applications. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 514Use of Ammonia-Oxidizing Bacteria as Indicator Organisms. . . . . . . . . . . . . . . . . 515

CONCLUSION . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 516From Black Box to theE. coli of Molecular Microbial Ecology . . . . . . . . . . . . . . . 516

INTRODUCTION

Chemolitho-Autotrophic Ammonia Oxidation

Chemolithotrophic nitrification is a two-step process, consisting of the conversionof ammonia to nitrite, which is in turn converted to nitrate. These steps are carriedout by two different groups of organisms, the ammonia-oxidizing bacteria and thenitrite-oxidizing bacteria (AOB and NOB), respectively (205, 211). There are noknown autotrophic bacteria that can catalyze the production of nitrate from am-monia. This review focuses on AOB and the development of molecular tools tostudy these organisms, which all have in common the ability to utilize ammoniaas a sole source of energy and carbon dioxide as the chief source of carbon (72).AOB are chemolitho-autotrophs and are obligate aerobes, although some speciesmay be highly tolerant of low oxygen or anoxic environments (26). Further to theirphysiological uniformity, Head et al (66) used 16S rRNA gene sequence data toindicate that all known aerobic AOB isolates were restricted to two evolutionar-ily distinct lineages of the class Proteobacteria. Furthermore, it was demonstratedthat all strains isolated from terrestrial and freshwater environments belonged toa single (monophyletic) evolutionary group within theβ-subclass of the class

14 Aug 2001 17:41 AR AR135K-19.tex AR135K-19.SGM ARv2(2001/05/10) P1: GSR

MOLECULAR ECOLOGY OF AMMONIA OXIDIZERS 487

Proteobacteria, providing evidence of their descent from a single chemolithoau-totrophic ammonia-oxidizing ancestor.

The unique conversion of ammonia to nitrite by AOB involves a compoundreaction, proceeding via hyroxylamine, as depicted in the following two reactions.(A) 2H+ + NH3 + 2e− + O2→ NH2OH+ H2O (B) NH2OH+ H2O→ HONO+ 4e− + 4H+.

It is generally accepted that ammonia (NH3) and not ammonium (NH4+) isused as substrate, and the ammonia/ammonium ratio may therefore affect growth(180, 215). A membrane-bound, multisubunit enzyme, ammonia monooxygenase(AMO), catalyzes the first reaction; a periplasm-associated enzyme, hydroxy-lamine oxidoreductase (HAO), catalyzes the second. Two of the electrons pro-duced in the second reaction are used to compensate for the electron input of thefirst reaction, whereas the other two are passed via an electron transport chain tothe terminal oxidase, thereby generating a proton motive force. (C) 2H+ + 1/2O2

+ 2e− → H2O. Reaction D gives the sum reaction: (D) NH3+ 11/2O2→ HONO+ H2O.

Despite the inability to grow on organic substrates, hydrolysis of urea(Equation E) can be used by some species as a primary ammonia source. (E )(NH2)2C = O+ H2O→ 2NH3+ CO2.

At least two possibilities exist for the anaerobic oxidation of ammonia. The firsthas been demonstrated with typical aerobic ammonia-oxidizing strains, which canreduce nitrite or nitrogen dioxide using hydroxylamine or ammonia as electrondonor (23, 158). Although this process is not thought to support cell growth, itmay provide sufficient energy to allow survival under anaerobic conditions. Asecond anaerobic-autotrophic ammonia-oxidation mechanism, coined the Anam-mox process, has been attributed to a group of organisms within a deep-branchingclade of thePlanctomycetales(156, 178). Reactors dominated by these obligateanaerobes produce N2 from ammonia and nitrate (123) via the intermediates hy-droxylamine and hydrazine. It is not yet known to what extent anaerobic ammoniaoxidation occurs in natural environments. These exciting processes receive onlycursory attention in this chapter, and readers are referred to recent reviews byJetten et al for expert information on these topics (80a, 81).

Distinct from autotrophic ammonia oxidation, ammonia can also be transformedby a number of heterotrophic fungi and bacteria (104, 106, 168). The physiologicalconsequences of heterotrophic nitrification and its importance in various environ-ments remain topics of debate falling outside the scope of this review (47, 71, 88,89, 93, 151, 188, 209). Despite the controversy surrounding techniques used toseparate the effects of autotrophic versus heterotrophic nitrification, such as use ofselective inhibitors such as acetylene (78),15N dilution techniques (14), and otherprocess-oriented approaches (100), it is generally accepted that chemolithotrophsare the primary nitrifiers in many systems. Methane-oxidizing bacteria (MOB) alsopossess the ability to transform ammonia owing to the broad substrate specificityof the particulate methane monooxygenase enzyme (pMMO) (72). The cotrans-formation of ammonia by MOB may have a significant impact on nitrification

14 Aug 2001 17:41 AR AR135K-19.tex AR135K-19.SGM ARv2(2001/05/10) P1: GSR

488 KOWALCHUK ¥ STEPHEN

activities in environments where these organisms are present in high numbers(25, 27).

Role of Ammonia-Oxidizing Bacteria in Nitrogen Cycling

The main natural sources of environmental ammonia are nitrogen fixation andmineralization, with the latter being linked most directly to autotrophic ammoniaoxidation. In addition, the amount of ammonia deposited either directly or indi-rectly because of agricultural and industrial activities has dramatically increasedin recent decades (21, 58, 62).

Ammonia oxidation is the primary step in the oxidation of ammonia to nitrateand is therefore central to the global nitrogen cycle (Figure 1). Ammonia oxidationis thought to be the rate-limiting step for nitrification in most systems, as nitrite israrely found to accumulate in the environment (45, 49, 53, 141). Nitrification canpositively or negatively affect nitrogen retention in a system, depending on theenvironmental conditions. Ammonia is highly volatile and therefore readily lostfrom some systems. In such cases, nitrification may facilitate nitrogen retention

Figure 1 Nitrogen cycling and turnover: 1, atmospheric input; 2, nitrogen fixation; 3, im-mobilization (assimilation); 4, mineralization (ammonification); 5, aerobic ammonia oxida-tion; 6, nitrite oxidation; 7, denitrification; 8, anaerobic ammonia oxidation (annamox); 9,volatilization; 10, nitrate leaching. (After Tietema et al 181a.)

14 Aug 2001 17:41 AR AR135K-19.tex AR135K-19.SGM ARv2(2001/05/10) P1: GSR

MOLECULAR ECOLOGY OF AMMONIA OXIDIZERS 489

by oxidizing ammonia to less volatile nitrogen forms. Nitrification can lead to anet loss of nitrogen from environmental systems via at least three different mecha-nisms. First, the eventual production of nitrate can fuel denitrification, which is themain pathway for the elimination of fixed nitrogen from environmental systems.Second, the product of ammonia oxidation, nitrite, can undergo chemodenitrifica-tion, and it is not known to what extent this process affects nitrogen losses. A thirdimportant route to nitrogen loss is through nitrate leaching from soil. Whereas an-ionic NH4

+molecules bind well to cationic soil particles, NO3− is far more mobile

and therefore readily lost to adjacent watercourses by leaching.The majority of estimates regarding the role of nitrification are based upon mea-

surements of key compounds within the nitrogen cycle (Figure 1). However, suchmeasurements do not reflect the dynamics of nitrogen transformations, thus pro-viding a net result at best. Standing nitrate levels are not indicative of nitrificationlevels owing to nitrate uptake by plants, microbial turnover, immobilization, andnumerous other factors (169). Gross measurements or process-oriented approachesoffer a much better measurement of actual activity.

Economic Importance and Utilizationof Ammonia-Oxidizing Bacteria

Biological ammonia oxidation is the first step in the removal of nitrogen during thetreatment of waste. With the dramatic increase in nitrogenous wastes due to the ex-pansion of animal husbandry, nitrogen-producing industries, and human activities,the handling of nitrogenous wastes has become a critical factor in environmentalmanagement. Ammonia is the predominant inorganic nitrogen source of influentmaterial, and nitrogen loss is typically due to the processes of nitrification, followedby denitrification. The removal of nitrogen from waste treatment is of extremeenvironmental importance, as release of untreated waste can result is devastatingeutrophication of the environment, especially around highly populated areas. Evenin cases where treatment does not lead to successful denitrification, nitrificationhelps avoid environmental contamination with potentially toxic ammonia salts(134). Waste treatment is by far the most important biotechnological applicationof AOB. The treatment of such wastes typically takes place in specialized reac-tors of various design and purpose, whose reliability and cost-effectiveness are ofprime concern to industry and public health (see below).

The broad specificity of the ammonia monooxygenase complex, common to allAOB, often permits the co-oxidation of numerous recalcitrant aliphatic, aromatic,and halogenated molecules (72). An initial oxidation step is often the rate-limitingtransformation in the microbial degradation of many recalcitrant hydrocarbons,and AOB may therefore facilitate bioremediation processes. Limited evidencefrom pure culture studies suggests that AOB may play a role in methane oxidation(175), thus facilitating removal of this important greenhouse gas, but results fromthe field have yet to show significant methane consumption by AOB (25, 27).

AOB may also play an essential role in biofilter systems, which typically containmicrobial consortia that utilize influent substrates to sustain a relatively constant

14 Aug 2001 17:41 AR AR135K-19.tex AR135K-19.SGM ARv2(2001/05/10) P1: GSR

490 KOWALCHUK ¥ STEPHEN

biomass and produce an effluent free from input contaminants. In such systems,AOB may interact closely with other consortium members to consume influentcompounds such as ammonia or CO2, recycle nutrients, and maintain mass bal-ance. Biofilter systems have been used for a variety of purposes to date, generallyinvolving the elimination of odors associated with waste treatment and compost-ing (28) but also for purposes as exotic as providing long-term filtering capacitysuitable for manned spacecraft (84).

Problems Associated with Nitrification andAmmonia-Oxidizing Bacteria

Despite being a natural process, nitrification can have a negative impact on manyenvironments, especially those exposed to elevated levels of nitrogen deposition(141, 186). One of the greatest problems associated with nitrification is nitratecontamination of ground- and freshwater supplies due to nitrate leaching fromagricultural sites. Human and animal consumption of nitrate can lead to healthrisks, and government regulations strictly limit the amount of nitrate permitted indrinking water sources. Nitrate infiltration in freshwater systems may also leadto the eutrophication of such environments, stimulating the growth of some pho-totrophic and/or heterotrophic organisms. This can lead to decreased biodiversityand the creation of anoxic conditions, respectively.

The high productivity of modern agricultural practices is dependent upon theuse of nitrogen-rich fertilizers, which are generated either chemically or via therecycling of organic wastes, typically animal excreta. Nitrogen loss from suchfertilizers, either before or after application, compromises their effectiveness. Ni-trification can lead to nitrogen loss from such systems via two major mechanisms:(a) increasing nitrogen leaching owing to conversion of ammonia to nitrate (seeabove) or (b) linking nitrification with denitrification activities (4).

Incomplete denitrification can also lead to the production of the ozone-depletinggas NO or the greenhouse gas N2O. AOB can also produce these deleterious gaseseither at low levels as by-products of normal nitrification (218) or at high levels (upto 20 mol%) (31, 60, 109, 139) via partial-denitrification processes under reduced-oxygen conditions (39). Several groups of microorganisms contribute significantlyto the production of these trace gases (30), but the relative contributions of differ-ent microbial groups is not yet fully understood. Laboratory experiments suggestthat the role of AOB in NO and N2O production is small, but this has yet to bedemonstrated in situ. Thus, although methods have been proposed to distinguishbetween the various sources of these gases, such as denitrifiers (87), it is unknownto what extent AOB contribute to the global production of these gases. However, itis now known that production of N2O under suboptimal aeration varies drasticallybetween AOB species (83), providing further urgency to the understanding of thefactors regulating population size and structure of environmental AOB.

The process of ammonia oxidation leads to a net acidification of the envi-ronment. Where ammonia deposition and nitrification levels are high, this maycontribute to a lowering of the environmental pH (17). The acidification of forest

14 Aug 2001 17:41 AR AR135K-19.tex AR135K-19.SGM ARv2(2001/05/10) P1: GSR

MOLECULAR ECOLOGY OF AMMONIA OXIDIZERS 491

soils can have a detrimental effect on tree health, and high levels of nitrificationmay intensify problems involving the effects of acid rain. Furthermore, N trans-formations that lead to an increased proton load can lead to the release of metalssuch as aluminum, which can contribute to root damage and forest decline. Suchproblems can be exacerbated by the role of nitrification in changing the relativepools of NH4

+ and NO3−, the most common nitrogen forms taken up by plants.

Some tree species show a strong preference for the uptake of NH4+ over NO3

−,and thus nitrification may affect seed-dormancy breaking, germination, seedlingestablishment, and growth (101). In contrast, many weed species show a prefer-ence for NO3

− and may thrive under conditions of elevated nitrification. AOB maytherefore affect vegetational succession by the relative abundance of inorganic Nsources (101).

The activity of AOB has also been implicated in the corrosion of natural stones,historical monuments, and building materials by generating elevated levels ofnitrous acid (119, 167, 206).

A shift from the use of chlorites to chloramines (quaternary ammonium com-pounds) in the sterilization of drinking-water storage and distribution systems hashighlighted the importance of AOB in maintaining potability of purified water(13, 108, 117, 214). Chloramines penetrate biofilms more efficiently than tradi-tional chlorites, but chloramine degradation results in the release of free ammonia,inducing growth of AOB. Nitrite generated by this activity reacts with remainingchoramine residuals, permitting rapid growth of other bacterial populations, in-cluding coliform bacteria. The concomitant nitrification and growth of coliformsresult in a drastic drop in water quality, such that water must be removed and thesystem resterilized by further treatment, constituting a major financial burden onthe water industry.

Isolation of Ammonia-Oxidizing Bacteria from the Environment

AOB are found in most aerobic environments where ammonia is available throughthe mineralization of organic matter or anthropogenic nitrogen sources, such asfertilizers and waste. Carbon source, CO2, is not thought to be limiting to ammoniaoxidizer activity and growth, in most cases. AOB are ubiquitous in soils, freshwater,and marine environments and have been isolated from these and other habitats (94)(Table 1, follow the Supplemental Material link in the online version of this chapteror at http://www.annualreviews.org/). (Table updates will be available via the fol-lowing internet address: http://www.nioo.knaw.nl/projects/AOB.) Although AOBare generally considered to be aerobic, they have also been isolated or enrichedfrom low-oxygen environments such as brackish water (164), essentially anoxicsoil and sediment layers, and the subsurface portions of building materials (22).Some isolated species tend to be restricted to particular environments (18; seebelow).

AOB pure cultures are typically obtained by picking colonies from a solidmedium (56) or by using dilution methods in liquid culture (2, 157). The selectivemedium used is critical and must be free of organic carbon sources and contain

14 Aug 2001 17:41 AR AR135K-19.tex AR135K-19.SGM ARv2(2001/05/10) P1: GSR

492 KOWALCHUK ¥ STEPHEN

inhibitors of heterotrophic organisms, an ammonia source, and essential traceelements (112). AOB are notorious for their slow growth and low maximum-growth yield, making their isolation and maintenance in pure culture difficult andtime-consuming (157). Also, the culture media may select for particular strainsthat may not necessarily represent the most abundant populations present (85,174, 192, 193). Furthermore, as with other microorganisms, the fraction of viablecells in an environmental sample that is actually amenable to laboratory cultureconditions may be quite small (10, 192).

Culture-dependent techniques such as selective plating (56) and the most prob-able number (MPN) method (6) have been used for the enumeration of AOB;however, such techniques are thought to underestimate actual cell numbers. In ad-dition to medium selectivity and bias, MPN underestimation may also stem frominadequate suspension of cells from solid substrates in the environmental sampleor dispersal of flocks and microcolonies (1, 46, 196). Cell damage due to rigorousdisruption methods or osmotic shock and the possible dependence on inter- orintraspecies interactions for growth may also generate inaccuracies.

The study of pure cultures of AOB has obviously advanced our understand-ing of their physiology; however, the physiological properties of isolated strainsoften cannot explain the nitrification activities found in the environments fromwhich they were recovered. Two main possibilities can be invoked to explain thisdichotomy. (a) Laboratory conditions used for physiological experiments are notrepresentative of the in situ environmental conditions. In order to have the activitiesobserved in the environment, AOB may require other organisms for protection,substrate provisions, or intra- or interspecific signal molecules, or may requireparticular physical properties of the environment (i.e., adhesion, microsites).(b) The organisms typically isolated in pure culture are not representative of thedominant physiological types of AOB in those environments. Clearly, sole relianceon pure culture studies of AOB is insufficient to describe their numbers, diversity,and activities.

DIVERSITY AND DISTRIBUTION OF AMMONIA-OXIDIZINGBACTERIA BASED UPON 16S rDNA ANALYSES

Phylogeny of Culture Collection

Traditionally, all obligate chemolithotrophic nitrifying bacteria have been groupedinto the gram-negative familyNitrobacteraceae(207). However, 16S rRNA geneanalysis has clearly demonstrated multiple lines of descent for both autotrophicammonia- and nitrite-oxidizing bacteria. Currently we know of three distinctgroupings of autotrophic AOBs, namely, obligate aerobes as monophyletic groupswithin the β- andγ -subclasses of the Proteobacteria and anaerobes within thePlanctomycetales (66, 156, 178, 181). To date, such analyses also demonstratethat all aerobic AOB strains isolated from nonmarine ecosystems fall within a

14 Aug 2001 17:41 AR AR135K-19.tex AR135K-19.SGM ARv2(2001/05/10) P1: GSR

MOLECULAR ECOLOGY OF AMMONIA OXIDIZERS 493

monophyletic group in theβ-subclass of the Proteobacteria (Figure 2), the focusof this review. This phylogenetic radiation comprises two genus-level divisions, forwhich the namesNitrosospiraandNitrosomonastake precedent (Nitrosococcusmobilisfalls within theNitrosomonasradiation. It has not as of yet been formallyrenamed and should not be confused withNitrosococcus oceani, a marine AOBwithin theγ -proteobacteria).

It is the monophyletic nature of theβ-subclass ammonia oxidizers that has fa-cilitated the use of molecular biological techniques in the study of their ecology.As all currently identified organisms within theNitrosomonas-Nitrosospiraradia-tion share the common phenotype of being chemolithotrophic ammonia oxidizers,it can be inferred that any 16S rDNA sequence recovered from an environmen-tal sample that falls within that radiation was originally carried by an organismthat also shared that physiology. Furthermore, this observation has led to the de-velopment of molecular probes and polymerase chain reaction (PCR) techniquesspecifically targeting the 16S rRNA and rDNA ofβ-subclass proteobacteriumammonia oxidizers in environmental samples independent of culture analyses.Funding to exploit this potential has been made available due to the ecological andindustrial importance of theβ-subclass proteobacteria AOBs and the inadequacyof available culturing methods to monitor them in an economic and timely fashion.

Development of Phylogenetically Based Primers and Probes

Ideally, PCR primers designed to target theNitrosomonas-Nitrosospiraradia-tion should fulfill three criteria: (a) A sufficient portion of the 16S rDNA geneshould be recovered to allow robust phylogenetic analysis of recovered sequences.(b) They should target all members of the clade. (c) They should not amplify 16SrDNA from organisms outside this clade. Although the first criterion is easily ac-commodated, neither of the latter can be satisfactorily fulfilled by laboratory andin silico analysis of pure cultures. It is not known a priori how well cultured strainsrepresent the target clade, nor do we know how well the closest cultured nontargetrelatives of the clade represent the closest nontarget organisms. Thus, as with allmolecular detection strategies, primer design and testing depends upon the breadthand representation of extant sequence information. Several research groups havedeveloped oligonucleotide primers and probes with various levels of specificitywithin theNitrosomonas-Nitrosospiraradiation. Although such approaches haveled to many new insights, as outlined below, primer and probe development re-mains an ongoing process as novel sequence information recovered from newlyisolated strains and environmental clones continually demands the refinement of16S-dependent detection strategies (183).

The first report of evidence for novelβ-subclass AOB based on 16S rDNAsequence data was provided by McCaig et al (114). In this study, PCR primerswere designed with the aim of recovering a large portion (1.1 kb) of the 16SrRNA gene from allβ-subclass ammonia oxidizers present in a sample. In order toachieve this aim, the primers were designed with a sufficient lack of specificity to

25 Aug 2001 17:10 AR AR135K-19.tex AR135K-19.SGM ARv2(2001/05/10) P1: GSR

494 KOWALCHUK ¥ STEPHEN

14 Aug 2001 17:41 AR AR135K-19.tex AR135K-19.SGM ARv2(2001/05/10) P1: GSR

MOLECULAR ECOLOGY OF AMMONIA OXIDIZERS 495

permit amplification of not only the target organisms but also their closest relativesoutside the ammonia-oxidizing radiation, such that novel bona fide members of theammonia-oxidizing clade were unlikely to be missed. This proved highly effectivein the recovery of ammonia oxidizer-like 16S rDNA from mixed cultures of ma-rine bacteria after selective enrichment and demonstrated the existence of novellines of descent within the genusNitrosomonas. When applied to DNA extractedfrom agricultural soils of varying pH and marine sediments, novel lineages withinthe generaNitrosomonasandNitrosospirawere discovered (174). While the re-laxed specificity of these primers was advantageous in the generation of novelsequence data, clone libraries generated from these soils also contained a highproportion of sequences (up to 70%) that fell outside the target clade, a seriousencumbrance to progress. This factor proved to be even more serious in analysisof some marine communities where none of the sequences screened from somelibraries fell within the ammonia oxidizer radiation (137). This has also been thecase in studying sewage sludge (G.A. Kowlachuk, J.R. Stephen, unpublished ob-servations). Similarly, it is possible to screen environmental clones generated witheubacteria-specific primers for the presence of nitrifier-like 16S rDNA sequences(85). However, this is only feasible in environments where nitrifiers represent asignificant proportion of the entire bacterial community. Where this is not thecase, domain-level PCR and cloning approaches generally provide little informa-tion specific to AOB because of the relative dominance of other bacterial groups(29, 115; for exceptions see 38, 152).

Parallel work has sought to target the sister generaNitrosomonasand Ni-trosospira together or separately, again using PCR primers based on sequencedata from cultured organisms (67, 191, 203). Use of these primers in direct andnested PCR strategies has detected trends in the distribution of ammonia oxidizerpopulations in the environment (see below). Confirmation of product integrityhas not generally involved sequence analysis of amplified products. In some cases,probing of PCR products at AOB-specific internal binding sites has helped confirmthe AOB origin of the recovered products (64, 65). In such studies, the strength ofthe hybridization signal observed should correspond to the amount of PCR productrecovered, thus suggesting that all recovered products originated from the desiredtarget group. Such approaches have not generally focused on the recovery of novelsequences but rather have been informative in a diagnostic capacity, providing yesor no answers to the presence of specific ammonia oxidizer lineages. Critical to thisdiagnostic approach is establishing the sensitivity of the assay, which can dependon a number of factors including environmental contaminants, target to nontargetratio, and type of PCR strategy used (96 and references cited therein).

←−−−−−−−−−−−−−−−−−−−−−−−−−−−−−−−−−−−−−−−−−−−−−−−−−−−−−−−−−Figure 2 Phylogeny of autotrophic nitrifiers. Schematic representation (not to scale)of the known autotrophic nitrifiers based on 16S rRNA gene sequences. Aerobic ammo-nia oxidizers are shown in dark gray, nitrite oxidizers in black, and anaerobic ammoniaoxidizers are striped. Representative species (or isolates) are named for each group orsequence cluster. See Table 1 for details ofβ-subgroup AOBs.

14 Aug 2001 17:41 AR AR135K-19.tex AR135K-19.SGM ARv2(2001/05/10) P1: GSR

496 KOWALCHUK ¥ STEPHEN

During the recovery of sequence information from ammonia oxidizer–like en-vironmental clones, it soon became apparent that clone sequences tended to formdistinct groupings. In the first large-scale sequence analysis of AOB 16S rDNAclones, Stephen et al (174) combined the sequences from 111 cloned gene frag-ments derived from soil and coastal marine sediments with available sequence in-formation from culture strains (Figure 2). This initial characterization has enduredrelatively well as database entries have accrued, although the current RibosomalDatabase Project classification recognizes four lineages within the genusNitro-somonas(113) and does not subdivide the genusNitrosospira. Any such classifica-tion system must be open to change as novel data accrues. For instance, subsequentstudies have shown the groupings proposed by Stephen et al (174) to be valuablein describing the distribution of differentNitrosospirapopulations (95, 97–99,173). However, the group defined as cluster 6 within theNitrosomonasgenus isinsufficiently robust to provide a convenient framework for the classification ofclonedNitrosomonas–like 16S sequences. Most environmentalNitrosomonas–like16S clones fall within this group, and its subdivision has become essential if thisscheme is to remain useful (42, 80, 138, 141b, 165). This has been largely achievedby the work of Purkhold et al (141b), who divided the genusNitrosomonasintofour clusters and also proposed the addition of a fifth subdivision in the genusNitrosospira. This recent paper is essential reading for those involved in the char-acterization of ammonia-oxidizing communities.

A further benefit from this cataloging approach was the opportunity to in-clude environmental clone sequence data in the design of more specific PCRprimers. This was accomplished by Kowalchuk et al (97), who combined theenhanced specificity of a new primer set (coined the CTO primers) with denatur-ing gradient gel electrophoresis (DGGE) (124). This provided a robust methodfor the rapid and simultaneous comparison of multiple samples. However, theenhanced specificity and the technical requirements of DGGE demand a com-promise on the length of sequence that is recoverable. The resulting decreasein the phylogenetic stability of recovered sequences can be counteracted by theapplication of phylogenetically informative oligonucleotide probes to the inter-pretation of DGGE patterns (173). Although this primer set is also prone to therecovery of some non-AOB-like sequences, the interpretation of DGGE resultsat the sequence or hybridization level still allows valid analysis of detected am-monia oxidizer–like populations (95). These developments have led directly tonew understandings of the associations between the 16S rDNA sequence clus-ters and numerous environments and environmental parameters (12, 34, 80, 95,98, 99, 105, 116, 135a, 137, 166, 173). These findings are summarized in Table 2(see the Supplemental Material link in the online version of this chapter or athttp://www.annualreviews.org/). (Table updates will be available via the follow-ing internet address: http://www.nioo.knaw.nl/projects/AOB.)

Reverse array hybridization strategies typically use directly extracted nucleicacids as a “probe” against a battery of oligonucleotides designed for the specificdetection of various taxonomic units (190, 192, 193). Although such strategies areunable to reveal novel diversity outside the oligonucleotide binding sites, they

14 Aug 2001 17:41 AR AR135K-19.tex AR135K-19.SGM ARv2(2001/05/10) P1: GSR

MOLECULAR ECOLOGY OF AMMONIA OXIDIZERS 497

do avoid potential biases of PCR-based techniques, thus providing quantitativelyrobust data. A number of probes specific for various lineages within theβ-subclassammonia oxidizers have been developed for this purpose (74, 122). When using ahierarchical approach, it is possible to not only detect populations that match thespecific oligonucleotide sequences used but also to test the coverage of probes forthe higher-level taxa being examined. Ideally, the summed hybridization signalsof all known lineages within a particular parent taxon should add up to the totalsignal for the parent taxon. Frequently, this does not seem to be the case, providingevidence for the existence of phylogenetic lineages that are not covered by currentlyavailable probes (190, 192, 193). Recovery of sequence information from suchphylogenetic gaps can lead to probe refinement, as has been the case for probestargeting AOB (122). In cases where AOB do not represent a large proportion of thetotal microbial community, it has been necessary to introduce PCR amplificationsteps into hybridization strategies (35). Although addition of PCR step(s) priorto hybridization can increase the sensitivity of slot-blot hybridization assays, thisalso compromises the ability to obtain quantitative results.

Some of the oligonucleotide sequences used for reverse hybridization assayshave sufficient specificity for use as probes for in situ hybridization analyses.Wagner et al (195) presented the first use of in situ hybridization for the visual-ization of specific ammonia oxidizer populations in sewage treatment plants. Asour knowledge of oligonucleotide probes for specific ammonia oxidizer lineageshas grown, so too has the usefulness of such in situ hybridization strategies, whichhave now been implemented in a number of recent studies (e.g., 85, 130, 160, 161).These techniques not only allow for the detection of ammonia oxidizer cells butalso provide information about the structural arrangement of bacterial populations,thus providing insight into microbial interactions. Also, in situ hybridization pro-vides a means of enumerating cells of a particular phylogenetic lineage with respectto other populations. Although the complex 3-D structure of particulate matter andmicrobial aggregates may hamper such efforts, technological improvements suchas the use of confocal laser-scanning microscopy are helping to circumvent some ofthese problems (19). To date, the application of in situ hybridization techniques toAOB has been most effective where AOB represent a large proportion of the totalbacterial community and activity (e.g., sewage treatment plants). However, im-provements in sampling and hybridization techniques may make such techniquesavailable to the study of AOB in other environments.

The Characterization of Ammonia-Oxidizing Bacteriain the Environment Based upon 16S Analyses:What Have We Learned So Far?

SEWAGE, BIOREACTORS, BIOFILMS, AND BIOFILTERS The environments for whichthe ammonia-oxidizing communities have best been characterized are those inter-nal to bioreactors and other systems used for the treatment of nitrogen-rich waste.Reasons for this high level of knowledge are the relatively high cell counts andactivities of AOB in such environments and the financial and social importance

14 Aug 2001 17:41 AR AR135K-19.tex AR135K-19.SGM ARv2(2001/05/10) P1: GSR

498 KOWALCHUK ¥ STEPHEN

of ammonia oxidation as the first step in nitrogen removal from such systems.In contrast to many other environments, the relatively low C-N ratios present inbiological waste-treatment situations provide an environment in which AOB canthrive with respect to many heterotrophic organisms. The presence of nitrifyingbacteria as dominant populations in such systems has made approaches using di-rect rRNA hybridization and in situ hybridization particularly informative. Suchapproaches are particularly robust when combined with cloning and pure cultureisolation techniques in a so-called full-cycle approach (10, 58a, 85), which makesuse of sequence information derived from environmental clones or pure culturesto design specific oligonucleotides, which are in turn used to detect cells corre-sponding to these targets in situ.

The number of recent studies using molecular techniques in sewage treatmentsystems is considerable, and this subject certainly warrants a separate, detailedreview. However, some important findings should be mentioned. In situ hybridiza-tion, using probes for specific AOB and NOB lineages, has been used in a numberof studies to demonstrate the close proximity of AOB and NOB in microcoloniesor cell aggregates (85, 130, 159–161, 196), suggesting a syntrophic association.In combination with microsensor technology, such studies have demonstrated therelative distribution of AOB and NOB populations with respect to microscalegradients of key factors such as oxygen, ammonia, nitrite, and nitrate (40, 41,145). AOB populations tend to reside in the more external regions of particles orbiofilms, where oxygen levels are typically high. NOB cells typically reside inter-nal and adjacent to AOB aggregates, where lower oxygen conditions are detected.NOB aggregates tend to be more diffuse, which may be indicative of oxygen limi-tation (102, 161). AOB populations, and in some cases NOB populations, may alsoreside within deeper, anoxic biofilm layers. Because microsensor measurementssuggest that nitrification is mainly limited to the outer 100–150µm of aggregates(159), the contribution of such cells to total nitrification activities is thought tobe low.

Nitrosomonas europaeastrains are most frequently recovered in pure culturewhen using high-nitrogen-culturing conditions (141), and it has therefore beenassumed that nitrosomonads are the key AOB in high-nitrogen waste treatmentsystems. Indeed, a number of studies have shown the dominance of bacteria oftheN. europaealineage among the AOB in some wastewater treatment systems(130, 161, 195). However,Nitrosomonas marina–like andNitrosospira-like popu-lations are the dominant AOB populations in other systems (85, 160). Furthermore,multiple AOB populations, either within one genus or across genera, have beendetected in some systems, and shifts in the relative proportion of AOB populationshave been observed in response to system conditions (85, 130, 155). Although spa-tial separations have been observed between different AOB populations within abiofilm or bioreactor, the physiological differences between different populationsare unclear (179).

The above groups a number of different systems receiving high nitrogen loadsunder a single heading. However, such treatment systems obviously differ intheir purpose, design, influent material, method, duration of biomass retention,

14 Aug 2001 17:41 AR AR135K-19.tex AR135K-19.SGM ARv2(2001/05/10) P1: GSR

MOLECULAR ECOLOGY OF AMMONIA OXIDIZERS 499

and numerous other factors. The close coupling of ammonia and nitrite oxidationis often desired in such systems, but these activities may be physically separated(68, 179). Elimination of nitrogen in the form of N2 is often a key goal, and oxic-anoxic interfaces are thought to be critical for the efficient coupling of nitrificationand denitification. Methods for tracking denitrifying populations and their ac-tivities (90) should help further our understanding of bioreactor ecology. Also,numerous other organisms within bioreactors contribute to the ecology and per-formance of the system. Analyses targeting nitrifying bacteria have now providedspectacular detail regarding the identification, numbers, activities, and interactionsof different nitrifying populations in a number of different systems. However, itis not yet clear which organisms, if any, are typical for a particular reactor designor performance. Nevertheless, the ability to characterize and quantify nitrifyingpopulations in such controlled reactor systems is important to better predict andsteer the performance of treatment processes (150, 194).

TERRESTRIAL ENVIRONMENTS Soil habitats are highly complex and heteroge-neous. Nevertheless, a number of key environmental factors are thought to affectthe distribution and activities of microbial populations most profoundly. These in-clude soil moisture content, pH, nitrogen input, vegetation, and pollution. Culture-dependent techniques have suggested a dominance ofNitrosospirapopulationsover Nitrosomonasin most terrestrial environments (111). However, it is onlywith the influx of molecular approaches in recent years that we have been able tointerrogate terrestrial samples for specific AO populations to better understand theinfluence of critical environmental factors on this key functional group.

Soil pH no doubt has a profound effect on AOB, but how AOB populationsrespond to changes in the pH, specifically acidification, of terrestrial environ-ments has been an enigma. Under acidic conditions, ammonia (pKa = 9.25) occursmostly in its protonated form, ammonium, and is therefore not suitable as a sub-strate for ammonia monooxygenase (180). AOB isolated from acid soils fail to ni-trify in liquid culture below pH 5.5 (8, 48, 82, 141), nevertheless chemolithotrophicnitrification is commonly observed in acid soils (pHKCl< 4). The strainNitrosos-pira sp. AHB1 adapts to low pH when cell densities are high (44, 46), but it isnot clear how ammonia oxidation can proceed under environmental conditions. Anumber of theories have been proposed, including the existence of acid-tolerantstrains, microsites of relatively neutral pH, use of ureolytic activity as a sourceof ammonia, syntrophic associations with mineralizing organisms, and protectionvia aggregate formation (7, 43, 47, 48, 141). The elimination of nitrite may also beessential to avoid the toxic effects of nitric acid accumulation in acid environments.Nitrifying bacterial aggregates have been observed to contain an ammonia-oxidizer“core” encircled by NOB (Figure 3; 46).

Although the mechanisms for ammonia oxidization in acidic soils are not fullyunderstood, molecular surveys of AOB across pH gradients have identifiedNi-trosospiracluster 2 as containing putatively acidophilic AOB strains (Figure 2).Stephen et al (174) detected a trend from domination byNitrosospiracluster 3in neutral pH agricultural plots towardNitrosospiracluster 2 in acidic pH soil.

14 Aug 2001 17:41 AR AR135K-19.tex AR135K-19.SGM ARv2(2001/05/10) P1: GSR

500 KOWALCHUK ¥ STEPHEN

Figure 3 Transmission electron micrograph taken from De Boer et al (46) of aggre-gatedNitrosospira-like bacteria surrounded byNitrobacter-like bacteria, with arrowspointing out examples of the latter group. This aggregate was observed in a soil en-richment culture nitrifying at pH 4.

Further examination of the same fields, in which PCR-DGGE was coupled withhybridization analysis of resultant profiles, revealed this shift more convincingly(173). The dominance ofNitrosospiracluster 2 amongβ-subclass AOB has alsobeen observed in acidic forest soils (105; G.A. Kowalchuk & W. De Boer, un-published data). Some acidic forest soils contain a vertical pH gradient, wherethe upper-most organic layers are least acidic, and the relative proportion ofNi-trosospiracluster 2 has been observed to increase with soil depth for some suchsoils (G.A. Kowalchuk & W. De Boer, unpublished data). A similar shift wasobserved in calcium-poor coastal dunes (97), where pH decreases with respectto distance from the beach. An increased proportion ofNitrosospiracluster 2

14 Aug 2001 17:41 AR AR135K-19.tex AR135K-19.SGM ARv2(2001/05/10) P1: GSR

MOLECULAR ECOLOGY OF AMMONIA OXIDIZERS 501

was also detected in successional acidic stream valley grasslands, which decreasein pH with respect to time since adoption of “set-aside” land use (99). Theseresults are consistent with observations from culture and activity experiments.The 16S rDNA sequence of the acid soil isolateNitrosospirasp. AHB1 (153)places it withinNitrosospiracluster 2. The increase inNitrosospiracluster 2 de-tection in more acidic stream valley grassland soils also corresponds with a shiftin the optimum pH of nitrification, as determined in soil suspensions (177). Al-thoughNitrosospiracluster 2 appears to be affiliated with acidic soil conditions ina number of disparate terrestrial habitats, a dominance ofNitrosomonas europaea–like populations in acidic Belgian forest soils has also been observed (M. Carnol,G.A. Kowalchuk, W. De Boer, unpublished data). It is not yet clear if and howNitrosospiracluster 2 strains adapt to acidic soil conditions. However, physio-logical analysis of culture strains belonging to this sequence cluster (82, 182)as well as experimental manipulation of soil pH conditions may provide someinsights.

Major efforts have been taken in the past decades to counteract the negativeenvironmental effects of eutrophication and pollution. In attempts to reverse thedecrease in floral diversity resulting from the use of nitrogen-rich fertilizers, manyintensely used fields have been taken out of production in the past half centuryfor conversion to more natural species-rich grasslands by using strategies of im-poverishment management. The secondary plant succession that occurs during im-poverishment management is accompanied by decreases in nitrification, culturablenitrifier numbers, and net N mineralization (131, 177). Although the availability ofammonia is known to affect ammonia-oxidizer pure cultures differently (171), it isonly recently that molecular data have suggested that particular AOB populationsmay be selected under high- versus low-ammonia conditions in soil habitats.

Nitrosospiracluster 3 has been detected as the dominant ammonia-oxidizergroup in a number of neutral pH arable fields receiving fertilizers (34, 98, 99,120, 173, 174). In three separate grassland systems, dominantNitrosospiracluster3 populations have been replaced byNitrosospiracluster 4 in response to pro-longed periods without fertilizer use (34, 98, 99). The observed shifts coincidedwith increased plant diversity and decreased mineralization levels. It has been hy-pothesized that reduced ammonia supply, due to retardation of mineralization, mayselect forNitrosospiracluster 4 populations in fields that have not received fertil-izers for a long time. In such soils, AOB must compete with heterotrophic bacteriaand plant roots for available ammonia (213). It is not known ifNitrosospira-cluster-4 isolates display the physiological properties necessary to win this competition.

In studies of Italian agricultural plots, Hastings et al (64) and Ceccheriniet al (35) detected bothNitrosospira- andNitrosomonas-like sequences in plotsfertilized with swine manure yet only detectedNitrosospira-like sequences in non-fertilized plots. Although the nucleotide sequences of recovered DNA fragmentswere not determined, these results seem to support the more traditional idea thathigh-ammonia conditions stimulate nitrosomonad populations (18).

Most AOB in soils are thought to be associated with soil particles, to displaya very low turnover rate, and to have the ability to survive extended periods of

14 Aug 2001 17:41 AR AR135K-19.tex AR135K-19.SGM ARv2(2001/05/10) P1: GSR

502 KOWALCHUK ¥ STEPHEN

dormancy (18). Aakra et al (1) separated AOB by their level of attachment tosoil using dispersal-density-gradient centrifugation. PCR and sequence analysisof AOB populations recovered from soil suggested that distinctNitrosospira-likepopulations were differentially stimulated by the addition of urea or water, whichaffected adherence to soil particles.

AOB populations in terrestrial environments may also be inhibited by secondaryplant metabolites such as tannins, polyphenolics, and monoterpenes (146–149,201). Plants typical of later successional stages of seminatural grasslands havebeen hypothesized to inhibit nitrification via the production of such allelochemi-cals (148). Although strong evidence for such an inhibition is still lacking (176), itremains difficult to separate possible effects of vegetation from ammonia availabil-ity issues. Plant species composition may also affect AOB populations indirectlyby differentially affecting mineralization patterns, i.e., via production of variablelitter quality and quantity.

The effects of pollution on AOB have received increased attention in recentyears. Ivanova et al (80) surveyed the size, activity, and population structure of AOBin groundwater affected by the plume of a uranium-milling dump site containingU metal in solution, sulfate, and high levels of ammonia. Although molecularanalyses did help establish a link between nitrate levels and AOB numbers, noeffects could be detected with respect to changes in AOB populations present.Nitrosomonas-like sequences were most commonly recovered, and the authorssuggested that groundwater may represent a habitat more similar to freshwaterand freshwater sediment (see below) than to soil. In a study examining the roleof copper contamination, Smit et al (163) detected no copper-related changesin the AOB community composition via amplified ribosomal DNA restrictionanalysis (ARDRA) of PCR-amplified 16S rDNA fragments. Deni & Penninckx(51) observed a decrease in the number of AOB and NOB that could be detectedby PCR in response to hydrocarbon treatment. Changes were observed within thenitrifying populations with respect to hydrocarbon inhibition, but AOB were notfurther characterized.

AQUATIC ENVIRONMENTS A variety of different freshwater habitats have beensurveyed for the presence of AOB populations using molecular techniques. Atpresent, it is difficult to compare results obtained across different freshwater sam-ples because of differences in the techniques employed by different investigators.A number of studies suggested thatNitrosospira-like organisms dominate amongAOB populations in freshwater environments. Hiorns et al (67) first used primersets designed for specific amplification ofNitrosospiraand Nitrosomonas16SrDNA to detect these genera in lake water and sediment. These authors could onlydetect the presence ofNitrosospiraDNA in all samples;Nitrosomonasappearedto be relatively rare. Hastings et al (65) detected onlyNitrosospira-like 16S rDNAfragments in a seasonal study of a eutrophic lake using the same PCR and probingstrategies. Whitby et al (210) also detectedNitrosospirain a freshwater lake system.Nitrosomonas-like populations were also observed with a shift fromN. europaea–like bacteria in coastal sediments toN. eutropha-like bacteria in deeper sediment.

14 Aug 2001 17:41 AR AR135K-19.tex AR135K-19.SGM ARv2(2001/05/10) P1: GSR

MOLECULAR ECOLOGY OF AMMONIA OXIDIZERS 503

Hovanec & DeLong (74) failed to detectN. europaea–like 16S rRNA in freshwateraquaria, even in filters showing high levels of nitrification. These authors suggestedthat this negative result might indicate thatNitrosospira-like populations dominateamong AOB in these freshwater filters. These authors also failed to detect NOB re-lated to the genusNitrobacterand subsequently demonstrated thatNitrospira-likebacteria dominate among the NOB in such filters (75). In contrast to the studiessummarized above, Speksnijder et al (166) observed a dominance ofNitrosomonasureae–like (cluster 6a in Figure 2) sequences in a variety of freshwater lakes andsediments in the Netherlands. De Bie et al (42) detected similar populations withinupper-estuarine samples, essentially a freshwater environment. Interestingly, thespecific primer and probe systems used in previous studies (65, 67, 74, 210) offreshwater habitats would not have detected this group of sequences (141b, 166,183). Thus, it remains difficult to determine if such differences were due to actualdifferences in the AOB populations inhabiting the different lake systems examinedor to differences in the methodologies used. [An up-to-date critique of probe andprimer sets is given in (141b).]

Kowalchuk et al (95) detectedNitrosospiraclusters 3 and 4 in freshwater sedi-ments associated with the aerenchymatous plantGlyceria maxima. No differenceswere detected in the distribution of these sequence clusters between rhizosphere-associated sediments that are exposed to oxygen from the plant and bulk sedimentswhere low oxygen conditions prevail. A survey of other more- or less-oxygenatedsoils also did not reveal any trends with respect to oxygen, even though AOBpopulations in these samples did show differences in their affinity for oxygen andrecovery after exposure to anoxic conditions (26, 95). Interestingly, AOB of thesame shallow lakes were dominated byNitrosomonas-like sequences in the watercolumn and top layers of the sediment (166). Apparently, the water-influencedsediment and the plant-influenced sediment represent different niches that areinhabited by different AOB populations. As in other environments, anoxic/oxicinterfaces are important in coupling nitrification and denitrification activities (24),and N-removal in wastewater treatment efforts relies on this coupling. Rhizo-remediation exploits the anoxic/oxic interface of plant roots to stimulate nitrogenremoval, and a dominance ofNitrosospira-like 16S rDNA sequences has beenobserved in such a system (5).

MARINE ENVIRONMENTS The majority of inorganic nitrogen in marine environ-ment is in the form of nitrate, presumably generated by nitrification. It has long beenheld that autotrophic nitrification in marine environments is primarily carried outby members of theγ -proteobacteria related toNitrosococcus oceani(199, 207)owing to their recovery as pure cultures. However, molecular analysis of ma-rine enrichment cultures casts doubt on this belief becauseβ-subclass AOB 16SrDNA sequences were recovered from all nitrifying-enrichment cultures tested byMcCaig et al (114). This finding was strengthened when, in contrast to their re-sults for freshwater aquaria, Hovanec & DeLong (74) found thatNitrosomonaseuropaea–like bacteria were responsible for 20% of the total 16S rRNA extractedfrom filters of saltwater aquaria. Direct evidence that marine nitrification might be

14 Aug 2001 17:41 AR AR135K-19.tex AR135K-19.SGM ARv2(2001/05/10) P1: GSR

504 KOWALCHUK ¥ STEPHEN

largely dependent onβ-subclass AOB was only provided recently by Nold et al(127), who demonstrated thatβ-subclass AOB outnumberedγ -subclass AOBs inPacific Northwest sediments.

Theβ-subclass AOB populations of marine and freshwater environments ap-pear to be largely distinct from each other, althoughNitrosomonas-like sequencesbelonging to cluster 7 have been detected in both soil and marine environments(64, 80, 137, 210). Cluster 1Nitrosospiraand cluster 5Nitrosomonasappear to belimited to marine environments (12, 116, 137, 174). Interestingly, neither clusterwas detected by molecular analysis of a hypersaline inland lake (202) whereNi-trosomonascluster 7 appeared to be dominant, suggesting that some factor otherthan salinity might affect species boundaries. Members ofNitrosomonascluster 6have also been recovered from marine environments (114, 116, 174), but whetherthese are distinct from the cluster 6 organisms detected in soil, soil water, andfreshwater remains to be determined. No pure cultures ofNitrosospiracluster 1or Nitrosomonascluster 5 have been generated as yet, and it would seem that thismight be a priority in studying marine nitrification. Indeed,Nitrosospiracluster 1may be the dominant type ofβ-subclass AOB in the Arctic ocean (12), and theirdistribution is clearly global (12, 116, 127, 137, 174). Some factors determiningthe relative abundances of the different marineβ-subclass AOB clusters have beensuggested.Nitrosomonascluster 5 appeared to be enriched over clusters 1 and 6 byfish-farm pollutants (116), andNitrosomonascluster 7 enriched overNitrosospiracluster 1 in particle-associated, as opposed to free-living populations (137). Bano &Hollibaugh (12) have observed thatNitrosospiracluster 1 sequences dominatethe β-AOB population of Arctic oceanic waters, whereas a large proportion ofNitrosomonascluster 5 sequences appear in waters influenced by Pacific waters. DeBie et al (42) have witnessed a transition fromNitrosomonascluster 6 sequences,typical of freshwater, in the upper reaches of the eutrophic Schelde estuary tothe marine-associatedNitrosomonascluster 5 in more seaward regions, coincidentwith the major environmental gradients of the estuary.Nitrosospiraclusters 2, 3,and 4 sequences have not yet been recovered from marine environments, and whileit is tempting to speculate from these results that bona fide marine and terrestrialβ-subclass AOB exist, further work will be required before firm conclusions canbe drawn.

COMPOST The use of nitrogen-rich biofertilizers such as compost represent animportant alternative to the use of chemical fertilizers. Ammonia is the mostimportant nitrogenous compound available in composting materials (79). No ther-mophilic AOB are known, and autotrophic ammonia oxidation presumably is neg-ligible at the high temperatures typical of most composting processes. In contrast,thermophilic heterotrophs may play a role in ammonia oxidation during high-temperature composting phases (135). Autotrophic ammonia oxidation may, how-ever, be important before and after composting, possibly affecting the overallnitrogen balance in compost products. A number of AOB populations have been de-tected in various types of composting materials including commercial biofertilizer

14 Aug 2001 17:41 AR AR135K-19.tex AR135K-19.SGM ARv2(2001/05/10) P1: GSR

MOLECULAR ECOLOGY OF AMMONIA OXIDIZERS 505

products (96). Some detected populations did not appear to be highly active, asjudged by RT-PCR of 16S rRNA. Specific composting regimes consistently con-tained the same AOB populations, but no correlations were found between thephysical composition of the composting materials and the AOB populations de-tected. Some AOB populations may be able to survive high-temperature compost-ing steps or passage through the intestinal tracts of domesticated animals, andbothNitrosospira- andNitrosomonas-like 16S rDNA have been detected in swinemanure (35). Thus, compost not only may be subject to nitrogen transformationsinitiated by AOB, but it may also serve to inoculate soils with compost-borne AOB.

OTHER ENVIRONMENTS FOR AMMONIA-OXIDIZING BACTERIA Culture-dependentmethods have been used to demonstrate the presence of AOB in a number ofbuilding materials where their contribution to nitrous acid may lead to corrosion(22, 119, 206). However, molecular analyses of AOB inhabiting such environmentsare still lacking. R¨olleke et al (152) examined potentially corrosive microbialcommunities on ornamental glass and detected aNitrosospira-like sequence as afaint DGGE band in several samples (Table 2). That this population was detectedusing eubacterial primers suggested it represented one of the dominant bacterialpopulations present.

USE OF NON-16S PCR TECHNIQUESFOR STRAIN IDENTIFICATION

The amoA Gene as a Marker for Ammonia-Oxidizing Bacteria

The majority of molecular studies of AOB in the environment have targeted 16SrDNA or rRNA of β-subclass. In addition to the danger of focusing on a singlemolecular marker, these studies assume that members of theγ -subclass are absentor insignificant. This assumption is based on the fact thatγ -subclass ammoniaoxidizers have been detected as isolates only from marine environments. Thus,this assumption is clearly flawed in studies of marine and estuarine environmentsand more subtly flawed in the study of terrestrial systems. The basic justificationfor favoring molecular tools over traditional culture-based methods stems fromthe observation that the majority of environmental microorganisms do not growto pure colonies on standard laboratory media. In other words, the fact that noγ -subclass AOB have been isolated from terrestrial environments is, in fact, verypoor evidence for their unimportance. Nonculturableγ -subclass AOB might beexpected in saline terrestrial environments (85, 202), for instance, and indeed,molecular analyses have now shown thatγ -subclass AOBs can be important innonmarine systems (155).

The potential to exploit a second molecular marker, theα-subunit of ammo-nia monooxygenase, was provided by the first publication of anamoAgene se-quence fromN. europaea(118) derived following purification of the AmoA protein(77). The available database ofamoAgene sequences from culturedβ-AOBs has

14 Aug 2001 17:41 AR AR135K-19.tex AR135K-19.SGM ARv2(2001/05/10) P1: GSR

506 KOWALCHUK ¥ STEPHEN

expanded rapidly since then (Tables 1 and 3). This marker has some interesting po-tential advantages over the use of rRNA markers. First,amoAgenes fromβ- andγ -AOBs can be amplified simultaneously (69, 120, 127, 162). Second, theamoAgeneshares an evolutionary ancestor with the gene encoding theα-subunit of particu-late methane monooxygenase (pmoA) (69, 127). Sufficient sequence homology re-mains to allow bothamoAandpmoAgenes to be targeted simultaneously, allowingan assessment of the relative abundances of autotrophic ammonia and methane oxi-dizing bacteria in situ. TheamoAgenes ofβ-subgroup ammonia oxidizers can alsobe targeted specifically using primers that do not detectγ–proteobacterialamoAor pmoAwhen onlyβ-subclass ammonia oxidizers are under study (154, 172).Furthermore,amoAcontains a greater level of sequence variation than 16S rDNA,providing greater resolution of closely related strains. Findings from this approachare summarized in Table 3 (see the Supplemental Material link in the online versionof this chapter or at http://www.annualreviews.org/). (Table updates will be avail-able via the following internet address: http://www.nioo.knaw.nl/projects/AOB.)

Recovery of amoA-Like Sequences and Comparisonwith 16S-Based Results

The phylogenetic trees predicted by 16S rDNA andamoAsequences from cul-ture strains display a high level of congruence (69, 141b). A complete compar-ison of 16S andamoA-based phylogenies from cultured strains is not yet avail-able. Both 16S andamoAsequences are available for a number ofNitrosospirastrains, and Purkhold et al (141b) have recently provided a near-complete view ofthese for the genusNitrosomonas(see Table 1 at http://www.annualreviews.org/or http://www.nioo.knaw.nl/projects/AOB). The reader is referred to (141b) fora detailed description of the available data. Although these trees are quite con-gruent for both genes, they do suggest important differences in the placement ofstrains closely related toNitrosomonas marina. Such comparisons are only validfor sequence pairs known to have originated from the same organism, somethingthat is not possible when using PCR methods to recover sequences from nucleicacids directly extracted from the environment. Nevertheless, combining the useof amoA-directed primers with 16S rDNA analyses allows direct comparison ofthe β-AOB populations detected by these two markers. Generally, good agree-ment is found between the environmentalβ-AOB populations described by bothamoAand 16S analyses, particularly in the division between the generaNitro-somonasandNitrosospira(33, 73, 80, 120). Although finer-level comparisons canbe problematic, evidence exists that phylogenetic congruence may also be presentbelow the genus level. All AOB populations detected by Laverman et al (105) inacidic forest soils were closely related to the culture strainNitrosospirasp. AHB1,both for 16S rDNA andamoA. Similarly, Speksnijder et al (166; A.G.C.L.Speksnijder, G.A. Kowalchuk, S. De Jong, H.J. Laanbroek, unpublished data)detected a dominantNitrosomonas-like cluster in a number of freshwater habitats,with phylogenetic affinity toNitrosomonas ureaepredicted by both markers. The

14 Aug 2001 17:41 AR AR135K-19.tex AR135K-19.SGM ARv2(2001/05/10) P1: GSR

MOLECULAR ECOLOGY OF AMMONIA OXIDIZERS 507

phylogenetic placement ofamoAsequences from enrichment cultures derived fromthe highest positive dilutions in MPN counts and containing a single AOB 16SrDNA sequence type also supports the finer-level congruence between the treespredicted by these two markers (A.G.C.L. Speksnijder, G.A. Kowalchuk, unpub-lished data). Although there has been some speculation about lateral gene transferof amoAgenes between ribotypes (85, 162), whether it will prove to be a sourceof discrepancy between 16S rDNA- andamoA-predicted phylogenies is unknown.Other important findings derived fromamoA/pmoA-directed PCR strategies in-clude evidence suggesting thatβ-subclass AOB may outnumberγ -subclass AOBin some marine environments (127) and the observation that noγ -subclass AOBhave been detected in soil or sewage sludge as of yet (85, 120).

Use of Other Markers Within the rRNA Operon

The ribosomal internal transcribed spacer regions (ITS) can be highly useful inthe comparison of closely related organisms. Although this region may containshort stretches of coding sequence, typically for tRNAs, much of the sequence hasno coding function, allowing for the rapid accumulation of sequence differences.Highly conserved primers located with the large and small subunit rRNA genes canbe used to amplify the region spanning these variable regions. Aakra et al (2) ap-plied such a strategy to the characterization of several closely relatedNitrosospirastrains and compared ITS results with those obtained by 16S rDNA sequencing.Although both markers predicted comparable phylogenetic relationships betweenthe strains, the ITS-based analysis offered a higher resolution. Interestingly,Ni-trosomonasstrains appeared to have much shorter ITS regions, and it would beinteresting to see if this characteristic may serve as a diagnostic feature for genusdistinction. Although, this marker may be highly useful, it has yet to be combinedwith the use of primers designed to target AOB in complex environments.

The 28S rRNA gene (large subunit, LSU) potentially contains more phylo-genetic information than does the 16S rRNA gene (103). However, fewer AOBsequences are available for this gene, and no molecular strategies for AOB detec-tion have been based on this marker. Nevertheless, as sequencing efforts continue,new possibilities to exploit this gene for AOB-targeted molecular approaches maybecome available.

TRACKING AMMONIA-OXIDIZINGBACTERIA POPULATIONS

Development of Quantitative Techniques TargetingamoA and 16S rDNA

Molecular characterizations of microbial communities are of limited value if theresults cannot be correlated with the number of targets in the environment. PCRprotocols designed for the detection of specific genes or bacterial groups are readily

14 Aug 2001 17:41 AR AR135K-19.tex AR135K-19.SGM ARv2(2001/05/10) P1: GSR

508 KOWALCHUK ¥ STEPHEN

adapted to hold a quantitative aspect. In its simplest form, such adaptation involvesspiking DNA extracted from an environmental sample with a carefully quantifiedinternal standard containing the same PCR primer target sites as the gene of interestand modification (typically a small deletion) of the sequence between the primingsites. This approach is fraught with difficulty if the aim is to place accurate countsof cell numbers in a given sample. Cell lysis efficiency, DNA recovery, damage ofnative target DNA, free DNA in the sample from dead AOB, variable copies of thetarget per genome, the relative efficiency of amplification of the internal standardand target DNA, and primer specificity are all factors hard to quantify. However,there are also many instances in which resistance of the target organisms to culture,or a lack of a priori knowledge of the AOB species composition, may invalidateMPN analysis; low absolute numbers or high background fluorescence may ren-der fluorescence in situ hybridization (FISH) impractical. BothamoAand 16SrDNA have become targets of quantitative PCR (Q-PCR) using the internal stan-dard approach (competitive PCR, C-PCR) for quantification ofβ-subclass AOBin environmental samples (120, 136, 172). The targeting of 16S rDNA has oneadvantage overamoA, in that all knownβ-subclass AOB are thought to carry onlya single 16S rRNA gene per genome, whereasamoAcopy number may vary (usu-ally either 2 or 3). However, none of the 16S rDNA primer sets designed to bespecific to the entireβ-subclass AOB radiation are 100% specific in all samples(85, 95, 98, 120). Therefore, use of 16S rDNA as a target for C-PCR demands thatthe PCR products generated from a given sample lack sequences derived fromcoamplifiable non-AOB source organisms. Some of the primer pairs targetingamoAare totally specific and apparently more sensitive than those targeting 16SrDNA (96, 120). However, given our more limited knowledge of the full diversity ofamoAgenes, it is difficult to predict the coverage ofamoA-directed C-PCR strate-gies. Few studies have comparedβ-subclass AOB enumeration using both MPNand C-PCR, but in grassland soils and compost, the calculated number ofamoAtargets from the competitive PCR assay was 1.5–2 orders of magnitude higher thanthat observed with MPN (96, 98, 99). Such methods are extremely useful in labo-ratory studies in which defined AOB populations are monitored over the course ofcontrolled microcosm environments, but they are clearly more open to criticismin the comparison of AOB population sizes between different environmental sam-ple types. The application of methods combining C-PCR, particularly real-time 5′

nuclease assays, with community profiling techniques may also increase quanti-tative capacity of PCR-based community analyses for AOB (16, 20, 32, 54, 66a,180b). As an alternative to C-PCR, template DNA from a sample can be diluted toextinction, as judged by a negative PCR result, to estimate target numbers. SuchMPN-PCR methods have been applied to estimate both AOB and NOB numbers(50, 51).

Comparison of Enrichments with Direct Extraction Results

Culture-dependent techniques are biased in favor of those organisms that are mostamenable to the culture conditions used, and molecular methods provide the ability

14 Aug 2001 17:41 AR AR135K-19.tex AR135K-19.SGM ARv2(2001/05/10) P1: GSR

MOLECULAR ECOLOGY OF AMMONIA OXIDIZERS 509

to track AOB during the course of selective enrichment. Molecular analyses havenow demonstrated that culture methods often misrepresent the AOB populationsdominant within a sample (e.g., 64, 67, 85, 114, 116). Even in cases where alldetectable AOB populations appear to be cultureable, the relative recovery ofparticular AOB populations may be skewed when comparing culture-based anddirect molecular techniques (98, 99, 163a). These results are consistent with thecommon isolation ofNitrosospiracluster 3 andNitrosomonascluster 7 culturesfrom various environments, even where they are not suspected to be the mostdominant AOB present (18, 141). Furthermore, the relative recovery of differentAOB populations in different culture media can provide clues into their physiology.Kowalchuk et al (99) detected a higher proportion ofNitrosospiracluster 2 inenrichment cultures from acidic grassland soils when using an acid pH medium, ascompared to findings with a pH neutral medium. Similarly, low-ammonia medium(0.3 mM) yielded a higher proportion ofNitrosospiracluster 4 in enrichmentsfrom successional chalk grassland soil than did high-ammonia medium (5 mM),suggesting an adaptation to low-ammonia conditions (98).

Following the Course of Ammonia-OxidizingBacteria in Laboratory Experiments

Although the distribution of AOB populations across environmental gradients hasprovided important insights into the selective forces affecting AOB, controlledexperimental manipulation of environmental conditions is necessary to determineconclusively which factors drive AOB community structure. Community analysistechniques such as DGGE or RFLP are ideal tools for monitoring the response oftarget AOB populations in such experiments (68, 73, 97). Molecular approacheshave now been used to examine induced population shifts among AOB in a numberof experimental settings, including experimental reactors (68, 129, 140), biofilms(130, 155), soil microcosms (51, 172), and chemostats (A. Bollmann & H.J.Laanbroek, unpublished data). Substrate concentration, salinity, C/N ratio, ex-posure to various pollutants, and oxygen tension have been some of the key factorsexamined to date, and such approaches should continue to provide insight into theforces selecting for and against specific AOB populations.

FUTURE PERSPECTIVES

Future Advances in Molecular Techniques for β-SubclassAmmonia-Oxidizing Bacteria

CONTINUED PRIMER AND PROBE DEVELOPMENT AND IMPROVEMENTS OF EXISTING

TECHNOLOGIES As we continue to discover the extent of AOB diversity, PCR andprobing strategies will have to be adjusted to account for new information. Theusefulness of various primers and probes for lineages within theβ-subclass AOBhas recently been evaluated by Ut˚aker & Nes (183) and Purkhold et al (141b), andprimer/probe design should be seen as an ongoing effort. It is important that we

14 Aug 2001 17:41 AR AR135K-19.tex AR135K-19.SGM ARv2(2001/05/10) P1: GSR

510 KOWALCHUK ¥ STEPHEN

continue to remain open to the possibility that populations that are not recognizedby currently available primers and probes may be important in some environments.Recent results demonstrate that unexpected lineages play important roles in envi-ronments in which more well-known species were thought to dominate (85, 160).Molecular surveys of AOB have thus far been limited to a rather narrow rangeof environments and locations. As more environments are surveyed in terms ofthe AOB populations present, we will have a more complete picture of the globaldiversity of these organisms.

Although molecular techniques have changed our perceptions of ammonia-oxidizer ecology, this does not diminish the need for continued efforts and newtechniques in the culturing of AOB (3). Molecular- and culture-based techniquescan be highly complimentary in several ways: (a) Pure cultures are the best resourcefor sequence information that may help in the comparison of phylogenetic treesbased upon 16S rDNA sequences and protein-encoding markers such asamoA.(b) Molecular surveys provide useful information on the locations and conditionsmost useful for the isolation of novel isolates. (c) Molecular strategies providethe best opportunities to follow ammonia-oxidizer populations and activities inexperimental analyses.

One of the major problems with studying AOB in many environments is thatthey represent only a small fraction of the total bacteria present. This has beenovercome by the use of PCR employing specific primer systems, but such analysesalways rely on previous sequence information and may miss target populationsthat do not match the primer set. New methods to subtract dominant bacterialpopulations from PCR mixtures (189) may help in the recovery of additionalnovel diversity within the AOB from environments where they are not numericallydominant.

FURTHER USE OF amoA AND OTHER PROTEIN-ENCODING GENES The use ofamoAhas provided new possibilities for characterizing the diversity and size of AOB.Continued efforts in the sequencing of culture isolates and single ribotype-enrichment cultures are essential to allow for a more complete comparison of phy-logenetic trees based upon 16S rDNA andamoAsequences. Other targets within thegene cluster encoding ammonia monooxygenase are also suitable for such appli-cations (91), and their further development would help validate results based uponamoA. Of specific interest is the detection, characterization, and quantification ofspecific messenger RNA levels foramoAor other key enzyme-encoding genes.Analogous to the study of the NiFe hydrogenase gene ofDesulfovibriosp. (208),combining specificamoAamplification with DGGE may be helpful, but primer lim-itations have hampered the development of this technology. As with other protein-encoding targets, the binding sites foramoA-specific primers necessarily containdegenerate positions, which may result in multiple DGGE bands from a singletemplate because of the incorporation of multiple primer variants into the product(97). Such limitations may be partially circumvented by the use of inosine residuesat appropriate degenerate primer positions (185).amoAsequence variants have

14 Aug 2001 17:41 AR AR135K-19.tex AR135K-19.SGM ARv2(2001/05/10) P1: GSR

MOLECULAR ECOLOGY OF AMMONIA OXIDIZERS 511

also been detected by bisbenzimine derivative agarose gel retardation assays (156)and by the use of terminal restriction fragment length polymorphism (T-RFLP)analysis (73). The extremely high resolution and reproducibility of the T-RFLPmethod may make this an analytical method of choice in field-scale studies ofAOB communities and PCR-amplified environmental gene-fragment mixtures ingeneral (132), but further development of the method is necessary (141b).

AmoA is certainly not the only protein unique to AOB. AOB contain severalother protein-encoding targets potentially suitable for the specific characteriza-tion of AOB populations in environmental samples (121). Development of otherfunctional markers for AOB detection and characterization should provide newperspectives into the diversity of AOB populations.

LINKING IDENTIFICATION WITH ACTIVITY Molecular analyses targeting DNA canprovide some measure of the presence of particular bacterial groups, but they givelittle information with regard to activity. The amount of rRNA within a cell isthought to provide a first approximation of total cell activity (197). Kowalchuket al (96) used reverse transcriptase (RT-) PCR to target AOB-like rRNAs in com-post samples. Comparison with DNA-derived patterns allowed inferences to bedrawn about general activity of detected AOB populations. In situ hybridizationtechniques can also provide direct information regarding rRNA copy number, andcontinued improvements in these techniques and image analysis should continueto provide valuable data. Of course, 16S rRNA does not address the specific ac-tivities of AOB, and the use of RT-PCR strategies targeting specific mRNAs, suchas those coding for AmoA, may provide information more relevant to ammoniaoxidation.

Recently, a number of novel techniques have been applied to couple activitywith phylogenetic identification. Microautoradiography can be coupled with theuse of phylogenetic probes to identify organisms that incorporate labeled substrates(133). This approach has already been applied to the study of activated sludgebacteria (125). Isotope incorporation into DNA can also be used to recover DNAfrom active organisms, which in turn can be subjected to the variety of PCR-based approaches discussed above. Such an approach was recently developed byRadajewski et al (142), who used PCR and cloning to identify organisms fromenvironmental samples that had incorporated13C-labelled CO2, clearly a suitableapproach for the study of autotrophic nitrifiers. Heavy isotope labeling of wholecells or DNA holds great promise for the analysis of active and inactive AOBpopulations under varying conditions.

GENOME SEQUENCING, FUNCTIONAL GENOMICS, AND DNA CHIP TECHNOLOGY

Genome sequencing forN. europaeashould be completed by the time this reviewis printed, and it will no doubt provide new information on the metabolic poten-tial of this bacterium (data is available online from the Joint Genome ResearchInstitute at http://www.jgi.doe.gov). Given the vast potential of current sequence-based technologies, it is important that molecular studies be used to help recognize

14 Aug 2001 17:41 AR AR135K-19.tex AR135K-19.SGM ARv2(2001/05/10) P1: GSR

512 KOWALCHUK ¥ STEPHEN

AOB of particular ecological relevance to steer future efforts in structural- andfunctional-genomics approaches.

Microarrays (microchips with attached polynucleotide probes) are fast becom-ing the approach of choice for acquiring detailed information about gene expressionin complex systems (e.g., 212). This technology almost certainly offers the great-est potential for the study of microbial communities in situ, and its applicabilityto ammonia-oxidizer ecology has already been demonstrated by Wu et al (216)usingamoAgene fragments from marine ammonia oxidizers as reverse probes.Microarray approaches have the potential to eliminate the need for PCR, thus re-ducing risks of contamination and the biases inherent to PCR. Furthermore, asgenome sequences of additional AOB become available, true functional-genomicsapproaches should become possible, where gene expression within physiologicallydistinct strains is compared under different conditions. Such strategies may pro-vide the clues to niche differentiation among AOB: What defines the physiologicalproperties such as tolerance of acid, saline, or low oxygen conditions?

LIMITATIONS AND NECESSITY TO COMBINE METHODS As with all molecular-basedtechniques, 16S rRNA-based techniques targeting AOB are limited by the speci-ficity and coverage of the primers and probes used, as well as potential biasesand artifacts introduced by the techniques. However, as discussed above, the rapidincrease in available sequence information from culture strains and environmentalclones should aid future primer/probe development.

The pitfalls of PCR-based techniques have been well documented and includebiases in amplification efficiency (16, 180a) and intermolecular interactions duringPCR that may result in the recovery of composite DNA fragments (107, 165, 198).The detection of chimeric sequences is troublesome given the high level of 16SrDNA sequence identity within theβ-subgroup AOB. It is difficult to determineto what extent procedural artifacts may introduce artificial microheterogeneity(165) as typically observed in AOB PCR-directed cloning studies (174). How-ever, comparisons to community-profiling approaches suggest that these effectsmay be considerable. Cloning-based and community-profiling approaches typi-cally detect the same AOB sequence clusters in environmental samples. However,the former strategy rarely recovers a high number of perfectly identical clones,whereas DGGE results suggest the presence of a limited number of discrete ribo-types (105, 166, 173, 174).

That recovered sequences from AOB typically fall into robust groupings thatdiffer in their distribution across environmental gradients strongly suggests thatphysiological differences have evolved between such groupings. However, giventhat highly related bacterial populations can represent differentially selected eco-types (204), the physiological variation between closely related AOB may invali-date some of the generalizations hypothesized at the level of the sequence clustersgiven in Figure 2. Furthermore, no single technique can be expected to provide acomprehensive view of microbial populations in their environment, and a combi-nation of different techniques and molecular markers is always preferred.

14 Aug 2001 17:41 AR AR135K-19.tex AR135K-19.SGM ARv2(2001/05/10) P1: GSR

MOLECULAR ECOLOGY OF AMMONIA OXIDIZERS 513

Techniques for Other Nitrifiers

MOLECULAR CHARACTERIZATION OF γγ -SUBCLASS AMMONIA-OXIDIZING BACTERIA

All primer sets, which are designed to target a specific phylogenetic grouping,must be engineered based upon extant sequence information. Unfortunately, thereare very few available DNA sequences, 16S rDNA or otherwise, fromγ -subclassAOB. Ward et al (202) attempted the specific detection ofγ -subclass AOB in ahypersaline lake but recovered no detectable signal by PCR. However, the primersused were necessarily based upon the only two sequences available, and it is notyet known how well this system detects the full phylogenetic breadth of this clade.

Since the detection ofγ -subclass AOB in marine habitats by immunoflores-cent antibodies (200), little progress has been made in the understanding of thispotentially important group of bacteria, and we still know very little about itscontribution to the AOB populations inhabiting the world’s oceans. Nold et al(127) suggested thatβ-subclass AOB dominate the ammonia-oxidizing commu-nities of a coastal marine sediment; however, detailed knowledge of the globaldistribution ofβ-subclass andγ -subclass AOB is scant. The recovery of aNitroso-coccus oceani-likeγ -proteobacterialamoAsequence from a terrestrial gas-biofilterserves to remind us that theγ -proteobacterial AOBs should never be ignored, evenin nonmarine habitats (155). Critical to the development of molecular detectiontechniques forγ -subclass AOB is the isolation and sequence characterization ofadditional culture strains from this group.

ADVANCES IN THE CHARACTERIZATION OF NITRITE-OXIDIZING BACTERIA Thestudy of NOB has been hampered by many of the same obstacles encounteredwith the study of AOB. In addition to limitations in culture-based methods, devel-opment of molecular methods for NOB detection and characterization is hamperedby the polyphyletic nature of this functional group (181). The use of antibodies tospecifically detect particular NOB populations has furthered this field (200), andthe development of monoclonal antibodies for specific NOB lineages represents asignificant advance in the study of these bacteria (15). Until relatively recently, itwas believed thatNitrobacterwas the dominant, or even the only, nitrite-oxidizinggenus in most nonmarine environments (207), and Degrange & Bardin (50) devel-oped an MPN-PCR method to estimate the size ofNitrobacter-like populations insoils. However, many of the most ecologically important nitrite-oxidizing strainsmay not fall within this genus (85). Thus, the actual distribution of nitrite-oxidizinglineages in the environment remains a topic of future study. Several probes havebeen developed to detect various nitrite oxidizer populations, and they have beenused in the detailed analysis of interactions between AOB and NOB (see above).However, methods for the direct recovery of uncultured nitrite oxidizer nucleicacids from complex environmental samples are still lacking.

HETEROTROPHIC NITRIFIERS The trait of heterotrophic nitrification is distributedover a wide phylogenetic range of bacteria, complicating the development of

14 Aug 2001 17:41 AR AR135K-19.tex AR135K-19.SGM ARv2(2001/05/10) P1: GSR

514 KOWALCHUK ¥ STEPHEN

molecular assays specific for heterotrophic nitrifiers. Although it may be pos-sible to target specific groups of heterotrophic nitrifiers using molecular targets, itis probably impossible to target this broad and variable group of organisms by theuse of a single molecular strategy. The use of molecular techniques, including theuse of polar fatty acid analysis, has been instrumental in advancing the microbialecology of MOB, but such advances are still lacking for other groups capable ofheterotrophic ammonia oxidation. Furthermore, the range of metabolic activitiesfor heterotrophic nitrifiers is much more diverse than that of chemolithotrophicnitrifiers, and the conditions used to demonstrate the potential for heterotrophicnitrification are very different than those encountered in vivo. Thus, the detectionof potential heterotrophic nitrifiers does not ensure that these organisms actuallyplay a role in nitrogen conversions in the environment. It may be possible to targetfunctional genes coding for enzymes involved in heterotrophic nitrification. How-ever, the primary functions of these enzymes may vary widely, making such assaysless than specific for ammonia oxidation. In general, heterotrophic nitrifying bac-teria do not appear to represent a group that is readily accessible by molecularbiological strategies.

ANAMMOX Anammox bacteria are currently impossible to grow in pure cultureand typically exhibit an extremely slow growth rate, with a minimum doubling timeof 9 days (178). The mechanism of this reaction has been elucidated, and recentmolecular data (110, 178) have revealed that the bacteria that appear to be respon-sible for this reaction belong to the Planctomycetes division, a surprising findingas the few cultured members of this division are all organotrophs (Figure 2). Mostinterestingly from the point of view of this review, the Anammox organisms seemto represent a monophyletic lineage with genus level divisions, analogous to theNitrosomonas-Nitrosospirasubdivisions within the aerobicβ-subclass ammoniaoxidizers (156). This knowledge opens the possibility of using similar molecu-lar approaches to study the ecology of these organisms to those described forβ-subclass ammonia oxidizers here, and such efforts have indeed begun (161a).

Modeling and Applications

USE IN BIOTECHNOLOGY Studies attempting to describe the microbial commu-nities active in bioreactors aim primarily to elucidate the population structurescharacteristic of fully functional and failing systems. In this way, it is assumed thatreactor failure can be predicted before the established transforming community isseriously damaged or washed out of the system. Such approaches may target the en-tire bacterial community (9, 52, 126) or key components such as theβ-AOBs. Thelatter approach may have advantages owing to functional redundancy in the widermicrobial community, such that considerable changes in community structure maynot influence functional reactor stability (55, 63). However, modeling such seem-ingly simple environments can still be highly complex (134), and one must recog-nize that certain nitrifying lineages may be particularly well suited for a particular

14 Aug 2001 17:41 AR AR135K-19.tex AR135K-19.SGM ARv2(2001/05/10) P1: GSR

MOLECULAR ECOLOGY OF AMMONIA OXIDIZERS 515

reactor setup and waste input (179). Thus, the maintenance of specific nitrifyingpopulations may be essential to reactor health, and attempts to identify and modelthese communities are of considerable importance (150, 194). Similarly, if reactoractivity is to be enhanced by bioaugmentation with nitrifying bacteria, it followsthat the inoculum must consist of species that are suited to the reactor environment.The same is assumed to be true of soil systems, given that soil characteristics deter-mine the balance ofβ-AOB species types (34, 64, 98, 99, 173, 174). Although theprimary role ofβ-AOBs in biotechnology is currently in nitrogen transforma-tions for wastewater management, including soil-based wastewater treatment (rhi-zoremediation) (5), they also hold considerable potential for the degradation ofnumerous recalcitrant waste chemicals.

REMEDIATION OF RECALCITRANT POLLUTANTS Due to the broad substrate speci-ficity of ammonia monooxygenase, AOB have the potential to act upon a number ofsubstrates, some of which can be highly damaging and recalcitrant in the environ-ment. Anthropogenic pollutants oxidized byβ-AOBs include trichloroethylene,chloroethane, benzene, various halogenated aliphatics, dimethylether, and naturaland synthetic estrogens (11, 70, 76, 86, 143, 144, 184, 187, 217). AOB do not de-rive energy from the oxidation of any of these substrates, which can generally beconsidered as inhibitors of autotrophic ammonia oxidation. Furthermore, the redoxpotential of the environment can be critical to the bioavailability of many pollu-tants, including heavy metals, as well as to the activity of other microorganismsthat play important roles in the remediation of such compounds (36, 80, 172). Thus,the activity of AOB may have both direct and indirect effects on environmentalcleanup mechanisms. Monitoring the health and species composition ofβ-AOBsin polluted environments and bioreactors can be seen to be an important aspect ofmicrobial ecology. However, it should be noted that lab-based studies in this areahave all concentrated onN. europaea, a nitrifier not abundant in many environ-mental settings, and that environmental matrices affect the rate of co-oxidation ofpollutants (70).

Use of Ammonia-Oxidizing Bacteria as Indicator Organisms

It is clear that physiological differences exist among ammonia-oxidizer ecotypes,and certain populations seem to correlate with particular environmental factors.The degree to which ammonia-oxidizer populations specialize in and are indicativeof specific environmental factors is not yet known, but some areas of study offerhope for the use of AOB as indicator organisms.

A number of studies have demonstrated a relationship between the presence ofdistinct ammonia-oxidizer populations and specific environmental conditions. Asindicator organisms, changes in both AOB numbers and species composition havebeen postulated as in situ indicators of the biological impact of pollutants includingfish-farm waste, heavy metals, and hydrocarbons (51, 116, 172). Similarly, AOBcommunity structure may help in defining endpoints to the recovery of land from

14 Aug 2001 17:41 AR AR135K-19.tex AR135K-19.SGM ARv2(2001/05/10) P1: GSR

516 KOWALCHUK ¥ STEPHEN

agricultural perturbation (34, 98, 99) and bioreactor stability (150, 194). Whetherthese observations are robust enough to allow a general coupling of ecologicalparameters withβ-AOB sequence information can be determined only by futureobservations at disparate sites. To be useful as indicators, the methods employedto monitor fluctuations inβ-AOB community size and composition must be suf-ficiently rapid and accurate to predict deterioration in the community in time toallow remediative steps to be taken, such as shutting off reactor inflow. Severalmethods currently exist, including in situ hybridization (85, 122, 194–196), 16SPCR-DGGE (12, 97, 173), and terminal restriction fragment length polymorphismanalysis (T-RFLP) of PCR amplified fragments ofamoA(73).

CONCLUSION

From Black Box to the E. coli of Molecular Microbial Ecology

Theβ-subclass proteobacteria ammonia oxidizers have provided us with an idealgroup of microorganisms to develop molecular ecological methods. Althoughmuch has been learned about the organisms themselves, much has also been learnedabout the advantages, limitations, and pitfalls inherent in these approaches. Ratherthan superceding traditional culture-based analyses, these studies have providednew impetus to the generation of pure cultures. Only pure cultures can unequiv-ocally demonstrate that organisms belonging to, for example, the inferred cluster1 Nitrosospiraand cluster 5Nitrosomonasare not methodological artifacts andindeed have the capacity to grow as chemolithotrophic ammonia oxidizers.β-subclass chemolithotrophic ammonia oxidizers are certainly not the only

organisms of global importance amenable to targeted molecular analysis. Forexample,Achromatium oxaliferumremains uncultured and may be central tosulfur and iron metabolism in freshwater sediments (61); the SAR11 cluster ofα-proteobacteria and Group IArchaea(57, 58a) are abundant in the world’s oceans,yet molecular data is the only knowledge we have of their existence. Clearly, muchof the world’s nutrient cycling is carried out by microorganisms that have not beenisolated, and much of the world’s nutrient sink exists as the biomass of these orga-nisms. Although the molecular methods necessary to identify and model the activ-ities of these organisms are now largely in place, a crucial component is missing.No standard protocols have been laid down for the application of these approaches,most importantly for cell disruption and nucleic acid isolation and amplification.Despite the lack of these standards, the picture of the relationship betweenβ-AOB phylogeny and ecology emerging from disparate laboratories is reassuringlyconsistent. However, methodological differences mean that the results of somerelated studies are difficult to compare even in the field of AOB research. Standardprotocols, as exist for environmental chemistry, have not yet been established inmicrobial ecology (128), but their adoption is a necessary step if the scientific andcommercial potential of this area is to be fully exploited.

14 Aug 2001 17:41 AR AR135K-19.tex AR135K-19.SGM ARv2(2001/05/10) P1: GSR

MOLECULAR ECOLOGY OF AMMONIA OXIDIZERS 517

ACKNOWLEDGMENTS

The authors would like to thank Wietse de Boer (CTO-NIOO) and Arjen Speksni-jder (Univ. Aberdeen) for Figure 3 and useful discussion, Michael J. McGuire(McGuire Environmental Consultants, Inc) for advice on chloramines, JizhongZhou (ORNL) for advice on microarray technology, and Katie Dent and JuliaBruggemann (HRI) for careful reading of the manuscript. The authors’ work onAOBs has been funded by UK NERC grant GR3/8911, U.S. Department of EnergyNABIR grant DE-FCO2-96ER62278 (JRS), the Netherlands Royal Acadamy ofArts and Sciences, and the Netherlands Science Foundation (GAK). This repre-sents NIOO publication number #2717.

Visit the Annual Reviews home page at www.AnnualReviews.org

LITERATURE CITED

1. Aakra A, Hesselsoe M, Bakken LR.2000. Surface attachment of ammonia-oxidizing bacteria in soil.Microbial Ecol.39:222–35

2. Aakra A, Utaker JB, Nes IF. 1999. RFLPof rRNA genes and sequencing of the16S-23S rDNA intergenic spacer regionof ammonia-oxidizing bacteria: a phylo-genetic approach.Int. J. Syst. Bacteriol.49:123–30

3. Aakra A, Utaker JB, Nes IF, BakkenLR. 1999. An evaluated improvement ofthe extinction dilution method for isola-tion of ammonia-oxidizing bacteria.J. Mi-crobiol. Methods39:23–31

4. Abbassi MK, Adams WA. 1998. Loss ofnitrogen in compacted grassland soil bysimultaneous nitrification and denitrifica-tion. Plant Soil200:265–77

5. Abd El Haleem D, Von Wintzingerode F,Moter A, Moawad H, G¨obel UB. 2000.Phylogenetic analysis of rhizosphere-ass-ociatedβ-subclass proteobacterial ammo-nia oxidizers in a municipal wastewatertreatment plant based upon rhizoremedi-ation technology.Lett. Appl. Microbiol.31:34–38

6. Alexander M. 1982. Most probable num-ber method for microbial populations.In Methods of Soil Analysis 2nd ed,ed. AL Page, RH Miller, DR Keeney,

pp. 815–20. Madison, WI: Am. Soc.Agron.

7. Allison SM, Prosser JI. 1993. Ammo-nia oxidation at low pH by attached pop-ulations of nitrifying bacteria.Soil Biol.Biochem.25:935–41

8. Allison SM, Prosser JI. 1991. Ureaseactivity in neutrophilic autotrophic am-monia-oxidizing bacteria isolated fromacid soils.Soil Biol. Biochem.23:45–51

9. Almeida JS, Leung K, MacNaughton SJ,Flemming C, Wimpee MH, et al. 1998.Mapping changes in soil microbial com-munity composition signaling for biore-mediation.Bioremed. J.1(3):255–64

10. Amann RI, Ludwig W, Schleifer KH.1995. Phylogenetic identification andinsitu detection of individual microbialcells without cultivation.Microbiol. Rev.59:143–69

11. Arciero D, Vanelli T, Logan M, HooperAB. 1989. Degradation of trichloroethy-lene by the ammonia-oxidizing bacteriumNitrosomonas europaea. Biochem. Bio-phys. Res. Commun.159:640–43

12. Bano N, Hollibaugh JT. 2000. Diversityand distribution of DNA sequences withaffinity to ammonia-oxidizing bacteria oftheβ-subdivision of the class Proteobac-teria in the Arctic Ocean.Appl. Environ.Microbiol. 66:1960–69

14 Aug 2001 17:41 AR AR135K-19.tex AR135K-19.SGM ARv2(2001/05/10) P1: GSR

518 KOWALCHUK ¥ STEPHEN

13. Baribeau H, Kinner CA, Stephen JR, DeLeon R, Rochelle PA, Clark DI. 2000.Microbial population characterization ofsuspended and fixed biomass in drinkingwater reservoirs.Proc. Am. Water WorksAssoc.–Water Quality Tech. Conf. Biosta-bility in Distribution Networks I, Nov. 5–9 Salt Lake City, UT: Am. Water WorksAssoc.

14. Barraclough D, Puri G. 1995. The useof 15N pool dilution and enrichment toseparate the heterotrophic and autotrophicpathways of nitrification.Soil Biol. Bio-chem.27:17–22

15. Bartosch S, Wolgast I, Spieck E, BockE. 1999. Identification of nitrite-oxidizingbacteria with monoclonal antibodies rec-ognizing the nitrite oxidoreductase.Appl.Environ. Microbiol.65:4126–33

16. Becker S, Boger P, Oehlmann R, ErnstA. 2000. PCR bias in ecological analysis:a case study for quantitativeTaqnucleaseassays in analyses of microbial communi-ties.Appl. Environ. Microbiol.66:4945–53

17. Beiderbeck VO, Campbell CA, Ukrai-netz H, Curtin D, Bouman OT. 1996.Soil microbial and biochemical propertiesafter ten years of fertilization with anhy-drous ammonia.Can. J. Soil Sci.76:7–14

18. Belser LW. 1979. Population ecology ofnitrifying bacteria.Annu. Rev. Microbiol.33:309–33

19. Bloem J, Veninga M, Shephard J. 1995.Fully automatic discrimination of soilbacterium numbers, cell volumes, andfrequencies of dividing cells by confo-cal laser scanning microscopy and im-age analysis.Appl. Environ. Microbiol.6:926–36

20. Blok HJ, Gohlke AM, Akkermans AD.1997. Quantitative analysis of 16S rDNAusing competitive PCR and the QPCR sys-tem 5000.BioTechniques22:700–4

21. Bobbink R, Hornung M, Roelofs JGM.1998. The effects of air-borne nitrogenpollutants on species diversity in natural

and semi-natural European vegetation.J.Ecol.86:717–38

22. Bock E, Sand W. 1993. The microbiol-ogy of masonry biodeterioration.J. Appl.Bacteriol.74:503–14

23. Bock E, Schmidt I, Stuven R, Zart D.1995. Nitrogen loss caused by denitrify-ing Nitrosomonascells using ammoniumor hydrogen as electron donors and ni-trite as electron acceptor.Arch. Microbiol.163:16–20

24. Bodelier PLE, Duyts H, Blom CWPM,Laanbroek HJ. 1998. Interactions be-tween nitrifying and denitrifying bacte-ria in gnotobiotic microcosms plantedwith the emergent macrophyteGlyceriamaxima. FEMS Microbiol. Ecol.25:63–78

25. Bodelier PLE, Frenzel P. 1999. Contribu-tion of methanotrophic and nitrifying bac-teria to CH4 and NH4

+-oxidation in therhizosphere of rice plants as determinedby new methods of discrimination.Appl.Environ. Microbiol.65:1826–33

26. Bodelier PLE, Libochant JA, BlomCWPM, Laanbroek HJ. 1996. Dynam-ics of nitrification and denitrification inroot-oxygenated sediments and adapta-tion of ammonia-oxidizing bacteria tolow-oxygen or anoxic habitats.Appl. En-viron. Microbiol.62:4100–7

27. Bodelier PLE, Roslev P, Henckel T, Fren-zel P. 2000. Stimulation by ammonium-based fertilizers of methane oxidation insoil around rice roots.Nature403:421–24

28. Bohn H. 1992. Consider biofiltration fordecontaminating gases.Chem. Eng. Prog.88:34–40

29. Borneman J, Skroch PW, O’Sullivan KM,Palus JA, Rumjanek NG, et al. 1996.Molecular microbial diversity of an agri-cultural soil in Wisconsin.Appl. Environ.Microbiol. 62:1936–43

30. Bouwman AF. 1990. Exchange of green-house gases between terrestrial ecosys-tems and the atmosphere. InSoils and theGreenhouse Effect, ed. AF Bouwman, pp.61–127. New York: Wiley

14 Aug 2001 17:41 AR AR135K-19.tex AR135K-19.SGM ARv2(2001/05/10) P1: GSR

MOLECULAR ECOLOGY OF AMMONIA OXIDIZERS 519

31. Bremner JM, Blackmer AM. 1978. Ni-trous oxide: emission from soils duringnitrification of fertilizer nitrogen.Science199:295–96

32. Bruggemann J, Stephen JR, Chang Y-J,Macnaughton SJ, Kowalchuk GA, et al.2000. Competitive PCR-DGGE analysisof bacterial mixtures: an internal standardand an appraisal of template enumerationaccuracy.J. Microbiol. Methods40:111–23

33. Bruns MA, Fries MR, Tiedje JM, PaulEA. 1998. Functional gene hybridizationpatterns of terrestrial ammonia-oxidizingbacteria.Microbial Ecol.36:293–302

34. Bruns MA, Stephen JR, Kowalchuk GA,Prosser JI, Paul EA. 1999. Comparativediversity of ammonia oxidizer 16S rRNAgene sequences in native, tilled, and suc-cessional soils.Appl. Environ. Microbiol.65:2994–3000

35. Ceccherini MT, Castialdini M, PiovanelliC, Hastings RC, McCarthy AJ. et al.1998. Effects of swine manure on au-totrophic ammonia-oxidizing bacteria insoil. Appl. Soil Ecol.7:149–57

36. Celebi S, Aytac G. 1998. Toxicity ofheavy metals to selected soil bacteria.Fre-senius Environ. Bull.7:402–7

37. Chandler DP, Schreckhise RW, SmithJL, Bolten H Jr. 1997. Electroelution toremove humic compounds from soil DNAand RNA extracts.J. Microbiol. Methods28:11–19

38. Cho J-C, Kim S-J. 2000. Increase in bac-terial community diversity in subsurfaceaquifers receiving livestock wastewaterinput. Appl. Environ. Microbiol.66:956–65

39. Colliver BB, Stephenson T. 2000. Produc-tion of nitrogen oxide and dinitrogen ox-ide by autotrophic nitrifiers.Biotechnol.Adv.18:219–32

40. De Beer D, Schramm A, SantegoedsCM, Kuhl M. 1997. A nitrite microsen-sor for profiling environmental biofilms.Appl. Environ. Microbiol.63:973–77

41. De Beer D, Van Den Heuvel JC, Otten-

graf SPP. 1993. Microelectrode measure-ments of the activity distribution in nitri-fying bacterial aggregates.Appl. Environ.Microbiol. 59:573–79

42. De Bie MJM, Speksnijder AGCL, Kowa-lchuk GA, Schuurman T, Zwart G, et al.2001. Changes in community structure ofautotrophic ammonia-oxidizingβ-sub-class proteobacteria along the eutrophicSchelde estuary.Aquat. Microb. Ecol.23:225–36

43. De Boer W, Duyts H, Laanbroek HJ.1988. Autotrophic nitrification in a fer-tilized heath soil.Soil Biol. Biochem.20:845–50

44. De Boer W, Klein-Gunnewiek PJA,Laanbroek HJ. 1995. Ammonium-oxi-dation at low pH by a chemolitho-trophic bacterium belonging to the genusNitrosospira. Soil Biol. Biochem.27:127–32

45. De Boer W, Klein-Gunnewiek PJA,Troelstra SR. 1990. Nitrification in Dutchheathland soils II. Characteristics of ni-trate production.Plant Soil 127:193–200

46. De Boer W, Klein-Gunnewiek PJA,Veenhuis M, Bock E, Laanbroek HJ.1991. Nitrification at low pH by aggre-gated autotrophic bacteria.Appl. Environ.Microbiol. 57:3600–4

47. De Boer W, Kowalchuk GA. 2000. Nitri-fication in acid soils: microorganisms andmechanisms.Soil Biol. Biochem.33:853–66

48. De Boer W, Laanbroek HJ. 1989. Ure-olytic nitrification at low pH by Ni-trosospirasp.Arch. Microbiol.152:178–81

49. De Boer W, Tietema A, Klein-Gunne-wiek PJA, Laanbroek HJ. 1992. Thechemolithotrophic ammonium-oxidizingcommunity in a nitrogen-saturated acidforest soil in relation to pH dependentnitrifying activity. Soil Biol. Biochem.24:229–34

50. Degrange V, Bardin R. 1995. Detectionand counting ofNitrobacterpopulations

14 Aug 2001 17:41 AR AR135K-19.tex AR135K-19.SGM ARv2(2001/05/10) P1: GSR

520 KOWALCHUK ¥ STEPHEN

in soil by PCR.Appl. Environ. Microbiol.6:2093–98

51. Deni J, Penninckx MJ. 1999. Nitrifica-tion and autotrophic nitrifying bacteria ina hydrocarbon-polluted soil.Appl. Envi-ron. Microbiol.65:4008–13

52. Eichner CA, Erb RW, Timmis KN, Wag-ner-Dobler I. 1999. Thermal gradient gelelectrophoresis analysis of bioprotectionfrom pollutant shocks in the activatedsludge microbial community.Appl. Envi-ron. Microbiol.65:102–9

53. El-Demerdash ME, Ottow JCG. 1983.Einfluss einer hohen Nitratd¨ungung aufKinetik und Gaszusammensetzung derDenitrifikation in unterscheidlichen B¨o-den. InZeitschrift fur Pflanzenernahrungund Bodenkunde, 146:138–50. Wein-heim, Germany: Verlag Chemie GmbH

54. Felske A, Akkermans AD, De Vos WM.1998. Quantification of 16S rRNAs incomplex bacterial communities by multi-ple competitive reverse transcription-PCRin temperature gradient gel electrophore-sis fingerprints.Appl. Environ. Microbiol.64:4581–87

55. Fernandez AS, Hashsham SA, DollhopfSL, Raskin L, Glagoleva O, et al. 2000.Flexible community structure corre-lates with stable community functionin methanogenic bioreactor communitiesperturbed by glucose.Appl. Environ. Mi-crobiol. 66:4058–67

56. Ford DL, Curchwell RL, Kachtick JW.1980. Comprehensive analysis of nitri-fication of chemical processing waste-waters. J. Water Pollut. Control. Fed.52:2726–45

57. Fuhrman JA, McCallum K, Davis AA.1992. Novel major archaebacterial groupfrom marine plankton.Nature 356:148–49

58. Galloway JN. 1995. Acid deposition: per-spectives in time and space.Water Air SoilPollut. 85:15–24

58a. Gieseke A, Purkhold U, Wagner M,Amann R, Schramm A. 2001. Com-munity structure and activity dynamics

of nitrifying bacteria in a phosphate-removing biofilm.Appl. Environ. Micro-biol. 67:1351–62

59. Giovannoni SJ, Britschgi TB, MoyerCL, Field KG. 1990. Genetic diversityin Sargasso Sea bacterioplankton.Nature345:60–63

60. Goreau TJ, Kaplan WA, Wofsy SC, McE-lroy MB, Valois FW, Watson SW. 1980.Production of NO2

– and N2O by nitrify-ing bacteria at reduced concentrationsof oxygen. Appl. Environ. Microbiol.40:526–32

61. Gray ND, Howarth R, Rowan A, PickupRW, Jones JG, Head IM. 1999. Nat-ural communities ofAchromatium oxal-iferumcomprise genetically, morphologi-cally, and ecologically distinct subpopula-tions.Appl. Environ. Microbiol.65:5089–99

62. Grennfelt P, Hultberg H. 1986. Effects ofnitrogen deposition on the acidification ofterrestrial and aquatic environments.Wa-ter Air Soil Pollut.30:945–63

63. Hashsham SA, Fernandez AS, Doll-hopf SL, Dazzo FB, Hickey RF, et al.2000. Parallel processing of substrate cor-relates with greater functional stabilityin methanogenic bioreactor communitiesperturbed by glucose.Appl. Environ. Mi-crobiol. 66:4050–57

64. Hastings RC, Ceccherini MT, Miclaus N,Saunders JR, Bazzicalupo M, McCar-thy AJ. 1997. Direct molecular biologi-cal analysis of ammonia oxidising bac-teria populations in cultivated soil plotstreated with swine manure.FEMS Micro-biol. Ecol.23:45–54

65. Hastings RC, Saunders JR, Hall GH,Pickup RW, McCarthy AJ. 1998. Appli-cation of molecular biological techniquesto a seasonal study of ammonia oxidationin a eutrophic freshwater lake.Appl. Env-iron. Microbiol. 64:3674–82

66. Head IM, Hiorns WD, Embley TM,McCarthy AJ, Saunders JR. 1993. Thephylogeny of autotrophic ammonium-oxidizing bacteria as determined by

14 Aug 2001 17:41 AR AR135K-19.tex AR135K-19.SGM ARv2(2001/05/10) P1: GSR

MOLECULAR ECOLOGY OF AMMONIA OXIDIZERS 521

analysis of 16S ribosomal RNA gene se-quences.J. Gen. Microbiol.139:1147–53

66a. Hermansson A, Lindgren P-E. 2001.Quantification of ammonia-oxidizingbacteria in arable soil by real-time PCR.Appl. Environ. Microbiol.67:972–76

67. Hiorns WD, Hastings RC, Head IM, Mc-Carthy AJ, Saunders JR, et al. 1995. Am-plification of 16S ribosomal RNA genesof autotrophic ammonia-oxidizing bac-teria demonstrates the ubiquity ofNi-trosospiras in the environment.Microbi-ology141:2793–800

68. Holben WE, Noto K, Sumino T, SuwaY. 1998. Molecular analysis of bacte-rial communities in a three-compartmentgranular activated sludge system indicatescommunity-level control by incompatiblenitrification processes.Appl. Environ. Mi-crobiol. 64:2528–32

69. Holmes AJ, Costello A, Lidstrom ME,Murrell JC. 1995. Evidence that partic-ulate methane monooxygenase and am-monia monooxygenase may be evolu-tionarily related.FEMS Microbiol. Ecol.132:203–8

70. Hommes NG, Russell SA, BottomleyPJ, Arp DJ. 1998. Effects of soil onammonia, ethylene, chloroethane, and1,1,1-trichloroethane oxidation byNitro-somonas europaea. Appl. Environ. Micro-biol. 64:1372–78

71. Honda N, Hirai M, Ano T, Shoda M.1998. Antifungal effect of a heterotrophicnitrifier Alcaligenes faecalis. Biotechnol.Lett.20:703–5

72. Hooper AB, Vannelli T, Bergmann DJ,Arciero D. 1997. Enzymology of the ox-idation of ammonia to nitrite by bacteria.Antonie van Leeuwenhoek J. Gen. Mol.Microbiol. 71:59–67

73. Horz HP, Rotthauwe JH, Lukow T, Lie-sack W. 2000. Identification of major sub-groups of ammonia-oxidizing bacteria inenvironmental samples by T-RFLP anal-ysis ofamoAPCR products.J. Microbiol.Methods39:197–204

74. Hovanec TA, DeLong EF. 1996. Compar-

ative analysis of nitrifying bacteria asso-ciated with freshwater and marine aqua-ria. Appl. Environ. Microbiol.62:2888–96

75. Hovanec TA, Taylor LT, Blakis A, De-Long EF. 1998.Nitrospira-like bacteriaassociated with nitrite oxidation in fresh-water aquaria.Appl. Environ. Microbiol.64:258–64

76. Hyman MR, Russell SA, Ely RL, Willi-amson KJ, Arp DJ. 1995. Cometabolismof trichloroethylene byNitrosomonas eu-ropaea: inhibition, inactivation, and re-covery of ammonia-oxidizing activity.Appl. Environ. Microbiol.61:1480–87

77. Hyman MR, Wood PM. 1985. Suicidalinactivation and labeling of ammoniamonooxygenase by acetylene.Biochem.J. 227:719–25

78. Hynes RK, Knowles R. 1982. Effectof acetylene on autotrophic and het-erotrophic nitrification.Can. J. Microbiol.28:334–40

79. Inbar Y, Hadar Y, Chen Y. 1993. Re-cycling of cattle manure: the compostingprocess and characterization of maturity.J. Environ. Qual.22:857–63

80. Ivanova IA, Stephen JR, Chang Y-J,Bruggemann J, Long PE, et al. 2000.A survey of 16S rRNA andamoAgenesrelated to autotrophic ammonia-oxidisingbacteria of theβ–subdivision of the classProteobacteria in contaminated ground-water.Can J. Microbiol.46:1012–20

80a. Jetten MSM. 2001. New pathways for am-monia conversion in soil and aquatic sys-tems.Plant Soil230:9–19

81. Jetten MSM, Strous M, Van De Pas-Sch-oonen KT, Schalk J, Van Dongen UGJM,et al. 1998. The anaerobic oxidation ofammonium.FEMS Microbiol. Rev.22:421–37

82. Jiang QQ, Bakken LR. 1999. Comparisonof Nitrosospirastrains isolated from ter-restrial environments.FEMS Microbiol.Ecol.30:171–86

83. Jiang QQ, Bakken LR. 2000. Nitrousoxide production and methane oxidation

14 Aug 2001 17:41 AR AR135K-19.tex AR135K-19.SGM ARv2(2001/05/10) P1: GSR

522 KOWALCHUK ¥ STEPHEN

by different ammonia-oxidizing bacteria.Appl. Environ. Microbiol.65:2679–84

84. Joshi JA, Hogan JA, Cowan RM, StromPF, Finstein MS. 2000. Biological re-moval of gaseous ammonia in biofil-ters: space travel and earth-based applica-tions.J. Waste Manage. Assoc.50:1647–54

85. Juretschko S, Timmermann G, SchmidM, Schleifer KH, Pommerening-R¨oserA, et al. 1998. Combined molecular andconventional analyses of nitrifying bac-terium diversity in activated sludge:Ni-trosococcus mobilisand Nitrospira-likebacteria as dominant populations.Appl.Environ. Microbiol.64:3042–51

86. Keener WK, Arp DJ. 1994. Transforma-tions of aromatic compounds byNitro-somonas europaea. Appl. Environ. Micro-biol. 60:1921–41

87. Kester RA, De Boer W, Laanbroek HJ.1996. Short exposure to acetylene to dis-tinguish between nitrifier and denitrifiernitrous oxide production in soil and sed-iment samples.FEMS Microbiol. Ecol.20:111–20

88. Killham K. 1986. Heterotrophic nitrifica-tion. See Ref. 141a, pp. 117–26

89. Killham K. 1990. Nitrification in conif-erous forest soils.Plant Soil 128:31–44

90. Kloos K, Husgen UM, Bothe H. 1998.DNA-probing for genes coding for deni-trification, N2-fixation and nitrification inbacteria isolated from different soils.Z.Naturforsch. Teil C53:69–81

91. Klotz MG, Alzerreca J, Norton JM. 1997.A gene encoding a membrane protein ex-ists upstream of theamoA/amoBgenes inammonia oxidizing bacteria: a third mem-ber of theamooperon?FEMS Microbiol.Lett.150:65–73

92. Klotz MG, Norton JM. 1995. Sequenceof an ammonia monooxygenase subunitA-encoding gene fromNitrosospira sp.NpAV. Gene163:159–60

93. Knowles R. 2000. Factors affecting the lo-calization of transformations of nitrogen

and methane in soils. InMicrobial Biosys-tems: New Frontiers, Proc. 8th Int. Symp.Microbial Ecol., ed. CR Bell, M Brylin-sky, P Johnson-Green, pp. 9–15. Halifax,Can.: Atl. Can. Soc. Microbial Ecol.

94. Koops H-P, M¨oller UC. 1992. The litho-trophic ammonia-oxidizing bacteria. InThe Prokaryotes, ed. A Balows, HGTruper, M Dworkin, W Harder, K-H Schleifer, pp. 2625–37. New York:Springer-Verlag

95. Kowalchuk GA, Bodelier PLE, HeiligGHJ, Stephen JR, Laanbroek HJ. 1998.Community analysis of ammonia-oxidising bacteria, in relation to oxygen-availability in soils and root-oxygenatedsediments, using PCR, DGGE andoligonucleotide probe hybridisation.FEMS Microbiol. Ecol.27:339–50

96. Kowalchuk GA, Naoumenko ZS, DerikxPJL, Felske A, Stephen JR, Arkhip-chenko IA. 1999. Molecular analysis ofammonia-oxidising bacteria of theß-subdivision of the classProteobacteriaincompost and composting materials.Appl.Environ. Microbiol.65:396–405

97. Kowalchuk GA, Stephen JR, De BoerW, Prosser JI, Embley TM, WoldendorpJW. 1997. Analysis ofβ-Proteobacteriaammonia-oxidising bacteria in coastalsand dunes using denaturing gradient gelelectrophoresis and sequencing of PCRamplified 16S rDNA fragments.Appl. En-viron. Microbiol.63:1489–97

98. Kowalchuk GA, Stienstra AW, HeiligGHJ, Stephen JR, Woldendorp JW. 2000.Changes in the community structure ofammonia-oxidising bacteria during sec-ondary succession of calcareous grass-lands.Environ. Microbiol.2:99–110

99. Kowalchuk GA, Stienstra AW, HeiligGHJ, Stephen JR, Woldendorp JW. 2000.Molecular analysis of ammonia-oxidisingbacteria in soil of successional grasslandsof the Drentche A (The Netherlands).FEMS Microbiol. Ecol.31:207–15

100. Kreitinger JP, Klein TM, Novick NJ,Alexander M. 1985. Nitrification and

14 Aug 2001 17:41 AR AR135K-19.tex AR135K-19.SGM ARv2(2001/05/10) P1: GSR

MOLECULAR ECOLOGY OF AMMONIA OXIDIZERS 523

characteristics of nitrifying microorgan-isms in an acid forest soil.Soil Sci. Soc.Am. J.49:1407–10

101. Kronzuker HK, Siddiqi MY, Glass ADM.1997. Conifer root discrimination againstsoil nitrate and the ecology of forest suc-cession.Nature385:59–61

102. Laanbroek HJ, Woldendorp JW. 1995.Activity of chemolithotrophic nitrifyingbacteria under stress in natural soils.Adv.Microb. Ecol.1:275–304

103. Lane DJ. 1991. 16S/23S rRNA sequenc-ing. In Nucleic Acid Techniques in Bac-terial Systematics, ed. E Stackebrandt, MGoodfellow, pp. 115–75. New York: Wi-ley

104. Lang E, Jagnow G. 1986. Fungi of a for-est soil nitrifying at low pH values.FEMSMicrobiol. Ecol.38:257–65

105. Laverman AM, Speksnijder AGCL,Braster M, Kowalchuk GA, Verhoef HA,Van Verseveld HW. 2000. Spatio-tem-poral stability of an ammonia oxidizingcommunity in a nitrogen-saturated forestsoil. Microbial Ecol.41: In press

106. Lettl A. 1985. Heterotrophic nitrifyingbacteria in acid forest soils polluted bySO2. Folia Microbiol. 30:509–16

107. Liesack W, Weyland H, Stakebrandt E.1991. Potential risks of gene amplificationby PCR as determined by 16S rDNA anal-ysis of a mixed culture of strict barophilicbacteria.Microbial Ecol.21:192–98

108. Lieu NI, Wolfe RL, Means EG III. 1993.Optimizing chloramine disinfection forthe control of nitrification.J. Am. WaterWorks Assoc.85:84–90

109. Lipschultz F, Zafiriou OC, Wofsy SC,McElroy MB, Valois FW, Watson SW.1981. Production of NO and N2O by soilnitrifying bacteria.Nature294:641–43

110. Logemann S, Schantl J, Bijvank S, VanLoosdrecht MCM, Kuenen J, JettenMSM. 1998. Molecular microbial diver-sity in a nitrifying reactor system withoutsludge retentionFEMS Microbiol. Ecol.27:239–49

111. MacDonald RM. 1986. Nitrification in

soil: an introductory history. See Ref.141a, pp. 117–26

112. MacDonald RM, Spokes JR. 1980. A se-lective and diagnostic medium for ammo-nia oxidising bacteria.FEMS Microbiol.Lett.8:143–45

113. Maidak BL, Cole JR, Lilburn TG, ParkerCT Jr, Saxman PR, et al. 2000. The RDP(Ribosomal Database Project) continues.Nucleic Acids Res.28:173–74

114. McCaig AE, Embley TM, Prosser JI.1994. Molecular analysis of enrichmentcultures of marine ammonium oxidisers.FEMS Microbiol. Lett.120:363–68

115. McCaig AE, Glover LA, Prosser JI. 1999.Molecular analysis of bacterial commu-nity structure and diversity in unimprovedand improved upland grass pastures.Appl.Environ. Microbiol.65:1721–30

116. McCaig AE, Phillips CJ, Stephen JR,Kowalchuk GA, Harvey M, et al. 1999.Nitrogen cycling and community struc-ture of β-subgroup ammonia oxidisingbacteria within polluted, marine fish-farm sediments.Appl. Environ. Microbiol.65:213–20

117. McGuire MJ, Lieu NI, Pearthree MS.1999. Chlorite ion to control nitrification.J. Am. Water Works Assoc.91:52–61

118. McTavish H, Fuchs JA, Hooper AB.1993. Sequence of the gene coding for am-monia monooxygenase inNitrosomonaseuropaea. J. Bacteriol.175:2436–44

119. Meincke M, Kreig E, Bock E. 1989.Ni-trosovibriospp., the dominant ammonia-oxidizing bacteria in building sto-nes.Appl. Environ. Microbiol.55:2108–10

120. Mendum TA, Sockett RE, Hirsch PR.1999. Use of molecular and isotopictechniques to monitor the response ofautotrophic ammonia-oxidizing popula-tions of theβ-subdivision of the classProteobacteriain arable soils to nitro-gen fertilizer.Appl. Environ. Microbiol.65:4155–62

121. Miller DJ, Nicholas DJD. 1985. Char-acterization of a soluble cytochrome

14 Aug 2001 17:41 AR AR135K-19.tex AR135K-19.SGM ARv2(2001/05/10) P1: GSR

524 KOWALCHUK ¥ STEPHEN

oxidase-nitrite reductase fromNitro-somonas europaea. J. Gen. Microbiol.131:2851–54

122. Mobarry BK, Wagner M, Urbain V, Ritt-mann BE, Stahl DA. 1996. Phylogene-tic probes for analyzing abundance andspatial organization of nitrifying bacte-ria. Appl. Environ. Microbiol.62:2156–62. Erratum. 1997.Appl. Environ. Micro-biol. 63:815

123. Mulder A, van de Graaf AA, RobertsonLA, Kuenen JG. 1995. Anaerobic ammo-nium oxidation discovered in a denitrify-ing fluidized bed reactor.FEMS Micro-biol. Ecol.16:177–84

124. Muyzer G, De Waal EC, Uitterlinden AG.1993. Profiling of complex microbial pop-ulations by denaturing gradient gel elec-trophoresis analysis of polymerase chainreaction-amplified genes coding for 16SrRNA. Appl. Environ. Microbiol.59:695–700

125. Nielsen PH, Andreasen K, Lee N, Wag-ner M. 1999. Use of microautoradiogra-phy and fluorescentin situ hybridizationfor chracterization of microbial activityin activated sludge.Water Sci. Technol.39:1–9

126. Noble PA, Almeida JS, Lovell CR. 2000.Application of neural computing methodsfor interpreting phospholipid fatty acidprofiles of natural microbial communities.Appl. Environ. Microbiol.66:694–99

127. Nold SC, Zhou J, Devol AH, TiedjeJM. 2000. Pacific northwest marine sedi-ments contain ammonia-oxidizing bacte-ria in theβ–subdivision of the proteobac-teria.Appl. Environ. Microbiol.66:4532–35

127a. Norton JM, Low JM, Martin G. 1996.The gene encoding ammonia monooxy-genase subunit A exists in three nearlyidentical copies inNitrosospirasp. NpAV.FEMS Microbiol. Lett.139:181–88

128. Ogram A. 2000. Soil molecular microbialecology at age 20: methodological chal-lenges for the future.Soil Biol. Biochem.32:1499–504

129. Ohashi A, Viray De Silva DG, MobarryB, Manem JA, Stahl DA, Rittmann BE.1995. Influence of substrate C/N ratioon the structure of multi-species biofilmsconsisting of nitrifiers and heterotrophs.Water Sci. Technol.32:75–84

130. Okabe S, Satoh H, Watanabe Y. 1999.In situ analysis of nitrifying biofilms asdetermined byin situ hybridization andthe use of microelectrodes.Appl. Environ.Microbiol. 65:3182–91

131. Olff H, Berendse F, De Visser W.1992. Consequences of changing nutri-ent supply rates for grassland vegetation I.Seasonal dynamics of nitrogen mineral-ization, biomass and tissue nutrient con-centrations. InOn the mechanisms of veg-etation succession, pp. 63–80. PhD thesis.State Univ. Groningen, The Netherlands

132. Osborn AM, Moore ERB, TimmisKN. 2000. An evaluation of terminal-restriction fragment length polymorphismanalysis for the study of microbial com-munity structure and dynamics.Environ.Microbiol. 2:39–50

133. Ouverney CC, Fuhrman JA. 1999. Com-bined microautoradiography-16S rRNAprobe technique for determination of ra-dioisotope uptake by specific microbialcell typesin situ. Appl. Environ. Micro-biol. 65:1746–52

134. Painter HA. 1986. Nitrification in thetreatment of sewage and wastewaters.Spec. Publ. Soc. Gen. Microbiol.20:185–213

135. Pel R, Oldenhuis R, Brand W, VosA, Gottschal JC, Zwart KB. 1997.Stable-isotope analysis of a combinednitrification-denitrification sustained bythermophilic methanotrophs under low-oxygen conditions.Appl. Environ. Micro-biol. 63:474–81

135a. Phillips CJ, Harris D, Dollhopf SL,Gross KL, Prosser JI, Paul EA. 2000.Effects of agronomic treatments on struc-ture and function of ammonia-oxidizingcommunities.Appl. Environ. Microbiol.66:5410–18

14 Aug 2001 17:41 AR AR135K-19.tex AR135K-19.SGM ARv2(2001/05/10) P1: GSR

MOLECULAR ECOLOGY OF AMMONIA OXIDIZERS 525

136. Phillips CJ, Paul EA, Prosser JI. 2000.Quantitative analysis of ammonia oxi-dising bacteria using competitive PCR.FEMS Microbiol. Ecol.32:167–75

137. Phillips CJ, Smith Z, Embley TM,Prosser JI. 1999. Phylogenetic differencesbetween particle-associated and plank-tonic ammonia-oxidizing bacteria of theβ-subdivision of the classProteobacteriain the northwestern Mediterranean Sea.Appl. Environ. Microbiol.65:779–86

138. Pommerening-R¨oser A, Rath G, KoopsHP. 1996. Phylogenetic diversity withinthe genusNitrosomonas. Syst. Appl. Mi-crobiol. 19:344–51

139. Poth M, Focht DD. 1985. N15 Kinetic-analysis of N2O production byNitro-somonas europaea—an examination ofnitrifier denitrificationAppl. Environ. Mi-crobiol. 49:1134–41

140. Princic A, Mahne I, Megusar F, PaulEA, Tiedje JM. 1998. Effects of pH andoxygen and ammonium concentrations onthe community structure of nitrifying bac-teria from wastewater.Appl. Environ. Mi-crobiol. 64:3584–90

141. Prosser JI. 1989. Autotrophic nitrificationin bacteria.Adv. Microb. Physiol.30:125–81

141a. Prosser JI, ed. 1986.Nitrification. Ox-ford, UK: IRL Press

141b. Purkhold U, Pommerening-R¨oser A, Ju-retschko S, Schmid MC, Koops H-P,Wagner M. 2000. Phylogeny of all recog-nized species of ammonia oxidizers basedon comparative 16S rRNA andamoAse-quence analysis: implications for molec-ular diversity surveys.Appl. Environ. Mi-crobiol. 66:5368–82

142. Radajewski S, Ineson P, Parekh NR,Murrell C. 2000. Stable-isotope probingas a tool in microbial ecology.Nature403:646–49

143. Rashe ME, Hicks RE, Hyman MR,Arp DJ. 1990. Oxidation of monohalo-genated ethanes and n-chlorinated alkanesby whole cells ofNitrosomonas europaea.J. Bacteriol.172:5368–73

144. Rashe ME, Hyman MR, Arp DJ. 1991.Factors limiting aliphatic chlorocarbondegradation byNitrosomonas europaea:cometabolic inactivation of ammoniamonooxygenase and substrate specificity.Appl. Environ. Microbiol.57:2986–94

145. Revsbech NP. 1994. Analysis of microbialmats by the use of electrochemical mi-crosensors: recent advances. InMicrobialMats, ed. L Stal, P Caumette, pp. 135–47.Berlin: Springer-Verlag

146. Rice EL, Pancholy SK. 1972. Inhibitionof nitrification by climax ecosystems.Am.J. Bot.59:1033–40

147. Rice EL, Pancholy SK. 1973. Inhibitionof nitrification by climax ecosystems II.Additional evidence and possible role oftannins.Am. J. Bot.60:691–702

148. Rice EL, Pancholy SK. 1973. Inhibitionof nitrification by climax ecosystems. III.Inhibitors other than tannins.Am. J. Bot.61:1095–103

149. Rice EL. 1984.Allelopathy. London:Academic. 2nd ed.

150. Rittmann BE, Laspidou CS, Flax J,Stahl DA, Urbain V, et al. 1999. Molec-ular and modeling analyses of the struc-ture and function of nitrifying activatedsludge.Water Sci. Technol.39:51–59

151. Robertson LA, Kuenen JG. 1990. Com-bined heterotrophic nitrification and aer-obic denitrification inThiosphaera pan-totrophaand other bacteria.Antonie vanLeeuwenhoek J. Gen. Mol. Microbiol.57:139–52

152. Rolleke S, Gurtner C, Drewello U, Drew-ello R, Lubitz W, Weissmann R. 1999.Analysis of bacterial communities on his-torical glass by denaturing gradient gelelectrophoresis of PCR–amplified genefragments coding for 16S rRNA.J. Mi-crobiol. Methods36:107–14

153. Rotthauwe JH, De Boer W, LiesackW. 1995. Comparative analysis of genesequences encoding ammonia monooxy-genase ofNitrosospira sp. AHB1 andNitrosolobus multiformisC-71. FEMSMicrobiol. Lett.133:131–35

14 Aug 2001 17:41 AR AR135K-19.tex AR135K-19.SGM ARv2(2001/05/10) P1: GSR

526 KOWALCHUK ¥ STEPHEN

154. Rotthauwe JH, Witzel KP, Liesack W.1997. The ammonia monooxygenasestructural geneamoA as a functionalmarker: molecular fine-scale analysis ofnatural ammonia-oxidizing populations.Appl. Environ. Microbiol.63:4704–12

155. Sakano Y, Kerkhof L. 1998. Assess-ment of changes in microbial communitystructure during operation of an ammo-nia biofilter with molecular tools.Appl.Environ. Microbiol.64:4877–82

156. Schmid M, Twachtmann U, Klein M,Strous M, Juretschko S, et al. 2000.Molecular evidence for genus level di-versity of bacteria capable of catalyz-ing anaerobic ammonia oxidation.Syst.Appl. Microbiol.23:93–106

157. Schmidt EL, Belser LW. 1982. Nitrify-ing bacteria. See Ref. 6, pp. 1027–42

158. Schmidt I, Bock E. 1997. Anaerobic am-monia oxidation with nitrogen dioxidebyNitrosomonas eutropha. Arch. Micro-biol. 167:106–11

159. Schramm A, De Beer D, Van DenHeuvel JC, Ottengraf S, Amann R.1999. Microscale distribution of popu-lations and activities ofNitrosospiraandNitrospira spp. along a macroscale gra-dient in a nitrifying bioreactor: quantifi-cation by in situ hybridization and theuse of microsensors.Appl. Environ. Mi-crobiol. 65:3690–96

160. Schramm A, De Beer D, Wagner M,Amann R. 1998. Identification and ac-tivities in situ of Nitrosospiraand Ni-trospiraspp. as dominant populations ina nitrifying fluidized bed reactor.Appl.Environ. Microbiol.64:3480–85

161. Schramm A, Larsen LH, Revsbech NP,Ramsing NB, Amann R, Schleifer KH.1996. Structure and function of a ni-trifying biofilm as determined byinsitu hybridization and the use of micro-electrodes.Appl. Environ. Microbiol.62:4641–47

161a. Selenska-Pobell S. 2001. Diversity andactivity of bacteria in uranium wastepiles. In Microbiology and Radioactiv-

ity, ed. M Keith-Roach, F Livens. Ox-ford, UK: Elsevier

162. Sinigalliano CD, Kuhn DN, Jones RD.1995. Amplification of theamoAgenefrom diverse species of ammonium-oxidizing bacteria and from an indige-nous bacterial population from seawater.Appl. Environ. Microbiol. 61:2702–6.Erratum. 1995.Appl. Environ. Micro-biol. 61:4140

163. Smit E, Leefland P, Wernars K. 1997.Detection of shifts in microbial commu-nity structure and diversity in soil causedby copper contamination using ampli-fied ribosomal DNA restriction analysis.FEMS Microbiol. Ecol.24:249–61

163a. Smith Z, McCaig AE, Stephen JR,Embley TM, Prosser JI. 2001. Speciesdiversity of uncultured and cultured pop-ulations of soil and marine ammonia ox-idising bacteria.Microbial Ecol.In press

164. Smorczewski WT, Schmidt EL. 1991.Numbers, activities, and diversity of au-totrophic ammonia-oxidizing bacteria ina freshwater, eutrophic lake sediment.Can. J. Microbiol.37:828–33

165. Speksnijder AGCL, Kowalchuk GA, DeJong S, Kline E, Stephen JR, LaanbroekHJ. 2001. Micro-variation artifacts intro-duced by PCR and cloning of closely re-lated 16S rDNA sequences.Appl. Envi-ron. Microbiol.67:469–72

166. Speksnijder AGCL, Kowalchuk GA,Roest K, Laanbroek HJ. 1998. Recov-ery of aNitrosomonas-like 16S rDNA se-quence group from freshwater habitats.Syst. Appl. Microbiol.21:321–30

167. Spieck E, Miecke M, Bock E. 1992.Taxonomic diversity ofNitrosovibriostr-ains isolated from building sandstones.FEMS Microbiol. Ecol.102:21–26

168. Stams AJM, Flameling EM, Mar-nette ECL. 1990. The importance ofautotrophic versus heterotrophic oxida-tion of atmospheric ammonium in forestecosystems with acid soil.FEMS Micro-biol. Ecol.74:337–44

169. Stark JM, Hart SC. 1997. High rates of

14 Aug 2001 17:41 AR AR135K-19.tex AR135K-19.SGM ARv2(2001/05/10) P1: GSR

MOLECULAR ECOLOGY OF AMMONIA OXIDIZERS 527

nitrification and nitrate turnover inundisturbed coniferous forests.Nature385:61–64

170. Stehr G, B¨ottcher B, Dittberner P, RathG, Koops H-P. 1995. The ammonia oxi-dising nitrifying population of the RiverElbe estuary.FEMS Microbiol. Ecol.17:177–86

171. Stein LY, Arp DJ. 1998. Ammoniumlimitation results in the loss of ammo-nium-oxidising activity inNitrosomonaseuropaea. Appl. Environ. Microbiol.64:1514–21

172. Stephen JR, Chang Y-J, MacnaughtonSJ, Kowalchuk GA, Leung KT, et al.1999. Effect of toxic metals in indige-nous soilβ–subgroup proteobacteriumammonia oxidizer community structureand protection against toxicity by inocu-lated metal-resistant bacteria.Appl. En-viron. Microbiol.65:95–101

173. Stephen JR, Kowalchuk GA, Bruns M-AV, McCaig AE, Phillips CJ, et al.1998. Analysis of β-subgroup pro-teobacterial ammonia oxidizer popula-tions in soil by denaturing gradient gelelectrophoresis analysis and hierarchi-cal phylogenetic probing.Appl. Environ.Microbiol. 64:2958–65

174. Stephen JR, McCaig AE, Smith Z,Prosser JI, Embley TM. 1996. Molec-ular diversity of soil and marine 16SrDNA sequences related toβ-subgroupammonia-oxidizing bacteria.Appl. Env-iron. Microbiol. 62:4147–54

175. Steudler PA, Jones RD, Castro MS,Melillo JM, Lewis D. 1996. Microbialcontrols of methane oxidation in temper-ate forest agricultural soils. InThe Mi-crobiology of Atmospheric Trace Gases:Sources, Sinks and Global Change Pro-cess. NATO ASI Ser. Global Environmen-tal Change, ed. JC Murrell, DP Kelly, pp.69–84. Berlin: Springer-Verlag

176. Stienstra AW. 2000.Nitrification insemi-natural grasslands. PhD thesis.Univ. Gronigen, The Netherlands

177. Stienstra AW, Klein-Gunnewiek P, La-

anbroek HJ. 1994. Repression of nitri-fication in soils under a climax grass-land vegetation.FEMS Microbiol. Ecol.14:45–52

178. Strous M, Fuerst JA, Kramer EHM, Lo-gemann S, Muyzer G, et al. 1999. Miss-ing lithotroph identified as new planto-mycete.Nature400:446–49

179. Suwa Y, Sumino T, Noto K. 1997. Phy-logenetic relationships of activated slu-dge isolates of ammonia oxidizers withdifferent sensitivities to ammonium sul-fate.J. Gen. Appl. Microbiol.43:373–79

180. Suzuki I, Dular U, Kwok SC. 1974. Am-monia and ammonium ion as substratefor oxidation byNitrosomonascells andextracts.J. Bacteriol.176:6623–30

180a. Suzuki MT, Giovannoni SJ. 1996. Biascaused by template annealing in the am-plification of mixtures of 16S rRNAgenes by PCR.Appl. Environ. Microbiol.62:625–30

180b. Suzuki MT, Taylor LT, DeLong EF.2000. Quantitative analysis of small–subunit rRNA genes in mixed microbialpopulations via 5′-nuclease assays.Appl.Environ. Microbiol.66:4605–14

181. Teske A, Elm E, Regan JM, Toze S,Rittmann BE, Stahl DA. 1994. Evolu-tionary relationships among ammonia-and nitrite-oxidizing bacteria.J. Bacte-riol. 176:6623–30

181a. Tietema A, De Boer W, Riemer L, Ver-straten JM. 1992. Nitrate production innitrogen-saturated acid forest soils: ver-tical distribution and characteristics.SoilBiol. Biochem.24:235–240

182. Utaker JB, Bakken L, Jiang QQ,Nes IF. 1995. Phylogenetic analysis ofseven new isolates of the highly relatedammonia-oxidizing bacteria based on16S rRNA gene sequencing.Syst. Appl.Microbiol. 18:549–59

183. Utaker JB, Nes IF. 1998. A qualitativeevaluation of the published oligonucleo-tides specific for the 16S rRNA gene se-quences of the ammonia-oxidizing bac-teria.Syst. Appl. Microbiol.21:72–88

14 Aug 2001 17:41 AR AR135K-19.tex AR135K-19.SGM ARv2(2001/05/10) P1: GSR

528 KOWALCHUK ¥ STEPHEN

184. Vader JS, Van Ginkel CG, SperlingFMGM, De Jong J, De Boer W, et al.2000. Degradation of ethinyl estradiol bynitrifying activated sludge.Chemosphere41:1239–43

185. Vainio EJ, Hantula J. 2000. Direct anal-ysis of wood-inhabiting fungi using de-naturing gradient gel electrophoresis ofamplified ribosomal DNA.Mycol. Res.104:927–36

186. Van Breemen N, Van Dijk HFG. 1988.Ecosystem effects of atmospheric deposi-tion of nitrogen in the Netherlands.Envi-ron. Pollut.54:249–74

187. Vanelli T, Logan M, Arciero DM, Hoo-per AB. 1990. Degradation of halo-genated aliphatic compunds by the am-monia-oxidizing bacteriumNitrosomonaseuropaea. Appl. Environ. Microbiol.56:1169–71

188. Verstraete W. 1975. Heterotrophic nitri-fication in soils and aqueous media—areview. Bull. Acad. Sci. USSR Biol. Ser.4:515–30

189. Von Wintzingerode F, Landt O, EhrichA, Gobel UB. 2000. Peptide nucleic acid-mediated PCR clamping as a useful sup-plement in the determination of micro-bial diversity. Appl. Environ. Microbiol.66:549–57

190. Voordouw G. 1998. Reverse samplegenome probing of microbial communitydynamics.ASM News64:627–33

191. Voytek MA, Ward BB. 1995. Detec-tion of ammonium-oxidizing bacteria oftheβ–subclass of the class Proteobacteriain aquatic samples with the PCR.Appl.Environ. Microbiol. 61:1444–50. Erra-tum. 1995.Appl. Environ. Microbiol. 61:2811

192. Wagner M, Amann R, Lemmer H,Schleifer K-H. 1993. Probing acti-vated sludge withProteobacteria-specificoligonucleotides: inadequacy of culture-dependent methods for describing micro-bial community structure.Appl. Environ.Microbiol. 59:1520–25

193. Wagner M, Erhart R, Manz W, Amann R,

Lemmer H, et al. 1994. Development ofan rRNA-targeted oligonucleotide probespecific for the genusAcinetobacterandfirst application forin situ monitoring inactivated sludge.Appl. Environ. Micro-biol. 60:792–800

194. Wagner M, Noguera DR, JuretschkoS, Rath G, Koops H-P, Schleifer K-H. 1998. Combining fluorescentin situhybridization (FISH) with cultivation andmathematical modeling to study popula-tion structure and function of ammonia-oxidizing bacteria in activated sludge.Wa-ter Sci. Technol.37:441–49

195. Wagner M, Rath G, Amann R, KoopsH-P, Schleifer K-H. 1995.In situ iden-tification of ammonia-oxidizing bacteria.System. Appl. Microbiol.18:251–64

196. Wagner M, Rath G, Koops H-P, Flood J,Amann R. 1996.In situ analysis of nitri-fying bacteria in sewage treatment plants.Water Sci. Technol.34:237–44

197. Wagner R. 1994. The regulation of ribo-somal RNA synthesis and bacterial cellgrowth.Arch. Microbiol.161:100–6

198. Wang GC-Y, Wang Y. 1997. The fre-quency of formation of chimeric mole-cules as a consequence of PCR coampli-fication of 16S rRNA genes from mixedbacterial genomes.Appl. Environ. Micro-biol. 63:4645–50

199. Ward BB. 1986. Nitrification in marineenvironments. See Ref. 141a, pp. 157–84

200. Ward BB, Carlucci AF. 1985. Marineammonia- and nitrite-oxidizing bacteria:serological diversity determined by im-munofluorescence in culture and in theenvironment.Appl. Environ. Microbiol.50:194–201

201. Ward BB, Courtney KJ, Langenheim JH.1997. Inhibition of Nitrosomonas eu-ropaeaby monoterpenes from coastal red-wood (Sequoia sempervirens) in whole-cell studies.J. Chem. Ecol.23:2583–98

202. Ward BB, Martino DP, Diaz MC, JoyeB. 2000. Analysis of ammonia-oxidizingbacteria from hypersaline Mono Lake, on

14 Aug 2001 17:41 AR AR135K-19.tex AR135K-19.SGM ARv2(2001/05/10) P1: GSR

MOLECULAR ECOLOGY OF AMMONIA OXIDIZERS 529

the basis of 16S rRNA sequences.Appl.Environ. Microbiol.66:2873–81

203. Ward BB, Voytek MA, Witzel K-P. 1997.Phylogenetic diversity of natural popula-tions of ammonia oxidizers investigatedby specific PCR amplification.MicrobialEcol.33:87–96

204. Ward DM. 1998. A natural species con-cept for prokaryotes.Curr. Opin. Micro-biol. 1:271–77

205. Warrington R. 1878. On nitrification.J.Chem. Soc.33:44–51

206. Wasserbauer R, Zadak Z, Novotny J.1998. Nitrifying bacteria on the asbestos-cement roofs of stable buildings.Int.Biodeterior.24:153–65

207. Watson SW, Bock E, Harms H, KoopsH-P, Hooper AB. 1989. Nitrifying bacte-ria. InBergey’s Manual of Systematic Mi-crobiology, ed. JT Staley, MP Bryant, NPfennif, JG Holt, 3:1808–34. Baltimore,MD: Williams & Wilkins

208. Wawer C, Muyzer G. 1995. Genetic di-versity of Desulfovibriospp. in environ-mental samples analyzed by denaturinggradient gel electrophoresis of (NiFe) hy-drogenase gene fragments.Appl. Environ.Microbiol. 61:2203–10

209. Wehrfritz JM, Reilly A, Spiro S,Richardson DJ. 1993. Purification of hy-droxylamine oxidase fromThiosphaerapantotropha—identification of electronacceptors that couple heterotrophic nitri-fication to aerobic denitrification.FEBSLett.335:246–50

210. Whitby CB, Saunders JR, Rodriguez J,Pickup RW, McCarthy A. 1999. Phyloge-netic differentiation of two closely related

Nitrosomonasspp. that inhabit differentsediment environments in an oligotrophicfreshwater lake.Appl. Environ. Microbiol.65:4855–62

211. Winogradsky S. 1891. Recherches sur lesorganismes de la nitrification.Ann. Inst.Pasteur5:577–16

212. Wodicka L, Dong H, Mittmann M, Ho M,Lockhart DJ. 1997. Genome-wide expres-sion monitoring inSaccharomyces cere-visiae. Nat. Biotechnol.15:1359–67

213. Woldendorp JW, Laanbroek HJ. 1989.Activity of nitrifiers in relation to nitrogennutrition of plants in natural ecosystems.Plant Soil115:217–28

214. Wolfe RL, Ward RN, Olson BH. 1984.Inorganic chloramines as drinking waterdisinfectants: a review.J. Am. WaterWorks Assoc.76:74–88

215. Wood PM. 1986. Nitrification as a bac-terial energy source. See Ref. 141a, pp.39–62

216. Wu L, Thompson DK, Li G, Hurt RA,Tiedje JM, Zhou J. 2001. Developmentof functional gene arrays for analysis ofmicrobial communities in natural environ-ments.Nat. Biotechnol.In press

217. Yang L, Chang Y-F, Chou M-S.1999. Feasibility of bioremediation oftrichloroethylene contaminated sites bynitrifying bacteria through cometabolismwith ammonia.J. Hazard. Mat.69:111–26

218. Zart D, Schmidt I, Bock E. 2000. Signifi-cance of gaseous NO for ammonia oxida-tion by Nitrosomonas eutropha. Antonievan Leeuwenhoek J. Gen. Mol. Microbiol.77:49–55