Open Collections

UBC Theses and Dissertations

UBC Theses Logo

UBC Theses and Dissertations

Microbial communities in field-based biological reactors treating mining-influenced water Rezadehbashi, Maryam 2015

Your browser doesn't seem to have a PDF viewer, please download the PDF to view this item.

Item Metadata

Download

Media
24-ubc_2015_november_rezadehbashi_maryam.pdf [ 35.28MB ]
Metadata
JSON: 24-1.0165837.json
JSON-LD: 24-1.0165837-ld.json
RDF/XML (Pretty): 24-1.0165837-rdf.xml
RDF/JSON: 24-1.0165837-rdf.json
Turtle: 24-1.0165837-turtle.txt
N-Triples: 24-1.0165837-rdf-ntriples.txt
Original Record: 24-1.0165837-source.json
Full Text
24-1.0165837-fulltext.txt
Citation
24-1.0165837.ris

Full Text

Microbial Communities in Field-based Biological Reactors Treating Mining-influenced Water  by  MARYAM REZADEHBASHI   B.Sc., The University of Tehran, 2005 M.Sc., The University of Tehran, 2008   A THESIS SUBMITTED IN PARTIAL FULFILLMENT OF THE REQUIREMENTS FOR THE DEGREE OF   DOCTOR OF PHILOSOPHY  in  THE FACULTY OF GRADUATE AND POSTDOCTORAL STUDIES (Chemical and Biological Engineering)   THE UNIVERSITY OF BRITISH COLUMBIA  (Vancouver)  October 2015  © MARYAM REZADEHBASHI, 2015 	  	   ii	  ABSTRACT  	  Metals, sulfate, nitrate and ammonia are the main chemical constituents of mining-influenced water (MIW). Biological reactors are attractive for low-cost treatment of MIW. But their successful performance relies on having microbial consortia with the metabolic potential to remove the contaminants of concern. This study undertook to explore the microbial communities in two types of field-based systems that successfully treat MIW: Semi-passive biochemical reactors (BCR) removing metals and sulfate; Active biological reactors removing ammonia and nitrate in the presence of metals. Pyrotag sequencing of 16S rRNA genes was performed for 80 samples from four BCRs. Although they were located at different mine sites, the BCRs shared several taxonomic groups. Sulfate-reducing microorganisms (SRM) were restricted to Deltaproteobacteria and only a few Clostridium genera. Core SRM were specialized, often poorly characterized genera, also prevalent at other metal-contaminated sites. The BCRs were populated by both acetate-consuming SRM and methanogens. SRM were more prevalent in some BCRs than methanogens, which are potential competitors. The structure of the microbial community in a BCR containing pulp and paper biosolids as carbon source was different from that in the other BCRs. Network analysis revealed that the putative keystone microorganisms in this BCR were phylogenetically different from those in the other BCRs. According to correlation analysis of these keystone microorganisms, a hypothetical model proposed that keystone microorganisms in BCR3 employ different mechanisms (antagonistic) than keystones in the other BCRs to regulate the structure of microbial community.  The microbial communities within mine nitrogen-removing bioreactors shared several taxonomic groups with those in non-mine related nitrogen-removing facilities. The mine system selected for archaeal (rather than bacterial) ammonia oxidizer and denitrifying populations. The denitrifying population was enriched and a metagenomic study revealed genes encoding enzymes involved in metal metabolism (e.g. arsenite oxidation, mercury reduction) that were related to denitrifiers such as Rhodobacter sp., Thiobacillus sp., Burkholderia sp., Methylotenera sp., and Variovorax sp. Sequences related to taxa capable of aerobic denitrification, autotrophic denitrification, nitrifier denitrification, and anammox were found. The microbiology of the influent water had a significant impact on the microbial composition of the nitrogen-removing bioreactors. 	  	   iii	  PREFACE  Prof. Susan Baldwin supervised the project, contributed to the organization and communications for visiting the sites, collecting samples and gathering chemistry data. She assisted in analysis of sequences and interpretation of the data. She also contributed largely to the revision of dissertation. Prof. Susan Baldwin, Dr. Maryam Khoshnoodi and Dr. Marcus Taupp performed sampling in the Trail Wetland. Prof. Susan Baldwin, Dr. Parissa Mirjafari collected samples from Mount Polley Mine. Mike Polywkan collected samples from Nickel Plate Mine treatment facility. Annual water quality of influent and effluent of all the BCRs in this study were measured by personnel of mine site. Chemistry data for the Highland Valley Copper treatment system were extracted from several annual reports that were written by Heather Larratt. Illumina Mi-seq sequencing was done by Jon Taylor at the UBC Sequencing Centre at Pharmaceutical Sciences (http://sequencing.ubc.ca/). Roche GS-FLX Titanium sequencing of samples was done at Genome Quebec sequencing facility. Dr. Kishori Konwar provided the scripts that were used in Chapter 3.   Publications and Presentations:  My contribution to the following publication was preparing the samples for Roche GS-FLX Titanium pyrotag sequencing, comparing the 16S rRNA clone library sequences to those in the NCBI database and building of phylogenetic trees.  • Susan A. Baldwin, Maryam Khoshnoodi, Maryam Rezadehbashi, Marcus Taupp, Steven Hallam, All Mattes and Hamed Sanaei. The microbial community of a passive biochemical reactor treating arsenic, zinc, and sulfate-rich seepage. Frontiers in Bioengineering and Biotechnology. Volume 3, Article 27, March 2015.  A version of Chapter 3 has been prepared for submission. • Maryam Rezadehbashi, Susan A. Baldwin. Diversity of sulfate-reducing microorganisms in metal-removing semi-passive biochemical reactors and their co-occurrence with methanogens.    	  	   iv	  The outcome of this work has been published in international conferences as follows;  • Maryam Rezadehbashi, Susan A. Baldwin. “The co-occurrence of sulfate-reducing bacteria and methanogens in bioreactors treating metal and sulfate containing mine-affected water”. The 114th General Meeting of American Society for Microbiologists, Boston-MA, USA, May 17-20, 2014.  • Maryam Rezadehbashi, Susan A. Baldwin. “Analysis of microbial community of two bioreactors treating metal mine tailing seepage water”. The 4th World Congress on Biotechnology, North Carolina, USA, September 23-25, 2013.  • Maryam Rezadehbashi, Maryam Khoshnoodi, Marcus Taupp, Steven Hallam and Susan A. Baldwin. “Microbial diversity of a biochemical reactor removing metals from smelter waste leachate”. Canadian Society for Microbiologists Annual Conference, Vancouver, Canada, June 20-23, 2012.  I was a co-author in the following conference abstracts:  • Susan A. Baldwin, Parissa Mirjafari, Maryam Rezahdebashi, Gaurav Subedi, Jon Taylor, Luke Moger, Katie McMahen and Art Frye. Start-up of a passive remediation bioreactor for sulfate and selenium removal from mine tailings water. 10th International Conference on Acid Rock Drain. Santiago, Chile, April 21-24, 2015.   • Susan A. Baldwin, Maryam Rezadehbashi. “Seasonal microbial community changes in pulp mill biosolids submerged in metal and sulfate-rich leachate”. 5th International Conference Microbial Ecology and Water Engineering, Michigan, USA, July 7-10, 2013.   • Maryam Khoshnoodi, Maryam Rezadehbashi and Susan A. Baldwin. “Correlation of the Microbial Diversity in a biochemical reactor treating metal-rich water with environmental factors”. Canadian Society for Microbiologists Annual Conference, Vancouver, Canada, June 20-23, 2012.   • Susan A. Baldwin, Maryam Rezadehbashi, Marcus Taupp, Al Mattes, and Steven J. Hallam. “The genetic diversity of microbes in an anaerobic biological reactor treating metal-rich landfill leachate: What this tells us about treatment mechanisms”. 9th International Conference on Acid Rock Drainage, Ottawa, Canada, May 20-24, 2012.  • Parissa Mirjafari, Maryam Rezadehbashi and Susan, A. Baldwin. “Microbial communities associated with passive treatment of sulfate and selenium containing water”. 9th International Conference on Acid Rock Drainage, Ottawa, Canada, May 20-24, 2012. 	  	   v	  TABLE OF CONTENTS 	   ABSTRACT ......................................................................................................................................... ii PREFACE ........................................................................................................................................... iii TABLE OF CONTENTS ................................................................................................................... v LIST OF TABLES ........................................................................................................................... viii LIST OF FIGURES ........................................................................................................................... ix NOMENCLATURE ......................................................................................................................... xiv ACKNOWLEDGEMENTS ............................................................................................................. xv DEDICATION ................................................................................................................................. xvi CHAPTER 1: INTRODUCTION AND LITERATURE REVIEW ............................................... 1 1.1 Mining-influenced Water ........................................................................................................................ 1 1.2 Physical and Chemical Technologies for Treating Mining-influenced Water ................................... 2 1.3 Biochemical Reactors for Removal of Heavy Metals and Sulfate from MIW ................................... 4 1.4 Active Biological Reactors for Removal of Nitrogen Compounds from MIW .................................. 6 1.5 Microbially-mediated Processes in BCRs for Removal of Metals and Sulfate .................................. 7 1.6 Microbially-mediated Processes in Active Bioreactors for Removal of Nitrogen Compounds ...... 10 1.7 Microorganisms in BCRs Used for Removal of Metals and Sulfate ................................................. 12 1.8 Association of Sulfate-reducing Microorganisms with Methanogens ............................................... 14 1.9 Microbial Communities, Factors controlling Community Properties, Microbial Interactions, and Keystone Microorganisms ........................................................................................................................... 15 1.10 Emerging Methods for Studying Microbial Communities in Bioreactors ..................................... 17 1.10.1 High-throughput DNA Sequencing ................................................................................................ 17 1.10.2 Network Analysis Tools for Exploring Microbial Interactions and Keystone Species ................. 17 1.11 Knowledge Gaps and Thesis Objectives ............................................................................................ 22 1.12 Thesis Layout ....................................................................................................................................... 24 CHAPTER 2: SITE DESCRIPTION AND PERFORMANCE HISTORY OF THE FIELD-BASED BIOREACTORS SURVEYED IN THIS STUDY ........................................................... 26 2.1 Introduction ............................................................................................................................................ 26 2.2 Highland Valley Copper (Mine Site 1: BCR1 and BCR2 and the Algae Pond) .............................. 28 2.2.1 BCR1 and the Algae Pond ................................................................................................................ 28 2.2.2 BCR2 for Removal of Metals ........................................................................................................... 30 2.2.3 Inoculum Pond (IP) .......................................................................................................................... 31 2.3 Trail Wetland (Mine Site 2: BCR3) ..................................................................................................... 31 2.4 Mount Polley Mine (Mine Site 3: BCR4) ............................................................................................. 33 	  	   vi	  2.5 Nickel Plate Mine (Mine Site 4; Active Biological Reactors for Removal of Ammonia and Nitrate) ....................................................................................................................................................................... 33 CHAPTER 3: COMPARSION OF SEVERAL METAL-REMOVING BIOCHEMICAL REACTORS IN TERMS OF THEIR MICROBIAL POPULATIONS ...................................... 36 3.1 Synopsis .................................................................................................................................................. 36 3.2 Materials and Methods .......................................................................................................................... 37 3.2.1 Description of Biochemical Reactors and Sampling ........................................................................ 37 3.2.2 DNA Extraction, Amplification of 16S rRNA Genes, and Pyrosequencing .................................... 38 3.2.3 Sequence Analysis ............................................................................................................................ 39 3.2.4 Analysis of Co-occurrence and Network of Correlations ................................................................ 40 3.3 Results ..................................................................................................................................................... 41 3.3.1 Physiochemical Properties of the BCRs ........................................................................................... 41 3.3.2 Overall Microbial Community Structure ......................................................................................... 42 3.3.3 Distribution of Sulfate-reducing Microorganisms ........................................................................... 45 3.3.4 Distribution of Methanogen-related Sequences in Bioreactors ........................................................ 50 3.3.5 Network of Co-occurrence for the SRM- and Methanogen-related Taxa ........................................ 53 3.4 Discussion ............................................................................................................................................... 55 3.4.1 Metal-removing BCRs Share Several Microorganisms Even Though They Are at Different Geographical Locations ............................................................................................................................. 55 3.4.2 The Type of Carbon Source Used in BCRs Affects the Composition of the Microbial Community ................................................................................................................................................................... 56 3.4.3 BCRs are Distinct From Other Anaerobic Environments in Terms of Their SRM Community ..... 57 3.4.4 Sulfate-reducing Microorganisms Are Prevalent in Metal-removing Bioreactors More so Than Methanogens That Are Potential Competitors for Carbon Sources .......................................................... 59 CHAPTER 4: KEYSTONE MICROORGANISMS IN BIOCHEMICAL REACTORS TREATING MINING-INFLUENCED WATER .......................................................................... 61 4.1 Synopsis .................................................................................................................................................. 61 4.2 Materials and Methods .......................................................................................................................... 62 4.2.1 Dataset Preparation ........................................................................................................................... 62 4.2.2 Analysis of Co-occurrence and Construction of Network of Correlations ...................................... 62 4.2.3 Determination of Keystone Taxa in Each Network ......................................................................... 63 4.3 Results ..................................................................................................................................................... 63 4.3.1 Summary of Network Statistics for BCR3 Compared with the Other BCRs ................................... 63 4.3.2 Keystone Microbial Taxa in BCR3 and Their Correlations with Other Taxa .................................. 64 4.3.3 Keystone Microbial Taxa in Other BCRs and Their Correlations with Other Taxa ........................ 67 4.3.4 Relationship between Relative Abundance and the Values of Betweenness Centrality and Closeness Centrality .................................................................................................................................. 70 4.4 Discussion ............................................................................................................................................... 71 4.4.1 Limitations of the Study ................................................................................................................... 71 4.4.2 BCR3 Versus Other BCRs; Keystone Microorganisms and Their Correlations with Other Taxa .. 72 4.4.3 Model Describes Different Mechanisms that Regulate the Structure of Microbial Communities in Each Ecosystem ......................................................................................................................................... 74 	  	   vii	  CHAPTER 5: MICROBIAL COMMUNITIES IN NITROGEN-REMOVING BIOLOGICAL REACTORS TREATING MIW ...................................................................................................... 76 5.1 Synopsis .................................................................................................................................................. 76 5.2 Materials and Methods .......................................................................................................................... 78 5.2.1 Description of Nitrogen-removing Bioreactors and Sampling ......................................................... 78 5.2.2 DNA Extraction, Amplification of 16S rRNA Genes and Pyrosequencing ..................................... 79 5.2.3 Analysis of Sequences from Pyrotag Sequencing of the 16S rRNA Gene ...................................... 80 5.2.4 Enrichment of Metal Oxidizing Denitrifying Microorganisms ........................................................ 81 5.2.5 Whole Genome DNA Extraction ..................................................................................................... 82 5.2.6 MiSeq Library Preparation, Quality Control and Shotgun Sequencing ........................................... 83 5.2.7 Quality Control Checking, Assembly, Genome Annotation and Phylogenic Assignment .............. 83 5.3 Results ..................................................................................................................................................... 83 5.3.1 Performance of Bioreactors .............................................................................................................. 83 5.3.2 Phylogenetic Diversity of Microorganisms in Nitrogen-removing Bioreactors .............................. 84 5.3.3 Distribution and Diversity of Key Functional Groups ..................................................................... 90 5.3.4 Influent Water, its Seasonal Changes and its Impact on the System ............................................... 95 5.4 Discussion ............................................................................................................................................. 100 5.4.1 Nitrogen-removing Bioreactors Selects for some Microorganisms that are Adapted to this Particular Ecosystem ............................................................................................................................... 100 5.4.2 Microorganisms That are Capable of Ammonia and Nitrate Removal by Alternative Pathways are Present in Conventional MIW Nitrogen-removing Bioreactors .............................................................. 105 5.4.3 Seasonal Changes in the Chemistry and Microbiology of Influent Water Affects the Microbial Community Composition of the Bioreactors ........................................................................................... 106 CHAPTER 6: CONCLUSIONS .................................................................................................... 107 6.1 Overall Conclusions ............................................................................................................................. 107 6.1.1 Microbial Communities of Metal-removing BCRs Treating MIW ................................................ 107 6.1.2 Factors Regulating the Structure of Microbial Communities in BCRs .......................................... 108 6.1.3 Microbial Communities of Nitrogen-removing Bioreactors Treating MIW .................................. 108 6.1.4 Summary of Metabolically Important Microorganisms ................................................................. 109 6.2 Originality and Contributions to the Field ........................................................................................ 110 6.3 Limitations of the Research and Recommendations for Future Work .......................................... 112 REFERENCES ................................................................................................................................ 114 APPENDIX A .................................................................................................................................. 138 APPENDIX B .................................................................................................................................. 151      	  	   viii	  LIST OF TABLES 	  	  	  Table 1.1: Chemical and physical technologies for removal of heavy metals from wastewater……...3  Table 2.1: Selected properties of the BCRs studied……………………………………………....…28  Table 2.2: Chemical characteristic of influent water…………………………..…………………….35  Table 3.1: Physiochemical properties of the BCRs…………………………………………...……..41  Table 3.2: Diversity indices of samples used in this study. These were determined using the OTU tables based on 97%, 94% and 90% sequences similarity. ………………………..…………..…….42  Table 4.1: Description of datasets that were used to construct the co-occurrence networks…….…..64  Table 5.1: List of samples that were taken from full- and pilot-scale nitrogen-removing bioreactors in Nickel Plate mine……………………………………………………………………………….…79  Table 5.2: Performance of nitrification and denitrification bioreactors for removal of contaminant of concern……………………………………………………………………………………………….84  Table 5.3: Diversity indices of samples used in this study. These were determined using the OTU tables based on 97%, 94% and 90% sequences similarity……………………………………….…..85  Table 5.4: Potential denitrifying taxa based on 24 studies that have surveyed 16S rRNA genes and/or nirK and nirS genes from different environments. ………………………………………………….93  Table 5.5: Denitrifying-related taxa in full- and pilot-scale denitrification bioreactors by Pyrotag sequencing of variable region V6-V8 of 16S rRNA gene…………………….……………………..93  Table 5.6: Table showing ORFs that encode enzymes for denitrification reactions and their related microorganisms based on SEED database……………………………………………...…………....94   Table 5.7: Average physiochemical properties of the influent water during spring and winter…….95  	  	  	   ix	  LIST OF FIGURES 	  	  	  Figure 1.1: Schematic diagram of a semi-passive, up-flow biochemical reactor (BCR) used for removal of metals and sulfate……………………………………………………………………..…..5  Figure 1.2: Microbially-mediated processes in semi-passive BCRs for removal of metals and sulfate…………………………………………………………………………………………...……..7   Figure 1.3: Microbially-mediated processes in active bioreactors for removal of nitrogen compounds…………………………………………………………………………………………...10  Figure 1.4: Biotic and abiotic factors controlling the structure of microbial communities……….....15  Figure 1.5: A) Figure showing similarity in abundance pattern for OTU 1 and OTU 2. B) Figure showing nodes with high betweenness and closeness centralities in a network……………………..18  Figure 2.1: Map of British Columbia showing the location of each BCR. Photos of bioreactors that are operating in each location are presented……………………………………..…………………..27  Figure 2.2: A) Aerial view of Highland Valley Copper Mine showing the location of BCR1 (indicated by the red circle). B) Schematic diagram showing the actual location of the algae pond (AP) and BCR1 within the S5 treatment system. …………………………………………………...29   Figure 2.3: Influent and effluent concentrations of dissolved molybdenum and copper in the S5 treatment system at Highland Valley Copper. ……………………….…………………………...…29   Figure 2.4: Influent and effluent concentrations of sulfate in the S5 treatment system……………. 30   Figure 2.5: Influent and effluent concentrations of dissolved molybdenum and copper in BCR2 at Highland Valley Copper………………………………………………………………………….….30   Figure 2.6: Inoculum pond…………………………………………………………….………..……31   Figure 2.7: Influent and effluent concentrations of dissolved arsenic and zinc in BCR3…………...32   Figure 2.8: Influent and effluent concentrations of sulfate in BCR3……………………………..…32   Figure 2.9: Influent and effluent concentrations of dissolved selenium and sulfate in BCR4………33  	  	   x	  Figure 2.10: Process flow diagram showing the nitrification and denitrification circuits (bioreactors and settling chambers). Photographs are of the indicated bioreactors taken at the Nickel Plate Mine. ………………………………………………………………………………………………………..34  Figure 2.11: Average annual influent and effluent concentrations of ammonia and nitrate in the nitrification and denitrification circuits of the Nickel Plate Mine bioremediation system…………..35  Figure 3.1: Three-dimensional principal coordinate analysis based on unweighted UniFrac distances       between samples from different bioreactors. Axis 1 explained 28.92% of variation, axis 2, 8.34%, and axis 3 explained 4.48% of variation. Separation of microbial diversity between BCR3 versus other BCRs was observed.…………………………………………………………………………...43  Figure 3.2: Phylum-level taxonomic summary of 94% homology cut-off pyrotag reads. Phyla arranged based on the color from the bottom to the top, starting with Acidobacteria. Vertical axis represents percent reads, and the horizontal axis belongs to BCRs and environmental samples……44  Figure 3.3: A) Family-level (90% homology cut-off) distribution of SRM-related sequences in all BCRs as determined by pyrotag sequencing method. B) Number of reads assigned to different known taxonomic groups (family-level) in each of the features sampled. C) Number of reads assigned to different uncultured taxonomic groups within Deltaproteobacteria (family-level) in each of the features sampled……………………………………………………………………………....46  Figure 3.4: Genus-level heatmap demonstrates relative distribution of SRM-related sequences. OTU numbers and their Silva 111 assigned taxonomy are given on the right hand side. Sample names are column labels. It constructed by clustering the sequences in to OTUs based on 94% similarities. The relative abundance for each OTU in different sites is colored in shades of white (low relative abundance) to purple (high relative abundance)……………………………………………………..48  Figure 3.5: Phylogenetic dendrogram of core SRM-related communities revealed from sequencing of V6-V8 of 16S rRNA with their neighbors from NCBI database. The bar size is proportional to the number of sequences in each OTU in each bioreactor. The sequences clustered in to OTUs based on 97% homology…………………………………………………………………………………….....50 Figure 3.6: A) Family-level distribution of methanogen-related sequences in all BCRs combined. B) Average number of reads per sample assigned to different family-level taxonomic groups in each of the features sampled (based on the 90% homology cut-off OTU table)……………………………..51  Figure 3.7: Genus-level heatmap demonstrates relative distribution of methanogen-related sequences. OTU numbers and their Silva 111 assigned taxonomy are given on the right hand side. Sample names are column labels. It constructed by clustering the sequences in to OTUs based on 94% similarities………………………………………………………………………………….…..52 	  	   xi	   Figure 3.8: Phylogenetic dendrogram of core methanogen-related communities revealed from sequencing of V6-V8 of 16S rRNA with their neighbors from NCBI database. The circle size is proportional to the number of sequences in each OTU in each bioreactor. The sequences clustered in to OTUs based on 97% homology……………………………………………………………….…..53  Figure 3.9: Network of coexisting microbial lineages. Each square (node) represents a bacterial or archaeal 97% OTU. Color of the nodes represents their phylogenetic affiliations. Lines connecting two taxa indicate a significant co-occurrence relationship based on Pearson correlation coefficients (r) (r ≥ |0.8| and P-value ≤ 0.01). Pearson's correlations ≥ |0.8| were considered as connections to increase the confidence for detecting only robust co-occurring associations. The edge length is proportional to the value of correlation coefficient. Node size is proportional to the number of sequences in that OTU. The sequences clustered in to OTUs based on 97% homology…………....54  Figure 4.1: A) Co-occurrence network showing significant correlations of microbial taxa (based on 94% homology cut-off OTU-table) in BCR3. Lines connecting two microbial taxa represent the correlations between them (r ≥ |0.8| and p-value ≤ 0.05). The sizes of nodes in the network are proportional to the value of BC and colored based on their phylogenetic affiliation. The network is circular sorted based on the value of BC. The top 5% OTUs that possess the highest value for BC were shown. B) Heatmap showing positive and negative correlations of keystones with other taxa based on the value of r. Each cell colored based on the average of negative and positive correlations (r) that keystones make with corresponding taxa…………………………………………………....65  Figure 4.2: A) Co-occurrence network showing significant correlations of microbial taxa (based on 94% homology cut-off OTU-table) in BCR3. Lines connecting two microbial taxa represent the correlations between them (r ≥ |0.8| and p-value ≤ 0.05). The sizes of nodes in the network are proportional to the value of CC and colored based on their phylogenetic affiliation. The network is circular sorted based on the value of CC. The top 5% OTUs that possess the highest value for CC were shown. B) Heatmap showing positive and negative correlations of keystones with other taxa based on the value of r. Each cell colored based on the average of negative and positive correlations (r) that keystones make with corresponding taxa………………………………………………..…..66  Figure 4.3: A) Co-occurrence network showing significant correlations of microbial taxa (based on 94% homology cut-off OTU-table) in Other BCRs. Lines connecting two microbial taxa represent the correlations between them (r ≥ |0.8| and p-value ≤ 0.05). The sizes of nodes in the network are proportional to the value of BC and colored based on their phylogenetic affiliation. The network is circular sorted based on the value of BC. The top 5% OTUs that possess the highest value for BC were shown. B) Heatmap showing positive and negative correlations of keystones with other taxa based on the value of r. Each cell colored based on the average of negative and positive correlations (r) that keystones make with corresponding taxa……………………………………………………68 	  	   xii	   Figure 4.4: A) Co-occurrence network showing significant correlations of microbial taxa (based on 94% homology cut-off OTU-table) in other BCRs. Lines connecting two microbial taxa represent the correlations between them (r ≥ |0.8| and p-value ≤ 0.05). The sizes of nodes in the network are proportional to the value of CC and colored based on their phylogenetic affiliation. The network is circular sorted based on the value of CC. The top 5% OTUs that possess the highest value for CC were shown. B) Heatmap showing positive and negative correlations of keystones with other taxa based on the value of r. Each cell colored based on the average of negative and positive correlations (r) that keystones make with corresponding taxa…………………………………………………....69  Figure 4.5: Relationship between rank relative abundance data from the total number of OTUs at genus level found a) in BCR3 and BC, b) in BCR3 and CC, c) in other BCRs and BC and d) in other BCRs and CC……………………………………………………………………………………...…70  Figure 4.6: Schematic model proposing biotic and abiotic mechanisms that regulate the structure of microbial communities in BCRs……………………………………………………………………..75   Figure 5.1: Schematic diagram of bioreactors using active processes for removal of ammonia and nitrate………………………………………………………………………………………………...78  Figure 5.2: Rarefaction curve of bacterial and archaeal operational taxonomic units (OTUs) in samples from pilot and full-scale bioreactors and influent in winter and spring………………….…86  Figure 5.3: Order-level microbial community of influent water and full- and pilot-scale treatment plants during winter and spring………………………………………………………………………88  Figure 5.4: Heatmap showing the top 10 percent predominant genera (based on 94% OTU-table) in all samples……………………………………………………………………………………………89  Figure 5.5: Distribution of key functional groups in each sample. It is based on OTU-table with 97% homology cut-off…………………………………………………………………………………….91  Figure 5.6: Principal coordinate analysis (PCoA) plots from unweighted UniFrac distance samples from Influent and bioreactors during spring and winter. Samples taken during the winter are depicted by the red squares, and samples taken in the spring by blue closed circles………………………....97  Figure 5.7: The logarithmic heatmap (Log 10) showing species-level (97% OTU-table) distribution of sequences in the winter influent water, nitrification and denitrification bioreactors. OTU numbers and their Silva 111 assigned taxonomy are given on the right hand side. Sample names are column labels…………………………………………………………………………………………………98 	  	   xiii	   Figure 5.8: The logarithmic heatmap (Log 10) showing species-level (97% OTU-table) distribution of sequences in the spring influent water, nitrification and denitrification bioreactors. OTU numbers and their Silva 111 assigned taxonomy are given on the right hand side. Sample names are column labels…………………………………………………………………………………………………99                      	  	   xiv	  NOMENCLATURE 	  	  	  Anammox: Anaerobic Ammonium Oxidation ANFO: Ammonium Nitrate Fuel Oil  AOA: Ammonia-oxidizing Archaea  AOB: Ammonia-oxidizing Bacteria  AP: Algae Pond ASFF: Aerated Submerged Fixed Film BC: Betweenness Centrality BCR: Biochemical Reactor CC: Closeness Centrality CSTR: Continuously Stirred Tank Reactors DGGE: PCR-denaturing Gradient Gel Electrophoresis  DOC: Dissolved Organic Carbon FISH: Fluorescence in situ Hybridization HVC: Highland Valley Copper Mine ICP-MS: Inductively Coupled Plasma Mass Spectrometry  IP: Inoculum Pond MIW: Mining-influenced Water NCBI: National Center for Biotechnology Information NOB: Nitrite Oxidizing Bacteria ORF: Open Reading Frame OTUs: Operational Taxonomic Units PCR: Polymerase Chain Reaction RBC: Rotating Biological Contactors SCG: Soil-Crenarchaeota-Group SRM: Sulfate Reducing Microorganisms SSU rRNA: Small Subunit Ribosomal RNA  UPGMA: Unweighted Pair Group Method with Arithmetic Mean 	  	   xv	  ACKNOWLEDGEMENTS 	  	   	  My very special thanks go to Professor Susan Baldwin. Working with her has been always an honor, a pleasure, an experience that I will truly treasure. This project was made possible with support of many people who assisted me with the field sampling and collection of data. I acknowledge generous support of Heather Larratt and Highland Valley Copper staff and engineers for their help in organizing the field trips, samplings and mine tours. I appreciate the technical insights they brought into this research. I would like to thank Mike Polywkan and Vanessa Bell from Nickel Plate Mine. They were enormously helpful with organization and communications for visiting the site, collecting samples and chemistry data. I am also thankful to my committee members Dr. Curtis Suttle and Dr. Bhushan Gopaluni for providing valuable feedback on my research proposal. Many thanks to Dr. Steven Hallam and his post-doctoral fellow Kishori Konwar. Kishori was extremely helpful in providing bioinformatics tools for data analysis and customizing them for our work. I also appreciate assistance of Jon Taylor in the laboratory especially with preparation of samples for Illumina sequencing. I thank Dr. Hesam Movassagh, Mahyar Boo Ajdari and Saad Dara for help with field collection of samples and proofreading of my dissertation. Thanks to Dr. Maryam Khoshnoodi for being with me during the start of the project. Thanks my parents and brothers for their unconditional support with my studies. Last but not least, I would like to thank Dr. Hossein Dabiri who helped my dreams come true.              	  	   xvi	  DEDICATION  I dedicate this work to: My mother Khadijeh, My father Ghodratollah and Dr. Hossein Dabiri. Thank you all for giving me a chance to prove and improve myself.                        	  	   1	  CHAPTER 1 INTRODUCTION AND LITERATURE REVIEW  1.1 Mining-influenced Water Mining-influenced water (MIW) is defined as any water whose chemical composition has been affected by mining or mineral processing. It often contains metals, sulfate, ammonia and nitrate in concentrations above regulatory standards. By oxygenation of sulfide-bearing minerals (via exposure to air and water during mining) sulfuric acid containing water may be produced. Acidic water facilitates dissolution of sulfide minerals, thereby introducing metals into the aqueous phase [109]. Ammonium and nitrate ions result from the widespread use of ammonium-nitrate-fuel oil (ANFO) as a blasting agent, and the use of other nitrogen-containing reagents such as cyanide or amines. Discharging MIW that contains high concentrations of these chemical compounds can have detrimental impacts on receiving ecosystems. For example, mining operations at the Giant Mine (a large gold mine located in Yellowknife) created 8 open pits, 4 tailing ponds and 325,000 m3 of arsenic contaminated soils (http://www.nnsl.com/frames/newspapers/2006-07/jul10_06g.html). At this mine, there are 15 underground chambers containing a total of 237,000 tonnes of arsenic trioxide that is enough to kill every person on the planet a couple times over. Large amounts of arsenic released into Yellowknife Bay during flooding will have devastating impacts on the environment and the health of people living nearby (http://www.nnsl.com/frames/newspapers/2006-07/jul10_06g.html). Mining for copper from the Britannia Mine (British Columbia) stopped in 1974. However, large amounts of MIW containing copper, cadmium, iron and zinc were created because of the runoff and rainwater that flowed through the mine’s abandoned tunnels. This effluent made the area around Britannia Beach extremely polluted affecting 4.5 million salmons. Prior to 2001, as much as 450 kilograms of copper was entering Howe Sound daily [265]. There is now a chemical treatment process at the site that needs to be run in perpetuity.  	  	   2	  The total ammonia concentration in mine effluents varies from 10 to 40 mg/l [128]. The concentration of metals is site specific and the total concentration of nitrate varies widely from 25 to 300 mg/l [128].  Based on Canadian water quality guidelines, discharge limits for ammonia and nitrate are 1 and 2.9 mg/l, respectively [276, 277]. The aquatic life discharge limits for metals: zinc, copper, cadmium, arsenic and molybdenum are 0.03 mg/l, 0.006 mg/l, 0.0002 mg/l, 0.05 mg/l and 2 mg/l, respectively [266]. The discharge limit for sulfate is varied and depends on the hardness of the receiving water [267]. Because MIW at many mine sites contains compounds that are above these regulated water quality guidelines, it must be treated before discharge into the environment. This can be a very expensive undertaking due to the large volumes of water used at mine sites. Thus, many treatment systems have been developed based on chemical and biological mechanisms. In the following literature review, physical and chemical treatment processes are described briefly highlighting their advantages and disadvantages. This is followed by a detailed review of biological treatments used for MIW since these were the focus of this thesis. What is known about the microbially-mediated mechanisms responsible for successful biological treatment is described comprehensively. Finally, how molecular biology techniques can be used to study these systems is discussed leading to the overall research questions and hypotheses for this work. 1.2 Physical and Chemical Technologies for Treating Mining-influenced Water Physical and chemical methods have been developed for removal of constituents of concern from MIW as summarized in Table 1.1 [82, 132]. The Table also presents advantages and shortcomings of each technology, and gives some examples of where they have been applied.        	  	   3	  Table 1.1: Chemical and physical technologies for removal of heavy metals from wastewater. Treatment Technique Advantages Disadvantages Application: Removal of Removal Efficiency (%) Chemical precipitation Simplicity, low cost and ease of pH control Generates large volumes of relatively low density sludge, extra cost for sludge disposal Cu2+, Zn2+ , Pb2+ 100, >94, >92 [5] Ion exchange No sludge generation, less time consuming Limitation in application of resin for heavy metals Zn2+ 100 [7] Adsorption Broad sources, low-cost and rapid adsorption, removal of heavy metals from low concentration wastewater The separation of biosorbents would be difficult after adsorption Pb2+, Fe2+ 98, 84 -97 [106] Membrane filtration High efficiency High cost, process complexity, membrane fouling and low permeate flux Cd2+ 92–98 [105] Coagulation and Flocculation The produced sludge has good sludge settling and dewatering characteristics Method involves chemical consumption and increased sludge volume generation. Cr3+, Cu2+ ,Pb2+ 79-97 [110] Flotation High metal selectivity, high removal efficiency, high overflow rates, low detention periods, low operating cost and production of more concentrated sludge High initial capital cost, high maintenance and operation costs.  Cr3+  96.2 [168] Electrochemical treatment Rapid and well-controlled that require fewer chemicals, provide good reduction yields and produce less sludge. High initial capital investment and the expensive electricity supply Ni2+, Zn2+ 100, 100 [114] Although all the techniques in Table 1.1 have been used to remove heavy metals, chemical methods are favored by the mining industry because of their simplicity and low cost, and because of the inefficiency of other methods. However, chemical methods also come with shortcomings. For example, chemical precipitation is usually adapted to treat wastewater containing high concentrations of heavy metal ions and is less effective when metal ion concentrations are low. Also, chemical precipitation requires reagents and can produce large volumes of sludge that need to be stored or further treated [132].  	  	   4	  Some technologies combine physical, chemical and biological methods for removal of metals, ammonia, nitrate and sulfate based on active, passive or semi-passive processes. The term active treatment is used when a technology needs ongoing human operation, maintenance and monitoring, and uses external sources of reagents and energy. In contrast, passive treatment does not require frequent human intervention, operation or maintenance. Passive treatment usually employs natural construction materials (e.g. soils, clays, rock), natural treatment media (e.g. plant-derived residues such as wood chips, manure, compost), and may also promote growth of natural vegetation [69, 95, 99, 278]. Semi-passive treatment systems are constructed based on minimizing energy and chemical inputs such as by using complex organic carbon sources that are waste materials locally available, but they require some human intervention to add additional components or for monitoring [69, 268]. The common purpose of active, passive and semi-passive processes is to raise the pH, lower dissolved metal, sulfate, ammonia and nitrate concentrations [124]. Some examples of MIW treatment technologies that are based on active, passive or semi-passive processes are, fluidized bed bioreactors, packed bed bioreactors, constructed wetlands, phytotechnologies, permeable reactive barriers and biochemical reactors (BCRs) [69, 92, 116, 158].  1.3 Biochemical Reactors for Removal of Heavy Metals and Sulfate from MIW  Biochemical reactors (BCRs) are sub-surface flow, semi-passive systems that treat MIW by physical, chemical and biological methods [95, 269, 278]. Their design are usually site specific and influenced by several factors such as available space and required retention time [95].  Retention or residence time is the period of the time that the water to be treated must reside within the BCR. Retention time usually influences the size of the reactor. BCR design is also controlled by pH, flow, temperature, and the type and concentration of constituents of concern.  In general, BCRs are at least 3–4 feet deep. Where freezing of the system is a concern, they are usually deeper (6–8 feet). The configuration of BCRs can be either up-flow with bottom water inflow or down-flow with top inflow [270]. There are usually three geochemical zones in a biochemical reactor: an oxidizing zone at the influent, followed by an oxic-anoxic transition zone and finally an oxygen-depleted zone where most metal removal is deemed to take place due to precipitation with produced sulfides (Figure 1.1). Organic material within the BCR supports the growth and activity of microorganisms. Within the anaerobic zone, the major reactions of sulfate-reduction and metal precipitation occur [69, 95, 268, 270, 275]. 	  	   5	   Figure 1.1: Schematic diagram of a semi-passive, up-flow biochemical reactor (BCR) used for removal of metals and sulfate. Both geochemical and biochemical reactions take place in a BCR to achieve metal removal, but sustainable and successful treatment relies on the activity of microorganisms. BCRs have been effective for removing a variety of metals, including copper, arsenic, molybdenum, selenium and zinc [10]. BCRs use complex organic materials (e.g., wood chips, hay, pulp and paper biosolids, manure, for example) as carbon sources for microbes. Complex organic materials are usually waste from other industries and freely available to mining companies. Therefore, the costs for operation for passive BCRs are lower than those for active bioreactors requiring the addition of defined chemical reagents. Using defined chemical reagents (e.g., glucose, ethanol and lactate) in BCRs accelerate the microbial processes for metal and sulfate removal. But, it is expensive. Therefore, application of complex organic materials in BCRs would be better alternative. In general, complex organic materials derived from plant are composed of various components such as lignin, cellulose, hemicellulose, chitin, lipid, protein, sugars and some other small molecular components. Several studies have been conducted to determine the most effective and sustainable complex materials for BCR and to correlate certain properties of organic materials to sulfate reduction rate [43, 165, 248, 261]. Their results showed that the application of a mixture of organic materials is more effective than using one particular type [43, 248]. Materials with high proportion of lignin and cellulose that is harder to degradate, achieve slower sulfate reduction rate [44, 86]. In oppose materials that consist of mainly smaller molecular (e.g., lipid, protein) support the major processes in passive BCR [165]. In this system, because the source of nutrients (e.g., carbon, nitrogen, phosphorus) for microbial activity is from complex organic materials, calculation of the decomposition rate of these is needed Anoxic zone Cover layer (water, or soil and plants) Oxic-anoxic transition zone Sulfate                Sul!de + Metals  (Precipitation) Discharge In"ow Mine water "ows upward !"#$%&'(()$*"&+(,(	  	   6	  so as to determine the longevity of effective BCR operation. Determining this requires long-term monitoring studies, during which the possibility of BCR failure of operation might be high due to the difficulty in measuring the state of the organic material inside the BCR.  There are several case studies where BCRs have been used for removal of metals. The Golinsky Copper Mine operated between 1890s and 1930s (http://www.itrcweb.org/miningwaste-guidance/cs21_golinsky.htm). The total concentrations of zinc, copper and cadmium in the mine’s effluent ranged from 0.1 to 78 mg/l. Federal effluent limits for copper, zinc and cadmium at that time were 0.3, 1.5, and 0.1 mg/l, respectively. Removal of 99% of cadmium, copper, and zinc were achieved by using three BCRs using organic material. Furthermore, the Stowell Mine is an abandoned copper mine (http://www.itrcweb.org/miningwaste-guidance/cs23_stowell.htm). Contaminants of concern in the water that was generated at this site include copper, cadmium, and zinc. Removal of 99% of metals of concerns we achieved in a full-scale BCR. Other example of using biochemical reactors for treatment of MIW is Copper Basin Mining Site in southeastern Tennessee. MIW at these site contained acidity and high concentrations of sulfate and metals, including aluminum, arsenic, cadmium, copper, iron, nickel, selenium, and zinc. Passive or semi-passive BCRs at this site were constructed as trenches installed between the mine waste rock pile and receiving environment. (http://www.itrcweb.org/miningwaste-guidance/cs2_copper_basin.htm).   1.4 Active Biological Reactors for Removal of Nitrogen Compounds from MIW An active biological removal of nitrogen compounds from water was first installed in 1981 at Chateau-Landon in France; and the technology is now used at many mine sites [203]. The types of biological reactors used for nitrogen compound removal include fluidized bed reactors, continuously stirred tank reactors (CSTR), packed bed reactors, suspended sludge reactors, aerated submerged fixed film (ASFF) reactors, and rotating biological contactors (RBC) [98, 128, 169, 232, 247]. Typically nitrogen-removing systems consist of a series of different types of bioreactors. Some provide processes for removal of ammonia while other bioreactors are designed for removal of nitrate.  Defined carbon sources such as methanol, ethanol, acetic acid or a sugar-based nutrient medium are used as carbon sources.  One example where active biological reactors are used for removal of ammonia and nitrate is at the Nickel Plate Mine in British Columbia (Canada). At this site, 99% removal of ammonia and 95% 	  	   7	  removal of nitrate are achieved using methanol as the carbon source for denitrification [90].The application of active biological nitrogen-removal bioreactors is not limited to MIW and they have been used for removal of nitrogen components from drinking water and wastewaters generated in several different sectors (e.g. agricultural and municipal) [2].  1.5 Microbially-mediated Processes in BCRs for Removal of Metals and Sulfate Several microbial-mediated process take place within BCRs to meet their treatment objectives (Figure1.2) [84, 181].  Figure 1.2: Microbially-mediated processes in semi-passive BCRs for removal of metals and sulfate. O: oxidized, R: reduced.  !"#$%&%'(#)'(*+!"#$%&'()*&+&,()-%.&/%0010#2%)!%,%&($*+-$.",/#+0#/1*++0'#%2%'*+0#(3"3(+45++6+7-5+Primary Fermentative Microorganisms Manganese-Reducing  Microorganisms Metal-Reducing  Microorganisms Iron-Reducing  Microorganisms Sulfate-Reducing  Microorganisms Methanogens 34567) 356)8)9%$:0)))))))9%$:053)Secondary Fermentative Microorganisms Nitrate-Reducing  Microorganisms ;4<5)9%$:0)-+ 9%$:0)8+=46)=46)=46)=46)=-7)=46);6)9%$:0)-+9%$:0)-+9%$:0)-+9%$:0)-+9%$:0)8+9%$:0)8+9%$:0)8+9%$:0)8+!"#$%&%'()*+,>%8<) >%86)!"#$%&%'()*+,9') 9'86)9%$:0)4) 9%$:0)?)!"#$%&%'()*+,!"#$%&%'()*+,!"#$%&%'()*+,@AB)CDE)8)BFGH)BFBB)8)BFHB)5)BFHB)@0%/$"#')I#'#"2)	  	   8	  Macromolecules in the complex organic material, such as cellulose, hemicellulose, proteins and lipids are broken down into sugar, amino acid and fatty acid monomers by cellulolytic and primary fermentative microorganisms such as those in the genera Bacteriodes, Ruminococcus, Clostridium, Bacillus, and Cellulomonas, for example. This provides a slow-release and long-term source of electron donors for other microorganisms. Then, these monomers are fermented to organic acids, alcohols, acetate, CO2 and H2 by acetogens and acidogens such as Acetobacterium and Syntrophobacter [72].  Different groups of microorganisms use these low molecular weight degradation products to reduce oxidized forms of compounds such as nitrate, iron, metals and sulfate. According to thermodynamics, first nitrate reducing microorganisms such as Flavobacterium, Acidobacter, Achromobacter, Chloroflexi, Clostridium, and Rhodobium couple oxidation of carbon sources to nitrate reduction since nitrate (NO3−) is the preferred electron acceptor for oxidation of organic matter in the absence of molecular oxygen [72]. The next favorable thermodynamic biological reactions are iron and metal (such as Mn) reduction. Metal reduction is a very important process inside the BCR since this can radically change the speciation, bioavailability and solubility of metals. For example, there is evidence for direct reduction of chromium (VI), uranium (VI), selenium (VI) and molybdenum (IV) by sulfate-reducing microorganisms [145]. Lovely et al. (1993) described mechanisms of direct reduction of iron (III) (by Shewanella Desulfovibrio, Geobacter), manganese (IV) (by Peudomonas putrefaciens), uranium (VI) (by Vellinonella atypica), selenium (VI, IV and 0) (by Clostridium, Citrobacter, flavobacter), chromium (VI) (by Pseudomonas, Bacillus), mercury (II) (by Geobacter Metallireducens), tungsten (VII) (by Desulfovibrio gigas, Moraxella and Planococcus sp.), vanadium (V) (by Pseudomonas vanadium), molybdenum (VI) (by Thiobacillus ferrooxidans, Sulfolobus brierleyi, Pseudomonas guillermondii and  Micrococcus sp.,), copper (II) (by T. ferrooxidans), and auric (III, I) (by Bacillus subtilis, Aspergillus niger and Cholorella vulgaris) [152].  When most nitrate, iron and metals have been exhausted, electron acceptors such as sulfate couple oxidation of carbon sources or H2 to sulfate reduction (Figure 1.2). The process is also referred to as sulfidogenesis. Metal precipitation by sulfate reduction is one of the major processes for immobilization of metals. The process is mainly carried out by sulfate-reducing microorganisms (SRM). Due to sulfidogenesis, metals are removed through the bioprecipitation of metal sulfides. 	  	   9	  One example is immobilization of zinc and copper by formation of ZnS and Cu2S [21, 121, 133].  Oxidation of electron donors such as lactate (CH3CHOHCOO-) is coupled with reduction of sulfate (SO42-) to sulfide (HS-). The sulfide ion (S2-) reacts with positively charged metal ions (Metal2+) to form insoluble metal sulfide precipitates (Metal-S):   2 CH3CHOHCOO- + SO4-2 → 2 CH3COO- + 2 HCO3- + HS- + H+ (1) HS- → H+ + S2- (2) S2- + Metal2+ → Metal-S(s) (3) There is evidence that the extremely reducing conditions that develop during sulfate reduction can lead to chemical conversion of oxyanions, such as SeO4-2, to cationic species, such as Se+2, that are more easily precipitated or biosorbed [14, 22, 145, 152, 153, 171, 216, 239]. Biological sulfate reduction can increase pH by removal of sulfate as the strong sulfuric acid and its replacement by the weak bisulfide ion and remaining hydrogen ions. Where an organic substrate acts as the electron donor bicarbonate is generated. This also can raise the pH, enhance sulfide precipitation and lead to precipitation of hydroxides and carbonates of metals [84]. In the absence of nitrate, metals and sulfate, acetotrophic methanogens use acetate to produce methane or hydrogenotrophic methanogens use hydrogen and CO2 to produce methane. This process is called methanogenesis [181, 190, 228, 229, 236]. Methanogenesis is known to occur in BCRs; however, its involvement in metal removal is still unclear [25, 103, 186]. Methylation of metals such as mercury, tin and lead and metalloids such as arsenic, selenium and tellurium, is another process that is responsible for immobilization of metals. It can be mediated by a range of microorganisms, including those in the class Clostridia, methanogens and sulfate-reducing microorganisms under anaerobic conditions [18, 38, 254]. Microbial methylation influences the bioavailability and toxicity of many metals by changing their solubility and volatility. For instance, methylated selenium derivatives are volatile and less toxic than inorganic forms [84].  Microbial biomass by itself provides a metal sink, by biosorption to cell walls, extracellular polysaccharides and pigments, or by precipitation of metal compounds in/or around cells, hyphae or 	  	   10	  other structures. Microbial products (metal binding peptides and extracellular polymers) can have high affinities toward metals [72]. Another process that helps removal of metals inside BCRs is the release of metal precipitating substrates such as oxalate and phosphates by microorganisms [84]. Moreover accumulation of metals inside cells has been reported. A variety of uptake mechanisms are involved in accumulation of heavy metals within the cell including active transport, efflux and intracellular compartmentalization, and ion exchange [84]. Biosorption, uptake and accumulation have been described as biological mechanisms for removal of zinc, arsenic, copper and selenium [21, 245].  Furthermore, metal oxidation state changes in biochemical reactors as a result of microbial activity and the geochemical conditions. This can contribute to form insoluble metal species. For example biological arsenic removal is achieved through direct oxidation of arsenite to arsenate using arsenic (III)-oxidizing bacteria [16].  1.6 Microbially-mediated Processes in Active Bioreactors for Removal of Nitrogen Compounds Conventional removal of ammonia and nitrate is based on biological ammonia oxidation in a nitrification bioreactor followed by nitrate reduction in a denitrification bioreactor (Figure 1.3).  Figure 1.3: Microbially-mediated processes in active bioreactors for removal of nitrogen compounds. Aerobic Denitri!ers Nitrifying Denitri!ers Ammonia-Oxidizing  Bacteria Ammonia-Oxidizing Archaea Nitrite-Oxidizing Bacteria Denitrifying  Microorganisms Autotrophic Denitri!ers Anammox !"#$% !&'(% !&)(% !'%!&)(% !'%*+%*'&)',+%*#&-',+%*&)',%./%01'$+%"'%$%!&)(%!&'(%$%!"#$% !'%!23/24567.8% 91823/24567.8%6:%;.8<187.86=%>/.51??1?%@./%A1B.<6=%.@%CBB.826%68D%!23/631%%%E:%!.<1=%>/.51??1?%@./%A1B.<6=%.@%CBB.826%68D%!23/631%&'%!'%!"#$% !&'(% !&% !'&% !'%	  	   11	  Nitrification is a two-step process. First ammonia is converted to nitrite by ammonia-oxidizing bacteria (AOB) or ammonia-oxidizing archaea (AOA). Nitrosomonas, Nitrosococcus, Nitrosospira and Nitrosolobus within Proteobacteria are AOB genera [224]. All strains of the lineage Thaumarchaeota thus far identified and characterized are archaeal ammonia-oxidizers [227]. In the second step of nitrification, nitrite is converted to nitrate by nitrite oxidizing bacteria (NOB). All known NOB are affiliated to the four genera Nitrospina (within the Deltaproteobacteria), Nitrobacter (within the Alphaproteobacteria), Nitrococcus (within the Gammaproteobacteria) and Nitrospira (within a separate phylum Nitrospirae) [224].  Denitrification is microbially driven nitrate reduction that may ultimately produce gaseous nitrogen (N2) through a series of intermediate reactions. This respiratory reaction couples oxidation of an electron donor (e.g. organic matter) to reduction of oxidized forms of nitrogen. The process of denitrification is performed by a large group of organisms belonging to all main phylogenetic groups. Generally, several species of bacteria and archaea are involved in the complete reduction of nitrate to molecular nitrogen [77, 208, 209, 218]. Some of the most recognized species include Paracoccus denitrificans, Thiobacillus denitrificans, Alcaligenes xylosoxidans, Alcaligenes faecalis, many Pseudomonas species, Blastobacter denitrificans and Bradyrhizobium japonicum [111]. During the past several years, a number of new biochemical processes including aerobic denitrification, autotrophic denitrification, nitrifier denitrification, anaerobic ammonium oxidation (anammox) have been identified that may account for nitrogen removal (Figure 1.3) [2, 119, 123, 205]. After discovery of these processes, several novel technologies emerged that apply these processes for the removal of ammonia and nitrate [2].  Aerobic denitrification metabolically combines aerobic denitrification and heterotrophic nitrification. It occurs within single bacterial species. Some microorganisms are able to use oxygen or nitrate simultaneously as electron acceptor for doing this reaction [205, 206]. Generally, denitrification is an anaerobic process [211]. Aerobic denitrification is a process of complete denitrification that takes place at high dissolved oxygen concentrations [206]. This reaction has been reported in Thiosphaera pantotropha, Pseudomonas aeruginosa, Alcaligenes faecalis, Bacillus cereus, Paracoccus, Achromobacter and Methylotenera [39, 119, 159, 180, 205, 206]. 	  	   12	  Autotrophic denitrification is based on the oxidation of inorganic sources (instead of organic sources) that coupled to nitrogen oxide reduction. Some denitrifiers can utilize inorganic sources, i.e., reduced sulfur compounds ((HS−, H2S, S, S2O32−, S4O62− or SO32−, etc.), H2, or Fe2+, as the electron donor and grow chemoautotrophically instead of heterotrophically [233]. Currently, 14 validly described species within Alpha-, Beta-, Gamma-, Epsilon- Proteobacteria have been identified as capable of autotrophic denitrification. Up to the current date, Thiobacillus denitrificans and Sulfurimonas denitrificans (formerly known as Thiomicrospira denitrificans) are the most commonly reported autotrophic denitrifiers[218]. Autotrophic denitrification by Paracoccus, Pseudomonas, Thiosphaera, Nitrosomonas, Thialkalivibrio, and Ralstonia genera has been reported [2]. Nitrifier denitrification is the pathway of nitrification in which ammonia (NH3) is oxidized to nitrite (NO2−) followed by the reduction of NO2− to nitric oxide (NO), nitrous oxide (N2O) and molecular nitrogen (N2). This process of simultaneous nitrification and denitrification can be carried out by single species of autotrophic nitrifiers under fully oxic or anoxic conditions [2]. Nitrosomonas-like microorganisms are able to do nitrifier nitrification [2, 24]. Anammox is the process in which nitrite and ammonium are converted directly to N2 anaerobically. Ten anammox species have been reported so far. All have the taxonomical status of Candidatus within the phylum Planctomycetes [107]. 1.7 Microorganisms in BCRs Responsible for Removal of Metals and Sulfate The major microbial mechanisms responsible for metal removal in BCRs were described in Section 1.5. In this section, what is known about the microbial diversity in actual BCRs is reviewed.  There are some reports on microbial populations and their activity in laboratory-based BCRs treating synthetic metal containing water or mine drainage [28, 45, 53, 54, 102, 115, 116, 146, 263]. However, there are much fewer reports on studies of microorganisms in field-based BCRs used for treating MIW. Halberg et al. (2005) studied the microbiology of a field-based BCR that was constructed to remediate acidic, metal-rich mine drainage from a heavy metal mine, in Cornwall, England [96]. They used cultivation and enumeration methods to describe microorganisms in this system. A range of acidophiles including iron-oxidizing Acidithiobacillus ferrooxidans and Leptospirillum ferrooxidans, Halothiobacillus neapolitanus, Ferrimicrobium acidiphilum, 	  	   13	  Thiomonas intermedia, Propionibacterium acnes, and Acidobacterium capsulatum were identified [96]. Burns et al. (2012) studied microorganisms in a BCR treating coal-generated acid mine drainage in Southern Illinois [28]. Bacterial population analysis targeting 16S rRNA and dsrAB (dissimilatory sulfite reductase) gene indicated that the samples were dominated by bacteria related to iron-oxidizing Betaproteobacteia, sulfur-oxidizing Epsilonproteobacteria and complex carbon degrading Bacteroidetes and Firmicutes phyla [28]. Hiibel et al. (2008) compared the microbial populations of two field-based BCRs treating mine drainage by using cloning and quantitative polymerase chain reaction (q-PCR). The microorganisms in these two bioreactors comprised cellulose degraders, fermenters and sulfate-reducing microorganisms (SRM). One site was dominated by uncultured SRB most closely related to Desulfovibrio sp., while other bioreactor were dominated by Thiobacillus sp., [103]. Koschorreck et al. (2010) studied the structure and function of the microbial community in a field-based BCR that was treating MIW. Different groups of Deltaproteobacteria, (mainly the genera Desulfomonile, Desulfobacterium and a phylotype related to Geobacter), Gram-positive sulfate reducers of the genus Desulfosporosinus, acetogenic Acetobacteria, different fermenting bacteria as well as methylotrophic methanogens were identified [129]. The phylogenetic diversity of sulfate-reducing microorganisms (SRM) has been studied more frequently since these organisms are recognized as key players in BCRs (Section 1.5) [36, 52, 122, 130, 155, 176, 239, 252]. Dar et al. (2007) investigated diversity and activity of the sulfate-reducing bacterial population in a BCR using 16S rRNA and dsrB genes [54]. Based on their study, bioreactors that received either ethanol or isopropanol as an energy source showed the presence of SRM affiliated with Desulfobulbus rhabdoformis and/or Desulfovibrio sulfodismutans, as well as SRM related to the acetate-oxidizing Desulfobacca acetoxidans. The reactor that received wastewater containing a diverse mixture of organic compounds showed the presence of nutritionally versatile SRM affiliated with Desulfosarcina variabilis and another acetate-oxidizing sulfate reducers affiliated with Desulfoarculus baarsii. By performing whole-cell hybridization with fluorescently labeled oligonucleotide probes, they showed that Desulfobacca acetoxidans-like populations were most dominant relative to the total SRM populations, followed by Desulfovibrio-like populations, and Desulfobulbus-like populations [54].  	  	   14	  1.8 Association of Sulfate-reducing Microorganisms with Methanogens Several studies of the microbial populations in anaerobic bioreactors discovered that SRM are often associated with methanogens [25, 52, 103, 147, 187, 230]. Both mutualistic and competitive interactions between SRM and methanogens have been observed in laboratory-based bioreactors containing defined carbon sources. Sulfate reducers will compete with methanogens for the common substrates; hydrogen, formate and acetate. However, several syntrophic relations between SRM and methanogenic species have been reported [52, 229, 230].  Methanogens are a phylogenetically diverse group of strictly anaerobic Euryarchaeota. Since these microorganisms are physiologically specialized to form methane, they can grow only through using a few simple structured substrates, and therefore, depend on other microorganisms that degradate more complex compound for substrate supply [237]. Several reviews described obligatory or facultative syntrophic interaction of methanogenic archaea and bacteria (e.g. fermentative Acetobacterium, Syntrophobacter, Pelotomaculum or sulfate-reduces Desulfovibrio, Desulfotomaculum and Desulfobacterium) in anaerobic environment [87, 229, 230, 236, 244, 256] .  Although methanogens often co-occur with SRM, their impact on sulfate-reduction and consequently reactor performance is still unclear particularly in field-based BCRs containing complex organic carbon source. Under some circumstances, methanogens can have an overall negative impact on reactor performance if they outcompete SRM and exclude them from the system [52, 229, 230]. But, methanogens can be beneficial to sulfate-reducers in ethanol-based bioreactors where they consume acetate produced by SRM [102]. Otherwise, accumulation of acetate would inhibit sulfate-reduction [102]. Studies of BCRs using complex organic carbon sources such as woodchips and manure, detected methanogenic archaea together with SRM, and sulfate-reduction did not appear to be inhibited by the presence methanogens [147]. In defined media the ratio of electron donor-to-sulfate governs the outcome of competition between SRM and methanogens [52, 198]. In the presence of sulfate and sufficient amounts of carbon source, SRM outcompete methanogens, and methanogens do not limit the activity of SRM since the SRM have a greater affinity for the carbon substrate. There is some evidence that at high levels of sulfate concentration with excess amounts of carbon source SRM and methanogens co-exist [102, 228]. Under sulfate limiting conditions methanogens outcompete SRM and use simple structured carbon source for methane production [228, 230]. 	  	   15	  1.9 Microbial Communities, Factors controlling Community Properties, Microbial Interactions, and Keystone Microorganisms Microorganisms live in communities and “no single organism contains all the genes necessary to perform the diverse biogeochemical reactions that make up ecological community function” [41]. Microbial communities are defined as multi-species assemblages, in which organisms live together in a well-organized manner and interact with each other. The activities of communities of microbes affect biogeochemical transformations in natural and engineered ecosystems [126],[75]. Structure and function are two properties of microbial communities. Structure refers to the inventory of species within the community, where properties such as richness measure the number of different species in a given community. Functional properties describe the community’s behavior, for instance, how the community removes pollutants, responds to perturbations such as invasion from other species, and interacts with forces in its environment [143]. In a community of microorganisms there is a link between structure and function [37, 85, 143, 220]. Furthermore, the robustness of a given community refers to the ability of the community to maintain its functional and structural integrity in the face of perturbations. When a community adapts to changes and sustains its function even after being exposed to perpetuations, it is considered robust (i.e. stable).  Both abiotic and biotic factors determine the structure of microbial communities (Figure 1.4). The type of carbon source, temperature, oxygen levels, moisture content and pH are some examples of abiotic factors that control the structure of microbial communities in different environments [88, 223, 263],[6].  Figure 1.4: Biotic and abiotic factors controlling the structure of microbial communities. !"#$%&'(%)#*+& ,-"#$%&'(%)#*+&o  !"#$%&#'()*+,,-./$/%0()1)2,o  34'(&',,-5#114')%0)2,o  6'$%&#'()*+,-5#174**#'8,!"49%*#'8,!%"%)(*)12,o  5%":#',)#/"+4,o  ;<=&4',04>40,o  .#()$/"4,o  !?,o  @4174"%$/"4,.(+"#:(%0,5#11/'(*4),	  	   16	  Microorganisms interact with each other via antagonistic, protagonistic or benign mechanisms. Antagonistic interactions occur when microorganisms compete for a common resource or one organism is predator or parasite of another microorganism. For example, some microorganisms produce antimicrobial components such as antibiotics that antagonize other microorganisms, leading to decrease or increase of certain species. In protagonistic interactions (such as syntrophy), both organisms benefit from one another. It defined as beneficial relationship that occurs between syntrophs. Benign (commensalism) associations are those where one partner benefits from the other and the other partner neither is harmed nor benefits from the association [143]. These interactions among microorganisms are biotic mechanisms that can affect population dynamics and ultimately shape the structure of community [143]. For example, a predator can regulate the abundance of certain other species (by an antagonistic mechanism) that in turn regulates other species, providing a cascade of effects that result in radical changes in microbial community structure and function. In this way, certain particular species can radically transform the structure of entire ecosystems through biotic mechanisms. These species are not necessarily the most abundant organisms in the community, but their extensive influence on the community structure and function identifies them as keystone species [143]. The term keystone species was introduced by Robert T. Paine in 1966 [172, 183, 191]. Keystones were defined as the species that are necessary to establish and sustain an ecosystem’s function and stability. If they are removed from the community, there will be an abrupt change in community structure and function. Keystone species affect many other microorganisms in their communities and help to determine the types and numbers of various other species in the community. Their presence is crucial in maintaining the organization and diversity of their community. Keystone species are special targets in the effort to maximize biodiversity and protection of ecological function [172, 191]. Knowing what the keystone microorganisms are in an ecosystem, such as a bioreactor, is very important so their presence can be maintained to keep the microbial community stable.  	  	   17	  1.10 Emerging Methods for Studying Microbial Communities in Bioreactors 1.10.1 High-throughput DNA Sequencing  Techniques that have been used previously to discover which microbes are present in BCRs include culture-based techniques, fluorescence in situ hybridization (FISH), polymerase chain reaction (PCR)-denaturing gradient gel electrophoresis (DGGE), DNA microarrays, clone library construction, and quantitative-PCR [161], [68], [54], [155], [28, 103, 129], [186]. However, these techniques revealed only a few microbial groups and failed to detect low abundance microorganisms. In contrast, using high-throughput next-generation sequencing of hyper-variable regions of small subunit ribosomal RNA (SSU rRNA) genes with technology such as Roche 454 titanium pyrosequencing allows for much higher coverage of phylogenetic diversity. In addition more samples can be processed due to the reduced costs and multiplexing of many samples at the same time [199]. A common approach to describe the phylogenetic diversity of a microbial community is to cluster sequence reads into so-called operational taxonomic units (OTUs) according to a defined sequence similarity cut-off and then assign taxonomy to a representative sequence from each OTU by comparison of the representative sequences with those in curated databases.    Metagenomic sequencing of whole DNA, using either 454 or Illumina technology, reveals all genetic information from the entire community including functional genes. In addition to assessing microbial diversity, metagenomics allows us to uncover the functional genes to provide a means of studying un-cultivable microorganisms. The short reads produced by metagenomic whole DNA sequencing are assembled into longer contigs, from which putative protein sequences are obtained and compared with curated databases so as to assign putative functions. 1.10.2 Network Analysis Tools for Exploring Microbial Interactions and Keystone Species Exploring microbial community structure using the very large datasets that are generated by high-throughput pyrotag or Illumina sequencing needs new analytical methods. Recently, network analysis, used for studying systems comprised of many components that might be connected in some way, has been applied to SSU rRNA amplicon and metagenomic data [12, 160, 231, 262]. Components in the network, referred to vertices (or nodes), can be operational taxonomic units or even meta-data, such environmental variables (e.g. temperature, dissolved oxygen, or pH). 	  	   18	  Connections (edges) between vertices are made based on adjacency matrices composed of metrics that relate vertices to each other.  A simple example of an adjacency matrix for OTUs would their co-occurrence or abundance correlation throughout a dataset of many different samples. This could be either presence or absence or weighted according to relative abundance data based on the number of sequence reads for each OTU in each sample.  So far, the most widely used networking approaches for phylogenetic datasets use correlation-based adjacency tables. When a pair of OTUs displays a similar abundance pattern over multiple samples they are assumed to have a tendency to co-occur with each other (Figure 1.5 a).   Figure 1.5: A) Figure showing similarity in abundance pattern for OTU 1 and OTU 2 over multiple samples. B) Figure showing nodes with high betweenness and closeness centralities in a network. Circles called nodes or vertices in networking language are microbial taxa (OTUs). Lines connecting the circles are called edges in networking language and indicate positive or negative correlations according to the value of person correlation coefficient. There are several commonly used metrics that measure co-occurrence of two variables (OTUs) including Pearson coefficient, mutual information score, Kendall coefficient, and Spearman coefficient. Berry et al. (2014) tested in silico, using a mathematical model, how these different measures of correlation or dissimilarity affected the outcome of the predicted associations. Based on their assessment, Spearman and Pearson correlation coefficients were the most reliable for detecting the actual interactions in their mathematical model [19]. They also advised using samples from very similar environments to avoid habitat partitioning. In this thesis, Pearson’s Product-Moment Correlation Coefficient matrix was selected to analyze OTU relative abundance data from several !"#!!"$!!"%!!"&!!"'!!!"()*+,-+./"01/.2/3"'"01/.2/3"#"!"#$%&'()%)'*)%+$(&$$,,$--%.$,(/01'(2%!"#$%&'()%)'*)%.1"-$,$--%.$,(/01'(2%03% +3%456%7%456%8%90:;1$-%!"#$%&%<$0/-",=-%</"#>.(?@":$,(%A"//$10B",%A"$C.'$,(%D/3%%'("$%&)456-%E%F%A%G%H%9)"/($-(%;0()%.",,$.B,*%,"#$%E%("%,"#$%H%	  	   19	  MIW bioreactors.  Pearson’s Product-Moment Correlation Coefficient (r) measures the tendency of two variables to change in value together (i.e., to either both increase or decrease together). To do this, the sum of products of the two standardized variables are divided by the degree of freedom (n-1). The formula for the Pearson’s Product-Moment Correlation Coefficient (r) is:  Equation (1) ! = !! − !!! !! − !!!!!!! ! − 1  where, x and y are the read counts of a pair OTUs (for example),  s is the standard deviation of x and y over the whole dataset and n is sample size. Correlation coefficient, r, ranges in value from +1 to −1, where 1 is total positive correlation, 0 is no correlation, and −1 is total negative correlation.  The p-value is a measure that tells that correlation between two variables is likely due to random chance. That means smaller the p-value the less likely that there could be a random correlation.  Generally p-values less than 0.05 indicate the correlation is statistically significant and null hypothesis that there is no correlation between two variables could be rejected. In this study, for any value of r, the p-value was computed by calculating t and using Student's t-distribution. The formula for the conversion of r to t is: Equation (2) ! = !  × !!!!!!!   A powerful use of network analysis is the determination of metrics that reveal the importance of vertices within the network based on the types of connections that they have with other vertices in the network. Most useful for microbial community analysis are metrics used to determine the “centrality” of vertices in a network. Betweenness centrality (BC) measures how often a vertex appears on the shortest geodesic paths between other vertices in a network (Figure 1.5 b). It is equal 	  	   20	  to the number of shortest paths (edges; Pearson’s Product-Moment Correlation Coefficient (r)) from all vertices to all others that pass through that node. The betweenness centrality (BC) of a node ϑ  is given by the expression: Equation (3): !"   ! = !!"(!)!!"!!!!!  where !!" is the total number of shortest paths from node s to node t and !!"(!) is the number of those paths that pass through !. The inference is that vertices with high betweenness centrality exert substantial influence or control over the network since many vertices are connected through them.  Closeness centrality (CC) measures the proximity of a vertex to all other vertices in the network (Figure 1.5 b). The closeness centrality (CC) of a node !  is given by the expression: Equation (4): !!   ! =    1!(!, !)!!!  where !   !, !  is the shortest distance between node ! to node ! . In the application of centrality indices to microbial community networks, it is hypothesized that taxa with high betweenness centrality and high closeness centrality are critically important for maintaining the overall structure and function of the community and their absence might lead to community fragmentation [13, 160, 164, 183, 191, 192].  Some studies claim that taxa with high BC and CC might represent keystone taxa according to the definition of Paine [13, 164, 183, 191]. These species may play a key role in a network for different topological reasons. A high CC means that the species interacts closely with many other species (high CC), and a high BC means that the species connects otherwise unconnected sub-networks [164]. So far, most microbial ecology studies have assumed that the most relatively abundant species are also the most important for community function in different ecosystems [32]. This may not be the case as low abundant microorganisms could contribute significantly in ecosystem functioning and one reason for this might be that they are 	  	   21	  keystone taxa [195]. Ecologists are applying these network analysis tools to explore the significance of taxa in the ecosystem under study [192]. Another potential application of network analysis is to suggest the manipulation of microbial communities to enhance the presence of keystone species. The idea for future biotechnologies would be to, instead of engineering a single species, to engineer a whole community by species removal or addition [48, 60, 250].   Network analysis tools recently have been applied to identify microbial interactions and “keystone species” in various biological contexts such as the human microbiome, the bodies of certain animals, marine and soil systems [74], [58], [88], [12, 160]. Lupatini et al. (2014) computed soil bacterial community interactions in several soil samples by using network analysis tools [160]. In their study, Pearson rank correlations test and analysis of betweenness centrality and closeness centrality were performed to determine keystone microbial genera within soil. In this study, genera mainly belonging to phyla Proteobacteria and Actinobacteria were identified as keystone in soil. Faust et al. (2012) analyzed microbial co-occurrence relationships in the human microbiome [74]. In this study, they used different statistical measures (e.g. Pearson and Spearman correlations) on the 16S rRNA sequences generated from 454 FLX titanium sequencing. Samples were taken from 18 different body sites (gut, skin, etc.) within human body. Resulting network showed 3,005 significant co-occurrence and co-exclusion relationships between 197 clades occurring throughout the human microbiome. The network revealed microorganisms that belong to each body site are highly connected and most microbial associations occurring within body sites than between body sites. Gilbert et al. (2012) applied Pearson correlation coefficient test on large dataset from marine microbial community to study interactions within microbial taxa, within eukaryotic cells and between microbial taxa and environmental parameters[88]. Based on their findings, co-occurrence associations were stronger within bacterial taxa rather than between bacteria and eukaryotes, or between bacteria and environmental variables. Barberan et al. (2012) applied network analysis approaches to 16S rRNA gene barcoded pyrosequencing dataset containing more than 160,000 bacterial and archaeal sequences from 151 soil samples [12]. They demonstrated the potential of exploring taxa association to gain an understanding of microbial community structure and the rules that guiding community assembly. Karpinets et al. (2012) proposed a computational model based on Pearson correlation coefficient test to show relationships and regularities in microbial communities [118].  	  	   22	  1.11 Knowledge Gaps and Thesis Objectives  Reliability and long-term performance of bioreactors for treatment of MIW relies on microbial consortia with the metabolic potential to remove the contaminants of concern. Previous studies of microbial populations in semi-passive BCRs were based on studying microorganisms of bioreactors containing defined carbon sources in the controlled environment of laboratory [28, 45, 53, 54, 102, 115, 116, 146, 263]. Supplementation of bioreactors with simple structured carbon sources speeds up the process of metal removal via sulfate reduction; however, it results in higher operational costs. In the field, usage of complex carbon sources reduces the cost of operation, but provides an environment with increased complexity in terms of microbial populations, microbial interactions and biochemical processes. Thus, laboratory-based bioreactor studies may not yield useful information for actual scale-up and operation of field-based systems.  There are several studies on the microbial populations of nitrogen-removing bioreactors. The phylogenetic diversity of key functional groups including AOB, AOA, NOB and denitrifiers have been reported. However, there is no data on the phylogenetic diversity of microorganisms existing in nitrogen-removing bioreactors when the source of wastewater is MIW. It is still unclear if the particular chemistry of MIW affects the phylogenetic diversity of microorganisms in nitrogen-removing bioreactors.   Furthermore, previous studies used conventional techniques to study microbiology of biological reactors that come with limitations. For example, the microbial groups important in metal-removing BCRs are often low in relative abundance, since organic matter degraders are the most dominant members of the community. Therefore, deep SSU rRNA amplicon sequencing is essential to reveal taxonomic groups involved in metal remediation [170, 219]. Additionally, many of the microorganisms living in bioreactors at mine sites are unclassified and uncharacterized, and their metabolic potential has remained to be illustrated. Metagenomic sequencing of whole DNA is needed to reveal functional capabilities of the bioreactors microbial community that cannot be inferred from taxonomy. The overall objective of this thesis was to first study and compare microbial communities in several field-based metal-removing BCRs and nitrogen-removing bioreactors at different mine sites, and 	  	   23	  then to identify taxonomic groups that are key to these types of ecosystems. The overall research questions are addressed in three Chapters: Chapter 3: What are the microbial populations in metal-removing BCRs and are the microbial community structures in BCRs at different locations similar?  The following hypotheses were tested in Chapter 3 to answer this research question: 1. Metal-removing BCRs share several microorganisms even though they are at different geographical locations. 2. The type of carbon source used in BCRs affects the composition of the microbial communities.  3. BCRs are distinct from other anaerobic environments in terms of their SRM community. 4. Sulfate reducing microorganisms in metal-removing BCRs are restricted only to specific phylogenetic groups adapted to metal rich environment.  5. Sulfate reducing microorganisms are prevalent in metal removing BCRs more so than methanogens that are potential competitors for carbon sources. 	  Chapter 4: What biotic factor(s) might contribute to differentiation of microbial community structures in metal-removing BCRs? The following hypotheses were tested in Chapter 4: 1. Since the microbial community structure of BCR3 was different from those in the other BCRs, the keystone microbial groups must also be different.  2. The keystone microorganisms in BCR3 employ different mechanism(s) than those in the other BCRs to regulate the structure of community. Chapter 5: Are the microbial communities of nitrogen-removing bioreactors treating MIW different from those in other nitrogen-removing bioreactors treating low-metal concentration, non-mine-related effluents? The following hypotheses were tested in Chapter 5: 	  	   24	  1. The particular chemistry of MIW affects the phylogenetic diversity of microorganisms in nitrogen-removing bioreactors.   2. Microorganisms able to remove ammonia and nitrate by alternative pathways are present in MIW nitrogen-removing bioreactors.  3. Seasonal changes in the chemistry and microbiology of influent water affects the microbial community composition of the bioreactors.   1.12 Thesis Layout Chapter 2 introduces the field-based bioreactors at different mine-sites in British Columbia that were studied in this thesis work. It describes the locations, configurations, removal efficiencies and operational conditions of four BCRs removing metals and sulfate and two active bioreactors removing ammonia and nitrate from MIW. This information was obtained from reports (grey literature) produced by operators of these bioreactors.  In Chapter 3, four BCRs with some similarities (they are anaerobic, at neutral pH, have high sulfate concentrations, are organic rich and all treat metal-rich MIW), and some differences (they received different inocula, contain substrate sources from pulp mill waste versus wood chips, hay and manure and are of different ages) were selected to study. All the BCRs in this study were pilot-scale or full-scale processes at mine sites or near mineral processing facilities that had been operating successfully for several years. Several samples were taken from different depths and locations of BCRs. After DNA extraction, sequences obtained from pyrotag sequencing of 16S rDNA were analyzed. Since SRM are recognized as important microorganisms in BCRs and most often co-occur with methanogens their phylogenetic affiliations and relative abundance were the focus of Chapter 3. Based on the findings of Chapter 3, the microbial community structure of one BCR (BCR3) was very different from those in the other BCRs. In Chapter 4, in order to assess possible biotic factors that might influence these two different microbial community structures, two datasets were constructed: One containing OTU relative abundance data for BCR3 and the other dataset containing OTU relative abundance data from other BCRs combined. Co-occurrence networks were constructed by calculating the Pearson correlation coefficient values for each dataset. Topological features of each network such as betweenness centrality and closeness centrality were calculated to identify putative keystone taxonomic groups in each network. Ultimately, a schematic model was constructed 	  	   25	  to propose biotic and abiotic factors that could be involved in regulation of microbial communities’ structure in each ecosystem. In Chapter 5, the microbial community structure within MIW nitrogen-removing bioreactors was studied using SSU rRNA amplicon and whole DNA metagenomic sequencing. Data were collected in two seasons (winter and spring) for both full-scale and pilot-scale treatment plants. Phylogenetic diversity of microorganisms with these bioreactors was compared with other nitrogen-removing bioreactors that treat wastewater from other sectors. Microorganisms that were more adapted to MIW treatment system were enriched in culture media (denitrifying bacteria) and whole genome shotgun sequencing was applied for both enrichment cultures and samples from bioreactors. The influence of chemistry and microbiology of influent water on the structure of microbial community of bioreactors were studies. Chapter 6 presents the research outcomes and suggestions for the future studies.         	  	   26	  CHAPTER 2 SITE DESCRIPTION AND PERFORMANCE HISTORY OF THE FIELD-BASED BIOREACTORS SURVEYED IN THIS STUDY  2.1 Introduction  This Chapter describes the field-based bioreactors successfully treating MIW at different mine-sites in British Columbia that were studied in this thesis work. Their locations in British Columbia are shown in Figure 2.1. It includes four BCRs removing metals and sulfate at three different mine or mineral processing sites, and two active biological systems (one full-scale and another pilot-scale) removing ammonia and nitrate from MIW (Table 2.1). An algae pond was also included in this study since it was located within the same treatment system as BCR1, which removes metals and sulfate. 	  	   27	   Figure 2.1: Map of British Columbia showing the location of each BCR. Photos of bioreactors that are operating in each location are presented. (Pictures of BCRs at Mount Polley Mine and Trail were taken by Sue Baldwin and were used with her permission). The major chemical components in the MIW treated by these bioreactors and the configuration of each system are summarized in Table 2.1.      	  	   28	  Table 2.1: Selected properties of the BCRs studied  2.2 Highland Valley Copper (Mine Site 1: BCR1 and BCR2 and the Algae Pond)  Located 80 km southwest of Kamloops, Highland Valley Copper (HVC) is Canada’s largest base metal mine producing copper and molybdenum. Forty-five years of mining activity has resulted in the creation of several completed tailings ponds and developing pit lakes. Management of molybdenum, copper and sulfate is a long-term challenge for the mine operators. HVC operates two semi-passive metal removal BCRs to treat Highmont tailings seepage containing high concentrations of molybdenum, copper, and sulfate. These were used for this thesis study.  2.2.1 BCR1 and the Algae Pond The algae pond and BCR1 were located in a treatment system referred to as S5 on the mine-site adjacent to the Highmont tailings pond. S5 was constructed in 1998. This treatment system consists of a series of algae ponds with a sub-surface flow BCR at the end (Figure 2.2). The final horizontal flow BCR (BCR1) was constructed using a mixture of manure, woodchips and rocks. The bed volume for BCR1 is 2500 m3 with 2 meters depth. Treatable volume in this system is 60-70 l/min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	  	   29	  a)                        b)  Figure 2.2: A) Aerial view of Highland Valley Copper Mine showing the location of BCR1 (indicated by the red circle). B) Schematic diagram showing the actual location of the algae pond (AP) and BCR1 within the S5 treatment system. (Pictures a and b are from HVC annual report by Heather Larratt and were used with her permission) [100]. Water quality data from the influent and effluent streams shows that the system removes molybdenum and copper (Figure 2.3).   a) Molybdenum                                                             b) Copper                                          Figure 2.3: Influent and effluent concentrations of dissolved a) molybdenum and b) copper in the S5 treatment system at Highland Valley Copper. (Data were extracted from HVC annual report by Heather Larratt and were used with her permission) [100].   !"#"$"%"&"'!"'#"'$"'((&" '(((" #!!!" #!!'" #!!#" #!!)" #!!$" #!!*" #!!%" #!!+" #!!&" #!!(" #!'!",-./0.1234-."-5"6-789:0.;<"=<>?7@"A.5-B"-;C7-B"!"!#!$"!#!%"!#!&"!#!'"!#!("!#!)"$**+" $***" %!!!" %!!$" %!!%" %!!&" %!!'" %!!(" %!!)" %!!," %!!+" %!!*" %!$!"-./01/2345./".6"-.7713"89:;<=" >/6.?".@A<.?"	  	   30	  In addition to molybdenum and copper, the concentration of sulfate decreases during treatment (Figure 2.4).  Figure 2.4: Influent and effluent concentrations of sulfate in the S5 treatment system. (Data were extracted from HVC annual report by Heather Larratt and were used with her permission) [100]. 2.2.2 BCR2 for Removal of Metals  BCR2 is another sub-surface flow BCR that also used woodchips mixed with manure at the HVC mine site but at a different location on the mine. It was constructed in 2002 with the ability for either up-flow or down-flow configuration. The bed volume is 600 m3 with the bed depth of 4 meters. The treatable volume is 30-40 l/min. Water quality data from influent and effluent water shows that BCR2 removes molybdenum and copper (Figure 2.5).   a) Molybdenum                                                             b) Copper                                         Figure 2.5: Influent and effluent concentrations of dissolved a) molybdenum and b) copper in BCR2 at Highland Valley Copper. (Data were extracted from HVC annual report by Heather Larratt and were used with her permission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	  	   31	  2.2.3 Inoculum Pond (IP)  Both BCRs at HVC received inoculum from the same source: a local natural pond on the mine site (Figure 2.6). The inoculum pond is located on the same mine site as the BCRs, and is exposed to the same climatic conditions. However, it differs from the BCRs in that it is not affected by any MIW and is free of heavy metals and high sulfate concentrations. Unfortunately there are no data for the actual geochemical conditions of the sediments in the inoculation pond. Sediments taken from the inoculum pond were used as a non-MIW impacted control in this thesis study.   Figure 2.6: Inoculum pond. (The picture of the inoculum pond was taken by Sue Baldwin and was used with her permission). 2.3 Trail Wetland (Mine Site 2: BCR3) Another semi-passive treatment system (BCR3) used for this study was located near Trail B.C. It treated landfilled smelter waste seepage water containing elevated concentrations of arsenic, zinc, and sulfate. BCR3 was the first BCR in a series of two sub-surface flow anaerobic bioreactors followed by several surface flow vegetation cells [166]. Samples from the first BCR were used for this study since this BCR was responsible for the majority of metal removal. It was constructed in 1997 and rebuilt in 2002. The BCR3 matrix consisted of pulp mill biosolids that were a waste product from a local kraft pulp mill and iron-coated sand along with limestone gravel (Table 2.1). Water quality data from influent and effluent shows that BCR3 removed arsenic and zinc (Figure 2.7) [121].  	  	   32	   a) Arsenic                                                                    b) Zinc                                          Figure 2.7: Influent and effluent concentrations of dissolved a) arsenic and b) zinc in BCR3. (Data were collected by the operators) [10]. In addition to arsenic and zinc, the concentration of sulfate decreases during treatment, although not to a great extent (Figure 2.8).  Figure 2.8: Influent and effluent concentrations of sulfate in BCR3. (Data were collected by the operators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	  	   33	  2.4 Mount Polley Mine (Mine Site 3: BCR4) At the Mount Polley mine site, a pilot-scale subsurface flow system (BCR4) removes selenium, copper and molybdenum as well as some sulfate from tailings storage facility toe drain water. It was constructed in 2010. This BCR differed from the others in that it contained a water cover as opposed to a solid cover of sand and soil as was used for the other BCRs (Figure 2.1). BCR4 contained woodchips, manure and hay as the carbon sources. Water quality data from the influent and effluent flow streams shows that BCR4 was removing selenium and sulfate (Figure 2.9).   a) Selenium                                                                    b) Sulfate                                          Figure 2.9: Influent and effluent concentrations of dissolved a) selenium and b) sulfate in BCR4. (Data were collected by Susan Baldwin and were used with her permission) [234]. 2.5 Nickel Plate Mine (Mine Site 4; Active Biological Reactors for Removal of Ammonia and Nitrate) The Nickel Plate Mine was an open pit gold mining operation located near Hedley 50 km west of Penticton in south central British Columbia (Figure 2.1). Closure of the mine in 1996 required treatment and release of the wastewater stored in the tailings pond, mostly to remove cyanide and metals such as arsenic. Cyanide is converted to ammonia, and sequential nitrification denitrification bioreactors are used for conversion of ammonia and nitrate into dinitrogen gas (Table 2.1). In this thesis, the nitrification and denitrification circuits were studied. Both the full- and pilot-scale systems were sampled. The system configuration consists of two circuits each with bioreactors followed by a settling chamber (Figure 2.10). Effluents from the nitrification and denitrification bioreactors, respectively, overflow into the center of settling chambers where the small amount of !"!#!!$"!#!%"!#!%$"!#!&"!#!&$"!#!'"&!%!" &!%%" &!%&" &!%'" &!%(")*+,-+./01*+2"*3"4-5-+678"98:;5<"=+>*?"@7A5*?"!"#!!"$!!"%!!"&!!"'!!"(!!"$!#!" $!##" $!#$" $!#%" $!#&")*+,-+./01*+"*2"34520.-"67895:";+<*=">4?5*="	  	   34	  cationic flocculant is added. The sludge (containing the microbial biomass) settles in the chamber and is pumped back into respective bioreactor (Figure 2.10).  Figure 2.10: Process flow diagram showing the nitrification and denitrification circuits (bioreactors and settling chambers). Photographs are of the indicated bioreactors taken at the Nickel Plate Mine. Water quality data collected by the operators from influent and effluent flow streams indicates that the nitrification reactor removes ammonia and the denitrification reactor removes nitrate (Figure 2.11).  Settling Chamber!(Sludge Settles)!Effluent from nitrification circuit !enters to denitrification circuit!Influent!Sludge Return!Effluent!Nitrification!(Aerobic)!Settling Chamber!(Sludge Settles)!Denitrification!(Anaerobic)! Sludge Return!Nitrate           Nitrogen!Ammonia           Nitrate!Nitrification Circuit! Denitrification Circuit!!"#$%&'()#*++,*-"%."/()0$-+1"$.*)(%$.2+3"%."/()0$-+1"$.*)(%$.2+ 45##&'()#*++	  	   35	   a) Ammonia                                                                    b) Nitrate                                         Figure 2.11: Average annual influent and effluent concentrations of a) ammonia and b) nitrate in the nitrification and denitrification circuits of the Nickel Plate Mine bioremediation system. (Data were collected by operator in this facility and were used with his permission). Since the source of influent water is mine tailings water, it contains high concentration of metals, sulfate in addition to the N-compounds (Table 2.2).  Table 2.2: Chemical characteristics of influent water.   !"#"$!"$#"%!"%#"&!"&#"%!!%" %!!&" %!!'" %!!#" %!!(" %!!)" %!!*" %!!+" %!$!",-./0.1234-."-5"677-.83"97:;<=>?"@.AB0.1"<8128C/34-.",82/B81"DEB0.1"F0.8128C/34-.",82/B81"DEB0.1"!"#!"$!"%!"&!"'!!"'#!"'$!"#!!#" #!!(" #!!$" #!!)" #!!%" #!!*" #!!&" #!!+" #!'!",-./0.1234-."-5"6712310"89:;6<=>"?.@A0.1"67127B/34-.",72/A71"CDA0.1"E0.7127B/34-.",72/A71"CDA0.1"	  	   36	  CHAPTER 3 COMPARSION OF SEVERAL METAL-REMOVING BIOCHEMICAL REACTORS IN TERMS OF THEIR MICROBIAL POPULATIONS  3.1 Synopsis  In this Chapter, the microbial community structures of four different BCRs removing metals and sulfate from MIW were studied. As described in Chapter 2, these BCRs contained complex organic materials such as wood chips, hay, manure and pulp mill biosolids, and therefore were expected to have complex microbial communities. The bioreactors were all similarly constructed to be anaerobic, at neutral pH, and receiving metal-rich MIW influent water with high sulfate concentrations. All of the organic sources used in these bioreactors were lignocellulosic plant-derived materials. But there were some differences between them. One bioreactor (BCR3) obtained its substrate source from pulp mill waste versus fresh wood chips that were used at the other sites. At one of the sites, the wood chips were supplemented with hay (BCR4).  Other differences between the BCRs that may affect their microbial community compositions were their ages, their different inocula, and their different influent waters composition.  Two of the bioreactors were on the same mine site, whereas the others were located farther away. Also studied in this Chapter was an algae pond located just prior to BCR1. In this pond, algae growing on top of the ponds were the source for carbon in the sediments. The microbial community within the algae pond sediments was compared with those in the BCRs. As controls, samples were obtained from the natural pond from which sediments were used as inoculum for BCR1 and BCR2, and the soil adjacent to BCR2.  Since sulfate-reducing microorganisms (SRM) are the most important functional group for effective metal removal, this Chapter focuses on their phylogenetic diversity within the bioreactors, and explores their co-occurrence with methanogens. As explained in Chapter 1, whether SRM and methanogens co-occur or compete is an important question for these bioreactors.  	  	   37	  The overall research questions in this Chapter were: What are the microbial populations in metal-removing BCRs and are the microbial community structures in BCRs at different locations similar? The following hypotheses were tested to answer these research questions: 1. Metal-removing BCRs share several microorganisms even though they are at different geographical locations. 2. The type of carbon source used in BCRs affects the composition of the microbial communities.  3. BCRs are distinct from other anaerobic environments in terms of their SRM community. 4. Sulfate reducing microorganisms in metal-removing BCRs are restricted only to specific phylogenetic groups adapted to metal rich environment.  5. Sulfate reducing microorganisms are prevalent in metal removing BCRs more so than methanogens that are potential competitors for carbon sources. 	  3.2 Materials and Methods  3.2.1 Description of Biochemical Reactors and Sampling Three mine sites were selected for this study since they operate BCRs efficiently treating metal and sulfate-containing water using semi-passive treatment technologies (Chapter 2, Figure 2.1, Table 2.1). Mine site 1 operates two metal removal bioreactors (BCR1, BCR2) and several algae ponds (AP) employing microorganisms to treat tailings seepage containing molybdenum, copper and sulfate. Total of 7 samples were taken from the sludge layer of the final algae pond just prior to the BCR1 and 13 samples were taken from within the sub-surface flow BCR1. Samples were obtained during operation and on the same day from random horizontal and vertical locations using an excavator. Nine samples were collected from similar depths and different locations within BCR2, which is another sub-surface flow BCR also using wood chips mixed with manure at the Mine site 1 (Chapter 2, Figure 2.1, Table 2.1).  A total of 38 samples were collected during July 2008, April 2009 and October 2009 by others from BCR3 which is located at site 2 [10]. BCR3 was a sub-surface flow bioreactor built for removal of arsenic, zinc, and sulfate (Chapter 2) [166]. Briefly, the material through which the metal-rich water flowed was accessed by drilling through a soil cover.  A total of nine bore holes were drilled, three at 	  	   38	  different spatial locations at each sampling time. Cores were removed inside 2 cm internal diameter poly vinyl chloride pipes that were capped immediately to prevent any contact with oxygen and then frozen using liquid nitrogen. In the laboratory, the cores were sectioned into 5 cm intervals while still frozen and each section was homogenized by grinding under liquid nitrogen. This BCR differed from the others in that it contained pulp mill biosolids as the carbon source (Chapter 2, Table 2.1). Six samples from BCR4 at Mine site 3 were removed from random locations by excavation from the saturated organic matrix. Others did sampling from this site [234]. BCR4 contained woodchips, hay and manure as carbon sources.   In an attempt to understand to what extent the microbial populations in metal-removing BCRs are distinct from environmental samples in the same geographical region, 8 samples were taken from a natural non-mine impacted pond (HVC inoculum pond) and soil located nearby BCR1 and BCR2. The inoculum pond samples were from the sediments, which are anaerobic like the BCRs, but the inoculum pond is not impacted by metal and sulfate containing MIW. The inoculum pond sediments were the inoculum source for BCR1 and BCR2.  Collected samples from Mine 1 were placed in sterile WhirlPak sampling bags (Nasco, Modesto, CA) and frozen immediately using liquid nitrogen. In the laboratory, the samples were homogenized and analyzed as described in Section 3.2.2. Environmental parameters such as temperature, pH, dissolved oxygen and oxidation-reduction potential were measured in the BCR pore water at the time of sampling using an YSI Sonde (EXO1). Nitrate, ammonia, ferrous iron, phosphate and sulfide were measured in the BCR pore water immediately using CHEMetrics kits (https://www.chemetrics.com/index.php?Page=methods&tab=1, USA) according to the manufacture’s instructions.  Other constituents such as sulfate and metals were analyzed using inductively coupled plasma mass spectrometry (ICP-MS) at a commercial or mine laboratory.  3.2.2 DNA Extraction, Amplification of 16S rRNA Genes, and Pyrosequencing After homogenization, DNA was extracted using a Power Soil DNA Isolation Kit (MoBio Laboratories, Carlsbad, CA, USA, Cat No: 12888-100) from 0.5 gr of each sample. Isolated community DNA (2 ng) was subjected to polymerase chain reaction (PCR) amplification of variable region V6 to V8 of bacterial and archaeal 16S rRNA by using barcoded primers pair 926f (AAA 	  	   39	  CTY AAA KGA ATT GAC GG), 1392r (ACG GGC GGT GTG TRC). Primer 454T-RA: 25 nt A-adaptor (CCATCTCATCCCTGCGTGTCTCCGACTCAG), primer 454T-FB: 25 nt B-adaptor sequence (CCTATCCCCTGTGTGCCTTGGCAGTCTCAG). Variable regions 6 to V8 were selected for amplification since studies demonstrated these regions provide better microbial community coverage in complex ecosystems [131]. PCR was performed using an iCycler® (Biorad) thermocycler under conditions: 95oC for 3 min; 25 cycles of 95oC for 30s, 55oC for 45s, 72oC for 90s; and 72oC for 10 min. DNA concentrations and purity were measured on a NanoDrop® ND-2000 UV-Vis Spectrophotometer (NanoDrop Technologies, Wilmington, DE) and by running 1 µL of the PCR product on a 1% agarose gel. Purified PCR products were sent to the Genome Quebec and McGill University Innovation Centre (Montréal, Québec, Canada) for pyrosequencing using a Roche GS-FLX Titanium Series sequencer. 3.2.3 Sequence Analysis 	  Raw sequence data generated from pyrosequencing were processed using QIIME 1.8.0 [33]. QIIME is a python interface that connects a very large number of programs (e.g., pynast, uclust). Because QIIME runs programs that developed independently, it may offer higher performance than what is implemented in other tools (e.g., mothur). It also gives a better visual output [34]. Every raw pyrosequence had to pass a multistage quality inspection to remove low quality reads and minimize sequencing errors that can be introduced during the pyrosequencing process. Sequences were rejected if they: (i) did not have perfect match with the pyrosequence primers (ii) contained any nucleotide ambiguities (iii) had a typical length of 1 standard deviation from the mean length after removing primer sequences (iv) had average quality scores were below 25 (recommended for the Roche Genome Sequencer FLX System) and (v) had a homopolymer sequence longer than eight base pairs. The remaining high quality sequences were imported to the QIIME suite of Python scripts and associated dependencies were used to cluster the filtered reads into operational taxonomic units (OTUs) using 97%, 94% and 90% sequence similarity thresholds with the computer program usearch which corresponds approximately to taxonomic level of species, genus and family, respectively [33], [63], [127]. Representative sequences for each OTU were assigned taxonomy using BLASTn to the Silva version 111 representative set (http://www.arb-silva.de/documentation/background/release-111/) [194]. OTUs represented by one read only were removed. Phylogenetic trees with pyrotag OTUs and nearest neighbours picked by using BLASTn to 	  	   40	  the NCBI nucleotide database were constructed by trimming the NCBI 16S rRNA sequences to the same region as the pyrotag amplicons and aligning these using MUSCLE version 3.8.31 to the reverse complement of the pyrotag OTUs representative sequences followed by tree building with PHYML (nucleotide substitution model HKY, 100 bootstraps) [62], [94]. Phylogenetic trees were constructed using the Bosque computer program and visualized on the Interactive tree of life website or by using Figtree V1.3.1 (Appendix B) [139, 196]. To assess the community properties (e.g. structure, Chapter 1, Section 1.9) and look for community-level differences in the samples, alpha- and beta-diversity analysis were performed using QIIME 1.8.0. Alpha diversity (measures the structure of microbial community in a given sample) metrics were: Chao1, Observed Species, Shannon, Simpson and PD-Whole_tree [134, 241]. Beta-diversity metrics such as weighted and unweighted UniFrac were used to assess the differences between microbial communities based on their composition [156]. UniFrac, measures the phylogenetic distance between sets of taxa in a phylogenetic tree as the fraction of the branch length of the tree that leads to descendants from either one environment or the other, but not both. This method can be used to determine whether communities are significantly different, and to measure the relative contributions of different factors, such as chemistry and geography, to similarities between samples [156]. The results showing similarities or dissimilarities between samples visualized with analysis such as Principal Coordinates Analysis (PCoA). UPGMA (Unweighted Pair Group Method with Arithmetic Mean) method was used for classification of samples on the basis of their pairwise similarities in sulfate-reducing and methanogenic communities [271]. The rarefied 94% and 97% sequence similarity OTU-tables were used to produce heatmaps in R. 3.2.4 Analysis of Co-occurrence and Network of Correlations For the analysis of co-occurrence of SRM and methanogen taxa, the 97% sequence similarity OTUs were used. Only those OTUs with more than five sequences were considered. Correlation within and/or between the SRM- and methanogen-related communities were examined by calculating all possible Pearson rank correlations between bacterial and archaeal species using equation 1 (Chapter 1, Section 1.10.2) and the script enterotypes_graph.pl (https://github.com/kishori82/Correlation_Network). The p-values were calculated using Equations 2 (Chapter 1, Section 1.10.2) using the script enterotypes_graph.pl obtained from https://github.com/kishori82/Correlation_Network. A valid interaction event was considered to be a 	  	   41	  robust correlation if the Pearson correlation coefficient (p) was either equal or greater than 0.8 or less than or equal to −0.8 and statistically significant (p-value equal or smaller than 0.05). The cutoff correlation of 0.8 or −0.8 was chosen to increase the confidence for strong taxa co-occurrence. The network structure was explored and visualized with the Cytoscape version 3.0.1 using the Edge-weighted (correlation) spring embedded layout (Appendix B) [217]. 3.3 Results  3.3.1 Physiochemical Properties of the BCRs The pH and dissolved oxygen concentrations measured in the BCRs at the time of sampling indicated they were all operating under neutral pH and anaerobic to sub-oxic conditions (Table 3.1). Concentrations of sulfate in the BCRs ranged from 321 mg/l to 600 mg/l. The chemical composition of the solids indicated high concentrations of arsenic and zinc in BCR3, and high levels of copper and molybdenum in AP, BCR1 and BCR2 showing that these systems are indeed accumulating metals. Information about performance of these BCRs for removal of metals and sulfate at the time of sampling was presented earlier (Chapter 2). Table 3.1: Physiochemical properties of the BCRs. (ND stands for not determined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	  	   42	  3.3.2 Overall Microbial Community Structure A total of 517,509 reads were determined by pyrotag sequencing of 16S rRNA amplicons. The numbers of OTUs that were generated based on 97%, 94% and 90% sequence similarity, equivalent to species, genus and family level, respectively, are presented in Table 3.2.  Table 3.2: Diversity indices of samples used in this study. These were determined using the OTU tables based on 97%, 94% and 90% sequence similarity.  Alpha rarefaction demonstrated that the highest level of diversity was found in BCR1 and BCR4 and the lowest level for BCR3 (Appendix A, Figure A.1). The environmental samples (soil, IP) had lower levels of diversity compared to those in the BCRs except for BCR3. For further analysis, the 94% similarity cut-off OTUs were rarefied to 3700 reads each per sample and then trimmed to exclude OTUs that were represented by fewer than 2 read counts each.   !"#$%&'()'*)%+,-.$%&'()'*)%$/0$12$%)-$3)!+,-.$&'()'*)456%)789:;&'()'*)456%)78<:;&'()'*)456%)78=:;>?@A !" #$%& "$#' (!(& !&()>?@B & )"*$ (*!) !)$' !"*'CD % *()& (!(* !"&# !!)">?@E "# *#") !#$& !!)* #')>?@< ) )&$$ ('%* !(#% !)%$FD $ *$$& !!)# )&' )$"!'". $ )&&# )%# $'' ")*	  	   43	  Comparing the microbial communities from each sample using unweighted UniFrac analysis (Figure 3.1) revealed that the algae pond (AP), BCR1, BCR2 and BCR4 were similar to each other, whereas BCR3 had a very different microbial community. The environmental samples, especially the inoculum pond (IP) had a microbial community structure similar to that of BCR1, BCR 2 and BCR4. The soil and some of the BCR1 samples were similar to each other and somewhat different from the other BCR samples. Comparing the microbial communities of samples from each depth (layer) revealed samples did not cluster with respect to depth (Appendix A, Figure A.2).   Figure 3.1: Three-dimensional principal coordinate analysis based on unweighted UniFrac distances       between samples from different bioreactors. Axis 1 explained 28.92% of variation, axis 2, 8.34%, and axis 3 explained 4.48% of variation. Separation of microbial diversity between BCR3 versus other BCRs was observed.       PC2 - Percent variation explained 8.34%PC2- Percent variation explained 8.34%PC3 - Percent variation explained 4.48% 	  	   44	  Despite having different geographical locations, similarities in major taxonomic groups (phyla) were observed for AP, BCR1, BCR2 and BCR4 (Figure 3.2 and Appendix A, Figure A.3). In these bioreactors, most OTUs were classified in the Phyla: Euryarchaeota, Bacteroidetes, Firmicutes, Proteobacteria and Chloroflexi. BCR3 was also dominated by sequences related to Euryarchaeota, Firmicutes, and Bacteroidetes, but with different proportions than the other BCRs. Distinctive from other BCRs, large numbers of sequences in BCR3 were affiliated to Spirochetes. Samples from the inoculum pond were dominated by same phyla as in the BCRs, however soil samples were dominated by Proteobacteria- and Acidobacteria-related sequences.   Figure 3.2: Phylum-level taxonomic summary of the 94% sequence similarity OTUs. Phyla arranged based on the color from the bottom to the top, starting with Acidobacteria. Vertical axis represents percent reads, and the horizontal axis identifies the source of the samples. AP and IP stand for algae pond and inoculum pond, respectively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	  	   45	  3.3.3 Distribution of Sulfate-reducing Microorganisms Since SRMs are the most important functional group in BCRs treating metal- and sulfate-rich MIW, a subset of the overall dataset was selected to study their phylogenetic diversity. Since the number of samples available for each BCR varied greatly, a subset of 6 samples from each were chosen randomly for the comparison of the SRM and methanogenic communities.  Taxonomic groups that are known to contain sulfate-reducing microorganisms are: Desulfobacterales, Desulfuromonadales, Desulfovibrionales and Syntrophobacterales (Deltaproteobacteria), genera Desulfotomaculum, Desulfosporomusa, and Desulfosporosinus (Firmicutes), species Thermodesulfovibrio (Nitrospirae), genera  Archaeoglobus, Thermocladium and Caldivirga (Archaea) and two phyla Thermodesulfobacteria, Thermodesulfobium [15]. To report on the distribution and phylogeny of sulfate-reducing microorganisms in these BCRs, the OTU representative sequence datasets were searched for all OTUs that were assigned to any of the aforementioned taxa. Subsequently, it was observed that the SRM-related sequences were restricted to a certain phylogenetic linage in that 93% of all sequences retrieved from all BCRs, were assigned to Orders within the Class Deltaproteobacteria. The only non-Proteobacteria sulfate reducer-related OTUs were a few that were assigned to the genus Desulfosporosinus in the order Clostridiales (Firmicutes phylum) (Appendix A, Figure A.4). In spite of this phylum-level restriction, high prevalence of SRM-related sequences with high level of within-phylum diversity was observed. Figure 3.3 shows the overall family-level distribution of SRM-related sequences in all BCRs (Figure 3.3 a) and each site (Figure 3.3 b). There are several environmental groups classified within the Deltaproteobacteria class. Although, it is not known if any of these contain sulfate-reducers since they are uncultured and uncharacterized, presence of representatives from the environmental groups in the BCR samples was observed (Figure 3.3 c).      	  	   46	   a)    b)                                                                              c) Figure 3.3: A) Family-level (90% homology cut-off) taxonomic summary of SRM and Deltaproteobacteria-related environmental groups in all BCRs combined. B) Number of reads assigned to different known taxonomic groups (family-level) in each of the features sampled. C) Number of reads assigned to different uncultured taxonomic groups within Deltaproteobacteria (family-level) in each of the features sampled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	  	   47	  Excluding BCR3, the SRM-related communities within the BCRs were very similar to each other despite the different geographical locations of the BCRs and the different sources of inoculum (as in the case of BCR4, which did not receive inoculum from the Highland Valley Copper inoculum pond) (Figure 3.4). This similarity was also demonstrated in a phylogeny-based dendrogram derived from the UniFrac analysis that classified samples on the basis of their pairwise similarities in their sequences (Appendix A, Figure A.5). In the BCRs containing large numbers of sequences assigned to sulfate reducers, the predominant OTUs were classified in families Desulfobacteraceae, Desulfobulbaceae, Desulfuromonadaceae, Sva0081_sediment_group, Geobacteraceae and Syntrophaceae. In BCR3, which was different from the other BCRs, far fewer SRM-related sequences were detected and those that were found formed a quite different phylogenetic lineage with OTUs classified within the Peptococcaceae and in the environmental group Sh765B-TzT-39 (within Deltaproteobacteria). A limited number of reads from BCR3 were assigned to uncultured Syntrophs and Desulfovibrio.  When the putative sulfidogenic microbial populations of the BCRs were compared with those in the environmental samples, the orders Desulfobacterales, Desulfuromonadales (specially Geobacter) and Desulfovibrionales were commonly represented in the natural inoculum pond and in the mine affected constructed systems: AP, BCR1, BCR2 and BCR4.  However, these groups were present in the BCRs in different proportions. Uncultured sequences affiliated to Desulfobacterium occurred with less abundance in IP. Sequences affiliated to genus Desulfosporosinus were detected in all bioreactors but not in the IP, and larger numbers of Desulfarculaceae-related sequences were observed in the IP and not in the bioreactors. No Desulfovibrio-related sequences were observed in the IP. The relative abundance of sequences affiliated to the uncultured- environmental group Sh765B-TzT-29 was higher in bioreactors specially BCR1 and BCR3. In contrast, sequences affiliated to the uncultured-uncharacterized group Sva0485-sediment-group were prevalent in the inoculum pond but not in the bioreactors. The soil samples contained very few SRM-related sequences except for some assigned to uncultured species of Nitrospina (Figure 3.4).  	  	   48	   Figure 3.4: Genus-level logarithmic heatmap (log 2) demonstrates relative distribution of SRM-related sequences. OTU numbers and their Silva 111 assigned taxonomy are given on the right hand side. Sample names are column labels. The heatmap was constructed by clustering the sequences in to OTUs based on 94% sequence similarity. The relative abundance for each OTU in different sites is colored in shades of white (low relative abundance) to purple (high relative abundance). BCR4.4BCR4.3BCR4.2BCR4.1BCR4.5BCR3.5BCR3.1BCR3.3BCR3.6BCR3.2BCR3.4Soil.2Soil.1BCR1.5AP.4AP.2AP.6AP.3AP.5BCR2.2BCR2.1BCR1.3BCR2.3BCR2.4BCR2.6BCR2.5IP.3IP.1IP.2BCR1.5BCR1.1BCR1.4AP.6AP.1BCR1.2297_un_Desulfosporosinus3902_Un_Desulfosporosinus2585_Un_Desulfosporosinus3211_Un_Desulfosporosinus5277_Un_Desulfosporosinus4054_Un_Desulfosporosinus4313_Un_43F-­1404R1097_Un_43F-­1404R4944_Un_43F-­1404R2817_Un_43F-­1404R1704_Un_Desulfarculaceae4523_Un_Desulfarculaceae2859_Un_Desulfarculaceae3102_Un_Desulfarculaceae2732_Un_Desulfarculaceae2376_Un_Desulfarculaceae1182_Un_Desulfarculaceae2447_Un_Desulfarculaceae5060_Un_Desulfarculaceae807_Un_Desulfatiferula16_Un_Desulfatirhabdium5178_Un_Desulfatirhabdium733_Un_Desulfatirhabdium4631_Un_Desulfatirhabdium1367_Un_Desulfatirhabdium504_Un_Desulfatirhabdium505_Un_Desulfatirhabdium1028_Un_Desulfatirhabdium2343_Un_Desulfatirhabdium3880_Un_Desulfatirhabdium4110_Un_Desulfatirhabdium244_Un_Desulfobacter4155_Un_Desulfobacter2269_Un_Desulfobacterium368_Un_Desulfobacterium337_Un_Desulfobacterium4956_Un_Desulfobacterium341_Un_Desulfobacula5079_Un_Desulfobacula216_Un_Desulfobacula2136_Un_Desulfococcus_biacutus3482_Un_Desulfobacula2295_Un_Desulfonema158_Un_Desulfosalsimonas975_Un_Desulfosalsimonas1078_Un_Desulfosalsimonas3266_Un_Desulfosarcina1393_Un_SEEP-­SRB13556_Un_SEEP-­SRB13143_Un_SEEP-­SRB12259_Un_SEEP-­SRB15280_Un_Sva0081_sediment_group140_Un_Sva0081_sediment_group5015_Un_Sva0081_sediment_group1745_Un_Sva0081_sediment_group3193_Un_Sva0081_sediment_group2419_Un_Sva0081_sediment_group3067_Un_Sva0081_sediment_group4478_Un_Sva0081_sediment_group1339_Un_Sva0081_sediment_group2684_Un_Desulfobacteraceae4535_Un_Desulfobacteraceae388_Un_Desulfobacteraceae5057_Un_Desulfobacteraceae5050_Un_Desulfobacteraceae5127_Un_Desulfobacteraceae2483_Un_Desulfobacteraceae3011_Un_Desulfobacteraceae66_Un_Desulfobacteraceae4796_Un_Desulfobacteraceae1023_Un_Desulfobacteraceae1504_Un_Desulfobacteraceae1358_Un_Desulfobacteraceae451_Un_Desulfobacteraceae4553_Un_Desulfobacteraceae441_Un_Desulfobacteraceae813_Un_Desulfobulbus131_Un_Desulfobulbus220_Un_Desulfocapsa2094_Un_Desulfocapsa302_Un_Desulfocapsa512_Un_Desulfocapsa2355_Un_Desulfopila125_Un_Desulfurivibrio4549_Un_Desulfurivibrio4854_Un_Desulfobulbaceae4942_Un_Desulfobulbaceae74_Un_Desulfobulbaceae2876_Un_Desulfobulbaceae825_Un_Desulfobulbaceae1828_Un_Desulfobulbaceae1925_Un_Desulfobulbaceae30_Un_Desulfobulbaceae3071_Un_Desulfobulbaceae1421_Un_Desulfobulbaceae4852_Un_Nitrospina850_Un_Nitrospina3263_Un_Nitrospina861_Un_Nitrospina1460_Un_Nitrospina3817_Un_Nitrospina4472_Un_Nitrospina1077_Un_Desulfomicrobium4708_Desulfovibrio_sp._A-­1924_Desulfovibrio_sp._LG-­2009392_Un_Desulfovibrio4972_Un_Desulfovibrio330_Un_Desulfovibrio716_Un_Desulfovibrio968_un_Desulfuromonadales4776_un_Geobacter_sp.952_un_Geobacter_sp.2924_un_Geobacter_sp.265_Un_Desulfuromonas5128_Un_Desulfuromonas4172_Un_Desulfuromusa311_Un_Geobacter3092_Un_Geobacter1983_Un_Geobacter3942_Un_Geobacter4855_Un_Geothermobacter3937_Un_GR-­WP33-­582994_Un_GR-­WP33-­58181_Un_Desulfuromonadales42_Un_Desulfuromonadales4634_Un_DTB1202057_Un_DTB1204224_Un_DTB1203769_Un_DTB1203729_Un_GR-­WP33-­301682_Un_GR-­WP33-­30520_Un_GR-­WP33-­304768_Un_GR-­WP33-­302947_Un_GR-­WP33-­304430_Un_GR-­WP33-­303057_Un_GR-­WP33-­303153_Un_GR-­WP33-­303719_Un_GR-­WP33-­302214_Un_GR-­WP33-­304525_Un_Syntrophorhabdus814_Un_Syntrophorhabdus2531_Un_Syntrophorhabdus483_Un_Syntrophorhabdus3928_Un_Syntrophorhabdus856_Un_Syntrophorhabdus4221_Un_Syntrophorhabdus4454_Un_Syntrophorhabdus1547_Un_Syntrophorhabdus2563_Un_Syntrophorhabdus4083_Un_Sh765B-­TzT-­294340_Un_Sh765B-­TzT-­293581_Un_Sh765B-­TzT-­293975_Un_Sh765B-­TzT-­293135_Un_Sh765B-­TzT-­292772_Un_Sh765B-­TzT-­29918_Un_Sh765B-­TzT-­29141_Un_Sh765B-­TzT-­29659_Un_Sh765B-­TzT-­292661_Un_Sh765B-­TzT-­293536_Un_Sh765B-­TzT-­29478_Un_Sh765B-­TzT-­293462_Un_Sh765B-­TzT-­294386_Un_Sva04852165_Un_Sva04852063_Un_Sva0485912_Un_Sva04851416_Un_Sva04853624_Un_Sva04851087_Un_Sva04853734_Un_Desulfobacca411_Un_Desulfobacca4976_Un_Desulfobacca4995_Desulfomonile_tiedjei1605_Un_Desulfomonile515_Un_Desulfomonile3399_Un_Desulfomonile172_Un_Smithella5061_Un_Smithella2951_Un_Smithella3551_Un_Smithella3391_Un_Smithella3715_Un_Smithella950_Un_Smithella264_Un_Smithella404_Un_Syntrophus2570_Un_Syntrophus5038_Un_Syntrophus5219_Un_Syntrophus319_Un_Syntrophus1962_Un_Syntrophus3815_Un_Syntrophus1264_Un_Syntrophus2908_Un_Syntrophus2261_Un_Syntrophus4254_Un_uncultured818_Un_uncultured1694_Un_uncultured1785_Un_uncultured2518_Un_uncultured5004_Un_Desulfovirga2013_Un_Desulfovirga2919_Syntrophobacter_wolinii2853_Un_Syntrophobacter3983_Un_Syntrophobacter3966_Un_Syntrophobacter5215_Un_Syntrophobacteraceae1259_Un_Syntrophobacteraceae1523_Un_SyntrophobacteraceaeDesulfosporosinus43F-­1404RDesulfarculaceaeDesulfatirhabdiumDesulfobacteriumDesulfosalsimonasSva0081_sediment_groupDesulfobacteraceaeDesulfobulbusDesulfocapsaDesulfurivibrioDesulfobulbaceaeDesulfovibrioUn_DesulfuromonadalesSyntrophorhabdusSh765B-­TzT-­29SmithellaSyntrophusSyntrophobacteraceae! " # $ % &!! " # $ % &!	  	   49	  From Figure 3.4, it appeared that some 94% OTUs were common to the BCRs, especially BCR1, BCR2 and BCR4, as well as some of the algae and inoculum pond samples. A bipartite network analysis was used to select these core OTUs, which were defined as the OTUs that are present in at least 50% of all the samples across all bioreactors (Appendix A, Figure A.6). In the bipartite network, AP, BCR1, BCR2 and BCR4 clustered closer together since they shared more SRM-related OTUs (the dark purple vertices in the center, which represent the shared taxa) than with BCR3. Representative sequences for the core OTUs (the dark purple vertices on Figure A.4) were compared using blastn against the NCBI 16S ribosomal RNA sequence databases to find their close cultured and environmental clone relatives so that their phylogenetic relationships could be explored. The core SRM community, which consisted of OTUs related to genera Desulfosporosinus, Geobacter, Desulfatirhabdium, Desulfocapsa and Smithella, was not limited to any particular phylogenetic linage but was distributed across all orders (Figure 3.5).   	  	   50	   Figure 3.5: Phylogenetic dendrogram of core SRM-related communities revealed from sequencing of V6-V8 of 16S rRNA with their neighbors from NCBI database. The bar size is proportional to the number of sequences in each OTU in each bioreactor. The sequences clustered in to OTUs based on 97% homology. 3.3.4 Distribution of Methanogen-related Sequences in Bioreactors A total of 19,996 methanogen-related sequences were found. Overall, 34% of all reads were classified in the order Methanobacteriales, 37% as Methanomicrobiales and 29% as Methanosarcinales. The pyrotag sequences were clustered into 76 97% similarity OTUs taxonomically distributed among 20 different families (Figure 3.6 a). Most (90%) of the reads assigned to Methanobacteriaceae, were found in BCR3 (Figure 3.6 b). The second most highly represented family with five 97% OTUs was Methanocorpusculaceae, which was also predominant in BCR3.   0.01Legend:Dataset  Read  countAlgae_pondBCR1BCR2BCR3BCR461100220Uncul  Desulfocapsa  sp  clone  GE7GXPU01CCD2H  OIl  res  bioreactor  [HM972525]Uncul  Desulfocapsa  sp  clone  F5OHPNU07ILWRC  Oil  sands  tailings  [HQ081546]8QFXO'HVXOIRWDOHDVSFORQH5&&)HDQGVXOIDWHíULFKZZ>(8@Desulfocapsa  sp  Cad626  meromictic  Lake  Cadagno  [AJ511275]Desulfocapsa  thiozymogenes  strain  Bra2    [NR  029306]125Uncul  Desulfovibrio  sp  clone  F5OHPNU07IBHS0  Oil  sands  tailings  [HQ081647]131Uncul  Desulfobulbus  sp  clone  GG5QJA201E07EM  Oil  res  bioreactor  [HM970588]Desulfobulbus  propionicus  strain  DSM  2032    [NR  074930]  074308QFXOVXOIDWHíUHGXFLQJEDFWHULXP/DNH&DGDJQR>$-@Uncul  bacterium  clone  HS4850B  10F    [JX434243]Desulfotalea  sp  NA22  16S  Tidal  flat  sediment  [AJ866933]Desulfobacterium  catecholicum    [EF442982]22618QFXO6\QWURSKXVVSFORQH)2+318,,;:2LOVDQGVWDLOLQJV>+4@8QFXO6\QWURSKDFHDHEDFWHULXPFORQH%DQR[LFODNHVHG>+4@Lake  Cadagno  clone  FR7290858QFXO6\QWURSKXVVSFORQH)0<$$/3.2LOVDQGVWDLOLQJV>+4@3198QFXO6PLWKHOODVSFORQH*(*;38$682LOUHVELRUHDFWRU>+0@8QFXO6PLWKHOODVSFORQH*E+&FRQWDTXLIHU>-4@2648QFXO6\QWURSKDFHDH+JíFRQWVHG>+(@Smithella  propionicaVWUDLQ/<3>15@Syntrophus  aciditrophicus  strain  SB    [NR  102776]2658QFXO'HVXOIXURPRQDVVSFORQH)2+318+9'.2LOVDQGVWDLOLQJV>+4@Desulfuromonas  acetoxidans  strain  DSM  684    [NR  121678]Uncul  bacterium  clone  EMIRGE  OTU  s8b4e  1987  acetate  amd  SS  [JX225607]4418QFXO'HVXOIREDFWHULXPVSFORQH)2+318+<<2LOVDQGVWDLOLQJV>+4@Desulfatitalea  tepidiphila    strain  S28OL1  tidal  flat  sediment  [AB719403]168QFXORUJDQLVPFORQH$*&$íEDFWHULRJHQLF)HR[LGHV>+4@'HQLWULI\LQJEDFWHULXPHQULFKPHQWFXOWXUHFORQH12$(>)-@8QFXO'HVXOIDWLUKDEGLXP+JíFRQWVHG>+(@Desulfatirhabdium  butyrativorans  strain  HB1  [NR  043578]66Weddell  Sea  cold  seep  clone  [FN429793]8QFXO'HVXOIREDFWHUDFHDHEDFWHULXPFORQH&$QR[LFODNHVHG>+4@8QFXOGHOWDSURWHREDFWHULXPFORQHDí8í$FLGPLQHGUDLQDJH>$<@Uncul  bacterium  clone  EMIRGE  OTU  s7t4e  2847  acetate  amd  SS  [JX224776]Uncul  Desulfobacteraceae  bacterium  clone  F5OHPNU07IFCMV  Oil  sands  tailings  [HQ067675]15842Geobacter  psychrophilus  strain  P35  [NR  043075]952*HREDFWHUVS3O\)HíUHGXFLQJHQY>()@297Desulfosporosinus  burensis  strain  BSREI1    [NR  109421]8QFXO'HVXOIRVSRURVLQXVVSFORQH6í2DNIRUHVWVRLO-;689990828987606910081100100100 947283100100599970991009299549810093100786310099651006499100100DesulfobacteralesSyntrophabacteralesDesulfobacteralesDesulfuro-­monadalesDesulfuro-­monadalesClostrid-­alesLog2(read  count) 4 6 8 1016 64 2561024Read  count	  	   51	  a)                                                                               b) Figure 3.6: A) Family-level distribution of methanogen-related sequences in all BCRs combined. B) Average number of reads per sample assigned to different family-level taxonomic groups in each of the features sampled (based on the 90% homology cut-off OTU table). In contrast to BCR3, the methanogenic communities observed in the algae and inoculum ponds, BCR1 and BCR4 were less prevalent. BCR2 contained more reads associated with methanogens than with SRM taxa.  Methanogens in BCR1, BCR2 and BCR4 were affiliated with genera Methanosarcina, Methanoregula and Methanosaeta (Figure 3.7). Methanogen-related OTUs in BCR2 were also found in the inoculum pond. In contrast, the large numbers of methanogen-related sequences in BCR3 were affiliated to different genera Methanobacterium and Methanocorpusculum. The methanogenic community within BCR3 was very different from that in the other bioreactors. Methanogens were rare in the soil samples.    !"#$%&'(%)#"*+%)"%",-./,!"#$%&')'*012)13%)"%",-4/,!"#$%&'5+)*'(+%)"%",./,!"#$%&'*"613%)"%",7/,!"#$%&'20+*+33%)"%",8/,!"#$%&'2%"#%)"%",49/,!"#$%&'2%*)+&%)"%",4-/,!"#!!"$!!!"$#!!"%!!!"%#!!" &'"  ()*"$" ()*"%" ()*"+" ()*","  -'"  ./01"*234")/567"829":3;812"(0/923<7/9:"=27>36/:39<063<232"=27>36/:3273<232"=27>36/:8090113<232"=27>36/92?513<232"=27>36/;0<9/@03<232"=27>36/</985:<513<232"=27>36/@3<72903<232"	  	   52	    Figure 3.7: Genus-level logarithmic heatmap (log 2) demonstrates relative distribution of methanogen-related sequences. OTU numbers and their Silva 111 assigned taxonomy are given on the right hand side. Sample names are column labels. It constructed by clustering the sequences in to OTUs based on 94% similarities. A bipartite network was constructed to visualize the core and reactor-specific methanogen-related OTUs (Appendix A, Figure A.7). This confirmed that some methanogen related 94% OTUs were common to BCR1, BCR2, BCR4 and the inoculum and algae ponds (dark purple vertices).  BCR3.3BCR3.1BCR3.6BCR3.2BCR3.4BCR3.5AP.2BCR1.5BCR1.6Soil.2AP.1BCR1.2AP.7AP.4Soil.1AP.3BCR1.3BCR1.1AP.6AP.5BCR1.4BCR4.2BCR4.4BCR4.3BCR4.1BCR4.5BCR2.3BCR2.1IP.3IP.1IP.2BCR2.5BCR2.4BCR2.6BCR2.2643_uncultured_Methanobacterium3206_uncultured_Methanobacterium1979_uncultured_Methanobacterium4358_uncultured_Methanobacterium142_uncultured_Methanobacterium3600_uncultured_Methanobacterium2148_uncultured_Methanobacterium4767_uncultured_Methanobacterium4_uncultured_Methanobacterium5026_uncultured_Methanobrevibacter3497_uncultured_Methanobrevibacter1995_uncultured_Methanobrevibacter3370_uncultured_Methanothermobacter1287_uncultured_Methanocellaceae4885_uncultured_Methanocellaceae562_uncultured_Methanocellaceae396_uncultured_Methanocorpusculum1769_uncultured_Methanocorpusculum2158_uncultured_Methanocorpusculum2_uncultured_Methanocorpusculum4596_uncultured_Methanocorpusculum5_uncultured_Methanoregula4094_uncultured_Methanoregula3334_uncultured_Methanoregula3287_uncultured_Methanoregula5040_uncultured_Methanoregula527_uncultured_Methanoregula4777_uncultured_Methanoregula4406_uncultured_Methanoregula3995_uncultured_Methanoregula3994_uncultured_Methanoregula3544_uncultured_Methanoregula892_uncultured_Methanospirillum898_uncultured_Methanospirillum166_uncultured_Methanospirillum707_uncultured_Methanospirillum223_uncultured_archaeon_WCHA2-­08827_uncultured_archaeon4752_Methanoculleus_sp._LH32_Methanoculleus_sp._LH655_uncultured_Methanoculleus3979_uncultured_Methanoculleus2744_uncultured_Methanoculleus3998_Methanofollis_ethanolicus112_uncultured_Methanosphaerula2598_uncultured_GOM_Arc_I391_uncultured_Methanosaeta2944_uncultured_Methanosaeta2715_uncultured_Methanosaeta6_uncultured_Methanosaeta1274_uncultured_Methanosaeta669_uncultured_Methanosaeta3948_uncultured_Methanosaeta2107_uncultured_Methanosaeta3946_uncultured_Methanosaeta4002_uncultured_Methanosaeta1646_uncultured_Methanosaeta3764_uncultured_Methanosaeta3212_uncultured_Methanosarcina1480_uncultured_Methanosarcina3228_uncultured_Methanosarcina1908_uncultured_Methanosarcina17_uncultured_Methanosarcina2522_uncultured_Methanosarcina5390_uncultured_Methanosarcina23_uncultured_Methanosarcina2024_uncultured_Methanosarcina1683_uncultured_Methanosarcina358_uncultured_Methanosarcina4124_uncultured_Methanosarcina2462_uncultured_Methanosarcina3131_uncultured_Methanosarcina4825_uncultured_Methanosarcina410_uncultured_Methanosarcina3560_uncultured_Methanosarcina5420_uncultured_MethanosarcinaMethanosarcinaMethanosaetaMethanoculeusMethanospirillumMethanoregulaMethanocorpusculumMethanobrevibacterMethanocellaceaeMethanobacterium! " # $ % !&& ' $ ( !" !)	  	   53	  Representative sequences for these core methanogen-related OTUs were compared against the NCBI 16S ribosomal RNA sequences databases using Blastn to find close cultures and environmental clone relatives in order to explore their phylogeny. Three clades of core methanogens were observed affiliated with Methanosaeta, Methanosarcina and Methanocorpusculum species (Figure 3.8).   Figure 3.8: Phylogenetic dendrogram of core methanogen-related communities revealed from sequencing of V6-V8 of 16S rRNA with their neighbors from NCBI database. The bubble size is proportional to the number of sequences in each OTU in each bioreactor. The sequences clustered in to OTUs based on 97% homology. 3.3.5 Network of Co-occurrence for the SRM- and Methanogen-related Taxa According to total read counts for each functional group (SRM and methanogens) it was apparent that some BCRs comprised more SRM than methanogens (BCR1 and BCR4), some contained comparable amounts of both SRM and methanogens (BCR2) and others were composed of mostly methanogen-related sequences (BCR3). In order to find out if specific SRM taxa were more likely to co-occur with specific methanogen taxa, a Pearson’s correlation coefficient test was applied to the 97% OTUs (Figure 3.9). Methanogenic taxa (red to pink colored nodes) tend to co-occur with other methanogenic taxa, similarly for SRM-taxa (blue to yellow). The only observed co-occurring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	  	   54	  taxa and methanogens belong to SRM affiliated to Syntrophobacterales (Syntrophobacter) and Methanosaeta.    Figure 3.9: Network of coexisting microbial lineages. Each square (node) represents a bacterial or archaeal 97% OTU. Color of the nodes represents their phylogenetic affiliations. Lines connecting two taxa indicate a significant co-occurrence relationship based on Pearson correlation coefficients (r) (r ≥ |0.8| and P-value ≤ 0.01). Pearson's correlations ≥ |0.8| were considered as connections to increase the confidence for detecting only robust co-occurring associations. The edge length is proportional to the value of correlation coefficient. Node size is proportional to the number of sequences in that OTU. The sequences clustered in to OTUs based on 97% homology.    337_Un_Desulfobacterium172_Un_Smithella504_Un_Desulfatirhabdium5178_Un_Desulfatirhabdium216_Un_Desulfobacula388_Un_Desulfobacteraceae392_Un_Desulfovibrio4110_Un_Desulfatirhabdium16_Un_Desulfatirhabdium975_Un_Desulfosalsimonas330_Un_Desulfovibrio3011_Un_Desulfobacteraceae66_Un_Desulfobacteraceae2570_Un_Syntrophus5057_Un_Desulfobacteraceae1504_Un_Desulfobacteraceae1339_Un_Sva0081_sediment_group411_Un_Desulfobacca29_Un_Nitrospiraceae118_Un_Nitrospiraceae1259_Un_Syntrophobacteraceae3193_Un_Sva0081_sediment_group74_Un_Desulfobulbaceae505_Un_Desulfatirhabdium140_Un_Sva0081_sediment_group5015_Un_Sva0081_sediment_group512_Un_Desulfocapsa131_Un_Desulfobulbus1358_Un_Desulfobacteraceae950_Un_Smithella158_Un_Desulfosalsimonas244_Un_Desulfobacter358_Un_Methanosarcina23_Un_Methanosarcina223_Un_WCHA2-08142_Un_Methanobacterium1421_Un_Desulfobulbaceae818_Un_Syntrophaceae3334_Un_Methanoregula3287_Un_Methanoregula6_Un_Methanosaeta5_Un_Methanoregula5040_Un_Methanoregula4752_Un_Methanoculleus32_Un_Methanoculleus 396_Un_Methanocorpusculum112_Un_Methanosphaerula391_Un_Methanosaeta827_Un_Methanomicrobiales3994_Un_Methanoregula3544_Un_Methanoregula3995_Un_Methanoregula669_Un_Methanosaeta	  	   55	  3.4 Discussion  The bioreactors surveyed in this study contained microbial consortia much more diverse than those previously described for laboratory-based bioreactors [28, 45, 53, 54, 102, 115, 116, 146, 263]. This suggests a much wider metabolic potential and occurrence of many more metabolic processes in field-based BCRs. This is likely due to the wide range of carbon sources produced from complex organic matter degradation. Additionally, field-based bioreactors are exposed to many avenues for colonization by naturally-occurring microbial species within the organic materials, inocula and also in the influent water.  A growing body of evidence indicates that a more diverse community provides a larger contribution to ecosystem functions and service compared with a less diverse counterpart [17]. There are two principles to explain how biodiversity affects the ecosystem functioning, and ecosystem productivity. First, different species use slightly different resources. Therefore, communities with higher diversity are more productive because more of the overall resource is used.  Secondly, communities with higher diversity are more productive because it is more probable that they contain species with a large effect on ecosystem functioning [151].  3.4.1 Metal-removing BCRs Share Several Microorganisms Even Though They Are at Different Geographical Locations Except for BCR3, the bioreactors contained very phylogenetically similar taxonomic groups and shared several species-level (97%) OTUs despite being located at different mines and inoculated with different sources of inoculum. Also, several sequences within the core group of SRM had close relatives also found in metal contaminated sites (Figure 3.5); for instance, Hg polluted sediment (e.g. HE648179.1, unpublished), sulfate-zinc-producing biofilm in acid mine drainage system containing heavy metals [133], bioreactor treating metals at low temperature [31], and passively operated compost-based bioreactor for remediation of acidic [197], highly iron, metal and sulfate-rich industrial waste water [51]. This suggests that there are core sulfate-reducers adapted to metal-rich environments that are phylogenetically closely related to each other even though the metal-rich environments are geographically dispersed. 	  	   56	  3.4.2 The Type of Carbon Source Used in BCRs Affects the Composition of the Microbial Community Despite operating under similar conditions of pH, temperature and anoxia, the structure of the microbial community in BCR3, which used pulp mill biosolids, was very different from the other BCRs where woodchips and manure (and hay for BCR4) were used. Although there were other differences between the bioreactors, and there are not enough data in the current study to attribute the distinct microbial community of BCR3 specifically to the nature of its carbon source there are some studies in the literature that have shown that carbon source type does influence microbial community structure [10, 25, 59, 102, 212, 263]. Schmidtova et al. (2011) tested under the same chemical conditions several different organic materials that are typically used in BCRs for their ability to support the growth of sulfate-reducing microorganisms and the process of sulfate reduction [212]. A comparison of microbial communities using 16S rRNA gene clone library sequencing revealed that the pulp mill biosolids supported a very different microbial community and only Desulfovibrio desulfuricans-related sequences were found in pulp mill biosolids dataset. Comparing silage, compost, hay, cattails and pulp mill biosolids in their study, lowest rate of sulfate reduction was also observed in presence of pulp mill biosolids [212]. In contrast, it has been shown in the field and in the laboratory that pulp mill biosolids can support a diverse SRM community and measurable sulfate-reduction when used fresh from the mill (unpublished data from Baldwin Lab). Chemical analysis of pulp mill biosolids revealed the presence of resin acids and phenolic compounds (unpublished data from Baldwin Lab), but it is not known if these are toxic to microorganisms. Hiibel et al. (2011) also studied the effect of three organic substrates (ethanol, hay and pine wood chips, and corn) on the microbial community structure in metal removing BCRs [102]. Based on their results, different organic matters had different microbial compositions and changing the carbon source shifted the relative quantities of total microorganisms and sulfate-reducing bacteria [102]. Bioreactors that contained more readily available carbon sources were dominated by sulfate-reducing microorganisms, while more complex structured carbon sources such cellulose, hemicellulose and lignin supported the growth methanogens and fewer SRM [102]. Some experiments measured the incorporation of radio-labelled C13 into peptides of different microorganisms [235]. They found that there were three different functional groups of fermentative microorganisms including Bacteroidetes, Firmicutes (Clostridiales) and Chlorobi, respectively, that displayed time-specific abundances implying the existence of a cascade of carbon utilization with 	  	   57	  time. First, a Clostridial group fermented complex structured components such as benzene while fixing CO2, then a sulfate-reducing group (mainly within Deltaproteobacteria) used metabolites released during fermentation, and eventually a scavenger group affiliated with Bacteroidetes/Chlorobi become predominant. Drawing on the conclusions of that study, it is likely that the microbial communities with the BCRs in this study may have evolved significantly over time from when they were commissioned until when they were sampled.  In fact, the degree of decompostion of biosolids within BCR3 correlated with the presence of Methanogens 29. Additionally, several studies of bacterial communities in lakes document the dependency of Bacteroidetes species on types of organic carbon. Bacteroidetes flourished during periods or at sites where dissolved organic carbon (DOC) was high [65, 66].  3.4.3 BCRs are Distinct From Other Anaerobic Environments in Terms of Their SRM Community By comparing the BCRs at Site 1 with a natural anaerobic organic rich environment at the same mine site (inoculum pond) the results suggested that the overall microbial community structures were similar at the phylum level, whereas the BCRs environment selected for specific sulfate-reducing microorganisms at the genus level that were more adapted to the presence of metals. The SRM community common to the metal-rich bioreactors in this study was phylogenetically diverse but comprised distinct taxa (within the class Deltaproteobacteria) associated with other metal-rich or saline environments indicating that they might be specialists at surviving under these conditions. Most BCR-associated SRM were affiliated to the metabolically versatile Desulfobacteraceae. The distinctive characteristic of members of this family is complete oxidation of a broad range of substrates. Sequences found in the BCRs were mostly affiliated to acetate-oxidizing sulfate-reducing genera Desulfatirhabdium, Desulfobacterium, Desulfobacula [11, 175, 252]. The genus Desulfatirhabdium in this family appears to be particularly important since quite large numbers of reads affiliated with Desulfatirhabdium were frequently detected in all bioreactors. Cultured species within the Desulfobacteraceae family have capabilities such as resistance against viral attacks, anaerobic degradation of a variety of complex structure components, degradation of protein compounds, resistance to oxygen stress, which gives them a competitive advantage over other SRM in the habitat [253]. Strains in this family possess several enzymes that are responsible for oxygen 	  	   58	  detoxification that assure its survival in oxic-anoxic transition zone that exist in BCRs (Chapter 1, Figure 1.1) [253].  The second most prevalent family among the core SRMs in the BCRs was Desulfobulbaceae (sequences mostly affiliated to genera Desulfocapsa and Desulfotalea). Species within this family are specialized in disproportionating inorganic sulfur compounds make them suitable candidate for metal-removing bioreactors [79, 80]. The disproportionation of sulfur and thiosulfate, is an important step in sulfur transformation since, they are cleaved to sulfate and sulfide without any oxido-reduction reaction. This reaction provides sulfate and sulfide for both activity of SRM and precipitation of metals by sulfide [79, 80]. This capability has been reported for very few other bacterial strains and most importantly, continuous removal of the produced sulfide by mechanisms such as precipitation with metal ions, is essential to keep the disproportionation reaction exergonic [80, 104].  Several capabilities make Geobacteraceae (other prevalent family in BCRs) suitable for bioremediation including; i) chemotaxic toward metals ii) capability for direct interspecies electron transfer and iii) anaerobic complete oxidation of wide variety of organic compounds and metals including iron, radioactive metals and petroleum compounds coupled to the reduction of wide variety of metals and humic matters [145, 154, 163, 177, 200]. The Desulfosporosinus-related sequences were the only non-Deltaproteobacteria found in all BCRs and the algae pond. Some Desulfosporosinus species are able to tolerate copper concentrations of up to 236 mM, which is much higher than the concentrations reported for other SRM so far [1]. Other cultured species are resistant to highly acidic conditions [176]. Desulfosporosinus species have the capability to use a variety of electron acceptors other than sulfate such as metals As (V) and Fe (III) as well as nitrate has been reported [137, 144, 189]. These qualities allow them to grow and be active under the harsh conditions (high metal concentrations) of biochemical reactors. Reasons given for their tolerance of metals include the ability to sorb metal ions, such as Cu2+ to their cell walls [245, 249]. One response mechanism that many SRM use to protect themselves from heavy metals is to release extracellular polymeric substances that contain negatively charged moieties, which act as metal-binding ligands [245, 249]. Beside, researchers suggested two CopA-like CPx-type ATPases (DOT_2451 and DOT_2536) and a polyphosphate kinase-phosphatase couple, DOT_3559 and DOT_4690, systems play key role in Desulfosporosinus sp. OT, to tolerate high concentrations of copper [1].  	  	   59	  Several Deltaproteobacteria environmental groups were prevalent and core in BCRs, but not in the environmental samples, revealing that the repertoire of metal-adapted SRM genera is more expansive than previously thought. OTUs classified as Sh765B-TzT-29 were present in most BCRs. The original clone Sh765B-TzT-29 came from uranium mining waste piles and mill tailings (Genbank Accession number AJ519630). Many clones closely related to the most prevalent BCR Sh765B-TzT-29 OTU (OTU number 141) came from activated sludge bioreactors treating various types of wastes. OTUs affiliated with this group have been found in sulfate-rich environments such as methane seeps where it was suspected that they might be involved in sulfate reduction coupled with anaerobic oxidation of methane [221]. Attempts made to culture one affiliate with methane were not successful. Previous to our study, Desulfovibrio species were the most often identified SRM in BCRs treating metal-rich water in defined and complex carbon sources [103, 187]. However, these studies were on SRM in laboratory bioreactors and not field-based pilot-scale systems. In laboratory bioreactors, the experimental setup, including inoculum used, type of carbon source, type of bioreactor, and apparatus configuration, can influence the microbial community [25], which may be very different from those found in field bioreactors. Thus laboratory-based bioreactor studies may not yield information useful for actual scale-up and operation of field-based systems. This and other studies are revealing that a diverse community exists of SRM adapted to metals than previously thought. It will be important for future reactor experiments to use consortia containing these organisms as inocula to see if metal removal efficiency is improved.  3.4.4 Sulfate-reducing Microorganisms Are Prevalent in Metal-removing Bioreactors More so Than Methanogens That Are Potential Competitors for Carbon Sources  Based on our study, SRM and methanogens appear not to be mutually exclusive and are able to co-exist since all BCRs contain both. However, sulfate-reducing bacteria were more prevalent in most of the bioreactors than methanogens. The results obtained from the network of co-occurrence study is in agreement with the finding from the relative abundance comparison since it did not show any negative association between SRM and methanogens. In those bioreactors where SRM and methanogens coexisted but with fewer methanogen- than SRM-related sequences, the majority of SRM (Desulfobacterium, Desulfatirhabdium, Desulfobacula) and methanogens (Methanosaeta and Methanoregula) were acetotrophic. In biochemical reactors containing complex organic carbon, a 	  	   60	  large diversity of microbial groups degrade the broad range of, mostly lignocellulose, substrates and as a result produce acetate. Only a few microorganisms are able to oxidize acetate. Acetotrophic SRM can couple its oxidation to sulfate reduction. Acetate can be used also by acetotrophic methanogens to produce methane. Accumulated acetate reduces alkalinity and inhibits sulfate reducing and acetate producing microorganism, two phenomena that are unwanted in wastewater treatment. Therefore, one model that can be proposed is; existence of acetotrophic sulfate reducing bacteria are beneficial for BCRs where they simultaneously remove undesirable acetate and couple this with the desired reaction for metal removal. However, if acetotrophic SRM are not present in the system or the BCR is sulfate-depleted, acetotrophic methanogens would benefit by using acetate. This finding is in agreement with some other reports describing advantages of having low proportion of methanogens in sulfidogenic bioreactors where there might not be acetotrophic SRM (such as Desulfovibrio) [102].    	  	   61	  CHAPTER 4 KEYSTONE MICROORGANISMS IN BIOCHEMICAL REACTORS TREATING MINING-INFLUENCED WATER   4.1 Synopsis The microbial community structure of BCR3 was different from those in the other BCRs (Chapter 3). BCR3 used pulp mill biosolids, whereas the other BCRs used wood chips and/or hay as carbon sources. This might have been one factor that contributed to the very different microbial community of BCR3 versus the other BCRs. Emerging evidence suggests that the type of carbon sources in an ecosystem is one important abiotic factor that determines the structure of its microbial community (Chapter 3, Section 3.4.2) [10, 25, 78, 102, 212, 235, 263]. There were other differences between the BCRs that might have also contributed to their microbial community dissimilarities. The BCRs received inocula from different sources, the influent water contributed to a different geochemical environment and their microbial communities might have evolved somewhat during their operation. Both abiotic and biotic factors regulate the structure of microbial communities (Chapter 1, Figure 1.4) [6, 88, 143, 223]. The work described in this Chapter attempted to address which biotic factor(s) may affect the structure of communities in these BCRs by using network analysis. In this Chapter, two datasets were constructed: one contained OTU relative abundance data for BCR3 and the other contained OTU relative abundance data from BCR1, BCR2 and BCR4 combined. Co-occurrence networks were constructed by calculating the Pearson correlation coefficient values for each dataset. Topological features of each network such as betweenness centrality (BC) and closeness centrality (CC) were calculated to identify putative keystone taxonomic groups in each network (ecosystem). Ultimately, a schematic model was constructed to propose biotic and abiotic factors that could be involved in regulation of microbial communities’ structure in each ecosystem (methanogen-rich (BCR3) or sulfate-reducing bacteria-rich (other BCRs)).  	  	   62	  The following hypotheses were tested: 1. Since the microbial community structure of BCR3 was different from those in the other BCRs, the keystone microbial groups must also be different.  2. The keystone microorganisms in BCR3 employ different mechanism(s) than those in the other BCRs to regulate the structure of community. 4.2 Materials and Methods  4.2.1 Dataset Preparation The first dataset for BCR3 contained 207,352 sequences from 38 samples, and the second dataset contained 235,168 sequences from 42 samples from BCR1, BCR2, and BCR4. The rationale for combining the samples from BCR1, BCR2, and BCR4 into one dataset was because of their similarities in microbial community structure (Chapter 3). Sequences were clustered into OTUs based on 94% sequence similarity (equivalent to genus level), and only OTUs with more than five sequence reads in total were considered for further analysis (Appendix B.1). We chose 94% similarities in sequences (but not 97% that is equivalent to species level) to exclude artifact that may occurs because of inaccurate assignment. 4.2.2 Analysis of Co-occurrence and Construction of Network of Correlations  For each dataset, co-occurrence between microbial taxa were examined by calculating all pair-wise Pearson rank correlations between microbial OTUs using Equation 1 (Chapter 1) and the script enterotypes_graph.pl obtained from https://github.com/kishori82/Correlation_Network. A valid interaction event was considered to be a robust correlation if the Pearson correlation coefficient (r) was either equal to or greater than 0.8 or, equal to or less than −0.8 and statistically significant (p≤0.05). The p-values were calculated with Equation 2 (Chapter 1) using the script enterotypes_graph.pl obtained from https://github.com/kishori82/Correlation_Network. The cut off of r ≥  |0.8| was chosen to increase the confidence for taxa co-occurrence. Two networks (one for each dataset) were constructed. The network structure was explored and visualized with the Cytoscape computer software version 3.0.1 using edge-weighted (according to the correlation coefficient) spring embedded layout (Appendix B.2) [217].  	  	   63	  4.2.3 Determination of Keystone Taxa in Each Network To determine putative keystone taxa, the values of betweenness centrality (BC) and closeness centrality (CC) were calculated using Equations 3 and 4 (Chapter 1) with the script compute_centrality.py obtained from https://github.com/kishori82/Correlation_Network and Cytoscape (Appendix B.3) [217]. In each network, the top 5% OTUs with the highest values for BC and the top 5% OTUs with the highest value for CC were selected for further investigation as possible keystones [164, 242, 257]. This cut-off (top 5%) is arbitrary and was chosen to investigate only those OTUs that possess the highest values for centrality in the network. Scatter plots were created in Cytoscape to show the relationship between taxon abundance vs. BC and taxon abundance vs. CC for each network. Keystones OTUs were highlighted in the co-occurrence network using the circular sorted (according to BC or CC rank) layouts in Cytoscape [217]. Heatmaps were constructed to show positive and negative correlations of keystone taxa with other taxonomic groups. 4.3 Results  4.3.1 Summary of Network Statistics for BCR3 Compared with the Other BCRs The value of Pearson’s correlation coefficient, r, was calculated for each pair of OTUs and a network of co-occurrence was constructed for each dataset (Appendix A, Figure A.8 and Figure A.9). In both networks, a similar percentage of the total 94% OTUs (26% in BCR3 versus 25% in the other BCRs) appeared in each network having a significant positive or negative correlation with other OTUs (Table 4.1). In both networks, the numbers of positive correlations were higher than negative correlations (77% of all interactions were positive in BCR3 and 99% of interactions in the other BCRs were positive). More negative correlations were observed in BCR3 (23%) than in other BCRs (1%).    	  	   64	  !"#$ %&'()*!"#+,-&./*0123()*-4*+.25/(+ $6 78,-&./*0123()*-4*+(91(0:(+ 8;<=$>8 8$>=?@6,-&./*0123()*-4*%,A+ B@< ?=<8;,-&./*0123()*-4*0-C(+ 8>B 7$7,-&./*0123()*-4*+DE0D4D:.0&*:-))(/.&D-0+ ?=@B? 7=?;>,-&./*0123()*-4*+DE0D4D:.0&*5-+D&DF(*:-))(/.&D-0+ ?=$;B 7=;@8,-&./*0123()*-4*+DE0D4D:.0&*0(E.&DF(*:-))(/.&D-0+ $<7 7$Table 4.1: Description of datasets that were used to construct the co-occurrence networks.    4.3.2 Keystone Microbial Taxa in BCR3 and Their Correlations with Other Taxa OTUs with the highest values for BC were restricted to certain phylogenetic lineages mostly within Actinobacteria (Corynebacterium), Firmicutes, and Alpha-Proteobacteria in BCR3 (Figure 4.1 a). The keystone OTUs (top 5% OTUs with the highest values for BC) had more negative correlations (57%) than positive ones with other taxa in this bioreactor (Figure 4.1 b). These negative correlations were mostly between keystone taxa and taxa affiliated to Bacteroidetes, Proteobacteria and Planctomycetes.  	  	   65	   Figure 4.1: A) Co-occurrence network showing significant correlations of microbial taxa (based on 94% sequence similarity OTU-table) in BCR3. Lines connecting two microbial taxa represent the correlations between them (r ≥ |0.8| and p-value ≤ 0.05). The sizes of nodes in the network are proportional to the value of BC and colored based on their phylogenetic affiliation. The network is circular sorted based on the value of BC. The top 5% OTUs that possess the highest value for BC were shown. B) Heatmap showing positive and negative correlations of keystones with other taxa based on the value of r. Each cell is colored based on the average of negative and positive correlations (r) that keystones make with corresponding taxa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	  	   66	  OTUs with the highest values for CC were mostly assigned to Actinobacteria, Firmicutes and Spirochetes (Treponema) in BCR3 and were positively and negatively correlated with other taxa (Figure 4.2).  Figure 4.2: a) a) Co-occurrence network showing significant correlations of microbial taxa (based on 94% sequence similarity OTU-table) in BCR3. Lines connecting two microbial taxa represent the correlations between them (r ≥ |0.8| and p-value ≤ 0.05). The sizes of nodes in the network are proportional to the value of CC and colored based on their phylogenetic affiliation. The network is circular sorted based on the value of CC. The top 5% OTUs that possess the highest value for CC were shown. B) Heatmap showing positive and negative correlations of keystones with other taxa based on the value of r. Each cell is colored based on the average of negative and positive correlations (r) that keystones make with corresponding taxa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	  	   67	  4.3.3 Keystone Microbial Taxa in Other BCRs and Their Correlations with Other Taxa In the other BCRs, OTUs with the highest values for BC were dispersed across several phylogenetic lineages mainly affiliated to Beta-, Gamma-, Delta - Proteobacteria and Bacteroidetes (Figure 4.3 a). The keystone OTUs (top 5% OTUs with the highest values for BC) had mostly positive correlations (94%) with other taxa (Figure 4.3 b). 	  	   68	   Figure 4.3: a) Co-occurrence network showing significant correlations of microbial taxa (based on 94% homology cut-off OTU-table) in Other BCRs. Caption is same as Figure 4.1. 000000000000000000000000000000000000000000000000000000 000 0 00000000000000000000000000000000000000000 000000000000000000000000000000000000000000000000000000000000000000000000000 000000000 0 000 0 00000000000000000000000 0!"#$""%%"&&'("%#)*+,#-'./0' ./.0'./.1' ./.2'!"#$"%&'()*+%,-./0%'(#"0*1#.'(./0%'(#"0*324'1#.'(./0%'(#"0*334546474'80%'(#."9('()*+%"9./0%'(#"0*:;<.#.=(>"*?&#@0#%;0(.'0*AB"#.%;('()*C';(#)*!*4'54'! " # $ !% !& !' !" &%!"!#!$%&!' !"!#!( !"!#!$)&!'!"#$%&'"$()*+,!"#$%#&'%()*%+&*,-).%&*(#,%/%&*(#),0(*(12'3)#)43(5,2$%-).%&*(#,%6,#7,&"*(183%-&*)7$&(*(8#)*().%&*(#,%9:,#)&'(*(;<<;=>',?).,"7 !" !# !" !$ !! !% !! !$ !"@AB=C-&"3*"#(0 !! !& !' !" !" ( !) !' !!;<D@=E%#,)F)#%5 !! !& !" !" !" !! !% !* !!AG;H=+&'#)7).%&*(# !! !& !' !" !" ( !% !* !$@H=I(1"34"#)7)-%1 !! !! !' !% !% !! !" !' !!<J=I(1"34"#).%&*(# !& !& !" !% !% !! !" !* !!;@@=9'<HB/KL?LKDG !& !& !" !" !" !& !" !' !!;BJ=97,*'(33% !! !& !' !% !% !& !% !' !$;AMM=81("0)7)-%1 !$ !! !" !* !* !! !" !' !!DJ<J=63%F).%&*(#,"7 !& !& !' !" !* !! !" !* !$DJ<J=63%F).%&*(#,"7 !! !& !' !% !* !! !" !* !$DDA=+&,0).%&*(# !& !& !% !% !% !! !% !% !!;D;J=C-&"3*"#(0 !$ !& !" !" !" !& !" !" !&A@D=N(:*)3,-(% !) !& !" !" !" !& !" !" !&ABM=C-&"3*"#(0 !) !& !" !" !" !! !" !" !!@;DJ=N)O,-,3,(% !$ !& !" !" !" !& !" !" !&D;@A=2%-0,0%*(P0,F,1,)-=Q8;; !) !& !" !" !" ( !" !" !$;MMB=C-&"3*"#(0 !& !& !% !% !% !& !% !% !&<<M=R(77%*,7)-%1 !& !) !! !& !% ( !% !& !$HM;=C-&"3*"#(0 !& !) !% !% !% ( !% !% !$	  	   69	  OTUs with the highest values for CC in the other BCRs were mainly assigned to Beta-, Gamma-, Delta - Proteobacteria and Euryarchaeota and had mostly positive correlations with other taxa (Figure 4.4).  Figure 4.4: a) Co-occurrence network showing significant correlations of microbial taxa (based on 94% sequence similarity OTU-table) in other BCRs. Caption is same as Figure 4.2. 000000000 0 0 0 00000000000000000000000000000000000000000000000000000000000000 0 000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000!"#$%&%$$'!%&()*"+(,'-./' /.-'/.0' /.1'!"#$"%&'()*+%,-./0%'(#"0*1#.'(./0%'(#"0*223'1#.'(./0%'(#"0*234546473'80%'(#."9('()*+%"9./0%'(#"0*:;<.#.=(>"*?&#@0#%;0(.'0*AB"#.%;('()*C';(#)*!*3'43'! " # $ !% !& !' !" &%!"!#!$%&!' !"!#!( !"!#!$)&!'!"#$%&'"$()**+!"#$%#&'%()*%+&*,-).%&*(#,%/%&*(#),0(*(12'3)#)43(5,2$%-).%&*(#,%6,#7,&"*(183%-&*)7$&(*(8#)*().%&*(#,%9:,#)&'(*(;<=>?(*'$3)*(-(#% !" !" !# !" !# !! !# !$ %@;A>97,*'(33% !" % !# !& !& % !' !# !'B;B>C(1"34)7)-,3( % ( !# !) !) !' !) !# !"@<;>D().%&*(# % % !$ !$ !$ !" !$ !( !";=E>9$-*#):')#'%.0"1 !" ( !) !) !) !" !) !# !"<F>"-&"3*"#(0 % % !* !) !) !" !) !# !"@@GB>"-&"3*"#(0 % !" !) !) !) !! !# !# !"EA;@>?(*'$3)7,&#).,"7 !" !" !$ !& !& !" !& !) %@@F>2%-0,0%*"1>8%#H%#&'%("7 !) !' !" !" !" !) !" ( !&G>?(*'%-)1%(*% !# !) % !" !" !) !" ( !)@=<@>2%-0,0%*"1>8%#H%#&'%("7 !& !& !" !" !" !& !" % !)@G;I>63%H).%&*(#,"7 % !" !) !& !& % !& !* !"IBF>8#)3,5,.%&*(# ( !! !# !) !) % !) !) %I<=>63%H).%&*(#,"7 ( !" !) !$ !$ % !$ !$ %IEE>J-&"3*"#(0 % !" !# !) !) % !) !# !!;;<>J-&"3*"#(0 % !! !) !) !) !" !) !# !!@FI>J-&"3*"#(0 !" % !& !) !) !" !) !# %GB>J-&"3*"(0 !" !" !! !' !' !' !) !' !'BG>2%-0,0%*(>0,H,1,)->K8@@ !' !) !) !& !" !' !& !& !&@IGE>J-&"3*"#(0 !! !! !' !" !) !' !" !" !"	  	   70	  4.3.4 Relationship between Relative Abundance and the Values of Betweenness Centrality and Closeness Centrality To be able to understand how taxon abundance and the centrality measures (BC, CC) are related, scatter plots of the relative abundances of OTUs versus the values of BC and scatter plots of the relative abundances of OTUs versus the values of CC were constructed for each network (Figure 4.5). There were no strong correlations between taxon abundance and centrality measures in these networks. As the plots show, both abundant and not-abundant taxa can possess high value of BC and CC. Remarkably, the most highly relatively abundant OTUs in the networks were not among those identified as putative keystones.   a)                                                                                   b)  c)                                                                                   d) Figure 4.5: Relationship between rank relative abundance data from the total number of OTUs at genus level found a) in BCR3 and BC, b) in BCR3 and CC, c) in other BCRs and BC and d) in other BCRs and CC. 	  	   71	  4.4 Discussion  	  4.4.1 Limitations of the Study Network analysis is emerging as a powerful approach for using high-throughput sequencing datasets to develop hypotheses to explain microbial community structure. However, there are some limitations inherent in the methods that are used to make these networks that are important to be aware of prior to interpretation of the results. These include:  (I) According to Taylor et al. (1990) a correlation coefficient that measures an association between two variables is an abstract measurement for which there is no direct precise interpretation. Low values of r do not account for significant variation in the value of the dependent variable[202]. Less stringent cutoffs decrease the reliability of the results. Therefore, in this study to increase the confidence for detecting only strong correlations, conservative cutoffs  (r ≥ |0.8| (p ≤ 0.05)) were used to generate the networks. (II) Because taxon assignment at lower taxonomic levels is not always precise, a link between two closely related taxa may be entirely due to mutual cross-assignment. Therefore, we only considered OTUs based on 94% sequence similarities to disentangle ecological signals and artifacts that may occur due to inaccurate assignment. (III) Uneven amounts of abundance-yielding material (e.g. the percentage of high-quality reads, the amount of extracted and sequenced DNA) can distort results and give some artificial correlations. Therefore, samples of equal mass were processed and equal concentrations of DNA were used for sequencing, which was done using the same method and machine. (IV) A feature of microbial relative abundance data is the presence of a large percentage of zeros in the relative abundance matrix. This could be problematic because of their ambiguous interpretation: a zero in a data set obtained from 16S or metagenomic sequencing can either mean that the taxon is indeed absent from the sample or, more likely, that its abundance is below the detection level. We alleviated this problem by data reduction through filtering out the data by setting a cutoff for a minimum number of non-zero columns for an OTU. For an OTU to be used in the analysis, the minimum percentage of non-zero columns was 50%.  	  	   72	  (V) Faust et al. (2012) reported that the choice of correlation metric has a large impact on the resulting network [73]. Berry et al. (2014) evaluated several commonly used metrics used to make networks. Based on their findings, Spearman and Pearson correlation coefficients were the most reliable methods for detecting interactions in meta-communities from relatively similar communities and abundance data [19]. Therefore, we selected Pearson correlation coefficient test to calculate associations.  VI) To provide a summary of the best methodological considerations for construction of networks, Berry et al. (2014) examined the effects of sample number, species richness and species evenness (alpha- and beta-diversity) on co-occurrence network sensitivity and specificity. Based on their results, the specificity of the network was increased with an increasing number of samples until it plateaued at about 25 samples [19]. Therefore, we constructed two networks one with 38 samples from BCR3 and another with 42 samples from BCR1, BCR2 and BCR4. The similarity of communities had a large effect on network sensitivity and samples with relatively high similarities in microbial community were used for constructing the networks [19]. We combined samples from BCR1, BCR2 and BCR4 that are similar phylogenetically (Chapter 4, Figure 3.2). Community evenness did not directly affect co-occurrence network sensitivity and specificity [19]. 4.4.2 BCR3 Versus Other BCRs; Keystone Microorganisms and Their Correlations with Other Taxa  Network analyses predicted that possible keystone microorganisms in BCR3 were phylogenetically different from those in the other BCRs and were negatively correlated with several other taxonomic groups (Figure 4.1 b). Antagonistic associations between species affiliated to Firmicutes, Actinobacteria and Alpha-Proteobacteria (keystones in BCR3) with other microbial species have been reported [89, 93, 148, 149]. Long et al. (2013) studied the regulation of species dynamics within a microbial mat and reported antagonistic interactions between bacteria in the phylum Firmicutes against isolates clustered in different phylogenetic groups. This was postulated as a potential biotic mechanism that regulates community structure within hypersaline microbial mats [149]. They isolated species from the microbial mat that were classified in the phylum Firmicutes and found that these produced secondary metabolites with antimicrobial properties. The authors suggested that these isolates could alter community structure and function by inhibiting the growth of some species and opening niches for others [149]. The production of these antagonistic 	  	   73	  metabolites might result from competitive pressure for limited substrates and can potentially lead to changes in the nature of the biogeochemistry within the microbial mats [149]. Grossart et al. (2004) studied the microbial community structure, inter-species interactions and antagonistic activities of isolates from microbial aggregates on organic matter from German Wadden Sea [93]. Based on their findings, growth of abundant isolates belonging to the phyla Bacteroidetes and Proteobacteria was affected by the presence of low abundance microorganisms affiliated to Actinobacteria, Firmicutes and Alpha-Proteobacteria because of their high antagonistic activities. They reported that inhibitory activity of species within these phylogenetic groups significantly influences inter-species interactions and may impact microbial community structure and function in terms of degradation of organic matter in aquatic environments [93]. Giudice et al. (2007) also reported high antagonistic activity for Actinobacteria and Alpha-Proteobacteria against other isolates by investigation of interspecies interactions among Antarctic isolates [89]. Actinobacteria in particular Corynebacterium generally inhibited isolates clustered in different phylogenetic groups by production of a broad spectrum of antibacterial agents [89]. According to studies on microbial interactions, the antagonistic interactions by Firmicutes, Actinobacteria and Alpha-Proteobacteria against other microorganisms may become a key factor in regulating of bacterial populations specially in microbial aggregates [91, 148].  In the other BCRs, possible keystone microorganisms were mostly affiliated to taxonomic groups containing species with versatile metabolic potentials, i.e. those that can grow under a broad range of nutritional and environmental conditions and are able to make protagonistic (syntrophic) interactions with other microorganisms [210]. For example, species of Variovorax are able to degrade a variety of components such as organic sulfur compounds (e.g. sulfolane and 3-sulfolene, 2-mercaptosuccinic acid (MS), 3,3-thiodipropionic acid (TDP), aromatic sulfonates), aromatic compounds (e.g. dimethyl terephthalate, polychlorinated biphenyls, 2,4-dinitrotoluene, homovanillic acid, veratraldehyde, linuron, 2,4-dichlorophenoxyacetic acid, and (2,4-D) anthracene), polymers (e.g. poly (3-hydroxybutyrate), chitin, and cellulose), under a wide range of environmental conditions (e.g. at 10oC or below) [210], [4, 20, 26]. Their roles in removal of metal ions (e.g. arsenite, yttrium) and many other toxic compounds from polluted water have been reported [210], [83, 117, 162]. Many of the chemical compounds metabolized by species within these taxonomic groups are present in BCRs as degradation products from the plant-derived carbon sources. Few microorganisms are able to 	  	   74	  consume the complex organic compounds that result from degradation of plant matter (listed above). Species with the genus Variovorax can be mediators for taking place of several biochemical reactions in BCRs, by providing ready to use nutrients for other microorganisms. One of the possible keystone taxa identified in BCR1, BCR2 and BCR4 was Variovorax-related. Syntrophic degradation of linuron by a bacterial consortium containing linuron-degrading Variovorax strains has been reported [55]. McInerney et al. (2008) listed the phylogenetic affiliations of all known microorganisms capable of syntrophic metabolism [167]. Several keystone microorganisms in BCR1, BCR2 and BCR4 were correlated within syntrophic microorganisms within the genera Syntrophorhabdus, Smithella, and Geobacter. A syntrophic relationship between Syntrophorhabdus aromaticivorans and Methanospirillum hungatei for complete mineralization of phenol has been reported [178, 193].  Syntrophorhabdus aromaticivorans also co-operates with Desulfovibrio sp. for degradation of isophthalate coupled to sulfate reduction [178, 193]. The relationship between these species are based on cross-feeding, in which one strain (Syntrophorhabdus) degrades the primary energy source and excretes an intermediate that is used as energy source by the second strain (Methanospirillum hungatei, or Desulfovibrio sp.,) [178, 193].  Some of the potential keystones in BCR1, BCR2 and BCR4 (i.e. those that were Flavobacterium-related) are related to microbial species known to produce antagonistic agents, and were sensitive to antagonistic interactions [148]. These observations suggest that the possible keystone microorganisms in BCR3 might employ different mechanisms than those postulated for the other BCRs in order to regulate the structure of its community. Protagonistic (syntrophic) interactions were postulated to be the potential biotic mechanisms that regulated the microbial community structures in the other BCRs. Deviation of the microbial community structure in BCR3 from those in the other BCRs might be due to activity of keystone microorganisms that regulate the structure of community through antagonistic interactions.  4.4.3 Model Describes Different Mechanisms that Regulate the Structure of Microbial Communities in Each Ecosystem Ultimately, a hypothetical model was proposed that describes how abiotic and biotic factors might regulate the structure of microbial communities in the BCRs studied in this thesis (Figure 4.6). The carbon source might be an abiotic factor that regulates the structure of microbial community by supporting the growth of keystone microorganisms with distinctive abilities to use the degradation products from a particular carbon source (Chapter 3, Section 3.4.2). Then these keystone 	  	   75	  microorganisms use different mechanisms to shape the structure of microbial communities. For example, pulp mill biosolids support the growth of keystones affiliated to Firmicutes, Actinobacteria and Alpha-Proteobacteria in BCR3 in the examined sites. Then, keystone microorganisms in this BCR regulate the structure of communities by antagonistic activities; whereas hay, manure and woodchips support the growth of different keystone microorganisms, which are able to make protagonistic interactions with other microorganisms and shape the structure of community in other BCRs.   Figure 4.6: Schematic model proposing biotic and abiotic mechanisms that regulate the structure of microbial communities in BCR3 and other BCRs.  	  	   76	   CHAPTER 5 MICROBIAL COMMUNITIES IN NITROGEN-REMOVING BIOLOGICAL REACTORS TREATING MIW  5.1 Synopsis  The microbial populations in nitrogen-removing systems have been studied and the phylogenetic diversity of key functional groups including AOB, AOA, NOB and denitrifiers have been reported [49, 50, 135, 136, 179, 188, 224, 259]. However, there are no data on the phylogenetic diversity of microorganisms existing in nitrogen-removing bioreactors where the influent is MIW. Nitrogen-containing wastewater generated from mining activities has distinctive characteristics that set it apart from wastewater that is generated from other sectors such as municipal, agricultural or food industries. Wastewater generated by the latter districts is frequently contaminated with significant levels of plant organic materials, solvents, pesticides, antibiotics, growth hormones, petrochemicals, coloring material, and detergents. However, MIW contains elevated concentrations of metals such as copper and arsenic in addition to sulfate and nitrogen-containing pollutants (e.g. ammonia, nitrate, nitrite, and cyanate). For example, in the system that was investigated in this study (Chapter 2, Figure 2.10), the concentrations of arsenic, copper and sulfate in the influent were 0.5-3.0, 0.3-2.0, and 2500-4000 mg/l, respectively, while the permitted limits for these compounds are 0.2, 0.2 and 1800 mg/l, respectively. The concentration of cyanide in the tailings pond water was 2-7 mg/l, cyanate 200-350 mg/l, thiocyanate 500-1400 mg/l, ammonia 50-100 mg/l, nitrate 10-50 mg/l and nitrite 1 mg/l. The process was designed to remove these chemical compounds to below the following levels:  0.2 mg/l, 5 mg/l, 5 mg/l, 10 mg/l, 10 mg/l and 1 mg/l, respectively. Knowledge about taxonomic groups that are adapted to nitrogen-removing bioreactors that treat MIW will help customize future bioreactors for improved performance.  	  	   77	  Moreover, several biological reactions such as aerobic denitrification, autotrophic denitrification, nitrifier denitrification, and anammox have been described recently in some microorganisms (Chapter 1, Figure 1.4). These findings raise the question of whether microbial communities in a conventional nitrogen-removing system, such as the Nickel Plate mine treatment system, have the metabolic potential for these novel processes. The influent water into the Nickel Plate mine treatment system comes from a storage pond and there are some notable differences in water quality during the winter and spring. Under these circumstances, an important question is whether microorganisms entering with the influent water as well as seasonal differences in influent water chemistry affect the microbial community composition of bioreactors or not. Therefore, the microbial community structure within MIW nitrogen-removing bioreactors was studied using SSU rRNA amplicon and whole DNA metagenomic sequencing. The samples were collected in the two seasons of winter and spring for both full-scale and pilot-scale treatment plants. Phylogenetic diversity of the microbial populations was compared with those described for nitrogen-removing bioreactors that treat wastewater from other sectors. Microorganisms that were more adapted to MIW treatment system were enriched and whole genome shotgun sequencing was applied for both enrichment cultures and samples from the bioreactors to obtain better insights into metabolic potential of adapted microorganisms. The overall research question in this Chapter was: Are the microbial communities of nitrogen-removing bioreactors treating MIW different from those of other nitrogen-removing bioreactors treating low-metal concentration, non-mine-related effluents?  The following hypotheses were tested to answer this research question: 1. The particular chemistry of MIW affects the phylogenetic diversity of microorganisms in nitrogen-removing bioreactors.   2. Microorganisms able to remove ammonia and nitrate by alternative pathways are present in MIW nitrogen-removing bioreactors.  3. Seasonal changes in the chemistry and microbiology of influent water affects the microbial community composition of the bioreactors.  	   	  	  	   78	  Settling Chamber!(Sludge Settles)!Effluent from nitrification circuit !enters to denitrification circuit!Influent!Sludge Return!Effluent!Aerobic!Settling Chamber!(Sludge Settles)!Anaerobic! Sludge Return!Nitrate           Nitrogen!Ammonia           Nitrate!Nitrification Circuit! Denitrification Circuit!!" #"#"$"%"$"5.2 Materials and Methods  5.2.1 Description of Nitrogen-removing Bioreactors and Sampling Sequential nitrification and denitrification bioreactors at Nickel Plate mine using active processes for removal of ammonia and nitrate (Chapter 1, Section 1.4) were selected for a comprehensive study of the phylogenetic diversity of their microorganisms. Figure 5.1 is a process flow diagram showing influent water entering the nitrification circuit (nitrification bioreactor and settling chamber). After oxidation of ammonia and nitrite, water flows to the denitrification bioreactor for reduction of nitrate to gaseous nitrogen. Effluents from the nitrification and denitrification bioreactors enter settling chambers for sludge settling and recycling.   Figure 5.1: Schematic diagram of bioreactors using active processes for removal of ammonia and nitrate. Two-liter samples were taken from the underflow and overflow of each bioreactor in two seasons of winter (December, 2011) and spring (May, 2012). The sample points are indicated in the flowchart (Figure 5.1). Samples were collected from the influent and effluent water. Samples were collected from the sludge and effluent of two settling chambers. All of the following described procedures were applied to a total of 28 samples removed from both the pilot- and full-scale systems (Table 5.1). Each sample was dispensed into three sterile 50 ml Eppendorf tubes and centrifuged at 14,000 g for 20 min. After decanting the supernatant, the pellets were subjected to DNA extraction.  	  	   79	  Table 5.1: List of samples that were taken from full- and pilot-scale nitrogen removing bioreactors at the Nickel Plate mine.  5.2.2 DNA Extraction, Amplification of 16S rRNA Genes and Pyrosequencing After homogenization, DNA was extracted using the Power Soil DNA Isolation Kit (MoBio Laboratories, Carlsbad, CA, USA, Cat No: 12888-100) from 0.5 gr of each sample. Two nanograms (ng) of isolated community DNA was subjected to polymerase chain reaction (PCR) to amplify the V6 to V8 variable region of bacterial and archaeal 16S rRNA genes using barcoded primers 926f (AAA CTY AAA KGA ATT GAC GG), 1392r (ACG GGC GGT GTG TRC). Primer 454T-RA: 25 nt A-adaptor (CCATCTCATCCCTGCGTGTCTCCGACTCAG), primer 454T-FB: 25 nt B-adaptor sequence (CCTATCCCCTGTGTGCCTTGGCAGTCTCAG). The PCR was performed using a !"#$%&'()"*&%!&"'+,-%",.(./$&-0+1&''2&'103$.3+,4 !"#$%& '()) *#+)(%#$ *#+)(%#$5 !"#$%& '()) ,"$&"+"-.$"/# 0%$$)"#12-3.45%&2.+$%&2$3%2#"$&"+"-.$"/#25"/&%.-$/&5 !"#$%& '()) ,"$&"+"-.$"/# 0%$$)"#12-3.45%&2.+$%&2$3%2#"$&"+"-.$"/#25"/&%.-$/&6 !"#$%& '()) 6%#"$&"+"-.$"/# 6%#"$&"+"-.$"/#25"/&%.-$/&7/8%&+)/96 !"#$%& '()) 6%#"$&"+"-.$"/# 6%#"$&"+"-.$"/#25"/&%.-$/&7(#:%&+)/96 !"#$%& '()) 6%#"$&"+"-.$"/# 6%#"$&"+"-.$"/#25"/&%.-$/&7/8%&+)/96 !"#$%& '()) 6%#"$&"+"-.$"/# 6%#"$&"+"-.$"/#25"/&%.-$/&7(#:%&+)/97 !"#$%& '()) 6%#"$&"+"-.$"/# 0%$$)"#12-3.45%&2.+$%&2$3%2:%#"$&"+"-.$"/#25"/&%.-$/&7 !"#$%& '()) 6%#"$&"+"-.$"/# 0%$$)"#12-3.45%&2.+$%&2$3%2:%#"$&"+"-.$"/#25"/&%.-$/&58 !"#$%& ;")/$ ,"$&"+"-.$"/# ,"$&"+"-.$"/#25"/&%.-$/&7/8%&+)/958 !"#$%& ;")/$ ,"$&"+"-.$"/# ,"$&"+"-.$"/#25"/&%.-$/&7(#:%&+)/968 !"#$%& ;")/$ 6%#"$&"+"-.$"/# 6%#"$&"+"-.$"/#25"/&%.-$/&7/8%&+)/968 !"#$%& ;")/$ 6%#"$&"+"-.$"/# 6%#"$&"+"-.$"/#25"/&%.-$/&7(#:%&+)/978 !"#$%& ;")/$ 6%#"$&"+"-.$"/# 0%$$)"#12-3.45%&2.+$%&2$3%2:%#"$&"+"-.$"/#25"/&%.-$/&78 !"#$%& ;")/$ 6%#"$&"+"-.$"/# 0%$$)"#12-3.45%&2.+$%&2$3%2:%#"$&"+"-.$"/#25"/&%.-$/&49 0<&"#1 '()) *#+)(%#$ *#+)(%#$59 0<&"#1 '()) ,"$&"+"-.$"/# 0%$$)"#12-3.45%&2.+$%&2$3%2#"$&"+"-.$"/#25"/&%.-$/&59 0<&"#1 '()) ,"$&"+"-.$"/# 0%$$)"#12-3.45%&2.+$%&2$3%2#"$&"+"-.$"/#25"/&%.-$/&69 0<&"#1 '()) 6%#"$&"+"-.$"/# 6%#"$&"+"-.$"/#25"/&%.-$/&7/8%&+)/969 0<&"#1 '()) 6%#"$&"+"-.$"/# 6%#"$&"+"-.$"/#25"/&%.-$/&7(#:%&+)/979 0<&"#1 '()) 6%#"$&"+"-.$"/# 0%$$)"#12-3.45%&2.+$%&2$3%2:%#"$&"+"-.$"/#25"/&%.-$/&79 0<&"#1 '()) 6%#"$&"+"-.$"/# 0%$$)"#12-3.45%&2.+$%&2$3%2:%#"$&"+"-.$"/#25"/&%.-$/&598 0<&"#1 ;")/$ ,"$&"+"-.$"/# ,"$&"+"-.$"/#25"/&%.-$/&7/8%&+)/9598 0<&"#1 ;")/$ ,"$&"+"-.$"/# ,"$&"+"-.$"/#25"/&%.-$/&7(#:%&+)/9698 0<&"#1 ;")/$ 6%#"$&"+"-.$"/# 6%#"$&"+"-.$"/#25"/&%.-$/&7/8%&+)/9698 0<&"#1 ;")/$ 6%#"$&"+"-.$"/# 6%#"$&"+"-.$"/#25"/&%.-$/&7(#:%&+)/9798 0<&"#1 ;")/$ 6%#"$&"+"-.$"/# 0%$$)"#12-3.45%&2.+$%&2$3%2:%#"$&"+"-.$"/#25"/&%.-$/&798 0<&"#1 ;")/$ 6%#"$&"+"-.$"/# 0%$$)"#12-3.45%&2.+$%&2$3%2:%#"$&"+"-.$"/#25"/&%.-$/&	  	   80	  thermocycler (iCycler® ,Biorad) under the following conditions: 95oC for 3 min; 25 cycles of 95oC for 30s, 55oC for 45s, 72oC for 90s; and 72oC for 10 min. DNA concentrations and purity were measured on a NanoDrop® ND-2000 UV-Vis Spectrophotometer (NanoDrop Technologies, Wilmington, DE) and by running 1 µL of the PCR product on a 1% agarose gel. Purified PCR products were sent to Genome Quebec and McGill University Innovation Centre (Montréal, Québec, Canada) for pyrosequencing using a Roche GS-FLX Titanium Series sequencer. 5.2.3 Analysis of Sequences from Pyrotag Sequencing of the 16S rRNA Gene Raw sequence data generated from pyrosequencing were processed using QIIME 1.8.0 [33]. Every raw pyrosequence had to pass a multistage quality inspection to remove low quality reads and minimize sequencing errors that can be introduced during the pyrosequencing process. Sequences were rejected if they: (i) did not have perfect match with the pyrosequence primers (ii) contained any nucleotide ambiguities (iii) had a typical length of 1 standard deviation from the mean length after removing primer sequences (iv) had average quality scores were below 25 (recommended for the Roche Genome Sequencer FLX System) and (v) had a homopolymer sequence longer than eight base pairs. The remaining high quality sequences were imported to the QIIME suite of Python scripts and associated dependencies were used to cluster the filtered reads into operational taxonomic units (OTUs) using 97%, 94% and 90% sequence similarity thresholds with the computer program usearch which corresponds approximately to taxonomic level of species, genus and family, respectively [33], [63], [127]. Representative sequences for each OTU were assigned taxonomy using BLASTn to the Silva version 111 representative set (http://www.arb-silva.de/documentation/background/release-111/) [194]. OTUs represented by one read only were removed. Phylogenetic trees with pyrotag OTUs and nearest neighbours picked by using BLASTn to the NCBI nucleotide database were constructed by trimming the NCBI 16S rRNA sequences to the same region as the pyrotag amplicons and aligning these using MUSCLE version 3.8.31 to the reverse complement of the pyrotag OTUs representative sequences followed by tree building with PHYML (nucleotide substitution model HKY, 100 bootstraps) [62], [94]. Phylogenetic trees were constructed using the Bosque computer program and visualized on the Interactive tree of life website or by using Figtree V1.3.1 (Appendix B) [139, 196]. To assess the community properties (e.g. diversity) and evaluate community-level differences in the samples, alpha and beta diversity analyses were performed using QIIME 1.8.0. Alpha diversity metrics were: Chao1, Observed 	  	   81	  Species, Shannon, Simpson and PD-Whole_tree [134, 241]. Beta-diversity metrics such as weighted and unweighted UniFrac were used to assess the differences between microbial communities based on their composition [156]. UniFrac measures the phylogenetic distance between sets of taxa in a phylogenetic tree as the fraction of the branch length of the tree that leads to descendants from either one environment or the other, but not both. This method can be used to determine whether communities are significantly different, and to measure the relative contributions of different factors, such as chemistry and geography, to similarities between samples [156]. The results showing similarities or dissimilarities between samples are visualized with analysis such as Principal Coordinates Analysis (PCoA). UPGMA (Unweighted Pair Group Method with Arithmetic Mean) method was used for classification of samples on the basis of their pairwise similarities in microbial communities [271]. The rarefied 94% sequence similarity OTU table was used to produce a heatmap in R. 5.2.4 Enrichment of Metal Oxidizing Denitrifying Microorganisms  Culture media were designed based on two studies of optimization of culture media for denitrifying bacteria and media for isolation of anaerobic metal oxidizing denitrifying bacteria[101] [201]. Denitrifying enrichment media based on a mineral salt medium under anaerobic conditions with pH adjusted to 7.2 was prepared in which As (III) and nitrate were used as the exclusive electron donor and  acceptor, respectively. A liter of medium made in distilled deionized water contained: 7.9 g Na2HPO4·7H2O, 1.5 g KH2PO4, 0.3 g NH4Cl, 0.1 g MgSO4·7H2O, 5 ml of trace elements solution and 10 ml of vitamins solution. Per liter, the trace element solution contained: 50 g EDTA, 22 g ZnSO4·7H2O, 5.54 g CaCl2, 5.06 g MnCl2·4H2O, 4.99 g FeSO4·7H2O, 1.1 g (NH4)6Mo7O24·4H2O, 1.57 g CuSO4·5H2O and 1.61 g CoCl2·7H2O. The vitamin solution contained: 0.002 g biotin, 0.002 g folic acid, 0.01 g pyridoxine hydrochloride, 0.005 g nicotinic acid, 0.005 g pantothenic acid, 0.0001 g B12, 0.005 g p-aminobenzoic acid, and 0.005 g thioctic acid, per liter. The medium was amended with 10 mM HCO3– (NaHCO3) as a carbon source, 5, 10, 20 mM As (III) (NaAsO2) and 3, 6 mM NO3– (KNO3). All incubations were carried out in the dark at room temperature. The headspace of the vials was replaced with argon (Appendix A, Figure A.10) [101] , [201].    	  	   82	  Under anaerobic conditions, enrichment cultures were inoculated by adding nitrification and denitrification sludge to mineral salts mediums (10% wt/vol). Aliquots of 100 ml inoculated medium were dispensed into 160 ml serum bottles. The bottles were sealed with rubber stoppers and aluminum crimp seals, and incubated at room temperature. Sterile controls, autoclaved three times on consecutive days, were also included. To ensure anaerobic conditions, all sampling and amendments were carried out using sterile plastic syringes flushed with argon. After 7 days of incubation, a 50% dilution of the active cultures were made into fresh medium and incubated as previously described. After the active enrichment cultures were serially transferred to the point where there was no more sludge, they were filtered for DNA isolation and preserved in 50% glycerol-medium solution (no As) at −80°C [201]. These cultures were initiated from both the aerobic and anaerobic circuit sludge. Activity of the cultures for denitrification was evaluated using the Griess test [225] . In the Griess test, nitrite is detected and analyzed by formation of a red pink color upon treatment of a nitrite-containing sample with reagent. Griess reagent contains 0.2% naphthylethylenediamine dihydrochloride, and 2% sulphanilamide in 5% phosphoric acid. Selection of samples for metagenomic analysis was based on the result of the Griess denitrification test (Appendix A, Figure A.10). Out of 48 enrichment cultures, 3 demonstrating the maximum denitrification activity were selected for whole genome shotgun sequencing. They were: 1) samples from the nitrification settling chamber inoculated into the medium containing 3 mM NO3- and 5 mM arsenic, 2) samples from the denitrification settling chamber inoculated into the medium containing 6 mM NO3- and 10 mM arsenic, and 3) samples from the denitrification settling chamber inoculated into the medium containing 3 mM NO3- and 10 mM arsenic. Three samples that were collected directly from the nitrification and denitrification bioreactors were subjected to whole genome shotgun sequencing as well.  5.2.5 Whole Genome DNA Extraction  Six samples, (one from the nitrification circuit, two from the denitrification circuit and 3 from their enrichment cultures) were subjected to DNA extraction. After centrifuge (5000 rpm, 20 min), pellet was washed with TE buffer and suspended in 500 µl TE buffer. 10 µl lysozyme (150 mg/mL) were added and incubated at 56oC for 1 hr. 75µl 10% SDS and 5 µl Proteinase K (20 mg/ml), and 5 µl RNase were added and incubated at 56oC for 1 hr. 400 µl of tris saturated phenol (pH:8) were added 	  	   83	  and were centrifuged at 10,000 rpm for 10 min. 200 µl of tris saturated phenol and 200 µl of Chloroform:isoamyl alcohol (24:1) were added to supernatant and were centrifuged at 10,000 rpm for 10 min. Four hundreds microliters of Chloroform:isoamyl alcohol (24:1) were added to supernatant and were centrifuged at 10,000 rpm for 10 min.  0.1 volume of 3M sodium acetate and 2 volume of absolute ethanol chilled to -20oC were added to supernatant and incubated at -20oC overnight. Samples were centrifuged at 15,000 rpm for 30 min at 0oC. The supernatant was decanted and pellet washed with 600 µl 70% ethanol followed by centrifugation at 10,000 rpm for 15 min. The supernatant was decanted and the final pellet dried at room temperature. Pellets of DNA were suspended in nano-pure water and stored at -20oC.  5.2.6 MiSeq Library Preparation, Quality Control and Shotgun Sequencing The Nextera XT DNA sample preparation kit (GA09115) was used to prepare high molecular weight whole genomic DNA libraries for sequencing. Droplet digital PCR method was used for quality control of sequencing libraries. Illumina MiSeq technology was used to produce paired-end 300 base reads at the UBC Sequencing Centre at Pharmaceutical Sciences (http://sequencing.ubc.ca/).  5.2.7 Quality Control Checking, Assembly, Genome Annotation and Phylogenic Assignment  After quality control checking by FASTQ quality filter and removal of reads shorter than 100 bp and quality score less than 25, Illumina reads were assembled using velvet and Metavelvet (http://metavelvet.dna.bio.keio.ac.jp/).  Taxonomic classification and gene prediction were done by using the Metapathways 2.0 pipeline. The pipeline uses databases KEGG, COG, METACYC, and SEED, to give functional context to sequences. Databases Greengene, Silva and NCBI  were used for taxonomic classification [272] [194] [273]. 5.3 Results  5.3.1 Performance of Bioreactors The total removal efficiencies for ammonia and nitrate were 99.6% and 95.4% in the system at the time of sampling (Table 5.2). Besides ammonia and nitrate, removal of cyanide, thiocyanate, dissolved/total copper, dissolved cobalt, dissolved/total iron and total arsenic occurred with the efficiencies of 79.2%, 100%, 58/25%, 16%, 21/75% and 99.8%, respectively. The percent of 	  	   84	  removal of these constituents indicated that this wastewater treatment plant was operating efficiently. Ammonia and nitrite concentrations in discharge water were less than 0.08 and 0.2 mg/l, meeting discharge requirements.  Table 5.2: Performance of nitrification and denitrification bioreactors for removal of contaminant of concern. NA stands for not applicable. ND stands for not determined.  5.3.2 Phylogenetic Diversity of Microorganisms in Nitrogen-removing Bioreactors A total of 172,954 sequenced reads were determined by pyrotag sequencing of 16S rRNA. Based on 97% sequence similarities, these reads were binned into 1393 OTUs (42 archaeal (12.3% of retrieved sequences) and 1351 bacterial classified OTUs (87.7% of retrieved sequences)). The alpha-diversity analysis showed the number of gene sequences was adequate for the analysis of microbial communities in the sites (Table 5.3).        !"#$%&'&()*+,-./01,,#'"&*+,-./*&2*303"(4&()*+,-./*&2*3056/7&()*+,-./08"22#/9):*;##<)4*+,-./0!#(&/*;#<<)4*+,-./08"22#/9):*;#<<)4*+,-./08"22#/9):*=4#'*+,-./0!#(&/*=4#'*+,-./08"22#/9):*142)'"$*+,-./0!#(&/*142)'"$*+,-./0='7/6)'(* !"#$%& '($#)# *$!) '%+% %$%%( %$%%(+ %$"(") %$%!& %$!"" %$'(!) %$!*)(3"(4"7"$&("#'*>"#4)&$(#4*?77/6)'(* %$%! %$%%# +($&'+ '*"* %$%%+( %$*'+ %$"&"+ %$%'' "'$#*( ,- %$'&*'!"#$%&&'(#)*"%&'+,- @AA BBCD 31 31 @ECF 31 GCF FECE 31 38 HIJ8)'"(4"7"$&("#'*>"#4)&$(#4*?77/6)'(* %$%! %$%%# %$%! '*'+ %$%%*+ %$%%+* %$"%!) %$%'" %$%)# ,. ,-!"#$%&&'(#)*"%&'+,- @AA BBCD @AA 31 KICF EFCK @HCE KFCF FI 38 38	  	   85	  Table 5.3: Diversity indices of samples used in this study. These were determined using the OTU tables based on 97% (species), 94% (genus) and 90% (family) sequences similarity.   Based on the rarefaction curves and PD-Whole tree, a higher level of diversity was observed in the denitrification bioreactors in comparison with nitrification bioreactors. Also, the pilot-scale had a higher level of diversity compared to full-scale (Figure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	  	   86	      Figure 5.2: Rarefaction curve of bacterial and archaeal 97% operational taxonomic units (OTUs) in samples from pilot and full-scale bioreactors and influent in winter and spring.  In the denitrification bioreactor, 93.6% of the pyrotag reads were derived from bacteria and 6.4% from archaea. Bacteroidetes was the predominant phylum of bacteria, comprising 35.3% of all detected OTUs (97%) (Figure 5.3). Proteobacteria, Chloroflexi, Candidate-division OP11, Candidate-division OD1, Firmicutes and Planctomycetes were the subdominant bacterial groups, each containing 31.8%, 6.1%, 4.9%, 3.7%, 2.5% and 1.9% of detected sequences, respectively. Methanomicrobia were the most abundant phylum of archaea, constituting 4.4% of all reads followed by Soil_Crenarchaeotic_Group (SCG) (Thaumarchaeota) (1.9%) and Halobacteria 	  	   87	  (1.7%). Fifty percent of detected sequences were affiliated to the five orders Sphingobacteriales, Hydrogenophilales, Soil_Crenarchaeotic_Group, Bacteroidales and Clostridiales (Figure 5.3).  To some extent, the structure of microbial community in nitrification bioreactors is different from that in denitrification bioreactors (Figure 5.3). In the nitrification bioreactor, an elevated level of archaeal communities was observed in that 78.5% reads were derived from bacteria and 21.5% from archaea. Proteobacteria were the predominant phylum of bacteria, constituting 35.1% of all reads. Bacteroidetes, Chloroflexi, Candidate-division OD1 and Firmicutes were the subdominant bacterial groups, each containing 16.3%, 8.4%, 3.9%, and 3.4% of detected microorganisms, respectively. Soil_Crenarchaeotic_Group (SCG) (Thaumarchaeota) was the most abundant phylum of archaea, constituting 17.6% of all detected microorganisms followed by Methanomicrobia (3.3%). Approximately 50% of detected sequences affiliated to four orders Hydrogenophilales (Betaproteobacteia), Soil_Crenarchaeotic_Group, Sphingobacteriales and Caldilineales. 	  	   88	   Figure 5.3: Order-level microbial community of influent water and full- and pilot-scale treatment plants during winter and spring.  To some extent, the winter and spring bioreactors carry the same predominant taxa. However, the relative abundance of predominant taxa changes by the season. For example, Hydrogenophilales, (Thiobacillus), SCG-related, Bacteroidales and Caldilineales were abundant taxa in nitrification bioreactors during spring and winter, but SCG-related and Bacteroidales-related sequences are more prevalent during winter and Hydrogenophilales-related and Caldilineales-related sequences are more prevalent during spring (Figure 5.3 and Figure 5.4).   These similarities in populations of dominant microorganisms between winter and spring bioreactors were higher when samples were from the same scale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	  	   89	  1 2 2 3 3 3 3 4 4 2B 2B 3B 3B 4B 4B 1’ 2’ 2’ 3’ 3’ 4’ 4’ 2’B2’B3’B3‘B4’B4’BChitinophagaceaeMethyloteneraMethanosarcinaceaeCandidatus_ParvarchaeumCandidat_division_OD1CaldilineaceaeChitinimonasCyanobacteriaFlavobacteriumAcinetobacterMariprofundusScenedesmusAlgoriphagusGOBB3-­C201PedobacterSediminibacteriumSphingomonasCandidat_divisionOP11XanthomonadaceaeComamonadaceaeComamonadaceaeArenicellaChitinophagaceaeCandidat_division_OD1AlphaproteobacteriaAlphaproteobacteriaSphingobacterialesCandidat_divisionOP11ThiobacillusGemmatimonadaceaeRhodobacteraceaePlanctomycetesAzohydromonasCandidat_division_OD1RhodocyclaceaeThaueraDesulfocapsaCandidat_division_OD1MethylophilaceaeMethylophilaceaeMethylophilaceaeCandidat_divisionOP11EubacteriaceaeSaprospiraceaeAnaerolineaceaeSaprospiraceaeVerrucomicrobiaCaldilineaceaeAlteromonadaceaeBacteroidesTuricibacterCitrobacterChristensenellaceaeRuminococcaceaeSpirochaetaMethanobacteriaceaeMethanocopusculaceaeAnaerolineaceaeSediminibacteriumSinobacteraceaeDHVEG-­6SediminibacteriumProteobacteriaHelicobacteraceaeThaumarchaeotaHydrogenophilaceaeAchromobacterAnaerolineaceaeAchromobacterCaldilineaceaeCaldilineaceaeNitrospiraThaumarchaeotaHydrogenophilaceaeHyphomicrobiumCaldilineaceaeCandidat_division_OD1GemmatimonadaceaeAchromobacterAchromobacterWinter SpringFull-Scale Full-ScalePilot-Scale Pilot-ScaleIn!uent Nitri"cation Denitri"cation Nitri"cationNitri"cationNitri"cationIn!uent Denitri"cationDenitri"cationDenitri"cation Figure 5.4: Logarithmic heatmap (log 2) showing the top 10 percent predominant genera (based on 94% OTU-table) in all samples.  ! " # $ % &!! " # $ % &!	  	   90	  5.3.3 Distribution and Diversity of Key Functional Groups  5.3.3.1 Distribution and Diversity of Nitrifier-related Sequences  Both archaea and bacteria are able to oxidize ammonia to nitrite (the first step of nitrification). Ammonia oxidizing bacteria (AOB) are affiliated to genera Nitrosomonas, Nitrosospira, Nitrosolobus within the Betaproteobacteia and Nitrosococcus within the Gammaproteobacteria. In this study, 0% (winter and spring influents) to 1.6 % (spring pilot scale denitrification bioreactor) of recovered sequences were affiliated to AOB, mainly Nitrosomonas (Figure 5.5).  In both seasons, large populations of ammonia oxidizing archaea (AOA)-related sequences (from 0% in the winter and spring influent to 43% in the winter and spring full-scale process) were detected in the nitrification bioreactors where only few AOB were retrieved.   	  	   91	   Figure 5.5: Distribution of key functional groups in each sample. It is based on OTU-table with 97% homology cut-off.  Nitrite oxidizing bacteria (NOB) that are responsible for the second step of nitrification are more diverse (phylogenetically) than ammonia oxidizing microorganisms. All known NOB belong to one of four different genera: Nitrospina within the Deltaproteobacteria, Nitrobacter within the Alphaproteobacteria, Nitrococcus within the Gammaproteobacteria and Nitrospira within a separate phylum Nitrospirae. In this study, 0% (winter and spring influents) to 3.2 % (winter, full-scale nitrification bioreactor) of recovered sequences were affiliated to NOB, mainly Nitrospira (Nitrospirales) (Figure 5.5). Few Nitrobacter-related sequences were detected in spring in the pilot scale samples. In contrast to AOB-related populations that were high in denitrification bioreactors, !"#$%&'()'*+)(",*-.'"/0-123345-6#7'"8"9":)8-;<4-6#7'"8"*"**/8-1=45-6#7'"80#')-<;4-6#7'">)*7('-<4-6#7'"*"**/8-?4-!"#$%#&'#()*+#,$-.$)"#$!-)*/$0#*1,$2#&$3*42/#$	  	   92	  NOB (alongside AOA) are more prevalent in nitrification bioreactor in both seasons and in both full scale and pilot scales. Since pyrotag sequencing revealed Nitrospira and Nitrosomonas as the major bacterial taxa that were involved in nitrification, Nitrospira- and Nitrosomonas- related genes were studied in the dataset from whole genome Illumina sequencing (Appendix A, Table A.1). The majority of Nitrospira-related sequences were affiliated to Candidatus N. defluvii. Several of the Nitrospira- and Nitrosomonas-related open reading frames (ORF) were related to membrane proteins that transport metals. Arsenite oxidase in addition to heavy metal efflux system-related proteins in the whole metagenomic dataset were closely related to those in the genome of Candidatus N. defluvii. Several Nitrosomonas-related ORFs were detected that are responsible for the resistance to copper, oxidation of copper and heavy metal transportation. 5.3.3.2 Distribution and Diversity of Denitrifier-related Sequences (Denitrification Reaction) Denitrifiers are represented in all main phylogenetic groups, with the vast majority belonging to the Betaproteobacteia, Gammaproteobacteria, and Alphaproteobacteria, Anaeromyxobacter, Flavobacterium, Geothrix, Propionibacterium, and Paenibacullus. Denitrification has also been described for several archaeal species as well as bacterial species [30]. A study by Spain et al. (2011) reported potential denitrifying taxa found in different environments, particularly contaminated wastewater, based on 24 previous studies that surveyed 16S rRNA genes and/or nirK and nirS genes [226]. Taxa presented in Table 5.4 comprise the collective known potential denitrifying community among all sites.       	  	   93	   Table 5.4: Potential denitrifying genera based on 24 studies that have surveyed 16S rRNA genes and/or nirK and nirS genes from different environments [226]. This table was adapted from Spain et al. (2011).  Based on our analysis of the V6-V8 16S rRNA pyrotags, the most frequently detected potential denitrifying taxa in the full- and pilot-scale denitrification bioreactors were related to Alpha-, Beta-, and Gamma-Proteobacteria (Table 5.5).  Table 5.5: Denitrifying-related taxa in full- and pilot- scale denitrification bioreactors by pyrotag sequencing of variable region V6-V8 of 16S rRNA gene.  The dataset that was prepared by whole genome shotgun sequencing was investigated for genes that are related to denitrification reactions (Appendix A, Figure A.11) and the taxa that carry them. Table 5.6 presents the assembled genes that are involved in denitrification reactions based on the SEED database and the list of taxa that carry them. The majority of predicted proteins were closely related to proteins in the database from Alpha-, Beta- and Gamma-Proteobacteria . Many of the predicted denitrification reaction-related proteins were related to Purple non-sulfur bacteria. ORFs that are related to denitrification reactions such as NorD and NorE (Appendix A, Figure A.11) were related to Rhodobacter, Hyphomicrobium, Methylotenera, Thiobacillus, and some other taxa were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	  	   94	  determined (Table 5.6).  Copper containing nitrite reductase and some other denitrification reaction genes belonged to Thialkalivibrio, Ralstonia and Nitrosomonas and some other taxa.  Table 5.6: Table showing ORFs that encode enzymes for denitrification reactions and their related microorganisms based on SEED database. 1; Alphaproteobacteria 2; Betaproteobacteia 3; Gammaproteobacteria 4; Purple non-sulfur bacteria 5; Chloroflexi 6; Bacteroidetes 7; Euryarchaeota 8; Aquificae 9; Nitrospirae 10; Spirochaeta 11; Actinobacteria.  Many predicted proteins related to enzymes involved in transportation and oxidation-reduction of heavy metals were found in the same taxa carrying genes for denitrification. Table A.2 (Appendix A) summarizes some of the potential denitrifier-related ORFs which are involved in metal metabolism. Large numbers of membrane protein-related genes that are responsible for translocation of cadmium, copper, cobalt, zinc, arsenic and lead were found. Recovered ORFs in the denitrification bioreactors were mostly related to Hyphomicrobium, Rhizobiales, Rhodobacter, Variovorax, Thiobacillus, Nitrosomonas, Ralstonia, Methylotenera and some other denitrifiers (Appendix A, Table A.2) Moreover, several genes that are related to enzymes responsible for arsenite or copper oxidation (e.g. -related to Rhodobacter), mercury reduction (e.g. -related to Burkholderia), metal chaperons (e.g. –related to Purple-non sulfur bacteria), heavy metal detoxification and methylation, and genes for several regulatory proteins were sequenced. Interestingly, in the dataset, the majority of retrieved genes that were involved in metal metabolism were also related to denitrifying communities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	  	   95	  5.3.4 Influent Water, its Seasonal Changes and its Impact on the System 5.3.4.1 Seasonal Changes of Physiochemical Properties of Influent Water In the winter, most of the surface water of the influent water storage pond is frozen and flow rates into the treatment system are very low, thus making the influent water to be composed of mostly ground water, which is more contaminated and has fewer organic components. During this season, the total dissolved solids, ammonia, nitrate, and sulfate in the influent were 2205, 18.8, 4.3 and 1060 mg/l, respectively (Table 5.7). Temperature of the influent water during the winter operation ranges from 5 to 8oC. In the spring, with the runoff, the composition of the influent water consists mostly of surface water that is less contaminated with metals. The spring influent water is rich in dissolved organic carbon. Algal blooms that also affect the available carbon source were observed during spring. Water temperature in spring operation varies from 15 to 18oC. In all seasons, the pH is constant between 7-8, and methanol addition to the denitrification bioreactor is also constant at 380 l/min.   Table 5.7: Average physiochemical properties of the influent water during spring and winter.  5.3.4.2 Seasonal Changes in Microbial Community Composition of the Influent Water  At the phylum level, in the influent water, 94.6% of sequences were derived from bacteria and 5.4% from archaea clustered into a total of 324 operational taxonomic units with 97% homology cut-off. Bacteroidetes was the predominant phylum of bacteria, constituting 36.5% of all detected reads. Proteobacteria, Firmicutes and Cyanobacteria were the subdominant bacterial groups, containing 23.6%, 15.1%, and 12% of detected reads, respectively. Methanomicrobia were the most abundant phylum of archaea, constituting 5.2% of all detected reads. More than 50% of detected sequences affiliated to three orders of Bacteroidales, Clostridiales and Chloroplast (Cyanobacteria).  !"#$%&'())"%*+,&)"%(,!+-./$#0/+&123456!"#$%&37$8(,+!("97$8$#+:--"8($&1-;<%&$)&=4=(#/$#+&1-;<%&$)&=4>0%?$#+&1-;<%4'())"%*+,&3"..+/&1-;<%4!"#$%&3"..+/&1-;<%4'())"%*+,&3"@$%#&1-;<%4'())"%*+,&A/"8&1-;<%4!"#$%&A/"8&1-;<%4'())"%*+,&:/)+8(9&1-;<%4!"#$%&:/)+8(9&1-;<%4B(8#+/&A8?%0+8#& !!"# #$$% &'& "'( !#& )* ('+ )"," "'") "'+$!'" "'#* "'"! "'!, "')% "'#$+'">./(8;&A8?%0+8# +!"( )#$)% &'( -. !,& !+ "'# *+( -. "'") -. /"'") "'#+ -. "',!	  	   96	  The main differences between winter and spring influent water were mainly due to the presence of large number of Bacteroidales (47%), Clostridiales (29%), Methanomicrobiales (7%), Spirochetales (6%) in the winter influent, (Figure 5.3), and large number of Chloroplast (24%), Flavobacteriales (17%), Burkholderiales (9%), Sphingobacteriales (8%), Sphingomonadales (8%), Rhodobacrterales (7%), Hydrogenophilales in the spring influent.  5.3.4.3 Impact of Influent Water on Microbial Community Composition of Bioreactors Based on the alpha diversity rarefaction curves and PD whole tree diversity metrics, a higher level of diversity was observed for the winter influent in comparison with the spring influent. In bioreactors, a higher level of diversity was also observed for the sample from winter in comparison to spring (Figure 5.2). Principal component analysis of unweighted UniFrac phylogenetic dissimilarities revealed variation in microbial community composition of influent water by changes in season and variation in microbial community composition of bioreactors by changes of influent water (Figure 5.6).  	  	   97	   Figure 5.6: Principal coordinate analysis (PCoA) plot from unweighted UniFrac distance samples from influent and bioreactors during spring and winter. Samples taken during the winter are depicted by the red squares, and samples taken in the spring by blue closed circles. A large proportion of functionally important phylogenetic groups (e.g. denitrifying genera of Thiobacillus) enter the bioreactors from spring and winter influent waters. The heatmap in Figure 5.7 demonstrates species-level distribution of OTUs in the winter bioreactors that are present in winter influent waters as well. Sequences related to microorganisms such as Methanosarcina, Methanobacterium, Soil_Crenarchaeotic_Group, Terrimonas, Pseudomonas, Desulfobulbus, Nitrospira, Smithella, were present in both winter bioreactors and influent water. Some of these taxa (e.g. Nitrospira, Terrimonas, Soil_Crenarchaeotic_Group, Methanosarcina) increased proportionally in the bioreactors however, other OTUs appeared in the bioreactors at a lower ratio in comparison to influent water. As can be seen in Figure 5.7, both nitrification and denitrification bioreactors carry microbes that came from the influent water. 	  	   98	   Figure 5.7: The logarithmic heatmap (Log 10) showing species-level (97% OTU-table) distribution of sequences in the winter influent water, nitrification and denitrification bioreactors. OTU numbers and their Silva 111 assigned taxonomy are given on the right hand side. Sample names are column labels. The heatmap in Figure 5.7 demonstrates species-level distribution of OTUs in the spring bioreactors that are present in spring influent waters as well. Sequences related to microorganisms such as Flavobacterium, Nitrospira, Rhodobacter, Alcaligenaceae, Acidovorax, Hydrogenophaga, Smithella, Terrimonas, Thiobacillus, and Pseudomonas were present in both spring bioreactors and 4 1 23B34B 2B72-­Methanobacterium46-­Methanocorpusculum433-­Methanosaeta100-­Methanomethylovorans5-­Methanosarcinaceae1787-­Soil_Crenarchaeotic_Group(SCG)1-­Soil_Crenarchaeotic_Group(SCG)147-­-­Bacteroides96-­-­Bacteroides15-­-­Bacteroides208-­-­Bacteroides706-­-­Bacteroides54-­-­Parabacteroides301-­-­Petrimonas122-­-­Proteiniphilum196-­-­Porphyromonadaceae614-­-­Paraprevotella0-­-­Terrimonas62-­-­Christensenellaceae1548-­-­Christensenellaceae83-­-­Clostridium1037-­-­Pseudobutyrivibrio452-­-­Pseudobutyrivibrio276-­-­Lachnospiraceae207-­-­Lachnospiraceae496-­-­Desulfosporosinus984-­-­Peptococcaceae32-­-­Anaerofilum317-­-­Hydrogenoanaerobacterium123-­-­Ruminococcaceae751-­-­Ruminococcaceae456-­-­Oscillibacter859-­-­Ruminococcus1178-­-­Ruminococcaceae563-­-­Ruminococcaceae493-­-­Ruminococcaceae374-­-­Ruminococcaceae594-­-­Clostridiales530-­-­Clostridiales400-­-­Veillonellaceae33-­-­Nitrospira511-­-­Caulobacterales1615-­-­Albidiferax191-­-­Simplicispira315-­-­Desulfobulbus678-­-­Smithella6-­-­Citrobacter521-­-­Enterobacter601-­-­Acinetobacter163-­-­Pseudomonas881-­-­Pseudomonas565-­-­RF3322-­-­RF3338-­-­Spirochaetes934-­-­Spirochaeta88-­-­Spirochaeta339-­-­Treponema429-­-­Treponema836-­-­Spirochaetaceae471-­-­Synergistaceae272-­-­Mollicutes391-­-­ThermotogaceaeIn!uentDenitri"cationNitri"cationDenitri"cationNitri"cation! " # $ % &! " # $ % &	  	   99	  influent water. Some of these taxa (e.g. Thiobacillus, Terrimonas and Smithella) increased proportionally in the bioreactors however; other OTUs appeared in the bioreactors at a lower ratio in comparison to influent water. As can be seen in Figure 5.8, both nitrification and denitrification bioreactors carry microbes that came from the influent water.  Figure 5.8: The logarithmic heatmap (Log 10) showing species-level (97% OTU-table) distribution of sequences in the spring influent water, nitrification and denitrification bioreactors. OTU numbers and their Silva 111 assigned taxonomy are given on the right hand side. Sample names are column labels. 1’ 2’ 3’ 4’4’B2’B3’B1006-­-­Candidatus_Aquiluna140-­-­Candidatus_Aquiluna1478-­-­Flexibacter620-­-­Flexibacter52-­-­Flavobacterium796-­-­Flavobacterium735-­-­Flavobacterium624-­-­Flavobacterium369-­-­Flavobacterium0-­-­Terrimonas693-­-­Chitinophagaceae170-­-­Chitinophagaceae92-­-­Sphingobacteriaceae109-­-­Candidate_division_OD134-­-­Candidate_division_OD1769-­-­Caldilineaceae603-­-­Chloroflexi443-­-­Chloroflexi874-­-­Chloroplast686-­-­Chloroplast212-­-­Chloroplast332-­-­Ruminococcaceae107-­-­Nitrospira319-­-­Planctomycetaceae260-­-­Planctomycetaceae305-­-­Rhodobacter75-­-­Rhodobacter578-­-­Caedibacter409-­-­Holospora1039-­-­Sphingomonadales64-­-­Novosphingobium81-­-­Alcaligenaceae2316-­-­Alcaligenaceae1681-­-­Acidovorax172-­-­Acidovorax21-­-­Hydrogenophaga728-­-­Hydrogenophaga1744-­-­Comamonadaceae2148-­-­Thiobacillus673-­-­Thiobacillus847-­-­Thiobacillus9-­-­Hydrogenophilaceae2045-­-­Methylophilaceae2040-­-­Methylophilaceae362-­-­Methylophilaceae1199-­-­Azospira117-­-­Rhodocyclaceae114-­-­Mariprofundus676-­-­Bacteriovorax294-­-­Desulfuromonadales14-­-­Smithella11-­-­Helicobacteraceae24-­-­Aeromonas4-­-­Arenicella163-­-­Pseudomonas881-­-­Pseudomonas181-­-­Pseudomonas423-­-­PseudomonasDenitri!cationIn"uentNitri!cation Nitri!cationDenitri!cation! " # $ % &! " # $ % &	  	   100	  In influent water, large numbers of Bacteroidetes, SCG-related sequences during winter were replaced by Hydrogenophilales-related and Caldilineales-related sequences during spring. In bioreactors this seasonal variation in relative abundance of predominant taxa was observed as well. SCG-related and Bacteroidales-related sequences were more prevalent during winter and Hydrogenophilales-related and Caldilineales-related sequences were more prevalent during spring.  5.4 Discussion  5.4.1 Nitrogen-removing Bioreactors Selects for some Microorganisms that are Adapted to this Particular Ecosystem One of the main questions in nitrogen-removing bioreactors where the source of influent is MIW is whether differences in chemistry of the mine wastewater with other influents results in differences in microbial community composition. Sequences belonging to several major taxa that were retrieved from the nitrification and denitrification bioreactors in this study were not restricted to this system and are present in diverse natural and engineered ecosystems. For example, the genus Sediminibacterium contains species present in activated sludge from wastewater treatment at a petroleum refinery, and bioreactors treating municipal wastewater [8, 136]. A Sinobacteraceae-related strain that was closely related to sequences that were detected in our study was first isolated from farmland soil near a chemical factory. The soil had been polluted by different kinds of insecticides and herbicides over a long period [264]. Tully et al. (2013) detected Sinobacteraceae-related sequences from deep-sea ferromanganese/polymetallic nodules [240]. Members of the genus Terrimonas (Chitinophagaceae) first were isolated from bulking sludge collected from a municipal wastewater treatment plant [108]. Later the same genus was found in bioreactors removing nitrate and phosphate from domestic wastewater [76]. Terrimonas lutea is a species that is also found in many different environments [108]. It is a slow-growing nitrifier [35]. Based on a 16S rRNA gene library analysis, Yoon et al. (2010) reported members of the subdivision Anaerolineae were dominant in the activated sludge of municipal wastewater treatment plants [259]. In another study, clones affiliated with the subdivision Anaerolineae (esp. class Anaerolineae) and Caldilineae were the most abundant in the Chloroflexi-specific 16S rRNA gene libraries of activated sludge in municipal treatment systems [173]. Chloroflexi are filamentous microorganisms that have been 	  	   101	  shown to be major members of various wastewater treatment processes [77, 204], [23]. They usually make the backbone of activated sludge [23]. Among all bacteria that are capable of ammonia oxidation (first step of nitrification reaction, Chapter 1, Section 1.6), our study (by pyrotag sequencing) suggests Nitrosomonas (mainly Nitrosomonas europaea) as the major ammonia oxidizing ones (AOB). Nitrosomonas spp. has been introduced as the main ammonia oxidizer in bioreactors treating municipal wastewater and other environments as well [56, 64, 97, 135, 184]. Wang et al. (2010) investigated ammonia oxidizing community in activated sludge collected from eight full-scale and pilot-scale wastewater treatment systems from different sectors using PCR followed by terminal restriction fragment length polymorphism (T-RFLP), cloning, and sequencing of the alpha subunit of the ammonia monooxygenase gene (amoA). Their study indicated that all the dominant AOB in their systems were closely related to Nitrosomonas sp., not to Nitrosospira sp., [246]. This result is consistent with most of the previous studies of nitrogen-removing systems [56, 64, 97, 120, 135, 141, 142, 184, 188]. Nitrosomonas sp. such as N. europaea, can have a maximum specific growth rate (μmax) as high as 0.088/hr in pure culture, whereas Nitrosospira sp., has a μmax ranging from 0.033 to 0.035/hr [224]. Higher growth rate may favor Nitrosomonas over Nitrosospira as the major species in mine and other wastewater treatment systems. However, in our study (by whole genome shotgun sequencing) the large numbers of recovered sequences were related to N. europaea that carry several metal metabolism ORFs that are related to resistance to copper, oxidation of copper and heavy metal transportation. Being equipped with metabolism of metals, N. europaea might be one of the key players in MIW treatment systems.  Among nitrite oxidizing communities (NOB) our study (by pyrotag sequencing) clearly showed dominance of Nitrospira and to lesser extend Nitrobacter in this system. Recent studies indicate that Nitrospira and Nitrobacter were the dominant NOB in most municipal wastewater treatment plants and activated sludge as well [27, 46, 215]. The high affinity toward oxygen that has been observed in the strains of Nitrospira can be the reason for this dominance [224]. Juretschko et al. (1998) investigated industrial wastewater treatment plant samples with high ammonia concentration; Burrell et al. (1999) used a reactor spike with nitrite, and Schramm et al. (1999) studied NOB in biofilms and fludized bed reactors. In their work, they found Nitrospira and Nitrobacter as major NOB in all municipal wastewater treatment plants samples [29, 113, 213, 215]. Coskuner et al. (2002) also 	  	   102	  detected Nitrospira and Nitrobacter in samples from activated-sludge system receiving domestic wastewater. They suggested that Nitrospira may be more diverse than what the databases currently describe, and low temperature may enhance the diversity of this NOB genus [29, 47, 61, 113, 182, 213–215, 224]. Siripong et al. (2007) examined the diversity of nitrite oxidizing bacterial communities in seven water-reclamation plants. The plants vary in types of influent waste stream, plant size, water temperature, and retention time. They observed coexisting Nitrobacter and Nitrospira genera and reported among the factors that varied among the plants; only the seasonal temperature variation seemed to change the nitrifying community [224]. The variation in NOB-related communities by season was observed in our study as well. In spring samples, Nitrobacter-related sequences were detected alongside Nitrospira. However no Nitrobacter-related sequences were found in either the full- or pilot-scale bioreactors during the winter. Moreover, winter pilot- and full-scale plants carry different operational taxonomic units of Nitrospira from the Nitrospira-related OTUs in the spring samples. However, in our study (by whole genome shotgun sequencing) the large numbers of recovered sequences were related to nitrite-oxidizing Candidatus Nitrospira defluvii that carry several metal metabolism ORFs. On the basis of metagenomic analysis of Ca. N. defluvii, researchers recently reported that Ca. N. defluvii differs extensively from other nitrite oxidizers in the composition of the respiratory chain, in the key enzyme nitrite oxidoreductase (NXR), and in the reactions used for autotrophic carbon fixation [157]. Their findings shows that enzyme NXR constantly expressed in Ca. N. defluvii and unusual pathways for the transport, oxidation, and assimilation of organic materials facilitate adaptation of Ca. N. defluvii to substrate-limited conditions Assuming that there is link between prevalence and activity, the role of archaea in ammonia oxidation is more significant than that of bacteria in the studied MIW nitrogen-removing system since the relative abundance of AOA 16S rRNA genes is several orders of magnitude higher than that of AOB. The contribution of AOA and AOB to the nitrification in various environments is still controversial. Mussmann et al. (2011) screened 52 municipal and industrial wastewater treatment plants for the presence of AOA and AOB. According to their results, the presence of AOB was more significant than AOA, and AOA were detected in high abundance in just four industrial plants [179].  However, quantitative analysis of amoA gene copies (ammonia monooxygenase; its amplification is a method to estimate the distribution and relative abundance of ammonium-oxidizing 	  	   103	  microorganisms) in several other studies has indicated that AOA can predominate over AOB in soil environment [40, 138]. Erguder et al. (2009) proposed that AOA might be major ammonia oxidizers in sulfide-containing environments and under low nutrient conditions [27, 70]. It has been reported that some capabilities of AOA, such as having high substrate affinity; being capable of doing ammonia oxidation over a wide range of ammonia concentrations, temperatures and pH; and their differential adaptation for ammonia oxidation inhibitors, gives AOA the advantage to outcompete AOB [112]. Since AOA appear to grow preferably under sub-oxic conditions, AOA may also be responsible for both nitrification and denitrification “nitrifier denitrification” in oxygen minimum zones or under alternating aerobic/anaerobic conditions [185, 260]. By production of variety of protective molecules and enzymes, many archaeans thrive in conditions that would kill other organisms. They have been found in extreme conditions such as boiling or frozen water, super salty ponds, sulfur-spewing volcanic vents, and highly acidic environments [207, 274]. On the other hand, at low temperature, ammonia-oxidizing bacteria tend to go dormant [251]. This may justifies the dominance of ammonia-oxidizing archaea over bacteria particularly during winter in studied nitrogen removing bioreactors. The denitrifying populations (mostly belonging to Alpha-, Beta- and Gamma-Proteobacteria) in the studied nitrogen-removing system were slightly distinctive from those in other nitrogen-removing bioreactors treating low metal concentration, non-mine-related effluent, and were highly similar to denitrifying communities detected in other heavy metal contaminated sites [71, 226]. Spain et al. (2011) reviewed 24 studies examining the microbial communities from bioremediation sites with several different geochemical conditions, such as different levels of nitrate contamination, pH and stimulated with ethanol as well as denitrifying fluidized bed reactors treating high-nitrate groundwater, in an effort to describe the overall potential denitrifying community composition at these sites. The denitrifying community determined in our study is similar to communities detected in denitrifying reactors contaminated with high levels of radionuclotides, heavy metals and nitrate  [226]. Their study revealed that under high concentrations of nitrate and heavy metals Betaproteobacteia genera, (Burkholderia, Thiobacillus and Acidovorax) were both dominant in 16S rRNA gene surveys, and are active in situ [226]. These same Betaproteobacteria genera were the most predominant denitrifiers in our bioreactors. Some of the other major denitrifiers in our study, Burkholderia, Hydrogenophaga, and genera in families Alcaligenaceae and Rhodanobacter also 	  	   104	  were the most metabolically active nitrate-reducing genera in uranium-contaminated sites according to the study of 16S rRNA molecules that amplified and cloned from reverse-transcribed total RNA extracts as well [3, 174]. Our finding suggest that this mine wastewater treatment system selects for a particular denitrifying community, therefore members of this community should possess important metabolic potential that makes them suitable for a mine environment.  In the dataset prepared by whole genome shotgun sequencing, many ORFs that were annotated as enzymes involved in metabolism of metals were closely related to denitrifiers such as the purple non-sulfur bacteria. Purple non-sulfur bacteria are found among the Alpha- and Beta- subgroups of Proteobacteria. According to the literature, members of non-sulfur bacteria such as Rhodobacter are particularly adapted for the removal of heavy metals and sulfate or nitrate (three abundant components in MIW) simultaneously by several mechanisms (e.g. direct enzymatic reduction, adsorption, sulfide precipitation) that work under aerobic, anaerobic, carbon-limiting condition and in dark-light transition zones [9, 67, 125, 243]. Species of this group are able to switch their energy metabolism to adapt to high concentrations of heavy metals [9, 67, 125, 243]. Some ORFs assembled from the mine bioreactor whole DNA sequencing reads (in this study) were closely related to Variovorax- and Rhodobacter- related arsenite oxidases. This enzyme is responsible for direct oxidation of the more toxic form of arsenic, namely, arsenite to arsenate. The capability of species of Variovorax to oxidize arsenite has been reported before [210]. Diverse metabolic capabilities of Variovorax species for removal of toxins make them promising species for use in bioremediation of metals [238].  Also, some Burkholderia-related ORFs were annotated as mercury reductase (in this study). Burkholderia species are denitrifying microorganisms in other nitrate, heavy metal and uranium contaminated sites, such as Oak Ridge (USA) [226]. Wide distribution of arsenic-related genes in Burkholderiales has been reported [140]. Methylotenera-related ORFs were retrieved extensively and, in addition to genes involved in denitrification, several genes for transportation of heavy metals that were also Methylotenera-related were found in our bioreactors. It has been reported that species within this genus link denitrification to the metabolism of methanol [180]. Some species of Methylotenera are capable of aerobic denitrification [180]. Aquificae-related nitrous oxide reductases, which use N2O to generate N2, were found in the mine bioreactor (in this study). 	  	   105	  Chemotrophic denitrifier Aquificae with nitrous oxide reductase, also carry mercury resistance and mercuric reductase activities [81].  5.4.2 Microorganisms that are Capable of Ammonia and Nitrate Removal by Alternative Pathways are Present in Conventional MIW Nitrogen-removing Bioreactors  Some sequences in the mine bioreactor metagenome were related to predicted proteins from microorganisms known to perform nitrification/denitrification through alternative pathways (Chapter 1, Figure 1.3 b). Many ORFs were taxonomically classified as being from Alcaligenes, Methylotenera, Paracoccus, and Pseudomonas genera, which contain species capable of aerobic denitrification. This might be one reason for the extensive presence of Methylotenera in both the nitrification (aerobic) and denitrification (anaerobic) bioreactors. The presence of ORFs related to Thiobacillus, Paracoccus, Pseudomonas, Thiosphaera, Nitrosomonas, Thialkalivibrio, and Ralstonia genera suggests that the metabolic potential for autotrophic denitrification exists in these bioreactors [2]. Several ORFs were affiliated to Planctomycetes, some of which are known to perform Anammox. Sequences related to Nitrosomonas were found extensively, especially in the dataset prepared by shotgun sequencing suggestive of a nitrifier denitrification reaction [2]. Removal of ammonia and nitrate by novel processes based on these alternative pathways for nitrification and denitrification would offer several advantages. Operating costs will decrease if either autotrophic denitrification or anammox were used because there would be no need for external carbon source addition. These processes would be particularly valuable for treatment of MIW that usually contains a low amount of organic components [2]. If nitrate is used as an electron acceptor for anammox, then oxygen would not be needed. Less energy would be required since no sparging or mixing would be required.  Less sludge is produced in autotrophic processes [57, 222]. It has been shown that application of the anammox process in the treatment of ammonia- and nitrate-rich water diminishes the cost of operation up to 60%, and lowers CO2 emissions[150]. Using a single microbial species, as in aerobic denitrification, might improve the efficiency of wastewater treatment. Wastewater treatment facilities often fail to establish stable nitrification due to unpredictable microbial community dynamics when a mixed culture of organisms is used. Cultivation of autotrophic nitrifiers is extremely time-consuming and difficult because nitrifying bacteria are autotrophic and grow slowly [255]. The aerobic denitrification process (based on 	  	   106	  application of single organisms) that combines aerobic denitrification and heterotrophic nitrification (use oxygen and nitrate simultaneously) may help to overcome this limitation. Heterotrophic nitrifiers grow fast and they occur directly in aerated bioreactors [47]. 5.4.3 Seasonal Changes in the Chemistry and Microbiology of Influent Water Affects the Microbial Community Composition of the Bioreactors  The chemistry (Table 5.7), the level of diversity (Figure 5.2) and microbial composition (Figure 5.6) of influent water changed by the season. Influent waters had impact on the microbiology of bioreactors by bringing several functionally important microorganisms such as denitrifying genera of Thiobacillus, Acidovorax, Burkholderia, Hydrogenophaga, Comamonas, Rhodobacter, Pseudomonas and Soil_Crenarchaeotic_Group into the bioreactors. Based on the principal coordinate plot from UniFrac analysis (Figure 5.6), seasonal changes in the microbial community structure of influent water impacted the microbial structure of bioreactors. Other reports describe the strong influence of influent microorganisms in the diversity of microorganisms in treatment systems [42, 258]. In our study, groups that were more prevalent in the spring influent were more prevalent in bioreactors during spring (but not winter) and during this season, algal blooms occurred that might have caused a reduction in microbial diversity in comparison to the winter. Lee et al. (2013) studied the effect of algal growth on wastewater bacterial communities and reported that algal growth significantly shifted the bacterial community structure and reduced bacterial diversity[136].    	  	   107	  CHAPTER 6 CONCLUSIONS  6.1 Overall Conclusions  This research investigated the microbial communities in two types of MIW treatment systems: Semi-passive biochemical reactors for removal of metals and sulfate, and active biological reactors for removal of nitrogen-containing components. Diagrams showing major biochemical processes that take place in each system were prepared (Chapter 1, Section 1.5 and Section 1.6) and some of microorganisms that are capable of doing those reactions were introduced. Also the few available studies that describe microbial communities of biological reactors treating MIW were reviewed. 6.1.1 Microbial Communities of Metal-removing BCRs Treating MIW  Four metal-removing BCRs of a similar configuration and geochemical environment with some differences that were successfully treating MIW at different mine-sites in British Columbia were studied. The results from surveying and comparing microbial populations in these BCRs were presented in Chapters 3 and 4. In summary, the bioreactors contained microbial consortia much more diverse than those previously described for laboratory-based BCRs. Considering principles for how biodiversity affects ecosystem functioning (Chapter 3, Section 3.4), this high level of diversity suggests a much wider metabolic potential in field-based BCRs when the source of carbon is from complex organic components. Metal-removing BCRs share several microorganisms even when they are at different geographical locations. The sulfate-reducer community common to this type of bioreactor was phylogenetically diverse but comprised distinct taxa (within the class Deltaproteobacteria) mostly in the family Desulfobacteraceae associated with other metal-rich or saline environments, indicating that they might be specialists at surviving under these conditions. Taking into account the link between structure and function (Chapter 1, Section 1.9), this similarity in microbial community structure and the presence of core sulfate-reducers in bioreactors suggest similarities in major biological processes that take place inside BCRs.  	  	   108	  SRM and methanogens appear not to be mutually exclusive and are able to co-exist in the studied BCRs. However, sulfate-reducing microorganisms are more prevalent than methanogens as potential competitors for carbon sources in field-based BCRs. In the bioreactors that contain both, the majority of SRM and methanogens were acetotrophic. Therefore, a theoretical model was proposed to justify the presence of methanogens in BCRs that has always been questioned (Chapter 1, Section 1.8). It also explains how the existence of acetate-consuming sulfate-reducing microorganisms and methanogens are beneficial for BCRs (Chapter 3, Section 3.4.4).  6.1.2 Factors Regulating the Structure of Microbial Communities in BCRs The structure of a microbial community of bioreactors with different carbon sources (pulp mill biosolids versus woodchips, hay and manure) was very different. Emerging evidence suggest that the type of carbon source in an ecosystem is an important abiotic factor that determines the structure of its microbial community. Network analysis tools were used to develop hypotheses about biotic factors regulating the structure of microbial communities in the BCRs. Network analysis suggested the keystone microorganisms in BCR3 were phylogenetically different from those in the other BCRs. In each ecosystem, the type of correlations (positive and negative) that keystone microorganisms make with other microorganisms was calculated by using correlation tests. Correlation analysis revealed that the putative keystone microorganisms in BCR3 might employ different mechanisms (antagonistic) than in the other BCRs to regulate the structure of community. A hypothetical model was developed that describes how abiotic and biotic factors regulate the structure of microbial communities in BCRs (Chapter 4, Figure 4.6).  6.1.3 Microbial Communities of Nitrogen-removing Bioreactors Treating MIW The microbial community structure within MIW nitrogen-removing bioreactors was studied using SSU rRNA amplicon and whole DNA metagenomic sequencing. The experiment was done in two seasons, winter and spring, for both full-scale and pilot-scale treatment plants. The phylogenetic diversity of microorganisms was reported in Chapter 5 and was compared with other nitrogen-removing bioreactors that treat wastewater from other sectors. Microorganisms that were more adapted to MIW treatment systems were enriched and whole genome shotgun sequencing was 	  	   109	  applied for both enrichment cultures and samples from bioreactors. This study concludes that a mine nitrogen-removing system constitutes several microorganisms that are present in other nitrogen-removing facilities; however, several key microorganisms in this system exist that offer important processes that are required for this MIW treatment system. Among key functional groups (AOA, AOB, NOB and denitrifiers) this system selects for archaeal ammonia oxidizer rather than bacterial. Also, the denitrifying population in this system was distinct from that in other systems treating non-MIW. The capacity to metabolize metals allowed the particular nitrifying and denitrifying taxonomic groups to exist in the MIW treatment system. Presence of OTUs related to taxa with the metabolic potential for aerobic denitrification, autotrophic denitrification, nitrifier denitrification, and anammox, suggests that there is the potential to use the microbial communities in this conventional system to develop new biotechnologies for more efficient treatment. Based on UniFrac-based principal coordinates analysis, the chemistry and microbiology of the influent water impacted the microbial community structure within the nitrogen-removing bioreactors. 6.1.4 Summary of Metabolically Important Microorganisms  One of the most important operational problems for field-based BCRs is low temperature during the cold seasons. This study showed the presence of sequences related to microorganisms such as Variovorax in BCRs. This is a significant finding since species of Variovorax paradoxus have the potential for removal of a variety of toxins and degradation of many organic compounds at temperatures of 10°C or below. In field-based BCRs the carbon sources are usually cellulose, chitin, and lignin derived from plants. They are not easily accessible; therefore, having microorganisms with versatile metabolic potential that are able to couple oxidation of aromatic and complex-structured components to sulfate or metal reduction is favorable. Microorganisms such as species in the family Desulfobacteraceae possess remarkable and diverse metabolic abilities and are promising choices for applications since they can degrade complex organic carbon compounds. In BCRs carbon source availability relies on hydrocarbon-degrading microorganisms, which can be a rate-limiting step in metal removal. Metal-removing microorganisms such as Acidobacteria or Rhodobacter adapt to carbon-limiting conditions. Species within the genera Desulfosporosinus, Desulfobacula, and Rhodobacter, can tolerate high concentrations of toxins and cope with the stress of hostile conditions of BCRs. Being able to cope with oxygen stress (for example in Rhodobacter sp., and Variovorax sp.,) is another significant quality since field-based bioreactors are not entirely 	  	   110	  oxygen sealed. Sulfate-reducing bacteria precipitate metals by hydrogen sulfide production, however, they generate sulfide only under strict anaerobic conditions. The presence of microorganisms such as Treponema denticola, Rhodobacter sp. and Pseudomonas that also possess the capacity for hydrogen sulfide production and precipitation of metals under aerobic conditions is an advantage for the system. Rhodobacter sp. is able to produce hydrogen sulfide and precipitate metals anaerobically (photoautotrophic and photoheterotrophic metabolisms) and aerobically (chemoheterotrophic) and in dark-light transition zones. The efficiency of the system can be improved by the presence of microorganisms such as Achromobacter that can co-metabolize several toxins and remove them simultaneously. Therefore, finding sequences related to the aforementioned microorganisms in this study would emphasize the significance of their presence in BCRs. Microorganisms with such capabilities are not necessarily predominant ones. Therefore the significance of most of them were not well understood by using conventional methods that usually capture only the dominant components of the communities.  6.2 Originality and Contributions to the Field The performance of bioreactors for treatment of MIW relies on microbial consortia with the metabolic potential to remove the contaminants of concern, but very little is known about the microbial community composition of these particular bioreactors. The complexity of this particular ecosystem has forced scientists and engineers to largely ignore the underlying details of what microorganisms are present and what their metabolic potentials are and use a “black box” approach to design and operation. This results in variable performance and a lack of understanding when the system is not working as expected. Removal of a particular set of constituents of concern requires the right microbial consortium to drive the desirable biochemical reactions; therefore we need to know what these microbial consortia are and how biotic and abiotic factors influence their structure and function. This knowledge is lacking in the current literature, and this thesis strives to fill in this gap by providing new knowledge on microbial community composition of MIWs at different sites. Previous studies of microbial populations in semi-passive BCRs were based on laboratory bioreactors and not field-based systems. In laboratory bioreactors, the experimental setup, including inoculum used, type of carbon source, type of bioreactor, and apparatus configuration, can influence the microbial community, which may be very different from those found in field bioreactors. Thus 	  	   111	  laboratory-based bioreactor studies may not yield information useful for actual scale-up and operation of field-based systems. At the time of writing, there were no comprehensive microbial community studies of field-based BCRs treating MIW. It was not known if BCRs at different mine sites contained the same phylogenetic groups or if they were very site specific. There is little information available on how carbon source, inoculum, and age impact BCR microbial communities. Since SRM are key for successful metal removal, a comprehensive phylogenetic study of the types of sulfate-reducing microorganisms present in BCRs was needed. Previous to our study, Desulfovibrio species were the most often identified SRM in BCRs treating metal-rich water in defined and complex carbon sources [103, 187]. This work provides new information on the SRM core to BCRs and revealed that these taxa are associated with metal-rich environments. Their phylogenetic diversity suggests that there are more metal-tolerant SRM than previously thought. The metabolic potential of these core SRM can be exploited to treat many different types of MIW. Also, there are several studies on the microbial populations of nitrogen-removing bioreactors, but there was no information about the phylogenetic diversity of microorganisms in nitrogen-removing bioreactors when the source of wastewater is MIW. It was still unclear if the particular chemistry of MIW affects the phylogenetic diversity of microorganisms in nitrogen-removing bioreactors.   Previous studies also used conventional techniques to study the microbiology of biological reactors; these conventional techniques come with limitations. For example, the microbial groups important in metal-removing BCRs are often low in relative abundance, since organic matter degraders are the most dominant members of the community. Therefore, deep SSU rRNA amplicon sequencing was essential to reveal taxonomic groups involved in metal remediation. Additionally, many of the microorganisms living in bioreactors at mine sites are unclassified and uncharacterized, and we do not know what their metabolic potential is. Metagenomic sequencing of whole DNA was needed to reveal functional capabilities of the bioreactors’ microbial community that cannot be inferred from taxonomy. Network analysis tools have been used in several ecosystems to interpret large datasets that generated by high throughput sequencing. Researchers have used microbial relative abundance data to identify putative interactions between microorganisms in communities [12, 19, 160, 231, 257, 262]. However, this data have never been used to develop hypotheses about microbial correlation 	  	   112	  and factors regulating the structure of microbial community in BCRs. It is postulated that signatures of microbial interactions become imprinted in microbial survey datasets [73]. 6.3 Limitations of the Research and Recommendations for Future Work The biggest limitation to this study was the difficulty in assigning putative functions to the OTUs that were identified. More work is required to characterize microorganisms that are adapted to field-based bioreactors since many of the OTUs found in these systems could not be assigned to known cultured species. Supplementation of SSU rRNA amplicon sequencing with whole DNA metagenomics studies will provide information about the metabolic potential of the unknown groups, even those that are uncultivable. To be able to precisely compare microbial compositions between different ecosystems it is recommended to measure each community with the same tools. In this study, pyrotag sequencing of V6-V8 SSU rRNA amplicons was used to compare samples from several biological reactors. But to compare the results of this study with other reports we encountered some limitations. This was mostly due to differences in library size (e.g. sequencing depth). Therefore, it is recommended to build datasets with a larger number of samples from different treatment systems by using the same methods. Having this larger dataset is more appropriate for correlation tests and network analysis studies as well. A much larger dataset for mine-based BCRs will allow us to obtain more accurate correlations between environmental factors and microbial community composition.  The Pearson correlation coefficient test was selected to calculate co-occurrence associations. Research showed that the choice of the measure impacts on the resulting network. It would be valuable to apply different methods on the same series of data and compare the consequential networks. Many of the putative keystone OTUs revealed through the correlation network analysis were unclassified and therefore their metabolic potential is unknown. Since they may play critical roles in maintaining microbial consortia structure, there is an urgent unmet need to reveal more information about their function.  Network analysis tools and correlation tests are suitable to generate hypotheses. Therefore, the 	  	   113	  models that are proposed in this study are purely theoretical and need to be validated with other types of analysis and experimentation.                   	  	   114	  REFERENCES  1.  Abicht HK, Mancini S, Karnachuk O V, Solioz M (2011) Genome sequence of Desulfosporosinus sp. OT, an acidophilic sulfate-reducing bacterium from copper mining waste in Norilsk, Northern Siberia. J Bacteriol 193:6104–5. doi: 10.1128/JB.06018-11 2.  Ahn Y-H (2006) Sustainable nitrogen elimination biotechnologies: A review. Process Biochem 41:1709–1721. doi: 10.1016/j.procbio.2006.03.033 3.  Akob DM, Mills HJ, Kostka JE (2007) Metabolically active microbial communities in uranium-contaminated subsurface sediments. FEMS Microbiol Ecol 59:95–107. doi: 10.1111/j.1574-6941.2006.00203.x 4.  Allison N, Turner JE, Wait R (1995) Degradation of homovanillate by a strain of Variovorax paradoxus via ring hydroxylation. FEMS Microbiol Lett 134:213–219. doi: 10.1111/j.1574-6968.1995.tb07940.x 5.  Alvarez MT, Crespo C, Mattiasson B (2007) Precipitation of Zn(II), Cu(II) and Pb(II) at bench-scale using biogenic hydrogen sulfide from the utilization of volatile fatty acids. Chemosphere 66:1677–83. doi: 10.1016/j.chemosphere.2006.07.065 6.  Andrew DR, Fitak RR, Munguia-Vega A, Racolta A, Martinson VG, Dontsova K (2012) Abiotic factors shape microbial diversity in Sonoran Desert soils. Appl Environ Microbiol 78:7527–37. doi: 10.1128/AEM.01459-12 7.  Athanasiadis K, Helmreich B (2005) Influence of chemical conditioning on the ion exchange capacity and on kinetic of zinc uptake by clinoptilolite. Water Res 39:1527–32. doi: 10.1016/j.watres.2005.01.024 8.  Ayarza JM, Figuerola ELM, Erijman L (2014) Draft Genome Sequences of Type Strain Sediminibacterium salmoneum NJ-44 and Sediminibacterium sp. Strain C3, a Novel Strain Isolated from Activated Sludge. Genome Announc. doi: 10.1128/genomeA.01073-13 9.  Bai H-J, Zhang Z-M, Yang G-E, Li B-Z (2008) Bioremediation of cadmium by growing Rhodobacter sphaeroides: kinetic characteristic and mechanism studies. Bioresour Technol 99:7716–22. doi: 10.1016/j.biortech.2008.01.071 10.  Baldwin SA, Khoshnoodi M, Rezadehbashi M, Taupp M, Hallam S, Mattes A, Sanei H (2015) The microbial community of a passive biochemical reactor treating arsenic, zinc and sulfate-rich seepage. Front Bioeng Biotechnol. doi: 10.3389/fbioe.2015.00027 11.  Balk M, Altinbaş M, Rijpstra WIC, Sinninghe Damsté JS, Stams AJM (2008) Desulfatirhabdium butyrativorans gen. nov., sp. nov., a butyrate-oxidizing, sulfate-reducing 	  	   115	  bacterium isolated from an anaerobic bioreactor. Int J Syst Evol Microbiol 58:110–5. doi: 10.1099/ijs.0.65396-0 12.  Barberán A, Bates ST, Casamayor EO, Fierer N (2012) Using network analysis to explore co-occurrence patterns in soil microbial communities. ISME J 6:343–51. doi: 10.1038/ismej.2011.119 13.  Barth M Betweenness Centrality in Large Complex Networks. 1–6. 14.  Barton LL, Fauque GD (2009) Biochemistry, physiology and biotechnology of sulfate-reducing bacteria., 1st ed. Adv Appl Microbiol. doi: 10.1016/S0065-2164(09)01202-7 15.  Barton LL, Fauque GD (2009) Biochemistry, Physiology and Biotechnology of Sulfate‐ Reducing Bacteria. Adv Appl Microbiol 68:41–98. doi: http://dx.doi.org/10.1016/S0065-2164(09)01202-7 16.  Battaglia-Brunet F, Dictor M-C, Garrido F, Crouzet C, Morin D, Dekeyser K, Clarens M, Baranger P (2002) An arsenic(III)-oxidizing bacterial population: selection, characterization, and performance in reactors. J Appl Microbiol 93:656–667. doi: 10.1046/j.1365-2672.2002.01726.x 17.  Bell T, Newman JA, Silverman BW, Turner SL, Lilley AK (2005) The contribution of species richness and composition to bacterial services. Nature 436:1157–60. doi: 10.1038/nature03891 18.  Bentley R, Chasteen TG (2002) Microbial methylation of metalloids: arsenic, antimony, and bismuth. Microbiol Mol Biol Rev 66:250–71. 19.  Berry D, Widder S (2014) Deciphering microbial interactions and detecting keystone species with co-occurrence networks. Front Microbiol 5:219. doi: 10.3389/fmicb.2014.00219 20.  Bers K, Sniegowski K, Albers P, Breugelmans P, Hendrickx L, De Mot R, Springael D (2011) A molecular toolbox to estimate the number and diversity of Variovorax in the environment: application in soils treated with the phenylurea herbicide linuron. FEMS Microbiol Ecol 76:14–25. doi: 10.1111/j.1574-6941.2010.01028.x 21.  Bijmans MFM, van Helvoort P-J, Buisman CJN, Lens PNL (2009) Effect of the sulfide concentration on zinc bio-precipitation in a single stage sulfidogenic bioreactor at pH 5.5. Sep Purif Technol 69:243–248. doi: 10.1016/j.seppur.2009.07.023 22.  Biswas KC, Woodards N a, Xu H, Barton LL (2009) Reduction of molybdate by sulfate-reducing bacteria. Biometals 22:131–9. doi: 10.1007/s10534-008-9198-8 23.  Bjornsson L, Hugenholtz P, Tyson GW, Blackall LL (2002) Filamentous Chloroflexi (green non-sulfur bacteria) are abundant in wastewater treatment processes with biological nutrient removal. Microbiology 148:2309–2318. 	  	   116	  24.  Bock E, Schmidt I, St�ven R, Zart D (1995) Nitrogen loss caused by denitrifying Nitrosomonas cells using ammonium or hydrogen as electron donors and nitrite as electron acceptor. Arch Microbiol 163:16–20. doi: 10.1007/BF00262198 25.  Breazeal MVR, Reardon KF, Reisman DJ, Pruden A, Hiibel SR, Pereyra LP (2011) Effect of Organic Substrate on the Microbial Community Structure in Pilot-Scale Sulfate-Reducing Biochemical Reactors Treating Mine Drainage. Environ Eng Sci 28:563–572. doi: 10.1089/ees.2010.0237 26.  Breugelmans P, Leroy B, Bers K, Dejonghe W, Wattiez R, De Mot R, Springael D (2010) Proteomic study of linuron and 3,4-dichloroaniline degradation by Variovorax sp. WDL1: evidence for the involvement of an aniline dioxygenase-related multicomponent protein. Res Microbiol 161:208–18. doi: 10.1016/j.resmic.2010.01.010 27.  Brown MN, Briones A, Diana J, Raskin L (2013) Ammonia-oxidizing archaea and nitrite-oxidizing nitrospiras in the biofilter of a shrimp recirculating aquaculture system. FEMS Microbiol Ecol 83:17–25. doi: 10.1111/j.1574-6941.2012.01448.x 28.  Burns AS, Pugh CW, Segid YT, Behum PT, Lefticariu L, Bender KS (2012) Performance and microbial community dynamics of a sulfate-reducing bioreactor treating coal generated acid mine drainage. Biodegradation 23:415–29. doi: 10.1007/s10532-011-9520-y 29.  BURRELL P, KELLER J, BLACKALL L (1999) Characterisation of the bacterial consortium involved in nitrite oxidation in activated sludge. Water Sci Technol 39:45–52. doi: 10.1016/S0273-1223(99)00121-3 30.  Cabello P, Roldán MD, Moreno-Vivián C (2004) Nitrate reduction and the nitrogen cycle in archaea. Microbiology 150:3527–46. doi: 10.1099/mic.0.27303-0 31.  Callbeck CM, Dong X, Chatterjee I, Agrawal A, Caffrey SM, Sensen CW, Voordouw G (2011) Microbial community succession in a bioreactor modeling a souring low-temperature oil reservoir subjected to nitrate injection. Appl Microbiol Biotechnol 91:799–810. doi: 10.1007/s00253-011-3287-2 32.  Campbell BJ, Kirchman DL (2013) Bacterial diversity, community structure and potential growth rates along an estuarine salinity gradient. ISME J 7:210–20. doi: 10.1038/ismej.2012.93 33.  Caporaso JG, Kuczynski J, Stombaugh J, Bittinger K, Bushman FD, Costello EK, Fierer N, Peña AG, Goodrich JK, Gordon JI, Huttley GA, Kelley ST, Knights D, Koenig JE, Ley RE, Lozupone CA, Mcdonald D, Muegge BD, Pirrung M, Reeder J, Sevinsky JR, Turnbaugh PJ, Walters WA, Widmann J, Yatsunenko T, Zaneveld J, Knight R (2010) correspondEnce QIIME allows analysis of high- throughput community sequencing data Intensity normalization improves color calling in SOLiD sequencing. Nat Publ Gr 7:335–336. doi: 10.1038/nmeth0510-335 	  	   117	  34.  Caporaso JG, Kuczynski J, Stombaugh J, Bittinger K, Bushman FD, Costello EK, Fierer N, Peña AG, Goodrich JK, Gordon JI, Huttley GA, Kelley ST, Knights D, Koenig JE, Ley RE, Lozupone CA, McDonald D, Muegge BD, Pirrung M, Reeder J, Sevinsky JR, Turnbaugh PJ, Walters WA, Widmann J, Yatsunenko T, Zaneveld J, Knight R (2010) QIIME allows analysis of high-throughput community sequencing data. Nat Methods 7:335–6. doi: 10.1038/nmeth.f.303 35.  Capua F Di (2013) Nitrogen Removal From Low pH and Heavy Metal Contaminated Mine Wastewater.  36.  Cardenas E, Wu W-M, Leigh MB, Carley J, Carroll S, Gentry T, Luo J, Watson D, Gu B, Ginder-Vogel M, Kitanidis PK, Jardine PM, Zhou J, Criddle CS, Marsh TL, Tiedje JM (2008) Microbial communities in contaminated sediments, associated with bioremediation of uranium to submicromolar levels. Appl Environ Microbiol 74:3718–29. doi: 10.1128/AEM.02308-07 37.  Chandler DP, Jarrell AE, Roden ER, Golova J, Chernov B, Schipma MJ, Peacock AD, Long PE (2006) Suspension array analysis of 16S rRNA from Fe- and SO(4)2- reducing bacteria in uranium-contaminated sediments undergoing bioremediation. Appl Environ Microbiol 72:4672–87. doi: 10.1128/AEM.02858-05 38.  Chasteen TG, Bentley R (2003) Biomethylation of selenium and tellurium: microorganisms and plants. Chem Rev 103:1–25. doi: 10.1021/cr010210+ 39.  Chen F, Xia Q, Ju L-K (2003) Aerobic Denitrification of Pseudomonas aeruginosa Monitored by Online NAD(P)H Fluorescence. Appl Environ Microbiol 69:6715–6722. doi: 10.1128/AEM.69.11.6715-6722.2003 40.  Chen X-P, Zhu Y-G, Xia Y, Shen J-P, He J-Z (2008) Ammonia-oxidizing archaea: important players in paddy rhizosphere soil? Environ Microbiol 10:1978–87. doi: 10.1111/j.1462-2920.2008.01613.x 41.  Chisholm S (2001) Ecological genomics: the application of genomic sciences to understanding the structure and function of marine ecosystems. NSF Work Mar Microb Genomics 1–20. 42.  Cho D-H, Ramanan R, Heo J, Kang Z, Kim B-H, Ahn C-Y, Oh H-M, Kim H-S (2015) Organic carbon, influent microbial diversity and temperature strongly influence algal diversity and biomass in raceway ponds treating raw municipal wastewater. Bioresour Technol 191:481–487. doi: 10.1016/j.biortech.2015.02.013 43.  Cocos IA, Zagury GJ, Clément B, Samson R (2002) Multiple factor design for reactive mixture selection for use in reactive walls in mine drainage treatment. Water Res 36:167–77. 	  	   118	  44.  Coetser SE, Pulles W, Heath RGM, Cloete TE (2006) Chemical characterisation of organic electron donors for sulfate reduction for potential use in acid mine drainage treatment. Biodegradation 17:169–79. doi: 10.1007/s10532-005-7567-3 45.  Cole M, Wrubel J, Henegan P, Janzen C, Holt J, Tobin T (2011) Development of a small-scale bioreactor method to monitor the molecular diversity and environmental impacts of bacterial bio fi lm communities from an acid mine drainage impacted creek. J Microbiol Methods 87:96–104. doi: 10.1016/j.mimet.2011.07.015 46.  Coskuner G, Curtis TP (2002) In situ characterization of nitrifiers in an activated sludge plant: detection of Nitrobacter Spp. J Appl Microbiol 93:431–7. 47.  Coskuner G, Curtis TP (2002) In situ characterization of nitrifiers in an activated sludge plant: detection of Nitrobacter Spp. J Appl Microbiol 93:431–437. doi: 10.1046/j.1365-2672.2002.01715.x 48.  Curtis TP, Head IM, Graham DW (2003) Theoretical ecology for engineering biology. Environ Sci Technol 37:64A–70A. 49.  Dang Y, Ye J, Mu Y, Qiu B, Sun D (2013) Effective anaerobic treatment of fresh leachate from MSW incineration plant and dynamic characteristics of microbial community in granular sludge. Appl Microbiol Biotechnol 97:10563–74. doi: 10.1007/s00253-013-4792-2 50.  Dang Y, Ye J, Mu Y, Qiu B, Sun D (2013) Effective anaerobic treatment of fresh leachate from MSW incineration plant and dynamic characteristics of microbial community in granular sludge. Appl Microbiol Biotechnol. doi: 10.1007/s00253-013-4792-2 51.  Dann AL, Cooper RS, Bowman JP (2009) Investigation and optimization of a passively operated compost-based system for remediation of acidic, highly iron- and sulfate-rich industrial waste water. Water Res 43:2302–16. doi: 10.1016/j.watres.2009.02.030 52.  Dar S a, Kleerebezem R, Stams AJM, Kuenen JG, Muyzer G (2008) Competition and coexistence of sulfate-reducing bacteria, acetogens and methanogens in a lab-scale anaerobic bioreactor as affected by changing substrate to sulfate ratio. Appl Microbiol Biotechnol 78:1045–55. doi: 10.1007/s00253-008-1391-8 53.  Dar SA, Bijmans MFM, Dinkla IJT, Geurkink B, Lens PNL, Dopson M (2009) Population dynamics of a single-stage sulfidogenic bioreactor treating synthetic zinc-containing waste streams. Microb Ecol 58:529–37. doi: 10.1007/s00248-009-9509-9 54.  Dar SA, Yao L, van Dongen U, Kuenen JG, Muyzer G (2007) Analysis of diversity and activity of sulfate-reducing bacterial communities in sulfidogenic bioreactors using 16S rRNA and dsrB genes as molecular markers. Appl Environ Microbiol 73:594–604. doi: 10.1128/AEM.01875-06 	  	   119	  55.  Dejonghe W, Berteloot E, Goris J, Boon N, Crul K, Maertens S, Hofte M, De Vos P, Verstraete W, Top EM (2003) Synergistic Degradation of Linuron by a Bacterial Consortium and Isolation of a Single Linuron-Degrading Variovorax Strain. Appl Environ Microbiol 69:1532–1541. doi: 10.1128/AEM.69.3.1532-1541.2003 56.  Dionisi HM, Layton AC, Harms G, Gregory IR, Robinson KG, Sayler GS (2002) Quantification of Nitrosomonas oligotropha-Like Ammonia-Oxidizing Bacteria and Nitrospira spp. from Full-Scale Wastewater Treatment Plants by Competitive PCR. Appl Environ Microbiol 68:245–253. doi: 10.1128/AEM.68.1.245-253.2002 57.  Van Dongen U, Jetten MS, van Loosdrecht MC (2001) The SHARON-Anammox process for treatment of ammonium rich wastewater. Water Sci Technol 44:153–60. 58.  Dougal K, de la Fuente G, Harris PA, Girdwood SE, Pinloche E, Newbold CJ (2013) Identification of a core bacterial community within the large intestine of the horse. PLoS One 8:e77660. doi: 10.1371/journal.pone.0077660 59.  Dunaj SJ, Vallino JJ, Hines ME, Gay M, Kobyljanec C, Rooney-Varga JN (2012) Relationships between soil organic matter, nutrients, bacterial community structure, and the performance of microbial fuel cells. Environ Sci Technol 46:1914–22. doi: 10.1021/es2032532 60.  Dunham MJ (2007) Synthetic ecology: a model system for cooperation. Proc Natl Acad Sci U S A 104:1741–2. doi: 10.1073/pnas.0611067104 61.  E M, A O, E A, A B, M W, P W (2000) Effect of long-term idle periods on the performance of sequencing batch reactors.  62.  Edgar RC (2004) MUSCLE: multiple sequence alignment with high accuracy and high throughput. Nucleic Acids Res 32:1792–1797. doi: 10.1093/nar/gkh340 63.  Edgar RC (2010) Search and clustering orders of magnitude faster than BLAST. Bioinformatics 26:2460–1. doi: 10.1093/bioinformatics/btq461 64.  Egli K, Langer C, Siegrist H-R, Zehnder AJB, Wagner M, van der Meer JR (2003) Community Analysis of Ammonia and Nitrite Oxidizers during Start-Up of Nitritation Reactors. Appl Environ Microbiol 69:3213–3222. doi: 10.1128/AEM.69.6.3213-3222.2003 65.  Eiler A, Bertilsson S (2004) Composition of freshwater bacterial communities associated with cyanobacterial blooms in four Swedish lakes. Environ Microbiol 6:1228–43. doi: 10.1111/j.1462-2920.2004.00657.x 66.  Eiler A, Bertilsson S (2007) Flavobacteria blooms in four eutrophic lakes: linking population dynamics of freshwater bacterioplankton to resource availability. Appl Environ Microbiol 73:3511–8. doi: 10.1128/AEM.02534-06 	  	   120	  67.  Ellington MJK (2003) Rhodobacter capsulatus gains a competitive advantage from respiratory nitrate reduction during light-dark transitions. Microbiology 149:941–948. doi: 10.1099/mic.0.26090-0 68.  Engel AS, Meisinger DB, Porter ML, Payn RA, Schmid M, Stern LA, Schleifer KH, Lee NM (2010) Linking phylogenetic and functional diversity to nutrient spiraling in microbial mats from Lower Kane Cave (USA). ISME J 4:98–110. doi: 10.1038/ismej.2009.91 69.  EPA U US Environmental Protection Agency.  70.  Erguder TH, Boon N, Wittebolle L, Marzorati M, Verstraete W (2009) Environmental factors shaping the ecological niches of ammonia-oxidizing archaea. FEMS Microbiol Rev 33:855–69. doi: 10.1111/j.1574-6976.2009.00179.x 71.  Etchebehere C, Errazquin MI, Dabert P, Muxí L (2002) Community analysis of a denitrifying reactor treating landfill leachate. FEMS Microbiol Ecol 40:97–106. doi: 10.1111/j.1574-6941.2002.tb00941.x 72.  Faulwetter JL, Gagnon V, Sundberg C, Chazarenc F, Burr MD, Brisson J, Camper AK, Stein OR (2009) Microbial processes influencing performance of treatment wetlands: A review. Ecol Eng 35:987–1004. doi: 10.1016/j.ecoleng.2008.12.030 73.  Faust K, Raes J (2012) Microbial interactions: from networks to models. Nat Rev Microbiol 10:538–50. doi: 10.1038/nrmicro2832 74.  Faust K, Sathirapongsasuti JF, Izard J, Segata N, Gevers D, Raes J, Huttenhower C (2012) Microbial co-occurrence relationships in the human microbiome. PLoS Comput Biol 8:e1002606. doi: 10.1371/journal.pcbi.1002606 75.  FE C (1916) Plant Succession: An Analysis of the Development of Vegetation. Carnegie Inst. Washington, D.C.  76.  Feng C-J, Zhang Z-J, Wang S-M, Fang F, Ye Z-L, Chen S-H (2013) Characterization of microbial community structure in a hybrid biofilm-activated sludge reactor for simultaneous nitrogen and phosphorus removal. J Environ Biol 34:489–99. 77.  Fernández N, Sierra-Alvarez R, Field JA, Amils R, Sanz JL (2008) Microbial community dynamics in a chemolithotrophic denitrification reactor inoculated with methanogenic granular sludge. Chemosphere 70:462–74. doi: 10.1016/j.chemosphere.2007.06.062 78.  Findlay SEG, Sinsabaugh RL, Sobczak W V., Hoostal M (2003) Metabolic and structural response of hyporheic microbial communities to variations in supply of dissolved organic matter. Limnol Oceanogr 48:1608–1617. doi: 10.4319/lo.2003.48.4.1608 	  	   121	  79.  Finster K, Liesack W, Thamdrup B (1998) Elemental Sulfur and Thiosulfate Disproportionation by Desulfocapsa sulfoexigens sp. nov., a New Anaerobic Bacterium Isolated from Marine Surface Sediment. Appl Envir Microbiol 64:119–125. 80.  Finster KW, Kjeldsen KU, Kube M, Reinhardt R, Mussmann M, Amann R, Schreiber L (2013) Complete genome sequence of Desulfocapsa sulfexigens, a marine deltaproteobacterium specialized in disproportionating inorganic sulfur compounds. Stand Genomic Sci 8:58–68. doi: 10.4056/sigs.3777412 81.  Freedman Z, Zhu C, Barkay T (2012) Mercury resistance and mercuric reductase activities and expression among chemotrophic thermophilic Aquificae. Appl Environ Microbiol 78:6568–75. doi: 10.1128/AEM.01060-12 82.  Fu F, Wang Q (2011) Removal of heavy metal ions from wastewaters: a review. J Environ Manage 92:407–18. doi: 10.1016/j.jenvman.2010.11.011 83.  Futamata H, Nagano Y, Watanabe K, Hiraishi A (2005) Unique kinetic properties of phenol-degrading variovorax strains responsible for efficient trichloroethylene degradation in a chemostat enrichment culture. Appl Environ Microbiol 71:904–11. doi: 10.1128/AEM.71.2.904-911.2005 84.  Gadd GM (2010) Metals, minerals and microbes: geomicrobiology and bioremediation. Microbiology 156:609–643. doi: 10.1099/mic.0.037143-0 85.  Gentile ME, Lynn Nyman J, Criddle CS (2007) Correlation of patterns of denitrification instability in replicated bioreactor communities with shifts in the relative abundance and the denitrification patterns of specific populations. ISME J 1:714–28. doi: 10.1038/ismej.2007.87 86.  Gibert O, de Pablo J, Luis Cortina J, Ayora C (2004) Chemical characterisation of natural organic substrates for biological mitigation of acid mine drainage. Water Res 38:4186–4196. doi: 10.1016/j.watres.2004.06.023 87.  Gieg LM, Fowler SJ, Berdugo-Clavijo C (2014) Syntrophic biodegradation of hydrocarbon contaminants. Curr Opin Biotechnol 27:21–29. doi: 10.1016/j.copbio.2013.09.002 88.  Gilbert J a, Steele J a, Caporaso JG, Steinbrück L, Reeder J, Temperton B, Huse S, McHardy AC, Knight R, Joint I, Somerfield P, Fuhrman J a, Field D (2012) Defining seasonal marine microbial community dynamics. ISME J 6:298–308. doi: 10.1038/ismej.2011.107 89.  Lo Giudice A, Brilli M, Bruni V, De Domenico M, Fani R, Michaud L (2007) Bacterium-bacterium inhibitory interactions among psychrotrophic bacteria isolated from Antarctic seawater (Terra Nova Bay, Ross Sea). FEMS Microbiol Ecol 60:383–96. doi: 10.1111/j.1574-6941.2007.00300.x 90.  Given B, Meyer S (1998) Biological treatment of tailings solution at the Nickel Plate Mine.  	  	   122	  91.  Gram L, Grossart H-P, Schlingloff A, Kiørboe T (2002) Possible quorum sensing in marine snow bacteria: production of acylated homoserine lactones by Roseobacter strains isolated from marine snow. Appl Environ Microbiol 68:4111–6. 92.  Green FB, Lundquist TJ, Quinn NWT, Zarate MA, Zubieta IX, Oswald WJ (2003) Selenium and nitrate removal from agricultural drainage using the ALWPS technology. Water Sci Technol 48:299–305. 93.  Grossart H-P, Schlingloff A, Bernhard M, Simon M, Brinkhoff T (2004) Antagonistic activity of bacteria isolated from organic aggregates of the German Wadden Sea. FEMS Microbiol Ecol 47:387–96. doi: 10.1016/S0168-6496(03)00305-2 94.  Guindon S, Delsuc F, Dufayard J-F, Gascuel O (2009) Estimating maximum likelihood phylogenies with PhyML. Methods Mol Biol 537:113–137. doi: 10.1007/978-1-59745-251-9_6 95.  Gusek JJ (2008) Passive Treatment 101: An Overview of the Technologies. In: U.S. EPA/National Groundw. Assoc. Remediat. Abandon. Mine, Lands. Denver, CO, pp 1–13 96.  Hallberg KB, Johnson DB (2005) Microbiology of a wetland ecosystem constructed to remediate mine drainage from a heavy metal mine. Sci Total Environ 338:53–66. doi: 10.1016/j.scitotenv.2004.09.005 97.  Hallin S, Lydmark P, Kokalj S, Hermansson M, Sörensson F, Jarvis A, Lindgren P-E (2005) Community survey of ammonia-oxidizing bacteria in full-scale activated sludge processes with different solids retention time. J Appl Microbiol 99:629–40. doi: 10.1111/j.1365-2672.2005.02608.x 98.  Hamoda MF, Zeidan MO, Al-Haddad AA (1996) Biological nitrification kinetics in a fixed-film reactor. Bioresour Technol 58:41–48. doi: 10.1016/S0960-8524(96)00082-X 99.  Hashim MA, Mukhopadhyay S, Sahu JN, Sengupta B (2011) Remediation technologies for heavy metal contaminated groundwater. J Environ Manage 92:2355–2388. doi: 10.1016/j.jenvman.2011.06.009 100.  Heather L (2010) Passive Sulphate Reducing Bacteria Treatment Ponds: Research and Operation at Highmont Tailings. Annu. Reclam. Rep. Volume II: 101.  Heylen K, Vanparys B, Wittebolle L, Verstraete W, Boon N, De Vos P (2006) Cultivation of denitrifying bacteria: optimization of isolation conditions and diversity study. Appl Environ Microbiol 72:2637–43. doi: 10.1128/AEM.72.4.2637-2643.2006 102.  Hiibel SR, Pereyra LP, Breazeal MVR, Reisman DJ, Reardon KF, Pruden A (2011) Effect of Organic Substrate on the Microbial Community Structure in Pilot-Scale Sulfate-Reducing Biochemical Reactors Treating Mine Drainage. Environ Eng Sci 28:563–572. doi: 10.1089/ees.2010.0237 	  	   123	  103.  Hiibel SR, Pereyra LP, Inman LY, Tischer A, Reisman DJ, Reardon KF, Pruden A (2008) Microbial community analysis of two field-scale sulfate-reducing bioreactors treating mine drainage. Environ Microbiol 10:2087–2097. doi: 10.1111/j.1462-2920.2008.01630.x 104.  Holmes DE, Bond DR, Lovley DR (2004) Electron Transfer by Desulfobulbus propionicus to Fe(III) and Graphite Electrodes. Appl Environ Microbiol 70:1234–1237. doi: 10.1128/AEM.70.2.1234-1237.2004 105.  Huang J-H, Zeng G-M, Zhou C-F, Li X, Shi L-J, He S-B (2010) Adsorption of surfactant micelles and Cd2+/Zn2+ in micellar-enhanced ultrafiltration. J Hazard Mater 183:287–93. doi: 10.1016/j.jhazmat.2010.07.022 106.  I. EA, F. OE (2014) Batch adsorption of heavy metals (Cu, Pb, Fe, Cr and Cd) from aqueous solutions using coconut husk. African J Environ Sci Technol 8:239–246. doi: 10.5897/AJEST2013.1577 107.  Jetten MSM, Niftrik L van, Strous M, Kartal B, Keltjens JT, Op den Camp HJM (2009) Biochemistry and molecular biology of anammox bacteria. Crit Rev Biochem Mol Biol 44:65–84. doi: 10.1080/10409230902722783 108.  Jin D, Wang P, Bai Z, Jin B, Yu Z, Wang X, Zhuang G, Zhang H (2013) Terrimonas pekingensis sp. nov., isolated from bulking sludge, and emended descriptions of the genus Terrimonas, Terrimonas ferruginea, Terrimonas lutea and Terrimonas aquatica. Int J Syst Evol Microbiol 63:1658–64. doi: 10.1099/ijs.0.036848-0 109.  Johnson DB, Hallberg KB (2005) Acid mine drainage remediation options: a review. Sci Total Environ 338:3–14. doi: 10.1016/j.scitotenv.2004.09.002 110.  Johnson PD, Girinathannair P, Ohlinger KN, Ritchie S, Teuber L, Kirby J (2008) Enhanced removal of heavy metals in primary treatment using coagulation and flocculation. Water Environ Res 80:472–9. 111.  Jones CM, Stres B, Rosenquist M, Hallin S (2008) Phylogenetic analysis of nitrite, nitric oxide, and nitrous oxide respiratory enzymes reveal a complex evolutionary history for denitrification. Mol Biol Evol 25:1955–66. doi: 10.1093/molbev/msn146 112.  Jung M-Y, Park S-J, Min D, Kim J-S, Rijpstra WIC, Sinninghe Damsté JS, Kim G-J, Madsen EL, Rhee S-K (2011) Enrichment and characterization of an autotrophic ammonia-oxidizing archaeon of mesophilic crenarchaeal group I.1a from an agricultural soil. Appl Environ Microbiol 77:8635–47. doi: 10.1128/AEM.05787-11 113.  Juretschko S, Timmermann G, Schmid M, Schleifer K-H, Pommerening-Roser A, Koops H-P, Wagner M (1998) Combined Molecular and Conventional Analyses of Nitrifying Bacterium Diversity in Activated Sludge: Nitrosococcus mobilis and Nitrospira-Like Bacteria as Dominant Populations. Appl Envir Microbiol 64:3042–3051. 	  	   124	  114.  Kabdaşli I, Arslan T, Olmez-Hanci T, Arslan-Alaton I, Tünay O (2009) Complexing agent and heavy metal removals from metal plating effluent by electrocoagulation with stainless steel electrodes. J Hazard Mater 165:838–45. doi: 10.1016/j.jhazmat.2008.10.065 115.  Kaksonen AH, Plumb JJ, Franzmann PD, Puhakka JA (2004) Simple organic electron donors support diverse sulfate-reducing communities in fluidized-bed reactors treating acidic metal- and sulfate-containing wastewater. FEMS Microbiol Ecol 47:279–89. doi: 10.1016/S0168-6496(03)00284-8 116.  Kaksonen AH, Plumb JJ, Robertson WJ, Riekkola-Vanhanen M, Franzmann PD, Puhakka JA (2006) The performance, kinetics and microbiology of sulfidogenic fluidized-bed treatment of acidic metal- and sulfate-containing wastewater. Hydrometallurgy 83:204–213. doi: 10.1016/j.hydromet.2006.03.025 117.  Kamijo M, Suzuki T, Kawai K, Murase H (1998) Accumulation of yttrium by Variovorax paradoxus. J Ferment Bioeng 86:564–568. doi: 10.1016/S0922-338X(99)80007-5 118.  Karpinets T V, Park BH, Uberbacher EC (2012) Analyzing large biological datasets with association networks. 40:1–8. doi: 10.1093/nar/gks403 119.  Kathiravan V, Krishnani KK (2014) Pseudomonas aeruginosa and Achromobacter sp.: nitrifying aerobic denitrifiers have a plasmid encoding for denitrifying functional genes. World J Microbiol Biotechnol 30:1187–98. doi: 10.1007/s11274-013-1543-6 120.  Kelly JJ, Siripong S, McCormack J, Janus LR, Urakawa H, El Fantroussi S, Noble PA, Sappelsa L, Rittmann BE, Stahl DA (2005) DNA microarray detection of nitrifying bacterial 16S rRNA in wastewater treatment plant samples. Water Res 39:3229–38. doi: 10.1016/j.watres.2005.05.044 121.  Khoshnoodi M, Dipple G, Baldwin S (2013) Mineralogical Study of a Biologically-Based Treatment System That Removes Arsenic, Zinc and Copper from Landfill Leachate. Minerals 3:427–449. doi: 10.3390/min3040427 122.  Kieu HTQ, Horn H, Müller E (2013) The effect of heavy metals on microbial community structure of a sulfidogenic consortium in anaerobic semi-continuous stirred tank reactors. Bioprocess Biosyst Eng. doi: 10.1007/s00449-013-1012-5 123.  Kim JK, Park KJ, Cho KS, Nam S-W, Park T-J, Bajpai R (2005) Aerobic nitrification-denitrification by heterotrophic Bacillus strains. Bioresour Technol 96:1897–906. doi: 10.1016/j.biortech.2005.01.040 124.  Klein R, Tischler JS, Mühling M, Schlömann M (2014) Geobiotechnology I. Adv Biochem Eng Biotechnol. doi: 10.1007/978-3-642-54710-2 125.  KOKU H (2002) Aspects of the metabolism of hydrogen production by Rhodobacter sphaeroides. Int J Hydrogen Energy 27:1315–1329. doi: 10.1016/S0360-3199(02)00127-1 	  	   125	  126.  Konopka A (2009) What is microbial community ecology? ISME J 3:1223–30. doi: 10.1038/ismej.2009.88 127.  Konstantinidis KT, Tiedje JM (2007) Prokaryotic taxonomy and phylogeny in the genomic era: advancements and challenges ahead. Curr Opin Microbiol 10:504–509. doi: 10.1016/j.mib.2007.08.006 128.  Koren DW, Gould WD, Bédard P (2000) Biological removal of ammonia and nitrate from simulated mine and mill effluents. Hydrometallurgy 56:127–144. doi: 10.1016/S0304-386X(99)00088-2 129.  Koschorreck M, Geller W, Neu T, Kleinsteuber S, Kunze T, Trosiener A, Wendt-Potthoff K (2010) Structure and function of the microbial community in an in situ reactor to treat an acidic mine pit lake. FEMS Microbiol Ecol 73:385–395. doi: 10.1111/j.1574-6941.2010.00886.x 130.  Kosinska K, Miskiewicz T PRECIPITATION OF HEAVY METALS FROM INDUSTRIAL WASTEWATER BY DESULFOVIBRIO DESULFURICANS. doi: 10.5277/epe120205 131.  Kumar PS, Brooker MR, Dowd SE, Camerlengo T (2011) Target region selection is a critical determinant of community fingerprints generated by 16S pyrosequencing. PLoS One 6:e20956. doi: 10.1371/journal.pone.0020956 132.  Kurniawan TA, Chan GYS, Lo W-H, Babel S (2006) Physico–chemical treatment techniques for wastewater laden with heavy metals. Chem Eng J 118:83–98. doi: 10.1016/j.cej.2006.01.015 133.  Labrenz M, Banfield JF (2004) Sulfate-reducing bacteria-dominated biofilms that precipitate ZnS in a subsurface circumneutral-pH mine drainage system. Microb Ecol 47:205–17. doi: 10.1007/s00248-003-1025-8 134.  Lande R (1996) Statistics and Partitioning of Species Diversity, and Similarity among Multiple Communities. Oikos 76:5. doi: 10.2307/3545743 135.  LaPara TM, Ghosh S (2006) Population Dynamics of the Ammonia-Oxidizing Bacteria in a Full-Scale Municipal Wastewater Treatment Facility. Environ Eng Sci 23:309–319. doi: 10.1089/ees.2006.23.309 136.  Lee J, Lee J, Lee TK, Woo S-G, Baek GS, Park J (2013) In-depth characterization of wastewater bacterial community in response to algal growth using pyrosequencing. J Microbiol Biotechnol 23:1472–7. 137.  Lee Y-J, Romanek CS, Wiegel J (2009) Desulfosporosinus youngiae sp. nov., a spore-forming, sulfate-reducing bacterium isolated from a constructed wetland treating acid mine drainage. Int J Syst Evol Microbiol 59:2743–6. doi: 10.1099/ijs.0.007336-0 	  	   126	  138.  Leininger S, Urich T, Schloter M, Schwark L, Qi J, Nicol GW, Prosser JI, Schuster SC, Schleper C (2006) Archaea predominate among ammonia-oxidizing prokaryotes in soils. Nature 442:806–9. doi: 10.1038/nature04983 139.  Letunic I, Bork P (2011) Interactive Tree Of Life v2: online annotation and display of phylogenetic trees made easy. Nucleic Acids Res 39:W475–8. doi: 10.1093/nar/gkr201 140.  Li X, Zhang L, Wang G (2014) Genomic evidence reveals the extreme diversity and wide distribution of the arsenic-related genes in Burkholderiales. PLoS One 9:e92236. doi: 10.1371/journal.pone.0092236 141.  Limpiyakorn T, Kurisu F, Sakamoto Y, Yagi O (2007) Effects of ammonium and nitrite on communities and populations of ammonia-oxidizing bacteria in laboratory-scale continuous-flow reactors. FEMS Microbiol Ecol 60:501–12. doi: 10.1111/j.1574-6941.2007.00307.x 142.  Limpiyakorn T, Shinohara Y, Kurisu F, Yagi O (2005) Communities of ammonia-oxidizing bacteria in activated sludge of various sewage treatment plants in Tokyo. FEMS Microbiol Ecol 54:205–17. doi: 10.1016/j.femsec.2005.03.017 143.  Little AEF, Robinson CJ, Peterson SB, Raffa KF, Handelsman J (2008) Rules of engagement: interspecies interactions that regulate microbial communities. Annu Rev Microbiol 62:375–401. doi: 10.1146/annurev.micro.030608.101423 144.  Liu A, Garcia-Dominguez E, Rhine ED, Young LY (2004) A novel arsenate respiring isolate that can utilize aromatic substrates. FEMS Microbiol Ecol 48:323–32. doi: 10.1016/j.femsec.2004.02.008 145.  Lloyd JR (2003) Microbial reduction of metals and radionuclides. FEMS Microbiol Rev 27:411–425. doi: 10.1016/S0168-6445(03)00044-5 146.  Logan M, Ahmann D, Figueroa L (2003) ASSESSMENT OF MICROBIAL ACTIVITY IN ANAEROBIC COLUMNS TREATING SYNTHETIC MINE DRAINAGE 1.  147.  Logan M V, Reardon KF, Figueroa L a, McLain JET, Ahmann DM (2005) Microbial community activities during establishment, performance, and decline of bench-scale passive treatment systems for mine drainage. Water Res 39:4537–51. doi: 10.1016/j.watres.2005.08.013 148.  Long RA, Azam F (2001) Antagonistic interactions among marine pelagic bacteria. Appl Environ Microbiol 67:4975–83. doi: 10.1128/AEM.67.11.4975-4983.2001 149.  Long RA, Eveillard D, Franco SLM, Reeves E, Pinckney JL (2013) Antagonistic interactions between heterotrophic bacteria as a potential regulator of community structure of hypersaline microbial mats. FEMS Microbiol Ecol 83:74–81. doi: 10.1111/j.1574-6941.2012.01457.x 	  	   127	  150.  Loosdrecht V (2008) Innovative nitrogen removal. Biol wastewater Treat Princ Model Des IWA Publ London 139–155. 151.  Loreau M, Hector A (2001) Partitioning selection and complementarity in biodiversity experiments. Nature 412:72–6. doi: 10.1038/35083573 152.  Lovley DR (1993) Dissimilatory metal reduction. Annu Rev Microbiol 47:263–90. doi: 10.1146/annurev.mi.47.100193.001403 153.  Lovley DR, Phillips EJ (1994) Reduction of Chromate by Desulfovibrio vulgaris and Its c(3) Cytochrome. Appl Environ Microbiol 60:726–8. 154.  Lovley DR, Ueki T, Zhang T, Malvankar NS, Shrestha PM, Flanagan KA, Aklujkar M, Butler JE, Giloteaux L, Rotaru A-E, Holmes DE, Franks AE, Orellana R, Risso C, Nevin KP (2011) Geobacter: the microbe electric’s physiology, ecology, and practical applications. Adv Microb Physiol 59:1–100. doi: 10.1016/B978-0-12-387661-4.00004-5 155.  Loy A, Lehner A, Lee N, Adamczyk J, Meier H, Ernst J, Schleifer K-H, Wagner M (2002) Oligonucleotide microarray for 16S rRNA gene-based detection of all recognized lineages of sulfate-reducing prokaryotes in the environment. Appl Environ Microbiol 68:5064–5081. doi: 10.1128/AEM.68.10.5064-5081.2002 156.  Lozupone C, Knight R (2005) UniFrac: a new phylogenetic method for comparing microbial communities. Appl Environ Microbiol 71:8228–35. doi: 10.1128/AEM.71.12.8228-8235.2005 157.  Lücker S, Wagner M, Maixner F, Pelletier E, Koch H, Vacherie B, Rattei T, Damsté JSS, Spieck E, Le Paslier D, Daims H (2010) A Nitrospira metagenome illuminates the physiology and evolution of globally important nitrite-oxidizing bacteria. Proc Natl Acad Sci U S A 107:13479–84. doi: 10.1073/pnas.1003860107 158.  Ludwig RD, McGregor RG, Blowes DW, Benner SG, Mountjoy K A permeable reactive barrier for treatment of heavy metals. Ground Water 40:59–66. 159.  Ludwig W, Mittenhuber G, Friedrich CG (1993) Transfer of Thiosphaera pantotropha to Paracoccus denitrificans. Int J Syst Bacteriol 43:363–7. 160.  Lupatini M, Suleiman AKA, Jacques RJS, Antoniolli ZI, de Siqueira Ferreira A, Kuramae EE, Roesch LFW (2014) Network topology reveals high connectance levels and few key microbial genera within soils. Front Environ Sci. doi: 10.3389/fenvs.2014.00010 161.  Lyew D, Knowles R, Sheppard J (1994) Biological treatment of acid mine drainage under continuous flow conditions in a reactor. Process Saf Environ Prot Trans Inst Chem Eng Part B 72:42–47. 	  	   128	  162.  Macur RE, Jackson CR, Botero LM, Mcdermott TR, Inskeep WP (2004) Bacterial Populations Associated with the Oxidation and Reduction of Arsenic in an Unsaturated Soil. Environ Sci Technol 38:104–111. doi: 10.1021/es034455a 163.  Malvankar NS, Lovley DR (2012) Microbial nanowires: a new paradigm for biological electron transfer and bioelectronics. ChemSusChem 5:1039–46. doi: 10.1002/cssc.201100733 164.  Martín González AM, Dalsgaard B, Olesen JM (2010) Centrality measures and the importance of generalist species in pollination networks. Ecol Complex 7:36–43. doi: 10.1016/j.ecocom.2009.03.008 165.  Martins M, Faleiro ML, Barros RJ, Veríssimo AR, Costa MC (2009) Biological sulphate reduction using food industry wastes as carbon sources. Biodegradation 20:559–67. doi: 10.1007/s10532-008-9245-8 166.  Mattes A, Evans LJ, Gould DW, Duncan WFA, Glasauer S (2011) The long term operation of a biologically based treatment system that removes As, S and Zn from industrial (smelter operation) landfill seepage. Appl Geochem Sources, Transp Fate Trace Toxic Elem Environ -IAGS 2009 26:1886–1896. 167.  McInerney MJ, Struchtemeyer CG, Sieber J, Mouttaki H, Stams AJM, Schink B, Rohlin L, Gunsalus RP (2008) Physiology, ecology, phylogeny, and genomics of microorganisms capable of syntrophic metabolism. Ann N Y Acad Sci 1125:58–72. doi: 10.1196/annals.1419.005 168.  Medina BY, Torem ML, de Mesquita LMS (2005) On the kinetics of precipitate flotation of Cr III using sodium dodecylsulfate and ethanol. Miner Eng 18:225–231. doi: 10.1016/j.mineng.2004.08.018 169.  Melcer H, Nutt SG (1988) Nitrogen Control of Complex Industrial Wastewaters. J Environ Eng 114:166–178. doi: 10.1061/(ASCE)0733-9372(1988)114:1(166) 170.  Metzker ML (2010) Sequencing technologies - the next generation. Nat Rev Genet 11:31–46. doi: 10.1038/nrg2626 171.  Meulepas RJW, Stams AJM, Lens PNL (2010) Biotechnological aspects of sulfate reduction with methane as electron donor. Rev Environ Sci Bio/Technology 9:59–78. doi: 10.1007/s11157-010-9193-8 172.  Mills LS, Soule ME, Doak DF (1993) The keystone-species concept in ecology and conservation . 219:1–8. 173.  Miura Y, Watanabe Y, Okabe S (2007) Significance of Chloroflexi in Performance of Submerged Membrane Bioreactors (MBR) Treating Municipal Wastewater. Environ Sci Technol 41:7787–7794. doi: 10.1021/es071263x 	  	   129	  174.  Mohanty SR, Kollah B, Hedrick DB, Peacock AD, Kukkadapu RK, Roden EE (2008) Biogeochemical Processes In Ethanol Stimulated Uranium-contaminated Subsurface Sediments. Environ Sci Technol 42:4384–4390. doi: 10.1021/es703082v 175.  Montoya L, Celis LB, Gallegos-García M, Razo-Flores E, Alpuche-Solís ÁG (2013) Consortium diversity of a sulfate-reducing biofilm developed at acidic pH influent conditions in a down-flow fluidized bed reactor. Eng Life Sci 13:302–311. doi: 10.1002/elsc.201200047 176.  Moreau JW, Zierenberg RA, Banfield JF (2010) Diversity of dissimilatory sulfite reductase genes (dsrAB) in a salt marsh impacted by long-term acid mine drainage. Appl Environ Microbiol 76:4819–28. doi: 10.1128/AEM.03006-09 177.  Morita M, Malvankar NS, Franks AE, Summers ZM, Giloteaux L, Rotaru AE, Rotaru C, Lovley DR (2011) Potential for direct interspecies electron transfer in methanogenic wastewater digester aggregates. MBio 2:e00159–11. doi: 10.1128/mBio.00159-11 178.  Morita M, Malvankar NS, Franks AE, Summers ZM, Giloteaux L, Rotaru AE, Rotaru C, Lovley DR (2011) Potential for direct interspecies electron transfer in methanogenic wastewater digester aggregates. MBio 2:e00159–11. doi: 10.1128/mBio.00159-11 179.  Mussmann M, Brito I, Pitcher A, Sinninghe Damsté JS, Hatzenpichler R, Richter A, Nielsen JL, Nielsen PH, Müller A, Daims H, Wagner M, Head IM (2011) Thaumarchaeotes abundant in refinery nitrifying sludges express amoA but are not obligate autotrophic ammonia oxidizers. Proc Natl Acad Sci U S A 108:16771–6. doi: 10.1073/pnas.1106427108 180.  Mustakhimov I, Kalyuzhnaya MG, Lidstrom ME, Chistoserdova L (2013) Insights into denitrification in Methylotenera mobilis from denitrification pathway and methanol metabolism mutants. J Bacteriol 195:2207–11. doi: 10.1128/JB.00069-13 181.  Muyzer G, Stams AJM (2008) The ecology and biotechnology of sulphate-reducing bacteria. Nat Rev Microbiol 6:441–54. doi: 10.1038/nrmicro1892 182.  Okabe S, Satoh H, Watanabe Y (1999) In Situ Analysis of Nitrifying Biofilms as Determined by In Situ Hybridization and the Use of Microelectrodes. Appl Envir Microbiol 65:3182–3191. 183.  Paine RT (1966) Food web complexity and species diversity. Am Nat 100:65–75. 184.  Park H-D, Noguera DR Evaluating the effect of dissolved oxygen on ammonia-oxidizing bacterial communities in activated sludge. Water Res 38:3275–86. doi: 10.1016/j.watres.2004.04.047 185.  Park H-D, Wells GF, Bae H, Criddle CS, Francis CA (2006) Occurrence of ammonia-oxidizing archaea in wastewater treatment plant bioreactors. Appl Environ Microbiol 72:5643–7. doi: 10.1128/AEM.00402-06 	  	   130	  186.  Pereyra L, Hiibel S (2010) Detection and quantification of functional genes of cellulose-degrading, fermentative, and sulfate-reducing bacteria and methanogenic archaea. Appl … 76:2192–202. doi: 10.1128/AEM.01285-09 187.  Pereyra LP, Hiibel SR, Pruden A, Reardon KF (2008) Comparison of microbial community composition and activity in sulfate-reducing batch systems remediating mine drainage. Biotechnol Bioeng 101:702–13. doi: 10.1002/bit.21930 188.  Persson F, Wik T, Sörensson F, Hermansso M (2002) Distribution and activity of ammonia oxidizing bacteria in a large full-scale trickling filter. Water Res 36:1439–48. 189.  Pester M, Bittner N, Deevong P, Wagner M, Loy A (2010) A “rare biosphere” microorganism contributes to sulfate reduction in a peatland. ISME J 4:1591–602. doi: 10.1038/ismej.2010.75 190.  Plugge CM, Zhang W, Scholten JCM, Stams AJM (2011) Metabolic flexibility of sulfate-reducing bacteria. Front Microbiol 2:81. doi: 10.3389/fmicb.2011.00081 191.  Power, M. E. and LSM (1995) The keystone cops meet in Hilo. Trends Ecol Evol 10:182–184. 192.  Proulx SR, Promislow DEL, Phillips PC (2005) Network thinking in ecology and evolution. Trends Ecol Evol 20:345–53. doi: 10.1016/j.tree.2005.04.004 193.  Qiu Y-L, Hanada S, Ohashi A, Harada H, Kamagata Y, Sekiguchi Y (2008) Syntrophorhabdus aromaticivorans gen. nov., sp. nov., the first cultured anaerobe capable of degrading phenol to acetate in obligate syntrophic associations with a hydrogenotrophic methanogen. Appl Environ Microbiol 74:2051–8. doi: 10.1128/AEM.02378-07 194.  Quast C, Pruesse E, Yilmaz P, Gerken J, Schweer T, Yarza P, Peplies J, Glöckner FO (2013) The SILVA ribosomal RNA gene database project: improved data processing and web-based tools. Nucleic Acids Res 41:D590–6. doi: 10.1093/nar/gks1219 195.  Rafrafi Y, Trably E, Hamelin J, Latrille E, Meynial-Salles I, Benomar S, Giudici-Orticoni M-T, Steyer J-P (2013) Sub-dominant bacteria as keystone species in microbial communities producing bio-hydrogen. Int J Hydrogen Energy 38:4975–4985. doi: 10.1016/j.ijhydene.2013.02.008 196.  Ramírez-Flandes S, Ulloa O (2008) Bosque: integrated phylogenetic analysis software. Bioinformatics 24:2539–41. doi: 10.1093/bioinformatics/btn466 197.  Ramos-Padrón E, Bordenave S, Lin S, Bhaskar IM, Dong X, Sensen CW, Fournier J, Voordouw G, Gieg LM (2011) Carbon and sulfur cycling by microbial communities in a gypsum-treated oil sands tailings pond. Environ Sci Technol 45:439–46. doi: 10.1021/es1028487 	  	   131	  198.  Raskin L, Rittmann BE, Stahl DA (1996) Competition and coexistence of sulfate-reducing and methanogenic populations in anaerobic biofilms. Appl Environ Microbiol 62:3847–3857. 199.  Rastogi G, Sani RK (2011) Microbes and Microbial Technology. 29–58. doi: 10.1007/978-1-4419-7931-5 200.  Reguera G, McCarthy KD, Mehta T, Nicoll JS, Tuominen MT, Lovley DR (2005) Extracellular electron transfer via microbial nanowires. Nature 435:1098–1101. doi: 10.1038/nature03661 201.  Rhine ED, Phelps CD, Young LY (2006) Anaerobic arsenite oxidation by novel denitrifying isolates. Environ Microbiol 8:899–908. doi: 10.1111/j.1462-2920.2005.00977.x 202.  Richard T (1990) Interpretation of the correlation-coefficient - a basic review. J Diagn Med Sonogr 6:35–39. 203.  Richarf Y TP (1992) Biological removal of nitrates- report on 7 years of operation and progress. Water Supply 10:151. 204.  Rivière D, Desvignes V, Pelletier E, Chaussonnerie S, Guermazi S, Weissenbach J, Li T, Camacho P, Sghir A (2009) Towards the definition of a core of microorganisms involved in anaerobic digestion of sludge. ISME J 3:700–14. doi: 10.1038/ismej.2009.2 205.  Robertson LA, Kuenen JG (1984) Aerobic denitrification: a controversy revived. Arch Microbiol 139:351–354. doi: 10.1007/BF00408378 206.  Robertson LA, Kuenen JG (1990) Combined heterotrophic nitrification and aerobic denitrification in Thiosphaera pantotropha and other bacteria. Antonie Van Leeuwenhoek 57:139–152. doi: 10.1007/BF00403948 207.  Rothschild LJ, Mancinelli RL (2001) Life in extreme environments. Nature 409:1092–101. doi: 10.1038/35059215 208.  Sahu AK, Conneely T, Nüsslein K, Ergas SJ (2009) Hydrogenotrophic denitrification and perchlorate reduction in ion exchange brines using membrane biofilm reactors. Biotechnol Bioeng 104:483–91. doi: 10.1002/bit.22414 209.  Satoh T, Hoshino Y, Kitamura H (1976) Rhodopseudomonas sphaeroides forma sp. denitrificans, a denitrifying strain as a subspecies of Rhodopseudomonas sphaeroides. Arch Microbiol 108:265–269. doi: 10.1007/BF00454851 210.  Satola B, Wübbeler JH, Steinbüchel A (2013) Metabolic characteristics of the species Variovorax paradoxus. Appl Microbiol Biotechnol 97:541–60. doi: 10.1007/s00253-012-4585-z 211.  Schink B (2013) The Prokaryotes. doi: 10.1007/978-3-642-30123-0 	  	   132	  212.  Schmidtova J, Baldwin SA (2011) Correlation of bacterial communities supported by different organic materials with sulfate reduction in metal-rich landfill leachate. Water Res 45:1115–28. doi: 10.1016/j.watres.2010.10.038 213.  Schramm A, De Beer D, Gieseke A, Amann R (2000) Microenvironments and distribution of nitrifying bacteria in a membrane-bound biofilm. Environ Microbiol 2:680–6. 214.  Schramm A, de Beer D, van den Heuvel JC, Ottengraf S, Amann R (1999) Microscale Distribution of Populations and Activities of Nitrosospira and Nitrospira spp. along a Macroscale Gradient in a Nitrifying Bioreactor: Quantification by In Situ Hybridization and the Use of Microsensors. Appl Envir Microbiol 65:3690–3696. 215.  Schramm A, de Beer D, Wagner M, Amann R (1998) Identification and Activities In Situ of Nitrosospira and Nitrospira spp. as Dominant Populations in a Nitrifying Fluidized Bed Reactor. Appl Envir Microbiol 64:3480–3485. 216.  Senko JM, Zhang G, McDonough JT, Bruns MA, Burgos WD (2009) Metal Reduction at Low pH by a Desulfosporosinus species: Implications for the Biological Treatment of Acidic Mine Drainage. Geomicrobiol J 26:71–82. doi: 10.1080/01490450802660193 217.  Shannon P, Markiel A, Ozier O, Baliga NS, Wang JT, Ramage D, Amin N, Schwikowski B, Ideker T (2003) Cytoscape: a software environment for integrated models of biomolecular interaction networks. Genome Res 13:2498–2504. doi: 10.1101/gr.1239303 218.  Shao M-F, Zhang T, Fang HH-P (2010) Sulfur-driven autotrophic denitrification: diversity, biochemistry, and engineering applications. Appl Microbiol Biotechnol 88:1027–42. doi: 10.1007/s00253-010-2847-1 219.  Shendure J, Ji H (2008) Next-generation DNA sequencing. Nat Biotechnol 26:1135–1145. doi: 10.1038/nbt1486 220.  Shrestha M, Abraham W-R, Shrestha PM, Noll M, Conrad R (2008) Activity and composition of methanotrophic bacterial communities in planted rice soil studied by flux measurements, analyses of pmoA gene and stable isotope probing of phospholipid fatty acids. Environ Microbiol 10:400–12. doi: 10.1111/j.1462-2920.2007.01462.x 221.  Siegert M, Krüger M, Teichert B, Wiedicke M, Schippers A (2011) Anaerobic Oxidation of Methane at a Marine Methane Seep in a Forearc Sediment Basin off Sumatra, Indian Ocean. Front Microbiol 2:249. doi: 10.3389/fmicb.2011.00249 222.  Siegrist H, Salzgeber D, Eugster J, Joss A (2008) Anammox brings WWTP closer to energy autarky due to increased biogas production and reduced aeration energy for N-removal. Water Sci Technol 57:383–8. doi: 10.2166/wst.2008.048 	  	   133	  223.  Singh BK, Dawson LA, Macdonald CA, Buckland SM (2009) Impact of biotic and abiotic interaction on soil microbial communities and functions: A field study. Appl Soil Ecol 41:239–248. doi: 10.1016/j.apsoil.2008.10.003 224.  Siripong S, Rittmann BE (2007) Diversity study of nitrifying bacteria in full-scale municipal wastewater treatment plants. Water Res 41:1110–20. doi: 10.1016/j.watres.2006.11.050 225.  Smibert R (1994) Phenotypic characterization. In: Methods for General and Molecular Bacteriology. Gerhardt P, Murray RGE, Wood WA, Krieg NR Eds. Am Soc Microbiol Washington, DC, 607–654. 226.  Spain AM, Krumholz LR (2011) Nitrate-Reducing Bacteria at the Nitrate and Radionuclide Contaminated Oak Ridge Integrated Field Research Challenge Site: A Review. Geomicrobiol J 28:418–429. doi: 10.1080/01490451.2010.507642 227.  Stahl DA, de la Torre JR (2012) Physiology and diversity of ammonia-oxidizing archaea. Annu Rev Microbiol 66:83–101. doi: 10.1146/annurev-micro-092611-150128 228.  Stams a J (1994) Metabolic interactions between anaerobic bacteria in methanogenic environments. Antonie Van Leeuwenhoek 66:271–94. 229.  Stams AJM, Plugge CM (2009) Electron transfer in syntrophic communities of anaerobic bacteria and archaea. Nat Rev Microbiol 7:568–77. doi: 10.1038/nrmicro2166 230.  Stams AJM, Plugge CM, de Bok FAM, van Houten BHGW, Lens P, Dijkman H, Weijma J (2005) Metabolic interactions in methanogenic and sulfate-reducing bioreactors. Water Sci Technol 52:13–20. 231.  Steele JA, Countway PD, Xia L, Vigil PD, Beman JM, Kim DY, Chow C-ET, Sachdeva R, Jones AC, Schwalbach MS, Rose JM, Hewson I, Patel A, Sun F, Caron DA, Fuhrman JA (2011) Marine bacterial, archaeal and protistan association networks reveal ecological linkages. ISME J 5:1414–25. doi: 10.1038/ismej.2011.24 232.  Stoner DL (1993) Biotechnology for the Treatment of Hazardous Waste. CRC Press 233.  Straub KL, Benz M, Schink B, Widdel F (1996) Anaerobic, nitrate-dependent microbial oxidation of ferrous iron. Appl Environ Microbiol 62:1458–60. 234.  Susan A. Baldwin, Parissa Mirjafari, Maryam Rezahdebashi GS and JT (2015) Start-up of a passive remediation bioreactor for sulfate and selenium removal from mine tailings water. 10th Int. Conf. Acid Rock Drain.  235.  Taubert M, Vogt C, Wubet T, Kleinsteuber S, Tarkka MT, Harms H, Buscot F, Richnow H-H, von Bergen M, Seifert J (2012) Protein-SIP enables time-resolved analysis of the carbon flux in a sulfate-reducing, benzene-degrading microbial consortium. ISME J 6:2291–301. doi: 10.1038/ismej.2012.68 	  	   134	  236.  Thauer RK, Kaster A-K, Seedorf H, Buckel W, Hedderich R (2008) Methanogenic archaea: ecologically relevant differences in energy conservation. Nat Rev Microbiol 6:579–91. doi: 10.1038/nrmicro1931 237.  Thauer RK, Kaster A-K, Seedorf H, Buckel W, Hedderich R (2008) Methanogenic archaea: ecologically relevant differences in energy conservation. Nat Rev Microbiol 6:579–591. doi: 10.1038/nrmicro1931 238.  Thomas RA, Lawlor K, Bailey M, Macaskie LE (1998) Biodegradation of metal-EDTA complexes by an enriched microbial population. Appl Environ Microbiol 64:1319–22. 239.  Tucker MD, Barton LL, Thomson BM (1998) Reduction of Cr, Mo, Se and U by Desulfovibrio desulfuricans immobilized in polyacrylamide gels. J Ind Microbiol Biotechnol 20:13–9. 240.  Tully BJ, Heidelberg JF (2013) Microbial communities associated with ferromanganese nodules and the surrounding sediments. Front Microbiol 4:161. doi: 10.3389/fmicb.2013.00161 241.  Veech JA, Summerville KS, Crist TO, Gering JC (2002) The additive partitioning of species diversity: recent revival of an old idea. Oikos 99:3–9. doi: 10.1034/j.1600-0706.2002.990101.x 242.  Vick-Majors TJ, Priscu JC, Amaral-Zettler LA (2014) Modular community structure suggests metabolic plasticity during the transition to polar night in ice-covered Antarctic lakes. ISME J 8:778–89. doi: 10.1038/ismej.2013.190 243.  Volpicella M, Costanza A, Palumbo O, Italiano F, Claudia L, Placido A, Picardi E, Carella M, Trotta M, Ceci LR (2014) Rhodobacter sphaeroides adaptation to high concentrations of cobalt ions requires energetic metabolism changes. FEMS Microbiol Ecol 88:345–57. doi: 10.1111/1574-6941.12303 244.  Walker CB, Redding-Johanson AM, Baidoo EE, Rajeev L, He Z, Hendrickson EL, Joachimiak MP, Stolyar S, Arkin AP, Leigh J a, Zhou J, Keasling JD, Mukhopadhyay A, Stahl D a (2012) Functional responses of methanogenic archaea to syntrophic growth. ISME J 6:2045–55. doi: 10.1038/ismej.2012.60 245.  Wang J, Li Q, Li M-M, Chen T-H, Zhou Y-F, Yue Z-B (2014) Competitive adsorption of heavy metal by extracellular polymeric substances (EPS) extracted from sulfate reducing bacteria. Bioresour Technol 163:374–6. doi: 10.1016/j.biortech.2014.04.073 246.  Wang X, Wen X, Criddle C, Wells G, Zhang J, Zhao Y (2010) Community analysis of ammonia-oxidizing bacteria in activated sludge of eight wastewater treatment systems. J Environ Sci 22:627–634. doi: 10.1016/S1001-0742(09)60155-8 	  	   135	  247.  Watanabe Y, Masuda S, Ishiguro M (2011) Simultaneous Nitrification and Denitrification in Micro-Aerobic Biofilms.  248.  Waybrant KR, Blowes DW, Ptacek CJ (1998) Selection of Reactive Mixtures for Use in Permeable Reactive Walls for Treatment of Mine Drainage. Environ Sci Technol 32:1972–1979. doi: 10.1021/es9703335 249.  Wei X, Fang L, Cai P, Huang Q, Chen H, Liang W, Rong X (2011) Influence of extracellular polymeric substances (EPS) on Cd adsorption by bacteria. Environ Pollut 159:1369–74. doi: 10.1016/j.envpol.2011.01.006 250.  Werner JJ, Knights D, Garcia ML, Scalfone NB, Smith S, Yarasheski K, Cummings TA, Beers AR, Knight R, Angenent LT (2011) Bacterial community structures are unique and resilient in full-scale bioenergy systems. Proc Natl Acad Sci U S A 108:4158–63. doi: 10.1073/pnas.1015676108 251.  Wes R Nitrification in Cold Weather. In: http://www.lagoonsonline.com/ripple3.htm. http://www.lagoonsonline.com/ripple3.htm. Accessed 10 Oct 2015 252.  Widdel F (1987) New types of acetate-oxidizing, sulfate-reducing Desulfobacter species, D. hydrogenophilus sp. nov., D. latus sp. nov., and D. curvatus sp. nov. Arch Microbiol 148:286–291. doi: 10.1007/BF00456706 253.  Wöhlbrand L, Jacob JH, Kube M, Mussmann M, Jarling R, Beck A, Amann R, Wilkes H, Reinhardt R, Rabus R (2013) Complete genome, catabolic sub-proteomes and key-metabolites of Desulfobacula toluolica Tol2, a marine, aromatic compound-degrading, sulfate-reducing bacterium. Environ Microbiol 15:1334–55. doi: 10.1111/j.1462-2920.2012.02885.x 254.  Wood JM, Cheh A, Dizikes LJ, Ridley WP, Rakow S, Lakowicz JR (1978) Mechanisms for the biomethylation of metals and metalloids. Fed Proc 37:16–21. 255.  Wrage N, Velthof G., van Beusichem M., Oenema O (2001) Role of nitrifier denitrification in the production of nitrous oxide. Soil Biol Biochem 33:1723–1732. doi: 10.1016/S0038-0717(01)00096-7 256.  Wrede C, Dreier A, Kokoschka S, Hoppert M (2012) Archaea in symbioses. Archaea 2012:596846. doi: 10.1155/2012/596846 257.  Wright JJ, Konwar KM, Hallam SJ (2012) Microbial ecology of expanding oxygen minimum zones. Nat Rev Microbiol 10:381–94. doi: 10.1038/nrmicro2778 258.  Yi T, Lee E-H, Kang S, Shin J, Cho K-S (2012) Structure and dynamics of microbial community in full-scale activated sludge reactors. J Ind Microbiol Biotechnol 39:19–25. doi: 10.1007/s10295-011-0994-8 	  	   136	  259.  Yoon D-N, Park S-J, Kim S-J, Jeon CO, Chae J-C, Rhee S-K (2010) Isolation, characterization, and abundance of filamentous members of Caldilineae in activated sludge. J Microbiol 48:275–83. doi: 10.1007/s12275-010-9366-8 260.  You J, Das A, Dolan EM, Hu Z (2009) Ammonia-oxidizing archaea involved in nitrogen removal. Water Res 43:1801–9. doi: 10.1016/j.watres.2009.01.016 261.  Zagury GJ, Kulnieks VI, Neculita CM (2006) Characterization and reactivity assessment of organic substrates for sulphate-reducing bacteria in acid mine drainage treatment. Chemosphere 64:944–54. doi: 10.1016/j.chemosphere.2006.01.001 262.  Zhang S-W, Wei Z-G, Zhou C, Zhang Y-C, Zhang T-H (2013) Exploring the interaction patterns in seasonal marine microbial communities with network analysis. In: 2013 7th Int. Conf. Syst. Biol. IEEE, pp 63–68 263.  Zhao Y-G, Wang A-J, Ren N-Q (2010) Effect of carbon sources on sulfidogenic bacterial communities during the starting-up of acidogenic sulfate-reducing bioreactors. Bioresour Technol 101:2952–9. doi: 10.1016/j.biortech.2009.11.098 264.  Zhou Y, Zhang Y-Q, Zhi X-Y, Wang X, Dong J, Chen Y, Lai R, Li W-J (2008) Description of Sinobacter flavus gen. nov., sp. nov., and proposal of Sinobacteraceae fam. nov. Int J Syst Evol Microbiol 58:184–9. doi: 10.1099/ijs.0.65244-0 265.  Zis T, Ronningen V, Scrosati R (2004) Minor improvement for intertidal seaweeds and invertebrates after acid mine drainage diversion at Britannia Beach, Pacific Canada. Marine Pollution Bulletin, Vol. 48 266.  Canadian Environmental Quality Guidelines. http://www.ccme.ca/en/resources/canadian_environmental_quality_guidelines/index.html. Accessed 15 Mar 2015 267.  Ambient Water Quality Guidelines for Sulphate. http://www.env.gov.bc.ca/wat/wq/BCguidelines/sulphate/sulphate.html. Accessed 18 Jun 2015 268.  http://www.itrcweb.org/miningwaste-guidance/to_bioreactors.htm.  269.  http://www.itrcweb.org/miningwaste-guidance/cs23_stowell.htm.  270.  Passive Treatment of Mining Influenced Water: An Overview of Available Technologies and the Periodic Table. https://www.oilgasandmining.com/volume1/issue2/85-v1n2-gusek. Accessed 30 Mar 2015 271.  Numerical Ecology, Volume 24, Third Edition (Developments in Environmental Modelling): P. Legendre, Loic F J Legendre: 9780444538680: Amazon.com: Books. 	  	   137	  http://www.amazon.com/Numerical-Ecology-Developments-Environmental-Modelling/dp/0444538682. Accessed 18 Aug 2014 272.  greengenes.secondgenome.com. http://greengenes.secondgenome.com/. Accessed 6 Oct 2014 273.  National Center for Biotechnology Information. https://www.ncbi.nlm.nih.gov/. Accessed 6 Oct 2014 274.  (2001) Archaea: Ancient Microbes, Extreme Environments, and the Origin of Life. Gulf Professional Publishing 275.  (2008) Tailings and Mine Waste ’08. CRC Press 276.  (2011) Environment Canada - Pollution and Waste - Metal Mining Investigation of Cause Workshop.  277.  (2011) Metal Mining - DEV - Pollution and Waste - Environment Canada.  278.  (2014) Reference Guide to Treatment Technologies for Mining-Influenced Water. EPA, United States Enviornmental Prot. Agency              	  	   138	  APPENDIX A 	  	  	  	   a)                                                                                 b)  c)                                                                                 d) Figure A.1: Alpha diversity rarefaction curves for the BCRs and environmental samples. Diversity indices are based on a) Observed-species b) Phylogenetic diversity (PD) - Whole tree analysis, c) Simpson and d) estimated (Chao1) OTU richness for each site using an OTU threshold of 94% identity.    	  	   139	   	   Figure A.2: Three-dimensional principal coordinate analysis based on unweighted UniFrac distances       between samples from different depths (layers). Axis 1 explained 28.92% of variation, axis 2, 8.34%, and axis 3 explained 4.48% of variation.   	  	   140	    Figure A.3: Order-level heatmap demonstrates relative distribution of sequences across sites. OTU numbers and their Silva 111 assigned taxonomy are given on the right hand side. Sample names are column labels. The relative abundance for each OTU in different sites is colored in shades of blue (low relative abundance) to red (high relative abundance). Numbers on the key are fractions of reads in each of the feature.  BCR 3Inoculum_Pond(IP)BCR 1Algae_pond (AP)BCR 2BCR 4Archaea_Euryarchaeota_Methanomicrobia_MethanosarcinalesBacteria_Proteobacteria_DeltaproteobacterLDB6K%ï7]7ïBacteria_ArmatimonadetesBacteria_Spirochaetes_Spirochaetales_SpirochaetaceaeArchaea_EuryDUFKDHRWDB7KHrPRSODVPDWDB7KHrmoplasmatalesBacteria_Firmicutes_Bacilli_BacillalesBacteria_Actinobacteria_Micrococcales_DermatophilaceaeBacteria_Proteobacteria_Alphaproteobacteria_Rhi]obialesArchaea_Euryarchaeota_Methanobacteria_MethanobacterialesBacteria_Firmicutes_Clostridia_ClostridialesBacteria_Bacteroidetes_Bacteroidia_BacteroidalesArchaea_Euryarchaeota_Methanomicrobia_MethanomicrobialesBacteria_Bacteroidetes_vadinHA17_soilBacteria_Firmicutes_Erysipelotrichi_ErysipelotrichalesBacteria_Fibrobacteres_Fibrobacteria_FibrobacteralesBacteria_Planctomycetes_Planctomycetacia_PlanctomycetalesBacteria_Proteobacteria_Gammaproteobacteria_B38Bacteria_Proteobacteria_Gammaproteobacteria_MethylococcalesBacteria_Proteobacteria_Deltaproteobacteria_SyntrophobacteralesBacteria_Acidobacteria_Holophagae_SJ$ïBacteria_Bacteroidetes_vadinHA17BacterLDB%DFWHURLGHWHVB6%ïBacteria_Verrucomicrobia_Opitutae_OpitutalesBacteria_Acidobacteria_Holophagae_HolophagalesBacterLDB%DFWHURLGHWHVB:&+%ïBacterLDB%DFWHURLGHWHVB6%ïBVRLOBacteria_Bacteroidetes_BSV13Bacteria_Bacteroidetes_BSV13_FukuN63Bacteria_Proteobacteria_Betaproteobacteria_RhodocyclalesBacteria_Candidate_division_WS3Bacteria_Chlorobi_Chlorobia_ChlorobialesBacteria_Candidate_division_OD1Archaea_Euryarchaeota_Halobacteria_HalobacterialesBacteria_Candidate_division_OP3Bacteria_Candidate_division_OP11BacterLDB%DFWHURLGHWHVB6%ïB%DFWHURLGHWHVBacteria_Chlorobi_Ignavibacteria_IgnavibacterialesBacteria_Candidate_division_OD1_candidate_division_WWE3Bacteria_Nitrospirae_Nitrospira_NitrospiralesBacteria_Chloroflexi_Caldilineae_CaldilinealesBacteria_Proteobacteria_Betaproteobacteria_BurkholderialesBacteria_Bacteroidetes_Cytophagia_CytophagalesBacteria_Proteobacteria_Betaproteobacteria_HydrogenophilalesBacteria_Proteobacteria_Deltaproteobacteria_MyxococcalesBacteria_ActinobacterLDB7KHrmoleophilia_GaiellalesBacteria_Chloroflexi_Anaerolineae_AnaerolinealesBacteria_Proteobacteria_Deltaproteobacteria_DesulfobacteralesBacteria_Proteobacteria_Gammaproteobacteria_XanthomonadalesBacteria_Proteobacteria_Gammaproteobacteria_ChromatialesBacteria_Proteobacteria_Alphaproteobacteria_Sphingomonadales$UFKDHDB7KDXPDUFKDHRWDB0LVFHOODQHRXVB&UHQDUFKDHRWLFB*URXSBacteria_Proteobacteria_Deltaproteobacteria_DesulfuromonadalesBacteria_Proteobacteria_Betaproteobacteria_MethylophilalesBacteria_Cyanobacteria_SubsectionIII_FamilyIBacteria_Proteobacteria_Betaproteobacteria_NitrosomonadalesBacteria_Bacteroidetes_Sphingobacteriia_SphingobacterialesBacteria_Cyanobacteria_ChloroplastBacteria_Gemmatimonadetes_Gemmatimonadales_GemmatimonadaceaeBacteria_Acidobacteria_Acidobacteriales_AcidobacteriaceaeBacteria_Actinobacteria_Micrococcales_MicrobacteriaceaeBacteria_Bacteroidetes_Flavobacteria_FlavobacterialesBacteria_Proteobacteria_Alphaproteobacteria_RhodobacteralesBacteria_Actinobacteria_Acidimicrobiia_AcidimicrobialesBacteria_Proteobacteria_Gammaproteobacteria_PseudomonadalesBacterLDB:&+%ïBacteria_Candidate_division_OP11_anaerobicBacteria_Candidate_division_OP11_eubacterLXPB:&+%ïBacteria_RF3Bacteria_Proteobacteria_Alphaproteobacteria_RhodospirillalesBacteria_Candidate_division_WS6Bacteria_Bacteroidetes_Class_Incertae_Sedis_Order_Incertae_SedisBacteria_Firmicutes_Bacilli_Lactobacillales(ï(ï(ï(ï	  	   141	   Figure A.4: Order-level distribution of SRM-related sequences in all BCRs as determined by pyrotag sequencing method.       	  	   142	    Figure A.5: Phylogeny-based dendrogram derived from the UniFrac analysis clusters samples based on similarity/dissimilarity of SRM-Methanogenic composition. The OTU-table used for this analysis generated based on the 97% homology cut-off. UniFrac uses phylogenetic information of present microbial population to compare samples. UPGMA (Unweighted Pair Group Method with Arithmetic Mean), which is a hierarchical clustering method used to classify samples on the basis of their pairwise similarities in their sequences.     BCR1BCR1APAPAPBCR4BCR4BCR4BCR4BCR4BCR4BCR1BCR1APBCR1BCR1BCR1APBCR2BCR2BCR2BCR2IPIPIPBCR2BCR2BCR2SOILSOILSOILSOILBCR3BCR3BCR3BCR3BCR3BCR3	  	   143	    Figure A.6: Bipartite network showing SRM-related OTUs found in the BCRs and AP. OTUs and BCRs are designated as two types of nodes (OTU-nodes (squares for cultured and diamond for environmental taxa) and bioreactor-nodes (circles)) in a bipartite network in which OTU-nodes are connected via edges to bioreactor-nodes in which their sequences were found. In the network BCRs clustered together according to their shared OTUs – BCRs that share more OTUs cluster closer together. This is weighted according to the number of sequences within an OTU. The darker purple OTU nodes were found in more bioreactors and the OTU-nodes with lighter purple to white color were restricted to certain reactor. It provides visual display of shared versus unique OTUs and displays how OTUs are partitioned between bioreactors. To cluster the OTUs and bioreactors in the network, a stochastic spring-embedded algorithm was used. The OTU table used for this analysis generated based on the 97% homology cut-off. 398365931354976326613931097452539661925107745531421216 2661494286129242684495622692343975330237641101028505504241922141087527718119834549BCR328593769 4855404422438151182411503825705280515813301145354340397519624254814146033912643902168271647825181418182013850152338801339924499525313624287625634083392810234768277235563462417217043102371950152951206343139123193479648542259291917452355BCR1150416052136107850049182947281721652994125942213536445444728255050AP1404631311355138171367968512743865204972126849443263169429083482807321141553937451BCR4 856425057337169523025061135836838828532483950244125477615473923153297667331725178483226130922651583197434130BCR24412201313071512	  	   144	    Figure A.7: Bipartite network showing methanogen-related OTUs found in BCRs and AP. OTUs and BCRs are designated as two types of nodes (OTU-nodes (squares) and bioreactor-nodes (circles)) in a bipartite network in which OTU-nodes are connected via edges to bioreactor-nodes in which their sequences were found. The OTU table used for this analysis was generated based on the 97% homology cut-off. Caption is same as Figure A.4.     2148BCR 36431662158337017694358655419794767274432064752539014803583334168341244101 723 21423212222522AlgaePondBCR 1BCR 4394829443998128711266919082 35391645964825399439636003995409452789840022024199539794885322835605420313156239462462477789237642598BCR 221075040827Innoc.Pond34971646707271544061274328735445026	  	   145	     Figure A.8: Co-occurrence network showing significant interactions of microbial taxa (based on 94% homology cut-off OTU-table) in BCR3. Lines connecting two OTUs represent the interaction between them (r ≥  |0.8| and p-value ≤ 0.05). The sizes of nodes in the network are proportional to the value of betweenness centrality and they are colored based on the value of closeness centrality. (Shades of green (low CC) to red (high CC)). Numbers are the OTUs identification number. 13416715825370155151261814201169611818880981607320511818731324725558104141458725221020412262115262373123610724313912719621224089153841542021942341922291931252352151983414111418314763301125724217161186942474920617622612124971061442186514095137172177822211111169016324821312716637180173647712417975855527228189725919225822023357423119117532108992091001951625228130219136965723217469811651610911318424615139119714641202511109113523910468762385644716811786131123623013354211208291051561781121682031201287824122210313819918583791501021904025092182149332334320093119142216223245224502174525	  	   146	     Figure A.9: Co-occurrence network showing significant interactions of microbial taxa (based on 94% homology cut-off OTU-table) in other BCRs. Lines connecting two OTUs represent the interaction between them (r ≥  |0.8| and p-value ≤ 0.05). The sizes of nodes in the network are proportional to the value of betweenness centrality and they are colored based on the value of closeness centrality (shades of green (low CC) to red (high CC)).  Numbers are the OTUs identification number.  5963883361423713412842953542373904212661949918513416917811325064535211114724141937527816736027143341834335078124698318601703102686731749219925326921115212111893400276153 219136732624626354166157761538327251374338314114156256428175362303734302244221398117339141640210171 2801453402013871162262041832932063094261224321024820036804293643652603471031042022915142723815892312222447210018240326323227531432553768225160333137111442511078725140965217228303913727431621624832822991713313939230843217632914166386164346863202473513371777330734540523230575033564315499719016125355414106424420315024240812089110207129267211501124071892853762642433301652963483848358751921341762844061091313125232212334443611863009636311727441527518313396195349229429229417252653853703281321553932401461722343893811331612791922038229118082351965298230149319404552453682839421229830617433285297254236198472052902201273251262863992891442332733243211935319338418711513633410209138179130412181972312356181154239342203344313043162772772151682623692613883594114261902740130242337832140188258395135397223184140380148571632814123661053441022130125928727070128379357108	  	   147	   Figure A.10: Preparation of culture media (top), measurement of denitrification (bottom). Nitrite is detected and analyzed by formation of a red pink color upon treatment of a NO2- -containing sample with the Griess reagent. When sulphanilic acid is added (in the picture its sulphonamide is shown instead), the nitrites form a diazonium salt. When the azo dye agent (N-alpha-naphthyl-ethylenediamine) is added a pink color develops. (Griess reaction is from http://en.wikipedia.org/wiki/Griess_test).    Table A.1: Average number of raw sequences retrieved from each site.       !"#$% &'()&'(*+, &'(-&'(./, !0"120340546784%$9:$;<$%4=$64%7>=1$ !"#$% &'%# $()! )$'( &!#! )&#! $#!%!"#$% &'()*$'#+"#,"("-.#"/'0$'"#,"("-.#"/'!$##)"'1234.56$,+"#,"("-.#"/'0$'"#,"("-.#"/'!$##)"'1234.56$,&'()*$'#+"#,"("-.#"/'0$'"#,"("-.#"/'!$##)"'1234.56$,+"#,"("-.#"/'0$'"#,"("-.#"/'!$##)"'1234.56$,+/72/(2,.82%$9*$'-$%2:$,2%.5:)$ ;<;= =>?@ A?B> =>CD ;B?@ AB<@ =C?@ A;=@ ABBD ;?BD ;>AD =D<? A;== CA>=	  	   148	   Table A.1: Metal metabolism related genes that are related to Nitrospira sp. and Nitrosomonas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	  	   149	   a)   b)  Figure A.11: a) Denitrification pathway. B) Genes involved in denitrification pathway based on SEED database.    	  	   150	  Table A.2: Denitrifying bacterial-related ORFs that are related to metal metabolism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`-#3(",/0=&FQ&Q"--#/)#&(#-'"0'",0&ES&("-4&68&:&QU&H&V&WSN6RF5&6GM<<H&_8R9WRa?NN?R8bW954Q8?RZS&T;T<;T;T'+'4A&9+'4&I#()@.&'2P#@/O+),'O'#$()*(&"II@*J&/-#,012-/"-&1"-("#0"&KGLH:Y&C"#B.?("/#@&"J12-/"-%&c_8'211"-?-"0)0/#,'"&1-2/"),%&9214&I#()@.&1-2/"),;1)'%<%)1= c"#B.&("/#@&"II@*J&1*(1&9+'4&71-"'*-02-=&KGLH:Y&C"#B.?("/#@&"J12-/"-%&c_8&I#()@.'+'4&C"#B.&("/#@&"II@*J&1*(1&9+'4&KGLH:Y&C"#B.?("/#@&"J12-/"-%&c_8&I#()@.#-0",)/"&2J)$#0"%&@#-3"&0*P*,)/&KGM:><&2J)$#0"&0*P*,)/&\89]T;HG;YM;T^9211"-?/-#,012-/),3&6?/.1"&456#0"&FQ&RC)+2P)*(&@"3*(),20#-*(&PB;&B)')#"&ES&#'/6&68&T&QU&V&WSN6RF5&XYZ>U:&:;<;:;A?RZS&:;<;:;AP)@"&#')$]02$)*(&0.(12-/"-&KG::H>&#-0",)/"&/-#,012-/"-%&49R:&I#()@.6-2P#P@"&'211"-?/-#,012-/),3&456#0"&Q.,4&FQ&Q.,"'C2'2''*0&01;&70/-#),&4599&HLTAA&O&699&<:GT&Q4WE&TAGHOT=&'+/R&C"#B.&("/#@&-"012,0"&-"3*@#/2-&KGL<<>&/d2?'2(12,",/&0.0/"(%&F(1R&I#()@.%&'211"-&-"0)0/#,'"&1C201C#/"&-"3*@2,&-"3*@#/2-&9*0R>+/0+)'(3//1 '21QH&/d2?'2(12,",/&-"3*@#/2-.&0.0/"(&0",02-.&C)0/)$),"&e),#0"&KGL<AA&0.0/"(%&F(1R&I#()@.%&C"#B.&("/#@&0",02-&e),#0"&9*0Q?$'%*1('//+8, '21QH&/d2?'2(12,",/&-"3*@#/2-.&0.0/"(&0",02-.&C)0/)$),"&e),#0"&KGL<AA&0.0/"(%&F(1R&I#()@.%&C"#B.&("/#@&0",02-&e),#0"&9*0Qc"#B.&("/#@&RSb&"II@*J&2*/"-&("(P-#,"&1-2/"),%&9+'9&I#()@.C"#B.&("/#@&"II@*J&1*(1&9+'4(#3,"0)*(O'2P#@/&/-#,012-/"-&92-4&KG:HMA&("/#@&)2,&/-#,012-/"-%&_N5&I#()@.@'2)%8%&%A18 (*@/)'211"-&2J)$#0"&/.1"&:1'2f&'211"-&-"0)0/#,'"&1-2/"),&f&1-"'*-02-&KGLH::51/82%A'1 Q)3(#>A&01"')I)'&/-#,0'-)1/)2,#@&-"3*@#/2-&FQ&R#@0/2,)#&("/#@@)$*-#,0&<;A;T;<?RZS&<;A;T;<?$1+3)1,-5$%7%("(/1(3134 '+'bH&1*/#/)B"&'2P#@/?+),'?'#$()*(&-"0)0/#,'"&1-2/"),&KG:HY>&'#/)2,&"II@*J&0.0/"(&1-2/"),%&9b`&I#()@.B32$"/%23A3)1 C"#B.&("/#@&/-#,0@2'#/),3&6?/.1"&456#0"&KGT>:A&9$H[OD,H[?"J12-/),3&456#0"&\89]:;<;:;:&:;<;:;>^,20b&1*/#/)B"&'211"-&P),$),3&1"-)1@#0()'&1-2/"),&KGLHTM&,)/-2*0&2J)$#0"&#''"002-.&1-2/"),KGHGGM&'2P#@/O,)'e"@&/-#,012-/&0.0/"(&1"-("#0"&1-2/"),(#3,"0)*(&'2P#@/&/-#,012-/&1-2/"),&92-4&KG:HMA&("/#@&)2,&/-#,012-/"-%&_N5&I#()@.'+'4H&92P#@/?+),'?'#$()*(&-"0)0/#,'"&("(P-#,"&'2(12,",/%&9#/)2,&"II@*J&0.0/"(&KGLH:Y&C"#B.?("/#@&"J12-/"-%&c_8&I#()@.C$)%&12'1(313,-D1&&1#)%23%*1(23)'14 9+'4&I#()@.&C"#B.&("/#@&RSb&"II@*J&1-2/"),&KGLLML&9*7N=O437N=&0.0/"(&("(P-#,"&1-2/"),&9*04C"#B.&("/#@&"II@*J&1*(1%&9+'4&I#()@.C"#B.&("/#@&"II@*J&1*(1%&9+'4&I#()@.&KGLH:Y&C"#B.?("/#@&"J12-/"-%&c_8&I#()@./d2&'2(12,",/&C"#B.&("/#@&-"012,0"&/-#,0'-)1/)2,#@&-"3*@#/2-%&'211"-&-"0)0/#,'"&1C201C#/"&-"3*@2,&-"3*@#/2-&9*0Rc"#B.&("/#@&/-#,0@2'#/),3&6?/.1"&456#0"&FQ&f-*'"@@#&("@)/",0)0&P)2/.1"&H&70/-#),&4599&H:A>L=&:;T;H;HT5$%71A%*1(23), C"#B.&("/#@&"II@*J&1*(1&9+'4&KGLH:Y&C"#B.?("/#@&"J12-/"-%&c_8&I#()@.E/1<%*1(23)'+&,-91(23)%'732384 9+'4&I#()@.&C"#B.&("/#@&"II@*J&1*(1&KGLLML&9*7N=O437N=&0.0/"(&("(P-#,"&1-2/"),&9*04c"#B.&("/#@&/-#,012-/O$"/2J)I)'#/)2,&1-2/"),F+)#/3,A%AG8+/0+),*1(23)'1 c"#B.&("/#@&/-#,0@2'#/),3&6?/.1"&456#0"&FQ&f-*'"@@#&("@)/",0)0&P)2/.1"&H&70/-#),&4599&H:A>L=&:;T;H;HT#-0",)/"&2J)$#0"%&@#-3"&0*P*,)/&KGM:><&2J)$#0"&0*P*,)/&\89]T;HG;YM;T^4-0",)'&"II@*J&1*(1&1-2/"),&\RC2$210"*$2(2,#0&1#@*0/-)0&9E4GGY^#-0",)/"?#'/)B#/"$&456#0"&4-04&789]:;<;:;T<=&KGT>>T&#-0",)/"?/-#,012-/),3&\89]:;<;:;T<^4-0",)/"&("/C.@/-#,0I"-#0"&FQ&c2(2&0#1)",0&ES&4Q:_5&68&T&QU&:&V&WSN6RF5&XYcfKY&H;T;T;T:M?RZS&_"/#9.']_FSF_8R?TG>HT&H;T;T;T:L#-0",#/"&-"$*'/#0"&7#+*-),=&789]T;HG;YM;T=&KGM:><&#-0",)/"&2J)$#0"&@#-3"&0*P*,)/&\89]T;HG;YM;T^!"#$%&'#$()*(%&+),'&("-'*-.&/-#,012-/),3&456#0"&789&:;<;:;:=&:;<;:;>=&9211"-?/-#,0@2'#/),3&6?/.1"&:;<;:;A=&\Q/)3(#/"@@#&#*-#,/)#'#&bgAO:?T^C"#B.&("/#@&"II@*J&1*(1%&9+'4&I#()@.&KGLLML&9*7N=O437N=&0.0/"(&("(P-#,"&1-2/"),&9*04C"#B.&("/#@&"II@*J&1*(1%&9+'4&I#()@.&KGLH:Y&C"#B.?("/#@&"J12-/"-%&c_8&I#()@.1*/#/)B"&'C"#B.&("/#@&P),$),3&1-2/"),&KGLHT:&'211"-&'C#1"-2,"C"#B.&("/#@&0",02-&0)3,#@&/-#,0$*'/)2,&C)0/)$),"&e),#0"&KGL<AA&/d2?'2(12,",/&0.0/"(%&F(1R&I#()@.%&e),#0"&9*0Q&\89]H;L;T:;:^456#0"%&8T?8H&/.1"]'211"-?/-#,0@2'#/),3&6?/.1"&456#0"]C"#B.&("/#@&/-#,0@2'#/),3&456#0"&KGT>::&9*H[?"J12-/),3&456#0"&\89]:;<;:;A^'+'4&'2P#@/?+),'?'#$()*(&-"0)0/#,'"&'#/)2,&"II@*J&0.0/"(&/-#,0("(P-#,"&1-2/"),&KGLH:Y&C"#B.?("/#@&"J12-/"-%&c_8&I#()@.C"#B.&("/#@&/-#,0@2'#/),3&6?/.1"&456#0"&KGT>::&9*H[?"J12-/),3&456#0"&\89]:;<;:;A^#'-8&#-0",)/"&"II@*J&1*(1&49R:&KG::H>&/-#,012-/"-%&49R:&I#()@.(#3,"0)*(&'2P#@/&/-#,012-/&1-2/"),&92-4&KG:HMA&("/#@&)2,&/-#,012-/"-%&_N5&I#()@.C"#B.&("/#@&/-#,012-/O$"/2J)I)'#/)2,&1-2/"),c"#B.?("/#@?#002')#/"$&$2(#),&7S?/"-(),*0=&("(P-#,"?P2*,$"$&'./2'C-2("&P)23","0)0&'.'D?@)e"&$2(#),/d2&'2(12,",/&C"#B.&("/#@&-"012,0"&/-#,0'-)1/)2,#@&-"3*@#/2-%&F(1R&I#()@.%&'211"-&-"0)0/#,'"&1C201C#/"&-"3*@2,&-"3*@#/2-&9*0RB32$1A%81)('A1,-H+)"1)($13214 12/#00)*(O'211"-?/-#,012-/),3&456#0"&789]:;<;:;TG=&KGT>AT&c[OK[?"J'C#,3),3&\89]:;<;:;TG^'2P#@/&4f9&/-#,012-/"-&456#0"&0*P*,)/&KGHGG<&'2P#@/O,)'e"@&/-#,012-/&0.0/"(&456?P),$),3&1-2/"),B"(%*1(23)'+&,-.(2'A%*1(23)'14 _*@/)'211"-&2J)$#0"1*/#/)B"&#-0",)/"&Q?#$",20.@("/C.@/-#,0I"-#0"&789]H;T;T;T:L=>#$13)%*1(23),-C$/%)%0/3='4 C"#B.&("/#@&/-#,0@2'#/),3&6?/.1"&456#0"&KGT>:A&9$H[OD,H[?"J12-/),3&456#0"1*/#/)B"&("/#@&'#/)2,&/-#,012-/"-I3#2%8#')1(313,->#')%($132134 C"#B.&("/#@&"II@*J&1*(1%&9+'4&I#()@.&KGLLML&9*7N=O437N=&0.0/"(&("(P-#,"&1-2/"),&9*04	  	   151	  APPENDIX B 	   1. Workflow for Processing Pyrotag Sequences in QIIME and Making OTU-table.  To start, following files are needed: 1. fasta file with all sequences (these must have the barcode included) 2. qual file with the sequence quality (same length as the fasta file) 3. a mapping file with information about each sample (including metadata)  First step to check the mapping file: • check_id_map.py -m mapping.txt –o mapping_output   Then split the sequences according to barcodes to the samples they belong to: • split_libraries.py -m mapping.txt -f pyrotag_sequences.fna -q pyrotag_sequences.qual -o split_library_output –L 500 -b 10  Then pick OTU and make an OTU-table: • pick_otus.py –m usearch -i split_library_output/seqs.fna -o otus/ –s 0.97 –db_filepath gold.fa –word_length 64 • pick_rep_set.py –i out/seqs_otu.txt –f split_library_output/seqs.fna –m most_abundant –o rep_set.fna • align_seqs.py -i rep_set.fna –t $HOME/core_set_aligned.fasta.imputed -o rep_set_aligned.fasta  • assign_taxonomy.py –i rep_set.fna –m blast –t 97_Silve_111_taxa_map.txt –r 97_Silva_111_rep_set.fasta • make_otu_table.py -i seqs_otus.txt -t tax_assignments.txt -o otu_table.biom -e rep_set_aligned/rep_set_failures.fasta • filter_otus_from_otu_table.py –I out_table.biom –o out_table_filtered.biom –n 2% removes singletons 	  	   152	  • print_biom_table_summary.py –I out_table.biom –o out_table_summary.txt • biom convert –I otu_table.biom –o out_table.txt –b –header-key=taxonomy • summarize_taxa_through_plots.py -i otu_table.biom –m mapping.txt –o taxa summary • filter_alignment.py –i rep_set_aligned.fasta –o rep_set_aligned_pfiltered.fata • make_phylogeny.py -i rep_set_aligned_pfiltered.fasta -o otu_rep_set.tre • make_otu_heatmap.py -i otu_table.biom -o otus/OTU_heatmap  2. Network of Co-occurrence Where all Nodes in the Network are OTUs and Lines (Edges) are Pearson Correlation Coefficient Value.  1- OTU table preparation • Sequences were clustered in to OTUs based on 97% (for species) or 95% (for genus) homology  (Detailed description for making OTU-table can be found in Appendix B.1)  • Only those OTUs with more than five sequences were considered in the following analysis  2- Calculation of Pearson correlation coefficient and p-value The formula for the correlation coefficient (r) is:  ! = !! − !!! !! − !!!!!!! ! − 1   x and y are variables, s is standard deviation and n is sample size. r always ranges in value from +1 to −1, where 1 is total positive correlation, 0 is no correlation, and −1 is total negative correlation.  • The Script used to calculate all Pearson correlation coefficient (r) and p-values for each pair of OTUs is:  $./enterotypes_graph.pl -i OUT_table.txt -o Output_file.txt -max_negative_correl -0.8 -min_positive_correl 0.8 -Label_file mapping.txt -min_count 50 -max_p_value 0.05  	  	   153	  -i: input file: OTU_table.txt -o: output file: Output_file.txt -max_negative_correl -0.8 & -min_positive_correl 0.8: valid interaction event was considered to be a robust correlation if the Pearson correlation coefficient was either equal or greater than 0.8 or −0.8  -max_p_value 0.05: statistically significant (p-value equal or smaller than 0.05) -min_count 50: cutoff for minimum number of non-zero columns for an OTU to be used; 50% of the total columns  • It produces two files 1- node.txt and 2- edge.txt  3- Visualization of the network in Cytoscape  • Install and open Cytoscape • File -> Import -> Network from file Select: edge.txt file. You’ll see a Preview of the table   	  	   154	  • Click: Show Text File Import Options • Click: Transfer first line as attribute names • Select: Source Interaction = Column 1 ( what is listed in the “from” column). Sources will be OTUs • Select: Target Interaction = Column 2  (what is listed in the “to” column). Targets will be the OTUs • Click the headers for all other columns to import them (they will turn blue) • Now you should see your network.  In the next step, import attributes for each of our nodes from node.txt file File -> Import -> node from Table • Select: node.txt file • Click: Show Text File Import Options • Click: Transfer first line as attribute name  • Click: Import As a result, you should get successful import message.  3.1. Formatting lines (edges) in the network  Changing the Layout format when the lines length are proportional to the value of  r  • In the upper tool bar window, Select: Layout -> Edge-Weighted Spring Embedded -> All Nodes -> CORRELATION 	  	   155	    Changing the line colors connect OTUs • In the edge.txt file add a column with the title of “type of interaction” then group samples based on r vale (Positive or negative) • Click: VizMapper in the Cytoscape • Select: Edge Color -> type of interaction  • Mapping Type: Discrete Mapping • Change the positive correlations to blue, and the negative correlations to red 	   3.2. Coloring and re-sizing nodes in the network  Change the size of nodes based on number of sequences in each OTU.  • Click: VizMapper in the Cytoscape • Select: Node Size. Can be “read count”   	  	   156	  Change the color of nodes by “phylogenetic affiliation”  • In the node.txt file create a column that group nodes based on “Phylum”, “Class”, “Order” or etc. • Click: VizMapper in the Cytoscape • Click: Node Color • Click: select a value!  Select: Phylum, Class, Order or etc. • Click: select a mapping type! • Select: Discrete You should now see the phylogenetic groups pop-up just below Node Color. The nodes in this network consist of OTUs.   • Click: Control, then click on the “…” next to it This will bring up a Pick a Color window – let’s make the Desulfobacterales blue. Click: OK Repeat this to make nodes different color based on phylogenetic affiliation  3. Network of Co-occurrence Where Nodes (OTUs) are Colored, Resized or Sorted Based on the Value of Betweenness Centrality or Closeness Centrality.  After steps 1 (OUT-table preparation) and step 2 (Calculation of Pearson correlation coefficient)  3- Calculation of value of betweenness centrality and closeness centrality for each node  The betweenness centrality of a node !  is given by the expression:  !" ! = !!"(!)!!"!!!!!   where !!" is the total number of shortest paths from node s to node t and !!"(!) is the number of those paths that pass through !.  	  	   157	  The closeness centrality of a node !  is given by the expression:  !!   ! =    1!(!, !)!!!  where !   !, !  is the shortest distance between node ! to node ! .  • The Script used to calculate all centrality indices (BC and CC) for each OTU is:  • $./compute_centrality.py --edges edges.txt --nodes nodes.txt  Repeat all steps for visualization of network in Cytoscape with following adjustments: In node.txt file: • Add a column “Betweenness centrality (BC)” and insert BC values nodes  • Add a column “Closeness centrality (BC)” and insert BC values nodes  • In step “Coloring and re-sizing nodes in the network” For node size for keystones select: Betweenness centrality For node color for keystones select: Closeness centrality                 	  	   158	  Alternative method for Calculation of value of BC and CC for each node  To analyze the network in Cytoscape Select: Tools -> Network Analyzer -> Network Analysis -> Analyze Network    A result panel window opens Press visualize parameters A window opens that allow you to adjust the color and size of nodes based of their value of BC or CC.       	  	   159	  	    • In the upper tool bar window, Select: Layout -> Attribute Circle Layout -> Betweenness Centrality • In the upper tool bar window, Select: Layout -> Attribute Circle Layout -> Closeness Centrality    	  	   160	  	  	  4. Bipartite Network in Which OTU-nodes are Connected via Lines (Edges) to Bioreactor-Nodes in Which their Sequences were Found (Shows shared or bioreactor specific OTUs).  After steps 1 (OUT-table preparation) • Run the QIIME script: make_otu_network.py -i otu_table.biom -m map.txt -o otu_network It generates two files: real_edge_table.txt and real node file real_node_table.txt  OTUs and bioreactors are designated as two types of nodes in a bipartite network in which OTU-nodes are connected via edges to bioreactor-nodes in which their sequences were found.  The real_edge_table.txt contains the columns “from”, “to”, “eweight” and “consensus_lin”,. The edges (lines connecting the nodes) will span “from” the bioreactor_ids “to” the otu_ids listed in the OTU table.  	  	   161	    Repeat all steps for visualization of network in Cytoscape with following adjustments:  • Select: Source Interaction = Column 1 ( what is listed in the “from” column). The sources will be the OTUs • Select: Target Interaction = Column 2  (what is listed in the “to” column). The targets will be the bioreactors   The visual output of this analysis is a clustering of bioreactors according to their shared OTUs. bioreactors that share more OTUs cluster closer together.   • Changing the Layout format based on e-weight.  • In the upper tool bar window, Select: Layout -> Cytoscape Layouts -> Edge-Weighted Spring Embedded -> e-weight  	  	   162	    E-weight is the number of sequences within an OTU that were found in a particular bioreactor.   5. Workflow for Making Phylogenetic Trees  1. Use filter_fasta.py in Qiime to select the representative sequences for the tree based on your list of OTUs that you want to include on the tree. 2. Gather a list of the GenBank accession numbers for the nearest neighbours that you want to include on the tree, including closely related cultured species and environmental clones. 3. Get the fasta files for these by batch downloading them using Batch Entrez on the NCBI site. 4. Create an Excel spreadsheet using these fasta files to get their accession numbers into one column and their names into another column. 5. Customize the names and save the Excel file as a tab-delimited-text file: Keep the original Excel file for easy modification of the leaf names if needed. 6. Remove all punctuation from the accession numbers and from the custom names i.e. remove | , . ; 	  	   163	  : [ ] ( ).  7. Remove all punctuation from the fasta files i.e. remove | , . ; : [ ] ( ).  8. Replace spaces in the custom names with underscore. 9. Combine the fasta files into one big file with all of the fasta sequences. 10. Align the sequences in this fasta file using MUSCLE with the -clw option for output in clustal format as a .aln file. 11. Convert the aligned file into Phylip format using the website:  sing.ei.uvigo.es/alter. 12. Build the tree with phyML using 100 bootstraps. 13. Open the tree in Figtree and then save it as a NEXUS tree file. 14. Replace the accession numbers on the NEXUS tree file with the custom names using my program mytest2.pl. 15. Import into iToL together with the data for number of reads per OTU. * There are several other softwares to construct and visualize the phylogenetic tree. One example is Bosque. It is a multi-platform software that performs standard phylogenetic analysis either remotely or locally on servers. Bosque offers a powerful environment that integrates all the steps of phylogenetic analyses (e.g. sequence alignments and graphical visualization and editing of trees). If you need to add some sequences for closely related cultured species or environmental clones that you’ve identified from Blastn to Genbank, then you’ll have to trim these first. The Genbank sequences are partial or full length 16S rRNA that are longer than pyrotag reads. To include them in a tree, they must be trimmed to the same region as pyrotags (i.e. the V6-V8 region). To do this align the Genbank sequences to the full length one for E. coli and then trim them to the 920-1400base region (base numbering according to E. coli). Then they need to be reversed and complemented using any on-line tool.  Then they will be ready to include with pyrotag reads for aligning and tree building.  

Cite

Citation Scheme:

        

Citations by CSL (citeproc-js)

Usage Statistics

Share

Embed

Customize your widget with the following options, then copy and paste the code below into the HTML of your page to embed this item in your website.
                        
                            <div id="ubcOpenCollectionsWidgetDisplay">
                            <script id="ubcOpenCollectionsWidget"
                            src="{[{embed.src}]}"
                            data-item="{[{embed.item}]}"
                            data-collection="{[{embed.collection}]}"
                            data-metadata="{[{embed.showMetadata}]}"
                            data-width="{[{embed.width}]}"
                            async >
                            </script>
                            </div>
                        
                    
IIIF logo Our image viewer uses the IIIF 2.0 standard. To load this item in other compatible viewers, use this url:
http://iiif.library.ubc.ca/presentation/dsp.24.1-0165837/manifest

Comment

Related Items