Engineering bacterial magnetic nanoparticles

ENGINEERING BACTERIAL MAGNETIC NANOPARTICLES By Walter Nevondo A thesis submitted in partial fulfillment of the requirements for the degree of Magister Scientiae (M.Sc.) Department of Biotechnology University of the Western Cape Bellville Supervisor: Professor D.A. Cowan January 2013 i TABLE OF CONTENTS Table of contents...................................................................................................................................................ii Abstract ................................................................................................................................................................ iv Acknowledgments...............................................................................................................................................vii List of Figures ....................................................................................................................................................viii List of Tables ........................................................................................................................................................ ix List of Abbreviations ............................................................................................................................................ x CHAPTER 1 1. Introduction.................................................................................................................................................... 1 1.1. Magnetic nanoparticles (MNPs) ................................................................................................................ 1 1.2. Magnetotactic bacteria .......................................................................................................................... 4 1.3. Improving MNPs................................................................................................................................... 8 1.4. Metals and bacteria.................................................................................................................................. 10 1.4.1. 1.4.1.1. Non-specific Ni2+ and Co2+ uptake in bacteria ........................................................................... 14 1.4.1.2. Ni2+ ABC type transporters ........................................................................................................ 15 1.4.1.3. Ni2+ uptake permeases ................................................................................................................ 17 1.4.1.4. Co2+ ABC type transporters........................................................................................................ 19 1.4.1.5. Co2+ uptake permeases ............................................................................................................... 20 1.4.2. 1.5. Metals uptake in bacteria ................................................................................................................ 11 Metals efflux in bacteria ................................................................................................................. 21 1.4.2.1. P-type ATPase efflux systems.................................................................................................... 22 1.4.2.2. Ni2+ and Co2+ specific RND driven efflux systems .................................................................... 23 Study rationale and aim ........................................................................................................................... 26 1.5.1. The specific objectives are to:......................................................................................................... 27 CHAPTER 2 2. Materials and methods ................................................................................................................................. 28 2.1. Chemicals and reagents ........................................................................................................................... 28 2.2. Bacterial strains and plasmids ................................................................................................................. 28 2.3. Growth media .......................................................................................................................................... 30 2.4. Ni2+ and Co2+ tolerance............................................................................................................................ 30 2.5. Chemical competent cells........................................................................................................................ 31 2.6. Genomic DNA of source organisms........................................................................................................ 31 2.7. Plasmid DNA isolation............................................................................................................................ 32 2.8. PCR ......................................................................................................................................................... 33 ii 2.9. Agarose gel electrophoresis..................................................................................................................... 35 2.10. Cloning and transformation ..................................................................................................................... 35 2.10.1. Cloning of PCR product in pJET 1.2/ blunt.................................................................................... 35 2.10.2. Restriction digests........................................................................................................................... 35 2.10.3. Sub-cloning in pBAD ..................................................................................................................... 36 2.10.4. Sequencing...................................................................................................................................... 36 2.10.5. Transformation ............................................................................................................................... 36 2.11. Expression and SDS-PAGE..................................................................................................................... 37 2.11.1. Gene expression.............................................................................................................................. 37 2.11.2. Cell lysis ......................................................................................................................................... 37 2.11.3. SDS-PAGE ..................................................................................................................................... 38 2.12. Metal accumulation experiment .............................................................................................................. 39 2.12.1. Accumulation at stationary phase ................................................................................................... 39 2.12.2. Accumulation at exponential phase ................................................................................................ 39 2.13. Nitric acid digest and ICP-OES analysis ................................................................................................. 40 CHAPTER 3 3. Results and discussions................................................................................................................................ 41 3.1. Introduction ............................................................................................................................................. 41 3.2. Ni2+ and Co2+ tolerance............................................................................................................................ 44 3.3. Growth curves in the presence of Ni2+ or Co2+ ........................................................................................ 46 3.4. 16S rRNA PCR Analysis......................................................................................................................... 49 3.5. Cloning of the metal uptake genes........................................................................................................... 50 3.6. Expression of metal uptake proteins........................................................................................................ 51 3.7. Ni2+ and Co2+ accumulation..................................................................................................................... 53 3.7.1. Ni2+ accumulation....................................................................................................................... 56 3.7.1.1. Ni2+ accumulation at stationary phase ........................................................................................ 56 3.7.1.2. Intracellular iron and magnesium at different Ni2+ concentrations............................................. 60 3.7.1.3. Intracellular Ni2+, iron and magnesium in conditions for magnetosome doping ........................ 64 3.7.1.4. Suitable strain for Ni2+ doping of the magnetosome .................................................................. 67 3.7.2. Co2+ accumulation .......................................................................................................................... 68 3.7.2.1. Co2+ accumulation at stationary phase........................................................................................ 68 3.7.2.2. Intracellular iron and magnesium at stationary phase in the presence of Co2+ ........................... 70 3.7.2.3. Intracellular Co2+, iron and magnesium at exponential phase .................................................... 72 3.7.2.4. Suitable strain for Co2+ doping of the magnetosomes ................................................................ 74 4. Conclusions.................................................................................................................................................. 75 5. References.................................................................................................................................................... 78 6. Appendix.................................................................................................................................................... 100 iii ABSTRACT Magnetosomes, produced by magnetotactic bacteria (MTB), are the most attractive alternative source of non-toxic biocompatible magnetic nanoparticles (MNPs). A magnetosome contains Fe2O4 magnetite with properties superior to MNPs synthesized by the traditional chemical route. However, synthesis of magnetosomes on large scale has not been achieved yet because magnetotactic bacteria are fastidious to grow. In addition, magnetosomes are generally “soft” magnetic materials which can only be used for some applications, while other applications require “hard” magnetic materials. Here at the Institute of Microbial Biotechnology and Metagenomic (IMBM), a study is being conducted on cloning and expression of the magnetosome gene island (MIA), the genetic machinery for magnetosome formation, in an easy to culture E. coli strain. The magnetic properties of the magnetosome can be manipulated by doping with divalent metals such as Ni2+ or Co2+ for a variety of applications. The specific objective of this study was to genetically engineer E. coli strains which accumulate intracellular Ni2+ or Co2+ in order to manipulate the magnetic properties of the magnetosomes. Three E. coli mutants and a wild type strain were transformed with high affinity Ni2+ or Co2+ uptake genes and evaluated for intracellular accumulation at different medium concentrations of NiCl2 or CoCl2. Cellular iron and magnesium were also evaluated because iron is the major component of the magnetosome and magnesium is important for cell growth. The wild type strain, EPI 300 habouring Ni2+ uptake permease the hoxN gene or Co2+ uptake ABC type transporter cbiKMQO operon was found to accumulate the most intracellular Ni2+ or Co2+ in medium conditions most likely to induce magnetosome formation and magnetite manipulation. This strain can be used to co-express the MIA and iv Ni2+ or Co2+ uptake gene for mass production of magnetosome with altered magnetic properties. v Declaration I declare that “Engineering bacterial magnetic nanoparticles” is my own work, that it has not been submitted for any degree or examination in any other university, and that all the sources I have used or quoted have been indicated and acknowledged by complete references. Mr. Walter Nevondo January 2013 .............................. vi ACKNOWLEDGMENTS I wish to express my grateful thanks to ASSOC. PROF. MARLA TUFFIN for giving me a chance to work on this project, PROF. DONALD COWAN, for affording me the opportunity to attain my MSc, DR. WESLEY LOFTY EATON, for helping me to get this project off the ground, the late DR. ZAMA MTSHALI, for being such a good influence in science and research, and Dr. LUCAS BLACK who put extra effort and time to help me finish this work. Special thanks to DR. HEIDE GOODMAN and LONNIE VAN ZYL. vii LIST OF FIGURES Figure 1: General representation of ferromagnetism of magnetic materials using hysteresis loop. ....................... 2 Figure 2: General representation of soft ferromagnetic material and hard ferromagnetic material. ....................... 4 Figure 3: Phylogenetic relationships of the cultured and uncultured MTB. ........................................................... 5 Figure 4: Model of magnetosome membrane and magnetite formation.. ............................................................... 7 Figure 5: Illustration of iron transport across bacterial membrane. ...................................................................... 12 Figure 6 A: Illustration of a general ABC type transporter. ................................................................................. 13 Figure 7: Schematic representation of nickel transporters in E. coli. ................................................................... 15 Figure 8: A: Summary of Ni2+ and Co2+ transporters in bacteria.......................................................................... 17 Figure 9: Topological representation of the C. metallidurans H16 Ni2+ permease (HoxN).. ............................... 19 Figure 10: Model for genes arrangement in the czc and cnr system of C. metallidurans CH34. ......................... 23 Figure 11: Nickel homeostasis mechanism through NikR-RcnR system. ............................................................ 26 Figure 12: Diagrammatic representation of nickel equilibrium bioaccumulation in E. coli ................................. 42 Figure 13A-B: Ni2+ and Co2+ tolerance curves. .................................................................................................... 45 Figure 14 A-C: Growth curves of E coli strains ................................................................................................... 47 Figure 15: Agarose gel electrophoresis of 16S rRNA amplicon........................................................................... 49 Figure 16: Agarose gel electrophoresis of PCR products of hoxN, nhlF and cbiKMQO...................................... 50 Figure 17: Agarose gel electrophoresis of pBAD carrying hoxN, nhlF or cbiKMQO .......................................... 51 Figure 18: SDS-PAGE analysis of E. coli EPI 300 cells harbouring pBADhoxN................................................ 52 Figure 19 A-B: Representative growth curves of E. coli EPI 300 recombinant strains ........................................ 55 viii LIST OF TABLES Table 1: Plasmids used in this study..................................................................................................................... 28 Table 2: Bacterial strains used in this study.......................................................................................................... 29 Table 3: PCR conditions used for amplifying 16S rRNA..................................................................................... 33 Table 4: Primers used in this study....................................................................................................................... 34 Table 5: Preparation of 10% separating gel and stacking gel ............................................................................... 38 Table 6: Ni2+ and Co2+ concentrations used for metals accumulation .................................................................. 39 Table 7: Lag and doubling time of E. coli strains used in this study .................................................................... 48 Table 8: Stationary phase intracellular Ni2+ in E. coli strains harboring hoxN ..................................................... 59 Table 9: Stationary phase intracellular iron and magnesium in E. coli strains harboring hoxN ........................... 63 Table 10: Intracellular Ni2+, iron and magnesium of EPI 300 harboring hoxN at higher iron concentration ....... 66 Table 11: Stationary phase intracellular Co2+ in E. coli strains harboring cbiKMQO or nhlF ............................. 69 Table 12: Stationary phase intracellular iron and magnesium in E. coli strains harboring cbiKMQO or nhlF..... 71 Table 13: Exponential phase intracellular Co2+, iron and magnesium in EPI 300 harboring cbiKMQO or nhlF . 73 ix LIST OF ABBREVIATIONS μ Micro Δ Delta Oe Oersteds ADP Adenosine diphosphate Amp Ampicilin ATP Adenosine triphosphate A/m Ampere per meter APS Ammonium persulphate BLAST Basic local alignment search tool bp Base pare Cam Chloramphenecol CPG Casamino peptone glucose dH2O Demineralised water dATP Deoxy-adenosine 5’- triphosphate DNA Deoxyribonucleic acid dNTPs Deoxyribonucleiotides et al. et alia (and others) EDTA Ethylenediamine tetra-acetic acid ICP Inducible couple plasma IPTG Isopropyl-b-D-thiogalactopyranoside kDA Kilo Dalton LB Luria Bertani MIA Magnetosome gene island MIC Minimum inhibitory concentration MIT Metal inorganic transporter MNP Magnetic nanoparticle MTB Magnetotactic bacteria x OD Optical density OES Optical emmision spectroscopy PAGE Polyacrylamide gel electrophoresis PCR Polymarase chain reaction rpm Revolution per minute SBD Substrate binding domain SDS Sodium dodecyl sulphate TAE Tris acetic acid EDTA TBDP TonB Dependent protein TEMED N,N,N’N’-Tetramethylethylenediamine TE Tris EDTE TMD Transmembrane domain Tris Tris hydroymethyl-aminomethane U Units v/v Volume per volume v/w Volume per weight xi CHAPTER ONE 1. INTRODUCTION 1.1. Magnetic nanoparticles (MNPs) Nanotechnology is one of the fastest growing fields of science. The electronic evolution of the 1970s resulted in an increased demand for smaller devices. Since then, researchers have developed numerous methods of producing materials as small as 5 nm in diameter with electrochemical properties completely different from those of bulk material. Scanning electron microscopy (SEM) and atomic force microscopy (AFM) not only have allowed nano-scale materials to be visualized, but also made manipulation of particles at the atomic level possible, resulting in an increased understanding of electrical and physiochemical properties of nanoparticles. Magnetic nanoparticles (MNPs) are a class of nanoparticles which has been receiving increased research interest over the past years. The most common MNPs are those consisting of divalent metals such as iron, Ni2+, chromium, Co2+ and their oxides (Yamamoto et al. 2011). These particles are used in a variety of applications, including magnetic resonance imaging (MRI) (Frimpong and Hilt, 2010), targeted drug delivery, hyperthermia treatment (de Souza et al. 2011), gene delivery (Kami et al. 2011) and biological separation. MNPs are also used in information media as components of storage materials such as magnetic stripes, compact discs and in current flow control materials such as transformers and generators (Buhler et al. 2000). 1 The applications of MNPs are mainly determined by their chemical composition, size and shape (Artus et al. 2011). Based on their magnetic behaviours, MNPs are classified in three groups: ferromagnetic, paramagnetic and superparamagnetic (Kirchmayr, 1996). Paramagnetic materials are those that lose magnetisation immediately after the removal of the external magnetic field (Mahmoudi et al. 2011). As the paramagnetic materials decrease in size, their magnetic saturation increases and they become superparamagnetic, but can still be easily demagnetised to their zero magnetic state. Magnetic materials which retain magnetisation even independent of external magnetic field are ferromagnetic materials. These materials can be completely demagnetised to their zero state by the use of negative magnetic force called magnetic coercive force (Figure 1), measured in ampere per meter (kA/m) or Oesteds (Oe) (Mahmoudi et al. 2011). Figure 1: General representation of ferromagnetism of magnetic materials using hysteresis loop. Point “a” represents a point of magnetic saturation when magnetic force H is applied. When magnetic force H is reduced to zero, the curve will move from point “a” to point "b". At this point, it can be seen that some magnetisation remains in the material even though magnetic force H is zero. In order to completely demagnetise this material, reverse magnetic force (Coercive force) is required (Jiles and Atherton, 1986). 2 Ferromagnetic materials which require little coercive force to completely demagnetise are called soft magnetic materials (Chakraverty and Bondyopadhyay, 2007). These materials are easily magnetised by exposure to electrical current and can also be completely demagnetised by little coercive force (Figure 2). Soft magnets are used to control the flow of current in countless electronic devices such as MP3 players, computers, tape decks, transformers or as the magnetic cores of relays and inductors where current alternates frequently (Mchenry and Laughlin, 2000). Magnetic materials which require higher coercive force to completely demagnetise are termed hard magnets (Chakraverty and Bondyopadhyay, 2007). Unlike soft magnetic materials, hard magnetic materials have low permeability (they are difficult to magnetise) and are mostly naturally magnetised, although they can also be magnetised through an external magnetic field in the production or maintenance of magnetic field (Chakraverty and Bondyopadhyay, 2007). They are generally used to create magnetic fields in devices such as generators and motors of automobiles (Mchenry and Laughlin, 2000). MNPs are found in nature. They are produced by birds (Hanzlik et al. 2000), fish (Lohmann et al. 2008), ants (Acosta-Avalos et al. 1999), honeybees (Hsu et al. 2007) and bacteria (Blakemore, 1975). Interest in MNPs from biological sources has recently been increasing to replace complex and expensive chemical synthetic routes with more environmentally friendly and cost effective ones (Van Belle et al. 2007). In addition, chemically synthesised metal ferrite nanoparticles have been shown to be toxic and biologically incompatible (Berry and Curtis, 2003). 3 Figure 2: General representation of soft ferromagnetic material and hard ferromagnetic material. The loop of soft material is thin and held close to the central point while the loop of hard material is wide and away from the central point (Jiles and Atherton, 1986). More reverse magnetic energy is reguired to bring the loop of hard ferromagnetic material to zero than the energy required for soft ferromagnetic material 1.2. Magnetotactic bacteria Magnetotactic bacteria (MTB) are Gram negative, motile, microaerophilic and aquatic microorganisms, which synthesize magnetic nanoparticles (MNPs) within a membrane bound organelle called the magnetosome (Blakemore, 1975). These bacteria were first described by Bellini, (1965) and were later rediscovered and characterized by Blakemore1975 who named them Magnetotactic bacteria (MTB). The 16S rRNA analysis revealed that most known cultured and uncultured MTB belong to the Proteobacteria; Alphaproteobacteria, Gammaproteobacteria and Deltaproteobacteria (Amann et al. 2006), although some uncultured species have been shown to belong to the Nitrospira (Vali et al. 1987) (Figure 3). 4 Figure 3: Phylogenetic relationships of the cultured and uncultured MTB. Most cultured MTB (shown in bold) belongs to alphaproteobacteria and only one (Desulfovibro magneticus RS-1) belongs to gammaproteobacteria (Yan et al. 2012) MTB are found mainly in the anoxic-oxic interface (AOI) of water or sediments (Blakemore, 1975). They have been reported to inhabit diverse environmental conditions, including higher pH (pH~9) and temperature (63ºC) (Lefevre et al. 2011). Different studies have also reported diverse metabolic activities in MTBs. Most Magnetospirillum species have been shown to use organic acid as carbon and electron source (Schleifer et al. 1991) while others are capable of nitrogen fixation (Bazylinski et al. 2000). Magnetococcus marinus has been reported to use reduced sulphur compounds as an electron source (Bazylinski et al. 2004). MTB produce 5 membrane bound organelles called magnetosomes (Ragsdale and Kumar, 1996). Magnetosomes are intracellular vesicle like structures which contain iron oxide magnetite (Fe3O4) or iron sulphide greigite (Fe3S4) (Blakemore, 1982). This structure is a signature feature of MTB and helps them align to earth’s magnetic field (Vali et al. 1987). Magnetites of the magnetosome have been shown to be of species-specific sizes ranging from 35 nm to 120 nm (de Souza et al. 2011). Magnetosome particles show three defined morphologies, roughly cuboidal (Mann et al. 1994); elongated-prismatic (Bazyliski et al, 2004) and bullet shaped (Lefevre et al. 2011). Magnetosome formation is still not well understood, however significant progress has been made over the past few years. Membrane proteins of magnetosomes were reported to be similar to cytoplasmic membrane proteins (Tanaka et al. 2006). A study by Bazylinski and Schubbe (2007) suggests that magnetosome membranes are derived from the cytoplasmic membrane. This is because the magnetosome membrane of Magnetospirillum magneticum AMB-1 was found attached to the cytoplasmic membrane. Magnetosomes were also observed located adjacent to the cytoplasmic membrane (Bazylinski and Schubbe, 2007) (Figure 4). 6 Figure 4: Model of magnetosome membrane and magnetite formation. During the initial stage of cytoplasmic membrane invagination to form magnetosome membrane, periplasmic iron flows freely into the new magnetosome vesicle. When the mature magnetosome detaches from the cytoplasmic membrane, MagA is required to transport iron across the magnetosome membrane. Mms 16, MpsA and Mms24 are some of the proteins suggested to facilitate magnetosome formation (Yan et al. 2012). Genetic studies of Magnetospirillum magneticum AMB-1 revealed that about nineteen genes are essential in the formation of magnetosomes (Murat et al. 2010). Of these genes, mamI, mamL, mamQ and mamB were reported to be unique to MTB and are located in a 105 kb DNA fragment called the magnetosome gene island (Nahvi et al. 2004). Deletion of mamI and mamL showed no magnetosome formation in the cytoplasm (Murat et al. 2010). Analysis of the mamB sequence revealed that MamB has high homology to heavy metal transporter proteins, implying that this protein is involved in the transportation of iron into the magnetosome (Murat et al. 2010). Another protein, MagA has been implicated in iron 7 transport across the magnetosome membrane (Tanaka et al. 2006). However no additional studies have been done to confirm the involvement of these proteins in iron transport. Bacterial magnetosomes are an ideal alternative for a biological source of MNP with properties superior to MNPs produced by the traditional chemical route. However magnetosomes are generally soft magnetite which require low magnetic coercive force to demagnetise (Bazyliski and Schubbe, 2007) making them suitable for some applications while other applications require hard magnetic materials (Buhler et al. 2000). 1.3. Improving MNPs Manipulation of the coercive force of MNPs from bacteria and chemical synthetic routes has been the focus of most research (Chakraverty and Bandyopadhyay, 2007, Hsu et al. 2007; Van Belle et al. 2007). This was motivated by a growing discovery of more potential applications which require magnetic material with specific magnetic properties mainly in biomedicine and biotechnology (Wu and Zeng, 2010). Numerous methods of synthesising MNPs with specific coercive force have been developed in the chemical synthetic route (Hyeon, 2002; Van Belle et al. 2007). These methods generally involve doping of pure ferrite MNPs with divalent metal. The most common divalent metal used to increase magnetic coercive force of ferromagnetic material is Co2+, although other metals such as Ni2+, copper, zinc and manganese are also used (Chakraverty and Bandyopadhyay, 2007). Co2+ is the metal of choice because it has the highest Curie temperature known (1121ºC) (Bruno, 1991) and highest coercive force (1.5 kA/m) compared to other divalent metals (Friedberg and Paul, 1975), making it a difficult metal to 8 demagnetise. Doped MNPs include Co2+ ferrite (CoFe2O4) (Maaz et al. 2007) and Ni2+ ferrite magnetic nanoparticles (NiFe2O4) (Maaz et al. 2010) which have coercive force of up to 95. 9 kA/m and 23.0 kA/m respectively. In bacteria, different studies have reported manipulated coercive force of the magnetosome. Magnetosomes of three cultured MTB have been doped with up to 20 μM Co2+ resulting in Co2+ ferrite oxide nanoparticles (CoFe2O4) with increased coercive force (Staniland et al. 2008). The coercive force of the Magnetospirillum magnetotacticum MS-1 magnetosome has been changed by doping with zinc and Ni2+ (Kundu et al. 2009). Work by Perez-Genzalez et al. (2010) reported the incorporation of 1.14 % of manganese into the magnetosome of Magnetospirillum gryphiswaldense MSR-1. Recently, Tanaka et al. (2012) reported the incorporation of 3.0 % Co2+, 2.7 % manganese and 15.6% copper into the magnetosome of M. magneticum AMB-1. The Tanaka et al. (2012) study also revealed that doping with copper and manganese reduces the coercive force while Co2+ increases the coercive force (Tanaka et al. 2012). Metal doping of the magnetosome has proved to be more challenging than synthesizing metal doped MNP in chemical reactors. This is generally because at elevated concentration metals are toxic to all forms of life. This toxicity presents a limitation when high concentrations of metals need to be incorporated inside the magnetosome. For example doped CoFe2O4 with magnetic coercive force of 95.9 kA/m at room temperature has been produced by the chemical route using 0.2 M CoCl2 and 0.4 M FeCl2 (Maaz et al. 2007). In bacteria, the amount of metals to be incorporated into the magnetosome is determined by bacterial metal tolerance, transportation of metals from the environment into the cell and regulation thereof. These activities are all controlled at the genetic level. 9 1.4. Metals and bacteria Bacteria require metals for different metabolic activities (Cvetkovic et al. 2010). About 40 % of all cellular enzymes require metals as co-factors, of these 16 % require magnesium, 9 % require zinc, 8 % require iron, and about 1 % require Co2+ or Ni2+ (Waldron et al. 2009). Ni2+ enzymes include urease (Watt and Ludden, 1999), NiFe hydrogenase (Eitinger and Mandrand Berthelot, 2000), carbon monoxide dehydrogenase (Ragsdale and Kumar, 1996); and superoxide dismutase (Youn et al. 1996). Co2+ is required for the synthesis of methionine aminopeptidase (Chang et al. 1989), proline dipeptidase (Browne and O’Cuinn, 1983) and some nitrile hydratases (Komeda et al. 1997). Zn2+ is required for structural scaffolding of RNA polymerase and tRNA synthatase and is also a co-factor for more than three hundred other enzymes (Coleman, 1998). Magnesium functions as a co-factor of DNA polymerase and ATP. Manganese is an important component of chlorophyll and plays a role in photosynthesis (Waldron et al. 2009). However, at elevated concentrations, metals are known to inhibit cell growth (Baek and An, 2011). High concentrations of transition metals affect the lactoperoxidase sensitivity of Escherichia coli (Sermon et al. 2005). High Co2+ has been implicated in competitive binding to iron-sulphur proteins in E. coli resulting in the production of dysfunctional protein (Ranquet et al. 2007). Zinc accumulation was reported to affect the activity of dehydrogenases of Bacillus, Salmonella and Arthrobacter genera (Viret et al. 2006). Ni2+ accumulation was shown to inhibit DNA replication and translation by binding to proteins and nucleic acid (Macomber and Hausinger, 2011). 10 Bacteria acquire metals from the surrounding environment through metals uptake systems (Nies, 1999). Because metals are toxic to bacterial cells, these uptake systems are highly regulated through (Rowe et al. 2005). However, when metals are accumulated through nonspecific uptake systems which are usually controlled by global regulatory systems (Papp and Maguire, 2004) metal toxicity is prevented through metal efflux systems (Silver and Phung, 2005). These metal efflux systems transport metals back to the extracellular environment. 1.4.1. Metals uptake in bacteria Translocation of metals across bacterial membranes has been well studied (Silver and Phung, 2005). Most metals cross the outer membrane through porins (Schauer et al. 2007), while siderophores (iron complexes) and Ni2+ complexes have been shown to cross the outer membranes through TonB dependent transport (TBDT) (Noinaj et al. 2010) (Figure 5). In the periplasm, metals are transported across the cytoplasmic membrane by high affinity transporter proteins or low affinity transporter proteins (Agranoff and Krishna, 1998). Low affinity metal transporters usually have affinity to a variety of metals, although they show high affinity to one specific metal. These transporters are usually regulated by global regulatory systems such as the PhoPQ system (Newcombe et al. 2005). 11 Figure 5: Illustration of iron transport across bacterial membrane. Iron in the environment onment form forms complexes with siderophores which cross the outer membrane membra through TBDT (Noinaj et al. 2010). High affinity metal transporters are ei either multi component systems which consist of multiple proteins, or single component permeases which transverse the membrane several times (Mulrooney and Hausinger, 2003 2003). The multi component systems are ATP binding cassettes (ABC) which generally consist of two transmembrane domains (TMD) that form pores for metal passage, and two nucleotide binding domains (NBD) that hydrolyse ATP (Figure 6). The NBD is believed to o bind and hydrolyse ATP, thereby generating conformational onformational changes that lead to metal translocation (Tomii and Kanehisa, 1998). This system may als also consist of 12 a periplasmic substrate binding domain (SBD) which serves as an initial metal sensor (Eitinger and Mandrand-Berthelot, Berthelot, 2000). 2000 A B FIGURE 6 A: Illustration of a general ABC type transporter. The transporter consists of two major domains, the transmembrane domains (TMD) and the nucleotide binding domain (NBD). The signature motif and different characteristic features of the NBDs of these transporters are also shown. The signature gnature motifs consist of the aromatic amino acids, the walker A and B sequence, the Q-loop, the ABC signature siguence, the D and H loop. B: Illustration of the he ABC type transporter tra mechanism. Two ATP molecules bind ind to the NBD aand cause conformational changes which result in the release of substrate from the SBP to the TMD. ATP is hydrolysed, resulting in the release of substrate bstrate into the cytoplasm together with two ADP molecules (Linton Linton and Higgins, 2007). 13 Permeases are widely distributed in bacteria and are also found in Archaea and fungi (Wolfram and Eitinger, 1995). They include metal-specific as well as permeases with mixed specificity. Permeases have been observed to have affinities higher than those of ABC transporters, suggesting that they are adapted to supply minerals at lowest environmental levels (Mobley et al. 1995). The work presented in this study pertains to Ni2+ and Co2+ accumulation in E. coli. Therefore the details regarding these metals will be presented, with little reference to mechanisms involved in the transportation and regulation of other divalent metals. 1.4.1.1. Non-specific Ni2+ and Co2+ uptake in bacteria CorA is the best studied magnesium transporter. This transporter was initially reported to be a primary importer of Mg2+ in E. coli and Salmonella typhimurium (Papp and Maguire, 2004) but was later also shown to import Ni2+ and Co2+ with low affinity and specificity (Nelson and Kennedy, 1971). The CorA protein belongs to the fast non-specific metals inorganic family of transporters (MIT) and is constitutively expressed (Niegowski and Eshaghi, 2007). This protein accumulates Ni2+ and Co2+ in the cell even when the cytoplasmic concentration is high, resulting in intracellular metal build-up and toxicity (Nies, 1999). The E. coli CorA was shown to bind magnesium, Co2+ and Ni2+ with an affinity of 10-15 μM, 20-40 μM and 200-400 μM respectively (Niegowski and Eshaghi, 2007). The corA mutant has been shown to grow in 200 μM Co2+ concentration while a wild type with functional corA showed drastically reduced growth (Ranquet et al. 2007). In a metals tolerance bacterium Cupriavidus metallidurans CH34, mutation of the corA gene increased the minimum inhibitory concentration (MIC) of Co2+ from 55 mM to 70 mM (Kirsten et al. 2011). 14 1.4.1.2. Ni2+ ABC type transporters Ni2+ ABC transporters have been reported in a number of different organisms, including E. coli (Navarro et al. 1993) and Helicobacter pylori (Hendricks and Mobley, 1997). In E. coli (Figure 7), this transporter consists of five proteins: A soluble periplasmic Ni2+ binding protein NikA; two transmembrane proteins NikB and NikC; and two ATP hydrolysis proteins NikD and NikE (Mulrooney and Hausinger, 2003). The genes coding these proteins constitute the nik operon and have been reported to be expressed under anaerobic growth conditions to provide Ni2+ for the three hydrogenases involved in anaerobic metabolism (Rowe et al. 2005). Figure 7: Schematic representation of nickel transporters in E. coli. Nickel crosses the outer membrane through porins and TBDP. Once inside the periplasm, nickel is accumulated into the cytoplasm through NikABCDE and CorA system. Low level of nickel inside the cytoplasm is mainly maintained through RcnA (Lee and Hausinger, 2011). 15 NikA is the initial Ni2+ sensor and is also responsible for a negative chemotaxic response to high levels of Ni2+ (de Pina et al. 1995). The nikA gene was shown to be down regulated by high Co2+ concentrations, suggesting that NikA might also have some affinity to Co2+ (Fantino et al. 2010). Fluorescence studies of NikA reveal that this protein binds Ni2+ with a Kd value >0.1 µM (de Pina et al. 1995). A study by Addy et al. (2007) reported a much higher NikA Ni2+ binding affinity, a Kd value of about 10 µM. Although both studies reported different values, these values are much lower than that reported for the non-specific CorA transporter (Niegowski and Eshaghi, 2007). An E. coli nikA mutant was shown to have a drastically decreased Ni2+ uptake rate under low Ni2+ concentration compared to the wild type (Navarro et al. 1993). No significant Ni2+ uptake difference was detected between the nikA mutant and the wild type grown in high Ni2+ concentrations (Navarro et al. 1993), confirming that Ni2+ is also taken up by other low affinity transporters such as CorA at elevated media concentration (Nelson and Kennedy, 1971). The Ni2+ ABC importer of E. coli was reported to be positively controlled by a fumarate nitrate regulatory protein (FNR) and negatively controlled by NikR (Rowe et al. 2005). This regulation occurs through protein-DNA interactions at the nikABCDE promoter in response to intracellular Ni2+ concentration and oxygen tension (Chivers and Sauer, 2002). A decrease in oxygen tension was shown to activate FNR and up regulate nikABCDE expression, while the presence of excess Ni2+ activates NikR, which overrides the action of FNR and results in repression of nikABCDE transcription (Rowe et al. 2005). Ni2+ ABC transporters have been reported in other microorganisms (Figure 8), including the nik(MN)QO system of R. capsulatus (Rodionov et al. 2006), yntABCDE of a Gram-negative 16 ureolytic bacterium Yersinia pseudotuberculosis (Sebbane et al. 2002), and the abcABCD of Gram-negative human pathogen H. pylori (Hendricks and Mobley, 1997). Figure 8: A: Summary of Ni2+ and Co2+ transporters in bacteria. Ni2+ operon (nikABCDE) and Co2+ operon (cbiMNQO) represented here are from S. typhimurium, R. capsulatus and Brucella melitensis. B: The diversity of Ni2+ and Co2+ ABC type transporters in prokaryotes (Gelfand and Rodionov, 2008). 1.4.1.3. Ni2+ uptake permeases In the Gram-negative aquatic and soil bacterium C. metallidurans H16, a 33.1 kDa permease called HoxN has been reported to import Ni2+ in extremely low Ni2+ concentrations (Wolfram and Eitinger, 1995). C. metallidurans H16 produces two types of Ni2+-containing 17 hydrogenases which mediate the hydrogenesis of H2 under anaerobic conditions. The synthesis of these enzymes requires the presence of Ni2+ which is transported into the cell through HoxN (Wolfram and Eitinger, 1995). HoxN is a transmembrane protein which transverses the membrane eight times (Figure 9). Expression of hoxN in E. coli lead to 15 fold increase in Ni2+ accumulation compared to a wild type strain without hoxN. The binding affinity of HoxN to Ni2+ was reported to be Kd of 20 nM, showing a much higher affinity than ABC type transporters (0.1-10 µM) (de Pina et al. 1995, Addy et al. 2007). In H. pylori, another single component Ni2+ permease called NixA has been reported (Bauerfeind et al. 1996). Insertion inactivation of the nixA gene reduces urease (a Ni2+ containing enzyme) activity by about 42 %, with the remaining activity suggested to be arising from another ABC type Ni2+ transport system (Bauerfeind et al. 1996). The E. coli strain containing a functional H. pylori nixA has been shown to transport about 1250 pmol of Ni2+ per minute per 108 cells, whereas the control strain lacking nixA only transported an equivalent of 140 pmol Ni2+, showing an 8.9 fold increase. A transport constant of 11 nM of Ni2+ has been reported for NixA confirming that it is a high affinity transporter (Mobley et al. 1995). Ni2+ permeases have also been reported in Bradyrhizobium japonicum (Fu et al. 1994), Mycobacterium tuberculosis (Cole et al. 1998), Yersinia pestis (Eitinger et al. 2000) Sulfolobus sulfataricus P2 (She et al. 2001), and Yersinia pseudotuberculosis (Sebbane et al. 2002). 18 Figure 9: Topological representation of the C. metallidurans H16 Ni2+ permease (HoxN). This protein transverses the membrane eight times, with five loops in the periplasm and three loops in the cytoplasm. Both the N- terminus and C- terminus of this protein are in the cytoplasm (Wolfram and Eitinger, 1995). 1.4.1.4. Co2+ ABC type transporters Co2+ ABC type transporters have been reported for a number of different organisms (Figure 8). A gene cluster designated cbiMNQO was reported in Salmonella enterica, a Gramnegetive bacterium and Rhodobacter capsulatus, a purple non-sulfur photosynthetic bacterium capable of growing rapidly under non-photosynthetic conditions (Rodionov et al. 2006). Sequence alignment of the genes cluster show that it is the most widely distributed divalent metal uptake system in prokaryotes (Rodionov et al. 2006). The CbiMNQO system consists of transmembrane protein CbiQ and the ATP hydrolysis protein CbiO. The CbiN has 19 been proposed to be the substrate binding protein while the exact function of CbiM is still ellusive (Rodionov et al. 2006). The CbiMNQO transport systems from S. enterica and R. capsulatus show different Co2+ import capacities when expressed in E. coli in the presence of 500 nM Co2+ (Rodionov et al. 2006). The R. capsulatus CbiMNQO was found to transport 8.8 fold more Co2+ than S. enterica CbiMNQO (Rodionov et al. 2006). Although these different capacities have been credited to different expression efficiency, the possibility of differences in the actual transport capacity has not yet been investigated fully. Regulation of the cbiMNQO gene cluster has been suggested to occur through the B12 riboswitch elements located within certain mRNA (Nahvi et al. 2004), which is not surprising since Co2+ is required in this organism mainly for the synthesis of vitamin B12 (Rodionov et al. 2003). Other Co2+ ABC type transporters include Cbi(MN)QO of Streptococcus salivarius (Chen and Burne, 2003) and Actinobacillus pleuropneumoniae (Bosse´ et al. 2001), CbiKMQO of Magnetospirillum magneticum AMB-1 (Matsunaga et al. 2005), and CbiJKL of Sinorhizobium meliloti (Cheng et al, 2011). 1.4.1.5. Co2+ uptake permeases A homologue of A. eutrophus H16 HoxN called NhlF was found in the gram-positive actinomycete Rhodococcus rhodochrous J1 which uses nitriles as both carbon and nitrogen source (Rodionov et al. 2006). NhlF is a product of a single component gene nhlF located in the nitrile hydrogenase (NHase) operon (Komeda et al. 1997). NHase (Co2+ containing protein) assay using benzonitrile as a substrate showed that the presence of nhlF yields 20 catalytically active NHase even at low Co2+ concentrations (1-5 µM). In particular NhlF increases the NHase activity 3.7 fold in the presence of 1 µM CoCl2, showing a much higher affinity. Co2+ uptake by this protein was shown not to be affected by Mn2+; Fe2+ or Cu2+, but the presence of Ni2+ led to a marked decrease in Co2+ uptake suggesting that NhlF might also transport Ni2+, although giving more preference to Co2+ (Komeda et al. 1997). The amino acid sequence and the topology model reveal that NhlF is a markedly hydrophobic protein containing eight transmembrane helices with the N-terminus in the cytoplasm (Komeda et al. 1997). NhlF is the only single component high affinity Co2+ importer that has been demonstrated experimentally to import Co2+ so far, although permeases such as CbtA and CbtC have also been suggested to import Co2+ (Gelfand and Rodionov, 2008) (Figure 8). 1.4.2. Metals efflux in bacteria Some bacteria have been shown to tolerate higher metals concentrations. The E. coli isolated from sewage water was shown to tolerate up to 4 mM Ni2+ (Rubikas et al.1997), four times higher than what E. coli K12 has been reported to tolerate (Nies, 1999). A bacterial strain of Bradyrhizobium isolated from nodules of a legume plant was reported to grow in a medium containing 15 mM Ni2+ chloride (Chaintreuil et al. 2007). Bacterial mechanisms for metals resistance have been the subject of most research (Silver and Phung, 2005). These resistance mechanisms involve metals efflux proteins which detoxify the cell by exporting metals to the external environment when cytoplasmic concentration increases above homeostasis limits (Nies and Silver, 1994). Efflux proteins have been reported for variety of metals, and include highly specific proteins and proteins which transport multiple metals. Most metals efflux genes are located in plasmids (Nies, 21 1999) although chromosomal metals efflux genes are also known. The most studied metals efflux proteins are the P-type ATPase and the Resistance Nodulation cell Division (RND) system (Nies and Silver, 1994). The P-type ATPase depends on the phosphorylation of ATP for energy while the RND systems are generally chemiosmotic systems which use proton motive force (Andrews et al. 2003). 1.4.2.1. P-type ATPase efflux systems The P-type ATPases are integral membrane proteins which transverse the membrane eight to ten times (Nies and Silver, 1994). These proteins are ATP driven efflux with dual energy states (Odermatt et al. 1993). These proteins have been identified in archaea, prokaryotes and eukaryotes (Nies and Silver, 1994). Although these proteins differ between species, the general metal transport mechanism is common. The increase in cytoplasmic metals concentration results in the metal-protein interaction which is coupled with ATP dependent phosphorylation of the protein. This phosphorylation induces conformational changes which cause translocation of metal across the membrane (Nies and Silver, 1994). The binding of metals on the periplasmic domain is also believed to result in the pumping of secondary metals in to the cytoplasm (Nies and Silver, 1994). Ni2+ and Co2+ P-type ATPases have not yet been reported. However, they have been reported for other metals, including the copper pump CopB of H. pylori (Odermatt et al. 1993); the cadmium pump CadA of Bacillus subtilis (Silver et al. 1993); the zinc efflux pump ZntA and the arsenic resistance pump ArsA of E. coli (Kaur and Rosen, 1993; Rensing et al., 1997). 22 1.4.2.2. Ni2+ and Co2+ specific pecific RND driven efflux systems Ni2+ and Co2+ resistance mechanisms through RND systems have been well studied in heavy metals resistance bacterium C. metallidurans m (Nies et al. 1987; Liesegang et al. 1993). This bacterium has two megaplasmids with at least seven metals resistance determinan determinants (Nies et al. 1987). Three of these determinants, determinants cnrABCHRY; nccABCHY and czcABCY ABCY have been shown to be involved in Ni2+ and Co2+ homeostasis (Nies et al. 1987). Amino acid sequence identity shows that the Czc and Cnr systems share 48% and 28% identity on two transmembrane domains while the substrate binding domains have 30% identity (Figure 10), suggesting a common ancestral operon (Liesegang et al. 1993). The cnrABCHRY ABCHRY and nccABCR have been reported to share an average sequence identity of 66% in the regulatory loci andd 79% in the structural loci (Tibazarwa et al. 1999). Figure 10: Model for genes arrangement in the czc and cnr system of C. metallidurans lidurans CH34. The transmembrane components of these two systems show 46% and 28% homology while the substrate binding components show 30% homology (Liesegang et al. 11993). 23 The cnrABCHRY gene cluster encodes Ni2+ and Co2+ efflux proteins, CnrABCRY. The CnrAB has a transmembrane domain; CnrC has a substrate binding domain while the CnrHRY is involved in regulation of Cnr system (Liesegang et al. 1993). Liesegang et al. (1993) have also demonstrated that CnrABC of C. metallidurans CH34 is only expressed under higher Ni2+ concentration. Grass et al. (2000) have shown that cnrABC expression follows a saturation curve with half- maximum activation occurring at 49 μM Ni2+. The cnr system has been shown to enable C. metallidurans CH34 to grow in 3 mM Ni2+ and 5 mM Co2+ (Sensfuss and Schlegel, 1988). The nccABCR gene cluster encodes the Ni2+, Co2+ and cadmium efflux proteins, NccABCR (Schmidt and Schlegel, 1994). The ncc system enables Achromobacter xylosoxidans 31A to grow in the presence of 40 mM Ni2+, 20 mM Co2+ and 1 mM cadmium (Schmidt and Schlegel, 1994). The czcABCY cluster encodes the Co2+, zinc and cadmium efflux protein complex CzcABCY (Siddiqui and Schlegel, 1989). CzcAB was suggested to contain a potential transmembrane domain while CzcY has a regulatory domain (Liesegang et al. 1993). The CzcABC was shown to have an affinity of 22 μM to zinc, 140 μM to cadmium and 10 μM to Co2+ (Nies et al. 1987). Deletion of czcC resulted in a decrease of Co2+ and cadmium efflux. The deletion of czcA and czcB resulted in the complete loss of efflux activity of this system (Nies et al. 1989). The presence of czcABC in a recombinant E. coli strain has been shown to increase resistance to zinc and Co2+ 100 fold and 10 fold to cadmium (Nies et al. 1987). In E. coli, a protein called RcnA (Figure 11) has been reported to export Ni2+ and Co2+ back to the periplasm when their intracellular concentrations increase above homeostasis limits 24 (Iwig et al. 2006). Transcription of rcnA was reported to occur at high intracellular Ni2+ and Co2+ concentrations and is repressed by a transcriptional repressor RcnR. Deletion of rcnR resulted in constitutive expression of rcnA and decreased intracellular amount of Ni2+ and Co2+. Interestingly, increased intracellular Ni2+ concentration was shown to decrease the activity of Ni2+ uptake regulatory protein NikR, suggesting that NikR might be the first level of maintaining Ni2+ homeostasis in E. coli and rcnA is activated as the amount of Ni2+ continues to increase (Iwig et al. 2006). The rcnA mutation was shown to increase intracellular amount of Ni2+ in a medium containing 5 μM and 50 μM of Co2+ and Ni2+ respectively (Rodrigue et al. 2005). 25 Figure 11: Nickel homeostasis mechanism through NikR-RcnR system. In a natural environment, nickel is accumulated into the cytoplasm through NikABCDE. As the concentration of nickel increases inside the cytoplasm, NikR binds to the promoter of nikABCDE preventing transcription of the operon. However, when the concentration of nickel continues to increase due to non-specific importers, rcnR is expressed. RcnR promotes the expression of rcnA which encodes RcnA, a transmembrane permease which exports nickel and cobalt back to the periplasm, thereby maintaining low levels of nickel in the cytoplasm (Hyeon, 2002). 1.5. Study rationale and aim Bacterial magnetosomes are the most attractive source of magnetic nanoparticles. However, magnetosomes are chemically “soft” magnetic materials and can only be used for some 26 applications while other applications require “hard” magnetic materials. In addition, magnetotactic bacteria (MTBs) which synthesise the magnetosome are fastidious growers, making the synthesis of magnetosomes on an industrial scale difficult. Studies elsewhere (Qi et al. 2012), and here at Institute for Microbial Biotechnology and Metagenomics (IMBM) are being done to clone the genetic machinery for magnetosome synthesis, the magnetosome gene island (MIA) in E. coli in order to synthesise magnetosomes in an easy to culture organism. The aim of this study was to genetically engineer E. coli strains which accumulate high intracellular Ni2+ or Co2+. These strains will be used to express MIAs for the synthesis of chemically altered magnetosomes which can be used for a variety of applications. 1.5.1. The specific objectives are to:  Clone different Ni2+ and Co2+ uptake genes in three mutants and a wild type strain  Express the uptake genes in the presence or Ni2+ of Co2+  Study the cellular accumulation of Ni2+ and Co2+  Evaluate cellular iron and magnesium at different concentrations of Ni2+ and Co2+ in relation to iron 27 CHAPTER TWO 2. MATERIALS AND METHODS 2.1. Chemicals and reagents Chemicals used in this study were supplied by Merck Chemicals and Sigma-Aldrich. All chemicals were of analytical grade. Culture media were supplied by Oxoid and Biolabs. Enzymes, cloning kits and DNA size markers were supplied by Fermentas. Oligonucleotide primers for polymerase chain reaction (PCR) were supplied by Inqaba laboratories. dNTPs were supplied by Life Technologies. 2.2. Bacterial strains and plasmids TABLE 1: Plasmids used in this study Plasmid Relevant genotype/ characteristic Source pBAD (28) F- ΔlacX74 (Pvu II) Δara714 leu::Tn10 (CAMR) Invitrogen pJET 1.2 blunt rep (pMB1) bla (ApR) eco47IR PlacUV5 Fermentas pBADhoxN pBAD (28) harboring hoxN from C. metallidurans CH34 This study pBADnhlF pBAD (28) harboring nhlF from R. rhodochrous J1 This study pBADcbiKMQO pBAD (28) harboring cbiKMQO from M. magneticum AMB-1 This study 28 TABLE 2: Bacterial strains used in this study E. coli strains Relevant genotype Reference Source JW 3789-1 (ΔcorA) λ rph-1 ΔcorA761::kan ΔcorA761::kan hsdR514 Rowe et al. (2005) CGSC (USA) JW 3441-1 (ΔnikA) λ ΔnikA730::kan rph-1 Δ(rhaD-rhaB)568 hsdR514 Wu et al. (1994) CGSC (USA) JW 2093-1 (ΔrcnA) λ ΔrcnA731::kan rph-1 Δ(rhaD-rhaB)568 hsdR514 Koch et al. (2006) CGSC (USA) EPI-300 D(ara, leu)7697 galU galK 1- rpsL nupG trfA tonA dhfr University of the Western Cape, IMBM culture collection (SA) Other strains Capriavidus metallidurans CH34 University of the Western Cape, IMBM culture collection (SA) Rhodococcus rhodochrous J1 University of Stellenbosch culture collection (SA) 29 2.3. Growth media All E. coli strains were grown in Luria Bertani (LB) medium or M9 medium. LB broth was prepared by dissolving 10 g Tryptone; 5 g yeast extract and 10 g NaCl2 in 1 litre of distilled deionised water (ddH2O). LB agar was prepared by adding 15 g bacteriological agar in LB broth. M9 broth was prepared by dissolving 6 g Na2HPO4; 3 g KH2PO4; 0.5 g NaCl2 and 1 g NH4Cl in 880 ml ddH2O and supplemented with 20 ml of filter sterilized solution containing 2 g/L glucose; 0.5 g/L MgSO4.7H2O ;0.0152 g/L CaCl2.2H2O; 0.01 g/L sodium thiamine; 0.01 g/L Leucine and 0.001 g/L FeSO4.7H2O. Where indicated, ampicillin (amp) and chloramphenicol (cam) were added to a concentration of 200 mg/l and 170 mg/l respectively. C. metallidurans CH34 was grown in Casamino peptone glucose (CPG) medium containing 1 g Casamino acid; 10 g peptone; 5 g glucose dissolved in 1 litre ddH2O. Rhodococcus rhodochrous J1 was grown in nutrient broth prepared by dissolving 10 g nutrient broth powder in 1 L ddH2O. All media were sterilised by autoclave at 121ºC for 30 minutes. 2.4. Ni 2+ and Co2+ tolerance Tolerance of E. coli strains on Ni2+ and Co2+ was determined by growing cells at different concentrations of NiCl2 or CoCl2 and measuring absorbance at 600 nm. Glycerol stock culture was streaked on LB agar plate and grown overnight 37ºC. A single colony was inoculated in 5 ml M9 broth which was also grown overnight at 37ºC. A 50 μl of overnight culture was inoculated in 50 ml of M9 broth in 250 ml flasks supplemented with 0; 20; 40; 80; 160; 320; 640; 1280; 2580; or 5120 μM of NiCl2 or CoCl2. The flasks were incubated at 37ºC with shaking at 150 rpm until stationary phase. Cell density was measured using UV visible spectrophotometer at the wavelength of 600 nm. 30 2.5. Chemically competent cells The E. coli strain from glycerol stock was streaked on LB plates and incubated overnight at 37ºC. A single colony from an overnight culture was transferred to 5 ml LB broth in 50 ml falcon tubes and incubated overnight at 37ºC with shaking at 150 rpm. This culture was used to inoculate pre-warmed 100 ml LB broth in a 250 ml flask. The flask was incubated at 37ºC with shaking at 150 rpm until optical absorbance at 600 nm reached 0.4-0.6. The cells were transferred to a 500 ml centrifuge tube, incubated on ice for 15 minutes and harvested by centrifugation at 4500 x g (Backman J-26 XP, USA) for 5 minutes. The pellet was gently resuspended in 10 ml of ice cold 0.1 M CaCl2, incubated on ice for 30 minutes and harvested by centrifugation as above. The pellet was again gently re-suspended in 5 ml of ice cold solution of 0.1 M CaCl2 containing 15 % glycerol. Aliquots of 100 μl were transferred into 1.5 ml tubes and stored at -80ºC. 2.6. Genomic DNA of source organisms Genomic DNA of Magnetospirillum magneticum AMB-1 was kindly provided by Dr. Sarah Staniland from University of Leeds, England. Isolation of C. metallidurans CH34 and R. rhodochrous J1 genomic DNA was done using phenol/chloroform method. A single colony from overnight culture was inoculated in 2 ml CPG broth or nutrient broth and incubated for 16 hours at 30ºC with shaking at 150 rpm in a 10 ml tube. Cells were harvested by centrifugation at 10000 x g for 5 minutes, supernatant was discarded and the pellet was resuspended in 500 μl TE buffer (0.005 M Tris base and 0.002 M EDTA, pH 8.0) . An amount of 30 μl of 10% SDS and 5 μl of proteinase K (20 mg/ml) were added to the cells followed by incubation at 37ºC for 1 hour. A 100 μl mixture of phenol/chloroform solution (50:50) was 31 added followed by centrifugation at 10000 x g for 5 minutes. The upper aqueous phase of the solution was transferred to a clean tube followed by addition of another 100 μl of phenol/chloroform solution. This was centrifugated at 10000 x g for 10 minutes. A 50 μl volume of sodium acetate (5 M sodium acetate and 5% acetic acid) and 100 μl of isopropanol were added to the extracted upper phase and mixed gently to precipitate the DNA. Using a clean sterile glass rod, a suspension of DNA was spooled out of the solution and washed by dipping the end of the rod into 1 ml of 70 % ethanol for 30 seconds. DNA was re suspended in 200 μl TE buffer and stored at -20ºC. DNA concentration was quantified using a Nanodrop ND-1000 spectrophotometer at 260 nm (Nanodrop, Delaware USA). 2.7. Plasmid DNA isolation A glycerol stock of E. coli culture was streaked on LB-amp or LB-cam agar plate and grown overnight at 37ºC. A single colony of overnight cells was inoculated in 2 ml LB- amp or LBcam broth in a 50 ml tube. The culture was incubated overnight at 37ºC with shaking at 150 rpm. Cells were transferred to a sterile tube followed by harvesting through centrifugation at 10000 x g for 5 minutes. The pellet was re- suspended in 200 μl GTE buffer (50 mM glucose; 25 mM Tris-Cl, pH 8.0 and 10 mM EDTA, pH 8.0). A 200 μl solution of NaOH/SDS (0.2 N NaOH and 1 % SDS) and 200 μM solution of potassium acetate solution (3 M potassium acetate and 11.5 % (v/v) acetic acid) were added. The tube was centrifuged at 10000 x g for 5 minutes and 500 μl of supernatant was transferred to a new 2 ml tube. A 100 μl of ice cold 100% ethanol was added followed by incubation at -20ºC for 30 minutes. The tube was centrifuged at 10000 x g for 15 minutes. The supernatant was discarded and 100 μl of 70% ethanol was added to the pellet. The tube was again centrifuged for 2 minutes at 10000 x g and ethanol was carefully removed. Ethanol was removed by leaving the tube open upside 32 down on a paper towel for two hours. A 50 μl volume of TE buffer was added to dissolve the pellet and plasmid DNA was stored at 4ºC. The concentration was quantified using Nanodrop ND-1000 spectrophotometer at 260 nm (Nanodrop, Delaware USA). Plasmid isolation for sequencing and cloning was done using a Qiagen miniprep kit according to manufacturer’s recommendations. 2.8. PCR The PCR was carried out in a 50 μl reaction using Dream Taq or Phusion polymerase according to manufacturer’s recommendations. A typical reaction mixture contained 5 μl of 10 X Dream Taq buffer or 10 μl of 5 X Phusion buffer; 5 μl of 2 mM dNTP; 2 μl of 10 μM forward primer; 2 μl of 10 μM reverse primer; 1 μl of 100 ng/μl genomic DNA and 0.25 μl of Dream Taq polymerase or 0.5 μl of Phusion polymerase. Reaction mixture was adjusted to a final volume of 50 μl with sterile super quality (Milli Q) H2O. Amplification was carried out in an automated thermal cycler under conditions in Table 3. All primers used in this study are listed in Table 4. TABLE 3: PCR conditions used for amplifying 16S rRNA PCR Temperature (°C) Time (min) Cycles Initial denaturation 95 5 1 Denaturation 95 0.45 35 Annealing 55 0.45 35 Extension 72 See table 5 35 Final extension 72 5 1 33 TABLE 4: Primers used in this study Primers Primer sequences Extension time (min) Tm (ºC) %GC Target gene Reference E9F 5’ GAGTTTGATCCTGGCTCAG 3’ 1 58 52.6 Bacterial 16S rRNA Hansen et al, 1998 U1510R 5’ GGTTACCTTGTTGTTACACTT 3’ 1 58 38.1 Bacterial 16S rRNA Baker et al. 2003 nCH34F 5' TCTAGAAGGAGGAGCGGAATGGAAG 3' 0.15 57.5 52 hoxN This study nCH34R 5' AAGCTTCGCAGCTTAAACGCTACG 3' 0.15 55.4 50 hoxN This study cRJ1F 5' GGTACCAAGGACAAGCGTTTGAC 3' 0.17 55.2 52.2 nhlF This study cRJ1R 5' AAGCTTCGTCAGCGTTATGGGTAT 3' 0.17 55.8 53.7 nhlF This study cAMB1F 5’ GGTACCAGAACGAAGTCGCA 3’ 0.57 52.5 55 cbiKMQO This study cAMB1R 5’ AAGCTTTGCCTCCAGACCAC 3’ 0.57 52.5 55 cbiKMQO This study 34 2.9. Agarose gel electrophoresis Visualisation of specific DNA fragments was done on 0.7% [w/v] of agarose gel prepared in 0.5 X TAE buffer (0.2 % [v/v] Tris base; 0.5 % [v/v] glacial acetic acid and 1% [v/v] 5 M EDTA, pH 8.0) and 10 μl/L of 0.5 μg/ml ethidium bromide. DNA was mixed with a loading dye (60% [v/v] glycerol and 0.25% [w/v] Orange G) and loaded into the wells of cast gel. DNA molecular marker (lambda DNA restricted with PstI) was used as standard. The samples were electrophoresed at 100 V in 0.5 X TAE buffer. 2.10. Cloning and transformation 2.10.1. Cloning of PCR product in pJET 1.2/ blunt The PCR products were cloned in pJET 1.2/ blunt (Fermentas) according to the manufactures recommendations. The reaction mixture contained 10 μl of 2 X reaction buffer, 2 μl PCR product (55 ng/μl), 1 μl pJET vector, 5-10 U of T4 DNA ligase per μg of DNA and 6 μl ddH2O. The mixture was incubated at room temperature for 20 minutes and then used to transform the host strains. 2.10.2. Restriction digests The pJETcbiMNQO, pJETnhlF and pBAD(cam) were double digested with HindIII and KpnI. pJEThoxN and pBAD(cam) were digested with HindIII and XbaI . Digestion was carried out in a sterile microfuge tubes in reaction volume of 20 μl. The reactions contained 35 2 μl of 10 X reaction buffer, 2 μl of plasmid DNA (100 ng/ μl), 5 units of enzyme per μg of DNA, and were made up to 20 μl with sterile ddH2O. The digestion products were analysed by gel electrophoresis as indicated above. 2.10.3. Sub-cloning in pBAD The amplified and restricted hoxN, cbiKMQO or nhlF were sliced out from agarose gel using sterile surgical blade on a 365 waveleangth UV transluminator and placed in a clean sterile microtube. The DNA fragment was extracted from agarose gel using the Nucleospin gel extraction kit (Macherey-Nagel) according to the manufactures recommendations. Directional cloning was done in a 20 μl reaction containing: 10 μl of 2 X reaction buffer, 2 μl of double digested gene of interest (55 ng/μl), 1 μl of double digested pBAD vector, 10 U of T4 DNA ligase per μg of DNA and made up to 20 μl with ddH2O. The mixture was incubated at room temperature for 20 minutes and then used to transform host strains. 2.10.4. Sequencing Sequencing was carried out using an automated DNA sequencer 373 at the University of Cape Town sequencing facility. Sequencing was performed with fluorescein labelled primers (Applied Biosystems, USA). 2.10.5. Transformation An aliquot of 5 μl plasmid was placed in a 1.5 ml microtube followed by adding 100 μl of E. coli chemical competent cells. The mixture was incubated on ice for 30 minutes followed by 36 heat shock at 42ºC for 2 minutes. A 900 μl volume of LB broth was added on cells and incubated for 30 minutes at 37ºC. Fifty microliters of the expression mix was aseptically spread on LB plates containing antibiotic and incubated for 16 hours at 37ºC. 2.11. Expression and SDS-PAGE 2.11.1. Gene expression Expressional analysis was carried out according to the modified method of Auer et al. (2001). The cells harboring pBAD-hoxN from the glycerol stock were grown overnight at 37ºC on LB-cam plates. A single colony of overnight cells was inoculated in 2 ml LB-cam broth in a 50 ml tube. The tube was incubated overnight at 37ºC with shaking at 150 rpm. Cells were transferred to 10 ml LB-cam broth in a 50 ml flask and incubated as above until OD600=0.5-0.6. L-arabinose was added to a final concentration of 0, 0.0002, 0.002, 0.02, 0.2, and 2% and cells were incubated at 15ºC for 2 hours with shaking. An aliquot of 2 ml was transferred to a microtube and cells were harvested by centrifugation at 10000 x g for 3 minutes. 2.11.2. Cell lysis The cell pellet was re-suspended in 100 μl of lysis buffer (50 mM Tris-HCl pH 8, 0.5 mM Phenylmethanesulfonyl floride (PMSF), and 400 mM CaCl2), subjected to 3 cycles of sonication at 10% power for 10 second using a Sonoplus HD-070 sonicator (Bandelin, Germany) and stored at -20ºC. 37 2.11.3. SDS-PAGE SDS-PAGE was carried out on a Mighty SmallTM SE 280 vertical slab unit (Hoefer Inc, USA). Separating gel (10%) and stacking gel was prepared as indicated in Table 5. A sample of 10 μM SDS-PAGE sample buffer (50 mM Tris-HCl, pH 6.8; 2 % SDS; 1 % βmercaptoethanol; 10 % glycerol; 12.5 mM EDTA and 0.02 % bromophenol blue) was added to 20 μl of sample and was loaded on a gel and run at 80 V for 2 hours. PageRulerTM prestained protein marker (Fermentas) was used to estimate the molecular weight of proteins in the sample. The gel was stained in Coomassie Brilliant Blue G-250 staining solution at room temperature for 16 hours and de-stained in a buffer containing 10% (v/v) acetic acid and 50% (v/v) methanol for 16 hours. TABLE 5: Preparation of 10% separating gel and stacking gel Reagent Separating gel (ml) Stacking gel (ml) 1.5 M Tris-HCl, pH 8.8 2.5 - 0.5 M Tris-HCl 6.8 - 1.250 20% (w/v) SDS 0.05 0.025 Acrylamide/Bis-acrylamide (30%/0.8% w/v) 3.3 0.67 10% (w/v) ammonium persulfate 0.05 0.025 TEMED 0.005 0.005 ddH2O 4.1 3.075 38 2.12. Metal accumulation experiment 2.12.1. Accumulation at stationary phase Bacterial cells from glycerol stock were grown overnight at 37°C in LB-cam plates. A single colony was inoculated into 300 ml of M9-cam broth in a 1 L flask and incubated at 37ºC with shaking at 150 rpm until OD600 reached 0.8 -1.0. L-arabinose was at a concentration of 0.2%. NiCl2 or CoCl2 (0-1 mM) was added as indicated in Table 6. Cells were incubated for two hours followed by harvesting by centrifugation at 10000 x g for 5 minutes at 4ºC. The pellet was washed twice with 1 mM EDTA and once with sterile ddH2O. TABLE 6: Ni2+ and Co2+ concentrations used for metals accumulation E. coli strain Ni2+/ Co2+ concentrations (mM) EPI 300 0.005 0.05 0.1 JW 2093-1 0.001 0.01 0.02 JW 3789-1 0.1 0.5 1 JW 3441-1 0.005 0.1 0.5 2.12.2. Accumulation at exponential phase Bacterial cells from glycerol stock were grown overnight at 37°C in LB-cam plates. A single colony was inoculated into 300 ml of M9-cam broth in a 1 L flask and incubated at 37ºC with shaking at 150 rpm until OD600 reached 0.1 - 0.2. L-arabinose was added at a concentration of 0.2%. NiCl2 or CoCl2 was added to the final concentration of 0.1 mM and FeSO4 was added to a final concentration of 0.36 mM. Cells were grown up to OD600 39 between 0.6-0.7 followed by harvesting by centrifugation at 10000 x g for 5 minutes at 4ºC. The pellet was washed twice with 1 mM EDTA and once with sterile ddH2O. 2.13. Nitric acid digest and ICP-OES analysis The washed pellet (section 2.12.2) was dried at 100ºC for 3 hours in an oven. Dry weight was measured and recorded. Dried cells were transferred into a clean 25 ml flask (cleaned by soaking overnight in 10 % nitric acid and washed with distilled deionised water). Cell digestion was carried out by heating dried cells in 5 ml of 70 % nitric acid at 180ºC for 1 hour. Metals analysis was done at the chemistry department of the University of the Western Cape by inductively coupled plasma optical emission spectroscopy (ICP-OES) using metal standard prepared in 2 % nitric acid. 40 CHAPTER THREE 3. RESULTS AND DISCUSSIONS 3.1. Introduction The objective of this study was to engineer E. coli strains which accumulate intracellular Ni2+ and Co2+. These strains would be used to co-express the magnetosome gene island (MIA) and metal uptake genes for the synthesis of chemically altered magnetosomes. To achieve this objective, the following strategies were employed: I. High affinity Ni2+ or Co2+ uptake protein encoding genes were cloned and expressed in E. coli to improve intracellular accumulation as demonstrated by Wolfram et al. (2005). A Ni2+ uptake gene hoxN from C. metallidurans CH34; a Co2+ uptake gene nhlF from R. rhodochrous J1 and a Co2+ ABC type gene cluster cbiKMQO from M. magneticum AMB-1 were used in this study. HoxN was used in this study because it has been fairly well studied compared to other Ni2+ permeases (Wolfram et al. 1995) from a family of rare NiCoT permeases, which include NixA from H. pylori, HupH from B. japonicum and NhlF (Rodionov et al. 2006). NhlF is the only Co2+ uptake permease which has been experimentally investigated and shown to accumulate Co2+ (Rodionov et al. 2006). The CbiKMQO gene cluster from M. magneticum AMB-1 was included to compare its capacity with the Co2+ permease NhlF. 41 II. Ni2+ and Co2+ accumulation were studied in Ni2+ or Co2+ concentrations ranging from 0 to 1 mM. This was done to identify the lowest medium concentration which results in the most intracellular accumulation. A study by Deng et al. (2003) demonstrated that intracellular metal accumulation in bacteria increases as the medium concentration increased, until a point of bioaccumulation equilibrium is reached where increasing the medium concentration does not increase the intracellular accumulation (Figure 12). Equilibrium bioaccumulation occurs when the amount of metals being imported into the cell is in equilibrium with the amount of metals being exported out of the cell through a metal efflux mechanism (Deng et al. 2003). FIGURE 12: Diagrammatic representation of nickel equilibrium bioaccumulation in E. coli grown at increasing nickel concentrations (Deng et al. 2003). 42 III. Ni2+ and Co2+ intracellular accumulation was also conducted in three Ni2+/ Co2+ transport E. coli mutants. Three systems have been reported to transport Ni2+ and Co2+ in E. coli: The CorA transport systems of magnesium; Ni2+ and Co2+ (Nies, 1999); the NikABCDE high affinity ABC type Ni2+ transporter (Rodrigue et al. 2005) and the RcnA system which exports both Ni2+ and Co2+ from the cytoplasm (Rodrigue et al. 2005). The ΔcorA mutant has been shown to grow at high Ni2+ and Co2+ concentrations (Nies, 1999). This mutant was selected for this study in order to determine Ni2+ and Co2+ intracellular accumulation through high affinity transporters in medium concentrations which inhibit the growth of the wild type strain. The ΔnikA mutant was used to study the Ni2+ uptake capacity of NikABCDE and HoxN permeases. This was done to identify the most efficient Ni2+ uptake system for intracellular accumulation. The ΔnikA mutant was transformed with hoxN gene and compared for intracellular Ni2+ with the ΔcorA mutant. The ΔrcnA mutant was used in order to determine whether higher intracellular accumulation could be achieved by preventing export. This strain does not export Ni2+ and Co2+ from the cell cytoplasm (Rodrigue et al. 2005); hence it is expected to contain high concentrations of these metals. IV. Intracellular accumulation studies were also conducted using metals concentrations which have been shown to result in magnetosome doping. In Co2+ doping, Staniland et al. (2008) demonstrated that the medium Fe:Co ratio of 1:4 resulted in 1.4% Co2+ inside the magnetosome. In Ni2+ doping, Kundu et al. (2009) used Fe: Ni ratio of 2:1 to incorporate Ni2+ in the magnetosome. 43 3.2. Ni 2+ and Co2+ tolerance The mutants selected for this study have been reported to have different phenotypes when grown in the presence of Ni2+ or Co2+. The CorA mutant has been shown to be more tolerant to Ni2+ and Co2+ that the wild type (Rowe et al. 2005). The ΔnikA mutant confers high tolerance to Ni2+ (Wu et al. 1994) while the ΔrcnA mutant has been shown to be more sensitive to both Ni2+ and Co2+ (Koch et al. 2006). In this study, we first verified this tolerance by conducting Ni2+ and Co2+ tolerance experiments. All strains were first transformed with pBAD(cam) as controls. Figure 13A-B shows tolerance curves of EPI 300 and the three mutants grown in M9 (cam) medium at increasing concentrations of NiCl2 or CoCl2. The ΔcorA mutant was able to grow in up to 5.120 mM of both NiCl2 and CoCl2, the highest tolerance compared to the other strains. This is consistent with previous studies which reported high minimal inhibitory concentration (MIC) for a ΔcorA mutant compared to wild type strain (Wu et al. 1994). The ΔnikA mutant grows in 2.560 mM NiCl2 but was inhibited by 1.280 mM CoCl2. This Co2+ tolerance is similar to that of EPI 300, but EPI 300 shows less tolerance to Ni2+. As expected, the ΔrcnA mutant has the lowest tolerance for both Ni2+ and Co2+. This mutant did not grow on 0.64 mM NiCl2 and 0.31 mM CoCl2. These tolerance curves show that intracellular accumulation of Ni2+ and Co2+ in ΔcorA can be conducted in higher medium concentrations of these metals than other strains without affecting toxicity. 44 NiCl2 μM 5120 2560 1280 640 320 EPI 300 0 5120 2560 1280 640 320 160 80 40 20 EPI 300 ΔrcnA 160 ΔrcnA ΔnikA 0.8 0.6 0.4 0.2 0.0 80 ΔnikA 0.8 0.6 0.4 0.2 0.0 ΔcorA 40 ΔcorA 2.0 1.8 1.6 1.4 1.2 1.0 20 2.0 1.8 1.6 1.4 1.2 1.0 Cells density OD600 B 0 Cells density OD600 A CoCl2 μM Figure 13A-B: Ni2+ and Co2+ tolerance curves of E. coli mutants and EPI 300 grown on M9-cam medium. A Nickel tolerance curves and B Cobalt tolerance curves. All strains reached cell density of over 1.75 in the absence of either nickel or cobalt. The results shown here are average of three independent experiments. Error bars represent the standard deviation of the three experiments. 45 3.3. Growth curves in the presence of Ni2+ or Co2+ Longer lag time in the presence of Ni2+ or Co2+ was observed in EPI 300, the ΔnikA and the ΔrcnA mutant when conducting tolerance curves. Lag time was directly linked to the amount of NiCl2 or CoCl2 in the medium. Mathematical models for metals concentrations against lag time have been proposed in studies elsewhere (Gikas et al. 2009). Delayed lag time has been reported in other studies such as Cabrera et al. (2005) who reported longer lag phase in Desulfovibrio vulgaris growth on minimal medium supplemented with different metals. The growth of E. coli V38 in medium supplemented with 4 mM Ni2+ was also delayed compared to strain grown without Ni2+ (Rubikas and Matulis, 1997). Although the dose response lag phase is not yet fully understood, it has been reported that when exposed to high metals concentrations, bacterial cells prioritise the use of energy on expressing metals efflux proteins at lag phase, hence the long lag time in the presence of metals (Ralfe et al. 2011). To investigate how Ni2+ or Co2+ will affect the growth of mutant strains used in this study, growth curves were conducted in M9 (cam) medium containing of 0.1 mM NiCl2 or CoCl2. Figure 14A-C and Table 7 show summaries of results obtained from growth profiles. E. coli strains, EPI 300; ΔnikA and ΔrcnA show a prolonged lag phase in the presence of 0.1 mM NiCl2 or CoCl2. This delayed growth was followed by slow log phase (Table 7). The growth of the ΔnikA mutant was not significantly affected at 0.1 mM NiCl2 but was affected by 0.1 mM CoCl2, confirming a previous report which shows that nikA mutation reduces cell sensitivity to Ni2+ (Eitinger and Mandrand-Berthelot, 2000). 46 A WT ΔnikA B ΔrcnA C 10.0 10.0 1.0 0.1 Cells density OD600 Cells demsity OD600 Cells density OD600 10.0 1.0 0.1 No Metal No Metal 0.1 mM Ni 0.1 mM Ni 0.1 mM Co 0.0 4 8 12 Time (h) 17 21 0.1 0.1 mM Ni 0.1 mM Co 0.0 0 1.0 0.1 mM Co 0.0 0 5 10 15 Time (h) 20 96 99 102 105 108 112 115 Time (h) Figure 14 A-C: Growth curves of E coli strains: EPI 300 (A), ΔnikA mutant (B) and ΔrcnA mutant in M9-cam supplemented with 0.1 mM NiCl2 or CoCl2 and without NiCl2. 47 Table 7: Lag and doubling time of E. coli strains used in this study E. coli strains No Ni2+ or Co2+ 0.1 mM NiCl2 0.1 mM CoCl2 Lag time (h) Doubling time (min) Lag time (h) Doubling time (min) Lag time (h) Doubling time (min) EPI 300 5 ± 0.14 55 ± 2.7 9 ± 0.51 73 ± 3.3 11 ± 0.04 75 ± 6.5 ΔcorA 5 ± 0.22 54 ± 5.3 5 ± 0.33 54 ± 3.9 5 ± 0.025 54 ± 6.1 ΔnikA 5 ± 0.09 52 ± 3.1 6 ± 0.14 58 ± 1.8 11 ± 0.52 72 ± 5.8 ΔrcnA 7 ± 0.11 61 ± 2.9 96 ± 2.66 86 ± 3.8 96 ± 1.98 91 ± 8.6 Growth curves were conducted in M9 (cam) medium with or without 0.1 mM NiCl2 or CoCl2 supplimentation. The data presented here represents the mean ± standard deviation of three independent experiments. 48 3.4. 16S rRNA PCR Analysis Anal Before amplification of metals uptake genes, source organisms were first verified by amplification of the 16S rRNA gene gene. The amplicons were analysed in agarose gel (Figure 15) and were all found to be about 1.5 kb. These were cloned in pJET 1.2/ blunt and sequenced. Sequence alignment was done with sequences in the GenBank database using NCBI tool to confirm that the organisms were correct. correct M 1 2 3 FIGURE 15: Agarose garose gel of 16S rRNA amplicon. Lane M is a phage λ DNA marker digested with PstI, lane 1 to 3 is 1.5 kb 16S rRNA of C. metallidurans durans CH36, R. rhodochrous J1 and M. magneticum AMB-1. AMB 49 3.5. Cloning of the metal uptake genes The hoxN, nhlF and cbiKMQO genes were amplified (Figure 16) and cloned in pJET 1.2/ blunt to confirm rm the sequences before they were finallyy cloned in pBAD (Figure 117). M 1 2 M 3 4 M 5 6 FIGURE 16: Agarose gel electrophoresis trophoresis of PCR products of hoxN, nhlF and cbiKMQO genes genes. Lane M shows a λ phage DNA marker digested with PstI, lane 1and 2 shows a 1kb nhlF PCR product uct lane 3 and 4 shows a 834 bp hoxN PCR product duct , and la lane 5 and 6 shows a 3kb cbiKMQO PCR product. 50 M 1 M 2 M 3 FIGURE 17: Agarose gel electrophore ectrophoresis of pBAD carrying hoxN, nhlF or cbiKMQO genes. All plasmids were first linearised by digesting with HindIII. Lane M is a λ phage DNA marker digested with PstI, lane 1 shows a linear 6.6 kb pBADhoxN DNA band, lane 2 shows a linear 6.8 kb pBADnhlF DNA band and lane 3 shows a linear 9 kb pBADcbiKMQO band. ba 3.6. Expression of metal uptake up proteins The pBAD expression system has been used for the expression of various membrane proteins under the control of the arabinose promoter (Wang et al. 2003). Expressional xpressional level under this promoter can be optimised by varying the concentration of arabinose, making this system suitable for expression of recombinant membrane proteins (Khlebnikov et al. 2001) which are known to be toxic to the cell at higher concentration (Wang et al. 2003). Metals uptake proteins, like all membrane proteins are difficult to isolate from the ccell due to a number of factors. Thiss includes their high susceptibility to proteolytic degradati degradation (Auer et al. 2001), and incorporation into o the inclusion bodies, which reduces the amount of isolated protein (Wang et al. 2003) making them invisible in SDS-PAGE. Different methods have 51 been reported for expression and isolation of high concentration of membrane protein from E. coli (Wang et al. 2003). The efficiency of these methods depends on the size and nature of the membrane protein. Generally, different studies recommend that post induction time and incubation temperature be reduced to minimise proteolysis degradation and formation of inclusion bodies (Wang et al. 2003). Here, a modified method of Auer et al. (2001) was used to verify expression of metal uptake protein using a pBAD system. Figure 18 shows the SDS-PAGE analysis of the expression of hoxN under the arabinose promoter at different concentrations of L- arabinose. The highest expression was achieved with 2% L-arabinose. At L-arabinose of 0.02 and 0.2% also give good expression. M 1 2 3 4 5 6 55 kDa Putative HoxN 35 kDa 25 kDa FIGURE 18: SDS-PAGE analysis of E. coli EPI 300 cells harbouring pBADhoxN. Cells were induced with different concentrations of L-arabinose, harvested, disrupted in lysis buffer and loaded onto the gel. Lane M contains protein molecular weight ladder, lane 1 to 6 contains disrupted cells which were induced with 0; 0.0002; 0.002; 0.02; 0.2; and 2% L-arabinose respectively. The expected product is 30 kDa. The highlighted band indicates putative HoxN monomer. 52 3.7. Ni 2+ and Co2+ accumulation Different methods for metal bioaccumulation in recombinant E. coli expressing metal uptake genes have been previously reported. These methods differ depending on the general objective of the experiment. However in most of these methods the metal of interest is added during late logarithmic to stationary growth phase and the post-harvest time is commonly one to two hours. For example Krishnaswamy and Wilson (2000) induced expression of Ni2+ uptake gene, nixA in pGPMT3 plasmid vector when the cell density reached OD600 of 0.5-0.7, adding Ni2+ when OD600 was 1 and harvesting cells an hour later. The same method was used elsewhere (Deng et al. 2003). The idea of inducing expression at logarithmic growth is to get high amount of recombinant protein because at this stage cells are metabolically active. However, at logarithmic growth cells also express high amounts of proteases which degrade membrane proteins (Wang et al. 2003). Auer et al. (2001) induced the expression of membrane protein at OD600 of 1 to minimise this proteolytic degradation. Addition of metals only when the cells approach stationary phase prevents long lag times and allows for the harvest of a high quantity of biomass. However, in magnetosome doping, metals were added to the growth medium prior to bacterial inoculation (Staniland et al. 2008; Perez-Gonzalez et al. 2010), allowing incorporation into the magnetosome throughout magnetosome formation. According to a recent study by Ralfe et al. (2011), E. coli accumulates the most intracellular Ni2+ and Co2+ at stationary phase as shown by the up regulation of these specific uptake proteins during this time. The Ralfe et al. (2011) study also shows that unlike Ni2+ and Co2+, 53 iron is accumulated at lag phase. In this current study two approaches were employed to analyse intracellular Ni2+ and Co2+ accumulation: i. A modified method of Auer et al. (2001) where metal uptake was induced at late log phase (Figure 19A) was used. This was done to minimize proteolytic degradation and to obtain high biomass. High biomass was important because of the low detection limits of the ICP-OES technique. As indicated in the Ralfe et al. (2011) study, highest intracellular Ni2+ and Co2+ are accumulated during stationary phase, suggesting that a proper comparison of intracellular Ni2+ or Co2+ between the wild type strain and recombinant strain harboring Ni2+ or Co2+ uptake genes can be made at stationary phase. ii. Accumulation was also done in E. coli EPI 300 in conditions most likely to allow magnetosome formation and doping as demonstrated in previous studies (Staniland et al. 2008; Kundu et al. 2009). These conditions include a suitable ratio of Fe: Dopant (metal to be doped into the magnetosome) in the growth medium and adding the dopant prior to bacterial inoculation. Staniland et al. (2008) incoporated 1.4 % of Co2+ into the magnetosome using medium containing Fe: Co ratio of 1:4. The M9 growth medium used in this study contains 0.036 mM iron concentration. Unlike MTB, E. coli has no known requirement for Co2+. Therefore the Fe: Co ratio used by Staniland et al. (2008) was adjusted in this study to 1:3 by adding 0.1 mM CoCl2 in the growth medium which contains 0.036 mM iron. Kundu et al. (2009) incorporated Ni2+ into the magnetosome using a medium containing Fe:Ni ratio of 2:1. The amount of iron was twofold higher than Ni2+ mainly because Ni2+ is a strong competitive inhibitor of iron uptake. In this study a Fe:Ni ratio of 4:1 was used. Kundu et al. (2009) used low Ni2+ concentration (14 μM) in the medium because MTB are poorely tolerant to metals. In this study, concentrations of up to 1 mM were used because E. coli tolerate metals better 54 that MTB. At these high medium concentrations, Ni2+ can inhibit iron uptakes, which can result in Ni2+ toxicity and inhibition of magnetosome formation. To avoid this, the concentration of iron was raised fourfold and not twofold. However here metals where added at initial exponential phase to avoid extended lag time (Figure 19B). Cells density OD600 A Harvest 10.0 1.0 NiCl2 / CoCl2 + L arabinose 0.1 0.0 0 3 6 9 13 16 Time (h) B Cells density OD600 10.0 1.0 Harvest 0.1 NiCl2 / CoCl2 + L arabinose + Fe 0.0 0 3 6 9 13 16 Time (h) Figure 19 A-B: Representative growth curves of E. coli EPI 300 and its recombinant strains harbouring pBADhoxN, pBADnhlF and pBADcbiKMQO. The curves illustrate the point of induction, metals addition and harvest. A: Induction and metals addition were done at early stationary phase and cells were harvested at stationary phase, B: Induction and metals addition were done at early log phase and cells were harvested at late log phase. 55 3.7.1. Ni2+ accumulation 3.7.1.1. Ni2+ accumulation at stationary phase Table 8 shows that the intracellular amount of Ni2+ in all strains increased as the amount of Ni2+ in the medium increased, consistent with previous reports (Pal and Paul, 2010; Bleriot et al. 2011). A point of highest intracellular accumulation, where an increase in medium concentration does not increase intracellular accumulation was expected as shown in the study of Deng et al. (2003). Deng et al. (2003) demonstrated that E. coli JM109 grown in increasing concentrations of Ni2+ reached a point of equilibrium bioaccumulation when the cellular amount is 1.52 mg/g cells dry weight. This value is within the same range with those obtained in this study for ΔnikA harboring hoxN at 0.5 mM NiCl2 (1.289 ± 0.278 mg/g) and ΔcorA harboring hoxN at 1 mM NiCl2 (1.109 ± 0.107 mg/g). It is most likely that ΔnikA reached equilibrium bioaccumulation at medium concentration between 0.5 mM and 1 mM NiCl2 while ΔcorA habouring hoxN reached equilibrium bioaccumulation at just over 1 mM NiCl2 in the medium. A further increase in medium concentration up to 2 mM in case of the ∆corA mutant and up to 1 mM in case of the ∆nikA mutant would have provided a clear equilibrium bioaccumulation. However, because of limited time available to complete this work, this was not done. The presence of hoxN increased intracellular Ni2+ by up to 31 fold (Table 8). However when the amount of NiCl2 in the medium was increased the contribution of HoxN to intracellular Ni2+ was not as significant. The E. coli EPI 300 harboring hoxN accumulated 11 fold more Ni2+ than the wild type without hoxN in a medium containing 5 μM NiCl2 (Table 8). This is consistent with a study by Wolfram et al. (1995) who showed that E. coli CC118 harboring 56 C. metallidurans CH34 hoxN accumulates 15 fold increased cellular Ni2+ compared to the wild type at 5 μM Ni2+ concentration (Wolfram et al. 1995). Wolfram’s finding and the results of this study also agree with a report by Eitinger and Mandrand-berthelot (2000) who showed that HoxN is a high affinity protein which scavenge Ni2+ at trace environmental concentrations. The ΔnikA mutant does not show a significantly decrease in Ni2+ accumulation. This was demonstrated by comparable amount of intracellular Ni2+ in EPI 300 and the ΔnikA mutant (Table 8 and appendix 1). This was surprising because the tolerance study showed that ΔnikA mutant is more tolerant to NiCl2 than EPI 300, suggesting that this mutant does not accumulate as much intracellular Ni2+ as EPI 300. However the growth curves of this two strains in the present of 0.1 mM NiCl2 were quite similar (Figure 14A-B), suggesting that at 0.1 mM NiCl2 mutation at nikA gene does not significantly decrease the intracellular accumulation. This is also in line with the finding of Navarro et al. (1993) as indicated in the previous chapter. The ΔcorA mutant has a significantly reduces intracellular amount of Ni2+ compared to the wild type strain (EPI 300) and ΔnikA mutant (Table 8), confirming previous reports that corA also transports Ni2+ (Snavely et al. 1990). The ΔrcnA mutant accumulates the least cellular Ni2+ compared to the other strains, although its intracellular contant is comparable with that obtained in EPI 300 at 5 μM NiCl2. This strain was expected to contain the most intracellular Ni2+ since it lacks Ni2+ and Co2+ specific efflux proteins. As shown in the growth profile done at 0.1 mM NiCl2 and Ni2+ tolerance studies, this strain has difficulties growing on Ni2+. 57 Under the conditions used in this study, the HoxN Ni2+ uptake system was shown to be more efficient than the NikABCDE system. This was demonstrated by almost 6 fold differences between the ΔnikA mutant harboring hoxN and the ΔcorA mutant with functional NikABCDE at 0.1 mM NiCl2. When both the HoxN and the NikABCDE systems are present (EPI 300 + hoxN), there was no significant increase in intracellular Ni2+ compared to strain with only hoxN (ΔnikA + hoxN). This may be because the NikABCDE system is down regulated in the presence of Oxygen as reported by de Pina et al. (1995). 58 Table 8: Stationary phase intracellular Ni2+ in E. coli strains harboring hoxN Cellular Ni mg/g dry weight NiCl2 mM EPI 300 EPI300+hoxN INC ΔnikA No metal 0.000 ± 0.000 0.000 ± 0.000 0 0.000 ± 0.000 0.005 0.002 ± 22E-5 0.022 ± 82E-5 11 0.003± 6E-6 INC ΔcorA ΔcorA +hoxN INC ΔrcnA ΔrcnA +hoxN INC 0.000 ± 0.000 0 0.000 ± 0.000 0.000 ± 0.000 0 0.000 ± 0.000 0.000 ± 0.000 0 0.031 ± 41E-5 31 0.003 ± 3E-6 0.063 ± 0.002 21 0.01 0.005 ± 27E-5 0.089 ± 0.002 17.8 0.02 0.019 ± 0.009 0.091 ± 0.001 4.8 0.05 0.028 ± 21E-7 0.126 ± 0.007 4.5 0.1 0.154 ± 0.035 0.394 ± 0.031 2.4 0.5 ΔnikA +hoxN 0.135 ± 0.084 0.366 ± 0.034 2.7 0.064 ± 0.002 0.265 ± 0.026 4.1 1.087 ± 0.240 1.289 ± 0.278 1.2 0.132 ± 0.016 0.500 ± 0.034 3.8 0.673 ± 0.093 1.109 ± 0.107 1.6 1 Intracellular Ni2+ was measured in cells grown in M9 (cam) medium. The culture was grown until OD600 reached 0.8-1, followed by addition of up to 0.2% L-arabinose and NiCl2 and harvested at the end of stationary phase. INC represents the fold increase between control strain and strain harboring hoxN .The results shown here represent the mean ± standard deviation of ICP-OES analysis of three independent experiments. 59 3.7.1.2. Intracellular iron and magnesium at different Ni 2+ concentrations Iron and magnesium are important trace metals required by most bacteria for various metabolic activities (Snavely et al. 1990). In magnetotactic bacteria (MTB), iron is the major component of the magnetosome (Blakemore, 1975). Outten and O’ Halloran, (2001) reported that E. coli is able to maintain a cellular iron “quota” of 180 μM. In MTB intracellular iron concentration has been shown to be up to 21 mg per gram of cell dry mass in cells grown under micro-aerobic conditions and 10.8 mg/g when cells were grown under aerobic conditions (Heyen and Schuler, 2003). The results in Table 9 and Appendix 2 show that the intracellular iron concentration was not reduced by the addition of up to 0.1 mM NiCl2. However at 0.5 mM NiCl2, intracellular iron was significantly reduced, suggesting higher Ni2+ concentrations inhibit iron uptake. This is consistent with the study of Bereswill et al. (2000) who demonstrated that high Ni2+ concentration inhibits iron uptake. The ΔcorA mutant showed low intracellular iron, suggesting that CorA is important in iron transport. Hantke, (1987) also reported that CorA is involved in iron uptake in E. coli and S. typhimurium. However Papp and Maguire (2004) reported contrary findings, showing that CorA does not transport iron. Papp and Maguire conducted their study in a solution containing extremely low iron concentration (0.001 mM). It is unlikely in their case that the wild type strain and the ΔcorA mutant will have any significant difference in intracellular iron content because iron is not a specific substrate of CorA. As a result, magnesium and other divalent metals in the medium whose concentrations are higher will inhibit iron import through CorA. In this current study, M9 medium contained 0.036 mM iron, a significantly 60 higher concentration than in the Papp and Maguire study. This may be the reason for the different results. The amount of cellular iron per gram of dry weight of bacterial cell has been previously studied. Barton et al. (2007) reported that bacterial cells contain 0.015g/ 100g of dry weight cellular iron. In E. coli W3110 grown in M9 medium supplemented with up to 16 μM iron citrate, the amount of iron was 0.18 mg/g at stationary phase and 0.2 mg/g at logarithmic growth (Abdul-Tehrani et al. 1999). As shown in Table 9, the amounts obtained in this current study are slightly higher, most likely due to the high iron concentration in the medium (0.036 mM). However, these amounts are still within the same range as those of previous studies. Cellular magnesium in all strains, with the exception of the ΔcorA mutant, was not reduced in the presence of all concentrations of NiCl2 tested (Table 9 and Appendix 3). Previous data show that E. coli K12 grown in LB medium contain 2.3 mg of magnesium per gram of cell dry weight at early log phase, 1.3 mg/g at late log phase and 0.86 mg/g at stationary phase (de Medicis et al. 1986). E. coli DG336 grown in minimal salt medium containing 0.6 mM magnesium ion was reported to accumulate 2.8 mg of magnesium per gram of cells dry weight measured at late log phase (Kung et al. 1975). The results in Table 9 and Appendix 3 show different values from those reported above. Cellular magnesium content of between 2.835 ± 0.338 mg/g and 3.048 ± 0.144 mg/g dry weight was observed in EPI 300 and the ΔrcnA mutant at all concentrations of Ni2+ tested. These values are higher than those reported previously at stationary phase (0.86 mg/g). The differences can be attributed to two factors: LB medium contains chelating agents which reduce intracellular bioaccumulation of divalent metals, hence lower magnesium content in the Medicis et al. (1986) report. High 61 magnesium concentration in the medium is most likely to increase intracellular concentration. M9 medium used in this experiment contains 2 mM magnesium ion, three fold higher than the 0.6 mM used in the study by Kung et al. (1975). The ΔnikA and ΔrcnA mutants show the same intracellular magnesium as the wild type, indicating that these genes do not import magnesium. Cellular magnesium was very low in the ΔcorA mutant compared to the other strains, which was not surprising since CorA is the main magnesium transporter (Rodrigue et al. 2005). 62 Table 9: Stationary phase intracellular iron and magnesium in E. coli strains harboring hoxN Cellular Fe and Mg mg/g dry weight NiCl2 mM ΔnikA + hoxN EPI 300 + hoxN ΔcorA + hoxN ΔrcnA + hoxN Fe Mg Fe Mg Fe Mg Fe Mg 0 0.329 ± 0.036 3.126 ± 0.341 0.369 ± 0.088 2.988 ± 0.158 0.084 ± 0.006 0.594 ± 0.008 0.340 ± 0.037 2.982 ± 0.093 0.005 0.373 ± 0.006 3.049 ± 0.183 0.336 ± 0.015 2.900 ± 0.172 0.351 ± 0.093 2.946 ± 0.129 0.01 0.300 ± 0.089 2.855 ± 0.245 0.02 0.282 ± 0.083 2.814 ± 0.110 0.05 0.372 ± 0.001 2.945 ± 0.003 0.1 0.320 ± 0090 2.919 ± 0.043 0.5 0.332 ± 0.002 2.906 ± 0.138 0.074 ± 0.008 0.582 ± 0.088 0.082 ± 0.002 2.953 ± 0.108 0.055 ± 0.003 0.551 ± 0.009 0.033 ± 0.008 0.541 ± 0.046 1 Intracellular Iron was measure in cells grown in M9-cam broth until OD600 reached 0.8-1. L-arabinose and NiCl2 were added and cells were harvested at stationary phase. The results shown here represent the mean ± standard deviation of ICP-OES analysis of three independent experiments. 63 3.7.1.3. Intracellular Ni2+, iron and magnesium in conditions for magnetosome doping Intracellular Ni2+ accumulation in EPI 300 and its recombinant strain harboring hoxN was also studied at conditions most likely to induce magnetosome formation. EPI 300 was used in this experiment because it accumulated the most cellular Ni2+ at 0.1 mM NiCl2 without affecting cellular magnesium and iron (Table 8-9 and appendix 2) at stationary phase. The concentration of FeSO4 in M9 (cam) medium was increased tenfold from 0.036 mM to 0.36 mM to simulate conditions for magnetosome formation and doping as shown by Kundu et al. (2009). Table 10 shows the amount of intracellular Ni2+, iron and magnesium in EPI 300 grown until mid-exponential phase in medium containing 0.36 mM iron and 0.1 mM Ni2+ concentrations. Intracellular Ni2+ in EPI 300 was 1.7 fold lower than the amount obtained at stationary phase, consistent with an earlier study which shows that Ni2+ uptake occurs at stationary phase (Rolfe et al. 2011). The same study also showed that the rcnA gene which encodes the Ni2+ efflux protein is upregulated at the later stage of lag phase (Rolfe et al. 2011), further explaining the lower intracellular Ni2+ at log phase. This intracellular Ni2+ was further reduced 10 fold in the presence of 0.36 mM FeSO4. However, iron did not seem to interfere with Ni2+ accumulation through HoxN, suggesting that these metals only compete for low affinity and non-specific metal proteins. Intracellular Ni2+ was increased in the presence of HoxN, showing a 2.95 fold increase. This was also the case in the presence of 0.36 mM iron, further suggesting that HoxN does not import iron. 64 Intracellular iron was slightly higher than values obtained at stationary phase at 0.036 mM iron and more than 1.5 fold higher when iron in the medium was increased to 0.36 mM iron, consistent with previous study which reported that iron is accumulated at lag phase (Ralfe et al. 2011). In E. coli W3110, cellular iron content at logarithmic phase was reported to be 0.27 mg/g dry weight in cells grown in LB medium supplemented with 17 μM iron citrate. In M9 medium supplemented with 16 μM iron citrate, cellular iron was 0.2 mg/g dry weight (AbdulTehrani et al. 1999). The amount of iron in the medium in this study was significantly higher (0.36 mM), which is most likely to be the reason for higher values obtained intracellularly. Intracellular magnesium was slightly higher at log phase, which also suggests that magnesium might be accumulated at lag phase of growth. Magnesium is required for growth hence a higher magnesium concentration would be expected at log phase than at stationary phase. The report by de Medics et al. (1989) shows that E. coli contains 2.3 mg/g dry weight at early log phase and 1.3 mg/g at late log phase, showing a decrease as the cells approach stationary phase. In this current study, a consistent average of 3.6 mg/g of dry weight was obtained. As indicated earlier, this higher amount is most probably due to high magnesium concentration in M9 medium. 65 Table 10: Intracellular Ni2+, iron and magnesium of EPI 300 harboring hoxN at higher iron concentration in the medium Metals Cellular Ni (mg/g dry weight) Cellular Fe (mg/g dry weight) Cellular Mg (mg/g dry weight) EPI 300 EPI 300 + hoxN EPI 300 EPI 300 + hoxN EPI 300 EPI 300 + hoxN 0 mM 0.000 ± 0.000 0.000 ± 0.000 0.390 ± 0.024 0.391 ± 0.043 3.661 ± 0.336 3.670 ± 0.451 0.1 mM Ni 0.087 ± 0.007 0.257 ± 0.015 0.397 ± 0.037 0.320 ± 0.096 3.606 ± 0.211 3.700 ± 0.495 0.1 mM Ni+ 0.36 mM Fe 0.008 ± 12E-9 0.255 ± 0.008 0.616 ± 0.042 0.610 ± 0.008 3.698 ± 0.083 3.602 ± 0.172 Intracellular Ni2+, iron and magnesium were determined in cells grown until early log phase in M9 (cam) medium followed be addition of Ni2+ or Ni2+ and iron. The results shown here represent the mean ± standard deviation of ICP-OES analysis of three independent experiments 66 3.7.1.4. Suitable strain for Ni2+ doping of the magnetosome The results presented in Table 8 shows that the ΔnikA mutant habouring hoxN accumulates the most intracellular Ni2+ at 0.5 mM NiCl2 in the medium (1.289 ± 0.278 mg/g). This is significantly higher than 0.213 mg/g obtained by Koch et al. (2006). However, at this medium concentration, the amounts of cellular iron were decreased. Iron is important for magnetosome formation, suggesting that intracellular Ni2+ accumulation for magnetosome doping should be conducted at lower medium Ni2+ concentration. At 0.1 mM NiCl2 in the medium, EPI 300 habouring hoxN accumulates the most intracellular Ni2+ (0.394 mg/g). This value was reduced to 0.255 mg/g in the presence of 0.36 mM iron in the medium when the cells were harvested at exponential phase. This decrease was coupled with an increase in intracellular iron, from 0.320 ± 0.096 mg/g to 0.610 ± 0.008 mg/g (Table 10). Unfortunately Ni2+ accumulation experiments were conducted in strains without the MIA. As a result, it is difficult to indicate with certainty the amount of intracellular Ni2+ which will result in magnetosome doping. However, the amount obtained in EPI 300 harboring hoxN at 0.1 mM NiCl2 in the medium is the highest obtained compared to all the other conditions tested. Of all the strains studied, this strain is the most suitable strain for Ni2+ doping in the magnetosomes. 67 3.7.2. Co2+ accumulation 3.7.2.1. Co2+ accumulation at stationary phase Intracellular Co2+ accumulation was studied in E. coli strains harboring cbiKMQO or nhlF. The results in Table 11 show that the amount of intracellular Co2+ increases as the amount of CoCl2 is increased in the medium, as was shown for Ni2+ accumulation and other studies (Pal and Paul, 2010; Bleriot et al. 2011). The comparable intracellular amount of Co2+ in the ∆nikA mutant habouring cbiKMQO or nhlF at 0.5 mM CoCl2 and the ∆corA mutant harboring nhlF at 1 mM CoCl2 (Table 11) suggest an equilibrium bioaccumulation. However, this cannot be stated with certainty since the medium concentration was raised only up to 1 mM. Most certainly equilibrium bioaccumulation would have been obtained if the medium Co2+ concentration was raised up to at least 2 mM. The ΔcorA mutant has low intracellular Co2+ compared to EPI 300 and the ∆nikA mutant (Table 11 and appendix 4-5), indicating that CorA is also important for Co2+ transport (Nies, 1999). The ΔrcnA mutant has the least cellular Co2+ compared to the other strains. This is because this strain grows poorly in the presence of metals. The CbiKMQO protein cluster was more efficient than NhlF. This is clearly indicated by the folds increase as shown in Table 11. However, at higher medium concentration of CoCl2, both CbiKMQO and NhlF did not make significant difference in intracellular accumulation, as was also seen for Ni2+ accumulation. This suggests that at high medium concentration, Co2+ is accumulated into the cells through other non-specific transporters (Nies, 1999). 68 Table 11: Stationary phase intracellular Co2+ in E. coli strains harboring cbiKMQO or nhlF Cellular Co mg/g dry weight Co mM EPI 300+ cbi INC EPI 300+nhlF INC ΔnikA + cbi INC ΔnikA+nhlF INC ΔcorA + cbi INC ΔcorA+nhlF INC ΔrcnA + cbi INC ΔrcnA+nhlF INC 0 0.000 ± 0.000 0 0.000 ± 0.000 0 0.000 ± 0.000 0 0.000 ± 0.00 0 0.000 ± 0.00 0 0.000 ± 0.00 0 0.00 ± 0.000 0 0.000±0.00 0 0.005 0.049 ± 3E-2 7 0.022 ± 31E-4 3.4 0.038 ± 0.001 9.5 0.020 ± 1E-4 5 0.088 ± 3E-1 22 0.052 ±5E-4 13 0.01 0.107 ± 3E-4 17 0.081 ± 3E-1 13.5 0.02 0.206 ± 0.02 15 0.133 ± 3E-2 10 0.05 0.216 ± 0.006 5.7 0.187 ± 0.003 4.9 0.1 0.673 ± 0.011 5.3 0.425 ± 0.01 3.3 0.5 0.693 ± 0.093 5.4 0.477 ± 0.02 2.9 0.294 ± 0.02 5.5 0.191 ± 0.08 3.6 1.492 ± 0.185 1.09 1.45± 0.185 1.05 0.866 ± 0.13 2.6 0.730 ± 0.14 2.2 1.527 ± 0.09 2.3 1.197 ± 0.08 1.8 1 Intracellular Co2+ was measured in cells grown in M9-cam medium. The cells were grown until OD600 reached 0.8-1, followed by addition of up to 0.2% L-arabinose and CoCl2 and harvested at the end of stationary phase. INC represents the fold increase between control strain without recombinant Co2+ uptake gene and strain harboring cbiKMQO or nhlF. cbi represent cbiKMQO. The results shown here represent the mean ± standard deviation of ICP-OES analysis of three independent experiments. 69 3.7.2.2. Intracellular iron and magnesium at stationary phase in the presence of Co2+ Intracellular iron and magnesium have been studied in all strains grown at different concentration of CoCl2 (Table 12 and appendices 6-9). Co2+ did not affect intracellular magnesium for all the medium concentrations tested. However, the ΔcorA mutant has significantly lower intracellular iron and magnesium, showing that CorA is important for both iron and magnesium import. The ΔnikA and ΔrcnA mutants accumulated similar magnesium and iron as EPI 300 (appendix 6+8), showing that NikA and RcnA do not transport iron or magnesium. The cbiKMQO and nhlF genes did not result in transport of either iron or magnesium, confirming that these systems are Co2+ specific transporters (Rodionov et al. 2005). This was shown by similar amount of cellular iron and magnesium in strains harboring nhlF or cbiKMQO and the untransformed strains (Appendices 6-9). 70 Table 12: Stationary phase intracellular iron and magnesium in E. coli strains harboring cbiKMQO or nhlF Cellular Fe and Mg (mg/g dry weight) ΔnikA + cbiKMQO CoCl2 mM ΔnikA + nhlF ΔcorA + cbiKMQO ΔcorA + nhlF Fe Mg Fe Mg Fe Mg Fe Mg 0 0.307 ± 0.022 3.198 ± 0.207 0.323 ± 0.005 3.265 ± 0.125 0.051 ± 0.003 0.589 ± 0.019 0.056 ± 0.004 0.614 ± 0.069 0.005 0.381 ± 0.009 2.975 ± 0.109 0.379 ± 0.080 3.251 ± 0.265 0.1 0.349 ± 0.067 2.939 ± 0.195 0.320 ± 0.009 2.975 ± 0.170 0.058 ± 0.004 0.639 ± 0.043 0.052 ± 0.004 0.623 ± 0.137 0.5 0.068 ± 0.002 2.913 ± 0.137 0.080 ± 0.003 2.941 ± 0.246 0.055 ± 0.008 0.679 ± 0.141 0.031 ± 0.001 0.352 ± 0.067 0.050 ± 0.002 0.471 ± 0.103 0.022 ± 0.008 0.461± 0.206 0.01 0.02 0.05 1 Cellular Iron was measured in cells grown in M9-cam broth until OD600 reached 0.8-1. L-arabinose and CoCl2 were added and cells were harvested at stationary phase. The results shown here represent the mean ± standard deviation of ICP-OES analysis of three independent experiments. 71 3.7.2.3. Intracellular Co2+, iron and magnesium at exponential phase Table 13 is a summary of results of intracellular Co2+, iron and magnesium obtained in cells grown until exponential phase in 0.1 mM CoCl2 and 0.36 mM iron. The amount of cellular Co2+ in E. coli EPI 300 at 0.1 mM CoCl2 was almost 3 fold lower compared to the value obtained at stationary phase. This may be due to the up regulation of the Co2+ efflux protein at log phase (Rolfe et al. 2011). The amount of intracellular iron was slightly increased compared to values obtained at stationary phase, indicating that iron is accumulated at lag phase and exponential phase (Ralfe et al. 2011). Increasing iron concentration in the medium resulted in the increase in intracellular accumulation. However, unlike with Ni2+, the amount of intracellular Co2+ was not decreased in the presence of 0.36 mM medium iron concentration, suggesting that this concentration of iron does not inhibit Co2+ uptake in EPI 300. Both NhlF and CbiKMQO do not seem to transport iron and magnesium. Magnesium slightly increased from 2.905 ± 0.103 mg/g at stationary phase to 3.656 ± 0.215 mg/g at exponential phase, indicating that most of this metal is accumulated at exponential phase. 72 TABLE 13: Exponential phase intracellular Co2+, iron and magnesium in EPI 300 habouring cbiKMQO or nhlF Metals Cellular Co (mg/g dry weight) Cellular Fe (mg/g dry weight) Cellular Mg (mg/g dry weight) EPI 300 + nhlF EPI 300 + cbiKMQO EPI 300 + nhlF EPI 300 + cbiKMQ EPI 300 + nhlF EPI 300 + cbiKMQ 0 mM 0.000 ± 0.000 0.000 ± 0.000 0.410 ± 0.553 0.411 ± 0.023 3.711 ± 0.523 3.656 ± 0.321 0.1 mM Co 0.321 ± 0.185 0.239 ± 0.004 0.470 ± 0.050 0.406 ± 0.008 3.705 ± 0.608 3.642 ± 0.061 0.1 mM Co+ 0.361 mM Fe 0.399 ± 0.014 0.368 ± 0.039 0.577 ± 0.269 0.517 ± 0.022 3.575 ± 0.892 3.697 ± 0.100 Cellular Co2+, iron and magnesium were determined in cells grown until early log phase in M9-cam medium followed by addition of Co2+ or Co2+ and iron. The results shown here represent the mean ± standard deviation of ICP-OES analysis of three independent experiments. 73 3.7.2.4. Suitable strain for Co2+ doping of the magnetosomes The most cellular Co2+ was accumulated by the ΔcorA mutant (1.527 ± 0.086 mg/g) harboring cbiKMQO at 1 mM CoCl2, although no of the other mutants was tested at this Co2+ concentration. This amount is comparable with accumulation by the ΔnikA mutant harboring either cbiKMQO (1.492 ± 0.185 mg/g) or nhlF (1.450 ± 0.185 mg/g) at 0.5 mM CoCl2. However, at 0.5 mM Co2+ concentration, the amount of intracellular iron was decreased in ΔnikA, suggesting that this concentration of Co2+ may not be suitable for magnetosome doping. At 0.1 mM CoCl2, the ΔnikA (cbiKMQO) and EPI 300 (cbiKMQO) strains accumulates the most intracellular Co2+ (0.693 ± 0.093 mg/g and 0.673 ± 0.011 mg/g respectively) compared to other strains. As with the Ni2+ accumulating strains, these experiments were conducted in strains without the MIA making it difficult to indicate with certainty the amount of intracellular Co2+ which will result in magnetosome doping. However, the amount obtained in EPI 300 and ∆nikA mutant harboring cbiKMQO at 0.1 mM CoCl2 is the best obtained and therefore the most suitable for magnetosome doping. 74 CHAPTER FOUR 4. CONCLUSIONS The magnetosomes, produced by MTBs, are the most attractive alternative source of nontoxic and biocompatible magnetic nanoparticles (MNPs). However, magnetosomes are generally “soft” magnetic materials which can be used for some applications while other applications require chemically altered magnetic nanoparticles (Li et al. 2009). Magnetotactic bacteria (MTB) are fastidious in growth (Blakemore, 1975), making the production of magnetosomes on a commercial scale difficult. Synthesis of the magnetosome in an easy to culture E. coli strain by expressing the magnetosome gene island (MIA) is one of the possibilities for commercial scale synthesis. The magnetic properties of this recombinant magnetosome may be altered by doping with divalent metals such as Ni2+ (Kundu et al. 2009) or Co2+ (Staniland et al. 2008). In this study two E. coli strains which accumulate high intracellular Ni2+ or Co2+ were developed to be used as hosts for the MIA in order to produce magnetosomes with altered magnetic properties. These strains were studied for intracellular Ni2+ and Co2+ accumulation in growth medium metal concentrations known to induce magnetosome doping. The amount of intracellular Ni2+ or Co2+ accumulated by these strains was found to be higher than the amounts obtained in other published studies. However, these strains were not experimentally studied for the synthesis of chemically altered magnetosomes because they still lack the MIA. This makes it imposible to state conclusively if the amount of Ni2+ or Co2+ accumulated can result in the alteration of the magnetic properties of the magnetosomes. A conclusive study would have involved co-expressing the MIA and metals uptake gene followed by analyzing magnetic properties of the magnetosome. 75 Although the techniques and strategies employed in this study were sufficient to develop and screen for strains which accumulate high intracellular Ni2+ and Co2+, some improvement These include Ni2+ and Co2+ intracellular could have provided more valuable data. accumulation experiments conducted under microaerobic conditions. Intracellular Ni2+ and Co2+ accumulation in this study was conducted under aerobic conditions. This could be a problem because magnetosome formation occurs under microaerobic conditions. The intracellular ratio of free metals and protein bound metals changes under microaerobic conditions. For example, Ni2+ is required in E. coli mainly for the synthesis of hydrogenase enzymes which are produced during anaerobic and microaerobic metabolism (Rowe et al. 2005). This suggests that the metabolic demand of Ni2+ increases under microaerobic conditions, which may result in low concentration of free Ni2+ ions available for incorporation into the magnetosome. In addition, measuring the amount of intracellular free metals would have also provided a strong indication of the metals available in the cell for incorporation into the magnetosome. The amounts of free metals in the cell are important in this study because they represent the cellular “quota” of metals available for magnetosome doping. Free metal ions can be determined by using electron paramagnetic resonance spectroscopy (EPS). This study used the ICP-OES technique which determines the total amount of free metals in the cell. This technique could have been combined with the EPS technique to quantify the amount of free Ni2+ or Co2+ and the total amount of these metals inside the cell. The total intracellular amount of Ni2+ or Co2+ in MTB has not yet been studied under any conditions, making it difficult to make rational analysis on the amount of these metals required to induce magnetosome doping. It would be interesting to conduct a study in MTB 76 to determine the amount of intracellular Ni2+ and Co2+ and the amount contained in the magnetosome. Using EPS and ICP-OES techniques, the total intracellular amount of Ni2+ or Co2+ in MTB; the amount of Ni2+ or Co2+ inside the magnetosome; and the total amount of free Ni2+ or Co2+ radicals can be quantified. In general, the synthesis of magnetosomes in recombinant E. coli is still at its early research stage. Although E. coli metabolic activity has been well studied, most of the information about magnetosome formation is still hypothetical, and there is no experimental evidence that recombinant magnetosomes can be synthesized in hosts such as E. coli. However, with more studies on this area, the possibilities of recombinant magnetosome may become more attainable. The use of other organisms which share more metabolic similarities with MTB for magnetosome synthesis, other than using E. coli which is clearly different from MTB, might be an alternative. In addition, since magnetosomes are intracellular vesicle structures, using organisms which produce vesicles may also be a good approach. This study has produced two E. coli strains which accumulate high intracellular Ni2+ or Co2+. Provided the magnetosome gene island is successfully cloned and expressed in these strains, it is highly likely that the amount of metals accumulated can result in chemically altered magnetosomes which can be used for a variety of applications. Further studies are still required to analyse the amount of free metals in both MTB and E. coli. A key study would be to quantify the amount of intracellular free metals ions and the degree to which they change the magnetic coercive force of the magnetosome. 77 5. REFERENCES Abdul-Tehrani, H., A. J. Hudson, Y. S Chang, A. R. Timms, C. Hawkins, J. M. Williams, P. M. Harrison, J. R. Guest and S. C. Andrews. 1999. Ferritin mutants of Escherichia coli are iron deficient and growth impaired, and fur mutants are iron deficient. Journal of Bacteriology 181: 1415-1428. Acosta-Avalos, D., E. Wajnberg, P. S. Oliveira, I. I. Leal, M. Farina, and D. M. Esquivel. 1999. Isolation of magnetic nanoparticles from pachycondyla marginata ants. Journal of Experimental Biology 202:2687-2692. Addy, C., M. Ohara, F. Kawai, A. Kidera, M. Ikeguchi, S. Fuchigami, M. Osawa, I. Shimada, S. Y. Park, J. R. Tame, and J. G. Heddle. 2007. Ni2+ binding to NikA: an additional binding site reconciles spectroscopy, calorimetry and crystallography. Acta Crystallographica: Biological Crystallography 63:221-229. Agranoff, D. D., and S. Krishna. 1998. Metal ion homeostasis and intracellular parasitism. Molecular Microbiology 28:403-412. Amann, R., J. Peplies, and D. Schuler. 2006. Diversity and taxonomy of Magnetotactic bacteria. Applied Microbiology 3:25-35. Andrews, S.C., A. K. Robinson, F. Rodriguez-Quinones. 2003. Bacterial iron homeostasis. FEMS Microbiology Reviews 27: 215-237. 78 Artus, M., L. B. Tahar, F. Herbst, L. Smiri, F. Villain, Y. Yaacoub, J. M. Gren`eche, S. Ammar, and F. Fievet. 2011. Size-dependent magnetic properties of CoFe2O4 nanoparticles prepared in polyol. Journal of Physics 23:9-19. Auer, M., M. J.Kim, M. J. Lemieux, A. Villa, J. Song, X. D. Li, and D. N. Wang. 2001. High-yield expression and functional analysis of Escherichia coli Glycerol-3-phosphate transporter. Biochemistry 40: 6628-6635. Baek, Y. W., and Y. J. An. 2011. Microbial toxicity of metal oxide nanoparticles (CuO, NiO, ZnO, and Sb2O3) to Escherichia coli, Bacillus subtilis, and Streptococcus aureus. Science of the total Environment 409:1603-1608. Baker, G. C., Smith, J. J. and Cowan, D. A. 2003. Review and re-analysis of domainspecific 16S primers. Journal of Microbiological Methods 55: 541-555. Barton, L.L., F. Goulhen, M. Bruschi, N. A. Woodards, R. M. Plunkett, and F. J. M. Rietmeijer. 2007. The bacterial metallome: composition and stability with specific reference to the anaerobic bacterium Desulfovibrio desulfuricans. Biometals 20: 291-302 Bauerfeind, P., R. M. Garner, and L. T. Mobley. 1996. Allelic exchange mutagenesis of nixA in Helicobacter pylori results in reduced Ni2+ transport and urease activity. Infectious Immunology 64:2877-2880. 79 Bazylinski, D. A., A. J. Dean, D. Schuler, E. J. P. Phillips, and D. R. Lovley. 2000. N2dependent growth and nitrogenase activity in the metal-metabolising bacteria, Geobacter and Magnetospirillum species. Environmental Microbiology 2:266-273. Bazylinski, D. A., A. J. Dean, T. J. Williams, L. Kimble Long, S. L. Middleton, and B. L. Dubbels. 2004. Chemolithoautotrophy in the marine Magnetotactic bacteria strains MV-1 and MV-2. Archives in Microbiology 182:373-387. Bazylinski, D. A., R. B. Frankel, and H. W. Jannasch. 1988. Anaerobic production of magnetite by marine magnetotactic bacteria. Nature 334:518-519. Bazylinski, D. A., and S. Schubbe. 2007. Controlled biomineralisation and application of magnetotactic bacteria. Advanced Applied Microbiology 62:21-62. Bellini, S. 1963. On a unique behavior of fresh water bacteria. Institution of Microbiology, University of Pavia. Berreswill, S., S. Greiner, A. H. M. Van Vliet, B. Waidner, F. Fassibinder, E. Schiltz, and J. G. Kusters. 2000. Regulation of ferritin-mediated cytoplasmic iron storage by the ferric uptake regulator homolog (Fur) of Helicobacter pylori. Journal of Bacteriology 182: 5948-5953. Berry, C. C., and A. S. G. Curtis. 2003. Functionalisation of magnetic nanoparticles for applications in biomedicine. Journal of Physics 36:198-206. 80 Blakemore, R. P. 1975. Magnetotactic bacteria. Science 190:377-379. Blakemore, R. P. 1982. Magnetotactic bacteria. Annual Review of Microbiology 36:217238. Bleriot, C., G. Effantin, F. Lagarde, M. Mandrand-Berthelot, and A. Rodrigue. 2011. RcnB Is a Periplasmic Protein Essential for Maintaining Intracellular Ni and Co Concentrations in Escherichia coli. Journal of Bacteriology 193: 3785-3793. Bosse´, J. T., H. D. Gilmour, and J. I. MacInnes. 2001. Novel genes affecting urease acivity in Actinobacillus pleuropneumoniae. Journal of Bacteriology. 183:1242–1247. Browne, P., and G. O'Cuinn. 1983. The purification and characterization of a proline dipeptidase from guinea pig brain. Journal of Biology and Chemistry 258:6147-6154. Bruno, P. 1991. Theory of the curie temperature of Co2+ based ferromagnetic ultrathin films and multilayers. Proceedings of Magneto-Optical Recording lnternational Symposium 91:1520. Buhler, C., S. Yunoki, and A. Moreo. 2000. Magnetic domains and stripes in a spinfermion model for cuprates. Phycal Review Letters 84:2690-2693. Cabrera G., R. Perez, J. M. Gomez, A. Abalos, D. Cantero. 2005. Toxic effects of dissolved heavy metals on Desulfovibrio vulgaris and Desulfovibrio sp. strains. Journal of Hazardous Materials. 135: 40-46. 81 Chaintreuil, C., F. Rigault, L. Moulin, T. Jaffre´, J. Fardoux, E. Giraud, B. Dreyfus, and X. Bailly. 2007. Ni2+ resistance determinants in Bradyrhizobium strains from nodules of the endemic New Caledonia legume Serianthes calycina. Applied Environmental Microbiology: 8018-8022. Chakraverty, S., and M. Bandyopadhyay. 2007. Coercivity of magnetic nanoparticles: a stochastic model. Journal of Applied Physics 19:16-32. Chang, S. Y., E. C. McGary, and E. Chang. 1989. Methionine aminopeptidase gene of Escherichia coli is essential for cell growth. Journal of Bacteriology 171:4071–4072. Chen, Y. Y. M., and R. A. Burne. 2003. Identification and characterization of the Ni2+ uptake system for urease biogenesis in Streptococcus salivarius 57.I. Journal of Bacteriology. 185:6773–6779 Cheng, J., B. Poduska, R. A. Morton, and T. M. Finan. 2011. An ABC-type Co2+ transport system is essential for growth of Sinorhizobium meliloti at trace metal concentrations. Journal of Bacteriology. 193: 4405-4416 Chivers, P. T., and R. T. Sauer. 2002. NikR repressor: High-affinity Ni2+ binding to the Cterminal domain regulates binding to operator DNA. Chemistry and Biology 9:1141-1148. 82 Cole, S. T., R. Brosch, J. Parkhill, T. Garnier, C. Churcher, D. Harris, S. V. Gordon, K. Eiglmeier, S. Gas, C. E. Barry, F. Tekaia, K. Badcock, D. Basham, et al. 1998. Deciphering the biology of Mycobacterium tuberculosis from the complete genome sequence. Nature 393:537-544 Coleman, J. E. 1998. Zinc enzymes. Current Opinion in Chemistry and Biology 2:222-234. Cvetkovic, A., A. L. Menon, M. P. Thorgersen, et al. 2010. Microbial metalloproteomes are largely uncharacterized. Nature 466:779-782. de Medicis, E., J. Paquette, J. J. Gauther, and D. Shapcott. Magnesium and manganese content of hellophilic bacteria. 1986. Applied Environmental Microbiology 52: 567-573 Deng, X., Q.B. Li, Y.H. Lu, D.H. Sun, Y.L. Huang, X.R. Chen. 2003. Bioaccumulation of Ni2+ from aqueous solutions by genetically engineered Escherichia coli. Water Research. 37: 2505-2511. de Pina, K., C. Navarro, L. McWalter, D. H. Boxer, N. C. Price, S. M. Kelly, M. A. Mandrand-Berthelot, and L. F. Wu. 1995. Purification and characterization of the periplasmic Ni2+-binding protein NikA of Escherichia coli K12. European Journal of Biochemistry 227:857-865. de Souza, K. C., N. D. S. Mohallem, and E. M. B. de Sousa. 2011. Magnetic nanocomposites: Potential for applications in biomedicine. Quimica Nova 34:1692-1703. 83 Eitinger, T., and M. A. Mandrand-Berthelot. 2000. Ni2+ transport systems in microorganisms. Archives in Microbiology 173:1-9. Fantino, J. R., B. Py, M. Fontecave, and F. Barras. 2010. A genetic analysis of the response of Escherichia coli to Co2+ stress. Environmental microbiology 12:2846–2857. Friedberg, R., and D. I. Paul. 1975. New Theory of Coercive Force of Ferromagnetic Materials. Physical Review Letters 34:1415-1415. Frimpong, R. A., and J. Z. Hilt. 2010. Magnetic nanoparticles in biomedicine: synthesis, functionalization and applications. Nanomedicine (Lond) 5:1401-1414. Fu, C., S. Javedan, F. Moshiri, and R. J. Maiel. 1994. Bacterial genes involved in incorporation of Ni2+ into a hydrogenase enzyme. Proceedings of the National Academy of Science of the United States of America 91:5099-5103. Gelfand, S. M., and D. A. Rodionov. 2008. Comparative genomics and functional annotation of bacterial transporters. Physics of Life Review 5:22-49. Gikas, P., S.S. Sengor, T. Ginn, J. Moberly, and B. Peyton. 2009. Effect of heavy metals and temperature on microbial growth and lag phase. Global Nest Journal. 11: 325-332. Grass, G., C. Grosse, and D. H. Nies. 2000. Regulation of the cnr Co2+ and Ni2+ resistance determinant from Ralstonia sp. strain CH34. Journal of Bacteriology. 182:1390–1398 84 Grass, G., S. Franke, N. Taudte, D. H. Nies, L. M. Kucharski, M. E. Maguire, and C. Rensing. 2005. The metal permease ZupT from Escherichia coli is a transporter with a broad substrate spectrum. Journal of Bacteriology 187: 1604-1611 Hansen, M.C., Tolker-Neilson, T., Givskow, M. and Molin, S. 1998. Biased 16S rDNA PCR amplification caused by interference from DNA flanking template region. FEMS Microbiology Ecology 26: 141-149. Hantke, K. 1987. Ferrous iron transport mutants in Escherichia coli K-12. FEMS Microbiology Letters 44:53–57. Hanzlik, M., C. Heunemann, E. Holtkamp-Rötzler, M. Winklhofer, N. Petersen, and G. Fleissner. 2000. Superparamagnetic magnetite in the upper beak tissue of homing pigeons. Biological Metals 13:325-331. Hendricks, J. K., and H. L. Mobley. 1997. Helicobacter pylori ABC transporter: effect of allelic exchange mutagenesis on urease activity. Journal of Bacteriology 179:5892-5902. Heyen, U., and D. Schuler. 2003. Growth and magnetosome formation by microaerophilic Magnetospirillum strains in an oxygen-controlled fermentor. Applied microbial Biotechnology 61: 536-544. Hsu, C. Y., F. Y. Ko, C. W. Li, K. Fann, and J. T. Lue. 2007. Magnetoreception system in honeybees (Apis mellifera). Plos One 2:1-11. 85 Hyeon, T. 2002. Chemical synthesis of magnetic nanoparticles. The Royal Society of Chemistry 8: 297-234. Iwig, J. S., J. L. Rowe, and P. T. Chivers. 2006. Ni2+ homeostasis in Escherichia coli: The rcnR-rcnA efflux pathway and its linkage to NikR function. Molecular Microbiology 61:252– 262. Jiles, D. C., and D. L. Atherton. 1986. Theory of ferromagnetic hysteresis. Journal of Magnetism and Magnetic Materials 61:48-60. Kami, D., S. Takeda, Y. Itakura, S. Gojo, M. Watanabe, and M. Toyoda. 2011. Application of magnetic nanoparticles to gene delivery. International Journal of Molecular Science 12:3705-3722. Kaur, P. and B. P. Rosen. 1993. Complementation between nucleotide binding domains in an anion-translocating ATPase. Journal of Bacteriology. 175: 351-357. Kirchmayr, H. R. 1996. Permanent magnets and hard magnetic materials. Journal of Physics 29:2763–2778. Kirsten, A., M. Herzberg, A. Voigt, J. Seravalli, G. Grass, J. Scherer, and D. H. Nies. 2011. Contributions of five secondary metal uptake systems to metal homeostasis of Cupriavidus metallidurans CH34. Journal of Bacteriology 10:5293-5311. 86 Khlebnikov, A., K. A. Datsenko, T. Skaug, B. L. Wanner, and J. D. Keasling. 2001. Homogeneous expression of the pBAD promoter in Escherichia coli by constitutive expression of the low-affinity high-capacity AraE transporter. Microbiology 147: 3241-3247. Koch, D., D. H. Nies, and G. Grass. 2006. The RcnRA (YohLM) system of Escherichia coli: A connection between Ni2+, Co2+ and iron homeostasis. Biological Metals 20: 759-771 Komeda, H., M. Kobayashi, and S. Shimizu. 1997. A novel transporter involved in Co2+ uptake. Applied Biological Science 94:36-41. Krishnaswamy, R., and D. B. Wilson. 2000. Construction and characterisation of an Escherichia coli strain genetically engineered for Ni(II) bioaccumulation. Applied and Environmental Microbiology 66:8383-8386. Kung, F. C., J. Roymind, and D. A. Glaser. 1975. Metal iron content of Escherichia coli versus cell algae. Journal of Bacteriology 126: 1089-1095. Kundu, S., A. A. Kale, A. G. Banpurkar, G. R. Kulkarni, and S. B. Ogale. 2009. On the change in bacterial size and magnetosome features for Magnetospirillum magnetotacticum (MS-1) under high concentrations of zinc and Ni2+. Biomaterials 30: 4211-4218. Lee Macomber, L., and R. P. Hausinger. 2011. Mechanisms of Ni2+ toxicity in microorganisms. Metallomics 3:1153-1162. 87 Lefevre, C. T., F. Abreu, M. L. Schmidt, U. Lins, R. B. Frankel, B. P. Hedlund, and D. A. Bazylinski. 2010. Moderately thermophilic magnetotactic bacteria from hot spring in Nevada USA. Applied environmental Microbiology 76:3740-3743. Lefevre, C. T., R. B. Frankel, M. Posfai, R. B. Frankel, and D. A. Bandyopadhyay. 2011. Isolation of obligately alkaliphilic magnetostactic bacteria from extremely alikaline environments. Environmental Microbiology 13: 1462-1468 Li, J.H, Y. X. Pan, Q. S. Liu, H. F. Qin, C. L. Deng, R. C. Che, and X. A. Yang. 2009. A comparative study of magnetic properties between whole cells and isolated magnetosomes of Magnetospirillum magneticum AMB-1. Chinese Science Bulletin 55: 38-44. Liesegang, H., K. Lemke, R.A. Siddiqui and H. G. Schlegel. 1993. Characterization of the inducible Ni2+ and Co2+ resistance determinant cnr from pMOL28 of Alcaligenes eutrophus CH34. Journal of Bacteriology 175: 767-778. Linton, K. J., and C. F. Higgins. 2007. Structure and function of ABC transporters: the ATP switch provides flexible control. European Journal of Physiology 453:555-567. Lohmann, K. J., N. F. Putman, and C. M. F. Lohmann. 2008. Geomagnetic imprinting: A unifying hypothesis of long-distance natal homing in salmon and sea turtles. Proceedings of National Academy of Science of the United State of America 105:19096–19101. 88 Maaz, K., A. Mumtaz, S. K. Hasanain, and M. F. Bertino. 2010. Temperature dependent coercivity and magnetization of Ni2+ ferrite nanoparticles. Journal of Magnetism and Magnetic Material 322:2199-2202. Maaz, K., A. Mumtaz, S. K. Hasanain, and A. Ceylan. 2007. Synthesis and magnetic properties of Co2+ ferrite (CoFe2O4) nanoparticles prepared by wet chemical route. Journal of Magnetism and Magnetic Material 308:289-295. Macomber, L., and R. P. Hausinger. 2011. Mechanisms of Ni2+ toxicity in microorganisms. Metallomics 3:1153-1162. Mahmoudi, M., S. Sant, B. Wang, S. Laurente, and T. Sen. 2011. Superparamagnetic iron oxide nanoparticles (SPIONs): Development, surface modification and applications in chemotherapy. Advanced Drug Delivary Review 63:24-46. Mann, S., R. B. Frankel, and R. P. Blakemore. 1994. Structure, morphology and crystal growth of bacterial magnetite. Nature 405:405-407. Matsunaga, T., Y. Okamura, Y. Fukuda, A. T. Wahyudi, Y. Murase, and H. Takeyama. 2005. Complete genome sequence of the facultative anaerobic magnetotactic bacterium magnetospirillum sp. strain AMB-1. DNA Research 12: 157-166. McHenry, M. E., and D. E. Laughlin. 2000. Nano scale materials development for future magnetic applications. Acta Matter 48:223-238. 89 Mobley, H. L., R. M. Garner, and P. Bauerfeind. 1995. Helicobacter pylori Ni2+-transport gene nixA: Synthesis of catalytically active urease in Escherichia coli independent of growth conditions. Molecular Microbiology 16:97-109. Mulrooney, S. B., and R. P. Hausinger. 2003. Ni2+ uptake and utilization by microorganisms. FEMS Microbiology Reviews 27:239-261. Murat, D., A. Quinlan, H. Vali, and A. Komeili. 2010. Comprehensive genetic dissection of the magnetosome gene island reveales the step wise assembly of a prokaryotic organelle. National Academic of Science, USA 107:5593-5598. Nahvi, A., J. E. Barrick, and R. R. Breaker. 2004. Coenzyme B12 riboswitches are widespread genetic control elements in prokaryotes. Nucleic Acids Research 32:143-150. Navarro, C., L. F. Wu, and M. A. Mandrand-Berthelot. 1993. The nik operon of Escherichia coli encodes a periplasmic binding-protein-dependent transport system for Ni2+. Molecular Microbiology 9:1181-1191. Nelson, D. L., and E. P. Kennedy. 1971. Magnesium transport in Escherichia coli. Inhibition by Co2+ous ion. Journal of Biological Chemistry 246:3042-3049. Newcombe, J., J. C. Jeynes, E. Mendoza, J. Hinds, G. L. Marsden, R. A. Stabler, M. Marti, and J. J. McFadden. 2005. Phenotypic and transcriptional characterization of the meningococcal PhoPQ System, a magnesium-sensing two-component regulatory system that 90 controls genes involved in remodeling the meningococcal cell surface. Journal of Bacteriology 187: 4797-4975. Niegowski, D., and S. Eshaghi. 2007. The CorA family: structure and function revisited. Cellular and Molecular Life Sciences 64:2564-2574. Nies, D. H., M. Mergeay, B. Friedrich and H.G. Schlegel. 1987. Cloning of plasmid gene coding resistance to cadmium, zinc, and Co2+ from Alcaligenes eutrophus CH34. Journal of Bacteriology 167: 4865-4868. Nies, D. H., L. Chu and S. Silver. 1989. Expression and nucleotide sequence of a plasmiddetermined divalent cation efflux system from Alcaligenes eutrophus. National Academy of Science USA 86: 7351-7355. Nies, D. H. 1992. CzcR and CzcD, gene products affecting regulation of resistance to Co2+, zinc, and cadmium (czc system) in Alcaligenes eutrophus. Journal of Bacteriology 174:81028110. Nies, D. H. and S. Silver. 1994. Ion efflux systems involved in bacterial metal resistances. Journal of Industrial Microbiology 14: 186-199. Nies, D. H. 1995. The Co2+, zinc, and cadmium efflux system CzcABC from Alcaligenes eutrophus functions as a cation-proton antiporter in Escherichia coli. Journal of Bacteriology 177:2707–2712. 91 Nies, D. H. 1999. Microbial heavy-metal resistance. Applied Microbiology and Biotechnology 51:730-750. Noinaj, N., M. Guillier, T. J. Barnard, and S. K. Buchanan. 2010. TonB-Dependent Transporters: Regulation, Structure, and Function. Annual Review of Microbiology 64:4360. Odermatt, A., H. Suter, R. Krapf and M. Solioz. 1993. Primary structure of two P-type ATPases involved in copper homeostasis in Enterococcus hirae. Journal of Biological Chemistry 268: 12775-12777. Outten, C.E., and T.V. O’Halloran. 2001. Femtomolar sensitivity of metalloregulatory proteins controlling zinc homeostasis. Science 292: 2488-2492 Pal, A., and A. K. Paul. 2010. Ni2+ uptake and intracellular localization in Cupriavidus pauculus KPS 201, native to ultramafic ecosystem. Advanced Bioscience and Biotechnology 1: 276-280. Papp, K. M., and M. E. Maguire. 2004. The CorA Mg2+ Transporter Does Not Transport Fe2+. Journal of Bacteriology 186:7653-7658. Perez-Gonzalez, T., T. Prozorov, A. Yebra-Rodriguez, D. A. Bazyliski, and C. Jimenezlopez. 2010. Mn incorporation in magnetosomes: New possibilities for the nanotechnological applications of biomagnetite. Resumen SEM 171-172. 92 Qi L, J. Li, W. Zhang, J. Liu, C. Rong, et al. 2012. Fur in Magnetospirillum gryphiswaldense Influences magnetosomes formation and directly regulates the genes involved in iron and oxygen metabolism. PLoS ONE 7(1): e29572. doi:10.1371/journal.pone.0029572 Ragsdale, W. E., and M. Kumar. 1996. Ni2+-containing carbon monoxide dehydrogenase/acetyl-CoA synthase. Chemistry Reviews 96:2515-2539. Rolfe, D. M., C. J. Rice, S. Lucchini, C. Pin, A. Thompson, A. D. S. Cameron, M. Alston, M. F. Stringer, R. P. Betts, J. Baranyi, M. W. Peck, and J. C. D. Hinton. 2011. Lag phase is a distinct growth phase that prepares bacteria for exponential growth and involves transient metal accumulation. Journal of Bacteriology 194: 686-701. Ranquet, C., S. Ollagnier-de-Choudens, L. Loiseau, F. Barras, and M. Fontecave. 2007. Co2+ stress in Escherichia coli: The effect on the iron-sulfur proteins. Journal of Biological Chemistry 282:30442-30451. Rensing, C., Mitra, B., and P.B. Rosen. 1997. The zntA gene of Escherichia coli encodes a Zn(II)-translocating P-type ATPase. Proceedings of the National Academy of Sciences. 94: 14326-14331. Rodionov, D. A., A. G. Vitreschak, A. A. Mironov, and M. S. Gelfand. 2003. Comparative genomics of the vitamin B12 metabolism and regulation in prokaryotes. Journal of Biological Chemistry 278:41148-41159. 93 Rodionov, D. A., P. Hebbeln, M. S. Gelfand, and T. Eitinger. 2006. Comparative and functional genomic analysis of prokaryotic Ni2+ and Co2+ uptake transporters: Evidence for a Novel Group of ATP-Binding Cassette Transporters. Journal of Bacteriaology 188:317-327. Rodrigue, A., G. Effantin, and M. A. Mandrand-Berthelot. 2005. Identification of rcnA (yohM), a Ni2+ and Co2+ resistance gene in Escherichia coli. Journal of Bacteriology 187:2912–2916. Rowe, J. L., G. L. Starnes, and P. T. Chivers. 2005. Complex transcriptional control links NikABCDE-dependent Ni2+ transport with hydrogenase expression in Escherichia coli. Journal of Bacteriology 187:6317-6323. Rubikas, J., D. Matulis, A. Leipus, and D. Urbaitiene. 1997. Ni2+ resistance in Escherichia coli V38 is dependent on the concentration used for induction. FEMS Microbiology Letters 155:193-198. Schauer, K., B. Gouget, M. Carriere, A. Labigne, and H. de Reuse. 2007. Novel Ni2+ transport mechanism across bacterial outer membrane energized by the TonB/ExbB/ExbD machinery. Molecular Biology 64: 1054-1068. Schleifer, K. H., D. Schuler, S. Spring, M. Weizenegger, R. Amann, W. Ludwig, and M. Kohler. 1991. The genus Magnetospirillum, description of Magnetospirillum gryphiswaldense and transfer of Aquaspirillum magnetotacticum to Magnetospirillum magnetotacticum comb. Applied Microbiology 14:379-385. 94 Schmidt, T. and H. G. Schlegel. 1994. Combined Ni2+-Co2+-cadmium resistance encoded by the ncc locus of Alcaligenes xylosoxidans 31A. Journal of Bacteriology. 176: 7045-7054. Sebbane, F., M. Mandrand-Berthelot, M. Simonet. 2002. Genes encoding specific Ni2+ transport systems flank the chromosomal urease locus of pathogenic yersiniae. Journal of Bacteriology. 184: 5706-5713. Sengor, S. S., S. Barua, P. Gikas, T. R. Ginn, B. Peyton, R. K. Sani, and N. F. Spycher. 2009. Influence of Heavy Metals on Microbial Growth Kinetics Including Lag Time: Mathematical Modeling and Experimental Verification. Environmental Toxicology and Chemistry 28:2020-2029. Sensfuss, C. and H. G. Schlegel. 1988. Plasmid pMOL28-encoded resistance to Ni2+ is due to specific efflux. FEMS Microbiology Letters. 55: 295-298. Sermon, J., E. M. Wevers, L. Jansen, P. D. Spiegeleer, K. Vanoirbeek, A. Aertsen, and C. W. Michiels. 2005. CorA affects tolerance of Escherichia coli and Salmonella enterica Serovar Typhimurium to the lactoperoxidase enzyme system but not to other forms of oxidative stress. Applied Environmental Microbiology 71:6515-6523. She, Q., R. K. Singh, F. Confalonieri, Y. Zivanovic, G. Allard, M. J. Awayez, C. C. Y. Chan-Weiher, I. G. Clausen, B. A. Curtis, A. De Moors, G. Erauso, C. Fletcher, P. M. K. Gordon, I. Heikamp-de Jong, A. C. Jeffries, C. J. Kozera, N. Medina, X. Peng, and J. Van der Oost. 2001. The complete genome of the crenarchaeon Sulfolobus solfataricus P2. 95 Proceedings of the National Academy of Science of the United States of America 98:78357840. Siddiqui, R. A., K. Benthin and H. G. Schlegel. 1989. Cloning of pMOL28-encoded Ni2+ resistance genes and expression of the genes in Alcaligenes eutrophus and Pseudomonas spp. Journal of Bacteriology 171: 5071-5078. Silver, S., G. Ji, S. Brer, S. Dey, D. Dou and B. P. Rosen. 1993. Orphan enzyme or patriarch of a new tribe: the arsenic resistance ATPase of bacterial plasmids. Molecular Microbiology. 8: 637-642. Silver, S., and T. Phung 2005. A bacterial view of the periodic table: genes and proteins for toxic inorganic ions. Journal of Industrial Microbiology and Biotechnology 32:587-605. Snavely, D. M., S. A. Gravina, T. T. Cheung, C. G. Miller, and M. Maguire. 1990. Magnesium transport in Salmonella typhimurium. Journal of Biological Chemistry 266: 824829. Staniland, S., W. Williams, N. Telling, G. Van Der Laan, A. Harrison, and B. Ward. 2008. Controlled Co2+ doping of magnetosomes in vivo. Nature Nanotechnology 3:158-162. Tanaka, M., R. Brown, N. Hondow, A. Arakati, T. Matsunaga, and S. Staniland. 2012. Highest levels of Cu, Mn and Co doped into nanomagnetic magnetosomes through optimized biomineralisation. Journal of Materials Chemistry 22:11919-11921. 96 Tanaka, M., Y. Okamura, A. Arakati, T. Tanaka, H. Takeyama, and T. Matsunaga. 2006. Origin of magnetosome membranes: Proteomics analysis of magnetosome membrane and comparison with cytoplasmic membrane. Proteomics 6:5234-5247. Tibazarwa, C., S. Wuertz, M. Mergeay, L. Wyns, and D. Van der Lilie. 1999. Regulation of the cnr Co2+ and Ni2+ resistance determinant of Ralstonia eutropha (Alcaligenes eutrophus) CH34. Journal of Bacteriology. 182: 1399-1409. Tommi, K. M. and T. Kamehisa. 1998. A comprehensive analysis of ABC transporters in complete microbial genomes. Genome Research 8: 1045-1059. Vali, H., O. Forster, G. Amarantidis, and N. Petersen. 1987. Magnetotactic bacteria and their magnetofossils in sediments. Earth Planet Science Letter 86:389-400. Van Belle, F., T. J. Hayward, and J. A. C. Blanda. 2007. Coercivity engineering of exchange biased magnetic multilayer samples for digital encoding applications. Journal of Applied Physics 102:17-27. Viret, H., O. Pringault, and R. Duran. 2006. Impact of zinc and Ni2+ on oxygen consumption of benthic microbial communities assessed with microsensors. Science of the Total Environment 367:302-311. Waldron, K. J., J. C. Rutherford, D. Ford, and N. J. Robinson. 2009. Metalloproteins and metal sensing. Nature 460:823-830. 97 Wang, D.N., M. Safferling, M. J. Lemieux, H. Griffith, Y. Chen, X.D. Li. 2003. Practical aspects of overexpressing bacterial secondary membrane transporters for structural studies. Biochemisty and Biophysics Acta. 1610: 23-36. Watt, R. K., and P. W. Ludden. 1999. Ni(2+) transport and accumulation in Rhodospirillum rubrum. Journal of Bacteriology 181:4554-4560. Wolfram, L., F. B. and T. Eitinger. 1995. The Alcaligenes eutrophus protein HoxN mediates Ni2+ transport in Escherichia coli. Journal of Bacteriology 177:1840-1843. Wu, L. F., C. Navarro, K. de Pina, M. Quénard, M. Mandrand. 1994. Antagonistic effect of Ni2+ on the fermentative growth of Escherichia coli K-12 and comparison of Ni2+ and Co2+ toxicity on the aerobic and anaerobic growth. Environmental Health Perspectives. 102: 297300. Wu, A., P. Ou, and L. Zeng. 2010. Biomedical application of magnetic nanoparticles. Nanotechnology Review 5:245-270. Yamamoto, T., Y. Shimotsuma, M. Sakakura, M. Nishi, K. Miura, and K. Hirao. 2011. Intermetallic magnetic nanoparticle precipitation by femtosecond laser fragmentation in liquid. The ACS Journal of Surfaces and Colloids 27:8359-8364. Yan, L., S. Zhang, P. Chen, H. Liu, H. Yin, and H. Li. 2012. Magnetotactic bacteria, magnetosomes and their application. Microbialogy Research: 167: 2507-519. 98 Youn, H. D., E. J. Kim, J. H. Roe, Y. C. Hah, and S. O. Kang. 1996. A novel Ni2+containing superoxide dismutase from Streptomyces spp. Biochemistry Journal 318:889-896. 99 6. APPENDICES APPENDIX 1: Stationary phase intracellular Ni2+ accumulation in E. coli strains harboring hoxN Cellular Ni mg/g dry weight NiCl2 mM EPI 300 EPI 300+hoxN ΔnikA ΔnikA +hoxN ΔcorA ΔcorA +hoxN ΔrcnA ΔrcnA +hoxN 0 0.000 ± 0.000 0.000 ± 0.000 0.000 ± 0.000 0.000 ± 0.000 0.000 ± 0.000 0.000 ± 0.000 0.000 ± 0.000 0.000 ± 0.000 0.005 0.002 ± 22E-5 0.022 ± 82E-5 0.003± 6E-6 0.031 ± 41E-5 0.003 ± 3E-6 0.063 ± 0.002 0.01 0.005 ± 27E-5 0.089 ± 0.002 0.02 0.019 ± 0.009 0.091 ± 0.001 0.05 0.028 ± 21E-7 0.126 ± 0.007 0.1 0.154 ± 0.035 0.394 ± 0.031 0.5 0.135 ± 0.084 0.366 ± 0.034 0.064 ± 0.002 0.265 ± 0.026 1.087 ± 0.240 1.289 ± 0.278 0.132 ± 0.016 0.500 ± 0.034 0.673 ± 0.093 1.109 ± 0.107 1 Cellular Ni2+ was measure in cells grown in M9-cam medium. The culture was grown until OD600 reached 0.8-1, followed by addition of up to 0.2% L-arabinose and NiCl2 and harvested at the end of stationary phase. The results shown here represent the mean ± standard deviation of ICP-OES analysis of three independent experiments. 100 APPENDIX 2: Stationary phase intracellular Fe in E. coli strains harboring hoxN in the presence of Ni2+ Cellular Fe mg/g dry weight NiCl2 mM EPI 300 hoxN ΔnikA ΔnikA + hoxN ΔcorA ΔcorA + hoxN ΔrcnA ΔrcnA + hoxN No metal 0.307 ± 0.071 0.329 ± 0.036 0.330 ± 0.013 0.369 ± 0.088 0.054 ± 0.004 0.084 ± 0.006 0.335 ± 0.072 0.340 ± 0.037 0.005 0.360 ± 0.050 0.373 ± 0.006 0.389 ± 0.059 0.336 ± 0.015 0.315 ± 0.063 0.351 ± 0.093 0.01 0.358 ± 0.068 0.300 ± 0.089 0.02 0.378 ± 0.015 0.282 ± 0.083 0.05 0.350 ± 0.014 0.372 ± 0.001 0.1 0.298 ± 0.041 0.320 ± 0.090 0.5 0.344 ± 0.038 0.332 ± 0.002 0.043 ± 0.002 0.074 ± 0.008 0.092 ±0.002 0.082 ± 0.002 0.028 ± 0.001 0.055 ± 0.003 0.016 ± 0.006 0.033 ± 0.008 1 Cellular Iron was measure in cells grown in M9-cam broth until OD600 reached 0.8-1. L-arabinose and NiCl2 were added and cells were harvested at late stationary phase. The results shown here represent the mean ± standard deviation of ICP- OES analysis of three independent experiments. 101 APPENDIX 3: Stationary phase intracellular Mg in E. coli strains harboring hoxN in the presence of Ni2+ Cellular Mg mg/g dry weight NiCl2 mM EPI 300 hoxN ΔnikA ΔnikA + hoxN ΔcorA ΔcorA + hoxN ΔrcnA ΔrcnA + hoxN 0 3.050 ± 0.079 3.126 ± 0.341 2.880 ± 0.013 2.988 ± 0.158 0.615 ± 0.117 0.594 ± 0.008 3.048 ± 0.144 2.982 ± 0.093 0.005 3.012 ± 0.055 3.049 ± 0.183 2.951 ± 0.170 2.900 ± 0.172 2.945 ± 0.039 2.946 ± 0.129 0.01 2.802 ± 0.114 2.855 ± 0.245 0.02 2.875 ± 0.284 2.814 ± 0.110 0.05 2.933 ± 0.089 2.945 ± 0.003 0.1 2.835 ± 0.338 2.919 ± 0.043 0.5 2.950 ± 0.232 2.906 ± 0.138 0.572 ± 0.019 0.582 ± 0.088 2.915 ± 0.060 2.953 ± 0.108 0.571 ± 0.067 0.551 ± 0.009 0.423 ± 0.046 0.541 ± 0.046 1 Cellular magnesium was measured in cells grown in M9-cam broth until OD600 reached 0.8-1. L-arabinose and NiCl2 were added and cells were harvested at late stationary phase. The results shown here represent the mean ± standard deviation of ICP-OES analysis of three independent experiments. 102 APPENDIX 4: Stationary phase intracellular Co2+ accumulation in E. coli strains harboring cbiKMQO Cellular Co mg/g dry weight CoCl mM EPI 300 EPI 300+ cbi ΔnikA ΔnikA + cbi ΔcorA ΔcorA + cbi ΔrcnA 0 0.000 ± 0.000 0.000 ± 0.000 0.000 ± 0.000 0.000 ± 0.000 0.000 ± 0.000 0.000 ± 0.000 0.000 ± 0.000 0.000 ± 0.000 0.005 0.007 ± 5E-5 0.049 ± 0.001 0.004± 8E-4 0.038 ± 0.001 0.004 ± 22E-5 0.088 ± 0.001 0.01 0.006 ± 91E-5 0.107 ± 0.004 0.02 0.013 ± 0.001 0.206 ± 0.002 0.05 0.038 ± 0.007 0.216 ± 0.006 0.1 0.127 ± 0.005 0.673 ± 0.011 0.5 0.161± 0.053 0.693 ± 0.093 0.053 ± 17E-5 0.294 ± 0.012 1.372 ± 0.256 1.492 ± 0.185 0.332 ± 0.047 0.866 ± 0.126 0.667 ± 0.021 1.527 ± 0.086 1 ΔrcnA + cbiKMQO Cellular Co2+ was measure in cells grown in M9-cam medium. The cells were grown until OD600 reached 0.8-1, followed by addition of up to 0.2% L-arabinose and CoCl2 and harvested at the end of stationary phase. The results shown here represent the mean ± standard deviation of ICP-OES analysis of three independent experiments. 103 APPENDIX 5: Stationary phase intracellular Co2+ accumulation in E. coli strains harboring nhlF Cellular Co mg/g dry weight CoCl2mM EPI 300 EPI 300 + nhlF ΔnikA ΔnikA + nhlF ΔcorA ΔcorA + nhlF ΔrcnA 0 0.000 ± 0.000 0.000 ± 0.000 0.000 ± 0.000 0.000 ± 0.000 0.000 ± 0.000 0.000 ± 0.000 0.000 ± 0.000 0.000 ± 0.000 0.005 0.007 ± 2E-5 0.022 ± 31E-4 0.004± 21E-5 0.020 ± 1E-4 0.004 ± 33E-5 0.052± 46E-4 0.01 0.006 ± 2E-5 0.081 ± 3E-4 0.02 0.013 ± 0.001 0.133 ± 0.002 0.05 0.038 ± 0.007 0.187 ± 0.003 0.1 0.127 ± 0.005 0.425 ± 0.013 0.5 0.161± 0.053 0.477 ± 0.024 0.053 ± 13E-4 0.191 ± 0.017 1.372 ± 0.256 1.450 ± 0.185 0.332 ± 0.047 0.730 ± 0.138 0.667 ± 0.021 1.197 ± 0.077 1 ΔrcnA + nhlF Cellular Co2+ was measure in cells grown in M9-cam medium. The cells were grown until OD600 reached 0.8-1, followed by addition of up to 0.2% L-arabinose and CoCl2 and harvested at the end of stationary phase. The results shown here represent the mean ± standard deviation of ICP-OES analysis of three independent experiments. 104 APPENDIX 6: Stationary phase intracellular iron in E. coli strains harboring cbiKMQO in the present of Co2+ Cellular Fe mg/g dry weight CoCl2 mM EPI 300 EPI 300+ cbi ΔnikA ΔnikA + cbi ΔcorA ΔcorA + cbi ΔrcnA ΔrcnA + cbi 0 0.307 ± 0.071 0.326 ± 0.069 0.330 ± 0.013 0.307 ± 0.022 0.054 ± 0.014 0.051 ± 0.003 0.335 ± 0.003 0.343 ± 0.010 0.005 0.341 ± 0.024 0.395 ± 0.035 0.380 ± 0.061 0.381 ± 0.009 0.361 ± 0.092 0.386± 0.094 0.01 0.321 ± 0.059 0.369 ± 0.046 0.02 0.318 ± 0.029 0.333 ± 0.041 0.05 0.368 ± 0.018 0.318 ± 0.064 0.1 0.369 ± 0.069 0.388 ± 0.042 0.5 0.325 ± 0.697 0.349 ± 0.067 0.061 ± 0.005 0.058 ± 0.004 0.058 ± 0.003 0.068 ± 0.002 0.053 ± 0.002 0.055 ± 0.008 0.055 ± 0.008 0.050 ± 0.002 1 Cellular Iron was measure in cells grown in M9-cam broth until OD600 reached 0.8-1. L-arabinose and CoCl2 were added and cells were harvested at late stationary phase. The results shown here represent the mean ± standard deviation of ICP OES analysis of three independent experiments 105 APPENDIX 7: Stationary phase intracellular iron in E. coli strains harboring nhlF in the present of Co2+ Cellular Fe mg/g dry weight CoCl2 mM EPI 300 EPI 300+ nhlF ΔnikA ΔnikA + nhlF ΔcorA ΔcorA + nhlF ΔrcnA ΔrcnA + nhlF 0 0.307 ± 0.071 0.322 ± 0.045 0.330 ± 0.013 0.323 ± 0.005 0.051 ± 0.004 0.056 ± 0.004 0.335 ± 0.003 0.372 ± 0.075 0.005 0.341 ± 0.004 0.300 ± 0.011 0.380 ± 0.061 0.379 ± 0.080 0.361 ± 0.092 0.361 ± 0.012 0.01 0.321 ± 0.059 0.363 ± 0.059 0.02 0.318 ± 0.029 0.331 ± 0.060 0.05 0.368 ± 0.018 0.360 ± 0.054 0.1 0.369 ± 0.069 0.395 ± 0.022 0.5 0.325 ± 0.097 0.320 ± 0.009 0.036 ± 0.005 0.052 ± 0.004 0.088 ± 0.003 0.080 ± 0.003 0.020 ± 0.002 0.031 ± 0.001 0.019 ± 0.008 0.022 ± 0.008 1 Cellular Iron was measure in cells grown in M9-cam broth until OD600 reached 0.8-1. L-arabinose and CoCl2 were added and cells were harvested at late stationary phase. The nhlF gene shows no effect on cellular iron. The results shown here represent the mean ± standard deviation of ICP-OES analysis of three independent experiments 106 APPENDIX 8: Stationary phase intracellular magnesium in E. coli strains harboring cbiKMQO in the presence of Co2+ Cellular Mg mg/g dry weight CoCl2 mM EPI 300 EPI 300+ cbi ΔnikA ΔnikA + cbi ΔcorA ΔcorA + cbi ΔrcnA 0 2.906 ± 0.050 3.355 ± 0.545 2.880 ± 0.013 3.198 ± 0.207 0.615 ± 0.117 0.589 ± 0.019 3.048 ± 0.144 3.030 ± 0.193 0.005 2.963 ± 0.313 2.966 ± 0.249 2.939 ± 0.090 2.975 ± 0.109 2.925 ± 0.048 3.047 ± 0.110 0.01 2.950 ± 0.150 2.902 ± 0.118 0.02 2.959 ± 0.089 2.943 ± 0.228 0.05 2.956 ± 0.038 2.930 ± 0.198 0.1 2.905 ± 0.103 2.918 ± 0.113 0.5 2.981 ± 0.071 2.939 ± 0.195 0.414 ± 0.177 0.639 ± 0.043 2.888 ± 0.025 2.913 ± 0.137 0.303 ± 0.018 0.679 ± 0.141 0.287 ±0.036 0.471 ± 0.103 1 ΔrcnA + cbi Cellular Magnesium was measure in cells grown in M9-cam broth until OD600 reached 0.8-1. L-arabinose and CoCl2 were added and cells were harvested at late stationary phase. The results shown here represent the mean ± standard deviation of ICP-OES analysis of three independent experiments . 107 APPENDIX 9: Stationary phase intracellular magnesium in E. coli strains harboring nhlF in the presence of Co2+ Cellular Mg mg/g dry weight CoCl2 mM EPI 300 EPI 300+ nhlF ΔnikA ΔnikA + nhlF ΔcorA ΔcorA + nhlF ΔrcnA 0 2.906 ± 0.050 3.140 ± 0.063 2.880 ± 0.013 3.265 ± 0.125 0.615 ± 0.117 0.614 ± 0.069 3.048 ± 0.144 3.211 ± 0.089 0.005 2.963 ± 0.313 2.864 ± 0.008 2.939 ± 0.090 3.251 ± 0.265 2.925 ± 0.048 3.386 ± 0.554 0.01 2.950 ± 0.150 3.019 ± 0.475 0.02 2.959 ± 0.089 3019 ± 0.475 0.05 2.956 ± 0.038 3.180 ± 0.156 0.1 2.905 ± 0.103 3.302 ± 0.417 0.5 2.981 ± 0.071 2.975 ± 0.170 0.414 ± 0.177 0.623 ± 0.137 2.888 ± 0.025 2.941 ± 0.246 0.303 ± 0.018 0.352 ± 0.067 0.287 ±0.036 0.461± 0.206 1 ΔrcnA + nhlF Cellular Magnesium was measure in cells grown in M9-cam broth until OD600 reached 0.8-1. L-arabinose and CoCl2 were added and cells were harvested at late stationary phase. The results shown here represent the mean ± standard deviation of ICP-OES analysis of three independent experiments. 108 APPENDIX 10: Log phase intracellular Co2+, iron and magnesium in E. coli EPI 300 in the presence of iron and Co2+ Metals Cellular Co (mg/g dry weight) Cellular Fe (mg/g dry weight) Cellular Mg (mg/g dry weight) EPI 300 EPI 300 + cbiKMQO EPI 300 EPI 300 + cbiKMQQ EPI 300 EPI 300 + cbiKMQ No metal 0.000±0.000 0.000 ± 0.000 0.420 ± 0.001 0.411 ± 0.023 3.790 ± 0.235 3.656 ± 0.321 0.1 mM Co 0.043 ± 13E-6 0.239 ± 0.004 0.420 ± 0.076 0.406 ± 0.008 3.656 ± 0.215 3.642 ± 0.061 0.1 mM Co+ 0.361 mM Fe 0.057 ± 0.001 0.368 ± 0.039 0.585 ± 0.088 0.517 ± 0.022 3.631 ± 0.299 3.697 ± 0.100 Cellular Co2+, iron and magnesium were determined in cells grown until early log phase in M9-can medium followed be addition of Co2+ or Co2+ and iron. Results shown here represent the mean ± standard deviation of ICP-OES analysis of three independent experiments. 109