Development and Application of Prokaryotic Biosensor Systems for the Evaluation of Toxicity of Environmental Water Samples

Home , Anne Glover (biologist)

DEVELOPMENT AND APPLICATION OF PROKARYOTIC BIOSENSOR SYSTEMS FOR THE EVALUATION OF TOXICITY OF ENVIRONMENTAL WATER SAMPLES

Report to the Water Research Commission by

B PILLAY1 and D PILLAY2

1Discipline of Microbiology Faculty of Science and Agriculture University of KwaZulu-Natal Westville Campus

2Department of Biotechnology Faculty of Engineering, Science and the Built Environment Durban University of Technology

WRC Report No 1286/1/07 ISBN 978-1-77005-538-4

DECEMBER 2007 Obtainable from: [email protected]

Water Research Commission Private Bag X03 Gezina, Pretoria 0031 South Africa

This publication emanates from a Water Research Commission research project entitled “Application of Biosensors for the eco-toxicity testing of water sources” (WRC project number K5/1286).

DISCLAIMER

This report has been reviewed by the Water Research Commission (WRC) and approved for publication. Approval does not signify that the contents necessarily reflect the views and policies of the WRC, nor does mention of trade names or commercial products constitute endorsement or recommendation for use.

TABLE OF CONTENTS

TABLE OF CONTENTS………………………………………………………………. i LIST OF FIGURES...... iv LIST OF TABLES...... x ACKNOWLEDGEMENTS...... xii EXECUTIVE SUMMARY…………………………………………………………….. xiii

CHAPTER ONE: SCOPE OF THIS STUDY…………………………………………. 1 1.1 GENERAL BACKGROUND AND MOTIVATION……………………..…….. 1 1.2 ORIGINAL OBJECTIVES OF THE STUDY …………………………………. 1 1.3 FINAL OBJECTIVES OF THE STUDY……………………………..………… 2 1.4 LAYOUT OF THE REPORT……………………………..…………….……….. 3

CHAPTER TWO: LITERATURE REVIEW...... …………………………………... 4 2.1 INTRODUCTION……………………………………………………………….. 4 2.2 ENVIRONMENTAL POLLUTION...... 5 2.3 FUNCTIONS ENCODED BY THE lux GENES...... 9 2.3.1 Luciferase...... 9 2.3.2 Aldehyde biosynthesis...... 10 2.4 APPLICATIONS OF lux GENES...... 10 2.4.1 Fusions of lux genes to specific stress promoters...... 11 2.4.1.1 fabA: lux-based biosensor...... 11 2.4.1.2 recA: lux-based biosensor...... 13 2.4.1.3 katG: lux-based biosensor...... 13 2.4.1.4 uspA: lux and grpE:lux-based biosensors...... 14 2.4.2 ipb:lux-based biosensor...... 15 2.4.3 pUCD607 based biosensor systems...... 15 2.4.3.1 BTEX contaminated sites...... 16 2.4.3.2 Mini-Tn5 luxCDABE transposon...... 18 2.4.3.3 Toxicity of chlorophenols...... 19 2.4.3.4 Copper availability in malt whisky distillery effluent...... 19 2.4.3.5 Assessment of the bioavailability of heavy metals...... 20 2.4.3.6 Toxicity of herbicides in freshwater...... 21 2.4.4 Versatile biosensors for the detection of mercury and arsenic...... 21 2.4.4.1 mer:lux-based biosensors...... 21 2.4.4.2 ars:luxAB-based biosensor...... 22

2.4.5 tfdRPDII ¯ luxCDABE-based biosensor...... 23 2.4.6 lux: nah fusion...... 23

i 2.4.7 Bioavailability of middle-chain alkanes in groundwater...... 24 CHAPTER THREE: DESIGN AND CONSTRUCTION OF PROKARYOTIC BIOSENSOR SYSTEMS...... 26 3.1 INTRODUCTION...... 26 3.2 MATERIALS AND METHODS...... 28 3.2.1 Growth and maintenance of bacterial cultures...... 28 3.2.2 Plasmid DNA isolation...... 28 3.2.3 Cloning of pLux...... 30 3.2.3.1 Extraction of restriction fragments...... 31 3.2.3.2 Ligation of DNA fragments...... 31 3.2.3.3 Preparation of electrocompetent cells...... 32 3.2.3.4 Transformation...... 32 3.2.4 Screening of transformed cells...... 33 3.2.4.1 Restriction analysis...... 33 3.2.4.2 Polymerase chain reaction amplification...... 34 3.2.5 Assaying for expression of the luxCDABE operon...... 34 3.2.6 Freeze-drying of prokaryotic biosensor cells...... 35 3.2.7 Resuscitation of freeze-dried prokaryotic biosensor cells...... 35 3.3 RESULTS...... 36 3.3.1 Cloning of pLux...... 36 3.3.2 Isolation of plasmid DNA...... 36 3.3.3 Polymerase chain reaction amplification...... 37 3.3.4 Construction of prokaryotic biosensor systems...... 37 3.3.5 Assaying for expression of the luxCDABE operon...... 38 3.3.6 Resuscitation of freeze-dried prokaryotic biosensor cells...... 38 3.4 DISCUSSION...... 39

CHAPTER FOUR: STANDARDISATION OF THE TOXICITY TESTS...... 41 4.1 INTRODUCTION...... 41 4.2 MATERIALS AND METHODS...... 42 4.2.1 Preparation of toxicant samples...... 42 4.2.2 Toxicity test...... 43 4.2.3 Calculation of bioluminescence and EC values...... 45 4.3 RESULTS...... 46 4.3.1 Selection of the most sensitive biosensor systems……………………… 48 4.3.2 S. sonnei pLux biosensor...... 53 4.3.3 E. coli DH5α uspA: lux biosensor...... 57 4.3.4 V. fischeri-based BioToxTM kit...... 61

4.3.5 Daphnia LC50 toxicity test...... 64 4.4 DISCUSSION...... …………………………… 65

CHAPTER FIVE: APPLICATION OF THE PROKARYOTIC BIOSENSOR SYSTEMS...... 69 5.1 INTRODUCTION...... 69 5.2 MATERIALS AND METHODS...... 70 5.2.1 Collection of wastewater effluent samples...... 70 5.2.2 Analysis of wastewater effluent samples...... 71 5.2.3 Calculation of bioluminescence and EC values...... 71 5.3 RESULTS...... 72 5.3.1 S. sonnei pLux biosensor...... 74 5.3.2 E. coli DH5α uspA: lux biosensor...... 75 5.3.3 V. fischeri-based BioToxTM kit...... 79 5.4 DISCUSSION...... 81

CHAPTER SIX: CONCLUSIONS AND RECOMMENDATIONS...... 84 6.1 SUMMARY OF MAJOR FINDINGS AND CONCLUSIONS REACHED.…. 84 6.2 RECOMMENDATIONS FOR FUTURE USE...... 85 6.3 RECOMMENDATIONS FOR TECHNOLOGY TRANSFER……………….. 86

REFERENCES...... 87

APPENDIX 1...... 92

APPENDIX 2...... 120

APPENDIX 3...... 123

iii LIST OF FIGURES Fig. 2.1 Activities that result in groundwater contamination (Prescott et al., 1996)...... 6

Fig. 2.2 An aerial photograph of the Amanzimtoti Wastewater Treatment Works (Ethekwini Wastewater, 2003)...... 6

Fig. 2.3 A schematic representation of a typical wastewater treatment works, showing the various treatment processes (Richman, 1997)...... 8

Fig. 2.4 The bioluminescent reactions encoded for by the luxCDABE operon. The luxAB genes convert an aldehyde substrate to a carboxyl group, generating visible light (Equation 1).

The luxCDE genes use NADPH2 and ATP to generate the aldehyde (Equation 2) (Burlage and Kuo, 1994)...... 10

Fig. 2.5 An Atlantic flashlight fish which contains luminous bacteria in the light organ under its eye (Prescott et al., 1996)...... 12

Fig. 2.6 Daphnia pulex (water flea) is a tiny shellfish with a transparent body covering. It is used in the Daphnia LC50 toxicity test at Umgeni Water as a standard bioassay (World Book, 1992)...... 12

Fig. 2.7 The pUCD607 plasmid which contains the V. fischeri luxCDABE operon (Shaw and Kado, 1986)...... 16

Fig. 2.8 P. fluorescens engineered to express the bacterial lux operon constitutively by using the pUCD607 plasmid. The flask on the left was photographed in the light. Although the flask on the right was photographed in the dark, it is visible due to emission of bioluminescence (Belkin et al., 1996)...... 17

Fig. 2.9 Bioluminescence of P. fluorescens /pUCD607 in response to increasing concentrations of toxin, e.g., copper sulphate (Belkin et al., 1996)...... 17

Fig. 3.1 Partial restriction map of the 11 335 bp pLux plasmid which contains the luxCDABE operon ...... 36

Fig. 3.2 Restriction analysis of pLux and plasmid DNA of the E. coli DH5α pLux, E. coli 38 DH5α recA: lux, E. coli DH5α fabA:lux, E. coli DH5α uspA:lux biosensor systems. Lane 1: pUC19 restricted with Sal I; lane 2: pUC19; lane 3: pUCD607 restricted with Sal I; lane 4: pUCD607; lane 5: λ DNA restricted with Hind III (molecular weight marker M II); lane 6: pLux restricted with Sal I; lane 7: pLux; lane 8: pRecALux; lane 9: pFabALux; and lane 10: pUspALux2......

Fig. 3.3 Polymerase chain reaction amplification of the plasmids of the E. coli DH5α pLux, E. coli DH5α recA: lux, E. coli DH5α fabA: lux, E. coli DH5α uspA: lux biosensor systems. Lane 1: negative control (without plasmid DNA); lanes 2 - 4: positive control (pUCD607); lanes 5 - 7: pLux; lanes 8 - 10: pRecALux; lane 11: pBR328 DNA restricted with Bgl I + pBR328 DNA restricted with Hinf I (molecular weight marker VI); lanes 12 - 14: pFabALux; and lanes 15 -17: pUspALux2...... 39

Fig. 4.1 A schematic representation of the Escherichia coli soft metal ion-translocating ATPases (Gatti et al., 2000)...... 42

Fig.4.2 Microtitre plate template used in the Fluoroskan Ascent FL where A, B, C or D represent the various pollutant samples...... 44

Fig 4.3 Bioluminescent response of the S. sonnei pLux biosensor in the presence of various concentrations of chromium trioxide……...... 48

Fig 4.4 Bioluminescent response of the S. flexneri pLux biosensor the presence of various concentrations of chromium trioxide...... 48

Fig 4.5 Bioluminescent response of the E. coli DH5α pLux biosensor in the presence of various concentrations of chromium trioxide...... 49

Fig 4.6 Bioluminescent response of the Enteropathogenic E. coli pLux biosensor in the presence of various concentrations of chromium trioxide...... 49

Fig 4.7 Bioluminescent response of the E. coli DH5α recA: lux biosensor in the presence of various concentrations of chromium trioxide...... 49

Fig 4.8 Bioluminescent response of the E. coli DH5α fabA: lux biosensor in the presence of various concentrations of chromium trioxide...... 50

Fig 4.9 Bioluminescent response of the E. coli DH5α uspA: lux biosensor in the presence of various concentrations of chromium trioxide...... 50

Fig 4.10 Bioluminescent response of V. fischeri to various concentrations of chromium trioxide...... 50

Fig 4.11 Bioluminescent response of the S. sonnei pLux biosensor in the presence of various concentrations of xylene...... 51

v Fig 4.12 Bioluminescent response of the S. flexneri pLux biosensor the presence of various concentrations of xylene...... 51

Fig 4.13 Bioluminescent response of the E. coli DH5α pLux biosensor in the presence of various concentrations of xylene...... 51

Fig 4.14 Bioluminescent response of the E. coli DH5α recA: lux biosensor in the presence of various concentrations of xylene...... 52

Fig 4.15 Bioluminescent response of the E. coli fabA: lux biosensor in the presence of various concentrations of xylene...... 52

Fig 4.16 Bioluminescent response of the E. coli uspA: lux biosensor in the presence of various concentrations of xylene...... 52

Fig 4.17 Bioluminescent response of the Enteropathogenic E. coli pLux biosensor in the presence of various concentrations of xylene...... 53

Fig 4.18 Bioluminescent response of V. fischeri in the presence of various concentrations of xylene...... 53

Fig 4.19 Bioluminescent response of the S. sonnei pLux biosensor in the presence of various concentrations of lead acetate...... 54

Fig 4.20 Bioluminescent response of the S. sonnei pLux biosensor in the presence of various concentrations of zinc sulphate...... 55

Fig 4.21 Bioluminescent response of the S. sonnei pLux biosensor in the presence of various concentrations of copper sulphate...... 55

Fig 4.22 Bioluminescent response of the S. sonnei pLux biosensor in the presence of various concentrations of cadmium acetate...... 55

Fig 4.23 Bioluminescent response of the S. sonnei pLux biosensor in the presence of various concentrations of potassium dichromate...... 56

Fig 4.24 Bioluminescent response of the S. sonnei pLux biosensor in the presence of various concentrations of ethylbenzene...... 56

vi Fig 4.25 Bioluminescent response of the S. sonnei pLux biosensor in the presence of various 56 concentrations of toluene......

Fig. 4.26 Bioluminescent response of the S. sonnei pLux biosensor in the presence of various concentrations of benzene...... 57

Fig 4.27 Bioluminescent response of the E. coli DH5α uspA: lux biosensor in the presence of various concentrations of copper sulphate...... 58

Fig 4.28 Bioluminescent response of the E. coli DH5α uspA: lux biosensor in the presence of various concentrations of cadmium acetate...... 58

Fig 4.29 Bioluminescent response of the E. coli DH5α uspA: lux biosensor in the presence of various concentrations of zinc sulphate...... 58

Fig 4.30 Bioluminescent response of the E. coli DH5α uspA: lux biosensor in the presence of various concentrations of lead acetate...... 59

Fig 4.31 Bioluminescent response of the E. coli DH5α uspA: lux biosensor in the presence of various concentrations of potassium dichromate...... 59

Fig 4.32 Bioluminescent response of the E. coli DH5α uspA: lux biosensor in the presence of various concentrations toluene...... 60

Fig 4.33 Bioluminescent response of the E. coli DH5α uspA: lux biosensor in the presence of various concentrations of ethylbenzene...... 60

Fig 4.34 Bioluminescent response of the E. coli DH5α uspA:lux biosensor in the presence of various concentrations of benzene...... 60 . Fig 4.35 Bioluminescent response of V. fischeri in the presence of various concentrations of cadmium acetate...... 62

Fig 4.36 Bioluminescent response of V. fischeri in the presence of various concentrations of lead acetate...... 62

Fig 4.37 Bioluminescent response of V. fischeri in the presence of various concentrations of copper sulphate...... 62

vii Fig 4.38 Bioluminescent response of V. fischeri in the presence of various concentrations of potassium dichromate...... 63

Fig 4.39 Bioluminescent response of V. fischeri in the presence of various concentrations of zinc sulphate...... 63

Fig 4.40 Bioluminescent response of V. fischeri in the presence of various concentrations of ethylbenzene...... 63

Fig 4.41 Bioluminescent response of V. fischeri in the presence of various concentrations of toluene...... 64

Fig 4.42 Bioluminescent response of V. fischeri in the presence of various concentrations of benzene...... 64

Fig. 5.1 Geographical locations (●) of the five wastewater treatment works investigated in this study...... 71

Fig. 5.2 Bioluminescent response of the S. sonnei pLux biosensor in the presence of various concentrations of the New Germany wastewater effluent sample...... 75

Fig. 5.3 Bioluminescent response of the S. sonnei pLux biosensor in the presence of various concentrations of the Kwa-Mashu wastewater effluent sample...... 76

Fig. 5.4 Bioluminescent response of the S. sonnei pLux biosensor in the presence of various concentrations of the Northern wastewater effluent sample...... 76

Fig. 5.5 Bioluminescent response of the S. sonnei pLux biosensor in the presence of various concentrations of the Phoenix wastewater effluent sample...... 76

Fig. 5.6 Bioluminescent response of the S. sonnei pLux biosensor in the presence of various concentrations of the Amanzimtoti wastewater effluent sample...... 77

Fig. 5.7 Bioluminescent response of the E. coli DH5α uspA: lux biosensor in the presence of various concentrations of the New Germany wastewater effluent sample...... 77

Fig. 5.8 Bioluminescent response of the E. coli DH5α uspA: lux biosensor in the presence of various concentrations of the Kwa-Mashu wastewater effluent sample...... 77

viii Fig. 5.9 Bioluminescent response of the E. coli DH5α uspA: lux biosensor in the presence of various concentrations of the Northern wastewater effluent sample...... 78

Fig. 5.10 Bioluminescent response of the E. coli DH5α uspA: lux biosensor in the presence of various concentrations of the Phoenix wastewater effluent sample...... 78

Fig. 5.11 Bioluminescent response of the E. coli DH5α fabA: lux biosensor in the presence of various concentrations of the Amanzimtoti wastewater effluent sample...... 78

Fig. 5.12 Bioluminescent response of V. fischeri in the presence of various concentrations of the New Germany wastewater effluent sample...... 80

Fig. 5.13 Bioluminescent response of V. fischeri in the presence of various concentrations of the Kwa-Mashu wastewater effluent sample...... 80

Fig. 5.14 Bioluminescent response of V. fischeri in the presence of various concentrations of the Northern wastewater effluent sample...... 80

Fig. 5.15 Bioluminescent response of V. fischeri in the presence of various concentrations of the Phoenix wastewater effluent sample...... 81

Fig. 5.16 Bioluminescent response of V. fischeri in the presence of various concentrations of the Amanzimtoti wastewater effluent sample...... 81

ix LIST OF TABLES Table 2.1 Health implications of heavy metals and organic pollutants (Holmes, 1996)...... 7

Table 2.2 General quality limits for acceptance of trade effluent for discharge into the sewage disposal system and sea outfalls (Ethekwini Wastewater, 2003)...... 8

Table 2.3 Toxicity of various chemicals to E. coli strains DPD2544 and DPD2543, as a b determined using EC50 and EC200 values ( Kaiser and Palabrica, 1999; Bechor et al., 2002)...... 13

Table 2.4 Comparison of the E. coli uspA: lux and grpE: lux-based biosensors (Van Dyk et al., 1995)...... 15

Table 2.5 EC50 values of the two Pseudomonas Tn5 luxCDABE biosensors (Weitz et al., 2001)...... 19

Table 2.6 Copper concentrations and percentage bioluminescence of contaminated sites (Paton et al., 1995)...... 20

Table 2.7 Comparison of mean EC50 values of P. fluorescens strains 10586s/FAC510 and 10586s/pUCD607, in the presence of selected potentially toxic chemicals (Paton et al., 1995)...... 21

Table 2.8 Inhibition of octane-induced luminescence in E. coli DH5α (pGEC74, pJAMA7) between different groups of related compounds (Sticher et al., 1997)...... 25

Table 3.1 Bacterial strains used in this study...... 29

Table 3.2 Antibiotics and other media supplements used in this study...... 29

Table 3.3 Plasmids used or constructed in this study...... 33

Table 3.4 PCR parameters for amplification of the luxA gene...... 34

Table 3.5 Prokaryotic biosensor systems constructed in this study...... 37

Table 4.1 The various heavy metal compounds used in this study...... 43

Table 4.2 The various organic solvents used in this study (Cole, 1994)...... 45

x Table 4.3 EC50 values of the luxCDABE-marked bacterial systems for the heavy metal compounds...... 46

Table 4.4 EC50 values of the luxCDABE-marked bacterial systems for the volatile organic 47 solvents......

Table 4.5 *EC20 values of the E. coli DH5α pLux biosensor for the volatile organic solvents...... 47

Table 4.6 *EC100 values of the luxCDABE-marked bacterial biosensors for the volatile organic solvents...... 47

Table 4.7 Daphnia LC50 toxicity test results obtained by the Umgeni Water Microbiology Laboratory after 48h, using Daphnia pulex as the eukaryotic test organism...... 65

TM Table 5.1 EC50 values of S. sonnei pLux and the V. fischeri-based BioTox kit for the five wastewater effluent samples...... 73

Table 5.2 *EC20 values of E. coli DH5α pLux for the five wastewater effluent samples...... 73

Table 5.3 *EC100 values of E. coli DH5α luxCDABE-based biosensors for the five wastewater effluent samples...... 73

Table 5.4 Analyses performed by Umgeni Water Analytical Services Department for the five wastewater effluent samples...... 74

xi ACKNOWLEDGEMENTS

The Steering Committee responsible for this project consisted of the following persons:

Mr HM du Plessis Water Research Commission (Chairperson) Prof J Lin University of Durban-Westville Prof R Bharathram University of Durban-Westville Mr IW Bailey Umgeni Water Mrs K Milford Umgeni Water Ms T Zokufa Institute for Water Quality Studies, Dept. of Water Affairs and Forestry Dr S Oelofse Water Quality Management, Dept. of Water Affairs and Forestry Mr DK Chetty Durban Institute of Technology Mr K Permual Durban Institute of Technology Ms JL Slabbert Environmentek, CSIR

The financing of the project by the Water Research Commission and the contribution of the members of the Steering Committee is gratefully acknowledged.

The authors, furthermore, wish to record their sincere thanks to:

Umgeni Water for the chemical analyses. Mr Mohammed Dildar and staff at the eThekwini Wastewater Works for their assistance in the collection of wastewater samples. Dr Tina Van Dyk from I. E. du Pont de Nemours and Company, Wilmington, United States of America, for generously supplying the stress promoters used in this research. Dr Anne Glover from the University of Aberdeen, Scotland, for providing the pUCD607 plasmid. All members of the Research team and students, especially Ms D Seepersad, Mr ND Sanpal, Mr A Govender, Ms S S Naiker and Dr V Guddera, for their contributions to this project.

xii EXECUTIVE SUMMARY

General background and motivation of project Pollution of water systems with chemicals and heavy metals poses a severe threat to human health and is of serious environmental concern. The detection of pollutants in the environment is time-consuming and expensive. Bacterial biosensors expressing the lux gene provide an alternative means of pollutant detection in the environment. This is made possible by the versatility of the metabolic and physicochemical characteristics of microorganisms. These biosensors offer a simple and convenient method to measure the acute toxicity of pollutants and are efficient tools in determining changes associated with complex chemical mixtures undergoing bioremediation. Microbial biosensors offer many advantages over chemical methods and other methods of ecotoxicity testing. Assays using microbial biosensors are rapid, sensitive, reproducible and cost-effective. These tests, based on the production of light by bacteria, reflect the effect of bio-available pollutants on the metabolic activity of the cells. The amount of light emitted is an indication of the presence of non-toxic or toxic substances which may induce or inhibit metabolism, respectively.

Original objectives of the study i) To conduct a comparative study of existing luminescence-based assays in order to evaluate the cost-effectiveness as well as the relevance and appropriateness of these assays for the: Determination of the bio-availability of pollutants in soil, groundwater and other water sources, Detection of viable-but-non-culturable bacteria in soil, water supplies, groundwater and other water sources, and Assessment of remediation and bioremediation potential;

It can be concluded from the literature survey that bacterial hosts incorporating the luxCDABE operon have proven to be a rapid and sensitive reporter system for the detection and monitoring of pollutants in environmental samples in developed countries. Therefore, the development of prokaryotic biosensors for local application would, in addition, also provide a cost-effective method of detecting environmental pollutants.

ii) To develop luminescence-based assays that would be appropriate, and cost- effective for the above applications; iii) To compare the luminescence-based assay with standard water testing methods; and iv) To transfer technology and develop local skills in luminescence-based assays. Among the above-mentioned original objectives, (i) was covered in the Literature Review (Chapter Two), while (ii) - (iv) were addressed in the final objectives below.

xiii Final objectives of the study

i) To construct prokaryotic biosensor systems using the Vibrio fischeri luxCDABE operon. ii) To construct Escherichia coli DH5α, Shigella sonnei, Shigella flexneri, and Enteropathogenic E. coli biosensor systems using lac, recA, fabA and uspA promoters fused to the V. fischeri luxCDABE operon; iii) To evaluate the sensitivity (minimum detection levels) of the biosensors’ in the presence of laboratory standards of heavy metals and chemical pollutant. iii) To compare the sensitivity of all the biosensor systems to existing acute toxicity

tests, e.g., the Daphnia LC50 toxicity test and the Vibrio fischeri-based BioToxTM kit. v) To evaluate the sensitivity of the bacterial biosensors to heavy metals and chemical pollutants in wastewater effluent samples, from various Ethekwini Wastewater Treatment Works.

All the final objectives of this study were met. It should be pointed out that the final objectives represent a slight modification of the original objectives since these changes make them more relevant in terms of the outcomes of this research project. The approach to this study commenced with a literature review in Chapter Two. Thereafter, the study focused on achieving the final objectives which are reported in Chapters Three (i and ii), Four (iii and iv) and Five (v).

Methodology Conventional methods and molecular methods were used. The conventional methods used were standard microbiological techniques, freeze-drying and measurement of bioluminescence. Molecular methods included the polymerase chain reaction (PCR) and standard molecular cloning techniques.

Summary of major findings and conclusions reached Several different bacterial biosensors with the ability to emit a readily detectable signal (light) in the presence of a wide range of environmental pollutants were developed. These included luxCDABE-containing E. coli DH5α, Enteropathogenic E. coli, S. flexneri and S. sonnei bacterial biosensor systems. The biosensors represent a fusion of bacterial bioluminescence (lux) genes, as a reporter, to selected bacterial gene promoters. The promoterless V. fischeri luxCDABE operon, responsible for light output, from pUCD607, was successfully integrated into the cloning vector pUC19, to yield the multi-copy 11 335 bp pLux plasmid.

xiv Plasmids pRecALux, pFabALux and pUspALux2 carrying the lux gene fused to stress inducible promoters were isolated from E. coli DPD2794, E. coli DPD2540 and E. coli DE135, respectively.

E. coli DH5α, Enteropathogenic E. coli, S. flexneri and S. sonnei were successfully transformed with the plasmids pLux, pRecALux, pFabALux and pUspALux2 to create the bioluminescent biosensors: E. coli DH5α pLux, Enteropathogenic E.coli pLux, S. flexneri pLux, E. coli DH5α recA:lux, E. coli DH5α fabA:lux, E. coli DH5α uspA:lux and S. sonnei pLux.

The prokaryotic biosensors were successfully freeze-dried using trehalose and Luria Bertani (LB) broth. Freeze-drying in trehalose consistently yielded uniform and stable freeze-dried products and only required 30 min of resuscitation, without agitation. It seemed that trehalose maintained the viability and biosensing activity of these biosensors, as seen by the high bioluminescence values after resuscitation in LB broth.

E. coli DH5α pLux, Enteropathogenic E.coli pLux, S. flexneri pLux, and S. sonnei pLux exhibited a general inhibition in bioluminescence in the presence of heavy metals. However, the biosensors containing the stress inducible promoters viz., E. coli DH5α recA:lux, E. coli DH5α fabA:lux and E. coli DH5α uspA:lux exhibited a general induction in biolumescence at low concentrations of heavy metals.

The different biosensors exhibited varying degrees of toxicity to the range of heavy metals tested. Cu (II) was most toxic to E. coli DH5α fabA: lux, E. coli DH5α uspA: lux, S. flexneri pLux, and Enteropathogenic E.coli pLux. Cr (VI) was most toxic to E. coli DH5α pLux while Zn (II) was most toxic to E. coli DH5α recA: lux.

A comparison of the EC50 and LC50 with heavy metal compounds indicated that S. sonnei pLux was more sensitive than the commercially available Vibrio fischeri-based BioToxTM kit and the traditional ecotoxicity test using Daphnia.

The toxic effect of the different BTEX compounds varied among the biosensors. Xylene was most toxic and benzene least toxic to E. coli DH5α pLux, E. coli DH5α fabA: lux, E. coli DH5α uspA: lux and S. sonnei pLux. Xylene was also most toxic to S. flexneri pLux and Enteropathogenic E.coli pLux, while toluene was least toxic to these biosensors. In the case of E. coli DH5α recA: lux, ethylbenzene was most toxic and toluene least toxic.

Effluent from the New Germany, Kwa-Mashu, Phoenix, Northern and Amanzimtoti wastewater treatment works induced bioluminescence in E. coli DH5α recA:lux, E. coli

xv DH5α fabA:lux and E. coli DH5α uspA:lux. However, the same wastewater effluent samples inhibited E. coli DH5α pLux, S. flexneri pLux, Enteropathogenic E. coli pLux and S. sonnei pLux.

All biosensors were able to detect pollutants in the wastewater effluent samples at

concentrations too low for detection with the Daphnia LC50 toxicity test.

The data generated in this research demonstrates that the biosensors constructed in this study are potentially useful for the evaluation of environmental water samples and pollution management.

Review of the project in terms of the final objectives All five of the final objectives of the study were achieved.

Recommendations for future research Biosensors constructed in this study have the potential application to monitor environmental pollution. These whole cell biosensors hold a great deal of promise for continuous on-line monitoring of pollutant concentrations in environmental applications.

The main application of these biosensor systems may be for the prescreening of environmental samples for toxic agents. Suspicious findings that indicate the presence of pollutants may then be verified using established physico-chemical methods utilised at environmental laboratories. This will reduce the costs of standard toxicity tests, e.g.,

Daphnia toxicity LC50 test. Therefore, efforts to design portable field devices that incorporate the biosensors constructed in this study will form the basis for future research.

The ultimate aim will be to design a kit which can be used in the field by semi-skilled people with a basic microbiological background. The freeze-dried biosensors can be resuscitated on-site and the bioluminescent response of the wastewater samples measured directly using the portable 1254 - 001 LUMINOVA luminometer (Bio-Orbit Oy, Finland).

Probably one of the greatest advantages of using these biosensors in toxicity evaluation of environmental pollution is that they can indicate the bioavailability of pollutants in a way that chemical analyses cannot. Continued improvement of these biosensors can meet the urgent need to not only quantify bioavailable pollutants, but also to perform in situ monitoring of biodegradation (Neilson et al., 1999). These biosensors have the

xvi potential to offer a risk assessment strategy to predict the level to which a contaminated site may be bioremediated.

The growing interest in employing whole-cell biosensors for the early detection of specific substance reinforces other potential uses apart from bioremediation and bioavailability assays. These applications include environmental hazard evaluations; prosecution and defence of chemical-related activities in environmental litigation; management of the discharge of municipal and industrial waste; and corporate industrial decisions on product development, manufacture and commercialisation so as to avert potential pollution. Therefore, the successful integration of the powerful applications of biosensor technology in pollution management may be one alternative to reverse the years of global environmental mistreatment.

Recommendations for technology transfer A hands-on workshop of 1-2 days for all stakeholders in the water, waste-water and health sectors.

A simplified manual of this method and its applications should be published and distributed to all stakeholders.

Development of a low-cost and user-friendly kit for on-site use.

Capacity-building and corrective action WRC-funding of this project has contributed to both capacity-building and corrective action. As a HDI the former University of Durban-Westville catered largely for the needs of disadvantaged students. The research initiatives undertaken within the scope of this project have been in keeping with the University’s Mission Statement as well as our Departmental Mission Statement, i.e. to develop and train scientists from previously disadvantaged communities and the establishment of a research culture at the University.

The following were achieved: Research culture: A vibrant and sustainable research culture was developed within the Department of Microbiology. The project facilitated the development and training of disadvantaged students and staff. Furthermore, technology was transferred from other institutions and research collaboration was fostered. This project has also contributed to increasing the critical mass of our Departmental research team.

xvii Students: Postgraduate student numbers have increased as a consequence of the implementation of the project. Twenty nine postgraduate students have successfully completed their studies within this project. Details of these students are included in Appendix Two.

Improvements in primary, secondary and higher education: Through the Microbiology Students we have been able to make Microbiology accessible to schools. Pupils, Science teachers and Career Counselors are invited to the Department and exposed to various aspects of Microbiology (curriculum, entrance requirements, job opportunities and entrepreneurial opportunities.

Community outreach: A research programme with relevance to the wider community has been established in Water Microbiology. This will now form one of the focus areas of the Department. Under-developed areas with poor sanitation and contaminated water supplies will be targeted with a view to improving the quality of life for disadvantaged sectors of the population.

Publications and conference proceedings These are listed in Appendix Three.

xviii CHAPTER ONE: SCOPE OF THIS STUDY

1.1 GENERAL BACKGROUND AND MOTIVATION Pollution of water systems with chemicals and heavy metals poses a severe threat to human health and is of serious environmental concern. The detection of pollutants in the environment is time-consuming and expensive. Bacterial biosensors expressing the lux gene provide an alternative means of pollutant detection in the environment. This is made possible by the versatility of the metabolic and physicochemical characteristics of microorganisms. These biosensors offer a simple and convenient method to measure the acute toxicity of pollutants and are efficient tools in determining changes associated with complex chemical mixtures undergoing bioremediation. Microbial biosensors offer many advantages over chemical methods and other methods of ecotoxicity testing. Assays using microbial biosensors are rapid, sensitive, reproducible and cost-effective. These tests, based on the production of light by bacteria, reflect the effect of bio-available pollutants on the metabolic activity of the cells. The amount of light emitted is an indication of the presence of non-toxic or toxic substances which may induce or inhibit metabolism, respectively.

1.2 ORIGINAL OBJECTIVES OF THE STUDY 1.2.1 To conduct a comparative study of existing luminescence-based assays in order to evaluate the cost-effectiveness as well as the relevance and appropriateness of these assays for the: 1.2.1.1 Determination of the bio-availability of pollutants in soil, groundwater and other water sources, 1.2.1.2 Detection of viable-but-non-culturable bacteria in soil, water supplies, groundwater and other water sources, and 1.2.1.3 Assessment of remediation and bioremediation potential;

Therefore, the development of prokaryotic biosensors for local application would, in addition, also provide a cost-effective method of detecting environmental pollutants.

- 1 - 1.2.2 To develop luminescence-based assays that would be appropriate, and cost-effective for the above applications; 1.2.3 To compare the luminescence-based assays with standard water testing methods; and 1.2.4 To transfer technology and develop local skills in luminescence-based assays.

Among the above-mentioned, original objectives (1.2.1) was covered in the literature review (Chapter Two), while (1.2.2) - (1.2.4) were addressed in the final objectives below.

1.3 FINAL OBJECTIVES OF THE STUDY 1.3.1 To construct prokaryotic biosensor systems using the V. fischeri luxCDABE operon. 1.3.2 To construct Escherichia coli DH5α, Shigella sonnei, Shigella flexneri, and Enteropathogenic E. coli biosensor systems using lac, recA, fabA and uspA promoters fused to the V. fischeri luxCDABE operon; 1.3.3 To evaluate the biosensors’ sensitivity (minimum detection levels) in the presence of heavy metal and chemical pollutant laboratory standards. 1.3.4 To compare the sensitivity of all the biosensor systems to existing acute toxicity tests, TM e.g. the Daphnia LC50 toxicity test and the Vibrio fischeri-based BioTox kit. 1.3.5 To evaluate the sensitivity of the bacterial biosensors to heavy metals and chemical pollutants in wastewater effluent samples, from various eThekwini Wastewater Treatment Works.

The project commenced with a literature review which appears in Chapter Two. Thereafter, objectives 1.3.1 and 1.3.2 which dealt with the design and construction of prokaryotic biosensor systems was dealt with in Chapter Three. Objectives 1.3.3 and 1.3.4, dealing with the evaluation of the biosensors’ sensitivity and comparison with existing commercial tests, is reported in Chapter Four. Chapter Five evaluates the application of biosensors for the detection of heavy metals and chemical pollutants in wastewater effluents, as stated in objective 1.3.5. The report ends with a general discussion and conclusion in Chapter Six.

- 2 - 1.4 LAYOUT OF THE REPORT The report comprises 6 chapters and an Executive Summary.

Executive Summary: a short description of the methodology, the original and final objectives, as well as the reasons why the objectives had been amended and an indication as to whether or not the final objectives had been met. Also includes a summary of major findings and conclusions reached.

Chapter One: background, motivation, objectives and a brief layout of the report.

Chapter Two: literature review of the research study undertaken.

Chapter Three: construction of the seven prokaryotic biosensor systems using standard molecular techniques viz., plasmid DNA isolation, molecular cloning, PCR and bioluminescence detection.

Chapter Four: standardization of the toxicity test procedure using freeze-dried biosensor cells and various heavy metal and chemical standards.

Chapter Five: application of the seven prokaryotic biosensor systems for the evaluation of wastewater effluent samples using bioluminescent detection.

Chapter Six: synopsis of the major findings and conclusions reached.

- 3 - CHAPTER TWO: LITERATURE REVIEW

2.1 INTRODUCTION Pollution of the environment arises as a result of human activities, largely industrial, although agricultural, waste treatment and sewage disposal also contribute to the introduction of contaminants at problem levels (Sousa et al., 1998). The problems of use and management of these contaminated sites necessitate the development of reliable ecotoxicity tests (dose-response assays) to assess the significance of the contamination and the potential for bioremediation (Paton et al., 1997). In the last decade, the biological sciences have been used in exciting new directions, including biotechnology. The use of biosensor technology to monitor processes via biological outputs has become one of the newest dimensions of this revolution (Purohit, 2003). Biosensors, whether or not involving genetic modification, offer a very promising way in the provision of reliable ecotoxicity tests. Biosensors are biological materials which, when exposed to an analyte (e.g. air, water, soil), provide an information linked response via a suitable transducer. The biological material can comprise plants (cells/organs or whole plants), vertebrates, microorganisms/microbial tissue, enzymes, nucleic acid probes, antibodies and various other types of biological receptors. Compared to most biological materials, microorganisms are generally simple and cheaper to culture than higher organisms, can be freeze-dried for storage and respond rapidly to toxins. Furthermore, the selection of microorganisms for biosensors can be made to ensure environmental relevance. Genetic modification can greatly enhance detection of microbial response, primarily through use of reporter genes (e.g. lux genes). The reporter genes may either be fused to genes involved in the response to a particular toxin or used to indicate overall metabolic status. Biosensor detection of specific heavy metals has been achieved through transcriptional fusion of lux reporter genes to appropriate heavy metal resistance promoters (i.e., light output is switched on by the presence of particular heavy metals). Detection of specific xenobiotics (i.e., pollutants) through fusion of reporter genes such as lux, with catabolic genes encoding degradation of the organic compound in question, provides a potentially powerful role for genetically modified microbial biosensors. However, many recalcitrant organics are not toxic to microorganisms. Fusion of reporter genes to specific stress and catabolic responsive promoters (i.e., switching reporter genes on in the presence of particular organic pollutants) will therefore be the preferred option in biosensor development (Paton et al., 1997).

- 4 - Ultimately, environmental pollutants, be they readily degradable or recalcitrant, are expected to continue to be widespread in the ecosystems over the next 20 years, causing serious effects on human health and the respective ecosystems (Keane et al., 2002).

2.2 ENVIRONMENTAL POLLUTION Comprising over 70% of the Earth’s surface, water is undoubtedly the most precious natural resource that exists on our planet. Without this invaluable compound comprised of hydrogen and oxygen, life on Earth would be non-existent since it is essential for everything on our planet to grow and prosper. Although humans recognize this fact, they disregard it by polluting wetlands and groundwater. In addition to innocent organisms dying off, drinking water has become greatly affected, as is the ability to use water for recreational purposes. In order to combat water pollution, one must understand the problems and become part of the solution (Richman, 1997). There is a continuing need for monitoring the environment due to the serious human health implications of environmental pollutants (Table 2.1).

A wide variety of land use activities can lead to groundwater contamination, which must be managed and controlled. Contaminants are released through agriculture, landfills, chemical wastes, urban areas, rural septic tanks, deforestation and mining (Fig. 2.1). Most of these industrial effluents are sent to wastewater treatment works (Fig. 2.2).

The major sources of water pollution can be classified as municipal, industrial, and agricultural. Municipal water pollution consists of wastewater from homes and commercial establishments. Raw sewage includes waste from sinks, toilets, and industrial processes. Treatment of the sewage is required before it can be safely buried, used, or released back into local water systems. In a treatment plant (Fig. 2.3), the waste is passed through a series of screens, chambers, and chemical processes to reduce its bulk and toxicity. The three general phases of treatment are primary, secondary, and tertiary. During primary treatment, a large percentage of the suspended solids and inorganic material is removed from the sewage. The focus of secondary treatment is reducing organic material by accelerating natural biological processes. Tertiary treatment is necessary when the water will be reused. During this phase 99% of solids are removed and various chemical processes are used to ensure the water is as free from impurity as possible. Treated wastewater is released directly into receiving waters like rivers and the sea (Richman, 1997).

- 5 -

Fig. 2.1 Activities that result in groundwater contamination (Prescott et al., 1996).

Fig. 2.2 An aerial photograph of the Amanzimtoti Wastewater Treatment Works (eThekwini Wastewater, 2003).

- 6 - Table 2.1 Health implications of heavy metals and organic pollutants (Holmes, 1996)

Pollutants Health effects in humans *TWQ (mg/l) Copper (II) Gastrointestinal disturbances; possible 30 liver, kidney and red blood cell damage Zinc (II) Acute toxicity and gastrointestinal 50 irritation; nausea and vomiting Chromium (VI) Gastrointestinal cancer 1 Chromium (III) lower incidences of cancer than 1 chromium(VI) Lead(II) neurological damage to children and 10 foetuses Cadmium(II) kidney damage and possible fatality 0.02 Benzene central nervous system (CNS) 5 disorders; cancer and anaemia Toluene CNS damage; nausea; kidney damage; 1 impairment of hearing and vision Ethylbenzene CNS damage; liver and kidney 0.7 damage; and fatigue Xylene CNS damage; liver and kidney 10 damage; and disturbance of balance *EPUSA (2003) recommended that the target water quality (TWQ) not be exceeded, due to potentially deleterious health effects.

With increasing awareness of growing pollution problems, stringent conditions have been enforced in South Africa, Durban, by EThekwini Wastewater (2003), viz., no trade effluent shall be accepted for discharge into the sewage disposal system and sea outfalls unless it complies with the conditions outlined in Table 2.2.

- 7 -

Fig. 2.3 A schematic representation of a typical wastewater treatment works, showing the various treatment processes (Richman, 1997).

Table 2.2 General quality limits for acceptance of trade effluent for discharge into the sewage disposal system and sea outfalls (eThekwini Wastewater, 2003)

Possible pollutants Large works Small works Sea outfall (mg/l) (mg/l) (mg/l ) Oils, grease, waxes 50 50 50 of mineral origin Copper 50 5 3 Zinc 50 5 20 Lead 20 5 5 Cadmium 20 5 1.5 Total chromium 20 5 3

- 8 - 2.3 FUNCTIONS ENCODED BY THE lux GENES Almost all luminous bacteria have been classified into the genera Vibrio, Photobacterium and Xenorhabus, with most species being marine in nature. Only Xenorhabus species infect terrestrial organisms. The light-emitting bacteria that have been investigated are Vibrio fischeri, V. harveyi, P. phosphoreum, P. leiognathi, Photorhabus luminescens and X. luminescence. The lux genes from V. fischeri and V. harveyi have been used extensively for gene fusions. The lux gene cassette is composed of five genes, luxCDABE, comprising about 7 kb of DNA (Burlage and Kuo, 1994). These genes are recognized as a convenient reporter system. Lux-marked biosensors offer great environmental relevance for luminescence based testing in aquatic systems which utilizes the response of the luminescent bacteria. An important use of bioluminescence is as a sensor for the detection of chemicals in the environment. New technologies have created various public health and environmental problems. Important goals in addressing these problems are the detection and elimination of pollutants, including various hydrocarbons, aromatic substances, polymers and toxic metals. Detection previously relied on methods of analytical chemistry. The disadvantage is that these methods are unable to distinguish pollutants that are available in the environment from the inert, unavailable forms. It has been determined that effective biosensors for the determination of bioavailable metals and other pollutants should contain sensitive receptor components. Therefore, bioluminescence (lux) genes have been recognized as a convenient reporter system

2.3.1 Luciferase Enzymes responsible for light production are called luciferases. Bacterial luciferase is a heterodimeric enzyme of 77 kDa, composed of α and β subunits with molecular masses of 40 kDa and 37 kDa, respectively. The α and β subunits are encoded for by the luxA and luxB genes, rspectively. The light-emitting reaction in bacteria involves

the oxidation of reduced riboflavin phosphate (FMNH2) and a long-chain fatty aldehyde with the emission of blue-green light (Fig. 2.4, Equation 1).

The reaction is highly specific for FMNH2. Differences in aldehyde specificity do exist between different luciferases. Particularly noteworthy re the high luminescent responses of V. harveyi luciferase to nonanal and decanal at saturating substrate concentrations. Higher light intensity can be obtained with dodecanal for V. fischeri luciferase. This property may be very important in terms of expression of light emission in vivo in prokaryotic cells missing the aldehyde substrate, because decanal appears to cross the cell membrane much more readily than longer-chain aldehydes do. Luciferase is produced in very large amounts in the marine bacteria, and the luxA and

- 9 - luxB genes can be readily expressed in E. coli, providing abundant sources of protein for purification and/or application (Meighen, 1991).

luxAB

1. R-CHO + FMNH2 + O2 R-COOH + FMN + H2O + Light (λ490nm) luxCDE

2. R-CHO + NADP + AMP + PP R-COOH + NADPH2 + ATP

Fig. 2.4 The bioluminescent reactions encoded for by the luxCDABE operon. The luxAB genes convert an aldehyde substrate to a carboxyl group, generating visible light (Equation 1). The luxCDE

genes use NADPH2 and ATP to generate the aldehyde (Equation 2) (Burlage and Kuo, 1994).

2.3.2 Aldehyde biosynthesis During the bioluminescent reaction, a multienzyme fatty acid reductase complex catalyses the synthesis of aldehydes. This complex comprises three proteins: a reductase (54 kDa), a transferase (42 kDa) and a synthase (33 kDa), which are encoded for by luxC, luxD and luxD, respectively. The transferase subunit catalyses the transfer of activated fatty acyl groups to water as well as oxygen and thiol receptors. The primary reaction catalysed by the fatty acid reductase complex is the reduction of fatty acids to aldehydes (Fig. 2.4, Equation 2). The reaction is catalysed by the reductase and the synthase components. The synthase activates the fatty acid, resulting in the formation of a fatty acyl-AMP intermediate that is tightly bound to the enzyme. In the presence of reductase, the acyl group is transferred first to the reductase before being

reduced by NADPH2 to aldehyde (R-CHO) (Meighen, 1991).

2.4 APPLICATIONS OF lux GENES Fish containing luminescent bacteria can be used for toxicity testing (Fig. 2.5).

Currently most South African Water Boards use the Daphnia LC50 toxicity test to analyse water samples (Fig. 2.6). However, these tests are expensive and time consuming. Thus, whole-cell microbial biosensors offer a powerful new approach to environmental monitoring. These biosensors provide an indication of pollutant bioavailability, rather than total concentrations obtained by traditional analytical techniques. Biosensors are also able to monitor very low levels of pollutants, can work in complex matrices and have very fast response times. Lux-marking of bacteria offers scope for the construction of environmentally relevant, whole-cell biosensors which can either be toxin specific or general indicators of pollutant toxicity. Biosensors have been used to monitor environmental contamination by heavy metals, industrial effluents,

- 10 - benzene, toluene, ethylbenzene, xylene (BTEX) compounds and chlorinated aromatics. Lux-marked biosensors offer greater environmental relevance for luminescence-based testing in various aquatic and terrestrial contaminated sites compared to other tests, e.g. MicrotoxTM, which utilises the response of the naturally luminescent marine bacterium V. fischeri (Sinclair et al., 1999).

Another significant advantage of microbial biosensors is that they are self-replicating, allowing for economical production of these sophisticated and complex biosensor systems. Moreover, cell based systems offer the possibility of identifying specific pollutants present in complex mixtures without pretreatment of environmental samples. Since bacterial bioluminescence is directly linked to cellular respiration, any inhibition of cellular metabolism due to toxicity results in a decrease in the light emission of affected biosensor cells (Keane et al., 2002). Conversely, in the case of catabolic and stress promoters fused to lux genes, there is an induction of light emission of affected biosensor cells in the presence of specific contaminants.

2.4.1 Fusions of lux genes to specific stress promoters Recent years have seen significant advances in the development of rapid and convenient micro-scale bioassays, aimed at the detection of either global parameters such as toxicity or genotoxicity of specific organic and inorganic contaminants. The amenability of bacteria to genetic manipulation is an additional advantage, which is being utilised in the development of novel bioassays. Thus, bacterial strains have been constructed that emit a readily detectable signal in the presence of pre-determined toxicants or groups of toxicants. Several of the more promising microbial strains are based upon the fusion of bacterial bioluminescence (lux) genes as a reporter to selected bacterial gene promoters (Bechor et al., 2002). Genes of the luxCDABE operon of V. fischeri have been fused to various promoter genes under the control of several global regulatory circuits, including rpoH, soxRS, oxyR, fadR, crp, uspA and recA (Ben-Israel et al., 1998).

2.4.1.1 fabA:lux-based biosensor The plasmid pfabALux6, containing an operon fusion of the fabA promoter, was constructed and transformed into the E. coli strains DPD2544 and DPD2543. The fabA gene codes for β-hydroxydecanoyl-ACP dehydrase, a key enzyme in the synthesis of unsaturated fatty acids, and is induced when fatty acid biosynthesis pathways are disrupted. The fabA gene controlled a dose-dependent and highly sensitive bioluminescent response to a variety of chemicals (Table 2.3).

- 11 -

Fig. 2.5 An Atlantic flashlight fish which contains luminous bacteria in the light organ under its eye (Prescott et al., 1996).

Fig. 2.6 Daphnia pulex (water flea) is a tiny shellfish with a transparent body covering. It is

used in the Daphnia LC50 toxicity test at Umgeni Water as a standard bioassay (World Book, 1992).

- 12 - The toxicity for the fabA: lux DPD2544 and DPD2543 biosensors were reported as an

EC200 value (i.e., the sample concentration causing a two-fold increase in luminescence compared with the control). Data concerning MicrotoxTM toxicity in Table 2.3 was derived from Kaiser and Palabrica (1991). The V. fischeri-based MicrotoxTM toxicity

test measured a decrease in luminescence, and was reported as EC50 values (i.e., the sample concentration causing a 50 percent decrease in luminescence).

Table 2.3 Toxicity of various chemicals to E. coli strains DPD2544 and DPD2543, as a b determined using EC50 and EC200 values ( Kaiser and Palabrica, 1999; Bechor et al., 2002)

Aromatics and Microtox EC50 DPD2544 EC200 DPD2543 EC200 detergents (mg/l)a (mg/l)b (mg/l)b Benzene 2 - 236 29 - Acetone 13 300 - 29 100 2220 1490 pXylene 5.7 299 208 Sodium 0.4 - 3 12 0.12 dodecyl-sulphate

2.4.1.2 recA:lux-based biosensor Bacterial repair of DNA damage is mediated partly by the recA-dependent, lexA- controlled SOS response. Upon SOS induction, the recA gene product is converted into an active protease. The activated RecA protein cleaves the LexA repressor protein, resulting in the transcriptional depression of several genes, among them recA. Activation of such repair systems is a measure of the mutagenic and genotoxic effects of various chemical and physical treatments. The recA promoter was fused to the promoterless V. fischeri luxCDABE operon present within the broad-host-range, multi-copy plasmid pUCD615. This fusion in E. coli allowed visualisation of the transcriptional responses induced by DNA damage, without the need to perform enzyme assays. DNA-damaging agents included mitomycin and UV irradiation. Bioluminescence was measured in real time over extended periods. This biosensor reported the presence of genotoxic doses of toxicants by an increase in the production of light (Vollmer et al., 1997).

2.4.1.3 katG: lux-based biosensor Active oxygen species are highly damaging to all living organisms, and aerobically grown cells cannot survive without adequate protection against the toxic effects of these molecules. The importance of oxidative stress and damage in biology, agriculture, and medicine is gaining more recognition. Of the oxygen species the free

- 13 - radicals are usually considered the most hazardous. Hydrogen peroxide, although chemically less reactive, is nevertheless a threat to the structure and function of proteins, nucleic acids, lipids, and membranes, whether it is added externally or produced intracellularly. The damage inflicted by hydrogen peroxide, as well as many organic peroxides, is due to intrinsic oxidative activity or to the production of hydroxyl radicals. Several enzymatic systems have evolved to eliminate peroxides and combat their deleterious effects. In both E. coli and Salmonella typhimurium, the oxyR regulon is the best studied of these defense circuits. Exposure to hydrogen peroxide leads to the induction of at least 30 proteins, 9 of which are under direct positive control of the OxyR protein. These include a catalase (HPI) that is encoded by the katG gene. A plasmid containing a transcriptional fusion of the E. coli katG promoter to the truncated V. fischeri luxCDABE operon was constructed. An E. coli strain bearing this plasmid (strain DPD2511) exhibited low basal levels of luminescence, which increased up to 1,000-fold in the presence of hydrogen peroxide, organic peroxides, redox- cycling agents (methyl viologen and menadione), a hydrogen peroxide-producing enzyme system (xanthine and xanthine oxidase), and cigarette smoke. An oxyR deletion abolished hydrogen peroxide-dependent induction, confirming that oxyR controlled katG-lux luminescence. Light emission was also induced by ethanol by an unexplained mechanism. A marked synergistic response was observed when cells were exposed to both ethanol and hydrogen peroxide. The level of luminescence measured in the presence of both inducers was much higher than the sum of the level of luminescence with ethanol and the level of luminescence observed with hydrogen peroxide. It is suggested that this construction or similar constructions may be used as a tool for assaying oxidant and antioxidant properties of chemicals, as a biosensor for environmental monitoring, and as a tool for studying cellular responses to oxidative hazards (Belkin et al., 1996).

2.4.1.4 uspA: lux and grpE: lux-based biosensors The E. coli uspA gene encodes the universal stress protein A (UspA). Conditions that limit cell growth, including nutrient starvation and exposure to toxic agents, induce uspA transcription. A large variety of environmental challenges trigger the heat shock response. Accordingly, the E. coli heat shock promoter grpE, fused to the lux reporter, experiences an increase in bioluminescence in response to many chemicals. A transcriptional fusion of the E. coli uspA to luxCDABE was characterised and compared with the heat shock-responsive grpE: lux fusion (Van Dyk et al., 1995). Similarities in range and rank order of inducing conditions were observed (Table 2.4).

- 14 - Table 2.4 Comparison of the E. coli uspA: lux and grpE: lux-based biosensors (Van Dyk et al., 1995)

Inducer E. coli uspA:lux biosensor E. coli grpE:lux biosensor Concentration for maximum Concentration for maximum induction (µg/ml) induction (µg/ml) Copper sulphate 800 800 Pentachlorophenol 37.5 37.5 Hydrogen peroxide 50 50 2,4-Dinitrophenol 125 63 Cadmium chloride 12 23 Mitomycin C 1.25 1.25

2.4.2 ipb:lux-based biosensor The isopropylbenzene (cumene) catabolism operon (ipb), located on plasmid pRE4 in Pseudomonas putida RE204, specifies the catabolism of alkylbenzenes, such as toluene, ethylbenzene, isopropylbenzene and n-butylbenzene. The ipb operon and the luxCDABE operon from V. fischeri were used to create an ipbRo/Pa-luxCDABE reporter fusion plasmid, pOS25. E. coli HMS174 (pOS25) produced light in the presence of inducers of the ipb operon. These inducers were shown to be hydrophobic compounds and to include monoalkylbenzenes, substituted benzenes and toluenes, some alkanes and cycloalkanes, chlorinated solvents and naphthalenes. Complex hydrocarbon mixtures such as gasoline, diesel fuel, jet fuels and creosote were also inducers of ipb: lux. Bacteria carrying the ipb: lux reporter may be useful as bioindicators of hydrocarbon pollution in the environment and may be particularly valuable for examining the bioavailability of inducing pollutants (Selifonova and Eaton, 1996).

2.4.3 pUCD607 based biosensor systems The multi-copy pUCD607 plasmid, constructed by Shaw and Kado (1986), contains the V. fischeri luxCDABE operon (Fig. 2.7).

- 15 -

Sal I luxC

AmpR luxD

Bgl II

pUCD607 luxA 20.30 kb

luxB

KmR luxE

Sal I

Fig. 2.7 The pUCD607 plasmid, which contains the V. fischeri luxCDABE operon (Shaw and Kado, 1986).

2.4.3.1 BTEX contaminated sites In response to the multitude of pollution that constitutes imminent health hazards; clean-up processes based on a wide variety of technologies have been developed. Physical and chemical processes are frequently employed to treat contaminated sites but often do not destroy contaminants. Bioremediation is being employed increasingly as a cost-effective and efficient alternative to traditional treatment technologies. Bioremediation is typically practiced by optimising environmental conditions and to stimulate the degradation of pollutants by naturally occurring organisms. However, bioremediation is often limited in the materials that it can treat and by conditions at the treatment site. The absence of catabolic pathways for a xenobiotic appears to be one of the obstacles to its biodegradative cleanup. However, biochemical pathways constantly evolve and microorganisms have great capacity to adapt to adverse conditions and develop the ability to degrade new molecules with time.

- 16 -

Fig. 2.9 Bioluminescence of P. fluorescens /pUCD607 in response to increasing concentrations of toxin, e.g. copper sulphate (Belkin et al., 1996).

Rates of pollutant biodegradation are often limited by environmental constraints such as restrictive pH and the concentration of toxic substances and inhibitors so high at the site

- 17 - that microbial proliferation and metabolism is prevented. Therefore, success will only be attained if these constraints are identified and means are devised to alleviate them to such an extent where bioremediation can effectively proceed (Sousa et al., 1998). While microorganisms capable of biodegradation of BTEX compounds are usually present at these sites, there is a need to know whether or not conditions are favourable for biodegradation to occur. A recent approach to determine whether compounds are available and what conditions are favourable for degradation is the use of whole-cell bioluminescent reporters. A P. fluorescens 10586s pUCD607 biosensor was used to screen sediment and groundwater samples obtained from a BTEX contaminated site in order to identify constraints to site remediation (Fig. 2.8 and Fig. 2.9). Conventional chemical analysis (gas chromatography, inductively coupled plasma mass spectroscopy) was used to confirm the reliable performance of the biosensor and to identify its potential contribution to site management to ensure effective remediation. Untreated samples caused reductions in percentage bioluminescence from 10 - 95%. Water sample W1, containing a total BTEX concentration of 30.595 µg/l, caused the highest decrease in bioluminescence (5.78%). This toxicity was significantly reduced after elimination of the volatile organic compounds (42.01%) and bioluminescence was further increased to 87.8% after removal of total organic matter. This suggested that volatile organic compounds were not the only toxic organopollutants present. The water samples, which gave high values of bioluminescence, contained BTEX concentrations of 17 - 31 µg/l (Sousa et al., 1998). A tod-luxCDABE fusion was constructed and introduced into the chromosome of P. putida F1, yielding the strain TVA8. This strain was used to examine the induction of the tod operon when exposed to BTEX compounds and aqueous solutions of JP-4 jet fuel constituents. Since this system contained the complete lux cassette (luxCDABE), bacterial bioluminescence in response to putative chemical inducers of the tod operon was measured on-line in whole cells without added aldehyde substrate. There was an increasing response to toluene concentrations from 30 - 50 µg/l, which began to saturate at higher concentrations. The detection limit was 30 µg/l. There was a significant light response to benzene, m- and p-xylenes, phenol, and water-soluble JP-4 jet fuel components, but there was no bioluminescence response upon exposure to o- xylene. The transposon insertion was stable and had no negative effect on cell growth (Applegate et al., 1998).

2.4.3.2 Mini-Tn5 luxCDABE transposon A mini-Tn5 transposon was modified to introduce the luxCDABE operon from pUCD607, to form the mini-Tn5 luxCDABE transposon. The mini-Tn5 luxCDABE transposon was introduced into P . fluorescens 8866 and P. putida F1, to develop

- 18 - bioluminescence-based biosensors for environmental toxicity testing. Both P. fluorescens 8866 Tn5 luxCDABE and P. putida F1 Tn5 luxCDABE were used to assess the toxicity of standard solutions as well as copper and 3, 5- dichlorophenol spiked groundwater samples (Table 2.5). The standard solutions included copper, zinc and 3, 5-DCP. They were successfully used for bioluminescence-based bioassays and the potential value of using different bacterial biosensors for ecotoxicity testing was shown (Weitz et al., 2001).

Table 2.5 EC50 values of the two Pseudomonas Tn5 luxCDABE biosensors (Weitz et al., 2001) Pollutant P. fluorescens 8866 Tn5 P. putida F1 Tn5 luxCDABE luxCDABE biosensor biosensor EC50 (mg/l) EC50 (mg/l) Copper 0.3 0.17 Zinc 0.1 0.04 3,5-DCP 4.82 5.55 Groundwater + copper 0.14 0.11 Groundwater + 3,5-DCP 6.74 7.96

2.4.3.3 Toxicity of chlorophenols E. coli HB101 and P. fluorescens were lux-marked with pUCD607. These two lux biosensors and the MicrotoxTM kit were exposed to 2,4 dichlorophenol (2,4-DCP). Increasing 2,4-DCP concentrations caused a decrease in light output in all three biosensors with an order of sensitivity (in terms of luminescence decrease over the 2,4- DCP concentration range) of P. fluorescens < E. coli HB101 < MicrotoxTM (Sinclair et al., 1999).

2.4.3.4 Copper availability in malt whisky distillery effluent The whisky industry of Scotland plays an important role in sustaining the highland economy. There is concern, however, that if the waste products of the distillation process are not adequately treated, there may be breaches of environmental legislation and the potential contamination of drinking water. The whisky distillation process uses copper stills and it is the treatment of copper contaminated post-distillation residue that may be of concern. Regulatory bodies ensure that the whisky producers follow strict discharge guidelines. However, it is costly for small distilleries to regularly send discharge samples for analysis. The development of a rapid and inexpensive bioassay

- 19 - may be a more cost-effective option for small companies to consider in preference to costly comprehensive analysis. Samples were taken from upstream, influent, effluent and downstream locations of a whisky distillery in North East Scotland. The concentration of inorganic pollutants in Table 2.6 was determined using inductively coupled plasma mass spectrometry (IC- MPS). The principal contaminant was found to be copper, and bioluminescence-based assays were carried out to assess the bioavailability of copper. One assay involved the use of the naturally luminescent marine bacterium Photobacterium phosphoreum and the other involved the use of the P. fluorescens 10586s pUCD607 biosensor. Use of the luminescence-marked biosensors was found to be most sensitive and reproducible (Paton et al., 1995).

Table 2.6 Copper concentrations and percentage bioluminescence of contaminated sites (Paton et al., 1995) Sample ICP-MS P. phosphoreum P. fluorescens pUCD607 (µg/l copper) Bioluminescence (%) Bioluminescence (%) Upstream 70 100.0 100 Influent 1 830 9.0 8.3 Effluent 640 92.5 65.1 Downstream 90 95.6 82.4

2.4.3.5 Assessment of the bioavailability of heavy metals The bioluminescence of genetically modified (lux-marked) bacteria to potentially toxic chemicals (PTEs) was monitored. The P. fluorescens 10586s/FAC510, which has the V. fischeri luxCDABE operon integrated into the chromosome, and P. fluorescens 10586s/pUCD607 were used. Bioluminescence, involving either plasmid or chromosomally encoded lux genes, decreased as the metal concentration increased. The pUCD607 marked construct was significantly more sensitive to all metals except chromium (Table 2.7). The order of metal sensitivity was found to be copper = zinc > cadmium > chromium > nickel for the chromosomally marked construct and copper = zinc > cadmium > nickel > chromium for the plasmid marked construct. The very sensitive response of lux-marked bacteria to PTEs identified the potential for a rapid and flexible assay for assessing the pollution of soil or fresh water environments (Paton et al., 1995).

- 20 - Table 2.7 Comparison of mean EC50 values of P. fluorescens strains 10586s/FAC510 and 10586s/pUCD607, in the presence of selected potentially toxic chemicals (Paton et al., 1995) P. fluorescens Potentially toxic chemicals strains EC50 (mg/l) zinc copper cadmium nickel chromium 10586s/FAC510 0.89 0.76 0.98 2.49 1.28 1058/pUCD607 0.09 0.09 0.17 0.28 1.46

2.4.3.6 Toxicity of herbicides in freshwater E. coli HB101, P. fluoresecns 10586r and P. putida F1 were all marked with the luxCDABE operon, from plasmid pUCD607. Rhizotox-C, a Rhizobium-based biosensor (Chaudri et al., 2000), was chromosomally marked with luxAB, and required the addition of n-decyl aldehyde to enable luminescence expression. These three biosensor systems were tested in the presence of the herbicides: atrazine, simazine, propazine, mecoprop, MCPA, diuron and paraquat. In general, E. coli HB101, marked with the luxCDABE operon was the most sensitive to all compounds tested, except mecoprop. Rhizotox-C was the least sensitive for all herbicides investigated (Strachan et al., 2001).

2.4.4 Versatile biosensors for the detection of mercury and arsenic The development of metal-specific biosensor tools functioning on the basis of bioluminescent reporter systems, has been receiving increasing attention. These biosensors are considered as key assets both for characterising the extent of contaminated areas and for following up the success or failure of bioremediation operations of large areas contaminated with heavy metals. They are of particular relevance for the assessment of remediation strategies based on in situ immobilisation of heavy metals.

2.4.4.1 mer:lux-based biosensors Biosensors for the detection of pollutants in the environment can complement analytical methods by distinguishing bioavailable contaminants from inert (unavailable forms) contaminants. Many studies have used whole-cell biosensors to detect and measure the presence of metals in complex environments, many of which focused on the detection of bioavailable mercury under both laboratory and environmental conditions. By using fusions of the well-understood Tn21 mercury resistance operon (mer) with promoterless luxCDABE from V. fischeri, three biosensors for Hg(II) have been constructed and tested. Bioluminescence specified by pRB28 (carrying merRo/pT'), by

- 21 - pOS14 [mediating active transport of Hg(II)], and pOS15 (containing the intact mer operon), were measured in rich and minimal media. The highest sensitivities were achieved in minimal medium and were 1, 0.5, and 25 nM Hg(II) for pRB28, pOS14, and pOS15, respectively. The utility of the biosensors in natural waters was demonstrated with freshwater, rain, and estuarine samples supplemented with Hg(II). mer:lux carried by pBR28 and pOS14 responded to Hg(II) in mercury contaminated water samples collected from a fresh water pond. Semiquantitative analyses based on light emission in samples collected from inlet (analytically determined total mercury, ~20nM) and outlet (total mercury, ~7nM) of the pond showed bioavailable mercury at approximately 20 and 1 - 2 nM, respectively. Thus, the biosensors described here semiquantitatively detect bioavailable inorganic mercury (at a nanomolar to micromolar concentration range) in contaminated water (Selifononva et al., 1993). Mercury, in the form of methyl mercury, is an environmental pollutant of great risk to human health. Measuring bioavailable mercury is essential for calculating methylation rates of mercury, and thereby predicting the bioaccumulation of methyl mercury in different environments. A whole-cell biosensor construct was made by fusing the

mercury inducible promoter, Pmer, and its regulatory gene, merR, from transposon Tn21 with the reporter genes from the luxCDABE. In E. coli this mer:lux biosensor construct responded to low levels of mercury by producing light. Since the response was quantitative, it was used to quantify bioavailable mercury in different environments. The mer: lux biosensor construct was cloned into a mini-Tn5 delivery vector, thus enabling its transfer into P. putida KT2440. This new biosensor was used to quantify water-extractable mercury in contaminated soil. It was found that 50 ng of mercury was water-extractable from water spiked with 2.5 µg/g soil (Hansen and Sorenson, 2000).

2.4.4.2 ars:luxAB-based biosensor The increase in arsenic concentration is partly due to the extensive use of a chromated copper arsenate (CCA) wood preservative. Wood intended for marine use receives 24 kg to 40 kg CCA per cubic metre of wood to prevent its destruction by bacteria, fungi and insects. Each of the three chemicals in CCA is known to be toxic to aquatic biota and found to be leached from the treated wood in both freshwater and seawater. Pathologic and genotoxic effects have been observed in oysters living on CCA-treated wood. In addition, CCA was shown to affect the growth of polychlorinated phenol- degrading bacterial species and their ability to degrade polychlorinated phenols (Wall and Stratton, 1994). Thus, monitoring of bioavailable amounts of CCA released by treated wood is important in order to detect and rectify its toxic effects. Among the three chemical constituents of CCA, arsenic is the most abundant in the environment, and known to have carcinogenic and teratogenic effects on humans upon chronic

- 22 - exposure. Therefore, the focus of Cai and DuBow’s (1997) study was on the biomonitoring of arsenic toxicity. The arsB: luxAB fusion, which was constructed using the arsenic oxyanion-inducible E. coli chromosomal operon (arsRBC), was induced in a concentration-dependent manner by arsenic salts. The bacterial biosensor, E. coli LF20012 (arsRBC), was exposed to increasing concentrations of CCA, as well as sodium arsenate and chromated copper solution (CC). Analysis of the luciferase activity revealed that the arsB: luxAB fusion was expressed in response to CCA and sodium arsenate, but not to CC. The detection limit of arsenic was found to be 0.01 µg arsenic/ml. A greater induction of luminescence by arsenate was observed when cells were limited for phosphate, as phosphate can act as a competitive inhibitor of arsenate ions. Ultimately, the E. coli LF20012 (arsRBC) biosensor has a promising future as a sensitive biosensor for monitoring bioavailable levels and toxicity of arsenic near sites where CCA-treated wood has been used (Cai and DuBow, 1997).

2.4.5 tfdRPDII ¯ luxCDABE-based biosensor The herbicide 2, 4-dichlorophenoxyacetic acid (2, 4-D) is widely used in both agricultural and domestic weed control applications. While it is rapidly degraded in most environments, the initial step in the degradation of 2, 4-D yields 2, 4-DCP (mentioned earlier in Section 1.3.3.3). At concentrations ranging from 120 - 250 µM, 2, 4-DCP is known to be toxic to 2, 4-D degraders and other microorganisms. This gives rise to the concern over the fate of 2, 4-D in the environment. As both 2, 4-D and 2, 4-DCP are non-polar molecules, they have the tendency to partition into organic matter. This reduction in bioavailability is difficult to assess with traditional analytical approaches but is an important factor affecting the longevity of these compounds in the environment. Hay et al. (2000) reported on the development of a bioluminescent reporter for the detection of 2, 4-DCP degradation in aqueous samples and its use in slurries containing

aged 2, 4-D residues. A bioreporter was made containing a tfdRPDII ¯ luxCDABE fusion in a modified mini-Tn5 construct. When it was introduced into the chromosome of Ralstonia eutropha JMP134, the resulting strain, JMP134-32, produced a sensitive bioluminescent response to 2, 4-D at concentrations of 2.0 - 5.0 µM. A sensitive response was also recorded in the presence of 2, 4-DCP at concentrations below 1.1 x 102 mM. A significant bioluminescent response was also recorded when strain JMP134-32 was incubated with soils containing aged 2, 4-D residues (Hay et al., 2000).

2.4.6 lux:nah fusion A bioluminescent reporter plasmid for naphthalene catabolism (pUTK21) was developed by tranposon (Tn4431) insertion of the lux gene cassette from V. fischeri into a naphthalene catabolic plasmid in Pseudomonas fluorescens. The insertion site of

- 23 - the lux transposon was the nahG gene encoding for the salicylate hydroxylase. Luciferase-mediated light production from P. fluorescens strains harbouring this plasmid was induced on exposure to naphthalene or the regulatory inducer metabolite, salicylate. In continuous culture, light induction was rapid (15 min) and was highly responsive to dynamic changes in naphthalene exposure. Strains harbouring pUTK21 were responsive to aromatic hydrocarbon contamination in Manufactured Gas Plant soils and produced sufficient light to serve as biosensors of naphthalene exposure and reporters of naphthalene biodegradative activity (King et al., 1990). An optical whole-cell biosensor based on a genetically engineered bioluminescent catabolic reporter bacterium was developed for continuous on-line monitoring of naphthalene and salicylate bioavailability and microbial catabolic activity in waste streams. The bioluminescent bacterium, P. fluorescens HK44, carries a transcriptional nahG-luxCDABE fusion for naphthalene and salicylate catabolism. Exposure to either compound resulted in inducible bioluminescence. The reporter culture was immobilised onto the surface of an optical light guide by using strontium alginate. This biosensor probe was then inserted into a measurement cell, which simultaneously received the waste stream solution and a maintenance medium. Exposure under defined conditions to both naphthalene and salicylate resulted in a rapid increase in bioluminescence. The magnitude of the response and the response was observed during repetitive perturbations with either salicylate or naphthalene. Exposure to other compounds, such as glucose and complex nutrient medium or toluene, resulted in either minor bioluminescence increases after significantly longer response times compared with naphthalene or no response, respectively. The environmental utility of the biosensor was tested, by using real pollutant mixtures. A specific bioluminescence response was obtained after exposure to either an aqueous solution saturated with JP-4 jet fuel or an aqueous leachate from manufactured-gas plant soil, since naphthalene was present in both pollutant mixtures (Heitzer et al., 1994).

2.4.7 Bioavailability of middle-chain alkanes in groundwater Although many contaminants are readily biodegradable, they often persist in the environment because they are degraded at rates too slow for efficient cleanup. One major factor that limits biodegradation in the environment is the insufficient accessibility of pollutants to microbial attack. This is especially true of hydrophobic compounds, such as those occurring in diesel oil contaminants. Diesel oil consists mainly of linear and branched alkanes with different chain lengths and contains a variety of aromatic compounds. A microbial whole-cell biosensor was developed, and its potential to measure water- dissolved concentration of middle-chain-length alkanes and some related compounds

by bioluminescence was characterised. This biosensor strain E. coli DH5α (pGEC74,

- 24 - pJAMA7) carried the regulatory gene alkS from Pseudomonas oleovorans on plasmid

pGEC74, and the luxAB from V. harveyi on pJAMA7. In standardised assays the biosensor cells were readily inducible with octane, a typical inducer of the alk system. The biosensor responded to middle-chain-length alkanes but not to acyclic or aromatic compounds. In order to test its applicability for analysing environmentally relevant samples, the biosensor was used to detect the bioavailable concentration of alkanes in heating oil-contaminated groundwater samples. The biosensor tended to underestimate the alkane concentration in the groundwater samples by about 25%, possibly because of the presence of unknown inhibitors. This was corrected by spiking the samples with an octane standard (Table 2.8). Biosensor measurements of alkane concentrations were further verified by comparing them with the results of chemical analyses (Sticher et al., 1997).

Table 2.8 Inhibition of octane-induced luminescence in E. coli DH5α (pGEC74, JAMA7) between different groups of related compounds (Sticher et al., 1997)

Substance class Compounds (5 µM) Bioluminescence (%) inhibition Linear alkanes Pentane, hexane, heptane, decane, 6 dodecane Alkane mixtures Petroleum ether (low and high 11 boiling) Branched alkanes Heptamethylnonane, 3- 0 methylheptane, pristane Aromatic hydrocarbons Benzene, toluene, m-xylene 12 Polycyclic aromatic 1-Methylnaphthalene 0 hydrocarbons Alkylbenzenes Hexylbenzene 14

The luxCDABE has proved to be a convenient reporter system. It allows the rapid and cost efficient way of detecting pollutants in an environment. This chapter highlights these areas of use, together with the introduction and literature review building up towards this research area. The literature review provided the relevant information to conduct a comparative study of existing luminescence-based assays in order to evaluate their cost effectiveness and relevance and appropriateness for the determination of pollutants in water sources.

- 25 -

CHAPTER THREE: DESIGN AND CONSTRUCTION OF PROKARYOTIC BIOSENSOR SYSTEMS

3.1 INTRODUCTION Awareness of the magnitude of existing environmental pollution problems has been steadily increasing in recent years. With this awareness grows the need for sensitive and effective means for efficient environmental monitoring to assess the hazards involved in present contamination levels, as well as to serve as a warning system against future pollution. In view of the need for real-time toxicity monitoring, recent years have seen significant advances in the use of bacteria as test organisms (Belkin et al., 1997). Microorganisms can be described as specific and sensitive devices for sensing the bioavailability of a particular pollutant or pollutant class. This is based on the ability of pollutants to invoke nonspecific (e.g. toxicity or stress) or specific (e.g. activation of a degradative pathway) responses in microorganisms. The signaling pathway thus activated will regulate the expression of one or more sets of genes. The extent of this gene expression serves as a measure of the available (“sensed”) concentration of the compound (Sticher et al., 1997). Furthermore, as is becoming increasingly obvious, bacteria are endowed with an additional characteristic, which further augments their attractiveness: they are readily amenable to genetic manipulation. Thus, by relatively simple molecular biology techniques, bacterial strains can be “tailored” to emit a detectable signal upon a specified change in environmental conditions. Therefore, there has been a rapid development of biosensors based on genetically engineered bacteria. Such microorganisms typically combine a promoter-operator, which acts as the sensing element, with reporter gene(s) coding for easily detectable proteins. These biosensors have the ability to detect global parameters such as stress conditions, toxicity or DNA- damaging agents such as organic and inorganic compounds (Kohler et al., 2000). Thus, a rapid and sensitive way to measure such gene expression is by using whole-cell biosensor systems. Whole-cell biosensors are constructed by fusing the relevant promoter sequences and promoterless reporter genes such as those for bacterial luciferases. This genetic fusion, located on a plasmid vector, is then inserted into a bacterial strain, where it may be integrated into the chromosome or remain in plasmid form. The engineered genetic fusion is then replicated along with the bacterial cell’s normal DNA (Steinberg et al., 1995). Since light emission is easy to monitor and quantify (Stewart and Williams, 1992), bioluminescence genes are excellent reporter elements. This approach has recently been utilised in several cases to construct microbial tools capable of sensitively

- 26 - reporting on the presence of specific compounds (Belkin et al., 1997). Researchers who have used bacterial luciferase as a reporter gene have used either the complete luxCDABE operon or the luxAB genes. However, since the multi-enzyme fatty acid reductase responsible for this process is not present within the shortened gene construct, an aldehyde (usually n-decanal) must be added exogenously in order for the bioluminescence reaction to proceed (Keane et al., 2002). A wide range of promoters were used to create bacterial constructs that emit light in response to specific or general stress factors. Some biosensor systems were constructed to detect toxic compounds by coupling the lux genes with stress-inducible promoters like recA (DNA-damage sensitive), fabA (cell membrane-damage sensitive), uspA (general stress-sensitive) and lac (general stress-sensitive) (Ben-Israel et al., 1998 and Belkin et al., 1996). Most biosensor constructs use the host bacterium E. coli. However, Paton et al. (1997) found that different bacterial strains had different sensitivities to metals. It is therefore likely that some bacterial strains are better biosensor hosts than others (Hansen and Sorenson, 2000). For a more practical application of bacterial biosensors, the biosensing cells must be maintained during storage and transported to field sites without loss of luciferase activity. Therefore, freeze-drying would be the most suitable method since it allows for ease of storage and transport (Choi and Gu, 2002). The addition of solutes, like trehalose and skim milk, are known to increase the number of viable cells in a freeze- dried sample. However, viability remains below that of the initial culture used. Damage to biological systems during freeze-drying can be attributed to changes in the physical state of membrane lipids and changes in the structure of sensitive proteins. Removal of hydrogen-bonded water from the headgroup region of phospholipid bilayers, increases the headgroup packing and forces the acyl chains together, increasing the probability of Van der Waals interactions. As a result, the lipid may undergo a transition from liquid crystalline to gel phase. Upon rehydration, dry membranes, which are in gel phase at room temperature, undergo a transition from gel to liquid crystal phase. As the membranes pass through this phase transition, there are regions with packing defects, making the membranes leaky. Adding a disaccharide such as trehalose before drying lowers the transition temperature of the dry membranes by replacing the water between the lipid headgroups, preventing the phase transition and its accompanying leakage upon rehydration. In addition, trehalose is able to preserve both structure and function of isolated proteins during drying. This ability to stabilise proteins during drying results from the disaccharides forming hydrogen bonds with the proteins when water is removed, thus preventing protein denaturation. Based on this, it seems that trehalose could be used to preserve intact cells during freeze- drying (Leslie et al., 1995).

- 27 - This chapter describes the construction of luxCDABE-containing E. coli DH5α, Enteropathogenic E. coli, S. flexneri and S. sonnei biosensor systems.

3.2 MATERIALS AND METHODS 3.2.1 Growth and maintenance of bacterial cultures Bacterial strains used in this study are listed in Table 3.1. E. coli DH5α, Enteropathogenic E. coli, S. flexneri and S. sonnei cells were grown in Luria-Bertani (LB) broth (10 g tryptone, 5 g yeast extract and 10 g NaCl per litre), for 16 h. Working stocks of these bacterial strains were sub-cultured on agar plates at 4-week intervals. When required, antibiotics or other supplements were added at the appropriate final concentrations (Table 3.2). During transformation, the chromogenic substrate X-gal (5- bromo-4-chloro-3-indolyl-β-D-galactopyranoside) and the inducer IPTG (isopropyl-β- D-thiogalactopyranoside) were used for blue-white colony selection. All bacterial cultures were successfully preserved at -70°C using Microbank PL60 vials (Davies Diagnostics), according to the manufacturer’s instructions. These bacterial cultures were also preserved by stab inoculations of LB agar deeps in air-tight Bijou bottles and stored at room temperature.

3.2.2 Plasmid DNA isolation Transformants containing recombinant plasmids were selected by picking single colonies from LB plates containing ampicillin. Plasmid DNA from these colonies was isolated using a modified alkaline lysis method of Birnboim and Doly (1979). A 1.5 ml aliquot of an overnight bacterial culture was pelleted in an Eppendorf tube and centrifuged at 13 000 rpm for 3 min. The supernatant was aspirated and the pellet re- suspended in 100 Pl Solution A (25 mM Tris.HCl; 50 mM glucose; 10 mM EDTA, pH 8.0) containing 100 Pg/ml RNaseA. The cell suspension was incubated for 10 min at room temperature. Two hundred microlitres of freshly prepared Solution B (0.2 M NaOH and 1% SDS) was added. The Eppendorf was immediately inverted several times to lyse the cell suspension. SDS is an anionic detergent that disrupts cell membranes.

- 28 - Table 3.1 Bacterial strains used in this study Strain Plasmid Characteristics Reference/Source

S. sonnei none Wild type, South UDW* S. flexneri Africa Enteropathogenic E. coli E. coli DH5α none NalR Sambrook et al., 1989 E. coli DPD2794 pRecALux KanR, AmpR, Vollmer et al., 1997 DNA-damage sensitive E. coli DPD2540 pFabALux KanR, AmpR, Belkin et al., 1997 membrane-amage sensitive E. coli DE135 pUspALux2 KanR, AmpR, general Van Dyk et al., stress-sensitive 1995 *UDW University of Durban-Westville, Department of Microbiology, South Africa NalR nalidixic acid resistant AmpR ampicillin resistant KanR kanamycin resistant

Table 3.2 Antibiotics and other media supplements used in this study Supplement Stock concentration Working concentration Ampicillin 100 mg/ml 100 - 200 Pg/ml IPTG 1 M 1 mM Nalidixic acid 25 mg/ml (pH 11 with NaOH) 25 Pg/ml X-gal 20 mg/ml (in *DMF) 50 Pg/ml *DMF is dimethylformamide

Liberated DNA was denatured by NaOH and cellular RNA was degraded by RNaseA. After incubation for 5 min at 4°C, 150 Pl cold Solution C (3 M sodium acetate, pH 4.8) was added to the clear cell lysate. This precipitated out denatured chromosomal DNA and cellular proteins. The Eppendorf tube was immediately inverted several times and incubated for 15 min at 4°C. Precipitated material was pelleted by centrifugation for 15 min at 13 000 rpm and discarded. Two volumes of cold 100% ethanol (± 850 Pl) were added to the supernatant and incubated for 15 min at -70°C. Ethanol precipitates out cellular proteins and solubilises lipids present in bacterial cell walls and membranes. Precipitated DNA was pelleted by centrifugation for 15 min at 13 000 rpm and washed

- 29 - once in 70% ethanol, to remove residual salts. The purified plasmid DNA pellet was air-dried, re-suspended in 20 Pl of sterile de-ionised water (pH 7.0) and stored at -20°C.

Large scale isolations were essentially scaled-up versions of the mini-prep procedure. Transformants containing recombinant plasmids were inoculated into 100 ml LB medium and grown to saturation at 37°C in an orbital shaker. Cells were harvested by centrifugation at 8 500 rpm for 10 min at 10°C in a Beckman J2-21 centrifuge (JA-14 rotor). The supernatant was discarded and the bacterial pellet re-suspended in 10 ml of Solution A, containing 50 Pl RNaseA and incubated at room temperature. Twenty millilitres of freshly prepared Solution B was added and the centrifuge bottle was immediately inverted 10 times to obtain a viscous solution. After incubation for 4 min at 4°C, a viscous cell lysate was obtained. The cell lysate was incubated for a further 5 min at 4°C. Fifteen millilitres of cold Solution C was added to the cell lysate and immediately inverted 10 times. During subsequent incubation, for 15 min at 4°C, the centrifuge bottle was gently inverted 2 times at 2 min intervals. Precipitated material and cell debris were pelleted by centrifugation at 14 000 rpm for 30 min at 10°C, and discarded. Twenty eight millilitres of isopropanol was added to the supernatant, gently mixed, and incubated for 10 min at room temperature. During this step denatured chromosomal DNA and cellular proteins were precipitated out. Precipitated DNA was pelleted by centrifugation at 11 000 rpm for 10 min at 24°C. Residual proteins were precipitated by the addition of 5 ml of 2 M ammonium acetate (pH 7.4) and incubated for 10 min at 4°C. Cellular proteins were pelleted by centrifugation at 10 000 rpm for 10 min at 4°C and discarded. Six milliliters of isopropanol was added to the supernatant and incubated for 10 min at 4°C to precipitate out plasmid DNA. Precipitated plasmid DNA was pelleted by centrifugation at 11 000 rpm for 10 min at 4°C, washed once in 70% ethanol, air-dried, re-suspended in 500 Pl of sterile de- ionised water (pH 7.0) and stored at -20°C. The presence of cloned genes was verified by restriction endonuclease digestion and subsequent electrophoresis, polymerase chain reaction amplification and bioluminescence detection using the 1254-001 LUMINOVA luminometer (Bio-Orbit Oy, Finland).

3.2.3 Cloning of pLux Plasmids pUCD607 (Shaw and Kado, 1986) and pUC19 were restricted with Sal I. After restriction endonuclease digestion, 5 Pl gel loading buffer (40% sucrose; 0.25% bromophenol blue) was added to restriction mixtures and analysed by horizontal agarose gel electrophoresis. DNA was loaded in 0.7 - 1.0% agarose gels and electrophoresis was carried out at 60 - 80 V for 2 - 3 h in 1 x TAE electrophoresis buffer (40 mM Tris-acetate; 20 mM glacial acetic acid; 2 mM EDTA). After

- 30 - electrophoresis, gels were stained for 15 min in 0.5 Pg/ml ethidium bromide. Fluorescent DNA bands were visualised on a UV transilluminator (UVP Inc.). The sizes of the restriction fragments were calculated using the SW5000 Gel Documentation System (UVP Inc.) by comparison with sizes of molecular weight marker II (phage λ DNA cleaved with Hind III) and molecular weight marker VI (pBR328 DNA restricted with Bgl I + pBR328 DNA restricted with Hinf I).

3.2.3.1 Extraction of restriction fragments Agarose gel slices containing the 8 649 bp luxCDABE and the 2 686 bp pUC19 fragment were excised and transferred into separate pre-weighed Eppendorf tubes. The gel slices were weighed and their volume estimated (1 g ± 1 ml). The QIAEX II Agarose Gel Extraction Kit (Qiagen GmBH and Qiagen Inc., USA) was used to isolate DNA fragments from agarose gel. Three volumes of Buffer QXI were added to 1 volume of gel for DNA fragments 100 - 4 000 bp. After being vortexed for 30 sec, 10 Pl QIAEX II was mixed with the sample, and incubated for 10 min at 50°C to solubilise the agarose and bind the DNA to the QIAEX II particles. If the colour of the mixture turned orange or purple, indicating an increase in pH, 10 Pl of 3 M sodium acetate (pH 5.0) was added, and the incubation time was increased to 15 min. This ensured a pH ≤ 7.5 during which efficient absorption of DNA to QIAEX II particle occurred. The reaction was mixed at 2 min intervals to keep QIAEX II in suspension. The reaction was centrifuged for 30 sec at 13 000 rpm and the supernatant carefully aspirated. The pellet was washed with 500 Pl Buffer QXI and centrifuged for 30 sec at 13 000 rpm to remove residual agarose contaminants. The pellet was washed twice with 500 Pl Buffer PE, centrifuged for 30 sec at 13 000 rpm to remove residual salt contaminants, air-dried until it turned white and re-suspended in 20 Pl DNA dilution buffer (Roche Biochemicals). The DNA elution was centrifuged for 30 sec at 13 000 rpm and the eluate containing the purified DNA was transferred to a sterile Eppendorf tube and stored at -20°C.

3.2.3.2 Ligation of DNA fragments DNA molecules having compatible sticky ends were ligated as follows: a typical ligation reaction consisted of 4 Pl linearised pUC19, 4 Pl of the luxCDABE fragment, 1 µl 10X ligation buffer, and 1 Pl of T4 DNA ligase (Roche Biochemicals). The reaction mixture was added to a sterile Eppendorf and incubated for 20 h at 25°C. The ligation reaction was used directly for transformation of electrocompetent E. coli DH5α, Enteropathogenic E. coli, S. flexneri and S. sonnei cells.

3.2.3.3 Preparation of electrocompetent cells

- 31 - E. coli DH5α, Enteropathogenic E. coli, S. flexneri and S. sonnei cells electrocompetent cells were prepared using the same method. Three millilitres of an overnight bacterial culture was added to 100 ml of LB broth and incubated with agitation until the cell

density reached 0.5 absorbance units at 600nm. The culture was immediately placed at 4°C to stop further growth. For all steps the cells were maintained at 4°C and pelleted by centrifugation for 10 min at 10 000 rpm. The pellet was gently re-suspended in 100 ml cold 10% (v/v) glycerol and centrifuged, resuspended in 50 ml cold 10% (v/v) glycerol and centrifuged, and re-suspended in 5 ml cold 10% (v/v) glycerol and centrifuged. Pelleted cells were finally re-suspended in 2.5 ml cold 10% (v/v) glycerol. Sixty microlitre aliquots of electrocompetent cells were dispensed into Eppendorf tubes, snap frozen in liquid nitrogen and stored at -70°C.

3.2.3.4 Transformation Recombinant plasmids present in the ligation reaction mixture (Section 3.2.3.2), pRecALux, pFabALux and pUspALux2 (Table 3.3) were introduced into competent E. coli DH5α, Enteropathogenic E. coli, S. flexneri and S. sonnei cells by transformation. The electroporation procedure was used. Sixty microlitres of electrocompetent cells were gently thawed at 4°C. Sterile 0.2 cm electroporation cuvettes and the white chamber slide were also placed at 4°C. Two microlitres of plasmid DNA was added to the electrocompetent cells, gently mixed and placed at 4˚C. The Gene Pulser apparatus (BIORAD) was set at 25 µF and 2.5 kV. The Pulse controller was set at 200 Ω. The electrocompetent cell and plasmid DNA mixture was transferred to the pre-chilled electroporation cuvette. The cuvette was gently tapped to ensure that the plasmid DNA/electrocompetent cell mixture was in contact with both aluminium sides of the cuvette. The cuvette was placed in the chilled safety chamber slide and pushed until the cuvette made contact with the electrodes in the base of the chamber. One pulse was applied at the settings specified above to allow plasmid DNA to enter the electrocompetent cells. The slide with the cuvette was removed from the chamber, and 900 Pl of LB broth was immediately added to the cuvette. The rapid addition of LB broth to the cuvette was very important in maximising the recovery of transformants.

- 32 - Table 3.3 Plasmids used or constructed in this study Plasmid Characteristics Source/Reference pUCD607 20.3 kb plasmid, AmpR Shaw and Kado, 1986 pUC19 2.686 kb plasmid, AmpR Stratagenea pLux 11.335 kb plasmid, AmpR This study pRecALux > 23 kb plasmid, KanR, AmpR Vollmer et al., 1997 pFabALux > 23 kb plasmid, KanR, AmpR Belkin et al., 1997 pUspALux2 > 23 kb plasmid, KanR, AmpR Van Dyk et al., 1995 AmpR ampicillin resistant KanR kanamycin resistant aStratagene: La Jolla, California, USA.

The electroporated cells were incubated for 1 h at 37°C with agitation. This allowed for replication of recombinant plasmids before a selective medium was used. One hundred microlitres of the cell suspension was spread onto LB agar plates (supplemented with 100 Pl/ml ampicillin) containing 20 Pl of 1 M IPTG and 50 Pl of a 2% solution of X-gal (Table 2.2).

3.2.4 Screening of transformed cells Insertional inactivation of the lacZ gene allowed for early differentiation of bacterial cells harbouring recombinant plasmids (cream colonies) from those carrying only pUC19 (blue colonies). Cream colonies growing on LB agar plates supplemented with 100 Pg/ml ampicillin were isolated and sub-cultured on master plates. Individual colonies, from master plates were inoculated into 3 ml of LB broth (supplemented with 100 Pg/ml ampicillin). Cell suspensions were incubated for 16 h at 37°C in an orbital shaker. E. coli DH5α, Enteropathogenic E. coli, S. flexneri and S. sonnei cells were tested for the presence of recombinant plasmids by plasmid DNA isolation as described previously in Section 3.2.2, restriction analysis, PCR amplification and agarose gel electrophoresis as described previously in Section 3.2.3.

3.2.4.1 Restriction analysis Only transformed E. coli DH5α, Enteropathogenic E. coli, S. flexneri and S. sonnei cells containing the putative pLux recombinant plasmid (cream colonies) were subjected to restriction analysis. Plasmid DNA was isolated from these bacterial cells, restricted with Sal I and subjected to agarose gel electrophoresis.

- 33 - 3.2.4.2 Polymerase chain reaction amplification Plasmid DNA was isolated from transformed E. coli DH5α, Enteropathogenic E. coli, S. flexneri and S. sonnei cells containing the putative pLux (Section 3.2.3) recombinant plasmid (cream colonies), pRecALux, pFabALux and pUspALux2 (Table 3.3), and subjected to PCR amplification and agarose gel electrophoresis. The PCR primers LUX A1 (Forward) and LUX A2 (Reverse), derived from the V. fischeri luxA gene, were used (Hira, 2002).

Primer LUX A1 (Forward): 5´-CCGAC(A/T)G(C/A)(G/C/A)CA(C/A)CC(A/T)G(C/T)(A/T)CG-3´

Primer LUX A2 (Reverse): 5´-CC(C/A)GT(G/C/T)GCATCAATAT(C/T)(A/T)(C/T)-3´

Plasmid DNA was amplified in 20 Pl reaction mixtures. Each reaction contained 100 ng template plasmid DNA, 0.25 µM of each primer, 500 PM of each dNTP, 1 x

Supertherm Taq buffer, 2.5 U Supertherm Taq polymerase and 0 - 3.75 mM MgCl2. The PCR procedure was performed according to the parameters in Table 3.4. PCR products were subjected to agarose gel electrophoresis as described in Section 3.2.3.

Table 3.4 PCR parameters for amplification of the luxA gene Steps Temperature (°C) Time (min) Initial denaturation 94 5.0 35 cycles of denaturation 92 0.5 35 cycles of annealing 50 1.0 35 cycles of primer extension 72 1.0 Final primer extension 72 5.0

3.2.5 Assaying for expression of the luxCDABE operon The 1254 - 001 LUMINOVA luminometer (Bio-Orbit Oy, Finland) was used to determine whether the luxCDABE operon was expressed in the five prokaryotic biosensor systems constructed in this study (Table 3.5). This instrument is a photon counting device with a photomultiplier tube. The single sample 1254 - 001 luminometer is sensitive and able to detect 0.05 femtomoles (0.03 pg) of ATP per bioluminescence test in a 1 ml luminometer cuvette. This luminometer displayed results as relative light units (RLU), which was a value directly correspondent to the

- 34 - light intensity emitted by a sample in a 1 ml luminometer cuvette, during a defined measurement period. Single colonies of the five prokaryotic biosensors were each inoculated into 3 ml of LB broth and incubated at 37°C. After 16 h the biosensor broth cultures were allowed to equilibrate to 26°C. The bioluminescence of the respective prokaryotic biosensors was measured using 1 ml luminometer cuvettes.

3.2.6 Freeze-drying of prokaryotic biosensor cells E. coli DH5α pLux, Enteropathogenic E. coli pLux, S. flexneri pLux and S. sonnei pLux biosensor systems (Table 3.5) were freeze-dried using the following procedures. Six millilitres of an overnight bacterial culture was added to 200 ml of LB broth supplemented with 200 Pl of a 100 mg/ml ampicillin stock. The culture was incubated with agitation at 37C until the respective cell densities of the biosensor systems

reached 0.6 - 0.7 absorbance units at 600nm. The culture was immediately placed on ice to stop further growth. Cells were pelleted by centrifugation for 30 min at 4 000 rpm at 4C. After centrifugation two different procedures were followed. During the first procedure the pellet was gently resuspended in 20 ml cold LB and 20 ml cold 24% trehalose (4.8 g trehalose in 20 ml deionised water). One millilitre of this suspension was dispensed into each sterile 2.5 ml freeze-drying vial. Freeze-drying could also be performed using the skim milk method of Malik (1992). Prior to freeze- drying the vials were prepared as follows: a 20% skim milk / 5% meso-inositol solution was prepared by dissolving 2 g skim milk powder and 0.5 g meso-inositol in 10 ml de- ionised water. Half a millilitre of this solution was added to each vial. The vials were placed onto freeze-drying trays, without the rubber stoppers, and wrapped in foil. The tray of vials was autoclaved at 115ºC for 13 min and placed at 4ºC, until required. In this procedure the cell pellet was re-suspended in 10 ml of 5% meso-inositol. Half a millilitre of the cell suspension was placed into each vial over the skim milk using sterile glass pipettes. The vials from both procedures were partially sealed with rubber stoppers, frozen in liquid nitrogen for 2 min and freeze-dried for 20 h at -55C. Only those vials containing a fine white, uniform freeze-dried product were sealed and stored at 4C.

3.2.7 Resuscitation of freeze-dried prokaryotic biosensor cells One millilitre of LB broth was dispensed into each vial containing the freeze-dried biosensor cells and the vial was re-sealed. The biosensor cells freeze-dried in trehalose were resuscitated for 30 min at 26C, without agitation. The biosensor cells freeze- dried in skim milk were resuscitated for 60 min at 26C, with agitation. One millilitre of resuscitated biosensor cells was diluted in 9 ml of LB broth that had been preheated for 30 min at 26C. For optimal conditions, the resuscitated freeze-dried biosensor

- 35 - cells were washed in 0.1 M potassium chloride (KCl) to remove excess LB broth and skim milk and utilised within 30 min of washing.

3.3 RESULTS 3.3.1 Cloning of pLux The promoterless V. fischeri luxCDABE operon, from pUCD607, was successfully integrated into the cloning vector pUC19, to yield the multi-copy 11 335 bp pLux plasmid (Fig. 3.1). The insertion of the luxCDABE operon into pLux was confirmed by restriction analysis. Restriction of pLux with Sal I produced the 2 686 bp pUC19 vector and the 8 649 bp luxCDABE-containing fragments (Fig. 3.2).

SalI

luxC

ampr

luxD plux SalI 11.69 kb

BglII luxE

luxA

luxB pUC19

Fig. 3.1 Partial restriction map of the 11 335 bp pLux plasmid which contains the luxCDABE operon

3.3.2 Isolation of plasmid DNA Mini-prep and large scale plasmid DNA isolations, from E. coli DH5α pLux, Enteropathogenic E. coli pLux, S. flexneri pLux and S. sonnei pLux, E. coli DPD2794, E. coli DPD2540 and E. coli DE135 reproducibly yielded DNA of high quality. The purity of plasmid DNA isolated using both procedures, was tested by restriction

analysis (Fig. 3.2) and yielded plasmid DNA with an OD260/OD280 ratio greater than 1.7. pLux 11 335 bp

- 36 - 3.3.3 Polymerase chain reaction amplification PCR amplification using the LUX A1 (Forward) and LUX A2 (Reverse) primer set for pLux, pRecALux, pFabALux and pUspALux2 all yielded the approximately 700 bp PCR product (Fig. 3.3).

3.3.4 Construction of prokaryotic biosensor systems E. coli DH5α, Enteropathogenic E. coli, S. flexneri and S. sonnei were successfully transformed with pLux, pRecALux, pFabALux and pUspALux2 to create the bioluminescent biosensors: E. coli DH5α pLux, Enteropathogenic E. coli pLux, S. flexneri pLux and S. sonnei pLux biosensor systems, E. coli DH5α recA: lux, E. coli DH5α fabA: lux, E. coli DH5α uspA: lux and S. sonnei pLux.

Table 3.5 Prokaryotic biosensor systems constructed in this study Biosensor Plasmid Promoter fusion S. sonnei pLux pLux lac: luxCDABE S. flexneri pLux pLux lac: luxCDABE E. coli DH5α pLux pLux lac: luxCDABE E. coli DH5α recA: lux pRecALux recA: luxCDABE E. coli DH5α fabA: lux pFabALux fabA: luxCDABE E. coli DH5α uspA: lux pUspALux2 uspA: luxCDABE Enteropathogenic E. coli pLux lac: luxCDABE pLux

- 37 -

1 2 3 4 5 6 7 8 9 10

23 130 bp → ← 20 300 bp 9 416 bp → ← 8 649 bp

← 2 686 bp 2 322 bp →

Fig. 3.2 Restriction analysis of pLux and plasmid DNA of the E. coli DH5α pLux, E. coli DH5α recA: lux, E. coli DH5α fabA: lux, E. coli DH5α uspA: lux biosensor systems. Lane 1: pUC19 restricted with Sal I; lane 2: pUC19; lane 3: pUCD607 restricted with Sal I; lane 4: pUCD607; lane 5: λ DNA restricted with Hind III (molecular weight marker M II); lane 6: pLux restricted with Sal I; lane 7: pLux; lane 8: pRecALux; lane 9: pFabALux; and lane 10: pUspALux2.

3.3.5 Assaying for expression of the luxCDABE operon Bioluminescence was most conveniently measured using the 1254 - 001 LUMINOVA luminometer. At 20 - 26°C, all five prokaryotic biosensor systems (Table 3.5) produced greater than 900 000 RLU.

3.3.6 Resuscitation of freeze-dried prokaryotic biosensor cells Biosensor cultures (Table 3.5) freeze-dried in skim milk did not consistently produce a uniform freeze-dried product. However, both trehalose and skim milk freeze-dried biosensor cultures were successfully resuscitated at 26°C, in LB broth. Maximum bioluminescence (greater than 900 000 RLU) was produced 30 min and 60 min after commencement of resuscitation of the trehalose and skim milk biosensor cultures, respectively.

- 38 -

1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17

2 176 bp →

1 033 bp → ← 700 bp 653 bp →

Fig. 3.3 Polymerase chain reaction amplification of the plasmids of the E. coli DH5α pLux, E. coli DH5α recA:lux, E. coli DH5α fabA:lux, E. coli DH5α uspA:lux biosensor systems. Lane 1: negative control (without plasmid DNA); lanes 2 - 4: positive control (pUCD607); lanes 5 - 7: pLux; lanes 8 - 10: pRecALux; lane 11: pBR328 DNA restricted with Bgl I + pBR328 DNA restricted with Hinf I (molecular weight marker VI); lanes 12 – 14: pFabALux; and lanes 15 - 17: pUspALux2.

3.4 DISCUSSION The integration of the entire luxCDABE operon in the E. coli DH5α pLux, EPEC pLux, S. flexneri pLux and S. sonnei pLux biosensor systems, E. coli DH5α recA:lux, E. coli DH5α fabA:lux, E. coli DH5α uspA:lux and S. sonnei pLux biosensor systems was advantageous. Apart from achieving reproducible bioluminescence results, it was economical as it negated the exogenous addition of decanal. Previously, Blouin et al. (1996) used a biosensor with the luxAB genes alone. An immediate advantage of this approach was the simplification and potential stabilisation of the biosensor’s expression systems by having eliminated the complex enzymatic subprocesses involved in the production and recycling of the natural luciferase substrate. However, this shorter gene construct, although mechanistically simpler, required a substantial excess of decanal to be exogenously added, to ensure that it was not limiting. At these higher concentrations decanal became strongly inhibitory, thereby complicating the interpretation of the results and compromising reproducibility of the experiments. Actively growing bacteria cannot meet the requirements for a portable toxicity biosensor, i.e., using biosensor cells on-site to evaluate environmental pollution. Thus,

- 39 - various approaches of maintaining bioluminescent biosensors have previously been used. These include immobilisation of exponentially growing cells to calcium-alginate beads (Davidov et al., 2000) and glass beads in agar (Gu and Chang, 2001). The biosensors constructed in this research (Table 3.5) were developed with the intention to be used as an on-site acute toxicity test. The use of freeze-drying for producing inocula for ecotoxicity testing has previously been used for the bioluminescent-based MicrotoxTM and BioToxTM kits (Kohler et al., 2000). These tests were based on V. fischeri, without any genetic manipulation. However, since the marine bacterium V. fischeri is not found naturally in freshwater and wastewater samples and required a saline environment, E. coli DH5α and S. sonnei were preferred hosts for the integration of the luxCDABE operon. During the freeze-drying of these genetically engineered biosensor cells (Table 3.5) the maintenance, production and metabolic functions of the luxCDABE genes was of primary concern. However, freeze-drying in skim milk did not consistently yield uniform freeze-dried products and required 60 min for resuscitation, with agitation, to create a homogenous test solution (Section 3.3.5). For these reasons the biosensor systems freeze-dried in skim milk did not indicate a practical application as a potential on-site biosensor. On the other hand, freeze-drying in trehalose consistently yielded uniform freeze-dried products and only required 30 min of resuscitation, without agitation. It seemed that trehalose maintained the viability and biosensing activity of these biosensors, as seen by the high bioluminescence values after resuscitation in LB broth (Section 3.3.5). Trehalose probably increased the tolerance of the biosensor cells to drying, by its ability to lower the temperature of the dry biosensor membrane phase transition, and maintain the general protein structure in the dry state (Leslie et al., 1995). Although all five prokaryotic biosensor systems produced maximum bioluminescence (greater than 900 000 RLU) at 30 - 37°C, the bacterial luciferase was most stable at 20 - 26°C. Therefore, the biosensor cells were maintained at 20 - 26°C, after resuscitation. This is advantageous for potential on-site applications of these biosensor systems for the detection of pollutants, since ambient temperature falls within this temperature range. Thus, the five freeze-dried trehalose biosensor systems (Table 3.5), could be used for water toxicity monitoring since they all have fast resuscitation rates, and produced reliable and rapid bioluminescence responses (Section 3.3.5). The DNA- damage, membrane damage and general toxicity responses of E. coli DH5α pLux, E. coli DH5α recA: lux, E. coli DH5α fabA: lux, E. coli DH5α uspA: lux and S. sonnei pLux will possibly allow for the evaluation of a range of environmental pollutants in Chapters Four and Five.

- 40 - CHAPTER FOUR: STANDARDISATION OF THE TOXICITY TESTS

4.1 INTRODUCTION The widespread use of petroleum products and the current regulations requiring underground storage tanks to be up-graded, replaced or closed, has increased the number of petroleum-contaminated sites. Of particular concern are the more water- soluble components such as benzene, toluene, ethylbenzene and xylene (BTEX). These organic solvents present a concern in many sites worldwide due to their high toxicity, carcinogenic and teratogenic effects, even at low concentrations. Heavy metals such as copper, zinc, chromium, lead and cadmium are metals with a density above 5 g/cm3, and cause deleterious health effects at high concentrations (Holmes, 1996). Although some heavy metal ions are essential trace elements since they can form complex compounds, most heavy metals are toxic at higher concentrations (Nies, 1999). However, the challenge of lower concentrations of heavy metals has led to the evolution of transport mechanisms for heavy metal ion homeostasis and detoxification. Thus, E. coli and other bacteria are able to catalyse the extrusion of low concentrations of metal ions, from the cytosol, via these transport mechanisms. In E. coli and other bacteria, soft metal ion-translocating ATPases (transport pumps) catalyse uptake or confer resistance to copper, zinc, cadmium, arsensic and lead. Fig. 4.1 shows the soft metal ATPases of E. coli. Zinc ions, (right-hand side of Fig. 4.1) are accumulated by the ZnuABC ATPase and an ABC transporter. Excess Zn (II) is extruded by the ZntA P-type ATPase. The opposing activities of these two pumps provide for zinc homeostasis. ZntA also confers resistance to Pb (II) and Cd (II). NDH-II is a putative Cu (II) reductase. The CopA Cu (I)-translocating P-type ATPase (middle of Fig. 4.1) contributes to copper homeostasis. The proteins responsible for uptake and intracellular transport of copper have not been identified in E. coli. Arsenic oxyanions, (left-hand side of Fig. 4.1) cycle through the cell by uptake of arsenate by a phosphate- translocating ATPase, an ABC transporter. The ArsC arsenate reductase reduces As (V) to As (III), which is then extruded from the cell by the ArsAB arsenite- translocating ATPase (Gatti et al., 2000). The increasing awareness of the environmental problems caused by industrial and agricultural pollution has created a demand for progressively more sophisticated detection methods. Thus, an increasingly varied set of bioassays has also been under continuous development for environmental monitoring purposes. A variety of organisms (cellular or sub-cellular systems) have been employed for these purposes, from whole-organism assays such as fish and Daphnia toxicity testing, frog lethality and bacterial inhibition testing, to immunological determination of specific pesticides (Kohler et al., 2000).

- 41 -

Fig. 4.1 A schematic representation of the Escherichia coli soft metal ion-translocating ATPases (Gatti et al., 2000).

Among the test organisms, bacteria hold a special niche. The advantages offered by microbial toxicity testing include high sensitivity, low costs, large homogeneous test populations, and rapid responses (Belkin et al., 1997). Thus, in this chapter, the relative toxicities of freeze-dried E. coli DH5α pLux, EPEC pLux, S. flexneri pLux, E. coli DH5α recA: lux, E. coli DH5α fabA: lux, E. coli DH5α uspA: lux and S. sonnei pLux, to standard concentrations of heavy metals and volatile organic solvents, were evaluated.

4.2 MATERIALS AND METHODS 4.2.1 Preparation of toxicant samples The heavy metal compounds (Table 4.1) and organic solvents (Table 4.2) were of analytical grade and were used without further purification. The heavy metals and organic solvents were dissolved separately in sterile de-ionised water to yield five working concentrations of 0.01 mg/l, 0.1 mg/l, 1 mg/l, 10 mg/l and 100 mg/l. Two hundred and seventy microlitres of each of these five working concentrations were dispensed in triplicate into the wells of a microtitre plate, as shown in Fig. 4.2. The 0 mg/l control sample simply contained 270 µl of sterile de-ionised water without toxicant. The heavy metal compounds (Table 4.1) and organic solvents (Table 4.2)

- 42 - were handled with latex gloves. When working with the organic solvents the added precautions of working with face-masks in a fume cupboard were necessary, due to the emission of toxic fumes.

4.2.2 Toxicity test The biosensors (Table 3.5) were freeze-dried using the protocols in Sections 3.2.8 and 3.2.9. One millilitre of the diluted resuscitated biosensor cells (Section 3.2.1) was added to a sterile Eppendorf tube and centrifuged for 2 min at 13 000 rpm and re- suspended in 1 ml of 0.1 M KCl. Thirty microlitres of this biosensor cell suspension was added to all 72 of the microtitre plate wells used in Fig. 4.2. The microtitre plate containing both the toxicant and biosensor cells was inserted into the Fluroskan Ascent FL instrument for 60 min. The Fluroskan Ascent FL (Thermo Labsystems) was programmed to: measure the RLU of all 72 microtitre plate wells at 15 min intervals for 60 min; shake the microplate before each measurement to ensure thorough mixing of samples and biosensor cells; and perform all measurements at an ambient temperature of 20 - 26ºC.

Table 4.1 The various heavy metal compounds used in this study Heavy metal compound Heavy metal oxidation state Copper sulphate Cu(II) Zinc sulphate Zn(II) Potassium dichromate Cr(III) Lead acetate Pb(II) Cadmium acetate Cd(II) Chromium trioxide Cr(VI)

- 43 -

control control control control control control control control control control control control 0 mg/l 0 mg/l 0 mg/l 0 mg/l 0 mg/l 0 mg/l 0 mg/l 0 mg/l 0 mg/l 0 mg/l 0 mg/l 0 mg/l

Sample A Sample A Sample A Sample B Sample B Sample B Sample C Sample C Sample C Sample D Sample D Sample D

0.01mg/l 0.01mg/l 0.01mg/l 0.01mg/l 0.01mg/l 0.01mg/l 0.01mg/l 0.01mg/l 0.01mg/l 0.01mg/l 0.01mg/l 0.01mg/l

Sample A Sample A Sample A Sample B Sample B Sample B Sample C Sample C Sample C Sample D Sample D Sample D

0.1 mg/l 0.1 mg/l 0.1 mg/l 0.1 mg/l 0.1 mg/l 0.1 mg/l 0.1 mg/l 0.1 mg/l 0.1 mg/l 0.1 mg/l 0.1 mg/l 0.1 mg/l

Sample A Sample A Sample A Sample B Ssample B Sample B Sample C Sample C sample C Sample D Sample D Sample D 1 mg/l 1 mg/l 1 mg/l 1 mg/l 1 mg/l 1 mg/l 1 mg/l 1 mg/l 1 mg/l 1 mg/l 1 mg/l 1 mg/l

Sample A Sample A Sample A Sample B Sample B Sample B Ssample C Sample C Sample C Saample D Sample D Sample D 10 mg/l 10 mg/l 10 mg/l 10 mg/l 10 mg/l 10 mg/l 10 mg/l 10 mg/l 10 mg/l 10 mg/l 10 mg/l 10 mg/l

Sample A Sample A Sample A Sample B Sample B Sample B Sample C Sample C Sample C Sample D Sample D Sample D 100 mg/l 100 mg/l 100 mg/l 10 mg/l 100 mg/l 100 mg/l 100 mg/l 100 mg/l 100 mg/l 100 mg/l 100 mg/l 100 mg/l

Fig. 4.2 Microtitre plate template used in the Fluoroskan Ascent FL where A, B, C or D represent the various pollutant samples.

- 44 - Table 4.2 The various organic solvents used in this study (Cole, 1994)

Organic solvent Molecular Solubility in Viscosity weight (g) water at 20ºC (Centistokes) (mg/l) Benzene 78.11 1.791 0.5 Toluene 92.14 515 0.5 Ethylbenzene 106.17 775 0.6 Xylene 106.17 150 0.6

4.2.3 Calculation of bioluminescence and EC values Since all bioluminescence values were taken as the average of triplicate samples, percentage bioluminescence was defined as follows;

Bioluminescence (%) = Average bioluminescence at x min at y mg/l x 100 Average initial bioluminescence

Where: initial bioluminescence = bioluminescence at 1 min at 0 mg/l; x = specific time at either 1, 15, 30, 45 or 60 min; and y = 0.01, 0.1, 1, 10 or 100 mg/l, of heavy metal compounds and BTEX. The bioluminescence profiles of E. coli DH5α pLux, E. coli DH5α recA: lux, E. coli DH5α fabA: lux, E. coli DH5α uspA: lux, S. sonnei pLux, S. flexneri pLux and

Enteropathogenic E. coli (EPEC) were determined using EC50, *EC20, and *EC100 values, which were defined as follows:

EC50 – Effective concentration of heavy metal compounds and BTEX that resulted in 50% inhibition in bioluminescence, as compared to the initial percentage bioluminescence.

*EC20 – Effective concentration of BTEX that resulted in 20% induction in bioluminescence, as compared to the initial percentage bioluminescence.

*EC100 – Effective concentration of BTEX that resulted in 100% induction in bioluminescence, as compared to the initial percentage bioluminescence.

- 45 - 4.3 RESULTS The toxicity responses of E. coli DH5α pLux, E. coli DH5α recA: lux, E. coli DH5α fabA: lux, E. coli DH5α uspA: lux, S. sonnei pLux, S. flexneri pLux, EPEC pLux, the TM V. fischeri-based BioTox kit and Daphnia LC50 toxicity test, to heavy metal

compounds and BTEX, were represented as EC50, *EC20 and *EC100 values and summarised in Tables 4.3 - 4.6. Standard deviations and average percentage bioluminescence values for Figs 4.3 - 4.42 were calculated using the Fluoroskan Ascent FL Software (Thermo Labsystems), and were tabulated in Appendix One. The low standard deviation values lent to the reproducibility and sensitivity of the bioluminescence tests when using the Fluoroskan Ascent FL Software (Thermo Labsystems).

Table 4.3 EC50 values of the luxCDABE-marked bacterial systems for the heavy metal compounds

luxCDA EC50 values of heavy metal compounds (mg/l) BE- copper zinc potassium lead cadmium chromium marked sulphate sulphate dichromat acetate acetate trioxide bacteria e S. sonnei 0.01 0.01 0.01 0.01 0.01 0.01 pLux (15-30 (15-30 (30-45 (15-30 (30-45 (30-45 min) min) min) min) min) min) S. flexneri 0.5 0.5 0.01 0.01 0.05 0.5 pLux (45-60 (45-60 (30-45 (15-30 (45-60 (45-60 min) min) min) min) min) min) E. coli 10 1 10 10 1 1 DH5α (1-15 min) (0-1 min) (0-1 min) (0-1 min) (1-15 min) (0-1 min) pLux E. coli 1 1 10 10 1 1 DH5α (1-15 min) (0-1 min) (15-30 (0-1 min) (0-1 min) (0-1 min) recA: lux min) E. coli 1 1 10 10 1 1 DH5α (1-15 min) (1-15 min) (0-1 min) (0-1 min) (1-15 min) (0-1 min) fabA: lux E. coli 1 1 100 10 1 1 DH5α (1-15 min) (0-1 min) (0-1 min) (0-1 min) (1-15 min) (1-15 min) uspA: pLux EPEC 0.01 0.01 0.01 0.01 0.5 0.1 pLux (45-60 (45-60 (30-45 (30-45 (45-60 (45-60 min) min) min) min) min) min) V. fischeri- 10 100 100 1 1 10 BioToxTM (1-15 min) (1-15 min) (0-1 min) (60 min) (60 min) (1-15 min) kit

Values in parentheses represent the time during which the EC50 values were obtained.

- 46 - Table 4.4 EC50 values of the luxCDABE-marked bacterial systems for the volatile organic solvents

luxCDABE- EC50 values of volatile organic solvents (mg/l) marked Benzene Toluene Ethylbenzene Xylene bacteria S. sonnei 0.01 0.01 0.01 0.01 pLux (15-30 min) (15-30 min) (15-30 min) (15-30 min) S. flexneri 0.01 0.01 0.01 0.01 pLux (1-15 min) (15-30 min) (15-30 min) (15-30 min) EPEC 0.01 0.01 0.01 0.1 pLux (15-30 min) (30-45 min) (30-45 min) (30-45 min) V. fischeri- 100 100 10 100 BioToxTM kit (0-1 min) (0-1 min) (0-1 min) (0-1 min)

Values in parentheses represent the time during which the EC50 values were obtained.

Table 4.5 *EC20 values of the E. coli DH5α pLux biosensor for the volatile organic solvents

*EC20 values of volatile organic solvents (mg/l) Benzene Toluene Ethylbenzene Xylene 10 10 10 10 (1-15 min) (1-15 min) (15-30 min) (15-30 min)

Values in parentheses represent the time during which the *EC20 values were obtained.

Table 4.6 *EC100 values of the luxCDABE-marked bacterial biosensors for the volatile organic solvents

luxCDABE- * EC100 values of volatile organic compounds (mg/l) marked Benzene Toluene Ethylbenzene Xylene biosensors E. coli DH5α 0.1 0.1 10 0.1 recA: lux (15-30 min) (15-30 min) (30-45 min) (15-30 min)

E. coli DH5α 10 10 10 10 fabA: lux (15-30 min) (15-30 min) (15-30 min) (15-30 min)

E. coli DH5α 100 100 10 10 uspA: pLux (15-30 min) (15-30 min) (1-15 min) (1-15 min)

Values in parentheses represent the time during which the *EC100 values were obtained.

- 47 - 4.3.1 Selection of the most sensitive biosensor systems E. coli DH5α pLux, EPEC pLux, S. flexneri pLux and S. sonnei pLux, E. coli DH5α recA: lux, E. coli DH5α fabA: lux, E. coli DH5α uspA: lux and S. sonnei pLux were all tested in the presence of chromium trioxide (Figs 4.3 – 4.10) and xylene (Figs 4.11 – 4.18). E. coli DH5α pLux, EPEC pLux, S. flexneri pLux and S. sonnei pLux biosensor systems all displayed similar bioluminescence profiles, while E. coli DH5α recA: lux, E. coli DH5α fabA: lux and E. coli DH5α uspA: lux displayed similar bioluminescence profiles. Therefore, only the S. sonnei pLux, E. coli DH5α uspA: lux and V. fischeri- based BioToxTM kit figures were selected for all following toxicity tests in Chapters Four and Five.

120

100 0mg/l 80 0.01mg/l 0.1mg/l 60 1mg/l 40 10mg/l 100mg/l

Bioluminescence (%) 20

0 1 15304560 Time (min)

Fig. 4.3 Bioluminescent response of the S. sonnei pLux biosensor in the presence of various concentrations of chromium trioxide.

120

100 0mg/l 80 0.01mg/l 0.1mg/l 60 1mg/l 40 10mg/l 100mg/l

Bioluminescence (%) 20

0 115304560 Time (min)

Fig. 4.4 Bioluminescent response of the S. flexneri pLux biosensor in the presence of various concentrations of chromium trioxide.

- 48 -

120 100 0 mg/l 80 0.01 mg/l 0.1 mg/l 60 1 mg/l 40 10 mg/l 100 mg/l Bioluminescence (%) 20 0 1 15304560

Time (min)

Fig. 4.5 Bioluminescent response of the E. coli DH5α pLux biosensor in the presence of various concentrations of chromium trioxide.

120

100 0 mg/l 80 0.01 mg/l 0.1 mg/l 60 1 mg/l 40 10 mg/l 100 mg/l Bioluminescence (%) 20 0 1 15304560 Time (min)

Fig. 4.6 Bioluminescent response of the Enteropathogenic E. coli pLux biosensor in the presence of various concentrations chromium trioxide

300

250 0 mg/l 200 0.01 mg/l 0.1 mg/l 150 1 mg/l 100 10 mg/l 100 mg/l Bioluminescence (%) Bioluminescence 50

0 1 15304560 Time (min)

Fig. 4.7 Bioluminescent response of the E. coli DH5α recA:lux biosensor in the presence of various concentrations of chromium trioxide.

- 49 -

250 200 0 mg/l 0.01 mg/l

150 0.1 mg/l 1 mg/l 100 10 mg/l 100 mg/l 50 Bioluminescence (%)

0 1 15304560 Time (min)

Fig. 4.8 Bioluminescent response of the E. coli DH5α fabA: lux biosensor in the presence of various concentrations of chromium trioxide.

450 400 350 0 mg/l 300 0.01 mg/l 250 0.1 mg/l 200 1 mg/l 150 10 mg/l 100 100 mg/l Bioluminescence (%) 50

1 15304560

Time (min)

Fig. 4.9 Bioluminescent response of the E. coli DH5α uspA: lux biosensor in the presence of various concentrations chromium trioxide.

120 100 0 mg/l

80 0.01 mg/l 0.1 mg/l 60 1 mg/l 40 10 mg/l 100 mg/l

Bioluminescence (%) 20

0 0 15304560 Time (min)

Fig. 4.10 Bioluminescent response of V. fischeri to various concentrations of chromium trioxide.

- 50 - 120 450 400100 350 0 0mg/l mg/l 30080 0.010.01 mg/l mg/l 250 0.10.1 mg/l mg/l 60 1 mg/l 200 1 mg/l 150 10 mg/l 40 10 mg/l 100 100 mg/l

Bioluminescence (%) 100 mg/l Bioluminescence (%) 5020 0 0 1 15304560 115304560Time (min) Time (min)

Fig. 4.11 Bioluminescent response of the S. sonnei pLux biosensor in the presence of various concentrations of xylene.

120

100 0 mg/l 80 100 mg/l 10 mg/l 60 1 mg/l 40 0.1 mg/l 0.01 mg/l

Bioluminescence (%) 20

0 115304560 Time (min)

Fig. 4.12 Bioluminescent response of the S. flexneri pLux biosensor in the presence of various concentrations of xylene.

160 140 0 mg/l 120 0.01 mg/l 100 0.1 mg/l

80 1 mg/l

60 10 mg/l

40 100 mg/l

Bioluminescence (%) 20 0 1 15304560 Time (min)

Fig. 4.13 Bioluminescent response of the E. coli DH5α pLux biosensor in the presence of various concentrations of xylene.

- 51 - 300

250 0 mg/l 200 0.01 mg/l 0.1 mg/l 150 1 mg/l

100 10 mg/l 100 mg/l

Bioluminescence (%) Bioluminescence 50

0 115304560 Time (min)

Fig. 4.14 Bioluminescent response of the E. coli DH5α recA: lux biosensor in the presence of various concentrations of xylene.

300

250 0 mg/l 200 0.01 mg/l 0.1 mg/l 150 1 mg/l 100 10 mg/l 100 mg/l

Bioluminescence (%) 50

0 1 15304560 Time (min)

Fig. 4.15 Bioluminescent response of the E. coli DH5α fabA: lux biosensor in the presence of various concentrations of xylene.

600

500 0 mg/l 400 0.01 mg/l 0.1 mg/l 300 1 mg/l 200 10 mg/l 100 mg/l

Bioluminescence (%) Bioluminescence 100

0 115304560 Time (min)

Fig. 4.16 Bioluminescent response of the E. coli DH5α uspA: lux biosensor in the presence of various concentrations of xylene.

- 52 - 140

120 0 mg/l 100 100 mg/l 80 10 mg/l 60 1 mg/l 0.1 mg/l 40 0.01 mg/l Bioluminescence (%) 20

0 1 15304560 Time (min)

Fig. 4.17 Bioluminescent response of the Enteropathogenic E. coli pLux biosensor in the presence of various concentrations of xylene

120

100 0 mg/l 80 0.01 mg/l 0.1 mg/l 60 1 mg/l 40 10 mg/l 100 mg/l

Bioluminescence (%) 20

0 1 15304560 Time (min)

Fig. 4.18 Bioluminescent response of V. fischeri in the presence of various concentrations of xylene

4.3.2 S. sonnei pLux biosensor

S. sonnei pLux produced an EC50 value of 0.01 mg/l for all six heavy metal compounds (Table 4.3). With increasing time, in the presence of all heavy metal compounds, S. sonnei pLux exhibited a gradual decrease in percentage bioluminescence (Figs 4.3, 4.19 - 4.23). The relative sensitivity of S. sonnei pLux to the six heavy metal compounds was: lead acetate > zinc sulphate > copper sulphate > chromium trioxide > cadmium acetate > potassium dichromate. This implied that the biosensor was most sensitive to lead acetate and least sensitive to potassium dichromate. In the presence of 0.01 - 0.1 mg/l concentrations of lead acetate, a gradual decrease in percentage bioluminescence was observed (Fig. 4.19). However, for zinc sulphate (Fig. 4.20) and copper sulphate (Fig. 4.21), S. sonnei pLux exhibited a higher percentage bioluminescence at 0.1 mg/l than at 0.01 mg/l.

- 53 - As the concentration of lead acetate, zinc sulphate and copper sulphate (Figs 4.19 - 4.20) increased from 1 - 100 mg/l, S. sonnei pLux exhibited the most marked decrease in bioluminescence. For 1 mg/l and 10 - 100 mg/l concentrations of lead acetate, zinc sulphate and copper sulphate, there was at least 88% and 99% decrease in bioluminescence, respectively. At 10 - 100 mg/l of both chromium trioxide (Fig. 4.3) and cadmium acetate (Fig. 4.22) the biosensor exhibited an overall decrease in percentage bioluminescence of at least 99%. For potassium dichromate (Fig. 4.23), this biosensor exhibited higher percentage bioluminescence at 0.1 mg/l and 1 mg/l, than at 0 mg/l. However, when tested in 10 - 100 mg/l of potassium dichromate (Fig. 4.23), S. sonnei pLux followed the general trend as seen for lead acetate, zinc sulphate, copper sulphate, chromium trioxide and cadmium acetate and exhibited a 72 - 99% decrease in bioluminescence.

S. sonnei pLux produced an EC50 value of 0.01 mg/l for BTEX (Table 4.4). The relative sensitivity of S. sonnei pLux to BTEX was: xylene > ethylbenzene > toluene > benzene (Figs 4.11; 4.24 - 4.26). This implied that the biosensor was most sensitive to xylene (Fig. 4.11) and least sensitive to benzene (Fig. 4.26). After 30 - 45 min, at 10 - 100 mg/l BTEX, a slight increase in percentage bioluminescence was observed. Although this biosensor exhibited a decrease in percentage bioluminescence at 1 mg/l of BTEX, the percentage bioluminescence was higher than that of 0 mg/l. S. sonnei pLux displayed a similar decrease in percentage bioluminescence at 0 mg/l and 1 mg/l of BTEX (Figs 4.11, 4.24 - 4.26).

120

100 )

80 0 mg/l 60 0.01 mg/l 0.1 mg/l 40 1 mg/l

Bioluminescence (% 10 mg/l 20 100 mg/l

0 1 15304560 Time (min)

Fig. 4.19 Bioluminescent response of the S. sonnei pLux biosensor in the presence of various concentrations of lead acetate.

- 54 - 120

100 0 mg/l 80 0.01 mg/l 0.1 mg/l 60 1 mg/l 40 10 mg/l 100 mg/l

Bioluminescence (%) Bioluminescence 20

0 1 15304560 Time (min)

Fig. 4.20 Bioluminescent response of the S. sonnei pLux biosensor in the presence of various concentrations of zinc sulphate.

120

100 0 mg/l 80 0.01 mg/l 0.1 mg/l 60 1 mg/l 40 10 mg/l 100 mg/l

Bioluminescence (%) Bioluminescence 20

0 1 15304560 Time (min)

Fig. 4.21 Bioluminescent response of the S. sonnei pLux biosensor in the presence of various concentrations of copper sulphate.

120

100 0 mg/l 80 0.01 mg/l 0.1 mg/l 60 1 mg/l 40 10 mg/l 100 mg/l

Bioluminescence (%) Bioluminescence 20

0 115304560 Time (min)

Fig. 4.22 Bioluminescent response of the S. sonnei pLux biosensor in the presence of various concentrations of cadmium acetate.

- 55 - 120

100 0 mg/l 80 0.01 mg/l 0.1 mg/l 60 1 mg/l 40 10 mg/l 100 mg/l

Bioluminescence (%) Bioluminescence 20

0 115304560 Time (min)

Fig. 4.23 Bioluminescent response of the S. sonnei pLux biosensor in the presence of various concentrations of potassium chromate.

120 100 0 mg/l 80 0.01 mg/l 0.1 mg/l 60 1 mg/l

40 10 mg/l 100 mg/l

Bioluminescence (%) 20 0 1 15304560 Time (min)

Fig. 4.24 Bioluminescent response of the S. sonnei pLux biosensor in the presence of various concentration of ethylbenzene.

120

100 0 mg/l 80 0.01 mg/l 0.1 mg/l 60 1 mg/l 40 10 mg/l 100 mg/l Bioluminescence (%) 20 0 1 15304560 Time (min)

Fig. 4.25 Bioluminescent response of the S. sonnei pLux biosensor in the presence of various concentrations of toluene.

- 56 -

120 100 0 mg/l 80 0.01 mg/l 0.1 mg/l

60 1 mg/l

40 10 mg/l 100 mg/l

Bioluminescence (%) Bioluminescence 20

1 15304560 Time (min)

Fig. 4.26 Bioluminescent response of the S. sonnei pLux biosensor in the presence of various concentrations of benzene.

4.3.3 E. coli DH5α uspA: lux biosensor

E. coli DH5α uspA: lux produced EC50 values of 1 mg/l for copper sulphate, zinc sulphate, cadmium acetate and chromium trioxide; 100 mg/l for potassium dichromate; and 10 mg/l for lead acetate (Table 4.3). The relative sensitivity of E. coli DH5α uspA:lux the six heavy metal compounds was: copper sulphate > chromium trioxide > cadmium acetate > zinc sulphate > lead acetate > potassium dichromate (Figs 4.8, Figs 4.27 - 4.31). This implied that the biosensor was most sensitive to copper sulphate (Fig. 4.27) and least sensitive to potassium dichromate (Fig. 4.31). At 1 mg/l copper sulphate, chromium trioxide and cadmium acetate (Figs 4.9, 4.27 – 4.28), overall decreases in bioluminescence of 80%, 20% and 32% were observed, respectively. The biosensor exhibited a 93% decrease in bioluminescence for 10 - 100 mg/l zinc sulphate (Fig. 4.29). At 100 mg/l lead acetate and potassium dichromate (Figs 4.30 - 4.31), at least 95% decrease in bioluminescence was observed. At 0.01 - 0.1 mg/l of all heavy metal compounds (Figs 4.27 - 4.31), E. coli DH5α uspA: lux continued to exhibit a gradual increase in percentage bioluminescence that remained less than the percentage bioluminescence at 0 mg/l.

- 57 - 450 400 350 0 mg/l 300 0.01 mg/l 250 0.1 mg/l 200 1 mg/l 150 10 mg/l 100 100 mg/l Bioluminescence (%) Bioluminescence 50 0 1 15304560 Time (min)

Fig. 4.27 Bioluminescent response of the E. coli DH5α uspA: lux biosensor in the presence of various concentrations of copper sulphate.

450 400 350 0 mg/l 300 0.01 mg/l 250 0.1 mg/l 200 1 mg/l 150 10 mg/l 100 mg/l 100 Bioluminescence (%) 50 0 1 15304560 Time (min)

Fig. 4.28 Bioluminescent response of the E. coli DH5α uspA: lux biosensor in the presence of various concentrations of cadmium acetate.

400 350 300 0 mg/l 0.01 mg/l 250 0.1 mg/l 200 1 mg/l 150 10 mg/l 100 100 mg/l Bioluminescence (%) 50 0 1 15304560 Time (min)

Fig. 4.29 Bioluminescent response of the E. coli DH5α uspA: lux biosensor in the presence of various concentrations of zinc sulphate.

- 58 - 450 400 350 0 mg/l 300 0.01 mg/l 250 0.1 mg/l 200 1 mg/l 150 10 mg/l 100 100 mg/l Bioluminescence (%) 50 0 115304560 Time (min)

Fig. 4.30 Bioluminescent response of the E. coli DH5α uspA: lux biosensor in the presence of various concentrations of lead acetate.

450 400 0 mg/l 350 0.01 mg/l 300 0.1 mg/l 250 1 mg/l 200 150 10 mg/l 100 100 mg/l

Bioluminescence (%) 50 0 1 15304560 Time (min)

Fig. 4.31 Bioluminescent response of the E. coli DH5α uspA: lux biosensor in the presence of various concentrations of potassium dichromate.

E. coli DH5α USP:lux achieved *EC100 values of 100 mg/l for benzene and toluene; and 10 mg/l for ethylbenzene and xylene (Table 4.6). The relative sensitivity of E. coli DH5α uspA: lux to BTEX was: xylene > ethylbenzene > toluene > benzene (Figs 4.16; Figs 4.32-4.34). This implied that this biosensor was most sensitive to xylene (Fig. 4.16) and least sensitive to benzene (Fig. 4.34). With increasing time, for all concentrations of BTEX the biosensor experienced a gradual increase in percentage bioluminescence. However, at the 0.01 – 1 mg/l concentrations of BTEX the percentage bioluminescence experienced by biosensor was less than that at 0 mg/l. At 100 mg/l and 10 mg/l benzene (Fig. 4.34) the biosensor experienced 687% and 581% bioluminescence, respectively. At 100 mg.l and 10 mg/l toluene (Fig. 4.32) the biosensor experienced 596% and 486% bioluminescence, respectively. For xylene and ethylbenzene (Figs 4.16 – 4.33) the biosensor showed the highest percentage bioluminescence at 10 mg/l. At 100 mg/l xylene and ethylbenzene (Figs 4.16 – 4.33) the biosensor experienced at least 483% bioluminescence.

- 59 -

700

600 0 mg/l 500 0.01 mg/l 400 0.1 mg/l

(%) 300 1 mg/l

200 10 mg/l 100 100 mg/l 0

Bioluminescence 1 15304560 Time (min)

Fig. 4.32 Bioluminescent response of the E. coli DH5α uspA: lux biosensor in the presence of various concentrations of toluene.

700

600 0 mg/l 500 0.01 mg/l 400 0.1 mg/l 1 mg/l 300 10 mg/l 200 100 mg/l

Bioluminescence (%) 100 0 1 15304560 Time (min)

Fig. 4.33 Bioluminescent response of the E. coli DH5α uspA: lux biosensor in the presence of various concentrations of ethylbenzene.

800 700 0 mg/l 600 0.01 mg/l 500 0.1 mg/l

400 1 mg/l 300 10 mg/l

200 100 mg/l

Bioluminescence (%) Bioluminescence 100 0 115304560 Time (min)

Fig. 4.34 Bioluminescent response of the E. coli DH5α uspA: lux biosensor in the presence of various concentrations of benzene.

- 60 - 4.3.4 Vibrio fischeri-based BioToxTM kit The sensitivity of the V. fischeri-based BioToxTM kit to the six heavy metal compounds was: cadmium acetate > chromium trioxide > lead acetate > copper sulphate > potassium dichromate > zinc sulphate (Figs 4.35– 4.39). This implied that V. fischeri most sensitive to cadmium acetate (Fig. 4.35) and least sensitive to zinc sulphate (Fig. 4.39). With increasing time, in the presence of all heavy metal compounds, V. fischeri exhibited a gradual decrease in percentage bioluminescence. For all metals, except 0.01 mg/l chromium trioxide (Fig. 4.10), 0.01 mg/l copper sulphate (Fig. 4.37) and 0.1 mg/l zinc sulphate (Fig. 4.39) V. fischeri exhibited highest% bioluminescence at 0 mg/l. For all heavy metal compounds V. fischeri exhibited greater than 89% decrease in bioluminescence at 100 mg/l. At 10 mg/l cadmium acetate (Fig. 4.35) and copper sulphate (Fig. 4.37), V. fischeri experienced greater than 81% decrease in bioluminescence. In the presence of 10 mg/l chromium trioxide (Fig. 4.10), lead acetate (Fig. 4.36), potassium dichromate and zinc sulphate (Fig. 4.39), V. fischeri experienced greater than 79%, 59%, 40% and 30% decrease in bioluminescence,

respectively. The V. fischeri-based BioTox kit achieved EC50 values of 1 mg/l for lead acetate and cadmium acetate; 10 mg/l for copper sulphate and chromium trioxide; and 100 mg/l for zinc sulphate and potassium dichromate (Table 4.3). The relative sensitivity of the V. fischeri-based BioTox kit to BTEX was: xylene > ethylbenzene > toluene > benzene (Figs 4.18; 4.40 – 4.42). This implied that this biosensor was most sensitive to xylene (Fig. 4.18) and least sensitive to benzene (Fig. 4.42). V. fischeri experienced a gradual decrease in % bioluminescence at 0.01 - 1 mg/l BTEX. For 100 mg/l BTEX there was a gradual decrease in V. fischeri percentage bioluminescence that remained below the percentage bioluminescence at 0 mg/l. During the first 30 min in the presence of BTEX V. fischeri experienced a gradual increase in percentage bioluminescence. For the last 15 min, in the presence of BTEX, V. fischeri experienced a gradual decrease in percentage bioluminescence that remained greater than that at 0 mg/l (Figs 4.18; 4.40 – 4.42).

- 61 -

120

100

60 0 mg/l 0.01 mg/l 40 0.1 mg/l 1 mg/l

Bioluminescence (%) 20 10 mg/l 0 100 mg/l 1 15304560 Time (min)

Fig. 4.35 Bioluminescent response of V. fischeri in the presence of various concentrations of cadmium acetate.

120

100 0 mg/l 80 0.01 mg/l 0.1 mg/l 60 1 mg/l 40 10 mg/l 100 mg/l

Bioluminescence (%) 20

0 1 15304560 Time (min)

Fig. 4.36 Bioluminescent response of the V. fischeri in the presence of various concentrations of lead acetate.

120

100 0 mg/l 80 0.01 mg/l 0.1 mg/l 60 1 mg/l 40 10 mg/l 100 mg/l

Bioluminescence (%) 20

0 1 15304560 Time (min)

Fig. 4.37 Bioluminescent response of V. fischeri in the presence of various concentrations of copper sulphate.

- 62 - 120

100 0 mg/l

80 0.01 mg/l 0.1 mg/l 60 1 mg/l 40 10 mg/l 20 100 mg/l Bioluminescence (%) Bioluminescence 0 1 15304560 Time (min)

Fig. 4.38 Bioluminescent response of V. fischeri in the presence of various concentrations of potassium dichromate.

120 100 0 mg/l

80 0.01 mg/l

0.1 mg/l 60 1 mg/l

40 10 mg/l

100 mg/l

Bioluminescence (%) 20

0 1 15304560 Time (min)

Fig. 4.39 Bioluminescent response of V. fischeri in the presence of various concentrations of zinc sulphate.

120 100 0 mg/l 80 0.01 mg/l 0.1 mg/l 60 1 mg/l 40 10 mg/l 100 mg/l

Bioluminescence (%) 20

1 15304560

Time (min)

Fig. 4.40 Bioluminescent response of V. fischeri in the presence of various concentrations of ethylbenzene

- 63 -

120

100 0 mg/l 80 0.01 mg/l 0.1 mg/l 60 1 mg/l 40 10 mg/l 100 mg/l Bioluminescence (%) 20 0 1 15304560 Time (min)

Fig. 4.41 Bioluminescent response of V. fischeri in the presence of various concentrations of toluene

120

100 0 mg/l 80 0.01 mg/l 0.1 mg/l 60 1 mg/l 40 10 mg/l 100 mg/l

Bioluminescence (%) 20

0 1 15304560

Time (min)

Fig. 4.42 Bioluminescent response of V. fischeri in the presence of various concentrations of benzene.

4.3.5 Daphnia LC50 toxicity test

In the Daphnia toxicity test, the LC50 value represented the lethal concentration which resulted in 50% death of the Daphnia pulex test organisms. The relative sensitivity of

the Daphnia LC50 toxicity test to the six heavy metal compounds was: chromium trioxide > cadmium acetate > copper sulphate > potassium dichromate > zinc sulphate > lead acetate. This implied that Daphnia pulex was most sensitive to chromium trioxide and least sensitive to lead acetate.

The sensitivity of the Daphnia LC50 toxicity test followed the trend: xylene > ethylbenzene > toluene > benzene. This implied that these toxicity tests were most sensitive to xylene and least sensitive to benzene (Table 4.7).

- 64 - Table 4.7 Daphnia LC50 toxicity test values obtained by the Umgeni Water Microbiology Laboratory after 48 h, using Daphnia pulex as the eukaryotic test organism

Toxicant Daphnia LC50 value (mg/l) Copper sulphate 0.04 Zinc sulphate 0.20 Potassium dichromate 0.17 Lead acetate 0.79 Cadmium acetate 0.03 Chromium trioxide 0.02 Benzene 55.80 Toluene 8.20 Ethylbenzene 3.49 Xylene 1.18

4.4 DISCUSSION There was a general decrease in percentage bioluminescence of all five biosensors at high concentrations of 10 - 100 mg/l, for the six heavy metal compounds. At 10 mg/l copper sulphate, E. coli DH5α recA:lux, E. coli DH5α fabA:lux, E. coli DH5α uspA:lux, S. sonnei pLux and the V. fischeri-based BioToxTM kit all exhibited greater than 94% decreases in bioluminescence. Since all these luxCDABE-marked bacteria are sensitive to less than 30 mg/l copper sulphate (Table 2.1), they are effective copper- toxicity determinants. Weitz et al. (2001) and Paton et al. (1995) also reported a high sensitivity of luxCDABE-containing bacterial constructs to copper (II). They found that P. fluorescens 8866 Tn5 luxCDABE, P. putida F1 Tn5 luxCDABE, P. fluorescens

10586s/FAC510 and P. fluorescens 1058/pUCD607 produced EC50 values of 0.3 mg/l, 0.17 mg/l, 0.76 mg/l and 0.09 mg/l for copper (II), respectively. However, S. sonnei pLux was more sensitive than these bioluminescent bacterial constructs as it produced

an EC50 value of 0.01 mg/l (Table 4.3) for copper sulphate [which contains copper (II) as shown in Table 4.1]. This implies that S. sonnei pLux could be useful in the detection of copper (II) in polluted water samples. At 10 mg/l zinc sulphate all five biosensors exhibited greater than 84% decreases in bioluminescence. At 10 mg/l zinc sulphate the V. fischeri-based BioToxTM kit exhibited a significant decrease in bioluminescence of 99%. At 1 mg/l zinc sulphate the E. coli DH5α recA:lux and S. sonnei pLux biosensors exhibited greater than 85% decreases in bioluminescence. Since all these luxCDABE-marked bacteria are sensitive

- 65 - to less than 50 mg/l zinc sulphate (Table 2.1) they are effective zinc-toxicity determinants. In the presence of zinc(II), P. fluorescens 8866 Tn5 luxCDABE, P. putida F1 Tn5 luxCDABE, P. fluorescens 10586s/FAC510 and P. fluorescens

1058/pUCD607 have been found to produce EC50 values of 0.1 mg/l, 0.04 mg/l, 0.89 mg/l and 0.09 mg/l, respectively (Weitz et al., 2001; Paton et al., 1995). However, S. sonnei pLux was much more sensitive than these bioluminescence bacterial constructs as it produced an EC50 value of 0.01 mg/l (Table 4.3) for zinc sulphate [which contains zinc(II) as depicted in Table 4.1]. At 10 mg/l potassium dichromate E. coli DH5α pLux and S. sonnei pLux exhibited greater than 74% decreases in bioluminescence. Since all these luxCDABE-marked bacteria are sensitive to less than 50 mg/l potassium dichromate (Table 2.1), they are effective chromium-toxicity determinants. In the presence of chromium (III), Paton et al. (1995) found that P. fluorescens 10586s/FAC510 and P. fluorescens

1058/pUCD607 produced EC50 values of 1.28 mg/l and 1.46 mg/l, respectively. However, S. sonnei pLux was much more sensitive than these bioluminescence bacterial constructs as it produced an EC50 value of 0.01 mg/l (Table 4.3) for potassium dichromate [which contains chromium (III) as depicted in Table 4.1]. At 10 mg/l lead acetate only E. coli DH5α pLux and S. sonnei pLux exhibited greater than 74% decreases in bioluminescence. At 1 mg/l lead acetate, S. sonnei pLux exhibited a greater than 93% decrease in bioluminescence. Since both these luxCDABE-marked bacteria are sensitive to 10 mg/l lead acetate (Table 2.1), they are effective lead- toxicity determinants. Only S. sonnei pLux was sensitive enough to detect less than 0.02 mg/l cadmium acetate (Table 2.1), and be an effective cadmium-toxicity determinant. In the presence of cadmium (II), Paton et al. (1995) found that P. fluorescens 10586s/FAC510 and P. fluorescens 1058/pUCD607 produced EC50 values of 0.98 mg/l and 0.17 mg/l, respectively. However, S. sonnei pLux was more sensitive than these bioluminescent bacterial constructs as it produced an EC50 value of 0.01 mg/l (Table 4.3) for cadmium acetate [which contains cadmium (II) as depicted in Table 4.1]. Nies (1999) best explained the exact mechanism of heavy metal toxicity to bacteria. According to his research, once inside the bacterial cell, the heavy metal cations tend to bind to sulphur-hydrogen (SH) groups. By binding to SH groups, the metal ions may inhibit the activity of sensitive enzymes. Heavy metal cations can also bind to glutathione in gram negative bacteria (e.g. E. coli and S. sonnei luxCDABE-based biosensors in Table 3.5). The resulting bisglutathionato complexes tend to react with molecular oxygen and form oxidized bisglutathione, the metal cations and hydrogen peroxide. Heavy metal ions cause considerable stress since the oxidised bisglutathione has to be reduced again and the metal cations have to immediately bind to another two glutathione molecules. Since percentage bioluminescence is directly proportional to

- 66 - metabolic activity, this implied that all the biosensors (Table 3.5) exhibited physiological and/or toxic effects in the presence of the heavy metal compounds. These physiological effects may include DNA damage (genotoxicity), membrane damage and general protein damage, since these are the stresses that the pRecALux, pFabALux and pUspALux2 plasmids (Table 3.3) found in E. coli DH5α recA: lux, E. coli DH5α fabA: lux and E. coli DH5α uspA: lux, respectively, responded to. A decrease in the percentage bioluminescence of the E. coli DH5α pLux and S. sonnei pLux biosensors also indicated general protein damage in the presence of all heavy metal compounds. However, at the lower heavy metal concentrations of 0.01 - 0.1 mg/l certain biosensors exhibited higher percentage bioluminescence than that at 0 mg/l. The S. sonnei pLux biosensor exhibited higher percentage bioluminescence for 0.01 - 0.1 mg/l chromium trioxide and cadmium acetate than at 0 mg/l. The E. coli DH5α pLux biosensor exhibited higher percentage bioluminescence at the 0.1 mg/l than the 0 mg/l for all six heavy metals. The E. coli DH5α recA: lux biosensor exhibited higher percentage bioluminescence at 0.01 - 0.1 mg/l copper sulphate than at 0 mg/l. The E. coli DH5α fabA: lux biosensor exhibited greater percentage bioluminescence at 0.1 mg/l lead acetate and potassium dichromate than at 0 mg/l. The V. fischeri-based BioToxTM kit exhibited higher percentage bioluminescence at 0.01 mg/l chromium trioxide, 0.1 mg/l zinc sulphate and 0.01 mg/l copper sulphate than at 0 mg/l. All the E. coli DH5α biosensors (Table 3.5) exhibited bioluminescence induction at 0.01 - 0.1 mg/l copper sulphate. The E. coli DH5α recA: lux, E. coli DH5α fabA: lux and E. coli DH5α uspA: lux biosensors exhibited bioluminescence induction at 0.01 - 0.1 mg/l zinc sulphate. According to Nies (1999), this can be attributed to complexation of the heavy metal compounds, which is a metal-ion homeostasis mechanism evolved by bacteria to detoxify heavy metals at low concentration in the cells cytoplasm. Complexation is extremely important since heavy metals cannot be degraded like toxic organic compounds (e.g. BTEX). The decreased toxicity at lower concentrations of zinc sulphate, cadmium acetate and lead acetate may also be due to the ZntA P-type ATPase transport pump in E. coli, which may have extruded these heavy metal ions to the cytosol (Gatti et al., 2000). It also appears that E. coli DH5α recA: lux, E. coli DH5α fabA: lux, E. coli DH5α uspA: lux exhibited the physiological effects of DNA damage (genotoxicity), membrane damage and general protein damage at lower heavy metal concentrations, as indicated by an induction of the bioluminescent response. However, at high metal concentrations there was a decrease in percentage bioluminescence of all biosensors (Table 3.5) since the physiological damage may have been too severe.

Based from the EC50 values (Table 4.3) and the Daphnia LC50 values (Table 4.7), the relative sensitivities of the biosensors (Table 3.5), the V. fischeri-based BioToxTM kit and the Daphnia LC50 toxicity test to the six heavy metal compounds were: S. sonnei

- 67 - pLux > S. flexneri pLux > EPEC pLux > Daphnia LC50 toxicity test > E. coli DH5α recA: lux > E. coli DH5α fabA: lux > E. coli DH5α pLux > E. coli DH5α uspA: lux > V. fischeri-based BioToxTM kit. Therefore, the biosensor most sensitive to heavy metal compounds was S. sonnei pLux. Although the Daphnia LC50 toxicity test was also sensitive, the S. sonnei pLux biosensor was able to detect heavy metal and chemical pollutants more rapidly, i.e., Daphnia LC50 toxicity test results were only available after 48 h and the S. sonnei pLux bioluminescent results were available after 1 h. The V. fischeri-based BioToxTM kit was least sensitive, possibly since it is a marine bacterium and unlike the biosensors constructed in this research required a saline environment for all toxicity tests. The sensitivity of E. coli DH5α pLux, E. coli DH5α fabA: lux, S. sonnei pLux, the V. TM fischeri-based BioTox kit and the Daphnia LC50 toxicity test all followed the same trend for the BTEX compounds and were least sensitive to benzene and most sensitive to xylene. According to Cole (1994), decreased toxicity of organic solvents can be attributed to low molecular weight, low viscosity and high solubility in water; while increased toxicity of organic solvents can be attributed to high molecular weight, high viscosity and low solubility in water. Benzene was least toxic since it had the lowest molecular weight and viscosity, and the highest solubility in water, as compared to BTEX. Xylene was most toxic since it had the highest molecular weight and viscosity, and lowest solubility in water, as compared to BTEX (Table 4.2). At 100 mg/l and 10 mg/l of the BTEX compounds, E. coli DH5α pLux, E. coli DH5α recA: lux, E. coli DH5α fabA: lux, E. coli DH5α uspA: lux and the V. fischeri-based BioToxTM kit exhibited an induction in percentage bioluminescence. However, only S. sonnei pLux was sensitive enough to achieve both EC50 values and detect less than 5 mg/l benzene, 1 mg/l toluene, 0.7 mg/l ethylbenzene and 10 mg/l xylene (Table 4.4). Thus, only S. sonnei pLux appeared to be an effective reporter of BTEX toxicity. The results of this chapter demonstrated that the E. coli DH5α pLux, E. coli DH5α recA: lux, E. coli DH5α fabA: lux, E. coli DH5α uspA: lux, S. sonnei pLux, EPEC pLux and S. sonnei pLux biosensors were able to detect the toxicity of heavy metal compounds and BTEX rapidly and reproducibly. However, further characterisation was required to determine the advantages of using this range of biosensors to assess their applicability in water toxicity testing. Therefore, the standard bioluminescence profiles (Figs 4.3 - 4.42) obtained for the freeze-dried biosensors in the presence of heavy metal compounds and BTEX could be used to evaluate their bioluminescence profiles in the presence of unknown toxic water samples. Thus, these freeze-dried biosensors offered potential versatility and environmental relevance for the testing of wastewater effluent samples described in Chapter Five.

- 68 - CHAPTER FIVE: APPLICATION OF THE PROKARYOTIC BIOSENSOR SYSTEMS

5.1 INTRODUCTION Advances in industrial technology, has led to increasing levels of diverse pollutants in the environment. As public awareness on environmental issues has recently increased, much focus has been on detecting potentially hazardous pollutants in the environment. Industrialisation and new technologies have not only made life more convenient for humans, but also have created various environmental problems, potentially posing serious health problems to living organisms. Most heavy metals originating from industrial production bioaccumulate in the aquatic food chain to the extent that the consumption of fish by birds and animals in certain areas pose a serious health threat. Their most distinguishing feature is that they are not biodegradable. Each year, several thousands of new chemicals are generated by human activities. These include the organochlorine pesticides such as lindane, endrin, dichlorodiphenyltrichloroethane (DDT), heptachlor, heptachlor epoxide, aldrin and dieldrin. These culprit organics are known to have toxic, mutagenic and carcinogenic activities such as intrauterine growth retardation, neurological developmental disorders, foetal death, and central nervous system damage (EPUSA, 2003). Organic compounds, originating from the widespread use of petroleum products (e.g. BTEX) are highly toxic and cause concern for soil and drinking water quality (Hansen and Sorenson, 2001). Strict legislation warns against BTEX compounds in wastewater treatment works (WWTW). According to EThekwini Wastewater (2003), no trade influent received by WWTW should contain volatile flammable solvents, or any substance that yields a flammable vapour at 21˚C (e.g. BTEX). However, certain large industries still discharge BTEX compounds into WWTW. Wastewater workers can be exposed to these toxic gaseous chemicals and experience abnormal health effects as described previously in Table 2.1. In response to heavy metal and BTEX pollutants, many chemical and physical methodologies were developed along with the required analytical equipment (Kohler et al., 2000). The monitoring of environmental pollutants involves techniques such as spectroscopy (UV, visible, IR), high-performance liquid chromatography, gas chromatography (GC), GC-mass spectrometry (MS) and atomic emission and absorption techniques for determination of metals (flame atomic absorption spectroscopy, inductively coupled plasma atomic emission spectroscopy). However, these powerful, accurate and sensitive techniques are also costly and require specialised laboratories. In addition, they fail to provide data on the bioavailability and persistence

- 69 - of the pollutant in the environment, its effects on living organisms or its potential synergistic/antagonistic behaviour in mixtures (Sousa et al., 1998). For wastewater characterisation, conventional chemical-based methods alone have been found to be inadequate for ensuring that the effluent has no significant effects on the receiving works performance. Consequently, it has been necessary to develop toxicity tests that can give a more selective indication of the potential effects of effluent discharges on a sewage works. Many types of bioassays are available which include the fish and water flea (Daphnia pulex) toxicity tests. These tests are time consuming and the use of higher organisms such as fish as a test species may also be ethically undesirable (Farre et al., 2001). Commercially available kits such as Microtox™ and BioTox™, which adopt the luminous marine bacterium V. fischeri, are frequently used today. These bioassays, rather than measuring long-term effects on viability, determine short-term (normally fifteen minutes) effects on light production (Belkin et al., 1997). One of the limitations of these assays, however, is that it is not ecologically representative of freshwater systems since the saltwater bacterium V. fischeri is used (Strachan et al., 2001). One way to resolve this is to use another bioluminescent-based assay that is of low cost, which is able to cope with large sample runs and yet remain environmentally relevant. The development of the E. coli DH5α pLux, EPEC pLux, S. flexneri pLux, E. coli DH5α recA: lux, E. coli DH5α fabA: lux, E. coli DH5α uspA: lux and S. sonnei pLux biosensors are a relevant approach. An understanding of the acute toxicity imposed by combinations of pollutants is essential in interpreting the fate of environmental contaminants (Strachan et al., 2001). Thus, in this chapter, these five biosensor systems (Table 3.5) were used to test relevant WWTW effluent samples from KwaZulu - Natal containing possible mixtures of pollutants.

5.2 MATERIALS AND METHODS 5.2.1 Collection of wastewater effluent samples New Germany wastewater effluent (NGWWE), Kwa-Mashu wastewater effluent (KMWWE) Northern wastewater effluent (NWWE), Phoenix wastewater effluent (PWWE) and Amanzimtoti wastewater effluent (AWWE) samples (Fig. 5.1) were collected in duplicate. All wastewater effluent samples were collected in sterile Schott bottles before the chlorination process (Fig. 2.3).

- 70 -

Fig. 5.1 Geographical locations (●) of the five wastewater treatment works investigated in this study.

5.2.2 Analysis of wastewater effluent samples Samples were stored at 4 ºC prior to use and analysed within 48 h. One set of the duplicate samples was sent to the Umgeni Water Analytical Services Department for both heavy metal and chemical analysis. One hundred percent, 10%, 1%, 0.1% and 0.01% (v/v) of the other set of duplicate samples were subjected to the toxicity test outlined in Section 3.2.2. The 0% sample contained 270 µl of sterile deionised water without any wastewater effluent sample.

5.2.3 Calculation of bioluminescence and EC values Since all bioluminescence values were taken as the average of triplicate samples,

- 71 - Bioluminescence (%) = Average Bioluminescence at x min at yx % 100 Average Initial Bioluminescence Where: Initial bioluminescence = Bioluminescence at time 0 min at 0 mg/l; x = either 1, 15, 30, 45 or 60 min; and y = either 0.01, 0.1, 1, 10 or 100% of wastewater effluent sample.

The bioluminescence patterns of E. coli DH5α pLux, EPEC pLux, S. flexneri pLux, E. coli DH5α recA: lux, E. coli DH5α fabA: lux, E. coli DH5α uspA: lux and S. sonnei

.EC100 values٭ EC20, and٭ ,pLux were determined using EC50

EC50 – This value was equivalent to the lowest percentage of wastewater effluent sample that resulted in 50% bioluminescence inhibition, as compared to the initial percentage bioluminescence.

EC20 – This value was equivalent to the percentage of wastewater effluent sample that٭ first resulted in 20% bioluminescence induction, as compared to the initial percentage bioluminescence.

EC100 – This value was equivalent to the percentage of wastewater effluent sample٭ that first resulted in 100% bioluminescence induction, as compared to the initial percentage bioluminescence.

5.3 RESULTS The toxicity responses of E. coli DH5α pLux, EPEC pLux, S. flexneri pLux, E. coli DH5α recA: lux, E. coli DH5α fabA: lux, E. coli DH5α uspA: lux, S. sonnei pLux and the V. fischeri-based BioTox kit, to wastewater effluent samples, were represented as

EC100 values and summarized in Tables 5.1 – 5.3. Wastewater٭ EC20 and٭ ,EC50 effluent samples were also analysed by Umgeni Water Analytical Services Department for the presence and concentrations of heavy metal and BTEX compounds (Table 5.4). The Standard deviations and average percentage bioluminescence values for Figs 5.2 – 5.16 were calculated using the Fluoroskan Ascent FL Software, and are tabulated in the Appendix One.

- 72 - TM Table 5.1 EC50 values of S. sonnei pLux and the V. fisheri – based Biotox kit for the five wastewater effluent samples

luxCDABE- EC50 values for the various samples (%) marked bacteria NGWWE KMWWE NWWE PWWE AWWE S. sonnei 0.01 0.01 0.01 0.01 0.01 (15-30 min) (30 min) (1-15 min) (1-15 min) (1-15 min) S. flexneri 100 100 0.01 0.01 0.01 (> 60 min) (> 60 min) (1-15 min) (15-30 min) (45-60 min) EPEC 0.01 0.01 100 0.01 0.01 (30-45 min) (30-45 min) (> 60 min) (30-45 min) (45-60 min) V. fischeri- 100 100 based none (45-60 min) none none (60 min) BioToxTM kit The value in parenthesis represents the time during which the EC50 value was obtained.

Table 5.2 *EC20 values of E. coli DH5D pLux for the five wastewater effluent samples

luxCDABE- *EC20values for the various samples (%) marked biosensors NGWWE KMWWE NWWE PWWE AWWE E. coli 10 10 10 100 DH5α pLux (15 min) none (1-15 min) (15 min) (1-15 min)

*The value in parenthesis represents the time during which the EC20value was obtained.

Table 5.3 *EC100 values of E. coli DH5D lux CDABE – based biosensors for the five wastewater effluent samples

luxCDABE- marked *EC100 values for the various samples (%) biosensors NGWWE KMWWE NWWE PWWE AWWE E. coli DH5α recA:lux 10 1 10 1 10 (15-30 min) (15-30 min) (15-30 min) (15-30 min) (15-30 min)

E. coli DH5α fabA:lux 10 1 10 1 10 (30-45 min) (30-45 min) (30-45 min) (30-45 min) (15-30 min)

E. coli DH5α uspA:lux 10 0.1 1 1 1 (30-45 min) (30-45 min) (15-30 min) (15-30 min) (30-45 min) The value in parenthesis represents the time during which the EC100 value was obtained.

- 73 - Table 5.4 Analyses performed by Umgeni Water Analytical Services Department for the five wastewater effluent samples

Determinant Samples (100%) NGWWE KMWWE NWWE PWWE AWWE Chromium 0.02 < 0.003 < 0.003 < 0.003 0.0031 (mg/l) Cadmium < 0.001 < 0.001 < 0.001 < 0.001 < 0.001 (mg/l) Copper < 0.05 < 0.05 < 0.05 < 0.05 < 0.05 (mg/l) Lead 0.033 0.0046 < 0.004 < 0.004 0.0042 (mg/l) Zinc 0.22 0.04 0.04 < 0.03 0.11 (mg/l) Total 1.387 < 0.02 0.256 < 0.02 < 0.02 Organochlorine pesticides (µg/l) Chemical Oxygen 67.4 27.4 28.2 51.9 73.5 Demand (mg O2/l) Daphnia LC50 non-toxic non-toxic non-toxic non-toxic non-toxic toxicity test Benzene < 0.2 < 0.2 < 0.2 < 0.2 < 0.2 (µg/l) Toluene < 0.08 0.34 < 0.08 < 0.08 < 0.08 (µg/l) Ethylbenzene < 0.03 0.1 < 0.03 < 0.03 < 0.03 (µg/l) Xylene < 0.15 0.35 < 0.15 < 0.15 < 0.15 (µg/l)

5.3.1 S. sonnei pLux biosensor With increasing time, S. sonnei pLux resulted in a gradual decrease in percentage bioluminescence at all concentrations of the NGWWE, KMWWE, NWWE, PWWE

and AWWE samples (Figs 5.2 – 5.6). An EC50 value of 0.01% was obtained for all samples (Table 5.1). The 0% sample resulted in higher percentage bioluminescence than all percentages of the NGWWE sample (Fig. 5.2). The 0% sample maintained a higher percentage bioluminescence than the 0.01%, 1% and 10% KMWWE samples. However, the 100% and 0.1% KMWWE samples resulted in higher percentage bioluminescence for the first 30 min and entire 60 min, respectively (Fig. 5.3). At 1 min and 15 min, the 1% and 10% NWWE samples resulted in the highest percentage

- 74 - bioluminescence. From 30 min to 60 min the 0% sample maintained the highest percentage bioluminescence response for the NWWE sample (Fig. 5.4). For the first 15 min the 1% PWWE sample resulted in the highest % bioluminescence. From 30 min to 60 min the 0% sample resulted in the highest percentage bioluminescence (Fig. 5.5). The 0% sample resulted in higher percentage bioluminescence than the 0.01%, 0.1% and 100% AWWE samples. The 1% AWWE sample resulted in the highest bioluminescence of 111% at 1 min, while the 10% AWWE sample resulted in the highest percentage bioluminescence from 15 – 45 min. The 0% sample only resulted in the highest percentage bioluminescence at 60 min (Fig. 5.6).

5.3.2 E. coli DH5α uspA: lux biosensor E. coli DH5α uspA: lux biosensor exhibited an induction in percentage bioluminescence at all concentrations of the five wastewater effluent samples, except the 100% KMWWE sample (Figs 5.7 – 5.11). The 0.1% NGWWE and 10% NGWWE samples resulted in the highest bioluminescence of 446% and 263% at 60 min and45 min, respectively (Fig. 5.7). At 15 min and 45 min the 0.1% KMWWE sample produced the highest bioluminescence of 165% and 255% (Fig. 5.8). At 60 min the 1% NGWWE sample resulted in the highest bioluminescence of 386% (Fig. 5.7). At 60 min the 10% NWWE and 1% PWWE samples resulted in the highest bioluminescence of 489% and 467%, respectively (Figs 5.9 – 5.10). At 1 min, 15 min and 60 min the 0.1% the AWWE sample resulted in the highest bioluminescence of 106%, 155% and 394%, respectively. At 30 min and 45 min the 0% and 1% AWWE samples resulted in the highest bioluminescence of 178% and 261%, respectively (Fig. 5.11). The NWWE, PWWE and AWWE samples produced an *EC100 value of 1%; while the NGWWE and KMWWE samples produced *EC100 values of 10% and 0.1%, respectively (Table 5.3).

120

100 0% 80 0.01% 0.10% 60 1% 40 10% 100%

Bioluminescence (%) Bioluminescence 20

0 1 15304560 Time (min)

Fig. 5.2 Bioluminescent response of the S. sonnei pLux biosensor in the presence of various concentrations of the New Germany wastewater effluent sample.

- 75 - 120

100 0% 80 0.01% 0.10% 60 1% 40 10% 100%

Bioluminescence (%) Bioluminescence 20

0 1 15304560 Time (min)

Fig. 5.3 Bioluminescent response of the S. sonnei pLux biosensor in the presence of various concentrations of the Kwa-Mashu wastewater effluent sample.

120

100

80 0% 60 0.01% 0.10% 40 1% 10%

Bioluminescence (%) 20 100% 0 1 15304560 Time (min)

Fig. 5.4 Bioluminescent response of the S. sonnei pLux biosensor in the presence of various concentrations of the Northern wastewater effluent sample.

120

100 0% 80 0.01% 0.10% 60 1% 40 10% 100%

Bioluminescence (%) Bioluminescence 20

0 1 15304560 Time (min)

Fig. 5.5 Bioluminescent response of the S. sonnei pLux biosensor in the presence of various concentrations of the Phoenix wastewater effluent sample.

- 76 - 120

100 0% 80 0.01% 0.10% 60 1% 40 10% 100%

Bioluminescence (%) Bioluminescence 20

0 1 15304560 Time (min)

Fig. 5.6 Bioluminescent response of the S. sonnei pLux biosensor in the presence of various concentrations of the Amanzimtoti wastewater effluent sample.

500 450 ) 400 0% 350 0.01% 300 0.10% 250 1% 200 150 10% 100 100% Bioluminescence (% Bioluminescence 50 0 1 15304560 Time (min)

Fig. 5.7 Bioluminescent response of the E. coli DH5α uspA: lux biosensor in the presence of various concentrations of the New Germany wastewater effluent sample.

450 400 350 0% 300 0.01% 250 0.10% 200 1% 150 10% 100 100% Bioluminescence (%) Bioluminescence 50 0 1 15304560 Time (min)

Fig. 5.8 Bioluminescent response of the E. coli DH5α uspA:lux biosensor in the presence of various concentrations of the Kwa-Mashu wastewater effluent sample.

- 77 - 500 450 400 0% 350 0.01% 300 0.10% 250 1% 200 150 10% 100 100% Bioluminescence (%) Bioluminescence 50 0 1 15304560 Time (min)

Fig. 5.9 Bioluminescent response of the E. coli DH5α uspA:lux biosensor in the presence of various concentrations of the Northern wastewater effluent sample

450 400 350 0% 300 0.01% 250 0.10% 200 1% 150 10% 100 100% Bioluminescence (%) Bioluminescence 50 0 1 15304560 Time (min)

Fig. 5.10 Bioluminescent response of the E. coli DH5α uspA:lux biosensor in the presence of various concentrations of the Phoenix wastewater effluent sample.

450 400 350 0% 300 0.01% 250 0.10% 200 1% 150 10% 100 100% Bioluminescence (%) 50 0 1 15304560 Time (min)

Fig. 5.11 Bioluminescent response of the E. coli DH5α fabA: lux biosensor in the presence of various concentrations of the Amanzimtoti wastewater effluent sample.

- 78 - 5.3.3 Vibrio fischeri-based BioToxTM kit V. fischeri experienced a gradual decrease in percentage bioluminescence at all concentrations of the five wastewater effluent samples, except the 100% NGWWE sample (Figs 5.12 – 5.16). The 10% NGWWE sample resulted in the highest percentage bioluminescence. Although the 100% sample resulted in the lowest percentage bioluminescence overall, it did result in a slight induction in % bioluminescence from 30 – 60 min. At 1 min the 0.01 - 10% NGWWE samples resulted in a slight percentage bioluminescence induction (Fig. 5.12). At 1 min the 100% KMWWE sample resulted in the highest induction in % bioluminescence. The 100% and 10% KMWWE sample resulted in the highest % bioluminescence for 1 – 15 min and 30 min, respectively. From 45 – 30 min the 0% sample resulted in the highest bioluminescence. From 30 - 60 min the 100% KMWWE sample resulted in the lowest % bioluminescence (Fig. 5.13). After 1 min the 100% NWWE sample resulted in the highest induction in percentage bioluminescence. The 100% and 10% NWWE sample maintained the highest percentage bioluminescence for 1 - 45 min and 45 – 60 min, respectively. However, the 10% NWWE sample maintained a higher percentage bioluminescence response than the 0% sample for the entire 60 min (Fig. 5.14). The 100% PWWE sample resulted in the highest percentage bioluminescence response for the first 45 min. At 60 min, the 10% PWWE sample resulted in the highest percentage bioluminescence. However, the 10% PWWE sample maintained a higher percentage bioluminescence response than the 0% sample for the entire 60 min. The 0.01 - 1% PWWE sample resulted in lower percentage bioluminescence than the 0% sample for the entire 60 min. The 0.01% PWWE sample resulted in the lowest percentage bioluminescence (Fig. 5.15). The 100% and 10% AWWE sample maintained the highest percentage bioluminescence for 1 – 15 min and 30 - 60 min, respectively. The 10% AWWE sample maintained a higher percentage bioluminescence response than the 0% sample for the entire 60 min (Fig. 5.16).

V. fischeri only experienced an EC50 value of 100% for the KMWWE and AWWE samples (Table 5.1).

- 79 -

120

100 0% 80 0.01% 0.10% 60 1% 40 10% 20 100% Bioluminescence (%) 0 1 15304560 Time (min)

Fig. 5.12 Bioluminescent response of V. fischeri in the presence of various concentrations of the New Germany wastewater effluent sample.

140 120 0% 100 0.01% 80 0.10% 60 1% 40 10% 100% Bioluminescence (%) 20 0 115304560 Time (min)

Fig. 5.13 Bioluminescent response of V. fischeri in the presence of various concentrations of the Kwa-Mashu wastewater effluent sample.

140 120 0% 100 0.01% 80 0.10% 60 1% 40 10% 20 100% Bioluminescence (%) 0 115304560 Time (min)

Fig. 5.14 Bioluminescent response of V. fischeri in the presence of various concentrations of the Northern wastewater effluent sample.

- 80 - 140 120 0% 100 0.01% 80 0.10% 60 1% 10% 40 100%

Bioluminescence (%) Bioluminescence 20 0 1 15304560 Time (min)

Fig. 5.15 Bioluminescent response of V. fischeri in the presence of various concentrations of the Phoenix wastewater effluent sample.

140 120 0% 100 0.01% 80 0.10% 60 1% 10% 40 100%

Bioluminescence (%) Bioluminescence 20 0 1 15304560 Time (min)

Fig. 5.16 Bioluminescent response of V. fischeri in the presence of various concentrations of the Amanzimtoti wastewater effluent sample.

5.4 DISCUSSION E. coli DH5α recA: lux, E. coli DH5α fabA: lux and E. coli DH5α uspA: lux all exhibited an overall increase in percentage bioluminescence in the presence of all five wastewater effluent samples (Figs 5.2 – 5.16). This percentage bioluminescence induction implied that at least 0.01 mg/l heavy metals and microgram concentrations of organic compounds were present in the various wastewater effluent samples. This was deduced since the profiles of the E. coli DH5α recA: lux, E. coli DH5α fabA: lux and E. coli DH5α uspA: lux graphs (Figs 5.7 – 5.11) were similar to the heavy metal compound and BTEX standard graphs generated in Chapter four (Figs 4.8; 4.27 – 4.31). On the other hand, S. sonnei pLux, E. coli DH5α pLux (Figs 5.2 – 5.11) and the V. fischeri-based BioToxTM kit (Figs 5.12 – 5.16) exhibited an overall decrease in percentage bioluminescence in the presence of all wastewater effluent samples. This percentage bioluminescence inhibition also indicated that at least 0.01 mg/l heavy

- 81 - metals or microgram concentrations of BTEX compounds were present. This was deduced since the shape of the S. sonnei pLux, (Figs 5.2 – 5.6) and V. fischeri-based BioTox kit (Figs 5.12 – 5.16) graphs were similar to the heavy metal compound and BTEX standard graphs generated in Chapter Four (Figs 4.3 – 4.11; Fig 4.18; Figs 4.19 – 4.26; Figs 4.10; Figs 4.35 – 4.42). The analyses from Umgeni Water Analytical Services Department (Table 5.4) confirmed the presence of appreciable concentrations of zinc in all the wastewater effluent samples, except the PWWE sample. The NGWWE sample also contained appreciable concentrations of total chromium and lead (Table 5.4). The metal binding capacity of the gram-negative E. coli cell has been studied in detail before. The phosphoryl groups of the outer membrane lipopolysaccharides are responsible for the high affinity of divalent metal ions (e.g. copper, zinc, lead, cadmium). The peptidoglycan layer of E. coli, which is most probably one molecule thick, binds metal ions via the carboxyl group of the D-glutamic acid of the peptide stem and the hydroxyl groups of the glycan backbone (Kotrba et al., 1999). This mode of metal ion interaction may also have occurred in the gram-negative S. sonnei cell. The bacterial biosensors may have inevitably accumulated metal ions present at higher concentrations, by the above mode of action, which combined with their proteins, often their sulphydral groups, and inactivated them. The metal ions may have also precipitated out cell proteins. Microgram quantities of BTEX compounds and organochlorine pesticides were found in all wastewater effluent samples (Table 5.4). This accounted for the high EC100 values of E. coli DH5α recA:lux, E. coli DH5α fabA:lux and E. coli DH5α uspA:lux

(Table 5.3) and the EC20 values of E. coli DH5α pLux. However, E. coli DH5α pLux did not achieve an EC20 value for the KMWWE sample (Table 5.2). Possibly, the higher combined concentrations of BTEX compounds in the KMWWE sample, as compared to the other wastewater samples, could have accounted for the increased toxicity of the KMWWE sample to E. coli DH5α pLux (Table 5.4). The monoaromatic compounds such as BTEX are generally easier to break down than higher molecular weight organochlorine pesticides. However, the organic compounds may be toxic to the microorganisms at higher concentrations. This may account for the lowest percentage bioluminescence exhibited at the 100% NGWWE samples by E. coli DH5α recA:lux, E. coli DH5α uspA:lux and the V. fischeri-based BioToxTM kit; the 100% KMWWE sample by E. coli DH5α recA:lux, E. coli DH5α fabA:lux, E. coli DH5α uspA:lux, S. sonnei pLux and the V. fischeri-based BioToxTM kit; the 100% PWWE sample by E. coli DH5α fabA:lux and E. coli DH5α uspA:lux; and the 100% AWWE sample by E. coli DH5α recA:lux, E. coli DH5α uspA:lux and V. fischeri-based BioToxTM kit. These low percentage bioluminescence values may have been due to high genotoxic stress and general protein damage exhibited at 100% of the wastewater effluent samples.

- 82 - However, at lower concentrations of the wastewater samples, the organochlorine pesticides, BTEX and heavy metal compounds were present in too low concentrations to result in high genotoxic and general protein damage. This may account for the lowest percentage bioluminescence exhibited at the 0.1% NGWWE sample by E. coli DH5α fabA:lux; the 0.01% NWWE sample by E. coli DH5α recA:lux, E. coli DH5α fabA:lux, E. coli DH5α uspA:lux; the 0.01% PWWE sample by E. coli DH5α recA:lux; and the 0.01% AWWE sample by E. coli DH5α fabA:lux. The decreased bioluminescence response at lower heavy metal concentrations may also be due to the metal-ion homeostasis mechanisms of complexation and ZntA P-type ATPase transport pumps (Nies et al., 1999 and Gatti et al., 2000). An understanding of the toxicity imposed by combinations of pollutants is essential for interpreting the fate of environmental contaminants (Strachan et al., 2001). In direct contrast with the biosensor results in this chapter, the Daphnia LC50 toxicity test reflected that none of the wastewater samples were toxic (Table 5.4). Interestingly, the

0.22 mg/l zinc present in the NGWWE sample did not result in an LC50 value, yet an

LC50 value of 0.2 mg/l was previously obtained for zinc sulphate (Table 4.7). These results indicated that S. sonnei pLux was more sensitive than the Daphnia LC50 toxicity test, especially when a mixture of heavy metals were present. S. sonnei pLux obtained TM EC50 for all wastewater effluent samples, while the V. fischeri-based BioTox kit only obtained EC50 values for the KMWWE and AWWE samples (Table 5.1). Thus, S. sonnei pLux biosensor was more sensitive to all wastewater effluent samples than the V. fischeri-based BioToxTM kit. This may have been due to the limitation of the V. fischeri marine bacterium in a freshwater toxicity test because of its requirement for saline conditions. Overall the decreasing trend in percentage bioluminescence induction observed in the presence of the wastewater effluent samples was E. coli DH5α uspA:lux, E. coli DH5α recA:lux and E. coli DH5α fabA: lux. This implied that in the presence of the various heavy metal and organic pollutants present in the wastewater effluent samples, the general stress exhibited by the E. coli DH5α biosensors was mostly due to DNA damage rather than membrane damage (Table 3.1). This chapter highlighted the use of E. coli DH5α pLux, E. coli DH5α recA: lux, E. coli DH5α fabA:lux, E. coli DH5α uspA: lux, S. flexneri pLux, Enteropathogenic E. coli plux and S. sonnei pLux as rapid, efficient, acute toxicity tools for detecting heavy metals and organic compounds in wastewater effluent samples. E. coli DH5α pLux, E. coli DH5α recA: lux, E. coli DH5α fabA: lux, E. coli DH5α uspA: lux and S. sonnei pLux were advantageous as they allowed for the sensitive measurement of environmental pollutants at concentrations not detectable by the standard Daphnia LC50 toxicity test.

- 83 - CHAPTER SIX: CONCLUSIONS AND RECOMMENDATIONS

6.1 SUMMARY OF MAJOR FINDINGS AND CONCLUSIONS REACHED All of the objectives of the study were successfully met. Several different bacterial biosensors with the ability to emit a readily detectable signal (light) in the presence of a wide range of environmental pollutants were developed. The biosensors represent a fusion of bacterial bioluminescence (lux) genes, as a reporter, to selected bacterial gene promoters. These included luxCDABE-containing E. coli DH5α, Enteropathogenic E. coli, S. flexneri and S. sonnei bacterial biosensor systems.

A comparison of the EC50 and LC50 with heavy metal compounds indicated that S. sonnei, developed in this study, was more sensitive than the commercially available Vibrio fischeri-based BioToxTM kit and the traditional ecotoxicity test using Daphnia. The toxic effect of the different BTEX compounds varied among the biosensors. The different biosensors also exhibited varying degrees of toxicity to the range of heavy metals tested. The biosensors demonstrated their potential to detect the presence of the heavy metals and other pollutants in effluent from the New Germany, Kwa-Mashu, Phoenix, Northern and Amanzimtoti wastewater treatment works. All biosensors were able to detect pollutants in the wastewater effluent samples at concentrations too low for

detection with the commercially available Daphnia LC50 toxicity test. The data generated in this research demonstrates that these biosensors are potentially useful for the evaluation of environmental water samples and pollution management.

Economics It is proposed that the biosensors developed in this study should be used as a first line of detection for the presence of inhibitory or toxic pollutants in water and wastewater. Thereafter, established physical and chemical methods may be used to verify findings obtained with the biosensors. Consequently, this will reduce the need to test large numbers of water and wastewater samples by the use of expensive physical and chemical methods. The biosensors developed in this study have also demonstrated their

enhanced sensitivity over the standard toxicity tests, e.g. Daphnia LC50 toxicity test, which are significantly more expensive.

Feasibility Biosensors are a rapid and convenient way to measure acute toxicity. These microbial biosensors offer advantages in that they are sensitive, reproducible and relatively inexpensive when compared to currently used ecotoxicity tests, for example Daphnia

- 84 - test. Biosensors have the added advantage that they can be easily adapted for use in the field or on-site testing without any laboratory facilities. For such an application the only requirements would be freeze-dried biosensor cultures, a portable luminometer and a semi-skilled personnel.

Test procedure Freeze dried biosensor cells are first resuscitated for 30-60 min at 26ºC. 1 ml of resuscitated cells is diluted with 9 ml of LB broth, washed in 0.1 ml of KCL (this step is optional) and utilized within 30 min. In the field, 1 ml of biosensor cells is added to 9 ml of sample and the bioluminescence recorded with a portable luminometer. Alternatively, the assay may be carried out with a Fluoroskan Ascent FL instrument. In this case 30 Pl of these cells were added to the microplate wells containing the various concentrations of toxicants or water samples. The Fluoroskan Ascent FL instrument measures the bioluminescence at preset intervals over a specific period of time. The end result is a bioluminescence curve as a function of time from which the response of the biosensors is assessed with respect to the toxicant present.

Application potential The greatest advantage of biosensors is that they can be used in toxicity evaluation of environmental pollution and can indicate the bioavailability of pollutants in a manner that chemical analyses cannot. Biosensors have the potential to offer a risk assessment strategy to predict the level at which a contaminated site may be bioremediated. Therefore, the successful integration of the powerful applications of biosensor technology in pollution management may be one alternative to reverse global environmental mistreatment.

6.2 RECOMMENDATIONS FOR FUTURE USE Biosensors constructed in this study have the capacity to monitor environmental pollution. These whole cell biosensors hold a great deal of promise for continuous on- line monitoring of pollutant concentrations in environmental applications.

The main application of these biosensor systems may be for the prescreening of environmental samples for toxic agents. Suspicious findings that imply the presence of pollutants may then be verified using established physical-chemical methods utilised at environmental laboratories. This will reduce the costs of standard toxicity tests, e.g.

Daphnia toxicity LC50 test. Therefore, efforts to design portable field devices that incorporate the biosensors constructed in this study will be initiated.

- 85 - The freeze-dried biosensors can be resuscitated directly, using the wastewater samples and the bioluminescent response measured using the portable 1254 - 001 LUMINOVA luminometer (Bio-Orbit Oy, Finland). The ultimate aim will be to design a portable system, which can be used by semi-skilled people without a microbiological background.

The growing interest in employing whole-cell biosensors for sentinel or first alert detection of specific substance attests to the fact that they have several potential uses aside from bioremediation and bioavailability assays. These applications include environmental hazard evaluations; prosecution and defense of chemical-related activities in environmental litigation; permitting for the discharge of municipal and industrial waste; and corporate industrial decisions on product development, manufacture and commercialisation so as to avert potential pollution. Therefore, the successful integration of the powerful applications of biosensor technology in pollution management may be one alternative to reverse the years of global environmental mistreatment. The bright, if not brilliant, future of biosensor technology is assured.

6.3 RECOMMENDATIONS FOR TECHNOLOGY TRANSFER A hands-on workshop for all in the wastewater, water and health sectors. The publishing and distribution of a simplified manual of this method and its applications. The development of a low-cost and user friendly kit for on-site use. Awareness and education will most assuredly continue to be the two most important ways to prevent water pollution. This requires knowledgeable and informed science and engineering practitioners to nurture components of environmental biotechnology.

- 86 - REFERENCES

APPLEGATE BM, KEHRMEYER SR and SAYLER GS (1998) A chromosomally based tod- luxCDABE whole-cell reporter for benzene, toluene, ethylbenzene, and xylene (BTEX) sensing. Appl. Environ. Microbiol. 49 103-119. BECHOR OD, SMULSKI R, VAN DYK TK, LAROSSA RA and BELKIN S (2002) Recombinant microorganisms as environmental biosensors: pollutant detection by Escherichia coli bearing fabA::lux fusions. J. Biotechnol. 94 125-132. BELKIN S, SMULSKI DR, DADON S, VOLLMER AC, VAN DYK TK and LAROSSA RA (1997) A panel of stress responsive luminous bacteria of selected classes of toxicants. Water Resour. 31 3009-3016. BELKIN S, VAN DYK TK, VOLLMER AC, SMULSKI DR and LAROSSA RA (1996) Monitoring subtoxic environmental hazards by stress-responsive luminous bacteria. Environ. Toxicol. Water Qual. 11 179-185. BEN-ISRAEL O, BEN-ISRAEL H and ULITZUR S (1998) Identification and quantification of toxic chemicals by use of Escherichia coli carrying lux genes fused to stress promoters. Appl. Environ. Microbiol. 64 4346-4352. BIRNBOIM HC, and DOLY J (1979) A rapid alkaline extraction procedure for screening recombinant plasmid DNA. Nucleic Acids Res. 7 1513-1523. BLOUIN KS, WALKER G, SMIT J and TURNER RFB (1996) Characterisation of in vivo reporter systems for gene expression and biosensor applications based on luxAB luciferase. Appl. Environ. Microbiol. 62 2013-2021. BURLAGE RS and KUO CT (1994) Living biosensors for the management and manipulation of microbial consortia. Ann. Rev. Microbiol. 48 291-309. CAI J and DUBOW MS (1997) Use of a luminescent bacterial biosensor for biomonitoring and characterisation of arsenic toxicity of chromated copper arsenate (CCA). Biodegradation 8 105-111. CHOI SH and GU MB (2002) A portable toxicity biosensor using freeze-dried recombinant bioluminescent bacteria. Biosens. Bioelectron. 17 433-440. COLE MG (1994) Assessment and remediation of petroleum contaminated sites. Press. Boca Raton, FL, 63. In: OBT® Oil Degradation Treatment, Bioremediation and contaminants, United-Tech, Inc (http://www.united-tech.com).

- 87 - COOK N, SILCOCK DJ, WATERHOUSE RN, PROSSER JI, GLOVER LA and KILHAM K (1993) Construction and detection of bioluminescent strains of Bacillus subtilis. J. Appl. Bacteriol. 75 350-359. DAVIDOV Y, ROZEN R, SMULSKI DR, VAN DYK TK, VOLLMER AC, ELSEMORE DA, LAROSSA RA and BELKIN S (2000) Improved bacterial SOS promoter: lux fusions for genotoxicity detection. Mutation Res. 466 97-107. DONG S and CHEN X (2002) Some new aspects in biosensors. Rev. Mol. Biotechnol. 2 303- 323. EPUSA (2003) National drinking water regulations; Announcement of completion of EPA’s review of existing water standards; Notice. Federal register. U. S. Environmental Protection Agency, Washington D. C (USA) 68 42908-42929. ETHEKWINI WASTEWATER (2003) http://www.durban.gov.za/wastewater/branches. FARRE M, PASINI O, ALONSO MC, CASTILLO M and BARCELO D (2001) Toxicity assessment of organic pollution in wastewaters using a bacterial biosensor. Anal. Chim. Acta 425 155-165. GATTI D, MITRA B and ROSEN BP (2000) Escherichia coli soft metal ion-translocating ATPases. J. Biol. Chem. 275 34009-34012. GU MB and CHANG ST (2001) Soil biosensor for the detection of PAH toxicity using an immobilised recombinant bacterium and a biosurficant. Biosens. Bioelectron. 16 667- 674. GU MB, CHOI SH and KIM SW (2001) Some observations in freeze-drying of recombinant bioluminescent Escherichia coli for toxicity monitoring. J. Biotechnol. 88 95-105. HANSEN LH and SORENSON SJ (2000) Versatile biosensor vectors for detection and quantification of mercury. FEMS Microbiol. Lett. 193 123-127. HAY AG, RICE JF, APPLEGATE BM, BRIGHT NG and SAYLER GS (2000) A bioluminescent whole-cell reporter for detection of 2,4-dichlorophenoxyacetic acid and 2,4-dichlorophenol in soil. Appl. Environ. Microbiol. 66 4589-4594. HEITZER A, MALACHOWSKY K, THONNARD JE, BIENKOWSKI PR, WHITE DC and SAYLER GS (1994) Optical biosensor for environmental on-line monitoring of naphthalene and salicylate bioavailability with an immobilised bioluminescent catabolic reporter bacterium. Appl. Environ. Microbiol. 60 1487-1484. HIRA KG (2002) Application of bioluminescence to determine the fate of plasmids in viable but non-culturable bacteria. M.Sc. Dissertation. University of Durban-Westville, South Africa. HOLMES S (1996) South African water quality guidelines 2nd ed. In: Domestic use. Published by the Department of Water Affairs and Forestry, South Africa 1 36-106.

- 88 - KAISER KLE and PALABRICA VS (1991) Photobacterium phosphoreum toxicity data index. Wat. Pol. Res. J. Canada 26 361-431. KEANE A, PHOENIX P, GHOSHAL S and LAU PCK (2002) Exposing culprit organic pollutants: A review. J. Microbiol. Methods 49 103-119. KING JMH, DIGRAZIA PM, APPLEGATE B, BURLAGE R, SANSEVERINO J, DUNBAR P, LARIMER F and SAYLER GS (1990). Rapid sensitive bioluminescent reporter technology for naphthalene exposure and biodegradation. Science 249:778-781. KOHLER S, BELKIN S and SCHMID RD (2000) Reporter gene assays in environmental analysis. Fresenius. J. Anal. Chem. 366 769-779. KOTRBA P, DOLECKOVA L, DE LORENZO V and RUML T (1999) Enhanced bioaccumulation of heavy metal ions by bacterial cells due to surface display of short metal binding peptides. Appl. Environ. Microbiol. 65 1092-1098. LESLIE SB, ISRAEL E, LIGHTHART B, CROWE JH and CROWE LM (1995) Trehalose and sucrose protect both membranes and proteins in intact bacteria during drying. Appl. Environ. Microbiol. 61 3592-3597. LEVEAU JHJ and LINDOW SE (2002) Bioreporters in microbial ecology. Curr. Opin. Microbiol. 5 259-265. MALIK KA (1992) Technical information for culture collection curators in developing countries. Unesco/WFCC-educations committee. Braunschweig. Federal Republic of Germany, pp 24-27. MEIGHEN EA (1991) Molecular biology of bacterial bioluminescence. Microbiol. Rev. 55 123-142. NEILSON JW, PIERCE SA and MAIER RM (1999) Factors influencing expression of luxCDABE and nah genes in Pseudomonas putida RB1353(NAH7, pUTK9) in dynamic systems. Appl Environ Microbiol. 65 3473-3482. NIES DH (1999) Microbial heavy-metal resistance. Appl. Microbiol. Biotechnol. 51 730-750. PATON GI, RATTRAY EAS, CAMPBELL CD, CRESSER MS, GLOVER LA, MEEUSSEN JCL and KILLHAM K (1997) Use of genetically modified microbial biosensors for soil ecotoxicity testing. In: Bioindicators of Soil Health (Pankhurst, C. S., B. Doube, and V. Gupta., Eds.), CAB International, pp 397-417. PATON GI, CAMPBELL CD, GLOVER LA and KILLHAM K (1995) Assessment of bioavailability of heavy metals using lux modified constructs of Pseudomonas fluorescens. Lett. Appl. Microbiol. 20 52-56. PRESCOTT LM, HARLEY JP and KLEIN DA (1996) Microbiology 3rd ed. USA, Chicago: Wm. C. Brown Publishers, pp 855. PUROHIT HJ (2003) Biosensors as molecular tools for use in bioremediation. J. Clean. Prod. 11 293-301.

- 89 - RICHMAN MIND (1997) Water pollution. Wastewater 5 24-29. SAMBROOK J, FRITSCH EF and MANIATIS T (1989) Molecular Cloning: a laboratory manual. Cold Spring Harbor Laboratory Press, Cold Spring Harbour, New York. SAYLER GS, FOX R and BLACKBURN JW (1991) Environmental biotechnology for waste treatment. Environ. Sci. Res. 41 9-13. SELIFONOVA O, BURLAGE R and BARKAY T (1993) Bioluminescent sensors for the detection of bioavailable Hg (II) in the environment. Appl. Environ. Microbiol. 59 3083-3090. SELIFONOVA OV and EATON RW (1996) Use of an ipb-lux fusion to study regulation of the isopropylbezene catabolism operon of Pseudomonas putida RE204 and to detect hydrophobic pollutants in the environment. Appl. Environ. Microbiol. 62 778-783. SHAW JJ and KADO CI (1986) Development of a Vibrio bioluminescence gene-set to monitor phytopathogenic bacteria during the ongoing disease process in a nondisruptive manner. Bio/Technol. 4 560-564. SINCLAIR GM, PATON GI, MEHARG AA and KILLHAM K (1999) Lux-biosensor assessment of pH effects on microbial sorption and toxicity of chlorophenols. FEMS Microbiol. Lett. 174 273-278. SOUSA S, DUFFY C, WEITZ H, GLOVER LA, BAR E, HENKLER R and KILLHAM K (1998) Environ. Toxicol. Chem. 17 1039-1045. STEINBERG SM (1995) A review of environmental applications of bioluminescence measurements. Chemosphere 30 2155-2197. STEWART GSAB and WILLIAMS P (1992) lux genes and the applications of bacterial bioluminescence. J. Gen. Microbiol. 138 1289-1300. STICHER P, JASPERS MCM, STEMMLER K, HARMS H, ZEHNDER AJB and VAN DER MEER JR (1997) Development and characterisation of a whole-cell bioluminescent sensor for bioavailable middle-chain alkanes in contaminated groundwater. Appl. Environ. Microbiol. 63 4053-4060. STRACHAN G, PRESTON S, MACIEL H, PORTER AJR and PATON GI (2001) Use of bacterial biosensors to interpret the toxicity and mixture toxicity of herbicides in freshwater. Water Res. 35 3490-3495. VAN DYK TK, SMULSKI DR, REED TR, BELKIN S, VOLLMER AC and LAROSSA RA (1995) Responses to toxicants of an Escherichia coli strain carrying a uspA: lux genetic fusion and an E. coli strain carrying a grpE: lux fusion are similar. Appl. Environ. Microbiol. 61 4124-4127. VOLLMER AC, BELKIN S, SMULSKI DR, VAN DYK TK and LAROSSA RA (1997) Detection of DNA damage by use of Escherichia coli carrying recA: lux, uvrA: lux, or alkA: lux reporter plasmids. Appl. Environ. Microbiol. 63 2566-2571.

- 90 - WALL AJ and STRATTON GW (1994) Effects of a chromated-copper-arsenate wood preservative on the bacterial degradation of pentachlorophenol. Can J Microbiol. 40 388-392. WEITZ HJ, RITCHIE JM, BAILEY DA, HORSBURGH AM, KILLHAM K and GLOVER LA (2001) Construction of a modified mini-Tn5 luxCDABE transposon for the development of bacterial biosensors for ecotoxicity testing. FEMS Microbiol. Lett. 197 159-165. World Book (1992) Water flea. In: The World Book Encyclopedia. Published by World Book International, USA 21 85-86.

- 91 - APPENDIX 1 STANDARD DEVIATIONS OF BIOLUMINESCENCE VALUES

Appendix one consists of the standard deviations of the triplicate bioluminescence values and the averages of the triplicate bioluminescence values of S. sonnei pLux, S. flexneri pLux, E. coli DH5α pLux, E. coli DH5α recA: lux, E. coli DH5α fabA:lux, E. coli DH5α uspA:lux, Enteropatogenic E. coli pLux and V. fischeri, in the presence of heavy metal and organic compounds.

Table 1a Standard deviations of the triplicate bioluminescence values of the S. sonnei pLux biosensor in the presence of chromium trioxide (Fig. 4.3) Time Standard deviations of the various concentrations (min) 0 mg/l 0.01mg/l 0.1 mg/l 1 mg/l 10 mg/l 100 mg/l 1 0.005868 0.014031 0.005493 0.005162 4.54E-05 0 15 0.001035 0.008404 0.00257 0.001524 0.000233 0 30 0.002084 0.009704 0.00457 0.001259 0.000106 0 45 0.001744 0.003253 0.003386 0.000217 4.28E-05 0.000154 60 0.002018 0.004241 0.002321 0.002013 2.22E-05 2.36E-05

Table 1b Averages of the triplicate % bioluminescence values of the S. sonnei pLux biosensor in the presence of chromium trioxide (Fig. 4.3) Time Average bioluminescence (%) at various concentrations (min) 0 mg/l 0.01 mg/l 0.1 mg/l 1 mg/l 10 mg/l 100 mg/l 1 100 100.7063 96.99038 77.93846 31.92579 1.703316 15 72.52589 74.00509 69.26819 41.40587 0.677955 0.060123 30 55.62708 61.51964 55.75991 30.75454 0.233448 0 45 47.40331 49.43087 47.2355 28.11593 0.204519 0.079493 60 37.72907 40.37445 38.5572 21.96198 0.041004 0.010817

Table 2a Standard deviations of the triplicate bioluminescence values of the S. flexneri pLux biosensor in the presence of chromium trioxide (Fig. 4.4) Time Standard deviations of the various concentrations (min) 0 mg/l 0.01mg/l 0.1 mg/l 1 mg/l 10 mg/l 100 mg/l 1 0.008258 0.009528 0.006066 0.002381 0.003473 0.000176 15 0.005752 0.02013 0.006787 0.000674 0.00526 9.39E-05 30 0.005459 0.015248 0.005094 0.002433 0.005136 3.02E-04 45 0.005588 0.011688 0.004051 0.001029 5.87E-03 0.000191 60 0.003514 0.009202 0.003627 0.001749 3.81E-03 0.00024

- 92 - Table 2b Averages of the triplicate % bioluminescence values of the S. flexneri pLux biosensor in the presence of chromium trioxide (Fig. 4.4) Time Average bioluminescence (%) at various concentrations (min) 0 mg/l 0.01 mg/l 0.1 mg/l 1 mg/l 10 mg/l 100 mg/l 1 100 100.6243 102.608 96.69068 89.68605 67.96184 15 76.07519 74.10754 77.28508 65.75009 50.77255 42.393 30 68.68836 71.93874 70.91774 64.19364 48.40507 38.95282 45 64.44609 61.54276 64.68161 57.8208 42.67526 34.41288 60 54.46213 52.64916 52.74547 48.88449 36.79183 28.9025

Table 3a Standard deviations of the triplicate bioluminescence values of the E. coli DH5α pLux biosensor in the presence of chromium trioxide (Fig. 4.5) Time Standard deviations of the various concentrations (min) 0 mg/l 0.01 mg/l 0.1 mg/l 1 mg/l 10 mg/l 100 mg/l 1 0.004155 0.015733 0.004627 0.005147 0.014783 0.001549 15 0.004633 0.008193 0.004938 0.002646 0.000146 3.46E-05 30 0.002827 0.004152 0.001846 0.000587 3.82E-05 0 45 0.000673 0.005151 0.001409 0.002244 8.97E-05 2.90E-05 60 0.003884 0.00316 0.001901 0.00098 7.76E-05 3.73E-05

Table 3b Averages of the triplicate % bioluminescence values of the E. coli DH5α pLux biosensor in the presence of chromium trioxide (Fig. 4.5) Time Average bioluminescence (%) at various concentrations (min) 0 mg/l 0.01 mg/l 0.1 mg/l 1 mg/l 10 mg/l 100 mg/l 1 100 83.88291 90.23148 28.98623 0.19909 0.000624 15 105.333 89.35158 102.3392 7.162871 0.086543 0 30 94.32946 84.46897 100.5022 6.358814 0 0 45 82.27034 75.2423 89.09986 6.927999 0.03703 0.145833 60 73.62696 68.59878 81.58248 7.519652 0.072604 0.056794

Table 4a Standard deviations of the triplicate bioluminescence values of the Enteropathogenic E. coli pLux biosensor in the presence of chromium trioxide (Fig. 4.6) Time Standard deviations of the various concentrations (min) 0 mg/l 0.01mg/l 0.1 mg/l 1 mg/l 10 mg/l 100 mg/l 1 0.008258 0.009528 0.006066 0.002381 0.003473 0.000176 15 0.005752 0.02013 0.006787 0.000674 0.00526 9.39E-05 30 0.005459 0.015248 0.005094 0.002433 0.005136 3.02E-04 45 0.005588 0.011688 0.004051 0.001029 5.87E-03 0.000191 60 0.003514 0.009202 0.003627 0.001749 3.81E-03 0.00024

- 93 - Table 4b Averages of the triplicate % bioluminescence values of the Enteropathogenic E. coli pLux biosensor in the presence of chromium trioxide (Fig. 4.6) Time Average bioluminescence (%) at various concentrations (min) 0 mg/l 0.01 mg/l 0.1 mg/l 1 mg/l 10 mg/l 100 mg/l 1 100 97.13482 95.88069 91.6606 77.8391 62.9772 15 89.76613 86.61262 84.91981 80.0356 59.6126 51.6468 30 76.38265 74.59899 74.24722 65.9186 50.1348 42.9101 45 61.33505 63.9226 55.13861 53.7082 41.2954 33.4336 60 55.6426 52.04014 49.0989 44.5382 35.0711 28.1237

Table 5a Standard deviations of the triplicate bioluminescence values of the E. coli DH5α recA: lux biosensor in the presence of chromium trioxide (Fig. 4.7) Time Standard deviations of the various concentrations (min) 0 mg/l 0.01 mg/l 0.1 mg/l 1 mg/l 10 mg/l 100 mg/l 1 0.000281 0.000648 0.000651 0.00035 0.00016 0.000116 15 0.000869 0.000448 5.94E-05 0.000165 5.13E-05 0.000123 30 0.000258 0.000582 0.000329 0.00012 0 0.000121 45 0.000393 0.000199 0.000174 9.38E-05 0.000683 1.97E-05 60 0.00018 0.000512 0.0008 0.000221 5.22E-05 8.49E-05

Table 5b Averages of the triplicate % bioluminescence values of the E. coli DH5α recA:lux biosensor in the presence of chromium trioxide (Fig. 4.7) Time Average bioluminescence (%) at various concentrations (min) 0 mg/l 0.01 mg/l 0.1 mg/l 1 mg/l 10 mg/l 100 mg/l 1 100 84.25218 98.25328 31.70033 9.579695 2.893012 15 170.6878 142.1534 117.9312 28.13864 2.510916 6.509279 30 184.9618 167.6719 167.0442 45.97434 0 3.04312 45 241.1436 188.1414 213.7964 69.25491 17.00328 0.491267 60 247.7893 243.2724 248.458 99.48144 2.456334 2.715606

Table 6a Standard deviations of the triplicate bioluminescence values of the E. coli DH5α fabA: lux biosensor in the presence of chromium trioxide (Fig. 4.8) Time Standard deviations of the various concentrations (min) 0 mg/l 0.01 mg/l 0.1 mg/l 1 mg/l 10 mg/l 100 mg/l 1 0.000286 0.000331 0.000678 9.55E-05 6.91E-05 0.00023 15 0.000371 0.000308 0.000237 0.000325 0.000772 0.00023 30 0.000618 8.22E-05 0.000421 0.00039 0.000256 0.000189 45 0.00036 0.000667 0.000704 0.000183 0.000126 2.10E-05 60 0.000499 0.000456 0.000242 9.68E-05 0.000186 0.000136

- 94 - Table 6b Averages of the triplicate % bioluminescence values of the E. coli DH5α fabA: lux biosensor in the presence of chromium trioxide (Fig. 4.8) Time Average bioluminescence (%) at various concentrations (min) 0 mg/l 0.01 mg/l 0.1 mg/l 1 mg/l 10 mg/l 100 mg/l 1 100 92.61136 71.19485 52.10049 18.68272 15.42115 15 129.449 113.7957 90.55309 18.60883 14.53452 17.56386 30 178.3618 141.5031 105.7843 27.62297 8.644714 10.97741 45 183.9666 160.2702 105.8898 32.52058 5.552039 13.85899 60 204.5598 189.5292 137.3971 48.21618 13.384 12.26515

Table 7a Standard deviations of the triplicate bioluminescence values of the E. coli DH5α uspA: lux biosensor in the presence of chromium trioxide (Fig. 4.9) Time Standard deviations of the various concentrations (min) 0 mg/l 0.01 mg/l 0.1 mg/l 1 mg/l 10 mg/l 100 mg/l 1 0.000588 0.000249 0.000349 0.000193 1.92E-05 0.000147 15 0.000221 0.000825 0.000312 8.60E-05 0.000101 0.000204 30 0.000714 0.000262 0.00026 0.000215 6.98E-05 0.000142 45 0.000404 0.000412 0.000237 0.000321 0.000153 3.86E-05 60 0.000373 0.00048 0.00184 0.000416 4.06E-05 5.74E-05

Table 7b Averages of the triplicate % bioluminescence values of the E. coli DH5α uspA: lux biosensor in the presence of chromium trioxide (Fig. 4.9) Time Average bioluminescence (%) at various concentrations (min) 0 mg/l 0.01 mg/l 0.1 mg/l 1 mg/l 10 mg/l 100 mg/l 1 100 100.4115 102.4502 69.89619 19.53615 6.359299 15 126.4379 122.7345 118.947 25.36238 5.00327 6.40606 30 171.6169 145.8337 139.3528 19.5455 1.842324 3.310573 45 241.2419 201.356 196.6427 30.88001 6.097445 3.413444 60 423.4176 395.801 330.9361 79.42579 0 2.066771

Table 8a Standard deviations of the triplicate bioluminescence values of the V. fischeri-based BioTox kit in the presence of chromium trioxide (Fig. 4.10) Time Standard deviations of the various concentrations (min) 0 mg/l 0.01 mg/l 0.1 mg/l 1 mg/l 10 mg/l 100 mg/l 1 0.091037 0.043808 0.119978 0.184416 0.133057 0.001335 15 0.059824 0.069559 0.039982 0.044811 0.077164 0.000469 30 0.052022 0.069254 0.042211 0.066068 0.079509 4.01E-05 45 0.042241 0.061381 0.039627 0.052159 0.067945 8.09E-05 60 0.053817 0.051285 0.038503 0.050084 0.048101 0.000121

- 95 - Table 8b Averages of the triplicate % bioluminescence values of the V. fischeri-based BioTox kit in the presence of chromium trioxide (Fig. 4.10) Time Average bioluminescence (%) at various concentrations (min) 0 mg/l 0.01 mg/l 0.1 mg/l 1 mg/l 10 mg/l 100 mg/l 0 100 105.9481 103.9896 104.2371 89.24543 59.48369 15 85.38308 88.76499 86.27825 79.05713 48.92564 10.45606 30 86.10664 88.44138 86.04629 70.07382 32.78139 2.125542 45 81.10762 82.41538 80.54875 61.58321 23.84901 0.58978 60 75.96249 76.26328 75.06959 56.31505 20.20715 0.310422

Table 9a Standard deviations of the triplicate bioluminescence values of the S. sonnei pLux biosensor in the presence of xylene (Fig. 4.11) Time Standard deviations of the various concentrations (min) 0 mg/l 0.01 mg/l 0.1 mg/l 1 mg/l 10 mg/l 100 mg/l 1 0.003747 0.001889 0.003463 0.002658 0.003684 0.001502 15 0.001955 0.001227 0.002906 0.001178 0.00201 0.001123 30 0.000228 0.003323 0.003393 0.000783 0.002114 0.000393 45 0.00151 0.00231 0.001884 0.00262 0.001488 0.001833 60 0.000541 0.000667 0.001791 0.003096 0.000487 0.000657

Table 9b Averages of the triplicate % bioluminescence values of the S. sonnei pLux biosensor in the presence of xylene (Fig. 4.11) Time Average bioluminescence (%) at various concentrations (min) 0 mg/l 0.01 mg/l 0.1 mg/l 1 mg/l 10 mg/l 100 mg/l 1 100 92.43907 100.708 87.96734 45.72493 11.51835 15 69.97483 66.00925 70.95047 71.35546 57.35135 10.14909 30 56.21715 51.04844 55.5401 59.85737 56.08195 12.42698 45 44.35571 41.18465 46.07753 50.21373 51.23296 15.21005 60 36.43804 34.87638 35.74805 38.69475 44.76642 16.53792

- 96 - Table 10a Standard deviations of the triplicate bioluminescence values of the S. flexneri pLux biosensor in the presence of xylene (Fig. 4.12)

Time Standard deviations of the various concentrations (min) 0 mg/l 0.01mg/l 0.1 mg/l 1 mg/l 10 mg/l 100 mg/l 1 0.000768 0.000806 0.001257 0.000775 0.000639 0.001644 15 0.000172 0.000824 0.0003 0.00123 0.000184 0.001287 30 0.000475 0.000977 0.000298 0.000764 0.000585 0.001169 45 0.000845 0.000463 0.000183 0.000133 0.000841 0.000727 60 0.000955 0.000799 0.00136 0.000178 0.001147 0.000531

Table 10b Averages of the triplicate % bioluminescence values of the S. flexneri pLux biosensor in the presence of xylene (Fig. 4.12)

Time Average bioluminescence (%) at various concentrations (min) 0 mg/l 0.01 mg/l 0.1 mg/l 1 mg/l 10 mg/l 100 mg/l 1 100 66.83702 93.00381 97.74013 97.5833 96.44133 15 63.71339 55.96497 73.95789 65.56557 63.58074 66.19644 30 40.43901 46.38054 58.35096 43.05147 43.25044 39.29373 45 28.91828 39.4474 44.45718 32.38 34.29527 32.10662 60 27.54168 36.74111 32.27809 22.95895 24.35981 25.53259

Table 11a Standard deviations of the triplicate bioluminescence values of the E. coli DH5α pLux biosensor in the presence of xylene (Fig. 4.13) Time Standard deviations of the various concentrations (min) 0 mg/l 0.01 mg/l 0.1 mg/l 1 mg/l 10 mg/l 100 mg/l 1 0.005901 0.002065 0.006155 0.004182 0.004862 0.000596 15 0.000924 0.002429 0.002295 0.001291 0.008023 0.002101 30 0.001567 0.002733 0.002182 0.00108 0.007905 0.001073 45 0.000952 0.00083 0.001516 0.001485 0.004604 0.002578 60 0.001345 0.002339 0.00198 0.001625 0.00553 0.004119

Table 11b Averages of the triplicate % bioluminescence values of the E. coli DH5α pLux biosensor in the presence of xylene (Fig. 4.13) Time Average bioluminescence (%) at various concentrations (min) 0 mg/l 0.01 mg/l 0.1 mg/l 1 mg/l 10 mg/l 100 mg/l 1 100 79.04274 89.0496 84.49043 43.68191 10.8099 15 108.6993 98.20424 105.3189 124.9783 102.6364 31.86434 30 92.53956 86.62563 90.60646 111.2218 133.4931 57.38922 45 82.15899 75.10459 81.084 97.92634 135.6354 88.36698 60 77.21644 67.05937 74.25038 87.98173 138.2064 110.3644

- 97 - Table 12a Standard deviations of the triplicate bioluminescence values of the E. coli DH5α recA: lux biosensor in the presence of xylene (Fig. 4.14)

Time Standard deviations of the various concentrations (min) 0 mg/l 0.01 mg/l 0.1 mg/l 1 mg/l 10 mg/l 100 mg/l 1 0.00038 0.000268 0.000349 0.000287 0.000422 0.000269 15 0.000305 4.25E-05 0.000293 0.000255 0.00052 0.000195 30 0.000404 4.69E-05 0.000279 0.000219 0.000672 0.000105 45 0.000256 0.000216 0.000546 0.000277 0.00045 0.000137 60 0.000479 0.000215 0.000243 0.00022 0.0007 0.000185

Table 12b Averages of the triplicate % bioluminescence values of the E. coli DH5α recA: lux biosensor in the presence of xylene (Fig. 4.14) Time Average bioluminescence (%) at various concentrations (min) 0 mg/l 0.01 mg/l 0.1 mg/l 1 mg/l 10 mg/l 100 mg/l 1 100 59.47272 133.6603 165.236 121.6432 112.3237 15 107.2042 93.71552 149.9693 152.7284 97.4862 150.3066 30 135.3771 101.3489 203.74 205.3955 156.775 183.0472 45 193.5929 171.7045 290.3127 211.7106 214.9908 203.6174 60 229.0926 244.9417 367.7805 293.9608 265.7879 249.2642

Table 13a Standard deviations of the triplicate bioluminescence values of the E. coli DH5α fabA: lux biosensor in the presence of xylene (Fig. 4.15) Time Standard deviations of the various concentrations (min) 0 mg/l 0.01 mg/l 0.1 mg/l 1 mg/l 10 mg/l 100 mg/l 1 0.000474 8.61E-05 0.000734 0.000569 0.000253 0.000865 15 0.000511 0.000386 0.001225 6.06E-05 0.000311 0.00042 30 0.000822 0.000411 0.000593 0.000349 0.000523 0.000434 45 0.000837 0.001197 0.000714 0.000387 0.000403 9.07E-05 60 0.00114 0.00096 0.000635 0.001116 0.000334 0.000342

Table 13b Averages of the triplicate % bioluminescence values of the E. coli DH5α fabA: lux biosensor in the presence of xylene (Fig. 4.15) Time Average bioluminescence (%) at various concentrations (min) 0 mg/l 0.01 mg/l 0.1 mg/l 1 mg/l 10 mg/l 100 mg/l 1 100 97.83832 103.3047 96.43471 90.94463 41.94255 15 174.5158 125.1643 135.1199 157.6607 174.0124 59.9467 30 182.061 151.5842 168.3625 172.6384 233.4024 88.76518 45 228.7415 173.6689 184.0983 201.3385 271.6789 111.3118 60 234.2197 200.7166 205.5611 222.8901 263.08 108.2736

- 98 - Table 14a Standard deviations of the triplicate bioluminescence values of the E. coli DH5α uspA: lux biosensor in the presence of xylene (Fig. 4.16) Time Standard deviations of the various concentrations (min) 0 mg/l 0.01 mg/l 0.1 mg/l 1 mg/l 10 mg/l 100 mg/l 1 0.000873 0.000597 0.000315 0.000146 0.000102 0.000183 15 0.00109 0.000426 0.000478 0.000705 0.001054 0.000592 30 0.001075 0.000592 0.000708 0.000551 0.00054 0.000396 45 0.00026 0.001298 0.00078 0.000821 0.00044 0.000573 60 0.000122 0.002535 0.000743 0.001122 7.11E-06 0.000855

Table 14b Averages of the triplicate % bioluminescence values of the E. coli DH5α uspA: lux biosensor in the presence of xylene (Fig. 4.16) Time Average bioluminescence (%) at various concentrations (min) 0 mg/l 0.01 mg/l 0.1 mg/l 1 mg/l 10 mg/l 100 mg/l 1 100 91.33802 90.74697 93.22328 78.70172 67.95068 15 164.6184 122.3785 135.7281 150.1783 214.0732 126.6993 30 209.5078 150.2802 162.0402 176.7349 286.3752 216.5597 45 277.0305 212.0452 227.2903 216.1724 386.6504 304.0151 60 438.0623 407.7041 378.3451 394.8028 555.8765 487.4656

Table 15a Standard deviations of the triplicate bioluminescence values of the Enteropathogenic E. coli pLux biosensor in the presence of xylene (Fig. 4.17) Time Standard deviations of the various concentrations (min) 0 mg/l 0.01mg/l 0.1 mg/l 1 mg/l 10 mg/l 100 mg/l 1 0.010594 0.013551 0.009946 0.011048 0.00447 0.001484 15 0.008392 0.00936 0.006197 0.009201 0.007897 0.001042 30 0.005871 0.011779 0.006551 0.007489 0.007362 0.000771 45 0.004814 0.006532 0.002093 0.008121 0.005629 0.002384 60 0.004647 0.005779 0.004789 0.007416 0.005801 0.001624

Table 15b Averages of the triplicate % bioluminescence values of the Enteropathogenic E. coli pLux biosensor in the presence of xylene (Fig. 4.17) Time (min) Average bioluminescence (%) at various concentrations 0 mg/l 0.01 mg/l 0.1 mg/l 1 mg/l 10 mg/l 100 mg/l 1 100 17.88266 50.44539 115.4173 99.07964 128.9686 15 72.47081 21.77254 60.26856 87.23455 67.60907 95.07171 30 56.44466 24.29991 59.82293 68.92446 53.33876 78.84581 45 48.45026 26.46768 55.8911 60.05611 45.94932 66.44296 60 44.91546 28.27252 53.46615 52.64598 39.64667 57.92026

- 99 - Table 16a Standard deviations of the triplicate bioluminescence values of the V. fischeri-based BioTox kit in the presence of xylene (Fig. 4.18) Time Standard deviations of the various concentrations (min) 0 mg/l 0.01 mg/l 0.1 mg/l 1 mg/l 10 mg/l 100 mg/l 1 0.21478 0.019952 0.122356 0.174028 0.121281 0.025493 15 0.168983 0.079082 0.060985 0.078542 0.074311 0.032867 30 0.139467 0.101009 0.031896 0.054394 0.054503 0.038714 45 0.118089 0.081906 0.008602 0.055771 0.062714 0.048188 60 0.121206 0.06594 0.02985 0.039956 0.054807 0.037717

Table 16b Averages of the triplicate % bioluminescence values of the V. fischeri-based BioTox kit in the presence of xylene (Fig. 4.18) Time Average bioluminescence (%) at various concentrations (min) 0 mg/l 0.01 mg/l 0.1 mg/l 1 mg/l 10 mg/l 100 mg/l 1 100 96.19194 99.18171 95.50601 59.9934 19.04983 15 102.4405 98.12459 100.6748 102.0569 85.93559 34.25286 30 96.52143 93.89005 95.71445 97.86692 94.89064 48.96148 45 88.22861 85.93786 87.25405 89.69613 91.9569 58.85152 60 80.44867 78.96336 80.17342 82.42095 87.51125 64.40353

Table 17a Standard deviations of the triplicate bioluminescence values of the S. sonnei pLux biosensor in the presence of lead acetate (Fig. 4.19) Time Standard deviations of the various concentrations (min) 0 mg/l 0.01mg/l 0.1 mg/l 1 mg/l 10 mg/l 100 mg/l 1 0.006292 0.005122 0.006896 0.008514 0.006355 0.000823 15 0.001842 0.001527 0.002408 0.001093 0.000154 0.000251 30 0.003223 0.000743 0.002601 0.000724 0.000205 0.000157 45 0.001546 0.003464 0.004189 0.000202 5.20E-05 8.50E-05 60 0.000216 0.000841 0.001258 0.000639 3.36E-05 2.41E-05

Table 17b Averages of the triplicate % bioluminescence values of the S. sonnei pLux biosensor in the presence of lead acetate (Fig. 4.19) Time Average bioluminescence (%) at various concentrations (min) 0 mg/l 0.01 mg/l 0.1 mg/l 1 mg/l 10 mg/l 100 mg/l 1 100 95.33059 89.50374 60.90522 20.3178 4.339032 15 69.51874 63.51946 61.18838 17.35224 0.805671 0.624036 30 57.02929 51.00341 49.05876 10.6751 0.145357 0.183339 45 45.25855 42.80793 40.58181 7.646203 0.044557 0.037739 60 36.24838 34.58227 31.59161 6.193122 0.029217 0.010713

- 100 - Table 18a Standard deviations of the triplicate bioluminescence values of the S. sonnei pLux biosensor in the presence of zinc sulphate (Fig. 4.20) Time Standard deviations of the various concentrations (min) 0 mg/l 0.01mg/l 0.1 mg/l 1 mg/l 10 mg/l 100 mg/l 1 0.004155 0.015733 0.004627 0.005147 0.014783 0.001549 15 0.004633 0.008193 0.004938 0.002646 0.000146 3.46E-05 30 0.002827 0.004152 0.001846 0.000587 3.82E-05 0 45 0.000673 0.005151 0.001409 0.002244 8.97E-05 2.90E-05 60 0.003884 0.00316 0.001901 0.00098 7.76E-05 3.73E-05

Table 18b Averages of the triplicate % bioluminescence values of the S. sonnei pLux biosensor in the presence of zinc sulphate (Fig. 4.20) Time Average bioluminescence (%) at various concentrations (min) 0 mg/l 0.01 mg/l 0.1 mg/l 1 mg/l 10 mg/l 100 mg/l 1 100 86.41901 97.24428 83.1465 35.52949 5.090194 15 67.56471 57.38894 65.43171 29.27534 0.259503 0.012634 30 50.35528 44.432 50.3284 19.38536 0.032087 0 45 40.13338 35.93039 41.03122 16.32905 0.09245 0.047128 60 33.91833 27.7414 32.71466 12.59393 0.085833 0.023464

Table 19a Standard deviations of the triplicate bioluminescence values of the S. sonnei pLux biosensor in the presence of copper sulphate (Fig. 4.21) Time Standard deviations of the various concentrations (min) 0 mg/l 0.01mg/l 0.1 mg/l 1 mg/l 10 mg/l 100 mg/l 1 0.001793 0.016798 0.003998 0.006645 0.005814 0.002189 15 0.001411 0.0126 0.003003 0.001168 0.000246 0 30 0.002201 0.006957 0.002262 0.000442 2.22E-05 0.000105 45 0.000808 0.003963 0.000773 0.001021 4.11E-05 4.39E-05 60 0.000562 0.006257 0.001952 0.000403 0.000204 6.42E-05

Table 19b Averages of the triplicate % bioluminescence values of the S. sonnei pLux biosensor in the presence of copper sulphate (Fig. 4.21) Time Average bioluminescence (%) at various concentrations (min) 0 mg/l 0.01 mg/l 0.1 mg/l 1 mg/l 10 mg/l 100 mg/l 1 100 87.47951 99.18681 86.25717 48.02563 6.52586 15 68.77692 58.90152 67.40272 22.38549 0.360278 0 30 49.14454 45.78127 50.72811 11.41496 0.012355 0.03826 45 39.86279 35.50696 41.34038 7.8235 0.014945 0.015941 60 32.53965 28.56782 33.26419 5.906333 0.136698 0.023314

- 101 - Table 20a Standard deviations of the triplicate bioluminescence values of the S. sonnei pLux biosensor in the presence of cadmium acetate (Fig. 4.22) Time Standard deviations of the various concentrations (min) 0 mg/l 0.01mg/l 0.1 mg/l 1 mg/l 10 mg/l 100 mg/l 1 0.006118 0.005002 0.003802 0.001918 0.00636 0.000828 15 0.002431 0.007394 0.001833 0.001087 0.00031 0.000179 30 0.00158 0.007259 0.001324 0.001103 0.000152 0 45 0.000666 0.007588 0.003423 0.002462 8.08E-05 0 60 0.000902 0.004137 0.001301 0.001387 6.75E-05 0

Table 20b Averages of the triplicate % bioluminescence of the S. sonnei pLux biosensor in the presence of cadmium acetate (Fig. 4.22) Time Average bioluminescence (%) at various concentrations (min) 0 mg/l 0.01 mg/l 0.1 mg/l 1 mg/l 10 mg/l 100 mg/l 1 100 94.9791 97.86713 91.91676 44.76614 2.290059 15 70.40352 72.24435 70.58957 49.48212 0.926068 0.153424 30 55.59824 57.86772 57.74693 37.25367 0.265165 0 45 47.67595 48.66026 46.79605 33.66942 0.328694 0 60 37.50202 38.83459 38.00723 27.74564 0.030886 0

Table 21a Standard deviations of the triplicate bioluminescence values of the S. sonnei pLux biosensor in the presence of potassium chromate (Fig. 4.23) Time Standard deviations of the various concentrations (min) 0 mg/l 0.01mg/l 0.1 mg/l 1 mg/l 10 mg/l 100 mg/l 1 0.032169 0.027481 0.033894 0.031928 0.021244 0.001633 15 0.024601 0.021757 0.025973 0.026116 0.015937 0 30 0.01988 0.018471 0.021832 0.021535 0.012265 0 45 0.015643 0.014494 0.018633 0.01815 0.010242 0 60 0.012197 0.011126 0.014488 0.014192 0.008622 6.40E-05

Table 21b Averages of the triplicate % bioluminescence values of the S. sonnei pLux biosensor in the presence of potassium chromate (Fig. 4.23) Time Average bioluminescence (%) at various concentrations (min) 0 mg/l 0.01 mg/l 0.1 mg/l 1 mg/l 10 mg/l 100 mg/l 1 100 85.42733 105.3609 99.24897 66.03701 5.075454 15 76.47342 67.63271 80.73978 81.18368 49.54183 0 30 61.79734 57.4198 67.86706 66.9417 38.12693 0 45 48.6273 45.05405 57.92287 56.41967 31.83676 0 60 37.91529 34.58711 45.03732 44.11594 26.80068 0.198845

- 102 - Table 22a Standard deviations of the triplicate bioluminescence values of the S. sonnei pLux biosensor in the presence of ethylbenzene (Fig. 4.24) Time Standard deviations of the various concentrations (min) 0 mg/l 0.01 mg/l 0.1 mg/l 1 mg/l 10 mg/l 100 mg/l 1 0.003906 0.005162 0.00499 0.000824 0.002326 0.001609 15 0.000918 0.002581 0.000566 0.0014 0.003097 0.001163 30 0.002436 0.000573 0.002415 0.001817 0.002795 0.00142 45 0.001101 0.000963 0.000741 0.002544 0.001597 0.00125 60 0.000658 0.001776 0.000474 0.000501 0.001852 0.001592

Table 22b Averages of the triplicate % bioluminescence values of the S. sonnei pLux biosensor in the presence of ethylbenzene (Fig. 4.24) Time Average bioluminescence (%) at various concentrations (min) 0 mg/l 0.01 mg/l 0.1 mg/l 1 mg/l 10 mg/l 100 mg/l 1 100 93.00192 99.48752 84.86646 39.95229 11.75771 15 71.12574 65.11138 70.76545 71.17739 52.86493 10.28533 30 55.39985 51.09606 55.07804 60.05149 53.10915 12.16814 45 43.97715 40.73904 47.39182 49.25896 47.91588 14.37529 60 36.06957 34.84799 34.96257 38.74322 43.85008 16.09718

Table 23a Standard deviations of the triplicate bioluminescence values of the S. sonnei pLux biosensor in the presence of toluene (Fig. 4.25) Time Standard deviations of the various concentrations (min) 0 mg/l 0.01 mg/l 0.1 mg/l 1 mg/l 10 mg/l 100 mg/l 1 0.00726 0.006786 0.004592 0.009464 0.002497 0.001601 15 0.002729 0.00113 0.002445 0.003698 0.000483 0.000572 30 0.001768 0.001593 0.002442 0.00317 0.000192 0.000184 45 0.000537 0.001552 0.002653 0.001681 0.00143 0.000599 60 0.001913 0.000545 0.001581 0.000968 0.000566 0.000494

Table 23b Averages of the triplicate % bioluminescence values of the S. sonnei pLux biosensor in the presence of toluene (Fig. 4.25) Time Average bioluminescence (%) at various concentrations (min) 0 mg/l 0.01 mg/l 0.1 mg/l 1 mg/l 10 mg/l 100 mg/l 1 100 92.08499 90.53074 84.53115 40.49534 13.89384 15 66.3737 63.79548 68.00875 70.4553 48.13787 5.445615 30 53.6426 48.93388 52.52937 56.95892 48.95671 5.191011 45 42.99659 38.0725 43.72094 47.39776 45.43145 6.809068 60 35.24112 31.01058 35.13723 36.92442 40.86962 8.466632

- 103 - Table 24a Standard deviations of the triplicate bioluminescence values of the S. sonnei pLux biosensor in the presence of benzene (Fig. 4.26) Time Standard deviations of the various concentrations (min) 0 mg/l 0.01 mg/l 0.1 mg/l 1 mg/l 10 mg/l 100 mg/l 1 0.004786 0.013766 0.005523 0.008763 0.00643 0.001178 15 0.002298 0.00771 0.001552 0.00091 0.004037 0.000533 30 0.001974 0.009802 0.000884 0.000665 0.00158 0.000692 45 0.002692 0.007604 0.003335 0.001823 0.001165 0.002115 60 0.00095 0.0044 0.001624 0.000812 0.000675 6.78E-05

Table 24b Averages of the triplicate % bioluminescence values of the S. sonnei pLux biosensor in the presence of benzene (Fig. 4.26) Time Average bioluminescence (%) at various concentrations (min) 0 mg/l 0.01 mg/l 0.1 mg/l 1 mg/l 10 mg/l 100 mg/l 1 100 94.80449 98.22102 91.4078 53.43209 22.17888 15 65.95196 65.51565 66.05215 72.86556 55.82378 3.582569 30 51.52173 52.26618 52.31294 55.65444 57.74692 5.756102 45 43.16977 41.58877 41.73286 46.27972 50.05953 10.04333 60 34.13753 33.72509 34.78991 37.0471 45.63928 13.34999

Table 25a Standard deviations of the triplicate bioluminescence values of the E. coli DH5α uspA:lux biosensor in the presence of copper sulphate (Fig. 4.27) Time Standard deviations of the various concentrations (min) 0 mg/l 0.01 mg/l 0.1 mg/l 1 mg/l 10 mg/l 100 mg/l 1 0.000156 0.000367 0.00073 0.001175 0.000578 2.01E-05 15 0.001367 0.000857 0.000581 0.000285 0.000184 4.00E-05 30 0.000593 3.78E-05 0.00062 5.11E-05 0.000265 0.000161 45 0.000555 0.000221 0.000268 8.39E-05 0.000107 0.000183 60 0.002385 0.001243 0.000382 0.000422 0.000127 0.000119

Table 25b Averages of the triplicate % bioluminescence values of the E. coli DH5α uspA: lux biosensor in the presence of copper sulphate (Fig. 4.27) Time Average bioluminescence (%) at various concentrations (min) 0 mg/l 0.01 mg/l 0.1 mg/l 1 mg/l 10 mg/l 100 mg/l 1 100 95.93943 96.31417 64.56879 30.1078 8.480496 15 175.9035 158.6037 138.5678 21.40657 6.683782 5.487685 30 211.268 204.9949 198.3573 9.876797 5.636555 7.823412 45 274.6255 240.3439 251.5149 12.39159 8.018483 5.975361 60 401.1602 353.9684 354.3483 20.21561 5.621153 8.085219

- 104 - Table 26a Standard deviations of the triplicate bioluminescence values of the E. coli DH5α uspA: lux biosensor in the presence of cadmium acetate (Fig. 4.28) Time Standard deviations of the various concentrations (min) 0 mg/l 0.01 mg/l 0.1 mg/l 1 mg/l 10 mg/l 100 mg/l 1 0.000345 0.000239 0.000111 0.000266 0.000278 0.000121 15 5.86E-05 9.23E-05 0.00043 0.00045 0.00013 0.000144 30 5.42E-05 0.000777 0.000556 0.000267 7.98E-05 0.000124 45 0.00047 0.00013 0.001142 0.000488 0.000248 2.44E-05 60 0.000429 0.001129 0.000174 0.000356 0.000108 9.49E-05

Table 26b Averages of the triplicate % bioluminescence values of the E. coli DH5α uspA: lux biosensor in the presence of cadmium acetate (Fig. 4.28) Time Average bioluminescence (%) at various concentrations (min) 0 mg/l 0.01 mg/l 0.1 mg/l 1 mg/l 10 mg/l 100 mg/l 1 100 91.6504 95.95386 75.20852 22.54658 2.697426 15 118.6602 130.6744 104.0639 25.49246 5.323865 5.119787 30 157.7906 156.5484 106.8767 21.22449 3.726703 3.070098 45 221.9343 188.6868 161.6415 29.58296 7.515527 3.531497 60 395.1824 376.362 255.1996 68.42059 1.748001 2.999113

Table 27a Standard deviations of the triplicate bioluminescence values of the E. coli DH5α uspA:lux biosensor in the presence of zinc sulphate (Fig. 4.29) Time Standard deviations of the various concentrations (min) 0 mg/l 0.01 mg/l 0.1 mg/l 1 mg/l 10 mg/l 100 mg/l 1 0.001558 0.001873 0.000235 0.000278 0.00011 8.06E-05 15 0.001456 0.001417 0.000309 0.000263 5.61E-05 0.000289 30 0.001119 0.001854 0.000703 9.19E-05 8.20E-05 6.94E-05 45 0.001111 0.000513 0.000736 0.000391 0.000157 0.000106 60 0.001183 0.000629 0.001961 0.000743 0.000193 5.32E-05

Table 27b Averages of the triplicate % bioluminescence values of the E. coli DH5α uspA:lux biosensor in the presence of zinc sulphate (Fig. 4.29) Time Average bioluminescence (%) at various concentrations (min) 0 mg/l 0.01 mg/l 0.1 mg/l 1 mg/l 10 mg/l 100 mg/l 1 100 99.63091 97.49681 48.24209 15.82833 5.522172 15 159.3243 129.2623 93.48412 15.40245 1.746086 7.39602 30 204.4811 177.7126 132.3333 18.25108 4.765061 6.037953 45 252.4774 222.5807 194.6434 37.43434 5.692522 5.115227 60 354.2472 330.2938 315.0383 105.0206 6.577395 6.733548

- 105 - Table 28a Standard deviations of the triplicate bioluminescence values of the E. coli DH5α uspA:lux biosensor in the presence of lead acetate (Fig. 4.30) Time Standard deviations of the various concentrations (min) 0 mg/l 0.01 mg/l 0.1 mg/l 1 mg/l 10 mg/l 100 mg/l 1 0.001309 0.002275 0.000435 0.000462 0.000417 8.67E-05 15 0.001018 0.004364 0.000582 0.000337 0.00012 0.000243 30 0.000779 0.005871 0.000515 0.00011 0.000132 0.00013 45 0.001714 0.007138 0.000331 0.000219 9.70E-05 0.000143 60 0.000693 0.011217 0.000861 0.000285 0.000186 0.000135

Table 28b Averages of the triplicate % bioluminescence values of the E. coli DH5α uspA:lux biosensor in the presence of lead acetate (Fig. 4.30) Time Average bioluminescence (%) at various concentrations (min) 0 mg/l 0.01 mg/l 0.1 mg/l 1 mg/l 10 mg/l 100 mg/l 1 100 70.80996 95.55859 72.42181 40.25062 6.027996 15 173.3616 98.90354 141.3774 115.4363 53.13021 5.618719 30 228.2502 127.9016 182.4162 132.5198 68.48062 7.513519 45 283.6289 168.991 234.3543 180.7238 100.4901 5.810722 60 413.1931 247.6431 362.2409 262.3872 151.9731 6.442323

Table 29a Standard deviations of the triplicate bioluminescence values of the E. coli DH5α uspA:lux biosensor in the presence of potassium dichromate (Fig. 4.31) Time Standard deviation at various concentrations (min) 0 mg/l 0.01 mg/l 0.1 mg/l 1 mg/l 10 mg/l 100 mg/l 1 0.000832 0.000283 0.0005 0.00041 0.000484 7.51E-05 15 0.000725 0.000632 0.000217 0.000461 0.000479 5.32E-05 30 0.001533 0.000476 0.001646 0.000873 0.000302 0.000138 45 0.001098 0.001009 0.000185 0.000602 0.000586 2.67E-05 60 0.001063 0.000212 0.002194 0.000254 0.001021 8.39E-05

Table 29b Averages of the triplicate % bioluminescence values of the E. coli DH5α uspA:lux biosensor in the presence of potassium dichromate (Fig. 4.31) Time Average bioluminescence (%) at various concentrations (min) 0 mg/l 0.01 mg/l 0.1 mg/l 1 mg/l 10 mg/l 100 mg/l 1 100 94.46576 99.56574 85.48778 53.51949 3.938599 15 175.8786 146.6118 156.1654 162.6035 76.21693 5.741268 30 224.3638 189.7142 200.9594 197.2884 95.08685 7.038986 45 293.1478 251.434 267.9865 268.3147 138.6185 6.554234 60 423.0007 362.6643 404.3027 383.2208 221.9097 5.428197

- 106 - Table 30a Standard deviations of the triplicate bioluminescence values of the E. coli DH5α uspA:lux biosensor in the presence of toluene (Fig. 4.32) Time Standard deviations of the various concentrations (min) 0 mg/l 0.01 mg/l 0.1 mg/l 1 mg/l 10 mg/l 100 mg/l 1 0.000227 0.000534 0.000342 0.000352 0.000227 0.000193 15 0.000591 0.000149 0.000656 0.00021 0.000247 0.000327 30 0.000395 0.000623 0.000596 0.000517 0.000699 0.00102 45 0.000645 0.000474 0.000175 0.000561 0.001476 0.001365 60 0.001079 0.00094 0.000752 0.000806 0.000872 0.00044

Table 30b Averages of the triplicate % bioluminescence values of the E. coli DH5α uspA:lux biosensor in the presence of toluene (Fig. 4.32) Time Average bioluminescence (%) at various concentrations (min) 0 mg/l 0.01 mg/l 0.1 mg/l 1 mg/l 10 mg/l 100 mg/l 1 100 77.54678 77.81138 89.19864 106.6528 80.02268 15 149.7354 110.5651 110.9053 116.4147 213.7498 162.8142 30 171.4988 121.461 153.4587 154.5549 245.7664 258.7413 45 262.8142 160.4234 188.8868 226.6112 320.2986 398.7715 60 423.436 304.3848 317.9928 353.5438 485.7592 595.6634

Table 31a Standard deviations of the triplicate bioluminescence values of the E. coli DH5α uspA:lux in the presence of ethylbenzene (Fig. 4.33) Time Standard deviations of the various concentrations (min) 0 mg/l 0.01 mg/l 0.1 mg/l 1 mg/l 10 mg/l 100 mg/l 1 0.000683 0.000201 0.000155 0.000493 0.000556 0.000437 15 0.000347 0.000334 0.00033 0.000488 0.000215 0.000184 30 0.000397 0.000146 0.000355 0.0008 0.000149 0.000413 45 0.000333 0.000372 0.000418 0.001317 0.000466 0.000429 60 0.000863 0.000563 0.000767 0.001166 0.000815 0.000207

Table 31b Averages of the triplicate % bioluminescence values of the E. coli DH5α uspA:lux biosensor in the presence of ethylbenzene (Fig. 4.33) Time Average bioluminescence (%) at various concentrations (min) 0 mg/l 0.01 mg/l 0.1 mg/l 1 mg/l 10 mg/l 100 mg/l 1 100 85.27416 99.44389 89.2448 92.97075 69.45835 15 171.1044 129.6185 144.7336 165.1429 236.9369 126.8379 30 211.2891 166.4442 185.6301 191.5471 322.8784 204.2709 45 295.1396 214.8704 243.221 252.1299 450.406 306.8513 60 452.8195 394.8393 421.3769 441.2412 637.3163 482.9496

- 107 - Table 32a Standard deviations of the triplicate bioluminescence values of the E. coli DH5α uspA:lux biosensor in the presence of benzene (Fig. 4.34) Time Standard deviations of the various concentrations (min) 0 mg/l 0.01 mg/l 0.1 mg/l 1 mg/l 10 mg/l 100 mg/l 1 0.000486 0.000161 0.000989 0.000308 0.000763 0.001183 15 0.000206 0.000836 0.000212 0.000261 0.000721 0.001036 30 0.000681 0.000261 0.000673 0.000485 0.001705 0.000757 45 0.000704 0.000315 0.000308 0.0009 0.000782 0.000817 60 0.00089 0.00105 0.000343 0.000284 0.000909 0.001316

Table 32b Averages of the triplicate % bioluminescence values of the E. coli DH5α uspA:lux biosensor in the presence of benzene (Fig. 4.34) Time Average bioluminescence (%) at various concentrations (min) 0 mg/l 0.01 mg/l 0.1 mg/l 1 mg/l 10 mg/l 100 mg/l 1 100 73.85862 75.18409 98.664 122.3648 109.5624 15 159.8359 116.4317 117.6099 145.3819 249.3373 189.7644 30 194.1405 144.7822 158.8365 186.4927 289.2594 275.5838 45 253.3768 201.9356 188.6808 238.5689 397.7593 441.3002 60 418.178 343.825 377.5826 399.2005 580.9179 686.9982

Table 33a Standard deviations of the triplicate bioluminescence values of the V. fischeri-based BioTox kit in the presence of cadmium acetate (Fig. 4.35 ) Time Standard deviations of the various concentrations (min) 0 mg/l 0.01 mg/l 0.1 mg/l 1 mg/l 10 mg/l 100 mg/l 1 0.043666 0.036142 0.029048 0.039176 0.032606 0.052985 15 0.048174 0.012164 0.032512 0.022136 0.019403 0.013471 30 0.06094 0.015683 0.040192 0.027607 0.002539 0.005256 45 0.060141 0.010603 0.037445 0.028484 0.008657 0.002813 60 0.045627 0.009553 0.038983 0.025963 0.007262 0.001712

Table 33b Averages of the triplicate % bioluminescence values of the V. fischeri-based BioTox kit in the presence of cadmium acetate (Fig. 4.35) Time Average bioluminescence (%) at various concentrations (min) 0 mg/l 0.01 mg/l 0.1 mg/l 1 mg/l 10 mg/l 100 mg/l 1 100 102.0772 102.9575 103.9134 101.4793 88.00323 15 87.08879 87.50123 87.63202 76.23978 52.67485 15.4629 30 88.30845 87.83024 88.00048 65.03657 26.02532 3.097383 45 83.5435 82.65643 82.98918 56.3752 12.84947 0.892577 60 78.29043 76.45319 77.6902 52.57858 8.716087 0.448906

- 108 - Table 34a Standard deviations of the triplicate bioluminescence values of the V. fischeri-based BioTox kit in the presence of lead acetate (Fig. 4.36) Time Standard deviations of the various concentrations (min) 0 mg/l 0.01 mg/l 0.1 mg/l 1 mg/l 10 mg/l 100 mg/l 1 0.039165 0.118105 0.155228 0.222102 0.190185 0.039372 15 0.047706 0.103114 0.074609 0.027796 0.012721 0.01209 30 0.07617 0.129607 0.059607 0.016761 0.009004 0.009222 45 0.089229 0.05708 0.019785 0.052578 0.035713 0.110231 60 0.089531 0.117371 0.039989 0.010104 0.006196 0.000455

Table 34b Averages of the triplicate % bioluminescence values of the V. fischeri-based BioTox kit in the presence of lead acetate (Fig. 4.36) Time Average bioluminescence (%) at various concentrations (min) 0 mg/l 0.01 mg/l 0.1 mg/l 1 mg/l 10 mg/l 100 mg/l 1 100 96.56168 96.24383 88.00299 71.23044 28.4867 15 99.20058 97.09527 96.16883 69.00715 57.44169 10.0539 30 93.1941 92.31353 91.68701 63.45243 52.10552 9.477849 45 86.9408 85.78719 85.21956 57.82973 45.36581 8.154663 60 82.249 80.2443 80.39049 52.83144 40.56114 6.795185

Table 35a Standard deviations of the triplicate bioluminescence values of the V. fischeri-based BioTox kit in the presence of copper sulphate (Fig. 4.37) Time Standard deviations of the various concentrations (min) 0 mg/l 0.01 mg/l 0.1 mg/l 1 mg/l 10 mg/l 100 mg/l 1 0.133969 0.087558 0.171932 0.182672 0.029105 0.038099 15 0.105422 0.062522 0.13584 0.142235 0.008447 0.000169 30 0.102254 0.072038 0.133536 0.134095 0.002351 0.000718 45 0.094265 0.056655 0.115038 0.122504 0.002022 0.000636 60 0.092616 0.051389 0.110257 0.110644 0.001974 0.000324

Table 35b Averages of the triplicate % bioluminescence values of the V. fischeri-based BioTox kit in the presence of copper sulphate (Fig. 4.37) Time Average bioluminescence (%) at various concentrations (min) 0 mg/l 0.01 mg/l 0.1 mg/l 1 mg/l 10 mg/l 100 mg/l 1 100 98.49933 97.47142 96.9378 56.1128 25.24466 15 102.6694 104.609 101.8545 100.1265 8.019045 0.054583 30 97.37183 98.83701 97.36137 96.36021 0.86071 0.044621 45 90.51238 91.82859 90.5882 90.7271 0.271731 0.043939 60 85.24372 86.46273 85.22169 85.39337 0.256805 0.034022

- 109 - Table 36a Standard deviations of the triplicate bioluminescence values of the V. fischeri-based BioTox kit in the presence of potassium dichromate (Fig. 4.38) Time Standard deviations of the various concentrations (min) 0 mg/l 0.01 mg/l 0.1 mg/l 1 mg/l 10 mg/l 100 mg/l 1 0.177113 0.125622 0.058937 0.113347 0.115756 0.345804 15 0.108512 0.093427 0.037028 0.062479 0.057603 0.179731 30 0.074108 0.067403 0.024855 0.049472 0.045018 0.156925 45 0.093543 0.071381 0.023833 0.068335 0.033231 0.131529 60 0.095834 0.13151 0.052842 0.014709 0.008517 0.00345

Table 36b Averages of the triplicate % bioluminescence values of the V. fischeri-based BioTox kit in the presence of potassium dichromate (Fig. 4.38) Time Average bioluminescence (%) at various concentrations (min) 0 mg/l 0.01 mg/l 0.1 mg/l 1 mg/l 10 mg/l 100 mg/l 1 100 96.00648 98.24617 94.88559 88.06092 37.43551 15 100.6376 97.36821 100.3294 96.0915 84.22506 14.38998 30 94.96993 92.77269 95.49376 90.63662 77.66665 13.51388 45 88.43181 86.5262 88.89369 82.99446 67.50219 11.9753 60 83.67729 81.49528 83.5902 76.22637 59.38547 10.14388

Table 37a Standard deviations of the triplicate bioluminescence values of the V. fischeri-based BioTox kit in the presence of zinc sulphate (Fig. 4.39) Time Standard deviations of the various concentrations (min) 0 mg/l 0.01 mg/l 0.1 mg/l 1 mg/l 10 mg/l 100 mg/l 1 0.03457 0.008409 0.041495 0.053168 0.066268 0.114411 15 0.054707 0.026082 0.08785 0.036001 0.067764 0.004042 30 0.076434 0.0236 0.067888 0.035232 0.053549 0.000632 45 0.061895 0.016094 0.061487 0.046692 0.06652 0.000685 60 0.049584 0.028642 0.075233 0.041245 0.067195 0.000508

Table 37b Averages of the triplicate % bioluminescence values of the V. fischeri-based BioTox kit in the presence of zinc sulphate (Fig. 4.39) Time Average bioluminescence (%) at various concentrations (min) 0 mg/l 0.01 mg/l 0.1 mg/l 1 mg/l 10 mg/l 100 mg/l 1 100 97.29716 97.94101 95.81192 95.16075 62.97465 15 102.57 101.4247 102.1846 97.52347 81.41312 5.252932 30 97.66838 96.91502 97.37971 93.24132 73.22213 1.256759 45 90.83169 90.55546 90.90646 87.73435 69.33509 0.647085 60 85.85339 84.68945 85.86022 83.92372 68.31103 0.438606

- 110 - Table 38a Standard deviations of the triplicate bioluminescence values of the V. fischeri-based BioTox kit in the presence of ethylbenzene (Fig. 4.40) Time Standard deviations of the various concentrations (min) 0 mg/l 0.01 mg/l 0.1 mg/l 1 mg/l 10 mg/l 100 mg/l 1 0.063628 0.030754 0.070046 0.073162 0.041747 0.037128 15 0.050942 0.0398 0.031795 0.011784 0.111587 0.062504 30 0.046113 0.036264 0.02304 0.02298 0.094227 0.079455 45 0.022994 0.023379 0.024918 0.045307 0.094647 0.083795 60 0.031821 0.031931 0.01461 0.033692 0.073182 0.085158

Table 38b Averages of the triplicate % bioluminescence values of the V. fischeri-based BioTox kit in the presence of ethylbenzene (Fig. 4.40) Time Average bioluminescence (%) at various concentrations (min) 0 mg/l 0.01 mg/l 0.1 mg/l 1 mg/l 10 mg/l 100 mg/l 1 100 93.29021 97.93556 92.24595 50.50669 18.46012 15 101.9525 95.66139 99.66444 100.6672 77.62968 31.68742 30 95.8947 92.15931 94.66753 96.89347 89.75258 44.28517 45 87.7054 84.593 86.7199 88.5371 89.39188 53.33458 60 80.25563 78.20614 79.71974 81.29329 85.7658 58.7852

Table 39a Standard deviations of the triplicate bioluminescence values of the V. fischeri-based BioTox kit in the presence of toluene (Fig. 4.41) Time Standard deviations of the various concentrations (min) 0 mg/l 0.01 mg/l 0.1 mg/l 1 mg/l 10 mg/l 100 mg/l 1 0.15281 0.079321 0.112528 0.073133 0.064355 0.044942 15 0.124509 0.067842 0.100982 0.03149 0.066308 0.057316 30 0.12068 0.056216 0.085748 0.035859 0.053791 0.057835 45 0.11301 0.048572 0.088804 0.035883 0.068986 0.071877 60 0.096365 0.053093 0.086558 0.045396 0.065831 0.091053

Table 39b Averages of the triplicate % bioluminescence values of the V. fischeri-based BioTox kit in the presence of toluene (Fig. 4.41) Time Average bioluminescence (%) at various concentrations (min) 0 mg/l 0.01 mg/l 0.1 mg/l 1 mg/l 10 mg/l 100 mg/l 1 100 95.04492 97.30009 96.86059 69.69621 28.14404 15 105.8532 100.68 103.0797 106.5624 96.77114 40.45504 30 100.1656 96.94317 99.64562 102.2243 104.0238 53.48269 45 91.54763 89.45949 91.0882 93.88556 100.3166 63.67083 60 83.7689 82.46889 83.65539 85.73814 93.44754 70.82604

- 111 - Table 40a Standard deviations of the triplicate bioluminescence values of the V. fischeri-based BioTox kit in the presence of benzene (Fig. 4.42) Time Standard deviations of the various concentrations (min) 0 mg/l 0.01 mg/l 0.1 mg/l 1 mg/l 10 mg/l 100 mg/l 1 0.100196 0.139108 0.041159 0.102119 0.066346 0.038726 15 0.176565 0.144076 0.088334 0.162771 0.142152 0.124181 30 0.158454 0.117007 0.085748 0.163106 0.119923 0.122452 45 0.137203 0.107474 0.099781 0.155776 0.116513 0.127384 60 0.13994 0.084511 0.077734 0.151481 0.118145 0.146679

Table 40b Averages of the triplicate % bioluminescence values of the V. fischeri-based BioTox kit in the presence of benzene (Fig. 4.42) Time Average bioluminescence (%) at various concentrations (min) 0 mg/l 0.01 mg/l 0.1 mg/l 1 mg/l 10 mg/l 100 mg/l 1 100 93.61097 93.85806 96.33307 89.40364 52.02073 15 102.6575 96.72831 99.79974 102.2907 103.4715 67.83073 30 97.88088 93.29136 95.81139 97.93332 104.7343 85.13677 45 89.90441 85.72033 88.15149 90.01559 97.7917 94.5487 60 82.64593 79.24339 80.99279 82.54065 90.38113 97.8948

Table 41a Standard deviations of the triplicate bioluminescence values of the S. sonnei pLux biosensor in the presence the New Germany wastewater works effluent sample (Fig. 5.2)

Time Standard deviations of the various concentrations (min) 0% 0.01% 0.10% 1% 10% 100% 1 0.00481 0.004774 0.006152 0.000923 0.00238 0.008868 15 0.004068 0.003064 0.001326 0.002816 0.005208 0.003707 30 0.004703 0.002973 0.001756 0.001006 0.00154 0.000931 45 0.003933 0.00135 0.002438 0.002267 0.000853 0.002419 60 0.002458 0.000917 0.001992 0.002357 0.002638 0.001808

Table 42b Averages of the triplicate % bioluminescence values of the S. sonnei pLux biosensor in the presence of the New Germany wastewater works effluent sample (Fig. 5.2)

Time Average bioluminescence (%) at various concentrations (min) 0% 0.01% 0.10% 1% 10% 100% 1 100 92.14433 90.45577 89.30374 98.26815 90.12322 15 61.65723 52.81878 52.51571 51.89105 55.54611 52.00337 30 48.97317 41.63136 42.15772 42.80148 44.82737 41.91 45 43.02453 35.59875 34.50832 35.99534 37.41477 35.4475 60 35.44292 30.3303 29.06832 29.97434 31.67966 30.63995

- 112 - Table 43a Standard deviations of the triplicate bioluminescence values of the S. sonnei pLux biosensor in the presence of the Kwa-Mashu wastewater works effluent sample (Fig. 5.3) Time Standard deviations of the various concentrations (min) 0% 0.01% 0.10% 1% 10% 100% 1 0.005832 0.00159 0.001507 0.002985 0.006569 0.004506 15 0.002117 0.000856 0.001034 0.002924 0.003032 0.001658 30 0.000688 0.003977 0.001481 0.000445 0.00131 0.000797 45 0.003143 0.002304 0.00352 0.000977 0.001348 0.00124 60 0.001803 0.001691 0.000459 0.001838 0.003361 0.002702

Table 43b Averages of the triplicate % bioluminescence values of the S. sonnei pLux biosensor in the presence of the Kwa-Mashu wastewater works effluent sample (Fig. 5.3) Time Average bioluminescence (%) at various concentrations (min) 0% 0.01% 0.10% 1% 10% 100% 1 100 97.56234 103.4877 96.12654 96.99852 103.5153 15 62.82721 61.59166 67.12892 60.61249 60.63618 63.08446 30 48.82454 50.91636 52.94905 49.24053 47.24497 46.98603 45 43.75395 43.13771 44.43999 42.46648 39.54449 37.44121 60 37.34003 37.00788 37.87729 35.03312 32.30823 30.57069

Table 44a Standard deviations of the triplicate bioluminescence values of the S. sonnei pLux biosensor in the presence of the Northern wastewater works effluent sample (Fig. 5.4) Time Standard deviations of the various concentrations (min) 0% 0.01% 0.10% 1% 10% 100% 1 0.006553 0.002096 0.004616 0.002494 0.003696 0.008853 15 0.001609 0.001352 0.000586 0.001667 0.000585 0.00344 30 0.00042 0.00082 0.001037 0.002169 0.00163 0.003564 45 0.001455 0.003023 0.000945 0.002031 0.00194 0.001542 60 0.000749 0.001712 0.001732 0.00183 0.001105 0.00047

Table 44b Averages of the triplicate % bioluminescence values of the S. sonnei pLux biosensor in the presence of the Northern wastewater works effluent sample (Fig. 5.4) Time Average bioluminescence (%) at various concentrations (min) 0% 0.01% 0.10% 1% 10% 100% 1 100 89.10873 103.0301 103.8897 102.6418 87.29975 15 72.0437 45.25008 64.00307 71.35805 75.42483 59.12033 30 57.33754 33.75844 47.76214 57.15676 56.77959 45.97002 45 47.58513 25.50647 37.54224 42.71117 45.18085 37.41072 60 38.19428 19.62012 29.10887 35.66595 36.28054 32.09603

- 113 - Table 45a Standard deviations of the triplicate bioluminescence values of the S. sonnei pLux biosensor in the presence of the Phoenix wastewater works effluent sample (Fig. 5.5) Time Standard deviations of the various concentrations (min) 0% 0.01% 0.10% 1% 10% 100% 1 0.000715 0.003176 0.002303 0.002782 0.003586 0.001612 15 0.001019 0.000173 0.002205 0.001247 0.001926 0.00114 30 0.000598 0.001163 0.001261 0.001405 0.000969 0.001588 45 0.000591 0.001086 0.002356 0.003082 0.002127 0.000986 60 0.000525 0.000477 0.001379 0.001264 0.000209 0.002304

Table 45b Averages of the triplicate % bioluminescence values of the S. sonnei pLux biosensor in the presence of the Phoenix wastewater works effluent sample (Fig. 5.5) Time Average bioluminescence (%) at various concentrations (min) 0% 0.01% 0.10% 1% 10% 100% 1 100 84.17832 100.8986 108.2896 102.2641 86.93411 15 70.44223 45.99748 58.07525 71.50591 70.78145 62.86456 30 55.31942 33.27297 48.29375 53.82792 54.04076 49.82037 45 45.11184 25.29739 36.40273 42.44118 43.87709 40.83544 60 37.93806 23.18268 29.40569 35.75472 36.89135 34.65802

Table 46a Standard deviations of the triplicate bioluminescence values of the S. sonnei pLux biosensor in the presence of the Amanzimtoti wastewater works effluent sample (Fig. 5.6) Time Standard deviations of the various concentrations (min) 0% 0.01% 0.10% 1% 10% 100% 1 0.002774 0.000289 0.000816 0.002434 0.002576 0.002162 15 0.001609 0.001801 0.001137 0.000649 0.001344 0.001984 30 0.000271 0.000251 0.001769 0.000662 0.000256 0.002049 45 0.000349 0.000621 0.001851 0.001443 0.001342 0.001128 60 0.000263 0.000151 0.000898 0.000407 0.000919 0.000377

Table 46b Averages of the triplicate % bioluminescence of the S. sonnei pLux biosensor in the presence of the Amanzimtoti wastewater works effluent sample (Fig. 5.6) Time Average bioluminescence (%) at various concentrations (min) 0% 0.01% 0.10% 1% 10% 100% 1 100 86.19989 93.02849 110.7368 103.0633 91.45279 15 79.67953 44.88554 59.78503 78.341 83.68274 70.96754 30 62.05451 26.06148 42.04103 57.87356 66.13027 57.21998 45 49.32635 16.65356 31.22349 45.83984 50.5308 48.14571 60 41.73814 13.03928 22.96812 35.98283 41.64312 37.99949

- 114 - Table 47a Standard deviations of the triplicate bioluminescence values of the E. coli DH5α uspA:lux biosensor in the presence the New Germany wastewater works effluent sample (Fig. 5.7) Time Standard deviations of the various concentrations (min) 0% 0.01% 0.10% 1% 10% 100% 1 0.000114 0.000765 0.000693 0.000393 8.76E-05 0.000442 15 0.000381 0.000263 0.000516 0.00035 0.000594 0.000604 30 0.000407 0.001104 0.000273 9.01E-05 0.000617 0.000733 45 0.000755 0.000526 0.000216 0.000632 0.000427 0.000983 60 0.002666 0.001025 0.00205 0.000749 0.0007 0.00239

Table 47b Averages of the triplicate % bioluminescence values of the E. coli DH5α uspA:lux biosensor in the presence of the New Germany wastewater works effluent sample (Fig. 5.7) Time Average bioluminescence (%) at various concentrations (min) 0% 0.01% 0.10% 1% 10% 100% 1 100 97.02488 106.7823 86.22592 92.20911 87.21887 15 132.4281 115.3466 136.9652 134.066 130.0603 102.8685 30 164.4063 136.4079 149.7056 145.4468 186.3702 139.0393 45 223.9837 197.6136 231.4938 215.1235 262.5732 196.0282 60 382.297 377.6125 445.522 353.2801 415.99 365.213

Table 48a Standard deviations of the triplicate bioluminescence values of the E. coli DH5α uspA:lux biosensor in the presence of the Kwa-Mashu wastewater works effluent sample (Fig. 5.8) Time Standard deviations of the various concentrations (min) 0% 0.01% 0.10% 1% 10% 100% 1 0.000641 0.000411 0.000678 0.000327 0.000304 0.000292 15 0.000218 0.000437 9.04E-05 0.00051 0.000531 8.29E-05 30 0.000522 0.000381 0.000302 0.000128 0.000231 0.000258 45 0.000599 0.00023 0.000372 0.000419 0.00067 0.000586 60 0.000943 0.000344 0.000627 0.000968 0.00087 0.000351

Table 48b Averages of the triplicate % bioluminescence values of the E. coli DH5α uspA:lux biosensor in the presence of the Kwa-Mashu wastewater works effluent sample (Fig. 5.8) Time Average bioluminescence (%) at various concentrations (min) 0% 0.01% 0.10% 1% 10% 100% 1 100 94.73915 87.40319 95.41137 53.16382 48.85284 15 144.2642 138.7988 164.9423 130.7614 100.7891 71.44527 30 188.9961 152.1116 175.8732 159.2869 119.5236 118.2084 45 217.0101 206.4153 254.8736 222.0166 196.6535 100.263 60 376.1216 327.1811 349.5835 386.0734 310.9747 108.9288

- 115 - Table 49a Standard deviations of the triplicate bioluminescence values of the E. coli DH5α uspA:lux biosensor in the presence of the Northern wastewater works effluent sample (Fig. 5.9) Time Standard deviations of the various concentrations (min) 0% 0.01% 0.10% 1% 10% 100% 1 0.000603 0.000385 0.000366 0.0009 0.000338 0.000247 15 0.000228 0.00053 0.000408 0.000693 0.000658 0.000385 30 0.000486 0.000413 0.000712 0.000295 0.000486 0.000508 45 0.000578 0.001164 0.000352 0.000628 0.000595 0.000222 60 0.000481 0.000225 0.000619 0.000418 0.000378 0.00112

Table 49b Averages of the triplicate % bioluminescence values of the E. coli DH5α uspA:lux biosensor in the presence of the Northern wastewater works effluent sample (Fig. 5.9) Time Average bioluminescence (%) at various concentrations (min) 0% 0.01% 0.10% 1% 10% 100% 1 100 60.30658 93.98951 87.72354 98.76294 56.05755 15 150.5446 112.7336 143.055 171.2518 162.7404 119.161 30 207.1265 167.4734 193.734 212.1823 204.518 166.7205 45 238.537 228.3179 249.3075 255.7214 271.7763 199.2605 60 368.5491 337.1521 390.4531 367.1777 488.6244 387.105

Table 50a Standard deviations of the triplicate bioluminescence values of the E. coli DH5α uspA:lux biosensor in the presence of the Phoenix wastewater works effluent sample (Fig. 5.10) Time Standard deviations of the various concentrations (min) 0% 0.01% 0.10% 1% 10% 100% 1 0.000267 0.000548 8.74E-05 0.000557 0.000488 0.000335 15 0.001159 5.05E-05 0.000208 0.000463 0.000906 0.000573 30 0.000114 0.000536 0.000267 0.000638 0.000256 0.000573 45 0.000642 0.000471 0.000218 0.000158 0.000859 0.00064 60 0.000405 0.00027 0.00085 0.000395 0.000685 0.000942

Table 50b Averages of the triplicate % bioluminescence values of the E. coli DH5α uspA:lux biosensor in the presence of the Phoenix wastewater works effluent sample (Fig. 5.10) Time Average bioluminescence (%) at various concentrations (min) 0% 0.01% 0.10% 1% 10% 100% 1 100 100.0595 85.26118 95.75776 57.6073 75.92427 15 148.9642 135.9005 138.1505 151.2638 123.144 99.48459 30 179.8493 161.0665 158.5985 211.9536 144.7914 114.4712 45 262.5136 254.3067 247.2396 259.933 208.9305 170.79 60 411.5968 356.408 423.937 467.3703 402.359 264.3473

- 116 -

Table 51a Standard deviations of the triplicate bioluminescence values of the E. coli DH5α uspA:lux biosensor in the presence of the Amanzimtoti wastewater works effluent sample (Fig. 5.11) Time Standard deviations of the various concentrations (min) 0% 0.01% 0.10% 1% 10% 100% 1 0.000668 0.000171 0.000513 0.000337 0.000491 0.000317 15 0.000345 0.00051 0.00095 0.000581 0.000478 0.000197 30 0.000683 0.00099 0.000635 0.000669 0.000865 0.00066 45 0.001368 0.000818 0.00071 0.001383 0.000431 0.000419 60 0.001675 0.000603 0.000708 0.00071 0.000976 0.000434

Table 51b Averages of the triplicate % bioluminescence of the E. coli DH5α uspA:lux biosensor in the presence of the Amanzimtoti wastewater works effluent sample (Fig. 5.11) Time Average bioluminescence (%) at various concentrations (min) 0% 0.01% 0.10% 1% 10% 100% 1 100 73.48417 105.5267 92.38061 86.82704 68.27938 15 127.2581 126.8288 154.5609 140.6367 104.3731 94.96512 30 178.3134 170.1574 161.7689 165.7038 131.7922 120.9801 45 240.3148 228.5012 248.4886 260.8925 198.2561 137.0327 60 385.6555 322.7508 394.0351 421.1948 380.9247 186.8449

Table 52a Standard deviations of the triplicate bioluminescence values of the V. fischeri-based BioToxTM kit in the presence the New Germany wastewater works effluent sample (Fig. 5.12) Time Standard deviations of the various concentrations (min) 0% 0.01% 0.10% 1% 10% 100% 1 0.019222 0.011939 0.026715 0.023688 0.065012 0.078782 15 0.039659 0.016472 0.025078 0.033408 0.070208 0.075015 30 0.025141 0.015417 0.018131 0.049129 0.078088 0.066738 45 0.02711 0.015837 0.014918 0.035094 0.07153 0.06879 60 0.024951 0.018227 0.0277 0.035981 0.062705 0.073901

Table 52b Averages of the triplicate % bioluminescence values of the V. fischeri-based BioToxTM kit in the presence of the New Germany wastewater works effluent sample (Fig. 5.12) Time Average bioluminescence (%) at various concentrations (min) 0% 0.01% 0.10% 1% 10% 100% 1 100 101.2387 102.1093 102.3931 102.4862 85.08338 15 91.10797 91.4985 93.61184 93.63522 94.59952 72.35285 30 91.4923 91.81706 93.10928 93.44772 95.15501 77.16373 45 86.54718 85.98829 87.20013 88.14919 91.44951 77.94075 60 81.4192 81.68154 82.44091 83.29969 87.4343 78.10448

- 117 - Table 53a Standard deviations of the triplicate bioluminescence values of the V. fischeri-based BioToxTM kit biosensor in the presence of the Kwa-Mashu wastewater works effluent sample (Fig. 5.13) Time Standard deviations of the various concentrations (min) 0% 0.01% 0.10% 1% 10% 100% 1 0.058212 0.088496 0.084893 0.040129 0.075191 0.052361 15 0.03034 0.057934 0.054482 0.013047 0.036658 0.022507 30 0.031101 0.052541 0.055496 0.016418 0.039121 0.035898 45 0.026028 0.053959 0.045655 0.018569 0.025852 0.025587 60 0.039316 0.045029 0.048827 0.028826 0.027269 0.026693

Table 53b Averages of the triplicate % bioluminescence values of the V. fischeri-based BioToxTM kit in the presence of the Kwa-Mashu wastewater works effluent sample (Fig. 5.13) Time Average bioluminescence (%) at various concentrations (min) 0% 0.01% 0.10% 1% 10% 100% 1 100 102.6704 101.4519 100.1049 111.6667 127.5394 15 91.91549 91.39477 91.10013 90.87941 101.251 104.6561 30 88.51335 88.01868 86.763 86.02432 91.24139 74.58518 45 81.12068 79.00643 78.13989 77.27393 78.35241 52.60185 60 75.03481 72.68416 72.0421 70.49709 68.35434 40.32266

Table 54a Standard deviations of the triplicate bioluminescence values of the V. fischeri-based BioToxTM kit in the presence of the Northern wastewater works effluent sample (Fig. 5.14) Time Standard deviations of the various concentrations (min) 0% 0.01% 0.10% 1% 10% 100% 1 0.043666 0.02263 0.072191 0.083454 0.064499 0.093264 15 0.031214 0.033869 0.048527 0.065589 0.040455 0.06773 30 0.022358 0.036232 0.047542 0.057024 0.024103 0.068407 45 0.019384 0.034195 0.035212 0.051429 0.033986 0.057145 60 0.019538 0.026121 0.044907 0.040358 0.034875 0.058112

Table 54b Averages of the triplicate % bioluminescence values of the V. fischeri-based BioToxTM kit in the presence of the Northern wastewater works effluent sample (Fig. 5.14) Time Average bioluminescence (%) at various concentrations (min) 0% 0.01% 0.10% 1% 10% 100% 1 100 104.4392 101.7518 100.7805 115.0385 128.0918 15 86.42951 83.6782 83.24451 84.35097 98.10919 108.4828 30 82.60211 79.82047 79.40336 79.58429 90.04806 93.48833 45 74.67829 71.72877 70.98586 70.64823 79.26104 72.39234 60 68.97184 65.19956 65.04527 64.55653 71.31532 59.25386

- 118 - Table 55a Standard deviations of the triplicate bioluminescence values of the V. fischeri-based BioToxTM kit in the presence of the Phoenix wastewater works effluent sample (Fig. 5.15) Time Standard deviations of the various concentrations (min) 0% 0.01% 0.10% 1% 10% 100% 1 0.111095 0.018294 0.04291 0.091857 0.08909 0.052049 15 0.061089 0.033034 0.026655 0.068874 0.091095 0.066337 30 0.066189 0.020921 0.013056 0.043829 0.092549 0.064276 45 0.067391 0.01556 0.023589 0.056951 0.080316 0.066695 60 0.073646 0.032452 0.013961 0.047827 0.094537 0.055337

Table 55b Averages of the triplicate % bioluminescence values of the V. fischeri-based BioToxTM kit in the presence of the Phoenix wastewater works effluent sample (Fig. 5.15) Time Average bioluminescence (%) at various concentrations (min) 0% 0.01% 0.10% 1% 10% 100% 1 100 98.87827 98.49222 96.85385 104.2554 124.6555 15 97.94628 93.91052 95.08748 95.30577 102.2438 118.6924 30 96.85441 93.90149 94.34695 94.20031 99.94888 109.6263 45 90.24003 87.67247 88.52874 87.76075 93.21279 95.15519 60 85.18672 82.21121 83.18037 82.36713 86.94832 84.56107

Table 56a Standard deviations of the triplicate bioluminescence values of the V. fischeri-based BioToxTM kit in the presence of the Amanzimtoti wastewater works effluent sample (Fig. 5.16) Time Standard deviations of the various concentrations (min) 0% 0.01% 0.10% 1% 10% 100% 1 0.139141 0.143572 0.05827 0.096873 0.133775 0.078678 15 0.049494 0.11558 0.059096 0.041483 0.079401 0.023058 30 0.040927 0.112763 0.058285 0.039004 0.087039 0.033806 45 0.031773 0.107875 0.048373 0.040378 0.090179 0.023294 60 0.026863 0.098719 0.04341 0.03525 0.106027 0.009213

Table 56b Averages of the triplicate % bioluminescence of the V. fischeri-based BioToxTM kit in the presence of the Amanzimtoti wastewater works effluent sample (Fig. 5.16) Time Average bioluminescence (%) at various concentrations (min) 0% 0.01% 0.10% 1% 10% 100% 1 100 102.666 99.66266 100.6737 108.2135 123.601 15 88.91507 86.99533 88.44233 89.53518 99.74889 111.0952 30 88.5611 86.68462 88.27167 88.51776 95.33721 86.99811 45 82.49198 80.852 81.95163 82.19886 85.09602 65.14977 60 77.02223 75.7725 77.18879 77.72297 78.13062 52.12483

- 119 - APPENDIX 2

POSTGRADUATE STUDENTS ASSOCIATED WITH THE PROJECT

PhD Thesis submitted to the University of KwaZulu-Natal (Westville Campus), Faculty of Science and Agriculture

Olaniran, A. 2005. Biodegradation of cis- and trans- dichloroethenes by indigenous African bacterial isolates.

M.Sc Dissertations submitted to the University of Durban-Westville, Faculty of Science and Engineering

Hira, K.G. 2002. Application of bioluminescence to determine the fate of plasmids in viable but non-culturable bacteria.

Naidu, Y. 2002. Enumeration and identification of bacterial pathogens in groundwater supplies.

Lazarus, E.D. 2002. Bioremediation of transformer oil by a microbial consortium isolated from contaminated soil.

Seepersad, D. 2003. Development and application of prokaryotic biosensor systems for the evaluation of toxicity of environmental water samples.

Moodley, R. 2003. Molecular epidemiology of clinical and environmental isolates of Vibrio cholerae isolated from local outbreaks of cholera is KZN, SA (2000-2001).

B.Sc. (Honours) Dissertations submitted to the University of Durban- Westville, Faculty of Science and Engineering

Naiker, S. S. 2001. Luminescence-based detection of chemical pollutants using Pseudomonas putida carrying the lux gene.

- 120 - Moodley, K. 2002. The analysis of genetic differences between pathogenic and non- pathogenic strains of Vibrio cholerae.

Singh, S. 2002. Evaluation of the sensitivity of prokaryotic biosensor systems to heavy metal and chemical pollutants.

Ramlall, J. 2002. Bioremediation of transformer oil using a bacterial consortium.

Abbai, N.S. 2002. Characterisation of plasmids in Vibrio chlorae.

Sanpal, N. D. 2003. The evaluation of prokaryotic pLux biosensor systems for the detection of chemical and heavy metal pollutants in wastewater effluent.

Mfumo, N. H. 2003. Aerobic biodegradation of dichloroethenes by bacterial consortia.

Singh R. 2003. PCR characterization of the toxin gene cassette of Vibrio cholerae.

Govender, T. 2004. Detection of pollutants in a paint.

Letsoala, S. S. 2004. Detection of pesticides using the luminescent pLux system.

Rajpal, D. A. 2004. Antimicrobial properties of ursolic acid from Calistemon viminalis.

Moodley, M. 2004. Application of the lux CDABE operon to monitor chemical compounds present in brewery effluent.

Pursutham, S. 2004. Application of S. flexneri biosensor for detecting heavy metals in explosive and chemical industrial effluent.

Singh, S. 2004. Development of a bacterial biosensor that can detect pollutants in pulp and paper industry effluent.

Naidoo, N. 2004. Use of a whole cell biosensor for evaluating the efficiency of a wastewater treatment plant for heavy metal reduction.

- 121 - Tshabe, V. E. 2004. Detection of chlorinated compounds using biosensor system constructed with environmental bacterial isolates.

Naidoo, S. 2004. Aerobic biodegradation of 1,3-dichloropropene by bacteria isolated from wastewater effluent.

Masango, M. G. 2004. Aerobic biodegradation of 1,2-dichloroethane by bacteria isolated from wastewater.

Toolsi, R. 2005. The impact of antimicrobial soaps on the development of antibiotic resistance.

Muendi, P. 2005. Microbial degradation of 1, 2-dichloroethane by bacterial consortia obtained from industrial effluents.

Buthelezi, S. P. 2005. The isolation and characterization of bacterial bioflocculant.

Nkomo, P. 2005. Microbiological quality of bottled water.

Current Students: PhD Govender, A. Characterisation of chlorinated-hydrocarbon-degrading genes of bacteria.

MSc Rajpal, D.A. Bioremediation of soil contaminated with a mixture of chlorinated aliphatic hydrocarbons.

Buthelezi, S. P. Application of bacterial bioflocculants for wastewater and river water treatment.

- 122 - APPENDIX 3

PUBLICATIONS AND CONFERENCE PROCEEDINGS

Hira, K.G., D. Pillay and B. Pillay. Fate of plasmids in viable but non-culturable cells. Xth International Congress of Bacteriology and Applied Microbiology. 27th July – 1st August 2002. Paris, Le Palais des Congres de Paris.

Hira, K.G., B. Pillay and D. Pillay. Fate of plasmids in viable but non-culturable (VBNC) cells. 12th Biennial Congress of the South African Society for Microbiology. 2-5 April 2002. University of the Free State, Bloemfontein.

Lazarus, E.D., D. Pillay and B. Pillay. Biodegradation of transformer oil by a consortium of bacteria isolated from oil-contaminated soil. 12th Biennial Congress of the South African Society for Microbiology. 2-5 April 2002. University of the Free State, Bloemfontein.

Moodley, R., D. Pillay and B. Pillay. Molecular typing of Vibrio cholerae strains isolated during cholera outbreaks in KwaZulu-Natal, South Africa (2000-2001). 12th Biennial Congress of the South African Society for Microbiology. 2-5 April 2002. University of the Free State, Bloemfontein.

Seepersad, D., B. Pillay and D. Pillay. The design and application of prokaryotic and eukaryotic biosensor systems for the evaluation of ecotoxicity testing. 12th Biennial Congress of the South African Society for Microbiology. 2-5 April 2002. University of the Free State, Bloemfontein.

Naidu, Y., D. Pillay and B. Pillay. Enumeration and identification of viable but non- culturable bacteria in rural groundwater supplies. 12th Biennial Congress of the South African Society for Microbiology. 2-5 April 2002. University of the Free State, Bloemfontein.

- 123 -