Distribution and Dynamics of Pyrene-Degrading Mycobacteria in Freshwater Sediments Contaminated with Polycyclic Aromatic Hydrocarbons

University of Tennessee, Knoxville TRACE: Tennessee Research and Creative Exchange

Doctoral Dissertations Graduate School

8-2008

Distribution and dynamics of pyrene-degrading Mycobacteria in freshwater sediments contaminated with polycyclic aromatic hydrocarbons

Jennifer M. DeBruyn University of Tennessee - Knoxville

Follow this and additional works at: https://trace.tennessee.edu/utk_graddiss

Part of the Ecology and Evolutionary Biology Commons

Recommended Citation DeBruyn, Jennifer M., "Distribution and dynamics of pyrene-degrading Mycobacteria in freshwater sediments contaminated with polycyclic aromatic hydrocarbons. " PhD diss., University of Tennessee, 2008. https://trace.tennessee.edu/utk_graddiss/424

This Dissertation is brought to you for free and open access by the Graduate School at TRACE: Tennessee Research and Creative Exchange. It has been accepted for inclusion in Doctoral Dissertations by an authorized administrator of TRACE: Tennessee Research and Creative Exchange. For more information, please contact [email protected]. To the Graduate Council:

I am submitting herewith a dissertation written by Jennifer M. DeBruyn entitled "Distribution and dynamics of pyrene-degrading Mycobacteria in freshwater sediments contaminated with polycyclic aromatic hydrocarbons." I have examined the final electronic copy of this dissertation for form and content and recommend that it be accepted in partial fulfillment of the requirements for the degree of Doctor of Philosophy, with a major in Ecology and Evolutionary Biology.

Gary S. Sayler, Major Professor

We have read this dissertation and recommend its acceptance:

Gary McCraken, Jennifer Schweitzer, Mark Radosevich

Accepted for the Council:

Carolyn R. Hodges

Vice Provost and Dean of the Graduate School

(Original signatures are on file with official studentecor r ds.) To the Graduate Council: I am submitting herewith a dissertation written by Jennifer M. DeBruyn entitled “Distribution and dynamics of pyrene-degrading Mycobacteria in freshwater sediments contaminated with polycyclic aromatic hydrocarbons.” I have examined the final electronic copy of this dissertation for form and content and recommend that it be accepted in partial fulfillment of the requirements for the degree of Doctor of Philosophy, with a major in Ecology & Evolutionary Biology.

Gary S. Sayler Gary S. Sayler, Major Professor We have read this dissertation and recommend its acceptance:

Gary McCraken Gary McCracken

Jennifer Schweitzer Jennifer Schweitzer

Mark Radosevich Mark Radosevich

Acceptance for the Council:

Carolyn R. Hodges Carolyn R. Hodges, Vice Provost & Dean of the Graduate School

(Original signatures are on file with official student records.)

DISTRIBUTION AND DYNAMICS OF PYRENE-DEGRADING

MYCOBACTERIA IN FRESHWATER SEDIMENTS CONTAMINATED WITH

POLYCYCLIC AROMATIC HYDROCARBONS

A Dissertation Presented for the Doctor of Philosophy Degree The University of Tennessee, Knoxville

JENNIFER M. DEBRUYN

AUGUST 2008

ACKNOWLEDGEMENTS

I would like to thank my advisor, Dr. Gary Sayler, as well as my committee members Dr. Mark Radosevich, Dr. Gary McCracken, and Dr. Jennifer Schweitzer for their guidance throughout this project. I also want to thank Dr. L. McKay, Dr. V. Vulava, and Dr. F. Menn for their invaluable advice and assistance.

This work would not have been possible without those that have directly assisted me in field sampling: D. Castle, C. Chewning, J. Rainey, J. Easter, M. Scholz, D. Williams, and C. Lalande all braved hot weather and voracious insects to retrieve malodorous mud samples from Chattanooga Creek. My gratitude goes to Dr. Johanna Rinta-Kanto who shared her sediment samples from Lake Erie. I am especially indebted to Chris Chewning and Tommy Mead for their patience and hard work with me in the laboratory.

Thank you to my coworkers at the Center for Environmental Biotechnology and the University of Tennessee with whom I’ve had the honour to work. You provided advice, support, and friendship, and ensured that my time here was never boring.

Last but not least, thank you to my family: To my father, Ed, for instilling an insatiable curiosity about the world around me. To my mother, Mary, and sisters Emily and Rebecca, for their love, friendship and support. And finally, to Steven, for his wisdom, inspiration, and unwavering belief in me.

ABSTRACT

Microbial biodegradation of polycyclic aromatic hydrocarbons (PAHs) is the primary means of attenuation of these toxic and carcinogenic compounds from contaminated soils and sediments. The documented toxicity and carcinogenicity of many PAHs demands remedial action for PAH-contaminated soils and sediments. This is especially important for historically contaminated sites, where higher molecular weight PAHs (HMW) are recalcitrant. Recently, fast-growing Mycobacteria have been identified that can degrade HMW PAHs, such as pyrene and benzo[a]pyrene. These bacteria have been isolated from a variety of geographical locations, indicating a cosmopolitan distribution.

This dissertation work was driven by the need for a better understanding of the ecology, distribution, and dynamics of indigenous microbial populations that can biodegrade PAHs, with an emphasis on Mycobacteria. Both culture-independent molecular approaches and cultivation were used to 1) determine the presence of pyrene-degrading Mycobacterium genotypes and compare dynamics to proteobacterial naphthalene-degradation genotypes; 2) determine the distribution of these genotypes across environments, including different geographical locations (Chattanooga Creek and Lake Erie) and different physical environments (sediments vs. suspended particles); and 3) provide a link between functional pyrene genotypes and phylogenetic identity of isolated pyrene-degrading organisms.

The results of these studies indicate the pyrene-degrading Mycobacteria have broad, cosmopolitan distribution in contaminated sediments and suspended particles. Isolation of pyrene-degrading organisms from both Chattanooga Creek and Lake Erie has provided strong evidence for horizontal transfer of pyrene- degrading genes between diverse genera. This work demonstrates the prevalence of pyrene-degrading organisms in contaminated sediments and implicates an integral role in natural attenuation of HMW PAHs. iii

TABLE OF CONTENTS

PART I...... 1

LITERATURE REVIEW ...... 1 1. Biodegradation of Polycyclic Aromatic Hydrocarbons ...... 2 1.1 Polycyclic aromatic hydrocarbons...... 2 1.2 Microbial degradation of PAHs ...... 5 1.3 Bioremediation and natural attenuation ...... 8 1.4 Ecology and biogeography of PAH-degrading organisms ...... 11 2. Diversity of PAH Dioxygenases: Applications in Ecology of Biodegrative Microorganisms ...... 13 2.1 Ring-hydroxylating dioxygenase structure and specificity ...... 15 2.2 Genetic organization of PAH-catabolism genes ...... 17 2.3 Diversity of PAH dioxygenases ...... 18 2.4 Ecological insight from dioxygenase-based studies ...... 22 3. PAH-Degrading Mycobacteria ...... 25 3.1 Identification of pyrene-degrading organisms – Mycobacterium ...... 25 3.2 Distribution and ecology ...... 27 3.3 Pyrene dioxygenase genes: nidA and nidA3 ...... 28 4. Methods in Molecular Microbial Ecology ...... 31 4.1 Extracting DNA from environmental samples ...... 31 4.2 Real-time PCR ...... 32 4.3 16S rDNA Methods ...... 35 4.4 PCR bias ...... 36 4.5 Community fingerprinting: T-RFLPs ...... 38 4.6 Statistical analysis of community fingerprinting data ...... 43 5. General Objectives and Hypotheses ...... 49 6. Study Areas...... 52 6.1 Chattanooga Creek ...... 52 6.2 Lake Erie...... 53

PART II...... 56

COMPARATIVE QUANTITATIVE PREVALENCE OF MYCOBACTERIA AND

FUNCTIONALLY ABUNDANT NIDA, NAHAC AND NAGAC DIOXYGENASE GENES IN

COAL TAR CONTAMINATED SEDIMENTS ...... 56 1. Abstract ...... 57 2. Introduction ...... 58 3. Methods ...... 61 3.1 Sampling ...... 61 3.2 PAH quantification ...... 61 3.3 Mineralization Assays ...... 63 3.4 DNA Extraction and Real-time PCR ...... 63 3.5 Mycobacterium 16S clone library ...... 65 4. Results ...... 66 4.1 Optimization and specificity of the nidA real time PCR assay ...... 66 4.2 Quantification of nidA, nagAc and nahAc in sediments of Chattanooga Creek ...... 66 4.3 PAH concentrations ...... 69 4.4 PAH mineralization ...... 72 4.5 Correlations among catabolic genotypes and mineralization potential ...... 72 4.6 Diversity of Mycobacteria ...... 75 5. Discussion...... 81

PART III...... 86

MICROBIAL COMMUNITY STRUCTURE AND BIODEGRADATION ACTIVITY

ASSOCIATED WITH PARTICLE-ASSOCIATED BACTERIA IN A COAL TAR

CONTAMINATED CREEK ...... 86 1. Abstract ...... 87 2. Introduction ...... 88 3. Materials and Methods ...... 91 3.1 Sample collection ...... 91 3.2 SEM ...... 91 3.3 Mineralization assays ...... 91 3.4 DNA extraction and real-time PCR ...... 93 3.5 T-RFLPs ...... 95 3.6 T-RFLP data analysis and statistics ...... 96 4. Results ...... 99 4.1 Particle characterization and PAH mineralization ...... 99

4.2 Specificity of the Myco16S real time PCR assay ...... 99 4.3 Genotypic characterization of PAH degrading populations in PAB communities ...... 102 4.4 Phylogenetic comparisons between sediments and particle-associated communities (T-RFLP) ...... 104 5. Discussion...... 113

PART IV...... 118

ISOLATION OF PYRENE-DEGRADING ORGANISMS FROM GEOGRAPHICALLY

DISPARATE PAH-CONTAMINATED SEDIMENTS: EVIDENCE FOR HORIZONTAL

TRANSFER OF NIDA AND NIDA3 PYRENE DIOXYGENASE GENES BETWEEN

ACTINOMYCETES ...... 118 1. Abstract ...... 119 2. Introduction ...... 120 3. Materials and Methods ...... 123 3.1 Sampling and sediment enrichments ...... 123 3.2 Staining and imaging ...... 124 3.3 Molecular characterization ...... 124 3.4 PFGE ...... 126 4. Results ...... 127 4.1 Isolation and identification of pyrene-degrading organisms ...... 127 4.2 Catabolic genes nidA and nidA3 ...... 138 4.3 Genomic architecture of five pyrene-degrading Mycobacterium spp. 138 5. Discussion...... 147

PART V...... 152

PAH BIODEGRADATIVE GENOTYPES IN LAKE ERIE SEDIMENTS: EVIDENCE FOR

BROAD GEOGRAPHICAL DISTRIBUTION OF PYRENE-DEGRADING MYCOBACTERIA ...... 152 1. Abstract ...... 153 2. Introduction ...... 154 3. Methods ...... 157 3.1 Sampling and DNA extraction ...... 157 3.2 PAH quantification ...... 157 3.3 16S clone library ...... 158 3.4 Real-time PCR ...... 159

4. Results ...... 160 4.1 CTD data and PAH concentrations ...... 160 4.2 Station 84 clone library...... 160 4.3 Quantification of PAH catabolic genes ...... 170 5. Discussion...... 178

PART VI...... 182

SUMMARY AND CONCLUSIONS ...... 182 1. Microbial ecology of PAH-degrading organisms ...... 183 2. Naphthalene biodegradative genotypes ...... 184 3. Evolution and biogeography of pyrene degrading organisms ...... 186

VITA ...... 216

vii

LIST OF TABLES

Table I-1. Structures and solubilities of 16 priority polycyclic aromatic hydrocarbons (PAHs) identified by the US EPA (Keith & Telliard 1979)...... 3 Table I-2. Pyrene degrading Mycobacterium previously isolated...... 26 Table II-1. Concentrations of 16 priority PAHs measured in Chattanooga Creek surface sediments...... 71

Table II-2. Spearman rank correlation coefficients (rs) matrix...... 73 Table III-1. Percent of added 14C-naphthalene or 14C-pyrene mineralized and first-order biodegradation rate constants (k) by bacterial communities associated with > 5 μm particles...... 101 Table III-2. Results of ANOSIM analysis...... 112 Table IV-1. Primers used in this study...... 125 Table IV-2. Characteristics of pyrene-degrading isolates...... 136 Table IV-3. Genomic characteristics of five pyrene-degrading Mycobacterium spp...... 141 Table V-1. Concentrations of PAHs in Lake Erie sediments...... 162

viii

LIST OF FIGURES

Figure I-1. Intra- vs. extradiol cleavage...... 14 Figure I-2. Dihydroxylation of naphthalene catalyzed by naphthalene 1,2- dioxygenase...... 14 Figure I-3. Phylogenetic relationship of α subunit of aromatic ring- hydroxylating dioxygenases, showing taxonomic affiliations of the organisms...... 19 Figure I-4. Phylogenetic relationships of the α subunit of aromatic ring- hydroxylating dioxygenases, grouped by primary substrate type...... 20 Figure I-5. Pyrene degradation pathway in Mycobacterium vanbaalenii (Kim et al. 2007)...... 29 Figure I-6. Real-time PCR using a dual-labeled probe...... 34 Figure I-7. An example of real time PCR data output (left) and standard curve (right)...... 35 Figure I-8. Schematic illustrating the terminal restriction fragment length polymorphism (T-RFLP) method...... 40 Figure II-1. Map of sampling locations on Chattanooga Creek...... 62 Figure II-2. Threshold cycles (CT) over a range of template concentrtations for both cloned nidA (◊) and environmental samples (■)...... 67 Figure II-3. Quantities of 16S (black), nidA (white), nagAc (hatched) and nahAc (grey) at the three sites for each sampling date...... 68 Figure II-4. Quantities of nidA (left), nagAc (center) and nahAc (right) measured at each sampling event (● October 2004, ○ April 2005, ▼ July 2005 and ∆ October 2005)...... 70 Figure II-5. PAH mineralization (40 hour incubation) in relation to measured concentrations of naphthalene (A) and pyrene (B)...... 74

and nagAc (rs = 0.755)...... 76 Figure II-7. Functional relationship between PAH-dioxygenase gene abundances (● nidA, ○ nagAc, ▼nahAc) and in vitro pyrene mineralized...... 77 Figure II-8. Neighbor-joining tree of 16S rDNA sequences from Chattanooga Creek retrieved using Mycobacterium specific primers. .. 78 Figure III-1. Map of Sampling locations on Chattanooga Creek...... 92 Figure III-2. Environmental ESM images of particles trapped on 5 μM polycarbonate filters...... 100 Figure III-3. Gene abundances in particle-associated communities...... 103 Figure III-4. Gene abundances in particle associated communities compared to abundances in sediment communities...... 105 Figure III-5. Relationship between Mycobacterium biomass and nidA copies in sediments (black) and particle-associated communities (white)...... 106 Figure III-6. Positive correlation between nidA and nagAc abundance in sediments (black) was not observed in the PAB communities (white). 106 Figure III-7. CCA ordination plots of T-RFLP profile data...... 108 Figure III-8. NMDS plots of T-RFLP profiles...... 110 Figure III-9. NMDS of T-RFLP profiles (HhaI digested) of particle- associated communities (top) and sediment communities (bottom). .... 111 Figure IV-1. SEM images of pyrene-degrading organism PY143, isolated from Lake Erie...... 128 Figure IV-2. SEM images of pyrene-degrading organism PY146, isolated from Lake Erie...... 129 Figure IV-3. SEM images of pyrene-degrading organism PY138, isolated from Lake Erie...... 130

Figure IV-4. SEM images of pyrene-degrading organism PY115, isolated from Chattanooga Creek...... 131 Figure IV-5. SEM images of pyrene-degrading organism PY116, isolated from Chattanooga Creek...... 132 Figure IV-6. SEM images of pyrene-degrading organism PY120, isolated from Chattanooga Creek...... 133 Figure IV-7. SEM images of pyrene-degrading organism PY129, isolated from Chattanooga Creek...... 134 Figure IV-8. SEM images of pyrene-degrading organism PY132, isolated from Chattanooga Creek...... 135 Figure IV-9. Phylogeny of pyrene-degrading isolates...... 137 Figure IV-10. Pulse field gel electrophoresis gels of undigested genomic DNA from pyrene degrading isolates...... 139 Figure IV-11. Phylogenetic relationships of pyrene dioxygenase large subunits...... 140 Figure IV-12. Gene organization for 5 strains of pyrene-degrading Mycobacteria...... 143 Figure IV-13. GC content (%) of Mycobacterium sp. KMS chromosome (outer circle) and plasmids...... 145 Figure IV-14. Comparison of Mycobacterium sp. strain MCS plasmid 1 and Mycobacterium sp. strain KMS plasmid pMKMS02 genes...... 146 Figure V-1. Sampling sites on Lake Erie...... 158 Figure V-2. Water column temperature and oxygen concentrations...... 161 Figure V-3. Rarefaction analysis of Lake Erie 16S clone library...... 164 Figure V-4. Phylogenetic classifications of 16S clones recovered from Station 84 sediments...... 165 Figure V-5. Neighbor-joining phylogenetic tree of all 351 16S rDNA clones from sediment microbial community at station 84...... 166 Figure V-6. Neighbor-joining phylogenetic tree of proteobacteria 16S sequences from Lake Erie sediments, highlighting the β subdivision. . 167

Figure V-7. Neighbor-joining phylogenetic tree of proteobacteria 16S sequences from Lake Erie sediments, highlighting the γ subdivision. .. 168 Figure V-8. Neighbor-joining phylogenetic tree of proteobacteria 16S sequences from Lake Erie sediments, highlighting the α and δ subdivisions...... 169 Figure V-9. Phylogenetic neighbor-joining tree of Actinobacteria sequences recovered from Lake Erie...... 171 Figure V-10. Phylogenetic neighbor-joining tree of Verrucomicrobia sequences from Lake Erie...... 172 Figure V-11. Dilution series to determine PCR inhibition in Lake Erie sediments...... 173 Figure V-12. Comparison of microbial genotypes in 0-2 cm and 2-4 cm core sections...... 174 Figure V-13. Microbial gene abundances in Lake Erie sediments...... 176 Figure V-14. Relationship between nidA and Mycobacterial 16S rDNA gene abundances...... 177

xii

LIST OF ABBREVIATIONS

ANOSIM Analysis of similarities ANOVA Analysis of variance ARDRA Amplified ribosomal DNA restriction analysis ARISA Automated ribosomal intergenic spacer analysis BSM Basal salts medium CA Correspondence analysis CCA Canonical correspondence analysis CTD Conductivity-Temperature-Depth (water column profiler) DDT Dichloro-diphenyl-trichloroethane DGGE Denaturing gradient gel electrophoresis ESEM Environmental scanning electron microsopy FAD Flavin adenine dinucleotide FAM 6-carboxyfluorescein FRET Fluorescence resonance energy transfer HMW High molecular weight ISP Iron sulfur protein LH-PCR Amplicon length heterogeneity polymerase chain reaction LMW Low molecular weight MW Molecular weight MTBE Methyl tert-butyl ether NADPH Nicotinamide adenine dinucleotide phosphate NAPL Non-aqueous phase liquid NMDS Nonmetric multidimensional scaling PAB Particle associated bacteria PAHs Polycyclic aromatic hydrocarbons

xiii

PCBs Polychlorinated biphenyls PCA Principle components analysis PCR Polymerase chain reaction PFGE Pulsed field gel electrophoresis qPCR Quantitative polymerase chain reaction RDA Redundancy analysis rDNA Ribosomal DNA (i.e. gene encoding ribosomal RNA) ROX Carboxy – X – rhodamine SEM Scanning electron microscopy SSCP Single strand conformation polymorphism TAMRA 6-carboxytetramethylrhodamine TCA Tricarboxylic acid cycle T-RFLPs Terminal restriction fragment length polymorphisms

xiv

Part I.

Literature Review

1. Biodegradation of Polycyclic Aromatic Hydrocarbons

Polycyclic aromatic hydrocarbons (PAHs) are persistent environmental contaminants. Aerobic microbial biodegradation is the primary degradation fate pathway of these toxic and mutagenic chemicals in the environment. It is the catabolic conversion by microorganisms that results in appreciable attenuation of PAHs, and thus is of great interest with respect to bioremediation. As such, it is imperative to understand the dynamics of microbial biodegradation processes in the environment. Microbial biodegradation of PAHs has been well-documented; however complex ecological interactions means there are still many unknowns, particularly with respect to the distribution and dynamics of high molecular weight PAH degrading organisms, that have only been relatively recently described.

1.1 Polycyclic aromatic hydrocarbons

Polycyclic aromatic hydrocarbons (PAHs) are a class of chemicals that are ubiquitous and persistent environmental pollutants (Keith & Telliard 1979), many of which have toxic, mutagenic and carcinogenic effects on organisms. They are formed from incomplete combustion of organic materials. While there are many natural sources of PAHs, such as forest fires, the highest concentrations (and most hazardous for human health) tend to be associated with wastes from industrial coal facilities, such as coking and coal gasification plants, and in wood preservatives such as creosote. Coal tar creosote is a mix of approximately 150- 200 different compounds, consisting of 85% PAHs, 10% phenolic compounds and 5% N-, S-, and O- heterocyclics (Mueller et al. 1989). Sixteen PAHs have been identified by the US EPA as priority pollutants (Keith & Telliard 1979), and their structures and aqueous solubilities are shown in Table I-1.

Table I-1. Structures and solubilities of 16 priority polycyclic aromatic hydrocarbons (PAHs) identified by the US EPA (Keith & Telliard 1979). MW = Molecular Weight. Solubilities from ChemFinder (http://chemfinder.cambridgesoft.com/)

Aqueous PAH Structure Formula MW Solubility (mg/L H2O)

Naphthalene C10H8 128.1732 31.7

Acenaphthylene C12H8 152.1952 3.93

Acenaphthene C12H10 154.211 3.47

Fluorene C13H10 166.222 0.19

Phenanthrene C14H10 178.233 1.18

Anthracene C14H10 178.233 0. 043

Fluoranthene C16H10 202.255 0.265

Pyrene C16H10 202.255 0.013

Benzo[a] C H 228.2928 0. 014 anthracene 18 12

Table I-1. Continued

Aqueous PAH Structure Formula MW Solubility (mg/L H2O)

Chrysene C18H12 228.2928 0. 0018

Benzo[b] C H 252.3148 0. 0012 fluoranthene 20 12

Benzo[k] C H 252.3148 0. 00055 fluoranthene 20 12

Benzo[a] pyrene C20H12 252.3148 0. 0038

Indeno[1,2,3- C H 276.3368 0. 062 c,d] pyrene 22 12

Dibenzo[a,h] C H 278.3526 0. 0005 anthracene 22 14

Benzo[g,h,i] C H 276.3368 0. 00026 perylene 22 12

PAH contamination is of concern because of the documented genotoxic, cytotoxic and carcinogenic effects. In general, the higher molecular weight (HMW) PAHs (> 3 benzene rings) have higher lipophilicity and thus tend to be more toxic than the lower molecular weight compounds. Carcinogenic effects have been reported for HMW PAHs, due to cytochrome P450 monooxygenase-mediated epoxide formation (Harvey 1996). PAH exposure has also recently been linked to cardiovascular disease (Kim et al. 2005a).

The environmental persistence and toxicity of PAHs is due to the fused aromatic ring structure: this structure is nonpolar and very thermodynamically stable because of delocalization of the double bond π-electrons, resulting in low solubility and resistance to nucleophilic attack. The low aqueous solubilities tend to decrease with increasing molecular weight (Table I-1). Because of these low solubilities, these compounds are poorly bioavailable, and thus environmentally recalcitrant (Cerniglia 1992, Priddle & MacQuarrie 1994). Bioavailability is further reduced in soils and sediments, where there is limited mixing and hydrophobic PAHs may be adsorbed onto particles or sequestered into pore spaces.

1.2 Microbial degradation of PAHs

In fungi and other eukaryotic organisms, PAHs are transformed as a detoxification mechanism, with the aim of increasing solubility to aid in excretion. In contrast, microbial biodegradation of PAHs typically results in carbon assimilation and energy production (the exception being cometabolic degradation, where the transformation does not produce carbon or energy for the microbe).

PAHs arise from incomplete combustion or oxidation of plant products; as such their existence in the environment is ubiquitous. Microorganisms have had a relatively long time to adapt and evolve a diverse array of degradation pathways for these compounds (Phale et al. 2007). The biodegradation of naphthalene by soil microorganisms was first observed and published in 1927 (Tattersfield 1927), and the catabolic pathway determined in 1964 (Davies & Evans 1964). Following that, microbial degradation pathways were elucidated for phenanthrene and anthracene (Evans et al. 1965, Barnsley 1983, Kiyohara et al. 1994, Moody et al. 2001, Prabhu & Phale 2003), and fluorene (Casellas et al. 1997). More recently, microbial pathways for the degradation of high molecular weight PAHs have been characterized. Because of their high thermodynamic stability and hydrophobicity, these compounds were long thought to be resistant to microbial biodegradation (Herbes & Schwall 1978, Cerniglia 1992, Kanaly & Harayama 2000). Pathways have been determined for the degradation of fluoranthene (Rehmann et al. 2001, Lee et al. 2007), and pyrene (Kim et al. 2007). Transformation of benzo[a]pyrene has been reported (Schneider et al. 1996), however few organisms are currently known that have the metabolic capability to mineralize such a large PAH; as a result degradation pathways have only been partially elucidated (Moody et al. 2004).

The acquisition of compounds like PAHs by microorganisms in the environment is dependent on several factors. First, the bacteria must come in contact, or at least in proximity, with the compound, which requires dissolution in water. The rate of mass transfer of the compounds to the cells is described as bioavailability, and is commonly the limiting factor in biodegradation of PAHs (Bosma et al. 1997, Harms & Bosma 1997). This also explains why the higher molecular weight PAHs, which are more hydrophobic, tend to be extremely recalcitrant and resistant to degradation in the environment. In addition to chemical constraints, physical separation also contributes to reduced bioavailability in soils and sediments: adsorption to particles, pore space sequestration, and dissolution in

non-aqueous phase liquids (NAPL) greatly minimizes the chance of contact between organisms and PAHs. Reduced bioavailability is also exacerbated by the age of the sediments, with older contaminants being more resistant to biodegradation (Hatzinger & Alexander 1995). Some PAH-degrading organisms have evolved strategies to overcome bioavailability challenges; for example, production of biosurfactants, extracellular polymeric substances, and/or biofilm formation (Alexander 1999, Samanta et al. 2002, Johnsen & Karlson 2004, Bamforth & Singleton 2005, Johnsen et al. 2005).

The second step in bacterial metabolism of contaminants is the actual uptake of the compounds. In the case of PAHs, their hydrophobicity allows them to pass across a bacterial cell wall with relatively high frequency. Therefore uptake is not thought to be a limiting step in bacterial metabolism of PAHs (Bateman et al. 1986). This observation is reinforced by molecular biological evidence: biodegradation gene operons do not contain genes for transport systems, and aromatic catabolism genes have been successfully cloned into E. coli, which then exhibited efficient metabolism without any special uptake system (Zylstra et al. 1989).

Once the PAHs are within the bacterial cell, the PAHs are metabolized via enzymatic degradation pathways. In aerobic PAH degradation, the initial attack involves the addition of two oxygen molecules by a dioxygenase enzyme, which polarizes and destabilizes the aromatic ring structure. (Dioxygenases are discussed in more detail in Section 2.) After a dehydrogenation, the aromatic ring is slightly polarized by the added hydroxyl groups and is much more amenable to nucleophilic attack by other enzymes. Ring cleavage enzymes catalyze the opening of the ring. For naphthalene, a two ring PAH, these initial enzyme steps result in production of salicylate and catechol, which are broken down by another ring cleavage dioxygenase to yield TCA cycle intermediates. For larger PAHs such as pyrene, the degradation pathway involves repetition of these steps for

each aromatic ring: rings are sequentially substituted, then cleaved (Kim et al. 2007).

Not all biodegradation and biotransformation reactions lead to TCA cycle intermediates (i.e. complete assimilative mineralization). Some pathways serve to detoxify compounds by the addition of functional groups, but do not actually cleave the ring structure (Cerniglia 1984, Kim et al. 2007). PAHs can also be transformed to intermediates via cometabolism, that is, transformation of compounds that cannot be used a carbon source by the organism (Boonchan et al. 2000, Dean-Ross et al. 2002). While this is not the most energetically favorable in pure culture, it likely plays a role in mixed communities in the environment, where cometabolism may be involved in cooperation between organisms (Keck et al. 1989).

1.3 Bioremediation and natural attenuation

Physical removal of aromatic pollutants such as coal tar and creosote is often difficult and expensive, and the feasibility of removal of all or most of a pollutant is negligible, especially from soils and sediments. Chattanooga Creek, discussed in more detail in section 5, represents a prime example: contaminated sediments from one section had been previously excavated, however significant quantities of PAHs still remain. There is unlikely to be further active cleanup on the site and contaminant removal must therefore occur through natural attenuation, that is, “biodegradation, dispersion, dilution, volatilization, and/or chemical transformations which serve to reduce the mass, toxicity, mobility volume or concentration of contaminants” (US-EPA 1999). In aerobic environments such as Chattanooga Creek, microbial biodegradation is the primary process contributing to degradation of coal tar constituents and natural attenuation of PAHs (Mueller et

8 al. 1989). As such, it is of fundamental importance to understand contaminant degradation carried out by indigenous bacterial communities.

Bioremediation aims to employ biodegradative organisms for the purpose of contaminant transformation or reduction in concentration. The strategies generally include biostimulation of indigenous microbial populations or bioaugmentation through the introduction of nonindigenous biodegradative strains or consortia. Biostimulation may include alterations of temperature, oxygen, redox, moisture content or nutrients in the contaminated zone to encourage the growth and biodegradation of the native community (Bamforth & Singleton 2005). Bioaugmentation strategies introduce biodegradative strains and/or soils with active contaminant degradation activity to a contaminated site, in attempt to encourage the spread of the beneficial organisms (Cerniglia 1992).

Unfortunately, bioaugmentation is often ineffective. Introductions of biodegradative organisms to contaminated sites are unsuccessful because the appropriate in situ conditions required for growth are not supplied, largely due to ignorance: Laboratory studies often focus on one organism and one compound, however polluted sites present a mixed consortia with a mix of contaminants and other carbon sources which may be more readily used. Bioaugmentation strategies often fail because of a lack of understanding of in situ processes, including: metabolism in the presence of other carbon sources, preferential degradation of compounds within a mixture, complex nutrient requirements, transcriptional regulatory responses, and/or cooperative interactions between organisms (Shuttleworth & Cerniglia 1996, Bouchez et al. 1999, Pelz et al. 1999, Leblond et al. 2001, Shingler 2003, Watanabe & Hamamura 2003). For example, addition of a PAH-degrading Mycobacterium strain to creosote contaminated soil showed statistically similar degradation activity to soil treated with nutrients alone (Juhasz et al. 2005). In another example, attempts to prime PAH-contaminated soil with bioremediated soil dominated by Mycobacterium spp. indicated that over

9 time, biodegradation rates were comparable between primed and unprimed soil (Johnsen et al. 2007). Enhanced biodegradation was observed at a coal gasification site when indigenous microbes were reinoculated to the site at higher concentrations (Grosser et al. 1991). These observations support the idea that indigenous microbial communities in contaminated soil often already have the catabolic capacity for biodegradation: this highlights the need to understand the ecology of biodegradative strains, populations, and communities as a basis for any decisions regarding bioremediation.

Isolation of PAH-degrading organisms has provided valuable insight into PAH metabolism and genetics. However, the application of this knowledge to remediation of contaminated sites is dependent on also understanding the ecology and dynamics of microbial communities in situ, in a culture-independent manner. Early studies quantified microbial community PAH transformation rates in the environment, demonstrating an inverse relationship between molecular weight and transformation rates (Herbes & Schwall 1978, Herbes 1981). These studies also indicated that community transformation rates were independent of PAH concentrations and total microbial biomass, paving the way for studies on bioavailability of PAHs as well as the role of particular functional populations within a community. Current ecological biodegradation studies aim to identify and quantify functional groups of microorganisms responsible for biodegradation, ultimately relating these groups back to biodegradation activity (Watanabe & Hamamura 2003). This dissertation work is essentially a continuation of these early PAH transformation studies, centered on the need for a better understanding of indigenous communities, and employing molecular tools to monitor population dynamics in PAH contaminated sediments.

1.4 Ecology and biogeography of PAH-degrading organisms

The presence of bioavailable PAHs in the environment will select for organisms that can use these contaminants as carbon source. As a result, microbial communities adapt and change depending on the type and concentration of PAHs (Ringelberg et al. 2001, Muckian et al. 2007, Uyttebroek et al. 2007). These community changes have been documented in terms of total biomass, phenotypes (i.e. types of organism) and functional genotypes (i.e. types of catabolic genes).

Microbial genera associated with PAH degradation have been isolated and examined in situ from a variety of environments. The fact that similar organisms are found in multiple locations supports Martinus Beijerinck’s early 20th century hypothesis that “everything is everywhere—the environment selects”, stemming largely from Beijerinck’s discovery that similar microbes could be isolated from diverse locations using enrichment cultures (Beijerinck 1913). This hypothesis essentially states that microbial diversity, distributions and abundances are the product of current environmental influences, and not historical factors (i.e. genetic isolation). It does not imply a random distribution of all microbes, in fact, there have been numerous studies documenting the nonrandom biogeographical distribution of microbial communities and genotypes (Fulthorpe et al. 1998, Cho & Tiedje 2000). Rather, the “everything everywhere” hypothesis addresses the cause of these biogeographical patterns by suggesting that dispersal rates are high enough that the explanation for the current biota is found in the current habitat: “the environment selects.” In the case of PAH-degrading microbes, this implies that under similar environmental or enrichment conditions, similar organism should be found.

Isolation of similar biodegradative organisms from geographically disparate locations strongly supports the idea that biodegradative organisms are widely dispersed (Fries et al. 1994, Mueller et al. 1997, Poelarends et al. 2000, Leys et al.

2005a). Further evidence for the “everything everywhere” hypothesis comes from observations of community shifts and isolation of biodegradative strains in the presence of similar contaminants (i.e. enrichments). In particular, observations from a variety of locations have indicated that enrichment on naphthalene frequently leads to an increase in γ-proteobacteria, especially Pseudomonas (Stoffels et al. 1998, Singleton et al. 2005, Castle et al. 2006, Ní Chadhain et al. 2006); enrichment with fluoranthene and phenanthrene often results in a dominance by Burkholderia and Sphingomonas (Mueller et al. 1997, Ní Chadhain et al. 2006). Pyrene-contaminated or pyrene-enriched sediments result in shifts in abundances of Mycobacteria (Dean-Ross & Cerniglia 1996, Churchill et al. 1999, Jouanneau et al. 2005, Leys et al. 2005b, Hilyard et al. 2008), as well as Burkholderia, Sphingomonas, and Stenotrophomonas (Ní Chadhain et al. 2006). While there are exceptions to these general observations, the fact that multiple studies have revealed similar patterns from diverse locations and communities is testament to the idea that many biodegradative strains are cosmopolitan in distribution.

2. Diversity of PAH Dioxygenases: Applications in Ecology of Biodegrative Microorganisms

Aerobic microbial biodegradation of aromatic compounds is mediated by oxygenase enzymes which add either one (monooxygenase, hydroxylase or mixed function oxidase) or two (dioxygenase) hydroxyl groups to the aromatic ring. Dioxygenases fall into two classes: the first are ring-hydroxylating dioxygenases, which catalyze the addition of hydroxyl groups onto an aromatic ring. The second class, ring-cleavage dioxygenases, catalyze the addition of hydroxyl groups onto aromatic diols (e.g. catechol) followed by either intra- or extradiol cleavage of the ring structure (Figure I-1). Ring cleavage dioxygenases were first discovered by O. Hayashi, who demonstrated that a ring cleavage enzyme converting catechol to cis,cis-muconic acid involved incorporation of atmospheric

dioxygen (O2) (Hayaishi et al. 1955), contrary to the prevailing thought that biology obtained oxygen from water . The same year, H. S. Mason also published similar results using a monooxygenase enzyme (Mason et al. 1955). Later, the activity of ring hydroxylating enzymes, was elucidated by D. T. Gibson (Gibson et al. 1970). This class of dioxygenases add hydroxyl groups to typically unsubstituted aromatic rings, which destabilize the ring structure, or reduces the aromaticity, allowing further degradation of the compound, including attack by ring cleavage dioxygenases. An example is naphthalene 1,2 dioxygenase , which catalyzes the conversion of naphthalene to cis-1,2,-dihydroxy-1,2- dihydronaphthalene (Figure I-2).

Figure I-1. Intra- vs. extradiol cleavage.

Figure I-2. Dihydroxylation of naphthalene catalyzed by naphthalene 1,2- dioxygenase.

2.1 Ring-hydroxylating dioxygenase structure and specificity

Aromatic ring dioxygenases are multicomponent enzymes that consist of a catalytic hydroxylase component (terminal dioxygenase) and electron transport chain components (iron-sulfur flavoprotein reductase and an iron-sulfur ferredoxin). The terminal dioxygenase, or the hydroxylase component, has a large (α) and small (β) subunit, generally around 50 and 20 kDa, respectively.

The subunits are arranged in an (αβ)n configuration; for example, naphthalene

dioxygenase from Pseudomonas is a trimer of dimers, i.e. α3β3 (Kauppi et al. 1998). The α subunit has two cofactors: a mononuclear nonheme iron and a

conserved [Fe2-S2] Rieske iron-sulfur center. These cofactors are responsible for the activation of dioxygen, which is normally not reactive. Because of these cofactors, the terminal dioxygenase is also sometime referred to as the iron sulfur protein (ISP).

Ring-hydroxylating dioxygenases require NAD(P)H as an electron donor (ring cleavage dioxygenases do not); the reducing power is passed from NAD(P)H to the terminal dioxygenase along a short electron transport chain. During

hydroxylation, NAD(P)H is oxidized by a flavoprotein (NADH-ferredoxinNAP reductase), transferring the electrons to FAD of the ferredoxin reductase. The FAD then provides electrons to the iron-sulfur center of the ferredoxin, which then transfers them to the iron-sulfur center in the active site of the terminal dioxygenase. The terminal dioxygenase catalyzes the addition of oxygen to the aromatic ring yielding a cis-dihydrodiol product which is then dehydrogenated by a cis-dihydrodiol dehydrogenase. Once slightly polarized by the added hydroxyl groups, the aromatic ring can be oxidized and cleaved by a ring cleavage dioxygenase, either between the hydroxyl groups (intradiol cleavage, ortho degradation pathway) or adjacent to the hydroxyl groups (extradiol cleavage, meta degradation pathway) (Harayama et al. 1992).

Many of the dioxygenases are able to transform a range of compounds. In particular, naphthalene dioxygenases demonstrate activity on a wide range of substrates (Sanseverino et al. 1993a), although shows preferential degradation of naphthalene over other PAHs (Leblond et al. 2001). Ring-hydroxylating dioxygenases are induced by the presence of the substrate; thus can be used as a biomarker of potential biodegradation capabilities or exposure to bioavailable compounds. Upon exposure to a compound, such as naphthalene, members of a microbial community with the biochemical pathways and naphthalene dioxygenase genes will be able to take advantage of the carbon source and proliferate in the community. Thus, PAH degradation genotypes (i.e. dioxygenase genes) within a community are indicative of PAH degrading populations.

The α subunit of the terminal dioxygenase contains the active site, which includes a number of hydrophobic amino acids and the highly conserved Rieske iron-sulfur center, and thus conveys the substrate specificity (Parales et al. 2000). Amino acid sequence alignments between the subunits of terminal ring hydroxylating dioxygenases indicate that the α subunit is much more conserved between organisms than the β subunit, though similarities between the two suggest a common evolutionary origin (Harayama et al. 1992). Therefore, because of the substrate specificity, researchers wishing to monitor dioxygenases in the environment generally target the genes encoding the α subunit (e.g. nahAc in Pseudomonas sp.) for the development of molecular tools. The characterization and quantification of dioxygenase genes in mixed communities can provide insight into the ecology of degradative organisms, populations, and communities.

2.2 Genetic organization of PAH-catabolism genes

The diversity of aromatic compounds and the diversity of the organisms that can degrade them necessarily means that PAH catabolism genes are arranged in a variety of genomic architectures. Often PAH catabolism genes are clustered into large operons. The genes for the multicomponent ring hydroxylating dioxygenase are generally adjacent to each other, with genes for the α- and β- subunits, ferredoxin and flavoprotein reductase in a single operon or gene cluster. It is not uncommon, especially in the characterized naphthalene degrading organisms, for the genes to be organized into two pathways: one operon will code the upper pathway enzymes (conversion of naphthalene to salicylate), and a second operon will code the lower pathway enzymes (salicylate to TCA cycle intermediates) (Yen & Gunsalus 1982). In Pseudomonas, these two operons are regulated by a trans-acting positive control regulator (NahR) (Schell 1985, Yen & Gunsalus 1985). The order of the upper pathway genes are very conserved between Pseudomonas spp.; a similar organization is also seen in other organisms (such as Ralstonia sp. U2) despite different gene sequences (Habe & Omori 2003).

Horizontal gene transfer has played an integral role in the acquisition and evolution of bacterial biodegradative pathways (Tsuda et al. 1999, Top et al. 2002, Top & Springael 2003, van der Meer & Sentchilo 2003, Nojiri et al. 2004, Springael & Top 2004, Phale et al. 2007). PAH catabolism genes have been found on large plasmids (Sayler et al. 1990, Romine et al. 1999, Iida et al. 2002, Basta et al. 2004, Dennis & Zylstra 2004, Dennis 2005, Habe et al. 2005, Jussila et al. 2007), and associated with transposons and genomic islands (integrative genetic elements) (van der Meer & Sentchilo 2003, Springael & Top 2004, Gaillard et al. 2006). Plasmid-encoded PAH catabolism genes have been shown to play a role in naphthalene degradation (Sanseverino et al. 1993a, Dennis & Zylstra 2004, Li et al. 2004). These mobile elements generally range from 10 – 500 kb in size and often contain complete catabolic operons as well as transfer

genes (e.g. transposases). Catabolic plasmids tend to be self-transmissible broad host range plasmids (IncP-1) (Nojiri et al. 2004). In particular, the NAH plasmids, carrying naphthalene catabolism gene clusters, have been found in a large number of naphthalene degrading Pseudomonas spp. Similar patterns of gene organization between phylogenetically disparate organisms suggest common evolutionary origins for many PAH dioxygenase genes (Goyal & Zylstra 1997, Habe & Omori 2003). Other organisms, such as some Sphingomonas strains, display degradation genes scattered in different gene clusters throughout the genome, suggesting some shuffling of these genes in the past (Pinyakong et al. 2003).

2.3 Diversity of PAH dioxygenases

Figure I-3 and Figure I-4 shows the phylogenetic relationships of the large (α) subunits of aromatic ring hydroxylating dioxygenases. These neighbor-joining trees are based on alignments of 149 translated sequences currently in NCBI’s GenBank. Most of these sequences are PAH dioxygenase genes, however some other aromatic dioxygenases have also been included for comparison (e.g. phthalate, carbazole, and benzoate dioxygenases). Phylogenetic taxonomic affiliations of the organisms are displayed in Figure I-3, showing that the majority of known PAH degrading organisms are γ-proteobacteria. The primary substrates catalyzed by the enzymes are displayed in Figure I-4, showing that the majority of known PAH dioxygenase enzymes use naphthalene and phenanthrene as a primary substrate. As predicted, there is a much greater diversity in low molecular weight PAH dioxygenases, which are found across phylogenetic groups, compared with the high molecular weight dioxygenases which have currently only been observed in gram positive organisms, particularly Mycobacteria.

p H o N

b 5 4 D 1 A T I 2 I 1 T Z 6 X Y 8 c 9 1 L Z 2 7 D 1 1 b C Y s P 8 p s 0 a s N Y a C 8 h s A a 3 s a C n L l 2 t3 a n s a 8 n s L a H 3 U o n a n 1 o s I o a S N S D o I P 1 m n o n a C 4 P n n m A 8 m o 2 1 o m n I 6 - D a m o I 6 a p o I 2 c 1 o 6 T A 3 d o o I 0 h h m o 8 1 a s d 3 3 2 d m Z N u 3 F A A d 5 s r e o d m Y 2 3 u A 2 u o s 81 P 2 e u a a M d c d u d u o C A e 9 5 9 e d a s 3 s d n a C e i d u e d i L 4 t A s a 9 A s u t n B o A s R 8 o P P s I s 3 Y D e u o n Y t P e u 1 a P a P 4 4 e P s m d 9 0 8 P s e C o - 0 7 p s 7 A l p m o n i 7 s 6 s 6 t 4 9 c 8 P a 1 4 6 N m u 4 P o A F t 4 o d 1 8 a 2 P n S 8 0 o 6 P o C 1 1 t 2 6 4 d u 8 7 l 8 9 m o 6 i 4 1 7 i s 1 da d T 9 d 1 g 9 1 0 u i e o A p 5 8 2 6 9 6 4 t 6 u T 4 s m B a 4 4 d F t a 9 e e 7 5 7 e 6 2 5 1 ut e I d 0 o A 2 1 u P N 9 4 s u r 9 P 2 n 1 1 n P 4 0 4 p s d a C 0 4 o 3 c P 6 6 e 1 s e M 6 4 P 2 5 A1 3 0 9 3 P u d N i n Q s i m e s F 0 P f Y 9 P e t a S 2 f Y 0 3 a a lu u P i H 6 5 2 P l B D 3 s u B 1 o A C 6 d n u A 4 i t 0 a D o d A Y Y 7 p t 2 d O 1 F 0 P a b t b o R Y 6 b c N 96 A A 8 r o c 4 1 u d 8 4 M 0 c A A 7 i i i e r c A 0 4 P 1 e 5 l B 6 A 0 4 4 p t a 8 A 3 i e m A 2 A c 1 y s s A 3 3 u t 4 0 A c 6 9 9 d A D B a s 3 B 2 4 4 i c D8 E c 2 h h c P p t s 1 o h A F 4 C 0 n c 1 h 4 6 8 3 c o e o A c 9 u n a P n a h A M 6 7 S a 4 b i 5 e a A 2 P e p 0 a h U Y 6 p D n d s nahAc AF306432 P putidaPR1MN1 A 0 6 n 0 r n c 7 a A a a h 4 A n n a a x h P n Y 1 a s O 5 0 s n a h B A 3 P b y c 5 s a C 0 n 3 a 4 t il 8 s n x A 13 7 id e S e lX e a p 0 C n a A c A F 0 4 4 t r o s D d A 8 r B 0 u n o A F 9 8 u o m a C b F i 5 A n ah h A c 4 0 u n C to 1 p 0 M P u K A t d A 1 2 p fl o o d h 6 6 m id 1 n a h A c F 0 4 d sS C A A r M c n a h 2 P P u m a 1 1 F 1 0 4 e D i a 0 A A F 0 4 e o n 1 1 4 7 s E n U n a h C c A 0 1 9 s d o S 1 m Y 4 D 8 4 in h e n a o F 3 P u m a 1 cb 18 8 8 9 7 o A 9 t C 0.1 n A 3 A 9 9 e o s A t 2 4 8 S P v 5 o C d h 3 3 0 s d o 5N a od 4 96 0 p o 5 b n A 2 2 3 05 P u in s tc A C 5 2 h p r a 2 na h C M 4 8 2 se g na bA J 1 P P 1 i u a c 2 a o Y 4 5 P ru o a 00 J p p R n ti n t n d B A 4 0 1 e m U 6 0 u u go d s e n o F 8 5 a o T1 Te 1 30 49 ti ti er m a C r d c A 44 0 P ud Z cA 52 7 9 da da y M A n A c F 8 7 e 7 sL 1 98 R 6 D M th on t- 1 h A A 44 04 Ps aG na 3 b U7 P al P O L ro a 2 0 na h c F 8 5 id o Y1 ph 80 se sto pu T 2 p s a A A 44 9 ut om sC A1 99 u n ti T o C n h c F 63 p ud na 14 D B do ia da 1 lis B a A A 9 P se o Y i 321 ur m JS F 3 n ah c J4 49 P om sLC pbA 42 kh on 70 1 n hA A 39 46 ud na 1 U2 Rh old as 5 a Ac 8 92 se mo 1 2 4 od eri P5 n h M 17 3 P do s4N bpd 277 oc aP 1 na Ac Q 16 seu ona C1 R occ S1 ah c D 694 2 P om D 1 U27 ery us 2 n hA Y 16 eud aSC 591 thro RH na c A 694 Ps inos b bph Rh pol A1 hA AY 392 rug phA1 A1 X odo isB na Ac 96 P ae as D173 800 cocc D2 ah AJ4 49 omon 19 P 41 R usM n Ac 4480 olar bph seudo glob 5 nah AF 32 P onas A U47 mon erul hAc 1949 larom 1-17 637 C asKK us na c AY 74 Po lderiaS bphA1 testost S102 nahA 1674 Burkho M83673 eroniB Ac DQ 448048 OD-5b P pseud -356 nag hAc AF holderiaS bphA oalcalige na 8053 Burk 1 M86348 B nes hAc AF44 urkholderiaLB na alstoniaU2 d 400 Ac AF036940 R bdCa AB121977 Xanth nag obacter tAc U62430 BurkholderiaRASC dxnA1 X72850 dn SphingomonasRW1 ntdAc U49504 PseudomonasJS42 thnA1 AF157565 Sphingopyxis d ntAc AF169302 B cepacia rabacterDBF63 n R34 1 AB084235 Ter i bzAc AF3796 phtA cterkeyser nah 38 Comamona 3 Arthroba PYR1 Ac AB06 sJS765 AF33104 baalenii 8 phnAc 6442 Rals phtAa M van B1203 AY56 toniaNI1 365117 NCIM N phnAc 8278 C htAa AY occus us1B n AY3 testos p odoc ococc us ahAc 67788 teroni 63 Rh od opac pa AY1 Delf F0826 612 Rh 1 R 00 hAc 949 tia ac rAa A J401 4688 sP2 n pa AF 31 C idovo na a A Q8 ccu I24 do hA 2525 test rans narA Aa D oco cus 0 na C2 c A 50 oster nar hod coc 40 hA AF B00 C te oneC 4 R do sP 14 na c1 004 405 stos J2 42 ho ccu N n hAc A 28 9 P tero 392 5 R co TK us a F30 3 P pu ni AY 190 do us ac Y1 n hA AF 64 flu tid rAa 12 ho ac op P 8 S ah c 29 41 or aO na AF 3 R op R m ZH L 1 na A AF 5 P es US dA 42 R 3 riu C J 0 h c 3 03 pu cen 82 ni 92 1 63 te m m 1 5 A AF 06 3 M tid sA Y3 67 0 ac riu riu H 6 na 3 3 42 a a5 TT A 06 11 b te te Z S S n h D8 06 0 rin II C1 Aa B2 B co c c C m C 1 a Ac 4 4 P ob AN 74 ar A A y a a m u M P n h 1 30 b a H 83 n A dA M ob ob vu ri 1 n a A A 46 P al cte id o 45 c c il te m H B S a h c F ea r n d 7 y y g c iu M B 5 n hA n A A 0 P p ri N 4 M M a r M d a a c Y 3 a ut c CE 9 1 M b te m m 6 1 h h A 1 9 e id aL 3 J4 4 98 2 o c iu u K S H 9 p b Ac c A 9 53 ru a S 1 9 r v n 1 f F g 3 4 2 A 7 00 94 yc a e il m m 1 phnAh A A c 3 68 3 i II 01 1 3 b t g iu A 2 Y n F A 0 P n IA A 5 33 7 M o c r iu mC 2 p p A B AF 0 6 2 os 2 o Q 3 c a r F H P p p p h F 4 7 s d Y 5 4 y M e u YR H K p h 0 5 0 t a p D b t e i e C p h h h n c 5 2 2 P u P A Q 0 t r 6 n n 3 t c s P B 1 h 5 a M o 7 C p d 7 z A 9 1 n n A c 9 7 c H I p h n A 4 3 s K A a p A e e c d 6 4 n m p a n t D te m L p 5 i d 2 a m h b A A 9 3 7 P u r 1 i y b G h n A c A Y n 4 3 8 i A c e u m C h h c 0 5 9 A 0 u 4 h A 7 3 t b Y h t n d 5 o u i i n c c 3 s z i 1 M 8 a g i H n A c A A Y 3 N 1 ya 6 o r u F n A 5 t e 9 N F c r r i r KP7 n

n 5 c A u n 4 R e A 6 6 0 b H s H 3 T r 1 A 7 y c e r s A c c A A Y N t 5 e e s A A i e 1 3 t R A c 7 z 6 0 3 y o t Y a h A Y 67785 n e t A c A P 3 F H ku B Y 3 3 C e A 7 M c b c t e a c A Y A 7 e a n i c c r Y F m c c p

n s s r d Y A M s c A 1 6 8 r s Y a 3 p i 9 y a c d n A 0 i 0 a F r e L sP a i e o A Y 1 A u

n 7 k P o u 5 7 6 h p A 6 a 0 5 i b a t i o a s A B 1 2 u S A b A M Y 3 t h

1 r n b o i n 5 7 b m e c 4 A 1 0 o i c F 1 4 7 d ho M d n

b n B o i Y t m 5 B i e Y 0 8 a d o n Y 4 8 h

0 9 3 m t 3 7 o c n 1 i u 5 N c o d a 0 9 u n a o i 4 A d o i 6 0 c o 3 5 c r 1 3 7 B a v n 6 e ya

y s 1 5 8 y o r c b l 1 5 4 4 r t or v 3 e sce 6 3 l b 3 y g u 6 3 k A 4 a 5 6 4 ur e t o

5 8 r y m a B g o 5 5 4 3 e M M a a 1 h d Y 5 87 f l 6 6 r a r S o c n g 4 i r 4 4 D r a M o v 4 u k n i 3 6 B

5 4 o n F M ve a c H n A 3 r i i 7 2 7 A o o g 4 1 n n 3 r hol l B s 3 M h g 3 0 2 6 B m a b 3 u A c d 0 6 l 3 e h o 5 e l k s n 6

4 a A i l N 6 B B c F f p n i 6 5 r h e 8 Q 5 8 4 n t 5 o d 3 5 T p i 1 B r k s a

A c der 6 i 3 p b u o r n 1 h 4 2 R a 9 D 9 7 7 4 S a o 6 h p l h i 3 n S y B h a a 3 M 0 B a r i l 4 5 p d 4 v 2 g g o d 9 B k 2 p a h C s B A m e 2 B 3 m s i

8 2 1 C p 1 s u l h e l e E 9 3 9 7 5 p l aC

d e d s a 4 S u n p 6 s i F 5 o r n o i H 4 M g r o Q 1 4 7 6 1 1 p h e i r

6 t n n 5 g t a i 8 6 r k e a A J v h r l - 2 l h3 3 1 6 1 o r D F h e a d a

F Q 8 a z 4 4 l h il E 1 2 e s i 1 8 A 2 o a n e a 5 t e

A e n z i lu A 7 D 3 E t o h A 1 9 n Ch t i 4 9 h i f c r d B0 5 N S a i

n a 1 h 2 2 a i 6 f l n i m d i A 1 i 5 Q 1 e d u a A 3 a z -1 n A a P e 8 e

A 1 0 i d 7 6 i n d F i e u m H D 1 Q 3 b i c i D 2 Q z i i

o 1 n J 1 - A 1 s m n A A H r g B a 6 o 1 n i d 0 D i 5 H

n o i D h 3 B d A 3 r u l J a 1 T g H i A p 3 g s i 9 l

h o Q p 5 g m A R A u a 1 d A A 2

1 g 1

i p b 7 d 4 D n A m t n P H 1 4 n 1 e p a 3 0 h h 0 0

3 4 H H g

1 A p F p Q 0 A 0 1 A n g g A 7

1 d D 6 i n h 8 1 f

p 1 n h c 1

6 p A A

n h

h p

A p b

p b

Figure I-3. Phylogenetic relationship of α subunit of aromatic ring- hydroxylating dioxygenases, showing taxonomic affiliations of the organisms. The phylogenetic groups are γ-proteobacteria (▼), β-proteobacteria (■), α- proteobacteria (♦) and Gram positive organisms (●). Branch names list gene name, GenBank accession number and organism.

p H o N

b 5 4 D 1 A T I 2 I 1 T Z 6 X Y 8 9 1 c Z L 2 7 D 1 1 b C Y s P 8 p s 0 a s N Y a C 8 h s A a 3 s a C n L l 2 t a n s a 8 3 n s L H 3 U o n a n 1 o s I o a Sa N S D o I P 1 m n o n a C 4 P n n m A 8 m o o - 2 1 m n I 6 D a m o I 6 a p o I 2 c 1 o 6 T A 3 d o o I 0 h h m o 8 1 a s d 3 3 2 d m Z N u 3 F A A d 5 s r e o d m Y 2 u A 3 2 u o s 81 P e a M 2 u a d c d u d u o C A e 9 5 9 e d a s 3 s d n a C e i d u e d i L 4 t A s a 9 A s u t n B o A s R 8 o P P s I s 3 Y D e u o n Y t P e u 1 a P a P 4 4 e P s m d 9 0 8 P s e C o - 7 p s 7 A 0 l p m o n i 7 s 6 s 6 t 4 9 c 8 P a 1 4 6 N m 4 P o A F t u 4 o d 1 8 a 2 P S 0 o 6 P n C 8 6 o 1 1 t 2 d u 8 7 l 8 4 9 m 6 i 4 o 1 7 i s 1 da d T 9 d 1 9 1 0 u i e o A p 5 g 8 2 6 9 6 4 t 6 t u T 4 m B a 4 4 s d F t a e e 7 7 e 6 9 2 5 5 1 e I d 0 o A 2 1 u P N 9 4 s u 9 P 2 n r 1 1 n P 4 0 4 pu s d a C 0 4 o 3 c P 6 6 e 1 s e M 6 4 P 2 5 A 3 0 9 3 P u d N i n Q s i m e s F 0 P f Y 9 P e t a S 2 1 f Y 0 3 a a lu u P i H 6 5 2 P l B D 3 s u B 1 o A C 6 d n u A 4 i t 0 a D o d A Y Y 7 p t 2 d O 1 F 0 P a b t b o R Y 6 b c N 96 A 8 r o c 4 1 u d 8 4 M A 0 c A A 7 i i i e r c A 0 4 P 1 e 5 l B 6 A 0 4 4 p t a 8 A 3 i e m A 2 A c 1 y s s A 3 3 u t 4 0 A c 6 9 d A D B a s 3 B 2 4 4 i c D E c 2 h h c P p t s 1 o h A F 4 C n c 1 h 49 6 8 3 c 0 o e o A c 9 u n a 8 P n a M 6 7 S a 4 b i 5 e a h A A 2 P e 0 a h U Y 6 p D p n d A s nahAc AF306432 P putidaPR1MN1 0 6 n 0 r 7 a A a n a h 4 c A n n a a x h P s n B A Y 1 a s O 5 0 n a h 3 3 P b a y c 5 s s a C 0 n 3 4 t il 8 n x A F 1 7 id e S e lX e a p 0 C n a A c 0 4 4 t r o s D d A 8 0 n o hA 9 o a C b F r B u 5 A n h A c A F 0 8 u m C t M P iu d a 4 2 p lu o n o 1 p 0 K A t a h A c A 1 f d o sS d A h 6 6 r m M c id 1 n F 0 4 P u a C A 4 e 0 n a h A 2 A 0 P m 1 1 F 1 0 7 D in a F 4 e o n 1 Y 1 4 4 s E U n a h C c A 0 1 9 s d o S 1 m 1 4 D 8 7 in h e n a o F 9 3 P u m a 1 cb 8 8 8 9 o A 9 t C 0.1 n d A 3 A 9 e o s N A t 2 4 8 S P v 5 o C h 3 3 0 s d o 5 a od 4 9 0 p o 5 b n a A 2 2 3 05 P u in s tc A C 5 6 2 h p r a 2 n h C M 4 8 2 se g a b J 1 P P 1 i u a c 2 a o Y 4 5 P ru on Aa 0 J p R n t n t n B 4 0 e m U 06 0 pu u g id s e d o A F 8 51 a o 1 Te 1 30 49 t ti e o a C r n d c A 4 0 P d ZT cA 52 7 9 id da ry m M A n A F4 8 7 eu 7 sL 1 9 R 6 aD M th o t 1 h c A 4 4 s G na 3 b U 8 a P O r n -2 0 a A F4 80 P da o 1 ph 78 Ps lst p T L2 o a n ah c A 4 95 ti m CY A1 09 eu on ut T po sC n hA c F4 3 pu do as 4 D 9 B do ia id 1 li B a A A 96 P eu on Y1 i 32 ur m JS aF s 3 n h c J4 49 Ps om sLC pbA 142 kh on 70 1 na A A 9 46 ud na 1 U R old as 5 ah c 83 92 se mo 1 2 24 hod eri P5 n hA M 17 3 P do s4N bpd 277 oc aP 1 na Ac 16 seu ona C1 R occ S1 ah c DQ 694 2 P om D 1 U27 ery us 2 n hA Y 16 eud aSC 591 thro RH na c A 694 Ps inos b bph Rh pol A1 hA AY 392 rug phA1 A1 X odo isB na Ac 96 P ae as D173 800 cocc D2 ah AJ4 49 omon 19 P 41 R usM n Ac 4480 olar bph seudo glob 5 nah AF 932 P onas A U47 mon erul hAc 194 larom 1-17 637 C asKK us na c AY 74 Po lderiaS bphA1 testost S102 nahA 1674 Burkho M83673 eroniB Ac DQ 448048 OD-5b P pseud -356 nag hAc AF holderiaS bphA oalcalige na 8053 Burk 1 M86348 Bur nes hAc AF44 kholderiaLB na alstoniaU2 d 400 Ac AF036940 R bdCa AB121977 Xanth nag obacter tAc U62430 BurkholderiaRASC dxnA1 X72850 dn SphingomonasRW1 ntdAc U49504 PseudomonasJS42 5 Sphingopyxis thnA1 AF15756 dntAc AF169302 erDBF63 B cepaciaR34 84235 Terrabact nbzAc AF phtA1 AB0 keyseri 379638 Coma rthrobacter R1 nahAc AB monasJS765 F331043 A aleniiPY phn 066442 R phtAa A vanba 2038 Ac AY5 alstoniaN 65117 M CIMB1 N phnA 68278 I1 Aa AY3 ccusN us1B c AY C testo pht doco cocc us nahA 36778 steron 3 Rho odo opac p c AY 8 De i 08266 2 Rh 1 R 0 ahA 194 lftia a a AF 0161 688 P20 p c AF 931 cidov narA AJ4 Q84 cus 24 nd ahA 252 C tes orans arAa a D coc usI n oC2 c A 550 tost n arA odo occ 00 ah A B0 C eron n Rh oc sP4 4 n Ac F00 040 testo eCJ2 24 hod cu N1 ah 1 A 42 59 ste 924 5 R oc K s na Ac F3 83 P p roni AY3 90 doc sT cu 1 n hA A 06 P f utid Aa 121 ho acu pa PY 8 a c F2 44 luo aO nar AF R op o m H LS n hA A 95 1 P res US A 23 R R riu CZ J 01 ah c F3 03 p ce 82 nid 24 1 33 te m m 1 A A 06 3 uti nsA 39 67 06 ac iu iu H 65 n 3 F3 4 Ma da T AY 06 11 b er er Z S S ah D 0 20 ri 5I TC a 2 B co ct ct C C na A 84 64 P no IAN 17 arA AB A y a a m um 1 n h c 1 3 b ba H 48 n A A M b b u ri M P 1 n a A AF 4 0 al ct 3 id od 5 co co ilv e m H B a h c 6 P ea er n d 74 y y g ct iu M B S n h n A A 0 P p ri N 4 M M a r M 5 d a a c Y 3 a ut ca CE 9 1 M b te m m 6 1 h A h A 1 95 e id L 3 J4 4 98 2 o c iu u K S H 9 p b A c A 9 3 ru a S 1 9 r v 1 f F g 3 4 2 A 7 00 94 yc a e il m m n 1 phnAh A c A c 3 68 3 in II 01 1 3 b t g iu A 2 Y n F A 0 P IA A 5 33 7 M o c r iu mC R 2 p p A B A 0 6 2 o 2 o Q 3 c a r F H P p p p h F F 4 7 s s d Y 5 4 y M e u Y H K p h 0 5 0 t a p D b t e i e C p h h h n c 5 2 2 P u P A Q 0 t 6 n n 3 t c s P B 1 h 5 a M o 7 C p d 7 z A 9 H 1 n n A c 4 9 7 D c I p h n A 3 s K A a p A e e c d 6 4 n m p a n t ter m L p 5 i d 2 a m h b A A 9 3 7 P u r 1 i y b G h n A c A Y n 4 3 8 i A c e u m C h h c 0 5 9 AN 0 u 4 h A t 7 3 t b Y h n d 5 o u i i n c c 3 s z i 1 M 8 a g i H n A c A A Y 3 u 1 ya 6 o r F n A 5 t e 9 N F c r i r KP7 n

r n 5 c u n 4 R e A A 6 6 0 H s H A 3 T r 1 A 7 y c e r s A c c A Y N t 5 e e s A A i 3 e t R 1 A c 7 z 6 0 3 y ob t Y a h A Y n

e t A c A 67785 P 3 F H ku B Y 3 C e 3 A 7 M c b c t e a c A Y A 7 e a n i c

c r Y F

c p m c n s s d Y A M s c A 1 6 8 r r s Y 3 p a i 9 y a c d n 0 i 0 a A F r e L sP a i e o A Y 1 A k o u

n P u 5 7 6 h p 7 A 6 a 0 5 i b a t i o a s A B 1 2 u S A b A 3 t h 1 r n b o i n Y 5 7 b A 1 0 o i c m e c F 1 4 7 d h M d b n 4 B

n o i Y t m 5 B i e Y 0 8 a d o n Y 4 8 h a

0 9 3 M m t 1 3 7 o o c n i u 5 N c o d a 0 9 u n a o i A o i 6 4 c o 3 5 d c r 1 3 0 7 B a v n 6 y

y e s 1 5 8 y o r 5 c b l 1 4 4 r t v 3 e sce 6 3 l o b 3 y g u 6 3 k A 4 a 5 6 4 ur e t o

5 8 r y m a B g o 5 r 5 4 3 e M M a a 1 h d Y 5 87 f l 6 6 r a r S o c n g 4 i r 4 4 D r a M o v 4 u k n i 3 6 B

5 4 o ns F M ve a c H n A 3 r i i 7 2 7 A o o g 4 1 n n 3 r hol l B 3 M h g 3 0 2 6 B m a b 3 u A c d 0 6 l 3 e h o 5 e l k s n 6

4 a A i l N 6 B B c F f p n i 6 5 r h e 8 Q n t 5 o d 5 3 8 4 T i 1 B r 5 s k

A c a der 6 i 3 p b u o r n 1 h R 4 2 D 9 7 7 4 a S a 9 h i o 6 h l 3 n Sp y B p h a a 3 M 0 B a r i l 4 5 v p d d 4 9 2 g g o B 2 p a h k C s B A m e 2 B

8 3 m s i 2 C p 1 9 1 s u l h e ld e E 9 3 7 5 p l aC

d a e s

4 S u n p 6 s F 5 o ri n o i H 4 M g r o Q 1 4 7 6 1 1

p h e r 6 n ni 5 g t t a i A 8 6 r k e a J v h r l - 2 3 1 6 1 l o r h3 D F h e a d a

F Q 8 a z 4 4 l h il E 1 2 e s i 1 o a e 8 A 2 5 n a t e

A e n z i lu A 7 D 3 E t o h A 1 9 n t i C 4 9 h i f c r d 5 S i N n a 1

h 2 2 a i 6 f l n i m a d i A 1 i 5 Q 1 e d u a h A 3 a z - n AB0 a P e 8 e

A 1 0 i d 7 1 6 i n d F i e u m H D 1 Q 3 b i c i D 2 Q z i i

o 1 n J 1 - A 1 s m n H A A r g B a 6 o 1 n i d 0 D i 5 H n i

3 o D B d h 3 r u l A J a 1 g H T i A p 3 g s i l 9

h o Q p 5 g m A R A u a 1

d A A 2 1 g 1 i p b 7 d 4 D n A m t n P H 1 4 n 1 e p a 3 0 h h 0 0

3 4 H H g

1 A p F p Q 0 A 0 1 A n g g A 7

D 1 d 6 i n h 8 1 f

p 1 n h c 1

6 p A A

n h

h p

A p b

p b

Figure I-4. Phylogenetic relationships of the α subunit of aromatic ring- hydroxylating dioxygenases, grouped by primary substrate type. The groups are: low molecular weight PAHs (e.g. naphthalene, phenanthrene) (○), high molecular weight PAHs (e.g. pyrene) (●), biphenyls (▼) and other aromatics (e.g. phthalate, carbazole, benzoate etc.) (∆). Each branch lists the gene name, GenBank accession number and organism name.

Naphthalene dioxygenases are the best studied and comprise the most diverse group of PAH dioxygenases. Naphthalene has a higher solubility and smaller structure compared to other PAHs and is therefore most amenable to biodegradation. The observed diversity in naphthalene dioxygenases compared to other PAH dioxygenases is due to several intertwined factors: First, because of the higher water solubility of naphthalene, it is a more bioavailable carbon source. This means there is likely a greater diversity of microorganisms exposed to it in the environment, providing increased selection for evolution naphthalene degradation pathways. Second, the smaller size and reduced structural complexity of naphthalene means that the metabolic requirements, including number of enzymes and genes, are less than for larger PAHs. As a result, naphthalene degradation operons are likely more easily maintained by organisms that acquire them (via evolution or horizontal gene transfer). Third, it should be noted that the observed diversity of naphthalene dioxygenases can also be partly attributed to their presence in easily culturable organisms. Traditionally, organisms such as naphthalene-degrading Pseudomonads have been successfully isolated in liquid culture or on agar (e.g. spray plate methods). Recent evidence, however, suggests that isolates acquired using a hydrophobic surface or biofilm may be more representative of environmental PAH-degrading organisms (Bastiaens et al. 2000, Stach & Burns 2002). These approaches favour the isolation of gram positives and other organisms that have the ability to access hydrophobic substrates such as PAHs.

Dioxygenases that have only been discovered and characterized more recently, such as pyrene dioxygenase nidA, show much less diversity. This likely partly attributed to methodological limitations: generally nidA sequences recovered from the environment are done so using molecular probes based on known nidA sequences, and will therefore be similar in sequence. There is evidence that there is a great diversity of initial dioxygenases that has yet to be revealed (Taylor et al. 2002, Ní Chadhain et al. 2006, Gomes et al. 2007). Fortunately, the advent of

genomic, metagenomic and proteomic approaches circumvent this problem: mass sequencing and subsequent searching for dioxygenase motifs or activities should continue to reveal a diversity of PAH catabolism genes not currently represented in our cultures or sequence databases (Eyers et al. 2004, Ono et al. 2007).

2.4 Ecological insight from dioxygenase-based studies

Because of the substrate-specificity of bacterial ring hydroxylating dioxygenases, these enzymes, and the genes encoding them, can be used to monitor the PAH- degradation capabilities of microbes and microbial communities. Indeed many researchers have used PAH dioxygenases or dioxygenase genes to provide functional insight into the role of both indigenous and artificial microbial communities. Detection and/or quantification of dioxygenases (proteins, genes or mRNA) indicate that the microbial communities are adapted to biodegradation of those particular compounds targeted by the dioxygenases (i.e. there is, or was, bioavailable substrate) and are an important marker of biodegradative function. Early DNA-based studies employed DNA:DNA hybridization as a means to detect catabolic genes and plasmids in PAH-degrading isolates and nucleic acids (Sayler et al. 1985, Stapleton & Sayler 1998, Stapleton et al. 2000a), and could relate PAH catabolism genotypes to activity and exposure (Sayler et al. 1985). As molecular methods have advanced, so too has the ability to probe PAH-degrading communities to answer questions about specific substrate pathways. Molecular detection of catabolic genes has revealed much about the ecology of PAH- degrading organisms, and have ultimately contributed to a better understanding of natural attenuation processes.

Probing for catabolic genes as indicators of microbial community function has been employed in areas other than PAH degradation: For example, microbial communities have been examined via quantification of genes for alkane monooxygenase (Powell et al. 2006), benzylsuccinate synthase (Beller et al.

2002), atrazine catabolic operon (Devers et al. 2004) and other degradation genes (Watanabe & Hamamura 2003).

The vast diversity of PAH dioxygenases almost entirely precludes the use of a single molecular probe or primer set for the detection the whole range of dioxygenases. Estimates of diversity of dioxygenases within a community have generally employed multiple primer sets (e.g. Baldwin et al. 2003). Ní Chadhain et al. (2006) is one of the only to have used a single primer set that was successful in amplifying a broad diversity of dioxygenases, albeit with a very small amplicon: they designed a single primer set to target 78 base pairs of the conserved Rieske motif (Ní Chadhain et al. 2006). For the most part, studies on dioxygenase genes have targeted gene families (e.g. γ- or β-proteobacteria) in which the PAH-dioxygenase genes are conserved enough to reliably design molecular probes.

Naphthalene dioxygenases, particularly the NahAB dioxygenases in γ- proteobacteria, are arguably the best-studied class of PAH dioxygenases. Naphthalene dioxygenases are upregulated in the presence of substrate (naphthalene or phenanthrene) (Schell 1985, King et al. 1990, Marlowe et al. 2002) and correlate to naphthalene degradation activity (Sanseverino et al. 1993b) making these dioxygenases a useful biomarker of naphthalene biodegradation. In addition, correlations between naphthalene dioxygenase mRNA and DNA gene copies show that the presence of the genes in an organism or community is indicative of activity (Fleming et al. 1993, Sanseverino et al. 1993b). Naphthalene dioxygenase genes, particularly Pseudomonas nahAc, have been used extensively as a biomarker for biodegradation (Sanseverino et al. 1993b, Langworthy et al. 1998, Stapleton & Sayler 1998, Ahn et al. 1999, Hamann et al. 1999, Wilson et al. 1999, Baldwin et al. 2003, Gomes et al. 2007).

Dioxygenases genes have also been used to provide insight into the dynamics and roles of other PAH-degrading organisms, such as the phenanthrene dioxygenase phnAc from Burkholderia (Laurie & Lloyd-Jones 2000), nagAc from β- proteobacteria such as Ralstonia (Dionisi et al. 2004), and pyrene dioxygenase nidA from Mycobacterium (Margesin et al. 2003, Hall et al. 2005).

Some studies have taken into consideration the fact that PAHs are a mixed group of chemicals, and that it is a mixed community adapted to biodegradation of these chemicals. In order to understand the dynamics and relative roles of different functional groups within a community in a mixed system, these studies have targeted multiple PAH-catabolism genes (Langworthy et al. 1998, Laurie & Lloyd-Jones 2000, Ringelberg et al. 2001, Baldwin et al. 2003). Monitoring of multiple catabolic genes is also the basis of functional gene arrays, aimed at profiling community shifts with hundreds (thousands) of biodegradation gene targets (Rhee et al. 2004). A comprehensive approach involving multiple gene targets has been employed in all four sections of this work, and has provided valuable insight as to the relative roles of three PAH dioxygenase genotypes in PAH-biodegrading communities.

3. PAH-Degrading Mycobacteria

3.1 Identification of pyrene-degrading organisms – Mycobacterium

In contrast to low molecular weight PAHs such as naphthalene and phenanthrene, high molecular weight PAHs (4 or more rings) are much more resistant to biodegradation and thus tend to be much more recalcitrant in the environment (Herbes & Schwall 1978, Bossert & Bartha 1986, Cerniglia 1992). This is of concern as these larger PAHs tend to have greater toxic, mutagenic and carcinogenic effects. The environmental recalcitrance of high molecular weight PAHs is largely due to very low aqueous solubility (as aforementioned, solubility decreases with increasing molecular weight) reducing the bioavailability of these compounds.

The first organism isolated that could degrade and use pyrene as a sole carbon source was isolated by C. Cerniglia, and belonged to the genus Mycobacterium (Heitkamp & Cerniglia 1988, Heitkamp et al. 1988a). Formerly strain PYR-1, now reclassified as M. vanbaalenii (Khan et al. 2002), this strain can mineralize pyrene, fluoranthene, phenanthrene, anthracene and benzo[a]pyrene (Kelley et al. 1993, Moody et al. 2001, Moody et al. 2004). Other Mycobacterium stains have been isolated from a variety of soils and sediments which can also use pyrene as a sole carbon source (Table I-2). Unlike many of the opportunistic pathogenic Mycobacteria common in the environment (Primm et al. 2004, Thorel et al. 2004), these pyrene-degrading Mycobacteria are fast-growing and are phylogenetically and biochemically distinct from slow-growing pathogens (Stahl & Urbance 1990, Adekambi & Drancourt 2004). There is also evidence that genera other than Mycobacterium can degrade pyrene, including Rhodococcus sp. (Walter et al. 1991), Burkholderia cepacia (Juhasz et al. 1997), and Stenotrophomonas maltophilia (Juhasz et al. 2000). Stable isotope probing of pyrene degrading

Table I-2. Pyrene degrading Mycobacterium previously isolated. NR=Not reported.

Mycobacterium Location Contamination Reference strain profile PYR-1 Redfish Bay, Texas Oil (Heitkamp & Cerniglia 1988) Mycobacterium sp. Former coal gasification NR (Grosser et al. 1991) site, Illinois M. flavescens Grand Calumet River, Pyrene-enriched (Dean-Ross & Indiana Cerniglia 1996) RJGII-135 Former coal-gasification NR (Schneider et al. site 1996) CH1 Great Lakes PAH- (Churchill et al. Environmental Research contaminated, 1999) Laboratories, Michigan pyrene-enriched LB501T NR NR (Bastiaens et al. 2000) JLS, KMS, MCS Champion International Creosote (Miller et al. 2004) Superfund Site, Libby, Montana S65 Sept-Iles, Quebec Jet fuel (Sho et al. 2004) PYR100, PYR11, National Natural Park, None (Kim et al. 2005b) PYR300, PYR200, Schwäbische Alb, PYR103, PYR212, Germany PYR210, PYR211, PYR110, PYR102, PYR213 PYR400 Former coal gasification NR (Kim et al. 2005b) plant site, Germany 6PY-1, M. Lake sediment, French Pyrene-enriched (Jouanneau et al. austroafricanum Alps 2005) CK1, HF, HH1, Mangrove sediments, NR (Zhou et al. 2006) HH2, HH3, HH4, Hong Kong, China HH5, HH6, MC, MD, MPHC, TE, TI, TM, TU, FLO SNP11 Grass-land soil, France NR (Pagnout et al. 2007) 40 isolates Elizabeth River Oil; PAH- (Hilyard et al. 2008) sediments, Norfolk, enrichment Virginia

populations has indicated that there may also be some uncultivated γ- and β- proteobacteria that can degrade pyrene (Singleton et al. 2006).

3.2 Distribution and ecology

Pyrene-degrading Mycobacteria have been isolated from a variety of contaminated and uncontaminated soils and sediments implicating a broad distribution (Table I-2). This is consistent with observations of wide biogeographical distributions of other aromatic-degrading organisms previously discussed. Fast growing Mycobacteria have also been isolated or identified which can degrade compounds than PAHs, such as toluene (Tay et al. 1998) and methyl tert-butyl ether (MTBE) (Francois et al. 2002, Ferreira et al. 2006).

The ability of Mycobacteria to access poorly bioavailable pyrene from the environment is enhanced by several key adaptations. First, Mycobacterial cell walls are highly hydrophobic and undergo hydrophobicity changes in the presence of PAHs: these properties translate to an increased affinity for PAHs which may aid in diffusion of pyrene into the cell (Wick et al. 2002b, Wick et al. 2003, MacLeod & Daugulis 2005). Second, studies have demonstrated that Mycobacterium spp. form biofilms on the surface of PAHs (Wick et al. 2002a), the attachment aided by the hydrophobic mycolic acids of the cell wall (Bendinger et al. 1993, Bastiaens et al. 2000). Attachment to a PAH surface has been shown to increase with decreasing aqueous solubility (Johnsen & Karlson 2004), implicating biofilm formation as a mechanism to increase bioavailability of higher molecular weight PAHs. In addition, there is evidence that Mycobacteria may have a selective advantage over other organisms because of long term persistence in the environment (Tay et al. 2001).

3.3 Pyrene dioxygenase genes: nidA and nidA3

Since the initial isolation and characterization of pyrene-degrading Mycobacteria, the pyrene degradation pathways have been elucidated both genetically and biochemically (Figure I-5) (Kim et al. 2007). Pyrene biodegradation in Mycobacteria is initiated by a pyrene ring-hydroxylating dioxygenase, NidAB, which is encoded by the genes nidA (α-subunit) and nidB (β-subunit) (Khan et al. 2001, Brezna et al. 2003, Kim et al. 2007). NidAB amino acid sequences are only 40-56% similar to PhdAB and NarAaAb from Nocardioides sp. strain KP7 and Rhodococcus sp. strain NCIMB12038 respectively, two other PAH-degrading gram positive organisms (Habe & Omori 2003). In the genome of Mycobacterium, nidB is upstream of nidA, an arrangement that is unique for dioxygenase subunit genes: in the majority of characterized PAH-degraders, the dioxygenase genes are arranged with the large subunit gene first.

In addition to this biodegradation pathway, a secondary pyrene detoxification pathway has been indentified, initiated by an alternate ring-hydroxylating enzyme, NidA3B3 (Kim et al. 2007). It is up-regulated in the presence of pyrene, and catalyzes the conversion of naphthalene, anthracene and benz[a]anthracene to their respective cis-diols. NidA3, the α subunit, has about 52% sequence homology to NidA, and 61.9% to a dioxygenase subunit from Nocardioides sp. strain KP7 (Kim et al. 2006).

Figure I-5. Pyrene degradation pathway in Mycobacterium vanbaalenii (Kim et al. 2007). The metabolic intermediates are: P1, pyrene cis-4,5-dihydrodiol; P2, 4,5- dihydroxypyrene; P3, phenanthrene-4,5-dicarboxylate; P4, phenanthrene-4- carboxylate; P5, cis-3,4-dihydroxyphenanthrene-4-carboxylate; P6, 3,4- dihydroxyphenanthrene; P7, 2-hydroxy-2H-benzo[h]chromene-2-carboxylate; P8, 1-hydroxy-2-naphthaldehyde; P9, 1-hydroxy-2-naphthoate; P10, trans-2'- carboxybenzalpyruvate; P11, 2-carboxylbenzaldehyde; P12, phthalate; P13, phthalate 3,4-dihydrodiol; P14, 3,4-dihydroxyphthalate; P15, proto- catechuate; P16, ß-carboxy-cis,cis-muconate; P17, γ-carboxymuconolactone; P18, ß- ketoadipate enol-lactone; P19, ß-ketoadipate; P20, ß-ketoadipyl-CoA; P21, pyrene cis-1,2-dihydrodiol; P22, 1,2-dihydroxypyrene; P23, 1-methoxy-2- hydroxypyrene; P24, 1-hydroxy-2-methoxypyrene; P25, 1,2-dimethoxypyrene.

4. Methods in Molecular Microbial Ecology

From Winogradsky’s column and Beijerinck’s selective plating to the current metagenomic and metaproteomic sequencing, the focus of microbial ecology has been linking phylogenetic identities with functional roles within complex microbial communities. The advent of molecular techniques has revolutionized the field of microbial ecology by allowing researchers to answer questions about microbial identities and activities in complex communities without the prerequisite cultivation and maintenance in the laboratory. Reviewed here are a few of the techniques pertinent to this dissertation work.

4.1 Extracting DNA from environmental samples

The advances in molecular microbial ecology were made possible by the development of efficient methods for extracting nucleic acids from environmental samples. As the majority of microorganisms cannot be cultured in a laboratory setting (Torsvik et al. 1990), acquiring DNA from an environmental sample can provide a wealth of phylogenetic and functional information in a culture- independent manner.

Methods to extract DNA from environmental samples aim to find the appropriate balance between being rigorous enough to lyse all organisms, while not being so harsh that the resulting DNA is too fragmented and degraded to provide useful information. Sediment and soil samples present additional challenges: DNA or organisms can be sorbed to clay and organic particles and can be resistant to separation (Ogram et al. 1988, Frostegard et al. 1999); humic acids and other organics can be extracted along with the nucleic acids which can present problems for downstream applications e.g. inhibition of Taq polymerase during PCR (Tsai & Olson 1992). There are two approaches generally used when extracting DNA 31

and RNA from soils and sediments: The original indirect method involved initially separating organisms from the soil particles through fractional centrifugation, and then lysing the cells and purifying the DNA (Torsvik 1980). In contrast, the direct method does not attempt to first separate the organisms, but instead subjects the whole sample to in situ lysing, then removes the soil particles through centrifugation and purifies the DNA (Ogram et al. 1987). In general, the direct method has been found to be more effective at retrieving DNA from a wide variety of organisms, though it is slightly harsher and results in smaller fragments of DNA. For this work, the FastDNA Spin Kit for Soil (Qbiogene, MP Biomedicals, Solon, OH) was used, which follows a direct DNA isolation method, and has previously been evaluated for extraction efficiency (Burgmann et al. 2001, Martin-Laurent et al. 2001, Dionisi et al. 2003, Dionisi et al. 2004, Mumy & Findlay 2004). The isolation is initiated by bead-beating the whole sample to lyse the cells. Bead-beating has been found to be more efficient at lysing cells than chemical lysis alone, and results in higher quality DNA than when sonication is used (Krsek & Wellington 1999). This is followed by removal of the sediment particles through centrifugation and column purification of the DNA.

4.2 Real-time PCR

The development of the polymerase chain reaction (PCR) (Mullis & Faloona 1987) revolutionized molecular biology. Initially developed as a method for producing many copies of a desired gene or DNA fragment, PCR is now also used to determine starting quantities of a given template: quantitative PCR methods take advantage of the exponential nature of the target gene amplification during PCR. Real time quantitative PCR employs a detectable signal which can be measured during each cycle as an indicator of double-stranded DNA quantity. The increase in signal corresponds to exponential increase in template concentration – the cycle at which this signal crosses a determined threshold (i.e.

32 threshold cycle or Ct) is directly proportional to the amount of template in the starting sample. By comparing the threshold cycle to a standard curve comprised of known copy numbers of the template DNA, one can ascertain the number of copies in the original sample.

There are several detection chemistries that can be used for real-time PCR. The most inexpensive is SYBR Green I which emits fluorescence upon binding to double stranded DNA. This approach can be used with any set of primers, however is susceptible to inaccuracy because it will bind any nonspecific PCR products that are formed (e.g. primer dimers). Probe-based quantitative PCR methods, such as molecular beacons or dual labeled probes, use a probe that anneals at a location internal to the primers and is specific to the template sequence, provided an added level of target specificity. The basic method of hydrolysis-based (TaqMan probe) real time PCR is displayed in Figure I-6. Internal probes are designed with a 5’ fluorophore, e.g. FAM (6- carboxyfluorescein), and a 3’ quencher dye, e.g. TAMRA (6- carboxytetramethylrhodamine) which absorbs light at the same wavelength emitted by the 5’ fluorophore. While intact, these probes do not emit light because of fluorescence resonance energy transfer (FRET) from the 5’ flurorophore to the 3’ quencher molecule. During the annealing step of PCR, the probes anneal to the target gene between the two primers. Typically, these primers are designed to produce a relatively short amplicon (100-200 base pairs) to reduce possible errors associated with longer amplicons. During elongation, the 5’ to 3’ exonuclease activity of Taq polymerase degrades the probe, releasing the fluorophore from proximity to the quencher, eliminating FRET, and emitting a detectable light signal. Molecular beacons work in a similar fashion: the probes are longer and designed to form a hairpin loop, bringing the two ends (with the fluorophore and the quencher) into proximity. Upon denaturation and annealing to the template, these dyes are separated, allowing a detectable light signal to be produced (Giulietti et al. 2001).

Figure I-6. Real-time PCR using a dual-labeled probe.

Real-time PCR data is quantitative because of the inclusion of internal standards consisting of known concentrations of the target gene. Most often, these standards are the target gene cloned into a plasmid, allowing for production of large quantities of the gene and long term storage. A dilution series (e.g. 101 – 108 copies/reaction) is done alongside samples. After the PCR is complete, the threshold is set at a fluorescence level above the background noise, but still within the exponential amplification stage of each sample i.e. before the template amounts (and thus fluorescence signal) starts to plateau (Figure I-7). The threshold cycles for the internal standards are plotted to create a standard curve against which all the samples are then compared.

Figure I-7. An example of real time PCR data output (left) and standard curve (right). The threshold (dashed line) is set above background noise, within exponential increase. The cycle at which each signal crosses the threshold is termed the threshold cycle and is proportional to the amount of original template.

4.3 16S rDNA Methods

Historically, the microscopic size and morphological simplicity of bacteria and archaea has meant that many of the classification approaches used on higher organisms, for example, species concepts and morphology-based phylogenetic assignment, had little applicability to microbial communities. Now the use of the small subunit RNA sequence as phylogenetic marker has revolutionized how ecologists classify organisms and rewritten the tree of life (Woese et al. 1990). Because of their highly conserved nature, the small subunit ribosomal DNA sequences (16S for prokaryotes and 18S for eukaryotes) have become the standard phylogenetic marker for microorganisms. Recognized early as the way to elucidate evolutionary relationships (Lane et al. 1985, Weisburg et al. 1991), the current Ribosomal Database Project (RDP v.9) now contains almost half a 35

million 16S ribosomal sequences (Cole et al. 2007). The molecular phylogenetic approach has been particularly useful in microbial ecology, lending insight into evolutionary relationships between microorganisms and providing a glimpse at the previously underestimated vastness of microbial diversity (Pace 1997, Tiedje et al. 1999, Curtis et al. 2002).

Ribosomes, integral to the translation of proteins in the cell, are highly conserved, which is why the small subunit RNA has been selected as a standard phylogenetic marker. The gene encoding the ribosomal RNA is approximately 1520 nucleotides long and contains nine hypervariable (V) regions (Neefs et al. 1993). “Universal” PCR primers used to amplify the 16S genes target conserved regions of the gene, and the variable regions provide taxonomic resolution between organisms. In general, 16S amplification methods that span several of the variable regions are able to resolve differences at a finer taxonomic level. For applications such as real time PCR or T-RFLPs which demand shorter amplicon lengths, selection of variable region can influence the measured genetic diversity (Schmalenberger et al. 2001).

4.4 PCR bias

One challenge in using whole community 16S methods, such as clone libraries or phylogenetic fingerprinting, is that most of them rely on an initial PCR step. This step can potentially introduce biases when using mixed community DNA as a template. These amplification biases can result in inaccurate estimation of population abundances and can misinform estimates of diversity (Frey et al. 2006). It is important to recognize the possibility of PCR bias so that steps may be taken to minimize it.

Differential amplification of mixed templates has been observed, and may be due to several phenomena. Nonspecific priming (Reysenbach et al. 1992), and PCR drift due to random events early in PCR (Wagner et al. 1994) have been shown to result in differential amplification. The location of the gene can also influence PCR amplification because of interference from flanking regions (Hansen et al. 1998) or differing accessibility within the genome (Farrelly et al. 1995). Differences in priming due to variable energies of degenerate primers based on GC content are also purported to lead to differential amplification (Wagner et al. 1994, Polz & Cavanaugh 1998), however this was not the case for moderately degenerate primers (1-2 degenerate sites) when tested on a model community (Hartmann & Widmer 2008). Choice of variable regions can also lead to biased results in mixed template amplification (Schmalenberger et al. 2001). In addition, preferential amplification of rarer sequences can arise from faster reannealing of predominant templates during PCR, resulting in a shift towards equal amounts of templates in later cycles (Mathieu-Daude et al. 1996, Suzuki & Giovannoni 1996, Kurata et al. 2004).

In addition to problems that can influence abundance ratios of templates and result in inaccurate representations of original template proportions, there can also be sequence artifacts introduced during amplification which may inaccurately represent the actual sequences. For example, PCR can generate chimeric sequences or heteroduplexes (Kopczynski et al. 1994, Wang & Wang 1997, Qiu et al. 2001, Thompson et al. 2002, Hugenholtz & Huber 2003, Ashelford et al. 2005), make errors in nucleotide incorporation (Cariello et al. 1991, Acinas et al. 2005), and/or produce single stranded amplicons which can adversely influence downstream applications (Egert & Friedrich 2003).

4.5 Community fingerprinting: T-RFLPs

Sequencing 16S ribosomal genes from the environment provides valuable information about the members of the microbial community. Microbial ecology, however, is often interested in comparing communities, and developing clone libraries and sequencing multiple communities can become unwieldy and expensive using current technology. As a result, whole community phylogenetic fingerprinting methods have been developed that take advantage of patterns generated by differences in 16S ribosomal DNA sequences. These methods, such as terminal restriction fragment length polymorphisms (T-RFLPs), denaturing gradient gel electrophoresis (DGGE), automated ribosomal intergenic spacer analysis (ARISA), amplified ribosomal DNA restriction analysis (ARDRA), single strand conformation polymorphism (SSCP) and amplicon length heterogeneity PCR (LH-PCR), involve analyzing the whole community in a single sample. These approaches are inexpensive compared with sequencing individual members of every community, allowing for a more practical approach for comparing multiple communities. In addition, these approaches offer more information about the structure of a community than traditional diversity estimates (Hartmann & Widmer 2006). While each technique has its own set of limitations and advantages, comparisons between them have indicated that there is little difference in terms of resolving overall microbial community structure patterns (Casamayor et al. 2002, Hartmann et al. 2005).

Terminal restriction fragment length polymorphisms (T-RFLP) is a high throughput, high resolution, culture-independent method for comparing microbial communities (Liu et al. 1997, Osborn et al. 2000, Marsh 2005). One of the main advantages of T-RFLPs over other fingerprinting methods is that the data are collected using a capillary electrophoresis instrument – this makes the technique very repeatable, and outputs the data in digital format which is very amenable to downstream multivariate statistical analyses.

The basic method for T-RFLPs is diagrammed in Figure I-8, which illustrates a simplified three-species community. Microbial community DNA is first extracted from an environmental sample. 16S ribosomal genes are then amplified using universal eubacterial primers, one (or both) of which are labeled with a fluorescent dye such as carboxyfluorescein (FAM). The PCR is stopped while still in the exponential phase (around 25 cycles) to reduce potential PCR bias: stopping the reaction at an earlier cycle results in PCR products in proportions which better reflect the original community (Mathieu-Daude et al. 1996, Suzuki & Giovannoni 1996, Hartmann & Widmer 2008). The resulting pool of mixed 16S PCR products, all labeled at one (or both) ends, are then column-purified to remove any polymerase which could continue lengthening fragments during the digestion step (Hartmann et al. 2007). The pool of products is then subject to restriction enzyme digestion using enzymes that are frequent cutters, typically with four base pair recognition sites, such as Alu I and Hha I. Using a single restriction enzyme can miss much of the diversity, therefore two (or more) restriction enzymes are typically used in separate digests (Engebretson & Moyer 2003). Differences in 16S sequences (i.e. phylogenetically different members of the community) mean that the restriction enzyme recognition sites will be in different locations. The restriction digest will thus yield fragments of different lengths. After reaction cleanup to remove restriction enzymes and inclusion of an internal standard (known fragment sizes labeled with a different fluorescent dye, often Carboxy-X-rhodamine (ROX)), the fragments are separated out using capillary electrophoresis. The labeled fragments are detected and compared to the internal standard, and size (base pairs) and abundance (peak height or area) are reported for each fragment detected. The resulting information (number of peaks, sizes, and abundances) can be used as a community fingerprint for comparison purposes.

Theoretically, one can use the length of the fragment and the known primer and restriction sequences to determine a probable phylogenetic assignment for the

Figure I-8. Schematic illustrating the terminal restriction fragment length polymorphism (T-RFLP) method. This example uses a three-species community (species A, B, and C). An example electropherogram shows the standard peaks in red and the labeled restriction fragments in blue.

organism represented by a given peak. Web-based phylogenetic assignment tools for phylogenetic assignment have been developed to compare fragment sizes against sequence databases (Marsh et al. 2000, Kent et al. 2003, Saari et al. 2007), and these have been employed with some accuracy (Mills et al. 2003). Practically, however, there are several issues which reduce the reliability of these predictions. First, with T-RFLPs and other phylogenetic fingerprinting methods it is very possible that more than one organism will have the same fragment length, and thus a single peak may actually represent multiple organisms. Likewise, a single fragment length may match several organisms in the database. Second, oligonucleotide mobility in capillary electrophoresis can be influenced by sequence composition and temperature (via creation of secondary structures) resulting variable migration rates (Kaplan & Kitts 2003). Third, different fluorophores can result in varied mobility: it has been shown that FAM (the sample dye) and ROX (the internal standard dye) migrate slightly differently (Kaplan & Kitts 2003, Marsh 2005, Pandey et al. 2007). Fourth, it is possible for artifact peaks to be added because of undigested single stranded DNA (Egert & Friedrich 2003) or chimeras (Qiu et al. 2001), though this was found only to a minimal extent in T-RFLP of a model community (Hartmann & Widmer 2008). Some of these problems associated with phylogenetic assignment have been alleviated via inclusion of alternate approaches. For example Grant and Ogilvie created a 16S rDNA clone library in parallel to T-RFLP profiling, then applied T- RFLP to screen clone library to identify taxa associated with the major peaks in the T-RFLP profile (Grant & Ogilvie 2004).

There also exists other biases introduced during T-RFLPs which may influence relative abundance of ribotypes (species). PCR bias, discussed earlier, can potentially lead to differences in amplification efficiency, although evaluation of this with a model community could not pinpoint any of the specific reasons (i.e. large amplicons, degenerate primers, genomic location influences, GC content or random stochastic events) as an explanation for amplification differences in a

41 model community (Hartmann & Widmer 2008). Other confounding factors may include residual polymerase activity during restriction digest (Hartmann et al. 2007), and differences in ribosomal 16S gene copy numbers between species (Farrelly et al. 1995). These have all been reviewed previously (von Wintzingerode et al. 1997, Hartmann & Widmer 2008), and are important considerations when interpreting T-RFLP data. In particular, there is a considerable amount of debate in the literature about using T-RFLPs and other community profiling methods to calculate diversity indices (Danovaro et al. 2006, Fierer & Jackson 2006, Bent et al. 2007, Blackwood et al. 2007, Fierer 2007). While these measures may provide some information about the community, profiling methods are unlikely to give accurate diversity indices as they will undoubtedly miss rare species, either due to lack of PCR amplification, or during data analysis when small peaks representing rare species may be indistinguishable from the background noise.

Despite these challenges with absolute assignment and diversity measures, T- RFLPs are still very useful for microbial community comparison, and have been found to be reliable in detection of relative changes in community structure (Watts et al. 2001, Hartmann & Widmer 2008). T-RFLPs have been successfully used in microbial ecology for documenting temporal and spatial shifts in microbial community composition. For example, T-RFLPs have illustrated spatial differences in communities between environment types (Dunbar et al. 2000, Buckley & Schmidt 2001, Moeseneder et al. 2001, Kuske et al. 2002, LaMontagne & Holden 2003, Mummey & Stahl 2003), and changes in response to environmental gradients or perturbations, such as redox (Braker et al. 2001), nutrients (Kennedy et al. 2004, Eldridge et al. 2007), salinity (Casamayor et al. 2002), land use (Buckley & Schmidt 2001, Widmer et al. 2006), viruses (Schwalbach et al. 2004) or contaminant concentrations (Cordova-Kreylos et al. 2006, Muckian et al. 2007). T-RFLPs have also revealed temporal community shifts, for example seasonal changes (Hullar et al. 2006) or succession during

42 biodegradation (Watts et al. 2001, Mills et al. 2003, Ayala-del-Rio et al. 2004, Pesaro et al. 2004, Fedi et al. 2005). It should be noted that T-RFLPs need not be limited to phylogenetic analysis – this method has been employed with functional genes instead of 16S ribosomal genes (Braker et al. 2001).

4.6 Statistical analysis of community fingerprinting data

The power of T-RFLPs and other molecular fingerprinting methods to document changes in community structure lies in the ability to apply multivariate statistical analyses to the data generated. The data matrices resulting from these techniques are large and multivariate clustering and ordination approaches are essential to reducing them to interpretable patterns. T-RFLP profiles consist of fragment sizes (ribotypes) and estimates of peak height or peak area (abundance), comparable to species types and counts used in macroecological community studies. Data is first standardized for comparison, then clustering and ordination techniques can then be used to determine communities that are more or less similar.

Prior to standardization, replicate profiles are aligned and any peaks not occurring in all three are removed. This ensures that spurious PCR products are not included in the analysis. Standardization of T-RFLP data is then necessary to adjust for differences in amount of DNA loaded onto each capillary. There are several approaches: Fluorescence standardization involves summing the total fluorescence (either peak height or peak area) as a proxy for total DNA loaded. Peak heights in replicate profiles are then proportionally reduced so that all have the same total fluorescence (Dunbar et al. 2001). For example, if the total fluorescence for replicates A, B, and C was 10000, 8000 and 12000, all of the peak heights in A would be multiplied by 0.8, and all of the heights in C by 0.67, so that all three profiles sum to 8000. If this standardization results in peaks

dropping below a predetermined noise threshold (20-25 fluorescence units), those peaks are removed. In contrast, peak height standardization uses the same number of most abundant peaks in each profile: the number of peaks in the profile with the fewest peaks is used as the standard, and all other profiles are reduced to that number (Kuske et al. 2002). For example, if the profile that has the fewest peaks has 100 peaks, for all of the other profiles, the 100 most abundant peaks are selected for analysis and the rest removed. These standardization methods inevitably result in rare phylotypes dropping out and therefore represent a conservative approach: they are based on the assumption that similar communities will have the same dominant phylotypes. The presence or absence of a few rare phylotypes should have limited impact on overall community similarity and dissimilarity analyses. Once standardized, replicate profiles are averaged, and data is put into a matrix where each row is a sample and each column is a ribotype (terminal restriction fragment). Peaks are considered the same ribotype if they are within 0.5 base pairs. This matrix can then be used as input for clustering and ordination statistics.

Hierarchical clustering measures are a means of arranging communities based on their differences to other communities, based on a calculated distance measure. Data are first transformed to similarity measure, such as Jaccard distances, which are based on presence or absence of ribotypes (“species”), where a sample is given a value of 1 for ribotypes present, and 0 for those that are not. Others take into account differences in abundances, using either peak height or peak area as an indication of quantity of a particular ribotype, and factors that into the calculation of distance. The data is then clustered in an agglomerative hierarchical fashion, that is, nested groupings are made based on linkage rules from most to least similarity. The clusters are typically visualized using a dendogram (horizontal tree). T-RFLP data has been previously analyzed using clustering based Jaccard distances (Dunbar et al. 2001, Kuske et al. 2002),

44 however these approaches are limited in that they are primarily exploratory, not hypothesis-driven (Ramette 2007).

There are a variety of more informative ordination techniques commonly used in analyzing community fingerprinting data. Linear approaches include principle components analysis (PCA) (unconstrained) and redundancy analysis (RDA) (constrained). Unimodal approaches include correspondence analysis (CA) (unconstrained) and canonical correspondence analysis (CCA) (constrained). It should also be noted that there are also some techniques which make no assumptions regarding linear or unimodal distributions, such as nonmetric multidimensional scaling (NMDS).

Linear ordination approaches, such as PCA and RDA, are based on the assumption that a species’ abundance will change linearly along an environmental gradient, and ordination in multivariate space is based on Euclidean distances between samples. PCA is unconstrained and therefore an exploratory approach: the axes are synthesized to maximize variance explanation, and thus any inference to environmental gradients is done post hoc (indirect gradient analysis). PCA is commonly used to analyze T-RFLP data, however may not always be the most informative or appropriate technique, especially in the context of large environmental gradients (Ramette 2007).

In contrast to linear methods, unimodal ordination techniques, such as CA and CCA, assume an underlying unimodal abundance distribution along a gradient, that is, that species have optima within the environmental gradient and abundance decreases on either side. This is the distribution predicted by niche theory, and may be more ecologically relevant than a linear model. It is a fair assumption for microbial communities at PAH-contaminated sites; studies have indicated that microbial diversity tends to be highest at intermediate levels of contamination

(Langworthy et al. 2002, Andreoni et al. 2004), suggesting that individual species most likely have an optimum along a contamination gradient.

Canonical correspondence analysis (CCA) is the constrained version of CA: it is based on a unimodal model of species distribution, however the determination of axes is based on a linear combination of environmental parameters (constrained, or direct gradient analysis) (ter Braak 1986). CCA is based on χ2 distances, and maximizes the correspondence between the species and the environmental parameters. As such, this technique is not simply exploratory but hypothesis- driven, and is particularly useful for analyzing community differences in relationship to environmental factors (Palmer 1993). Unimodal ordination has the added advantage that it necessarily standardizes the samples to model relative abundances – this is good for T-RFLPs which are susceptible to variations based on DNA extraction efficiency, loading etc. In addition, unimodal models are more accurate than linear estimates for datasets with many zeros (Ramette 2007). Because of the ecologically-relevant unimodal distribution, and the constrained approach allowing for comparisons with environmental data, CCA has been frequently used in the interpretation of T-RFLP, DGGE and other fingerprinting data (Salles et al. 2004, Yannarell & Triplett 2005, Cordova-Kreylos et al. 2006, Gleeson et al. 2006, Klaus et al. 2007, Muckian et al. 2007, Sapp et al. 2007).

Another advantage of constrained ordinations such as CCA is that significance testing of the relationship between species and the environmental parameters is possible; the null hypothesis being that the species distributions (response) are independent of the environmental parameters. The statistical significance of the relationship between species distributions and environmental parameters may be tested using Monte Carlo permutation tests, a resampling method which randomly shuffles (permutes) the data set, recalculating the test statistic for each permutation. If the null hypothesis of no relationship is true, then there should be little change in the test statistic; the shuffled datasets should give the same results

as the original. The p value is estimated based on the percentage of permutations resulting in a test statistic equal to or higher than the original dataset. That is, for a significance level of α = 0.05, the test statistic calculated for the original data is considered significant if it is greater than 95% of the values generated for reshuffled data. Resampling techniques, such as permuation, bootstrap or jackknife testing, are advantagous in that they are nonparametic approaches which do not require comparison to theoretical probability distributions (Lepš & Šmilauer 2003, Shaw 2003).

Nonmetric multidimensional scaling (NMDS) is another unconstrained ordination technique. NMDS uses ranked Bray-Curtis similarity measures between samples, and is thus well suited to cases where the dataset is not multivariate normal (Shaw 2003). Unlike linear and unimodal ordination models like PCA and CCA, no assumptions are made about the relationships between variables. This has an advantage when the environment is fragmented and does not present a continuous gradient (Ramette 2007). It is one of the simplest representations of distances (dissimilarities) between communities, and has been used for the analysis of T- RFLP data (Casamayor et al. 2002, Hullar et al. 2006, Mannisto et al. 2007). NMDS iteratively optimizes goodness of fit by maximizing the correlation between Bray-Curtis scores and distance in ordination space (reduces stress) (Kruskal 1964). It is different from other ordination techniques in that the number of axes is user-selected, and ordination changes based on the subjective choice of number of axes. In addition, the first axis is not necessarily more important than the others, as it is with other ordination techniques. Like other unconstrained methods, NMDS is an exploratory technique, but can provide valuable insight into community similarities.

Ordination plots are useful in reducing complex datasets down so that groups of similar communities are more easily visualized. These groups can also, however, be statistically compared using procedures such as ANOSIM. ANOSIM is a

nonparametric technique analogous to a univariate ANOVA that compares predetermined groups of samples (e.g. treatment, date, site) to determine if there is a significant difference between their ranked distances (Clarke 1993, Clarke & Gorley 2006). ANOSIM has been used in conjunction with community profiling techniques including T-RFLPs (Bisson et al. 2007, Kent et al. 2007, Nelson & Mele 2007).

5. General Objectives and Hypotheses

Microbial metabolism of PAHs is the primary degradation fate of these persistent environmental contaminants. Despite a large body of research, there are still many questions that demand further investigation. The disconnect between laboratory and field studies has long been recognized, however accurate measurements of in situ activity has been largely dependent on the development of sensitive tools and techniques for probing communities in a culture- independent manner. This has left researchers with a less than complete picture of the evolution and dynamics of indigenous microbial communities which are responsible for natural attenuation of contaminants, and has resulted in many failed attempts at mimicking these processes for the purpose of bioremediation. Fortunately, advances in molecular approaches in microbial ecology has allowed researchers to better answer some of the questions regarding the roles and dynamics of particular populations and genotypes within a community. This dissertation focuses on pyrene-degrading Mycobacteria, a recently characterized group of organisms which have the potential for significant contributions to attenuation of high molecular weight PAHs from the environment. Despite the isolation of many strains within this group, there are still unanswered questions regarding their biogeographical distribution, evolution of PAH catabolism, and their dynamics with respect to other PAH degrading organisms. The four following chapters present studies that are united by several overarching objectives:

To determine the presence and role of pyrene-degrading Mycobacteria in contaminated sediments and to compare populations to other PAH- degradation genotypes, namely naphthalene dioxygenase genes nahAc and nagAc

To determine the distribution of these genotypes in the environment through examination across different contaminated sites (Chattanooga Creek and Lake Erie) and across different environments within a site (sediment and suspended particles in the water column)

To provide a link between the functional pyrene dioxygenase gene nidA and the phylogenetic identity of the organisms that carry this genotype through isolation and molecular characterization of pyrene-degrading organisms from both Chattanooga Creek and Lake Erie

The first study (Part II) tested the specific hypothesis that fast-growing Mycobacterium nidA genes are enriched in the coal-tar contaminated sediments of Chattanooga Creek. This study is the first to quantitatively measure copies of this gene and relate the abundances to biodegradation parameters.

The second study (Part III) addressed the general hypothesis that particle- associated bacterial communities are contributing to natural attenuation of PAHs from Chattanooga Creek. Specifically, it was hypothesized that these communities exhibit PAH biodegradation (both actual and genotypic potential) and are functionally and structurally (phylogenetically) different from sediment communities at the site.

Part IV presents the third study, which addressed the general observation that biodegradative organisms tend to have broad geographical distributions (the “everything is everywhere” hypothesis). It specifically tested the hypothesis that pyrene degrading organisms (Mycobacterium sp.) can be isolated from different locations, namely Chattanooga Creek and Lake Erie, and that the pyrene dioxygenase nidA genotype can be found in these isolates.

The final study (Part V) continues on the theme of determining geographical distributions, and tested the hypothesis that biodegradative genotypes (specifically nidA) are prevalent in Lake Erie sediments.

6. Study Areas

6.1 Chattanooga Creek

Chattanooga Creek runs through the south part of Chattanooga, Tennessee, USA, originating in Georgia and emptying into the Tennessee River. This area of Chattanooga is a mix of industry and lower income residential areas. Industrial activity in the area has rendered Chattanooga Creek contaminated with mixed organics and metals. In particular, the Tennessee Products coal carbonization (coking) plant, in operation from 1918 – 1987 deposited large volumes of waste coal tar into the creek and floodplain (Vulava et al. 2007). Because of its viscosity and hydrophobicity, coal tar is an especially recalcitrant sediment contaminant. In 1995, the extent of contamination was federally recognized and the former Tennessee Products site and a 4 km portion of the Chattanooga Creek were placed on the National Priorities List. A community-based initiative from local residents resulted in emergency clean up action: Three years later, a 1.6 km section of the Creek was dredged to the bedrock, and approximately 20,000 tons of coal tar -contaminated sediments removed. Despite this remediation effort, concentrations of PAHs in both the remediated and remaining section of the Superfund site remain high (Dionisi et al. 2004, DeBruyn et al. 2007). In addition, measurement of PAHs in the floodplain soils indicated that contaminated sediments are being moved and redistributed around the creek and floodplain (Dickerson 2005).

Three sampling sites were selected for sampling sediments along Chattanooga Creek (Figure II-1). Site W, next to the Wilson Road bridge, is upstream of the Superfund Site and represents the background or control site. Site DO is the section of creek that was excavated in 1998, and is located at the end of an access road originating at the end of E41st Street. Site CF is in the lower section of the

creek – this area was excavated in 2006-2007, however all sampling was completed prior to this and thus represents an unremediated, contaminated sampling site. CF is located at the back of the property of Crabtree Farms, located at the end of E30th Street. These three sites present a gradient of contamination levels (and PAH concentrations) from low to high moving downstream (i.e. W < DO < CF).

6.2 Lake Erie

Lake Erie, the smallest and shallowest of the Laurentian Great Lakes, is also the most highly impacted. In addition to agricultural uses, 17 major urban areas (approximately 33 million people) are also within the Lake Erie basin. Many of these sites are historic and current locations of heavy industry, including Detroit, MI, Toledo, OH, Cleveland, OH, and Buffalo, NY. In the 1950s and 1960s, increased fertilizer use in agriculture, combined with human waste and detergent runoff from cities culminated in severe eutrophication of the lake. Lake Erie’s problems were not limited to biological production – in the early 20th century, polychlorinated biphenyls (PCBs) and dichloro-diphenyl-trichloroethane (DDT) came into usage and found their way into Lake Erie sediments, often via tributaries such as the Detroit, Maumee, and Cuyahoga Rivers, which run through Detroit, Toledo and Cleveland, respectively. The Great Lakes Water Quality Agreement signed by the United States and Canada in 1972 served to reduce the phosphorus inputs and thus the eutrophication problems.

As a result of heavy industry surrounding Lake Erie, the sediments of the lake have been found to contain a mixture of organic contaminants, such as PAHs (Eadie 1984). These are entering Lake Erie via a combination of fluvial and atmospheric deposition from nearby industry, which includes coking coal plants in Detroit and Windsor which support the area’s steel mills. Isotopic patterns of PAHs confirm that the majority of these contaminants enter the lake fluvially and 53 are transported around the lake via circulation patterns (Smirnov et al. 1998). Total PAH concentrations (i.e. the sum of 16 priority PAHs) around the lake range from 0.2 – 5.3 mg/kg sediments, with highest concentrations observed near the major cities and tributaries (Detroit, Toledo, Cleveland, Buffalo) and the lowest concentrations along the northern shore, where land use is primarily agriculture and few major tributaries run (Smirnov et al. 1998). Secondary sources of PAHs include deposition of contaminated dust and precipitation (Cortes et al. 2000, Buckley et al. 2004, Sun et al. 2006).

PAHs can be toxic to higher organisms, and there is evidence that PAHs in Lake Erie are moving into the food chain. Benthic invertebrates , including oligochaetes, chironomids, dreissenid mussels, amphipods and crayfish , have been found to contain PAHs (Gewurtz et al. 2000, Marvin et al. 2000, Gewurtz et al. 2003, Pickard et al. 2006) indicating that these contaminants are bioavailable in the sediments and are likely to be transferred to higher trophic levels (Gewurtz et al. 2000). Metabolites of PAHs have been measured in fish in Lake Erie and its tributaries, with concentrations of metabolites correlating to PAH concentrations in the sediments, indicating that these contaminants are reaching and impacting higher trophic levels (Metcalfe et al. 1997, Baumann 1998, Yang & Baumann 2006).

Despite the knowledge of the presence of PAHs and their toxicological impacts, very little work has been done to determine the biodegradative capabilities of the indigenous microbial populations which may be contributing to natural attenuation in Lake Erie. Sediment microbial communities have demonstrated the potential for monoaromatic and PCB degradation (Myers et al. 1994, Hoostal et al. 2002), suggesting they may also be adapted to PAH degradation. PAH- degrading microorganisms have been isolated from Detroit River sediments (McNally et al. 1998), however have not been characterized either taxonomically or in terms of functional genotypes.

Lake Erie is undeniably a valuable freshwater resource that demands careful management and scientifically-informed policy making. Considering the integral role of microorganisms in carbon and nutrient cycling in most environments, it is surprising that our understading of indigenous Lake Erie microbial communities is lacking. Bacterioplankton communities in the water column of Lake Erie have been examined in terms of their role in nutrient and carbon cycling (Hwang & Heath 1997, DeBruyn et al. 2004, Lavrentyev et al. 2004, Gouvea et al. 2006), trace metal scavenging (Twiss & Campbell 1998), seasonal hypoxia (Wilhelm et al. 2006), and harmful algal blooms (Ouellette et al. 2006). Less work has been done on the sediment communities: the functional dynamics of sediment bacteria as it relates to DOM cycling (Hoostal & Bouzat 2008), and interactions with Dreissena mussels (Frischer et al. 2000, Lohner et al. 2007) have been examined previously.

Part II.

Comparative quantitative prevalence of Mycobacteria and functionally abundant nidA, nahAc and nagAc dioxygenase genes in coal tar contaminated sediments

This section is a lightly revised version of a paper published in the journal Environmental Science & Technology:

DeBruyn JM, Chewning CS, Sayler GS (2007) Comparative quantitative prevalence of Mycobacteria and functionally abundant nidA, nahAc, and nagAc dioxygenase genes in coal tar contaminated sediments. Environmental Science & Technology 41:5426-5432

1. Abstract

The Chattanooga Creek Superfund site is heavily contaminated with metals, pesticides, and coal tar with sediments exhibiting high concentrations of polycyclic aromatic hydrocarbons (PAHs). High molecular weight PAHs are of concern because of their toxicity and recalcitrance in the environment; as such there is great interest in microbes, such as fast-growing Mycobacterium spp., capable of degradation of these compounds. Real-time quantitative PCR assays were developed targeting multiple dioxygenase genes to assess the ecology and functional diversity of PAH-degrading communities. These assays target the Mycobacterium nidA, β-proteobacteria nagAc, and γ-proteobacteria nahAc with the specific goal of testing the hypothesis that Mycobacteria catabolic genes are enriched and may be functionally associated with high molecular weight PAH biodegradation in Chattanooga Creek. Dioxygenase gene abundances were quantitatively compared to naphthalene and pyrene mineralization, and temporal and spatial PAH concentrations. nidA abundances ranged from 5.69 x 104 - 4.92 x 106 copies per gram sediment; nagAc from 2.42 x 103 – 1.21 x 107 and nahAc from below detection to 4.01 x 106 copies per gram sediment. There was a significantly greater abundance of nidA and nagAc at sites with the greatest concentrations of PAHs. In addition, nidA and nagAc were significantly positively correlated (r = 0.76), indicating a coexistence of organisms carrying these genes. A positive relationship was also observed between nidA and nagAc and pyrene mineralization indicating that these genes serve as biomarkers for pyrene degradation. A 16S rDNA clone library of fast-growing Mycobacteria indicated that the population is very diverse and likely plays an important role in attenuation of high molecular weight PAHs from Chattanooga Creek.

2. Introduction

Microbial biodegradation is the primary degradation fate pathway for polycyclic aromatic hydrocarbons (PAHs) in soils and sediments. PAHs are toxic, mutagenic, carcinogenic and have been associated with inflammatory vascular responses (Kim et al. 2005a). Many studies focus on aerobic dioxygenase- mediated degradation of lower molecular weight PAHs, such as naphthalene and phenanthrene, which are more soluble, and thus more easily biodegraded. More recently, Mycobacterium capable of using high molecular weight (HMW) PAHs (e.g. pyrene and fluoranthene) as a sole carbon source have been isolated (Heitkamp & Cerniglia 1988, Heitkamp et al. 1988b). These PAH-degrading Mycobacteria are fast-growers, a distinct biochemical and phylogenetic taxonomic division within the genus (Stahl & Urbance 1990). Their ability to degrade HMW PAHs is partially due to the highly hydrophobic mycobacterial cell wall (Wick et al. 2003, MacLeod & Daugulis 2005) suggesting a selective advantage in obtaining PAHs from the environment. Mycobacterium spp. have been isolated and characterized from PAH-contaminated environments (Heitkamp & Cerniglia 1988, Schneider et al. 1996, Churchill et al. 1999, Bastiaens et al. 2000, Dean-Ross et al. 2002, Miller et al. 2004, Kim et al. 2005b) and in vivo biochemical pathways of pyrene degradation have been identified (Liang et al. 2006, Kim et al. 2007). The genes encoding the large (α) and small (β) subunits (nidA and nidB) of the Rieske-type dioxygenase responsible for initial pyrene dihydroxylation have been sequenced in several strains (Brezna et al. 2003).

Recent advances in molecular microbial ecology have allowed for broader analysis of biodegradative organisms in the environment. Molecular methods targeting 16S ribosomal (rDNA) genes have been effective in resolving phylogenetic diversity in PAH-degrading Mycobacterium populations (Cheung & Kinkle 2001, Leys et al. 2005b). However, 16S methods make no inference as to

the function of these communities. Identification and quantification of catabolic genes is one way to functionally characterize microbial communities. In PAH- contaminated sites, previous studies have revealed an enrichment of PAH- catabolism genes, specifically naphthalene dioxygenases. For example, γ- proteobacteria naphthalene dioxygenase genes (nahAc) have been related to naphthalene concentrations and/or naphthalene mineralization rates in the environment (Fleming et al. 1993, Sanseverino et al. 1993b, Langworthy et al. 1998). Dionisi et al. (Dionisi et al. 2004) found a positive correlation between β- proteobacteria naphthalene dioxygenase genes (nagAc) and naphthalene concentrations in Chattanooga Creek sediments. Quantification of catabolic genes can provide insight into the role of particular groups of bacteria in natural attenuation of PAHs.

The Chattanooga Creek Superfund Site (Chattanooga, TN, USA) was placed on the National Priorities list in 1995 due to mixed priority pollutants and heavy coal tar contamination; predominantly from a nearby coking coal plant (Vulava et al. 2007). This contamination has resulted in high concentrations of PAHs in the sediments (Dionisi et al. 2004). HMW PAHs are of particular concern because of their toxicity, carcinogenicity and recalcitrance due to low solubility, high thermodynamic stability and slow microbial degradation (Cerniglia 1992). Tar- rich sediments were excavated from a section of Chattanooga Creek in 1998, however despite these remediation efforts PAH concentrations have remained high (Dionisi et al. 2004). In addition, measurable concentrations of PAHs on the floodplain indicate that there is likely frequent redeposition of contaminated sediments during flood events (Dickerson 2005, Vulava et al. 2007). The unsuccessful physical remediation and dynamic nature of this system indicates that further understanding of natural attenuation of PAHs by microorganisms is still needed.

To test the hypothesis that fast-growing Mycobacterium nidA genes are enriched in contaminated sediments and are functionally associated with HMW PAH degradation in Chattanooga Creek, a quantitative real-time PCR assay was designed to quantify pyrene dioxygenase gene (nidA) abundance in indigenous microbial communities. While primers have been designed to detect this gene in isolates (Miller et al. 2004) and in soil (Hall et al. 2005), this study is the first to quantitatively measure copies of nidA and relate catabolic gene frequency to other parameters (such as mineralization rates and PAH concentrations) in order to evaluate this target as a biomarker of biodegradation. In addition, a Mycobacterial 16S clone library has been developed to assess the diversity of Mycobacteria indigenous to this contaminated site. 16S rDNA genes have been a target for other studies of fast-growing Mycobacterium (Cheung & Kinkle 2001, Leys et al. 2005b); this study presents the first Mycobacterium clone library from the Chattanooga Creek Superfund Site.

3. Methods

3.1 Sampling

Sediment samples were collected from Chattanooga Creek, a shallow, lotic system in October 2004, and April, July, and October 2005. Samples were collected from a site upstream of the Superfund Site (W), from a section that was excavated in 1998 (DO), and from an unremediated, highly contaminated site (CF) (Figure II-1) (corresponding to sites 1, 4, and 6, respectively, from Dionisi et

al. 2004). The control site (W) is upstream of the highly contaminated area, however still exhibits low levels of contamination due to heavy industry and traffic in the area. At all sites, surface sediments were fine sand, with no visible coal tar (unlike deeper sediments). Triplicate sediment samples were taken from the top 5-10 (aerobic zone), stored in Whirlpak bags, and either frozen immediately on dry ice for DNA extraction or kept at ambient temperature for mineralization assays.

3.2 PAH quantification

PAHs were extracted according to EPA method 3546 using the Microwave Automated Reaction System from CEM (Matthews, NC). Briefly, 10 g sediment were combined with Base/Neutral Surrogate Standard (UltraScientific, North

Kingstown, RI) and 5 g Na2SO4. Samples were ground to < 1 mm particle sizes, extracted twice in GreenChem vessels with 25 ml of a 1:1 hexanes:acetone mixture (HPLC grade, Fisher Scientific, Pittsburg, PA) according to manufacturer’s protocol (100°C for at least 20 minutes). The solvent was concentrated and run on Hewlett-Packard gas chromatograph (model 6890) equipped with a mass-selective detector (model 5973) and Hewlett-Packard 61

Figure II-1. Map of sampling locations on Chattanooga Creek. The Tennessee River is pictured in the top left hand corner; the Tennessee- Geoergia border is along the bottom. Zone I is upstream of point sources of contamination (sources include the Tennessee products coking coal carbonization facility, in operation from 1918-1987). Zone II sediments were excavated in 1998; Zone III is currently being excavated (2006-2007).

62 autosampler complying with Environmental Protection Agency (EPA) standard method 8270D. Sediment dry weight was determined following baking of 5 g of wet soil at >80°C for at least 48 hours.

3.3 Mineralization Assays

To determine mineralization rates, three vials (and one killed control) were used for each sample containing 2 g sediment slurried with 1 ml dH2O in a 40 ml EPA vial with a septum. As a positive control, a 1 ml overnight culture of Mycobacterium flavescens (ATCC 700033) (pyrene) or Pseudomonas fluorescens

HK44 (naphthalene) were used in place of sediment. One ml 2N H2SO4 was added to each killed control; an 8 ml vial with 0.5 ml 0.5N NaOH was placed in 14 each to serve as a CO2 trap. Either 400,000 dpm naphthalene-1- C (31.3 mCi/mmol, Sigma, St. Louis, MO) or 200,000 dpm pyrene-4,5,9,10-14C (55 mCi/mmol, Sigma, St. Louis, MO) dissolved in acetone were added to each. Vials were dark-incubated at room temperature with shaking for 40 hours. The NaOH was mixed with water and ReadySafe liquid scintillation cocktail (Beckman Coulter, Fullerton, CA) and assayed on a Packard TriCarb 2900TR Liquid Scintillation Analyzer (PerkinElmer, Downers Grove, IL). 14C efficiencies (0-156 keV) ranged from 94.37 to 96.18.

3.4 DNA Extraction and Real-time PCR

Total DNA was purified in triplicate from sediment samples using the FastDNA SPIN kit for soil (Qbiogene, Morgan Irvine, CA) with minor modifications as described previously (Dionisi et al. 2004). Primer and probe sets for quantifying dioxygenase genes targeted genes encoding the large (α) dioxygenase subunit (nidA, nagAc, and nahAc). For the nidA TaqMan® probe quantitative PCR assay,

primers and probes target conserved regions determined from a multiple alignment of nidA from several PAH-degrading Mycobacterium: M. flavescens (AF548343), M. fredrickbergense (AF548345), M. gilvum (AF548347), and M. sp. strains PYR-1 (AF249301), 6PYR (AJ494745), JLS (AY330098), KMS (AY330100), MCS (AY33010), and MHP-1 (AB179737). The TaqMan® probe (5’-FAM-TCCTACCCGTCGCCGGTACA-BHQ1) contained several base pair differences to closely related Rhodococcus sp. NCIMB narA (AF082663) gene. Forward and reverse primer sequences were 5’- TTCCCGAGTACGAGGGATAC, and 5’-TCACGTTGATGAACGACAAA, respectively. The real-time PCR reactions used QuantiTect Probe PCR Master Mix (Qiagen, Valencia, CA) and were performed on an MJ Opticon thermocycler (Bio-Rad, Hercules, CA) using the following protocol: 50°C for 2 minutes, 95°C for 15 minutes, then 40 cycles of denaturing at 94°C for 15 seconds and annealing at 56°C for 30 seconds. nidA from Mycobacterium flavescens (ATCC 700033) cloned into TOPO-TA-PCR4.1 (Invitrogen, Carlsbad, CA) vector was used to create an internal eight-point standard curve ranging from 101 to 108 gene copies per reaction. To ensure specificity of the assay, PCR products were cloned. Twenty clones were sequenced using M13 reverse primer on an ABI3100 genetic analyzer (Applied Biosystems, Foster City, CA) at the Molecular Biology Resource Facility, University of Tennessee (Knoxville, TN).

A TaqMan® probe quantitative PCR assay targeting conserved regions of nahAc genes encoding naphthalene dioxygenase in γ-proteobacteria was also developed. Eighteen Pseudomonas were used to design degenerate primers and probe, including, P. stutzeri (AF306425), P. balearica (AF306428), P. fluorescens (AY125981), P. putida strains (AF306430, AF306441, AF306439), and related Pseudomonas sp. Forward and reverse primer sequences were 5’- CAGAGCGTYCCRTTYGAAAA and 5’-TCGAAGCAACCRTARATGAA, respectively. The TaqMan® probe sequence was 5’-FAM- TGGGGTTGAAAGAAGTCGCTCG-BHQ1. Real-time PCR assays were

64 performed as described above, using an annealing temperature of 52°C, and cloned nahA from Pseudomonas putida G7 as the internal standard. Copies of nagAc-like genes in β-proteobacteria as well as universal 16S rDNA genes were quantified as previously described (Harms et al. 2003, Dionisi et al. 2004).

Statistical comparisons of gene abundance were done in NCSS (Number Cruncher Statistical Systems, Kaysville, UT) using a nonparametric one-way ANOVA on the ranked data, and Kruskal-Wallis Multiple Comparison Z-test with Bonferroni adjustment. Comparisons were made between sites (W, DO, CF) and between sampling dates; ANOVA results with p < 0.01 were considered statistically significant.

3.5 Mycobacterium 16S clone library

Primers designed by Leys et al. (Leys et al. 2005b) to target an approximately 530 bp region of the 16S ribosomal gene in fast-growing Mycobacterium (corresponding to positions 66 and 600 in Escherichia coli) were used to create a clone library using DNA extracted from site DO. A touchdown PCR protocol was used (95°C for 10 minutes, then 20 cycles of denaturing at 95°C for 30s, annealing at 55°C - 1°C/cycle for 30s and extension 72°C for 30s, followed by 10 cycles with an annealing step at 45°C, and a final extension of 72°C for 10 minutes). PCR products were cloned into a TOPO-TA-PCR4.1 vector (Invitrogen, Carlsbad, CA). Sequencing was done as described above. Alignments and phylogenetic trees were done using MEGA (Molecular Evolutionary Genetics Analysis) software version 3.1 (http://www.megasoftware.net/). Sequences were deposited in the NCBI GenBank, accession numbers EF438310-EF438383.

4. Results

4.1 Optimization and specificity of the nidA real time PCR assay

Initially, PCR was performed with a gradient of annealing temperatures to determine optimal temperature (59ºC); and different concentrations of primers and probe were combined to determine the optimal combination of 300 nM probe and 600 nM each primer per reaction (data not shown). For both cloned nidA and total sediment DNA extracted, the assay yielded an expected linear response in threshold cycle between 1 ng and 100 ng per reaction, indicating no PCR inhibition (Figure II-2).

To ensure specificity of the nidA real-time PCR assay, 20 cloned products were sequenced, all showing 99 – 100% similarity with PAH-degrading Mycobacterium sp. nidA genes (data not shown). This is expected as this assay targets a conserved region of nidA, and this gene is highly homologous between species (Miller et al. 2004).

4.2 Quantification of nidA, nagAc and nahAc in sediments of Chattanooga Creek

16S rDNA genes were quantified in all samples as a proxy for bacterial biomass (Harms et al. 2003). In October 2004 and July and October 2005, abundance of 16S rDNA copies (biomass) ranged from 2.64 x 108 - 3.70 x 109 copies per gram dry weight sediment (Figure II-3). In April, there was approximately an order of magnitude more biomass, ranging from 1.72 x 109 - 9.67 x 109 copies per gram sediment. A nonparametric one-way ANOVA indicated that there was no significant biomass difference between the three sampling sites, confirming that differences in dioxygenase gene abundance between sites were not simply due to differences in biomass.

50 M. flavescens (ATCC 700033) nidA (cloned) 45 y = 13.098 - 2.148x (r2 = 0.966) Total DNA from contaminated sediments (DO) 40 y = 34.444 - 2.101x (r2 =0.844) )

T 35

Threshold cycle (C 15

0 110100

Template concentration (ng DNA/rxn)

Figure II-2. Threshold cycles (CT) over a range of template concentrtations for both cloned nidA (◊) and environmental samples (■). Error bars represent the standard deviation of triplicate PCRs.

1010 Oct 2004 Apr 2005 109 108 107 106 105 104 103 102 101 1010 Jul 2005 Oct 2005 109 108 107 106 5 Gene copies / g sediment (dry weight) g sediment / copies Gene 10 104 103 102 101

WDOCF 16S WDOCF nidA Site Site nagAc nahAc

Figure II-3. Quantities of 16S (black), nidA (white), nagAc (hatched) and nahAc (grey) at the three sites for each sampling date. Bars are means of triplicate PCR on triplicate DNA extractions from triplicate sediment samples taken at each site. Error bars are ± standard deviation (n = 27).

Between sampling sites there was a significant difference in quantities of nidA and nagAc: both exhibited significantly lower copies at site W. In addition, there were no significant differences in quantities of nidA or nagAc between dates at site W. These observations are consistent with the fact that site W has not experienced the historical contamination and typically exhibits lower concentrations of PAHs than site DO and CF. nidA copies were not significantly different between DO and CF, and nagAc copies were highest at DO. These patterns are also observed in Figure II-4, which shows quantities of each gene by sampling event. nahAc abundances were quite variable: Quantities varied about three orders of magnitude between October 2004 and October 2005, and in some cases, quantities were below the detection limit of the assay (>100 copies/g sediment). This observation is in contrast to nidA and nagAc, which were detected in all samples.

Across sampling dates there was not as clear of a pattern. There were significant differences in copies of nidA: greatest abundances were observed in October 2004 and April 2005, followed by July and October 2005. No significant differences existed in nagAc quantities; however, there was a significantly greater abundance of nahAc in October 2004 compared to other sampling dates.

4.3 PAH concentrations

Sixteen priority PAHs were measured at each site and each sample date (Table II-1). At site W, naphthalene was not detected in April or July 2005; in October 2004, we measured 0.07 mg naphthalene/kg sediment (dry weight). Average concentrations of pyrene ranged from 0.30 – 0.44 mg/kg and concentrations of benzo[a] pyrene up to 0.33 mg/kg. At site DO, average naphthalene concentrations ranged from undetectable to 0.07 mg/kg; pyrene from 0.69 – 9.75 mg/kg; and benzo[a]pyrene from 1.77 – 8.57 mg/kg. Site CF had the highest PAH concentrations with naphthalene ranging from 0.34 – 2.63 mg/kg; pyrene 69

108 nidA nagAc nahAc 107 106 105 104 103 102 October 2004 April 2005 101 July 2005 October 2005 100 Gene copies / ug sediment dry weight WDOCF WDOCF WDOCF

Figure II-4. Quantities of nidA (left), nagAc (center) and nahAc (right) measured at each sampling event (● October 2004, ○ April 2005, ▼ July 2005 and ∆ October 2005). Points are means of triplicate PCR on triplicate DNA extractions from triplicate sediment samples taken at each site. Error bars are ± standard deviation (n = 27).

Table II-1. Concentrations of 16 priority PAHs measured in Chattanooga Creek surface sediments. Concentrations are average mg PAH/kg sediment dry weight, ± standard deviation). - = Not detected October 2004 April 2005 July 2005 PAH W DO CF W DO CF W DO CF Naphthalene 0.07 0.08 2.63 - - 0.08 - 0.07 0.34 ±0.05 ±0.07 ±1.45 ±0.07 ±0.03 ±0.13 Acenaphthylene ------2.66 3.97 ±0.51 1.48 Acenaphthene ------

Fluorene 0.04 0.02 0.98 0.03 0.09 0.92 0.05 0.53 15.74 ±0.01 ±0.01 ±0.17 ±0.01 ±0.08 ±0.36 ±0.02 ±0.06 ±14.3 Phenanthrene 0.23 0.09 3.49 0.19 0.34 1.76 0.34 3.83 5.116 ±0.09 ±0.01 ±0.87 ±0.04 ±0.19 ±0.71 ±0.11 ±0.88 ±1.90 Anthracene 0.35 0.71 24.08 0.31 1.98 14.78 0.11 3.55 13.81 ±0.09 ±0.11 ± ±0.08 ±1.30 ±6.57 ±0.03 ±1.44 ±8.38 Fluoranthene 0.36 0.68 27.10384 0.47 1.89 13.20 0.72 11.44 38.02 ±0.09 ±0.15 ±7.49 ±0.31 ±0.24 ±5.86 ±0.27 ±2.88 ±5.72 Pyrene 0.30 0.69 22.43 0.34 1.60 9.16 0.44 9.75 21.94 ±0.07 ±0.11 ±5.88 ±0.19 ±0.47 ±5.82 ±0.15 ±2.42 ±8.79 Benzo[a] 0.21 0.71 24.13 0.25 1.57 8.94 0.29 9.59 22.15 anthracene ±0.07 ±0.17 ±5.61 ±0.16 ±0.66 ±5.52 ±0.13 ±2.27 ±9.97 Chrysene 0.20 0.40 18.77 0.20 1.23 6.84 0.26 7.43 15.60 ±0.06 ±0.11 ±3.52 ±0.11 ±0.68 ±4.27 ±0.09 ±2.07 ±8.25 Benzo[b] 0.17 0.92 19.38 0.18 1.48 7.64 0.26 11.42 23.69 fluoranthene ±0.05 ±0.27 ±4.08 ±0.11 ±0.94 ±2.65 ±0.14 ±3.07 ±9.17 Benzo[k] 0.06 0.29 7.09 0.07 0.51 2.49 - 4.42 10.80 fluoranthene ±0.02 ±0.09 ±1.52 ±0.05 ±0.35 ±0.87 ±0.47 ±3.43 Benzo[e]pyrene 0.09 0.39 7.17 0.09 0.68 3.10 0.13 5.48 11.97 ±0.03 ±0.11 ±1.51 ±0.06 ±0.49 ±1.05 ±0.07 ±0.76 ±4.27 Benzo[a]pyrene 0.28 1.77 36.34 0.33 4.11 15.25 - 8.57 18.46 ±0.09 ±0.53 ±7.86 ±0.22 ±4.03 ±6.02 ±2.30 ±8.66 Indeno[1,2,3- 0.08 0.52 9.93 0.10 0.79 3.83 0.07 6.93 16.01 c,d] pyrene ±0.02 ±0.17 ±2.06 ±0.06 ±0.57 ±1.62 ±0.06 ±1.63 ±5.26 Dibenzo[a,h] 0.02 0.10 1.92 0.02 0.17 0.82 0.02 2.39 1.58 anthracene ±0.01 ±0.03 ±0.39 ±0.01 ±0.11 ±0.27 ±0.01 ±0.55 ±0.68 Benzo[g,h,i] 0.08 0.39 8.02 0.09 0.70 3.01 0.09 5.11 12.92 perylene ±0.02 ±0.12 ±1.63 ±0.05 ±0.52 ±0.99 ±0.06 ±1.21 ±4.36

ranging from 9.16 – 22.43 mg/kg; and benzo[a]pyrene from 15.25 – 36.34 mg/kg. Concentrations of PAHs were mostly consistent with previously measured concentrations at this site, with the exception of naphthalene which was lower in this study (Dionisi et al. 2004). Concentrations of PAHs with 3 or more rings were strongly correlated (Table II-2).

4.4 PAH mineralization

Naphthalene mineralization was observed in all samples over 40 hours. At site W, 1.2 – 35.6% of added naphthalene was mineralized; at site DO, 7.1 – 99.2% was mineralized, and at site CF, 41.8 – 80.7%. Pyrene mineralization was very low at site W, ranging from 0.1 – 0.8% of pyrene added, but higher at sites DO (11.1 – 25.8%) and CF (16.6 – 26.9%). The amount of pyrene mineralized corresponds with observations made by Cheung and Kinkle (Cheung & Kinkle 2001): inoculated Mycobacterium mineralized less than 40% of added pyrene, even after 100 days. Figure II-5 shows the curvilinear relationship of measured mineralization to PAH concentrations in Chattanooga Creek sediments: biodegradative capabilities are positively related to PAH concentration only at low PAH concentrations.

4.5 Correlations among catabolic genotypes and mineralization potential

To determine the relationship between dioxygenase gene quantities and other variables which may be related to the microbial community’s degradative potential, a correlation matrix of Spearman rank correlation coefficients (rs) was developed (Table II-2). As expected, a significant correlation was observed

between naphthalene and pyrene mineralization (rs = 0.783), and mineralization is positively correlated to concentrations of most individual PAHs (Table II-2). A significant correlation also exists between nidA and nagAc copies per gram dry

Table II-2. Spearman rank correlation coefficients (rs) matrix. NapMin and PyrMin are the percentage of added naphthalene and pyrene mineralized in 40 hours. 16S, nid, nag and nah are abundance of 16S rDNA, nidA, nagAc, and nahAc genes, respectively, normalized to g dry weight sediment (/g), μg DNA extracted (/μg), or to 16S rDNA abundances (/16S). Naph, phen, pyr, benz-a, chry, and benz-gh are concentrations of naphthalene, phenanthrene, pyrene, benzo[a]pyrene, chrysene and benzo[g,h,i]perylene, respectively, measured in mg PAH per kg dry weight sediment. Significant correlations are shaded in grey (n = 12, rs > 0.591, α = 0.05).

Nap Pyr 16S 16S nid nid nid nag nag nag nah nah nah nap phe pyr benz chry Min Min /g /μg /g /μg /16S /g /μg /16S /g /μg /16S h n -a NapMin 1.00 PyrMin 0.78 1.00 16S/g 0.14 0.35 1.00 16S/μg -0.15 0.15 0.85 1.00 nid/g 0.23 0.49 0.36 0.21 1.00 nid/μg 0.02 0.35 0.51 0.43 0.93 1.00 nid/16S 0.33 0.28 -0.33 -0.48 0.66 0.45 1.00 nag/g 0.67 0.81 0.52 0.34 0.76 0.69 0.45 1.00 nag/μg 0.36 0.57 0.68 0.66 0.58 0.69 0.10 0.78 1.00 nag/16S 0.60 0.55 0.41 0.27 0.48 0.52 0.34 0.80 0.85 1.00 nah/g -0.31 -0.01 -0.16 -0.09 0.21 0.27 0.26 0.06 -0.19 -0.18 1.00 nah/ug -0.37 -0.09 -0.24 -0.09 0.19 0.25 0.29 0.01 -0.20 -0.20 0.98 1.00 nah/16S -0.21 -0.09 -0.58 -0.45 0.07 0.06 0.44 -0.09 -0.32 -0.16 0.86 0.89 1.00 naph 0.28 0.58 -0.37 -0.31 0.23 0.16 0.34 0.45 -0.06 0.03 0.65 0.51 0.70 1.00 phen 0.82 0.58 -0.38 -0.53 -0.17 -0.45 0.22 0.25 -0.20 0.18 -0.01 -0.14 0.26 0.49 1.00 pyr 0.75 0.73 -0.48 -0.53 0.30 0.05 0.60 0.57 0.08 0.38 0.18 0.08 0.54 0.76 0.80 1.00 benz-a 0.63 0.78 -0.32 -0.32 0.27 0.17 0.47 0.63 0.23 0.45 0.33 0.22 0.60 0.85 0.68 0.93 1.00 chry 0.73 0.62 -0.52 -0.63 0.15 -0.17 0.47 0.37 -0.12 0.22 0.04 -0.09 0.39 0.66 0.88 0.95 0.82 1.00 benz-gh 0.80 0.78 -0.50 -0.52 0.23 -0.03 0.55 0.53 0.07 0.35 0.14 0.06 0.53 0.73 0.83 0.98 0.92 0.93

100 A

20 % Added PAH Mineralized

0 0.00.51.01.52.02.53.0

Naphthalene (mg/kg dry weight sediment)

100 B

20 % AddedPAH Mineralized

0 0 5 10 15 20 25 Pyrene (mg/kg dry weight sediment)

Figure II-5. PAH mineralization (40 hour incubation) in relation to measured concentrations of naphthalene (A) and pyrene (B). Error bars represent the standard deviation of triplicate mineralization assays (● naphthalene and ○ pyrene) on triplicate sediment samples from each site.

weight sediment (rs = 0.755, p > 0.001) (Figure II-6). Weak but significant correlations between dioxygenase genes and PAHs include nidA (normalized per

biomass) and pyrene (rs = 0.600), nagAc and benzo[a]pyrene (rs = 0.633), nahAc

(normalized per biomass) and benzo[a]pyrene (rs=0.600) and nahAc and

naphthalene (rs = 0.647).

One goal of this study was to assess dioxygenase genes as markers for active biodegradation in this system. The relationship between catabolic gene quantities and pyrene mineralization is shown in Figure II-7. Both nagAc and nidA show positive relationships to pyrene mineralization. Best fit lines for each data set are described by logarithmic nonlinear regression (% pyrene mineralized = y0 + a ln (gene copies /g)): for nidA (% pyrene mineralized = -41.47 + 4.12 ln (nidA/g)), r2 = 0.74 and for nagAc (% pyrene mineralized = -27.55 + 3.46 ln (nagAc/g)), r2 = 0.93.

4.6 Diversity of Mycobacteria

Detection of nidA in Chattanooga Creek sediments implicates the presence of fast-growing, potentially PAH-degrading Mycobacteria. In order to further investigate Mycbacterial diversity, fast-growing Mycobacterium 16S rDNA genes were cloned. Seventy-four clones were successfully sequenced, all showing high sequence similarity (96% - 100%) to other published fast-growing Mycobacterium. Few clones were duplicate sequences: instead the majority was unique, indicating a high diversity of Mycobacterium in this system. A neighbor- joining tree using the Kimura 2-parameter distance model (Kimura 1980) shows the phylogenetic relationship of these sequences (Figure II-8). As expected, most clustered apart from M. ulcerans, a slow-growing Mycobacterium sp. Most sequences fall into the same cluster as M. vanbaalenii, M. austroafricanum, and M. flavescens, species that have all been characterized as high molecular weight

108 nagAc nahAc 107

106

105 copies/g sediment dry weight dry sediment copies/g 104 nagAc

3 or 10 nahAc 102 104 105 106 107 108 nidA copies/g sediment dry weight

Figure II-6. Comparison of Mycobacterium nidA to related naphthalene dioxygenase genes (● β-proteobacteria nagAc and ○ γ-proteobacteria nahAc) in Chattanooga Creek sediments, indicating a coexistence of nidA and nagAc (rs = 0.755). Data points represent the mean of triplicate PCR on triplicate DNA extractions from sediments.

10 % Pyrene mineralized (40 hours) (40 mineralized % Pyrene

5 nidA nagAc nahAc

0 103 104 105 106 107 Gene copies / g dry weight sediment

Figure II-7. Functional relationship between PAH-dioxygenase gene abundances (● nidA, ○ nagAc, ▼nahAc) and in vitro pyrene mineralized. Best fit lines are described by logarithmic nonlinear regression (% pyrene mineralized = y0 + a ln (gene copies /g)): for nidA (% pyrene mineralized = - 41.47 + 4.12 ln (nidA/g)), r2 = 0.74 (long dashes) and for nagAc (% pyrene mineralized = -27.55 + 3.46 ln (nagAc/g)), r2 = 0.93 (short dashes).

Figure II-8. Neighbor-joining tree of 16S rDNA sequences from Chattanooga Creek retrieved using Mycobacterium specific primers. Included are published 16S rDNA sequences from other fast-growing PAH- degrading Mycobacterium (▲), fast-growing non-PAH-degrading M. mucogenicum (■), slow-growing M. ulcerans (●) and Pseudomonas putida as an outgroup (○). Boostrap values (for 5000 iterations) over 50 % are indicated on branches.

DOS-MYC-101 DOS-MYC-118 DOS-MYC-46 DOS-MYC-115 DOS-MYC-43 DOS-MYC-32 63 DOS-MYC-3 74 DOS-MYC-35 DOS-MYC-24 72 DOS-MYC-50 DOS-MYC-74 60 DOS-MYC-73 99 DOS-MYC-120 DOS-MYC-62 AF480578 - M. flavescens 76 AY083217 - Mycobacterium sp KMS 98 AF387803 - Mycobacterium sp MCS AF387804 - Mycobacterium sp JLS DOS-MYC-68 100 DOS-MYC-12 DOS-MYC-13 DOS-MYC-27 DOS-MYC-58 86 X93182 - M. austroafricanum 85 DOS-MYC-102 68 DOS-MYC-64 DOS-MYC-93 77 AY636002 - M. vanbaalenii DOS-MYC-20 DOS-MYC-124 DOS-MYC-9 99 DOS-MYC-31 DOS-MYC-23 71 DOS-MYC-121 50 DOS-MYC-37 DOS-MYC-108 99 DOS-MYC-59 65 DOS-MYC-107 66 DOS-MYC-139 71 DOS-MYC-125 68 DOS-MYC-89 62 AJ276274 - M. fredricksbergense 92 DOS-MYC-56 DOS-MYC-48 89 61 DOS-MYC-57 DOS-MYC-30 54 99 DOS-MYC-17 65 DOS-MYC-19 DOS-MYC-105 97 DOS-MYC-136 DOS-MYC-10 99 X81996 - M. gilvum DOS-MYC-70 50 DOS-MYC-14 72 DOS-MYC-52 57 DOS-MYC-61 99 DOS-MYC-65 72 DOS-MYC-92 DOS-MYC-5 DOS-MYC-21 96 DOS-MYC-141 DOS-MYC-4 DOS-MYC-26 95 DOS-MYC-45 DOS-MYC-11 DOS-MYC-29 DOS-MYC-69 85 DOS-MYC-140 50 DOS-MYC-18 57 DOS-MYC-25 DOS-MYC-91 DOS-MYC-1 100 DOS-MYC-83 52 DOS-MYC-88 77 DOS-MYC-126 63 DOS-MYC-28 DOS-MYC-72 DOS-MYC-84 AF544230 - Mycobacterium sp S65 99 DOS-MYC-15 DOS-MYC-2 AJ627393 - M. mucogenicum DOS-MYC-128 57 DOS-MYC-7 95 X88926 - M. ulcerans AY686638 Pseudomonas putida strainKL33 79 0.05 PAH degraders (Kim et al. 2005b), suggesting that these clones represent other PAH-degrading Mycobacterium. Two clones were most closely related (97 - 98 %) to opportunistic pathogens such as Mycobacterium mucogenicum (AJ627393).

5. Discussion

Catabolic genes represent the degradative potential of a bacterial community, and are indicative of the community’s response to bioavailable substrate. It is hypothesized that catabolic genes can serve as markers of actual function: in the case of PAH-degrading communities, strong positive correlations have previously been found between nahAc gene copies and transcripts (Fleming et al. 1993, Sanseverino et al. 1993b) indicating that the presence of genes in a community is related to their activity. In this study we quantified three different PAH dioxygenase genotypes to better understand the relationship to each other within a community, as well as to actual function (PAH mineralization) of these communities.

While many studies have used molecular tools such as PCR to detect single catabolic genes during biodegradation (Stapleton et al. 1998, Dionisi et al. 2004), multiple targets give a more comprehensive picture of community-level responses, and can provide insight into relationships between populations carrying the genotypes. Multiple catabolic gene targets have been successfully used to establish microbial biodegradative potential in an aquifer (Stapleton & Sayler 1998) and within a BTEX plume (Stapleton et al. 2000b), and have also provided insight into microbial communities exposed to PAHs (Langworthy et al. 1998, Laurie & Lloyd-Jones 2000, Ringelberg et al. 2001). In this study, naphthalene dioxygenase genes nagAc and nahAc (from β- and γ-proteobacteria), as well as pyrene dioxygenase gene nidA (from Mycobacterium) were not only identified, but also quantified. Establishment of dioxygenase genes within these communities combined with measured mineralization indicates that aerobic PAH- degradation is occurring; this observation is consistent with the creek’s frequent aeration due to flood disturbances and its lotic nature.

Abundances of nidA and nagAc were lower at site W, which has the lowest levels of contamination and serves as the control site for this study. Ralstonia-like nagAc abundances measured in our study (103 – 107 copies/g sediment) correspond with those previously measured in Chattanooga Creek by Dionisi et al. (Dionisi et al. 2004) (105 – 107 copies/g sediment), however do not show the same strength of correlation with naphthalene concentrations (r = 0.45 in this study compared to r = 0.77 by Dionisi et al.). This is likely a result of lower naphthalene concentrations (sometimes below detection) measured during our study.

A significant positive correlation was observed between copies of nidA and nagAc in Chattanooga Creek sediments indicating that Mycobacterium and Ralstonia sp. U2-like bacteria coexist and are likely responding to similar ecological cues (such as contaminant concentrations and bioavailability). Enrichment and coexistence of genotypes for low- and high molecular weight PAH degradation are consistent with the complex contaminant mixture in Chattanooga Creek.

In contrast to the relationship between nagAc and nidA, no strong relationship was observed between nahAc and either of the other catabolic genes or mineralization rates; however, there was a significant positive correlation between nahAc and naphthalene concentrations. Sanseverino et al. (Sanseverino et al. 1993b) found a similar positive relationship between nahAc abundance and naphthalene concentrations above a threshold of about 100 mg/kg soil, postulating that at low concentrations of naphthalene, organisms with nahAc may be inactive. While many cultured PAH-degrading organisms have nahAc-like genes, these results suggest that γ-proteobacteria carrying nahAc may not play a major role in Chattanooga Creek sediment communities in terms of PAH degradation. In microcosms amended with low molecular weight PAHs, Laurie and Lloyd-Jones (Laurie & Lloyd-Jones 2000) observed an increase in phnAc genes (from

Burkholderia), but did not detect nahAc, suggesting that this genotype may not be ecologically relevant in all environments. This is also supported by other studies indicating that nah genes alone underestimate total biodegradative capacity (Ahn et al. 1999). In addition, populations containing nahAc have been found to be transient with respect to contaminant concentrations (Stapleton et al. 2000b), possibly explaining the variability in nahAc abundance seen in this study.

Mineralization of pyrene was positively correlated to abundance of nagAc and nidA. As this is, to our knowledge, the first quantitative detection of nidA in PAH-degrading communities, this also presents the first demonstration of a link between nidA abundance to function (pyrene mineralization) in indigenous microbial communities. The observed correlation is consistent with work by Wang et al. (1996), who found a positive correlation between pyrene mineralization and abundance of inoculated Mycobacterium sp. PYR-1 in soil slurries. Our observations suggest that nagAc and nidA may be valuable as molecular biomarkers for high molecular weight PAHs degradation; especially of interest in historically contaminated systems similar to Chattanooga Creek. The presence and enrichment of fast-growing Mycobacterium in Chattanooga Creek sediments has been established through quantification of pyrene dioxygenase genes. As all nidA-carrying organisms identified thus far have been fast-growing Mycobacteria, sequencing of a fast-growing Mycobacterium 16S rDNA clone library was done to examine the phylogenetic distribution of the population with these functional genes. Most clones sequenced were unique, revealing a high diversity of fast-growing Mycobacterium that is only partially represented by our clone library. The majority of sequences clustered with other cultured PAH-degrading species: this observation, combined with detection of nidA genes in this community, suggests that many of these uncultured Mycobacteria could also be capable of high molecular weight PAH degradation. Other studies have also found a high diversity of Mycobacteria in PAH- contaminated soils (Cheung & Kinkle 2001, Leys et al. 2005b). Interestingly, two

83 sequences were most similar to M. mucogenicum, a fast-growing opportunistic human pathogen which has been reported to be frequently found in surface waters (Covert et al. 1999). Finding these sequences is not surprising, considering highly populated area of Chattanooga nearby.

The demonstrated diversity of Mycobacteria, combined with the presence of nidA genes, indicates that PAH-degrading Mycobacterium spp. are present in Chattanooga Creek sediments, and likely playing an important role in attenuation of PAHs.

ACKNOWLEDGEMENTS

This work was supported by the Center for Environmental Biotechnology, Research Center of Excellence, University of Tennessee and the Waste Management Research and Education Institute at the University of Tennessee. Special thanks to F. Menn, J. Easter, J. Rainey, D. Williams, C. Lalande and V. Vulava for assistance in field sampling, and to L. McKay, J. Sanseverino, and three anonymous reviewers for constructive comments on this manuscript. DNA sequencing was done by Joseph May at the Molecular Biology Resource Facility at the University of Tennessee.

Part III.

Microbial community structure and biodegradation activity associated with particle-associated bacteria in a coal tar contaminated creek

1. Abstract

The Chattanooga Creek Superfund site (Chattanooga, TN) is one of the most polluted waterways in the Southeastern USA with high concentrations of toxic and carcinogenic polycyclic aromatic hydrocarbons (PAHs) in the sediments. The sediment communities have been characterized in terms of PAH degradation genotypes (DeBruyn et al. 2007), and contribute to natural attenuation of PAHs from Chattanooga Creek, however less is known about suspended sediments in the water column. Hydrophobic contaminants such as PAHs will associate with these suspended solids, and during flood events may be redeposited onto the floodplain. These suspended particles represent an interesting but understudied environment for PAH-degrading microbial communities. This study tested the hypotheses that particle-associated bacterial (PAB) communities also have the genotypic potential (i.e. PAH-dioxygenase genes) and activity (i.e. naphthalene and pyrene mineralization), and contribute to natural attenuation of PAHs in Chattanooga Creek. Upstream of the Superfund site, mineralization ranged from 0.2 – 2.0 % of added 14C-naphthalene and 0 – 0.1% of added 14C-pyrene (after 40 hour incubation). Mineralization was significantly greater in the PAB communities within the contaminated zone, with 11.8 – 31.2% of added 14C- naphthalene and 1.3 – 6.6% of 14C-pyrene mineralized. In addition, abundances of nagAc (naphthalene dioxygenase) and nidA (pyrene dioxygenase) genes indicate that PAB communities harbor populations with the genetic potential for both low- and high-molecular weight PAH degradation. The PAB communities were compared to sediment communities at the same sites. Microbial communities in these two environments are phylogenetically distinct (determined using T-RFLPs), however share some functional similarities, namely PAH catabolic genotypes, mineralization capabilities, and community structuring along a contamination gradient.

2. Introduction

In an aquatic system, hydrophobic contaminants like polycyclic aromatic hydrocarbons (PAHs) will partition out of the dissolved (aqueous) phase and into the sediments, where microbial biodegradation is well-documented and generally the primary route of attenuation of these carcinogenic and mutagenic pollutants (Mueller et al. 1989). In addition to lake- or stream-bottom sediments, PAHs will also associate with particles in the water column, either through diffusion into the water and subsequent sorption onto suspended particles, or through disturbance and suspension of contaminated stream-bottom sediments (Dachs et al. 1997, Leppard et al. 1998, Pohlman et al. 2002, Marvin et al. 2004, Mackintosh et al. 2006).

The Chattanooga Creek Superfund Site, located in an industrial and residential area of Chattanooga, Tennessee, was placed on the National Priorities list in 1995 due to mixed priority pollutants and heavy coal tar contamination, predominantly from a nearby coking coal plant (Vulava et al. 2007). This contamination has resulted in high concentrations of PAHs in the sediments, despite physical remediation efforts involving excavation of contaminated sediments (Dionisi et al. 2004, DeBruyn et al. 2007). In addition to contaminated creek-bottom sediments, repetitive deposition of polycyclic aromatic hydrocarbons (PAHs) onto the floodplain soils has be documented (Dickerson 2005). It has been demonstrated that flood events can redistribute organic contaminants (Hilscherova et al. 2007), and the frequent flooding of Chattanooga Creek is likely contributing to the observed PAHs on the flooplains.

Particles supended in the water column act as a vehicle for contaminant transport both within an aquatic system (Pohlman et al. 2002) and onto the flooplain during

flood events. This leads to the hypothesis that microbial communities associated with these particles represent one pathway of PAH attenuation. Biodegradative potential, through quantification of both low-and high- molecular weight PAH dioxygenase genes (nidA, nagAc, and nahAc), as well as mineralization activity have already been documented for the sediment communities in Chattanooga Creek (Dionisi et al. 2004, DeBruyn et al. 2007), and this study aims to address the hypothesis that suspended particle-associated bacterial (PAB) communities are also involved in natural attenuation of PAHs from the site.

Suspended particles, also referred to as aggregrates or flocs, have been recognized as a unique environment within the aquatic system, comprised of both inorganic (e.g. clay) and organic (e.g. detritus) materials, supporting a community of particle-associated bacteria (PAB) (Kirchman 1993, Droppo 2001). These PAB communities are important to carbon metabolism and cycling in a variety of aquatic environments (Simon et al. 2002). In both marine and freshwater environments, PAB communities tend have higher densities (Caron et al. 1982, Turley & Mackie 1994) and are phylogenetically distinct compared to free-living bacteria (DeLong et al. 1993, Crump et al. 1999, Moeseneder et al. 2001, Riemann & Winding 2001, Allgaier & Grossart 2006b, Zhang et al. 2007). In addition, PAB generally display higher metabolic and enzymatic activity per cell, compared to free-living bacteria (Kirchman 1983, Lemarchand et al. 2006, Ghiglione et al. 2007).

Because of the tendency for hydrophobic contaminants to associate with suspended particles, these particles may be an environment suited to biodegradation: higher densities of bacteria on the particles increases contact with PAHs and may aid bacteria in accessing insoluable PAHs (Johnsen & Karlson 2004). In addition, suspension in the water column, especially in a lotic creek such as Chattanooga Creek, results in an oxygenated environment conducive to aerobic biodegradation, compared to the bottom sediments where oxygen can

become limited. There is evidence that biodegradation of PAHs increases with increasing suspended sediment loads (Xia et al. 2006), indicating that these understudied communities are likely contributing to PAH attenuation in aquatic systems.

While other studies have contrasted free-living bacteria and PAB communities, with an emphasis on phylogenetic relationships, this study compares PAB with sediment communities, with an emphasis of functional (biodegradative) similarities and differences. Suspended particle systems are very dynamic, however very little baseline information exists regarding molecular biodegradative potential of these communities. This study addresses the hypotheses that a) particle associated bacterial communities are unique in structure (phylogeny) and function (low- and high-molecular-weight PAH biodegradation) when compared to the sediment communities, and that b) these communities are contributing to the natural attenuation of PAHs from Chattanooga Creek.

3. Materials and Methods

3.1 Sample collection

Surface (top 20 cm) water samples were collected from Chattanooga Creek at sites W, DO, and CF (Figure III-1). Samples were collected in April, July and October of 2005. One L of water was filtered through 47 mm diameter 5 μM polycarbonate filters (Millipore, Billerica, MA). At least seven filters were collected for each site at each date, four were used immediately for mineralization assays, and at least three were stored at -20°C for DNA extraction. Suspended material dry weight was determined after baking filters at 80°C for > 48h. Sediment samples were collected as previously described in DeBruyn et al. 2007.

3.2 SEM

Particles captured on the 5 μm filtures were imaged using environmental SEM (ESEM) in a method similar to Droppo (2001). Circles 12 mm in diameter were punched and mounted on an SEM sample mount. Images were taken with a Hitachi S 4300 Scanning Electron Microscope operating at 20 kV.

3.3 Mineralization assays

Filters collected for each site were cut into small pieces using sterile scissors and slurried with 2 ml sterile dH2O in a 40 mL EPA vial with a septum. For each site, three vials were used for mineralization measurements and the fourth was mixed with 1 ml of 2 N H2SO4 to serve as a killed control. An 8 ml vial with 0.5 ml 0.5

N NaOH was placed in each 40 ml vial as a CO2 trap. As a positive control, a 1 ml overnight culture of Mycobacterium flavescens (ATCC 700033) (pyrene) or 91

Figure III-1. Map of Sampling locations on Chattanooga Creek. The Tennessee River is pictured in the top left-hand corner; the Tennessee- Georgia border is along the bottom. Zone I is upstream of point sources of contamination (sources include the Tennessee Products coking coal carbonization facility). Zone II sediments were excavated in 1998; Zone III is currently being excavated (2006-2007).

Pseudomonas fluorescens HK44 (naphthalene) were used in place of filters. Either 400,000 dpm naphthalene-1-14C (31.3 mCi/mmol, Sigma, St. Louis, MO) or 200,000 dpm pyrene-4,5,9,10-14C (55 mCi/mmol, Sigma, St. Louis, MO) dissolved in acetone were added to each. Vials were dark-incubated at room temperature with shaking for 40 hours. The NaOH was mixed with water and ReadySafe liquid scintillation cocktail (Beckman Coulter, Fullerton, CA) and assayed on a Packard TriCarb 2900TR Liquid Scintillation Analyzer (PerkinElmer, Downers Grove, IL). 14C efficiencies (0-156 keV) ranged from 94.37 to 96.18.

3.4 DNA extraction and real-time PCR

Total DNA was extracted from the filters using the FastDNA SPIN kit for soil (Qbiogene, Morgan Irvine, CA). One 47mm filter was inserted into each lysing matrix tube, and minor modifications were made to the extraction protocol as described previously (Dionisi et al. 2004).

To quantify abundance of PAH catabolic genes in particle-associated communities, TaqMan probe quantitative PCR assays targeting genes that encode the large (α) subunit of the terminal dioxygenase, responsible for the initial dihydroxylation of the aromatic ring. The gene targets included: pyrene dioxygenase genes (nidA from Mycobacterium), and naphthalene dioxygenase genes (nahAc from γ-proteobacteria and nagAc from β-proteobacteria) ; assays were designed and applied as described previously (Dionisi et al. 2004, DeBruyn et al. 2007). Eubacterial 16S rDNA genes were also quantified as a proxy for bacterial biomass as previously described (Harms et al. 2003). In addition, a real time PCR assay targeting fast-growing Mycobacterium 16S rDNA genes was designed as a proxy for fast-growing Mycobacterial biomass. The assy used forward primer 5’-CCTTATGTCCAGGGCTTCA, and reverse primer 5’-

TAGCGACTCCGACTTCACG, which amplified 136 nucleotides of the V7 region of the 16S rDNA gene (approximately 1208 – 1344, E. coli numbering). The internal dual-labelled probe (5’- FAM- TTGAAAGGATTCGCTCCACCTCA-BHQ-1) targeted nucleotides 1273-1296. Gradient PCR was done to determine the optimal annealing temperature (data not shown). The real-time PCR reactions used QuantiTect Probe PCR Master Mix (Qiagen, Valencia, CA) and were performed on an MJ Opticon thermocycler (Bio-Rad, Hercules, CA) using the following protocol: 50°C for 2 minutes, 95°C for 15 minutes, then 40 cycles of denaturing at 94°C for 15 seconds and annealing at 57°C for 30 seconds. As a negative control, the assay was tested against genomic DNA extracts of slow-growing environmental species of Mycobacteria, namely M. marinum, M. terrae, and M. triviale.

16S rDNA was amplified from Mycobacterium flavescens (ATCC 700033) using universal 16S primers 8F and 1492R, and cloned into TOPO-TA-PCR4 (Invitrogen, Carlsbad, CA) vector. These clones were used to create an internal eight-point standard curve ranging from 101 to 108 gene copies per reaction. To ensure specificity of the assay, PCR products were also cloned. Twenty PCR product-containing clones were sequenced using M13 reverse primer on an ABI3100 genetic analyzer (Applied Biosystems, Foster City, CA) at the Molecular Biology Resource Facility, University of Tennessee (Knoxville, TN).

Statistical comparisons of gene abundance were done in NCSS (Number Cruncher Statistical Systems, Kaysville, UT) using a nonparametric one-way ANOVA on the ranked data, and Kruskal-Wallis Multiple Comparison Z test with Bonferroni adjustment. ANOVA results with p < 0.01 were considered statistically significant.

3.5 T-RFLPs

Terminal restriction fragment length polymorphisms (T-RFLP) profiles were generated to compare phylogenetic fingerprints of both sediment and PAB communities in the July and October samples. Triplicate PCRs were set up for each sample to amplify 16S rDNA genes from extracted total DNA. For each 25 μl reaction, 10 ng of DNA was mixed with 400 nM each of the universal eubacterial primers 8F (tagged with FAM) and 907R (Lane et al. 1985, Reysenbach et al. 1994) and puRe Taq Ready-to-go PCR beads (GE Healthcare). The following thermocycler program was followed: 95°C for 10min, followed by 15 touchdown cycles (95°C for 30s, 60°C – 1°C/cycle for 30s, and 72°C for 1 min), followed by 10 cycles of 95°C for 30s, 45°C for 30s and 72°C for 1 min, and finished with a final extension at 72°C for 10 min. Only 25 total cycles were used, in order to minimize PCR bias (Suzuki & Giovannoni 1996). PCR products were purified using a QIAquick PCR Purification Kit (Qiagen, Valencia, CA) according to the manufacturer’s instructions to remove any polymerase which can potentially alter the length of DNA fragments during restriction enzyme digestion (Hartmann et al. 2007). Each purified PCR product was divided into three for the restriction digest. AluI and HhaI were chosen because preliminary T-RFLPs on Chattanooga Creek sediment communities indicated these two would provide the resolution required (i.e. gave more peaks that other enzymes tested, data not shown). Twenty μl restriction digests consisted of 15 ul PCR product (about 50- 100ng; if yield was low, 2 PCR products were combined), 10U of restriction enzyme (AluI or HhaI, Promega, Madison, WI), and buffer and BSA according to manufacturer’s instructions. One third of each PCR product was mixed as above, without the restriction enzyme, as an uncut control. Digests were incubated at 37°C for at least 3 hours. This was determined to be a sufficient about of time for full digestion of the PCR product, as no change the T-RFLP profile was observed after 3 h digestion time (data not shown). After digestion, DNA was cleaned again using the PCR Purification Kit to remove buffers and enzyme, then

completely dried on a Savant SpeedVac (Thermo Scientific). 0.75 μl of ROX500 size standard (Applied Biosystems, Foster City, CA) and 20-30 μl deionized formamide was added to each sample before denaturing at 96°C for 5 min followed immediately by 10 min on ice. Samples were then transferred to 96-well plates and analysed on an ABI Prism 3100 Genetic Analyzer (Applied Biosystems, Foster City, CA) using default GeneScan (fragment analysis) parameters.

3.6 T-RFLP data analysis and statistics

T-RFLP profiles of replicate PCRs were aligned by peak size: Peaks that did not occur in all three replicates were removed. Total fluorescence (i.e. the sum of all peak heights, as a relative measure of total amount of DNA loaded) for a trace was used to normalize replicates in a reiterative manner similar to Dunbar et al. (Dunbar et al. 2001), and adjusted peaks giving a fluorescence of < 20 were removed. A consensus trace for a sample was calculated using the average peak size (nucleotides) and height of the three replicate PCRs. Consensus traces from all samples for each enzyme were then aligned and peaks within 0.5 nucleotides were binned together. Samples which did not contain a peak of a given size were given a “0” value for that peak. The resulting matrix was used for multivariate analysis.

Constrained multivariate ordination was done using CANOCO for Windows version 4.54. Canonical correspondence analysis (CCA) was selected as the optimal approach for constrained multivariate analysis of the T-RFLP data: CCA uses a unimodal species distribution model, and is not susceptible to biases caused by high multicollinearity in the dataset (Palmer 1993). In addition, CCA compares relative species abundances, so will not represent artificial differences between communities due to differences in DNA concentrations (an artifact of T-

RFLP which may unduly influence total abundances). Predictor (“environmental”) variables included PAH concentration variables, including Total PAH (sum of 16 EPA priority PAH concentrations in mg/kg, LMW PAH (low molecular weight PAHs, sum of PAHs with three rings or less), HMW PAH (high molecular weight PAHs, sum of PAHs with more than three rings), LMW:HMW (ratio), Naph (Naphthalene concentration in mg/kg), Pyr (pyrene concentration in mg/kg). These data were measured in the sediments only and were reported previously (DeBruyn et al. 2007) – there was not enough material on the filters to accurately measure particle-associated PAHs. To determine how this would affect the analysis, all CCAs were run both with a predictor matrix that included PAH measures and one that did not; in all cases the relationships between communities in ordination space remained the same. PAH mineralization data were also used as predictor variables, including Naph Min (% of 14C-naphthalene mineralized in 40 hours) and PyrMin (% 14C-pyrene mineralized in 40 hours). Also used as predictor variables were gene quantities measured using real time PCR, normalized for total bacterial biomass (16S rDNA gene copies), including nidA (copies of pyrene dioxgyenase), nagAc (copies of naphthalene dioxygenase) and myco (copies of fast-growing Mycobacterial biomass). Hill’s scaling with a focus on inter-sample distances and downweighting of rare species was used. A Monte Carlo permutation test was used to test the null hypothesis that the ordination patterns were unrelated to environmental parameters (499 random permutations).

In addition to the constrained CCA analysis, data was also subjected to an unconstrained nonmetric multidimensional scaling (NMDS) using Primer v6.1 software (PRIMER-E Ltd., Plymouth, UK). NMDS does not make assumptions about the relationships between species and environmental gradients and is therefore a straightforward representation of similarities between communities, useful for verification of constrained analyses such as CCA. Abundance data was first standardized by sample total, then transformed either into binary data or

square root. Bray-Curtis similarity scores were calculated and used to construct the NMDS plots. To test for statistically significant differences between communities at different sites and dates, ANOSIM was done on a matrix of ranked Bray-Curtis similarities using Primer v6.1 software (PRIMER-E Ltd., Plymouth, UK) (Clarke & Gorley 2006). The following groups were tested: environment (particle-associated vs. sediment communities), site (W, DO, or CF) and date (July and October).

4. Results

4.1 Particle characterization and PAH mineralization

Average sediment loads ranged from 4.25 – 17.2 mg of ≥ 5 μm suspended particles (dry weight) per L for all dates and sites sampled, with no significant differences between loadings at different sites or dates (ANOVA, p < 0.001). Environmental SEM (E-SEM) images show that these particles are complex matrices of organic and inorganic materials, characteristic of suspended sediment (Droppo 2001) (Figure III-2).

Table III-1 gives the percentages of labeled PAHs that were mineralized by the particle associated communities after 40 hours, as well as the first order biodegradation rate constants. In general the particles collected from site DO and CF, which have higher concentrations of PAHs in the sediments, exhibit greater mineralization of PAHs. As expected, more of the naphthalene was mineralized compared to pyrene.

4.2 Specificity of the Myco16S real time PCR assay

The real time assay targeting fast-growing Myocbacterium 16S rDNA genes was tested against slow-growing Mycobacterium spp. as a negative control to ensure specificity. The assay did not amplify M. marinum or M. terrae 16S, however there was some amplification of M. triviale (data not shown). M. triviale is one of the deepest branching of the slow-growing group (i.e. more closely related to the fast-growers) and is the only known slow growing strain that can grow on 5% NaCl, a characteristic of the fast growers (Stahl & Urbance 1990). The assay was also tested for specificity by cloning and sequencing real time PCR products. The

Figure III-2. Environmental ESM images of particles trapped on 5 μM polycarbonate filters. Water samples were taken from site DO in October 2005. Pores are visible as black circles in the upper photo.

100

Table III-1. Percent of added 14C-naphthalene or 14C-pyrene mineralized and first-order biodegradation rate constants (k) by bacterial communities associated with > 5 μm particles. Assay was carried out for 40 hours. Percentages and rates are mean ± standard deviation (n=3).

14C-Napthalene Site July 2005 October 2005 Percent Rate (day -1) Percent Rate (day-1) W 2.02 % ± -2.202 ± 0.24 % ± -2.61 ± 1.87 % 2.043 x 10-3 0.04 % 0.413 x 10-4 DO 22.36 % ± -0.0244 ± 11.75 % ± -0.0128 ± 15.87 % 0.0173 10.57 % 0.0115 CF 24.93 % ± -0.0272 ± 31.21 % ± -0.0341 ± 5.07 % 0.553 11.14 % 0.0122

14C- Pyrene Site July 2005 October 2005 Percent Rate (day -1) Percent Rate (day -1) W 0.05 % ± -2.700 ± 0.00 % ± 0 ± 0.03 % 1.446 x10-5 0.11 % 5.913 x 10-5 DO 1.42 % ± -7.727 ± 1.27 % ± -6.935 ± 0.38 % 2.071 x 10-4 0.53 % 2.907 x 10-4 CF 2.59 % ± -1.411 ± 6.61 % ± -3.607 ± 0.93 % 0.5047 x 10-3 1.02 % 5.542 x 10-3

101 primers amplified sequences from fast growing Mycobacterium spp., along with some Rhodococcus and other Actinomycetes, however, all of these non- Mycobacterial PCR products contained at least 4 base mismatches to the probe, indicating that they not contributing to the detectable light signal (i.e. not a false positive).

4.3 Genotypic characterization of PAH degrading populations in PAB communities

Figure III-3 shows the abundances of 16S rDNA, nidA, nagAc, nahAc, and Mycobacteria 16S rDNA measured per litre of water on each of the three sampling dates (nahAc and Mycobacterial 16S were not measured for the April samples). nidA and nagAc were detected in all PAB communities, indicating that these populations do have the catabolic potential for pyrene and naphthalene mineralization. In particular, quantification of both nidA and Mycobacterial 16S indicates that fast-growing Mycobacterium are part of the particle-associated bacterial communities.

Total eubacteria and Mycobacterium biomass remained consistent between sites: there were no significant differences in quantities of total 16S rDNA or Mycobacteria 16S rDNA (Kruskal-Wallis one-way ANOVA, p > 0.05). nidA abundances were also similar between sites: there was no significant difference in quantities (Kruskal-Wallis one-way ANOVA, p > 0.05). In contrast, nagAc abundances were significantly lower at site W compared to the other two sites (DO and CF) (Kruskal-Wallis One-way ANOVA, p < 0.001). nahAc abundances were more varied : abundances were four-fold higher in July compared with the October samples, and at site CF in October, nahAc was not detected (Figure III-3).

102

April July October 109 109 109

8 8 8 10 10 10 j-16S j-nid 7 7 7 10 10 10 j-beta j-gamma 106 106 106 j-myco

105 105 105

104 104 104

103 103 103

102 102 102

Gene copies/L filtered water 101 101 101

100 100 100 WDOCF WDOCF WDOCF

16S nidA nagAc nahAc myco 16S Figure III-3. Gene abundances in particle-associated communities. Gene copies measured per L of water filtered (> 5 μm), including eubacterial 16S rDNA (black), nidA (white), nagAc (hatched), nahAc (light grey), and Mycoabacteria 16S (dark grey). nahAc and Mycoabacterial 16S was not quantified for the April samples. Bars are means of triplicate PCR on triplicate filters taken at each site (n=9). Error bars are ± standard deviation.

103

Gene abundances in the PAB communities were compared to measurements made on the sediment communities (DeBruyn et al. 2007), to determine functional similarities or differences between communities in these two environments (Figure III-4). While both nidA and nagAc were detected in all samples for both PAB and sediment communities, the proportions and relationships between these genes varied. There was a significantly lower abundance of nidA (as a percentage of both total eubacterial biomass and mycobacterial biomass) in the PAB communities compared to the sediments (Mann-Whitney U test, p < 0.001), and no significant correlation existed between nidA copy numbers in the sediments and PAB communities. Fast-growing Mycobacterium biomass ranged from 0.30% - 1.32% of total eubacterial biomass in the PAB and 0.80% - 3.22% in the sediments. Figure III-5 shows that while Mycobacteria biomass was similar in sediments and PAB, nidA abundances were lower in PABs.

In contrast, there was no significant difference in abundance of nagAc between sediment and PAB communities, and a positive relationship was observed between particle and sediment nagAc quantities (Figure III-4). There was no detectable relationship between nahAc in the sediments and PAB, largely due to the variability in abundance observed for this genotype. Positive relationships that had previously been observed in the sediment communities between nidA and nagAc (DeBruyn et al. 2007) were not observed for PAB (Figure III-6).

4.4 Phylogenetic comparisons between sediments and particle-associated communities (T-RFLP)

16S rDNA T-RFLPs were used to obtain phylogenetic fingerprints of the sediment and PAB communities. The combination of T-RFLP profiles for all samples resulted in a total of 227 restrition fragment lengths (ribotypes) when digested with AluI and 232 for HhaI, indicating that both enzymes gave a similar 104

10-1

10-2

10-3

10-4

10-5

10-6

-7 10 nidA nagAc 10-8 nahAc Mycobactium 16S

10-9 Particle-associated gene copies / 16S rDNA copies rDNA 16S / copies gene Particle-associated 10-6 10-5 10-4 10-3 10-2 10-1 100

Sediment gene copies / 16S rDNA copies Figure III-4. Gene abundances in particle associated communities compared to abundances in sediment communities. Gene copies are normalized to total Eubacterial biomass (16S rDNA copies) for the community.

105

1e+8

p-myco/L vs p-nidA/L s-myco/g vs s-nidA/g 1e+7

1e+6

1e+5

1e+4

copies (particles) /L or (sediments) /g 1e+3 nidA 1e+2 1e+4 1e+5 1e+6 1e+7 1e+8 1e+9 Mycobacterium 16S rDNA copies / L(particles) or /g (sediments) Figure III-5. Relationship between Mycobacterium biomass and nidA copies in sediments (black) and particle-associated communities (white).

Suspended particles 1e+7 Sediments

1e+6

1e+5

1e+4

copies /g (sediment) or /L (water) /g (sediment) copies 1e+3

nagAc 1e+2 1e+3 1e+4 1e+5 1e+6 1e+7 nidA copies /g (sediment) or /L (water) Figure III-6. Positive correlation between nidA and nagAc abundance in sediments (black) was not observed in the PAB communities (white).

106

level of community resolution. There were no significant differences in ribotype richness (number of T-RFLP fragments) between environments (PAB vs. sediments), sites (W, DO, or CF) or dates (July, October) (one-way ANOVA, p > 0.05) (data not shown). These data were used for canonical correspondence analysis (CCA) along with a matrix of predictor variables. Figure III-7 shows the ordination plots from canonical correspondence analysis for both the AluI and HhaI datasets. The similarity in arrangement of the samples in multivariate space for both datasets confirms the observed relationships between communities are not spurious. In addition, Monte Carlo permutation tests were significant (p < 0.01) confirming that the distribution of samples with respect to the environmental variables is nonrandom. The first canonical correspondence axis explains the greatest amount of variance (27.4% for AluI and 30.8% for HhaI). There is a separation between PAB and sediment communities along this axis, indicating that phylogenetic differences are greatest between the two environments (compared to differences between sites or dates). This first axis, or gradient, along which these two communities separate is correlated to some of the biological predictor variables, especially nidA and nagAc abundances.

The second canonical correspondence axis explains the second most amount of variance – the first two axes account for 33.3% and 34.8% of the total variance for AluI and HhaI respectively. This axis is strongly correlated to the PAH predictor variables (Total PAH, LMW (low molecular weight) PAHs, HMW (high molecular weight) PAHs, naphthalene and pyrene concentrations), indicating that PAH concentration is a determinant of community structure within each environment. For both PAB and sediment communities, the community follows the general trend of increasing PAHs (W < D < CF), that is, the CF samples plot further along the gradient of PAHs than the other sites.

NMDS analysis was done on both AluI and HhaI matrices. All configurations had 2-D stress values < 0.14. Both binary and square root transformations yielded

107

Figure III-7. CCA ordination plots of T-RFLP profile data. Ordination plot of the CCA, using Hill’s scaling for alu matrix (top) and hha matrix (bottom). Sediment communities are represented by grey circles, particles represented by open circles. Predictor variables include: Total PAH (16 EPA Priority PAHs), LMW PAH (low molecular weight PAHs), HMW PAH (high molecular weight PAHs, Naph (Naphthalene), Pyr (Pyrene), NaphMin (% 14C- Naphthalene mineralized in 40 hours), PyrMin (% 14C-Pyrene mineralized in 40 hours), nidA (copies of pyrene dioxygenase nidA, normalized for bacterial biomass), nagAc (copies of naphthalene dioxygenase nagAc, normalized for bacterial biomass), and myco (fast-growing Mycobacterial biomass, normalized for total bacterial biomass). Coding of samples: first letter refers to date of sample (J=July, O=October), second letter(s) refer to sampling site (W, D or CF).

108

similar plots with stress values within 0.05, only the square root transformed data NMDS analysis was done on both AluI and HhaI matrices. All configurations had 2-D stress values < 0.14. Both binary and square root transformations yielded similar plots with stress values within 0.05, only the square root transformed data is shown in Figure III-8 and Figure III-9. The plots of the whole matrices (Figure III-8) shows a clear separation between particle and sediment communities, confirming the separation along the primary axis seen in the CCA. The particle and sediment communities were also separately subjected to NMDS. Figure III-9 shows the NMDS plots of the separate datasets, using the HhaI matrices as input.

ANOSIM was used to determine if there were significant differences between environments, sites, or dates based on Bray-Curtis similarity scores. The results are shown in Table III-2. The ANOSIM generates the R statistic; when R > 0.75, groups are well separated, R > 0.5 indicates groups are separated but overlapping, R > 0.25 groups are barely separable (Clarke & Gorley 2006). Both the HhaI and AluI matrices were used as input, and the data was either binary transformed (presence-absence) or square root transformed. In all cases, the two matrices (HhaI and AluI) were always in agreement; the two types of data transformation also yielded very similar analysis results. When the whole data set is used, there is a significant separation between the type of environment, i.e. particle associated vs. sediment communities (R = 0.905- 0.997, p < 0.001). This confirms the separation along the primary axis observed in both the CCA (Figure III-7) and NMDS (Figure III-8) plots. When the data set is split by environment type, the sediment communities show some significant separation by site (R = 0.38 – 0.46, p < 0.001) but little by date. In contrast, the particle-associated communities show some significant separation by date (R = 0.54 – 0.86, p < 0.001), but little separation by site. The clusters are evident in the NMDS plots of the particle- associated and sediment communities (Figure III-9).

109

1.2 alu - sqrt stress = 0.07 1.0 OCFF1 OCFF6 OCFF4 0.8

OWSB 0.6

0.4 OCFSCOCFSA 0.2 JWSAOWSA OWF5ODF4 ODF2 OWF3 JWSB OWSC ODF6OWF2 Axis 2 JWSC JCFSB 0.0 JCFSC JCFSAODSBODSC JSDCJDSBODSA JCFF1 JDF6JCFF2JCFF7 -0.2 JDSA JWF4 JWF2 OCFSB -0.4 JDF4 JWF5 JDF1 -0.6

-0.8 -1.5 -1.0 -0.5 0.0 0.5 1.0 1.5 Axis 1

1.0 hha - sqrt - stress = 0.06 0.8 OCFSB OCFF4 ODF2 0.6 OCFF1 OCFF6 OCFSC

0.4 OCFSA ODF4 OWF5 0.2 ODF6 JDSA ODSBODSC ODSA OWF2 OWF3 JWSCOWSC JDSB Axis 2 0.0 JDSC JWSB OWSA JDF6JCFF1 JWSA -0.2 JWF5JWF4 JWF2 JCFSBOWSB JCFF7JCFF2 JCFSA JCFSC -0.4 JDF4

-0.6 JDF1

-0.8 -1.5 -1.0 -0.5 0.0 0.5 1.0 1.5 Axis 1 Figure III-8. NMDS plots of T-RFLP profiles. Plots for AluI (top) and HhaI (bottom) digested samples. Coding of samples: first letter refers to date of sample (J=July, O=October), second letter(s) refer to sampling site (W, D or CF).

110

1.5 hha - part - sqrt - stress = 0.06

OWF2 1.0 OWF3

OWF5 0.5 JCFF7 JCFF2 ODF2 ODF4 ODF6 Axis 2 Axis JCFF1JDF6 JDF4 0.0 JDF1 JWF5

JWF4 -0.5 OCFF4 OCFF6 JWF2 OCFF1

-1.0 -1.5 -1.0 -0.5 0.0 0.5 1.0 1.5 2.0 Axis 1

1.5 hha - sed - sqrt - stress = 0.13 JDSC 1.0

JDSA ODSA 0.5 JDSB ODSB ODSC JCFSC OCFSB JCFSB JWSC 0.0 JWSB

Axis 2 OWSC OWSA

OCFSC JCFSA JWSA -0.5 OCFSA

-1.0 OWSB

-1.5 -3 -2 -1 0 1 2 Axis 1 Figure III-9. NMDS of T-RFLP profiles (HhaI digested) of particle- associated communities (top) and sediment communities (bottom). Coding of samples: first letter refers to date of sample (J=July, O=October), second letter(s) refer to sampling site (W, D or CF).

111

Table III-2. Results of ANOSIM analysis. ANOSIM determines if there are significant differences between groups, based on Bray-Curtis similarities. The sediment and particle community data were tested as separate data sets, then combined and tested as a whole. Grouping was by site (W, DO, or CF), by date (July or October), or by environment (sediment and particle). Data was transformed either to binary (presence-absence) or by square root. Reported here is the Global R calculated by ANOSIM, and the estimated p value. Grey boxes show where significant differences (p < 0.01) between groups were observed.

Data set Grouping Restriction Binary transformed Square root variable enzyme transformed Global R p value Global R p value Whole Environment Alu I 0.945 0.001 0.997 0.001 (sediment Hha I 0.905 0.001 0.996 0.001 and Date Alu I 0.073 0.078 0.083 0.072 particles) Hha I 0.136 0.02 0.142 0.021 Site Alu I 0.023 0.26 0.019 0.263 Hha I 0.009 0.327 0.002 0.384 Sediment Date Alu I 0.097 0.065 0.118 0.035 Hha I 0.082 0.097 0.134 0.024 Site Alu I 0.375 0.001 0.390 0.001 Hha I 0.42 0.001 0.456 0.001 Particles Date Alu I 0.535 0.001 0.619 0.001 Hha I 0.758 0.001 0.861 0.001 Site Alu I 0.109 0.120 0.151 0.059 Hha I 0.049 0.256 0.012 0.378

112

5. Discussion

The microbial communities associated with suspended particles in the water column of coal tar-contaminated Chattanooga Creek have both the genotypic potential (PAH dioxygenase genes) and actual PAH mineralization (naphthalene and pyrene) activity, indicating that these communities are likely contributing to natural attenuation of PAHs from the site. The abundances of nidA and fast- growing Mycobacterium 16S rDNA genes were not significantly different between sites, suggesting that a population of pyrene-degrading Mycobacteria make up a stable portion of the particle-associated bacterial (PAB) community. This is consistent with observations of nidA abundances in the sediments of Chattanooga Creek (DeBruyn et al. 2007). In contrast to nidA abundances, Ralstonia-like naphthalene dioxygenase genes (nagAc) were significantly enriched in PAB collected from sites DO and CF compared to the background site W. Interestingly, the strong correlation previously seen between nidA and nagAc in sediment communities of Chattanooga Creek (DeBruyn et al. 2007) was not observed in the PAB communities, suggesting that while the Mycobacteria- and Ralstonia-like populations do coexist, they may not be responding to similar ecological cues.

This is the first study (to our knowledge) indicating that fast-growing PAH- degrading Mycobacteria are part of the suspended particle-associated communities. In combination with detection of pyrene dioxygenase nidA genes, this observation implicates suspended particles as an environment conducive to pyrene degradation – this of importance as the higher molecular weight PAHs tend to more toxic and more recalcitrant than lower molecular weight PAHs (such as naphthalene) (Cerniglia 1992). The presence of Mycobacteria is not unexpected: Actinomycetes have previously been found to be associated with suspended particles in other freshwater aquatic systems (Crump et al. 1999,

113

Allgaier et al. 2007); in particular, Allgaier et al. 2007 recovered Mycobacteriaceae sequences associated with particles which were not observed free-living in the water column. The presence of Ralstonia–like nagAc genes associated with PAB also corresponds to other studies: β-proteobacteria have been found to be associated with suspended particles in freshwater systems, and in many cases dominate PAB assemblages (Weiss et al. 1996, Grossart & Simon 1998, Schweitzer et al. 2001, Lemarchand et al. 2006).

In comparing PAB and sediment communities in Chattanooga Creek, it was noted that these communities do not significantly differ in abundances of Mycobacterium, however there is significantly higher nidA abundances in the sediments. The difference in Mycobacterium: nidA ratios may be because PAB communities may have more fast growing Mycobacteria that do not have the nidA genotype, and/or the nidA gene may be carried by other organisms (non- Mycobacterium) in the sediments. To date, nidA genes have only been associated with fast-growing Mycobacterium.

16S T-RFLP profiles indicated that sediment and PAB communities are phylogenetically distinct: the greatest phylogenetic differences observed were between physical environments, as opposed to temporal or spatial differences. In the ordination plots of the canonical correspondence analysis, this difference is reflected along the first ordination axis, and ANOSIM confirmed this was a significant separation. In the constrained CCA ordination, some biological predictor variables included in the analysis correlated with this first axis, suggesting that the biological (functional) differences in communities are related to the phylogenetic (structural) differences observed.

Within each of the environments (particle-associated and sediment) some structuring along the second CCA axis is apparent. The second ordination axis explains the next most amount of variance, and is very strongly correlated to PAH

114

concentrations: communities in both PAB and sediments separate out along this axis spatially (low PAHs at W > D > CF high PAHs) indicating that within each physical environments, contaminant concentration is an important predictor of community structure. Unconstrained NMDS and ANOSIM confirmed this pattern for the sediment communities: There was significant separation of communities by site, indicating that environmental parameters chosen (i.e. PAH concentrations) were adequate at predicting species distributions. Microbial community changes in relationship to PAH chemistry is well-documented: communities shifts have been observed in response to both the type (Ní Chadhain et al. 2006, Muckian et al. 2007) and concentration (Langworthy et al. 1998, Cheung & Kinkle 2001, Ringelberg et al. 2001, Langworthy et al. 2002, Vinas et al. 2005) of PAHs present. In this study, there was no change in richness (i.e. no significant differences between number of ribotypes) in the samples, indicating that community shifts are due proliferation of different populations, also observed at other PAH-contaminated sites (Muckian et al. 2007).

The only disagreement between CCA and NMDS was with respect to the particle- associated communities. When axes were constrained to environmental parameters (CCA), there was a noticeable spatial structuring of communities. However, unconstrained analysis (ANOSIM) showed no significant spatial differences by site, only temporal differences by sampling date. This suggests there are likely other environmental factors structuring PAB communities that were not accounted for in this analysis. It is important to do simultaneous constrained and unconstrained multivariate approaches; constrained analyses relate species distributions to environmental parameters, and will therefore only show biological variation explained by environmental parameters. Unconstrained ordination, like NMDS, maximizes variance explanation, and provides confirmation of constrained ordination patterns (Ramette 2007). The observed spatial separation of sediment communities is consistent with the fact that soil environments tend to be much more spatially heterogeneous because of limited

115

transport. In contrast, the temporal separation of PAB communities is likely due to greater spatial homogeneity of the water column (especially in a shallow, lotic system like Chattanooga Creek) which may be more susceptible to seasonal variability. Seasonal variation in PAB communities has been previously observed (Allgaier & Grossart 2006b).

While PAB communities in Chattanooga Creek are phylogenetically distinct compared to the sediment communities, there are functional similarities (dioxygenase genes, mineralization of PAHs) that indicate that PAB are contributing to natural attenuation of both low- and high-molecular weight PAHs.

116

ACKNOWLEDGEMENTS

This work was supported by the Center for Environmental Biotechnology, Research Center of Excellence, University of Tennessee and the Waste Management Research and Education Institute at the University of Tennessee. Special thanks to C. Chewning, F. Menn, J. Easter, J. Rainey, D. Williams, C. Lalande and V. Vulava for assistance in field sampling, and to J. Schweitzer for statistical advice. ESEM images were taken at the electron microscopy facility at the University of Tennessee. DNA sequencing was done by Joseph May at the Molecular Biology Resource Facility at the University of Tennessee.

117

Part IV.

Isolation of pyrene-degrading organisms from geographically disparate PAH-contaminated sediments: evidence for horizontal transfer of nidA and nidA3 pyrene dioxygenase genes between Actinomycetes

118

1. Abstract

Biodegradation of high molecular weight polycyclic aromatic hydrocarbons (PAHs), such as pyrene and benzo[a]pyrene has only been observed in a few genera, namely fast-growing Mycobacterium and Rhodococcus. In Mycobacterium, catabolism of pyrene is initiated by the pyrene ring hydroxylating dioxygenases NidAB or NidA3B3. In this study, pyrene-degrading organisms were isolated from the sediments of two aquatic systems: coal-tar contaminated Chattanooga Creek Superfund Site (TN, USA) and the western basin of Lake Erie. Organisms that could use pyrene as a sole carbon source were phylogenetically identified (16S rDNA sequencing) and probed for the presence of Mycobacteria nidA and nidA3 genes, which encode the large subunits of the initial pyrene dioxygenases. All of the isolates from Lake Erie were most closely related to fast-growing Mycobacterium spp. (order Actinomycetales, family Mycobacteriaceae), a group that has been previously associated with pyrene degradation. These isolates all contained the nidA and nidA3 pyrene dioxygenase genes, as well as a large (220-230 kb) plasmid. In contrast, the isolates from Chattanooga Creek were phylogenetically most closely related to other Actinomycetes; members of the families Microbacteriaceae and Intrasporangiaceae. These isolates had homologous nidA genes, providing the first evidence for horizontal transfer of this pyrene dioxygenase gene between Actinomycetales genera. Some of these Chattanooga Creek isolates also had nidA3. None appeared to have a large plasmid. The high homology of nidA sequences between different genera as well as an analysis of the genomic architecture of pyrene-degrading Mycobacteria reinforces the hypothesis that pyrene dioxygenase genes have been horizontally transferred between genera.

119

2. Introduction

High molecular weight polycyclic aromatic hydrocarbons (HMW PAHs), comprised of at least three fused aromatic rings, tend to be resistant to microbial biodegradation because of their hydrophobicity and chemical stability. As such, these compounds are generally recalcitrant in soils and sediments (Cerniglia 1992) which is of concern because of their documented toxicity, mutagenicity and carcinogenicity, as well as their association with inflammatory vascular responses (Kim et al. 2005a). Several microorganisms have, however, been identified that can use HMW PAHs such as pyrene as a sole carbon source. Most of these have been fast-growing Mycobacterium spp., isolated from both contaminated and uncontaminated freshwater sediments at geographically disparate locations (Table I-2).

The primary PAH catabolic degradation pathway in aerobic bacteria involves an initial dihydroxylation of the aromatic ring, followed by subsequent ring cleavage. Molecular methods probing biodegradative populations and organisms often target the α-subunit of the terminal Rieske dioxygenase, as this is the enzyme which is responsible for the initial dihydroxylation and therefore conveys specificity for particular PAH substrates. In pyrene degradation by Mycobacterium, the nidA gene encodes the α- subunit of the terminal dioxygenase which adds hydroxyl groups at C-4 and C-5 positions (Heitkamp et al. 1988b, Khan et al. 2001, Brezna et al. 2003), and has been employed as a biomarker to identify pyrene-degraders and track pyrene-degrading Mycobacterium populations (Margesin et al. 2003, Hall et al. 2005, DeBruyn et al. 2007). More recently, an alternate dioxygenase, NidA3B3, has been identified in Mycobacterium vanbaalenii PYR-1 (Kim et al. 2006) , which performs the initial dihydroxylation step in an alternate detoxification pathway (Kim et al. 2007). It is unknown

120

whether the nidA3B3 genes encoding the alternative dioxygenase gene is present in organisms other than Mycobacterium vanbaalenii.

Previous work has established the prevalence of fast growing Mycobacteria and nidA genotypes in PAH-contaminated sediments using culture-independent approaches (Hall et al. 2005, DeBruyn et al. 2007). The objective of this study is to cultivate pyrene-degrading organisms to provide the link between phylogeny (Mycobacteria) and function (nidA genotypes and pyrene degradation). In particular, this study aims to isolate pyrene-degrading organisms from two historically contaminated locations. The first is Chattanooga Creek, a Superfund site in Chattanooga, Tennessee which is heavily contaminated with coal tar and exhibits high concentrations of PAHs, ranging from approximately 50 - 250 mg/kg total PAHs (Dionisi et al. 2004, DeBruyn et al. 2007). The second is the western basin of Lake Erie, which has a history of mixed contamination, including PAHs (Eadie 1984). Concentrations of PAHs here are lower than Chattanooga Creek, with total PAHs ranging from 1.5 – 5.3 mg/kg (Smirnov et al. 1998).

Freshwater Actinobacteria have been identified in planktonic and sediment communities from diverse geographical locations (Zwart et al. 1998, Glockner et al. 2000, Zwart et al. 2002, Hahn et al. 2003, Allgaier & Grossart 2006a, Briee et al. 2007). 16S rDNA clone libraries made using fast-growing Mycobacterium specific primers (Leys et al. 2005b) have revealed that PAH-contaminated Chattanooga Creek contains a diverse population of fast-growing Mycobacteria, closely related to other described PAH-degrading Mycobacteria (DeBruyn et al. 2007). In Lake Erie, water column bacterioplankton communities were shown to be dominated by Actinobacteria sequences, including several closely related to Mycobacterium hodlerii (Cupp et al.). This evidence, combined with the isolation of PAH-degrading Mycobacteria from other locations, suggests that both

121

Chattanooga Creek and Lake Erie may contain fast-growing Mycobacteria and/or related organisms capable of HMW PAH degradation.

This study tests the hypothesis that Mycobacteria capable of HMW PAH degradation are present in sediments from two geographically disparate locations. To date, the only identified pyrene-degrading organisms have been Mycobacterium and Rhodococcus; this study also addresses the hypotheses that other genera are capable of pyrene degradation and have nidA and/or nidA3 pyrene dioxygenase genes homologous to those identified in M. vanbaalenii PYR-1 (Kim et al. 2007).

122

3. Materials and Methods

3.1 Sampling and sediment enrichments

Sediments (top 5 cm) were collected from Chattanooga Creek from the summer of 2005 from site CF. This site was the most contaminated of the three sampling sites, with total PAH concentration in the upper 5 cm of sediments ranging from 100-200 mg/kg (DeBruyn et al. 2007). Sediments from the long term monitoring station 357 in the center of the western basin Lake Erie were collected in August 2006 using a box core and subcored to collect the top 5 cm of sediments. Total PAHs have been measured for Lake Erie previously, with a concentration at station 357 of 2.59 mg/kg (Smirnov et al. 1998).

For both Chattanooga Creek and Lake Erie samples, 5 g sediment were slurried with about 10 ml sterile H2O and 500 mg/L pyrene and shaken at 30°C. A spray plate method was used to obtain pyrene-degrading isolates (Shuttleworth & Cerniglia 1997): Aliquots from both enrichments were taken after 20, 30 and 60 days, diluted and plated onto Basal Salts minimal media plates that had been coated with pyrene. To coat each plate, pyrene was dissolved in acetone, then sprayed onto the plates using an aerosolizer. Acetone was allowed to dry, and spraying was repeated until an opaque layer of pyrene was visible. Plates were sealed with parafilm to prevent drying and incubated at 30°C. After approximately 7-10 days, colonies which produced a zone of clearing in the surrounding pyrene layer were transferred to to an LB agar plate for further analysis. Single colonies were picked off of the LB plate and transferred back to another pyrene-coated BSM plate to confirm degradation. Isolates capable of pyrene degradation were grown in liquid LB at 30°C for 2-5 days until a thick culture had grown, which was then used for staining, imaging and DNA extraction. Genomic DNA was extracted from the isolates using the Wizard

123

Genomic DNA kit (Promega, Madison, WI) using the manufacturer’s gram positive bacteria protocol.

3.2 Staining and imaging

Acid fast staining of isolates was done to determine the presence of cell wall mycolic acids using the Ziehl-Neelsen (hot) method. Eight of the isolates were imaged using scanning electron microscopy (SEM). The cells were fixed in phosphate buffered 3% gluteraldehyde for 60 minutes, rinsed three time in buffer then post-fixed in phosphate buffered 2% osmium tetroxide for 60 minutes. Cells were then washed in water and allowed to settle for 10 minutes onto a silicon chip (Ted Pella) before dehydration in a graded ethanol series. Following dehydration cells were critical point dried in a LADD critical point dryer. Just before examination cells were sputter coated with gold in an SPI sputter coater. Cells were examined in a Zeiss 1525 scanning electron microscope operating at 3KV. Images were recorded digitally.

3.3 Molecular characterization

Isolates were screened for presence of the nidA gene using primers targeting a conserved 141 bp region of the gene (nidAfw and nidArv, DeBruyn et al. 2007). Genomic DNA was combined with primers and PuReTaq Ready-To-Go PCR beads (GE Healthsciences), and PCR performed using the following thermocycler program: 95°C for 5 min, then 40 cycles of denaturing at 95°C for 1 minute, annealing 56°C for 1 minute, and extension at 72°C for 1 minute, followed by a 10 minute final extension at 72°C. M. flavescens (ATCC 700033) genomic DNA was used as a positive control. PCR products were visualized on a 1% agarose gel stained with ethidium bromide.

124

Isolates were then subjected to subsequent cloning and sequencing of both the nidA gene and 16S rDNA genes. nidAfullfw and nidAfullrv primers (Table IV-1) were used to amplify the full nidA gene (approximately 1365 bp), using the same PCR conditions as above, except with an annealing temperature of 75°C. Universal eubacterial 16S primers 8F and 1392R were used to amplify a section of the 16S rDNA gene, using an annealing temperature of 56°C. PCR products were cloned into PCR4.0 sequencing vector using the TOPO TA kit (Invitrogen, Carlsbad, CA) according to manufacturer’s protocol. Products were sequenced on an ABI 3730 DNA Analyzer (Applied Biosystems, Foster City, CA) at the Molecular Biology Resource Facility, University of Tennessee (Knoxville, TN), using both M13 forward and M13 reverse primers to sequence the entire insert. Alignments and phylogenetic trees were done using MEGA (Molecular Evolutionary Genetics Analysis) software version 3.1 (http://www.megasoftware.net/).

Table IV-1. Primers used in this study. Size refers to approximate size of amplicon in base pairs (bp). Size Primer Target Sequence (5’Æ3’) Reference (bp) name

nidA nidAfw TTCCCGAGTACGAGGGATAC DeBruyn et al. 2007 141 nidA rv TCACGTTGATGAACGACAAA DeBruyn et al. 2007

nidAfull fw ATGACCACCGAAACAACCG This study nidA 1365 nidAfull rv TCAAGCACGCCCGCCGA This study

nidA3 fw CCTGATGCGACGACAATG This study nidA3 1329 nidA3 rv TCATACGTTCTCATCCCTTCTC This study

16S 8F AGAGTTTGATCMTGGCTCAG Lane et al. 1985 1364 rDNA 1392R ACGGGCGGTGTGTACA Lane et al. 1985

125

3.4 PFGE

Pulsed field gel electrophoresis was done on undigested genomic DNA to determine if the isolates carried large plasmids. A method similar to Stinear et al. was followed (Stinear et al. 2005). Briefly, cells grown in liquid culture were inactivated in 70% ethanol for 30 minutes, followed by washing the cell pellet in 1% Triton X-100. Cells were resuspended in TE buffer and mixed with an equal volume of 2% SeaKem Gold low melting temperature agarose at 45°C. Once the agarose had solidified, the cells in the agarose plugs were lysed in a solution of 0.5 M EDTA (pH 8.0), 0.5% Sarkosyl, 2 mg/ml of deoxycholic acid, and 3.33 mg/ml lysozyme for 18 h at 37°C. Plugs were washed once, then incubated in as solution of 0.5 M EDTA (pH 8.0), 0.5% Sarkosyl, and 1 mg/ml of proteinase K for 7 days at 50°C. Seven day incubations have been found to be optimal for recovery of DNA from Mycobacterium spp. (Hughes et al. 2001). After washing in TE, the plugs were inserted into a 1% agarose gel and run on a BioRad CHEF DRII system (BioRad) in 0.5X TBE buffer at 6 V/cm3 with 3 to 15.5 second switch times for 12 hours followed by 15.5 to 60 second switch times for 6 hours. DNA was visualized by staining with 0.5 µg/ml of ethidium bromide.

126

4. Results

4.1 Isolation and identification of pyrene-degrading organisms

After at least a 20 day enrichment in pyrene, colonies that could produce a clear zone in the pyrene-coated minimal media plates were apparent. These clear zones are considered to be the result of localized metabolism, diffusion and solubilization of the substrate (pyrene). Of the organisms that could degrade pyrene on the minimal media plates, 18 (7 from Chattanooga Creek, 11 from Lake Erie) could also been consistently grown on LB and were selected for further characterization. The 11 Lake Erie isolates were all gram positive, acid fast rods; SEM images were acquired for three of the strains (Figure IV-1 through Figure IV-3). Their rod-shaped morphology is similar to other characterized pyrene- degrading Mycobacterium spp. The seven Chattanooga Creek isolates were gram positive, non-acid fast rods; SEM images of five of the strains are shown in Figure IV-4 through Figure IV-8.

Table IV-2 summarizes the characteristics of the pyrene-degrading isolates, including their phylogenetic affiliations (at the family level), whether they were acid fast, and whether they had nidA and nidA3 genes. The isolates were all the order Actinomycetales (Phylum Actinobacteria); Figure IV-9 shows the phylogenetic relationships between the isolates. The isolates from Lake Erie were all acid fast and most were very closely related to other fast growing, pyrene- degrading Mycobacteria, indicating these isolates were all Mycobacterium spp. One Lake Erie isolate, py136, clustered away from the others – it is most closely related to M. smegmatis and M. moriokaense, which are fast-growing pathogenic species, not previously associated with PAH degradation. In contrast, the isolates from Chattanooga Creek were not acid fast, and fell into two different families: Isolates py114, py115, py116, py120, py122 were most closely related to

127

Figure IV-1. SEM images of pyrene-degrading organism PY143, isolated from Lake Erie.

128

Figure IV-2. SEM images of pyrene-degrading organism PY146, isolated from Lake Erie.

129

Figure IV-3. SEM images of pyrene-degrading organism PY138, isolated from Lake Erie.

130

Figure IV-4. SEM images of pyrene-degrading organism PY115, isolated from Chattanooga Creek.

131

Figure IV-5. SEM images of pyrene-degrading organism PY116, isolated from Chattanooga Creek.

132

Figure IV-6. SEM images of pyrene-degrading organism PY120, isolated from Chattanooga Creek.

133

Figure IV-7. SEM images of pyrene-degrading organism PY129, isolated from Chattanooga Creek.

134

Figure IV-8. SEM images of pyrene-degrading organism PY132, isolated from Chattanooga Creek.

135

Table IV-2. Characteristics of pyrene-degrading isolates. Organisms isolated from Chattanooga Creek (CC) or Lake Erie (LE). Closest cultured organisms show RDP seqmatch score (number of unique 7-base oligomers shared divided by the lowest number of unique oligos) (Cole et al. 2007) followed by NCBI BLAST Max identity %. + indicates that the gene was amplified from the isolate and sequenced. – indicates gene could not be amplified. Plas = plasmid size (determined by PFGE) ND = Not detected. Phylogenetic Isolate Acid nid nid Plas From assignment Closest cultured organism name Fast A A3 (kb) (family) Phycicoccus dokdonensis Py114 CC Intrasporangiaceae N + - ND (0.874, 97%) Knoellia subterranean Py115 CC Intrasporangiaceae N + + ND (0.798, 94%) Phycicoccus dokdonensis Py116 CC Intrasporangiaceae N + + ND (0.873, 97%) Phycicoccus dokdonensis Py120 CC Intrasporangiaceae N + - ND (0.869, 97%) Phycicoccus dokdonensis Py122 CC Intrasporangiaceae N + - ND (0.837, 95%) Leifsonia sp. PTX1 Py129 CC Microbacteriaceae N + + ND (0.915, 98%) Leifsonia sp. PTX1 Py132 CC Microbacteriaceae N - - ND (0.952, 99%) Mycobacterium barrassiae Py136 LE Mycobacteriaceae Y + + 230 (0.954, 98%) M. vanbaalenii PYR-I (0.990, 99%) Py137 LE Mycobacteriaceae Y + + 230 M. austroafricanum (0.990, 99%) Mycobacterium vanbaalenii PYR-I Py138 LE Mycobacteriaceae Y + + 230 (0.992, 99%) Mycobacterium vanbaalenii PYR-I Py139 LE Mycobacteriaceae Y + + 230 (0.997, 99%) Mycobacterium austroafricanum Py140 LE Mycobacteriaceae Y + + 230 (0.996, 99%) Mycobacterium austroafricanum Py142 LE Mycobacteriaceae Y + + 230 0.987, 99%) Mycobacterium austroafricanum Py143 LE Mycobacteriaceae Y + + 230 (0.991, 99%) Mycobacterium austroafricanum Py145 LE Mycobacteriaceae Y + + 230 (0.995, 98%) Mycobacterium vanbaalenii PYR-I Py146 LE Mycobacteriaceae Y + + 230 (0.970, 99%) Mycobacterium austroafricanum Py147 LE Mycobacteriaceae Y + + 230 (0.985, 99%) Mycobacterium vanbaalenii PYR-I Py148 LE Mycobacteriaceae Y + + 230 (0.941, 98%)

136

py142-3 py148-1 py138-4 58 py139-2 X84977-Mycobacterium vanbaalenii 83 py146-2 100 py137-1 X93182-Mycobacterium austroafricanum py143-1 82 py140-1 57 Mycobacteriaceae py145-1 py147-2 CP000580-Mycobacterium spJLS 100 CP000384-Mycobacterium spMCS CP000518-Mycobacterium spKMS 100 py136-2 AF456240-Mycobacterium marinum 59 X55599-Mycobacterium gilvum 61 X52933-Mycobacterium fortuitum 88 77 AJ276274-Mycobacterium frederiksbergense D87974-Nocardioides KP7 py115-1 100 py129-1 97 py132-1 100 X77450-Leifsonia aquatica Microbacteriaceae 78 AB016985-Leifsonia xyli Y17239-Microbacterium terregens 95 AB030911-Tetrasphaera elongata 73 AY944176-Terrabacter terrae 95 AJ566282 Intrasporangium calvum 51 EF493848-Phycicoccus aerolatum EF493847-Phycicoccus aerophilum Intrasporangiaceae 100 py122-1 59 py116-1 100 py114-1 py120-1 AJ276351-Bacillus subtilis

0.02

Figure IV-9. Phylogeny of pyrene-degrading isolates. Neighbor joining tree showing the phylogenetic relationships between the pyrene- degrading isolates (16S rDNA, 8-1392). Characterized Actinomycete strains are included for reference; Bacillus subtilis is included as an outgroup. Bootstrap values (% of 1000 repetitions) over 50% are shown on branches.

137

Phycicoccus sp., in the family Intrasporangiaceae. py129 and py132 were most closely related to Leifsonia sp., in the family Microbacteriaceae.

Pulsed field gel electrophoresis of undigested genomic DNA from the isolates revealed that all of the Lake Erie isolates (Mycobacteriaceae) had large plasmids, around 220 – 230 kb in size (Figure IV-10). No plasmids were detected for any of the Chattanooga Creek isolates.

4.2 Catabolic genes nidA and nidA3

All 18 isolates had the pyrene dioxygenase nidA gene, and the full gene (approximately 1365 bp) was amplified and sequenced. The nidA genes from the 18 isolates had 98-99% sequence identity at the nucleotide level. The nidA3 gene, an alternate pyrene dioxygenase, was found in all the Lake Erie Mycobacteria isolates. However, only three of the Chattanooga Creek isolates (py115, py116, py129) carried nidA3. The sequence similarity between nidA3 genes was 97- 99%. nidA and nidA3 genes from the isolates are highly similar to nidA and nidA3 from characterized Mycobacterium spp. The phylogenetic relationships between these α-subunit genes are shown in Figure IV-11.

4.3 Genomic architecture of five pyrene-degrading Mycobacterium spp.

The five species of fast-growing pyrene-degrading Mycobacteria that have had their genome sequenced (Mycobacterium sp. strain KMS, MCS, JLS, M. gilvum PYR-GCK, M. vanbaalenii PYR-1) show variability in genome architecture in terms of plasmids and localization of pyrene dioxygenase genes (Table IV-3). Strain JLS and M. vanbaalenii have no plasmids, MCS has one large linear plasmid, KMS has two large circular plasmids, and M. gilvum has two small

138

Figure IV-10. Pulse field gel electrophoresis gels of undigested genomic DNA from pyrene degrading isolates.

139

nidA-py137 nidA AY330102 MycobacteriumMCS nidA-py140 nidA AF249301 MycobacteriumPYR1 nidA AY330100 MycobacteriumKMS nidA-py143 nidA DQ537942 M gilvumCZH101 nidA AF548343 M flavescensPYR GCK nidA-py115 nidA-py139 nidA-py145 nidA-py142 nidA-py138 nidA-py120 nidA-py116 nidA-py129 56 nidA AY330098 MycobacteriumJLS nidA-py148 64 nidA-py146 60 nidA-py147 nidA-py122 nidA-py114 99 54 nidA-py136 nidA AF548347 M gilvumBB1 nidA AF546904 MycobacteriumS65 pdoA1 AJ494745 MycobacteriumPY1 pdoA AF546905 MycobacteriumS65 64 nidA3-py140 nidA3-py138 nidA3-py146 nidA3-py136 nidA3-py137 nidA3 DQ028634 M vanbaaleniiPYR1 nidA3-py147 nidA3-py115 nidA3-py148 nidA3-py142 nidA3-py139 nidA3-py143

95 nidA3-py129 nidA3-py116 99 nidA3-py145 93 pdoA DQ118530 TerrabacterHH4 phdA AB017794 NocardioidesKP7 83 53 pdoA2 AJ494743 Mycobacterium6PY1 narAa AJ401612 Rhodococcus1BN 99 narAa AY392423 RhodococcusP400 phtA1 AB084235 TerrabacterDBF63

0.1 Figure IV-11. Phylogenetic relationships of pyrene dioxygenase large subunits. Neighbor joining tree showing the phylogenetic relationships between pyrene dioxygenase genes (nidA and nidA3) amplified from pyrene-degrading organisms isolated from Chattanooga Creek (▲) and Lake Erie (●). Translated amino acid sequences were used for the alignment, and characterized dioxygenase genes from Mycobacterium and Rhodococcus are included for reference (□); a phthalate dioxygenase gene (phtA1) from Terrabacter is included as an outgroup. Bootstrap values (% of 1000 repetitions) over 50% are shown on branches. 140

Table IV-3. Genomic characteristics of five pyrene-degrading Mycobacterium spp. Refseq and Genbank are NCBI accession numbers. + = one copy of gene, ++ = 2 copies of gene

Chromosome/ Refseq Size nid nid Mycobacterium GC plasmid Genbank (kb) A A3 Mycobacterium NC_008726 vanbaalenii Chromosome CP000511 6,491 67% + + PYR-1 NC_009338 Chromosome 5,619 67% ++ + CP000656 Plasmid NC_009339 Mycobacterium 16 65% pMFLV03 CP000657 gilvum PYR- Plasmid NC_009340 GCK 25 64% pMFLV02 CP000658 Plasmid NC_009341 321 65% pMFLV01 CP000659 Mycobacterium NC_09077 Chromosome 6,048 68% ++ + sp. JLS CP000580 NC_008146 Chromosome 5,705 68% + + Mycobacterium CP000384 sp. MCS NC_008147 Plasmid1 215 66% + CP000385 NC_008705 Chromosome 5,737 68% + + CP000518 Mycobacterium Plasmid NC_008703 302 65% + + sp. KMS pMKMS01 CP000519 Plasmid NC_008704 216 66% + pMKMS02 CP000520

141 circular plasmids plus one large linear. The location of the catabolic genes nidA and nidA3, as determined by sequence analysis, is just as varied: nidA and nidA3 are chromosomal in all five species. Two copies of nidA are in the chromosomes of M. gilvum PYR-GCK and strain JLS. One copy of nidA exists on a plasmid in KMS. nidA3 exists on the plasmid of MCS and on both plasmids of KMS. This variation between closely related species suggests a certain amount of intracellular gene shuffling and/or multiple horizontal transfer events have occurred.

The organization of pyrene dioxygenase gene clusters is also indicative of probable horizontal transfer events (Figure IV-12). In all 5 species, integrase and transposase genes can be found in the regions upstream and downstream of nidA and nidA3. Associated with these terminal dioxygenase genes are the requisite dehydrogenase and ferredoxin genes that encode the electron transport chain components of the dioxygenase. They are also associated with multiple copies of other aromatic ring-hydroxylating dioxygenase genes, identifiable on the basis of a conserved Rieske iron-sulfur center motif. The large (α) subunits encoded by these genes range from < 41% to 71% amino acid similarity to nidA from M. vanbaalenii PYR-1.

The inference of horizontal transfer is further strengthened by an analysis of GC content of these regions. The larger gene clusters containing nidA3 shown in Figure IV-12 range from 62.2 – 63.0 % GC; three of the smaller chromosomal clusters ranged from 63.9 – 65% GC. This is notable lower than the GC content of the chromosomes (67-68%) and the plasmids (65% - 66%). Plots of the genomic GC content of Mycobacterium sp. KMS illustrate the low GC of these gene clusters compared to surrounding regions (Figure IV-13). This pattern is also evident for the gene clusters from the other four Mycobacterium spp. (data not shown).

142

Figure IV-12. Gene organization for 5 strains of pyrene-degrading Mycobacteria. nidAB and nidA3B3 genes are in blue and green, respectively. Mobile genetic elements (integrase or transposase genes) are in pink. Dehydrogenases and ferredoxin genes are in brown and black, respectively. Other aromatic ring- hydroxylating dioxygenases (ARHDO) are in light purple, genes for the α subunits include % amino acid similarity to nidA from M. vanbaalenii PYR-1. Intra- or extradiol aromatic ring cleavage dioxygenases (ARCDO) are in dark purple.

143

144

Figure IV-13. GC content (%) of Mycobacterium sp. KMS chromosome (outer circle) and plasmids. Regions containing catabolic genes nidA and nidA3 are indicated by blue lines.

145

It is of particular interest to note the difference in organization between the two plasmids of Mycobacterium sp. KMS. Plasmid pMKMS01 (a 302 kb plasmid) has nidA and nidA3 arranged in a manner similar to that of the chromosomal arrangement in KMS and the other species. In contrast, plasmid pMKMS02 has only nidA3 and the gene arrangement is most similar to that observed on plasmid1 of Mycobacterium sp. MCS. These plasmids are also similar in size (pMKMS01 is 216 kb, plasmid1 is 215 kb), GC content (both 66%), and have a high degree of homologous genes (Figure IV-14), implicating a shared evolutionary origin.

Figure IV-14. Comparison of Mycobacterium sp. strain MCS plasmid 1 and Mycobacterium sp. strain KMS plasmid pMKMS02 genes. Red indicate homologous regions, blue are regions which are homologous but inverted, yellow indicates the homologous region containing the pyrene catabolic genes (nidA3). Alignment was done using blastn, graphic was generated by ACT (Artemis Comparison Tool).

146

5. Discussion

Pyrene-degrading isolates from Lake Erie were all Mycobacteria (order Actinomycetales, suborder Corynebacterineae, family Mycobacteriaceae), closely related to Mycobacterium vanbaalenii PYR-1, a strain which has been well- characterized in terms of pyrene degradation capabilities, genetics, and catabolic pathways (Brezna et al. 2003, Brezna et al. 2006, Kim et al. 2006, Kim et al. 2007). Like M. vanbaalenii and the other pyrene-degrading Mycobacteria which have had their genome sequenced, these strains all had highly homologous nidA and nidA3 genes.

Pyrene-degrading isolates obtained from Chattanooga Creek were more phylogenetically related to organisms in the families Intrasporangiaceae and Microbacteriaceae (order Actinomycetales, suborder Micrococcineae). Phylogenetically similar organisms have previously been isolated from activated sludge (e.g. Tetraspheara spp.) (Maszenan et al. 2000): considering the mixed containment profile of Chattanooga Creek, it is not surprising to recover organisms belonging to these families. This is, however, the first study to demonstrate that members of the Intrasporangiaceae and Microbacteriaceae families degrade pyrene and carry nidA (and some, nidA3) genes encoding pyrene dioxygenases.

Xenobiotic degradation capabilities have been observed in other members of the family Intrasporangiaceae, which includes the genera Intrasporangium, Janibacter, Knoellia, Terrabacter, Tetrasphaera, Serinicoccus and others. Intrasporangium spp. have been associated with a naphthalene-degrading consortium (Yu & Chu 2005). A varietyof Janibacter spp. have been associated with biodegradation: Several can grow on dibenzofuran as a sole carbon source (Noumura et al. 2004, Jin et al. 2006), others have been shown to degrade trichloroethylene (Imamura et al. 2000), monochlorinated dibenzo-p-dioxin (Iwai 147

et al. 2005), and PCBs (Sierra et al. 2003); one Janibacter strain was able to grow on fluorene and dibenzothiophene as sole sources of carbon and cometabolize some PAHs (dibenzo-p-dioxin, phenanthrene, and anthracene) (Yamazoe et al. 2004). Terrabacter spp. have also been associated with PAH degradation (specifically fluorene) (Habe et al. 2005), and have dioxygenase genes similar to nidA3 (Zhou et al. 2006). Some Terrabacter strains also metabolize phthalate (Habe et al. 2003), dibenzofuran (Kasuga et al. 1997), chlorinated dioxins (Habe et al. 2002), and 1,1-dichloro-2,2-bis(4-chlorophenyl)ethylene (DDE)(Aislabie et al. 1999).

Two of the isolates from Chattanooga Creek were phylogenetically grouped within the family Microbacteriacea which includes the genera Microbacterium, Agromyces, Clavibacter, Curtobacterium, Leifsonia and others. The isolates were most similar to other Leifsonia spp., a recently reclassified genus of actinomycetes (Suzuki et al. 1999). This genus includes seven described species, isolated from a variety of environments, including soil, plants and water (Janssen et al. 2002).

The nidA and nidA3 genes from all the isolates show high homology to each other and cultured pyrene-degrading Mycobacterium spp. Highly homologous nidA and nidA3 genes in disparate geographical locations (Chattanooga Creek and Lake Erie) and in phylogenetically disparate hosts (from families Mycobacteriaceae, Intrasporangiaceae, and Microbacteriaceae) provides strong evidence for recent horizontal transfer of nidA and nidA3 genes. nidA3 is a novel pyrene dioxygenase identified by Kim et al. (2006) involved an alternate pyrene detoxification pathway. This gene appears in all five of the published fast-growing, pyrene degrading Mycobacterium genomes that are now publically available, and this is the first study to show that nidA3, like nidA, is not limited to Mycobacteria. nidA3 was not found in all nidA-containing isolates indicating that this gene can be horizontally transferred independently of nidA. This is corroborated by the

148 presence of integrases and transposase genes between nidA and nidA3 regions in three of the five Mycobacterium spp. genomes examined.

Horizontal gene transfer is an important process in adaptation and evolution of microbial communities. Mobile genetic elements can include plasmids, transposons, integrons, genomic islands and phage (Osborn & Boltner 2002), and has been shown to play an integral role in evolution of bacterial pathways. In particular, these elements have contributed to development of xenobiotic catabolism pathways. Evidence for a number of mobile genetic elements containing catabolic genes has been extensively reviewed elsewhere (Sayler et al. 1990, Tan 1999, Tsuda et al. 1999, Habe & Omori 2003, Top & Springael 2003, van der Meer & Sentchilo 2003, Nojiri et al. 2004, Springael & Top 2004, Phale et al. 2007). There is evidence that some of these mobile catabolic elements can be transferred between genera: Pseudomonas plasmid IncP-1 pJP4 (2,4-D degradative genes) was found to be conjugatively transferred to α, β and γ proteobacteria from an activated sludge community (Bathe et al. 2004). In a similar study, the Pseudomonas plasmid encoding carbazole degradative genes (IncP-7) was shown to transfer to Stenotrophomonas in a natural aquatic setting (Shintani et al. 2007). Both Alcaligenes and Pseudomonas strains were found to have closely related 2,4-D biodegradative plasmids (Ka & Tiedje 1994).

All eleven of the Lake Erie isolates had large plasmids that were approximately 220 – 230kb. In the five Mycobacterium spp. genomes examined, three had plasmids, and pyrene dioxygenase genes were found on the plasmids of two. Two of the species had duplicate copies of nidA on the chromosome. This varied arrangement suggests intracellular rearrangement or multiple transfer and acquisition events and has also be observed for catabolic genes in other bacteria: Pseudomonas naphthalene dioxygenase genes are located on the chromosome in some strains and plasmids in others (Herrick et al. 1997, Izmalkova et al. 2005) .

149

Atrazine degradation genes have also been observed on the chromosomes and/or plasmids of atrazine degrading organisms (Devers et al. 2007).

The arrangement of the gene clusters in the five pyrene-degrading Mycobacterium spp. whose genome sequences are currently available presents additional evidence for horizontal gene transfer events in the evolution of pyrene degradation genes and pathways. Previous analysis of whole chromosome restriction patterns as well as sequences of particular genomic regions have revealed evolutionary relationships between these five strains (Miller et al. 2007). Here, a look at the overall genetic architecture of the region provides more information, most notably that nidA and nidA3 genes are in regions of multiple aromatic dioxygenase genes bounded by mobile genetic elements, such as transposases and integrases. This close association of catabolic genes with transposases and integrases has been observed for PAH dioxygenase genes, as well as other xenobiotic degradative genes, in other species (Zhou et al. 2006, Devers et al. 2007).

Duplication of nidA occurred in two of the species, and similarity of arrangement and GC content between chromosome and plasmid clusters further suggests high mobility of these regions. Integration of transmissible catabolic plasmids into genomes has been demonstrated for 2,4-D degradation gene- containing plasmids in Alcaligenes paradoxus (Ka & Tiedje 1994). Rhodococcus spp. have also been observed to have numerous dioxygenase genes, large plasmids and a large genome prone to hyperrecombination, contributing to their metabolic versatility (Larkin et al. 2005).

This study has demonstrated pyrene degradation by organisms in the families Intrasporangiaceae and Microbacteriaceae and provides the first evidence for horizontal transfer of the pyrene dioxygenase genes nidA and nidA3 between genera of Actinomycetes.

150

ACKNOWLEDGEMENTS

This work was supported by the Center for Environmental Biotechnology, Research Center of Excellence, University of Tennessee, and the Waste Management Research and Education Institute at the University of Tennessee. Special thanks to C. Chewning, J. Easter, D. Williams, and C. Lalande for assistance in Chattanooga Creek sampling. Lake Erie samples were graciously collected by J. M. Rinta-Kanto and the crew of the CCGS Limnos. DNA sequencing was done by Joseph May at the Molecular Biology Resource Facility at the University of Tennessee. SEM images were taken by John R. Dunlap at the Electron Microscopy Facility at the University of Tennessee.

151

Part V.

PAH biodegradative genotypes in Lake Erie sediments: evidence for broad geographical distribution of pyrene-degrading Mycobacteria

152

1. Abstract

Despite a long history of contamination of Lake Erie sediments, little work has been done to understand the potential for PAH biodegradation by indigenous microbial communities. Pyrene-degrading Mycobacterium are prevalent in many polycyclic aromatic hydrocarbon (PAH)-contaminated fresh water sediments, and are of interest for their ability to degrade environmentally-recalcitrant high molecular weight PAHs. This study tested the hypothesis that pyrene-degrading Mycobacteria are prevalent in Lake Erie; an additional aim was to gain a baseline picture of the sediment microbial communities through construction and sequencing of a 16S rDNA clone library. The biodegradation potential of indigenous Lake Erie Mycobacterium populations was assessed through quantification of pyrene dioxygenase genes (nidA) and Mycobacterium biomass (16S rDNA genes) using quantitative real time PCR. nidA was detected at all seven sampling sites across Lake Erie, with abundances ranging from 2.09 to 70.4 x106 copies per gram sediment, with the highest abundances at the site with the greatest concentration of pyrene (Cleveland Harbor). This is in contrast to naphthalene dioxygenase genes: proteobacteria dominated the Lake Erie sediment 16S rDNA clone libraries (>50% of clones), however nahAc (from γ- proteobacteria) was not detected anywhere, and nagAc (from β-proteobacteria) was detected only at the most contaminated site (Cleveland Harbor). The prevalence of Mycobacterium nidA genotypes corroborate previous studies indicating that PAH-degrading Mycobacterium have a cosmopolitan distribution and suggests they play an important but overlooked role in natural attenuation and cycling of PAHs in Lake Erie.

153

2. Introduction

Lake Erie is the smallest of the Laurentian Great Lakes, and the most impacted by human activities. Lake Erie sediments have been found to contain a mix of persistent organic contaminants, including polycyclic aromatic hydrocarbons (PAHs) (Eadie 1984). PAHs are the result of incomplete combustion of organic material, and are often associated with industrial activity. Not surprisingly, PAH concentrations in Lake Erie tend to be highest in sediments near major cities (Detroit, Cleveland, and Buffalo), including sediments of Detroit River, which provides the majority of water inflow (Fallon & Horvath 1985, Drouillard et al. 2006). Isotopic patterns of PAHs indicate they are primarily fluvially deposited, then transported around the lake via circulation (Smirnov et al. 1998). Secondary sources of PAHs include deposition of contaminated dust and precipitation (Cortes et al. 2000, Buckley et al. 2004, Sun et al. 2006).

PAHs are of concern because of their toxic, mutagenic and carcinogenic effects on organisms. Benthic invertebrates in the Great Lakes have been found to contain PAHs (Gewurtz et al. 2000, Marvin et al. 2000, Gewurtz et al. 2003, Pickard et al. 2006). This indicates that these contaminants are bioavailable; there is also evidence that PAHs may have toxicological impacts on higher trophic levels (i.e. fish) in Lake Erie and its tributaries (Metcalfe et al. 1997, Baumann 1998, Yang & Baumann 2006).

Despite the presence of measurable concentrations of PAHs and their toxic effects on aquatic organisms, little work has been done to examine the role of microbial PAH biodegradation in Lake Erie sediments. The long history of PAH deposition will have resulted in microbial communities adapted to exploit these as a carbon source. As aerobic biodegradation is the primary degradation fate pathway of PAHs from sediments (Mueller et al. 1989), it is very likely that these indigenous

154 communities are playing an important but overlooked role in PAH attenuation and contaminant cycling. Lake Erie sediment microbial communities have been shown to degrade simple aromatics (Myers et al. 1994) and PCBs (Hoostal et al. 2002). PAH-biodegrading organisms have been previously isolated from the Detroit River and Lake Superior, indicating that indigenous communities have the capacity for PAH degradation (McNally et al. 1998), however these isolates were not identified taxonomically or characterized in terms of functional (PAH- degradation) genotypes. This study is the first, to our knowledge, to examine PAH-degradation potential in Lake Erie sediment communities using molecular approaches.

Aerobic PAH catabolism in bacteria is initiated by substrate-specific ring- hydroxylating dioxygenases, which catalyze the addition of hydroxyl groups, destabilizing the aromatic ring for further nucleophilic attach. In other PAH- contaminated sediments, detection and quantification of the genes encoding the large (α) subunits of these dioxygenases has provided valuable information regarding genotypic potential for PAH mineralization by microbial populations (Laurie & Lloyd-Jones 2000, Baldwin et al. 2003, Watanabe & Hamamura 2003, Dionisi et al. 2004, Nyyssönen et al. 2006, DeBruyn et al. 2007).

In an aged-contamination setting such as Lake Erie, high molecular weight (HMW) PAHs, such as pyrene and benzo[a]pyrene can be recalcitrant (Cerniglia 1992). Fast-growing Mycobacteria have been isolated from contaminated sediments which can degrade these HMW PAHs (Heitkamp & Cerniglia 1988). Pyrene-degrading Mycobacteria have been isolated from a variety of locations, suggesting a cosmopolitan distribution (Miller et al. 2004, Kim et al. 2005b, Zhou et al. 2006). This study aims to test the hypothesis that microbial communities in Lake Erie sediments have the genotypic potential for PAH degradation, with particular emphasis on determining the presence of HMW PAH-degrading Mycobacteria. While some work has focused on the microbial community

155

dynamics in the water column of Lake Erie (DeBruyn et al. 2004, Wilhelm et al. 2006, Cupp et al. 2008), limited information is currently available regarding the microbial communities in the sediments, so an additional objective of this study was to gain a basic picture of the indigenous community through sequencing of a 16S rDNA clone library. This is, to our knowledge, the first large clone library constructed for Lake Erie sediments.

156

3. Methods

3.1 Sampling and DNA extraction

Samples were collected from seven stations on Lake Erie (Figure V-1) aboard the CCGS Limnos in August 2006. A CTD profiler was used to collect temperature and oxygen concentrations. A box core sampler was used to grab bottom sediments at each station, and box cores were subsequently subcored. The top 4 cm of each sub core was used for DNA extractions. Total DNA was purified in triplicate from triplicate sediment cores (i.e. nine DNA extractions per station) using the FastDNA SPIN kit for soil (Qbiogene, Morgan Irvine, CA) with minor modifications as described previously (Dionisi et al. 2004). To determine if there is variation in gene quantities within the top 4 cm, one subcore from each site was sectioned into 0-2 cm and 2-4 cm depths, with these half cores extracted in triplicate as above. The DNA from these sectioned cores was used as templates for three of the real time PCR assays.

3.2 PAH quantification

Figure V-1. Sampling sites on Lake Erie.

method 8270D. Sediment dry weight was determined following baking of 5 g of wet soil at > 80° C for at least 48 hours.

3.3 16S clone library

A 16S rDNA clone library was constructed using DNA extracted from station 84 sediment. 16S genes were amplified using universal eubacterial primers 519F (5’-CAGCAGCCGCGGTAATAC) and 926R (5’- CGTCAATTCCTTTRACTTT) (Lane et al. 1985), and the following thermocycler program: Denaturing at 95°C for 5 minutes, followed by 40 cycles of 95°C for 30 seconds, 45°C for 30 seconds, and 72°C for 30 seconds, and a final extension at 72°C for 10 minutes.

158

PCR products were cloned into a TOPO-TA-PCR4 vector (Invitrogen, Carlsbad, CA). Sequencing was done by the Clemson University Genomics Institute. Alignments and phylogenetic trees were done using MEGA (Molecular Evolutionary Genetics Analysis) software version 3.1 (http://www.megasoftware.net/). Trees were all built using a Kimura 2-parameter distance model (Kimura 1980) and neighbor-joining clustering method with bootstrap analysis. Phylogenetic assignment was done using the Ribosomal Database Project II (release 9.59) (Cole et al. 2007). Phylotype grouping for rarefaction analysis was done using FastGroupII (Yu et al. 2006).

3.4 Real-time PCR

To quantify abundance of PAH catabolic genes in Lake Erie sediments, TaqMan probe quantitative PCR assays targeting genes that encode the large (α) subunit of the terminal dioxygenase, responsible for the initial dihydroxylation of the aromatic ring. The gene targets included pyrene dioxygenase genes (nidA from Mycobacterium), and naphthalene dioxygenase genes (nahAc from γ- proteobacteria and nagAc from β-proteobacteria) and assays were performed as described previously (Dionisi et al. 2004, DeBruyn et al. 2007). Statistical comparisons of gene abundance were done in NCSS (Number Cruncher Statistical Systems, Kaysville, UT) using a nonparametric one-way ANOVA on the ranked data, and Kruskal-Wallis Multiple Comparison Z-test with Bonferroni adjustment.

To check for the presence of PCR inhibitors in the Lake Erie DNA extractions, one replicate from each station was diluted 10 and 100 times (i.e. 2.5 ng and 0.25 ng DNA template per PCR, respectively). nidA copies were quantified using the real time PCR assay.

159

4. Results

4.1 CTD data and PAH concentrations

Water column temperature and oxygen concentrations are shown in Figure V-2. As the western basin of Lake Erie is shallow (8-10 m), it remains mixed (unstratified) in the summer. As a result, the three western basin stations (357, 969, 974) had bottom water oxygen concentrations of at least 4.6 mg/L. The central basin is deeper (20-25 m) and stratifies during the summer, resulting in bottom water oxygen concentrations of > 1.5 mg/L at stations 84 and 961. Station 958, just offshore of Cleveland, OH, is shallow and remains mixed, exhibiting oxygen concentrations > 7.5 mg/L throughout the water column. Similarly, station 308, in Long Point Bay, is also shallow (approximately 9 m deep) and remains mixed, exhibiting temperatures > 23°C throughout the water column (data not shown). Oxygen data was not available for station 308.

Fluoranthene and pyrene were detected at the three western basin stations (357, 974, 969) (Table V-1). Only fluoranthene was measurable at the central basin stations (84 and 961). Cleveland Harbor (958) had the highest level of PAH contamination: 15 of the 16 priority PAHs were measureable with a total PAH concentration of 53.53 mg/kg sediment (dry weight). This is approximately 10 times higher than the most contaminated pelagic stations, measured both in this study and previously (Smirnov et al. 1998). In contrast, none of the 16 priority PAHs were detected in Long Point Bay (308).

4.2 Station 84 clone library

To get a picture of the sediment microbial communities at station 84, a 16S rDNA clone library was constructed. 351 clones were sequenced, and FastGroupII was 160

O2 (mg/L) O2 (mg/L) O2 (mg/L) 024681012140246810121402468101214

0 Station 357 Station 974 Station 969

O2 -2 Temperature

-4

Depth (m) Depth -6

-8

-10 10 12 14 16 18 20 22 24 26 10 12 14 16 18 20 22 24 26 10 12 14 16 18 20 22 24 26 Temperature (°C) Temperature (°C) Temperature (°C) O (mg/L) O2 (mg/L) 2 O2 (mg/L) 02468101214024681012140 2 4 6 8 10 12 14

0 Station 84 Station 961 Station 958

O2 -5 Temperature

-10

Depth (m) Depth -15

-20

-25

10 12 14 16 18 20 22 24 26 10 12 14 16 18 20 22 24 26 10 12 14 16 18 20 22 24 26 Temperature (°C) Temperature (°C) Temperature (°C)

Figure V-2. Water column temperature and oxygen concentrations. Water column temperatures (black) and oxygen concentrations (white) measured in August 2006. The western basin stations 357, 974, 969 are shown in the top panel and central basin stations 84 and 961 as well as station 958 (Cleveland) are shown in the bottom panel.

161

Table V-1. Concentrations of PAHs in Lake Erie sediments. Concentrations are average mg/kg ± standard deviation, n=3. - =Not detected, LMW PAH = low molecular weight PAHs (3 rings or less), HMW PAH = high molecular weight PAH (more than 3 rings). Western basin Central basin Cleve- Long land Point PAH 357 974 969 84 961 958 308 Naphthalene - - - - - 0.22 ± - 0.02 Acenaphthylene - - - - - 0.28 ± - 0.01 Acenaphthene ------Fluorene - - - - - 0.80 ± - 0.07 Phenanthrene - - - - - 2.31 ± - 0.13 Anthracene - - - - - 16.79 ± - 1.17 Fluoranthene 0.31 ± 0.20 ± 0.18 ± 0.10 ± 0.07 ± 4.20 ± - 0.02 0.02 0.01 0.01 0.002 0.36 Pyrene 0.36 ± 0.22 ± 0.20 ± - - 3.86 ± - 0.03 0.01 0.01 0.27 Benzo[a] - - - - - 2.42 ± - anthracene 0.15 Chrysene - - - - - 1.94 ± - 0.11 Benzo[b] - - 0.14 ± - - 1.96 ± - fluoranthene 0.004 0.11 Benzo[k] - - - - - 0.73 ± - fluoranthene 0.05 Benzo[e]pyrene - - - - - 1.13 ± - 0.07 Benzo[a]pyrene - - 5.02 ± - - 15.09 ± - 0.27 0.97 Indeno[1,2,3-c,d] - - - - - 0.81 ± - pyrene 0.06 Dibenzo[a,h] - - - - - 0.22 ± - anthracene 0.02 Benzo[g,h,i] - - - - - 0.81 ± - perylene 0.08 Total PAHs 0.67 0.42 5.54 0.10 0.07 53.53 0 LMW PAHs 0.31 0.20 0.18 0.10 0.07 24.61 0 HMW PAHs 0.36 0.22 5.36 0 0 28.97 0

162

used for rarefaction analysis (Yu et al. 2006). Groups were determined for 80, 85, 90, 95, and 97% sequence identity, which grouped the sequences into 61, 96, 149, 207, and 240 phylotype groups, respectively; rarefaction curves are shown in Figure V-3. The predicted Chao1 values indicate that there is a great deal of diversity in these sediment communities, and this library reflects only a portion of it: at 95% sequence identity, the observed phylotypes (207) make up only 38% of predicted diversity (Chao1 = 549).

The phylogenetic affiliations of the clones sequenced are shown in Figure V-4. Major divisions of bacteria common to sediment and aquatic microbial communities were represented, including Proteobacteria, Actinobacteria, Acidobacteria, Nitrospira, Firmicutes, Bacterioidetes, Planctomycetes, and Cyanobacteria. Phylogenies were determined using RDP (Cole et al. 2007) with a confidence threshold of 80%.

A neighbor-joining phylogenetic tree of the major divisions recovered in the clone library is shown in Figure V-5. Over half (51%) of the clones in the library were proteobacteria. Within the proteobacteria, 30.3% were β-proteobacteria. Some were similar to Rhodocycales, Burkholderiales, and Commamonas, however many fell into clades which were closely related to sequences from other environmental clone libraries which had no cultured representatives (Figure V-6). 27.5% of the proteobacteria clones were γ-proteobacteria, falling into orders Methyococcales, Xanthomonadales, Chromatiales, and Legionellales, also with a large number of unclassified γ-proteobacteria matching only other environmental (uncultured) sequences in the databases (Figure V-7). 10.7 % of the proteobacteria sequences were α-proteobacteria, closely related to Sphingomonas, Methylocycstis, and Rhodobacter (Figure V-8). 15.2% were δ-proteobacteria, falling into two distinct clades: one group was closely related to Geobacter and Desulfobacter, while the other group was more closely related to Bacteriovorax and Desulfovibrio (Figure V-8). 16.3% were unclassified proteobacteria.

163

250 97% identity (Chao1 = 764) 95% identity (Chao1 = 549) 90% identity (Chao1 = 345) 200 85% identity (Chao1 = 171) 80% identity (Chao1 = 118)

150

100 Number of ribotypes 50

0 100 200 300

Number of sequences Figure V-3. Rarefaction analysis of Lake Erie 16S clone library. Curves were calculated for 80, 85, 90, 95, and 97% sequence identity. Chao1 values are listed for each.

164

Verrucomicrobia Cyanobacteria (6) (18) 5% 2% WS3 (2) 1% Nitrospira (11) 3%

Unclassified Bacteroidetes Chloroflexi (3) 1% bacteria (67) 19% (23) 6% Firmicutes (4) 1% OP10 (2) 1% Planctomycetes (5) 1% Actinobacteria (20) 6% Acidobacteria(12) 3%

Proteobacteria (178) 51%

Figure V-4. Phylogenetic classifications of 16S clones recovered from Station 84 sediments. 351 clones were sequenced. Phylogenetic divisions are based on an 80% confidence threshold. The number of sequences for each division are listed in brrackets.

165

c t o s

a m l p

o i r

c lo r h

o . c b s

U e p i p a

n t s

c e a

l r

a c i

s y e

s s

i m f o

i i 3 c e o

r 2 t a d - l e a

t i

3 c u b r

c S a A a E l

a e E

c t S b P

W t

e c

1 d

9 e a

i i a

S - f a 8 i b ri 3 s e s o t 6 a 1 l n c i a - c t b 2 ria n 1 e c d S ct U a ie E d b A if fie s ssi U s cla nc la Un la c ss n ifi U ia ed ter b ac ac b te no ria E ri ya cte S8 a C ba -5 ified OP lass 10 Unc lexi rof -16 Clo ium) ES6 strid A (Clo cidob es eria acter icut bact ia irm ssified F Uncla eria ed bact Al nclassifi pha pr U oteob ES13-11 acteria Bacteroidetes assified) Proteobacteria (uncl Uncl assified bacteria ed) lassifi ES11-2 ia (unc bacter Proteo a eri act ob ote pr elta D s) ad on N m itr ro o fu sp ul ir es a

D )

( s

a i d ri ia r

e a

t r e c n t a te

ob c o c

te a m a o b

r b o p o h o

a t t e e

l t n e t

o a

D r o X r

p (

p a 4 a

t - i r a e 3

1 e B t m S c

a m b a

o G

e t

o r

Figure V-5. Neighbor-joining phylogenetic tree of all 351 16S rDNA clones from sediment microbial community at station 84. Phylogenetic classifications based on 80% confidence level. A euryarchaeota was used as the outgroup. Size of the wedge represents the number of clones recovered within that phylogenetic classification. “Unclassified” groups could not be assigned to a particular phylogenetic group. Clones not falling into a distinguishable clade are represented by their clone name (e.g. ES6-16).

166

57 ES9-6 60 Acidovorax facilis AF078765 ES16-6 Leptothrix mobilis X97071 Variovorax paradoxus AJ420329 ES13-8 Polaromonas naphthalenivorans AY166684 ES18-9 ES2-21 ES10-19 65 ES15-14 Comamonas testosteroni M11224 ES3-16 ES9-10 ES14-11 ES20-8 ES2-8 ES4-6 65 ES4-19 70 ES6-12 ES5-23 ES15-5 ES16-9 ES3-19 99 ES14-8 ES20-5 7760 ES16-4 ES19-13 ES9-15 ES13-4 Azoarcus evansii X77679 Dechloromonas denitrificans AJ318917 ES5-3 90 ES14-2 Burkholderia glathei Y17052 Alcaligenes faecalis AJ242986 Beta proteobacteria 63 Rals tonia s yzygii U28237 72 Oxalobacter formigenes U49757 Limnobacter thiooxidans AJ289885 ES5-1 Sterolibacterium denitrificans AJ306683 ES8-18 ES11-19 ES3-10 86 ES11-11 96 ES14-4 Methylophilus methylotrophus L15475 ES8-22 Nitrosomonas oligotropha AJ298736 ES12-7 61 ES14-3 9376 ES5-18 81 ES5-21 98 ES3-21 ES2-1 ES4-1 ES18-8 ES5-22 58 ES10-7 66 89 ES4-5 ES13-15 ES18-6 ES6-6 ES3-4 98 ES19-9 64 ES3-13 58 ES10-12 ES8-12 ES8-10 56 ES19-6 Gamma proteobacteria (Xanthomonads)

Gamma proteobacteria

Delta proteobacteria

Alpha proteobacteria

Delta proteobacteria (Geobacter)

Unclassified proteobacteria 100 ES12-15 ES13-13 Unclassified proteobacteria 83 Halobacterium salinarum

0.05 Figure V-6. Neighbor-joining phylogenetic tree of proteobacteria 16S sequences from Lake Erie sediments, highlighting the β subdivision. 167

Beta proteobacteria

Xanthomonas pisi Y10758 ES18-14 ES3-5 ES16-10 Gamma proteobacteria (Xanthomonads) 91 ES15-3 58 ES14-10 5857 ES6-10 ES11-3 ES5-8 ES19-14 99 ES13-19 ES9-3 ES6-2 68 ES11-8 99 Legionella oakridgensis X73397 ES13-5 100 ES19-2 ES2-2399 Methylococcus capsulatus X72770 ES5-17 ES12-12 50 Pseudomonas aeruginosa X06684 64 Pseudomonas putida D37923 92 96 Pseudoalteromonas denitrificans X82138 Shewanella denitrificans AJ311964 ES3-24 99 ES2-19 76 ES9-2 ES12-17 87 ES2-3 ES6-7 85 ES11-14 87 54 ES13-17 ES17-19 89 ES14-13 Gamma proteobacteria 98 ES18-13 ES19-15 81 ES8-19 ES11-7 ES9-8 ES18-19 ES16-12 ES17-17 58ES18-15 ES13-14 99 ES2-13 84 ES8-13 65 ES2-14 ES10-8 100 ES12-6 ES8-21 80 ES10-24 ES12-8 73 ES14-20 52 ES11-15 ES11-9 100 ES8-23 ES15-13 ES13-16 66 ES15-16 ES3-22 ES18-2 ES16-19

Delta proteobacteria

Alpha proteobacteria

Delta proteobacteria (Geobacter)

Unclassified proteobacteria 100 ES12-15 ES13-13 Unclassified proteobacteria 83 Halobacterium salinarum

0.05 Figure V-7. Neighbor-joining phylogenetic tree of proteobacteria 16S sequences from Lake Erie sediments, highlighting the γ subdivision.

168

Beta proteobacteria

Gamma proteobacteria (Xanthomonads)

Gamma proteobacteria

98 ES3-9 97 ES20-9 86 ES12-3 ES2-18 ES2-2 Syntrophobacter wolinii X70905 Bacteriovorax litoralis AF084859 ES4-18 100 ES15-1 99 ES9-5100 ES19-12 Delta proteobacteria Desulfovibrio aespoeensis X95230 ES17-11 ES18-1 ES6-4 81 ES11-18 50 ES20-11 ES14-7 52 ES2-12 90 ES16-5 91 ES17-9 ES11-1075 Roseococcus thiosulfatophilus X72908 100 ES6-11 82 ES14-5 Rhodospirillum photometricum AJ222662 ES13-6 Caulobacter fusiformis AJ227759 Rhizobium leguminosarum AY509899 53 ES3-8 Rhodobacter sphaeroides X53853 63 ES6-20 62 ES14-16 ES9-16 ES17-2 ES13-2 ES19-17 Alpha proteobacteria 77 83 ES8-15 Hyphomicrobium denitrificans Y14308 ES15-17 ES11-16 Sphingomonas paucimobilis U20776 ES18-18 ES20-3 85 ES8-20 57 ES17-7 Rhodopseudomonas palustris D25312 Beijerinckia indica subsp. indica M59060 ES6-5 92 ES5-12 52 Methylocystis echinoides AJ458473 96 ES4-386 98 ES13-10 68 ES2-10 ES16-1 74 Geobacter sulfurreducens U13928 ES13-9 Desulfuromonas thiophila Y11560 60 ES20-2 100 ES15-15 97 ES17-3 ES16-18 Delta proteobacteria (Geobacter) 99 ES9-18 Desulfobacter curvatus M34413 88 ES5-15 ES17-8 ES8-2 ES12-2 96 ES8-9 ES10-2 ES10-4 ES17-10 Unclassified proteobacteria 100 ES12-15 ES13-13 Unclassified proteobacteria 83 Halobacterium salinarum 0.05 Figure V-8. Neighbor-joining phylogenetic tree of proteobacteria 16S sequences from Lake Erie sediments, highlighting the α and δ subdivisions.

169

Twenty sequences were recovered that fell into the phylum and class Actinobacteria (Figure V-9). Of these, 10 were in the order Rubrobacterales, which included Rubrobacterineae and Conexibacter. Seven were Actinomycetes, including Actinosynnemataceae, Mycobacteriacea and Frankineae. The two sequences most closely related to Mycobacterium had 98% sequence identity to other fast-growing, environmental Mycobacterium spp. Verrucomicrobia was represented by 18 sequences, falling into subdivision 3, Opitutus, Xiphinematobacteriaceae and Verrucomicrobiaceae (Figure V-10). The clone that was most closely related to a cultured representative (ES4-7) had 97% sequence identity to Opitutus sp. VeSm13.

Other sequences of interest include clones most closely related to candidate divisions WS3 and OP10, monophyletic groups that currently have no cultured representatives. Chloroplast sequences were also recovered: eleven clones had 98-99% identity to Aulacoseira and Thalassiosira spp. chloroplast sequences.

4.3 Quantification of PAH catabolic genes

To check for PCR inhibitors in the Lake Erie DNA extractions, 10 and 100 fold dilutions were used as templates (i.e. 2.5 ng and 0.25 ng DNA template per PCR, respectively). Quantitative PCR on dilutions yielded nidA abundances that were not significantly different from abundances measured using 25 ng per reactions (Student’s t-test, n=3, p > 0.05, Figure V-11), indicating that the DNA extraction method effectively removed PCR inhibitors.

Sectioned cores (0-2 cm and 2-4 cm) were collected to determine if there was variation in genotype abundances at this scale. 16S rDNA, Mycobacteria 16S rDNA and nidA were quantified (Figure V-12). There were no significant differences in abundances of any of the genes between the 0-2 and 2-4 cm

170

55 Cellulomonas cellasea X79459 69 Microbacterium aurum D21340 Intrasporangium calvum AJ566282 81 Terrabacter terrae AY944176 77 Micrococcus lylae X80750 96 Arthrobacter globiformis X80736 Brevibacterium casei X76564 53 Kineosporia aurantiaca X87110 Corynebacterium durum Z97069 Rhodococcus erythropolis X81929 Mycobacterium vanbaalenii X84977 91 ES2-22 99 100 ES10-13 100 ES12-13 67 ES12-20 ES12-1 Geodermatophilus obscurus X92356 ES2-5 58 96 99 ES9-19 Nocardioides jensenii Z78210 ES5-4 Streptomyces arenae AJ399485 Rubrobacter radiotolerans X87134 100 ES4-14 95 ES6-21 86 ES15-6 ES18-5 99 ES6-22 ES19-5 87 ES11-5 55 ES14-14 100 ES8-4 99 ES16-2 ES8-17 ES12-4 100 ES9-13 93 ES13-7 100 ES14-1 ES4-23 98 52 ES2-17 ES10-10 ES3-1 Coriobacterium glomerans X79048 Halobacterium salinarum

0.05

Figure V-9. Phylogenetic neighbor-joining tree of Actinobacteria sequences recovered from Lake Erie. Type strains are included for reference, although most sequences were unclassifiable i.e. no cultured representative

171

94 ES12-14 ES10-3 85 ES17-16 100 86 ES8-8 ES15-18 Rubritalea marina DQ302104 50 Verrucomicrobium spinosum X90515 85 Prosthecobacter vanneervenii U60013 ES9-14 71 ES10-17 100 75 ES10-22 ES4-22 ES9-23 65 73 99 Xiphinematobacter rivesi AF217461 89 ES10-1 ES20-4 74 ES4-15 99 ES10-16 92 ES9-9 ES13-1 62 ES18-3 62 58 90 Verrucomicrobia subdiv3 AY395425 ES4-7 54 100 Opitutus terrae AJ229235 ES16-15 ES19-8 ES4-4 94 100 ES17-20 ES5-19 Verrucomicrobia subdiv5 AY114320 86 ES16-7 73 ES16-14 Halobacterium salinarum

0.05 Figure V-10. Phylogenetic neighbor-joining tree of Verrucomicrobia sequences from Lake Erie. Cultured and uncultured (Verrucomicrobia subdiv3 and subdiv5) Verrucomicrobia sequences are included for reference.

172

105 No dilution (25 ng/rxn) 10 x dilution (2.5 ng/rxn) 100 x dilution (0.25 ng/rxn)

104

103 copiesng / 25 extracted DNA nidA 102

Stn 84 Stn 308 Stn 958 Stn 974 Stn 961 Stn 969 Stn 357 Station

Figure V-11. Dilution series to determine PCR inhibition in Lake Erie sediments. Copies of nidA per 25 ng of extracted DNA at each station, using 25 ng (black circles), 2.5 ng (white triangles) and 0.25 ng (grey squares) of environmental DNA template per reaction. There was no significant differences between the dilutions (Student’s t-test, n=3, p>0.05), indicating that PCR inhibition is not occurring.

173

1012

1011

1010

109

108 0-2 cm 16S 2-4 cm 16S 107 109

108

107 Copies/g dry weight 106 0-2 cm myco16S 2-4 cm myco16S 105 1010 109 108 107 106 0-2 cm nidA 2-4 cm nidA 105 357 974 969 961 84 958 308 Station Figure V-12. Comparison of microbial genotypes in 0-2 cm and 2-4 cm core sections. 16S rDNA, mycobacterium 16S and nidA were quantified in the split cores: 0-2 cm (closed circles), 0-4 cm (open circles). There were no significant differences between the 0-2 cm and 2-4 cm sections of the cores (Student’s T-test, n=9, p>0.01).

174

sections (Student’s T-test, n=9, p>0.01), indicating that the 0-4 cm cores used for the remainder of the study captured a similar community (i.e. there were no major differences in community abundance or structure within the 0-4 cm section in terms of degradation genotypes).

Gene abundances for the seven stations are shown in Figure V-13. Total eubacterial biomass (16S gene copies/g dry weight) did not vary significantly between the pelagic stations (357, 974, 969, 961, 84), ranging from 9.72 x 109 to 3.87 x 1010 copies per gram sediment (dry weight). Eubacterial biomass was, however, significantly lower in Cleveland Harbor (958) and Long Point Bay (308) (Kruskal-Wallis one way ANOVA p < 0.001), with mean abundances of 5.55 ± 4.62 x 107 and 1.53 ± 0.57 x 109 copies/gram dry weight, respectively. Mycobacterium biomass was also measured, using an assay specific to fast- growing Mycobacteria. Stations 357 and 961 had the highest Mycobacterial biomass at 5.86 ± 2.25 x 107 and 8.07 ± 4.69 x 107 copies per gram sediment, respectively. The lowest was observed at station 958, at 4.79 ± 4.30 x 105 copies per gram sediment.

nagAc (from β-proteobacteria) was only detected in Cleveland Harbor (958), with abundances averaging 1.04 ± 0.79 x 104 copies per gram dry weight sediment. All other stations were below detection limit (about 1000 copies per gram sediment). nahAc was also below detection at all stations. The absence of detectable nagAc and nahAc was also observed in sediment samples taken from Lake Erie in July 2004 (data not shown).

In contrast to the low-molecular weight PAH dioxygenases, nidA was detected at all stations in abundances ranging from 2.09 x 106 to 7.04 x 107 copies per gram sediment (Figure V-13). A Kruskal-Wallis one way ANOVA and post hoc multiple comparison Z value test showed that nidA abundances at station 84 and 961 are significantly higher than the rest of the stations (p < 0.001), which do not

175

1e+12 16S rDNA 1e+11 nidA Myco 16S rDNA 1e+10 nagAc

1e+9

1e+8

1e+7

1e+6

1e+5 Copies/g dry weight sediment weight dry Copies/g 1e+4

1e+3 357 974 969 961 84 958 308

Figure V-13. Microbial gene abundances in Lake Erie sediments. Quantities of 16S rDNA (black), pyrene dioxygenase gene nidA (grey), Mycobacteria 16S rDNA (white) and naphthalene dioxygenase gene nagAc (hatched) are shown for each station. nahAc was below detection at all stations; nagAc was below detection at all but 958 (Cleveland harbor).

176 differ significantly from each other. When normalized to Mycobacteria biomass (Mycobacterium 16S copies) or total Eubacterial biomass (eubacterial 16S copies) nidA copies are significantly higher in Cleveland Harbor (Station 958) (Kruskal- Wallis one way ANOVA, p < 0.001). This is due to a reduced abundance of both Mycobacterium and total Eubacterial biomass at this site.

The relationship between nidA copies and Mycobacterial biomass is shown in Figure V-14. There is a positive relationship between the two: where greater Mycobacteria biomass is observed, nidA is enriched, suggesting that Mycobacterium spp. are the primary carriers of nidA genotypes in Lake Erie.

100

10-1

10-2

10-3

10-4 nidA / Total Eubacterial 16S rDNA

10-5 10-4 10-3 10-2 10-1

Mycobacterium 16S / Total Eubacterial 16S rDNA

Figure V-14. Relationship between nidA and Mycobacterial 16S rDNA gene abundances. The positive relationship between nidA gene abundances and fast-growing Mycobacteria biomass (16S rDNA copies) indicates that nidA likely occurs primarily in Mycobacterium spp. 177

5. Discussion

PAH concentrations have been previously measured for the five pelagic stations sampled (974, 969, 357, 961, 84) (Smirnov et al. 1998). Where PAHs were quantified in our samples, concentrations were comparable to those measured previously. For example, fluoranthene concentrations in our samples ranged from 0.18 – 0.31 mg/kg and pyrene from 0.20 – 0.36 mg/kg for the western basin stations; Smirnov et al. measured 0.16 – 0.19 mg/kg and 0.18 – 0.20 mg/kg, repectively. The major difference in this study is that PAHs were not always detected in this study, while Smirnov detected most of the 16 priority PAHs compounds at all five sites; this difference may be due to some attenuation of PAHs over time, or differences in detection sensitivity due to methodology. The persistence of fluoranthene and pyrene without any detectable low molecular weight PAHs (i.e. naphthalene) suggests that these are aged contaminants. The most contaminated site was Cleveland harbor (958), with total PAH concentrations of 54 mg/kg sediment. This contamination is not unexpected, as this site is at the mouth of the Cuyohoga River, an EPA-designated Area of Concern. The high level of contamination at Cleveland Harbor and low level seen in Long Point Bay (308) is consistent with previous measures showing higher levels of contamination near cities and lowest in the northern parts of the lake (Smirnov et al. 1998).

The 16S rDNA clone library representing the microbial community in the sediments at station 84 revealed a great deal of diversity. Phylogenetic groups common to other freshwater sediment habitats were represented in this clone library, including δ-proteobacteria and Acidobacteria (Spring et al. 2000). The largest phylogenetic grouping observed was the β-proteobacteria, representing about 15% of the clones in the library. This coincides with many other freshwater sediment microbial community studies, which report a high percentage of β-

178

proteobacteria in their clone libraries (Spring et al. 2000, Briee et al. 2007). In contrast to other libraries, more γ-proteobacteria were detected here: Our library had about 14%, while other studies typically find 0-9% of clones are γ- proteobacteria (Spring et al. 2000). In addition to the expected taxa in the community, clones representing the novel phylogenetic lineages OP10 and WS3 were also detected. These candidate divisions have no cultured representatives, but appear to be broadly distributed in freshwater environments (Dojka et al. 1998, Stein et al. 2002). This community also contained sequences most closely related to Verrucomicrobia. This too is a group that has few cultured representatives, but is commonly retrieved from sediment aquatic communities (Zwart et al. 1998). The recovery of chloroplast and cyanobacterial sequences in sediment clone libraries is not unexpected: In particular, Aulacoseira chloroplast sequences in our library is consistent with observed increases in Aulacoseira spp. in Lake Erie following Dreissena invasion (Barbiero et al. 2006).

Pyrene dioxygenase nidA genes were detected at all stations sampled. This is consistent with previous observations of consistent nidA genotype abundances in PAH-contaminated sediments (DeBruyn et al. 2007) and suspended particles (Part III). Abundances in Lake Erie ranged from 2.09 x 106 to 7.04 x107 copies per gram sediment. This is slightly higher than abundances observed in Chattanooga Creek sediments (5.69 x 104 - 4.92 x 106 copies per gram sediment) (DeBruyn et al. 2007). These observations, combined with other studies that have detected nidA at a variety of geographical locations (Margesin et al. 2003, Hall et al. 2005, Miller et al. 2007), implies a cosmopolitan distribution of pyrene-degrading organisms (i.e. Mycobacteria) carrying nidA. Coinciding with the detection of nidA pyrene dioxygenase genes across the lake is the recovery of two sequences in the clone library that were most closely related to fast-growing Mycobacterium spp. This indicates that while Mycobacteria populations likely do not dominate numerically in these communities, they are present, and they do have the functional genotypes for contaminant conversion. The low abundance seen in the

179

clone library may be in part due to difficulties in amplifying Actinomycetes using universal primers, which has been observed elsewhere (Farris & Olson 2007). nidA and Mycobacteria biomass were positively related to each other, suggesting that Mycobacterium spp. are the primary carriers of the nidA genotype in Lake Erie. This is in contrast to Chattanooga Creek, where no relationship was observed between the two (Part IV).

Despite the prevalence of γ- and β-proteobacteria in the clone library, the naphthalene dioxygenase genes nahAc and nagAc normally associated with these groups was notably absent. This is also inconsistent with previous attempts to isolate PAH-degrading organisms from the Detroit River, which yielded gram negative organisms only (McNally et al. 1998). nagAc from β-proteobacteria was detected only in Cleveland Harbor (station 958), also the only station with detectable concentrations of low molecular weight PAHs (naphthalene, phenanthrene). Despite the measureable concentrations of low molecular weight PAHs at 958, nahAc was below detection here (and the rest of the stations), and no sequences similar to Pseudomonas spp. were recovered in the clone library from station 84. Much research has been done on PAH-degrading Pseudomonas with the nahAc genotype, and NahA is often considered the “classic” naphthalene dioxygenase. However, recent work on PAH-degrading communities has suggested that this genotype may not be as environmentally relevant as some of the other dioxygenase genes (e.g. nagAc or phnAc) (Laurie & Lloyd-Jones 2000, Rhee et al. 2004, DeBruyn et al. 2007, Gomes et al. 2007).

In summary, the prevalence of fast growing Mycobacteria rDNA sequences and nidA pyene dioxygenase genes at all stations sampled shows a broad distribution of these genotypes and implicates that Mycobacterium spp. are likely playing a role in natural attenuation of high molecular weight PAHs from Lake Erie sediments.

180

ACKNOWLEDGEMENTS

This work was supported by the Center for Environmental Biotechnology, Research Center of Excellence, University of Tennessee, and the Waste Management Research and Education Institute at the University of Tennessee. Lake Erie sediment samples were graciously collected by J. M. Rinta-Kanto and the crew of the CCGS Limnos. DNA sequencing was done by Joseph May at the Molecular Biology Resource Facility at the University of Tennessee and by the Clemson University Genomics Institute (CUGI).

181

Part VI.

Summary and Conclusions

182

1. Microbial ecology of PAH-degrading organisms

Microbial ecologists strive to determine important functional groups in complex microbial communities. In the case of biodegradation of PAHs, these functions contribute to the attenuation of these compounds and remediation of contaminated environments. Therefore an understanding of the functional ecology of PAH- degrading microbial communities is integral to determining the environmental fate of contaminants as well as making informed remediation and management decisions.

The work presented in this dissertation has employed both culture-independent molecular tools as well as more traditional cultivation of microorganisms to examine the dynamics of indigenous PAH-degrading populations in situ. In particular, it has highlighted fast-growing Mycobacterium spp. which are capable of using high molecular weight PAHs, such as pyrene and benzo[a]pyrene, as a sole carbon source. Mycobacterium populations and pyrene degradation genotypes were observed to have consistent abundances across different environments, indicating they play an important role in attenuation of high molecular weight PAHs from historically contaminated aquatic systems.

183

2. Naphthalene biodegradative genotypes

In this examination of Mycobacterium and related pyrene-degrading populations, abundance of naphthalene degradation genotypes were also monitored. Some of the best characterized PAH-degrading organisms are Pseudomonas, originally isolated on naphthalene, and their naphthalene dioxygenase, NahAc, is arguably the best studied PAH catabolism enzyme. As a result, many studies have focused on naphthalene dioxygenase genotypes as a biomarker for PAH degradation. In sites such as Chattanooga Creek and Lake Erie, the contaminants are typically aged, and significant attenuation of low molecular weight PAHs has occurred. In Chattanooga Creek, Pseudomonas naphthalene dioxygenase genes were only sporadically detected, and did not relate significantly to activity or naphthalene concentrations. In Lake Erie, this genotype was not detected at any of the seven stations. This suggests that Pseudomonas populations may not be present and/or functionally relevant in all PAH-contaminated settings, and that nahAc may not as universal of a PAH-degradation biomarker as it was once thought. This also highlights the importance of using culture-independent approaches to understanding the dynamics of PAH-degrading microbial populations in situ, rather than relying solely on lab-based studies.

In Chattanooga Creek, it had previously been shown that Ralstonia-like nagAc genotypes were prevalent and related to naphthalene concentrations (Dionisi et al. 2004), and this dissertation work confirms that observation: nagAc genotypes were detected at all sites. Further confirmation comes from observations from Lake Erie, where this genotype was detected at the one site where low molecular weight PAHs (naphthalene, phenanthrene) were measurable. Other studies have also demonstrated enrichment of nagAc genotypes in naphthalene-contaminated soils and sediments. For example, in a survey of dioxygenase diversity during enrichment, the indigenous community in PAH-contaminated soil had a much

184

higher abundance of Comamonas-like nagAc compared to Pseudomonas nahAc (Ní Chadhain et al. 2006). Sequencing of naphthalene dioxygenase genes from contaminated mangrove sediments revealed only one sequence similar to cultured organisms, and that was nagAc (Gomes et al. 2007). β-proteobacteria are typically dominant in freshwater sediment microbial communities and these observations suggest that β-proteobacteria, such as Ralstonia or Commomonas, play a key role in attenuation of low molecular weight PAHs.

This work has highlighted the dynamics of three PAH dioxygenase genotypes, however it should be kept in mind that studies have indicated that there is a great diversity of PAH dioxygenase genes which has yet to be uncovered (Ahn et al. 1999, Taylor et al. 2002, Ní Chadhain et al. 2006, Gomes et al. 2007). The real- time PCR assays used for this work is quite stringent, targeting sequences with ≥ 98% sequence identity. While this is ideal for quantifying specific genotypes, it certainly excludes related dioxygenase genes. Exemplifying this fact was the isolation of one pyrene-degrading organism that did not appear to have either the nidA or the nidA3 genes, indicating that there is likely other initial pyrene dioxygenase genes in this organism. As more genome sequences become available, high throughput methods such as functional gene arrays (e.g. Rhee et al. 2004) will provide a more comprehensive picture of community dynamics during PAH biodegradation.

185

3. Evolution and biogeography of pyrene degrading organisms

The observations presented in this dissertation of a persistent population of pyrene-degrading organisms over time and space suggests a cosmopolitan biogeographic distribution of these organisms. This work demonstrated their presence in PAH-contaminated aquatic sediment and suspended particles; other studies have also isolated similar organisms from both contaminated and uncontaminated soils and sediments (Table I-2). These observations indicate that pyrene-degraders are widely distributed in the environment. The consistency of detection of pyrene degrading population is particularly interesting in contrast to the other two naphthalene genotypes (nahAc and nagAc) monitored throughout these studies which showed greater variability in abundances.

Examination of the genomes of Mycobacteria indicates that the evolution of pyrene catabolic pathways has involved extensive horizontal gene transfer and intracellular gene shuffling. This has been observed in other organisms for other catabolic genes, and this work now presents evidence for horizontal gene transfer in Mycobacteria. The first pyrene degrading organism, Mycobacterium vanbaalenii PYR-1, was isolated about 20 years ago, relatively recently compared to other biodegradative strains. Now, this work has demonstrated that other genera can use pyrene as a sole carbon source, most likely evolving this metabolic capability through horizontal transfer of pyrene catabolic genes. Future work will likely continue to reveal the importance of these organisms to contaminant cycling and attenuation in a variety of contaminated environments, particularly those which have a long history of contamination and have aged, recalcitrant, higher molecular weight compounds.

186

REFERENCES

187

Acinas SG, Sarma-Rupavtarm R, Klepac-Ceraj V, Polz MF (2005) PCR-induced sequence artifacts and bias: Insights from comparison of two 16S rRNA clone libraries constructed from the same sample. Applied and Environmental Microbiology 71:8966-8969 Adekambi T, Drancourt M (2004) Dissection of phylogenetic relationships among 19 rapidly growing Mycobacterium species by 16S rRNA, hsp65, sodA, recA and rpoB gene sequencing. International Journal of Systematic and Evolutionary Microbiology 54:2095-2105 Ahn Y, Sanseverino J, Sayler GS (1999) Analyses of polycyclic aromatic hydrocarbon-degrading bacteria isolated from contaminated soils. Biodegradation 10:149-157 Aislabie J, Davison AD, Boul HL, Franzmann PD, Jardine DR, Karuso P (1999) Isolation of Terrabacter sp. strain DDE-1, which metabolizes 1,1- dichloro-2,2-bis(4-chlorophenyl)ethylene when induced with biphenyl. Applied and Environmental Microbiology 65:5607-5611 Alexander M (1999) Biodegradation and Bioremediation. Academic Press, San Diego Allgaier M, Bruckner S, Jaspers E, Grossart H-P (2007) Intra- and inter-lake variability of free-living and particle-associated Actinobacteria communities. Environmental Microbiology 9:2728-2741 Allgaier M, Grossart HP (2006a) Diversity and seasonal dynamics of Actinobacteria populations in four lakes in northeastern Germany. Applied and Environmental Microbiology 72:3489-3497 Allgaier M, Grossart HP (2006b) Seasonal dynamics and phylogenetic diversity of free-living and particle-associated bacterial communities in four lakes in northeastern Germany. Aquatic Microbial Ecology 45:115-128 Andreoni V, Cavalca L, Rao MA, Nocerino G, Bernasconi S, Dell'Amico E, Colombo M, Gianfreda L (2004) Bacterial communities and enzyme activities of PAHs polluted soils. Chemosphere 57:401-412 Ashelford KE, Chuzhanova NA, Fry JC, Jones AJ, Weightman AJ (2005) At least 1 in 20 16S rRNA sequence records currently held in public repositories is estimated to contain substantial anomalies. Applied and Environmental Microbiology 71:7724-7736 Ayala-del-Rio HL, Callister SJ, Criddle CS, Tiedje JM (2004) Correspondence between community structure and function during succession in phenol- and phenol-plus-trichloroethene-fed sequencing batch reactors. Applied and Environmental Microbiology 70:4950-4960

188

Baldwin BR, Nakatsu CH, Nies L (2003) Detection and enumeration of aromatic oxygenase genes by multiplex and real-time PCR. Applied and Environmental Microbiology 69:3350-3358 Bamforth SM, Singleton I (2005) Bioremediation of polycyclic aromatic hydrocarbons: Current knowledge and future directions. Journal of Chemical Technology and Biotechnology 80:723-736 Barbiero RP, Rockwell DC, Warren GJ, Tuchman ML (2006) Changes in spring phytoplankton communities and nutrient dynamics in the eastern basin of Lake Erie since the invasion of Dreissena spp. Canadian Journal of Fisheries and Aquatic Sciences 63:1549-1563 Barnsley EA (1983) Phthalate pathway of phenanthrene metabolism: Formation of 2'-carboxybenzalpyruvate. Journal of Bacteriology 154:113-117 Basta T, Keck A, Klein J, Stolz A (2004) Detection and characterization of conjugative degradative plasmids in xenobiotic-degrading Sphingomonas strains. Journal of Bacteriology 186:3862-3872 Bastiaens L, Springael D, Wattiau P, Harms H, deWachter R, Verachtert H, Diels L (2000) Isolation of adherent polycyclic aromatic hydrocarbon (PAH)- degrading bacteria using PAH-sorbing carriers. Applied and Environmental Microbiology 66:1834-1843 Bateman JN, Speer B, Feduik L, Hartline RA (1986) Naphthalene association and uptake in Pseudomonas putida. Journal of Bacteriology 166:155-161 Bathe S, Lebuhn M, Ellwart JW, Wuertz S, Hausner M (2004) High phylogenetic diversity of transconjugants carrying plasmid pJP4 in an activated sludge- derived microbial community. FEMS Microbiology Letters 235:215-219 Baumann PC (1998) Epizootics of cancer in fish associated with genotoxins in sediment and water. Mutation Research-Reviews in Mutation Research 411:227-233 Beijerinck M (1913) De Infusies en de Ontdekking der Backterien. Jaarboek van de Koninklijke Akademie v. Wetenschappen Muller, Amsterdam Beller HR, Kane SR, Legler TC, Alvarez PJJ (2002) A real-time polymerase chain reaction method for monitoring anaerobic, hydrocarbon-degrading bacteria based on a catabolic gene. Environmental Science & Technology 36:3977-3984 Bendinger B, Rijnaarts HHM, Altendorf K, Zehnder AJB (1993) Physiochemical cell surface and adhesive properties of coryneform bacteria related to the presence of chain-length of mycolic acids. Applied and Environmental Microbiology 59:3973-3977 Bent SJ, Pierson JD, Forney LJ, Danovaro R, Luna GM, Dell'Anno A, Pietrangeli B (2007) Measuring species richness based on microbial community fingerprints: the emperor has no clothes. Applied and Environmental Microbiology 73:2399-2401 Bisson IA, Marra PP, Burtt EH, Sikaroodi M, Gillevet PM (2007) A molecular comparison of plumage and soil bacteria across biogeographic, ecological, and taxonomic scales. Microbial Ecology 54:65-81

189

Blackwood CB, Hudleston D, Zak DR, Buyer JS (2007) Interpreting ecological diversity indices applied to terminal restriction fragment length polymorphism data: insights from simulated microbial communities. Applied and Environmental Microbiology 73:5276-5283 Boonchan S, Britz ML, Stanley GA (2000) Degradation and mineralization of high-molecular-weight polycyclic aromatic hydrocarbons by defined fungal-bacterial cocultures. Applied and Environmental Microbiology 66:1007-1019 Bosma TNP, Middeldorp PJM, Schraa G, Zehnder AJB (1997) Mass transfer limitation of biotransformation: Quantifying bioavailability. Environmental Science & Technology 31:248-252 Bossert ID, Bartha R (1986) Structure-biodegradability relationships of polycyclic aromatic hydrocarbons in soil. Bulletin of Environmental Contamination and Toxicology 37:490-495 Bouchez M, Blanchet D, Bardin V, Haeseler F, Vandecasteele JP (1999) Efficiency of defined strains and of soil consortia in the biodegradation of polycyclic aromatic hydrocarbon (PAH) mixtures. Biodegradation 10:429- 435 Braker G, Ayala-del-Rio HL, Devol AH, Fesefeldt A, Tiedje JM (2001) Community structure of denitrifiers, Bacteria, and Archaea along redox gradients in pacific northwest marine sediments by terminal restriction fragment length polymorphism analysis of amplified nitrite reductase (nirS) and 16S rRNA genes. Applied and Environmental Microbiology 67:1893-1901 Brezna B, Khan AA, Cerniglia CE (2003) Molecular characterization of dioxygenases from polycyclic aromatic hydrocarbon-degrading Mycobacterium spp. FEMS Microbiology Letters 223:177-183 Brezna B, Kweon O, Stingley RL, Freeman JP, Khan AA, Polek B, Jones RC, Cerniglia CE (2006) Molecular characterization of cytochrome P450 genes in the polycyclic aromatic hydrocarbon degrading Mycobacterium vanbaalenii PYR-1. Applied Microbiology and Biotechnology 71:522-532 Briee C, Moreira D, Lopez-Garcia P (2007) Archaeal and bacterial community composition of sediment and plankton from a suboxic freshwater pond. Research in Microbiology 158:213-227 Buckley DH, Schmidt TM (2001) The structure of microbial communities in soil and the lasting impact of cultivation. Microbial Ecology 42:11-21 Buckley DR, Rockne KJ, Li A, Mills WJ (2004) Soot deposition in the Great Lakes: Implications for semi-volatile hydrophobic organic pollutant deposition. Environmental Science & Technology 38:1732-1739 Burgmann H, Pesaro M, Widmer F, Zeyer J (2001) A strategy for optimizing quality and quantity of DNA extracted from soil. Journal of Microbiological Methods 45:7-20 Cariello NF, Thilly WG, Swenberg JA, Skopek TR (1991) Deletion mutagenesis during polymerase chain reaction: dependence on DNA polymerase. Gene 99:105-108 190

Caron DA, Davis PG, Madin LP, Sieburth JM (1982) Heterotrophic bacteria and bacterivorous protozoa in oceanic macroaggregates. Science 218:795-797 Casamayor EO, Massana R, Benlloch S, Ovreas L, Diez B, Goddard VJ, Gasol JM, Joint I, Rodriguez-Valera F, Pedros-Alio C (2002) Changes in archaeal, bacterial and eukaryal assemblages along a salinity gradient by comparison of genetic fingerprinting methods in a multipond solar saltern. Environmental Microbiology 4:338-348 Casellas M, Grifoll M, Bayona JM, Solanas AM (1997) New metabolites in the degradation of fluorene by Arthrobacter sp. strain F101. Applied and Environmental Microbiology 63:819-826 Castle DM, Montgomery MT, Kirchman DL (2006) Effects of naphthalene on microbial community composition in the Delaware estuary. FEMS Microbiology Ecology 56:55-63 Cerniglia CE (1984) Microbial metabolism of polycyclic aromatic hydrocarbons. Advances in Applied Microbiology 30:31-71 Cerniglia CE (1992) Biodegradation of polycyclic aromatic hydrocarbons. Biodegradation 3:351-368 Cheung PY, Kinkle BK (2001) Mycobacterium diversity and pyrene mineralization in petroleum-contaminated soils. Applied and Environmental Microbiology 67:2222-2229 Cho JC, Tiedje JM (2000) Biogeography and degree of endemicity of fluorescent Pseudomonas strains in soil. Applied and Environmental Microbiology 66:5448-5456 Churchill SA, Harper JP, Churchill PF (1999) Isolation and characterization of a Mycobacterium species capable of degrading three- and four-ring aromatic and aliphatic hydrocarbons. Applied and Environmental Microbiology 65:549-552 Clarke KR (1993) Nonparametric multivariate analyses of changes in community structure. Australian Journal of Ecology 18:117-143 Clarke KR, Gorley RN (2006) PRIMER v6: User Manual/Tutorial. PRIMER-E Ltd., Plymouth Cole JR, Chai B, Farris RJ, Wang Q, Kulam-Syed-Mohideen AS, McGarrell DM, Bandela AM, Cardenas E, Garrity GM, Tiedje JM (2007) The ribosomal database project (RDP-II): Introducing myRDP space and quality controlled public data. Nucleic Acids Research 35:D169-D172 Cordova-Kreylos AL, Cao YP, Green PG, Hwang HM, Kuivila KM, LaMontagne MG, Van De Werfhorst LC, Holden PA, Scow KM (2006) Diversity, composition, and geographical distribution of microbial communities in California salt marsh sediments. Microbial Ecology 72:3357-3366 Cortes DR, Basu I, Sweet CW, Hites RA (2000) Temporal trends in and influence of wind on PAH concentrations measured near the Great Lakes. Environmental Science & Technology 34:356-360 Covert TC, Rodgers MR, Reyes AL, Stelma GN, Jr. (1999) Occurrence of nontuberculous Mycobacteria in environmental samples. Applied and Environmental Microbiology 65:2492-2496 191

Crump BC, Armbrust EV, Baross JA (1999) Phylogenetic analysis of particle- attached and free-living bacterial communities in the Columbia River, its estuary, and the adjacent coastal ocean. Applied and Environmental Microbiology 65:3192-3204 Cupp AR, Bullerjahn GS, Dooley KC, Wilhelm SW Microbial community composition during transition to hypoxia in the central basin of Lake Erie. In revision Cupp AR, Bullerjahn GS, Dooley KC, Wilhelm SW (2008) Microbial community composition during transition to hypoxia in the central basin of Lake Erie. Submitted Curtis TP, Sloan WT, Scannell JW (2002) Estimating prokaryotic diversity and its limits. Proceedings of the National Academy of Sciences 99:10494-10499 Dachs J, Bayona JM, Raoux C, Albaiges J (1997) Spatial, vertical distribution and budget of polycyclic aromatic hydrocarbons in the western Mediterranean seawater. Environmental Science & Technology 31:682-688 Danovaro R, Luna GM, Dell'Anno A, Pietrangeli B (2006) Comparison of two fingerprinting techniques, terminal restriction fragment length polymorphism and automated ribosomal intergenic spacer analysis, for determination of bacterial diversity in aquatic environments. Applied and Environmental Microbiology 72:5982-5989 Davies JI, Evans WC (1964) Oxidative metabolism of naphthalene by soil Pseudomonads. The ring-fission mechanism. Biochemistry Journal 91:251-261 Dean-Ross D, Cerniglia CE (1996) Degradation of pyrene by Mycobacterium flavescens. Applied Microbiology and Biotechnology 46:307-312 Dean-Ross D, Moody J, Cerniglia CE (2002) Utilization of mixtures of polycyclic aromatic hydrocarbons by bacteria isolated from contaminated sediment. FEMS Microbiology Ecology 41:1-7 DeBruyn JM, Chewning CS, Sayler GS (2007) Comparative quantitative prevalence of Mycobacteria and functionally abundant nidA, nahAc, and nagAc dioxygenase genes in coal tar contaminated sediments. Environmental Science & Technology 41:5426-5432 DeBruyn JM, Leigh-Bell JA, McKay RML, Bourbonniere RA, Wilhelm SW (2004) Microbial distributions and the impact of phosphorus on bacterial activity in Lake Erie. Journal of Great Lakes Research 30:166-183 DeLong EF, Franks DG, Alldredge AL (1993) Phylogenetic diversity of aggregate-attached vs. free-living marine bacterial assemblages. Limnology and Oceanography 38:924-934 Dennis JJ (2005) The evolution of IncP catabolic plasmids. Current Opinion in Biotechnology 16:291-298 Dennis JJ, Zylstra GJ (2004) Complete sequence and genetic organization of pDTG1, the 83 kilobase naphthalene degradation plasmid from Pseudomonas putida strain NCIB 9816-4. Journal of Molecular Biology 341:753-768

192

Devers M, El Azhari N, Kolic NU, Martin-Laurent F (2007) Detection and organization of atrazine-degrading genetic potential of seventeen bacterial isolates belonging to divergent taxa indicate a recent common origin of their catabolic functions. FEMS Microbiology Letters 273:78-86 Devers M, Soulas G, Martin-Laurent F (2004) Real-time reverse transcription PCR analysis of expression of atrazine catabolism genes in two bacterial strains isolated from soil. Journal of Microbiological Methods 56:3-15 Dickerson DS (2005) Distribution and transportation of coal tar contaminants in the Chattanooga Creek floodplain. M.S., University of Tennessee, Knoxville, TN Dionisi HM, Chewning CS, Morgan KH, Menn F-M, Easter JP, Sayler GS (2004) Abundance of dioxygenase genes similar to Ralstonia sp. strain U2 nagAc is correlated with naphthalene concentrations in coal tar-contaminated freshwater sediments. Applied and Environmental Microbiology 70:3988- 3995 Dionisi HM, Harms G, Layton AC, Gregory IR, Parker J, Hawkins SA, Robinson KG, Sayler GS (2003) Power analysis for real-time PCR quantification of genes in activated sludge and analysis of the variability introduced by DNA extraction. Applied and Environmental Microbiology 69:6597-6604 Dojka MA, Hugenholtz P, Haack SK, Pace NR (1998) Microbial diversity in a hydrocarbon- and chlorinated-solvent-contaminated aquifer undergoing intrinsic bioremediation. Applied and Environmental Microbiology 64:3869-3877 Droppo IG (2001) Rethinking what constitutes suspended sediment. Hydrological Processes 15:1551-1564 Drouillard KG, Tomczak M, Reitsma S, Haffner GD (2006) A river-wide survey of polychlorinated biphenyls (PCBs), polycylic aromatic hydrocarbons (PAHs), and selected organochlorine pesticide residues in sediments of the Detroit River - 1999. Journal of Great Lakes Research 32:209-226 Dunbar J, Ticknor LO, Kuske CR (2000) Assessment of microbial diversity in four southwestern United States soils by 16S rRNA gene terminal restriction fragment analysis. Applied and Environmental Microbiology 66:2943-2950 Dunbar JM, Ticknor LO, Kuske CR (2001) Phylogenetic specificity and reproducibility and new method for analysis of terminal restriction fragment profiles of 16S rRNA genes from bacterial communities. Applied and Environmental Microbiology 67:190-197 Eadie BJ (1984) Distribution of polycylic aromatic hydrocarbons in the Great Lakes. In: Nriagu JO, Simmons MS (eds) Toxic Contaminants in the Great Lakes, Vol 14. John Wiley & Sons, Inc., Washington, D.C., p 195-211 Egert M, Friedrich MW (2003) Formation of pseudo-terminal restriction fragments, a PCR-related bias affecting terminal restriction fragment length polymorphism analysis of microbial community structure. Applied and Environmental Microbiology 69:2555-2562

193

Eldridge ML, Cadotte MW, Rozmus AE, Wilhelm SW (2007) The response of bacterial groups to changes in available iron in the Eastern subtropical Pacific Ocean. Journal of Experimental Marine Biology and Ecology 348:11-22 Engebretson JJ, Moyer CL (2003) Fidelity of select restriction endonucleases in determining microbial diversity by terminal-restriction fragment length polymorphism. Applied and Environmental Microbiology 69:4823-4829 Evans WC, Fernley HN, Griffith E (1965) Oxidative metabolism of phenanthrene and anthracene by soil Pseudomonads - Ring-fission mechanism. Biochemical Journal 95:819-& Eyers L, George I, Schuler L, Stenuit B, Agathos SN, El Fantroussi S (2004) Environmental genomics: Exploring the unmined richness of microbes to degrade xenobiotics. Applied Microbiology and Biotechnology 66:123- 130 Fallon ME, Horvath FJ (1985) Preliminary assessment of contaminants in soft sediments of the Detroit River. Journal of Great Lakes Research 11:373- 378 Farrelly V, Rainey FA, Stackebrandt E (1995) Effect of genome size and rrn gene copy number on PCR amplification of 16S ribosomal RNA genes from a mixture of bacterial species. Applied and Environmental Microbiology 61:2798-2801 Farris MH, Olson JB (2007) Detection of Actinobacteria cultivated from environmental samples reveals bias in universal primers. Letters in Applied Microbiology 45:376-381 Fedi S, Tremaroli V, Scala D, Perez-Jimenez JR, Fava F, Young L, Zannoni D (2005) T-RFLP analysis of bacterial communities in cyclodextrin- amended bioreactors developed for biodegradation of polychlorinated biphenyls. Research in Microbiology 156:201-210 Ferreira NL, Maciel H, Mathis H, Monot F, Fayolle-Guichard F, Greer CW (2006) Isolation and characterization of a new Mycobacterium austroafricanum strain, IFP 2015, growing on MTBE. Applied Microbiology and Biotechnology 70:358-365 Fierer N (2007) Tilting at Windmills: A response to a recent critique of terminal restrion fragment length polymorphism data. Applied and Environmental Microbiology 73:8041-8042 Fierer N, Jackson RB (2006) The diversity and biogeography of soil bacterial communities. Proceedings of the National Academy of Science 103:626- 631 Fleming JT, Sanseverino J, Sayler GS (1993) Quantitative relationship between naphthalene catabolic gene-frequency and expression in predicting PAH degradation in soils at town gas manufacturing sites. Environmental Science & Technology 27:1068-1074 Francois A, Mathis H, Godefroy D, Piveteau P, Fayolle F, Monot F (2002) Biodegradation of methyl tert-butyl ether and other fuel oxygenates by a

194

new strain, Mycobacterium austroafricanum IFP 2012. Applied and Environmental Microbiology 68:2754-2762 Frey JC, Angert ER, Pell AN (2006) Assessment of biases associated with profiling simple, model communities using terminal-restriction fragment length polymorphism-based analyses. Journal of Microbiological Methods 67:9-19 Fries MR, Zhou JH, Cheesanford J, Tiedje JM (1994) Isolation, characterization, and distribution of denitrifying toluene degraders from a variety of habitats. Applied and Environmental Microbiology 60:2802-2810 Frischer ME, Nierzwicki-Bauer SA, Parsons RH, Vathanodorn K, Waitkus KR (2000) Interactions between zebra mussels (Dreissena polymorpha) and microbial communities. Canadian Journal of Fisheries and Aquatic Sciences 57:591-599 Frostegard A, Courtois S, Ramisse V, Clerc S, Bernillon D, Le Gall F, Jeannin P, Nesme X, Simonet P (1999) Quantification of bias related to the extraction of DNA directly from soils. Applied and Environmental Microbiology 65:5409-5420 Fulthorpe RR, Rhodes AN, Tiedje JM (1998) High levels of endemicity of 3- chlorobenzoate-degrading soil bacteria. Applied and Environmental Microbiology 64:1620-1627 Gaillard M, Vallaeys T, Vorholter FJ, Minoia M, Werlen C, Sentchilo V, Puhler A, van der Meer JR (2006) The clc element of Pseudomonas sp. strain B13, a genomic island with various catabolic properties. Journal of Bacteriology 188:1999-2013 Gewurtz SB, Lazar R, Haffner GD (2000) Comparison of polycyclic aromatic hydrocarbon and polychlorinated biphenyl dynamics in benthic invertebrates of Lake Erie, USA. Environmental Toxicology and Chemistry 19:2943-2950 Gewurtz SB, Lazar R, Haffner GD (2003) Biomonitoring of bioavailable PAH and PCB water concentrations in the Detroit River using the freshwater mussel, Elliptio complanata. Journal of Great Lakes Research 29:242-255 Ghiglione J, Mevel G, Pujo-Pay M, Mousseau L, Lebaron P, Goutx M (2007) Diel and seasonal variations in abundance, activity, and community structure of particle-attached and free-living bacteria in NW Mediterranean Sea. Microbial Ecology 54:217-231 Gibson DT, Hensley M, Yoshioka H, Mabry TJ (1970) Formation of (+)-cis-2,3- dihydroxy-1-methylcyclohexa-4,6-diene from toluene by Pseudomonas putida. Biochemistry 9:1626-1630 Giulietti A, Overbergh L, Valckx D, Decallonne B, Bouillon R, Mathieu C (2001) An overview of real-time quantitative PCR: Applications to quantify cytokine gene expression. Methods 25:386-401 Gleeson D, McDermott F, Clipson N (2006) Structural diversity of bacterial communities in a heavy metal mineralized granite outcrop. Environmental Microbiology 8:383-393

195

Glockner FO, Zaichikov E, Belkova N, Denissova L, Pernthaler J, Pernthaler A, Amann R (2000) Comparative 16S rRNA analysis of lake bacterioplankton reveals globally distributed phylogenetic clusters including an abundant group of actinobacteria. Applied and Environmental Microbiology 66:5053-5065 Gomes NCM, Borges LR, Paranhos R, Pinto FN, Krogerrecklenfort E, Mendonca-Hagler LCS, Smalla K (2007) Diversity of ndo genes in mangrove sediments exposed to different sources of polycyclic aromatic hydrocarbon pollution. Applied and Environmental Microbiology 73:7392-7399 Gouvea SP, Melendez C, Carberry MJ, Bullerjahn GS, Wilhelm SW, Langen TA, Twiss MR (2006) Assessment of phosphorus-microbe interactions in Lake Ontario by multiple techniques. Journal of Great Lakes Research 32:455- 470 Goyal AK, Zylstra GJ (1997) Genetics of naphthalene and phenanthrene degradation by Comamonas testosteroni. Journal of Industrial Microbiology & Biotechnology 19:401-407 Grant A, Ogilvie LA (2004) Name that microbe: rapid identification of taxa responsible for inidividual fragments in fingerprints of microbial community structure. Molecular Ecology Notes 4:133-136 Grossart HP, Simon M (1998) Bacterial colonization and microbial decomposition of limnetic organic aggregates (lake snow). Aquatic Microbial Ecology 15:127-140 Grosser RJ, Warshawsky D, Vestal JR (1991) Indigenous and enhanced mineralization of pyrene, benzo[a]pyrene, and carbazole in soils. Applied and Environmental Microbiology 57:3462-3469 Habe H, Chung JS, Ishida A, Kasuga K, Ide K, Takemura T, Nojiri H, Yamane H, Omori T (2005) The fluorene catabolic linear plasmid in Terrabacter sp. strain DBF63 carries the beta-ketoadipate pathway genes, pcaRHGBDCFIJ, also found in proteobacteria. Microbiology-SGM 151:3713-3722 Habe H, Ide K, Yotsumoto M, Tsuji H, Yoshida T, Nojiri H, Omori T (2002) Degradation characteristics of a dibenzofuran-degrader Terrabacter sp. strain DBF63 toward chlorinated dioxins in soil. Chemosphere 48:201-207 Habe H, Miyakoshi M, Chung J, Kasuga K, Yoshida T, Nojiri H, Omori T (2003) Phthalate catabolic gene cluster is linked to the angular dioxygenase gene in Terrabacter sp. strain DBF63. Applied Microbiology and Biotechnology 61:44-54 Habe H, Omori T (2003) Genetics of polycyclic aromatic hydrocarbon metabolism in diverse aerobic bacteria. Bioscience Biotechnology and Biochemistry 67:225-243 Hahn MW, Lunsdorf H, Wu QL, Schauer M, Hofle MG, Boenigk J, Stadler P (2003) Isolation of novel ultramicrobacteria classified as Actinobacteria from five freshwater habitats in Europe and Asia. Applied and Environmental Microbiology 69:1442-1451 196

Hall K, Miller CD, Sorensen DL, Anderson AJ, Sims RC (2005) Development of a catabolically significant genetic probe for polycyclic aromatic hydrocarbon-degrading Mycobacteria in soil. Biodegradation 16:475-484 Hamann C, Hegemann J, Hildebrandt A (1999) Detection of polycyclic aromatic hydrocarbon degradation genes in different soil bacteria by polymerase chain reaction and DNA hybridization. FEMS Microbiology Letters 173:255-263 Hansen MC, Tolker-Nielsen T, Givskov M, Molin S (1998) Biased 16S rDNA PCR amplification caused by interference from DNA flanking the template region. FEMS Microbiology Ecology 26:141-149 Harayama S, Kok M, Neidle EL (1992) Functional and evolutionary relationships amond diverse oxygenases. Annual Review of Microbiology 46:565-601 Harms G, Layton AC, Dionisi HM, Gregory IR, Garrett VM, Hawkins SA, Robinson KG, Sayler GS (2003) Real-time PCR quantification of nitrifying bacteria in a municipal wastewater treatment plant. Environmental Science & Technology 37:343-351 Harms H, Bosma TNP (1997) Mass transfer limitation of microbial growth and pollutant degradation. Journal of Industrial Microbiology & Biotechnology 18:97-105 Hartmann M, Enkerli J, Widmer F (2007) Residual polymerase activity-induced bias in terminal restriction fragment length polymorphism analysis. Environmental Microbiology 9:555-559 Hartmann M, Frey B, Kolliker R, Widmer F (2005) Semi-automated genetic analyses of soil microbial communities: Comparison of T-RFLP and RISA based on descriptive and discriminative statistical approaches. Journal of Microbiological Methods 61:349-360 Hartmann M, Widmer F (2006) Community structure analyses are more sensitive to differences in soil bacterial communities than anonymous diversity indices. Applied and Environmental Microbiology 72:7804-7812 Hartmann M, Widmer F (2008) Reliability for detecting composition and changes of microbial communities by T-RFLP genetic profiling. FEMS Microbiology Ecology 63:249-260 Harvey RG (1996) Mechanisms of carcinogenesis of polycyclic aromatic hydrocarbons. Polycyclic Aromatic Compounds 9:1-23 Hatzinger PB, Alexander M (1995) Effect of aging of chemicals in soil on their biodegradability and extractability. Environmental Science & Technology 29:537-545 Hayaishi O, Katagiri M, Rothberg S (1955) Mechanism of the pyrocatechase reaction. Journal of the American Chemical Society 77:5450-5451 Heitkamp MA, Cerniglia CE (1988) Mineralization of polycyclic aromatic hydrocarbons by a bacterium isolated from sediment below an oil-field. Applied and Environmental Microbiology 54:1612-1614 Heitkamp MA, Franklin W, Cerniglia CE (1988a) Microbial metabolism of polycyclic aromatic hydrocarbons - Isolation and characterization of a

197

pyrene-degrading bacterium. Applied and Environmental Microbiology 54:2549-2555 Heitkamp MA, Freeman JP, Miller DW, Cerniglia CE (1988b) Pyrene degradation by a Mycobacterium sp. - Identification of ring oxidation and ring fission products. Applied and Environmental Microbiology 54:2556- 2565 Herbes SE (1981) Rate of microbial transformation of polycyclic aromtatic hydrocarbons in water and sediments in the vicinity of a coal-coking wastewater discharge. Applied and Environmental Microbiology 41:20-28 Herbes SE, Schwall LR (1978) Microbial transformation of polycyclic aromatic hydrocarbons in pristine and petroleum-contaminated sediments. Applied and Environmental Microbiology 35:306-316 Herrick JB, StuartKeil KG, Ghiorse WC, Madsen EL (1997) Natural horizontal transfer of a naphthalene dioxygenase gene between bacteria native to a coal tar-contaminated field site. Applied and Environmental Microbiology 63:2330-2337 Hilscherova K, Dusek L, Kubik V, Cupr P, Hofman J, Klanova J, Holoubek I (2007) Redistribution of organic pollutants in river sediments and alluvial soils related to major floods. Journal of Soils and Sediments 7:167-177 Hilyard EJ, Jones-Meehan JM, Spargo BJ, Hill RT (2008) Enrichment, isolation, and phylogenetic identification of polycyclic aromatic hydrocarbon- degrading bacteria from Elizabeth River sediments. Applied and Environmental Microbiology 74:1176-1182 Hoostal MJ, Bouzat JL (2008) The modulating role of dissolved organic matter on spatial patterns of microbial metabolism in Lake Erie sediments. Microbial Ecology 55:358-368 Hoostal MJ, Bullerjahn GS, McKay RML (2002) Molecular assessment of the potential for in situ bioremediation of PCBs from aquatic sediments. Hydrobiologia 469:59-65 Hugenholtz P, Huber T (2003) Chimeric 16S rDNA sequences of diverse origin are accumulating in the public databases. International Journal of Systematic and Evolutionary Microbiology 53:289-293 Hughes VM, Stevenson K, Sharp JM (2001) Improved preparation of high molecular weight DNA from pulsed-field gel electrophoresis from Mycobacteria. Journal of Microbiological Methods 44:209-215 Hullar MAJ, Kaplan LA, Stahl DA (2006) Recurring seasonal dynamics of microbial communities in stream habitats. Applied and Environmental Microbiology 72:713-722 Hwang SJ, Heath RT (1997) Bacterial productivity and protistan bacterivory in coastal and offshore communities of Lake Erie. Canadian Journal of Fisheries and Aquatic Sciences 54:788-799 Iida T, Mukouzaka Y, Nakamura K, Kudo T (2002) Plasmid-borne genes code for an angular dioxygenase involved in dibenzofuran degradation by Terrabacter sp strain YK3. Applied and Environmental Microbiology 68:3716-3723 198

Imamura Y, Ikeda M, Yoshida S, Kuraishi H (2000) Janibacter brevis sp nov., a new trichloroethylene-degrading bacterium isolated from polluted environments. International Journal of Systematic and Evolutionary Microbiology 50:1899-1903 Iwai S, Yamazoe A, Takahashi R, Kurisu F, Yagi O (2005) Degradation of monochlorinated dibenzo-p-dioxins by Janibacter sp strain YA isolated from river sediment. Current Microbiology 51:353-358 Izmalkova TY, Sazonova OI, Sokolov SL, Kosheleva IA, Boronin AM (2005) Diversity of genetic systems responsible for naphthalene biodegradation in Pseudomonas fluorescens strains. Microbiology 74:60-68 Janssen PH, Yates PS, Grinton BE, Taylor PM, Sait M (2002) Improved culturability of soil bacteria and isolation in pure culture of novel members of the divisions Acidobacteria, Actinobacteria, Proteobacteria, and Verrucomicrobia. Applied and Environmental Microbiology 68:2391- 2396 Jin SW, Zhu T, Xu XD, Xu Y (2006) Biodegradation of dibenzofuran by Janibacter terrae strain XJ-1. Current Microbiology 53:30-36 Johnsen AR, Karlson U (2004) Evaluation of bacterial strategies to promote the bioavailability of polycyclic aromatic hydrocarbons. Applied Microbiology and Biotechnology 63:452-459 Johnsen AR, Schmidt S, Hybholt TK, Henriksen S, Jacobsen CS, Andersen O (2007) Strong impact on the polycyclic aromatic hydrocarbon (PAH)- degrading community of a PAH-polluted soil but marginal effect on PAH degradation when priming with bioremediated soil dominated by Mycobacteria. Applied and Environmental Microbiology 73:1474-1480 Johnsen AR, Wick LY, Harms H (2005) Principles of microbial PAH-degradation in soil. Environmental Pollution 133:71-84 Jouanneau Y, Willison JC, Meyer C, Krivobok S, Chevron N, Besombes JL, Blake G (2005) Stimulation of pyrene mineralization in freshwater sediments by bacterial and plant bioaugmentation. Environmental Science & Technology 39:5729-5735 Juhasz AL, Britz ML, Stanley GA (1997) Degradation of fluoranthene, pyrene, benz[a]anthracene and dibenz[a,h]anthracene by Burkholderia cepacia. Journal of Applied Microbiology 83:189-198 Juhasz AL, Stanley GA, Britz ML (2000) Microbial degradation and detoxification of high molecular weight polycyclic aromatic hydrocarbons by Stenotrophomonas maltophilia strain VUN 10,003. Letters in Applied Microbiology 30:396-401 Juhasz AL, Walter N, Lease C, Bentham R, Stewart R (2005) Pilot scale bioremediation of creosote contaminated soil --efficacy of enhanced natural attenuation and bioaugmentation strategies. Bioremediation Journal 9:141-157 Jussila MM, Zhao J, Suominen L, Lindstrom K (2007) TOL plasmid transfer during bacterial conjugation in vitro and rhizoremediation of oil compounds in vivo. Environmental Pollution 146:510-524 199

Ka JO, Tiedje JM (1994) Integration and excision of a 2,4-dichlorophenoxyacetic acid degradative plasmid in Alcaligenes paradoxus and evidence of its natural intergenic transfer. Journal of Bacteriology 176:5284-5289 Kanaly RA, Harayama S (2000) Biodegradation of high-molecular-weight polycyclic aromatic hydrocarbons by bacteria. Journal of Bacteriology 182:2059-2067 Kaplan CW, Kitts CL (2003) Variation between observed and true Terminal Restriction Fragment length is dependent on true TRF length and purine content. Journal of Microbiological Methods 54:121-125 Kasuga K, Nojiri H, Yamane H, Kodama T, Omori T (1997) Cloning and characterization of the genes involved in the degradation of dibenzofuran by Terrabacter sp strain DBF63. Journal of Fermentation and Bioengineering 84:387-399 Kauppi B, Lee K, Carredano E, Parales RE, Gibson DT, Eklund H, Ramaswamy S (1998) Structure of an aromatic-ring-hydroxylating dioxygenase- naphthalene 1,2-dioxygenase. Structure 6:571-586 Keck J, Sims RC, Coover M, Park K, Symons B (1989) Evidence for cooxidation of polynuclear aromatic hydrocarbons in soil. Water Research 23:1467- 1476 Keith LH, Telliard WA (1979) Priority pollutants I - a perspective view. Environmental Science & Technology 13:416-423 Kelley I, Freeman JP, Evans FE, Cerniglia CE (1993) Identification of metabolites from the degradation of fluoranthene by Mycobacterium sp. strain PYR-1. Applied and Environmental Microbiology 59:800-806 Kennedy N, Brodie E, Connolly J, Clipson N (2004) Impact of lime, nitrogen and plant species on bacterial community structure in grassland microcosms. Environmental Microbiology 6:1070-1080 Kent AD, Smith DJ, Benson BJ, Triplett EW (2003) Web-based phylogenetic assignment tool for ananlysis of terminal restriction fragment length polymorphism profiles of microbial communities. Applied and Environmental Microbiology 2003:6768-6776 Kent AD, Yannarell AC, Rusak JA, Triplett EW, McMahon KD (2007) Synchrony in aquatic microbial community dynamics. ISME Journal 1:38- 47 Khan AA, Kim SJ, Paine DD, Cerniglia CE (2002) Classification of a polycyclic aromatic hydrocarbon-metabolizing bacterium, Mycobacterium sp strain PYR-1, as Mycobacterium vanbaalenii sp nov. International Journal of Systematic and Evolutionary Microbiology 52:1997-2002 Khan AA, Wang RF, Cao WW, Doerge DR, Wennerstrom D, Cerniglia CE (2001) Molecular cloning, nucleotide sequence, and expression of genes encoding a polcyclic aromatic ring dioxygenase from Mycobacterium sp strain PYR-1. Applied and Environmental Microbiology 67:3577-3585 Kim JH, Yamaguchi K, Lee SH, Tithof PK, Sayler GS, Yoon JH, Baek SJ (2005a) Evaluation of polycyclic aromatic hydrocarbons in the activation

200

of early growth response-1 and peroxisome proliferator activated receptors. Toxicological Sciences 85:585-593 Kim SJ, Kweon O, Freeman JP, Jones RC, Adjei MD, Jhoo JW, Edmondson RD, Cerniglia CE (2006) Molecular cloning and expression of genes encoding a novel dioxygenase involved in low- and high-molecular-weight polycyclic aromatic hydrocarbon degradation in Mycobacterium vanbaalenii PYR-1. Applied and Environmental Microbiology 72:1045- 1054 Kim SJ, Kweon O, Jones RC, Freeman JP, Edmondson RD, Cerniglia CE (2007) Complete and integrated pyrene degradation pathway in Mycobacterium vanbaalenii PYR-1 based on systems biology. Journal of Bacteriology 189:464-472 Kim YH, Engesser KH, Cerniglia CE (2005b) Numerical and genetic analysis of polycyclic aromatic hydrocarbon-degrading Mycobacteria. Microbial Ecology 50:110-119 Kimura M (1980) A simple method for estimating evolutionary rate of base substitutions through comparative studies of nucleotide sequences. Journal of Molecular Evolution 16:111-120 King JMH, Digrazia PM, Applegate B, Burlage R, Sanseverino J, Dunbar P, Larimer F, Sayler GS (1990) Rapid, sensitive bioluminescent reporter technology for naphthalene exposure and biodegradation. Science 249:778-781 Kirchman D (1983) The production of bacteria attached to particles suspsended in a fresh-water pond. Limnology and Oceanography 28:858-872 Kirchman DL (1993) Particulate detritus and bacteria in marine environments. In: Ford TE (ed) Aquatic microbiology -- an ecological approach. Blackwell, Oxford, p 321-341 Kiyohara H, Torigoe S, Kaida N, Asaki T, Ilda T, Hayashi H, Takizawa N (1994) Cloning and characterization of a chromosomal gene cluster, pah, that encodes the upper pathway for phenanthrene and naphthalene utilization by Pseudomonas putida OUS82. Journal of Bacteriology 176:2439-2443 Klaus JS, Janse I, Heikoop JM, Sanford RA, Fouke BW (2007) Coral microbial communities, zooxanthellae and mucus along gradients of seawater depth and coastal pollution. Environmental Microbiology 9:1291-1305 Kopczynski ED, Bateson MM, Ward DM (1994) Recognition of chimeric small- subunit ribosomal DNAs composed of genes from uncultivated microorganisms. Applied and Environmental Microbiology 60:746-748 Krsek M, Wellington EMH (1999) Comparison of different methods for the isolation and purification of total community DNA from soil. Journal of Microbiological Methods 39:1-16 Kruskal JB (1964) Multidimensional scaling by optimizing goodness of fit to a nonmetric hypothesis. Psychometrika 29:1-27 Kurata S, Kanagawa T, Magariyama Y, Takatsu K, Yamada K, Yokomaku T, Kamagata Y (2004) Reevaluation and reduction of a PCR bias caused by

201

reannealing of templates. Applied and Environmental Microbiology 70:7545 - 7549 Kuske CR, Ticknor LO, Miller ME, Dunbar JM, Davis JA, Barns SM, Belnap J (2002) Comparison of soil bacterial communities in rhizospheres of three plant species and the interspaces in an arid grassland. Applied and Environmental Microbiology 68:1854-1863 LaMontagne MG, Holden PA (2003) Comparison of free-living and particle- associated bacterial communities in a coastal lagoon. Microbial Ecology 46:228-237 Lane DJ, Pace B, Olsen GJ, Stahl DA, Sogin ML, Pace NR (1985) Rapid determination of 16S ribosomal RNA sequences for phylogenetic analyses. Proceedings of the National Academy of Science 82:6955-6959 Langworthy DE, Stapleton RD, Sayler GS, Findlay RH (1998) Genotypic and phenotypic responses of a riverine microbial community to polycyclic aromatic hydrocarbon contamination. Applied and Environmental Microbiology 64:3422-3428 Langworthy DE, Stapleton RD, Sayler GS, Findlay RH (2002) Lipid analysis of the response of a sedimentary microbial community to polycyclic aromatic hydrocarbons. Microbial Ecology 43:189-198 Larkin MJ, Kulakov LA, Allen CCR (2005) Biodegradation and Rhodococcus - masters of catabolic versatility. Current Opinion in Biotechnology 16:282- 290 Laurie AD, Lloyd-Jones G (2000) Quantification of phnAc and nahAc in contaminated New Zealand soils by competitive PCR. Applied and Environmental Microbiology 66:1814-1817 Lavrentyev PJ, McCarthy MJ, Klarer DM, Jochem F, Gardner WS (2004) Estuarine microbial food web patterns in a Lake Erie coastal wetland. Microbial Ecology 48:567-577 Leblond JD, Schultz TW, Sayler GS (2001) Observations on the preferential biodegradation of selected components of polyaromatic hydrocarbon mixtures. Chemosphere 42:333-343 Lee SE, Seo JS, Keum YS, Lee KJ, Li OX (2007) Fluoranthene metabolism and associated proteins in Mycobacterium sp JS14. Proteomics 7:2059-2069 Lemarchand C, Jardillier L, Carrias JF, Richardot M, Debroas D, Sime-Ngando T, Amblard C (2006) Community composition and activity of prokaryotes associated to detrital particles in two contrasting lake ecosystems. FEMS Microbiology Ecology 57:442-451 Leppard GG, Flannigan DT, Mavrocordatos D, Marvin CH, Bryant DW, McCarry BE (1998) Binding of polycyclic aromatic hydrocarbons by size classes of particulate in Hamilton Harbor water. Environmental Science & Technology 32:3633-3639 Lepš J, Šmilauer P (2003) Multivariate Analysis of Ecological Data using CANOCO. Cambridge University Press, Cambridge Leys NM, Ryngaert A, Bastiaens L, Top EM, Verstraete W, Springael D (2005a) Culture independent detection of Sphingomonas sp. EPA 505 related 202

strains in soils contaminated with polycyclic aromatic hydrocarbons (PAHs). Microbial Ecology 49:443-450 Leys NM, Ryngaert A, Bastiaens L, Wattiau P, Top EM, Verstraete W, Springael D (2005b) Occurrence and community composition of fast-growing Mycobacterium in soils contaminated with polycyclic aromatic hydrocarbons. FEMS Microbiology Ecology 51:375-388 Li W, Shi JD, Wang XG, Han YN, Tong W, Ma L, Liu B, Cai BL (2004) Complete nucleotide sequence and organization of the naphthalene catabolic plasmid pND6-1 from Pseudomonas sp strain ND6. Gene 336:231-240 Liang Y, Gardner DR, Miller CD, Chen D, Anderson AJ, Weimer BC, Sims RC (2006) Study of biochemical pathways and enzymes involved in pyrene degradation by Mycobacterium sp. strain KMS. Applied and Environmental Microbiology 72:7821-7828 Liu WT, Marsh TL, Cheng H, Forney LJ (1997) Characterization of microbial diversity by determining terminal restriction fragment length polymorphisms of genes encoding 16S rRNA. Applied and Environmental Microbiology 63:4516-4522 Lohner RN, Sigler V, Mayer CM, Balogh C (2007) A comparison of the benthic bacterial communities within and surrounding Dreissena clusters in lakes. Microbial Ecology 54:469-477 Mackintosh CE, Maldonado JA, Ikonomou MG, Gobas F (2006) Sorption of phthalate esters and PCBs in a marine ecosystem. Environmental Science & Technology 40:3481-3488 MacLeod CT, Daugulis AJ (2005) Interfacial effects in a two-phase partitioning bioreactor: degradation of polycyclic aromatic hydrocarbons (PAHs) by a hydrophobic Mycobacterium. Process Biochemistry 40:1799-1805 Mannisto MK, Tiirola M, Haggblom MM (2007) Bacterial communities in Arctic fjelds of Finnish Lapland are stable but highly pH-dependent. FEMS Microbiology Ecology 59:452-465 Margesin R, Labbe D, Schinner F, Greer CW, Whyte LG (2003) Characterization of hydrocarbon-degrading microbial populations in contaminated and pristine alpine soils. Applied and Environmental Microbiology 69:3085- 3092 Marlowe EM, Wang JM, Pepper IL, Maier RM (2002) Application of a reverse transcription-PCR assay to monitor regulation of the catabolic nahAc gene during phenanthrene degradation. Biodegradation 13:251-260 Marsh TL (2005) Culture-independent microbial community analysis with terminal restriction fragment length polymorphism. Methods in Enzymology 397:308-329 Marsh TL, Saxman PR, Cole JR, Tiedje JM (2000) Terminal restriction fragment length polymorphism analysis program, a web-based research tool for microbial community analysis. Applied and Environmental Microbiology 66:3616-3620

203

Martin-Laurent F, Philippot L, Hallet S, Chaussod R, Germon JC, Soulas G, Catroux G (2001) DNA extraction from soils: Old bias for new microbial diversity analysis methods. Applied and Environmental Microbiology 67:2354-2359 Marvin C, Allan L, Bryant D, McCarry B (2000) Use of the zebra mussel (Dreissena polymorpha) as a bioindicator for aromatic hydrocarbons in Hamilton Harbour. Water Quality Research Journal of Canada 35:59-72 Marvin CH, Leppard GG, West MM, Stern GA, Boden AR, McCarry BE (2004) Refined tunable methodology for characterization of contaminant-particle relationships in surface water. Journal of Environmental Quality 33:2132- 2140 Mason HS, Fowlks WL, Peterson E (1955) Oxygen transfer and electron transport by the phenolase complex. Journal of the American Chemical Society 77:2914-2915 Maszenan AM, Seviour RJ, Patel BKC, Schumann P, Burghardt J, Tokiwa Y, Stratton HM (2000) Three isolates of novel polyphosphate-accumulating Gram-positive cocci, obtained from activated sludge, belong to a new genus, Tetrasphaera gen. nov., and description of two new species, Tetrasphaera japonica sp, nov and Tetrasphaera australiensis sp nov. International Journal of Systematic and Evolutionary Microbiology 50:593-603 Mathieu-Daude F, Welsh J, Vogt T, McClelland M (1996) DNA rehybridization during PCR: The 'C(O)t effect' and its consequences. Nucleic Acids Research 24:2080-2086 McNally DL, Mihelcic JR, Lueking DR (1998) Polycyclic aromatic hydrocarbon degrading microorganisms in Great Lakes sediments. Journal of Great Lakes Research 24:392-403 Metcalfe CD, Metcalfe TL, Riddle G, Haffner GD (1997) Aromatic hydrocarbons in biota from the Detroit River and western Lake Erie. Journal of Great Lakes Research 23:160-168 Miller CD, Child R, Hughes JE, Benscai M, Der JP, Sims RC, Anderson AJ (2007) Diversity of soil Mycobacterium isolates from three sites that degrade polycyclic aromatic hydrocarbons. Journal of Applied Microbiology 102:1612-1624 Miller CD, Hall K, Liang YN, Nieman K, Sorensen D, Issa B, Anderson AJ, Sims RC (2004) Isolation and characterization of polycyclic aromatic hydrocarbon-degrading Mycobacterium isolates from soil. Microbial Ecology 48:230-238 Mills DK, Fitzgerald K, Litchfield CD, Gillevet PM (2003) A comparison of DNA profiling techniques for monitoring nutrient impact on microbial community composition during bioremediation of petroleum- contaminated soils. 54:57-74 Moeseneder MM, Winter C, Herndl GJ (2001) Horizontal and vertical complexity of attached and free-living bacteria of the eastern Mediterranean Sea,

204

determined by 16S rDNA and 16S rRNA fingerprints. Limnology and Oceanography 46:95-107 Moody JD, Freeman JP, Doerge DR, Cerniglia CE (2001) Degradation of phenanthrene and anthracene by cell suspensions of Mycobacterium sp strain PYR-1. Applied and Environmental Microbiology 67:1476-1483 Moody JD, Freeman JP, Fu PP, Cerniglia CE (2004) Degradation of benzo[a]pyrene by Mycobacterium vanbaalenii PYR-1. Applied and Environmental Microbiology 70:340-345 Muckian L, Grant R, Doyle E, Clipson N (2007) Bacterial community structure in soils contaminated by polycyclic aromatic hydrocarbons. Chemosphere 68:1535-1541 Mueller JG, Chapman PJ, Pritchard PH (1989) Creosote-contaminated sites - Their potential for bioremediation. Environmental Science & Technology 23:1197-1201 Mueller JG, Devereux R, Santavy DL, Lantz SE, Willis SG, Pritchard PH (1997) Phylogenetic and physiological comparisons of PAH-degrading bacteria from geographically diverse soils. Antonie Van Leeuwenhoek International Journal of General and Molecular Microbiology 71:329-343 Mullis KB, Faloona FA (1987) Specific synthesis of DNA in vitro via a polymerase-catalyzed chain reaction. Methods in Enzymology 155:335- 350 Mummey DL, Stahl PD (2003) Spatial and temporal variability of bacterial 16S rDNA-based T-RFLP patterns derived from soil of two Wyoming grassland ecosystems. FEMS Microbiology Ecology 46:113-120 Mumy KL, Findlay RH (2004) Convenient determination of DNA extraction efficiency using an external DNA recovery standard and quantitative- competitive PCR. Journal of Microbiological Methods 57:259-268 Myers CR, Alatalo LJ, Myers JM (1994) Microbial potential for the anaerobic degradation of simple aromatic compounds in sediments of the Milwaukee Harbor, Green Bay and Lake Erie. Environmental Toxicology and Chemistry 13:461-471 Neefs JM, Vandepeer Y, Derijk P, Chapelle S, Dewachter R (1993) Compiliation of small ribosomal subunit RNA structures. Nucleic Acids Research 21:3025-3049 Nelson DR, Mele PM (2007) Subtle changes in rhizosphere microbial community structure in response to increased boron and sodium chloride concentrations. Soil Biology & Biochemistry 39:340-351 Ní Chadhain SM, Norman RS, Pesce KV, Kukor JJ, Zystra GJ (2006) Microbial dioxygenase gene population shifts during polycyclic aromatic hydrocarbon biodegradation. Applied and Environmental Microbiology 72:4078-4087 Nojiri H, Shintani M, Omori T (2004) Divergence of mobile genetic elements involved in the distribution of xenobiotic-catabolic capacity. Applied Microbiology and Biotechnology 64:154-174

205

Noumura T, Habe H, Widada J, Chung JS, Yoshida T, Nojiri H, Omori T (2004) Genetic characterization of the dibenzofuran-degrading Actinobacteria carrying the dbfA1A2 gene homologues isolated from activated sludge. FEMS Microbiology Letters 239:147-155 Nyyssönen M, Piskonen R, Itävaara M (2006) A targeted real-time PCR assay for studying naphthalene degradation in the environment. Microbial Ecology V52:533-543 Ogram A, Sayler GS, Barkay T (1987) The extraction and purification of microbial DNA from sediments. Journal of Microbiological Methods 7:57-66 Ogram A, Sayler GS, Gustin D, Lewis RJ (1988) DNA adsorption to soils and sediments. Environmental Science & Technology 22:982-984 Ono A, Miyazaki R, Sota M, Ohtsubo Y, Nagata Y, Tsuda M (2007) Isolation and characterization of naphthalene catabolic genes and plasmids from oil- contaminated soil by using two cultivation-independent approaches. Applied Microbiology and Biotechnology 74:501-510 Osborn AM, Boltner D (2002) When phage, plasmids, and transposons collide: Genomic islands, and conjugative- and mobilizable-transposons as a mosaic continuum. Plasmid 48:202-212 Osborn AM, Moore ERB, Timmis KN (2000) An evaluation of terminal- restriction fragment length polymorphism (T-RFLP) analysis for the study of microbial community structure and dynamics. Environmental Microbiology 2:39-50 Ouellette AJA, Handy SM, Wilhelm SW (2006) Toxic microcystis is widespread in Lake Erie: PCR detection of toxin genes and molecular characterization of associated cyanobacterial communities. Microbial Ecology 51:154-165 Pace NR (1997) A molecular view of microbial diversity and the biosphere. Science 276:734-740 Pagnout C, Frache G, Poupin P, Maunit B, Muller JF, Ferard JF (2007) Isolation and characterization of a gene cluster involved in PAH degradation in Mycobacterium sp strain SNP11: Expression in Mycobacterium smegmatis mc2155. Research in Microbiology 158:175-186 Palmer MW (1993) Putting things in even better order - The advantages of canonical correspondence analysis. Ecology 74:2215-2230 Pandey J, Ganesan K, Jain RK (2007) Variations in T-RFLP profiles with differing chemistries of fluorescent dyes used for labeling the PCR primers. Journal of Microbiological Methods 68:633-638 Parales RE, Lee K, Resnick SM, Jiang HY, Lessner DJ, Gibson DT (2000) Substrate specificity of naphthalene dioxygenase: Effect of specific amino acids at the active site of the enzyme. Journal of Bacteriology 182:1641- 1649 Pelz O, Tesar M, Wittich RM, Moore ERB, Timmis KN, Abraham WR (1999) Towards elucidation of microbial community metabolic pathways: unravelling the network of carbon sharing in a pollutant-degrading

206

bacterial consortium by immunocapture and isotopic ratio mass spectrometry. Environmental Microbiology 1:167-174 Pesaro M, Nicollier G, Zeyer J, Widmer F (2004) Impact of soil drying-rewetting stress microbial communities and activities and on degradation of two crop protection products. Applied and Environmental Microbiology 70:2577-2587 Phale PS, Basu A, Majhi PD, Deveryshetty J, Vamsee-Krishna C, Shrivastava R (2007) Metabolic diversity in bacterial degradation of aromatic compounds. OMICS A Journal of Integrative Biology 11:252-278 Pickard SW, Clarke JU, Lotufo GR (2006) Bioavailability of polycyclic aromatic hydrocarbons (PAHs) from surficial Lake Erie sediments. Bulletin of Environmental Contamination and Toxicology 76:791-798 Pinyakong O, Habe H, Omori T (2003) The unique aromatic catabolic genes in sphingomonads degrading polycyclic aromatic hydrocarbons (PAHs). Journal of General and Applied Microbiology 49:1-19 Poelarends GJ, Zandstra M, Bosma T, Kulakov LA, Larkin MJ, Marchesi JR, Weightman AJ, Janssen DB (2000) Haloalkane-utilizing Rhodococcus strains isolated from geographically distinct locations possess a highly conserved gene cluster encoding haloalkane catabolism. Journal of Bacteriology 182:2725-2731 Pohlman JW, Coffin RB, Mitchell CS, Montgomery MT, Spargo BJ, Steele JK, Boyd TJ (2002) Transport, deposition and biodegradation of particle bound polycyclic aromatic hydrocarbons in a tidal basin of an industrial watershed. Environmental Monitoring and Assessment 75:155-167 Polz MF, Cavanaugh CM (1998) Bias in template-to-product ratios in multitemplate PCR. Applied and Environmental Microbiology 64:3724- 3730 Powell SM, Ferguson SH, Bowman JP, Snape I (2006) Using real-time PCR to assess changes in the hydrocarbon-degrading microbial community in Antarctic soil during bioremediation. Microbial Ecology V52:523-532 Prabhu Y, Phale PS (2003) Biodegradation of phenanthrene by Pseudomonas sp. strain PP2: Novel metabolic pathway, role of biosurfactant and cell surface hydrophobicity in hydrocarbon assimilation. Applied Microbiology and Biotechnology 61:342-351 Priddle MW, MacQuarrie KTB (1994) Dissolution of creosote in groundwater - an experimental and modeling investigation. Journal of Contaminant Hydrology 15:27-56 Primm TP, Lucero CA, Falkinham JO (2004) Health impacts of environmental Mycobacteria. Clinical Microbiology Reviews 17:98-106 Qiu XY, Wu LY, Huang HS, McDonel PE, Palumbo AV, Tiedje JM, Zhou JZ (2001) Evaluation of PCR-generated chimeras: Mutations, and heteroduplexes with 16S rRNA gene-based cloning. Applied and Environmental Microbiology 67:880-887 Ramette A (2007) Multivariate analyses in microbial ecology. FEMS Microbiology Ecology 62:142-160 207

Rehmann K, Hertkorn N, Kettrup AA (2001) Fluoranthene metabolism in Mycobacterium sp strain KR20: Identity of pathway intermediates during degradation and growth. Microbiology-SGM 147:2783-2794 Reysenbach AL, Giver LJ, Wickham GS, Pace NR (1992) Differential amplification of ribosomal RNA genes by polymerase chain reaction. Applied and Environmental Microbiology 58:3417-3418 Reysenbach AL, Wickham GS, Pace NR (1994) Phylogenetic analysis of the hyperthermophilic pink filament community in Octopus Spring, Yellowston National Park. Applied and Environmental Microbiology 60:2113-2119 Rhee SK, Liu X, Wu L, Chong SC, Wan X, Zhou J (2004) Detection of genes involved in biodegradation and biotransformation in microbial communities by using 50-mer oligonucleotide microarrays. Applied and Environmental Microbiology 70:4303-4317 Riemann L, Winding A (2001) Community dynamics of free-living and particle- associated bacterial assemblages during a freshwater phytoplankton bloom. Microbial Ecology 42:274-285 Ringelberg DB, Talley JW, Perkins EJ, Tucker SG, Luthy RG, Bouwer EJ, Fredrickson HL (2001) Succession of phenotypic, genotypic, and metabolic community characteristics during in vitro bioslurry treatment of polycyclic aromatic hydrocarbon-contaminated sediments. Applied and Environmental Microbiology 67:1542-1550 Romine MF, Stillwell LC, Wong KK, Thurston SJ, Sisk EC, Sensen C, Gaasterland T, Fredrickson JK, Saffer JD (1999) Complete sequence of a 184-kilobase catabolic plasmid from Sphingomonas aromaticivorans F199. Journal of Bacteriology 181:1585-1602 Saari TA, Saari SK, Campbell CD, Alexander IJ, Anderson IC (2007) FragMatch - A program for the analysis of DNA fragment data. Mycorrhiza 17:133- 136 Salles JF, van Veen JA, van Elsas JD (2004) Multivariate analyses of Burkholderia species in soil: Effect of crop and land use history. Applied and Environmental Microbiology 70:4012-4020 Samanta SK, Singh OV, Jain RK (2002) Polycyclic aromatic hydrocarbons: Environmental pollution and bioremediation. Trends in Biotechnology 20:243-248 Sanseverino J, Applegate BM, King JMH, Sayler GS (1993a) Plasmid-mediated mineralization of naphthalene, phenanthrene and anthracene. Applied and Environmental Microbiology 59:1931-1937 Sanseverino J, Werner C, Fleming J, Applegate B, King JMH, Sayler GS (1993b) Molecular diagnostics of polycyclic aromatic hydrocarbon biodegradation in manufactured gas plant soils. Biodegradation 4:303-321 Sapp M, Wichels A, Wiltshire KH, Gerdts G (2007) Bacterial community dynamics during the winter-spring transition in the North Sea. FEMS Microbiology Ecology 59:622-637

208

Sayler GS, Hooper SW, Layton AC, King JMH (1990) Catabolic plasmids of environmental and ecological significance. Microbial Ecology 19:1-20 Sayler GS, Shields MS, Tedford ET, Breen A, Hooper SW, Sirotkin KM, Davis JW (1985) Application of DNA-DNA colony hybridization to the detection of catabolic genotypes in environmental samples. Applied and Environmental Microbiology 49:1295-1303 Schell MA (1985) Transcriptional control of the NAH and SAL hydrocarbon- degradation operons by the NahR gene product. Gene 36:301-309 Schmalenberger A, Schwieger F, Tebbe CC (2001) Effect of primers hybridizing to different evolutionarily conserved regions of the small-subunit rRNA gene in PCR-based microbial community analyses and genetic profiling. Applied and Environmental Microbiology 67:3557-3563 Schneider J, Grosser R, Jayasimhulu K, Xue W, Warshawsky D (1996) Degradation of pyrene, benz[a]anthracene, and benzo[a]pyrene by Mycobacterium sp strain RJGII-135, isolated from a former coal gasification site. Applied and Environmental Microbiology 62:13-19 Schwalbach MS, Hewson I, Fuhrman JA (2004) Viral effects on bacterial community composition in marine plankton microcosms. Aquatic Microbial Ecology 34:117-127 Schweitzer B, Huber I, Amann R, Ludwig W, Simon M (2001) alpha- and beta- Proteobacteria control the consumption and release of amino acids on lake snow aggregates. Applied and Environmental Microbiology 67:632-645 Shaw PJA (2003) Multivariate Statistics for the Environmental Sciences. Hodder Arnold, London Shingler V (2003) Integrated regulation in response to aromatic compounds: from signal sensing to attractive behaviour. Environmental Microbiology 5:1226-1241 Shintani M, Fukushima N, Tezuka M, Yamane H, Nojiri H (2007) Conjugative transfer of the IncP-7 carbazole degradative plasmid, pCAR1, in river water samples. Biotechnology Letters Sho M, Hamel C, Greer CW (2004) Two distinct gene clusters encode pyrene degradation in Mycobacterium sp strain S65. FEMS Microbiology Ecology 48:209-220 Shuttleworth KL, Cerniglia CE (1996) Bacterial degradation of low concentrations of phenanthrene and inhibition by naphthalene. Microbial Ecology 31:305-317 Shuttleworth KL, Cerniglia CE (1997) Practical methods for the isolation of polycyclic aromatic hydrocarbon (PAH)-degrading microorganisms and the determination of PAH mineralization and biodegradation intermediates. In: Hurst CJ, Knuden GR, McInerney MJ, Stetzenbach LD, Walter MV (eds) Manual of Environmental Microbiology. American Society for Microbiology, Washington, D. C., p 766-775 Sierra I, Valera JL, Marina ML, Laborda F (2003) Study of the biodegradation process of polychlorinated biphenyls in liquid medium and soil by a new isolated aerobic bacterium (Janibacter sp). Chemosphere 53:609-618 209

Simon M, Grossart HP, Schweitzer B, Ploug H (2002) Microbial ecology of organic aggregates in aquatic ecosystems. Aquatic Microbial Ecology 28:175-211 Singleton DR, Powell SN, Sangaiah R, Gold A, Ball LM, Aitken MD (2005) Stable-isotope probing of bacteria capable of degrading salicylate, naphthalene, or phenanthrene in a bioreactor treating contaminated soil. Applied and Environmental Microbiology 71:1202-1209 Singleton DR, Sangaiah R, Gold A, Ball LM, Aitken MD (2006) Identification and quantification of uncultivated Proteobacteria associated with pyrene degradation in a bioreactor treating PAH-contaminated soil. Environmental Microbiology 8:1736-1745 Smirnov A, Abrajano TA, Smirnov A, Stark A (1998) Distribution and sources of polycyclic aromatic hydrocarbons in the sediments of Lake Erie, - Part 1. Spatial distribution, transport, and deposition. Organic Geochemistry 29:1813-1828 Spring S, Schulze R, Overmann J, Schleifer KH (2000) Identification and characterization of ecologically significant prokaryotes in the sediment of freshwater lakes: Molecular and cultivation studies. FEMS Microbiology Reviews 24:573-590 Springael D, Top EM (2004) Horizontal gene transfer and microbial adaptation to xenobiotics: New types of mobile genetic elements and lessons from ecological studies. Trends in Microbiology 12:53-58 Stach JEM, Burns RG (2002) Enrichment versus biofilm culture: a functional and phylogenetic comparison of polycyclic aromatic hydrocarbon-degrading microbial communities. Environmental Microbiology 4:169-182 Stahl DA, Urbance JW (1990) The division between fast-growing and slow- growing species corresponds to natural relationships among the Mycobacteria. Journal of Bacteriology 172:116-124 Stapleton RD, Bright NG, Sayler GS (2000a) Catabolic and genetic diversity of degradative bacteria from fuel-hydrocarbon contaminated aquifers. Microbial Ecology 39:211-221 Stapleton RD, Ripp S, Jimenez L, Cheol-Koh S, Fleming JT, Gregory IR, Sayler GS (1998) Nucleic acid analytical approaches in bioremediation: site assessment and characterization. Journal of Microbiological Methods 32:165-178 Stapleton RD, Sayler GS (1998) Assessment of the microbiological potential for the natural attenuation of petroleum hydrocarbons in a shallow aquifer system. Microbial Ecology 36:349-361 Stapleton RD, Sayler GS, Boggs JM, Libelo EL, Stauffer T, MacIntyre WG (2000b) Changes in subsurface catabolic gene frequencies during natural attenuation of petroleum hydrocarbons. Environmental Science & Technology 34:1991-1999 Stein LY, Jones G, Alexander B, Elmund K, Wright-Jones C, Nealson KH (2002) Intriguing microbial diversity associated with metal-rich particles from a freshwater reservoir. FEMS Microbiology Ecology 42:431-440 210

Stinear TP, Hong H, Frigui W, Pryor MJ, Brosch R, Garnier T, Leadlay PF, Cole ST (2005) Common evolutionary origin for the unstable virulence plasmid pMUM found in geographically diverse strains of Mycobacterium ulcerans. Journal of Bacteriology 187:1668-1676 Stoffels M, Amann R, Ludwig W, Hekmat D, Schleifer KH (1998) Bacterial community dynamics during start-up of a trickle-bed bioreactor degrading aromatic compounds. Applied and Environmental Microbiology 64:930- 939 Sun P, Backus S, Blanchard P, Hites RA (2006) Annual variation of polycyclic aromatic hydrocarbon concentrations in precipitation collected near the Great Lakes. Environmental Science & Technology 40:696-701 Suzuki K, Suzuki M, Sasaki J, Park YH, Komagata K (1999) Leifsonia gen nov, a genus for 2,4-diaminobutyric acid-containing actinomycetes to accommodate "Corynebacterium aquaticum" (Leifson 1962) and Clavibacter xyli subsp cynodontis (Davis et al. 1984). Journal of General and Applied Microbiology 45:253-262 Suzuki MT, Giovannoni SJ (1996) Bias caused by template annealing in the amplification of mixtures of 16S rRNA genes by PCR. Applied and Environmental Microbiology 62:625-630 Tan HM (1999) Bacterial catabolic transposons. Applied Microbiology and Biotechnology 51:1-12 Tattersfield F (1927) The decomposition of naphthalene in the soil and the effect upon its insecticidal action. Annals of Applied Biology 15:67 Tay STL, Hemond FH, Krumholz LR, Cavanaugh CM, Polz MF (2001) Population dynamics of two toluene degrading bacterial species in a contaminated stream. Microbial Ecology 41:124-131 Tay STL, Hemond HF, Polz MF, Cavanaugh CM, Dejesus I, Krumholz LR (1998) Two new Mycobacterium strains and their role in toluene degradation in a contaminated stream. Applied and Environmental Microbiology 64:1715-1720 Taylor PM, Medd JM, Schoenborn L, Hodgson B, Janssen PH (2002) Detection of known and novel genes encoding aromatic ring-hydroxylating dioxygenases in soils and in aromatic hydrocarbon-degrading bacteria. FEMS Microbiology Letters 216:61-66 ter Braak CJF (1986) Canonical correspondence analysis - A new eigenvector technique for multivariate direct gradient analysis. Ecology 67:1167-1179 Thompson JR, Marcelino LA, Polz MF (2002) Heteroduplexes in mixed-template amplifications: formation, consequence and elimination by 'reconditioning PCR'. Nucleic Acids Research 30:2083-2088 Thorel MF, Falkinham JO, Moreau RG (2004) Environmental Mycobacteria from alpine and subalpine habitats. FEMS Microbiology Ecology 49:343-347 Tiedje JM, Asuming-Brempong S, Nusslein K, Marsh TL, Flynn SJ (1999) Opening the black box of soil microbial diversity. Applied Soil Ecology 13:109-122

211

Top EM, Springael D (2003) The role of mobile genetic elements in bacterial adaptation to xenobiotic organic compounds. Current Opinion in Biotechnology 14:262-269 Top EM, Springael D, Boon N (2002) Catabolic mobile genetic elements and their potential use in bioaugmentation of polluted soils and waters. FEMS Microbiology Ecology 42:199-208 Torsvik V, Goksoyr J, Daae FL (1990) High diversity in DNA of soil bacteria. Applied and Environmental Microbiology 56:782-787 Torsvik VL (1980) Isolation of bacterial DNA from soil. Soil Biology & Biochemistry 12:15-21 Tsai YL, Olson BH (1992) Detection of low numbers of bacterial cells in soils and sediments by polymerase chain reaction. Applied and Environmental Microbiology 58:754-757 Tsuda M, Tan HM, Nishi A, Furukawa K (1999) Mobile catabolic genes in bacteria. Journal of Bioscience and Bioengineering 87:401-410 Turley CM, Mackie PJ (1994) Biogeochemical significance of attached and free- living bacteria and the flux of particles in the NE Atlantic Ocean. Marine Ecological Progress Series 115:191-203 Twiss MR, Campbell PGC (1998) Trace metal cycling in the surface waters of Lake Erie: Linking ecological and geochemical fates. Journal of Great Lakes Research 24:791-807 US-EPA (1999) Use of Monitored Natural Attenuation at Superfund, RCRA Corrective Action, and Underground Storage Tank Sites. OSWER Directive 9200.4-17P, April 21, 1999. http://www.epa.gov:80/ordntrnt/ORD/WebPubs/biorem/D9200417.pdf Uyttebroek M, Spoden A, Ortega-Calvo JJ, Wouters K, Wattiau P, Bastiaens L, Springael D (2007) Differential responses of eubacterial, Mycobacterium, and Sphingomonas communities in polycyclic aromatic hydrocarbon (PAH)-contaminated soil to artificially induced changes in PAH profile. Journal of Environmental Quality 36:1403-1411 van der Meer JR, Sentchilo V (2003) Genomic islands and the evolution of catabolic pathways in bacteria. Current Opinion in Biotechnology 14:248- 254 Vinas M, Sabate J, Espuny MJ, Solanas AM (2005) Bacterial community dynamics and polycyclic aromatic hydrocarbon degradation during bioremediation of heavily creosote-contaminated soil. Applied and Environmental Microbiology 71:7008-7018 von Wintzingerode F, Gobel UB, Stackebrandt E (1997) Determination of microbial diversity in environmental samples: Pitfalls of PCR-based rRNA analysis. FEMS Microbiology Reviews 21:213-229 Vulava VM, McKay LD, Driese SG, Menn F-M, Sayler GS (2007) Distribution and transport of coal tar-derived PAHs in fine-grained residuum. Chemosphere 68:554-563 Wagner A, Blackstone N, Cartwright P, Dick M, Misof B, Snow P, Wagner GP, Bartels J, Murtha M, Pendleton J (1994) Surveys of gene families using 212

polymerase chain reaction: PCR selection and drift. Systematic Biology 43:250-261 Walter U, Beyer M, Klein J, Rehm HJ (1991) Degradation of pyrene by Rhodococcus sp. UW1. Applied Microbiology and Biotechnology 34:671- 676 Wang GCY, Wang Y (1997) Frequency of formation of chimeric molecules is a consequence of PCR coamplification of 16S rRNA genes from mixed bacterial genomes. Applied and Environmental Microbiology 63:4645- 4650 Wang R-F, Luneau A, Cao W-W, Cerniglia CE (1996) PCR Detection of polycyclic aromatic hydrocarbon-degrading Mycobacteria. Environmental Science & Technology 30:307-311 Watanabe K, Hamamura N (2003) Molecular and physiological approaches to understanding the ecology of pollutant degradation. Current Opinion in Biotechnology 14:289-295 Watts JEM, Wu QZ, Schreier SB, May HD, Sowers KR (2001) Comparative analysis of polychlorinated biphenyl-dechlorinating communities in enrichment cultures using three different molecular screening techniques. Environmental Microbiology 3:710-719 Weisburg WG, Barns SM, Pelletier DA, Lane DJ (1991) 16S ribosomal DNA amplification for phylogenetic study. Journal of Bacteriology 173:697-703 Weiss P, Schweitzer B, Amann R, Simon M (1996) Identification in situ and dynamics of bacteria on limnetic organic aggregates (Lake Snow). Applied and Environmental Microbiology 62:1998-2005 Wick LY, Pasche N, Bernasconi SM, Pelz O, Harms H (2003) Characterization of multiple-substrate utilization by anthracene-degrading Mycobacterium frederiksbergense LB501T. Applied and Environmental Microbiology 69:6133-6142 Wick LY, Ruiz de Munain A, Springael D, Harms H (2002a) Responses of Mycobacterium sp 501T to the low bioavailability of solid anthracene. Applied Microbiology and Biotechnology 58:378-385 Wick LY, Wattiau P, Harms H (2002b) Influence of the growth substrate on the mycolic acid profiles of Mycobacteria. Environmental Microbiology 4:612-616 Widmer F, Rasche F, Hartmann M, Fliessbach A (2006) Community structures and substrate utilization of bacteria in soils from organic and conventional fanning systems of the DOK long-term field experiment. Applied Soil Ecology 33:294-307 Wilhelm SW, Bullerjahn GS, Eldridge ML, Rinta-Kanto JM, Poorvin L, Bourbonniere RA (2006) Seasonal hypoxia and the genetic diversity of prokaryote populations in the central basin hypolimnion of Lake Erie: Evidence for abundant cyanobacteria and photosynthesis. Journal of Great Lakes Research 32:657-671 Wilson MS, Bakermans C, Madsen EL (1999) In situ, real-time catabolic gene expression: Extraction and characterization of naphthalene dioxygenase 213

mRNA transcripts from groundwater. Applied and Environmental Microbiology 65:80-87 Woese CR, Kandler O, Wheelis ML (1990) Towards a natural system of organisms: proposal for the domains archaea, bacteria and eucarya. Proceedings of the National Academy of Sciences of the United States of America 87:4576-4579 Xia XH, Yu H, Yang ZF, Huang GH (2006) Biodegradation of polycyclic aromatic hydrocarbons in the natural waters of the Yellow River: Effects of high sediment content on biodegradation. Chemosphere 65:457-466 Yamazoe A, Yagi O, Oyaizu H (2004) Degradation of polycyclic aromatic hydrocarbons by a newly isolated dibenzofuran-utilizing Janibacter sp strain YY-1. Applied Microbiology and Biotechnology 65:211-218 Yang X, Baumann PC (2006) Biliary PAH metabolites and the hepatosomatic index of brown bullheads from Lake Erie tributaries. Ecological Indicators 6:567-574 Yannarell AC, Triplett EW (2005) Geographic and environmental sources of variation in lake bacterial community composition. Applied and Environmental Microbiology 71:227-239 Yen KM, Gunsalus IC (1982) Plasmid gene organization - naphthalene salicylate oxidation. Proceedings of the National Academy of Sciences of the United States of America-Biological Sciences 79:874-878 Yen KM, Gunsalus IC (1985) Regulation of naphthalene catabolic genes of plasmid NAH7. Journal of Bacteriology 162:1008-1013 Yu CP, Chu KH (2005) A quantitative assay for linking microbial community function and structure of a naphthalene-degrading microbial consortium. Environmental Science & Technology 39:9611-9619 Yu Y, Breitbart M, McNairnie P, Rohwer F (2006) FastGroupII: A web-based bioinformatics platform for analyses of large 16S rDNA libraries. BMC Bioinformatics 7:57 Zhang R, Liu B, Lau SCK, Ki J-S, Qian P-Y (2007) Particle-attached and free- living bacterial communities in a contrasting marine environment: Victoria Harbor, Hong Kong. FEMS Microbiology Ecology 61:496-508 Zhou HW, Guo CL, Wong YS, Tam NFY (2006) Genetic diversity of dioxygenase genes in polycyclic aromatic hydrocarbon-degrading bacteria isolated from mangrove sediments. FEMS Microbiology Letters 262:148- 157 Zwart G, Crump BC, Agterveld M, Hagen F, Han SK (2002) Typical freshwater bacteria: an analysis of available 16S rRNA gene sequences from plankton of lakes and rivers. Aquatic Microbial Ecology 28:141-155 Zwart G, Hiorns WD, Methe BA, Van Agterveld MP, Huismans R, Nold SC, Zehr JP, Laanbroek HJ (1998) Nearly identical 16S rRNA sequences recovered from lakes in North America and Europe indicate the existence of clades of globally distributed freshwater bacteria. Systematic and Applied Microbiology 21:546-556

214

Zylstra GJ, Wackett LP, Gibson DT (1989) Trichloroethylene degradation by Escherichia coli containing the cloned Pseudonmonas putida F1 tolulene dioxygenase genes. Applied and Environmental Microbiology 55:3162- 3166

215

Vita

Jennifer Mary DeBruyn was born in Winnipeg, Manitoba, Canada, on November 2, 1980, to parents Edwin and Mary DeBruyn. Her early years were split between Winnipeg, where she attended Oakenwald School, and the Experimental Lakes Area, a Canadian government field research station in Northern Ontario where her father worked as a fisheries biologist. Her family later moved to Grimsby, Ontario, where Jennifer attended Smithville District Christian High School, graduating in 1998 as class valedictorian. She was awarded a full scholarship to attend Queen’s University in Kingston, Ontario. During the summers, she had the opportunity to work as a marine technician on the Great Lakes aboard the Canadian Coast Guard Ship Limnos; one project included a study of microbial community dynamics in Lake Erie for her Honours thesis, entitled “Microbial distributions in Lake Erie and the impact of phosphorus loading on bacterial production and proliferation” (DeBruyn et al. 2004). She graduated from Queen’s University in 2003 with an Honours Bachelor of Science degree in Biology. She received her Doctorate in Ecology & Evolutionary Biology from The University of Tennessee in 2008.

216