Optimisation of a High-throughput Screening Assay for Nitrile Hydratase Enzymes by Ateret Ben-David (587747) Dissertation Submitted in fulfilment of the requirements for the degree Master of Science in Molecular and Cell Biology in the Faculty of Science, University of the Witwatersrand, Johannesburg, South Africa Supervisor: Professor Karl Rumbold Co-supervisor: Professor Dean Brady June 2022 ii DECLARATION I, Ateret Ben-David (587747), am a student registered for the degree of Master’s by Dissertation (MSc) in the academic year 2021. I hereby declare the following: · I am aware that plagiarism (the use of someone else’s work without their permission and/or without acknowledging the original source) is wrong. · I confirm that the research proposal submitted for assessment for the above degree is my own unaided work except where explicitly indicated otherwise and acknowledged. · I have followed the required conventions in referencing the thoughts and ideas of others. · I understand that the University of the Witwatersrand may take disciplinary action against me if there is a belief that this is not my own unaided work or that I have failed to acknowledge the source of the ideas or words in my writing. Signature: 6 June 2022 iii ABSTRACT Nitrile hydratases (NHase) are important enzymes that can transform nitriles, which are low- cost intermediates, into high-value industrial chemicals like acrylamide. To increase the use of NHase as a biocatalyst in industry, it needs to be engineered to work in harsher conditions. A high-throughput screening assay must be developed before indirect evolution can be conducted so that enzyme activity of the mutants can be determined. In this study, a coupled-enzyme assay was employed to convert the amide generated by the NHase to hydroxamic acid, which could then be colorimetrically identified. The NHase was successfully transformed and expressed prior to assay optimisation. Multiple steps were engaged in the optimisation process, including examining the influence of substrate and product inhibition on the NHase and amidase, as well as the effect of temperature, pH, and organic solvent on the assay. The optimal conditions were determined to be 100 mM acetonitrile and 1.0 µg/µl NHase which were incubated for 10 minutes. After which 0.4 units of amidase and 20 mM hydroxylamine were added to the well and incubated for a further 30 minutes. The optimal incubation temperature and pH were determined to be 35˚C and pH 8. The study resulted in a sensitive, high-throughput screening assay for NHase enzymes. iv ACKNOWLEDGEMENTS I would like to thank my supervisors Prof. Karl Rumbold and Prof. Dean Brady for their guidance and support along the journey. I gratefully acknowledge that this project would not have been possible without the financial support from the National Research Fund (NRF). The biggest thank you to Sasha Richardson and Dylan Moodley for their patience, guidance and most importantly their friendship. Thank you to all my colleagues in Industrial Microbiology Laboratory (IMBL) and the Microbiology and Biotechnology department. Particularly thanks to Michael Tobin, who was always ready to help and made sure everything ran smoothly. Thank you to Dr Botes and Dr Meyer, the postgraduate coordinators, who guided me and where always ready to answer my questions. I would like to thank G-D for bringing me through this journey as without Him this would have not been possible. Thank you to my husband, Ory, who encourages and supports me in being a better person every day. This master is in dedication to you. To my beautiful, smart, kind kids, Noam, Eliyah and Yemima thank you for allowing me to do my Masters and for your patience throughout it. I love you all so much. I hope that my hard work encourages you to do hard things in life. Thank you to my parents Daniel and Raelene Ben-David who have been there throughout everything for me, providing me with encouragement, helping with the kids and helping financially when needed. To my dearest sister, Batya, thank you for everything. v This work is based on the research supported in part by the National Research Foundation of South Africa (Grant Numbers: 122062). vi TABLE OF CONTENTS DECLARATION ..................................................................................................................... ii ABSTRACT ............................................................................................................................. iii ACKNOWLEDGEMENTS ................................................................................................... iv LIST OF FIGURES .............................................................................................................. viii LIST OF SCHEMES ............................................................................................................... x LIST OF TABLES .................................................................................................................. xi LIST OF ABBREVIATIONS ............................................................................................... xii Chapter 1: Introduction .......................................................................................................... 1 1.1 Literature review .............................................................................................................. 1 1.1.1 The nitrile hydratase enzyme ..................................................................................... 1 1.1.2 Enzyme engineering .................................................................................................. 7 1.1.3 Screening and selection techniques ........................................................................... 8 1.2 Rationale......................................................................................................................... 10 1.3 Aim and objectives ......................................................................................................... 12 1.3.1 Aim .......................................................................................................................... 12 1.3.2 Objectives ................................................................................................................ 12 Chapter 2: Protein expression of Nitrile Hydratase from R. rhodochrous ATCC BAA- 870 using a two-plasmid vector system ................................................................................ 13 2.1 Introduction .................................................................................................................... 13 2.1.1 Recombinant protein expression and its effect on biocatalysis ............................... 13 2.1.2 Two-plasmid vector system used for recombination of the R. rhodochrous ATCC BAA-870 Nitrile Hydratase .............................................................................................. 13 vii 2.1.3 The host system used for Nitrile Hydratase expression .......................................... 15 2.2 Methods and materials: .................................................................................................. 17 2.2.1 Materials .................................................................................................................. 17 2.2.2 Methods ................................................................................................................... 20 2.3 Results and Discussion ................................................................................................... 24 2.3.1 Conformation of transformation .............................................................................. 24 2.3.2 Comparison of the growth curves and growth rates for each plasmid as well as the co-transformed cells ......................................................................................................... 24 2.3.2 Co-expression of the R. rhodochrous ATCC BAA-870 Nitrile Hydratase ............. 28 2.3.3 Transformation efficiencies for co-transformed and individually transformed plasmids ............................................................................................................................ 31 Chapter 3: Optimising a high-throughput screening assay for the R. rhodochrous ATCC BAA-870 nitrile hydratase enzyme ...................................................................................... 34 3.1 Introduction .................................................................................................................... 34 3.2 Materials and methods ................................................................................................... 36 3.2.1 Materials .................................................................................................................. 36 3.2.2 Methods ................................................................................................................... 37 3.3 Results and discussion .................................................................................................... 42 3.3.1 The effect of hydroxylamine on the assay ............................................................... 42 3.3.2 The optimal concentration of acetohydroxamic acid for maximum assay sensitivity. .......................................................................................................................................... 47 3.3.3 Optimisation of the amidase activity within the assay ............................................ 49 3.3.4 Optimisation of the NHase activity within the assay .............................................. 57 3.3.5 Optimisation of the assay temperature .................................................................... 63 3.3.7 The effect of organic solvents on the assay ............................................................. 66 3.3.8 The effect of different nitrile substrates on the assay .............................................. 69 3.3.9 Analysis of the kinetic parameters for NHase in this assay .................................... 71 Chapter 4: General discussion and conclusion ................................................................... 77 viii 4.1 Future work and suggestions…………………………………………………………81 REFERENCES ....................................................................................................................... 82 LIST OF FIGURES Figure 1.1: The active site of a Co-type nitrile hydratase from Pseudonocardia thermophila….3 Figure 1.2: A ribbon drawing of the tertiary structure of the Co-type nitrile hydratase from P. thermophila……………………………………………………………………………………4 Figure 2.1 Plasmid maps of pET-21a(+) with beta subunit insert and pRSFDuet-1 with alpha and chaperon inserts…………………………………………………………………………..19 Figure 2.2: Growth curves representing the change in biomass of differentaily transformed T7 Express competent E. coli cells in LB media measured at OD600……………………………26 Figure 2.3 Growth rates of the transformants in T7 Express competent E. coli……………….27 Figure 2.4: Co-Expression and individual expression of the nitrile hydratase alpha and beta subunits in T7 Express competent E. coli…………………………………………………………..30 Figure 2.5: Transformation effeciencies of co-transformed T7 Express competent E. coli with varied plasmid ratios, as well as independently transformed cells…………………………..32 Figure 2.6: Co-expression of T7 Express competent E. coli cells that were transformed using a 4:1 and a 1:1 ratio of pRSFDuet-1 to pET21a(+) plasmids…………………………………..33 Figure 3.1: The effect of hydroxylamine concentration on NHase activity…………………..44 Figure 3.2: The effect of hydroxylamine concentration on amidase activity…………………46 Figure 3.3: Standard curve to determine acetohydroxamic acid concentration………………48 Figure 3.4: Amidase activity as a function of acetamide concentration……………………….51 Figure 3.5: Amidase activity as a function of acetonitrile concentration……………………..53 Figure 3.6: The effect of amidase concentration on the rate of change of absorbance in time…56 Figure 3.7: The effect of acetonitrile concentration on the activity of NHase…………………58 Figure 3.8: The effect of acetamide concentration on the activity of NHase…………………60 ix Figure 3.9: The effect of NHase concentration on the rate of change of absorbance in time….62 Figure 3.10: Determining the temperature optimum for the assay……………………………64 Figure 3.11: Determining the pH optimum for the assay…………………………………….66 Figure 3.12: The effect of organic solvents on the assay……………………...………………68 Figure 3.13: The effect of different nitrile substrates on the assay……………………………70 Figure 3.14: The effect of reaction time on acetohydroxamic acid produced using different concentrations of acetonitrile…………………………………………………………………72 Figure 3.15: The Lineweaver-Burk plot for the NHase………………………………………74 Figure 3.16: The Eadie-Hofstee plot for the NHase………………………………………….74 Figure 3.17: The Hanes-Woolf plot for the NHase……………………………………………75 x LIST OF SCHEMES Scheme 1.1: Reactions catalysed by nitrile enzymes…………………………………………1 Scheme 3.1: NHase and amidase catalyse the conversion of nitrile to hydroxamic acid……36 Scheme 3.2: The effect of hydroxylamine on the assay. Amidase transfers the acyl group from the amide to hydroxylamine……………………………………………………………..……42 Scheme 3.3: Hydroxamic acid is the final, measurable product in the assay……………..…..47 Scheme 3.4: The amidase enzyme uses amides as substrates, and the nitrile substrate could be present during the process……………………………………………………………………49 Scheme 3.5: The amidase enzyme is the second enzyme in the coupled assay………………54 Scheme 3.6: Nitriles are substrates of the NHase enzyme while amides are the reaction’s products………………………………...…………………………………………………….57 Scheme 3.7: NHase is coupled with a second enzyme in order to determine its activity…….61 xi LIST OF TABLES Table 2.1: Summary of characteristics for the pRSFDuet-1 and pET21a(+) plasmids………………………………………………………………………………………15 Table 2.2: Summary of the characteristics of T7 Express competent E. coli and E. coli BL21(DE3) cells……………………………………………………………………………...17 Table 2.3: The E. coli strains used as hosts in this study……………………………………..18 Table 2.4: shows the concentration of antibiotics employed in the LB broth medium and LB agar plates for each transformant………………….………………………………………….20 Table 3.1 A summary of kinetic parameters for the NHase………………………………….76 Table 4.1: Summery of ideal parameters for performing the assay …………………………..81 xii LIST OF ABBREVIATIONS BRENDA Braunschweig ENzyme Database CFU colony forming units Co-NHase non-corrin cobalt-type nitrile hydratase CL crude lysate CP crude protein ER-PCR error prone -PCR E. coli Escherichia Coli ESS enzyme substrate substrate complex Fe-NHase nonheme iron-type nitrile hydratase FACS fluorescence-activated cell sorting HIV-1 HIV type 1 protease H-NHase high molecular mass nitrile hydratase HPLC high-performance liquid chromatography HTS high throughput screening IB inclusion bodies IPTG isopropyl β-D-1-thiogalactopyranoside kDa kilo Daltons Km Michaelis-Menten constant L-NHase Low molecular mass nitrile hydratase LB Luria Bertani min minute NHase nitrile hydratase xiii NMR nuclear magnetic resonance OD optical density P. thermophila Pseudonocardia thermophila PDB protein data base R. rhodochrous Rhodococcus rhodochrous RBS ribosome binding sight RET resonance energy transfer rpm rounds per minute SI substrate inhibition PAGE polyacrylamide electrophoresis SDS sodium dodecyl sulphate T7 RNAP T7 RNA polymerase TLC Thin-layer chromatography TLP-ste thermolysin-like protease UV ultra-violate Vmax maximum reaction rate 1 Chapter 1: Introduction 1.1 Literature review Many plants and soil microorganisms use nitriles in their metabolic processes therefore there is a plethora of enzymes found in nature that are able to metabolise nitrile compounds efficiently (Martínková et al., 2014; Yamada and Kobayashi, 1996). Nitrilase and nitrile hydratase (NHase), which both use nitrile as a substrate, are the two most important enzymes for converting nitrile. (Kobayashi et al., 1989). The enzymes use two distinct routes for nitrile hydrolysis (Scheme 1.1). Nitrilase catalyses the conversion of nitriles to carboxylic acid in the first pathway, while NHase catalyses the conversion of nitriles to amides in the second pathway, which are then swiftly converted to carboxylic acid in the presence of a second enzyme amidase (Kobayashi et al., 1993). Scheme 1.1: Reactions catalysed by nitrile enzymes. Nitrile hydratase converts nitrile into amides which are consequently converted to carboxylic acids via amidase. Alternatively, nitrilase can convert nitriles to carboxylic acid directly. 1.1.1 The nitrile hydratase enzyme The focus of this study was on NHase (EC 4.2.1.84) a metalloenzyme. NHase is composed of alpha and beta subunits that usually combine to form a α2β2 heterotetramer (Sasaki et al., 2002). The molecular masses of the alpha and beta subunits are 23 kDa and 26 kDa respectively. The amino acid sequences of the two subunits are not similar, although the alpha and beta subunits from different species are highly comparable. The beta subunit contains two highly conserved arginine residues which form a bond with the sulfur atoms in the cystine residues found in the alpha subunits. In this way a strong bond is formed ensuring the subunits are binding in a precise manner (Liu et al., 2012). The active site of the nitrile hydrates has conserved cysteine residues, CXXCSC, that are involved with the binding of the metal ion in 2 the centre. NHase was first discovered in Arthrobacter sp. J1 which was later renamed Rhodococcus rhodochrous (Asano et al., 1980). Although since then many other bacterial genera have been discovered to produce NHase, R. rhodochrous has remained the primary source of novel NHases. 1.1.1.1 Cobalt and iron-type NHases NHase is a metalloenzyme that requires metal ions in its active site. In order to have catalytic activity, a nonheme iron (Fe-NHase) or non-corrin cobalt (Co-NHase) is required. The metal ion is responsible for the catalytic hydration of the enzyme as well as the structural folding of the enzyme (Gong et al., 2017). The two types of NHases have a significant degree of sequence similarity, and their alpha subunits have the three conserved cysteine residues in the active area. Furthermore, they are structurally similar and share a common response mechanism (Gong et al., 2017). The alpha subunit provides all the protein ligands for the metal ion which is found at the interface between the alpha and beta subunits in the central cavity. The metal ion is located in a claw-like setting to deprotonated peptide nitrogens, a cysteine thiolate, sulfur atoms in the cysteine residues and a water molecule (Gumataotao et al., 2013; Hopmann, 2014). The cysteine residues are post translationally modified to cysteine sulfunic acid and cysteine sulfenic acid (Gumataotao et al., 2013; Hopmann, 2014). 1.1.1.1.1 Amino acid sequence in catalytic site of the cobalt and iron- type NHase NHases have a conserved sequence made up of 7 amino acids in their active site (Figure 1.1). However, the Fe-type and Co-type NHases differ by two amino acid residues at positions 109 and 114. The Fe-type NHase has a 'CSLCSCT' sequence in its active site, whereas the Co-type NHase has a 'CTLCSCY' sequence (Payne et al., 1997). The S and T in the Fe-type NHase stand for Ser109 and Thr114, respectively, whereas the T and Y in the Co-NHase stand for Thr109 and Tyr114. The Thr109 in the Co-type NHase forms a hydrophobic interaction with the side chain of Val36, however, in the Fe-type NHase the equivalent amino acid residue, Ser109, does not associate with Val36 (Frederick et al., 2006; Payne et al., 1997). In addition, the Tyr114 residue in the Co-type NHase establishes a hydrogen bond with Leu119 and Leu121 via a water molecule (Frederick et al., 2006; Payne et al., 1997). On the other hand, in the Fe-type NHase, the equivalent residue at position 114, Thr114, establishes a hydrogen bond with Ser113 found in the cysteine cluster. The hydrogen bond between Thr114 and Ser113 causes the residues to move closer together, making the active 3 site more open compared to the Co-type NHase (Frederick et al., 2006; Payne et al., 1997). The difference in conformation of the active site could be the reason for different substrate preferences between the two NHases. Cobalt-type NHase prefer aliphatic nitriles whereas the Fe-type NHase are able to catalyse aromatic nitrile substrates. Figure 1.1: The active site of a Co-type nitrile hydratase from Pseudonocardia thermophila. The metal ion forms 6 bonds with two deprotonated peptide nitrogens, a cysteine thiolate, two sulfur atoms in cysteine residues at the active site, and a water molecule in a claw- like configuration. The figure is reprinted from Mitra and Holz (2007). 1.1.1.1.2 Crystal structure of the Co-type NHase There are 39 X-ray crystal structures of NHases from bacterial species deposited in the PDB. Despite the number of NHases discovered in the Rhodococcal species, no Co-type NHase from this species has yet to be crystallized. The reason for this is that the Co-type NHase from Rhodococcal species may be a difficult protein target to crystallize. Below in Figure 1.2, is a ribbon drawing of the tertiary structure of the Co-type NHase from P. thermophila based on a crystal structure determined by Miyanaga (2001). 4 Figure 1.2: A ribbon drawing of the tertiary structure of the Co-type nitrile hydratase from P. thermophila. The NHase is a heterotetramer made up of α2β2 subunits. The alpha subunits are in orange and pink, while the beta subunits are in purple and green. The pink and purple unit and the green and orange unit represent one heterodimer. The image was reprinted from the protein data bank (PDB) (Bank, n.d.) and was originally constructed by Miyanaga (2001). 1.1.1.1.3 Catalytic mechanism Several catalytic mechanisms for NHase have been proposed based on synthetic and theoretical modelling studies, molecular dynamic simulations, substrate docking studies and X-ray crystal structures. One method that was proposed suggested that the nitrile compound binds directly to the metal ion in the centre of the active site causing the water molecule to be displaced. In this way the metal ion acts as a Lewis acid, by facilitating a nucleophilic attack by the water molecule on the carbon of the nitrile (Desai and Zimmer, 2004) 5 1.1.1.2 Low and high molecular mass NHase Within R. rhodochrous J1 two types of Co-NHase have been discovered, these being the high molecular mass nitrile hydratase (H-NHase) and a low molecular mass nitrile hydratase (L- NHase) (Kobayashi et al., 1991). The L-NHase is a 130 kDa protein made up of two alpha- beta heterodimers, each with one cobalt ion. Conversely, the H-NHase is a 520 kDa protein containing nine to ten pairs of alpha-beta subunits. The two proteins share a 42.89% sequence similarity (Frederick et al., 2020) and therefore the H-NHase is thought to have evolved through horizontal gene transfer and gene duplication. Although the amino acid residues, CTLCSC, that are located in the active site of all Co-NHases are conserved in both (Frederick et al., 2020). 1.1.1.3 Evaluation of the substrate profile of the low and high molecular mass NHase 1.1.1.3.1 Substrate profile of H-NHase The substrate profile of the H-NHase was performed by Nagasawa (1991) using 61 small nitrile compounds. The study showed that the H-NHase favoured short unbranched aliphatic nitriles, like acetonitrile and acrylonitrile. The catalytic activity decreased to less than 5% as the size of the chain increased and there was no activity against the branched and aromatic nitriles. 1.1.1.3.2 Substrate profile of L-NHase Studies evaluating the substrate profile for L-NHase showed that this enzyme was able to catalyse short aliphatic nitrile compounds as well as longer chained compounds such as n- butyronitrile and methacrylonitrile. According to Wieser (1998), L-NHase may catalyse branching aliphatic nitriles as well as aromatic nitriles like benzonitrile. Furthermore, the enzyme has a high catalytic activity for substituents of benzonitrile (Wieser et al., 1998). Since the L-NHase has a larger substrate profile than the H-NHase, it is more in demand to synthetic chemists. The L-NHase used in this study comes from R. rhodochrous ATCC BAA-870. 1.1.1.3.3 Substrate profile of the L-NHase from R. rhodochrous ATCC BAA-870 A recent study by Mashweu (2020) looked at the substrate profile of the L-NHase enzyme from R. rhodochrous ATCC BAA-870 a wide range of nitriles of varying sizes, degrees of branching and with different substituents. The results revealed that the L-NHases active site was accommodating and was able to catalyse aromatic compounds with large substituents adjacent to the nitrile group. However, substituents on either side of the nitrile group were sterically 6 hindered. In addition, branching aromatic compounds were not catalysed, which again indicates that steric hindrance around the cyano group is the primary barrier to substrate conversion (Mashweu et al., 2020). This could be because the active site is located deep within the alpha subunit, and the channel leading to it is made up of bulky aromatic amino acids that can limit substrate access (Pravda et al., 2014). Although electrophilicity has shown to affect the rate of the reaction, it cannot prevent the overall reaction (Mashweu et al., 2020). 1.1.1.4 Application of NHase 1.1.1.4.1 Biotransformation in industry Nitriles are simple to synthesis through several organic chemical pathways from inexpensive substrates, therefore, they have been recognised as important intermediates in the production of high-value industry compounds (Martínková et al., 2014). The present chemical processes for converting nitriles to commercial chemicals, however, have a number of limitations. Adverse reaction circumstances include strong acidic or basic solutions, high temperatures, hazardous by-product production, and significant volumes of salt, to name a few (Kaul and Banerjee, 2008). Since enzymes are naturally evolved to perform under mild reaction conditions using them as biocatalysts is a more environmentally friendly and cost-effective way of producing industrial compounds. Furthermore, because enzymes are stereo- and regio- selective, utilizing them as a biocatalyst allows for the production of novel substrates that chemical processes cannot achieve (Banerjee et al., 2003). The most successful application of NHase in industry is the manufacturing of acrylamide, which is one of the most important industrial chemicals, used in the production of coagulators, stock adhesives, paints, soil conditioners and recovering agents (Yamada and Kobayashi, 1996). In the late 1980s, Mitsubishi Rayon Company (Japan) generated 6,000 tons of acrylamide per year using a NHase from Pseudomonas chlororaphis B23 (Yamada and Kobayashi, 1996). Using a NHase from R. rhodochrous J1, the yield was increased to almost 30,000 tons per year (Kobayashi et al., 1991). In addition whole-cell NHase is used as a catalyst in the pharmaceutical and food additives industry (Bhalla et al., 2018). An interesting application of nicotinamide in industry is its use in treating arthritis as it aids in cartilage production (Bhalla et al., 2018). Therefore, nitrile-converting enzymes have become crucial in the organic chemistry industry. 1.1.1.4.2 Bioremediation Since nitriles are used as intermediate compounds and as organic solvents in industry, massive amounts of nitrile-containing chemical waste are produced. Nitriles contain a toxic 7 cyano group that has mutagenic and carcinogenic qualities that can have negative health consequences (Gong et al., 2017). Therefore, a method to dispose of industrial levels of nitrile is in huge demand. Microorganisms utilise nitriles in their metabolic processes, making them ideal candidates for nitrile waste treatment. Bacterial mixed cultures comprising of NHases and nitrilaseshave been shown to breakdown 99% of hazardous nitrile compounds in both continuous and batch culture systems (Wyatt and Knowles, 1995). 1.1.2 Enzyme engineering 1.1.2.1 NHase as a target for enzyme engineering Biocatalysts are valuable tools for the organic synthesis of industrial chemicals as they are superior to current chemical methods as they portray a number of key benefits, such as, high selectivity, high yield, mild reaction conditions, energy efficiency and the procedures are environmentally friendly (Riva and Fessner, 2014). However, although enzymes have high capabilities, they are evolved towards the needs of their natural role and host and may lack certain requirements that are needed for an enzyme to be an industrial catalyst. Even when the industrial process works well, the incentive to look for a better biocatalyst to improve the productivity remains. Therefore, enzyme engineering is being used to optimise the catalytic efficiency, specificity and substrate range of enzymes. Furthermore, there are efforts to improve the stability of the enzyme in order to resist harsh production conditions such as elevated temperatures, very acidic or basic pH values and high substrate/product concentration. This is necessary in order to maintain high work cycles and increase product yields so that the enzymes can be utilised in industry. 1.1.2.2 Methods used for enzyme engineering Enzyme engineering methods can be categorized into rational design and indirect evolution. 1.1.2.2.1 Rational design Rational design methods generally involve the comparison of the amino acid sequence or the tertiary structure of the protein with a homologous protein that has shown to have the desired function. Then using site-directed mutagenesis the promising amino acid residues are implemented into the protein of interest (Strompfová et al., 2008). Making use of rational design TLP-ste, a thermolysin-like protease from Bacillus stearothermophilus, was engineered to be eight times more thermostable and functional at 100°C than the wild-type by substituting thermostability-related amino acids (Riva and Fessner, 2014). 8 However, the use of rational design is limited since it necessitates a considerable deal of knowledge about the protein sequence, structure, catalytic mechanism, structure-function mechanism, and structure-ligand link (Brenner and Lerner, 1992). Our current understanding of how to adjust protein structure in order to improve protein function is insufficient, therefore relying solely on this method could hinder the development of potential mutants. 1.1.2.2.2 Indirect evolution Indirect evolution mimics the natural evolution process that occurs through random mutagenesis and sexual recombination. This technique, unlike rational design, requires limited previous understanding of the protein sequence and structure. The gene encoding the desired protein is randomly altered during indirect evolution, resulting in a huge library of mutants. After that, the mutant library must be screened or selected in order to find the mutant with the desired trait. The most commonly used methods to introduce random mutations include error- prone PCR (EP-PCR), chemical mutagens and UV radiation. The improved activity level of each mutation in the library can be conceptualised as climbing a fitness landscape (Packer and Liu, 2015). The aim of protein engineering is to climb this landscape towards peak activity levels through taking mutational steps (Packer and Liu, 2015). Over several generations of protein engineering there should be an accumulation of beneficial mutations that ultimately result in an improved phenotype. However, since indirect evolution generates large mutant libraries, the success of this technique relies on the availability of a high-throughput selection or screening method that is able to detect mutants with the desired feature. It is critical that the technique used to select or screen enzymes is sensitive and specific to the desired enzyme property. As a result, most screening and selection procedures try to identify the phenotype associated with the mutant gene. Consequently, the lack of high-throughput screening and selection methods available limits the search for desirable mutations. 1.1.3 Screening and selection techniques The screening and selection step is the bottleneck in the overall process of directed evolution. There is continues research being done to improve screening and selection techniques for probing the vast protein sequence space. Screening is the process of inspecting the phenotype of each individual library member in order to determine the variant with the desired biocatalytic activity (Packer and Liu, 2015). In 9 contrast selection bypasses the need to inspect individual phenotypes by linking the biocatalytic activity of interest to the ability of the organism to survive (Packer and Liu, 2015). 1.1.3.1 Selection techniques If the enzyme is able to fulfil a metabolic function a selection method that separates active library members that contain the gene of interest can be developed (Packer and Liu, 2015). Although NHase performs a metabolic activity that aids in the organism survival, it requires amidase to convert amide products into carboxylic acid, which is the nitrogen compound used by the cell. As a result, both enzymes would need to be linked which could result in random mutations to the amidase, thus leading to optimisation of the amidase over the NHase. In addition, since the selection technique is based on survival, it would not be able to differentiate which enzyme has evolved. 1.1.3.2 Screening techniques In screening techniques, the gene variants must be individually expressed on either a solid media or in multi-well liquid plate. The individual expression allows for spatial separation of the variants which in turn allows for the phenotype of the variants to be linked to the genotype. 1.1.3.2.1 Current screening techniques Current screening techniques available to screen for NHase include, nuclear magnetic resonance (NMR), mass spectroscopy, gas chromatography and high-performance liquid chromatography (HPLC). These techniques are not high-throughput and would be tedious and inefficient at screening large mutant libraries of at least 10,000 mutants per round of evolution. 1.1.3.2.2 Current high-throughput screening techniques Techniques such as fluorescence-activated cell sorting (FACS) has been employed to screen bulk populations of mutant cells. FACS identifies and isolates cells that contain a desired variant based on the presence of a fluorescent reporter molecule. Since this process does not require spatially separating out the variant it is a much faster, high-throughput screening process that can screen up to 108 library members in less than 24 hours (Santoro and Schultz, 2002). Resonance energy transfer (RET) is a technique that relies on the transfer of energy between two chromo- or fluorophores. The transfer of energy between the two fluorophores allows for the study of protein interactions and conformations. Coupled with FACS this technique has been used as a high-throughput screening method for enzyme engineering (Olsen 10 et al., 2000). Another technique called cell surface display uses anchoring motifs to display the protein of interest on the surface of the cell (Lee et al., 2003). This technique can be useful for enzyme engineering of the substrate selectivity of an enzyme since the protein can react with the substrate on the surface of the cell. Coupling this technique with FACS it can be used for high-throughput screening of mutant enzymes. The current high-throughput screening techniques discussed above require expensive equipment and reagents. Furthermore, the research required to get the technique to work for the NHase enzyme for all the different parameters that can be evolved (pH, temperature, substrate selectivity profile) would be tedious. 1.1.3.3 Microwell plates for high-throughput screening Using a microwell plate, traditional enzyme activity experiments can be carried out by miniaturising the reaction components. The components can be added manually (so that it can be carried out in any laboratory), or the throughput can be exponentially enhanced using robotic systems that are available. Among the enzyme assays available, colorimetric or fluorometric assays are the most convenient (Packer and Liu, 2015). The colour can be measured spectroscopically for a quantitative assay, or the colour can be assed visually for a qualitative assay. Furthermore, because colorimetric assays may be performed continuously, information concerning enzyme kinetics can be easily derived. 1.1.3.3.1 Colorimetric assay for detection of nitrile hydratase by Banerjee Banerjee (2003) developed a colorimetric assay which had a pH indicator dye (bromothymol blue) within it that changed colour based on the pH. Upon the release of hydrogen ions during the nitrilase catalytic reaction the colour changed from blue to green indicating nitrilase activity. Sahu (2019) then adapted this assay so that it could be used to detect NHase activity via the accumulation of amides. However, when this experiment was attempted (results not shown), it was discovered that the amides are not significantly acidic, and their accumulation did not result in a colour change. Furthermore, other components within the cell, such as ammonia, could be causing the colour shift, making this approach insufficiently specific. 1.2 Rationale NHases are needed for the production of high-value industry compounds such as amides. Using NHases as a biocatalyst to manufacture industry compounds instead of chemical methods that need harsh conditions offers several advantages, including being more 11 environmentally friendly and cost effective. Enzymes are also stereo- and regio-selective, therefore using them as a biocatalyst allows for the creation of novel substrates that chemical processes cannot (Benerjee et al., 2002). Furthermore, NHases are being used in bioremediation to remove toxic nitrile waste that is produced in industry (Gong et al., 2017). Since microorganisms employ nitriles in their natural metabolic processes, using their nitrile hydrolysing enzymes for nitrile waste treatment is ideal. However, despite their tremendous capabilities, enzymes have evolved to meet the needs of their natural role and may lack some characteristics required for an enzyme to function as an industrial catalyst. Even when the industrial process works well, the desire to find a superior biocatalyst to boost productivity continues. Therefore, enzyme engineering is being used to optimise the catalytic efficiency, specificity and substrate range of enzymes. Furthermore, attempts are being made to increase the enzyme's stability so that it can withstand tough manufacturing circumstances such as high temperatures, very basic or acidic pH values and substrate/product concentration. Enzyme stability is critical for maintaining intense work cycles and increasing product yield. There are two main enzyme engineering methods. The first is the rational design method, which examines the protein's tertiary structure or homologous amino acid sequence to discover prospective amino acid residues that might be swapped to increase enzyme activity. However, a great amount of knowledge about the protein sequence and structure is required (Brenner and Lerner, 1992) and our current understanding of how to adjust protein structure in order to improve protein function is insufficient. As a result, relying only on this strategy could stifle the emergence of possible mutants. The second enzyme engineering method is indirect evolution. In this method mutations are added randomly to the DNA sequence and results in large libraries with mutant enzymes. Therefore, the success of indirect evolution relies on the availability of a high-throughput selection or screening method that is able to detect mutants with the desired feature. Current screening and selection methods are inefficient, complicated and require expensive equipment that is not available in all laboratories. Therefore, in order to optimise the search for desirable mutations that could be utilised in industry, a screening method needed to be identified and optimised. 12 This project will result in a screening method for NHase enzymes, specifically the NHase from R. rhodochrous ATCC BAA-870, as well as improve the body of knowledge in the fields of biocatalysis and green chemistry. The R. rhodochrous species has been the key source of new NHases and thus developing an assay based on the R. rhodochrous ATCC BAA-870 strain will allow strains with homologous sequences to be easily adapted to this assay. 1.3 Aim and objectives 1.3.1 Aim Identify and optimise a high-throughput screening assay for detecting nitrile hydratase activity. 1.3.2 Objectives 1) Co-transform T7 Express competent E. coli cells with the pRSFDuet-1 plasmid carrying the alpha and chaperon subunits, as well as the pET21a(+) plasmid containing the beta subunit. 2) Successfully co-express the two subunits of the nitrile hydratase enzyme. 3) Identify and optimise a high-throughput screening technique to measure nitrile hydratase activity. 13 Chapter 2: Protein expression of Nitrile Hydratase from R. rhodochrous ATCC BAA- 870 using a two-plasmid vector system 2.1 Introduction Prior to developing a screening assay, expression of the Nitrile Hydratase (NHase) was carried out. One of the most crucial criteria for effective application of enzymes is that the expression and isolation of the required enzyme be repeatable. 2.1.1 Recombinant protein expression and its effect on biocatalysis The development of recombinant protein expression has revolutionised biochemistry and the use of biocatalysts for industrial and therapeutic applications. The advent of recombinant protein expression has allowed proteins to be genetically engineered, synthesised and then purified quickly and in large amounts. Expression of specific proteins outside of the native cell allows for the protein’s biochemistry to be studied independent of other host cell metabolic processes which could interfere with the enzyme characterisation. Although in theory recombinant protein expression seems straightforward: inserting a gene of interest into a vector, transforming a host cell, induction followed by protein purification; in practice, there are several things that can go wrong, such as, poor growth of host, losing the plasmid, no expression of the protein, inclusion body (IB) formation and protein inactivity (Rosano and Ceccarelli, 2014). There are many factors that need to be taken into consideration to ensure the ultimate aim of large volumes of soluble, active recombinant protein can be achieved. 2.1.2 Two-plasmid vector system used for recombination of the R. rhodochrous ATCC BAA-870 Nitrile Hydratase The plasmids were designed by Joni Frederick at the University of Cape Town (2006). First attempts were made by him to clone the subunits and the chaperone protein into one plasmid, specifically the pET28a(+) or the pET20b(+) plasmids. However, in both plasmids only the beta and chaperone subunits were cloned, with the chaperone subunit being overexpressed compared to the beta subunit. This indicated that the ribosome binding site (RBS) of the chaperone was being recognised by the E. coli while the RBS for the subunits needed further investigation. Upon investigation the alpha subunit’s RBS was identified as a rare site which made it unrecognisable by the ribosomes of the E. coli. In order to increase the recognition of the transcript of the alpha subunit the RBS was mutated to a sequence identical to that of the 14 chaperone. However, this did not seem to increase the expression of the alpha subunit. Interestingly, upon investigation of enzyme activity, there was some conversion of benzonitrile to benzamide, indicating some activity was present despite the very low expression (Frederick et al., 2006). Based on these findings Frederick investigated the use of a two-plasmid system. The use of a two-plasmid system ensures that the subunits are expressed independent of each other as transcription of each subunit is under the control of each plasmid’s expression promoter. Since the two-plasmid method produced more equal expression of the subunits, it was decided to adopt the two-plasmid vector system for protein expression in this study. The beta subunit of the L-NHase from R. rhodochrous ATCC BAA-870 was cloned into a pET21a(+) plasmid while the alpha subunit as well as the chaperone protein was cloned into a pRSFDuet-1 plasmid. Protein expression can be tricky when working with only one plasmid so adding a second plasmid can add a level of complexity. Many factors need to be taken into consideration when selecting plasmids for a two-plasmid expression system such as plasmid compatibility, copy number, selection markers and expressivity. 2.1.2.1 Plasmid compatibility of the pRSFDuet-1 and pET21a(+) plasmids Plasmid compatibility as well as copy number lies within the replicon of the plasmid. The replicon is a genetic element that consists of the origin of replication as well as other cis-acting elements. For a multi-plasmid expression system to work the plasmids must not fall within the same incompatibility group which means that they must have different origins so that the plasmids are not competing for the same replication machinery within the cells. In this study the pET21a(+) plasmid was used, which possess the pMB1 replicon, a derivative of ColE1, and has low copy number with 20-60 copies per cell. The pRSFDuet-1 plasmid contains the RSF 1030 (NTP1) replicon and has a high-copy number. The difference in copy number can affect equal expression of the plasmids and will be discussed in more detail in Section 2.3. 2.1.2.2 Selection markers on the pRSFDuet-1 and pET21a(+) plasmids To prevent the growth of plasmid-free cells, a plasmid must contain a resistance marker such as antibiotic resistance genes (Rosano and Ceccarelli, 2014). Since two plasmids are used in the expression host in order to ensure that both plasmids are maintained, each plasmid must have a different resistance marker (Brzoska and Firth, 2013; Sninsky et al., 1981). The pRSFDuet-1 plasmid contains the genes for kanamycin resistance, whereas pET21a(+) contains the genes for ampicillin resistance. 15 2.1.2.3 pRSFDuet-1 and pET21a(+) plasmids both use the T7 promoter system The T7 promoter system is popularly used for protein expression as it can result in over 50% of the E. coli’s total cell protein accounting for the target protein (Baneyx, 1999; Graumann and Premstaller, 2006). The method operates by inserting the gene of interest near the promoter, which is recognized by T7 RNA polymerase (T7 RNAP). The gene encoding the T7 RNAP is under the control of the LacUV5 promoter. In the absence of an inducer the lac repressor protein binds to the operator next to the LacUV5 promoter and inhibits expression of the T7 RNAP. When lactose or a lactose analogue such as isopropyl β-D-1-thiogalactopyranoside (IPTG) is present it binds the lac repressor thus allowing for transcription of the T7 RNAP which is then available to bind at the T7 promoter and transcribe the gene of interest. The T7 expression system is used in both the pRSFDuet-1 and pET21a(+) plasmids, making subunit expression as straightforward as possible. Table 2.1 below summarises the main characteristics of the two expression plasmids. Table 2.1: Summary of characteristics for the pRSFDuet-1 and pET21a(+) plasmids pRSFDuet-1 pET21a(+) Subunit expression Alpha subunit and chaperone protein Beta subunit Size 3829 bp 5443 bp Insert size 618 bp and 444 bp 675 bp Copy number >100 20- 60 Promoter T7lac T7lac Replicon RSF 1030 (NTP1) pBR322, derived from ColEI Selection marker Kanamycin Ampicillin Expressed protein size 22.7 kDa and 16.7 kDa 25.1 kDa 2.1.3 The host system used for Nitrile Hydratase expression There is a wide variety of different hosts and vectors that can be used in recombinant protein expression. Deciding on which such options to use requires consideration and understanding of their elements as well as understanding what is required for the protein of interest. The R. rhodochrous ATCC BAA-870 NHase has previously been expressed in the E. coli BL21(DE3) cell line (Frederick et al., 2006). The BL21(DE3) strain of E. coli is the most used expression 16 strain. BL21 cells are deficient in the Lon protease which is responsible for degrading foreign proteins (Gottesman, 1996). They are also deficient in the gene coding for the outer membrane protease OmpT, whose function is to degrade extracellular proteins (Grodberg and Dunn, 1988). Furthermore, it has the hsdSB mutation which prevents plasmid loss by preventing DNA methylation which signals for DNA cleavage by endogenous restriction endonucleases (Rosano and Ceccarelli, 2014). The DE3 indicates that this strain of E. coli BL21 has a λDE3 prophage inserted within its chromosome which carries the T7 RNA polymerase under the control of the lacUV5 promoter. However, previous expression of the R. rhodochrous ATCC BAA-870 NHase in E. coli BL21(DE3) cells showed that there was uneven expression of the beta subunit compared to the low expression of the alpha subunit. Expression of the NHase in E. coli BL21(DE3) cells was also initially attempted in this study (results not shown), however despite changing expression conditions there was still uneven expression of the subunits. Therefore, transformation of the plasmid into a T7 Express competent E. coli strain was done in order to see if the strain of E. coli that is used for expression could improve the outcome. The T7 Express competent E. coli strain is an enhanced BL21(DE3) derivative, with the main difference being that the gene carrying the T7 RNA polymerase is inserted into the lac operon on the E. coli chromosome and is expressed under the control of the lac promoter (New England Biolabs). This provides more control for the induction of the T7 RNA polymerase and consequently more control of transcription of genes downstream of the T7 promotor located on the plasmid. The tighter the control of these genes the less ‘leakiness’ or basal expression occurs in the absence of an inducer. In addition, the T7 Express competent E. coli strain has a higher transformation efficiency than the BL21(DE3) strain even with the use of toxic clones (Anton et al., 2016). A summary of difference between the two strains is seen in Table 2.2 below. 17 Table 2.2: Summary of the characteristics of T7 Express competent E. coli and E. coli BL21(DE3) cells Strain T7 Express competent E. coli BL21(DE3) B strain Yes Yes Transformation efficiency 0.6-1 x 109 cfu/μg pUC19 1–5 x 107 cfu/µg pUC19 DNA T7 RNA polymerase location In the lac operon- no λ prophage λ prophage carrying the T7 RNA polymerase gene Protease deficiencies Lon and OmpT Lon and OmpT Resistant to phage T1 (fhuA2) Yes Yes Animal product Yes Yes Restrict methylated DNA (McrA-, McrBC-, EcoBr-m- , Mrr-) No Yes Genotype fhuA2 lacZ::T7 gene1 [lon] ompT gal sulA11 R(mcr- 73::miniTn10--TetS)2 [dcm] R(zgb-210::Tn10--TetS) endA1 Δ(mcrC- mrr)114::IS10 fhuA2 [lon] ompT gal (λ DE3) [dcm]∆hsdS λ DE3 = λ sBamHIo ∆EcoRI- B int::(lacI::PlacUV5::T7 gene 1) i21 ∆nin5 2.2 Methods and materials: 2.2.1 Materials 2.2.1.1 Chemicals and reagents The chemicals were acquired from Sigma-Aldrich (SA) and Merck Chemicals. All chemicals were of analytical grade. 2.2.1.1 E. coli strains Two E. coli strains were used in this study and are listed below in Table 2.3. The T7 Express competent E. coli strain was used for protein expression while the DH5α strain was used for storage of the plasmids. 18 Table 2.3: The E. coli strains used as hosts in this study Strain Feature Supplier T7 Express competent E. coli Expression strain Novagen DH5α Cloning strain Novagen 2.2.1.2 Recombinant expression systems Based on the protocol developed by Frederick (2006) a two-construct expression system was used and the plasmids were ordered from GenScript (USA). The same plasmids were used in this study. The gene sequence of the alpha subunit as well as the chaperone protein were inserted as a single construct into the pRSFDuet-1 plasmid in the second multiple cloning site (MCS) between the Nde1 and Pac1 restrictions sites. The gene sequence of the beta subunit was inserted into the pET-21a(+) plasmid within the MCS between the Nde1 and Xho1 restriction sites. At the carboxy-terminal (C-terminal) the native stop codon of the beta subunit was removed in order to incorporate the His-tag from the vector for easy purification. With the intention of optimising expression, the native ribosome binding sites (RBS) of the alpha and beta subunits were removed to allow the RBS of the plasmids to be the only site where ribosome binding could occur. Since the genetic sequence of the chaperone protein is within the alpha subunit its RBS could not be altered. 19 Figure 2.1 Plasmid maps of pET-21a(+) with beta subunit insert and pRSFDuet-1 with alpha and chaperon inserts. The maps was originally constructed by Schmid (2020) using Benching. 3.2.1.3 Media The Luria Bertani (LB) medium was produced using 5% NaCl, 5% yeast extract and 10% tryptone. The LB agar was produced using the same ingredients at the LB medium except 1% agar was added. All media and agar were sterilised by autoclaving at 120℃, 1 atm for 20 min. The LB medium was stored at room temperature while the LB agar plates were stored at -4 ˚C. To avoid contamination and antibiotic degradation, the medium and plates were only stored for two weeks. 3.2.1.4 Antibiotics Kanamycin and ampicillin were used to maintain the plasmid within the transformed cells. The cells transformed with just pRSFDuet-1 had a final concentration of 25 mg/ml of kanamycin in the LB broth or agar while the cells transformed with the pET21a(+) plasmid had a final concentration of 50 mg/ml of ampicillin (Table 2.4). When the plasmids were co- transformed both ampicillin and kanamycin was used. 20 Table 2.4: shows the concentration of antibiotics employed in the LB broth medium and LB agar plates for each transformant. Transformant Antibiotics Final concentration αβc Kanamycin and Ampicillin 25 mg/ml and 50 mg/ml α Kanamycin 25 mg/ml β Ampicillin 50 mg/ml 2.2.2 Methods 2.2.2.1 Transformation of pRSFDuet-1 and pET21a(+) plasmids 2.2.2.1.1 Competent cell preparation Super-competent T7 Express competent E. coli cells were used for the transformation. Competent cells were prepared by taking a single colony from a fresh streak plate to inoculate a 5 ml starter culture of LB broth. The cells were allowed to grow overnight at 37˚C while shaking. A 1 in 100 dilution of the starter culture was then used to inoculate 50 ml of fresh LB broth. The culture was allowed to grow at 37˚C while shaking until an optical density measured at 600 nm (OD600) of ~ 0.5 using the S-20 Spectrophotometer (Boeco, Germany) was reached. From here on all glassware, buffers and disposables were chilled before use. The cells were placed on ice for 20 minutes before being centrifuged at 4˚C at 5000 rpm for 10 minutes. The pellet was then resuspended in 10 ml of ice cold 0.1 M MgCl2 and was incubated on ice for 30 minutes. The cells were then collected via centrifugation again at 4˚C at 4000 rpm for 10 minutes. The pellet was then resuspended in 1 ml ice cold 0.1 M CaCl2 supplemented with 15% glycerol and was incubated on ice for 10 minutes. Subsequently, 50 µl of cells were aliquoted into cryogenic vials (Sigma-Aldrich) and stored at -80˚C. 2.2.2.1.2 Transformation of T7 Express competent E. coli by heat shock The super-competent cells were thawed on ice and transferred from cryogenic vials to Eppendorf tubes (Eppendorf, Hamburg, Germany). Transformation was performed by adding 100 ng of each plasmid to the tubes and allowing the cells and plasmids to incubate together on ice for 40 minutes. This incubation period allows for the plasmids to come into close contact with the porous cell walls. The tubes were then placed in a Dri-Block DB-2D (Techne, United Kingdom) at 45˚C for 1 minute. Immediately thereafter the cells were placed back on ice for 2 minutes in order to close the pores in the membrane and retain the plasmids. After which 1 ml 21 of LB broth with no antibiotics was placed in the tubes and the tubes were incubated at 37˚C for 1 hour with shaking. The time spent without antibiotics allowed the transformed cells to start producing antibiotic resistant genes. Lastly, the cells were plated at various dilutions and grown over night at 37˚C. The number of colony forming units (CFU) were counted on the following day and the transformation efficiencies were calculated 2.2.2.1.3 Confirmation of transformations using sequencing Selected colonies that were grown on LB agar plates with antibiotics (which is already an indication that transformation was successful) were grown over night in LB broth medium with antibiotics at 37˚C while shaking at 200 rpm. The ZyppyTM plasmid miniprep kit (Zymo Research, USA) was used to extract the plasmids from the cells. The extracted plasmids were then subjected to sequencing. Inqaba Biotech (Pretoria, South Africa) was employed to sequence selected colonies that contained both plasmids to ensure the presence of both the plasmid and the insert within the transformed cells. 2.2.2.2 Glycerol stocks of the transformed cells A colony for each type of transformed cell (cell containing either the pRSFDuet-1 or the pET21a(+) plasmid or a cell containing both plasmids) was grown in LB broth medium with the relevant antibiotic at 37˚C while shaking at 200 rpm until an OD(600) = 0.5 was reached. An aliquot of 500 µl of broth was transferred into a Cryotubes (Lasec SA, Johannesburg, Gauteng, South Africa) followed by the addition of 500 µl of 50% glycerol solution. The tubes were quickly agitated by hands to ensure mixing of the broth with the glycerol. The aliquots were stored at -80˚C until needed. 2.2.2.3 Expression of the subunits and chaperone proteins 2.2.2.3.1 Plate culture of the transformed cells Glycerol stocks of the transformed cells were streaked onto LB agar plates containing the relevant antibiotics (50 mg/ml ampicillin and 25 mg/ml kanamycin) using sterile techniques. The plates were incubated at 37˚C overnight. The cultures were then maintained at 4˚C but only colonies from freshly streaked cultures (streaked bi-weekly) were used to prevent contamination, mutation or loss of plasmid. 22 2.2.2.3.2 Growth of starter culture A colony of transformed cells from a freshly streaked plate was inoculated into 5 ml LB broth medium with the relevant antibiotic. The starter culture was either grown over night or in the morning on the same day that induction was to take place in order to limit the loss of either plasmid due to depletion of antibiotics or toxicity from metabolic biproducts. The broth was incubated at 37˚C while shaking at 200 rpm until the broth was opaque. 2.2.2.3.3 Expansion of starter culture From the starter culture 2 ml was added to a further 200 ml LB broth medium with antibiotics so that a 1:100 ratio was achieved. A 1 L baffled Erlenmeyer flask was used so that there was good aeration of the shake culture. The broth was incubated at 37˚C while shaking at 200 rpm until an OD(600) = 0.3-0.5 was achieved at which point the culture was induced. 2.2.2.3.4 Induction of the culture Once an OD(600) = 0.3-0.5 was achieved the culture was then induced by addition of 800 µl of IPTG to give a final concentration of 0.4 mM. Thirty minutes prior to induction 200 µl of cobalt chloride was added so that there was a final concentration of 0.1 mM. The culture was placed in a 16˚C incubator for 20 hours with shaking at 200 rpm. The cells were harvested by centrifugation for 10 minutes at 8000 x g at 4˚C. The supernatant containing the broth was discarded and the pellet was washed using 0.5 mM Tris- HCl buffer (pH 7.2) after which the cells were stored at -20˚C overnight. Freezing of the pellet overnight helped with sheering of the cell walls during sonication. 2.2.2.4 Cell disruption via sonication The frozen pellet was weighed and was resuspended in lysis buffer (0.5 mM Tris- HCl buffer, pH 7.2) using a ratio of 10 ml lysis buffer to 1 g of pellet to get the appropriate amount of lysis buffer for sonication. Sonication was done using Soniprep Ultrasonic Instrument (Matthews Studio Equipment, California, USA) at a power outage of 17 Watts. The resuspended pellet was placed in ice and was exposed to discontinuous blasts of ultrasound (10 cycles of 30 seconds on and 30 seconds off) so that there was a total of 5 minutes of ultrasound exposure. Sonication was done on ice and was cycled so that the heat produced by the sonication process would not denature the proteins that were released. 23 2.2.2.5 Harvesting of the crude protein extract A sample of the crude lysate, which contained both the soluble and insoluble proteins, was taken for future analysis of the subunit solubility. The crude lysate was then centrifuged for 20 minutes at 20 000 x g at 4˚C. The supernatant which contained the cell-free crude protein extract was appropriately stored for downstream NHase assays and/or purification. The pellet was stored at -20˚C for future analysis of the insoluble protein fractions. 2.2.2.6 Storage of the crude protein extract The supernatant that contained only the soluble fractions of protein, also referred to as the crude protein extract, was stored at 4˚C on ice. Freezing of the protein was discouraged as the freeze- thawing process could affect the activity of the enzyme. The crude protein was stored at a high concentration as higher concentrations of protein prevent aggregation. Prior to each use the crude protein lysate was centrifuged to remove any protein aggregates (seen as white particles) and the concentration of protein was redetermined. 2.2.2.7 Determination of protein concentration The Qubit 2.0 Fluorometer (Invitrogen by Life Technologies, USA) was used to quantify the protein concentration. The protocol involved using 199 µl of the Qubit protein buffer and the addition of 1 µl of Qubit protein reagent into Qubit tubes. The tubes were vortexed to homogenise the solution and 2 µl of the solution was removed. Thereafter, 2 µl of a diluted protein sample in distilled water was added to the solution, so that the total volume in the tubes was 200 µl. After homogenising the tubes by vortex they were incubated at room temperature for 15 minutes. The tubes were then placed inside the fluorometer, and the concentration was given based on a standard curve. 2.2.2.8 Visualisation of protein on sodium dodecyl sulphate-polyacrylamide electrophoresis (SDS-PAGE) gel Electrophoresis was carried out using the Laemmli method (Laemmli, 1970). A 12% polyacrylamide gel was prepared using 40% acrylamide/bisacrylamide, 1.5 M Tris-HCl (pH 8.8), 10% SDS, 10% ammonium persulfate (APS) and tetramethylethylenediamine (TEMED). The protein samples were mixed with a 2x sample buffer (0.5 mM Tris-HCl (pH 6.8), 10% SDS, 1% bromophenol blue, 50% glycerol and 5% 2-mercaptoethanol), and boiled for 5 minutes at 95˚C using a Dri-Block DB-2D (Techne, United Kingdom) so that the protein was linearised. The protein samples were then loaded into the wells and 1x running buffer (3% Tris 24 Base, 14% glycine and 1% SDS) was poured into the chambers. An Unstained Protein Ladder (Thermo Fisher Scientific, USA) with a molecular weight range between 10 and 200 kDa was used. The gels were stacked at 80 V and once the protein samples were aligned the gels were run at 120 V. The polyacrylamide gels were stained with 0.1% solution of Coomassie Blue R- 250 (40% methanol, 10% glacial acetic acid) overnight with shaking. The gels were then de- stained (20% methanol, 10% glacial acetic acid) until the gels were clear. 2.2.2.9 Growth curves The starter culture and expansion culture were done using the same protocol described in Sections 2.2.2.3.2 and 2.2.2.3.3. However, once the expansion culture was initiated, 500 µl of the culture was extracted every 30 minutes and the absorbance was measured using S-20 Spectrophotometer (Boeco, Germany). The absorbance values were recorded and were used to construct growth curves. 2.2.2.10 Statistical Analysis Duplicate experiments were done, and the data was analysed using the Student t-test in excel. The differences were considered significant for * p < 0.05 or ** p <0.01 2.3 Results and Discussion 2.3.1 Conformation of transformation Transformation conformation was achieved by submitting the plasmids to Inqaba Biotech for sequencing. BioEdit Sequence Alignment Editor was used to align the sequences returned from Inqaba with the gene sequences of each subunit and chaperon protein. There was perfect alignment indicating that the plasmids with the correctly orientated gene inserts where successfully transformed into the host cells. 2.3.2 Comparison of the growth curves and growth rates for each plasmid as well as the co-transformed cells 2.3.2.1 Growth curves Growth curves were constructed in order to better understand the transformed cells growth patterns as well as how the growth changes once expression is induced. Understanding the transformed cells' growth patterns and how expression affects them is a crucial initial step in determining the ideal conditions for protein expression. 25 The growth curves were created by measuring the optical density of a culture at 30-minute intervals while it was incubated at 37°C. The induction of the co-transformed cells was initiated at the beginning of the exponential phase which was achieved at 120 minutes after incubation. This is in line with a previous study that was done using the same plasmids, excepts they were transformed into E. coli BL21(DE3) cells (Soobben, 2020). The expression was induced by the addition of 0.4 mM IPTG and 0.1 mM CoCl3. The overexpression of protein within recombinant host cells causes a metabolic load due to energy and resources being directed towards protein production. This is illustrated by the blue curve in Figure 2.2 which has slower growth over time compared to the co-transformed cells that were not induced (orange curve). Although the induced recombinant cells grew slower at 37°C, the rate of growth was still steady at this temperature, which could have resulted in the cells entering the death phase before there was sufficient time for protein expression. In addition, had the cells entered the death phase, the number of cells available to express the protein would have dropped dramatically, and the dying cells would have lysed, releasing the protein into the external environment. As a result, lower temperatures, such as 16°C, were chosen as a good expression temperature over 37°C since a lower temperature slows down the cell growth rate. The two plasmids were individually cloned into T7 Express competent E. coli in order to determine the plasmid burden that each one has on the growth of the cell. Although the pRSFDuet-1 plasmid, which contains the alpha and chaperone subunits, appears to grow faster at first (purple curve) than the pET21a(+) plasmid (yellow curve), they eventually achieve the same biomass. 26 Figure 2.2: Growth curves representing the change in biomass of differentaily transformed T7 Express competent E. coli cells in LB media measured at OD600. A starter culture of the transformed cells where grown to an OD of 0.1 prior to being added to 200 ml of fresh broth in order to ensure equal starting biomass. 500 µl of culture was removed in 30- minute intervals in order to measrure the OD at 600 nm. Induction of the co-transformed cells took place at 120 minutes using 0.4 mM IPTG. The green curve represents untransformed cell; the orange curve represents co-transformed cells that were not induced; the purple curve represents cells transformed with both the alpha and chaperone subunits; the yellow curve represents cells transformed with the beta subunit; the blue curve represents co-transformed cells that were induced with addition of 0.4 mM IPTG at 120 minutes. The standard deviation between duplicate measurements is indicated by the error bars. 2.3.2.2 Growth rates Growth rates were used as a metric to better understand the metabolic and replication burden that each plasmid has on T7 Express competent E. coli cells. The growth rates were determined using just the exponential phase from the growth curves in Figure 2.2. In Figure 2.3 the green bar graph depicts untransformed T7 Express competent E. coli cells, which has the maximum growth rate, as expected. When comparing the growth rates of uninduced co-transformed cells (orange bar) to induced co-transformed cells (blue graph), a significant difference is observed. This again indicates that there is a metabolic burden associated with protein expression. 27 The pRSFDuet-1 plasmid which contains the alpha and chaperone subunits is the smaller of the two plasmids yet has a higher copy number. The pET21a(+) plasmid which contains the beta subunit is a larger plasmid but has a low copy number. Due to replication and maintenance of the additional plasmid DNA, both the copy number and the size of the plasmid have an impact on the level of plasmid burden. Although the pRSFDuet-1 plasmid appears to attain higher biomasses earlier than the pET21a(+) plasmid in Figure 2.2, their growth rates are similar as seen in Figure 2.3. Although the copy number and size of the plasmids are different neither one significantly affects the growth rate of the cell. Figure 2.3 Growth rates of the transformants in T7 Express competent E. coli. The growth rates of the transformants were calculated using only the exponential phase from the growth curves. The growth rates are expressed as relative activity (%) taking the growth rate of the untransformed T7 Express competent E. coli cells at 100%. ** indicates significant difference of p < 0.01. The standard deviation between duplicate measurements is indicated by the error bars. The two plasmids used in this study have similar growth rates and have minimal burden on the host. 0 20 40 60 80 100 120 αβc not induced αβc Induced αc plasmid β plasmid Untransformed T7 cells R el at iv e gr o w th r at es ( % ) 28 2.3.2 Co-expression of the R. rhodochrous ATCC BAA-870 Nitrile Hydratase The appropriate expression parameters were established based on the growth patterns of the recombinant cells as well as previous studies and litriture. There are a number of expression paramaters that influnce the quantity and quality of the expressed recombonint protein. This includes IPTG concentration, temperature, time and optical density of the culture at time of induction. Aside from expressing the protein successfully, the expression parameters must be optimised for high levels of protein activity. If the expression paramaters are not correct it can lead to protein instability and aggregation which will compromise the proteins activity. When recombinant proteins are expressed in E. coli host cells the spatio-temporal control of its expression is lost. The host's microenvironment may differ from the protein's initial source in terms of pH, osmolarity, cofactors, and folding processes. Furthermore, high levels of expression result in a large concentration of hydrophobic regions within polypeptides, which may form contacts and cause incorrect folding and aggregation. Build up of these insoluble protein aggregates are called inclusion bodies (IBs). Altering the various parameters that influence the degree of expression, such as temperature, IPTG concentration, and time, can slow down the expression rate, reducing molecular crowding of the cell, which leads to unfavourable protein folding conditions. Since the induced recombinant cells showed a steady increasing growth pattern at 37°C, a lower temperature was chosen, which would reduce the growth rate and provide sufficient time for protein expression before the cells entered the death phase. In addition, previous expression trials showed that the best expression temperature for the nitrile hydratase protien was at 16˚C for 20 hours (Schmid, 2020). This is in line with our knowledge of protein expression and literature. Slower protein synthesis gives the freshly synthesized peptide strand more time to fold appropriately. Furthermore, low temperature reduces aggregation by removing temperature-dependent hydrophobic interactive forces (Vera et al., 2007). The expression of the NHase enzyme was conducted at 16˚C for 20 hours using 0.4 mM IPTG and 0.1 mM cobalt chloride. The cultures were then centrifuged to remove the broth and 50 mM Tris-HCl lysis buffer was added. The cells were lyssed via sonication. The protein subunits were run on an SDS-PAGE gel, which separates proteins by size, to see if the proteins of interest were successfully expressed. In order to determine the level of aggregation, crude lysate (CL), which contains both the soluble and insoluble fractions, and 29 crude protein (CP), which only contains the soluble fraction after seperation by centrifugation, were run on the gel in parallel. Lanes 3 and 4, which reflect the CL and CP of the alpha subunit expression, respectively, reveal a good deal of alpha protein expression (Figure 2.4). Since the bands in lanes 3 and 4 are equally dense, the alpha protein is soluble, as evidenced by the bulk of the protein found in the supernatant. This implies that the alpha protein is primarily stable and folding properly. On the other hand, the band in lane 5 that represents the CL of the beta subunit is slightly larger than the band in lane 6 that represents the CP of the beta subunit. This implies that at the current expression parameters, the beta subunit is slightly insoluble indicating instability and aggregation. Frederick (2006) and Schmid (2020) did co-expresssian studies of the same plasmids and found similar results. It is possible that the beta subunit is aggregating because it is not expressed with the chaperone protein, which has been shown to help in protein folding (Frederick et al., 2006; Hartl and Hayer-Hartl, 2002). However, because of the modest aggregation, there were still sufficent soluble beta subunits in the supernatant available to form α2β2 hetrodimers. Looking at lanes 1 and 2, which represent the CL and CP of co-expressed cells, it can be seen that the beta subunit is expressed at a higher concentration than the alpha subunit. When the level of expression of the alpha subunit during co-expression is compared to the level of expression when it is expressed independently, it is clear that the alpha subunit is expressed at lower levels during co-expression. Previous study by Frederick (2006) and Schmid (2020) validated these findings. It is particularly intriguing because the alpha subunit was cloned into the pRSFDuet-1 plasmid, which has a high copy number and hence should contain a large amount of alpha protein. However, although it is reasonable to assume that a high copy number will result in a high protein yield, a high plasmid number can cause metabolic stress, which can lead to plasmid instability. It is possible that when two genes are co-expressed, twice as many polypeptides are produced, resulting in protein instability. However, the width of the alpha subunit bands in lanes 1 and 2 are the same, indicating that its solubility is unaffected, suggesting that the reduced expression of the alpha subunit during co-expression is unrelated to increased polypeptide instability. In addition, protein toxicity of the alpha subunit towards the T7 Express competent E. coli host must be considered. Protein toxicity occurs when recombinant protein performs an unneeded and harmful function in the host cell. However, because both the alpha and beta subunits of 30 the protein were successfully synthesized in equal amounts when individually expressed, the protein poses no risk to E. coli, making it a suitable host. Although, the co-expression of the subunits affects the expression levels of the alpha subunit, co-expression of the two subunit proteins in the same cell has various advantages over expression in two different systems with post-expression reconstitution of protein partners. The proper folding of each subunit is due to subunit interactions in vivo. The importance of protein subunit interaction during co-expression has showed increased amounts of properly folded proteins and decreased proteolytic degradation proteins (Sørensen and Mortensen, 2005). Figure 2.4: Co-Expression and individual expression of the nitrile hydratase alpha and beta subunits in T7 Express competent E. coli cells. Cultures containing T7 Express competent E. coli transformed with either the pRSFDuet-1 or the pET21a(+) plasmids or co- transformed with both plasmids were grown until an OD600 = 0.3. The expression of the subunits was induced by the addition of 0.4 mM IPTG and 0.1 mM CoCl2, for 20 hours at 16˚C. The solution after sonication is called the crude lysate (CL) which contains both soluble and insoluble proteins. The crude protein (CP) contains only the soluble protein once the lysate is centrifuged. Lane 1 and 2: CL and CP of co-transformed cells; Lane 3 and 4: CL and CP of 9 8 7 6 54 3 2 1 MW 50 40 30 25 20 15 31 alpha and chaperone subunits; lane 5 and 6: CL and CP of beta subunit; Lane 7 and 8: CL and CP of co-transformed cells that were not induced and Lane 9: untransformed T7 Express competent E. coli cells. The green box indicates the beta subunit which is 27 kDa and the red box indicates the 23 kDa alpha subunit 2.3.3 Transformation efficiencies for co-transformed and individually transformed plasmids Based on a suggestion by Frederick (2006) in order to try increase the expression of the alpha subunit during co-expression a 4:1 ratio of pRSFDuet-1 to pET21a(+) plasmid was transformed into super-competent cells using heat-shock. The transformation efficiencies were calculated as seen in Figure 2.5. The transformation efficiency of the 4:1 ratio of pRSFDuet-1 to pET21a(+) plasmid was significantly lower than the 1:1 ratio, which was expected as an increase in plasmid concentration can hinder the movement of plasmid into the pores of the cell. However, only a single colony was needed to determine whether an increased ratio of the alpha subunit would even out the co-expression of the subunits. The SDS-PAGE analysis of the expression (Figure 2.6) showed that a 4:1 ratio did not increase alpha subunit concentration during co-expression. This result was in contrast to the result obtained by Frederick (2006) who showed that the 4:1 ratio increased the alpha subunit expression. Figure 2.5 shows that the transformation efficiencies of the individually transformed plasmids are significantly different. The pRSFDuet-1 plasmid that contains the alpha subunit is a smaller plasmid compared to the pET21a(+) plasmid and thus has an advantage of moving through the pores during heat-shock more easily. 32 Figure 2.5: Transformation effeciencies of co-transformed T7 Express competent E. coli cells with varied plasmid ratios, as well as independently transformed cells. Heat-shock transformations were done using a 1:1 and a 4:1 ratio of pRSFDuet-1 plasmid carrying the alpha and chaperone subunits to pET21a(+) plasmid containing the beta subunit. Each plasmid was individually transformed into T7 Express competent E. coli, yielding transformation effcieincies for each plasmid. The standard deviation between duplicate measurements is indicated by the error bars.** indicates significant difference of p < 0.01. 33 Figure 2.6: Co-expression of T7 Express competent E. coli cells that were transformed using a 4:1 and a 1:1 ratio of pRSFDuet-1 to pET21a(+) plasmids. The expression of the subunits were induced by the addition of 0.4 mM IPTG and 0.1 mM CoCl2, for 20 hours at 16˚C. The solution after sonication is called the crude lysate (CL) which contains both soluble and insoluble proteins. The crude protein (CP) contains only the soluble protein once the lysate is centrifuged. Lane 1 and 2: CL and CP of 4:1 ratio of co- transformed cells; Lane 3 and 4: CL and CP of 1:1 ratio of co-transformed cells. The green box indicates the beta subunit which is 27 kDa and the red box indicates the 23 kDa alpha subunit 34 Chapter 3: Optimising a high-throughput screening assay for the R. rhodochrous ATCC BAA-870 nitrile hydratase enzyme 3.1 Introduction Developing a sensitive enzyme assay that is suitable for high-throughput screening (HTS) requires identification of relevant substrate forms, buffers, characterisation of co-factors, measurement of kinetic parameters and choice of a detection technology for the final assay (Robinson, 2015). In addition, the cost of reagents, the amount of enzyme required, and the simplicity of implementation within the laboratory must all be addressed while establishing the enzyme assay (Onyeogaziri and Papaneophytou, 2019). HTS assays need to be able to test millions of samples, often over the course of numerous enzyme preparations and weeks of assay repetition. Therefore, HTS assays need to be designed to be highly repeatable. HTS assays must meet stringent requirements that are more demanding than those required by typical bench-top activity assays. The chemical elements of the high- throughput enzymatic assay, such as the buffer and substrates, must remain stable over a long period, and must not deteriorate or be impacted by the equipment and automated machinery used. To ensure high repeatability, reliability and quality, it is of utmost importance to obtain a thorough grasp of the enzymology and biochemistry of the chosen enzyme within the context of the specific assay. Enzyme activity is strongly influenced by a number of factors such as the concentration of substrate, product and enzyme, temperature, pH and ion strength (Bisswanger, 2014). Thus, in order to ensure that the assay is highly accurate, sensitive and reproducible these parameters must be optimised, and the protocol developed must be rigorously maintained. Enzymes have their highest activity under optimal conditions and deviations from those decrease their activity. The higher the divergence, the higher the impact on activity. Since each enzyme is unique, few general rules apply, and scientific research for each enzyme is required for an assay specific application. Current techniques used to detect amide production include high-performance liquid chromatography (HPLC), thin-layer chromatography (TLC), nuclear magnetic resonance (NMR) spectroscopy and mass spectroscopy (Hertzberg and Pope, 2000). These techniques are cumbersome, time consuming, relatively difficult and cannot perform at high-throughput levels (Hertzberg and Pope, 2000). In this study a coupled enzyme assay was used because the products of nitrile hydrolysis by a nitril hydratase (NHase) are amides which are difficult to 35 detect with regular techniques. By coupling the NHase with an amidase enzyme the amide produced by the NHase becomes a substrate for the amidase, which results in production of a more easily detectable compound, hydroxamic acid, that could then be determined with a colorimetric reaction (Wyk and Caroline, 2008). The colour can be visualised easily by eye for qualitative assessment, or the absorbance can be measured quantitatively by a spectrometer. Colorimetric assays can be miniaturised easily which means that multi-well plates and a microplate reader can be combined to create high-throughput abilities. With automated equipment thousands to millions of samples can be tested rapidly. A prominent coupled enzyme assay used in the past for other NHase’s is the phenol- hypochlorite method, developed by Fawcett and Scott (Fawcett and Scott, 1960). This method relies on forming a coloured complex between the phenol-hypochlorite and ammonia group that is released when the amidase catalyses the amide to its respective carboxylic acid (Cameron et al., 2005; Okamoto and Eltis, 2007). However, this assay is not ideal for a high- throughput screening technique for the random mutagenesis of the NHase enzyme because crude protein is used in the assay which means that the ammonia produced endogenously by the microbial cells will be present in the samples. Therefore, this assay can result in false positives. To ensure accuracy and given that the optimal screening strategy is to evaluate the product produced by the enzyme directly (Aharoni et al., 2005) another assay method had to be identified and optimised. The hydroxamic acid assay is usually used to test an amidases acyl transfer ability by measuring the production of hydroxamic acid that results due to the acyl group transfer from the amide to a hydroxylamine (Bhatia et al., 2016; Fournand et al., 1998). By adding an additional incubation step, the NHase activity can be measured indirectly (Scheme 3.1). This assay is more accurate than the phenol-hypochlorite method since the final product is dependent on an amide which would only be present if the NHase was successful. The amidase used in this assay was purchased from Sigma-Aldrich and originates from the species R. rhodochrous ATCC BAA-870. 36 Scheme 3.1: NHase and amidase catalyse the conversion of nitrile to hydroxamic acid. There are various challenges in successfully accomplishing a coupled enzyme assay, as two reactions are addressed one after the other in the same reaction medium. The general issues encountered are differences in the optimal conditions for both enzymes such as pH or temperature change (Mouafo Tamnou et al., 2021). In addition, the enzyme kinetics of coupled enzyme assays are more complicated and open to inaccuracies. Thus, understanding the enzymes kinetics of each enzyme in the coupled assay is essential to ensuring that a sensitive and accurate assay is developed. An enzyme-coupled assay for the high-throughput screening of the NHase from R. rhodochrous ATCC BAA-870 is presented here. In particular, the parameters that effect enzyme activity and the accuracies of the enzyme kinetics were investigated in detail, to: (1) ensure compatibility of the assays, (2) optimise the concentration of substrate for each enzyme to increase the assays sensitivity, (3) identify the concentration and incubation time for each enzyme to optimally detect the reaction kinetics, (4) define the necessary pH and temperature conditions and (5) examine the effects of various organic solvents and nitrile substates on enzymatic activity. 3.2 Materials and methods 3.2.1 Materials Chemicals were purchased from Sigma-Aldrich (SA) and Merck Chemicals. All chemicals were of analytical grade. Care was taken in the preparation of the iron (III) and hydroxamic acid solutions to prevent extensive hydrolysis which could affect the absorption values (Knight and Sylva, 1974; Monzyk and Crumbliss, 1979). The preparation of the nitrile substrates was done on the day of experimentation and were kept in the dark as nitriles are sensitive molecules. 37 3.2.2 Methods The physiological conditions that exist within the organism that produces the enzyme, or literature relating to homologous enzymes, can be used as a guide for setting up the initial assay, offering starting conditions for initial test parameters such as buffer, salt concentrations, pH and temperature. The assay was developed so that the total volume of the NHase incubation step remained at 100 µl while the volume of the amidase incubation step was performed with a further 100 µl volume to result in at 200 µl total volume. Afterwards, 50 µl of acidic ferric chloride solution (0.1 M FeCl3 in 0.1 M hydrochloric acid) was added, and the total acetohydroxamic acid concentration was measured at a volume of 250 µl. This ensured that all concentrations were determined using the same volume as according to Beer’s law the length of the pathway can affect absorbance values. The optimisation of the assay was performed sequentially, so that the results from the previous experiment were used for optimising the next step. 3.2.2.1 Investigation into the effect of hydroxylamine on the assay Hydroxylamine is added to the wells and acts as a nucleophile by accepting the acyl group from the amide. Any chemical in the reaction can influence the activity of the enzyme and therefore needs to be investigated. 3.2.2.1.1 Determining the effect of hydroxylamine on NHase activity in biocatalytic reactions The uncoupled assay involved two incubation steps. The first was with the NHase and the second with the amidase. In the first incubation step, 50 µl of increasing concentrations of hydroxylamine, ranging from 0-100 mM (prepared in 50 mM Tris-HCl pH 7.2), were added to the wells of a 96 well plate. This was followed by the addition of 50 µl of 50 mM acetonitrile (dissolved in 50 mM Tris-HCl, pH 7.2) and lastly 1 µg/µl of crude NHase enzyme (prepared in 50 mM Tris-HCl, pH 7.2) was added to the wells which started the reaction. Additional, 50 mM Tris-HCl, pH 7.2, was added to the wells, so that the total volume of the first reaction was 150 µl. The reaction took place at 37˚C and ran for 30 minutes. The second incubation step involved the addition of 50 µl of 1 unit amidase enzyme (Sigma-Aldrich) (prepared in 50 mM Tris-HCl, pH 7.2), which brought the total assay volume up to 200 µl. The reaction was then allowed to react for a further 30 minutes before being terminated by the addition of 50 µl of an acidic ferric chloride solution (0.1 M FeCl3 in 0.1 M hydrochloric acid). The reddish/brownish 38 colour that formed was measured by the Victor Nivo Multimode Microplate Reader (PerkinElmer, USA) at 510 nm. 3.2.2.1.2 Determining the effect of hydroxylamine on amidase activity in biocatalytic reactions Since specifically the amidase reaction was being investigated here only one incubation step was needed. 50 µl hydroxylamine was added to the wells in increasing concentrations ranging from 0–100 mM (prepared in 50 mM Tris-HCl, pH 7.2). This was followed by the addition of 50 µl of 50 mM acetamide (Sigma- Aldrich) (prepared in 50 mM Tris-HCl, pH 7.2). Lastly, 50 µl of 1 unit amidase enzyme (prepared in 50 mM Tris-HCl, pH 7.2) was added which started the reaction. The 50 mM Tris-HCl, pH 7.2 was added to the wells so that the total volume of the reaction was 200 µl. The reaction was incubated at 37˚C for 30 minutes before it was terminated through addition of acidic ferric chloride solution (0.1 M FeCl3 in 0.1 M hydrochloric acid). The coloured compound was then measured spectroscopically at 510 nm. 3.2.2.2 Determining the optimal concentration of the final product, acetohydroxamic acid, in order to maximise the assay’s sensitivity The ideal concentration of acetohydroxamic acid produces a sensitive signal while the required reactant concentrations are kept to a minimum. To determine the optimal concentration a standard curve was constructed. Increasing concentrations of acetohydroxamic acid (Sigma- Aldrich) ranging from 0 to 50 mM were used. The acetohydroxamic acid was prepared in 50 mM Tris-HCl, pH 7.2 and 200 µl of the solution was added to the wells. An acidic ferric chloride solution (50 µl of 0.1 M FeCl3 in 0.1 M hydrochloric acid) was added to the well. The absorbance of the formed coloured complex was measured by the Victor Nivo Multimode microplate reader (PerkinElmer, USA) at 510 nm. 3.2.2.3 Optimisation of the amidase activity within the assay 3.2.2.3.1 Determining the optimal acetamide concentration on amidase activity Only one incubation step was required because specifically the amidase enzyme was investigated. The acetamide (Sigma- Aldrich) was prepared in 50 mM Tris-HCl, pH 7.2 and 50 µl of increasing concentrations ranging from 0-100 mM were added to the well. In addition, 50 µl of 20 mM hydroxylamine and 50 µl of 50 mM Tris-HCL, pH 7.2, was added to the wells. Lastly, 50 µl of 1 unit amidase was added to the wells to start the reaction. The total volume of the reaction in the well was 200 µl. The reaction was incubated for 30 minutes at 37˚C. The https://www.perkinelmer.com/product/victor-nivo-advanced-f-abs-filter-lu-hh35000500 https://www.perkinelmer.com/product/victor-nivo-advanced-f-abs-filter-lu-hh35000500 39 reaction was then terminated by the addition of 50 µl of 0.1 M FeCl3 in 0.1 M hydrochloric acid and the absorbance was measured at 510 nm. 3.2.2.3.2 Determining the effect of acetonitrile concentration on the amidase activity Since acetonitrile from the first incubation step could be left in the wells during the second, the effect of acetonitrile on the amidase enzyme was investigated. 50 µl of increasing concentrations of acetonitrile ranging from 0 to 100 mM (prepared in 50 mM Tris-HCl, pH 7.2) were added to each well. In addition, the wells contained 50 µl of 100 mM acetamide and 50 µl of 20 mM hydroxylamine which were both prepared in Tris-HCl, pH 7.2. The reaction was initiated with 50 µl of 1 unit amidase (prepared in 50 mM Tris-HCl, pH 7.2). The 96 well plate was incubated for 30 minutes at 37˚C before the reaction was terminated with 50 µl of 0.1 M FeCl3 in 0.1 M hydrochloric acid. The absorbance was measured at 510 nm. 3.2.2.3.3 Determining the optimised concentration of amidase for optimal timely product conversion The optimal amidase concentration was determined by using a 0.2 to 1 unit range of amidase. 50 µl of amidase was added to the wells already containing 50 µl of 100 mM acetamide and 50 µl of 20 mM hydroxylamine. The amidase enzyme and the two substrates were prepared in 50 mM Tris-HCl, pH 7.2. In order to maintain a consistent volume (200 µl) in which the amidase reaction takes place, 50 µl of 50 mM Tris-HCl was also added to the well. The reaction was incubated at 37˚C and was stopped at 10-minute intervals by adding 50 µl of 0.1 M FeCl3 in 0.1 M hydrochloric acid. The absorption was measured at 510 nm. 3.2.2.4 Optimisation of the NHase activity within the assay 3.2.2.4.1 Determining the optimal acetonitrile concentration on the NHase activity The uncoupled assay involved two incubation steps, the first with NHase and second with amidase. In the first incubation step, 50 µl of acetonitrile concentrations ranging from 0 to 100 mM were mixed with 1 µg/µl of crude NHase enzyme. The first incubation reaction was maintained at 100 µl by adding Tris-HCl (50 mM, pH 7.2). The reaction was incubated at 37˚C for 30 minutes. The NHase reaction was inhibited by the addition of 50 µl of 20 mM hydroxylamine. The amidase incubation step began after 50 µl (0.4 units) of amidase was added. All the enzymes and substrates were prepared using 50 mM Tris-HCl, pH 7.2. The reaction was incubated at 37˚C for another 30 minutes before being terminated by addition of 40 50 µl of 0.1 M FeCl3 in 0.1 M hydrochloric acid. The brownish/red colour was measured at 510 nm. 3.2.2.4.2 Determining the effect of acetamide concentration on NHase activity Acetamide is the product of NHase and thus the effect of product inhibition needed to be investigated. In order to evaluate this, varying concentrations of acetamide needed to be present during the NHase incubation step. The wells contained 50 µl of 100 mM acetonitrile, and 40 µl of increasing acetamide concentrations ranging from 0 to 100 mM, and 1 µg/µl of crude Nhase enzyme. In order to maintain a volume of 100 µl during the NHase incubation step extra Tris-HCl (50 mM, pH 7.2) was added to make up the missing volume, depending on the volume of crude NHase solution that was used. The reaction was incubated at 37˚C for 30 minutes. This was followed by the addition of 50 µl of 20 mM hydroxylamine and 50 µl o