@prefix vivo: . @prefix edm: . @prefix ns0: . @prefix dcterms: . @prefix skos: . vivo:departmentOrSchool "Applied Science, Faculty of"@en, "Chemical and Biological Engineering, Department of"@en ; edm:dataProvider "DSpace"@en ; ns0:degreeCampus "UBCV"@en ; dcterms:creator "Khoshnoodi, Maryam"@en ; dcterms:issued "2014-09-25T19:36:15Z"@en, "2014"@en ; vivo:relatedDegree "Doctor of Philosophy - PhD"@en ; ns0:degreeGrantor "University of British Columbia"@en ; dcterms:description """Biogeochemical cycling of arsenic and speculation on mechanisms of arsenic removal are interest in the environmental remediation of contaminated sites. In the present study, combination of metagenomic molecular biology techniques with mineralogical analyses were used to study a biochemical reactor (BCR) that was successfully removing arsenic, zinc, copper and cadmium. First the metal and mineralogical content of the BCR solids was investigated. X-ray diffraction (XRD) and automated quantitative evaluation of minerals by scanning electron microscopy (QEMSCAN) were used for mineralogical characterization. Analysis indicated that sulfates and sulfides were the predominant types of Zn and As minerals formed in the BCR. Arsenic minerals were detected as sulfides (arsenopyrite, tennantite), arsenates(wihelmkleinite), oxides (unknown zinc arsenic oxides) and zincarsenic sulfides, which showed evidence of metal adsorption on the surfaces of other solids such as silicates. Energy-dispersive X-ray spectroscopy verified that arsenic was associated with iron, zinc and sometimes cadmium as arsenopyrite-type minerals. Using a SSU rRNA survey of the site, the following taxa were correlated with high metal content: Bacteroidetes, Synergistaceae, Victivallales, methanogens (Methanocorpusculum, Methanospirillum, Methanosarcina) and new phyla such as VadinHA17, M2PB4-65, candidate division WS6, RF3 and TM6. Next, enrichment culturing and arsenic chemical speciation monitoring were performed to assess potential for arsenic species transformations in the BCR. Most predominant groups in the As(III) and As(V) media, were Simplicispira (β-proteobacterium) and Sedimentibacter (Firmicutes), respectively. Chemical arsenic speciation monitoring of the enrichments suggested that arsenite oxidation and arsenate reduction occurred. These genera were not previously reported for arsenic transformation. Finally, functional metagenomic workflow was applied to study arsenic resistance genes. Functional screening and end-sequencing of large insert fosmid libraries demonstrated that arsenic(V) resistance genes were taxonomically widespread and different class of arsenic resistance genes related to periplasmic arsenate reduction, arsenite efflux, bioaccumulation (phosphate, metal transporters) and arsenite oxidation were present. Fewer genes were associated with dissimilatory arsenate reduction and arsenic volatilization mechanisms. Methanomicrobia were predominant in the BCR and identification of methanogen-related arsenic resistance genes indicated that methanogens potentially played a role in arsenic removal inside the BCR."""@en ; edm:aggregatedCHO "https://circle.library.ubc.ca/rest/handle/2429/50430?expand=metadata"@en ; skos:note """MICROBES INVOLVED INARSENIC REMOVAL IN PASSIVETREATMENT SYSTEMSbyMaryam KhoshnoodiB.Sc., University of Tehran, 2004M.Sc., University of Tehran, 2008A THESIS SUBMITTED IN PARTIAL FULFILLMENT OFTHE REQUIREMENTS FOR THE DEGREE OFDOCTOR OF PHILOSOPHYinTHE FACULTY OF GRADUATE AND POSTDOCTORAL STUDIES(Chemical and Biological Engineering)THE UNIVERSITY OF BRITISH COLUMBIA(Vancouver)September 2014c© Maryam Khoshnoodi 2014AbstractBiogeochemical cycling of arsenic and speculation on mechanisms of arsenicremoval are interest in the environmental remediation of contaminated sites.In the present study, combination of metagenomic molecular biology tech-niques with mineralogical analyses were used to study a biochemical reactor(BCR) that was successfully removing arsenic, zinc, copper and cadmium.First the metal and mineralogical content of the BCR solids was inves-tigated. X-ray diffraction (XRD) and automated quantitative evaluationof minerals by scanning electron microscopy (QEMSCAN) were used formineralogical characterization. Analysis indicated that sulfates and sulfideswere the predominant types of Zn and As minerals formed in the BCR.Arsenic minerals were detected as sulfides (arsenopyrite, tennantite), ar-senates (wihelmkleinite), oxides (unknown zinc arsenic oxides) and zinc-arsenic sulfides, which showed evidence of metal adsorption on the surfacesof other solids such as silicates. Energy-dispersive X-ray spectroscopy ver-ified that arsenic was associated with iron, zinc and sometimes cadmiumas arsenopyrite-type minerals. Using a SSU rRNA survey of the site, thefollowing taxa were correlated with high metal content: Bacteroidetes, Syn-ergistaceae, Victivallales, methanogens (Methanocorpusculum, Methanospir-illum, Methanosarcina) and new phyla such as VadinHA17, M2PB4-65, can-didate division WS6, RF3 and TM6.Next, enrichment culturing and arsenic chemical speciation monitor-ing were performed to assess potential for arsenic species transformationsin the BCR. Most predominant groups in the As(III) and As(V) media,were Simplicispira (β-proteobacterium) and Sedimentibacter (Firmicutes),respectively. Chemical arsenic speciation monitoring of the enrichmentssuggested that arsenite oxidation and arsenate reduction occurred. Thesegenera were not previously reported for arsenic transformation.Finally, functional metagenomic workflow was applied to study arsenicresistance genes. Functional screening and end-sequencing of large insertiifosmid libraries demonstrated that arsenic(V) resistance genes were taxo-nomically widespread and different class of arsenic resistance genes relatedto periplasmic arsenate reduction, arsenite efflux, bioaccumulation (phos-phate, metal transporters) and arsenite oxidation were present. Fewer geneswere associated with dissimilatory arsenate reduction and arsenic volatiliza-tion mechanisms. Methanomicrobia were predominant in the BCR andidentification of methanogen-related arsenic resistance genes indicated thatmethanogens potentially played a role in arsenic removal inside the BCR.iiiPrefaceAll manuscript-based chapters of this thesis were co-authored with the re-search supervisor Prof. Susan Baldwin. My contribution was the data collec-tion, analysis of the data, and the writing of the manuscripts. Prof. Bald-win contributed largely to the analysis and the discussion of results, andthe revision of drafts. Two of the chapters in this thesis were manuscriptsco-authored with authors in addition to Prof. Baldwin. This work includedfield studies; the assistance received from Al Mattes of NatureWorks Reme-diation Corporation during samples collection. Dr. Stephane Brienne andJeanine Powell assisted for field sampling permission. Prof. Baldwin, JanaSchmidtova, Dr. Marcus Taupp and myself were involved in data collectionat the site.Chapter 3 is based on the work conducted in different laboratories.Prof. Gregory Dipple (Professor of Department of Earth, Ocean and At-mospheric Sciences at UBC) contributed in geochemical and mineralogicalguidance. Dr. Gareth Chalmers advised in mineralogical tools and tech-niques. I was in charge of searching for the equipment and the laboratories,preparation of the samples, performing the tests, collecting and analysis thedata. The work was conducted in the following centers with support of thefollowing names: Mario Beaudoin from Advanced Materials and Process En-gineering Lab (AMPEL)- Nanofabrication Facility (ANF), UBC; Dr. MatiRaudsepp and Jenny Lai from Electron Microbeam / X-Ray Diffraction Fa-cility, Department of Earth, Ocean and atmospheric Sciences, UBC.In addition, services were received from Calgary Rock and MaterialsServices Inc. (thin section preparation), Geological Survey of Canada - Cal-gary Division (RockEval 6 analysis), ALS Environmental (samples chem-istry analysis) and electron microscopy study was performed during a visitto Kamloops at ALS, Metallurgy. Xiaowen (Wendy) Ma, manager of min-eralogy department at the time, contributed for data collection.A version of Chapter 3 has been published [71]. It was co-authoredivwith Prof. Gregory Dipple and Prof. Susan Baldwin. I conducted all thetesting and wrote most of the manuscript. Dr. Gregory Dipple assisted forcompiling the research results and revising the manuscript. Prof. Baldwincontributed to the discussion and paper revision.Chapter 4 is based on work conducted in Chemical and Biological Engi-neering Department laboratory. I was responsible for setting up the bacte-rial enrichment cultures and performing the tests with analytical technique(HPLC-UV). Maryam Rezadehbashi and Pariisa Mirjafari assisted for se-quence submission to Centre d′innovation, Genome Quebec et UniversiteMcGill (Montreal, Quebec).A version of Chapter 4 has been prepared for publication. I conductedall the testings and wrote the manuscript. Prof. Baldwin was the co-authorand largely contributed for the bioinformatic analysis and revision of thedraft.Chapter 5 is based on the collaboration with Dr. Steven Hallam’s lab-oratory in the Department of Microbiology and Immunology at UBC. Thedevelopment of experimental design for the metagenomic screening was un-der supervision of Dr. Steven Hallam. I was responsible for all the exper-iments, tests, data collection and analysis. Dr. Marcus Taupp contributedin metagenomic library production. Keith Medwis assisted significantly fordata analysis and interpretation. Jacky Chan (co-op student) helped forwriting the programming algorithm for data analysis. Dr. Kishori Kon-war and Niels Hanson provided help in performing integrated analysis ofsequences (MetaPathways software). Material presented in Chapter 5 hasbeen prepared for publication.vTable of ContentsAbstract . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . iiPreface . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . ivTable of Contents . . . . . . . . . . . . . . . . . . . . . . . . . . . . viList of Tables . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . xList of Figures . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . xiiList of Abbreviations . . . . . . . . . . . . . . . . . . . . . . . . . xvAcknowledgments . . . . . . . . . . . . . . . . . . . . . . . . . . . xviiDedication . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . xviii1 Introduction . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 11.1 Arsenic General Description . . . . . . . . . . . . . . . . . . 11.2 Passive Treatment System . . . . . . . . . . . . . . . . . . . 31.3 The Removal of Arsenic in Biochemical Reactors . . . . . . . 31.3.1 Geochemistry of BCRs, Eh Versus pH Diagrams . . . 41.3.2 Arsenic Complexation with Iron in Solutions . . . . . 51.3.3 Arsenic Sulfide Precipitation . . . . . . . . . . . . . . 71.4 Microbe-Metal Interactions in Mining-Affected Environments 91.5 Microbial Responses to Arsenic in Mining-Affected Environ-ments . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 111.5.1 Dissimilatory Arsenate-Respiration/Reduction . . . . 121.5.2 Chemoautotrophic or Heterotrophic Arsenite Oxida-tion . . . . . . . . . . . . . . . . . . . . . . . . . . . 151.5.3 Other Arsenic Resistance Mechanisms . . . . . . . . 171.6 Project Study Site - Background Information . . . . . . . . . 181.7 Thesis Objectives . . . . . . . . . . . . . . . . . . . . . . . . 201.8 Thesis Layout . . . . . . . . . . . . . . . . . . . . . . . . . . 21vi2 Study Site, Sample Collection and Analysis . . . . . . . . . 242.1 Scope . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 242.2 Project Study Site . . . . . . . . . . . . . . . . . . . . . . . . 242.3 Sample Removal . . . . . . . . . . . . . . . . . . . . . . . . . 252.4 Analyses . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 342.5 Molecular Biology Analysis . . . . . . . . . . . . . . . . . . . 342.6 Chemistry Analysis . . . . . . . . . . . . . . . . . . . . . . . 342.6.1 Porewater Analysis . . . . . . . . . . . . . . . . . . . 342.6.2 Core Solid Analysis . . . . . . . . . . . . . . . . . . . 352.7 Mineralogy Analysis . . . . . . . . . . . . . . . . . . . . . . . 363 Mineralogical Study of the Biochemical Reactor, Identifica-tion of Arsenic and Zinc Minerals . . . . . . . . . . . . . . . 373.1 Scope . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 373.2 Experimental Section . . . . . . . . . . . . . . . . . . . . . . 383.2.1 Sampling Site Characterization . . . . . . . . . . . . 383.2.2 Core Solids Chemistry . . . . . . . . . . . . . . . . . 393.2.3 Pretreatment of Core Samples: Organic Matter (OM)Removal . . . . . . . . . . . . . . . . . . . . . . . . . 393.2.4 Mineralogical Analysis . . . . . . . . . . . . . . . . . 403.2.5 Geochemical Modeling . . . . . . . . . . . . . . . . . 413.2.6 Correlation with Microbial Communities . . . . . . . 413.3 Results . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 423.3.1 Geochemical Predictions . . . . . . . . . . . . . . . . 423.3.2 Mineral Species Based on XRD Analysis . . . . . . . 433.3.3 QEMSCAN and X-Ray Analysis Observations . . . . 453.3.4 BCR Microbial Groups Possibly Associated with MetalRemoval . . . . . . . . . . . . . . . . . . . . . . . . . 493.4 Discussion . . . . . . . . . . . . . . . . . . . . . . . . . . . . 513.4.1 Evidence for Biotic Mineralization . . . . . . . . . . . 513.4.2 Other Possible Mechanisms for Mineral Formation inthe BCR . . . . . . . . . . . . . . . . . . . . . . . . . 533.5 Conclusions . . . . . . . . . . . . . . . . . . . . . . . . . . . . 544 Enrichment of Arsenic Resistant Microbes from the Bio-chemical Reactor . . . . . . . . . . . . . . . . . . . . . . . . . . 604.1 Scope . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 604.2 Material and Methods . . . . . . . . . . . . . . . . . . . . . . 614.2.1 Sampling Site . . . . . . . . . . . . . . . . . . . . . . 614.2.2 Enrichment Culturing . . . . . . . . . . . . . . . . . . 61vii4.2.3 Analytical Techniques . . . . . . . . . . . . . . . . . . 624.2.4 DNA Extraction, Polymerase Chain Reaction Ampli-fication and Phylogenetic Analysis . . . . . . . . . . . 644.3 Results . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 654.3.1 Biochemical Reactor Environment . . . . . . . . . . . 654.3.2 Enrichment Culture Characterization: Taxonomic Pro-file . . . . . . . . . . . . . . . . . . . . . . . . . . . . 664.3.3 Detection of Arsenic Transformation . . . . . . . . . 684.3.4 Prevalence of Putative CAO and Arsenate ReducingMicrobes in the BCR . . . . . . . . . . . . . . . . . . 694.4 Discussion . . . . . . . . . . . . . . . . . . . . . . . . . . . . 695 Functional Metagenomic Screening of Arsenic . . . . . . . 785.1 Scope . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 785.2 Experimental Section . . . . . . . . . . . . . . . . . . . . . . 795.3 Sampling Site, Fosmid Library Construction . . . . . . . . . 795.4 Sequence- Based Fosmid Library Screening . . . . . . . . . . 805.4.1 Gene Finding and Open Reading Frame Prediction . 805.4.2 Phylogenetic Analysis . . . . . . . . . . . . . . . . . . 805.5 High Throughput Functional Screening . . . . . . . . . . . . 805.5.1 Phenotype Characterization of Resistant Clones . . . 815.5.2 Full Fosmid Sequencing . . . . . . . . . . . . . . . . . 815.5.3 Phylogenetic Analysis . . . . . . . . . . . . . . . . . . 825.6 Results . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 835.7 Discussion . . . . . . . . . . . . . . . . . . . . . . . . . . . . 865.8 Concluding Remarks . . . . . . . . . . . . . . . . . . . . . . 916 Conclusions . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 956.1 Overall Conclusions . . . . . . . . . . . . . . . . . . . . . . . 956.2 Significance of the Research . . . . . . . . . . . . . . . . . . . 1046.3 Contributions to the Field . . . . . . . . . . . . . . . . . . . 1066.4 Limitations of the Research and Recommendations for FutureInvestigations . . . . . . . . . . . . . . . . . . . . . . . . . . 108Bibliography . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 110AppendicesA Supplementary Data Chapter 1 . . . . . . . . . . . . . . . . 130viiiB Supplementary Data Chapter 2 . . . . . . . . . . . . . . . . . 140C Supplementary Data Chapter 3 . . . . . . . . . . . . . . . . . 156D Supplementary Data Chapter 4 - Part 1 . . . . . . . . . . . 166E Supplementary Data Chapter 4 - Part 2 . . . . . . . . . . . 189F Supplementary Data Chapter 4 - Part 3 . . . . . . . . . . . 197F.1 Program . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 197F.1.1 Sample Data . . . . . . . . . . . . . . . . . . . . . . . 199G Experimental and Analytical Protocols . . . . . . . . . . . . 201G.1 Preparation of Pre-reduced Media for the Enrichment Culture 201G.2 MIC Determination in L.B Agar Medium . . . . . . . . . . . 202G.3 MIC Test/Fosmids Growth Rate Determination in L.B Broth 202G.4 Analytical Methods . . . . . . . . . . . . . . . . . . . . . . . 203ixList of Tables1.1 Common arsenic-bearing minerals . . . . . . . . . . . . . . . 21.2 Mechanisms of arsenic- microbe interactions, pathway distri-bution . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 131.3 As(V)-reducing bacteria . . . . . . . . . . . . . . . . . . . . . 141.4 As(III)-oxidizing bacteria . . . . . . . . . . . . . . . . . . . . 162.1 Recovered core solid samples and analysis . . . . . . . . . . . 293.1 Pore water and bulk solids chemistry measured in the coresections collected from the BCR . . . . . . . . . . . . . . . . 433.2 Mineralogical composition of the BCR samples . . . . . . . . 484.1 Enrichment cultures sequenced . . . . . . . . . . . . . . . . . 63A.1 Chemical analysis of the BCR core solid samples . . . . . . . 130A.2 Carbon analysis of the BCR core solid samples . . . . . . . . 133A.3 Carbon analysis of the BCR core solid samples-2 . . . . . . . 134A.4 Parameters definition of carbon analysis . . . . . . . . . . . . 137B.1 Total arsenic, iron and sulfur in the BCR influent and effluent 141B.2 Total arsenic, iron and sulfur in the BCR lysimeter . . . . . . 142B.3 Total arsenic, iron and sulfur in the borehole porewater . . . 143B.4 Aqueous and solid species equilibrium for Eh-pH diagram . . 144B.5 Results of geochemical modeling of the As-O-H-S-Fe system . 147B.6 Predominance of high metal concentration associated micro-bial groups . . . . . . . . . . . . . . . . . . . . . . . . . . . . 148B.7 SRB-related and iron-reducing bacteria enrichment culturerelated OTUs found in the BCR . . . . . . . . . . . . . . . . 151C.1 Accession numbers for arsenic enrichment culture sequence . 157C.2 Pore water and solid chemistry from the bore holes used totake samples for the enrichment cultures . . . . . . . . . . . . 158C.3 Arsenic speciation and HPLC analysis-As(III) medium . . . 162xC.4 Arsenic speciation and HPLC analysis-As(V) medium . . . . 163D.1 MIC analysis for EPI300 (expression host) . . . . . . . . . . 166D.2 Comparison of fosmid growth rates at different As(V) cons. . 168D.3 Summary of arsenic resistance genes identified in completefosmid sequences . . . . . . . . . . . . . . . . . . . . . . . . . 172D.4 Summary of arsenic resistance genes identified in fosmid in-sert sequence ends . . . . . . . . . . . . . . . . . . . . . . . . 173E.1 MLtreeMap taxonomy analysis . . . . . . . . . . . . . . . . . 190F.1 Example used for the written program to combine fosmid endsequences-tblastn entries . . . . . . . . . . . . . . . . . . . . 198xiList of Figures1.1 Eh-pH diagram for As-Fe-S-O-H . . . . . . . . . . . . . . . . 51.2 Mechanisms of metal - microbe interactions . . . . . . . . . . 101.3 Mechanisms of arsenic - microbe interactions . . . . . . . . . 131.4 Conceptual model for biogeochemical processes in BRC . . . 202.1 Schematic and photograph of the BCR . . . . . . . . . . . . . 262.2 Schematic sampling map of the BCR . . . . . . . . . . . . . . 272.3 Sample collection from Trail biochemical reactor . . . . . . . 283.1 Eh-pH diagram for As-Fe-S-O-H at 25 ◦C and 1 atm pressure 443.2 BSE images of zinc-arsenic sulfides on the surface of amphi-bole and feldspar . . . . . . . . . . . . . . . . . . . . . . . . . 473.3 BSE images and EDX spectra of some other mineral particlesfrom BCR . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 563.4 BSE images and EDX spectra of some other mineral particlesfrom BCR . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 573.5 PCA plots with As and Zn concentrations . . . . . . . . . . 583.6 PCA plots with highly correlated microbial groups . . . . . . 594.1 Most highly represented phyla in enrichment cultures . . . . . 674.2 Heatmap of enrichment culture samples . . . . . . . . . . . . 764.3 Prevalence of putative CAO and arsenate reducing microbesin the BCR . . . . . . . . . . . . . . . . . . . . . . . . . . . . 775.1 Summary of arsenic resistance genes identified from MG2 . . 925.2 Circos representation of complete fosmid sequences- GM2 . . 935.3 Megan 5 analysis of the full fosmid sequences . . . . . . . . . 946.1 Schematic diagram for As removal mechanism in the BCR . . 97A.1 Schematic of dissimilatory reduction process . . . . . . . . . . 138A.2 Arsenite oxidase (aox) gene cluster . . . . . . . . . . . . . . . 139xiiB.1 Schematic and photograph of the BCR . . . . . . . . . . . . . 149B.2 Eh-pH diagrams for As-O-H-S-Fe system . . . . . . . . . . . . 150C.1 As calibration curves and example chromatograms . . . . . . 159C.2 Rarefaction statistics (PD whole tree and Chao1) for diversityof enrichment samples . . . . . . . . . . . . . . . . . . . . . . 160C.3 Distribution of pyrotag reads among genera . . . . . . . . . . 161C.4 Prevalence of putative CAO and DARPs in the BCR . . . . . 164C.5 Prevalence of putative CAO and DARPs in the BCR-legend . 165D.1 Phenotype characterization of the resistant fosmid clones-1 . 174D.2 Phenotype characterization of the resistant fosmid clones-2 . 175D.3 Summary of arsenic resistance genes identified from MG1 . . 176D.4 Summary of arsenic resistance genes identified from MG3 . . 177D.5 Circos representation of all complete fosmid sequences . . . . 178D.6 Circos representation of complete fosmid sequences, MG1 -blast comparison . . . . . . . . . . . . . . . . . . . . . . . . . 179D.7 Circos representation of aox positive fosmids- MG2 . . . . . . 180D.8 Circos representation of aox positive fosmids- MG1 . . . . . 181D.9 Circos representation of aox positive fosmids- MG3 . . . . . 182D.10 Circos representation of complete fosmid sequences- MG3 . . 183D.11 Circos representation of complete fosmid sequences- MG1 . . 184D.12 Megan 5 analysis of the full fosmid sequences - phylum level . 185D.13 Megan 5 analysis of the full fosmid sequences - species level . 186D.14 Putative ArsC identified from Full fosmids and fosmids endsand its homology to ArsC . . . . . . . . . . . . . . . . . . . . 187D.15 Putative ArsC identified from Full fosmids and fosmids endsand its homology to ArsC - combined fosmid ends . . . . . . 188E.1 Potential pathways identified by metagenomic functional -KEGG profile . . . . . . . . . . . . . . . . . . . . . . . . . . . 189E.2 KEGG functional annotation of ORFs - metal transportation 193E.3 ABC transporters - KEGG functional annotation . . . . . . 194E.4 RefSeq functional annotation of ORFs - metal transportation 195E.5 Enzymes potentially involved in sulfur cycle, KEGG analysis 196F.1 Visual representation of fosmid end sequence analysis . . . . 199F.2 Fosmid end sequence analysis . . . . . . . . . . . . . . . . . . 200G.1 Analytical Method 1 . . . . . . . . . . . . . . . . . . . . . . . 204G.2 Analytical Method 2 . . . . . . . . . . . . . . . . . . . . . . . 205xiiiG.3 Analytical Method 3 . . . . . . . . . . . . . . . . . . . . . . . 206G.4 Analytical Method 4 . . . . . . . . . . . . . . . . . . . . . . . 207G.5 Analytical Method 5 . . . . . . . . . . . . . . . . . . . . . . . 208xivList of AbbreviationsATP-adenosine triphosphateARM-arsenic resistance microbesBCR-biochemical reactorBSE-backscattered electronsbp-base pairBMA-bulk mineral analysisCW-constructed wetlandCAO-chemoautotrophic arsenite oxidizersDO-disolved oxigenDARP-dissimilatory arsenate respiring prokaryotesDMA(V)-dimethylarsenateDMA(III)-dimethylarseniteDMAA-dimethylarsine acid◦C-degree CelsiusEh-reduction potentialEDX-energy dispersive X-rayEPS-extracellular polymeric substanceg-gramHPLC-high performance liquid chromatographyHAO-heterotrophic arsenite oxidizerskg-kilogramkb-kilobasekDa-kilodaltonLTA- low temperature ashingmin-minutesml-millilitermM-milimolarMG-metagenomic libraryµL- microliterMMA(As(V))-monomethyl arsenateMMA(As(III))-monomethyl arseniteMINC-mineral carbonxvnm-nanometerNCBI-national center for biotechnology informationOTU-operational taxonomic unitORF-open reading frameOM-organic matterOD-optical densityORP-oxygen reduction potentialPMA-particle mineral analysisPCA-principal coordinate analysisPC-pyrolyzed carbonQEMSCAN - quantitative evaluation of minerals by scanning electron mi-croscopyRF-radio frequencyrRNA-ribosomal ribonucleic acidRT-residence time (seconds)SRB-sulfate reducing bacteriaSEM-scanning electron microscopeT-temperatureTOC-total organic carbonTMAO- trimethylarsine oxideW-wattXRD- X-ray diffractionxviAcknowledgmentsFirst and foremost, I am immensely thankful for ongoing advice, guidance,understanding, support, motivation and patience of my adviser Prof. SusanBaldwin.I am so grateful to Dr. Steven Hallam for accommodating me in his Laband his advice. Thanks to Keith Medwis who provided valuable discussionsgenerating valuable ideas. Dr. Marcus Taupp for his guidance in beginningstage of my research. Eugene Kuatsjah, Elena Zaikova, Sam Khairandish,Melanie Scofield, Payal Sipahimalani for assisting with lab work and molecu-lar techniques. Charles Howes, Kishori Konwar, Niels Hanson for supportingmy work using bioinformatics tools.I am so thankful to Prof. Gregory Dipple, Head, Depart. of Earth, Oceanand Atmospheric Sciences at UBC, encourage and educate me in the field ofgeo- microbiology. His colleagues Dr. Gareth Chalmers, Jenny Lai providedinvaluable insights, discussions to steer my research. Thanks to them fordedicating their time, thoughts and energy to this work.I am thankful to Prof. Majid Mohseni, for reviewing my research pro-posal, and providing valuable comments to strengthen it. I am also thankfulfor Dr. Marcello Veiga and Dr. Christopher Teaf reviews of this dissertationand subsequent examinations. Thanks to Dr. Louise Creagh for providingaccess to her biotechnology laboratory and using the equipments.Thanks to Ekaterina Vassilenko, Parissa Mirjafari, Gaurav Subedi fortheir friendship and providing their time and minds to tackling issues, Icould not have solved easily otherwise.I acknowledge the funding contributed by Teck Mining Company, Na-tureWorks Remediation Corporation, Genome British Columbia and theNatural Sciences and Engineering Research Council of Canada (NSERC).xviiDedicationThis work is dedicated to my father, Prof. Mohammad Khoshnoodi, whohad considerable impact on me to pursue higher education in applied micro-biology and biotechnology. He passed on me: a love of reading and interestin our environment.xviiiChapter 1Introduction1.1 Arsenic General DescriptionArsenic is relatively ubiquitous in the earth’s crust (0.0001%) where itis associated with igneous and sedimentary rocks [31]. Natural phenom-ena such as weathering, biological activity, and volcanic activity lead towidespread distribution of arsenic species among different subsurface layers,some of which have been used for drinking water with disastrous conse-quences [31][117].Anthropogenic activities, such as mining, mineral smelting, fossil fuelcombustion, wood preservation and pesticide application can increase theconcentration of arsenic species in the aqueous environment and cause ar-senic pollution [144]. In seawater, the concentration of arsenic is usually lessthan 2 µg L−1, the mean arsenic concentrations in sediments range from 5to 3000 mg kg−1 and the levels of arsenic in unpolluted surface water andgroundwater vary typically from 1-10 µg L−1 [42]. The World Health Or-ganization (WHO) has set a guideline of 10 µg L−1 as the drinking waterstandard [42]. However, arsenic imposes significant risks to the health ofpeople in many different countries (for example arsenic conc. (µg L−1) re-ported; Western USA 1-48,000 (Arseniferous pyrite mining area), Romania1-176, India ≤50-23,080) [144]. Arsenic levels reported in untreated waterat Manitoba, Canada [163] ranged from 65-70 µg L−1.Arsenic exists in four oxidation states: As(-III), As(0), As(III) andAs(V). Elemental (native) form occurs rarely and As(V) and As(III) arethe two most abundant forms of inorganic arsenic in aqueous systems. Inan aerobic environment, arsenate (As(V) as H3AsO4, H2AsO−4 , HAsO2−4 ,AsO3−4 ) is dominant, whereas arsenite (As(III) as H3AsO3, H2AsO−3 andHAsO2−3 ) is more prevalent under anaerobic conditions. Arsenate possessesthe ability to co-precipitate with or absorb into minerals such as iron oxyhy-droxides and alumina. Arsenite absorb or co-precipitate with metal sulfides.Indeed, arsenite adsorbs less strongly and to fewer minerals, which makes it11.1. Arsenic General Descriptionthe more mobile and reactive oxyanion [117]. Arsine (AsH3) is consideredto be the most acutely toxic form of arsenic, followed by the arsenites, ar-senates and organic arsenic compounds[42].Arsenic is mostly introduced into water through the dissolution of rocks(e.g minerals, ores) and from industrial effluents, including mining wastes.It is commonly associated with the ores of metals like copper, lead, gold,zinc, silver [135]. Some common arsenic minerals that are frequently as-sociated with base metal ores and concentrates are presented in Table 1.1.Inevitably, some of the arsenic contained in these minerals become availableduring metallurgical processes and metal extraction.Table 1.1: Common arsenic-bearing minerals [158],[63]Type Mineral FormulaArsenides Nickeline NiAs2Niccolite NiAsSafflorite (Co,Fe,Ni)As2Skutterudite CoAs3Sulfides Arsenopyrite FeAsSArsenical pyrite Fe(As, S)2Orpiment As2S3Realgar As4S4Sulfosaltes Cobaltite CoAsSEnargite Cu3AsS4Tennantite (Cu,Fe)12As4S13Gersdorffite NiAsSOxides Arsenolite As2O3Claudite As2O3Arsenates Olivenite Cu2(AsO4)OHScorodite FeAsO4H2OArseniosiderite Ca2Fe3O2(AsO4)3.3H2OKankite FeAsO43.5H2OParasymplesite Fe3(AsO4)28H2OPharmacosiderite K[Fe4(OH)4(AsO4)3]6.5H2OKo¨ttigite Zn3(AsO4)28H2OThe effluents from treatment facilities, seepage and surface drainage ef-fluent undergo strict governmental regulations. Therefore, mining effluent21.2. Passive Treatment Systemcontaminated by arsenic or other metals must be collected and treated toremove metals before being discharged into environment.1.2 Passive Treatment SystemPassive treatment systems, which are natural or engineered systems thatmake use of naturally occurring geochemical and biological processes fordetoxification of water and soils, have received more attention than anyother strategy for arsenic remediation. This is mainly because of relativelylow construction and operating costs as well as their potential for effectiveremoval of dissolved metals and metalloids [109]. Passive treatment systemsare usually part of wetland systems.Constructed wetland systems using organic waste substrates (e.g., resid-uals) and limestone or lime were first used to treat acid rock drainage atcoal mines [166]. They have since been adopted for treatment of many othermine-influenced waters [57][80][131][95]. Most effective for removal of metalssuch as arsenic are subsurface flow anaerobic systems currently referred toas biochemical reactors (BCRs) [17][33].1.3 The Removal of Arsenic in BiochemicalReactorsIn biochemical reactors, study of metal removal mechanisms, particularlyarsenic, is challenging. Removal under aerobic conditions by sorption to orco-precipitation with iron oxyhydroxide and possibly by scorodite formation,as well as removal under anaerobic condition as sulfides, are reported for pas-sive treatment of arsenic [109] [56]. However, in order to accurately assessthe various metal removal mechanisms, geochemical modeling, solid phasespeciation analysis, as well as mineralogical and microbial characterization,are needed. Furthermore, this will enable the elucidation of fundamentalphysiochemical and microbiological reactions that occur in anaerobic biore-actors with complex natural organic substrates. Environmental genomicsthat provide access to an entire pool of genetic information can additionallyhelp to understand arsenic removal from BCRs.31.3. The Removal of Arsenic in Biochemical Reactors1.3.1 Geochemistry of BCRs, Eh Versus pH DiagramsFor chemical removal mechanisms, geochemical modeling can to a certaindegree predict the minerals formation based on the oxidation-reduction po-tential (Eh), pH and chemistry of the BCR environment [161]. Variousversions of Eh-pH diagrams for arsenic have previously been constructed[115], [87],[161],[79],[34]. The Eh-pH diagrams slightly differ from each othermainly due to differences in thermodynamic properties of solids, aqueousspecies and addition of certain arsenic species. The most recent arsenicEh-pH diagrams at 25◦C and 1 bar are presented in Lu et al. [87] (Fig-ure 1.1). In the system of As-Fe-S-H-O, orpiment, realgar, arsenopyrite,scorodite and native arsenic are solid phases that could be predicted withgeochemical modeling. The total concentration of Fe, As and S can effectthe stability fields of the solids. In general, under oxidative conditions,scorodite is present and under reducing conditions, which applies to theBCR, orpiment, realgar and arsenopyrite are possible arsenic-bearing min-erals (Figure 1.1). Sulfide minerals, such as orpiment (As2S3) can be stableunder slightly reducing conditions, but mainly at acidic pHs, whereas, re-alger (As2S2) is the more stable mineral formed under strongly reducingconditions [161]. Furthermore, in the presence of Fe and under reducingand alkaline conditions, arsenopyrite (FeAsS) can occur [161][87]. However,prediction of mineral formation based on thermodynamic modeling is ofteninaccurate in BCRs because bacterial processes can greatly alter the chem-ical environment and result in the formation of minerals outside the bulkthermodynamic geochemical stability zones. In addition, amorphous miner-als are more dominantly formed in the environment than crystalline formsand therefore log K cannot be determined accurately. Therefore, systems inBCR may be dynamic and not at thermodynamic equilibrium [11],[109].Geochemical speciation and plotting can be performed with several pro-grams and databases. For example, PhreePlot (http://www.phreeplot.org/)is a program used for making geochemical plots and PHREEQC, which isavailable from the US Geological Survey (USGS), can be used for fitting datawith geochemical models. Hydra and Medusa chemical equilibria comput-ing software packages (http://www.kth.se/che/medusa/) [126] [127] is an-other example. Thermodynamic databases such as WATEQ4F, MINTEQA1are available for these programs. It should be noted that thermodynamicdatabases may require evaluation as new data become available for arsenicspecies [79][87].41.3. The Removal of Arsenic in Biochemical ReactorsHAsO4-AsO4-H3AsO4ScoroditeH2AsO3-H3AsO3OrpimentRealgarNative As ArsenopyriteH2AsO4-Eh(volts)pH1.510.50-0.5-10 4 6 108 12 142Figure 1.1: Eh-pH diagram for As-Fe-S-O-H at 25◦C and 1 atm pressure.Total S, total Fe, total As set at 10−3 M, 10−6 M and 10−3 M respectively.Grey shaded areas denotes the solid phase [87].1.3.2 Arsenic Complexation with Iron in SolutionsIn mine water treatment operations, soluble arsenic is most commonly re-moved by sorption to or coprecipitation with iron oxyhydroxides (HFO) un-der oxidative conditions. As(V) can be precipitated from process solutionswith Fe(III) to form crystalline ferric arsenate (scorodite)[79]. This processoccurs at relatively high concentrations of Fe(III) and As(V)(≥0.001M) andlow pH (about pH=2)[138]. At lower concentrations of As(V) and higherFe(III) concentrations the co-precipitation of arsenic with ferrioxyhydroxidesis predominant [138]. For example, in one laboratory-scale study related toburied mine tailings in northern Saskatchewan, Canada, precipitation of90-98% of As(V) as scorodite at pH 2-3 was obtained and adsorption of re-maining As(V) by a precipitated FO phase at higher pH was reported [79].51.3. The Removal of Arsenic in Biochemical ReactorsScorodite is the least soluble arsenate phase in many mine tailing systemsand in this case study, formation and solubility of scorodite was evaluated.Although Fe(III)-As(V) complexes can be formed under oxic conditions,oxygen alone is not able to oxidize the arsenic and therefore oxidation ofarsenite is affected by the presence of the ions such as sulfide and ferrous(Fe(II)). The hypothesis is that ferrous ions are oxidized by dissolved oxygenor other oxidants and then the ferric ions (Fe(III)) oxidizes arsenite resultingin quick oxidation and co-precipitation of As by Fe(III) [151]. Understand-ing the solubility and stability of Fe(III)-As(V) complex is important interms of controlling arsenic release to pore water [79]. Many studies focuson re-evaluating of solubility of this complex [136][79][74].Reduction of As(V) and reductive dissolution of ironhydroxides, eitherbiologicaly or with transition from aerobic to anaerobic conditions couldresult in arsenic desorption. As(V) reduction appears to be the domi-nant process controlling As release [156]. In a desorption study of ar-senic from ferrihydrite-, goethite-, and hematite-coated sand, using mu-tants of Shewanella sp. ANA-3, capable of Fe(III)- and/or As(V)-reduction,columns with As(V) reduction, showed higher desorption of arsenic com-pared to columns with Fe(III) reduction alone. As(III) is also desorbed morerapidly from iron hydrooxides despite being adsorbed to a greater extentthan As(V)[156]. In addition, dissimilatory Fe(III)-reducing microorgan-isms might contribute to this release, as they are thought to be the majorcatalysts for Fe(III) reduction in anoxic environments [84]. Nevertheless,bacterial formation of Fe oxy-hydroxides is possible under anoxic conditionswith nitrate as electron acceptor for iron oxidizing bacteria, and one suchmicrobe was isolated from the BCR used for this study [96]. The microbialroles in formation or dissolution of the minerals are discussed in anothersection.In most cases in the mining industry arsenic, is precipitated by com-plexion of As(V) with Fe(III). The problems associated with the presentpractices of arsenic disposal are related to generation of large volumes offerric oxyhydroxide wastes and the long-term stability of the waste [34].Scorodite and amorphous ferric arsenate are the two forms that can formunder oxidative condition, whereas resolubilization of iron and release of ar-senic into pore water usually result under anaerobic conditions. In otherwords, arsenic remains adsorbed to iron oxy-hydroxides (FO) as long as61.3. The Removal of Arsenic in Biochemical Reactorsthe environment is sufficiently oxidized. High input of organic carbon tothe systems, such as residuals used in chemical bioreactors, usually resultsin oxygen depletion and lowering of redox potentials. Sulfide precipitationbecomes an important mechanism to stabilize the arsenic in an anoxic envi-ronment Section 1.3.3.1.3.3 Arsenic Sulfide PrecipitationAn alternative process for removing soluble arsenic from the environment isthe selective precipitation of As as sulfides. This technology has advantagesover the arsenic co-precipitation with iron hydroxides: less waste produc-tion and lower lime and iron consumption. Following are a few examples ofbench scale column sulfide reactors.In one example, precipitation of trivalent arsenic sulfide (As2S3) wasachieved in a fixed-film bioreactor under sulfate-reducing conditions at lowpH [7]. The bioreactor was fed continuously with As(V), glycerol and/orhydrogen at pH values between 2.7 and 5. The overall As removal rate withglycerol was in the range of 1.5-2 mg L−1 h−1, and it reached 2.5 mg L−1 h−1with hydrogen. Although the final concentration of arsenic was lower usingH2 feed, glycerol gave better kinetic results. It has been suggested that thegreatest efficiency for arsenic removal was under a limited sulfate-reducingcondition because this avoids the formation of soluble thioarsenic species(e.g. AsS(OH)(SH)−, AsS−2 , HAsS2). Both glycerol and H2 are electrondonors; however, with H2, the sulfate-reducing rate is greatly stimulatedand results in arsenic species release from the solid phase [7].In another study, microbial sulfate reduction and subsequent precipita-tion of arsenic and other metals were demonstrated in an upflow anaerobicpacked bed reactor filled with silica sand. A mixed population of SRB wasused and the columns were able to remove 77.5% of As [68]. However, astudy of sulfate reducing chemical bioreactors (BCR) operating in down-ward flow, indicated a low fraction of metals removed as sulfides (up to 15%of total metals recovered in the reactive mixtures) [110]. In fact, organicmatter and presence of (oxy)hydroxides resulted in competitive metal ad-sorption/complexation and precipitation in this system [110].A field application of this technology has been documented in a con-structed wetland (CW) at the Wheal Jane mine in the UK. The wetland71.3. The Removal of Arsenic in Biochemical Reactorswas designed to remove dissolved metals and metalloids from mine wasteleachate and results indicated sulfate removal in the range of 3−38% andmetal removal to below detection limits. An anaerobic bioreactor containingSRB was effective to remove metals (Zn, Cu, Cd) by sulfide precipitation[165].Numerous other studies aside from these few examples have been pub-lished, showing effective application of this method. The drawback of thisapplication is that the precipitated arsenic sulfide is unstable or unsuitablefor storage because arsenic sulfide minerals may become soluble again as aresult of pH changes. The reported solubility of orpiment, As(III) sulfide, is28.5 mg/L in the pH range 1-4 [137]. Therefore, the sulfate reducing processcan be efficient in the low pH conditions. In SRB-columns studies when thepH was kept in the low range, good results were reported; Recovery of nickelsulfide was observed in a SRB bioreactor working continuously, which waskept at pH 5 and fed with hydrogen [14]. Moreover, selective precipitation ofcopper or zinc was occurred at pH values between 2 and 5 with continuouslyfed sulphidogenic bioreactor [108].Metal sulfide precipitation is expected in sulfate-reducing bioreactors.Sulfate-reducing bacteria (SRB) oxidize organic matter using sulfate as anelectron acceptor and the metabolic product hydrogen sulfide forms spar-ingly soluble precipitate with metals. SRB also produce bicarbonate, whichcan neutralize water acidity [110]. In a recent study, SRB enrichment cul-ture from gold mine tailings sediment revealed formation of FeS precipi-tate which was later identified as mixture of pyrite, marcasite, mackinawiteand griegite, and was found to be associated with Deltaproteobacteria andFirmicutes-related sequences[167].Microbial groups other than SRB play an important roles and can effecton metal removal. A number of studies have indicated both microbial sulfateand iron reduction control the mobility and bioavailability of metals [50], [49],[123]. Fe(III)-reducing bacteria closely related to Acidiphilium spp., werefound in acidic (high sulfate) and Fe-rich sediments [76]. Geobacteraceaespp. identified from enrichment cultures indicated that bacterial Fe(III) re-duction can occur in mining-impacted lake sediments. Geochemical datafrom the sediment showed iron sulfide and magnetite as common productsof bacterial Fe(III) hydroxide reduction[32]. Iron reducing Shewanella algawere also shown to incorporate metals in siderite, FeCO3 precipitation, re-sulting in significant solid phase capture of metals [120]. In the next section81.4. Microbe-Metal Interactions in Mining-Affected Environmentsmicrobial roles in metal removal will be discussed further.1.4 Microbe-Metal Interactions inMining-Affected EnvironmentsBecause of the toxic potential and the widespread presence of metals inthe environment, microbes have developed unique and diverse ways of deal-ing with toxic metals. Some microorganisms have mechanisms to sequesterand immobilize metals, whereas others enhance metal solubility in the en-vironment. Microbial resistance mechanisms influence the fate of metals inthe environment, which is applicable to metal mining and remediation ofmetal-contaminated sites. This was the focus of various research and stud-ies [58],[12],[117]. In Figure 1.2, the major types of metal-microbe interac-tions are summarized. This schematic diagram was derived from differentstudies [85],[119][63].Some resistance mechanisms are plasmid encoded and are specific forparticular metals while others represent general conferred resistance to avariety of metals. First of all, in general mechanisms of metal resistance,binding of metals to extracellular materials results in immobilization of themetal and prevents its entry into the cell. Another term for this process isbiosorption. Figure 1.2 describes metabolism-independent sorption of heavymetals to biomass. Both living and dead biomass are capable of biosorptionand many ligands are involved in metal binding [99]. An example of extra-cellular binding sites are slime layers and exopolymers (EPSs). Exopoly-mer functional groups are generally negatively charged and consequentlybinding of cation metals is pH-dependent. Biosorption characteristics werestudied for arsenic-resistant Arthrobacter sp. [124]. Other examples ofextracellular molecules produced microbially are siderophores [51] and bio-surfactants [107]. These molecules are capable of complexing with metalsnonspecifically and thereby play a role in reducing metal toxicity.Secondly, resistance mechanisms that are dependent on a specific metalinvolve many intracellular metal resistance mechanisms. One of the mecha-nisms is metal sequestration (bioaccumulation in Figure 1.2) by metalloth-ioneins or similar proteins. Their production is induced by the presence ofa metal, and they serve a primary role in metal detoxification [16]. Metalbinding by metallothionein-like proteins can result in intracellular accumula-91.4. Microbe-Metal Interactions in Mining-Affected EnvironmentsFigure 1.2: Mechanisms of metal - microbe interactions, adapted from[85],[119],[63].tions, an example of metabolism-dependent bioaccumulation. Metal effluxsystems are other energy-dependent mechanisms that remove metal ionsfrom the cells. Toxic ions are effectively pumped out via active transportmediated by membrane bound enzymes such as ATPase pump or diffusion[113]. Metal methylation is also considered as a metal resistance mechanism.Metal volatilization facilitates metal diffusion away from the cell, effectivelyreducing metal toxicity. This process is important for metal removal fromcontaminated surface water, sewage and soils. Furthermore, this mechanismis carried out by pathways involving either: S-adenosylmethionine, methyl-cobalamin or N-methyltetrahydrofolate [93] (Figure 1.2).In addition to the mechanisms outlined above, biologically-controlledmineralization [12],[143] plays an important role in environmental reme-101.5. Microbial Responses to Arsenic in Mining-Affected Environmentsdiation of metals by microorganisms (Figure 1.2). Some microorganismspromote nucleation and growth of minerals at specific locations on the cellor EPS. The metals can precipitate as carbonates and hydroxides, wherebyproton influx antiport to metal efflux results in localized alkalinization atthe cell surface. Minerals formed by this mechanism are usually character-ized with well ordered crystals (e.g., not amorphous). Alternatively, metalscan precipitate with enzymatic generated ligands, e.g., sulfides [48] or phos-phates [102]. Beveridge et al. [13] also suggested two step mechanisms forthe development of metal precipitation in Bacillus subtilis. Microbial cellsin this process, which is also referred to microbially enhanced chemosorp-tion, first precipitate biomineral of one metal (priming deposit) and thenthe priming deposit acts as a nucleation site for the subsequent depositionof the target metal[143]. The priming deposit is based on stoichiometricinteraction of the metal ions with reactive chemical groups on the bacterialsurface [13][143] or initially made by the sulfide or phosphate biomineraliza-tion routes described earlier [102]. Bacteria can also produce minerals insidethe cell. Magnetotactic bacteria are well known examples of such groups thatproduce magnetosomes (contain crystals of magnetite (Fe3O4)). It has alsobeen shown that magnetite(Fe3O3), greigite (Fe3S4) and pyrrhotite(Fe7S8)can be formed in non-magnetotactic bacteria [30]. For example Shewanellaputrefaciens can produce interacellular iron oxides by dissimilatory reduc-tion of iron [53].Enzymatically catalyzed biotransformation [117] is another importantmechanism that allows microorganisms to mobilize metals, metalloids byreduction and oxidation processes. In other words, microorganisms can af-fect metal speciations and change solubility of metals. For instance, theaddition of organic electron donors to promote microbial reduction of sol-uble U (VI) to less soluble U (IV) has shown to be an effective methodfor immobilizing uranium in uranium-contaminated groundwater [85], Fig-ure 1.2. Arsenic biotransformation is a major response to arsenic stress inmany organisms.1.5 Microbial Responses to Arsenic inMining-Affected EnvironmentsArsenate (As(V)) and arsenite (As(III)) are two soluble inorganic forms ofarsenic, present widely in the mining-affected environment. Although these111.5. Microbial Responses to Arsenic in Mining-Affected Environmentsare potential toxic components, some bacteria can gain energy by oxidiz-ing As(III) or respiring As(V), while other bacteria can respond to arsenicvia resistance mechanisms. Organisms associated with arsenic are taxo-nomically diverse and metabolically versatile and can be categorized intofour groups; arsenate-resistance microbes (ARM), dissimilatory arsenate-respiring prokaryotes (DARP), chemoautotrophic arsenite oxidizers (CAOs)and heterotrophic arsenite oxidizers (HAOs)[59]. Figure 1.3 and Table 1.2summarize possible microbial interactions with arsenic [119][63].1.5.1 Dissimilatory Arsenate-Respiration/ReductionDARPs utilize As(V) as a terminal electron acceptor and belong to severalphylogenetic groups, including the γ-, δ- and - Proteobacteria, Firmicutes,Chrysiogenes and Crenoarchaea [117]. Arsenate reduction is coupled withoxidation of organic compounds (e.g., lactate, acetate, formate, pyruvate,citrate) which are notably present in passive treatment systems as interme-diates in metabolic cycles. Some of the arsenate reducers can also respireon sulfate, which shows their metabolic diversity. Desulfotomaculum au-ripigmentum and Desulfomicrobium strain Ben-RB are examples of thesegroups[112]. It should be noted that arsenic reduction can also take placeinside the cell in order to detoxify arsenic by converting arsenate to arsenite,which is eliminated from the cell via the arsenic efflux pumps (see below).Examples of some of isolated As(V)-reducing bacteria are presented in Ta-ble 1.3.Respiratory arsenate reductase proteins (ArrA and ArrB) have beenidentified in DARPs (Biotransformation- Figure 1.3). arrA encodes a 95.2kDa protein and binds to molybdenum cofactor (dimethyl sulfoxide reduc-tase DMSO family enzymes) and arrB is encodes a 25.7 kDa iron-sulfurprotein[148],[59].121.5. Microbial Responses to Arsenic in Mining-Affected EnvironmentsFigure 1.3: Mechanisms of arsenic - microbe interactions, adaptedfrom [85][119][63]Table 1.2: Mechanisms of arsenic- microbe interactions, pathway distribu-tionNo. As resistance pathways List of genes comments1 Bioaccumulation pit,pst,glpF,phn phosphate transporters2 Chr,Znu,Cbi,Fhu,Nham4423,4424metal transporters3 Periplasmic arsenate reduction arsC, acr2p4 arsR,arsD,arsT,arsX,arsHregulators for ars operon5 Arsenite efflux pumps arsA,arsB,acr36 Volatilization arsM7 Biosorption NA8 Biomineralization NA9 Arsenite oxidation aroA,asoA,aoxB/aroB,asoB,aoxA/aoxSaox operon10 Dissimilatory arsenate reduction arrA,arrB arr operon131.5. Microbial Responses to Arsenic in Mining-Affected EnvironmentsThe gene arrA has been used as a biomarker for detecting the presenceof dissimilatory As(V)-reducing bacteria in a diverse of subsurface environ-ments [91][52]. Indeed, DARPs are able to reduce solid phase As(V) (arsenicadsorbed to iron oxyhydroxide minerals) and mobilize arsenic. Therefore,this gene and its product can be used to identify the presence and activity ofAs(V)-respiring bacteria in arsenic-contaminated iron-rich sediments[91][52].Table 1.3: As(V)-reducing bacteriaOrganism Phylogeny RefCitrobacter sp. TSA-1 γ-Proteobacteria [59]Shewanella sp. ANA-3 γ-Proteobacteria [59]Chrysiogenes arsenatis str. BAL-1T Chrysiogenetes [90]Wolinella succinogenes -Proteobacteria [149]Sulfurospirillum deleyianum -Proteobacteria [59]Sulfurospirillum barnesii -Proteobacteria [59]Sulfurospirillum arsenophilum -Proteobacteria [150]Sulfurospirillum halorespirans -Proteobacteria [59]Sulfurospirillum multivorans -Proteobacteria [59]Clostridium sp. OhILAs Firmicutes [59]Geospirillum arsenophilus -Proteobacteria [59]MLMS-1 δ-Proteobacteria [59]Bacillus arsenicoselenatis Firmicutes [150]Bacillus selenitireducens Firmicutes [59],[150]Bacillus sp. JMM-4 Firmicutes [59]Bacillus sp. HT-1 Firmicutes [59]Bacillus macyaestr. JMM-4 Firmicutes [142]Desulfotomaculum auripigmentum Firmicutes [59]Desulfitobacterium frappieri Firmicutes [118]Desulfitobacterium hafniense GBFH Firmicutes [8]Desulfitobacterium hafniense JH1 Firmicutes [83]Deferribacter desulfuricans Deferribacteres [35]Desulfosporosinus sp.Y5 Firmicutes [149]Halanaerobiaceae bacterium SLAS Firmicutes [59]Aquificales str. HGM-K1 Aquificae [59]Thermus sp. HR13 Deinococcus-Thermus [59]Alkaliphilus metalliredigenes Firmicutes [149]Alkaliphilus oremlandii Firmicutes [46]141.5. Microbial Responses to Arsenic in Mining-Affected EnvironmentsAlthough the As(V) reductases share a common reaction substrate andproduct, they can be different in structure and function. In Shewanella sp.stain ANA-3, a periplasmic protein complex, ArrAB, accepts electrons de-livered by c-type cytochromes in the cytoplasmic membrane. Figure A.1indicates the schematic diagram for dissimilatory metal reduction. This is adirect enzymatic mechanism where bacteria utilize metals as terminal elec-tron acceptors and has been discussed in more detail for Geobacter strains[133],[23].Several anaerobic arsenate-respiring bacteria have also been isolatedfrom arsenic-contaminated mining environments. For example, α- protobac-teria such as Caulabacter, Sphingomonas, and Rhizobium-like bacteria wereisolated from contaminated mine tailing (abandoned copper smelter)[89].Chrysiogenes arsenatis str. BAL-1T and Bacillus macyae str. JMM-4 arearsenate respirers isolated from anoxic mud in a tailing pond[90][142].1.5.2 Chemoautotrophic or Heterotrophic ArseniteOxidationChemoautotrophs (CAOs) oxidize As(III) to As(V) using oxygen, nitrateor chlorate as electron acceptors. They obtain carbon from the fixation ofcarbon dioxide and energy from the oxidation of inorganic compounds. Thisis in contrast to heterotrophs (HAOs) that cannot fix carbon dioxide andinstead use organic carbon for making cell material[59]. Examples of some ofthe isolated As(III)-oxidizing bacteria and their metabolism are presented inTable 1.4. Most known species are classified as α-, β-, and γ-Proteobacteria,Deinococcus-Thermus, or Crenarchaeota.The arsenite oxidizing enzymes are heterodimers consisting of a largesubunit (90 kDa) with a molybdenum center and a (3Fe-4S) cluster: AroA,AsoA, and AoxB and a small subunit containing a Rieske-type (2Fe-2S):AroB, AsoB, and AoxA. The large subunit is similar to the DMSO (dimethylsulfoxide reductase) family; however, it is phylogenetically separate anddistinct from respiratory arsenate reductase (ArrA) or other proteins ofDMSO family, Nar/Nap (assimilatory and periplasmic nitrate reductases(Nas/Nap)), respiratory nitrate reductase (Nar) and DMSO reductase (Dor)[149]. The genes encoding the two subunits have been sequenced, but un-fortunately were given different names. Figure A.2 shows the organizationof the aox operon in few arsenic metabolizing microrganisms, with different151.5. Microbial Responses to Arsenic in Mining-Affected Environmentsannotation. In the recent study of Santini et al. [63] these two enzymesare referred to as Aio and Arx. In fact, these proteins which are locatedin the periplasm can oxidize arsenite to arsenate. In the next step, theelectrons are transferred to other proteins involved in the electron transportchain which results in the production of ATP and the reduction of oxygen,nitrate or chlorate (Biotransformation- Figure 1.3). In addition, aoxRS, atwo-component signal-transduction system was found upstream from aoxABin most of the Proteobacteria order (Figure A.2)[106]. Similar gene arrange-ment (aoxRSABC-moeA) has been identified in NT-26 and Agrobacteriumtumefaciens [140].Arsenate oxidizers have been isolated from arsenic rich environmentssuch as Sainte-marie-aux-Mines (France) [60] and arsenopyrite-containinggold mining environments (Australia)[141][117].Table 1.4: As(III)-oxidizing bacteriaOrganism Phylogeny Metabolism RefNT-2 α- Proteobacteria CAO [59]NT-3 α- Proteobacteria CAO [59]NT-5 α- Proteobacteria HAO [59]NT-4 α- Proteobacteria CAO/HAO [141]NT-14 β- Proteobacteria CAO/HAO [141]NT-6 β- Proteobacteria HAO [59]NT-25 α- Proteobacteria CAO [59]NT-26 α- Proteobacteria CAO [140]Hydrogenophaga sp. CL3 β- Proteobacteria HAO [59]NT-10 β- Proteobacteria HAO [141]BEN-4 β- Proteobacteria HAO [141]BEN-5 α- Proteobacteria CAO [141]Alcaligenes faecalis β- Proteobacteria HAO [59]Agrobacterium albertimagni α- Proteobacteria HAO [59]Agrobacterium tumefaciens α- Proteobacteria HAO [59]Thermus thermophilus str. HB8 Deinococcus-Thermus HAO [59]Thermus sp. str. HR13 Deinococcus-Thermus HAO [59]Thermus equaticus YT1 Deinococcus-Thermus HAO [59]GM1 β- Proteobacteria HAO [118]Leptothrix sp. str. S1.1 β- Proteobacteria HAO [8]Continued on next page161.5. Microbial Responses to Arsenic in Mining-Affected EnvironmentsTable 1.4 – continued from previous pageOrganism Phylogeny Metabolism RefRalstonia sp. str. 22 β- Proteobacteria HAO [83]Sinorhizobium sp. str. M14 α- Proteobacteria HAO [35]Sinorhizobium sp. DAO10 α- Proteobacteria CAO [59]Thiomonas arsenivorans β- Proteobacteria CAO [9]Thiomonas sp. str. 4As β- Proteobacteria CAO [38]Thiomonas sp. NO115 β- Proteobacteria CAO [59]Variovorax paradoxus β- Proteobacteria HAO [8]Acinetobacter junni γ- Proteobacteria HAO [27]Herminiimonas arsenicoxydans β- Proteobacteria CAO [59]Marinobacter sp. γ- Proteobacteria HAO [27]Azoarcus sp. DAO1 β- Proteobacteria CAO [59]MLHE-1 γ- Proteobacteria CAO [59]Bosea sp. WAO α- Proteobacteria CAO [59]Ancylobacter sp. OL1 α- Proteobacteria CAO [59]Thiobacillus sp. S1 β- Proteobacteria CAO [59]Sulfolobus tokodaii Crenarchaeota CAO [148]Aeropyrum Pernix Crenarchaeota [148]Nitrobacter hamburgensis β- Proteobacteria CAO [147],[148]Rhodoferax Ferrireducens β- Proteobacteria HAO [44],[148]1.5.3 Other Arsenic Resistance MechanismsThe most well-studied resistance mechanism in arsenate-resistance microbes(ARM) is the periplasmic arsenate reduction that occurs through the arsoperon [119],[149],[59]. This comprises the arsenate reductase (ArsC) andthe efflux pump (ArsB). As As(V) enters the cells, either through the phos-phate transport system (e.g Pit, Pst) or through diffusion, it is reduced toAs(III) inside the cell (i.e., in the cytoplasm) and As(III) is then pumpedout of the cell through ArsB coupled to ArsA-mediated ATP hydrolysis(Figure 1.3). As(III) can also be pumped directly via ArsB. ArsR reg-ulator and ArsH putative flavoprotein are also reported to be related tothis operon[106]. Indeed, novel arsenic resistant genes are being identifiedas part of the arsenic resistance operon. For example, arsTX encoding athioredoxin(TrX) system has been documented recently in Microbacteriumsp. A33 [1].It should be noted that arsenic resistant enzymes are also present in eu-171.6. Project Study Site - Background Informationkaryotic cells where they are referred to as Acr2p (arsenate reductase) andAcr3 (arsenite pump). Prokaryotic arsenate reductases (ArsC) have beencategorized into two groups: one family is typified by the E. coli plasmidR773 ArsC, that uses Grx and GSH (Glutathione) as reductants and thesecond family is represented by the Staphylococcus aureus plasmid pI258ArsC, that uses thioredoxin (Trx) as a reductant [105]. ArsC is an approx-imately 150-residue long arsenate reductase that uses reduced glutathione(GSH) to convert arsenate to arsenite with the active site containing a re-dox active cysteine residue. ArsC forms an active quaternary complex withGSH, arsenate, and glutaredoxin 1 (Grx1).The presence of arsenic resistance genes can serve as biomarkers for ar-senic contamination or presence in the environment, as well as being animportant ecotoxicological tool [47][25][3]. The acr3/arsB genes, which en-code arsenite pumps, have been tracked in the study of in situ uraniumbioremediation [52].On the other hand, some organisms detoxify arsenic by methylation(Volatilization- Figure 1.3). This produces monomethyl arsenate (MMA(V)),methylarsenite (MMA(III)), dimethylarsinate (DMA(V)), dimethylarsenite(DMA(III)) and trimethylarsine oxide (TMAO), as well as volatile arsines[149]. As(III) S-adenosylmethionine methyltransferase (arsM ) gene is re-sponsible for the removal of arsenic from the bacteria in the volatile arsinesform [149]. Moreover, multiple metal-efflux operons have also present inother ARMs. For instance, in Herminiimonas arsenicoxydans and Ralsto-nia metallidurans cadmium-zinc-cobalt-resistance operons (czc operons),copper-resistance operons and chromate-resistance operons were reported [106].1.6 Project Study Site - Background InformationThe BCR used in this study was the first in a series of two subsurfaceflow and three surface flow wetland treating arsenic containing seepage fromlandfilled waste in southern British Columbia, Canada [94][36]. The BCR1was reconstructed in June 2002 and demonstrated good performance in re-moving As and other metals (Zn, Cd). The BCR had flow rate of 20,000 Lper day and treated more than 19,100 m3 of contaminated water during a5-year period (2003-2007). The BCR removed more than 10,700 kg of As,Zn and sulfate at an average BCR influent concentrations (mg/L) of: As,181.6. Project Study Site - Background Information58.6; Zn, 51.9 and SO2−4 , 781.5 [96]. During a sixteen month sampling period(25 June 2008 to 2 October 2009), the average total concentrations of ar-senic (T/As), zinc (T/Zn) and sulfate in the BCR influent were ≈ 33.2, 71.1and 654.1 mg/L, respectively. The average total concentrations of arsenic(T/As) and zinc (T/Zn) in the effluent decreased to ≈ 8.8 and 34.3 mg/Lduring this time period, respectively, which indicated a high metal removalefficiency within the BCR1. The schematic diagram of the BCR1 is pre-sented in Figure 2.1 and features of the BCR, sampling methods as well asanalysis are described in Chapter 2.This British Columbia site was of particular interest to us due to itsunique features. Although several previous studies have been published, bi-ological and mineralogical analyses are limited. For microbiological analysis,Duncan et al. [36] performed PLFA (phospholipid fatty acids analysis) andindicated high bacterial populations in the system. DGGE (Denaturing GelGradient Electrophoresis)/PCR was also performed and demonstrated evi-dence of SRBs in the BCR. Preliminary analysis with SEM/EDS - scanningelectron microscope, with energy dispersive X-ray spectroscopy indicatedzinc and iron sulfides presence in the BCR samples [36]. In addition, anautotrophic iron oxidizer bacterium was isolated from this site using ironoxidizing media. The bacterium coupled iron oxidation with nitrate reduc-tion [96]. The characteristics of this BCR and the constructed wetland canbe found in [37],[95]. In another study, mineralogical analysis, XRD (X-raydiffraction) and XANES (X-ray absorption near edge structure) were usedin bench scale reactors as well as representative samples collected from theBCR to demonstrate As sulfide precipitation as a mechanism for As removal.Results indicated the potential presence of a mixture of As(III)-O and disor-dered As2S3 from the bench scale reactors. Moreover, it was observed thatzinc enhanced arsenic removal, a conclusion that resulted from reactors thatcontained zinc in comparison with reactors that did not. Representativefield samples were collected for six months in Mesh tubes that were packedwith biosolids and lowered into water sampling wells [66]. Notably, rep-resentative samples were not directly obtained from BCR matrix. A widesurvey in mineralogy and microbiology for the BCR has not been performedand the mechanisms for metal removal, as well as the microbial populationpresent in BCR, are unclear.191.7. Thesis ObjectivesFigure 1.4: Conceptual model for all possible biogeochemical processesfor As-transformation and removal in the BCR, Abbreviations: DMA,dimethyl arsin; TMA, trimethyl arsin; SS, soluble sulfides; DARP, dis-similtory arsenic-respiring prokaryotes; ARM, arsenate resistance microbes;CAO, chemoautotrophic arsenic oxidizers, IRB, iron reducing bacteria; IOB,iron oxidizing bacteria.1.7 Thesis ObjectivesThe overall objective of this work was to investigate metal removal mech-anisms in biochemical reactors, focusing on arsenic. The most commonlyused BCRs for metal removal are operated anaerobically. Microbial activ-ity within the organic-rich matrix creates alkalinity and reducing conditionsthat are conducive for metal precipitation, for example, metal sulfides. But,it is unknown if other microbial processes, particularly those involved in ar-senic transformation, also occur in BCRs that contribute to metal removal.A conceptual model for all possible biogeochemical processes for arsenic201.8. Thesis Layoutremoval in the BCR is given in Figure 1.4. Both biotic and abiotic mech-anisms contribute to arsenic transformation and removal from the BCR.One hypothesis suggests arsenic accumulation in the site by changing thevalence state, converting As(III) to As(V), which is less soluble and lessbioavailable. Arsenite oxidizers (CAOs) can immobilize arsenic by couplingthe reaction with intermediate molecules in metabolic cycles that act aselectron acceptors (e.g. nitrate), whereas arsenate reducers (ARM, DARP)activity can result in the release of arsenic. Furthermore, precipitation ofarsenite As(III) by biologically produced sulfides is an important mechanismfor arsenic removal; a hypothesis that is evident from the detection of sulfideminerals in the BCR. The geochemical processes inside the BCR could alsoresult in the formation of metal sulfides and As(V) co-precipitation withferrioxyhydroxides. Nevertheless, taxonomically diverse and metabolicallyversatile organisms present at the site could effectively remove arsenic fromthe system. For example, iron oxidizers provide ferric ion, which can facil-itate arsenite oxidation and precipitation. Other examples are As-resistantmicrobes that important in uptake of arsenic and accumulate arsenic insidethe cell.Therefore, with respect to metal-microbial interactions indicated in Fig-ure 1.3, the research objective was to elucidate the microbial mechanisms,to identify microbes, genes and minerals that are involved in arsenic removalfrom mine influenced water, by developing and applying environmental ge-nomic tools as well as advanced mineralogical techniques.The specific research objectives are :1. What arsenic bearing minerals exist at the BCR?2. What arsenic resistance microbes can be cultured from the BCR?3. What arsenic resistance genes can be found at BCR ?4. What arsenic functional genes exist at BCR ?5. What is the main mechanism for arsenic removal, and which microbesare potentially involved?1.8 Thesis LayoutThe previous objectives were addressed throughout the course of the re-search. This dissertation compiles the research results and discusses them211.8. Thesis Layoutin the form of three unique manuscripts, as given below. Not all the re-search results are presented in the manuscripts and have been included inthe appendixes, as Supplementary Data. Supplementary Data are assignedto specific Chapters. Following is a detailed description of how each of theobjectives has been met with respect to their presentation in the followingChapters.Chapter 3, contributes primarily to meeting objective 1. It reports onthe types of minerals present in the BCR and discusses arsenic bearing min-erals with an analysis of the mineralogical forms expected from biogenicmechanisms. An automated mineralogical method QEMSCAN (quantita-tive evaluation of minerals by scanning electron microscopy) was appliedto detect the low abundance arsenic-, zinc-, cadmium-, copper- containingminerals. Geochemical modeling of the BCR conditions was also performed.In addition, the correlation of these results with a taxonomic profile of mi-croorganisms identified in this BCR which may putatively be involved inbiomineralizaton, was discussed in this Chapter.Chapter 4, contributes primarily to meeting objective 2. It presents iden-tification of microbes from the BCR that were capable of growing in highconcentration of As(III) and As(V) media using enrichment techniques. Infact, the main objectives of enrichment cultures were to identify microor-ganisms capable of reducing or oxidizing arsenic compounds by monitoringarsenic species concentration over time. The results indicated taxonomicgroups particularly involved in arsenate reduction and arsenite oxidation inthe BCR. Further, a comparative analysis of a microbial profile inside theBCR ( [6] ) was reported that indicated abundance of enriched microorgan-isms.Chapter 5, contributes primarily to meeting objectives 3 and 4. In thisChapter, two metagenomic screening strategies were presented; (1) func-tional metagenome screening, in which clones of interest were screened andselected through the detection of heterologous expression of arsenic(V) re-sistance trait, based on growth in high arsenic concentration medium. Full-length large DNA fosmid inserts were sequenced and the growth rate ofAs(V) resistant clones was measured. In the second approach (2), the 700-800 base ends of all the large DNA fosmid inserts were screened for homol-ogy to known As resistance genes. From both of these approaches, arsenictransformation gene markers identified were found to be related to bioac-cumulation, volatilization, biotransformation and other biotic mechanisms221.8. Thesis Layout(Figure 1.3, Table 1.2) in response to arsenic. This Chapter demonstratedmetagenomic approach for identifying potential metal interactions in a sys-tematic way inside BCRs.The final or concluding Chapter summarizes the overall findings in termsof the potential metal-microbe interactions in this BCRs and the main mech-anism for arsenic removal. The conclusions drawn from this particular BCRand the application of this work to remediation of other arsenic contami-nated sites is discussed.23Chapter 2Study Site, SampleCollection and Analysis2.1 ScopeIn this Chapter, the BCR site used for this thesis work, the study design,samples collected and analyses performed are described. First, features ofthe BCR are given and then the methodology for sample collection is pre-sented. In addition, environmental variables that were measured at the siteduring sampling are described. A list of samples collected is presented. Thesamples were analyzed to obtain chemical and biological information. Thespecific methods used for analysis are described. The chemical and biologi-cal data of the samples, used in subsequent studies, is described in Chapters3, 4 and 5.2.2 Project Study SiteTrail in British Columbia, Canada, has been the site of continuous smeltingoperation since 1897. A treatment system was constructed to treat sea-page originating from a smelter waste leachate (capped 2002) and a historicAs storage facility, which previously flowed into a creak (i.e., Stoney Creakdrains into the Columbia River). The leachate contained high concentra-tion of As, Zn, Cd and So2−4 . Nature Works Remediation Corporation wascontracted by Teck Cominco in 1997 to design and build this treatment sys-tem. The engineered wetland consists of six cells: two anaerobic bioreactors(BCR1 and BCR2) followed by three horizontal sub surface flow (HSSF)cells and a pond wetland cell. BCR1 is 30m x 18m and contains 750 m3 ofsubstrate. BCR2 is 25m x 18m and contains 600 m3 substrate. HSSF1 andHSSF2 are 50 m2 and contain plants; dominantly Tripsicum dactyloides andCalamagrostis canadensis. HSSF3, with area of 300 m2, is planted with cat-tails (Typhala tifolica). BCR1 was constructed in 1997 and rebuilt in June2002 and is used for this study. BCR1 contains, pulp mill biosolids; a waste242.3. Sample Removalproduct of the pulp and paper industry (60 wt % of the starting material).More specifically, BCR matrix consisted of: 60% pulp mill biosolids, 35%sand and 5% manure along with limestone gravel. The schematic diagramof BCR1 is presented in Figure 2.1. Details on the construction are availablein other PhD theses of this site [36] [94].2.3 Sample RemovalSamples were collected from the biochemical reactor by drilling into thereactor matrix at three different locations and on three different days: 8July 2008, 21 April 2009, 20 October 2009. Schematic sampling map of thebiochemical reactor, given in Figure 2.2, shows the locations and boreholenumbers. For sample removal, a portable diamond drilling rig was used toallow the withdrawal of the cores inside 2.0 cm internal diameter poly vinylchloride (PVC) pipe. In the field, poly vinyl chloride (PVC) pipe containingthe organic sludge was cut into 30-40 cm length and then the ends werecapped and frozen with liquid N2. In the laboratory each core was cut into5 cm intervals, homogenized under liquid N2 and stored at -80◦C. Table 2.1shows the recovered core samples from the BCR and Figure 2.3 describesthe sample collection from the site. In Table 2.1, all the core samples withtheir origin, collection date, core length, core number, borehole number arepresented. Selected samples were analyzed for molecular biology, mineralogyand chemistry analysis. These analyses were referred to as DNA, Mineral-ogy and Chem, respectively (Table 2.1).252.3. Sample Removal - Mixture of 60% kraft pulp mill biosolids residuals, 35% sand and 5% manureupflowseepage inflow BCR outflowgeotextile linersandsoilvegetationsurrounding natural subsurfacesoil3-7 m deep18m X 30mFigure 2.1: Schematic and photograph of the BCR. Diagram adopted from[37]. Photograph taken by Maryam Khoshnoodi on 21 April 2009.262.3. Sample RemovalTrail, B.C.BCR # 1+ +1 meter1 A B C B2cores of length30-90 cm2 cm ID PVC pipeCentral lysimeterKeyLocation of drill holes July 2008, Boreholes: TS1(A)-TS2(B)-TS3(C) April 2009, Boreholes: TS4(A)-TS5(B2)-TS6(C) October 2009, Boreholes: TS7(A)-TS8(B)-TS9(C)++Figure 2.2: Schematic sampling map of the BCR and photographs of theTeck Cominco/NatureWorks treatment wetland at Trail, B.C.272.3. Sample RemovalFigure 2.3: Sample collection from Trail biochemical reactor. A portablediamond drilling rig was used to allow the withdrawal of the core samplesfrom the biochemical reactor, Trail, B.C.282.3.SampleRemovalTable 2.1: Recovered core solid samples and analysisCoreID Date Length Core Section-cm Borehole Analysis20080708-02-0-5 July2008 30 02 0-5 TS1 Chem(R)120080708-02-5-10 July2008 30 02 5-10 TS1 Chem(R)20080708-02-10-15 July2008 30 02 10-15 TS1 Chem(R)20080708-02-15-20 July2008 30 02 15-20 TS1 Chem(R)20080708-02-20-25 July2008 30 02 20-25 TS1∗ DNA(P,C)Mineralogy Chem(R,M)20080708-02-25-30 July2008 30 02 25-30 TS1∗ DNA(P,C)Mineralogy Chem(R,M)20080708-01-30-35 July2008 30 01 0-5 TS1 Chem(R)20080708-01-35-40 July2008 30 01 5-10 TS1 Chem(R)20080708-01-40-45 July2008 30 01 10-15 TS1 Chem(R)20080708-01-45-50 July2008 30 01 15-20 TS1 Chem(R)20080708-01-50-55 July2008 30 01 20-25 TS120080708-03-0-5 July2008 30 03 0-5 TS2 Chem(R)20080708-03-5-10 July2008 30 03 5-10 TS2 Chem(R)20080708-03-10-15 July2008 30 03 10-15 TS2 Chem(R)20080708-03-15-20 July2008 30 03 15-20 TS2 Chem(R)20080708-04-20-25 July2008 30 04 0-5 TS2∗ DNA(P,C)Mineralogy Chem(R)20080708-04-25-30 July2008 30 04 5-10 TS2∗ DNA(P,C)Mineralogy Chem(R)1Abbreviation: 16S clone library (C), Roche454 titanium sequencing (P), metagenomic library (MG), metal analysis (M) and carbonanalysis using RochEval 6 Technology (R). (*) indicates that the two core sections were combined for analysis.292.3.SampleRemovalTable 2.1 – continued from previous pageCoreID Date Length Core Section-cm Borehole Analysis20080708-04-30-35 July2008 30 04 10-15 TS2 Chem(R)20080708-04-35-40 July2008 30 04 15-20 TS2 Chem(R)20080708-04-40-45 July2008 30 04 20-25 TS2 Chem(R)20080708-04-45-50 July2008 30 04 25-30 TS2 Chem(R)20080708-05-0-5 July2008 30 05 0-5 TS3 Chem(R)20080708-05-5-10 July2008 30 05 5-10 TS3 Chem(R)20080708-05-10-15 July2008 30 05 10-15 TS3 Chem(R)20080708-05-15-20 July2008 30 05 15-20 TS3 Chem(R)20080708-05-20-25 July2008 30 05 20-25 TS3∗ DNA(P,C)Mineralogy Chem(R,M)20080708-05-25-30 July2008 30 05 25-30 TS3∗ DNA(P,C)Mineralogy Chem(R,M)20080708-06-30-35 July2008 30 06 0-5 TS3 Chem(R)20080708-06-35-40 July2008 30 06 5-10 TS3 DNA(MG1)Chem(R)20080708-06-40-45 July2008 30 06 10-15 TS3 Chem(R)20080708-05-45-50 July2008 30 05 15-20 TS3 Chem(R)20080708-05-50-55 July2008 30 05 20-25 TS320090421-01-0-5 Apr2009 25 01 0-5 TS6 Chem(R)20090421-01-5-10 Apr2009 25 01 5-10 TS6 Chem(R)20090421-01-10-15 Apr2009 25 01 10-15 TS6 Chem(R)20090421-01-15-20 Apr2009 25 01 15-20 TS6 Chem(R)20090421-01-20-25 Apr2009 25 01 20-25 TS6 Chem(R)302.3.SampleRemovalTable 2.1 – continued from previous pageCoreID Date Length Core Section-cm Borehole Analysis20090421-02-25-30 Apr2009 25 02 0-5 TS6∗ DNA(P,C)Mineralogy Chem (R,M)20090421-02-30-35 Apr2009 25 02 5-10 TS6∗ DNA(P,C)Mineralogy Chem (R,M)20090421-02-35-40 Apr2009 25 02 10-15 TS6 Chem(R)20090421-02-40-45 Apr2009 25 02 15-20 TS620090421-03-0-5 Apr2009 30 03 0-5 TS420090421-03-5-10 Apr2009 30 03 5-10 TS420090421-03-10-15 Apr2009 30 03 10-15 TS420090421-03-15-20 Apr2009 30 03 15-20 TS420090421-03-20-25 Apr2009 30 03 20-25 TS4∗ DNA(P,C,MG2)Mineralogy Chem (R,M)20090421-04-25-30 Apr2009 20 04 0-5 TS4∗ DNA(P,C)Mineralogy Chem(R,M)20090421-04-30-35 Apr2009 20 04 5-10 TS420090421-04-35-40 Apr2009 20 04 10-15 TS420090421-04-40-45 Apr2009 20 04 15-20 TS420090421-05-0-5 Apr2009 15 05 0-5 TS5 Chem(R)20090421-05-5-10 Apr2009 15 05 5-10 TS5 Chem(R)20090421-05-10-15 Apr2009 15 05 10-15 TS5 Chem(R)20090421-06-15-20 Apr2009 20 06 0-5 TS5 Chem(R)20090421-06-20-25 Apr2009 20 06 5-10 TS5∗ DNA(P,C)Mineralogy Chem(R,M)312.3.SampleRemovalTable 2.1 – continued from previous pageCoreID Date Length Core Section-cm Borehole Analysis20090421-06-25-30 Apr2009 20 06 10-15 TS5∗ DNA(P,C)Mineralogy Chem(R,M)20090421-06-30-35 Apr2009 20 06 15-20 TS5 Chem(R)20090421-07-35-40 Apr2009 15 07 0-5 TS5 Chem(R)20090421-07-40-45 Apr2009 15 07 5-10 TS520091020-01-0-5 Oct2009 35 01 0-5 TS7 Chem(R)20091020-01-5-10 Oct2009 35 01 5-10 TS7 Chem(R)20091020-01-10-15 Oct2009 35 01 10-15 TS7 Chem(R)20091020-01-15-20 Oct2009 35 01 15-20 TS7 Chem(R)20091020-01-20-25 Oct2009 35 01 20-25 TS7∗ DNA(P,C)Mineralogy Chem(R,M)20091020-01-25-30 Oct2009 35 01 25-30 TS7∗ DNA(P,C)Mineralogy Chem(R,M)20091020-01-30-35 Oct2009 35 01 30-35 TS720091020-02-0-5 Oct2009 25 02 0-5 TS8 Chem(R)20091020-02-5-10 Oct2009 25 02 5-10 TS8 Chem(R)20091020-02-10-15 Oct2009 25 02 0-15 TS8 Chem(R)20091020-02-15-20 Oct2009 25 02 15-20 TS8 Chem(R)20091020-03-20-25 Oct2009 35 03 0-5 TS8∗ DNA(P,C)Mineralogy Chem(R,M)20091020-03-25-30 Oct2009 35 03 5-10 TS8∗ DNA(P,C)Mineralogy Chem(R,M)20091020-03-30-35 Oct2009 35 03 10-15 TS8 Chem(R)20091020-03-35-40 Oct2009 35 03 15-20 TS8 Chem(R)322.3.SampleRemovalTable 2.1 – continued from previous pageCoreID Date Length Core Section-cm Borehole Analysis20091020-03-40-45 Oct2009 35 03 20-25 TS8 Chem(R)20091020-03-45-50 Oct2009 35 03 25-30 TS8 Chem(R)20091020-03-50-55 Oct2009 35 03 30-35 TS820091020-04-0-5 Oct2009 30 04 0-5 TS920091020-04-5-10 Oct2009 30 04 5-10 TS920091020-04-10-15 Oct2009 30 04 10-15 TS920091020-04-15-20 Oct2009 30 04 15-20 TS920091020-04-20-25 Oct2009 30 04 20-25 TS9∗ DNA(P,C,MG3)Mineralogy Chem(R,M)20091020-05-25-30 Oct2009 45 05 0-5 TS9∗ DNA(P,C)Mineralogy Chem(R,M)20091020-05-30-35 Oct2009 45 05 5-10 TS920091020-05-35-40 Oct2009 45 05 10-15 TS920091020-05-40-45 Oct2009 45 05 15-20 TS920091020-05-45-50 Oct2009 45 05 20-25 TS920091020-05-50-55 Oct2009 45 05 25-30 TS920091020-05-55-60 Oct2009 45 05 30-35 TS920091020-05-60-65 Oct2009 45 05 35-40 TS9332.4. Analyses2.4 Analyses2.5 Molecular Biology AnalysisFor microbial community analysis of the BCR, DNA extracted from the coresamples was used for 16S clone library construction and Roche454 titaniumsequencing (pyrotaq analysis). This part was done by other laboratory re-searchers and was described previously [6]. The samples that were used for16S clone library and pyrotaq analysis were indicated by abbreviations (C)and (P) respectively in the Table 2.1.In addition, three metagenomic libraries were constructed (MG1, MG2,MG3) from three core samples (Table 2.1) in Dr. Hallam’s laboratory at theDepartment of Microbiology and Immunology, UBC [153]. The large DNAinsert clone libraries (Fosmid libraries) were used to identify functional ar-senic genes and to investigate taxonomic diversity in the BCR. The analysisand results are presented in Chapter 5.2.6 Chemistry Analysis2.6.1 Porewater AnalysisThe pore water environmental variables measured in the field were dissolvedoxygen (DO), pH, oxidation/reduction potential (ORP) and temperature(T). These parameters were measured with a YSI 6920 multi-probe sonde(YSI Inc. Yellow Springs, Ohio, U.S.A.). Concentrations of chemical con-stituents such as ammonium-N, nitrite-N, nitrate-N, sulfate, sulfide and totalphosphorous were measured at the site using Hach (Mississauga, Ontario)and Chemets kits. The porewater measurements in the BCR is presented inTable C.2.Dissolved and total metals were monitored biweekly from the influent,effluent, lysimeter and borehole pore water in the BCR (Tables B.1 B.2 B.2in the Supplemental Information). These measurements were performed inan industrial laboratory and the results were used for the construction ofEh-pH diagrams as described in Chapter 4.Furthermore, slurry samples, taken from different boreholes (TS1 TS6)30-90 cm deep into the BCR organic substrate, were withdrawn into anaero-bic serum bottles for microbial culturing in the laboratory and are described342.6. Chemistry Analysisin Chapter 4.2.6.2 Core Solid AnalysisOrganic Matter Characterization-Carbon AnalysisThe carbon compound composition of the core samples was examined at theGeological Survey of Canada - Calgary Division, 3303 - 33rd Street usinga pyrolysis technique called Rock-Eval 6. This method, used for character-izing the organic matter, obtained information about the different forms ofcarbon present in the samples. Table 2.1 shows the samples that were usedfor this chemistry analysis and is indicated as Chem (R). Carbon analysis ofthe BCR core solid samples with RockEval 6 technology is summarized inTables A.2and A.3. The quantity of pyrolysable carbon (PC), total organiccarbon (TOC), as well as mineral carbon (MinC) (insoluble organic matter)present in a sample were extracted from this method. Pyrolysable carbon(PC) is the mass lost upon heating at 650◦C and residual carbon (RC) isthe mass of carbon left over after heating at 850◦C. The released hydrocar-bons were monitored by flame ionization detection (FID), and released COand CO2 were monitored with an infra-red detection during the pyrolysisand combustion (oxidation) steps of Rock-Eval 6 apparatus. More detailedcomputation of these parameters is given [10] and parameters definition ofcarbon analysis is presented in Table A.4. The nature of the organic carboncompounds present in the BCR is important since easily degradable carbonavailability affects the nature of microorganisms present and long term op-eration of the system.Metal Content of the CoresThe core solid samples recovered from the BCR were analyzed for metalcontent, (Chem(M), Table 2.1). Determination of the metal content in thecore samples was important for detecting the amount of arsenic accumu-lated inside the BCR and its correlation with microbial communities. Char-acterization of major elements, base metals, trace elements, volatiles, wasdetermined at the ALS Minerals Division analytical laboratory, Geochem-istry Group (2103 Dollarton Highway, North Vancouver, B.C., Canada) andis presented in Table A.1. Base metals, such as Cu and Zn, were decom-posed using a four acid digestion (HNO3-HClO4-HF-HCl) followed by induc-tively coupled plasma-atomic emission spectroscopy (ICP-AES) (Analytical352.7. Mineralogy AnalysisMethod 1). For major compounds (such as Fe2O3, SiO2, Al2O3) analy-sis, the samples were decomposed by lithium metaborate/lithium tetrab-orate (LiBO2/Li2B4O7) fusion and then analyzed using ICP-AES (Ana-lytical Method 2). Aqua regia digestion was used for arsenic detectionwith inductively-coupled plasma-mass spectroscopy (ICP-MS) (AnalyticalMethod 3). In addition, for total sulfur and carbon analysis, samples weredecomposed in a Leco Furnace (Leco Corp., St Joseph, MI, U.S.A.) andmeasured with infrared spectroscopy (Analytical method 4). For trace ele-ments, samples were mixed with lithium borate flux and were melted in afurnace at 1000◦C. The mixture was dissolved in HNO3/ HCl solution andthen analyzed with inductively coupled plasma - mass spectrometry (ICP -MS) (Analytical method 5).2.7 Mineralogy AnalysisThe recovered core samples were also examined for the presence of minerals.Table 2.1 represents the samples that were selected for this analysis. XRD(X-ray Diffraction), QEMSCAN (Quantitative Evaluation of Minerals byScanning Electron Microscopy) bulk mineral analysis (BMA) and particlemineral analysis (PMA) were used and are described in Chapter 3.36Chapter 3Mineralogical Study of theBiochemical Reactor,Identification of Arsenic andZinc Minerals3.1 ScopeTo determine the extent of biomineralization occurring in the BCR, miner-alogical analysis was applied to identify the chemical forms of metal precip-itates retained in the solid phase. In fact, precipitation of arsenite by bio-logically produced sulfides can be present under reducing conditions (thesishypothesis). SRB is an example of the group of bacteria that is presentunder reducing condition and consume organic carbon source supplementedat the site. Sulfide minerals, orpiment, realgar, arsenopyrite (Table 1.1) aswell as other form of metal sulfides, such as sphalerite can provide evidenceof this mechanism (Figures 1.3-biomineralization and Figure 1.4).Physicochemical conditions, present at the site (e.g. reducing environ-ment), can also help in arsenic removal from the BCR. Formation of metalsulfides and coprecipitation of arsenic with ferrioxyhydroxide are abiotic pro-cesses that possibly occur in the BCR. In this research, a detailed look inthe mineralogy of samples were undertaken. Minerals in BCRs were presentin trace amounts and the solids were mostly composed of organic materials.This limited the number of techniques that could be applied [109]. To studymetal bearing minerals, different mineralogical analysis and approaches wereperformed to deal with the low concentration of metals.In order to test these thesis hypotheses, mineralogical analysis was per-formed to study metal removal mechanisms in the BCR. The objectives wereto (1) identify the types of minerals present in the BCR (2) identify the ar-373.2. Experimental Sectionsenic and zinc bearing minerals with more focus on the mineralogical formsexpected from biogenic mechanisms and (3) correlate these findings with ataxonomic profile of microorganisms identified in the BCR [6], putativelyinvolved in biomineralizaton/precipitation.3.2 Experimental Section3.2.1 Sampling Site CharacterizationThe biochemical reactor (BCR) selected for this study received metal andsulfate-rich seepage from a smelter-waste landfill [95]. A schematic of theBCR is included in the Supplemental Information, Figure B.1. The BCRwas reconstructed in 2002 with a mixture of pulp mill biosolids, iron-coatedsand and limestone. Samples were collected from cores removed by drillinginto the reactor matrix at three different locations and on three differentdays: 22 July 2008 (samples ts1, ts2 and ts3), 19 April 2009 (samples ts4,ts5 and ts6) and 20 October 2009 (samples ts7, ts8 and ts9). Considerableeffort was needed to access the BCR biosolids, which were below more than1 m of soil and sand. A portable diamond drilling rig was used to allow thewithdrawal of the cores inside 2.0 cm internal diameter polyvinyl chloride(PVC) pipe that was then sectioned into approximately 30 to 40 cm lengths.The core samples were frozen using liquid N2 immediately after removal. Inthe laboratory, each core was sectioned into 5 cm intervals, each of whichwas homogenized by grinding under liquid N2 and used for chemistry, phy-logenetic and mineralogical analysis. In the field, pore water environmentalvariables, such as pH, dissolved oxygen (DO), oxidation/reduction poten-tial (ORP) and temperature, were measured with a YSI 6920 multi-probesonde (YSI Inc. Yellow Springs, Ohio, U.S.A.). The average total concentra-tions of arsenic (T/As), zinc (T/Zn) and sulfate in the BCR influent duringthe sampling period of 25 June 2008 to 2 October 2009 were 33.2, 71.1and 654.1 mg/L, respectively. The average total concentrations of arsenic(T/As) and zinc (T/Zn) decreased to 8.8 and 34.3 mg/L, respectively, whichdemonstrated a high metal removal efficiency within the BCR. The overalltreatment system of two anaerobic BCRs, three plant cells and a settlingpond achieved 99.97% total As removal [95]. The sulfate concentrationswere highly variable during this period. The average total suspended solidsentering the BCR between July 2008 and October 2009, was 168.5 mg/L,which decreased to 82.7 mg/L in the effluent.383.2. Experimental Section3.2.2 Core Solids ChemistryCharacterization of major elements, base metals, trace elements, and volatiles,as well as total nitrogen, carbon and sulfur, were determined at the ALSMinerals Division analytical laboratory, Geochemistry Group (2103 Dollar-ton Highway, North Vancouver, B.C., Canada). Base metals, such as Cu andZn, were decomposed using a four acid digestion (HNO3-HClO4-HF-HCl)followed by inductively coupled plasma-atomic emission spectroscopy (ICP-AES). For major compounds (such as Fe2O3) analysis, the samples weredecomposed by (lithium metaborate/lithium tetraborate) (LiBO2/Li2B4O7)fusion and then analyzed using ICP-AES. Aqua regia digestion was used forarsenic detection with inductively-coupled plasma-mass spectroscopy (ICP-MS). For total sulphur analysis, the sample was decomposed in a Leco Fur-nace (Leco Corp., St Joseph, MI, U.S.A.) and measured with infrared spec-troscopy.3.2.3 Pretreatment of Core Samples: Organic Matter (OM)RemovalThe presence of significant amounts of organic matter (OM) in samples formineralogical analysis can interfere with X-ray diffraction and scanning elec-tron (SEM) microscopy analysis; therefore, an effective method for the re-moval of OM from the solid samples without changing their inorganic phaseswas required before mineralogical assessment. Core samples obtained fromthe biochemical reactor contain high amounts of organic matter (from 1.7wt % to 13.2 wt % total organic carbon (TOC)). For this purpose, low tem-perature ashing (LTA) was applied, which uses radio frequency radiation toexcite oxygen to remove OM from the samples at temperatures below 100◦C [2]. A plasma asher (Planer EtchII, Technics West Inc., Cranston, R.I.,U.S.A.) at the Materials and Process Engineering Laboratory (AMPEL)-Nanofabrication Facility (ANF), University of British Columbia, was used,and the LTA reaction was carried out at 0.550 torr (0.73 mbar), 25 ◦C andat a radio frequency (RF) power of 70 W for 1 h. Radio frequency radia-tion dissociates, ionizes and excites oxygen molecules into chemically excitedatoms, which removes OM from the samples. For optimum results, the re-action was stopped after 30 min, and samples were stirred and reloaded foran additional 30 min.393.2. Experimental Section3.2.4 Mineralogical AnalysisXRD (X-ray Diffraction)Mineralogical analysis was used to identify the chemical forms of the metalsin the solid phase. For these types of samples, the collection of accuratemineralogical data is limited by the poor crystallinity of the precipitatesand their rarity, particularly for arsenic minerals, in the sample. A D8 FocusBragg-Brentano diffractometer (Bruker , Karlsruhe, Germany ) and a D5000Bragg-Brentano diffractometer (Siemens (Bruker) , Karlsruhe, Germany) atthe Electron Microbeam/X-Ray Diffraction Facility, Department of Earth,Ocean and atmospheric Sciences, University of British Columbia, were usedfor collecting diffraction data for qualitative analysis of the treated samplesand preliminary mineral characterization. Diffraction data were processedby automated “search-match” software, which makes use of the Interna-tional Centre for Diffraction Data PDF-4+ database (International Centrefor Diffraction Data (ICDD), Newtown Square, PA, USA). The mineralsidentified by these methods were compared with the quantitative evalua-tion of minerals by scanning electron microscopy (QEMSCAN) bulk mineralanalysis results (method described below).QEMSCAN (Quantitative Evaluation of Minerals by ScanningElectron Microscopy)QEMSCAN is a scanning electron microscope (SEM) that provides rapid au-tomated quantitative mineral analysis using Backscattered Electrons (BSE)(Bruker, Karlsruhe, Germany) and up to four energy dispersive X-ray (EDX)spectrometers. Nine representative subsamples taken from the BCR coresamples were mounted into epoxy resin blocks. Standard sized, polished,thin sections were carbon coated and analyzed at the ALS Metallurgy Lab-oratory (2957 Bowers Place, Kamloops, B.C., Canada) using a QEMSCANQuanta 650W (FEI Academy, Brisbane, Australia). Two modes of QEM-SCAN analysis were used: bulk mineral analysis (BMA) and particle min-eral analysis (PMA). BMA provides information on mineral composition,and PMA identifies the details of the mineral fragmentation characteristics(liberation and association) of minerals. In fact, the PMA mode createsdigital images by using BSE and EDX signals from SEM, and each pixel inthe image corresponds to mineral species in a small region under the elec-tron beam. The individual particles section was scanned by this method,and BSE and EDX photons emanating from a given point were used to403.2. Experimental Sectionidentify the elements present. The measured spectra were compared againsta database of known spectra, and a mineral or phase name was assignedto each measurement by the QEMSCAN computer software. X-ray spotanalysis was applied to specific mineral forms for further identification andverification of mineral phases present. More details on the operation of thesystem is given in References [122] and [21]. In keeping with the goal of theBCR, the focus was on arsenic and zinc-bearing minerals.3.2.5 Geochemical ModelingIn order to predict the formation of arsenic-bearing minerals under thephysico-chemical conditions present at the site, Eh (redox potential) -pH dia-grams for the system of As-Fe-S-O-H were constructed using the Spana (pre-viously Hydra and Medusa) chemical equilibria computing software packageand database, version 2.0, 10 April 2013 [126],[127]. Arsenic species equilib-ria constants were updated based on the most recent literature, as reportedin Tables 1 and 2 of Lu and Zhu [87]. Predominant As species in the As-Fe-S-O-H system were predicted over the range of total As, Fe and S concentra-tions measured in the BCR influent, effluent, lysimeter and borehole porewater (Tables B.1 B.2 B.2in the Supplemental Information). The input filecontaining all the species, their log(K) values and stoichiometric coefficientsin terms of H+, e−, HS−, As(OH)3 and Fe2+ are shown in Table B.4 of theSupplemental Information.3.2.6 Correlation with Microbial CommunitiesThe microbial community composition in each of the BCR samples was de-termined by extracting DNA and using Roche454 titanium sequencing of theamplified V6 to V8 region of the small subunit ribosomal ribonucleic acids(SSU rRNA), as described previously [132],[6]. The sequences were binnedinto so-called operational taxonomic units (OTUs) based on 97% similar-ity, which represent putative microbial species. A representative sequencefrom each OTU was compared with a curated database, Silva111 [128], soas to assign a taxonomic classification. The closest cultured relatives tothe OTUs found in the BCR were found by using the program, Blastn, tosearch the National Center for Biotechnology Information (NCBI) 16S ribo-somal RNA sequences (Bacteria and Archeae) database [125]. Phylogeneticcompositions of the samples were compared using UNIFRAC [86] and vi-413.3. Resultssualized with principal component analysis implemented with the QIIMEsuite of Python scripts [26]. UNIFRAC evaluates the similarities betweenmicrobial communities from different samples or sites based on their phylo-genetic relatedness. Particular microbial groups that were more prevalent inhigh metal content samples were identified. Furthermore, the types of puta-tive sulfate-reducing bacteria and iron-reducing environment-related groupspresent in the BCR were determined based on taxonomic classification ofthe OTU representative sequences.3.3 ResultsThe BCR under study had been operating successfully for ten years. Duringthe five-year period from 2003 to 2007, 2,691 kg of arsenic, 6,930 kg zincand 85 kg cadmium were removed from approximately 25,550 m3 of watertreated [95]. Biweekly sampling undertaken as part of the present studyfrom July 2008 to June 2009 revealed average concentrations of dissolvedarsenic and zinc entering the BCR of 2.26±1.0 mg/L and 39.5±7.5 mg/L,respectively. These decreased to 1.22±1.1 mg/L and 27.5±7.4 mg/L, re-spectively, in the BCR effluent. At the time of sampling, conditions in theBCR were reducing and the pH was neutral to slightly basic (Table 3.1).The ORP was challenging to measure during the April sampling and nostable values were obtained. During the spring and fall samplings there wassome oxygen measured in the pore water, whereas during the summer nodetectable oxygen was present. The temperatures varied from 19.4◦C in thesummer to 6.5◦C in the spring. Some of the samples contained significantlymore metals (ts4 and, to a lesser extent, ts7) than the others. The organiccontent of the BCR solids varied widely with 1.74 to 13.28% total organiccarbon (TOC).3.3.1 Geochemical PredictionsGeochemical modelling of the As-Fe-S-O-H system using total As, Fe and Svalues within the ranges measured at the site revealed that realgar (AsS),elemental As and arsenopyrite (FeAsS) were all possible predominant formsof As under the Eh and pH conditions in the bioreactor (Table B.5 andFigure B.2, Supplemental Information) depending on the total As and Feconcentrations. The total S concentration was kept at 7 mM for the geo-chemical modelling study. The effect of increasing Fe concentrations was423.3. ResultsTable 3.1: Pore water and bulk solids chemistry measured in the core sections collectedfrom the BCRSampleID ts1 ts2 ts3 ts4 ts5 ts6 ts7 ts8 ts9Date 22 July 2008 21 April 2009 20 October 2009Pore water chemistrypH 5.6 5.6 6.9 6.9 6.9 7.0 6.7 6.2 7.5ORP (mV) −133 −112 −238 NA 1 NA NA −118 −244 −130DO (mg/L) 0 0 0 1.5 1.1 0.7 1.2 1.3 1.3T (◦C) 17.6 15.9 19.4 7.6 6.5 7.1 11.2 11.5 11.2Solids analysisAs (ppm) 2.7 2.9 1.8 >250 16.6 18.9 45.8 9.3 18.9Zn (ppm) 108 122 118 1735 180 177 705 127 380Cu (ppm) 32 15 11 79 22 19 21 12 12Fe2O3 (%) 1.58 1.36 1.86 NA 2.12 2.24 1.29 2.35 1.88S (%) 0.24 0.19 0.11 NA 0.31 0.38 0.27 0.15 0.09TOC (mg/L) 4.46 4.29 3.17 5.20 8.70 8.00 13.28 2.69 1.74Note: 1 not available; data were not measured.to increase the stability range for arsenopyrite, such that, at the averageBCR influent concentrations of total As, Fe and S (0.44, 0.23 and 7.00 mM,respectively), arsenopyrite was the only predicted predominant form (Fig-ure 3.1). The formation of orpiment (As2S3) and realgar (AsS) could bepossible under less reducing and more acidic conditions. At lower total Feconcentrations, modelling predicted that realgar and orpiment would be pre-dominant over wider pH ranges, including the pH values measured in theBCR in the case of realgar (Figure B.2, Supplemental Information).3.3.2 Mineral Species Based on XRD AnalysisXRD analysis revealed that the principal minerals were silicates and carbon-ates. Quartz, calcite, feldspar (albite, orthoclase), amphibole (actinolite,ferro-actinolite), chlorite (clinochlore), hauyne and muscovite were presentin almost all samples. Small amounts of sulfides, sulfates and arsenateswere also detected. Zinc sulfide (sphalerite) and iron sulfides (arsenopy-rite, pyrite) were identified by XRD in some of the samples. Oxides andhydroxides were also identified, such as manganoan franklinite and mag-netite (Table 3.2). This analysis is a qualitative approach, and detecting alldiffraction spectra of the trace minerals depends on the detection limit ofthe XRD analyzer. Therefore, XRD cannot be used to definitively detectthe formation of sphalerite or other sulfide and sulfate rare minerals, unless433.3. Resultsthey are present in high enough amounts (i.e., above 2-5%wt). To verify theformation of these minerals, further analysis was performed, as describednext.Figure 3.1: Eh (redox potential)-pH diagram for As-Fe-S-O-H at 25 ◦C and1 atm pressure. Predominant arsenic species predicted for the chemistry ofthe BCR pore water at: total arsenic, 0.44 mM; total sulphur, 7.00 mM; totaliron, 0.23 mM. Solid species and stable (equilibrium) phases are indicatedwith the dark field and the Eh and pH conditions of the BCR indicated withdashed box. Diagram produced from Spana software [126, 127].443.3. Results3.3.3 QEMSCAN and X-Ray Analysis ObservationsBy using QEMSCAN bulk mineral analysis (BMA), a more thorough min-eralogical composition was obtained for some subsamples by quantifying allthe mineral phases identified. Confirming the XRD analysis, samples con-sisted mainly of silicates (quartz, feldspars, chlorite, amphibole, epidote,muscovite, hauyne, biotite, clay, talc and sphene) and carbonates (calcite,dolomite). The main minerals, quartz, calcite and feldspar (plagioclase andalkali feldspar), represented between 9.2% and 29.9% of the subsamples.Phosphates (apatite) were also detected and represented around 0.3%. TheBMA analysis quantified the amounts of trace minerals, such as pyrite,arsenopyrite and iron oxides. Small amounts (<0.1%) of zinc-, arsenic-,copper- and lead-bearing minerals were found in all samples. Possible formsof zinc minerals are as carbonates, arsenic oxides, sulfates, sulfides and phos-phates. Arsenic was present as oxides and sulfides. Table 3.2 shows theabundance of these minerals. Sample TS2, in particular, contained signif-icant amounts of Zn and As minerals (sphalerite, arsenopyrite, tennantiteand zinc sulfate). Iron oxides were highest in sample TS5 and includedlimonite, goethite, magnetite and hematite. This sample also had a signifi-cant amount of arsenopyrite.To further evaluate the forms of the arsenic- and zinc-bearing minerals,which were challenging to detect, due to their relatively low concentrations,a hybrid particle mineral analysis (PMA) was performed on each sampleusing QEMSCAN supported by EDX analysis. The PMA analysis revealedthat some mineralogical forms were located in close proximity to each otheror in association with each other. Figures 3.2, 3.3 and 3.4 present selectedparticles that were analyzed by this method. In general, the trace mineralswere grouped into four different classes: zinc-arsenic oxides; and zinc-leadsulphosalts/sulfates, zinc-arsenic sulphosalts/sulfates and zinc phosphates.In one particle, zinc and arsenic were present as oxides. This particlecontained up to 25.6% arsenic and 27.4% zinc and was associated with thebright phases of a multiplex grain. X-ray analysis suggested co-occurrencewith feldspar (data not shown).Zinc sulfate minerals were identified, such as zinc-aluminum sulfate, zinc-lead sulfates and zinc sulfates. Zinc-aluminum sulfates corresponded to themineral, zinc woordite ((Zn1−xAlx)(OH)2)(SO4)x/2), according to the XRDspectra.453.3. ResultsZinc-arsenic sulphosalts/sulfates, particularly sulfides, were relativelypredominant in the samples compared to other minerals identified withQEMSCAN analysis: sphalerite was a common mineral observed in many ofthe samples, either as liberated sphalerite particles (zinc and sulphur spectrasolely detected) or in association with other particles, such as iron oxides orsilicate minerals. An interesting observation was the coating of sphalerite(24.5%–26.3% Zn and 12.6%–15% S) on amphibole (Figure 3.2a). The X-ray analysis revealed that As was also present in this mineral phase up to2.2%. Similarly, Figure 3.2c shows a feldspar particle with a bright coatingpresent that contained zinc and sulphur at 36.1% and 14.7%, respectively,that was also associated with arsenic (3.0% As). An arsenic peak was absentfrom the point analysis spectra obtained from inside of the amphibole andfeldspar particles, verifying arsenic’s association with the sphalerite coating.The QEMSCAN BMA reported arsenopyrite in some particles (Table 3.2),such as the one with an irregular shape and average diameter estimated at33.44 µm in Figure 3.3a. EDX spectra indicated high amounts of arsenic inthis particle (2.6%–21.8%), as well as Zn (1.1%–29.4%), Fe (1.5%–15.5%),S (0–6.2%) and, at some positions, cadmium (0–6%). The composition ofthese elements varied throughout the particle, making it difficult to asso-ciate them with any one mineral type.Copper-iron sulphosalts/sulfates were detected. One with 15.1% Cu,12.1% Fe and 7.0% S is shown in Figure 3.3c. This particle was associatedwith an aggregate of irregular-shaped non-mineral phases and could rep-resent chalcopyrite (CuFeS2) or guildite (CuFe+3(SO4)(OH)4H2O). Indeed,XRD and EDX analysis detected copper minerals, such as guildite, chal-copyrite, chalcocite and tennantite. There were some challenges with findingtennantite with XRD analysis, since the tennantite peak overlaps with otherpeaks. With QEMSCAN BMA analysis, only small amounts of guildite werepresent, but the arsenic-bearing tennantite, (Cu11.9Fe0.1)As4S13, was moreprevalent at 2.2% and 7.4% of the arsenic-bearing minerals in samples TS1and TS2, respectively (Table 3.2).463.3. Results(a) (b)(c) (d)Figure 3.2: Backscattered electron microscopy photomicrographs of zinc-arsenic sulfides on the surface of (a) amphibole and (c) feldspar andone example EDX spectrum ((b) and (d)) from the bright mineralphase around the outside of the amphibole or feldspar (red arrow onthe photomicrographs). The small white-boarded triangles indicate thepositions where EDX spectra were obtained. The particles detected in sam-ple TS4.473.3.ResultsTable 3.2: Mineralogical composition of the BCR samplesSampleID ts1 ts2 ts3 ts4 ts5 ts6 ts7 ts8 ts9ChemicalformMineral (retrievedfrom XRD)Formula (retrieved from EDX) BMA(Qemscan)Carbonates Calcite CaCO3 16.3 21.9 16 24.3 15.3 15 28.1 29.9 10.2Minrecordite/dolomite Ca(Zn,Mg)(CO3)2 0.3 0.3 0.6 0.6 0.6 0.5 0.8 1.3 0.2Oxides, Hy-droxidesIron oxides a FexOx 0.2 0.2 0.5 0.2 1.1 0.5 0.2 0.2 0.3not detected ZnAsOxSulfate Zincwoodwardite-3R ((Zn1−xAlx)(OH)2)(SO4)x/2) P P P PGuildite CuFe+3(SO4)(OH)4H2O P P Pnot detectedb Zn− SO4 1.8 2.2 1.3 2.2 1.4 0.9 3.1 1.3 1.3Zn− PbSOxSulfide Sphalerite b ZnS 1.9 8.3 1.9 0.8 3.6 1.5 1 0.5 1Arsenopyritec FeAsS 0 34.6 4.1 0 25.7 0 0 7.4 0Pyrite FeS2 <0.1 0.2 <0.1 <0.1 <0.1 <0.1 0.1 0.1 <0.1Iron sulfies d FeS2 or (Fe,Ni)S0.9not detected FeAsS-Zn or FeAsS-Zn-CdTennantitec (Cu,Fe)12As4S13 2.2 7.4Chalcopyrite CuFeS2not detected CuCoNiZnSChalcocite Cu2SPhosphates,Arsenates,VanadateWilhelmkleinite ZnFe2(AsO4)2(OH)2 Pe P P Pnot detected ZnPOx or ZnFe− POxalimonite, goethite, magnetite and hematitebthe values represent the percentage of this particle in total of zinc bearing minerals in the samplecthe values represent the percentage of this particle in total of arsenic bearing minerals in the sampledMarcasite, mackinawiteethe particle were not quantified and only identified by XRD, (P;Present)483.3. ResultsZinc phosphate particles were also identified by PMA, as shown in Fig-ure 3.4a, where the spectra clearly show phosphorus and zinc peaks (6.5%and 16.3%, respectively). One interesting observation was that metalliczinc particles (64.9%–65.2% Zn) were present on the outside of a dark ob-ject composed of carbon, approximately 100 µm long. In Figure 3.4a, zinciron sulfides were also detected.QEMSCAN PMA identified arsenate oxide particles, such as wilhelmk-leinite ZnFe2(AsO4)2(OH)2. Figure 3.4c presents an example, where thebright phases are related to zinc, iron and arsenic peaks. The compositionmeasured at one location was 10.8% Zn, 15.2% Fe and 23.9% S, respectively.Wilhelmkleinite was identified also by XRD analysis and QEMSCAN BMA(Table 3.2).3.3.4 BCR Microbial Groups Possibly Associated withMetal RemovalComparison of the microbial community compositions of the core sampleswas based on their phylogenetic distances from each other, and they wereclustered according to three principal coordinates. The two coordinates rep-resenting the most variation are presented in Figure 3.5, where the samplesare coloured according to the amount of arsenic and zinc measured in thesolids (values from Table 3.1). The microbial communities found in thehigh metal concentration samples cluster together towards the top left ofthe principal coordinate diagram, whereas those in the low metal concentra-tion samples are grouped together towards the bottom right. The microbialgroups more prevalent in the metal-rich samples (Figure 3.6) included thepoorly characterized taxonomic groups of Bacteroidetes-related SB-1, Vad-inHA17 and M2PB4-65 environmental groups, candidate division WS6, RF3and TM6, as well as the better known Synergistaceae family, Victivallalesorder and methanogen genera Methanocorpusculum, Methanospirillum andMethanosarcina. The number of sequences or reads binned into each OTUprovides an indication of the prevalence of the organism from which thosesequences came in the BCR samples (Table B.6, Supplemental Informa-tion). The Methanospirillum-related OTU was the most highly representedfollowed by an OTU classified in the WS6 phylum and a M2PB4-65 termitegroup-related OTU.Although not highly correlated with the high metal concentration sam-493.3. Resultsples, sequences related to sulfate-reducing and iron-reducing bacteria, espe-cially those represented in the Deltaproteobacteria class, were detected in thecore samples. The Deltaproteobacteria-related operational taxonomic unitsor OTUs (bins of sequences more than 97% homologous to each other) wereclassified into the orders of Bdellovibrionales, Desulfobacterales, Desulfovib-rionales, Desulfuromonadales, Myxococcales and Syntrophobacterales andenvironmental clone groups GR-WP33-30, Sh765B-TzT-29 and Sva0485 orwere unclassified sequences labeled as Incertae sedis (Table B.7, Supplemen-tal Information).The most prevalent groups of putative sulfate-reducing bacteria iden-tified in the BCR were classified into the orders Desulfovibrionales andDesulfobacterales (Table B.7), Supplemental Information), which were mostclosely related to cultured species, Desulfobulbus elongatus strain FP, Desul-fovibrio mexicanus strain Lup1, Desulfovibrio paquesii ,Desulfovibrio aminophilus strain ALA-3 and Desulfovibrio idahonensis. Themost prevalent Deltaproteobacteria related OTUs were classified into the en-vironmental group Sh765B-TzT-29 (40% of all Deltaproteobacteria-relatedsequences). There were no cultured species closely related to the BCRSh765B-TzT-29 OTUs. In addition to the Deltaproteobacteria-related OTUs,putative SRBs were classified in five genera in the Firmicutes phylum. EachOTU was rare (less than 10 reads per OTU), but the total number of se-quences assigned to the Desulfosporosinus and Desulfotomaculum generaindicate that these SRB may have been important sulfate reducers in theBCR.None of the BCR sequences were classified in genera of known iron-reducing bacteria, such as Shewanella or Geobacter. Very few sequenceswere classified as Thiobacillus and Acidithiobacillus. However, the BCRphylogeny indicated that many microbial groups were similar to those foundpreviously in iron-reducing enrichments, such as Gracilibacter thermotoler-ans JW/YJL-S1 (97% homology), Proteiniphilum acetatigenes (94% homol-ogy), Bacteroides graminisolvens strain XDT-1 (97% homology) and theanaerobic digester-related environmental clone, VadinBC27 (accession num-ber U81676). However, all of the previously mentioned cultured species arefermentative, proteolytic or cellulose degraders, and none of them are knowniron-reducers.503.4. Discussion3.4 DiscussionSince dissolved arsenic and zinc concentrations decreased through the BCR,there is evidence that adsorption and mineralization were occurring andcontributing to metal removal, in addition to filtration of the incoming solidforms of arsenic and zinc. Ko¨ttigite (Zn3(AsO4)28H2O) that was detectedin the BCR influent previously was not identified in any of the BCR coresamples subjected to XRD in this study. One possibility could have beenthat all the ko¨ttigite was removed due to filtration deeper in the BCR nearthe inlet. Another explanation might be that ko¨ttigite was transformed tothe closely related zinc-iron-arsenate mineral, Wilhelmkleinite, that was de-tected by XRD and QEMSCAM BMA.Nevertheless, it is not impossible that some of the solids entering theBCR could have dissolved and reformed as different mineral types. Arsenate-reducing bacteria were present in the BCR [72] and might have contributedto the release of arsenic from arsenate mineral forms.The main drivers for mineralization in the BCR were the geochemicalreducing conditions, processes mediated by microorganisms and adsorptionto surfaces. It is not possible to determine definitively whether microbiolog-ical processes were direct or indirect, causal or incidental to mineralization.However, mineralogical forms were detected that were consistent with bio-logical mechanisms playing a role together with geochemistry, as is discussedbelow.3.4.1 Evidence for Biotic MineralizationMinerals that are consistent with biomineralization processes were detected,namely sulfides and carbonates. Sphalerite was a common mineral observedin many of the samples, and the microbial community analysis of the BCRsupports the hypothesis of a biological role in that SRBs were present andlikely provided the source of sulfide. Biologically-induced ZnS (sphalerite)formation by SRB has been demonstrated in other studies under laboratoryconditions and in biofilms exposed to mine effluent [5, 77, 104]. Thus, theBCR was successful in creating environmental conditions for SRB to thrive.The forms of arsenic sulfide found in the BCR were tennantite and ar-senopyrite, the latter being the predominant form predicted by As-Fe-S-O-H513.4. Discussiongeochemical modelling for high Fe conditions. This indicates that iron playedan important role in arsenic removal in this system. The role that biologymight have played to make the formation of arsenopyrite possible would bearsenate reduction to arsenite by arsenate reducing bacteria and the reduc-tion of sulfate to sulfide by SRB, since both microbial groups were identifiedin the BCR. Arsenite may have adsorbed to pre-existing iron sulfides toform arsenopyrite. Arsenic adsorption to solids is the major mechanism re-sponsible for reducing aqueous arsenic concentrations in natural soils andsediments [162]. Arsenite sorbs strongly to iron sulfides, pyrite (FeS2) andpyrrhotite (FeS), especially at circumneutral pH [18]. X-ray absorption spec-troscopy of iron sulfides that had been reacted with arsenic revealed that Ason the mineral surface was coordinated with both Fe and S, demonstratingthe formation of arsenopyrite [18]. Iron sulfides were present in the BCR;therefore, such a mechanism for arsenopyrite formation was possible. Ac-cording to geochemical predictions, conditions in the BCR were favorablefor the formation of other arsenic sulfides (realgar or orpiment) under condi-tions of low Fe (Figure B.2, Supplemental Information), but neither of theseAs minerals were observed in our samples. In laboratory SRB reactors, re-algar and orpiment were the predominant arsenic precipitates at low pH,which was attributed to the presence of acid-tolerant SRB, such as thoseof the Desulfosporosinus genus [7]. Sequences closely related to these SRBwere also found in the BCR core samples.In summary, the microbial activities of many groups in the BCR playedindirect roles in mineralization by consuming all the oxygen entering withthe influent and creating a reducing environment, as well as by increasing al-kalinity through carbonate formation, which raised the pH. Combined withbiogenic sulfide formation, this provided favorable geochemical conditionsfor metal sulfide precipitation.Carbonates could have formed during treatment. Both dolomiteCaMg(Co3)2 and minrecordite CaZn(CO3)2 were detected in the BCR sam-ples and could have been mineralogical sinks for some of the Zn removed dur-ing treatment. As dolomite is formed, Zn could substitute for Mg, formingminrecordite. Furthermore, solid solution from Mg-dolomite, CaMg(CO3)2,towards Zn-dolomite is possible [139]. Zn-dolomite has been identified asa secondary mineral in wastes associated with historical Zn-Pb ore min-ing [24]. Microbiologically-mediated dolomite formation has been shown tooccur in anaerobic sediments and in laboratory experiments, where sulfate-reducing bacteria were found to play a specific role in the formation of the523.4. Discussiondolomite [160, 164]. Thus, it is feasible for carbonate formation to be an-other biologically-mediated mineralization mechanism for Zn in the BCR.Microorganisms, such as Synergistetes, that were highly correlated withmetal content suggest a role for these microbes in mineralization, althoughthere are no studies available supporting their bioprecipitation of arsenicand zinc. The most closely related cultured species was Thermanaerovib-rio acidaminovorans (90% homology), which is a thermophilic amino acidfermenter, which was the first member of the Synergistetes phylum to tohave its complete genome sequenced [29]. The Synergistetes-related BCRsequences were distantly related to the species, Dethiosulfovibrio peptidovo-rans, which has been associated with metal interactions, such as corrosionof steel in oil wells [78], and closely related clones were prevalent in copper-polluted sediment in Chile [121]. Dethiosulfovibrio peptidovorans are strictlyanaerobic and known to reduce thiosulfate and might be important for metalsulfide precipitation.3.4.2 Other Possible Mechanisms for Mineral Formation inthe BCRTextural association of Zn- and As-bearing minerals with other particles, asrevealed with the QEMSCAM PMA imaging, suggested that adsorption andcoprecipitation may have contributed to immobilization of arsenic and zincin the BCR. This was observed on feldspar and amphibole (Figure 3.2c),where a bright coating was related to sphalerite with trace amounts of ar-senic. The feldspars and amphibole, which are silicate minerals, were preva-lent in the samples, due to the limestone that was included in the BCRmatrix. In fact, feldspar has been used for the treatment of arsenite(III) inother studies, where a maximum adsorption of 90.19% was obtained [146].Clay minerals, such as illite, kaolinite [92] and mullite [159], have been usedfor arsenate adsorption [116]. Kaolinite was present at 0.3% in some of theBCR samples; however, adsorption of As or Zn was not detected on theseminerals.Oxy-hydroxides, which were found in the BCR solids (such as those ofFe(III), Mn(III,IV) and Al(III)), are adsorbents for metal cations [144] andarsenic [111], most often under oxidative conditions. For example, removalof As(V) from solution by adsorption to iron oxy-hyroxides has been demon-strated in mine tailings at a uranium mine, where poorly crystalline scorodite533.5. Conclusionswas precipitated below pH 3 [79]. In many cases, anaerobic conditions leadto resolubilization of iron, due to bacterially enhanced iron reduction toFe(II), concomitant with As release [123]. Nevertheless, bacterial formationof Fe oxy-hydroxides is possible under anoxic conditions with nitrate as theelectron acceptor for iron oxidizing bacteria, and one such microbe was iso-lated from the BCR [96]. Although evidence of arsenic or zinc adsorptionon iron oxide particles was not observed in the samples used for this study,iron oxides, such as magnetite, were present, and in one case, particle X-rayanalysis found magnetite adjacent to sphalerite. Therefore, it is possible formetal adsorption on iron oxy-hydroxides to occur inside the BCR.Moreover, organic matter is also known to affect metal mobility in thatdegradation products, such as humic and fulvic acids, strongly adsorb arsenicat near neutral pH, for example [111],[144]. The pulp mill biosolids used inthe BCR matrix (65 wt % of the starting material) could have been a potentadsorbent for metal cations. Although the organics had been removed fromthe samples, rare elements were found that were associated with complexparticles and aggregates, possibly remnants of organic material. However,adsorption sites reach saturation relatively soon after start-up in these typesof bioreactors. Mattes estimated from the total mass of biosolids used inthe BCR that all adsorption sites would be occupied within the first fourmonths of operation [94]. This suggests that adsorption is not a reliablemechanism for long-term metal removal in these systems.3.5 ConclusionsA comprehensive mineralogical assessment using X-ray diffraction (XRD)and quantitative evaluation of minerals by scanning electron microscopy(QEMSCAN) was performed to identify arsenic- andzinc-bearing minerals in a biochemical reactor treating metal-rich leachate,so as to gain new insights into the chemical and biological processes thatmediate metal removal. Identification of these arsenic- and zinc-bearingminerals was very challenging, due to their rarity in the mostly organic-richmatrix, and was overcome through a low temperature ash pretreatment. Ahybrid particle QEMSCAN analysis (PMA) proved to be an effective ap-proach to find trace and low abundance arsenic- and zinc-bearing minerals.Many mechanisms, both biotic and abiotic, were involved in the mineral-543.5. Conclusionsization of arsenic and zinc as oxides, sulfides, sulfates and phosphates. Theevidence of arsenic and zinc-bearing minerals coating amphibole and feldsparparticles suggested that adsorption followed by precipitation. QEMSCAMBMA indicated that zinc and arsenic sulfides were predominant forms, whichwere attributed to the presence of SRB verified by the presence of sequencesrelated to known SRB found together with the precipitates. Inclusion ofzinc in carbonates was another possible biologically-mediated mechanism.Particular groups of microbes were highly correlated with metal-richsamples. Some of these were taxonomically classified in uncharacterizednovel phyla of environmental sequences. These organisms may play someas yet unknown role in metal precipitation. Sequences classified in the Syn-ergistetes phylum were prevalent and strongly correlated with metal-richsamples. Some members of this phylum are thiosulfate-reducing and couldpotentially play a role in metal sulfide formation.553.5. Conclusions(a) (b)(c) (d)Figure 3.3: Backscattered electron microscopy photomicrographs and EDXspectra of some other mineral particles: (a) arsenic-zinc-iron- ± cadmium-bearing particles; (b) EDX spectra for the point indicated by the red arrowin (a); (c) copper-iron-sulphide particles; (d) EDX spectra for the positionindicated by the red arrow in (c). 563.5. ConclusionsZinc-phosphorousMetallic zincZn-Fe-SulphidesCarbon(a) (b)(c) (d)Figure 3.4: Backscattered electron microscopy photomicrographs and EDXspectra of some other mineral particles: (a) zinc-phosphorus, metallic zincand zinc-iron-sulphide particles; (b) EDX spectra for position indicated bythe red arrow in (a); (c) arsenic-iron-zinc-oxide particle and (d) EDX spectrafor the position indicated by the red arrow in (c). 573.5. ConclusionsTS4TS7TS9TS3TS1TS2TS8TS5TS6AsTS4TS7TS9TS3TS1TS2TS8TS5TS6ZnFigure 3.5: Microbial community principal coordinate analysis with arsenicand zinc concentrations coloured from blue (high) to red (low).583.5. ConclusionsBacteria:Bacteroidetes:SB-1 uncultured bacteriumBacteria:Candidate Division WS6: uncultured bacteriumBacteria:Lentisphaeria:Lentisphaeria:VictivallalesBacteria: Synergistetes:Synergistia:SynergistalesBacteria:TM6:uncultured bacteriumBacteria: RF3: uncultured bacteriumTS4TS7TS9TS3TS1TS2TS8TS5TS6TS4TS7TS9TS3TS1TS2TS8TS5TS6TS4TS7TS9TS3TS1TS2TS8TS5TS6TS4TS7TS9TS3TS1TS2TS8TS5TS6TS4TS7TS9TS3TS1TS2TS8TS5TS6TS4TS7TS9TS3TS1TS2TS8TS5TS6Figure 3.6: Microbial community principal coordinate analysis with highlycorrelated microbial groups coloured blue (high read count) to red (low readcount)59Chapter 4Enrichment of ArsenicResistant Microbes from theBiochemical Reactor4.1 ScopeIn biologically active treatment systems, microorganisms capable of arsenictransformation are present and greatly influence arsenic speciation and mo-bility (Figure 1.3-biotransformation and Figure 1.4). There are microbesthat can respire on arsenic such as those referred to as dissimilatory arse-nate respiring prokaryotes (DARPs), which couple arsenate reduction withoxidation of organic compounds such as acetate, formate, pyruvate, citrateand many other simple organic compounds, that are notably present inBCRs as intermediates in metabolic cycles. Additionally, arsenate resistantmicroorganisms are also able to reduce arsenate via periplasmic arsenatereduction (Figure 1.3). In contrast to the previous groups, some microbesgain energy via chemolithotrophic arsenite oxidation (CAO). This mecha-nism can couple reduction of oxygen under aerobic conditions or reduction ofnitrate under anaerobic conditions. In this Chapter the objectives were: (1)to test the hypothesis regarding potential of arsenate reduction or arseniteoxidation or both in an operating BCR (Figure 1.3-biotransformation andFigure 1.4), (2) monitor chemical arsenic speciation, and (3) assess preva-lence of putative arsenic-transforming microbes in the actual BCR by theirread counts in a SSU rRNA library sequenced [6] from core samples takenfrom within the BCR.604.2. Material and Methods4.2 Material and Methods4.2.1 Sampling SiteThe BCR used in this study was the first in a series of two subsurface flowand three surface flow wetlands treating arsenic containing seepage fromlandfilled smelter waste in southern British Columbia, Canada [95]. TheBCR contained a mixture of pulp mill biosolids, sand and limestone throughwhich the metal-laden water percolated upwards. A cover of impermeablemembrane, sand and soil on top of the BCR prevented oxygen infiltrationso as to maintain anaerobic conditions. Slurry samples were taken using aperistaltic pump from six different boreholes (TS1 to TS6, Table 4.1) 30-90cm deep into the BCR organic substrate and withdrawn into sterile 100mlanaerobic serum bottles without air contact. The serum bottles were cappedwith no headspace and transported at the in-situ temperature (8◦C) to thelaboratory within 24 hrs. Environmental conditions inside the BCR such aspH, temperature, dissolved oxygen and oxidation/reduction potential weredetermined with a YSI 6920 multi-probe sonde (YSI Inc., Yellow Springs,OH, USA).4.2.2 Enrichment CulturingIn the laboratory, 0.5 ml of the liquid suspension was transferred into twodifferent growth media to enrich for chemoautotrophic arsenite oxidizing(CAO) microbes and heterotrophic arsenate reducing microbes, respectively.The arsenite oxidizing bacteria minimal salt enrichment medium recipe wasadapted from Santini el al. [141] with the addition of 5 mM sodium arsenite(NaAsO2), Na2SO4 0.5 g/L and KNO3 0.5 g/L (henceforth referred to asAs(III) medium). The anoxic minimal salts medium used to enrich for arse-nate reducing bacteria was adapted from Zhang et al. [169] and contained5 mM sodium arsenate (HAsNa2O4.7H2O) and lactate 1.121 g/L (referredto as As(V) medium). The media were dispensed into 25 ml anaerobictubes and in order to maintain the enrichment cultures, 10% of the originalsample was used as inoculum for the subsequent passage. After 12 daysincubation, 2 ml of the initial slurry was transferred to 18 ml fresh mediumand incubated for another 12 days. The cultures were incubated in thedark at ambient temperature (19-21◦C) in a Coy (Grass Lake Michigan)anaerobic glove chamber under N2/CO2/H2. By using florescence stainingwith SYBR Green (Biorad, Hercules, California) combined with flow cytom-etry to measure cell density, cultures with high cell counts were selected forDNA extraction and sequencing. Table 4.1 shows the samples selected for614.2. Material and Methodssequencing, their site of origin in the BCR and from which passage theywere taken. Liquid samples were taken periodically from each enrichmentculture for As speciation analysis and concentration measurements. See G.1.4.2.3 Analytical TechniquesArsenic speciation was measured using high performance liquid chromatog-raphy (HPLC, Waters 1525 binary HPLC pump) with ultraviolet (UV) de-tection (Waters 2487 dual lambda absorbance detector) for each culturejust before each passage. Detection at 191nm was used to determine As(III)and As(V) concentrations and any transformations in arsenic concentrationand/or speciation. Chromatographic separation setup and reagents wereadapted from the protocol in Jedynak et al. [67] with some modifications.A Waters IC-Pak A HC anion column was used with the mobile phase con-taining 10 mM sodium phosphate (NaH2PO4) buffer at pH 6.0. The flow ratewas 2.0 ml/min and injection volume was 200 µl. Limits of detection weredetermined from calibration plots in concentration range 0.107-0.293 mM(8-22 ppm) for As(III) and 0.067-0.0600 mM (5-45 ppm) for As(V) (Fig-ure C.1). Standards of 0.138 mM (As(III)) and 0.064 mM (As(V)) were runwith each sample analysis. Samples were diluted so that the total expectedarsenic concentration was within the analysis range.Biochemical reactor influent, effluent and pore water properties andchemistry were measured using 1) YSI Inc (Yellow Springs, Ohio) probesin situ for temperature, pH, and dissolved oxygen; 2) Hach (Mississauga,Ontario) kits immediately after sample removal for ammonium-N, nitrite-N,nitrate-N, ferrous-Fe, and sulfide, and 3) total and dissolved metals using in-ductively coupled plasma mass spectrometry (ICP-MS). Analysis describedin 1) and 2) were performed in the field.624.2.MaterialandMethodsTable 4.1: Enrichment cultures sequenced: Their sample IDs, site of origin (pore water boreholes), passagenumber and the number of OTUs (97%) obtained when the total number of reads in each sample was rarefied to5460.Arsenite Medium Arsenate MediumSample ID Origin Number ofpassagesNumber ofOTUsSample ID Origin Number ofpassagesNumber ofOTUsAsIII-1P2 TS1 2 197 AsV-1P2 TS1 2 211AsIII-1P3 TS1 3 252 AsV-1P3 TS1 3 232AsIII-1P7 TS1 7 205 AsV-2P3 TS2 3 240AsIII-2P3 TS2 3 284 AsV-3P3 TS3 3 205AsIII-3P3 TS3 3 217 AsV-3P8 TS3 8 214AsIII-4P3 TS4 3 165 AsV-4P3 TS4 3 134AsIII-4P8 TS4 8 174 AsV-4P8 TS4 8 196AsIII-5P3 TS5 3 207 AsV-5P3 TS5 3 201AsIII-5P7 TS5 7 203 AsV-6P3 TS6 3 97AsIII-6P3 TS6 3 120634.2. Material and Methods4.2.4 DNA Extraction, Polymerase Chain ReactionAmplification and Phylogenetic AnalysisDNA was extracted from 0.5ml samples of culture using a PowerMax SoilDNA isolation kit (Mo Bio Laboratories Inc., Carlsbad, CA, USA). Smallsubunit ribosomal nucleic acid amplicons for sequencing were prepared withpolymerase chain reaction (PCR) of the target DNA using primers specificfor the V6-V8 variable region: 926f (5 AAACTYAAAKGAATTGRCGG 3)and 1392r (5 ACGGGCGGTGTGTRC 3), using an iCycler (Biorad) ther-mocycler under conditions: 95◦C for 3min; 25 cycles of 95◦C for 30s, 55◦Cfor 45s, 72◦C for 90s; and 72◦C for 10min. Attached to the primers were bar-codes and linker-adapters for the LibA chemistry used for Roche GS-FLX Ti-tanium Series sequencing. Amplicons were purified using the QIAquick PCRpurification kit (Qiagen). DNA concentrations and purity were measured ona NanoDrop ND-2000 UV-Vis Spectrophotometer (NanoDrop Technologies,Wilmington, DE) and by running 1L of the PCR product on a 0.8% agarosegel. In total 19 samples were sequenced by the Centre dinnovation GnomeQubec et Universit McGill (Montreal, Quebec). Pyrotag reads were filteredusing the following quality control criteria: length between 200 and 500bp, no ambiguous base reads, no missing quality scores, mean quality scoremore than 25, no more than 6 nucleotide length homopolymer runs and nomismatches in reverse primer. The Qiime suite of Python scripts [26] and as-sociated dependencies were used to cluster the filtered reads into operationaltaxonomic units using the method usearch [40] with a 97% homology cut-off. Representative sequences for each OTU were assigned taxonomy usingBLASTn to the Silva version 111 representative set [128]. Very rare OTUsrepresented by only one sequence each in the entire dataset were filtered out.Samples were rarefied to the same sequencing depth (5460 reads) using the(rarify seqs.py) script of Qiime. The most highly represented OTUs foundin each enrichment sample were selected as those represented by more than5% of the total number of reads per sample. Their prevalence in each ofthe samples was compared by producing a heatmap of read counts per OTUusing the heatmap function in the R statistics software package (R 2.13.0).In order to locate the OTUs in the BCR a V6-V8 SSU rRNA pyrotaglibrary [6] obtained directly from core samples removed from the same boreholes at the BCR as those that the enrichment samples came from was used.This was done to determine the representation and distribution in the ac-tual BCR of the putative arsenic-transforming microbes identified in theenrichment cultures. Both the enrichment culture and BCR core sample644.3. ResultsSSU rRNA pyrotag sequences were run through the Qiime suite of scriptsat the same time and clustered into OTUs so that read counts for the mostprevalent enrichment culture OTUs in the BCR core samples could be ob-tained. Phylogenetic trees with representative sequences for the prevalentenrichment OTUs, their most closely related cultured species and environ-mental clones, picked by using BLASTn to the NCBI nucleotide databases,plus sequences for known CAOs and arsenate reducing microbes, were con-structed by trimming the NCBI 16S rRNA sequences to the same region asthe pyrotag amplicons, aligning these using MUSCLE version 3.8.31 [39] tothe reverse complement of the pyrotag OTUs representative sequences fol-lowed by tree building with PHYML (nucleotide substitution model HKY,100 bootstraps) [55]. The final phylogenetic tree was imported into the In-teractive Tree of Life (iTOL) (http://itol.embl.de/) [82] together with datafor the number of reads per each OTU in the As(III), As(V) enrichmentsand BCR samples.Sequences obtained from Roche 454 Titanium sequencing of the en-richment cultures and the BCR core samples were submitted to the Na-tional Centre for Biotechnology Information (NCBI) Sequence Read Archive(SRA) under the Project Accession Number SRP038769. Sample, experi-ment and run accession numbers for each of the samples are tabulated inthe Supplemental Information Table C.1.4.3 Results4.3.1 Biochemical Reactor EnvironmentAt the time of sample collection, the environment in the bioreactor wasanoxic (DO ≤ 1.5 mg/L), almost neutral in pH (6.9) and at a temperatureof 7.1 ◦C. When samples were taken for enrichment culturing both field ORPprobes failed to stabilize. Measurements taken of the pore water at othertimes of the year revealed conditions to be reducing (ORP ≤ -112 mV). Sul-fate concentration was 470 mg/L, but no sulfide was detected. Total phos-phorous, nitrite and nitrate were low (Table C.2). Ammonia/ammonium-Nwas measured at 61 mg/L. The pore water contained measurable amounts ofdissolved arsenic (0.94 mg/L). Slurry core samples from the same bore holescontained high concentrations of arsenic in the solids (Table C.2) indicatingthat arsenic removal had been taking place. Zinc was another metal presentin high concentrations in the pore water and in the core solids (Table C.2).Mineralogical study of the BCR core solids revealed arsenopyrite (FeAsS)654.3. Resultsand sphalerite (ZnS) as the predominant As and Zn minerals, respectively(Chapter3),[71]. Iron oxide minerals were also detected in the BCR, but,for the samples processed, no As was found associated with these.4.3.2 Enrichment Culture Characterization: TaxonomicProfileNineteen enrichment cultures were sequenced, 10 from the As(III) mediumand 9 from the As(V) medium, after several passages from 2 to 8 (Table 4.1).A total number of 105,045 reads were clustered into 986 OTUs (97% homol-ogy cutoff) after quality screening. Diversity of the microbial communitiesassessed by rarefaction with the Chao-1 (Chao, 1984) and PD whole treediversity [41] indices, showed that the depth of sequencing was adequate todescribe the diversity of these enrichments (Figure C.2). The number ofobserved species (number of OTUs) per sample varied from 97 for the leastdiverse sample AsV-6P3 to 284 for the most diverse sample AsIII-2P3. Onaverage the As(III) enrichments were more diverse than those in the As(V)medium, but this was not statistically significant. The earlier passages wereslightly more diverse than the later passages (Figure C.2).First, the overall microbial community structure in all the cultures foreach arsenic species medium was characterized. The As(III) medium en-richment OTUs were assigned to a total of 29 phyla. Ninety eight per-cent of all As(III) medium reads were grouped into five primary phyla:Proteobacteria, Euryarchaeota, Bacteroidetes, Fibrobacteres and Firmicutes(Figure 4.1a). The Betaproteobacteria were the most predominant (66%)taxonomic group. Whereas for the As(V) enrichment cultures, of the to-tal 25 genera phyla represented, 99% of all reads were assigned to the fiveprimary phyla of Firmicutes, Proteobacteria, Bacteroidetes, Euryarchaeotaand Spirochaetes, with Clostridia being the dominant Class (Figure 4.1b).As(III) enrichment sequences were classified into 24 genera (Figure C.3a)containing 94.7% of all As(III) enrichment culture pyrotag sequence reads.Many of these sequences were classified as Simplicispira-related (51.4% oftotal reads). Methanogen taxa, Methanocorpusculum and Methanoregula,were highly represented among the top As(III) enrichment reads. AlthoughFirmicutes-related sequences were found in these enrichments, they werenot represented in the most prevalent genera. The top most highly repre-sented As(V) enrichment OTUs were taxonomically classified into 23 dif-ferent genera containing 97.3% of all As(V) enrichment culture reads (Fig-664.3. Resultsure C.3b). Sedimentibacter was the most highly represented genus (34% ofthe sequences) for the As(V) enrichments. Methanogens, Methanocorpus-culum-related, were also associated with As(V) enrichments but to a lesserdegree than in the As(III) cultures. Sulfur metabolism related bacteria,Desulfitobacterium and Desulfosporosinus were represented in the arsenateenrichments. Based on this phylogenetic analysis, the As(III) and As(V)enrichment microbial communities were quite distinct from each other.The most highly represented OTUs (97% homology cutoff) were chosenfor detailed study of the enrichment culture phylogeny. The prevalence ofspecific OTUs enriched in each one of the enrichment cultures are shown ona heatmap in Figure 4.2. As expected, the samples clustered according tothe medium that they were grown in, and to a certain extent due to theirpassage number. The As(III) cultures were all enriched with Simplicispira-related OTUs and clustered into two groups. Many of the early passagesalso contained some methanogen-related OTUs: Methanocorpusculum andMethanoregula boonei related. The late passages were grouped togetherwith some of the early passages and were distinguished by the presence ofAcidovorax, Ramlibacter, Ottawia and Lutibacter related OTUs.Firmicutes1% Alpha-proteobacteria7%Beta-proteobacteria66%Delta-proteobacteria0.33%Gamma-proteobacteria5%Bacteroidetes6%Euryarchaeota11%Other2%Fibrobacteres2%Proteobacteria;ARKICE-900%A)Firmicutes;Bacilli8%Firmicutes;Clostridia69%Firmicutes;Erysipelotrichi0.06%Alpha-proteobacteria0.3%Betaproteobacteria5%Delta-proteobacteria0.4%Epsilon-proteobacteria0.07%Gamma-proteobacteria3%Proteobacteria;Milano-WF1B-440.003%Proteobacteria;SK2590.03%Bacteroidetes9%Euryarchaeota3%Spirochaetes2% Other1%B)Figure 4.1: Most highly represented phyla in the (a)As(III) and (b) As(V)enrichment culturesFive Sedimentibacter related OTUs were highly prevalent in most of the674.3. ResultsAs(V) enrichments, except for AsV-6P3. Sulfate-reducing bacteria Desul-fitobacterium hafniense and D. chlororespirans related OTUs were morepredominant in the latter sample that had an overall lower diversity thanthe other As(V) cultures. Uniquely found in AsV-6P3 was a cluster of fiveOTUs: two UCT-N117 environmental group Betaproteobacteria related; twoPaenibacillus related and an unclassified Clostridium-related OTU. Desul-fitibacterium related OTUs were found in seven out of the nine As(V) enrich-ments. The two As(V) enrichments that did not have any Desulfitibacteriumrelated OTUs contained methanogen-related OTUs that were largely absentfrom most of the Desulfitibacterium containing samples. The two As(V) latepassages were most similar to each other and differed from the other As(V)cultures in that they contained also some Simplicispira related OTUs thatwere the most predominant genera in the As(III) enrichments. The lateAs(V) passages contained proportionally less of the Sedimentibacter relatedOTUs than the earlier passages and were distinguished by two highly preva-lent Proteiniclasticum related OTUs. The other As(V) enrichment culturethat stood out from the rest was AsV-5P3 in which three OTUs were pre-dominant. These were classified in the Lachnospiraceae family. Most As(III)and As(V) enrichments had unclassified Bacteroidetes Chlorobi group bac-teria related OTUs in common.4.3.3 Detection of Arsenic TransformationChemical speciation analysis performed on the As(III) enrichment culturesrevealed that the size of the As(III) peak at 0.98s on the HPLC chro-matogram diminished compared with that in the control (medium only)concomitant with the appearance of a As(V) peak at 6.66s (Table C.3).Quantification of the As(III) and As(V) concentrations could not be per-formed accurately on all of the As(III) medium samples due to either a verysmall As(III) peak or because the As(V) peak was very broad. Nevertheless,evidence suggested that arsenite oxidation was taking place in the As(III)enrichment cultures.Although some arsenate reduction was measured in controls (mediumonly) (Table C.4) showing that chemical reduction was taking place, formost As(V) medium enrichment cultures, As(V) concentrations were lessthan that of the control, indicating that microbes in the cultures contributedto arsenate reduction. Similarly, As(III) was detected in the As(V) mediumonly control but the As(III) concentrations in the enrichments were higherthan that in the control supporting the observation that microbes in the684.4. DiscussionAs(V) medium reduced As(V) to As(III) (Table C.4).4.3.4 Prevalence of Putative CAO and Arsenate ReducingMicrobes in the BCRSome of the SSU rRNA sequence reads from the enrichments and the BCRclustered together into the same OTUs. Figure 4.3 displays the phylogeneticdiversity of the enrichment culture OTU representative sequences with thecoloured bars indicating the log2(readcount) for each OTU in the As(III)enrichment (red), As(V) enrichment (green) and BCR core (blue) sequencelibraries. Cultured species most closely related to the enrichment OTUs areincluded on the phylogenetic tree, as are some selected 16S rRNA sequencesfor known CAOs (red text) and arsenate reducing microbes (green text). Op-erational taxonomic units from the As(III) and As(V) media were present inthe BCR, particularly those that were prevalent in both media such as themethanogen, Proteiniphilum and Spirochaeta related OTUs. Of the As(III)enrichment OTUs, those related to Acidovorax and Albidiferax spp. werepresent in the BCR. The predominant Simplicispira related OTUs identifiedin the As(III) medium were not detected in the BCR; however, there was anOTU in the BCR taxonomically classified as Simplicispira related (labeledas T3098 on the Tree in Suppl Info C.4). In contrast, Sedimentibacter re-lated OTUs that were highly prevalent in the As(V) growth medium (OTUs386, 822 and 30 on Figure 4.3 ) were represented within the BCR. Anotherpredominant As(V) enrichment OTU (249) that was Bacteroidales-relatedappeared to be highly represented in the BCR. Other As(V) OTUs presentin the BCR were the Atopococcus related OTU 335, the Paenibacillus relatedOTU 819 and the Betaproteobacterium UCT N117 related OTU 802. TheMethanomicrobia (Methanocorpusculum and Methanoregula) related OTUswere highly prevalent in the BCR and in the enrichments, but they were, asshown in Figure 4.2, eliminated in later passages.4.4 DiscussionThere was evidence for the potential of both biological arsenite oxidation andbiological arsenate reduction in the BCR. The change in chemical speciationthat was measured demonstrated that the enrichment cultures contained mi-crobes capable of oxidizing As(III) and reducing As(V). All enrichments weremixed cultures comprised of more than one operational taxonomic unit, so694.4. Discussionit is not possible to ascribe arsenite oxidation or arsenate reduction to anyparticular taxonomic group. Likely most of the enrichment culture microbeswere working in concert to oxidize As(III) or reduce As(V) as part of detox-ification mechanisms in order to survive in the media or to obtain energy forgrowth. Arsenic resistance is ubiquitous in nature and a wide diversity ofmicroorganisms contain homologs of genes known to be involved in arsenicdetoxification [60].The enrichment cultures contained microbes closely related to knownCAOs and arsenate reducing microbes. For instance, a Rhizobium relatedOTU (number 740 in Figure 4.3) was closely affiliated with CAO isolatesNT-25 and NT-26 obtained from arsenopyrite (FeAsS) rock taken from agold mine [140]. NT-26 was found to grow aerobically in minimal mediumwith arsenite as the electron donor, and CO2 or HCO−3 as carbon source. Assuch this was the first isolate shown to derive energy from growth on arsen-ite. NT-26 was also capable of arsenite oxidation in the presence of organicmatter. NT-25 and NT-26 form their own branch within the Agrobacterium-Rhizobium branch of the Alphaproteobacteria. Whereas the Rhizobium re-lated OTU identified in the enrichments in this study was more closely re-lated to the cultured species Rhizobium selenitireducens strain B1, which wasisolated from a bioreactor reducing selenite to elemental selenium [64]. Thesequence used for taxonomic classification of OTU 740, Rhizobium sp. p49(accession number HQ652582) (Figure C.4) was isolated from a magnetitemine, but no further information on its metabolic potential was available atthe time of writing. Species within this genus are metabolically versatile andtheir capability for transformation of metals contributes to detoxification ofsoils and enables metal-resistance in plants.Several Thiomonas spp. strains that use arsenite as an electron donorhave been isolated from mine sites, such as the two included on the phyloge-netic tree in Figure 4.3 [9][38]. Not all Thiomonas spp. strains are arsenicresistant or have the aox operon despite their phylogenetic similarity [20].An As(III) medium OTU (536) that was classified in the Comamonadaceaefamily was phylogenetically related, although only distantly, to the CAOThiomonas strains. Its closest cultured species relatives were Aquabacteriumspp., which are ubiquitous in the oligotrophic environment of drinking wa-ter. The closest environmental clone relatives to OTU 536 were found in agypsum-treated oil sands tailings pond [129], river sediments associated withmining (accession number KC541171, 98% homology) and a freshwater ironseep (JQ906323, 99%), all of which are metal contaminated environments.704.4. DiscussionThus OTU 536 might belong to another genus within the Comamonadaceaefamily able to oxidize arsenic.The two Acidovorax related OTUs, numbers 783 and 621 in Figure 4.3,that were very predominant in the As(III) medium, including also the latepassages, were closely related to Acidovorax sp. strain NO1, which was iso-lated from arsenic-contaminated soil at a gold mine. Acidovorax sp. strainNO1 was found to oxidize As(III) to As(V) under both under aerobic andanaerobic conditions [62]. An aio arsenite oxidation operon was revealedin its genome [62]. Nitrate and nitrite served as electron acceptors underanaerobic growth. This suggests that it would be likely to survive withinthe anoxic conditions of the BCR. Indeed, of all of the As(III) enrichmentOTUs, the two Acidovorax related OTUs were the most predominant in theBCR cores. This provides evidence that the potential for biological As(III)oxidation exists in the BCR with Acidovorax related microbes likely CAOs.In contrast to the As(III) medium OTUs, most from the As(V) mediumwere not closely related to any previously identified arsenate reducing mi-crobes, except for the Desulfitobacterium related OTUs that were predomi-nant in both the early and late As(V) medium passages. Interestingly, theAs(V) enrichments that did not have Desulfitobacterium related sequences,or those in which they were present in low amounts, contained Methanocor-pusculum and Methanoregula related sequences (Figure 4.2) indicating thatthese two groups may compete for electron donors and be mutually exclusive.OTUs 605 and 599, and 451 and 624 were classified as D. chlororespiransand D. hafniense, respectively, which are spore-forming sulfate-reducers inthe Firmicutes phylum. D. auripigmentum was the first sulfate-reducingbacterium shown to reduce As(V) as well as sulfate, and even did so prefer-entially [112]. Since then other Desulfitobacterium strains have been foundthat reduce arsenate, plus other metals such as As(V), Fe(III), Se(VI), andMn(IV), as well as sulfur species. Strain GBFH [114] from a lake affected bymining was also closely related to cultured species D. hafniense DCB2, thattoo is capable of respiration on metals, as well as chlorinated aromatic hy-drocarbons. These microbes are implicated in mobilization of arsenic. Theother sulfate-reducing bacterium found in the As(V) enrichments was Desul-fosporosinus related. Sulfate-reducing microbes related to this genus wereassociated with arsenic sulfide precipitation in a bioreactor [7] indicatingthat the presence of As(V) reducers does not necessarily imply arsenic mo-bilization. Co-current with sulfidogenesis, arsenic minerals can be formed,such as arsenopyrite [71]. No As(V) enrichment OTUs were related to the714.4. DiscussionDeltaproteobacteria sulfate reducers known to also reduce arsenic, such asstrains Desulfomicrobium BenRB and Desulfovibrio BenRA. Although thesulfate-reducing Clostridia related microbes were prevalent in the As(V) en-richments, they were undetectable in the actual BCR. The rarity of sulfate-reducers in the BCR was surprising given the high sulfate concentrations(greater than 600 mg/L in the influent) [95]. Sulfate reducers that weredetected by pyrotag sequencing in the BCR were Desulfovibrio and Desul-fobulbus related and the formation of sulfide minerals inside the BCR verifiedthat sulfate-reducers were active during some periods over the BCRs lifetimeif not at the time of sampling [71].The enrichments in this study also revealed taxonomic groups not previ-ously known to have the potential for arsenite oxidation or arsenate reduc-tion. In contrast to the Thiomonas and Acidivorax related OTUs discussedpreviously, the most prevalent As(III) medium OTUs classified as Simpli-cispira, a genus also within the Comamonadaceae family, were not related toany previously characterized CAOs. The closest cultured relatives of OTUs474 and 12 (Figure 4.3) were species associated with activated sludge removalof phosphorous [88]. Other highly prevalent Simplicispira related OTUs, 673and 92, were closely related to a phototrophic purple non-sulfur bacteriumRhodoferax fermentas strain FR2 and another Rhodoferax species knownto reduce ferric iron [45]. The strain to which the prevalent SimplicispiraOTUs (12, 474 and 298) were taxonomically assigned, BetaproteobacteriumNOS8 (AB076846), belongs to a group of Comamonadaceae strains capa-ble of degrading complex organic molecules, such as biodegradable plastics,in combination with denitrification [70]. Environmental clones closely re-lated (100% homology) to the As(III) medium Simplicispira related OTUswere found in high sulfur environments (JQ723662) and metal contaminatedaquifers (KC113247). Therefore there is some evidence that microbes withinthis branch of Comamonadaceae have versatile metabolisms that might in-volve metal transformations. However, this is the first report of Simpli-cispira related microbes being involved in arsenic oxidation. Nevertheless,the Simplicispira related OTUs that were prevalent in the As(III) mediumwere not detected in the BCR core samples, indicating that they were veryrare in that environment, whereas the Acidovorax -related CAOs were moreadapted to the BCR conditions.Many OTUs that were predominant in the As(V) medium were notclosely related to any known arsenate reducing microbes. The Sedimen-tibacter related OTUs that were the most prevalent As(V) reducers, were724.4. Discussiononly very distantly related to Clostridium sp. OhILAs (Figure 4.3), or Al-kaliphilus oremlandii as strain OhILAs has been named, which is an ar-senic respiring spore-forming bacterium isolated from the Ohio River [46].The closest cultured relative to OTUs 30, 822 and 386, Sedimentibactersaalensis, was isolated from a 2,4,6-trichlorophenol-dehalogenating enrich-ment culture [19]. Other closely related environmental clones also came fromdechlorinating environments (Figure C.4). Interestingly, one closely relatedenvironmental clone came from an iron reducing enrichment culture derivedfrom a creek contaminated with heavy metals and uranium (HM992484,[22] ) (Figure C.4). In this enrichment culture, Sedimentibacter as well asDesulfosporosinus and Desulfitobacterium were the predominant genera rep-resented echoing the communities enriched in the As(V) enrichments of thisstudy. That creek contained high concentrations of copper, cadmium, cobaltand zinc. Thus, there is strong evidence that the Sedimentibacter relatedOTUs in the As(V) enrichments might be involved in metal transformationsor detoxification in order to survive those environments. The OTUs 30, 822and 386 were also found in the BCR, indicating that the potential for As(V)reduction exists in the BCR, which might, together with iron reduction,contribute to As remobilization.Betaproteobacterium UCT N117 classified OTU 802 was closely relatedto acid-tolerant Clostridium uligenosum strain CK55, and was detected inthe BCR as well as in the As(V) enrichment. Clostridium uligenosum hasnot been linked with arsenate reduction previously, to our knowledge. TheAtopococcus classified OTU (335) that was present in both the As(V) en-richments and in the BCR was closely related to cultured species in the Tri-chococcus genus, which consists of filamentous species that dominate sewagesystems (Figure C.4). Cow manure was used as inoculum for the BCR,which may have introduced feces-related microbes, some of which could beresistant to arsenic. One particular As(V) medium culture, AsV.6P3, wasenriched with Paenibacillus related OTUs, which were not found in any ofthe other As(V) enrichments. Species within the Paenibacillus genus arewidespread and have great potential for biotechnology applications sincethey produce extracellular enzymes and antibiotics. The latter might ex-plain why this enrichment culture was so different from the rest. Inter-estingly, the Desulfitobacterium related OTUs were the only ones that co-cultured with the Paenibacillus related OTUs. Paenibacillus are known tobe plant growth promoting rhizobacteria since they facilitate nutrient de-livery to the plant, and some species are metal (e.g. chromium) resistant[69]. Presence of Paenibacillus related OTUs in arsenate reducing enrich-734.4. Discussionments suggests another possible role for these very versatile microbes. Pro-teiniclasticum related OTUs were predominant in the late passage As(V)enrichments and to a lesser extent in some of the late passage As(III) en-richments. The only known Proteiniclasticum species was isolated fromyak rumen undergoing cellulose biodegradation [168]. The closest culturedspecies related to OTU 657 was Youngiibacter fagilis, which was isolatedfrom coal bed methane process water [81]. Other amino acid utilizing gen-era were represented in both the As(V) and As(III) media as well as in theBCR, such as the Proteiniphilum classified OTUs 491 and 1136. Like theProteiniclasticum related OTUs the Proteiniphilum associated OTUs werealso phylogenetically similar to isolates and environmental clones from de-graded oil reservoirs (e.g. Petrimonas sulfuriphilia strain B, [54]). Previousresearchers of these environments have suggested that the co-occurrence ofFirmicutes and Methanogens in the microbial communities of these sitessuggests a symbiotic relationship between two groups. Indeed both groupsare prevalent within the BCR. However, in the enrichments, amino acid de-grading taxa did not co-occur with methanogens, rather Desulfitobacteriumrelated OTUs were found together with Proteiniclasticum related OTUsin the same cultures. Nevertheless, both methanogenic and fermentativetaxonomic groups were highly represented in both As(III) and As(V) en-richments and in the BCR. Spirochaetes and Bacteroidales were anothertwo Firmicutes taxa represented in both enrichments and the BCR. Methy-lation of metalloids by methanogens is known to occur [155][101] and theBCR methanogens, Methanocorpusulum and Methanoregula related OTUs,might methylate arsenic to volatile forms, which would be one likely reasonfor their survival in the As rich media. If so, they would be possibly signif-icant contributors to arsenic bioremediation in the BCR as they were themost predominant taxa in the BCR [6].This work suggests that the capability to oxidize arsenite and reducearsenate could be more widespread among many diverse metabolic groupsthan previously thought. Other genetic studies of enrichments from ar-senic contaminated environments have corroborated this. The characteri-zation of aoxB genes from enrichments of samples taken from an arsenic-contaminated mine site revealed much phylogenetic diversity implicatingmany groups hitherto not known to transform arsenic [60]. The microbialcommunity of an arsenic contaminated stream sediment was just as diverseas that in a relatively pristine nearby stream sediment [130]. Although thepresent study revealed additional taxonomic groups suspected of arseniteoxidation and arsenate reduction, the caveat should be mentioned that the744.4. Discussionenrichments were mixed cultures, and the measured arsenic transformationscannot be directly attributed to any particular organisms. Nevertheless, wecan hypothesize that survival in the arsenic-rich media necessitated geneticmachinery for detoxification, which might involve arsenic transformation.This study suggests that putative CAOs and arsenate reducing microbeswere present in the BCR and that the potential for biological arsenic cyclingexisted, akin to the model suggested by Stolz and Oremland for Mono Lake[150]. Firmicutes and Bacteroidetes related taxonomic groups were predom-inant in the BCR [6] and products of fermentation such as acetate, butyrateand propionate, likely would have been available as electron donors for het-erotrophic arsenate reducing microbes. Nitrate was present in the feed tothe BCR and would have been available as an electron acceptor for CAOs.Arsenic cycling would mean that both species were present and althoughadsorption affinities and release rates differ between As(III) and As(V),mineral formation from either chemical species is possible. Both As(III)and As(V) bind to iron (hydr)oxides: As(III) binds to a greater extent, butAs(V) remains more strongly bound [156]. Biological As(V) reduction wasfound to promote arsenic release from iron (hydr)oxides in column leachingexperiments [156]. Iron oxides were detected in the BCR solids, but, for thesamples that were analyzed, no evidence for their association with arsenicwas found [71]. Iron sulfides such as pyrite (FeS2) were present in the BCR[71]. As(III) is known to adsorb with high affinity to iron sulfides eventuallyforming covalent bonds with both Fe and S suggestive of arsenopyrite [18].Thus formation of As(III) does not necessarily lead to arsenic mobilizationin all anaerobic environments.754.4. DiscussionAsV.6P3AsV.4P8AsV.3P8AsV.5P3AsV.3P3AsV.1P2AsV.4P3AsV.2P3AsV.1P3AsIII.3P3AsIII.2P3AsIII.6P3ASIII.1P2AsIII.1P3AsIII.5P3AsIII.4P3AsIII.5P7AsIII.1P7AsIII.4P8Simplicispira_beta_proteobacterium_NOS8_OTU298Rhizobium_sp_p49_OTU740Simplicispira_beta_proteobacterium_NOS8_OTU12Simplicispira_beta_proteobacterium_NOS8_OTU474Pseudomonas_fluorescens_OTU76Uncultured_Bacteroidetes_Chlorobigroupbacterium_OTU1136Uncultured_Bacteroidetes_Chlorobigroupbacterium_OTU491Sedimentibacter_sp_MO_SED_OTU30Sedimentibacter_uncultured_Peptostreptococcaceae_OTU331Sedimentibacter_unculturedPeptostreptococcaceae_OTU653Sedimentibacter_unculturedPeptostreptococcaceae_OTU822Sedimentibacter_unculturedbacterium_OTU386Desulfosporosinus_sp_A10_OTU313Lachnospiraceae_unculturedbacterium_OTU314Lachnospiraceae_unculturedbacterium_OTU700Lachnospiraceae_unculturedbacterium_OTU111Sedimentibacter_unculturedbacterium_OTU773Proteiniclasticum_unculturedbacterium_OTU657Proteiniclasticum_unculturedbacterium_OTU176Clostridium_unculturedbacterium_OTU67UCT_N117_unculturedbetaproteobacterium_OTU562Paenibacillus_sp_D9_OTU436Paenibacillus_ehimensis_OTU819UCT_N117_unculturedbetaproteobacterium_OTU802Desulfitobacterium_hafniense_DCB2_OTU699Desulfitobacterium_chlororespirans_OTU605Desulfitobacterium_hafniense_DCB2_OTU624Desulfitobacterium_chlororespirans_OTU599Desulfitobacterium_hafniense_DCB2_OTU451Methanoregula_boonei_6A8_OTU40Methanoregula_boonei_6A8_OTU226Archaea_Methanocorpusculum_unculturedarchaeon_OTU36Archaea_Methanocorpusculum_unculturedarchaeon_OTU425Lutibacter_Flavobacteriaceaebacterium_cloneMT1779_OTU720Simplicispira_unculturedbacterium_OTU92Simplicispira_unculturedbacterium_OTU673Ottowia_unculturedbacterium_OTU77Acidovorax_delafieldii_2AN_OTU783Ramlibacter_unculturedbacterium_OTU114Acidovorax_unculturedbacterium_OTU621Figure 4.2: Heatmap of prevalent 97% homology cut-off OTUs (rows) inthe enrichment culture samples (columns). The OTU numbers and theirtaxonomic assignments are presented on the right hand vertical axis, andthe sample names are given beneath the lower horizontal axis. Heatmapwas produced using the heatmap function in R version 2.15.2. Dendogramsare based on the Euclidian distance matrices for OTUs and samples, respec-tively.764.4. Discussion1545 AlbidiferaxSimplicispira psychrophila strain CA1 NR_028712Simplicispira metamorpha strain DSM 1837 NR_044941474 Simplicispira1639 Comamonas1252 SimplicispiraSimplicispira limi strain EMB325 NR_043773Rhodoferax fermentans strain FR2 NR_025840673 SimplicispiraRhodoferax ferrireducens T118 strain DSM15236 NR_074760783 Acidovorax621 AcidovoraxAcidovorax caeni strain R24608 NR_042427Acidovorax temperans strain PHL NR_028715Aquabacterium commune strain B8 NR_024875536 ComamonadaceaeThiomonas arsenivorans strain b6 AY950676Thiomonas NO115 AY455807Alkalilimnicola ehrlichii MLHE1 NR_074775Pseudomonas gessardii strain CIP NR_02492876 PseudomonasShewanella ANA3 AF136392Citrobacter TSA1 AF463533Thermomonas fusca strain R10289385 ThermomonasDesulfomicrobium BenRB AF131233Desulfovibrio BenRA AF131234Rhizobium selenitireducens strain B1 NR_044216740 RhizobiumNT25 AF159452NT26 AF159453Sulfurospirillum barnesii AF038843Sulfurospirillum barnesii SES3 NR_102929Sulfurospirillum arsenophilum strain MIT-13 NR_044806Wolinella succinogenes AF463534Chrysiogenes arsenatis DSM 11915 NR_029283822 Sedimentibacter386 Sedimentibacter30 SedimentibacterSedimentibacter saalensis strain ZF2 NR_025498Clostridium OhILAs DQ250645Clostridium propionicum strain X2 NR_029269700 Lachnospiraceae657 ProteiniclasticumYoungiibacter fagilis JF262039Anaerobacter polyendosporus strain PS-1 NR_026496Clostridium disporicum strain DS1 NR_026491Clostridium uliginosum strain CK55 NR_028920802 Beta-proteobacteria UCT N117Desulfosporosinus lacus strain STP 12 NR_042202313 DesulfosporosinusDesulfitobacterium chlororespira NR_026038451 Desulfitobacterium605 DesulfitobacteriumDesulfitobacterium hafniense AJ307028Desulfitobacterium hafniense DCB NR_074996Desulfotomaculum auripigmentum U85624 Bacillus macyae strain JMM4 AY032601Bacillus arseniciselenatis AF064705Bacillus HT1 AF463535T. patagoniensis strain PMagG1 NR_041841T. collinsii strain 37AN 3 NR_042061335 AtopococcusTrichococcus flocculiformis strain DSM 2094Trichococcus pasteurii strain KoTa2 NR_036793 Bacillus selenitireducens MLS10 NR_028707Paenibacillus ehimensis strain K NR_025666819 PaenibacillusThermus HR13 AF384168Sphaerochaeta globus str Buddy NR_074808178 SpirochaetaParabacteroides merdae strain JC NR_041343249 Bacteroidales S247Proteiniphilum acetatigenes strain TB107 NR_043154491 Proteiniphilum1136 ProteiniphilumPetrimonas sulfuriphila strain B NR_042987Lutibacter litoralis strain CLTF NR_043301720 LutibacterMethanoregula boonei 6A8 strain NR_074180 226 Methanoregula boonei425 MethanocorpusculumMethanocorpusculum labreanum NR_074173Methanocorpusculum bavaricum NR_042787Alphaproteobacteria Delta-Proteo Gammaproteobacteria Betaproteobacteria Epsilonproteobacteria Bacili Clostridia Other0 16 256 4096Read count scaleBCRAs(V)As(III)Figure 4.3: Prevalence of putative CAO and arsenate reducing microbes inthe BCR. Phylogenetic tree of 97% homology cut-off OTUs represented inboth the As(III) and As(V) enrichments as well as the biochemical reactor,together with closely related cultured species and some known CAOs (redfont) and arsenate reducing microbes (green font). Bar length representsthe log2(read count) of that OTU in the biochemical reactor (BCR) (blue),As(V) medium enrichments (green) and As(III) medium enrichments (red)77Chapter 5Functional MetagenomicScreening of Arsenic5.1 ScopeTo identify microbial-mediated mechanisms for removing metals from thebiochemical reactor, metabolic responses of microorganisms against metalsand metalloids need to be investigated. Since some metal-microbes inter-actions are enzymatic and therefore genetically coded (Figure 1.3), metage-nomics analysis can be used to access microbial community metabolic po-tential in the absence of cultivation, and functional metagenomic screenscan be developed to identify specific metabolic processes in a heterologoushost system such as E.coli.Upon exposure to heavy metals such as arsenic, microorganisms respondeither by sequestering or immobilizing metals or by enhancing their solu-bility (thesis hypothesis). In addition, methylation, periplasmic arsenatereduction and arsenite expulsion from the cell are reactions catalyzed en-zymatically and associated with catabolic operons (Figure 1.3). For thework described in this Chapter, the aim was to identify fosmid clones thatharbored genes and metabolic pathways mediating arsenic detoxification.These heterologous genes, when expressed in E.coli, allowed growth in highconcentration of arsenic and thus could potentially contribute to arsenicremoval/detoxification in biological-based chemical reactors. The more in-formation about the microbial communities and their metabolic powers thatwe gain, the more insight about the system function we achieve, which couldhelp to improve bioreactor performance and troubleshooting the systems.It should be noted that different culture media were used to identifyarsenite oxidizers and arsenate reducers (Figure 1.3-biotransformation) asdiscussed in Chapter 4. Indeed, with purely phylogenetic studies (Chap-ter 4), function can only be inferred based on what is known about closely785.2. Experimental Sectionrelated cultured species, which is highly inferential. Identifying homologsfor arsenic genes goes further in demonstrating actual metabolic potential.Therefore, here emphasis is placed on the identification and quantificationof these genes associated with a specific phenotype trait (arsenic resistance)using a combination of marker gene identification and functional metage-nomic screening.In this Chapter the objectives were (1) to screen the fosmid insert se-quence ends for functional arsenic gene homolog anchors (sequence basedscreening) (2) to screen the large insert DNA metagenomic library for ar-senic resistant clones, examine the gene expression potential, e.g. phenotypecharacterization and identify functional genes (function based screening), (3)to perform a taxonomy analysis of the arsenic resistance clones and finally(4) to postulate on the possible mechanisms for arsenic removal in the BCR.5.2 Experimental Section5.3 Sampling Site, Fosmid Library ConstructionThe biochemical reactor (BCR) selected for this study received metal andsulfate-rich seepage from a smelter-waste landfill [95]. Samples were col-lected from cores of the reactor matrix at three different locations and onthree different days: 22 July 2008, 19 April 2009 and 20 October 2009.Corresponding fosmid libraries were constructed from these core samples,which are cataloged as MG1, MG2 and MG3 (Chapter 2, Table 2.1). Thehigh molecular weight (HMW) DNA extracted from homogenized core sam-ples were used to construct fosmid libraries (Done by Marcus Taupp, Dr.Hallam laboratory, Department of Microbiology and Immunology at UBC).Environmental DNA was cloned into the pCC1 copy control system and ex-pressed in E.coli EPI300 (Epicentre, Madison, WI) as previously described[153]. Three fosmid libraries yielded a total of 18432 fosmid harboring cloneswith average insert size of 42 kilobase (kb) pairs. Bi-directional Sanger end-sequencing was performed on all fosmid clones at the Canadian MichaelSmith Genome Sciences Centre, Vancouver, BC, Canada. (More details onfosmid end sequencing can be found [100]. See Table A.1 for chemical anal-ysis of the core solid samples, Table A.2 for carbon analysis and Table C.2for chemistry of the pore water during the sampling times.)795.4. Sequence- Based Fosmid Library Screening5.4 Sequence- Based Fosmid Library Screening5.4.1 Gene Finding and Open Reading Frame PredictionBased on previous studies, a database of arsenic resistance genes was cre-ated [148][106][1][59][60]. This included a total of 382 gene sequences withaverage length of 823 nucleic acids and 274 amino acid polypeptides. Thegene sequences were chosen from eight As resistant classes that belonged todiverse microorganisms (arsenite oxidation, ars operon regulators, periplas-mic As(V) reduction, arsenite efflux pumps, dissimilatory As(V) reduction,phosphate transporters, metal transporters and methylation). The fosmid-end sequences were queried against the custom As gene database using thetblastn program included in the NCBI blastall v.2.2.20 package. From se-quence comparison, we picked the top hits that have E-value (expectationvalue) less than 1E-06 and identity scores of more than 30%. The E-valueis a parameter that describes the number of hits one can expect to see bychance when searching a database of a particular size. The lower the E-value the more significant the match between the two translated proteins(http://www.ncbi.nlm.nih.gov/blast).5.4.2 Phylogenetic AnalysisAn algorithm (Section F) was written to combine multiple fosmid-end genes,if they were identical, into one tblastn entry and each unique fosmid-endgene was compared to a custom arsenic database. The protein sequencespredicted from Prodigal (http://prodigal.ornl.gov/) [65] were aligned withClustal X 2.1. Neighbor joining (NJ) clustering algorithm was used to createphylogenetic tree.5.5 High Throughput Functional ScreeningMinimum inhibitory concentration (MIC) of arsenic was determined forE.coli prior to screening of clone libraries. The viability of E.coli was moni-tored in the range of 0.1 - 6 mM for As(III) and 1-270 mM for As(V). AfterMIC analysis for E.coli, 5.5 mM of As(III) and 130 mM of As(V) were usedto select arsenic resistant fosmid clones. The MIC determination in L.B.agar medium is given in Section G.2. Using a Q-Soft XP gridding robot,fosmids were inoculated into agar medium in an extra large (XL) square805.5. High Throughput Functional Screeningbiodish and incubated for a week. After recovery of fosmids that grew on130 mM As(V), their specific growth rates during the logarithmic phase werecalculated from optical density (OD) measurements at 600 nm.5.5.1 Phenotype Characterization of Resistant ClonesBased on optical density measurements, growth curves were generated forfosmids at varying As(V) concentrations (0, 20, 30, 40, 50 and 60 mM) dur-ing 20 h time period. The specific growth rates for each positive fosmidwere determined from the slope of growth curves from the exponential re-gion. The performance index is defined as µ MAX n / µ MAX a, where ’n’and ’a’ are the slopes of growth curves when fosmids are grown with andwithout arsenic (control) respectively. Further details for the determinationof growth rates of fosmids are given in Section G.3. After characterizinggrowth phenotypes of resistant clones, 92 fosmid clones were sequenced toidentify genes that conferred arsenic resistance.5.5.2 Full Fosmid SequencingOnce active clones were identified, fosmid DNA was extracted using theFosmidMax DNA preparation kit (Epicentre, Madison WI) according to themanufacturer’s instructions and further treated with PlasmidSafe DNase(Epicentre) to remove contamination of E.coli chromosomal DNA. To avoiddegradation of nicked fosmids, the DNase was inactivated after use by heat-ing to 70◦C for 30 min. DNA concentrations were measured using Quant-iTPicoGreen (Invitrogen,Carlsbad,CA). For full fosmid sequencing, 500 ng ofeach fosmid was sent to the Canadian Michael Smith Genome Sciences Cen-tre (Vancouver, BC). Fosmid samples from 92 arsenic resistance clones werebarcoded and sequenced on a single lane of an Illumina GAIIx sequencer(Illumina, San Diego, CA). Contigs were assembled for each well using thebarcoded sequences with ABySSv1.2 [145]. In the next step, a .fasta file wascreated from all contigs (from all wells), which have size larger than 5kb.The vectors were trimmed out and the contigs were compared to the fosmidends using BLASTN to verify their identity with the particular fosmid.The full fosmid sequences (.fasta files) were translated into amino acidsequences using Prodigal v1.20 [65], which yielded a total of 2880 ORF from815.5. High Throughput Functional Screening89 full fosmid sequences (i.e., the number of prodigal output per full fos-mid sequences was varied from 22 to 53 ORFs). For gene prediction andannotation, the ORFs were compared against the custom arsenic resistancegenes database using NCBI BLASTP analysis. Arsenic resistance gene as-signments were selected with bit score (50) and identity (30%) cutoffs. Thegene assignments were also compared with MetaPathway pipeline annota-tions (http://hallam. microbiology. ubc. ca/ MetaPathways/ ) [73]. In thispipeline, translated ORFs were queried against user-defined reference pro-tein databases including KEGG, COG, RefSeq, and MetaCyc. In addition,the gene assignments were compared with HMMER (hidden Markov mod-els) annotations (http://hmmer.janelia.org/search/phmmer) [43]. Finally,arsenic resistance gene assignments were retained that are common in mostof these programs. The position of the identified genes on the fosmids wasrepresented using a diagram generated with the computer program Circos(http://circos.ca/) [75].5.5.3 Phylogenetic AnalysisIn order to establish taxonomic assignments for ORFs, the DNA sequenceswere compared using Blastp to a database of reference sequences (NCBI-NR), which resulted in a 22.6 MB file containing approximately 11 thousandblast hits. The resulting blast file was later directly imported into Meganversion 5.1.5. Megan is the MEtaGenome Analyzer program and it canbe used to analyze the taxonomy and functional gene/protein content of ametagenomic dataset. MEGAN uses blast output (BlastX, BlastN, BlastP,etc) to bin all reads of a given dataset into different NCBI taxonomy groups.MEGAN parse the blast output file and automatically perform a taxonomicclassification, and if desired, also a functional classification using the KEGG,COG, metacyc classification. To deal with inconsistencies in the taxonomicassignment for each ORF, the Megan approach involves setting up of thresh-olds for different parameters including min-score, top-percent, min-support.First, alignment must achieve a min-score threshold to be considered in thecalculations. For high length reads (length 800 bp), a min-score of 100 issuitable. Second, the top-percent filter sets a threshold to help distinguishbetween hits based on sequence identity and homology. A useful range ofvalues is between 10 to 20 percent. Third, the min-support parameter isused to specify the minimum number of reads that must be assigned to ataxon before that taxon is considered. Following these analyses, all readsthat do not meet these cut-offs, and hence not assigned to an existing taxon,825.6. Resultsare reassigned to a special taxon. In a taxonomic analysis, reads are mappedto nodes of the NCBI taxonomy, using the LCA (Lowest Common Ancestor)algorithm. The results of the metagenome data (.rma file) can be viewedand inspected with MEGAN. Here, a threshold of min score=120, top per-cent=10 and min support=5, which enables conservative assignments, wasused and .rma file, which is about 50 MB in size was obtained. In addi-tion, full fosmids sequences were submitted individually to ML-TreeMap forfurther phylogenetic analysis. ML-TreeMap searches sequences for suitablephylogenetic marker genes, and uses maximum likelihood analysis for phy-logenetic analysis (http://mltreemap.org/).5.6 ResultsFunction-based metagenomic screening approaches include heterologous ex-pression, in which clones that express the desired function are identified.In this study, E.coli Epi300 was selected as a fosmid host. MIC analysisdemonstrated that E.coli Epi300 host was able to grow at conditions up to4 mM and 270 mM of As(III) and As(V), respectively (Table D.1). Theincubation time is important for acquiring arsenic resistance, as the hostwas found to tolerate higher arsenate concentration when incubated longer.The hosts harboring fosmids were grown on XL L.B. agar plates containing5.5 mM of As(III) and 130 mM of As(V) for two weeks. While no growthwas observed on the As(III) plates, we found colonies growing on As(V)plates after two weeks.The 104 positive fosmid clones identified to confer As(V) resistance weretransferred to liquid medium, where their resistance to arsenic was veri-fied and phenotypes (e.g. growth rate) were characterized. Fosmid clonesexhibited different phenotypes at different As(V) concentrations. The ma-jority of fosmid clones grew in the 20 mM As(V) liquid medium, Table D.2.Figure D.1 presents a few examples of their growth curves over 20 ≈ h.Some fosmid clones, such as FOS6205-B1, were highly resistant to As(V)and grew in concentrations up to 60 mM As(V), Figure D.1. In addition,some fosmid clones were fast growing, i.e., started growing within 5 hoursof inoculation, such as fosmid Area1-F77. Other fosmids such as FOS6206-B10 demonstrated a higher resistance toward 50 mM arsenate in comparisonwith lower As concentrations (Figure D.2, Figure D.1). The growth perfor-mance for 104 fosmid clones identified from screening of three metagenomic835.6. Resultslibraries are summarized (Table D.2).Arsenic resistance pathways and the associated genes are listed in theintroduction (Figure 1.3, Table 1.2). For each pathway, the frequenciesof the genes in the As(V) resistant full fosmid DNA sequences and in thefosmid-end As-related sequences were compared (the raw data are presentedin Tables D.3 and D.4). Screening results showed that metagenome data ob-tained from analysis of BCR samples are enriched in arsenic resistance genes(Figure 5.1, Figure D.3 and Figure D.4). Genes potentially related to bioac-cumulation (phosphate transporters (e.g. pstA, phnC ), metal transporters(e.g. Chr, Znu), periplasmic arsenate reduction (e.g. arsC, arsR, arsT ), ar-senite efflux pumps (arsA, arsB) and arsenite oxidation mechanisms ( aoxR,aoxB) were detected. In contrast, genes pertaining to volatilization (e.g.arsRM ) and dissimilatory arsenate reduction (e.g. arrA) mechanisms (Fig-ure 5.1) were not detected as frequently. It should be noted that the bio-mineralization and bio-adsorption mechanisms could not be investigated byfunctional screening analysis since they are not enzymatically driven.Position of the genes related to the ars operon, including ars A/B, ars C,regulators and also genes that were assigned as arsRM (volatilization mech-anism), are presented in Figure 5.2. Since this operon activates by arsenic(V) uptake either specifically or nonspecifically through transporters, genesassociated with the transporter systems are also included in the Figure 5.2.Genes assigned to these two classes of resistance from three metagenomiclibraries are presented in Figure 5.2, Figure D.10 and Figure D.11.Although, for the most part, each class of arsenic resistance genes (In-troduction, Figure 1.3, Table 1.2) were found on separate fosmids, therewere a few examples of fosmids that harbored ars related genes togetherwith transporters genes. For instance, carsR, arsRM and phnC were identi-fied in FOS6209-M3 (Figure D.11). The genebank blast protein identity ofthese ORFs were 67%, 61% and 86% respectively, thus there was high confi-dence in their annotation. Presence of these genes conferred higher toleranceto arsenate (grew in 60 mM arsenate), as indicated by phenotype charac-terization of the fosmids (Table D.2), Methanocorpusculum labreanum wasthe taxonomic assignment for this fosmid. Metagenomic analysis showedthat this organism possessed both As(V) reduction and As(III) volatiliza-tion potential (Figure D.11, Table E.1). In addition, ORFs identified inFOS6205-B4 and FOS6206-O13 were closely related to the ars operon anda transporter (phnC). The blast identity of two ORFs that were assigned845.6. Resultsto the ars operon determined 86% and 89% in both fosmids. 48% identitywas also obtained for ORFs assigned phnC. Although, these fosmids werepartially similar in term of DNA sequence (Figure D.5), they did not haveidentical growth characteristics. FOS6205-B4 was able to grow in low con-centrations of arsenate only, whereas FOS6206-O3 showed growth at higherarsenate concentrations (50 mM) Table D.2. All fosmids sequenced from thethree libraries are given in Figure D.5, showing regions of nucleotide homol-ogy at greater than 90% similarity across intervals of more than 300 bp.Genes potentially related to the aox operon were also identified (Fig-ures D.7, D.9 and D.8). The aoxB (65% identity), oxyS and aoxR geneswere detected in FOS6231-G15. Notably, the genes conferred a low growthrate phenotype in expression host in the presence of arsenic(V). The resultswere consistent with other fosmids such as FOS6232-A3, FOS6230-F14 andFOS6239-AO7 that contained aox B/A genes. Indeed, fosmid clones thatwere transferred to As(III) agar plates did not show growth during twoweeks of the experiments. However, when the fosmid clones were grown indifferent concentrations of As(V) liquid media growth was detected only in20 mM As(V) (low As(V) concentration), Table D.2. The identification ofthese fosmids demonstrated presence of the arsenite oxidation genotype andgene expression potential in the BCR.Functional screening of the BCR metagenome and identification of thearsenic resistance clones also showed that they were highly associated withtaxa within the Euryarchaeota phylum (1332 reads). Megan 5 taxonomicanalysis of the As(V) resistant fosmids resulted in total 11140 closest blasthits and 7254 unassigned reads. Firmicutes (467), Bacteroidetes (449), Pro-teobacteria (397) and Fibrobacteres (142) were major taxonomic groups thatwere predominant after Euryarchaeota, Figure D.12. Among Euryarchaeota,Methanomicrobiales (618), Methanobacteriales (228) and Methanosarcinales(168) were highly abundant. Taxonomy profile of BCR that were identifiedfrom functional screening at genus and species levels are presented in Fig-ures 5.3 and D.13 respectively. Figures show the assignment of reads to theNCBI taxonomy and each node is labeled by a taxon and the number ofreads assigned to it.Phylogenetic analysis of the arsenic resistance genes revealed informa-tion about distribution of these genes across different taxa. Preliminaryanalysis of ArsC that were identified on fosmid ends and full fosmids andtheir closest ArsC from diverse taxa are given in figure D.15. ORF of855.7. DiscussionFOS6217.CPR1-NO6 which assigned to ArsC, for example, was closely re-lated to ArsC Clostridium acetobutylicum. Similarly, ArsC FOS6236.CPRF17 was related to Spirochaeta genus. In contrast, ArsC of FOS6205-B1and FOS6209-B8, which identified from functional screening (Full fosmidanalysis), showed similarity to a hypothetical protein of Methanocorpuscu-lum labreanum. In fact, many genes have been associated to this species aswill be discussed later. By assembling some of the fosmid end sequences,the contig (i.e., hybrid fosmid) was clustered with ArsCa Herminiimonasarsenicoxydansherminii D.14.MLTreeMap analysis of the As(V) resistant clones indicated some othertaxonomic assignments in addition to Methanocorpusculum labreanum andMethanospirillum hungatei, such as Methanosphaera stadtmanae and Bac-teroides thetaiotaomicron. According to MLTreeMap fosmids: FOS6208-E24, FOS6209-B19, FOS6209-M3 and FOS6232-C4 were assigned toMethanocorpusculum labreanum with 100% confidence. However, FOS6228-P13, FOS6226-K15, FOS6230-K14 were phylogentically assigned toMethanospirillum hungatei, Bacteroides thetaiotaomicron andMethanosphaera stadtmanae, respectively (Table E.1).The major physiological functions represented by the annotated As(V)resistant fosmid ORFs were investigated based on their sequence comparisonwith KEGG pathways, SEED and COG databases using Megan. Figure E.1displays the KEGG profile of the arsenic resistant clones obtained from theBCR. Carbohydrate, energy and amino acid metabolism were predominantpathways represented on the whole fosmids. Furthermore, relatively highnumber of reads were assigned to membrane transport functions, Figure E.1.5.7 DiscussionIn this study, the aim was to understand the composition and function ofmicrobial consortia involved in arsenic resistance in the BCR through func-tional screening and sequence analysis of the metagenome. The functionalmetagenomic experimental design relied on genetic selection and survival ofthe host, as measured by growth or no growth, in the presence of arsenic.This approach proved to be an effective screen for arsenic resistance andwhould be useful for other phenotype traits. The combination of functionalscreening and sequence analysis resulted in the identification of expression865.7. Discussionpotential of arsenic gene biomarkers associated with the BCR environment.Functional screening of other environments was reported previously for efflu-ent treatment plants (in the pesticide industry) screening arsenic resistanceclones [28] and from a plant rhizosphere adapted to acid mine drainage,searching for nickel resistance genes [103].The full fosmid analysis of the arsenate resistance clones and metagenomefosmid ends demonstrated presence of different arsenic resistance mecha-nisms inside the BCR. In one pathway for arsenic resistance (Figure 1.3),As(V) enters the microbial cell, most probably via the phosphate transportsystems (e.g Pit, Pst). These transporters were predominantly associatedwith functional screening (Figure E.3), suggesting a potential for arsenateaccumulation inside the microbial biomass of the BCR. Periplasmic arsenatereduction also is an important mechanism that was identified in the BCR.In this process, arsenate that enters the microbial cells is reduced to As(III)inside the periplasm. In the next step, As(III) is pumped out of the cellthrough ArsB coupled to ArsA-mediated ATP hydrolysis. The genes asso-ciated with arsenate reduction and efflux pumps were detected in the As(V)resistant fosmids and fosmid ends. Circos diagrams represent the abun-dance of the transporters and ars related genes in different fosmids fromthree libraries (Figure 5.2, Figure D.10 and Figure D.11). Notably, As(III)is released from the microbial cells can further be immobilized as sulfideminerals inside the BCR. Indeed, in the Chapter 3, [71], arsenic was de-tected as sulfide minerals (arsenopyrite, tennantite) and microbial processessuch as periplasmic arsenate reduction and arsenate efflux mechanisms po-tentially facilitated this arsenic sulfide formation in the BCR by providingthe As(III). Through a dissimilatory arsenate reduction mechanism whichwas detected with functional screening, As(V) is reduced to As(III) underanaerobic condition and released As(III) could also combine with the sul-fides and be precipitated in the BCR. However, this mechanism was not themajor mechanism identified in the arsenic resistant clones and was presentto a lesser degree, Figure 5.1, Figure D.3and Figure D.4. Detection of thegenes related to arsenite oxidation indicated microbial-mediated processesfor arsenite oxidation inside the BCR. The As(III) volatilization mechanismwas an alternative mechanism for arsenic removal from the BCR, despitelow presence of genes related to this mechanism, Figure 5.1, Figure D.3 andFigure D.4.Three fosmid libraries were analyzed for identification of arsenic resistantgenes constructed from three separate core sections as described in Chap-875.7. Discussionter 2 (Table 2.1). Solid analysis of the samples recovered from boreholes TS3and TS9, which used for MG1 and MG3, revealed lower As concentrationscompared to MG2, which was obtained from borehole TS4. Arsenic concen-trations associated with MG1, MG2 and MG3 were 1.8 ppm, ≥ 250 ppm and18.9 ppm respectively, which could effect recovery rates of arsenic resistancegenes. Indeed, selection of an appropriate environment for gene discoveryplays an important role in the recovery of functional genes [154]. In thisstudy, metagenomic libraries with a total of 14206 clones were screened forthe arsenate resistance genes. The recovery rates of arsenic resistance geneswere approximately 1/238, 1/ 64 and 1/245 clones in MG 1, MG2 and MG3,respectively. The higher recovery rate from MG2, which sourced from April2009 BCR sampling time (Table D.2), was attributed to the relatively highconcentration of arsenic measured in the core solid samples. Metal analysisand pore water chemistry of the metagenomic libraries can be found in Ta-bles A.1 and C.2.Arsenic resistance pathways identified from MG2, which had higherAs resistant genes recovery rate, included bioaccumulation (e.g., pst, phn,Nham-4423 /heavy metal translocating ATPase), periplasmic arsenate re-duction (e.g. arsC, arsR, arsT, arsO), arsenite efflux pumps (arsA/B),volatilization (arsRM), arsenite oxidation (aoxR, aoxB, axyS) and dissim-ilatory arsenate reduction (arrA). Arsenic resistance pathways were alsodetected via functional screening of MG3 and MG1; however, mechanismssuch as periplasmic arsenate reduction (e.g. arsC) and dissimilatory arsen-ate reduction (e.g., arr) were not detected in MG3 and MG1, respectively. Adissimilatory arsenate reduction mechanism was only detected by functionalscreening of MG2. Comparison of three metagenomic libraries also showedthat the periplasmic arsenate reduction mechanism was more predominantin MG2 than other libraries, including a higher number of arsC and arsoperon regulators. Moreover, efflux pump genes identified from fosmid-endsscreening were higher than functional screening in all three metagenomiclibraries, indicating this mechanism to be predominant in the importanceof whole metagenome and metal transport in metal contaminated environ-ments.The metal transporters that were identified in this work are capable ofassimilating many metals [157]. Metal transporters for nickel, iron, zinc,cobalt and copper were identified through functional analysis of the full fos-mids, Figures E.2, E.4, demonstrating richness of these transporters in theBCR. The observations were consistent with results from genomic analy-885.7. Discussionsis of AMD environments [157] [103]. The variety of proton efflux systems(H+) ATPase, antiporters and symporters as well as resistance to copper,cobalt, arsenite, mercury, zinc, silver and cadmium (arsR, arsB, merA)were reported in an AMD microbial biofilm study [157]. Figure E.2 andE.4 present gene predictions based on KEGG and NCBI-NR databases. Itshould be noted that the full fosmid ORFs were compared to both the cus-tom As resistant gene and Genbank databases. Figure D.6 shows percentidentities for the best hits in Genbank and the custom As gene databases.Higher percent identities were obtained for the Genebank matches probablyis due to small size of the arsenic database and low number of arsenic genes.The arsenic resistant fosmid clones that were taxonomically affiliatedto Methanocorpusculum labreanum, (Table E.1), a species which was ableto tolerate arsenic, possess following arsenic resistant genes identified bycomparative analysis of the ORFs to the custom arsenic database: arsT(pfam:Glutaredoxin) (FOS6208-E24), ABC transporters (pstA, phnC ) andarsH (FOS6209-B19) and arsRM, arsR, transporters (phnC )(FOS6209-M3).The identified genes are also given in Figure D.11. Furthermore, arsR,arsA and Nham4423 (i.e., metal transporter) were identified in a fosmidclone (FOS6230-K14) assigned to Methanosphaera stadtmanae, Figure 5.2.Gene arsRM also identified in fosmid clone (FOS6228-P13) affiliated toMethanospirillum hungatei. The genes discovered were proven to be func-tionally active and were resistant up to 60 mM As (Table D.2). The mul-tiple As-resistance mechanisms that were identified in these methanogensare unexplored and not discussed much at all in the literature. In onestudy, Methanocaldococcus jannaschii was investigated for the thioredoxin(Trx)-based oxidative redox regulation and the results suggested that Trxassists methanogens under oxidative stress and coordinates metabolic activi-ties with availability of reductant [152]. Gene arsT (HMMER:Glutaredoxin)that was identified in methanogens via functional screening in this thesiswork also supports the importance of this system in metal-rich environments.Some of the arsenate reductase enzymes (ArsC) use reduced thioredoxin(Trx) to convert arsenate to arsenite with redox active cystein residue in theactive site. Co-expression of a thioredoxin system and arsenate reductasewould enhance the efficiency of arsenate reduction [1]. Annotation of the fos-mid ORFs using Megan indicated the presence of phosphoadenylyl-sulfatereductase (thioredoxin) (KEGG 1.8.4.8) (Figure E.5) in some of the fullfosmid sequences, which was closely related to a hypothetical protein fromMethanospirillum hungatei JF-1. In addition, protein tyrosine phosphatase(Mlab-0819) and arsenical-resistance protein (Mlab-0818) were detected pre-895.7. Discussionviously in Methanocorpusculum labreanum type strain Z [4]. This species washighly prevalent, represented by a total of 106 ORFs, via functional screen-ing (Figure 5.3). Methanocorpusculum labreanum type strain Z belongsto the order Methanomicrobiales and was isolated from surface sedimentsof Tar Pit Lake [4]. Methylation of arsenite also reported for Methanobac-terium species [98]. Interestingly, the methanogens Methanocorpusculum,Methanospirilium and Methanosarcina were very predominant in the BCR[6] and were highly correlated with metal content (Chapter 3-Table B.6).This observation, plus the predominance of methanogen-related ORFs inthe As(V) resistant fosmid clones, indicated methanogenic microorganismsmight have played a role in arsenic transformation and its subsequent re-moval as As(III) precipitates in the BCR. Thus, they might be involved inbiogeochemical cycling of arsenic.Bacteroidetes was the next most highly represented taxonomic groupin the arsenate resistant clones (Table E.1, Figure 5.3). Genes potentiallyinvolved in aox operon (e.g moeA, aoxR) were identified in an arsenateresistant clone (FOS6226-K15) that was assigned to Bacteroides thetaio-taomicron, Figure D.7. No growth was observed at arsenate concentrationshigher than > 30 mM for this fosmid clone, suggesting its potential rolein As(III) detoxification. Transporter encoding genes (e.g.phnC ) and genespotentially involved in aox operon (e.g. moeAA, aoxB) were identified inarsenate resistant clones that were assigned to lower common ancestor ofBacteroides, Porphyromonas and Parabacteroides (FOS6211-O24, FOS6228-I24 and FOS6233-j24) (Table E.1, Figure D.7 and Figure D.8). The moeAgene was previously reported as part of the As(III) oxidation [140] mecha-nism; however, this needs further studies, such as transposon mutagenesis,to determine the actual function encoded for by the genes discovered in thecurrent study. Bacteroidetes (e.g Flavobacterium) and Fibrobacteres (e.gFibrobacter succinogenes), related to As(V) resistance genes, were highlydetected with the functional screening approach. Glycoside hydrolase (GH)was identified in a previous functional screening study of the same BCR.These were related to the phyla Bacteroidetes, Firmicutes, Fibrobacteres,and Chloroflexi suggesting that these groups are involved in cellulose degra-dation (and other recalcitrant material) in the BCR system [100]. ParticularBacteroidetes-related SSU rRNA sequences from the BCR correlated withthe high metal content samples (Chapter 3-Table B.6). Microbes assignedto the phylum of Bacteroidetes were highly represented in the BCR [6].Another phylum that was represented in the As(V) resistant genes wasSpirochaetes, which was also very prevalent in the BCR. An arsenate reduc-905.8. Concluding Remarkstase (ArsC- thioredoxin-like fold) was taxonomically related to Spirochaetasmaragdinae [97]. This bacterium is strictly anaerobic and is able to fer-ment numerous polysaccharides. Spirochaeta smaragdinae is only speciesof the family Spirochaetaceae known to produce sulfide by reducing thio-sulfate or element sulfur [97]. The BCR influent contained high concen-trations of sulfate (400-600 mg/L), therefore microbial sulfur cycling waspossible. Spirochaeta smaragdinae was detected through functional screen-ing (53reads), as shown on Figures 5.3 and D.13.5.8 Concluding RemarksIn summary, the functional metagenomic workflow described here was ef-fective in identifying potential microbial responces to arsenic within the mi-crobial community in the metal remediation system throwing further lighton the biological-based mechanisms. Based on the functional screening andfosmid ends sequence analysis of the three metagenomic libraries, the fol-lowing mechanisms were present in the BCR: bioaccumulation, periplas-mic arsenate reduction, arsenite efflux pumps, volatilization, arsenite oxida-tion, and dissimilatory arsenate reduction. Processes mediated by microor-ganisms (periplasmic arsenate reduction, transporters including phosphatetransporters, metal transporters, arsenate efflux pumps and dissimilatoryarsenate reduction) potentially facilitate arsenic removal via arsenic sul-fide formation in the BCR. High presence of ars related genes, phosphateand metal transporters and arsenite efflux pumps indicated predominanceof periplasmic arsenate reduction mechanisms compared to volatilizationmechanism. In addition, arsenic resistance genes related to methanogens,such as transporters, periplasmic arsenate reduction, methylation suggestedmethanogens may play a role in arsenic detoxification inside the BCR.915.8. Concluding Remarkspstphnmetal transporterFull fosmid screeningFosmid end screening2% 5% 10% 15%Figure 5.1: Summary of arsenic resistance genes identified in full fosmidsand fosmid ends, that were potentially involved in arsenic resistance mecha-nisms, Metagenomic Library 2. Circles area represents percentage of arsenicresistant genes identified in the Metagenomic Library 2 (number of As genesin total identified As genes in the MG2).925.8. Concluding RemarksArea1-F3102030FOS6225-C1251015202530 FOS6225-F151015202530 FOS6225-G151015202530FOS6225-M15101520253035FOS6225-P85101520253035FOS6226-H1351015202530FOS6226-K15102030FOS6228-I24102030FOS6230-B24102030FOS6230-E2251015202530FOS6230-F141020 30FOS6230-F16510 15 20 2530FOS6230-G175 10 15 20 25 30FOS6230-H245 10 15 20FOS6230-I235 10 15 20 25 30FOS6230-K185 10 15 20 25 30FOS6230-M145 1015 2025 30FOS6231-E2251015 2025FOS6231-H15102030FOS6231-N185101520253035FOS6232-A351015202530FOS6232-C3 51015202530FOS6232-C4 102030FOS6232-D22 51015202530FOS6232-I15 51015202530FOS6232-I245101520253035FOS6232-K145101520253035FOS6232-P0251015202530100 Pst Phn metal transporters efflux pumps Ars A/BArsCArs regulators methylation Ars RM These are shown whitin the fosmid boxes (1th circle) These are shown in the next inner circle (2th circle) (3th circle)Figure 5.2: Circos representation of complete fosmid sequences using Illu-mina sequencing. Circos plot representing arsenic resistant positive fosmidsidentified from metagenomic library 2 (April 2009). The outer circle repre-sents fosmids, labeled FOS6232-P02, FOS6232-K14, etc. Numbers on theouter circle show size scale in kilobases (kb). Colored bars within fosmidsand inner circles show locations of As genes. Black histograms indicate thepercent identity of the blast hits. Identity below 30% were not shown.935.8. Concluding RemarksNo hits; 11Not assigned; 7254unclassified sequences; 11Arabidopsis; 11Tetrapoda; 5Eukaryota; 8 Pyrococcus; 24Methanopyrus; 6Methanosarcina; 160Methanospirillum; 106Methanoregula; 5Methanoplanus; 5Methanocorpusculum; 438Methanomicrobiales; 55Methanomicrobia; 13Methanococcus; 15Methanocaldococcus; 15Methanococcales; 0Methanothermobacter; 89Methanobacterium; 130Methanobacteriaceae; 8Archaeoglobus; 24Euryarchaeota; 220Thermotoga; 13Treponema; 52Spirochaeta; 53Borrelia; 10Spirochaetaceae; 20Vibrio; 7Pseudomonas; 13Pasteurellaceae; 6Yersinia; 5Salmonella; 5Escherichia; 26Enterobacteriaceae; 20Alteromonadales; 5Gammaproteobacteria; 27Helicobacter; 7Deltaproteobacteria; 11delta/epsilon subdivisions; 0Neisseria; 5Ralstonia; 15Betaproteobacteria; 0Sinorhizobium; 6Rhizobium/Agrobacterium group; 7Rhizobiaceae; 4Mesorhizobium; 10Hyphomicrobium; 9Bradyrhizobiaceae; 8Rhizobiales; 70Caulobacter; 5Alphaproteobacteria; 35Proteobacteria; 76Fusobacterium; 12Thermoanaerobacter; 5Acetivibrio; 6unclassified Lachnospiraceae; 6Eubacteriaceae; 6Clostridium; 82Clostridiales; 54Clostridia; 14Lactococcus; 5Staphylococcus; 6Brevibacillus; 22Listeria; 8Bacillus; 79Bacillales; 25Bacilli; 24Firmicutes; 82Fibrobacter; 142Deinococcus; 8Chroococcales; 5Nostoc; 16Cyanobacteria; 6Chloroflexales; 5Anaerolinea; 22Chloroflexi ; 6Verrucomicrobia; 6Lentisphaeria; 5Chlamydiae/Verrucomicrobia group; 2Sphingobacteriaceae; 13Flavobacteriaceae; 13Cytophagaceae; 6Prevotella; 10Porphyromonas; 5Paludibacter; 9Dysgonomonas; 11Porphyromonadaceae; 16Anaerophaga; 8Bacteroides; 89Bacteroidales; 138Bacteroidetes; 115Mycobacterium; 10Bacteria; 502cellular organisms; 234root; 2Figure 5.3: Megan 5 analysis of the full fosmid sequences against the NCBI-NR database. Numbers in taxonomic tree indicate the number of assignedreads to each taxon collapsed at genus level of NCBI taxonomy.94Chapter 6Conclusions6.1 Overall ConclusionsThis research investigated the metal removal mechanisms with respect tometal-microbe interactions in a biochemical reactor (BCR) treating watercontaminated with arsenic, zinc and cadmium. The conceptual model pro-posed for this BCR (Chapter 1-Figure 1.4) included all possible biogeochem-ical processes involved in arsenic transformation and arsenic accumulationinside the BCR. The model proposed in Figure 6.1 depicted arsenic removalmechanism based on the direct and indirect evidences that were found fromthe BCR. Overall, the main drivers responsible for immobilizing arsenicin this study were (i) the geochemical reducing conditions, (ii) processesmediated by microorganisms (dissimilatory arsenate reduction, periplasmicarsenate reduction, arsenite efflux mechanism) which facilitate arsenic sul-fide formation, (iii) adsorption and coprecipitation on the surfaces of othersolids such as silicates (e.g., textural association of Zn- and As-bearing min-erals with amphibole particles) and (iv) mineralization of arsenic and zinc asoxides, sulfates and phosphates. Other processes may contribute to arsenicimmobilization in the BCR; however, they are not reliable mechanisms forlong term arsenic removal in the system. Metal adsorption to organic matterand biosolids is possible in early stages of the BCR operation and adsorp-tion sites are occupied within short time [94]. An alternative mechanism isfiltration of the incoming solid forms of arsenic and zinc, which helps for ac-cumulation of metals. Nevertheless, some of the solid forms (e.g., ko¨ttigite)could dissolve and reform as different mineral types. Biotic processes (e.g.,DARP, ARM) can contribute to arsenic release from the arsenate-bearingsolids such as ko¨ttigite.According to the data collected from the site, conditions in the BCRwere reducing, and based on chemical analysis of the BCR solids, sam-ples had high organic content (i.e., TOC measured up to 13.28%, TableA.2). In addition, 60% biosolid residuals were used as starting materials forBCR construction. Therefore, high organic content inside the BCR provided956.1. Overall Conclusionsgeochemical reducing conditions for arsenate reduction, i.e., conversion ofAs(V) to As(III). The reaction also was influenced by microbial groups suchas arsenate resistance microbes (periplasmic arsenate reduction, arsenite ef-flux pumps mechanisms) and dissimilatory arsenate respiring prokaryotes(dissimilatory arsenate reduction mechanism). The arsenate reducers iden-tified from enrichment cultures and detection of the potential genes relatedto these mechanisms from metagenomic studies indicated potential bioticrole for transformation of As(V) to As(III). Methanogenic archaea (e.g.Methanocorpusculum labreanum and Methanospirillum hungatei JF-1 ) andSedimetibacter, are examples of the groups that were involved in the arsen-ate reduction (Figure 6.1). In the following sections, these groups and thegenes identified will be discussed further.The arsenite species, in the next step can be precipitated as arsenic sul-fides. QEMSCAN analysis revealed presence of arsenic sulfides, includingarsenopyrite and tennantite. This seems to be the main mechanism oper-ating inside the BCR. As shown in Figure 1.4, arsenite can be bind to ironsulfides (e.g., pyrite (FeS2), pyrrhotite (FeS)) and arsenopyrite can be formedin the BCR. Reducing conditions inside the BCR and microbially enhancediron reduction also lead to resolubilization of iron from Fe-coated sands. Iron(i.e., Fe(II)) can react with sulfide to form iron sulfides. In fact, iron sul-fides were detected in the BCR; therefore, arsenite sorption to iron sulfidesfor arsenopyrite formation was possible. Arsenopyrite compound was alsopredicted by geochemical modeling under higher Fe concentrations. Thepossible role of biology in the formation of arsenopyrite is arsenite reductionby ARM and DAPR and sulfate reduction by SRB. The identification of thefunctional genes related to the arsenite efflux mechanism and detection ofsulfate reducing bacteria also supported formation of arsenic sulfides in theBCR. Desulfitobacterium spp. and Desulfosporosinus spp. were enriched inAs(V) cultures. Desulfobulbus spp. and Desulfovibrio spp. were found fromBCR samples (Table B.7, Supplemental Information).Transformation of As(III) to As(V) is also a possible mechanism in theBCR, although less likely than As(V) reduction (Figure 1.4). Identificationof arsenite oxidation genes and arsenite oxidizers in enrichment culturesshowed potential presence in the BCR. This might contribute to arsenicprecipitation with iron oxides. Although As(V) can precipitate from so-lution with Fe(III) to form scorodite, it was not detected in mineralogicalanalysis. Iron oxides, such as magnetite, were detected; however, evidenceof adsorption and coprecipitation of arsenic with iron oxides was not ob-966.1. Overall ConclusionsFigure 6.1: Schematic diagram for As removal mechanism in the BCR.Abbreviations: ARM, arsenate resistance microbes; SRB, sulfate reducingbacteria.served, possibly due to arsenate reducers and iron reducing bacteria, whichcontribute to As and Fe release. Indeed, iron-reducing bacteria enrichmentculture-related OTUs were found in the BCR (Table B.7, Supplemental In-formation). Simplicispira and Acidovorax species are examples of the groupsthat may play a role in arsenite oxidation. Acidovorax was also predomi-nant in BCR core samples, which couples arsenite oxidation with nitratereduction.Functional metagenomic study also revealed some other mechanismsthat play important role for arsenic removal in the BCR. For example,arsenite methylation was present but not as predominant as other mech-anisms. Arsenite methylation was associated with microbial groups such asmethanogens. Identification of transporter genes (phosphate transporters,metal transporters) associated with arsenic resistance was also widespread976.1. Overall Conclusionsamong the BCR’s taxonomic groups suggesting potential metal bioaccumu-lation mechanisms inside the microbial cells.Since different methodologies were conducted to understand arsenic re-moval from the BCR, the following sections describe different approachesand present the main conclusions, followed by a summary to relate the re-search results.First, a geochemical model was developed to predict the formation ofarsenic-bearing minerals under the physico-chemical conditions present atthe site. Here are the conclusions obtained by producing Eh-pH diagramsfor As-Fe-S-O-H at 25 ◦C and 1 atm pressure, from Spana software.• Predominant arsenic species based on the chemistry of the BCR porewater revealed realgar (AsS), crystalline As and arsenopyrite (FeAsS)were all possible to form under the Eh and pH condition of the BCR(pH 5.6-7.5, ORP -100 to -300 mV) when total As, Fe and S valuesare chosen within the rages measured at the site.• Total As and Fe concentrations effect stability ranges for arsenic species.At lower total Fe concentrations, modelling predicted realgar and or-piment to be predominant, whereas at higher Fe concentrations, ar-senopyrite becomes predominant.• At the BCR influent concentrations of total As, Fe and S (0.44, 0.23and 7.00 mM, respectively), arsenopyrite was the only predicted pre-dominant form of arsenic sulfides at geochemical reducing conditions.Indeed, one of the limitations of these predictions is that many el-ements are not included in the model (e.g., Cd, Cu, Zn etc.) andsolubility products of arsenic-bearing minerals is still uncertain.To investigate the actual minerals formed in the BCR, the metal andmineral content of the BCR core solids were measured.• QEMSCAN bulk mineral analysis (BMA) revealed that the samplesconsisted mainly of silicate and carbonate minerals. The minerals,quartz, calcite and feldspar represented up to 29.9% in the samples.The trace minerals (zinc-, arsenic-, copper- and lead-bearing minerals)were found in all the samples and represented less than 0.1%. This iscomparable with core solid chemistry analysis where concentration of986.1. Overall Conclusionsmetals were not very high and 1735, 79 and ≥ 250 ppm were obtainedfor Zn, Cu and As, respectively.• To further study the forms of the arsenic- and zinc-bearing minerals,which was the particular research interest, a hybrid particle mineralanalysis (PMA) was performed on each sample using QEMSCAN sup-ported by EDX analysis. The trace minerals were grouped into fourdifferent classes: zinc-arsenic sulphosalts/sulfates; zinc-lead sulphos-alts/sulfates; zinc phosphates and zinc-arsenic oxides. The overallconclusions are:– Possible arsenic minerals were oxides (unknown zinc arsenic ox-ides), sulfides (arsenopyrite, tennantite), arsenates (Wihelmklei-nite) and zinc-arsenic sulfides (coating around silicates).– An interesting observation made by PMA analysis was the coat-ing of sphalerite on amphibole and feldspar. X-ray analysis re-vealed presence of As in this mineral, indicating arsenic’s asso-ciation with the sphalerite coating. EDX analysis also verifiedarsenic association with iron, zinc and sometimes cadmium asarsenopyrite-type minerals.– Zinc sulfate minerals were identified such as zinc-aluminum sul-fate (Zinc woodwardite ((Zn1−xAlx)(OH)2)(SO4)x/2)), zinc-leadsulfates and zinc sulfates (unknown minerals). Zinc phosphateparticles were also present.This textural association of Zn- and As-bearing minerals with other par-ticles suggested adsorption and coprecipitation contributions for immobi-lization of arsenic and zinc in the BCR, as revealed with the QEMSCAMPMA imaging. In addition, it was highlighted that sulfates and sulfides arethe predominant types of Zn and As minerals formed in the BCR. In orderto address the reasons for association of these minerals with BCR microbialgroups, metal content of the samples were correlated with the BCR micro-bial community and particular microbial groups that were more prevalentin high metal content samples were identified. The microbial communitycomposition in each of the BCR samples was determined by other lab re-searchers. The conclusions are as follows:• The microbial groups more prevalent in the metal-rich samples in-cluded the poorly characterized taxonomic groups of Bacteroidetes-related SB-1, VadinHA17 and M2PB4-65 environmental groups, can-didate division WS6, RF3 and TM6.996.1. Overall Conclusions• The other predominant microbial groups associated with metals wereSynergistaceae family, Victivallales order and methanogen generaMethanocorpusculum, Methanospirillum and Methanosarcina. Sulfate-reducing bacteria and iron-reducing environment-related groups werefound in the BCR, although not highly correlated with the high metalconcentration samples.The above conclusions generally address the types of minerals presentedat the BCR and potential roles of less characterized bacterial groups interact-ing with metals. Furthermore, evidence for biotic mineralization (Figure 1.3-Biomineralization ) was obtained by mineralogical studies. Sulfides (spha-lerite) and carbonates (dolomite CaMg(CO3)2, minrecordite CaZn(CO3)2)are the minerals that were predominant and were consistent with biominer-alization processes. The forms of arsenic sulfide found in the BCR were ten-nantite and arsenopyrite, the latter being the predominant form predictedby As-Fe-S-O-H geochemical modelling for high Fe conditions. However, thepossible role of biology in formation of arsenic sulfide (e.g., arsenopyrite) asdiscussed earlier would be arsenate reduction to arsenite and the reductionof sulfate to sulfide by microbes (e.g., SRB). Periplasmic arsenate reductionmechanisms (Figure 1.3- mechanism 3,4,5) were identified predominantlythrough functional metagenomic screening and SRB were identified in theBCR.The mineralogical study confirmed that arsenic minerals were beingformed in the BCR. However, the role of microorganisms in mineraliza-tion or mobilization was unknown. A literature review revealed that manymicrobes have the capacity to transform arsenic species, either as part ofresistance mechanisms or for energy. Enrichments of As(III) and As(V)media identified putative arsenite oxidizing and arsenate reducing microbesin the BCR. The arsenic enrichment cultures sourced from the BCR wereestablished in the laboratory, under anaerobic setup, followed by deep 16SrRNA gene pyrosequencing. This was coupled to chemical measurements ofAs(III)/As(V) during enrichments. Key conclusions include:• Taxonomic profile of the enrichment microbes revealed the BCR ar-senic transforming microbes to be closely related to the previouslyshown arsenate redusers and arsenite oxidizers. However, Simplicispira(β- proteobacterium) (CAOs) and Sedimentibacter (Clostridia, Fir-micutes) (arsenate reducing) were the most predominant enrichmentculture OTUs. These OTUs have not been associated with arsenictransformation until now.1006.1. Overall Conclusions• The As(III) medium enrichment OTUs were assigned to Proteobacte-ria, Euryarchaeota, Bacteroidetes, Fibrobacteres and Firmicutes.Whereas, primary phyla Firmicutes, Proteobacteria, Bacteroidetes, Eu-ryarchaeota and Spirochaetes were identified for the As(V) enrichmentcultures.• The Methanomicrobia (Metanocorpusculum and Metanoregula) relatedto Euryarchaeota, were prevalent in the arsenic enrichment cultures.These methanogens were present in earlier passages and were absentin late passages.• Sulfate-reducing bacteria Desulfitobacterium and Desulfosporosinus,were present in the arsenate enrichments. Desulfitobacterium such asD. auripigmentum and Desulfosporosinus species are found to reduceAs(V) as well as sulfate [7],[112]. Co-occurrence of As(V) reductionwith sulfidogenesis could attribute to detection of arsenic minerals(e.g., arsenopyrite) in the BCR.• These sulfate-reducing bacteria found in the enrichments are likelyto compete with methanogens, since the As(V) enrichments that didnot have any Desulfitibacterium related OTUs contained methanogenrelated OTUs. These two groups may compete for electron donors(lactate, molecular hydrogen ) or electron acceptors (arsenate). Nev-ertheless, the presence of both groups in enrichments demonstratedtheir high capabilities to respond to arsenic.• Chemical arsenic speciation monitoring of the enrichments also demon-strated potential of arsenite oxidation and arsenate reduction in theAs(III) and As(V) media, respectively. The genera listed above are as-sociated with arsenic transformation, as arsenic species concentrationwas monitored over time. Some of the taxonomic groups were alsoidentified with functional screening; however, it is not possible to as-cribe arsenic oxidation/ reduction to any particular taxonomic groupsand these species should be further investigated to verify involvementin arsenic transformation. HPLC-UV was an effective method for ar-senic speciation measurements.Moreover, prevalence of these putative arsenic-transforming microbes(putative CAO and arsenate reducing microbes) were assessed in the BCRand the following conclusions were obtained:1016.1. Overall Conclusions• The methanogens, Bacteroidetes (Proteiniphilum) and Spirochaeta re-lated OTUs which were prevalent in both As(III) and As(V) enrich-ments, were identified in the BCR. Notably, they were also majorgroups identified by phylogeny analysis of the resistance clones (func-tional screening method).• As(III) enrichment OTUs, those related to Acidovorax and Albidiferaxspp. (β-proteobacterium) were present in the BCR. From the litera-ture, Acidovorax species (e.g., NO1) were able to oxidize arsenite un-der aerobic and anaerobic conditions, which entails an aio arseniteoxidation operon [62], which indicates potential biological arsenite ox-idation in the BCR by Acidovorax species. As(V) OTUs present inthe BCR were Bacteroidales-related OTUs, Sedimentibacter -relatedOTUs, Atopococcus-related OTUs, the Paenibacillus-related OTUsand the BetaproteobacteriumUCT N117-related OTUs. Although Sed-imentibacter -related OTUs were found in highly metal contaminatedenvironments [22], these taxonomic groups have not been linked witharsenate reduction previously.The above conclusions showed that both arsenate reduction and arseniteoxidation took place in the biochemical reactor and that arsenic speciationmay be cyclic. Moreover, it can be inferred that arsenic resistance microbescan be cultured from the BCR. To address presence of other mechanismsand investigation of arsenic biotransformation mechanism, metagenomic li-braries were screened for the functional genes (arsenic biomarkers) involv-ing bioaccumulation, periplasmic arsenate reduction, arsenite efllux pumps,volatilization, arsenate oxidation and dissimilatory arsenate reduction mech-anisms (Figure 1.3- mechanisms 1-6,9,10). The functional metagenomescreening and fosmid end sequence screening for arsenic resistance genesled to the following conclusions:• Analysis of three metagenomic libraries with a total of 14206 clones,revealed that arsenic resistance genes are widespread. The recoveryrates of arsenic resistance genes from functional metagenomic screen-ing were approximately 1/238, 1/ 64 and 1/245 clones in the metage-nomic library 1, 2 and 3, respectively. The higher recovery rate frommetagenomic library 2, might be attribute to the relatively higherconcentration of arsenic measured in the core solid samples in thismetagenomic library.1026.1. Overall Conclusions• The full fosmid analysis of the arsenate resistance clones and fosmidends demonstrated the presence of several mechanisms involved in ar-senic removal from the site. Phosphate transporters were found tobe predominant, suggesting potential of arsenate uptake by micro-bial biomass. High number of genes detected were associated withperiplasmic arsenate reduction and efflux pumps. Indeed, As(III) thatis released from the microbial cells can further be immobilized as sul-fide minerals inside the BCR. Dissimilatory arsenate reduction mecha-nism which also helps in this processes (i.e., provide the arsenite), wasdetected in lower degrees. Identification of these genes is importantto biomarkers showing presence of biotic processes that help arsenicpercipitation in the BCR when sulfidogenesis is also present in the en-vironment. Detection of putative genes related to As(III) methylationmechanism, also showed arsenite volatilization from the BCR.• Comparison of three metagenomic libraries also showed that periplas-mic arsenate reduction mechanism was more detected in MG2 thanother libraries, including higher number of ArsC and ars operon reg-ulators. Dissimilatory arsenate reduction was only detected via func-tional screening of MG2. The findings were consistent with high ar-senic genes recovery rate from MG2.• The fosmid clones that were taxonomically affiliated to methanogens,Methanocorpusculum labreanum and Methanospirillum hungatei JF-1,also contained putative genes related to arsenate reduction mechanism(for example: arsT, arsR, arsH ), methylation (arsRM ) and trans-porters (pstA, phnC, arsA). The multiple As resistance mechanismsthat were identified in these methanogens are unexplored and not dis-cussed much in the literature. Identification of methanogen-relatedarsenic resistance genes, plus the predominance of methanogens in theBCR, indicated their role in arsenic transformation in the BCR.• The phylogeny analysis showed that, besides the Euryarchaeota phy-lum (Methanomicrobiales, Methanobacteriales and Methanosarcinales),which were dominant in arsenic resistant fosmid clones, Firmicutes,Bacteroidetes, Proteobacteria and Fibrobacteres, were also highly as-sociated with arsenate resistant fosmid clones. Bacteroidetes, Firmi-cutes were predominant in the BCR and they were involved in complexorganic degradation (i.e., recalcitrant material).1036.2. Significance of the Research• Taxonomical comparison of the arsenate resistance clones with ar-senate enrichment cultures also showed common taxonomic groups:Methanocorpusculum, Metanoregual, Spirochaeta, Pseudomonas,Clostridium and most of the genera related to porphyromonadaceae,including Bacteroidetes, Cytophagaceae, Bacteroidetes, Flavobacteri-aceae and Sphingobacteriaceae.The above conclusions emphasized arsenate conversion to arsenite, possiblyvia microbial processes such as periplasmic arsenate reduction, dissimilatoryarsenate reduction and arsenite release via the arsenite efflux mechanism.Methanogens were most likely associated with these mechanisms. They alsocorrelated with high metal content samples of the BCR and were identifiedin arsenic enrichment cultures. There is also potential for bioaccumula-tion mechanisms and transformation of arsenite to arsenate inside the BCR.These conclusions also emphasized the role of functional screening and se-quence analysis of the metagenome to determine the mechanisms for arsenicremoval, specially by identification of the functional genes in resistant fosmidclones, where arsenic resistance expression potential was verified by growthobservation in the As-rich media.6.2 Significance of the ResearchThis work shows that physicochemical, culturing and metagenomicapproaches can be combined to reveal the main mechanisms involved inarsenic immobilization. This is the first time these methods have been com-bined together on the same system. First, it has been found that throughmetagenomic studies arsenic resistance is widespread. Indeed, detection ofdifferent class of metal resistance genes, represented different arsenic removalpathways inside the BCR. The wide diversity of mechanisms makes BCR arobust treatment system. Secondly, both arsenic oxidation and arsenic re-duction are possible in the BCR, due to presence of both arsenic reducingand oxidizing microbes and wide diversity of taxonomic groups that havethe capacity for arsenic transformation. As redox cycling indicates the BCRis a dynamic system which can benefits the BCR in term of As removal.According to the proposed As removal mechanism in the BCR (Figure 6.1),conversion of As(V) to As(III) is possible in the BCR (e.g., identificationof putative genes related to periplasmic arsenate reduction mechanism, dis-similatory arsenate reduction). As(III) can be immobilized by binding to1046.2. Significance of the Researchiron-sulfides and FeAsS can be formed. In addition, As(V) can precipi-tate with Fe(III) (e.g., from Fe-coated sands). Finally, the presence of ironand sulfide are most important since FeAsS is the main form of immobilizedarsenic, indicating the arsenic remediation using microbes under anoxic con-ditions with Fe and S as feasible arsenic treatment mechanisms.This work also includes novel aspects, for example a comprehensive sur-vey of all types of zinc minerals formed in the BCR had not been doneprior to the current study. Metagenomic studies of arsenic gene markerswere performed for the first time in biological based metal remediation sys-tems and the most predominant enrichment bacteria (Simplicispira (CAOs)and Sedimentibacter (arsenate reducing)) that was identified in this studyhave not been associated with arsenic transformation before. In this study,arsenic removal was investigated in the field scale reactor, considering site-specific characterizations such as Eh-pH conditions, chemistry parameters,etc., which distinguished this work form bench scale arsenic removal studies.Beside the new findings, which are of particular interest to metal biore-mediation, contaminated sites reclamations and biosolid waste managementfields, the methodologies such as functional metagenomic screening and min-eralogical QEMSCAN analysis provide practical tools to study BCRs andsimilar sites, and make this study significant.The QEMSCAN, an automated mineralogical method, has been con-ducted for characterization of soil, sediments in different studies. It is mostlyapplied for ore characterization, mineral exploration and in the few cases foridentification of bioaccessible compounds in waste materials to assess poten-tial for AMD production [15],[21]. Here, this automated approach was usedfor identification of arsenic and zinc minerals and study of metal immobliza-tion in the mine generated effluent treatment system, which demonstratedan effective tool for comprehensive survey of the mineralogical forms presentin the complex treatment systems such as BCRs.Functional metagenomic workflow described here was also effective to in-vestigate the mechanisms involved in arsenic remediation in the biological-based metal remediation system. The ultimate goal of the metagenomicanalyses (e.g., metatranscriptomics, metaproteomics) are to understand theroles of microbes in eco-geological processes by studying the gene expressionor proteins. Here, arsenic genes in the metagenomic libraries (fosmid ends)was searched and using functional screening these genes in the expressed1056.3. Contributions to the Fieldgenotype were looked at. Functional metagenomics highlighted the majorAs biotransformation potential and informed us of the taxonomic groupswith this potential. Functional screening is suggested as the powerful toolto study the phenotype of interest in the BCRs or similar environments.6.3 Contributions to the FieldThis research was a very comprehensive study of a biological-based treat-ment system, treating high metal (As, Zn, Cd) sulfate leachate by usingorganic residuals. The major scientific contributions of this research re-vealed: (i) different minerals formed in the BCR (i.e., what arsenic, zinc,copper bearing minerals could form or precipitate in the system), which isimportant to know when managing the residuals; (ii) new hitherto unknownspecies associated with arsenite oxidation and arsenate reduction. Thesespecies identified with enrichment experiments should be further investi-gated to verify their involvement in arsenic transformation (e.g., metage-nomics, isolate culturing); (iii) the most prevalent As resistance mechanismwas As reduction or methylation with methanogen-related genes. This is anew role suggested for methanogens in bioremediation of arsenic. Most ar-senic resistant clones were taxonomically assigned to methanogens and theywere also identified in arsenic enrichments, demonstrating their role in thearsenic biogeochemical cycle inside BCRs.This research was also directed to provide relevant information for practi-cal applications and to address industry concerns related to arsenic removal.The sulfate precipitation processes is one of the promising approaches thathas been considered as clean technologies for metal removal and even recov-ery of dissolved metals from wastewater. One of the issues related to thismethod is the solubility and leachability of metal sulfides. For example, thequestion that may be asked in this field is: could base metal sulfides gen-erated through passive treatment systems leach in the future if the waterflowing through them ever becomes oxidizing?Here we studied a BCR which worked under reducing condition (ox-idation reduction potential measured -150 mV) and demonstrated thatprecipitation of metal sulfides occurs under anaerobic condition. Mineralog-ical analysis on the BCR’s samples showed that sulfates and sulfides werethe predominant types of metal-bearing minerals [71]. The method worked1066.3. Contributions to the Fieldbecause both biotic and abiotic reactions were involved in As sulfides pre-cipitation. However, if the condition of the BCR changes, for example waterflowing through BCR becomes oxidizing and oxygen enters the system, themicrobes consume oxygen (e.g., oxygen can be used by arsenite oxidizersas electron acceptor). Nevertheless, If excess oxygen get introduced to thesite (i.e., demolishing the BCR) the minerals may be oxidized and results inarsenic release from sulfides. Therefore in future it is important to maintainreducing conditions of the BCR.This research also contributed to development of ecotoxicological toolsto evaluate metal bioremediation in the sites. One important question inthe remediation field is how we can evaluate remediation of contaminatedsediments and what are the proper monitoring techniques for site remedia-tion. One way to address this is to look for the presence of specific genesthat convey resistance to toxic metals or the ability to degrade toxic organiccompounds. If these genes are present, this provides at least preliminaryevidence that the organisms may be exposed to bioavailable contaminantsand are capable of using the metal in a variety of ways.For example, research focus at the New Bedford Harbor (NBH), a sitemanaged by the U.S. Environmental Protection Agency (U.S. EPA), was ondeveloping ecological biomarkers of contaminant exposure for use in monitor-ing remediation of contaminated sites [47]. arsA, arsB, arsC were chosen inthis research to evaluate metal resistance. The expression profile were stud-ied. However, optimizing RNA extraction from sediments for evaluation ofgene expression is a time consuming and sensitive step. In other research,the transcriptomics approach was developed to determine which arrA (Asreduction), arsB and acr3 (As resistance) sequences were being transcribedduring the in situ bioremediation experiment. mRNA was extracted fromthe groundwater and cDNA libraries were made from arrA, arsB and acr3transcripts [52].Here we described the development and implementation of a high-throughput functional metagenomic screen incorporating bioinformatics thatprovides a general paradigm for detection of a different class of metal resis-tance genes. Moreover, we demonstrated how functional screening methodas well as sequence base screening of metagenomic library could be usedto detect the different diverse classes of arsenic resistance genes (e.g., arsoperon, aox operon, arr operon, methylation and etc.) and their frequency.The screening of metagenome sequences (fosmid ends) and resistant clones1076.4. Limitations of the Research and Recommendations for Future Investigations(functional screening) not only tell us about the quantitative aspects ofthese genes, but also tell us about the major microbial functions. Func-tional screening is a rapid and informative tool that bears great potentialfor use as a ecotoxicological tool to evaluate metal resistance in contami-nated sites.6.4 Limitations of the Research andRecommendations for Future Investigations1- With reference to Chapter 3, we showed preliminary evidence for bioticmineralization, sulfides (sphalerite, tennantite) and carbonates (dolomite,minerecordite); however, QEMSCAN results need to be confirmed withother methods. Biologically mediated mineralization mechanism for Zn havebeen showed in literature, even though the evidences that we identified arenot sufficient and further studies are needed to verify biologically mediatedmineralization. The Synchrotron-based STXM (Scanning transmission X-ray microscopy) approach can be used to study cell-mineral aggregates (i.e.,close proximity of cells and minerals) [61]. It is also recommended that theminerals that were presented in the Table 3.2 and were unknown be furtherstudied.2- In Chapter 4, the Simplicispira and Sedimentibacter were in the ar-senic enrichment cultures and were associated with arsenite oxidation andarsenate reduction, respectively. The arsenic transformation by these groupswere not confirmed and need metagenomic sequencing of these bacteria orisolation and characterization of these bacteria.3- DNA isolated form arsenic enrichment cultures was PCR amplifiedusing primers specific for the V6 variable region of the SSU rRNA gene.Pyrosequencing of the samples was used to empirically estimate bacterialspecies richness. Alternatively, Illumina sequencing of enrichment samples isalso proposed to identify whole bacterial genome present in enrichment sam-ples. Higher advance methods such as single-cell genomics can also be usedfor functional and phylogenetic diversity. In this approach, cells can be iso-lated with high-throughput single-cell flow sorting, and after whole genomeamplification, single amplified genome (SAG) is constructed by genome se-quencing assembly. SAG can be screened for functional or phylogeny genemarkers [134].1086.4. Limitations of the Research and Recommendations for Future Investigations4- In Chapter 4, enrichment cultures were established in order to identifymicroorganisms capable of reducing or oxidizing arsenic compounds. Twodifferent growth media to enrich chemoautotrophic arsenite oxidizing (CAO)microbes and heterotrophic arsenate reducing microbes were used, whichcontained nitrate and lactate, respectively. Although lactate acts as an elec-tron donor and carbon source for DARPs, there is a limitation to distinguishdissimilatory arsenate reduction from periplasmic arsenate reduction. It issuggested that, based on the taxonomic profile of the enrichments, differentmedia be used to understand their nutrient requirements.5- Biological contribution to arsenic removal is still hypothetical. Moreexperiments are suggested to investigate if microbes are needed for effectiveAs mineralization. For example, biotic versus abiotic experimental reactorsin the laboratory and inoculation with different species or enrichments (ar-senic oxidizing or arsenic reducing) can be helpful in this respect. The alter-native method is applying different design for functional screening targetingbiomineralization genes. In Chapter 5, functional metagenomic workflowis described to investigate the possible mechanisms involved in arsenic re-sponses in the biological based metal remediation system; however, we didnot investigate biomineralization and biosorption mechanisms in details. Infact, the design of functional metagenomic screening is an important step tointerpret the target genes or mechanisms. By providing the media containingthe required nutrients and substrates, we can examine a broad range of mech-anisms. In addition, alternative screening host such as archean methanogenis recommended.6- Regarding Chapter 5, to examine further identified genes related toars cluster or aox cluster, the transposon mutagenesis method can be em-ployed to determine the function and to study the encoding genes. Q-PCRof the aresenic resistant genes from core samples is also suggested to studythe mechanisms and arsenic removal pathways at genomic level.109Bibliography[1] A. Achour-Rokbani, A. Cordi, P. Poupin, P. Bauda, and P. Bil-lard. Characterization of the ars gene cluster from extremely arsenic-resistant Microbacterium sp. strain A33. Applied and EnvironmentalMicrobiology, 76(3):948–55, February 2010.[2] A. Adegoroye, P. Uhlik, O. Omotoso, Z. Xu, and J. Masliyah. AComprehensive Analysis of Organic Matter Removal from Clay-SizedMinerals Extracted from Oil Sands Using Low Temperature Ashingand Hydrogen Peroxide. Energy & Fuels, 23(7):3716–3720, July 2009.[3] C. R. Anderson and G. M. Cook. Isolation and characterization ofarsenate-reducing bacteria from arsenic-contaminated sites in NewZealand. Current microbiology, 48(5):341–7, May 2004.[4] I.J. Anderson, M. Sieprawska-Lupa, E. Goltsman, A. Lapidus,A. Copeland, T. Glavina Del Rio, H. Tice, E. Dalin, K. Barry,S. Pitluck, L. Hauser, M. Land, S. Lucas, P. Richardson, W.B Whit-man, and N.C Kyrpides. Complete genome sequence of Methanocor-pusculum labreanum type strain Z. Standards in Genomic Sciences,1(2):197–203, January 2009.[5] L.G.M. Baas Becking and D. Moore. Biogenic sulfides. EconomicGeology, 56(2):259–272, March 1961.[6] S.A. Baldwin, M. Rezadehbashi, M. Taupp, A. Mattes, and S. Hal-lam. The genetic diversity of microbes in a biochemical reactor treat-ing metal-rich landfill leachate: what this tells us about treatmentmechanisms. In 9th International Conference on Acid Rock Drainage(ICARD), page 153, Ottawa, Canada, 2012.[7] F. Battaglia-Brunet, C. Crouzet, A. Burnol, S. Coulon, D. Morin, andC. Joulian. Precipitation of arsenic sulphide from acidic water in afixed-film bioreactor. Water Research, 46(12):3923–33, August 2012.110Bibliography[8] F. Battaglia-Brunet and Y. Itard. A simple biogeochemical process re-moving arsenic from a mine drainage water. Geomicrobiology Journal,23:201–211, 2006.[9] F. Battaglia-Brunet, C. Joulian, F. Garrido, M.C. Dictor, D. Morin,K. Coupland, D Barrie Johnson, K. B Hallberg, and P. Baranger.Oxidation of arsenite by Thiomonas strains and characterization ofThiomonas arsenivorans sp. nov. Antonie van Leeuwenhoek, 89(1):99–108, January 2006.[10] F. Behar, V. Beaumont, and H.L. De B. Penteado. Rock-Eval 6 Tech-nology: Performances and Developments. Oil & Gas Science and Tech-nology, 56(2):111–134, March 2001.[11] C.M. Bethke, R.A. Sanford, M.F. Kirk, Q. Jin, and T.M. Flynn. Thethermodynamic ladder in geomicrobiology. American Journal of Sci-ence, 311(3):183–210, June 2011.[12] T.J. Beveridge. Role of Cellular Design in Bacterial Metal Accumula-tion and Mineralization. Annual Review of Microbiology, 43:147–171,1989.[13] T.J. Beveridge and R.G. Murray. Uptake and retention of metals bycell walls of Bacillus subtilis. Journal of Bacteriology, 127(3):1502–18,September 1976.[14] M.F.M. Bijmans, P.J. Van Helvoort, S.A. Dar, M. Dopson, P.N.L.Lens, and C.J.N. Buisman. Selective recovery of nickel over iron froma nickel-iron solution using microbial sulfate reduction in a gas-liftbioreactor. Water Research, 43(3):853–61, February 2009.[15] R. Blaskovich, S. Brienne, R.D. Beckie, K.U. Mayer, C.A. Arande, andC. Haupt. Quantitative mineralogy to assess waste rock weatheringat the Antamina mine. In 9th International Conference on Acid RockDrainage (ICARD), page 383, Ottawa, Canada, 2012.[16] C.A. Blindauer, M.D. Harrison, J.A. Parkinson, A.K. Robinson, J.S.Cavet, N.J. Robinson, and P.J. Sadler. A metallothionein containinga zinc finger within a four-metal cluster protects a bacterium fromzinc toxicity. Proceedings of the National Academy of Sciences of theUnited States of America, 98(17):9593–8, August 2001.111Bibliography[17] E.P. Blumenstein and J.J. Gusek. Designing a Biochemical Reac-tor for Selenium and Thallium Removal, from Bench Scale TestingThrough Pilot Construction. In P.R. Young Taylor, C.G. Anderson,and Y. Choi, editors, Hydrometallurgy 2008: Proceedings of the SixthInternational Symposium, pages 117–129, Phoenix, AZ, USA, 2008.Society for Mining, Metallurgy, Exploration.[18] B.C. Bostick and S. Fendorf. Arsenite sorption on troilite (FeS)and pyrite (FeS2). Geochimica et Cosmochimica Acta, 67(5):909–921,March 2003.[19] A. Breitenstein, J. Wiegel, C. Haertig, N. Weiss, J.R. Andreesen, andU. Lechner. Reclassification of Clostridium hydroxybenzoicum as Sed-imentibacter hydroxybenzoicus gen. nov., comb. nov., and descriptionof Sedimentibacter saalensis sp. nov.. International Journal of Sys-tematic and Evolutionary Microbiology, 52(Pt 3):801–7, May 2002.[20] C.G. Bryan, M. Marchal, F. Battaglia-Brunet, V. Kugler, C. Lemaitre-Guillier, D. Lie`vremont, P.N. Bertin, and F. Arse`ne-Ploetze. Carbonand arsenic metabolism in Thiomonas strains: differences revealeddiverse adaptation processes. BMC Microbiology, 9(1):127, January2009.[21] M. Buckwalter-Davis, H. Jaggardi, S. Morman, G. Plumlee, andH. Jamieson. Using automated mineralogy to evaluate bioaccessibilityof Pb-bearing mine waste. In Goldschmidt, pages 24–29, Montreal,Canada, 2012.[22] E.M. Burkhardt, S. Bischoff, D.M. Akob, G. Bu¨chel, and K. Ku¨sel.Heavy metal tolerance of Fe(III)-reducing microbial communities incontaminated creek bank soils. Applied and Environmental Microbiol-ogy, 77(9):3132–6, May 2011.[23] J.E. Butler, N.D. Young, and D.R. Lovley. Evolution of electron trans-fer out of the cell: comparative genomics of six Geobacter genomes.BMC Genomics, 11(1):40, January 2010.[24] J. Cabala, B. Zogala, and R. Dubiel. Geochemical and GeophysicalStudy of Historical Zn-Pb Ore Processing Waste Dump Areas (South-ern Poland). Journal of Environmental Studies, 17(5):693–700, 2008.112Bibliography[25] L. Cai, G. Liu, C. Rensing, and G. Wang. Genes involved in ar-senic transformation and resistance associated with different levels ofarsenic-contaminated soils. BMC microbiology, 9(1):4, January 2009.[26] J.G. Caporaso, J. Kuczynski, J. Stombaugh, K. Bittinger, F.D. Bush-man, E.K. Costello, N. Fierer, A.G. Pen˜a, J.K Goodrich, J.I Gordon,G.A Huttley, S.T Kelley, D. Knights, J.E. Koenig, R.E Ley, C.A.Lozupone, D. McDonald, B.D. Muegge, M. Pirrung, J. Reeder, J.R.Sevinsky, P.J. Turnbaugh, W.A. Walters, J. Widmann, T. Yatsunenko,J. Zaneveld, and R. Knight. QIIME allows analysis of high-throughputcommunity sequencing data. Nature Methods, 7(5):335–6, May 2010.[27] J.S. Chang, J.H. Lee, and I.S. Kim. Bacterial aox genotype fromarsenic contaminated mine to adjacent coastal sediment: evidencesfor potential biogeochemical arsenic oxidation. Journal of HazardousMaterials, 193:233–42, October 2011.[28] N.S. Chauhan, R. Ranjan, H.J. Purohit, V.C. Kalia, and R. Sharma.Identification of genes conferring arsenic resistance to Escherichia colifrom an effluent treatment plant sludge metagenomic library. FEMSmicrobiology Ecology, 67(1):130–9, January 2009.[29] M. Chovatia, J. Sikorski, M. Schro¨der, A. Lapidus, M. Nolan, H. Tice,T. Glavina Del Rio, A. Copeland, J.F. Cheng, S. Lucas, F. Chen,D. Bruce, L. Goodwin, S. Pitluck, N. Ivanova, K. Mavromatis,G. Ovchinnikova, A. Pati, A. Chen, K. Palaniappan, M. Land,L. Hauser, Y.J. Chang, C.D. Jeffries, P. Chain, E. Saunders, J.C.Detter, T. Brettin, M. Rohde, M. Go¨ker, S. Spring, J. Bristow, Vi.Markowitz, P. Hugenholtz, N.C. Kyrpides, H.P. Klenk, and J.A. Eisen.Complete genome sequence of Thermanaerovibrio acidaminovoranstype strain (Su883T). Standards in Genomic Sciences, 1(3):254–261,November 2009.[30] V.S. Coker, R.A.D. Pattrick, G.V.D. Laan, and J.R. Lloyd. Forma-tion of Magnetic Minerals by Non-Magnetotactic Prokaryotes. In DirkSchu¨ler, editor, Magnetoreception and Magnetosomes in Bacteria, vol-ume 3 of Microbiology Monographs, pages 275–300. Springer BerlinHeidelberg, Berlin, Heidelberg, 2007.[31] W.R. Cullen and K.J. Reimer. Arsenic speciation in the environment.Chemical Reviews, 89:713–764, May 2002.113Bibliography[32] D. E. Cummings, A. W. March, B. Bostick, S. Spring, F. Caccavo,S. Fendorf, and R. F. Rosenzweig. Evidence for Microbial Fe(III)Reduction in Anoxic, Mining-Impacted Lake Sediments (Lake Coeurd’Alene, Idaho). Applied and Environmental Microbiology, 66(1):154–162, January 2000.[33] A.L Dann, R.S Cooper, and J.P Bowman. Investigation and optimiza-tion of a passively operated compost-based system for remediation ofacidic, highly iron- and sulfate-rich industrial waste water. WaterResearch, 43(8):2302–16, May 2009.[34] B. Daus, J. Mattusch, R. Wennrich, and H. Weiss. Investigation onstability and preservation of arsenic species in iron rich water samples.Talanta, 58(1):57–65, 2002.[35] L. Drewniak, R. Matlakowska, and A. Sklodowsska. Arsenite andarsenate metabolism of Sinorhizobium sp. M14 living in the extremeenvironment of the Zloty Stok gold mine. Geomicrobiology Journal,25:363–370, 2008.[36] W. F. A Duncan. Long term operation of engineered anaerobic biore-actors and wetland cells treating zinc, arsenic and cadmium in seepage- results, longevity, cost and design issue. PhD thesis, University ofVictoria, 2010.[37] W.F.A. Duncan, A. Mattes, W.D. Gould, and F. Goodazi. Multi-stageBiological Treatment System for Removal of Heavy Metal Contami-nants. In 43rd Annual Conference of Metallurgists, pages 22–25, ON,Hamilton, 2004.[38] K. Duquesne, A. Lieutaud, J. Ratouchniak, D. Muller, M.C. Lett, andV. Bonnefoy. Arsenite oxidation by a chemoautotrophic moderatelyacidophilic Thiomonas sp.: from the strain isolation to the gene study.Environmental Microbiology, 10(1):228–37, January 2008.[39] R.C. Edgar. MUSCLE: multiple sequence alignment with high ac-curacy and high throughput. Nucleic Acids Research, 32(5):1792–7,January 2004.[40] R.C. Edgar. Search and clustering orders of magnitude faster thanBLAST. Bioinformatics, 26(19):2460–1, October 2010.[41] D.P. Faith and A.M. Baker. Phylogenetic diversity (PD) and biodi-versity conservation: some bioinformatics challenges, 2006.114Bibliography[42] J.K. Fawell and R. Mascarenhas. Arsenic in Drinking-water , Back-ground document for development of WHO Guidelines for Drinking-water Quality. Technical report, WHO Press,World Health Organiza-tion.[43] R.D. Finn, J. Clements, and S.R. Eddy. HMMER web server: inter-active sequence similarity searching. Nucleic Acids Research, 39(WebServer issue):W29–37, July 2011.[44] K. T. Finneran. Rhodoferax ferrireducens sp. nov., a psychrotolerant,facultatively anaerobic bacterium that oxidizes acetate with the reduc-tion of Fe(III). International Journal of Systematic and EvolutionaryMicrobiology, 53(3):669–673, May 2003.[45] K.T. Finneran, C.V. Johnsen, and D.R. Lovley. Rhodoferax ferrire-ducens sp. nov., a psychrotolerant, facultatively anaerobic bacteriumthat oxidizes acetate with the reduction of Fe(III). International Jour-nal of Systematic and Evolutionary Microbiology, 53(Pt 3):669–73,May 2003.[46] E. Fisher, A. M. Dawson, G. Polshyna, J. Lisak, B. Crable, E. Perera,M. Ranganathan, M. Thangavelu, P. Basu, and J.F. Stolz. Transfor-mation of inorganic and organic arsenic by Alkaliphilus oremlandii sp.nov. strain OhILAs. Annals of the New York Academy of Sciences,1125:230–41, March 2008.[47] T. Ford, J. Jay, A. Patel, M. Kile, P. Prommasith, T. Galloway,R. Sanger, K. Smith, and M. Depledge. Use of ecotoxicological toolsto evaluate the health of New Bedford Harbor sediments: a microbialbiomarker approach. Environmental Health perspectives, 113(2):186–91, February 2005.[48] D. Fortin and T. J. Beveridge. Microbial sulfate reduction withinsulfidic mine tailings: Formation of diagenetic Fe sulfides. Geomicro-biology Journal, 14(1):1–21, January 1997.[49] D. Fortin, B. Davis, and T.J. Beveridge. Role of Thiobacillus andsulfate-reducing bacteria in iron biocycling in oxic and acidic minetailings. FEMS Microbiology Ecology, 21(1):11–24, September 1996.[50] D. Fortin and T. Praharaj. Role of microbial activity in Fe and Scycling in sub-Oxic to anoxic aulfide-rich mine tailings: A mini-review.Journal of Nuclear and Radiochemical Sciences, 6(1):39–42, 2005.115Bibliography[51] T. Gaonkar and S. Bhosle. Effect of metals on a siderophore produc-ing bacterial isolate and its implications on microbial assisted biore-mediation of metal contaminated soils. Chemosphere, 93(9):1835–43,November 2013.[52] L. Giloteaux, D.E. Holmes, K.H Williams, K.C. Wrighton, M.J.Wilkins, Alison P Montgomery, Jessica A Smith, Roberto Orellana,Courtney A Thompson, Thomas J Roper, Philip E Long, and Derek RLovley. Characterization and transcription of arsenic respiration andresistance genes during in situ uranium bioremediation. The ISMEJournal, 7(2):370–83, February 2013.[53] S. Glasauer, S. Langley, and T.J. Beveridge. Intracellular iron mineralsin a dissimilatory iron-reducing bacterium. Science, 295(5552):117–9,January 2002.[54] A. Grabowski, B.J. Tindall, V. Bardin, D. Blanchet, and C. Jeanthon.Petrimonas sulfuriphila gen. nov., sp. nov., a mesophilic fermenta-tive bacterium isolated from a biodegraded oil reservoir. InternationalJournal of Systematic and Evolutionary Microbiology, 55(Pt 3):1113–21, May 2005.[55] S. Guindon, J.F. Dufayard, V. Lefort, M. Anisimova, W. Hordijk,and O. Gascuel. New algorithms and methods to estimate maximum-likelihood phylogenies: assessing the performance of PhyML 3.0. Sys-tematic Biology, 59(3):307–21, May 2010.[56] J.J. Gusek. A Periodic Table Of Passive Treatment For Mining Influ-enced Water. In National Meeting of the American Society of Miningand Reclamation, Billings, 2009. ASMR.[57] J.J. Gusek, J.T. Gormley, and J.W. Scheetz. Design and constructionaspects of pilot-scale passive treatment systems for acid rock drainageat metal mines. In Hydrometallurgy, pages 777–793. Springer Nether-lands, Dordrecht, 1994.[58] G. Haferburg and E. Kothe. Microbes and metals: interactions in theenvironment. Journal of Basic Microbiology, 47(6):453–67, December2007.[59] X. Han and J.D Gu. Sorption and transformation of toxic metalsby microorganisms. In R. Mitchell and D. Gu, J., editors, Environ-116Bibliographymental Microbiology, chapter 7, pages 153–176. John Wiley & Sons,Chichester, 2010.[60] A. Heinrich-Salmeron, A. Cordi, C. Brochier-Armanet, D. Halter,C. Pagnout, E. Abbaszadeh-fard, D. Montaut, F. Seby, P.N Bertin,P. Bauda, and F. Arse`ne-Ploetze. Unsuspected diversity of arsenite-oxidizing bacteria as revealed by widespread distribution of theaoxB gene in prokaryotes. Applied and Environmental Microbiology,77(13):4685–92, July 2011.[61] C. Hohmann, M. Obst, K. Benzerara, G. Morin, and A. Kappler.Influence of Arsenic on the precipitation of Iron-oxides by Acidovoraxsp. BoFeN1. Technical report, Canadian light source, 2009.[62] Y. Huang, H. Li, C. Rensing, K. Zhao, L. Johnstone, and G. Wang.Genome sequence of the facultative anaerobic arsenite-oxidizing andnitrate-reducing bacterium Acidovorax sp. strain NO1. Journal ofBacteriology, 194(6):1635–6, March 2012.[63] K. Hudson-Edwards and J. Santini. Arsenic-Microbe-Mineral Interac-tions in Mining-Affected Environments. Minerals, 3(4):337–351, Oc-tober 2013.[64] W.J. Hunter, L.D. Kuykendall, and D.K. Manter. Rhizobium selenire-ducens sp. nov.: a selenite-reducing alpha-Proteobacteria isolated froma bioreactor. Current Microbiology, 55(5):455–60, November 2007.[65] D. Hyatt, G.L. Chen, P.F. Locascio, M.L. Land, F.W. Larimer, andL.J. Hauser. Prodigal: prokaryotic gene recognition and translationinitiation site identification. BMC Bioinformatics, 11(1):119, January2010.[66] C. Kira Jackson, Iris Koch, and Kenneth J. Reimer. Mechanisms ofdissolved arsenic removal by biochemical reactors: A bench- and field-scale study. Applied Geochemistry, 29:174–181, 2013.[67] L. Jedynak, P. Kieroski, J. Kowalska, and J. Golimowski. Specia-tion analysis of arsenic by HPLC-UV in highly contaminated watersamples. Chemia Analityczna, Vol. 53, N:557–568, 2008.[68] T. Jong and D.L. Parry. Removal of sulfate and heavy metals bysulfate reducing bacteria in short-term bench scale upflow anaerobicpacked bed reactor runs. Water Research, 37(14):3379–89, August2003.117Bibliography[69] N. Khan, A. Mishra, P.S. Chauhan, Y.K. Sharma, and C.S. Nau-tiyal. Paenibacillus lentimorbus enhances growth of chickpea (Cicerarietinum L.) in chromium-amended soil. Antonie Van Leeuwenhoek,101(2):453–9, February 2012.[70] S.T. Khan, Y. Horiba, M. Yamamoto, and A. Hiraishi. Members ofthe family Comamonadaceae as primary poly(3-hydroxybutyrate-co-3-hydroxyvalerate)-degrading denitrifiers in activated sludge as revealedby a polyphasic approach. Applied and Environmental Microbiology,68(7):3206–14, July 2002.[71] M. Khoshnoodi, G. Dipple, and S.A. Baldwin. Mineralogical Study ofa Biologically-Based Treatment System That Removes Arsenic, Zincand Copper from Landfill Leachate. Minerals, 3(4):427–449, December2013.[72] M. Khoshnoodi, M. Taupp, K. Mewis, S. Hallam, and S.A. Baldwin.Screening for Arsenate and Arsenite Resistant Microbes in PassiveTreatment Systems for Metal Leachate Remediation. In the 9th In-ternational Conference on Acid Rock Drainage, pages 20–24 May, Ot-tawa, Canada, 2012.[73] K.M Konwar, N.W Hanson, A.P. Page´, and S.J. Hallam. MetaPath-ways: a modular pipeline for constructing pathway/genome databasesfrom environmental sequence information. BMC Bioinformatics,14(1):202, January 2013.[74] E. Kruse and V. A. Ettel. Solubility and stability of scorodite,FeAsO42H2O: new data and further discussion. The American Min-eralogist, 73(7-8):850–854, 1988.[75] M. Krzywinski, J. Schein, I. Birol, J. Connors, R. Gascoyne, D. Hors-man, S.J. Jones, and M.A. Marra. Circos: an information aesthetic forcomparative genomics. Genome Research, 19(9):1639–45, September2009.[76] K. Kusel, T. Dorsch, G. Acker, and E. Stackebrandt. Microbial Reduc-tion of Fe(III) in Acidic Sediments: Isolation of Acidiphilium cryptumJF-5 Capable of Coupling the Reduction of Fe(III) to the Oxidation ofGlucose. Applied and Environmental Microbiology, 65(8):3633–3640,August 1999.118Bibliography[77] M. Labrenz, G.K. Druschel, T. Thomsen-Ebert, B. Gilbert, S.A.Welch, K.M. Kemner, G.A. Logan, R.E. Summons, G.D. Stasio, P.L.Bond, B. Lai, Shelly D. Kelly, and Jillian F. Banfield. Formationof Sphalerite (ZnS) Deposits in Natural Biofilms of Sulfate-ReducingBacteria. Science, 290(5497):1744–1747, December 2000.[78] K. Labutti, S. Mayilraj, A. Clum, S. Lucas, T. Glavina Del Rio,M. Nolan, H. Tice, J.F. Cheng, S. Pitluck, K. Liolios, N. Ivanova,K. Mavromatis, N. Mikhailova, A. Pati, L. Goodwin, A. Chen,K. Palaniappan, M. Land, L. Hauser, Y.J. Chang, C.D Jeffries,M. Rohde, S. Spring, M. Go¨ker, T. Woyke, J. Bristow, J.A. Eisen,V. Markowitz, P. Hugenholtz, N.C. Kyrpides, H.P. Klenk, andA. Lapidus. Permanent draft genome sequence of Dethiosulfovibriopeptidovorans type strain (SEBR 4207). Standards in Genomic Sci-ences, 3(1):85–92, January 2010.[79] D. Langmuir, J. Mahoney, and J. Rowson. Solubility products of amor-phous ferric arsenate and crystalline scorodite (FeAsO42H2O) andtheir application to arsenic behavior in buried mine tailings. Geochim-ica et Cosmochimica Acta, 70(12):2942–2956, 2006.[80] H. Larratt, M. Freburg, and R. Hamaguchi. Developing TailingsPonds and Pit Lakes as Bioreactors and Habitat Cost-Effective Suc-cesses at Highland Valley Copper. In British Columbias 31st An-nual Mine Reclamation Symposium, Squamish, Canada, 2007. TheBritish Columbia Technical and Research Committee on Reclamation(TRCR).[81] P.A. Lawson, B. Wawrik, T.D. Allen, C.N. Johnson, C.R. Marks, R.S.Tanner, B.H. Harriman, D. Strapoc, and A.V. Callaghan. Youngiibac-ter fragilis gen. nov., sp. nov., isolated from natural gas production-water and reclassification of Acetivibrio multivorans as Youngiibactermultivorans comb. nov. International Journal of Systematic and Evo-lutionary Microbiology, 64(1):198–205, January 2014.[82] I. Letunic and P. Bork. Interactive Tree Of Life (iTOL): an onlinetool for phylogenetic tree display and annotation. Bioinformatics,23(1):127–8, January 2007.[83] A. Lieutaud, R. van Lis, S. Duval, L. Capowiez, D. Muller, R. Lebrun,S. Lignon, M. Fardeau, M.C. Lett, W. Nitschke, and B. Schoepp-Cothenet. Arsenite oxidase from Ralstonia sp. 22: characterization of119Bibliographythe enzyme and its interaction with soluble cytochromes. The Journalof Biological Chemistry, 285(27):20433–41, July 2010.[84] D.R. Lovley. Dissimilatory metal reduction. Annual Review of Micro-biology, 47:263–90, January 1993.[85] D.R. Lovley. Environmental Microbe-metal Interactions. ASM, 2000.[86] C. Lozupone and R. Knight. UniFrac: a new phylogenetic method forcomparing microbial communities. Applied and Environmental Micro-biology, 71(12):8228–35, December 2005.[87] P. Lu and C. Zhu. Arsenic EhpH diagrams at 25C and 1 bar. Envi-ronmental Earth Sciences, 62(8):1673–1683, July 2010.[88] S. Lu, S.H. Ryu, B.S. Chung, Y.R. Chung, W. Park, and C.O. Jeon.Simplicispira limi sp. nov., isolated from activated sludge. Interna-tional Journal of Systematic and Evolutionary Microbiology, 57(Pt1):31–4, January 2007.[89] R.E. Macur, J.T. Wheeler, T.R. McDermott, and W.P. Inskeep. Mi-crobial populations associated with the reduction and enhanced mo-bilization of arsenic in mine tailings. Environmental Science & Tech-nology, 35(18):3676–82, September 2001.[90] J.M. Macy, K. Nunan, K.D. Hagen, D.R. Dixon, P.J. Harbour,M. Cahill, and L.I. Sly. Chrysiogenes arsenatis gen. nov., sp. nov.,a new arsenate-respiring bacterium isolated from gold mine wastew-ater. International Journal of Systematic Bacteriology, 46(4):1153–7,October 1996.[91] D. Malasarn, C.W. Saltikov, K.M. Campbell, J.M. Santini, J.G. Her-ing, and D.K. Newman. arrA is a reliable marker for As(V) respiration.Science, 306(5695):455, October 2004.[92] B.A. Manning and S. Goldberg. Adsorption and Stability of Ar-senic(III) at the Clay Mineral - Water Interface. Environmental Sci-ence & Technology, 31:2005–2022, 1997.[93] R.P. Mason. The Methylation of Metals and Metalloids in AquaticSystems. In Anica Dricu, editor, Methylation - From DNA, RNA andHistones to Diseases and Treatment. InTech, Rijeka, Croatia, Novem-ber 2012.120Bibliography[94] A. Mattes. The Trail Biochemical Reactor and the Removal of Arsenicwith a Focus on the Specific Role of Iron. PhD thesis, The Universityof Guelph, August 2013.[95] A. Mattes, L.J. Evans, Douglas G.W., W.F.A. Duncan, andS. Glasauer. The long term operation of a biologically based treatmentsystem that removes As, S and Zn from industrial (smelter operation)landfill seepage. Applied Geochemistry, 26(11):1886–1896, November2011.[96] A. Mattes, D. Gould, M. Taupp, and S. Glasauer. A Novel Au-totrophic Bacterium Isolated from an Engineered Wetland SystemLinks Nitrate-Coupled Iron Oxidation to the Removal of As, Zn andS. Water, Air, & Soil Pollution, 224(4):1490, March 2013.[97] K. Mavromatis, M. Yasawong, O. Chertkov, A. Lapidus, S. Lucas,M. Nolan, T. Del Rio, H. Tice, J.F. Cheng, S. Pitluck, K. Liolios,N. Ivanova, R. Tapia, C. Han, D. Bruce, L. Goodwin, A. Pati, A. Chen,K. Palaniappan, M. Land, L. Hauser, Y.J. Chang, C.D. Jeffries, J.C.Detter, M. Rohde, E. Brambilla, S. Spring, M. Go¨ker, J. Sikorski,T. Woyke, J. Bristow, J.A. Eisen, V. Markowitz, P. Hugenholtz, H.P.Klenk, and N.C. Kyrpides. Complete genome sequence of Spirochaetasmaragdinae type strain (SEBR 4228). Standards in Genomic Sci-ences, 3(2):136–44, January 2010.[98] B.C. McBride and R.S. Wolfe. Biosynthesis of dimethylarsine bymethanobacterium. Biochemistry, 10(23):4312–4317, November 1971.[99] A.P. McHale and S. McHale. Microbial biosorption of metals: po-tential in the treatment of metal pollution. Biotechnology Advances,12(4):647–52, January 1994.[100] K. Mewis, Z. Armstrong, Y.C. Song, S.A. Baldwin, S.G. Withers, andS.J. Hallam. Biomining active cellulases from a mining bioremediationsystem. Journal of Biotechnology, 167(4):462–471, 2013.[101] K. Michalke, E.B. Wickenheiser, M. Mehring, A.V. Hirner, andR. Hensel. Production of volatile derivatives of metal(loid)s by mi-croflora involved in anaerobic digestion of sewage sludge. Applied andEnvironmental Microbiology, 66(7):2791–6, July 2000.[102] L.J. Michel, L.E. Macaskie, and A.C. Dean. Cadmium accumula-tion by immobilized cells of a Citrobacter sp. using various phosphate121Bibliographydonors. Biotechnology and Bioengineering, 28(9):1358–65, September1986.[103] S. Mirete, C.G. de Figueras, and J.E. Gonza´lez-Pastor. Novel nickelresistance genes from the rhizosphere metagenome of plants adaptedto acid mine drainage. Applied and Environmental Microbiology,73(19):6001–11, October 2007.[104] J.W. Moreau, R.I. Webb, and J.F. Banfield. Ultrastructure,aggregation-state, and crystal growth of biogenic nanocrystalline spha-lerite and wurtzite. American Mineralogist, 89(7):950–960, July 2004.[105] R. Mukhopadhyay and B.P. Rosen. Arsenate reductases in prokaryotesand eukaryotes. Environmental Health Perspectives, 110 Suppl:745–8,October 2002.[106] D. Muller, C. Me´digue, S. Koechler, V. Barbe, M. Barakat, E. Talla,V. Bonnefoy, E. Krin, F. Arsene-Ploetze, C. Carapito, M. Chan-dler, B. Cournoyer, S. Cruveiller, C. Dossat, S. Duval, M. Hey-mann, E. Leize, A. Lieutaud, D. Lievremont, Y. Makita, S. Mangenot,W. Nitschke, P. Ortet, N. Perdrial, B. Schoepp, P. Siguier, D. D. Sime-onova, Z. Rouy, B. Segurens, E. Turlin, D. Vallenet, A. V. Dorsselaer,S. Weiss, J. Weissenbach, M.C. Lett, A. Danchin, and P.N Bertin.A tale of two oxidation states: bacterial colonization of arsenic-aichenvironments. PLoS Genetics, 3:0519–0530, 2007.[107] C.N. Mulligan, R.N. Yong, and B.F. Gibbs. Heavy metal removalfrom sediments by biosurfactants. Journal of Hazardous Materials,85(1):111–125, 2001.[108] I. Nancucheo and D. B. Johnson. Selective removal of transition metalsfrom acidic mine waters by novel consortia of acidophilic sulfidogenicbacteria. Microbial Biotechnology, 5(1):34–44, January 2012.[109] C.M. Neculita, G.J. Zagury, and B. Bussie`re. Passive treatment of acidmine drainage in bioreactors using sulfate-reducing bacteria: criticalreview and research needs. Journal of Environmental Quality, 36(1):1–16, January 2007.[110] C.M. Neculita, G.J. Zagury, and B. Bussie`re. Effectiveness of sulfate-reducing passive bioreactors for treating highly contaminated acidmine drainage: I. Effect of hydraulic retention time. Applied Geo-chemistry, 23(12):3442–3451, December 2008.122Bibliography[111] D.K. Newman, D. Ahmann, and F.M.M. Morel. A brief review ofmicrobial arsenate respiration. Geomicrobiology Journal, 15(4):255–268, October 1998.[112] D.K. Newman, E.K. Kennedy, J.D. Coates, D. Ahmann, D.J. Ellis,D.R. Lovley, and F.M. Morel. Dissimilatory arsenate and sulfate re-duction in Desulfotomaculum auripigmentum sp. nov. Archives ofMicrobiology, 168(5):380–8, November 1997.[113] D.H. Nies and S. Silver. Ion efflux systems involved in bacterial metalresistances. Journal of Industrial Microbiology, 14(2):186–199, Febru-ary 1995.[114] A. Niggemyer, S. Spring, E. Stackebrandt, and R.F. Rosenzweig. Iso-lation and characterization of a novel As(V)-reducing bacterium: im-plications for arsenic mobilization and the genus Desulfitobacterium.Applied and Environmental Microbiology, 67(12):5568–80, December2001.[115] D. K. Nordstrom and D.G. Archer. Arsenic thermodynamic data andenvironmental geochemistry. In Alan H. Welch and Kenneth G. Stol-lenwerk, editors, Arsenic in Ground Water, pages 1–25. Kluwer Aca-demic Publishers, Boston, 2003.[116] P.A. O’Day. Chemistry and Mineralogy of Arsenic. Elements, 2(2):77–83, April 2006.[117] R.S. Oremland and J.F. Stolz. The ecology of arsenic. Science,300(5621):939–44, May 2003.[118] T.H. Osborne, H.E. Jamieson, K.A. Hudson-Edwards, D.K. Nord-strom, S.R. Walker, S.A. Ward, and J.M. Santini. Microbial oxidationof arsenite in a subarctic environment: diversity of arsenite oxidasegenes and identification of a psychrotolerant arsenite oxidiser. BMCMicrobiology, 10(1):205, January 2010.[119] D. Pa´ez-Espino, J. Tamames, V. de Lorenzo, and D. Ca´novas. Mi-crobial responses to environmental arsenic. Biometals, 22(1):117–30,February 2009.[120] N. Parmar, L.A. Warren, E.E. Roden, and F.G. Ferris. Solid phasecapture of strontium by the iron reducing bacteria Shewanella algastrain BrY. Chemical Geology, 169(3-4):281–288, September 2000.123Bibliography[121] J.P. Pavissich, M. Silva, and B. Gonza´lez. Sulfate reduction, molecu-lar diversity, and copper amendment effects in bacterial communitiesenriched from sediments exposed to copper mining residues. Environ-mental Toxicology and Chemistry, 29(2):256–64, February 2010.[122] D. Pirrie, A. R. Butcher, M. R. Power, P. Gottlieb, and G. L. Miller.Rapid quantitative mineral and phase analysis using automated scan-ning electron microscopy (QemSCAN); potential applications in foren-sic geoscience. Geological Society, 232(1):123–136, January 2004.[123] T. Praharaj and D. Fortin. Seasonal variations of microbial sulfateand iron reduction in alkaline PbZn mine tailings (Ontario, Canada).Applied Geochemistry, 23(12):3728–3740, December 2008.[124] K.S. Prasad, A.L. Ramanathan, J. Paul, V. Subramanian, andR. Prasad. Biosorption of arsenite (As(+3)) and arsenate (As(+5))from aqueous solution by Arthrobacter sp. biomass. EnvironmentalTechnology, 34(17-20):2701–8, 2013.[125] K.D. Pruitt, T. Tatusova, G.R. Brown, and D.R. Maglott. NCBI Ref-erence Sequences (RefSeq): current status, new features and genomeannotation policy. Nucleic Acids Research, 40(Database issue):D130–5, January 2012.[126] I. Puigdomenech. MEDUSA (make equilibrium diagrams using so-phisticated algorithms) programs. Royal, 2002.[127] I. Puigdomenech. HYDRA: Hydrochemical Equilibrium-ConstantDatabase Software. Sweden: Royal Institute of Technology, 2004.[128] C. Quast, E. Pruesse, P. Yilmaz, J. Gerken, T. Schweer, P. Yarza,J. Peplies, and F.O. Glo¨ckner. The SILVA ribosomal RNA genedatabase project: improved data processing and web-based tools. Nu-cleic Acids Research, 41(Database issue):D590–6, January 2013.[129] E. Ramos-Padro´n, S. Bordenave, S. Lin, I.M. Bhaskar, X. Dong, C.W.Sensen, J. Fournier, G. Voordouw, and L.M. Gieg. Carbon and sul-fur cycling by microbial communities in a gypsum-treated oil sandstailings pond. Environmental Science & Technology, 45(2):439–46,January 2011.[130] M.P. Reis, F.A.R. Barbosa, E. Chartone-Souza, and A.M.A. Nasci-mento. The prokaryotic community of a historically mining-impacted124Bibliographytropical stream sediment is as diverse as that from a pristine streamsediment. Extremophiles, 17(2):301–9, March 2013.[131] D. Reisman, T. Ruttcowski, P. Smart, and J.J. Gusek. The Construc-tion and Instrumentation of a Pilot Treatment System at the StandardMine Superfund Site, Crested Butte, CO. In Proceedings of the 2008National Meeting of the American Society of Mining and Reclamation,Richmond, VA, USA, 2008.[132] M. Rezadehbashi, M. Khoshnoodi, M. Taupp, S. Hallam, and S.A.Baldwin. Microbial diversity of a biochemical reactor removing metalsfrom smelter waste leachate. In Canadian Society of Microbiology,Vancouver, 2012.[133] K. Richter, M. Schicklberger, and J. Gescher. Dissimilatory reductionof extracellular electron acceptors in anaerobic respiration. Appliedand Environmental Microbiology, 78(4):913–21, February 2012.[134] C. Rinke, P. Schwientek, A. Sczyrba, N.N. Ivanova, I.J. Anderson, J.F.Cheng, A. Darling, S. Malfatti, B.K. Swan, E.A. Gies, J.A. Dodsworth,B.P. Hedlund, G. Tsiamis, S.M Sievert, Wen-Tso Liu, Jonathan AEisen, Steven J Hallam, Nikos C Kyrpides, Ramunas Stepanauskas,Edward M Rubin, Philip Hugenholtz, and Tanja Woyke. Insights intothe phylogeny and coding potential of microbial dark matter. Nature,499(7459):431–7, July 2013.[135] P.A. Riveros, J.E. Dutrizac, and P. Spencer. Arsenic Disposal Prac-tices in the Metallurgical Industry. Canadian Metallurgical Quarterly,40(4):395–420, January 2001.[136] R.G. Robins. Solubility and stability of scorodite, FeAsO4. 2H2O:Discussion. American Mineralogist, 72:842–844, 1987.[137] R.G. Robins. Arsenic hydrometallurgy. In R.G. Reddy, J.L. Hendrix,and P.B. Queneau, editors, Arsenic Metallurgy: Fundamentals andApplications, pages 215–247. Minerals, Metals, & Materials Society,PA, USA, 1988.[138] R.G. Robins, T. Nishimura, and P. Singh. Removal of Arsenic fromDrinking Water by Precipitation, Adsorption or Cementation. InM.Feroze Ahmed, M.Ashraf Ali, and Zafar Adeel, editors, Interna-tional Workshop on Technologies for Arsenic Removal from Drink-ing Water, pages 31–42. Bangladesh University of Engineering and125BibliographyTechnology, Dhaka The United Nations University, Tokyo, Dhaka,Bangladesh, 2001.[139] P.E. Rosenberg and F.F. Foit. The stability of transition metaldolomites in carbonate systems: a discussion. Geochimica et Cos-mochimica Acta, 43:951–955, 1979.[140] J.M. Santini, L.I. Sly, R.D. Schnagl, and J.M. Macy. A newchemolithoautotrophic arsenite-oxidizing bacterium isolated from agold mine: phylogenetic, physiological, and preliminary biochemicalstudies. Applied and Environmental Microbiology, 66(1):92–7, Jan-uary 2000.[141] J.M. Santini, L.I. Sly, and A. Wen. New Arsenite-Oxidizing Bacte-ria Isolated from Australian Gold Mining Environments–PhylogeneticRelationships. Geomicrobiology, 19:67–76, 2002.[142] J.M. Santini, I.C.A. Streimann, and R.N. Vanden Hoven. Bacillus ma-cyae sp. nov., an arsenate-respiring bacterium isolated from an Aus-tralian gold mine. International Journal of Systematic and Evolution-ary Microbiology, 54(Pt 6):2241–4, November 2004.[143] S. Schultze-Lam, D. Fortin, B.S. Davis, and T.J. Beveridge. Mineral-ization of bacterial surfaces. Chemical Geology, 132(1):171–181, 1996.[144] V.K. Sharma and M. Sohn. Aquatic arsenic: toxicity, specia-tion, transformations, and remediation. Environment International,35(4):743–59, May 2009.[145] J.T. Simpson, K. Wong, S.D. Jackman, J.E. Schein, S.J.M. Jones,and I. Birol. ABySS: a parallel assembler for short read sequencedata. Genome Research, 19(6):1117–23, June 2009.[146] D.B. Singh, R.K. Tiwary, and B.K. Tewary. Feldspar for the Treatmentof As(III) Contaminated Water. Water Quality Research Journal ofCanada, 37(4):757–771, 2002.[147] S.R. Starkenburg, F.W. Larimer, L.Y. Stein, M.G. Klotz, P.S.G.Chain, L.A. Sayavedra-Soto, A.T. Poret-Peterson, M.E. Gentry, D.J.Arp, B. Ward, and P.J. Bottomley. Complete genome sequence ofNitrobacter hamburgensis X14 and comparative genomic analysis ofspecies within the genus Nitrobacter. Applied and Environmental Mi-crobiology, 74(9):2852–63, May 2008.126Bibliography[148] J.F. Stolz, P. Basu, J.M. Santini, and R.S. Oremland. Arsenic andSelenium in Microbial Metabolism. Annual review of microbiology,60:107–130, 2006.[149] J.F. Stolz, P. Basu, J.M. Santini, and R.S Oremland. Arsenic andselenium in microbial metabolism. Annual Review of Microbiology,60:107–30, January 2006.[150] J.F. Stolz and R.S. Oremland. Bacterial respiration of arsenic andselenium. FEMS Microbiology Reviews, 23(5):615–627, October 1999.[151] U. Stottmeister, S. Buddhawong, P. Kuschk, A. Wiessner, and J. Mat-tusch. Constructed wetlands and their performance for treatment ofwater contaminated with arsenic and heavy metals. In Irena Twar-dowska, Herbert E. Allen, Max M. Ha¨ggblom, and Sebastian Stefa-niak, editors, Soil and Water Pollution Monitoring, Protection andRemediation, volume 69 of NATO Science Series, pages 417–432.Springer Netherlands, Dordrecht, 2006.[152] D. Susanti, J.H. Wong, W.H. Vensel, U. Loganathan, R. DeSantis,R.A Schmitz, M. Balsera, B.B. Buchanan, and B. Mukhopadhyay.Thioredoxin targets fundamental processes in a methane-producingarchaeon, Methanocaldococcus jannaschii. Proceedings of the NationalAcademy of Sciences of the United States of America, 111(7):2608–13,February 2014.[153] M. Taupp, S. Lee, A. Hawley, J. Yang, and S.J. Hallam. Large in-sert environmental genomic library production. Journal of VisualizedExperiments, (31), January 2009.[154] M. Taupp, K. Mewis, and S.J. Hallam. The art and design of functionalmetagenomic screens. Current Opinion in Biotechnology, 22(3):465–72, July 2011.[155] F. Thomas, R.A. Diaz-Bone, O. Wuerfel, B. Huber, K. Weidenbach,R.A. Schmitz, and R. Hensel. Connection between multimetal(loid)methylation in methanoarchaea and central intermediates of methano-genesis. Applied and Environmental Microbiology, 77(24):8669–75, De-cember 2011.[156] K.J. Tufano, C. Reyes, C.W. Saltikov, and S. Fendorf. ReductiveProcesses Controlling Arsenic Retention: Revealing the Relative Im-127Bibliographyportance of Iron and Arsenic Reduction. Environmental Science &Technology, 42(22):8283–8289, November 2008.[157] G.W. Tyson, J. Chapman, P. Hugenholtz, E.E. Allen, R.J. Ram, P.M.Richardson, V.V. Solovyev, E.M. Rubin, D.S. Rokhsar, and J.F. Ban-field. Community structure and metabolism through reconstructionof microbial genomes from the environment. Nature, 428(6978):37–43,March 2004.[158] A.E. Valenzuela. Arsenic Management in the Metallurgical Industry:.PhD thesis, University of Laval, 2000.[159] E.E. Van der Hoek, P.A. Bonouvrie, and R.N.J. Comans. Sorption ofAs and Se on mineral components of fly ash: Relevance for leachingprocesses. Applied Geochemistry, 9(4):403–412, July 1994.[160] C. Vasconcelos, J.A. McKenzie, S. Bernasconi, D. Grujic, and A.J.Tiens. Microbial mediation as a possible mechanism for naturaldolomite formation at low temperatures. Nature, 377(6546):220–222,September 1995.[161] B.W. Vink. Stability relations of antimony and arsenic compounds inthe light of revised and extended Eh-pH diagrams. Chemical Geology,130(1-2):21–30, August 1996.[162] S. Wang and C.N. Mulligan. Natural attenuation processes for re-mediation of arsenic contaminated soils and groundwater. Journal ofHazardous Materials, 138(3):459–70, December 2006.[163] S. Wang and C.N. Mulligan. Occurrence of arsenic contamination inCanada: sources, behavior and distribution. The Science of the TotalEnvironment, 366(2-3):701–21, August 2006.[164] R. Warthmann, Y. Van Lith, C. Vasconcelos, J.A. McKenzie, andA.M. Karpoff. Bacterially induced dolomite precipitation in anoxicculture experiments. Geology, 28(12):1091–1094, December 2000.[165] P.G. Whitehead, B.J. Cosby, and H. Prior. The Wheal Jane wetlandsmodel for bioremediation of acid mine drainage. Science of The TotalEnvironment, 338(1):125–135, 2005.[166] R. K. Wieder. A survey of constructed wetlands for acid coal minedrainage treatment in the Eastern United States. Wetlands, 9(2):299–315, December 1989.128[167] S. Winch, H.J. Mills, J.E Kostka, D. Fortin, and D.R.S. Lean. Iden-tification of sulfate-reducing bacteria in methylmercury-contaminatedmine tailings by analysis of SSU rRNA genes. FEMS microbiologyEcology, 68(1):94–107, April 2009.[168] K. Zhang, L. Song, and X. Dong. Proteiniclasticum ruminis gen. nov.,sp. nov., a strictly anaerobic proteolytic bacterium isolated from yakrumen. International Journal of Systematic and Evolutionary Micro-biology, 60(Pt 9):2221–5, September 2010.[169] X. Zhang, Y. Jia, X. Wang, and L. Xu. Phylogenetic analysis and ar-senate reduction effect of the arsenic-reducing bacteria enriched fromcontaminated soils at an abandoned smelter site. Journal of Environ-mental Sciences (China), 20(12):1501–7, January 2008.129Appendix ASupplementary DataChapter 1Table A.1: Chemical analysis of the BCR core solid samples.Year 2008 2008 2008 2009 2009 2009 2009 2009 2009Month July July July Apr Apr Apr Oct Oct OctBorehole TS1 TS2 TS3 TS4 TS5 TS6 TS7 TS8 TS9Core 02 04 05 03/04 06 01/02 01 03 05/04Section-cm 20-300-10 20-3020-25/0-55-15 20-25/0-520-300-10 20-25/0-5SiO2(%) 22 43.00 32.70 47.20 NA 45.10 37.70 40.40 43.2 62.00Al2O3(%) 2 7.94 6.21 8.67 NA 8.38 6.57 6.87 7.49 12.20Fe2O3(%) 2 1.58 1.36 1.86 NA 2.12 2.24 1.29 2.35 1.88CaO(%) 2 15.65 24.10 17.00 NA 11.80 17.80 7.02 17.10 6.63MgO(%) 2 1.09 0.86 1.00 NA 1.74 1.12 1.10 1.72 0.86Na2O(%) 2 1.94 1.60 2.15 NA 2.05 1.68 1.82 1.80 3.19K2O(%) 2 1.92 1.56 2.14 NA 2.01 1.61 1.74 1.98 3.38Cr2O3(%) 2 0.02 0.01 0.02 NA 0.03 0.02 0.02 0.02 0.02TiO2(%) 2 0.25 0.18 0.24 NA 0.27 0.20 0.21 0.31 0.23MnO(%) 2 0.04 0.06 0.06 NA 0.08 0.15 0.05 0.04 0.05P2O5(%) 2 0.13 0.19 0.14 NA 0.25 0.27 0.33 0.28 0.15SrO(%) 2 0.03 0.04 0.05 NA 0.04 0.03 0.02 0.06 0.07BaO(%) 2 0.07 0.06 0.08 NA 0.08 0.07 0.07 0.09 0.13C(%) 4 9.39 10.65 6.23 NA 10.6 12.5 18.3 7.12 3.32S(%) 4 0.24 0.19 0.11 NA 0.31 0.38 0.27 0.15 0.09Ba(ppm) 5 641 533 727 479 716 584 612 774 1125Ce(ppm) 5 43.80 41.00 47.30 41.90 46.20 35.50 29.30 71.70 45.40Cr(ppm) 5 130 80 130 250 160 140 110 110 150Cs(ppm) 5 1.00 0.86 1.09 0.88 1.03 0.85 0.87 0.73 1.39Continued on next page2Numbers refer to analytical methods given in appendix F130Appendix A. Supplementary Data Chapter 1Table A.1 – continued from previous pageYear 2008 2008 2008 2009 2009 2009 2009 2009 2009Month July July July Apr Apr Apr Oct Oct OctBorehole TS1 TS2 TS3 TS4 TS5 TS6 TS7 TS8 TS9Core 02 04 05 03/04 06 01/02 01 03 05/04Section-cm 20-300-10 20-3020-25/0-55-15 20-25/0-520-300-10 20-25/0-5Dy(ppm) 5 1.57 1.52 1.87 1.75 1.69 1.49 1.35 2.06 1.78Er(ppm) 5 1.06 0.85 1.13 0.93 1.03 1.01 0.77 1.15 1.00Eu(ppm) 5 0.61 0.56 0.71 0.58 0.74 0.51 0.56 0.88 0.80Ga(ppm) 5 10.00 8.10 10.90 7.70 10.20 8.10 8.00 9.20 14.60Gd(ppm) 5 2.65 2.51 2.92 2.49 2.61 1.95 1.96 3.5 2.60Hf(ppm) 5 2.30 3.40 3.30 2.30 2.40 2.90 2.10 3.50 2.90Ho(ppm) 5 0.33 0.28 0.36 0.34 0.36 0.32 0.26 0.37 0.33La(ppm) 5 23.90 23.00 25.9 24.2 25.60 20.00 16.60 40.70 26.10Lu(ppm) 5 0.13 0.13 0.15 0.14 0.13 0.14 0.11 0.16 0.14Nb(ppm) 5 11.60 10.80 14.10 10.00 11.70 10.80 9 23.10 17.30Nd(ppm) 5 17.20 15.60 18.6 16.5 17.80 13.60 11.4 26.80 16.90Pr(ppm) 5 4.90 4.63 5.33 4.75 5.14 3.97 3.42 7.71 5.02Rb(ppm) 5 53.30 45.20 60.30 37.70 51.10 44.30 46.80 54.10 98.40Sm(ppm) 5 2.84 2.55 3.24 3.06 2.89 2.18 1.94 4.17 2.85Sn(ppm) 5 4.00 4.00 3.00 10.00 5.00 6.00 9.00 2.00 3.00Sr(ppm) 5 372 346 423 250 412 328 320 495 569Ta(ppm) 5 0.70 0.70 0.90 0.80 0.80 0.70 0.6 1.2 1.0Tb(ppm) 5 0.36 0.03 0.39 0.34 0.35 0.28 0.26 0.44 0.33Th(ppm) 5 6.10 7.38 6.89 4.58 7.85 5.53 4.16 9.60 7.88TI(ppm) 5 <0.5 <0.5 <0.5 <0.5 <0.5 <0.5 <0.5 <0.5 <0.5Tm(ppm) 5 0.17 0.16 0.18 0.26 0.22 0.21 0.19 0.18 0.17U(ppm) 5 1.90 1.65 1.62 1.84 1.74 1.91 1.66 1.83 1.8V(ppm) 5 36 22 31 28 41 24 28 37 33W(ppm) 5 1.00 1.00 1.00 56.00 1.00 2.00 1.00 2.00 5.00Y(ppm) 5 8.60 7.80 9.90 9.40 8.80 8.70 7.0 10.40 8.90Yb(ppm) 5 0.79 0.82 1.04 0.81 0.88 0.94 0.73 1.02 0.94Zr(ppm) 5 88.00 132 123 86 88 106 80 135 104As(ppm) 3 2.70 2.90 1.80 >250 16.6 18.9 45.8 9.3 18.9Bi(ppm) 3 0.11 0.12 0.09 0.27 0.20 0.19 0.21 0.07 0.11Hg(ppm) 3 0.079 0.099 0.044 0.287 0.134 0.156 0.297 0.047 0.035Sb(ppm) 3 0.70 1.40 0.63 3.11 2.02 2.08 0.85 0.38 1.28Continued on next page131Appendix A. Supplementary Data Chapter 1Table A.1 – continued from previous pageYear 2008 2008 2008 2009 2009 2009 2009 2009 2009Month July July July Apr Apr Apr Oct Oct OctBorehole TS1 TS2 TS3 TS4 TS5 TS6 TS7 TS8 TS9Core 02 04 05 03/04 06 01/02 01 03 05/04Section-cm 20-300-10 20-3020-25/0-55-15 20-25/0-520-300-10 20-25/0-5Se(ppm) 3 0.30 0.20 0.20 0.80 0.30 0.30 0.30 0.20 0.20Te(ppm) 3 0.02 0.02 0.01 0.03 0.03 0.02 0.02 0.01 0.01LoI(%) 25.10 29.50 18.60 NA 24.8 30.1 38.7 23.1 8.49Ag(ppm) 1 <0.5 <0.5 <0.5 1.1 <0.5 <0.5 <0.5 <0.5 <0.5Cd(ppm) 1 1.70 2.00 1.60 59.40 2.80 2.80 4.70 1.10 3.70Co(ppm) 1 2.00 3.00 1.00 7.00 5.00 3.00 2.00 4.00 3.00Cu(ppm) 1 32.00 15.00 11.00 79.00 22.00 19.00 21.00 12.00 12.00Mo(ppm) 1 1.00 1.00 1.00 14.00 1.00 1.00 2.00 1.00 <1Ni(ppm) 1 12.00 11.00 10.00 34.00 34.00 13.00 15.00 19.00 11.00Pb(ppm) 1 23.00 40.00 24.00 62.00 59.00 57.00 25.00 16.00 26.00Zn(ppm) 1 108 122 118 1735 180 177 705 127 380132Appendix A. Supplementary Data Chapter 1Table A.2: Carbon analysis of the BCR core solid samples with RockEval6 technology.Year 2008 2008 2008 2009 2009 2009 2009 2009 2009Month July July July Apr Apr Apr Oct Oct OctBorehole TS1 TS2 TS3 TS4 TS5 TS6 TS7 TS8 TS9Core 02 04 05 03/04 06 01/02 01 03 05/04Section-cm 20-300-10 20-3020-25/0-55-15 20-25/0-520-300-10 20-25/0-5PC(%) 2 1.73 1.21 2.33 3.18 2.9 5.59 1.14 0.68TOC 4.46 4.29 3.17 5.2 8.7 8 13.28 2.69 1.74RC(%) 2.46 2.56 1.96 2.87 5.52 5.1 7.69 1.55 1.06MINC(%) 2.75 5.35 3.22 3.82 3.21 3.6 1.79 4.88 1.09S1 6.89 6.29 3.69 8.51 11.57 10.14 16.96 2.98 1.77S2 14.11 12.26 9.13 15.97 23.12 20.82 43.71 8.72 5.44PI 0.33 0.34 0.29 0.35 0.33 0.33 0.28 0.25 0.25S3 4.57 4.01 3.02 5.32 6.63 7.07 10.79 3.21 1.89Tmax 324 323 314 320 316 316 338 327 321Tpeak 362 361 352 358 354 354 376 365 359S3CO 2.31 1.24 1.01 2.57 1.92 1.86 4.72 1.32 0.42HI 316 286 288 307 266 260 329 324 313OICO 52 29 32 49 22 23 36 49 24OI 102 93 95 102 76 88 81 119 109S4CO 3.91 5.05 5.55 6.12 17.28 13.95 18.09 2.9 2.76S4CO2 84.06 85.84 63.13 95.63 175.35 165.02 253.36 52.37 34.5RCCO(%) 0.17 0.22 0.24 0.26 0.74 0.6 0.78 0.12 0.12RCCO2(%) 2.29 2.34 1.72 2.61 4.78 4.5 6.91 1.43 0.94133Appendix A. Supplementary Data Chapter 1Table A.3: Carbon analysis of the BCR core solid samples with RockEval6 technology. Complete analysis of the all recovered samplesYear 2008 2008 2008 2008 2008 2008Month July July July July July JulyBorehole TS1 TS1 TS1 TS1 TS1 TS1Core 02 02 01 01 03 03Section 00-10 10-20 00-10 10-20 00-10 10-20PC(%) 1.38 1.57 0.39 2.01 2.55 1.54TOC 3.09 3.52 0.99 5.39 6.16 3.86RC(%) 1.71 1.95 0.6 3.38 3.61 2.32MINC(%) 4.18 3.26 5.52 1.68 6.33 5.81S1 5.14 5.47 1.03 6.22 9.29 5.76S2 9.44 10.92 2.65 14.74 17.78 10.65PI 0.35 0.33 0.28 0.3 0.34 0.35S3 3.18 3.53 1.67 5.05 5.56 3.62Tmax 316 324 330 327 321 318Tpeak 354 362 368 365 359 356S3CO 1.35 1.65 0.43 2.02 2.35 1.07HI 306 310 268 273 289 276OICO 44 47 43 37 38 28OI 103 100 169 94 90 94S4CO 3.59 3.59 2.66 10.18 8.08 4.38S4CO2 57.21 65.93 18.1 107.81 119.59 78.05RCCO(%) 0.15 0.15 0.11 0.44 0.35 0.19RCCO2(%) 1.56 1.8 0.49 2.94 3.26 2.13Year 2008 2008 2008 2008 2008 2008Month July July July July July JulyBorehole TS2 TS2 TS3 TS3 TS3 TS3Core 04 04 05 05 06 05Section 10-20 20-30 00-10 10-20 5-10 10-20PC(%) 1.21 2.34 2.7 2.26 1.62 1.5TOC 3.06 6.01 6.78 5.78 4.32 3.9RC(%) 1.71 1.95 0.6 3.38 3.61 2.32MINC(%) 6.54 3.4 3.12 1.59 4.64 5.68S1 4.23 8.04 9.13 7.52 5.22 5S2 8.77 16.95 19.95 16.96 11.89 11.1PI 0.33 0.32 0.31 0.31 0.31 0.31Continued on next page134Appendix A. Supplementary Data Chapter 1Table A.3 – continued from previous pageYear 2008 2008 2008 2008 2008 2008Month July July July July July JulyBorehole TS2 TS2 TS3 TS3 TS3 TS3Core 04 04 05 05 06 05Section 10-20 20-30 00-10 10-20 5-10 10-20S3 2.93 5.33 5.85 5.09 4.12 3.44Tmax 313 324 315 314 313 316Tpeak 351 362 353 352 351 354S3CO 0.67 1.98 2.03 1.38 1.07 0.81HI 287 282 294 293 275 285OICO 22 33 30 24 25 21OI 96 89 86 88 95 88S4CO 5.29 9.78 7.37 10.1 3.84 6.21S4CO2 59.48 119.24 137.77 113.13 93 77.95RCCO(%) 0.23 0.42 0.32 0.43 0.16 0.27RCCO2(%) 1.62 3.25 3.76 3.09 2.54 2.13Year 2009 2009 2009 2009 2009 2009Month Apr Apr Apr Apr Apr AprBorehole TS6 TS6 TS6 TS5 TS5 TS5Core 01 01 02 05 05-06 07-06Section-cm 00-10 10-20 5-10 00-10 10-15/0-515-20/0-5PC(%) 0.66 1.5 1.26 1.64 1.28 2.45TOC 1.91 4.44 3.28 3.89 3.15 6.48RC(%) 1.25 2.94 2.02 2.25 1.87 4.03MINC(%) 2.18 5.86 4.88 2.14 4.48 3.39S1 1.71 4.81 4.12 5.74 4.5 8.82S2 4.88 10.8 9.09 11.96 9.06 17.7PI 0.26 0.31 0.31 0.32 0.33 0.33S3 2.19 4.11 3.27 3.57 2.91 5.3Tmax 314 317 312 319 331 329Tpeak 352 355 350 357 369 367S3CO 0.65 0.94 0.86 1.06 1.28 1.63HI 255 243 277 307 288 273OICO 34 21 26 27 41 25OI 115 93 100 92 92 82S4CO 2.78 7.76 5.75 7.01 5.26 11.36Continued on next page135Appendix A. Supplementary Data Chapter 1Table A.3 – continued from previous pageYear 2009 2009 2009 2009 2009 2009Month Apr Apr Apr Apr Apr AprBorehole TS6 TS6 TS6 TS5 TS5 TS5Core 01 01 02 05 05-06 07-06Section-cm 00-10 10-20 5-10 00-10 10-15/0-515-20/0-5S4CO2 41.49 95.67 64.75 71.65 60.08 129.72RCCO(%) 0.12 0.33 0.25 0.3 0.23 0.49RCCO2(%) 1.13 2.61 1.77 1.95 1.64 3.54Year 2009 2009 2009 2009 2009 2009Month Oct Oct Oct Oct Oct OctBorehole TS7 TS7 TS8 TS8 TS8 TS8Core 01 01 02 02 03 03section-cm 00-10 10-20 00-10 10-20 10-20 20-30PC(%) 0.53 0.62 1.78 2.97 2.95 1.9TOC 1.4 1.62 4.12 6.83 6.65 4.67RC(%) 0.87 1 2.34 3.86 3.7 2.77MINC(%) 9.2 7.28 2.67 3.19 3.19 4.14S1 1.46 1.71 5.25 10.11 10.26 5.46S2 4.04 4.65 13.04 21 20.86 15PI 0.27 0.27 0.29 0.32 0.33 0.27S3 1.41 1.66 4.47 6.51 6.55 4.48Tmax 314 315 319 321 319 319Tpeak 352 353 357 359 357 357S3CO 0.36 0.72 2.38 3.8 3.34 1.13HI 289 287 317 307 314 321OICO 26 44 58 56 50 24OI 9.2 7.28 2.67 3.19 3.19 4.14S4CO 2.58 2.76 4.91 9.24 7.18 6.82S4CO2 27.77 32.09 78.28 127.02 124.35 90.95RCCO(%) 0.11 0.12 0.21 0.4 0.31 0.29RCCO2(%) 0.76 0.88 2.13 3.46 3.39 2.48136Appendix A. Supplementary Data Chapter 1Table A.4: Parameters definition of carbon analysis with RockEval 6 tech-nology, calculated and acquisition parameters adapted from [10] Table 2aand 2b.Calculatedparame-tersUnit Formula NameS1 mg HC/g rock Free hydrocarbonsS2 mg HC/g rock Oil potentialS3 mg CO2/g rock CO2 organic sourceS3CO mg CO/g rock CO organic sourceS4CO2 mg CO2/g rock CO2 organic sourceS5 mg CO2/g rock CO2 mineral sourceS4CO mg CO/g rock CO organic sourceTpS2 ◦C (T) of peak S2 maxTmax ◦C TpS2-δTmax TmaxPI S1/(S1+S2) Production indexPC wt% 3 Pyrolysable org4. carbonRC CO wt% (S4CO*12/28)/10 Residual org. carbon(CO)RCCO2 wt% (S4CO2*12/44)/10 Residual org.carbon(CO2)RC wt% RC CO+RC CO2 Residual org. carbonTOC wt% PC+RC Total org carbonHI mg HC/g TOC (S2*100/TOC ) Hydrogen indexOI mgCO2/g TOC (S3*100/TOC) Oxygen indexOI CO mg CO/g TOC S3CO*100/TOC Oxygen index COPyroMinC wt% 5 Pyrolysis mineral carbonOxiMinC wt% (S5*12/14)/10 Oxidation mineral carbonMinC wt% PyroMinC+OxiMinC Mineral carbon3[(S1+S2)*0.83]+[S3*12/44]+[(S3CO2+S3′CO/2)*12/28]/104org:organics5(S3′*12/44)+(S3′/2)*12/28)137Appendix A. Supplementary Data Chapter 1Glucose 2 Pyruvate Krebs Cycle 2 Acetyl CoA e- O2 Fe (III) NO3 - SO4 2- As(V),etc e- Cytochrome system NADH FADH e- e- e- e- H2O Fe (II) N2, S 2- As(III) etc Figure A.1: Schematic of Dissimilatory Reduction process, adapted from[133],[23]138Appendix A. Supplementary Data Chapter 1A : Herminiimonas asenicoxydansB: Alcaligenes faecalisC: Agrobacterium tumefaciensD: Rhodoferax ferrireducensE: Burkholderia multivoransF: Xanthobacter autotrophicusG: Roseovarius sp217,H: Nitrobacter hamburgensisI: Chlorobium phaerobacteroidesJ: Chloroexus aurentiacusK: Thermus thermophilus HB8L: Aeropyrum pernixM: Sulfolobus tokodaiN: Environmental sample 1O: Environmental sample 2ABCDEFGHIJKLMNOFigure A.2: Organization of the arsenite oxidase (aox ) gene cluster invarious arsenic-metabolizing microorganisms [106]139Appendix BSupplementary DataChapter 2140Appendix B. Supplementary Data Chapter 2Table B.1: Total arsenic, iron and sulfur in the BCR influent and effluentfrom 25 June 2008 to 2 October 2009Total metal concentrations (mM)Date BCR influent BCR effluentAs Fe S As Fe S25 June 2008 0.534 0.041 8.33 0.095 0.287 9.383 July 2008 0.108 0.008 6.25 0.160 0.502 6.258 July 2008 0.071 0.233 6.25 1.028 0.323 6.2516 July 2008 0.601 0.013 6.25 0.134 0.109 9.3829 July 2008 0.347 0.008 8.33 0.028 0.066 9.3813 August 2008 0.561 0.007 8.65 0.011 0.039 5.2127 August 2008 0.174 0.102 7.29 0.016 0.065 9.3810 September 2008 0.059 0.002 6.25 0.035 0.073 6.2524 September 2008 0.160 0.004 6.25 0.587 1.971 6.258 October 2008 0.427 0.038 5.21 0.115 0.394 6.2522 October 2008 0.547 0.109 5.21 0.019 0.030 6.255 November 2008 0.174 0.007 8.33 0.160 0.095 9.3818 November 2008 1.135 0.573 7.29 0.012 0.057 10.422 December 2008 0.734 0.013 7.29 0.059 0.059 7.2917 December 2008 0.267 0.052 6.25 0.079 0.090 7.2915 January 2009 0.227 0.004 5.21 0.023 0.017 5.2129 January 2009 0.174 0.008 4.17 0.019 0.010 4.1711 February 2009 0.280 0.013 7.29 0.053 0.015 6.2525 February 2009 0.214 0.018 6.67 0.064 0.116 7.0810 March 2009 0.174 0.006 6.67 0.043 0.018 6.4624 March 2009 0.320 0.358 6.56 0.507 2.509 7.926 April 2009 0.174 0.045 7.19 0.021 0.014 7.0824 April 2009 2.537 0.048 6.67 0.016 0.006 8.445 May 2009 0.587 0.032 6.88 0.019 0.005 7.8122 May 2009 0.467 0.079 6.67 0.025 0.020 7.085 June 2009 0.454 0.007 7.08 0.096 0.043 7.8119 June 2009 0.387 0.006 6.98 0.020 0.006 7.5030 June 2009 0.307 0.016 7.60 0.043 0.034 8.854 September 2009 0.614 0.065 8.54 0.017 0.011 9.172 October 2009 0.467 0.018 nd 0.075 0.039 ndMaximum 2.537 0.573 8.646 1.028 2.509 10.417Minimum 0.059 0.002 4.167 0.011 0.005 4.167Continued on next page141Appendix B. Supplementary Data Chapter 2Table B.1 – continued from previous pageTotal metal concentrations (mM)BCR influent BCR effluentAs Fe S As Fe SAverage 0.443 0.064 6.814 0.119 0.234 7.428Std. dev. 0.46 0.12 1.07 0.22 0.56 1.54Table B.2: Total arsenic, iron and sulfur in the BCR lysimeterDate T/As (mM) T/Fe (mM) T/S (mM)10 September 2008 0.008 0.048 3.1308 October 2008 0.011 0.043 3.1312 November 2008 0.010 0.057 3.8510 December 2008 0.009 0.086 4.3807 January 2009 0.008 0.077 4.1713 February 2009 0.007 0.070 4.5812 March 2009 0.006 0.057 4.5808 April 2009 0.012 0.143 5.0009 May 2009 0.010 0.104 5.0012 June 2009 0.007 0.072 4.7907 July 2009 0.005 0.034 4.7910 August 2009 0.006 0.023 4.69Maximum 0.012 0.143 5.000Minimum 0.005 0.023 3.125Average 0.008 0.068 4.340Std. dev. 0.002 0.03 0.66142Appendix B. Supplementary Data Chapter 2Table B.3: Total arsenic, iron and sulfur in the borehole porewaterSample origin Date T/As (mM) T/Fe (mM) T/S (mM)TS1 8 July 2008 0.003 13.2 2.08TS2 8 July 2008 0.001 16.36 0.83TS3 8 July 2008 0.003 2.51 1.56TS4 21 April 2009 0.013 0.14 4.90TS5 21 April 2009 0.005 0.01 6.35TS7 20 October 2009 0.112 4.48 6.25TS8 20 October 2009 0.013 4.30 2.19TS9 20 October 2009 0.036 0.73 2.81Maximum 0.112 4.48 6.35Minimum 0.001 0.01 0.83Average 0.023 2.03 3.37Std. dev. 0.038 2.04 2.166T/Fe of TS1 and TS2 (13.2 and 16.3 mM) may be artificially high due to the closeproximity of iron pipes143Appendix B. Supplementary Data Chapter 2Table B.4: Aqueous and solid species equilibrium reactions and constantsused for geochemical modeling of the As-O-H-S-Fe system with Spana. Thisis the input file. First column gives the chemical species, second column isthe log(K) value and the final five columns give the stoichiometry in termsof the species H+, e−, HS−, As(OH)3 and Fe2+. Equilibrium constants forAs species modified according to Tables 1 and 2 in [87].Chemical species log(k) H+ e− HS− As(OH)3 Fe2+H2AsO3− 9.17 1 0 0 1 0HAsO32− 23.27 2 0 0 1 0AsO33− 38.27 3 0 0 1 0H4AsO3+ 0.305 1 0 0 1 0As3S4(HS)2− 72.23 5 0 6 3 0AsS(OH)(HS)− 18.008 1 0 2 1 0H2AsO4− 21.65 3 2 0 1 0HAsO42− 28.64 4 2 0 1 0AsO3− 40.43 5 2 0 1 0H3AsO4 19.35 2 2 0 1 0Fe3+ 13.02 0 1 0 0 1Fe(HS)2 8.95 0 0 2 0 1Fe(HS)3− 10.987 0 0 3 0 1Fe(OH)2 20.8 2 0 0 0 1Fe(OH)3− 33.4 3 0 0 0 1Fe(OH)42− 46.35 4 0 0 0 1FeOH+ 10.2 1 0 0 0 1H2 3.15 2 2 0 0 0H2(g) 0 2 2 0 0 0H2O2 59.601 2 2 0 0 0H2S 6.994 1 0 1 0 0H2S (g) 7.991 1 0 1 0 0H2S4 7.879 2 6 4 0 0H2S5 8.994 3 8 5 0 0HS4− 3.678 3 6 4 0 0HS5− 5.094 4 8 5 0 0O2 86.08 4 4 0 0 0O2(g) 83.12 4 4 0 0 0O3 156.05 6 6 0 0 0O3 (g) 153.25 6 6 0 0 0Continued on next page144Appendix B. Supplementary Data Chapter 2Table B.4 – continued from previous pageChemical species log(k) H+ e− HS− As(OH)3 Fe2+OH− 14 1 0 0 0 0S 2− 19 1 0 1 0 0S 2− 5.903 2 3 2 0 0S22− 12.734 2 2 2 0 0S2O32− 28.793 8 8 2 0 0S32− 7.065 3 4 3 0 0S42− 3.022 4 6 4 0 0S52− 1.007 5 8 5 0 0S62− 0.844 6 10 6 0 0SO32− 37.089 7 6 1 0 0SO42− 33.692 9 8 1 0 0Fe(OH)2+ 18.69 2 1 0 0 1Fe(OH)3 25.58 3 1 0 0 1Fe(OH)4− 34.62 4 1 0 0 1Fe(SO4)2− 75.024 18 17 2 0 1Fe2(OH)24+ 28.99 2 2 0 0 2Fe3(OH)5+4 45.36 4 3 0 0 3FeHSO2+4 42.244 8 9 1 0 1FeHSO+4 30.624 8 8 1 0 1FeO42− 125.87 8 4 0 0 1FeOH2− 15.21 1 1 0 0 1FeS2O+3 39.833 8 9 2 0 1FeSO4 31.442 9 8 1 0 1FeSO+4 42.672 9 9 1 0 1H2S2O3 26.521 6 8 2 0 0H2SO4 33.692 7 8 1 0 0HO2− 71.251 3 2 0 0 0HS2O3− 27.108 7 8 2 0 0HSO3− 29.869 6 6 1 0 0HSO4− 31.712 8 8 1 0 0S2O62− 75.428 14 14 2 0 0S2O82− 133.645 18 18 2 0 0S4O62− 54.881 16 18 4 0 0SO2 (aq) 28.029 5 6 1 0 0SO2 (g) 28.349 5 6 1 0 0FeAsS (c) 25.629 2 3 1 1 1Continued on next page145Appendix B. Supplementary Data Chapter 2Table B.4 – continued from previous pageChemical species log(k) H+ e− HS− As(OH)3 Fe2+As (c) 12.5253 3 3 0 1 0As2S3 (c) 46.3 3 0 3 2 0AsS (c) 20.11 2 1 1 1 0Fe (c) 16.097 0 2 0 0 1Fe(OH)2 (c) 12.996 2 0 0 0 1Fe3O4 (c) 37.077 8 2 0 0 3FeS (am) 3.915 1 0 1 0 1FeS (c) 4.648 1 0 1 0 1FeS2 (c) 18.479 2 2 2 0 1S (c) 2.145 1 2 1 0 0As2O3 (s) 1.432 0 0 0 2 0As2O5 (c) 46.96 4 4 0 2 0Fe(OH)3 (am) 17.911 3 1 0 0 1Fe2(SO4)3 (c) 130.696 27 26 3 0 2Fe2O3 (cr) 26.448 6 2 0 0 2Fe3(OH)8 (c) 46.262 8 2 0 0 3Fe3S4 (c) 18.995 4 2 4 0 3FeAsO4 : 2H2O (c) 27.35 5 3 0 1 1FeOOH (cr) 14.02 3 1 0 0 1FeSO4 : 7H2O (c) 26.08 9 8 1 0 1H3OFe3(SO4)2(OH)6(c)101.054 23 19 2 0 3FeAsO4 39.32 5 3 0 1 1FeHAsO4 31.76 4 3 0 1 1FeH2AsO42+ 30.47 3 3 0 1 1FeH2AsO4+ 18.94 3 2 0 1 1FeHAsO4 25.34 4 2 0 1 1FeAsO−4 32.88 5 2 0 1 1FeH2AsO2+3 14.98 1 1 0 1 1146Appendix B. Supplementary Data Chapter 2Table B.5: Results of geochemical modeling of the As-O-H-S-Fe systemusing the equilibrium data in Table B.4 over the concentration ranges mea-sured in the BCR (Tables B.1, B.2, B.3)Concentrations (M) Predominant Arsenic solid speciesAs(OH)3 Fe2+ HS− (pH 5.6-7.5, ORP -100 to -300 mV)0.000011 0.002509 0.007 FeAsS0.000011 0.000064 0.007 FeAsS0.000011 0.000234 0.007 FeAsS0.000011 0.000002 0.007 AsS As0.002537 0.002509 0.007 FeAsS0.002537 0.000064 0.007 AsS As0.002537 0.000234 0.007 AsS As0.002537 0.000002 0.007 AsS As0.000443 0.000002 0.0070.000119 0.000002 0.0070.000443 0.002509 0.007 FeAsS0.000119 0.002509 0.007 FeAsS0.000443 0.000234 0.007 FeAsS0.000119 0.000234 0.007 FeAsS0.000443 0.000064 0.007 AsS As147Appendix B. Supplementary Data Chapter 2Table B.6: Predominance of high metal concentration associated microbialgroups in the BCR core samples. Concentrations are in ppm.Taxonomic classification Number of sequences (Reads)ts1 ts2 ts3 ts4 ts5 ts6 ts7 ts8 ts9 TotalMethanocorpusculum 0 1 2 32 2 1 11 3 6 58Methanospirilium 0 3 36 1913 255 1042 468 358 663 4738Methanosarcina 0 1 3 162 9 18 539 1 209 942Bacteroidetes SB-1 0 0 3 3 0 0 2 0 2 10M2PB4-65 7 5 16 166 263 504 86 129 29 37 1235VadinHA178 0 0 6 90 8 30 25 4 8 171TM6 0 0 0 9 0 0 0 2 8 19RF3 7 2 16 78 3 4 5 11 12 138WS6 0 5 6 165 4 722 480 2 70 1454Victivallaceae 0 0 1 6 1 16 13 2 8 47Synergistales 0 0 5 57 2 2 1 1 1 69reads9 4000 4000 4000 4000 4000 4000 4000 4000 4000Arsenic 2.7 2.9 1.8 ≥ 250 16.6 18.9 45.8 9.3 18.9Zinc 108 122 118 1735 180 177 705 127 3807M2PB4-65 termite group8VadinHA17 anaerobic digester group9Total number of reads in each sample148Appendix B. Supplementary Data Chapter 2 - Mixture of 60% kraft pulp mill biosolids residuals, 35% sand and 5% manureupflowseepage inflow BCR outflowgeotextile linersandsoilvegetationsurrounding natural subsurfacesoil3-7 m deep18m X 30mFigure B.1: Schematic and photograph of the BCR. Diagram adopted from[37]. Photograph taken by Maryam Khoshnoodi on 21 April 2009.149Appendix B. Supplementary Data Chapter 2Eh(V)1.00.50.0-0.5-1.02 4 6 8 10 12pHHAsO42-As(OH)3ScoroditeArsenopyriteRealgarOrpiment H2AsO4-H3AsO4H2AsO3-H2AsO4-1.00.50.0-0.5-1.02 4 6 8 10 12pHEh(v)As(c)HAsO42-ScoroditeRealgarOrpiment As(OH)3H2AsO4-H3AsO4H2AsO3-As(OH)3H2AsO4-1.00.50.0-0.5-0.12 4 6 8 10 12Eh(v )pHArsenopyriteOrpimentRealgarScoroditeH3AsO4H2AsO4- HAsO42-H2AsO3-1.00.50.0-0.5-1.02 4 6 8 10 12ScoroditeAs(OH)3HAsO42-H3AsO4H2AsO4-H2AsO3-RealgarOrpimentEh(v)pH As(c)A) B)C) D)Figure B.2: Eh-pH diagrams for As-Fe-S-O-H at 250 ◦C and 1 atm pressure.Predominant arsenic species predicted for the chemistry of the BCR porewater at total concentration of arsenic 0.44 mM; sulfur 7.00 mM and totaliron (A) 0.002 mM, (B) 0.064 mM, (C) 1.29 mM, and (D) 2.51 mM. TheEh and pH ranges of the BCR are within the dashed rectangle150AppendixB.SupplementaryDataChapter2Table B.7: SRB-related and iron-reducing bacteria enrichment culture related OTUs found in the BCR. The tablecolumns are OTU No, Number of sequences, Taxonomic classification according to Blastn to Silva111 database,Most closely related cultured species and Percent similarity respectfully.OTU Reads Taxa Species Similarity2066 44 Bacteria; Proteobacteria; Deltaproteobacteria; Desul-fobacterales; Desulfobulbaceae; Desulfobulbus; uncul-tured bacteriumDesulfobulbus elon-gatus strain FP996440 10 Bacteria; Proteobacteria; Deltaproteobacteria; Desul-fobacterales; Nitrospinaceae; uncultured; unculturedbacteriumDesulfovibrio mexi-canus strain Lup1841605 31 Bacteria;Proteobacteria; Deltaproteobacteria; Desul-fovibrionales; Desulfovibrionaceae; Desulfovibrio;Desulfovibrio idahonensisDesulfovibrio mexi-canus strain Lup1964663 10 Bacteria;Proteobacteria; Deltaproteobacteria; Desul-fovibrionales; Desulfovibrionaceae; Desulfovibrio;Desulfovibrio idahonensisDesulfovibrio mexi-canus strain Lup197259 2 Bacteria; Proteobacteria; Deltaproteobacteria; Desul-fovibrionales; Desulfovibrionaceae; Desulfovibrio;Desulfovibrio paquesiiDesulfovibrio paque-sii strain SB1983241 8 Bacteria; Proteobacteria; Deltaproteobacteria; Desul-fovibrionales; Desulfovibrionaceae; Desulfovibrio;Desulfovibrio sp. A-1Desulfovibrioaminophilus strainALA-397728 9 Bacteria; Proteobacteria; Deltaproteobacteria; GR-WP33-30; uncultured bacteriumGeobacter daltoniiFRC-32904896 724 Bacteria;Proteobacteria; Deltaproteobacteria;Sh765B-TzT-29; uncultured bacteriumHalothiobacillusneapolitanus c285151AppendixB.SupplementaryDataChapter2Table B.7 – continued from previous pageOTU Reads Taxa Species Similarity6722 217 Bacteria; Proteobacteria; Deltaproteobacteria;Sh765B-TzT-29; uncultured delta proteobacteriumHalothiobacillusneapolitanus c285850 10 Bacteria; Bacteroidetes; Bacteroidia; Bacteroidales;Bacteroidaceae; Bacteroides; iron-reducing enrich-ment clone Cl-A12Bacteroides gramini-solvens strain XDT-1934336 18 Bacteria; Bacteroidetes; Bacteroidia; Bacteroidales;Bacteroidaceae; Bacteroides; iron-reducing enrich-ment clone Cl-A12Bacteroides gramini-solvens strain XDT-1975861 113 Bacteria; Bacteroidetes; Bacteroidia; Bacteroidales;Porphyromonadaceae; Petrimonas; iron-reducing bac-terium enrichment culture clone HN-HFO84Proteiniphilum ac-etatigenes strainTB1079411776 10 Bacteria; Bacteroidetes; Bacteroidia; Bacteroidales;Porphyromonadaceae; Petrimonas; iron-reducing bac-terium enrichment culture clone HN-HFO84Proteiniphilum ac-etatigenes strainTB107954452 53 Bacteria; Bacteroidetes; Bacteroidia; Bacteroidales;Rikenellaceae; vadinBC27 wastewater-sludgegroup;iron-reducing bacterium enrichment cultureclone HN7Marinifilum fragileCECT 7942 StrainJC2469876640 92 Bacteria; Bacteroidetes; Bacteroidia; Bacteroidales;Rikenellaceae; vadinBC27 wastewater-sludgegroup;iron-reducing bacterium Enrichment Cul-ture Clone HN7Marinifilum fragileCECT 7942 StrainJC246987152AppendixB.SupplementaryDataChapter2Table B.7 – continued from previous pageOTU Reads Taxa Species Similarity2065 274 Bacteria; Firmicutes; Clostridia; Clostridiales; Gra-cilibacteraceae; Gracilibacter; iron-reducing bac-terium enrichment culture clone FEA2F1GracilibacterthermotoleransJW/YJL-S1971197 63 Bacteria; Firmicutes; Clostridia; Clostridiales; Gra-cilibacteraceae; Gracilibacter; iron-reducing bac-terium enrichment culture clone FEA2F1GracilibacterthermotoleransJW/YJL-S1971550 70 Bacteria; Firmicutes; Clostridia; Clostridiales; Gra-cilibacteraceae; Gracilibacter; iron-reducing bac-terium enrichment culture clone FEA2F1GracilibacterthermotoleransJW/YJL-S1974012 10 Bacteria; Firmicutes; Clostridia; Clostridiales; Gra-cilibacteraceae; Gracilibacter; iron-reducing bac-terium enrichment culture clone FEA2F1GracilibacterthermotoleransJW/YJL-S1972714 12 Bacteria; Firmicutes; Clostridia; Clostridiales; Lach-nospiraceae; Caldicoprobacter; iron-reducing bac-terium enrichment culture clone FEA2A2Tindallia texco-conensis strainIMP-300901398 13 Bacteria; Firmicutes; Clostridia; Clostridiales; Lach-nospiraceae; uncultured; iron-reducing bacterium en-richment culture clone HN109Bacillus thuringien-sis Bt407884392 140 Bacteria; Firmicutes; Clostridia; Clostridiales; Ru-minococcaceae; Incertae Sedis; iron-reducing bac-terium enrichment culture clone FEA2H3Clostridium suf-flavum strain CDT-1987291 53 Bacteria; Firmicutes; Clostridia; Clostridiales; Ru-minococcaceae; Incertae Sedis; iron-reducing bac-terium enrichment culture clone FEA2H3Clostridium suf-flavum strain CDT-198153AppendixB.SupplementaryDataChapter2Table B.7 – continued from previous pageOTU Reads Taxa Species Similarity980 28 Bacteria; Firmicutes; Clostridia; Clostridiales; Ru-minococcaceae; uncultured; iron-reducing bacteriumenrichment culture clone HN3Oscillibacter valerici-genes Sjm18-2089Several 31 Bacteria; Firmicutes; Clostridia; Clostridiales; Pepto-coccaceae; Desulfosporosinus; uncultured bacteriumSeveral 13 Bacteria; Firmicutes; Clostridia; Clostridiales; Pepto-coccaceae; Desulfotomaculum; uncultured Bacillus sp.8578 178 Bacteria; Synergistetes; Synergistia; Synergistales;Synergistaceae; uncultured; uncultured bacteriumThermanaerovibrioacidaminovoransDSM 6589908930 123 Bacteria; Synergistetes; Synergistia;Synergistales;Synergistaceae; uncultured; uncultured bacteriumThermanaerovibrioacidaminovoransDSM 65899012548 68 Bacteria; Synergistetes; Synergistia; Synergistales;Synergistaceae; uncultured; uncultured bacteriumThermanaerovibrioacidaminovoransDSM 6589923668 10 Bacteria; Synergistetes; Synergistia; Synergistales;Synergistaceae; uncultured; uncultured bacteriumAminobacteriumcolombiense DSM12261914310 32 Bacteria; Synergistetes; Synergistia; Synergistales;Synergistaceae; uncultured;uncultured bacteriumThermovirga lieniiDSM 17291946986 46 Bacteria; Synergistetes; Synergistia; Synergistales;Synergistaceae; uncultured; uncultured bacteriumThermanaerovibrioacidaminovoransDSM 658989154AppendixB.SupplementaryDataChapter2Table B.7 – continued from previous pageOTU Reads Taxa Species SimilarityTotal10 1360010These data were obtained from sequencing of a total of 48 samples from the BCR, including the nine that were used for metal andmineralogical analysis155Appendix CSupplementary DataChapter 3156Appendix C. Supplementary Data Chapter 3Table C.1: National Center for Biotechnology Information (NCBI) SequenceRead Archive (SRA) accession numbers for arsenic enrichment culture se-quenceProject Accession number: SRP038769NCBI SRABioSample Ac-cession NumberSample Name Experiment Acces-sion NumberRun AccessionNumberAs EnrichmentsSRS584982 AsIII-1P2 SRX505694 SRR1211190SRS584983 AsIII-1P3 SRX505695 SRR1211191SRS584984 AsIII-1P7 SRX505696 SRR1211192SRS584985 AsIII-2P3 SRX505697 SRR1211194SRS586271 AsIII-3P3 SRX507312 SRR1213084SRS584987 AsIII-4P3 SRX505699 SRR1211235SRS584999 AsIII-4P8 SRX505711 SRR1211236SRS585001 AsIII-5P3 SRX505714 SRR1211239SRS585005 AsIII-5P7 SRX505719 SRR1211242SRS585007 AsIII-6P3 SRX505721 SRR1211245SRS585008 AsV-1P2 SRX505724 SRR1211249SRS585012 AsV-1P3 SRX505727 SRR1211250SRS585016 AsV-2P3 SRX505731 SRR1211252SRS585018 AsV-3P3 SRX505734 SRR1211255SRS585023 AsV-3P8 SRX505738 SRR1211257SRS585028 AsV-4P3 SRX505744 SRR1211260SRS585030 AsV-4P8 SRX505746 SRR1211263SRS585031 AsV-5P3 SRX505748 SRR1211267SRS585033 AsV-6P3 SRX505751 SRR1211270157AppendixC.SupplementaryDataChapter3Table C.2: Pore water and solid chemistry from the bore holes used to take samples for the enrichment culturesand cores for in situ microbial phylogenetic analysisDate of coresamplingJuly8, 2008 April 21,2009 October 20,2009core: 02 04 06 03/04 06 01/02 01 03 04/05Bore holenumber:TS1 TS2 TS3 TS4 TS5 TS6 TS7 TS8 TS9PorewaterparametersaDO(mg/L) 0.0 0.0 0.0 1.5 1.1 0.7 1.2 1.3 1.3pH 5.6 5.6 6.9 6.9 6.9 7.0 6.7 6.2 7.5ORP b -132.5 -112.1 -238.1 NA NA NA -117.5 -244.2 -130.3Temperature(◦C) 17.6 15.9 19.4 7.6 6.5 7.1 11.2 11.5 11.2PorewaterchemistrycAs 0.19 0.06 0.03 0.94 0.94 0.94 0.14 0.21 0.51Cd 0.06 0.02 ≤ 0.01 ≤ 0.01 ≤ 0.01 ≤ 0.01 ≤ 0.01 ≤ 0.01 ≤ 0.06Mn 78.00 110.00 3.90 7.60 7.60 7.60 1.50 22.00 5.70Fe 740 910 50.00 7.60 7.60 7.60 1.00 4.25 0.30Sb ≤ 0.01 ≤ 0.01 ≤ 0.01 0.01 0.01 0.01 0.00 0.00 0.00Zn 6.10 1.30 0.52 0.06 0.06 0.06 0.05 0.09 0.28Nutrientsd Sulfide 1.1 2.6 3.5 ≤ 0.1 ≤ 0.1 ≤ 0.1 ≤ 0.1 ≤ 0.1 ≤ 0.1Total phosphorous 21.0 20.0 30.0 0.5 0.5 0.5 24.0 14.0 6.3Sulfate 200 80 150 470 470 470 600 150 270Nitrite/Nitrate NA/5.2 NA NA 0.015/1.30.015/1.30.015/1.30.023/5.50.109/5.50.01/1.5Ammonium-N 460.0 340.0 490.0 61.0 61.0 61.0 3.0 3.8 1.8Core solidschemistryeTotal C (%) 9.39 10.65 6.23 NSS 10.60 18.30 18.30 7.12 3.32Total S (%) 0.24 0.19 0.11 NSS 0.31 0.27 0.27 0.15 0.09As (ppm) 2.7 2.9 1.8 ≥ 250 16.6 18.9 45.8 9.3 18.9Cd (ppm) 1.70 2.00 1.60 59.40 2.80 2.80 4.70 1.10 3.70Zn (ppm) 108 122 19 1735 180 177 705 127 380LoI f 25.0 29.5 118.0 NSS 24.8 30.1 38.7 23.1 8.5ameasured in the field using the YSI probebOxidation/reduction potential(mV)cmeasured in industry labdmeasured in the field(mg/L)emeasured by ASL commercial laboratoryfLoss on ignition(%)(1000 ◦C)158Appendix C. Supplementary Data Chapter 35 10 15 20 2530000060000090000012000001500000As(III) Concentration (mg/L)Area under the 0.98s peak5 10 15 20 25 30 35050000100000150000200000250000As(V) Concentration (mg/L)Area under the 6.62s peak y = 7780.4x - 24618R² = 0.99119y = 66605x - 9883.6R² = 0.9831Figure C.1: As calibration curves and example chromatograms.159Appendix C. Supplementary Data Chapter 3  As(V) redAs(III) blue2 red3 blue 7 orange8 green(a) (b)(c) (d)(e) (f )Figure C.2: Rarefaction statistics (PD whole tree and Chao1) for diversityof samples: (a) and (b) for each sample; (c) and (d) according to arsenicspecies medium; (e) and (f ) according to passage number.s.160Appendix C. Supplementary Data Chapter 351%7%6%4%3%3%2%2%2%2%2% 2%2% 1%1%1%1%1%1%1% 0%0%0%0%5%As(III) mediumSimplicispira(51.4%)Methanocorpusculum(6.83%)Rhizobium(6.34%)Ottowia(3.76%)Acidovorax(3.18%)Ramlibacter (2.79%)Pseudomonas(2.39%)Methanoregula(2.03%)Lutibacter(1.96%)Acidovorax delafieldii 2AN(1.78%)Uncutured, KD2- 123(1.74%)Thermomonas(1.71%)Proteiniphilum(1.68%)Aquabacterium(1.49%)Chryseobacterium(0.92%)Enterobacter(0.85%)Sedimentibacter(0.64%)Methanosaeta(0.62%)Methanosarcina(0.59%)Spirochaeta(0.55%)Thiobacillus(0.39%)uncultured,Comamonadaceae(0.39%)Chlorochromatium(0.39%)Ensifer(0.32%)Others34%11%9%7%6%6%4%3%2%2%2% 2%2%1% 1%1%1%1%0%0% 0%0%0%3%As(V) medium Sedimentibacter(34%)Desulfitobacterium(11%)Proteiniclasticum(9.1%)Proteiniphilum(7.1)Incertae Sedis(6.1%)Clostridium(6%)Atopococcus(4.4%)Simplicispira(2.9%)Pseudomonas(2.3%)Spirochaeta(2.2%)UCT N117(2.2%)Methanocorpusculum(2%)Paenibacillus(1.9%)Trichococcus(1.2%)Desulfosporosinus(1%)Parabacteroides(0.9%)uncultured Lachnospiraceae(0.5%)Sphingobacterium(0.5%)Deltaproteobacteria,GR- WP33- 30 (0.4%)Lactobacillus(0.3%)Methanoregula(0.3%)Providencia(0.3%)Rhodovulum(0.3%)othersFigure C.3: Distribution of pyrotag reads among genera161AppendixC.SupplementaryDataChapter3Table C.3: Arsenic speciation and concentration analysis using high-pressure liquid chromatography (HPLC): (a)As(III) mediumSample ID Origin No. ofpas-sagesRT,As(III)At,As(V)Areaarsenite(µV ∗sec)Areaarsenate(µV ∗sec)As(III)ppmAs(V)ppmAs(III)oxidationrate %Control Medium 2 0.986 6.615 893026 123713 13.56 NA -8.3AsIII- 1P2 TS1 2 0.984 6.618 665661 31444 10.14 7.2 18.96AsIII- 2P2 TS2 2 0.983 6.619 PN 214391 NAa 30.72 NAAsIII- 3P2 TS3 2 0.987 6.561 777762 64521 11.82 11.46 5.518AsIII- 4P2 TS4 2 0.985 6.622 368069 161287 5.67 23.9 54.66AsIII- 5P2 TS5 2 0.987 6.607 724146 40261 11.02 8.33 11.95AsIII- 6P2 TS6 2 0.983 6.617 813192 217288 12.36 31.1 1.27Control Medium 3 0.984 4.656 1124082 121628 17.03 NA -2.3AsIII- 1P3 TS1 3 PN 6.792 PN 975599 NA 128.56 NAAsIII- 2P3 TS2 3 PNb Pc PN P NA NA NAAsIII- 3P3 TS3 3 PN P PN P NA NA NAAsIII- 4P3 TS4 3 0.984 6.738 489466 P 7.5 NA 54.95AsIII- 5P3 TS5 3 0.986 6.75 640301 P 9.76 NA 41.34AsIII- 6P3 TS6 3 0.985 6.766 946432 18503 14.36 5.54 13.72Control Medium 4 1.007 4.656 1329133 121628 20.1 P 0AsIII- 1P4 TS1 4 PN P PN P NA NA NAAsIII- 2P4 TS2 4 PN P PN P NA NA NAAsIII- 3P4 TS3 4 PN P PN P NA NA NAAsIII- 4P4 TS4 4 PN P PN P NA NA NAAsIII- 5P4 TS5 4 PN P PN P NA NA NAAsIII- 6P4 TS6 4 1.005 6.617 873345 649267 13.26 86.62 34.04aNA : Was not possible to accurately determine the concentrationbPN : A peak was visible, but was too small to be quantified accuratelycP: A peak was present, but too broad to be quantified accurately162AppendixC.SupplementaryDataChapter3Table C.4: Arsenic speciation and concentration analysis using high-pressure liquid chromatography (HPLC): (b)As(V) mediumSample ID Origin No. ofpas-sagesRT,As(III)At,As(V)Areaarsenite(µV ∗sec)Areaarsenate(µV ∗sec)As(III)ppmAs(V)ppmAs(V)reductionrate %Control Medium 2 0.985 6.629 797952 68861 PN 12.02 0AsV- 1P2 TS1 2 0.983 6.618 937957 9522 14.23 4.39 63.5AsV- 2P2 TS2 2 0.984 6.655 1021922 4632 15.5 3.76 68.73AsV- 3P2 TS3 2 0.983 6.579 993048 11480 15.06 4.64 61.41AsV- 4P2 TS4 2 0.983 6.652 927344 61983 14.07 11.13 7.361AsV- 5P2 TS5 2 0.981 6.602 924467 PNa 14.03 NAb NAAsV- 6P2 TS6 2 0.987 6.642 812824 7777 12.35 4.16 65.37Control Medium 3 0.982 6.739 NA 54008 PN 10.11 0AsV- 1P3 TS1 3 0.984 6.749 835243 24129 12.69 6.26 38AsV- 2P3 TS2 3 0.982 6.754 1030308 15574 15.62 5.17 48.88AsV- 3P3 TS3 3 0.981 6.73 902170 11289 13.7 4.61 54.33AsV- 4P3 TS4 3 0.978 6.744 625343 15650 9.54 5.17 48.78AsV- 5P3 TS5 3 0.98 6.739 716755 52984 10.91 9.97 1.3AsV- 6P3 TS6 3 0.986 6.759 630821 4034 9.62 3.68 63.56Control Medium 4 0.989 6.638 806544 47497 PN 9.27 0AsV- 1P4 TS1 4 0.997 6.666 Pc 14785 NA 5.06 45.36AsV- 2P4 TS2 4 1.003 6.676 P 7754 NA 4.16 55.11AsV- 3P4 TS3 4 0.99 6.648 884734 9449 13.43 4.39 52.76AsV- 4P4 TS4 4 0.998 6.681 998621 PN 15.14 NA NAAsV- 5P4 TS5 4 0.991 6.626 positive 8031 NA 4.2 54.72AsV- 6P4 TS6 4 1.004 PN 794895 PN 12.08 NA NAaPN : A peak was visible, but was too small to be quantified accuratelybNA : Was not possible to accurately determine the concentrationcP : A peak was present, but was too broad to be quantified accurately163AppendixC.SupplementaryDataChapter30.1E621E783EC Gypsum treated oil sands tailings HQ069949T6600 E77EC As oxidizing bioreactor GU557154EC As contaminated sediment JQ810460Acidovorax cattleyae strain ICMP2826 NR 041756Acidovorax spG3DM83 EU037287E114T4024T2263EC As contaminated sediment JQ810421EC Subsurface aquifer sediment JX225745E92Beta proteobacterium NOS8 AB076846E474EC Uncultured Simplicispira S carrier bioreactor JQ723662T3098E1392T3223E536EC U contaminated sediment DQ316800EC Subsurface aquifer sediment As mobilization FJ205137Aquabacterium spP136 AM412127T4549Thermomonas fusca strain R 10289 NR 025577 T335Acinetobacter spULV16 DQ357691E26T2114E76Pseudomonas spAKS2 HQ416710E80Geobacter strain CdA2 Y19190T612EC Anoxic marine fjord FJ615142Rhizobium spp492011 HQ652582E740E1250T8260EC Gypsum treated oilsands tailings HQ044209E700Clostridium propionicum strain X2 NR 029269EC Trichloroethene dechlorinating enrichments AY217423 T3314T8801E894EC Gypsum treated oilsands tailings HQ062420T1371EC As reducing enrichment industrial soil HQ896299T4299Clostridium beijerinckii strain E080 JX267098 EC Heavy metal contaminated site estuarine sediment HQ132386 E67E802EC Gypsum treated oilsands tailings HQ092874E176T4404EC Fe reducing enrichments estuary sediment DQ677011EC Cr contaminated site primary settling pond HM468016E189T5077T1607E30EC Heavy metal and U contaminated creek soil HM992484E386E388Sedimentibacter spJLN1 JQ918080Sedimentibacter spD7 AY766467Trichococcus flocculiformis strain DSM2094 NR 042060 E600E335EC Dioxin dehalogenating enrichments AM933652Trichococcus spJ9 GQ406052EC FeOOH reducing culture AB755786T5417Desulfitobacterium chlororespirans strain Co23 NR 026038E605EC As reducing enrichments HQ896302E451 EC As contaminated sediment JQ810475Desulfosporosinus hippei DSM8344 Y11571T3158Desulfosporosinus lacus strain STP12 NR 042202E313 T332 EC Natural acidic rock drainage sediment JQ420024EC anaerobic methanogenic reactor EU591655T9559E990EC Biogas fermentation enrichments GU476607T3594E178EC Bioreactor simulating low Tem oil reservoir HM520917EC uncultured candidate division WS6 GU179944T7265EC Gypsum treated oil sands tailings HQ076788 E720Flavobacterium spC2 HQ188912 E100 EC Gypsum treated oilsands tailings HQ104348T10004EC oil sands tailings pond EF420217T377E43EC Heavy metal contaminated site U mill tailings AJ582209T4931E491EC Pilot scale, HC bioremediation process AM936244T4148T3046E331uncultured candidate division WS6 HQ057802E676T6072EC microbial fuel cell JF309189EC Gypsum treated oilsands tailings HQ046508T7803E616T7353EC Gypsum treated oil sands tailings HQ059809E226EC Anaerobic digester sludge AB494244E586EC Coal bed water methane formation EU168262T6206Figure C.4: Prevalence of putative CAO and DARPs in the BCR164Appendix C. Supplementary Data Chapter 3Dataset Enrichment OTUs abundance As(V) media As(III) mediaTrail OTUs abundanceColor ranges:Betaproteobacteria_ComamonadaceaeFibrobaccteres_KD2−123Gammaproteobacteria_XanthomonadaceaeBacteroidetes_SphingobacteriaceaeGammaproteobacteria_PseudomonadaceaeDeltaproteobacteria_GeobacteraceaeFirmicutes_LachnospiraceaeRF3Firmicutes_ClostridiaceaeBetaproteobacteria_UCT N117Firmicutes_Family XI Incertae SedisBacteroidetes_SphingobacteriaceaeFirmicutes_PeptococcaceaeSynergistaceaeSpirochaetaceaeBacteroidetes_FlavobacteriaceaeBacteroidetes_vadinHA17Bacteroidetes_SB−1Bacteroidetes_PorphyromonadaceaeCandidate division WS6VictivallaceaeCandidate division OP11Archaea_FamilyIncertaeSedisArchaea_MethanocorpusculaceaeFigure C.5: Legend - figure C.4165Appendix DSupplementary DataChapter 4 - Part 1Table D.1: Minimum inhibitory concentration (MIC) analysis of As(V) andAs(III) for Esherichia coli EPI300 (expression host)Arsenic Type Conc.(mM) Growth ObservationAs(III) 0.1 >24 hAs(III) 0.5 >24 hAs(III) 1.00 >24 hAs(III) 2.00 >24 hAs(III) 2.50 >24 hAs(III) 3.00 >24 hAs(III) 3.50 >24 hAs(III) 4.00 >24 hAs(III) 4.06 −As(III) 4.60 −As(III) 5.00 −As(III) 5.50 −As(V) 0.1 >24 hAs(V) 0.5 >24 hAs(V) 1.00 >24 hAs(V) 5.00 >24 hAs(V) 8.00 >24 hAs(V) 15 >24 hAs(V) 20 >24 hAs(V) 30 >24 hAs(V) 40 >24 hAs(V) 50 >24 hAs(V) 100 >3 daysAs(V) 105 >3 daysContinued on next page166Appendix D. Supplementary Data Chapter 4 - Part 1Table D.1 – continued from previous pageArsenic Type Conc.(mM) Growth ObservationAs(V) 110 >3 daysAs(V) 115 >3 daysAs(V) 120 >3 daysAs(V) 125 >3 daysAs(V) 130 >6 daysAs(V) 150 >6 daysAs(V) 170 >7 daysAs(V) 210 >10 daysAs(V) 230 >18 daysAs(V) 250 >18 daysAs(V) 270 >18 days167Appendix D. Supplementary Data Chapter 4 - Part 1Table D.2: Comparison of fosmid growth rates at different As(V) cons.µ1 =µ max for fosmid growth w/o As(V), µ2 = µ max at 20 mM As(V), µ3 = µmax at 30 mM As(V), µ4 = µ max at 40 mM As(V), µ5 = µ max at 50 mMAs(V), µ6 = µ max at 60 mM As(V), Empty cells= No / low growth was ob-served. Fosmids highlighted with Green represent the metagenomic library1 (July 2008, Core 05, section 35-40cm). Blue, represent the metagenomiclibrary 2 (April 2009, core 03/04, section 20-30cm (Sample TS4)). Violetrepresent the metagenomic library 3 (October 2009, core 04/05, section 20-30cm (Sample TS9)). Fosmids with no ID were not sequenced. Fosmid IDswere confirmed by comparison with fosmid ends ID, using nblast program.Fosmid Id Fosmid µ2/µ1 µ3 /µ1 µ4/µ1 µ5/µ1 µ6 /µ1FOS6228-D1 1 0.426 0.161 0.176 0.073FOS6228-D1-B* 1FOS6225-C12 2Area1-F3 3 0.378 0.187 0.132 0.189FOS6228-I24 4 0.490FOS6225-F1 5 0.323 0.123FOS6225-F2 6 0.277FOS6225-G1 7 0.262FOS6226-H13 8 0.250FOS6226-K15 9 0.297 0.141FOS6225-M1 10 0.230FOS6228-P6 11 0.304 0.145 0.101FOS6225-P8 12 0.298FOS6227-P11 13 0.222 0.108 0.094FOS6228-P13 14 0.240 0.173 0.187 0.133 0.107FOS6232-A3 15 0.237FOS6230-B24 16 0.236FOS6231-C24 17 0.328 0.328FOS6232-C4 18 0.270 0.206 0.190 0.111 0.159FOS6232-C3 19 0.271 0.11420FOS6232-D3 21 0.321FOS6232-D22 22 0.393 0.321 0.303 0.286 0.0285FOS6230-D23 23 0.343 0.283 0.224 0.239FOS6231-E22 24FOS6230-E22 25 0.240Continued on next page168Appendix D. Supplementary Data Chapter 4 - Part 1Table D.2 – continued from previous pageFosmid Id Fosmid µ2/µ1 µ3 /µ1 µ4/µ1 µ5/µ1 µ6 /µ1FOS6230-E24 26 0.167 0.283FOS6230-F14 27 0.274FOS6230-F16 28 0.390 0.238 0.322 0.322 0.203FOS6231-G24 29 0.271FOS6231-G15 30 0.392FOS6232-G05 31 0.316FOS6231-H15 32 0.308FOS6230-G17 33 0.190FOS6230-H24 34 0.255FOS6230-H14 35 0.600 0.475 0.400FOS6230-I23 36 0.690 0.500 0.286 0.310 0.191FOS6232-I24 37 0.32838 0.233FOS6232-I15 39 0.327 0.184 0.286 0.143FOS6230-K18 40 0.490 0.367 0.388 0.367 0.286FOS6231-K18 41 0.299 0.175FOS6230-N24 42 0.757 0.298FOS6230-M14 43 0.270 0.173FOS6231-N18 44 0.349FOS6230-N13 45 0.553 0.404 0.17046FOS6230-N13 47 0.674 0.419 0.325 0.326 0.232FOS6231-N18 48 0.467 0.400 0.334 0.334 0.289FOS6232-PO2 49 0.423 0.288 0.365 0.308 0.231FOS6233-J24 50 0.308 0.290 0.192 0.230FOS6234-K4 51 0.718 0.436 0.038 0.282FOS6235-L20 52 0.32053 0.055 0.030 0.031 0.01254 0.228FOS6239-A07 55 0.500 0.363 0.330 0.241FOS6238-IO1 56 0.500 0.320 0.300FOS6239-A07-Bprime-A57 0.430FOS6239-I21 57FOS6240-M19 58 0.637 0.450 0.361 0.310FOS6237-P04 58Continued on next page169Appendix D. Supplementary Data Chapter 4 - Part 1Table D.2 – continued from previous pageFosmid Id Fosmid µ2/µ1 µ3 /µ1 µ4/µ1 µ5/µ1 µ6 /µ1Area4-F58-B 58Area4-F59-B 59 0.385 0.315 0.229 0.136FOS6240-M19 59FOS6237-P04 59FOS6243-E20 60 0.379 0.310 0.237 0.165FOS6244-F1 61 0.663 0.302 0.332 0.278FOS6243-F7 62 0.474 0.334 0.229FOS6243-G5 63 0.363 0.227 0.152FOS6244-G20 64 0.383 0.207 0.17865 0.616 0.457 0.374 0.11766 0.171 0.128FOS6247-C17 67 0.446 0.385 0.261 0.246 0.170FOS6247-L12 68 0.449 0.275 0.190FOS6247-L12 69 0.368 0.265 0.162 0.132FOS6245-N10 70 0.536 0.482 0.232FOS6245-N10 71 0.400 0.200 0.147 0.133FOS6248-O7 72 0.347 0.142 0.0163FOS6247-P11-A-B 73 0.407 0.270 0.170FOS623-B02 74 0.518 0.375 0.429 0.375 0.375FOS624-G24 75 0.267 0.083FOS6247-L12 76 0.490 0.191FOS624-I29 76Area1-F77 77 0.426 0.333 0.204FOS621-N24 78 0.327 0.250FOS6205-A1 79 0.461 0.292 0.18580 0.240FOS6205-B1 81 0.423 0.407 0.322 0.322 0.288FOS6208-B1 82 0.241 0.172 0.024 0.172FOS6205-B4 83 0.245FOS6205-B10 84 0.232 0.105FOS6207-A15 85 0.28986FOS6208-E24 87 0.231 0.09688 0.261FOS6206-O13 89 0.500 0.435 0.217 0.261FOS6207-O14 90 0.254Continued on next page170Appendix D. Supplementary Data Chapter 4 - Part 1Table D.2 – continued from previous pageFosmid Id Fosmid µ2/µ1 µ3 /µ1 µ4/µ1 µ5/µ1 µ6 /µ191 0.20492 0.203FOS6211-A18 93 0.383 0.333 0.167FOS6209-B8 94 0.260 0.120FOS6209-B19 95 0.237 0.286 0.186 0.18696 0.333FOS6210-I1 97 0.357 0.143FOS6216-H19 98 0.360 0.297 0.188 0.266 0.156FOS6209-M3 99 0.391 0.266 0.281 0.219 0.219FOS6211-O13 100 0.351 0.263 0.250 0.210 0.281FOS6211-O24 101 0.412 0.382 0.279 0.206FOS6209-P8 102 0.406 0.297 0.250 0.203 0.172103 0.220 0.147N10-F104 104 0.351 0.298 0.228 0.210 0.210171Appendix D. Supplementary Data Chapter 4 - Part 1Table D.3: Summary of arsenic resistance genes identified in complete fos-mid sequencesYear 2008 2009 2009Month July April OctBorehole TS3 TS4 TS9Core 06 03 04Section-cm 5-10 20-15 20-25Metagenomic library MG1 MG2 MG3Pst 5 30 7Phn 17 0 0metal transporters 5 16 6ArsC 2 4 0ArsA/B 3 9 3Ars regulators 16 30 4methylation ArsRM 4 9 2Aox 16 51 10Arr 0 1 0Potential As genes 68 150 32Hypothetical genes11 11 26 10Total number of genes12 79 176 4211Not annotated12Total number of genes identified in complete fosmid sequences, i.e., annotated ORFsplus hypothetical genes172Appendix D. Supplementary Data Chapter 4 - Part 1Table D.4: Summary of arsenic resistance genes identified in fosmid insertsequence endsYear 2008 2009 2009Month July April OctBorehole TS3 TS4 TS9Core 06 03 04Section-cm 5-10 20-15 20-25Metagenomic library MG1 MG2 MG3Pst 50 74 58Phn 83 83 64metal transporters 40 60 15ArsC 15 10 6ArsA/B 38 55 30Ars regulators 43 72 42methylation ArsRM 1 2 2Aox 63 87 39Arr 8 16 6Potential As genes 341 459 262Blast output hits13 1293 1535 80813 Total number of top hits identified from comparison of fosmid end sequences againstthe custom As gene database. Top hits have E-value less than 1E-06 and identity scoresof more than 30%.173Appendix D. Supplementary Data Chapter 4 - Part 100.10.20.30.40.50.60 5 10 15 20 25600nm absurbanceIncubation hour00.10.20.30.40.50.60 5 10 15 20 25600nm AbsurbanceIncubation hour00.10.20.30.40.50.60 5 10 15 20 25600nm AbsorbanceIncubation HourFosmid Growth w/o As(V)Fosmid Growth in 20mM As(V)Fosmid Growth in 30mM As(V)Fosmid Growth in 40mM As(V)Fosmid Growth in 50mM As(V)Fosmid Growth in 60mM As(V)FOS6228-D1FOS6205-B1FOS6205-B10Figure D.1: Phenotype characterization of the resistant fosmid clones, fewexamples of the fosmid clones are presented174Appendix D. Supplementary Data Chapter 4 - Part 100.10.20.30.40.50.60 5 10 15 20 25600nm AbsorbanceIncubation HourFosmid Growth w/o As(V)Fosmid Growth in 20mM As(V)Fosmid Growth in 30mM As(V)Fosmid Growth in 40mM As(V)Fosmid Growth in 50mM As(V)Fosmid Growth in 60mM As(V)00.10.20.30.40.50.60 5 10 15 20 25600nm absurbance600nm absurbanceIncubation hourArea 1-f7700.10.20.30.40.50.60 5 10 15 20 25600nm absurbanceIncubation hourFOS6209-M3Figure D.2: Phenotype characterization of the resistant fosmid clones, fewexamples of the fosmid clones are presented175Appendix D. Supplementary Data Chapter 4 - Part 1pstphnmetal transporterFull fosmid screeningFosmid end screening2% 5% 10% 15%Figure D.3: Summary of arsenic resistance genes identified in full fosmidsand fosmid ends, that were potentially involved in arsenic resistance mecha-nisms, Metagenomic library 1. Circles area represents percentage of arsenicresistant genes identified in the Metagenomic Library 1 (number of As genesin total identified As genes in the MG1)176Appendix D. Supplementary Data Chapter 4 - Part 1pstphnmetal transporterFull fosmid screeningFosmid end screening2% 5% 10% 15%Figure D.4: Summary of arsenic resistance genes identified in full fosmidsand fosmid ends, that were potentially involved in arsenic resistance mecha-nisms, Metagenomic library 3. Circles area represents percentage of arsenicresistant genes identified in the Metagenomic Library 3 (number of As genesin total identified As genes in the MG3).177Appendix D. Supplementary Data Chapter 4 - Part 1Area1-F3Area1-F77FOS6205-A1FOS6205-B1FOS6205-B10FOS6205-B4FOS6206-O13FOS6207-A15FOS6207-O14FOS6208-B1FOS6208-E24FOS6209-B19FOS6209-B8FOS6209-M3FOS621-N24FOS6210-I1FOS6211-A18FOS6211-O13FOS6211-O24FOS6211-P8FOS6216-H19FOS6225-C12FOS6225-F1FOS6225-F2FOS6225-G1FOS6225-M1FOS6225-P8FOS6226-H13FOS6226-K15FOS6227-P11FOS6228-D1-B*FOS6228-I24FOS6228-P13FOS623-B02FOS6230-B24***FOS6230-D23FOS6230-E22FOS6230-E24FOS6230-F14FOS6230-F16FOS6230-G17FOS6230-H14FOS6230-H24FOS6230-I23FOS6230-K14FOS6230-K18FOS6230-M14FOS6230-N13FOS6230-N18FOS6230-N24FOS6231-C24FOS6231-E22FOS6231-G15FOS6231-G24FOS6231-H15FOS6231-K18FOS6231-N10FOS6231-N18FOS6232-A3FOS6232-C3FOS6232-C4FOS6232-D22FOS6232-D3FOS6232-G5FOS6232-I15FOS6232-I24FOS6232-K14FOS6232-P02FOS6233-j24FOS6234-K4FOS6235-L20FOS6237-P04FOS6238-IO1FOS6239-A07FOS6239-A07-BPrime_AFOS6239-I21FOS624-G24FOS624-I24FOS6240-M19FOS6243-E20FOS6243-F7FOS6243-G5FOS6244-F1FOS6244-G20FOS6245-N10FOS6247-C17FOS6247-L12FOS6247-P11-A_BFOS6248-O7FOs6228-P6Figure D.5: Circos representation of all complete fosmid sequences. Circosplot representing arsenic resistant positive fosmids identified through func-tional screening. The outer circle represents fosmids, labeled FOS6232-I15,FOS6232-D22 etc. Connections in the center show regions of nucleotidehomology at greater than 90% similarity across intervals of more than 300bp.178Appendix D. Supplementary Data Chapter 4 - Part 1FOS6205-A1510152025303540 FOS6205-B1510152025303540FOS6205-B10510152025303540455055FOS6205-B45101520253035FOS6206-O135101520253035FOS6207-A155101520253035FOS6208-B15 1015 20253035FOS6208-E245101520253035FOS6209-B195 10 15 20 25 30 35FOS6209-B85 10 15 20 25 30 35FOS6209-M35 1015 20 25 30 35FOS621-N245101520253035FOS6210-I1510152025303540FOS6211-A185101520253035FOS6216-H19 510152025303540FOS624-G24 510152025303540FOS624-I24 510152025303501000100 Pst Phn metal transporters efflux pumps Ars A/BArsCArs regulators methylation Ars RM These are shown whitin the fosmid boxes (1th circle) These are shown in the next inner circle (2th circle) (3th circle)Figure D.6: Circos representation of complete fosmid sequences using Illu-mina sequencing. Circos plot representing arsenic resistant positive fosmidsidentified from metagenomic library 1 (July2008). The outer circle repre-sents fosmids, labeled FOS624-I24, FOS624-G24 etc. Numbers on the outercircle show size scale in kilobases (kb). Colored bars within fosmids and in-ner circles show locations of As genes. Black histograms indicate the percentidentity of the blast hits. The two layers indicate percent Identity againstthe genebank database (COG, KEGG, RefSeq) (histogram outer layer) andarsenic database (histogram inner layer).179Appendix D. Supplementary Data Chapter 4 - Part 1Area1-F35101520253035 FOS6225-F251015202530 FOS6225-G151015202530FOS6225-M15101520253035FOS6226-K155101520253035FOS6227-P11510152025303540FOS6228-I245101520253035FOS6230-D235 10152025FOS6230-E245 10 1520 2530FOS6230-F145 10 1520 25 3035FOS6230-F165 10 15 20 25 30FOS6230-H245 10 15 20FOS6230-N185 10 15 20 25 30 35 40FOS6230-N245 10 15 20 25 30FOS6231-G15 510152025FOS6231-G24510152025FOS6231-N185101520253035FOS6232-A3 51015202530FOS6232-D22 51015202530FOS6232-D3 51015202530FOS6232-I245101520253035FOS6232-P025101520250100Figure D.7: Circos representation of complete fosmid sequences potentiallyassociated with aox, identified from metagenomic library 2 (April2009).The outer circle represents fosmids, labeled FOS6232-P02, FOS6232-I24 etc.Numbers on the outer circle show size scale in kilobases (kb). Green barswithin fosmids show locations of genes corresponds to aox operon. Blackhistograms indicate the percent identity of the blast hits.180Appendix D. Supplementary Data Chapter 4 - Part 1FOS6205-B10510152025303540455055FOS6207-A155101520253035FOS6208-B15101520253035FOS6208-E245 10 1520253035FOS6210-I151015 20 25 30 35 40FOS6211-O13510152025FOS6211-O24510152025303540FOS623-B02 510152025300100Figure D.8: Circos representation of complete fosmid sequences poten-tially associated with aox, identified from metagenomic library 1 (July2008).The outer circle represents fosmids, labeled FOS623-B02, FOS6211-O24 etc.Numbers on the outer circle show size scale in kilobases (kb). Green barswithin fosmids show locations of genes corresponds to aox operon. Blackhistograms indicate the percent identity of the blast hits.181Appendix D. Supplementary Data Chapter 4 - Part 1FOS6237-P0451015202530FOS6238-IO1510152025FOS6239-A075101520253035FOS6239-I215 10 1520253035FOS6243-E2051015 20 25 30FOS6243-F751015202530FOS6244-F15101520253035FOS6247-L12 5101520250100Figure D.9: . Circos representation of complete fosmid sequences poten-tially associated with aox, identified from metagenomic library 3 (Oct 2009).The outer circle represents fosmids, labeled FOS6247-L12, FOS6244-F1, etc.Numbers on the outer circle show size scale in kilobases (kb). Green barswithin fosmids show locations of genes corresponds to aox operon. Blackhistograms indicate the percent identity of the blast hits.182Appendix D. Supplementary Data Chapter 4 - Part 1FOS6233-j24510152025303540FOS6234-K451015202530FOS6235-L20510152025FOS6239-I215101520253035FOS6240-M195 101520253035FOS6243-E205 101520 2530FOS6243-G55 10 1520 25FOS6244-F15 10 15 20 25 30 35FOS6244-G205 10 15 20 25 30 35FOS6245-N10510152025 3035FOS6247-C1751015202530FOS6247-L12 510152025FOS6247-P11-A_B 5101520253035FOS6248-O7 510152030100 Pst Phn metal transporters efflux pumps Ars A/BArsCArs regulators methylation Ars RM These are shown whitin the fosmid boxes (1th circle) These are shown in the next inner circle (2th circle) (3th circle)Figure D.10: Circos representation of complete fosmid sequences using Illu-mina sequencing. Circos plot representing arsenic resistant positive fosmidsidentified from metagenomic library 3 (Oct 2009). The outer circle repre-sents fosmids, labeled FOS6248-O7, FOS6247-P11-A-B, etc. Numbers onthe outer circle show size scale in kilobases (kb). Colored bars within fos-mids and inner circles show locations of As genes. Black histograms indicatethe percent identity of the blast hits. Identity below 30% were not shown.183Appendix D. Supplementary Data Chapter 4 - Part 1FOS6205-A1510152025303540 FOS6205-B1510152025303540FOS6205-B10510152025303540455055FOS6205-B45101520253035FOS6206-O135101520253035FOS6207-A155101520253035FOS6208-B15 1015 20253035FOS6208-E245101520253035FOS6209-B195 10 15 20 25 30 35FOS6209-B85 10 15 20 25 30 35FOS6209-M35 1015 20 25 30 35FOS621-N245101520253035FOS6210-I1510152025303540FOS6211-A185101520253035FOS6216-H19 510152025303540FOS624-G24 510152025303540FOS624-I24 510152025303530100 Pst Phn metal transporters efflux pumps Ars A/BArsCArs regulators methylation Ars RM These are shown whitin the fosmid boxes (1th circle) These are shown in the next inner circle (2th circle) (3th circle)Figure D.11: Circos representation of complete fosmid sequences using Illu-mina sequencing. Circos plot representing arsenic resistant positive fosmidsidentified from metagenomic library 1 (July 2008). The outer circle rep-resents fosmids, labeled FOS624-I24, FOS624-G24, etc. Numbers on theouter circle show size scale in kilobases (kb). Colored bars within fosmidsand inner circles show locations of As genes. Black histograms indicate thepercent identity of the blast hits. Identity below 30% were not shown.184Appendix D. Supplementary Data Chapter 4 - Part 1No hits; 11Not assigned; 7254unclassified sequences; 11Streptophyta; 24Chordata; 5Eukaryota; 8 Euryarchaeota; 1332Thermotogae ; 15Spirochaetes; 138Proteobacteria; 397Fusobacteria; 12Firmicutes; 467Fibrobacteres; 142Deinococcus-Thermus; 8Cyanobacteria; 28Chloroflexi ; 33Verrucomicrobia; 6Lentisphaerae; 5Chlamydiae/Verrucomicrobia group; 2 Bacteroidetes; 449Actinobacteria ; 20Bacteria; 502cellular organisms; 234root; 2Figure D.12: Megan 5 analysis of the full fosmid sequences against theNCBI-NR database. Numbers in taxonomic tree indicate the number ofassigned reads to each taxon collapsed at phylum level of NCBI taxonomy.185Appendix D. Supplementary Data Chapter 4 - Part 1No hitsNot assignedhybridArabidopsis thalianaTetrapodaEukaryota Pyrococcus horikoshiiMethanopyrus kandleriMethanosarcina barkeriMethanosarcina acetivoransMethanosarcina Methanospirillum hungateiMethanoregula booneiMethanoplanus petroleariusMethanocorpusculum labreanumMethanomicrobialesMethanomicrobiaMethanococcusMethanocaldococcus jannaschiiMethanococcales Methanothermobacter thermautotrophicusMethanobacterium sp. SWAN-1Methanobacteriaceae Archaeoglobus fulgidusEuryarchaeotaThermotoga maritimaTreponema succinifaciensTreponema pallidumTreponema brennaborenseTreponema Spirochaeta smaragdinaeBorrelia burgdorferiSpirochaetaceae Vibrio choleraePseudomonas aeruginosaPasteurellaceaeYersinia pestisSalmonellaEscherichia coliEnterobacteriaceae AlteromonadalesGammaproteobacteriaHelicobacter pyloriDeltaproteobacteriadelta/epsilon subdivisions Neisseria meningitidisRalstonia solanacearumBetaproteobacteria Sinorhizobium melilotiRhizobium/Agrobacterium groupRhizobiaceaeMesorhizobium lotiHyphomicrobiumBradyrhizobiaceaeRhizobialesCaulobacter vibrioidesAlphaproteobacteriaProteobacteriaFusobacterium nucleatumThermoanaerobacterAcetivibrio cellulolyticusLachnospiraceae bacterium 3_1_57FAA_CT1EubacteriaceaeClostridium acetobutylicumClostridialesClostridiaLactococcus lactisStaphylococcus aureusBrevibacillus brevisListeria innocuaBacillus subtilisBacillus haloduransBacillusBacillalesBacilliFirmicutesFibrobacter succinogenesDeinococcus radioduransChroococcalesNostoc sp. PCC 7120Cyanobacteria ChloroflexalesAnaerolinea thermophilaChloroflexi VerrucomicrobiaLentisphaeriaChlamydiae/Verrucomicrobia groupSphingobacteriaceaeFlavobacteriaceaeCytophagaceaePrevotellaPorphyromonasPaludibacter propionicigenesDysgonomonasPorphyromonadaceae Anaerophaga thermohalophilaBacteroides fragilisBacteroidalesBacteroidetesMycobacterium tuberculosisBacteriacellular organismsrootFigure D.13: Megan 5 analysis of the full fosmid sequences against theNCBI-NR database. Numbers in taxonomic tree indicate the number ofassigned reads to each taxon collapsed at species level of NCBI taxonomy.186Appendix D. Supplementary Data Chapter 4 - Part 1gi|260161442_274-681_ArsC_Microbacteriumgi|33867057_136876-137292_ArsC_Rhodococcus_erythropolis0.1210.136gi|119952551_59392-59814_ArsC_Arthrobacter_aurescens0.0090.139gi|119952551_62445-63101_ArsC_Arthrobacter_aurescens0.0600.233gi|260161442_710-1705_ArsRC_Microbacterium0.0370.276gi|33867057_138348-138758_ArsC_Rhodococcus_erythropolis0.0590.323gb|AE001438.3|_111324-111734_ArsC_Clostridium_acetobutylicumFOS6217.CPR.1_N06_Fosmid end MG20.2050.235gi|125973625| protein tyrosine phosphatase_Clostridium thermocellum 0.0420.2220.0230.120gi|225184640_2655322-2655741_ArsC_Bacillus_subtilisgi|251796195| arsenate reductase_Paenibacillus0.2140.210gi|15282445_465181-465615_ArsC_Aquifex_aeolicus0.0600.288gi|134093294_3263875-3264375_ArsC_Herminiimonas_arsenicoxydansgi|134093294_3167176-3167682_ArsC_Herminiimonas_arsenicoxydans0.1380.151gi|134093294_503274-503789_ArsC_Herminiimonas_arsenicoxydans0.0180.147gi|94308945_c350232-349720_ArsC2_Cupriavidus_metallidurans0.0240.179gi|134093294_1981471-1981965_ArsCa_Herminiimonas_arsenicoxydans0.0390.244FOS6236.CPR_F17_JAN_Fosmid end MG3gi|302340043| protein-tyrosine phosphatase_Spirochaeta smaragdinae0.2230.218gi|224367295| protein ArsC1_Desulfobacterium autotrophicum0.0480.261gi|15679354| arsenate reductase_Methanothermobacter thermautotrophicusgi|219853132| protein-tyrosine phosphatase_Methanosphaerula palustris0.2670.2560.0410.0370.1200.0150.0560.0100.034gi|62423175_ACR3_Brevibacterium_linensFOS6231.CPF_J14_Fosmid end MG2FOS6228.CPR_C22_Fosmid end MG2FOS6224.CPF_G05_Fosmid end MG20.1810.1780.2270.0410.2650.037gi|134093294_ACR3_Herminiimonas_arsenicoxydansBIG_HYBRID0.2820.2920.1030.0740.0750.106FOS6230-K180.0090.457FOS6209-B8FOS6205-B1hypothetical_Methanospirillum labreanum0.2860.294FOS6221.CPF.1_P12_Fosmid end MG2FOS6247.CPR_I09_Fosmid end MG30.4550.445FOS6232-D220.0090.4400.1500.2Figure D.14: Putative ArsC identified from Full fosmids and fosmids endsand its homology to ArsC in various arsenic-metabolizing microorganisms.Some of the fosmid ends combined to one big hybride. Prodigal output pro-tein sequences were aligned with ClustalX 2.1 and clustering NJ (Neighborjoining) algorithm was used for build the phylogenetic tree. MG stands forMetagenomic library.187Appendix D. Supplementary Data Chapter 4 - Part 1FOS6209-B8FOS6205-B1hypothetical_ Methanocorpusculum labreanum Z0.2890.290FOS6232-D220.1520.446FOS6247.CPR_I09_Fosmid end MG3FOS6230-K180.4470.4330.029gi|116668568_ArsC_Arthrobactergi|119960487_ArsC_Arthrobacter_aurescens0.0260.031gi|260161442_ArsC_Microbacterium0.1100.121gi|33867057_ArsC_Rhodococcus_erythropolis0.139gi|260161442_ArsRC_Microbacterium0.1000.277gi|33867057_ArsC_Rhodococcus_erythropolis0.0430.325gb|AE001438.3|_ArsC_Clostridium_acetobutylicumFOS6217.CPR.1_N06_Fosmid end MG20.2060.235gi|125973625| protein tyrosine phosphatase_Clostridium 0.0430.2200.0310.115gi|225184640_ArsC_Bacillus_subtilisgi|251796195| arsenate reductase Paenibacillus0.2110.213gi|15282445_ArsC_Aquifex_aeolicus0.0630.284gi|134093294_ArsC_Herminiimonas_arsenicoxydansgi|134093294_ArsC_Herminiimonas_arsenicoxydans0.1380.151gi|134093294_ArsC_Herminiimonas_arsenicoxydans0.0180.146gi|94308945_ArsC2_Cupriavidus_metallidurans0.0260.177gi|134093294_ArsCa_Herminiimonas_arsenicoxydans0.0380.244FOS6236.CPR_F17_Fosmid end MG3gi|302340043| protein-tyrosine phosphatase_Spirochaeta0.2260.215gi|224367295| ArsC1 Desulfobacterium autotrophicum0.0500.259gi|15679354| arsenate reductase_Methanothermobacter gi|219853132| protein-tyrosine phosphatase_Methanosphaerula 0.2720.2510.0450.0330.1250.0090.0470.0140.0150.0150.077FOS6228.CPR_C22_Fosmid end MG2FOS6231.CPF_J14_Fosmid end MG20.1930.201gi|62422723_ACR3_Brevibacterium_linens0.0690.268FOS6224.CPF_G05_Fosmid end MG2FOS6211.CPR_A06_Fosmid end MG10.021FOS623.CPF_P22_Fosmid end MG1FOS6212.CPR_F10_Fosmid end MG10.0300.0280.1800.165gi|134093294_ACR3_Herminiimonas_arsenicoxydans0.1020.2870.0650.041FOS6221.CPF.1_P12_Fosmid end MG20.1270.4400.0190.1Figure D.15: Putative ArsC identified from Full fosmids and fosmids endsand its homology to ArsC in various arsenic-metabolizing microorganisms.Prodigal output protein sequences loaded in ClustalX 2.1 for alignment andclustering NJ (Neighbor joining) algorithm was used for build the phyloge-netic tree, MG stands for Metagenomic library.188Appendix ESupplementary DataChapter 4 - Part 2051050100Carbohydrate MetabolismEnergy MetabolismLipid MetabolismNucleotide MetabolismAmino Acid MetabolismMetabolism of Other Amino AcidsGlycan Biosynthesis and MetabolismMetabolism of Cofactors and VitaminsMetabolism of Terpenoids and PolyketidesBiosynthesis of Other Secondary MetabolitesXenobiotics Biodegradation and MetabolismTranscriptionTranslationFolding, Sorting and DegradationReplication and RepairMembrane TransportSignal TransductionTransport and CatabolismCell MotilityCell Growth and DeathEndocrine SystemNervous SystemCancersImmune DiseasesNeurodegenerative DiseasesSubstance DependenceCardiovascular DiseasesEndocrine and Metabolic DiseasesInfectious DiseasesUnclassified95 85114410015246215 16 1456428 425331137141631 1 1 1313451Figure E.1: Potential pathways identified by metagenomic functionalscreening method form the CBR. Full fosmid sequences were blast againstthe protein databases using MetaPathway pipeline. Numbers in KEGG pro-file indicate the number of assigned reads (log scale) to each pathway.189Number of reads (log scale)KEGGAppendix E. Supplementary Data Chapter 4 - Part 2Table E.1: MLtreeMap taxonomy analysisstart end COG score14 taxonomy15FOS6208-B1 33703 34416 mcrAref 402 mcrAFOS6208-E24 31923 32795 COG0552 546 Methanocorpusculumlabreanum32809 34083 COG0541 7696209 7795 COG0018 955FOS6209-B19 7102 7596 COG0197 305 MethanocorpusculumlabreanumFOS6209-M3 18967 19278 COG0016 163 Methanocorpusculumlabreanum19281 20381 COG0016 644FOS6211-O13 141 545 COG0102 126 LCA18 of Thermococcuskodakarensis589 984 COG0103 159 Pyrococcus horikoshiiFOS6211-O24 1281 2213 COG0016 458 LCA of BacteroidesthetaiotaomicronPorphyromonas gingivalisParabacteroides distasonisFOS6226-K15 1440 1754 COG0103 165 Bacteroides thetaiotaomi-cron1790 2209 COG0102 217584 1231 COG0052 344FOS6228-I24 29053 30300 COG0172 484 LCA of BacteroidesthetaiotaomicronPorphyromonas gingivalisParabacteroides distasonisFOS6228-P13 2606 3154 COG0049 285 Methanospirillum hungatei3199 3624 COG0048 26835504 36526 COG0533 5378247 9881 COG0018 759FOS6230-K14 6924 8102 COG0012 577 Methanosphaera stadt-manae20565 20939 K025867 65.9 nifDFOS6231-E22 853 2100 COG0172 428 7 groups19Continued on next page18LCA: Lower common ancestor19Assignment of query to the lowest common ancestor of Bacteroides thetaiotaomicron190Appendix E. Supplementary Data Chapter 4 - Part 2Table E.1 – continued from previous pagestart end COG score16 taxonomy17FOS6231-H15 10385 11056 COG0052 331 LCA of Bacteroidesthetaiotaomicron9444 9854 COG0102 188 Porphyromonas gingivalis9447 9866 COG0102 188 Parabacteroides distasonis9905 10267 COG0103 136FOS6232-C3 22136 2302 COG0552 211 LCA of Borrelia burgdor-feriTreponema pallidumFOS6232-C4 2 1900 COG0495 1210 MethanocorpusculumlabreanumFOS6232-K14 6924 8102 COG0012 577 Methanosphaera stadt-manae20222 20596 K02586 65.9 nifDFOS6233-j24 8054 8167 16srRNA 139 LCA of Parabacteroidesdistasonis8289 9031 16srRNA 436 Bacteroides acidifaciens9087 9581 16srRNA 504 Porphyromonas bennonisFOS6239-I21 1370 2083 mcrAref 402 mcrAFOS6210-I1 36509 38485 bssAref 155 20 bssA-nmsA-assA(226186), Porphyromonas gingivalis (242619), Cytophaga hutchinsonii (269798), Salini-bacter ruber (309807), Flavobacterium johnsoniae (376686), Gramella forsetii (411154)and Parabacteroides distasonis (435591). (14.29%each)20(1)Assignment of query to the lowest common ancestor of Azoarcus sp. T bssA (1),Geobacter metallireducens GS-15 bssA (2), Contaminated Testfeld Sued aquifer sedimentsclone B49bss 020 bssA (3), Contaminated Testfeld Sued aquifer sediments clone B49bss087 bssA (4), Contaminated Flingern aquifer sediments clone D12 09 bssA (5), Contam-inated Flingern aquifer sediments clone D12 31 bssA (6), Contaminated Pasing aquifersediments clone Pb112 15 bssA (7), Desulfosarcina cetonica bssA (8), Desulfobacula tolu-olica Tol2 bssA (9), Geobacter grbiciae bssA (10), Desulfotomaculum sp. Ox39 bssA(11), Geobacter toluenoxydans bssA (12), Sulfate-reducing bacterium TRM1 bssA (13),Toluene degrading methanogenic consortium clone bssA-2 bssA (14), Toluene degradingmethanogenic consortium clone bssA-1 bssA (15), Sulfate-reducing bacterium PRTOL1bssA (16), Coal tar waste-contaminated groundwater clone bssAC03 bssA (17), Coal tarwaste-contaminated groundwater clone bssAC05 bssA (18), Coal tar waste-contaminatedgroundwater clone bssAC14 bssA (19), Coal tar waste-contaminated groundwater clonebssAC15 bssA (20), Coal tar waste-contaminated groundwater clone bssAC16 bssA(21), Coal tar waste-contaminated groundwater clone bssAC18 bssA (22), Coal tarwaste-contaminated groundwater clone bssAD03 bssA (23), Coal tar waste-contaminatedgroundwater clone bssAD10 bssA (24), BTEX-contaminated afquifer enrichment cloneB1B6 bssA (25), BTEX-contaminated afquifer enrichment clone B1C1 bssA (26), BTEX-191Appendix E. Supplementary Data Chapter 4 - Part 2contaminated afquifer enrichment clone B2B2 bssA (27), Toluene-degrading SIP micro-cosms clone LA07bs01 bssA (28), Toluene-degrading SIP microcosms clone LA07bs07 bssA(29), Toluene-degrading SIP microcosms clone LA07bs11 bssA (30), Toluene-degradingSIP microcosms clone LA07bs16 bssA (31), Casper Wyoming contaminated aquifer sedi-ment clone OTU1 bssA (32), Casper Wyoming contaminated aquifer sediment clone OTU4bssA (33), Fort Lupton contaminated aquifer sediment clone OTU1 bssA (34), CasperWyoming contaminated aquifer sediment clone OTU6 bssA (35), Thauera sp. DNT-1bssA (36), Magnetospirillum sp. TS-6 bssA (37), Azoarcus sp. DN11 bssA (38), Thaueraaromatica bssA (39), Aromatoleum aromaticum EbN1 bssA (40), Georgfuchsia toluolicabssA (41), Geobacter sp. FRC-32 bssA (42), Deltaproteobacterium NaphS2 nmsA (43),Deltaproteobacterium NaphS3 nmsA (44), Deltaproteobacterium NaphS6 nmsA (45) andDeltaproteobacterium N47 nmsA (46).192Appendix E. Supplementary Data Chapter 4 - Part 2Metabolism; 346Genetic Information Processing; 135K02020 - molybdate transport system substrate-binding p.; 2K02018 - molybdate transport system permease p.; 2K02012 - iron(III) transport system substrate-binding p.; 2K10108 - maltose/maltodextrin transport system p; 2K10112 - maltose/maltodextrin transport system ATP p.; 1K10440 - ribose transport system permease p.; 1K10441 - ribose transport system ATP-binding p.; 1K02040 - phosphate transport system substrate-binding p.; 2K02038 - phosphate transport system permease p.; 2K02036 - phosphate transport system ATP-binding p.; 2K01999 - branched-chain amino acid transport system P; 1K13893 - microcin C transport system substrate-binding p.; 2K13894 - microcin C transport system permease p.; 1K02016 - Fe complex transport system substrate-binding P;2K02015 - Fe complex transport system permease p.; 4K02013 - Fe complex transport system ATP-binding p.;6K09815 - zinc transport system substrate-binding p.; 1K09816 - zinc transport system permease p.; 1K09817 - zinc transport system ATP-binding p.; 1K02007 - cobalt/nickel transport system permease p.; 1K02008 - cobalt/nickel transport system permease p.; 2K02006 - cobalt/nickel transport system ATP-binding p.; 3K09686 - antibiotic transport system permease p.; 2K09687 - antibiotic transport system ATP-binding p.; 3K11085 - ATP-binding cassette, subfamily B, MsbA; 1K06148 - ATP-binding cassette, subfamily C ; 1ABC transporters; 0Bacterial secretion system; 4Membrane Transport; 0K07636 - two-component system, OmpR family, PhoR ; 1K02040 - phosphate transport system substrate-binding p.; 2K07798 - Cu(I)/Ag(I) efflux system membrane p. CusB/SilB;2K07787 - Cu(I)/Ag(I) efflux system membrane p.CusA/SilA; 1K02405 - RNA polymerase sigma factor, flagella, FliA; 1K02313 - chromosomal replication initiator protein; 1K07696 - two-component system, NarL, regulator NreC; 1K01915 - glutamine synthetase [EC:6.3.1.2]; 4K00575 - chemotaxis protein methyltransferase CheR; 1K03408 - purine-binding chemotaxis p. CheW; 3K03407 - two-component system, sensor kinase CheA ; 3K03413 - two-component system, response regulator CheY;4K03412 - two-component system, response regulator CheB;2K07718 - two-component system, histidine kinase, YesM ; 3K08738 - cytochrome c; 1Two-component system; 0Phosphatidylinositol signaling system; 1Signal Transduction; 0Environmental Information Processing; 0Cellular Processes; 52Organismal Systems; 7Human Diseases; 17Unclassified; 451KEGG; 0Figure E.2: KEGG functional annotation of ORFs potentially involved inmetal transportation. Full fosmid sequences were blast against the proteindatabases using MetaPathway pipeline and analyzed with Megan 5. Num-bers indicate the number of assigned reads to each protein. p stands forprotein193Appendix E. Supplementary Data Chapter 4 - Part 2Figure E.3: ABC transporters identified from KEGG functional annota-tion. Full fosmid sequences were blast against the protein databases usingMetaPathway pipeline and analyzed with Megan 5. Numbers indicate thenumber of assigned reads to each protein.194Appendix E. Supplementary Data Chapter 4 - Part 2Carbohydrates; 90Cofactors, Vitamins, Prosthetic Groups, Pigments; 75Virulence; 30Sulfur Metabolism; 3Stress Response; 18Protein Metabolism; 65RNA Metabolism; 36Motility and Chemotaxis; 26Secondary Metabolism; 2Amino Acids and Derivatives; 54Nitrogen Metabolism; 6Clustering-based subsystems; 27Fatty Acids, Lipids, and Isoprenoids; 15Nucleosides and Nucleotides; 21Cell Wall and Capsule; 47Metabolism of Aromatic Compounds; 2DNA Metabolism; 55Phages, Prophages, Transposable elements; 2Substrate-specific component NikM of nickel ECF transporter; 1ATPase component NikO of energizing module of nickel ECF transporter; 1Transmembrane comp. NikQ of energizing module of Ni ECF transporter; 2Transport_of_Nickel_and_Cobalt; 0 ABC_transporter_branched-chain_amino_acid_(TC_3.A.1.4.1); 1ABC transporters; 1 Kup system potassium uptake protein; 1Large-conductance mechanosensitive channel; 1Trk system potassium uptake protein TrkA; 2putative Glutathione-regulated potassium-efflux system protein KefB; 2Potassium_homeostasis; 0Mn-dependent transcriptional regulator MntR; 2Manganese transport protein MntH; 1Manganese ABC transporter, periplasmic-binding protein SitA; 1Manganese ABC transporter, ATP-binding protein SitB; 1Manganese ABC transporter, inner membrane permease protein SitC; 1Manganese ABC transporter, inner membrane permease protein SitD; 1Transport_of_Manganese; 0 Transmembrane comp. NikQ of energizing module of Ni ECF transporter; 2ATPase component NikO of energizing module of nickel ECF transporter; 1Substrate-specific component NikM of nickel ECF transporter; 1ATPase component of general energizing module of ECF transporters; 2ECF_class_transporters; 0Zinc ABC transporter, ATP-binding protein ZnuC; 1Zinc ABC transporter, periplasmic-binding protein ZnuA; 1Transport_of_Zinc; 0Membrane Transport; 0Respiration; 34Miscellaneous; 10Cell Division and Cell Cycle; 17Dormancy and Sporulation; 1Regulation and Cell signaling; 11Phosphorus Metabolism; 8Potassium metabolism; 1Not assigned; 10496No hits; 11SEED; 0Figure E.4: RefSeq (NCBI-NR) functional annotation of ORFs potentiallyinvolved in metal transportation. Full fosmid sequences were blast againstthe protein databases using MetaPathway pipeline and analyzed with Megan5. Numbers indicate the number of assigned reads to each protein.195Appendix E. Supplementary Data Chapter 4 - Part 2Figure E.5: Enzymes potentially involved in sulfur cycle. The analysisis based on KEGG database annotation. Numbered rectangles representdifferent enzymes that are shaded on a scale from white (corresponding to0 reads) to dark green to indicate the number of reads assigned to eachenzyme.196Appendix FSupplementary DataChapter 4 - Part 3F.1 ProgramAn algorithm program was written in C++. The purpose of the program isto combine multiple tblastn entries into one, if they are of the same fosmidand gene (e.g., ORF). The algorithm is outlined in this section. Refer totable F.1 as a theoretical set of entries that have been simplified to onlyshow the information that matter, and Figure F.1 is a visual representationof the data in Table F.1.1. Mark the longest entry, (b)2. Mark all other entries in respect to the longest entry as, contained,partial contained, extend, mismatch.Begin by marking each Q and S as extend, contained or partial con-tained.extend: entry marked as extend if its sStart, sEnd, qStart and qEndare all greater or less than that of the longest entry.contained: if sStart, sEnd, qStart, qEnd of entry is between that ofthe longest entry. i.e., c.partial contained: cases other than extend or contained.Now check S and Q for agreementi.e., d will be marked as contained for fosmid, but extend for gene, sothey disagree and the entry will be flagged as mismatch.If S and Q agree which is the case for all the other entries, they willbe flagged a such.Now check for direction of transcription.a entry and Black are transcribed in opposite directions so orange will197F.1. Programbe marked as mismatch.Finally, those that have been marked as extend need to be checkedif they extend in the right place. i.e., h entry will be considered amismatch and have its flag changed from extend to mismatch.3. Repeat step two for the entries considered ”extend”. f, g and i will beconsidered extend, and loaded into another vector to be fed back intothe function (a recursive function used in algorithm). Yellow will beconsidered longest, f considered extend and g a partial extend.4. This is repeated until no more entries are considered extend, if oneentry is considered extend, the function will return that one function.5. So now, the function has b, f and i to work with. As it uses a recur-sive function, it begins unwinding backwards by combining green andyellow together and returns it to its caller, who then combines blackto f+i. And then this is returned to its caller, the main program. Soultimately, only entries considered extend matter, all the others arediscarded one way or another.Table F.1: Example used for the written program to combinefosmid end sequences-tblastn entriesentry alignment length q. start q. end s. start s. enda 80 10 90 80 0b 190 100 290 80 270c 60 140 200 130 190d 50 230 280 300 350e 80 260 320 250 310f 70 330 400 320 390g 40 380 420 380 420h 40 390 430 30 70i 80 410 490 410 490198F.1. ProgramFigure F.1: Visual representation of fosmid end sequence analysisF.1.1 Sample DataThe program was used on three libraries of fosmid end sequences for themetal remediation project. Figure F.2 shows three samples of the data be-fore and after running the program, the top table is before, and the lowertable is after the program has been run.199F.1. ProgramFigure F.2: Fosmid end sequence analysis using the written program, threeexamples of the data before and after running the program200Appendix GExperimental and AnalyticalProtocolsG.1 Preparation of Pre-reduced Media for theEnrichment CultureProcedure1. Place tubes, stoppers in glove box one day before using them.2. Weigh out dry ingredients of the media, and place them in proper sizeflasks, add distilled water (consider the volume of arsenic solution that willbe added later).3. Add resazurin (final Cons. 0.001%)4. Bring the culture medium to boil in flask with chimney until the resazurinturns from blue to pink.5. Remove flask from heat and immediately replace chimney with a 2-holestopper and a cannula, which delivers a stream of oxygen-free N2 into themedium. The flow of N2 should cause gentle bubbling.6. After autoclave at 121◦C, under the glove box, add arsenite solution orarsenate solution to the media at or near room temperature.7. Add sodium carbonate, L-Cysteine to arsenite minimal salt enrichmentmedium and arsenate anoxic minimal salts medium respectively.8. Add vitamins and trace element solutions (final cons. 0.1 µ/ml).9. To attain final pH of about 7 for both media, only few drops of HCL-2Nis needed for arsenate medium. (It should be noted that, the pH and DOwere measured before and after autoclave and also prior to inoculation).10. Distribute prepared media into tubes along with inoculum (10%). Sealwith butyl rubber stopper, screw-cap and store in the dark until the nextpassage.201G.2. MIC Determination in L.B Agar MediumG.2 MIC Determination in L.B Agar MediumProcedure1. Prepare desire amount of L.B agar medium, and adjust pH in 7.2.2. Prepare arsenate and arsenite stock solution in advance and sterilize themwith 0.22 µm filter.3. After autoclave of L.B agar medium, add As solution to the medium (Themedium temperature is around 55◦C), just before the medium be solid, andmixed well.4. Pour the media to the regular Petri dish/Omni tray and incubatedovernight at 37◦C to dry.5. Use 100 µ l of E.coli culture (OD 600 =0.01) for Petri dish agar mediumand evenly spread the bacteria with sterile, L- shaped glass rod on the sur-face of agar plate (For Omni trays use the manual stamp for transferringthe bacteria).6. After incubation at 37◦C, check the plates in different time for growthobservation. Record the time for growth observation.G.3 MIC Test/Fosmids Growth RateDetermination in L.B BrothProcedure1. Prepare desire amount of L.B broth and adjust pH at 7.2.2. Prepare arsenic stock solution, arabinose and chloramphenicol in advanceand sterilize with 0.22 µm filter. Arabinose and chloramphenicol are addedto the medium at the final concentration of 100 µg/ml and 12.5 µg/ml. (Addition of arabinose and chloramphenicol are not applicable for MIC testsin L.B broth ).4. Distribute media to 96/384 micro well plates. ( Q-fill machine can beused for this purpose).5. Inoculate the plates with overnight E.coli (host) culture / Fosmids frommetagenomic libraries, at OD600 ≥0.4 . (Automated approach, QSoft-XPreplicating program).6. Incubate the plates at 37◦C, read at OD600 with Microplate reader atrequired time.202G.4. Analytical MethodsG.4 Analytical Methods203G.4. Analytical Methods Revision 01.00 March 16, 2012 Geochemical Procedure ME-4ACD81 Trace Level Methods Using Conventional ICP-AES Analysis Sample Decomposition: HNO3-HClO4-HF-HCl digestion, HCl Leach (GEO-4ACID) Analytical Method: Inductively Coupled Plasma - Atomic Emission Spectroscopy (ICP - AES) Base metals can be reported with the ME-MS81 by a four acid digestion. The four acid digestion is preferred when the targets include more resistive mineralization such as that associated with nickel and cobalt. A prepared sample (0.25 g) is digested with perchloric, nitric, hydrofluoric and hydrochloric acids. The residue is topped up with dilute hydrochloric acid and the resulting solution is analyzed by inductively coupled plasma-atomic emission spectrometry. Results are corrected for spectral inter-element interferences. NOTE: Four acid digestions are able to dissolve most minerals; however, although the term “near-total” is used, depending on the sample matrix, not all elements are quantitatively extracted. Element Symbol Units Lower Limit Upper Limit Default Overlimit Method Silver Ag ppm 0.5 100 Ag-OG62 Arsenic As ppm 5 10000 Cadmium Cd ppm 0.5 500 Cobalt Co ppm 1 10000 Co-OG62 Copper Cu ppm 1 10000 Cu-OG62 Molybdenum Mo ppm 1 10000 Mo-OG62 Nickel Ni ppm 1 10000 Ni-OG62 Lead Pb ppm 2 10000 Pb-OG62 Scandium Sc ppm 1 10000 Zinc Zn ppm 2 10000 Zn-OG62 Figure G.1: Analytical Method 1, four acid digestion (HNO3-HClO4-HF-HCl) followed by inductively coupled plasma-atomic emission spectroscopy(ICP-AES). Reproduced with permission of the ALS Minerals Division an-alytical laboratory. 204G.4. Analytical Methods Revision 05.00 Mar 06, 2006 Whole Rock Geochemistry ME- ICP06 and OA- GRA05 Analysis of major oxides by ICP- AES ME- ICP06 Sample Decomposition: Lithium Metaborate/Lithium Tetraborate (LiBO2/Li2B4O7) Fusion* (FUS-LI01) Analytical Method: Inductively Coupled Plasma - Atomic Emission Spectroscopy (ICP-AES) A prepared sample (0.200 g) is added to lithium metaborate/lithium tetraborate flux (0.90 g), mixed well and fused in a furnace at 1000°C. The resulting melt is then cooled and dissolved in 100 mL of 4% nitric acid/2% hydrochloric acid. This solution is then analyzed by ICP-AES and the results are corrected for spectral inter-element interferences. Oxide concentration is calculated from the determined elemental concentration and the result is reported in that format. Element Symbol Units Lower Limit Upper Limit Aluminum Al2O3 % 0.01 100 Barium BaO % 0.01 100 Calcium CaO % 0.01 100 Chromium Cr2O3 % 0.01 100 Iron Fe2O3 % 0.01 100 Magnesium MgO % 0.01 100 Manganese MnO % 0.01 100 Phosphorus P2O5 % 0.01 100 Potassium K2O % 0.01 100 Silicon SiO2 % 0.01 100 Sodium Na2O % 0.01 100 Figure G.2: Analytical Method 2, lithium metaborate/lithium tetrabo-rate (LiBO2/Li2B4O7) fusion and analyzed by inductively coupled plasma-atomic emission spectroscopy (ICP-AES). Reproduced with permission ofthe ALS Minerals Division analytical laboratory. 205G.4. Analytical Methods Revision 02.00 Nov 23, 2006 Geochemical Procedure ME- MS42 Single Element Trace Level Methods Using ICP- MS Sample Decomposition: Aqua Regia Digestion (GEO-AR01) Analytical Method: Inductively Coupled Plasma - Mass Spectrometry (ICP-MS) This method is an effective option when analytical results for one or only a few elements are required. With this method you can create your own package of elements specific to your exploration program. A prepared sample is digested with aqua regia for in a graphite heating block. After cooling, the resulting solution is diluted to 12.5 mL with deionized water, mixed and analyzed by inductively coupled plasma-atomic emission spectrometry. The analytical results are corrected for inter-element spectral interferences. NOTE: In the majority of geological matrices, data reported from an aqua regia leach should be considered as representing only the leachable portion of the particular analyte. Element Symbol Units Lower Limit Upper Limit Silver Ag ppm 0.01 25 Arsenic As ppm 0.1 250 Bismuth Bi ppm 0.01 250 Mercury Hg ppm 0.005 25 Molybdenum Mo ppm 0.05 250 Antimony Sb ppm 0.05 250 Selenium Se ppm 0.2 250 Tellurium Te ppm 0.01 250 Thallium Tl ppm 0.02 250 Uranium U ppm 0.05 250 Figure G.3: Analytical Method 3, Aqua regia digestion plus inductively-coupled plasma-mass spectroscopy (ICP-MS) analysis. Reproduced withpermission of the ALS Minerals Division analytical laboratory.206G.4. Analytical MethodsS-IR08.pdf Revision 02.01 Jan 12, 2004 Specialty Assay Procedure C- IR07 & S- IR08 Evaluation of Ores and High Grade Materials Sample Decomposition: Leco Furnace Analytical Method: Infrared Spectroscopy The sample is analyzed for total sulfur and/or carbon using a Leco analyzer. While a stream of oxygen passes through a prepared sample (0.05 to 0.6g), it is heated in a furnace to approximately 1350°C. Sulfur dioxide and carbon dioxide released from the sample are measured by an infrared detection system and the total sulfur and/or carbon result is provided. ALS Chemex Method Code Element Detection Symbol Upper Limit Units Limit C-IR07 Carbon C 0.01 100 % S-IR08 Sulfur S 0.01 50 % S-IR08t Sulfur S 0.01 100 % Figure G.4: total sulphur and carbon analysis, Leco Furnace and measure-ment with infrared spectroscopy. Reproduced with permission of the ALSMinerals Division analytical laboratory. 207G.4. Analytical Methods Revision 06.02 March 16, 2012 Geochemical Procedure ME-MS81 Ultra-Trace Level Methods Sample Decomposition: Lithium borate Fusion (FUS -LI01) Analytical Method: Inductively Coupled Plasma - Mass Spectroscopy (ICP - MS) A prepared sample (0.200 g) is added to lithium borate flux (0.90 g), mixed well a nd fused in a furnace at 1000°C. The resulting melt is then cooled and dissolved in 100 mL of 4% HNO 3 / 2% HCl solution. This solution is then analyzed by inductively coupled plasma - mass spectrometry. Element Symbol Units Lower Limit Upper Limit Barium Ba ppm 0.5 10000 Cerium Ce ppm 0.5 10000 Cobalt* Co ppm 0.5 10000 Chromium Cr ppm 10 10000 Cesium Cs ppm 0.01 10000 Dysprosium Dy ppm 0.05 1000 Erbium Er ppm 0.03 1000 Europium Eu ppm 0.03 1000 Gallium Ga ppm 0.1 1000 Gadolinium Gd ppm 0.05 1000 Hafnium Hf ppm 0.2 10000 Holmium Ho ppm 0.01 1000 Lanthanum La ppm 0.5 10000 Lutetium Lu ppm 0.01 1000 Figure G.5: Lithium borate Fusion, Inductively Coupled Plasma - MassSpectroscopy (ICP - MS). Reproduced with permission of the ALS MineralsDivision analytical laboratory.208"""@en ; edm:hasType "Thesis/Dissertation"@en ; vivo:dateIssued "2014-11"@en ; edm:isShownAt "10.14288/1.0135561"@en ; dcterms:language "eng"@en ; ns0:degreeDiscipline "Chemical and Biological Engineering"@en ; edm:provider "Vancouver : University of British Columbia Library"@en ; dcterms:publisher "University of British Columbia"@en ; dcterms:rights "Attribution-NonCommercial-NoDerivs 2.5 Canada"@en ; ns0:rightsURI "http://creativecommons.org/licenses/by-nc-nd/2.5/ca/"@en ; ns0:scholarLevel "Graduate"@en ; dcterms:title "Microbes involved in arsenic removal in passive treatment systems"@en ; dcterms:type "Text"@en ; ns0:identifierURI "http://hdl.handle.net/2429/50430"@en .