The Distribution and Diversity of Functional Gene Pathways

Louisiana State University LSU Digital Commons

LSU Master's Theses Graduate School

2012 The Distribution and Diversity of Functional Gene Pathways Controlling Sulfur Speciation in Lower Kane Cave, Wyoming Audrey Tarlton Paterson Louisiana State University and Agricultural and Mechanical College, [email protected]

Follow this and additional works at: https://digitalcommons.lsu.edu/gradschool_theses Part of the Earth Sciences Commons

Recommended Citation Paterson, Audrey Tarlton, "The Distribution and Diversity of Functional Gene Pathways Controlling Sulfur Speciation in Lower Kane Cave, Wyoming" (2012). LSU Master's Theses. 2877. https://digitalcommons.lsu.edu/gradschool_theses/2877

This Thesis is brought to you for free and open access by the Graduate School at LSU Digital Commons. It has been accepted for inclusion in LSU Master's Theses by an authorized graduate school editor of LSU Digital Commons. For more information, please contact [email protected].

THE DISTRIBUTION AND DIVERSITY OF FUNCTIONAL GENE PATHWAYS CONTROLLING SULFUR SPECIATION IN LOWER KANE CAVE, WYOMING

A Thesis

Submitted to the Graduate Faculty of the Louisiana State University and Agricultural and Mechanical College in partial fulfillment of the requirements for the degree of Master of Science

The Department of Geology and Geophysics

by Audrey Tarlton Paterson B.S., Louisiana State University, 2009 May 2012

ACKNOWLEDGEMENTS

I would like to acknowledge Dr. Annette Summers Engel who was the major advisor for the project. Thank you for bringing me into your lab group as an undergrad, introducing me to an amazing group of people, accepting me as a graduate student, and encouraging me to ‘stick my neck out.’ I cannot imagine where I would be without your guidance. I sincerely appreciate the time, effort, and funding you provided that made this project an enjoyable and memorable experience for me.

I also especially thank Dr. Amitava Roy, Dr. Huiming Bao, Dr. Sam Bentley, and Dr.

Rachel Beech for patience and guidance while serving on my thesis committee. I am very proud to have such inspiring, interdisciplinary committee members. Technical support at CAMD was provided by Greg Merchan, Henning Lichtenberg, and Lisa Bovenkamp. Partners in the field and lab included Kathleen Brannen, Terri Brown, Shane Cone, Chang Liu, Ben Maas, Axita Neema,

Sarah Keenan, and Brendan Headd. Thank you all so much for your brilliance, energy and enthusiasm.

To my family, Malcolm and Dorothy Paterson, Andrew Paterson, Helen and Alec

Paterson, Damon and Dot Slator, Terry Huffington and the Dittman family, William Grierson, the Wilds, the Rickmans, the Bakers, the Youngs, the Skinners, the O’Connors, the Herkenhoffs, thank you for emotional and technical support, love and encouragement!

This research was partially supported by two grants from the National Science

Foundation (DEB-0640835 and EAR-0844364) and a grant from the Louisiana Board of Regents

Pilot fund for New Research Program [NSF(2010)-PFUND-174]. Support for the project also came from receiving the James G. Mitchell award for best student presentation at the 2011

National Speleological Society convention, Glenwood Springs, Colorado. ii

TABLE OF CONTENTS

ACKNOWLEDGEMENTS ...... ii

LIST OF TABLES ...... v

LIST OF FIGURES ...... vi

ABSTRACT ...... vii

CHAPTER 1: INTRODUCTION ...... 1 Research Objectives and Hypotheses ...... 1 Scientific Importance ...... 5 Thesis Organization ...... 6

CHAPTER 2: REVIEW OF LITERATURE ...... 7 Introduction ...... 7 Lower Kane Cave ...... 7 Microbial Roles in the Sulfur Cycle ...... 9 The Sulfur Oxidation Multienzyme (Sox) System ...... 10 Adenosine-5’-phosphosulfate (APS) Reductase ...... 13 Biological Sulfur Speciation in Biological and Geological Materials ...... 14 The Sulfur Cycle in Karst Systems ...... 16

CHAPTER 3: MATERIALS AND METHODS ...... 18 Sample Collection and Aqueous Geochemistry ...... 18 XANES Spectroscopy, Spectral Fitting, and Quantitative Analyses ...... 21 Microbial Mat DNA Extraction ...... 22 Functional Gene PCR Amplification and Cloning ...... 24 Functional Gene Sequence Analysis ...... 28 Pyrosequencing of 16S rRNA Genes ...... 28 Statistical Analyses ...... 30

CHAPTER 4: RESULTS ...... 32 Upper Spring Microbial Mat Morphology and Aqueous Geochemistry ...... 32 Sulfur Speciation in the Microbial Mats ...... 34 Functional Gene Diversity and Distribution ...... 36 soxB ...... 36 aprA ...... 38 Diversity of 16S rRNA Gene Sequences ...... 46 Statistical Comparisons ...... 50

CHAPTER 5: DISCUSSION ...... 57 Links between Geochemistry and Biodiversity ...... 57 Importance of Missing Functional Representation for Epsilonproteobacteria ...... 59 iii

Microbial Metabolisms in Industrial Settings ...... 60 Conclusions and Future Directions ...... 60

REFERENCES ...... 62

APPENDIX A: LOWER KANE CAVE SPRING GEOCHEMISTRY ...... 74

APPENDIX B: DIVERSITY OF 16S RRNA GENE SEQUENCES ...... 77

VITA ...... 109

LIST OF TABLES

3.1 Functional gene-targeting primers used in PCR assays in this study ...... 25

4.1 Geochemical results from Lower Kane Cave, listed in meters from the back of the cave to the front of the cave ...... 33

4.2 Percentages of chemical sulfur species contribution based on the best fittings of the XANES analyses of microbial mat samples ...... 34

4.3 Lower Kane Cave soxB sequence diversity, as number of sequences with maximum shared identity to GenBank sequences with taxonomic affiliations ...... 39

4.4 Lower Kane Cave soxB sequence diversity, as number of sequences with maximum shared identity to GenBank sequences with and without taxonomic affiliations ...... 48

4.5 Distribution of 16S rRNA pyrosequences among classes of Proteobacteria ...... 50

LIST OF FIGURES

2.1 Location of Lower Kane Cave in Wyoming, USA, and schematic diagram of the cave with springs and mat types ...... 9

3.1 Images from the Upper Spring area of Lower Kane Cave ...... 19

3.2 Sulfur K-edge XANES spectra (normalized and stacked) for standard reference compounds used for linear combination fittings of microbial mat sample spectra ...... 23

4.1 Dissolved oxygen and sulfide concentrations from Upper Spring, shown in mmol/L .... 33

4.2 Upper Spring microbial mat XANES spectra ...... 35

4.3 Rarefaction curves for functional genes, based on OTUs defined from sequences ...... 40

4.4 Distribution of soxB OTUs among phyla in Upper Spring white filaments and webs .... 41

4.5 Distribution of aprA OTUs among phyla in Upper Spring white filaments and webs .... 44

4.6 Rarefaction curves for 16S rRNA gene sequences based on 96% sequence similarity.....45

4.7 Microbial community diversity as normalized abundance of 16S rRNA gene pyrosequences within the orders of major phyla or classes ...... 47

4.8 CCA plot of soxB OTUs related to major geochemical gradients described by Axes 1 and 2, which explain 34.2% and 29.6% of the variation among samples ...... 52

4.9 CCA plot of soxB OTUs related to major geochemical gradients described by Axes 2 and 3, which explain 29.6% and 25.4% of the variation among samples ...... 53

4.10 CCA plot of aprA OTUs related to major geochemical gradients ...... 55

ABSTRACT

Defining the linkages between microbial metabolic activity and environmental geochemistry can be used to understand how carbonate dissolution in sulfidic karst systems proceeds. In Lower

Kane Cave, Wyoming, sulfidic springs support taxonomically distinct microbial communities that are also associated with putative sulfur metabolisms (e.g., oxidation of reduced sulfur compounds, sulfate reduction, and sulfur disproportionation) that influence carbonate dissolution. The distribution of diverse microbial groups in other terrestrial subsurface environments is not well understood, making Lower Kane Cave ideal for study. Molecular genetics techniques were used to uncover diversity of 16S rRNA genes and bacterial soxB and aprA functional genes. Aqueous geochemistry was compared to sediment sulfur speciation based on X-ray absorption near-edge structure spectroscopy. The distribution of functional genes significantly correlated to gradients of dissolved oxygen and sulfide concentrations compared to the relative contributions of sediment elemental sulfur, either in the cyclo-octasulfur (S8) or polymeric sulfur forms. Despite Epsilonproteobacteria being the most abundant putative sulfur- oxidizing bacteria within the communities, particularly in upstream mats, Gammaproteobacteria of the order Thiothricales and Betaproteobacteria represented soxB gene diversity downstream and correlated to dissolved oxygen and stored S8. Deltaproteobacterial aprA groups, some of which could be linked to disproportionation, were almost exclusively associated with upstream mats and correlated to high sulfide concentrations, as well as to cysteine, stored sulfate, and methionine gradients within the sediments. Microbial storage of elemental sulfur in downstream mats precludes proton generation and diminishes the potential for limestone dissolution by those microbial groups, supporting that carbonate dissolution linked to microbial oxidation of reduced sulfur compounds would be expected upstream. vii

CHAPTER 1

INTRODUCTION

Project Overview

Lower Kane Cave, located in northern Wyoming, contains several springs that discharge into the cave and form outflow streams. The springs contain dissolved hydrogen sulfide (H2S) and support diverse and flourishing communities of microorganisms and invertebrates. Lower

Kane Cave has been the focus of many geochemistry and ecology studies in the past fifty years, predominately because of its accessibility but also unusual cave formation mechanism involving sulfuric acid speleogenesis promoted by abiotic and microbial sulfide oxidation (Egemeier, 1981;

Engel et al., 2004b). Recent biological assessments of the microbial mats in the cave, based on describing the microbial diversity to the species level from sequenced 16S rRNA genes, suggest the presence of taxonomically distinct organisms capable of oxidizing reduced sulfur compounds, reducing sulfate, and disproportionating elemental sulfur, thiosulfate, and sulfite

(Engel et al., 2010). One of the limitations of 16S rRNA based analyses, however, is being able to interpret metabolic functions in natural systems because taxonomy of this gene does not provide a direct link to physiology. Although the previous research indicated that microbial metabolic activity contributes to the overall aqueous and sediment geochemistry in the cave, the diversity of metabolic pathways involved in the sulfur cycle has not been investigated.

Research Objectives and Hypotheses

The goal of this study was to constrain the effects of microbial metabolic activity on local geochemistry in Lower Kane Cave that could affect carbonate dissolution and cave formation, particularly associated with sulfur cycling. To meet this goal, the following objectives were met:

(1) Apply molecular genetics techniques to uncover bacterial functional gene diversity for

pathways associated with the oxidation of reduced sulfur compounds (herein referred to as

sulfur-oxidizing bacteria or sulfur-oxidizers) compared to overall 16S rRNA gene sequence

diversity because these microbial groups have been shown previously to play a role in

carbonate dissolution in sulfidic karst (Engel et al., 2004b; Steinhauer et al., 2010), and

(2) Evaluate the distribution of microbially derived ‘biosulfur’ compounds, such as ‘sulfur

globules,’ compared to abiotic sulfur compounds. Globules are sulfur metabolic byproducts

that are stored intracellularly or formed extracellularly, but biosulfur compounds could

include a range of chemical species that have a distinct chemical signature compared to

abiotically produced elemental sulfur.

These interdisciplinary approaches uncovered the relationships and interactions among metabolism, substrate availability, and sulfur recycling by different naturally occurring microbial groups. The methods used included making descriptions of overall microbial diversity from 454 tag pyrosequencing of 16S rRNA genes, isolation and analysis of metabolic soxB and aprA functional gene diversity, characterization of aqueous geochemistry, and evaluation of sulfur speciation from X-ray Absorption Spectroscopy (XAS) of mats and sediments. The bulk of the research focused at the Upper Spring cave stream location because this was the site where microbial mats were abundant and there were extensive historical data from previous studies

(e.g., Engel et al., 2003; Engel et al., 2004a; Engel et al., 2007; Engel et al., 2010). Microbial communities in these mats have been implicated in causing limestone dissolution and cave development (Engel et al., 2004b; Steinhauer et al., 2010), but confirmation of the genetic basis for metabolic activities has not been done. To test functional gene amplification using newly designed primers and to compare the distribution of functional genes among bacteria in sulfidic 2

environments, samples from throughout Lower Kane Cave were used, as well as samples from the Edwards Aquifer (Gray, 2010) and a marine lucinid gill symbiont (Green-Garcia, 2008).

The first research objective was to describe the genetically controlled metabolic capabilities of putative sulfur-oxidizers by amplifying and characterizing functional genes from environmental DNA using primers designed for several distinct oxidation pathways. Analysis of resulting clone sequences included identifying the organisms to which the genes most likely belonged and then interpreting the relevance of their spatial distribution to overall and local environmental conditions. Two functional genes were chosen, soxB for the sulfur oxidation multienzyme (Sox) complex, and aprA for the adenosine-5-phosphosulfate reductase (Apr) pathway. Because both genes are highly conserved among the different sulfur oxidation metabolic pathways (Meyer et al., 2007; Meyer and Kuever, 2007), the first hypothesis tested was that the presence and diversity of functional genes correlate to substrate availability based on geochemistry. In other words, if the microbial communities within the Upper Spring mats utilize reduced sulfur compounds to gain energy, then the presence of functional genes, and the distribution and diversity of microorganisms capable of that metabolism, should correlate to specific sulfur-based geochemical gradients. The metabolic and geochemical gradients directly influence proton balance locally and can lead to carbonate dissolution (e.g., Engel et al., 2004b).

The second research objective was to identify the chemical signatures suggestive of either abiotic oxidation or biotic reduced sulfur compound (e.g., H2S) oxidation from an independent method. This objective was predicated on the fact that the presence and abundance of a functional gene amplified from environmental DNA using end-point PCR methods does not necessarily indicate active expression of the sequenced genes encoding for those pathways

(Saleh-Lakha et al., 2005; Smith and Osborn, 2009). To know the levels of expression, RNA- 3

based analyses (e.g., transcriptomics) would need to have been done (Poretsky et al., 2005), but this was beyond the scope of my investigation. The second hypothesis tested in this research was that functional activities result in specific microbial mat and sediment geochemistry based on the accumulation of metabolic byproducts. Although a ‘cause’ of geochemical conditions cannot be proven with the methods available for the realm of this study (i.e. mRNA gene analyses;

(Poretsky et al., 2005), it is reasonable to pursue testing this hypothesis by assuming that a strong correlation between microbial mat sulfur geochemistry and functional gene distribution and diversity will provide sufficient evidence to warrant further investigation of the expression of microbial metabolic genes and function. For the purpose of modeling geomicrobial roles in the sulfidic karst system, statistically insignificant correspondence between a gene and geochemical dataset would invalidate the proposed relationship. If microorganisms utilize reduced sulfur compounds predictably based on the functional pathways, then the sulfur chemical signature and sulfur speciation of the mats should contain evidence of that metabolism, such as from the accumulation of metabolic intermediate species or byproducts in sediments.

X-ray Absorption Near-Edge Structure spectroscopy (XANES), a type of XAS method that probes atomic (i.e. valence and geometry) structures of solids, liquids, or gases, provides an alternative approach to characterize microbial function in natural materials and the environment.

It is a non-destructive method that allows for in situ characterization of oxidation states and chemical structures within specific energy ranges (Prange, 2008). The spectra cannot differentiate between abiotically and biologically produced sulfur, but can provide information about microbial activity that give rise to the specific chemical speciation within a sample (Prange et al., 1999; Lee et al., 2007; Franz et al., 2009). There have been an increasing number of studies that characterize sulfur species with XANES from a variety of settings and from cultured 4

organisms (Lee et al., 2007; Prange, 2008; Franz et al., 2009; He et al., 2012), with some finding that elemental sulfur produced by microorganisms is geochemically and physically distinct from inorganically produced sulfur “flower” (Janssen et al., 1999).

Scientific Importance

Broadly, this research expands our knowledge of functional gene diversity that can be used to constrain evolutionary relationships and the timing of early earth metabolic evolution, particularly associated with the sulfur cycle. For instance, distance analyses of 16S rRNA gene sequences indicate that non-photosynthetic sulfur-oxidizing bacteria likely originated as early as at least ~1.44 Ga, perhaps earlier, but most microbes dependent on the availability of free oxygen

(O2) likely arose ~600 to 800 Ma during or after the “Great Oxidation Events” (Canfield and

Teske, 1996). The soxB gene is found in a variety of sulfur-oxidizing bacteria with diverse metabolisms, some dependent on O2, others that can use alternative electron acceptors like nitrate, and others that use a specific species of sulfur (e.g., thiosulfate), while others use many forms of reduced sulfur metabolically (Meyer et al., 2007). Therefore, evaluating the phylogenetic differences among genes, as they are linked to habitat preference, geochemistry, and overall metabolism, will help to understand diversification events in earth’s history.

Environments like Lower Kane Cave, where sulfide naturally enters the system and is transformed and removed by a flourishing community of sulfur-oxidizers, are extremely valuable natural analogs to engineered and industrial settings, including downstream hydrocarbon and petrochemical wastewater treatment systems (EPA, 1993, 1995; Khanal, 2008), agricultural and tanning wastes (EPA, 2003), acid mine drainage and refractory mineral bioleaching (Tang et al.,

2009), and bioenergy development (Khanal, 2008). Hydrogen sulfide is a cytotoxic gas that causes asphyxiation in humans (EPA, 2002), but that can also be harmful to the ambient 5

environment (EPA, 2002). Referred to as biological sulfide oxidation (BSO) in process industry,

BSO remediation is gaining popularity due to the relatively high efficiency, economic favorability, and reduction in secondary pollutants compared to chemical or physical stripping

(Khanal, 2008). Traditional BSO systems often employ cultured microbes, such as the well- studied sulfide-oxidizing Thiobacillus spp. (Oh et al., 1998). Thiobacilli use sulfide for energy conservation, but are susceptible to toxicity at concentrations of 5-30 mg/L (Buisman et al.,

1991). For engineered and industrial applications, ideally microbes should tolerate high sulfide concentrations or pulses in substrate availability. Recently, metabolically diverse, chemolithoautotrophic sulfur-oxidizers have been the focus of research on sulfide removal

(Reyes-Avila et al., 2004; Vannini et al., 2008; Ji et al., 2009; Li et al., 2012). Therefore, the naturally aphotic (cave) microbial communities, comprised of chemolithoautotrophs that convert sulfide to byproducts (Engel et al., 2010), can help to improve modern biotechnology aimed at remediating industrial wastewater problems.

Thesis Organization

This thesis is divided into chapters that describe the aqueous and solid geochemistry and bacterial gene diversity of Upper Spring in Lower Kane Cave. Chapter 2 provides a literature review covering the topics of the Lower Kane Cave geologic setting, microbial sulfur metabolisms, sulfur oxidation genes and metabolic byproducts, and a review of karst sulfur geochemistry. Chapter 3 outlines methods used for collection and analysis of aqueous geochemistry, amplification of functional and 16S rRNA genes and diversity analyses, and examination of sulfur chemical species by XANES. The results are described in Chapter 4, and

Chapter 5 is the discussion, which includes a more detailed explanation of the importance of the research to broad areas of science and technology, and a summary of future research directions. 6

CHAPTER 2

REVIEW OF LITERATURE

Introduction

Lower Kane Cave is located in the Bighorn County of northern central Wyoming, along the western perimeter of the Bighorn Mountains. The cave is forming within Mississippian-age, upper Madison Limestone that has been folded into an anticline from sulfuric acid speleogenesis

(SAS). This chapter describes the geological setting of Lower Kane Cave, and provides the background and literature review for current knowledge of major biological oxidation pathways within the sulfur cycle, metabolic intermediates produced by bacterial reduced sulfur compound

2- (e.g. H2S, S2O3 ) oxidation, and microbial activities in sulfidic karst systems.

Lower Kane Cave

The Madison Limestone is the reservoir for the Madison Aquifer, which is an important supply of drinking water in Wyoming, as well as significant for hydrocarbon production

(McCaleb and Wayhan, 1969; Downey, 1984; Greene, 1993; Stacy and Huntoon, 1994;

Westphal et al., 2004). The Madison Limestone (or Madison Group, where differentiated) was originally deposited in a shelf-and-basin system over a period of ~12 million years (357-345

Ma), covering an area that approximately extends from Arizona to western Canada (Downey,

1984; Plummer et al., 1990; Busby et al., 1991; Fischer et al., 2005). Marine transgression and regression cycles supported abundant sediment accumulation that led to cyclical reservoir stacking (Busby et al., 1991; Westphal et al., 2004). Dolomitization and anhydrite precipitation in shelf and lagoonal environments, and evidence of halite precipitation during regression, have been described (Busby et al., 1991). Significant uplift and exposure of parts of the Madison

Limestone began during the Laramide Orogeny (70-80 Ma). Although there is evidence of paleokarst (Huntoon, 1985), major folds, faults, and fractures are the primary structural controls on modern groundwater movement on a regional or local scale and the distribution of karst observed in the area (Stacy and Huntoon, 1994; Whitehead, 1996).

The geochemistry of water in Lower Kane Cave is controlled by the source and transport history of the aquifer. In general, recharge areas for the aquifer are at high elevation, such as the crests of anticlines or upthrown fault blocks, including the Bighorn Mountains to the east

(Whitehead, 1996). Wells drilled into the Madison Formation either access the water or hydrocarbon resources, which are among some of the largest reservoirs in the Bighorn Basin

(McCaleb and Wayhan, 1969; Busby et al., 1991; Greene, 1993; Westphal et al., 2004). Overall

Madison Limestone porosity ranges from 0-35% (McCaleb and Wayhan, 1969; Greene, 1993;

Westphal et al., 2004), with higher values being a consequence of dolomitization, dedolomitization, and karstification, and lower values linked to lithologic remineralization. The uppermost Madison is described as cavernous (Greene, 1993).

Aquifer heterogeneity is reflected by the geochemically distinct compositions of water upwelling at different spring orifices inside Lower Kane Cave (Figure 2.1) compared to springs outside the cave. Major aquifer flow patterns and water quality have been described for large regions of the Madison Aquifer (Huntoon, 1985; Plummer et al., 1990; Busby et al., 1991). But, localized groundwater flow in the aquifers west of the Bighorn Mountains, where Lower Kane

Cave is found, is not well described. Several mixing models have been proposed for the regional groundwater transport (Downey, 1984; Huntoon, 1985; Plummer et al., 1990; Stacy and

Huntoon, 1994; Putnam and Long, 2007), with the source of H2S in this and similar systems to the south and west of the area being attributed to the proximity of hydrocarbon reservoirs and 8

geothermal regions. The sulfide in the waters at Lower Kane Cave has been isotopically linked to microbial dissimilatory sulfate reduction compared to thermochemical sulfate reduction, likely as a result of mixing meteoric fluids with basinal brines associated with the hydrocarbon fields

(Engel et al., 2007).

Figure 2.1: Location of Lower Kane Cave in Wyoming, USA, and schematic diagram of the cave with springs and mat types (Engel et al., 2003). Locations in the cave are numbered in meters from the back of the cave, so the spring sites are located at 118 m (Fissure Spring), 189 m (Upper Spring), and 248 m (Lower Spring).

Microbial Roles in the Sulfur Cycle

All transformations within the sulfur cycle are controlled by microbial activity, and specifically by physiologically distinct oxidation and reduction pathways that are associated with taxonomically diverse groups of Bacteria and Archaea. All metabolic transformations, such as the oxidation of reduced sulfur compounds and dissimilatory sulfate reduction, are controlled by a number of distinct enzyme-coding genes. The oxidation of reduced sulfur compounds by microbes is complex enzymatically, in terms of possible substrates and intermediates that can be oxidized, and also because of the diversity of taxonomic groups. In contrast, dissimilatory sulfate reduction has been linked to two lineages of sulfate-reducing Archaea and five lineages of

Bacteria, the most common and diverse being members of the class Deltaproteobacteria within the phylum Proteobacteria (Muyzer and Stams, 2008).

Sulfur oxidation is primarily performed by Bacteria and Archaea; oxidation associated with Eukarya is performed by endosymbiotic chemolithoautotrophic bacteria (Grieshaber and

Volker, 1998; Friedrich et al., 2005). Although archaeal sulfur oxidation is primarily performed by thermoacidophilic Sulfolobales (Friedrich et al., 2005), oxidation of sulfur species by chemolithoautotrophic and anaerobic phototrophic bacteria is more ecologically diverse

(Friedrich et al., 2001; Friedrich et al., 2005). Phototrophic sulfur bacteria are referred to as either purple sulfur bacteria (PSB), encompassing members of Chromatiaceae and

Ectothiorhodospiraceae families of the class Gammaproteobacteria, or green sulfur bacteria

(GSB), which includes members of the family Chlorobiaceae of phylum Chlorobi. Phototrophic sulfur bacteria grow anaerobically where light and reduced sulfur compounds (e.g., sulfide, thiosulfate, or elemental sulfur) are available, although some exceptions have been documented

(Henshaw et al., 1998; Stout et al., 2006; Ogawa et al., 2008; Azai et al., 2009; Holkenbrink et al., 2010; Ogawa et al., 2010; Sakurai et al., 2010). The function of sulfur-oxidizing bacteria and the genes that control them have been studied extensively from pure cultures in the last decade, but an understanding of the types and distribution of pathways in nature, relative to substrate bioavailability, is still lacking.

The Sulfur Oxidation Multienzyme (Sox) System

The Sox multienzyme system is perhaps the best known oxidation pathway and is ubiquitous among all known thiosulfate-oxidizers, as well as others that are not known to utilize thiosulfate (Meyer et al., 2007). First described from cultures of the facultative chemolithoautotroph Paracoccus pantotrophus (Alphaproteobacteria), the Sox system is 10

versatile in substrate utility and consists of many genes that code for enzymes controlling distinct reactions (Petri et al., 2001; Bardischewsky et al., 2005; Rother et al., 2005; Bagchi and Ghosh,

2006; Meyer et al., 2007). At the most basic level, thiosulfate-oxidizers utilizing the Sox system are categorized into two groups based on whether sulfur globules form as metabolic intermediates (Welte et al., 2009). Other studies suggest different distinctions within the Sox system according to the presence or absence of certain gene-coded protein components because not all organisms capable of oxidizing thiosulfate via the Sox pathway possess the same genetic components of the multienzyme complex (Petri et al., 2001; Meyer et al., 2007; Meyer and

Kuever, 2007). Of the Sox system complexes that have been studied, the dinuclear manganese cluster protein, formed from the expression of the soxB gene (Epel et al., 2005), is most highly conserved and is generally thought to be essential for function (Petri et al., 2001; Meyer et al.,

2007; Ghosh et al., 2009). Phylogenies of soxB genes have been used to characterize the Sox system in a variety of habitats, e.g. agricultural crop soils (Anandham et al., 2008), coastal marine environments (Krishnani et al., 2010a; Krishnani et al., 2010b), acid mine drainage

(Bhowal and Chakraborty, 2011), deep-sea communities (Sievert et al., 2008; Harada et al.,

2009; Yamamoto et al., 2010), and caves (Chen et al., 2009). Adaptations to extreme environments and lateral gene transfer are among the causes for variations in the Sox pathway and distribution of genes among taxa (Meyer et al., 2007).

In decreasing order of reactivity, hydrogen sulfide, elemental sulfur, thiosulfate, and sulfite can be oxidized in the P. pantotrophus-type Sox system (Rother et al., 2001), although the mechanism for thiosulfate oxidation is probably best understood. At least 15 genes are present in the P. pantotrophus-type system (Friedrich et al., 2005), but other genes have been discovered in different organisms more recently (Welte et al., 2009; Ogawa et al., 2010). Four periplasmic 11

complexes are generally responsible for a thiosulfate oxidation in the P. pantotrophus-type system, SoxAX, SoxYZ, SoxB and Sox(CD)2 (Meyer et al., 2007). SoxAX, controlled by soxA and soxX genes, is composed of diheme cytochrome c SoxA and monoheme SoxX proteins.

Together, they oxidatively couple the sulfane sulfur of thiosulfate to SoxYZ (Bagchi and Ghosh,

2005; Meyer et al., 2007; Ogawa et al., 2008). SoxY can, however, covalently bind sulfur of various oxidation states (Friedrich et al., 2005). In the thiosulfate oxidation pathway, after thiosulfate is coupled to the SoxY-cysteine-sulfhydryl group of SoxYZ, SoxB, coded for by the soxB gene, cleaves off the terminal sulfone (Quentmeier et al., 2003; Meyer et al., 2007).

Sox(CD)2 further oxidizes the sulfane sulfur from the remaining SoxY-cysteine persulfide, and the product is hydrolyzed from SoxB as sulfate (Quentmeier et al., 2003; Meyer et al., 2007).

Other components of the complex code for other functions, such as the soxR gene that codes for a repressor protein of the ArsR family (Friedrich et al., 2005), and soxS that is essential for complete expression of Sox genes (Rother et al., 2005; Bagchi and Ghosh, 2006). Organisms with this type of Sox pathway do not form intracellular or extracellular elemental sulfur globules, and instead store sulfur as intermediate metabolic products (Welte et al., 2009). Because thiosulfate is unstable under certain environmental conditions, such as at acidic pH, and the Sox oxidation pathway in extreme chemolithoautotrophic organisms, such as Acidithiobacillus spp.,

Halothiobacillus spp., and Thermobacillus spp., commonly utilize polythionates instead of thiosulfate (Meyer et al., 2007). These types of organisms contain tetrathionate hydrolase and thiosulfate dehydrogenase enzymes unique from other organisms (Meyer et al., 2007).

Sox mechanisms among sulfur-oxidizing bacteria that form intra- and extracellular sulfur globules differ in function from the P. pantotrophus-type pathways because the components of the complex are spatially separated, lack Sox(CD)2, and possess another gene that can oxidize 12

the stored sulfur, dissimilatory sulfite reductase (dsrAB/dsrMKJOP) (Meyer et al., 2007). From microorganisms without the soxC and soxD genes, this Sox pathway may also be related to the reverse-acting enzymes of the cytoplasm sulfate reduction pathway that utilize the reverse dissimilatory sulfite reductase, adenosine-5-phosphosulfate-reductase (APS reductase), ATP sulfurylase, and the sulfite acceptor oxidoreductase (Meyer et al., 2007; Meyer and Kuever,

2007). Deletions of clusters of dissimilatory sulfite reductase or of sulfide:quinone oxidoreductase (sqr) genes in Chlorobaculum tepidum, a type of green sulfur bacteria, prevent or impair formation and consumption of sulfur globules (Holkenbrink et al., 2010).

Adenosine-5’-phosphosulfate (APS) Reductase

The alpha subunit of the adenosine-5-phosphosulfate reductase (aprA) gene is ~390 base

2- pairs in length and controls an energy-yielding step in dissimilatory sulfate reduction (SO4 

H2S) in Deltaproteobacteria. The reverse reaction of this same gene controls the oxidation of

2- reduced sulfur compounds (e.g., H2S  SO4 ) in Gammaproteobacteria (e.g., Thiothrix spp.) and other groups (Woyke et al., 2006; Harada et al., 2009; Hügler et al., 2010), including controlling the reaction of sulfite to adenylylsulfate that is then converted to sulfate by gammaproteobacterial symbionts in deep-sea bivalves (Harada et al., 2009). Other sulfur- oxidizing bacteria actively expressing the reverse reaction are typically associated with strictly anaerobic or at least facultative anaerobic lifestyles, such as a few Chlorobiaceae and most members of the Chromatiaceae families, as well as the genus Thiobacillus within the class

Betaproteobacteria (Meyer and Kuever, 2007). In general, gene sequence studies from environmental samples and complete genome sequencing of novel organisms cultured from sulfidic environments, including deep-sea vent communities (Harada et al., 2009; Hügler et al.,

2010; Yamamoto et al., 2010) and coastal aquaculture (Krishnani et al., 2010a; Krishnani et al., 13

2010b), have contributed to a growing gene database of bacterial Apr and Sox multienzyme systems. Recent studies have focused on identifying the presence or absence of these genes and comparing their diversity and phylogenetic relationships to those retrieved from 16S rRNA gene sequence phylogenies (Kubo et al., 2011). From multiple gene phylogenies, horizontal gene transfer (HGT) events common among bacteria can be constrained (Meyer et al., 2007; Meyer and Kuever, 2007).

Biological Sulfur Speciation in Biological and Geological Materials

2- During the oxidation of reduced sulfur compounds, the formation of sulfate (SO4 )

(Equation 2.1) yields more energy than if elemental sulfur (S0) is formed as an intermediate

(Equation 2.2). In general, with unlimited oxygen, sulfate is the major metabolic product of reduced sulfur oxidation (Equation 2.1) (Khanal, 2008), but biologically produced S0, also known as biosulfur or sulfur globules, can be produced intracellularly or extracellularly as metabolic intermediates. Eventual oxidation of S0 yields energy that, in combination with storage, is comparable to the overall oxidation of sulfide to sulfate (Equation 3).

2- + H2S + 2O2  SO4 + 2H Equation 2.1

0 H2S + ½ O2 S + H2O Equation 2.2

0 2- + S + H2O + ½ O2  SO2 + 2H Equation 2.3

Understanding of the structural properties and compositional diversity of biologically produced sulfur structure is primarily based on in situ measurement of pure cultures. Biosulfur compounds are chemically and structurally distinct from abiotically produced sulfur compounds, such as crystalline orthorhombic sulfur ‘flower’ (Janssen et al., 1999; Kleinjan et al., 2003).

Sulfur globules are considered to be colloids, typically <1.0 µm (Kleinjan et al., 2003), but have also been reported to be as large as 3 µm (George et al., 2008). The primary composition of 14

sulfur globules is as cyclo-octasulfur (S8, or ring-sulfur) or polymeric sulfur (S - sulfur) forms (Prange et al., 1999; Dahl and Prange, 2006; Engel et al., 2007; Prange, 2008).

Biogenic sulfur globules are hydrophilic, attributed to organic end groups and adsorbed organic polymers (Janssen et al., 1999), compared to the hydrophobic nature of crystalline inorganic S0

(Kleinjan et al., 2003). In environmental samples, the formation and consumption of sulfur globules are dependent on substrate availability and microbial community structure. Until recently, studies of sulfur speciation in natural samples had been conducted in sediment samples

(Vairavamurthy et al., 1995; Zhao et al., 2006), but the structure and composition of sulfur species in natural chemolithoautotrophically-sustained communities and associated sediments were lacking (Engel et al., 2007).

Globule formation during sulfide oxidation is dependent on the metabolic pathway utilized by organisms. From bioengineering research using pure cultures, the formation of S0 as a metabolic byproduct will only proceed under high sulfide or oxygen-limiting conditions

(Kuenen, 1975; Khanal, 2008). Oxidation of sulfide without sulfur globule formation has been attributed to the periplasmic Sox pathway. However, organisms may possess genes encoding multiple oxidation pathways; for example, during thiosulfate oxidation Allocromatium vinosum

(phototrophic sulfur Gammaproteobacteria, Chromatiacea) obligately forms intracellular sulfur globules but has SoxXA, SoxYZ, and soxB genes (Grimm et al., 2008). Other anaerobic anoxygenic phototrophs within the families Chromatiacea, Ectothiorhodospiraceae, and

Chlorobiaceae also form obligate sulfur intermediates during thiosulfate oxidation (Meyer et al.,

2007). Aerobic chemolithoautotrophic sulfur-oxidizing bacteria, such as from the genera

Beggiatoa, Thiothrix, Thiobacillus, Thiomicrospira, obligately produce intracellular sulfur globules (Meyer et al., 2007). Conversely, members of the class Epsilonproteobacteria have 15

putative soxB genes for reduced sulfur compounds oxidation using either O2 or nitrate as electron acceptors, but none of the members of this class are known to form intracellular sulfur globules;

Arcobacter spp. can form sulfur extracellularly in filaments (Sievert et al., 2007).

Most XANES studies of sulfur have been done from pure cultures (e.g., Prange et al.,

1999; Dahl and Prange, 2006). But, from a study describing XANES at the S K-edge from natural microbial mat samples in Lower Kane Cave, there is a range of naturally co-occurring sulfur species, as well as specific species that were identified from specific regions of the mats

(Engel et al., 2007). Natural environmental samples highlight the complexity of the sulfur cycle.

When the concentration of sulfate was inversely proportional to the elemental sulfur content in some of the samples (Engel et al., 2007), this suggested that the production of elemental sulfur as globules was occurring in some regions of the mats, but that oxidation of reduced sulfur to sulfate was occurring in other locations. Scanning electron microscopy showed that gypsum

(CaSO4•2H2O) accumulated in biofilms dominated by Epsilonproteobacteria, despite gypsum undersaturation in the surrounding water (Engel et al., 2004b). The results suggested that sulfur- oxidizing microbial communities locally influence aqueous and solid-phase geochemistry, which affects overall cave formation.

The Sulfur Cycle in Karst Systems

In Lower Kane Cave, the oxidation of reduced sulfur compounds promotes the replacement of calcite with gypsum, which rapidly dissolves in the undersaturated cave stream

(Egemeier, 1981). This process, SAS (Equation 2.4), has been described in other caves, such as

H2SO4 + CaCO3 + H2O  CaSO4∙2H2O + CO2 Equation 2.4

Cueva de Villa Luz in Mexico (Hose and Pisarowicz, 1999; Hose et al., 2000), the Frasassi cave system in Italy (Galdenzi and Maruoka, 2003; Galdenzi et al., 2008; Macalady et al., 2008), 16

Movile Cave in Romania (Chen et al., 2009; Puşcaş et al., 2010), and caves in Iraq (Iurkiewicz and Stevanovic, 2010). Even caves that are no longer forming, like the Guadalupe caves, New

Mexico (Jagnow et al., 2000; Polyak and Provencio, 2000), have evidence to support origins involving SAS. The effective rate of carbonate dissolution in phreatic and vadose zones of SAS caves have been reported as an average of ~20 mg/cm2/year (Forti et al., 2002).

Many active SAS caves contain microbial communities living in conditions that would otherwise be considered hostile due to high sulfide concentrations and acidic pH. Biodiversity studies in active SAS caves reveal the presence of sulfur-oxidizers and sulfate-reducers (Chen et al., 2009; Engel et al., 2010; Iurkiewicz and Stevanovic, 2010), which are thought to be the major organisms capable of metabolically influencing local geochemical conditions. The microbial communities in the Frasassi Caves share similar biodiversity with Lower Kane Cave, including Epsilonproteobacteria, Gammaproteobacteria, and Betaproteobacteria (Galdenzi and

Maruoka, 2003; Engel et al., 2010). However, some differences exist, such as the presence of

Bacteroidetes, Chlorobi, Verrucomicrobia, and Planctomycetes in the Frasassi Caves that are less abundant in Lower Kane Cave mats (Galdenzi and Maruoka, 2003; Engel et al., 2010).

Epsilonproteobacteria inhabit a number of sulfidic environments, such as SAS caves and deep- sea vent communities (Forti et al., 2002; Sievert et al., 2008; Chen et al., 2009; Kern and Simon,

2009; Yamamoto et al., 2010). The distribution of different members of Epsilonproteobacteria in

Lower Kane Cave is linked to sulfide and electron acceptor availability (Kern and Simon, 2009).

CHAPTER 3

MATERIALS AND METHODS

Sample Collection and Aqueous Geochemistry

Although historical geochemical and microbiological results were used from previous studies in Lower Kane Cave, additional field sampling to obtain fresh materials as part of this thesis work took place in January 2011. A total of nine biological samples were aseptically collected in triplicate (27 total samples), predominately at the Upper Spring orifice and downstream within the stream channel discharging from the spring (Figure 3.1). Sampling locations in the cave are numbered in meters from the back of the cave. Microbial mat samples were immediately preserved with ethanol, sealed, and placed on ice for transport. Samples were stored at -20o C until use.

Aqueous geochemical conditions were measured at all spring orifices and at least every few meters along the stream that discharged from each orifice. At each location, pH, temperature, and conductivity were measured with standard electrode methods (APHA, 1998), and dissolved sulfide and oxygen concentrations were measured spectophotometrically using

CHEMetrics colorimetric chemistries (APHA, 1998). Water samples were manually filtered to

alkalinity. Alkalinity was determined from end-point seeking titration to pH 4.3 in the field

(APHA, 1998). Based on the pH of the waters (~7.2), alkalinity was considered to be predominately as the bicarbonate species (pKa @ 22o C, 6.32) (APHA, 1998). The cation samples were acid- preserved, and both sets of filtered samples were stored on ice for transport

Figure 3.1 (next page): Images from the Upper Spring area of Lower Kane Cave. (A) Looking downstream and across the Upper Spring orifice pool; white arrow notes the spring orifice. (B) Microbial filaments attached to gray sediment at the bottom of the Upper Spring orifice pool. The image is ~30 cm across, making filaments on the order of a few cm to 10’s of cm long. (C) Looking downstream from the Upper Spring orifice pool at the microbial filaments pictures in B. (D) Looking upstream toward the Upper Spring pool (noted with arrow) and at white filamentous micorbial mats in the stream channel eminating from pool. The stream channel ranges from ~ 0.75 to 3 m across and ~5-10 cm deep. Person in yellow shirt in foreground is at ~203 m distance. (E) Sampling microbial filaments at ~194 m distance along the stream channel, looking downstream from the orifice pool.

A B

Upper Spring orifice

D E

Upper Spring orifice

and kept at 4o C until use. Ion concentrations were measured on a Dionex Ion Chromatography system (ICS)-3000 at Louisiana State University.

XANES Spectroscopy, Spectral Fitting, and Quantitative Analyses

To determine the chemical speciation of sulfur compounds in the microbial mats from

Lower Kane Cave, representative mat samples of each morphology and location were analyzed using synchrotron-based X-ray Absorption Near Edge Structure Spectroscopy (XANES). All sample preparations and XANES measurements were performed at Louisiana State University’s

Center for Advanced Microstructures and Devices (CAMD) in Baton Rouge.

Portions of microbial biomass were spread uniformly on a 1-cm2 piece of sulfur-free filter paper and placed between mylar and self-adhesive kapton® film. From previous experience

(e.g., Prange et al., 1999; Engel et al., 2007), no oxidation artifacts were expected from sample storage or preparation, including from sample preparation in ambient air conditions. Samples were scanned four to six times as freshly prepared specimens, as well as one day after preparation, to identify possible speciation due to sample dehydration or the presence of ethanol.

No significant differences in spectra were observed, but scans of freshly prepared samples were selected for analysis over dry mat scans. Control measurements of the filter, mylar, and kapton® film were also done and used as additional reference spectra.

XANES spectra at the S K-edge were recorded under reduced air pressure or in a helium atmosphere at the DCM beamline at CAMD, which operated at an energy of 1.2 - 1.5 GeV with electron currents between 150 – 300 mA. The synchrotron radiation was monochromatized by a modified Lemmonier-Bonn double-crystal X-ray monochrometer equipped with InSb (111) crystals. Measurements were measured in fluorescence mode by flushing the sample chamber with helium upstream of the first ionization chamber. 21

Spectral energy calibration was done based on the spectrum of zinc sulfate by setting the maximum of the first resonance (white line) to an energy of 2481.44eV, following procedures previously described (Prange et al., 1999). According to the step width, this value is reproducible to ±0.1 eV. Spectra were scanned with step widths varying from 0.5 eV between 2440 and 2468 eV, 0.1 eV between 2468 and 2485 eV, and 0.3 eV between 2485 and 2450 eV with an integration time of 1 second per point. A linear background was determined as described previously (Engel et al., 2007), whereby the Origin Program (Origin Lab Corporation, North

Hampton, MA) was used to correct the spectra from contributions of higher shells and other materials. Spectra were normalized to 2510 eV in SigmaPlot®.

Five reference compounds with characteristic absorption energies were used (Figure 3.2): cyclo-octasulfur (S8), polymeric (Sµ), methiosulfone, cysteic acid, and zinc sulfate. These compounds have been used and described previously (Prange et al., 1999; Engel et al., 2007).

They represent the sulfur atom atomic environment, and include two common elemental sulfur forms (S8 and Sµ) and three oxidized forms typically found in biological systems (Prange et al.,

2002; Engel et al., 2007).

For quantitative analysis, the fitting and plotting package Athena XAS Data Analysis

Software was used (Ravel and Newville, 2005) to produce linear combination fits for each sample scan. Errors for contributions for each of the spectral fittings were < ±9%.

Microbial Mat DNA Extraction

DNA was extracted from biological samples within two weeks of collection using

PowerSoil® DNA Isolation Kits (MO BIO Laboratories, Inc.), following the manufacturer’s instructions. The concentration of extracted DNA was determined by using a NanoDrop ND-

1000 spectrophotometer to compare measured absorbance ratios at 260/280 nm and 260/230 nm 22

Figure 3.2: Sulfur K-edge XANES spectra (normalized and stacked) for standard reference compounds used for linear combination fittings of microbial mat sample spectra. (a) cyclo- octasulfur, (b) polymeric sulfur, (c) methiosulfone, (d) cysteic acid, and (e) zinc sulfate.

wavelengths, and the quality of DNA was determined visually from TBE agarose gel electrophoresis followed by ethidium bromide staining. Extracted DNA was stored in TE buffer at -20o C until use for PCR amplification of functional genes and 454 tag pyrosequencing.

Functional Gene PCR Amplification and Cloning

Specific PCR primer sets and conditions for the amplification of each of the functional genes were selected from the literature, and conditions were developed for newly designed primers for this study (Table 3.1). Between 50 and 75 ng of extracted DNA were used in combination with varying concentrations of components required for reactions using

DNAPerfectTAQ DNA polymerase (5 PRIME Inc., Gaithersburg, MD, USA). Due to weak or incorrect amplification of soxC and soxEF genes to test Lower Kane Cave samples collected previously in 2004, only soxB and aprA amplifications were selected as the foci for this study.

Several degenerate soxB primer sets have been described (Petri et al., 2001), and the primer set selected for this study was soxB693F for base pair sites 693-713 and soxB1446B for 1446-1428 sites (Meyer et al., 2007). These primers were expected to amplify one ~750 bp fragment.

Another forward primer for soxB has been described (soxB432F) and has the potential to amplify soxB genes from a broader diversity of sulfur-oxidizers (Petri et al., 2001; Meyer et al., 2007), but amplifications using this forward primer and the reverse primer (soxB1446B) typically result in multiple fragments ~1000 bp long due to priming mismatches compared to the target DNA sites. Therefore, based on the high levels of potential sequence ambiguities, diversity among retrieved soxB genes in this study was suspected to be less than what is known for the full diversity of bacteria possessing the Sox multienzyme system from whole genome analyses (e.g.,

Meyer et al., 2007).

Table 3.1: Functional gene-targeting primers used in PCR assays in this study.

Primer Sequences (5’-3’) Target References soxB693F ATCGGNCARGCNTTYCCNTA soxB Petri et al. (2001) soxB1446R CATGTCNCCNCCRTGYTG soxEF_Forward1 TGYGCIGGITGYCAYGGIAC soxEF Verte et al. (2002) soxEF_Reverse1 GCIGCRTGCCAYTCRATCA soxC_forward GGCRYBVTKRYWCCGVVWGRC soxC This study soxC_reverse GCCSBYATCHNDGSTMAYWTC aprA-1-FW TGGCAGATCATGATYMAYGG aprA Meyer and Kuever (2007) aprA-5-RV GCGCCAACYGGRCCRTA

PCR reaction concentrations and conditions for soxB gene amplifications were modified after Petri et al. (2001). Three 25 µl reactions were prepared for each sample using 2.5 µl 10x buffer solution, 2 µl dNTPs (10 mM stock), 1.25 µl each primer (20 µm stock), 3.75 µl bovine serum albumin, 0.25 µl TAQ (5 U/µl; 5 PRIME Inc.), and a balance of DI water. An initial denaturation step for 10 min (94⁰ C) was followed by 10 cycles of 30 sec denaturation (94⁰ C),

40 sec annealing (55-60⁰ C, each temperature range was tested separately for each sample after fragment amplification by visually inspecting the optimal product band size from TBE agarose electrophoresis gels), and 30 sec elongation (72⁰ C). The first 10 cycles were followed by 25 additional cycles of 30 sec of denaturation (94⁰ C), 40 sec annealing (47⁰ C), and 30 sec elongation (72⁰ C). A final elongation at 72⁰ C was for 10 min. Products of the correct size were visually observed on ethidium bromide-stained TBE agarose gels, extracted from ethidium bromide-stained TAE agarose gels, and purified for cloning and sequencing (see details below).

Amplification of soxEF genes were performed using soxEF_Forward1 and soxEF_Reverse1 primers (Table 3.1) (Verte et al., 2002; Hall et al., 2008) in two 25 µl replicate reactions composed of 2.5 µl 10x buffer solution, 2.5 µl dNTPs (10 mM stock), 2.5 µl each primer (20 µm stock), 3 µl bovine serum albumin, 0.25 µl TAQ (5 U/ µl; 5 PRIME Inc.), and

balance of DI water. An initial five min denaturation (94⁰ C) step was followed by 30 cycles of

30 sec denaturation (94⁰ C), 60 sec annealing (53⁰ C), and 60 sec elongation (72⁰ C); a final elongation at 72⁰ C for 10 min. Products of the correct size were visually observed on agarose gels followed by ethidium bromide staining and purified for cloning and sequencing (see details below).

Primers used to amplify soxC genes were soxC_forward and soxC_reverse (Table 3.1).

These primers were designed specifically to target Epsilonproteobacteria, based on knowing that this group appears to be underrepresented in soxB functional gene studies (Chen et al., 2009), likely due to high levels of sequence ambiguities. Consensus and degenerate primer target sites conserved across all relevant taxa were determined by aligning 76 soxC-containing sequence regions (DNA and translated amino acids) from candidate genomes, then calculating primer efficiency from, for example, melting temperature, GC content, and the stability of the primer

DNA duplex, using the program CODEHOP (Rose et al., 1998; Rose et al., 2003). Two primers were identified to amplify ~450 bp region of the soxC gene. Reaction concentrations applied for soxC PCR amplification initially followed the protocol for soxB, including a denaturation step for 10 min (95⁰ C) was followed by 20 cycles of 60 sec denaturation (95⁰ C), 60 sec annealing

(60-50⁰ C, “touchdown” for improved primer specificity and sensitivity (Korbie and Mattick,

2008), and 90 sec elongation (72⁰ C); the first 30 cycles were followed by 15 additional cycles of

60 sec denaturation (94⁰ C), 60 sec annealing (50⁰ C), and 90 sec elongation (72⁰ C). A final elongation (72⁰ C) step was done for 10 min. Products of the correct size were visually observed on TBE agarose gels followed by ethidium bromide staining, and products were purified for cloning (see section below).

Amplification of aprA genes was performed using the primer set aprA-1-FW and aprA-5-

RV and PCR conditions previously described (Meyer et al., 2007; Meyer and Kuever, 2007).

Products of the correct size were visually observed on TBE agarose gels followed by ethidium bromide staining, and purified for cloning and sequencing (see details below).

Replicate PCR amplification products for each sample to be analyzed were combined and purified on TAE agarose electrophoresis gels (after visual inspection on TBE agarose gels).

Products of the correct base pair size were cut out and purified from the gels using the Wizard®

PCR Preps DNA Purification System (Promega Corporation, Madison, WI, USA), following manufacturer’s instructions. Purified PCR products were cloned using the Original TA Cloning© kit (Invitrogen™, Carlsbad, CA, USA), following manufacturer’s instructions. Positive clone colonies were individually picked and separated into TE buffer, and cells were lysed at 95oC for

10 min. Unpurified clone DNA was amplified using the Invitrogen M13 (8F and 1510R) primers in preparation for Sanger sequencing on ABI3730xl capillary sequencers by the High-

Throughput Genomics Unit© at the University of Washington, Seattle, Washington

(http://www.htseq.org/). Cloned products for the soxEF and soxC genes were sequenced from the forward and reverse positions. The aprA and most of the soxB genes were sequenced from the forward position only, depending on the coverage.

Functional gene sequences were first inspected for quality using the ABI sequencer- specific software. Poor quality reads or sequences without accurate gene amplification based on chromatograph reads were removed from the dataset prior to further analysis. Resulting functional gene sequences were (if necessary) assembled in Contig Express, a computation component of the Vector NTI Advance 10.3.0 software (Invitrogen Corporation, Carlsbad, CA).

Functional Gene Sequence Analysis

Sequences were compared to GenBank sequences using the National Center for

Biotechnology Information (NCBI) (http://www.ncbi.nlm.nih.gov/genbank/) Basic Local

Alignment Search Tool (BLAST) to identify closest relatives of cultured and environmental uncultured microbes (http://blast.ncbi.nlm.nih.gov/Blast.cgi). Sequences were initially aligned with Muscle (©European Bioinformatics Institute 2012) and manually edited in BioEdit

Sequence Alignment Editor v7.1.3 (Hall, 1999). The computer program mothur (ver. 1.22) was used for clustering and calculating operational taxonomic unit (OTU) assignments, according to methods previously described (Schloss et al., 2011). As a caveat, assigning OTUs from raw functional gene DNA sequences likely overestimates the taxonomic diversity of functional genes because the differences among DNA sequences are conserved and may not result in changes at the level of the translated amino acid sequences or even to variation in protein function.

Pyrosequencing of 16S rRNA Genes

Although functional soxB and aprA are among commonly reported functional genes describing sulfur oxidation pathways in natural, environmental microbial communities or in pure cultures manipulated in the laboratory or in bioengineering applications (Meyer and Kuever,

2007; Anandham et al., 2008; Harada et al., 2009; Hügler et al., 2010; Krishnani et al., 2010a;

Luo et al., 2011), much remains to be determined with respect to the rate of occurrence and expression of these genes. Relative to rRNA genes that are more commonly sequenced to describe phylogenetic diversity (Ludwig and Schleifer, 1994), the practice of functional gene sequencing to describe diversity and apparent function is less common. Consequently, there are relatively few studies to which results from this study could be compared. Also, because of the highly diverse taxonomic representation for functional gene homologs, the probability of 28

identifying the taxonomic group to which a sequence belongs at a high sequence similarity and confidence level is fairly low. Therefore, an understanding of the full microbial diversity from the microbial mats was needed to understand how representative functional gene taxonomies were to overall mat diversity. A type of next-generation sequencing, 454 tag pyrosequencing, of

16S rRNA genes from the mats was done. This method rapidly and cost-effectively produces a high number of short gene sequences (Ahmadian et al., 2006; Amend et al., 2010; Kumar et al.,

2011; Sun et al., 2011). These data were used in conjunction with the historical bacterial diversity results obtained from Lower Kane Cave (Engel et al., 2004b; Engel et al., 2010).

For the 454 tag pyrosequencing, DNA extracted from the microbial mats was sent unpurified to the Research and Testing Laboratory (RTL) LLC (Lubbock, TX, USA) for 454 pyrosequencing using a Roche 454 FLX genome sequencer system. Amplifications were done after purification by RTL of the V1-V3 hypervariable region of 16S rRNA of E. coli

(Sun et al., 2011), whereby variations and similarities in the resulting amplicons represented distinct taxonomic groups as OTUs. Raw pyrosequence reads were trimmed to a minimum length of greater than or equal to 170 bp after removing primers, and were evaluated with the maximum edit distance for forward primers to 2, the number of ambiguous codons (N’s) set to 0, and the minimum quality score for each pyrosequence set to 20. Quality reads were aligned with the Ribosomal Database Project (RDP) database (http://rdp.cme.msu.edu) (Cole et al., 2009), and taxonomic classifications were performed after filtering, screening, and removing pyrosequences that were considered to be chimera. For taxonomic analyses, pyrosequence reads were compared to the reference database of known 16S rRNA genes and classified according to the RDP

Classifier using the computer program mothur (version 1.22) (Schloss et al., 2009). OTU classification and clustering were done using methods previously described (Schloss et al., 2011; 29

Sun et al., 2011). mothur was also used to determine the depth of sequence sampling based on rarefaction curves, and OTU richness indicators, including Shannon Index (H’) and Simpson’s

Index (D) for describing diversity and evenness, and Chao 1 index.

Statistical Analyses

Hypothesis testing was based on the statistical significance of nonparametric correlations between geochemical data and gene diversity using PAST v. 2.14 (Hammer et al., 2001)

(http://folk.uio.no/ohammer/past) and an online statistics calculator (Wessa, 2008), and from

Canonical Correspondence Analysis (CCA) for multivariate correlations of the normalized functional gene OTU distribution to major environmental gradients, also using PAST.

To assess confidence in the classification of functional genes based on GenBank sequence relatives, normalized abundance of taxonomic classifications of functional genes and

16S rRNA gene sequences were correlated to environmental gradients individually using

Kendall’s rank correlation coefficient tau (Kendall’s tau,) values and collectively using CCA.

Kendall’s τ values, where -1< τ <1, were calculated in PAST correlation tables, and statistical significance was calculated as a two-sided P-value (Hill, 1973; Best and Gipps, 1974; Kendall,

1976; Valz and Thompson, 1994; Davidson and Hinkley, 1997). Kendall’s τ equal to 0 or 1 indicate variable independence or perfect dependence, respectively (Kendall, 1976). Correlation coefficients were considered statistically significant at P < 0.05. The normalized abundance of broadly classified functional gene sequences distributed among samples was correlated with multivariate geochemical gradients, as well as the normalized abundance of taxa of 16S rRNA gene sequences for CCA. Strength of a relationship between gene distribution and major geochemical gradients were interpreted based on plotted proximity of OTUs (i.e. scatter points), environmental gradients (i.e. vectors that point in the direction of maximal change, and the 30

angles between environmental gradient vectors and the ordination axes that explain the amount of variation observed among sites (i.e. sampling units) (Palmer, 1993; Ramette, 2007) .

Statistical significance for each ordination axis was given by performing P tests based on 999

Monte Carlo permutations, and P-values <0.05 was considered significant. OTUs distributed in

CCA plots were interpreted based on taxonomic assignments of closest functional gene sequence relatives, supported by the significance of Kendall’s τ for major represented taxa correlations to

16S rRNA genes, and interpreted gene function for broad taxonomic groups. Multiple CCA plots were used to demonstrate geochemical and OTU distribution relationships to multiple axes, despite statistical significance of the axes, if nonrandom OTU distributions and associated taxa among sites were observed.

CHAPTER 4 RESULTS Upper Spring Microbial Mat Morphology and Aqueous Geochemistry

Sulfidic water discharges at the Upper Spring’s orifice, which is 189 m from the back of the cave, and collects in a large pool several meters in diameter (Figure 3.1). The Upper Spring pool is visibly lined with a thick, nearly continuous gray microbial mat, above which long white filaments form and are suspended in the water column (Figure 3.1A, B, and C). At ~191 m, the pool narrows into a shallow stream where gray mats become less visible and the density of white filamentous biomass increases (Figure 3.1C, E). Samples of three distinct mat types, long white filaments (denoted as sample LKC_11_191_1-3), gray mats (sample LKC_11_191_4-6), and short white filaments (sample LKC_11_191_7-9), were collected at this location (Figure 3.1A,

B). White filaments collected at 194 m (sample LKC_11_194_1-3) (Figure 3.1E) and 196 m

(sample LKC_11_196_1-3) (Figure 3.1D) were generally shorter in length and populated shallow sediments in the narrow stream. At 198 m, the stream widens into a larger, shallow area that is heavily populated by white filamentous webs (sample LKC_11_198_1-3). Knobby white filamentous mats and bundles of filaments, referred to as tufts or ‘fluffies,’ were collected at

~201 m (sample LKC_11_201_1-3). Collected biomass at 203 m (sample LKC_11_203_1-3) was comprised of white webs and tufts in a yellow gel-like matrix (Figure 3.1D).

Field and laboratory IC measurements were done every 4 to 5 meters to coincide with microbiological sample sites (Table 4.1). Measurements corresponded to historical data

(Appendix A). Water temperature (~22.1⁰ C) and pH (~7.2) were relatively constant along the flowpath. Lower Spring (248 m) was slightly warmer and pH was slightly lower than Upper

Spring. Dissolved oxygen and sulfide concentrations were measured in the field every

Table 4.1: Geochemical results from Lower Kane Cave, listed in meters from the back of the cave to the front of the cave. Fissure Spring (128 m), Upper Spring (189, 194, 198, 203 m), and Lower Spring (248 m) (Figure 2.1) All ion concentrations are in mmol/L, except temperature (⁰C) and pH (standard units). BDL = below detection limit.

Parameters 128 189 194 198 203 248 Temperature (⁰C) 21.4 22.1 22.1 22.1 20 23.7 pH 6.94 7.24 7.25 7.16 7.2 6.8 - HCO3 3.41 3.25 3.33 3.09 3.58 3.10 2- SO4 1.36 1.21 1.20 1.20 1.21 1.23 Cl- 0.11 0.10 0.10 0.10 0.17 0.10 - NO3 BDL BDL BDL BDL BDL 0.002 Ca2+ 2.05 1.88 1.91 1.89 1.90 1.92 Mg2+ 1.18 1.21 1.18 1.17 1.16 1.16 Na+ 0.42 0.49 0.41 0.40 0.40 0.40 K+ 0.04 0.04 0.04 0.04 0.04 0.04

Figure 4.1: Dissolved oxygen and sulfide concentrations from Upper Spring, shown in mmol/L.

meter from 189 to 208 m (Figure 4.1). Sulfide concentrations were highest (~20 µM) nearest the orifice and decreased with increasing distance. Oxygen concentrations increased along the

flowpath. The observed rates of sulfide loss and oxygen increase were similar to historical modeling results (Engel et al., 2004a; Engel, 2010).

Sulfur Speciation in the Microbial Mats

The chemical speciation of sulfur in the microbial mats differed by location and mat type

(Table 4.2; Figure 4.2). All Upper Spring mats were closely fit with reference spectra, although the fits generated for the Lower Spring orifice long white filaments (248 m) indicate further analysis with additional references may be necessary. As was previously noted by Engel et al.

(2007) for microbial mats from Lower Kane Cave, iron-sulfur minerals, such as pyrite, were absent from the mats analyzed because none of the spectra had lower energies denoting a sulfide

S K-edge white line. All of the samples had between 91 - 100% elemental sulfur as S8 (cyclo- octasulfur or ring sulfur) + Sµ (polymeric or chain sulfur), with low (<9%, or within the

2- analytical error) relative contributions of sulfate, methionine sulfone (e.g., C-S-C or R-SO3

- bonds), and cysteic acid (e.g., R-SO3 bonds).

Table 4.2: Percentages of chemical sulfur species contribution based on the best fittings of the XANES analyses of microbial mat samples. Error is within ±9%

Sample Description Cyclo- Polymeric Sulfate Cysteic Methionine R-factor (m) octasulfur Sulfur Acid Sulfone 191_3 Long white filaments 85 15 0 0 0 0.0004 191_6 Gray mats 76 15 7 0 2 0.008 191_8 Short white filaments 64 34 1 2 0 0.001 194 White filaments 57 37 3 2 0 0.004 196 White filaments 69 26 3 2 0 0.0008 198 White filaments 90 5 3 2 0 0.0005 201 White knobby or filaments 49 48 2 0 0 0.003 203 White webs and yellow gel 63 36 0 0 0 0.0008

Figure 4.2: Upper Spring microbial mat XANES spectra. Samples collected in 2011 from within Upper Spring of Lower Kane Cave for (A) 191 m long white filaments, 191 L, (B) 191 m gray mats, 191 G, (C) 191 m short white filaments, 191 S, (D) 194 m white filaments, (E) 196 m white filaments, (F) 198 m white filaments (G) 201 m white knobby filamentous bundles, (H) 203 m white tufts in a yellow gel-like matrix.

Functional Gene Diversity and Distribution

soxB

PCR amplification of environmental DNA using the soxB693F and soxB1446B primers generally produced a single clean product size of ~750 bp when viewed on electrophoresis gels.

Amplification products were not obtained for the gray mats at 191 m or short filaments at 191 m or 196 m, no matter the conditions or optimization of the PCR reactions. However, the other samples (191L, 194, 198, 201, and 203) yielded a total of 242 soxB clones from the Upper

Spring mats that were sequenced and compared to other sequences in GenBank using BLAST

(Table 4.3). Rarefaction curves indicate coverage of functional gene diversity would have increased with the addition of sequences to the dataset for both soxB and aprA (Figure 4.3).

Primers designed for soxC did not produce clear PCR products of the correct size for any samples, except for LKC_04_198, and soxEF primers similarly resulted in weak amplifications for the mat samples. Multiple PCR product sizes were generated with soxC primers, and only the anticipated band size was excised and purified for cloning. Cloning and sequencing of the

LKC_04_198 soxC products produced few sequences, all of which were weakly related to soxB genes according to comparisons with the BLAST database. soxEF primers successfully amplified products from all Edwards Aquifer samples and one sample from Jack Island; the soxEF sequences were most closely related to environmental clones deposited in the BLAST database that were also similar to those identified from the soxB sequences.

The relative frequencies of taxa among GenBank sequences most closely related to

Lower Kane Cave mat soxB sequences, weighted by BLAST identity scores, were determined. In general, ~80% of the Lower Kane Cave soxB gene sequence homologs were closely related to other environmental samples, and so interpretations of phyla were initially based only on 36

previously reported phylogenetic associations. Homolog sequence similarity in overlapping coverage regions with the BLAST closest relatives ranged from 95% homology to 78% with the most sequences having maximum similarity of <90% identity (Table 4.3). These results imply that there may be some ambiguity in assigning taxonomic affiliations to the some gene OTUs, particularly those with low homology. Therefore, the taxonomy of some OTUs will be referred to herein as “-like sequences.”

OTUs were assigned identities based on homology to sequences deposited in GenBank

(Table 4.3). The most frequently encountered closest relative to the Lower Kane Cave sequences was to the environmental clone MCSoxBA02 from Movile Cave (Table 4.3), interpreted by

Chen et al. (2009) to be from a Thiothrix-like species within the class Gammaproteobacteria.

Diversity among the soxB genes based on the closest BLAST relatives and phylogenetic interpretations was lowest at 191 m (with 98% of the soxB clones being related to

Gammaproteobacteria-like sequences) and highest at 203 m (with 58% of the soxB clones being related to Gammaproteobacteria-like sequences, 38% to Betaproteobacteria-like sequences, and

4% to Alphaproteobacteria-like sequences) (Table 4.3). Of all the soxB raw DNA sequences, 45

OTUs could be assigned, of which 26 OTUs represented Gammaproteobacteria-like sequences,

16 Betaproteobacteria-like OTUs, and one Alphaproteobacteria-like OTU. However, based on the rarefaction curve for soxB OTUs (Figure 4.3), OTU coverage was not saturated and additional clones would need to be sequenced to increase represented diversity.

In Figure 4.4, OTU group “β1” shared 90-92% similarity with a Movile Cave environmental clone MCSoxBB09 (Chen et al., 2009), 83% similarity with Sulfuritalea hydrogenivorans, 80-82% similarity with Thiobacillus denitrificans, and 81% similarity with

Thiobacillus thioparus, the last three organisms being cultured representatives. The OTUs named 37

“β*” and “β**” on Figure 4.4 represent the betaproteobacterial sequence with the highest similarity (although 76-80% is considered weak) to the Burkholderiales Leptothrix cholodnii.

Sequences in these OTUs also shared 80% similarity with a soxB sequence from cultured

Sphaerotilus natans ‘subspecies sulfidovorans,’ also within the Burkholderiales and 82% homology to an uncultured bacterium from a sulfide-removing bioreactor community (Luo et al.,

2011). The betaproteobacterial group “β2” (Figure 4.4) had ~84-90% similarity to the Movile

Cave environmental clone MCSoxBB09. The soxB OTU groups affiliated with the

Gammaproteobacteria “ɣ1” and “ɣ2” were homologous to Thiothrix sp. 12730, but ɣ1 had higher homology (89-95%) to the environmental (Thiothrix-like) clone MCsoxBA02 from Movile Cave and ɣ2 had lower homology (83-92%) (Table 4.3). In general, Thiobacillus-like soxB sequences occurred primarily downstream where sulfide concentrations were lower but oxygen was higher compared to upstream where there was almost no oxygen and the highest dissolved sulfide concentrations (Figure 4.1).

aprA

All mat types and locations in yielded successful aprA gene amplifications, with the typical aprA product size being either 395 or 404 bp, although retrieved product ranges were 165 to 428 bp. In total, 368 aprA gene sequences were obtained (Table 4.4). aprA representation of taxa was higher than the diversity retrieved from soxB gene comparisons to GenBank, mostly because gene homologies for sulfur-oxidizing and sulfate-reducing or sulfur, thiosulfate, or sulfite disproportionating bacterial groups were represented. The closest relatives with the highest homologies to the retrieved aprA sequences ranged from 97% to Thiothrix nivea

(Gammaproteobacteria) to 76% to Desulfarculus baarsii (Deltaproteobacteria). The three proteobacterial classes with the highest frequency among aprA matches were the 38

Table 4.3: Lower Kane Cave soxB sequence diversity, as number of sequences with maximum shared identity to GenBank sequences with taxonomic affiliations. 191L, G, S = Long white filaments, gray mats, and short white filaments from 191 m location, respectively.

Phylogenetic NCBI Closest Relative from Identity (%) 191L 194 198 201 203 Association Accession No. BLAST database Proteobacteria Gammaproteobacteria FJ604832.1 MCSoxBA02 uncultured bacterium clone 89-95 50 26 41 3 21 FJ604828.1 MCSoxBC02 uncultured bacterium clone 83-92 3 6 6 1 1 FJ604831.1 MCSoxBA01 uncultured bacterium clone 90-94 1 12 EF618608.1 Thiothrix sp. 12730 soxB 83-89 13 20 1 Betaproteobacteria FJ604846.1 MCSoxBB09 uncultured bacterium clone 85-86 3 13

AB552844.1 Sulfuritalea hydrogenivorans soxB 82-84 2 14 CP001013.1 Leptothrix cholodnii SP-6, complete 80 1 genome CP000245.1 Ramlibacter tataouinensis TTB310, 78 1 complete genome AP009385.1 Burkholderia multivorans complete 76 1 genome Alphaproteobacteria AY005800.2 Rhodovulum sulfidophilum sulfur 82 2 compounds oxidation operon and soxR soxS Cyanobacteria CP002199.1 Cyanothece sp. PCC 7822 plasmid 79 1

Total soxB Sequences per Sample = 55 48 50 50 40

100

Number of OTUs (96% Similarity) OTUs of Number

0 0 100 200 300 400 500 Number of Sequences soxB aprA

Figure 4.3: Rarefaction curves for functional genes, based on OTUs defined from sequences.

25 = Sulfuritalea hydrogenivorans 20 = MCSoxBB09 = Thiothrix sp. 12730 15 = MCSoxBA02 = MCSoxBC02 = MCSoxBA01 10

Number of of sequences Number 0 α β① ɣ❶ ɣ❷

ɣ❸ β②

Figure 4.4: Distribution of soxB OTUs among phyla in Upper Spring white filaments and webs. The horizontal axis represents OTUs (at 96% similarity) that were grouped by taxonomic associations, including Alphaproteobacteria (α), Betaproteobacteria (β), and Gammaproteobacteria (γ), based on the closest GenBank sequence homologs. See text for notation explanations.

Gammaproteobacteria (52% of all aprA sequences), Betaproteobacteria (13% of all aprA sequences), and Deltaproteobacteria (10% of all aprA sequences). Many of the cultured

Deltaproteobacteria that had closest relatives to Lower Kane Cave aprA genes, such as

Desulfocapsa thiozymogenes (Janssen et al., 1996) and Desulfvibrio fructosovorans (Tsu et al.,

1998), have the ability to disproportionate sulfur, thiosulfate, or sulfite. Additionally, 20% of total aprA sequences from Lower Kane Cave were also related to uncultured and unclassified bacterial clones (Table 4.4). Two retrieved sequences were homologous to Firmicutes.

aprA homologies to Deltaproteobacteria were primarily restricted to the samples nearest the Upper Spring orifice, and none of the sequences from 198 through 203 m samples were related to previously cultured organisms (Table 4.4). The gray mats aprA sequences had the greatest abundance and diversity among Deltaproteobacteria. White filaments from 191 m, 194 m, and 196 m also had some sequences homologous to Deltaproteobacteria (Table 4.4). In contrast, long white filaments had more relatives affiliated with the Gammaproteobacteria, primarily to Thiothrix spp., which oxidize thiosulfate (Rosetti et al., 2003). Other aprA sequences homologous to common sulfur-oxidizers included betaproteobacterial Thiobacillus spp. from 191 m short filaments, 198 m white filaments, and 203 m white filamentous webs. The distribution of aprA-betaproteobacterial sequences reflected the distribution of soxB- betaproteobacterial sequences that were only retrieved from the oxidized portion of the mats

(Table 4.4, Figure 4.5).

A total of 78 OTUs were defined for aprA sequences. Based on the rarefaction curve for aprA OTUs versus the number of sequences obtained (Figure 4.3), coverage was not reached, and the addition of new sequences would potentially yield additional underrepresented diversity.

Sequence similarities to GenBank aprA sequences were not sufficiently high to differentiate the 42

Table 4.4: Lower Kane Cave aprA sequence diversity, as number of sequences with maximum shared identity to GenBank sequences with and without taxonomic affiliations. * denotes a sequence matching a complete genome. 191L, G, S = Long white filaments, gray mats, and short white filaments from 191 m location, respectively. Phylogenetic NCBI Closest Relative from Identity 191 191 191 194 196 198 201 203 Association Accession No. BLAST Database (%) L G S Proteobacteria Gammaproteobacteria EF641919.1 Thiothrix nivea 88-97 25 3 4 28 4 40 31 43 EF641918.1 Thiothrix sp. 92 1 1 1 AM228902.1 Gammaproteobacteria 88 5 4 EU864035.2 Gammaproteobacteria symbionts of Robbea sp. 88 5 FM879002.1 Uncultured Gammaproteobacteria 84-85 12 Betaproteobacteria EF641924.1 Thiobacillus denitrificans 87-89 7 15 DQ825796.1 Uncultured Thiobacillus sp. 87-94 1 23 CP000572.1 Burkholderia pseudomallei* 77 1 Deltaproteobacteria EF442938.1 Desulfocapsa thiozymogenes 89-93 2 1 3 15 AF418166.1 D. thiozymogenes 88-90 2 AF418109.1 Desulfovibrio fructosovorans 88 4 AF418146.1 Desulfobulbus elongatus 88 1 5 FR695872.1 Uncultured Desulfobacterium sp. 89 1 1 CP002085.1 Desulfarculus baarsii* 76 1 Firmicutes EF442956.1 Desulfotomaculum australicum* 82 1 CP000860.1 Candidatus Desulfordudis audaxviator* 79 1 Unknown EU156164.1 Uncultured prokaryote clone DH_R123 90-92 3 1 7 3 CT025836.2 Uncultured sulfate-reducing bacterium 86 1 1 GU472445.1 Uncultured bacterium 88 2 1 FJ389351.1 Uncultured bacterium 87-88 1 1 GU472447.1 Uncultured bacterium 89-96 1 2 3 HQ191165.1 Uncultured bacterium 86 1 1 FJ468527.1 Uncultured bacterium 86 3 8 FJ468542.1 Uncultured bacterium 84 1 FJ389344.1 Uncultured bacterium 89 1 EU722732.1 Uncultured bacterium 80 1 GU472463.1 Uncultured bacterium 83 2 GU472543.1 Uncultured bacterium 96 1 GU472456.1 Uncultured bacterium 96 2 GU472553.1 Uncultured bacterium 80 1 GU472512.1 Uncultured bacterium 82 1 GU197403.1 Uncultured bacterium 87 2 AB645550.1 Uncultured bacterium 91 18 HM461006.1 Uncultured bacterium 89 1 EF551642.1 Uncultured bacterium 87 2 1 Total aprA sequences per sample = 39 19 42 59 43 57 34 75 43

= ɣ-proteobacteria symbiont (*) = Thiobacillus denitrificans = Thiothrix nivea = Desulfobulbus elongatus = Desulforudis audaxviator* (F) = Desulfocapsa thiozymogenes =Uncultured =Desulfovibrio fructovorans =Desulfotomaculum australicum 20 =Desulfarculus baarsii

0 Number ofsequences Number

ɣ❶ ɣ❷

β① δ ɣ❸ β②

Figure 4.5: Distribution of aprA OTUs among phyla in Upper Spring white filaments and webs. See text for notation explanation.

1000

800

600

400

Number of OTUs Number 200

0 0 1000 2000 3000 4000 5000 Number of Sequences

191 L 191 G 191 S 194 196 198 201 203 248

Figure 4.6: Rarefaction curves for 16S rRNA gene sequences based on 96% sequence similarity.

taxonomy among the different OTU groups compared to the soxB homologies, partly because of the relatively poor taxonomic resolution for non-deltaproteobacterial groups. But, three OTU groupings were defined for the Gammaproteobacteria (“ɣ1, ɣ2, ɣ3”), two for Betaproteobacteria

(“β1, β2”), and Deltaproteobacteria were grouped together with the exception of an OTU indicated with closest relative to the Burkholderia pseudomallei complete genome (Figure 4.5).

Diversity of 16S rRNA Gene Sequences

To determine if the taxonomic representation retrieved from the functional gene sequences was representative of the total microbial diversity for bacteria capable of putative sulfur oxidation or sulfate reduction pathways, 454 tag pyrosequencing of 16S rRNA genes was performed for Upper Spring mat samples (191-203 m) and one Lower Spring mat sample (248 m). Pyrosequencing of Upper Spring and Lower Spring samples resulted in libraries with a total of 99,468 sequences, 35,299 of which were unique. Bacteria were dominant and Archaea represented less than 1% of all sequences in all samples. Proteobacteria, Bacteroidetes,

Chloroflexi, and Firmicutes were the most abundantly represented phyla (Figure 4.7, Appendix

B). Based on the rarefaction curves for OTUs from the 16S rRNA genes (Figure 4.6), coverage was generally high for more than half of the samples; steeper curves indicated that some samples

(e.g., 191G, thin brown solid line) needed to have additional sequencing because OTU diversity was high.

All five classes of Proteobacteria were retrieved from all mat types, but the ratios of the classes compared to each other differed (Table 4.5). Although the Epsilonproteobacteria were not represented in either the soxB or aprA functional gene libraries, members of this class were the most abundant organisms represented in the 454 pyrotag data, especially from upstream white filaments (191L, 191S, 194, 196) (Figure 4.7). Gammaproteobacteria accounted for <10% of the sample diversity where Epsilonproteobacteria were dominant, but the ratios of the relative abundance of Epsilonproteobacteria to Gammaproteobacteria approached 1:1 in 198 m white filaments. Sample 191 G of gray mats contained the highest abundance of Deltaproteobacteria, which were predominately related to sulfate-reducers. Deltaproteobacterial diversity was high overall, with the number of represented orders ranging from 5-9 among mat types. The 46

100% ɣ-proteobacteria+*

Ԑ-proteobacteria

δ-proteobacteria *

β-proteobacteria +*

α-proteobacteria +

Total Sequences Total of of Firmicutes *

Chloroflexi Composition Composition

Bacteroidetes

0% Other 191 L 191 G 191 S 194 196 198 201 203

Figure 4.7: Microbial community diversity as normalized abundance of 16S rRNA gene pyrosequences within the orders of major phyla or classes. Orders represented by alternating shades within taxonomic groups representing >10% total diversity in at least one mat sample. + = taxa represented among soxB sequence relatives (Table 4.3). * = taxa represented among aprA sequence relatives.

Table 4.4: 191L, 191G, 191S = long white filaments, gray mats, and short white filaments, respectively, from 191 m sample location.

Phylogenetic NCBI Closest Relative from Identity 191 191 191 194 196 198 201 203 Association Accession No. BLAST Database (%) L G S Proteobacteria Gammaproteobacteria EF641919.1 Thiothrix nivea 88-97 25 3 4 28 4 40 31 43 EF641918.1 Thiothrix sp. 92 1 1 1 AM228902.1 Gammaproteobacteria 88 5 4 EU864035.2 Gammaproteobacteria symbionts of Robbea sp. 88 5 FM879002.1 Uncultured Gammaproteobacteria 84-85 12 Betaproteobacteria EF641924.1 Thiobacillus denitrificans 87-89 7 15 DQ825796.1 Uncultured Thiobacillus sp. 87-94 1 23 CP000572.1 Burkholderia pseudomallei* 77 1 Deltaproteobacteria EF442938.1 Desulfocapsa thiozymogenes 89-93 2 1 3 15 AF418166.1 D. thiozymogenes 88-90 2 AF418109.1 Desulfovibrio fructosovorans 88 4 AF418146.1 Desulfobulbus elongatus 88 1 5 FR695872.1 Uncultured Desulfobacterium sp. 89 1 1 CP002085.1 Desulfarculus baarsii* 76 1 Firmicutes EF442956.1 Desulfotomaculum australicum* 82 1 CP000860.1 Candidatus Desulfordudis audaxviator* 79 1 Unknown EU156164.1 Uncultured prokaryote clone DH_R123 90-92 3 1 7 3 CT025836.2 Uncultured sulfate-reducing bacteriaum 86 1 1 GU472445.1 Uncultured bacterium 88 2 1 FJ389351.1 Uncultured bacterium 87-88 1 1 GU472447.1 Uncultured bacterium 89-96 1 2 3 HQ191165.1 Uncultured bacterium 86 1 1 FJ468527.1 Uncultured bacterium 86 3 8 FJ468542.1 Uncultured bacterium 84 1 FJ389344.1 Uncultured bacterium 89 1 EU722732.1 Uncultured bacterium 80 1 GU472463.1 Uncultured bacterium 83 2 GU472543.1 Uncultured bacterium 96 1 GU472456.1 Uncultured bacterium 96 2 GU472553.1 Uncultured bacterium 80 1 GU472512.1 Uncultured bacterium 82 1 GU197403.1 Uncultured bacterium 87 2 AB645550.1 Uncultured bacterium 91 18 HM461006.1 Uncultured bacterium 89 1 EF551642.1 Uncultured bacterium 87 2 1 Total aprA sequences per sample = 39 19 42 59 43 57 34 75

dominance of Epsilonproteobacteria, Gammaproteobacteria, Betaproteobacteria and

Deltaproteobacteria was similarly observed in a previous study of Lower Kane Cave microbial communities based on Sanger sequencing of 16S rRNA genes (Engel et al., 2003; Engel et al.,

2004a).

16S rRNA sequences classified as non-proteobacterial phyla included Bacteroidetes,

Chloroflexi, and Firmicutes (Figure 4.7). At 198 m, Bacteroidetes were approximately equal to the abundance of Proteobacteria. Bacteroidetes were generally more abundant in downstream communities (198-203 m) than upstream (191-196 m). Chloroflexi and Firmicutes were less abundant in all samples compared to Proteobacteria, and only represented >10% of the total sequences in gray mats (191 m) and knobby white filamentous webs (201 m). Phyla represented by <10% of sequences included Acidobacteria (61 sequences), Actinobacteria (34), Aquificae

(80), Caldiserica (216), Chlamydiae (12), Chrysiogenetes (21), Deferribacteres (297),

Deinococcus-thermus (18), Fibrobacteres (1), Fusobacteria (9), Gemmatinmonadetes (13),

Lentisphaerae (449), Nitrospira (52), Planctomycetes (604), Spirochaetes (300), Synergistetes

(152), Tenericutes (43), Thermodesulfobacteria (43), Thermotogae (9), Verrucomicrobia (576),

Cyanobacteria (6), and unclassified phyla (575 sequences among 7 unclassified phyla).

Interestingly, also representing <10% of mat communities from these results, Chlorobi, photoautotrophic sulfide-oxidizers that can also grow chemolithoautotrophically in the absence of light (Henshaw et al., 1998), have also been previously noted in Lower Kane Cave microbial mats (Engel et al., 2010).

Table 4.5: Distribution of 16S rRNA pyrosequences among classes of Proteobacteria.

191 L 191 G 191 S 194 196 198 201 203 Proteobacteria 4772 1866 4242 5029 3114 2116 1983 3139 Gammaproteobacteria 240 86 97 97 78 666 315 911 Alteromonadales 1 1 Acidithiobacillales 1 1 Aeromonadales 1 Cardiobacteriales 2 7 3 Chromatiales 1 33 13 7 17 43 10 86 Enterobacteriales 1 Gamma-sedis 7 5 24 3 27 17 64 Legionellales 31 Methylococcales 2 1 1 1 8 8 Oceanospiralles 3 22 1 10 5 128 Pseudomonadales 1 1 12 3 1 11 5 Thiotrichales 229 37 18 87 50 563 275 530 (Thiothrix spp.) 228 27 13 82 41 554 273 459 Xanthomonadales 3 2 1 4 58

Betaproteobacteria 2 108 214 24 39 87 281 604 Burkholderiales 1 41 168 6 19 37 75 329 Hydrogenophilales 10 2 1 4 4 12 (Thiobacillus spp.) 9 4 3 11 Neisseriales 2 6 1 2 1 2 7 Methylophilales 10 Rhodocyclales 1 53 38 16 18 45 200 256

Deltaproteobacteria 93 1242 446 278 480 644 429 933

Alphaproteobacteria 24 25 31 4 29 84 250 527

Epsilonproteobacteria 4413 405 3454 4626 2488 635 707 164 Total in 16S Library 5301 4876 5954 5643 4589 4543 3990 4635

Statistical Comparisons

Although CCA assumes a unimodal distribution for organisms in response to environmental conditions, it can be robust for other ecological biomodal responses to gradients, unequal ranges, and unequal maxima, and it is well suited for hypothesis testing based on ecological datasets where major gradients are known (Ramette, 2007). Major environmental gradients along the selected transect of Upper Spring that were expected to relate to the distribution of functional genes included dissolved sulfide and dissolved oxygen concentrations, measured in the field, and the composition of stored sulfur species interpreted from XANES 50

spectra for sediments and mats. Weak environmental gradients along the Upper Spring transect that showed no significant relationship to distribution of OTUs were removed from CCA data tables upon preliminary review. Betaproteobacteria and Gammaproteobacteria soxB genes, and

Gammaproteobacteria, Betaproteobacteria, Deltaproteobacteria, and unclassified aprA sequences, were grouped separately for nonparametric correlation to environmental geochemistry. The results were used to support multivariate environmental gradients from CCA relating functional gene OTU taxonomic associations and distributions among sample locations for both soxB (Figures 4.8 and 4.9) and aprA (Figure 4.10) genes.

The normalized abundance of 16S rRNA genes associated with taxonomic orders were correlated with functional gene distributions, and the results indicated several significant relationships. The distribution of major represented classes of Proteobacteria among functional gene relatives significantly correlated to the distribution of the most abundant families of respective proteobacterial classes, an observation that supported confidence for the interpretations of functional gene taxonomic assignments. The distribution of

Betaproteobacteria-like soxB genes was significantly related to the abundance of Rhodocyclales

(τ = 0.96, P-value = 0.0001), whereas the distribution of Betaproteobacteria-like aprA genes was significantly related to the abundance of Burkholderiales (τ = 0.75, P-value = 0.03). The

Betaproteobacteria-like soxB gene distributions were also significantly correlated to the availability of dissolved oxygen (τ = 0.76, P-value = 0.008) and sulfide (τ = -0.82, P-value =

0.005). Gammaproteobaceria aprA genes were significantly related with the abundance and distribution of Thiothrichales (τ = 0.79, P-value =0.006), which includes the genus Thiothrix.

The distribution of Gammproteobacteria-like soxB genes, Gammaproteobacteria-like aprA genes, and Betaproteobacteria-like aprA genes did not significantly correlate (P-value >0.05) to 51

) 3

2 S (aq)

value = 0.08 = value

- P 1 Cys 1911

198 194 203 0

-2 -1.5 -1 -0.5 0 0.5 1 1.5 2 Axis 2 (29.6 %, %, (29.6 2 Axis SO4 (s) -1 DO = Thiothrix sp. 12730 201 = MCSoxBA02 -2 = MCSoxBC02 Sµ = MCSoxBA01 =Sulfuritalea hydrogenivorans -3 = MCSoxBB09 Axis 1 (34.2%, P-value = 0.02) =Alphaproteobacteria

Figure 4.8: CCA plot of soxB OTUs related to major geochemical gradients described by Axes 1 and 2, which explain 34.2% and 29.6% of the variation among samples. OTUs are represented by colored symbols categorized taxonomically, based on closest relative sequences in GenBank. Squares and circles represent OTUs associated with Gammaproteobacteria and Betaproteobacteria, respectively. Axis 1 is statistically significant (P-value <0.05), and Axis 2 is related to Axis 3 in Figure 4.9. DO = dissolved oxygen; S(aq) = dissolved sulfide; Cys = cysteine as measured from XANES spectra; SO4(s) = sulfate solid in sediment or mats.

) 2 S (aq) 1.5 1911

value = 0.54 = value

- P 201 0.5 S µ 0 S8 -2 -1.5 -1 -0.5 0 1940.5 1 1.5 2 -0.5 SO4 (s) 198

Axis 3 (25.4 %, %, (25.4 3 Axis -1 203 Cys -1.5

= Thiothrix sp. 12730 -2 = MCSoxBA02 DO = MCSoxBC02 -2.5 = MCSoxBA01 -3 =Sulfuritalea hydrogenivorans = MCSoxBB09 Axis 2 (29.6 %, P-value = 0.08) =Alphaproteobacteria

Figure 4.9: CCA plot of soxB OTUs related to major geochemical gradients described by Axes 2 and 3, which explain 29.6% and 25.4% of the variation among samples. OTUs are represented by colored symbols categorized taxonomically, based on closest relative sequences in GenBank. Squares and circles represent OTUs associated with Gammaproteobacteria and Betaproteobacteria, respectively. DO = dissolved oxygen; S(aq) = dissolved sulfide; Cys = cysteine as measured from XANES spectra; SO4(s) = sulfate solid in sediment or mats.

the dissolved oxygen or sulfide concentrations, nor to the percent contribution of S8 or Sµ in the mats. Because significance of Kendall’s τ may be affected by underrepresentation of functional gene sequence diversity for some samples, which was suggested from the rarefaction curves

(Figure 4.3), or because of methodological error, like PCR amplification biases (Smith and

Osborn, 2009), OTUs and geochemical gradients were plotted collectively following CCA analysis to interpret relationships in the overall datasets.

The results of CCA based on soxB OTUs and dissolved oxygen, sulfide, S8, Sµ, cysteine, and solid sulfate indicate relationships predominantly described by three ordination axes, representing 34.2%, 29.6%, and 25.4% of the total variation among sites. CCA triplots are shown as axis 1 vs. axis 2 (Figure 4.8) and axis 2 vs. axis 3 (Figure 4.9). Downstream sites 198 m and

203 m samples were most closely affiliated with S8, but 201 m was closely associated with Sµ.

The 194 m sample site score was affiliated with the cysteine and sulfate gradients. The ordination of the majority of OTUs was nearer to the direction of maximum change of dissolved oxygen than sulfide. According to taxonomic interpretations based on GenBank relatives and assignments to OTUs (Figure 4.4), soxB OTUs most closely related to Movile Cave uncultured environmental clones consistently plotted in the direction of maximum change in S8; OTUs related to soxB sequences of cultured organisms Thiothrix sp. 12730 and Sulfuritalea hydrogenivorans were more closely associated with the 201 m site and maximal change in Sµ. A similar relationship resulted for OTU distributions for axes 2 and 3 (Figure 4.9), in which S0 species were strongly related to the ordination axis 2, and sulfate, dissolved oxygen, and cysteine were comparatively less related to the ordination axes describing variation. Although the ordination axes 2 and 3 were not statistically significant, the association of OTUs related to cultured organisms and sample site 201 m clustered in close proximity to the Sµ gradient vector. 54

2 Cys 1.5 191S S

1 µ

value = 0.53) = value -

P 0.5 DO 196 203191L 0 194 -3 -2 -1 201 0 1 S 2 (20.6%, (20.6%, (aq) -0.5 S8

-1 Axis 2 2 Axis -1.5

=Gammaproteobacteria -2 =Betaproteobacteria 191G SO4 -2.5 =Deltaproteobacteria Meth =Firmicutes -3 =Unknown Axis 1 (24.4%, P-value = 0.046)

Figure 4.10: CCA plot of aprA OTUs related to major geochemical gradients. OTUs are represented by colored symbols categorized taxonomically, based on closest relative sequences in GenBank. DO = dissolved oxygen; S(aq) = dissolved sulfide; Cys = cysteine as measured from XANES spectra; SO4(s) = sulfate solid in sediment or mats; Meth = methionine.

The CCA likely over-represents the strength of the relationship for soxB 191L and the S8 gradient because the represented taxonomy comprises considerably less abundant

Gammaproteobacteria compared to other samples downstream.

The nonrandom ordination of aprA OTUs indicated strong relationships with interpreted taxonomic associations and geochemistry, which also corresponded predictably to gene function

(Figure 4.10). The significant ordination axis was strongly related to the dissolved oxygen and sulfide gradients, and weakly to the S8 gradient. Downstream samples comprised of

Gammaproteobacteria-like and Betaproteobacteria-like OTUs generally plotted in close proximity to the dissolved oxygen and S8 gradients. Conversely, upstream samples were more closely associated with the maximal direction of sulfide, and the interpreted taxonomy of the proximal OTUs are predominantly Deltaproteobacteria, although several

Gammaproteobacteria-like OTUs were affiliated with these sites and gradients, which reflected the relatively even distribution of Gammaproteobacteria observed in Figure 4.5.

CHAPTER 5

DISCUSSION

Links between Geochemistry and Biodiversity

Defining a relationship between the distribution of microbial metabolic genes and environmental geochemistry in sulfidic karst systems can be used to understand how carbonate dissolution in sulfidic karst proceeds. This thesis stems from ongoing research to understand sulfidic groundwater interactions, with Lower Kane Cave being an important site that serves as a proxy for less accessible sulfidic aquifers. The Upper Spring cave stream is also valuable natural analog to wastewater treatment systems used for sulfide removal.

The hypotheses of this study, specifically that functional gene distributions would relate to aqueous geochemical substrate availability and metabolic products, were supported by the statistically significant correlations among the distributions of soxB and aprA genes to dissolved oxygen, as well as to the S8 and Sµ gradient vectors according to CCA (Figures 4.8-4.10). The distribution of OTUs among soxB functional genes predominantly represented diversity in the mats for several groups of Gammaproteobacteria and Betaproteobacteria, which are classes with known sulfur-oxidizes within predominately the genera Thiothrix and Thiobacillus spp. Storage of elemental sulfur as intracellular globules is typical for these groups based on what is known of their Sox pathways. Moreover, the lack of amplification of soxC genes, and the failure to amplify soxB genes in the upstream mats, may indicate that sulfur oxidation, if occurring, does not result in the storage of elemental sulfur. Interestingly, storage of sulfur in downstream samples, based on the relationship of Gammaprotobacteria and Betaproteobacteria OTUs to the S0 gradient vectors, as well as geochemical sediment sulfur species data from XANES, is occurring

downstream. Betaproteobacteria-like and Gammaproteobacteria-like aprA OTUs also occurred in downstream mats and were also statistically correlated to dissolved oxygen availability and stored S8 (with a few exceptions, Figure 4.10). Considering the geochemical reaction (Equation

2.2), elemental sulfur storage does not promote carbonate dissolution because the metabolic pathway does not produce acidity like sulfur-oxidizers that oxidize reduced sulfur to sulfate

(Equations 2.1 or 2.3) (Engel et al., 2004b; Steinhauer et al., 2010).

The distribution of aprA deltaproteobacterial OTUs also supported the hypotheses that functional gene distribution would correlate to substrate availability and metabolic intermediates or products because the deltaproteobacterial aprA OTUs were almost exclusively associated with upstream sample sites and sulfide, cysteine, stored sulfate, and methionine gradients. The taxonomic affiliations of OTUs associated with deltaproteobacterial aprA were closely related to sulfate-reducers or disproportionators (Meyer and Kuever, 2007; Harada et al., 2009), such as

Desulfocapsa thiozymogenes and Desulfobulbus elongatus. However, because ~30% of aprA sequences were closely related to uncultured bacteria that could not be identified taxonomically, most of the OTUs assigned to raw aprA DNA sequences were difficult to classify based solely on GenBank closest relatives. Multiple horizontal gene transfer events identified by comparisons of functional gene and 16S rRNA gene phylogenies (Friedrich, 2002) are a plausible explanation of a lack of statistically significant relationships among the broadly classified aprA genes.

Therefore, although this study uncovered novel diversity among aprA (and soxB) sequences, with limited understanding of the organisms possessing these genes, it is difficult to know just how novel the organisms from Lower Kane Cave are metabolically.

Importance of Missing Functional Representation for Epsilonproteobacteria

Overall diversity is low in Lower Kane Cave mats nearest the Upper Spring orifice where

Epsilonproteobacteria comprise the largest taxonomic group represented by 16S rRNA gene sequence diversity; however, Epsilonproteobacteria, which are considered to be putative sulfur- oxidizers that are also found in a variety of aphotic sulfidic environments, are not represented among the taxonomic assignments of soxB or aprA sequences. Instead, Gammaproteobacteria, less abundant in 16S rRNA gene sequence results, are the most commonly identified closest relatives of both soxB and aprA gene sequences. This conflict in the two datasets (i.e. 16S rRNA and funcational genes) highlights that the more research needs to be done. Based on previous work with pure cultures, Epsilonproteobacteria are expected to oxidize sulfide and have been known to possess the genes required for the Sox system (Hügler et al., 2010; Yamamoto et al.,

2010). In deep-sea vents (Logatchev field), where sulfidic waters support chemolithoautotrophically-based ecosystems, communities are dominated by

Epsilonproteobacteria and have different genera of Gammaproteobacteria (e.g., Beggiatoa spp.).

These groups are hypothesized to use different pathways for carbon fixation and sulfur oxidation, based on amplification of soxB and aprA genes, respectively (Hügler et al., 2010).

Oxidation of sulfide without sulfur globule formation has been attributed to the periplasmic Sox pathway, and the Sox genes in Epsilonproteobacteria strains are separated into two clusters, which is distinct from other known chemolithoautotrophs (Yamamoto et al., 2010).

Specifically, Epsilonproteobacteria have soxB genes for reduced sulfur compound oxidation using O2 and nitrate, but the genes are distantly related to those from other organisms (Meyer et al., 2007; Sievert et al., 2008). In a previous study, the epsilonproteobacterial sulfite:acceptor oxidoreductase pathway was proposed to function as a direct oxidation pathway, but APS 59

reductase did not appear functional (Takai et al., 2005). Alternative explanations for the lack of

Epsilonproteobacteria represented among the closest relatives to LKC soxB sequences could include a deficiency in GenBank of sequences related to Epsilonproteobacteria, but another more plausible explanation is that the specificity of primers used to amplify soxB excluded

Epsilonproteobacteria soxB genes.

Microbial Metabolisms in Industrial Settings

Cultured organisms, such as Thiobacillus spp. (Oh et al., 1998) or Chlorobi groups

(Henshaw et al., 1998), are commonly used in bioreactors for the removal of hydrogen sulfide.

However, reactor failure due to problems, such as susceptibility to pulses in sulfide concentrations (Buisman et al., 1991) and impaired phototrophy due to turbidity, are drawbacks for using such organisms for bioremediation (Khanal, 2008). Constructing wastewater treatment systems with microbial communities specifically designed for optimum performance in the presence of traditional stressors is advantageous, and recent studies have focused exploiting the metabolic functions of microbial communities in the presence of sulfide (Khanal, 2008; Vannini et al., 2008; Ji et al., 2009; Wang et al., 2010; Li et al., 2012).

Conclusions and Future Directions

The distribution of Lower Kane Cave microbial mat functional gene OTUs representing relatives of sulfur oxidizers with putative sulfur storing pathways was significantly correlated to

0 higher dissolved oxygen in the cave stream and S composition, as S8 and Sµ stored in the mats, supporting the hypotheses that functional gene distribution was related to the availability of metabolic substrates and accumulation of metabolic products. Sulfur storage in downstream mat samples suggests that sulfur oxidation does not contribute locally to carbonate dissolution and speleogenesis. aprA OTUs representing sequences related to sulfur disproportionaters distributed 60

almost exclusively among upstream samples (correlating to highest sulfide in the cave stream), the lack of soxB amplification in most upstream samples, relatively low S0 storage in mats, and the abundance of Epsilonproteobacteria upstream suggests that if microbial metabolic sulfur oxidation occurs utilizing the periplasmic Sox system, the net reaction may contribute to localized carbonate dissolution. Further analysis of microbial mat geochemical data based on

XANES spectra using additional reference compounds (e.g. cysteine) to determine the composition of stored sulfur species may refine the statistical significance of results and improve our understanding of the metabolic activity, nutrient cycling, and net effect of geochemical process that affect rates of karstification in sulfidic systems.

In order to further constrain the function of soxB and aprA genes described in this study, several additional data processing steps will be necessary in future work. In addition to increasing the resolution of variation in geochemical data, by supplementing XANES spectral fittings with additional reference compounds and performing multivariate statistical analyses such as Principal Component Analysis (PCA), functional and 16s rRNA gene sequence dataset can be further analyzed. First, the amino acid translations for each gene sequence should be performed and aligned with other known soxB or aprA amino acid sequences to more precisely predict gene function. Characterizing the phylogenies of functional genes, based on DNA or amino acids, will help identify the taxonomic associations of sequences that were unknown in this study; analysis of phylogenies of multiple functional genes in conjunction with 16S rRNA genes may help identify events of horizontal gene transfer as well as timing of evolution of early

Earth metabolisms.

REFERENCES

Ahmadian, A., Ehn, M., and Hober, S., 2006, Pyrosequencing: History, biochemistry and future: Clinica Chimica Acta, v. 363, p. 83-94.

Amend, A.S., Seifert, K.A., and Bruns, T.D., 2010, Quantifying microbial communities with 454 pyrosequencing: does read abundance count?: Molecular Ecology, v. 19, p. 55-65.

American Public Health Association (APHA), 1998, in Clersceri, L.S., Greenberg, A.E., Eaton, A.D. (eds.), Standard Methods for the Examination of Water and Wastewater, 20th ed., US Environmental Protection Agency, American Public Health Association, the American Water Works Association, and the Water Environmental Federation, p.1220.

Anandham, R., Indiragandhi, P., Madhaiyan, M., Ryu, K.Y., Jee, H.J., and Sa, T.M., 2008, Chemolithoautotrophic oxidation of thiosulfate and phylogenetic distribution of sulfur oxidation gene (soxB) in rhizobacteria isolated from crop plants: Research in Microbiology, v. 159, p. 579-589.

Azai, C., Tsukatani, Y., Harada, J., and Oh-oka, H., 2009, Sulfur oxidation in mutants of the photosynthetic green sulfur bacterium Chlorobium tepidum devoid of cytochrome c-554 and SoxB: Photosynthesis Research, v. 100, p. 57-65.

Bagchi, A., and Ghosh, T.C., 2005, A structural study towards the understanding of the interactions of SoxY, SoxZ, and SoxB, leading to the oxidation of sulfur anions via the novel global sulfur oxidizing (sox) operon: Biochemical and Biophysical Research Communications, v. 335, p. 609-615.

Bagchi, A., and Ghosh, T.C., 2006, Structural insight into the interactions of SoxV, SoxW and SoxS in the process of transport of reductants during sulfur oxidation by the novel global sulfur oxidation reaction cycle: Biophysical Chemistry, v. 119, p. 7-13.

Bardischewsky, F., Quentmeier, A., Rother, D., Hellwig, P., Kostka, S., and Friedrich, C., 2005, Sulfur dehydrogenase of Paracoccus pantotrophus: the heme-2 domain of the molybdoprotein cytochrome c complex is dispensible for catalytic activity: Biochemistry, v. 44, p. 7024-7034.

Best, D.J., and Gipps, P.G., 1974, Algorithm AS 71: The upper tail probabilities of Kendall's tau Applied Statistics, v. 23, p. 98-100.

Bhowal, S., and Chakraborty, R., 2011, Five novel acid-tolerant oligotrophic thiosulfate- metabolizing chemolithotrophic acid mine drainage strains with the genus Burkholderia of Betaproteobacteria and identification of two novel soxB gene homologues: Research in Microbiology, v. 162, p. 436-445.

Buisman, C.J.N., Lettinga, G., Paasschens, C.W.M., and Habets, L.H.A., 1991, Biotechnological sulphide removal from effluent: Water Science and Technology, v. 24, p. 347-356.

Busby, J.F., Plummer, L.N., Lee, R.W., and Hanshaw, B.B., 1991, Geochemical evolution of water in the Madison Aquifer in parts of Montana, South Dakota, and Wyoming: U.S. Geological Survey Professional Paper 1273-F.

Canfield, D.E., and Teske, A., 1996, Late Proterozoic rise in atmospheric oxygen concentration inferred from phylogenetic and sulphur-isotope studies: Nature, v. 382, p. 127-132.

Chen, Y., Wu, L., Boden, R., Hillebrand, A., Kumaresan, D., Moussard, H., Baciu, M., Lu, Y., and Colin Murrell, J., 2009, Life without light: microbial diversity and evidence of sulfur- and ammonium-based chemolithotrophy in Movile Cave: The ISME Journal, v. 3, p. 1093-1104.

Cole, J.R., Wang, Q., Cardenas, E., Fish, J., Chai, B., Farris, R.J., Kulam-Syed-Mohideen, A.S., McGarrell, D.M., Marsh, T., Garrity, G.M., and Tiedje, J.M., 2009, The Ribosomal Database Project: improved alignments and new tools for rRNA analysis: Nucleic Acids Research, v. 37, p. D141-D145.

Dahl, C., and Prange, A., 2006, Bacterial sulfur globules: occurrence, structure and metabolism in Shively, J., ed., Volume 1: Microbiology Monographs, Springer Berlin / Heidelberg, p. 21-51.

Davidson, A.C., and Hinkley, D.V., 1997, Bootstrap Methods and Their Application, Cambridge University Press, Cambridge.

Downey, J.S., 1984, Geohydrology of the Madison and associated aquifers in parts of Montana, North Dakota, South Dakota, and Wyoming: U.S. Geological Survey Professional Paper 1273-G.

Egemeier, S.J., 1981, Cavern development by thermal waters: The NSS Bulletin v. 43, p. 31-51.

Engel, A.S., 2010, Microbial Diversity of Cave Ecosystems, in Barton, L.L., ed., Geomicrobiology: Molecular and Environmental Perspective, Springer Science + Business Media B.V., p. 219-238.

Engel, A.S., Lee, N.M., Porter, M.L., Stern, L.A., Bennett, P.C., and Wagner, M., 2003, Filamentous "Epsilonproteobacteria" dominate microbial mats from sulfidic cave springs: Applied and Environmental Microbiology, v. 69, p. 5503-5511.

Engel, A.S., Lichtenberg, H., Prange, A., and Hormes, J., 2007, Speciation of sulfur from filamentous microbial mats from sulfidic cave springs using X-ray absorption near-edge spectroscopy: FEMS Microbiology Letters, v. 269, p. 54-62.

Engel, A.S., Meisinger, D.B., Porter, M.L., Payn, R.A., Schmid, M., Stern, L.A., Schleifer, K., and Lee, N.M., 2010, Linking phylogenetic and functional diversity to nutrient spiraling in microbial mats from Lower Kane Cave (USA): The ISME Journal, v. 4, p. 98-110.

Engel, A.S., Porter, M.L., Stern, L.A., Quinlan, S., and Bennett, P.C., 2004a, Bacterial diversity and ecosystem function of filamentous microbial mats from aphotic (cave) sulfidic springs dominated by chemolithoautotrophic 'Epsilonproteobacteria': FEMS Microbiology Ecology, v. 51.

Engel, A.S., Stern, L.A., and Bennett, P.C., 2004b, Microbial contributions to cave formation: New insights into sulfuric acid speleogenesis: Geology, v. 32, p. 369.

EPA, 1993, Report to Congress on Hydrogen Sulfide Air Emissions Associated with Extraction of Oil and Natural Gas: EPA-453-R-93-045.

—, 1995, Compilation of Air Pollutant Emission Factors, AP 42. Petroleum Industry: Natural Gas Processing 5.3.1.

—, 2002, Hydrogen Sulfide: Interim Acute Exposure Guideline Levels (AEGLS) for NAS/COT Subcommittee for AEGLs, in Agency, E.P., ed., Volume 2010.

—, 2003, Toxicological Review of Hydrogen Sulfide CAS No. 7783-06-4. In Support of Summary Information on the Integrated Risk Information System (IRIS), in Agency, E.P., ed.

Epel, B., Schafer, K.O., Quentmeier, A., Friedrich, C., and Lubitz, W., 2005, Multifrequency EPR analysis of the dimanganese cluster of the putative sulfate thiohydrolase SoxB of Paracoccus pantotrophus: Journal of Biological Inorganic Chemistry, v. 10, p. 636-642.

Fischer, D.W., Smith, S.A., Peck, W.D., LeFever, J.A., LeFever, R.D., Helms, L.D., Sorensen, J.A., Steadman, E.N., and Harju, J.A., 2005, Sequestration potential of the Madison of the Northern Great Plains Aquifer System, in Dakota, U.o.N., ed., Energy and Environmental Research Center p. 1-19.

Foriel, J., Philippot, P., Susini, J., Dumas, P., Somogyi, A., Salomé, M., Khodja, H., Ménez, B., Fouquet, Y., Moreira, D., and Lópe -Garc a, P., 2004, High-resolution imaging of sulfur oxidation states, trace elements, and organic molecules distribution in individual microfossils and contemporary microbial filaments: Geochimica et Cosmochimica Acta, v. 68, p. 1561-1569.

Forti, P., Galdenzi, S., and Sarbu, S.M., 2002, The hypogenic caves: a powerful tool for the study of seeps and their environmental effects Continental Shelf Research, v. 22, p. 2373- 2386.

Franz, B., Lichtenberg, H., Dahl, C., Hormes, J., and Prange, A., 2009, Utilization of 'elemental' sulfur by different phototrophic sulfur bacteria (Chromatiaceae, Ectothiorhodospiraceae): A sulfur K-edge XANES spectroscopy study: Journal of Physics: Conference Series, v. 190.

Friedrich, C., Bardischewsky, F., Rother, D., Quentmeier, A., and Fischer, J., 2005, Prokaryotic sulfur oxidation: Current Opinion in Microbiology, v. 8, p. 253-259.

Friedrich, C.G., Rother, D., Bardischewsky, F., Quentmeier, A., and Fischer, J., 2001, Oxidation of reduced inorganic sulfur compounds by bacteria: emergence of a common mechanism?: Applied and Environmental Microbiology, v. 67, p. 2873-2882.

Friedrich, M.W., 2002, Phylogenetic analysis reveals multiple lateral transfers of adenosine-5'- phosphosulfate reductase genes among sulfate-reducing microorganisms: Journal of Bacteriology, v. 184, p. 278-289.

Galdenzi, S., Cocchioni, M., Morichetti, L., Amici, V., and Scuri, S., 2008, Sulfidic groundwater chemistry in the Frasassi Caves, Italy: Journal of Cave and Karst Studies, v. 70, p. 94-107.

Galdenzi, S., and Maruoka, T., 2003, Gypsum deposits in the Frasassi Caves, central Italy: Journal of Cave and Karst Studies, v. 65, p. 111-125.

George, G.N., Gnida, M., Bazylinski, D., Prince, R.C., and Pickering, I.J., 2008, X-ray Absorption Spectroscopy as a probe of microbial sulfur biochemistry: the nature of bacterial sulfur globules revisited: Journal of Bacteriology, v. 190, p. 6376-6383.

Ghosh, W., Mallick, S., and DasGupta, S.K., 2009, Origin of the Sox multienzyme complex system in ancient thermophilic bacteria and coevolution of its constituent proteins: Research in Microbiology, v. 160, p. 409-420.

Gray, C., 2010, Influences of microbial diversity on carbonate geochemistry across a transition from fresh to saline water in the Edwards Aquifer, Texas: Baton Rouge, Louisiana State University.

Greene, E.A., 1993, Hydraulic properties of the Madison aquifer system in the Western Rapid City area, South Dakota: U.S. Geological Survey Water Resources Investigation Report 93-4008.

Grieshaber, M.K., and Volker, S., 1998, Animal adaptations for tolerance and exploitation of poisonous sulfide: Annual Review of Physiology, v. 60, p. 33-53.

Grimm, F., Franz, B., and Dahl, C., 2008, Thiosulfate and sulfur oxidation in purple sulfur bacteria, in Dahl, C., and Friedrich, C.G., eds., Microbial Sulfur Metabolism, Springer, p. 101-115. 65

Hall, J.R., Mitchell, K.R., Jackson-Weaver, O., Kooser, A.S., Cron, B.R., Crossey, L.J., and Takacs-Vescbach, C.D., 2008, Molecular characterization of the diversity and distribution of a thermal spring microbial community using rRNA and metabolic genes: Applied and Environmental Microbiology, v. 74, p. 4910-4922.

Hall, T.A., 1999, BioEdit: a user-friendly biological sequence alignment editor and analysis program for Windows 95/98/NT: Nucleic Acids Symposium Series, v. 41, p. 95-98.

Hammer, O., Harper, D.A.T., and Ryan, P.D., 2001, PAST: Paleontological Statistics software package for education and data analysis., Palaeontologia Electronica, v. 4, p. 9.

Harada, M., Yoshida, T., Kuwahara, H., Shimamura, S., Takaki, Y., Kato, C., Miwa, T., Miyake, H., and Maruyama, T., 2009, Expression of genes for sulfur oxidation in the intracellular chemoautotrophic symbiont of the deep-sea bivalve Calyptogena okutanii: Extremophiles, v. 13, p. 895-903.

He, H., Xia, J.-l., Hong, F.-f., Tao, X.-x., Leng, Y.-w., and Zhao, Y.-d., 2012, Analysis of sulfur speciation on chalcopyrite surface bioleached with Acidithiobacillus ferrooxidans: Minerals Engineering, v. 27–28, p. 60-64.

Henshaw, P.F., Bewtra, J.K., and Biswas, N., 1998, Hydrogen sulphide conversion to elemental sulphur in a suspended-growth continuous stirred tank reactor using Chlorobium limicola: Water Research, v. 32, p. 1769-1778.

Hill, I.D., 1973, Algorithm AS 66: The Normal Integral Applied Statistics, v. 22, p. 424-427.

Holkenbrink, C., Barbas, S.O., Mellerup, A., Otaki, H., and Frigaard, N.-U., 2010, Sulfur globule oxidation in green sulfur bacteria is dependent on the dissimilatory sulfite reductase system: Microbiology, v. 157, p. 1229-1239.

Hose, L.D., Palmer, A.N., Palmer, M.V., Northup, D.E., Boston, P.J., and DuChene, H.R., 2000, Microbiology and geochemistry in a hydrogen-sulphide-rich karst environment: Chemical Geology, v. 169, p. 399-423.

Hose, L.D., and Pisarowicz, J.A., 1999, Cueva de Villa Luz, Tabasco, Mexico: Reconnaissance study of an active sulfur spring cave: Journal of Cave and Karst Studies, v. 61, p. 13-21.

Hügler, M., Gärtner, A., and Imhoff, J.F., 2010, Functional genes as markers for sulfur cycling and CO2 fixation in microbial communities of hydrothermal vents of the Logatchev field: FEMS Microbiology Ecology, v. 73, p. 526-537.

Huntoon, P.W., 1985, Rejection of recharge water from Madison Aquifer along eastern perimeter of Bighorn Artesion Basin, Wyoming: Ground Water, v. 23, p. 345-353.

Iurkiewicz, A.A., and Stevanovic, Z.P., 2010, Reconnaissance study of active sulfide springs and cave systems in the southern part of the Sulaimani Governorate (NE Iraq): Carbonates and Evaporites, v. 25, p. 203-216.

Jagnow, D.H., Hill, C.A., Davis, D.G., and DuChene, H.R., 2000, History of the Sulfuric Acid Speleogenesis in the Guadalupe Mountains, New Mexico: Journal of Cave and Karst Studies, v. 62, p. 54-59.

Janssen, A.J.H., Lettinga, G., and de Keizer, A., 1999, Removal of hydrogen sulphide from wastewater and waste gases by biological conversion to elemental sulphur: Colloidal and interfacial aspects of biologically produced sulphur particles: Colloids and Surfaces A: Physicochemical and Engineering Aspects, v. 151, p. 389-397.

Janssen, P.H., Schuhmann, A., Bak, F., and Liesack, W., 1996, Disproportionation of inorganic sulfur compounds by the sulfate-reducing bacterium Desulfocapsa thiozymogenes gen. nov., sp. nov: Archives of Microbiology, v. 166, p. 184-192.

Ji, G., Liao, B., Tao, H., and Lei, Z., 2009, Analysis of bacteria communities in an up-flow fixed- bed (UFB) bioreactor for treating sulfide in hydrocarbon wastewater: Bioresource Technology, v. 100, p. 5056-5062.

Kendall, M.G., 1976, Rank Correlation Methods. 4th ed., Griffin.

Kern, M., and Simon, J., 2009, Electron transport chains and bioenergetics of respiratory nitrogen metabolism in Wolinella succinogenes and other Epsilonproteobacteria: Biochimica et Biophysica Acta (BBA) - Bioenergetics, v. 1787, p. 646-656.

Khanal, S.K., 2008, Bioenergy recovery from sulfate-rich waste streams and strategies for sulfide removal, Anaerobic Biotechnology for Bioenergy Production: Principles and Applications: Ames, Wiley-Blackwell, p. 133-160.

Kleinjan, W.E., De Keizer, A., and Janssen, A.J.H., 2003, Biologically produced sulfur: Topics in Current Chemistry, v. 230, p. 167-188.

Korbie, D.J., and Mattick, J.S., 2008, Touchdown PCR for increased specificity and sensitivity in PCR amplification: Nat. Protocols, v. 3, p. 1452-1456.

Krishnani, K.K., Gopikrishna, G., Pillai, S.M., and Gupta, B.P., 2010a, Abundance of sulphur- oxidizing bacteria in coastal aquaculture using soxB gene analyses: Aquaculture Research, v. 41, p. 1290-1301.

Krishnani, K.K., Kathiravan, V., Natarajan, M., Kailasam, M., and Pillai, S.M., 2010b, Diversity of sulfur-oxidizing bacteria in greenwater system of coastal aquaculture: Applied Biochemistry and Biotechnology, v. 162, p. 1225-1237.

Kubo, K., Knittel, K., Amann, R., Fukui, M., and Matsuuma, K., 2011, Sulfur-metabolizing bacterial populations in microbial mats of the Nakabusa hot spring, Japan: Systematic and Applied Microbiology, v. 34, p. 293-302.

Kuenen, J.G., 1975, Colourless sulfur bacteria and their role in the sulfur cycle: Plant Soil, v. 43, p. 49-76.

Kumar, P.S., Brooker, M.R., Dowd, S.E., and Camerlengo, T., 2011, Target region selection is a critical determinant of community fingerprints generated by 16S pyrosequencing PLoS ONE, v. 6, p. e20956.

Lee, Y.-J., Prange, A., Lichtenberg, H., Rohde, M., Dashti, M., and Wiegel, J., 2007, In situ analysis of sulfur species in sulfur globules produced from thiosulfate by Thermoanaerobacter sulfurigignens and Thermoanaerobacterium thermosulfurigenes: Journal of Bacteriology, v. 189, p. 7525-7529.

Li, W., Wang, L.-Y., Duan, R.-Y., Liu, J.-F., Gu, J.-D., and Mu, B.-Z., 2012, Microbial community characteristics of petroleum reservoir production water amended with n- alkanes and incubated under nitrate-, sulfate-reducing and methanogenic conditions: International Biodeterioration & Biodegradation, v. 69, p. 87-96.

Ludwig, W., and Schleifer, K.H., 1994, Bacterial phylogeny based on 16S and 23S rRNA sequence analysis: FEMS Microbiology Reviews, v. 15, p. 155-173.

Luo, J.-F., Lin, W.-T., and Guo.Y., 2011, Functional genes based analysis of sulfur-oxidizing bacteria community in sulfide removing bioreactor: Applied Microbiology and Biotechnology, v. 90, p. 769-778.

Macalady, J.L., Dattagupta, S., Schaperdoth, I., Jones, D.S., Druschel, G.K., and Eastman, D., 2008, Niche differentiation among sulfur-oxidizing bacterial populations in cave waters: ISME Journal, v. 2, p. 590-601.

McCaleb, J.A., and Wayhan, D.A., 1969, Geologic reservoir analysis, Mississippian Madison Formation, Elk Basin Field, Wyoming-Montana: AAPG Bulletin, v. 53, p. 2094-2113.

Meyer, B., Imhoff, J.F., and Kuever, J., 2007, Molecular analysis of the distribution and phylogeny of the soxB gene among sulfur-oxidizing bacteria - evolution of the Sox sulfur oxidation enzyme system: Environmental Microbiology, v. 9, p. 2957-2977.

Meyer, B., and Kuever, J., 2007, Molecular analysis of the diversity of sulfate-reducing and sulfur-oxidizing prokaryotes in the environment, using aprA as functional marker gene: Applied and Environmental Microbiology, v. 73, p. 7664-7679.

Muyzer, G., and Stams, A., J.M., 2008, The ecology and biotechnology of sulphate-reducing bacteria: Nature Reviews Microbiology, v. 6, p. 441-454. 68

Ogawa, T., Furusawa, T., Nomura, R., Seo, D., Hosoya-Matsuda, N., Sakurai, H., and Inoue, K., 2008, SoxAX binding protein, a novel component of the thiosulfate-oxidizing multienzyme system in the green sulfur bacterium Chlorobium tepidum: Journal of Bacteriology, v. 190, p. 6097-6110.

Ogawa, T., Furusawa, T., Shiga, M., Seo, D., Sakurai, H., and Inoue, K., 2010, Biochemical studies of a soxF-encoded monomeric flavoprotein purified from the green sulfur bacterium Chlorobaculum tepidum that stimulates in vitro thiosulfate oxidation: Bioscience Biotechnology and Biochemistry, v. 74, p. 771-780.

Oh, K., Kim, D., and Lee, I., 1998, Development of effective hydrogen sulfide removing equipment using Thiobacillus sp. IW: Environmental Pollution, v. 99, p. 87-92.

Palmer, M.W., 1993, Putting things in even better order: the advantages of Canonical Correspondence Analysis: Ecology, v. 74, p. 2215-2230.

Petri, R., Podgorsek, L., and Imhoff, J.F., 2001, Phylogeny and distribution of the soxB gene among thiosulfate-oxidizing bacteria: Fems Microbiology Letters, v. 197, p. 171-178.

Plummer, L.N., Busby, J.F., Lee, R.W., and Hanshaw, B.B., 1990, Geochemical modeling of the Madison Aquifer in parts of Montana, Wyoming, and South Dakota: Water Resources Research, v. 26, p. 1981-2014.

Polyak, V.J., and Provencio, P.P., 2000, Summary of the timing of sulfuric acid speleogenesis for the Guadalupe caves based on ages of alunite: Journal of Cave and Karst Studies, v. 62, p. 72-74.

Poretsky, R.S., Bano, N., Buchan, A., LeCleir, G., Kleikemper, J., Pickering, M., Pate, W.M., Moran, M.A., and Hollibaugh, J.T., 2005, Analysis of microbial gene transcripts in environmental samples: Applied and Environmental Microbiology, v. 71, p. 4121-4126.

Prange, A., 2008, Speciation analysis of microbiologically produced sulfur by X-ray Absorption Near Edge Structure Spectroscopy, in Dahl, C., and Friedrich, C.G., eds., Microbial Sulfur Metabolisms: Berlin Heidelberg, Springer-Verlag, p. 259-272.

Prange, A., Arzberger, I., Engemann, C., Modrow, H., Schumann, O., Truper, H.G., Steudel, R., Dahl, C., and Hormes, J., 1999, In situ analysis of sulfur in the sulfur globules of phototrophic sulfur bacteria by X-ray absorption near edge spectroscopy: Biochimica et Biophysica Acta, v. 1428, p. 446-454.

Puşcaş, C.M., Onac, B.P., and Tămaş, T., 2010, The mineral assemblage of caves within Şălitrari Mountain (Cerna Valley, SW Romania): depositional environment and speleogenetic implications: Carbonates and Evaporites, v. 25, p. 107-115.

Putnam, L.D., and Long, A.J., 2007, Analysis of ground-water flow in the Madison Aquifer using fluorescent dyes injected in spring Creek and Rapid Creek near Rapid City, South Dakota: U.S. Geological Survey Scientific Investigations Report 2007-5137.

Quentmeier, A., Hellwig, P., Bardischewsky, F., Grelle, G., Kraft, R., and Friedrich, C.G., 2003, Sulfur oxidation in Paracoccus pantotrophus: interaction of the sulfur-binding protein SoxYZ with the dimanganese SoxB protein: Biochemical and Biophysical Research Communications, v. 312, p. 1011-1018.

Ramette, A., 2007, Multivariate analyses in microbial ecology FEMS Microbiology Ecology, v. 62, p. 142-160.

Ravel, B., and Newville, M., 2005, Athena, Artemis, Hephaestus: data analysis for X-ray absorption spectroscopy using Ifeffit: Journal of Synchrotron Radiation, v. 12, p. 537-541.

Reyes-Avila, J., Razo-Flores, E.a., and Gomez, J., 2004, Simultaneous biological removal of nitrogen, carbon and sulfur by denitrification: Water Research, v. 38, p. 3313-3321.

Rose, T., Henikoff, J.G., and Henikoff, S., 2003, CODEHOP (COnsensus-DEgenerate Hybrid Oligonucleotide Primer) PCR primer design: Nucleic Acids Research, v. 31, p. 3763- 3766.

Rose, T., Schultz, E., Henikoff, J., Pietrokovski, S., McCallum, C., and Henikoff, S., 1998, Concensus-degenerate hybrid oligonucleotide primers for amplification of distantly related sequenes Nucleic Acids Research, v. 27, p. 1628-1635.

Rosetti, S., Blackall, L.L., Levantesi, C., Uccelletti, D., and Tandoi, V., 2003, Phylogenetic and physiological characterization of a heterotrophic chemolithoautotrophic Thiothrix strain isolated from activated sludge: International Journal of Systematic and Evolutionary Microbiology, v. 53, p. 1271-1276.

Rother, D., Henrich, H.J., Quentmeier, A., Bardischewsky, F., and Friedrich, C.G., 2001, Novel genes of the sox gene cluster, mutagenesis of the flavoprotein SoxF, and evidence for a general sulfur-oxidizing system in Paracoccus pantotrophus GB17: Journal of Bacteriology, v. 183, p. 4499-4508.

Rother, D., Orawski, G., Bardischewsky, F., and Friedrich, C.G., 2005, SoxRS mediated regulation of chemotrophic sulfur oxidation in Paracoccus pantrophus: Microbiology, v. 151, p. 1707-1716.

Sakurai, H., Ogawa, T., Shiga, M., and Inoue, K., 2010, Inorganic sulfur oxidizing system in green sulfur bacteria: Photosynthesis Research, v. 104, p. 163-176.

Saleh-Lakha, S., Miller, M., Campbell, R.G., Schneider, K., Elahimanesh, P., Hart, M.M., and Trevors, J.T., 2005, Microbial gene expression in soil: methods, applications and challenges: Journal of Microbiological Methods, v. 63, p. 1-19.

Schloss, P.D., Gevers, D., and Westcott, S.L., 2011, Reducing effects of PCR amplification and sequencing artifacts on 16s rRNA-based studies: PloS one, v. 6.

Sievert, S.M., Hugler, M., Taylor, C.D., and Wirsen, C.O., 2008, Sulfur oxidation at deep-sea hydrothermal vents, in Dahl, C., and Friedrich, C., eds., Microbial Sulfur Metabolisms, Springer, p. 238-258.

Smith, C.J., and Osborn, A.M., 2009, Advantages and limitations of quantitative PCR (Q-PCR)- based approaches in microbial ecology: FEMS Microbiology Ecology, v. 67, p. 6-20.

Stacy, M.E., and Huntoon, P.W., 1994, Karstic groundwater circulation in the fault-severed Madison Aquifer in the Casper Mountain area of Natrona County, Wyoming: WWRC- 94-26.

Steinhauer, E.S., Omelon, C.R., and Bennett, P.C., 2010, Limestone corrosion by neutrophilic sulfur-oxidizing bacteria: A coupled microbe-mineral system: Geomicrobiology Journal, v. 27, p. 723-738.

Stout, J., De Smet, L., Panjikar, S., Weiss, M.S., Savvides, S.N., and Van Beeumen, J., 2006, Crystallization, preliminary crystallographic analysis and phasing of the thiosulfate- binding protein SoxY from Chlorobium limicola f. thiosulfatophilum: Acta Crystallographica Section F-Structural Biology and Crystallization Communications, v. 62, p. 1093-1096.

Sun, Y., Wolcott, R.D., and Dowd, S.E., 2011, Tag-encoded FLX amplicon pyrosequencing for the elucidation of microbial and functional gene diversity in any environment, in Kwon, Y.M., and Ricke, S.C., eds., High-throughput Next Generation Sequencing: Methods and Applications, Methods in Molecular Biology, Volume 83, p. 129-141.

Takai, K., Campbell, B.J., Cary, C., Suzuki, M., Oida, H., Nunoura, T., Hirayama, H., Nakagawa, S., Suzuki, Y., Inagaki, F., and Horikoshi, K., 2005, Enzymatic and genetic characterization of carbon and energy metabolisms by deep-sea hydrothermal chemolithoautotrophic isolates of Epsilonproteobacteria: Applied and Environmental Microbiology, v. 71, p. 7310-7320.

Tang, K., Baskaran, V., and Nemati, M., 2009, Bacteria of the sulphur cycle: An overview of microbiology, biokinetics and their role in petroleum and mining industries: Biochemical Engineering Journal, v. 44, p. 73-94.

Tsu, I.H., Huang, C.Y., Garcia, J.L., Patel, B.K.C., Cayol, J.-L., Baresi, L., and Mah, R.A., 1998, Isolation and characterization of Desulfovibrio senezii sp. nov., a halotolerant sulfate 71

reducer from a solar saltern and phylogenetic confirmation of Desulfovibrio fructosovorans as a new species: Archives of Microbiology, v. 170, p. 313-317.

Vairavamurthy, M.A., Wang, S., Khandelwal, B., Manowitz, B., Ferdelman, T., and Fossing, H., 1995, Sulfur transformations in early diagenetic sediments from the Bay of Concepcion, off Chile, in Vairavamurthy, M.A., and Schoonen, M.A.A., eds., Geochemical Transformations of Sedimentary Sulfur: Washington DC, American Chemical Society, p. 38-58.

Valz, P.D., and Thompson, M.E., 1994, Exact inference for Kendall S and Spearman rho: Journal of Computational and Graphical Statistics, v. 3, p. 459-472.

Vannini, C., Munz, G., Mori, G., Lubello, C., Verni, F., and Petroni, G., 2008, Sulphide oxidation to elemental sulphur in a membrane bioreactor: Performance and characterization of the selected microbial sulphur-oxidizing community: Systematic and Applied Microbiology, v. 31, p. 461-473.

Verte, F., Kostanjevecki, V., De Smet, L., Meyer, T.E., Cusanovich, M.A., and Van Beeumen, J.J., 2002, Identification of a thiosulfate utilization gene cluster from the green phototrophic bacterium Chlorobium limicola: Biochemistry, v. 41, p. 2932-2945.

Wang, J., Shen, S., Kang, J., Li, H., and Guo, Z., 2010, Effect of ore solid concentration on the bioleaching of phosphorus from high-phosphorus iron ores using indigenous sulfur- oxidizing bacteria from municipal wastewater: Process Biochemistry, v. 45, p. 1624- 1631.

Welte, C., Hafner, S., Kratzer, C., Quentmeier, A., Friedrich, C.G., and Dahl, C., 2009, Interaction between Sox proteins of two physiologically distinct bacteria and a new protein involved in thiosulfate oxidation: Febs Letters, v. 583, p. 1281-1286.

Wessa, P., 2008, Kendall tau Rank Correlation (v1.0.10) in Free Statistics Software (v1.1.23-r7), Resa R& D--Office for Research Development and Education .

Westphal, H., Eberli, G.P., Smith, L.B., Grammer, G.M., and Kislak, J., 2004, Reservoir characterization of the Mississippian Madison Formation, Wind River basin, Wyoming: AAPG Bulletin, v. 88, p. 405-432.

Whitehead, R.L., 1996, Groundwater atlas of the United States, Montana, North Dakota, South Dakota, Wyoming: U.S. Geological Survey HA 730-I.

Woyke, T., Teeling, H., Ivanova, N., Huntemann, M., Richter, M., Gloeckner, F.O., Boffelli, D., Anderson, I.J., Barry, K.W., Shapiro, H.J., Szeto, E., Kyrpides, N.C., Mussmann, M., Amann, R., Bergin, C., Ruehland, C., Rubin, E.M., and Dubilier, N., 2006, Symbiosis insights through metagenomic analysis of a microbial consortium: Nature, v. 443, p. 950- 955. 72

Yamamoto, M., Nakagawa, S., Shimamura, S., Takai, K., and Horikoshi, K., 2010, Molecular characterization of inorganic sulfur-compound metabolism in the deep-sea epsilonproteobacterium Sulfurovum sp. NBC37-1. : Environmental Microbiology, p. 1- 10.

Zhao, F.J., Lehmann, J., Solomon, D., Fox, M.A., and McGrath, S.P., 2006, Sulphur speciation and turnover in soils: evidence from the sulphur K-edge XANES spectroscopy and isotope dilution studies: Soil Biology and Biochemistry, v. 38, p. 1000-1007.

APPENDIX A

LOWER KANE CAVE SPRING GEOCHEMISTRY

LIST OF FIGURES FOR APPENDIX A

A1 Lower Kane Cave spring geochemistry...... 76

Figure A1: Lower Kane Cave spring geochemistry. Piper Diagrams showing (a) recent and historical compositions of individual Lower Kane Cave springs, solid shapes representing recent analyses and open shapes depicting historical analyses. In (b) all Lower Kane Cave samples are displayed green for comparison with other well water chemical analyses in the Madison Aquifer west of the Bighorn Mountains. Compared to the generalized evolution of groundwater chemistry expected in a carbonate aquifer, Lower Kane Cave samples are more typical of discharge compositions than recharge compositions (Busby et al., 1991).

APPENDIX B

DIVERSITY OF 16S RRNA GENE SEQUENCES

LIST OF TABLES FOR APPENDIX B

B1 Taxonomic diversity of pyrosequenced 16S rRNA genes based on RDP Classifier analyses...... 77

Table B1: Taxonomic diversity of pyrosequenced 16S rRNA genes based on RDP Classifier analyses.

Phylum Class Order Family Genus 191L 191G 191S 194 196 198 201 203 [Archaea] 2 1 5 1 2 Crenarchaeota 2 1 1 Thermoprotei 2 1 1 Desulfurococcales 2 1 1 Desulfurococcaceae Ignisphaera 2 1 1 Stetteria 1 2 1 Euryarchaeota 2 1 3 1 Methanomicrobia 1 Methanosarcinales 1 Methanosarcinaceae 1 Methanosalsum 1 Halobacteria 1 1 3 Halobacteriales 1 1 3 Halobacteriaceae 1 1 3 Haloquadratum 1 1 2 Halosarcina 1 Methanobacteria 1 Methanobacteriales 1 Methanobacteriaceae Methanosphaera 1 1

[Bacteria] 5301 4874 5953 5643 4584 4543 3989 4633 Acidobacteria 16 86 21 3 8 32 6 61 Acidobacteria_Gp2 12 Acidobacteria_Gp3 3 2 2 Acidobacteria_Gp6 2 1 Acidobacteria_Gp7 1 1 2 2 4 3 18 Acidobacteria_Gp10 8 Acidobacteria_Gp13 13 Acidobacteria_Gp16 2 Acidobacteria_Gp21 8 79

Phylum Class Order Family Genus 191L 191G 191S 194 196 198 201 203 Acidobacteria_Gp22 5 Acidobacteria_Gp23 3 Holophagae 15 41 18 1 1 28 3 34 Holophagales 15 41 18 1 1 28 3 34 Holophagaceae 15 41 18 1 1 28 3 34 Geothrix 10 12 12 6 2 9 Holophaga 5 29 6 1 1 22 1 25 Actinobacteria 65 23 6 19 2 18 34 Actinobacteria 65 23 6 19 2 18 34 Acidimicrobiales 6 5 3 2 2 9 8 Acidimicrobiaceae 2 3 3 1 Ferrithrix 2 3 3 1 Iamiaceae 4 2 2 2 8 8 Iamia 4 2 2 2 8 8 Unclassified 1 1 5 Unclassified 1 1 5 Ilumatobacter 1 1 5 Actinomycetales 47 7 2 13 6 16 Acidothermaceae 3 1 1 Acidothermus 3 1 1 Dermabacteraceae 1 3 1 2 1 8 Dermabacter 3 2 1 7 Devriesea 1 1 Helcobacillus 1 Dermatophilaceae 6 1 Kineosphaera 6 1 Unclassified 1 Fodinicola 1 Intrasporangiaceae 1 Arsenicicoccus 1 Kineosporiaceae 1 1 1 Quadrisphaera 1 1 1 Microbacteriaceae 5

Phylum Class Order Family Genus 191L 191G 191S 194 196 198 201 203 Klugiella 2 Yonghaparkia 3 Micrococcaceae 24 2 Acaricomes 24 2 Micromonosporaceae 2 Polymorphospora 2 Nakamurellaceae 1 Humicoccus 1 Nocardiaceae 1 Skermania 1 Promicromonosporaceae 1 Myceligenerans 1 Propionibacteriaceae 7 3 1 6 2 2 Propioniferax 1 Micropruina 7 3 1 6 1 2 Pseudonocardiaceae 1 Actinomycetospora 1 Bifidobacteriales 10 1 1 5 Bifidobacteriaceae 10 1 1 5 Metascardovia 8 3 Parascardovia 1 1 1 2 Scardovia 1 Coriobacteriales 8 1 1 Coriobacteriaceae 8 1 1 Gordonibacter 1 Paraeggerthella 1 1 Asaccharobacter 6 Denitrobacterium 1 Nitriliruptorales 1 Nitriliruptoraceae 1 Nitriliruptor 1 Solirubrobacterales 4 1 1 Conexibacteraceae 3 1 1

Phylum Class Order Family Genus 191L 191G 191S 194 196 198 201 203 Conexibacter 3 1 1 Solirubrobacteraceae 1 Solirubrobacter 1 Aquificae 49 5 3 19 2 2 Aquificae 49 5 3 19 2 2 Aquificales 49 5 3 19 2 2 Aquificaceae 13 1 2 2 1 Aquifex 12 1 2 Hydrogenobaculum 1 2 1 Unclassified 1 Thermosulfidibacter 1 Desulfurobacteriaceae 29 7 Balnearium 29 7 Hydrogenothermaceae 6 4 3 10 1 Venenivibrio 6 4 3 10 1 Bacteroidetes 362 217 385 217 317 1622 977 446 Unclassified 6 5 14 3 11 11 27 9 Unclassified 6 5 14 3 11 11 27 9 Unclassified 6 5 14 3 11 11 27 9 Fulvivirga 6 7 2 7 9 13 6 Marinifilum 3 7 4 1 13 Prolixibacter 2 1 1 1 3 Bacteroidia 225 111 145 118 164 806 317 66 Bacteroidales 225 111 145 118 164 806 317 66 Marinilabiaceae 16 22 5 12 40 30 22 9 Alkaliflexus 16 22 5 12 40 30 22 9 Anaerophaga 3 1 1 Porphyromonadaceae 143 57 122 90 123 565 176 15 Tannerella 2 1 1 3 2 124 8 6 Odoribacter 1 1 3 Paludibacter 139 41 48 57 83 432 135 7 Petrimonas 4 9 5 12 2 2 1 Proteiniphilum 2 11 63 24 26 7 28 1

Phylum Class Order Family Genus 191L 191G 191S 194 196 198 201 203 Prevotellaceae 1 Paraprevotella 1 Rikenellaceae 66 27 14 12 209 114 37 Alistipes 18 11 5 6 90 75 5 Rikenella 48 16 9 6 119 39 32 Bacteroidaceae 5 4 4 2 4 4 Bacteroides 5 4 4 2 4 4 Unclassified 1 1 Phocaeicola 1 1 Sphingobacteria 63 75 150 44 85 191 386 169 Sphingobacteriales 63 75 150 44 85 191 386 169 Chitinophagaceae 18 8 4 5 6 4 37 Parasegetibacter 1 Sediminibacterium 1 Segetibacter 1 1 4 3 Terrimonas 14 Filimonas 7 8 2 5 4 1 Flavisolibacter 17 Lacibacter 10 1 1 1 Cyclobacteriaceae 12 1 8 1 1 17 8 19 Aquiflexum 10 1 5 1 11 6 18 Belliella 2 3 1 6 2 1 Flammeovirgaceae 41 36 79 25 43 117 246 41 Fabibacter 6 1 1 1 23 62 14 Roseivirga 2 1 Thermonema 1 5 1 3 1 Flexithrix 23 15 7 9 8 74 140 2 Limibacter 9 1 9 Perexilibacter 1 6 9 10 1 Persicobacter 5 18 50 5 24 18 24 23 Reichenbachiella 5 1 1 8 Rhodothermaceae 15 1 3 1 1 Rhodothermus 15 1 3 1 1

Phylum Class Order Family Genus 191L 191G 191S 194 196 198 201 203 Saprospiraceae 2 25 48 Haliscomenobacter 2 25 48 Cytophagaceae 5 1 17 9 3 22 57 58 Adhaeribacter 3 1 9 1 7 1 25 Larkinella 1 2 Leadbetterella 1 11 45 8 Persicitalea 1 Sporocytophaga 7 8 3 3 11 22 Effluviibacter 1 1 Sphingobacteriaceae 5 4 37 3 5 29 22 13 Nubsella 5 4 36 3 4 15 20 9 Pseudosphingobacterium 1 2 Solitalea 1 12 2 4 Flavobacteria 68 26 76 52 57 614 247 202 Flavobacteriales 68 26 76 52 57 614 247 202 Cryomorphaceae 4 12 37 37 36 146 62 20 Brumimicrobium 1 Crocinitomix 2 7 11 8 3 1 Cryomorpha 3 2 2 2 119 6 6 Fluviicola 1 1 2 9 13 2 Lishizhenia 1 4 16 21 21 5 29 3 Owenweeksia 2 2 2 7 Sediminitomix 1 10 3 5 8 11 2 Flavobacteriaceae 64 14 39 15 21 468 185 182 Actibacter 12 1 4 1 25 36 Cloacibacterium 3 1 1 Coenonia 11 1 1 13 6 1 Croceibacter 1 9 1 1 3 2 Dokdonia 1 1 Flagellimonas 1 Flaviramulus 1 Flavobacterium 148 28 100 Formosa 1

Phylum Class Order Family Genus 191L 191G 191S 194 196 198 201 203 Galbibacter 4 1 1 1 2 9 Aestuariicola 1 Gilvibacter 2 Jejuia 1 2 1 Joostella 2 2 1 Kaistella 3 4 7 8 Kordia 3 4 3 2 Leptobacterium 7 1 3 46 19 2 Lutaonella 1 1 Lutibacter 2 Lutimonas 8 3 31 27 Mesoflavibacter 1 Mesonia 2 Myroides 2 Ornithobacterium 1 1 2 Persicivirga 1 Psychroflexus 8 3 38 17 Psychroserpens 1 Robiginitalea 2 2 1 Sandarakinotalea 7 6 7 71 14 28 Sediminibacter 1 1 Subsaxibacter 1 Ulvibacter 1 Wautersiella 2 Weeksella 1 3 Zhouia 1 1 2 7 3 Capnocytophaga 1 Zunongwangia 1 2 3 11 6 11 Cellulophaga 62 6 16 Caldiserica 80 42 12 32 6 24 20 Caldisericia 80 42 12 32 6 24 20 Caldisericales 80 42 12 32 6 24 20 Caldisericaceae 80 42 12 32 6 24 20

Phylum Class Order Family Genus 191L 191G 191S 194 196 198 201 203 Caldisericum 80 42 12 32 6 24 20 Chlamydiae 238 36 6 50 1 14 12 Chlamydiae 238 36 6 50 1 14 12 Chlamydiales 238 36 6 50 1 14 12 Parachlamydiaceae 228 15 3 41 1 4 9 Neochlamydia 221 14 3 36 2 4 Parachlamydia 7 1 5 1 2 5 Simkaniaceae 10 21 3 9 10 3 Simkania 10 21 3 9 10 3 Chlorobi 3 142 33 19 21 21 153 25 Chlorobia 3 142 33 19 21 21 153 25 Chlorobiales 3 142 33 19 21 21 153 25 Chlorobiaceae 3 142 33 19 21 21 153 25 Chloroherpeton 3 135 27 15 13 18 91 17 Prosthecochloris 7 6 4 8 3 62 8 Chloroflexi 31 538 303 125 199 130 503 381 Chloroflexi 2 1 Chloroflexales 2 1 Chloroflexaceae 2 1 Chloroflexus 2 1 Dehalococcoidetes 20 1 2 3 Unclassified 20 1 2 3 Unclassified 20 1 2 3 Dehalogenimonas 20 1 2 3 Anaerolineae 31 476 300 123 191 119 502 379 Anaerolineales 31 476 300 123 191 119 502 379 Anaerolineaceae 31 476 300 123 191 119 502 379 Anaerolinea 1 2 3 82 3 Bellilinea 2 271 51 17 55 6 38 52 Leptolinea 1 16 5 1 7 3 22 38 Levilinea 70 75 16 37 1 21 46 Longilinea 28 118 169 89 90 106 339 240 Caldilineae 32 2 4 11 1 2

Phylum Class Order Family Genus 191L 191G 191S 194 196 198 201 203 Caldilineales 32 2 4 11 1 2 Caldilineaceae 32 2 4 11 1 2 Caldilinea 32 2 4 11 1 2 Thermomicrobia 8 Sphaerobacterales 7 Sphaerobacteraceae 7 Sphaerobacter 7 Thermomicrobiales 1 Thermomicrobiaceae 1 Thermomicrobium 1 Chrysiogenetes 19 1 1 Chrysiogenetes 19 1 1 Chrysiogenales 19 1 1 Chrysiogenaceae 19 1 1 Chrysiogenes 19 1 1 Cyanobacteria 3 1 2 Cyanobacteria 3 1 2 Unclassified 3 1 2 Chloroplast 3 1 2 Bangiophyceae 2 1 Chlorarachniophyceae 1 1 1 Deferribacteres 128 51 24 57 2 27 8 Deferribacteres 128 51 24 57 2 27 8 Deferribacterales 128 51 24 57 2 27 8 Deferribacteraceae 5 1 1 1 1 5 Calditerrivibrio 3 1 1 1 1 1 Denitrovibrio 1 Mucispirillum 2 3 Unclassified 123 50 23 57 1 26 3 Caldithrix 123 50 23 57 1 26 3 Deinococcus-Thermus 6 2 1 7 1 1 Deinococci 6 2 1 7 1 1 Deinococcales 2 1 1 3 1

Phylum Class Order Family Genus 191L 191G 191S 194 196 198 201 203 Trueperaceae 2 1 1 3 1 Truepera 2 1 1 3 1 Thermales 4 1 4 1 Thermaceae 4 1 4 1 Marinithermus 1 Vulcanithermus 4 1 3 1 Fibrobacteres 1 Fibrobacteria 1 Fibrobacterales 1 Fibrobacteraceae 1 Fibrobacter 1 Firmicutes 16 575 255 82 279 350 88 262 Bacilli 19 32 7 11 211 16 26 Lactobacillales 7 19 5 6 1 6 2 Aerococcaceae 2 5 4 4 3 Dolosicoccus 2 Globicatella 1 5 4 4 1 Ignavigranum 1 Carnobacteriaceae 4 14 1 2 2 Lacticigenium 3 2 2 Allofustis 4 9 1 Atopostipes 2 Enterococcaceae 1 1 Pilibacter 1 1 Lactobacillaceae 1 1 Paralactobacillus 1 1 Streptococcaceae 1 Streptococcus 1 Bacillales 12 13 2 5 210 10 24 Thermoactinomycetaceae 1 1 Shimazuella 1 Thermoflavimicrobium 1 Paenibacillaceae 1

Phylum Class Order Family Genus 191L 191G 191S 194 196 198 201 203 Thermicanus 1 Sporolactobacillaceae 1 Tuberibacillus 1 Bacillaceae 11 10 2 4 209 10 24 Halalkalibacillus 2 2 1 1 Jeotgalibacillus 3 204 5 15 Paucisalibacillus 1 8 Saccharococcus 1 2 1 Thalassobacillus 1 1 1 4 Vulcanibacillus 2 5 3 5 1 Filobacillus 1 Planococcaceae 2 Filibacter 2 Clostridia 16 549 221 72 264 139 71 228 Thermoanaerobacterales 1 305 86 46 61 7 20 86 Thermoanaerobacteraceae 108 20 9 37 2 5 10 Ammonifex 1 4 Gelria 3 4 1 1 Mahella 2 8 1 8 1 4 Tepidanaerobacter 51 3 1 3 1 Thermacetogenium 7 2 2 1 Thermanaeromonas 36 7 2 23 5 Thermovenabulum 1 Carboxydibrachium 3 1 Fervidicola 4 Thermodesulfobiaceae 1 197 66 37 24 5 15 76 Thermodesulfobiu 1 197 66 37 24 5 15 76 m Clostridiales 15 231 134 25 201 126 27 128 Eubacteriaceae 4 3 6 Alkalibacter 1 2 Anaerofustis 3 6 Garciella 1 Unclassified 1 5 89

Phylum Class Order Family Genus 191L 191G 191S 194 196 198 201 203 Anaerovorax 1 5 Unclassified 3 8 9 1 11 4 1 2 Anaerovirgula 4 1 3 Blautia 3 2 1 1 1 4 1 Dethiosulfatibacter 1 Proteiniborus 1 2 1 Proteocatella 1 4 6 1 1 Unclassified 1 1 Symbiobacterium 1 1 Peptococcaceae 1 23 2 2 8 2 1 Cryptanaerobacter 1 Peptococcus 1 Dehalobacter 2 1 1 Desulfitibacter 1 11 1 1 4 1 Desulfonispora 1 Desulfurispora 9 1 3 Syntrophomonadaceae 1 19 37 3 12 1 3 Pelospora 1 3 1 1 1 3 Syntrophothermus 19 34 2 11 Veillonellaceae 100 23 1 21 12 11 33 Acetonema 2 2 Dendrosporobacter 11 1 1 Phascolarctobacterium 2 2 3 1 Schwartzia 2 1 1 Sporotalea 1 1 Succiniclasticum 4 7 1 5 3 Succinispira 2 4 3 1 12 Zymophilus 1 Allisonella 72 12 1 7 6 1 6 Anaeroarcus 2 3 3 Anaeroglobus 2 1 Anaeromusa 1 3 Anaerovibrio 6

Phylum Class Order Family Genus 191L 191G 191S 194 196 198 201 203 Centipeda 1 Gracilibacteraceae 24 11 3 14 6 3 Gracilibacter 1 Lutispora 23 11 3 14 6 3 Heliobacteriaceae 1 2 Heliobacillus 1 2 Lachnospiraceae 3 11 2 4 4 5 1 4 Lachnobacterium 1 1 1 Marvinbryantia 3 Parasporobacterium 1 1 Roseburia 3 4 1 2 Syntrophococcus 1 1 Anaerostipes 1 Catonella 2 1 Coprococcus 1 1 2 2 1 Dorea 1 1 1 Ruminococcaceae 24 5 6 7 5 2 43 Acetanaerobacterium 1 2 Lactonifactor 1 Papillibacter 2 2 2 2 1 26 Sporobacter 1 Subdoligranulum 1 Acetivibrio 7 1 2 11 Anaerotruncus 1 3 1 Butyricicoccus 12 3 1 2 Ethanoligenens 1 Faecalibacterium 1 4 1 Clostridiaceae 1 12 8 4 7 35 4 21 Natronincola 1 7 Thermohalobacter 6 2 35 1 5 Thermotalea 1 1 Caloranaerobacter 15 Caminicella 1 3

Phylum Class Order Family Genus 191L 191G 191S 194 196 198 201 203 Clostridiisalibacter 1 2 5 3 Clostridium 1 Geosporobacter 1 1 Unclassified 5 31 1 114 40 13 Sporanaerobacter 1 1 Tepidimicrobium 4 28 110 40 Anaerosphaera 1 1 1 Finegoldia 1 Parvimonas 1 10 Soehngenia 1 2 2 Unclassified 6 1 1 17 Acidaminobacter 5 1 12 Fusibacter 1 4 Guggenheimella 1 1 Halanaerobiales 9 1 6 23 12 Halobacteroidaceae 9 1 6 23 12 Halanaerobaculum 2 1 1 22 12 Natroniella 5 5 1 Selenihalanaerobacter 2 Natranaerobiales 4 1 2 1 2 Natranaerobiaceae 4 1 2 1 2 Dethiobacter 1 1 Natranaerobius 1 Natronovirga 2 1 2 2 Erysipelotrichi 7 2 3 3 1 8 Erysipelotrichales 7 2 3 3 1 8 Erysipelotrichaceae 7 2 3 3 1 8 Allobaculum 6 Bulleidia 2 1 Sharpea 3 1 Solobacterium 1 Turicibacter 2 1 3 3 1 Thermolithobacteria 1

Phylum Class Order Family Genus 191L 191G 191S 194 196 198 201 203 Thermolithobacterales 1 Thermolithobacteraceae 1 Thermolithobacter 1 Fusobacteria 2 1 5 1 Fusobacteriales 2 1 5 1 Fusobacteriaceae 2 1 5 1 Psychrilyobacter 2 1 5 1 Gemmatimonadetes 8 2 2 1 Gemmatimonadetes 8 2 2 1 Gemmatimonadales 8 2 2 1 Gemmatimonadaceae 8 2 2 1 Gemmatimonas 8 2 2 1 Lentisphaerae 190 74 11 65 7 72 30 Lentisphaeria 190 74 11 65 7 72 30 Lentisphaerales 19 13 8 4 2 Lentisphaeraceae 19 13 8 4 2 Lentisphaera 19 13 8 4 2 Victivallales 171 61 11 57 7 68 28 Victivallaceae 171 61 11 57 7 68 28 Victivallis 171 61 11 57 7 68 28 Nitrospira 16 9 5 1 10 11 Nitrospira 16 9 5 1 10 11 Nitrospirales 16 9 5 1 10 11 Nitrospiraceae 16 9 5 1 10 11 Nitrospira 16 9 5 1 10 11 Planctomycetes 244 132 24 130 3 19 52 Planctomycetacia 244 132 24 130 3 19 52 Planctomycetales 244 132 24 130 3 19 52 Planctomycetaceae 244 132 24 130 3 19 52 Blastopirellula 4 1 1 Gemmata 1 4 1 1 5 Isosphaera 22 4 4 Pirellula 41 34 3 23 6 20

Phylum Class Order Family Genus 191L 191G 191S 194 196 198 201 203 Planctomyces 1 2 Rhodopirellula 16 24 8 2 5 Schlesneria 129 34 21 87 2 8 9 Singulisphaera 27 23 7 1 6 Zavarzinella 3 9 3 1 1 Proteobacteria 4772 1866 4242 5029 3114 2116 1982 3139 Alphaproteobacteria 24 25 31 4 29 84 250 527 Sphingomonadales 3 4 Erythrobacteraceae 3 3 Croceicoccus 3 3 Sphingomonadaceae 1 Blastomonas 1 Unclassified 1 Unclassified 1 Novispirillum 1 Caulobacterales 3 2 1 2 41 Caulobacteraceae 1 Brevundimonas 1 Hyphomonadaceae 3 2 1 2 40 Hellea 2 2 1 Henriciella 2 7 Hirschia 1 19 Maribaculum 1 Maricaulis 13 Kiloniellales 1 Kiloniellaceae 1 Kiloniella 1 Kordiimonadales 1 Kordiimonadaceae 1 Kordiimonas 1 Rhizobiales 3 10 4 1 7 6 136 Aurantimonadaceae Martelella 1 1 1 1

Phylum Class Order Family Genus 191L 191G 191S 194 196 198 201 203 Phyllobacteriaceae Defluvibacter 1 1 Rhizobiaceae 16 Ensifer 1 Rhizobium 11 Sinorhizobium 4 Unclassified 1 Amorphus 1 Rhodobiaceae 1 2 Afifella 1 Parvibaculum 2 Xanthobacteraceae 1 Pseudolabrys 1 Beijerinckiaceae 5 2 1 6 Beijerinckia 1 Chelatococcus 2 Methylocapsa 1 3 Methylovirgula 3 1 1 2 Bradyrhizobiaceae 1 1 89 Agromonas 1 Balneimonas 1 87 Bosea 1 Rhodoblastus 1 Cohaesibacteraceae 2 Cohaesibacter 2 Hyphomicrobiaceae 1 1 5 2 13 Prosthecomicrobium 6 Zhangella 1 Cucumibacter 1 Hyphomicrobium 1 Methylorhabdus 1 5 1 4 Pedomicrobium 2 Methylobacteriaceae 3 1 8

Phylum Class Order Family Genus 191L 191G 191S 194 196 198 201 203 Meganema 2 1 Microvirga 1 8 Methylocystaceae 1 1 Methylocystis 1 Terasakiella 1 Rhodobacterales 1 3 4 1 76 Rhodobacteraceae 1 3 4 1 76 Ahrensia 1 1 Haematobacter 1 47 Pannonibacter 1 Paracoccus 1 Pelagibaca 2 1 Pseudorhodobacter 1 Rhodobacter 23 Roseisalinus 2 Salinihabitans 1 1 2 Rhodospirillales 20 10 19 4 20 66 241 267 Acetobacteraceae 7 11 2 54 Granulibacter 2 Neoasaia 7 36 Paracraurococcus 2 Rhodopila 1 Rhodovarius 5 2 1 Swaminathania 1 Tanticharoenia 4 Acidisphaera 2 1 1 Acidomonas 2 6 Belnapia 1 Rhodospirillaceae 20 3 19 4 9 64 241 213 Rhodocista 2 Rhodovibrio 1 Telmatospirillum 1 14 25 105 Thalassobaculum 2 2 1

Phylum Class Order Family Genus 191L 191G 191S 194 196 198 201 203 Caenispirillum 6 Tistrella 1 Defluviicoccus 1 1 Inquilinus 2 41 Marispirillum 20 2 17 4 3 13 3 4 Nisaea 3 Oceanibaculum 34 209 56 Rickettsiales 4 3 3 3 3 Anaplasmataceae 1 1 Anaplasma 1 1 Rickettsiaceae Orientia 4 1 1 3 2 Candidate Division SAR11 1 2 Pelagibacter 1 2 Betaproteobacteria 2 108 214 24 39 87 281 604 Burkholderiales 1 41 168 6 19 37 75 329 Alcaligenaceae 1 1 4 12 Achromobacter 1 10 Azohydromonas 4 2 Brackiella 1 Burkholderiaceae 26 137 2 3 6 17 Wautersia 103 2 Chitinimonas 1 Cupriavidus 1 33 1 Limnobacter 1 1 Paucimonas 13 Thermothrix 24 2 3 6 Unclassified 10 10 1 1 4 13 231 Rubrivivax 4 Thiobacter 1 Aquincola 5 1 Ideonella 2 2 93 Inhella 1

Phylum Class Order Family Genus 191L 191G 191S 194 196 198 201 203 Leptothrix 8 5 Methylibium 2 2 1 1 2 17 Mitsuaria 2 9 109 Paucibacter 2 Rhizobacter 1 Comamonadaceae 1 2 19 3 14 27 58 69 Hydrogenophaga 9 11 4 Malikia 6 4 Ottowia 1 2 8 3 14 41 5 Pseudacidovorax 6 1 Pseudorhodoferax 6 2 28 Ramlibacter 1 Rhodoferax 4 2 5 Roseateles 1 Schlegelella 2 1 1 Xenophilus 2 Caenimonas 4 Caldimonas 2 17 Oxalobacteraceae 3 1 Duganella 2 Oxalicibacterium 1 1 Hydrogenophilales 10 2 1 4 4 12 Hydrogenophilaceae 10 2 1 4 4 12 Hydrogenophilus 2 1 1 Tepidiphilus 1 1 Thiobacillus 9 4 3 11 Methylophilales 2 Methylophilaceae 2 Methylotenera 2 Neisseriales 2 6 1 2 1 2 7 Neisseriaceae 2 6 1 2 1 2 7 Formivibrio 1 2 Gulbenkiania 1

Phylum Class Order Family Genus 191L 191G 191S 194 196 198 201 203 Leeia 1 Microvirgula 1 2 3 Paludibacterium 1 Silvimonas 1 1 1 Bergeriella 1 Chitinibacter 4 Chitinilyticum 1 Rhodocyclales 1 53 38 16 18 45 200 256 Rhodocyclaceae Azoarcus 1 53 38 16 18 45 200 256 Sterolibacterium 1 Thauera 1 41 25 15 11 4 197 14 Uliginosibacterium 1 15 Zoogloea 2 Azonexus 2 4 1 36 2 183 Azospira 24 Denitratisoma 4 Ferribacterium 1 9 6 1 Methyloversatilis 2 Propionivibrio 7 1 7 2 4 1 3 Deltaproteobacteria 93 1242 446 278 480 644 429 933 Bdellovibrionales 4 16 16 1 11 42 58 44 Bacteriovoracaceae 3 16 15 1 11 35 58 37 Bacteriovorax 1 4 1 2 1 Peredibacter 2 16 11 9 35 58 36 Bdellovibrionaceae 1 1 7 7 Bdellovibrio 1 1 7 7 Unclassified 1 15 6 4 20 4 Syntrophorhabdaceae 1 15 6 4 20 4 Syntrophorhabdu 1 15 6 4 20 4 s Desulfarculales 5 3 Desulfarculaceae 5 3 Desulfarculus 5 3 99

Phylum Class Order Family Genus 191L 191G 191S 194 196 198 201 203 Desulfobacterales 51 290 164 106 196 323 144 391 Desulfobacteraceae 28 232 90 42 136 304 127 380 Desulfatibacillum 3 Desulfofaba 2 1 Desulfoluna 6 2 7 2 8 61 4 4 Desulfonema 1 85 28 9 25 20 16 157 Desulforegula 19 21 5 2 13 205 13 37 Desulfosarcina 2 Desulfospira 2 14 11 24 1 56 46 Desulfatiferula 18 7 6 13 12 7 Desulfatirhabdiu 2 33 7 4 21 3 10 10 m Desulfobacter 1 3 Desulfobacterium 4 1 1 3 Desulfobacula 12 4 17 2 23 21 Desulfobotulus 45 5 1 10 3 91 Desulfocella 3 3 3 3 Desulfococcus 12 1 1 1 Desulfobulbaceae 23 58 74 64 60 19 17 11 Desulfobulbus 1 Desulfocapsa 21 42 63 48 42 13 6 3 Desulfofustis 1 5 5 13 17 2 1 1 Desulfopila 1 1 Desulforhopalus 7 3 3 4 10 5 Desulfurivibrio 4 3 1 Desulfovibrionales 39 18 13 15 2 7 14 Desulfohalobiaceae 3 Desulfonatronovibrio 2 Desulfovermiculus 1 Desulfovibrionaceae 36 18 13 15 2 7 14 Bilophila 3 Desulfocurvus 17 17 12 15 4 10 Lawsonia 16 1 1 2 3 4 Desulfurellales 82 6 1 11 4 10 100

Phylum Class Order Family Genus 191L 191G 191S 194 196 198 201 203 Desulfurellaceae 82 6 1 11 4 10 Hippea 82 6 1 11 4 10 Desulfuromonadales 31 66 60 9 29 239 35 89 Desulfuromonadaceae 4 5 3 4 1 1 Desulfuromonas 2 Malonomonas 2 5 3 4 1 1 Geobacteraceae 31 62 55 6 25 239 34 88 Geoalkalibacter 2 1 Geobacter 30 6 10 1 103 20 5 Geopsychrobacte 1 51 42 6 16 134 10 77 r Geothermobacter 3 3 7 2 4 6 Myxococcales 6 507 114 61 99 33 164 356 Haliangiaceae 4 35 128 Haliangium 4 35 128 Cystobacteraceae 3 63 46 15 25 8 11 46 Anaeromyxobacter 44 30 11 12 6 2 32 Archangium 5 4 3 5 1 1 2 Hyalangium 3 6 9 1 7 1 8 7 Melittangium 8 3 1 5 Kofleriaceae 4 1 2 1 3 23 Kofleria 4 1 2 1 3 23 Myxococcaceae 8 4 1 2 1 1 Pyxidicoccus 8 4 1 2 1 1 Nannocystaceae 3 21 14 14 2 21 Enhygromyxa 1 Nannocystis 2 7 13 11 19 Plesiocystis 1 14 1 3 2 1 Phaselicystidaceae 1 88 10 3 8 2 11 5 Phaselicystis 1 88 10 3 8 2 11 5 Polyangiaceae 2 341 32 28 48 15 83 153 Byssovorax 26 20 20 23 1 15 33 Chondromyces 4 2 3 1 13 10 Sorangium 2 311 10 8 22 13 55 110 101

Phylum Class Order Family Genus 191L 191G 191S 194 196 198 201 203 Syntrophobacterales 222 62 83 96 5 13 29 Syntrophaceae 139 24 5 31 3 2 12 Desulfobacca 25 2 4 3 1 2 Desulfomonile 5 1 6 7 Smithella 101 23 3 21 1 3 Syntrophus 8 Syntrophobacteraceae 83 38 78 65 2 11 17 Desulfoglaeba 13 1 2 Desulforhabdus 26 30 72 49 2 6 12 Desulfovirga 42 7 4 14 5 5 Syntrophobacter 1 Thermodesulforhabdus 1 2 Epsilonproteobacteria 4413 405 3454 4626 2488 635 707 164 Campylobacterales 4253 381 3245 4439 2331 548 675 136 Hydrogenimonaceae 1286 107 927 1371 724 148 204 17 Hydrogenimonas 1286 107 927 1371 724 148 204 17 Campylobacteraceae 57 24 64 45 30 23 11 4 Arcobacter 3 1 Sulfurospirillum 57 21 64 45 30 22 11 4 Helicobacteraceae 2910 250 2254 3023 1577 377 460 115 Sulfuricurvum 41 11 1 1 2 21 Sulfurimonas 210 45 2 22 3 49 Sulfurovum 2651 174 2241 3018 1566 339 457 32 Wolinella 8 20 10 5 10 14 13 Nautiliales 160 24 209 187 157 87 32 28 Nautiliaceae 160 24 209 187 157 87 32 28 Lebetimonas 2 Nitratifractor 156 12 156 179 121 49 30 14 Nitratiruptor 2 11 52 6 34 38 1 14 Thioreductor 2 1 1 2 1 Gammaproteobacteria 240 86 97 97 78 666 315 911 Enterobacteriales 1 Enterobacteriaceae 1

102

Phylum Class Order Family Genus 191L 191G 191S 194 196 198 201 203 Escherichia/Shigella 1 Legionellales 31 Legionellaceae 31 Legionella 30 Tatlockia 1 Methylococcales 2 1 1 1 8 8 Methylococcaceae 2 1 1 1 8 8 Methylothermus 1 1 1 2 Methylocaldum 1 Methylococcus 2 7 5 Methylosarcina 1 Oceanospirillales 3 22 1 10 5 128 Hahellaceae 1 Endozoicomonas 1 Halomonadaceae 1 Halotalea 1 Oceanospirillaceae 3 1 2 1 Oceanobacter 1 1 Neptuniibacter 1 2 1 Oceaniserpentilla 1 Oceanospirillales_incertae_sedis 1 123 Spongiispira 1 123 Oleiphilaceae 3 19 8 3 3 Oleiphilus 3 19 8 3 3 Pseudomonadales 1 1 12 3 1 11 5 Moraxellaceae 1 12 3 1 11 4 Enhydrobacter 1 12 2 1 11 4 Perlucidibaca 1 Unclassified 1 1 Dasania 1 1 Thiotrichales 229 37 18 87 50 563 275 530 Piscirickettsiaceae 2 3 2 Cycloclasticus 1 1 1

103

Phylum Class Order Family Genus 191L 191G 191S 194 196 198 201 203 Hydrogenovibrio 1 2 Sulfurivirga 1 Thiotrichaceae 229 35 18 84 50 560 274 514 Beggiatoa 8 5 2 9 4 1 50 Leucothrix 1 2 5 Thiothrix 228 27 13 82 41 554 273 459 Thiotrichales_incertae_sedis 1 1 16 Caedibacter 1 16 Fangia 1 Xanthomonadales 3 2 1 4 7 58 Sinobacteraceae 3 2 1 2 6 Alkanibacter 3 2 2 Steroidobacter 1 2 4 Xanthomonadaceae 1 3 5 52 Aquimonas 1 3 31 Lysobacter 1 17 Rudaea 1 Arenimonas 1 1 Aspromonas 1 2 Ignatzschineria 1 1 Acidithiobacillales 1 1 Thermithiobacillaceae 1 1 Thermithiobacillus 1 1 Aeromonadales 1 Succinivibrionaceae 1 Succinivibrio 1 Alteromonadales 1 1 Alteromonadaceae 1 Marinimicrobium 1 Unclassified 1 Teredinibacter 1 Cardiobacteriales 2 3 3 Cardiobacteriaceae 2 3 3

104

Phylum Class Order Family Genus 191L 191G 191S 194 196 198 201 203 Dichelobacter 2 3 3 Chromatiales 1 33 13 7 17 43 10 86 Chromatiaceae Thiococcus 4 7 3 8 5 3 14 Thioflavicoccus 1 1 Thiohalocapsa 1 2 2 1 1 6 Thiophaeococcus 3 1 2 2 1 1 Chromatium 1 1 Thiorhodovibrio 4 1 3 1 Halochromatium 3 Lamprocystis 1 Nitrosococcus 1 1 Rhabdochromatium 1 Ectothiorhodospiraceae 1 Thiorhodospira 1 29 6 3 9 37 7 48 Alkalispirillum 4 1 1 4 1 5 Aquisalimonas 3 1 7 1 Ectothiorhodosinus 4 1 2 Natronocella 15 1 28 6 34 Thioalkalispira 3 3 Granulosicoccaceae 1 3 4 1 6 Granulosicoccus 1 1 7 1 1 7 Halothiobacillaceae 17 Thioalkalibacter 1 Thiofaba 16 Unclassified 7 5 24 3 27 17 64 Unclassified 7 5 24 3 27 17 64 Thiohalophilus 1 5 22 1 12 10 53 Methylonatrum 6 2 15 7 10 Sedimenticola 2 1 Spirochaetes 16 96 44 25 44 19 10 46 Spirochaetes 16 96 44 25 44 19 10 46 Spirochaetales 16 96 44 25 44 19 10 46

105

Phylum Class Order Family Genus 191L 191G 191S 194 196 198 201 203 Leptospiraceae 15 30 24 20 25 11 6 28 Leptonema 12 9 1 7 1 2 23 Turneriella 15 18 15 19 18 10 4 5 Spirochaetaceae 1 33 8 2 12 7 4 14 Spirochaeta 29 2 4 2 3 Treponema 1 4 6 2 8 5 4 11 Unclassified 33 12 3 7 1 4 Exilispira 33 12 3 7 1 4 Synergistetes 52 50 13 27 9 1 Synergistia 52 50 13 27 9 1 Synergistales 52 50 13 27 9 1 Synergistaceae 52 50 13 27 9 1 Aminiphilus 7 3 2 3 1 Thermovirga 2 Aminomonas 4 3 1 1 1 Cloacibacillus 11 Jonquetella 1 Pyramidobacter 27 44 10 24 7 Tenericutes 3 1 Mollicutes 3 1 Anaeroplasmatales 1 Anaeroplasmataceae 1 Asteroleplasma 1 Haloplasmatales 2 1 Haloplasmataceae 2 1 Haloplasma 2 1 Thermodesulfobacteria 25 1 16 1 Thermodesulfobacteria 25 1 16 1 Thermodesulfobacteriales 25 1 16 1 Thermodesulfobacteriaceae 25 1 16 1 Caldimicrobium 9 1 1 1 Thermodesulfatator 16 15 Thermotogae 5 1 2 1

106

Phylum Class Order Family Genus 191L 191G 191S 194 196 198 201 203 Thermotogae 5 1 2 1 Thermotogales 5 1 2 1 Thermotogaceae 5 1 2 1 Geotoga 1 1 Kosmotoga 5 1 1 Verrucomicrobia 170 177 25 123 2 33 46 Opitutae 121 167 25 106 2 30 26 Opitutales 110 163 25 101 2 26 23 Opitutaceae 110 163 25 101 2 26 23 Alterococcus 92 91 23 80 2 16 2 Opitutus 18 72 2 21 10 21 Puniceicoccales 11 4 5 4 3 Puniceicoccaceae 11 4 5 4 3 Cerasicoccus 1 4 2 Coraliomargarita 10 4 1 2 3 Spartobacteria 1 5 Unclassified 4 Unclassified 4 Xiphinematobacter 4 Subdivision3 12 5 8 1 6 Subdivision5 1 Verrucomicrobiae 35 5 4 2 14 Verrucomicrobiales 35 5 4 2 14 Verrucomicrobiaceae 35 5 4 2 14 Akkermansia 6 2 2 1 Persicirhabdus 14 1 1 Verrucomicrobium 15 2 1 2 13 Candidate Division OD1 28 48 6 20 5 10 22 Candidate Division OP10 3 2 2 2 13 9 Candidate Division OP11 1 Candidate Division SR1 85 3 10 19 215 9 20 Candidate Division TM7 1 Candidate Division WS3 13 9 2 1 5

107

Phylum Class Order Family Genus 191L 191G 191S 194 196 198 201 203 Candidate Division BRC1 5 3 2 2

Ktedonobacteria 1 Unclassified Ktedonobacteria 1 Ktedonobacterales 1 Ktedonobacteraceae 1 Ktedonobacter 1 Total Sequences 5301 4876 5954 5643 4589 4543 3990 4635

108

VITA

Audrey Tarlton Paterson was born in London, England, and raised in Houston, Texas. Audrey graduated from Episcopal High School in Bellaire, Texas, in 2004. With a strong interest in environmental sciences, Audrey declared biology as her major discipline upon enrollment at

Louisiana State University. Audrey later developed an interest in geology and began an independent study course with Dr. Annette Summers Engel. In December 2009, she earned dual

Bachelor of Science degrees in Biology and Geology from Louisiana State University. Audrey continued work as a Contingent Research Assistant with Dr. Engel during the spring of 2010, and later that year, Audrey enrolled in the Master’s program with the Department of Geology and Geophysics.

109