4-chlorophenol biodegradation by

Arthrobacter chlorophenolicus A6

Karolina Nordin

Doctoral Thesis

Department of Biochemistry and BiophysicsStockholm University 2004

© Karolina Nordin 2004ISBN: 91-7265-875-4 (p. 1-72)Intellecta DocuSys AB, Stockholm

‘Never lose that screwdriver of yours,’ Mr Litmuswhispered. ‘And never stop questioning or exploring.

But most of all never stop enjoying what you find.’

From “The Adventures of Endill Swift”by Stuart McDonald

5

Doctoral dissertation 2004 Arrhenius Laboratories for Natural SciencesDepartment of Biochemistry and Biophysics Stockholm University

S-106 91 Stockholm, Sweden

Abstract

Microorganisms have the ability to adapt to a variety of environmental conditions. Theirversatility makes them useful for various biotechnological applications. One such applicationis the use of microorganisms for removal of pollutants from the environment, so-calledbioremediation.

The first goal of the present work was to isolate a microorganism which could grow onthe model pollutant 4-chlorophenol. From an enrichment culture, a microorganism wasobtained which could grow on up to 350 µg/ml (2.7 mM) 4-chlorophenol, the highestconcentration reported for bacterial growth in pure culture. In addition to 4-chlorophenol, thestrain could grow on a number of other para-substituted phenols and unsubstituted phenol.Taxonomic studies showed that the microorganism constituted a novel species within thegenus Arthrobacter and it was named Arthrobacter chlorophenolicus A6.

A second objective was to study 4-chlorophenol degradation in soil by A.chlorophenolicus A6. Therefore, the strain was chromosomally tagged with either the gfpgene, encoding the green fluorescent protein (GFP), or the luc gene, encoding fireflyluciferase. When the tagged cells were inoculated into 4-chlorophenol contaminated soil theycould completely remove 175 µg/g 4-chlorophenol within 10 days, whereas no loss of 4-chlorophenol was observed in the uninoculated control microcosms. During theseexperiments the gfp and luc marker genes allowed monitoring of cell number and metabolicstatus.

When A. chlorophenolicus A6 was grown on mixtures of phenolic compounds, the strainexhibited a preference for certain phenols over others. In mixtures, 4-nitrophenol waspreferred over 4-chlorophenol, which in turn was preferred over phenol. Analysis of data oncell growth and degradation of the target compounds indicated that the same enzyme systemwas used for removal of 4-chlorophenol and 4-nitrophenol. Studies with a mutant strain whereone enzyme in the chlorophenol degradation pathway was dysfunctional corroborated this,and indicated that degradation of unsubstituted phenol was mediated by another or anadditional enzyme system. The luc-tagged A. chlorophenolicus A6 gave valuable informationabout growth, substrate depletion and toxicity of the phenolic compounds in substratemixtures.

The 4-chlorophenol degradation pathway in A. chlorophenolicus A6 was elucidated. Themetabolic intermediate subject to ring cleavage was found to be hydroxyquinol, instead of themost commonly found 4-chlorocatechol. Two pathway branches led from 4-chlorophenol tohydroxyquinol; one via hydroquinone and one via 4-chlorocatechol. A gene cluster involvedin 4-chlorophenol degradation was cloned from A. chlorophenolicus A6. The clustercontained two functional hydroxyquinol 1,2-dioxygenase genes and a number of other openreading frames presumed to encode enzymes involved in 4-chlorophenol catabolism. Analysisof the DNA sequence suggested that the gene cluster had partly been assembled by horizontalgene transfer.

In summary, 4-chlorophenol degradation by A. chlorophenolicus A6 was studied from anumber of angles. This organism has several interesting and useful traits such as the ability todegrade high concentrations of 4-chlorophenol and other phenols alone and in mixtures, anunusual and effective 4-chlorophenol degradation pathway and demonstrated ability toremove 4-chlorophenol from contaminated soil.

6

This thesis is based on the following publications:

I. Karolina Westerberg, Annelie M. Elväng, Erko Stackebrandt and Janet K.Jansson. 2000. Biodegradation of high concentrations of 4-chlorophenol:Isolation and characterisation of Arthrobacter chlorophenolicus sp. nov.International Journal of Systematic and Evolutionary Microbiology 50:2083-2092.

II. Annelie M. Elväng, Karolina Westerberg, Cecilia Jernberg and Janet K.Jansson. 2001. Use of green fluorescent protein and luciferase biomarkers tomonitor survival and activity of Arthrobacter chlorophenolicus A6 cells duringdegradation of 4-chlorophenol in soil. Environmental Microbiology 3:32-42.

III. Karolina Nordin, Cecilia Jernberg, Maria Unell, John Stenström and Janet K.Jansson. Degradation of mixtures of phenolic compounds by Arthrobacterchlorophenolicus A6. Manuscript.

IV. Karolina Nordin, Maria Unell and Janet K. Jansson. Molecular and biochemicalcharacterization of 4-chlorophenol biodegradation via a hydroxyquinol pathwayin Arthrobacter chlorophenolicus A6. Submitted.

The publications will be referred to by their Roman numerals in the text.

Additional publications:

Karolina Westerberg, Cecilia Jernberg, Annelie M. Elväng and Janet K. Jansson.Arthrobacter chlorophenolicus A6 – the complete story of a 4-chlorophenol degradingbacterium. Proceedings of the Sixth International In Situ and On-Site BioremediationSymposium. Battelle Press, 2001.

William W. Mohn, Karolina Westerberg, William R. Cullen and Kenneth J. Reimer.1997. Aerobic biodegradation of biphenyl and polychlorinated biphenyls by arctic soilmicroorganisms. Applied and Environmental Microbiology 63:3378-3384.

7

Contents

1 Introduction 9

2 The issue: aromatic pollutants in the environment 102.1 Spread of aromatic pollutants 102.2 Human exposure 112.3 Human health effects 122.4 Restrictions of use 122.5 Focus on phenols 12

3 Bioremediation 143.1 Intrinsic bioremediation 153.2 Biostimulation 15

3.2.1 Oxygenation 153.2.2 Addition of nutrients 163.2.3 pH adjustment 163.2.4 Temperature 163.2.5 Improving bioavailability 16

3.3 Bioaugmentation 173.3.1 Microorganisms for bioaugmentation 173.3.2 Process design 18

3.4 Fungal bioremediation 193.5 Phytoremediation 193.6 Conclusions 19

4 Tools for obtaining microorganisms for bioremediation 204.1 Selective enrichment 204.2 Patchwork assembly of degradation pathways 234.3 Protein engineering 244.4 Comparison of the methods 25

5 Monitoring of bacterial inoculants for bioremediation 265.1 Monitoring using intrinsic markers 26

5.1.1 Approaches to monitoring of intrinsic markers 265.2 Introduced markers 29

5.2.1 Marker gene tagging 295.2.2 Marker genes for monitoring of bioremediation inoculants 30

6 Aerobic degradation of aromatic pollutants by bacteria 356.1 Reactions and enzymes in the metabolism of aromatic compounds 35

6.1.1 Convergent pathways produce a few common ring cleavage substrates 356.1.2 Aromatic ring cleavage 376.1.3 Ring cleavage products are funneled into the citric acid cycle of the cell 39

6.2 Aerobic 4-chlorophenol degradation 40

8

7 Microbial degradation of pollutant mixtures 457.1 Diauxic growth 457.2 Simultaneous degradation of substrates 467.3 Rate deceleration 487.4 Rate acceleration 48

8 Evolution of genes and operons for degradation of pollutants 508.1 Modes of evolution 50

8.1.1 Mutational drift 508.1.2 Genetic rearrangement within a cell 508.1.3 Horizontal gene transfer 51

8.2 Evolution of genes and operons for intradiol ring cleavage 54

9 Conclusions 57

10 Future perspectives 58

11 Acknowledgments 60

12 References 61

9

1 Introduction

“Bacteria represent the great success story of life's pathway. They occupy a widerdomain of environments and span a broader range of biochemistries than any othergroup. They are adaptable, indestructible and astoundingly diverse. We cannot evenimagine how anthropogenic intervention might threaten their extinction, although weworry about our impact on nearly every other form of life…. This is the 'age ofbacteria'---as it was in the beginning, is now and ever shall be.”-Stephen Jay Gould (1994).

Bacteria are found everywhere on the planet, from deserts in Antarctica to deep-seathermal vents, from high up in the atmosphere to several kilometers into the Earth’scrust. Their metabolism is amazingly versatile and they can grow in a wide range ofenvironmental conditions. As humans, we depend on bacteria for our existence; forexample they colonize our skin and digestive tract as part of our immune system,certain bacteria in our gut provide us with vitamin K, and bacteria were initiallyresponsible for the oxygenation of the Earth’s atmosphere.

The versatility of bacteria can be harnessed in a number of biotechnologicalapplications. For example, microorganisms can be used for production of substancessuch as insulin in the pharmaceutical industry, for manufacture of biodegradableplastics and as sources of novel enzymes with activities at temperature extremes.

The nutritional versatility of microorganisms can also be exploited forbiodegradation of environmental pollutants. This process is called bioremediation andis based on the capability of certain microorganisms to metabolize toxic pollutants,obtaining energy and biomass in the process. Ideally, the chemicals are transformedinto harmless compounds such as carbon dioxide and water. Harnessingmicroorganisms to degrade harmful compounds is an attractive option for clean up ofpolluted environments. However, despite the apparent simplicity of microorganisms,the different strategies for dealing with pollutants are as diverse as the organismsthemselves. The process of biodegradation must therefore be investigated on severallevels; biochemical, genetic and physiological.

The focus of this thesis was biodegradation of the target pollutant 4-chlorophenol. A bacterium which could grow on and degrade 4-chlorophenol wasisolated. The isolate was designated strain A6, and 4-chlorophenol degradation by thismicroorganism was studied from a number of angles.

The main questions addressed in this thesis are:o How much 4-chlorophenol can strain A6 tolerate and degrade?o Which other phenols can the microorganism grow on?o Which genus and species does A6 belong to?o Can strain A6 be used to clean up 4-chlorophenol contaminated soil?o Is it possible to monitor cell number and metabolic status of A6 during

remediation of 4-chlorophenol contaminated soil?o Can strain A6 degrade mixtures of different phenols?o How is 4-chlorophenol degraded in this strain?o Which genes encode 4-chlorophenol degradation by strain A6?

10

2 The issue: aromatic pollutants in the environment

Aromatic pollutants pose a serious hazard to human health and the environment. Thearomatic structure itself is not foreign to nature – the benzene ring is the second mostcommon monomer in the environment, prevalent for example as a component of ligninin wood. However, in modern industrial practices, various substitutions may be addedto the aromatic ring, creating compounds which do not naturally occur in theenvironment. Such man-made compounds are termed xenobiotic. Furthermore,aromatic compounds may become pollutants when they are transported to anenvironment where they do not normally occur. Petroleum oil, for example, is anatural product containing aromatic compounds, but it is a pollutant on the Earth’ssurface.

2.1 Spread of aromatic pollutants

Often it is the same property that makes an aromatic compound attractive for use thatcauses the environmental problems. Some compounds, such as polychlorinatedbiphenyls (PCBs) (Fig. 1) and polybrominated diphenyl ethers (PBDEs) (Fig. 1) arechemically inert, and therefore have been used for example as flame retardants(Rahman et al., 2001; Swoboda-Colberg, 1995). Unfortunately, these compounds areinert in the environment as well, and although the production of PCBs was completelyphased out in the 1990s, they can still be found all over the world, even in remoteareas (Breivik et al., 2002; Harner et al., 1998; Swoboda-Colberg, 1995).

Other aromatic compounds are toxic to weeds and pests and have been used asbiocides. However, many such substances are toxic to humans and wildlife as well,and some are also resistant to biological and chemical degradation. Aromatics used asbiocides are for example 1,1,1-Trichloro-2,2-bis-(4'-chlorophenyl)ethane (DDT),chlorophenols and creosote (creosote is a complex mixture of polyaromatichydrocarbons (PAHs)) (Swoboda-Colberg, 1995) (Fig. 1).

Some aromatic compounds in the environment originate from spills or leakage. Forexample, benzene, toluene, ethylbenzene and xylenes (collectively called BTEXcompounds) (Fig. 1) are components in gasoline and used as solvents for example inthe plastics industry (Swoboda-Colberg, 1995; Lynge et al., 1997). Even though only asmall fraction of these compounds escape to the environment, the amount is large inabsolute numbers. PAHs may also enter the environment in this way. For example, at aformer gas work site in Värtan, Stockholm, some 20 PAHs were found in the soil, withlevels of total PAHs ranging between below detection limit up to 5000 mg PAH/kgsoil (Eriksson et al., 2000).

Finally, some aromatic pollutants are formed as by-products of various industrialprocesses. Chlorophenols are for example formed as by-products in paper mills, andPAHs, dioxins (polychlorinated dibenzo-p-dioxins (PCDDs) (Fig. 1) and

11

polychlorinated dibenzofurans (PCDFs)) are formed in combustion processes(Everaert and Bayens, 2002; Pritchard et al., 1987).

Cl

Cl

Cl

CH3 CH3CH3

Cl ClCl

ClCl

DDT

BrBr

Br

Br Br

O

Br Br

Br

BrBr

O

O

Cl

Cl

Cl

Cl

TCDD, a dioxin

OH

Cl

OH

NO2

2,5,4'-Trichlorobiphenyl, a PCB

Phenanthrene, a PAH

Benzene Toluene ortho-Xylene Ethylbenzene

Decabromodiphenyl ether, a PBDE

4-Chlorophenol

4-Nitrophenol

Fig. 1. Some aromatic pollutants mentioned in the text.

2.2 Human exposure

Humans and wildlife are exposed to aromatic pollutants in the environment, as hasbeen shown by the detection of these compounds in samples from many differentorganisms. Furthermore, pollutants may be metabolized in the body, giving rise topotentially more harmful compounds. For example, PCBs and their metabolites havebeen found both in human breast milk and blood plasma (Hovander et al., 2002; Norénand Meironyté, 2000). Also, DDT and its metabolite 1,1-Dichloro-2,2-bis(4'-chlorophenyl)ethylene (DDE) can be detected in breast milk from exposed women, butin areas where the compound has been banned levels have declined strongly in recentyears (Curtis and Lines, 2000; Norén and Meironyté, 2000). Whereas the levels ofmost organochlorine compounds are declining in samples from Swedish individuals,the amount of PBDEs in breast milk is greatly on the increase, and multiplied 60 timesin the time period between 1972 and 1997 (Norén and Meironyté, 2000).

12

2.3 Human health effects

Many aromatic compounds exert their hazardous effect on humans by impacting thehormonal system, which is serious as low amounts can have a profound effect. Forexample, some PCB and DDT congeners can mimic estrogen (Toppari et al., 1996).PBDEs and their metabolites as well as phenols have also been found to have potentialendocrine disrupting properties (Meerts et al., 2001; van den Berg, 1990). Othercompounds are carcinogenic; phenanthrene, benzo(a)pyrene and benzo(a)anthraceneare among the PAHs shown to cause cancer (Samanta et al., 2002; Finlayson-Pitts andPitts, 1997). Benzene is also a well-documented carcinogen, and toluene and xyleneare both implicated in cancer development (Lynge et al., 1997). Some dioxins arehighly toxic and for example 2,3,7,8-tetrachlorodibenzo-p-dioxin (TCDD) is a knownhuman carcinogen (Huwe, 2002; Parzefall, 2002).

2.4 Restrictions of use

For many years, there has been a growing awareness about the adverse effects ofaromatic compounds on both human health and the environment. The StockholmConvention, for example, is a global project to phase out a number of persistentorganic chemicals (www.pops.int). Many of the compounds mentioned above havealready been restricted or banned. The use of DDT was restricted in Sweden in 1970and totally banned in 1975 (Regulation on certain active substances (pesticides),1993), but use continues for malaria control in some African, Asian and SouthAmerican countries (Curtis and Lines, 2000). Improved combustion techniques havealso lead to the reduction of by-products such as dioxins (Huwe, 2002). As mentionedabove, the use of PCBs was completely phased out in the 1990s. Unfortunately,brominated aromatic compounds such as PBDEs have largely taken the place of PCBsin industrial applications and are presently widely used. As a result, theirconcentrations in environmental samples are now frequently found to be higher thanthose of PCBs (Rahman et al., 2001). However, regulations of these compounds arecoming; the European Union has decided to ban the selling and use of any preparationsand objects containing more than 0.1% penta- and octa-BDE starting from August2004 (Directive 2003/11/EG) and investigations on deca-BDE are under way (SwedishNational Chemicals Inspectorate, www.kemi.se).

2.5 Focus on phenols

The focus of this thesis is biodegradation of phenolic compounds, especially 4-chlorophenol. Phenol, methylphenols and chlorophenols in general have been used inthe petrochemical industry (oil/gas industry, refineries and production of basicchemicals). They have also been used in pesticides, in disinfectants, as paintpreservatives and for wood preservation (Swoboda-Colberg, 1995). 4-chlorophenolspecifically is formed from chlorination of waste water, from chlorine bleaching ofpulp and from breakdown of phenoxy herbicides such as 2,4-dichlorophenoxyaceticacid (2,4-D) (Pritchard et al., 1987). 4-chlorophenol is also a product of anaerobic

13

degradation of more highly chlorinated phenols, such as pentachlorophenol (Madsenand Aamand, 1992; Woods et al, 1989), which have been used extensively forpreservation of lumber (Swoboda-Colberg, 1995). 4-nitrophenol was also a targetcompound in the present work, and nitroaromatic compounds in general are used forexample in the explosives industry (Spain, 1995; Swoboda-Colberg, 1995). Moreover,4-nitrophenol may also be formed in soil by hydrolysis of organophospourousinsecticides such as parathion and methylparathion (Munnecke, 1976; Nelson, 1982).

Phenolic compounds, especially halogenated phenolic compounds, have been found tohave possible endocrine-disrupting effects, exerted by interference with the transportof thyroid hormones (van den Berg, 1990). The exposure of humans to phenoliccompounds becomes apparent from the results of a recent study, where some 50brominated and chlorinated phenols were found in plasma from Swedish blood donors(Hovander et al., 2002).

Phenols are also toxic to individual cells, including bacteria, as they uncouple thecell’s respiration (Escher et al., 1996; McLaughlin and Dilger, 1980; Terada, 1990).The chemical basis of this effect is that phenols are fat-soluble and thus may transversethe cell membrane, together with the fact that they are also weak acids, forming aphenolate ion when the hydrogen of the hydroxyl group dissociates from the parentcompound. The type of the substituents on the phenol will affect the dissociationconstant (pKa) of the compound. For example, unsubstituted phenol has a pKa of 10.0,whereas it is 9.4 for 4-chlorophenol and as low as 7.1 for 4-nitrophenol. Thus, aphenolate ion may take up a proton on the outside of the cell membrane, pass throughthe membrane and deposit the proton on the cytosolic side. The phenolate ion maythen diffuse out through the cell membrane, completing the cycle. To furthercomplicate matters, phenols have also been found to form dimers which take part inthis uncoupling cycle (Escher et al., 1996). The effect of the uncoupling activity ofphenols is that these compounds are lethal to microorganisms over a certainconcentration level, and this level varies with both the phenol and microorganism inquestion. In the present work, one of the aims was to obtain a microorganism thatcould tolerate high concentrations of 4-chlorophenol. The goal was achieved as theisolated strain, Arthrobacter chlorophenolicus A6, could use up to 350 µg/ml of 4-chlorophenol for growth (paper I). This is the highest 4-chlorophenol concentrationreported to be used as a growth substrate for a microorganism in pure culture (paper I).

In addition to phenols being toxic when present as single compounds, two differentlysubstituted phenols may have synergistic deleterious effects on uncoupling (Escher etal., 2001). This effect is due to a formation of a dimer between a phenol and aphenolate ion and is most likely to arise when the two phenols have widely differingpKa-values (Escher et al., 2001). Because of this, mixtures of differently substitutedphenols may have a more serious toxic effect than estimated by the effect of eachsingle compound. The issue of mixtures of phenols was addressed in paper III (seealso chapter 7, Microbial degradation of mixed pollutants).

14

3 Bioremediation

The use of biological systems to remove pollutants from the environment is calledbioremediation. The technique may be used instead of, or as a complement to,conventional remediation techniques such as pump-and-treat or excavation andincineration. The advantage of using bioremediation is that it is potentially more cost-effective than traditional cleaning techniques such as waste incineration; possiblesavings have been estimated to 65-85% (Zechendorf, 1999). As any technology,bioremediation has its limitations, in that not all materials can be treated, conditions atthe treatment site may be restrictive or require costly tailor-made applications, and thetime required for treatment may be too long for the method to be useful (Hart, 1996:Zechendorf, 1999).

Bioremediation using bacteria takes advantage of the ability of microorganisms todegrade organic compounds and their ability to evolve degradation pathways forpollutants that are foreign (xenobiotic) in the environment. Some compounds areeasily degraded by microorganisms, whereas others persist in nature – these are calledrecalcitrant. In general, xenobiotics, especially halocarbons, tend to be recalcitrant.Some compounds can only be degraded together with another substrate(cometabolism), such as the solvent trichloroethane and highly chlorinated PCBs.Branched and polyaromatic organic compounds are as a rule more resistant todegradation than straight-chain compounds and monoaromatics. Hydrocarbons andalcohols with low or medium molecular mass are relatively easily biodegraded.

A disadvantage of bioremediation as a soil treatment strategy is the need forcontinuous monitoring of the process to ensure its effectiveness. Both depletion of thetarget pollutant(s) and impact of the treatment on soil biological activities should beexamined. As the environment is often heterogenous, effective monitoring can be verydifficult, and thus reduce the cost-effectiveness of bioremediation. To overcome thislimitation, the use of microbial cells as biosensors for monitoring environmentalconditions is gaining in interest (Jansson, 2003). In this technique, bacteria arenormally engineered with a luminescent or fluorescent reporter gene that is expressedunder specific environmental conditions, such as in the presence of pollutants(Jansson, 2003). For example, Behor et al. (2002) constructed an Escherichia colistrain with luciferase genes fused behind a promoter which was induced in thepresence of toxic compounds. This resulted in light output in response to a variety ofenvironmental pollutants. Microbial biosensors are also useful in that they measure thebioavailability of the pollutant to microbial cells, not just presence of the compound(Sayler and Ripp, 2000; Jansson, 2003). In papers II and III, luminescence output fromluciferase gene (luc)-tagged A. chlorophenolicus A6L gave information about themetabolic state of the cells. In this strain, the luc reporter gene was cloned behind aconstituitive promoter. Therefore the light output reflected the general metabolic statusof the cells, in contrast to biosensors where the reporter gene is fused to induciblepromoters. In paper III, the luciferase activity of A. chlorophenolicus A6L yielded

15

information about the toxicity of the phenols, growth on phenols, and the time point oftheir depletion from the medium.

3.1 Intrinsic bioremediation

Intrinsic bioremediation, also called natural attenuation, relies on the catabolicactivities of the indigenous microflora already present at a contaminated site todegrade the pollutants. The only intervention needed is monitoring of thebiodegradation processes. For example, in a study by Landmeyer et al. (2001), theindigenous microbial community present in a groundwater system was able to degradethe gasoline additive methyl tert-butyl ether, which is a suspected human carcinogen,and no additional treatment was needed as long as oxygen was available. Theadvantage of intrinsic bioremediation is the low cost of the treatment. However, amajor concern with the method is how to obtain quality data to ensure thatbioremediation really is taking place and implementing a long-term monitoringprogram of the process (Atlas and Unterman, 1999; Smets et al., 2002). In addition,the pollutant and its potential metabolites should not pose any danger to plants,animals, humans or natural ecosystems during treatment and spread of the pollutantmust be prevented (Alexander, 1999; Hart, 1996; Smets et al., 2002).

3.2 Biostimulation

Biostimulation is a bioremediation approach that relies on supplementing the naturalbiodegrading microflora with nutrients or other factors that are otherwise limiting.Often a natural degrading microbial community exists at a contaminated site, but oneor several environmental conditions are inadequate. Critical factors may be identifiedby physical and chemical measurements of the site conditions. Some treatments thatcan be employed include the following:

3.2.1 Oxygenation

Oxygen is often a limiting factor in aerobic bioremediation processes (Adriaens andVogel, 1995). Common pollutants such as oil or gasoline, for example, are normallydegraded under aerobic conditions, and therefore oxygenation is an important issue forbiostimulation. Addition of oxygen may be achieved by drainage, tilling, addition ofchemicals that release oxygen (such as hydrogen peroxide) or by injection of air intothe contaminated matter (Adriaens and Vogel, 1995; Atlas and Unterman, 1999).Bioventing is a process that involves injection of air into soil above the water table,and has been extensively used in hydrocarbon remediation (Adriaens and Vogel, 1995;Alexander, 1999). Biosparging involves injection of air below the water table, andhere the purpose is not only to provide O2 as a terminal electron acceptor but also totransfer pollutants into the zone above the water table, which may contain a populationof microorganisms with appropriate degradation capabilities (Alexander, 1999; Hart,1996).

16

3.2.2 Addition of nutrients

If nutrients at the site are lacking, fertilization may enhance biodegradation. Nitrogenand phosphorous fertilization have been found to be beneficial for oil bioremediationand was used for treatment of the Exxon Valdez oil spill on the Alaskan coast (Atlasand Unterman, 1999; Head and Swannell, 1999; Romantschuk et al., 2000).Fertilization is more problematic when the goal is to stimulate subsurface orgroundwater populations, since the nutrients have to reach the microorganisms in anefficient way. For example, electric currents have been used, either directly orfollowing soil fracturing, to create channels in the soil and increase availability ofnutrients (Atlas and Unterman, 1999).

3.2.3 pH adjustment

Most microorganisms are pH-sensitive, and near neutral conditions are typicallyrequired for activity, with variation between 6.5 and 8 or 8.5 usually being acceptable(Bowlen and Kosson, 1995). However, degradation of aromatic compounds has beenreported at pH values as low as 2.0 (Stapleton et al., 1998). If pH is a limiting factorfor bioremediation, it may be adjusted by addition of chemicals; basic soils can betreated with ammonium sulfate or ammonium nitrate, and acidic soils can be treated byliming with CaCO3 or MgCO3 (Bowlen and Kosson, 1995).

3.2.4 Temperature

Low temperature may limit bioremediation by having a negative impact on enzymeactivity and the ability of cells to take up the pollutant (Leung et al., 1997; Nedwell,1999). The best degradation rate for hydrocarbons is generally obtained at around 30°C (Leahy and Colwell 1990; Zhou and Crawford, 1995). Thus, low temperature is apotential problem for bioremediation in many parts of the world. There are twoapproaches to this problem; either heating of the site, for example by pumping warmwater through soil (Ensley and DeFlaun, 1995), or by employing cold-tolerantmicroorganisms. The latter approach has been shown to enable bioremediation of soileven at temperatures prevailing in Arctic climate zones (Mohn and Stewart, 2000;Mohn et al., 1997).

3.2.5 Improving bioavailability

Low biodegradation rates in soil may be due to limited accessibility of the pollutants.A major reason for inaccessibility is slow diffusion in the soil matrix (Harms andBosma, 1997), due to absorption of hydrophobic contaminants to soil particles.Weathering also often makes the pollutant unavailable to the microbial population,presumably due to entrapment of the chemical into soil pores (Atlas and Unterman,1999; Harms and Bosma, 1997). Different treatment strategies may remedy thisproblem; for example biodegradation of PAHs in soil was increased by both heatingand treatment with acetone, presumably due to increased bioavailability (Bonten et al.,1999). Also, the addition of surface-active agents (surfactants) can be used to improvethe bioavailability of oils by increasing the oil/water surface area, or to enhancebioavailability of poorly soluble compounds (Alexander, 1999). However, use ofsurfactants is not unproblematic, as they may inhibit microbial adhesion tohydrophobic pollutants, thus negatively impacting bacterial growth (Stelmack et al.,

17

1999). Some bacteria produce surfactants of their own, called biosurfactants, allowingthem to access hydrophobic compounds as growth substrates, and these bacteria or thesurfactants they produce may be added to soil to improve pollutant bioavailability(Romantschuk et al., 2000; Rosenberg and Ron, 1999).

3.3 Bioaugmentation

When the intrinsic microorganisms at a polluted site are not sufficient to degrade apollutant, inoculation with an exogenous strain or strains, so-called bioaugmentation,may be a feasible approach. The technique may also be used where the intrinsicpopulation is able to degrade the pollutant, but the rate of degradation needs to beincreased. Bioaugmentation may also be combined with the strategies forbiostimulation described above. Bioaugmentation for soil remediation has beenpracticed both in agriculture, forestry and wastewater treatment for a number of years(Vogel, 1996). The method has great potential for remediation but also manyproblems. The success of bioaugmentation is determined by a large number of factors.In addition to the characteristics of the contaminant itself and environmentalconditions at the site, the efficiency of bioaugmentation is also affected by themicrobial ecology of the site (such as grazing by predators or interspeciescompetition), the characteristics of the inoculant (such as stability of degradativecapability and the effect of cosubstrates) and inoculation methodology (van Veen etal., 1997; Vogel, 1996). Soil generally acts as a “buffer” against maintenance ofintroduced microorganisms, but if there is a strong selective advantage for theinoculant, the hostility of soil is more easily overcome, allowing establishment of theintroduced strain (van Veen et al., 1997).

3.3.1 Microorganisms for bioaugmentation

A variety of microorganisms are known which have special properties that make thempromising candidates for bioaugmentation. Such properties include the ability todegrade chemicals that are extremely resistant to biological and chemical degradation,such as chlorinated dibenzofurans and dibenzo-p-dioxins (Habe et al., 2001). Anotherdesirable trait may be tolerance to pollutant concentrations that are inhibitory to othermicroorganisms, such as that displayed by 4-chlorophenol degrading A.chlorophenolicus A6 (paper I). However, catabolic ability alone is not enough toensure efficient bioremediation. An inoculant also needs to be able to survive andperform its task in the contaminated environment, for example soil or ground water.This is usually first evaluated in microcosms set up in the laboratory, designed tomimic environmental conditions, and if these initial experiments are successful,eventually in field trials. Microcosm studies with luc- and gfp-tagged A.chlorophenolicus A6 demonstrated that the microorganism was able to remove 4-chlorophenol from artificially contaminated, non-sterile soil (paper II). All of the 4-chlorophenol (175 µg/g) was removed within 10 days after inoculation whereas therewas negligible removal of the pollutant in uninoculated soil microcosms (paper II). Inaddition, A. chlorophenolicus A6 has the ability to degrade 4-chlorophenol at lowtemperatures. The microorganism can survive and degrade 4-chlorophenol in soil attemperatures as low as 5 °C and also tolerates temperature fluctuations between 5 °C

18

and 28 °C (Backman and Jansson, 2004). This is an important trait for anymicroorganism with potential use for bioremediation in temperate climates. Othermicrocosm studies have shown less promising results; for example Sphingomonas sp.UG30 which can degrade a mixture of pentachlorophenol and 4-nitrophenol in pureculture had only a marginal effect in soil perfusion columns contaminated by the twocompounds (Cassidy et al., 1999). In fact, there are a great number of studies wherebioaugmentation of contaminated soil failed to improve biodegradation (Alexander,1999).

There are several ways in which bioaugmentation may be enhanced by improvingsurvival and activity of the inoculant. For example, encapsulation of the cells intopolymeric material such as alginate or chitosan may improve their survival (Vogel,1996). Encapsulation of a 4-nitrophenol degrading Moraxella sp. was found to protectthe strain from grazing by protozoa (Leung et al., 2000). Another approach is to adjustthe environment to better suit the inoculant (Vogel, 1996). For example, 4-chlorophenol degradation by A. chlorophenolicus A6 was optimized by buffering thesoil pH (paper II). Another strategy for enhancing inoculant survival is to isolate apollutant-degrading microorganism from the polluted site itself. In theory, thismicroorganism will already be adapted to the prevailing environmental conditions andcan be added back to the site in greater numbers. Alternatively, a “geneticbioaugmentation” approach can be adapted, where a (non-degrading) dominantmember of the natural microbial population is isolated from a contaminated site, andgenes encoding the desired catabolic properties are then transferred to the isolate,which is subsequently re-introduced into the contaminated environment (Watanabe etal., 2002). An alternative approach was used by Strong et al. (2000) whooverexpressed an enzyme for atrazine degradation in E. coli, killed the cells and thenapplied them to soil contaminated with atrazine. The method was successful in that upto 77% of the atrazine was removed in the inoculated soil, whereas there was noatrazine removal in non-inoculated soil. Methods involving genetic engineering arefurther discussed in chapter 4, Tools for obtaining microorganisms for bioremediation.Finally, for bioaugmentation applications it is highly advantageous to be able tomonitor the survival and activity of the inoculant, and today a number of techniquesare available for this (see chapter 5, Monitoring of bacterial inoculants forbioremediation).

3.3.2 Process design

Bioaugmentation may be employed both in situ, by adding the cells directly to thecontaminated site, or in bioreactors at or outside the contaminated site, called ex situ.In situ treatment has the advantage that excavation is not needed, but on the other handefficient application of the microorganisms may be difficult, and inherentheterogeneity of the site, especially apparent in soil remediation, may cause problems(Vogel, 1996). In an ex situ approach, on the other hand, mixing and addition ofoxygen and nutrients may be efficiently accomplished for example in soil slurries,compost piles or in biofilm reactors (Adriaens and Vogel, 1995; Vogel, 1996).However, there is always a risk of mobilizing the contaminant, for example byvolatilization, when excavating a site for subsequent treatment, and in this sense, in

19

situ degradation is preferred (Romantschuk et al., 2000). Additional things to considerin setting up a bioaugmentation treatment include determination of the amount ofbacteria that should be added, as well as formulation of the cells for field application(e.g. freeze-drying or suspending in a liquid) (Vogel, 1996).

3.4 Fungal bioremediation

White-rot fungi such as Phanerochaete and related genera have the ability to produceextracellular peroxidase enzymes that act on a variety of organic compounds. Theseenzymes have originally evolved for degradation of lignin and the reactions theyperform are non-specific. Phanerochaete chrysosporium is the most widely studiedbioremediation fungus and has been reported to degrade PAHs, PCBs, DDT, lindanesand PCP, to mention just a few (Alexander, 1999; Cameron et al., 2000; Reddy andGold, 2000). The use of white-rot fungi for bioremediation has several potentialadvantages in that no acclimatization to the substrates is necessary as the system isnon-specific, the method is applicable to insoluble compounds and the non-specificityof the system allows treatment of a mixture of compounds (Cameron et al., 2000). Adisadvantage is that bioremediation using fungi tends to be slow (Alexander, 1999).This problem may be overcome by the addition of large amounts of inoculum, whichcould however be costly (Adriaens and Vogel, 1995; Alexander, 1999).

3.5 Phytoremediation

Phytoremediation is a technique that employs plants to accumulate and eventuallyremove environmental contaminants. The technique is mainly used for soilremediation, but also for sediment (Alexander, 1999). Plants may also be used in“rhizosphere remediation”, where they provide favorable soil characteristics andsuitable environments for biodegrading bacteria that colonize the rhizosphere(Alexander, 1999; Romantschuk et al., 2000).

3.6 Conclusions

Bioremediation is a viable alternative (or supplement) to conventional remediationtechniques, but the success of the methodology depends on many factors, both bioticand abiotic. These factors span over a great range, from physico-chemical properties ofthe contaminant to genetics of the biodegrading microorganisms. Gathering ofknowledge about these factors is important as it will increase the accuracy ofpredicting the outcome of a bioremediation strategy. Two strategies can be discerned,a “top-down” approach used by engineers, mainly relying on stimulating the intrinsicpopulation in order to reduce contaminant levels, and a “bottom-up” approachemployed by many researchers, investigating in detail the properties of a singlebiodegrading microorganism, or a single factor at a polluted site. When these twoapproaches can be fully integrated, the power of bioremediation as a treatment strategywill greatly increase.

20

4 Tools for obtaining microorganisms for bioremediation

Although the microorganisms already present at a polluted site can often adapt todegrade the contaminants if given sufficient time, the indigenous microflora issometimes limited in a way that hampers biodegradation. Such limitations couldinclude a block in the degradation pathway leading to accumulation of intermediates,sensitivity to high concentrations of the pollutant, or incompatibility of differentcomponents in a mixture of pollutants. Often these limitations can be overcome usinga variety of laboratory tools to obtain optimized microbial strains. Such “improved”strains may then be added back to the polluted site.

4.1 Selective enrichment

The classical way to obtain a microorganism that can degrade a specific compound isvia selective enrichment. In this approach, an environmental sample (such as soil) issuspended in medium with the pollutant of interest as the sole carbon source. Thiscreates a selective advantage for those cells that can use the compound for growth(Alexander, 1999). After the start of an enrichment culture, there is typically anacclimation period during which very little degradation occurs. Then after a period ofhours to months, degradation begins (Alexander, 1999; van der Meer et al., 1992).When the enrichment culture is exposed to higher and higher concentrations of thepollutant, there is further selection for microorganisms that are highly tolerant to thecompound. Acquisition of degradative abilities by selective enrichment has been seenin laboratory ecosystems for many organic compounds and heavy metals (Alexander,1999; van der Meer et al., 1992).

Different events may cause an adaptive response to a pollutant in an enrichmentculture. Such events could be induction of enzymes already present in the community,growth of a specific subpopulation able to metabolize the xenobiotic, or selection formutants with altered or novel abilities to metabolize the xenobiotic (Alexander, 1999;van der Meer et al., 1992). The isolation of Burkholderia cepacia AC1100, which isable to grow on the herbicide 2,4,5-trichlorophenoxyacetic acid (2,4,5-T) constitutes atypical example of an enrichment process (Daubaras et al., 1996a). Samples from toxicdump sites were inoculated into a chemostat, and after an incubation period of severalmonths, 2,4,5-T degradation could be detected. Subsequent selection with 2,4,5-Tyielded a pure culture and upon studies of the genetics of the microorganism, it wasfound to contain a unique assortment of 2,4,5-T catabolic genes. Furthermore, atransposable element was implicated in their acquisition.

In paper I the aim was to isolate a microorganism with the ability to tolerate anddegrade high concentrations of the pollutant 4-chlorophenol. An enrichment culturewas inoculated with soil and the 4-chlorophenol concentration was set to 50 ppm, ashigher concentrations of the compound could not be degraded by the microorganismspresent in the soil inoculum. The 4-chlorophenol concentration in the enrichment was

21

monitored over time, and when all of the 4-chlorophenol had been degraded, moresubstrate was added (Fig. 2). The process was repeated, and the 4-chlorophenolconcentration was increased in a step-wise fashion to select for microorganisms thatcould degrade high amounts of the compound. In this way, the bacterial communitywas adapted to be able to remove 350 ppm 4-chlorophenol at the end of the selectiveenrichment. The rate of degradation also increased during the process (Fig. 2), as istypically seen in an enrichment culture (Alexander, 1999). The enrichment strategyproved successful as a microorganism isolated from the community, strain A6, couldgrow on up to 350 µg/ml 4-chlorophenol (paper I). This is the highest 4-chlorophenolconcentration reported to be used for growth by any microorganism growing in pureculture. In addition to 4-chlorophenol, the microorganism can also grow on 4-nitrophenol, 4-bromophenol, 4-fluorophenol and unsubstituted phenol (paper I, paperIII).

Fig. 2. Removal of 4-chlorophenol by a soil community during enrichment withincreasing concentrations of 4-chlorophenol. Arrows show the time points when theculture was spiked with 4-chlorophenol.

A serious concern of the use of enrichment in liquid batch cultures is the inherent biasof the method as it is selective for strains with a high growth rate (Harder andDijkhuizen, 1982). Although a high growth rate may be desirable, the strains thusobtained are not necessarily the ones best suited for bioaugmentation of contaminatedenvironments. Furthermore, some types of degradative genes may be favored overother types in an enrichment culture, as was seen in a comparison of different methodsto obtain 2,4-dichlorophenoxyacetic acid (2,4-D) degrading isolates. Enrichment inbatch cultures yielded almost exclusively strains containing genes highly similar to thewell-known tfd genes, which encode enzymes for 2,4-D catabolism in Ralstonia

22

eutropha JMP134 (pJP4). Direct isolation on plates, on the other hand, yielded threedifferent groups of 2,4-D degradation genes, present in approximately equalabundance (Dunbar et al., 1997).

Instead of applying selective pressure to an entire bacterial community, a desiredcharacteristic of a specific strain may also be enhanced by sequential rounds ofmutation and selection in pure culture. This method is widely used for development ofcommercial microorganisms (Vinci and Byng, 1999). The selection process may beenhanced by increasing the frequency of mutations. For example, Selifonova et al.(2001) introduced so-called mutator plasmids in three different Esherichia strains.This enabled the authors to rapidly produce strains with an increased tolerance toformaldehyde. When the process was completed, the strains were cured of theplasmids.

The 4-chlorophenol degrading isolate described in paper I, strain A6, was studied inmore detail to classify it to genus and species levels. Since bacteria multiply by binaryfission, the species concept is not as straight-forward as for larger animals where twoindividuals are considered to belong to the same species if they have the ability toproduce a fertile offspring. Traditionally, classification of bacteria has instead reliedentirely on phenotypic characteristics, such as cell wall type, morphology, motility,nutritional requirements and fatty acid profile. Although these characteristics certainlygive useful information for bacterial taxonomy, the problem is that individualmicroorganisms may exchange genes with one another via horizontal gene transfer.Therefore, genes for characteristics such as the ability to grow on a certain carbonsource may be transferred between distantly related strains and complicate theclassification.

An alternative approach to bacterial taxonomy is to study the genotype of a strain byanalysis of its nucleic acids. A widely used method is cloning and sequencing of the16S rRNA genes. These genes have become a useful tool for phylogenetic analyses asthey are present in all microorganisms, they are long enough to provide taxonomicinformation but short enough to be easily PCR-amplifed, and they contain conservedregions as targets for PCR primers. The sequences of a large number of 16S rRNAgenes are available in public databases and a novel strain can easily be compared toalready studied microorganisms. Moreover, the 16S rRNA sequence can be used totrace the phylogenetic history (lineage) of an organism. However, analysis of the 16SrRNA gene is in itself not enough to assign a strain as a novel species, more data isnecessary. The “gold standard” of bacterial phylogenetics is instead DNA:DNAhybridization between the total genomic DNA of two related microorganisms. Here,the entire genomes of two strains are compared, and this method is actually used todefine a bacterial species; two strains with 70% or greater DNA similarity asdetermined by DNA:DNA hybridization are considered to belong to the same species(Wayne et al., 1987). However, as cloning and sequencing of 16S rRNA genes istechnically much simpler and cheaper than DNA:DNA hybridization, analysis of 16SrRNA has its place as an initial screening method to determine which the closestrelatives of a novel strain are (Stackebrandt and Goebel, 1994). The level of 16S rRNA

23

similarity is also an indication of whether the strain may constitute a separate speciesand should be investigated further. Comparison of 16S rRNA gene sequences withDNA:DNA hybridization data indicates that when two strains have 16S rRNA geneswhich are identical to less than 97.5%, they are likely to belong to different species(Stackebrandt and Goebel, 1994).

Generally, a polyphasic approach to bacterial systematics is favored, combiningphenotypic and genotypic characteristics. This strategy was used to classify strain A6in paper I. Gram staining showed that the microorganism was Gram-positive, and theobservation that A6 was obligately aerobic, had a rod-coccus life cycle and did notform spores indicated that the strain belonged to the genus Arthrobacter. The molarguanosine plus cytosine content was found to be 65.1 %, which was consistent with anArthrobacter species. That strain A6 was an Arthrobacter was further affirmed byanalysis of the cellular fatty acid profile and peptidoglycan type, although speciesclassification was not possible based on these characteristics alone. Cloning andsequencing of the 16S rRNA gene of strain A6 showed that the strain shared 97.5%sequence identity with its two closest relatives, Arthrobacter oxydans andArthrobacter polychromogenes. The fact that these two species were the closestrelatives to A6 was reinforced by the finding of an unusual type of peptidoglycaninterpeptide bridge in A6 which only was found in A. oxydans and A.polychromogenes within the genus Arthrobacter. As the 16S rRNA gene of strain A6was 97.5% identical to its closest relatives, DNA:DNA hybridization was performed todetermine if the strain constituted a separate species. The result of this analysis wasthat strain A6 was 52% and 41% similar to A. oxydans and A. polychromogenes,respectively, and thus that A6 did indeed constitute a novel species, which was namedArthrobacter chlorophenolicus (paper I).

At this point, one may wonder why bacteria should be classified into genus andspecies at all, when individual cells can exchange genes with one another by horizontalgene transfer, making the classification diffuse at best. Bacterial taxonomy is stilluseful, however, as it can be used to derive a measure of similarity, which allows theresearcher to create some order in the vast diversity of the microbial world. Theconcept of genus and species in bacteria is also useful for identification of isolates, andallows prediction of certain characteristics of a strain, such as medical or ecologicalsignificance. For example, Arthrobacter strains are usually nutritionally versatile, andthis was found to be true also for strain A6 that could grow on a wide variety of carbonsources (paper I).

4.2 Patchwork assembly of degradation pathways

Limitations in the degradation capabilities of naturally occurring microorganisms maybe overcome using genetic engineering techniques, bringing together desirablepathways or enzymes from different organisms. For example, peripheral and centralpathway sequences may be combined in a suitable host to create hybrid degradationpathways (Dua et al., 2002; Reineke, 1998). This may be accomplished by the transferof entire natural plasmids, which is technically simple, or by cloning of appropriate

24

genes, generating novel plasmids (Reineke, 1998; Timmis and Pieper, 1999). Genesmay also be introduced into the chromosome of a cell using transposons as deliveryvectors (Timmis and Pieper, 1999). Erb et al. (1997) used genetic engineering toovercome the problem that few natural bacterial strains are able to metabolize chloro-and methylaromatic compounds simultaneously. A Pseudomonas strain with a novelortho-cleavage pathway, enabling the microorganism to degrade chloro- andmethylphenols simultaneously, could protect the natural microbial community in asewage plant from the deleterious effects of shock loads of chloro- and methylphenols.

Even though strains can be constructed to harbor suitable combinations of genes, thismay not be enough to accomplish growth on a certain compound. Haro and deLorenzo (2001) used a combination of genes from different sources to construct abacterial strain that could degrade 2-chlorotoluene. No natural microorganisms hadpreviously been found which could degrade this compound. However, although thepathway was functional, 2-chlorotoluene could not be used as a sole carbon source bythe constructed strain, indicating that degradation, although enzymatically possible,was physiologically unfavorable. Therefore, the bottleneck limiting successfuldegradation lay elsewhere than enzymatic capability.

Improved biodegradation may also be obtained by other means than transfer of theactual degradative genes. For example, Ohtsubo et al. (2003) were able to enhance thePCB-degrading activity of a Pseudomonas strain by replacing the regulated operonpromotor with a constitutive promotor to increase transcription. The expression of adegradation pathway in another microorganism may also enhance the efficiency ofbiodegradation if the new host is more fit for survival at a particular contaminated site(Chen et al., 1999).

4.3 Protein engineering

The performance of enzymes involved in degradation may be enhanced by a numberof methods. One method for enzyme tailoring is site-directed mutagenesis, wherespecific amino acids in a protein are exchanged. This technique has been used forexample to increase the catalytic activity of haloalkane dehalogenases (Pieper andReineke, 2000). Also, hybrid enzymes may be created by exchanging subunits. Forexample, a and b subunits have been exchanged between biphenyl dioxygenases,altering the substrate range of the enzymes (Furukawa, 2000). Enzymes may also beoptimized by creating variation by mutation and/or recombination, followed byselection for desired traits of the novel enzymes (Harayama, 1998). Mutationaccomplished by error-prone PCR, can be followed by DNA-shuffling (Harayama,1998). Also called “sexual PCR”, DNA-shuffling is a method for rapid evolution ofgenes based on recombination (Stemmer, 1994). Genes are randomly fragmented andthen reassembled by PCR, and recombination occurs as fragments prime each other’selongation based on similarity. Raillard et al. (2001) used DNA-shuffling on twotriazine hydrolases and were able to increase the transformation rate of the enzyme150 times as well as expand the enzyme substrate range. In fact, DNA-shuffling maybe performed without the need to isolate the corresponding organisms or enzymes.

25

Instead, the genes of interest may be PCR-amplified from the environment usingprimers annealing to conserved regions and inserting them into flanking regions of arelated gene (Pieper and Reineke, 2000). Thus, genes from uncultured microorganismsmay be used to create novel versions of enzymes.

4.4 Comparison of the methods

The methods described above all have their advantages and drawbacks. An enrichmentculture is a “black-box approach” in that numerous events are likely to occur duringthe selection process that are difficult to monitor, such as mutation, recombination andhorizontal gene transfer. This is evident from genetic studies of strains obtained byselective enrichment such as Burkholderia cepacia AC1100 (Daubaras et al., 1996a)and A. chlorophenolicus A6 (paper IV). On the other hand, the technique is simple andno advance information about genes and enzymes involved is required.Microorganisms with new degradation abilities harboring unique catabolic genes arecontinually being reported, so natural sources are in no way exhausted, as the resultspresented in this thesis also show. Furthermore, certain problems encountered withmore specifically constructed strains are avoided by the use of enrichment cultures,such as that encountered in the construction of a strain to degrade 2-chlorotoluenedescribed above where degradation was enzymatically but not physiologically feasible(Haro and de Lorenzo, 2001). On the other hand, the introduction of specific genesinto microorganisms to expand or enhance degradation capabilities is an attractivealternative to the enrichment approach as genes from phylogenetically distant strainsmay be combined directly. Also, important knowledge may be gained about thefunction and interaction of different pathways and enzymes. Protein engineering bysite-directed mutagenesis allows investigation of reactions and regulatory mechanisms,but is hampered by the limited sequence space which can be explored in a reasonabletime. In this respect, irrational approaches such as DNA-shuffling are attractive sincethey do not require extensive structural or biochemical information and together with awell-designed screening approach may generate desired enhancements rapidly.

Another angle to consider when comparing these methods is risk assessment. Anystrain that is the product of the more directed methods such as protein engineering orcombination of different genes in the laboratory constitutes a genetically modifiedorganism (GMO). The strains obtained from enrichment cultures, on the other hand,are not regarded as GMOs. There is great concern both among the general public andregulators about the ecological impact of GMOs, and field releases of GMOs arestrictly regulated (Gustafsson, 2000). Therefore, field-scale applications of GMOs arelikely to be much more difficult than using an non-modified strain. However,considering how easily microorganisms can exchange genes with one another byhorizontal gene transfer (discussed in chapter 8, Evolution of genes and operons fordegradation of pollutants), one may wonder why a strain modified by the introductionof a few well-studied genes by a directed method would be more dangerous than amicroorganism which is the product of the uncontrolled exchange of genes occurringin an enrichment culture.

26

5 Monitoring of bacterial inoculants for bioremediation

What happens to a microorganism that is released into the environment to perform acertain task, for example degradation of a pollutant? Will it survive, die, or evenmultiply? Furthermore, will it persist after the pollutant has been removed? As a singlegram of soil may contain 109 - 1010 bacteria (Torsvik et al., 1990), it is no mean feat todistinguish a specific strain amidst this background. However, techniques now existwhich can give information about the survival, dispersal and metabolic activity of aninoculant, as well as the level of transcription of genes encoding catabolic enzymes.Based on such data, the bioaugmentation strategy can be adjusted to optimize removalof pollutants. To assess the fate of a microorganism, markers that naturally occur in thestrain may be used. These are referred to as intrinsic markers. Alternatively, markergenes that facilitate monitoring may be introduced into the cells.

5.1 Monitoring using intrinsic markers

The advantage of using intrinsic markers for tracking of specific microorganisms isthat no previous modification of the cells is needed before they can be applied forbioremediation. This is also a benefit from a regulatory point of view, since in contrastto adding an external marker gene (described below) the use of an intrinsic marker is away to avoid creating a genetically modified organism. However, when using intrinsicmarkers, there is a risk of background, as the same marker may be present in theindigenous microbial population. Generally, the targets for monitoring bacterialinoculants degrading pollutants are the catabolic genes themselves, the 16S rRNAgenes of the cells, intrinsic antibiotic resistance, or combinations of these (Tas andLindström, 2000).

5.1.1 Approaches to monitoring of intrinsic markers

5.1.1.1 Hybridization techniquesThe presence of a gene or specific RNA may be detected by extraction of DNA orRNA, respectively, followed by slot-blot or Southern blot hybridization with specificprobes. Alternatively, culturable cells may be grown on plates and assayed forpresence of the target gene by colony hybridization. Slot-blot hybridization was forexample used to measure the metabolic activity of dehydroabietic-degradingSphingomonas sp. DhA-33 in a mixed sludge community (Muttray et al., 2000). RNAand DNA were extracted and species-specific probes targeted to the 16S rRNA and the16S rRNA genes were used to measure rRNA and DNA encoding rRNA genes,respectively, thus allowing quantification of the RNA:DNA ratio as a measure ofactivity. A drawback of hybridization-based methods is that the sensitivity of detectionmay be low (Jansson and Prosser, 1997).

27

5.1.1.2 PCRThe presence of a specific DNA sequence can be monitored using PCR by designingspecific primers annealing to the target sequence. This method is generally consideredto be more sensitive than hybridization techniques, but the sensitivity is dependent onthe sample quality, DNA extraction approach and the target gene copy number. Thesensitivity of PCR may be further enhanced by the use of nested PCR, whereby asecond round of PCR amplification is performed using the same or internal primers, orby combining PCR with hybridization of specific probes (Tas and Lindström, 2000;Widada et al., 2002). PCR may also be performed in a quantitative manner, providinginformation about the number of target genes present. One quantitative PCR method iscompetitive PCR, where the amount of target DNA amplified is related to the amountof a competitor DNA of known concentration added as an internal standard (Janssonand Leser, 1996). This method was used, for example, to quantify cyanobacteria inBaltic Sea sediment (Möller and Jansson, 1997). PCR may also be renderedquantitative based on the standard most probable number technique (MPN-PCR) (vanElsas et al., 2000). Real-time quantitative PCR has also emerged as a promisingcandidate for monitoring microorganisms in the environment. The method is basedupon detecting amplification products by fluorescence as they accumulate in real-time.This technique was for example used to quantify a methyl tert-butyl ether – degradingmicrobial strain in ground water (Hristova et al., 2001).

RNA may be monitored instead of DNA by the use of reverse-transcription PCR (RT-PCR) (Widada et al., 2002). In this technique, RNA extracted from an environmentalsample is transformed to DNA by reverse transcription, followed by standard PCRamplification. The advantage of monitoring RNA rather than DNA is that the actualtranscription of the gene in question is monitored (or presence of ribosomes, in thecase of rRNA), as opposed to just the presence of the gene when standard PCR isemployed. A major disadvantage of RNA-based methods is that the sensitivity is oftenlow. This is caused by the inherent instability of RNA (especially prokaryotic mRNA)and degradation of RNA by RNases, which are abundant in the environment (vanElsas et al., 2000).

5.1.1.3 Fingerprinting methodsA number of PCR-based methods exist for analysis of microbial communities withoutthe need for isolation of the individual strains. These methods are commonly basedupon amplification of DNA from environmental samples using random primers orprimers targeted to conserved sequences. A pattern of DNA fragments of differentsizes is produced, and changes in the community structure can be assessed over time.Such methods include randomly primed amplified polymorphic DNA (RAPD)analysis, denaturing gradient gel electrophoresis (DGGE) and single strandconformation polymorphism (SSCP). However, these methods are not commonly usedfor monitoring of specific strains introduced into an environment, as the bandingpatterns produced are too complex. In contrast, one fingerprinting technique that isalso applicable to monitoring of specific microorganisms is based on terminalrestriction fragment length polymorphism (T-RFLP). The method consists ofextracting DNA from the environment and PCR-amplifying 16S rRNA genes with one

28

of the PCR primers carrying a fluorescent tag. The PCR products are then digestedwith a restriction enzyme and analyzed on a sequencing gel, where only the terminalfragments, which are carrying the fluorescent tag, are detected (Liu et al., 1997). Thesequencing gel enables very precise size discrimination of the fragments and one base-pair differences in length may be distinguished. Therefore, T-RFLP can be used tomonitor a specific inoculum in soil based on its known terminal restriction fragmentsize. To monitor a specific strain, the restriction enzyme has to be chosen so that noindigenous microorganisms yield the same fragment length as the monitoredmicroorganism. For example, T-RFLP was demonstrated to be useful for monitoringof A. chlorophenolicus A6 cells when the strain was inoculated into 4-chlorophenolcontaminated soil (Jernberg and Jansson, 2002).

5.1.1.4 Fluorescent in situ hybridization (FISH)In this technique, fluorescently labeled oligonucleotide probes are targeted to the 16SrRNA of ribosomes in whole cells. The samples may be analyzed by fluorescencemicroscopy, which also enables study of the positioning of the cells relative tosurrounding matter such as soil particles or crystalline substrate. The probes can beconstructed to be specific for organisms at different phylogenetic levels (Jansson andProsser, 1997). A problem with FISH is that cells from nutrient-limited environmentssuch as soil may contain few ribosomes and thereby also low levels of rRNA, andnutrient activation may be needed prior to applying FISH (Jansson and Prosser, 1997).The background from autofluorescence of soil and organic matter may also be aproblem, but this can be circumvented by the use of confocal scanning lasermicroscopy, a technique which collects detailed images by scanning a laser beamthrough sections of the material (Jansson and Prosser, 1997). The images can becompiled into three-dimensional images using digital imaging software. Tani et al.(2002) used FISH to monitor the fate of Ralstonia eutropha KT1, a microorganismthat can degrade trichloroethylene, after injection of the strain into ground water. Thetechnique was used together with PCR, using primers targeted to the phenol oxidasegene. The PCR was performed on whole, permeabilized cells and afterwards afluorescent probe was hybridized to the PCR product. Thus, both the presence of R.eutropha KT1 16S rRNA and presence of the phenol oxidase gene were monitored.The authors found that after a week fewer cells were detected by FISH than by PCR,which was presumably due to a decline in ribosomal content of the cells.

5.1.1.5 Selective platingAn intrinsic phenotype of an inoculant may be used to specifically identify anorganism of interest. Such a phenotype could be antibiotic resistance, ability to growon certain carbon sources or resistance to heavy metals. In this approach,microorganisms are extracted from an environmental sample and spread on agar platescontaining the selective agent, such as an antibiotic. The main problem in this methodis the specificity, in that there is a risk that other microorganisms in an environmenthave the same phenotype. This risk is especially high if the cells are to be added backto the habitat from which they were originally isolated.

29

5.2 Introduced markers

The idea behind introducing a marker gene into a bacterial strain is to “tag” the cellwith a unique phenotype and/or genotype. This characteristic can then be used tomonitor the strain in the environment. Alternatively, the presence of the marker geneDNA may be monitored by several of the methods described above for monitoring ofintrinsic markers. The ideal marker gene should be specific, sensitive, non-toxic to thecell and the environment, allow quantification of cell number, be suitable for in situdetection and preferably also give information about the metabolic status of the cell.No marker gene fulfills all these criteria, and instead a suitable marker must be chosenon a case-by-case basis. Also, different marker genes may be combined within a singlecell to expand the amount of information gained.

5.2.1 Marker gene tagging

A marker gene may be maintained in the cell on a plasmid, or it may be integrated inthe chromosome. For some bacterial genera, well-studied plasmids exist which can bestably maintained in the host cell and in these cases it is usually easiest to introduce themarker gene on a plasmid, commonly accompanied by an antibiotic resistance gene forthe selection of transformants. An advantage of using plasmids for introduction of amarker gene is that the sensitivity of detection can be improved, if the plasmid ispresent in multiple copies, due to the large amount of protein expressed. A majordraw-back of keeping the marker on a plasmid, however, is that most bacterial strainstend to lose plasmids in the absence of selection and an appropriate selection pressure(e.g. with antibiotics) cannot usually be maintained in a bioaugmentation scenario. Toensure that the marker is stably inherited, it is therefore necessary to introduce themarker gene into the chromosome. Also, for many bacterial species, no well-characterized plasmids are available which can serve as vectors for marker genes.

A marker gene may be introduced into the genome of a microorganism by homologousrecombination or by the use of transposons. For homologous recombination, a deliveryvector is constructed where the marker gene is flanked with sequences from a non-essential part of the genome of the recipient, and usually an antibiotic resistance geneis included to select for recombinants. The delivery vector, which may be a plasmidunable to replicate in the new host, is introduced into the target cell, enabling themarker gene to be integrated into the genome by homologous recombination. Thismethod was used, for example, to chromosomally tag Synechocystis cells with the lucgene encoding firefly luciferase (Möller et al., 1995) (see section 5.2.2.5 for adescription of this marker gene). The marker gene may alternatively be cloned into atransposon, which is then introduced into the target strain and allowed to integrateitself into the chromosome. Again, an antibiotic resistance gene is usually included tofacilitate selection of recombinants. A problem with using transposons for marker genetagging is that the transposon may move again within the genome of the tagged strain,and thus the marker is not completely stably inherited. In response to this issue, so-called mini-transposons have been constructed (de Lorenzo et al., 1990; Herrero et al.,1990), where the transposase is not included in the part of the transposon that is

30

integrated into the genome of the tagged cell. The transposase is instead presentoutside the transferred part, and the entire construct is located on a plasmid that isunable to replicate in the recipient cell. Therefore, when the plasmid is introduced, themarker gene must be transposed into the genome of the recipient, or it is lost. As thetransposase enzyme is not transferred, there is no risk of subsequent transpositionevents after the marker gene has been introduced. Such mini-transposons were usedfor tagging A. chlorophenolicus A6 with the luc or gfp gene (paper II).

5.2.2 Marker genes for monitoring of bioremediation inoculants

5.2.2.1 Antibiotic resistance genesGenes encoding resistance to antibiotics were the first introduced marker genes, andthe phenotype is easy to detect by plating on selective media and counting colonies.The nptII gene, encoding kanamycin or neomycin resistance is commonly used. Otherantibiotic resistance genes used as markers include genes encoding resistance tochloramphenicol, tetracyline, streptomycin, ampicillin and others (Jansson, 2002;Jansson, 2003). However, background from the indigenous microflora could be aserious concern with an antibiotic resistance marker, and therefore antibiotic resistanceis often used together with another marker gene to enable enumeration of colonyforming units (CFUs) on selective plates. This technique was used in paper II, whereA. chlorophenolicus A6 cells extracted from soil were enumerated on selective platesbased on kanamycin resistance conferred by the introduced nptII gene, together withthe luminescent or fluorescent phenotypes conferred by the introduced luc- or gfpgenes (described below). However, there is substantial concern about the spreading ofresistance genes in nature (Kruse and Jansson, 1997). Therefore, genes encodingresistance to antibiotics used in clinical or veterinary practice should preferably beexcised from the tagged strain before a field release. This can be accomplished forexample by site-specific deletion of the resistance gene (Kristensen et al., 1995).

5.2.2.2 Heavy metal resistance genesMarker genes encoding heavy metal resistance are attractive alternatives to antibioticresistance markers, as the potential hazard of releasing heavy metal resistance genesinto the environment is small. Heavy metals used for selection are for examplemercury, zinc, copper and cadmium (Meregay, 1995). However, there is a seriousdrawback with heavy metal resistance genes in that most of the metals used forselection are toxic to humans and require care in medium preparation as well as specialdisposal methods that are both laborious and costly. An option is to use tellurite forselection, a compound that is non-toxic at the concentrations used in growth media.Strains tagged with tellurite resistance genes may be easily visualized on agar platescontaining tellurite, as they turn black from the production of elemental tellurium(Jansson and deBruijn, 1999). Sanchez-Romero et al. (1998) combined genes for thedegradation of toluene from P. putida mt-2 (pWW0) with a tellurite resistance geneand inserted the cassette into another P. putida strain, creating a microorganism thatcould grow on toluene as sole carbon source and was selectable with tellurite.

31

5.2.2.3 Metabolic/chromogenic markersMetabolic or chromogenic markers encode enzymes that can transform a certainsubstrate into a colored product. One such marker is the lacZ gene which encodes b-galactosidase. This gene is often combined with lacY encoding lactose permease(Jansson, 2002; Jansson, 2003). When cells expressing b-galactosidase are grown on amedium containing X-gal (5-bromo-4-chloro-3-indolyl-beta-D-galactopyranoside), thecolonies turn blue. A disadvantage with lacZY as a marker is that the background inthe environment may be high (Jansson, 2002; Jansson, 2003). Another metabolicmarker that has been used for monitoring pollutant-degrading inoculants is the xylEgene which encodes catechol 2,3-dioxygenase. This enzyme cleaves catechol to yieldhydroxymuconic semialdehyde, which is bright yellow. The marker is easy to assay asno special growth medium is needed, instead a catechol-containing solution is sprayeddirectly on the colonies and the yellow color develops within minutes. An advantagewith xylE is that the catechol 2,3-dioxygenase activity is rapidly destroyed outside thecell, therefore only viable cells are detected (Saunders et al., 1995). The xylE gene wasused, for example, to monitor Sphingomonas sp. RW1 while degrading dibenzofuran,dibenzo-p-dioxin and 2-chlorodibenzo-p-dioxin in soil microcosms (Halden et al.,1999).

5.2.2.4 Green fluorescent proteinOriginally isolated and cloned from the jellyfish Aequorea victoria (Chalfie et al.,1994), the green fluorescent protein (GFP) has now become one of the most widelyused biological markers for both prokaryotic and eukaryotic cells. The reason for itspopularity is that the protein is easy to detect as it fluoresces green upon excitationwith UV-light and it does not need any cofactors for fluorescence except oxygen forthe initial formation of the chromophore (Zimmer, 2002). Furthermore, the GFPprotein is extremely stable, with a half-life of approximately 26 h for the wild-typeprotein when measured in mouse cells (Corish and Tyler-Smith, 1999). The GFPprotein, encoded by the gfp gene, has been shown to have a barrel-like structureconsisting of 11 beta-sheets, with a central helix enclosing the chromophore (Ormö etal., 1996; Yang et al., 1996). Today, many mutant GFP variants exist, which areoptimized for use in different organisms (Zimmer, 2002). Furthermore, GFP isavailable in many different colors such as red, cyan, yellow and magenta, allowingtracking of several microbial strains tagged with different GFP variants. However,GFP does have some drawbacks as a marker. The chromophore forms slowly aftertranslation, the protein cannot be used in completely anaerobic environments due tothe oxygen requirement for chromophore formation and it can be difficult todistinguish GFP fluorescence from background fluorescence (Zimmer, 2002).

Because of the stability of GFP once it is formed, and the fact that no co-factors areneeded for expression of the marker phenotype once fluorescence is established, theprotein is ideal for microbial ecology applications as even starved cells may bedetected (Backman et al., 2004; Tombolini and Jansson, 1998; Tombolini et al., 1997,

32

Unge et al., 1999). GFP fluorescence may also be detected in viable but non-culturable(VBNC) cells and in dead cells before membrane integrity is lost (Lowder et al.,2000). Unstable variants have also been created to enable study of transient geneexpression (Andersen et al., 1998).

GFP fluorescence may be detected by a variety of methods. Colonies expressing thephenotype may be distinguished directly with a lamp emitting light in the near-UVrange. Microcolonies of bacteria expressing GFP, for example colonizing a seedsurface, may be studied using a stereomicroscope equipped with a UV-lamp(Tombolini et al., 1999; Unge and Jansson, 2001). Single cells may be detected byepifluorescence microscopy or confocal scanning laser microscopy; the lattertechnique gives very high-resolution images (Tombolini et al., 1999, Unge andJansson, 2001). Confocal microscopy was for example used to visualize GFP-taggedMycobacterium spp. on PAH crystals, making use of the fact that PAHs also fluoresce,but at a different wavelength than GFP (Wattiau et al., 2002). Another technique fordetection of single gfp-tagged cells is by flow cytometry (Backman et al., 2004;Tombolini et al., 1997, Unge et al., 1999). In this approach, the cells are passedbetween a laser beam and a detector, one by one, and a variety of parameters may bemeasured such as fluorescence intensity, cell size and cell shape. Thousands of cellscan be analyzed per second, and thus the method is very efficient.

In paper II, gfp-tagged A. chlorophenolicus A6 was monitored by flow cytometry andselective plating while degrading 4-chlorophenol in soil microcosms. Differentinformation was obtained with the different methods used, highlighting the advantageof combining several monitoring strategies. After an initial increase, approximately 10times more cells were detected by flow cytometry than by selective plating. This couldbe due either to dead cells that retained the GFP protein and thus were counted in theflow cytometer, or to gfp-tagged cells that were indeed alive, but had lost their abilityto grow on rich medium, by entering into a VBNC or dormant state. Recently, this hasbeen investigated in more detail, and the results indicate that A. chlorophenolicus A6can indeed enter into a dormant state in soil (Backman et al., 2004). This wasespecially obvious at an incubation temperature of 5 °C and may be a mechanism forlong-term survival of the microorganism (Backman et al., 2004).

5.2.2.5 Bioluminescence markersBioluminescence markers encode enzymes that produce light at the expense of ATP orreducing power. Luciferase enzymes are produced both by prokaryotes andeukaryotes. Bacterial luciferase is found predominantly in the genera Vibrio,Photobacterium and Xenorhabdus. The reaction catalysed by the enzyme is RCHO +FMNH2 + O2 Æ RCOOH + FMN + H2O + light, where R is a long-chain fattyaldehyde. The luciferase enzyme in bacteria is encoded by the luxAB genes, and thealdehyde by the luxCDE genes. The luxAB genes are often introduced to mark cellswith bacterial luciferase, and the aldehyde then needs to be added externally. For this,n-decanal is often used, as the compound is volatile which makes application simple,and it also diffuses easily through the cell membrane. Alternatively, cells may betagged with the entire lux operon so that the aldehyde substrate is produced by the

33

cells. However, this places an additional burden on the cell and reduced viability hasbeen seen in strains tagged with the entire lux operon (Jansson and deBruijn, 1999).Eukaryotic luciferase, encoded by the luc gene, is produced by fireflies and catalysesthe reaction luciferin + ATP + O2 Æ oxyluciferin + PPi + CO2 + light. Click beetlesalso produce a luciferase which emits light at a different wavelength than the fireflyluciferase (Jansson and deBruijn, 1999). The quantum yield of eukaryotic luciferase isvery high and in Bacillus spp. the light output was found to be 2-3 times higher fromeukaryotic luciferase than from bacterial luciferase (Lampinen et al., 1992).

Luminescence produced by colonies or cultures may be detected on X-ray film, by aCCD camera, or even by eye in a dark room (Möller and Jansson, 1998).Luminescence may also be detected microscopically from single cells if the gene ispresent on a multi-copy plasmid and long camera exposure times are used (Silcock etal., 1992). A very sensitive way to quantitate luminescence is by the use of aluminometer. In such an instrument, photons are collected and amplified and output isgiven as quanta/s or relative light units, RLU. This method was used to quantifyluciferase activity of luc-tagged A. chlorophenolicus A6L, both in pure culture andafter extraction of the cells from soil (papers II and III).

As mentioned above, both bacterial and eukaryotic luciferase enzymes are dependenton the energy reserves of the cell for light output: FMNH2 for bacterial luciferase andATP for eukaryotic luciferase. Luciferase activity therefore reflects the metabolicstatus of the tagged cells and the light output may vary with the growth stage of thecells and presence of growth substrate (Jansson, 2003). Firefly luciferase encoded bythe luc gene was used to tag A. chlorophenolicus A6 in paper II, enabling themetabolic status of the cells to be monitored during remediation of 4-chlorophenolcontaminated soil. In that study, both CFU, as determined by selective plating of thetagged strain, and luciferase activity, were found to initially decrease and then to laterstabilize. In a more recent study, Jernberg and Jansson (2002) showed that theluciferase activity of luc-tagged A. chlorophenolicus was higher in 4-chlorophenolcontaminated soil than in uncontaminated soil, reflecting the increased availability ofenergy reserves in the cells utilizing 4-chorophenol as a substrate.

Luciferase activity also gave valuable information during growth of luc-tagged A.chlorophenolicus A6 on mixtures of phenolic compounds in pure culture (paper III).Light yield and cell density were closely correlated during growth on a mixture of 4-nitrophenol and 4-chlorophenol, but when these compounds were exhausted,luminescence immediately dropped and stabilized at a lower level. This enabled rapiddetermination of the time point when the phenolic compounds were depleted. Such adependence of luciferase activity on metabolic status of populations has also been seenwith other microorganisms constitutively expressing luciferase enzymes (Shaw et al.,1999; Unge et al., 1999). Moreover, the toxicity of phenolic compounds to A.chlorophenolicus A6 was evident in paper III, as luminescence was repressed at aconcentration of phenols over a certain threshold. The repression remained until theconcentration of the phenolic compounds had decreased below this threshold. Thisrepression of luminescence could be caused by the uncoupling effects of the phenols

34

as a disrupted or disturbed membrane potential would affect intracellular ATP levels ina negative way. The direct effect on membrane potential and cellular ATP contentexerted by pentachlorophenol has been studied in Sphingomonas sp. UG25 and UG30.The addition of a non-lethal dose of pentachlorophenol was found to de-energize thecell membrane and cause a drop in ATP levels, both of which were subsequentlyrestored during pentachlorophenol degradation by the strain (Lohmeier-Vogel et al.,2001).

35

6 Aerobic degradation of aromatic pollutants by bacteria

Knowledge about pollutant degradation pathways is important for a number ofreasons. In a bioremediation process, formation of potential toxic metabolites or dead-end products may be predicted, or degradation may be optimized by modifying theenvironment to suit microorganisms having desirable degradation pathways. Also,novel catabolic genes and enzymes may have biotechnological applications, forexample in construction of genetically engineered strains designed to degradeotherwise unreactive compounds. Finally, knowledge of existing degradation pathwayscan aid in designing novel xenobiotic compounds so that they are easilybiodegradable. A database for biodegradation pathways is available on the Internet,where a large number of catabolic reactions may be explored (Ellis and Wackett,1997).

6.1 Reactions and enzymes in the metabolism of aromatic compounds

By combining the findings from elucidation of a number of degradation pathways,some generalizations can be made about the catabolism of aromatic compounds bymicroorganisms, as well as which types of enzymes are involved in the different steps.The degradation process can roughly be divided into three steps; formation of a ringcleavage substrate, cleavage of the aromatic ring, and assimilation into centralmetabolism (Fig. 3).

6.1.1 Convergent pathways produce a few common ring cleavage substrates

The aromatic ring with its shared resonance electrons is a very stable structure andthus resistant to enzymatic attack. Halogen substituents on the aromatic ring stabilizesthe structure further, as the halogens create a steric hindrance to enzymes and alsohave an electron withdrawing effect (Copley, 1997). Before the aromatic ring can becleaved, microorganisms must invest energy in the form of reducing power to modifythe ring. Under aerobic conditions, this usually occurs by the formation of a ringcleavage substrate that contains at least two hydroxyl groups. These are usuallysituated ortho or para to each other (Pieper and Reineke, 2000). The hydroxyl groupsmay be viewed as a sort of “handle” which enables enzymes to cleave the ring. Byintroduction of hydroxyl groups, a variety of different starting compounds aretransformed into a few common intermediates, such as catechol (1,2-dihydroxybenzene) (Fig. 3) (Häggblom, 1992; Harayama and Timmis, 1989).

36

Cl

ClCl

OH

CH3

Cl

Cl

Cl

OHOH

Aliphatic compounds

Citric acid cycle

Fig. 3. Schematic drawing of the general strategies used by microorganisms fordegradation of aromatic pollutants.

6.1.1.1 MonohydroxylationIn general, compounds that already contain a hydroxyl group, such as phenols,salicylate and hydroxybenzoate, are monohydroxylated as the initial transformation(Häggblom, 1992, Mishra et al., 2001). The addition of a single hydroxyl group to anaromatic ring is typically performed by a monooxygenase enzyme, which reduces twoatoms of dioxygen to one hydroxyl group and one H2O molecule by the concomitantoxidation of NAD(P)H (Häggblom 1992; Harayama et al., 1992). Monooxygenases, aswell as dioxygenases (below), require cofactors capable of reacting with dioxygen asreactions directly between dioxygen and carbon in organic compounds are spin-forbidden (Harayama et al., 1992). This cofactor is commonly iron (Harayama et al.,1992). Several types of monooxygenases exist; they may be one-component, two-component, or multi-component (Harayama et al., 1992; Kadiyala and Spain, 1998;Prieto and Garcia, 1994; Takizawa et al., 1995). Open reading frames for both a one-component and a two-component monooxygenase were found in A. chlorophenolicusA6, located in a gene cluster shown to be involved in 4-chlorophenol metabolism(paper IV).

6.1.1.2 DihydroxylationThe simultaneous addition of two hydroxyl groups to an aromatic ring is commonlyseen in the degradation of pollutants such as benzenes, PCBs and PAHs that do notalready contain a hydroxyl group (Gibson and Parales 2000; Häggblom 1992). This

37

dihydroxylation is usually accomplished by a dioxygenase enzyme, which incorporatesboth atoms of dioxygen into the substrate, forming a dihydrodiol, with NAD(P)H as anelectron donor for the reaction (Gibson and Parales 2000, Harayama et al., 1992). In asecond step, the dihydrodiol is dehydrogenated, yielding catechol (Gibson and Parales2000; Harayama et al., 1992). The dioxygenases are multi-component enzymes andmay be divided into different classes based on the number of subunits and cofactorrequirements, but they all contain components that form an electron transport chainfrom NAD(P)H to a terminal oxygenase, and the oxygenase acts on the substrate(Gibson and Parales, 2000; Harayama et al., 1992; van der Meer, 1997).

6.1.1.3 Substituent modificationSubstituents present on the starting aromatic compound may be targeted for chemicalmodification in the formation of ring cleavage intermediates. For example, duringtoluene degradation by Pseudomonas mendocina KR1, the methyl group substituent isoxidized in several steps, eventually forming a carboxyl group (Whited and Gibson,1991). Substituents may also be completely removed. For example, an NO2 group maybe reduced to NH2 that may then be eliminated via the formation of NH3 (Spain, 1995;Takenaka et al., 2003).

For many highly halogenated compounds, it is necessary to remove halogens prior toring cleavage. Under aerobic conditions, this often occurs during mono- ordihydroxylation (above) when a hydroxyl group is introduced in such a way that itreplaces a halogen substituent, resulting in an oxidative dehalogenation of the substrate(Copley, 1997). Furthermore, halogen atoms may be removed through reductivedehalogenation, where a hydrogen atom replaces the substituent, a reaction which is ofspecial importance for degradation of highly chlorinated pollutants, as addition ofmany hydroxyl groups would lead to a product which is very prone to oxidation(Copley, 1997). Aerobic reductive dehalogenation is seen for example duringpentachlorophenol metabolism in Sphingomonas chlorophenolica ATCC 39723 (Xunet al., 1992) and degradation of 2,4,5-trichlorophenoxyacetic acid by Burkholderiacepacia AC1100 (Zaborina et al., 1998). Such a reaction was also inferred in 4-chlorophenol degradation by A. chlorophenolicus A6 (Fig. 5b, paper IV). Finally,dehalogenation may also be hydrolytic, where the chlorine group is replaced by ahydroxyl group derived from water (Copley, 1997), seen for example during 4-chlorobenzoate degradation by Pseudomonas sp. CBS3 (Yang et al., 1994).

6.1.2 Aromatic ring cleavage

The next step in degradation of aromatic compounds is breakage of the aromatic ring,which is performed by ring cleavage dioxygenases. These enzymes are distinct fromring-hydroxylating dioxygenases. All of the ring-cleavage dioxygenases incorporatetwo atoms of dioxygen into the substrate without the need for an external reductantsuch as NAD(P)H. Non-heme iron is required for activity, as for ring-hydroxylatingdioxygenases (Que and Ho, 1996). Two major mechanisms can be distinguished forring cleavage, and they are performed by enzymes that have evolved independently ofone another (Eltis and Bolin, 1996; Que and Ho, 1996) (Fig. 4).

38

OHOH

COOHCOOH

CHO

OH

COOH extradiol cleavage

intradiol cleavage

Fig. 4. Extradiol (meta) and intradiol (ortho) cleavage of catechol.

6.1.2.1 Cleavage proximal to one hydroxyl groupOne of the two major mechanisms for aromatic ring cleavage is called extradiol ormeta cleavage. In this reaction, the ring is opened just outside the hydroxyl groups(Fig. 4). The product of meta cleavage of catechol is hydroxymuconic semialdehyde(Fig. 4). This ring-cleavage mechanism is commonly found in degradation ofmethylated monoaromatics, in degradation of PAHs and for cleavage of the first ringin PCB degradation (Mishra et al., 2001). Extradiol enzymes usually employ ferrousiron (Fe(II)) for their catalytic activity, ligated to the enzyme by one glutamate andtwo histidine residues (Que and Ho, 1996). However, there are also Mn- and Mg-dependent variants (Boldt et al., 1995; Gibello et al., 1994; Hatta et al., 2003; Que etal., 1981). The enzyme reaction mechanism is based upon a nucleophilic attack onbound substrate by O2

- (Que and Ho, 1996).

The catechol 2,3-dioxygenase enzyme encoded by the xylE gene on the TOL plasmidis a well-studied extradiol dioxygenase. The enzyme consists of four identical subunitswith one catalytically essential Fe(II) per subunit (Harayama et al., 1992). The productof catechol cleavage by this enzyme is 2-hydroxymuconic semialdehyde, which has ayellow color, easily visualized by eye (Fig. 4) (Harayama et al., 1992). Because of thisproperty, the xylE gene has been used as a marker or reporter gene for variousapplications in molecular biology and microbial ecology (see section 5.2.1.3, above).

There are enzymes that mechanistically resemble catechol 2,3-dioxygenase, but act oncompounds which have the hydroxyl groups placed in the para position to oneanother. The PcpA enzyme, for example, cleaves dichlorohydroquinone in thepentachlorophenol degradation pathway and although there is some uncertainty as tothe cleavage position (Ohtsubo et al., 1999; Xu et al., 1999; Xun et al., 1999), theenzyme appears to be related to extradiol-cleaving enzymes (Xu et al., 1999). Anotherenzyme that cleaves an aromatic ring with two hydroxyl groups in para position isgentisate 1,2-dioxygenase. This enzyme is involved in degradation of anthralinate, 3-and 4-hydroxybenzoate and salicylate and catalyses the cleavage of the C1-C2-bond of

39

gentisate (2,5-dihydroxybenzoate) (Harayama et al., 1992; Que and Ho, 1996).However, there are indications that gentisate 1,2-dioxygenases should be considered asdistinct from the extradiol dioxygenases (Werwath et al., 1998).

6.1.2.2 Cleavage between two hydroxyl groupsThe second major ring cleavage mechanism is called intradiol or ortho cleavage andtakes place between two hydroxyl groups on an aromatic ring. The product of intradiolcleavage of catechol is cis,cis-muconate (Fig. 4). Ortho cleavage is the dominant ringcleavage mechanism in degradation of chlorinated compounds, as extradiol cleavageof halocatechols may produce dead-end or suicide metabolites (Bartels et al., 1984;Häggblom, 1992). Still, there are examples of productive extradiol cleavage ofhalocatechols. For example, the chlorocatechol 2,3-dioxygenase of Pseudomonasputida GJ31 is resistant to suicide inactivation (Mars et al., 1997).

Catechol 1,2-dioxygenase and protocatechuate 3,4-dioxygenase are well-studiedintradiol cleaving enzymes and their DNA sequences indicate that they are derivedfrom a common ancestor (van der Meer, 1997). The activity of these enzymes isdependent on a ferric iron (Fe(III)) ligated by two tyrosyl and two histidyl residues ofthe enzyme, and the reaction mechanism is an electrophilic attack on bound substrateby O2 (Que and Ho, 1996). Catechol 1,2-dioxygenases of Gram-negative bacteria maybe classified into two types, where type I have little or no activity on chlorocatechols,and type II, also called chlorocatechol 1,2-dioxygenases, have broader specificity andare active on both catechols and chlorocatechols (Häggblom, 1992; Harayama et al.,1992). More recently, hydroxyquinol 1,2-dioxygenases have also been described,which are related to catechol 1,2-dioxygenases and protocatechuate 3,4-dioxygenasesand which use the same reaction mechanism (Daubaras et al., 1995; Travkin et al.,1997). Hydroxyquinol 1,2-dioxygenases have been found to be involved indegradation of phenols and phenoxyacetic acids with two or more chlorine atoms(Daubaras et al., 1995; Häggblom, 1992; Latus et al., 1995; Louie et al., 2002;Zaborina et al., 1995). Furthermore, two functional hydroxyquinol 1,2-dioxygenaseswere found in the cph gene cluster in A. chlorophenolicus A6 and at least one of theseenzymes is vital for 4-chlorophenol degradation in this microorganism (paper IV).

6.1.3 Ring cleavage products are funneled into the citric acid cycle of the cell

After ring cleavage, the products are transformed into compounds that can be used inthe central metabolism of the cell. The hydroxymuconic semialdehyde produced bymeta cleavage (or its derivative, depending on the substitutions on the catechol) istransformed into 2-hydroxypent-2,4-dienoate (or derivative), and finally, depending onthe starting compound, into compounds such as pyruvate, acetaldehyde, oxaloacetate,malate or succinate semialdehyde (Fig. 5a) (Harayama and Timmis, 1989). Orthocleavage of catechol and protocatechuate yields products that are transformed into 3-oxoadipate, eventually yielding acetyl-CoA and succinyl-CoA (Fig 5a) (Harayama andTimmis, 1989). Hydroxyquinol ortho cleavage yields maleylacetate, which in turn isreduced to 3-oxoadipate and thus the pathway converges with the pathway for ortho

40

cleavage of catechol (Armengaud et al., 1999; Daubaras et al., 1996b; Latus et al.,1995; Louie et al., 2002; Murakami et al., 1999; Zaborina et al., 1995; paper IV).

When the ring cleavage substrate is halogenated, these substituents have to beremoved after cleavage. Ortho cleavage of mono- to trichlorocatechol yieldschlorinated cis,cis-muconic acid. One chlorine is removed concurrently with alactonization of the compound by chloromuconate cycloisomerase (Fig. 5a). Thelactone is then acted upon by dienelactone hydrolase, yielding maleylacetate, which, ifnon-chlorinated, is reduced to 3-oxoadipate by maleylacetate reductase (Fetzner andLingens, 1994). If chlorine atoms still remain on the maleylacetate, they may beremoved by reductive dechlorination by maleylacetate reductase, yieldingmaleylacetate (Fetzner and Lingens, 1994).

For complex hydrocarbons such as PCBs and PAHs, several rounds of ring cleavageare needed in order to obtain mineralization. For example, PCB degradation yieldschlorobenzoates, which may themselves pose a pollution problem.

6.2 Aerobic 4-chlorophenol degradation

4-chlorophenol biodegradation, the subject of this thesis, has mainly been studied inGram-negative microorganisms (Bae et al., 1996b; Chih-Jen et al., 1996; Fava et al.,1995; Hollender et al., 1994; Häggblom, 1992). Degradation of this compound isusually accomplished by oxidation of the substrate to 4-chlorocatechol, followed byortho cleavage of the aromatic ring (Fig. 5a) (Chih-Jen et al., 1996; Farrell and Quilty,2002; Fava et al., 1995; Häggblom, 1992; Park and Kim, 2003). Some microorganismsprocess 4-chlorocatechol by meta cleavage, although it is less common (Fig. 5a) (Baeet al., 1996b; Farrell and Quilty, 1999; Hollender et al., 1994). After ring cleavage, thechlorine atom is removed and the carbon skeleton is transformed into products that areassimilated into the central metabolism of the cell (Fig. 5a).

41

OH

Cl

OH

Cl

OH

OH

OHOH

Cl

OHOH

Cl

OHOH

ClOH

OH

COO-

Cl

CHO

OHOH

OH

COO-

-OOC

Cl

OH

COO-

CHO

COO-

O

COO-

OO

COO-

OH

COO-

COO-

O

COO-

COO-

OH O

COO-

O

COO-

O-

O

O

4-chlorophenol

4-chlorocatechol

3-chloro-cis,cis-muconate

cis-4-4carboxymethylene-but-2-ene-4-olide

maleylacetate

3-oxoadipate

5-chloro-2-hydroxy-muconic semialdehyde

2-hydroxymuconicsemialdehyde

2-hydroxy-2,4-pentadienoate

4-hydroxy-2-oxovalerate

pyruvate

4-chlorophenol

4-chlorocatechol

maleylacetate

hydroquinone

hydroxyquinol

5-chlorohydroxy-quinol

a. b.

ortho cleavage meta cleavage

Fig. 5. 4-chlorophenol degradation pathways.a) Conventional pathways (The University of Minnesota Biocatalysis/BiodegradationDatabase, Ellis and Wackett, 1997). b) The 4-chlorophenol degradation pathway in A.chlorophenolicus A6 (paper IV).

42

There are a few microorganisms that do not degrade 4-chlorophenol via cleavage of 4-chlorocatechol. Arthrobacter ureafaciens CPR706 (Bae et al., 1996a) was found totransform 4-chlorophenol to hydroquinone (1,4-dihydroxybenzene), but the pathwaywas not characterized further. Similarly, Nocardioides sp. NSP41 (Cho et al., 1998)also produced hydroquinone from 4-chlorophenol, but the entire pathway was notelucidated in that strain either. In A. chlorophenolicus A6, 4-chlorocatechol wasproduced from 4-chlorophenol, but the compound was not the ring cleavage substrate.Instead, 4-chlorocatechol was further transformed into hydroxyquinol, which wassubsequently cleaved to yield maleylacetate (paper IV, Fig. 5b). Also, a secondpathway branch was found, where 4-chlorophenol was transformed into hydroquinonewhich in turn was hydroxylated to yield hydroxyquinol (Fig. 5b). 4-chlorophenoldegradation had not previously been found to proceed via hydroxyquinol. However,hydroxyquinol and its chlorinated derivatives are commonly ring-cleavageintermediates in degradation of more highly chlorinated phenols (Häggblom, 1992;Häggblom and Valo, 1995; Latus et al., 1995; Padilla et al., 2000; Uotila et al., 1995;Zaborina et al., 1998). Hydroxyquinol has also been found to be the ring cleavagesubstrate in degradation of 4-nitrophenol (Jain et al., 1994) and 4-aminophenol(Takenaka et al., 2003).

From the general description of degradation pathways and enzymes above, severaltypes of enzymes would be expected to be involved in the 4-chlorophenol degradationpathway in A. chlorophenolicus A6 (Fig. 5b). These enzymes would be amonooxygenase for the initial transformations of 4-chlorophenol, an enzyme forreductive dechlorination of 5-chlorohydroxyquinol, a hydroxyquinol dioxygenase andfinally a maleylacetate reductase for funneling the product of ring cleavage into thecentral metabolism of the cell. Genes and open reading frames encoding these types ofenzymes were indeed found in the cph gene cluster in A. chlorophenolicus A6 (paperIV). The cluster was found to harbor two putative monooxygenases, twohydroxyquinol 1,2-dioxygenases, two putative maleylacetate reductases and twoputative regulatory genes. The fact that most of the enzymes appeared to be present intwo non-identical versions was surprising, but may be explained by the finding thathorizontal gene transfer appeared to have played a role in assembling of the genecluster (further discussed in chapter 8, Evolution of genes and operons for degradationof pollutants). However, a dechlorinase enzyme performing the transformation from 5-chlorohydroxyquinol to hydroxyquinol (Fig. 5b) was not found. Since the cph genecluster is not completely sequenced to date, additional genes may eventually beidentified.

A mutant A. chlorophenolicus A6 strain where one of the hydroxyquinol 1,2-dioxygenase genes was disrupted by a transposon was used to confirm that the cphgene cluster was involved in 4-chlorophenol degradation and also shed some light onthe degradation pathways for other phenolic compounds. The mutant strain, calledT99, grew poorly on 4-chlorophenol, showing that the hydroxyquinol 1,2-dioxygenasegene (cphA-I) was vital for effective growth on the compound (paper III, paper IV).Furthermore, when challenged with 4-nitrophenol or 4-bromophenol, strain T99 grewas poorly as when 4-chlorophenol was the sole carbon source, indicating that the same

43

enzyme system was used for degradation of all three compounds (paper III). This, inturn, would indicate that 4-nitrophenol and 4-bromophenol also are degraded viahydroxyquinol in A. chlorophenolicus A6. Interestingly, mutant T99 grew well onunsubstituted phenol, showing that another or an additional enzyme system isresponsible for metabolism of this compound.

The existence of two pathway branches leading from 4-chlorophenol to hydroxyquinolapparently active concurrently in A. chlorophenolicus A6 is intriguing. Thesimultaneous operation of two pathways for the degradation of a single aromaticcompound appears to be a fairly unusual trait. However, there are some reports of thisphenomenon, especially in the degradation of methylbenzene derivatives. Three upperpathways have been shown to operate at the same time for the transformation oftoluene in Burkholderia cepacia JS150 (Johnson and Olsen, 1997) and Rhodococcusrhodocrous OFS (Vanderberg et al., 2000). Another Rhodococcus, strain B3(Bickerdike et al., 1997), was found to degrade o-xylene both via oxidation of thearomatic ring and via oxidation of a methyl group, again with both pathways operatingsimultaneously. Several degradation pathways have also been shown to proceedsimultaneously in metabolism of polyaromatic hydrocarbons (PAHs) as exemplifiedby 1- and 2-methylnaphthalene degradation in P. putida CSV86 (Mahajan et al., 1994)and naphthalene degradation in Bacillus thermoleovorans Hamburg 2 (Annweiler etal., 2000). This type of metabolic redundancy may perhaps facilitate adaptation tocertain conditions or environments, giving the strain a selective advantage in nature.Alternatively, the existence of several pathway branches leading from 4-chlorophenolto hydroxyquinol in A. chlorophenolicus A6 may reflect the fact that twomonooxygenases appears to be encoded by the cph gene cluster. Possibly, themonooxygenases could be responsible for the transformations in one pathway brancheach. Alternatively, one monooxygenase could perform ortho-hydroxylations in bothbranches, and the other enzyme para-hydroxylations.

There are several possible explanations to why a novel 4-chlorophenol degradationpathway was found in A. chlorophenolicus A6. First of all, metabolism ofchlorophenols has mainly been studied in Gram-negative microorganisms such asPseudomonas and Ralstonia, and the 4-chlorophenol degradation pathway in A.chlorophenolicus A6 may simply reflect the most common strategy for degradation ofmonochlorophenols in Gram-positives. Another possibility stems from the fact that themicroorganism was isolated from an enrichment culture that was adapted to increasingamounts of 4-chlorophenol. During this process, the most efficient 4-chlorophenoldegrader was selected, and perhaps the hydroxyquinol pathway in A. chlorophenolicusA6 gave the strain a selective advantage, for example by providing high flux ofmetabolites through the catabolic sequence, or by avoiding accumulation of toxicmetabolites. Chlorocatechols are known to be toxic to microorganisms, and the ringcleavage reaction is often the bottleneck in degradation of aromatic compounds,especially when oxygen supply is suboptimal. Thus, by removing the chlorine atomand transforming 4-chlorocatechol to hydroxyquinol, accumulation of toxicchlorocatechols could perhaps be avoided in A. chlorophenolicus A6. Probably, as

44

more microorganisms from different genera and different environments are studied,there will be an increase in the diversity of the described degradation pathways.

45

7 Microbial degradation of pollutant mixtures

The ultimate goal of biodegradation research is often application of the process forbioremediation of contaminants in the environment. However, laboratory studies ofbiodegradation are usually performed with a single pollutant, whereas a polluted siteoften is contaminated with more than one compound. Pollutants that may be present asmixtures are for example BTEX compounds, pesticides (e.g. benzene derivatives,organophospourous compounds and organometals) and chemicals used for woodpreservation such as chlorobenzenes, chlorophenols and polyaromatic hydrocarbons(PAHs) (Swoboda-Colberg, 1995). Moreover, additional carbon-containingcompounds can serve as potential growth substrates and complicate degradation ofpollutants. For example, in a soil environment, carbon sources may become availablevia hydrolysis of naturally occurring polymeric material such as cellulose, lignin,chitin, proteins, nucleic acids and waxes (Egli, 1995). Numerous studies have shownthat mixed growth substrates often affect each other’s degradation (Kovárová-Kovarand Egli, 1998). The effects vary both with the microorganism involved and thegrowth substrates and are therefore difficult to predict. Thus, substrate interactionsneed to be carefully considered in order to optimize the efficacy of bioremediation.

7.1 Diauxic growth

In diauxic growth, one growth compound or its metabolites represses the synthesis ofthe enzymes required for degradation of a second substrate, delaying metabolism ofthe second substrate until the first is exhausted. Diauxic growth was previouslythought to be the norm for degradation of mixed substrates (Kovárová-Kovar and Egli,1998). The most famous example is the lac operon in Esherichia coli, which encodesenzymes for lactose metabolism. Expression of the genes in the lac operon is repressedin the presence of glucose, and hence glucose is degraded before lactose in a mixtureof the two growth compounds. Diauxie is typically accompanied by a biphasic growthcurve, where the cell density reaches a plateau when the first compound is exhausted.During this time, the genes for degradation of the second compound are induced, andgrowth eventually recommences. Diauxic growth may also occur in pollutantbiodegradation, as the presence of organic acids and carbohydrates has been shown torepress enzymes for catabolism of pollutants. For example, a number of organic acids(such as succinate) and carbohydrates (such as glucose) repress transcription of thegenes for phenol degradation in Pseudomonas putida H (Müller et al., 1996). Inanother study, the presence of benzoate completely abolished phenol degradation byRalstonia eutropha (Ampe et al., 1998). Moreover, the presence of two or morepollutants may also cause diauxic growth. This was seen during growth ofComamonas testosteroni JH5 on a mixture of 4-methylphenol and 4-chlorophenol(Hollender et al., 1994), and by Pseudomonas fluorescens PC18 on a mixture ofphenol and methylphenol (Heinaru et al., 2001).

46

7.2 Simultaneous degradation of substrates

Two compounds that are incompatible at high concentrations are often degradedsimultaneously at low concentrations (Kovárová-Kovar and Egli, 1998). This may bean adaptation to life in nutrient-limited environments such as soil, where the availablegrowth substrates are often a myriad of carbon sources, each present at a very lowconcentration. In fact, carbon starvation itself has been shown to induce expression ofa number of catabolic enzymes, presumably so that the cells are ready if thesecompounds become available (Egli 1995; Kovárová-Kovar and Egli, 1998). Twocompounds may also be degraded simultaneously at high concentrations, and this isoften accompanied by an increased maximum specific growth rate as compared togrowth on each compound by itself (Kovárová-Kovar and Egli, 1998). Such anincreased growth rate may be due to the induction of two separate degradationpathways at the same time, thus increasing the flux of carbon into the centralmetabolism of the cell. The degradation of two pollutants simultaneously by a singlemicroorganism has been seen for example during degradation of a mixture of phenoland 4-methylphenol by Pseudomonas mendocina PC1 (Heinaru et al., 2001) anddegradation of toluene and benzene by Pseudomonas putida F1 (Reardon et al., 2000).

The effect of substrate mixtures on degradation of phenols by A. chlorophenolicus A6was addressed in paper III. In addition to 4-chlorophenol, the microorganism can alsogrow on 4-nitrophenol, 4-bromophenol, 4-fluorophenol and unsubstituted phenol assole carbon sources (paper I, Fig. 6 a and b). When the microorganism was grown onmixtures of phenolic compounds, a distinct pattern of preference for the differentsubstrates could be seen. In mixtures of 4-nitrophenol and 4-chlorophenol, 4-nitrophenol was the preferred substrate (Fig. 6c, paper III). When unsubstituted phenolwas also present in the cultures, this compound was removed last, with degradationstarting when 4-chlorophenol was nearly depleted (paper III). On the other hand, 4-chlorophenol and 4-bromophenol were clearly degraded simultaneously (paper III).

When a degradation pattern as the one in Figure 6c is seen, the question immediatelyarises whether the observed behavior is due to diauxic growth. Two approaches wasused in paper III to address this question. First, the growth and degradation data wasinvestigated. If diauxie occurred, there would likely be a plateau in OD in thetransition between growth on the different phenols. No such plateau could be detectedby a visual inspection of the data (Fig. 6c). In order to study this more closely, thegrowth and degradation data was subjected to kinetic analysis which also indicatedthat diauxie did not occur. Secondly, the A. chlorophenolicus A6 mutant T99 wasemployed. This strain carries a transposon in a hydroxyquinol 1,2-dioxygenase genewhich leads to severely impaired growth on 4-chlorophenol (paper IV). Experimentsshowed that the T99 mutant behaved in exactly the same way when 4-nitrophenol or 4-bromophenol was the sole carbon source, indicating that the same enzyme system wasused as for 4-chlorophenol degradation. Thus diauxie would not occur.

47

0

0.2

0.4

0.6

0.8

1

0 1 2 3 4 5 6 7 8

time (h)

conc

entr

atio

n (m

M)

0.08

0.1

0.12

0.14

0.16

0.18

0.2

0.22

0.24

OD

(600 nm)

a

0

0.2

0.4

0.6

0.8

1

0 1 2 3 4 5 6 7 8time (h)

conc

entr

atio

n (m

M)

0.08

0.1

0.12

0.14

0.16

0.18

0.2

0.22

0.24

OD

(600 nm)

0

0.2

0.4

0.6

0.8

1

0 1 2 3 4 5 6 7 8

time (h)

conc

entr

atio

n (m

M)

0.08

0.1

0.12

0.14

0.16

0.18

0.2

0.22

0.24

OD

(600 nm)

c

b

Fig. 6. Growth of A. chlorophenolicus A6 on a) 4-chlorophenol b) 4-nitrophenol c) amixture of 4-chlorophenol and 4-nitrophenol. Filled circles: OD. Open squares: 4-chlorophenol. Open triangles: 4-nitrophenol. The brackets denote standard deviation(three replicates).

Contrary to the results with 4-nitrophenol and 4-bromophenol, the T99 mutant grewwell on unsubstituted phenol. Also, phenol as a growth substrate appeared to affect A.chlorophenolicus A6 differently than the substituted phenols tested: much higherconcentrations could be tolerated and growth was slower on this compound (paper III).

48

If the same enzyme system is used for degradation of 4-chlorophenol and 4-nitrophenol, why is one compound preferred over the other? A possible explanationcould be that the first enzyme in the degradation pathway has a higher affinity to 4-nitrophenol than to 4-chlorophenol. Alternatively, there could be transport-levelinteractions. For example, if A. chlorophenolicus A6 only takes up the phenolate ionform of the phenols, the differing pKa values of the compounds would affect theirorder of degradation. 4-nitrophenol has a pKa of 7.1, whereas it is 9.4 for 4-chlorophenol as well as 4-bromophenol and 10.0 for phenol. This means that at pH 7.0– 7.2 where the experiments were performed, 50% of the 4-nitrophenol was present inits dissociated form, i.e. as a phenolate ion, whereas only 0.6% of the 4-chlorophenoland 0.2% of the phenol were present in this form. Presuming that only phenolate ionswere taken up into the cells, 4-nitrophenol would be preferred over 4-chlorophenol.This hypothesis could even be extended to encompass phenol. Also, the fact that 4-chlorophenol and 4-bromophenol, with the same pKa, were degraded simultaneouslyfits this hypothesis.

7.3 Rate deceleration

In addition to causing diauxic growth, the presence of several growth substrates mayalso affect the rate of degradation of the individual compounds. Even if an inhibitoryeffect does not lead to diauxie, degradation of one or several compounds may bedecelerated in a mixture. For example, ethanol concentrations over a certain thresholdwere shown to have a negative effect on the degradation of benzene by Pseudomonasputida F1 (Lovanh et al., 2002), and phenanthrene degradation by two Pseudomonasspecies was subject to competitive inhibition by several other PAHs (Stringfellow andAitken, 1995). Rate deceleration was observed in substrate mixtures for A.chlorophenolicus A6 as well; for example 4-nitrophenol degradation was slower when4-chlorophenol was present as an additional substrate than when 4-nitrophenol was thesole substrate. This was reflected in a decreasing specific growth rate with increasingconcentrations of 4-chlorophenol and 4-nitrophenol (paper III). The reason for slowerdegradation rates of pollutants in mixtures may be the increased toxicity caused by theadditional compounds, the formation of intermediates that inhibit enzymes vital fordegradation, or competition for uptake into the cell or for the active binding site of anenzyme (Alexander, 1999).

7.4 Rate acceleration

The presence of one compound may also enhance the degradation of another, as in theenhanced mineralization of carbazole in groundwater when fluorene was present(Millette et al., 1995), the improved biodegradation of phenanthrene and fluorene in amixture also containing naphthalene (Guha et al., 1999) and 3-CP degradation byAlcacligenes sp. A7-2 which was made possible by the addition of phenol (Menke andRehm, 1992). This enhancement of removal of a pollutant may be due to the greaterbiomass formed because of the additional carbon source available, or one compoundmay induce enzymes beneficial for degradation of the second compound (Alexander,

49

1999). For degradation of some pollutants, the presence of another substrate is aprerequisite. This is called cometabolism and is usually due to the fact that the sameenzyme system is active on both compounds, but the compound being co-metabolizeddoes not serve as an inducer. Instead, its degradation is a fortuitous reaction, oftenwithout benefit to the cell. For example, highly chlorinated PCBs are degraded bycometabolism with biphenyl or another suitable inducer such as carvone.

50

8 Evolution of genes and operons for degradation of pollutants

During the last century, microorganisms have acquired the ability to degradexenobiotic compounds such as DDT, PCBs and chlorobenzenes. How did these noveldegradation pathways come into existence? Some biochemical pathways appear tohave been built up in a backward fashion (Horowitz, 1945), for example the glycolyticpathway (Fothergill-Gilmore and Michels, 1993). The reasoning behind thishypothesis is that the first microorganisms lived in a rich organic soup, but eventuallysome compounds became depleted and an organism that could synthesize thismolecule from a precursor came to have a selective advantage. In the next step, therewas selection for a microorganism that could synthesize the precursor from anothercompound, and so on until an entire pathway was built up. However, todaymicroorganisms do not live in a rich organic soup, but instead in harsh environmentssuch as soil or ground water, where competition for carbon sources is fierce. Instead ofbeing built up in a backward fashion such as described by Horowitz, the degradationpathways for xenobiotic compounds are thought to evolve by the recruitment ofenzymes from other metabolic pathways and alteration of the corresponding genes bymutation and recombination (Copley, 2000; van der Meer, 1997). Well-studied centralbiochemical pathways such as the Krebs cycle are also thought to have evolved in thisfashion (Meléndez-Hevia et al., 1996).

8.1 Modes of evolution

8.1.1 Mutational drift

The substrate range or effector specificity of an enzyme may be altered by pointmutations in its corresponding gene (van der Meer et al., 1992). For example,Seffernick et al. (2001) found that only 9 amino acids differed between two enzymesperforming distinct biochemical reactions, with no overlap of activity between theenzymes. In another study, mutagenic PCR targeted to the sensor domain of theregulatory protein DmpR could be used to alter the specificity of the regulatory systemin order to detect priority pollutant phenols (Wise and Kuske, 2000). Othermechanisms which give rise to changes in the primary DNA sequence are, forexample, slippage of DNA polymerase during replication, erroneous DNA repair andgene conversion (van der Meer, 1997; van der Meer et al., 1992). However, as theresults of these changes usually are relatively minor, they are not likely to account forthe large changes that allow microbes to exploit new environments (Ochman et al.,2000; van der Meer et al., 1992).

8.1.2 Genetic rearrangement within a cell

Recruitment of genes to a novel pathway may occur with the help of the cell’srecombination system. DNA segments can be rearranged or relocated to a newposition in the genome whenever there is sequence similarity between them, and assmall regions of sequence identity as 4 base pairs may be sufficient for recombination(van der Meer, 1997). Genes may also be recruited via gene duplication followed by

51

divergence of one of the copies. A selective advantage may rapidly be gained if theenzyme encoded by the duplicated gene already has a low activity in another reaction(O’Brien and Herschlag, 1999). Larger regions of DNA may also be duplicated. Forexample, Nakatsu et al. (1998) found that in several replicate populations of 2,4-Ddegrading Ralstonia sp. TFD41, a large segment of the plasmid coding for 2,4-Ddegradation had been duplicated after 1000 generations. The movement of IS elementswithin a cell may also serve to recruit genes to a novel pathway since they oftencontain promotor-like elements which may activate silent genes (van der Meer et al.,1992).

8.1.3 Horizontal gene transfer

Horizontal gene transfer is viewed as the major source of bacterial evolution (Kooninet al., 2001; Ochman et al., 2000). Genome sequencing has shown that bacteriacontain mosaics of horizontally acquired genes, and the proportion of genes of foreignorigin has been shown to vary greatly, from almost none in, for example, Rickettsiaprowazekii and Mycobacterium genitalium to almost 17% in a Synechocystis strain(Ochman et al., 2000). In bacteria, genes encoding enzymes of a degradation pathwayare often found clustered together in the genome, and different gene clusters may becombined as modules (for degradation of aromatic compounds, genes for peripheralpathways may for example be combined with different ring cleavage enzyme genes)(van der Meer et al., 1992). An effect of this rearrangement is that operons withdegradative genes often are observed to have many genes in common, but in differentcombinations (van der Meer, 1997).

Plasmids are the most common vehicles for horizontal transfer of genes encodingenzymes for pollutant degradation (Alexander, 1999). Natural plasmids are often verylarge (75-100 kb) and frequently self-transmissible (Reineke, 1998; Tan, 1999; van derMeer et al., 1992). Genes may also be transferred by transposons, and so-calledcatabolic transposons, which contain entire catabolic operons, have been found (Tan,1999). Moreover, transposons are frequently located on self-transmissible plasmids,facilitating their dissemination, and their movement can lead to capture andmobilization of other genes (Reineke, 1998; Tan, 1999). Natural gene transfer amongbacteria can also occur via uptake of naked DNA (transformation) and via phages(transduction) (Yin and Stotzky, 1997).

Although it is difficult to witness gene transfer events occurring in nature, clues maybe found in a DNA sequence that such an event has taken place. Indications ofhorizontal gene transfer may be found by the observation of differences in thetransferred DNA compared to the vertically transmitted (ancestral) DNA. For example,widely differing molar percentage of guanosine and cytosine (G+C) betweenneighboring genes is an indication of a past gene transfer event since although thepercentage of G+C varies greatly between species, the value tends to be consistentwithin a strain (Ochman et al., 2000; Shapiro, 1999). Also, a difference in the patternof codon bias between genes may indicate a past horizontal gene transfer event(Ochman et al., 2000; Shapiro, 1999). The rationale behind codon bias is that although

52

several three-letter codons may in most cases be used to specify a single amino acid,many strains do not show equal use of every codon. Codon bias may also be linked tothe G+C percentage of the microorganism’s DNA (Bibb et al., 1984). Clues indicatingpast horizontal gene transfer events may also be found in the regions flanking thetransferred DNA, since these often contain genes or remnants of genes affectingintegration of DNA, such as parts of IS elements or transposons, attachment sites ofphage integrases and transfer origins of plasmids (Ochman et al., 2000).

An example of a set of genes where all of the indications of horizontal gene transfermentioned above were found is the cph gene cluster which is involved in 4-chlorophenol degradation by A. chlorophenolicus A6 (top of Fig. 7, paper IV). Thepercent G+C content of certain genes in the cluster was much lower than that of thegenome, the codon bias of the same genes was lower than surrounding genes and aresolvase pseudogene probably originating from a transposon was found in the vicinityof these genes (Fig. 7, paper IV). As A. chlorophenolicus A6 was isolated from anenrichment culture gradually adapted to increasing amounts of 4-chlorophenol, it ispossible that new genes were acquired by the strain from another organism in the soilcommunity during the adaptation period. Another indication of horizontal genetransfer to this microorganism is that several of the genes and open reading framesappear to encode enzymes that perform the same type of reactions. For example, twofunctional hydroxyquinol 1,2-dioxygenases were found, as well as open readingframes for two monooxygenases and two maleylacetate reductases (paper IV). Geneduplication did not appear to have caused this, as the DNA sequences encodingenzymes with similar functions were distinct from one another, and sometimesappeared to be completely unrelated (paper IV).

53

F

ig. 7. Gene clusters w

ith hydroxyquinol 1,2-dioxygenase genes. The m

icroorganism the cluster w

as cloned from and the com

poundw

hich the organism degrades are noted on the right. G

enBank/E

MB

L accession num

bers as follows: A

Y131335 for the cph genes from

A.

chlorophenolicus A6, accession num

ber pending for the npd genes from A

rthrobacter sp. JS443 (L. L

. Perry and G. J. Z

ylstra, personal comm

unication),A

B016258 for the O

RFs from

Arthrobacter sp. B

A-5-17, A

F498371 for the tcp genes from R

alstonia eutropha JMP134, D

86544 for the had genes fromR

alstonia pickettii DT

P0602, U19883 for the tft genes from

Burkholderia cepacia A

C1100 and X

72850 for the dxn genes from Sphingom

onas sp. RW

1.

54

8.2 Evolution of genes and operons for intradiol ring cleavage

Biodegradation of halogenated aromatic compounds commonly involves enzymescatabolizing intradiol ring cleavage. An ancestral pathway for degradation of naturallyoccurring haloaromatics has been proposed to be the origin of a set of chlorocatecholcatabolic genes which then presumably spread in the environment after the advent ofindustrial use of chloroaromatics in the 20th century (Schlömann 94). Interestingly,adaptation to chlorocatechol metabolism appears to have occurred independently inGram-negative and Gram-positive microorganisms (Eulberg et al., 1998) and resultspresented by Moisheeva et al. (2002) indicate that chloroaromatic metabolism mayhave arisen at least twice in Gram-positives.

The best studied intradiol dioxygenases are protocatechuate 3,4-dioxygenase, catechol1,2-dioxygenase and chlorocatechol 1,2-dioxygenase, which all show sequencesimilarity to one another (van der Meer, 1997). Protocatechuate 3,4-dioxygenaseconsists of two different subunits which may have arisen through gene duplication(van der Meer, 1997). The catechol 1,2-dioxygenases of Gram-positive and Gram-negative bacteria form separate branches on a phylogenetic tree of these enzymes(Eulberg et al., 1998).

Hydroxyquinol 1,2-dioxygenases were found in 1995 to constitute a separate class ofintradiol dioxygenases in prokaryotes (Daubaras et al., 1995). Since then, severaladditional hydroxyquinol 1,2-dioxygenases have been found, many as hypotheticalproteins from genome sequencing projects. To date, the function of nine of theseproteins has been verified (Armengaud et al., 1999; Daubaras et al., 1996b; Hatta etal., 1999; Johnson et al., 2002; Louie et al., 2002, Murakami et al., 1999; L.L. Perryand G.J. Zylstra, personal communication; paper IV). A phylogenetic tree constructedfrom the amino acid sequences of these hydroxyquinol 1,2-dioxygenases shows thatthe four sequences from Gram-positive microorganisms cluster together (CphA-I,NpdB, CphA-II and ORF2 in Fig. 8). Incidentally, all these sequences originate fromArthrobacter strains, so it is not possible to determine if this clustering represents aspecial branch of the phylogenetic tree for Gram-positive microorganisms or if it justreflects the fact that all of the sequences originate from the same genus.

Many of the studied hydroxyquinol 1,2-dioxygenases are believed to have beenacquired by horizontal gene transfer. For example, ORF3 and TftH (Fig. 8) are bothcloned from Burkholderia cepacia, and would normally be expected to cluster togetherin the tree, as they originate from the same species and perform the same reaction.However, they do not cluster together and the hypothesis that ORF3 and possibly alsoTftH were acquired via lateral gene transfer, explains this phenomenon (Daubaras etal., 1996a; Johnson et al., 2002). Other examples of gene clusters where horizontalgene transfer is implicated is the cph gene cluster of A. chlorophenolicus A6 (paperIV) and the dxn genes in Sphingomonas sp. RW1 (Armengaud et al., 1999). Both ofthese strains contain transposon-derived genes in the vicinity of the hydroxyquinol 1,2-dioxygenase genes, indicating a past horizontal gene transfer.

55

Fig. 8. Phylogenetic tree of hydroxyquinol 1,2-dioxygenases. The proteins are labeledusing trivial abbreviation and strain. Catechol 1,2-dioxygenase from Pseudomonasputida mt-2 was used as an outgroup. The numbers on the branches indicate how manytimes the group outside the fork occurred among 100 trees generated in a bootstrapanalysis. The bar represents 0.1 expected changes per amino acid position.GenBank/EMBL accession numbers as follows: CAA51371 for hydroxyquinol 1,2-dioxygenase fromSphingomonas sp. RW1 (DxnF RW1), I40182 for hydroxyquinol 1,2-dioxygenase from Burkholderiacepacia AC1100 (TftH AC1100), AAL50018 for putative hydroxyquinol oxygenase fromBurkholderia cepacia R34 (ORF3 R34), BAA13107 for hydroxyquinol 1,2-dioxygenase fromRalstonia pickettii DTP0602 (HadC DTP0602), AAM55216 for 6-chloro-hydroxyquinol 1,2-dioxygenase from Ralstonia eutropha JMP134 (TcpC JMP134), BAA82713 for hydroxyquinol 1,2-dioxygenase from Arthrobacter sp. BA-5-17 (ORF2 BA-5-17), S66469 for catechol 1,2-dioxygenasefrom Pseudomonas putida mt-2 (CatA mt-2) and accession number is pending for hydroxyquinol 1,2-dioxygenase from Arthrobacter sp. JS443 (NpdB JS443) (L. L. Perry and G. J. Zylstra, personalcommunication).

A comparison of the architecture of the gene clusters harboring hydroxyquinol 1,2-dioxygenases illustrates the point that similar types of catabolic genes may becombined in different order in different organisms (Fig. 7). (The gene cluster fromBurkholderia cepacia R34 was excluded from the figure as the hydroxyquinol 1,2-dioxygenase in the cluster is thought to have no function in the actual degradation

56

pathway (Johnson et al., 2002)). Genes for maleylacetate reductases, monooxygenasesand oxidoreductases (oxidoreductases may transfer electrons to monooxygenases) arefound in many of the gene clusters in Figure 7, but the order of the genes is notconserved. Some trends may be discerned, however, using the hydroxyquinol 1,2-dioxygenase gene as a starting point. In four clusters, an oxidoreductase is founddirectly upstream of the hydroxyquinol 1,2-dioxygenase, and in three clusters, amaleylacetate reductase is found in this position. Interestingly, in A. chlorophenolicusA6, the only organism found to contain two hydroxyquinol 1,2-dioxygenases, one ofthese genes is preceded by an oxidoreductase and the other by a maleylacetatereductase (paper IV). Another striking feature of Figure 7 is that the npd gene clusterfrom Arthrobacter sp. JS443 and part of the cph cluster have a very similar gene order.In addition to the genes labeled in the figure, cphS and cphR are thought to beregulatory genes, as is npdR (L. L. Perry and G. J. Zylstra, personal communication;paper IV). This similarity in gene order is interesting as there is evidence that severalof the cph genes in A. chlorophenolicus A6 were acquired via horizontal gene transfer(paper IV). Perhaps some of the genes for metabolism of aromatic compounds in A.chlorophenolicus A6 and Arthrobacter sp. JS443 originate from the same ancestor.

57

9 Conclusions

Major findings in this thesis are:

® Arthrobacter chlorophenolicus A6 is a novel species.

® A. chlorophenolicus A6 can grow on unusually high concentrations of 4-chlorophenol without the need for any cofactors or vitamins.

® A. chlorophenolicus A6 was successfully tagged with the gfp or luc genes andmonitored during 4-chlorophenol degradation in soil microcosms.

® The luc-tagged and gfp-tagged A. chlorophenolicus A6 strains can remove 180µg/g 4-chlorophenol from non-sterile soil microcosms.

® A. chlorophenolicus A6 can grow on several phenolic compounds, and whenthese are present in mixtures, distinct patterns of substrate preferences are seen.

® In A. chlorophenolicus A6, 4-chlorophenol is degraded via an unusual pathwaywhere hydroxyquinol is the ring cleavage substrate.

® A gene cluster involved in 4-chlorophenol degradation has been cloned andsequenced. Analysis of the DNA sequence of the gene cluster indicated thathorizontal gene transfer has been involved in its assembly.

58

10 Future perspectives

An immediately obvious continuation of the work presented in this thesis would be toclone and sequence more genes in the cph cluster. Part of a monooxygenase gene isstill missing, and judging by the reactions in the 4-chlorophenol degradation pathwaythere should be a dechlorinase gene present as well. However, a dechlorinase genewould not necessarily need to be present in the cph cluster but could be locatedelsewhere in the genome of A. chlorophenolicus A6.

Furthermore, it would be highly interesting to study the initial steps of the 4-chlorophenol degradation pathway, from 4-chlorophenol to hydroxyquinol, in moredetail. This could be accomplished by overexpression of the two putativemonooxygenases found in the cph gene cluster. In the degradation pathway leadingfrom 4-chlorophenol to hydroxyquinol, both a para-hydroxylation and an ortho-hydroxylation occur, and perhaps one monooxygenase performs the para-hydroxylation and the other the ortho-hydroxylation. Alternatively, the enzymes couldbe involved in one pathway branch each. Overexpression of the enzymes andspectrophotometric study of the reactions as performed for the two hydroxyquinol 1,2-dioxygenases should be a suitable approach as both substrates and products absorblight in the UV range.

Perhaps it would be possible to expand the substrate range of A. chlorophenolicus A6by gene shuffling or some other protein engineering method. The monooxygenasesinvolved in the transformation of 4-chlorophenol to hydroxyquinol could maybe bealtered so that 2-chlorophenol and 3-chlorophenol could also be transformed intohydroxyquinol and further metabolized.

Another interesting field of study would be to investigate the pigment formation by A.chlorophenolicus A6 more closely. The strain produces a pigment from severalphenols such as 2-chlorophenol and 3-chlorophenol, and also from 4-chlorophenol ifthe cultures are not sufficiently oxygenated. The pigment is initially red in color andeventually darkens to black, and its chemical characteristics is consistent with apolyphenolic structure, related to melanins. Perhaps 2-chlorophenol and 3-chlorophenol are transformed to 2-chlorohydroquinone, but this compound cannot betransformed further to hydroxyquinol because of the chloro-substituent.Hydroquinones form benzoquinones spontaneously in the minimal medium used forgrowing A. chlorophenolicus A6 on 4-chlorophenol and as benzoquinones are knownto be reactive they could initiate a polymerization process leading to pigmentformation. The T99 mutant where the hydroxyquinol 1,2-dioxygenase gene cphA-I hasbeen disrupted may be useful in the study of pigment formation. This mutant producespigment from 4-chlorophenol even in cultures which are sufficiently oxygenated. Anintriguing fact is that pigment production by strain T99 is stronger at low 4-chlorophenol concentrations than at high concentrations.

59

It would also be of interest to determine how 4-chlorophenol degradation in A.chlorophenolicus A6 is regulated. We know that some sort of regulation is takingplace as 4-chlorophenol degradation does not occur while some other substrates, suchas succinate, are present. However, we do not know how this regulation is broughtabout. Is 4-chlorophenol or an intermediate the effector, and are CphR and CphSinvolved as their deduced amino acid sequences imply?

When A. chlorophenolicus A6 was grown on a mixture of 4-nitrophenol and 4-chlorophenol, the strain exhibited a peculiar behavior in that 4-nitrophenol waspreferred over 4-chlorophenol. One hypothesis put forward to explain thisphenomenon is that the phenolate ions, and not the phenols, are taken up into the cellsand metabolized. If this is true, the pH of the medium should influence the degree ofoverlap of 4-nitrophenol and 4-chlorophenol degradation. At a higher pH, the overlapof degradation of the compounds should also increase. Thus this hypothesis could betested, supposing that A. chlorophenolicus A6 would be able to grow at a pH of around8.0 or higher.

Maybe the basis of the extreme 4-chlorophenol tolerance exhibited by A.chlorophenolicus A6 is that the microorganism is able to slow the compound enteringthe cell. This would limit the uncoupling effect of 4-chlorophenol. To investigate this,NMR could be used to study deenergization of the cell membrane and depletion ofATP upon exposure to 4-chlorophenol, and A. chlorophenolicus A6 could becompared with other microorganisms. In this context, it would also be interesting toinvestigate if A. chlorophenolicus A6 alters its cellular fatty acid content in response to4-chlorophenol, as other microorganisms have been shown to use such a strategy toprotect themselves from phenolic compounds.

60

11 Acknowledgments

Thanks to my supervisor Janet Jansson, for providing a stimulating researchenvironment, allowing me freedom to explore my ideas and not laughing at my wilderones… Thanks for taking me on in the research group when I showed up in your officeone day wanting to work with “Not E. coli”. I have learned so much in your lab!

Thanks to Stefan Nordlund at the Department of Biochemistry and Biophysics atStockholm University and Eilert Karpberg at the Department for Natural Sciences atSödertörn University College, and the administrative and technical staff at both places,for creating the environment where this research could take place. I especially want tomention Ann, Kicki and Anki at Biochemistry, SU, you have been very helpfulthrough moves and other changes.

Thanks to John Stenström and Erko Stackebrandt for fruitful collaboration.

Thanks to Bill Mohn, supervisor for my “examensarbete” at University of BritishColumbia, Vancouver. Thanks for believing in my abilities more than I believed inthem myself; without the knowledge and self-confidence I gained in your lab thisthesis would have taken much longer time to finish.

Thanks to the people who got me interested in research in the first place by taking meon as a summer student: Marie Öhman, Patrick Young and Britt-Marie Sjöberg at theDepartment of Molecular Biology, Stockholm University. Without you, who knows ifI would have done a Ph. D. degree at all?

Thanks to Mia Unell for continuing my research projects in such a talented andsuccessful way and good luck in the future!

Thanks to all the members of Janet’s research group in Stockholm, past and present,for answering all my questions and being there making the lab an inspiring and funwork-place; Annelie, Riccardo, Annika, Cia, Ninwe, Agneta, Anna, Veronika, Lu,Berndt, Elisabeth, our frequent visitor Kersti and our neighbors Sara and Fredrik. Aspecial thanks to “the girls” – Annika, Cia, Ninwe and Agneta – whom I’ve sharedoffices with for so many years. No issue is too big or too small to discuss with you!

To all my friends and relatives outside science who endured hours of my obsessingabout problems with my “DNA-soup” (as someone called it)… I’m done now!

Last but not least, I thank my family, your support has been the most valuable. Thanksmamma for always being willing to discuss my research through ups and downs.Finally, a special thanks to Lars-Arvid and Isak who both helped me finish this thesis –you’re the best!

61

12 References

Adriaens, P., and T. M. Vogel. 1995. Biological treatment of chlorinated organics, p. 435-486. InL. Y. Young and C. E. Cerniglia (ed.), Microbial Transformation and Degradation of ToxicOrganic Chemicals. Wiley-Liss, New York, N.Y.

Alexander, M. 1999. Biodegradation and Bioremediation, 2nd ed. Academic Press, San Diego,CA.

Ampe, A., D. Léonard, and N. D. Lindley. 1998. Repression of phenol catabolism by organicacids in Ralstonia eutropha. Applied and Environmental Microbiology 64:1-6.

Andersen, J. B, C. Sternberg, L. K. Poulsen, S. P. Björn, M. Givskov, and S. Molin. 1998. Newunstable variants of green fluorescent protein for studies of transient gene expression inbacteria. Applied and Environmental Microbiology 64:2240-2246.

Annweiler, E, H. H. Richnow, G. Antranikian, S. Hebenbrock, C. Garms, S. Franke, W. Francke,and W. Michaelis. 2000. Naphthalene degradation and incorporation of naphthalene-derivedcarbon into biomass by the thermophile Bacillus thermoleovorans. Applied and EnvironmentalMicrobiology 66:518-523.

Armengaud, J., K. N. Timmis, and R. M. Wittich. 1999. A functional 4-hydroxysalicylate/hydroxyquinol degradative pathway gene cluster is linked to the initialdibenzo-p-dioxin pathway genes in Sphingomonas sp. strain RW1. Journal of Bacteriology181:3452-3461.

Atlas R. M., and R. Unterman. 1999. Bioremediation, p. 666-681. In A. C. Demain and J. E.Davies (ed.), Manual of Industrial Microbiology and Biotechnology, 2nd ed. ASM Press,Washington, D. C.

Backman, A., and J. K. Jansson. 2004. Degradation of 4-chlorophenol at low temperature andduring extreme temperature fluctuations by Arthrobacter chlorophenolicus A6. MicrobialEcology (in press).

Backman, A., N. Maraha, and J. K. Jansson. 2004. Impact of temperature on the physiologicalstatus of a potential bioremediation inoculant, Arthrobacter chlorophenolicus A6. Applied andEnvironmental Microbiology (in press).

Bae, H. S., J. M. Lee, and S.-T. Lee. 1996a. Biodegradation of 4-chlorophenol via a hydroquinonepathway by Arthrobacter ureafaciens CPR706. FEMS Microbiology Letters 145:125-129.

Bae, H. S., J. M. Lee, Y. B. Kim, and S.-T. Lee. 1996b. Biodegradation of the mixtures of 4-chlorophenol and phenol by Comamonas testosteroni CPW301. Biodegradation 7:463-469.

Bartels, I., H.-J. Knackmuss, and W. Reineke. 1984. Suicide inactivation of catechol 2,3-dioxygenase from Pseudomonas putida mt-2 by 3-halocatechols. Applied and EnviromentalMicrobiology 47:500-505.

Behor, O., D. R. Smulski, T. K. Van Dyk, R. A. LaRossa, and S. Belkin. 2002. Recombinantmicroorganisms as environmental biosensors: pollutants detection by Escherishia coli bearingfabA’::lux fusions. Journal of Biotechnology 94:125-132.

Bibb, M. J., P. R. Findlay, and M. W. Johnson. 1984. The relationship between base compositionand codon usage in bacterial genes and its use for the simple and reliable identification ofprotein-coding sequences. Gene 30:157-166.

Bickerdike, S. R., R. A. Holt, and G. M. Stephens. 1997. Evidence for metabolism of o-xylene bysimultaneous ring and methyl group oxidation in a new soil isolate. Microbiology 143:2321-2329.

Boldt, Y. R., M. J. Sadowsky, L. B. Ellis, L. Que Jr, and L. P. Wackett. 1995. A manganese-dependent dioxygenase from Arthrobacter globiformis CM-2 belongs to the major extradioldioxygenase family. Journal of Bacteriology 177:1225-1232.

Bonten, L. T. C., T. C. Grothenius, and W. H. Rulkens. 1999. Enhancement of PAHbiodegradation in soil by physicochemical pretreatment. Chemosphere 38:3627-3636.

62

Bowlen, G. F., and D. S. Kosson. 1995. In situ processes for bioremediation of BTEX andpetroleum fuel products, p. 515-544. In L. Y. Young and C. E. Cerniglia (ed.), MicrobialTransformation and Degradation of Toxic Organic Chemicals.Wiley-Liss, New York, N.Y.

Breivik, K., A. Sweetman, J. M. Pacyna, and K. C. Jones. 2002. Towards a global historicalemission inventory for selected PCB congeners – a mass balance approach 1. Globalproduction and consumption. The Science of the Total Environment 290:181-198.

Cameron, M. D., S. Timofeevski, and S. D. Aust. 2000. Enzymology of Phanerochaetechrysosporium with respect to the degradation of recalcitrant compounds and xenobiotics.Applied Microbiology and Biotechnology 54:751-758.

Cassidy, M. B., H. Lee, J. T. Trevors, and R. B. Zablotowicz. 1999. Chlorophenol andnitrophenol metabolism by Sphingomonas sp UG30. Journal of Industrial Microbiology andBiotechnology 23:232-241.

Chalfie, M., T. Yuan, G. Euskirchen, W. W. Ward, and D. C. Prasher. 1994. Green fluorescentprotein as a marker for gene expression. Science 263:802-805.

Chen, W., F. Brühlmann, R. D. Richins, and A. Mulchandani. 1999. Engineeering of improvedmicrobes and enzymes for bioremediation. Current Opinion in Biotechnology 10:137-141.

Chih-Jen, L., L. Chi-Mei, and H. Chiou-Zong. 1996. Biodegradation of chlorophenols byimmobilized pure-culture microorganisms. Water Science and Technology 34:67-72.

Cho, Y.-G., J.-H. Yoon, Y.-H. Park, and S.-T. Lee. 1998. Simultaneous degradation of p-nitrophenol and phenol by a newly isolated Nocardioides sp. Journal of General and AppliedMicrobiology 44:303-309.

Copley, S. 1997. Diverse mechanistic approaches to difficult chemical transformations: microbialdehalogenation of chlorinated aromatic compounds. Chemistry & Biology 4:169-174.

Copley. S. D. 2000. Evolution of a metabolic pathway for degradation of a toxic xenobiotic: thepatchwork approach. Trends in Biochemical Sciences 25:261-265.

Corish, P., and C. Tyler-Smith. 1999. Attenuation of green fluorescent protein half-life inmammalian cells. Protein Engineering 12:1035-1040.

Curtis, C. F., and J. D. Lines. 2000. Should DDT be banned by international treaty? ParasitologyToday 16:119-121.

Daubaras, D. L., C. D. Hershberger, K. Kitano, and A. M. Chakrabarty. 1995. Sequence analyisof a gene cluster involved in metabolism of 2,4,5-trichlorophenoxyacetic acid by Burkholderiacepacia AC1100. Applied and Environmental Microbiology 61:1279-1289.

Daubaras, D. L., C. E. Danganan, A. Hübner, R. W. Ye, W. Hendrickson, and A. M. Chakrabarty.1996a. Biodegradation of 2,4,5-trichlorophenoxyacetic acid by Burkholderia cepacia AC1100:evolutionary insight. Gene 179:1-8.

Daubaras, D. L., K. Saido, and A. M. Chakrabarty. 1996b. Purification of hydroxyquinol 1,2-dioxygenase and maleylacetate reductase: the lower pathway of 2,4,5-trichlorophenoxyaceticacid metabolism by Burkholderia cepacia AC1100. Applied and Environmental Microbiology62:4276-4279.

de Lorenzo, V., M. Herrero, U. Jakubzik, and K. N. Timmis. 1990. Mini-Tn5 transposonderivatives for insertion mutagenesis, promotor probing, and chromosomal insertion of clonedDNA in Gram-negative eubacteria. Journal of Bacteriology 172:6568-6572.

Dua, M., A. Singh, N. Sethunathan, and A. K. Johri. 2002. Biotechnology and bioremediation:successes and limitations. Applied Microbiology and Biotechnology 59:143-152.

Dunbar, J., S. White, and L. Forney. 1997. Genetic diversity through the looking glass: effect ofenrichment bias. Applied and Environmental Microbiology 63:1326-1331.

Egli, T. 1995. The ecological and physiological significance of the growth of heterotrophicmicroorganisms with mixtures of substrates. Advances in Microbial Ecology 14:305-386.

Ellis, L. B. M. and L.P. Wackett (1997) The University of Minnesota Biocatalysis/BiodegradationDatabase. http://umbbd.ahc.umn.edu/

Eltis, L. D., and J. T. Bolin. 1996. Evolutionary relationships among extradiol dioxygenases.Journal of Bacteriology 178:5930-5937.

63

Ensley, B. D., and M. F. DeFlaun. 1995. Hazardous chemicals and biotechnology: past successesand future promise, p. 603-630. In L. Y. Young and C. E. Cerniglia (ed.), MicrobialTransformation and Degradation of Toxic Organic Chemicals.Wiley-Liss, New York, N.Y.

Erb, R. W., C. A. Eichner, I. Wagner-Döbler, and K. N. Timmis. 1997. Bioprotection of microbialcommunities from toxic phenol mixtures by a genetically designed pseudomonad. NatureBiotechnology 15:378-382.

Eriksson, M., G. Dalhammar, and A.-K. Borg-Karlsson. 2000. Biological degradation of selectedhydrocarbons in and old PAH/creosote contaminated soil from a gas work site. AppliedMicrobiology and Biotechnology 53:619-626.

Escher, B. I., M. Snozzi, and R. P. Schwarzenbach. 1996. Uptake, speciation, and uncouplingactivity of substituted phenols in energy transducing membranes. Environmental Science andTechnology 30:3071-3079.

Escher, B. I., R. W. Hunziker, and R. P. Schwarzenbach. 2001. Interaction of phenolic uncouplersin binary mixtures: concentration-additive and synergistic effects. Environmental Science andTechnology 35:3905-3914.

Eulberg, D., E. M. Kourbatova, L. A. Golovleva, and M. Schlömann. 1998. Evolutionaryrelationship between chlorocatechol catabolic enzymes from Rhodococcus opacus 1CP andtheir counterparts in proteobacteria: sequence divergence and functional convergence. Journalof Bacteriology 180:1082-1094.

Everaert, K., and J. Baeyens. 2002. The formation and emission of dioxins in large scale thermalprocesses. Chemosphere 46:439-448.

Farrell, A., and B. Quilty. 1999. Degradation of mono-chlorophenols by a mixed microbialcommunity via a meta-cleavage pathway. Biodegradation 10:353-362.

Farrell, A. and B. Quilty. 2002. Substrate-dependent autoaggregation of Pseudomonas putida CP1during the degradation of mono-chlorophenols and phenol. Journal of Industrial Microbiology& Biotechnology 28:316-324.

Fava, F., P. M. Armenante, and E. Kafkewitz. 1995. Aerobic degradation and dechlorination of 2-chlorophenol, 3-chlorophenol and 4-chlorophenol by a Pseudomonas pickettii strain. Letters inApplied Microbiology 21:307-312.

Fetzner, S., and F. Lingens. 1994. Bacterial Dehalogenases: Biochemistry, Genetics andBiotechnological Applications. Microbiological Reviews 58:641-685.

Finlayson-Pitts, B. J., and J. N. Pitts Jr. 1997. Tropospheric air pollution: ozone, airborne toxics,polycyclic aromatic hydrocarbons, and particles. Science 276:1045-1052.

Fothergill-Gilmore, L.A., and P. A. M. Michels. 1993. Evolution of glycolysis. Progress inBiophysics and Molecular Biology 59:105–235.

Furukawa, K. 2000. Engineering dioxygenases for efficient degradation of environmentalpollutants. Current Opinion in Biotechnology 11:244-249.

Gibello, A., E. Ferrer, M. Martin, and A. Garrido-Pertierra. 1994. 3,4-dihydroxyphenylacetate2,3-dioxygenase from Klebsiella pneumoniae, a Mg2+-containing dioxygenase involved inaromatic catabolism. Biochemical Journal 301:145-150.

Gibson, D. T., and R. E. Parales. 2000. Aromatic hydrocarbon dioxygenases in environmentalbiotechnology. Current Opinion in Biotechnology 11:236-243.

Gould, S. J. 1994. The evolution of life on earth. Scientific American October:85-91.Guha, S., C. A. Peters, and R. P. Jaffé. 1999. Multisubstrate biodegradation kinetics of

naphthalene, phenanthrene, and pyrene mixtures. Biotechnology and Bioengineering 65:491-499.

Gustafsson, K. 2000. Regulatory aspects, p. 153-162. In J. K. Jansson, J. D. van Elsas and M. J.Bailey (ed.), Tracking genetically-engineered microorganisms. Landes Bioscience,Georgetown, TX.

Habe, H., J.-S. Chung, J.-. Lee, K. Kasuga, T. Yoshida, H. Nojiri, and T. Omori. 2001.Degradation of chlorinated dibenzofurans and dibenzo-p-dioxins by two types of bacteria

64

having angular dioxygenases with different features. Applied and Environmental Microbiology67: 3610-3617.

Häggblom, M. M. and R. J. Valo. 1995. Bioremediation of chlorophenol wastes, p. 389-434. In L.Y. Young and C. E. Cerniglia (ed.), Microbial Transformation and Degradation of ToxicOrganic Chemicals.Wiley-Liss, New York, N.Y.

Häggblom, M. M. 1992. Microbial breakdown of halogenated aromatic pesticides and relatedcompounds. FEMS Microbiology Reviews 103:29-72.

Halden, R. U., B. G. Halden, and D. F. Dwyer. 1999. Removal of dibenzofuran, dibenzo-p-dioxin,and 2-chlorodibenzo-p-dioxin from soils inoculated with Sphingomonas sp. strain RW1.Applied and Environmental Microbiology 65:2246-2249.

Harayama, S., and K. N. Timmis. 1989. Catabolism of aromatic hydrocarbons by Pseudomonas,p. 151-174. In D. A. Hopwood and K. F. Chater (ed.) Genetics of bacterial diversity. AcademicPress, London.

Harayama, S. 1998. Artificial evolution by DNA shuffling. Trends in Biotechnology 16:76-82.Harayama, S., E. L. Neidle, and M. Kok. 1992. Functional and evolutionary relationships among

diverse oxygenases. Annual Review of Microbiology 46:565-601.Harder, W., and L. Dijkhuizen. 1982. Strategies of mixed substrate utilization in microorganisms.

Philosophical transactions of the Royal society of London, series B 297:459-480.Harms, H., and T. N. P. Bosma 1997. Mass transfer limitation of microbial growth and pollutant

degradation. Journal of Industrial Microbiology & Biotechnology 18:97-105.Harner, T., H. Kylin, T. F. Bidleman, C. Halsall, W. M. J. Strachan, L. A. Barrie, and P. Fellin.

1998. Polychlorinated naphthalenes and coplanar polychlorinated biphenyls in Arctic air.Environmental Science and Technology 32:3257-3265.

Haro, M.-A., and V. de Lorenzo. 2001. Metabolic engineering of bacteria for environmentalapplications: construction of Pseudomonas strains for biodegradation of 2-chlorotoluene.Journal of Biotechnology 85:103-113.

Hart, S. 1996. In situ bioremediation: defining the limits. Environmental Science and Technology30:398-401

Hatta, T., G. Mukerjee-Dhar, J. Damborsky, H. Kiyohara, and K. Kimbara. 2003.Characterization of a novel thermostable Mn(II)-dependent 2,3-dihydroxybiphenyl 1,2-dioxygenase from a PCB and naphthalene-degrading Bacillus sp JF8. Journal of BiologicalChemistry 278:21483-92.

Hatta, T., O. Nakano, N. Imai, N. Takizawa, and H. Kiyohara. 1999. Cloning and sequenceanalysis of hydroxyquinol 1,2-dioxygenase gene in 2,4,6-trichlorophenol-degrading Ralstoniapickettii DTP0602 and characterization of its product. Journal of Bioscience andBioengineering 87:267-272.

Head, I. M., and R. P. J. Swannell. 1999. Bioremediation of petroleum hydrocarbon contaminantsin marine habitats. Current Opinion in Biotechnology 10:234-239.

Heinaru, E., S. Viggor, E. Vedler, J. Truu, M. Merimaa, and A. Heinaru. 2001. Reversibleaccumulation of p-hydroxybenzoate and catechol determines the sequential decomposition ofphenolic compounds in mixed substrate cultivations in pseudomonads. FEMS MicrobiologyEcology 37:79-89.

Herrero, M., V. de Lorenzo, and K. N. Timmis 1990. Transposon vector containing non-antibioticresistance selection markers for cloning and stable chromosomal insertion of foreign genes inGram-negative bacteria. Journal of Bacteriology 172:6557-6567.

Hollender, J., W. Dott, and J. Hopp. 1994. Regulation of chloro- and methylphenol degradation inComamonas testosteroni JH5. Applied and Environmental Microbiology 60:2330-2338.

Horowitz, N. H. 1945. On the evolution of biochemical synthesis. Proceedings of the NationalAcademy of Sciences USA 31:153-157.

Hovander, L., T. Malmberg, M. Athanasiadou, I. Athanassiadis, S. Rahm, Å. Bergman, and E.Klasson Wehler. 2002. Identification of hydroxylated PCB metabolites and other phenolic

65

halogenated pollutants in human blood plasma. Archives of Environmental Contamination andToxicology 42:105-117.

Hristova, K. R., C. M. Lutenegger, and K. M. Scow. 2001. Detection and quantification of methyltert-butyl ether-degrading strain PM1 by real-time TaqMan PCR. Applied and EnvironmentalMicrobiology 67:5154-5160.

Huwe, J. K. 2002. Dioxins in food: a modern agricultural perspective. Jounal of Agricultural andFood Chemistry 50:1739-1750.

Jain, R. K., J. H. Dreisbach, and J. C. Spain. 1994. Biodegradation of p-nitrophenol via 1,2,4-benzenetriol by an Arthrobacter sp. Applied and Environmental Microbiology 60:3030-3032.

Jansson, J. K., and F. J. deBruijn. 1999. Biomarkers and bioreporters to track microbes andmonitor their gene expression, p. 651-665. In A. L. Demain and J. E. Davies (ed.), Manual ofindustrial microbiology and biotechnology. ASM Press, Washington, DC.

Jansson, J. K., and J. I. Prosser. 1997. Quantification of the presence and activity of specificmicroorganisms in nature. Molecular Biotechnology 7:103-120.

Jansson, J. K., and T. Leser. 1996. Quantitative PCR of environmental samples, p. 2.7.4. In A. D.L. Akkermans, J. D. van Elsas and F. J. deBruijn (ed.), Molecular Microbial EcolologyManual. Kluwer, Dordrecht, The Netherlands.

Jansson, J.K. 2002. Antibiotic, Chromogenic, and Luminescent Markers for Bacteria, p. 604-614.In C. J. Hurst, R. L. Crawford, G. R. Knudsen, M. J. McInerney, and L. D. Stetzenbach (ed.)Manual of Environmental Microbiology, 2nd Edition. ASM Press, Washington, D.C.

Jansson, J. K. 2003. Marker and reporter genes: illuminating tools for environmentalmicrobiologists. Current Opinion in Microbiology 6:310-316.

Jernberg, C. and J. K. Jansson. 2002. Impact of 4-chlorophenol contamination and/or inoculationwith the 4-chlorophenol-degrading strain, Arthrobacter chlorophenolicus A6L, on soilbacterial community structure. FEMS Microbiology Ecology 42:387-397.

Johnson, G. R., and R. H. Olsen. 1997. Multiple pathways for toluene degradation inBurkholderia cepacia sp. strain JS150. Applied and Environmental Microbiology 63:4047-4052.

Johnson, G. R., R. K. Jain, and J. C. Spain. 2002. Origins of the 2,4-dinitrotoluene pathway.Journal of Bacteriology 184:4219-4232.

Kadiyala, V., and J.C. Spain. 1998. A two-component monooxygenase catalyzes both thehydroxylation of p-nitrophenol and the oxidative release of nitrite from 4-nitrocatechol inBacillus sphaericus JS905. Applied and Environmental Microbiology 64:2479-2484.

Koonin, E. V., K. S. Makarova, and L. Aravind. 2001. Horizontal gene transfer in prokaryotes:quantification and classification. Annual Review of Microbiology 55:709-742.

Kovárová-Kovar. K., and T. Egli. 1998. Growth kinetics of suspended microbial cells: fromsingle-substrate-controlled growth to mixed-substrate kinetics. Microbiology and MolecularBiology Reviews 62:646-666

Kristensen, C. S., L. Eberl, J. M. Sanchez-Romero, M. Givskov, S. Molin, and V. de Lorenzo.1995. Site-specific deletions of chromosomally located DNA segments with the multimerresolution system of broad-host-range plasmid RP4. Journal of Bacteriology 177:52-58.

Kruse, H., and J. K. Jansson. 1997. The use of antibiotic resistance genes as marker genes ingenetically modified organisms. Norwegian Pollution Control Authority. Report 97:03.

Lampinen, J., L. Koivisto, M. Wahlsten, P. Mäntsälä, and M. Karp. 1992. Expression ofluciferase genes from different origins in Bacillus subtilis. Molecular Genetics 232:498-504.

Landmeyer, J. E., R. H. Chapelle, H. H. Herlong, and P. M. Bradley. 2001. Methyl tert-butylether biodegradation by indigenous aquifer microorganisms under natural and artificial oxicconditions. Environmental Science and Technology 35:1118-1126.

Latus, M., H.-J. Seitz, J. Eberspächer and F. Lingens. 1995. Purification and characterization ofhydroxyquinol 1,2-dioxygenase from Azotobacter sp. strain GP1. Applied and EnvironmentalMicrobiology 61:2453-2460.

66

Leahy, J. G., and R. R. Colwell. 1990. Microbial degradation of hydrocarbons in the environment.Microbiological Reviews 54:305-315.

Leung, K., D. Errampalli, M. Cassidy, H. Lee, J. T. Trevors, H. Okamura, H. J. Bach, and B. Hall.1997. A case study of bioremediation of polluted soil: Biodegradation and toxicity ofchlorophenols in soil. In J. D. van Elsas, J. T. Trevors, and E. M. H. Wellington (ed.), Modernsoil microbiology. Marcel Dekker, New York, N.Y.

Leung, K. T., J.-S. So, M. Kostrzynska, H. Lee, and J. T. Trevors. 2000. Using a greenfluorescent protein gene-labeled p-nitrophenol-degrading Moraxella strain to examine theprotective effect of alginate encapsulation against protozoan grazing. Journal ofMicrobiological Methods 39:205-211.

Liu, W.-T., T. L. Marsh, H. Cheng and L J. Forney. 1997. Characterization of microbial diversityby determining terminal restriction fragment length polymorphisms of genes encoding 16SrRNA. Applied and Environmental Microbiology 63:4516-4522.

Lohmeier-Vogel, E. M., K. T. Leung, H. Lee, J. T. Trevors, and H. J. Vogel. 2001. Phosphorus-31 nuclear magnetic resonance study of the effect of pentachlorophenol (PCP) on thephysiologies of PCP-degrading microorganisms. Applied and Environmental Microbiology67:3549-3556.

Louie, T. M., C. M. Webster, and L. Xun. 2002. Genetic and biochemical characterization of a2,4,6-trichlorophenol degradation pathway in Ralstonia eutropha JMP134.Journal of Bacteriology 184:3492-3500.

Lovanh, N., C. S. Hunt, and P. J. J. Alvarez. 2002. Effect of ethanol on BTEX biodegradationkinetics: aerobic continuous culture experiments. Water Research 36:3739-3746.

Lowder, M., A. Unge, N. Maraha, J. K. Jansson, J. Swiggett, and J. D. Oliver. 2000. Effect ofstarvation and the viable-but-nonculturable state on green fluorescent protein (GFP)fluorescence in GFP-tagged Pseudomonas fluorescens A506. Applied and EnvironmentalMicrobiology 66:3160-3165.

Lynge, E., A. Anttila, and K. Hemminki. 1997. Organic solvents and cancer. Cancer Causes andControl 8:406-419.

Madsen, T., and J. Aamand. 1992. Anaerobic transformation and toxicity of trichlorophenols in astable enrichment culture. Applied and Environmental Microbiology 58:557-561.

Mahajan, M. C., P. S. Phale, and C. S. Vaidyanathan. 1994. Evidence for the involvement ofmultiple pathways in the biodegradation of 1- and 2-methylnaphthalene by Pseudomonasputida CSV86.

Mars, A. E., T. Kasberg, S. R. Kaschabek, M. H. van Agteren, D. B. Janssen, and W. Reineke.1997. Microbial degradation of chloroaromatics: use of the meta-cleavage pathway formineralization of chlorobenzene. Journal of Bacteriology 179:4530-4537.

McLaughlin, S. G., and J. P. Dilger. 1980. Transport of protons across membranes by weak acids.Physiological Reviews 60:825-863.

Meerts I. A. T. M., R. J. Letcher, S. Hoving, G. Marsh, Å. Bergman, J. G. Lemmen, B. van derBurg and A. Brouwer. 2001. In vitro estrogenicity of polybrominated diphenyl ethers(PBDEs), hydroxylated PBDEs. Environmental Health Perspectives 109:399-407.

Meléndez-Hevia, E., T. G. Waddell, and M. Cascante. 1996. The puzzle of the Krebs citric acidcycle: assembling the pieces of chemically feasible reactions, and opportunism in the design ofmetabolic pathways during evolution. Journal of Molecular Evolution 43:293-303.

Menke, B., and H.-J. Rehm. 1992. Degradation of mixtures of monochlorophenols and phenol assubstrates for free and immobilized cells of Alcaligenes sp. A7-2. Applied Microbiology andBiotechnology 37:655-661.

Meregay, M. 1995. Heavy metal resistance in microbial ecosystems, p. 6.1.7:1-17. In A. D. L.Akkermans, J. D. van Elsas and F. J. deBruijn (ed.), Molecular Microbial Ecolology Manual.Kluwer, Dordrecht, The Netherlands.

67

Millette, D., J. F. Barker, Y. Comeau, B. J. Butler, E. O. Frind, B. Clément, and R. Samson. 1995.Substrate interaction during aerobic biodegradation of creosote-related compounds: a factorialbatch experiment. Environmental Science and Technology 29:1944-1952.

Mishra, V., R. Lal, and Srinivasan. 2001. Enzymes and operons mediating xenobiotic degradationin bacteria. Critical Reviews in Microbiology 27:133-166.

Mohn, W. W., and G. R. Stewart. 2000. Limiting factors for hydrocarbon degradation at lowtemperature in Arctic soils. Soil Biology and Biochemistry 32:1161-1172.

Mohn, W. W., K. Westerberg, W. R. Cullen, and K. J. Reimer. 1997. Aerobic biodegradation ofbiphenyl and polychlorinated biphenyls by Arctic soil microorganisms. Applied andEnvironmental Microbiology 63:3378-3384.

Moisheeva, O. V., I. P. Solyanikova, S. R. Kaschabek, J. Gröning, M. Thiel, L. A. Golovleva, andM. Schlömann. 2002. A new modified ortho cleavage pathway of 3-chlorocatecholdegradation by Rhodococcus opacus 1CP: genetic and biochemical evidence. Journal ofBacteriology 184:5282-5292.

Möller, A., A.-M. Norrby, K. Gustafsson, and J. K. Jansson. 1995. Luminometry and PCR-basedmonitoring of gene-tagged cyanobacteria in Baltic Sea microcosms. FEMS MicrobiologyLetters 129:43-50.

Möller, A., and J. K. Jansson. 1998. Detection of firefly luciferase-tagged bacteria inenvironmental samples, p. 269-283. In R. LaRossa (ed.), Methods in molecular biology vol102: Bioluminescence methods and protocols. Humana Press, Totowa, NJ.

Möller, A., and J. K. Jansson. 1997. Quantification of genetically-tagged cyanobacteria in BalticSea sediment by competitive PCR. Biotechniques 22:512-518.

Müller, C., L. Petruschka, H. Cuypers, G. Burchhardt, and H. Herrmann. 1996. Carbon cataboliterepression of phenol degradation in Pseudomonas putida is mediated by the inhibition of theactivatior protein PhlR. Journal of Bacteriology 178:2030-2036.

Munnecke, D. M. 1976. Enzymatic hydrolysis of organophosphate insecticides, a possiblepesticide disposal method. Applied and Environmental Microbiology 32:7-13.

Murakami, S., T. Okuno, E. Matsumura, S. Takenaka, R. Shinke, and K. Aoki. 1999. Cloning of agene encoding hydroxyquinol 1,2-dioxygenase that catalyzes both intradiol and extradiol ringcleavage of catechol. Bioscience, Biotechnology and Biochemistry 63:859-865.

Muttray, A. F., and W. W. Mohn. 2000. Quantitation of the population size and metabolic activityof a resin acid degrading bacterium in activated sludge using slot-blot hybridization to measurethe rRNA:rDNA ratio. Microbial Ecology 38:348-357.

Nakatsu, C. H., R. Korona, R. E. Lenski, F. J. deBruijn, T. L. Marsh, and L. J. Forney. 1998.Parallel and divergent genotypic evolution in experimental populations of Ralstonia sp. Journalof Bacteriology 180:4325-4331.

Nedwell, D. B. 1999. Effect of low temperature an microbial growth: lowered affinity forsubstrates limits growth at low temperatures. FEMS Microbiology Ecology 30:101-111.

Nelson L.M. 1982. Biologically-induced hydrolysis of parathion in soil: isolation of hydrolyzingbacteria. Soil Biology and Biochemistry 14:219-222.

Norén, K., and D. Meironyté. 2000. Certain organochlorine and organobromine contaminants inSwedish human milk in perspective of past 20-30 years. Chemosphere 40:1111-1123.

O’Brien, P. J., and D. Herschlag. 1999. Catalytic promiscuity and the evolution of new enzymaticactivities. Chemistry & Biology 6:R91-R105.

Ochman, H., J. G. Lawrence, and E. A. Groisman. 2000. Lateral gene transfer and the nature ofbacterial innovation. Nature 405:299-304.

Ohtsubo, Y., K. Miyauchi, K. Kanda, T. Hatta, H. Kiyohara, T. Senda, Y. Nagata, Y. Mitsui, andM. Takagi. 1999. PcpA, which is involved in the degradation of pentachlorophenol inSphingomonas chlorophenolica ATCC39723, is a novel type of ring-cleavage dioxygenase.FEBS Letters 459:395-398.

Ohtsubo, Y., M. Shimura, M. Delawary, K. Kimbara, M. Takagi, T. Kudo, A. Ohta, and Y.Nagata. 2003. Novel approach to the improvement of biphenyl and polychlorinated biphenyl

68

degradation activity: promotor implantation by homologous recombination. Applied andEnvironmental Microbiology 69:146-153.

Ormö, M., A. B. Cubitt, K. Kallio, L. A. Gross, R. Y. Tsien, and S. J. Remington. 1996. Crystalstructure of the Aequorea victoria Green Fluorescent Protein. Science 273:1392-1395.

Padilla, L., V. Matus, P. Zenteno and B. González. 2000. Degradation of 2,4,6-trichlorophenol viachlorohydroxyquinol in Ralstonia eutropha JMP134 and JMP222. Journal of BasicMicrobiology 4:243-249.

Park, H.-J., and E.-S. Kim. 2003. An inducible Streptomyces gene cluster involved in aromaticcompound metabolism. FEMS Microbiology Letters 226:151-157.

Parzefall, W. 2002. Risk assessment of dioxin contamination in human food. Food and ChemicalToxicology 40:1185-1189.

Pieper, D., and W. Reineke. 2000. Engineering bacteria for bioremediation. Current Opinion inBiotechnology 11:262-270.

Prieto, M. A., and J. L. Garcia. 1994. Molecular characterization of 4-hydroxyphenylacetate 3-hydroxylase of Escherishia coli. Journal of Biological Chemistry 269-22823-22829.

Pritchard, P. H., E. J. O’Neill, C. M. Spain, and D. J. Ahearn. 1987. Physical and biologicalparameters that determine the fate of p-chlorophenol in laboratory test systems. AppliedEnvironmental Microbiology 53:1833-1838.

Que L. Jr, and R. Y. N. Ho. 1996. Dioxygen activation by enzymes with mononuclear non-hemeiron active sites. Chemical Reviews 96:2607-2624.

Que L. Jr, Widom J., and Crawford R. L. 1981. 3,4-Dihydroxyphenylacetate 2,3-dioxygenase. Amanganese(II) dioxygenase from Bacillus brevis. Journal of Biological Chemistry 256:10941-10944.

Rahman, F., K. H. Langford, M. D. Scrimshaw, and J. N. Lester. 2001. Polybrominated diphenylether (PBDE) flame retardants. The Science of the Total Environment 275:1-17.

Raillard, S-A., A. Krebber, Y. Chen, J.E. Ness, E. Bermudez, R. Trinidad, R. Fullem, C. Davis,M. Welch, J. Seffernick, L.P. Wackett, W.P.C. Stemmer, and J. Minshull. 2001. Novel enzymeactivities and functional plasticity revealed by recombining homologous enzymes. Chemistryand Biology 125:1-9.

Reardon, K. F., D. C. Mosteller, and J. D. Bull Rogers. 2000. Biodegradation kinetics of benzene,toluene, and phenol as single and mixed substrates for Pseudomonas putida F1. Biotechnologyand Bioengineering 69:385-400.

Reddy, G. V. B., and Gold, M. H. 2000. Degradation of pentachlorophenol by Phanerochaetechrysosporium: intermediates and reactions involved. Microbiology 146:405-413.

Regulation on certain active substances (pesticides). 1993. KIFS 1993:5, Swedish NationalChemicals Inspectorate.

Reineke, W. 1998. Development of hybrid strains for the mineralization of chloroaromatics bypatchwork assembly. Annual Review of Microbiology 52:287-331.

Romantschuk, M., I. Sarand, T. Petänen, R. Peltola, M. Jonsson-Vihanne, T. Koivula, K. Yrjälä,and K. Haahtela. 2000. Means to improve the effect of in situ bioremediation of contaminatedsoil: an overview of novel approaches. Environmental Pollution 107:179-185.

Rosenberg, E., and Ron, E. Z. 1999. High- and low-molecular mass microbial surfactants.Applied Microbiology and Biotechnology 52:154-162.

Samanta, S. K., O. V. Singh, and R. K. Jain. 2002. Polycyclic aromatic hydrocarbons:environmental pollution and bioremediation. Trends in Biotechnology 20:243-248.

Sanchez-Romero, J. M., R. Diaz-Orejas, and V. de Lorenzo. 1998. Resistance to tellurite as aselection marker for genetic manipulations of Pseudomonas strains. Applied andEnvironmental Microbiology 64:4040-4046.

Saunders, J. R., R. W. Pickup, J. A. Morgan, C. Winstanley, and V. A. Saunders. 1995. xylE as amarker gene for microorganisms, p. 6.1.3:1-12. In A. D. L. Akkermans, J. D. van Elsas and F.J. deBruijn (ed.), Molecular Microbial Ecolology Manual. Kluwer, Dordrecht, TheNetherlands.

69

Sayler, G. S., and Ripp, S. 2000. Field applications of genetically engineered microorganisms forbioremediation processes. Current Opinion in Biotechnology 11:286-289.

Schlömann, M. 1994. Evolution of chlorocatechol catabolic pathways. Conclusions to be drawnfrom comparisons of lactone hydrolases. Biodegradation 5:301-21.

Seffernick, J. L, M. L. DeSouza, M. J. Sadowsky, and L. P. Wackett. 2001. Melamine deaminaseand atrazine chlorohydrolase: 98 identical but functionally different. Journal of Bacteriology183:2405-2410.

Selifonova, O., F. Valle, and V. Schellenberger. 2001. Rapid evolution of novel traits inmicroorganisms. Applied and Environmental Microbiology 67:3645-3649.

Shapiro, J. A. 1999. Views about evolution are evolving. ASM News 65:201-207.Shaw, L. J., Y. Beaton, L. A. Glover, K. Kilham, and A. A. Meharg. 1999. Development and

characterization of a lux-modified 2,4-dichlorophenol-degrading Burkholderia sp. RASC.Environmental Microbiology 1:393-399.

Silcock, D. J., R. N. Waterhouse, L. A. Glover, J. I. Prosser, and K. Kilham. 1992. Detection of asingle genetically modified bacterial cell in soil by using charge coupled device-enhancedmicroscopy. Applied and Environmental Microbiology 58:2444-2448.

Smets, B. F., S. D. Siciliano, and W. Verstraete. 2002. Natural attenuation: extant microbialactivity forever and ever? Environmental Microbiology 4:315-317.

Spain, J. C. 1995. Biodegradation of nitroaromatic compounds. Annual Reviews of Microbiology49:523-555.

Stackebrandt, E. and B. M. Goebel. 1994. Taxonomic note: a place for DNA-DNA reassociationand 16S rRNA sequence analysis in the present species definition in bacteriology. InternationalJournal of Systematic Bacteriology 44:846-849.

Stapleton R. D., D. C. Savage, G. S. Sayler, and G. Stacey. 1998. Biodegradation of aromatichydrocarbons in an extremely acidic environment. Applied and Environmental Microbiology64:4180-4184.

Stelmack, P. L., M. R. Gray, and M. A. Pickard. 1999. Bacterial adhesion to soil contaminants inthe presence of surfactants. Applied and Environmental Microbiology 65:163-168.

Stemmer, W. P. C. 1994. DNA shuffling by random fragmentation and reassembly: In vitrorecombination for molecular evolution. Proceedings of the National Academy of Sciences,USA. 91:10747-10751.

Stringfellow, W. T., and M. D. Aitken. 1995. Competitive metabolism of naphthalene,methylnaphthalenes, and fluorene by phenanthrene-degrading pseudomonads. Applied andEnvironmental Microbiology 61:357-362.

Strong, L.C., H. McTavish, M. J. Sadowsky, and L. P. Wackett 2000. Field-scale remediation ofatrazine-contaminated soil using recombinant Escherichia coli over-expressing atrazinechlorohydrolase. Environmental Microbiology 2:91-98.

Swoboda-Colberg, N. G. 1995. Chemical contamination of the environment: Sources, types, andfate of synthetic organic compounds, p. 27-74. In L. Y. Young and C. E. Cerniglia (ed.),Microbial Transformation and Degradation of Toxic Organic Chemicals. Wiley-Liss, NewYork, N.Y.

Takenaka, S., S. Okugawa, M. Kadowaki, S. Murakami, and K. Aoki. 2003. The metabolicpathway of 4-aminophenol in Burkholderia sp. strain AK-5 differs from that of aniline andaniline with C-4 substituents. Applied and Environmental Microbiology 69:5410-5413.

Takizawa, N., H. Yokoyama, K. Yanagihara, T. Hatta and H. Kiyohara. 1995. A locus ofPseudomonas pickettii DTP0602, had, that encodes 2,4,6-trichlorophenol-4-dechlorinase withhydroxylase activity, and hydroxylation of various chlorophenols by the enzyme. Journal ofFermentation and Bioengineering 80:318-326.

Tan, H.-M. 1999. Bacterial catabolic plasmids. Applied Microbiology and Biotechnology 51:1-12.

70

Tani, K., M. Muneta, K. Nakamura, K. Shibuya, and M. Nasu. 2002. Monitoring of Ralstoniaeutropha KT1 in groundwater in an experimental bioaugmentation field by in situ PCR.Applied and Environmental Microbiology 68:412-416.

Tas, É., and Lindström, K. 2000. Detection of bacteria by their intrinsic markers, p. 53-68. In J.K. Jansson, J. D. van Elsas and M. J. Bailey (ed.), Tracking genetically-engineeredmicroorganisms. Landes Bioscience, Georgetown, TX.

Terada, H. 1990. Uncouplers of oxidative phosphorylation. Environmental Health Perspectives87:213-218

Timmis, K. N., and D. H. Pieper. 1999. Bacteria designed for bioremediation. Trends inBiotechnology 17:201-204.

Tombolini, R., and J. K. Jansson. 1998. Monitoring of GFP tagged bacterial cells. Methods inMolecular Biology 102:285-298.

Tombolini, R., A. Unge, M. E. Davey, F. J. deBruijn, and J. K. Jansson. 1997. Flow cytometricand microscopic analysis of GFP-tagged Pseudomonas fluorescens bacteria. FEMSMicrobiology Ecology 22:17-28.

Tombolini, R., D. J. van der Gaag, B. Gerhardson, and J. K. Jansson. 1999. Colonization patternof the biocontrol strain Pseudomonas chlororaphis MA 342 on barley seeds visualized byusing green fluorescent protein. Applied and Environmental Microbiology 65:3674-3680.

Toppari, J., J. C. Larsen, P. Christiansen, A. Giwercman, P. Grandjean, L. J. Guillette, B. Jégou,T. K. Jensen, P. Jouanner, N. Keiding, H. Leffers, J. A. McLachlan, O. Meyer, J. Müller, E.Rajpert de-Meyts, T. Scheike, R. Sharpe, J. Sumpter, and N. E. Skakkebæk. 1996. Malereproductive health and environmental xenoestrogens. Environmental Health Perspectives1041(suppl 4):741-803.

Torsvik, V., J. Goksøyr, and F. L. Daae. 1990. High diversity in DNA of soil bacteria. Appliedand Environmental Microbiology 56:782-787.

Travkin, V. M., A. P. Jadan, F. Briganti, A. Scozzafava, and L. A. Golovleva. 1997.Characterization of an intradiol dioxygenase involved in the biodegradation of thechlorophenoxy herbicides 2,4-D and 2,4,5-T. FEBS Letters 407:69-72.

Unge, A., and J. K. Jansson. 2001. Monitoring population size, activity, and distribution of gfp-luxAB-tagged Pseudomonas fluorescens SBW25 during colonization of wheat. MicrobialEcology 41:290-300.

Unge, A., R. Tombolini, L. Mølbak, and J. K. Jansson. 1999. Simultaneous monitoring of cellnumber and metabolic activity of specific bacterial populations with a dual gfp-luxAB markersystem. Applied and Environmental Microbiology 65:813-821.

Uotila, J. S., V. H. Kitunen, T. Coote, T. Saastamoinen, M. Salkinoja-Salonen and J. H. A.Apajalahti. 1995. Metabolism of halohydroquinones in Rhodococcus chlorophenolicus PCP-1.Biodegradation 6:119-126.

van den Berg K. J. 1990. Interaction of chlorinated phenols with thyroxine binding sites of humantransthyretin, albumin and thyroid binding globulin. Chemico-Biological Interactions 76:63-75.

Vanderberg, L. A., R. Krieger-Grumbine, and M. N. Taylor. 2000. Evidence for diverseoxidations in the catabolism of toluene by Rhodococcus rhodocrous strain OFS. AppliedMicrobiology and Biotechnology 53:447-452.

van der Meer, J. R. 1997. Evolution of novel metabolic pathways for the degradation ofchloroaromatic compounds. Antonie Van Leeuwenhoek 71:159-78.

van der Meer, J. R., W. M. de Vos, S. Harayama, and A. J. B. Zehnder. 1992. Molecularmechanisms of genetic adaptation to xenobiotic compounds. Microbiological Reviews 56:677-694.

van Elsas, J. D., K. Smalla, and C. C. Tebbe. 2000. Extraction and analysis of microbialcommunity nucleic acids from environmental matrices, p. 29-52. In J. K. Jansson, J. D. vanElsas and M. J. Bailey (ed.), Tracking genetically-engineered microorganisms. LandesBioscience, Georgetown, TX.

71

van Veen, J. A., L. S. van Overbeek, and J. D. van Elsas. 1997. Fate and activity ofmicroorganisms introduced into soil. Microbiology and Molecular Biology Reviews 61:121-35.

Vinci, V. A., and G. Byng. 1999. Strain improvement by nonrecombinant methods, p. 103-113. InA. C. Demain and J. E. Davies (ed.), Manual of Industrial Microbiology and Biotechnology,2nd ed. ASM Press, Washington, D. C.

Vogel, T. M. 1996. Bioaugmentation as a soil remediation approach. Current Opinion inBiotechnology 7:311-316.

Watanabe, K., M. Teramoto, and S. Harayama. 2002. Stable augmentation of activated sludgewith foreign catabolic genes harboured by an indigenous dominant bacterium. EnvironmentalMicrobiology 4:577-583.

Wattiau, P., D. Springael, S. N. Agathos, and S. Wuertz. 2002. Use of the pAL5000 replicon inPAH-degrading mycobacteria: application for strain labelling and promotor probing. AppliedMicrobiology and Biotechnology 59:700-705.

Wayne, L. G., D. J. Brenner, R. R. Colwell, P. A. D. Grimont, O. Kandler, M. I. Krichevsky, L.H. Moore, W. E. C. Moore, R. G. E. Murray, E. Stackebrandt, M. P. Starr, and H. G. Trüper.1987. Report of the ad hoc committee on reconciliation of approaches to bacterial systematics.International Journal of Systematic Bacteriology 37:463-464.

Werwath, J., H.-A. Arfmann, D. H. Pieper, K. N. Timmis, and R.-M. Wittich. 1998. Biochemicaland genetic characterization of a gentisate 1,2-dioxygenase from Sphingomonas sp. strainRW5. Journal of Bacteriology 180:4171-4176.

Whited, G. M., and D. T. Gibson. 1991. Separation and partial characterization of the enzymes ofthe toluene-4-monooxygenase catabolic pathway in Pseudomonas mendocina KR1. Journal ofBacteriology 173:3017-3020.

Widada, J., H. Nojiri, and T. Omori. 2002. Recent developments in molecular techniques foridentification and monitoring of xenobiotic-degrading bacteria and their catabolic genes inbioremediation. Applied and Environmental Biotechnology 60:45-59.

Wise, A. A., and C. R. Kuske. 2000. Generation of novel bacterial regulatory proteins that detectpriority pollutant phenols. Applied and Environmental Microbiology 66:163-9.

Woods, S. L., J. F. Ferguson, and M. M. Benjamin. 1989. Characterization of chlorophenol andchloromethoxybenzene biodegradation during anaerobic treatment. Environmental Science andTechnology 23:62-68.

Xu, L., K. Resing, S. L. Lawson, P. C. Babbitt, and S. D. Copely. 1999. Evidence that pcpAencodes 2,6-dichlorohydroquinone dioxygenase, the ring cleavage enzyme required forpentachlorophenol degradation in Sphingomonas chlorophenolica strain ATCC 39723.Biochemistry 38:7659-7669.

Xun, L., E. Topp, and C. S. Orser. 1992. Purification and characterization of a tetrachloro-p-hydroquinone reductive dehalogenase from a Flavobacterium sp. Journal of Bacteriology174:8003-8007.

Xun, L., J. Bohuslavek, and M. Cai. 1999. Characterization of 2,6-dichloro-p-hydroquinone 1,2-dioxygenase (PcpA) of Sphingomonas chlorophenolica ATCC 39723. Biochemical andBiophysichal Research Communications 266:322-325.

Yang, F., L. G. Moss, and G. N. Phillips Jr. 1996. The molecular structure of green fluorescentprotein. Nature Biotechnology 14:1246-1251.

Yang, G., P.-H. Liang, and D. Dunaway-Mariano. 1994. Evidence for nucleophilic catalysis in thearomatic substitution reaction catalyzed by (4-chlorobenzoyl)coenzyme A dehalogenase.Biochemistry 33:8527-8531.

Yin, X., and G. Stotzky. 1997. Gene transfer among bacteria in natural environments. Advancesin Applied Microbiology 45:153-212.

Zaborina, O., D. L. Daubaras, A. Zago, L. Xun, K. Saido, T. Klem, D. Nikolic, and A. M.Chakrabarty. 1998. Novel pathway for conversion of chlorohydroxyquinol to maleylacetate inBurkholderia cepacia AC1100. Journal of Bacteriology 180:4667-4675.

72

Zaborina, O., M. Latus, J. Eberspächer, L. A. Golovleva, and F. Lingens. 1995. Purification andcharacterization of 6-chlorohydroxyquinol 1,2-dioxygenase from Streptomyces rochei 303:comparison with an analogous enzyme from Azotobacter sp. strain GP1. Journal ofBacteriology 177:229-234.

Zechendorf, B. 1999. Sustainable development: how can biotechnology contribute? Trends inBiotechnology 17:219-225.

Zhou, E., and R. L. Crawford. 1995. Effects of oxygen, nitrogen, and temperature on gasolinebiodegradation in soil. Biodegradation 6:127-140.

Zimmer, M. 2002. Green fluorescent protein (GFP): applications, structure, and relatedphotophysical behavior. Chemical Reviews 102:759-781.