This item is only available as the following downloads:
1 DIVERSITY AND DISTRIBUTION OF AROMATIC RING DIOXYGENASES AMONG SOIL ACTINOBACTERIA By CHRISTOPHER ADAM WEIDOW A THESIS PRESENTED TO THE GRADUATE SCHOOL OF THE UNIVERSITY OF FLORIDA IN PARTIAL FULFILLMENT OF THE REQUIREMENT S FOR THE DEGREE OF MASTER OF SCIENCE UNIVERSITY OF FLORIDA 2013
2 2013 Christopher Adam Weidow
3 To my mother, Rebecca, and my brother, Matthew, for all their love and support these last four years. Their presence in my l ife during this time has been invaluable and often one of the few things that has kept me driven and confident in myself as a scientist and as a person. I would especially like to thank my mother, without whose love, understanding, and not too infrequent brow beatings I would have ended up a greeter at Wal Mart.
4 ACKNOWLEDGMENTS I would like to acknowledge the following people who have supported me in the years it took me to finish my degree and to those who have labored as long and as hard as I have to see this task completed. First, I would like to thank the members of my committee: Dr. Ashvini Chauhan, from Florida A&M University, for being so willing to go out of his way to participate in my research; Dr. John Thomas, of the Department of Soil and Water Sciences, here at the University of Florida for not only being a soundboard for bouncing ideas for my research back and forth, but also for being a source of much needed amusement and levity during intermittent periods of difficulty in my research and my writing; to my advisor, Dr. Andrew V. Ogram, of the Department of So il and Water Sciences here at UF, without whose enthusiastic support and unbelievably endless supply of patience I would have never made it past my second year. I would like to thank all my friends and co workers, here at the University of Florida, with whom I have had the honor and privilege of working these last four years: Moshe Doron, Hiral Gohil, Cassandra Medvedeff, Kimberly Johnson, Manmeet Waria, Alexandra Rosin, Piyasha Ghosh, Xi ao Rong, Haryun Kim, Stefani Schwarz, Elise Morrison, Michelle Amit, and others. I would like to give a very special thanks to Lisa Stanley who was pretty much my mom away from home the first two years of my research. I would like to thank Dr. Hee Sung B ae, without whose knowledge about molecular work this project would never have gotten off the ground. I would like to thank Dr. Abid Al Agely, our laboratory manager, for his friendliness and the opportunity I had to teach the microbial ecology lab course here at UF.
5 LIST OF ABBREVIATIONS ARD/s Aromatic Ring Dioxygenase/s MAH/s Monocyclic Aromatic Hydrocarbon/s PAH/s Polycyclic Aromatic Hydrocarbon/s PCoA Principal Component Analysis LMW Low Molecular Weight HMW High Molecular Weight OM Organic Matter pH Negative Log of the Hydronium Ion Concentration TKN Total Kjeldahl Nitrogen NH 3 N Ammonia based Nitrogen NO x N Nitrates/Nitrites based Nitrogen Fe Iron P Phosphorous MW McCarty Woods Soil, University of Florida, Gainesville EV Elk Valley Soil, Tennessee R14 Regal 14 Theater Soil, Archer Rd., Gainesville AS Scrubs Carwash on Archer Rd. Soil, Gainesville MS Saginaw River & Bay Site Soil, Tittabawassee River, Midland, Michigan NATL Natural Area Teaching Laboratory, University of Florida, Gainesville MUAC May sampled Unburned Arredondo Soil Control, NATL MUMC May sampled Unburned Millhopper Soil Control, NATL MBAA May
6 MBMA May MBMB May sampled Burn JUAC June sampled Unburned Arredondo Soil Control, NATL JUMC June sampled Unburned Millhopper Soil Control, NATL JBAA June JBMA June NATL JBMB June NBAA November
7 TABLE OF CONTENTS page ACKNOWLEDGMENTS ................................ ................................ ................................ .. 4 LIST OF ABBREVIATIONS ................................ ................................ ............................. 5 LIST OF TABLES ................................ ................................ ................................ ............ 9 LIST OF FIGURES ................................ ................................ ................................ ........ 10 ABSTRACT ................................ ................................ ................................ ................... 12 CHAPTERS 1 INTRODUCTION AND THE CONCERN WITH PAHs ................................ ......... 14 2 REVIEW OF THE LITERATURE ................................ ................................ ............ 18 Basic Cell Morphology ................................ ................................ ............................ 18 Environmental Distribution ................................ ................................ ...................... 19 Range of Metabolic Activities ................................ ................................ .................. 19 Naphthalene Degradation Pathways among Bacteria ................................ ............ 20 Enzymology of the Naphthalene Dioxygenase Enzyme ................................ ......... 24 Genome of Rhodococcus opacus Strain M213 ................................ ....................... 29 Effects of Forest F ire on PAH Levels and Distribution and Microbial Population Shifts ................................ ................................ ................................ .................... 31 3 DEVELOPMENT AND EVALUATION OF PCR PRIMERS TARGETING ACTINOBACTERIAL AROMATIC RING DIOXYGENASES ................................ ... 37 Approach/Hypothesis ................................ ................................ .............................. 37 Materials and Methods ................................ ................................ ............................ 38 Primer Development: PCR and Clone Libr ary Generation ............................... 38 Soil Descriptions ................................ ................................ ............................... 41 Measuring Soil Chemical Properties ................................ ................................ 44 Statistical Analyses of Data ................................ ................................ .............. 45 Comparison of narAa and nahAc in Select Soils ................................ .............. 46 Results ................................ ................................ ................................ .................... 47 Discussion ................................ ................................ ................................ .............. 53 4 EFFECT OF BURN TREATMENT ON ARD GENE DISTRIBUTION IN SOILS ...... 80 Approach/ Hypothesis ................................ ................................ .............................. 80 Materials and Methods ................................ ................................ ............................ 81 Soil Descriptions ................................ ................................ ............................... 81 Mole cular Analyses ................................ ................................ .......................... 83
8 Measuring Soil Chemical Properties ................................ ................................ 84 Statistical Analyses ................................ ................................ .......................... 84 Results ................................ ................................ ................................ .................... 85 Discussion ................................ ................................ ................................ .............. 92 5 ARD GENE DISTRIBTION SHIFTS IN SOILS INTENTIONALLY EXPOSED TO PAH VAPOR ................................ ................................ ................................ ......... 114 Approach/Hypothesis ................................ ................................ ............................ 114 Materials and Methods ................................ ................................ .......................... 114 Soil Descriptions ................................ ................................ ............................. 114 Soil Incubations ................................ ................................ .............................. 115 Molecular Analysis ................................ ................................ ......................... 116 Measuring Soil Chemic al Properties ................................ .............................. 117 Results ................................ ................................ ................................ .................. 118 Discussion ................................ ................................ ................................ ............ 121 6 CONCLUDING STATEM ENTS AND FUTURE WORK ................................ ......... 126 Conclusions ................................ ................................ ................................ .......... 126 Future Work ................................ ................................ ................................ .......... 127 LIST OF REFERENCES ................................ ................................ ............................. 130 BIOGRAPHICAL SKETCH ................................ ................................ .......................... 136
9 LIST OF TABLES Table page 2 1 A comparison of amino acid residues of the naphthalene dioxygenase alpha subunit in Pseudomonas putida G7 and several Actinobacteria. ........................ 36 3 1 List of microorganisms, accession numbers, a nd genes used in the deve lopment of degenerate primers. ................................ ................................ 58 3 2 Shannon Weaver and Chao1 diversity values for experimental soils (5% cutoff value). ................................ ................................ ................................ ................. 71 3 3 Soil chemical properties for Gainesville, Florida soils. ................................ ........... 71 3 4 Direct comparison of results of Mantel analyses for nahAc and narAa like ARD clone sequences from R14, AS, MBAA, MBMB, and JBAA soils. ...................... 78 3 5 Mantel results for soil nahAc like and Actinobacterial ARD like sequence clone libraries ................................ ................................ ................................ ............... 79 3 6 Degenerate codon abbreviations and their meanings. ................................ ........... 79 4 1 Summary of the changes in distribution and diversity of Actinobacterial ARD like sequences among NATL clone lib raries from May to June, 2012 .............. 110 4 2 Shannon Weaver and Chao1 diversity values for experimental soils. .................. 112 4 3 Soil chemical propert ies for May and June sampled NATL west soils. ............... 112 4 4 R and significance values for Mantel analyses of soil chemical properties for all NATL and Gainesville soil clone libraries (excluding M W, EV, and MS soils). 113 4 5 R and significance values for Mantel analyses of soil chemical properties for all NATL soil clone libraries without Gainesville soil clone libraries (also exclud ing MW, EV, and MS soils). ................................ ................................ .... 113 5 1 Comparison of the relative percentage of unexposed control and diesel exposed NBAA clone sequences that are divided between dif ferent phylogenetic groupings ................................ ................................ ................... 125
10 LIST OF FIGURES Figure page 2 1 Catabolic pathway for bacterial metabolism of naphthalene by the NAH7 plasmid of Pseudomonas putida G7. ................................ ................................ 35 3 1 Neighbor joining tree based on aligned nucleotide sequences for PAH degrading bacteria. ................................ ................................ ............................. 66 3 2 Binding sites for ActND O F2 and ActNDO R2 primers against Rhodococcus sp. NCIMB 12038 ................................ ................................ ................................ ..... 67 3 3 Comparison of PCR products derived from MW soil DNA and M213 colony PCR. ................................ ................................ ................................ ................... 68 3 4 Phylogenetic associations of R. opacus M213 PCR product and MW soil cloning results with known ARD sequences. ................................ ...................... 69 3 5 Geographical map of the continental United States showing the relative locations of sampling sites where soils were obtained for this portion of the research. ................................ ................................ ................................ ............ 70 3 6 PCoA analysis for experimental soils Generated by UniFrac. ................................ 71 3 7 Phylogenetic tree for clone library of R14 soil ARD sequences ............................. 72 3 8 Phylogenetic tree for clone library of AS soil ARD sequences. .............................. 73 3 9 Phylogenetic tree for clone library of EV soil ARD sequences. .............................. 74 3 10 Phylogenetic tree for clone library of MS soil ARD se quences. ........................... 7 5 3 11 Phylogenetic tree showing the associations of R14 nahAc like clone sequences with known Proteobacterial aromatic degradation genes from GenBank. ................................ ................................ ................................ ........... 76 3 12 Direct comparison of PCoA of clone libraries for R14, AS, MBAA, MBMB, and JBAA nahAc and narAa like ARD sequences. ................................ .................. 77 4 1 Location of NATL West on University of Florida Campus (ht tp://natl.ifas.ufl.edu/index.php). ................................ ................................ ................. 96 4 2 Locations of Burned (Red) and Unburned Control (Blue) sampling sites in NATL West (http://natl.ifas.ufl.edu/index.php). ................................ .............................. 97 4 3 Phylogenetic tree for clone library of MUAC soil ARD sequences. ........................ 98 4 4 Phylogenetic tree for clone library of MUMC soil ARD sequences. ........................ 99
11 4 5 Phylogenetic tree for clone library of MBAA soil ARD sequences. ....................... 100 4 6 Phylogenetic tree for clone library of MBMA soil ARD sequences. ...................... 101 4 7 Phylogenetic tree for clone library of MBMB soil ARD sequences. ...................... 102 4 8 Phylogenetic tree for clone library of JUAC soil ARD sequences. ....................... 103 4 9 Phylogenetic tree for clone library of JUMC soil ARD sequences. ....................... 104 4 10 Phylogenetic tree for clone library of JBAA soil ARD seque nces. ...................... 105 4 11 Phylogenetic tree for clone library of JBMA soil ARD sequences. ..................... 106 4 12 Phylogenetic tree for clone library of JBMB soil ARD sequences. ..................... 107 4 13 Phylogenetic tree for all May sampled NATL soil clone libraries against reference sequences. ................................ ................................ ....................... 108 4 14 Phylogenetic tree for all June sampled NATL soil clone libraries against reference sequences. ................................ ................................ ....................... 109 4 15 PCoA of May and June sampled NATL soil clone libraries against Gainesville a nd EV soil clone libraries. ................................ ................................ ............... 111 4 16 PCoA of all NATL soil clone libraries excluding additional regional soils. .......... 111 4 17 Shann on Weaver diversity plotted as a function of soil OM%. ........................... 113 5 1 Phylogenetic associations of unexposed control and diesel exposed NBAA clone ARD like sequences w ith known reference sequences. ........................ 124 5 2 PCoA analysis of Actinobacterial ARD like clone sequences for unexposed control and diesel exposed NBAA, MBAA, JBAA, R14, AS, and MW soils. ..... 125
12 Abstract of Thesis Presented to the Graduate School of the University of Florida in Partial Fulfillment of the Requirements for the Degree of Master of Science DIVERSITY AND DISTRIBUTION OF AROMATIC RING DIOXYGENASES AMONG SOIL ACTIN OBACTERIA By Christopher Adam Weidow May 2013 Chair: Andrew V. Ogram Major: Soil and Water Science Polycyclic aromatic hydrocarbons (P AHs) are composed of two or more fused aromatic (typically benzene) rings They result from the incomplete combustion of fossil fu els ( oil, gasoline ) and natural vegetation ( forest fires ) The Actinobacteria are environmentally ubiquitous Gram positive bacteria with a high G+C DNA c ontent that, while less well studied than the Proteobacteria ( i.e., Pseudomonads) have been implicated as important in the de gradation of various PAHs in soils. W e designed de generate PCR primers ( ActNDO F2 and ActNDO R2) around various, aligned Actinobacterial naphthalene dioxygenase (NDO) gene s that allow for the detection of multiple Rie ske type aromatic ring dioxygenases (ARDs) from soil DNA (i.e., naphthalene, phthalate dioxygenases ) S oils for testing were obtained from Gainesville ( FL ) Tennessee, and Michigan and represent ed a range of exposure to anthropogenic PAHs in addition to s everal paired burned/unburned forest soils PCR and clone library development revealed differences in the diversity and distribution of ARD like sequences across these s oils It was determined that the detectable distribution and diversity of Actinobacte rial ARDs was higher in the more directly impacted urban soils
13 than the less impacted forested soils (whether burned or unburned) Also, s equence diversity and distribution increased over time in soils subjected to perscribed burning compared to unburned soils Mantel analyses of soil clone libraries against select soil chemic al properties implicated soil organic matter content as a s ignificant driver of the diversity between the more directly impacted urban soils and the less/indirectly impacted forested soils. S ignificant correlation s (alpha = 0.05) between soil chemical properties could not be determined for grouped urban/directly impacted soils or forested/indirectly impacted soil s when considered separately
14 CHAPTER 1 INTRODUCTION AND THE CON CERN WITH PAHs Since the American Industrial Revolution, the widespread combustion of fossil fuels in the United States has steadily increased over time. Fossil fuels such as coal, oil, and refined gasoline are used to generate electricity and heat our ho mes as well as power the automobiles that many of us use as a means of personal transportation. Byproducts of fossil fuel processing are also widely used in other areas of our lives, roleum derivatives used in plastics. Along with this increased usage has come the subsequent increased contamination of our air, water, and soil with the byproducts of fossil fuel combustion, including sulfate and sulfide, various nitroxides, carbon monox ide and methane, and a broad range of compounds known as polycyclic aromatic hydrocarbons (PAHs). In addition to the combustion of fossil fuels, the incomplete combustion of naturally occurring organic matter, such as trees, shrubs, and other woody plants with a high lignin content, as in the case of forest fires, can also result in the production of PAHs (Kim et al., 2011; Vergnoux et al., 2011). PAHs are hydrophobic (lipophilic) compounds composed of two or more fused aromatic rings in linear, angular, or clustered patterns. Naphthalene is the most volatile of all the known PAHs and is included in a list of 16 PAHs compiled by the U.S. Environmental Protection Agency (EPA) for concentrated bioremediation efforts due to its potential as a carcinogen ( Mas trangelo et al., 1997; Preuss et al., 2003; Schreiner, 2003) Certain microscopic organisms are capable of utilizing these compounds as a source of energy and carbon. These organisms include the Bacteria and Archaea as well as both micro and macroscopic algae and fungi (Cerniglia et al., 1992).
15 There is one order of bacteria in particular which contains several groups of microorganisms that have been shown to metabolize these compounds as either sole carbon substrates or when supplemented with a second ary carbon source; the Actinobacteria. Research concerning the role of the Actinobacteria as agents in the bioremediation of environmental PAH contamination has become an important subject in environmental studies as our understanding of their biological and ecological roles has been expanded over the past decades. Many areas of research on the roles of PAH degradation b y the Actinobacteria are comparatively new, compared to research done with other groups of bacteria (such as the Pseudomonads), and inclu de kinetics, genomics, and ecological interactions between each other and other clades of microbes. One topic of particular interest to us is the diversity and distribution of Actinobacterial genes in soils that are related to the degradation of aromatic compounds, specifically those that are potential substrates and intermediates in the degradation of higher molecular weight PAHs, such as naphthalene. The enzymes responsible for the degradation of naphthalene and other high molecular weight (HMW) PAHs in the Actinobacteria may be evolutionarily related. In fact, many of these enzymes that degrade naphthalene also target additional HMW PAHs with similar chemical structures, such as anthracene, phenanthrene, and pyrene (Parales et al., 2003; Kim et al., 2008; Perez Pantoja et al., 2009). We hypothesize that the presence, distribution, and diversity of the genes that encode the enzymes of Actinobacterial aromatic ring dioxygenases (ARDs) responsible for the degradation of naphthalene and chemically simila r PAHs can be determined by utilizing basic molecular techniques based around the genetic similarities shared by these genes. Our
16 overall objectives are three fold: 1) identify regions of sequence similarity among these various Actinobacterial ARD genes a nd develop PCR primers that target as broad a range of these sequences as possible; 2) apply these primers in PCR for DNA extracted from geographically and chemically diverse soils representing a wide range of exposure to ant hropogenic PAHs; 3) a pply thes e primers in PCR conducted on non anthropogenically impacted soils that have been subjected to prescribed burn treatments. Our third objective is based on experimental evidence that shows that fire can affect the levels and distribution of PAHs in soils ( Kim et al., 2011; Vergnoux et al., 2011). Research has previously been conducted on examining the phylogenetic associations of enzymes related to the degradation of aromatics among various groups of bacteria (Perez Pantoja et al., 2009). However, these studies appear to have been limited to phylogenetic comparisons of sequence data from pure cultures entered into online national databases (such as GenBank at the National Institute of Health). To the best of our knowledge, this is the first attempt to de velop techniques for determining diversity and distribution of these genes in soil DNA. This research will extend the groundwork for a broad spectrum approach to examining soils for the presence of a broad diversity of Actinobacterial aromatic degrading b acteria. The data generated from this research will provide important insights into the phylogeny of these aromatic degraders present in soils and the diversity and distribution of their degradation genes, as well as to what environmental factors may be c ontrolling that diversity and distribution. We will begin this thesis by providing a brief overview of the information currently available in the literature so that the reader has the necessary background to
17 understand our experimental approach to our res earch. We will then outline and describe the methods and results of two phases of our research: soil molecular analyses of the diversity and distribution of Actinobacterial ARD like gene sequences; and the potential effect of forest fire on the distributi on of these sequences.
18 CHAPTER 2 REVIEW OF THE LITERATURE Basic Cell Morphology The term Actinobacteria refers to both the phylum and the class of Gram positive prokaryotes within the domain Bacteria that are characterized by a high guanine and cytosine co ntent in their DNA. Single cell morphology varies from spherical (coccus/coccoid), short rods with distinguishably rounded ends (rod coccoid/ rhodococcal), to longer, more rod shaped bacilli. More complex cellular associations can form either short or te mporary hyphal structures or more permanent, branching mycelia (Goodfellow & Williams, 1983; Ventura et al., 2007). Examples of Actinobacteria that display the rounded, coccoid form include the genera Micrococcus while the rod coccoid pattern can be found among the Arthrobacter. Nocardia (Goodfellow & Williams, 1983; Ventura et al., 2007) is representative of those Actinobacteria that form short or temporary hyphal chains. This particular type of morphology is common enough among the Actinobacteria that has become descriptive of any Actinobacteria which forms these short hyphal chains and which can fragment into individual coccoid or rod coccoid elements (Bell et al., 1998). In addition to the genus Nocardia from which the term w as derived, these Nocardioform Actinobacteria also include the genera Rhodococcus, Corynbacterium, and Mycobacterium (Bell et al., 1998). There is a particular type of branching morphology in Actinobacteria that is most well exemplified by, and most predom inant in, members of the Streptomyces This type of branching is complex and highly differentiated among individual groupings of cells. It is also permanent, with no observed natural fragmentation into smaller bodies (Ventura et al., 2007).
19 Environmental Distribution Members of the Actinobacteria are ubiquitous. The Actinobacteria are extensive in the terrestrial environment and numbers as high as 1 million Actinobacterial cells per gram of soil are attested (Goodfellow & Williams, 1983). Actinobacteria are also constituents of freshwater (Sekar et al., 2003; Warnecke et al., 2004; Holmfeldt et al., 2009) and marine microbial communities (e.g. Bull et al. 2005; Maldonado et al., 2005 ) though there is debate as to whether these Actinobacteria represent tr ue aquatic species or are simply terrestrial organisms that can survive under freshwater/marine conditions (Goodfellow & Williams, 1983). Almost all known genera of the Actinobacteria characterized to date can be found in soils. Those that can be found in both freshwater and marine sediments (as well as soils) are members of the Micromonspora, Rhodococcus, and Streptomyces while members of the Nocardia and Mycobacterium have been reported in marine sediments (Goodfellow & Williams, 1983). Actinobacteria have also been known to grow well in arid soils or soils with high pH and salinity as found in deserts or sand dunes near coastal areas. Under conditions of osmotic stress these bacteria produce compatible solutes such as the amino acid glutamine, or sug ars such as trehalose, which allow the organism to survive by maintaining isotonicity with its environment (Killham & Firestone, 1983; Avonce et al., 2006; Sudnitsyn, 2009). Range of Metabolic Activities The Actinobacteria are important in carbon cycling f or their ability to degrade recalcitrant substrates such as lignocelluloses and chitin (Phelan et al., 1979; Zimmerman, 1990; McCarthy & Williams, 1992) which are typically resistant to abiotic breakdown and difficult for microorganisms to decompose. Exce ptions to the bacterial
20 degradation of recalcitrant lignocelluloses are the white rot fungi, which can outcompete all known bacteria in the degradation of recalcitrant lignin (Zimmerman, 1990; McCarthy & Williams, 1992). Many Actinobacteria have been show n to be capable of degrading the same or similar compounds and in this manner contribute to the long term replenishment of endogenous carbon sources (McCarthy & Williams, 1992). In a related capacity, the Actinobacteria tend to be efficient degraders of a romatic compounds of anthropogenic origin (such as the aforementioned PAHs in Chapter 1). Because these pollutants often contain aromatics (and the chemical structure of lignin is composed of aromatic s cross linked with carbohydrates ), the decomposition o f PAHs and lignin by the Actinobacteria may involve similar mechanisms which rely on the presence of molecular oxygen and the activity of ring hydroxylating oxygenase enzymes (Cerniglia et al., 1992; Husain, 2008; Bugg et al., 2011; Mallick et al., 2011). Certain Actinobacteria are well known for producing a variety of both environmentally and commercially desirable surfactants. Surfactants produced naturally by microbes can either be low molecular weight compounds that function in lowering surface and int erfacial tensions between two immiscible substances, or higher molecular weight compounds which bind tightly to surfaces (Ron & Rosenberg, 2001). All surfactants, whether biological or anthropogenic in origin, are amphipathic compounds with both hydrophi lic and hydrophobic regions (Fietcher, 1992). Naphthalene Degradation Pathways among Bacteria What is currently known about the aerobic microbially mediated degradation of naphthalene, and other structurally similar PAHs (anthracene, phenanthrene, and py rene), is based primarily on work done with the Pseudomonads. Among these
21 to follow a generally consistent pattern with few, if any, variations. Additional research w ith the Actinobacteria, as well as other Proteobacteria besides the Pseudomonads, has revealed that naphthalene is not always degraded in the same manner and that differences, not only in the enzymes utilized, but the intermediates formed from naphthalene do occur. While the location, arrangement, and nucleotide/amino acid makeup of the genes that encode similar enzymes may vary among organisms, generally their functionality is similar if not identical. Also co nf ounding the issue is that many organisms so metimes contain enzymes/pathways that are not present in phylogenetically related species. Based on all the data that has been collected to date it can be difficult to effectively discuss the different ways that soil bacteria have to metabolize aromatic compounds without getting sidetracked by overly detailed discussions about the many enzymes that can be involved in any given mechanism, the genetics of those enzymes, and the phylogeny of organisms that have them. The task of reviewing the process in a c lear and simplistic manner, for all bacteria, can be truly daunting and may be impossible. In order to avoid any confusion as a result of discussing every known pathway for naphthalene degradation, this section will deal exclusively with naphthalene degra dation in a single model bacterium, Pseudomonas putida G7 (hereafter PpG7) as it is encoded on the NAH7 plasmid. Why do we choose PpG7 out of so many other hundreds of organisms? We do this simply because the PpG7system represents the most thoroughly stu died model for naphthalene degradation and the reader should keep that in mind when considering the broader issue. We recognize two things about using this model for our discussion: 1) PpG7, though being well studied, may not be
22 representative of all Pseu domonads and Proteobacterial systems, though similarities are likely to exist; 2) Actinobacterial systems show greater variation, in general, than the Proteobacteria in the degradation of naphthalene and, therefore, applying the details of this discussion to those systems should probably be avoided. Aromatic rings are resistant to cleavage because their resonance structures ( oscillating distributions of pi electrons through conjugated C=C double bonds) provide chemical stability. In order to break the ri ng open, so that terminal carbons can be reactive side groups. The following discussion is based on information found in Goyal and Zylstra (1997) and Habe and Omari (2003 ). The upper pathway of naphthalene degradation in PpG7 is summarized in Figure 2.1. The initial step in naphthalene degradation (also anthracene, phenanthrene, and pyrene) is the addition of two hydroxyls (OH ) to carbons 1 and 2 of the aromatic ring. This occurs through a process which involves a dearomatization and then rearomatization of the ring. First, naphthalene 1,2 dioxygenase ( nahAcAd ) adds both molecules of O 2 to the ring as cis naphthalene 1,2 dihydrodiol. This is then transformed into 1,2 dihydroxynaphthalene by naphthalene 1,2 dihydrodiol dehydrogenase ( nahB ) which rearomaticises the ring. By adding hydroxyls (which are a type of reactive oxygen species, ROS) to the 1 and 2 carbons of the ring, the electrons of 1,2 dihydroxynaphthalene are redistributed in a manner that makes the ring more susceptible to cleavage. But before the ring can actually be cleaved open, another reaction must occur. 1,2 dihydroxynaphthalene dioxygenase ( nahC ) converts 1,2 dihydroxynaphthalene into 2 hydroxy 2H ch romene 2 carboxylate in which one atom of oxygen is now substituted directly into the ring. Now
23 the chemical structure of the substrate has been modified enough to permit cleavage of the ring itself which is performed by 2 hydroxy 2H chromene 2 carboxylat e isomerase ( nahD ) to produce trans o hydroxybenzyldene pyruvate. trans o hydroxybenzyldene pyruvate hydratase aldolase ( nahE ) converts this into salicylaldehyde, which is finally converted into salicylic acid by salicylaldehyde dehydrogenase ( nahF ). The previous reactions constitute the upper pathway of naphthalene degradation in PpG7. The degradation of salicylate into acetyl CoA can then be carried out through two alternative intermediates, catechol and gentisate. For the catechol pathway, salic ylate hydroxylase ( nahG ) converts salicylate into catechol. Catechol itself can then be cleaved either ortho (between the two diols, which is a non NAH7 system) or meta (adjacent to the two diols; encoded on NAH7 plasmid). The meta cleavage pathway is mo re common in PpG7 and the enzymes involved are as follows: 1) catechol 2,3 dioxygenase ( nahH ); 2) hydroxymuconic semialdehyde hydrolase ( nahN ); 3) hydroxymuconic semialdehyde dehydrogenase ( nahI ); 4) 4 oxalocrotonate isomerase ( nahJ ); 5) 4 oxalocrotonate d ecarboxylase ( nahK ); 6) 2 oxopent 4 enoate hydrotase ( nahL ); 7) 2 oxo 4 hydroxypentanoate aldolase ( nahM ); and 8) acetaldehyde dehydrogenase ( nahO ) which produces acetyl CoA. The less common ortho cleavage pathway is not encoded on the NAH7 plasmid and ut ilizes different enzymes than found in the meta pathway: 1) catechol 1,2 dioxygenase; 2) a cis,cis muconate lacton izing ketoadipate enol ketoadipate:succinyl ketoadipyl CoA thiolase which produces both succinyl and acetyl CoA.
24 The gentisate pathway is also encoded on the NAH7 plasmid of PpG7 but only has enzymes that convert salicylate to pyruvate and fumarate and not all the way to acetate. Those enzymes are most likely encoded elsewhere. The first step is a conversion of salicylate to gentisate by salicylate 5 hydroxylase ( nagGHAaAb ). Gentisate is converted to maleylpyruvic acid by gentisate 1,2 dioxygenase ( nagI ) and then to fumarylpyruvic acid by maleylpyruvic acid isomerase ( nagL ). Finally, pyruvate and fumarate are made from fumarylpyruvic acid by fumarylpyruvate hydr olase ( nagK ). Additional degradation to acetyl CoA is carried out by enzymes encoded elsewhere. Once acetyl CoA is generated it is then channeled into the tricarboxylic acid (TCA), or Krebs cycle in aerobic organisms. The TCA cycle provides various impor tant substrates and compounds for cell growth and metabolism including amino acid precursors as well as the reducing agent NADH. The TCA cycle is also an important process in producing cellular ATP because it is coupled to the oxidative phosphorylation pa thway in cell membranes (the electron transport pathway) though this only occurs in aerobic organisms. Although detoxification and elimination of potentially deadly and mutagenic compounds may be one purpose of aromatic metabolism by bacteria, it is also the contribution of important metabolites to cell growth and energy production that makes the microbially mediated degradation of aromatics beneficial to the cell. Enzymology of the Naphthalene Dioxygenase Enzyme Both the Pseudomonas and Rhodococcus napht halene dioxygenase are Rieske nonheme iron oxygenase enzymes characterized by a unique 2Fe 2S electron transfer site and a mononuclear iron catalytic site that is not typical of other enzymes involved in the degradation of aromatics (such as flavoprotein m onooxygenases and soluble diiron oxygenases) (Perez Pantoja et al., 2009). This enzyme is designated as ISP NAR (for
25 I ron S ulfur P rotein NA phthalene R hodococcus ) (Larkin et al., 1999) and may be used as a model for other Actinobacterial naphthalene dioxyge nases (NDOs). The enzyme is a trimer dimer in which 3 large alpha subunits composes one half of the dimer and 3 small beta subunits composes the other half. Larkin et al. (1999) found that ISP NAR made up approximately 6.6% of the total protein found in N CIMB 12038 while the corresponding concentration of ISP NAP ( I ron S ulfur P rotein NA phthalene P seudomonas ) in Pseudomonas sp. NCIMB 9816 4 was approximately 0.9% of the total cell protein. The relatively higher percentage of cell protein dedicated to PAH me tabolism may, in part, support the claim that the Rhodococci might be more efficient than the Pseudomonads as agents of PAH bioremediation. Parales (2003) states that the ISP NAP system is a multienzyme system with three components. The first component i s an iron sulfur center containing flavoprotein reductase. The second component is a Rieske ferredoxin protein containing enzyme where the active site is composed of a [2Fe 2S] cluster. Both the iron sulfur flavoprotein reductase and the Rieske ferredoxi n are responsible for the transfer of electrons from reduced nicotinamide adenine dinucleotide phosphate [ NAD(P)H ] to the dioxygenase component of the system, which constitutes the trimer dimer of alpha and beta subunits mentioned above. Each of the three alpha subunits of the dioxygenase component contains a Rieske [2Fe 2S] center in addition to a mononuclear iron active site located at junctions connecting the three alpha subunits. The same structure for the ISP NAR enzyme was determined by Larkin et al. (1999) At this time, there is little experimental evidence as to the specific function of the beta subunit. Larkin et al. (1999) showed that degradation of naphthalene could occur in the presence of whole,
26 cell free enzyme or in the presence of purifie d alpha subunit only. However, when purified beta subunit was tested they did not observe naphthalene degradation suggesting that its presence it not necessary for catalysis. Overall, the ISP NAR/NAP enzyme appears to be structurally related to the cytochr ome bc 1 complexes involved in the cellular Q cycle of electron transport. The amino acid residues which make up the Rieske center and the mononuclear iron active site of the dioxygenase component of ISP NAR/NAP are highly conserved. When the ISP NAR sequen ce of Rhodococcus sp. NCIMB 12038 is aligned against the ISP NAP sequence of Pseudomonas and other naphthalene degrading bacteria (see Table 2.1) the portion of the sequences that encode the Rieske and enzyme active sites are identical in both amino acid co mposition and positioning from the sequence starting point except that Pseudomonads utilize glutamate at amino acid position 229 while Actinobacteria utilize aspartate (Larkin et al., 1999). More variation between the ISP NAR and ISP NAP sequences can be se en in the amino acid residues between those encoding essential catalytic functions. The high degree of amino acid homology for residues encoding the active sites across phylogenetically diverse Pseudomonads and Actinobacteria, where they occur, may be evi dence that even single residue substitutions could result in a loss of enzyme function. It was noted by Parales (2003) that the naphthalene dioxygenase of Pseudomonas sp. can oxidize over 60 aromatic compounds. Multiple substrate usage by several Rieske nonheme iron oxygenase enzymes among a broad range of microorganisms was further emphasized in review by Perez Pantoja et al. (2009). It has also been shown, in certain Mycobacterium species, that Rieske type enzymes are
27 responsible for the degradation o f naphthalene, pyrene, and phthalic acid (Perez Pantoja et al., 2009; Kim et al., 2008; Kweon et al., 2010). These Rieske enzymes are said to attack various mono and polycyclic aromatic compounds where structural similarities occur. Based on the conclus ions of Larkin et al. (1999), it may be reasonable to apply the same considerations to ISP NAR A typical Rhodococcus genome is very large when compared to many other bacteria. Rhodococcus sp. strain RHA1 has a chromosome that is 9.7 million bases in size. In addition to the chromosome, RHA1 harbors three linear plasmids (pRHL1, pRHL2, and pRHL3) which range between 330,000 and 1 million basepairs in size and contain many transposable elements (Larkin et al., 2005). Gene redundancy appears to play a role in the metabolic versatility and efficiency of many Rhodococci In RHA1, the chromosome and three plasmids encode at least 6 ring hydroxylating dioxygenases (RHDOs) and ten cytochromes P450 in addition to many other mono and dioxygenases. Other Rhodococ ci genomes are comparable in size and complexity, with R. aetherovorans strain I24 and R. erythropolis strain PR4 both having chromosomes around 7 million bases (Larkin et al., 2005). Furthermore, transcription and translation of these enzymes in the Rhod ococci proceeds in a manner that appears to have an increased effect on enzyme efficiency. The catalytic dioxygenase enzyme in naphthalene degradation requires the assistance of a reductase and ferredoxin protein (Parales, 2003). Larkin et al. (2005) po ints out that cytochrome P450 redox partners (such as various ferredoxins and reductases) are typically transcribed and translated separately from the cytochrome itself in many bacteria. However, the cytochrome P450 of R. rhodocrous strain 11Y has its ass ociated flavodoxin bound to the N terminus of its
28 own sequence, effectively generating the required co substrate polypeptide along with its own. This may facilitate more rapid enzyme response than would be found in microorganisms that generate their cytoc hromes P450 and reduced co substrates separately on the chromosome/plasmid. Kulakov et al. (2005) also report ed on the variability in gene arrangement in the Rhodococci for naphthalene degradation. Specifically, they contrast ed the relative stability of the genes responsible for naphthalene degradation in P. putida with those found in several Rhodococci by looking at how these genes are physically arranged within the genome. In Pseudomonas putida strain PpG7, naphthalene degradation is encoded by the NAH 7 plasmid. This plasmid contains both the upper pathway of naphthalene degradation, which converts naphthalene to salicylate, and the lower pathway, which converts salicylate to pyruvate and acetyl CoA, in a back to back position on the plasmid. In contr ast, the upper pathway of naphthalene degradation in Rhodococci which takes naphthalene down to salicylate (or phthalate) appears to be arranged separately from each other in the chromosome or plasmids (Kulakov et al., 2005). The genes for the lower pathw ays of naphthalene degradation in Rhodococci, as induced by naphthalene and the operon on which they are located have not yet been described to the best of our knowledge. Kulakov et al. (2005) comment that the naphthalene degradation genes in the Rhodococ ci are often flanked by open reading frames (orfs) whose sequences often resemble transposable elements. The large number of these orfs in the genomes of various Rhodococci may have played a critical role in the development of such highly flexible and var iable genomes. In contrast to the
29 larger diversity of gene arrangements in the Rhodococci (having genes present in both the chromosome and on plasmids) the enzymes responsible for PAH degradation in Pseudomonads appear to be located exclusively on plasmi ds (Barnesly, 1976). The NAH7 plasmid of P. putida G7 (a representative naphthalene degrading Pseudomonad ; Uz et al., 2000; Filonov et al., 2004), which contains the sequences for the ISP NAP enzyme, is only 83 thousand bases in size and contains only one copy of the gene per plasmid (Izmalkova et al., 2006). The discontinuous arrangement, or mosaic pattern, of aromatic degradation genes in the chromosome of the Rhodococci species described above appears to be common among all Nocardioform Actinobacteria. This same type of organization has been well studied in other organisms such as Mycobacterium vanbaalenii PYR 1 (Kim et al. 2011) for which the actual location of genes encoding enzymes in both the upper and lower pathway of aromatic degradation are know n. Genome of Rhodococcus opacus Strain M213 Our research is part of a larger, ongoing project examining the prevalence of naphthalene degradation pathways in soils which are related to a naphthalene degradation pathway in Rhodococcus opacus M213 (hereafter M213) Unlike the PpG7 pathway and other Actinobacteria ( Rhodococcus sp. NCIMB12038), M213 does not appear to grow on salicylic acid nor produce it as an intermediate in naphthalene degradation. Instead, M213 appears to degrade naphthalene to o phthali c acid and then to protocatechuate. In 2000, at the time that this research was first started, it was believed that this phthalic acid pathway was novel among the Actinobacteria and soil bacteria in general. Since then, aromatic degradation with phthalic acid and protocatechuate as intermediates has been found among various groups of bacteria, including the Proteobacteria, though the specific mechanisms through which it is
30 metabolized varies between taxonomic groups (Kim et al., 2008; Perez Pantoja et al. 2009; Kweon et al., 2010). In addition to our research, collaborators of ours from the Florida A&M University (FAMU) School of the Environment (SE) have been working specifically on examining the genome of R. opacus M213 and ascribing metabolic functio ns to sequence data. As part of a manuscript currently in preparation, Pathak et al. (2013) have determined much about the genome of M213. In addition to presence of several linear and circular plasmids, Pathak et al. showed that the chromosome of M213 i s approximately 9.2 million bases in length with approximately 8,942 putative genes being encoded by the chromosome alone. Heat mapping and principal component analysis PCoA against the genomic data from eight Rhodococci species on Genbank indicate d that 75.14% of the remaining 24.06% of the DNA does not associate with any known functions. Of the 75% of protein encoding DNA, approximately 22% appears to be related to cell me tabolism with an estimated 401 genes similar to those found in the degradation of multiple PAHs [naphthalene, phenanthrene, anthracene, and b enzo(a)pyrene] as well as halogenated aromatics and aromatic hydrocarbons, including several pesticides ( including DDT and atrazine). As far as naphthalene metabolism by M213 is concerned, Pathak et al. identif ied the presence of genes similar to those involved in the production and metabolism of both phthalic acid and salicylate, even though a previous study indicated that salicylate was unlikely to be produced or utilized by M213 (Uz et al., 2000). The previous study by Uz et al. was based largely on culture dependent/selective media methods and
31 techniques and did no t involve any genetic analyses. The presence of ge nes related to both phthalate and salicylate metabolism in M213, as determined by Pathak et al., suggests that both pathways for naphthalene degradation are possibly present in M213 though an explanation for why salicylate utilization is not readily observ ed by pure cultures of M213 on selective media is still forthcoming. Effects of Forest Fire on PAH Levels and Distribution and Microbial Population Shifts In C hapter 1 we introduced the concept that forest fires, whether naturally occurring or as a result of human activities, often result in the production of PAHs as a result of the incomplete combustion of aromatic rich organic matter. The production of PAHs from this process would obviously impact the PAH concentrations and distribution in the upper leve ls of the soils where these fires have the most direct interaction. In response to these changes we may even expect the overall microbial community structures of impacted soils to change as new carbon substrates are added and environmental conditions chan ged. Several studies conducted over the last decade have provided us with interesting data on the overall impacts of forest fire to changes in PAH concentration and microbial community structure in soils. Kim et al. (2011) looked at the concentrations of 16 priority PAHs from forest sites at risk for fire at five locations in the mountains along the east coast of South Korea, well away from any urban/industrial sources of PAHs. These samples, in addition to control samples, were collected at 1, 5, and 9 months after fire from 0 5 cm below the soil surface Samples were both collected and stored in Teflon bags and they were stored at 5 C until analyses were
32 performed. Briefly, within a month after burning, sites A, B, and C showed a dramatic increase in the concentrations of PAHs with the majority being made up of PAHs with 2 to 3 rings (3 of the16 PAHs not detected in sites A and B after the first month were present after 5 months). After five months, overall PAH concentrations decreased significantly across all 3 soils and the approximate proportion of PAHs with 2 3 rings to higher molecular weight PAHs also decreased. Three of the16 PAHs that were detectable after 1 month in site C samples were not detected at 5 months. Nine months after fire, the overall concentrations of PAHs had increased slightly in sites A and B from values obtained after five months, but site C showed a n even further decrease in PAH concentrations. There was still a decrease across all sites with respect to the relative proportion of low molecular weight PAHs to higher molecular weight compounds. Kim et al. concluded that forest fire was a significant contributor of low and high molecular weight PAHs to soils and that the majority of these PAHs were initially composed of 2 to 3 rings. They also concluded that the lower molecular weight compounds would be rapidly mineralized or volatilized from soils, o r even physically removed by wind or rain, within a nine month period after burning. Those PAHs remaining in burned soils after nine months would be dominated by the most recalcitrant compounds and be composed of 4, or more, rings. Kim et al. (2011) int erpret ed their data in support of PAH concentrations returning to approximately control levels after nine mon ths. A closer look at their data shows that few PAHs decrease to control levels, even 9 months after fire, and that the concentrations of total PA Hs in all three soils (A, B, and C) still exceed the control by 4 times. This data more appropriately support the conclusion that forest fire results in
33 both a rapid increase, and then decrease, in PAH concentrations with degradation rates slowing over ti me. While Kim et al. (2011) studied the effects of forest fire on the production and distribution of PAHs in soils, Khodadad et al. (2010) examined the effect of biochar addition on the structures of soil microbial communities. Khodadad et al. employed th ree types of biochar: grass based biochar, combusted at 650 C (Grass650) ; oak based biochar, combusted at 250 C (Oak250) ; and oak based biochar, combusted at 650 C (Oak650) Two types of soil were utilized: a soil from an upland hardwood forest with no hi story of burning (unburned); a soil from a longleaf slash pine palmetto forest that received periodic burn treatments (burned). Separate samples of unburned forest soil were amended with Grass650, Oak250, and Oak650 biochar, respectively. Burned soil was only amended with Oak650 biochar. Unamended soils from both sites were utilized as controls. Three types of analyses were performed on these soils: 1) CO 2 evolution measurement (as an indicator of microbial respiration rates); 2) 16S rRNA gene copy numb er (as an indicator of gene copy numbers ; 3) formation of colony forming units (CFUs) on solid media (also an indicator of diversity). Briefly, the results of Khodadad et al. (2010) indicate that there is a combination of effects based on the source of t he biochar used and whether a soil has had previous exposure to fire. When unburned soils were exposed to biochars formed at lower temperatures (Grass250; Oak250) the results showed a decrease in microbial diversity and an increase in overall respiration rates. The authors suggest that the lower
34 aliphatic compounds when compared to woody materials. The increased bioavailability in the lower temperature biochars leads to a combination of increased degradation of carbon substrate (increased respiration rates) and a decrease in microbial diversity as the labile substrates select for more ra pidly growing species of soil bacteria at the expense of other species. Interestingly, when higher temperature biochar (Oak650) was added to unburned soil it resulted in both increased respiration and biodiversity. Although overall respiration rates in crease compared to unamended soil, the increase is smaller than with amendments of low temp biochars. The authors g a ve the following explanation: addition of high temp biochar to soil increases the total carbon pool available for microbial respiration and therefore also increases measurable respiration rates. However, because the high temp biochar is more complex/less labile than the low temp biochar, faster growing microbes that are typically able to outcompete slower growing microbes for a single carbon source no longer have a metabolic advantage, allowing slower growing organisms, or those with more metabolic diversity, an opportunity to grow in numbers which results in higher microbial diversity. Addition of biochar to previously burned soil did not r esult in a statistically significant increase in respiration rates but did increase microbial diversity compared to unamended control. The microbial community composition of these soils showed a trend of increasing numbers of Actinobacteria and Gemmatimo nadetes for addition of high temp erature, recalcitrant biochars.
35 Figure 2 1. Catabolic pathway for bacterial metabolism of naphthalene by the NAH7 plasmid of Pseudomonas putida G7 Genes written in red represent the action of the following enzymes: na hAcAd large and small subunits of naphthalene 1,2 dioxygenase; nahB cis naphthalene dihydrodiol dehydrogenase; nahC 1,2 dihydroxynaphthalene dioxygenase; nahD 2 hydroxy 2H chromene 2 carboxylate isomerase; nahE trans o hydroxybenzyldene pyruvate hydra tase aldolase; nahF salicylaldehyde dehydrogenase; nahG salicylate hydroxylase; nagGHAaAb salicylate 5 hydroxylase [t aken and modified from Goyal & Z ylstra. 1997 Genetics of naphthalene and phenanthrene degradation by Comamonas testosteroni (Page 402, Figure 1). Society for Industrial Microbiology).
36 Table 2 1. A comparison of amino acid residues of the naphthalene dioxygenase alpha subunit in Pseudomonas putida G7 and sever al Actinobacteria. R esidue numbering is based on our ClustalX2 alignments of NDO large, alpha subunits. Rieske Center and Active Site Residues Dioxygenase Active Site Residues Pseudomonas putida G7 Rhodococcus, Mycobacterium, Arthrobacter Pseudomonas putida G7 Rhodococcus, Mycobacterium, Arthrobacter Cysteine 103 Cysteine 103 As paragine 230 Asparagine 230 Histidine 105 Histidine 105 Histidine 237 Histidine 237 Arginine 106 Arginine 106 Histidine 242 Histidine 242 Cysteine 123 Cysteine 123 Aspartate 234 Aspartate 234 Histidine 126 Histidine 126 **Aspartate 394 **Aspartate 394 Tryptophan 128 Tryptophan 128 Valine 139 Valine 139 *Glutamate 229 *Aspartate 229 *Pseudomonads utilize glutamate while Actinobacteria utilize aspartate at position 229 of the Rieske site of the alpha subunit. **Asp 394 is 1 residue further away from Asp 234 in our determinations than the positionin g given by Larkin et al. (1999)
37 CHAPTER 3 DEVELOPMENT AND EVALUATION OF PCR PRIMERS TARGETING ACTINOBACTERIAL AROMATIC RING DIOXYGENASES Approach/Hypothesis Our overarching goal in this part of the w ork was to evaluate the diversity of ring hydroxylating enzymes involved in the degradation of PAHs by soil Actinobacteria. Specifically, we would like to develop reliable methods which will allow us to analyze soils for the presence of DNA sequences simi lar to those encoding aromatic ring dioxygenases for PAH degradation by Actinobacteria and to evaluate the ir distribution in soils. The development of these techniques will primarily revolve around detection of genes encoding individual enzymes; however, a secondary purpose of these techniques will be to allow us to evaluate the distribution of specific genotypes in soils collected from disparate sources. The development of such techniques targeting diverse aromatic degrading genes in the Actinobacteria, as a whole, is a relatively unexplored area in current research as many past experiments have focused on the ability of an organism to metabolize single compounds. Despite this more singular focus of past research, some representative Actinobacteria have been examined over the past few decades with respect to their ability to contribute to the overall degradation of PAHs in soils. Therefore, we proceeded in the development of molecular methods based on the following rationale: genetic data (DNA and protei n sequences) for a range of Actinobacteria capable of degrading PAHs is currently available from a public database (GenBank; http://www.ncbi.nlm.nih.gov/genbank/ ); these data can be compiled and compare d such that regions of high sequence similarity in functional genes can be identified; these sequences can be used in developing PCR primers for aromatic dioxygenase genes specific to the Actinobacteria.
38 We hypothesize that the broad range of enzymes respo nsible for the degradation of various PAHs (in both Gram negative and Gram positive bacteria) have evolved from degrading enzymes to retain various sequence motifs and protein structure, regardless of substrate specificity. By designing degenerate primers based on comparisons of functional genes present in various PAH degradation pathways we will be able to test soils for the presence of a broad range of PAH degrading Actinobacteria and pathways, and analyze information associated with the evolutionary relationships among different PAH degrading enzymes. This research will also be useful for detecting the presence of multiple Actinobacterial ARD sequences as well as showing, in a semi quantitative man ner, the distribution of these sequences across a range of geographically and chemically diverse soils. Application of these primers to soils, and analysis of the clone libraries resulting from their use, will provide us a useful means for examining the p resence of sequences similar to genes known to be important in the microbially mediated degradation of contaminating PAHs of anthropogenic origin. Materials a nd Methods Primer Development: PCR and Clone Library Generation ISP NAR belongs to a large class of enzymes known as the Rieske non heme iron oxygenases All of these enzymes share a similar arrangement of subunit proteins (a trimer dimer of enzymes because they possess a highly conserved [ 2Fe 2S ] electron transfer center coordinated by 2 histidine and 2 cysteine residues (respectively) They also have a mononuclear iron cat alytic site responsible for the introduction of 2 oxygen atoms from molecular O 2 (Parales et al., 2003; Perez Pantoja et al., 2009). They have been shown
39 to play a role in the degradation of multiple polycyclic aromatic compounds such as naphthalene, phen anthrene, and pyrene and monocyclic aromatics such as phthalate and protocatechuate (Kim et al., 2008; Perez Pantoja et al., 2009; Kweon et al., 2010). Over 100 amino acid and nucleotide based sequences from a range of phylogenetic groups, including Act inobacteria, were retrieved from GenBank which were related to enzymes responsible for metabolizing the following substrates: naphthalene; phthalic acid; salicylic acid; gentisate; catechol; methyl catechol; cinnamic acid; and similar sequences for unspeci fied aromat ics and PAHs (see Table 3 1). Sequence alignments were generated using ClustalX2 (Thompson et al., 1997) and phylogenetic relationships were determined by neighbor joining trees generated using MEGA 5.0 (Tamura et al., 2011). Nucleotide sequen ces from the group highlighted in red font in Figure 3 1 were used to create a pair of degenerate PCR primers based on, and supposedly targeting, the Rieske protein and mononuclear iron enzyme active site of the large alpha subunit of the Actinobacterial n aphthalene dioxygenase (ISP NAR ) (Figure 3 2) Primer specificity was verified by using our own Touchdown PCR protocol conducted on DNA from Rhodococcus opacus M213 ( narAa positive control) and soil DNA from a nearby soil on campus. Concentrations of DNA in extracts were not quantified. A 0.1 dilution of soil DNA was used in PCR. For M213 PCR and cloning, colony material was added to PCR reagents using a flame sterile inoculation needle. DNA was stored at 20 C for the duration of the research in the e lution buffer that was provided in the DNA extraction kit (QIAGEN Inc., Valencia, CA, USA) For PCR, we used Promega GoTaq 2x Green Master Mix with the following addition of reagents: 12.0 L Master Mix; 5.0 L of 10 M ActNDO F2 (2.0 M final concentrati ons); 5.0 L of 10
40 M ActNDO R2 (2.0 M final concentrations); 2.5 L of 0.1 soil DNA extract or water (for both negative control and M213 colony PCR). The final volume of each PCR reaction was 25 L. Cycling was carried out on a Bio Rad model T100 therm ocycler (Bio Rad, Hercules, CA, USA) as follows: 1 cycle of 94 C for 3 minutes; 6 cycles of 94 C for 40 seconds, 55 C 49 C for 30 seconds (decrease 1 C per cycle), and 72 C for 1 minute; 30 cycles of 94 C for 40 seconds, 50 C for 30 seconds, and 72 C for 1 minutes; and 1 cycle of 72 C for 7 minutes. Gel electrophoresis was performed on PCR products using a 1.5% w/v agarose gel in 1x TAE buffer. PCR products from soil DNA were purified us ing QIAquick Gel Extraction Kit (QIAGEN Inc., Valencia, CA USA) w hile PCR products from M213 were purified using QIAquick PCR Purification Kit ( QIAGEN Inc., Valencia, CA USA ) A clone library of PCR products from our testing soil was developed using a TOPO TA Cloning Kit utilizi ng pCR4 TOPO cloning vector and One Shot TOP10 Electrocompetent E. coli culture from Invitrogen, Life Technologies Corporation (Grand 4 L of gel purified PCR product was incubated for 12 to 15 hours at room temperature in a mixture with 1 L cloning vector and 1 L 0.25 x MgCl 2 solutions. Afterwards, 2 L of this ligation mixture was added to 50 L ice thawed One Shot TOP10 E. coli and electroporated at 1.62 kV in a c hilled electroporation cuvette. The electroporated culture was then transferred back into One Shot culture vial along with 250 L of room temperature S.O.C. medium and allowed to incubate in a spinning hybridizer for 60 minutes at 37 C. After this incuba tion, a small subsample of the culture was diluted in a
41 separate vial and aliquots were spread plated on 1.5% agar Luria Bertani plates containing, per liter of deionized water, 10 g tryptone, 10 g sodium chloride, 5 g yeast extract, and 50 mg kanamycin. The remaining, unused culture was stored at 80 C in 100 L glycerin. The plates were incubated overnight until development of individual colonies occurred. Patch plates containing the same concentration of antibiotic were prepared by selecting individua l colonies from the dilution spread plates prepared the previous evening and were once again left to incubate overnight at 37 C. The following day, colony PCR utilizing the same reaction formula and cycling conditions used to generate the original PCR pr oducts was performed using randomly selected individuals from patch plates to verify the presence of the insert. Forty eight individual colonies were selected from these patch plates and re cultured in the wells of a 96 well plate containing 200 L of 50 mg/L k anamycin Luria Bertani broth and 8% v/v glycerin. The 96 well plate was left to incubate, overnight, at 37 C. The following day, this 96 well plate with cultures was submitted, along with purified M213 PCR product, to the Sanger Sequencing Lab at t he Interdisciplinary Center for Biotechnology Research (ICBR) at U niversity of Florida, Gainesville, FL for sequence determination. Sequencing was done utilizing our own forward primer, ActNDO F2. The results of the MW clone library, together with retur ned M213 sequence, were compared to sequences on GenBank as part of the validation process. Soil Descriptions Soils were collected from geographically disparate sources (see Figure 3.5 ) and represent a range in exposure to anthropogenic sources of PAHs. M cCarty Woods (MW) is a small, forested site located at the intersection of Museum Rd. and Newell Dr.
42 on campus at the University of Florida. The nearby intersection is heavily traversed by cars trucks, and city buses and, therefore, consistently exposed to fossil fuel based PAHs in the form of exhaust deposition. DNA extracted from t his soil was used during the initial testing phase of the degenerate primers that we developed Regal 14 Theater (R14) soil was obtained from a parking lot drainage ditch f or a local movie theater in Butler Plaza on Archer Rd., Gainesville, Florida. The theater receives high levels of automotive traffic, both from vehicles driving through the parking lot itself and from traffic on adjacent streets. PAH inputs into the drai nage ditch soil are from a combination of parking lot runoff, during rain events, and exhaust deposition. Another drainage ditch soil was obtained from a Scrubs carwash (AS) on Archer Rd., Gainesville, Florida. This soil was composited from samples taken from several drainage ditches situated, inline, on a slight downward slope leading away from the carwash. In addition to PAHs, which are input as a combination of exhaust deposition and runoff caused by both rain events and customer usage, the soil also receives high inputs of various soaps, detergents, and other chemicals common in automotive maintenance. Additionally, s oil was obtained from beneath a long standing (more than 50 years) pile of coal from a densely forested soil situated behind an isolate d residence in the mountainous region of Elk Valley (EV), Tennessee. The soil that was obtained had visible rock fragments of various sizes as well as varying sizes of chipped coal The area was well isolated from any roads and receives very little autom otive or human traffic. PAH inputs into the soil are likely from coal leachate produced by physical and chemical wearing of coal and washed into the soil during rain events. Finally, soil was collected from an area adjacent to the Saginaw River & Bay Sit e of the Tittabawassee
43 River in Midland, Michigan (MS). This urban river is heavily trafficked by private and commercial watercraft and has a history of exposure to dioxins and dioxin like compounds (DLCs). Dioxins are structurally similar to PAHs in tha t they are aromatic molecules with various numbers and arrangements of rings but where 2 carbons in the ring are replaced by oxygen. DLCs are also aromatic compounds that behave similarly to dioxins, but have only 1 oxygen substitution (as in furans) or n on e at all (as in biphenyls) Dioxins and DLCs are usually much more toxic and carcinogenic than PAHs (Schecter et al., 2006). Additional soils, not listed here, that were used in our comparison of narAa and nahAc like ARD sequences (see Comparison of narAa and nahAc in Select Soils section, below) are described in Chapter 4 because they were primarily collected for work that is specifically relevant to that chapter. Gainesville soils were collected by hand using a small hand shovel that was marked up to 15 cm from the tip. All s amples were taken from approximately the top 10 cm of the soil surface, as judged from the markings on the hand shovel and transported back to laboratory in a small cooler, inside plastic whirl pak bags. Soils were then sieve d through a clean, 2 mm metal mesh screen and then divided into two stocks, one for molecular work and one for potential culture dependent analyses. Stocks to be used for molecular work were stored at 20 C or 80 C when available. Stocks for culture bas ed work were stored inside a sealed cooler on a bench. It was believed that storing soils inside a cooler, at slightly less than room temperature, would reduce the impacts on soil microbial communities resulting from temperatures that were too low or too high. EV soil was collected using a large two handed shovel due to the rocky nature of the mountain soil. The exact depth of the samples taken was not measured but attempts
44 were made to sample less than 15 cm below the soil surface after removing the ov erlying litter layer. EV soil was transported in a cooler, inside a large, clear trash bag and processed in the same manner as Gainesville soils upon returning to the lab. MS soil was collected by a third party and the exact details of the sampling were not recorded MS soil sample was processed in the same manner as Gainesville and EV soils. During sieving, a large, clean rubber stopper was used to crush soil aggregates larger than 2 mm so that all mineral and organic matter except for rocks and plant matter larger than 2 mm were collected Sieved soils were then divided for use in molecular work and for determination of soil physical/chemical properties. Both sources were stored at 80 C, whenever available, or at 20 C when not in use to prevent ch anges in both soil physical/chemical properties and microbial community structure. Measuring Soil Chemical Properties Subsamples of soils were submitted to the Soil Analytical Services Lab, University of Florida, Gainesville, to be tested for the following : 1) percent organic matter (OM); 2) pH; 3) total Kjeldahl nitrogen content (TKN); 4) ammonia based nitrogen content (NH 3 N); 5) nitrate /nitrite based nitrogen content ( NO x N); 6) extractable iron content (Fe); and 7) extractable phosphorous content (P). EV and MS soils were not analyzed because too little sample remained to be tested for the above parameters. Approximately 30 g (wet weight) samples of a different sampling of MW (from that used for clone library production) and R14 soils (same as the clon e library) were sent to Test America (Tampa, Florida, USA) for determination of naphthalene and o phthalic acid concentrations by gas chromatography mass spectrometry (GCMS) utilizing SW846 method 8270C or 8270C LL (for low levels)
45 Statistical Analyses o f Data Soil DNA extraction, Touchdown PCR, and clone library development for all tested soils were performed in the same manner as outlined above for MW in preliminary primer testing phase. DOTUR (Schloss & Handelsman, 2005) was used to generate rarefacti on curves for all clone libraries and for determining our sampling limit as well as Shannon Weaver and Chao1 diversity indices Chao1 diversity is a measure of the richness of operational taxonomic units (OTUs) within the sample, but Shannon Weaver includ distribution among different OTUs. Distribution and diversity of ARD genes across all clone libraries was analyzed using the online UniFrac ( mailto:http://bmf2.colo rado.edu/unifrac/index.psp ) and Fast UniFrac ( http://bmf2.colorado.edu/fastunifrac/ ) Statistical Analysis Engine (Lozupone et al., 2007) and included PCoA and distance matrix calculations. Mantel anal ysis with an alpha set at 0.05, was performed for each of the physical/chemical properties mentioned above, independently, against the distance matrix generated in UniFrac and (R Core Team, 2012) utilizing the vegan package. Although phyl ogenetic data for MW, EV and MS soils allowed us to include these soils in our PCoA analysis, they had to be excluded from Mantel analyses because soil chemical properties could not be analyzed for EV and MS soils and data for MW soils was obtained for a different sample than that used for clone library development (making a comparison between the two inappropriate ) Mantel analysis utilizing naphthalene and phthalate concentration data was not performed for this portion of our research because only one o f our two tested soils, R14, had measureable naphthalene concentrations. A Mantel analysis cannot be performed for a single clone library because the Mantel analysis requires a bacterial distance matrix The distance matrix, itself requires at
46 least two libraries so that differences in sequence diversity/divergence can be measured. Comparison of narAa and nahAc in Select Soils In addition to examining the phylogenetic diversity and distribution of Actinobacterial ARDs, we were also interested in how th ese sequences compared to the diversity and distribution of Proteobacterial ARDs. Because the diversity and distribution of Pseudomonad like ARDs was not a fundamental component of our research, our work in this matter was not extensive and was among the last tasks that we performed before finalizing this research. We only performed some preliminary comparisons of our Actinobacterial based ARD data from several soils to data concerning a much more limited range of Pseudomonad ARDs. These comparisons revo lved mostly around the use of PCoA and Mantel analyses to determine whether or not the nahAc and narAa like ARD sequences from soils showed any similarity to each other in terms of the extent of their diversity and distribution as well as whether or not c ertain environmental factors, described above, correlated to these similarities. In order to do this, we first chose previously published degenerate PCR primers designed by Ferrero et al. (2002) to target the gene sequence encoding the large subunit of va rious Pseudomonad naphthalene dioxygenases, nahAc These primers CCC YGG CGA CTA TGT CTC RGG CAT GTC TTT TTC 3 A table defining degenerate codon notation is provided in Table 3 6 We utilized the PCR cyclin g conditions of Lloyd Jones et al. (1999) in order to amplify the required PCR product, which proceeds as follows: 1 cycle of 94 C, for 5 minutes; 40 cycles of 94 C for 2 minutes, 55 C for 1 minute, and 72 C for 1 minute; followed by a final elongation ste p at 72 C for 10 minutes. We then proceeded to gel
47 purify, clone, and sequence in the same manner as described above for our Actinobacterial work. The data that we obtained from these clone libraries was input into the UniFrac and Fast UniFrac online sta tistical analysis engine to determine vegan package was used to perform Mantel analysis using the data for soil physical/chemical properties listed above. All data generated was then finally compared against that generated for Actinobacterial based ARD data from the same soils and excluding those soils for which we did not produce nahAc clone libraries. Soils that were utilized in this work include Regal 14 Theater (R14) par king lot soil and Archer Rd. Scrubs Carwash (AS) drainage ditch soil (described previously) as well as May sampled (MBMB), and June BAA). An additional Mantel analyses was performed comparing Proteobacterial and Actinobacterial soil clone libraries against each other. Full descriptions of MBAA, MBMB, and JB AA soils are given in Chapter 4. Results Our work in the initial stages of pri mer development resulted in the design of two pairs of degenerate PCR primers. Our initial PCR primers, ActNDO TGY CCN TAY CAC GGN TG TCN RYR TCR TCY TGK TCV AA were unsuccessful in amplifying any PCR products from our soil DNA. After reducing the amount of degeneracy in the previous set of primers we obtained the primer pair ActNDO TGY CCS TAY CAC GGB TGG TCV RCR TCR TCY TGK TCV AA NAR sequenc e for
48 Rhodococcus sp. NCIMB 12038 (Larkin et al., 1999) shows that the primers theoretically bind so that the product encompasses the majority of the Rieske [2Fe 2S] domain protein and enzyme mononuclear iron active site. This region of DNA begins and end s at positions 264 and 1116, respectively in Rhodococcus sp. NCIMB12038 (see Figure 3.2 ). The product of the M213 colony PCR was a single band occurring at 800 bp, against standard DNA ladder, in accordance with our theoretical product size determination d iscussed above (refer to Figure 3.2 ). The products of the MW soil DNA PCR were two bands closely flanking either side of the 800 bp marker at what appears to be approximately 775 to 825 bp (see Figure 3.3 for direct comparison). BLASTn results for submit ted samples indicated that the PCR product obtained from R. opacus M213 was 99% identical to the naphthalene, or naphthalene inducible, dioxygenases of 7 Rhodococcus species (including R. opacus B4). Of the 48 MW soil narAa clones submitted for sequencing 31 showed sufficient similarity (>75%) with Actinobacterial aromatic dioxygenase sequences. Although none of the soil clones from the MW library wer e related to M213 or Rhodococcus naphthalene d ioxygenases, all of the returned sequences showed high simi larity to Actinobacterial ph thalate dioxygenases with most being associated with Mycobacteria ( Figure 3.4 ). Shannon Weaver and Chao1 values (for 5% cutoff) were 2.45 and 27.45, respectively. The combined ability of these primers to amplify gene sequences for PAH naphthalene (in M213) and MAH (monocyclic aromatic hydrocarbon) phthalate dioxygenases (in soil DNA) along with their exclusions of Gram negative bacteria and the Firmicutes was taken as justification for their use
49 investigating the diversity and distribution of multiple Actinobacterial aromatic ring dioxygenase (ARD) genes in soils. For all other soils tested, phylogenetic associations with known reference sequences are summarized in Figures 3 7 to 3 1 0 We submitted 48 clones from each soil to b e sequenced, and the number of returned sequences that showed significant similarity to published sequences on Gen B ank varied greatly with each library. The number of usable MW clone sequences was provided previously, but for the remaining soils the distr ibution of usable sequences is: EV (37/48); R14 (25/48); AS (26/48); and MS (4/48). Rarefaction curves were generated for all libraries at 3%, 5%, and 10% similarity cutoff where differences were seen For all libraries, results of rarefaction indicated that further sampling would not return a reasonable number of unique sequences and that all plateaus were approximated. Shannon Weaver/Chao1 diversity values for EV, R14, AS, and MS, respectively, are as follows: 2.54/25.23; 2.08/14.00; 1.28/13.92; and 0 .35/3.03. Shannon Weaver and Chao1 d iversity indices are summarized in Table 3 2. For EV and MS soils, initial PCR resulted in products that showed similar sizes consistent with the results observed in initial work with MW at approximately 775 and 825 b p. Randomly selected clones showed inserts that approximated the size of either the lower or the higher of the two bands, but never both in a single clone, just as in MW. The appearance of a single insert per clone was consistent throughout our research and we never saw multiple bands in individual clones. In contrast to MW, EV, and MS soils, R14 and AS PCR products and clones showed a greatly reduced range in size, grouping more closely in size to the M213 positive control in initial PCR with clones
50 app earing to be of the same size as M213 control The results of the PCoA analysis, for all soils, are presented in Figure 3 6 Although excluded from soil physical/chemical analyses, EV and MS soils have been included here for PCoA. We can clearly see fr om Figure 3 6 that there is separati on of the less impacted clone libraries of MW and EV from both each other and the more impacted clone libraries of R14, AS, and MS. Both MW and EV clone libraries contain sequences that cluster exclusively with Actinob acterial phthalate dioxygenases For MW, the majority of these sequences are most closely associated with the Mycobacteria The clone sequences of EV soil are unique among all the soils analyzed in one regard; even though all of its sequences cluster exc lusively with the Actinobacterial phthalate dioxygenases, the majority of those sequences cluster outside the GenBank reference genes with very few being included under any one reference. Because these sequences still show a bootstrap value of 100 with th e remaining clones and reference genes they are obviously phylogenetically related, but because they are grouped outside the references they may represent an unincluded genus of Actinobacteria or other genetically distinct species not present in the tree. Unlike the clone libraries for EV and MW, R14, AS, and MS soil clone libraries include additional, non phthalate dioxygenase like sequences, the majority of which cluster with the unspecified Mycobacterium ARDs with occasional sequences associated with My cobacterial naphthalene inducible/pyrene (Nid) dioxygenases. Overall, the Mycobacteria are well represented by the returned clone sequences from all soils with few, if any, representatives found among other Actinobacterial genera. The results of the soil chemical properties, for testable soils, are presented in Table 3 3 MW data was measured but is not reported here because the chemical data
51 comes from a separate soil sample than that used to generate the clone library. As a result of the data coming fr om two different samples, attempting to correlate the two would be erroneous. O f the two testable soils (R14 and AS), AS soil had a lower pH and higher %OM than R14 ( pH = 7.23; %OM = 5.06; and pH = 7.83; %OM = 3.96 respectively). P content was lower in R 14 (63.36 mg/Kg) than in AS (93.4 mg/Kg ) while Fe content was fairly c onsistent for both soils ( 173.8 and 167.2 mg/ K g for R14 and AS respectively ). TKN and NOx N contents were highest in R14 soil (14489 and 210 mg/K g) compared to AS soil (5162 and 131.9 mg/Kg) while NH 3 N content was similar in both soils (2.83 and 3.17 mg/kg for R14 and AS respectively ). Refer to Table 3 3 for more direct comparison. Of all the soil chemical parameters tested for, no Mantel values could be calculated for the distance matrix containing only R14 and AS soil clone library sequences. Naphthalene concentrations, as determined by GCMS, were 480 g/Kg of soil for R14 and below the limit of detection in MW. O phthalic acid concentrations, in both R14 and MW, were below the l imit of detection. This limited amount of data was insufficient to perform a Mantel analysis to determine whether there was a significant impact of naphthalene on the diversity and distribution of both Pseudomonad and Actinobacterial ARD like sequences in our soils for this portion of our research. However, naphthalene concentration data for R14 soil was included along with naphthalene concentration data for soils to be discussed in Chapter 5, of this thesis, to perform a Mantel analyses solely for the Ac tinobacterial ARD like clone libraries. For PCR with the Pseudomonad nahAc primers of Ferrero et al. (2002), soil DNA from the soils previously listed produced several PCR products of various sizes. The
52 theoretical size of the target product is approximat ely 866 bp and a band approximating this size was present in R14, AS, MBAA, MBMB, and JBAA soils. After gel excision and purification of the target band, clone libraries were developed for each soil as they were for our work with Actinobacterial primers. Initially we submitted 48 clones from each library to be sequenced (R14 and JBAA soils) but rarefaction analysis for these libraries indicated a smaller sampling size was sufficient and only 24 clones were submitted for MBAA, MBMB, and JBAA soils. A samp le phylogenetic tree, for R14 soil nahAc like sequences, is provided in Figure 3 1 1 Here, all 46/46 usable clone sequences (that showed significant similarity to known Proteobacterial nahAc like sequences on GenBank) cluster exclusively with the naphthal ene 1,2 dioxygenases of Pseudomonas, Comamonas, Burkholderia, and Ralstonia nahAc and nagAc sequences. The closest, most significant association was with nagAc from Burkholderia sp. The phylogenetic distribution of clone sequences for the remaining 4 lib raries showed a similar distribution of sequences although a greater percentage of the total number of nahAc like sequences in the MBAA, MBMB, and JBAA libraries was distributed among Burkholderia Comamonas and Ralstonia sp. Overall, our nahAc like sequ ence diversity and distribution is in good agreement with results published by Ni Chadhain et al. (2006) who performed a more thorough study of Pseudomonad Rieske type aromatic degradation genes. A side by side comparison of PCoA and Mantel analyses for n ahAc and narAa like sequences for R14, AS, MBAA, MBMB, and JBAA soil clone libraries is presented in Figure 3 1 2 and Table 3 4 respectively. A direct visual compar ison of the phylogenetic trees and PCoA plots, coupled with Mantel analysis for soil prop erties an d clone libraries for nahAc and ARD like
53 sequences, reveals that the diversity and distribution of Proteobacterial nahAc like sequences shifts independently from the Actinobacterial ARDs Mantel analys es shows that both sets of clone libraries a re only associated with a P value of 0. 901 (Table 3 5 ) which is insignificant with our alpha value of 0.05. Additionally, none of the soil chemical properties analyzed appears to be significantly correlated to the differences seen in the nahAc like seque nces between clone libraries, although soil pH and percent organic matter content are significant factors affecting the differences seen in the Actinobacterial ARD like seque nce libraries (Table 3 4 ) In the Actinobacterial ARD libraries, organic matter c ontent is the most significant at a P value of 0.009. Additional factors that may be suggestive but not significant, are soil iron content, total Kjeldahl nitrogen, and nitrite/nitrates content with P values of 0.06, 0.071, and 0.085 respectively. D iscussion This research was initiated with the intention of designing a method for estimating the diversity and distribution of Actinobacterial aromatic ring dioxygenases in soils. Since our concern is primarily with PAHs, the PCR primers were based on th e Actinobacterial narAa gene, responsible for encoding the large subunit of a dioxygenase targeting naphthalene. Although no M213 like or other Actinobacterial narAa like sequences were detected in any of the target soils we did amplify the narAa gene in M213 itself as well as sequences similar to the Rieske type Nid (naphthalene inducible) genes of select non Rhodococcus Actinobacteria. This suggests three potential implications: 1) naphthalene is not present in biologically relevant concentrations; 2) t he Actinobacteria are not the predominant PAH degraders in these soils; or 3) naphthalene
54 degradation is dominated by non Rhodococcus Actinobacteria utilizing other Rieske type dioxygenases other than the M213 or narAa type The first two cases are diffi cult to either prove or disprove in light of the following facts: w e have limited, and questionable, PAH concentration dat a for MW, R14, MBAA, and JB AA soils but not for EV, MS, AS or MBMB soils so we canno t argue the first point; our data concerning the presence and distribution of ARD sequences related to other clades of bacter ia (namely the Pseudomonads) is much more limited than for the Actinobacteria which makes it difficult to argue the second point. De spite the limited data on Pseudomonad nahAc lik e sequences in our soils, we were still able to detect the presence of these sequence s in the select soils tested including the urban, anthropogenically impacted R14 and AS soils as well as the more isolated, forested soils MBAA, MBMB, and JBAA (see Chapt er 4 for more data and descriptions for these three soils). N o M213 or Rhodococcus like NDO like sequences were detected in any of these soils where nahAc like sequences were found Coupled with the previous data, t h is strongly suggests that nahAc like gene diversity changes independently of the shifts that are seen in the Actin obacteria and that the presence of naphthalene degrading Proteobacteria is higher than those for Actinobacteria in our sampled soils under in situ conditions. However, the data w e have does not provide us with any definitive information that might allow us to state which group of bacteria is the more contributory to PAH degradation. The third point made above, is at least plausible given that a significant relative percentage of the returned phthalate and non phthalate sequences were related to non Rhodococcus Actinobacteria.
55 More extensive and inclusive PAH concentration data would allow us to determine whether or not naphthalene, our representative PAH, is a constituent compoun d of our soils and would allow us to validate or discredit the first assumption above. More extensive s oil clone libraries of Pseudomonad like ARDs, in combination with soil PAH data, would allow for a more quantitative comparison between the two groups o f bacteria and offer more insight into which group likely predominates under low impact versus high impact conditions. Further molecular and chemical analyses are required to resolve the question with any reasonable degree of certainty. What we have sho wn is that the degenerate PCR primers that were designed for this research are useful in detecting the presence of multiple Rieske type Actinobacterial ARD sequences as well as showing the distribution of these sequences across several geographically and c hemically diverse soils. Since our primers were designed around the sequences highlighted in red in Figure 3.1 (which are all Rieske type enzymes), our results are in good agreement with previously published results dealing with the phylogeny and substrat e specificity of this class of enzymes (Kim et al., 2008; Perez Pantoja et al., 2009; Kweon et al., 2010). Our appli cation of these primers to the soils described, and the resulting clone library data have provided us a useful means for examining the pr esence of sequences similar to genes known to be important in the microbially mediated degradation of contaminating PAHs of anthropogenic origin. It should be noted, however, that we cannot state with absolute certainty that any given sequence that we hav e determined in this research does, in fact, code for the enzyme that it has been shown to cluster with. Nor can we say with certainty that these genes, if present, are being actively
56 expressed. Therefore, it may be best to limit our conclusions of this work by saying that the PCR system that we have developed may prove to be a useful tool in determining Actinobacterial degradation genotypes in a given soil by providing insight into the diversity and distribution of highly similar DNA sequences. Of th e soil chemical properties that we have currently analyzed, we could not calculate any values to determine significance in explaining the distribution of ARD sequences across the submitted clone libraries for R14 and AS However, the results change when w e include sequence data from soils MBAA, JBAA, and MBMB from Chapter 4 (refer to Chapter for further discussion) For Actinobacterial ARD like sequences from R14, AS, MBAA, JBAA, and MBMB, percent organic matter is implicated as the principal driver of di versity. For the corresponding Proteobacterial nahAc like sequences for the same soils the results were insignificant, as they were when considering the ARD libraries for R 14 and AS alone. When Proteobacterial clone libraries were compared against Actino bacterial libraries, for select soils we saw that there was no correlation between shifts in one set of libraries and shifts in the other. They are clearly independently controlled. Since the primers targeting nahAc like sequences are less degenerate th an those targeting Actinobacterial ARD like sequences, it is to be expected that differences in nahAc like clone libraries will be correspondingly smaller than those found in ARD like libraries. W e will see again, in Chapter 4 that four of the environm ental variables measured (OM %, pH, TKN, and NOx N) are significantly correlated to differences in Actinobacterial ARD like sequen ce diversity and distribution between the more directly impacted city soils and forested soils that are not directly impacted. Further testing for
57 naphthalene and o ther organic chemicals content might prove useful in determining whether there are significant differences in these properties across our soils to further explain the variation in ARD gene distribution that we have se en. Based on previously published materials, it seems reasonable to expect that an analysis of ARD genes in soils would return measurably different results for city soils with more direct exposure to fossil fuel based PAHs than for a forested soil with le ss direct exposure to similar PAHs. It is also reasonable to expect that a highly trafficked urban soil is more likely to contain genes for PAH degradation than forested soil with little or no PAH exposure. The results of our study confirm this, in part, by showing that there is greater diversity and distribution of ARD genes in urban soil s with direct exposure to PAHs than in soil s with lower expected exposures. Additionally, the limited work that we have done concerning the diversity and distribution of Proteobacterial ARD like sequences in soils (compared to Actinobacterial ARDs) has produced preliminary data that indicate that Proteobacterial ARD like sequences become less diverse and more homogenous under conditions of increased nutrient inputs and PA H exposure while they become more diverse and more highly distributed among different genera in soils with comparatively lower levels of nutrients and less exposure to PAHs. The converse appears to be true for the Actinobacterial ARD like sequences in the same soils.
58 Table 3 1. List of microorganisms, accession numbers, and genes used in the development of degenerate primers. Accession numbers are those of for nucleotide sequences from Genbank (presented in the same order given by the phylogenetic tre e in Figure 3.1). Accession Number Microorganism Gene and/or protein DQ157863 Mycobacterium sp. CH 2 nidA / naphthalene inducible dioxygenase (large, alpha subunit) AY330098 Mycobacterium sp. JLS nidA / naphthalene inducible dioxygenase (large, alpha sub unit) AY330102 Mycobacterium sp. MCS nidA / naphthalene inducible dioxygenase (large, alpha subunit) GU586859 Mycobacterium tuberculosis SE12 Unspecified dioxygenase, large subunit FJ032197 Pseudoxanthomonas sp. RN402 nidA/ naphthalene inducible dioxygen ase (large, alpha subunit) FJ032196 Diaphorobacter sp. KOTLB clone 1 nidA / naphthalene inducible dioxygenase (large, alpha subunit) AF548343 Mycobacterium flavescens strain PYR GCK nidA / naphthalene inducible dioxygenase (large, alpha subunit) AF548345 Mycobacterium frederiksbergense strain FAn9T nidA / naphthalene inducible dioxygenase (large, alpha subunit) DQ537941 Mycobacterium pallens strain czh 8 nidA / naphthalene inducible dioxygenase (large, alpha subunit) AB179737 Mycobacterium sp. MHP 1 nidA / naphthalene inducible dioxygenase (large, alpha subunit) HM049718 Mycobacterium sp. py136 nidA / naphthalene inducible dioxygenase (large, alpha subunit)
59 Table 3 1. Continued Accession Number Microorganism Gene and/or protein AF249301 Mycobacterium sp. PYR 1 nidA / naphthalene inducible dioxygenase (large, alpha subunit) AY330100 Mycobacterium sp. KMS nidA / naphthalene inducible dioxygenase (large, alpha subunit) AF546904 Mycobacterium sp. S65 nidA/ putative initial ring hydroxylating dioxygenase large subunit HM049713 Bacterium py114 nidA / naphthalene inducible dioxygenase (large, alpha subunit) AB626849 Mycobacterium sp. NJS P pdoA / ring hydroxylating dioxygenase, large subunit AB031319 Nocardioides sp. KP7 phdA / iron sulfur protein, large subunit AB017794 Nocardioides sp. KP7 phdA / iron sulfur protein, large subunit DQ028634 Mycobacterium vanbaalenii PYR 1 nidA3 / PAH ring hydroxylating dioxygenase, large subunit DQ118530 Terrabacter sp. HH4 pdoA / large subunit of PAH dioxygenase; ISP alpha CP000 511 Mycobacterium vanbaalenii PYR 1 benzoate 1,2 dioxygenase, alpha subunit AJ494743 Mycobacterium sp. 6PY1 pdoA2 / PAH ring hydroxylating dioxygenase, large subunit 2 CP002385 Mycobacterium sp. Spyr1 ring hydroxylating dioxygenase, large subunit CP00051 8 Mycobacterium sp. KMS ring hydroxylating dioxygenase, large subunit CP000580 Mycobacterium sp. JLS ring hydroxylating dioxygenase, large subunit
60 Table 3 1. Continued Accession Number Microorganism Gene and/or protein EF026099 Mycobacterium sp. SNP 11 phdA / putative ring hydroxylating dioxygenase large subunit CP000384 Mycobacterium sp. MCS ring hydroxylating dioxygenase, large subunit DQ358754 Mycobacterium sp. CH 1 pdoA 2/ putative PAH ring hydroxylating dioxygenase large subunit 2 CP002380 Arthr obacter phenanthrenivorans Sphe3 plasmid pASPHE301 ring hydroxylating dioxygenase, large terminal subunit AF082663 Rhodococcus sp. NCIMB12038 narAa / naphthalene dioxygenase large subunit AB024936 Rhodococcus sp. CIR2 rnoA 3/ iron sulfur protein AJ401612 Rhodococcus sp. 1BN narA / putative cis naphthalene 1,2 dioxygenase, large and small subunits DQ846881 Rhodococcus opacus narAa / naphthalene dioxygenase large subunit AB206671 Rhodococcus opacus nidA / naphthalene inducible dioxygenase (large, alpha subuni t) AP011117 Rhodococcus opacus B4 plasmid pROB02 nidA / naphthalene inducible dioxygenase (large, alpha subunit) GQ503240 Rhodococcus sp. B13 narAa / naphthalene dioxygenase large subunit AB110633 Rhodococcus opacus plasmid pWK301 orf6 dodA / terminal diox ygenase, large subunit AF121905 Rhodococcus sp. I24 nidA / naphthalene inducible dioxygenase (large, alpha subunit) GQ503239 Rhodococcus sp. DB11 narAa / naphthalene dioxygenase large subunit
61 Table 3 1. Continued Accession Number Microorganism Gene and/ or protein GQ503237 Rhodococcus sp. SMB38 narAa / naphthalene dioxygenase large subunit GQ848233 Gordonia sp. CC NAPH129 6 narAa / putative naphthalene dioxygenase large subunit AY392424 Rhodococcus sp. P200 narAa / naphthalene dioxygenase large subunit G Q503241 Rhodococcus sp. B2 1 narAa / naphthalene dioxygenase large subunit AY392423 Rhodococcus sp. P400 narAa / naphthalene dioxygenase large subunit ACWD01000051 Bacillus sp. 2_A_57_CT2 cont1.51 hypothetical protein HMPREF1013_03489 AB113649 Bacillus sp JF8 bphA 1/ large subunit of biphenyl dioxygenase CP001878 Bacillus pseudofirmus OF4 PAH dioxygenase large subunit CP002293 Geobacillus sp. Y4.1MC1 Aromatic ring hydroxylating dioxygenase, alpha subunit like protein NZ_ABIN01000073 Mycobacterium intrac ellulare ATCC 13950 Phenylpropionate dioxygenase and related ring hydroxylating dioxygenases, large terminal subunit NZ_ACBV01000012 Mycobacterium kansasii ATCC 12478 phthalate 3,4 dioxygenase alpha subunit AB084235 Terrabacter sp. DBF63 phtA 1/ oxygenase large subunit of phthalate dioxygenase NC_014814 Mycobacterium sp. Spyr1 chromosome phthalate 3,4 dioxygenase alpha subunit FJ641978 Mycobacterium sp. CH1 transposase gene phtAa / phthalate dioxygenase large subunit
62 Table 3 1. Continued Accession Numb er Microorganism Gene and/or protein AY372762 Mycobacterium flavescens PYR GCK phtAa / phthalate dioxygenase large subunit AY365117 Mycobacterium vanbaalenii strain PYR 1 phtAa / phthalate dioxygenase large subunit AB048709 Rhodococcus sp. RHA1 tic ring hydroxylation dioxygenase E DQ007994 Rhodococcus sp. TFB phtAa / phthalate dioxygenase large subunit DQ119329 Rhodococcus coprophilus strain G9 phtAa / phthalate dioxygenase large subunit FJ528993 Rhodococcus sp. JDC 11 phtAa / phthalate dioxygen ase large subunit CP000456 Arthrobacter sp. FB24 plasmid 2 phthalate 3,4 dioxygenase, alpha subunit FJ528990 Arthrobacter sp. JDC 12 phtAa / phthalate dioxygenase large subunit FJ528989 Arthrobacter sp. JDC 8 phtAa / phthalate dioxygenase large subunit F J528991 Arthrobacter sp. JDC 9 phtAa / phthalate dioxygenase large subunit AF331043 Arthrobacter keyseri plasmid pRE1 phtAa / phthalate dioxygenase large subunit FJ528988 Arthrobacter sp. JDC 1 phtAa / phthalate dioxygenase large subunit AF004284 Pseudomon as putida ndoC 2/ naphthalene dioxygenase iron sulfur protein, large subunit NZ_ACNO01000069 Rhodococcus erythropolis SK121 contig00088 Salicylate hydroxylase CP002017 Bacillus tusciae DSM 2912 Aromatic ring hydroxylating dioxygenase, large subunit CP001 276 Thermomicrobium roseum DSM 5159 plasmid 3 phenylpropionate dioxygenase alpha subunit
63 Table 3 1. Continued Accession Number Microorganism Gene and/or protein CP000656 Mycobacterium gilvum PYR GCK ring hydroxylating dioxygenase, alpha subunit CP0019 62 Thermus scotoductus SA 01 biphenyl 2,3 dioxygenase, alpha subunit CP001743 Meiothermus ruber DSM 1279 Aromatic ring hydroxylating dioxygenase, alpha subunit NZ_ADUJ01000002 Pseudonocardia sp. P1 PP100002 ring hydroxylating dioxygenase, alpha subunit NC_008269 Rhodococcus jostii RHA1 plasmid pRHL1 padAa 1/ phthalate 3,4 dioxygenase, alpha subunit AY502076 Rhodococcus sp. DK17 plasmid pDK3 putative phthalate dioxygenase NC _008268 Rhodococcus jostii RHA1 chromosome hmgA / homogentisate 1,2 dioxygenase NC_012522 Rhodococcus opacus B4 Unspecified dioxygenase NC_013159 Saccharomonospora viridis DSM 43017 chromosome homogentisate 1,2 dioxygenase NC_008711 Arthrobacter aurescens TC1 gtdA / gentisate 1,2 dioxygenase NC_003450 Corynebacterium glutamicum ATC C 13032 gentisate 1,2 dioxygenase NZ_GG657746A Streptomyces sp. AA4 Gentisate 1,2 dioxygenase NC_014830 Intrasporangium calvum DSM 43043 chromosome homogentisate 1,2 dioxygenase NZ_ACLY01000050 Bacillus cereus R309803 contig00054 Gentisate 1,2 dioxygena se DQ267826 Nocardia sp. C 14 1 catechol 1,2 dioxygenase NZ_AEKG01000133 Dietzia cinnamea P4 contig00133 catechol 1,2 dioxygenase NZ_ADUJ01000082 Pseudonocardia sp. P1 PP100085 catechol 1,2 dioxygenase NC_013441 Gordonia bronchialis DSM 43247 chromosom e catechol 2,3 dioxygenase
64 Table 3 1. Continued Accession Number Microorganism Gene and/or protein NZ_ACNO01000001 Rhodococcus erythropolis SK121 contig00034 catA / catechol 1,2 dioxygenase NZ_ACLI01000117 Corynebacterium efficiens YS 314 contig00133 pr obable salicylate monooxygenase NZ_ABYA01000149 Streptomyces ghanaensis ATCC 14672 salicylate monooxygenase NC_008268 Rhodococcus jostii RHA1 chromosome gentisate 1,2 dioxygenase NC_012522 Rhodococcus opacus B4 Putative oxidoreductase FM202432 Rhodococ cus sp. PY11 hpoO / putative salicylate 1 monooxygenase DQ846882 Rhodococcus opacus genH / gentisate 1,2 dioxygenase AB186916 Rhodococcus opacus rnoH rnoH / gentisate dioxygenase HM852512 Rhodococcus sp. NCIMB 12038 narI / gentisate 1,2 dioxygenase NZ_ACLI 01000117 Corynebacterium efficiens YS 314 contig00133 gtdA / gentisate 1,2 dioxygenase AB112586 Streptomyces sp. WA46 sdgD / gentisate 1,2 dioxygenase NZ_GG753629 Streptomyces sp. e14 gentisate 1,2 dioxygenase NZ_GG657758 Streptomyces griseoflavus Tu4000 gentisate 1,2 dioxygenase NC_011983 Agrobacterium radiobacter K84 gentisate 1,2 dioxygenase NC_009338 Mycobacterium gilvum PYR GCK chromosome catechol 1,2 dioxygenase NZ_GG657746 Streptomyces sp. AA4 gentisate 1,2 dioxygenase AB109791 Arthrobacter sp. BA 5 17 catA II / catechol 1,2 dioxygenase AY613438 Nocardia sp. H17 1 catA / catechol 1,2 dioxygenase
65 Table 3 1. Continued Accession Number Microorganism Gene and/or protein NC_012522 Rhodococcus opacus B4 catA / catechol 1,2 dioxygenase NC_008268 Rhod ococcus jostii RHA1 chromosome catA 1/ catechol 1,2 dioxygenase AF043741 Rhodococcus rhodochrous catA / catechol 1,2 dioxygenase AB167712 Rhodococcus sp. AN 22 catA 2/ catechol 1,2 dioxygenase EU004427 Rhodococcus gordoniae isolate AK38 catA 1/ catechol 1,2 dioxygenase
66 Figure 3 1. Neighbor joining tree based on aligned nucleotide sequences for PAH degrading bacteria. Sequences in red were used to design degenerate primers targeting Actinobacterial NDOs.
67 Figure 3 2. Binding sites for A ctNDO F2 and ActNDO R2 primers against Rhodococcus sp. NCIMB 12038
68 Figure 3 3. Comparison of PCR products derived from MW soil DNA and M213 colony PCR.
69 Figure 3 4. Phylogenetic associations of R. opacus M213 PCR product and MW soil cloning resul ts with known ARD sequences.
70 Figure 3 5. Geographical map of the continental United States showing the relative locations of sampling sites where soils were obtained for this portion of the research
71 Table 3 2. Shannon Weaver and Chao1 diversity val ues for experimental soils (5% cutoff value). SOIL No. of Clones Shannon Weaver Chao1 MW 31 2.45 27.45 EV 37 2.54 25.23 R14 25 2.08 14.00 AS 27 1.28 13.92 MS 4 0.35 3.03 Table 3 3 Soil chemical properties for Gainesville, Florida soils. SOIL % O M pH TKN (mg/kg) NH3 N (mg/kg) NOx N (mg/kg) Fe (mg/kg) P (mg/kg) R14 3.96 7.83 14489 2.83 210 173.8 63.36 AS 5.06 7.23 5162 3.17 131.9 167.2 93.4 Figure 3 6 PCoA analysis for experimental soils Generated by UniFra c.
72 Figure 3 7 Phylogenetic t ree for c lone library of R14 soil ARD sequences
73 Figure 3 8 Phylogenetic tree for clo ne library of AS soil ARD sequences.
74 Figure 3 9 Phylogenetic tree for clone library of EV soil ARD sequences.
75 Figure 3 1 0 Phylogenetic tree for clone librar y of MS soil ARD sequences.
76 Figure 3 1 1 Phylogenetic tree showing the associations of R14 nahAc like clone sequences with known Proteobacterial aromatic degradation genes from GenBank.
77 Figure 3 1 2 Direct comparison of PCoA of clone libraries fo r R14, AS, MBAA, MBMB, and JBAA nahAc and narAa like ARD sequences.
78 Table 3 4 Direct comparison of results of Mantel analyses for nahAc and narAa like ARD clone sequences from R14, AS, MBAA, MBMB, and JBAA soils. pH OM% P mg/Kg Fe mg/Kg TKN mg/ Kg NH 4 mg/Kg NO x mg/Kg nahAc R Value 0.2228 0.3973 0.1946 0.01348 0.1864 0.08115 0.05565 Significance 0.7 0.851 0.617 0.47 0.419 0.369 0.573 narAa R Value 0.7843 0.9611 0.3419 0.7647 0.5459 0.2658 0.8168 Significance 0.04 4 0.009 0.989 0.06 0.071 0.76 0.085
79 Table 3 5 Mantel results for soil nahAc like and Actinobacterial ARD like sequence clone libraries nahAc vs. AARD R Value Significance MANTEL 0.4028 0.901 Table 3 6 Degenerate codon abbreviations a nd their meanings. Degenerate Codon Meaning Base Usage W Weak A, T S Strong C, G M Amino A, C K Keto G, T R Purine A, G Y Pyrimidine C, T B C, G, T D A, G, T H A, C, T V A, C, G N Any Base A, C, G, T
80 CHAPTER 4 EFFECT OF BURN TREATMENT ON ARD GENE DISTRIBUTION IN SOILS Approach/Hypothesis In C hapter 1, we introduced the concept that the incomplete combustion of organic m atter (such as in forest fires) has been implicated in the production and input of PAHs into soil. This input of PAHs can also have an effect on the soil microbial community composition. In C hapter 2 we reviewed two papers dealing specifically with how the production and distribution of PAHs in soils is affected by forest fire and how different types of biochar (incompletely combusted organic matter) affected soil microbial communities. A significant amount of past and current research being conducted o n the subject of PAH contamination of soils has been focused on the effects of fossil fuel based sources of contamination on soil and environmental quality. Studies examining the impacts of forest fire based PAH contamination are much fewer in number and more recent in the literature. In Khodadad et al. (2010) we saw that the incomplete combustion of woody plant matter at high temperatures (650 C, in this case) resulted in biochars that were composed of highly recalcitrant, aromatic carbon compounds. Whe n added to soils, these biochars significantly impacted soil microbial communities with a detectable shift towards Actinobacterial dominated communities. For our research, we too were also interested in looking at the possible impacts of burning on soil microbial community structure. For that purpose, we utilized the same techniques that were developed and utilized for the research discussed in chapter 3 in examining paired burned and unburned forest soil plots, here at the University of Florida. By per forming PCR analysis and clone library development on paired burned and unburned soil plots we believed that we could detect measureable differences in
81 the community structures of both types of soils over a short period of time. We hypothesized, based on the materials reviewed in chapter 2, that the soil microbial communities of burned soil plots would show an increase in both the diversity and distribution of Actinobacterial ARD like sequences, compared to unburned plots. We analyzed samples taken from 2 unburned control plots and 3 burned plots in May of 2012, three weeks after a prescribed burn treatment of the area, and in June of 2012, approximately one and a half months after burning. Materials and Methods Soil Descriptions Soils were collected from the Natural Area Teaching Laboratory (NATL) field site, at the University of Florida. A schematic representation of the relevant area and surrounding facilities is presented in Figures 4.1 and 4.2 Additional information concerning the site, its history, and management can be accessed publicly online at http://natl.ifas.ufl.edu/index.php On April 24, 2012 selected sites in the NATL were subjected to prescribed burn treatments that have occurred once ev ery three years since 1996. Those sites are highlighted in red, in Figure 4.2 Two sites utilized as paired unburned controls are located n orth and to the e ast of the burned plots have not received burning since 1996, and are highlighted in blue, in Fig ure 4.2 April 24, 2012 burning was conducted during a northern wind to minimize deposition of ash and smoke borne particulates to unburned plots adjacent to burned zones. As one of our control soils, we obtained samples from a small unburned plot of h ammock type forest (mixed pines and oak) situated immediately n orth of, and outside, NATL west. Geographically this site is s outh of the Florida Museum of Natural History and w est of the Performing Arts Center, on Hull Road, and e ast of the Doyle Conner
82 Building on 34 th Street. The underlying soil series is Arredondo and was designated UAC for 3. Our second control site was a n unburned plot situated immediately w est of the Old Field Succession plots of NATL West and east of soils receiving burn treatments. The v egetation of this site represents a transition from predominantly longleaf pine to mixed pine and oak (hammock). A previous soil survey of the area described t he u nderlying s oil series as beginning as Millhopper at the west most boundary of the plot and becomes Lochloosa as one moved East towards the Old Field Succession plots. This soil was designated UMC for Our first burn plo t wa s situated immediately south of UAC and w est of UMC. The v egetation is representative of an upland pine ecosystem and predominantly longleaf pine with a few interspersed oaks and an understory of mixed saw palmetto, thorny shrubs, and wire grasses. T he underlying soil series is Arredondo and the site was designated BAA for Our second burn soil e ncompasses a large area s outh of BAA with a small portion bo rdering BAA to the northwest. This site has the same vegetation as BAA and the underlying soil series is Millhopper. The site was designated BMA for the largest of the sampling plot s and was situated s outh of Site A (BAA and BMA) This site is separ ated by a firebreak (a man made path of physically disrupted soil a nd vegetation between two burned areas) about 2.0 3.0 meters wide Vegetation here is the same as BMA with additions of some invasive species not found in Site A. The underlying soil se ries is Millhopper and the site was designated BMB for
83 Both Arredondo and Millhopper soils are sandy, siliceous, hyperthermic, Grossarenic paleudults with no discernible taxonomic differences in the upper 10 15 cm of soil Replicate soil samples (n = 6) were collected from each of the five sites above (n = 30) on May 4, 2012 (1.5 weeks after burn treatment) and again on June 2, 2012 (total = 60 soil samples). Sampling was conducted in the same manner for soils used in ch apter 3. Briefly: each individual replicate sample, from each of the sites, was collected either with a hand shovel or punch coring device from the top 10 cm of the soil after clearing away the overlying litter layer. When a hand shovel was used, larger samples were obtained. When a cylindrical punch corer was used, each replicate sample was comprised of a composite of 2 3 cores taken closely together and collected in the same bag as 1 replicate. Replicate samples were brought back to the lab and dry si eved using a 2 mm steel mesh screen. Individual replicates were divided in two for use in soil physical/chemical analyses and molecular work, respectively. Both types of samples were stored at either 20 C or 80 C (when available) when not in use. Molec ular Analyses DNA extraction, processing and storage as well as narAa PCR and clone library development for each sampling site was conducted in the same manner as outline in chapter 3. Briefly: DNA from sieved replicate samples was extracted in duplicate or triplicate 0.25 0.3 g subsamples and composited, in 1:1 ratios, after extraction; an aliquot of DNA extract was diluted 1:10 for PCR and cloning, and the remaining unused extract was stored at 20 C for the duration of the research. DNA extraction, PC R, and cloning materials and methods used for this portion of our research were the same as used in Chapter 3.
84 Measuring Soil Chemical Properties Subsamples of soils were submitted to the Soil Analytical Services Lab, University of Florida, Gainesville, t o be tested for the following: 1) percent organic matter (OM); 2) pH; 3) total Kjeldahl nitrogen content (TKN); 4) ammonia nitrogen content (NH 3 N); 5) nitrates nitrogen content ( NO x N); 6) extractable iron content (Fe); and 7) extractable phosphorous cont ent (P). Approximately 30 g (wet weight) samples of MUMC, MBAA, and JBAA soils were sent to Test America (Tampa, Florida, USA) for determination of o phthalic acid concentrations in the same manner as the samples in Chapter 3 were. Statistical Analyses St atistical analyses were performed as in Chapter 3 excluding MW, EV, and MS soils for reasons previously described Briefly: rarefaction curves and Shannon Weaver and Chao1 diversity indices were generated for returned clone libraries using DOTUR to ensu re we approximated our sampling plateaus; PCoA and soil microbial distance matrices were generated using the UniFrac and Fast UniFrac online statistical engine; Mantel analyses were conducted for each of the soil chemical properties mentioned, individually and together against the distance matrix generated by UniFrac conducted for all NATL soils due to the large amount of sample from each site which allowed us to analyze all soils for the soil physical/chemical properties described above. Mantel analysis utilizing phthalate concentration data was not performed for reasons discussed below. SAS/STAT(R) and the GLM procedure were utilized to determine the significance of burn versus unburned in affecting soil clone library sequence distribution and diversity (SAS Institute, Cary NC).
85 Results Phylogenetic trees for individual NATL clone libraries against reference sequences are presented in Figures 4 3 to 4 12 Phylogenetic trees sh owing the distribution of all May and June sampled NATL soil clone libraries are presented in Figures 4 13 and 4 14, respectively. 48 individual clones were submitted for sequencing from each library. The number of returned sequences that showed a reaso nable degree of similarity with known sequences on GenBank varied widely between all libraries for soils sampled in May and June. There are a total of 161 clone sequences across all May sampled NATL soil clone libraries and a total of 175 clone sequences across the June sampled libraries. Both groups of libraries returned similar numbers of sequences per sample site with one exception, the MUMC/JUMC pair (MUAC/JUAC = 42/39; MUMC/JUMC = 26/36; MBAA/JBAA = 35/37; MBMA/JBMA = 36/37; MBMB/JBMB = 22/26). Tabl e 4 1 summarizes the relative distributions of sequences between phthalate and non phthalate related dioxygenases, for all NATL libraries. For MUAC library, 42/48 sequences showed reasonable similarity with Genbank sequences with 36 being related to Acti nobacterial phthalate dioxygenases and 6 being related to an unspecified Frankia aromatic dioxygenases and a Mycobacterium Rieske protein. Shannon Weaver and Chao1 values for this library were 2.37 and 18.81, respectively. For MUMC library, only 26/48 se quences showed similarity to known sequences on GenBank and they were all related to Actinobacterial phthalate dioxygenases. The diversity values, for Shannon Weaver and Chao1, for MUMC were 2.22 and 17.00, respectively. For MBAA library, 35/48 sequences showed similarity with published sequences on GenBank with 33 being related to Actinobacterial
86 phthalate dioxygenases and 2 being related to an unspecified Frankia aromatic dioxygenases and a Mycobacterium Rieske protein. Shannon Weaver and Chao1 values were 2.66 and 41.03. For MBMA library, 36/48 clone sequences showed similarity to sequences published on Genbank with 24 sequences being related to Actinobacterial phthalate dioxygenases and 12 being related to an unspecified Frankia aromatic dioxygenases and a Mycobacterium Rieske protein. Diversity values, for MBMA, were 2.80 and 38.19 for Shannon Weaver and Chao1, respectively. MBMB showed the least returned sequences for all soils sampled in May, with only 22/48 sequences showing reasonable similarit y to sequences published on Genbank. Out of these 22, 15 were related to Actinobacterial phthalate dioxygenases and 7 were related to an unspecified Frankia aromatic dioxygenases and a Mycobacterium Rieske protein. Diversity values were 2.25 and 21.66, f or Shannon Weaver and Chao1, respectively. For JUAC clone library, 39/48 clones showed sequence similarity with known sequences on Genbank with 36 being related to Actinobacterial phthalate dioxygenases and 3 being related to an unspecified Frankia aromati c dioxygenases and a Mycobacterium Rieske protein. Shannon Weaver and Chao1 diversity values were 2.44 and 30.68, respectively. JUMC library had 36/48 clone sequences that showed reasonable similarity to known sequences on Genbank with 34 being related t o Actinobacterial phthalate dioxygenases, 1 being related to unspecified Mycobacterium, Nocardioides, and Terrabacter aromatic ring dioxygenases, and 1 sequence being related to an unspecified Frankia aromatic dioxygenases and a Mycobacterium Rieske protei n. Diversity values for Shannon Weaver and Chao1 were 2.90 and 38.69, respectively. JBAA clone library had 37/48 clone sequences that were similar to
87 sequences on Genbank with 36 related to Actinobacterial phthalate dioxygenases and 1 being related to an unspecified Frankia aromatic dioxygenases and a Mycobacterium Rieske protein. Shannon Weaver and Chao1 diversity values were 2.44 and 30.68, respectively. 37/48 clone sequences for JBMA library showed similarity to Genbank sequences with 32 being relate d to Actinobacterial phthalate dioxygenases, 3 being related to an unspecified Frankia aromatic dioxygenases and a Mycobacterium Rieske protein, and 1 related to unspecified Mycobacterium, Nocardioides, and Terrabacter aromatic ring dioxygenases. Shannon Weaver and Chao1 diversity values were 2.66 and 30.12, respectively. Only 26/48 clone sequences from JBMB clone library showed reasonable similarity to sequences published on Genbank. 19 of these sequences were related to Actinobacterial phthalate dioxyg enases and 7 were related to an unspecified Frankia aromatic dioxygenases and a Mycobacterium Rieske protein. Diversity values for Shannon Weaver and Chao1 were 2.62 and 35.99, respectively. As in Chapter 3, the majority of the returned sequences for the clone libraries developed in this portion of the research were related to non Rhodococcus Actinobacteria with most being related to the Mycobacteria A summary of clone sequence numbers and Shannon Weaver and Chao1 diversity values is provided in Table 4 1. For site UAC, there was an increase in the percentage of sequences related to phthalate dioxygenases relative to those not related to phthalate. This indicated a decrease in the distribution of ARD sequences in site UAC, from May to June, while diver sity appeared unaffected as the same group of non phthalate dioxygenases is still represented in the tree. For site UMC, there was an increase in both sequence
88 distribution and diversity. From May to June, sequence distribution and diversity shifted from being 100% phthalate related to including sequences among two groups of non phthalate related sequences that are phylogenetically distinct. Site BAA was similar to UAC, with an increase in the number of sequences related to phthalate dioxygenases with a smaller number still present in the same group of non phthalate related sequences. Site BMA is curious in that the relative distribution of sequences decreased but diversity increased. From May to June, more sequences appear ed that are related to phthala te dioxygenases, but the remaining non phthalate sequences are divided between two phylogenetically distinct groups instead of appearing in only one group as they did previously. Finally, site BMB shows a decrease in distribution with an increase in the r elative percentage of phthalate dioxygenase related sequences while still maintaining a small number of sequences in the same group of non phthalate dioxygenases. PCoA analysis comparing May and June sampled NATL soil clone libraries against Gainesville and Elk Valley (EV) soils (from Chapter 3), excluding Michigan soil (MS), is presented in Figure 4 1 5 PCoA compa ring all NATL soil clone libraries by themselves is presented in Figure 4 1 6 (excluding all other soils ). The PCoA analyses have be e n broken up in this manner in order to best view differences in sequence diversity and distribution within the NATL, independent of other soils, as well as the relationships between the NATL libraries and those of other regional soils. MS was excluded from conside ration in all cases because of the comparatively low number of returned sequences in the clone library (4) compared with the next lowest, MBMB (22).
89 Figure 4 1 5 clearly show s that the sequences present in May and June sampled NATL soil clone librar ies, when considered against other soils, are distinct and do not group with Gainesville soils or EV soil. Perhaps this reflects both the fact that they are geographically related soils from within a larger site and that they are not directly impacted by fossi l fuel based PAHs while the other soils are. The effects of burning cannot be clearly determined from such a densely compacted arrangement on the PCoA but differences between the two treatments do not appear to be dramatic Figure 4 1 6 considers only the NATL soil clone libraries and does not include any of the Gainesville soils, EV, or MS soil clone libraries. Here, May sa mpled NATL clone libraries appear to be more diverse than June sampled libraries, as can be seen by comparing the sp atial arrangement of the May and June libraries E ach of the May sampled libraries appears to form its own distinct group with little association with any other site with the possible exception of MUAC and MBAA which are closer together on the plot than any other May samp led libraries. The other three sites are situated well away from each other indicated highly significant differences in sequence distribution. The same is seen, to a lesser degree, for the June sampled NATL clone libraries. Unlike the May sampled librar ies, which are widely separated on the PCoA, the June sampled libraries, overall, are considerable closer together indicating greater homogeneity in sequence distribution among June sampled libraries. It should be noted that there is less distinction betw een burned and unburned sites sampled in June, which we will attempt to explain next. In order to more clearly analyze the effect of burning on the distribution of ARD sequences for both May and June sample NATL clone libraries, we compared the
90 relative p ercent contribution of phthalate dioxygenase like ARDs from each individual NATL clone library burned and unburned, utilizing SAS/STAT(R) and the GLM procedure. After analysis we could not determine any significant differences between the burned and unb urned treatments and determined that perscribed burning had no impact on clon e sequence distribution within the NATL Table 4 2 summarizes and compares t he results of the soil chemical properties measured in our soils (TKN, NOx N, NH 3 N, P, Fe, OM%, and p H) for both the Ma y and June samplings. A s eparate M antel analys is was performed for each individual property as well as a comprehensive Mantel analysis considering all properties together This was done for both the PCoA that compares the NATL soil libr aries to the Gainesville soil libraries as well as for the NATL clone libraries alone. Tables summarizing the Mantel r and signif icance values for soil chemical properties for each of the PCoA conducted are presented in Tables 4 3 and 4 4 MW and EV soil clone libraries were included in PCoA due to the large number of returned clone sequences (37) but excluded from Mantel analyses because soil chemical properties were not measured (EV) or from a different sampling (MW) All phthalate concentration data o btained for soils MUMC, MBAA, and JBAA were below the limits of detection and therefore we could not perform a Mantel analysis for this variable. Table 4 3 summarizes the Mantel r and significance values for analyses performed using the UniFrac distance matrix for clone libraries of May sampled NATL soils and soils R14 and AS (excluding MW, EV and MS from Mantel). Although all properties taken together were significant (R = 0.015), of the se ven chemical properties tested four were determined to be espec ially significantly correlated (alpha = 0.05) with
91 the distribution of ARD sequences among clone libraries. These properties (from most to least significant) are: organic matter, OM (P = 0.00 1 ); pH (0.004); total Kjeldahl nitrogen TK N (P = 0.0 06); and ni trates/nitrites nitrogen, NO x N (0.0 15 ). All other properties had values that were greater than 0.10 and not significant. Table 4 4 summarizes the Mantel r and significance values for analyses performed using the UniFrac distance matrix for NATL clone li braries only ( excluding all other soils ). Unlike the previous analyses, none of the seven soil chemical properties were significantly correlated to the differences in the distribution of ARD sequences among the different clone libraries within the NATL. A combination of the phylogenetic data and PCoA and Mantel analyses makes it easier for us to begin to make some assumptions about what soil properties are influencing the differences in the distribution and diversity of clone ARD sequences between NATL soi l and the more directly impacted Gainesville soils MW, R14, and AS (for both May and June libraries). However, these same analyses make it less clear on what is influencing those differences among the NATL soils, themselves. A plot of the Shannon Weaver diversity values for Actinobacterial ARD like sequences against OM% for all sites tested (including NATL, R14, and AS soils) is presented in Figure 4 17. In this figure, the percent OM in soils increases to the right of the x axis while Shannon Weaver div ersity values increase moving up on the y axis. By plotting the data in this manner it may be possible to make further inferences about the importance of organic matter to the diversity of Actinobacterial ARD like sequences in soils. It would appear tha t an initial increase in soil OM% increases the diversity of ARD like sequences followed by a subsequent decrease in sequence diversity with
92 further increases in OM%. The current data do not clearly illustrate at what OM% value diversity begins to decline and such a relationship can only be inferred at this point. The observed response regarding an increase in diversity with increasing organic matter concentrations followed by a decrease in diversity at higher organic matter concentrations may be related to shifting nutrient limitations (Carrick et al., 2008). As soil OM% increases, the diversity of the microbial community increases in response. As organic carbon limitation s are alleviated, the hosts of these genes become affected by limitations in another nutrient. This results in a decrease in diversity at high er organic matter contents. Discussion At the start of this chapter, we hypothesized that the distribution and diversity of ARD sequences among clone libraries generated from burned soil plots wou ld be significantly greater (more diverse; more highly distributed) than those of the unburned control plots. We performed several comparisons of the data from the May and June sampled NATL clone libraries against similar data obtained from the more impa cted Gainesville soils, but this was primarily done to further aid our interpretation of the NATL data, itself, which has been more unclear. Soils sampled in May, 1.5 weeks after burn treatment, should be theoretically more impacted than the soils sampled in June, one month later, though this short period of time between sampling may make such distinctions more difficult to make. However, if we maintain that soils sampled closer to the time of burning are more directly impacted than samples taken at a lat er date, then May soils should show the greatest diversity and distribution of ARD like sequences when compared to samples taken in June.
93 As previously stated, the number of usable clone sequences from each library is comparable for any site for both Ma y and June samplings (except the MUMC/JUMC pair) so a direct comparison of the distribution and diversity of sequences between May and June clone libraries, based on sampling site, can be reasonably made and have otherwise been normalized as percentage of overall sequences distributed between phthalate related sequences and non phthalate related sequences (and how many different groups) (Table 4 2). The highest percentage of clone sequences retrieved, for both May and June, are related to known phthalate d ioxygenases, the majority of which were non Rhodococcal In the one month between May and June, UMC site went from containing only phthalate dioxygenase related sequences to including additional ARD sequences associated with two reference groups distingui shable from each other and the phthalate dioxygenases. BMA site, likewise, went from having representatives in two reference groups (phthalate and one non phthalate) to including additional ARD sequences associated with a third non phthalate reference gro up that was distinct. The remaining three libraries (UAC, BAA, and BMB) showed only minor shifts in the percentage distribution of clone ARD sequences among reference groups already represented in May libraries. Although the total number of sequences rela ted to phthalate dioxygenases increased in samples obtained in June, a closer look at the individual phylogenetic trees indicates that there was a decrease in the diversity of those phthalate related sequences coupled with the emerging presence of non phth alate related dioxygenases not previously seen in the May samples. This may indicate an initial enrichment for phthalate degraders close to the initial time of burning. However, despite differences in
94 phylogeny that might be inferred from the trees, PCoA Mantel, and additional statistical treatment of the sequence data (based on phthalate dioxygenase like ARD composition in each treatment ) has indicated that there are no significant differences in the diversities and distributions of ARD like sequences b etwee n burned and unburned soils This lack of difference between the burned and unburned libraries is more apparent when all NATL libraries are considered against the libraries of the remaining regional soils where the NATL forms a distinct, closely rela ted collection of sequences. Our Mantel analyses performed for NATL only soil clone libraries found that none of the soil chemical properties tested could account for the differences between clone libraries for May and June samples within the NATL site itself. However, when we compare May and June sampled NATL libraries against libraries from selected soils fro m Chapter 3 for which we have soil chemical data (R14 and AS ), our Mantel analysis clearly indicates that there are four environmental variable s that are significantly correlated to the differences in Actinobacterial ARD like sequence diversity and distribution that we see between these soils and those of the NATL. These variables are percent organic matter (OM %), pH, nitrate/nitrite nitrogen c ontent (NOx N), and total Kjeldahl nitrogen (TKN), a measurement of organic nitrogen and ammonia content. There is good reason to associate an increase in overall soil organic matter content with an increase in diversity and distribution of ARD like gene sequences. These genes function in the sequential breakdown of PAHs and MAHs in the environment and these compounds are most definitely a type of organic matter. As the overall organic matter content of a soil increases it is reasonable to expect that t he diversity and relative
95 percentages of different types of organic molecules will also increase. Among these compounds may be aromatic substrates of varying bioavailability. That is another point to be made for associating an increase in organic matter content with an increase in ARD like sequence diversity and distribution; generally speaking, increasing the concentration of an organic substrate increases its bioavailability to the organisms that make up any particular microbial assemblage. And of cou rse, because TKN is a measurement of organic nitrogen content, and because pH is related to changes in the relative concentrations of H+ and OH in soils, an increase in soil organic matter content will directly affect these two measurements as well (both as a source of the organic nitrogen and as a source of H+ and OH from organic molecule functional groups). Additional data, including burn temperatures, and duration times, for each site at the time of treatment would be helpful in understanding the sign ificance of burning to our data (we know that different burn temperatures and durations as well as different vegetation types produce different types and amounts of PAHs; see Chapter 2).
96 Figure 4 1. Location of NATL West on University of Florida Campu s ( http://natl.ifas.ufl.edu/index.php )
97 Figure 4 2. Locations of Burned (Red) and Unburned Control (Blue) sampling sites in NATL West ( http://natl.ifa s.ufl.edu/index.php )
98 Figure 4 3. Phylogenetic tree for clone library of MUAC soil ARD sequences.
99 Figure 4 4. Phylogenetic tree for clone library of MUMC soil ARD sequences.
100 Figure 4 5. Phylogenetic tree for clone library of MBAA soil ARD seq uences.
101 Figure 4 6. Phylogenetic tree for clone library of MBMA soil ARD sequences.
102 Figure 4 7. Phylogenetic tree for clone library of MBMB soil ARD sequences.
103 Figure 4 8 Phylogenetic tree for clone library of JUAC soil ARD sequences
104 Figur e 4 9. Phylogenetic tree for clone library of JUMC soil ARD sequences.
105 Figure 4 10. Phylogenetic tree for clone library of JBAA soil ARD sequences.
106 Figure 4 11. Phylogenetic tree for clone library of JBMA soil ARD sequences.
107 Figu re 4 12. Phylog enetic tree for clone library of JBMB soil ARD sequences.
108 Figure 4 13. Phylogenetic tree for all May sampled NATL soil clone libraries against reference sequences.
109 Figure 4 14. Phylogenetic tree for all June sampled NATL soil clone libraries a gainst reference sequences.
110 Table 4 1. Summary of the changes in distribution and diversity of Actinobacterial ARD like sequences among NATL clone libraries from May to June, 2012. Values are normalized in terms of percent of total sequences present in clone libraries. SAMPLE SITE Location in NATL % Pht Sequences % Non Pht Sequences Non Pht Sequences >1 Group (Y/N) MUAC NATL North of Site A 85.7 14.3 NO MUMC NATL East of Site A 100 0.0 N/A MBAA NATL Site A Northmost 94.3 5.7 NO MBMA NATL Sit e A Southmost 66.67 33.33 NO MBMB NATL Site B South 68.18 31.82 NO JUAC NATL North of Site A 92.31 7.69 NO JUMC NATL East of Site A 94.44 5.56 YES JBAA NATL Site A Northmost 97.28 2.72 NO JBMA NATL Site A Southmost 89.19 10.81 YES JBMB NATL Site B South 73.08 26.92 NO
111 Figure 4 15. PCoA of May and June sampled NATL soil clone libraries against Gainesville and EV soil clone libraries. Figure 4 1 6 PCoA of all NATL soil clone libraries excluding additional regional soils
112 Tabl e 4 2 Shannon Weaver and Chao1 diversity values for experimental soils. SOIL No. of Clones Shannon Weaver Chao1 MUAC 42 2.37 18.81 MUMC 26 2.22 17.00 MBAA 36 2.66 41.03 MBMA 36 2.80 38.19 MBMB 22 2.25 21.66 JUAC 37 2.44 30.68 JUMC 36 2.90 38.69 J BAA 37 2.44 30.68 JBMA 37 2.66 30.12 JBMB 26 2.62 35.99 Table 4 3 Soil chemical properties for May and June sampled NATL west soils. SOIL % OM pH TKN (mg/kg) NH3 N (mg/kg) NOx N (mg/kg) Fe (mg/kg) P (mg/kg) MUAC 1. 5 5. 1 744.9 7. 0 0. 6 162.7 76. 1 MUMC 1. 6 5. 7 681.3 4. 5 1. 5 136.2 85. 2 MBAA 1. 8 5. 8 555.9 4. 0 1. 3 137.2 56. 2 MBMA 2 6. 1 814.8 4. 7 4.4 132.0 87. 3 MBMB 1. 5 6. 4 829.4 1. 1 10.2 114.5 87. 4 JUAC 1. 8 4. 8 800.0 11. 6 7. 1 188.0 91. 3 JUMC 2. 1 5. 4 973.0 1. 3 24 143.2 124.4 JBAA 1. 8 5. 8 577.4 1 7 11. 8 133.4 209.5 JBMA 2 5. 5 706.4 1. 5 10. 6 124.3 79. 3 JBMB 2 5. 6 577.6 1. 5 12. 7 110.7 70. 3
113 Table 4 4. R and significance values for Mantel analyses of soil chemical properties for all NATL and Gainesville soil clone libraries (excluding MW, E V, and MS soils). MANTEL All NH 4 NOx TKN OM pH Fe P R 0.6868 0.219 0.8622 0.7323 0.9168 0.8008 0.2301 0.1685 Sig. 0.015 0.987 0.015 0.006 0.00 1 0.004 0.11 0.766 Table 4 5 R and significance values for M antel analyses of soil chemical properties f or all NATL soil clone libraries without Gainesville soil clone libraries ( also excluding MW, EV and MS soils) MANTEL All NH 4 NOx TKN OM pH Fe P R 0.15 5 0.092 0.182 0.042 0.055 0.157 0.091 0.155 Sig. 0.808 0.583 0.172 0.575 0.588 0.789 0.301 0.73 1 Figure 4 17. Shannon Weaver diversity plotted as a function of soil OM%.
114 CHAPTER 5 ARD GENE DISTRIBTION SHIFTS IN SOILS INTENTIONALLY EXPOSED TO PAH VAPOR Approach/Hypothesis In Chapters 1 and 2 we introduced the co ncept that the contamination of soils with PAHs can occur as a result of indirect exposure to the fumes and exhaust put off by the combustion of fossil fuels such as gasoline as well as direct contamination with PAHs in liquid form In Chapter 3 we des cribed one of the soils utilized in this research, McCarty Woods (MW) soil, as being situated such that it is consistently exposed to automobile exhaust given off by daily traffic from a busy intersection on the University of Florida campus nearby Altho ugh we were unable to detect any ARD like sequences in MW soil related to the degradation hig her molecular weight aromatics ( such as naphthalene) we were still interested in how the ARD like sequence profiles of soils might shift when exposed to PAHs in va por form. We hypothesized that the Actinobacterial ARD like sequence profile of a previously unexposed soil would shift to more closely resemble the sequence profiles on soils known to be contaminated if they were exposed to PAH vapor In order to test t his hypothesis, we intentionally exposed samples of a soil not directly impacted by PAHs of anthropogenic origin to diesel fumes and then compared the resulting ARD sequence profiles of the exposed soil samples to those of unexposed control samples. Ma terials a nd Methods Soil Descriptions For this experiment we decided to return to the University of Florida Natural Area Teaching Laboratory (NATL) west site (see Figure 4 1) and resample from the burned
115 west ern portion of the site. A detailed description of the sampling site is provided in Chapter 4, where this soil featured in work looking at the effects of prescribed burning on the Actinobacterial ARD like sequence profiles of soils. Sampling occurred in November, 2012, approximately six and a half months after the prescribed burning of April 24 and much of the understory vegetation that had been completely cleared during the burn treatments had regrown, resulting in fairly dense underbrush Sampling was conducted in the same manner outlined in Chapter 4: 2 3 samples were obtained from each of six random locations spread out across the BAA site by using a steel punch coring tool. The se individual subsamples, 2 3 from each of the six locations sampled were homogenized by sieving through a 2 mm steel mesh followed by vigorous shaking by hand. The result was six bulk samples of soil which were then d ivided for use in both molecular work and soil incubations. Bulk soil samples for molecular work that w ere not being used were stored frozen at 20 C and at 80 C, when available. Leftover soil s from batches used for incubations were stored on a bench top in a sealed cooler to reduce temperature effects on soil microbial populations in the event that these soils would be used in a repeat experiment. Soil Incubations Working soils for incubations were prepared in the following manner: a large volume of working stock was prepared, from each of the six bulk homogenized samples, by adding equal, wet weight, am ounts of soil in a separate container in a 1:1:1:1:1:1 ratio followed by vigorous shaking by hand. After determining the percent moisture cont ent of this homogenized mixture, 1 00 grams (wet weight) was added to each of 6 clean non sterile glass mason jar s The mason jars were hand washed, in the lab, using antibacterial soap and deionized water and left to air dry, upside down. Sterile
116 deionized water was added directly to the soils in these jars to artificially increase the soil moisture content to app roximately 5%. We decided to raise the moisture content of the soils in our incubations to ensure that there was enough water available for adequate micro bial respiration and growth for the duration of the incubations. Three of the six mason jars were de signated as unexposed controls and were sealed and incubated without diesel fumes. The remaining three mason jars were designated as our exposed test soils and a clean, 50 mL glass beaker containing 20 mL of commercial boating diesel was pressed down into the center o f the soil, inside the jars, which were then sealed tightly. All six mason jars were incubated for 29 days, in the dark, in a closed bench drawer. During the incubation the jars were left undisturbed and the lids were not opened. After 29 d ays, all six mason jars were sampled for use in molecular analyses (see following section for details). Molecular Analysis Molecular analyses were performed in the same manner as they were for BAA samples discussed in Chapter 4 with some minor changes Briefly, subsamples from each of the original six homogenized bulk samples were stored frozen at 20 C, or 80 C, in the event that we decided to use them in DNA extraction and PCR analysis to determine the Actinobacterial ARD like sequence profiles of the November sampled BAA (NBAA) soil at time of sampling For incubated soils, subsamples from each of the three diesel exposed replicates and the three unexposed controls were used for DNA extraction. A single extraction was performed for each replicate, e xposed and unexposed, and then the resulting extracts were homogenized by treatment, so that a single source of soil DNA for both exposed and unexposed soils was produced. DNA extraction was performed using 0.250 0.350 grams of sample and with the Powe rSoil
117 DNA Isolation Kit from Mo Bio (Carlsbad, CA, USA) Touchdown PCR utilizing our degenerate prime rs, ActNDO F2 and ActNDO R2, PCR product purification, and clone library generation were performed in the same manner as outlined in Chapter s 3 and 4 (and utilizing the same kits and materials). 48 clones from each library were submitted to the Sanger Sequencing Lab at the Interdisciplinary Center for Biotechnology Research (ICBR), at UF, for sequence determination. Sequencing was done utilizing our own f orward primer, ActNDO F2. Measuring Soi l Chemical Properties The following soil chemical properties were not tested for in NBAA soils: TKN; NOx N; NH 3 N, OM%; pH; Fe; and P After our incubation study was completed, all three diesel exposed replicates were homogenized together and all three controls were homogenized as well. Subsamples from these homogenized mixtures were sent to Test America (Tampa, Florida, USA) in sealed glass containers, on ice, for measurement of naphthalene and phthalate concentratio ns. Statistical Analysis of Data Statistical analyses were performed similarly as in Chapter s 3 and 4 excluding MW, EV, and MS soils for reasons discussed previously PCoA and soil microbial distance matrices were generated using the UniFrac and Fast Un i Frac online statistical engine For these PCoA we included additional data from soils investigated in previous chapters, including the contaminated soils R14 and AS, the May and June sampled BAA clone libraries, and MW. Mantel analyses was conducted for naphthalene concentrations only for the soils that we have data for (diesel exposed and unexposed NBAA, R14 and MW) and against a separate distance matrix for only those soils, using
118 as in Chapter 3 and 4 M antel analysis utilizing phthalate concentration data was not performed for reasons discussed below. Results A single p hylogenet ic tree for unexposed control and diesel exposed NBAA soil clone sequences is presented in F igure 5 1 Of the 48 clone sequence s submitted for sequencing from the unexposed control library, 32 showed significant similarity with published gene sequences on GenBank. Out of these 32 sequences, 18 clustered with the Actinobacterial phthalate dioxygenases, 10 clustered with the unspec ified Frankia and Mycobacterium Rieske proteins, 2 clustered with M213 and the Actinobacterial naphthalene and naphthalene inducible (Nid) dioxygenases, and 1 sequences clustered with the unspecified Mycobacterium, Nocardioides and Terrabacter PAH ring di oxygenases. Of the 48 clones submitted for sequencing from the diesel exposed NBAA library, 41 showed significant similarity with published gene sequences on GenBank. Out of these 41 sequences, 27 clustered with the M213 and Actinobacterial naphtha lene/N id dioxygenases, 8 clustered with the Actinobacterial phthalate dioxygenases, 5 clustered with the unspecified Frankia and Mycobacterium Rieske proteins, and 1 sequence clustered with the Mycobacterium, Nocardioides, and Terrabacter PAH ring dioxygenases. Table 5 1 summarizes the relative percentage of clone sequences from both clone libraries that are associated with phthalate and with non phthalate dioxygenases, including the relative percentage of sequences that clusters with M213 and the Actinobacterial naphthalene/Nid dioxygenases. Of the 32 control sequences, 56.25% are related to known phthalate dioxygenases, 37.5% are related to non phthalate dioxygenases, and 6.25% of the total control sequences were associated with the
119 naphthalene/Nid dioxygenases Among the diesel exposed clone sequences, 80.49% were related to aromatic dioxygenases other than phthalate and 65.85% were associated with naphthalene/Nid dioxygenases. The results clearly indicate that soils that were exposed to diesel fumes for one month have a higher relative percentage of detectable ARD like sequences related to the degradation of naphthalene and similar PAHs (Nid often refers to pyrene degradation). The presence of naphthalene/Nid related sequences in the control incubations is i nteresting as we have never detected these sequences in the May and June sampled BAA soils. This may indicate that the mason jars in which the soils were incubated may not have been sealed tightly enough and that there was cross contamination with diesel fumes that occurred inside the bench drawer where the incubations were stored. PCoA analysis of unexposed control and diesel exposed NBAA soil clone libraries, along with MBAA, JBAA, R14, AS, and MW soil clone libraries is presented in Figure 5 2. Four d istinct group ings can be determined from Figure 5 2: 1) unexposed control NBAA clone sequences cluster closely with those of MBAA and JBAA; 2) MW soil forms its own distinct cluster; 3) R14 and AS soil clone libraries form a third cluster; 4) and the diese l exposed NBAA soil clone library forms its own distinct cluster well away from any of the other clusters. The results of the PCoA indicate that the Actinobacterial ARD like sequences present in the unexposed control NBAA clone library are comparable to t he libraries generated from May and June sampled BAA soils. The control NBAA library is slightly offset from the MBAA/JBAA cluster most likely due to the presence of a few sequences related to naphthalene/Nid dioxygenases, none of which were present in t he MB AA/JBAA libraries. The diversity and distribution of the
120 remaining clone sequences related to phthalate dioxyge nases across these three libraries appear to be similar enough to form a reasonably well associated cluster. The diesel exposed NBAA cl one library contains an even more diverse assortment of ARD like sequences than either R14 or AS libraries Furthermore, a higher relative percentage of these sequences are associated with non phthalate dioxygenases. As a result, this library is well sep arated on PCoA and does not correlate well with any other soil clone library A reasonable interpretation for this difference is that both the form and level of exposure to PAHs that this soi l was subjected to selected for Actinobacterial ARD like genotyp es unique to those conditions ; conditions which might not have been present in the other contaminated soils. The soil that had not previously been directl y impacted by fossil fuels presented with ARD like sequences similar to those encoding enzymes for the degradation of aromatic su bstrates Since we did not amend our soils with any bacterial cultures our results prove that the detectable diversity and distribution of these ARD like gene sequences in diesel exposed soil must have co me from endogenous bacteria that were already present in the soil prior to exposure The following naphthalene concentration data was obtained for the following soils: 1) diesel exposed NB AA, 2100 g/Kg soil; 2) unexposed control NBAA, 8.8 g/Kg soil; 3) R14 soil, 480 g/Kg soil; 4) MW soil, 4.7 g/Kg soil. The measured concentrations of naphthalene in MW and unexposed control NBAA soil were close to the method detection limit of 4.2 4.4 g/Kg soil as performed by Test America. A Mantel analysis was performed for these concentration data against a separate UniFrac
121 distance matrix which included sequences from all soils for which we have naphthalene concentration data but excluding unana lyzed MBAA and JBAA soil libraries. The resulting r value was 0.1871. The significance (P) value was calculated as 0.612 indicating that naphthalene concentrations were not significant with respect to the differences seen in the diversity and distributi on of ARD like sequences between clone libraries Further Mantel analyses concerning TKN, NOx N, pH and OM% are forthcoming and may be significant factors as they were for Chapter 3 and 4 results Discussion This experiment was designed to allow us to qu ickly and effectively, determine if the Actinobacterial ARD like sequence profiles of soils not impacted by PAHs of anthropogenic origin would shift so that they more closely resembled the profiles of soils that were known to be directly contaminated. A lthough the sequence profile of the soil that we exposed to diesel fumes did not closely resemble those of the urban soils, R14 and AS, it d iverge d dramatically from the control soil profile which itself, was most closely associated with the profiles of t he unimpacted MBAA and JBAA samples. As the only variable that was changed between our incubations of unexposed controls and diesel treated soil s was the presence of diesel vapors it seems likely that exposure to diesel fumes was the primary cause of the differences that we see in diversity and distribution of ARD like sequences between the clone libraries. It is interesting then, that a Mantel analysis of the clone library distance matrix against measured naphthalene concentrations showed that the diffe rences in naphthalene concentrations are insignificant in affecting these profile changes between soils. The situation is even more confusing considering that there were such a high percentage of diesel treated NBAA clone sequences that clustered with M2 13 and other
122 Actinobacterial naphthalene/Nid dioxygenases compared to only 2 such sequences in the control s and none in the R14, AS, or May and June sampled NATL soil libraries What we end up seeing, then, is that after exposing t hese soils to diesel va pors an increase in the overall percentage of ARD like sequences related to naphthalene/Nid dioxygenases occurs that is somehow independent of actual naphthalene concentrations f naphthalene concentrations are unrelated to th ese sequence shifts between soils clone libraries, then what other factors are present in the exposed soil that may driving the differences between them and why N, pH, and OM% correlated strongly with the differences we saw between different clone libraries with OM% consistently being the extremely significant. Since we do not currently have measurements for these parameters for either the November sampled BAA field samples or the unexposed control a nd diesel incubation we cannot yet state whether any of them can be correlated to ARD like sequence diversity and distribution differences between treatments. However, the data which suggested that these four soil properties are strongly correlated to the differences we see in our clone libraries, leads us to believe that there is a high probability that one, or all, of these properties may once again prove to be significant in control ling sequence diversity and distribution in soils with different exposures to PAHs. Further analysis of these properties for these soils is currently forthcoming. Despite the current lack of statistical certainty linking differences in ARD like soil clo ne sequences between soils to a given environmental factor, phylogenetic analysis alone
123 has shown that there is most definitely a shift that occurs in the ARD like sequence profiles of soils exposed to PAH vapors and that the sequence profiles of these ex posed soils are indicative of their exposure.
124 Figure 5 1 Phylogenetic associations of unexposed control and diesel exposed NBAA clone ARD like sequences with known reference sequences. Groupings highlighted in red are dominated by a higher relative percentage of diesel exposed clone sequences. Groupings highlighted in blue are dominated by a higher relative percentage of unexposed control clone sequences.
125 Table 5 1. Comparison of the relative percentage of unexposed control and die sel exposed NBAA clone sequences that are divided between different phylogenetic groupings. Pht DO = phthalate dioxygenase; NDO = naphthalene dioxygenase. SOIL % Pht DO % Non Pht DO % NDO Diesel exposed 19.51 80.49 65.85 Unexposed Control 56.25 37. 5 6.25 Figure 5 2. PCoA analysis of Actinobacterial ARD like clone sequences for unexposed control and diesel exposed NBAA, MBAA, JBAA, R14, AS, and MW soils.
126 CHAPTER 6 CONCLUDING STATEMENTS AND FUTURE WORK Conclusions Based on the results of the work conducted and discussed in Chapter 3, 4, and 5 of this thesis, we can make the following conclusions: 1. T he degenerate primers that we develo ped are useful tools in analyzing the distribution and diversity of sequences similar to the Actinobacterial Rie ske type non heme iron di oxygenases involved in both PAH and MAH degradation in soils. 2. W e can effectively amplify the narAa gene product from R. opacus M213 while also determining the Rieske like sequences in soils that are split between those related to the degradation of monocyclic aromatics, such as phthalate and those related to naphthalene or Nid like sequences (which may target multiple polycyclic aromatic substrates). 3. B ased on the phylogenetic associations of clone sequences with known reference genes, we can conclude that the Mycobacteria are the dominant Actinobacterial phthalate degraders and, potentially, PAH and nap hthalene degraders in the unimpacted NATL soils and the directly impacted R14 and AS soils. 4. Alternatively, the November sampl ed BAA soil that we intentio nally exposed to diesel fumes possessed Nid like sequences that were exclusive to the Arthrobacter Gordonia and Rhodococcus naphthalene/Nid dioxygenase references Sequences related to phthalate dioxygenase s were still dominat ed by the Mycobacteria This division of phthalate/non phthalate dioxygenase like ARD sequences among different genera of Actinobacteria in soils with different levels of exposure to PAHs indicates that there may be a selective effect on the microbial con sortia with detectable ARD like genes. 5. P hylogenetic analyses for the NATL and Gainesville soils indicates that more anthropogenically impacted soils show greater diversity in sequences based on substrate utilization while the distribution of these genes among different groups of Actinobacteria appears to be more limited. In all cases, the Mycobacteria are the most highly represented and fewer sequences from additional genera are found (i.e., Arthrobacter, Rhodococcus, etc. ) with the exception of our int entionally exposed NBAA soil which received a much more acute exposure to PAH fumes 6. A prelimnary comparison of Proteobacterial nahAc like sequences for select soils against their Actinobacterial ARD like counterparts and measured soil chemical properties indicates that Proteobacterial and Actinobacterial ARD sequence
127 diversity and distribution are controlled independently of each other. nahAc like sequences were present in all soils, where tested, without corresponding detection of M213 or Rhodococcus l ike NDO. Mycobacterial like PAH dioxygenases were detected, but were few in comparison to phthalate dioxygenase like ARDs and the Proteobacteria l NDOs This could indicate that the Proteobacteria are the most likely PAH degraders in the soils for which we have comparative data under in situ conditions and that naphthalene concentrations are the more likely driver of nahAc like diversity and distribution 7. For our burn treatment experiment, it appeared, at first, that burning had an impact on the distr ibution and diversity of Actinobacterial ARD like sequences co mpared to unburned controls for certain soils but not others. May sampled burned soils appeared enriched for more diverse, more highly distributed Actinobacterial phthalate dioxygenase like seq uences. After one month, the diversity and distribution of phthalate dioxygenase like sequences decreased and additional sequences associated with non phthalate dioxygenase like genes were detected. However, statistical analysis of the sequence data reve aled no statistically relevant impact of burning on ARD like sequence distribution and diversity appears to exist within the NATL. 8. Naphthalene concentrations alone do not appear to correlate to the differences that we have seen in impacted and un impacted soils including those soils that intentionally exposed to concentrated PAH vapors. Instead, s oil organic matter content appears to be the most significant driver of Actinobacterial ARD like gene sequence diversity and distribution between soils that are more directly impacted by PAHs and those that are less impacted. The significance of organic matter content cannot be seen when considering un impacted or impacted soils alone, but only by considering them side by side Higher soil organic matter conten t generally results in a greater diversity of organic substrates and increased bioavailability of these compounds to microorganisms with increasing concentrations. Consequently, we see the highest diversity and distribution of Actinobacterial ARD like gen e sequences in soils that have been more directly impacted by PAHs. Future Work In light of the work that has been discussed, many questions still remain unanswered and demand further effort in research. To begin with, we still have not fully answered the question about what is truly driving the diversity and distribution of Actinobacterial ARD like sequences in soils, whether they are directly impacted by PAHs of anthropogenic origin or not. Our data suggest that soil organic matter content
128 is the ove rriding variable in this situation but more extensive experimentation and soil chemical analyses are needed to verify this assumption Additionally, we still do not know whether it is the more well studied Proteobacteria or the Actinobacteria which are t he more significant contributors to PAH degradation in soils. Although much work has been done on that topic, some here but most elsewhere, our data and the literature currently available suggest that it is not an endent on both nutrient and PAH impact levels in soils. Furthermore, we now have to consider the question about o phthalic acid and its role as both an intermediate in the degradation of PAHs, such as naphthalene, anthracene, phenanthrene, etc. and as an indicator of both microbial diversity and degradation pathways among soil bacteria. Recent research has revealed that PAH degradation through pht halate is not as limited as we once thought and any future work on this topic should include a far more exten sive comparison of a broad range of Proteobacterial ARD like sequences alongside their Actinobacterial counterparts. Additionally, more complete soil nutrient and PAH analyses would more definitively correlate soil physical/chemical properties to the dist ribution and diversity of soil DNA sequences related to the degradation of multiple aromatic substrates from a very diverse array of bacteria. Many questions remain unanswered and the future of scientific research in this area of environmental science st ill offers many opportunities for scientists to more thoroughly elucidate the environmental variables that are controlling the diversity,
129 distribution, and activities of these microorganisms which may prove to be invaluable in future decades as soil contam ination with PAHs is likely to increase.
130 LIST OF REFERENCES Avonce, N., Mendoza Vargas, A., Morett, E., & Iturriaga, G. (2006). Insights on the evolution of trehalose biosynthesis. BMC evolutionary biology 6 (1), 109. Bell, K. S. Philp, J. C., Aw, D. W. J., & Christofi, N. (1998). The genus Rhodococcus. Journal of applied microbiology 85 (2), 195 210. Bull, A.T., Stach, J.E.M., Ward, A.C., & Goodfellow, M. 2005. Marine actinobacteria: perspectives, challenges, future directions An tonie van Leeuwenhoek. 87:65 79. Bugg, T. D., Ahmad, M., Hardiman, E. M., & Singh, R. (2011). The emerging role for bacteria in lignin degradation and bio product formation. Current opinion in biotechnology 22 (3), 394 400. Carrick, H. J., Lowe, R. L., & Rotenberry, J. T. (1988). Guilds of benthic algae along nutrient gradients: relationships to algal community diversity. Journal of the North American Benthological Society 117 128. Cerniglia, C. E. (1992). Biodegradation of polycyclic aromatic hydrocar bons. Biodegradation 3 (2), 351 368. Ferrero, M., Llobet Brossa, E., Lalucat, J., Garca Valds, E., Rossell Mora, R., & Bosch, R. (2002). Coexistence of two distinct copies of naphthalene degradation genes in Pseudomonas strains isolated from the western Mediterranean region. Applied and environmental microbiology 68 (2), 957 962. Fiechter, A. (1992). Biosurfactants: moving towards industrial application. Trends in Food Science & Technology 3 286 293. Filonov, A. E., Puntus, I. F., Karpov, A. V., Koshele va, I. A., Kashparov, K. I., Slepenkin, A. V., & Boronin, A. M. (2004). Efficiency of naphthalene biodegradation by Pseudomonas putida G7 in soil. Journal of Chemical Technology and Biotechnology 79 (6), 562 569. Goodfellow, M., & Williams, S. T. (1983). E cology of actinomycetes. Annual Reviews in Microbiology 37 (1), 189 216. Goyal, A. K., & Zylstra, G. J. (1997). Genetics of naphthalene and phenanthrene degradation by Comamonas testosteroni. Journal of industrial microbiology & biotechnology 19 (5), 401 4 07. Habe, H., & Omori, T. (2003). Genetics of polycyclic aromatic hydrocarbon metabolism in diverse aerobic bacteria. Bioscience, biotechnology, and biochemistry 67 (2), 225 243.
131 Holmfeldt, K., Dziallas, C., Titelman, J., Pohlmann, K., Grossart, H. P., & R iemann, L. (2009). Diversity and abundance of freshwater Actinobacteria along environmental gradients in the brackish northern Baltic Sea. Environmental microbiology 11 (8), 2042 2054. Hopwood, D. A., & Merrick, M. J. (1977). Genetics of antibiotic product ion. Bacteriological reviews 41 (3), 595. Husain, S. (2008). Literature overview: Microbial metabolism of high molecular weight polycyclic aromatic hydrocarbons. Remediation Journal 18 (2), 131 161. Hwang, H. M., Wade, T. L., & Sericano, J. L. (2003). Con centrations and source characterization of polycyclic aromatic hydrocarbons in pine needles from Korea, Mexico, and United States. Atmospheric environment 37 (16), 2259 2267. Izmalkova, T. Y., Mavrodi, D. V., Sokolov, S. L., Kosheleva, I. A., Smalla, K., T homas, C. M., & Boronin, A. M. (2006). Molecular classification of IncP 9 naphthalene degradation plasmids. Plasmid 56 (1), 1 10. I wahori, K., Tokutomi, T., Miyata, N., & Fujita, M. (2001). Formation of stable foam by the cells and culture supernatant of G ordonia ( Nocardia ) amarae Journal of bioscience and bioengineering 92 (1), 77 79. Johnsen, A. R., & Karlson, U. (2007). Diffuse PAH contamination of surface soils: environmental occurrence, bioavailability, and microbial degradation. Applied microbiology and biotechnology 76 (3), 533 543. Khodadad, C. L., Zimmerman, A. R., Green, S. J., Uthandi, S., & Foster, J. S. (2011). Taxa specific changes in soil microbial community composition induced by pyrogenic carbon amendments. Soil Biology and Biochemistry 43 (2), 385 392. Killham, K., & Firestone, M. K. (1984). Salt stress control of intracellular solutes in streptomycetes indigenous to saline soils. Applied and environmental microbiology 47 (2), 301 306. Kim, S. J., Kweon, O., Jones, R. C., Edmondson, R. D., & Cerniglia, C. E. (2008). Genomic analysis of polycyclic aromatic hydrocarbon degradation in Mycobacterium vanbaalenii PYR 1. Biodegradation 19 (6), 859 881. Kim, E. J., Choi, S. D., & Chang, Y. S. (2011). Levels and patterns of polycyclic aromatic hydroc arbons (PAHs) in soils after forest fires in South Korea. Environmental Science and Pollution Research 18 (9), 1508 1517. Kulakov, L. A., Chen, S., Allen, C. C., & Larkin, M. J. (2005). Web type evolution of Rhodococcus gene clusters associated with utiliz ation of naphthalene. Applied and environmental microbiology 71 (4), 1754 1764.
132 Kuyukina, M. S., Ivshina, I. B., Makarov, S. O., Litvinenko, L. V., Cunningham, C. J., & Philp, J. C. (2005). Effect of biosurfactants on crude oil desorption and mobilization in a soil system. Environment international 31 (2), 155 161. Kuyukina, M. S., Ivshina, I. B., Serebrennikova, M. K., Krivorutchko, A. B., Podorozhko, E. A., Ivanov, R. V., & Lozinsky, V. I. (2009). Petroleum contaminated water treatment in a fluidized bed bioreactor with immobilized Rhodococcus cells. International Biodeterioration & Biodegradation 63 (4), 427 432. Kuyukina, M., & Ivshina, I. (2010). Application of Rhodococcus in Bioremediation of Contaminated Environments. Biology of Rhodococcus 231 262. Kweon, O., Kim, S. J., Freeman, J. P., Song, J., Baek, S., & Cerniglia, C. E. (2010). Substrate specificity and structural characteristics of the novel rieske nonheme iron aromatic ring hydroxylating oxygenases NidAB and NidA3B3 from Mycobacterium vanbaale nii PYR 1. MBio 1 (2). Larkin, M. J., Allen, C. C., Kulakov, L. A., & Lipscomb, D. A. (1999). Purification and characterization of a novel naphthalene dioxygenase from Rhodococcus sp. strain NCIMB12038. Journal of bacteriology 181 (19), 6200 6204. Larkin, M. J., Kulakov, L. A., & Allen, C. C. (2005). Biodegradation and Rhodococcus masters of catabolic versatility. Current opinion in biotechnology 16 (3), 282 290. Larkin, M.J., Kulakov, L.A., & Allen, C.R.R. ( 2010 ) Rhodococcus In Handbook ofHydrocarbon a nd Lipid Microbiology, Part 14 (Springer Verlag Berlin Heidelberg), pp. 1345 1353. Lechevalier, H. A., & Lechevalier, M. P. (1967). Biology of actinomycetes. Annual Reviews in Microbiology 21 (1), 71 100. Lloyd Jones, G., Laurie, A.D., Hunter, D.W.F., & Fraser, R. ( 1999 ) Analysis of catabolic genes for naphthalene and phenanthrene degradation in contaminated New Zealand soils FEMS Microbi al Ecol ogy 29:69 79. Lozupone, C., Hamady, M., & Knight, R. (2006). UniFrac an online tool for comparing micr obial community diversity in a phylogenetic context. BMC bioinformatics 7 (1), 371. Maldonado, L. A., Stach, J. E., Pathom aree, W., Ward, A. C., Bull, A. T., & Goodfellow, M. (2005). Diversity of cultivable actinobacteria in geographically widespread mari ne sediments. Antonie van Leeuwenhoek 87 (1), 11 18. Mallick, S., Chakraborty, J., & Dutta, T. K. (2011). Role of oxygenases in guiding diverse metabolic pathways in the bacterial degradation of low molecular weight polycyclic aromatic hydrocarbons: A revi ew. Critical reviews in microbiology 37 (1), 64 90.
133 Marr, L. C., Kirchstetter, T. W., Harley, R. A., Miguel, A. H., Hering, S. V., & Hammond, S. K. (1999). Characterization of polycyclic aromatic hydrocarbons in motor vehicle fuels and exhaust emissions. E nvironmental science & technology 33 (18), 3091 3099. Mastrangelo, G., Fadda, E., & Marzia, V. (1996). Polycyclic aromatic hydrocarbons and cancer in man. Environmental health perspectives 104 (11), 1166. McCarthy, A. J., & Williams, S. T. (1992). Actinomy cetes as agents of biodegradation in the environment a review. Gene 115 (1), 189 192. Mummey, D. L., & Stahl, P. D. (2004). Analysis of soil whole and inner microaggregate bacterial communities. Microbial ecology 48 (1), 41 50. Ni Chad hain, S. M. Norman, R. S., Pesce, K. V., Kukor, J. J., & Zylstra, G. J. (2006). Microbial dioxygenase gene population shifts during polycyclic aromatic hydrocarbon biodegradation. Applied and environmental microbiology 72 (6), 4078 4087. Nojiri, H., Sota, M., & Shintani, M. ( 2009). Catabolic plasmids involved in the degradation of polycyclic aromatic hydrocarbons and heteroaromatic compounds. Microbial Megaplasmids 55 87. Parales, R. E. (2003). The role of active site residues in naphthalene dioxygenase. Journal of industrial microbiology & biotechnology 30 (5), 271 278. Pathak, A., Green, S. J., Ogram, A., & Chauhan, A. (2013). Draft genome sequence of Rhodococcus opacus strain M213 shows a diverse catabolic potential. Genome announcements 1 (1). Patrauchan, M. A., Florizone, C., Dosanjh, M., Mohn, W. W., Davies, J., & Eltis, L. D. (2005). Catabolism of benzoate and phthalate in Rhodococcus sp. strain RHA1: redundancies and convergence. Journal of bacteriology 187 (12), 4050 4063. Prez Pantoja, D., Donoso, R., Junca, H., Gonz lez, B., & Pieper, D. H. (2009). Phylogenomics of aerobic bacterial degradation of aromatics. Handbook of Hydrocarbon and Lipid Microbiology 1355 1397. Phelan, M. B., Crawford, D. L., & Anthony III, L. (1979). Isolation of lignocellulose decomposing acti nomycetes and degradation of specifically 14C labeled lignocelluloses by six selected Streptomyces strains. Canadian journal of microbiology 25 (11), 1270 1276. Phillips, D. H. (1999). Polycyclic aromatic hydrocarbons in the diet. Mutation Research/Genetic Toxicology and Environmental Mutagenesis 443 (1), 139 147.
134 Preuss, R., Angerer, J., & Drexler, H. (2003). Naphthalene an environmental and occupational toxicant. International archives of occupational and environmental health 76 (8), 556 576. Ron, E. Z., & Rosenberg, E. (2001). Natural roles of biosurfactants. Environmental microbiology 3 (4), 229 236. Samanta, S. K., Singh, O. V., & Jain, R. K. (2002). Polycyclic aromatic hydrocarbons: environmental pollution and bioremediation. TRENDS in Biotechnology 2 0 (6), 243 248. Schecter, A., Birnbaum, L., Ryan, J.J., & Constable, J.D. 2006. Dioxins: An overview Environ Research. 101(3):419 428. Schloss, P. D., & Handelsman, J. (2005). Introducing DOTUR, a computer program for defining operational taxonomic u nits and estimating species richness. Applied and environmental microbiology 71 (3), 1501 1506. Schreiner, C. (2003). Genetic toxicity of naphthalene: a review. Journal of Toxicology and Environmental Health Part B: Critical Reviews 6 (2), 161 183. Sekar, R., Pernthaler, A., Pernthaler, J., Warnecke, F., Posch, T., & Amann, R. (2003). An improved protocol for quantification of freshwater Actinobacteria by fluorescence in situ hybridization. Applied and Environmental Microbiology 69 (5), 2928 2935. Stackebran dt, E., Rainey, F. A., & Ward Rainey, N. L. (1997). Proposal for a new hierarchic classification system, Actinobacteria classis nov. International journal of systematic bacteriology 47 (2), 479 491. Sudnitsyn, I. I. (2009). Specificity of actinomycetes in salt affected soils. Eurasian Soil Science 42 (2), 235 236. Tamura, K., Peterson, D., Peterson, N., Stecher, G., Nei, M., & Kumar, S. (2011). MEGA5: molecular evolutionary genetics analysis using maximum likelihood, evolutionary distance, and maximum parsim ony methods. Molecular biology and evolution 28 (10), 2731 2739. Team, R. C. (2008). R: A language and environment for statistical computing. R Foundation Statistical Computing Technological Planning and Management Corporation. ( 2002 ) Final Report on C arcinogens Background Document for Naphthalene Prepared for the U.S. Department of Health and Human Services. Thompson, J. D., Gibson, T. J., Plewniak, F., Jeanmougin, F., & Higgins, D. G. (1997). The CLUSTAL_X windows interface: flexible strategies for multiple sequence alignment aided by quality analysis tools. Nucleic acids research 25 (24), 4876 4882.
135 Uz, I., Duan, Y. P., & Ogram, A. (2000). Characterization of the naphthalene degrading bacterium, Rhodococcus opacus M213. FEMS microbiology letters 185 (2), 231 238. Ventura, M., Canchaya, C., Tauch, A., Chandra, G., Fitzgerald, G. F., Chater, K. F., & Van Sinderen, D. (2007). Genomics of Actinobacteria: tracing the evolutionary h istory of an ancient phylum. Microbiology and molecular biology reviews 71 (3), 495 548. Vaillancourt, F.H., Bolin, J.T., & Eltis, L.D. 2004. Ring Cleavage Dioxygenases. In Pseudomonas: Biosynthesis of macromolecules and molecular metabolism Juan Luis Ramos, ed. (Kluwer Academic/Plenum Publishers), pg. 359. Vergnoux, A., Malleret, L., Asia, L., Doumenq, P., & Theraulaz, F. (2011). Impact of forest fires on PAH level and distribution in soils. Environmental Research 111 (2), 193 198. Volk, D.E., Thiviyan athan, V., Rice, J.S., et al. ( 2003 ) Solution Structure of a CisOpened (10 R ) N 6 Deoxyadenosine Adduct of (9 S ,10 R ) 9,10 Epoxy 7,8,9,10 tetrahydrobenzo[ a ]pyrene in a DNA Duplex 42:1410 1420. Ward, A. C., & Bora, N. (2006). Diver sity and biogeography o f marine actinobacteria. Current opinion in microbiology 9 (3), 279 286. Warnecke, F., Amann, R., & Pernthaler, J. (2004). Actinobacterial 16S rRNA genes from freshwater habitats cluster in four distinct lineages. Environmental microbiology 6 (3), 242 253. Watve, M. G., Tickoo, R., Jog, M. M., & Bhole, B. D. (2001). How many antibiotics are produced by the genus Streptomyces?. Archives of microbiology 176 (5), 386 390. Winslow, C. E., Broadhurst, J., Buchanan, R. E., Krumwiede Jr, C., Rogers, L. A., & Smith, G. H. (1920). The families and genera of the bacteria: final report of the committee of the Society of American Bacteriologists on characterization and classification of bacterial types. Journal of bacteriology 5 (3), 191 Zhi, X. Y., Li, W. J., & Stackeb randt, E. (2009). An update of the structure and 16S rRNA gene sequence based definition of higher ranks of the class Actinobacteria, with the proposal of two new suborders and four new families and emended descriptions of the existing higher taxa. Interna tional journal of systematic and evolutionary microbiology 59 (3), 589 608. Zimmermann, W. (1990). Degradation of lignin by bacteria. Journal of biotechnology 13 (2), 119 130.
136 BIOGRAPHICAL SKETCH Christopher Weidow was born in Manhattan, New York in July 1985 to parents Rebecca Lynn Nichting and Carl Henry Weidow II but has been living in Florida since 1994. He and his family have spent most of their time living in the Jacksonville area, where his mother and younger brother still live. After attending th e Florida Community College at Jacksonville (now Florida College at Jacksonville) where he obtained his Associate of Arts degree, h e came to Gainesville to attend the University of Florida in the fall of 2006. H e majored in m icrobiology and c ell s cience with two minors, one in c hemistry and the other in r eligion. He gradu ated with honors from the College of Agriculture and Life Sciences in December 2008 with his Bachelor of Science degree He then enrolled in the the fall of 2009 in the Department of Soil and Water Sciences. Originally intending to obtain a Doctor of Philosophy degree for research on the microbiological controls of methylmercury formation and degradation in the Florida Everglades, he switched his project in his second year to the study of the degradation of aromatic compounds by Actinobacteria in soils where he completed his Master of Science thesis research He graduate d in May 2 013