DISSERTATION O Attribution

o Attribution — You must give appropriate credit, provide a link to the license, and indicate if changes were made. You may do so in any reasonable manner, but not in any way that suggests the licensor endorses you or your use.

o NonCommercial — You may not use the material for commercial purposes.

o ShareAlike — If you remix, transform, or build upon the material, you must distribute your contributions under the same license as the original.

How to cite this thesis

Surname, Initial(s). (2012) Title of the thesis or dissertation. PhD. (Chemistry)/ M.Sc. (Physics)/ M.A. (Philosophy)/M.Com. (Finance) etc. [Unpublished]: University of Johannesburg. Retrieved from: https://ujcontent.uj.ac.za/vital/access/manager/Index?site_name=Research%20Output (Accessed: Date).

DIVERSITY AND BIOTECHNOLOGY

APPLICATIONS OF THERMOPHILIC BACTERIA

FROM HOT-SPRING WATER IN LIMPOPO

SOUTH AFRICA RELATING TO WASTEWATER

BIOREMEDIATION AND WATER SAFETY

Jocelyn Leonie Jardine

Thesis for the degree of Doctor of Technology

Department of Biotechnology

Faculty of Science

University of Johannesburg, South Africa

October 2017

Supervisor: Dr Eunice Ubomba-Jaswa, PhD

Co-supervisor: Dr Vuyo Mavumengwana, PhD

ABSTRACT

Hot-spring environments are valuable resources for novel and useful bacteria in biotechnology. In South Africa, hot springs in Limpopo Province have been investigated by metagenomics but not by culture-based methods. This study was performed in order to describe the diversity of cultured bacteria from hot springs, and to comment on their relevance to public health and antibiotic resistance. Furthermore, potential bacterial candidates were screened for characteristics associated with wastewater (WW) bioremediation and applications.

Cultured bacteria were identified by a combination of tools using the 16S rDNA sequence by comparison with public databases, percent guanine-cytosine content, amplified rDNA restriction analysis (ARDRA) and phylogeny. Resistance against ten antibiotics (carbenicillin, gentamicin, kanamycin, streptomycin, tetracycline, chloramphenicol, ceftriaxone, co-trimoxazole, nalidixic acid and norfloxacin) and eight heavy metal ions (Al, Cr, Cu, Fe, Hg, Mn, Ni and Pb) were tested using disk diffusion assays. Isolates were screened for production of enzymes and biophysical characteristics (bioflocculant and biosurfactant activities, biosorption, bioassimilation, anti-biofilm and antimicrobial properties) and selected cell free culture supernatants (CFCS) were processed by tandem LC-MS for identification of the bioactive molecules. CFCS were exposed to pollutants (pigmented food, textile dyes, polyaromatic hydrocarbons) and food WW samples to establish whether there was a reduction in turbidity or phenol levels.

Forty-three (43) Gram-positive isolates were of the phylum Firmicutes with the majority of the genus Bacillus (n = 31). Different species were identified as Anoxybacillus flavithermus, Anoxybacillus rupiensis, Bacillus subtilis, Bacillus licheniformis and Brevibacillus spp. Single isolates of Gram-positive Kocuria sp. and Arthrobacter sp., and Gram-negative Cupriavidus sp., Ralstonia sp., Cronobacter sp., Tepidimonas sp., Hafnia sp. and Sphingomonas sp. were identified, previously reported to be emerging opportunistic pathogens, and an absence of Legionella sp. was reported. Low levels of antibiotic resistance (AR) were reported. Only 2.5% (n = 40) of the isolates were multiple antibiotic resistant against >3 antibiotics, while 37.5% were found for both resistance against one or two antibiotics. Resistance was found against ceftriaxone (52.5%), nalidixic acid (37.5%) and carbenicillin (22.5%). All 29 isolates tested heavy metals were tolerant to ≥2 heavy-metal salts. No association was observed between antibiotic resistance and heavy-metal tolerance.

Amylase and protease positive isolates, and isolates that could discolorize bromothymol blue were detected by conventional agar plate assays. The CFCS of A. rupiensis 19S was selected for

tandem LC-MS analysis and potential bioremediation enzymes were identified (amylase, proteases, catalase peroxidase, superoxide dismutase, azoreductase, quinone oxidoreductase, ribonucleases and inorganic pyrophosphatase). Potential dehydrogenase enzymes for biomonitoring of environmental pollutants were also reported. Bioaccumulation of triphenylmethane dye (bromothymol blue) was a property of B. subtilis (9T), while B. subtilis (20S) was able to bioflocculate kaolin clay reducing turbidity by 30%. Biosorption resulted in reductions of Cr and Cu using CFCS of four isolates (7T, 9T, 30M and 83Li). Four different isolates (16S, 71T, 76S, 85Li), were positive for biosurfactant properties by emulsion activity of paraffin oil, sunflower seed oil and petroleum, and drop collapse assay. Brevibacillus sp. (16S) was able to inhibit biofilm formation of B. subtilis. The CFCS fraction of Bacillus mojavensis (biosurfactant positive) 76S inhibited growth of 76% of an AR environmental bacterial panel as well as human Gram-positive pathogens, but not Gram-negative pathogens and acid-fast Mycobacterium smegmatis. Tandem LC-MS identified the biosurfactant, subtilisin, and possible responsible biomolecules with antimicrobial activity of isolate 76S as proteases, subtilisin BM1 and bacillolysin.

A reduction in turbidity of dairy and brewery WW was observed when exposed to CFCS fractions of these isolates. Anoxybacillus sp. 19S and Brevibacillus sp. 84Li were found to be the most effective in reducing toxic phenol by at least half in phenol red broth media, food WW samples, in river water contaminated with industrial effluents, and poly aromatic hydrocarbons. In the simulated pollutant experiment, phenols were reduced by more than 54% in coffee, soya sauce, and a commercial textile dye.

This is the first report describing cultured aerobic bacteria from hot springs in South Africa. Identification using a combination of molecular tools was useful in discerning differences between the Bacillus and Bacillus-related bacteria. Emerging opportunistic pathogens were isolated and identified, having implications for water safety and public health. Low levels of AR could be useful as a baseline measure of intrinsic environmental AR. Isolates were identified that produced relevant enzymes, and with biophysical properties useful for WW bioremediation. Exposure of CFCS to pollutants and WW samples reduced the toxic phenol concentrations, substantiating the belief that South African hot springs are a valuable resource of potentially important bacteria that can be used for several applications in WW bioremediation.

iii

DECLARATION

I, Jocelyn Leonie Jardine, do hereby declare that this thesis is my own, unaided work. The thesis is presented in fulfilment of the requirement for the degree of D Tech in Biotechnology, Faculty of Science, University of Johannesburg, and it has not been submitted before any degree or examination in any other Technikon or University.

Signature:

Date: 23/11/2017

DEDICATION

To my parents with all my Love, Leo Paul Jardine and Sue Phyllis Jardine (1936-2017)

ACKNOWLEDGEMENTS

I wish to acknowledge and thank the following people for their contribution to this research.

Family (especially my mother who did not live long enough to see this completed) and friends who gave me constant encouragement, support and help.

Supervisor Dr Vuyo Mavumengwana who started the ball rolling, Dr Eunice Ubomba-Jaswa who kept it going, and for submission of manuscripts to journals.

From the department of Biotechnology UJ, Ms Malie King (retired departmental secretary) and Mr Erick Van Zyl (retired Head of Department).

For assistance with statistics : Dr Gill Hendry; RT-PCR : Dr King Abia; LC-MS/MS : Dr Stoyan Stoychev at CSIR Biosciences; DNA sequencing : Prof Michelle van der Bank and Ronny Kabongo at the African Centre for DNA Barcoding, UJ, SA.

University of Johannesburg for student funding in the form of Global Excellence Scholarship over a period of 3 years.

Conference travel bursaries awarded from South African society for Microbiology (SASM) 2016 and Centre for Microbial Ecology and Genomics (CMEG), University of Pretoria and Thermophiles 2017 conference committees. Sponsorship from Dr Patrik Njobeh of Food Technology, UJ, for attendance to Food Security and Safety (SFAS) 2016 conference.

Owners and managers of the hot spring resorts : Tshipise, Siloam, Mphephu, Lekkerrus and Libertas. Assistance from Dr Sudarshan Sekar and Dr Mavumengwana in field sample collection.

Dr Irwin Juckes for industrial wastewater collection from a stormdrain in Edenvale, South Africa, Mr Andrew Edwards at Douglasdale dairy for wastewater samples and Mr Vincent Le Roux of the Brew Keg, Kyalami for supply of brewing wastewater and sample of amylase.

Colleagues and staff in the Dept of Biotechnology and Food Technology, UJ; in the Dept Physics, UJ; and staff at Doornfontein and Auckland Park Libraries, Post Grad Centre UJ, Faculty of Science Offices, UJ and staff at Natural Resource Environments, CSIR Pretoria.

PUBLICATIONS

Jocelyn L. Jardine, Akebe Luther King Abia, Vuyo Mavumengwana, Eunice Ubomba-Jaswa. (2017). Phylogenetic analysis and antimicrobial profiles of cultured emerging opportunistic pathogens (phyla Actinobacteria and Proteobacteria) identified in hot springs. International Journal of Environmental Research and Public Health 14, 1070; doi10.3390/ijerph14091070.

PUBLICATIONS SUBMITTED

1. Identification of Bacillus and closely related Bacillus species from South African hot springs using culture-based and phylogenetic analysis. Jocelyn Jardine; Vuyo Mavumengwana; Eunice Ubomba-Jaswa.

2. Screening of potential bioremediation enzymes from hot spring bacteria using conventional plate assays and Liquid Chromatography –tandem Mass Spectrophotometry (LC- MS/MS). Jocelyn L Jardine, Stoyan Stoychev, Vuyo Mavumengwana, Eunice Ubomba-Jaswa.

3. Antibiotic and Heavy Metal Resistance of Bacillus species and opportunistic pathogens cultured from hot springs in South Africa. Jocelyn Jardine, Vuyo Mavumengwana, Eunice Ubomba-Jaswa.

4. Absorption of heavy metal ions, emulsion of polyaromatic hydrocarbon compounds and anti- biofilm properties of hot springs bacterial isolates has potential in removal of hazardous water pollutants. Jocelyn Jardine, VuyoMavumengwana, Stoyan Stoychev, Eunice Ubomba-Jaswa.

vii

PRESENTATIONS AT CONFERENCES

1 Jardine J, Mavumengwana V. and Ubomba-Jaswa E. 17 Jan – 20 Jan 2016. Isolation and identification of Bacillus and Bacillus-related bacteria from hot springs Limpopo, South Africa. Oral presentation at the 19th South African Society for Microbiology Conference, Umhlanga, South Africa S12.05. TRAVEL AWARD FOR ABSTRACT

2 Jardine J, Mavumengwana V. and Ubomba-Jaswa E. 16-18 May 2016. Antibiotic and heavy metal resistance profiles of potential food-borne and probiotic bacteria isolated from hot springs, Limpopo, South Africa. Oral presentation at the 2016 Autumn International Scientific Conference on Food Security and Safety. Johannesburg, South Africa. FIRST PRIZE FOR ORAL PRESENTATION.

3 Jardine J, Mavumengwana V., Abia AK. and Ubomba-Jaswa E. 25-28 September 2016. The potential of hot spring water as a reservoir for emerging opportunistic pathogens in South Africa. 6th Infection Control Africa Network Congress 2016. Johannesburg, South Africa. Poster Presentation

4. Jardine J, Mavumengwana V., Stoychev S and Ubomba-Jaswa E. 27 Aug -1 Sept 2017. Screening of potential bioremediation enzymes from hot springs bacteria using conventional plate assays and LC-MS. 14th International conference for Thermophiles 2017. Skukuza, South Africa. TRAVEL AWARD FOR ABSTRACT

5 Jardine J, Mavumengwana V., Stoychev S and Ubomba-Jaswa E. 4-7 April 2018. Antimicrobial activity of Bacillus mojavensis, isolated from hot springs, South Africa, against environmental antibiotic resistant bacteria and human pathogens. 20th South African Society for Microbiology Conference, Johannesburg, South Africa. Poster Presentation (accepted 30th February 2018)

viii

TABLE OF CONTENTS

ABSTRACT ...... ii DECLARATION ...... iv DEDICATION ...... v ACKNOWLEDGEMENTS ...... vi PUBLICATIONS ...... vii PUBLICATIONS SUBMITTED ...... vii PRESENTATIONS AT CONFERENCES ...... viii TABLE OF CONTENTS ...... ix LIST OF FIGURES ...... xvii LIST OF TABLES ...... xxi ABBREVIATIONS ...... xxiii CHAPTER ONE: INTRODUCTION ...... 1

1.1 PROBLEM STATEMENT ...... 1 1.2 AIMS AND OBJECTIVES OF THE STUDY ...... 4 1.3 HYPOTHESIS ...... 5

CHAPTER 2: LITERATURE REVIEW ...... 6

2.1 INTRODUCTION ...... 6 2.2 MICROBIAL DIVERSITY OF HOT SPRINGS ...... 6 2.3 PATHOGENS AND OPPORTUNISTIC PATHOGENS ...... 9 2.4 ANTIBIOTIC RESISTANCE OF PRISTINE SITES AND HOT SPRINGS ...... 13 2.5 BIOREMEDIATION OF WASTEWATER ...... 16 2.6 MECHANISMS FOR BIOREMEDIATION ...... 21 2.6.1 Bacterial enzyme activity...... 21 2.6.2 Biophysical characteristics ...... 24 2.6.3 Naturally occurring antimicrobials ...... 28

CHAPTER THREE: MATERIALS AND METHODS ...... 32

3.1 SAMPLING AND SAMPLING SITES ...... 32

3.1.1 Sampling from hot springs in Limpopo Province, SA ...... 32 3.1.2 Sampling of brewery and dairy wastewaters and river water contaminated with coloured industrial effluents in Gauteng Province, SA ...... 34

3.2 ISOLATION OF BACTERIA FROM WATER AND SEDIMENT ...... 37

3.3 DETERMINATION OF OPTIMAL TEMPERATURE, pH AND SALINITY FOR GROWTH OF THERMOPHILES...... 37 3.4 DNA EXTRACTION PROTOCOL AND 16S rDNA SEQUENCING ...... 38 3.5 GENBANK ACCESSION NUMBERS ...... 39

3.5.1 Phylum Firmicutes ...... 39 3.5.2 Phyla Actinobacteria and Proteobacteria ...... 39

3.6 PHYLOGENETIC ANALYSIS ...... 40 3.7 COMPUTER-SIMULATED PCR-RFLP OR AMPLIFIED rDNA RESTRICTION ANALYSIS (ARDRA) ...... 40 3.8 GUANINE-CYTOSINE (GC) CONTENT ...... 41 3.9 DETECTION OF LEGIONELLA BY REAL-TIME POLYMERASE CHAIN REACTION ...... 41 3.10 ANTIBIOTIC RESISTANCE/SUSCEPTIBILITY ASSAY ...... 42 3.11 HEAVY-METAL TOLERANCE ASSAY ...... 43 3.12 DETECTION OF BACTERIAL ENZYMES ...... 43

3.12.1 Plate assay for screening of potential enzymes (amylase, protease, lipase, pectinase, gelatinase, azoreductase and laccase) for bioremediation ...... 43 3.12.2 Concentration of amylase produced ...... 44 3.12.3 Gravimetric assay for the detection of cellulase ...... 44 3.12.4 Biochemical tube assay for the detection of laccase/peroxidase ...... 44 3.12.5 Biochemical tube assay for phenol reduction ...... 44

3.13 DETECTION OF BIOPHYSICAL CHARACTERISTICS: BIOASSIMILATION OF TRIPHENYLMETHANE DYE (BROMOTHYMOL BLUE) ...... 45 3.14 DETECTION OF BIOPHYSICAL CHARACTERISTICS: BIOSORPTION OF HEAVY METALS ...... 46 3.15 DETECTION OF BIOPHYSICAL CHARACTERISTICS: BIOFLOCCULANT ACTIVITY USING THE KAOLIN CLAY ASSAY...... 47 3.16 DETECTION OF BIOPHYSICAL CHARACTERISTICS: BIOSURFACTANT ACTIVITY ...... 47

3.16.1 Emulsion activity assay ...... 48

3.16.2 Drop-collapse assay using Parafilm M ...... 48

3.17 DETECTION OF BIOPHYSICAL CHARACTERISTICS: ANTI-BIOFILM ACTIVITY ...... 48

3.18 ANTIMICROBIAL ACTIVITY SCREEN ...... 49

3.18.1 Streak technique to identify antimicrobial activity ...... 49 3.18.2 Antimicrobial activity testing of reference human pathogens and antibiotic resistant environmental isolates crude extracts ...... 49

3.19 IDENTIFICATION OF PROTEIN BANDS BY LIQUID CHROMATOGRAPHY – TANDEM MASS SPECTROPHOTOMETRY (LC-MS/MS) ...... 50 3.20 REDUCTION OF COLOURATION OR TURBIDITY IN WASTEWATER SAMPLES ...... 51

3.20.1 Effect of temperature on wastewater colour or turbidity reduction ...... 52

3.21 STATISTICAL ANALYSIS...... 52

3.21.1 Determination of relationship between antibiotic resistance and heavy-metal resistance ...... 53 3.21.2 Determination of significance of difference between negative controls and samples ...... 53

CHAPTER FOUR: ISOLATION AND IDENTIFICATION OF BACILLUS AND CLOSELY RELATED BACILLUS BACTERIA ISOLATED FROM HOT SPRINGS IN LIMPOPO PROVINCE, SOUTH AFRICA ...... 54

4.1 INTRODUCTION ...... 54 4.2 METHODOLOGY ...... 54 4.3 RESULTS ...... 55

4.3.1 Bacterial isolation ...... 55

4.3.1.1 Concentration of mesophiles and thermophiles in hot-spring water ...... 55 4.3.1.2 Influence of isolation media ...... 56

4.3.2 Optimal growth conditions for bacterial isolation and growth ...... 56 4.3.3 16S rDNA sequencing ...... 57 4.3.4 Guanine-cytosine (GC) content (in percentage) ...... 60 4.3.5 Computer-simulated amplified ribosomal DNA restriction analysis (ARDRA) ...... 60 4.3.6 Phylogenetic analysis ...... 61

4.3.6.1 Family Bacillaceae genus Anoxybacillus ...... 62 4.3.6.2 Family Bacillaceae genus Bacillus ...... 63

4.3.7 Analysis of unknown isolates ...... 65

4.4 DISCUSSION ...... 65

4.4.1 Bacterial isolation ...... 65 4.4.1.1 Concentration of mesophiles and thermophiles in hot-spring water ...... 65

4.4.1.2 Influence of isolation media ...... 66 4.4.1.3 Optimal growth conditions for bacterial isolation and growth ...... 66

4.4.2 16S rDNA sequencing ...... 67 4.4.3 Guanine-cytosine (GC) content (in percentage) ...... 68 4.4.4 Computer-simulated amplified ribosomal DNA restriction analysis (ARDRA) ...... 69 4.4.5 Phylogenetic analysis ...... 70 4.4.5.1 Family Bacillaceae genus Anoxybacillus ...... 71 4.4.5.2 Family Bacillaceae genus Bacillus ...... 71 4.4.5.3 Family Paenibacillaceae genus Brevibacillus ...... 73 4.4.6 Analysis of unknown isolates ...... 73

CHAPTER FIVE: CULTURED EMERGING OPPORTUNISTIC PATHOGENS (PHYLA ACTINOBACTERIA AND PROTEOBACTERIA) IDENTIFIED IN HOT-SPRING WATER THROUGH ISOLATION AND PHYLOGENETIC ANALYSIS ...... 76

5.1 INTRODUCTION ...... 76 5.2 METHODOLOGY ...... 76 5.3 RESULTS ...... 77

5.3.1 Isolation of bacteria ...... 77 5.3.2 16S rDNA amplicon sequencing ...... 78 5.3.3 Phylogenetic analysis ...... 80 5.3.4. Legionella real-time polymerase chain reaction (RT-PCR) ...... 83 5.3.5 Antibiotic resistance ...... 84

5.4 DISCUSSION ...... 85

5.4.1 Phylum Actinobacteria ...... 85 5.4.2 Phylum Proteobacteria ...... 86

5.4.2.1 Alpha-Proteobacteria ...... 86 5.4.2.2 Beta-Proteobacteria ...... 86 5.4.2.3 Gamma-Proteobacteria ...... 87

5.4.3 Antibiotic resistance of opportunistic emerging pathogens ...... 88

xii

5.4.4 Relationship between waterborne pathogens & opportunistic emerging pathogens .889 5.4.5 Significance in water-scarce developing countries ...... 90

CHAPTER SIX: ANTIBIOTIC AND HEAVY-METAL TOLERANCE IN CULTURED BACILLUS SPECIES AND OPPORTUNISTIC PATHOGENS (KOCURIA SPECIES AND HAFNIA ALVEI) FROM HOT SPRINGS IN LIMPOPO PROVINCE, SOUTH AFRICA ...... 92

6.1 INTRODUCTION ...... 92 6.2. METHODOLOGY ...... 92 6.3. RESULTS AND DISCUSSION ...... 93

6.3.1 Isolation and identification of bacteria ...... 93 6.3.2 Antibiotic resistance ...... 95 6.3.3 Antibiotic resistance in the environment...... 96

6.3.3.1 Hot springs and pristine environments ...... 96 6.3.3.2 Antibiotic resistance of isolates from hot springs as a baseline measure of natural environmental AR without acquired AR due to human activities ...... 98 6.3.4 Antibiotic resistance of hot-spring bacteria ...... 101 6.3.4.1 Bacillus species ...... 101 6.3.4.2 Opportunistic pathogens ...... 102

6.3.5 Heavy-metal tolerance patterns in isolates from hot springs ...... 102 6.3.6 Correlation between heavy-metal resistance and antibiotic resistance ...... 104

CHAPTER SEVEN: SCREENING OF POTENTIAL BIOREMEDIATION ENZYMES FROM HOT-SPRING BACTERIA USING CONVENTIONAL PLATE ASSAYS AND LIQUID CHROMATOGRAPHY-TANDEM MASS SPECTROPHOTOMETRY (LC-MS/MS) ...... 108

7.1 INTRODUCTION ...... 108 7.2 METHODOLOGY ...... 109 7.3 RESULTS ...... 110

7.3.1 Identification of bacteria ...... 110 7.3.2 Plate assays ...... 110 7.3.3 Quantitative amylase assays ...... 113 7.3.4 Gravimetric assay for the detection of cellulase ...... 114 7.3.5 Biochemical tube assay for the detection of laccase and peroxidase ...... 114 7.3.6 Identification of proteins by LC-MSMS ...... 115

7.4 DISCUSSION ...... 117

xiii

7.4.1 Amylase ...... 118 7.4.2 Proteases/peptidases ...... 119 7.4.3 Cellulases ...... 119 7.4.4 Phosphatases ...... 120 7.4.5 Ribonuclease ...... 120 7.4.6 Dihydrolipoyl dehydrogenase ...... 121 7.4.7 Oxidoreductases ...... 121 7.4.8 Removal of lead and chromate ...... 123 7.4.9 Biomonitoring ...... 123 7.4.10 Comparison of assays...... 124

CHAPTER EIGHT: BIOPHYSICAL CHARACTERISTICS (BIOFLOCCULANTS, BIOSURFACTANTS, BIOSORPTION AND ANTI-BIOFILM) OF BACTERIA ISOLATED FROM HOT SPRINGS ...... 125

8.1 INTRODUCTION ...... 125 8.2 METHODOLOGY ...... 126 8.3 RESULTS AND DISCUSSION ...... 126

8.3.1 Bioassimiliation ...... 126 8.3.2 Biosorption ...... 130 8.3.3. Bioflocculant activity ...... 133 8.3.4 Biosurfactant assay ...... 135

8.3.4.1 Emulsion assay ...... 135 8.3.4.2 Drop-collapse assay for biosurfactant on Parafilm M ...... 136 8.3.4.3 Emulsion activity of selected isolates using petroleum and sunflower seed as substrate ...... 137 8.3.4.4 Identification of molecules with biosurfactant potential by LC-MS/MS ...... 138

8.3.5 Anti-biofilm activity ...... 140 8.3.6 16S rDNA identification of important isolates ...... 142

CHAPTER NINE: POTENTIAL OF HOT-SPRING WATER ISOLATES FROM SOUTH AFRICA IN WASTEWATER BIOREMEDIATION ...... 143

9.1 INTRODUCTION ...... 143 9.2 METHODOLOGY ...... 144 9.3 RESULTS ...... 145

9.3.1 Reduction of colouration or turbidity ...... 145

xiv

9.3.1.1 Reduction of colouration due to food pollutants (coffee and soya sauce) ...... 146 9.3.1.2 Reduction of colouration due to derivative of textile dyes (bromothymol blue and commercial textile dye Dye It No. 18) ...... 146 9.3.1.3 Reduction of colouration/turbidity of brewery and dairy wastewaters and river water contaminated with coloured industrial effluents ...... 147

9.3.2 Phenol reduction in phenol red broth media by CFCS extracts of isolates ...... 148 9.3.3 Reduction of phenol in pollutants and wastewater samples ...... 149

9.3.3.1 Reduction of phenol in coffee and soya sauce ...... 149 9.3.3.2 Reduction of phenol in bromothymol blue, crystal violet & commercial dye..150 9.3.3.3 Reduction of phenol in brewery and dairy wastewaters and river water contaminated with industrial effluents ...... 151 9.3.3.4 Reduction of phenol in paraffin oil and petroleum ...... 152 9.3.3.5 Effect of temperature on reduction of phenols ...... 152

9.4 DISCUSSION ...... 154

9.4.1 Reduction in colouration or turbidity ...... 154 9.4.2 Screening of isolates that reduced phenol in phenol red broth media ...... 156 9.4.3 Reduction in phenol and phenolics in coffee, soya sauce & brewery wastewater ....157 9.4.4 Reduction of phenol and phenolics in BB, CV, commercial dye and river water contaminated with industrial effluents ...... 158 9.4.5 Reduction in phenol and phenolics in dairy wastewater ...... 159 9.4.6 Reduction in phenol and phenolics in paraffin oil and petroleum ...... 159 9.4.7 Effect of temperature on reduction of phenol in river water contaminated with industrial effluents ...... 159 9.4.8 Identification of isolates with bioremediation potential ...... 160 9.4.9 Properties of isolates with bioremediation potential ...... 160

CHAPTER TEN: ANTIMICROBIAL ACTIVITY OF BACILLUS MOJAVENSIS, ISOLATED FROM HOT SPRINGS, SOUTH AFRICA, AGAINST ENVIRONMENTAL MULTIPLE ANTIBIOTIC-RESISTANT BACTERIA AND HUMAN PATHOGENS...... 162

10.1 INTRODUCTION ...... 162 10.2 METHODOLOGY ...... 163 10.3 RESULTS ...... 164

10.3.1 Biosurfactant activity: emulsion activity assay and drop-collapse assay using Parafilm M ...... 164 10.3.2 Antimicrobial activity ...... 165 10.3.2.1 T-streak technique to identify isolates with antimicrobial activity ...... 165 10.3.2.2 Antimicrobial activity testing of crude CFCS extracts against the panel of environmental isolates and human pathogens ...... 166 10.3.3 Antibiotic susceptibility testing ...... 168 10.3.4 Screening of proteins in CFCS of isolate 76S by LC-MS/MS ...... 169 10.3.5 Identification of isolate 76S using 16S rDNA sequence ...... 170

10.4 DISCUSSION ...... 171

10.4.1 Biosurfactant production ...... 171 10.4.2 Screening for antimicrobial activity ...... 172

CHAPTER ELEVEN: CONCLUSIONS ...... 176

11.1 Isolation and identification of Bacillus and closely related Bacillus bacteria from hot springs in Limpopo Province, South Africa ...... 176 11.2 Cultured emerging opportunistic pathogens (phyla Actinobacteria and Proteobacteria) identified in hot-spring water through isolation and phylogenetic analysis...... 177 11.3 Antibiotic and heavy-metal tolerance in cultured Bacillus species and opportunistic pathogens (Kocuria species and Hafnia alvei) from hot springs ...... 177

11.4 Screening of potential bioremediation enzymes from hot-spring bacteria using conventional plate assays and liquid chromatography-tandem mass spectrophotometry (LC- MS/MS) ...... 178

11.5 Bacterial isolates cultured from hot springs possess biophysical characteristics (bioflocculant, biosurfactant, biosorption and anti-biofilm activities) useful in water bioremediation ...... 179

11.6 Potential of hot-spring water isolates from South Africa in wastewater bioremediation ...... 180

11.7 Antimicrobial activity of Bacillus mojavensis, isolated from hot springs, South Africa, against environmental multiple antibiotic-resistant bacteria and human pathogens ...... 181

REFERENCES ...... 182

APPENDICES ...... 241

xvi

LIST OF FIGURES

Figure 2.1:World wide bacterial isolations made from hotsprings ...... 11 Figure 2.2: Annotated diagrams of persistent organic pollutants: a) phenol; b) phenol red; c) triphenylmethane dye bromothymol blue; d) triphenylmethane dye crystal violet; and e) melanoidin...... 18 Figure 3.1: Flow chart of summary of methods used in study ...... 32 Figure 3.2: Geographical location of sampling sites in Limpopo Province, SA, indicated on Google map. 1Tshipise; 2Siloam, 3Mphephu, 4Libertas and 5Lekkerrus ...... 33 Figure 3.3: Photographs (copyright J Jardine) of sampling sites A: Siloam; B Tshipise; C: Mphephu; D: Libertas; and E: Lekkerrus ...... 34 Figure 3.4: Sample of beer processing wastewater from The Beer Keg, Johannesburg ...... 35 Figure 3.5: Time course photographs taken on 8 February 2016 (photos copyright, permission from Irwin Juckes) showing a stormwater drain delivering regular input of coloured pollutant at a single site in the Modderfontein River ...... 36 Figure 3.6: Sampling site for industrial coloured wastewater entering the river from a stormwater drain in Modderfontein, Johannesburg ...... 36 Figure 3.7: Sampling site at dairy wastewater outlet of Douglasdale Dairy, Johannesburg ...... 37 Figure 4.1: Concentrations (CFU/400 mL) of mesophiles (isolated at 37 °C) and thermophiles (isolated at 53 °C) from five sampled hot springs ...... 55 Figure 4.2: Colony-forming units (CFU) per 100 mL of water showing the difference between NA and minimal Luria agar for mesophiles incubated at 37 °C, and thermophiles at 53 °C at three different hot-spring locations ...... 56 Figure 4.3: Determination of optimal conditions for growth of bacterial isolates: Optimal pH was 7 (Figure 4.3A), optimal temperature was 55 °C (Figure 4.3B) and optimal salinity was 5% NaCl (Figure 4.3C). Three of the 16 isolates tested for salinity could also grow at NaCl...... 57 Figure 4.4: Average GC content (%) of 16S rDNA sequence of isolates from this study and reference type strains of genera Aneurinibacillus and Brevibacillus (family Paenibacillaceae) and genera Anoxybacillus and Bacillus (family Bacillaceae) and unclassified genus Solibacillus showing that GC content can be used to distinguish between genera ...... 60 Figure 4.5: Circular neighbour-joining tree of ARDRA binary data of 16S rDNA fragments using HaeIII. Five main clusters (A-E) are indicated in the diagram ...... 61 Figure 4.6: A neighbour-joining phylogenetic tree of a 914 bp fragment of the 16S rDNA sequences between isolates from this study and representative members of type strains of Anoxybacillus, Bacillus, Brevibacillus and Aneurinibacillus. Bootstrap values (%) are based on

xvii

100 replicates and shown for branches with more than 50% bootstrap support. Bar indicates 0.02 substitutions per 100 nucleotide positions ...... 62 Figure 4.7: Neighbour-joining tree of only Bacillus isolates, with representative members of type strains from different Bergey’s groups A-K confirming that most isolates in this study fell into the group A (B. subtilis/B. licheniformis), and that two isolates matched B. pumilus/B. aerophilus (group A) and B. panaciterrae (group K), respectively. Bootstrap values were obtained from 100 replicates and the bar indicated 0.01 substitutions per 100 nucleotide positions ...... 64 Figure 5.1: Comparison of number of isolates (Actinobacteria and Proteobacteria) obtained on different media ...... 78 Figure 5.2: Unrooted parsimony tree for Actinobacteria showing placement of isolate 57T with Kocuria and isolate 58T with Arthrobacter, with bootstrap values ...... 81 Figure 5.3: Rooted parsimony tree for Proteobacteria showing grouping with alpha-, beta- and gamma-Proteobacteria isolates with bootstrap values ...... 82 Figure 5.4A: PCR cycling for the detection of L. pneumophila in water samples; a – g are the dilutions 100 – 10-6 with highest concentration a = 6 368 ng/µL and the lowest concentration g = 0.006368 ng/µL ...... 83 Figure 5.4B: High-resolution melt curve analysis for the identification of L. pneumophila...... 84

Figure 6.1: Antibiotic resistance of 40 hot-spring isolates against 10 antibiotics: carbenicillin (CAR), gentamicin (GEN), kanamycin (KAN), streptomycin (STP), tetracycline (TET), chloramphenicol (CHL), ceftriaxone (CEF), co-trimoxazole (COT), nalidixic acid (NAc) ...... 95 Figure 6.2: Number of isolates from hot springs expressing different MAR index values tested against 10 different antibiotics ranging from 0 to 0.3 ...... 96 Figure 6.3: Heavy-metal tolerance of 29 hot-spring isolates against eight heavy-metal salts: aluminium (Al), chromium (Cr), copper (Cu), iron (Fe), mercury (Hg), manganese (Mn), nickel (Ni) and lead (Pb) in mM (except for Hg which is in nM) ...... 103 Figure 6.4: Hot-spring isolates at different multiple heavy-metal resistance (MHMT) index values ...... 104 Figure 7.1: Summary of methods used in study ...... 109 Figure 7.2A: Amylase-producing colony on starch agar plate; Figure 7.2B: Motile bacterial isolate on starch agar plate; Figure 7.2C: Protease-producing colony on skim milk agar plate; Figure 7.2D: Isolate 84Li on bromothymol blue agar plate agar plate ...... 111 Figure 7.3: Reduction of soluble starch in starch broth by inoculation of thermophilic isolates 9T, 13S and 20S ...... 114

xviii

Figure 7.4A: Peroxidase assay positive control using turnip extract; Figure 7.4B: Peroxidase assay for isolates 16S and 84Li ...... 114 Figure 7.5: Coomassie blue stained sodium dodecyl sulphate polyacrylamide gel electrophoresis of crude extracts of bacterial supernatants ...... 115 Figure 8.1A: Cell pellet of thermophile isolate 9T (LHS) and mesophile isolate 84Li (RHS) after growth in 0.1% bromothymol blue (BB) in nutrient broth for 4 d showing the difference in colouration of the biomass ...... 127 Figure 8.1B: Cell pellets of isolate 9T grown at 37 °C (left) compared with growth at 53 °C (right) in duplicate ...... 127

Figure 8.2: Removal of bromothymol blue (BB) from media by B. subtilis isolate 9T determined by OD at 595 nm ...... 128 Figure 8.3: Biosorption activity of dead cell biomass in aqueous solutions of chromium (Cr), copper (Cu), iron (Fe) and nickel (Ni) ...... 131 Figure 8.4: Bioflocculant activity of isolates from hot springs, South Africa using the kaolin clay assay ...... 134 Figure 8.5: Emulsion activity of positive control (1% SLS in nutrient broth) (tube 1); negative control (nutrient broth only) (tube 2); high emulsion activity (tube 3 – isolate 4T); medium emulsion activity (tube 4 - isolate 7T); and low emulsion activity (tube 5 - isolate 13) using kerosene (paraffin oil) as substrate ...... 135 Figure 8.6: Emulsion activity showing biosurfactant activity of CFCS of isolates from hot springs, SA (green bars) with positive results in the drop-collapse assay (blue crosses) ...... 136 Figure 8.7A: Drop-collapse assay for biosurfactant assay on Parafilm M. Bromothymol blue added to view negative control (left) and positive control (right). Figure 8.7B: Screening of CFCSs on drop-collapse assay ...... 136 Figure 8.8: Anti-biofilm activities of isolates from hot springs in South Africa ...... 140 Figure 8.9: Crystal violet staining of anti-biofilm assay of isolate 16S, 21M, 71T and 75S including positive and negative controls ...... 140 Figure 9.1: Colouration as percentage of negative control of coffee and soya sauce treated with CFCS of hot spring isolates ...... 146 Figure 9.2: Colouration as percentage of negative control of commercial textile dye (Dye It No.18) treated with CFCS of hot spring isolates ...... 147 Figure 9.3: Turbidity as percentage of negative control of brewery wastewater and dairy WW treated with CFCS of hot spring isolates ...... 147

xix

Figure 9.4A: Phenol assay positive green tea (left) and negative water control (right); Figure 9.4B: Phenol assay with bacterial supernatant from isolate 84Li (left) and nutrient broth only (right) mixed with Phenol Red Broth media ...... 148 Figure 9.5: The relative phenol reduction by CFCS of isolates using the phenol as a substrate in Phenol Red Broth media ...... 149 Figure 9.6: Reduction of phenol in coffee and soya sauce ...... 150 Figure 9.7: Reduction of phenol in commercial dyestuff Dye It No. 18...... 150 Figure 9.8: Reduction of phenol concentrations in brewery WW and dairy WW. No brewery WW assays were done for isolates 4T, 7T and 16T ...... 151 Figure 9.9: Reduction of phenol in river water contaminated with industrial effluents (green bar), paraffin oil PF (yellow bar) and petroleum PT (brown bar). (*) denotes isolates with the ability to reduce phenol in contaminated river water. (^) denotes isolates with the ability to reduce phenol in paraffin oil and petroleum ...... 152

Figure 10.1: Emulsion activity of Anoxybacillus 4T against paraffin oil (tube 3), with positive (sodium lauryl sulphate) and negative (nutrient broth only) controls ...... 164 Figure 10.2: Agar plate showing T-streak method where isolate 54T, 77S and 85Li were inoculated perpendicular to isolate 73T. Growth inhibition of isolate 77S against isolate 73T was observed ...... 165 Figure 10.3: Agar disk diffusion assay showing inhibition of crude CFCS extracts from isolate 76S (Bacillus mojavensis) inhibiting the growth of Anoxybacillus sp. 7T ...... 167 Figure 10.4: Antibiotic resistance determined by disk diffusion assay showing resistance to carbenicillin ...... 169 Figure 10.5: Colony morphology of Bacillus mojavensis isolate 76S grown on nutrient agar at 37 °C ...... 170 Figure 10.6: Phylogenetic analysis using parsimony and 100 replicate bootstrapping placing isolate 76S as Bacillus mojavensis ...... 171

LIST OF TABLES

Table 2.1: List of emerging opportunistic pathogens that have been isolated from hot spring water ...... 12 Table 2.2: Summary and comparison of terminology used in bioremediation ...... 27 Table 3.1: Geographical location, site description, pH, temperature and heavy-metal concentrations of samples collected from hot springs in Limpopo Province, SA...... 33 Table 3.2: Location, GPS coordinates and physicochemical parameters of wastewater sampling sites in Gauteng Province, SA ...... 35 Table 3.3: Wavelength (nm) used to measure level of turbidity or colouration of pollutants in water and wastewater ...... 52

Table 4.1: Identification and grouping of isolates from five hot springs in Limpopo, SA where S (Siloam), T (Tshipise), M (Mphephu), Li (Libertas), Le (Lekkerrus) by comparison with 16S rDNA sequences from GenBank (BLAST), EzTaxon-e, GC content, ARDRA and phylogenetic analysis. Two isolation temperatures were used (37 °C and 53 °C) ...... 58 Table 5.1: List of Actinobacteria and Proteobacteria isolates with isolation conditions ...... 77 Table 5.2: Closest match of 16S rDNA sequence from hot-spring isolates with GenBank and EzTaxon-e with percentage similarity and associated accession numbers ...... 79 Table 5.3: Antibiotic resistance of 40 hot-spring isolates against 10 antibiotics: carbenicillin (CAR), gentamicin (GEN), kanamycin (KAN), streptomycin (STP), tetracycline (TET), chloramphenicol (CHL), ceftriaxone (CEF), co-trimoxazole (COT), nalidixic acid (NAc) and norfloxacin (NOR) in µg. Values of 0 indicate resistance while numerical values are zones of inhibition in millimetres denoting sensitivity ...... 84 Table 6.1: Isolates from five hot springs, Limpopo Province, SA, showing temperature and isolation media, and comparison of the 16S rDNA sequences with GenBank and accession numbers ...... 93 Table 7.1: Identification, isolation conditions and enzyme characterisation for hot-spring isolates, Limpopo Province, South Africa ...... 111 Table 7.2: LC-MS/MS identification of protein in crude bacterial supernatant of isolates 76S, 77S, 85Li and 19S ...... 115 Table 8.1: A comparison of biosurfactant activity of 12 isolates of emulsion activity against paraffin oil, petroleum and sunflower seed oil, and Parafilm M drop-collapse assay ...... 137 Table 8.2: Isolates from hot springs with biophysical attributes relevant to wastewater bioremediation ...... 142

xxi

Table 9.1: Summary of substrates tested against CFCS fractions of isolates for reduction of colouration or turbidity and/or phenol concentrations at 25 °C for 3 h ...... 145 Table 9.2: Identification of relevant isolates from hot springs that reduced phenol in simulated pollutants and wastewater samples ...... 153 Table 10.1: Biosurfactant activity of cell-free culture supernatants of five thermophilic and four mesophilic isolates determined by emulsion assay and drop-collapse assay ...... 164 Table 10.2: Crude CFCS extracts of hot-spring isolates tested for antimicrobial activity against panel of environmental isolates and human pathogens. Zones of inhibition are in millimetres (mm) ...... 166 Table 10.3: Antibiotic resistance profiles of panel of environmental isolates used to test antimicrobial activity of CFCS of isolate 76S ...... 168 Table 10.4: Characteristics of two biosurfactant proteases produced by B. mojavensis 76S identified by LC-MS/MS ...... 169

xxii

ABBREVIATIONS

µg microgram

°C degrees celcius aa amino acids

AIDS/ HIV acquired immune deficiency syndrome / human immunodeficiency virus

Al aluminium

ALAD aminolevulinate dehyratase

AMPs antimicrobial peptides

AR antibiotic resistance

ARDRA amplified rDNA restriction analysis

ARG antibiotic resistance genes

As arsenic

BB bromothymol blue

BB-NB bromothymol blue in nutrient broth bp base pairs

BLAST Basic Local Alignment Search Tool

CaCl2 calcium chloride

CAR carbenicillin

Cd cadium

CDC Centre for Disease Control

CEF ceftriaxone

CFCS cell free culture supernatant cfu colony forming units

CHL chloramphenicol

COT co-trimoxazole

xxiii

Cr chromium

Cu copper

CV crystal violet

DNA deoxyribose nucleic acid

DO dissolved oxygen

EISQ enzymatic indicator of sediment quality

EU European Union

FC Ficol-Ciocalteau

FDR false discovery rateFe iron

GC guanine-cytosine

GEN gentamicin

GRAS generally regarded as safe

GPS global positioning system

Hg mercury

HMT heavy metal tolerance

KAN kanamycin

LA luria agar

Le Lekkerus

LHS left hand side

Li Libertas

LC-MS liquid chromatography – Mass spectrophotometry

M Mphephu

MAR multiple antibiotic resistance

MHMT multiple heavy metal tolerance

Mn manganese

xxiv

mM millimolar

MRSA methicillin resistant Staphylococcus aureus

MUSCLE Multiple Sequence Comparison by Log-Expectation

NB nutrient broth nd not done n number

NAc nalidixic acid

NOR norfloxacin

Ni nickel

N-J neighbour joining

NA nutrient agar

NaCl2 sodium chloride

OD optical density

PAHs poly aromatic hydrocarbons

Pb lead

PBS phosphate buffered saline

PCR polymerase chain reaction

POP persistent organic pollutants r resistant

RFLP restriction fragment length polymorphism

RHS right hand side

RNA ribonucleic acid rpm revolutions per minute

S Siloam s sensitive

xxv

SA South Africa

SABS South African Bureau of Standards

SDS-PAGE sodium dodecyl sulphate polyacrylamide gel electrophoresis

SLS sodium lauryl sulphate

SOD superoxide dismutase

STP streptomycin

T Tshipise

TET tetracycline

USA United States of America

USEPA United States Environmental Protection Agency

USFDA United States Food and Drug Administration

WHO World Health Organisation

WWTPs wastewater treatment plants

WW wastewater

xxvi

CHAPTER ONE: INTRODUCTION

1.1 PROBLEM STATEMENT

South Africa (SA) is an arid water-scarce country with a large proportion of the rural population having no access to safe drinking water. It is imperative that SA’s water resources are well understood, well managed and maintained for overall human health, the country’s economy and natural biological diversity as a resource. All water bodies and springs that feed into groundwater and surface water therefore need to be monitored for microbial diversity and water quality. In SA, there are more than 80 hot springs, a third of which are located in the Limpopo Province in the southern Waterberg area and the northern Soutpansberg area. Only a small number of hot springs have been developed for human recreational purposes with water temperatures ranging from 25-67 °C (Olivier et al., 2011).

López-López et al. (2013) studied hot-spring environments and suggested that the diversity of this specialised habitat is yet to be fully appreciated with <1% of bacteria isolated and identified using traditional culture-based methods. The metagenomic profiles of these Limpopo (Tekere et al., 2011; 2012) and Western Cape hot springs (Selvarajan et al., 2017) have been previously investigated; however, this analysis only describes the variety of deoxyribonucleic acid (DNA) of species in a population of microorganisms, and does not take into account viability and/or circumstantial contamination. The importance of culturing an organism cannot be underestimated in order to understand the biochemical potential and production of bioactive molecules associated with viability. The harsh environments of hot springs include extreme conditions of temperature, pH, salinity and oxygen limitations. These environments result in unique properties of evolutionary and adapted cellular structures and biochemistry which may be useful in biotechnology such as polymers and extremozymes or new drug discoveries (Demirjian et al., 2001; Demorne et al., 2017). The aforementioned therefore, justifies this study for exploring (by traditional culture-based methods) the diversity of bacteria of hot springs.

Through the ages, hot springs have been known for their healing qualities or balneotherapy (Hamzah et al., 2013b; Boekstein, 2014). Although very few reports of human infections have been linked to hot spring water, Legionella, viral and protozoan pathogens have been sporadically documented (LeClerc et al., 2002). The potential of infection appears to be associated with the level of human activity and contact (Perestrelo et al., 2006; Singh et al.,

2013a). However, the microbial health risks of these pristine hot spring waters (without contact with human or human activity) have yet to be established because very little is known about their viable microbial population. The areas surrounding the springs are undeveloped and of rural setting, and the water feeds into the groundwater used by the local population for many domestic purposes including drinking water. In a population where >10% are AIDS/HIV positive (https://www.statssa.gov.za/publications/P0302/ P03022016.pdf), the aetiological agents of infection must therefore not just include the established waterborne pathogens such as Escherichia coli, Campylobacter, Vibrio cholera, Helicobacter, Shigella, Pseudomonas, nontuberculous mycobacterium and Legionella, and protozoa infections by Acanthamoeba and Naegleria fowleri (Nel & Markotter, 2004), but must also include opportunistic emerging pathogens. Pathogens that are both opportunistic and emerging have been described as Pseudomonas (Akhtyamova, 2013), Ralstonia (Ryan & Adley, 2014), Kocuria (Purty et al., 2014; Paul et al., 2015; Pulcrano et al., 2017), Sphingomonas (Dewan et al., 2014), Cupriavidus (Bittar & Rolain, 2010) and Cronobacter sakazakii (Hunter et al., 2008; Fakruddin et al., 2013).

Waterborne pathogens are not the only indicator of water quality to be monitored, and recently the increased levels of multiple antibiotic resistant (MAR) bacteria have also been considered to be a modern day emerging water pollutant (Coutinho et al., 2013; Sanderson et al., 2016). In SA, bacteria from drinking water in a rural setting were found to be 78% MAR (Samie et al., 2012), whilst bacteriological testing of samples from wastewater (WW) and dams in the North West Province showed that >70% of E. coli isolates were resistant to one or more antibiotic (Kinge et al., 2010). Both local and global studies suggest that antibiotic resistance (AR) is a significant problem in water resources, because of direct infections by AR pathogens or indirectly as a possible source of transfer of AR from ubiquitous harmless environmental bacteria to well established human pathogens. The prevalence of AR in isolates from hot springs has not been well investigated although the occurrence of AR-carrying plasmids has been reported (Imanaka et al., 1981).

Man has used bacteria for many purposes in biotechnology including the bioremediation of WW, and the discovery of new bacterial species or novel enzymes and bioactive molecules with unusual properties is highly desirable. Bioremediation is a process that uses living organisms such as microorganisms or plants for the degradation or transformation of contaminants into nonhazardous or less hazardous substances (Karigar & Rao, 2011; Facchin et al, 2013) and has several advantages over conventional treatment processes. This includes lower costs, green footprint, decreased toxicity to the environment, ease of application, specificity and flexible applications to a variety of possible substrates (Vidali, 2001). South Africa’s limited water

supply is threatened with water pollution by domestic, pharmaceutical and industrial waste, acid mine drainage and heavy-metal pollution. Problems arise when wastewater treatment plants (WWTPS) are unable to remove complex structured and recalcitrant pollutants (pharmaceuticals, textile dyes), water-soluble and toxic pollutants in small concentrations (heavy metals), or water-insoluble pollutants such as petroleum and polycyclic aromatic hydrocarbons (PAHs). Furthermore, pollutants may also increase the turbidity or change the colour of the water thereby reducing sunlight in a water body causing the natural ecosystem to become unbalanced. There are several avenues where microorganisms can be involved in WW bioremediation. Firstly, their specialised and numerous enzymes are involved in the breakdown of pollutants (Karigar & Rao, 2011; Facchin et al., 2013), and are mainly oxidoreductases (including catalase, peroxidase, azoreductase) and hydrolytic enzymes (including amylase, protease, lipase, glucosidase, phosphatase, cellulase). Secondly, biophysical characteristics of the microorganisms can assist in removal of unwanted pollutants. Biophysical characteristics include i) bioassimilation - active transport of the pollutant into the cell by active metabolism (Chojnacka, 2010); ii) biosorption - removal of substances from solution, both inorganic and organic, insoluble and soluble, by biological material, based on “high attractive forces” between the substances and the biomass; the biomass can be both living and none living cells (Abbas et al., 2014); iii) bioflocculation - coagulation and precipitation of suspended solid particles or heavy-metal ions by natural substances that facilitate the aggregation of particles to form flocs (Zhu et al., 2014; Agunbiade et al., 2016); and iv) biosurfactant activity - groups of amphipathic molecules that lower surface and interfacial tensions of liquids, form micelles and thereby enhance the solubility of poorly soluble compounds in aqueous environments (Pacwa- Plociniczak et al., 2011). The abovementioned bacterial properties would be useful for biotechnological applications. Bacteria cultured from hot springs are commonly spore-forming heat-tolerant Bacillus and Bacillus-related species that are found in several countries such as India (Pandey et al., 2015), Tunisia (Sayeh et al., 2010), Saudi Arabia (Khiyami et al., 2012), Jordan (Obeidat et al., 2012) and Bulgaria (Derekova et al., 2008). This group of bacteria is also very important in biotechnology due to their ease of culture, robustness and thermotolerance (Kumar et al., 2013; Chen & Jiang, 2018) making SA’s indigenous microflora of Bacillus spp. from hot springs a potential resource to deal with bioremediation of pollutants in water. The literature review revealed that these genera cultured from other hot springs are producers of enzymes that degrade food WW including amylase (Matpan Bekler & Güven, 2014; Zhang et al., 2015) and proteases (Al-Qodah et al., 2013; Bekler et al., 2015). Furthermore, Bacillus spp. isolated from hot springs have been found with biosorption (Al-dayhistan, 2012; Ghalib et al., 2014) and biosurfactant properties (Pakpitcharoen et al., 2008; Khairuddin et al., 2016). As

proof of concept, this versatile group of bacteria has been useful in bioremediation of many pollutants both in the laboratory (Gayathri & Vasudevan, 2010; Kavitha et al., 2014) and in WW samples (Ogunlaja et al., 2013; Bujang et al., 2013). As the level of pollution increases globally (Schwarzenbach et al., 2010), and the complexity of chemical pollutants grows, there is a constant demand for novel bacteria including those harbouring new bioactive molecules or established bioactive molecules that can optimally function in extreme settings (temperature, pH and salinity) that can address water bioremediation issues.

The culture-based study of the microbial communities in hot springs therefore has several outcomes that relate to increasing the knowledge base of indigenous diversity of specialised environmental and extreme niches, water safety and the presence of potential pathogens, as well as the isolation of strains that have a potential use in bioremediation of WW.

1.2 AIMS AND OBJECTIVES OF THE STUDY

Since isolates from hot-spring environments are relatively poorly characterised globally, and in SA only from a metagenomic perspective, the aim of this study was to use culture-based microbiology to describe the bacterial diversity of hot-spring water in the Limpopo Province, SA. Comment will be made on the implications for water safety, and possible use and applications in WW bioremediation according to the objectives set out below:

1. Identification of cultured bacteria from hot-spring water and sediment using genetic analysis. 2. Detection and identification of potential emerging and opportunistic pathogens. 3. Investigation of antibiotic resistance and heavy-metal tolerance in isolates. 4. Screening for potential bacterial enzymes useful for bioremediation. 5. Selection of bacterial isolates that have properties that assist in bioremediation, i.e. bioflocculant, biosurfactant, biosorption and anti-biofilm properties. 6. Testing of bacterial supernatants against pollutants and wastewater samples in vitro. 7. Determination of antimicrobial properties and applications to pathogens and the environment.

1.3 HYPOTHESIS

Since hot springs are a unique environmental site that has not been well investigated, it is hypothesised that novel bacteria or established bacteria with novel characteristics will be isolated and cultured. Furthermore, it is envisaged that bacteria with potential for use in WW bioremediation will be identified.

CHAPTER 2: LITERATURE REVIEW

2.1 INTRODUCTION

This literature review is divided into three sections. The first section (2.2 and 2.3) will describe previous findings on the microbial diversity of hot springs, and comment on the methodologies for identification and characterisation. The second section (2.3 and 2.4) will relate to water safety in terms of waterborne pathogens, and AR profiles from pristine sites in the environment including hot springs. The third section (2.5 – 2.6) will cover the application of bacteria to WW bioremediation (2.5), and the properties that such bacteria would require to be suitable candidates for use in the removal or degradation of unwanted pollutants including other bacteria (2.6).

2.2 MICROBIAL DIVERSITY OF HOT SPRINGS

Microbes are ubiquitous, able to grow in many extreme environments of temperatures, pH, salinity, pressure, nutrients, oxygen, toxins and radiation. These environments result in unique properties of evolutionary and adapted cellular structures and biochemistry which may be useful in biotechnology such as polymers and extremozymes (enzymes that are stable and active under extreme conditions), or new drug discoveries (Demirjian et al., 2001; Demorne et al., 2017).

The extreme limits for temperature, pH and salinity of thermophilic, alkaliphilic and halotolerant microbes, respectively, have been described (Demirjian et al., 2001; Gerday, 2002). Thermophiles grow optimally at 55-60 °C but the range for hyperthermophiles can extend to 110 °C. It is speculated that primitive life began in hot springs evolving from thermophiles (Di Giulio, 2007). Within the environmental niche of hot springs, it could be possible to isolate extremophilic microorganisms that flourish in extreme environments and have characteristics such as halotolerance, tolerance for extreme pH values, thermostability and can survive under high barometric pressures.

The premise that microorganisms did not exist unless they could be cultured was challenged when it became clear that many microorganisms could not be grown in pure culture (Handelsman, 2004). It became evident that the number of phyla with cultivable representatives had seriously lagged behind (Alain & Querellou, 2009). In a review on the phylogeny of thermophiles, 26 phyla of prokaryotes were described (Lebedinsky et al., 2007). López-López et al. (2013) more specifically studied hot-spring environments and suggested that the diversity of this specialised habitat is yet to be fully appreciated. They estimated that <1% of bacteria in hot

springs are isolated and identified using traditional culture-based methods. The advent of molecular techniques including polymerase chain reaction (PCR) and metagenomics has provided access to far more microbial diversity than was possible using culture-based methods as it gives insight into groups of prokaryotes that are otherwise entirely unknown. Analyses of 16S rDNA sequences yields a phylogenetic description of microorganisms, but it gives only scant information on the genomics, physiology and biochemistry of these organisms. Metagenomics, the genomic analysis of a population of microorganisms, provides a second tier of technical innovation that facilitates the study of the physiology and genetics of uncultured microorganisms (Handelsman, 2004). Although, metagenomics reveals a variety of genetic diversity, it does not take into account viability and circumstantial contamination. Therefore, any novel bacteria isolated from these unique environmental sites are essential contributions to the current database and general understanding of microbial communities. The importance of culturing an organism cannot be underestimated in order to understand the biochemical potential and production of bioactive molecules related to viability. The application of metagenomic sequence information will facilitate the design of better culturing strategies to link genomic analysis with pure culture studies (Handelsman, 2004). Worldwide studies on thermophiles from hot springs have been carried out through the use of metagenomics (López-López et al., 2013; Strazzulli et al, 2017; Sharma, et al., 2017) as well as through isolation and culture-based techniques (Mawadza & Zvauya, 1996; Derekova et al., 2008; Narayan et al., 2008; Cihan, 2013; Khiyami et al., 2012; Pandey et al., 2015).

The guanine-cytosine (GC) content of the DNA of bacterial genomes varies with different genera, and is useful in bacterial systematics. In addition, the GC content has been correlated with thermostability of a genome and is therefore higher in thermophiles (Wang et al., 2006). The use of amplified rDNA restriction analysis (ARDRA) and 16S rDNA sequencing allows for a more accurate, rapid and efficient identification compared with the more traditional microbiological and biochemical methods (Rajendhran & Gunasekaran, 2011). A computer- simulated restriction fragment length polymorphism (RFLP) analysis of the PCR-amplified 16S rDNA (also called ARDRA) is a valid means of identifying unknown organisms (Moyer et al., 1996). The 16S rDNA amplicon sequencing allows for maximum discrimination between individual isolates taking into account the entire 16S rDNA sequence, while ARDRA represents only variations in the restriction enzyme sites. The use of sequences in phylogenetic analysis allows differentiation between closely related isolates at every nucleotide site. Using metagenomic analysis and colony PCR to screen the microbes of hot springs, a predominance of the phyla Proteobacteria and Cyanobacteria has been reported in Malaysia (Goh et al., 2011), India (Sharma et al., 2014b) and SA (Tekere et al., 2011; 2012). More recent metagenomic

studies of hot springs have been recently reported in SA (Selvarajan et al., 2017), India (Gupta et al., 2017; Sharma et al., 2017), Pakistan (Ahmed et al., 2017) and Russia (Yu et al., 2017). However, by isolation and culture, the predominant bacteria are Gram-positive spore-forming Bacillus and closely related Bacillus bacteria. This has been reported in India (Pandey et al., 2015; Panda et al., 2016), Tunisia (Sayeh et al., 2010), Saudi Arabia (Khiyami et al., 2012), Armenia (Panosyan & Birkeland, 2014), Fiji (Narayan et al., 2008), Jordan (Obeidat et al., 2012), Bulgaria (Derekova et al., 2008), Turkey (Gulecal-Pekta, 2016) and Zimbabwe (Mawadza & Zvauya, 1996). In 2007, using studies based on the 16S rDNA gene sequences, several Bacillus species were reclassified into new genera phylogenetically related to the genus Bacillus (Ludwig et al., 2009). In addition, this is a highly diverse and expanding group, with 25 new genera being described in the past two years, and 38 new species of Bacillus described since August 2013 (Mandic-Mulec et al., 2015). Like the genus Bacillus, the relatively new genus Anoxybacillus, was established in 2000 and is growing rapidly with six new species being described since 2011 (Mandic-Mulec et al., 2015). Of the 115 endospore-forming Bacillus isolates from geothermal regions in Turkey, Anoxybacillus was the most abundant, being represented by 53 isolates (Cihan, 2013). This suggests that geothermal environments could be a niche for the discovery of new Anoxybacillus species. Because the genera Bacillus and Anoxybacillus have been reclassified, and novel species are being described at a rapid rate, there may be some incongruence and confusion when comparing the nomenclature of this group from studies prior to the reclassification, and between different studies where the new nomenclature is not taken into account.

Since Bacillus spp. are spore-forming, it is not uncommon to isolate both thermophilic and mesophilic strains from the same habitat. Despite the depletion of natural environments that potentially harbour undiscovered microorganisms, hot springs remain common sites for the isolation of such novel species of microorganisms. Novel cultured microorganisms can add to current microbial culture collections worldwide.

In SA, there are more than 80 hot springs, a third of which are located in the Limpopo Province in the southern Waterberg area and the northern Soutpansberg area. Only a small number of hot springs have been developed for human recreational purposes. In some of the hot springs the water is not potable due to the presence of bromide, fluoride and mercury (Olivier et al., 2011). A metagenomic study of four hot springs in the Soutpansberg area of the Limpopo Province was undertaken by Tekere et al. (2011; 2012) using the 16S rDNA sequence as a phylogenetic marker to determine the bacterial diversity of these hot springs. In those two studies it was shown that the Proteobacteria dominated in all the samples; however, the levels of the phyla

Cyanobacteria and Bacteroidetes differed between hot springs. The phylum Firmicutes which includes Bacillus and Bacillus-related species was detected in all the springs but in much lower abundance. Various other phyla were reported but in very small percentages of the total rDNA sequences (<0.2%). The discovery that these hot springs hold a great diversity of bacteria suggests that they might be refuges for potential thermophiles that could have novel biotechnological applications. Bacillus and Bacillus-related species have been used for biotechnology with various applications due to their ability to grow easily, their robustness in the environment and because they are generally regarded as harmless to humans (Kumar et al., 2013; Mandic-Mulec et al., 2015; Chen & Jiang, 2018). In this study, conventional culture techniques were used for the isolation of organisms and 16S rDNA sequence analysis was used for genotypic identification of the isolates.

2.3 PATHOGENS AND OPPORTUNISTIC PATHOGENS

In developing countries, many people do not have access to clean safe water that is essential to all life. Water resources are also further threatened by the presence and pollution of human pathogens, and waterborne diseases are globally well described and investigated (Sharma et al., 2003; Nel & Markotter, 2004; Pandey et al., 2014). Well established waterborne bacterial pathogens include E. coli, Campylobacter, Vibrio cholera, Helicobacter, Shigella, Pseudomonas, non-tuberculous mycobacterium and Legionella, while protozoa infections also include Acanthamoeba and Naegleria fowleri (Nel & Markotter, 2004). The majority of waterborne diseases result in diarrhoea, and globally more than 2.2 million children die each year as a result. In SA, diarrhoea is the main cause of mortality in children under five years old (DeWaal et al., 2010).

In general, water from hot springs is associated with healing properties called balneotherapy or hydrotherapy (Nasermoaddeli & Kagamimori, 2005; Lund, 2009). However, some studies have reported the isolation of the above-mentioned pathogens from swimming pools (Barna & Kádár, 2012), and thermal baths associated with hot springs (Perestrelo et al., 2006; Singh et al., 2013a). The potential of infection appears to be associated with the level of human activity and contact, and these organisms are not the dominant microorganisms that can potentially cause disease from pristine hot-spring waters, i.e. water that has had no previous contact with human or animal activity. Infections from hot springs are rare and sporadic and mostly associated with Legionella (Kurosawa et al., 2010) and free-living amoeba (Acanthamoeba and N. fowleri) (Sukthana et al., 2005). Other protozoa, Vittaforma (Fan et al., 2012) and fungi Ochroconis

gallopava (Yarita et al., 2007), have also been reported to cause infections associated with hot springs.

Emerging human pathogens are defined as aetiological agents of infectious diseases whose incidence have increased in the past 20 years and will probably do so in the future. In several older reviews on emerging waterborne pathogens, Legionella and Enterobacter sakazakii (Sharma et al., 2003) were mentioned. The latter, later renamed Cronobacter sakazakii, was isolated in baby formula and is an emerging pathogen of infants and neonates (Hunter et al., 2008; Fakruddin et al., 2013). On the other hand, opportunistic pathogens are described as pathogens causing infections that exploit opportunities not normally available, such as a host with a weakened immune system, an altered microbiota (such as a disrupted gut flora), or breached integumentary barriers (Berg et al., 2005). Pathogens that are both opportunistic and emerging have been described in more recent publications including Pseudomonas (Akhtyamova, 2013), Ralstonia (Ryan & Adley, 2014), Kocuria (Purty et al., 2013; Paul et al., 2015; Pulcrano et al., 2017), Sphingomonas (Dewan et al., 2014) and Cupriavidus (Bittar & Rolain, 2010). Opportunistic pathogens, by their nature, are not definitively harmful or harmless, and infections are rather a reflection of the host’s state of health and immunity. Commonly, these bacteria are ubiquitous in the environment, and are classified in the lowest level of biohazard, i.e. level one (Van Belkum, 2011). They can, and do cause infections in debilitated or immunocompromised individuals (including neonates, infants, the elderly, those infected with AIDS/HIV and cystic fibrosis patients) and include 27 genera as listed by Berg et al., (2005). They also commonly cause nosocomial infections in hospitals (Williams et al., 2013; Kanamori et al., 2016). Eight percent of acute gastrointestinal infections in the USA are from unknown causes (Cliver, 2000) and DeWaal et al. (2010) also reported that the causative agents of a large proportion of global diarrheal outbreaks remain unspecified. These opportunistic and emerging pathogenic bacteria may be the missing link in waterborne diseases of unknown aetiology. Cases studies and outbreaks due to these opportunistic pathogens are listed in Appendix A, showing the diversity of infections possible, global location of infections and risk factors associated with the host’s condition.

It is well known these bacteria can be isolated from environmental soil and water including C. sakazakii (Fakruddin et al., 2013), Hafnia (Janda & Abbott, 2006), Sphingomonas (Kim et al., 1998; Piccini et al., 2006) and Legionella (Parthuisot et al., 2010). Cupriavidus has been isolated from drinking water (Berthiaume et al., 2013), and Gulbenkiania sp. from WW (Vaz- Moreira et al., 2007). Not only do they exist ubiquitously in the environment but interestingly

are found in the plant rhizosphere on or within the plant, and play a role in production of plant growth factors (Akhtyamova, 2013; Berg et al., 2005).

Most cultured bacteria from hot springs in other studies (Khiyami et al., 2012; Obeidat et al., 2012; Panda et al., 2016) have been described as phylum Firmicutes genus Bacillus, due to their endospore-forming ability, thermotolerance and relatively simple nutritional requirements (Pandey et al., 2015). However, in metagenomic studies, the Gram-negative phylum Proteobacteria is equally as predominant in the microbiome as the phylum Firmicutes (Lebedinsky et al., 2007). In sporadic publications of bacterial isolations from hot springs, alpha-, beta- and gamma-Proteobacteria have been described including Hafnia, Ralstonia, Tepidimonas, Sphingomonas and Silanimonas (see Figure 1.1, Table 1.1). It appears that these non-spore-forming bacteria are also able to withstand harsh environmental conditions of temperature, pH, salinity and radioactivity. Extremophilic properties have been reported for Ralstonia (Lee & Lee, 2001), Legionella (Berjeaud et al., 2016), Kocuria (Asgarani et al., 2012; Gholami et al., 2015), Tepidimonas (Chen et al., 2006a) and Cronobacter (Jaradat et al., 2014). With the exception of Legionella and Pseudomonas, there have been no reported associated infections by the abovementioned bacteria from hot springs.

Figure 2.1:World wide bacterial isolations made from hot springs

Table 2.1: List of emerging opportunistic pathogens that have been isolated from hot spring water

Opportunistic pathogen Country

Free-living amoeba Taiwan (Kao et al., 2013; Tung et al., 2013); USA (Baumgartner et al., 2003)

Kocuria Iran (Asgarani et al., 2012; Gholami et al., 2015)

Arthrobacter Japan (Yoshinaka et al., 1973)

Sphingomonas China (Briggs et al., 2014)

Ralstonia Korea (Lee & Lee, 2001); Pakistan (Shah et al., 2015)

Gulbenkiania India (Jyoti et al., 2010; Verma et al., 2015)

Tepidimonas Taiwan (Chen et al., 2006a); Portugal (Moreira et al., 2000)

Legionella Taiwan (Hsu et al., 2006; Huang et al., 2010); Thailand (Sukthana et al., 2005); France (Chaabna et al., 2013); China (Qin et al., 2013); Japan (Kobayashi et al., 2014); USA (Sheehan et al., 2003); Poland (Walczak et al., 2016); Tunisia (Ghrairi et al., 2013)

Silanimonas Korea (Lee et al., 2005)

Cronobacter Malaysia (Jimat et al., 2015); Saudi Arabia (Khiyami et al., 2012)

Globally, in 2016 the prevalence of AIDS/HIV was 0.5% with 3.7 million infected out of a world population of 7.5 billion (http://www.unaids.org/sites/default/files/media_asset/global- AIDS-update-2016_en.pdf). The population of SA is vulnerable to opportunistic infections for various reasons. In 2016, the prevalence of AIDS/HIV in South Africa was 12.7% (https://www.statssa.gov.za/publications/P0302/P03022016.pdf). Of the 539 714 deaths, 27.9% were related to HIV. Approximately 10.4% of the population is <5 years old, while approximately 8% are >60 years old. Consequently 18% of the population is very young and elderly. Although 60 years of age might not be considered old in other countries, the life

expectancy in SA is 61 (males) and 64 (females) years old. Malnutrition also weakens the immune system and is a major concern in the country where the growth of one in five children under five is being stunted due to lack of nourishment (Said-Mohamed et al., 2015). In addition, SA is a water-scarce country, with people living in rural areas utilising groundwater for both domestic and drinking purposes thereby increasing the contact with untreated water. Springs, both hot and cold feed the system, and should therefore be monitored for their impact on water safety. All these factors result in a large proportion of the South African population being at an increased risk of infection due to opportunistic pathogens.

2.4 ANTIBIOTIC RESISTANCE OF PRISTINE SITES AND HOT SPRINGS

Large-scale use of antibiotics started in the 1930s when sulphonamide drugs were introduced, and continued through World War II to the current use in the fight against infections in humans and animals. Their use have been extended to control plant diseases, aquaculture and even as growth promoters in domestic stock animals (Al-Bahry et al., 2014). As a consequence, different antibiotics, and higher levels thereof, occur in the environment (Al-Bahry et al., 2014). Bacteria have developed resistance resulting in the emergence of MAR, and is a huge public health concern especially when pathogens acquire resistance rendering treatment of infection inadequate (Al-Bahry et al., 2015). Since antibiotic resistance genes (ARGs) via their microbial hosts are able to be transmitted through the environment in many ways, i.e. physically through wind- and water-borne particulate contaminants, and mechanically by human and animals via gut bacteria, the problem of the spread of AR is exacerbated when people or animals are in close proximity to hospitals and battery farms, respectively (West et al., 2010; Rodriguez-Mozaz et al., 2016). These sites together with wastewater treatment plants (WWTPs) become reservoirs for microorganisms carrying AR genes and create an opportunity for their transfer to potentially clinically important pathogens (Rizzo et al., 2013). Antibiotics are not removed by conventional WW treatment processes thus contributing to the increase in risk that AR strains will develop and spread. Antibiotic-resistant bacteria (ARB) or ARGs are now considered as pollutants in aquatic environments (Martinez, 2012; Sanderson et al., 2016).

Many studies have described AR in aquatic environments (Milić et al., 2013; Vaz-Moreira et al., 2014) all over the world including the USA (Bollin et al., 2015), Canada (Belliveau et al., 1991), Brazil (Coutinho et al., 2013), India (Tewari et al., 2013), Uganda (Soge et al., 2009) and Nigeria (Ayandiran et al., 2014). In SA, where a large proportion of the population do not have access to treated piped water, and often have to make use of untreated groundwater and surface

water for drinking and domestic use, the presence of ARB is likely to have a more pronounced negative impact on human health. In the North-West Province, Kinge et al. (2010) found that more than 70% of 230 isolates of E. coli were MAR in rivers. Similarly, in the KwaZulu-Natal Province, this figure was 72% of 113 enteric bacteria (Lin et al., 2004). In drinking water reservoirs of HIV-infected and AIDs patients, in the Limpopo Province, Samie et al. (2012) reported that 78% of coliform bacteria were MAR. An increase in exposure to AR opportunistic pathogens may be a contributor to the general health of a population, especially HIV- immunocompromised individuals, emphasising the need to monitor the levels of AR in the environment.

However, AR has occurred in the environment for billions of years, existing way before the existence of mankind (Allen et al., 2010). Not only is AR ancient, but intrinsic and natural in the environment (Allen et al., 2010; Martinez, 2012; Sengupta et al., 2013). Evidence of ARGs without gene expression has been reported in 30 000-year-old permafrost sediments (D’Costa et al., 2011), ancient arctic sediments (Perron et al., 2015) and glaciers of the world (Segawa et al., 2013). Genes that confer resistance are known as the environmental resistome. Antibiotic- resistant microbiota have been isolated extensively in environments that are pristine and antibiotic-free such as the waters of the Antarctic (De Souza et al., 2006; Miller et al., 2009), in caves that have been in isolation untouched by mankind for 4 million years (Bhullar et al., 2012), and in faecal, oral and skin microbiomes of native Amerindians who have never had contact with western civilisation and never used antibiotics (Clemente et al., 2015). Lima- Bittencourt et al. (2007) reported that 61% of the enterobacterial isolates from a remote and pristine stream in the Serra do Cipo National Park, Brazil were MAR. These examples illustrate some level of naturally occurring AR, and the existence of novel ARGs. The role of antibiotics in nature has yet to be fully described, and it is thought that these same molecules have a different function in their natural state, other than as an antibiotic (Perron et al., 2015). Intrinsic AR differs from acquired AR that is selected by an external source of antibiotics in the environment. There is a greater genetic diversity of intrinsic or natural AR, and as indicated by metagenomics studies, the genes are chromosomally located and often sub-inhibitory (Perron et al., 2015; Sengupta et al., 2013). Natural “antibiotics” function in quorum sensing and biofilm formation. They activate biochemical pathways associated with community structure, virulence and pathogenesis, and communication between host and parasite (Sengupta et al., 2013). Multidrug efflux pumps play an important role in processes such as intercellular signal trafficking, detoxification of metabolites and extrusion of plant compounds. These natural ARGs in environmental microbiota can be introduced into new bacterial hosts that do not normally express AR and do not have the original biochemistry, via mobile genetic elements

such as transposons and plasmids. When this happens, the transferred genes confer only AR and not the other cellular functions (Martinez, 2009a). For example, the genes for beta-lactamases (β-lactamases) have occurred in microorganisms for billions of years, and may merely have been involved in cell-wall biosynthesis prior to the exposure to man-made antibiotics being introduced into the environment (Martinez, 2012). However, with the introduction of antibiotics and transfer of AR to new bacterial hosts, bacterial resistance to β-lactam antibiotics occurs as the functionality of β-lactamases changes of breaking the β-lactam ring that allows antibiotics such as penicillin to work (Martinez, 2012). The problem arises when AR is introduced into animal and human pathogens, thereby leading to a decrease in effectiveness of infection control through antibiotics (Martinez, 2012). By comparing the role and functions of ARGs that occur naturally in environmental microorganisms prior to exposure to antibiotics in the environment with the antibiotic-resistant microbial populations post-exposure, one can gain better insight into the diverse information on the functionality and evolution of the mechanisms of AR (Martinez, 2012). This can be done through metagenomic analysis of different AR populations of environmental and pathogenic microorganisms. Although there are abundant sources of sampling possibilities for microorganisms in the post-antibiotic era, in modern times it is difficult to find a site in the environment that has not been exposed to man-made antibiotics.

Hot and cold freshwater springs are such locations. If sampled correctly by avoiding contact with the surface environment and human activity, these are also “pristine” environments. Water is commonly bottled from freshwater springs, and Mary et al. (2000) reported MAR in five brands of French bottled spring water. Falcone-Dias et al. (2012) confirmed MAR in a study of 238 cultured isolates from a brand of French bottled spring water, and two Portuguese brands, suggesting that these studies may reflect the naturally occurring intrinsic baseline levels of AR. Antibiotic resistance has been reported in hot-spring isolates in Turkey (Sariözlü et al., 2012), Jordan (Akel et al., 2008), India (Sen et al., 2010) and the USA (Hudson, 2012). Isolates from hot springs have been reported to carry plasmids (Munster et al., 1985; Khalil et al., 2003), and plasmids have been implicated in the transfer of ARGs. Higher lateral gene transfer was reported in thermophiles that grew at high temperatures, more than in halophiles that grew in high salt content (Rhodes et al., 2011). Venton (2013) reported that bacteria shed DNA at higher temperatures as reflected by the smaller genome size of isolates at >60 °C compared with that of isolates at <45 °C. The observation of a higher shedding rate of DNA at a high temperature is further justification for the study of microorganisms from hot springs in order to determine the dynamics of ARGs.

Antibiotic resistance and heavy-metal tolerance (HMT) has been correlated in studies of river isolates. In two independent studies of predominately Gram-negative bacteria from rivers in Turkey, Icgen and Yilmaz (2014) tested 240 isolates from the Kizilirmak River against 26 antibiotics and 17 heavy metals and found that 24 isolates were resistant to both. Similar results were found in 268 isolates from the Seyhan River (Matyar et al., 2014), suggesting co-selection of AR and that HMT was possible. Tewari et al. (2013) investigated 201 coliform waterborne isolates from rural areas in India with most of the isolates resistant to at least one antibiotic and to multiple metals. Of these, 12 were able to transfer both AR and HMT via plasmid to a susceptible recipient strain of E. coli K12. Curing of the plasmids resulted in a loss of resistance to both antibiotics and heavy metals. A similar study was done on a Gram-positive Bacillus isolate from municipal waste (Samanta et al., 2012a) where resistance was found against kanamycin, ampicillin and methicillin as well as chromium, and nickel. The loss of plasmids was associated with a loss of resistance to both antibiotics and heavy metals, and co-selection and co-transference was implicated. The aim of this study was to determine the level of resistance to antibiotics cultured bacteria from hot springs in Limpopo Province, SA. The detection of tolerance to heavy metals may also be an indicator of AR if there is an association. In a country where surface water is used for drinking and domestic purposes in rural settings, the natural AR of isolates in pristine water entering the environment has implications for infections by pathogens and opportunistic pathogens further downstream. It was also used to explore the possible use of hot springs to provide baseline levels of AR that occurs naturally in such a unique environment.

2.5 BIOREMEDIATION OF WASTEWATER

Agricultural, industrial and domestic activities generate WW containing numerous chemicals at varying concentrations (Schwarzenbach et al., 2010; Chen et al., 2015b). More specifically, these chemicals are pesticides, biocides, fertilisers, heavy metals, petroleum and oil residues, textiles dyes, antibiotics and hormones as pollutants. Inorganic pollutants, e.g. nitrogen and phosphorus from fertilisers can also indirectly cause eutrophication and toxic algal blooms. Waterborne pathogens are also regarded as pollutants (Schwarzenbach et al., 2010).

Persistent organic pollutants (POPs) are a large and diverse group of chemicals that enter the environment intentionally as a means for biological control (biocontrol) of unwanted insect pests or weeds, or inadvertently as by-products or waste (Xu et al, 2013). Due to their recalcitrant nature, they can be transported over great distances and are now found worldwide (Chakraborty & Das, 2016; Han & Currell, 2017). Because they are lipophilic and soluble in

fat, they are able to bioaccumulate in the food chain. They have been reported to cause abnormalities in wildlife including birds, fish and mammals (Han & Currell, 2017). In humans, POPs have been detected in human serum and breast milk, and linked to central and peripheral nervous system damage, reproductive and immune dysfunctions, hypersensitivity, birth defects, decreased intelligence and even death (Schwarzenbach et al., 2010; Xu et al., 2013; Han & Currell, 2017).

Polycyclic aromatic hydrocarbons (PAHs) are a major group within the POPs, defined as organic compounds with two or more fused benzene rings in linear, angular or cluster structural arrangements (Seo et al., 2009; Lu et al., 2011). Sixteen PAHs are listed as priority pollutants on the United States Environmental Protection Agency (USEPA) (Seo et al., 2009). They are ubiquitous in natural environments formed during geological reactions associated with fossil fuel and mineral production, during burning of vegetation in forest and bush fires (Seo et al., 2009). Petrogenic PAHs are derived from petroleum and petroleum products, and biogenic PAHs are aromatic amino acids, lignin compounds and their derivatives (Seo et al., 2009). Petrogenic PAHs, e.g. diesel and heavy oils, are a complex mixture of a range of molecules ranging from short-chain aliphatic and simple aromatic hydrocarbons to molecules with increasing carbon chains and complexity (Fan & Krishnamurthy, 1995).

Dyes that are found in leather, plastics, cosmetics and food industries are another category of POPs (Pokharia & Ahluwalia, 2013). Dyes are synthetic aromatic water-soluble organic colorants (Buntic et al., 2017) that are purposely produced to be very stable, not fade in light (Ramachandran et al., 2013), and therefore remain in the environment for long periods of time. They have a direct and indirect effect on the environment, being directly toxic or degrading into toxic metabolites, or resulting in a change in turbidity or colouration of the water. The latter results in decreased penetration of sunlight, and a knock-on effect where photosynthesis is blocked, oxygen transfer is inhibited and the underlying water ecosystem is disrupted (Santal & Singh, 2013). Trace amounts of dyes in water (<1 mg/L for some dyes) are highly visible, affecting the transparency in water bodies (Ramachandran et al., 2013) Often the dyes are discarded in combination with mixtures of other chemicals including heavy metals (Santal & Singh, 2013).

The textile industry receives particular attention from environmentalists because of the large consumption of water used in the processing that generates associated volumes of contaminated wastewater (Ghaly et al., 2014). It is estimated that 200 L of water are used to produce 1 kg of dyed textile. Different dyes are used for different fabrics and have been classified into groups (Ghaly et al., 2014). Protein fibres are coloured with acid dyes which include azo-dyes and

triphenylmethane dyes. Azo-dyes are the most commonly used dye accounting for 60-70% of all dye groups produced (Saranraj, 2013). Like most POPs, dyes cause health problems in humans, e.g. haemorrhage, ulceration of skin, nausea, skin irritation and dermatitis (Ghaly et al., 2014).

Some triphenylmethane dyes have been used to simulate “dye pollutants” or as dye “derivatives” in laboratory experiments testing the removal of textile dyes in WW. Mittal et al. (2009) used phenol red (Figure 2.2b), bromothymol blue (Figure 2.2c) was used by Haque & Muneer (2007) and Agarwal et al. (2016) and crystal violet (Figure 2.2d) was used by Chen et al. (2013a) and Buntic et al. (2017).

Br OH OH HO OH HO

(a) Phenol O SO3 S O O

(b) Phenol Red (c) Bromothymol Blue

O glc N HO O Cl R OR RO N OH OR glc O OH O N N OH glc O (d) Crystal Violet

(e) Melanoidin

Figure 2.2: Annotated diagrams of persistent organic pollutants: a) phenol; b) phenol red; c) triphenylmethane dye bromothymol blue; d) triphenylmethane dye crystal violet; and e) melanoidin

Melanoidins are natural dark brown nitrogen-containing complex biopolymers (Figure 2.2e) produced by a reaction between amino and carbonyl groups in organic substances (Santal & Singh, 2013). Due to the colouration, they also reduce transparency in water causing additional problems in the ecosystem by reducing photosynthesis and dissolved oxygen (Chandra et al., 2008). In addition, they have an offensive odour and are recalcitrant (Santal & Singh, 2013). Melanoidins are found in foods, drinks and WW from fermentations (Chandra et al., 2008; Santal & Singh, 2013). Some foods that contain melanoidins are coffee (Coelho et al., 2014), soya sauce (Dedin et al., 2006) and beer (Langner & Rzeski, 2014).

Phenol and phenolic compounds are stable natural aromatic hydrocarbons (Figure 2a) with one or more hydroxyl substitutes (Takeda et al., 2013); their structures are found in many toxic POPs, including those described above (Ahmad et al., 2011). Phenol levels in water arise from large-scale production, wide applicability and unrestricted release by industries (Vedula et al., 2013). According to USEPA, phenols are a priority pollutant with a limit of 0.02 mg/L in raw drinking water and 0.1 mg/L in WW discharge (Galgale et al., 2014). Even at very low concentrations, it leaves an odour, and photo-oxidises in sunlight resulting in toxic molecules. It is absorbed through the skin and causes eye and skin irritations, cyanosis, coma, convulsions and even death in surplus (Vedula et al., 2013). Phenols found naturally in foods, organic decomposition and animal and human waste, enter surface water via effluents of coal tar, plastic, rubber, coke, pharmaceutical, steel and chemical industries including domestic and agricultural runoff (Vedula et al., 2013; Galgale et al., 2014). Phenolics in WW from food production (Verheyen et al., 2009; Coelho et al., 2014), can be derived from the food source in addition to the synthetic chemicals used in the factories.

Bioremediation is defined as the use of biological matter including microorganisms, to reduce, eliminate, contain or transform benign contaminants in the environment (Adams et al., 2015). The advantages of bioremediation are its relatively low cost, low technology and effectiveness at reducing levels of a range of contaminants. The disadvantages are that time scales are relatively long and residual contaminant levels may not always be achieved (Vidali, 2001)

Bioremediation of POPs has been well reviewed (Arvind et al., 2015; Chakraborty & Das, 2016). With new technology and industrial development, the list of chemicals regarded as POPs is growing rapidly (Magulova & Priceputu, 2016). Conventionally, paths to bioremediation of pollutants processed by microbes are two-fold (Gowri et al., 2014) and extensively reviewed in Sections 2.6.1 and 2.6.2: enzymatic degradation, and biophysical properties such as biosurfactant and bioflocculant activity, and biosorption. Bioremediation of synthetic dyes has been reviewed (Ali, 2010; Khan et al., 2013) and more specifically reactive dyes or azo-dyes (Saranraj 2013; Gowri et al., 2014; Sudha et al., 2014). Bioremediation of PAHs has also been described extensively in the literature (Fan & Krishnamurthy, 1995; Seo et al., 2009; Lu et al., 2011).

However, the real-world test is the bioremediation of pollutants in the environment in WW samples. Although simulation of WW in the laboratory is useful, environmental samples will test whether the bioremediation will be successful in a range of temperatures, pH values, in complex mixtures of compounds and other variables (Ali, 2010). Enzymes from several biological sources including plants, fungi and bacteria, are peroxidases (Chiong et al., 2014) and

phenoloxidases (Duran & Esposito, 2000) that remove phenol from WW. Microbes have been used in textile effluent decolourisation (Ramachandran et al., 2013). Bioremediation of food effluent containing phenols includes fermentation processes of distilleries, breweries and wineries producing dark coloured soluble organic matter (Chiacchierini et al., 2004; Tiwari et al., 2012; Santal & Singh, 2013), olive oil WW (Chiacchierini et al., 2004) and coffee process WW (Torres et al., 2016). Polycyclic aromatic hydrocarbons (PAHs) have been degraded in the environment by enzymes that are different to those degrading dyes and natural coloured matter, i.e. oxygenases (Peixoto et al., 2014), and biosurfactants (Abbas et al., 2015).

An easy and simple colorimetric test for phenols and phenolic compounds uses the Folin- Ciocalteu (FC) reagent. Upon reaction with phenols, it produces a blue color; the concentration of the reduced Folin reagent is therefore measured by absorbance at 765 nm. It is very sensitive, but not specific (Agbor et al., 2014). This lack of specificity can be used as an advantage in complex environmental solutions like WW samples. Everette et al. (2010) tested the FC reagent with >80 different compounds including phenols, thiols, vitamins, amino acids, proteins, nucleotide bases, unsaturated fatty acids, carbohydrates, organic acids, inorganic ions, metal complexes, aldehydes and ketones. The study found reactivity with phenols, proteins, thiols, many vitamins, and ions (iron, manganese, iodine and sulphate). The nucleotide base guanine, and the trioses glyceraldehydes and dihydroxyacetone also showed reactivity. The authors (Everette et al., 2010) suggested that the FC reagent represents total antioxidants in a sample and gives a rough approximation of phenolic content due to this cross reactivity. The FC reagent has been used to measure phenols in rivers and lakes in the UK (Box, 1983), and in bio-oils from the lignin portion of biomass (Rover & Brown, 2014). Takeda et al. (2013) used the FC reagent to measure dissolved organic matter in concentrated river and sea water samples in Japan. It has been used for phenolic concentrations in milk (Vázquez et al., 2015), olive oil WW (Atanassova et al., 2005) and brewery WW (Tatullo et al., 2016). The FC reagent is comparable to liquid-liquid extraction methods in plant extracts (Rover & Brown, 2014) and chemiluminescent assay in food processing WW (Atanassova et al., 2005).

South Africa’s mining activity has always played a major role in the country’s economy ((https://www.brandsouthafrica.com/investments-immigration/economynews/sa-economy-key- sectors). Associated with this is WW as acid mine drainage which has an important effect on water quality lowering the pH due to oxidation of sulphide minerals (Naicker et al., 2003). Heavy-metal pollution of rivers in Thohoyandou (Okonkwo & Mothiba, 2005) and sediments in streams in Port Elizabeth (Binning & Baird, 2001) has been reported. In addition, groundwater resources are also threatened by pollution by agricultural fertilisers and pesticides (Schulz,

2001). Eutrophication and nutrients released in urban waters is also on the increase (Nyenje et al., 2010). As a result, SA’s scarce but most valuable resource is under threat.

The aim of this study was to determine whether the cell-free culture supernatant (CFCS) of these hot-spring isolates was able to reduce turbidity or colouration in food pollutants containing melanoidin, textile pollutants including precursor textile dyes bromothymol blue (BB), a commercial dye (Dye It, CDS1, SA) and three WW samples, namely brewery and dairy industry WW and river water polluted with industrial effluents. Since phenol is a priority environmental pollutant (Galgale et al., 2014), further testing was performed in order to determine whether the phenolic compounds in these samples were reduced.

2.6 MECHANISMS FOR BIOREMEDIATION

2.6.1 Bacterial enzymes

Bioremediation of WW has made use of the growing technology of industrial enzymology, and its advantages over chemical treatments. This includes lower costs, green footprint, decreased toxicity to the environment, ease of application, specificity and flexible applications to a variety of possible substrates. The enzymes specifically involved in WW treatment have been reviewed (Karigar & Rao, 2011; Facchin et al., 2013), and are mainly oxidoreductases (including catalase, peroxidase, azoreductase) and hydrolytic enzymes (including amylase, protease, lipase, glucosidases, phosphatases, cellulase). Amylase and protease make up more than half of economically important industrial enzymes (Kumar et al., 2013) holding the majority of patents submitted. The applications to WW from different industries are numerous and include mining, food industry, textile industry, pharmaceuticals and municipal wastewaters, and are reviewed in Section 2.5.

The microorganisms from extreme environments like hot springs have unusual biochemistry, and function with enzymes that not only can withstand harsh environmental conditions of pH and temperature, but are able to function optimally under such conditions (Demirjian et al., 2001; Baltaci et al., 2017; Demorne et al., 2017). As a result, interest has been shown for valuable and applicable bioactive molecules from such sites (Fooladi & Sajiadian, 2010; Obeidat et al., 2012; Sen & Maiti, 2014; Archna et al., 2015; Verma et al., 2015; Chen & Jiang, 2018). The water temperature of hot springs in Limpopo Province, SA can range from 42 to 68 °C and the pH between 7 and 9 (Olivier et al., 2012) thereby introducing the possibility of obtaining alkaliphilic and/or thermophilic enzymes. Previously mentioned in Section 2.2, metagenomic studies of hot springs in Limpopo have been conducted, showing a great diversity

of bacteria (Tekere et al., 2011; 2012). However, the phylum Firmicutes, genus Bacillus spp. and related bacteria predominate when the bacteria are cultured.

Bacillus spp. are well known for production of enzymes including amylase, and are hardy yet easy to culture, and versatile in applications (Kumar et al., 2013) including WW bioremediation. Several novel species of Anoxybacillus have been cultured from hot-spring environments globally (Mandic-Mulec et al., 2015). Producers of thermostable amylases of Anoxybacillus spp. and Bacillus spp. have been isolated from hot springs in Turkey (Matpan Bekler & Güven, 2014; Fincan et al., 2014; Acer et al., 2016), Pakistan (Asad et al., 2011), Iran (Fooladi & Saijiandian, 2010; Asoodeh et al., 2013) and China (Li et al., 2013; Zhang et al., 2015). Thermostable protease producers have been isolated from hot springs in Turkey (Bekler et al., 2015), Jordan (Al-Qodah et al., 2013), Uganda (Hawumba et al., 2002), Mongolia (Namsaraev et al., 2010), India (Panda et al., 2013) and Indonesia (Wang et al., 2012). It is well established that amylase and protease producers are common in hot-spring environments. Yet new discoveries can be made because enzymes from extreme environments are known to express unique properties (Nigam, 2013). Singular thermophilic Geobacillus sp. from two independent studies of hot springs in Pakistan (Zahoor et al., 2016) and in Tunisia (Thebti et al., 2016) were able to produce several important biotechnological enzymes. In addition, environmental pollutants are becoming more complex in structure and the discoveries of new enzymes with novel specificities are in demand. For example, in studies of the crystal structure of the alpha amylase of Anoxybacillus sp., a new subfamily of glycosyl hydrolase family GH13 was recently discovered which was thermostable and had an improved output of secondary products such as maltose (Chai et al., 2016). Sen and Maiti (2014) described several phosphatases producing Bacillus and Bacillus-related isolates from hot springs in India. A metagenomics study of hot springs in China revealed the presence of a thermostable superoxide dismutase (SOD) gene that was further characterised by cloning (He et al., 2007). Recently, Aneurinibacillus thermoaerophilus isolated from an Indian hot spring, was described producing a thermostable lipase (Sharma et al., 2016), and a thermostable halotolerant cellulose was discovered from an Icelandic hot-spring isolate (Zarafeta et al., 2016). These studies suggest that hot-spring isolates are an untapped resource of uncharacterised enzymes with potentially useful biotechnology applications.

The majority of studies have used conventional plate assays or biochemical tests, but with the introduction of new techniques e.g. liquid chromatography–mass spectrometry (LC-MS), more information about the proteins and enzymes can be obtained in one experiment (Fandi et al., 2012) although it is more expensive. Individual plate assays and biochemical tests only

investigate one enzyme-substrate at a time and do not reveal differences between similar acting enzymes, whilst LC-MS can reveal the presence of several different enzymes in combination all at one time. It is also more sensitive to very small changes of molecular weight of enzymes and can therefore distinguish between different enzymes that may have the same action in a plate assay. Using both low and high cost methodologies to screen and further identify useful enzymes will highlight advantages of both methods.

An indirect application of new microbial enzymes to WW bioremediation is their use in biomonitoring. Biomonitoring is the use of biological systems to detect pollutants and toxins in environmental niches, and has been reviewed in aquatic environments (Li et al., 2010), including the use of microbial enzymes (Logar & Vodovnik, 2007). Biosensors measure toxicity with respect to cell viability and DNA damage, and most commonly use genetically modified bacteria with reporter genes lux (luciferase) and gfp (green fluorescent protein), lacz (beta galactosidase) and phoA (alkaline phosphatase). Biosensor genes that detect geno-toxic and oxidative damage include katG (anti-oxidative enzyme) (Logar & Vodovnik, 2007). Lead (Pb) is highly toxic in the environment because of its ability to mimic biologically important metals and produce membrane damage. Delta-aminolevulinate dehydratase (ALAD) is a conserved metalloprotein in many organisms including bacteria, fishes, amphibians, reptiles, birds, mammals and humans (Konuk et al., 2010), and is very sensitive to Pb. In the search for a bacterial ALAD for potential use as a biosensor for Pb, the ALAD enzyme of Pseudomonas spp. (Korcan et al., 2007) and Pseudomonas putida (Ogunseitan et al., 2000) has been found to be useful for biomonitoring of Pb in contaminated water environments. Gram-negative E. coli and Gram-positive Bacillus subtilis were found to grow less in the presence of atrazine, a component of herbicides, and this was a result of oxidative stress and indicated by the induction of superoxide dismutase (SOD), catalase and glutathione S-transferase, suggesting that these enzymes could be useful indicators of pollution by herbicides (Zhang et al., 2012a). Filimon et al. (2013) studied sediment bacterial communities of water streams adjacent to a copper smelter complex in Serbia and found that copper (Cu), zinc (Zn), Pb and arsenic (As) were very high. Using an enzymatic indicator of sediment quality (EISQ), by measuring total and combined catalase, dehydrogenase, urease and phosphatase activity, EISQ could be used to monitor the level of inorganic pollution in aquatic environments. Whole cells of B. subtilis have been useful in colorimetric bioassays to determine biotoxicity assessment of heavy metals Cu, Pb, cadmium (Cd), Zn and nickel (Ni) in the laboratory and in effluent landfill water samples, electroplating WW and inorganic WW from a chemical laboratory (Fang et al., 2015).

2.6.2 Biophysical characteristics

As reviewed in Section 2.5, increasing human populations globally has resulted in escalating pollution from industry and agriculture including the environmental water systems, threatening natural ecosystems and human health (Schwarzenbach et al., 2010; Chen et al., 2015b). Pollutants include PAHs, petroleum, pesticides, pharmaceutical residues, textiles dyes, heavy metals and residues from food and beverage industries (Chen et al., 2015b; Liu et al., 2017). Many of these pollutants are complex structures and recalcitrant, resisting natural breakdown in the environment. Heavy metals may accumulate in the food chain posing health hazards even at very low concentrations (Chen et al., 2000). Heavy metals are grouped together because they have a density >5 g/cm3, an atomic number >20, and are toxic to humans at low concentrations. The most toxic or poisonous is Pb, Cd, mercury (Hg), and chromium (Cr) (Abbas et al., 2014).

Textile dyes including triphenylmethane dyes, in low concentrations of 1 mg/L affect visibility in water, are also recalcitrant and toxic to biological systems (Khan et al., 2013). Microbes have been fundamental in the treatment of extreme polluted water including WWs as described in Section 2.5. Not only organic or inorganic pollutants are problematic but biofouling due to biofilm formation in drinking and industrial water systems may be a physical hindrance and contribute to poor water quality in colour, odour and taste (Farkas et al., 2012). Biofilms form in water pipes, interfere with underwater electronic gadgets or upset the balance in WWTPs, sloughing off and resulting in regrowth of microorganisms’ post-disinfection (Farkas et al., 2012). Furthermore, but less obviously, biofilms have the ability to harbour waterborne pathogens, and encourage transfer of AR amongst its microbial communities (Proia et al., 2016).

Bioremediation is a process that uses living organisms such as microorganisms or plants for the degradation or transformation of contaminants into nonhazardous or less hazardous substances (Karigar & Rao, 2011; Adams et al., 2015). The process utilises both active metabolism and/or passive cellular properties of the microorganisms to perform such functions, and has many advantages over conventional treatments. Advantages include their eco-friendliness where the microbial inocula biomass is easily produced, the treatment process is simple, efficient and cost- effective (Vidali, 2001). Individual species of bacteria may employ several mechanisms to break down the pollutants, and consortiums of bacteria can be used, allowing for many diverse simultaneous reactions to take place in complex mixtures of different pollutants. The bacterial inoculum can be adapted for specific needs based on the pollutants in question. Often, there is little need for post-treatment to neutralise or remove additional chemicals (Vidali, 2001).

However, there are limitations; for example, chlorinated organic and high aromatic hydrocarbons are resistant to bacterial attack, and therefore have to be dealt with by other means.

Bacteria are very adaptable to their environment and are an enormous natural resource of novel enzymes and biomolecules useful for bioremediation of wastewater (Vidali, 2001). It is the opinion of the investigator that the terminology for the mechanisms used by bacteria in the bioremediation process is sometimes confusing in the literature as discussed in this review, and the terms will therefore be clearly defined for the purpose of this study.

The mechanism of bioassimilation is defined as intracellular accumulation of the pollutant or chemical, which occurs in two stages. The first is rapid adsorption onto the cell surface and a slower transport into the cells. The process is complex and requires metabolically active cells (Chojnacka, 2010). A disadvantage is that the system needs to be monitored for optimal conditions required to keep the bacteria alive and growing. On the otherhand, no biomass production prior to bioremediation is required, but occurs concurrently (Chojnacka, 2010).

Biosorption is defined as the removal of substances from solution, both inorganic and organic moeties, insoluble and soluble, by biological material. It is a physico-chemical process based on “high attractive forces” between the substances and the biomass. This biomass can be both living and none living cells however most studies refer to biosorption as a property of dead cells (Abbas et al., 2014).

There are several reasons why the terminology for biosorption and bioassimilation can be confused in reviewing the literature. Firstly, the bioassimilation process includes a first step of biosorption, and secondly, by definition biosorption is possible with both live and dead cells, unlike bioassimilation that requires exclusively live metabolically active cells (Abbas et al., 2014). The differences are well described in reviews by Chojnacka (2010) and Vijayaraghavan & Yun (2008). Biosorption is a passive process while bioassimilation is an active process; however, the end result of both processes will be the same, expressed as a reduction in pollutant concentration. In the past, there has been some confusion between the terminology, biosorption and bioassimilation, therefore this distinction is important in reviewing the literature. For example, in a laboratory study, Issazadeh et al. (2011) described “bioaccumulation” of heavy metals by Bacillus species. However, the bacterial biomass was grown prior to exposure to the heavy metals, and on microscopic observations, they showed that the cells had not grown or ruptured in the metal solutions. A reduction in heavy metals in the solution was reported. By definition, because there was no distinction between the removal process being active or passive, the use of both terms of bioassimilation or biosorption could be correct. In a study of

Geobacillus sp. and the removal of heavy metals (Ozdemir et al., 2012), the removal of heavy metals by actively growing cells were described as bioaccumulation and the removal by dead cell membranes was described as biosorption. As in the previously mentioned study, a removal of heavy metals by live active bacteria by definition could also be described as biosorption. Francis (1998) used the terminology bioaccumulate and biosorb interchangeably in the study of bacterial removal of uranium in radioactive waste. Both bioassimilation and biosorption can be useful to address heavy-metal pollution in aqueous environments (Zabochnicka-Swiatek & Krzywonos, 2014).

Bioassimilation and biosorption therefore require the presence of microbial cells. However, other mechanisms that utilise extracellular molecules in the CFCS can be produced in bio- fermenters and used independently of bacterial cells to treat polluted waters (Banat et al., 2014; Zhu et al., 2014). These processes include biodegradation by extracellular enzymes (Karigar & Rao, 2011; Facchin et al., 2013), biosurfactants (Das et al., 2009; Thavasi et al., 2011) and bioflocculant (François et al., 2011) production.

Bioflocculation is the coagulation and precipitation of suspended solid particles or heavy-metal ions by natural substances that facilitate the aggregation of particles to form flocs (Zhu et al., 2014; Agunbiade et al., 2016). It is usually an exopolysaccharide that occurs in CFCS unlike biosorption which utilises whole cell biomass. Bioflocculation is sometimes confused with the term “biosorption” because biosorption is also the first step of bioflocculation. However, bioflocculation ultimately results in the formation of heavier bacterial flocs that separate out of the aqueous solution (Czemierska et al., 2015). In this study, the term bioflocculant will refer to the non-cell associated substances that result in the removal of solid particles by floc formation and sedimentation.

Biosurfactants are important for bioremediation of water insoluble pollutants like petroleum, crude oil and PAHs (Pacwa-Plociniczak et al., 2011), food WW (Sanjana et al., 2017). These pollutants can be toxic, carcinogenic and mutagenic for humans, and are not bioavailable for degradation (Souza et al., 2014). Biosurfactants are a diverse group of molecules that lower surface and interfacial tensions of liquids, form micelles and thereby enhance the solubility of poorly soluble compounds in aqueous environments (Al-Araji et al., 2007). They are amphipathic compounds of biological origin containing a hydrophilic region and a hydrophobic region (lipid or fatty acid) (Santos et al., 2016). Biosurfactants have been classified according to their chemical composition and microbial origin being both low molecular weight (often glycolipids) (Liu et al., 2015a) and high molecular weight (polyanionic heteropolysaccharides) (Rahman & Gakpe, 2008; Singh, 2012).

By their nature, some biosurfactants are able to interfere with biofilm formation either by exhibiting antimicrobial, anti-adhesive or biofilm disrupting properties (Banat et al., 2014). However, not all anti-biofilm agents are biosurfactants. Enzyme degradation due to protease and amylase of the extracellular polysaccharides around the cell walls may also manifest as anti- biofilm agents (Gautam et al., 2013).

Bioremediation therefore can be divided into two main mechanisms: Firstly, degradation or biotransformation where the pollutant is broken down into smaller components or changed from a toxic to a nontoxic form or from a non-biodegradable form to a biodegradable form, respectively (Adams et al., 2015). This can be both intracellular requiring active transport system into the cell (bioassimilation), or it can be extracellular, for example the production of enzymes (Karigar & Rao, 2011; Facchin et al., 2013) or biosurfactants (Liu et al., 2015a). Secondly, there are mechanisms that only require the passive removal and transport of the pollutant out of solution through biosorption or bioflocculation.

Table 2.2: Summary and comparison of terminology used in bioremediation

Biosorption Bioassimilation Bioflocculation Biosurfactant Energy required Passive Active Passive Passive Location of Live or dead activity cells Live cells Extracellular Extracellular Processes Two steps, including Two steps, including One step biosorption biosorption One step Pollutants Heavy metals Heavy metals PAHs Dyes Dyes Heavy metals Petroleum

A diverse group of bacteria both Gram-negative and Gram-positive are involved in bioremediation of the environment. Gram-positive Bacillus and Bacillus –related species have been cultured in hot springs (Section 2.2) and are ideal candidates for biotechnological applications (Kumar et al., 2013). Bacillus spp. have been isolated from hot springs in Pakistan (Ghalib et al., 2014) and in Jordan (Al-dayhistani, 2012) for their biosorption capacities. Biosurfactant-producing Anoxybacillus species have been isolated and characterised from hot springs in Thailand (Pakpitcharoen et al., 2008) and Malaysia (Khairuddin et al., 2016). Appendix 2 summarises studies of Bacillus spp. and Anoxybacillus spp. which have bioflocculant, biosorption, biosurfactant and anti-biofilm properties.

This study will screen bacterial isolates from hot springs for biophysical characteristics that may be useful for WW bioremediation, with applications in addressing the current pollution

problems relevant to SA. This may lead to a cost-effective, eco-friendly and simple way of addressing WW in SA.

2.6.3 Naturally occurring antimicrobials

Widespread over-use and indiscriminate use of antibiotics to treat infections in humans and animals, control plant diseases, aquaculture and even as growth promoters in domestic stock animals has resulted in different antibiotics, and higher levels thereof, present in the environment (Section 2.4). Bacteria have developed resistance resulting in the emergence of MAR, and are now considered as “new age” pollutants in groundwater and surface water (Martinez, 2009b; Sanderson et al., 2016).

Because AR levels have increased dramatically within the human pathogen populations, it has rendered the previous advantage of the treatment with antibiotics severely ineffective (Al-Bahry et al., 2014). Both the Centers for Disease Control (CDC) and the World Health Organization (WHO) have declared that AR is a threat to USA public health and a national security issue (Ventola, 2015). In 2011 a survey of clinicians in the USA reported that >60% had seen a pan- resistant, untreatable bacterial infection within the prior year (Ventola, 2015). Methicillin- resistance Staphylococcus aureus (MRSA) was first identified five decades ago. Since then it has spread worldwide and in the USA, with 11 285 deaths annually are attributed to MRSA (Ventola, 2015). Vancomycin was introduced in 1972 for the treatment of MRSA infections, with resistance first reported two years later. Currently 20 000 (30%) of hospital-acquired enterococcal infections per year are due to vancomycin resistance (Ventola, 2015).

The discovery of novel natural antimicrobial drugs is critical to alleviate the burdens of infections caused by AR pathogens. Natural bacterial antimicrobial substances fall into three groups. Firstly, large complex molecules produced by multi enzymatic complexes are naturally occurring antibiotics. Secondly, ribosomally synthesised peptides called bacteriocins are produced by many microorganisms including Archaebacteria, Gram-negative and Gram- positive bacteria. Bacteriocins can be further differentiated into three classes as described in detail by Güllüce et al. (2013), and defined as peptides that are bacteriostatic or bactericidal against other related and unrelated microorganisms. Bacteriocins are about 30-60 amino acids (aa) long and act against a narrow spectrum of bacteria compared with antibiotics. The third group are non-ribosomonally synthesized peptides or bioactive peptides are different from bacteriocins. They are generally smaller in size (3-20aa) and are released by proteolysis with hormone-like activity (Güllüce et al., 2013). Since bacteriocins are between 30-60aa, antimicrobials produced by Bacillus spp. have been correctly classified as bacteriocins of 44aa

(Kindoli et al., 2012), 45aa (Lim et al., 2016) and 82aa (Naclerio et al., 1993). However, smaller molecules have been described as bacteriocin or bacteriocin-like by Aunpad et al. (2011) (2.5 kDa – 23aa) and Kaewklom and Aunpad (2012) (3.14 kDa – 28aa). Since the average molecular weight of an amino acid is 110 Da, the protein size of 3.3 kDa would be the conversion of 30aa, the size cut-off for bacteriocins. The main advantage of antimicrobial peptides (AMPs) over antibiotics therapy is that resistance is less likely to develop. This is because AMPs function by disrupting cell membranes resulting in osmotic lysis. The mode of action is determined by the composition of the cell wall (Hartmann et al., 2010). Gram-positive, Gram-negative and acid- fast bacteria all have different wall structures and this is fundamentally difficult to modify, unlike the specific molecular targets of antibiotics (Falanga et al., 2016). Examples of such target of antibiotics are quinolones that target DNA and chloramphenicols and tetracyclines that affect the ribosomes (Blair et al., 2015).

Sometimes AMPs have biosurfactant properties. Biosurfactants are amphipathic molecules with both hydrophilic and hydrophobic moieties that are able to reduce surface tension (Section 2.6.2). For example, a soil bacterial isolate that produced biosurfactants was also antimicrobial against human pathogens E. coli, S. aureus and Pseudomonas aeruginosa (Sarin et al., 2011). Similarly, biosurfactants of Lactobacillus and Bacillus inhibited the growth of Pseudomonas, E. coli, Salmonella typhi, Staphylococcus and Aspergillus (Sharma et al., 2014b). Based on their mode of action, it is therefore not surprising that biosurfactants can be antimicrobial, and that bacteriocins or bioactive peptides are biosurfactants (Harshada, 2014).

Bacillus are popular “work horses” in biotechnology (Kumar et al., 2013; Chen & Jiang, 2018) because they are regarded as non-pathogenic and non-toxic, categorised as “generally regarded as safe” (GRAS) by the United States Food and Drug Administration (USFDA) and therefore important in food preservation or as probiotics. They are ubiquitous in the environment, robust, easy to culture and produce good yields of enzymes. They are also prolific producers of a diverse array of biomolecules useful in food preservation (Nath et al., 2015). Bacillus and Bacillus-related bacteria have been cultured from hot springs (Section 2.2) and are potential producers of specialised enzymes (Section 2.6.1) and other bioactive molecules (Section 2.6.2) expressing heat, pH and salt tolerance. Several isolates from hot springs have been reported to have antimicrobial activity including Streptomyces in Saudi Arabia (Al-Dhabi et al., 2016), Geobacillus in Jordan (Alkhalili et al., 2015) and cyanobacteria in Oman (Dobretsov et al., 2010). In a study of 148 Bacillus-related isolates from hot springs in Thailand, three isolates produced the highest biosurfactant activity. The CFCS of these isolates exhibited variable

inhibitory effects against four bacterial and four fungal test organisms, however no set patterns could be observed (Pakpitchareon et al., 2008).

The most studied antimicrobial of Bacillus are the bacteriocins of which >45 different types have been described and are produced by a number of different species of bacteria (Sumi et al., 2015). They are classified according to chemical structure, heat stability, molecular mass, enzyme sensitivity, modification of amino acids and mode of action (Sumi et al., 2015), and most often in the context of food preservation (Baruzzi et al., 2011; Nath et al., 2015) and biocontrol of plant pathogens (Nagórska et al., 2007). For example, Bacillus cereus and Geobacillus bacteriocin inhibited other strains of B. cereus (Naclerio et al., 1993) and Bacillus licheniformis (Martirani et al., 2012), respectively. Bacillus pumilus inhibited mycobacteria (Hassi et al., 2012) and several bacteriocins controlled phytopathogens (Hammami et al., 2008; Berić et al., 2012).

Subtilin is a class 1 post-translationally modified bacteriocin or lantibiotic first described from B. subtilis (Jack et al., 1995). It is a 32aa cationic, pentacyclic AMP that is stable to acid and heat treatment, inhibiting a broad range of Gram-positive bacteria including other species of Bacillus (Baruzzi et al., 2011). Antimicrobial activity of AMPs has been reviewed by Sumi et al., 2015. Subtilisins are extracellular alkaline serine proteases that are involved in the processing of subtilin coded for by genes “apr” (Corvey et al., 2003). There are over 200 types of serine proteases which fall into six classes, of which 170 have been completely sequenced (Siezen & Leunissen, 1997). Subtilisin, together with non-ribosomal AMPs iturin and fengycin, has been reported to be antimicrobial (Wu et al., 2013a). Although surfactin is the most well known biosurfactin of Bacillus spp., subtilisin has been listed as a biosurfactant of Bacillus (Al- Araji et al., 2007; Sineriz, 2011). Iturin, fengycin and surfactin are the common smaller lipopeptides of Bacillus spp. that are both biosurfactant and also antimicrobial (Sumi et al., 2015).

Bacillus mojavensis, a close relative of B. subtilis, was previously classified as Bacillus axarquiensis (Ruiz-Garcia et al., 2005), and initially isolated from desert soil (Roberts et al., 1994). However, it is currently better known as an endophyte and useful in biocontrol of phytopathogens (Ghanney et al., 2016) and has been isolated from food products (Moe et al., 2015). B. mojavensis is useful for biocontrol of fungal mycotoxin producer Fusarium moniliforme (Ma & Hu, 2015), Pseudomonas savastanoi causing olive knot disease (Ghanney et al., 2016), and food poisoning pathogens, Listeria (Moe et al., 2015) and Vibrio parahaemolytics (Liu et al., 2015b). B. mojavensis, like the other Bacillus species, produces

subtilisin BM1 (Haddar et al., 2009), and iturin, fengycin and surfactin (Ma & Hu, 2015; Jasim et al., 2016). These molecules have also been with associated biosurfactant activities.

Besides the use of these bacteriocins for controlling plant pathogens (Shafi et al., 2017), they are also the “new age” antimicrobials against AR bacteria, and have good applications as therapeutic agents in the medical sector (Hassan et al., 2012). Some studies have just reported antimicrobial activity against panels of reference microorganisms, but others have included the AR profiles of the panel reference microorganisms including E. coli, S. aureus (MRSA) and Pseudomonas (Kaewklom & Aunpad, 2012; Sambanthamoorthy et al., 2014). Studies have focused on activity against foodborne pathogens (O’Connor et al., 2015) including Enterobacter (Cronobacter) (Tene et al., 2014), and antibiotic induced nosocomial diarrhoea due to Clostridium difficile (Mathur et al., 2014). The variable effectiveness of AMPs is associated with the composition of the outer cell wall layer, stress response, presence of efflux pumps and biofilm formation (Russell, 2003). Bacteriocidal activity is dependent on the cell wall permeability to the biocide (Lambert, 2002). Differences in the affective target organism groups can reveal the pathway and mode of action of AMPs (Falanga et al., 2016).

Bacillolysins are neutral metalloproteases of Bacillus spp. that are useful in food taste production, cell disassociation and as bio-insecticides. They are a diverse group of robust enzymes that require no further genetic manipulation before application; however, their use in industry has been low key (Ou & Zhu, 2012). They play a role in virulence factors of pathogens and have fibrinolytic activity (Narasaki et al., 2005). Their role as antimicrobials is not reported; however, bacillolysins are also biosurfactants (Zhang et al., 2010), and production of neutral proteases has been reported from an isolate of hot springs in Jordan (Youcef-Ali et al., 2012).

In this study, bacterial isolates from hot springs were screened for biosurfactant activity and antimicrobial activity against human pathogens and against a panel of AR environmental bacterial isolates from the same hot springs. Identification of the antimicrobial-producing bacteria and relevant associated biomolecules will be further elucidated.

CHAPTER THREE: MATERIALS AND METHODS

Figure 3.1: Flow chart of summary of methods used in study 3.1 SAMPLING AND SAMPLING SITES

3.1.1 Sampling from hot springs in Limpopo Province, SA

Due to the fact that the bioactive molecules produced by thermophilic bacteria are sought after for their potential applications in biotechnology, the hot springs with a cut-off water temperature of 43 °C were prioritised. Secondly, permission for sample collection was granted only at five hot-spring locations, being Tshipise (T), Siloam (S), Mphephu (M), Lekkerrus (Le) and Libertas (Li), all located in the Limpopo Province, SA and these were therefore the sites selected for this investigation. The global positioning system (GPS) coordinates for all the locations are listed in Table 3.1 and photographs are given in Figure 3.3. Temperature and pH were measured in situ using the portable YSI Professional Plus Multiparameter Water Quality Instrument (Xylem, USA). At all the sites, the dissolved oxygen (DO) concentration was below the limit of detection of the probe and could therefore not be determined.

With the exception of Siloam, the other springs have been developed into resorts for human recreational purposes. Samples were taken in the spring of 2014 (September 2014). Water samples were collected from each sampling site in sterile 1 L Schott Duran® glass bottles by hand approximately 30cm below the water surface of the sites closest to the source (Tshipise,

Mphephu, Libertas) and sediment samples in sterile plastic 50 mL Falcon™ tubes and processed separately. All the samples were taken directly from the spring water that had not been in contact with any human activity, except for sampling at Lekkerrus where the water was collected from the pipe that conveyed water directly into the swimming pool and flow was manually controlled depending on demand (Figure 3.3E). Water was sampled directly from the pipe as indicated in Figure 3.3A. As a result, no sediment sample was collected at Lekkerrus. Samples were transported to the laboratory in a cooler box kept at 4 °C, and processed within 72 h of collection.

Table 3.1: Geographical location, site description, pH, temperature and heavy-metal concentrations of samples collected from hot springs in Limpopo Province, SA

Sampling Temperature Cr * Cu * Mn * Ni * Pb * Hg * site GPS location pH (°C) Comments (µ/L) (µ/L) (µ/L) (µ/L) (µ/L) (µ/L) WHO/ EU limits for heavy metals 50 2000 500 20 10 1 SA limits for heavy metals (SABS) 100 1000 100 150 20 1 22° 36.521'S Open air Tshipise 30°10.345'E 8.63 55.2 enclosed section 0.7 0 0 37.19 0.08 0.33 22° 53.667'S 30° Pipe on private Siloam 11.7718'E 9 69 property 0.97 0 0.75 0 0.05 0.53 22° 54.225'S 30° Open air Mphephu 10.83'E 7.07 42.4 enclosed section 1.2 0 0 0 0.16 0.23 24° 28.04'S Lekkerrus 28°33.1' E 7.46 43.5 into pool nd nd nd nd nd nd 24°27'36"S Water pumped Libertas 28°34'11"E 7.44 52.1 at source nd nd nd nd nd nd

*Heavy-metal concentrations (µg/L) from Olivier et al. (2011). Cr (chromium), Cu (copper), Mn (manganese), Ni (nickel), Pb (lead) and Hg (mercury)

Figure 3.2: Geographical location of sampling sites in Limpopo Province, SA, indicated on Google map. 1Tshipise, 2Siloam, 3Mphephu, 4Libertas and 5Lekkerrus

Figure 3.3: Photographs (copyright J Jardine) of sampling sites A: Siloam; B Tshipise; C: Mphephu; D: Libertas; and E: Lekkerrus

3.1.2 Sampling of brewery and dairy wastewaters and river water contaminated with coloured industrial effluents in Gauteng Province, SA

The GPS coordinates of three different WW sites are indicated in Table 3.2. Volumes of 3-5 litres were collected in clean non-sterile plastic bottles by hand on site. Brewery WW was obtained at the last step of the process prior to being discharged down the drain (Figure 3.4). The river sampling site in Modderfontein, Johannesburg selected for sampling was earmarked as a possible site of industrial dye contamination as part of the Edenvale RiverWatch program (Dr Irwin Juckes, personal communication; http://www.edenvaleriverwatch.co.za/author/irwin/). Figure 3.5 shows pictures taken from a stationary video camera taken over time from this

particular storm drain sample site (copyright and permission from Dr I Juckes). The industrial source of this contaminant has not been identified. Since the “coloured” pollutant did not consistently enter the river but occurred in batches, sampling was done at the peak of such “colour run” from the water flowing out of the storm drain into the river rather than from stagnant water. The white coloured pollutant sunk to the bottom of the riverbed and appeared to be denser than water before it flowed further downstream (Figure 3.6). Wastewater from a local dairy was sampled at the outlet prior to discharge (Figure 3.7). However, prior to sampling, the overlying fat layer was separated and removed at another site preceding the outlet. All the water samples were tested with swimming pool Insta-TEST 6T La Motte dipsticks for chlorine, alkalinity, pH, total hardness and cyanuric acid. Samples were transported in a cooler box, stored at 4° C and tested within 72 h of collection.

Table 3.2: Location, GPS coordinates and physicochemical parameters of wastewater sampling sites in Gauteng Province, SA

Type of Chemical River sampling wastewater Parameters Dairy site Brewery GPS 26°S 2" 52,44" 26°S 6' 26.145" 26°S 1' 23.714" coordinates 28°E 0' 39.11" 28°E 10' 3.668" 27°E 56' 44.819" Geographic Founders Hill, location Fourways Modderfontein Kyasands Douglasdale Supplier Dairy na The Beer Keg FCl free Cl (mg/L) 0-0.5 0 0 TCl total Cl (mg/L) 0.5-1 0,5 0 alka alkalinity (mg/L) 180 120 0 pH 7.2-7.8 7.8 4 total hardness TH (mg/L) 100 100 100 cyanuric acid cyA (mg/L) 0 0 <0 Colour white white brown

Figure 3.4: Sample of beer processing wastewater from The Beer Keg, Johannesburg

Figure 3.5: Time course photographs taken on 8 February 2016 (photos copyright, permission from Irwin Juckes) showing a stormwater drain delivering regular input of coloured pollutant at a single site in the Modderfontein River

Figure 3.6: Sampling site for industrial coloured wastewater entering the river from a stormwater drain in Modderfontein, Johannesburg

Figure 3.7: Sampling site at dairy wastewater outlet of Douglasdale Dairy, Johannesburg

3.2 ISOLATION OF BACTERIA FROM WATER AND SEDIMENT

A 100 mL sample aliquot was passed through a 0.22 µm membrane filter (Millipore™ nitrocellulose GSWP04700) and the membrane filters were then placed on the surface of the following four different agar media (Himedia, India) as per manufacturer’s instructions: nutrient agar (NA), Actinomycete isolation agar, minimal Luria agar media(10% LA), and potato dextrose agar (PDA). Minimal Luria agar media was prepared as 10% of the concentration recommended by the manufacturer. The filters were removed before the plates were incubated aerobically for 48 h at 37 °C and 53 °C. Four plates per media were thus seeded with 100 mL/plate in quadruplicate. Bacterial concentrations were expressed as colony-forming units (CFU) per 100 mL (each Petri dish represented by a membrane filter through which 100 mL of sample water was filtered).

Bacterial isolates from sediment samples were obtained using the streak plate method. The small volume of water from the wet sediment was sampled after shaking, with an inoculation loop and streaked onto media for isolation of single colonies. All colonies with distinct and different morphology were sub-cultured at least three times until a pure culture was obtained and cultures were stored on NA slants at 4 °C.

3.3 DETERMINATION OF OPTIMAL TEMPERATURE, pH AND SALINITY FOR GROWTH OF THERMOPHILES

Pure cultures of the isolates that grew at 53 °C were studied for optimal conditions of growth relating to temperature, pH and salinity. The optimum pH for growth of the isolates at 53 °C

was determined, by growth in nutrient broth (NB) from pH 6 to 10 in intervals of one unit. A bacterial suspension at an optical density (OD) at 600 nm (OD600) of approximately 0.3 was made, and 1 mL volumes of NB were inoculated with 10 µL of the bacterial suspension, and then incubated for 24 h at 53 °C. The OD at 600 nm was measured using a spectrophotometer (Phillips PU8620 UV/VIS/NIR) to determine whether growth had occurred. Thereafter the optimum temperature for growth was determined for the same isolates by the method described above, and incubating these samples at temperatures of between 45 and 70 °C at temperature increments of 5 °C. Optimal salinity was determined in the same way, by inoculation of NB containing sodium chloride (NaCl) at concentrations ranging from 0 to 15% w/v.

3.4 DNA EXTRACTION PROTOCOL AND 16S rDNA SEQUENCING

The DNA was extracted by re-suspending bacterial colonies in sterile phosphate buffer saline, then boiled for 10 min and centrifuged at 10 000 rpm for 10 min (Dashti et al., 2009). The supernatant was used for PCR-based amplicon Sanger sequencing. Universal bacterial primers 8F (5’-AGAGTTTGATCCTGGCTCTCAG), 27F (5’-AGRGTTTGATCMTGGCTCAC), and 1472 R (5’-GGTTACCTTGTTACGACTT) as listed by Galkiewicz & Kellogg (2008), and 357F 5’-CTCCTACGGGAGGCAGCAG (Turner et al., 1999) for the 16S rDNA were obtained from Inqaba Biotechnology, SA, and used in different combinations in order to obtain a PCR product for direct sequencing. The PCR tube per sample contained 9.5 µL of water, 12.5 µL of PCR Master Mix (2x) (Thermo Scientific DreamTaqTM Green K1081), 1 µL of each primer at 10 µM, and 1 µL of genomic RNA at 1-10 ng determined by OD 260:280nm on NanoDrop Hach 6200 Spectrophotometer. The PCR machine, Bio-Rad MyCycler, was used with the thermal cycle profile as follows: initial denaturation at 94 °C for 5 min; 40 cycles of 94 °C for 30 s; 50 °C for 30 s; and 72 °C for 60 s; followed by 72 °C for 10 min for the final extension; and held at 4 °C until the machine was switched off. The PCR products were run on a 1% agarose gel in TAE buffer at 80-100V for 60 min together with molecular weight markers (Thermo Scientific SM1113 middle markers), stained with ethidium bromide and visualised and photographed with Bio-Rad Gel Doc™ EZ Imager.

The PCR products were cleaned enzymatically before sequencing with Big Dye Terminator 3.1 cycle sequencing kit (Applied Biosystems Inc, ABI) according to the manufacturer’s instructions, and run on the ABI capillary sequencer at the African Centre for DNA Barcoding (ACDB), University of Johannesburg. If primer 8F failed, only partial sequencing was obtained. A contiguous sequence was constructed with forward and reverse sequencing data resulting in a fragment of approximately 1 400bp, with DNA Baser Sequence Assembler v4 (2013) (Heracle BioSoft, www.DnaBaser.com). Sequences were compared with those in the NCBI database

(GenBank) using the Basic Local Alignment Search Tool (BLAST) (McGinnis & Madden, 2004), and EzTaxon-e (Kim et al., 2012). Isolates with a >99% match to the published sequences were identified to the species level, and those with a >97% match were identified to the genus level (Yarza et al., 2014).

3.5 GENBANK ACCESSION NUMBERS

3.5.1 Phylum Firmicutes

The 16S rDNA sequences of hot-spring isolates from SA were allocated accession numbers and deposited in GenBank as indicated in Table 4.1 (Section 4.3.2). Strains of Anoxybacillus spp. have accession numbers: MF037806, MF037807, MF037808, MF037809, MF037810, MF037811, MF037812 and MF037813. Strains of B. licheniformis were assigned numbers: MF037814, MF037815, MF037816, MF037817, MF037818, MF037819, MF037820, MF037821 and MF037822. Accession numbers: MF037827, MF037828, MF037829, MF037830, MF037831, MF037832, MF037833, MF037834, MF037835, MF037836, MF037837, MF037838 were given to B. subtilis strains. Bacillus spp. have GenBank accession numbers: MF038049, MF038050, MF038051 and MF039084. Single isolates were assigned the following accession numbers: B. pumilus MF038052, B. panaciterrae MF038053, B. methylotrophicus MF038054, Solibacillus sp. MF039085 and Aneurinibacillus sp. MF040218. MF038055, MF038056, MF038057 and MF038058 were numbers allocated to four Brevibacillus spp. All reference strains used for analyses are listed in Appendix 3 together with their associated GenBank accession numbers.

3.5.2 Phyla Actinobacteria and Proteobacteria

The 16S rDNA sequences of hot-spring isolates from SA were allocated accession numbers and deposited in GenBank as indicated in Table 5.2 (Section 5.3.2). Strains of Actinobacteria, Kocuria sp. and Arthrobacter sp. have accession numbers MF120234 and MF120235, respectively. Two alpha-Proteobacteria isolates were assigned numbers MF120236 and MF120239. Accession numbers: MF120227, MF120228, MF120229, MF120230, MF120231, MF120232, MF120233 and MF120237 were given to beta-Proteobacteria including Gulbenkiania mobilis, Cupriavidus gilardii, Tepidimonas fonticaldi and Ralstonia mannitolilytica. Four strains of gamma-Proteobacteria including Hafnia alvei have GenBank accession numbers of MF144571, MF144572, MF144573 and MF120238. All reference strains

used for analyses are listed in Appendix 4 together with their associated GenBank accession numbers.

3.6 PHYLOGENETIC ANALYSIS

Sequences of type strains were obtained from GenBank and included in phylogenetic analysis to confirm identification. Alignments were made by Clustal Omega (www.ebi.ac.uk), and manually refined using SeaView (Gouy et al., 2010). Phylogenetic analyses were performed for the phylum Firmicutes genera Bacillus and Bacillus-related bacteria (Chapter 4), for the phyla Actinobacteria and Proteobacteria (Chapter 5) and for the identification of an isolate (76S) that had antimicrobial properties (Chapter 10).

The outgroup in the Firmicutes tree was Pseudomonas aeruginosa ATCC 23993. Neighbour- joining (N-J) phylogenetic trees of a 914 bp fragment were drawn with SeaView (Gouy et al., 2010). Neighbour-joining phylogenetic trees of a 947 bp fragment for Actinobacteria and a 510 bp fragment for Proteobacteria were also constructed using SeaView. Methane-producing Archaea or methanoarchaea (GenBank DQ372975.1) was the outgroup in the Proteobacteria tree, while the Actinobacteria tree was unrooted for better resolution. Since fewer isolates were analysed to identify the unknown 76S, an unrooted maximum parsimony method was used to create the phylogenetic tree.

In all the constructed trees, statistical confidence in branching points was determined by 100 bootstrap replicates. Complete and partial sequences from this study were submitted to GenBank (Section 3.6). The GenBank accession numbers of the type strains used in the phylogenetic trees are listed in Appendix 3 (Chapter 4), Appendix 4 (Chapter 5) and Appendix 5 (Chapter 10).

3.7 COMPUTER-SIMULATED PCR-RFLP OR AMPLIFIED rDNA RESTRICTION ANALYSIS (ARDRA)

Computer-simulated polymerase chain reaction-restriction fragment length polymorphism (PCR-RFLP) patterns were generated from the approximately 1 400 bp fragment of the 16S rDNA amplicon (using the computer program RestrictionMapper version 3 “www.Restrictionmapper.org”) and restriction enzyme, HaeIII, and the results used to create a binary data file. Several bacterial strains from published data were included in the study, in order to determine the phylogenetic groups into which the isolates fell (Appendix 3). The

SeaView program was used to analyse the binary data and an N-J tree was created for graphic representation of clusters A, B, C, D and E.

3.8 GUANINE-CYTOSINE (GC) CONTENT

The GC content (in percentage) for the Firmicutes group was calculated with the 1 400 bp 16S rDNA sequences (Yamane et al., 2011) using the ENDMEMO GC calculating tool (www.endmemo.com/bio/gcratio) for all the isolates as described in Chapter 4. Reference strains and their GenBank accession numbers used for this analysis are listed in Appendix 6.

3.9 DETECTION OF LEGIONELLA BY REAL-TIME POLYMERASE CHAIN REACTION

A 300 mL aliquot of hot-spring water was filtered through a 0. 22 µm filter, washed off, centrifuged at 10 000 rpm, and DNA extracted with Zymo fungal/bacterial DNA MiniPrep kit (D6005) (Inqaba Biotec). The presence of Legionella spp. in the water samples was investigated using real-time PCR (Wellinghausen et al., 2001) on a Corbett Life Science Rotor-Gene™ 6000 Cycler (Qiagen, Hilden, Germany). Primers used in the PCR reaction were JFP (5´-AGG GTT GATAGG TTA AGA GC-3´) and JRP (5´-CCA ACA GCT AGT TGACAT CG-3´) (75). The PCR reaction was run in a total volume of 20 µL consisting of 10 µL 2 x SensiFAST™ HRM mix (Bioline GmbH, Germany), 0.2 µL of each primer (each at a final concentration of 0.2 µM), 5.6 µL of nuclease-free water and 4 µL of template DNA. Template DNA from L. pneumophila ATCC® 211-33-2 was used as positive control while the negative control had no template DNA. To test for the detection limit of the PCR reaction, the DNA concentration of the positive control was determined using a NanoDrop Hach 6200 Spectrophotometer, and then serially diluted up to 10-6. The samples and the positive control were run in triplicate.

Reaction conditions for the PCR were optimised as follows: initial activation at 95 °C for 10 mins; denaturation at 95 °C for 5 s; annealing at 57 °C for 5 s; and a final extension at 72 °C for 5 s, for a total of 40 cycles. The last cycle was followed by a second incubation period of 72 °C for 5 min. The second incubation was followed by a melt curve prepared by ramping up the melting temperature from 72 °C to 95 °C at a ramp rate of 1 °C at each step, a pre-melt hold of 90 s on the 1st step followed by a 5 s hold on each of the next steps. A Rotor-Gene™ 6000 Cycler (Corbett Life Science (Pty) Ltd, Sydney, Australia) was used to run the PCR assays. Reaction mixtures without template DNA were used as No Template Controls (NTC) in

each reaction. Melt curve analyses were performed using the Rotor-Gene™ Real-Time Analysis Software, Version 6.1 (Build 93) (Corbett Life Science (Pty) Ltd, Sydney, Australia).

3.10 ANTIBIOTIC RESISTANCE/SUSCEPTIBILITY ASSAY

Resistance against ten antibiotics representing six different classes: β-lactam (carbenicillin), aminoglycosides (gentamicin, kanamycin, streptomycin), tetracycline, amphenicols (chloramphenicol, ceftriaxone), sulphonamides (co-trimoxazole) and quinolones (nalidixic acid, norfloxacin) was tested using the Kirby-Bauer disk diffusion assay (Bauer et al., 1966). Commercial antibiotic disks (Oxoid, Basingstoke, UK) were used at the following potencies in µg/mL: gentamicin 10, tetracycline 30, co-trimoxazole 25, chloramphenicol 30, ceftriaxone 30 and norfloxacin 10. Carbenicillin 100 µg/mL (Sigma Aldrich C1389), kanamycin 30 µg/mL (Sigma Aldrich K4000), streptomycin 10 µg/mL (Sigma Aldrich S6501) and nalidixic acid 125 µg/mL (Sigma Aldrich N4382) were prepared from stock solutions, and dried on sterile Whatman No. 17 disks before placement on a lawn of bacteria. The different concentration of antibiotics used, was determined by recommended values listed in Clinical and Laboratory Standards Institute (2013). The inoculum was prepared from an overnight culture in NB, using a McFarland standard of 0.5. The disks were applied to lawns of bacteria on Mueller Hinton agar (HiMedia, India). Plates were scored after 24 h at 55 °C or 37 °C depending on the isolate. Antibiotic resistance was determined by measuring the inhibition zones around the antibiotic paper disks; resistance was described as an absence of inhibition while sensitivity was scored as a zone of inhibition.

Multidrug resistance (MDR) is defined as acquired non-susceptibility to at least one agent in three or more antimicrobial categories (Magiorakos et al., 2011). Multiple antibiotic resistance (MAR) will refer to MDR in this study. The multiple antibiotic resistance (MAR) index for each isolate was calculated as a ratio between the number of antibiotics to which the isolate is resistant to the total number of antibiotics against which the isolate was tested. Therefore an isolate that is resistant to all the antibiotics tested would have an MAR index of 1 in this study (Matyar et al., 2014; Kimiran-Erdem et al., 2015). The zone of inhibition was measured after a 24 h incubation period at 53 °C (thermophiles) or 37 °C (mesophiles).

3.11 HEAVY-METAL TOLERANCE ASSAY

Tolerance of isolates to heavy-metal salts was determined by disk diffusion assay (Belliveau et al., 1991). The following metal salts were used: aluminium sulphate (Al), chromium IV oxide (Cr), copper sulphate (Cu), iron sulphate (Fe), mercury chloride (Hg), manganese sulphate (Mn), nickel chloride (Ni) and lead nitrate (Pb). All metal salts were purchased from Saarchem Pty Ltd, SA.

Sterile heavy-metal solutions were dried onto autoclaved filter paper disks (Whatman No. 17) at concentrations of 10 mM and 40 mM per disk for all the metal salts except for mercury, which was tested at a concentration of 200 nM per disk. The bacterial inoculums were prepared as described above for the antibiotic resistance assay, before seeding plates of Mueller Hinton agar. Zones of inhibition were measured after a 24 h incubation period at 55 °C or 37 °C as required, where no zones of inhibition were scored as resistance.

3.12 DETECTION OF BACTERIAL ENZYMES

3.12.1 Plate assay for screening of potential enzymes (amylase, protease, lipase, pectinase, gelatinase, azoreductase and laccase) for bioremediation

Hydrolytic activities of pure cultures were tested qualitatively using the disk diffusion procedure on agar plates containing the various substrates: starch for amylase, skim milk for protease, pectin for pectinase (Usha et al., 2014), gelatine for gelatinase, and Tween 80 (Samad et al., 1989) or olive oil (Lee et al., 2015) for lipase, where a clearing around the colonies after growth at 37 °C or 53 °C for 24-48 h would indicate a positive presence of enzymes. The recipes for the media are listed in the Appendix 7. The degradation of starch was detected by flooding the plate with Lugol’s iodine solution post-incubation to observe the clearing zones around the colonies, whilst phenol red was included in the media with olive oil to indicate a change in pH.

For the detection of azoreductase, methyl red in NA (Leelakriangsak & Borisut, 2012) was used. For the detection of laccase, guaiacol (Sheikhi et al., 2012) or bromothymol blue (Tekere et al., 2001) NA was used, again with a clearing denoting positive for enzyme production (Appendix 8).

3.12.2 Concentration of amylase produced

Because a liquid assay was used to determine starch concentration, a standard curve was made with soluble starch (boiled for 1.5 h) 1% stock solution starting from 0.00001% to 0.1%, stained with 10% Lugol’s iodine. The OD was read at 660 nm in a Bio-Rad 96-well iMark microplate absorbance reader. The detailed method and standard curve are described in Appendix 9.

Three thermophilic isolate amylase producers were selected (see Chapter 7, Table 7.1), that had a range of clearing widths on starch agar plates of 3 mm (isolate 9T), 5 mm (isolate 20S) and 10 mm (isolate 13S). Conical flasks were inoculated in triplicate with the selected isolates 9T, 13S and 20S into 50 mL 1% starch broth and incubated at 53 °C for 24 h without agitation. A static experimental set up was chosen due to lack of equipment. The negative control was starch broth that was not inoculated. Supernatants were serially diluted and tested for starch after 24 h as indicated in the liquid assay described in Appendix 9.

3.12.3 Gravimetric assay for the detection of cellulase

The presence of cellulase in CFCS was determined gravimetrically with any loss of weight (mg) of cellulose filter paper (Whatman No. 17) post incubation with CFCS for a period of 24 h being a positive result (Shuangqi et al., 2011). A negative control was uninoculated sterile NB. There was no positive control.

3.12.4 Biochemical tube assay for the detection of laccase/peroxidase

The presence of laccase and peroxidase in the bacterial supernatant (post 48 h incubation) using the substrate guaiacol was determined spectrophotometrically (Bio-Rad 96 well iMark plate reader) with turnip as a positive control and sterile uninoculated NB as negative control (Desai et al., 2011). The substrate 2,2'-azino-bis (3-ethylbenzothiazoline-6-sulphonic acid) (ABTS) for laccase was also used (Li et al., 2008).

3.12.5 Biochemical tube assay for phenol reduction

The Folin-Ciocalteu (FC) reagent was used for the quantitation of phenol in Phenol Red Broth base media containing 0.018 g of phenol red per litre (Merck1.10987) (Agbor et al., 2014). This differential test media was not used to determine a change of pH rather that it contained phenol red (phenolsulfonphthalein), a phenolic molecule with three aromatic rings that would be a substrate for potential phenol degrading enzymes. The method was modified to increase

sensitivity by diluting the sample 1:1 instead of 1:16 as described by Strong & Burgess (2008). The Phenol Red Broth base was mixed in a 1:1 ratio with overnight CFCS and incubated at room temperature for 3 h, to establish whether the concentration of phenol was reduced. Green tea was used as a positive control and sterile uninoculated NB was used as a negative control.

When phenol concentrations were measured in food pollutants, textile dyes and wastewater samples, the same methodology was used except that the samples were used instead of phenol red broth media. All the experiments were done in triplicate.

Since phenol and phenolic compounds are a priority pollutant and found in many different WWs, in addition to determining a reduction in turbidity or colouration, a reduction in phenol due to the CFCS was also determined. The mixtures of 1:1 ratios of CFCS from different isolates mixed with test pollutants incubated at room temperature for 3 h (Section 3.16) were also subjected to a phenol concentration assay using the FC reagent. The results were compared with the negative control, a 1:1 mix of test pollutant with sterile uninoculated NB. The results were determined as a percentage of the negative control.

In addition to the test pollutants in the food category (coffee and soya sauce), textile dyes (BB, crystal violet (CV) and a commercial dye (Dye It), and wastewaters (brewery and dairy WW samples, and river water polluted with coloured industrial effluents), as well as PAHs: petroleum and paraffin, were also included to determine a reduction in phenols and phenolic compounds using FC.

3.13 DETECTION OF BIOPHYSICAL CHARACTERISTICS: BIOASSIMILATION OF TRIPHENYLMETHANE DYE (BROMOTHYMOL BLUE)

Two isolates (9T and 84Li) were previously shown to produce a clearing on NA- bromothymol blue (Section 3.12.1; Section 7.3.2; Figure 7.2D).

Preliminary experiments in liquid cultures were performed to establish whether the bacterial biomass changed colour during growth and whether it was influenced by temperature. Overnight cultures were used to inoculate 10 mL of 0.1% bromothymol blue in nutrient broth (BB-NB) and incubated at 37 °C (mesophile isolate 84Li) and 53 °C (thermophile isolate 9T) in glass McCartney bottles. They were incubated statically for 4 d and centrifuged at 10 000 rpm, and the biomass photographed. A static experimental set up was chosen due to lack of equipment.

Since isolate 9T has been previously shown to grow at both 37 °C and 53 °C, the same experiment was set up at these two temperatures, and photographed. Since isolate 84Li is mesophilic and does not grow at 53 °C, the effect of temperature was not performed on this isolate.

Decolourisation of BB by isolate 9T was measured by inoculation of 0.1% BB-NB with logarithmic cultures, incubated at 53 °C, and samples were taken at regular intervals up to 10 d and the concentration of BB was measured. The experiment was performed in triplicate, and the average and standard deviations calculated. Samples were centrifuged at 10 000 rpm to remove the cells and other particulate matter. The concentration of BB was determined spectrophotometrically at OD 595 nm (Bio-Rad 96-well iMark plate reader), as indicated by the standard curve in Appendix 9. Decolourisation was indicated by a decrease in absorbance at 595 nm. Results were compared to day “0” values at the start of the experiment. Decolourisation efficiency was expressed as a percentage of absorbance at 595 nm of sample / absorbance at 595 nm at day 0 x 100. Growth in 0.05% CV in NB, another triphenylmethane dye, was attempted, but no bacterial growth was observed suggesting that CV was bacteriostatic or bactericidal to isolate 9T.

3.14 DETECTION OF BIOPHYSICAL CHARACTERISTICS: BIOSORPTION OF HEAVY METALS

The biosorption assay was set up to determine whether dead bacterial cells would reduce the heavy metal content in aqueous solutions of Cr, Cu, Fe and Ni. Isolates were cultured for 48 h in NB statically and centrifuged to separate the biomass. The supernatants were removed and the cell pellet dried at 50 °C. Biomass was stored until enough dry weight could be accumulated and a sample was streaked onto NA for viability testing. The aqueous concentration of all heavy-metal ions could be determined spectrophotometrically and standard curves are indicated in Appendix 11. (http://cals.arizona.edu/extension/water_wagon/pdf- files/concentrationofcopperores.pdf). The optimal absorbance was determined by measuring the standard curves at wavelengths ranging between 415 nm and 750 nm. Absorbance at 415 nm was used for measuring the concentration of Cr, Fe and Ni ions, while the 750 nm wavelength was used to measure the concentration of Cu ions. The standard curves were used to determine the concentration of heavy-metal ions to be used in the biosorption assay that fell in the linear range of the standard curve. Copper solution was used at 50 mg/mL, Ni and Fe solutions were used at 5 mg/mL, while chromium solution was used at 1.67 mg/mL for the biosorption assay. Dry weight of bacterial growth was made up to a concentration of 0.1mg/µL concentration. A

100 µL aliquot of bacterial suspension was placed in an Eppendorf tube, and 500 µL of heavy metal solution added at concentrations of 50 mg/mL (Cu), 5 mg/mL (Ni, Fe) and 1.67 mg/mL (Cr). After mixing on a vortex, the tubes were left static for 5 h at room temperature. The tubes were thereafter centrifuged at 10 000 rpm, and the supernatant transferred to a 96-well plate for OD reading spectrophotometrically on Bio-Rad iMark microplate absorbance reader. Nine isolates were tested in duplicate for biosorption, and the averages and standard deviations were calculated. A negative control included heavy metal solutions in the absence of bacterial biomass. Because approximately 150 mL of growth culture was required to generate 0.1 g dry weight of bacteria, the experiment was conducted in duplicate instead of the recommended triplicate.

3.15 DETECTION OF BIOPHYSICAL CHARACTERISTICS: BIOFLOCCULANT ACTIVITY USING THE KAOLIN CLAY ASSAY

The assay used to measure the bioflocculant potential of bacterial isolates was conducted as described by Zhang et al. (2007) with slight modifications. Thirty-one isolates were grown in 100 mL NB in glass conical flasks statically for 48 h at either 37 °C or 53 °C, and centrifuged at 10 000 rpm to remove bacterial cells. The supernatant (CFCS) was tested in the flocculation assay. In a tube, 900 µL of kaolin clay at a concentration of 4 g/L, 90 µL of 10% calcium chloride (CaCl2) and 100 µL of CFCS were mixed and filled to the 100 mL mark with deionised water. The cylinder was gently mixed at room temperature and left for 5 min. The formation of visible aggregates was observed. By measuring the decrease in turbidity of the upper phase, the degree of flocculation could be measured. The OD of the upper phase was measured at 550 nm with a Bio-Rad iMark microplate absorbance reader. The positive control was alum (20 mg/mL) and the negative control was 1% sodium lauryl sulphate (SLS). The assays were performed in triplicate.

Flocculation activity % = B/A x 100% where: A is the negative control OD at 550 nm and B is the sample OD at 550 nm.

3.16 DETECTION OF BIOPHYSICAL CHARACTERISTICS: BIOSURFACTANT ACTIVITY

3.16.1 Emulsion activity assay

Bacterial isolates were grown in 100 mL NB in conical glass flasks for 48 h at either 37 °C or 53 °C. They were centrifuged at 10 000 rpm to remove the cell biomass. The CFCS was tested for emulsion activity using the method as described by Padmavathi and Pandian (2014). Preliminary screening used kerosene (paraffin oil) as a substrate. In 75 mm x 10 mm diameter glass test tubes, 1 mL of CFCS was mixed with an equal volume of kerosene (1 mL) by vortexing vigorously for 2 min, and then incubated at room temperature for 24 h. The negative control was sterile uninoculated NB and the positive control was 1% SLS (Samanta et al., 2012b; Hamzah et al., 2013a). The emulsification index was calculated as the height of the emulsion layer of the sample compared with the height of the emulsion layer of the positive control (1% SLS) as a percentage. The assay was done in triplicate. Twelve kerosene (paraffin oil)-emulsifying positive CFCS extracts were also tested against sunflower seed oil and petroleum for emulsifying activity and scored as “+” or “-“. Screening against other substrates for emulsion properties could reveal differences between isolates, and also reveal bioremediation potential of different pollutants in the environment.

3.16.2 Drop-collapse assay using Parafilm M

A method to determine biosurfactant activity using the drop-collapse assay on Parafilm M assay has been described by Kalyani et al. (2014). Using a micropipette, 10 µL of CFCS was carefully placed on Parafilm M. The shape of the drop on the surface was inspected after a minute. The diameter of the drops was evaluated, and for photographic purposes, BB was added for improved visualisation. The positive control was NB only, while the negative control included 1% SLS. The results were scored as “+” or “-“.

3.17 DETECTION OF BIOPHYSICAL CHARACTERISTICS: ANTI- BIOFILM ACTIVITY

Before the anti-biofilm assay could be performed, a bacterial isolate was required to form a robust biofilm that would serve as the target for anti-biofilm activity. Preliminary experiments were performed to select a candidate bacterial isolate from this hot-spring isolate collection that would form a robust biofilm that formed in a 96-well plate or glass test tube, and could withstand the washing and staining processes. Out of the four isolates (2T, 7T, 54T and 77S) tested, isolate B. subtilis 54T was selected as the best candidate. For the screening for anti-

biofilm properties, nine isolates that were positive as biosurfactant producers were selected. Positive control was 1% SLS in NB. A negative control was NB only.

This method uses staining of biofilm formed with CV, followed by washing. The CV colour is extracted and quantitatively measured at a specific absorption wavelength of 595 nm (O’Toole, 2011). Anti-biofilm activity of CFCS was tested against the ability of isolate 54T to form a biofilm in glass test tubes. An overnight culture of isolate 54T was used as a biofilm inoculum. The CFCS was prepared by centrifugation at 10 000 rpm for 4 min to remove the bacteria from a 48 h culture of the nine test isolates. Thereafter, 1 mL of NB was inoculated with 50 µL of B. subtilis 54T, and 1 mL of CFCS was added to the test tube. The tubes were incubated for 48 h at room temperature.

Media were carefully removed, the biofilm was rinsed once with phosphate buffered saline (PBS) and fixed with 2% sodium acetate (Magesh et al., 2013) and dried for 24 h. The biofilms were stained with 0.1% CV for 2-3 min and decolourised with 70% ethanol (Padmavathi & Pandian, 2014). Thereafter 200 µL was removed and placed in a 96-well plate to be read at OD 595 nm on the Bio-Rad iMark microplate absorbance reader. Samples were tested in quadruplicate.

3.18 ANTIMICROBIAL ACTIVITY SCREEN

3.18.1 Streak technique to identify antimicrobial activity

Preliminary antimicrobial screening was carried out by T-streak method on NA. Overnight cultures were inoculated in a straight line on agar plates, and test cultures were inoculated perpendicular to the first culture. Plates were incubated at 37 °C for mesophiles, or 53 °C for thermophiles. Antimicrobial activity was determined based on a zone of inhibition between two bacterial isolates. With this method thermophiles could not be tested against mesophiles, and motile bacteria were also not considered suitable for this method.

3.18.2 Antimicrobial activity testing of reference human pathogens and AR environmental isolates crude extracts

The AR and identification of the environmental isolates that was used to test for antimicrobial activity has been described in Section 3.10. Human pathogens were laboratory stocks of S. aureus, B. cereus, P. aeruginosa, E. coli, Enterococcus faecalis, Klebsiella pneumonia,

Mycobacterium smegmatis. Kocuria sp., B. subtilis, and B. licheniformis have been reported as opportunistic pathogens and were isolated from hot springs, from this study.

Antimicrobial activity for crude extracts was determined using an agar disk diffusion assay (Cladera et al., 2004). Crude extracts of CFCS were generated by centrifugation of bacterial cultures at 10 000 rpm to remove the bacteria and filtered through a 0.22 µm syringe filter for sterility.

Inocula of the test microorganisms were prepared from an overnight culture in NB, of McFarland standard of 0.5 or OD at 600 nm. A 25 µL aliquot of CFCS was placed on sterile disks (Whatman No 17) and then applied to inoculated lawns of bacteria on NA (Hi Media, India). A negative control included sterile uninoculated NB on the sterile disk. After incubation, plates were scored after 24 h at 55 °C or 37 °C depending on the isolate. Resistance was described as an absence of inhibition while sensitivity was scored as a zone of inhibition (Fandi et al., 2014).

3.19 IDENTIFICATION OF PROTEIN BANDS BY LIQUID CHROMATOGRAPHY – TANDEM MASS SPECTROPHOTOMETRY (LC-MS/MS)

Ten mL of NB was inoculated with 100 µl of an overnight culture and incubated at 37 °C or 53 °C for 24 h. The cultures were centrifuged at 10 000 rpm to remove the bacteria, and the supernatant was further processed. Two mL of each culture was dehydrated by freeze drying under vacuum and resuspended in loading buffer for sodium dodecyl polyacrylamide gel electrophoresis (SDS-PAGE), and 200 µL of the suspension was loaded in each well. Post-run, the gel was stained with Coomassie blue dye.

Protein bands of interest were in-gel trypsin digested as per the protocol described in

Shevchenko et al. (2007). In short, gel bands were de-stained using 50 mM NH4HCO3/50% methanol followed by in-gel protein reduction (50 mM DTT in 25 mM NH4HCO3) and alkylation (55 mM iodoacetamide in 25 mM NH4HCO3). Proteins were digested overnight at 37 °C using 5 – 50 µL, and 10 ng/µL trypsin, depending on the size of the gel piece. Digests were resuspended in 20 µL, 2% acetonitrile/0.2% formic acid and analysed using a Dionex UltiMate 3000 RSLC system coupled to an AB Sciex 6600 TripleTOF mass spectrometer. Peptides were first de-salted on an Acclaim PepMap C18 trap column (100 μm × 2 cm) for 2 min at 15 μL/min using 2% acetonitrile/0.2% formic acid, and then separated on Acclaim PepMap C18 RSLC column (300 μm × 15 cm, 3 µm particle size). Peptide elution was achieved using a flow-rate of

8 L/min with a gradient: 4-60% B in 15 min (A: 0.1% formic acid; B: 80% acetonitrile/0.1% formic acid). An electrospray voltage of 5.5 kV was applied to the emitter. The 6600 TripleTOF mass spectrometer was operated in data-dependent acquisition mode. Precursor MS scans were acquired from m/z 400-1500 using an accumulation time of 250 ms followed by 30 MS/MS scans, acquired from m/z 100-1800 at100 ms each, for a total scan time of 3.3 s. Protein Pilot (AB Sciex) was used for comparison of the obtained MS/MS spectra with a custom database containing sequences of Bacillus (UniProtKB/Swiss-Prot) as well as a list of sequences from common contaminating proteins. Peptide and proteins false discovery rate (FDR) cut-off was set to 1%.

3.20 REDUCTION OF COLOURATION OR TURBIDITY IN WASTEWATER SAMPLES

Reduction of colouration or turbidity was measured spectrophotometrically at different wavelengths. Standard curves are described in the Appendices showing a correlation between optical density and concentration of pollutant. Melanoidins have a maximum absorption at 405 nm (Langner & Rzeski, 2014) that allows for quantification in coffee and soya sauce (Appendix 11). Bromothymol blue has a maximum absorption at 598 nm (Narayan & Veeranna, 2014), and this was the wavelength used to measure colouration of BB (Appendix 10), CV and commercial dye (Appendix 12). Turbidity or light scatter by small particles has been used to measure bacterial concentrations with a spectrophotometer at 405 nm (Dominguez et al., 2001). The closest wavelength to 405 nm on the 96-well Bio-Rad plate reader with fixed wavelength filters was 415 nm, and therefore this value was used to measure turbidity in WW samples. Standard curves showing the relationship between turbidity of dairy WW and OD, and the standard curve for the measurement of phenol (Section 3.12.5) measured in brewery WW are listed in Appendix 13.

All experiments were performed at 25 °C to simulate bioremediation of water in the natural environment. Crude extracellular extracts were obtained from 48 h bacterial cultures that had been centrifuged at 10 000 rpm to remove the bacterial cells. This CFCS was mixed in a 1:1 ratio with the test sample and incubated for 3 h statically. The OD at either 415 nm or 595 nm, as indicated in Table 3.3, was used to determine whether there was a reduction in colouration or turbidity by comparison with the negative control, sterile uninoculated NB.

Table 3.3: Wavelength (nm) used to measure level of turbidity or colouration of pollutants in water and wastewater

Molecule/s responsible for Wavelength Potential Pollutant turbidity or colouration (nm) Detection 1% coffee melanoidin 415 colouration

soya sauce melanoidin 415 colouration

bromophenol blue triphenylmethane dye 595 colouration

Commercial Dye IT unknown 595 colouration

Brewery WW complex 415 turbidity

Dairy WW complex 415 turbidity

Industrial river WW complex 415 turbidity

3.20.1 Effect of temperature on WW colour or turbidity reduction

Isolates 2T, 9T, 17S and 19S were initially isolated at 53 °C, while isolates 71T, 77S, 83Li and 84Li were isolated at 37 °C. The aim of this study was to identify isolates that are important for water bioremediation, and hence a temperature of 25 °C (room temperature) was selected which most realistically simulates environmental conditions. Temperatures of 37 °C and 53 °C were used for comparison as temperatures at which the thermophilic and mesophilic bacteria were initially isolated.

Turbidity reduction was measured as described in Section 3.16 above in triplicate on brewery WW and river water contaminated with coloured industrial wastewater at 25 °C, 37 °C and 53 °C. The negative control was sterile uninoculated NB. The results were represented by OD at 415 nm in order to determine whether there was a difference between the controls and the three temperatures tested.

The effect of temperature on reduction in phenol levels in CFCS mixed with river water contaminated with coloured industrial wastewater was also determined using the phenol assay as described in Section 3.12.5. The experiment was performed in triplicate at 25 °C, 37 °C and 53 °C.

3.21 STATISTICAL ANALYSIS

3.21.1 Determination of relationship between antibiotic resistance and heavy-metal resistance

Data were analysed with the IBM SPSS statistics software (2015). The results were represented by “R” for resistance and “S” for sensitivity for both antibiotic and heavy-metal resistance parameters. The null hypothesis Chi-square test of independence was used to determine whether there was a statistically significant relationship between AR and HMT. Since the frequencies were low, Fisher’s exact test was used to determine the p values. Statistical significance was set at p<0.05.

3.21.2 Determination of significance of difference between negative controls and samples

Unless otherwise stipulated, experiments were performed in triplicate to allow calculation of means and standard deviations (MS Excel 2007). Statistical analysis was performed by one-way ANOVA followed by the Tukey’s post hoc test. The level of significance was accepted at p<0.05 with a 95% confidence level. All statistical analyses were performed using online analysis at (http://statspages.info/anova1sm.html).

CHAPTER FOUR: ISOLATION AND IDENTIFICATION OF BACILLUS AND CLOSELY RELATED BACILLUS BACTERIA ISOLATED FROM HOT SPRINGS IN LIMPOPO PROVINCE, SOUTH AFRICA

4.1 INTRODUCTION

Microbes from hot springs may be a resource of extremophiles and associated polymers and that have novel and useful properties in bioremediation of water (Demirjian et al., 2001; Demorne et al., 2017). It is estimated that <1% of bacteria in hot springs are isolated and identified by culture (López-López et al., 2013) although molecular methodologies suggest a huge diversity of different bacterial DNA are present (Lebedinsky et al., 2007). Gram-positive Bacillus spp. have previously been cultured from hot springs (Obeidat et al., 2012; Khiyami et al., 2012; Panda et al., 2016), and the Bacillus genus is commonly used in biotechnology (Kumar et al., 2013). With a view to identify bacteria that have a potential for WW bioremediation, this study firstly isolated and identified bacteria by culture using 16S rDNA sequencing. Identification was supported by four tools, i.e. % guanine-cytosine (GC) content (Yamane et al., 2011), amplified rDNA restriction analysis (ARDRA) (Rajendhran & Gunasekaran, 2011), comparison with published sequences in databases, GenBank (McGinnis & Madden, 2004) and EzTaxon-e (Kim et al., 2012), and phylogenetic analysis (Maughan & Van der Auwera, 2011).

4.2 METHODOLOGY

Water and sediment samples were taken from five hot springs in the Limpopo Province, South Africa, and their geographical locations are described in Section 3.1.1. Section 3.3 describes how the bacteria were isolated on five different media at two different incubation temperatures of 37 °C and 53 °C. Optimal conditions for the growth of the bacterial isolates (temperature, pH and salinity) were determined (Section 3.3).

DNA was extracted using a boiling technique. Using universal primers for the 16S rDNA fragment for PCR, the 16SrDNA amplicons were amplified and sequenced as indicated in Section 3.4. DNA sequences of the 16S rDNA amplicons were submitted to a public database, GenBank and accession numbers are listed (Section 3.5). Section 3.6 gives a description of how the partial 16S rDNA sequences were aligned and analysed for phylogenetic relationships.

ARDRA is a specialised PCR-RFLP using specifically rDNA sequences. In this study, fragments were generated using a computer program to cut at the restriction enzyme site Hae III

for the 16S rDNA sequences of the isolates. The restriction enzyme digest patterns were converted to binary data and displayed as a Neighbour-Joining (N-J) tree (Section 3.6).

The percent of GC present in the 16S rDNA fragment was calculated online as outlined in Section 3.8 for all the isolates. Reference strains were also included in the analysis.

4.3 RESULTS

4.3.1 Bacterial isolation

4.3.1.1 Concentration of mesophiles and thermophiles in hot-spring water

Figure 4.1: Concentrations (CFU/400 mL) of mesophiles (isolated at 37 °C) and thermophiles (isolated at 53 °C) from five sampled hot springs

The bacterial load in 400 mL of hot-spring water samples from five sites are shown in Figure 4.1. Siloam, the warmest hot spring at 68 °C, gave the lowest total counts while the lower water temperatures of ≤52 °C resulted in a higher bacterial load. The results are expressed as CFU/400 mL as this was the volume sampled from each site and plated onto different media. The bacterial concentrations were extremely low and therefore the results were not expressed conventionally as CFU/mL or CFU/100 mL. At hot-spring water temperatures of ≤52 °C for Libertas, Mphephu and Lekkerrus, the mesophilic bacterial load (isolated at 37 °C) was similar for all three sites (Figure 4.1). The two hottest springs, Tshipise and Siloam had the lowest total number of counts because they had the lowest number of mesophiles. The thermophilic bacterial load (isolated at 53 °C), was similar for hot springs Tshipise, Libertas and Mphephu, but much lower for Lekkerrus (red bar in Figure 4.1).

4.3.1.2 Influence of isolation media

300

250

200

150 Nutrient agar 100 minimal Luria agar 50

No. of bacteria No.per100ml water 0 37°C 53°C 37°C 53°C 37°C 53°C Siloam Tshipise Lekkerus Hot Spring Sample Site

Figure 4.2: Colony-forming units (CFU) per 100 mL of water showing the difference between NA and minimal Luria agar for mesophiles incubated at 37 °C, and thermophiles at 53 °C at three different hot-spring locations

At three of the five sampled sites, bacterial counts on 10% LA were higher than on NA (Figure 4.2). The colonies that developed on the agar plates of the other two sites, namely Mphephu and Libertas, were too numerous to count (>300 bacteria per 100 mL) and these values are therefore not given in Figure 4.2.

4.3.2 Optimal growth conditions for bacterial isolation and growth

The optimal growth conditions for the isolates were determined as the bacteria needed to be grown as inocula for further experiments and investigations. The various WW sampling sites differed in terms of temperature, salinity and pH; therefore, three parameters were measured for the isolated and cultured thermophiles (isolated at 53°C). The values for pH, temperature and salinity for each isolate are listed in tabular format in Appendix 14.

Figure 4.3: Determination of optimal conditions for growth of bacterial isolates: Optimal pH was 7 (Figure 4.3A), optimal temperature was 55 °C (Figure 4.3B) and optimal salinity was 5% NaCl (Figure 4.3C). Three of the 16 isolates tested for salinity could also grow at 10% NaCl

4.3.3 16S rDNA sequencing

The contiguous sequences were compared to two databases, namely GenBank and EzTaxon-e and the highest percentage similarities and accession numbers are listed in Table 4.1. Values >97% suggest a match to genus level, while a value of >99% suggests a match to species level

(Yarza et al., 2014). Where no PCR product was obtained, the sequencing was not determined (nd), and in some cases, sequencing was incomplete which did not allow for a full consensus sequence to be constructed. Sequences from this study were submitted to GenBank with their relevant accession numbers.

Isolati Isol on GC ate Temp GenBank GB Access EzTaxon-e Submit Access conte No. °C BLAST No. EzTaxon-e Access No. No. nt ARDRA Phylogeny Anoxybacillus Anoxybacillus

3T 53 rupiensis 94% KJ842629.1 rupiensis 99% AJ879076 MF037806 √ A1 Anoxybacillus Anoxybacillus AM988775. Anoxybacillus

4T 53 rupiensis 99% 1 rupiensis 99% AJ879076 MF037807 √ A1 Anoxybacillus Anoxybacillus Anoxybacillus

7T 53 sp. ATCC 99% KJ722458.1 rupiensis 99% AJ879076 MF037808 √ A1 Anoxybacillus Anoxybacillus Anoxybacillus

11T 53 sp. 94% KF254912.1 rupiensis 98% AJ879076 MF037809 X B Anoxybacillus Anoxybacillus FN432807. Anoxybacillus

13S 53 sp. 99% 1 rupiensis 98% AJ879076 MF037810 √ A1 Anoxybacillus Anoxybacillus Anoxybacillus flavithermus mongoliensis

17S 53 99% KF039883.1 98% EF654664 MF037811 √ C Anoxybacillus Anoxybacillus KP221933. Anoxybacillus

19S 53 sp. 99% 1 rupiensis 99% AJ879076 MF037812 √ A1 Anoxybacillus Anoxybacillus Anoxybacillus flavithermus KC503890. flavithermus AVGH0100004 S 43T 53 99% 1 99% 1 MF037813 √ A2 Anoxybacillus Bacillus Bacillus licheniformis HM631844. licheniformis Bacillus group

2T 53 99% 1 ATCC 98% AE017333 MF037814 √ D A Bacillus Bacillus licheniformis licheniformis Bacillus group

6T 53 99% KJ729823.1 ATCC 99.8% AE017333 MF037815 √ C A Bacillus Bacillus sp. GU132507. licheniformis Bacillus group

8T 53 98% 1 ATCC 97% AE017333 MF037816 √ D A Bacillus Bacillus licheniformis HM631844. licheniformis Bacillus group

10T 53 97% 1 ATCC 98% AE017333 MF037817 √ D A Bacillus Bacillus KC634086. licheniformis Bacillus group

20S 53 subtilis 99% 1 ATCC 99.8% AE017333 MF037818 √ nd A Bacillus Bacillus 28 licheniformis GQ340513. licheniformis Bacillus group

M 53 99% 1 ATCC 99.8% AE017333 MF037819 √ D A Bacillus Bacillus 30 licheniformis licheniformis Bacillus group

M 53 99% KJ526854.1 ATCC 99% AE017333 MF037820 √ D A Bacillus Bacillus licheniformis licheniformis Bacillus group

39T 37 99% KF879197.1 ATCC 99.7% AE017333 MF037821 √ B A Bacillus Bacillus licheniformis sonorensis AYTN0100001 Bacillus group

74T 37 99% JN366749.1 99.6% 6 MF037822 √ D A

Bacillus KC634086. Bacillus subtilis AMXN010000 Bacillus group

12S 53 subtilis 99% 1 99% 21 MF037827 √ B A Brevibacterium Bacillus sp. halotolerans Bacillus group 14S 53 98% CP11051,1 97% AM747812 MF037828 X B A 21 Bacillus Bacillus subtilis AMXN010000 Bacillus group

M 53 subtilis 99% JN585712.1 99.7% 21 MF037829 √ C A Bacillus 22 methylotrophi JQ765577. Bacillus AYTO0100004 Bacillus group

M 53 cus 99% 1 tequilensis 99% 3 MF037830 √ nd A Bacillus KC182058. Bacillus subtilis AMXN010000 Bacillus group

33Li 53 subtilis 98% 1 99.7% 21 MF037831 X B A Bacillus 40L Bacillus KP249695. tequilensis AYTO0100004 Bacillus group

e 37 subtilis 99% 1 99.6% 3 MF037832 √ C A B licheniformis KC429774. Bacillus subtilis AMXN010000 Bacillus group

41Li 37 97% 1 96% 21 MF037833 √ D A Bacillus KP249695. Bacillus subtilis AMXN010000 Bacillus group S 47Li 53 subtilis 99% 1 99.9% 21 MF037834 √ B A Bacillus NR_118486 Bacillus subtilis AMXN010000 Bacillus group S 48Li 53 subtilis 99% .1 99.9% 21 MF037835 √ D A Bacillus HM753614. Bacillus subtilis Bacillus group

54T 37 subtilis 99% 1 98.9% CP002905 MF037836 √ B A Bacillus sp. Bacillus subtilis ABQL0100000 Bacillus group

78S 37 99% KF984420.1 99% 1 MF037837 √ D A Bacillus Bacillus subtilis ABQL0100000 Bacillus group

83Li 37 subtilis 97% KF533727.1 96% 1 MF037838 √ D A Bacillus HM367735. Bacillus subtilis AMXN010000 Bacillus group

1T 53 subtilis 96% 1 96% 21 MF038049 X B A Bacillus Bacillus licheniformis HM055609. licheniformis B.

15S 53 94% 1 94% AE017333 MF038050 √ D licheniformis Bacillus Bacillus sp. GU132507. licheniformis B.

18S 53 96% 1 95% AE017333 MF038051 √ C licheniformis 52 Brevibacillus Brevibacillus Bacillus group S M 53 agric 94% JN812211.1 agri 94% D78454 MF039084 √ E A Bacillus 24 Bacillus aerophilus Bacillus group

M 53 pumilus 99% KJ526891.1 99.8% AJ831844 MF038052 √ C A Bacillus Bacillus 32L panaciterrae NR_041379 panaciterrae Bacillus group

e 53 99% .1 99% AB245380 MF038053 √ E K Bacillus Bacillus methylotrophi KP342210. methylotrophic Bacillus group

77S 37 cus 99% 1 us 99.9% JTKJ01000077 MF038054 √ B A Solibacillus Solibacillus

73T 37 sylvestris 95% KF441704.1 sylvestris 95% AJ006086 MF039085 X C Solibacillus Aneurinibacill us migulanus NR_113764 Aneurinibacillus

86Li 37 96% .1 migulanus 96% X94195 MF040218 √ E Brevibacillus Brevibacillus LN681596. Brevibacillus LDCN0100001

16S 53 sp. 99% 1 formosus 97% 5 MF038055 √ E Brevibacillus Brevibacillus KM403208. Brevibacillus

36Li 53 sp. 99% 1 agri 99% D78454 MF038056 √ E Brevibacillus Brevibacillus GQ497292. Brevibacillus

70T 37 sp. 99% 1 fluminis 99% EU375457 MF038057 √ E Brevibacillus Brevibacillus KP165013. Brevibacillus

85Li 37 formosus 97% 1 brevis 98% AB271756 MF038058 √ E Brevibacillus 53 Brevibacillus GQ497292. Brevibacillus S M 53 sp. 92% 1 fluminis 89% EU375457 nd √ E nd

Uncultured bacterium Kocuria

75S 37 94% KJ013386.1 sediminis 92% JF896464 nd x A2 nd

4.3.4 Guanine-cytosine (GC) content (in percentage)

The GC content (in percentage) and accession numbers for isolates and reference strains (from GenBank) are listed in Appendix 6 for the approximately 1 400 bp 16S rDNA fragment. Based on the GC content of the 16S rDNA sequences for the isolates in this study, they were grouped into four genera (Anoxybacillus, Bacillus, Aneurinibacillus and Brevibacillus). The average and standard deviations for the GC content were calculated and plotted and are given in Figure 4.4. The isolates that fell out of one standard deviation range of the average GC content were earmarked as potentially being different from the group, i.e. isolates 1T, 11T, 14S and 33Li (Appendix 5 as indicated by *). In all other respects, there was a general match with the GC contents and groupings into the four genera.

Figure 4.4: Average GC content (%) of 16S rDNA sequence of isolates from this study and reference type strains of genera Aneurinibacillus and Brevibacillus (family Paenibacillaceae) and genera Anoxybacillus and Bacillus (family Bacillaceae) and unclassified genus Solibacillus showing that GC content can be used to distinguish between genera

4.3.5 Computer-simulated amplified ribosomal DNA restriction analysis (ARDRA)

In this investigation, computer-generated HaeIII fragments were represented as a simulated gel pattern and converted to a binary sequence where “0” means no band and “1” represents a band of that size. Sequences were aligned and analysed using SeaView (Gouy et al., 2010) and

presented as a circular tree as indicated in Figure 4.5. Five clusters were identified A-E, including Anoxybacillus (cluster A), Bacillus sp (cluster B) and Bacillus spp. related to B. subtilis and B. licheniformis of Bergey’s Group A (cluster D). Cluster E held Brevibacillus and Aneurinibacillus whilst cluster C was a mixed undefined group.

E A.kestanbolinensisK4

30M Aneurinibacillus sp. U33 21M 8T 15S

A2 A1 B

Figure 4.5: Circular neighbour-joining tree of ARDRA binary data of 16S rDNA fragments using HaeIII. Five main clusters (A-E) are indicated in the diagram

4.3.6 Phylogenetic analysis

4.3.6.1 Family Bacillaceae genus Anoxybacillus

The N-J phylogenetic tree (Figure 4.6) grouped the eight Anoxybacillus isolates together, with convincing bootstrap values, a result that supported and confirmed the correlation of matches with GenBank BLAST. The phylogeny also confirms the four GC content (%) groupings of Aneurinibacillus and Brevibacillus, Anoxybacillus and Bacillus spp.

Figure 4.6: A neighbour-joining phylogenetic tree of a 914 bp fragment of the 16S rDNA sequences between isolates from this study and representative members of type strains of Anoxybacillus, Bacillus, Brevibacillus and Aneurinibacillus. Bootstrap values (%) are based on 100 replicates and shown for branches with more than 50% bootstrap support. Bar indicates 0.02 substitutions per 100 nucleotide positions

4.3.6.2 Family Bacillaceae genus Bacillus

The majority of the isolates in this study fell within the genus Bacillus, more specifically into Bergey’s Group A which includes B. subtilis and B. licheniformis, two very closely related species (Ludwig et al., 2009). These two species are commonly described as isolates from hot springs in many investigations. In order to ensure that the isolates did not fall into other Bergey’s groups not represented in the phylogenetic tree (Figure 4.6), another phylogenetic tree was drawn with only the Bacillus isolates from this study and reference type strains from Bergey’s Group B (Bacillus lentus), Group C (Bacillus megaterium), Group D (Bacillus cereus), Group E (Bacillus aquimaris), Group F (Bacillus coagulans), Group G (Bacillus halodurans), Group H (Bacillus arsenicus), Group I (Bacillus smithii) and Group J (Bacillus panaciterrae) (Figure 4.7).

The sequences were obtained from published databases as listed in Appendix 3 and Section 3.5. It confirmed that all the isolates in this study clustered with Bergey’s Group A: B. subtilis/B. licheniformis by phylogeny except for the single isolate 32Le. Single isolates that were not B. subtilis/B. licheniformis were identified by GenBank BLAST to be B. panaciterrae (isolate 32Le), B. aerophilus (isolate 24M) and B. methylotrophicus (isolate 77S). By GC content these three isolates could not be differentiated from the rest of the Bacillus reference strains or new isolates (Appendix 6). Type strains of Bacillus aerophilus/B. pumilus and B. methylotrophicus and the corresponding new isolates from this study fell into Bergey’s Group A with B. subtilis/B. licheniformis as confirmed by the N-J phylogenetic tree (Figure 4.7) and ARDRA clustering (Figure 4.5), while isolate 32Le together with B. panaciterrae branched separately into Bergey’s Group K (Ludwig et al., 2009) in the phylogenetic tree (Figure 4.6).

However, these classifications were not supported by ARDRA patterns where the isolates 32Le and 24M fell into the ARDRA cluster C and isolate 77S fell into ARDRA cluster B, but the reference type strain of B. panaciterrae fell into ARDRA cluster E, and the type species B. methylotrophicus and B. aerophilus/pumilus clustered with ARDRA cluster D that contained B. subtilis and B. licheniformis (Figure 4.5). This suggests that the three singular isolates may require more confirmation of their species identification, and that identification by GenBank BLAST may not be sufficient in this case.

Pseudomonas aeruginosaATCC23993 91 Bacillus haloduransATCC27557 (G) Bacillus arsenicusB3 (H) Bacillus coagulansATCC7050 (F) 32Le Bacillus lentusIAM12466 (B) Bacillus smithiiTBMI12 (I) Bacillus siralis171544 Bacillus panaciterrae Gsoil1517 Bacillus cereusATCC14579 (D) Bacillus megateriumATCC14581 (C) Bacillus aquimarisTF-12 (E) 52M 96 24M Bacillus aerophilusCRh28 Bacillus pumilusATCC7061 100 54T 2T 72 10T 15S 23M 60 18S 8T 79 74T 20S 30M 39T 28M Bacillus licheniformisATCC145808T 6T 1T 22M 41Li 78S 83Li 61 14S Bacillus subtilis subsp. spizizenii ATCC6633 33Li 21M Bacillus subtilis JCM 1465

40Le 12S 77S Bacillus subtilisATCC21331 Bacillus subtilisUAC50 Bacillus subtilisIAM12118 Bacillus subtilisBCRC10255 Bacillus tequilensis10b Bacillus methylotrophicus Bacillus siamensisIHBB16121 Bacillus subtilisDSM10

Figure 4.7: Neighbour-joining tree of only Bacillus isolates, with representative members of type strains from different Bergey’s groups A-K confirming that most isolates in this study fell into the group A (B. subtilis/B. licheniformis), and that two isolates matmatched B. pumilus/B. aerophilus (group A) and B. panaciterrae (group K), respectively. Bootstrap values were obtained from 100 replicates and the bar indicated 4.3.6.3 Family Paenibacillaceae0.01 substitutions genus per 100 Brevibacillus nucleotide positions

Four isolates, namely 16S, 36Li, 70T and 85Li were identified as Brevibacillus spp. by BLAST search (Table 4.1) and confirmed by ARDRA (Figure 4.5), phylogenetic analysis (Figure 4.6) and supported by GC content percentage values (Appendix 6).

4.3.7 Analysis of unknown isolates

Four isolates that had a >97% BLAST match to published 16S rDNA sequences were 11T, 1T, 14S and 33 Li. However, they were found to have a GC content and ARDRA clustering that did not fall into the corresponding group

Six isolates were deemed to be “unknown” due to their poor BLAST match to published sequences of <95%). These were 15S, 52M, 53M, 73T, 74S and 86Li. Three other “unknown Bacillus” isolated from hot springs in India (clone TPB_GMAT_AC4; GenBank HG327138.1), Indonesia (clone KSB12; GenBank JX047075.1) and China (clone DGG30; GenBank AY082370.1) from the GenBank database that remain, as yet, unidentified were also included and discussed in this analysis using a combination of the tools for identification.

4.4 DISCUSSION

4.4.1 Bacterial isolation

4.4.1.1 Concentration of mesophiles and thermophiles in hot-spring water

Since the 1950s, numerous studies have been conducted on the isolation and characterisation of thermophilic bacteria in hot springs located in many parts of the world (Narayan et al., 2008; Cihan, 2013; Khiyami et al., 2012; Pandey et al., 2015), because thermophiles are potential sources of specialised and adapted bioactive molecules such as enzymes and polymers, and these microbes are metabolically active under conditions of extreme temperatures (López-López et al., 2013; Chen & Jiang, 2018). However, in general, low bacterial loads are reported, since only a small percentage of the bacterial diversity present is cultured due to limited culture conditions of media, atmospheric pressure and oxygen. As a result, most studies do not provide an accurate quantitative assessment, and it may be a misrepresentation to try and compare bacterial loads observed in different studies due to different isolation conditions and methodologies.

The bacterial load of the hot-spring water of Siloam, the warmest hot spring at 68 °C, gave the lowest total counts while the lower water temperatures of ≤52 °C resulted in a higher bacterial load (Figure 4.1). At hot-spring water temperatures of ≤52 °C the mesophilic bacterial load (isolated at 37 °C) was the same for the three colder sites; however, the thermophilic bacterial load (isolated at 53 °C), was similar for hot springs Tshipise, Libertas and Mphephu, but much lower for Lekkerrus (red bar in Figure 4.1). Lekkerrus and Mphephu have similar water temperatures of 43.5 °C and 42.4 °C, respectively, while Libertas and Lekkerrus are very

closely situated geographically with the same geological formation found at their origin. However, it was observed that the thermophilic bacterial load of Lekkerrus was considerably lower compared with those observed in Libertas and Mphephu spring waters. A likely reason for this observation is that the Lekkerrus sampling source was a pipe that conveyed water into the swimming pool with manual flow control, and therefore not a true reflection of the actual thermophilic bacterial load, compared with direct sampling of a hot spring, as the temperature of the feed water standing in the pipe could drop to below 42 °C. Similarly, Khiyami et al. (2012) conducted a study of the microbial diversity of hot springs in Saudi Arabia, and found that the levels of mesophiles (bacteria incubated at 30 °C, 40 °C and 50 °C) remained constant in hot- spring water that ranged in temperature from 40 to 83 °C. They also reported that there was a decrease in bacterial load with temperature with respect to thermophiles (grown at 60 °C), as found in this study. Because the majority of isolated bacteria form spores, it is not uncommon or surprising to isolate mesophiles together with thermophiles where they exist tolerating, but not growing at higher water temperatures.

4.4.1.2 Influence of isolation media

Bacterial counts on 10% LA were higher than on NA (Figure 4.2). It is not surprising that these bacteria prefer a low-nutrient culture medium, being isolated from a very low-nutrient environment, and this should be a consideration in the future, as a factor to extend the diversity of bacteria isolated from hot springs. In a study of optimal culture conditions for microbes from Yellowstone hot springs, Meyer-Dombard et al. (2012) reported that simulating the trace elements of the original water samples resulted in higher diversity and faster growth of bacteria, and higher concentrations of trace elements actually inhibited growth and diversity.

4.4.1.3 Optimal growth conditions for bacterial isolation and growth

To maintain the stock cultures, it was necessary to determine the optimal temperature, pH and salinity conditions for growth. The results showed that most of the bacteria preferred a neutral pH of 7, an incubation temperature ranging between 50 °C and 55 °C, and salinity of 5% NaCl (w/v). Obeidat et al. (2012) tested eight Geobacillus spp. from hot springs in Jordan with temperatures ranging from 48 °C to 62 °C, and pH between 6 and 7, and found the optimal temperatures to be between 60 °C and 65 °C, and pH 6 to 8. Zhang et al. (2011) reported findings on two isolates of Anoxybacillus with an optimal growth temperature of 55 °C and pH of 8. It therefore appears that these spore-forming Bacillus and closely related Bacillus microorganisms are robust with a tolerance for a wide range of environmental conditions. Extremophiles isolated in this study include the alkaliphilic thermophile Anoxybacillus flavithermus with an optimal pH of 10 and a temperature of 50 °C (isolate 17S), thermophilic

Anoxybacillus rupiensis with an optimal temperature of 60 °C (isolate 13S), and halotolerant thermophiles of B. licheniformis that could still grow in 10% (w/v) NaCl (isolates 2T, 6T and 8T).

Seasonal changes also affect the microbial community composition of hot-spring water (Mackenzie et al., 2013; Briggs et al., 2014), and therefore sampling in another season may produce a different composition of isolates, or isolates with different tolerances to the environment, and this aspect should be further investigated in the future. At all sites sampled, the dissolved oxygen (DO) concentration was negligible as a result of the high water temperatures. Strict anaerobic microbes are another avenue that could be explored to increase the existing knowledge of the diversity of culturable microorganisms from hot-spring environments (Lebedinsky et al., 2007; Pagaling et al., 2012), and for the isolation of biotechnologically important microbes that could be useful in fermentations.

Knowledge of the diversity of culturable bacteria from hot springs can therefore be increased by isolation of anaerobes, use of different media, minimal media, seasonal sampling or enrichment for certain groups and inclusion of trace elements.

4.4.2 16S rDNA sequencing

Molecular techniques based on DNA sequencing have far surpassed the traditional methods to determine biochemical and phenotypic characteristics that had been used previously to identify prokaryotes such as bacteria. This is because the latter is dependent on environmental and external factors, and can be time-consuming, laborious and subject to error. Phenotypic characteristics related to colony morphology, biochemical reactions, serology, pathogenicity and AR can vary considerably. However, the DNA of a bacterial isolate remains unchanged. The 16S rDNA sequence has been used as the gold standard for identification of microorganisms, because it is present in all bacteria, the function of this gene has not changed over time and the 1 400 bp sequence is long enough to provide suitable information for analysis (Janda & Abbott, 2007; Rajendhran & Gunasekaran, 2011). Using universal primers, a fragment of approximately 1 400 base pairs (bp) is amplified. There are both conserved and variable areas scattered within this 16S rDNA sequence, allowing different universal PCR primers to be used but variable enough to allow discrimination between different species. The larger the fragment sequenced, the more accurate the result. To identify unknown bacterial isolates, the 16S rDNA sequences are compared with those in the GenBank database (BLAST) and the closest similarities are listed in Table 4.1. However, the disadvantage of this tool is that the database is created by public contributions, and therefore it is possible that unknown sequence may be compared to

misidentified or incorrectly named strains. This can be narrowed down by comparison with EzTaxon-e, a Web-based tool for the identification of prokaryotes based on 16S ribosomal RNA gene sequences; EzTaxon is an open access database that is produced and maintained by ChunLab USA, Inc. In 2012, EzTaxon was extended to include uncultured prokaryotes and the name of the database was changed to EzTaxon-e. This database is manually curated and quality controlled and thus less susceptible to be contaminated by erroneous species identifications made by the public and hence, it would be more accurate. For example, in the case of Bacillus species, the results of the BLAST search will not take into account the new reclassification of genera in 2009, and up-to-date changes within the group’s nomenclature, therefore the results will automatically BLAST the closest match which may be a “Bacillus” species by older classification. The advantage of this technique, however, is that the database is very large and extensive allowing one to have access to sequences worldwide in real time. The cut-off value for the percentage similarity is also critical. Yarza et al. (2014) and López-López et al. (2013) describe with statistical proof that >97-98% allows for determination at a species level. Other investigators have used >97% as a cut-off value (Belduz et al., 2003; Drancourt & Raoult, 2005). When the value is lower than 95%, the result cannot be accurate at the genus or species level.

4.4.3 Guanine-cytosine (GC) content (in percentage)

The guanine-cytosine (GC) content of a fragment of DNA or the whole genome refers to the proportion of DNA that is either G-C, but not A-T, in relation to all the bases present. The G-C bond is stronger than an A-T bond in DNA resulting in a more stable DNA molecule. The GC content of a bacterial genome and the GC content of the stem of the rDNA sequence have been correlated with optimal growth temperatures (Galtier & Lobry, 1997; Wang et al., 2006). Mann & Chen (2010) describe the GC content of bacteria correlating with environmental niches and lifestyle, e.g. free-living organisms in high-nutrient environments have high GC content while those living in nutrient-poor environments have a lower GC content. In addition, the GC content varies among different genera (Muto & Osawa, 1987) which has led to its inclusion as supportive information in the taxonomic classification of bacteria. Yamane et al. (2011) compared the GC content of the 16S rDNA in isolates from high-temperature oil wells and found isolates with high GC contents associated with oil deposits compared with other niches. The GC content of the 16S rDNA was included in this study to establish whether this percentage could be useful in showing which strains were similar, and whether it was useful in discriminating between different genera. Figure 4.4 of the GC content shows the family

Bacillaceae including genera Anoxybacillus and Bacillus, the family Paenibacillaceae including the genera Aneurinibacillus and Brevibacillus, and unclassified Bacillales genus Solibacillus. The standard deviations of the two genera within the families did not overlap and were therefore different. Therefore, Aneurinibacillus could be distinguished from Brevibacillus, and similarly Anoxybacillus could be distinguished from Bacillus based on GC content.

4.4.4 Computer-simulated amplified ribosomal DNA restriction analysis (ARDRA)

Essentially, amplified ribosomal DNA restriction analysis (ARDRA) is PCR-RFLP analysis of the amplified 16S rDNA (Rajendhran & Gunasekaran, 2011) and has been useful for genotyping of strains. It is based on the number and size of fragments that are generated when a PCR product of the 16S rDNA amplicon is digested with a restriction enzyme and the restriction fragments are separated according to their lengths by agarose gel electrophoresis. The discriminatory power of this technique is generally low because of the highly conserved nature of the 16S rDNA sequence; the generated fingerprinting and discriminatory power between different species are dependent on the length of the fragment and the restriction enzymes used. Although ARDRA is fast, and its patterns are highly reproducible, it is based on variations associated with restriction sites only, while sequencing near full-length 16S rDNA amplicons of approximately 1 400 bp has more discriminatory power. The use of computer-simulated fragments has been shown to be a valid assessment of genotyping (Moyer et al., 1996; Wei et al., 2007). Restriction enzyme HaeIII in ARDRA has been used to group bacteria (Rahmani et al., 2006) including Bacillus isolates from hot springs (Pagaling et al., 2012; Pathak & Rathod, 2015). Rai et al. (2015) reported that ARDRA with restriction enzymes HaeIII and AluI produced 100% similarity with the clustering of Bacillus spp., based on carbohydrate utilisation patterns.

The circular N-J tree using the binary data from the computer generated ARDRA results (Figure 4.5), shows clusters A to E. Clusters A1 and A2 included the Anoxybacillus, cluster B included Bacillus spp. and cluster C consisted of a mixed group that was not clearly defined. Cluster D included Bacillus species from Bergey’s classification Group A (Ludwig et al., 2009) which includes B. subtilis and B. licheniformis. Cluster E made up the genera Aneurinibacillus and Brevibacillus of the family Paenibacillaceae. Together, the newly sequenced isolates with a large database of reference strains indicated that the use of computer-simulated ARDRA patterns is a valid tool to differentiate bacteria in this study.

4.4.5 Phylogenetic analysis

Phylogenetic trees analyse the relationships between individual strains considering every base of the 16S rDNA sequence. Due to its flexibility, it can be applied to any prokaryotic organism because this gene is highly conserved and stable, and this approach is more accurate than a BLAST search. A comparison of the three molecular tools mentioned above (16S rDNA BLAST search, GC-content percentage and ARDRA) and phylogenetic tree analysis (Figure 4.6), shows that there was, in general, good correlation with the grouping of the phylum Firmicutes into family Bacillaceae with genera Bacillus (n = 31) and Anoxybacillus (n = 8), and family Paenibacillaceae genera Brevibacillus (n = 3) and Aneurinibacillus (n = 1).

The family Bacillaceae is distinguished by their ability to form heat tolerant endospores. As a result, they are abundant, robust and well distributed in many environmental niches, including hot springs. The prototype B. subtilis was first described in 1872, and prior to the 1990s, the genus Bacillus mainly constituted the family Bacillaceae. However, since then, many major taxonomic changes have occurred, which has resulted in new genera being described and several species formerly classified in Bacillus having been reclassified into other genera. Consequently, a comparison with older published literature revealed that the nomenclature of “Bacillus” would be used but in later publications it would appear as another genus. Mandic- Mulec et al. (2015) reviewed this group in detail. In addition, this family Bacillaceae is expanding extremely rapidly with 25 new genera having been described in 2013 and 2014, with a total of 62 genera listed in 2015. The genus Bacillus (family Bacillaceae) is the largest group with 226 species described in 2015, and it is expanding rapidly with 38 new species having been described between August 2013 and March 2015. Some of these species are represented by only one isolate making verification challenging and increasing the complexity of Bacillus phylogeny. This confusion regarding the phylogeny of the genus Bacillus was reported by Maughan and Van der Auwera (2011) who observed that phenotypic groupings are not congruent with 16S rDNA groupings because this group is phenotypically so variable. A comparison of publications on Bacillus phylogeny in 1991 (Rössler et al., 1991), in 2003 (Xu & Cote, 2003) and in 2009 (Ludwig et al., 2009) confirmed the exploding evolutionary changes in this group’s nomenclature. As a result, the nomenclature and classification of this group are challenging, and difficult to keep up to date.

Therefore, a literature review of Bacillus species isolated from hot springs will result in different nomenclature used depending on the date of publication. What may have been previously called Bacillus could be named “Geobacillus” or “Paenibacillus” (meaning “almost

Bacillus”) in later publications introducing incongruence between different studies. A significant proportion of publications on the identification of Bacillus species from hot springs rely on only one tool of identification, a comparison of 16S rDNA sequence to a public database, i.e. GenBank BLAST (Obeidat et al., 2012; Sharma et al., 2013a; Ghalib et al., 2014). However, this study will show that other means of genotyping, such as phylogenetic analysis, can disprove conclusions that are based only on the BLAST tool. The general consensus of a positive identification using the 16S rDNA sequence on GenBank BLAST, is a >97% match (Yarza et al., 2014), and if studies are not stringent in applying this cut-off value, and merely report bacterial identification based on any genetic similarity, this leads to more “misidentification” within this group.

4.4.5.1 Family Bacillaceae genus Anoxybacillus

As compared to other groups, the Anoxybacillus group is relatively new, having been established in 2000. Cihan et al. (2012) suggested that Anoxybacillus is the most dominant genus in hot springs. Twelve of the 15 novel species of Anoxybacillus listed in Appendix 3 were isolated from hot springs. Thirty-five of the 53 isolates of Anoxybacillus from hot springs in Turkey showed uniquely different patterns with ARDRA compared with 12 type species (Cihan, 2013) providing further evidence that novel species of Anoxybacillus can be found in hot springs and that differentiation from reference strains is discernible by phylogenetic analysis and ARDRA. A BLAST search confirmed that eight isolates from this study were Anoxybacillus spp. including A. flavithermus and A. rupiensis. The N-J phylogenetic tree grouped the eight isolates together, with convincing bootstrap values (Figure 4.6). However, the GC content percentage of isolate 11T differed by more than one standard deviation from published Anoxybacillus spp. data (Appendix 6), and the other Anoxybacillus isolates from this study. Results from ARDRA analysis confirmed that isolate 11T did indeed group separately from the Anoxybacillus cluster A (Figure 4.5), and requires further investigation. In order to establish whether there was a geographical discrimination by phylogeny, the isolates of A. rupiensis from this study and from China, Indonesia, India, Pakistan and Turkey (Appendix 6, Figure 4.5) were analysed by ARDRA. No geographically associated groupings were observed.

4.4.5.2 Family Bacillaceae genus Bacillus

N-J phylogenetic analysis (Figure 4.7) of only Bacillus spp. including reference strains from all Bergey’s groupings, confirmed that all the isolates in this study clustered with Bergey’s Group A with B. subtilis and B. licheniformis. Three isolates that were different were identified by

GenBank BLAST, i.e. B. panaciterrae (isolate 32Le), B. aerophilus (isolate 24M) and B. methylotrophicus (isolate 77S). By GC content these three isolates could not be differentiated from the rest of the Bacillus reference strains or new isolates (Appendix 5). Type strains of B. aerophilus/B. pumilus and B. methylotrophicus and the corresponding new isolates from this study fell into Bergey’s Group A with B. subtilis/B. licheniformis in the N-J phylogenetic tree (Figure 4.6) and ARDRA clustering (Figure 4.4), while isolate 32Le together with B. panaciterrae branched separately into Bergey’s Group K (Ludwig et al., 2009) in the overall phylogenetic tree (Figure 4.5). However, these classifications were not supported by ARDRA patterns where the isolates 32Le and 24M fell into the ARDRA-mixed cluster C and isolate 77S fell into ARDRA cluster B, but the reference type strain of B. panaciterrae fell into ARDRA cluster E, and the type species B. methylotrophicus and B. aerophilus/pumilus clustered with ARDRA cluster D that contained B. subtilis and B. licheniformis (Figure 4.4). This suggests that the three singular isolates may require more confirmation of their species identification, and that identification by GenBank BLAST may not reveal the correct identification.

B. panaciterrae is represented by only one type strain (Gsoil1517) isolated from a ginseng field (Ten et al., 2006), and has not been previously reported as an isolate from a hot-spring environment. However, only a tentative conclusion can be made that this is the first report of B. panaciterrae being isolated from hot springs because there is only one type strain represented in this group and therefore statistically inconclusive, and by ARDRA clustering, this isolate 32Le was different from the reference strain. However, its novelty and difference from the other Bacillus isolates begs further investigation.

Another example of the complexity of Bacillus identification relates to isolate 24M which, with a BLAST search, convincingly matched (99.85%) with B. aerophilus and B. stratosphericus. However, these reference strains were isolated (Shivaji et al., 2006) from samples of high altitude atmospheric cryotubes. Recently, Branquinho et al. (2015) suggested that the nomenclature of these be dropped from bacterial systematics as B. aerophilus and B. stratosphericus were not represented in any type of culture collection, and that they should be absorbed into the group of B. pumilus. Thereafter, Liu et al. (2015c) reported that B. aerophilus was actually B. altitudinis, and that B. stratosphericus was actually a Proteus sp. This finding is a prime example where a GenBank BLAST result matches up to nomenclature that is already dubious and questionable. Furthermore, isolate 24M does not cluster with B. pumilus or B. aerophilus using ARDRA, and by phylogenetic analysis it remains inconclusive. Bacillus aerophilus has not been isolated from hot springs although B. pumilus has been reported (Al- Qodah et al., 2013). One needs to be aware of the fact that a bacterial isolate that has a 99.85%

match to 16S rDNA sequences in databases within the public domain can possibly be a different species.

4.4.5.3 Family Paenibacillaceae genus Brevibacillus

Four isolates, namely 16S, 36Li, 70T and 85Li were identified as Brevibacillus spp. Paenibacillaceae refers to “nearly Bacillus”, and phylogenetically, the genus Brevibacillus has been shown to be distinct from the genus Bacillus (Xu & Cote, 2003). Brevibacillus is generally thought to be mesophilic although in this study isolates 16S and 36Li were isolated at 53 °C. Isolation of mesophilic Brevibacillus from higher temperatures is common due to the presence of heat-tolerant spores, and Brevibacillus sp. has been reported previously from hot springs (Inan et al., 2012; Cihan et al., 2012).

Even though there are challenges in precise identification of isolates, the aerobic Gram-positive spore-forming bacteria isolated in this study were similar to those reported in other investigations. Narayan et al. (2008) reported that of 104 isolates from hot springs in Fiji, 58% were A. flavithermus and 19% were B. licheniformis/Geobacillus stearothermophilus. Anoxybacillus, Brevibacillus, Geobacillus and Bacillus made up the 76 isolates cultured from hot springs in Turkey (Derekova et al., 2008). Of 115 isolates, Cihan et al. (2012) listed seven genera in hot springs in Turkey which included Anoxybacillus, Brevibacillus, Geobacillus and Paenibacillus. From hot springs in Morocco, Aanniz et al. (2015) found that 97.5% of 240 isolates were Bacillus sp. including B. licheniformis (n = 119), B. subtilis (n = 6) and B. pumilus (n = 3).

4.4.6 Analysis of unknown isolates

The GC content of four isolates (with a >97% BLAST match to published 16S rDNA sequences) listed in Appendix 5 was significantly different from within their group, i.e. isolate 11T within the Anoxybacillus group, and isolates 1T, 14S and 33Li within the Bacillus group. By ARDRA analysis, these differences were confirmed because all fell into the B cluster, a loosely defined group of isolates, and not into cluster A of Anoxybacillus or cluster C of Bacillus strains (Figure 4.4) as predicted by 16S rDNA comparison. Therefore, isolates that are clearly identified by BLAST could be further differentiated by ARDRA.

In order to attempt to assign unknown bacterial isolates that had a poor match to 16S rDNA sequences of known genera in public databases (<95%) listed in Table 4.1, a combination of three tools were used, namely BLAST to a public sequence database, GC content and ARDRA.

Included in the analysis were six unknown isolates from this study that gave a poor match (<94%) to published 16S rDNA sequences (isolates 15S, 52M, 53M, 73T, 75S and 86Li), as well as three “unknown Bacillus” isolated from hot springs in India (clone TPB_GMAT_AC4; GenBank HG327138.1), Indonesia (clone KSB12; GenBank JX047075.1) and China (clone DGG30; GenBank AY082370.1) from the GenBank database that remain unidentified.

Isolate 15S was confirmed Bacillus sp. by GC content, and with ARDRA analysis it was tentatively grouped with cluster D (B. subtilis/B. licheniformis). Similarly, isolate 52M was confirmed Brevibacillus sp. with GC content and ARDRA clustering (Appendix 5, Figure 4.4). However, isolate 53M was only 92% similar to Brevibacillus sp. by BLAST, did not cluster with Brevibacillus by ARDRA, but rather with Bergey’s Group C Bacillus, Bacillus siralis. Its GC content of 56.73% was a standard deviation different from that of Brevibacillus reference strains that had an average GC content of 55.8%, thus suggesting that isolate 53M was probably not Brevibacillus, and BLAST results would be misleading in this case.

Using BLAST, isolate 75S was found to have only a weak match (94%) with sequences of an “uncultured bacterium”. However, its high GC content of 57.43% suggested that it was not Bacillus (average 55.13%) or Brevibacillus (average 55.8%), but could be Anoxybacillus (average 56.4%). Using ARDRA, isolate 75S was grouped with Anoxybacillus supporting this tentative grouping according to GC content.

Isolate 86Li was 96% similar to sequences of Aneurinibacillus by BLAST, and supported by GC content percentage, but not by ARDRA. In the N-J phylogeny, it fell between Aneurinibacillus and Brevibacillus but differed from both, suggesting that further analysis is required in this case. This could include comparison of another gene sequence e.g. rpoB (Case et al., 2007) to confirm this phylogenetic placement, but was put on the back burner to focus on the biotechnology applications as a priority.

Solibacillus, an undefined member of the family Bacillaceae was included in this study because isolate 73T was found to have 95% similarity to sequences of Solibacillus using BLAST. Its “different” status was confirmed by a lower GC content of 53.91%, but by ARDRA it was grouped with the loosely defined C cluster rather than the reference strain of Solibacillus (Figure 4.5). In addition, by N-J phylogenetic analysis, it grouped with Aneurinibacillus sp (Figure 4.6) and could therefore not be assigned to any genus with any degree of certainty.

Three “unknown” sequences from previous hot-spring studies from India (uncultured Bacillus sp clone KSB12), Indonesia (uncultured Bacillus sp. clone TPB_GMAT_AC4) and China (uncultured Bacillus sp. clone DGG30) were obtained from GenBank, and analysed with the

four molecular tools to determine whether they could be placed in a cluster with reference strains or with isolates from this study. Clone DGG30 from China, had a GC content of 53.2% and a standard deviation different to that of Bacillus, Anoxybacillus and Brevibacillus. It clustered with the ARDRA group of reference strain of Solibacillus with a similar GC content of 53.2%, suggesting that there may be a relationship. Clone KSB12 from India was confirmed and grouped with Bacillus sp by GC content; however, by ARDRA, it clustered with Anoxybacillus. Indonesian isolate Clone TPB_GMAT_AC4 was confirmed to be Bacillus sp by GC content and ARDRA grouping with cluster D (B. subtilis/B. licheniformis). With the additional analysis, it is possible to further place this “unknown” into the Bergey’s Bacillus group A (Ludwig et al., 2009). Thus, previously obtained sequence data of uncultured microorganisms can be analysed in retrospect, and found to cluster with known bacteria at a genus level using a combination of tools, namely comparison of 16S rDNA sequences with GenBank sequences using BLAST, GC content percentage and computer-simulated ARDRA, adding to the pool of information regarding the bacterial diversity and population studies in hot springs.

CHAPTER FIVE: CULTURED EMERGING OPPORTUNISTIC PATHOGENS (PHYLA ACTINOBACTERIA AND PROTEOBACTERIA) IDENTIFIED IN HOT-SPRING WATER THROUGH ISOLATION AND PHYLOGENETIC ANALYSIS

5.1 INTRODUCTION

In developing countries, scarce water resources are used for drinking and domestic purposes, and monitoring of human waterborne pathogens is critical (Pandey et al., 2014). Generally bathing in hot spring water improves well being, or balneotherapy (Nasermoaddeli & Kagamimori, 2005). Infections are uncommon and associated with level of human activity (Singh et al., 2013a). Emerging opportunistic pathogens are important in populations that are immunocompromised through HIV/AIDs, malnutrition or age (very young or very old), even though these bacteria are commonly found in the environment (Berg et al., 2005). Metagenomics studies of hot springs in Limpopo, SA suggested that there is a large diversity of microorganisms as indicated by the presence of their DNA (Tekere et al., 2011; 2012); however, it is not indicative of viability which can only be discovered from isolation by culture. Apart from Legionella (Kurosawa et al., 2010) and Pseudomonas (Barna & Kadar, 2012), there have been no reported associated infections by emerging opportunistic bacteria from hot springs. The population of SA is afflicted with both a 12.7% prevalence of HIV, 10.4% that are very old or very young and a high rate of malnutrition resulting in potential susceptibility to emerging opportunistic infections (Section 2.3). As part of a diversity study of cultured bacteria from hot springs in the Limpopo, emerging opportunistic pathogens will be identified and their impact on water safety will be assessed.

5.2 METHODOLOGY

Water and sediment samples were taken from five hot springs in the Limpopo Province, SA as described in Section 3.1.1. The geographical locations are also indicated. Section 3.3 describes how the bacteria were isolated on five different media at two different incubation temperatures of 37 °C and 53 °C.

DNA was extracted and used to amplify the 16S rDNA with universal primers for the PCR. Partial sequences were obtained from the resulting amplicon as indicated in Section 3.4. DNA sequences of the 16S rDNA amplicons were submitted to a public database, GenBank and accession numbers are listed (Section 3.5). In Section 3.6 a description is given of how the

partial 16S rDNA sequences were aligned and analysed for phylogenetic relationships. DNA extracted from water samples was used for detection of Legionella by real-time PCR of the 16S rDNA as described in Section 3.9. The isolates were tested against ten different antibiotics using the Kirby-Bauer disk diffusion assay according to Section 3.10.

5.3 RESULTS

5.3.1 Isolation of bacteria

Since this was a study on microbial diversity and not a quantitative investigation, colonies were picked based on differences in colony morphology. The isolation conditions for 16 isolates are described in Table 5.1. The majority were isolated from water samples, although two isolates, namely 42T and 44M, were from the sediment.

Table 5.1: List of Actinobacteria and Proteobacteria isolates with isolation conditions

Isolation Isolate No Site temperature (° C) Sample Isolation media 57T Tshipise 37 Water Nutrient agar Actinomycete isolation 58T Tshipise 37 Water agar 87T Tshipise 25 Water minimal Luria agar 61T Tshipise 25 Water Cyanobacterial agar 72T Tshipise 37 Water Nutrient agar 80Le Lekkerrus 37 Water Nutrient agar 79M Mphephu 37 Water Nutrient agar 44M Mphephu 53 Sediment Nutrient agar 55M Mphephu 37 Water Potato dextrose agar Actinomycete isolation 37Li Libertas 53 Water agar 42T Tshipise 53 Sediment Nutrient agar 59Le Lekkerrus 37 Water minimal Luria agar 5T Tshipise 53 Water minimal Luria agar Actinomycete isolation 27M Mphephu 53 Water agar 69Le Lekkerrus 25 Water Cyanobacterial agar 31Le Lekkerrus 53 Water Nutrient agar

There was no correlation between the number of isolates and temperature, as the warmest hot spring, Siloam at 69 °C, yielded no isolated bacteria while the second warmest site, Tshipise at 55 °C, was found to have the highest number of isolates of seven. Furthermore, the coldest water temperature at Mphephu at 42.7 °C did not yield the highest number of isolates. Only one isolate was obtained from Libertas, while Lekkerrus and Mphephu yielded 4 isolates each. This was however not a quantitative study, rather qualitative where microbial diversity was investigated.

Figure 5.1: Comparison of number of isolates (Actinobacteria and Proteobacteria) obtained on different media

By comparing the different media used for isolation of all the isolates (Actinobacteria and Proteobacteria) from all five hot springs, NA performed the best resulting in the highest number of isolates, and PDA, which is generally used for the isolation of fungi, was the poorest performer (Figure 5.1). However, this is only an observation as Bacillus spp. and related species (discussed in Chapter 4) were isolated from the same plates and could have competed for space and nutrients.

5.3.2 16S rDNA amplicon sequencing

The consensus sequences were compared to two databases, GenBank and Ez-taxon-e and the highest similarities (as percentage similarities) and accession numbers are given in Section 3.5.2 and Table 5.2. Values >95% suggest a match in genus, while a value of >99% suggests a match in species. Similarities of 86-95% can only be identified to a family level (Yarza et al, 2014). As

indicated in Table 5.2, isolates of Sphingomonas echinoides, Hafnia alvei, Tepidimonas fonticaldi, Gulbenkiania mobilis, Ralstonia mannitolilytica, Cupriavidus gilardii, Kocuria turfanensis and Arthrobacter luteolus were identified to species level, while isolates of Cronobacter, Silanimonas, Tepidimonas and Zoogloea were identified only to genus level. Four isolates (55M, 59Le, 61T and 72T) were unknown as they matched poorly with the published sequences in the public domain. Sequences were submitted to GenBank with their relevant accession numbers listed in Table 5.2 (Section 3.5).

Table 5.2: Closest match of 16S rDNA sequence from hot-spring isolates with GenBank and EzTaxon-e with percentage similarity and associated accession numbers

Isolate GenBank PHYLUM; CLASS; SUBMIT No BLAST ACCESS No EzTaxon-e ACCESS No FAMILY GENUS SPECIES ACCESS No Kocuria Actinobacteria; turfanensis Actinobacteria; Kocuria

57T Kocuria sp. 93% HQ323439.1 99.91% DQ531634 Micrococcaceae Kocuria turfanensis MF120234 Arthrobacter Arthrobacter Actinobacteria; luteolus sp. luteolus Actinobacteria; Arthrobacter

58T 99% DQ486130.1 99.56% AJ243422 Micrococcaceae Arthrobacter luteolus MF120235 Proteobacteria; uncultured Skermanella alpha Azospirillum aerolata Proteobacteria;

61T 93% HG917273.1 92.14% DQ672568 Rhodospirillaceae MF120236 Proteobacteria; Sphingomona alpha Sphingomonas s echinoides Proteobacteria; Sphingomon Sphingomon

87T echinoides 99% KP208156.1 99.9 JH584237 Sphingomonadaceae as as echinoides MF120239 beta Proteobacteria; beta Proteobacteriu KM054705. Zoogloea Proteobacteria;

5T m 99% 1 98.14% U46748 Rhodocyclaceae Zoogloea MF120227 Gulbenkiania Proteobacteria; beta Gulbenkiania NR_042548. mobilis Proteobacteria; Gulbenkiania

27M mobilis 99% 1 99.93% AM295491 Neisseriaceae Gulbenkiania mobilis MF120228 Cupriavidus Proteobacteria; beta Burkholderia gilardii Proteobacteria; Cupriavidus

31Le sp. 99% AY005039.1 99.52% EU024163 Burkholderiaceae Cupriavidus gilardii MF120229 Proteobacteria; beta Tepidimonas Proteobacteria; Tepidimonas fonticaldi unclassified

37Li sp. 99% KF206375.1 98.01% JN713899 Burkholderiales Tepidimonas MF120230 Proteobacteria; beta uncultured Tepidimonas Proteobacteria; Tepidimonas fonticaldi unclassified Tepidimonas

42T 99% HF912299.1 99.81 JN713899 Burkholderiales Tepidimonas fonticaldi MF120231 Proteobacteria; beta uncultured Tepidimonas Proteobacteria; Tepidimonas fonticaldi unclassified

55M 92% HF912299.1 92.43% JN713899 Burkholderiales MF120232 Proteobacteria; beta uncultured Aquabacteriu Proteobacteria; Aquabacterium m fontiphilum unclassified 59Le 90% KF598759.1 90.51% EF626687 Burkholderiales MF120233 Ralstonia Proteobacteria; beta Ralstonia mannitolilytic Proteobacteria; mannitolilyti

69Le Ralstonia 97% CP010799.2 a 99.26% AJ270258 Burkholderiaceae Ralstonia ca MF120237 Proteobacteria; gamma Silanimonas sp. Silanimonas AUBD0100 Proteobacteria;

44M 99% KF206368.1 lenta 92.52% 0017 Xanthomonadaceae Silanimonas MF144571 Cronobacter Proteobacteria; uncultured sakazakii BAWU0100 gamma

72T bacterium 93% AY959011.1 89.64% 0071 Proteobacteria; MF144572

Enterobacteriaceae

Proteobacteria; Hafnia gamma Hafnia alvei paralvei Proteobacteria;

79M 99% KC210872.1 99.43% FM179943 Enterobacteriaceae Hafnia Hafnia alvei MF120238 Proteobacteria; Cronobacter gamma Cronobacter dublinensis Proteobacteria;

80Le sakazakii 95% HQ880369.1 91.38% EF059892 Enterobacteriaceae Cronobacter MF144573

5.3.3 Phylogenetic analysis

Most of the cultured isolates were phylum Firmicutes, genus Bacillus, which has been described and discussed in Chapter Four. To include all the isolates in the phylogenetic analysis, only partial sequences of the 16S rDNA sequences were used for Actinobacteria (947 bp) and Proteobacteria (510 bp) allowing alignment to the shortest fragment of the group.

A 947 bp fragment of the 16S rDNA amplicon after Multiple Sequence Comparison by Log- Expectation (MUSCLE) alignment for the Actinobacterial sequences were analysed as an unrooted parsimony tree (SeaView) with bootstrap values (Figure 5.2). Two isolates identified in this investigation, namely 57T (pink pigment) and 58T (yellow pigment) grouped with K. turfanensis and A. luteolus, respectively, as predicted from the BLAST results with GenBank and EzTaxon-e given in Table 5.2. Reference strains are included in the analysis (GenBank accession numbers listed in Table 5.2 and Section 3.5), and pathogenic isolates are Arthrobacter woluwensis, Arthrobacter cumminsii, Arthrobacter mysorens, K. rhizophila, K. marina, K. varians, K. kristinae and K. rosea (Purty et al., 2013). Three unknown sequences from hot- spring isolates isolated in other studies, and listed in the GenBank database, were included. Arthrobacter NCCP from hot springs in Pakistan grouped with the pathogenic group of A. woluwensis. Uncultured Arthrobacter clone BR5clone TPB_GMAT_5_1 grouped with the cultured Arthrobacter GM37 isolate from hot springs in India suggesting they could be closely related. Kocuria B38, an isolate from hot springs in India, grouped with K. flava.

Figure 5.2: Unrooted parsimony tree for Actinobacteria showing placement of isolate 57T with Kocuria and isolate 58T with Arthrobacter, with bootstrap values

Figure 5.3: Rooted parsimony tree for Proteobacteria showing grouping with alpha-, beta- and gamma- Proteobacteria isolates with bootstrap values

Following MUSCLE alignment (SeaView), a parsimony tree of the 16S rDNA sequence of the Proteobacteria isolates was drawn using partial sequence of 510 bp, with bootstrapping (Figure 5.3). A methanogenic archaeon (GenBank accession No DQ372975.1) was the outgroup used. The results of this investigation showed that three classes of the Proteobacteria were represented, namely alpha-Proteobacteria (n = 2), beta-Proteobacteria (n = 8), and gamma- Proteobacteria (n = 3). Reference type strains were included in the analysis (Appendix 4) and the opportunistic pathogens are Sphingomonas paucimobilis, C. sakazakii, H. alvei and H. paralvei, Tepidimonas arfidensis, R. mannitolilytica, Ralstonia pickettii, Cupriavidus metallidurans and C. gilardii.

From published sequences in GenBank, four isolates that have not been cultured but sequenced from hot springs in other studies, were included in the phylogenetic tree to establish whether they were similar to isolates obtained in this study, or to determine whether their identity could be further characterised.

5.3.4. Legionella real-time polymerase chain reaction (RT-PCR)

A total of four water samples were analysed in triplicate. All samples analysed were negative for the 16S rDNA amplified. A real-time PCR assay for Legionella is given in Figure 5.4A below showing the PCR cycling curve of the positive control at different concentrations indicating that the even at low resolution, the samples were negative. Figure 5.4B shows high- resolution melt curve analysis of the PCR reaction for the positive control with a mean melting temperature of the positive samples was 86.1 °C ± 0.1. If there was a peak at another temperature, this might indicate a false positive in the samples.

Figure 5.4A: PCR cycling for the detection of L. pneumophila in water samples; a – g are the dilutions 100 – 10-6 with highest concentration a = 6 368 ng/µL and the lowest concentration g = 0.006368 ng/µL

Figure 5.4B: High-resolution melt curve analysis for the identification of L. pneumophila

5.3.5 Antibiotic resistance

Table 5.3: Antibiotic resistance of 40 hot-spring isolates against 10 antibiotics: carbenicillin (CAR), gentamicin (GEN), kanamycin (KAN), streptomycin (STP), tetracycline (TET), chloramphenicol (CHL), ceftriaxone (CEF), co-trimoxazole (COT), nalidixic acid (NAc) and norfloxacin (NOR) in µg/ml. Values of 0 indicate resistance while numerical values are zones of inhibition in millimetres denoting sensitivity

Isolate CAR GEN KAN STR TET CEF CHL COT NAc NOR no Identification 100* 10* 30* 10* 30* 30* 30* 25* 30* 10*

57T Kocuria turfanensis 15 9 12 7 12 0 14 16 0 2 Arthrobacter 58T luteolus 0 5 3 4 9 1 6 11 0 2

79M Hafnia alvei 0 2 3 1 4 0 5 6 3 7

80Le Cronobacter sp. 1 4 5 5 6 2 6 6 3 11

unknown 72T Enterobacteriaceae 5 6 10 5 7 0 0 16 5 14

*Antibiotic concentrations are given in µg/mL

Isolates Kocuria, Arthrobacter and Hafnia spp. were resistant to two antibiotics in different combinations of CEF, NAc and CAR as indicated in Table 5.3. Isolate Cronobacter sp. was sensitive to all antibiotics while the unknown isolate (72T) belonging to the family Enterobacteriaceae, was also resistant to two antibiotics, CHL and CEF.

5.4 DISCUSSION

Metagenomic analysis of the water samples of these hot springs in Limpopo, SA, revealed that the dominant phylum present was the Proteobacteria, and formed up to 30% of the microbiome of Tshipise and Mphephu, while phylum Actinobacteria was found to occur at <1%. Of the Proteobacteria, the beta- and gamma-Proteobacteria were found to be the majority (Tekere et al., 2011), and as a result isolation of Proteobacteria would be expected under these conditions. However, metagenomic studies detect the genetic make-up of a population of microorganisms in these environmental samples, and the diversity is reported as proportions of different DNA that match with known published gene sequences. Currently, molecular assay, i.e. DNA-based detection does not distinguish between dead and viable cells. Gram-positive spore-forming Bacillus of the phylum Firmicutes are the most predominant bacteria isolated from hot springs globally (Khiyami et al., 2012; Obeidat et al., 2012; Pandey et al., 2015), although there are sporadic reports of isolation by culture of Pseudomonas spp. of the phylum Proteobacteria in Algeria (Yakhlef & Darbouche, 2012) and Iceland (Thorolfsdottir & Marteinsson, 2013).

5.4.1 Phylum Actinobacteria

From sediment samples, two pigmented mesophilic isolates 57T and 58T were identified to be K. turfanensis and A. luteolus, respectively, using a BLAST with GenBank and EzTaxon-e (Table 5.2) and confirmed with phylogenetic analysis (Figure 5.2).

Kocuria sp. ASB107 is closely related to K. polaris and K. rosea, which was isolated from Iranian Ab-e-Siah hot springs, showed resistance to gamma and UV rays with a temperature growth range of 0 to 37 °C (Asgarani et al., 2012). Another isolate of K. rosea MG2 from the same Iranian hot spring optimally grew at pH 9.2 and at a temperature of 28 °C. It could also tolerate15% NaCl salinity, high doses of radioactivity and grew in the presence of 4% hydrogen peroxide (Gholami et al., 2015), providing further proof that members of this genus are capable of multiple extreme resistance. An unpublished Kocuria sp. B38 was isolated from Bakreshwar hot springs in India, and included in the phylogenetic tree to establish whether there was any similarity between that one and the one isolated during this investigation. Kocuria B38 grouped with Kocuria flava, an airborne Actinobacteria (GenBank KC492107), with a bootstrap value of 90%, suggesting that it is possible to culture closely related Kocuria isolates that were previously reported uncultured from hot springs at other sites. Pathogenic Kocuria sp. are K. rhizophila, K. marina, K. rosea listed in reviews (Purty et al., 2013; Savini et al., 2010; Pulcrano et al., 2017), K. kristinae (Ma et al., 2005; Chen et al., 2015a), K. varians (Bhaiyat et

al., 2013). All, except for K. rosea fell into a separate group. Even though neither K. turfanensis nor K. flava have been reported as opportunistic pathogens, it does not exclude the possibility that they have the ability to acquire virulence factors.

Similarly, some radiation-resistant Arthrobacter spp. have been isolated from hot springs in Japan (Yoshinaka et al., 1973) and Tibet (Yang et al., 2015). Three 16S rDNA sequences from Arthrobacter sp. isolated from hot springs were included in the phylogenetic analysis (Figure 5.2). Interestingly Arthrobacter NCCP 1348 from Pakistan (GenBank Accession LC065375) was related to pathogenic A. woluwensis, while Arthrobacter sp. GM37AC3 K2 strongly grouped with uncultured Arthrobacter sp. clone TPB_GMAT_AC3. This latter observation was not surprising since both came from the same investigation of a hot spring study in India (Appendix 4). Three species of pathogenic Arthrobacter were included in the phylogenetic analysis and A. woluwensis and A. mysorens were closely related. Interestingly, A. cumminsii grouped closed with the pathogenic Kocuria species. A literature search did not reveal A. luteolus as an opportunistic and emerging pathogen; however, the type strain of A. luteolus was isolated originally from human clinical specimens (Wauters et al., 2000) and therefore this does not entirely exclude this species from being a potential opportunistic pathogen.

5.4.2 Phylum Proteobacteria

Fourteen Proteobacteria isolates were represented by three classes: alpha- (n = 2), beta- (n = 8) and gamma-Proteobacteria (n = 4) (Table 5.2). Of these, four isolates could not be classified up to genus level as they were <97% similar to GenBank database entries. Four could be identified up to genus level, and six identified up to species level.

5.4.2.1 Alpha-Proteobacteria

Isolate 87T was similar to S. echinoides by >99% and could therefore be identified to species level. However isolate 61T, matched poorly in BLAST (92%) and remains an unknown member of the alpha-Proteobacteria. Sphingomonas sp. has been isolated in the environment of soil (Zhou et al., 2014), pristine rivers (Chen et al., 2013b) and from hot springs in China (Briggs et al., 2014). Sphingomonas echinoides is related but different to pathogen S. paucimobilis (Figure 5.3).

5.4.2.2 Beta-Proteobacteria

Within this class, single isolates could be identified to a species level of H. alvei (isolate 79M), G. mobilis (isolate 27M), R. mannitolilytica (isolate 69Le) and C. gilardii (isolate 31Le).

Isolates 37Li and 5T could only be identified to a genus level as Tepidimonas sp. and Zoogloea sp. respectively. Two isolates, namely 55M and 59Le remain unknown due to their low match <97% with published data as indicated in Table 5.2.

Opportunistic pathogens within this class of beta-Proteobacteria group are included in the phylogenetic tree (Figure 5.3) as R. mannitolilytica (Ryan & Adley, 2014) and R. pickettii (Coenye, 2002), C. metallidurans (Langevin et al., 2011) and C. gilardii (Karafin et al., 2010) and T. arfidensis (Ko et al., 2005). Gulbenkiania mobilis has not been reported to be an opportunistic pathogen. As a result, three of the four isolates identified to species level have been reported as opportunistic pathogens. As indicated in Table 2.1, these genera have been reported previously isolated from hot-spring environments, suggesting that these common environmental bacteria have the capacity to adapt to harsh conditions of high temperature. Tepidimonas has been reported as thermophilic growing optimally at 55 °C (Chen et al., 2013c) and Sphingomonas as chlorine resistant (Sun et al., 2013). Cupriavidus has been isolated from drinking water (Van Belkum, 2011), but not hot springs.

The presence of unknown bacteria further supports the need to study such environments to contribute to the knowledge of bacteria diversity in hot spring niches. Including the two unknown isolates from this investigation, sequences of four beta-Proteobacteria isolates from hot springs (G. mobilis V28, Cupriavidus NCCP, uncultured Tepidimonas clone TPB_GMAT_5_1 and Tepidimonas BR5) were obtained from GenBank and included in the phylogenetic analysis to see if there was a similarity with isolates in this study or other reference strains. Gulbenkiania mobilis V28 from Indian hot springs was similar to our G. mobilis isolate 27M. No further identification to a species level could be obtained with Cupriavidus NCCP and uncultured Tepidimonas sp. clone TPB_GMAT_5_1. Isolate 42T which was identified as T. fonticaldi, grouped closely with Tepidimonas BR5, an isolate from hot springs in India. By extending the study to include retrospective 16S rDNA data from other hot springs investigations, a broader knowledge of the diversity of culturable hot spring microorganisms can be gained.

5.4.2.3 Gamma-Proteobacteria

This class includes Cronobacter, Hafnia, Silanimonas and Legionella. Isolate 79M was identified as H. alvei, isolate 80Lk as Cronobacter sp. and isolate 44M as Silanimonas sp. Isolate 72T was the single unknown in this group, matching only 89% with Cronobacter sp. sequence in the public domain. Gamma-Proteobacteria especially Legionella have been isolated from hot springs (Ishizaki et al., 2016). Other gamma-Proteobacteria isolated from hot springs

worldwide are listed in Table 2.1 including pathogenic C. sakazakii in Malaysian hot springs (Jimat et al., 2015). The reference strains of gamma-Proteobacteria opportunistic pathogens in the phylogenetic tree are indicated as H. alvei, H. paralvei and C. sakazakii (Appendix 4). Hafnia paralvei, formerly called H. alvei (Huys et al., 2010), is the only species in this genus.

Berjeaud et al. (2016) commented on the paradox that L. pneumophila is highly sensitive but still able to persist in the environment. Investigators have also observed that there is no correlation of Legionella loads with temperature, in water (Qin et al., 2013; Walczak et al., 2016). Although the aim of this study was to investigate the diversity of cultured microbes from hot springs, detection of L. pneumophila by RT – PCR was included because Legionella is the most common opportunistic pathogen of hot-spring water (Sukthana et al., 2005). The sensitivity of the assay was 0.006 ng/µL which excludes the possibility of lack of detection associated with the assay. There are several possible reasons to explain the lack of detection of Legionella in this study. From the literature, not all hot springs were positive for Legionella being as low as 3% in Japan (Kobayashi et al., 2014) to as high as 51% in China and 71% in Thailand. It is possible that the sample size was too small in this study, or that the presence of Legionella could be seasonal. Seasonal variations have been reported in wastewater sites in French river watershed samples (Parthuisot et al., 2010) as well as in water reservoirs in Taiwan (Kao et al., 2013). A follow-up investigation could include biofilm and sediment samples as Legionella replicates in protozoa that naturally colonise and persist in biofilms (Abdel-Nour et al., 2013).

5.4.3 Antibiotic resistance of opportunistic emerging pathogens

Antibiotic resistance profiles were determined for five isolates (57T, 58T, 72T, 79M and 80Le) as indicated in Table 5.3. Of those, four exhibited antibiotic resistance to two antibiotics, with one isolate being completely sensitive to the eight antibiotics tested. All four isolates differed in the antibiotics they were resistant to, and combinations of resistance towards CEF, NAc, CAR and CHL were observed. Resistance to CEF was reported in three of the four isolates while only one isolate showed resistance to CHL.

Antibiotic resistance of emerging opportunistic bacteria is poorly investigated (Savini et al., 2010). However, the results of this study corroborate previous findings as indicated below. Kocuria sp. isolate 57T was resistant to CEF and NAc and sensitive to CHL, TET, STREP, GEN and CAR in this study. In a review of antibiotic resistance in Kocuria (Savini et al., 2010), these same results were also found by other researchers. However, variability resistance was also reported to COT and KAN in previous studies, suggesting that Kocuria spp. are able to

acquire additional resistance genes than that reported in this study. Similarly, Hafnia spp. were reported to be resistant to CEF and sensitive to GEN, as found in this study with isolate 79M; however, resistance to TET and CHL has been previously reported (Abbott et al., 2011) again implying that Hafnia spp. are also able to acquire further antibiotic resistance mechanisms. The lack of resistance to any antibiotic of Cronobacter sp. isolate 80Le is consistent with that found in other studies of Cronobacter from foods in Brazil (Brandao et al., 2017). In a larger study of C. sakazakii from 78 domestic kitchens in the US, 67% were TET resistant and 48% NAc resistant. All were sensitive to GEN (Kilonzo-Nthenge et al., 2012). This suggests that multidrug resistance is highly likely in this group of bacteria. Antibiotic resistance of emerging opportunistic pathogens is important for two reasons. Firstly, it has implications in that should these isolates cause infections, therapy will be limited to the antibiotics to which they are sensitive. Secondly, by monitoring antibiotic resistance of environmental opportunistic microbial populations, one can also track these microbes as the etiological agent of an infection based on their antibiotic resistance profiles, to determine the source of nosocomial or community outbreaks.

This study provides evidence of the presence of different emerging opportunistic pathogens in hot springs in Limpopo, SA. To establish whether these bacteria have a significant impact on human health, literature on nosocomial infections, opportunistic infections of HIV patients and cystic fibrosis patients was reviewed. With the exclusion of Legionella, the reviews on opportunistic infections in HIV patients (Maartens, 2002; Holmes et al., 2003; Marais et al., 2006) and nosocomial opportunistic infections (Williams et al., 2013; Kanamori et al., 2016) do not mention the abovementioned bacteria in this study. However, Ralstonia sp. was mentioned as a risk factor for cystic fibrosis patients (Coenye, 2002; Coman et al., 2017).

5.4.4 Relationship between waterborne pathogens and opportunistic emerging pathogens

The question must be asked, if these opportunistic bacteria listed in Appendix 1 cause infections, then why are these not reflected in the literature review of waterborne pathogens? In fact, these environmental bacteria have been described in the literature as “rarely isolated, strange, atypical, uncommon and unusual” (Singh et al., 2013a; Dunn et al., 2011; Al-Grawi, 2008; Lai, 2001; Toh et al., 2011) pathogens which could be a clue as to why these bacteria are seemingly overlooked or underestimated in diagnosis as aetiological agents for infections. A more thorough literature search could confirm the significance of these microorganisms as being pathogenic; however, the sporadic reports listed in Appendix 1 should be of some concern and

alarm for clinicians, healthcare workers and medical diagnostic laboratories. It is possible that these bacteria are just not monitored because they are regarded as “harmless” in a healthy individual. However, this benchmark should be flexible to cover pathogens that may be emerging for a particular population, i.e. population with high levels of malnutrition, a high incidence of HIV/AIDS, and a large number of very young or elderly age groups. The designation of a pathogen as an emerging pathogen is vague as they are often based on subjective judgement of criteria. Pathogens are classified according to their level of biohazard (Van Belkum, 2011). The United States Centers for Disease Control and Prevention (CDC) categorises various diseases in levels of biohazard, Level 1 being minimum risk and Level 4 being extreme risk. Species classified as Biohazard Level 2 cause mild diseases which are unlikely to spread in the human population and for which an adequate therapy exists; they can be used safely in routine diagnostic work. Legionella, C. sakazakii, H. alvei, A. luteolus, Cupriavidus and R. mannitolilytica fall into this category. Biohazard Level 1 microbial species are non-infectious, and precautions are minimal. The Level 1 species have been demonstrated by in vitro and in vivo testing to be non-pathogenic. Tepidimonas, Kocuria, Gulbenkiania and Zoogloea fall into this lowest biohazard classification; however, they are also able to cause opportunistic infections. For example, Lewis et al. (2013) studied urinary tract infections in SA, and Kocuria sp. was isolated and cultured but was disregarded as the aetiological agent, rightly so according to Van Belkum’s (2011) classification. However, since this is in flux, the classification should be constantly reviewed and updated. Reviews of emerging infectious diseases published after the Netherlands Commission on Genetic Modification research report had appeared in 2011 (Van Belkum, 2011), named Sphingomonas (Dewan et al., 2014), Ralstonia (Ryan & Adley, 2014) and Kocuria (Purty et al., 2013; Paul et al., 2015). Another reason why these bacteria may have been rarely mentioned as pathogens is mis-identification as Sphingomonas echinoides was initially called Pseudomonas echinoides (Meric et al., 2009). Ralstonia and Tepidimonas were also previously called Pseudomonas (Coenye et al., 2003).

5.4.5 Significance in water-scarce developing countries

South Africa is a water-scarce country, and there is thus more human contact with groundwater in areas that do not have microbially safe tap water. Furthermore, these “pathogens” enter and may be concentrated in the food chain via the environment. Cronobacter sakazakii has been reported in infant formula (Witthuhn et al., 2007) and Hafnia in other foods (Janda & Abbott, 2006). Global warming has also resulted in climate change; Anttila et al. (2015) describe variations in the environment that generate outbreaks by opportunistic pathogens.

Further studies could look for virulence factors of the isolates and their interactions with human cells in culture, and to perform retrospective studies of local nosocomial outbreaks and cystic fibrosis patient samples to see whether these bacteria can be isolated from the clinical samples, or detected by molecular methods such as PCR. Since these microbes are common in the environment, trying to eradicate them would not be possible or reasonable; however, by monitoring their presence, health care workers can be aware of hot spring water as a reservoir should infections arise with unknown aetiology.

Metagenomic studies of these hot springs in Limpopo (Tekere et al., 2011), indicated that 0- 20% of the sequences were “unclassified bacteria’ suggesting that there is a great diversity of bacteria that has yet to be discovered and characterised. Four unknown bacteria were sequenced in this study (61T, 55M, 59Lk and 72T) as indicated in Table 5.2, and eight sequences (four Actinobacteria and four Proteobacteria) of partially identified isolates from other investigations of hot springs, were obtained from GenBank allowing comparison of 16S rDNA data. With such analysis, the uncultured Arthrobacter sp. clone TPB_GMAT_AC3 was found to be similar to Arthrobacter sp. GM37AC3 K2, and Arthrobacter sp. NCCP-1348 from a hot spring in Pakistan, was closely related to pathogenic A. woluwensis (Figure 5.2). Kocuria B38, from a hot spring in India, grouped with K. flava, an airborne bacterium from China, with a bootstrap value of 100%, suggesting that it was K. flava. This shows the value of 16S rDNA sequence data in identification of unknown bacteria in retrospect. Such information could give a broader view of the presence of potential opportunistic pathogens in other hot springs, and an understanding of how they may enter the water system via airborne or soil routes. Improvements could be made to this investigation by using different isolation culture media or enrichment methodology, and laboratory conditions to extend the number of isolates obtained. The dissolved oxygen of hot springs is negligible therefore we expect to find a large proportion of anaerobic microbes.

This study suggests that these ubiquitous environmental bacteria that are emerging opportunistic pathogens of humans are underestimated and overlooked. They occur in the plant rhizosphere (Berg et al., 2005), on recreational beaches (Mudryk et al., 2014; Yamahara et al., 2012), in mangrove estuaries (Ghaderpour et al., 2014) springs of devotion (Singh et al., 2013a), sauna floors (Kim et al., 2013) and hot springs (Sukthana et al., 2005) including those reported in this study.

CHAPTER SIX: ANTIBIOTIC AND HEAVY-METAL TOLERANCE IN CULTURED BACILLUS SPECIES AND OPPORTUNISTIC PATHOGENS (KOCURIA SPECIES AND HAFNIA ALVEI) FROM HOT SPRINGS IN LIMPOPO PROVINCE, SOUTH AFRICA

6.1 INTRODUCTION

Multiple antibiotic resistance (MAR) in bacteria especially pathogens is a huge and escalating problem for public health (Al-Bahry et al., 2014). Several studies have reported MAR in aquatic environments (Vaz-Moreira et al., 2014) and it is now regarded as a hazardous pollutant of water (Coutinho et al., 2013; Sanderson et al., 2016). However, AR is ancient, occurring in the environment for billions of years, intrinsically and naturally (Allen et al., 2010) reported in pristine antibiotic-free locations such as the Antarctic (Miller et al., 2009), remote isolated caves (Bhullar et al., 2012) and remote mountain streams (Lima-Bittencourt et al., 2007). This intrinsic AR differs from AR induced by the presence of external pharmaceutical sources, in terms of function and diversity (Sengupta et al, 2013; Perron et al., 2015). Hot and cold spring sites are sources of sampling possibilities for microorganisms that have had no previous exposure to man-made antibiotics and MAR has been investigated in hot springs (Akel et al., 2008; Sen et al., 2010; Sariözlü et al., 2012. A correlation between MAR and heavy-metal tolerance (HMT) has been reported in isolates from rivers (Icgen & Yilmaz, 2014; Matyar et al., 2014) with the co-selection and co-transference due to the presence of genes for MAR and HMT located on the same plasmid (Samanta et al., 2012a).

The aim of this study was to determine the level of resistance to antibiotics of cultured bacteria from hot springs in Limpopo Province, SA. A detection of tolerance to heavy metals may also be an indicator of AR if an association was found. The findings have implications for infections by pathogens and opportunistic pathogens further downstream. The study was also used to explore the possible use of hot springs to provide baseline levels of AR that occurs naturally in such a unique environment.

6.2. METHODOLOGY

The geographical location of the water and sediment samples sites is described in Section 3.1.1 including on-site analysis. The method used for isolation of bacteria from water and sediment, and generation of pure cultures is described in Section 3.2. Universal primers 8F, 27F and

1472R were used to amplify the 16S rDNA prior to Sanger sequencing (Section 3.4). The details of the disk diffusion assay to test isolates against ten different antibiotics (Section 3.10) and against eight heavy metals (Sections 3.11) is made clear. To determine if there was a statistical correlation between AR and HMT of the isolates, a Chi-squared test of independence was used (Section 3.21.1).

6.3. RESULTS AND DISCUSSION

6.3.1 Isolation and identification of bacteria

Forty bacterial isolates were used in this study (Table 6.1) consisting of two members of the family Micrococcaceae (genera Kocuria and Arthrobacter), one member of the family Enterobacteriaceae (genus Hafnia), and 37 members of the family Bacillaceae (genera Bacillus, Anoxybacillus, Aneurinibacillus and Brevibacillus). The majority of the isolates were obtained from water samples (87.5%), and 12.5% were isolated from sediment samples. Thirteen (32.5%) were isolated at 37 °C, while 67.5% were isolated at 53 °C.

Table 6.1: Isolates from five hot springs, Limpopo Province, SA, showing temperature and isolation media, and comparison of the 16S rDNA sequences with GenBank and accession numbers

Temp isolated ISOLATE* SAMPLE °C MEDIA** Phylum Genus GENBANK 57T water 37 NA Actinobacteria Kocuria Kocuria sp. HQ323439.1 Arthrobacter luteolus sp 58T water 37 Actino Actinobacteria Arthrobacter DQ486130.1 79M water 37 NA Proteobacteria Hafnia Hafnia alvei KC210872.1 Anoxybacillus rupiensis 3T water 53 NA Firmicutes Anoxybacillus KJ842629.1 Anoxybacillus rupiensis 4T water 53 10%LA Firmicutes Anoxybacillus AM988775.1 Anoxybacillus sp. ATCC 7T water 53 Actino Firmicutes Anoxybacillus KJ722458.1 Anoxybacillus sp. 13S water 53 NA Firmicutes Anoxybacillus FN432807.1 Anoxybacillus flavithermus 17S water 53 10%LA Firmicutes Anoxybacillus KF039883.1 Anoxybacillus sp. 19S water 53 Actino Firmicutes Anoxybacillus KP221933.1 Bacillus subtilis 1T water 53 NA Firmicutes Bacillus HM367735.1 Bacillus licheniformis 2T water 53 NA Firmicutes Bacillus HM631844.1 Bacillus licheniformis 6T water 53 10%LA Firmicutes Bacillus KJ729823.1 8T water 53 Actino Firmicutes Bacillus Bacillus sp. GU132507.1 9T water 53 PDA Firmicutes Bacillus Bacillus subtilis N366797.1 Bacillus licheniformis 10T water 53 PDA Firmicutes Bacillus HM631844.1 Bacillus subtilis 12S water 53 NA Firmicutes Bacillus KC634086.1

Bacillus licheniformis 15S water 53 10%LA Firmicutes Bacillus HM055609.1 Bacillus subtilis 21M water 53 NA Firmicutes Bacillus JN585712.1 Bacillus methylotrophicus 22M water 53 NA Firmicutes Bacillus JQ765577.1 Bacillus licheniformis 23M water 53 NA Firmicutes Bacillus KF737353.1 Bacillus licheniformis 30M water 53 PDA Firmicutes Bacillus KJ526854.1 Bacillus subtilis 33Li water 53 na Firmicutes Bacillus KC182058.1 Bacillus licheniformis 39T water 37 na Firmicutes Bacillus KF879197.1 Bacillus subtilis 40Le water 37 Actino Firmicutes Bacillus KP249695.1 Bacillus licheniformis 46S sediment 53 na Firmicutes Bacillus KP245784.1 Bacillus subtilis 47Li sediment 53 NA Firmicutes Bacillus KP249695.1 Bacillus subtilis 48Li sediment 53 NA Firmicutes Bacillus NR_118486.1 Bacillus tequilensis 51T sediment 53 PDA Firmicutes Bacillus KC992300.1 Bacillus subtilis 54T water 37 PDA Firmicutes Bacillus HM753614.1 Bacillus tequilensis 76S water 37 NA Firmicutes Bacillus KF025658.1 Bacillus methylotrophicus 77S water 37 NA Firmicutes Bacillus KP342210.1 78S water 37 Actino Firmicutes Bacillus Bacillus sp KF984420.1 Bacillus subtilis 83Li water 37 NA Firmicutes Bacillus KF533727.1 Solibacillus sylvestris 73T water 37 NA Firmicutes Solibacillus KF441704.1 Brevibacillus sp 16S water 53 10%LA Firmicutes Brevibacillus LN681596.1 Brevibacillus sp 36Li water 53 Actino Firmicutes Brevibacillus KM403208.1 Brevibacillus agric 52M sediment 53 PDA Firmicutes Brevibacillus JN812211.1 Brevibacillus formosus 85Li water 37 10%LA Firmicutes Brevibacillus KP165013.1 Aneurinibacillus 86Li water 37 10%LA Firmicutes Aneurinibacillus migulanus NR_113764.1

* Sample site of isolate where S (Siloam), T (Tshipise), M (Mphephu), Li (Libertas) and Le (Lekkerrus)

**Media used for initial isolation where NA (nutrient agar), PDA (potato dextrose agar), Actino (Actinomycete isolation agar), 10%LA (minimal Luria agar)

The family Bacillaceae are Gram-positive bacteria, typically forming heat-tolerant endospores, are easy to culture being non-fastidious, ubiquitous and in large numbers in the environment (Mandic-Mulec et al., 2015). They are therefore commonly isolated from hot springs. This includes Bacillus spp. (Sayeh et al., 2010), Anoxybacillus spp. (Cihan et al., 2012) and Brevibacillus spp. (Inan et al., 2012). Thirty-seven of the 40 isolates in this study belonged to this family with representation in the three abovementioned genera with predominantly Anoxybacillus spp. (n = 6) and Bacillus spp. (predominantly B. subtilis (n = 10) and B.

licheniformis (n = 8)). The sequences were deposited into GenBank and accession numbers are listed in Section 3.5. Because the majority of isolated bacteria form spores, it is not uncommon or surprising to isolate mesophiles preferring growth at 37 °C, together with thermophiles with optimal growth at 53 °C, where they exist tolerating, but not growing in higher water temperatures.

6.3.2 Antibiotic resistance

Forty bacterial isolates were tested against ten antibiotics representing six different classes: β- lactam (CAR), aminoglycosides (GEN, KAN, STP), TET, amphenicols (CHL, CEF), sulphonamides (COT) and quinolones (NAc, NOR) using the standard disk diffusion assay. Low levels of <10% were observed for STP (5%) and KAN (2.5%), while complete sensitivity was reported for CHL, GEN, TET, COT and NOR. The highest percentage of isolates was resistant to CEF (52.5%), while fewer isolates were found to be resistant to NAc (37.5%) and CAR (22.5%) (Figure 6.1).

MAR is defined as resistance to ≥3 antibiotics, and in this study ten antibiotics were tested. Consequently, an MAR index of 0.1 and 0.2 translates to antibiotic resistance towards one and two antibiotics, respectively. Most of the isolates fell into this range (Figure 6.2). Only a single

isolate 2T was resistant to three different antibiotics in this study (2.5%); equal numbers were resistant to one or two antibiotics (37.5% each), while 22.5% showed sensitivity to all the antibiotics tested (Figure 6.2).

Figure 6.2: Number of isolates from hot springs expressing different MAR index values tested against 10 different antibiotics ranging from 0 to 0.3

6.3.3. Antibiotic resistance (AR) in the environment

6.3.3.1 Hot springs and pristine environments

Since β-lactam resistance is ancient, widespread and natural in the environment, it is not surprising that some level of β-lactam resistance is found in hot-spring isolates. Three Bacillus isolates from Moroccan hot springs were β-lactam resistant (Filali et al., 1997). From Jordan hot springs, three Bacillus strains were tested against 18 antibiotics and all were sensitive to NOR and STP, similar to the findings of this study, but also sensitive to CAR (Akel et al., 2008). In another study, Sen et al. (2010) also reported that two isolates from hot springs in India were sensitive to ampicillin. A more recent study in India (Pathak & Rathod, 2014) of 10 Bacillus isolates and 19 antibiotics, found that 40% of isolates were resistant to penicillin and 10% to sulphatriad, but sensitive to all the other antibiotics. Sariözlü et al. (2012) found that 30% of isolates were resistant to ampicillin but they observed no resistance to TET and GEN. The results of this study corroborate the AR profiles reported for other hot springs, and confirm that lower AR levels were observed in this study than those reported in other aquatic bodies.

It is difficult to compare the AR profiles observed in one study with those reported for other pristine environments because different concentrations of antibiotics and methodologies might have been used. However, a comparison of the results of this study with those of other environments considered to be pristine reveals that remote cave microbiomes, Antarctic marine waters and pristine mountain rivers have higher AR levels, and that these sites are also influenced by wild animals, even migrating birds, that can play a role in AR dissemination. The Gram-positive subset of the cave biome (Bhullar et al., 2012) that had not been touched for 4 million years, showed 15% resistance to GEN, 25% to STP, 20% to TET and 38% to CHL. In this study, resistance against these abovementioned antibiotics was found to be less than 10% or completely sensitive. Bhullar et al. (2012) reported a 20% resistance to ampicillin, which is similar to the 23% observed in this study. However, the low-level resistance (10%) to ceftriaxone reported in that study (Bhullar et al., 2012) differed from that observed in our study as 56% of isolates were found to be resistant to CEF. A comparison of AR in Antarctic marine waters (De Souza et al., 2006) and the cave biome (Bhullar et al., 2012) showed higher percentage AR to all antibiotics tested in the Antarctic marine waters, and most isolates were resistant to three or four different antibiotics. In another study, Miller et al. (2009) described AR in seawater in the Antarctic and in penguin faeces with TET resistances varying from 1 – 14%. Enterobacteriaceae species from pristine freshwater mountain springs (n = 111) in Serra Do Capo National Park in Brazil, were studied by Lima-Bittencourt et al. (2007). They reported MAR of 61%, most frequently against β-lactam antibiotics and CHL. This discrepancy between “pristine” AR levels suggests that further investigation is required to define baseline intrinsic levels of environmental AR.

The environment of hot springs may be unique because high temperatures may introduce novel microbial genotypes due to the loss of DNA (Venton, 2013) or more lateral gene transfer that occurs at higher temperatures (Rhodes et al., 2011). In addition, biofilm formations encourage the production of naturally occurring “antibiotics” and the possibility of lateral gene transfer due to close proximity of different microorganisms (Yan et al., 2003). Thermus sp. isolated from Yellowstone hot springs, USA was found to carry plasmids not associated with antibiotic resistance transfer (Munster et al., 1985). Bacillus spp. and Streptococcus thermophilus from Himma and Maeen hot springs, respectively, in Jordan were found to have plasmids associated with the transfer of antibiotic resistance (Khalil et al., 2003), making lateral gene transfer a strong possibility.

In this study, Siloam is the only water sample that can be considered “truly pristine”, because the water is conveyed in a pipeline from a constantly flowing spring. Although water sampled

from Lekkerrus is also conveyed in a pipeline into a swimming pool, the flow is manually controlled, and water may be static for several hours. The other sites, Tshipise, Mphephu, and Libertas are open water bodies exposed to wild animals, runoff from rainfall and therefore susceptible to possible contamination. Sampling was carefully done to ensure that there was no contamination caused by human activities such as recreational swimming. The MAR of 2.5% (one isolate of 40) is far lower than that described for other sites regarded as “pristine”, e.g. Lima-Bittencourt et al. (2007) described 61% MAR for isolates from “pristine” mountain streams.

6.3.3.2 Antibiotic resistance (AR) of isolates from hot springs as a baseline measure of natural environmental AR without acquired AR due to human activities

Globally, AR is currently a growing public health concern because AR levels have increased in both benign and pathogenic bacteria (Woolhouse et al., 2016), and in the environment (Davies & Davies, 2010; Finley et al., 2013). Antibiotics are used in the treatment of human and animal infections, the control of spoilage of food produce, and in aquaculture (aqua-farming) for control of bacterial diseases in fish. As a result, AR levels have been associated with human activity. Direct evidence for this relationship has been an observed increase in AR levels at WWTPs and downstream thereof, compared with the AR of isolates upstream. The conclusion of this abovementioned study suggested that increased MAR of up to 98% was due to multiple drugs entering the system (Li et al., 2009a). Similarly, Rizzo et al. (2013) reported that WWTPs are hot spots for AR bacteria and genes spread into the environment. Once the AR bacteria enter the environment, the spread of antibiotic-resistant bacteria is equally complicated and is a multifactorial process, including waterborne transmission, airborne dispersion or via inter and intra-species contact with humans and animals, both domestic and wild (Allen et al., 2010).

This has led to the investigation of AR in non-clinical environments or in environments that are not heavily polluted (Walsh, 2013), including marine waters, rivers and soil samples. The presence of high levels of MAR at these sites suggests that AR is not localised to areas of drug contamination, but has spread globally. However, in order to gain a true understanding and to monitor the contribution of human activity on the growing AR level in the environment, some kind of a baseline level for AR needs to be established for comparison. When AR against a certain antibiotic reaches 100% globally, it will be very difficult to establish any further impact of human activity on that particular AR. Even when analysing seawater and penguin faecal samples collected at the remote Palmer Station (a United States research station in the

Antarctic), Miller et al. (2009) was able to show a direct correlation between human activity and AR.

One option to establish baseline levels is to use AR data that pre-date the discovery, the use and production of antibiotics. For example, Baker et al. (2015) described the value of a culture collection of Enterobacteriaceae from the pre-antibiotic era. Metagenomic studies of ancient DNA from 30 000-year-old Beringian permafrost sediments have identified genes encoding resistance to β-lactam, tetracycline and glycopeptide antibiotics, providing evidence of naturally occurring AR (D’Costa et al., 2011), but these ancient and old samples are not always accessible or available. Knapp et al. (2010) was able to track the trends of antibiotic resistance genes (ARGs) from soil samples in the Netherlands from 1940 to 2008 and showed a significant increase in ARGs over this period, especially regarding resistance against tetracycline. It was also suggested that basal environmental levels of ARGs are increasing.

An alternative option for baseline levels of AR in real time is to identify pristine sites not yet ruined by human encroachment or “belonging to an original state or condition”. Historically, examples of such sites are remote cave microbiomes (Bhullar et al., 2012), isolated ancient tribes (Clemente et al., 2015), Antarctic sediment samples (Perron et al., 2015) and clear mountain streams (Lima-Bittencourt et al., 2007). The rare event of humans removing samples from these “pristine sites” is very small compared with sites that are frequented by human activity. It is hypothesised by the investigators of this study that hot-spring sites are the only truly pristine environmental niches because the hot-spring water is not stagnant and has a constant flow in one direction. This means that even if intermittent contact with the water was made by human sampling activity, the point of sampling would always be new and by definition, “pristine”. AR from these sites should not increase over time due to the use of pharmaceutical antibiotics, but only due to natural evolution, and should therefore truly reflect intrinsic/natural AR.

In order to test this hypothesis, a comparison is made of other AR “pristine” sites and intrinsic AR with the results of this study. Natural, not acquired, resistance to β-lactam antibiotics is well documented, but documentation on resistance to other antibiotics in pristine environments including hot springs is relatively scant (see Section 6.3.3.1). In this study, AR levels between 20 and 55% were reported against NAc, CEF and CAR (β-lactam). Resistance against CAR, a β-lactam antibiotic, is not surprising, and almost expected, as β-lactamases are involved in common bacterial functions of cell wall biosynthesis, signalling molecules, detoxification or metabolites and other functions (Martinez, 2012). In the study of the microbiome of an isolated cave (Bhullar et al., 2012), resistance was observed against cephalosporins, cefotaxime and

cephalexin, as observed in this study as a resistance to CEF. However, in other studies of pristine environments of mountain streams (Lima-Bittencourt et al., 2007), Antarctic marine waters (De Souza et al., 2006) and ancient arctic soil (Perron et al., 2015), a natural level of AR against cephalosporins was not reported.

A follow-on study of AR of hot-spring isolates should investigate the AR of isolates in adjoining recreational swimming pools which would determine the effect of human activity. In a study of Pseudomonas in swimming pools in Ohio, USA, 96% of the isolates were found to be MAR. Twenty-two antimicrobials were tested and resistance against nine was reported (Lutz & Lee, 2011). In another survey, 462 isolates from swimming pools in Greece were studied and an MAR level of 74% was reported (Papadopoulou et al., 2008). Although both of these figures are far higher than the 2.5% MAR reported in this study, only tentative comparisons can be made, owing to variations due to seasons, human recreational activity load, and localised maintenance of such water bodies such as chlorination and other water treatments.

Not all hot springs feed into recreational swimming pools but directly into the surface water system of the area. A comparison is made of the AR of river isolates and hot-spring isolates in SA. Most studies on AR have concentrated on the Enterobacteriaceae and other enteric bacteria, including studies conducted in SA. Kinge et al. (2010) in North West Province, SA found that of the 230 isolates tested, 70-100% were MAR. These isolates originated from WW and dams in the Province. In KwaZulu-Natal Province, Lin et al. (2004) reported a similar MAR (72%) in 113 enteric bacteria, with 95% of bacteria being resistant to at least one antibiotic. In another study conducted by Samie et al. (2012) it was shown that even in drinking water in rural Limpopo Province, SA, 78% of the isolates (not only enteric bacteria) were MAR. The latter study was conducted in the same province as the sampling sites in this study. The MAR percentages in the abovementioned examples were all higher than the single isolate reported in this study. It suggests that hot-spring AR levels are lower than those in surface waters and other water sources that are contaminated by human activities. However, a bigger sample size study is required. The MAR index is useful for tracking sources of microbial contamination (Chitanand et al., 2010) where >0.2 MAR index implies the presence of antibiotic contamination in the environment and <0.2 the lack of external input of antibiotics (Krumperman, 1983; Matyar et al., 2014; Kimiran-Erdem et al., 2015). Monitoring of baseline information on intrinsic AR is as important as monitoring non-clinical, groundwater and surface water bodies, and other aquatic environments associated with human activities such as WWTPs.

100

6.3.4 Antibiotic resistance (AR) of hot-spring bacteria

6.3.4.1 Bacillus species

Most of the isolates (92.5%) cultured in this study were Gram-positive endospore-forming rods of the phylum Firmicutes, in the two main families, namely Bacillaceae (Bacillus and Anoxybacillus) and Paenibacillaceae (Aneurinibacillus and Brevibacillus). The Firmicutes are ubiquitous, numerous and common in the environment and are generally not considered to be pathogenic. As a result, less focus has been placed on the AR of Bacillus spp. as a group, unlike the pathogenic Enterobacteriaceae, and AR studies have related to pathogenic Bacillus spp. or part of bigger microbial communities. Andrews & Wise (2002) tested 13 isolates of Bacillus against TET, doxycycline, penicillin and ciprofloxacin, reporting resistance to penicillin and TET. Coonrod et al. (1971) tested 49 Bacillus isolates for AR representing six species. All were sensitive to TET, KAN, GEN and CHL. Susceptibility to penicillin and cephalothin was species related, being high for B. subtilis, and intermediate for B. pumilus and B. licheniformis. Results of this study are similar to the findings reported by Coonrod et al. (1971). Forty-six percent of 111 Bacillus isolates from marine sediment in Canada were β-lactam resistant (Belliveau et al., 1991) comparable to 50% resistance found against CAR in this study.

Bacillus AR has been described in two studies associated with aquaculture where antibiotics are added in the water to control bacterial growth. Balakrishnan et al. (2003) isolated 30 Bacillus strains from shrimp culture ponds. Resistance was observed for antibiotics that inhibited cell- wall synthesis, e.g. penicillin and ampicillin, while susceptibility was reported for inhibitors of protein synthesis (GEN, TET, KAN, CHL) and inhibitors of nucleic acid synthesis (NAc). In another study of six fish farms in China, Gao et al. (2012) reported that 63% of isolates were resistant to sulphonamides and 57% to TET, but TET resistance was not found in this study. Phylogenetic analysis comparing indigenous Bacillus to intestinal bacteria implied that gene transfer of the sul1 gene (sulphonamide resistance gene) occurred between these two groups, suggesting that increased human activity and use of antibiotics deliberately introduced into the water can result in changes in AR profiles of indigenous Bacillus spp.

The AR of pathogenic Bacillus anthracis and Bacillus cereus (Luna et al., 2007) is always a concern. However, Bacillus species that are not generally classified as pathogens, e.g. B. subtilis and B. licheniformis can cause opportunistic infections in the young, old and immunocompromised individuals (Sliman et al., 1967; Sietske & Diderichsen, 1991; Apetroaie- Constantin et al., 2009; Logan, 2011). Bacillus subtilis and B. licheniformis were the predominant bacterial species identified in this study from hot-spring water. The water feeds

101

into the surface water of rural areas where it is used for drinking and domestic purposes, and therefore enters the food chain or is in direct contact with individuals. This is further motivation for the study of AR of these particular environmental isolates, as their AR profiles have implications for water safety and public health.

In addition, there is the possibility of the transfer of ARGs between these non-pathogenic Bacillus to pathogenic bacteria since transfer of ARGs in natural environments can occur between phylogenetically distant bacterial genera, including between Gram-positive and Gram- negative bacteria (Courvalin, 1994).

6.3.4.2 Opportunistic pathogens

Two isolates were identified as Gram-positive Kocuria sp. (57T) and Gram-negative Hafnia alvei (79M). Both are potential opportunistic pathogens that have been reported to cause septicaemia, gastroenteritis and respiratory infections in compromised hosts with severe underlying disease (Klapholz et al., 1994; Purty et al., 2013). The AR profile of Kocuria sp. includes resistance to KAN, variable susceptibility to β-lactam antibiotics, quinolone, COT, and sensitivity to STP (Savini et al., 2010). Assuming that resistance to NAc (quinolone) and CEF from this hot-spring study represents intrinsic AR, it suggests that Kocuria species have the potential to become resistant to other antibiotics. Antibiotic resistance to β-lactam antibiotics in H. alvei has been reported (Klapholz et al., 1994). In this study, a single isolate was also resistant to the β-lactam antibiotics CAR, and CEF. Although conclusions cannot be drawn from the AR profiles of single isolates of Kocuria and Hafnia, these results provide preliminary justification that further investigation of intrinsic AR profiles of opportunistic pathogens is required prior to their exposure to environmental selection pressures.

6.3.5 Heavy-metal tolerance patterns in isolates from hot springs

Twenty-nine hot-spring isolates were tested against eight heavy-metal salts: aluminium sulphate (Al), chromium IV oxide (Cr), copper sulphate (Cu), iron sulphate (Fe), manganese sulphate (Mn), nickel chloride (Ni) and lead nitrate (Pb) at two concentrations of 40 and 10 mM. Mercuric chloride (Hg) was tested at only at a very low concentration of 200 nM. Heavy-metal tolerance describes the lack of susceptibility to the toxic effect of heavy metals by interference with enzyme functionality or with the DNA or RNA of cells; however, the definition has been imprecise and inconsistently used. This is because the assay used has not been standardised. It has been performed on solid (Icgen & Yilmaz, 2014; Onuoha et al., 2016) and liquid media

102

(Kimiran-Erdem et al., 2015), incubation times of 24 h (Kimiran-Erdem et al., 2015; Onuoha et al., 2016) or 48 h (Belliveau et al., 1991; Mustapha & Halimoon, 2015), defined as a comparison to a control strain (Matyar et al., 2014; Kimiran-Erdem et al., 2015), or as a cut-off value of resistance with growth in >2.5 mM (Belliveau et al., 1991) or no heavy metal present (Icgen & Yilmaz, 2014). The HMT assay is described as minimum inhibition concentrations values with different maximum concentrations used varying from 5 mM (Onuoha et al., 2016), 40 mM (Belliveau et al., 1991) or as high as 5 000 µg/mL (Kimiran-Erdem et al., 2015).

120

100

40 metal toleranceisolates -

20 % Heavy % 0

Al 40 Al Ni 10 Ni 40 Ni Cr 10 Cr 40 Cr Fe 10 Fe 40 Fe Al 10 Al Pb 10 Pb 40 Pb Cu 10 Cu 40 Cu Mn 10 Mn 40 Mn Hg 200 Hg Heavy-metal concentration (mM) except for Hg (nM)

Figure 6.3: Heavy-metal tolerance of 29 hot-spring isolates against eight heavy-metal salts: aluminium (Al), chromium (Cr), copper (Cu), iron (Fe), mercury (Hg), manganese (Mn), nickel (Ni) and lead (Pb) in mM (except for Hg which is in nM)

Due to limited resources only two concentrations were tested in this investigation (10 mM and 40 mM, except for Hg which was tested at 200 nM) based on concentrations used by Belliveau et al. (1991). Tolerance was defined for this study as growth at that concentration while sensitivity was defined as no growth. Initially, metal concentrations of 10 mM resulted in resistance varying from 86% (Fe) to 100% (Cr) as indicated in Figure 6.3, and this concentration did not give a discrepancy that was large enough in order to make a possible association with AR. For this reason, a higher concentration of 40 mM was used. At these concentrations, the highest resistance of >80% was recorded against Pb, Cu and Cr. Only 20 to 34% of the isolates were resistant to Fe and Al, while intermediate levels between 48 and 73% were observed for resistance against Hg, Mn and Ni (Figure 6.3). Mercury (Hg), being the most toxic of the heavy metals, was used at the lowest concentration of 200 nM, and half of the

103

isolates were inhibited at this concentration. In literature, Bacillus spp. range in HMT within concentrations of 50µg/mL to 300µg/mL (Singh et al., 2013b). The use of concentrations of HMT in this study was relatively high and above this range, however experimental conditions are not standardised for this assay resulting in errors in comparisons.

Figure 6.4: Hot-spring isolates at different multiple heavy-metal resistance (MHMT) index values

All isolates were resistant to at least two metals. The MHMT index was calculated with 40mM concentrations of metal ions (except for Hg which was tested at 200nM) as described above for the MAR index (Kimiran-Erdem et al., 2015) as depicted in Figure 6.4. The MHMT index was calculated as the number of heavy metals to which the isolate was resistant/total number of heavy metals against which the isolate was tested. The median MHMT index value for isolates was 0.625 with the majority of the isolates (45%) being resistant to five heavy metals. A single isolate (54T) was resistant to all eight heavy metals. Even though the MHMT index was high for the majority of the isolates, the measured concentrations of the heavy metals in the hot- spring water did not exceed the WHO, EU or SABS guideline values (Table 3.1). Environmental selection pressure of HMR would therefore be weak.

6.3.6 Correlation between heavy-metal resistance (HMT) and antibiotic resistance (AR)

A correlation between AR and HMR has been well documented and is a growing concern globally including in waterborne isolates (Oves & Hussain, 2016). There are several reasons why AR and HMT are associated, including co-selection where different resistance determinants are present on the same genetic element such as plasmids, and cross-resistance where the same genetic determinant is responsible for resistance to both antibiotics and metals

104

(Baker-Austin et al., 2006). Since genes for AR and HMT can occur on the same plasmid, they can also be transferred together during conjugation. With the increasing levels of water pollution, particularly heavy metals from mining industries, manufacturing and household activities, metal-driven co-selection with AR is the resulting environmental consequence. The positive correlation between the abundance of ARGs and the elevated concentrations of antibiotics and heavy metals in environments has been reported in several studies. Knapp et al. (2010) provided evidence that the presence of heavy metals in soil enhanced the abundance of ARGs. Antibiotic resistance affects the health of humans, animals and plants and therefore this problem should be addressed with great urgency.

Nalidixic acid (NAc) is a quinolone and this resistance can be useful to distinguish between natural or acquired resistance. This is because intrinsic quinolone resistance (qnr) genes that are chromosomally encoded have been detected in several waterborne bacteria (Sanchez et al., 2008), but when quinolone resistance is acquired, it is found integrated into plasmids (Martinez, 2009a; 2012). These differences could be useful to differentiate natural AR from acquired AR, and therefore also an indication of whether a location is truly pristine or not. A Bacillus isolate from municipal waste in India grew in the presence of Ni and Cr, and was resistant to KAN, ampicillin and methicillin. Curing of the plasmid resulted in a loss of both HMT and MAR (Samanta et al., 2012a). Similarly, 24 of 201 waterborne coliform isolates from rural Indian villages were able to transfer resistance against ampicillin, Ni, Cu and Cr to a susceptible laboratory strain E. coli K12, and this could also be associated with curing of plasmids (Tewari et al., 2013). Although some of the abovementioned environments are not heavily burdened by human activity, the correlation between AR and HMT has been established. If intrinsic AR had not yet made the evolutionary leap onto plasmids, an association between AR and HMT in more remote and pristine locations would not be possible. Eleven-hundred Bacillus strains were isolated from marine sediment in Nova Scotia, Canada and tested against 11 antibiotics, Hg, Cu and Zn, and screened for plasmids. No correlation between plasmid content and resistance to either antibiotics or Hg was observed (Belliveau et al., 1991). Munster et al. (1981) studied the plasmids of 48 Thermus spp. isolated from hot springs in Yellowstone National Park, USA and found no correlation between AR and HMT which supported the hypothesis that the cleaner the water environment, the lower the association of MHMT and MAR. El-Gayar et al. (2017) studied 2 isolates from hot springs of Saudi Arabia for AR and HMT however their results were inconclusive. Further investigation is recommended for additional data on AR and HMT of hot- spring isolates.

105

In this study, AR and HMT, based on the results obtained at metal concentrations of 40 mM (Appendix 14), were used for statistical analysis to determine whether AR was associated with HMT. The AR and HMT profiles of each of the isolates were determined separately and compared. Statistically, no association was found between AR and HMT (p>0.05), suggesting that there is no cross resistance. To definitively determine whether this is indeed intrinsic and not acquired resistance or plasmid-associated, further genetic confirmation is required. A significant correlation where p<0.05 using Fisher’s exact test was found between NAc resistance and susceptibility to Cu (p = 0.031) and Mn (p = 0.048); currently the reason for this finding is not known.

106

CHAPTER SEVEN: SCREENING OF POTENTIAL BIOREMEDIATION ENZYMES FROM HOT-SPRING BACTERIA USING CONVENTIONAL PLATE ASSAYS AND LIQUID CHROMATOGRAPHY-TANDEM MASS SPECTROPHOTOMETRY (LC-MS/MS)

7.1 INTRODUCTION

Bioremediation of WW has made use of microbially produced enzymes with advantages of lower costs, green footprint, decreased toxicity to the environment, ease of application, specificity and flexible applications to a variety of possible substrates (Karigar & Rao, 2011; Facchin et al., 2013). Microorganisms from extreme environments like hot springs are investigated for their enzymes that remain active at harsh environmental conditions of pH and temperature (Demirjian et al., 2001; Demorne et al., 2017). Species in the genus Bacillus and related bacteria predominate when the bacteria are cultured from hot-spring sites (Derekova et al., 2008; Panosyan & Birkeland, 2014; Panda et al., 2016). This group of bacteria are well known for their industrial applications and production of useful enzymes (Kumar et al., 2013). Bacillus species that produce amylase (Zhang et al., 2015; Acer et al., 2015), protease (Panda et al., 2013; Bekler et al., 2015) and phosphatase (Sen & Maiti, 2014) have been previously isolated from hot-spring sites. Even though this is well studied, novel enzymes that are active at 110 °C have been recently described in Geobacillus spp. isolated in Tunisian hot springs (Thebti et al., 2016).

More modern techniques or protein or enzyme detection including LC-MS/MS have been useful in expediting the identification and discovery of microbial enzymes (Fandi et al., 2012) although the conventional plate assays have their place in low-resource laboratories. An indirect application to WW bioremediation has been the use of microbial enzymes for biomonitoring (Logar & Vodovnik, 2007) in aquatic environments (Li et al., 2010). For example, lead (Pb) is highly toxic in the environment because of its ability to mimic biologically important metals and produce membrane damage. Delta-aminolevulinate dehydratase (ALAD) is a conserved metalloprotein in many organisms including bacteria, and is very sensitive to Pb (Konuk et al., 2010). Monitoring of this enzyme in Pseudomonas spp. (Korcan et al., 2007) has been used to detect Pb concentrations in contaminated water environments.

108

The aim of this study was to screen isolated and identified bacteria from Limpopo hot springs, SA for oxidoreductase and hydrolytic enzymes that could be potentially useful in wastewater bioremediation by conventional biochemical and gravimetric assays, and LC-MS/MS including oxidoreductase and hydrolytic enzymes, and enzymes potentially useful for biomonitoring.

7.2 METHODOLOGY

Figure 7.1: Summary of methods used in study

The location of five hot springs in the Limpopo Province, SA that were sampled is described in Table 1 in Section 3.1. Water and sediment samples were collected and processed in the laboratory. Using five different media and two different temperatures for incubation, bacteria were isolated from water and sediment samples as explained in Section 3.2. Bacterial isolates were identified based on the 16S rDNA sequencing using universal prokaryotic primers (Section 3.4).

Plate assays for the detection of amylase, protease, pectinase, gelatinase, lipase, azoreductase and laccase are described (Section 3.12.1). Three thermophilic isolates that produced differential

109

clearing on the amylase plate assay were used to quantify starch reduction in 1% starch liquid media using a colorimetric assay (Section 3.12.2). The presence of cellulase in CFCS was determined gravimetrically with a reduction in weight (mg) of cellulose (Section 3.12.3). The spectrophotometric tube assay for the detection of laccase and peroxidase is described in Section 3.12.4.

The detailed method for LC-MS/MS is described in Section 3.19 for the identification of proteins in the extracellular extract of isolates 19S, 76S, 77S and 85Li.

7.3 RESULTS

7.3.1 Identification of bacteria

Of the 56 isolates tested, only 44 were identified by 16S rDNA sequencing since no PCR product was obtained for 12 of the isolates. The majority were identified from the phylum Firmicutes which includes the predominant genus Bacillus. Two isolates were Kocuria sp. and Arthrobacter sp. of the phylum Actinobacteria. Seven isolates were Anoxybacillus spp. and five belonged to the phylum Proteobacteria. The isolate numbers, location of sample site, temperature of initial isolation and source (water or sediment) are listed in Table 7.1.

7.3.2 Plate assays

Using plate assays, 56% (n = 43), 68% (n = 38) and 16% (n = 31) were positive for amylase (Figure 7.2A), protease (Figure 7.2C) and bromothymol blue decolourisation (Figure 7.2D), respectively. Bromophenol blue (BB) decolourisation is indicated by a clearing around the colony. The results for the amylase and protease degrading isolates are given in Table 7.1 as colony size and clearing size in millimetres. Several authors (Bakri et al., 2012; Demirkan et al., 2017) report these results as relative or ratio, i.e. clearing/colony size; however, this is inaccurate when the bacteria are motile as indicated in Figure 7.2B.

Negative results were obtained for lipase using olive oil and Tween 80 at 1.5% and 2%, gelatinase, pectinase and laccase (with guaiacol at 0.5 mM and 2 mM as substrate) by plate assay. No growth was observed on methyl red plates for the detection of azoreductase.

110

A B

C D

Figure 7.2A: Amylase-producing colony on starch agar plate; Figure 7.2B: Motile bacterial isolate on starch agar plate; Figure 7.2C: Protease-producing colony on skim milk agar plate; Figure 7.2D: Isolate 84Li on bromothymol blue agar plate agar plate

Table 7.1: Identification, isolation conditions and enzyme characterisation for hot-spring isolates, Limpopo Province, South Africa

Isolate Temp Laccas Peroxida No. Sites °C Media Identification Amylase * Protease * BB Phenol e se colony colony # ** size clearing size clearing

1T Tshipise 53 NA Bacillus sp. 20 4 25 6 - - - - Bacillus 2T Tshipise 53 NA licheniformis motile na motile na + + - - Anoxybacillus 3T Tshipise 53 NA rupiensis 8 10 2 1 - - - - 10%L Anoxybacillus 4T Tshipise 53 A rupiensis 5 6 40 6 - nd - - 10%L Bacillus 6T Tshipise 53 A licheniformis 20 4 45 2 nd nd - - Anoxybacillus 7T Tshipise 53 Actino rupiensis 6 5 10 12 nd nd - - Bacillus 8T Tshipise 53 Actino licheniformis 15 2 nd - - - -

9T Tshipise 53 PDA Proteobacteria 22 3 13 6 + - - - Bacillus 10T Tshipise 53 PDA licheniformis motile na 1 1 nd - - - Anoxybacillus 11T Tshipise 53 PDA rupiensis 6 7 nd - nd mo 12S Siloam 53 NA Bacillus subtilis motile na 40 10 tile - - - Anoxybacillus 13S Siloam 53 NA rupiensis 6 10 1 3 - nd - - 10%L 14S Siloam 53 A Bacillus subtilis 18 4 7 11 - nd 10%L 15S Siloam 53 A Bacillus sp. 36 7 nd - - - - 10%L 16S Siloam 53 A Brevibacillus sp. 40 0 motile na - nd - -

111

10%L Anoxybacillus 17S Siloam 53 B flavithermus 5 0 nd nd nd - -

18S Siloam 53 Actino Bacillus sp. 20 6 nd nd nd Anoxybacillus 19S Siloam 53 Actino sp. 5 5 13 10 + + - - Bacillus 20S Siloam 53 PDA licheniformis 15 5 65 2 - nd - - Mpheph 22M u 53 NA Bacillus subtilis 23 4 motile na nd - - - Mpheph 23M u 53 NA nd 56 3 nd - - - - Mpheph 24M u 53 NA Bacillus pumilus 4 0 nd - nd Mpheph 25M u 53 NA nd 3 4 nd nd nd Mpheph Gulbenkiania 27M u 53 Actino mobilis 3 0 nd nd nd Mpheph Bacillus 28M u 53 Actino licheniformis 22 2 40 3 nd nd Mpheph 29M u 53 PDA nd 37 1 nd - nd

33Li Libertas 53 NA Bacillus subtilis 31 3 5 11 - nd - - 10%L 34Li Libertas 53 A nd 68 0 45 6 nd nd

35Li Libertas 53 Actino nd 9 1 nd nd - - - Brevibacillus ne 36Li Libertas 53 Actino agri 50 1 motile na g nd

38Li Libertas 53 PDA nd 37 1 nd nd nd Bacillus 39T Tshipise 37 Na licheniformis nd nd nd - - - Lekkerru 40Le s 37 Actino Bacillus subtilis nd nd - - - - 53 45T Tshipise Sed NA nd nd 6 2 + nd 53 46S Siloam Sed NA nd nd 65 0 nd nd 53 47Li Libertas Sed NA Bacillus subtilis nd motile na nd - - - 53 48Li Libertas Sed NA Bacillus subtilis motile na 50 8 nd - - - 53 49Li Libertas Sed NA nd nd 51 0 nd nd 53 50T Tshipise Sed PDA nd 4 0 5 10 nd - - - 53 51T Tshipise Sed PDA nd nd 8 11 nd nd Mpheph 53 52M u Sed PDA Bacillus sp. motile na 10 10 - - - - Mpheph 53 53M u Sed PDA unknown nd 4 1 nd - - -

54T Tshipise 37 PDA Bacillus subtilis nd motile na nd - - -

56S Siloam 37 NA nd nd 3 0 nd nd

57T Tshipise 37 NA Kocuria sp. 10 1 6 0 - - - - Arthrobacter 58T Tshipise 37 Actino luteolus 10 0 4 0 - nd - -

71T Tshipise 37 NA Bacillus sp. motile na 8 10 nd + - -

112

72T Tshipise 37 NA Proteobacteria motile na nd nd - - - Bacillus 76S Siloam 37 NA mojavensis nd nd nd - - - Bacillus methylotrophic 77S Siloam 37 NA us motile na 15 12 nd + - -

78S Siloam 37 Actino Bacillus subtilis nd 20 10 nd - - - Mpheph 79M u 37 NA Hafnia alvei nd nd - - - - Lekkerru 80Le s 37 NA Proteobacteria 3 0 nd nd nd - -

83Li Libertas 37 NA Bacillus subtilis motile na 12 10 nd + - - 10%L 84Li Libertas 37 A Bacillus sp. motile na 12 9 + + - - 10%L 85Li Libertas 37 A Brevibacillus sp. motile na motile na nd - - - # Hot-spring water temperature

“Sed” refers to isolates obtained from sediment

## Phenol-degrading bacteria were positive if they differed by 2 standard deviations from the control

* Motile bacteria grew over the entire surface of the plate and no results could be observed

**Media used for isolation where NA is nutrient agar, PDA is potato dextrose agar, 10%LA is minimal Luria agar and Actino is actinomycete isolation agar

7.3.3 Quantitative amylase assays

In order to quantify the efficiency of starch degradation by three thermophilic isolates that showed variable amylase production on the starch agar plate, it was necessary to draw a standard curve for soluble starch to ensure that the assay could be used qualitatively (Appendix 9) and OD at 660 nm would reflect starch concentration. Uninoculated broth was the control representing 100% starch content. Post-incubation, the starch content of the control and three inoculated flasks was tested for soluble starch. A photograph of the assay in Eppendorf tubes, corresponding with the bars in the graph, is inserted directly above the graph (Figure 7.3). The amount of starch left after inoculation with isolate 9T was 2% indicating that starch was reduced by 98% as indicated by the lack of colouration in the corresponding tube in Figure 7.3. Similarly 28% and 66% starch was left after inoculation with isolate 13S and isolate 20S, respectively, suggesting a reduction by 72% and 34% as shown in Figure 7.3.

113

Figure 7.3: Reduction of soluble starch in starch broth by inoculation of thermophilic isolates 9T, 13S and 20S 7.3.4 Gravimetric assay for the detection of cellulase

No positive results obtained for cellulase (n = 27).

7.3.5 Biochemical tube assay for the detection of laccase and peroxidase

7.4A

7.4B

Figure 7.4A: Peroxidase assay positive control using turnip extract; Figure 7.4B: Peroxidase assay for isolates 16S and 84Li

114

All the crude bacterial extracts were negative for laccase and peroxidase activities. Bacillus spp. have been previously reported to produce laccase and peroxidase (Archna et al., 2015; Min et al., 2015; Ezhilarasu, 2016).

7.3.6 Identification of proteins by tandem LC-MS

Figure 7.5: Coomassie blue stained sodium dodecyl sulphate polyacrylamide gel electrophoresis of crude extracts of bacterial supernatants

Table 7.2: LC-MS/MS identification of protein in crude bacterial supernatant of isolates 76S, 77S, 85Li and 19S

ISOLATE ENZYME FAMILY ENZYME GENE UNIQUE PEPTIDE ID % COVERAGE

19S Oxidoreductases NADH dehydrogenase ahpF (ndh) P42974 20.24

19S, 85Li Oxidoreductases Alanine dehydrogenase ald (spoVN) Q08352; Q08352 43.12; 8.73 FMN-dependent NADH- 19S Oxidoreductases azoreductase azoR Q9X4K2 15.64 2-oxoisovalerate bfmBAA 19S Oxidoreductases dehydrogenase (bfmB1a) P37941 40.06 bfmBB Alpha-keto acid (bfmB, 19S Oxidoreductases dehydrogenase complex bfmB2) P37942 9.91

19S Oxidoreductases Enolase eno Q65EN2 41.16 3-hydroxyacyl-[acyl-carrier- 19S Oxidoreductases protein] dehydratase fabZ Q9K6J4 17.14 Glyceraldehyde-3- 19S Oxidoreductases phosphate dehydrogenase gap P00362 21.19 Probable glycine 19S Oxidoreductases dehydrogenase gcvPB (yqhK) P54377 18.85 6-phosphogluconate 19S Oxidoreductases dehydrogenase gndA (yqjI) P80859 38.81 Glycerol-3-phosphate 19S Oxidoreductases dehydrogenase [NAD(P)+] gpsA (glyC) P46919 5.8

115

Cryptic catabolic NAD- specific glutamate 19S Oxidoreductases dehydrogenase gudB (ypcA) P50735 34.19 Inosine-5'-monophosphate 19S Oxidoreductases dehydrogenase guaB P21879 45.49

19S Oxidoreductases Imidazolonepropionase hutI P42084 4.513

19S Oxidoreductases Urocanate hydratase hutU Q9KBE5 6.08 Isocitrate dehydrogenase 19S Oxidoreductases [NADP] icd (citC) P39126 45.63 katG (cat, 19S Oxidoreductases Catalase-peroxidase perA) P14412 20.27

19S Oxidoreductases Leucine dehydrogenase ldh Q53560 25.82

19S, 85Li Oxidoreductases L-lactate dehydrogenase ldh (lctE) P00344; P13714 7.57; 27.19 19S, 76S, Q9X4K8; P49814; 27.24; 8.65; 85Li Oxidoreductases Malate dehydrogenase mdh (citH) P49814 8.65 Probable malate: quinone 19S Oxidoreductases oxidoreductase mqo Q9Z9Q7 2.6

19S Oxidoreductases NADPH dehydrogenase namA Q9KCT8 11.83 2-oxoglutarate 19T Oxidoreductases dehydrogenase complex odhB (citM) P16263 15.83 19S, 76S, Dihydrolipoyl pdhD (aceD, P11959; P21880; 40.85; 11.06; 77S Oxidoreductases dehydrogenase citL) P21880 19.57 Pyruvate dehydrogenase E1 component subunit 19S, 85Li Oxidoreductases beta pdhB (aceB) P21874, P21882 64; 10.15 Pyruvate dehydrogenase E1 component subunit 19S Oxidoreductases alpha pdhA P21873 30.89 Pyruvate dehydrogenase 19S Oxidoreductases complex pdhC (aceC) P21883 28.51 NADPH dependant 7 cyano 19S Oxidoreductases 7 deazaguanine reductase queF Q65KI3 19.39 1-pyrroline-5-carboxylate 19S, 85Li Oxidoreductases dehydrogenase rocA Q63GS0; Q65NN2 41.94; 4.26 Succinate dehydrogenase 19S Oxidoreductases flavoprotein subunit sdhA (citF) P08065 13.31

19S Oxidoreductases Superoxide dismutase [Mn] sodA P00449 89.22 Putative aldehyde 85Li Oxidoreductases dehydrogenase yfmT O06478 7.42 Probable NADH-dependent 19S Oxidoreductases butanol dehydrogenase 1 yugJ O05239 6.98 Glucose-6-phosphate 1- 19S Oxidoreductases dehydrogenase zwf (yqjJ) P54547 9.41 aprE (apr, 76S Hydrolytic: protease Subtilisin E aprA, sprE) P04189 37.01

77S, 85Li Hydrolytic: protease Subtilisin BPN' apr P00782; P00782 44.02; 44.02 19S, 77S, P35835; P35835; 4.72; 42.78; 85Li Hydrolytic: protease Subtilisin NAT aprN P35835 26.25

76S, 77S Hydrolytic: protease Bacillolysin nprE P68736; P06832 7.68; 22.26 Methionine 77S Hydrolytic: protease aminopeptidase 1 map P19994 7.66

19S Hydrolytic: protease Uncharacterized protein yqjE P54542 35.04 ATP-dependent Clp protease proteolytic 19S Hydrolytic: protease subunit clpP Q9K8F4 28.3

116

19S Hydrolytic: protease Putative aminopeptidase ysdC P94521 14.4

19S Hydrolytic: protease Peptidase T pepT Q65D74 12.92999983 76S, 77S, Q40507, Q6YK37, 8.77; 27.66; 85Li Hydrolytic: cellulase Glucuronoxylanase xynC (ynfF) Q6YK37 4.49

76S Hydrolytic: cellulase Beta-glucanase bglS (bgl, licS) P07980 18.41

85Li Hydrolytic: amylase Alpha-amylase amyE (amyA) P00691 7.28 Trehalose-6-phosphate 19S Hydrolytic: amylase hydrolase treA (treC) P39795 6.42

19S Hydrolase Oligo-1,6-glucosidase malL P29094 3.56 Hydrolytic: 19S phosphotase Inorganic pyrophosphatase ppa P19514 61.82 Hydrolytic: beta 19S lactamase Ribonuclease J1 rnjA (ykqC) Q45493 34.23 delta aminolevulinic acid 19S Lyase: dehydratase hemB Q9K8G2 14.94

The listed proteins in Table 7.2 have no more than 1% false discovery rate (FDR), where FDR indicates the reliability of identifying differentially expressed proteins. The UniprotKB protein accession number is the unique identifier assigned to the protein, and the percent coverage is calculated by dividing the number of amino acids in all found peptides by the total number of amino acids in the entire protein sequence.

7.4 DISCUSSION

Bacteria isolates from hot springs in Limpopo Province, SA prdocued a variety of enzymes that have potential use in bioremediation, and further characterisation might reveal unique properties of functionality under extreme conditions such as temperature (Baltaci et al., 2017; Demorne et al., 2017). Comparable to other studies of hot springs (Derekova et al., 2008; Panosyan & Birkeland, 2014; Panda et al., 2016), the majority of the isolates were identified as Bacillus and Bacillus-related species. Bacillus spp and Anoxybacillus spp have been reported to have many applications in biotechnology (Kumar et al., 2013; Goh et al., 2013). The two main categories of enzymes useful for bioremediation are hydrolytic enzymes including lipase, cellulase, amylase and protease, and oxidoreductases which include laccase and peroxidase (Karigar & Rao, 2011; Facchin et al., 2013). Other relevant hydrolases are phosphatase and beta-lactamase. These enzymes were identified by this investigation either by conventional plate assays or LC- MS/MS, and discussed further in separate categories. Several investigations have also sought useful industrial enzymes from hot spring environments (Thebti et al., 2016; Zahoor et al., 2016).

117

7.4.1 Amylase

The production of amylase and protease by Bacillus spp. is well known (Li et al., 2013), and amylase-producing Bacillus isolates from hot springs have been well documented. In this study, 56% of the isolates produced amylase by plate assay. Three thermophilic isolates (9T, 13S and 20S) were selected for a quantitative assay in liquid media as they produced a range of clearing on starch agar plates from 3 mm to 10 mm. Isolate 9T proved to be very efficient in reducing a solution of 1% starch by 98% at 53 °C. The poorest degrader was isolate 20S that reduced the starch by 34% under the same conditions. However, on the plate assay, isolate 9T gave the smallest zone clearing of 3 mm compared with isolates 13S (10 mm) and 20S (5 mm) as indicated in Table 7.1. This suggests that the plate assay for amylase is useful for qualitative screening, but may not reflect the quantitative potential that is revealed in liquid culture assay. Isolate 19S was positive for amylase and protease. Isolate 77S produced a protease. No results could be obtained for the amylase assay of isolate 77S and 85Li, and the protease assay for isolate 85Li due to the bacteria being motile and covering the entire plate.

By LC-MS/MS Anoxybacillus rupiensis 19S, was found to produce trehalose-6-phosphate hydrolase and oligo 1,6 glucosidase. Trehalose-6-phosphate hydrolase (EC 3.2.1.93) is related to alpha-amylase and also belongs to the glycoside hydrolase family GH13 (Reddy et al., 2003; Chuang et al., 2012). This enzyme hydrolyses trehalose, trehalose-6-phosphate and p- nitrophenyl-alpha-D-glucopyranoside, to glucose and glucose 6-phosphate, but not lactose, maltose, sucrose or sucrose-6-phosphate. This enzyme from B. licheniformis was cloned and expressed in E. coli (Chuang et al., 2012), and the associated genes have been reported in Anoxybacillus sp. (Belduz et al., 2015; Poli et al., 2015). However, the natural expression of the protein as found in this study allows further characterisation and functionality studies such as determination of molecular weight, enzyme kinetics of Km and Vmax for different substrates (Rimmele & Boos, 1994). Similarly, oligo-1,6-glucosidase (EC 3.2.1.10) is a subfamily of alpha-amylase, and catalysis of the hydrolysis of (1->6)-alpha-D-glycosidic linkages in some oligosaccharides produced from starch by alpha-amylase. It hydrolyses various disaccharides such as sucrose, maltose, and iso-maltose, and also longer maltodextrins. The enzyme has been reported in Bacillus stearothermophilus (Kobayashi et al., 2006) and Anoxybacillus strains from hot springs (Poli et al., 2015; Lim et al., 2015b). Both trehalose-6-phosphate hydrolase and oligo-1,6 glucosidase have biotechnological applications including the reduction of biomass (Li et al., 2009b).

118

Even though amylase production of Brevibacillus sp 85Li could not be determined by the plate assay because the bacteria was motile, by LC-MS/MS, it was found to produce alpha-amylase (1,4-alpha-D-glucan glucanohydrolase EC3.2.1.1). These amylase-producing bacteria are important in the bioremediation of waste from the food industry, and the paper and pulp industry, however knowledge of the extent of enzyme production is key to applications.

7.4.2 Proteases/peptidases

The second major group of proteins produced by Bacillus sp. is proteases and it is common component in detergents. Proteases are important in bioremediation hydrolysing proteins as waste from industries like poultry, fishery, meat, dairy, leather as well as household wastes (Karigar & Rao, 2011). Proteases are commonly found in detergents (Souza et al., 2014). Sixty eight percentof the isolates in this study were able to degrade skim milk. The LC-MS/MS technique was able to discern aminopeptidase, peptidase T and several types of subtilisins. Subtilisin is a serine protease (EC 3.4.21.62) that is non-specific, initiating the nucleophilic attack on the peptide (amide) bond through a serine residue at the active site (Siezen & Leunissen, 1997). It was produced by isolates Bacillus sp. 77S, Brevibacillus sp. 85Li and Anoxybacillus sp. 19S. Bacillus sp. 77S appeared to produce two types of subtilisins (subtilisin NAT and BPN) and the production of multiple subtilisins has been reported in Bacillus sp. previously (Takamura et al., 2007). The occurrence of three types of subtilisins (sub E, BPN and NAT), and the production of subtilisin NAT from three different species of hot-spring isolates could motivate for comparative studies in functionality between different subtilisins and between different species. The action of peptidase T (EC 3.4.11.4) is to cleave the N-terminal amino acid of tripeptides (Cha et al., 2000).

7.4.3 Cellulases

Cellulases are also important in bioremediation of animal feed, wine production, paper and textile industry, and the removal of lignin. Cellulase has been isolated from thermophilic Bacillus (Elkhalil et al., 2015; Rozanov et al., 2015) and Anoxybacillus (Genc et al., 2015). Glucuronoxylanase (EC 3.2.1.136) cleaves the beta-1,4-xylosidic bond in some glucuronoarabinoxylans. Glucuronoxylans are primary components of hemicelluloses of plant cell-wall polysaccharides, and their presence in the bacterial supernatant of isolates 77S, 85Li and 19S suggests a potential application for cellulose degradation.

119

Beta-glucanase or O-glycosyl hydrolases (EC 3.2.1.4) are part of the complex family of cellulases (Sadhu & Maiti, 2013), and non-specifically hydrolyse the glycosidic bond between two or more carbohydrates. Novel beta-glucanases have been isolated from Bacillus spp. (Mawadza & Zvauya, 1996).

7.4.4 Phosphatases

Not many studies of bacteria from hot springs have screened for phosphatase, although it has been reported by Sen and Maiti (2014) in a study in India where half of the Bacillus isolates were positive suggesting that it may be common. In this investigation by LC-MS/MS, Anoxybacillus rupiensis 19S produced inorganic pyrophosphatase (EC 3.6.1.1). This enzyme is involved in the conversion of one molecule of pyrophosphate to two phosphate ions. Phosphates in the environment may be regarded as pollutants especially as a result of agro-industrial activities, e.g. fertiliser. Phosphates need to be solubilised to prevent eutrophication as well as to make these nutrients available for plant uptake. Phosphatase enzymes have been implicated in this process of phosphate solubilisation (Mudryk & Podgorska, 2005), and production has been reported by several bacteria from the rhizosphere of plants (Islam et al., 2007; Kadiri et al., 2013; Karpagam & Nagalakshmi, 2014). In terms of bioremediation, these phosphatase positive microbes might sustainably manage phosphorus in poor agricultural soil (Sharma et al., 2013b), and have been reported to remove phosphate from WW (Kristnaswamy et al., 2009). Chen et al. (2006b) observed a drop in pH and production of organic acids correlated with phosphatase production. In Figure 9.5, five isolates in the phenol degradation assay showed that the nutrient broth pH changed from neutral to acidic based on the colour of phenol red, suggesting these isolates might produce phosphatase. It includes isolate 19S which was found to produce phosphatase by LC-MS/MS.

In addition, in this study, it was found that isolates 85Li and 19S produced 1-pyrroline-5- carboxylate dehydrogenase (EC 1.2.1.88). This oxidoreductase enzyme catalyses the chemical reaction between 1-pyrroline-5-carboxylate, NAD and water to produce glutamate, NADH and hydrogen ions. When this gene was cloned into E. coli from fungi Penicillin sp., an improvement of phosphate solubilisation was reported (Gong et al., 2014).

7.4.5 Ribonuclease

Ribonuclease J1 (EC 3.1) is an RNase that has endonuclease and 5’-‘3’ exonuclease activity. In Bacillus it belongs to the family of beta-lactamases (Even et al., 2005), enzymes that degrade

120

the antibiotics, beta-lactams. The occurrences of these bioactive molecules that infer antibiotic resistance are important in bioremediation of pharmaceutical waste and degradation of antibiotics in the environment (Mandal & Mandal, 2014). This removal of antibiotics has been reported by fungi and bacteria in aquatic environments (Nnenna et al., 2011). Antibiotics in aquatic environments are becoming an increasingly global problem (Martinez, 2009b). In the Msunduzi River of KwaZulu-Natal, SA, they have been found in low concentrations of <10 µg/L in surface water and up to 34.5 µg/L in WW (Matongo et al., 2015). By LC-MS/MS, ribonuclease was identified from isolate 19S. However, in a previous study this isolate was sensitive to the beta-lactam antibiotic carbenicillin and did not display resistance (Appendix 14). Further characterisation is required to determine the specificity of ribonuclease J1 of A. rupiensis 19S in this study to determine whether it is able to degrade beta-lactam molecules and exactly which antibiotics are suitable substrates.

7.4.6 Dihydrolipoyl dehydrogenase

Dihydrolipoyl dehydrogenase (EC 1.8.1.4) catalyses the oxidation of dihydrolipoamide to lipoamide, found in the extracellular extracts of isolates 76S, 77S and 19S. The gene that encodes this enzyme from Clostridium kluyveri was cloned into E. coli for characterisation and found to act on synthetic dyes such as 2,6 dichlorophenolindophenol and nitro blue tetrazolium suggesting a possible function in decolourisation of textile WW (Chakraborty et al., 2008).

7.4.7 Oxidoreductases

The oxidoreductases identified are: catalase peroxides (EC 1.11.1.21), super oxide dismutase (SOD EC 1.15.1.1), FMN-dependent NADH-azoreductase (EC 1.7.1.6) and probable malate: quinone oxidoreductase 1 (EC 1.1.5.4). The oxidoreductases will be discussed as a group as they are important in the biodegradation of recalcitrant pollutants including aromatic compounds such as textile dyes, petroleum pollutants such as PAHs and phenolic compounds in WW (Duran & Esposito, 2000; Singh & Eltis, 2015). All four enzymes were identified in the extracellular growth culture supernatant of Anoxybacillus sp. 19S by LC-MS/MS. The mechanism of bioremediation is through precipitation or transformation. Numerous publications report that Bacillus and related species are able to decolourise textile dyes (Husain, 2010; Gursahani & Gupta, 2011; Zhao et al., 2014a; Min et al., 2015)

The biotechnologically important peroxidases and applications are listed by Duran & Esposito (2000).

121

Catalase-peroxidase is a strong catalase with hydrogen peroxide as a donor, and a release of oxygen. Catalase peroxidase was reported to be involved in textile bleaching (Gudelji et al., 2001) and the removal of melanoidin in coffee waste water (Mahgoub et al., 2014).

Probable malate: quinone oxidoreductase 1 (EC 1.1.5.4) synthesises oxaloacetate from (S)- malate via quinone in the tricarboxylic acid cycle. The gene responsible for producing quinone oxidoreductase in Bacillus sp. was cloned into E. coli and discovered to decolour the lignin- model dye, Azure B (Bandounas et al., 2013).

Superoxide dismutase (SOD, EC 1.15.1.1) is an enzyme that alternately catalyses the dismutation of the superoxide (O2−) radical into either ordinary molecular oxygen (O2) or hydrogen peroxide (H2O2), and is reviewed by Scott et al. (1987). Bacillus spp. are well known to express this enzyme (Areekit et al., 2011; da Fonseca et al., 2015).

The molecular structure of azo dyes contains the 6-carbon ring of phenolic compounds. This phenol structure is prevalent in many pollutants, and the role of peroxidases in the degradation of phenols from wastewater has been reviewed (Zámocký et al., 2001; Kulkarni & Kaware, 2013; Chiong et al., 2014; Divate & Hinge, 2014; Pradeep et al., 2015). Phenolic compounds of crude oil have been removed by SOD (Onwurah & Eze, 2000). Peroxidases have been reported to degrade petroleum hydrocarbons and PAHs (Wang et al., 2000; Shekoohiyan et al., 2016), toluene (Abari et al., 2012) and thiazole (Alneyadi & Ashraf, 2016).

Phenolic compounds are abundant in many different pollutants in WW, and peroxidase, SOD and similar enzymes act directly on these compounds (Zámocký et al., 2001; Chiong et al., 2014). Five isolates in this study (2T, 19S, 77S, 83Li and 84Li) as indicated in Chapter 9, Figure 9.5, significantly reduced phenol compared with the control. Only isolate 19S degraded phenol and produced a drop in pH while the other pH values remained neutral. In situ, a reduction in phenol has been reported in wastewater (Kulkarni & Kaware, 2013; Divate & Hinge, 2014; Pradeep et al., 2015).

An FMN-dependent NADH-azoreductase (EC1.7.1.6) catalyses the reductive cleavage of the azo bonds in aromatic azo compounds to their corresponding amines. It requires NADH as an electron donor for its activity. The enzyme can reduce ethyl red and methyl red, but is not able to convert sulfonated azo dyes. Maier et al. (2004) reported that an NADH-dependent azoreductase of Bacillus sp. was responsible for decolourisation of azo dyes. Similarly, a closely related Brevibacillus also produces azoreductase that was able to degrade azo-dye, Reactive Black 5 (Ramlan et al., 2012).

122

Since the isolate 19S expresses catalase peroxidase, azoreductase, quinone oxidoreductase and super oxide dismutase, it may be a prime candidate for further investigations of its bioremediation applications in various substrates.

7.4.8 Removal of lead and chromate

Another function in bioremediation is the removal of lead (Pb) by SOD (So et al., 2001), and chromate by quinone oxidoreductase (Eswaramoorthy et al., 2012; Kabashima et al., 2013; Thatoi et al., 2014). Susceptibility studies against heavy metals have been reported in previous studies and isolate 19S was found to be resistant to Cr and Pb suggesting that these enzymes could be involved in such mechanisms.

7.4.9 Biomonitoring

The other new avenue where enzymes are useful for bioremediation is their use in biomonitoring and or biosensoring of pollutants. They are particularly useful in their specificity, range of detection and sensitivity (Li et al., 2010; Filimon et al., 2013; Lim et al., 2015a). Genetically modified microorganisms using microbial genes fuse reporter genes such as gfp (green fluorescent protein) with inducible promoters to create biomonitoring systems. However without genetic engineering, several microbial enzymes have been useful indicators or pollutants. Atrazine is an ingredient of a herbicide that is genotoxic causing breaks in DNA and is therefore undesirable in water resources. Zhang et al. (2012a) reported that levels of catalase and SOD of B. subtilis due to oxidative stress were induced by the presence of atrazine and this could be useful for biomonitoring of herbicide concentrations. Lead, a toxic heavy metal can be detected by three different enzymes: 1-pyrroline-5-carboxylate dehydrogenase, delta- aminolevulinic acid dehydratase and δ-aminolevulinate dehydratase (Ogunseitan et al., 2000; Korcan et al., 2007; Konuk et al., 2010). In a more general study, the microbial enzymes (catalase, dehydrogenase, urease and phosphatase) of sediment bacteria from sediment of water streams next to a copper smelting complex in Serbia was investigated (Filimon et al., 2013). An enzymatic indicator of sediment quality (EISQ) which included these four enzymes was established and it was found to increase with distance from the major sources of pollution of inorganic pollutants such as Cu, zinc, Pb and As, suggesting the EISQ could be useful as an environmental biomonitoring system. With a similar principle, the effect of various petroleum products on soil dehydrogenase enzymes was evaluated (Kaczyńska et al., 2015). Biodiesel, diesel oil and fuel oil stimulated the enzymes while petroleum was an inhibitor. Changes in lipase, catalase and dehyrogenase activity were investigated in crude oil polluted soil

123

remediated with Bacillus sp. (Ogbolosingha et al., 2015). Lipase increased, while catalase and dehydrogenase decreased as the volume of crude oil in the soil increased.

Isolate 19S from hot springs in SA was identified as Anoxybacillus sp. It was isolated at 53 °C and thermophilic, endospore forming and therefore robust in the environment, and facultatively anaerobic (Mandic-Mulec, 2015). These characteristics make this strain a potential candidate for use in biomonitoring of environmental pollutants. By LC-MS/MS, several useful enzymes for bioremediation were identified. The individual genes for the enzymes can be cloned and investigated further for their possible use in biomonitoring or bioremediation.

7.4.10 Comparison of assays

Conventional plate assay for enzyme detection is suitable for screening purposes, doesn’t require expensive equipment and skill and is easy to perform. However, the results are qualitative and not quantitative, and do not discriminate between different classes of amylase or protease. However, LC-MS/MS is very useful in obtaining a lot of information from a relatively small sample and can distinguish between small variations in chemical compounds. It is, however, expensive and requires specialised equipment and training. For example, isolate 85Li was motile on the agar plate and could not be assessed for amylase production, but alpha- amylase was identified by LC-MS/MS. Isolate 19S was positive for amylase on the plate assay, and by LC-MS/MS, was also found to produce trehalose-6-phosphate hydrolase and oligo-1,6 glucosidase. Similarly, the plate screening assay identified protease, but further analysis by LC- MS/MS identified aminopeptidase, peptidase T and several types of subtilisins. Other limitations of conventional assays are low concentrations of bioactive molecules, and sub- optimal incubation conditions for expression of enzymes and for the assays. Data generated from LC-MS can be used to search several databanks and can therefore be used in retrospect or compared with future analysis (Finoulst et al., 2011). According to Lu et al. (2014), LC-MS is a useful tool to identify secondary metabolites. Although amylase is very well researched, it is still possible to make new discoveries of amylase with different properties. Chai et al. (2016) recently reported the discovery of a new subfamily of glycosyl hydrolase (GH13) from the crystal structure of alpha-amylase of Anoxybacillus sp.

124

CHAPTER EIGHT: BIOPHYSICAL CHARACTERISTICS (BIOFLOCCULANTS, BIOSURFACTANTS, BIOSORPTION AND ANTI-BIOFILM) OF BACTERIAL ISOLATES CULTURED FROM HOT SPRINGS

8.1 INTRODUCTION

Environmental pollutants include PAHs, petroleum, pesticides, pharmaceutical residues, textile dyes, heavy metals and residues from food and beverage industries (Chen et al., 2015b; Liu et al., 2017). Indirect consequences of pollutants in water can also be the formation of biofilms (Farkas et al., 2012), or increased colouration or turbidity by pigment pollutants (Santal & Singh, 2013). All these factors can result in poor water quality such as colour, taste, odour and toxicity.

Bacteria use different methods for bioremediation (bioassimilation, biosorption, bioflocculation, biosurfactant) which are defined and described in detail in Section 2.6 (Vidali, 2001) for the degradation or transformation of contaminants into nonhazardous or less hazardous substances (Karigar & Rao, 2011). Potential extracellular enzymes identified from these hot-spring isolates in this study that are useful for WW bioremediation are discussed in Chapter 7.

The properties of the pollutants determine which method is best to implement. For example, heavy metals are removed by bioflocculation (Abbas et al., 2014) or biosorption (François et al., 2011; Ozdemir et al., 2012), while textile dyes are removed by biosorption (Liu et al., 2011) or enzyme degradation (Ramlan et al., 2012; Bandounas et al., 2013). Water-insoluble petroleum and PAHs are removed with biosurfactants (Pacwa-Plociniczak et al., 2011) and bacterial enzymes (Peixoto et al., 2011; Abbas et al., 2015). Biosurfactants (Banat et al., 2014) and enzymes (Gautam et al., 2013) also have the ability to interfere with biofilm formation.

South Africa is a water-scarce country with its most valuable resource being threatened by pollution including acid mine drainage (Adler et al., 2007) and agricultural fertilisers and pesticides (Schulz, 2001). This study screened bacterial isolates from hot springs for biophysical characteristics that may be useful for water bioremediation, with applications in addressing the current pollution problems in SA.

125

8.2 METHODOLOGY

The origins of the isolates and geographical hot-spring sampling locations are described in Section 3.1.1. The methods, media and incubation conditions used to isolate bacteria from water and sediment samples are explained in Section 3.2. Relevant isolates were identified by 16S rDNA amplicon sequencing following PCR with universal primers and Sanger sequencing as depicted in Section 3.4.

Bioassimilation of triphenylmethane dye (BB and CV) was investigated using isolate 9T mentioned in Section 3.13. Section 3.14 describes the method used for biosorption to screen the dry biomass of bacterial cells of the isolates for their ability to absorb heavy metals Cr, Cu, Fe and Ni in aqueous solutions with the reduction in heavy metals determined spectrophotometrically. The flocculating activity of the CFCS was determined using a suspension of kaolin clay; the details are found in Section 3.15. The positive control was 20 mg/mL alum and the negative control was 1% SLS. The biosurfactant activity of the CFCS extracts of the isolates was tested using paraffin oil (kerosene), petroleum and sunflower oil as substrates in an emulsion assay (Section 3.16.1). A second assay to determine biosurfactant activity was the drop-collapse assay on Parafilm M (Section 3.16.2). The assay used to measure anti-biofilm properties of the CFCS extracts of the isolates has been described in Section 3.16. To determine whether there was a statistical difference between all negative controls and samples, a one-way ANOVA analysis was performed (Section 3.21.2). Cell free culture supernatants were processed for LC-MS/MS according to Section 3.19, and compared to a protein database of Bacillus spp.

8.3 RESULTS AND DISCUSSION

8.3.1 Bioassimiliation

Using 16S rDNA sequencing, mesophile isolate 84Li was identified as Brevibacillus sp., and thermophile isolate 9T as B. subtilis (Table 4.1). In a previous study (Section 4.3.2), both showed the ability to cause clearing on BB-NA around the colony growth and were therefore selected for further analysis. When both isolates were cultured statically in liquid media with BB for four days, and centrifuged, the resulting cell pellets were obtained (Figure 8.1A). The observed increase in blue colouration of biomass of B. subtilis 9T compared with Brevibacillus

126

sp. 84Li, suggested that either biosorption or bioaccumulation of BB dye had occurred by B. subtilis 9T.

9T 84Li

Figure 8.1A: Cell pellet of thermophile isolate 9T (LHS) and mesophile isolate 84Li (RHS) after centrifugation of cultures grown in 0.1% bromothymol blue (BB) in nutrient broth statically for 4 d showing the difference in colouration of the biomass

Since isolate 9T was originally isolated at 53 °C, it was therefore grown at this high temperature for the preliminary experiment. However, bioremediation often occurs at lower temperatures in the environment and therefore a comparison was made between 37 °C and 53 °C under the same conditions. B. subtilis 9T grew at both temperatures, indicated by the presence of biomass at both temperatures. However, the cells were only pigmented at 53 °C and not at 37 °C (Figure 8.1B), suggesting that the blue colouration of the bacterial cells was associated with change in metabolism. The accumulation of dye was therefore suspected to be bioassimilation (metabolism dependant) rather than biosorption because growth occurred at both temperatures but differed in colouration of cell pellet.

37°C 53°C

Figure 8.1B: Cell pellets of isolate 9T grown at 37 °C (left) compared with growth at 53 °C (right) in duplicate

127

To determine whether B. subtilis 9T could remove BB dye from the media, 0.1% BB-NB was inoculated and sampled at several time intervals up to 10 d (Figure 8.2). One-way ANOVA showed that there was a significant difference between day 1 and day 7 (p<0.05), and BB can be reduced by 20% under these current experimental conditions. Unlike biosorption which does not require metabolism, bioassimilation is slower (Abbas et al., 2014). The removal of BB requires growth of the bacteria into an active state and a difference was only observed after 7 days; it was limited by experimental conditions of small scale and static conditions.

Figure 8.2: Removal of bromothymol blue (BB) from media by B. subtilis isolate 9T determined by OD at 595 nm

There are a number of different types of synthetic dyes that are used in textile industries. Azo- dyes are the most produced and most commonly used (Ali, 2010; Khan et al., 2013; Saranraj, 2013), and as a result also have been extensively studied. Triphenylmethane dye is a group of synthetic textiles dyes, of which BB is a derivative. These dyes are recalcitrant and therefore persistent in the environment. Even in small concentrations, they may be visible to the human eye in water, and result in discolouration and decreased visibility in water features. This blocks sunlight to aquatic fauna and flora and therefore has an additional hazardous effect in that it causes aquatic ecosystems to become unbalanced (Khan et al., 2013). Directly, the dyes and their degradation products are toxic, mutagenic and carcinogenic (Zabłocka-Godlewska et al., 2009) posing a risk to human health.

In this study, decolourisation was observed at 53 °C but not at 37 °C showing that a specific state of metabolism at 53 °C was required for the cell biomass to turn blue. This confirms that the process of decolourisation by B. subtilis 9T was indeed bioassimilation and not biosorption.

128

Many studies have described “decolourisation” of textile dyes by both fungi and bacteria. Fungi are very efficient in decolouring textile dyes. Aspergillus flavus decolourised triphenylmethane dye BB and malachite green (Singh & Singh, 2010; Subramanian et al., 2014). Bacterial decolourisation of triphenylmethane dyes also includes Gram-negative bacteria including Pseudomonas sp. (Wu et al., 2009) and Aeromonas sp. (Ren et al., 2006). Crystal violet, another triphenylmethane dye, has been shown to be decolourised by different bacterial genera (Ali, 2010), including Bacillus sp. (Kochher & Kumar, 2011). In this investigation, the growth of B. subtilis 9T was inhibited by the presence of CV, known to be bacteriostatic and bactericidal (Hoffmann & Rahn, 1944; Ogawa et al., 1988).

Zabłocka-Godlewska et al. (2009) showed that fungi removed >80% of thymol blue compared to <30% removal by bacteria. The discovery of a novel bacterial isolate that has the ability to bioaccumulate triphenylmethane textile dyes, and possibly other recalcitrant pigments, would contribute to the gaps in current literature of bacterial mechanisms involved. If textile dyes and their degradation products are toxic, biosorption is preferred for remediation of textile dyes since cell viability is not required, and is therefore the method of choice. However, the advantage of potentially using a bacterial isolate (B. subtilis 9T) is that it is not only resistant to the dyes but can also grow and actively bioaccumulate BB. That would reduce the need for growing biomass and inactivation of bacteria prior to the bioremediation process. This would therefore reduce costs, time and effort. An additional advantage of using this isolate 9T for bioremediation would be the possibility that it can be used not only to decolourise polluted water, but also to detoxify the environment; however, further toxicity studies are required.

Since bioassimilation is an active process requiring metabolism, unlike biosorption, optimisation of conditions is critical. Further investigation of B. subtilis 9T is required, in relation to temperature, pH, carbon and nitrogen source, and agitation conditions. Factors that have an influence on bacterial decolourisation are reviewed by Pearce et al. (2003) and Khan et al. (2013). In this study, BB was used as a derivative of triphenylmethane dyes; however, other pigments and dyes could be tested to determine its mode of activity and specificity. For example, melanoidin is a pollutant that also results in increased colouration of water, e.g. of the brewery or winery industry, sugar molasses processing, or coffee production (Chavan et al., 2006; Chandra et al., 2008), and also increases the turbidity or colouration of wastewater.

The lack of information on the bioassimilation of dyes by Bacillus spp. could be due to misuse of terminology because there is good evidence that this genus is important in bioremediation of textile dyes. In four reviews, only biosorption and biodegradation but not bioassimilation are mentioned as methods that bacteria use to decolourise textile dyes (Ali, 2010; Ong et al., 2011;

129

Khan et al., 2013; Sudha et al., 2014). Issazadeh et al. (2011) described bioaccumulation of Cu, Zn and Cd; however, the process was actually biosorption by definition. Numerous studies have described biodegradation of dyes by Bacillus spp. (Kochher & Kumar et al., 2011; Anjaneya et al., 2011; Sudha et al., 2014; Santos et al., 2014; Ezhilarasu, 2016). Other bacteria that are commonly isolated from hot springs are Brevibacillus spp. (Inan et al., 2012; Xian et al, 2016), and Anoxybacillus spp. (Cihan, 2013; Yohandini et al., 2015); both have been isolated in this study. Both of these genera have been reported to biodegrade textile dyes (Gursahani & Gupta, 2011; Ramlan et al., 2012). The production of extracellular enzymes that are associated with dye discolouration by these hot-spring isolates have been described in Chapter 7 and further comment will be made in Chapter 9 in relation to WW bioremediation.

Preliminary investigations show that B. subtilis 9T, a thermophilic isolate, can bioassimilate BB, a triphenylmethane dye at 53 °C and thus has the potential to be useful for bioremediation of textile industrial WWs.

8.3.2 Biosorption

South Africa is a water-scarce country with half its surface area receiving <200 mm rain annually. In addition, mining is a major contributor to the gross domestic product in this country resulting in many of the waterways inevitably being negatively affected by mining activity (https://www.brandsouthafrica.com/investments-immigration/economynews/sa-economy-key- sectors).

In this investigation, isolates from hot springs were screened for their ability to biosorb heavy- metal ions: chromium (Cr), copper (Cu), iron (Fe) and nickel (Ni). Standard curves (Appendix 11) were drawn to show that OD at 415nm (for Ni, Fe and Cr) or 750 nm (for Cu) could be used to indicate the concentrations of heavy-metal ions in water. All samples were compared to a negative control that was not in contact with bacterial biomass; only sterile NB present with the heavy-metal ions solution.

Using the one-way ANOVA (confidence interval of 95%, p<0.05 was regarded as significant) as described in Section 3.21.2, only significant reductions in heavy-metal ions relative to the control were observed as follows: With Cr: 7T (p = 0.0035), 9T (p = 0.0004), 16S (p = 0.0002), 20S (p = 0.0004), 30M (p = 0.0016), 71T (p = 0.0006), 83Li (p = 0.0006) and 84Li (p = 0.0025); with Fe: 20S (p = 0.0183), 30M (p = 0.0043), 83Li (p = 0.0022); and with Cu: 7T, (p = 0.0016), 9T (p = 0.0001). Nickel was not reduced by any of the isolates tested. Anoxybacillus

130

sp. isolate 7T and B. subtilis isolate 9T could reduce Cr and Cu from solution, while isolates B. subtilis 20S, B. licheniformis 30M and B. subtilis 83Li could reduce Cr and Fe from solution.

Figure 8.3: Biosorption activity of dead cell biomass in aqueous solutions of chromium (Cr), copper (Cu), iron (Fe) and nickel (Ni)

Biosorption is a very effective way of dealing with bioremediation because bacteria have a large surface area. The process is passive and does not require energy and time to keep the bacterial biomass viable. Both live and dead cells can be used, so inactivation is not necessary. Bacterial biosorption can address several pollutant problems simultaneously. For example, in this study, two different metals could be removed by isolates 7T, 9T, 20S, 30M and 83Li. Sadettin and Donmez (2007) also described thermophilic cyanobacteria from hot springs in Turkey that could remove textile dyes at the same time as Cr heavy-metal ions. Biosorption is favoured for the removal of heavy metals because heavy metals are toxic, and viable cells for bioassimilation processes would not survive. However, a disadvantage is that a large volume of bacterial cells is required to obtain the dry weight of biomass. In this study between 600 mg and 800 mg wet weight was obtained per litre bacterial culture after a growth period of 48 h. After drying, a 10- fold reduction in weight was observed to give the dry weight of 60-80 mg per litre bacterial culture, used in this assay. There would therefore be a cost associated with generating sufficient bacterial biomass for biosorption to be used in bioremediation. Another economic consideration is the processing of spent biomass where in the case of heavy metals the metals are desorbed into a liquid fraction, and the biosorbent can be reused. The concentrated toxic metals can either be safety disposed of, or recovered by electrolysis (Zabochnicka-Swiatek & Krzywonos, 2014).

131

In this study, the reductions of heavy metals by these selected isolates ranged between 10% (Cu) and 23% (Fe) with intermediate reductions of 11% for both Cr and Ni. From a review of the literature, other investigators have reported much larger reductions. Copper was reduced by 32% and Cr by 95% by B. licheniformis (Abbas et al., 2014). Bacillus sp. reduced Cu (3.332 mg/L) by 56% and Ni (3.8 mg/L) by 48% in batch cultures (Kumar et al., 2010). Anoxybacillus amylolyticus removed 40% of Cu (Ozdemir et al., 2013), and immobilised Bacillus spp. removed >50% of Cu, Cr and Ni (Al-Daghistani 2012). However, there are many factors that affect biosorption, i.e. pH, initial metal concentration, amount of biomass, temperature and contact time (Ozdemir et al., 2013). The values obtained by Abbas et al. (2014) were read after 48 h, while an exposure period of 5 h was used in this study. The biosorption of Cu was optimal at a pH of 4 for Anoxybacillus (Ozdemir et al., 2013), which could affect the biosorption abilities of isolates 4T, 7T and 19S in this investigation. The concentration of the initial heavy- metal solution is also a factor affecting biosorption. When an initial concentration of 500 µg/mL Cr (VI) ions was used with Bacillus cereus isolated from hot springs in Pakistan, only 20-30% was removed compared to 90-95% when an initial concentration of 100 µg/mL was used (Ghalib et al., 2014).

In this study, the concentration used was much higher at 1.67 mg/mL (1 670 µg/mL) based on the detection range within the standard curve (Appendix 11) of the experimental setup used. Lower concentrations in further studies might reveal a more efficient reduction in Cr. Sabae et al. (2006) reported that live cells of B. subtilis were more efficient than dead cells in the uptake of heavy metals, but Ozdemir et al. (2012) showed that dead cells of Geobacillus sp. were more efficient than live cells. Further studies are required to optimise conditions for biosorption of heavy metals by these isolates.

With the mining sector being extremely important to the economy of SA, it is inevitable that mining activities will give rise to heavy-metal pollution of the the country’s water resources. Dzoma et al (2010) found that heavy-metal contamination of uranium, As, Pb, Cd and aluminium was directly associated with mining activity in the North-West Province of SA. In addition, electronic waste, a relatively recent pollutant due to technology, contributes to toxic metals into the environments (Bhattacharya & Khare., 2016). This must be differentiated from natural heavy-metal contamination of groundwater from leaching of the earth’s crust. Heavy metals are defined as metals with a density >5 g/cm3 and atomic number >20, and are toxic and hazardous to human health (Abbas et al., 2014). Chrome (Cr) is found in textiles, dyeing, paints, pigments and steel fabrication causing cancer, mutations, teratogenic, epigastria pain, nausea, vomiting, diarrhoea, lung tumours. Copper (Cu) is used in plating, copper polishing, paint,

132

printing and causes neurotoxicity, dizziness and diarrhoea. Nickel (Ni) is found in porcelain enamelling, non-ferrous metal, paint, electroplating and causes chronic bronchitis, decreased lung function, and lung cancer. Iron (Fe) is not regarded as a highly toxic metal and is essential to life; however, in high concentrations it can cause blood disorders and cancers (Guo et al., 2015).

The Gram-positive, non-fastidious endospore-forming genus Bacillus has been well documented to remove heavy metals (Sabae et al,, 2006; Kumar et al., 2010; Abbas et al., 2014) and dyes, including triphenylmethane dyes (Ye et al., 2013) by biosorption. The surface layer (S-layer) proteins of the cell wall and teichoic acids are involved in this process (Allievi & Mariano, 2011). The more recently discovered Bacillus-related species Anoxybacillus is a facultative anaerobe, and has been reported to biosorb heavy metals (Ozdemir et al., 2013), and the responsible cell-wall components have been characterised (Zhao et al., 2014b).

Hot springs are good resources for isolation of useful Bacillus spp. in Pakistan (Ghalib et al., 2014) and Anoxybacillus spp. in Jordan (Al-Daghistani et al., 2012) for the removal of heavy metals. This was confirmed in this investigation where B. subtilis, B. licheniformis and Anoxybacillus spp. from hot springs in SA were able to reduce heavy-metal concentrations of Cr, Cu and Fe in aqueous solutions by biosorption, and its applications to wastewater remediation will be discussed in Chapter 9.

8.3.3. Bioflocculant activity

The kaolin clay assay is a standard test for bioflocculant activity, which determines the ability of extracellular bacterial components to form flocs of solid particles of clay allowing them to settle out (Zhang et al., 2007).

There was a significant difference between the positive (alum) and negative (SLS) controls where p<0.05.

Three isolates, namely B. subtilis 20S (p = 0.0000), B. licheniformis 30M (p = 0.0207) and Brevibacillus sp. 84Li (p = 0.0012) were significantly different from the negative control and could clarify particles of kaolin clay in an aqueous suspension ranging from 46% to 26% (Figure 8.4). Bioflocculant positive B. coagulans B8 reduced the particles by 37% in the kaolin clay assay (Bajlan et al., 2013) and Humudat et al. (2014) reported that the bioflocculant produced by Bacillus resulted in flocculation of kaolin clay suspensions comparable to alum, the positive control which is comparable to the results of this study. Further investigation is

133

required to optimise conditions for the use of B. subtilis 20S as a bioflocculant (Zhu et al., 2014).

Figure 8.4: Bioflocculant activity of isolates from hot springs, South Africa using the kaolin clay assay

Five isolates (7T, 15S, 23M, 35Li and 71T) were observed to disperse kaolin clay even more than the 1% SLS negative control (Figure 8.4). This might be a result of the presence of natural biosurfactant molecules which is discussed in Section 3.4

The cellular component responsible for biosorption is expressed on the cell wall. However, macromolecules, usually exopolysaccharides secreted into the culture media, are responsible for bioflocculant activity (Al-Wasify et al., 2015). They also have an absorbent nature, but this should not be confused with cell-wall biosorption described in Section 3.2. The characterisation of bioflocculants produced by B. subtilis (Yoon et al., 1998; Giri et al., 2015), B. licheniformis (Xiong et al., 2010; Karthiga & Natarajan, 2015) and Bacillus spp. (Nontembiso et al., 2011; Okaiyeto et al., 2015) has been described. The bridging mechanism of the polymer between the particles is described by Czemierska et al. (2015).

134

A bioflocculant therefore has biosorption properties but differs in that it is extracellular and forms flocs, resulting in clarification through sedimentation. Azzam & Tawfik (2015) correctly describe the removal of heavy metals by bacterial bioflocculants produced by B. subtilis and Pseudomonas since the CFCS was used in the assay. Commonly, the terms biosorption and bioflocculation are used interchangeably, especially in articles relating to heavy-metal removal (Francois et al., 2012; Abbas et al., 2014; Zabochnicka-Swiatek & Krzywonos, 2014).

The usefulness of bioflocculants in bioremediation is wide-ranging from dairy and brewery WW and polluted river water in the removal of suspended organic matter, reduction of turbidity and odour (Agunbiade et al., 2016), to industrial WW from mining, chemical, petroleum and metallurgical industries (Czemierska et al., 2015) in the removal and recovery of metal ions, and unwanted chemicals.

8.3.4 Biosurfactant assay

8.3.4.1 Emulsion assay 1 2 3 4 5

Figure 8.5: Emulsion activity of positive control (1% SLS in nutrient broth) (tube 1); negative control (nutrient broth only) (tube 2); high emulsion activity (tube 3 – isolate 4T); medium emulsion activity (tube 4 - isolate 7T); and low emulsion activity (tube 5 - isolate 13) using kerosene (paraffin oil) as substrate

Three Anoxybacillus species, namely isolates 4T, 7T and 19S had the highest emulsion activity in paraffin oil, of between 70 and 110% compared to the positive control (1% SLS) (Figure 8.6). However, they were not positive in the drop-collapse assay (as indicated by the blue cross on Figure 8.6). Eight isolates that were positive in the drop-collapse assay only had <40% of emulsion activity, suggesting that the bioactive agents responsible for the emulsion activity were not necessarily the same ones that resulted in the decrease of surface tension of droplets on

135

Parafilm M. Isolates 54T and 72T were negative for both emulsion activity of paraffin oil and the drop-collapse assay.

Figure 8.6: Emulsion activity showing biosurfactant activity of CFCS of isolates from hot springs, SA (green bars) with positive results in the drop-collapse assay (blue crosses)

8.3.4.2 Drop-collapse assay for biosurfactant on Parafilm M

A B

Figure 8.7A: Drop-collapse assay for biosurfactant assay on Parafilm M. Bromothymol blue added to view negative control (left) and positive control (right). Figure 8.7B: Screening of CFCSs on drop- collapse assay

136

The drop-collapse assay for biosurfactant activity is fast, easy and simple to perform. The results of this investigation are listed in Table 8.1. Kalyani et al. 2014 showed good correlation of this assay with the oil-spreading assay, a conventional and robust assay for biosurfactant detection. The drop-collapse method was recommended for primary screening followed by emulsion assay in a study of 45 different types of bacteria in eight different methods of biosurfactant testing (Satpute et al., 2008).

8.3.4.3 Emulsion activity of selected isolates using petroleum and sunflower seed as substrate

Table 8.1: A comparison of biosurfactant activity of 12 isolates of emulsion activity against paraffin oil, petroleum and sunflower seed oil, and Parafilm M drop-collapse assay where (-) is negative and (+) is positive

EMULSION DROP ACTIVITY COLLAPSE SUNFLOWER SEED ISOLATE NO. PARAFFIN OIL (%) PETROLEUM OIL PARAFILM M Pos 1% SLS + (100) + + + Neg water -(0) - - - 76S + (22) + + + 85Li + (19) + + + 71T + (19) + + + 16S + (22) + + + 50T + (31) + - + 84Li + (17) + - + 35Li + (39) + + + 4T + (115) - + - 7T + (72) - + - 19S + (92) - + - 77S + (25) + - - 54T - (0) - - - Total positive 11 7 8 7

Four isolates (16S, 71T, 76S and 85Li) were positive in all four assays (Table 8.1). Isolate 54T was negative for both emulsion assay with paraffin oil and the drop collapse assay and included as an additional negative control and confirmed negative with petroleum and sunflower oil as substrates. Variations of results were observed for the other isolates. This suggests that different types of biosurfactants may be present but further characterisation is required. When eight methods of biosurfactants were tested, Satpute et al. (2008) found that one method was not good enough to detect all candidates. In this investigation, the emulsion assay with paraffin oil identified 11 positives, while emulsion with petroleum and sunflower seed oil, and the drop- collapse assay only detected seven or eight of the positives. This suggests that the emulsion assay with paraffin oil was the most efficient test including being better than the drop-collapse assay. However, Satpute et al. (2008) found that the opposite was true. Because their

137

investigation used 45 strains of Acinetobacter and other bacteria, the difference might be explained by the difference in genera of bacteria used. To support this idea, Satpute et al. (2008) included five reference type strains of Bacillus sp. All five were positive by emulsion assay but only three were positive on the drop-collapse assay supporting the findings of this study. Three methods (oil spreading, drop collapse and blood agar lysis) were compared for 205 environmental isolates, and the oil spreading method was found to be the best although the drop-collapse assay was useful and recommended for a quick primary screening. However, the blood agar lysis assay was not found to be reliable as a detector of biosurfactant activity (Youssef et al., 2004). Akintokun et al. (2017) isolated 12 Bacillus spp. isolates from pharmaceutical WW which were all positive in both the emulsion assay using kerosene and the drop collapse assay. This was surprising since the group included of seven different species of Bacillus. Similiarly, two Bacillus strains from the Red Sea in Egypt were also both positive for the paraffin oil emulsion assay and drop collapse assay (Barakat et al., 2017) showing there is variability in different Bacillus spp. isolates as indicated in this study.

The substrate used in the emulsion assay is also a determining factor in biosurfactant detection. Using Pseudomonas sp., Ramani et al. (2012) found that emulsion activity ranged from 32% with paraffin oil (kerosene), 46% with petroleum and 60% with olive oil being the best substrate. Biosurfactants of Klebsiella and Citrobacter were tested against olive oil, paraffin oil, castor oil and coconut oil, bitter almond oil and cyclohexanol. Coconut oil was found to give the highest biosurfactant activity more than paraffin oil (Hassan et al., 2014). Petroleum was better as a substrate compared with paraffin oil for a Pseudomonas strain (Samanta et al., 2012b). Four different substrates were used by Perez et al. (2017) with two Bacillus spp. isolated from puba (fermented cassava product). No emulsion was observed with xylene, while one isolate was positive with mineral oil and the other, positive with biodiesel and toluene. A difference in emulsion activity of different substrates could give an indication of the presence of different biosurfactants with varied functionality.

8.3.4.4 Identification of molecules with biosurfactant potential by LC-MS/MS

The CFCS extracts of isolates 76S and 77S were processed by LC-MS/MS. Subtilisin NAT was identified in extracts from both isolates and subtilisin BPN was also identified in the extract from isolate 77S. Although surfactin is the most well known biosurfactin of Bacillus spp., subtilisin has also been listed as a biosurfactant of Bacillus (Al-Araji et al., 2007). Isolate 76S showed biosurfactant activity with all emulsion substrates tested and drop collapse assay, while isolate 77S, that expressed two types of subtilisins, was only positive with paraffin oil and

138

petroleum emulsion. New technology such as LC-MS/MS has been useful to identify both established and novel biosurfactants in Bacillus spp. (Plaza et al., 2015).

Microbial biosurfactants have been classified into five groups: A Glycolipids; B lipopeptides; C fatty acids, phospholipids and neutral lipids; D polymeric biosurfactants; and E particulate biosurfactants. Bacillus biosurfactants fall into group B, including subtilisin and surfactin from B. subtilis and lichenysin from B. licheniformis. B. subtilis can also produce glycolipid biosurfactants (Reis et al., 2013). More recently discovered biosurfactants, iturin and fengycin, have been described by Plaza et al. (2015). A biosurfactant producing strain of B. licheniformis was isolated from fermented food and found to produce three kinds of lipopeptide biosurfactants, including plipastatin and surfactin, and a novel biosurfactant BL 1193 (Thaniyavarn et al., 2003). The biochemistry of surfactins alone is complicated, consisting of at least 44 different compounds (Shaligram & Singhal, 2010; Liu et al., 2015a).

Anoxybacillus spp. that produced biosurfactants have been isolated from hot springs in Malaysia (Khairuddin et al., 2016) and Thailand (Techaoei et al., 2007; Pakpitchaeroon et al., 2008), similar to thermophilic isolates 4T, 7T and 19T in this study. Brevibacillus sp. 16S, B. methylotrophicus 71T, Bacillus sp. 76S and unknown 85Li were positive for emulsion activity against paraffin oil, petroleum and sunflower seed oil and positive in the drop collapse assay on Parafilm M. Three Anoxybacillus isolates (4T, 7T and 19T) had high emulsion activities with paraffin oil and sunflower seed, but were negative with petroleum and ability to collapse a drop on Parafilm M, suggesting that the biosurfactant of Anoxybacillus differed from that of Brevibacillus and Bacillus spp. biosurfactants.

Not only are biosurfactants useful in dealing with bioremediation of WW treatment, but several WW have been shown to be useful as substrates for biosurfactant production, e.g. potato cassava WW (Nitschke & Pastore, 2006), vinasse and frying oil WW (Oliveira & Garci-Cruz, 2013; Khairuddin et al., 2016).

139

8.3.5 Anti-biofilm activity

Figure 8.8: Anti-biofilm activities of isolates from hot springs in South Africa

Figure 8.9: Crystal violet staining of anti-biofilm assay of isolate 16S, 21M, 71T and 75S including positive and negative controls

Commonly, biosurfactants also have properties of preventing biofilm formation (Banat et al., 2014) and therefore biosurfactant positive isolates as determined in Section 3.4 above, were selected for screening for anti-biofilm properties. Although CFCS of Brevibacillus sp., isolate 16S visibly (Figure 8.8) produced the most active anti-biofilm activity amongst the nine isolates tested, it was not significantly different from the negative control (p = 0.4) (Figure 8.9). Anti- biofilm activity of isolate 16S was significantly different to isolate 4T (p = 0.0001), 19T (p = 0), isolate 21M (p = 0.0223), isolate 35Li and 71T (p = 0), but not to the negative (p = 0.4) and positive (p = 0.25) control. The CV staining in the assay, does not take into account a change in biomass formation, as there were concentrated buttons of biomass at the bottom of isolate 16S

140

test tube that took up the stain but the biofilm formation was clearly disrupted. Therefore, the potential biofilm-forming cells of isolate 54T were transformed into an adherent lump of compact cells which still gave a CV staining reading but no biofilm was visible. It was also previously mentioned by Bueno (2014) that errors in this assay could be due to incomplete removal of the unattached bacterial material from the surface, in the review describing methods to study biofilms. The investigator of this study recommends that both visual and quantitative results be noted, and that the assay be optimised further with more stringent washing steps. It is possible that biofilm promoting molecules are present in CFCS samples of isolates 4T, 19T, 35Li and 50T and their higher values compared with the negative control are genuine results and not due to an error in incomplete washing.

The isolates that produced the highest emulsion activity on paraffin oil (4T, 19T), shown in Figure 8.6, did not produce the maximum anti-biofilm activity (16S) shown in Figure 8.8, and vice versa. This suggests that the effect was not due to the quantity of biosurfactant, but the action of a specific biomolecule yet to be characterised.

Compounds that prevent biofilm formation are either anti-adhesive or antimicrobial and further tests are required to investigate the mechanism used by Brevibacillus sp. 16S. Because the CFCS extracts were crude and not purified, it is possible that the supernatants had one or more components with anti-biofilm properties. Enzymes such as protease and amylase are able to work in conjunction with biosurfactants (Samanta et al., 2012b; Bhange et al., 2016), and promote biofilm removal (Molobela, 2010). Antibiotic-like antimicrobial components have also been implicated as molecules that “prevent biofilm formation” (Banat et al., 2014).

The target bacteria that was used for biofilm formation in this study was B. subtilis 54T also isolated from hot springs, Limpopo, SA. Although this particular isolate was not well characterised, the biofilm of B. subtilis has been well investigated (Lemon et al., 2008) and is recommended for use as a “model biofilm”. The application of testing against a Bacillus biofilm is useful for environmental issues where biofilms are physical hindrances, or harbouring antibiotic-resistant organisms (Patel et al., 2013). It would be of interest to use different target bacteria for biofilm formation that included pathogens (Klebsiella, Mycobacterium, Aeromonas) (Sayem et al., 2014). Bacillus spp. (Wu et al., 2013b) and Brevibacillus spp. (Zimmer et al., 2013) have also been reported to actively inhibit biofilm formation.

141

8.3.6 16S rDNA identification of isolates useful for wastewater bioremediation

Table 8.2: Isolates from hot springs with biophysical attributes relevant to wastewater bioremediation

ISOLATE ISOLATION GENBANK NO LOCATION TEMP °C IDENTIFICATION ACC NO. COMMENT

4T Tshipise 55 Anoxybacillus rupiensis 99% AM988775.1 biosurfactant biosurfactant,

7T Tshipise 55 Anoxybacillus sp. 99% KJ722458.1 biosorbent bioaccumulation,

9T Tshipise 53 Bacillus subtilis 99% JN366797.1 biosorbent biosurfactant,

16S Siloam 53 Brevibacillus sp. 99% LN681596.1 anti-biofilm

19S Siloam 55 Anoxybacillus sp. 99% KP221933.1 biosurfactant bioflocculant,

20S Siloam 53 B. subtilis 99% KC634086.1 biosorbent

30M Mphephu 53 Bacillus licheniformis 99% KJ526854.1 biosorbent biofilm

54T Tshipise 37 B. subtilis 99% HM753614.1 formation

76S Siloam 37 Bacillus sp 98% KF956597.1 biosurfactant Bacillus methylotrophicus

77S Siloam 37 99% KP342210.1 biosurfactant

83Li Libertas 37 B. subtilis 97% KF533727.1 biosorbent biosurfactant,

84Li Libertas 37 Brevibacillus sp. 99% LN681596.1 biosorbent

85Li Libertas 37 Brevibacillus formosus 97% KP165013.1 biosurfactant

Thirteen isolates were sequenced for identification using the 16S rDNA sequence. Mesophilic (37 °C) and thermophilic (53 °C) were found to express biophysical properties for bioremediation purposes (Table 8.2); however, assays were generally performed at room temperature to simulate “natural” conditions. On the other hand, bacterial isolates with these activities that can function under high temperatures may be applicable to other fields of industry. Bacillus spp. (Kumar et al., 2013), Brevibacillus sp. (Panda et al., 2014) and Anoxybacillus sp. (Goh et al., 2013) were identified, and this is consistent with findings in the literature (see Section 2.6 Table 2.2). These bacteria are well known to be extremely useful in biotechnology including WW bioremediation.

142

CHAPTER NINE: POTENTIAL OF HOT-SPRING WATER ISOLATES FROM SOUTH AFRICA IN WASTEWATER BIOREMEDIATION

9.1 INTRODUCTION

Wastewater from agricultural, industrial and domestic sources contains numerous and complex pollutants that are both organic and inorganic components (Schwarzenbach et al., 2010).

Persistent organic pollutants (POPs) are a large and diverse group of recalcitrant chemicals that enter the environment (Xu et al., 2013). Not only are they transported great distances, but they also accumulate in the food chain (Han & Currell, 2017), and ultimately can cause many health issues in humans (Schwarzenbach et al., 2010). Poly aromatic hydrocarbons (PAHs), a priority group of POPs includes diesel, heavy oil and petroleum pollutants (Fan & Krishnamurthy, 1995). Dyes found in many products are synthetic aromatic water-soluble organic colourant POPs (Buntic et al., 2017) directly affect the environment by their toxicity, or indirectly, by increasing turbidity of the water and upsetting the natural ecosystem (Ramachandran et al., 2013; Santal & Singh, 2013). Triphenylmethane textile dye precurors (Figure 2) include phenol red (Mittal et al., 2009), BB (Agarwal et al., 2016) and CV (Buntic et al., 2017), and have been used in the laboratory for decolourisation experiments. Melanoidins are brown high-molecular- weight heterogeneous polymers found in WW of food, drinks and fermentation processes polluting water in similar ways to textile dyes (Chandra et al., 2008; Langner & Rzeski, 2014). The phenol structure of stable aromatic hydrocarbons with hydroxyl substitutions (Figure 2a) is found in many POPs including the ones mentioned above (Ahmad et al., 2011). At even very low concentrations of 0.02 mg/mL, they are toxic and have offensive odours in drinking water (Vedula et al., 2013). Because of the complex, diverse and ubiquitous nature of this group of pollutants, bioremediation by bacteria are highly desirable and well investigated (Arvind et al., 2015; Chakraborty & Das, 2016). The different modes of bacterial processes are covered in Sections 2.6, and in Chapters 7 and 8.

In this study, hot-spring isolates from Limpopo Province, South Africa, were found to produce numerous enzymes (Chapter 7), and express several biophysical properties (Chapter 8). Ultimately the litmus test for bacterial bioremediation is testing against samples of WW rather than single pollutants under controlled conditions in the laboratory, because there are so many contributing factors (Ali, 2010). This has been performed with WW from textile industries (Ramachandran et al., 2013), food processing (Chiacchierini et al., 2004; Tiwari et al., 2012;

143

Santal & Singh, 2013; Torres et al., 2016) and PAH polluted environmental samples (Peixoto et al., 2014; Abbas et al., 2015). Phenolic compounds have been measured in olive oil WW (Atanassova et al., 2005) and brewery WW (Tatullo et al., 2016) using the Folin-Ciocalteu (FC) reagent.

The aim of this study was the identification of previously isolated bacteria from hot springs in SA with biodegradation potential for application in WW bioremediation, as preliminary results had indicated that relevant enzymes and suitable biophysical characteristics were present. Cell- free culture supernatants (CFCS) were tested to establish whether they could reduce turbidity or colouration in food pollutants containing melanoidin, textile pollutants including precursor textile dyes BB, a commercial dye (Dye It, CDS1, SA) and three samples from brewery and dairy WW and river water contaminated with industrial effluents. Since phenol is a priority environmental pollutant (Galgale et al., 2014), further testing was performed in order to determine whether there was also a reduction in the phenolic compounds present in these samples.

9.2 METHODOLOGY

Water and sediment samples were collected from hot springs Limpopo, SA as described in Section 3.1.1. Section 3.4 describes the identification of bacterial isolates by PCR and gene sequencing of the 16S rDNA amplicon.

Section 3.1.2 explains how the collection of samples from brewery and dairy wastewaters and from a site on the river receiving industrial effluents was performed; GPS coordinates, physicochemical parameters and photographs are also listed. Environmental samples were processed in the laboratory for the isolation and maintenance of pure cultures (Section 3.2).

Spectrophotometry was used to determine whether there was a change in turbidity or colouration of WW when mixed with CFCS of different isolates (Section 3.20). Phenol concentrations were determined using the Folin-Ciocalteu (FC) reagent in both experiments with phenol red and different WW samples as described in Section 3.12.5.Since WW bioremediation occurs at room temperature and the isolates were obtained from hot springs, the affect of temperature on the reduction of WW phenol levels were determined as explained in Section 3.20.1.

Statistical differences between negative controls and samples were determined by one-way ANOVA analysis with a p<0.05 being significant (Section 3.21.2).

144

9.3 RESULTS

Table 9.1: Summary of the experiments performed with substrates tested against CFCS fractions of isolates for reduction of colouration or turbidity and/or phenol concentrations at 25 °C for 3 h where (nd) means not done.

Substrates Colouration or turbidity Phenol concentrations Coffee (Figure 9.1) (Figure 9.6) Soya sauce (Figure 9.1) (Figure 9.6) Bromothymol blue (Appendix 16) (Appendix 17) Crystal violet nd (Appendix 17) Commercial dye (Figure 9.2) (Figure 9.7) Brewery WW (Figure 9.3) (Figure 9.8) Dairy WW (Figure 9.3) (Figure 9.8) Polluted river water samples (Appendix 18) (Figure 9.9) Paraffin nd (Figure 9.9) Petroleum nd (Figure 9.9)

In this investigation, hot-spring isolates from Limpopo Province, SA produced enzymes (Chapter 7) and possess biophysical properties (Chapter 8) that could be useful for WW bioremediation. In this investigation, the CFCS fractions were tested against pigmented pollutants (melanoidin and triphenylmethane dyes) and three WW samples (brewery and dairy wastewaters and river water contaminated with industrial effluents) to establish whether there was a reduction in colouration or turbidity and/or phenolic compound concentrations as indicated in Table 9.1. In addition, phenolic compounds in paraffin and petroleum were also tested, and since they are both clear fluids, turbidity was not a problem. Values are described as a percentage of the control, means and standard deviations are indicated from the triplicate data. Using a one-way ANOVA statistical analysis, positive results were regarded as significantly different from the control when p<0.05, as indicated in the discussion below.

9.3.1 Reduction of colouration or turbidity

Reduction of colouration or turbidity was measured spectrophotometrically at different wavelengths as described in Section 3.20 and Table 3.3. To show that there was a linear relationship between OD at the appropriate wavelengths and the concentration of pollutants or WW samples, standard curves are given in Appendices 10, 12 and 13.

145

9.3.1.1 Reduction of colouration due to food pollutants (coffee and soya sauce)

* denotes a significant reduction in colouration of soya sauce

Figure 9.1: Colouration as percentage of negative control of coffee and soya sauce treated with CFCS of hot spring isolates.

Of the 28 isolates tested, none could reduce the colouration of coffee and two isolates (76S & 77S) reduced the colouration in soya sauce by 7-8%.

9.3.1.2 Reduction of colouration due to derivative of textile dyes (bromothymol blue and commercial textile dye Dye It No. 18)

The extracellular fractions of ten isolates, namely 2T, 7T, 9T, 20S, 30M, 71T, 77S, 83Li and 84Li, were tested to establish whether they could diminish colouration in 0.05 mg/mL BB; however, none were significantly different to the control (Appendix 16).

Two isolates 22M and 35 Li lessened the colouration of commercial textile dye (Figure 9.2) by 5-6%.

146

* denotes a significant reduction in colouration in commercial textile dye

Figure 9.2: Colouration as percentage of negative control of commercial textile dye (Dye It No.18) treated with CFCS of hot spring isolates.

9.3.1.3 Reduction of colouration/turbidity of brewery and dairy wastewaters and river water contaminated with coloured industrial effluents

* denotes significant difference in control and dairy WW

Figure 9.3: Turbidity as percentage of negative control of brewery wastewater and dairy WW treated with CFCS of hot spring isolates.

147

Brewery effluent samples were assayed only in singlicate, and therefore not statistically analysed. The isolates that were found to be most effective in reducing turbidity (>40%) were isolates 35Li (54%) and 72T (59%). Two isolates, namely 35Li and 50T lowered the turbidity of dairy WW significantly by 25%.

The CFCS extracts of six isolates were found not to decrease the turbidity of river water contaminated with industrial effluents (Appendix 17).

9.3.2 Phenol reduction in phenol red broth media by CFCS extracts of isolates

Because the phenolic (Figure 2.2) structures are common in many pollutants, the CFCS extracts of the isolates were screened to establish whether they could reduce the concentration of phenol in phenol red broth media. Phenol red broth media was not used conventionally as a pH indicator medium, but rather as a source of substrate of phenolic compounds to be tested for degradation or removal.

In order to optimise the phenol assay using the FC reagent (Strong & Burgess, 2008), a positive control of green tea and a negative control of water was used (Figure 9.4A). However, the method was adjusted in order to measure smaller differences between the negative control (phenol red broth media with sterile nutrient broth) compared with tests of different CFCS extracts (phenol red broth media with CFCS fractions) of the isolates (Figure 9.4B).

9.4A 9.4B

148

130 120 110

100 90 80 70 Percentagephenol 60 50

40 1T 3T 8T 9T 12S 15S 76S 78S 2T* 10T 39T 50T 54T 57T 72T 73T 35Li 22M 23M 53M 79M 19S* 77S* 71T* 83Li* 84Li* control Hot Spring isolates and control

*denotes isolates that were positive in their ability to reduce phenol. Pink bars denote a change in pH of media from neutral to acidic.

Figure 9.5: The relative phenol reduction by CFCS of isolates using the phenol as a substrate in Phenol Red Broth media.

Six isolates (2T, 19S, 71T, 77S, 83Li and 84Li) were positive in their ability to reduce phenol concentrations as indicated in Figure 9.5. The most efficient was isolate 71T that removed 53% of the phenol in phenol red broth media. The pink bars in Figure 9.5 indicate a shift from neutral pH to acidic, which has been discussed in Section 7.4.4 as a possible presence of phosphatase.

9.3.3 Reduction of phenol in pollutants and WW samples

9.3.3.1 Reduction of phenol in coffee and soya sauce

Only three isolates 19S, 71T and 84Li significantly lessened phenol concentrations in coffee. These three as well as 4T, 7T, 76S, 77S and 83Li also decreased phenol in soya sauce (Figure 9.6). Isolate 84Li was the most active reducing phenol in coffee to 70% and in soya sauce to 46%.

149

Figure 9.6: Reduction of phenol in coffee and soya sauce.

9.3.3.2 Reduction of phenol in bromothymol blue, crystal violet and commercial dye

Isolates 19S and 84Li were selected to test whether they could reduce phenol in two textile dye derivatives, BB and CV. Isolate 2T was included as a negative control, while results were compared to when only sterile nutrient broth was added. No significant difference was observed between the control and the isolates tested (Appendix 18) although isolate 84Li reduced phenol in both BB and CV by between 30 and 40%.

* denotes significant reduction in phenol

Figure 9.7: Reduction of phenol in commercial dyestuff Dye It No. 18.

150

As indicated in Figure 9.7, ten isolates (4T, 7T, 16S, 19S, 71T, 76S, 77S, 78S, 83Li and 84Li) were positive for phenol reduction in commercial textile dye. The two best performers were isolates 19S and 84Li that reduced phenol in Dye It No. 18 by 55% and 74% respectively.

9.3.3.3 Reduction of phenol in brewery and dairy wastewaters and river water contaminated with industrial effluents

Figure 9.8: Reduction of phenol concentrations in brewery WW and dairy WW. No brewery WW assays were done for isolates 4T, 7T and 16S. (*) denotes significant reduction of phenol in brewery WW. (^) denotes significant reduction in phenol in dairy WW.

Isolates 2T, 19S, 71T, 83Li and 84Li were positive for phenol reduction in brewery WW (Figure 9.8). Seven isolates (4T, 7T, 16S, 19S, 71T, 83Li and 84Li) reduced the phenols in dairy WW (Figure 9.8). In the latter, isolates 19S and 84Li resulted in residual phenol levels of 47% and 27% respectively. No brewery WW assays were done for isolates 4T, 7T and 16S as equipment and funding prevented further generation of CFCS.

Six isolates were tested against river water contaminated with industrial effluents and only two, namely 19S and 84Li, were effective in reducing phenol as indicated by “*” (Figure 9.9) by magnitudes of 55-56%.

151

Figure 9.9: Reduction of phenol in river water contaminated with industrial effluents (green bar), paraffin oil PF (yellow bar) and petroleum PT (brown bar). (*) denotes isolates with the ability to reduce phenol in contaminated river water. (^) denotes isolates with the ability to reduce phenol in paraffin oil and petroleum

9.3.3.4 Reduction of phenol in paraffin oil and petroleum

The results for six isolates tested are shown in Figure 9.9. Four of the isolates (19S, 77S, 83Li, 84Li) were also able to reducing phenol in paraffin oil and petroleum as indicated by “^”. The reduction in phenol in paraffin oil and petroleum was similar with isolate 84Li removing 60% of the phenol present.

9.3.3.5 Effect of temperature on reduction of phenols

Since the isolates were originally isolated from hot springs, and several isolates could significantly reduce phenols, the effect of temperature on phenol reduction in river water contaminated with industrial effluents was tested with six isolates. The isolates were selected based on their previous performance in reducing phenols in this investigation and 2T, 9T and 19S were isolated at 55 °C, while 77T, 83Li and 84Li were isolated at 37 °C. No difference was observed in the effect of temperature on reduction of phenols in river water contaminated with industrial effluents (Figure 9.10).

152

Figure 9.10: Effect of temperature on reduction of phenol concentrations in river water contaminated with industrial effluents

9.3.4 Summary of isolate identity and pollutant reduction

Table 9.2: Summary of relevant isolates from hot springs that reduced phenol in simulated pollutants and wastewater samples

Sample Isolation GenBank match Isolate site T (Accession No.) Enzymes* Biophysical** Phenol reduction

Bacillus licheniformis 2T Tshipise 55 °C (HM631844.1) phenol red, brewery WW

Anoxybacillus rupiensis amylase, soya sauce, commercial dye, 4T Tshipise 55 °C (AM988775.1) protease biosurfactant dairy WW

Anoxybacillus sp. amylase, biosurfactant, soya sauce, commercial dye, 7T Tshipise 55 °C (KJ722458.1) protease biosorbent dairy WW

Brevibacillus sp biosurfactant, soya sauce, commercial dye, 16S Siloam 55 °C (LN681596.1) anti-biofilm dairy WW phenol red, coffee, soya sauce, commercial dye, brewery WW, dairy WW, amylase, river water polluted with Anoxybacillus sp. protease, multi- industrial effluents, paraffin, 19S Siloam 55 °C (KP211933.1) enzyme biosurfactant petroleum

phenol red, coffee, soya Bacillus sp. sauce, commercial dye,

71T Tshipise 37 °C (HQ267759.1) protease brewery WW, dairy WW

Bacillus mojavensis 76S Siloam 37 °C (KF05492.1) biosurfactant soya sauce, commercial dye phenol red, soya sauce, Bacillus commercial dye, river water methylotrophicus polluted with industrial 77S Siloam 37 °C (KP342210.1) protease biosurfactant effluents, paraffin,

153

petroleum

Bacillus sp.

78S Siloam 37 °C (KC813167.1) protease soya sauce, commercial dye phenol red, soya sauce, commercial dye, brewery Bacillus sp. WW, dairy WW, paraffin, 83Li Libertas 37 °C (KF533727.1) protease biosorbent petroleum phenol red, coffee, soya sauce, commercial dye, brewery WW, dairy WW, river water polluted with Brevibacillus sp. biosurfactant, industrial effluents, paraffin, 84Li Libertas 37 °C (LN681596.1) biosorbent petroleum *From Chapter 7 results **From Chapter 8 results

Hot-spring bacterial isolates were identified by 16S rDNA sequencing described in detail in Chapter 3. These bacteria were screened for production of enzymes (Chapter 7) and for biophysical characteristics (Chapter 8) that could be potentially useful for WW bioremediation. Table 9.2 lists the relevant isolates, their identification from this study and their properties relating to bioremediation. The simulated pollutants and different sources of wastewater sampled tested against the CFCS of the isolates are also mentioned.

9.4 DISCUSSION

9.4.1 Reduction in colouration or turbidity

Since the microorganisms are ultimately to be used in the bioremediation of the environment with the least amount of processing to reduce costs, this investigation was performed at 25 °C on extracellular crude extracts of CFCS for a period of 3 h against substrates that were potential pollutants and WW samples.

Melanoidin is a brown pigment found in many food products, drinks and fermented products, includes both coffee and soya sauce (Agarwal et al., 2010). The negative effects of melanoidin as water pollutants have been described in Section 2.5. Previous researchers have used OD at 400 nm (Miyagi et al., 2013) and 475 nm (Chandra et al., 2008) to measure melanoidin. A wavelength of 415 nm was used in this investigation as the filters were fixed and not variable, and found to be suitable as determined by the standard curve showing coffee/soya sauce concentration and OD at 415 nm in Appendix 12. Instant coffee was also used to simulate coffee WW (Novita, 2016). There was no significant reduction in colouration of coffee by CFCS of the isolates screened; however, isolates 76S (p = 0.0018) and 77S (p = 0.0004) reduced

154

colouration in soya sauce by 7% and 8%, respectively (Figure 9.1). Brewery WW that also has a brown colour due to melanoidin was not reduced in colour by these two isolates, 76S and 77S, but was reduced by at least 40% by isolates 35Li and 72T (Figure 9.3) although the latter was not supported statistically. Possible reasons for this difference may be the complex nature of brewery WW and an acidic pH of 4 (Section 3.1.2, Table 3.2) compared with the pH of 7 in the single component experiment.

The turbidity of the WW is likely not only a result of melanoidin but also other components of sugars, fats and proteins (Thiel & du Toit, 1965; Inyang et al., 2012). A list of different bacteria that are able to decolourise distillery WW is reviewed (Agarwal et al., 2010) and it includes Bacillus spp. Most of the isolates in this investigation were identified as Bacillus spp. or related (Chapter 4). Agarwal et al. (2010) reported Bacillus spp. that decolourised distillery WW, with success between 36-58% and it took a period of 20 d by growth in the substrate suggesting that the process could require more than 3 h to show a significant difference. Thermotolerant B. subtilis was able to decolourise industrial distillery WW by 85% (Tiwari et al., 2012). Other microorganisms including Gram-negative bacteria, cyanobacteria and fungi have been reported to decolourise melanoidin in distillery WW (Chandra et al., 2008; Kharayat, 2012).

The second group of pigmented pollutants includes BB and a purple commercial dye (Dye It No. 18). Textile dyes are classified into groups which include azo-dyes and triphenylmethane dyes (Ali, 2010). Bromothymol blue is a triphenylmethane dye (Section 2.5, Figure 2.2c), and a derivative of textile dyes. An OD of 595 nm was used in this investigation as indicated by the standard curve in Appendix 10 to measure BB levels. Kumar et al. (2014) used an OD of 590 nm for BB while Kochher & Kumar (2011) used 523 nm for another triphenylmethane dye, CV. None of the ten isolates had any effect on BB (Appendix 16) while isolates 22M (p = 0.0146) and 35Li (p = 0.0394) reduced the colouration of commercial textile dye by a mere but significant 5-6% (Figure 9.2). Al Masud et al. (2015) isolated Bacillus strains from textile effluent and found one that could decolourise 98% of Terasil Green after 48 h at 35 °C in neutral pH. Another Bacillus isolate decolourised azo-dye Metanil Yellow completely within 27 h (Anjaneya et al., 2016). Marine B. subtilis degrade 70% of textile dyes under optimal conditions (Maheswari & Sivagami, 2016). Anoxybacillus sp. from hot springs in Spain was tested against azo-dye, reactive Black 5 in batch culture with 80% decolourisation within 24 h (Deive et al., 2010). Thermophilic A. rupiensis from hot springs in India decolourised 75% of textile dye effluent at 60 °C in 8 days (Gursahani & Gupta, 2011). Brevibacillus sp. could degrade textile dyes over a period of 7 days at 28 °C (Durve et al., 2012) or 24 h at 25 °C

155

(Alhassani et al., 2007). It appears that decolourisation requires more time and post-screening optimisation of conditions is required.

The sampling site for river water contaminated with industrial effluents was selected based on previous sightings of different coloured dyes entering the water at regular intervals (Chapter 3, Figure 3.5). Samples were taken when the flow through the storm drain was at maximum output of white coloured pollutant (Chapter 3, Figure 3.6). The CFCS extracts of six isolates did not reduce the turbidity of river water contaminated with industrial effluents (Appendix 17); however, the source of the pollution remains unknown and speculative.

Dairy WW was used as a pollutant in this investigation and its negative effects on the environment have been reviewed (Shete & Shinkar, 2013; Tikariha & Sahu, 2013). Isolates 35Li (p = 0.0034) and 50T (p = 0.0101) both significantly reduced turbidity by 25% (Figure 9.3) which can be further investigated. Bacteria have been useful in bioremediation of dairy WW (Chatterjee et al., 2015; Sanjana et al., 2017) including Bacillus spp. isolated from dairy WW sludge (Garcha et al., 2016). Bacterial biodegradation appears to address the organic compounds (sugars, proteins and fats) often found in food processing WW (Chatterjee et al., 2015).

9.4.2 Screening of isolates that reduced phenol in phenol red broth media

Phenolic compounds are toxic and abundant in many types of industrial WWs, and some are carcinogenic (Bazrafshan et al., 2016). Consequently, screening for isolates that can degrade and reduce phenols is sought after because removal of phenol from water resources is both critical and essential, and the different methods have been reviewed (Kulkarni & Kaware, 2013; Divate & Hinge, 2014; Pradeep et al., 2015).

Five isolates in this study (2T, 19S, 77S, 83Li and 84Li) as indicated in Figure 9.5, significantly reduced phenol compared with the control using phenol in phenol red broth media as substrate (0.009 g/L final concentration). Phenol was reduced to 53% by 71T and 59% by isolates 19S and 84Li. These values are associated with extracellular bioactive molecules in culture supernatants and suggest they are good candidates for further investigation in batch cultures for optimisation of conditions. Mangrove sediment isolates Pseudomonas, Actinobacter and Bacillus sp. were reported to completely deplete phenol by growing the bacteria in the presence of 0.4g/L phenol for seven days (Kafilzadeh & Mokhtari, 2013). Shaking batch cultures of 0.5g/L phenol over a period at 25° C, identified a strain of Actinobacter sp. AQ5NOL that completely removed phenol over 4 days, and could also grow in the presence of 1.5g/L phenol (Ahmad et al., 2011). Paisio et al. (2014) studied phenolic biodegradation in effluents from a

156

chemical industry and a tannery also under shaking batch culture conditions also reporting a strain of Actinetobacter RTE1.4 completely removing 1g/L 2-methoxyphenol in 5 days. Alkaliphilic Pseudomonas stutzeri isolated from a lake in India, was described by Tambekar et al. (2012) that reduced 87% of phenol in phenol-containing batch cultures over a period of 4 day. In a similar experimental setup, Tambekar et al. (2015) later reported that Bacillus flexus could remove 74% of phenol within 96 h. As reported with the dye degradation, other investigations commonly describe batch cultures for longer than 24 h periods of time. It is impossible to compare the results of this investigation to the abovemented studies due to the differences in experimental setup. The shortfall of this study not to include whole and viable cell removal of phenol was due to limited funds, however it does describe a simple and cost effective method of screening CFCS fractions quickly.

The FC reagent is useful for measuring the concentration of phenols and other antioxidants (Everette et al., 2010) in aqueous solutions (Box, 1983), milk (Vázquez et al., 2015), olive oil WW (Atanassova et al., 2005) and brewery WW (Tatullo et al., 2016), and therefore used in this investigation for both pollutants and for WW samples. The use of phenol red broth media as a substrate is simple and novel and has not been reported previously to screen isolates for phenol reduction.

9.4.3 Reduction in phenol and phenolics in coffee, soya sauce and brewery wastewater

Coffee and WW of coffee production contain phenols (Coelho et al., 2014; Torres et al., 2016) and plant enzymes particularly peroxidases have been used to degrade coffee phenols (Torres et al., 2016) in WW. In the simulated coffee pollutant experiment, phenols were reduced to 78% (isolate 71T), 76% (isolate 19S) and 70% (isolate 84Li). In soya sauce, the reduction ranged from 54% (isolate 84Li) to 10% (isolate 16S) for the nine positive phenol reducers (Figure 9.6). To the author’s knowledge, there is no literature on the use of bacterial degradation of phenols in specifically coffee or soya sauce WW, although Rojas et al. (2003) reported using Bacillus sp. inoculated into fresh coffee pulp in order to improve its nutritive value by decreasing total phenols (regarded as an antinutritional factor). This may also be another application for these phenol degraders in coffee in food technology can be investigated at a later date.

Brewery effluent contains 0.018 mg/L phenol (Inyang et al., 2012), while wine distillery can contain between 0.029 and 0.474 mg/L phenol. Molasses and vinasse WW contain phenol levels of 0.45 and 0.477 mg/L, respectively (Kharayat, 2012). Santal & Singh (2013) reported the

157

complex composition of acidic coloured distillery WW with phenolic compound levels of 4.2 mg/L.

The two top phenol degraders in brewery WW was 19S (Anoxybacillus sp) and 71T (Bacillus sp.) both reduced phenol by 48% in crude extracts (Figure 9.8). The isolates that reduced colouration in brewery WW was 35Li and 72T (Figure 9.3), were different from the ones that reduced phenol (19S and 71T), suggesting that the colouration was not due to phenols and that 35Li and 72T reduced other components of pollution that contributed to water colour. Only one of five isolates of Lactobacillus planarum that decolourised melanoidins in palm oil mill effluent also removed phenols suggesting that the two processes are more complex and independent (Limkhuansuwan & Chaiprasert, 2010). Usually a consortium of microbes including anaerobic and aerobic bacteria are involved in distillery WW bioremediation with a predominance of Pseudomonas spp. (Chavan et al., 2006; Bodike & Thatikonda, 2014). The various bacterial species involved have been reviewed and listed and include Bacillus sp. found by other researchers (Kharayat, 2012; Ogunlaja et al., 2013), although these were not specifically related to phenol degradation.

9.4.4 Reduction of phenol and phenolics in BB, CV, commercial dye and river water contaminated with industrial effluents

No isolate could be identified that reduced phenol in BB and CV (Appendix 17). Nine isolates were efficient in lowering phenols in the commercial dyestuff Dye It No.18; however, the two best performers were 19S (reduction of 55%) and 84Li (reduction of 74%) (Figure 9.7). These isolates were different from the two that could reduce colouration (Figure 9.2) and therefore it is suggested that the phenol fraction is not the main contributor to colouration in the commercial dye. The two isolates 19S and 84Li that reduced phenol in the commercial dye were also the same isolates that were positive in reducing phenol by 54-56% in the river water sample suspected of being contaminated with a dye industry effluent (Figure 9.9). This finding is supportive of the idea that the white colouration was due to a dye but it requires verification. Most studies have been done on decolourisation of textile effluent; however, Ramachandran et al. (2013) reported phenolic compound concentrations of 0.077 mg/L in textile WW confirming their occurrence.

158

9.4.5 Reduction in phenol and phenolics in dairy WW

Milk processing WW contains soluble and semivolatile phenol compounds (Verheyen et al., 2009); these compounds have been measured in milk processing wastewaters from different species (goat, human, cow, sheep) (Vázquez et al., 2015). Seven isolates were positive for phenol reduction in dairy WW, and the most efficient was 19S (47% residual) and 84Li (27% residual) as indicated in Figure 9.8. These two isolates were also responsible for decreased phenol in the commercial dye sample and in the river water sample contaminated with industrial effluent confirming their properties.

9.4.6 Reduction in phenol and phenolics in paraffin oil and petroleum

Polycyclic aromatic hydrocarbons (PAHs) as environmental pollutants have been described in Section 2.5. As shown in Figure 9.9, six isolates were tested against PAHs: petroleum and paraffin oil. Again, the best performers were 84Li (decrease of 70%) and 19S (decrease of 40%), with 77T and 83Li also showing positive results for the reduction of phenol. It is not clear why isolates Anoxybacillus sp. 19S and Brevibacillus sp. 84Li were the most successful based on the screening assays. However, both exhibited biosurfactant properties (Table 9.2) and both genera have been reported to degrade phenols by other investigators. A thermophilic Anoxybacillus sp. HBB16 was reported to produce a lipase (Bakir & Metin, 2017). Brevibacillus sp P6 degraded phenol at concentrations <200mg/ L (Yang & Lee, 2007). Bacillus spp. degraded toluene (Abari et al., 2013), diesel oil (Basu et al., 2014), oily WW (Bujang et al., 2013) but not specifically phenol. Bacillus sp. degraded phenols in four different soil types (Djokic et al., 2013). More clear-cut evidence is the activation of genes involved in the degradation of PAHs in the presence of phenol by B. subtilis (Tam et al., 2006).

9.4.7 Effect of temperature on reduction of phenol in river water contaminated with industrial effluents

Since the optimum temperature for growth of a bacterial culture can be different from the optimum temperature of functional enzymes, the CFCS from thermophilic isolates 19S (isolated at 53 °C) was tested at lower temperatures of 25 °C and 37 °C. Similarly, the CFCS of mesophilic isolates (77T, 83Li and 84Li) isolated at 37 °C, were tested at a higher (53 °C) and lower (25 °C) temperature. This effect on temperature on phenol reduction in river water contaminated with industrial effluents was determined (Figure 9.10). From initial results in this investigation, isolate 2T and 9T were included as CFCS bacterial negative controls while 19S,

159

77T, 83Li and 84Li were previously shown to reduce phenol (Figure 9.5). However, 19S was the only thermophile of the four positive isolates. Phenol reduction at 53 °C was not observed to be higher compared with the reduction achieved at 25 °C and at 37 °C for isolate 19S. Furthermore, phenol reduction for isolates 77T, 83Li and 84Li was not higher at lower temperatures despite their isolation at 37 °C. In this study, temperature did not affect phenol reduction. Since Bacillus and Bacillus-related bacteria are heat tolerant and spore-forming, it was hypothesised that temperature could determine phenol degrading activity. In experiments where conditions were optimised, temperature was an important factor for this genus. For example, the optimal temperature for Anoxybacillus sp. to decolourise textile dyes was 60 °C (Gursahani & Gupta, 2011) or 65 °C (Deive et al., 2010); and decolourisation of distillery WW by B. subtilis was found to be optimal at 45 °C (Tiwari et al., 2012). For Bacillus sp. that decolourised azo dye Matanil Yellow, the optimal temperature was 40 °C (Anjaneya et al., 2016) or for textile dye Terasil Green, it was 35 °C (Al Masud et al., 2015). Brevibacillus sp. decolourised textile dye optimally at 28 °C and although commonly isolated in hot springs, it is generally thought to be mesophilic. However, in the abovementioned optimization experiments of other investigators, bacterial cells were grown in batch cultures at various temperatures rather than only extracellular CFCS tested. As a result, different results of phenol reduction might be revealed if these four isolates were grown in the river water contaminated with industrial effluents.

9.4.8 Identification of isolates with bioremediation potential

Three genera are represented in Table 9.2 as Anoxybacillus, Bacillus and Brevibacillus. All have been alluded to throughout the text of this chapter as contributing microorganisms to WW bioremediation both through decolourisation of WW or by removal of phenols (Aysha & Mumtaj, 2014). The top candidates are thermophilic Anoxybacillus sp. 19S and mesophilic Brevibacillus sp. 84Li with the ability to address several substrates (Table 9.2) and maximum reduction of phenol levels in the assays described above. Since WW samples are commonly a complex mixture of different pollutants, generally a consortium of bacteria is required for bioremediation (Kharayat 2012; Vijayalakshmidevi & Muthukumar, 2015; Poi et al., 2017).

9.4.9 Properties of isolates with bioremediation potential

Previous chapters describe the processes required for microbial bioremediation of water which includes enzyme production (Chapter 7) and biophysical properties, e.g. biosurfactant activity (Chapter 8). Of the eleven isolates important in the reduction of phenol and other pollutants in

160

WWs (Chapter 9, Table 9.2), eight were also earmarked with relevant biophysical characteristics for bioremediation (Chapter 8, Table 8.2), suggesting that biosurfactant activity is a major factor in bioremediation of WW. It is therefore not surprising that both Anoxybacillus sp. 19S and Brevibacillus sp. 84Li produced biosurfactants. Biosurfactants are critical in addressing PAHs and oily pollutants (Marti et al., 2014; Kavitha et al., 2014; Abbas et al., 2015; Xia et al., 2015). Microbial enzymes are also important in degrading dyes (Zhang et al., 2012b; Santos et al., 2014), melanoidins in WW (Gonzalez et al., 2008), coffee production WW (Torres et al., 2016), petroleum (Peixoto et al., 2011) and PAHs (Shekoohiyan et al., 2016). Karigar & Rao (2012) reviewed the important oxidoreductases and hydrolases in bioremediation. Isolate 19S produced both amylase and protease by conventional plate assay, and this was confirmed by LC-MS/MS. Using LC-MS//MS, two enzymes related to alpha-amylase were identified (trehalose-6-phosphate hydrolase and oligo-1,6-glucosidase) and three proteases (aminopeptidase, peptidase T and subtilisin). Other important enzymes identified included catalase, peroxidase, superoxide dismutase and NADH-azoreductase that play a role in decolourisation (Gonzalez et al., 2008) and in phenol removal (Kulkarni & Kaware, 2013; Pradeep et al., 2015). The enzymes of isolate 19S are discussed in detail in Chapter 7. Based on the previous findings that isolate 19S produces both enzymes and expresses the necessary biophysical characteristics useful for WW bioremediation, and the results that show that its crude CFCS extracts can reduce phenolic pollutants in WW, it is suggested that it is a strong candidate for further investigations relating to isolation and characterisation of enzymes, and optimisation of conditions for their activity.

Observations of the different substrates used by different isolates, e.g. isolates 76S and 78S, showed that phenol reduction occurred in only soya sauce and commercial dyes but not in WW samples. However, these activities of isolates 76S and 78S were not detected during the initial screening using phenol in phenol red broth media only, compared to the other isolates, and implies that the enzymes produced are different or that they require different conditions for optimal activity. Further investigation is required. More sophisticated techniques (HPLC) will also be able to discern the process of degradation and decolourisation. Bioremediation potential in association with bacterial growth will also reveal other mechanisms not yet described, e.g. bioassimilation or intracellular degradation requiring metabolic pathways.

161

CHAPTER TEN: ANTIMICROBIAL ACTIVITY OF BACILLUS MOJAVENSIS, ISOLATED FROM HOT SPRINGS, SOUTH AFRICA, AGAINST ENVIRONMENTAL MULTIPLE ANTIBIOTIC-RESISTANT BACTERIA AND HUMAN PATHOGENS

10.1 INTRODUCTION

The increase in AR of pathogens has become a threat to public health (Ventola, 2015) prompting a hunt for novel natural antimicrobial biomolecules. Natural antimicrobials are classified into three groups: large complex antibiotics, bacteriocins (30-60 amino acids) and anti-microbial peptides (3-20 amino acids) (Hassan et al., 2012; Güllüce et al., 2013) with different modes of action (Blair et al., 2015; Falanga et al., 2016).

Biosurfactants are amphipathic molecules sometimes with antimicrobial properties against human pathogens (Sarin et al., 2011; Sharma et al., 2014a). Based on their mode of action, it is therefore not surprising that biosurfactants can be antimicrobial, and that bacteriocins or bioactive peptides are biosurfactants (Harshada, 2014). Antimicrobial thermophiles from hot springs have been reported including Geobacillus spp. (Alkhalili et al., 2015) and Bacillus spp. (Pakpitcharoen et al., 2008). In the latter, the isolates also exhibited biosurfactant properties.

The most commonly studied antimicrobial substances of Bacillus are the bacteriocins (Sumi et al., 2015). They are most often used for food preservation (Baruzzi et al., 2011; Nath et al., 2015) and biological control (biocontrol) of plant pathogens (Nagórska et al., 2007). Subtilin is a class 1 bacteriocin or cationic, pentacyclic antimicrobial peptide inhibiting a broad range of Gram-positive bacteria including other species of Bacillus (Baruzzi et al., 2011). Subtilisins are extracellular alkaline serine proteases that are involved in the processing of subtilin (Siezen & Leunissen, 1997). Subtilisin (Wu et al., 2013b) and surfactin (Al-Araji et al., 2007) of Bacillus spp. are both antimicrobial and biosurfactants. Endophyte Bacillus mojavensis, a close relative of B. subtilis, also produces subtilisin (Haddar et al., 2009), and is useful in the biocontrol of phytopathogens (Ghanney et al., 2016) and control of mycotoxin fungi (Ma & Hu, 2015). Bacteriocins are also the “new age” antimicrobials against AR pathogenic bacteria (Hassan et al., 2012) and further examples are given in Section 2.6.3.

162

Bacillolysins are neutral metalloproteases of Bacillus sp. that are used for production of taste in food, cell disassociation and as bioinsecticides (Narasaki et al., 2005). Although not known to be antimicrobial, they are also biosurfactant in nature (Zhang et al., 2010).

Hot springs have the potential to harbour microbes that produce antimicrobial agents (Mahajan & Balchandran, 2017), including Bacillus spp. from Jordanian hot springs (Fandi et al., 2014; Alkhalili et al., 2016). In this study, bacterial isolates from hot springs were screened for biosurfactant activity, for antimicrobial activity against human pathogens and against a panel of multiple resistant environmental bacterial isolates from the same hot springs. Identification of the biosurfactant-producing bacteria and relevant biomolecules will be further elucidated.

10.2 METHODOLOGY

The geographical locations of the five hot springs selected as sampling sites are described in Section 3.1.1 together with the procedure followed in processing the water and sediment samples. Mesophilic and thermophilic aerobic bacteria were isolated from water and sediment samples (Section 3.2) to obtain pure cultures. Following the isolation of bacteria (Section 3.2), DNA was extracted from pure cultures (Section 3.4), and identified using 16S rDNA amplicon (Section 3.4) and phylogenetically analysed (Section 3.6). The pure culture of isolate 76S was confirmed by Inqaba Biotechnology using the 16S rDNA sequence, and phylogenetically analysed with an unrooted maximum-parsimony and bootstrapped with 100 replications.

The AR (Chapter 6) and identification (Chapter 4) of the environmental isolates that were used to test for antimicrobial activity has been described. Human pathogens were laboratory stocks of S. aureus, B. cereus, P. aeruginosa, E. coli, E. faecalis, K. pneumonia, and M. smegmatis, as listed in Section 3.18.2. Environmental bacterial strains Kocuria sp., B. subtilis, and B. licheniformis were isolated during this study. The abovementioned bacteria have been reported as opportunistic pathogens.

Biosurfactant activity was determined by an emulsion assay using paraffin oil explained in Section 3.16.1 and a drop-collapse assay using Parafilm M (Section 3.16.2). Details of the T- streak method for preliminary antimicrobial screening are given in Section 3.18.1. Cell-free culture supernatants of the hot-spring isolates were clarified by centrifugation to remove bacteria and tested by disk diffusion against human pathogens and the panel of environmental bacteria as described in Section 3.18.2. Tandem LC-MS used to identify proteins in CFCS of isolate 76S has been described in Section 3.19.

163

10.3 RESULTS

10.3.1 Biosurfactant activity: emulsion activity assay and drop-collapse assay using Parafilm M

Table 10.1: Biosurfactant activity of cell-free culture supernatants of five thermophilic and four mesophilic isolates determined by emulsion assay and drop-collapse assay

Temp isolation Emulsion activity* Identification °C (%) Drop collapse 4T Anoxybacillus 53 + (115) - 7T Anoxybacillus 53 + (72) - 19S Anoxybacillus 53 + (92) - 30M Bacillus 53 + (11) - 50T Bacillus 53 + (31) + 57T Kocuria sp 37 + (20) - 73T Aneurinibacillus 37 + (11) - 76S Bacillus 37 + (22) + 77S Bacillus 37 + (25) - *Expressed as percentage of positive control (1% sodium lauryl sulphate)

Positive Negative 4T

Figure 10.1: Emulsion activity of Anoxybacillus 4T against paraffin oil (tube 3), with positive (sodium lauryl sulphate) and negative (nutrient broth only) controls

164

Nine isolates (five thermophiles isolated at 53 °C and four mesophiles isolated at 37 °C) were screened for biosurfactant activity using two methods. Figure 10.1 shows the emulsion activity using paraffin oil and positive control of 1% SLS, a negative control of water only and a test assay with CFCS of isolate Anoxybacillus sp. 4T. Results were expressed as a percentage of the positive control. Emulsion assays were done in triplicate for each CFCS. The crude extracts of Anoxybacillus spp. 4T, 7T and 19S were the best biosurfactant producers; however, they were not positive in the drop-collapse assay. The lowest producers 30M and 73T by the emulsion assay were also negative in the drop-collapse assay. Two isolates, namely 50T and 76S were positive in both assays (Table 10.1). The determination of positive in the drop-collapse assay is described in Chapter 8 (Figure 8.7A) as a reduction in surface tension and change in the shape of the drop on Parafilm M. Discrepancies in methods of biosurfactant assays have been reported (Satpute et al., 2008).

10.3.2 Antimicrobial activity

10.3.2.1 T-streak technique to identify isolates with antimicrobial activity

Figure 10.2: Agar plate showing T-streak method where isolate 54T, 77S and 85Li were inoculated perpendicular to isolate 73T. Growth inhibition of isolate 77S against isolate 73T was observed

The T-streak technique was used as a preliminary screen to observe the interaction between isolates (Figure 10.2) and the results for all the isolates screened are not shown. Limitations of

165

this method were that thermophiles could not be tested against mesophiles, and some results were inaccurate for motile bacteria, and therefore not included in the study.

10.3.2.2 Antimicrobial activity testing of crude CFCS extracts against the panel of environmental isolates and human pathogens

Table 10.2: Crude CFCS extracts of hot-spring isolates tested for antimicrobial activity against panel of environmental isolates and human pathogens. Zones of inhibition are in millimetres (mm)

Cell-free culture

supernatant 4T 7T 19S 30M 50T 57T 73S 76S 77S Environmental isolates

/ test organisms

1T - - - nd - - - + (2mm) - + 2T B. licheniformis - - - + * - - (1mm) + (3mm) - + 3T - - - + * (5mm) - - + (2mm) - + 6T B. licheniformis - - - + * (1mm) - nd + (1mm) nd + 7T - - - + * (4mm) - - + (3mm) + (1mm)

9T ------nd + (2mm) -

16S ------

21M B. subtilis ------nd + (1mm) nd

30M B. licheniformis - - - nd - - nd - nd + 45T (1mm) - - + * - - nd + (1mm) nd + 46S - - - - (1mm) - nd + (2mm) nd

51T ------nd + (2mm) nd

54T - - - nd - - - - - + 57T Kocuria ------(2mm) + (2mm) + (1mm) + 58T nd nd nd - - nd (2mm) - - + + 72T nd nd nd - (1mm) nd (2mm) + (3mm) + (1mm)

75S - - - nd - - - + (4mm) -

Mycobacterium smegmatis nd nd nd - - nd - - -

Klebsiella pneumonia nd nd nd - - nd - - -

Enterococcus faecalis nd nd nd - - nd - - - + Bacillus cereus nd nd nd + - nd (2mm) + (3mm) -

Escherichia coli nd nd nd - - nd - - - + Staphylococcus aureus nd nd nd - - nd (2mm) + (3mm) - * positive results from T-streak assay; nd – not done

166

From preliminary antimicrobial T-streak results, together with the biosurfactant activities of isolates indicated in Table 10.1, isolates listed in Table 10.2 were selected to be investigated further with regard to antimicrobial activity. The CFCS extracts of the isolates 4T, 7T, 19S, 30M, 50T, 57T, 73S, 76S and 77S were placed on sterile filter disks and placed on a lawn of bacteria as indicated in column 1 of Table 10.2. Antimicrobial activity was recorded as positive if there was a zone of inhibition of growth of bacteria as shown in Figure 10.3.

Figure 10.3: Agar disk diffusion assay showing inhibition of crude CFCS extracts from isolate 76S (Bacillus mojavensis) inhibiting the growth of Anoxybacillus sp. 7T

Seventeen hot-spring isolates were used as environmental test organisms (Table 10.2 Column 1). They included opportunistic pathogens Kocuria sp. (57T), B. subtilis (21M) and B. licheniformis (2T, 6T). Other test organisms were M. smegmatis, Gram-positive B. cereus and S. aureus, and Gram-negative K. pneumonia, E. coli and E. faecalis resulting in a total of 26 test organisms. Of the panel, three bacteria were Gram-negative (E. coli, E. faecalis, K. pneumonia) and one was acid-fast (Mycobacterium) while the majority were Gram-positive. Crude CFCS extracts of Anoxybacillus sp. 4T, 7T and 19S and Kocuria sp. (57T) were not active against most of the bacteria. The CFCS extract of isolate 77S only inhibited three test bacteria, while supernatants of isolates 30M and 73S inhibited six test organisms, and 50T inhibited five test organisms. The CFCS extract of isolate 76S was the most successful in inhibiting 15 of the 23 test microorganisms (65%). Of the 17 isolates on the environmental panel, inhibition activity of the CFCS extract of 76S was observed in 13 (76%).

167

10.3.3 Antibiotic susceptibility testing Table 10.3: Antibiotic resistance profiles of panel of environmental isolates used to test antimicrobial activity of CFCS of isolate 76S

Antimicrobial β- Tetrac Sulphona activity of 76S lactam Amino glycosides yclines Amphenicols mides Quinolones

Environ mental (inhibition in isolates mm) Genera CAR* GEN* KAN* STRP* TET* CEF* CHL* COT* NA* NOR*

3T p (2) Anoxybacillus sp. S S S S S R S S R S

7T p (3) Anoxybacillus sp. S S S S S R S S R S

1T p (2) Bacillus sp. S S S 5 6 R S S R S

9T p (2) Bacillus sp. S S S S S R S S R S

46S p (2) Bacillus sp. R S S S S S S S R S

51T p (2) Bacillus sp. R S S S S R S S S S

72T p (2) Bacillus sp. S S S S S R R S S S

75S p (4) Bacillus sp. S S S S S R R S S S

Bacillus 2T p (3) licheniformis R S S S S R S S R S

Bacillus 6T p (1) licheniformis S S S S S R S S R S

Bacillus 30M n licheniformis S S S S S R S S R S

20S n Bacillus subtilis R S S R S R S S R S

21T p (1) Bacillus subtilis S S S R S R S S S S

54T n Bacillus subtilis R S S S S R S S S S

16S n Brevibacillus S S R S S R S S S S

45T p (1) nd R S S S S R S S R S

58T n Actinobacteria R S S S S S S S R S

57T n Kocuria sp. S S S S S R S S R S

* Antibiotics tested are β-lactam (carbenicillin CAR), aminoglycosides (gentamicin GEN, kanamycin KAN, streptomycin STRP), tetracycline TET, amphenicols (chloramphenicol CHL, ceftriaxone CEF), sulphonamides (co-trimoxazole COT) and quinolones (nalidixic acid NAc, norfloxacin NOR)

Seventeen hot-spring isolates were used as the panel of environmental test organisms to test the antimicrobial activity of isolate 76S (Table 10.2). Table 10.3 shows the antibiotic resistance profiles towards 10 antibiotics of these environmental bacteria. These results have been presented and discussed in detail in Chapter 6. Antibiotic resistance was determined using a disk diffusion assay. An example of isolate 54T is shown in Figure 10.4 with a resistance to CAR. This panel was selected as two isolates (2T and 45T) were resistant to three antibiotics while the rest were resistant to two. The CFCS of isolate 76S inhibited 76% (13 of 17) of the

168

environmental bacteria that expressed antibiotic resistance. Multiple antibiotic resistance is defined as resistance towards ≥3 antibiotics (Magiorakos et al., 2012), and the two isolates 2T and 45T were both inhibited by the extracellular fraction of isolate 76S.

Figure 10.4: Antibiotic resistance determined by disk diffusion assay showing resistance to carbenicillin

10.3.4 Screening of proteins in CFCS of isolate 76S by LC-MS/MS

Following separation of proteins by SDS-PAGE, the bands were extracted and processed by LC-MS/MS (Chapter 7, Figure 7.5). The peptides were compared to published databases for Bacillus spp. (NCBI and UniProt) for identification. Two biosurfactants were identified as subtilisin and bacillolysin with a low molecular weight of between 6 and 16 kDa based on the SDS-PAGE analysis. The characteristics of subtilisin (alkaline serine protease) and bacillolysin (neutral metalloprotease) are listed in Table 10.4.

Table 10.4: Characteristics of two biosurfactant proteases produced by B. mojavensis 76S identified by LC-MS/MS

Subtilisin Bacillolysin

Closest match Subtilisin NAT OS = Bacillus subtilis P35835 Bacillolysin OS = Bacillus subtilis P68736

Bacillolysin OS = Bacillus subtilis subsp. Subtilisin E OS = Bacillus subtilis P04189 amylosacchariticus P38735 Neutral protease NprE OS = Bacillus Subtilisin BM1 = Bacillus mojavensis AGL7445 pumilus P68734 Bacillolysin OS = Bacillus amyloliquefaciens P06832

Activity serine-type endopeptidase activity metalloendopeptidase activity Cellular location secreted, extracellular secreted, extracellular

169

Binding metal ion binding metal ion binding Gene aprN & apr E npr E Belongs to the peptidase S8 family; Contains 1 Family peptidase S8 domain Belongs to the peptidase M4 family

10.3.5 Identification of isolate 76S using 16S rDNA sequence

The antibiotic-resistant environmental isolates were sequenced and identified as described in Chapter 4 and listed in Table 10.3.

Figure 10.5: Colony morphology of Bacillus mojavensis isolate 76S grown on nutrient agar at 37 °C

Isolate 76S was isolated from a water sample from Siloam hot springs on NA and a pure culture is shown in Figure 10.4. By sequencing, it was identified as Gram-positive endospore forming Bacillus mojavensis, a close relative of B. subtilis. A contiguous sequence of a partial 16S rDNA is listed in the Appendix 20, and was processed by BLAST (GenBank). GenBank results showed a 99% match with Bacillus axarquiensis CIP108772 (DQ993670.1), B. subtilis BI1 (FJ947049.1), B. mojavensis IARINIAW2-23 (KF05492.1) and B. mojavensis W1-2 (KC503924.1).

Sequences were aligned with Clustal Omega (www.ebi.ac.uk). Phylogenetic analysis was performed with a 927-base pair fragment using SeaView (Gouy et al., 2010) generating a maximum-parsimony tree analysis. The unrooted tree was evaluated by bootstrap with 100 replications. A table of the reference strains used for phylogenetic analysis is listed in Appendix 19. The identification of the isolate as B. mojavensis was confirmed by the LC-MS/MS result in a peptide BLAST search that matched its subtilisin 100% with subtilisin BM1 of B. mojavensis (Accession No AGL76445 Haddar A, unpublished).

170

Figure 10.6: Phylogenetic analysis using parsimony and 100 replicate bootstrapping placing isolate 76S as Bacillus mojavensis

10.4 DISCUSSION

10.4.1 Biosurfactant production

Biosurfactant activity was determined with an emulsion assay with paraffin oil as substrate as indicated in Figure 10.1 and described in Section 8.3.4.1 and 8.3.4.2. Five thermophiles and four mesophiles were tested for biosurfactant activity, including the genera Anoxybacillus (4T,

171

7T, 19S), Bacillus (30M, 76S, 77S), Kocuria (57T) and Aneurinibacillus (73T) (Table 15). The biosurfactant of Anoxybacillus spp. 4T, 7T and 19S appears to be different to the other isolates because they produce high emulsion activities (72-115%) but were negative in the drop-collapse assay. Other Bacillus sp. had lower emulsion activities of between 11 and 31%, and variable correlation with the drop-collapse assay. Only two isolates, 50T and 76S were positive for both assays. Satpute et al. (2008) compared eight different methods used to test for biosurfactants of 45 strains of bacteria. They recommend the drop collapse, tilted glass slide test and oil spread method as the first choice for preliminary screening followed by the emulsion assay, and concluded that no one method was good enough to detect all the producers because there are different classes of biomolecules responsible. In this investigation only two positives were detected by drop collapse out of the nine positives by emulsion assay supporting the possibility of different types of biosurfactants or different concentrations being present. The two methods used were selected on resources available.

Microbial biosurfactants are a large diverse group of molecules classified into five groups: Group A glycolipids; Group B lipopeptides; Group C fatty acids, phospholipids and neutral lipids; Group D polymeric biosurfactants; and Group E particulate biosurfactants. Bacillus biosurfactants fall into Group B, including subtilisin and surfactin from B. subtilis, and lichenysin from B. licheniformis. Bacillus subtilis can also produce glycolipid biosurfactants (Reis et al., 2013). Newer biosurfactants, such as iturin and fengycin, have been described by Plaza et al. (2015).

The biosurfactant activity of hot-spring isolates in this study has been discussed in detail in Chapter 8.

10.4.2 Screening for antimicrobial activity

Since biosurfactants can also have antimicrobial activity (Sarin et al., 2011; Sharma et al., 2014a), the isolates that were positive for biosurfactant production in this investigation were screened for antimicrobial activity. A comparison was made between the results of this study and those of a similar study (Pakpitcharoen et al., 2008) of 148 isolates from hot springs in Thailand. In that study, Anoxybacillus produced a biosurfactant and was inhibitory to E. coli. In this study, no inhibition of E. coli was observed. In the Thailand study, Aneurinibacillus was negative for both biosurfactant and antimicrobial activity. In this study, Aneurinibacillus isolate 73T produced biosurfactant and had a broad spectrum of antimicrobial activity against Gram- positive test cultures. In the Thailand study, variable results were reported where biosurfactant positives were both antimicrobial or not, but of the biosurfactant negatives, only one of 148

172

showed antimicrobial activity, suggesting that a majority of biosurfactants are responsible for antimicrobial properties, but fewer antimicrobial agents are non-biosurfactant within this group of bacteria. In this study (Table 10.2), the CFCS of 7T, 19S and 57T produced biosurfactant but had no antimicrobial activity against the panel of test isolates. Isolates 30M, 50T, 73T, 76S and 77S were positive for biosurfactant but were variable in effectiveness against the panel of test organisms varying from 5/23 to 15/23 as similarly described by Pakpitcharoen et al. (2008). This suggests the presence of heterogeneous biomolecules responsible for reducing surface tension, resulting in variable effectiveness of antimicrobial activity, but further investigations are required.

As the discovery of new antimicrobials mostly relates to controlling human pathogens or unwanted fungal or bacterial growth, especially in food spoilage, these groups have been targeted in most studies (Baruzzi et al., 2011; Güllüce et al., 2013; Cheung et al., 2014). Both Sarin et al. (2011) and Sharma et al. (2014a) reported effective antimicrobial properties of biosurfactants of Bacillus spp. against reference human pathogens. In this study, pathogenic and opportunistic pathogens were used as test cultures. Kocuria sp. (Purty et al., 2013), M. smegmatis (Wallace et al., 1998), B. licheniformis and B. subtilis (Sietske & Diderichsen, 1991) can cause infections in immunocompromised patients and hospitalised patients (Chapter 5). E. coli and E. faecalis cause enteric disease, K. pneumonia, respiratory infections, and S. aureus, septicaemia and wound infections.

In this study, crude extracts of isolate 50T inhibited one of two B. licheniformis isolates. The CFCS of isolate 77S only repressed Kocuria sp., while B. licheniformis and B. cereus, but not B. subtilis, were inhibited by the CFCS of isolate 30M. Isolate 73T crude extract exhibited a broader spectrum of inhibiting Gram-positive bacteria: B. licheniformis, B. cereus, Kocuria sp. and S. aureus. However, 76S was the most effective with antimicrobial activity against a broad range of all Gram-positive bacteria, but not Gram-negative and acid-fast bacteria (Mycobacterium). The results were variable and inconsistent as indicated in Table 10.2. Fandi et al. (2014) reported on ten bacterial isolates from hot springs of Jordan, and showed antimicrobial activity against Gram-positive S. aureus and B. subtilis in half of them. However, without testing against Gram-negative bacteria in that study, some degree of specificity cannot be deduced.

It has been documented that the antimicrobial activity of antimicrobial peptides (AMPs) can vary tremendously, unlike antibiotics that are very specific (Russell, 2003). The bactericidal and bacteriostatic nature of AMPs is a result of pores created in the cell membrane which results in cell lysis. Cell-wall targeting is a function of the physiological fitness of the cells and is

173

influenced by stressful environmental factors (Farkas et al., 2017). This would explain the variability in AMPs or the effectiveness of bacteriocins against different test cultures even of the same species.

Where there is a more clear-cut distinction between susceptible groups, e.g. CFCS of isolate 77S exclusively inhibited the Gram-positive phylum Actinobacteria genus Kocuria, and secondly CFCS of isolate 76S only inhibited all Gram-positive bacteria in the panel but not Gram-negative bacteria, suggests that further investigation could reveal a mode of action that involves differences in the cell-wall structure. Lambert (2002) describes the differences in permeability of cell walls of Gram-positive, Gram-negative and acid-fast bacteria. The cell wall of Gram-positive bacteria have a thick peptidoglycan layer that is permeable to most antimicrobials, while Gram-negative and acid-fast (mycobacterium) cell walls are more complex with a protective high molecular lipid layer that overlays a thin peptidoglycan layer (Falanga et al., 2016).

Isolate 76S matched 99% to sequences in Genbank: Bacillus axarquiensis CIP108772 (DQ993670.1), B. subtilis BI1 (FJ947049.1), B. mojavensis IARINIAW2-23 (KF05492.1) and B. mojavensis W1-2 (KC503924.1). Bacillus mojavensis was shown to be different from B. subtilis (Roberts et al., 1994) and the type strain was isolated from hot desert soil. This was confirmed by the phylogenetic analysis in Figure 10.6 where B. subtilis and B. mojavensis including isolate 76S clustered separately. Bacillus axarquiensis was reclassified as B. mojavensis (Wang et al., 2007). Bacillus mojavensis is better known as an endophyte, a producer of surfactins and as an agent for biocontrol (Bacon & Hinton, 2011; Bacon et al., 2012). Bacillus mojavensis is not a common isolate from hot-spring environments; however, it is common to find mesophilic bacterial spores in hot-spring environments due to their heat tolerance and present as viable but not in a growing state (Khiyami et al., 2012). It has also been found in sediment (Liu et al., 2015b), oysters (Ma & Hu, 2015) and processed foods (Moe et al., 2015) confirming their ability to tolerate heat. Bacon et al. (2012) reported that B. mojavensis strains were able to produce a combination of several surfactins of different sizes at the same time. This species has a wide range of action including antibacterial (Liu et al., 2015b; Moe et al., 2015; Jasim et al., 2016), antifungal (Bacon & Hinton, 2011; Ma et al., 2012; Youcef-Ali et al., 2014) and antidermatophytic (Galgóczy et al., 2014).

More specifically, B. mojavensis inhibits mycotoxin-producing Fusarium (Bacon & Hinton, 2011; Ma et al., 2012), Vibrio, E. coli and S. aureus (Liu et al., 2015b). In Liu’s study (2015b), B. mojavensis did not inhibit B. subtilis, Listeria, Aeromonas, Pseudomonas and Salmonella typhi, but there have been no reports of activity exclusively against Gram-positive bacteria as a

174

group, or applications for biocontrol of Gram-positive pathogens. Since subtilin commonly inhibits the same or closely related species, the finding of bacteriocin subtilisin BM1, supports biocontrol against other Bacillus spp. belonging to the the phylum Firmicutes in this study. Staphylococcus aureus also belongs to the phylum Firmicutes; however, Gram-positive Kocuria belongs to the phylum Actinobacteria. Since both subtilisin BM1 and bacillolysin were identified in the culture media, it is possible that other biosurfactants remain unidentified, and that AMPs are also able to work in combination. Dean and Ward (1991) showed that both an alkaline and a neutral protease produced by Bacillus was able to lyse Gram-negative E. coli, but a mutant deficient in both protease was not able to, suggesting that several protease mechanisms may be working concurrently or in combination.

Bacillolysin (neutral metalloproteases) identified in the supernatant of B. mojavensis, was earmarked as a possible antimicrobial agent due to its biosurfactant properties (Zhang et al., 2010); however, this is still to be elucidated. No reports have described bacillolysin to be antimicrobial. They have been described to function as pathogenic factors in the human pathogen B. anthracis (Chung et al., 2006) and insect pathogen, B. thuringiensis (Altincicek et al., 2007), and antimicrobial properties would be an interesting investigation to follow.

Since the overuse of antibiotics has resulted in an increase of AR globally (Al-Bahry et al., 2014), one of the applications of AMPs is the biocontrol of such AR pathogens. Several studies have shown activity against methicillin-resistant S. aureus (MRSA) (Kaewklom & Aupad, 2012; Sambanthamoorthy et al., 2014). In a study of B. subtilis biosurfactants, Fernandes et al. (2007) showed that all AR pathogens of E. coli, E. faecalis, S. aureus and Pseudomonas (29 isolates) were sensitive and killed by the Bacillus biosurfactant. This finding supports the notion that AMPs with biosurfactant properties could be a new genre to eliminate AR pathogens and possibly replace antibiotic therapy. Although there are numerous studies of these biosurfactants or AMPs inhibiting AR human pathogens, there is no literature on the effect of AMPs on environmental MAR isolates. This environmental microbial population, although not directly involved in causing human infections, remains important because they are a source of ARGs that can be transferred to human pathogens in water bodies and biofilms, and may themselves cause infections in immunocompromised populations. Table 10.3 shows that 17 environmental isolates from hot springs in this study were resistant to two or three antibiotics (mainly resistance to CEF and NAc). The CFCS extract of B. mojavensis 76S was able to inhibit 76% of these AR bacteria suggesting there may be an application in addressing environmental AR which is now regarded as a new water pollutant (Martinez, 2009b; Sanderson et al., 2015).

175

CHAPTER ELEVEN: CONCLUSIONS

11.1 Isolation and identification of Bacillus and closely related Bacillus bacteria from hot springs in Limpopo Province, South Africa

This is the first report describing the bacterial diversity in hot springs in Limpopo, SA as determined by both culture-based and molecular approaches. Forty-three isolates were cultured clustered into four genera: Anoxybacillus, Bacillus, Brevibacillus and Aneurinibacillus. The study confirms the presence of A. flavithermus, A. rupiensis, B. subtilis, B. licheniformis and Brevibacillus spp. Singular Bacillus species that are phylogenetically related to B. panaciterrae, B. pumilus and B. methylotrophicus were identified; however, these three isolates require further characterisation. All, except for B. panaciterrae have been previously isolated from hot-spring environments. In addition to a comparison of 16S rDNA sequences with those in public databases, other tools were applied (GC content, ARDRA and phylogenetic analysis). A combination of these analysis techniques proved to be very effective in detecting outliers. Using ARDRA, the single isolate of Aneurinibacillus grouped separately from Aneurinibacillus type isolate, and several other Bacillus isolates from this study also fell into miscellaneous clusters B and E. The results of this study therefore show that using one molecular tool may result in a misrepresentation of Bacillus and Bacillus-related identification, and that where possible a combination of tools should be used. The contribution of this study is that there is some merit in combining different molecular tools to discriminate between different isolates. One tool does not necessarily reflect the absolute identification, and the complications regarding the classification of the genus of Bacillus need to be taken into consideration.

This study confirms that only a small portion of the microbial diversity present can be cultured, compared with the more comprehensive assessment of microbial diversity obtained using the metagenomic approach. Three different types of extremophiles with different properties (alkaliphilic, thermophilic and halophilic) were isolated suggesting that hot-spring water is a source of potentially important bacteria that can be used in biotechnology. Local hot springs are important bioresources of “novel” microorganisms as well as “established” microorganisms with novel bioactive molecules, thus supporting the need to study, protect and maintain such indigenous and pristine sites.

176

11.2 Cultured emerging opportunistic pathogens (phyla Actinobacteria and Proteobacteria) identified in hot-spring water through isolation and phylogenetic analysis

Although metagenomics studies have reported the presence of the phyla Actinobacteria and Proteobacteria from hot springs in Limpopo Province, this is the first report of cultured isolates from this particular ecological niche. This study confirms the presence of the phylum Actinobacteria, genera Kocuria and Arthrobacter, and the phylum Proteobacteria, represented by the following three classes: alpha-Proteobacteria (Sphingomonas), beta-Proteobacteria (Zoogloea, Gulbenkiania, Cupriavidus and Tepidimonas) and gamma-Proteobacteria (Ralstonia, Silanimonas, Hafnia and Cronobacter). These are commonly found environmental bacteria, and a review of the literature confirms that they have the capacity to cause infections in hosts that are susceptible due to a weakened immune system or where the integumentary barrier has been disrupted. Although Legionella is considered a re-emerging pathogen, it was not detected in the waters of these hot springs. An outcome of this study is to raise awareness about these hot springs being potential reservoirs for emerging opportunistic pathogens, as spring waters form part of the groundwater commonly utilised by humans for drinking and domestic purposes. The results are relevant for public health officials to gain an understanding of the aetiology of the infections caused by emerging opportunistic pathogens that are less likely to be reported, and their role in nosocomial outbreaks and in immunocompromised patients. As such, they can put in place mechanisms to protect populations from infections related to pathogens in hot springs.

11.3 Antibiotic and heavy-metal tolerance in cultured Bacillus species and opportunistic pathogens (Kocuria species and Hafnia alvei) from hot springs

Because AR is a growing health problem globally, the determination of baseline levels in the environment is important but unclear, due to the widespread distribution of AR bacteria, even in marine waters and Antarctic locations. Forty bacterial isolates from hot springs in Limpopo Province, SA, were resistant to three antibiotics and up to eight heavy metals. Resistance to CAR, a β-lactam antibiotic, was not surprising since it is ancient, ubiquitous and occurs naturally. However, intermediate resistance to NAc and CEF was a new finding. The results revealed that levels of AR in hot springs were generally low, consistent with other “pristine” environments, and compared with environments involving human activity such as WWTPs where AR levels are very high. The MAR index ratios can give an indication of water quality and pollution levels. In this study the overall MAR ratio >0.2 was 2.5% with only a single

177

isolate being resistant to three antibiotics. A value of 61% was reported for MAR in pristine mountain rivers (Lima-Bittencourt et al., 2007). For South African rivers which are exposed to human activities, MAR was reported to be between 70% (Kinge et al., 2010), 72 % (Lin et al., 2004) and 78% (Samie et al., 2012). At WWTPs, MAR increases up to 98% (Li et al., 2009a). All the values were found to be way above 2.5% in this study. Two opportunistic pathogens, Kocuria sp. and H. alvei were both resistant to two antibiotics and to at least two metals. This has implications in terms of potential health risks for the rural populations that use this water for drinking and domestic purposes. More data from “pristine and antibiotic-free (clean)” environments need to be collected and collated to support the notion that MAR levels are bioindicators of water quality, to indicate the location of AR genes that will enable differentiation between intrinsic AR and acquired AR, and to highlight the significance of AR in emerging opportunistic waterborne pathogens. A bigger sample size could address the knowledge gap identified in this environmental niche that has not yet been well studied with regard to AR.

The results of this study emphasise the importance and need to investigate and protect “clean” environmental aquatic sites to fully reveal the global problem of AR. Hot-spring environments are untapped resources that can be used to provide baseline levels of AR in real time.

11.4 Screening of potential bioremediation enzymes from hot-spring bacteria using conventional plate assays and liquid chromatography-tandem mass spectrophotometry (LC- MS/MS)

Thermophilic bacteria including Bacillus species are an important source of novel enzymes for water bioremediation. In this study, 56 bacterial isolates which were cultured from five hot springs in SA were identified predominantly as Bacillus sp. or Bacillus-related spp. by 16S rDNA sequencing. These isolates were screened for potentially useful enzymes for water bioremediation. Using conventional agar plate assays, 56% (n=43), 68% (n=38) and 16% (n=31) were positive for amylase, protease and BB decolorisation respectively. In liquid starch culture, three amylase-positive isolates differentially degraded starch by 34% (isolate 20S) to 98% (isolate 9T). Phenol degradation revealed that five out of thirty reduced phenol up to 42% by colorimetric assay. A thermophilic strain of A. rupiensis 19S (optimal growth temperature of 50°C), which degraded starch, protein and phenol, was selected for further analysis by LC- MS/MS. This newer technique identified potential enzymes for water bioremediation relating to pollutants from food industry (amylase, proteases), polyaromatic hydrocarbons and dye pollutants (catalase peroxidase, superoxide dismutase, azoreductase, quinone oxidoreductase),

178

antibiotic residues (ribonucleases), solubilisation of phosphates (inorganic pyrophosphatase) and reduction of chromate and lead. In addition, potential enzymes for biomonitoring of environmental pollutants were also identified e.g. dehydrogenases were found to decrease as the level of inorganic heavy metals and petroleum increased in soil samples. This study suggests that bacteria found in South African hot springs are a potential source of novel enzymes with LC-MS/MS revealing substantially more information compared with conventional assays, which can be used for various applications of water bioremediation.

Further characterisation is required to elucidate the biochemical and physical properties of the enzymes, and to determine the exact site of action on various substrates. The ultimate test is to determine their ability to reduce pollutants in aqueous conditions, and to determine their effect on WW various industries in the natural environment. Both conventional plate assay and sophisticated LC-MS/MS have their advantages depending on financial resources and available skills; however, LC-MS/MS provides greater and more accurate information and identification than plate assays.

11.5 Bacterial isolates cultured from hot springs possess biophysical characteristics (bioflocculant, biosurfactant, biosorption and anti-biofilm activities) useful in water bioremediation

Hot springs are an excellent resource of Bacillus and Bacillus-related bacteria with a diverse range of biophysical characteristics that can be applied to water bioremediation. Pollutants from several industries, such as pharmaceutical, food, petroleum and agriculture, can be addressed with bioremediation and clarification or degradation by bacteria. As proof of concept, several hot-spring isolates were found to have such characteristics. Bioassimilation of triphenylmethane dye (BB) by B. subtilis (9T) was observed. Another B. subtilis isolate 20S could reduce turbidity generated by kaolin clay by 30% using bioflocculants. Four additional isolates including 9T could remove heavy metals Cr, and Cu but not Fe and Ni from aqueous solutions. The majority of isolates were able to emulsify paraffin oil indicating they possessed biosurfactant properties. Using additional substrates of petroleum and sunflower oil, and a drop- collapse assay, four Bacillus spp. (76S, 85Li, 71T, 16S) were positive in all biosurfactant assays performed. Finally, a single isolate of Brevibacillus sp. (16S) showed anti-biofilm activity in a spectrophotometric assay. More detailed analysis using LC-MS/MS was able to reveal the presence of enzymes and proteins that have biosurfactant properties. These preliminary results strongly suggest that bacteria with a diverse range of biophysical characteristics that are useful for WW bioremediation specifically WW from dye industries and the mining sector, can be

179

isolated and cultured from hot springs. Further characterisation is required to identify the causative bioactive molecules, and the optimisation of culture conditions for maximum output of biophysical features.

11.6 Potential of hot-spring water isolates from South Africa in wastewater bioremediation

In this investigation, isolates from hot springs were reported to produce potential enzymes (Chapter 7), reduce phenol (Chapter 9) and possess biophysical properties, e.g. biosurfactant production (Chapter 8) useful for water bioremediation. Furthermore, the CFCS extracts of the isolates were screened to determine whether they could reduce turbidity or colouration in food pollutants containing melanoidin, textile pollutants including precursor textile dyes BB and a commercial dye (Dye It, CDS1, SA) and three WW samples, namely brewery and dairy industry WWs and river water contaminated with industrial effluents. No reduction in turbidity was observed with coffee, and only marginally (7-8%) with soya sauce in two isolates (76S, 77S). However, the turbidity of brewery WW was reduced by up to 40% with isolates 54T and 72T. No reduction in colouration of BB for contaminated river water, suspected to be effluent discharges from a dye manufacturer based on previous observations at the sample site. A minor reduction in colour of 5-6% for a commercial textile dye, and 25% reduction in dairy WW were reported in this study. Different isolates were responsible. Since phenol is a priority environmental pollutant (Galgale et al., 2014), further testing was performed in order to determine whether the phenolic compounds in these abovementioned samples and PAHs (petroleum and paraffin oil) were reduced. Five isolates (2T, 19S, 77S, 83Li and 84Li) significantly reduced phenol in phenol red broth media with a maximum of 59% (19S, 84Li). In the simulated pollutant experiment, phenols were reduced by up to 78% in coffee, and by up to 54% in soya sauce. No isolate could be identified that reduced phenol in BB and CV but in the commercial dye, 84Li was able to reduce phenol by 74%. Two isolates, namely 19S and 84Li were the best performers in reducing phenol in brewery WW (48%), river water contaminated with industrial effluents (54-56%) and dairy WW (47-73%), identified as Anoxybacillus sp. and Brevibacillus sp., respectively. Again, the CFCS extracts of these two isolates were also the most effective at reduction of phenol in PAHs, petroleum and paraffin oil. There was no effect of temperature ranging between 25 °C, 37 °C and 53 °C on these processes.

This study shows that the crude extracellular extracts from culture supernatants, at room temperature for 3 h, were able to reduce turbidity in brewery and dairy WW. Reduction of phenol was even more dramatic showing significant results in coffee, soya sauce, commercial

180

textile dyes, petroleum and paraffin oil, and all three types of WW samples tested. The relevant isolates were identified as Bacillus-related and are therefore robust and heat-tolerant, and suitable candidates for WW bioremediation, especially in a country that is water-scarce and where skills and resources are limited. Two isolates, namely thermophilic Anoxybacillus sp. 19S and mesophilic Brevibacillus sp. 84Li require further characterisation of their bioactive molecules and mechanisms involved in bioremediation of polluted WWs.

11.7 Antimicrobial activity of Bacillus mojavensis, isolated from hot springs, South Africa, against environmental multiple antibiotic-resistant bacteria and human pathogens

Globally, AR is on the increase and there is a demand for the discovery of natural alternative antimicrobial drugs. Novel biomolecules are commonly found in extreme environments, e.g. hot springs, and in this study, isolates were screened for biosurfactant and antimicrobial properties. One isolate (76S) was found to inhibit growth of Gram-positive pathogens including B. cereus, Kocuria and S. aureus, but not Gram-negative pathogens. The CFCS of isolate 76S was also able to inhibit the growth of 71% of environmental bacteria that were AR, suggesting that it has applications for the treatment of both human infections directly, and to address the new issue of AR in the environment recently classified as an emerging water pollutant. Two biosurfactant molecules were identified in the CFCS by LC-MS/MS to be subtilisin BM1 (alkaline serine protease) and bacillolysin (neutral metalloprotease); the latter has not been previously described in B. mojavensis and not in any public database, suggesting that this is a novel finding. The isolate was identified to be B. mojavensis by 16S rDNA amplicon sequencing. This species is known to be useful for the biocontrol of fungal infections, and to produce several types of surfactins at the same time.

This study shows that hot springs are a source of bacteria that have the potential to be used in biocontrol through production of biosurfactants and antimicrobial agents. They can be applied to human pathogens and to environmental MAR populations to address the contamination of the environment by MAR. The components are natural, biodegradable and easy to produce without any further genetic or chemical manipulations. Further investigation is required to isolate the components and to characterise them, including optimisation of conditions of production.

181

REFERENCES

Aanniz, T., Ouadghiri, M., Melloul, M., Swings, J., Elfahime, E., Ibijbijen, J., Ismaili, M. & Amar, M. (2015). Thermophilic bacteria in Moroccan hot springs, salt marshes and desert soils. Brazilian Journal of Microbiology, 46(2):443-453.

Abari, A.H., Emtiazi, G., Ghasemi, S.M. & Roghanian, R. (2013). Isolation and characterization of a novel toluene-degrading bacterium ex-hibiting potential application in bioremediation. Jundishapur Journal of Microbiology (JJM), 6(3):256-261.

Abbas, A.H., Wsain, S.M., Al-Quraishy, D.A.A. & Fartoos, S.K. (2015). Using of biosurfactant and bacterial enzymes for treatment of the oily polluted soils. International Journal of Advanced Research (IJAR), 3(5):741-748.

Abbas, S.H., Ismail, I.M., Mostafa, T.M. & Sulaymon, A.H. (2014). Biosorption of heavy metals: A review. Journal of Chemical Science and Technology Oct, 3(4):74-102.

Abbott, S.L., Moler, S., Green, N., Tran, R.K., Wainwright, K. & Janda, J.M. (2011). Clinical and laboratory diagnostic characteristics and cytotoxigenic potential of Hafnia alvei and Hafnia paralvei strains. Journal of Clinical Microbiology, 49(9):3122-3126.

Abdel-Nour, M., Duncan, C., Low, D.E. & Guyard, C. (2013). Biofilms: The stronghold of Legionella pneumophila. International Journal of Molecular Sciences, 14(11):21660-21675.

Acer, Ö, Pırınççiğlu, H., Bekler, F.M. & Gül-Güven, R. (2015). Anoxybacillus sp. AH1, an α- amylase-producing thermophilic bacterium isolated from Dargeçit hot spring. Biologia, 70(7):853-862.

Adams, G.O., Fufeyin, P.T., Okoro, S.E. & Ehinomen, I. (2015). Bioremediation, biostimulation and bioaugmention: A review. International Journal of Environmental Bioremediation & Biodegradation, 3(1):28-39.

Adler R.A., Claassen M., Godfrey, L & Turton, A.R. (2007). Water, mining and waste: an historical and economic perspective on conflict management in South Africa. The Economics of Peace and Security Journal, 2 (2):33-41.

Agarwal, S., Sadegh, H., Monajjemi, M., Hamdy, A.S., Ali, G.A., Memar, A.O., Shahryari- ghoshekandi, R., Tyagi, I. & Gupta, V.K. (2016). Efficient removal of toxic bromothymol blue and methylene blue from wastewater by polyvinyl alcohol. Journal of Molecular Liquids, 218:191-197.

182

Agbor, G.A., Vinson, J.A. & Donnelly, P.E. (2014). Folin-Ciocalteau reagent for polyphenolic assay. International Journal of Food Science, Nutrition and Dietetics, 3(8):147-156.

Agunbiade, M.O., Pohl, C.H. & Ashafa, A.O. (2016). A review of the application of bioflocculants in wastewater treatment. Polish Journal of Environmental Studies, 25(4):1381- 1389.

Ahmad, S.A., Syed, M.A., Arif, N.M., Shukor, M. & Shamaan, N.A. (2011). Isolation, identification and characterization of elevated phenol degrading Acinetobacter sp. strain AQ5NOL1. Australian Journal of Basic and Applied Sciences, 5(8):1035-1045.

Ahmed, AA., Salam, N., Kim, B., Singh, D., Zhi, X., Xiao, M. & Li, W. (2017). Diversity and Distribution of Thermophilic Bacteria in Hot Springs of Pakistan. Microbial Ecology, 74 (1): 116–127.

Akel, H., Al-Quadan, F., Atoum, M. & Battikhi, M. (2008). Phenotypic and genotypic characterization of three novel halophilic Bacillus strains from Jordanian hot springs. Jordan Journal of Biological Sciences, 1(1):19-25.

Akhtyamova, N. (2013). Human pathogens-the plant and useful endophytes. Journal of Medical Microbiology & Diagnosis, 2:e121.

Akintokun, AK., Adebajo, SO. & Akinremi, CA. (2017). Potential biosurfactant-producing bacteria from pharmaceutical wastewater using simple screening methods in south-west Nigeria. Applied Environmental Research 39 (2) :41-54.

Al Masud, H.A., Mahbub, K.R. & Morshed, M. (2015). Biodecolourization of textile dyes by effluent adapted bacteria. IOSR Journal of Pharmacy and Biological Sciences 10 (4):88-95.

Alain, K. & Querellou, J. (2009). Cultivating the uncultured: Limits, advances and future challenges. Extremophiles, 13(4):583-594.

Al-Araji, L., Basri, M. & Salleh, A.B. (2007). Minireview: Microbial surfactant. Asia Pacific Journal of Molecular Biology and Biotechnology, 15(3):99-105.

Al-Bahry, S.N., Mahmoud, I.Y., Paulson, J.R. & Al-Musharafi, S.K. (2014). Antibiotic resistant bacteria in terrestrial and aquatic environments: A review. The International Arabic Journal of Antimicrobial Agents, 4(3):754-765

Al-dayhistani, H. (2012). Bioremediation of Cu, Ni and Cr from rotograuvure wastewater using immobilized, dead and live biomass of indigenous thermophilic Bacillus species. The internet Journal of Microbiology, 10 (1): 1.

183

Al-Dhabi, N.A., Esmail, G.A., Duraipandiyan, V., Arasu, M.V. & Salem-Bekhit, M.M. (2016). Isolation, identification and screening of antimicrobial thermophilic Streptomyces sp. al-dhabi-1 isolated from Tharban hot spring, Saudi Arabia. Extremophiles, 20(1):79-90.

Al-Grawi, J. (2008). Hafnia alvei urinary tract infection. Iraqi Postgraduate Medical Journal, 7 (1):1.

Alhassani, H.A., Rauf, M.A. & Ashraf, S.S. (2007). Efficient microbial degradation of toluidine blue dye by Brevibacillus sp. Dyes and Pigments, 75(2):395-400.

Ali, H. (2010). Biodegradation of synthetic dyes - a review. Water, Air, & Soil Pollution, 213(1- 4):251-273.

Alkhalili, R.N., Bernfur, K., Dishisha, T., Mamo, G., Schelin, J., Canbäck, B., Emanuelsson, C. & Hatti-Kaul, R. (2016). Antimicrobial protein candidates from the thermophilic Geobacillus sp. strain ZGt-1: Production, proteomics, and bioinformatics analysis. International Journal of Molecular Sciences, 17(8):1363.

Allen, H.K., Donato, J., Wang, H.H., Cloud-Hansen, K.A., Davies, J. & Handelsman, J. (2010). Call of the wild: Antibiotic resistance genes in natural environments. Nature Reviews Microbiology, 8(4):251-259.

Allievi, M.C. & Mariano, P.A. (2011). Metal biosorption by surface-layer proteins from Bacillus species. Journal of Microbiology and Biotechnology, 21(2):147-153.

Alneyadi, A.H. & Ashraf, S.S. (2016). Differential enzymatic degradation of thiazole pollutants by two different peroxidases–a comparative study. Chemical Engineering Journal, 303:S29- S38.

Al-Qodah, Z., Daghistani, H. & Alananbeh, K. (2013). Isolation and characterization of thermostable protease producing Bacillus pumilus from thermal spring in Jordan.African Journal of Microbiology Research, 7:3711-3719.

Altincicek, B., Linder, M., Linder, D., Preissner, K.T. & Vilcinskas, A. (2007). Microbial metalloproteinases mediate sensing of invading pathogens and activate innate immune responses in the lepidopteran model host Galleria mellonella. Infection and Immunity, 75(1):175-183.

Al-Wasify, R.S., Al-Sayed, A.A., Saleh, S.M. & Aboelwafa, A.M. (2015). Bacterial exopolysaccharides as new natural coagulants for surface water treatment. International Journal of PharmTech Research, 8(9):198-207.

184

Andrews, J.M. & Wise, R. (2002). Susceptibility testing of Bacillus species.The Journal of Antimicrobial Chemotherapy, 49(6):1040-1042.

Anjaneya, O., Souche, S.Y., Santoshkumar, M. & Karegoudar, T.B. (2011). Decolorization of sulfonated azo dye metanil yellow by newly isolated bacterial strains: Bacillus sp. strain AK1 and Lysinibacillus sp. strain AK2. Journal of Hazardous Materials, 190(1):351-358.

Anttila, J., Kaitala, V., Laakso, J. & Ruokolainen, L. (2015). Environmental variation generates environmental opportunist pathogen outbreaks. PloS One, 10(12):e0145511.

Apetroaie‐Constantin, C., Mikkola, R., Andersson, M.A., Teplova, V., Suominen, I., Johansson, T. & Salkinoja‐Salonen, M. (2009). Bacillus subtilis and B. mojavensis strains connected to food poisoning produce the heat stable toxin amylosin. Journal of Applied Microbiology, 106(6):1976-1985.

Archna, S., Priyank, V., Nath, Y.A. & Kumar, S.A. (2015). Bioprospecting for extracellular hydrolytic enzymes from culturable thermotolerant bacteria isolated from Manikaran thermal springs. Research Journal of Biotechnology, vol 10 (4): 33-42.

Areekit, S., Kanjanavas, P., Khawsak, P., Pakpitchareon, A., Potivejkul, K., Chansiri, G. & Chansiri, K. (2011). Cloning, expression, and characterization of thermotolerant manganese superoxide dismutase from Bacillus sp. MHS47. International Journal of Molecular Sciences, 12(1):844-856.

Arvind, M., Shreedharan, P.C. & Ambika, S.R. (2015). Bioremediation for environmental management. International Journal of Environmental Science and Development, 6(7):555.

Asad, W., Asif, M. & Rasool, S.A. (2011). Extracellular enzyme production by indigenous thermophilic bacteria: Partial purification and characterization of α-amylase by Bacillus sp. WA21. Pakistan Journal of Botany, 43(2):1045-1052.

Asgarani, E., Soudi, M.R., Borzooee, F. & Dabbagh, R. (2012). Radio-resistance in psychrotrophic Kocuria sp. ASB 107 isolated from Ab-E-Siah radioactive spring. Journal of Environmental Radioactivity, 113:171-176.

Asoodeh, A., Alemi, A., Heydari, A. & Akbari, J. (2013). Purification and biochemical characterization of an acidophilic amylase from a newly isolated Bacillus sp. DR90. Extremophiles, 17(2):339-348.

Atanassova, D., Kefalas, P. & Psillakis, E. (2005). Measuring the antioxidant activity of olive oil mill wastewater using chemiluminescence. Environment International, 31(2):275-280.

185

Aunpad, R., Sripotong, N., Khamlak, K., Inchidjuy, S., Rattanasinganchan, P. & Pipatsatitpong, D. (2011). Isolation and characterization of bacteriocin with anti-listeria and anti-MRSA activity produced by food and soil isolated bacteria. African Journal of Microbiology Research, 5(29):5297-5303.

Ayandiran, T.A., Ayandele, A.A., Dahunsi, S.O. & Ajala, O.O. (2014). Microbial assessment and prevalence of antibiotic resistance in polluted Oluwa River, Nigeria. The Egyptian Journal of Aquatic Research, 40(3):291-299.

Aysha, O.S. & Mumtaj, K. (2014). Degradation of phenol by selected strains of Bacillus species isolated from marine water. American Journal of Pharmacy and Health Science, 2(6):41-56.

Azzam, A.M. & Tawfik, A. (2015). Removal of heavy metals using bacterial bio-flocculants of Bacillus sp. and Pseudomonas sp. Journal of Environmental Engineering and Landscape Management, 1-7.

Bacon, C.W., Hinton, D.M., Mitchell, T.R., Snook, M.E. & Olubajo, B. (2012). Characterization of endophytic strains of Bacillus mojavensis and their production of surfactin isomers. Biological Control, 62(1):1-9.

Bacon, C.W. & Hinton, D.M. (2011). Bacillus mojavensis: its endophytic nature, the surfactins, and their role in the plant response to infection by Fusarium verticillioides. In: Maheshwari, D.K. (Editor), Bacteria in Agrobiology: Plant Growth Responses. Springer Heidelberg New York Dordrecht London. pp. 21-39.

Bajlan, J.W., Alden, S.B. & Al-Saffar, O.B. (2013). Characterisations of bioflocculant produced by Bacillus coagulans 8B and application in domestic wastewater treatment. Iraqi academic Scientific Journal, 5:47-55.

Baker, K.S., Burnett, E., McGregor, H., Deheer-Graham, A., Boinett, C., Langridge, G.C., Wailan, A.M., Cain, A.K., Thomson, N.R. & Russell, J.E. (2015). The Murray collection of pre- antibiotic era Enterobacteriacae: A unique research resource. Genome Medicine, 7(1):97-105.

Baker-Austin, C., Wright, M.S., Stepanauskas, R. & McArthur, J.V. (2006). Co-selection of antibiotic and metal resistance. Trends in Microbiology, 14(4):176-182.

Bakir, ZB. & Metin, K. (2017). Production and characterisation of an alkaline lipase from thermophilic Anoxybacillus sp. HBB16. Chemistry Biochemistry Engineering Q, 31 (3): 303- 312.

186

Bakri, Y., Ammouneh, H., El-Khouri, S., Harba, M. & Thonart, P. (2012). Isolation and identification of a new Bacillus strain for amylase production. Research in Biotechnology, 3(6):51-58.

Balakrishnan, S., John, K.R. & George, M.R. (2003). Antibiotic susceptibility of bacillus spp. isolated from shrimp (Penaeus monodon) culture ponds. Indian Journal of Marine Sciences, 32(1):81-84.

Baltaci, M.O., Genc, B., Arslan, S., Adiguzel, G. & Adiguzel, A. (2017). Isolation and characterization of thermophilic bacteria from geothermal areas in Turkey and preliminary research on biotechnologically important enzyme production. Geomicrobiology Journal, 34(1):53-62.

Banat, I.M., De Rienzo, Mayri A Díaz & Quinn, G.A. (2014). Microbial biofilms: Biosurfactants as antibiofilm agents. Applied Microbiology and Biotechnology, 98(24):9915- 9929.

Bandounas, L., Pinkse, M., de Winde, J.H. & Ruijssenaars, H.J. (2013). Identification of a quinone dehydrogenase from a Bacillus sp. involved in the decolourization of the lignin-model dye, azure B. New Biotechnology, 30(2):196-204.

Barakat, KM., Hassan, S. & Darwesh, O. (2017). Biosurfactant production by haloalkaliphilic Bacillus strains isolated from Red Sea, Egypt. Egyptian Journal of Aquatic Research 43: 205- 211.

Barna, Z. & Kádár, M. (2012). The risk of contracting infectious diseases in public swimming pools: A review. Annali dell'Istituto Superiore di Sanità, 48(4):374-386.

Baruzzi, F., Quintieri, L., Morea, M. & Caputo, L. (2011). Antimicrobial compounds produced by Bacillus spp. and applications in food. In: A. Méndez-Vilas (Editor) Science against Microbial Pathogens: Communicating Current Research and Technological Advances. Formatex Research Center, Badajoz Spain. pp. 1102-1111.

Basu, B., Ramasamy, R., Viswanathan, S., Priya, I.D., Jeyan, J., Reddy, A. & Kumar, V.V. (2014). Analysis of aromatic hydrocarbon degrading capacity by thermophilic bacteria isolated from oil contaminated soil. International Journal of Chemical Technology Research, 6(10): 4556-4563.

Bauer, A.W., Kirby, W.M., Sherris, J.C. & Turck, M. (1966). Antibiotic susceptibility testing by a standardized single disk method. American Journal of Clinical Pathology, 45(4):493.

187

Baumgartner, M., Yapi, A., Gröbner-Ferreira, R. & Stetter, K.O. (2003). Cultivation and properties of Echinamoeba thermarum n. sp., an extremely thermophilic amoeba thriving in hot springs. Extremophiles, 7(4):267-274.

Bazrafshan, E., Amirian, P., Mahvi, A.H. & Ansari-Moghaddam, A. (2016). Application of adsorption process for phenolic compounds removal from aqueous environments: A systematic review. Global NEST Journal, 18(1):146-163.

Bekler, F.M., Acer, Ö & Güven, K. (2015). Production and purification of novel thermostable alkaline protease from Anoxybacillus sp. KP1. Cellular and Molecular Biology, 61(4):113-120.

Belduz, A.O., Canakci, S., Chan, K., Kahar, U.M., Chan, C.S., Yaakop, A.S. & Goh, K.M. (2015). Genome sequence of Anoxybacillus ayderensis AB04 T isolated from the Ayder hot spring in Turkey. Standards in Genomic Sciences, 10(1):70-81.

Belduz, A.O., Dulger, S. & Demirbag, Z. (2003). Anoxybacillus gonensis sp. nov., a moderately thermophilic, xylose-utilizing, endospore-forming bacterium. International Journal of Systematic and Evolutionary Microbiology, 53(5):1315-1320.

Belliveau, B.H., Starodub, M.E. & Trevors, J.T. (1991). Occurrence of antibiotic and metal resistance and plasmids in Bacillus strains isolated from marine sediment. Canadian Journal of Microbiology, 37(7):513-520.

Berg, G., Eberl, L. & Hartmann, A. (2005). The rhizosphere as a reservoir for opportunistic human pathogenic bacteria. Environmental Microbiology, 7(11):1673-1685.

Berić, T., Kojić, M., Stanković, S., Topisirović, L., Degrassi, G., Myers, M., Venturi, V. & Fira, D. (2012). Antimicrobial activity of Bacillus sp. natural isolates and their potential use in the biocontrol of phytopathogenic bacteria. Food Technology and Biotechnology, 50(1):25-31.

Berjeaud, J., Chevalier, S., Schlusselhuber, M., Portier, E., Loiseau, C., Aucher, W., Lesouhaitier, O. & Verdon, J. (2016). Legionella pneumophila: The paradox of a highly sensitive opportunistic waterborne pathogen able to persist in the environment. Frontiers in Microbiology, 7: 486-508.

Berthiaume, C., Gilbert, Y., Fournier‐Larente, J., Pluchon, C., Filion, G., Jubinville, E., Sérodes, J., Rodriguez, M., Duchaine, C. & Charette, S.J. (2014). Identification of dichloroacetic acid degrading Cupriavidus bacteria in a drinking water distribution network model. Journal of Applied Microbiology, 116(1):208-221.

188

Bhaiyat, M.I., Hariharan, H., Chikweto, A., Tiwari, K., Sharma, R.N. & Kobayashi, Y. (2013). Isolation of coagulase-negative Staphylococcus spp. and Kocuria varians in pure culture from tissues of cases of mortalities in parrots in Grenada, West Indies. International Journal of Veterinary Medicine: Research & Reports (2013) Article ID 149634.

Bhange, K., Chaturvedi, V. & Bhatt, R. (2016). Simultaneous production of detergent stable keratinolytic protease, amylase and biosurfactant by Bacillus subtilis PF1 using agro industrial waste. Biotechnology Reports, 10:94-104.

Bhattacharya, A. & Khare, S.K. (2016). Sustainable options for mitigation of major toxicants originating from electronic waste. Current Science, 111(12):1946-1954.

Bhullar, K., Waglechner, N., Pawlowski, A., Koteva, K., Banks, E.D., Johnston, M.D., Barton, H.A. & Wright, G.D. (2012). Antibiotic resistance is prevalent in an isolated cave microbiome. PLoS One, 7(4):e34953.

Binning, K. & Baird, D. (2001). Survey of heavy metals in the sediment of the Swarkops River estuary, Port Elizabeth, South Africa. Water SA, 27(4):461-466.

Bittar, F. & Rolain, J. (2010). Detection and accurate identification of new or emerging bacteria in cystic fibrosis patients. Clinical Microbiology and Infection, 16(7):809-820.

Blair, J.M., Webber, M.A., Baylay, A.J., Ogbolu, D.O. & Piddock, L.J. (2015). Molecular mechanisms of antibiotic resistance. Nature Reviews Microbiology, 13(1):42-51.

Bodike, R. & Thatikonda, S.R. (2014). Biotreatment of brewery effluent using Pseudomonas species. IOSR-Journal of Environmental Science, Toxicology and Food Technology, 8(6):8-12.

Boekstein, M.S. (2014). Healing waters: Balneological classification of thermal springs in south africa: Tourism. African Journal for Physical Health Education, Recreation and Dance, 20:557- 568.

Bollin, M. (2014). Occurrence of multiple antibiotic resistant bacteria in aquatic environments in central Minnesota. American Journal of Undergraduate Research, 12 (3): 19-35.

Box, J.D. (1983). Investigation of the Folin-Ciocalteau phenol reagent for the determination of polyphenolic substances in natural waters. Water Research, 17(5):511-525.

Brandao, M.L.L., Umeda, N.S., Jackson, E., Forsythe, S.J. & de Filippis, I. (2017). Isolation, molecular and phenotypic characterization, and antibiotic susceptibility of Cronobacter spp. from Brazilian retail foods. Food Microbiology, 63:129-138.

189

Branquinho, R., Klein, G., Kämpfer, P. & Peixe, L.V. (2015). The status of the species Bacillus aerophilus and Bacillus stratosphericus. Request for an opinion. International Journal of Systematic and Evolutionary Microbiology, 65(3):1101.

Briggs, B.R., Brodie, E.L., Tom, L.M., Dong, H., Jiang, H., Huang, Q., Wang, S., Hou, W., Wu, G. & Huang, L. (2014). Seasonal patterns in microbial communities inhabiting the hot springs of Tengchong, Yunnan Province, China. Environmental Microbiology, 16(6):1579-1591.

Bueno, J. (2014). Anti-biofilm drug susceptibility testing methods: Looking for new strategies against resistance mechanism. Journal of Microbial & Biochemical Technology, (S3):1-9.

Bujang, M., Ibrahim, N.A. & Rak, A.E. (2013). Biodegradation of oily wastewater by pure culture of Bacillus cereus. ARPN Journal of Agricultural and Biological Sciences, 8(2):1-8.

Buntic, A.V., Pavlovic, M.D., Antonovic, D.G., Siler-Marinkovic, S.S., & Dimitrijevic- Brankovic, S.I. (2017). A treatment of wastewater containing basic dyes by the use of new strain Streptomyces microflavus CKS6. Journal of Cleaner Production, 148:347-354.

Case, Rj., Boucher, Y., Dahlloff, I., Holmstrom, C., Doolittle, WF. & Kjelleberg, S. (2007). Use of 16S rRNA and rpoB genes as molecular markers for microbial ecology studies. Applied and Environmental Microbiology, 73 (1): 278-288.

Cha, M.H., Yong, W.M., Lee, S.M., Lee, Y.S. & Chung, I.Y. (2000). The biochemical and molecular characterization of recombinant Bacillus subtilis tripeptidase (PepT) as a zinc- dependent metalloenzyme. Molecules and Cells, 10(4):423-431.

Chaabna, Z., Forey, F., Reyrolle, M., Jarraud, S., Atlan, D., Fontvieille, D. & Gilbert, C. (2013). Molecular diversity and high virulence of Legionella pneumophila strains isolated from biofilms developed within a warm spring of a thermal spa. BMC Microbiology, 13:17-27.

Chai, K.P., Othman, N.F., Teh, A.H., Ho, K.L., Chan, K.G., Shamsir, M.S., Goh, K.M. & Ng, C.L. (2016). Crystal structure of Anoxybacillus alpha-amylase provides insights into maltose binding of a new glycosyl hydrolase subclass. Nature Scientific Reports, 6:srep 23126.

Chakraborty, J. & Das, S. (2016). Molecular perspectives and recent advances in microbial remediation of persistent organic pollutants. Environmental Science and Pollution Research, 23(17):16883-16903.

Chakraborty, S., Sakka, M., Kimura, T. & Sakka, K. (2008). Characterization of a dihydrolipoyl dehydrogenase having diaphorase activity of Clostridium kluyveri. Bioscience, Biotechnology, and Biochemistry, 72(4):982-988.

190

Chandra, R., Bharagava, R.N. & Rai, V. (2008). Melanoidins as major colourant in sugarcane molasses based distillery effluent and its degradation. Bioresource Technology, 99(11):4648- 4660.

Chatterjee, S., Sharma, C., Mehtani, P. & Gupta, N. (2015). In situ bioremediation of dairy waste water – A novel approach in dairy waste management. IIS University Journal of Science and Technology, 4 (1): 6-10.

Chavan, M.N., Kulkarni, M.V., Zope, V.P. & Mahulikar, P.P. (2006). Microbial degradation of melanoidins in distillery spent wash by an indigenous isolate. Indian Journal of Biotechnology, 5(3):416-421.

Guo-Qiang Chen; Xiao-Ran Jiang. 2018. Next generation industrial biotechnology based on extremophilic bacteria. Current Opinion in Biotechnology. Volume 50, April 2018, Pages 94- 100

Chen, H., Chi, H., Chiu, N. & Huang, F. (2015a). Kocuria kristinae: A true pathogen in pediatric patients. Journal of Microbiology, Immunology and Infection, 48:80-84.

Chen, M., Xu, P., Zeng, G., Yang, C., Huang, D. & Zhang, J. (2015b). Bioremediation of soils contaminated with polycyclic aromatic hydrocarbons, petroleum, pesticides, chlorophenols and heavy metals by composting: Applications, microbes and future research needs. Biotechnology Advances, 33(6):745-755.

Chen, C., Kuo, J., Yang, H. & Chung, Y. (2013a). A coupled biological and photocatalysis pretreatment system for the removal of crystal violet from wastewater. Chemosphere, 92(6):695-701.

Chen, H., Jogler, M., Tindall, B.J., Klenk, H., Rohde, M., Busse, H. & Overmann, J. (2013b). Sphingomonas starnbergensis sp. nov., isolated from a prealpine freshwater lake. International Journal of Systematic and Evolutionary Microbiology, 63(3):1017-1023.

Chen, W., Huang, H., Chang, J., Han, Y., Guo, T. & Sheu, S. (2013c). Tepidimonas fonticaldi sp. nov., a slightly thermophilic betaproteobacterium isolated from a hot spring. International Journal of Systematic and Evolutionary Microbiology, 63(5):1810-1816.

Chen, T., Chou, Y., Chen, W., Arun, B. & Young, C. (2006a). Tepidimonas taiwanensis sp. nov., a novel alkaline-protease-producing bacterium isolated from a hot spring. Extremophiles, 10(1):35-40.

191

Chen, Y.P., Rekha, P.D., Arun, A.B., Shen, F.T., Lai, W. & Young, C.C. (2006b). Phosphate solubilizing bacteria from subtropical soil and their tricalcium phosphate solubilizing abilities. Applied Soil Ecology, 34(1):33-41.

Chen, C.Y., Stemberger, R.S., Klaue, B., Blum, J.D., Pickhardt, P.C. & Folt, C.L. (2000). Accumulation of heavy metals in food web components across a gradient of lakes. Limnology and Oceanography, 45(7):1525-1536.

Cheung, R.C.F., Wong, J.H., Pan, W.L., Chan, Y.S., Yin, C.M., Dan, X.L., Wang, H.X., Fang, E.F., Lam, S.K. & Ngai, P.H.K. (2014). Antifungal and antiviral products of marine organisms. Applied Microbiology and Biotechnology, 98(8):3475-3494.

Chiacchierini, E., Restuccia, D. & Vinci, G. (2004). Bioremediation of food industry effluents: Recent applications of free and immobilised polyphenoloxidases. Food Science and Technology International, 10(6):373-382.

Chiong, T., Lau, S.Y., Khor, E.H. & Danquah, M.K. (2014). Enzymatic approach to phenol removal from wastewater using peroxidases. OA Biotechnology, 10(1):3-9.

Chitanand, M.P., Kadam, T.A., Gyananath, G., Totewad, N.D. & Balhal, D.K. (2010). Multiple antibiotic resistance indexing of coliforms to identify high risk contamination sites in aquatic environment. Indian Journal of Microbiology, 50(2):216-220.

Chojnacka, K. (2010). Biosorption and bioaccumulation – the prospects for practical applications. Environment International, 36(3):299-307.

Chuang, T., Ong, P., Wang, T., Huang, H., Chi, M. & Lin, L. (2012). Molecular characterization of a novel trehalose-6-phosphate hydrolase, TreA, from Bacillus licheniformis. International Journal of Biological Macromolecules, 50(3):459-470.

Chung, M.C., Popova, T.G., Millis, B.A., Mukherjee, D.V., Zhou, W., Liotta, L.A., Petricoin, E.F., Chandhoke, V., Bailey, C. & Popov, S.G. (2006). Secreted neutral metalloproteases of Bacillus anthracis as candidate pathogenic factors. The Journal of Biological Chemistry, 281(42):31408-31418.

Cihan, A.C. (2013). Taxonomic classification of Anoxybacillus isolates from geothermal regions in Turkey by 16S rRNA gene sequences and ARDRA, ITS-PCR, rep-PCR analyses. Polish Journal of Microbiology, 62(2):149-163.

192

Cihan, A.C., Tekin, N., Ozcan, B. & Cokmus, C. (2012). The genetic diversity of genus Bacillus and the related genera revealed by 16S rRNA gene sequences and ARDRA analyses isolated from geothermal regions of Turkey. Brazilian Journal of Microbiology, 43(1):309-324.

Cladera‐Olivera, F., Caron, G.R. & Brandelli, A. (2004). Bacteriocin‐like substance production by Bacillus licheniformis strain P40. Letters in Applied Microbiology, 38(4):251-256.

Clemente, J.C., Pehrsson, E.C., Blaser, M.J., Sandhu, K., Gao, Z., Wang, B., Magris, M., Hidalgo, G., Contreras, M. & Noya-Alarcón, Ó (2015). The microbiome of uncontacted Amerindians. Science Advances, 1(3):e1500183.

Clinical and Laboratory Standards Institute. (2013). VET01-A4:performance standards for antimicrobial disk and dilution susceptibility tests for bacteria isolated from animals; Approved Standard – Fourth Edition. Wayne, USA. http://shopping.netsuite.com/s.nl.c1253739/it.A/id.1684/.f.

Cliver, D. (2000). Emerging pathogens on the rise: How can waterborne illness be prevented? California Agriculture, 54(5):78-79.

Coelho, C., Ribeiro, M., Cruz, A.C., Domingues, M.R.M., Coimbra, M.A., Bunzel, M. & Nunes, F.M. (2014). Nature of phenolic compounds in coffee melanoidins. Journal of Agricultural and Food Chemistry, 62(31):7843-7853.

Coenye, T. (2002). Infection by Ralstonia species in cystic fibrosis patients: Identification of R. pickettii and R. mannitolilytica by polymerase chain reaction. Emerging Infectious Diseases 8 (7): 692-696.

Coman, I., Bilodeau, L., Lavoie, A., Carricart, M., Tremblay, F., Zlosnik, J.E. & Berthiaume, Y. (2017). Ralstonia mannitolilytica in cystic fibrosis: A new predictor of worse outcomes. Respiratory Medicine Case Reports, 20:48-50.

Coonrod, J.D., Leadley, P.J. & Eickhoff, T.C. (1971). Antibiotic susceptibility of Bacillus species. The Journal of Infectious Diseases, 123(1):102-105.

Corvey, C., Stein, T., Düsterhus, S., Karas, M. & Entian, K. (2003). Activation of subtilin precursors by Bacillus subtilis extracellular serine proteases subtilisin (AprE), WprA, and vpr. Biochemical and Biophysical Research Communications, 304(1):48-54.

Cosa, S., Ugbenyen, A., Mabinya, L.V., Rumbold, K. & Okoh, A. I. (2013). Characteristaion and flocculation efficiency of a bioflocculant produced by a marine Halobacillus. Environment Technology, 34(18):2671-2679.

193

Courvalin, P. (1994). Transfer of antibiotic resistance genes between gram-positive and gram- negative bacteria. Antimicrobial Agents and Chemotherapy, 38(7):1447-1451.

Coutinho, F.H., Cardoso, A.M., de Oliveira Santoro, D., et al. (2013). Antibiotic Resistance in Aquatic Environments of Rio de Janeiro, Brazil INTECH Open Access Publisher.

Czemierska, M., Szczes, A. & Jarosz-Wilkolazka, A. (2015). Purification of wastewater by natural flocculants. BioTechnologia. Journal of Biotechnology Computational Biology and Bionanotechnology, 96(4): 272-278.

D’Costa, V.M., King, C.E., Kalan, L., Morar, M., Sung, W.W., Schwarz, C., Froese, D., Zazula, G., Calmels, F. & Debruyne, R. (2011). Antibiotic resistance is ancient. Nature, 477(7365):457- 461.

Da Fonseca, F.S., Angolini, C.F., Arruda, M.A.Z., Junior, C.A., Santos, C.A., Saraiva, A.M., Pilau, E., Souza, A.P., Laborda, P.R. & de Oliveira, P.F. (2015). Identification of oxidoreductases from the petroleum Bacillus safensis strain. Biotechnology Reports, 8:152-159.

Das, P., Mukherjee, S. & Sen, R. (2009). Substrate dependent production of extracellular biosurfactant by a marine bacterium. Bioresource Technology, 100(2):1015-1019.

Dashti, A.A., Jadaon, M.M., Abdulsamad, A.M. & Dashti, H.M. (2009). Heat treatment of bacteria: A simple method of DNA extraction for molecular techniques. Kuwait Medical Journal, 41(2):117-122.

Davies, J. & Davies, D. (2010). Origins and evolution of antibiotic resistance. Microbiology and Molecular Biology Reviews: MMBR, 74(3):417-433.

Demorne, K., Cordova, DC., Astorga-Elo, M & REnganathan, P. (2107). Extremozymes: a potential source for industrial applications. Journal of microbiology and Biotechnology 27(4): 649-659.

De Souza, M., Nair, S., Bharathi, P.L. & Chandramohan, D. (2006). Metal and antibiotic- resistance in psychrotrophic bacteria from Antarctic marine waters. Ecotoxicology, 15(4):379- 384.

Dean, C.R. & Ward, O.P. (1991). Nature of Escherichia coli cell lysis by culture supernatants of Bacillus species. Applied and Environmental Microbiology, 57(7):1893-1898.

Dedin, F.R., Fardiaz, D., Apriyantono, A. & Andarwulan, N. (2010). Isolation and characterization of soy sauce melanoidin and its role as antioxidant. Jurnal Teknologi dan Industri Pangan, 17(3):204.

194

Deive, F.J., Domínguez, A., Barrio, T., Moscoso, F., Morán, P., Longo, M.A. & Sanromán, M.A. (2010). Decolorization of dye reactive black 5 by newly isolated thermophilic microorganisms from geothermal sites in Galicia (Spain). Journal of Hazardous Materials, 182(1):735-742.

Demirjian -Varas, F. & Cassidy, C.S. (2001). Enzymes from extremophiles. Current Opinion in Chemical Biology, 5(2):144-151.

Demirkan, E., Sevgi, T. & Başkurt, M. (2017). Optimization of physical factors affecting the production of the α-amylase from a newly isolated Bacillus sp. M10 strain. Karaelmas fen ve Mühendislik Dergisi, 7(1):23-30.

Derekova, A., Mandeva, R. & Kambourova, M. (2008). Phylogenetic diversity of thermophilic carbohydrate degrading bacilli from Bulgarian hot springs. World Journal of Microbiology and Biotechnology, 24(9):1697-1702.

Desai, S.S., Tennali, G.B., Channur, N., Anup, A.C., Deshpande, G. & Murtuza, B.A. (2011). Isolation of laccase producing fungi and partial characterization of laccase. Biotechnology, Bioinformatics and Bioengineering, 1(4):543-549.

DeWaal, C.S., Robert, N., Witmer, J. & Tian, X.A. (2010). A comparison of the burden of foodborne and waterborne diseases in three world regions, 2008. Food Protection Trends, 30(8):483-490.

Dewan, E., Oberio, A. & Mathias, A. (2014). An emerging threat: a rare non fermenter in an immunocompromized patient Sphingomonas paucimobilis from a tertiary care centre. Indian Journal Medical Case Report, 3 (4):35-39

Di Giulio, M. (2007). The universal ancestor and the ancestors of archaea and bacteria were anaerobes whereas the ancestor of the eukarya domain was an aerobe. Journal of Evolutionary Biology, 20(2):543-548.

Divate, S.B. & Hinge, R.V. (2014). Review on research removal of phenol from wastewater by using different methods. International Journal of Scientific & Research Publications, 4 (5):1-3.

Djokic, L., Narancic, T., Biocanin, M., Saljnikov, E., Casey, E., Vasiljevic, B. & Nikodinovic- Runic, J. (2013). Phenol removal from four different natural soil types by Bacillus sp. PS11. Applied Soil Ecology, 70:1-8.

195

Dobretsov, S., Abed, R.M., Al Maskari, S.M., Al Sabahi, J.N. & Victor, R. (2011). Cyanobacterial mats from hot springs produce antimicrobial compounds and quorum-sensing inhibitors under natural conditions. Journal of Applied Phycology, 23(6):983-993.

Dominguez, M.C., de la Rosa, M. & Borobio, M.V. (2001). Application of a spectrophotometric method for the determination of post-antibiotic effect and comparison with viable counts in agar. Journal of Antimicrobial Chemotherapy, 47(4):391-398.

Drancourt, M. & Raoult, D. (2005). Sequence-based identification of new bacteria: A proposition for creation of an orphan bacterium repository. Journal of Clinical Microbiology, 43(9):4311-4315.

Dunn, R., Bares, S. & David, M.Z. (2011). Central venous catheter-related bacteremia caused by Kocuria kristinae: Case report and review of the literature. Annals of Clinical Microbiology and Antimicrobials, 10(1):31-36.

Duran, N. & Esposito, E. (2000). Potential applications of oxidative enzymes and phenoloxidase-like compounds in wastewater and soil treatment: A review. Applied Catalysis B: Environmental, 28(2):83-99.

Durve, A.A., Arvind, R. & Naphade, S.R. (2012). Decolourization of textile dyes and biological stains by bacterial strains isolated from industrial effluents. Advances in Applied Science Research, 3(5):2660-2671.

Dzoma, B.M., Moralo, R.A., Motsei, L.E., Ndou, R.V. & Bakunzi, F.R. (2010). Preliminary findings on the levels of five heavy metals in water, sediments, grass and various specimens from cattle grazing and watering in potentially heavy metal polluted areas of the North West province of South Africa. Journal of Animal and Veterinary Advances, 9(24):3026-3033.

El-Gayar, KE., Al Abboud, M. & Essa, AM. (2017). Characterisation of thermophilic bacteria isolated from two hot springs in Jazan, Saudi Arabia. Journal of Pure and Applied Microbiology 11 (2): 1-10.

Elkhalil, E.A., Gaffar, F.Y., El Siddig, M.A. & Osman, H.A. (2015). Isolation and molecular characterization of cellulolytic Bacillus isolates from soil and compost. American Journal of Microbiological Research, 3(2):55-58.

Eswaramoorthy, S., Poulain, S., Hienerwadel, R., Bremond, N., Sylvester, M.D., Zhang, Y., Berthomieu, C., Van Der Lelie, D. & Matin, A. (2012). Crystal structure of ChrR—a quinone reductase with the capacity to reduce chromate. PloS One, 7(4):e36017.

196

Even, S., Pellegrini, O., Zig, L., Labas, V., Vinh, J., Brechemmier-Baey, D. & Putzer, H. (2005). Ribonucleases J1 and J2: Two novel endoribonucleases in B. subtilis with functional homology to E. coli RNase E. Nucleic Acids Research, 33(7):2141-2152.

Everette, J.D., Bryant, Q.M., Green, A.M., Abbey, Y.A., Wangila, G.W. & Walker, R.B. (2010). Thorough study of reactivity of various compound classes toward the folin− ciocalteu reagent. Journal of Agricultural and Food Chemistry, 58(14):8139-8144.

Ezhilarasu, A. (2016). Textile industry dye degrading by bacterial strain Bacillus sp. International Journal of Advanced Research in Biological Sciences, 3(3):211-226.

Facchin, S., Alves, P.D.D., de Faria Siqueira, F., Barroca, T.M., Victória, J.M.N. & Kalapothakis, E. (2013). Biodiversity and secretion of enzymes with potential utility in wastewater treatment. Open Journal of Ecology, 3(1):34-47.

Fakruddin, M., Rahaman, M.M., Ahmed, M.M. & Hoque, M.M. (2013). Cronobacter sakazakii (Enterobacter sakazakii): An emerging foodborne pathogen. International Journal of Biomedical and Advance Research, 4(6):349-359.

Falanga, A., Lombardi, L., Franci, G., Vitiello, M., Iovene, M.R., Morelli, G., Galdiero, M. & Galdiero, S. (2016). Marine antimicrobial peptides: Nature provides templates for the design of novel compounds against pathogenic bacteria. International Journal of Molecular Sciences, 17(5):785-804.

Falcone-Dias, M.F., Vaz-Moreira, I. & Manaia, C.M. (2012). Bottled mineral water as a potential source of antibiotic resistant bacteria. Water Research, 46(11):3612-3622.

Fan, C. & Krishnamurthy, S. (1995). Enzymes for enhancing bioremediation of petroleum- contaminated soils: A brief review. Journal of the Air & Waste Management Association, 45(6):453-460.

Fan, N., Wu, C., Chen, T., Yu, W., Chen, C., Lee, S. & Lin, P. (2012). Microsporidial keratitis in patients with hot springs exposure. Journal of Clinical Microbiology, 50(2):414-418.

Fandi, K., Al-Muaikel, N. & Al-Momani, F. (2014) Antimicrobial activities of some thermophiles isolated from Jordan hot springs. International Journal of Chemical, Environmental & Biological Sciences, 2 (1): 57-61.

Fandi, K., Massadeh, M. & Laatsch, H. (2012) LC-MS/MS profiling-based metabolite screening of thermophilic bacteria from Jordanian hot springs. International conference on applied chemistry and Pharmaceutical Sciences, Penang Malaysia.

197

Fang, D., Yu, Y., Wu, L., Zhang, J. & Zhi, J. (2015). Bacillus subtilis-based colorimetric bioassay for acute biotoxicity assessment of heavy metal ions. Royal Society of Chemistry, Advances, 5: 59472-59479.

Farkas, A., Maróti, G., Kereszt, A. & Kondorosi, É (2017). Comparative analysis of the bacterial membrane disruption effect of two natural plant antimicrobial peptides. Frontiers in Microbiology, 8: 51-63.

Farkas, A., Ciataras, D. & Bocos, B. (2012). Biofilms impact on drinking water quality. In: Kostas Voudouris (Editor), Ecological Water Quality - Water Treatment and Reuse. InTechOpen. Pp. 142-162.

Fernandes, P.A.V., Arruda, I.R.d., Santos, Antônio Fernando Amatto Botelho dos, Araújo, A.A.d., Maior, A.M.S. & Ximenes, E.A. (2007). Antimicrobial activity of surfactants produced by Bacillus subtilis R14 against multidrug-resistant bacteria. Brazilian Journal of Microbiology, 38(4):704-709.

Filali, F.R., Zaid, A., Zekhnini, Z. & Frere, J.M. (1997). Thermophilic bacteria resistant to antibiotics in traditional public baths. Comparative Immunology, Microbiology and Infectious Diseases, 20(4):335-344.

Filimon, M.N., Nica, D.V., Ostafe, V., Bordean, D., Borozan, A.B., Vlad, D.C. & Popescu, R. (2013). Use of enzymatic tools for biomonitoring inorganic pollution in aquatic sediments: A case study (Bor, Serbia). Chemistry Central Journal, 7(1):1-14.

Fincan, S.A., Enez, B., Özdemir, S. & Bekler, F.M. (2014). Purification and characterization of thermostable α-amylase from thermophilic Anoxybacillus flavithermus. Carbohydrate Polymers, 102:144-150.

Finley, R.L., Collignon, P., Larsson, D.G., McEwen, S.A., Li, X.Z., Gaze, W.H., Reid-Smith, R., Timinouni, M., Graham, D.W. & Topp, E. (2013). The scourge of antibiotic resistance: The important role of the environment. Clinical Infectious Diseases: An Official Publication of the Infectious Diseases Society of America, 57(5):704-710.

Finoulst, I., Pinkse, M., Van Dongen, W. & Verhaert, P. (2011). Sample preparation techniques for the untargeted LC-MS-based discovery of peptides in complex biological matrices. Journal of Biomedicine & Biotechnology, 2011: Article ID 245291.

198

Fooladi, J. & Sajjadian, A. (2010). Screening the thermophilic and hyperthermophilic bacterial population of three Iranian hot-springs to detect the thermostable alpha-amylase producing strain. Iranian Journal of Microbiology, 2(1):46-50.

Francis, A.J. (1998). Biotransformation of uranium and other actinides in radioactive wastes. Journal of Alloys and Compounds, 271:78-84.

François, F., Lombard, C., Guigner, J., Soreau, P., Brian-Jaisson, F., Martino, G., Vandervennet, M., Garcia, D., Molinier, A. & Pignol, D. (2012). Isolation and characterization of environmental bacteria capable of extracellular biosorption of mercury. Applied and Environmental Microbiology, 78(4):1097-1106.

Galgale, A.D., Shah, N.B. & Shah, N.G. (2014). Treatment of wastewater containing high concentration of phenol and total dissolved solids in moving bed biofilm reactor. International Journal of Innovative Research in Science, Engineering and Technology, 3(4):10924-10930.

Galgóczy, L.N., Guba, M., Sajben-Nagy, E.I., Vörös, M., Takó, M. & Vágvölgyi, C. (2014). Antidermatophytic effect of Bacillus mojavensis SZMc 22228 and its secreted chymotrypsin- like protease. Acta Biologica Szegediensis, 58(2):157-162.

Galkiewicz, J.P. & Kellogg, C.A. (2008). Cross-kingdom amplification using bacteria-specific primers: Complications for studies of coral microbial ecology. Applied and Environmental Microbiology, 74(24):7828-7831.

Galtier, N. & Lobry, J.R. (1997). Relationships between genomic G C content, RNA secondary structures, and optimal growth temperature in prokaryotes. Journal of Molecular Evolution, 44(6):632-636.

Gao, P., Mao, D., Luo, Y., Wang, L., Xu, B. & Xu, L. (2012). Occurrence of sulfonamide and tetracycline-resistant bacteria and resistance genes in aquaculture environment. Water Research, 46(7):2355-2364.

Garcha, S., Verma, N. & Brar, S.K. (2016). Isolation, characterization and identification of microorganisms from unorganized dairy sector wastewater and sludge samples and evaluation of their biodegradability. Water Resources and Industry, 16:19-28.

Gautam, C.K., Srivastav, A.K., Bind, S., Madhav, M. & Shanthi, V. (2013). An insight into biofilm ecology and its applied aspects. International Journal of Pharmacy and Pharmaceutical Sciences, 5(4):69-73.

199

Gayathri, K.V. & Vasudevan, N. (2011). Enrichment of phenol degrading moderately halophilic bacterial consortium from saline environment. Journal of Bioremediation & Biodegradation, 1:104. doi:10.4172/2155-6199.1000104.

Genc, B., Nadaroglu, H., Adiguzel, A. & Baltaci, O. (2015). Purification and characterization of an extracellular cellulase from Anoxybacillus gonensis O9 isolated from geothermal area in Turkey. Journal of Environmental Biology, 36(6):1319.

Gerday, C. (2002). Extremophiles: Basic concepts. Knowledge for sustainable development.an insight into the encyclopedia of life support systems.UNESCO publishing/EOLSS Publishers, Oxford, pp. 1573-598.

Ghaderpour, A., Nasori, K.N.M., Chew, L.L., Chong, V.C., Thong, K.L. & Chai, L.C. (2014). Detection of multiple potentially pathogenic bacteria in Matang mangrove estuaries, Malaysia. Marine Pollution Bulletin, 83(1):324-330.

Ghalib, A.K., Yasin, M. & Faisal, M. (2014). Characterization and metal detoxification potential of moderately thermophilic Bacillus cereus from geothermal springs of Himalaya. Brazilian Archives of Biology and Technology, 57(4):554-560.

Ghaly, A.E., Ananthashankar, R., Alhattab, M. & Ramakrishnan, V.V. (2014). Production, characterization and treatment of textile effluents: A critical review. Journal of Chemical Engineering & Process Technology, 5:182-201.

Ghanney, N., Locantore, P., Nahdi, S., Ferchichi, A. & Iacobellis, N. S. (2016). Potential biocontrol effect of the phylloplane bacterium Bacillus mojavensis ABC7 on the olive knot disease. Journal of Plant Pathology and Microbiology, 7(3): e1000337.

Gholami, M., Etemadifar, Z. & Bouzari, M. (2015). Isolation a new strain of Kocuria rosea capable of tolerating extreme conditions. Journal of Environmental Radioactivity, 144:113-119.

Ghrairi, T., Chaftar, N., Jarraud, S., Berjeaud, J.M., Hani, K. & Frere, J. (2013). Diversity of legionellae strains from Tunisian hot spring water. Research in Microbiology, 164(4):342-350.

Giri, S.S., Harshiny, M., Sen, S.S., Sukumaran, V. & Park, S.C. (2015). Production and characterization of a thermostable bioflocculant from Bacillus subtilis F9, isolated from wastewater sludge. Ecotoxicology and Environmental Safety, 121:45-50.

Goh, K.M., Chua, Y.S. & Abdul, Raja Noor Zaliha Raja (2011). A comparison of conventional and miniprimer PCR to elucidate bacteria diversity in Malaysia Ulu Slim hot spring using 16S rDNA clone library. Romanian Biotechnological Letters, 16(3):8-17.

200

Goh, K.M., Kahar, U.M., Chai, Y.Y., Chong, C.S., Chai, K.P., Ranjani, V., Illias, R.M. & Chan, K. (2013). Recent discoveries and applications of Anoxybacillus. Applied Microbiology and Biotechnology, 97(4):1475-1488.

Gomaa, E. (2013). Antimicrobial activity of a biosurfactant produced by Bacillus licheniformis Strain M104 grown on whey. Brazilian Archives of Biology and Technology, 56(2):259-268.

Gong, M., Tang, C. & Zhu, C. (2014). Cloning and expression of delta-1-pyrroline-5- carboxylate dehydrogenase in Escherichia coli DH5α improves phosphate solubilization. Canadian Journal of Microbiology, 60(11):761-765.

Gonzalez, T., Terron, M. C., Yague, S., Junca, H., Carbajo, J. M., Zapico, E. J., Silva, R., Arana-Cuenca, A., Tellex, A. & Gonzalez A. E. (2008). Melanoidin containing wastewaters induce selective laccase gene expression in the white rot fungus Trametes sp. I 62. Research in Microbiology, 159:103-109.

Gouy, M., Guindon, S. & Gascuel, O. (2010). SeaView Version 4: A multiplatform graphical user interface for sequence alignment and phylogenetic tree building. Molecular Biology and Evolution, 27(2):221-224.

Gowri, R.S., Vijayaraghavan, R. & Meenambigai, P. (2014). Microbial degradation of reactive dyes - a review. International Journal of Current Microbiology and Applied Sciences, 3(3):421- 436.

Gudelji, M., Fruhwirth, G., Paar, A., Lottspeich, F., Robra, K., Cavaco-Paulo, A. & Gübitz, G. (2001). A catalase-peroxidase from a newly isolated thermoalkaliphilic Bacillus sp. with potential for the treatment of textile bleaching effluents. Extremophiles, 5(6):423-429.

Gulecal-Pektas, Y. (2016). Culture-dependent and culture-independent characterization of microbial communities in hot springs of Bursa, Turkey. Journal of Pure and Applied Microbiology, 10(2):1607-1613.

Güllüce, M., Karadayı, M. & Barış, Ö (2013). Bacteriocins: Promising natural antimicrobials. Microbial pathogens and strategies for combating them: Science, Technology and Education. Editor A Mendez-Vilas. Pp.1016-1028.

Guo, W., Zhang, J., Li, W., Xu, M. & Liu, S. (2015). Disruption of iron homeostasis and resultant health effects upon exposure to various environmental pollutants: A critical review. Journal of Environmental Sciences, 34:155-164.

201

Gupta, P., Manjula, A., Rajendhran, J., Gunasekaran, P. & Vakhlu, J. (2017). Comparison of Metagenomic DNA Extraction Methods for Soil Sediments of High Elevation Puga Hot Spring in Ladakh, India to Explore Bacterial Diversity. Geomicrobiology Journal, 34 (4): 289-299.

Gursahani, Y.H. & Gupta, S.G. (2011). Decolourization of textile effluent by a thermophilic bacteria Anoxybacillus rupiensis. Journal of Petroleum & Environmental Biotechnology, 2:111- 119.

Haddar, A., Agrebi, R., Bougatef, A., Hmidet, N., Sellami-Kamoun, A. & Nasri, M. (2009). Two detergent stable alkaline serine-proteases from Bacillus mojavensis A21: Purification, characterization and potential application as a laundry detergent additive. Bioresource Technology, 100(13):3366-3373.

Hammami, I., Rhouma, A., Jaouadi, B., Rebai, A. & Nesme, X. (2009). Optimization and biochemical characterization of a bacteriocin from a newly isolated Bacillus subtilis strain 14B for biocontrol of Agrobacterium spp. strains. Letters in Applied Microbiology, 48(2):253-260.

Hamzah, A., Sabturani, N. & Radiman, S. (2013a). Screening and optimization of bio surfactant production by the hydrocarbon-degrading bacteria. Sains Malaysiana, 42 (5):615-623.

Hamzah, Z., Rani, N.L.A., Saat, A. & Wood, A.K. (2013b). Determination of hot springs physico-chemical water quality potentially use for balneotherapy. Malaysian Journal of Analytical Sciences, 17(3):436-444.

Han, D. & Currell, M.J. (2017). Persistent organic pollutants in China's surface water systems. Science of the Total Environment, 580:602-625.

Handelsman, J. (2004). Metagenomics: Application of genomics to uncultured microorganisms. Microbiology and Molecular Biology Reviews: MMBR, 68(4):669-685.

Haque, M.M. & Muneer, M. (2007). TiO2-mediated photocatalytic degradation of a textile dye derivative, bromothymol blue, in aqueous suspensions. Dyes and Pigments, 75(2):443-448.

Harshada, K. (2014). Biosurfactant: A potent antimicrobial agent. Journal of Microbiology & Experimentation, 1:1-5.

Hartmann, M., Berditsch, M., Hawecker, J., Ardakani, M.F., Gerthsen, D. & Ulrich, A.S. (2010). Damage of the bacterial cell envelope by antimicrobial peptides gramicidin S and PGLa as revealed by transmission and scanning electron microscopy. Antimicrobial Agents and Chemotherapy, 54(8):3132-3142.

202

Hassan, M., Kjos, M., Nes, I.F., Diep, D.B. & Lotfipour, F. (2012). Natural antimicrobial peptides from bacteria: Characteristics and potential applications to fight against antibiotic resistance. Journal of Applied Microbiology, 113(4):723-736.

Hassan, M., Essam, T., Yassin, A.S. & Salama, A. (2014). Screening of bio-surfactant production ability among organic pollutants degrading isolates collected from Egyptian environment. Journal of Microbial & Biochemical Technology, 6:195-201.

Hassi, M., David, S., Haggoud, A., El Guendouzi, S., El Ouarti, A., Ibn Souda, S. & Iraqui, M. (2012). In vitro and intracellular antimycobacterial activity of a Bacillus pumilus strain. African Journal of Microbiology Research, 6 (10):2299-2304.

Hawumba, J.F., Theron, J. & Brözel, V.S. (2002). Thermophilic protease-producing Geobacillus from Buranga hot springs in western Uganda. Current Microbiology, 45(2):144- 150.

He, Y., Fan, K., Jia, C., Wang, Z., Pan, W., Huang, L., Yang, K. & Dong, Z. (2007). Characterization of a hyperthermostable fe-superoxide dismutase from hot spring. Applied Microbiology and Biotechnology, 75(2):367-376.

Hoffmann, C.E. & Rahn, O. (1944). The bactericidal and bacteriostatic action of crystal violet. Journal of Bacteriology, 47(2):177-187.

Holmes, C.B., Losina, E., Walensky, R.P., Yazdanpanah, Y. & Freedberg, K.A. (2003). Review of human immunodeficiency virus type 1-related opportunistic infections in sub-Saharan Africa. Clinical Infectious Diseases: An Official Publication of the Infectious Diseases Society of America, 36(5):652-662.

Hsu, B., Chen, C., Wan, M. & Cheng, H. (2006). Legionella prevalence in hot spring recreation areas of Taiwan. Water Research, 40(17):3267-3273.

Huang, S., Hsu, B., Wu, S., Fan, C., Shih, F., Lin, Y. & Ji, D. (2010). Water quality parameters associated with prevalence of Legionella in hot spring facility water bodies. Water Research, 44(16):4805-4811.

Hudson, C.L. (2012). Determination of antibiotic resistance in the bacterial community of boiling spring lake in Lassen Volcanic National Park. Masters of Science in Biology. [Unpublished]. Humboldt State University. USA.

Humudat, Y.R., Al-Naseri, S.K. & Kadhim, S.A. (2014). Production of highly efficient bacterial flocculant in water treatment. International Journal of Advanced Research, 2(9):297-301.

203

Hunter, C.J., Petrosyan, M., Ford, H.R. & Prasadarao, N.V. (2008). Enterobacter sakazakii: An emerging pathogen in infants and neonates. Surgical Infections, 9(5):533-539.

Husain, Q. (2010). Peroxidase mediated decolorization and remediation of wastewater containing industrial dyes: A review. Reviews in Environmental Science and BioTechnology, 9(2):117-140.

Huys, G., Cnockaert, M., Abbott, S.L., Janda, J.M. & Vandamme, P. (2010). Hafnia paralvei sp. nov., formerly known as Hafnia alvei hybridization group 2. International Journal of Systematic and Evolutionary Microbiology, 60(8):1725-1728.

Icgen, B. & Yilmaz, F. (2014). Co-occurrence of antibiotic and heavy metal resistance in Kızılırmak River isolates. Bulletin of Environmental Contamination and Toxicology, 93(6):735- 743.

Imanaka, T., Fujii, M. & Aiba, S. (1981). Isolation and characterization of antibiotic resistance plasmids from thermophilic bacilli and construction of deletion plasmids. Journal of Bacteriology, 146(3):1091-1097.

Inan, K., Canakci, S., Belduz, A.O. & Sahin, F. (2012). Brevibacillus aydinogluensis sp. nov., a moderately thermophilic bacterium isolated from Karakoc hot spring. International Journal of Systematic and Evolutionary Microbiology, 62(4):849-855.

Inyang, U.E., Bassey, E.N. & Inyang, J.D. (2012). Characterization of brewery effluent fluid. Journal of Engineering and Applied Sciences, 4:67-77.

Ishizaki, N., Sogawa, K., Inoue, H., Agata, K., Edagawa, A., Miyamoto, H., Fukuyama, M. & Furuhata, K. (2016). Legionella thermalis sp. nov., isolated from hot spring water in Tokyo, Japan. Microbiology and Immunology, 60(3):203-208.

Islam, M., Deora, A., Hashidoko, Y., Rahman, A., Ito, T. & Tahara, S. (2007). Isolation and identification of potential phosphate solubilizing bacteria from the rhizoplane of Oryza sativa L. cv. BR29 of Bangladesh. Zeitschrift für Naturforschung C, 62(1-2):103-110.

Issazadeh, K., Pahlaviani, M. & Massiha, A. (2011). Bioremediation of toxic heavy metals pollutants by Bacillus spp. isolated from Guilan Bay, north of Iran. In: Proceedings of the International Conference on Chemical, Biological and Environmental Engineering (ICBEE). The ICBEE 2011 Conference was held on 2-4 September 2011, Singapore.

Jack, R.W., Tagg, J.R. & Ray, B. (1995). Bacteriocins of Gram-positive bacteria. Microbiological Reviews, 59(2):171-200.

204

Janda, J.M. & Abbott, S.L. (2007). 16S rRNA gene sequencing for bacterial identification in the diagnostic laboratory: Pluses, perils, and pitfalls. Journal of Clinical Microbiology, 45(9):2761- 2764.

Janda, J.M. & Abbott, S.L. (2006). The genus Hafnia: From soup to nuts. Clinical Microbiology Reviews, 19(1):12-18.

Jaradat, Z.W., Al Mousa, W., Elbetieha, A., Al Nabulsi, A. & Tall, B.D. (2014). Cronobacter spp.--opportunistic food-borne pathogens. A review of their virulence and environmental- adaptive traits. Journal of Medical Microbiology, 63(8):1023-1037.

Jasim, B., Sreelakshmi, S., Mathew, J. & Radhakrishnan, E.K. (2016). Identification of endophytic Bacillus mojavensis with highly specialized broad spectrum antibacterial activity. 3 Biotech Journal, 6(2):187-197.

Jimat, D.N., Mohamed, I.B., Azmi, A.S. & Zainudin, Z. (2015). Isolation and characterization of thermophilic bacteria producing L-asparaginase from Malaysia hotspring and enzyme activity using different carbon and nitrogen sources. Journal of Applied Science and Agriculture, 10(5 (spe)):69-77.

Jyoti, V., Narayan, K.D. & Das, S.K. (2010). Gulbenkiania indica sp. nov., isolated from a sulfur spring. International Journal of Systematic and Evolutionary Microbiology, 60(5):1052- 1055.

Kabashima, Y., Sone, N., Kusumoto, T. & Sakamoto, J. (2013). Purification and characterization of malate: Quinone oxidoreductase from thermophilic Bacillus sp. PS3. Journal of Bioenergetics and Biomembranes, 45(1-2):131-136.

Kaczyńska, G., Borowik, A. & Wyszkowska, J. (2015). Soil dehydrogenases as an indicator of contamination of the environment with petroleum products. Water, Air, & Soil Pollution, 226(11):1-11.

Kadiri, D.D., Gorle, N., Peetala, K.V.R. & Peela, S. (2013). Isolation, screening and identification of phosphate solubilizing bacteria from different regions of Visakhapatnam and Araku Alley. International journal of Advanced Biotechnology and Research, 4:518-526.

Kaewklom, S. & Aunpad, R. (2012). The characterization of a novel bacteriocin with anti- MRSA activity produced by Bacillus sp. strain WARY9-1M. Thai Journal of Pharmacology Science, 36 (Sup):85-90.

205

Kafilzadeh & Mokhtari, S. (2013). Isolation and identification of phenol degrading bacteria from mangrove sediments in the Persian Gulf (Asaluyeh) and their growth kinetics assay. Biomedical & Pharmacology Journal, 6(2):189-196.

Kalyani, A., Naga Sireesha, G., Aditya, A., Girija Sankar, G. & Prabhakar, T. (2014). Production optimization of rhamnolipid biosurfactant by Streptomyces coelicoflavus (NBRC 15399T) using Plackett-Burman design. European Journal of Biotechnology and Bioscience, 1(5):7-13.

Kanamori, H., Weber, D.J. & Rutala, W.A. (2016). Healthcare outbreaks associated with a water reservoir and infection prevention strategies. Clinical Infectious Diseases: An Official Publication of the Infectious Diseases Society of America, 62(11):1423-1435.

Kao, P., Tung, M., Hsu, B., Hsueh, C., Chiu, Y., Chen, N., Shen, S. & Huang, Y. (2013). Occurrence and distribution of naegleria species from thermal spring environments in Taiwan. Letters in Applied Microbiology, 56(1):1-7.

Karafin, M., Romagnoli, M., Fink, D.L., Howard, T., Rau, R., Milstone, A.M. & Carroll, K.C. (2010). Fatal infection caused by Cupriavidus gilardii in a child with aplastic anemia. Journal of Clinical Microbiology, 48(3):1005-1007.

Karigar, C.S. & Rao, S.S. (2011). Role of microbial enzymes in the bioremediation of pollutants: A review. Enzyme Research, 2011: Article ID 805187. Doi:10.4061/2011/805187.

Karpagam, T. & Nagalakshmi, P.K. (2014). Isolation and characterization of phosphate solubilizing microbes from agricultural soil.International Journal of Current Microbiology and Applied Sciences, 3:601-614.

Karthiga, K. & Natarajan, K.A.(2015) Production and characterisation of bioflocculants for mineral processing applications. International Journal of Mineral Processing, 137:15-25.

Kavitha, V., Mandal, A.B. & Gnanamani, A. (2014). Microbial biosurfactant mediated removal and/or solubilization of crude oil contamination from soil and aqueous phase: An approach with Bacillus licheniformis MTCC 5514. International Biodeterioration & Biodegradation, 94:24- 30.

Khairuddin, N.F., Mulok, T.E., Khalil, K.A., Omar, W.S. & Saleh, S.H. (2016) Screening of medium with different range of waste frying oil, NaNO3, NaCl for biosurfactant production by thermophlic Anoxybacillus using fractional factoral design. In book: In CIEC 2015 Proceedings

206

of the International Civil and Infrastructure Engineering Conference. Springer publications. pp.9-19. doi:10.1007/978-981-10-0155-0_2.

Khalil, A.B., Anfoka, G.H. & Bdour, S. (2003). Isolation of plasmids present in thermophilic strains from hot springs in Jordan. World Journal of Microbiology and Biotechnology, 19(3):239-241.

Khan, R., Bhawana, P. & Fulekar, M.H. (2013). Microbial decolorization and degradation of synthetic dyes: A review. Reviews in Environmental Science and Bio/Technology, 12(1):75-97.

Kharayat, Y. (2012). Distillery wastewater: Bioremediation approaches. Journal of Integrative Environmental Sciences, 9(2):69-91.

Khiyami, M.A., Serour, E.A., Shehata, M.M. & Bahklia, A.H. (2012). Thermo-aerobic bacteria from geothermal springs in Saudi Arabia. African Journal of Biotechnology, 11(17):4053-4062.

Kilonzo-Nthenge, A., Rotich, E., Godwin, S., Nahashon, S. & Chen, F. (2012). Prevalence and antimicrobial resistance of Cronobacter sakazakii isolated from domestic kitchens in middle Tennessee, United States. Journal of Food Protection, 75(8):1512-1517.

Kim, B.S., Seo, J.R. & Park, D.H. (2013). Variation and characterization of bacterial communities contaminating two saunas operated at 64 ℃ and 76 ℃. Journal of Bacteriology and Virology, 43(3):195-203.

Kim, H., Nishiyama, M., Kunito, T., Senoo, K., Kawahara, K., Murakami, K. & Oyaizu, H. (1998). High population of Sphingomonas species on plant surface. Journal of Applied Microbiology, 85(4):731-736.

Kim, O.S., Cho, Y.J., Lee, K., Yoon, S.H., Kim, M., Na, H., Park, S.C., Jeon, Y.S., Lee, J.H., Yi, H., Won, S. & Chun, J. (2012). Introducing EzTaxon-e: A prokaryotic 16S rRNA gene sequence database with phylotypes that represent uncultured species. International journal of systematic and evolutionary microbiology, 62( 3):716-721.

Kimiran-Erdem, A., Arslan Aydoğdu, E.Ö, Gürün, S. & Altun, Ö (2015). Determination of multiple antibiotic and heavy metal resistance of the bacteria isolated from the Küçükçekmece lagoon, Turkey. Polish Journal of Environmental Studies, 24(3):1077-1084.

Kindoli, S., Lee, H.A., Heo, K. & Kim, J.H. (2012). Properties of a bacteriocin from Bacillus subtilis H27 isolated from Cheonggukjang. Food Science and Biotechnology, 21(6):1745-1751.

207

Kinge, C.N.W., Ateba, C.N. & Kawadza, D.T. (2010). Antibiotic resistance profiles of Escherichia coli isolated from different water sources in the Mmabatho locality, North West Province, South Africa. South African Journal of Science, 106(1-2):44-49.

Klapholz, A., Lessnau, K., Huang, B., Talavera, W. & Boyle, J.F. (1994). Hafnia alvei respiratory tract isolates in a community hospital over a three-year period and a literature review. CHEST Journal, 105(4):1098-1100.

Knapp, C.W., Dolfing, J., Ehlert, P.A. & Graham, D.W. (2009). Evidence of increasing antibiotic resistance gene abundances in archived soils since 1940. Environmental Science & Technology, 44(2):580-587.

Ko, K.S., Lee, N.Y., Oh, W.S., Lee, J.H., Ki, H.K., Peck, K.R. & Song, J. (2005). Tepidimonas arfidensis sp. nov., a novel Gram‐negative and thermophilic bacterium isolated from the bone marrow of a patient with leukemia in Korea. Microbiology and Immunology, 49(8):785-788.

Kobayashi, A., Tonozuka, T., Suyama, M., Sasaki, J., Nyamdawaa, B., Sakaguchi, M. & Sakano, Y. (2006). Molecular cloning and characterization of an enzyme hydrolyzing p- nitrophenyl α-D-glucoside from Bacillus stearothermophilus SA0301. Bioscience, Biotechnology, and Biochemistry, 70(2):495-499.

Kobayashi, M., Oana, K. & Kawakami, Y. (2014). Bath water contamination with Legionella and nontuberculous mycobacteria in 24-hour home baths, hot springs, and public bathhouses of Nagano Prefecture, Japan. Japanese Journal of Infectious Diseases, 67(4):276-281.

Kochher, M.S. & Kumar, J. (2011). Microbial decolourization of crystal violet by Bacillus subtilis. Biological Forum — An International Journal, 3:82-86.

Konuk, M., Ciğerci, İH. & Korcan, S.E. (2010). ALAD (δ-aminolevulinic acid dehydratase) as biosensor for Pb contamination. In: Vernon S. Somerset (Editor), Intelligent and Biosensors. InTech, Rijeka. pp. 363-376.

Korcan, S.E., Ciĝerci, İH. & Konuk, M. (2007). Screening of δ-aminolevulinic acid dehydratase from Pseudomonas strains as biosensor for lead and some other metals contamination. Environmental Monitoring and Assessment, 134(1-3):263-269.

Krumperman, P.H. (1983). Multiple antibiotic resistance indexing of Escherichia coli to identify high-risk sources of fecal contamination of foods. Applied and Environmental Microbiology, 46(1):165-170.

208

Kulkarni, S.J. & Kaware, D.J.P. (2013). Review on research for removal of phenol from wastewater. International Journal of Scientific and Research Publications, 3(4):1-4.

Kumar, A., Bisht, B.S. & Joshi, V.D. (2010). Biosorption of heavy metals by four acclimated microbial species, Bacillus spp., Pseudomonas spp., Staphylococcus spp. and Aspergillus niger. Journal of Biological and Environmental Sciences, 4(12):97-108.

Kumar, P., Patel, S.K., Lee, J. & Kalia, V.C. (2013). Extending the limits of Bacillus for novel biotechnological applications. Biotechnology Advances, 31(8):1543-1561.

Kurosawa, H., Fujita, M., Kobatake, S., Kimura, H., Ohshima, M., Nagai, A., Kaneko, S., Iwasaki, Y. & Kozawa, K. (2010). A case of legionella pneumonia linked to a hot spring facility in Gunma Prefecture, Japan. Japanese Journal of Infectious Diseases, 63(1):78-79.

Lai, K.K. (2001). Enterobacter sakazakii infections among neonates, infants, children, and adults: Case reports and a review of the literature. Medicine, 80(2):113-122.

Lambert, P.A. (2002). Cellular impermeability and uptake of biocides and antibiotics in gram‐positive bacteria and mycobacteria. Journal of Applied Microbiology, 92(s1):46S-54S.

Langevin, S., Vincelette, J., Bekal, S. & Gaudreau, C. (2011). First case of invasive human infection caused by Cupriavidus metallidurans. Journal of Clinical Microbiology, 49(2):744- 745.

Langner, E. & Rzeski, W. (2014). Biological properties of melanoidins: A review. International Journal of Food Properties, 17(2):344-353.

Lebedinsky, A.V., Chernyh, N.A. & Bonch-Osmolovskaya, E.A. (2007). Phylogenetic systematics of microorganisms inhabiting thermal environments. Biochemistry (Moscow), 72(12):1299-1312.

Leclerc, H. & Moreau, A. (2002). Microbiological safety of natural mineral water. FEMS microbiology reviews, 26(2):207-222.

Lee, E.M., Jeon, C.O., Choi, I., Chang, K.S. & Kim, C.J. (2005). Silanimonas lenta gen. nov., sp. nov., a slightly thermophilic and alkaliphilic gammaproteobacterium isolated from a hot spring. International Journal of Systematic and Evolutionary Microbiology, 55(1):385-389.

Lee, L.P., Karbul, H.M., Citartan, M., Gopinath, S.C., Lakshmipriya, T. & Tang, T. (2015). Lipase-secreting Bacillus species in an oil-contaminated habitat: Promising strains to alleviate oil pollution. BioMed Research International, 2015: Article ID 820575.

209

Lee, S. & Lee, S. (2001). Isolation and characterization of a thermotolerant bacterium Ralstonia sp. strain PHS1 that degrades benzene, toluene, ethylbenzene, and o-xylene. Applied Microbiology and Biotechnology, 56(1-2):270-275.

Leelakriangsak, M. & Borisut, S. (2012). Characterization of the decolorizing activity of azo dyes by Bacillus subtilis azoreductase AzoR1. Songklanakarin Journal of Science and Technology, 34(5):509-516.

Lemon, K.P., Earl, A.M., Vlamakis, H.C., Aguilar, C. & Kolter, R. (2008). Biofilm development with an emphasis on Bacillus subtilis. In: Romeo, Tony (Editor) Bacterial Biofilms. Springer Publishing Company, New York/Berlin/Heidelberg. pp. 1-16.

Lewis, D.A., Gumede, L.Y., van der Hoven, Louis A, de Gita, G.N., de Kock, E.J., de Lange, T., Maseko, V., Kekana, V., Smuts, F.P. & Perovic, O. (2013). Antimicrobial susceptibility of organisms causing community-acquired urinary tract infections in Gauteng Province, South Africa. SAMJ: South African Medical Journal, 103(6):377-381.

Li, A., Zhu, Y., Xu, L., Zhu, W. & Tian, X. (2008). Comparative study on the determination of assay for laccase of Trametes sp. African Journal of Biochemistry Research, 2(8):181-183.

Li, D., Yang, M., Hu, J., Zhang, J., Liu, R., Gu, X., Zhang, Y. & Wang, Z. (2009a). Antibiotic‐resistance profile in environmental bacteria isolated from penicillin production wastewater treatment plant and the receiving river. Environmental Microbiology, 11(6):1506- 1517.

Li, L., McCorkle, S.R., Monchy, S., Taghavi, S. & van der Lelie, D. (2009b). Bioprospecting metagenomes: Glycosyl hydrolases for converting biomass. Biotechnology for Biofuels, 2(1):10- 21.

Li, L., Zheng, B. & Liu, L. (2010). Biomonitoring and bioindicators used for river ecosystems: Definitions, approaches and trends. Procedia Environmental Sciences, 2:1510-1524.

Li, Y., Niu, D., Zhang, L., Wang, Z. & Shi, G. (2013). Purification, characterization and cloning of a thermotolerant isoamylase produced from Bacillus sp. CICIM 304. Journal of Industrial Microbiology & Biotechnology, 40(5):437-446.

Lim, K.B., Balolong, M.P., Kim, S.H., Oh, J.K., Lee, J.Y. & Kang, D. (2016). Isolation and characterization of a broad spectrum bacteriocin from Bacillus amyloliquefaciens RX7. BioMed Research International, 2016: Article ID 8521476. Doi:10.115/2016/8521476.

210

Lim, J.W., Ha, D., Lee, J., Lee, S.K. & Kim, T. (2015a). Review of micro/nanotechnologies for microbial biosensors. Frontiers in Bioengineering and Biotechnology, 3:61-73.

Lim, Y.L., Chan, K., Ee, R., Belduz, A.O., Canakci, S., Kahar, U.M., Yaakop, A.S. & Goh, K.M. (2015b). Complete genome of the potential thermozyme producer Anoxybacillus gonensis G2 T isolated from the Gönen hot springs in Turkey. Journal of Biotechnology, 212:65-66.

Lima-Bittencourt, C.I., Cursino, L., Gonçalves-Dornelas, H., Pontes, D.S., Nardi, R., Callisto, M., Chartone-Souza, E. & Nascimento, A. (2007). Multiple antimicrobial resistance in Enterobacteriaceae isolates from pristine freshwater. Genetics and Molecular Research, 6(3):510-521.

Limkhuansuwan, V. & Chaiprasert, P. (2010). Decolorization of molasses melanoidins and palm oil mill effluent phenolic compounds by fermentative lactic acid bacteria. Journal of Environmental Sciences, 22(8):1209-1217.

Lin, J., Biyela, P.T. & Puckree, T. (2004). Antibiotic resistance profiles of environmental isolates from Mhlathuze River, KwaZulu-Natal (RSA). Water SA, 30(1):23-28.

Liu, C., Yuan, H., Yang, J. & Li, B. (2011). Effective biosorption of reactive blue 5 by pH- independent lyophilized biomass of Bacillus megaterium. African Journal of Biotechnology, 10(73):16626-16636.

Liu, J., Mbadinga, S.M., Yang, S., Gu, J. & Mu, B. (2015a). Chemical structure, property and potential applications of biosurfactants produced by Bacillus subtilis in petroleum recovery and spill mitigation. International Journal of Molecular Sciences, 16(3):4814-4837.

Liu, X., Li, Y., Cai, L., Li, X., chen, J. & Lyu, S. (2015b). Isolation and characterisation of Bacillus spp. antagonistic to Vibrio parahaemolyticus for use as probiotics in aquaculture. World Journal of Microbiology and Biotechnology, 31:795-803

Liu Y, Kumar NR, Lai Q, Du J, Dobritsa AP, Samadpour M, et al. (2015c) Identification of strains Bacillus aerophilus MTCC 7304T as Bacillus altitudinis and Bacillus stratosphericus MTCC 7305T as a Proteus sp. and the status of the species Bacillus aerius Shivaji et al. 2006. Request for an opinion. International Journal of Systematic and Evolutionary Microbiology, 65(9):3228-3231.

Liu, S., Zeng, G., Niu, Q., Liu, Y., Zhou, L., Jiang, L., Tan, X., Xu, P., Zhang, C. & Cheng, M. (2017). Bioremediation mechanisms of combined pollution of PAHs and heavy metals by bacteria and fungi: A mini review. Bioresource technology, 224:25-33.

211

Logan, N.A. (2012). Bacillus and relatives in foodborne illness. Journal of Applied Microbiology, 112(3):417-429.

Logar, R.M. & Vodovnik, M. (2007). The application of microbes in environmental monitoring. Communicating Current Research and Educational Topics and Trends in Applied Microbiology, 577-585.

López-López, O., Cerdán, M.E. & González-Siso, M.I. (2013). Hot spring metagenomics. Life, 3(2):308-320.

Lu, L., Wang, J., Xu, Y., Wang, K., Hu, Y., Tian, R., Yang, B., Lai, Q., Li, Y. & Zhang, W. (2014). A high-resolution LC-MS-based secondary metabolite fingerprint database of marine bacteria. Scientific Reports, 4:6537-6544.

Lu, X., Zhang, T. & Fang, H.H. (2011). Bacteria-mediated PAH degradation in soil and sediment. Applied Microbiology and Biotechnology, 89(5):1357-1371.

Ludwig, W., Schleifer, K. & Whitman, W.B. (2009) Revised road map to the phylum Firmicutes. In: De Vos, P., Garrity, G., Jones, D., Krieg, N.R., Ludwig, W., Rainey, F.A., Schleifer, K.-H., Whitman, W. (Editors), Bergey’s Manual of Systematic Bacteriology, 2nd edition, Volume 3, The Firmicutes. Springer-Verlag Inc. pp. 1-13.

Luna, V.A., King, D.S., Gulledge, J., Cannons, A.C., Amuso, P.T. & Cattani, J. (2007). Susceptibility of Bacillus anthracis, Bacillus cereus, Bacillus mycoides, Bacillus pseudomycoides and Bacillus thuringiensis to 24 antimicrobials using sensititre automated microbroth dilution and etest agar gradient diffusion methods. The Journal of Antimicrobial Chemotherapy, 60(3):555-567.

Lund, J.W. (2001). Balneological use of thermal waters. Proceedings of International Geothermal Days, “Germany 2001”, Bad Urach, Germany, 17-22 September 2001.

Lutz, J.K. & Lee, J. (2011). Prevalence and antimicrobial-resistance of Pseudomonas aeruginosa in swimming pools and hot tubs. International Journal of Environmental Research and Public Health, 8(2):554-564.

Ma, E.S., Wong, C.L., Lai, K.T., Chan, E.C., Yam, W.C. & Chan, A.C. (2005). Kocuria kristinae infection associated with acute cholecystitis. BMC Infectious Diseases, 5:60-63.

Ma, Z. & Hu, J. (2015). Production and characterization of surfactin-type lipopeptides as bioemulsifiers produced by a pinctada martensii-derived Bacillus mojavensis B0621A. Applied Biochemistry and Biotechnology, 177(7):1520-1529.

212

Ma, Z., Wang, N., Hu, J. & Wang, S. (2012). Isolation and characterization of a new iturinic lipopeptide, mojavensin A produced by a marine-derived bacterium Bacillus mojavensis B0621A. The Journal of Antibiotics, 65(6):317-322.

Maartens, G. (2002). Opportunistic infections associated with HIV infection in Africa. Oral Diseases, 8(s2):76-79.

Mackenzie, R., Pedrós-Alió, C. & Díez, B. (2013). Bacterial composition of microbial mats in hot springs in northern Patagonia: Variations with seasons and temperature. Extremophiles, 17(1):123-136.

Magesh, H., Kumar, A., Alam, A., Priyam, Sekar, U., Sumantran, V.N. & Vaidyanathan, R. (2013). Identification of natural compounds which inhibit biofilm formation in clinical isolates of Klebsiella pneumoniae. Indian Journal of Experimental Biology, 51(9):764-772.

Magiorakos, A., Srinivasan, A., Carey, R.B., Carmeli, Y., Falagas, M.E., Giske, C.G., Harbarth, S., Hindler, J.F., Kahlmeter, G. & Olsson‐Liljequist, B. (2012). Multidrug‐resistant, extensively drug‐resistant and pandrug‐resistant bacteria: An international expert proposal for interim standard definitions for acquired resistance. Clinical Microbiology and Infection, 18(3):268- 281.

Magulova, K. & Priceputu, A. (2016). Global monitoring plan for persistent organic pollutants (POPs) under the Stockholm convention: Triggering, streamlining and catalyzing global POPs monitoring. Environmental Pollution, 21:782-84.

Mahajan, G.B. & Balachandran, L. (2016). Sources of antibiotics: Hot springs. Biochemical Pharmacology. doi.org/10.1016/j.bcp.2016.11.021.

Maheswari, N. & Sivagami, S. (2016). Biological degradation of textile dyes using marine Bacillus species. International Journal of Pure and Applied Bioscience 4 (4): 123-128.

Mahgoub, S., Tsioptsias, K., Likotrafiti, E. & Samaras, P. (2014). Removal of melanoidins by potential manganese peroxidase producing bacteria isolated from molasses effluent. Proceedings from Symbiosis International Conference 2014. Athens, Greece.

Maier, J., Kandelbauer, A., Erlacher, A., Cavaco-Paulo, A. & Gubitz, G.M. (2004). A new alkali-thermostable azoreductase from Bacillus sp. strain SF. Applied and Environmental Microbiology, 70(2):837-844.

Mandal, S. & Mandal, M.D. (2015). Can bacteria subsist on antibiotics? Journal of Forensic Toxicology and Pharmacology, 4:2-3.

213

Mandic-Mulec, I., Stefanic, P. & van Elsas, J.D. (2015). Ecology of Bacillaceae. Microbiology Spectrum, 3(2): TBS-0017-2013. doi:10.1128/micrbiolspec.TBS-0017—2013.

Mann, S. & Chen, Y.P. (2010). Bacterial genomic G C composition-eliciting environmental adaptation. Genomics, 95(1):7-15.

Marais, B.J., Rabie, H., Schaaf, S.H. & Cotton, M.F. (2006). updAIDS review: Co-mmon opportunistic infections in HIV infected infants and children, Part 1- respiratory infections. South African Family Practice, 48(10):52-56.

Marti, M.E., Colonna, W.J., Patra, P., Zhang, H., Green, C., Reznik, G., Pynn, M., Jarrell, K., Nyman, J.A. & Somasundaran, P. (2014). Production and characterization of microbial biosurfactants for potential use in oil-spill remediation. Enzyme and Microbial Technology, 55:31-39.

Martinez, J.L. (2012). Natural antibiotic resistance and contamination by antibiotic resistance determinants: The two ages in the evolution of resistance to antimicrobials. Frontiers in Microbiology, 3:1-3.

Martinez, J.L. (2009a). The role of natural environments in the evolution of resistance traits in pathogenic bacteria. Proceedings of the Royal Society B: Biological Sciences, 276(1667):2521- 2530.

Martinez, J.L. (2009b). Environmental pollution by antibiotics and by antibiotic resistance determinants. Environmental Pollution, 157(11):2893-2902.

Martirani, L., Varcamonti, M., Naclerio, G. & De Felice, M. (2002). Purification and partial characterization of bacillocin 490, a novel bacteriocin produced by a thermophilic strain of Bacillus licheniformis. Microbial Cell Factories, 1(1):1-5.

Mary, P., Defives, C. & Hornez, J.P. (2000). Occurrence and multiple antibiotic resistance profiles of non-fermentative Gram-negative microflora in five brands of non-carbonated French bottled spring water. Microbial Ecology, 39(4):322-329.

Mathur, H., Rea, M.C., Cotter, P.D., Ross, R.P. & Hill, C. (2014). The potential for emerging therapeutic options for Clostridium difficile infection. Gut Microbes, 5(6):696-710.

Matongo, S., Birungi, G., Moodley, B. & Ndungu, P. (2015). Pharmaceutical residues in water and sediment of Msunduzi River, KwaZulu-Natal, South Africa. Chemosphere, 134:133-140.

214

Matpan Bekler, F. & Güven, K. (2014). Isolation and production of thermostable α-amylase from thermophilic Anoxybacillus sp. KP1 from Diyadin hot spring in Ağri, Turkey. Biologia, 69(4):419-427.

Matyar, F., Gülnaz, O., Guzeldag, G., Mercimek, H.A., Akturk, S., Arkut, A. & Sumengen, M. (2014). Antibiotic and heavy metal resistance in Gram-negative bacteria isolated from the Seyhan Dam Lake and Seyhan River in Turkey. Annals of Microbiology, 64(3):1033-1040.

Maughan, H. & Van der Auwera, G. (2011). Bacillus taxonomy in the genomic era finds phenotypes to be essential though often misleading. Infection, Genetics and Evolution, 11(5):789-797.

Mawadza, C. & Zvauya, R. (1996). Some factors affecting endo‐β‐1, 4‐glucanase production by two Bacillus strains isolated from Zimbabwean hot springs. Journal of Basic Microbiology, 36(3):177-185.

McGinnis, S. & Madden, T.L. (2004). BLAST: At the core of a powerful and diverse set of sequence analysis tools. Nucleic Acids Research, 32(Web Server issue):20.

Meric, M., Willke, A., Kolayli, F., Yavuz, S. & Vahaboglu, H. (2009). Water-borne Sphingomonas paucimobilis epidemic in an intensive care unit. Journal of Infection, 58(3):253- 255.

Meyer-Dombard, D.R., Shock, E.L. & Amend, J.P. (2012). Effects of trace element concentrations on culturing thermophiles. Extremophiles, 16(2):317-331.

Miller, R.V., Gammon, K. & Day, M.J. (2009). Antibiotic resistance among bacteria isolated from seawater and penguin fecal samples collected near Palmer Station, Antarctica. Canadian Journal of Microbiology, 55(1):37-45.

Min, K., Gong, G., Woo, H.M., Kim, Y. & Um, Y. (2015). A dye-decolorizing peroxidase from Bacillus subtilis exhibiting substrate-dependent optimum temperature for dyes and β-ether lignin dimer. Nature Scientific Reports, 5:8245.

Mittal, A., Kaur, D., Malviya, A., Mittal, J. & Gupta, V.K. (2009). Adsorption studies on the removal of coloring agent phenol red from wastewater using waste materials as adsorbents. Journal of Colloid and Interface Science, 337(2):345-354.

Miyagi, A., Nabetani, H. & Nakajima, M. (2013). Decolorization of japanese soy sauce (shoyu) using adsorption. Journal of Food Engineering, 116(3):749-757.

215

Moe, N.K.T., Thwe, S.M., Suzuki, K., Nakai, R., Terahara, T., Imada, C. & Kobayashi, T. (2015). Production of an antibacterial substance by Bacillus mojavensis strain F412 isolated from a Myanmar shrimp product fermented with boiled rice. Fisheries Science, 81(4):795-802.

Molobela, I.P. (2010). Protease and amylase enzymes for biofilm removal and degradation of extracellular polymeric substances (EPS) produced by Pseudomonas fluorescens bacteria. African Journal of Microbiology Research 4 (14):1515-1524.

Moreira, C., Rainey, F.A., Nobre, M.F., da Silva, M.T. & da Costa, M.S. (2000). Tepidimonas ignava gen. nov., sp. nov., a new chemolithoheterotrophic and slightly thermophilic member of the beta-Proteobacteria. International Journal of Systematic and Evolutionary Microbiology, 50 (2):735-742.

Morikawa, M., Kagihiro, S., Haruki, M., Takano, K., Branda, S., Kolter, R. & Kanaya, S. (2006). Biofilm formation by a Bacillus subtilis strain that produces gamma polyglutamate. Microbiology, 152:2801-2807.

Moyer, C.L., Tiedje, J.M., Dobbs, F.C. & Karl, D.M. (1996). A computer-simulated restriction fragment length polymorphism analysis of bacterial small-subunit rRNA genes: Efficacy of selected tetrameric restriction enzymes for studies of microbial diversity in nature. Applied and Environmental Microbiology, 62(7):2501-2507.

Mudryk, Z.J. & Podgorska, B. (2006). Enzymatic activity of bacterial strains isolated from marine beach sediments. Polish Journal of Environmental Studies, 15(3):441-448.

Mudryk, Z., Gackowska, J., Skórczewski, P., Perliński, P. & Zdanowicz, M. (2014). Occurrence of potentially human pathogenic bacteria in the seawater and in the sand of the recreational coastal beach in the southern Baltic sea. Oceanological and Hydrobiological Studies, 43(4):366- 373.

Munster, M.J., Munster, A.P. & Sharp, R.J. (1985). Incidence of plasmids in Thermus spp. isolated in Yellowstone National Park. Applied and Environmental Microbiology, 50(5):1325- 1327.

Mustapha, M.U. & Halimoon, N. (2015). Screening and isolation of heavy metal tolerant bacteria in industrial effluent. Procedia Environmental Sciences, 30:33-37.

Muto, A. & Osawa, S. (1987). The guanine and cytosine content of genomic DNA and bacterial evolution. Proceedings of the National Academy of Sciences of the United States of America, 84(1):166-169.

216

Naclerio, G., Ricca, E., Sacco, M. & De Felice, M. (1993). Antimicrobial activity of a newly identified bacteriocin of Bacillus cereus. Applied and Environmental Microbiology, 59(12):4313-4316.

Nagórska, K., Bikowski, M. & Obuchowski, M. (2007). Multicellular behaviour and production of a wide variety of toxic substances support usage of Bacillus subtilis as a powerful biocontrol agent. Acta Biochimica Polonica - English Edition, 54(3):495-508.

Naicker, K., Cukrowska, E. & McCarthy, T.S. (2003). Acid mine drainage arising from gold mining activity in Johannesburg, South Africa and environs. Environmental Pollution, 122(1):29-40.

Namsaraev, Z.B., Babasanova, O.B., Dunaevsky, Y.E., Akimov, V.N., Barkhutova, D.D., Gorlenko, V.M. & Namsaraev, B.B. (2010). Anoxybacillus mongoliensis sp. nov., a novel thermophilic proteinase producing bacterium isolated from alkaline hot spring, central Mongolia. Microbiology, 79(4):491-499.

Narasaki, R., Kuribayashi, H., Shimizu, K., Imamura, D., Sato, T. & Hasumi, K. (2005). Bacillolysin MA, a novel bacterial metalloproteinase that produces angiostatin-like fragments from plasminogen and activates protease zymogens in the coagulation and fibrinolysis systems. The Journal of Biological Chemistry, 280(14):14278-14287.

Narayan, V.V., Hatha, M.A., Morgan, H.W. & Rao, D. (2008). Isolation and characterization of aerobic thermophilic bacteria from the Savusavu hot springs in Fiji. Microbes and Environment, 23(4):350-352.

Narayan, R.T. & Veeranna, K. D. (2014). Visual detection and determination of melamine using synthetic dyes. Journal of Applied Chemistry Article ID 457254. doi.org/10.1155/2014/457254

Nasermoaddeli, A. & Kagamimori, S. (2005). Balneotherapy in medicine: A review. Environmental Health and Preventive Medicine, 10(4):171-179.

Natarajan, K.A. (2015). Production and characterization of bioflocculants for mineral processing applications. International Journal of Mineral Processing, 137:15-25.

Nath, S., Chowdhury, S. & Dora, K.C. (2015). Application of Bacillus sp. as a biopreservative for food preservation. Journal of Engineering Research and Application, 5(4):85-95.

Nel, L.H. & Markotter, W. (2004). Emerging Infectious Waterborne Diseases: Bacterial Agents. Microbial Waterborne Pathogens. IWA Publishing, London, United Kingdom.

217

Nigam, P.S. (2013). Microbial enzymes with special characteristics for biotechnological applications. Biomolecules, 3(3):597-611.

Nitschke, M. & Pastore, G.M. (2006). Production and properties of a surfactant obtained from Bacillus subtilis grown on cassava wastewater. Bioresource Technology, 97(2):336-341.

Nnenna, F., Lekiah, P. & Obemeata, O. (2011). Degradation of antibiotics by bacteria and fungi from the aquatic environment. Journal of Toxicology and Environmental Health Sciences, 3(10):275-285.

Nontembiso, P., Sekelwa, C., Leonard, M.V. & Anthony, O.I. (2011). Assessment of bioflocculant production by Bacillus sp. gilbert, a marine bacterium isolated from the bottom sediment of Algoa Bay. Marine Drugs, 9(7):1232-1242.

Nourbakhsh, M. N., Kilicarslan, S., Ilhan, S. & Ozdag, H. (2002). Biosorption of Cr, Pb and Cu ions in industrial wastewater on Bacillus sp. Chemical Engineering Journal, 85:351-355.

Novita, E. (2016). Biodegradability simulation of coffee wastewater using instant coffee. Agriculture and Agricultural Science Procedia, 9:217-229.

Nyenje, P.M., Foppen, J.W., Uhlenbrook, S., Kulabako, R. & Muwanga, A. (2010). Eutrophication and nutrient release in urban areas of sub-Saharan Africa - a review. Science of the Total Environment, 408(3):447-455.

O’Connor, P.M., Ross, R.P., Hill, C. & Cotter, P.D. (2015). Antimicrobial antagonists against food pathogens: A bacteriocin perspective. Current Opinion in Food Science, 2:51-57.

Obeidat, M., Khyami-Horani, H., Al-Zoubi, A. & Otri, I. (2012). Isolation, characterization, and hydrolytic activities of Geobacillus species from Jordanian hot springs. African Journal of Biotechnology, 11(25):6763-6768.

Ogawa, T., Shibata, M., Yatome, C. & Idaka, E. (1988). Growth inhibition of Bacillus subtilis by basic dyes. Bulletin of Environmental Contamination and Toxicology, 40(4):545-552.

Ogbolosingha, A.J., Essien, E. & Ohiri, R. (2015). Variation of lipase, catalase and dehydrogenase activites during bioremediaton of crude oil polluted soil. Journal of Environment and Earth Science, 5(14):128-142.

Ogunlaja, A., Olomo, O.O. & Olukanni, D.O. (2013). Aerobic bacterial degraders in effluent from Itoku textile industry, Abeokuta. African Journal of Biotechnology, 12(50):6954-6958.

218

Ogunseitan, O.A., Yang, S. & Ericson, J. (2000). Microbial δ-aminolevulinate dehydratase as a biosensor of lead bioavailability in contaminated environments. Soil Biology and Biochemistry, 32(13):1899-1906.

Okaiyeto, K., Nwodo, U.U., Mabinya, L.V. & Okoh, A.I. (2015). Bacillus toyonensis strain AEMREG6, a bacterium isolated from South African marine environment sediment samples produces a glycoprotein bioflocculant. Molecules, 20(3):5239-5259.

Okonkwo, J.O. & Mothiba, M. (2005). Physico-chemical characteristics and pollution levels of heavy metals in the rivers in Thohoyandou, South Africa. Journal of Hydrology, 308(1):122- 127.

Oliveira, J.G.d. & Garcia-Cruz, C.H. (2013). Properties of a biosurfactant produced by Bacillus pumilus using vinasse and waste frying oil as alternative carbon sources. Brazilian Archives of Biology and Technology, 56(1):155-160.

Olivier, J., Venter, J.S. & Jonker, C.Z. (2011). Thermal and chemical characteristics of hot water springs in the northern part of the Limpopo Province, South Africa. Water SA, 37(4):427- 436.

Ong, S., Keng, P., Lee, W., Ha, S. & Hung, Y. (2011). Dye waste treatment. Water, 3(1):157- 176.

Onuoha, E., Okafor, C. & Aduo, B. (2016). Antibiotic and heavy metal tolerance of bacterial pathogens isolated from agricultural soil. World Journal of Medical Science, 13 (4):236-241.

Onwurah, I. & Eze, M.O. (2000). Superoxide dismutase activity in Azotobacter vinelandii in the disposition of environmental toxicants exemplified by Fenton reagent and crude oil. Toxic Substance Mechanisms, 19(2):111-123.

O’Toole, G. (2011). Microtiter dish biofilm formation assay. Journal of Visualized Experiments 47. doi: 10.3791/2437

Ou, J. & Zhu, M. (2012). An overview of current and novel approaches for microbial neutral protease improvement. International Journal of Modern Biology and Medicine, 2:1-31.

Oves, M. & Hussain, F. (2016). Antibiotics and heavy metal resistance emergence in waterborne bacteria. Journal of Investigative Genomics, 3(2):45-48.

Ozdemir, S., Kilinc, E., Poli, A., Nicolaus, B. & Guven, K. (2012). Cd, Cu, Ni, Mn and Zn resistance and bioaccumulation by thermophilic bacteria, Geobacillus toebii subsp decanicus

219

and Geobacillus thermoleovorans subsp stromboliensis. World Journal Microbiology and Biotechnology, 28:155-163.

Ozdemir, S., Kilinic, E., Poli, A. & Nicolaus, B. (2013). Biosorption of heavy metals (Cd, Cu, Co and Mn) by thermophilic bacteria, Geobacillus thermantarcticus and Anoxybacillyus amylolyticus: equilibrium and kinetic studies. Bioremediation Journal, 17(2):86-96.

Pacwa-Płociniczak, M., Płaza, G.A., Piotrowska-Seget, Z. & Cameotra, S.S. (2011). Environmental applications of biosurfactants: Recent advances. International Journal of Molecular Sciences, 12(1):633-654.

Padmavathi, A.R. & Pandian, S.K. (2014). Antibiofilm activity of biosurfactant producing coral associated bacteria isolated from Gulf of Mannar. Indian Journal of Microbiology, 54(4):376- 382.

Pagaling, E., Grant, W.D., Cowan, D.A., Jones, B.E., Ma, Y., Ventosa, A. & Heaphy, S. (2012). Bacterial and archaeal diversity in two hot spring microbial mats from the geothermal region of Tengchong, China. Extremophiles, 16(4):607-618.

Paisio, C.E., Quevedo, M.R., Talano, M.A., González, P.S. & Agostini, E. (2014). Application of two bacterial strains for wastewater bioremediation and assessment of phenolics biodegradation. Environmental Technology, 35(14):1802-1810.

Pakpitcharoen, A., Potivejkul, K., Kanjanavas, P., Areekit, S. & Chansiri, K. (2008). Biodiversity of thermotolerant Bacillus sp. producing biosurfactants, biocatalysts, and antimicrobial agents. Science Asia, 34(4):424-431.

Panda, A.K., Bisht, S.S., DeMondal, S., Kumar, N.S., Gurusubramanian, G. & Panigrahi, A.K. (2014). Brevibacillus as a biological tool: A short review. Antonie van Leeuwenhoek, 105(4):623-639.

Panda, A.K., Bisht, S.S., Mandal, S. & Kumar, N.S. (2016). Bacterial and archeal community composition in hot springs from Indo-Burma region, north-east India. AMB Express, 6(1):111- 123.

Panda, M.K., Sahu, M.K. & Tayung, K. (2013). Isolation and characterization of a thermophilic Bacillus sp. with protease activity isolated from hot spring of Tarabalo, Odisha, India. Iranian Journal of Microbiology, 5(2):159-165.

220

Pandey, A., Dhakar, K., Sharma, A., Priti, P., Sati, P. & Kumar, B. (2015). Thermophilic bacteria that tolerate a wide temperature and pH range colonize the Soldhar (95 °C) and Ringigad (80 °C) hot springs of Uttarakhand, India. Annals of Microbiology, 65(2):809-816.

Pandey, P.K., Kass, P.H., Soupir, M.L., Biswas, S. & Singh, V.P. (2014). Contamination of water resources by pathogenic bacteria. AMB Express, 4:51-67.

Panosyan, H. & Birkeland, N. (2014). Microbial diversity in an Armenian geothermal spring assessed by molecular and culture‐based methods. Journal of Basic Microbiology, 54(11):1240- 1250.

Papadopoulou, C., Economou, V., Sakkas, H., Gousia, P., Giannakopoulos, X., Dontorou, C., Filioussis, G., Gessouli, H., Karanis, P. & Leveidiotou, S. (2008). Microbiological quality of indoor and outdoor swimming pools in Greece: Investigation of the antibiotic resistance of the bacterial isolates. International Journal of Hygiene and Environmental Health, 211(3):385-397.

Parthuisot, N., West, N.J., Lebaron, P. & Baudart, J. (2010). High diversity and abundance of Legionella spp. in a pristine river and impact of seasonal and anthropogenic effects. Applied and Environmental Microbiology, 76(24):8201-8210.

Patel, I., Patel, V., Thakkar, A. & Kothari, V. (2013). Microbial biofilms: Microbes in social mode. International Journal of Biotechnology Research and Practice (IJBRP), 1(1): 19-34.

Pathak, A.P. & Rathod, M.G. (2015). Cultivable bacterial diversity of terrestrial thermal spring of Unkeshwar, India. Journal of Biochemical Technology, 5(4):814-818.

Paul, M., Gupta, R., Khushwaha, S. & Thakur, R. (2015). Kocuria rosea: An emerging pathogen in acute bacterial meningitis-case report. Journal of Microbiology and Antimicrobial Agents, 1(1): 4-7.

Pearce, C.I., Lloyd, J.R. & Guthrie, J.T. (2003). The removal of colour from textile wastewater using whole bacterial cells: A review. Dyes and Pigments, 58(3):179-196.

Peixoto, R.S., Vermelho, A.B. & Rosado, A.S. (2011). Petroleum-degrading enzymes: Bioremediation and new prospects. Enzyme Research, 2011: Article ID475193. doi:10.4061/2011/475193.

Pereira, J., Gudina, E., Costa, R., Vitorino, R., Teixeira, J., Coutinho, J & Rodrigues, L. R. (2013). Optimization and characterisation of biosurfactant production by Bacillus subtilis isolates towards microbial enhanced oil recovery applications. Fuel, 111:259-268.

221

Perestrelo, M.F., Norberg, A.N., Guerra-Sanches, F., Torres, A.C. & Pile, E. (2006). Microbiological analysis of water used in hydrotherapy. Journal of Venomous Animals and Toxins including Tropical Diseases, 12(3):418-422.

Perez, K., Viana, J., Lopes, F., Pereira, J, Dos Santos, E., Oliveira, J, Velho, R., Crispim, S., Nicoli, J, Brandelli, A. & Nardi, R. (2017). Bacillus spp. isolated from puba as an source of biosurfactants and antimicrobial lipopeptides. Frontiers in Microbiology, 8: 61. Doi:10.3389/fmicb.2017.00061.

Perron, G., Whyte, L., Turnbaugh, P., Hanage, W.P., Dantas, G. & Desai, M.M. (2013). Functional characterization of bacteria isolated from ancient arctic soil exposes diverse resistance mechanisms to modern antibiotics. PloS One,10(3):e0069533.

Piccini, C., Conde, D., Alonso, C., Sommaruga, R. & Pernthaler, J. (2006). Blooms of single bacterial species in a coastal lagoon of the southwestern Atlantic Ocean. Applied and Environmental Microbiology, 72(10):6560-6568.

Płaza, G., Chojniak, J., Rudnicka, K., Paraszkiewicz, K. & Bernat, P. (2015). Detection of biosurfactants in Bacillus species: Genes and products identification. Journal of Applied Microbiology, 119(4):1023-1034.

Pokharia, A. & Ahluwalia, S.S. (2013). Isolation and screening of dye decolorizing bacterial isolates from contaminated sites. Textiles and Light Industrial Science and Technology, 2(2):54- 62.

Poi, G., Aburto-Medina, A., Mok, P., Ball, AS. & Shasavari, E. (2017). Bioremediation of phenol-contaminated industrial wastewater using a bacterial consortium-from laboratory to field. Water, Air & Soil Pollution, 228:89

Poli, A., Nicolaus, B., Chan, K.G., Kahar, U.M., Chan, C.S. & Goh, K.M. (2015). Genome sequence of Anoxybacillus thermarum AF/04T, isolated from the Euganean hot springs in Abano Terme, Italy. Genome Announcements, 3(3):15-16.

Pradeep, N.V., Anupama, S., Navya, K., Shalini, H.N., Idris, M. & Hampannavar, U.S. (2015). Biological removal of phenol from wastewaters: A mini review. Applied Water Science, 5(2):105-112.

222

Proia, L., von Schiller, D., Snchez-Melsi, A., Sabater, S., Borrego, C.M., Rodrguez-Mozaz, S. & Balczar, J.L. (2016). Occurrence and persistence of antibiotic resistance genes in river biofilms after wastewater inputs in small rivers. Environmental Pollution, 210:121-128.

Pulcrano, G., Balzaretti, M., Grosini, A., Placentini, V. & Poggighe, D. (2017). First report of Kocuria marina bloodstream infection unrelated to a central venous catheter: a mini-review on an emerging and under-recognised opportunistic pathogen. Le Infezioni in Medicina (1): 71-74.

Purty, S., Saranathan, R., Prashanth, K., Narayanan, K., Asir, J., Devi, C.S. & Amarnath, S.K. (2013). The expanding spectrum of human infections caused by Kocuria species: A case report and literature review. Emerging Microbes & Infections, 2(10):e71-78.

Qin, T., Yan, G., Ren, H., Zhou, H., Wang, H., Xu, Y., Zhao, M., Guan, H., Li, M. & Shao, Z. (2013). High prevalence, genetic diversity and intracellular growth ability of Legionella in hot spring environments. PloS One, 8(3): e59018.

Rahman, P.K. & Gakpe, E. (2008). Production, characterisation and applications of biosurfactants-review. Biotechnology, 7(2):360-370.

Rahmani, S., Forozandeh, M., Mosavi, M. & Rezaee, A. (2006). Detection of bacteria by amplifying the 16s rRNA gene with universal primers and RFLP. Medical Journal of the Islamic Republic of Iran (MJIRI), 19(4):333-338.

Rai, P., Sharma, A., Saxena, P., Soni, A.P., Chakdar, H., Kashyap, P.L., Srivastava, A.K. & Sharma, A.K. (2015). Comparison of molecular and phenetic typing methods to assess diversity of selected members of the genus Bacillus. Microbiology, 84(2):236-246.

Rajendhran, J. & Gunasekaran, P. (2011). Microbial phylogeny and diversity: Small subunit ribosomal RNA sequence analysis and beyond. Microbiological Research, 166(2):99-110.

Ramachandran, P., Sundharam, R., Palaniyappan, J. & Munusamy, A.P. (2013). Potential process implicated in bioremediation of textile effluents: A review. Advances in Applied Science Research, 4:131-145.

Ramani, K., Jain, S.C., Mandal, A.B. & Sekaran, G. (2012). Microbial induced lipoprotein biosurfactant from slaughterhouse lipid waste and its application to the removal of metal ions from aqueous solution. Colloids and Surfaces B: Biointerfaces, 97:254-263.

Ramlan, M.A.M., Azizan, N.A., Han, B.H., Kim, L.C., Mohamad, S.E. & Ibrahim, Z. (2012). Decolourisation of reactive black 5 by azoreductase produced by Brevibacillus panacihumi ZBI. Jurnal Teknologi (Sciences and Engineering), 59(1):11-16.

223

Reddy, N.S., Nimmagadda, A. & Rao, K.S. (2003). An overview of the microbial α-amylase family. African Journal of Biotechnology, 2(12):645-648.

Reis, R.S., Pacheco, G.J., Pereira, A.G. & Freire, D. (2013). Biosurfactants: Production and applications. In: Rolando Chamy and Francisca Rosenkranz (Editors), Biodegradation: Life of Science. InTechOpen. ISBN 978-953-51-1154-2.

Ren, S., Guo, J., Zeng, G. & Sun, G. (2006). Decolorization of triphenylmethane, azo, and anthraquinone dyes by a newly isolated Aeromonas hydrophila strain. Applied Microbiology and Biotechnology, 72(6):1316-1321.

Rhodes, M.E., Spear, J.R., Oren, A. & House, C.H. (2011). Differences in lateral gene transfer in hypersaline versus thermal environments. BMC Evolutionary Biology, 11:199-212.

Rimmele, M. & Boos, W. (1994). Trehalose-6-phosphate hydrolase of Escherichia coli. Journal of Bacteriology, 176(18):5654-5664.

Rizzo, L., Manaia, C., Merlin, C., Schwartz, T., Dagot, C., Ploy, M.C., Michael, I. & Fatta- Kassinos, D. (2013). Urban wastewater treatment plants as hotspots for antibiotic resistant bacteria and genes spread into the environment: A review. Science of the Total Environment, 447:345-360.

Roberts, M.S., Nakamura, L.K. & Cohan, F.M. (1994). Bacillus mojavensis sp. nov., distinguishable from Bacillus subtilis by sexual isolation, divergence in DNA sequence, and differences in fatty acid composition. International Journal of Systematic and Evolutionary Microbiology, 44(2):256-264.

Rodriguez-Mozaz, S., Chamorro, S., Marti, E., Huerta, B., Gros, M., Sànchez-Melsió, A., Borrego, C.M., Barceló, D. & Balcázar, J.L. (2015). Occurrence of antibiotics and antibiotic resistance genes in hospital and urban wastewaters and their impact on the receiving river. Water Research, 69:234-242.

Rojas, J.U., Verreth, J., Amato, S. & Huisman, E.A. (2003). Biological treatments affect the chemical composition of coffee pulp. Bioresource Technology, 89(3):267-274.

Rössler, D., Ludwig, W., Schleifer, K.H., Lin, C., Mcgill, T.J., Wisotzkey, J.D., Jurtshuk, P. & Fox, G.E. (1991). Phylogenetic diversity in the genus Bacillus as seen by 16S rRNA sequencing studies. Systematic and Applied Microbiology, 14(3):266-269.

Rover, M.R. & Brown, R.C. (2013). Quantification of total phenols in bio-oil using the Folin– Ciocalteu method. Journal of Analytical and Applied Pyrolysis, 104:366-371.

224

Rozanov, A.S., Sushentseva, N.N., Malup, T.K., Goryachkovskaya, T.N., Demidova, E.V., Meshcheriakova, I.A., Demidov, E.A. & Peltek, S.E. (2015). Analysis of enzymes of the hemicellulose complex from Geobacillus stearothermophilus 22 VKPM B-11678 isolated from Garga hot spring, Russia. Journal of Molecular Catalysis B: Enzymatic, 116:159-165.

Ruiz-Garcia, C., Quesada, E., Martinez-Checa, F., Llamas, I., Urdaci, M.C. & Béjar, V. (2005). Bacillus axarquiensis sp. nov. and Bacillus malacitensis sp. nov., isolated from river-mouth sediments in southern Spain. International Journal of Systematic and Evolutionary Microbiology, 55(3):1279-1285.

Russell, A.D. (2003). Similarities and differences in the responses of microorganisms to biocides. The Journal of Antimicrobial Chemotherapy, 52(5):750-763.

Ryan, M.P. & Adley, C.C. (2014). Ralstonia spp.: Emerging global opportunistic pathogens. European Journal of Clinical Microbiology & Infectious Diseases, 33(3):291-304.

Sabae, S.Z., Hazaa, M., Hallim, S.A., Awny, N.M. & Daboor, S.M. Bioremediation of Zn, Cu and Fe using Bacillus subtilis D215 and Pseudomonas putida biovar AD 225. BioScience Research, 3(1):189-204.

Sadettin, S. & Dönmez, G. (2006). Bioaccumulation of reactive dyes by thermophilic cyanobacteria. Process Biochemistry, 41(4):836-841.

Sadhu, S. & Maiti, T.K. (2013). Cellulase production by bacteria: A review. British Microbiology Research Journal, 3(3):235-258.

Said-Mohamed, R., Micklesfield, L.K., Pettifor, J.M. & Norris, S.A. (2015). Has the prevalence of stunting in South African children changed in 40 years? A systematic review. BMC Public Health, 15:534-544.

Samad, M.Y.A., Razak, C.N.A., Salleh, A.B., Yunus, W.Z.W., Ampon, K. & Basri, M. (1989). A plate assay for primary screening of lipase activity. Journal of Microbiological Methods, 9(1):51-56.

Samanta, A., Bera, P., Khatun, M., Sinha, C., Pal, P., Lalee, A. & Mandal, A. (2012a). An investigation on heavy metal tolerance and antibiotic resistance properties of bacterial strain bacillus sp. isolated from municipal waste. Journal of Microbiology and Biotechnology Research, 2(1):178-189.

225

Samanta, A., Pal, P., Mandal, A., Sinha, C., Lalee, A., Das, M., Kaity, S. & Mitra, D. (2012b). Estimation of biosurfactant activity of an alkaline protease producing bacteria isolated from municipal solid waste. Central European Journal of Biology, 1:26-35.

Sambanthamoorthy, K., Feng, X., Patel, R., Patel, S. & Paranavitana, C. (2014). Antimicrobial and antibiofilm potential of biosurfactants isolated from lactobacilli against multi-drug-resistant pathogens. BMC Microbiology, 14:197-206.

Samie, A., Mashao, M.B., Bessong, P.O., NKgau, T.F., Momba, M.N. & Obi, C.L. (2012). Diversity and antibiograms of bacterial organisms isolated from samples of household drinking- water consumed by HIV-positive individuals in rural settings, South Africa. Journal of Health, Population, and Nutrition, 30(3):241-249.

Sanchez, M.B., Hernandez, A., Rodriguez-Martinez, J.M., Martinez-Martinez, L. & Martinez, J.L. (2008). Predictive analysis of transmissible quinolone resistance indicates Stenotrophomonas maltophilia as a potential source of a novel family of qnr determinants. BMC Microbiology, 8:148-163.

Sanderson, H., Fricker, C., Brown, R.S., Majury, A. & Liss, S.N. (2016). Antibiotic resistance genes as an emerging environmental contaminant. Environmental Reviews, 24(2):205-218.

Sanjana, MC., Yadav., KS., Malashree, L & Prabha, R. (2017). Bacterial biosurfactants – a boon to dairy industry. International Journal of Current Microbiology and Applied Sciences 6 (5): 608-612.

Santal, A.R. & Singh, N. (2013). Biodegradation of melanoidin from distillery effluent: Role of microbes and their potential enzymes. Biodegradation of Hazardous and Special Products, INTECH Open Access Publisher. Pp.71-100. doi.org/10.5772/56252.

Santos, A., Mendes, S., Brissos, V. & Martins, L.O. (2014). New dye-decolorizing peroxidases from Bacillus subtilis and Pseudomonas putida MET94: Towards biotechnological applications. Applied Microbiology and Biotechnology, 98(5):2053-2065.

Santos, D.K.F., Rufino, R.D., Luna, J.M., Santos, V.A. & Sarubbo, L.A. (2016). Biosurfactants: Multifunctional biomolecules of the 21st century. International Journal of Molecular Sciences, 17(3):401-432.

Saranraj, P. (2013). Bacterial biodegradation and decolourization of toxic textile azo dyes. African Journal of Microbiology Research, 7(30):3885-3890.

226

Sarin, S., Khamsri, B. & Sarin, C. (2011). Isolation of biosurfactant–producing bacteria with antimicrobial activity against bacterial pathogens. Environment Asia Journal, 4:1-5.

Sariözlü, N.Y., Demirel, R. & Kivanc, M. (2012). Microbial population of hot spring waters in Eskişehir, Turkey. Anadolu University Journal of Science and Technology – C Life Sciences and Biotechnology, 2(1):31-39.

Satpute, S.K., Bhawsar, B.D., Dhakephalkar, P.K. & Chopade, B.A. (2008). Assessment of different screening methods for selecting biosurfactant producing marine bacteria. Indian Journal of Marine Sciences, 37(3):243-250.

Savini, V., Catavitello, C., Masciarelli, G., Astolfi, D., Balbinot, A., Bianco, A., Febbo, F., D'Amario, C. & D'Antonio, D. (2010). Drug sensitivity and clinical impact of members of the genus Kocuria. Journal of Medical Microbiology, 59(12):1395-1402.

Sayeh, R., Birrien, J.L., Alain, K., Barbier, G., Hamdi, M. & Prieur, D. (2010). Microbial diversity in Tunisian geothermal springs as detected by molecular and culture-based approaches. Extremophiles, 14(6):501-514.

Sayem, S.M., Manzo, E., Ciavatta, L., Tramice, A., Cordone, A., Zanfardino, A., De Felice, M. & Varcamonti, M. (2011). Anti-biofilm activity of an exopolysaccharide from a sponge- associated strain of Bacillus licheniformis. Microbial Cell Factories, 10:74-86.

Schulz, R. (2001). Rainfall-induced sediment and pesticide input from orchards into the Lourens River, Western Cape, South Africa: Importance of a single event. Water Research, 35(8):1869- 1876.

Schwarzenbach, R.P., Egli, T., Hofstetter, T.B., von Gunten, U. & Wehrli, B. (2010). Global water pollution and human health. Annual Review of Environment and Resources, 35:109-136.

Scott, M.D., Meshnick, S.R. & Eaton, J.W. (1987). Superoxide dismutase-rich bacteria. paradoxical increase in oxidant toxicity. The Journal of Biological Chemistry, 262(8):3640- 3645.

Segawa, T., Takeuchi, N., Rivera, A., Yamada, A., Yoshimura, Y., Barcaza, G., Shinbori, K., Motoyama, H., Kohshima, S. & Ushida, K. (2013). Distribution of antibiotic resistance genes in glacier environments. Environmental Microbiology Reports, 5(1):127-134.

Selvarajan, R., Sibanda, T. & Tekere, M. (2017). Thermophilic bacterial communities inhabiting the microbial mats of “indifferent” and chalybeate (iron-rich) thermal springs:

227

Diversity and biotechnological analysis. Microbiology Open, 2017; e560 https://doi.org/ 10.1002/mbo3.560.

Sen, R. & Maiti, N.K. (2014). Genomic and functional diversity of bacteria isolated from hot springs in Odisha, India. Geomicrobiology Journal, 31(7):541-550.

Sen, S.K., Mohapatra, S.K., Satpathy, S. & Rao, G.T. (2010). Characterization of hot water spring source isolated clones of bacteria and their industrial applicability. International Journal of Chemical Research, 2(1):1-7.

Sengupta, S., Chattopadhyay, M.K. & Grossart, H. (2013). The multifaceted roles of antibiotics and antibiotic resistance in nature. Frontiers in Microbiology, 4: Article 47:1-13.

Seo, J., Keum, Y. & Li, Q.X. (2009). Bacterial degradation of aromatic compounds. International Journal of Environmental Research and Public Health, 6(1):278-309.

Shafi, J., Tian, H. & Ji, M. (2017). Bacillus species as versatile weapons for plant pathogens: a review. Biotechnology & Biotechnological equipment. 31 (3): 446-459.

Shah, A.A., Nawaz, A., Kanwal, L., Hasan, F., Khan, S. & Badshah, M. (2015). Degradation of poly (ε-caprolactone) by a thermophilic bacterium Ralstonia sp. strain MRL-TL isolated from hot spring. International Biodeterioration & Biodegradation, 98:35-42.

Shaligram, N.S. & Singhal, R.S. (2010). Surfactin--A review on biosynthesis, fermentation, purification and applications. Food Technology & Biotechnology, 48(2):119-134.

Sharma, A., Paul, D., Dhotre, K., Jani, A., Pandey, Y. & Shouche, S. (2017). Deep sequencing analysis of bacterial community structure of Soldhar hot spring, India. Microbiology, 86(1):136–142.

Sharma, A., Dhiman, K., Ansari, Z. & Shirkot, P. (2016). Thermostable lipase production by aneurinibacillus thermoaerophilus MBW2 strain isolated from Indian hot water spring. Journal of Pure and Applied Microbiology, 10(2):1197-1211.

Sharma, A., Soni, J., Kaur, G. & Kaur, J. (2014a). A study on biosurfactant production in lactobacillus and Bacillus sp. International Journal of Current Microbiology and Applied Sciences, 3(11):723-733.

Sharma, A., Jani, K., Shouche, Y.S. & Pandey, A. (2014b). Microbial diversity of the Soldhar hot spring, India, assessed by analyzing 16S rRNA and protein-coding genes. Annals of Microbiology, pp.1-10. doi:10.1007/s13213-014-0970-4.

228

Sharma, N., Vyas, G. & Pathania, S. (2013a). Culturable diversity of thermophilic microorganisms found in hot springs of Northern Himalayas and to explore their potential for production of industrially important enzymes. Scholars Academic Journal of Biosciences, 1(5):165-178.

Sharma, S.B., Sayyed, R.Z., Trivedi, M.H. & Gobi, T.A. (2013b). Phosphate solubilizing microbes: Sustainable approach for managing phosphorus deficiency in agricultural soils. SpringerPlus, 2(1):587-601.

Sharma, S., Sachdeva, P. & Virdi, J.S. (2003). Emerging water-borne pathogens. Applied Microbiology and Biotechnology, 61(5-6):424-428.

Sheehan, K., Fagg, J., Ferris, M. & Henson, J. (2003). Thermophilic amoebae and Legionella in hot springs in Yellowstone and Grand Teton National Parks. Geothermal Biology and Geochemistry in Yellowstone National Park: Proceeding of the Thermal Biology Institute Workshop, Yellowstone National Park, WY, October 2003. pp. 317-326.

Sheikhi, F., Ardakani, M.R., Enayatizamir, N. & Rodriguez-Couto, S. (2012). The determination of assay for laccase of Bacillus subtilis WPI with two classes of chemical compounds as substrates. Indian Journal of Microbiology, 52(4):701-707.

Shekoohiyan, S., Moussavi, G. & Naddafi, K. (2016). The peroxidase-mediated biodegradation of petroleum hydrocarbons in a H2O2-induced SBR using in-situ production of peroxidase: Biodegradation experiments and bacterial identification. Journal of Hazardous Materials, 313:170-178.

Shete, B.S. & Shinkar, N.P. (2013). Dairy industry wastewater sources, characteristics & its effects on environment. International Journal of Current Engineering and Technology, 3(5):1611-1616.

Shevchenko, A., Tomas, H., Havli, J., Olsen, J.V. & Mann, M. (2006). In-gel digestion for mass spectrometric characterization of proteins and proteomes. Nature Protocols, 1(6):2856-2860.

Shivaji, S., Chaturvedi, P., Suresh, K., Reddy, G., Dutt, C., Wainwright, M., Narlikar, J.V. & Bhargava, P.M. (2006). Bacillus aerius sp. nov., Bacillus aerophilus sp. nov., Bacillus stratosphericus sp. nov. and Bacillus altitudinis sp. nov., isolated from cryogenic tubes used for collecting air samples from high altitudes. International Journal of Systematic and Evolutionary Microbiology, 56(7):1465-1473.

229

Shuangqi, T., Zhenyu, W., Ziluan, F., Lili, Z. & Jichang, W. (2011). Determination methods of cellulase activity. African Journal of Biotechnology, 10(37):7122-7125.

Sietske, A. & Diderichsen, B. (1991). On the safety of Bacillus subtilis and B. amyloliquefaciens: A review. Applied Microbiology and Biotechnology, 36(1):1-4.

Siezen, R.J. & Leunissen, J.A. (1997). Subtilases: The superfamily of subtilisin‐like serine proteases. Protein Science, 6(3):501-523.

Sineriz, F., Hommel, R.K. & Kleber, H.P. (2001). Production of biosurfactants. Encyclopedia of Life Support Systems (EOLSS). Developed under the Auspices of the UNESCO, EOLSS Publishers, Oxford, UK.

Singh, R. & Eltis, L.D. (2015). The multihued palette of dye-decolorizing peroxidases. Archives of Biochemistry and Biophysics, 574:56-65.

Singh, R.N., Kaushik, R., Arora, D.K. & Szxena, A.K. (2013a). Prevalence of opportunist pathogens in thermal springs of devotion. Journal of Applied Sciences in Environmental Sanitation, 8(3):195-203.

Singh, Y., Ramteke, PW., Tripathy, A. & Shukla, PK. (2013b). Isolation and characterisation of Bacillus resistant to multiple heavy metals. International Journal of Current Microbiology 2(11): 525-530.

Singh, V. (2012). Biosurfactant – isolation, production, purification & significance. International Journal of Science and Research, 2:1-4.

Singh, L. & Singh, V.P. (2010). Biodegradation of textile dyes, bromophenol blue and congo red by fungus Aspergillus flavus. Environment & We: An International Journal of Science & Technology, 5:235-242.

Sliman, R., Rehm, S. & Shlaes, D.M. (1987). Serious infections caused by Bacillus species. Medicine, 66(3):218-223.

So, N., Rho, J., Lee, S., Hancock, I.C. & Kim, J. (2001). A lead-absorbing protein with superoxide dismutase activity from Streptomyces subrutilus. FEMS Microbiology Letters, 194(1):93-98.

Soge, O.O., Giardino, M.A., Ivanova, I.C., Pearson, A.L., Meschke, J.S. & Roberts, M.C. (2009). Low prevalence of antibiotic-resistant gram-negative bacteria isolated from rural southwestern Ugandan groundwater. Water SA, 35(3):343-347.

230

Souza, E.C., Vessoni-Penna, T.C. & de Souza Oliveira, Ricardo Pinheiro (2014). Biosurfactant- enhanced hydrocarbon bioremediation: An overview. International Biodeterioration & Biodegradation, 89:88-94.

Strazzulli, A., Fusco, S., Cobbucci-Ponzano, B., Moracci, M. & Contursi, P. (2017). Metagenomics of microbial and viral life in terrestrial geothermal environments. Review Environment Science Biotechnology 16:425-454.

Strong, P.J. & Burgess, J.E. (2008). Fungal and enzymatic remediation of a wine lees and five wine-related distillery wastewaters. Bioresource Technology, 99(14):6134-6142.

Subramanian, J., Ramesh, T. & Kalaiselvam, M. (2014). Degradation of triphenylmethane dye: Malachite green by Aspergillus flavus. World Journal of Pharmacy and Pharmaceutical Sciences, 3(8):44-50.

Sudha, M., Saranya, A., Selvakumar, G. & Sivakumar, N. (2014). Microbial degradation of azo dyes: A review. International Journal of Current Microbiology and Applied Sciences, 3(2):670- 690.

Sukthana, Y., Lekkla, A., Sutthikornchai, C., Wanapongse, P., Vejjajiva, A. & Bovornkitti, S. (2005). Spa, springs and safety. Southeast Asian Journal of Tropical Medicine and Public Health, 36(s4):10-17.

Sumi, C.D., Yang, B.W., Yeo, I. & Hahm, Y.T. (2015). Antimicrobial peptides of the genus Bacillus: A new era for antibiotics. Canadian Journal of Microbiology, 61(2):93-103

Sun, W., Liu, W., Cui, L., Zhang, M. & Wang, B. (2013). Characterization and identification of a chlorine-resistant bacterium, Sphingomonas TS001, from a model drinking water distribution system. Science of the Total Environment, 458:169-175.

Takeda, K., Moriki, M., Oshiro, W. & Sakugawa, H. (2013). Determination of phenolic concentrations in dissolved organic matter pre-concentrate using solid phase extraction from natural water. Marine Chemistry, 157:208-215.

Tam, L., Eymann, C., Albrecht, D., Sletmann, R., Schauer, F., Hecker, M. & Antelmann, H. (2006). Differential gene expression in response to phenol and catechol reveals different metabolic activities for the degradation of aromatic compounds in Bacillus subtilis. Environmental Microbiology, 8 (8):1408-1427.

231

Tambekar, D.H., Borse, P.S. & Gadakh, P.V. (2012). Biodegradation of phenol by native microorganisms isolated from Lonar Lake in Maharashtra state (India). International Journal of Life Science and Pharma Research, 2(4):26-30.

Tambekar, D.H., Tambekar, S.D., Borkar, P.R. & Laddha, Y.P. (2015). Molecular characterization and detoxification of phenol by haloalkaliphilic bacterium Bacillus flexus. International Journal of Advanced Biotechnology and Research, 6(2):232-237.

Tatullo, M., Simone, G.M., Tarullo, F., Irlandese, G., De Vito, D., Marrelli, M., Santacroce, L., Cocco, T., Ballini, A. & Scacco, S. (2016). Antioxidant and antitumor activity of a bioactive polyphenolic fraction isolated from the brewing process. Nature Scientific Reports, 6:36042. doi:10.1038/srep36042.

Techaoei, S., Leelapornpisid, P., Santiarwarn, D. & Lumyong, S. (2007). Preliminary screening of biosurfactant producing microorganisms isolated from hot spring and garages in northern Thailand. KMITL Science and Technology Journal, 7:38-43.

Tekere, M., Mswaka, A.Y., Zvauya, R. & Read, J.S. (2001). Growth, dye degradation and ligninolytic activity studies on Zimbabwean white rot fungi. Enzyme and Microbial Technology, 28(4):420-426.

Tekere, M., Prinsloo, A., Olivier, J., Jonker, N. & Venter, S.N. (2012). An evaluation of the bacterial diversity at Tshipise, Mphephu and Sagole hot water springs, Limpopo Province, South Africa. African Journal of Microbiology Research, 6(23): 4993-5004.

Tekere, M., Lötter, A., Olivier, J., Jonker, N. & Venter, S. (2011). Metagenomic analysis of bacterial diversity of Siloam hot water spring, Limpopo, South Africa. African Journal of Biotechnology, 10(78):18005-18012.

Ten, L.N., Baek, S.H., Im, W.T., Liu, Q.M., Aslam, Z. & Lee, S.T. (2006). Bacillus panaciterrae sp. nov., isolated from soil of a ginseng field. International journal of Systematic and Evolutionary Microbiology, 56(12):2861-2866.

Tene, N., Roche-Chatain, V., Rifflet, A., BOnnafe, E., Lefranc, B., Leprince, J. & Treilhou, M. (2014). Potent bactericidal effects of bicarinalin against strains of the Enterobacter and Cronobacter genera. Food Control 42:202-206.

Tewari, S., Ramteke, P.W., Tripathi, M., Kumar, S. & Garg, S.K. (2013). Plasmid mediated transfer of antibiotic resistance and heavy metal tolerance in thermotolerant water borne coliforms. African Journal of Microbiology Research, (7):130-136.

232

Thaniyavarn, J., Roongsawang, N., Kameyama, T., Haruki, M., Imanaka, T., Morikawa, M. & Kanaya, S. (2003). Production and characterization of biosurfactants from Bacillus licheniformis F2. 2. Bioscience, Biotechnology, and Biochemistry, 67(6):1239-1244.

Thatoi, H., Das, S., Mishra, J., Rath, B.P. & Das, N. (2014). Bacterial chromate reductase, a potential enzyme for bioremediation of hexavalent chromium: A review. Journal of Environmental Management, 146:383-399.

Thavasi, R., Sharma, S. & Jayalakshmi, S. (2011). Evaluation of screening methods for the isolation of biosurfactant producing marine bacteria. Journal of Petroleum and Environmental Biotechnology, S, 001. doi:10.4172/2157-7463.S1-001.

Thebti, W., Riahi, Y., Gharsalli, R. & Belhadj, O. (2016). Screening and characterization of thermo-active enzymes of biotechnological interest produced by thermophilic bacillus isolated from hot springs in Tunisia. Acta Biochimica Polonica, 63(3):581-587.

Thiel, P.G. & du Toit, P.J. (1965). The chemical composition of a brewery waste. Journal of the Institute of Brewing, 71(6):509-514.

Thorolfsdottir, B.O.T. & Marteinsson, V.T. (2013). Microbiological analysis in three diverse natural geothermal bathing pools in Iceland. International Journal of Environmental Research and Public Health, 10(3):1085-1099.

Tikariha, A. & Sahu, O. (2014). Study of characteristics and treatments of dairy industry waste water. Journal of Applied & Environmental Microbiology, 2(1):16-22.

Tiwari, S., Gaur, R., Rai, P. & Tripathi, A. (2012). Decolorization of distillery effluent by thermotolerant Bacillus subtilis. American Journal of Applied Sciences, 9(6):798-806.

Toh, H., Tay, H., Kuar, W., Weng, T., Tang, H. & Tan, C. (2011). Risk factors associated with Sphingomonas paucimobilis infection. Journal of Microbiology, Immunology and Infection, 44(4):289-295.

Torres, J.A., Chagas, P.M.B., Silva, M.C., Dos Santos, C.D. & Corrêa, A.D. (2016). Enzymatic oxidation of phenolic compounds in coffee processing wastewater. Water Science and Technology, 73(1):39-50.

Tung, M., Hsu, B., Tao, C., Lin, W., Tsai, H., Ji, D., Shen, S., Chen, J., Shih, F. & Huang, Y. (2013). Identification and significance of Naegleria fowleri isolated from the hot spring which related to the first primary amebic meningoencephalitis (PAM) patient in Taiwan. International Journal for Parasitology, 43(9):691-696.

233

Turner, S., Pryer, K.M., Miao, V.P. & Palmer, J.D. (1999). Investigating deep phylogenetic relationships among cyanobacteria and plastids by small subunit rRNA sequence analysis. Journal of Eukaryotic Microbiology, 46(4):327-338.

Usha, D., Kanimozhi, G. & Panneerselvam. A. (2014). Isolation and screening of pectin lyase producing fungi from soil sample of dead organic matters. World Journal of Pharmaceutical Research, 3(10):563-569.

Van Belkum A. (2011). Classification of Bacterial Pathogens. COGEM research report. The Netherlands Commission on Genetic Modification (COGEM), Bilthoven, the Netherlands.

Vaz-Moreira, I., Nobre, M.F., Nunes, O.C. & Manaia, C.M. (2007). Gulbenkiania mobilis gen. nov., sp. nov., isolated from treated municipal wastewater. International Journal of Systematic and Evolutionary Microbiology, 57(5):1108-1112.

Vaz-Moreira, I., Nunes, O.C. & Manaia, C.M. (2014). Bacterial diversity and antibiotic resistance in water habitats: Searching the links with the human microbiome. FEMS Microbiology Reviews, 38(4):761-778.

Vázquez, C.V., Rojas, M.G.V., Ramírez, C.A., Chávez-Servín, J.L., García-Gasca, T., Martínez, R.A.F., García, O.P., Rosado, J.L., López-Sabater, C.M. & Castellote, A.I. (2015). Total phenolic compounds in milk from different species. Design of an extraction technique for quantification using the Folin–Ciocalteu method. Food Chemistry, 176:480-486.

Vedula, R.K., Dalal, S. & Majumder, C.B. (2013). Bioremoval of cyanide and phenol from industrial wastewater: An update. Bioremediation Journal, 17(4):278-293. Ventola, C.L. (2015). The antibiotic resistance crisis: Part 1: Causes and threats. P & T: A Peer-Reviewed Journal for Formulary Management, 40(4):277-283.

Venton, D. (2013). Highlight: Bacteria shed DNA as they adapt to hot temperatures. Genome Biology and Evolution, 5(5):987-988.

Verheyen, V., Cruickshank, A., Wild, K., Heaven, M.W., McGee, R., Watkins, M. & Nash, D. (2009). Soluble, semivolatile phenol and nitrogen compounds in milk-processing wastewaters. Journal of Dairy Science, 92(7):3484-3493.

Verma, P., Yadav, A.N., Shukla, L., Saxena, A.K. & Suman, A. (2015). Hydrolytic enzymes production by thermotolerant Bacillus altitudinis IARI-MB-9 and Gulbenkiania mobilis IARI- MB-18 isolated from Manikaran hot springs. International Journal of Advanced Research, 3(9):1241-1250.

234

Vidali, M. (2001). Bioremediation. An overview. Pure and Applied Chemistry, 73(7):1163- 1172.

Vijayadeep, C. & Sastry, P. (2014). Effect of heavy metal uptake by E. coli and Bacillus spp. Journal of Bioremediation & Biodegradation, 5:5-8.

Vijayalakshmidevi, S.R. & Muthukumar, K. (2015). Improved biodegradation of textile dye effluent by co-culture. Ecotoxicology and Environmental Safety, 114:23-30.

Vijayaraghavan, K. & Yun, Y. (2008). Bacterial biosorbents and biosorption. Biotechnology Advances, 26(3):266-291.

Walczak, M., Burkowska‐But, A., Brzezinska, M.S. & Krawiec, A. (2016). Distribution of Legionella pneumophila in thermal pools. CLEAN–Soil, Air, Water,43 (9999):1-6.

Wallace Jr, R.J., Brown, B.A. & Griffith, D.E. (1998). Nosocomial outbreaks/pseudo outbreaks caused by nontuberculous mycobacteria. Annual Reviews in Microbiology, 52(1):453-490.

Walsh, F. (2013). Investigating antibiotic resistance in non-clinical environments. Frontiers in Microbiology, 4:19-24.

Wang, H., Xia, X. & Hickey, D. (2006). Thermal adaptation of the small subunit ribosomal RNA gene: A comparative study. Journal of Molecular Evolution, 63(1):120-126.

Wang, L., Lee, F., Tai, C., Yokota, A. & Kuo, H. (2007). Reclassification of Bacillus axarquiensis Ruiz-Garcia et al. 2005 and Bacillus malacitensis Ruiz-Garcia et al. 2005 as later heterotypic synonyms of Bacillus mojavensis Roberts et al. 1994. International Journal of Systematic and Evolutionary Microbiology, 57(7):1663-1667.

Wang, R.F., Wennerstrom, D., Cao, W.W., Khan, A.A. & Cerniglia, C.E. (2000). Cloning, expression, and characterization of the katG gene, encoding catalase-peroxidase, from the polycyclic aromatic hydrocarbon-degrading bacterium Mycobacterium sp. strain PYR-1. Applied and Environmental Microbiology, 66(10):4300-4304.

Wang, S., Lin, X., Huang, X., Zheng, L. & Zilda, D.S. (2012). Screening and characterization of the alkaline protease isolated from PLI-1, a strain of Brevibacillus sp. collected from Indonesia’s hot springs. Journal of Ocean University of China, 11(2):213-218.

Wauters, G., Charlier, J., Janssens, M. & Delmee, M. (2000). Identification of Arthrobacter oxydans, Arthrobacter luteolus sp. nov., and Arthrobacter albus sp. nov., isolated from human clinical specimens. Journal of Clinical Microbiology, 38(6):2412-2415.

235

Wei, W., Davis, R.E., Lee, M. & Zhao, Y. (2007). Computer-simulated RFLP analysis of 16S rRNA genes: Identification of ten new phytoplasma groups. International Journal of Systematic and Evolutionary Microbiology, 57(8):1855-1867.

Wellinghausen, N., Frost, C. & Marre, R. (2001). Detection of legionellae in hospital water samples by quantitative real-time LightCycler PCR. Applied and Environmental Microbiology, 67(9):3985-3993.

West, B.M., Liggit, P., Clemans, D.L. & Francoeur, S.N. (2011). Antibiotic resistance, gene transfer, and water quality patterns observed in waterways near CAFO farms and wastewater treatment facilities. Water, Air, & Soil Pollution, 217(1-4):473-489.

Williams, M.M., Armbruster, C.R. & Arduino, M.J. (2013). Plumbing of hospital premises is a reservoir for opportunistically pathogenic microorganisms: A review. Biofouling, 29(2):147- 162.

Witthuhn, R.C., Kemp, F. & Britz, T.J. (2007). Isolation and PCR detection of Enterobacter sakazakii in South African food products, specifically infant formula milks. World Journal of Microbiology and Biotechnology, 23(2):151-157.

Woolhouse, M., Waugh, C., Perry, M.R. & Nair, H. (2016). Global disease burden due to antibiotic resistance–state of the evidence. Journal of Global Health, 6(1):010306-010311.

Wu, W., Park, S. & Ahn, B. (2013a). Isolation and characterization of an antimicrobial substance from Bacillus subtilis BY08 antagonistic to Bacillus cereus and Listeria monocytogenes. Food Science and Biotechnology, 22(2):433-440.

Wu, Z., Ye, C., Guo, F., Zhang, S. & Yu, X. (2013b). Evidence for broad-spectrum biofilm inhibition by the bacterium Bacillus sp. strain SW9. Applied and Environmental Microbiology, 79(5):1735-1738.

Wu, L., Yao, J., Jain, A.K., Chandankere, R., Duan, X. & Richnow, H. H. (2014). An efficient thermotolerant and halophlic biosurfactant producing bacterium isolated from Dagang oil field for MEOR application. International Journal of Current Microbiology and Applied Sciences, 3(7):586-599.

Wu, J., Jung, B., Kim, K., Lee, Y. & Sung, N. (2009). Isolation and characterisation of Pseudomonas otitidis WL-13 and its capacity to decolorise triphenylmethane dyes. Journal of Environmental Sciences, 21: 960-964.

236

Xia, W., Dong, H., Zheng, C., Cui, Q., He, P. & Tang, Y. (2015). Hydrocarbon degradation by a newly isolated thermophilic Anoxybacillus sp. with bioemulsifier production and new alkB genes. RSC Advances, 5(124):102367-102377.

Xian, W., Yin, Y., Liu, L., Yuan, C., Hussain, F., Khan, I., Habib, N., Zhou, E. & Li, W. (2016). Brevibacillus sediminis sp. nov., isolated from a hot spring. International Journal of Systematic and Evolutionary Microbiology, 66(2):548-553.

Xiong, Y., Wang, Y., Yu, Y., Li, Q., Wang, H., Chen, R., and He, N. (2010). Production and characterisation of a novel bioflocculant from Bacillus licheniformis. Applied and Environmental Microbiology, 76 (9):2778-2782.

Xu, D. & Cote, J.C. (2003). Phylogenetic relationships between Bacillus species and related genera inferred from comparison of 3' end 16S rDNA and 5' end 16S-23S ITS nucleotide sequences. International Journal of Systematic and Evolutionary Microbiology, 53(3):695-704.

Xu, W., Wang, X. & Cai, Z. (2013). Analytical chemistry of the persistent organic pollutants identified in the Stockholm convention: A review. Analytica Chimica Acta, 790:1-13.

Yakhlef, W. & Darbouche, A. (2012) Metabolic diversity of thermophlic bacteria from hot springs in Algeria. Journal of Academica, 2:57-65.

Yamahara, K.M., Sassoubre, L.M., Goodwin, K.D. & Boehm, A.B. (2012). Occurrence and persistence of bacterial pathogens and indicator organisms in beach sand along the California coast. Applied and Environmental Microbiology, 78(6):1733-1745.

Yamane, K., Hattori, Y., Ohtagaki, H. & Fujiwara, K. (2011). Microbial diversity with dominance of 16S rRNA gene sequences with high GC contents at 74 and 98 degrees C subsurface crude oil deposits in Japan. FEMS Microbiology Ecology, 76(2):220-235.

Yan, L., Boyd, K.G., Adams, D.R. & Burgess, J.G. (2003). Biofilm-specific cross-species induction of antimicrobial compounds in bacilli. Applied and Environmental Microbiology, 69(7):3719-3727.

Yang, J., Li, X., Huang, L. & Jiang, H. (2015). Actinobacterial diversity in the sediments of five cold springs on the Qinghai-Tibet plateau. Frontiers in Microbiology, 6:1345-1354.

Yang,C. & Lee, C. (2007). Enrichment, isolation and characterisation of phenol-degrading Pseudomonas resinovorans strain P1 and Brevibacillus sp. strain P6. International Biodeterioration & Biodegradation 59: 206-210.

237

Yarita, K., Sano, A., Murata, Y., Takayama, A., Takahashi, Y., Takahashi, H., Yaguchi, T., Ohori, A., Kamei, K. & Miyaji, M. (2007). Pathogenicity of Ochroconis gallopava isolated from hot springs in Japan and a review of published reports. Mycopathologia, 164(3):135-147.

Yarza, P., Yilmaz, P., Pruesse, E., Glöckner, F.O., Ludwig, W., Schleifer, K., Whitman, W.B., Euzéby, J., Amann, R. & Rosselló-Móra, R. (2014). Uniting the classification of cultured and uncultured bacteria and archaea using 16S rRNA gene sequences. Nature Reviews Microbiology, 12(9):635-645.

Ye, J., Yin, H., Peng, H., Bai, J., Xie, D. & Wang, L. (2013). Biosorption and biodegradation of triphenyltin by Brevibacillus brevis. Bioresource Technology, 129:236-241.

Yohandini, H. (2015). Isolation and phylogenetic analysis of thermophile community within Tanjung Sakti hot spring, South Sumatera, Indonesia. HAYATI Journal of Biosciences, 22(3):143-148.

Yoon, S., Song, J., Go, S. & Ryu, J. (1998). Production of biopolymer flocculant by Bacillus subtilis TB11. Journal of Microbiology and Biotechnology, 8(6):606-612.

Yoshinaka, T., Yano, K. & Yamaguchi, H. (1973). Isolation of highly radioresistant bacterium, Arthrobacter radiotolerans nov. sp. Agricultural and Biological Chemistry, 37(10):2269-2275.

Youcef-Ali, M., Kacem-chaouche, N., Dehimat, L., Bataiche, I., Kara-Ali, M., Cawoy, H. & Thonart, P. (2014). Antifungal activity and bioactive compounds produced by Bacillus mojavensis and Bacillus subtilis. African Journal of Microbiology Research, 8(6):476-484.

Youssef, N.H., Duncan, K.E., Nagle, D.P., Savage, K.N., Knapp, R.M. & McInerney, M.J. (2004). Comparison of methods to detect biosurfactant production by diverse microorganisms. Journal of Microbiological Methods, 56(3):339-347.

Alexander Yu. MerkelNikolay V. PimenovIgor I. RusanovAlexander I. SlobodkinGalina B. SlobodkinaIvan Yu. TarnovetckiiEvgeny N. FrolovArseny V. DubinAnna A. PerevalovaElizaveta A. Bonch-Osmolovskaya. 2017. Microbial diversity and autotrophic activity in Kamchatka hot springs. Extremophiles Volume 21, Issue 2, pp 307–317

Zabłocka-Godlewska, E., Przystaś, W. & Grabińska-Sota, E. (2009). Decolourization of triphenylmethane dyes and ecotoxicity their end products. Environment Protection Engineering, 35(1):161-169.

238

Zabochnicka-Świątek, M. & Krzywonos, M. (2014). Potentials of biosorption and bioaccumulation processes for heavy metal removal. Polish Journal of Environmental Studies, 23(2):551-561.

Zahoor, S., Javed, H.M. & Babar, M.E. (2016). Characterization of a novel hydrolytic enzyme producing thermophilic bacterium isolated from the hot spring of Azad Kashmir-Pakistan. Brazilian Archives of Biology and Technology, 59:e16150662.

Zaki, S., Farag, S., Elreesh, G., Elkady, M., Nosier, M. & Abd El Haleem, D. (2011). Characterisation of bioflocculants produced by bacteria isolated from crude petroleum oil. International Journal of Environmental Science and Technology, 8(4):831-840.

Zámocký, M., Godočíková, J., Koller, F. & Polek, B. (2001). Potential application of catalase- peroxidase from Comamonas terrigena N3H in the biodegradation of phenolic compounds. Antonie van Leeuwenhoek, 79(2):109-117.

Zarafeta, D., Kissas, D., Sayer, C., Gudbergsdottir, S.R., Ladoukakis, E., Isupov, M.N., Chatziioannou, A., Peng, X., Littlechild, J.A. & Skretas, G. (2016). Discovery and characterization of a thermostable and highly halotolerant GH5 cellulase from an Icelandic hot spring isolate. PloS One, 11(1):e0146454.

Zhang, F., Yang, X., Geng, L., Zhang, Z., Yin, Y. & Li, W. (2015). Purification and characterization of a novel and versatile α‐amylase from thermophilic Anoxybacillus sp. YIM 342. Starch‐Stärke, 67:1-8.

Zhang, C., Diao, H., Lu, F., Bie, X., Wang, Y. & Lu, Z. (2012a). Degradation of triphenylmethane dyes using a temperature and pH stable spore laccase from a novel strain of Bacillus vallismortis. Bioresource Technology, 126:80-86.

Zhang, Y., Meng, D., Wang, Z., Guo, H. & Wang, Y. (2012b). Oxidative stress response in two representative bacteria exposed to atrazine. FEMS Microbiology Letters, 334(2):95-101.

Zhang, C., Huang, X., Pan, W., Zhang, J., Wei, K., Klenk, H., Tang, S., Li, W. & Zhang, K. (2011). Anoxybacillus tengchongensis sp. nov. and Anoxybacillus eryuanensis sp. nov., facultatively anaerobic, alkalitolerant bacteria from hot springs. International Journal of Systematic and Evolutionary Microbiology, 61(1):118-122.

Zhang, Y., Gu, J., Wang, S. & Fan, X. (2010). Preparation and properties of surfactant- bacillolysin ion-pair in organic solvents. Chemical Papers, 64(4):443-449.

239

Zhang, Z., Bo, L., Xia, S., Wang, X. & Yang, A. (2007). Production and application of a novel bioflocculant by multiple-microorganism consortia using brewery wastewater as carbon source. Journal of Environmental Sciences, 19(6):667-673.

Zhao, M., Sun, P., Du, L., Wang, G., Jia, X. & Zhao, Y. (2014a). Biodegradation of methyl red by Bacillus sp. strain UN2: Decolorization capacity, metabolites characterization, and enzyme analysis. Environmental Science and Pollution Research, 21(9):6136-6145.

Zhao, S., Cao, F., Zhang, H., Zhang, L., Zhang, F. & Liang, X. (2014b). Structural characterization and biosorption of exopolysaccharides from Anoxybacillus sp. R4-33 isolated from radioactive radon hot spring. Applied Biochemistry and Biotechnology, 172(5):2732-2746.

Zhou, L., Li, H., Zhang, Y., Han, S. & Xu, H. (2014). Development of genus‐specific primers for better understanding the diversity and population structure of Sphingomonas in soils. Journal of Basic Microbiology, 54(8):880-888.

Zhu, Y., Li, S., Li, D., Liu, C. & Ma, F. (2014). Bioflocculation behaviours of microbial communities in water treatment. Water Science & Technology, 2014 doi: 10.2166/wst.2013746.

Zimmer, K.R., Seixas, A., Conceicao, J.M., Zvoboda, D.A., Barros, M.P., Tasca, T., Macedo, A.J. & Termignoni, C. (2013). Cattle tick-associated bacteria exert anti-biofilm and anti- Tritrichomonas foetus activities. Veterinary Microbiology, 164(1):171-176.

240

APPENDICES

Appendix 1: Waterborne emerging opportunistic bacterial pathogens

PATHOGEN INFECTION HOST LOCATION REFERENCE epidemic in intensive care Meric et Sphingomonas sp pneumonia unit patients Turkey al., 2009 diabetes, alcoholism, nosocomial and catheter- community acquired and Toh et al., Sphingomonas related bacteraemia nosocomial Taiwan 2011 bacteremia in hypertension Sphingomonas diabetic male with kidney and Dewan et paucimobilis heart disease immunocompromised India al., 2014 Ralstonia pickettii and R. Coenye, mannitolilytica respiratory infections cystic fibrosis patients USA 2002 Ryan & Adley, Ralstonia sp. osteomyelitis, meningitis nosocomial (review) 2014 Tepidimonas Ko et al., arfidensis bone marrow sample leukaemia patient Korea 2005 Cupriavidus child with aplastic Karafin et gilardii fatal infection anaemia USA al., 2010 elderly immunocompromised Al-Grawi, Hafnia alvei urinary tract infection patient Iraq 2008 Enterobacter sakazakii sepsis and meningitis neonates and infants USA Lai, 2001 Witthuhn Enterobacter South et al., sakazakii isolated infant formula Africa 2007 Enterobacter Hunter et sakazakii neonates and infants (review) al., 2008 able to escape immune Jaradat et Cronobacter spp. system outbreaks (review) al., 2014 Fakruddin Cronobacter et al., sakazakii (review) 2013 post laparoscopic Hong Ma et al., Kocuria kristinae acute cholecystitis cholecystectomy Kong 2005 Savini et Kocuria spp. (review) al., 2010 catheter-related bacteraemia and pulmonary septic emboli Dunn et Kocuria kristinae endocarditis in pregnant female USA al., 2011 Purty et Kocuria spp. (review) al., 2013 Kurosawa Legionella diabetic male at hot et al., pneumophila pneumonia or Pontiac fever springs facility Japan 2010 Arthrobacter Huang et scleromae from swollen scleromata dermatosis patient China al., 2010

Appendix 2: References for Bacillus spp. with bioflocculant, biosorption, biosurfactant and anti-biofilm properties.

241

Bacillus spp. B. subtilis B. licheniformis Anoxybacillus spp. Nontembiso et Yoon et al., Bioflocculants al., 2011 1998 Xiong et al., 2010

Cosa et al., 2013 Zaki et al., 2011 Natarajan, 2015 Al-Wasify et al., 2015 Giri et al., 2015 Nourbakhsh et Sabae et al., Biosorption al., 2002 2006 Ozdemir et al., 2013 Kumar et al., Vijayadeep & 2010 Sastry, 2014 Al-Araji et al., Nitschke & Thaniyavarn et al., Pakpitcharoen et al., Biosurfactant 2007 Pastore, 2006 20 03 2008 Pereira et al., Khairruddin et al., 2013 Gomaa, 2013 2016

Wu et al., 2014 Kavitha et al., 2014 Padmavathi & Morikawa et al., Anti-biofilm Pandian, 2014 2006 Sayem et al., 2011

Appendix 3: List of GenBank accession numbers for 16S rDNA sequence data of Bacillus and Bacillus-related reference strains and unknown isolates used in this study. The letters in parenthesis denotes Bergey’s groupings as listed by Ludwig et al. (2009). The size of the 16S rDNA sequence in base pairs (bp) is also indicated.

STRAIN OR ISOLATE SIZE (bp) GenBank or NCBI accession number

Aneurinibacillus sp. U33 1451 GenBank: KJ725179.1

Aneurinibacillus danicus NBRB102444 1501 GenBank: AB681786.1

Aneurinibacillus migulanus strain ATCC 9999 1422 NR_115593

Aneurinibacillus tyrosinisolvens 1458 GenBank: AB899818.1

Anoxybacillus sp. ATCC BBA-2555 1548 GenBank: KJ722458.1

Anoxybacillus bogrovensis strain NBIMCC 8427=DSM 17956 1341 NCBI Reference Sequence: NR_115021.1

Anoxybacillus eryuanensis strain E-112 1519 NCBI Reference Sequence: NR_117229.1

Anoxybacillus flavithermus strain WK1 1524 NCBI Reference Sequence: NR_074667.1

Anoxybacillus flavithermus subsp. yunnanensis strain E13 1449 NCBI Reference Sequence: NR_117774.1

Anoxybacillus gonensis strain G2 1382 NCBI Reference Sequence: NR_025667.1

242

Anoxybacillus kaynarcensis strain D1021 1430 NCBI Reference Sequence: NR_108265.1

Anoxybacillus kestanbolinensis strain K4 1376 GenBank: AY248711.1

Anoxybacillus pushchinoensis strain k-1 1338 NCBI Reference Sequence: NR_037100.1

Anoxybacillus rupiensis strain TSSC-4 1503 GenBank: KC759325.1

Anoxybacillus rupiensis strain TS_04 1463 GenBank: KJ842629.1

Anoxybacillus rupiensis strain TS_01 1463 GenBank: KJ842627.1

Anoxybacillus rupiensis strain A3 1428 GenBank: KC310454.1

Anoxybacillus rupiensis strain JF82 1463 GenBank: KF254911.1

Anoxybacillus rupiensis strain JF83 1510 GenBank: KC849452.1

Anoxybacillus rupiensis strain FZWP-10 1319 GenBank: JX914493.1

Anoxybacillus salavatliensis strain A343 1397 NCBI Reference Sequence: NR_104492.1

Anoxybacillus tengchongensis strain T-11 1519 GenBank: FJ438370.1

Anoxybacillus tepidamans strain R-35643 1507 NCBI Reference Sequence: NR_116985.1

Anoxybacillus thermarum strain DSM 17141 1358 NCBI Reference Sequence: NR_118117.1

Anoxybacillus vitaminiphilus strain 3nP4 1511 NCBI Reference Sequence: NR_108379.1

Anoxybacillus voinovskiensis strain TH13 1506 NCBI Reference Sequence: NR_024818.1

Bacillus aerophilus strain CRh28 1345 GenBank: KR780465.1

Bacillus aquimaris strain TF-12 (E) 1507 NCBI Reference Sequence: NR_025241.1

Bacillus arsenicus strain B3 (H) 1515 GenBank: GQ304784.1

Bacillus cereus strain ATCC 14579 (D) 1482 NCBI Reference Sequence: NR_114582.1

Bacillus coagulans strain ATCC 7050 (F) 1549 NCBI Reference Sequence: NR_115727.1

Bacillus halodurans strain ATCC 27557 (G) 1508 NCBI Reference Sequence: NR_112056.1

Bacillus lentus strain IAM 12466 (B) 1486 NCBI Reference Sequence: NR_115527.1

Bacillus licheniformis strain ATCC 14580 1545 NCBI Reference Sequence: NR_074923.1

Bacillus megaterium strain ATCC 14581 (C) 1495 NCBI Reference Sequence: NR_117473.1

Bacillus methylotrophicus 1510 GenBank: HZ046623.1

Bacillus panaciterrae Gsoil1517 1476 AC AB245380;

Bacillus pumilus strain ATCC 7061 1434 ACCESSION NR_043242

Bacillus siamensis strain IHB B 16121 1516 GenBank: KM817270.1

Bacillus siralis strain 171544 1430 NCBI Reference Sequence: NR_028709.1

Bacillus smithii strain TBMI12 (I) 1453 GenBank: EF010852.1

243

Bacillus subtilis ATCC 21331 1504 GenBank: AB018487.1

Bacillus subtilis subsp. spizizenii strain ATCC 6633 1507 NCBI Reference Sequence: NR_112049.1

Bacillus subtilis strain BCRC 10255 1468 NCBI Reference Sequence: NR_116017.1

Bacillus subtilis strain IAM 12118 1553 NCBI Reference Sequence: NR_112116.1

Bacillus subtilis strain JCM 1465 1472 NCBI Reference Sequence: NR_113265.1

Bacillus subtilis strain DSM 10 1517 NCBI Reference Sequence: NR_027552.1

Bacillus tequilensis strain 10b 1456 ACCESSION NR_104919

Brevibacillus agri strain DSM 6348 1487 NCBI Reference Sequence: NR_040983.1

Brevibacillus fluminis strain CJ71 1413 GenBank: EU375457.1

Brevibacillus panacihumi strain DCY35 1473 GenBank: EU383033.1

Pseudomonas aeruginosa strain ATCC 23993 1458 GenBank: FJ652615.1

Solibacillus sylvestris HR3-23 1528 PUBMED; 10319505.

Uncultured Bacillus sp. clone KSB12 1504 GenBank: JX047075.1

Uncultured Bacillus sp. clone TPB_GMAT_AC4 1278 GenBank: HG327138.1

Uncultured Bacillus sp. clone DGG30 1569 GenBank: AY082370.1

Appendix 4: List of the accession numbers of all the strains from GenBank used in phylogenetic analysis of emerging potential pathogens.

NO OF BACTERIAL ISOLATE BASES GENBANK ACCESSION NO Arthrobacter luteolus strain CF-25 1499 ACCESSION NR_025362 Arthrobacter aurescens TC1 strain TC1; ATCC NCBI Reference Sequence: BAA-1386 1481 NR_074272.1 Uncultured Arthrobacter sp. clone TPB_GMAT_AC3. 1303 ACCESSION HG327131 Arthrobacter sp. GM37AC3 K2 1303 GenBank: HE798206.1 Arthrobacter sp. NCCP-1348 1465 ACCESSION LC065375 Arthrobacter woluwensis strain GQ-9 1432 ACCESSION KT072630 Arthrobacter woluwensis strain 1551 1477 ACCESSION NR_044894 Arthrobacter mysorens strain SBANHCu24 1385 ACCESSION KR152305 Arthrobacter mysorens strain DSM 12798 1491 ACCESSION NR_025613 Arthrobacter cumminsii strain DMMZ 445 1483 ACCESSION NR_044895 Arthrobacter cumminsii strain Z486 1414 GenBank: EU086827.1 NCBI Reference Sequence: Kocuria kristinae strain DSM 20032 1475 NR_026199.1

244

NCBI Reference Sequence: Kocuria varians strain ATCC 15306 1333 NR_114674.1 Kocuria marina 1434 GenBank: LC055502.1 NCBI Reference Sequence: Kocuria rhizophila strain TA68 1471 NR_026452.1 NCBI Reference Sequence: Kocuria himachalensis strain K07-05 1459 NR_043323.1 Kocuria polaris strain CMS 76or 1480 ACCESSION NR_028924 Kocuria rosea strain DSM 20447 1481 ACCESSION NR_044871 Kocuria aegyptia strain YIM 70003 1492 ACCESSION NR_043511 Kocuria sediminis strain FCS-11 1429 ACCESSION NR_118222 Kocuria flava strain HO-9041 1443 ACCESSION NR_044308 Kocuria sp. B38 1309 ACCESSION KC492107 Kocuria turfanensis strain HO-9042 1441 ACCESSION NR_043899 Uncultured methanogenic archaeon clone 6 1439 GenBank: DQ372975.1 NCBI Reference Sequence: Cupriavidus metallidurans strain CH34 1493 NR_027607.1 NCBI Reference Sequence: Cupriavidus gilardii strain CIP 105966 1315 NR_116146.1 NCBI Reference Sequence: Cupriavidus gilardii strain LMG 5886 1451 NR_114460.1 Cupriavidus sp. NCCP-1142 1557 ACCESSION LC065169 Gulbenkiania mobilis strain V28 1404 ACCESSION KC492099 Gulbenkiania mobilis E4FC31T 1498 ACCESSION AM295491 NCBI Reference Sequence: Ralstonia pickettii strain ATCC 27511 1491 NR_043152.1 Ralstonia syzygii ATCC 49543 1513 GenBank: AB021403.1 NCBI Reference Sequence: Ralstonia mannitolilytica strain LMG6866 1449 NR_025385.1 Tepidimonas sp. BR5 1257 GenBank: KF206377.1 NCBI Reference Sequence: Tepidimonas fonticaldi strain AT-A2 1457 NR_109514.1 Tepidimonas arfidensis 1442 AC AB206468; Uncultured Tepidimonas sp. clone TPB_GMAT_5_1 1494 ACCESSION HE575486 NCBI Reference Sequence: Hafnia paralvei strain ATCC 29927 1495 NR_116898.1 NCBI Reference Sequence: Hafnia alvei strain ATCC 13337 1448 NR_044729.2 NCBI Reference Sequence: Cronobacter sakazakii strain ATCC 29544 1438 NR_118449.1 NCBI Reference Sequence: Cronobacter sakazakii ATCC BAA-894 1542 NR_102490.1 NCBI Reference Sequence: Silanimonas lenta strain 25-4 1382 NR_025815.1 Skermanella aerolata KACC 11604 1355 ACCESSION DQ672568 NCBI Reference Sequence: Sphingomonas echinoides strain ATCC 14820 1463 NR_024700.1 NCBI Reference Sequence: Sphingomonas paucimobilis strain DSM 30198 1462 NR_104893.1

245

Appendix 5: Table of isolates and GenBank accession numbers for phylogenetic tree of B. mojavensis

SIZE GenBank or NCBI accession STRAIN OR ISOLATE (bp) number Bacillus aerophilus strain CRh28 1345 GenBank: KR780465.1 Bacillus aquimaris strain TF-12 NCBI Reference Sequence: (E) 1507 NR_025241.1 Bacillus arsenicus strain B3 (H) 1515 GenBank: GQ304784.1 Bacillus cereus strain ATCC NCBI Reference Sequence: 14579 (D) 1482 NR_114582.1 Bacillus coagulans strain ATCC NCBI Reference Sequence: 7050 (F) 1549 NR_115727.1 Bacillus halodurans strain NCBI Reference Sequence: ATCC 27557 (G) 1508 NR_112056.1 Bacillus lentus strain IAM NCBI Reference Sequence: 12466 (B) 1486 NR_115527.1 Bacillus licheniformis strain NCBI Reference Sequence: ATCC 14580 1545 NR_074923.1 Bacillus megaterium strain NCBI Reference Sequence: ATCC 14581 (C) 1495 NR_117473.1 Bacillus methylotrophicus 1510 GenBank: HZ046623.1 Bacillus mojavensis BCRC17531 1527 GenBank: DQ993678 Bacillus mojavensis CS27 1455 GenBank: KR780382 Bacillus mojavensis UEBFK 1405 GenBank: KC297104 Bacillus mojavensis IFO15718 1407 GenBank: NR_118290 Bacillus panaciterrae Gsoil1517 1476 AC AB245380; Bacillus pumilus strain ATCC 7061 1434 GenBank: NR_043242 Bacillus siamensis strain IHB B 16121 1516 GenBank: KM817270.1 NCBI Reference Sequence: Bacillus siralis strain 171544 1430 NR_028709.1 Bacillus smithii strain TBMI12 (I) 1453 GenBank: EF010852.1 Bacillus subtilis ATCC 21331 1504 GenBank: AB018487.1 Bacillus subtilis strain BCRC NCBI Reference Sequence: 10255 1468 NR_116017.1 Bacillus subtilis strain IAM NCBI Reference Sequence: 12118 1553 NR_112116.1 Bacillus subtilis strain JCM NCBI Reference Sequence: 1465 1472 NR_113265.1 Bacillus subtilis UAC50 1453 GenBank: KC634086 NCBI Reference Sequence: Bacillus subtilis strain DSM 10 1517 NR_027552.1 Bacillus tequilensis strain 10b 1456 GenBank: NR_104919

246

Appendix 6: Comparison of GC content (%) for the 16S rDNA sequences for reference type strains obtained from GenBank, and isolates in this study. (*) denotes isolates that grouped into a different ARDRA cluster, not Anoxybacillus and Bacillus.

FRAGMENT GC content ISOLATE SIZE (%) GENUS AVERAGE SD

Anoxybacillus sp. ATCC BBA-2555 1548 56.65

Anoxybacillus bogrovensis strain NBIMCC 8427=DSM 17956 1341 55.78

Anoxybacillus eryuanensis strain E-112 1519 56.09

Anoxybacillus gonensis strain G2 1382 56.58

Anoxybacillus kaynarcensis strain D1021 1430 55.94

Anoxybacillus kestanbolinensis strain K4 1376 55.6

Anoxybacillus pushchinoensis strain k-1 1338 55.9

Anoxybacillus tengchongensis strain T-11 1519 56.48

Anoxybacillus thermarum strain DSM 17141 1358 56.55

Anoxybacillus vitaminiphilus strain 3nP4 1511 56.78

Anoxybacillus voinovskiensis strain TH13 1506 55.71

Anoxybacillus tepidamans strain R-35643 1507 56.67

Anoxybacillus salavatliensis strain A343 1397 56.4

Anoxybacillus flavithermus WK1 strain WK1 1524 56.56

Anoxybacillus flavithermus subsp. yunnanensis strain E13 1449 56.73

Anoxybacillus rupiensis strain TSSC-4 1503 56.95

Anoxybacillus rupiensis strain TS_04 1463 56.93

Anoxybacillus rupiensis strain TS_01 1463 56.73

Anoxybacillus rupiensis strain A3 1428 56.09

Anoxybacillus rupiensis strain JF82 1463 56.57

Anoxybacillus rupiensis strain JF83 1510 56.76

Anoxybacillus rupiensis strain FZWP-10 1319 56.33 56.4 0.41

11T * 1060 55.28 Anoxybacillus

4T 1407 56.15 Anoxybacillus

17S 1407 56.22 Anoxybacillus

3T 1394 56.24 Anoxybacillus

43T 1390 56.26 Anoxybacillus

19S 1406 56.33 Anoxybacillus

13S 1398 56.51 Anoxybacillus

247

7T 1421 56.72 Anoxybacillus 56.21 0.5

75S 1393 57.43 unknown

Bacillus subtilis ATCC 21331 1504 55.12

Bacillus licheniformis strain ATCC 14580 1545 55.53

Bacillus tequilensis strain 10b 1456 55.22

Bacillus subtilis subsp. spizizenii strain ATCC 6633 1507 55.14

Bacillus subtilis strain BCRC 10255 1468 55.12

Bacillus subtilis strain IAM 12118 1553 55.05

Bacillus subtilis strain JCM 1465 1472 55.23

Bacillus subtilis strain DSM 10 1517 55.31

Bacillus pumilus strain ATCC 7061 1434 55.16

Bacillus panaciterrae Gsoil1517 1476 54.74

Bacillus methylotrophicus 1510 55.36

Bacillus siamensis strain IHB B 16121 1516 55.01

Bacillus siralis strain 171544 1430 54.72 55.13 0.22

1T * 1417 52.86 Bacillus

33Li * 1079 54.21 Bacillus

14S* 1370 54.23 Bacillus

78S 1001 54.45 Bacillus

12S 1083 54.48 Bacillus

39T 1060 54.52 Bacillus

47Li 1069 54.53 Bacillus

40Le 1369 54.57 Bacillus

41Li 1386 54.69 Bacillus

54T 1379 54.69 Bacillus

18S 1411 54.71 Bacillus

28M 1065 54.84 Bacillus

48Li 1409 54.86 Bacillus

74T 1388 54.97 Bacillus

10T 1402 55 Bacillus

21M 1408 55.04 Bacillus

8T 1406 55.05 Bacillus

15S 1379 55.11 Bacillus

30M 1358 55.15 Bacillus

2T 1413 55.4 Bacillus

6T 1380 55.43 Bacillus

83Li 1396 55.44 Bacillus 54.75 0.54

24M 1303 54.8 Bacillus

248

32Le 918 54.36 Bacillus

77S 1057 54.4 Bacillus

Uncultured Bacillus sp. clone KSB12 1504 55.053 unknown

Uncultured Bacillus sp. clone TPB_GMAT_AC4 1278 54.695 unknown

Brevibacillus agri strain DSM 6348 1487 55.68

Brevibacillus fluminis strain CJ71 1413 55.56

Brevibacillus panacihumi strain DCY35 1473 56.2 55.8 0.34

16S 1398 54.72

36Li 1399 55.54

52M 1396 55.09 unknown

70T 1396 55.3

85Li 1375 54.25 54.95 0.58

Aneurinibacillus tyrosinisolvens 1458 56.86

Aneurinibacillus sp. U33 1451 56.92

Aneurinibacillus danicus NBRB102444 1501 56.89 56.89 0.03

86Li 1392 56.82 56.82

53M 1389 56.73 unknown

Solibacillus sylvestris HR3-23 1528 53.2 53.2

Uncultured Bacillus sp. clone DGG30 1569 53.218 unknown

73T 1408 53.91 unknown

Appendix 7: Recipes for plate assays for the detection of amylase, protease, lipase, pectinase, gelatinase

A: Amylase (http://www.bd.com/europe/regulatory/assets/ifu/difco_bbl/272100.pdf)

Per litre, 3 g beef extract, 10 g soluble starch, 12 g agar.

Post incubation, stain with 1:10 Lugol’s iodine

B: Protease (http://microbiologyonline.org/teachers/preparation-of-media-and-cultures)

Per litre, 20 g skim milk (2%), 12 g agar

C: Lipase

C1: Tween 80 (Samad et al., 1989)

Per litre, 10 mL Tween 80, 10 g peptone, 5 g NaCl, 0.1 g CaCl2 and 20 g agar

249

C2: Olive oil (Lee et al., 2015)

Per litre, 1 mL olive oil (0.1%), 1 g CaCl2 (0.1%), 0.1 g phenol red (0.01%) and 20 g agar

D: Pectinase (Usha et al., 2014)

Per litre, 10 g pectin, 1.5 g beef extract and 15 g agar

E: Gelatinase (Smith & Goodner, 1958)

Per litre: 28 g gelatine in nutrient agar

Appendix 8: Plates assay recipes for detection of azoreductase and laccase

Media recipes for azoreductase and laccase

A: Azoreductase (Leelakriangsak & Borisut, 2012)

Per litre: 600 mg methyl red in nutrient agar

B1: Laccase using guaiacol (Sheikhi et al., 2012)

0.5 mM guaiacol in nutrient agar

B2: bromothymol blue (Tekere et al., 2001; opennetware.org/wiki/laccase_protocols)

Per litre: 0.1 g bromothymol blue in nutrient agar

Appendix 9: Starch assay and standard curve

Standard curve for starch

250

A 1% stock solution of starch was made and boiled for 1.5 h to solubilise the starch. A 10-fold dilution series was made in water to give concentrations of 0.1, 0.01, 0.001, 0.0001 and 0.00001% in a total volume of 500 µL. 500 µL of 10% Lugol’s iodine was added to each tube and the colour reaction was read at 660 nm with a Bio-Rad 96-well iMark plate reader.

Appendix 10: Standard curve for bromothymol blue and OD at 595 nm

Doubling dilutions of bromothymol blue were made in water with a stock of 1%. Optical density was read at 595 nm on the Bio-Rad iMark 96-well reader to display the standard curve.

Appendix 11

Standard curves were drawn for nickel and iron from 0 to 6.25 mg/mL and read at OD 415 nm. Copper was measured from 0 to 100 mg/mL and read at 750 nm. Chromium was measured from 0 to 3.125 mg/mL and read at 415 nm.

Standard curves for heavy metals determined spectrophotometrically:

Appendix 11.1: Standard curve for nickel

251

Appendix 11.2: Standard curve for iron

Appendix 11.3: Standard curve for chromium at OD 415 nm

252

Appendix 11.4: Standard curve for copper at OD 750 nm

Appendix 12

Appendix 12.1: Standard curve for coffee and OD at 415 nm

253

Appendix 12.2: Standard curve for soya sauce and OD at 415 nm

Appendix 12.3: Standard curve for commercial textile dye (Dye It) and OD at 595 nm

254

Appendix 13

Appendix 13.1: Standard curve for dairy wastewater at OD 415 nm

Appendix 13.2: Standard curve for phenol concentrations in brewery wastewater at OD 700 nm

255

Appendix 14

Optimal growth condition of temperature, pH and salinity for isolates cultured at 55 C. * indicates growth at 10% NaCl

Isolation Optimal temperature temperature Optimal Optimal salinity % Isolate No Site (°C) Sample (°C) ph NaCl 1T Tshipise 55 water 55 7 nd 2T Tshipise 55 water 50 7 5 * 3T Tshipise 55 water 45 6 0 4T Tshipise 55 water 50 8 0 6T Tshipise 55 water 50 7 5 * 7T Tshipise 55 water 50 8 1.25 8T Tshipise 55 water 55 8 5 * 9T Tshipise 55 water 50 6 3.8 10T Tshipise 55 water 55 7 5 39T Tshipise 37-53 water 55 7 5 43T Tshipise 55 sediment 50 no grow nd 45T Tshipise 55 sediment 55 7 nd 50T Tshipise 55 sediment 55 7 nd 51T Tshipise 55 sediment 55 6 nd 12S Siloam 55 water 55 7 2.5 13S Siloam 55 water 60 7 0 15S Siloam 55 water 55 8 5 16S Siloam 55 water 50 7 0

256

17S Siloam 55 water 50 10 nd 19S Siloam 55 water 50 8 0 20S Siloam 55 water 55 9 nd 46S Siloam 55 sediment 55 7 nd 21M Mphephu 55 water 55 7 nd 22M Mphephu 55 water 45 9 nd 23M Mphephu 55 water 50 8 5 27M Mphephu 55 water no grw 7 nd 29M Mphephu 55 water 55 8 nd 30M Mphephu 55 water 55 6 nd 52M Mphephu 55 sediment 55 6 nd 33Li Libertas 55 water 55 7 5 34Li Libertas 55 water 50 7 nd 35Li Libertas 55 water 55 6 nd 36Li Libertas 55 water 55 8 nd 38Li Libertas 55 water 55 8 nd 47Li Libertas 55 sediment 45 6 nd 48Li Libertas 55 sediment 50 7 nd 49Li Libertas 55 sediment 50 7 nd 40Le Lekkerrus 37-53 water no grw 8 nd

257

APPENDIX 15: Antibiotic and heavy-metal resistance of 40 isolates from hot springs, Limpopo Province, South Africa, where (R) is resistant, (S) sensitive and (nd) not determined

ISOLATE NO. CAR* GEN* KAN* STRP* TET* CEF* CHL* COT* NA* NOR* Al10# Al40# Cr10# Cr40# Cu10# Cu40# Fe10# Fe40# Mn10# Mn40# Ni10# Ni40# Pb10# Pb40# Hg200# Anoxybacillus spp 7T S S S S S R S S R S R S R R R R R S R S R S S R R 3T S S S S S R S S R S R R R R R R R S R S R R R R S 4T S S S S S S S S R S R S R S R R R S R S R R R R R 17S S S S S S S S S R S nd nd nd nd nd nd nd nd nd nd nd nd nd nd nd 13S S S S S S S S S R S R R R R R R R S S S R R S R S 19S S S S S S S S S S S R S R R R R R S R S R R S R S Bacillus Bergey group A spp 2T R S S S S R S S R S R S R R R S R S R nd R R R R R 6T S S S S S R S S R S nd nd nd nd nd nd nd nd nd nd nd nd nd nd nd 30 S S S S S R S S R S nd nd nd nd nd nd nd nd nd nd nd nd nd nd nd 8T S S S S S R S S S S nd nd nd nd nd nd nd nd nd nd nd nd nd nd nd 10T S S S S S R S S S S R S R R R R R R R S R R R R S 15S S S S S S R S S S S nd S nd R nd R nd S nd R nd R nd R nd 23M S S S S S R S S S S S S R S R R R S R R R R R R S 39T S S S S S S S S R S R S R R R R R S R R R S R R R 54T R S S S S R S S S S S R nd R R R S R R R R R S R R 78S R S S S S R S S S S S R R R R R R S R R R S R R R 21M S S S R S R S S S S nd nd nd nd nd nd nd nd nd nd nd nd nd nd nd 48Li S S S S S R S S S S R S R R R R R R R R R R R S S 33Li S S S S S R S S S S R S R R R R R S R S R S R S R 47Li S S S S S R S S S S R S R S R R R S R R R R R R S 12S S S S S S S S S R S R S R R R R R R R S R S R R R 22M S S S S S S S S S S R R R R R S R S R R R S R R R 40Le S S S S S S S S S S R S R R R R R S R R R R R R S 77S S S S S S S S S S S R R R R R R R R R R R R R R S Aneurinibacillus spp 73T R S S S S S S S S S R R R R R R R S R R R R R R S 86Li R S S S S S S S S S nd nd nd nd nd nd nd nd nd nd nd nd nd nd nd Brevibacillus spp 16S S S R S S R S S S S R R R R R R R S R R R R R S R 36Li S S S S S S S S S S R S R R R R R S R S R R R R R 85Li S S S S S S S S S S R R R R S R R R R R R R R S S Bacillus spp 46S R S S S S S S S R S nd nd nd nd nd nd nd nd nd nd nd nd nd nd nd 51T R S S S S R S S S S nd nd nd nd nd nd nd nd nd nd nd nd nd nd nd 1T S S S S S R S S R S R S R R R S R S R S R R R R R 83Li S S S R S S S S S S R R R R R R S S R R R R R R R 50T S S S S S S S S S S nd nd nd nd nd nd nd nd nd nd nd nd nd nd nd 52M S S S S S S S S S S S S R R R R R S R S R S R S S 76S S S S S S S S S S S R S R R R R R S R R R R R R S Actinobacteria 57T S S S S S R S S R S S S R R R S S S R R R R S R R 58T R S S S S S S S R S R S R S R S S S R R S R R S S Proteobacteria 9T S S S S S R S S R S nd nd nd nd nd nd nd nd nd nd nd nd nd nd nd 79M R S S S S R S S S S R S R R R R R S S R R S R R R *Antibiotics: carbenicillin (CAR), gentamycin (GEN), kanamycin (KAN), streptomycin (STP), tetracycline (TET), chloramphenicol (CHL), ceftriaxone (CEF), cotrimoxazole (COT), nalidixic acid (NA) and norfloxacin (NOR) in µg/ml #Heavy metals at 10mmol, 40mmol and 200nmol (Hg only): aluminium (Al), chromium (Cr), copper (Cu), iron (Fe), mercury (Hg), manganese (Mn), nickel (Ni) and lead (Pb) in mmol (except for Hg

258

Appendix 16

Colouration as % of negative control of bromothymol blue treated with CFCS of hot spring isolates

Appendix 17

Turbidity as % of negative control of contaminated river water samples treated with CFCS of hot spring isolates

Appendix 18

Reduction of phenol in bromothymol blue and crystal violet

259

Appendix 19

Contiguous sequence of partial 16S rDNA for isolate 76S Bacillus mojavensis

GGCTGGCTCCATAAAGGTTACCTCACCGACTTCGGGTGWTACAAACTCTGtggtgtgacgggcggtgtgtacaaggcccg ggaacgtattcaccgcggcatgctgatccgcgattactagcgattccagcttcacgcagtcgagttgcagactgcgatccgaacTGAGAACAGATTTGTGG GATTGGCTTAACCTCGCGGTTTCGCTGCCCTTTGTTCTGTCCATTGTAGCACGTGTGTAGCCCAGGTCATA AGGGGCATGATGATTTGACGTCATCCCCACCTTCCTCCGGTTTGTCACCGGCAGTCACCTTAGAGTGCCC AACTGAATGCTGGCAACTAAGATCAAGGGTTGCGCTCGTTGCGGGACTTAACCCAACATCTCACGACAC GAGCTGACGACAACCATGCACCACCTGTCACTCTGCCCCCGAAGGGGACGTCCTATCTCTAGGATTGTC AGAGGATGTCAAGACCTGGTAAGGTTCTTCGCGTTGCTTCGAATTAAACCACATGCTCCACCGCTTGTGC GGGCCCCCCGTCAATTCCTTTGAGTTTCAGTCTTGCGACCGTACTCCCCAGGCGGAGTGCTTAATGCGTT AGCTGCAGCACTAAGGGGCGGAAACCCCCTAACACTTAGCACTCATCGTTTACGGCGTGGACTACCAGG GTATCTAATCCTGTTCGCTCCCCACGCTTTCGCTCCTCAGCGTCAGTTACAGACCAGAGAGTCGCCTTCG CCACTGGTGTTCCTCCACATCTCTACGCATTTCACCGCTACACGTGGAATTCCACTCTCCTCTTCTGCACT CAAGTTCCCCAGTTTCCAATGACCCTCCCCGGTTGAGCCGGGGGCTTTCACATCAGACTTAAGGAACCGC CTGCGAGCCCTTTACGCCCAATAATTCCGGACAACGCTTGCCACCTACGTATTACCGCGGCTGCTGGCAC GTAGTTAGCCGTGGCTTTCTGGTTAGGTACCGTCAAGGTACCGCCCTATTCGAACGGTACTTGTTCTTCC CTAACAACAGAGCTTTACGATCCGAAAACCTTCATCACTCACGCGGCGTTGCTCCGTCAGACTTTCGTCC ATTGCGGAAGATTCCCTACTGCTGCCTCCCGTAGGAGTCTGGGCCGTGTCTCAGTCCCAGTGTGGCCGAT CACCCTCTCAGGTCGGCTACGCATCGTTGCCTTGGTGAGCCATTACCTCACCAACTAGCTAATGCGCCGC GGGTCCATCTGTAAGTGGTAGCCGAAGCCACCTTTTATGTTTGAACCATGCGGTTCAAACAAGCATCCG GTATTAGCCCCGGTTTCCCGGAGTTATCCCAGTCTTACAGGCAGGTTACCCACGTGTTACTCACCCGTCC GCCGCTAACATCAGGGAGCAAGCTCCCATCTGTCCGCTCGACTTGCATGTATTAGGCACGCCGCCAGCG TTCGTCCTGAGCWAKTYAAAAAYYTYATA

260