THEORETICAL AND EXPERIMENTAL STUDIES OF MULTISTEP ELECTROCHEMICAL BIOSENSORS By Neda Rafat A DISSERTATION Submitted to Michigan State University in partial fulfillment of the requirements for the degree of Chemical Engineering ŠDoctor of Philosophy 2020 ABSTRACT THEORETICAL AND EXPERIMENTAL STUDIES OF MULTISTEP ELECTROCHEMICAL BIOSENSORS By Neda Rafat Electrochemical biosensors are analytical devices that detect analytes by transforming a biochemical reaction into a quantitative, electrical signal. This class of biosensors has proven valuable in research, quality control, food safety, medical diagnosis, and monitoring of therapeutic efficacy. Electrochemical biosensors integr ate specificity of biological recognition molecules (e.g., antibodies) with the advantages of electrochemical detection techniques (reproducible, quantitative electrical output) to provide sensitive and specific analytical devices. Miniaturized amperometri c biosensors that use redox enzymes to generate an electric current in response to the voltage applied at a working electrode have been successfully commercialized . Mechanistic mathematical models that describe the multiple mass -transfer and chemical -react ion steps that give rise to the electrical output are needed to help design, optimize, and validate electrochemical biosensors for medical and environmental applications. In this work, experimental and theoretical studies of two type s of multistep electroc hemical biosensors were performed. An electrochemical immunosensor (EI) was fabricated on screen -printed electrodes (SPEs) for detection of a model protein (mouse IgG) by integrating princip les of an enzyme -labled immunosorbent assay (ELISA) using horserad ish peroxidase (HRP) as the labeling enzyme and an electrochemical transducer. Experimental conditions such as substrates concentration, pH, and applied voltage were optimized using a fractional factorial design. A mathematical model was developed to simul ate the EI™s steady -state signal by solving the non -linear ordinary differential equations including enzyme kinetics and diffusion -based mass transfer rates for all the reactants. A new concept, current -control coefficient, was introduced to measure the ex tend to each reaction step limit ed the current density. The model allow s the rate limiting step to be indentified and experimental condition s that optimize detetction sensitivity to be determined. In addition, experimental and theoretical studies of an inhibition -based bi -enzyme electrochemical biosensor (IBE) for a model inhibitor of acetylcholinesterase (AChE), phenylmethyl sulfonyl fluoride (PMSF), were conducted. The IBE was fabricated by co -immobilization of AChE and tyrosi nase (Tyr) on the gold working electrode of a SPE. Inclusion of a hydrolase enzyme (AChE) and an oxidase enzyme (Tyr) provided an amplification system which improved the biosensor™s sensitivity significantly. A comprehensive mathematical model was develope d to simulate time -dependent electrochemical signal in the IBE. The unsteady -state model was developed by solving a system of non -linear partial differential equations including enzymatic reactions, inhibition kinetics of AChE by an inhibitor (PMSF), and d iffusion -based mass transfer steps. The model successfully simulated the IBE™s response to the substrate (phenyl acetate) and the inhibitor. Using the model along with the current -control coefficient and sensitivity parameters, effect of the governing fact ors on the IBE™s performance were studied. The model allowed to optimize the governing factors to achieve optimum sensitivity for detection of the inhibitor and design the biosensor to achieve specific performance criteria. Copyright by NEDA RAFAT 2020 v Dedicated to my family vi ACKNOWLEDGMENTS I would like to express my sincere gratitude to my supervisor, Prof. Robert Mark Worden for his continuous support, encouragement , and guidance during my P.h.D research. Dr. Worden provided a positive and respectful work environment which helped me to grow my professional skills with his insight, patience, valuable knowledge, and positive attitude. Throughout my Ph.D. study, Dr. Worden always put his effort and time to teach me his great knowledge clearly and simply. I truly feel honored for having the opportunity to be in Dr. Worden™s research group. I would like to thank Dr. Paul Satoh for his sup port and valuable ideas during my Ph.D work. It was an honor to know him and learn a lot from his valuable experiences and knowledge in the field of biosensors and biochemistry. I would like to thank my committee members, Dr. Scott Calabrese Barton, Dr. Dana Spence, and Dr. David Hickey for their helpful ideas, support, and contributions to this work. I would like to thank our collaborators , Dr. Mohsen Zayernouri for his valuable insight and help on developing mathematical models of our biosensors , Dr. Jonathan Hardy for generously sharing his great knowledge and laboratory space with us to conduct research on Listeria monocytogenes , and Dr. Masamitsu Kanada for kindly tutoring me on extraction of extracellular vesicles from breast cancer -cells. vii Finally, I would like to thank my husband, Iman, for all his dedication and support , and my wonderful parents for being my inspiration and motivation to always seek higher goals. viii TABLE OF CONTENTS LIST OF TABLES ......................................................................................................................... xi LIST OF FIGURES ...................................................................................................................... xii KEY TO SYMBOLS AND ABBREVIATIONS ......................................................................... xv 1 Introduction ............................................................................................................................. 1 1.1 Overview ............................................................................................................................. 1 1.2 Enzyme Linked Immunosorbent Assay (ELISA) ............................................................... 4 1.3 Integrated Experimental and Theoretical Studies on an Electrochemical Immunosensor . 5 1.4 Theoretical and Experimental Studies of an Inhibition -based Bi -enzyme Electrochemical Biosensor (IBE) for Detection of Organophosphorus Compounds ............................................ 7 1.5 Use of Electrochemical Detection Techniques for Listeria Monocytogenes Ongoing Research ...................................................................................................................................... 9 2 Enzyme Linked Immunosorbent Assay (ELISA) ................................................................. 12 2.1 Introduction ....................................................................................................................... 12 2.2 Experimental Methods ...................................................................................................... 14 2.2.1 Materials and Instrumentation .................................................................................. 14 2.2.2 ELISA Development ................................................................................................. 15 2.2.3 Color Formation Reaction ......................................................................................... 22 2.3 Results and Discussion ..................................................................................................... 26 2.3.1 ELISA for Mouse IgG .............................................................................................. 26 2.3.2 ELISA for EVs from Breast Cancer Cells ................................................................ 29 2.4 Conclusions ....................................................................................................................... 31 3 Integrated Experimental and Theoretical Studies on an Electrochemical Immunosensor (EI) ––––––––––––––––––––––––––––––––––––32 3.1 Introduction ....................................................................................................................... 32 3.2 Experimental Methods ...................................................................................................... 36 3.2.1 Materials and Instrumentation .................................................................................. 36 3.2.2 Preparation of Immunosensing Layer ....................................................................... 36 3.2.3 Electrochemical Measurement of EI Signal ............................................................. 38 3.3 Optimization of EI Operating Conditions and Characterization of EI Performance Properties .................................................................................................................................. 39 3.4 Mechanistic Mathematical Model ..................................................................................... 40 3.4.1 Kinetics of Enzymatic and Electrochemical Reactions ............................................ 41 3.4.2 Mass Balance Equations ........................................................................................... 43 3.4.3 Boundary Conditions ................................................................................................ 44 3.5 Results and Discussion ..................................................................................................... 46 3.5.1 EI system™s Properties under Optimal Operating Conditions ................................... 46 3.5.2 Validation of Mechanistic Model ............................................................................. 47 ix 3.5.3 Integration of Dimensional Analysis and Flux Analysis to Determine Rate -Limiting Step ....––––––––––––––––––––––––––––––––– 53 3.6 Conclusions ....................................................................................................................... 59 APPENDI X ............................................................................................................................... 62 4 Theoretical and Experimental Studies of an Inhibition -based Bi -enzyme Electrochemical Biosensor (IBE) for Detection of Organophosphorus Compounds .............................................. 71 4.1 Introduction ....................................................................................................................... 71 4.2 Experimental Methods ...................................................................................................... 74 4.2.1 Materials and Instrumentation .................................................................................. 74 4.2.2 Enzyme Electrode Preparation .................................................................................. 74 4.2.3 PMSF Detection and Electrochemical Measurements .............................................. 75 4.3 Mathematical Model ......................................................................................................... 75 4.3.1 AChE Inactivation and Enzyme Kinetics ................................................................. 77 4.3.2 Mass Balance Equations ........................................................................................... 81 4.3.3 Boundary Conditions ................................................................................................ 82 4.3.4 Initial Conditions ...................................................................................................... 84 4.4 Results and Discussion ..................................................................................................... 84 4.4.1 Biosensor™s Response to PMSF ................................................................................ 84 4.4.2 Validation of the Mathematical Model and Simulation of the Biosensor™s Response ..–––––––––––––––––––––––––––––––..87 4.4.3 Signal Amplification ................................................................................................. 91 4.4.4 Biosensor™s Sensitivity .............................................................................................. 93 4.4.5 Rate Limiting Step .................................................................................................... 95 4.5 Conclusions ..................................................................................................................... 100 APPENDI X ............................................................................................................................. 101 5 Use of Electrochemical Detection Techniques for Listeria Monocytogenes Ongoing Research ...................................................................................................................................... 108 5.1 Introduction ..................................................................................................................... 108 5.2 Materials and Instrumentation ........................................................................................ 111 5.3 Experimental Methods .................................................................................................... 111 5.3.1 Cultivating LM ........................................................................................................ 111 5.3.2 Preparation of Chemically Defined Media for LM ................................................. 111 5.4 Motility Assay ................................................................................................................. 111 5.5 Results and Discussion ................................................................................................... 112 5.5.1 Optimization of agar concentration and temperature .............................................. 112 5.5.2 Motility Assay in Complex Media (BHI) ............................................................... 112 5.5.3 Motility Assay of LM in a Chemically Defined Media .......................................... 113 5.6 Conclusion ...................................................................................................................... 115 6 Summary and Recommendations for Future Work ............................................................ 116 6.1 Summary ......................................................................................................................... 116 6.2 Enzyme Linked Immunosorbent Assay (ELISA) ........................................................... 118 6.3 Integrated Experimental and Teoretical Studies on an Electrochemical Immunosensor (EI) ––––––––––––––––––––––––––––––––––... 118 6.4 Theoretical and Experimental Studies of an Inhibition -based Bi -enzyme Electrochemical Biosensor (IBE) for Detection of Organophosphorus Compounds ........................................ 119 x 6.5 Use of Electrochemical Detection Techniques for Listeria Monocytogenes Ongoing Research .................................................................................................................................. 119 REFERENCES ........................................................................................................................... 120 xi LIST OF TABLES Table 3.1. Design of Experiments in coded units suggested by MINITAB using half factorial design. For each factor, three levels, denoted low ( -1), center point (0), and high (+1), were selected: -0.05 V, -0.125 V, and -0.2 V for E; 1.0 mM, 4.5 mM, and 9.0 mM for [C]; 0.5 mM, 1mM, and 1.5 mM for [22]; and 6.2, 6.6, and 7.0 for pH, respectively. ................................ 40 Table 3.2. Values of constants used in the mechanistic model. The 0 and [HRP] values were fit to the experimental data obtained using a constant analyte conc entration of 40 ng/ml. .............. 48 Table 4.1. Parameters and variables used in the numerical simulation. ....................................... 89 Table 5.1. Chemical formula for chemically defined m edia of LM. .......................................... 114 xii LIST OF FIGURES Figure 2.1. Molecular structure of the ELISA for mouse IgG on the bottom of a well in a microplate. .................................................................................................................................... 16 Figure 2.2. Filtering media solution through a 0.22 µm filter ..................................................... 18 Figure 2.3. Filtering media solution through a filter with a 50 nm pore size filter using vacuum ....................................................................................................................................................... 19 Figure 2.4. Size distribution of the EVs. NTA analysis showed an average diameter of 100 nm for the collected EVs from breast cancer cells .............................................................................. 20 Figure 2.5. Molecular structure of the ELISA for EVs from breas t cancer cells on the bottom of a well in a microplate. ...................................................................................................................... 21 Figure 2.6. Reaction steps between TMB -H2O2 and HRP to form colored products. ................. 23 Figure 2.7. Reaction steps between o -quinone and MBTH for form pink colored products. ....... 25 Figure 2.8. Optical ELISA for the mouse IgG. Mouse I gG was detected in a sandwich ELISA with a detection antibody labeled with HRP using TMB -H2O2 substrates. ................................ 26 Figure 2.9. Optical ELI SA for the mouse IgG. Mouse IgG was detected in a sandwich ELISA with a detection antibody labeled with HRP using catechol -H2O2 substrates. ............................ 28 Figure 2.10. Control experiments for measuring the background signal cause by non -specific binding, H2O2 and catechol. ........................................................................................................ 29 Figure 2.11. ELISA for EVs from breast cancer cells using CD81 as the surface biomarker. ..... 30 Figure 2.12. ELISA for EVs from breast cancer cells using CD63 as the surface biomarker. ..... 31 Figure 3.1. Schematic diagram of immunosensing layer showing molecular sandwiches containing the capture antibody, the target analyte, and the HRP -tagged secondary antibody bound to the EI™s gold working electrode. .................................................................................... 38 Figure 3.2. Schematic representation of diffusional mass -transfer, enzyme catalysis and electrochemical reaction steps happening on the biosensor interface. ......................................... 41 Figure 3.3. The dose response for mouse IgG on gold Dropsens SPEs. The dose response for mouse IgG on gold Dropsens SPEs before ( [22]=1.5 mM, pH=7 , [C] = 7 mM, E -Eh = -0.3 V) and after optimization ( [22]=1 mM, pH=6.2, [C] = 8 mM, E -Eh = -0.35 V). Error bars show ± standard deviation from the mean of 3 replicates. ........................................................... 47 xiii Figure 3.4. Pareto chart showing the standardized effect (SE) of factors E, [C], pH, and [H_2 O_2] on the biosensor signal. The terms with an SE value greater than the threshold value marked with the dotted line (SE=2.09) exerted a statistically significant effect on biosensor signal at the 95% confidence level. ............................................................................................... 49 Figure 3.5. Effect of worki ng electrode overpotential (E -Eh) on the steady -state EI™s signal. [C]=8mM, [H_2 O_2] =1 mM, pH=6.2, [HRP]=0.5µM. ............................................................. 50 Figure 3.6. Eff ect of [C] on the steady -state EI™s signal: comparison of model prediction and experimental data. [H 2O2]=1 mM, pH=6.2, [HRP]=0.5µM, E -Eh = -0.35 V. .............................. 51 Figure 3.7. Effect of [H 2O2] on the steady -state EI™s signal: comparison of model prediction and experimental data. [C]=8mM, pH=6.2, [HRP]=0.5µM, E -Eh = -0.35V. ..................................... 52 Figure 3.8. Simulation of pH effect on the steady -state EI™s signal. [C] = 8mM, [H 2O2] = 1 mM, [HRP] = 0.5µM, E -Eh = -0.35 V. ................................................................................................. 53 Figure 3.9. A: Simulated [Q ](x=0) over a range of (E -Eh) values B: Error percentage caused by assuming [Q ](x=0) =0 as a function of (E -Eh). Error percentage = [(J assuming [Q ](x=0) =0 - J using calculated value of [Q ](x=0) ) / J using calculated value of []=0] *100. [C]=8mM, [22] =1 mM, pH=6.2, [HRP]=0.5µM. .................................................................................... 55 Figure 3.10. Predicted current density (J) and current -control coefficients for the electrochemical reaction at different E values. [C]=8mM, [22] =1 mM, pH=6.2, [HRP]=0.5µM. .................. 58 Figure 3.11. Sensitivity -control coefficient and sensitivity vs. E -Eh. [C]=8mM, [22] =1.0mM, pH=6.2, [HRP]=0.5µM. ................................................................................................................ 59 Figure 4.1. Schematic representation of reactions happening on the surface of gold working electrode. S1, S 2, S3, and S 4 denote phenyl acetate, phenol, catechol, o -quinone respectively. E 1, E2, and E 3 denote acetylcholinesterase, tyrosinase's phenolase activity, and tyrosinase™s catecholase activity. ...................................................................................................................... 76 Figure 4.2. Hydrolysis of phenylacetate with AChE. ................................................................... 77 Figure 4.3. Molecular structure of PMSF. .................................................................................... 78 Figure 4.4. Inhibition mechanism of AChE (E) with PMSF (I) in the presence of substrate (S1). ....................................................................................................................................................... 78 Figure 4.5. Scheme of phenol oxidation with tyrosinase to produce O -quinone. ......................... 80 Figure 4.6. Current vs. time response of the bi -enzyme biosensor to the addition of phenylacetate (S1) to obtain final phenylacetate (S 1) concentration of 0.9 mM followed by the add ition of inhibitor PMSF to obtain a final PMSF concentration of 0.17 mM. ............................................ 85 Figure 4.7. Control experiment to study the effect of p hosphate buffer addition on the bi -enzyme biosensor™s signal. ......................................................................................................................... 86 xiv Figure 4.8. Current vs PMSF concentration. Error bars indicate mean ± standard devi ation of 3 replicates. Phenylacetate: 0.9 mM. ............................................................................................... 87 Figure 4.9. Simulated the bi -enzyme biosensor™s signal vs. time. ................................................ 90 Figure 4.10. Simulated current density vs. PMSF concentration (I). ........................................... 91 Figure 4.11. The simulated current density with and without (E3=0) amplification system in the bi-enzyme biosensor. .................................................................................................................... 92 Figure 4.12. Signal amplification in bi -enzyme biosensor due to S3 recycling caused by catecholase activity (E3). .............................................................................................................. 93 Figure 4.13. Sensitivity vs. phenylacetate concentration (S 1). S 1 was normalized with K m,1,app . . 94 Figure 4.14. Sensitivity vs [AChE] at different tyrosinase concentrations. [AChE] was normalized with [AChE*] =3 µM. [I] = 0.3 mM. ......................................................................... 95 Figure 4.15. Current -control coefficient vs [AChE]/[AChE*]. .................................................... 97 Figure 4.16. Current -control coefficient for tyrosinase. Tyrosinase concentration was normalized with [AChE*]=3µM. ..................................................................................................................... 98 Figure 4.17. Current -control coefficient vs. applied voltage. ....................................................... 98 Figure 5.1. Motility assays of luminescent LM (1C) in 0.15% agar in BHI. ............................. 113 Figure 5.2. A: Motility assay of wild type LM in the unoptimized defined media. B: Motility assay of wild type LM in the optimized defined media. Some crystals were formed after the addition of magnesium sulfate in the optimized media. ............................................................. 115 xv KEY TO SYMBOLS AND ABBREVIATIONS ELISA Enzyme linked immunosorbent assay PMSF Phenylmethylsulfonylfluoride NTE Neuropathy target esterase OP Organophosphorus compound EI Electrochemical immunosensor IBE Inhibition -based bi -enzyme electrochemical biosensor LM Listeria monocytogenes EV Extracellular vesicle POC Point -of-care HRP Horseradish peroxidase AChE Acetylcholinesterase Tyr Tyrosinase EDC 1-Ethyl -3-(3-dimethylaminopropyl) carbodiimide NHS N-Hydroxy succinimide BSA Bovine serum albumin SPE Screen -printed electrode RSM Response surface methodology PCR Polymerase chain reaction EET External electron transfer CAGR Compound annual growth rate TMB 3,3',5,5'-Tetramethylbenzidine H2O2 Hydrogen peroxide MBTH 3-Methyl -2-benzothiazolinone hydrazone hydrochloride hydrate PBS Phosphate buffered saline NTA Nanoparticle tracking analysis xvi WT Wild type BHI Brain heart infusion E Applie d electrochemical potential on the working electrode C Catechol Q O-quinone Eh Standard electrochemical potential D Diffusion coefficient Diffusion coefficient in the diffusion layer Diffusion coefficient in the immunosensing/ enzyme -containing layer Maximum enzymatic reaction rate Catalytic rate constant in Michaelis -Menten kinetics Michaelis -Menten constant of HRP for catechol Michaelis -Menten constant of HRP for hydrogen -peroxide Current density n Number of transferred electrons in a redox reaction m Number of transferred protons in a redox reaction F Faraday constant R Universal gas constant T Temperature Midpoint potential of a voltammogram L Thickness of immunosensing/enzyme -containing layer Thickness of diffusion layer Partition coefficient Current -control coefficient Sensitivity S1 Phenyl acetate S2 Phenol xvii S3 Catechol S4 O-quinone E1 Acetylcholinesterase activity E2 Phenolase activity of tyrosinase E3 Catecholase activity of tyrosinase I Inhibitor of acetylcholinesterase pseudo -first -order rate constant for the inactivation of AChE with PMSF Reaction constant of deact ivation of acetylcholinesterase with PMSF Dissociation constant of PMSF Forward rate constant for binding of the inhibitor to AChE Backward rate constant for binding of the inhibitor to AChE Heterogeneous electron transfer rate AF Amplification factor Damkohler number 1 1 Introduction 1.1 Overview A biosensor is defined as a device that utilizes a biochemical mechanism to transfer concentration or presence of a specific sample component (analyte) to a detectable signal [1-3]. Biosensors have a wide range of applications including environmental monitoring, disease detection, food safety, drug discovery, etc [4]. A biosensor includes two major components: a biological recognition element or a bioreceptor and a transducer [3, 5, 6]. Biological recognition element or bioreceptor specifically targets the analyte by using a biochemica l mechanism for recognition. Bioreceptors can be generally divided into five categories: enzyme, antibody/antigen, nucleic acid/DNA, cellular structure/cell, and biomimetics. The enzymes and antibodies are the most commonly used type of bioreceptors in bio sensor applications [3, 7, 8]. The main categories of transduces in biosensor applications are electrochemical transducer, optical transducer, piezoelectric transducer, and gravimetric transducer [7, 9]. Electrochemical transducers report the presence or concentration of analyte in the form of an electrical signal. Electrochemical biosensors integrate the sensitivity of electrochemical transducers and their low limit of detection with the high specificit y of the bioreceptors. Electrochemical biosensors benefit from several advantages such as low cost, ease of use, portability, and simplicity of construction. These advantages make electrochemical biosensors great options for development of analytical devic es in different fields [10, 11]. The electrochemical biosensors can be divided in f our major categories based on the electrochemical technique which is used to measure the electrical signal produced by the biochemical mechanism: amperometric biosensors, potentiometric biosensors, conductometric biosensors, and impedimetric biosensors [12]. In amperometric biosensors, electric current flow between two electrodes is measured (usually at a fixed applied electrochemical potenti al on a working electrode) 2 when a redox reaction takes place on the working electrode [13]. In potentiometric biosensors, the electrochem ical potential difference between a working electrode and a reference electrode is measured. This potential difference is related to the analyte concentration [14]. Conductometric biosensors meas ure the electrical conductivity in the sample solution , which can be changed by changing the analyte concentration [15]. In impedimetric biosensors, an analyte is detected by measuring the change in the impedance of the syst em, which is caused by the biochemical reaction between the bioreceptor and the analyte [16]. While each electrochemical transducer has its unique advantages, in this work, we have been focused in amperometric transducers due to their high sensitivity, simplicity of their construction, relative low background signal and wide linear range [17]. This dissertation describes theoretical and experimental studies of two electrochemical biosensors: an electrochemical immunosensor (EI) for a model antigen and a n inhibition -based bi-enzyme electrochemical biosensor (IBE) for the detection of a model inhibitor of acetylcholinesterase . The EI was developed by the integration of an amperometric transducer with the principal of the enzyme -labeled immunosorbent assay (ELISA). The IBE was developed by including a hydrolase enzyme (acetylcholinesterase) and an oxidase enzyme (tyrosinase) . The i nclusion of the two enzymes provides an amplification system that improve s the biosensor™s sensitivity significantly . While the underlying theme of th is study is the development of electrochemical biosensors, each chapter in this thesis addresses a unique architecture or issue . Chapter 2 of this dissertation d iscusses the princip les of an optical ELISA , and this high throughput assay was used to find and optimize the type of bioreceptors before developing an EI. In Chap ter 3, the theoretical and experimental studies of an EI for a model antigen, mouse IgG, is discussed. This chapter reports a unique and novel mathematical model for the simulation and 3 optimization of the steady -state EI™s signal. A new concept, the current -control coefficient, is introduced to measure the extent that each reaction step is limiti ng the current density. The model allow s to predict the rate limiting step and optimize experimental conditions for improving the sensitivity of detection for the mouse IgG. Chapter 4 of this dissertation is devoted to the theoretical and experimental st udies of an IBE for the detection of organophosphorus compounds. IBE is fabricated by co -immobilization of acetylcholinesterase (AChE) and tyrosinase (Tyr) on the gold working electrode of an SPE. The inclusion of a hydrolase enzyme (AChE) and an oxidase e nzyme (tyrosinase) provide s an amplification system that significantly improved the biosensor™s sensitivity. A comprehensive mathematical model is presented to simulate the time -dependent electrochemical signal in the IBE. The unsteady -state model is developed by solving a system of non -linear partial differential equations, including enzymatic reactions, inhibition kinetics of AChE by an inhibitor (PMSF), and diffusion -based mass transfer steps. The model successfully simulate s the IBE™s response to the substrate (phenyl acetate) and the inhibitor. Using the model and the current -control coefficient and sensitivity parameters, the effect of the governing factors on the sensitivity are examined . The model provides a platform to optimize the governing f actors to achieve optimum sensitivity for detecting the inhibitor . Finally, in Chap ter 5, it will be discussed how the developed EI and previous research in Dr. Worden™s research group in the field of chromatids, can be applied for the ongoing research of Listeria monocytogenes . 4 1.2 Enzyme Linked Immunosorbent Assay (ELISA) The enzyme -linked immunosorbent assay (ELISA) is a commonly used analytical biochemistry assay that is developed based on the strong and specific antibody -antigen interactions [18]. This is a plate -based assay technique which is frequently used for detecting and quantifying peptides, proteins, antibodies, toxins, pathogens, and hormones [18, 19]. ELISAs are typically performed in 96-well polystyrene plates where the analyte is immobilized on the bottom of wells directly or with the aid of an antibody. Then, an enzyme conjugated antibody is used to detect the immobilized anal yte. The enzyme reacts with a substrate to produce a measurable optical signal which its intensity is related to the analyte concentration. ELISAs can be divided into three categories based on their binding structure of antibody and antigen: indirect ELISA , competitive ELISA and Sandwich ELISA [20]. In this work, the focus was on integration of electrochemical transducers with sandwich ELISAs. In a sandwich ELISA, a capture antibody against the analyte is coated on the plate to detect the analyte from the sample solution. Then a secondary antibody conjugated to an enzyme is added to detect the antibody -analyte complex. Finally, a substrate is added that can react with the immobilized enzyme to produce an optical signal which its intensity is related to the analyte concentration. This technique benefits from a high specificity as two antibodies are used to specifically detect and bind the analyte [21, 22]. Because conventional ELISAs in a 96 -well plate provide a high throughput standard platform to develop an immunoassay against a specific analyte, they were performed in this work prior to the development of the EI for a specific analyte. Developed optical ELISAs are discussed in Chap ter 2. An ELISA was performed against a model antigen, mouse IgG, to validate functionality of antibodies and optimizes the governing factors. Once a successful optical ELISA was developed wit h the proper antibodies, the principal of ELISA was integrated with an amperometric transducer to develop an EI. An optical ELISA was also developed for the detection of extracellular vesicles (EVs) from breast cancer cells. EVs are 5 membrane -bound vesicles that can be produced by any type of live cells , including breast cancer cells [23]. In this work, we were interested to develop an electroc hemical biosensor for the detection of EVs from breast cancer cells as an approach for detection of breast cancer cells, but before that, it was important to find a proper surface biomarker on the EVs to target them. Recent studies have shown that EVs carr y surface biomarkers which can be specific to their cell of origin. Some works have shown elevated concentrations of some of the tetraspanin proteins , including CD63 and CD81 on EVs from breast cancer cells [23, 24]. Before developing an EI for EVs from breast cancer cells , it was crucial to find out which surface biomarker is efficient for the detection of the EVs. Chapter 2 discusses the development of the optical ELISAs and how their results guided us for fabrication of EIs . 1.3 Integrated Experimental and Theoretical Studies on an Electrochemical Immunosensor Continuous moni toring and screening of biological and chemical pathogens, contaminants, and biomarkers play an important role in prevention of disease spread and pathologies, early diagnosis of cancers, and the study of the efficacy of treatments [25]. Electrochemical immunosensors benefits from several advantages, which make them a great option in point of care (POC) and on -site diagnostics. They are developed by integrating immune pr inciples and electrochemical transducers. High sensitivity, ease of use, low cost, portability, and having potential for automation and miniaturization are some of the advantages of electrochemical biosensors [26, 27]. Antibodies are universal biorecognition molecules that recognize their corresponding antigen to form highly specific antigen -antibody complexes. E xtreme affinity and selectivity of antigen -antibody interactions provide great sensitivity for electrochemical immunosensors [28, 29]. Sand wich immunoassay has been widely used in electrochemical immunosensors. In this assay, a capture 6 antibody immobilized on the electrode captures the antigen and form an antibody -antigen complex. Then, a labeled detection antibody is used to detect and quant ify antibody -antigen complexes [30]. Horseradish peroxidase (HRP) is the most commonly used enzyme label in sandwich immunoassay. This is mainly because of the high catalytic activity of HRP, its commercial availability, and its capacity for oxidation of a wide range of substrates [31]. Besides, due to its relatively small molecular size and stability to chemical modification, HRP is very suitab le for the labeling of immunological reagents [32]. HRP has been used as the labeling enzyme in a wide range of disposable electrochemical immunosensors . While some mathematical models have been developed to study the kinetics of HRP [33-37], there is a lack of a comprehensive mathematical model that can be used to study, optimize, and simulate the HRP induced electrochemical signal under different experimental conditions and predict the rate limiting step. Chapter 3 presents a mechanistic model to simulate the steady -state signal in an electrochemical immunosensor (EI) having HRP as the labeling enzyme. An electrochemical immunosensor was devel oped for mouse immunoglobulin G ( IgG), as a model antigen, using HRP as the labeling enzyme. Immunosensing layer was prepared by using EDC -NHS chemistry since this chemistry is widely used for the preparation of the immunosensing layer in electrochemical immunosensors. [38-40]. EDC (1-Ethyl -3-(3-dimethylaminopropyl)carbodii mide ) is a zero -length cross -linker, which cause s the coupling of primary amines to carboxylate groups. The a ddition of NHS (N-Hydroxysuccinimide ) to EDC reactions increases the efficiency of the coupling reaction [41]. Experimental variables such as substrates concentration, electrochemical potential, and pH affect a biosensor sensitivity [42]. Therefore, a statistical model was developed using response surface methodology (RSM) to optimize multiple experimental variables influencing our biosensor signal. RSM is an efficient statistical method for screening and optimizing multiple variables influencing a resp onse [43]. The 7 mechanistic model was developed by solving a set of non -linear differential equations , including diffusion equations coupled with non-linear enzymatic reactions. A bi -substrate ping -pong mechanism was assumed for the enzymatic reaction catalyzed by HRP [44-46]. Our mechanistic model provided a platform to study the effect of applied electrochemical potential, hydrogen peroxide concentration, catechol concentration , and pH on steady -state signal and it allowed simulating the signal under different experim ental conditions. Besides, it helped to understand and study the mass transfer steps and reactions happening on the biosensor interface. Concepts such as current -control coefficient, sensitivity, and Damkohler number were introduced in this chapter using t he mechanistic model to predict the rate limiting step and study the effect of the governing factors on sensitivity. Knowing such information would help to optimize governing factors for improving the EI™s performance. 1.4 Theoretical and Experimental Studies of an Inhibition -based Bi-enzyme Electrochemical Biosensor (IBE) for Detection of Organophosphorus Compounds Organophosphorus compounds (OPs) are the main group of insecticides (malathion, parathion, diazinon, fenthion, dichlorvos, chlorpyrifos, and others) and nerve gases (soman, sarin, tabun, and VX)[47]. OPs are synthetic chemicals first synthesized in early 1800 [48]. OPs have been commercially developed as pesticides for over five decades , and they still are used as pesticides and insecticides [49]. OPs eradicate pests by deactivating an important enzyme in the body ca lled acetylcholinesterase (AChE). AChE is a vital enzyme responsible for controlling nerve signals in the body[50, 51]. The widespread use of OPs cause their accumulation in s oil and aquatic organisms and poses a serious risk to non -target species , including humans and animals [26] . Extensive use of OPs in modern agriculture pesticides and th eir high toxicity requires development of analytical devices that can effectively monitor environmental samples and food samples [52]. 8 The current gold standard technique for the detection of OPs is based on the principles of chromatography [53, 54]. Although these techniques provide accurate results for detection of OPs, they are expensive, time -consuming, require special trained tec hnician s, and cannot be used for on-site applications [55]. In contrast, electrochemical biosensors offer several advantages such as high sensitivity, fast response, the potential for being miniaturized and portable, and bei ng cost effective. These advantages make electrochemical biosensors great options for the development of portable analytical devices for OPs detections [56]. In general, electrochemical biosensors for detection of OPs are developed based on three main principles: inhibition of acetylcholinesterase (AChE) or butyrylcholinesterase with OPs, inhibition of enzymes phosphatase with OPs, and direct electrochemical detection of OPs [56]. In this work, we have been interested in the development of an inhibition -based bi-enzyme electrochemical biosensor (IBE) for the detection of OPs using AChE. OPs can covalently bind to the active site of AChE to inhibit it. The amount o f inhibition in AChE activity is related to the OPs concentration. The inhibition of AChE with OPs is the principle of the detection method in inhibition -base electrochemical biosensors [57-60]. Acetylcholine and acetylthiocholine are two commonly used substrates in AChE electrochemical biosensors. AChE hydrolyses acetylcholine to produce c holine which can be detected by use a second enzyme, choline oxidase [59]. In the case of acetylthiocholine, product of the hydrolase reaction catalyzed by AChE is thiocholine which can be oxidized to produce an electric signal which is related to AChE activity. In this work, we were interested in combining AChE with an oxidase enzyme, tyrosinase (Tyr) , to develop an IBE with an amplification system for the detection of OPs. This biosensor is a modified version of a novel electrochemical biosenso r that was previously developed in Dr. Worden™s group to measure the activity of a hydrolase enzyme, 9 neuropathy target esterase [61]. Chapter 4 discusses theoretical and experimental studies of the developed IBE for the detection of a model inhibitor . It would be discussed how the inclusion of AChE and Tyr created an amplification system which improve the sensitivity of detection for a model inhibitor, phenyl methyl sulfonyl fluoride (PMSF). Besides, a comprehensive mathematical model is presented for simulation of the unsteady -state electric signal in the presented biosensor. The mathematical model provides a platform to estimate the rate limiting step in the biosensor and optimize the experimental condition in a way that maximum sensitivity is obtained. 1.5 Use of Electrochemical Detecti on Techniques for Listeria Monocytogenes Ongoing Research Chapter 5 of this dissertation discusses briefly how the previously developed techniques in Dr. Worden™s group in the field of chemotaxis and electrochemical biosensors can be applied to the Listeria monocytogens (LM) ongoing research. This work was perfor med with close collaboration with Dr. Jonathan Hardy (MSU Microbiology and Molecular Genetics Dept). LM is a gram -positive facultative intracellular pathogen, which can invade and multiply within mammalian cells [62]. Listeriosis, a serious infection caused by LM with a global mortality rate of 24 %, is most likely to infect high -risk population groups, including pregnant women, their fetuses, adults over 65 years old, and immunoco mpromised people [63]. According to the Centers for Disease Control and Prevention (CDC), 1600 people are diagnosed with listeriosis within the United States annually, 260 of which lead to death. LM can grow and survive under a wide range of environmental conditions , including high salt concentrations, anaerobic environments, refrigeration temperatures, and acidic conditions. Besides, LM can produce biofilms on food product ion equipment plants that allow their organism to survive for more than ten years. All these features favor LM as a foodborne pathogen and make it ubiquitous in the environment [63, 64]. To 10 avoid health risks associated with LM, it is important to be able to detect this pathogen in different environments and food samples and as well as to learn more about its survival mechanism in adverse growth conditions such as anaerobic conditions [65]. Some of the common detection methods of LM are culturing, biosensors, enzyme linked immunosorbent assay ( ELISA), and polymerase chain reaction -based method (PCR). Among these methods, biosensors are the latest techniques which provide a low detection limit [65]. Recently, it was discovered that LM can secrete biologically active extracellular vesicles (EVs) despite having a thick cell wall and lack of outer membrane [66-68]. These EVs, with a diameter ranging from 20 to 200 nm, can be used as toxin cargo to transport concentrated virulence factors to host cells [66]. Another recent study showed for the first time that under anaerobic conditions, LM uses a mechanism called ext racellular electron transfer (EET) to transfer electrons produced in respi ration to extracellular soluble and insoluble electron acceptors [69]. The s ame study investiga ted the effect of the EET mechanism for colonization of the gut. Because oxygen levels are low in the intestinal lumen, anaerobic growth capabilities strongly enhance microbial proliferation there. The g rowth of a mutant deficient in EET in a gut model was six -fold lower than that of the wild -type Listeria with full EET capability [69]. This finding suggests that the EET mechanism helps LM to respire under anaerobic and facilitate the development of Listeriosis following LM ingestion. However, it is currently unknown whether the newly discovered EV s also participate in EET. If so, strategies to inhibit EV -mediated EET might be effective in preventing or mitigating the severe health problems associated with Listeriosis. Previously, Dr. Worden™s research group showed for the first time that using the EET mechanism, Shewanella oneidensis cells can locate insoluble electron acceptors in an anaerobic environment [70]. Using a set of motility assays, it was shown that the cells use reduced riboflavin as both an electron shuttle and an attractant to tactically move toward local 11 insoluble electron acceptors to respire under anaerobic conditions. In this work , we were interested in test if LM tactically move s toward external electron acceptors (similar to Shewanella oneidensis), to develop electrochemical immunosensors for detection EVs from LM, and to test the hypothesis if these EVs p articipate in anaerobic respiration of LM via EET. To perform the needed experiments to test the mentioned hypothesis, we needed to isolate EVs from LM, confirm motility of the used strain s of LM, and develop a chemically defined media which allow s studyin g tactic movement (chemotaxis) of LM under a controlled growth condition. Chapter 5 discusses how these steps were performed. 12 2 Enzyme Linked Immunosorbent Assay (ELISA) 2.1 Introduction Immunoassays are a class of analytical techniques that are perhaps the most commonly used method for measuring biological compounds in translational and clinical research [71]. Immunoassays are developed based on the princip le of the immune system wherein a specific antigen (analyte) reacts with specific antibodies [72]. Epitopes on the antigen and the binding site of the antibody have specific chemical structure and spatial configuration. Therefore, an antibody can only bind to an antigen , which has the complementary epitope for the antibody™s binding site. This selective, specific, and stable formation of the antibody -antigen complex makes immunoassays highly specific [73]. Immunoassays are wi dely applied for clinical, pharmaceutical, and environmental applications due to their intrinsic advantages such as being high throughput, very specific and sensitive (due to the highly selective and stable antibody -antigen binding), adaptable to a wide ra nge of analytes, and relatively cost effective [74]. Enzyme -linked immunosorbent assay (ELISA) is a type of immunoass ay that uses an enzyme -linked antibody to measure and detect an analyte. In a conventional ELISA, the target an alyte is immobilized on the surface of a microplate (directly or via a capture antibody) and then complexed with an enzyme -linked antibody [18]. The amount of bound enzyme is directly related to the analyte concentration. The bound enzyme catalyzes a reaction in which its products produce a measurable signal which is related to the bound enz yme concentration. Therefore, by measuring the enzyme -induced signal, the analyte concentration can be measured. The global market for ELISA was valued at USD 1,583.4 million in 2016 and is estimated to reach USD 2.5 billion by 2025 with a compound annual growth rate (CAGR) of 5.1 %. ELISA market is pred icted to increase significantly in the next five years due to the high demand for cost -effective 13 diagnostic tools and the increasing incidence of infectious diseases and cancer [75]. For example, the massively increasing rate of COVID -19 cases has produced a high demand for antibody detection kits globally. The global market for COVID -19 antibody detection kits has a value of USD 5,406 million in 2020 , with a CAGR of 10.16 % [76]. In this work, we have been interested in integrating the advantages of an optical ELISA and an amperometric transducer to develop a sensitive and specific EI. Because a conventional optical ELISA in a 96 - microplate provides a high throughput platform to evaluate the efficiency of reagents and antibodies for detection of a specific antigen, we performed optical ELISAs to validate the functionality of antibodies and reagents which were going to be used in o ur EI. In this chapter , developed optical sandwich ELISAs for two different analytes will be discussed. Horseradish peroxidase (HRP) was used as the reporter enzyme for the following reasons: HRP is the most commonly used enzyme in ELISAs due to a relative ly high turnover number, smaller molecular size, and stability , which makes the antibody -conjugation process more effective; it is relatively cost -effective [77]; there is a versatile range of colorimetric and redox active substrates available for HRP whic h makes it suitable for being used in both optical ELISAs and electrochemical immunosensors [78]. An optical sandwich ELISA was developed using 3,3', 5,5fl-tetramethylbenzidine (TMB) - H2O2 and catechol -H2O2 system for detection of a model antigen (mouse IgG) . This chapter will discuss how the optical ELISA helped to evaluate the functionality of antibodies and study the source of background signal , which was observed in the EI. Another optical ELISA was dev eloped to detect extracellular vesicles (EVs) produced by breast cancer cells. EVs are lipid bilayer -delimited particles that are released by cells to the extracellular environment [79]. Several studies have shown that cancer cells release more EVs than normal cells 14 and cancer EVs contain surface biomarkers and cargo specific to their cell of origin [80-82]. Biological fluids contain a large amount of EVs , which have stable sources of biomarkers that are unique to their cell of origin. EVs derived from cancer cells can be used as a biomarker in liquid biopsy to provide a minimally invasive approach for cancer diagnosis [83]. In this project , we were interested to develop an optical ELISA for the detection of EVs from breast cancer cells. Studies have shown that the concentration of some tetraspanin proteins such as CD63 and CD81 are elevated in EVs produced by breast cancer cells , and they are used as classical biomarkers for cancer detection [84]. CD63 and CD81 are glycoproteins that are present on the membrane of EVs [85]. We have used the optical ELISA to evaluate which surface biomarker (CD63 or CD81) would be more efficient for detecting breast cancer cells. Once the proper surface biomarker is known, an electrochemical immunosensor would be developed to detect EVs from breast cancer cells. 2.2 Experimental Methods 2.2.1 Materials and Instrumentation Maxisorp Immuno Clear Standard Modules 96 microplates (Nunc, 469914) and 1 -StepŽ Ultra TMB -Blotting solution (37574) were purchased from Thermofisher Scientific. Mo use IgG, anti -mouse IgG antibody (ap124), HRP -conjugated -goat anti -mouse IgG (a5278), 3 -methyl -2-benzothiazolinone hydrazone (MBTH), catechol, Bovine serum albumin (BSA, a3059), and TWEEN 20 were obtained from Sigma Aldrich. CD63 Antibody -HRP conjugated (NBP2 -42225H) was obtained from Novus Biologicals. Anti -CD63 (215 -820) was obtained from Ancell corporation. CD81 antibodies were obtained from Cosmo Bio US. MF-MilliporeŽ Membrane Fil ter, 0.05 µm pore size (VMWP02500) and Millex -GP Syringe Filter Unit, 0.22 µm, polyethersulfone, 33 mm, gamma sterilized (SLGP033RB) were purchased from Mi llipore Sigma. -UV four -stage purifier (Barnstead 15 Ultrapure water was used in all aqueous solutions. A Synergy H1 hybrid multi -mode plate reader was used to measure the absorbance in the plates. 2.2.2 ELISA Development 2.2.2.1 ELISA for Mouse IgG In this section, we discuss the optical ELISA for a model antigen, mouse IgG. The optical ELISA was conducted to choose the best antibodies for the assay and optimize some of the experimental conditions such as antibody dilutio n. Once the optical ELISA was optimized, it was integrated with an electrochemical transducer to develop an electrochemical immunosensor. First, 200 µl of [1:333] dilution of the primary antibody (ap124) in 50 mM phosphate buffer at pH 7.4 was added to the wells of a 96 -well plate , and the microplate was incubated at 4 overnight. Next day, after being washed with PBS, wells were filled thoroughly with 1% BSA in phosphate buffer to block the sites, which have the potential for causing non -specific binding , and the plate was incubated at room temperature for one hour. After being washed with PBS, 200 ul of the samples containing different concentration s of mouse IgG prepared in 1% BSA were added to each well. The plate was incubated at room temperature for two hours. Then, wells were washed five times with a washing buffer containing 0.5% TWEEN 20 in PBS. Next, 200 µl of [1:333] dilution of the detection antibody (a5278) in 1% BSA in 50 mM phosphate buffer at pH 7.4 was added to the wells . The plate was incubated at room temperature for an hour (Note: because HRP is light sensitive, this step and the consequent steps were performed in a dark space to avoi d deactivation of the HRP). Next, wells were washed thoroughly for 5 times with the same washing buffer with 16 washing buffer (Figure 2.1). The optical signal was measured using TMB -H2O2 system or catechol -H2O2 system which have been discussed in the next section. 2.2.2.2 ELISA for EVs from Breast Cancer Cells All eukaryotes and prokaryotic cells release extracellul ar vesicles (EVs) into the extracellular environment. EVs are lipid bilayer -enclosed, cytosol -containing spheres that play an important role in intracellular transfer of signaling molecules, functional proteins, nucleic acids, lipids , and virulence factors [86]. Biological fluids contain a large amount of EVs which have stable sources Figure 2.1. Molecular structure of the ELISA for mouse IgG on the bottom of a well in a microplate. 17 of biomarkers that are unique to their cell of origin. EVs derived from cancer cells can be used as biomarker in liquid biopsy to provide a minimally invasive approach for cancer di agnosis [83]. Studies have shown that the concentration of some tetraspanin proteins such as CD63 and CD81 are elevated in EVs produced by breast cancer cells and they are used as classical biomarkers for cancer detection [84]. CD63 and CD81 are glycoproteins which are present on the membrane of EVs [85]. Herein, we present an optical ELISA to detect EVs from breast cancer cells using CD81 and CD81 as the surface biomarkers. To develop the ELISA, we needed to culture breast cancer cells to collect E Vs. The procedure for the collecting EVs from breast cancer cells is discussed in the next section. 2.2.2.2.1 Culturing Breast Cancer Cells MDA -MB-231 cells were seeded at a density of 1 -3 million cells /75 cm 2 flask in 10 mL of growth medium and then incubated for 48 h in medium containing 10% Exo -depleted FBS at . At least 5 flasks were used. The media was transferred from each plate to a separate 15 mL tube. The tubes were Centrifuged at 600 x g for 10 mi nutes. The supernatant from each tube was transferred to a new 15 mL tube and the pellet (included dead cells and cell debris) was discarded. Tubes were centrifuged at 2,000 x g for 30 minutes. The supernatant was combined into a 50 mL tube or stored at -20 C until ready to use. 2.2.2.2.2 Purification of EVs The supernatant was filtered through a 0.22 µm filter and the filtrate was collected in a 50 mL tube (Figure 2.2). The collected solution was passed through a vacuum filtration (QIAvac 24 Plus) using the 50 nm filter (Figure 2.3, the filter was changed for every 20 mL of solution). The vacuum filtration was continued until the solution was just above the filter. Then, 5 mL of the PBS was 18 added, and the filtration continued until three -quarters of the filter space contained PBS. The solution above the filter contained the concentrated EVs. The EVs were stored at -80 C. Figure 2.2. Filtering media solution through a 0.22 µm filter 19 2.2.2.2.3 Quantification of EVs by NTA Analysis Nanoparticle tracking analysis (NTA) was used to quantitate extracellu lar vesicles (EVs) in the collected sample solution from the previous step . NTA is a commonly used technique to determine the particle size distribution and concentration of a sample containing nanoparticles [87]. Figure 2.4 shows the size distribution of the EVs. The average size o f the EVs was found to be 100 nm. Figure 2.3. Filtering media solution through a filter with a 50 nm pore size filter using vacuum 20 2.2.2.2.4 ELISA Development for EVs from Breast Cancer Cells To begin, 100 ul of 20 ug/ml anti -CD63 or anti -CD81 in 50 mM Phosphate buffer pH 7.4 was added to each well in MAXisorp 96 -well plate (Nunc) and the microplate was incubated at 4 overnight. After being washed wi th PBS, 250 ul of 1% BSA was added to each well to block the sites, which have the potential for causing non -specific binding. Plate was incubated at room temperature for one hour. Wells were washed with PBS, 100 ul of the samples containing different conc entration of EVs was added to each well. Samples were prepared in 100 mM PBS pH 7.4. The plate was incubated at room temperature for eight hours on a microplate shaker for gentle mixing (Note: Because EVs are relatively larger and heavier particles, they c an precipitate when the sample is left stagnant). After being washed with PBS thoroughly (5 times), 100 ul of 20 ug/ml Figure 2.4. Size distribution of the EVs. NTA analysis showed an average diameter of 100 nm for the collected EVs from breast cancer cells 21 detection antibody -HRP in 2% BSA in 50 mM PBS buffer pH 7.4 was added to each well. The plate was incubated at room temperature for one h our. Then, wells were washed with PBS 5 times (Figure 2.5, Note: because EVs can be lysed by TWEEN 20, washing buffer was prepared without the detergent). Then, 100 ul of TMB -H2O2 was added to each well and enzymatic reaction between TMB -H2O2 and HRP happ ened for 10 minutes until a bright blue color was formed (Note: both TMB and the Figure 2.5. Molecular structure of the ELISA for EVs from breast cancer cells on the bottom of a well in a microplate . 22 products of the reaction catalyzed by HRP are light sensitive and this step must be done in a dark space). Then, 100 ul of 1 M sulfuric acid was added to each well to stop the reaction (yellow color formed). Finally, absorbance was read at 450 nM. 2.2.3 Color Formation Reaction Two different cosubstrates , TMB and catechol , were used for HRP. TMB is a commonly used substrate for HRP in optical ELISAs. Initially the optical ELISA was performed with TMB to optimize the experimental conditions. Then, an o ptical ELISA was performed with catechol as this substrate was going to be used in the EI. Oxidation of catechol with HRP in the presence of H 2O2 produces O -quinone which is an electroactive chemical. 2.2.3.1 TMB -H2O2 System In this case, 200 ul of TMB -H2O2 was ad ded to each well and enzymatic reaction between TMB -H2O2 and HRP happened for 10 minutes until a bright blue color was formed (Note: both TMB and the products of the reaction catalyzed by HRP are light sensitive and this step must be done in a dark space). Then, 100 ul of 1 M sulfuric acid was added to each well to lower pH and stop the reaction . Lowering pH induce formation of a relatively stable yellow colored product which can be measured at 450 nm [88]. Figure 2.6 shows the reaction the steps leading to formation of the colored pr oducts. 23 Figure 2.6. Reaction steps between TMB -H2O2 and HRP to form color ed products. 24 2.2.3.2 Catechol -H2O2 In the case of catechol -H2O2, 200 µL of a solution containing 5 mM catechol and 1.5 mM H2O2 was added to the wells and the plate was incubated in a dark space for 20 minutes. Then, 100 µM of 10 mM MBTH (prepared in deionized water) was added to the wells and absorbance was read at 505 nm. Catechol reacts with HRP in the presence of H 2O2 to produce o-quinones which react s with 3 -methyl -2 benzothiazolinone hydrazine (MBTH) to produce pink colored products (Figure 2.7) [89]. 25 Figure 2.7. Reaction steps between o -quinone and MBTH for form pink colored products. 26 2.3 Results and Discussion 2.3.1 ELISA for Mouse IgG One of the key parameters that had a significant effect on the performance of the ELISAs was dilution factor of the capture antibody and the detection antibody. Different dilutions of antibodies were used to design ELISAs (Figure 2.8). According to the results shown in Figure 2.8, a dilution of [1:333] gave the best sensitivity and this dilution was used in the design of EI. Figure 2.8. Optical ELISA for the mouse IgG. Mouse IgG was detected in a sandwich ELISA with a detection antibody labeled with HRP using TMB -H2O2 substrates. 27 An optical ELISA was also developed using catechol as the cosubstrate for the HRP as catechol was going to be used in the electrochemical immunosensor (Figure 2.9). Accor ding to this result , using catechol caused a relatively higher background signal compared to TMB -H2O2 system. A significant background signal was also observed in the EI using catechol -H2O2. Therefore, a set of control experiments were performed to investi gate the source of the high background current when using catechol as the cosubstrate (Figure 2.10). According to Figure 2.10, a significant background signal was observed for the case of catechol alone and the case of the catechol with H 2O2. These results can be attributed to the autoxidation of the catechol and the fact that a small amount of catechol might be oxidized in the presence of H 2O2. Because this background current was also observed in the case of the EI using catechol -H2O2 as the substrates, a design of experiment was performed in MINITAB (discussed in the next chapter) to optimize catechol and H 2O2 concentrations and thereby maximizing the signal to background ratio. 28 Figure 2.9. Optical ELISA for the mouse IgG. Mouse IgG was detected in a sandwich ELISA with a detection antibody labeled with HRP using catechol -H2O2 substrates. 29 Figure 2.10. Control experiments for measuring the background signal cause by non -specific binding, H2O2 and catechol. 2.3.2 ELISA for EVs from Breast Cancer Cells In this work, we were interested to develop an EI for detection of EVs from breast cancer cells. Before developing the biosensor, it was important to find a surface biomarker on the EVs that its concentration is high enough to detect EVs. In several studies, CD81 and CD63 were reported as two surface biomarkers on EVs from breast cancer cells that commercial antibodies wer e available for them. Optical ELISAs were developed to select the biomarker that gives a better sensitivity for detection EVs. In the first attempt, an ELISA was developed using antibodies against CD81 (Figure 2.11). Despite trying two sources of CD81 anti body and changing their concentrations 30 used in ELISA, the sensitivity did not improve. Therefore, antibodies against CD63 were used to develop the ELISA (Figure 2.12). Figure 2.12 shows that antibodies against CD63 significantly enhanced the dose response for detection of the EVs. These results suggest that antibodies against CD63 should be used in development of an EI. Figure 2.11. ELISA for EVs from breast cancer cells using CD81 as the surface biomarker. 31 2.4 Conclusions In this chapter, it was discussed how the conventional optical ELISA was used as a high throughput assay to optimize experimental conditions before developing EIs for an analyte of interest . Optical ELISAs was developed for mouse IgG to optimize antibody c oncentrations and to study the background signal observed in the EI . Optical ELISAs were also developed for detection of EVs from breast cancer cells. With the aid of optical ELISAs, a surface biomarker on EVs from breast cancer cells (CD63), was found. Th is surface biomarker provided a good sensitivity of detection for EVs from breast cancer cells. The results and finding from the optical ELISAs were crucial for development of the EI which is discussed in the next chapter. Figure 2.12. ELISA for EVs from breast cancer cells using CD63 as the surface biomarker. 32 3 Integrated Experimental and Theo retical Studies on an Electrochemical Immunosensor (EI) 3.1 Introduction Electrochemical biosensors are analytical devices that detect analytes by transforming a biochemical reaction into a quantita tive, electrical signal. This class of biosensors has proven valuable in research, quality control, food safety, medical diagnosis, and monitoring of therapeutic efficacy [25]. Miniaturized amperometric biosensors that use redox enzymes to generate a n electric current in response to voltage applied at a working electrode have been successfully commercialized; personalized blood glucose meters used by diabetics represented 85% the total biosensor market i n 2008 [90]. By 2013, the wo rldwide market for glucose -monitoring biosensor systems was estimated to be billion s of dollars per year, with screen -printed -electrode (SPE) arrays that served as single -use biosensor fistripsfl representing two -thirds of that market [91]. The dis posable , redox -enzyme -based biosensor market is being further expanded by commercializing glucose -monitoring systems for animals [92]. Optical immunoassays based on the exceptionally high binding selectivity and affinity of biological recognition molecules (predominantly antibodies, but also aptamers [93]) have been commercialized for applications in many fields, including environmental protection, food safety, and healthcare. The projected global market for lateral -flow immunoassays has risen at a compound annual growth rate of 8.1% since 2017 and is expected to reach $8 billion in 2022 [94]. Immunoassays typically involve a fisandwichfl mol ecular architecture, in which immobilized capture antibodies first bind target -analyte molecules present in the liquid sample . Then secondary antibodies labeled with reporter molecule s that generate an optical signal also bind the analyte molecules . The result ing molecular fisandwichesfl consist of an analyte molecule held between 33 primary - and secondary -antibody molecules . To estimate the analyte concentrati on, the surface concentration of bound reporter molecules is measured by the intensity of the optical signal they generate . A calibration curve is used to convert the reporter molecule™s concentration into the analyte concentration [30]. Commonly used reporter molecules for immunoassays include redox enzymes whose products can be measured optically, such a s horseradish peroxidase ( HRP ). HRP offers multiple advantages as a reporter . It is robust, has a relatively small molecular size, is inexpensive, is readily bound to antibodies in an active form, has a high turnover rate, and can oxidize a wide range of substrates to yield optically active products [31, 32]. Whereas virtually al l commercial immunoassay systems involve optical detection, the benefits of integrating electrochemical biosensors and immunoassays have been recognized [10]. Such hybrid electrochemical immunosensors (EI) have the potential to combine th e advantages of immunoassays (extremely high sensitivity and selectivity) with those of electrochemical biosensors (reproducible, quantitative, continuous electrical output). The electrical output is achieved by forming a sandwich molecular architecture on the working electrode , and the reporter molecule triggers an electrical signal. Redox enzymes are commonly used as EI reporters because some of their reaction products can be either oxidized or reduced at the working electrode , resulting in an electric cu rrent that serves as the EI™s output. This approach offers exceptional versatility because an EI biosensor could be developed for virtually any analyte for which antibodies can be developed. Also, inexpensive, disposable, SPE arrays designed to be read by portable meters similar to glucose meters EI could be mass -produced. The resulting EI platform would enable an extremely wide range of molecular and cellular analytes to be accurately measured with high sensitivity and selectivity , ease of use, low cost, and portability [26-29, 95]. 34 Prototype EI systems have been developed for heal thcare applications. Sanchez -Tirado et al. fabricated an EI to measure cytokines used as markers of inflammation [96]. Tallapragada et al. developed an EI for human epidermal growth factor receptor 2 (HER2) that had a detection limit of 4 ng/mL [97]. Dempsey et al. described a disposable, printed lateral flow EI for human cardiac troponin T (cTnT) [98]. The reporter used in all of these studies, HRP , gen erated an oxidized product that was electrochemically reduced at the working electrode , resulting in a continuous amperometric output . However, commercial implementation of EI systems has been hampered by the complexity of the multiple molecular mass -trans fer, binding, and reaction steps that give rise to the electrical signal. This complexity complicates efforts to design new EIs that achieve specified performance metrics, including the lower detection limit and sensitivity (defined as the change in output per unit change in analyte concentration). Fabrication methods and operating conditions needed to achieve these metrics are expected to vary between EI systems due to factors including analyte -antibody binding affinities, the concentrations of primary ant ibodies bound to the electrode, and the kinetics of both the reporter enzyme™s reaction and the electrochemical reaction. These kinetics will, in turn, be influenced by the liquid sample™s properties, including its pH and its concentrations of the analyte, and substrates for the enzymatic reaction. Moreover, the concentrations of redox -active interferents in the sample may limit the working electrode™ s voltage. The d evelopment of robust product -design algorithms for new EI systems that meet specified performance metrics would be aided by m echanistic mathematical models that quantitatively describe the rates of the key molecular mass -transfer, binding, and reaction steps. Such models would enable the step(s) that limits performance to be ident ified and guide strategies to overcome such limitation(s). To date, few mechanistic models of HRP -based EIs have been reported [33- 35 37], and these models have not been sufficiently comprehensive to predict how the output would vary with key independent variables, including the working electrode™ s applied voltage ( E), the pH, and the concentrations of HRP ™s substrates . Such models are needed to help design EIs, identify factors that limit their performance properties, and guide research strategies to optimize EI systems. Mechanistic models would a lso help support petitions for U.S. Food and Drug Administration (FDA) approval of EI systems for healthcare applications. The FDA requires that stringent accuracy and consistency standards be met by portable glucose monitoring systems while in the hands o f lay users [99], and similar requirements would be expected for EI s. Mechanistic models would enable rapid, in-silic o hypothesis testing, including fiwhat -iffl studies to assess whether non -standard use by lay users c ould result in dangerously incorrect readings. This chapter addresses the need for such mechanistic models by presenting a novel, integrated experimental and mathematical framework to characterize EI performance and then using the framework to optimize performance of a novel EI that can detect a target protein (mou se IgG) at the ng/ml level. The framework includes three components. The first is a detailed mechanistic model that can predict the rates of the individual mass -transfer and reaction steps that give rise to the EI™s amperometric output . The second is a sta tistical -design -of-experiments approach that generates an empirical, statistical model describing the effects o f key independent variables on the EI™s output. This statistical model is used both to optimize the EI system and to help validate the mechanisti c model . The third is an integration of dimensional analysis with principles of flux -control theory to quantify the extent to which individual mass -transfer and reaction steps limit the EI™s sensitivity and output current (J). The chapter concludes by disc ussing the utility of the 36 integrated experimental and mathematical framework for future design, optimization, and validation of EI system s. 3.2 Experimental Methods 3.2.1 Materials and Instrumentation Thioctic acid, sodium phosphate (monobasic and dibasic), mouse I gG, anti -mouse IgG antibody (ap124), HRP -conjugated -goat anti -mouse IgG (a5278), TWEEN 20, H2O2), C, and N -hydroxysulfosuccinimide sodium salt (NHS) were obtained from Sigma Aldrich. MES buffered saline packs , and 1 -ethyl -3-(3-dimethylaminopropyl carbodiimide hydrochloride) (EDC) were Nanopure -UV four -stage purifier (Barnstead International, Dubuque, IA); the purifier was solutions. Screen -printed electrodes were obtained from Conductive Technologies Inc. and Metrohm DropSens (models DRP -250BT and DRP -110SWCNT). 3.2.2 Preparation of Immunosensing Layer The immunosensing layer was prepared by using 1-ethyl -3-(3-dimethylaminopropyl (EDC) and N-hydroxysulfosuccinimide sodium salt (NHS) chemistry to attach the primary ( capture) antibodies covalently to carboxylate groups present on the DropSens array™s working electrodes. EDC -NHS chemistry has been widely used to fabricate the immunosensing layers of EIs [38-40]. EDC is a zero -length cross -linker that activates carboxylate groups for covalent coupling to primary amines. The addition of NHS with EDC results in an NHS ester intermediate that reacts rapidly with primary amine s, thereby increasing the efficiency of the coupling reaction [41]. Cleaned gold SPEs were dipped in 15 mM thioctic acid in ethanol for 1 h. The resulting carboxylated SPEs were washed with ethanol and dried under nitrogen. The carboxyl groups were 37 activated by incubating the SPEs in 100 mM MES buffer containing 5.0 mM EDC and 9.0 mM NHS at pH 4.6 for 1 h at room temperature. Electrodes were then rinsed with MES buffer and dipped in 6 µg/mL goat anti -mouse IgG antibody in 50 mM phosphate buffer at pH 7 for 2 h. The primary -antibody -functionalized SPEs were then washed with phosphate buffer. To block nonspecific binding of the target analyte (mouse IgG), the SPEs were incubated in 2 % BSA in phosphate buffer for 1 h at room temperature. The resulting f unctional SPEs were washed with phosphate buffer at pH 7 and stored in phosphate buffer at 4 oC. SPEs were each dipped in a standard solution having a known concentration of the target analyte (mouse IgG) in a 2% aqueous bovine serum albumin ( BSA ) solution in 50 mM phosphate buffer at pH 7 for 1 h at room temperature. The SPEs were then washed four times with washing buffer (0.05% TWEEN20 in 50 mM phosphate buffer at pH 7) and incubated in a [1:333] dilution of HRP -conjugated -goat anti -mouse IgG in pH 7, 50 mM phosphate buffer in 2% BSA (Figure 3.1). After 1 h, the electrodes were rinsed four times with washing buffer and stored in phosphate buffer at 4 oC until the electrochemical measurements were conducted. 38 Figure 3.1. Schematic diagram of immunosensing layer showing molecular sandwiches containing the capture antibody, the target analyte, and the HRP -tagged secondary antibody bound to the EI™s gold working electrode . 3.2.3 Electrochemical Measurement of EI Signal The EIs were removed from the refrigerator and allowed to equilibrate at room temperature. Forty µL of a solution (subsequently referred to as the fibulk solutionfl) containing 50 mM phosphate buffer, 1 mM , and 8 mM C were added to the SPE. Wire leads from a potentiometer (CHI 660, C.H. Instruments, USA) were connected to the EI™s working, reference, and auxiliary electrodes , and reduction potential of -0.2 V relative to a n Ag/AgCl reference electrode w ere applied to the working electrode. After about 1 min, t he reduction current (i.e. the EI™s signal (J)) reached a steady -state value, and the current level was recorded as the EI™s output for that set of experimental conditions. Each EI was used once. All electrochemical potentials given in this work are relati ve to a n Ag/AgCl reference electrode. 39 3.3 Optimization of EI Operating Conditions and Characterization of EI Performance Properties A statistical design of experiment (DOE) approach was used for two purposes: (1) to determine the values of key independent variables that optimized the EI™s signal and (2) to obtain an empirical equation that described the effects of the key independent variables on the EI™s signal to help validate the mechanistic model . The independent variables expe cted to most strongly affect the performance of the EI described above included (1) the working electrode™s E, (2) the bulk solution™s [C], (3) the bulk solution™s [H2O2], and (4) the bulk solution ™s pH [42]. A two -level half factorial design with center points and three replicates for each experiment was set up using Minitab ® software (Table 3.1) . For each factor, the following three levels, denoted low (-1), center point (0) , and high (+1), were chosen : -0.05 V, -0.125 V, an d -0.2 V for E; 1.0 mM, 4.5 mM, and 9.0 mM for [ C]; 0.5 mM, 1mM, and 1.5 mM for []; and 6.2, 6.6, and 7.0 for pH, respectively. To avoid electrical noise arising from the reduction of redox -active interferents in the bulk solution [100], the lowest E value was set to -0.2 V. T o control the rate of C autoxidation [101], 8 mM was selected as the highest [C] value. Experiments were conducted in triplicate for each combination of factors specified by Minitab ® using a constant analyte concentration of 40 ng/mL mouse IgG. Each EI™s signal was calculated as the difference between the J measured first in the absence of analyte and then in the presence of the analyte. All signal data were input to Minitab ®, which provided a statistical analysis of the results. The experimental conditions that Minitab ® indicated were optimal for the EI were used in subsequent experiments to characterize the EI™s performance properties. In these experiments, the EI signal was measured in triplicate for six concentrations of the analyte. 40 Table 3.1. Design of Experiments in coded units suggested by MINITAB using half factorial design. For each factor, three levels, denoted low ( -1), center point (0), and high (+1), were selected: -0.05 V, -0.125 V, and -0.2 V for E; 1.0 mM, 4.5 mM, and 9.0 mM for [C]; 0.5 mM, 1mM, and 1.5 mM for [HO]; and 6.2, 6.6, and 7.0 for pH, respectively. Run Order E pH C H2O2 Run Order E pH C H2O2 1 +1 -1 -1 +1 19 0 0 0 0 2 -1 +1 +1 -1 20 -1 -1 -1 -1 3 0 0 0 0 21 +1 +1 +1 +1 4 0 0 0 0 22 0 0 0 0 5 +1 -1 +1 -1 23 -1 +1 -1 +1 6 -1 +1 -1 +1 24 +1 -1 +1 -1 7 -1 -1 +1 +1 25 -1 +1 +1 -1 8 +1 +1 -1 -1 26 +1 -1 -1 +1 9 0 0 0 0 27 0 0 0 0 10 -1 -1 +1 +1 28 0 0 0 0 11 0 0 0 0 29 -1 +1 +1 -1 12 +1 +1 -1 -1 30 +1 -1 -1 +1 13 0 0 0 0 31 -1 +1 -1 +1 14 +1 +1 +1 +1 32 0 0 0 0 15 -1 -1 -1 -1 33 +1 -1 +1 -1 16 +1 +1 +1 +1 34 +1 +1 -1 -1 17 0 0 0 0 35 0 0 0 0 18 -1 -1 -1 -1 36 -1 -1 +1 +1 3.4 Mechanistic Mathematical Model The mechanistic mathematical model of the EI describes the transport and reaction processes involving catechol ( ), O-quinone (), and hydrogen peroxide () that generate a current (J) at the EI™s working electrode. Differential mass -balance equations describe the diffusion of these species in the x -direction (perpendicular to the electrod e) through two layers (Figure 3.2) that lie between the electrode™s surface at x=0 and the bulk solution : (1) the immunosensing layer between x=0 and x=L containing the antibodies and HRP , and (2) a stagnant , aqueous , diffusion layer between x=L and x=L+ . The HRP -catalyzed conversion of and to is a ssumed to occur uniformly throughout the immunosensing layer, and the electrochemical reducti on of Q to C is assumed to occur on the electrode™s surface. The bulk solution is assumed to be well -mixed, with the concentrations of all chemical species remaining constant at their initial values [102]. Mass 41 transfer is assumed to follow Fick™s law, with a diffusion coefficient ( D) that is assumed to be the same for , , and but to vary between the diffusion layer ( ) and the immunosensing layer ( ). The HRP concentration and maximum reaction rate constant () are assumed to be uniform throughout the immunosensing layer [103]. 3.4.1 Kinetics of Enzymatic and Electrochemical Reactions The non -linear, ping -pong kinetic mechanism describing HRP oxidation of C in the presence of is shown in reactions A Œ C [44-46, 103] : Figure 3.2. Schematic representation of diffusional mass -transfer, enzyme catalysis and electrochemical reaction steps happening on the biosensor interface. 42 HRP (Fe 3+ )+H 2O2 Compound(I) + H 2O (A) Compound (I) + Compound (II) + Q (B) Compound (II ) + HRP (Fe 3+ ) + Q (C) where compounds (I) and (II) are oxidized intermediates of HRP . The kinetic formula resulting from this mechanism [35, 104-107] is: =[][][]+[]+[][] (1) where is the reaction rate, is the maximum reaction rate constant ( = [ ]), and [ HRP ] are turnover number and HRP concentration within the immunosensing layer , respectively; and are the corresponding Michaelis -Menten constants, and [] and [] are and C concentrations, respectively. Molecules of Q produced by HRP can be reduced back to C at the surface of the working electrode in a two -electron, two proton reaction s hown in reaction (D) at a rate described by the Butler -Volmer equation (Eq 2) [108]: Q + 2e - + 2H + C (D) == [] ()[] ()() (2) where is the electric current density, is the number of transferred electrons ( n=2 for this reaction ), is the charge transfer coefficient (assumed 0.4), is the Faraday constant (96,485 C mol -1), is the apparent electron transfer rate constant for Q, is the universal gas constant (8.314 J K-1 mol -1), is the absolute temperature (298 K) , and is the redox potential for 43 electrochemical reduction of Q to C under the experimental conditions used (0.15 V at pH 6.2) . Values of for a given set of experimental conditions were determined as the midpoint potential ( ) between the cathodic peak (for Q reduction) and anodic peak (for C oxidation) of cyclic voltammogram s obtained under the same conditions [109]. The calculated value of J was taken to be the current generated by the EI. The effect pH o n is shown in Eq 3 [110, 111], in which m (=2) and n (=2) is the number of protons and electrons involved in the reduction of Q, respectively. This equation indicates that increasing the pH would make more negative and thereby reduce the working electrode™s overpotential, reaction rate, and EI ™s signal, according to the Butler -Volmer equation. To simulate the effect of pH on Eq 3 was incorporated in to the mechanistic model. ~ 2.303 pH (3) 3.4.2 Mass Balance Equations Assuming one -dimensional diffusion in the x-direction, the steady -state , differential, mass balance equations including diffusion and enzymatic reaction for , , and Q across the immunosensing layer (0 [ C] (SE=8.9) > pH (SE=4.6) > [ ] (SE=2.1). Figure 3.4. Pareto chart showing the standardized effect (SE) of factors E, [C], pH, and [H_2 O_2] on the biosensor signal. The terms with an SE value greater than the threshold value marked with the dotted line (SE=2.09) exerted a statistically significant effect on biosensor signal at the 95% confidence level. The strong increase in the EI™s signal with E, and thus the magnitude of (E-Eh), is apparent in both the experimental results and the model™s predictions (Figure 3.5). This trend is attributed to the 50 Butler -Volmer equation™s (Eq 2) exponential dependency of the EI™s amperometric signal on (E-Eh). Figure 3.5. Effect of working electrode overpotential (E -Eh) on the steady -state EI™s signal. [C]=8mM, [H_2 O_2] =1 mM, pH=6.2, [ HRP ]=0.5µM. The effects of the two HRP substrate concentrations , [C] and [], predicted by the model , are also similar to those observed experimentally (Figures 3.6 and 3.7, respectively). The increase in signal with an increase in each substrate™s concentration is consistent with the ping -pong kinetic model (Eq 1), which predicts that HRP ™s reaction rate would increase as either [C] or [] increase s. However, the SE for [C] is considerably stronger (SE=8.9) than that for [ ] (SE=2.1), possibly because u sed the [] used in the experiments was much greater than the value for HRP . 51 Figure 3.6. Effect of [C] on the steady -state EI™s signal: comparison of model prediction and experimental data. [H 2O2]=1 mM, pH=6.2, [HRP]=0.5µM, E -Eh = -0.35 V. 52 Figure 3.7. Effect of [H 2O2] on the steady -state EI™s signal: comparison of model prediction and experimental data. [C]=8mM, pH=6.2, [HRP]=0.5µM, E -Eh = -0.35V. Both the experimental results and the mechanistic model (Figure 3.8) indicated a slightly higher EI™s signal in a mildly acidic bulk solution (pH = 6.2 or 6.6) than at a neutral one (pH =7) . This trend is consistent with published reports that HRP oxidize d substrates more rapidly in the slightly acidic buffer than in neutral buffer [124]. One explanation for this effect is that pH (i.e., proton concentration) affects the thermodynamic driving force for the two -electron, two -proton electrochemical reduction of Q to C at the electrode. The value used in the model was measured as the midpoint potential ( ) of cyclic voltammograms of an aqueous solution containing C and Q. Eq 3 shows that increasing pH would make more negative, which would reduce the magnitude of () and thereby reduce the EI™s signal [110, 111]. 53 Figure 3.8. Simulation of pH effect on the steady -state EI™s signal. [C] = 8mM, [H 2O2] = 1 mM, [ HRP ] = 0.5µM, E -Eh = -0.35 V. 3.5.3 Integration of Dimensional Analysis and Flux Analysis to Determine Rate -Limiting Step Previous mathematical models developed to describe the kinetics of HRP on the electrodes [35, 104-107] focused on the enzyme™s kinetics or were based on the assumption that the J is mass -transfer limited. In contrast, our model explicitly calculates the rates of all key reaction s and mass transfer steps , all of which could limit the signal™s magnitude to some extent . Additionally, incorporation of Eqs 2 and 3 allows effects of (E-Eh) and pH, respectively, to be predicted, even under conditions in which the commonly used assumption that [] = 0 is invalid. Figure 3.9A shows that [] decreases as the magnitude of (E-Eh) and the reduc tion rate of [Q] increases. 54 To demonstrate the improvement in accuracy this extension of the model provides, we calculated the error that would result from assuming [] =0 (Figure 3.9) for an (E-Eh) range between -0.2 and -0.35V. The predicted error would have been about 15% for the ( E-Eh) value of -0.3 V used by Kohli et al. [115, 125] (Figure 3.9B). The smaller the absolute value of ( E-Eh), the greater the improvement in accuracy our extended model would provide . 55 The performance properties of an EI are controlled by the dynamics of the underlying transport and reaction steps that give rise to its J. We developed a mathematical framework that leverages dimensional analysis and the mechanistic model™s ability to predict the rates of the underlying steps to quantitatively assess the degree to which individual steps control the magnitude of the EI™s Figure 3.9. A: Simulated [Q ](x=0) over a range of (E -Eh) values B: Error percentage caused by assuming [Q ](x=0) =0 as a function of (E -Eh). Error percentage = [(J assuming [Q](x=0) =0 - J using calculated value of [Q](x=0) ) / J usi ng calculated value of [Q]] *100. [C]=8mM, [HO] =1 mM, pH=6.2, [HRP ]=0.5µM. 56 signal and its sensi tivity (defined as the change in J per unit change in analyte concentration). Examples of the approach are described below. The dimensionless Damkohler number ( ) shown in Eq 17 expresses the ratio of th e relative rates of enzymatic reaction ( ) and diffusional mass transfer () of HRP ™s substrates within the immunosensing layer [126]. Plugging constants from Table 3.2 into Eq 17 revealed that for C and were on the order of 10 -5, indicating that the diffusion could provide C and to the HRP orders of magnitude faster than the HRP could consume it [127, 128]. This result indicates that the EI™s signal is not significantly limited by the diffusion rate within the immunosensing layer. = (17) Flux -control analysis has been used t o determine the extent to which the rates of individual enzymatic reactions in a biochemical reaction pathway limit the overall mass flux through that pathway [129]. We used a similar approach to determine the relative degrees to which the enzymatic and electrochemical reaction steps limit the magnitude of EI™s signa l. We defined a current -control coefficient ( ) for each reaction step (V i) as the ratio of the percent change in the EI™s signal to the percent change in V i while holding all other independent variables constant (Eq 18). We used the mechanistic model to calculate an incremental c hange in J J) resulting from an incremental ) in either the enzymatic reaction rate (simulated by changing the [ HRP ] value) or the electrochemical reaction rate (simulated by changing the (E-Eh) value) . The incremental changes 57 ) were then used in place of the differentials ( and ) in Eq 18 to calculate the values for both the enzymatic r eaction and the electrochemical reaction across the range of ( E-Eh) values used in this study . = (18) The values calculated by making incremental changes in [ HRP ] remained virtually 1.0 across the entire range of ( E-Eh), for the [ HRP from 0.005 µM to 50 µM (results not shown). This resul t indicates that the EI™s signal is strongly limited by [ HRP ] over the entire range simulat ed. Consequently, the EI™s signal has the potential to be linearly correlated with the target analyte™s concentration, depending on the shape of the adsorption isotherm of the immobilized primary antibody for its target analyte. In contrast, the values for the electrochemical reaction varied significantly across the ran ge of overpotential used in this study (Figure 3.10) and exhibited a peak at about 3.3 at an (E-Eh) value of about -0.26V. Although the predicted EI™s signal curve increased monotonically as the magnitude of (E-Eh) increased, the curve exhibited an inflection point at about the same (E-Eh) value the curve peaked. This observation suggests that a transition occurs at this point. For lower (E-Eh) magnit udes, increasing the magnitude strongly increases the EI™s signal; however, for higher (E-Eh) magnitudes, further increases in the (E-Eh) magnitude offer diminishing returns, suggesting that the peak in may mark an optimal operating overpotential in the absence of other overriding considerations, such as the presence of electrochemical interferents. For significantly higher (E- Eh) magnitudes, the J asymptotically approaches a maximum value and the value approaches 0 . 58 Figure 3.10. Predicted current density (J) and current -control coefficients for the electrochemical reaction at different E values. [C]=8mM, [HO] =1 mM, pH=6.2, [HRP] =0.5µM. Because [ HRP ] would be expected to increase with the analyte concentration, the mechanistic model was also used to calculate the EI™s sensitivity ( S) to [HRP ] (defined in Eq 19 ) as well as sensitivity -control coefficients ( ) (defined in Eq 20 ). [HRP ] (19) = (20) The S and values were calculated in a manner similar to that used to calculate values. The model was used to calculate i ncremental HRP ] values. The incremental change values were substituted for differentials in Eqs 19 and 20. The resulting S 59 values and values (Figure 3.11) have shapes similar to the J and curves, respectively, shown in Figure 3.10. However, the peak in the curve occurs at a slightly different (E-Eh) value ( -0.23V) than the peak in the curve ( -0.26V). If an EI were operated near the peak of the curve, the sensitivity could be adjusted simply by making a relatively small change in the (E-Eh) value. Higher sensitivities would be desirable for ac curately measuring analyte concentr ations over a relatively small concentration range, whereas l ower sensitivities would be desirable for measuring ana lyte concentrations over a wide range. Figure 3.11. Sensitivity -control coefficient and sensitivity vs . E-Eh. [C]=8mM, [HO] =1.0mM, pH=6.2, [HRP ]=0.5µM. 3.6 Conclusions This study demonstrated the use of a novel, integrated experimental and modeling framework to analyze and optimize the performance of EIs. Th e experimental component included (1) deposition 60 of an EI interface on the working electrode of miniature SPE arrays ; (2) measurement of the performance properties of the resulting EIs for measuring the concentration of a surrogate protein antigen (mouse I gG); (3) use of a response -surface, statistical -design -of-experiments approach to optimize four independent variables: electrode overpotential, pH, and the concentrations of HRP ™s two substrates ( [C] and []); and (4) development of a statistical mode l of the experimental data that empirically describes the effect of the four independent variables on the EI™s signal. The modeling component included (1) development of a detailed, mechanistic model of the EI interface that described the rates of the mass -transfer and reaction steps that gave rise to the EI™s signal; (2) use of the statistical model of the experimental data to help validate the mechanistic model; and (3) integration of dimensional analysis, principles of flux -control analysis, and the mechanistic model™s predictive capabilities to obtain unprecedented insight into which steps control the magnitude of the EI™s signal and its sensitivity to the target analyte. The EI developed in this study had a limit of detection of 1 ng/mL and an inter -assay/intra -assay variation of less than 5%. The mechanistic model reproduced experimentally observed effects of the four independent variables on the EI™s signal. Calculation of Damko hler numbers indicated that diffusion of HRP ™s substrates in the biocatalytic layer did not limit the EI™s performance at the overpotential of -0.3 V . Calculation of current -control and sensitivity -control coefficients analyses provided new insight into to which the enzymatic and electrochemical reactions limited both the EI™s signal and its sensitivity over the experimentally relevant range of ( E-Eh) values. The novel, integrated experimental and modeling framework presented in this study provides unprecedented capabilities to design, optimize, and validate EIs for diverse applications. Its ability to quickly identify key mass transfer or reaction step(s) that limit(s) could guide strategies to overcome such limitation(s) and thereby reduce the time required to develop new commercial EI 61 systems. Also, the predictive power of the mechanistic model could, in principle, enable EIs to be designed a priori to meet specifications and enable rapid, in-silic o hypothesis testing that c ould accelerate FDA approval of EI systems for healthcare applications . 62 APPENDI X 63 MATLAB Codes %%%% This function returns the effect of overpotential on EI's signal%%%% function overpotential H=1e -6; %H2O2 bulk concentration in mol/cm3 C=8e -6; %Catechol bulk concentration in mol/cm3 KHm=2e -7; %Km value of HRP for H2O2 in mol/cm3 KCm=3e -6; %Km value of HRP for catechol in mol/cm3 Kcat=22000; %turnover number of HRP for catechol and H2O2 in 1/sec E=5e -10; %concentration of HRP in mol/cm3 L= 2.2e -6; %thickness of enzyme layer in cm Df= 2.28e -6; %cm2/s %diffusion coefficient in enzyme layer De= 2.2e -5; %cm2/s %diffusion coefficient in boundary layer Kp= 1; %partition coefficient del = 3e -3; %thickne ss of boundary layer in cm Ka=0.1e -6; %apparent electron transfer rate in cm/s R=8.314; %universal gas constant (8.314 J K -1 mol -1 T=298; %temperature in K area= 0.118; %area of the working electrode in cm2 electron= 2; %number of electron transferred in reduction of quinone F= 96485 %Faraday constant 96,485 C mol -1 x = linspace(0,L,100); function dydx = ode3(x,y) %this function returns all of the odes, concentrations have been normalized by catechol bulk concentration dy1dx = [ y(2); (Kcat* E*y(1)*y(3)/(Df*(((KHm*KCm)/C)+KCm*y(3)+KHm*y(1)+(y(1)*y(3)*C))))]; dy2dx = [ y(4); (Kcat*E*y(1)*y(3)/(Df*(((KHm*KCm)/C)+KCm*y(3)+KHm*y(1)+(y(1)*y(3)*C))))]; dy3dx = [ y(6); -(Kcat*E*y(1)*y(3)/(Df*(((KHm*KCm)/C)+KCm*y(3)+KHm*y(1)+(y(1)*y(3)*C))))]; dydx =[dy1dx;dy2dx;dy3dx]; end function res = ode3bc(ya,yb) %this function returns the BCs res1 = [ya(4); Df*yb(2) -((De/(Kp*del))*(Kp -yb(1)))];% res2 = [ya(2)+ya(6); Df*yb(4) -((De/(Kp*del))*(Kp*H/C -yb(3)))];% res3 = [ya(6) -((((ya(5)*Ka*Exp) -(ya (3)*Ka*EXP))/(area*Df))); Df*yb(6)+((De/(Kp*del))*yb(5))]; res=[res1;res2;res3]; 64 end S=linspace( -0.05, -0.25,10);%E(applied voltage range) i=1; for i=1:length(S) V=S(i); Exp=exp(( -F*0.8*((V -0.15)))/(R*T));%corresponds to butler -volmer EXP=exp(( F*1.2*((V -0.15)))/(R*T));%corresponds to butler -volmer eexp(i)=Exp; initialsolution = bvpinit(x,[1,0.001,1, -0.5,0.01, -0.06]);%initial guess solution = bvp4c(@ode3,@ode3bc,initialsolution); y = deval(solution,x); D(i)= y(6,1); J(i)=2*96485*Df*D(i)*C*(10000000); %current density in nA/mm2 end figure (1) hold on plot(S,J); xlabel('Potential(V)'); ylabel('Current density(nA/mm2)'); end %%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%% 65 %%%% This function returns the effect of pH on EI's signal%%%% function hydrogenperoxide H=1e -6; %H2O2 bulk concentration in mol/cm3 C=8e -6; %Catechol bulk concentration in mol/cm3 KHm=2e -7; %Km value of HRP for H2O2 in mol/cm3 KCm=3e -6; %Km value of HRP for catechol in mol/cm3 Kcat=220 00; %turnover number of HRP for catechol and H2O2 in 1/sec E=5e -10; %concentration of HRP in mol/cm3 L= 2.2e -6; %thickness of enzyme layer in cm Df= 2.28e -6; %cm2/s %diffusion coefficient in enzyme layer De= 2.2e -5; %cm2/s %diffusion coefficient in boundar y layer Kp= 1; %partition coefficient del = 3e -3; %thickness of boundary layer in cm Ka=0.1e -6; %apparent electron transfer rate in cm/s R=8.314; %universal gas constant (8.314 J K -1 mol -1 T=298; %temperature in K area= 0.118; %area of the working electrod e in cm2 electron= 2; %number of electron transferred in reduction of quinone F= 96485 %Faraday constant 96,485 C mol -1 V=-0.2; %applied voltage Exp=exp(( -F*0.8*((V -0.15)))/(R*T));%corresponds to butler -volmer EXP=exp((F*1.2*((V -0.15)))/(R*T));%cor responds to butler -volmer x = linspace(0,L,100); function dydx = ode3(x,y) %this function returns all of the odes, concentrations have been normalized by catechol bulk concentration dy1dx = [ y(2); (Kcat*E*y(1)*y(3)/(Df*(((KHm*KCm)/C)+KCm*y(3)+KHm*y(1)+(y( 1)*y(3)*C))))]; dy2dx = [ y(4); (Kcat*E*y(1)*y(3)/(Df*(((KHm*KCm)/C)+KCm*y(3)+KHm*y(1)+(y(1)*y(3)*C))))]; dy3dx = [ y(6); -(Kcat*E*y(1)*y(3)/(Df*(((KHm*KCm)/C)+KCm*y(3)+KHm*y(1)+(y(1)*y(3)*C))))]; dydx=[dy1dx;dy2dx;dy3dx]; end function res = ode3bc(ya,yb) %this function returns BCs res1 = [ya(4); Df*yb(2) -((De/(Kp*del))*(Kp -yb(1)))];% res2 = [ya(2)+ya(6); Df*yb(4) -((De/(Kp*del))*(Kp*H/C -yb(3)))];% res3 = [ya(6) -((((ya(5)*Ka*Exp) -(ya(3)*Ka*EXP))/(area*Df))); Df*yb(6)+((De/(Kp*del))*yb(5))]; 66 res=[res1;res2;res3]; end S=linspace(0.5e -6,1.5e -6,10); %H2O2 range i=1; for i=1:length(S) H=S(i); initialsolution = bvpinit(x,[1,0.001,1,0.05,0.001,0.06]); solution = bvp4c(@ode3,@ode3bc,initialsolution); y = deval(solution,x); D(i)= y(6,1); J(i)=2*96 485*Df*D(i)*C*(10000000); %current density end figure (1) hold on plot(S,J); xlabel('H2O2(mM)'); ylabel('Currentdensity(nA/mm2)'); end %%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%% 67 %%%% This function returns the effect of catechol concentration on the EI's signal%%%% function catechol H=1e -6; %H2O2 bulk concentration in mol/cm3 C=8e -6; %Catechol bulk concentration in mol/cm3 KHm=2e -7; %Km value of HRP for H2O2 in mol/cm3 KCm=3e -6; %Km value of HRP for catechol in mol/cm3 Kcat=22000; %turnover number of HRP for catechol and H2O2 in 1/sec E=5e -10; %concentration of HRP in mol/cm3 L= 2.2e -6; %thickness of enzyme layer in cm Df= 2.28e -6; %cm2/s %diffusion coefficient in enzyme layer De= 2.2e -5; %cm2/s %diffusion coefficient in boundary l ayer Kp= 1; %partition coefficient del = 3e -3; %thickness of boundary layer in cm Ka=0.1e -6; % apparent electron transfer rate in cm/s R=8.314; %universal gas constant (8.314 J K -1 mol -1 T=298; %temperature in K area= 0.118; %area of the working electrode in cm2 electron= 2; %number of electron transferred in reduction of quinone F= 96485 %Faraday constant 96,485 C mol -1 V=-0.2; %applied voltage Exp=exp(( -F*0.8*((V -0.15)))/(R*T));%corresponds to butler -volmer EXP=exp((F*1.2*((V -0.15)))/(R* T));%corresponds to butler -volmer x = linspace(0,L,100); function dydx = ode3(x,y) %this function returns all of the odes, concentrations have been normalized by catechol bulk concentration dy1dx = [ y(2); (Kcat*E*y(1)*y(3)/(Df*(((KHm*KCm)/C)+KCm*y(3)+KHm* y(1)+(y(1)*y(3)*C))))]; dy2dx = [ y(4); (Kcat*E*y(1)*y(3)/(Df*(((KHm*KCm)/C)+KCm*y(3)+KHm*y(1)+(y(1)*y(3)*C))))]; dy3dx = [ y(6); -(Kcat*E*y(1)*y(3)/(Df*(((KHm*KCm)/C)+KCm*y(3)+KHm*y(1)+(y(1)*y(3)*C))))]; dydx=[dy1dx;dy2dx;dy3dx]; end function res = od e3bc(ya,yb)%this function returns all of the BCs res1 = [ya(4); Df*yb(2) -((De/(Kp*del))*(Kp -yb(1)))];% res2 = [ya(2)+ya(6); Df*yb(4) -((De/(Kp*del))*(Kp*H/C -yb(3)))];% res3 = [ya(6) -((((ya(5)*Ka*Exp) -(ya(3)*Ka*EXP))/(area*Df))); Df*yb(6)+((De/(Kp*del))*yb(5 ))]; 68 res=[res1;res2;res3]; end S=linspace(1e -6,8e -6,10); %range of catechol i=1; for i=1:length(S) C=S(i); initialsolution = bvpinit(x,[1,0.001,1,0.05,0.001,0.06]); solution = bvp4c(@ode3,@ode3bc,initialsolution); y = deval(solution,x); D(i)= y(6,1); J(i)=2*96485*Df*D(i)*C*(10000000); %Current density in nA/mm2 end hold on plot(S*1e6,J); xlabel('Catechol(mM)'); ylabel('Current density(nA/mm2)'); end %%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%% 69 %%%% This function returns the effect of pH on EI's signal%%%% function hydrogenperoxide H=1e -6; %H2O2 bulk concentration in mol/cm3 C=8e -6; %Catechol bulk concentration in mol/cm3 KHm=2e -7; %Km value of HRP for H2O2 in mol/cm3 KCm=3e -6; %Km value of HRP for catechol in mol/cm3 Kcat=220 00; %turnover number of HRP for catechol and H2O2 in 1/sec E=5e -10; %concentration of HRP in mol/cm3 L= 2.2e -6; %thickness of enzyme layer in cm Df= 2.28e -6; %cm2/s %diffusion coefficient in enzyme layer De= 2.2e -5; %cm2/s %diffusion coefficient in boundary layer Kp= 1; %partition coefficient del = 3e -3; %thickness of boundary layer in cm Ka=0.1e -6; %apperant electron transfer rate in cm/s R=8.314; %universal gas constant (8.314 J K -1 mol -1 T=298; %temperature in K area= 0.118; %area of the working e lectrode in cm2 electron= 2; %number of electron transferred in reduction of quinone F= 96485 %Faraday constant 96,485 C mol -1 V=-0.2; %applied voltage Exp=exp(( -F*0.8*((V -0.15)))/(R*T));%corresponds to butler -volmer EXP=exp((F*1.2*((V -0.15)))/(R*T ));%corresponds to butler -volmer x = linspace(0,L,100); function dydx = ode3(x,y) %this function returns all of the odes, concentrations have been normalized by catechol bulk concentration dy1dx = [ y(2); (Kcat*E*y(1)*y(3)/(Df*(((KHm*KCm)/C)+KCm*y(3)+KHm*y (1)+(y(1)*y(3)*C))))]; dy2dx = [ y(4); (Kcat*E*y(1)*y(3)/(Df*(((KHm*KCm)/C)+KCm*y(3)+KHm*y(1)+(y(1)*y(3)*C))))]; dy3dx = [ y(6); -(Kcat*E*y(1)*y(3)/(Df*(((KHm*KCm)/C)+KCm*y(3)+KHm*y(1)+(y(1)*y(3)*C))))]; dydx=[dy1dx;dy2dx;dy3dx]; end function res = ode3bc(ya,yb) %this function returns BCs res1 = [ya(4); Df*yb(2) -((De/(Kp*del))*(Kp -yb(1)))];% res2 = [ya(2)+ya(6); Df*yb(4) -((De/(Kp* del))*(Kp*H/C -yb(3)))];% res3 = [ya(6) -((((ya(5)*Ka*Exp) -(ya(3)*Ka*EXP))/(area*Df))); Df*yb(6)+((De/(Kp*del))*yb(5))]; 70 res=[res1;res2;res3]; end S=linspace(0.5e -6,1.5e -6,10); %H2O2 range i=1; for i=1:length(S) H=S(i); initialsolution = bvpinit(x,[1,0.0 01,1,0.05,0.001,0.06]); solution = bvp4c(@ode3,@ode3bc,initialsolution); y = deval(solution,x); D(i)= y(6,1); J(i)=2*96485*Df*D(i)*C*(10000000); %current density end figure (1) hold on plot(S*1e6,J); xlabel('H2O2(mM)'); ylabel('Current density(nA /mm2)'); end %%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%% 71 4 Theoretical and Experimental Studies of an Inhibition -based Bi-enzyme Electrochemical Biosensor (IBE) for Detection of Organophosphorus Compounds 4.1 Introduction Amperometric biosensors detect chemicals at a constant electrochemical potential by measuring oxidation or reduction current produced by electroactive products of a biochemical reaction [130]. Low cost, high sensitivity, relatively fast response time, simplicity of design, being compact, and having the potential for being miniaturized make an amperometric biosensor a great choice for detecting a w ide range of chemicals [118, 131]. Amperometric biosensors based on the principle of enzyme inhibition have been extensively developed for environmental applications [132]. These biosensors can be developed for analytes that can act as inhibitors for a specific enzyme. Such analytes can interact with an enzyme or enzyme -substrate complex and inhibit the bioca talytic properties. Therefore, they can be detected and measured indirectly by measuring the change that they cause in the biocatalytic activity of an immobilized enzyme. The s usceptibility of most enzymes to a very low concentration of inhibitors makes th ese type s of biosensors very sensitive [133 ]. Inhibition -based amperometric biosensors have been frequently developed for the detection of Organophosphorus co mpounds (OPs) [56, 134, 135 ]. OPs are synthetic compounds that are widely used in pesticides and chemical warfare [136]. OPs work based on inhibition of acetylcholinesterase (AChE) in the central nervous system of ins ects and human s. OPs can reside in the environment for several years and pose serious health issue for the non -target species such 72 as human and animals . Inhibition of AChE by OPs in non -target species can cause severe health issues and even death [137 ]. The gold standard method for detecting OPs is gas/liquid chromatography combined with mass spectroscopy [138]. Chromatographic methods are very sensitive, specific, and reliable , but they suffer from some major drawbacks. Some of these disadvantages are high cost, complicated and time-consuming process for sample preparation, requiring highly trained technicians, and not being applicable for on -site or in -field applications [138]. In contrast, amperometric biosensors based on the principle of enzyme inhibition provide a fast and sensitive detection without sample preparation. Besides, they have the potential for being developed as miniatur ized and portable analytical devices for on -site applications [139]. AChE is the most commonly used enzyme in the fabrication of biosensors based on principles of enzyme inhibition for OPs [140, 141]. Despite a large amount of interest in the area of inhibition based amperome tric biosensors for detection of OPs, there is a lack of a comprehensive theoretical study in this field. Zhang et al . developed a theoretical model for immobilized enzyme inhibition biosensors under the assumption that the inhibition process is diffusion limited. This model was valid for the concentrations of the OPs , which were very low compared to the amount of enzyme available [142]. Choi et al . developed a mathematical model for a fiber -optic biosensor for the detection of OPs. In their study, a mathematical model of enzyme kinetics for the inhibition of AChE and transport phenomena was developed to analyze the effect of operating parameters. Using the mathematical model, they optimized AChE concentration and su bstrate concentration [143].Their model was able t o simulate the optical signal under different experimental conditions. Herein, we present a comprehensive mathematical model to simulate and study the time -dependent electric current in an inhibition -based bi -enzyme electrochemical biosensor (IBE) . The IBE is a 73 modified version of our previously developed biosensor. Previously, a novel bi -enzyme electrochemical biosensor was developed in Dr. Worden™s research group to measure the activity of an esterase enzyme, neuropathy target esterase (NTE) [61]. The biosensor included an esterase enz yme (NTE) and an oxidase enzyme (tyrosinase) to generate a substrate recycling system that could amplify the biosensor™s signal. The biosensor was fabricated in a conventional electrochemical cell format. A mathematical model was also developed to study th e steady -state electric signal in the biosensor [61, 115, 144]. The IBE includes two enzymes, AChE (neurological esterase) and tyrosinase (oxidase). Phenyl methylsulfonyl fluoride (PMSF) was used as a model inhibitor of AChE . AChE hydrolyzes a reactant to yield a substrate that is repeatedly oxidized by the tyrosinase (Tyr) and then reduced by the elec trode. This substrate recycling not only generates a current that reports the AChE™s activity but also amplifies the electric current to increase the biosensor™s sensitivity dramaticall y. The IBE detects PMSF by quantitative measurement of AChE™s activity , and the amount of loss in AChE™s activity is related to PMSF concentrations. Therefore, to achieve the maximum sensitivity, it is necessary to adjust the ratio of the two enzymes™ activity in a way that AChE™s activity is controlling the electric current. Besides, AChE™s activity cannot exceed a specific range. It should remain low enough to allow low PMSF concentrations to have a significant effect on its activity and high enough to generate a measurable electric current. Several factors , including substr ate concentration, AChE™s activity, Tyr ™s activity, applied voltage, and diffusion rate , can influence the sensitivity of the IBE for detection of the PMSF. Therefore, we have developed an unsteady -state model to simulate the irreversible inhibition of ACh E with PMSF and to study to what extend each parameter affects the biosensor™s sensitivity. The model includes unsteady -state mass balance equations , including diffusion -based mass -transfer steps, enzymatic reactions , and kinetics of 74 irreversible enzyme in hibition. We report parameters such as current -control coefficients, sensitivity , and Damkohler number s to quantify the effect of each factor on limiting the electric current and the biosensor™s sensitivity. The model provides a platform to study the effec t of each factor on controlling the electric current and the biosensor™s sensitivity ; therefore, allow one to optimize the governing factors affecting the biosensor™s performance. 4.2 Experimental Methods 4.2.1 Material s and Instrumentation Sodium phosphate (monoba sic and dibasic), Acetylcholinesterase (C2888, from Electrophorus electricus ), Tyrosinase (T3824, from mushroom), Bovine serum albumin (BSA), Glutaric dialdehyde (50 wt. % solution in water, Phenylmethylsulfonyl fluoride (PMSF), and Phenyl acetate were obt -UV four -stage purifier (Barnstead International, Dubuque, IA); the purifier was equipped with a UV ous solutions. Screen -printed electrodes were obtained from Conductive Technologies Inc. and Metrohm DropSens (models DRP -250AT). 4.2.2 Enzyme Electrode Preparation SPEs were cleaned by sonication in pure ethanol for 2 minutes , followed by rinsing with ultrapure water. To prepare the enzyme solution, 40 µl of 50 mM phosphate buffer pH 7, 20 µl of 20 mg/mL Tyr in phosphate buffer, 20 µl of 1 mg/mL AChE in phosphate buffer, 10 µl of 2.7 mg/mL BSA in phosphate buffer and 10 µl of 4 wt. % glutaraldehyde in water were mixed together just before starting the preparation procedure. Three microliters of enzyme solution (in the case of DropSens SPEs) or one microliter of enzyme solution (in the case of CTI SPEs) were deposited on the work ing electrode , -enzyme 75 modified SPEs were rinsed with ultrapure water , and then they were stored at phosphate buffer at 4.2.3 PMSF Detection and Electrochemical Measurements To perform electrochemical experiments for the detection of PMSF, 30 µL of 50 mM phosphate buffer (pH 7) was added on the working electrode of SPEs. A potential of -200 mV relative to a n Ag/AgCl reference electrode was maintained on the working electrode usin g a potentiometer ( CHI 660, C.H. Instruments, USA) . An aliquot of phenyl acetate solution was added , and after reaching a stable electrochemical signal , a known amount of PMSF was added , and an electrochemical current was recorded after 30 seconds. 4.3 Mathema tical Model The biosensor includes a working electrode onto which a thin layer containing AChE and Tyr is bound, a diffusion layer, and the bulk liquid. To design the mathematical model, it was assumed that the geometry of the electrode is symmetrical and mass transfer of all species take s place in one dimension (x). The model developed to describe the unsteady -state amperometric response of the IBE consist ing of a set of differential mass balances for all reacting components (phenylacetate (S1), phenol (S2), catechol (S3), O-quinone (S4), and PMSF ( I) over the spatial regions depicted in Figure 4.1, the enzyme -containing layer, the diffusion layer, and the bul k solution. PMSF concentration is zero throughout the biosensor interface before addition to the system ( <). 76 The mass balance equations describe diffusion of phenylacetate ( S1), phenol ( S2), catechol ( S3), O-quinone ( S4), and PMSF ( I) in one dimension ( x) through two layers (Figure 4.1) that lie between the electrode™s surface at x=0 and the bulk solution: an enzyme -containing layer between x=0 and x=L containing the AChE ( E1) and Tyr (E2, E3), and a stagnant aqueous layer between x=L and . The enzymes concentration and their maximum reactions rates were assumed to be uniform a cross the enzyme -containing layer, and the electrochemical reduction of O -quinone ( S4) to catechol ( S3) is assumed to occur on the electrode™s surface. Figure 4.1. Schematic representation of reactions happening on the surface of gold working electrode. S1, S 2, S3, and S 4 denote phenyl acetate, phenol, catechol, o -quinone respectively. E 1, E2, and E 3 denote acetylcholinesterase, tyrosinase's phenolase activity, and tyrosinase™s catecholase activity. 77 The bulk solution was assumed to be well -mixed, with the concentrations of all chemical species remaining constant at their initial values [102, 115 ]. (Note: PMSF ( I) bulk concentration is zero before the addition time (t= )). 4.3.1 AChE Inactivation and Enzyme Kinetics AChE is a hydrolase enzyme found in the synapse between nerve cells and muscle cells. AChE stops the signal pathway between nerve cells and muscle cells by hydrolyzing a neurotransmitt er called acetylcholine [145]. In this work, phenylacetate ( S1) was used as the substrate for AChE. The AChE ( E1) hydrolyzes phenylacetate ( S1) to produce phen ol ( S2) and acetate (Figure 4.2). PMSF ( I) is an irreversible inhibitor of AChE that covalently binds to the active site of the enzyme and modifies AChE™s activity ( E1) [146]. The sulfonyl group of PMSF (Figure 4.3) mimics the carbonyl group of phenylacetate transition state. The hydroxyl group of serine residue of the active site of AChE nucleophilically attacks the sulfonyl group of PMSF, resulting in irreversible sulfonylation of AChE [147]. In this mod el, we assumed that the rate of PMSF ( I) consumption equals the rate of AChE inactivation. Figure 4.2. Hydrolysis of phenylacetate with AChE . 78 The general scheme for the inactivation of AChE with PMSF ( I) in the presence of the substrate ( S1) is shown in Figure 4.4. Studies have shown that AChE inhibition with PMSF follows a pseudo -first order kinetics [147] (Eq 1): ,,= (1) Where , and , are maximum velocities of the enzymatic reaction for AChE in the absence of the inhibitor and when incubated with inhibitor for a time of . The is the pseudo -first -order rate constant for the inactivation of AChE with PMSF (Eq 2): =[]1(1)+[] (2) The affinity of PMSF for AChE is given by the Michaelis -Menten type constant, . (Note: has also been denoted as Kd and Ka in other studies) [110, 147]. Figure 4.3. Molecular structure of PMSF . Figure 4.4. Inhibition mechanism of AChE (E) with PMSF (I) in the presence of substrate (S1). 79 =+ (3) and are the forward and backward rate constant s for the formation of the Michaelis -Menten type complex and is the sulfonylation rate constant (Figure 4.4). The is given by Eq 4 Where , is the Michaelis -Menten constant for phenylacetate. =[][]+, (4) PMSF ( I) competes with phenyl acetate ( S1) for the active site of AChE ( E1), therefore changing the Michaelis -Menten constant ( ,) to the apparent , (Eq 5) [143]. ,=,(1+[]) (5) Eqs 6 -8 explain the enzymatic kinetics of AChE in the presence of the irreversible inhibition with PMSF. , is the turnover number of AChE for phenylacetate. By assuming that the rate of PMSF ( I) consumption equals the rate of enzyme inactivation, Eq 9 was derived to explain the rate of PMSF ( I) consumption. =,[,+[ (6) ,=, () (7) , =, E1 (8) =() (9) Tyr contains two enzyme activities: monophenolase activity , which catalyzes the hydroxylation of monophenols to produce o -diphenols (catechols) , and catecholase activity , which catalyzes 80 oxidation of catechols to O -quinones. Figure 4.5 shows the scheme for the two -step oxidation of phenol with Tyr . Figure 4.5. Scheme of phenol oxidation with tyrosinase to produce O -quinone . Studies have shown that the hydroxylation step (monophenolase activity) takes place much slower than the oxidation step (catecholase activity) and therefore l imits the rate of O -quinone production [148]. Therefore, we assumed that rate ( ) of O -quinone ( S4) production from phenol ( S2) can be obtained from Eqs 11 -12 where E2 is correspo nded to phenolase activity of Tyr [114]. The rate () of conversion of catechol ( S3) to O -quinone ( S4) can be given by Eqs 12 -13. E3 denotes catecholase activity of Tyr [115]. = ,[],+[ (10) , =, E2 (11) =,[],+[ (12) , =, E3 (13) Molecules of O-quinone (S4) produced by Tyr can be reduced back to catechol (S3) at the surface of the working electrode at a rate described by the Butler -Volmer equation (Eq 14): 81 == [] ()[] ()() (14) where is the electric current density, is the number of transferred electrons (e.g. , n=2 for the electrochemical reduction of O-quinone (S4), is the charge transfer coefficient (assumed 0. 35), is the Faraday constant (96485 C mol -1), is the apparen t electron transfer rate constant for O-quinone (S4), is the universal gas constant (8.314 J K -1 mol -1), is the absolute temperature (assumed 298 K), and is the redox potential for electrochemical reduction of O-quinone (S4) to catechol (S3) under the experimental conditions (0.15 V). Values of for a given set of experimental conditions were determined as the midpoint between the cathodic peak and anodic peak of cyclic voltammogram obtained under the same conditions. 4.3.2 Mass Balance Equation s Assuming one -dimensional diffusion in the x-direction, the mass balance equations including diffusion and enzymatic reaction for phenyl acetate ( S1), phenol ( S2), catechol ( S3), O -quinone ( S4), and PMSF ( I) across the enzyme -containing layer (0= 40000) && (TimeStep <= 40001) %this elseif coomand is to simulate the sharp drop in currentcaused by mixing when adding PMSF C2_k = C2_old*0.8; C3_k = C3_old*0.8; C4_k = C4_old*0.8; I_old(M,1)=0.17; %inhibitor concentration KP = ((I_old*K12)./(I_old+KI*ones(M,1))); INH=exp(-KP*(TimeStep-40000)*(DeltaT/60)); Vmax1=k1*E1*INH; kminh=ones(M,1)+(1/KI)*I_old; else I_old(M,1)=0.17; KP = ((I_old*K12)./(I_old+KI*ones(M,1))); INH=exp(-KP*(TimeStep-40000)*(DeltaT/60)); Vmax1=k1*E1*INH; kminh=ones(M,1)+(1/KI)*I_old; end for Iter=1:IterMax for j=2:M-1 % to solve the problem iteratively, starting from an initial guess, % then iteratively converge to the new solution 105 C1_kp1(j,1) = (1/(1 + 2*kappa*DeltaT/Dx^2))*( C1_old(j,1) + kappa*DeltaT*( C1_k(j+1,1) + C1_k(j-1,1) )/Dx^2 + constant1*DeltaT*( (Vmax1(j,1)*C1_old(j,1)/(K1*kminh(j,1) + Si*C1_old(j,1)) ) )); C2_kp1(j,1) = (1/(1 + 2*kappa*DeltaT/Dx^2))*( C2_old(j,1) + kappa*DeltaT*( C2_k(j+1,1) + C2_k(j-1,1) )/Dx^2 + constant2*DeltaT*( ((Vmax1(j,1)*C1_old(j,1)/(K1*kminh(j,1) + Si*C1_old(j,1)))- (Vmax2*C2_old(j,1)/(K2 + Si*C2_old(j,1)))) ) ); C3_kp1(j,1) = (1/(1 + 2*kappa*DeltaT/Dx^2))*( C3_old(j,1) + kappa*DeltaT*( C3_k(j+1,1) + C3_k(j-1,1) )/Dx^2 + constant3*DeltaT*( (Vmax3*C3_old(j,1)/(K3 + Si*C3_old(j,1)) ) ) ); C4_kp1(j,1) = (1/(1 + 2*kappa*DeltaT/Dx^2))*( C4_old(j,1) + kappa*DeltaT*( C4_k(j+1,1) + C4_k(j-1,1) )/Dx^2 + constant4*DeltaT*( (Vmax2*C2_old(j,1)/(K2 + Si*C2_old(j,1)))+(Vmax3*C3_old(j,1)/(K3 + Si*C3_old(j,1)) ) ) ); I_kp1(j,1) = (1/(1 + 2*kappa*DeltaT/Dx^2))*( I_old(j,1) + kappa*DeltaT*( I_k(j+1,1) + I_k(j-1,1) )/Dx^2 + constant5*DeltaT*(-KP(j,1)*E1*INH(j,1))); end % Applying the BCs at each time-step j=1; % C4_kp1(j,1) = 0; I_kp1(j,1) = I_kp1(j+1,1); C1_kp1(j,1) = C1_kp1(j+1,1); C2_kp1(j,1) = C2_kp1(j+1,1); C3_kp1(j,1) = C3_kp1(j+1,1) + C4_kp1(j+1,1) ; C4_kp1(j,1) = (C4_kp1(j+1,1) + (Dx/(area*Df))*Ka*C3_kp1(j,1)*EXP)/(1+(Dx/(area*Df))*Ka*Exp); j=M; C4_kp1(j,1) = ( 1/(1+Dx*KAPPA) )*C4_kp1(j-1,1); C3_kp1(j,1) = ( 1/(1+Dx*KAPPA) )*C3_kp1(j-1,1); C2_kp1(j,1) = ( 1/(1+Dx*KAPPA) )*C2_kp1(j-1,1); C1_kp1(j,1) = (C1_kp1(j-1,1) + KAPPA*Dx )/(1+ Dx*KAPPA); I_kp1(j,1) = (I_kp1(j-1,1) + I_old(M,1)*KAPPA*Dx )/(1+ Dx*KAPPA); I_k = I_kp1; C1_k = C1_kp1; C2_k = C2_kp1; C3_k = C3_kp1; C4_k = C4_kp1; end for j=2:M-1 I_new(j,1)=I_kp1(j,1); C1_new(j,1)=C1_kp1(j,1); 106 C2_new(j,1)=C2_kp1(j,1); C3_new(j,1)=C3_kp1(j,1); C4_new(j,1)=C4_kp1(j,1); end % Applying the BCs at each time-step j=1; % C4_new(j,1) = 0; I_new(j,1) = I_new(j+1,1); C1_new(j,1) = C1_new(j+1,1); C2_new(j,1) = C2_new(j+1,1); C3_new(j,1) = C3_new(j+1,1) + C4_kp1(j+1,1) ; C4_new(j,1) = (C4_new(j+1,1) + (Dx/(area*Df))*Ka*C3_new(j,1)*EXP)/(1+(Dx/(area*Df))*Ka*Exp); j=M; C4_new(j,1) = ( 1/(1+Dx*KAPPA) )*C4_new(j-1,1); C3_new(j,1) = ( 1/(1+Dx*KAPPA) )*C3_new(j-1,1); C2_new(j,1) = ( 1/(1+Dx*KAPPA) )*C2_new(j-1,1); C1_new(j,1) = (C1_new(j-1,1) + KAPPA*Dx )/(1+ Dx*KAPPA); I_new(j,1) = (I_new(j-1,1) + I_old(M,1)*KAPPA*Dx )/(1+ Dx*KAPPA); % updating the old field QM(TimeStep,1)=C4_new(1,1); QMM(TimeStep,1)=C4_new(2,1); IMM(TimeStep,1)=I_new(1,1); I_old=I_new; C1_old=C1_new; C2_old=C2_new; C3_old=C3_new; C4_old=C4_new; %%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%% end Initial2(:,1)=C1_new; Initial2(:,2)=C2_new; Initial2(:,2)=C3_new; Initial2(:,4)=C4_new; % xlswrite('initial.xls', Initial2, 'Sheet 1', 'A1'); AAA=QM; aaa=QMM; J=2*96485*Df*Si*(1000000)*(QMM-QM)/Dx; % xlswrite('current.xls', J); hold plot(Time,J) %normalized current density plot(Time, IMM) 107 xlabel('Time(s)') ylabel('Current Density (microamp/cm^2)') hold % all concentrations are normalizzed to bulk concentration of phenylacetate (Si) % figure; % plot(x/L,C1_new) %normalized phenyl acetate % % figure; % % plot(x/L,C2_new) %normalized phenol % % figure; % plot(x/L,C3_new) %normalized catechol % % % figure; % plot(x/L,C4_new) %normalized Quinone 108 5 Use of Electrochemical Detection Techniques for Listeria Monocytogenes Ongoing Research 5.1 Introduction This chapter will discuss how our previous research in the field of electrochemical immunosensors and chemotaxis can be applied for Listeria monocytogenes (LM) research . This research was conducted in collaboration with Dr. Jonathan Hardy™s lab, which has expertise in the foodborn e pathogenic bacterium LM. LM is a facultative anaerobic, gram -positive , pathogenic bacterium that causes the infection listeriosis [149]. Listeriosis, a serious infection that has a global mortality rate of 24 %, is most likely to infect high -risk population groups, including pregnant women, their fetuses, adults over 65 years old, and immunocompromised people [63]. According to the Centers for Disease Control and prevention (CDC), 1600 cases of listeria are diagnosed within the United States annually, resulting in 260 deaths . LM can grow and survive under a wide range of environmental conditions , including high salt concentrations, aerobic and anaerobic environments, refrigeration temperatures, and acidic conditions. Besides, LM can produce biofilms on food production equipment , thereby sheltering LM cells from sterilization methods. For these reasons, LM has proven to be a ubiquitous and persi stent foodborne pathogen [63, 64]. To help avoid health risks associated with LM, it is important to detect this pathogen in food -processing environments and food samples [65] and as well as to learn more about the mechanisms this bacterium uses to survive under adverse conditions . Some of the com mon detection methods of LM are culturing, biosensors, enzyme linked immunosorbent assay (ELISA), and polymerase chain reaction (PCR) [65]. 109 Recently, LM was shown to secrete biologically active extracellular vesicles (EVs) despite the cells™ having a thick cell wall and no outer membrane [66-68]. These EVs, with a diameter ranging from 20 to 200 nm, can be used as toxin cargo to transport concentrated virulence factors to host cells [66]. We hypothesized thatan EEIB could be developed for detection of EVs secreted by LM. To our knowledge, this woulf be the first time that an e lectrochemical biosensor would have been developed for that purpose . Moreover, a recent study has shown that LM can respire under anaerobic conditions using an extracellular electron transfer mechanism (EET). In EET, cells discard electrons generated during respiration by using molecular electron carriers to transporting the electrons from the cytosol to the exterior of the cell [150 ], rather than by reacting them with O2 to form H2O. This study also showed that FmnB, a flavin containing membrane protein, plays an important role in the LM™s EET mechanism [151]. Previously, a collaboration between the laboratories of Dr. Worden and Dr. James Ti edje demonstrated for the first time a new mechanism by which the motile, facultative anaerobe Shewanella oneidensis , locates extracellular electron acceptors needed to carry out EET. In this new mechanism, denoted mediated energy taxis , S. oneidensis secretes a reduces electron carrying molecule (e.g., the flavin derivative riboflavin ), which served as both an electron shuttle and a chemoattractant to direct cell movement toward local insoluble electron acceptors (IEA) . The reduced riboflavin diffuses away from the cells in all directions. Molecules that encounter an IEA (e.g. an iron oxide particle ) are oxidized. The re sulting oxidized riboflavin diffuses away from the IEA particle, creating a spatial gradient of the oxidi zed riboflavin, which serves as a chemoattractant for S. oneidensis cells that draws the cells to the IEA particle [152-155]. 110 Based on these recent discoveries , we hypothesized that LM also uses an oxidized flavin derivative as a chemoattractant to direct bacteria movement toward IEA particles . We also hypothesized that secreted EVs by LM may contain a reduced flavin or FmnB , and the cells can migrate toward secreted EVs that contained oxidized flavin or FmnB . To test this hypothesis, we developed develop swarm plate assays (motility assays) to study LM chemotaxis. In Dr. Worden™s previous research s warm plate assay s have provide d graphical measurement of chemotaxis rates and been used to validate mathematic al models of chemotaxis for E. coli and S. oneidensis [70, 152]. . This chapter will discuss the motility assa ys of LM in complex media and a defined media. Initial motility assays were performed in a semi -solid complex media, brain -heart infusion (BHI) to confirm motility of LM and optimize temperature and agar concentration for the motility assay. Then, a chemi cally defined medium was formulated for LM to study chemotaxis of LM. Having a chemical ly defined medium allows us to study the chemotactic behavior of LM under a controlled growth environment , and it allows to adjust the concentration of each chemical. The defined medium will be useful to investigate LM™s chemotactic properties, and possibly energy taxis, to riboflavin and external electron acceptors such as iron oxide . As described below, promising results were obtained for the motility assays of LM in a complex media and a chemically defined media . However , this project was paused due to the COVID -19 pandemic . Suggestions for the continuation of this project and future work will be discuss ed in the next chapter. 111 5.2 Materials and Instrumentation Brain heart infusion broth (BHI), RPMI 1640 amino acid solution (50X), RPMI -1640 vitamin solution mix (100X), thioctic acid, UltraPureŽ Agarose, M9 Minimal Salts, 5X, magnesium sulfate, L -glutamine, glu cose, and ferric citrate were purchased from Sigma Aldrich. Listeria monocytogenes WT strain 10403S and luminescent 10403S (C1) were used for the motility assays. 5.3 Experimental Methods 5.3.1 Cultivating LM A day before performing motility assays, LM was streaked out on a 4% BHI agar plate and A sterile platinum wire was used to transfer cultivated cells for inoculation of plates. 5.3.2 Preparation of Chemically Defined Media for LM Table 5.1 includes the final concentrations for a previously developed defined media [156] for LM and two different formula that were developed in our work to optimize the motility assays. M9 minimal media was prepared following the instruction provided by Sigma Aldrich. The optimized defined media was prepared by mixing the following solutions: 20 mL of 5X M9 media, 20 mL of 5X amino acid solution, 2 mL of 30 g/L L -glutamine, 10 mL of 100 g/L glucose, 1 mL of 0.2 g/L of ferric citrate, 1 mL of 40 g/L of magnesium sulfate, and 0. 5 mg/L thioctic acid in ethanol. The total volume was brought to 95 mL with DI water (5 mL was saved for the addition of agarose solution). 5.4 Motility Assay Agar plates (BHI or chemically defined media) were prepared with different agar concentrations: 0.4 %, 0.3 %, 0.2 %, 0.15 %, 0.1 %. A 6 % stock agar solution was prepared in DI water. To prepare an agar plate, an appropriate amount of the melted stock agar solution was added to the 112 media. 10 mL of the media containing agar solution was added to a 6 mm petri dish. The petri dish was stored at 4 or four hours , and afterward it was ready for inoculation. Next, each plate was inoculated with LM using the tip of an inoculation wire that contacted the cultivated LM. Then, before carrying the LM. To inoculate properly, it is important to prevent contacting of the wire from the bottom of the plate. The inoculation wire was inse rted in the agar plate for around 2 mm). 5.5 Results and Discussion 5.5.1 Optimization of agar concentration and temperature Temperature and agar concentration are two significant factors affecting motility of LM. We performed motility assays for wild type ( WT) stra in in BHI with different agar concentrations (0.4 %, 0.3 %, 0.2 %, concentration , for the motility of LM, and this condition was selected for the next assays. 5.5.2 Motility Assay in Complex Media (BHI) Motility (swarm plate ) assay provides qualitative observation of chemotaxis. Besides motility assays of wild type LM , we were also interested to study motility assays of a luminescent strain of LM because bioluminescence imagining of the plates would provide a quantitative measure of bacteria growth. In Dr. Hardy™s lab , WT strain 10403S was made luminescent by chromosol integration of lux -kan transpos on cassette. Using this technique, a motile luminescent strain (1C) was created [157 ]. Figure 5.1 shows the motility assay for luminescent strain (1C) in 0.15 % agar plate at 25 113 Figure 5.1. Motility assays of luminescent LM (1C) in 0.15% agar in BHI. 5.5.3 Motility Assay of LM in a Chemically Defined Media After confirming the motility of wild type and luminescent strains of LM in a complex media, we sought to devel op a chemically defined media that would allows chemotaxis of LM to be studied under a known and controlled growth environment. Table 5.1 shows the chemical concentrations in each defined media. Using the first developed media, we did not observe a success ful growth for LM (Figure 5.2.A). In the next defined media, we increase s amino acid concentrations , and we added magnesium sulfate and thioctic acid to the media. This media successfully supported LM growth in the motility assay (Figure 5.2.B). 114 Table 5.1. Chemical formula for chemically defined m edia of LM. Chemical MWB media This work (unoptimized) This work (optimized) KH2PO 4 6.56 g/L 15 g/L 15 g/L NaHPO 4.7H2O 30.96g/L 33.9 g/L 33.9 g/L MgSO 4.7H2O 0.41g/L 0 0.4 g/L Ferric citrate 0.088 g/L 0.02 g/L 0.02 g/L Glucose 10 g/L 50 mM glycerol 10 g/L L-Leucine 0.1 g/L 0.05 g/L 0.5 g/L L-Isoleucine 0.1 g/L 0.05 g/L 0.5 g/L L-Valine 0.1 g/L 0.025 g/L 0.25 g/L L-Methionine 0.1 g/L 0.015 g/L 0.15 g/L L-Arginine 0.1 g/L 0.2 g/L 2 g/L L-Cysteine 0.1 g/L 0.05 g/L 0.5 g/L L-Histidine 0 0.015 g/L 0.15 g/L L-Tryptophan 0 0.005 mg/L 0.05 g/L L-Glutamine 0.6 g/L 0.6 g/L 0.146 g/L Riboflavin 0.5 mg/L 0.2 mg/L 0.4 mg/L Thiamine 1.0 mg/mL 1 mg/L 2 mg/L Biotin 0.5 mg/L 0.2 mg/L 0.4 mg/L Thioctic acid 0.005 mg/L 0 0.005 mg/L 115 Figure 5.2. A: Motility assay of wild type LM in the unoptimized defined media. B: Motility assay of wild type LM in the optimized defined media. Some crystals were formed after the addition of magnesium sulfate in the optimized media. 5.6 Conclusion In this chapter, we developed motility assays of WT , and lum inescent strai ns of LM in a complex media ; the agar concentration and temperature were optimized for the se motility assays. Because we were interested in studying chemotaxis of LM, a chemically defined media was developed to study LM chemotaxis under a controlled growth condition. We developed a formulation for a chemically defined media which successfully supported WT LM growth and facilitated a motility assay for LM under known growth condition s. 116 6 Summary and Recommendations for Future Work 6.1 Summary This dissertation describes , experimental and theoretical studies of two type s of electrochemical biosensors. The first type, a n electrochemical immunosensor (EI) , was fabricated on screen -printed electrodes (SPEs) for the detection of a model analyte (mo use IgG) . The EI concept in integrat es the princip les of an enzyme -lab eled immunosorbent assay (ELISA) and an electrochemical transducer using horseradish peroxidase (HRP) as the labeling enzyme. High throughput optical ELISAs were used to validat e the functionality of antibodies against an analyte to aid in developing the EI. The experimental conditions , such as substrates concentrations, pH, and applied voltage , were optimized using a fractional factorial design. A mechanistic mathematical model was developed to simulate the steady -state signal in the EI by solving a system of coupled, non-linear ordinary differential mass -balance equations that described the rates of chemical reaction and diffusion -based mass transfer rates for all the reactants. A new dimensionless group , the current -control coefficient, was defined and used to characterize the exten t that each reaction and/or mass -transfer step limit s the current density. The mathematical model and associated new dimensionless groups provide powerful new tools fop to predict the rate -limiting step and optimize experimental condition s for improving the sensitivity of detection for EIs, and the current -control -coefficient concept could also be extended to other types of amperometric biosensors . The seco nd type of electrochemical biosensor, a bi-enzyme electrochemical biosensor containing AChE was also fabricated on SPE. It™s ability to detect a model AChE inhibitor, phenylmethylsulfonyl fluoride (PMSF), w as then characterized. The bi -enzyme biosensor h ad AChE and tyrosinase coimmobilized on the gold working electrode . The use of a substrate (phenylacetate) that was cleaved by AChE to produce phenol, together with the phenolase and 117 catecholase activities of tyrosinase, provided a redox -recycle signal amp lification system that significantly improved the biosensor™s sensitivit y. A mechanistic, unsteady -state mathematical model was developed to simulate the time-dependent electrochemical signal in the IBE. The model consisted of a system of coupled, non-line ar, partial differential mass -balance equations that described the rates of chemical reaction and diffusion -based mass transfer for all the reactants , including PMSF. The model was able to reproduce dynamics of the bi -enzyme amperometric biosensor™s respon se a step change in the phenylacetate and PMSF . Using the model and the current -control coefficient and sensitivity parameters, the effect s of the governing factors , (e.g., the relative concentrations of the AChE and tyrosinase enzymes on the working electrode) on the biosensor™s sensitivity w ere characterized . The model and associated dimensionless groups provide new insights that can facilitate efforts to design and optimize be -enzyme biosensors in general, and biosensors to AChE inhibitors specifically. Finally, we established the groundwork for developing an EI biosensor to detect EVs produced by Listeria monocytogenes (LM). Such a biosensor would be valuable for two purposes. First, it could be used to dete ct the presence of LM in a sample (e.g. a food product) . Second, it could be use to measure the role of LM EVs in extracellular electron transfer (EET) , which enables facultative anaerobic bacteri a like LM respire in the absence of oxygen by shuttling the electrons from the inside the cell to electron acceptors in the extracellular enviroment . We hypothesized that LM might use EVs to transport electrons produced in anaerobic respiration away from the cell and to help identify the location of nearby extracel lular electron acceptors. To help test this hypothesis, we developed motility assays for LMin both complex media and a chemically defined medi um. 118 6.2 Enzyme Linked Immunosorbent Assay (ELISA) In this chapter, optical ELISAs for detection of two different ana lytes, mouse IgG and extracellular vesicles from breast cancer cells, were presented . Because ELISAs provide a high throughput platform to develop standard immunoassays, they were used to find suitable bioreceptors to detect the analyte of interest before developing an electrochemical immunosensor. It is suggested that similar optical ELISA studies be conducted to detect EVs from LM. To our knowledge, no such ELISA assay has been developed . . Based on the literature review of pro teomics of EVs from LM, we suggest that ActA protein (a membrane protein found in LM™s EVs) be evaluated as a surface biomarker for detecting EVs secreted by LM. Antibodies against LM™s ActA protein are commercially available.. Once a successful ELISA is developed, the same antibody would be integrated with an electrochemical transducer to develop an electrochemical immunosensor for the detection of EVs from LM. 6.3 Integrated Experimental and Teoretical Studies on an Electrochemical Immunosensor (EI) In this chapter , theoretical and experimental studies of an EI for a model antigen, mouse IgG, were presented. The model is suitable to optimize the steady -state current in the EI. We recommend that the model be extended to include equations describing equilibrium partitioning of the antigen binding to the immobilized captured antibody and enzyme -labeled detection antibody. This extension would allow investigators to optimize antibody concentrations and estimate HRP concentration as a function of analyte concentrat ion. This capability would be useful for optimization of antibody -antigen kinetics and designing EIs that meet specific performance criteria. 119 6.4 Theoretical and Experimental Studies of an Inhibition -based Bi-enzyme Electrochemical Biosensor (IBE) for Detection of Organophosphorus Compounds In this chapter, theoretical and experimental studies of a bi-enzyme biosensor for a model inhibitor of AChE, PMSF, were presented. We recommend that the commercial prospects of an SPE to detect toxic OPs that are widely used in agriculture, such as methamidophos , be evaluated . Such a biosensor would enable food and environmental samples to be checked for the presence of OPs that pose serious health issues for non -targeted species such as human and animals. The mathematical model c ould be applied to an OP of interest by adding its inhibition kinetic constants to the model. Using the model, governing factors such as AChE concentration, Tyr concentration, applied voltage and substrate concentration c ould be optimized for achieving optimum sensitivity for the inhibitor. 6.5 Use of Electrochemical Detection Techniques for Listeria Monocytogenes Ongoing Research In this chapter we proposed that LM EVs are might be used to discard electrons produced during anaerobic respiration and that they might be involved in an energy -taxis mechanism to increase LM cells™ chances of survival under anaerobic conditions. To test this hypothes es, we recommend that an EI be developed to detect EVs generated by LM . We also recommend that the chemically defined media we formulated for LM motility assays be used to assess LM chemotaxis in the presence of oxidized flavins (e.g., riboflavin) and its own EVs after they have b een oxidized. These experiments could provide important insight into LM‚s persistence in food -processing plants and its pathogenicity. 120 REFERENCES 121 REFERENCES 1. Cammann, K., Bio -sensors based on ion -selective electrodes. Fresenius' Zeitschrift für Analytische Chemie, 1977. 287(1): p. 1 -9. 2. Karunakaran, C., R. Rajkumar, and K. Bhargava, Introduction to biosensors , in Biosensors and bioelectronics 2015, Elsevier. p. 1 -68. 3. Yoon, J. -Y., Introduction to biosensors: from electric circuits to immunosensors 2016: Springer. 4. Mehrotra, P., Biosensors and their applications ŒA review. Journal of oral biology and craniofacial research, 2016. 6(2): p. 153 -159. 5. Eggins, B.R., Biosensors: an int roduction 2013: Springer -Verlag. 6. Cass, T., Biosensors 2004: Oxford University Press. 7. Gorton, L., Biosensors and modern biospecific analytical techniques 2005: Elsevier. 8. Scheller, F. and F. Schubert, Biosensors 1991: Elsevier. 9. Knopf, G.K. and A. S. Bassi, Smart biosensor technology 2018: CRC press. 10. Ronkainen, N.J., H.B. Halsall, and W.R. Heineman, Electrochemical biosensors. Chemical Society Reviews, 2010. 39(5): p. 1747 -1763. 11. Eggins, B.R., Chemical sensors and biosensors . Vol. 28. 2008: John Wiley & Sons. 12. Grieshaber, D., et al., Electrochemical biosensors -sensor principles and architectures. Sensors, 2008. 8(3): p. 1400 -1458. 13. Ducheyne, P., Comprehensive biomaterials II 2017: Elsevier. 14. Pisoschi, A.M., Potentiometric biosensors: concept and analytical applications -an editorial. Biochem Anal Biochem, 2016. 5(3): p. 19 -20. 15. Pereira, A., M. Sales, and L. Rodrigues, Biosensors for Rapid Detection of Breast Cancer Biomarkers , in Advanced Biosensors for He alth Care Applications 2019, Elsevier. p. 71 -103. 16. A review on impedimetric biosensors. Artificial cells, nanomedicine, and biotechnology, 2016. 44(1): p. 248 -262. 122 17. Wang, J., Amperometric biosensors for clinical an d therapeutic drug monitoring: a review. Journal of pharmaceutical and biomedical analysis, 1999. 19(1-2): p. 47 -53. 18. Lequin, R.M., Enzyme immunoassay (EIA)/enzyme -linked immunosorbent assay (ELISA). Clinical chemistry, 2005. 51(12): p. 2415 -2418. 19. Borrebaeck, C.A., Antibodies in diagnostics Œfrom immunoassays to protein chips. Immunology today, 2000. 21(8): p. 379 -382. 20. Aydin, S., A short history, principles, and types of ELISA, and our laboratory experience with peptide/protein analyses using ELISA. Peptides, 2015. 72: p. 4 -15. 21. Crowther, J.R., ELISA: theory and practice . Vol. 42. 1995: Springer Science & Business Media. 22. Engvall, E., [28] Enzyme immunoassay ELISA and EMIT , in Methods in enzymology 1980, Elsevier. p. 419 -439. 23. Andaloussi, S.E., et al., Extracellular vesicles: biology and emerging therapeutic opportunities. Nature reviews Drug discovery, 2013. 12(5): p. 347 -357. 24. Rontogianni, S., et al., Proteomic profiling of extracellular vesicles allows for human breast ca ncer subtyping. Communications biology, 2019. 2(1): p. 1 -13. 25. Ricci, F., G. Adornetto, and G. Palleschi, A review of experimental aspects of electrochemical immunosensors. Electrochimica Acta, 2012. 84: p. 74 -83. 26. Wan, Y., et al., Development of electrochemical immunosensors towards point of care diagnostics. Biosensors and Bioelectronics, 2013. 47: p. 1 -11. 27. Applications of electrochemical immunosensors for early clinical diagnostics. Talanta, 2015. 132: p. 162 -174. 28. Ghindilis, A.L., et al., Immunosensors: electrochemical sensing and other engineering approaches. Biosensors and Bioelectronics, 1998. 13(1): p. 113 -131. 29. Skládal, P., Advances in electrochemical immunosensors. Electroanalysis, 1997. 9(10): p. 737-745. 30. Wu, J., et al., Biomedical and clinical applications of immunoassays and immunosensors for tumor markers. TrAC Trends in Analytical Chemistry, 2007. 26(7): p. 679 -688. 31. Wen, W., et al., Rece nt advances in electrochemical immunosensors. Analytical chemistry, 2016. 89(1): p. 138 -156. 123 32. Abuknesha, R.A., et al., Efficient labelling of antibodies with horseradish peroxidase using cyanuric chloride. Journal of immunological methods, 2005. 306(1-2): p. 211 -217. 33. Kulys, J. and R. Baronas, Modelling of amperometric biosensors in the case of substrate inhibition. Sensors, 2006. 6(11): p. 1513 -1522. 34. Nicell, J.A. and H. Wright, A model of peroxidase activity with inhibition by hydrogen peroxid e. Enzyme and Microbial Technology, 1997. 21(4): p. 302 -310. 35. Ruzgas, T., et al., Kinetic models of horseradish peroxidase action on a graphite electrode. Journal of Electroanalytical Chemistry, 1995. 391(1-2): p. 41 -49. 36. Kin etics of soluble and immobilized horseradish peroxidase -mediated oxidation of phenolic compounds. Biochemical engineering journal, 2007. 35(2): p. 126 -135. 37. Buchanan, I.D. and J.A. Nicell, removal from aqueous solution. Journal of Chemical Technology & Biotechnology: International Research in Process, Environmental & Clean Technology, 1999. 74(7): p. 669-674. 38. Raghav, R. and S. Srivastava, Immobilization strategy for enhancing sensitivity of immunose nsors: L -Asparagine ŒAuNPs as a promising alternative of EDC ŒNHS activated citrate ŒAuNPs for antibody immobilization. Biosensors and Bioelectronics, 2016. 78: p. 396-403. 39. Duan, C. and M.E. Meyerhoff, Immobilization of proteins on gold coated porous membranes via an activated self -assembled monolayer of thioctic acid. Microchimica Acta, 1995. 117(3-4): p. 195 -206. 40. Pei, Z., et al., Optimizing immobilization on two -dimensional carboxyl surface: pH dependence of antibody orientation and antig en binding capacity. Analytical biochemistry, 2010. 398(2): p. 161 -168. 41. Scientific, T. Carbodiimide Crosslinker Chemistry . Available from: https:// www.thermofisher.com/us/en/home/life -science/protein -biology/protein -biology -learning -center/protein -biology -resource -library/pierce -protein -methods /carbodiimide -crosslinker -chemistry.html . 42. Lomillo, M.A., J. Kauffmann, and M.A. Martinez, HRP -based biosensor for monitoring rifampicin. Biosensors and Bioelectronics, 2003. 18(9): p. 1165 -1171. 43. Hanrahan, G. and K. Lu, Application of factorial an d response surface methodology in modern experimental design and optimization. Critical Reviews in Analytical Chemistry, 2006. 36(3-4): p. 141 -151. 124 44. Choi, Y. -J., H.J. Chae, and E.Y. Kim, Steady -state oxidation model by horseradish peroxidase for the es timation of the non -inactivation zone in the enzymatic removal of pentachlorophenol. Journal of bioscience and bioengineering, 1999. 88(4): p. 368 -373. 45. Wu, Y., et al., Kinetic model for removal of phenol by horseradish peroxidase with PEG. Journal of Environmental Engineering, 1999. 125(5): p. 451 -458. 46. Mansouri Majoumerd, M. and H.R. Kariminia, Bisubstrate kinetic model for enzymatic decolorization of reactive black 5 by Coprinus cinereus Peroxidase. Iranian Journal of Chemistry and Chemical Engin eering (IJCCE), 2013. 32(2): p. 125 -134. 47. SATOH, T., Global epidemiology of organophosphate and carbamate poisonings , in Toxicology of Organophosphate & Carbamate Compounds 2006, Elsevier. p. 89 -100. 48. Soltaninejad, K. and S. Shadnia, History of the use and epidemiology of organophosphorus poisoning , in Basic and Clinical Toxicology of Organophosphorus Compounds 2014, Springer. p. 25 -43. 49. Sultatos, L.G., Mammalian toxicology of organophosphorus pestic ides. Journal of Toxicology and Environmental Health, Part A Current Issues, 1994. 43(3): p. 271 -289. 50. Davis, C.S. and R.J. Richardson, Organophosphorus compounds. Experimental and clinical neurotoxicology, 1980: p. 527 -544. 51. Pyridinium oximes in the treatment of poisoning with organophosphorus compounds , in Handbook of Toxicology of Chemical Warfare Agents 2020, Elsevier. p. 1145 -1159. 52. Singh, B.K. and A. Walker, Microbial degradation of organophosphorus compounds. FEMS microbiology reviews, 2006. 30(3): p. 428 -471. 53. Barceló, D., et al., Determination of organophosphorus compounds in Mediterranean coastal waters and biota samples using gas chromatography with nitrogen -phosphorus and chemical ionization mass spectrometric detection. International journal of environmental analytical chemistry, 1990. 38(2): p. 199 -209. 54. Hauser, B., M. Schellin, and P. Popp, Membrane -assisted solvent extraction of triazines, organochlorine, and organophosphorus compound s in complex samples combined with large -volume injection -gas chromatography/mass spectrometric detection. Analytical chemistry, 2004. 76(20): p. 6029 -6038. 55. Jaffrezic -Renault, N., New trends in biosensors for organophosphorus pesticides. Sensors, 2001 . 1(2): p. 60 -74. 125 56. Dhull, V., et al., Acetylcholinesterase biosensors for electrochemical detection of organophosphorus compounds: a review. Biochemistry research international, 2013. 2013. 57. Boublik, Y., et al., Acetylcholinesterase engineering for detection of insecticide residues. Protein Engineering, 2002. 15(1): p. 43 -50. 58. Turdean, G.L., et al., Sensitive detection of organophosphorus pesticides using a needle type amperometric acetylcholinesterase -based b ioelectrode. Thiocholine electrochemistry and immobilised enzyme inhibition. Journal of enzyme inhibition and medicinal chemistry, 2002. 17(2): p. 107 -115. 59. Palchetti, I., et al., Determination of anticholinesterase pesticides in real samples using a disposable biosensor. Analytica Chimica Acta, 1997. 337(3): p. 315 -321. 60. Skládal, P., et al., Detection of carbamate pesticides in vegetable samples using Electroanalysis, 1997. 9(14): p. 1083 -1087. 61. Kohli, N., et al ., Nanostructured biosensor for measuring neuropathy target esterase activity. Analytical chemistry, 2007. 79(14): p. 5196 -5203. 62. Pizarro -Cerda, J. and P. Cossart, Listeria monocytogenes: cell biology of invasion and intracellular growth. Microbiology spectrum, 2018. 6(6). 63. Farber, J. and P. Peterkin, Listeria monocytogenes, a food -borne pathogen. Microbiology and Molecular Biology Reviews, 1991. 55(3): p. 476 -511. 64. Swaminathan, B. and P. Gerner -Smidt, The epidemiology of human listeriosis. Micr obes and infection, 2007. 9(10): p. 1236 -1243. 65. Priyanka, B., R.K. Patil, and S. Dwarakanath, A review on detection methods used for foodborne pathogens. The Indian journal of medical research, 2016. 144(3): p. 327. 66. Coelho, C., et al., Listeria monocytogenes virulence factors, including listeriolysin O, are secreted in biologically active extracellular vesicles. Journal of Biological Chemistry, 2019. 294(4): p. 1202 -1217. 67. Lee, J.H., et al., extracellular membrane -derived vesicles in Listeria monocytogenes. PloS one, 2013. 8(8): p. e73196. 68. Vdovikova, S., et al., A novel role of Listeria monocytogenes membrane vesicles in inhibition of autophagy and cell death. Frontiers in cellular and infection microbiology, 2017. 7: p. 154. 126 69. Light, S.H., et al., A flavin -based extracellular electron transfer mechanism in diverse Gram -positive bacteria. Nature, 2018. 70. Li, R., et al., Soluble Electron Shuttles Can Mediate Energy Taxis toward Insoluble Electron Acceptors. Environmental Science & Technology, 2012. 46(5): p. 2813 -2820. 71. Kellogg, M.D., Measurement of Biological Materials , in Clinical and Translational Science 2017, Elsevier. p. 137 -155. 72. Matabaro, E., et al., Current immunoassay methods for the rapid detection of aflatoxin in milk and dairy products. Comprehensive reviews in food science and food safety, 2017. 16(5): p. 808 -820. 73. Gan, S.D. and K.R. Patel, Enz yme immunoassay and enzyme -linked immunosorbent assay. J Invest Dermatol, 2013. 133(9): p. e12. 74. Darwish, I.A., Immunoassay methods and their applications in pharmaceutical analysis: basic methodology and recent advances. International journal of biome dical science: IJBS, 2006. 2(3): p. 217. 75. medgadget. Available from: https:// www.medgadget.com/2019/11/enzyme -linked -immunosorbent -assay -elisa -market -2017-predicted -to-reach -2-3-billion -by-industry -size -till-2025-at-a-cagr -of-5-1-size -trends -revenue -analysis -by-pro ducts -techniques -therapeut.html . 76. MARKET STUDY REPORT. 77. Beyzavi, K., et al., Comparison of horseradish peroxidase and alkaline phosphatase -labelled antibodies in enzyme immunoassays. Annals of clinical biochemistry, 1987. 24(2): p. 145 -152. 78. Veitch, N.C., Horseradish peroxidase: a modern view of a classic enzyme. Phytochemistry, 2004. 65(3): p. 249 -259. 79. Raposo, G. and W. Stoorvogel, Extracellular vesicles: exosomes, microvesicles, and friends. Journal of Cell Biology, 2013. 200(4): p. 373 -383. 80. Van Doormaal, F., et al., Cell -derived microvesicles and cancer. Neth J Med, 2009. 67(7): p. 266 -73. 81. Buzas, E.I., et al., Emerging role of extracellular vesicles in in flammatory diseases. Nature Reviews Rheumatology, 2014. 10(6): p. 356 -364. 82. D'Souza -Schorey, C. and J.W. Clancy, Tumor -derived microvesicles: shedding light on novel microenvironment modulators and prospective cancer biomarkers. Genes & development, 20 12. 26(12): p. 1287 -1299. 127 83. Xie, H., et al., Extracellular vesicles based electrochemical biosensors for detection of cancer cells: A review. Chinese Chemical Letters, 2020. 84. Kumar, D., et al., Biomolecular characterization of exosomes released from cancer stem cells: possible implications for biomarker and treatment of cancer. Oncotarget, 2015. 6(5): p. 3280. 85. Rubinstein, E., et al., CD9, CD63, CD81, and CD82 are components of a surface integrins. European journal of immunology, 1996. 26(11): p. 2657 -2665. 86. Rybak, K. and S. Robatzek, Functions of extracellular vesicles in immunity and virulence. Plant physiology, 2019. 179(4): p. 1236 -1247. 87. Parsons, M.E., et al., A protocol for i mproved precision and increased confidence in nanoparticle tracking analysis concentration measurements between 50 and 120 nm in biological fluids. Frontiers in cardiovascular medicine, 2017. 4: p. 68. 88. Muhammad, N., et al., Engineering of the E. coli outer membrane protein FhuA to overcome the hydrophobic mismatch in thick polymeric membranes. Journal of nanobiotechnology, 2011. 9(1): p. 1 -9. 89. Nolan, L.C. and K.E. O™Connor, A spectrophotometric method for the quantification of an enzyme activity pr oducing 4 -substituted phenols: Determination of toluene -4-monooxygenase activity. Analytical biochemistry, 2005. 344(2): p. 224 -231. 90. Wang, J., Electrochemical glucose biosensors. Chemical Reviews, 2008. 108(2): p. 814 -825. 91. Heinemann, L. and D.C. Klonoff, Blood glucose meter market: this world is undergoing drastic changes. Journal of diabetes science and technology, 2013. 7(3): p. 584 -6. 92. Kang, M. -H., et al., Evaluation of four portable blood glucose meters in diabetic and non-diabetic dogs and cats. Veterinary Quarterly, 2016. 36(1): p. 2 -9. 93. Chen, A.L. and S.M. Yang, Replacing antibodies with aptamers in lateral flow immunoassay. Biosensors & Bioelectronics, 2015. 71: p. 230 -242. 94. Jiang, N., et al., Lateral and Ver tical Flow Assays for Point -of-Care Diagnostics. Advanced Healthcare Materials, 2019. 8(14). 95. Lisi, F., J.R. Peterson, and J.J. Gooding, The application of personal glucose meters as universal point -of-care diagnostic tools. Biosensors & Bioelectronics, 2020. 148. 128 96. Bahri, M., et al., Capacitance electrochemical biosensor based on silicon nitride transducer for TNF - Talanta, 2019: p. 120501. 97. Tallaprag ada, S.D., et al., Development of screen -printed electrode based immunosensor for the detection of HER2 antigen in human serum samples. Bioelectrochemistry, 2017. 118: p. 25 -30. 98. Dempsey, E. and D. Rathod, Disposable printed lateral flow electrochemica l immunosensors for human cardiac troponin T. IEEE Sensors Journal, 2018. 18(5): p. 1828-1834. 99. Katz, L.B., et al., Meeting the New FDA Standard for Accuracy of Self -Monitoring Blood Glucose Test Systems Intended for Home Use by Lay Users. Journal of d iabetes science and technology, 2020: p. 1932296820906184 -1932296820906184. 100. Belluzo, M.S., M.É. Ribone, and C.M. Lagier, Assembling amperometric biosensors for clinical diagnostics. Sensors, 2008. 8(3): p. 1366 -1399. 101. Branch, G. and M. Joslyn, The kinetics of the auto -oxidation of catechol in the presence of several foreign substances. Journal of the American Chemical Society, 1935. 57(12): p. 2388-2394. 102. Baronas, D., F. Ivanauskas, and R. Baronas, Mechanisms controlling the sensitivity of amperometric biosensors in flow injection analysis systems. Journal of mathematical chemistry, 2011. 49(8): p. 1521 -1534. 103. Loghambal, S. and L. Rajendran, Mathematical modeling in amperometric oxidase enzyme Œmembrane electrodes. Journal of Membrane Sci ence, 2011. 373(1-2): p. 20 -28. 104. Huang, J., W. Huang, and T. Wang, Catalytic and Inhibitory Kinetic Behavior of Horseradish Peroxidase on the Electrode Surface. Sensors, 2012. 12(11): p. 14556 -14569. 105. Ivanec -Goranina, R. and J. Kulys, Kinetic study of peroxidase -catalyzed oxidation of 1 -hydroxypyrene. Development of a nanomolar hydrogen peroxide detection system. Central European Journal of Biology, 2008. 3(3): p. 224. 106. Galende, P.P., et al., Kinet ics of Spanish broom peroxidase obeys a Ping -Pong Bi ŒBi mechanism with competitive inhibition by substrates. International journal of biological macromolecules, 2015. 81: p. 1005 -1011. 107. Deyhimi, F. and F. Nami, ay of hydrogen peroxide: A ping Œpong mechanism. International Journal of Chemical Kinetics, 2012. 44(10): p. 699 -704. 129 108. −imelevicius, D. and K. Petrauskas, Application of the Butler -Volmer Equation in Mathematical Modelling of Amperometric Biosensor. 109. Compton, R.G. and C.E. Banks, Understanding voltammetry 2011: World Scientific. 110. Richardson, R.J., et al., Neuropathy target esterase as a biomarker and biosensor of delayed neuropathic agents , in Handbook of Toxicology of Chemical Warfare Agents 2015, Elsevier. p. 935 -952. 111. Compton, R.G. and J. Wadhawan, Electrochemistry: Nanoelectrochemistry . Vol. 12. 2013: Royal Society of Chemistry. 112. Coche -Guerente, L., et al., Amplification of amperometric biosensor responses by electrochemical substr ate recycling: Part I. Theoretical treatment of the catechol Œpolyphenol oxidase system. Journal of Electroanalytical Chemistry, 1999. 470(1): p. 53 - 60. 113. Coche -Guerente, L., et al., Amplification of amperometric biosensor responses by electrochemical s ubstrate recycling: Part II. Experimental study of the catechol Œpolyphenol oxidase system immobilized in a laponite clay matrix. Journal of Electroanalytical Chemistry, 1999. 470(1): p. 61 -69. 114. Coche -Guerente, L., P. Labbe, and V. Mengeaud, Amplification of Amperometric Biosensor Responses by Electrochemical Substrate Recycling. 3. Theoretical and Hydrogels and La yer-by-Layer Self -Assembled Structures. Analytical chemistry, 2001. 73(14): p. 3206 -3218. 115. Kohli, N., et al., Theoretical and experimental study of bi -enzyme electrodes with substrate recycling. Journal of electroanalytical chemistry, 2010. 641(1-2): p. 104-110. 116. Indira, K. and L. Rajendran, Analytical Expressions Pertaining to the Concentration of Substrates and Product in Phenol -Polyphenol Oxidase System Immobilized in Laponite Hydrogels: A Reciprocal Competitive Inhibition Process. Advances in Physical Chemistry, 2012. 2012. 117. Achi, F., et al., Mathematical model and numerical simulation of inhibition based biosensor for the detection of Hg (II). Sensors and Actuators B: Chemical, 2015. 207: p. 413-423. 118. Meena, A. and L. Rajendran, Math ematical modeling of amperometric and potentiometric biosensors and system of non -linear equations ŒHomotopy perturbation approach. Journal of Electroanalytical Chemistry, 2010. 644(1): p. 50 -59. 119. Aıeris, V., et al., Modelling glucose dehydrogenase -based amperometric biosensor utilizing synergistic substrates conversion. Electrochimica Acta, 2014. 146: p. 752 -758. 130 120. BRENDA. The Comprehensive Enzyme Information System . Available from: https:// www.brenda -enzymes.org/ . 121. Amatore, C., et al., The real meaning of Nernst's steady diffusion layer concept under non-forced hydrodynamic conditions. A simple model based on Levich's seminal view of convection. Journal of Electroanalytical Chemis try, 2001. 500(1-2): p. 62 -70. 122. Fransaer, J. and M. Ammam, Mathematical modeling of the amperometric response to glucose of glucose oxidase films deposited by AC -electrophoresis. Journal of Sensor Technology, 2011. 1(02): p. 17. 123. Grunwald, P., Determination of effective diffusion coefficients Šan important parameters for the efficiency of immobilized biocatalysts. Biochemical education, 1989. 17(2): p. 99 -102. 124. Ximenes, V .F., et al., mediated degradation. Journal of pineal research, 2007. 42(3): p. 291 -296. 125. Kohli, N., et al ., Nanostructured biosensor for measuring neuropathy target esterase activity. Analytical Chemistry, 2007. 79(14): p. 5196 -5203. 126. Parthasarathy, P. and S. Vivekanandan, A numerical modelling of an amperometric -enzymatic based uric acid biosensor for G OUT arthritis diseases. Informatics in Medicine Unlocked, 2018. 12: p. 143 -147. 127. Ismail, I., G. Oluleye, and I. Oluwafemi, Mathematical modelling of an enzyme -based biosensor. Int J Biosen Bioelectron, 2017. 3(2): p. 265 -268. 128. Baronas, R., et al. , Effect of diffusion limitations on multianalyte determination from biased biosensor response. Sensors, 2014. 14(3): p. 4634 -4656. 129. Kacser, H., J.A. Burns, and D.A. Fell, The control of flux , 1995, Portland Press Limited. 130. Gerard, M., A. Chaubey , and B. Malhotra, Application of conducting polymers to biosensors. Biosensors and bioelectronics, 2002. 17(5): p. 345 -359. 131. Worsfold, P., et al., Encyclopedia of analytical science 2019: Elsevier. 132. Honeychurch, K.C. and M. Piano, Electrochemical (bio) sensors for environmental and food analyses , 2018, Multidisciplinary Digital Publishing Institute. 133. Bachan Upadhyay, L.S. and N. Verma, Enzyme inhibition based biosensors: A review. Analytical Letters, 2013. 46(2): p. 225 -241. 131 134. Gahlaut, A., Electrochemical biosensors for determination of organophosphorus compounds. Open Journal of Applied Biosensor, 2012. 1(01): p. 1. 135. Pundir, C. and A. Malik, Bio -sensing of organophosphoru s pesticides: A review. Biosensors and Bioelectronics, 2019. 140: p. 111348. 136. Zheng, Y., et al., A review on the detoxification of organophosphorus compounds by microorganisms. African Journal of Microbiology Research, 2013. 7(20): p. 2127 -2134. 137. Pundir, C.S. and N. Chauhan, Acetylcholinesterase inhibition -based biosensors for pesticide determination: A review. Analytical Biochemistry, 2012. 429(1): p. 19 -31. 138. Xu, M. -L., et al., Detection of pesticide residues in food using surface -enhanced R aman spectroscopy: a review. Journal of agricultural and food chemistry, 2017. 65(32): p. 6719-6726. 139. Palchetti, I., S. Laschi, and M. Mascini, Electrochemical biosensor technology: application to pesticide detection , in Biosensors and Biodetection 2009, Springer. p. 115 - 126. 140. Kumaran, S. and M. Morita, Application of a cholinesterase biosensor to screen for organophosphorus pesticides extracted from soil. Talanta, 1995. 42(4): p. 649 -655. 141. Trojanowicz, M. and M.L. Hitchman, Determination of p esticides using electrochemical biosensors. TrAC Trends in Analytical Chemistry, 1996. 15(1): p. 38 -45. 142. Zhang, S., H. Zhao, and R. John, Development of a quantitative relationship between inhibition percentage and both incubation time and inhibitor c oncentration for inhibition biosensors Štheoretical and practical considerations. Biosensors and Bioelectronics, 2001. 16(9-12): p. 1119 -1126. 143. Choi, J. -W., J. Min, and W. -H. Lee, Signal analysis of fiber -optic biosensor for the detection of organophosphorus compounds in the contaminated water. Korean Journal of Chemical Engineering, 1997. 14(2): p. 101 -108. 144. Richardson, R.J., et al., Neuropathy target esterase as a biomarker and biosensor of delayed neuropathic agents , in Handbook of Tox icology of Chemical Warfare Agents 2020, Elsevier. p. 1005 -1025. 145. Pohanka, M., Acetylcholinesterase inhibitors: a patent review (2008 Œpresent). Expert opinion on therapeutic patents, 2012. 22(8): p. 871 -886. 146. Irie, T., K. Fukushi, and M. Iyo, Evaluation of phenylmethanesulfonyl fluoride (PMSF) as a tracer candidate mapping acetylcholinesterase in vivo. Nuclear medicine and biology, 1993. 20(8): p. 991 -992. 132 147. Kraut, D., et al., Inactivation studies of acetyl cholinesterase with phenylmethylsulfonyl fluoride. Molecular Pharmacology, 2000. 57(6): p. 1243 -1248. 148. Coche -Guerente, L., P. Labbé, and V. Mengeaud, Amplification of Amperometric Biosensor Responses by Electrochemical Substrate Recycling. 3. Theoreti cal and Hydrogels and Layer -by-Layer Self -Assembled Structures. Analytical chemistry, 2001. 73(14): p. 3206 -3218. 149. Ramaswamy, V., et al., Listeria -review of epidemiolo gy and pathogenesis. Journal of Microbiology Immunology and Infection, 2007. 40(1): p. 4. 150. Shi, L., et al., Extracellular electron transfer mechanisms between microorganisms and minerals. Nature Reviews Microbiology, 2016. 14(10): p. 651 -662. 151. Light, S.H., et al., A flavin -based extracellular electron transfer mechanism in diverse Gram -positive bacteria. Nature, 2018. 562(7725): p. 140 -144. 152. Li, R., et al., Shewanella oneidensis MR -1 Chemotaxis in a Diffusion Gradient Chamber. Environmental S cience & Technology, 2011. 45(3): p. 1014 -1020. 153. Li, R., et al., Soluble electron shuttles can mediate energy taxis toward insoluble electron acceptors. Environmental science & technology, 2012. 46(5): p. 2813 -2820. 154. Kotloski, N.J. and J.A. Gralnick, Flavin electron shuttles dominate extracellular electron transfer by Shewanella oneidensis. MBio, 2013. 4(1). 155. Oram, J. and L.J. Jeuken, Tactic response of Shewanella oneidensis MR -1 toward insoluble electron acceptor s. MBio, 2019. 10(1). 156. Premaratne, R.J., W. -J. Lin, and E.A. Johnson, Development of an improved chemically defined minimal medium for Listeria monocytogenes. Applied and Environmental Microbiology, 1991. 57(10): p. 3046 -3048. 157. Hardy, J., et al., Extracellular replication of Listeria monocytogenes in the murine gall bladder. Science, 2004. 303 (5659): p. 851 -853.